From b4180139e0ac8d433c4393280e775b1eaae808ac Mon Sep 17 00:00:00 2001 From: Alican Gok Date: Thu, 5 Oct 2023 23:09:01 +0300 Subject: [PATCH 1/2] Add KWSv3 models with new dataset splits --- trained/ai85-kws20_v3-qat8-q.pth.tar | Bin 2058871 -> 2061986 bytes trained/ai85-kws20_v3-qat8.log | 66905 ++++++++++++------------ trained/ai85-kws20_v3-qat8.pth.tar | Bin 2062088 -> 2070584 bytes trained/ai87-kws20_v3-qat8-q.pth.tar | Bin 1815560 -> 1815458 bytes trained/ai87-kws20_v3-qat8.log | 67105 +++++++++++++------------ trained/ai87-kws20_v3-qat8.pth.tar | Bin 1822792 -> 1824052 bytes 6 files changed, 67306 insertions(+), 66704 deletions(-) diff --git a/trained/ai85-kws20_v3-qat8-q.pth.tar b/trained/ai85-kws20_v3-qat8-q.pth.tar index 80727202aafd09816f708442a1776e7058df4a59..55b9069ef511771ea19ef22c72f46a16a222844f 100644 GIT binary patch literal 2061986 zcmc$_37A~hS*L%TD2Zd7P=FHx1Sr6O1I8&W)&e0u)r7>*2^38Lv8ohX#CE(?U6ld> z43{-xuVEGYWfyz1i#4YN1hF@}*e|TbE@-~Ri) zbMCFBTkX;MQ}=~%^UdK2s~6rnoUHD9;KA3Pdg1*iXD9EPyyoGTu01<^%$`+0uZiPs*uW%~3x?l}I8-21WW zF5?(|kK&-(laobcJR>0540d&l2t^_ZV@>cLk(Fgx+82WO|JZ}k8}Jg{v1^2Z*1 z^|iil-Ss_k_x@w_eVYezzuUTg&#~WgOTV|BO`o?heY@k&U)}Fj51u?-y1jkr_CTRN zw104LKl_@#L!o|tNBxdczf+-p=h?KsF@2Zg@3ESCdgkP#vj;!r)QOYat=FFPj!oaS zYto(Du>*$(hiUR|rOEAGlXthtdz2>cIh)?GF@3>}nw-4w=Rtcm-GBUDR_`P~G$+b_9zHcau$TYnu|vo94-QQa7XCvW z|K0K*F8oJk)1w>HV>h}9k4^3>_kH@l*HIrmc4Tn(;pzK5aLX&gf!Xv$a^JsRuiy0g z>H8~GUR-+IGn>YZ>2US)e_-$PzxKp7hX;e{2b3MWq}#yXq5xa+!S z^r5}{M~@yndg$PR>1zuA!yW%4@;_SmADc}l8`J5HZosCLc2lqatpK^VlBX=J>bZGjiWecCy zZQ*gZ@acv9GiK9zWBQq^AJT(+9@5Y1$d?c4(SwH$4~|YhyO4iQNB+5ze_kPf{cL)6 zWBU24&*$K-hx7}sqdqp+@2&m9Lj6S@^%qP1C58G+XVdw{^xv(1eh#jDNWX0F;rIb` zm>1cXmnL7)HTg=Ld{t@k)wAijjp^6isL85_^lSGtIkx}6em>*|)2}N{zP@Yn4L144 z(&U?F(`IA(&8r{MgV#T#-?E3_Jsdo+|M2u%3;(xu{NFDBcNG5boK4SfOuy?!H(}=? z{qDWgJfugC@gDs>zLDNLn|`m{k;jyW^!r}X?;g_cFHL@6Hr?Eq{^06|^w8di^oREJ z$w%a;<`0)G{7AQjA9V{qR@i@hHodSh{fX5N>7hLj=}&g#%ZD`I$Wahju-r7oSLdj2G9>73!bwsDDA~Hx}w&oJ}unOn+(h^K)qBL;A~4 zIGp@J`75Q#uXas-%_hHIn*7FWy0tO=%^NjY^^pG76Pq07A^Ppo7nZ%(m&k8??ZZQ|KX$4KPvox-0}a3{C`^b|7Q7nqnohvkp9J9 z>SOzl9T*%uF#Ss((%Wa#zmj{vW6DGN*Xyagk^W8T@wc<-m5u4&9e>B2Z!E9*q|>Xu z>;CD7(te>?v6tH$2>od54nAU@zn%>O9F|Jo7%Pl^9mA^zX9 zD%=#pMs*X{Rag7aO2_JHs5Gsfj?8_lXQ0ih zo4IaO&*Zwg^@``Vx`pgLt-Uqe{z3H?Msn+0(Yn^pLbdg?k)`!pqRpxv*Ny6}xbA+x z+PPnQzZX<*P2!%eJkZB@9tPFh7|C64L+iRe2i306MV78_i#DsC$91E6JFcs5<+W$L zdV6yAHsmqi&lke#9gO5By?Q>X9V3tXQ(s@bBj)-%KkwB$p_k6@j5e$KTsNwB;kxSC z-}^ncdRNl+G(WbVho8@|>fMaw-QJGY?e^VK-R*lIOV9U2n^kvk-Kbu`b=7NP&$r*| zy~yfXUVB*_JGTGs1J!#Q$t~|h>sr1Ksx9wAmXn^gm@8`XZUs}6YAcjD>*IZteO z^zOseK_j{0A+)aH-KaJ^j4Tb0pv|hITsNv?Tvt89E59jM?@RV|$9v$w-Mmz*_cM}P zzX+{s{r;%7elfDNz6WhqMXnpwkn8H!t6n{8WbbL6U${q)4yq3@l3Tw7t!w>(sJ4D7 zvb25~+N`>l>qhlKTvxq%uK!kDeK3i8yB<4m=Qg z)p4#H)kkw(_3GLC1-$wg(ynX1|L9?Un5#b4NZ#$op>@0ccvN@$3CPm(1lp_`bKR&; za$WW6+4Hr$dKFn+%WJQmqx^_BsO~e8Ti%b>wLFDt%TGj>maj&eRS$68s2=3H>VS8B zS+5=<=ZOuE@B#1|Be~(jXkEicP;K}qvNU`QZB|XVZd6mQs~+K%U*W5Y?CXyAz=1=D z4jrvd8_BI_XkF_wsJ4DBvb6prv|07ZTsNxMab4Yd)vM=I$iA*M?~BI{AFV#sNN)XU zXkF{aQEmO{$kO^V&}LQ5b)))BuB%=>*MHrwK8wUXU5_5)-GaMYeYTO@^>fg=uAhr) z*Uv+iuCGU%RcEU&p!=RUyQl@ zfuQ;ljMDl`(Pq`0>qhnOxUPEn?ETT8`ZCh?bmyb?@S#Kd4_9AqByaaC(7Nq@C92!~ zDrD*T)o8Qo9M_HNYq+j@`Rw_Dq54{~x|ZF`hZDY=ufx^X8Obfb9<6Kn4XC#KMr3LE zO=z>K;kr?MGuKroyz9q^>RZTpLc_xcd62%buF_bG-S7FMeBAeUFjc`g_s3*58L}>+eUF)<1wYt2ViAR6oddb?a3x zpC2Oo`ql^bA3AcN`e7rv^^c%+t$!5N)<1?Ut$!SCR$bt_QT+thRWG0Ge-NpDlEgh- zefjV?Kd64nNbdU6XkFKzLAC48B1_jdpv|hpb)#BvU43(RUOpGe+0)Q(sQZ00{y8Ij z_-0OGHx?aDFYOmizmR`S)Hmfdk-KhS6>+!eQ)9c!gG}Rvxd}1el z;ZgmOk=*Hz(Yj85f@-HfMV3x~hBm8Ot{c^#bKU(8#1XFg>E{T&@}^|~9l^J@6-1ow3E{pK*gbgTY{5nc_?nN|OaYOCkGBHZ+O)&Imo zSx5a}sHNfmM&pOor`Yf&u4j((k7oH_$Ev4b9)GshgGcY?N4NunM^DXOd$4+{{@d_B zHkm#62>+x~o!I#&v1%XD$E&AtE&q0wL)Fu>_ZfQEpZ<7ta}L|-^z786dM4%v|7v<_ z$p6N5>XRoYPQLcl)h&4UY>~q}GT6_Tp|>!?77xs-TT$I&`DcjgS(x3wL{!g4FPnTz zG=4Va%Fm`;^FQNtZBCq?R&R~(iJkcEV)ZsgN~haUcXfIW=Cads(MzYdMdRmDuKYa8 z_00OS9#x_UmU?aJ@2t9Qg){`$IlCydhXozeIy zlq)}la(!s!Ru1zFR_{u<<9;{PUEH^0UgLgujKcjMXtU})xo%WP@a4* z`~<(Be}Jmqi%gI1-dXkDsP-$r;;in(y!NH$eK1PDyU=FU3%PDo1FkFfBKILaQ}+|@ zxDTN2;y#FZjr$Nr;l3MfRvqTLQ61rWX8j!=ua2TT`Cj;W>fjOX%P}(Tcl)e*UsU^* z--K1~hq?Um=0)hG-TR}>suy$JsP5sqVk=M|()LqnTm`mypKrhrUL7P<{ z$aSN7Dc3XW@A`Q4GL$#9lOspySlvrz*Y1N*%gNe=%wAu(Pq^vxNcM* z!ga+?jvgEE52r_~4<+1Dk5G3}e;DSHdIP;se>mE#dL`G5>La+GS%1Of)kmT{=}wLf z4jvr%W64L6X*-^;tzv1qgE;cd<{xy!Kt}lQBxa*P+d-PvN>zeJa;g zd%2rm)bpzUG{PPCa)0>S%2~4)n}u;iM<>d96rJi z4WC1%{rDvPTvYp&PtwoB>OM(dk6N0YMVnQh&vm2v0 z>WIGtwGe+P8b9oD<%d14XU-hIl~2ye$?D`ICth7Wc4l(A`ZD|$j}G#rm!5&su->`dhB|*57LVt=HdX{kHYztUq`CZP%Z-{&wqczy1#E&tHGX z^>)zW(m(@3H=#>vycbVEw(;-+TSe_4ir7YyE}1ZSP+{uzqm; z(E8o$hu4p+A6-AT{=Vz)xBjB__g{bUsWT_fe>=~UlN=7;S5(R}w`9qoMo|BjsY=KrZo^_@>vu&G8`vVv9T^<%kU<#|2fSA9SJWJB+q*M*$6^1QD5W!g`W`M{g=kLJ}+-njMd z-(3AvPFs0itAF40(@$2g>b!m?7py$5C;V>bXP<27o%4D_PJ5I8=36E59(dFLfLkpx z-;@7iY;`dwz1jby`g1oiQ5#_6jL=~r_4ibHzB8`ZDgc-Wmo`n8<2 z;*hR;llpb?9y)o0L)!ZfH@|VCzPsO}wsO*nLt6a~^_w?NUv)^omD5)o(i7gDe*4D5 z?i|wZoA>Ma-@8%Y-EU04pOaP`((1RR%QsG6bx41Z(^nkQ6W)^k z@W#XL9MT`aTIXEPmv^ z4}H5g{_Q^;enzX=h)xuiM<7}W$7csQT4Y(*F*kO?_8qqO7Z#iHv95l%2$7< z_l?KjQi6vcoJ<~k^wlRGIW>FWvHPpPCv^N-1YR*beq!x|UU}lf@BN^czOwoU+-J_* zHClhh&5wW8$&Z=6kRN%#=!Sp8|D5p5J%7Xh!sVZczjWD|C|2#ua*A|`{%#t|F`R3{t))xvcG?*{D0Ix zg!&$zy@ww9=I`Cdb0?nc**kXqv&X;B@|WH(UAz1L|Fh?#`TwB5f9?N&(Z7u$B+!Ty zLRb{^kP*~x0LH+47($EeA@$YwgwQ8v+@Vfzfk=&sG zAOY*-YLV8>7a?qc2J}GAC+1AnwO|59AkV%iyc1*smq3KaIjk}N0=NRs0hfEa3a)@{ za2X85klx@bxOJ__v*^wr7d$-|;RMeE7qCabnQkUmJRewRpOzZ;aL2`xubg9_1Bv)P z@@@xrf^FNz!#!_;F{pvFhG97)XP@AcHv*S60o1Xa zZ-XAuMGG2g2B5}oj(rk(K%b+JL0FXhCAazJXj`Wose=~V`xH8<-WwLXiKUk^`*3G2 z>?V~S_KgDDWxX+<%@*0l(px%GH28Gcwx=Njzb< zAlKo`y*L|lJfn!5z@hZnYe9{E8K2LreYWxWOeg$4zqf%&mwO)R^E?9P*4Xko*U~4i z{Tu50_}!bi&f5pdK&*#<`Fd)IyJ9Rb&t=X8m_NsN6`Tdm`|`E&I=GDeqKoCVQsehE zH3$3I41!M2BcxEbH`L4;=u+di(iC>%}X zywc7yBv-)K5KqAUZR1OzM%Q-%1a$Y}nVLVwHvkdHH^A>4_H*CPU`|V}`MYvc&o;9EYML~>4xZ$Qj^^)?o1uC)`8(5K{r3YRUG0-RND`J6#JpxZN`%jwKfBy^x}0tO&}mOQx{V2x*N|5}}% zF6)+O-zU!+edagFO?VP=Ow9Nks2lVB*LWky_x@f64G79bc_x%C`R3I4<=6xX+gV#M z0-lfY1h#;AVNu?*6VjggyuW>r=f>|0V-Sds@i)NxBY(v2`)vy*N_>}w<$ITT0oePt zi{1M?dL+mB?pQ1D-6QlGf1jKHmhaye@ZV0(Rb-+jkh2BMv6sCk!*V89kj`e_CVoF> zuOjVjj-N#XAeZ~v#(yt4e$Vt==J@S(6l|FTkDpCP!4={?;Q8+(o{;t$aZx9~fY zvhH%ukz=1RFsA{LT4&w_1JDP?YT#VLya|5KavPin);X^;d7kdC#pk~KE?}NHmyD5P z-w9~&J&$-pJiivOLn=MoTffWe5g&j#aHgQWD1FW%6Ey*84|8iUR$>QW&Kzulm`dM7 zJfAz#p2obF%9a{A++$cPvo&4ZJv8JS+w5Td2!8_3tZjd1o&eVb{}{OJopX_OEN5(9 zN@cc)-Yfu8UhB+@#FxE%UFa8m^nDl0z3?0d=$`w4SPSgeQfm)=J!0PfZQ%3np6oLp zmUA-SWliqc+4kB(-$FLT%s-EAZHwMS=0x8{kH|!hG4~OO*=GWb_kiE0e15zSee?;^ zb9FX(-Iuba_N+E=Uk{M`BBVaoC+A%=XW{dE*`MWGu8le z)%(aAna>qHqA&T(3$etU*H{qbIit4oyYE^F+Jr!g3H}zTzXr=aH=VBZygU=<^c;M? zT+3PIHpia!PC%cWb9_GEH9j%Iw&wt8KY1H;?TDW6$-{mvqnd7ejuv~Gm%o&vhPN9b}~K~G3+&vF3f=kP^V-5OQ^cE>1`UL+N8IT$11AQQ$0roj` z^ZV#7_vTE_v*f#kJs>WpbxoH;dY;+d`ttlpZF%h3?_f{+STC1z&A}!(5A?@jc};F1 z!&Cz2FZZcF0b}5E@7~< z$cS9p*64Sh7Afxl-QO^x2*bN@9m0Qmwj=Q{_qN63I|V|k4_ z-x%at{COdBUM#$FX%B#Nnk%nMT^Mr)`$uGpuLjO5zjhC4P6H<9g9zl#eE7}10OU#7 zzW-ReJO^{Fz({!a`vai5O?NcfDM1)fhIL~_Q%l4GuC7RjqYLtG4i_p(pSxP0!%CFgiia&jJajv0@{ ziD?f&i@tBI%qo8nE@J(QJMRm3qT4HiK#eQLZZ77qwr7krLi)P!8S@O>*Bm{jf~J%5 z_?fQE`zOaeO^`k_J|ouLioT7{&%+)v#PZ$BXD7cd@LS_v1NqKqPo+O&UqOz6oIV?4 zVl^CneAZk=+CSnG;UY-oOgB5fHS)NpKDvAnsco;uI%?+V0p9@V+s3|xp3t|@^Lb#) zF#_^5)*(H!{2qwzZ0>i2Y`a?bXnu{~eY=cZ#@@p3a;6Z8rxNSox7WZpK6|)E*yn+M zd)argDDi~c1kUB$&M$l|ZX#=Z@>~V>zT;xYi(fwPi@%HTT&-*2aj)LT2=;*nyCJ_J ze&<^GeWv+QkT3UuynOG`t)CDJ;znLJD;)-dg`&{IZqAD73M8-2k2uE zQsJAEC(eRMuFL+H$g!pe%ncEKavEen%B$aaLe7y9wCD|JW5_kmZN7Vmpr=hNCeCGn z&e}q%Zv#1;Z}}dy+`BVg0b9Ts`3$WK6?7HBuXXHS^h4hwu1yw|lhe*5&u^>um$ z$f2}8bEML5Ff7lg`?&zNK?9=PAS~b_Hz9pydKZgNUioWsdPB|>Bp}Zi=yS<2UX-(R z<`KHKyza9>B+%fMDzpU;B&gx-Mc>*C9N z{c=TQNPwkpfG;6y>hHXWy;ff9#(mAc5?>@X0g+svx7+dg+Ha`y%#91rPoHNsK@aG) zNJU@vP1rsYm&ozFx7EQ_>_q-~^lkLp*2;a=H^*nMgmez!OnuOz_mG}LO|H7RJ~IQb zi9Sa*^tHzXec8jgYv3IE8W0wxCaHrL5j(V2_SeD@@%71RW0~gyx_di||1$ar+3)m* z9C^)m-}>cr=16P{J)#?v*ICV}!4|sGrS7vg(7s4L(tT_%%KM%i&f`kxo5bY0jO-%^ zAjNWjBEGt_$4GNW$U2pKk?$#X4}VRKIdk+0xQ*T~TI)W0@%tP&gZpw{qd~4FS%dmZ;;M+3BR8mfn51E@onPkAqU7U} z+7Z(Glkn*up`XVW(4EKo*2FTy7P_-UY-exC>0w)M%se@i3CQ1KyZ5+OzKc%qdCo09 zzhn5h<-N4;0Np%$de-j4S$vIqhC==bcR&VW5rhcR&Uv3V51)Ov(UWm>`!wWfC*)Wk zvIU-vdK1h2+$7g{i|t;=*jpf6EYIx~^gg z8seVA&FCZik$6Mi1l{M#-risJfZuy&-nG{?c7(%z0o}cr-y-cZ0DiyI7s1&@Hw(Tw zw)0yjr}xo$>sa#K-vnPzKhNnlwz-L%fZj*nw^sItC(mu%RU=z+O5~vXI?HmDVPh<8J?lFKE%kw^7bUtJcX!k)Rrv~OK z^Zlh)=85Gz8vSs}Zy)>RIQ~Am^JE`&O)S2!DBMI)@-<&lD*Xq<)tz76eOgmvx4^S; z<$WcTSkA_p--^Iq*0eF_qEYFfRlH)mQD$KH-o!w$vZ?Bek!nWSN%X7_j*onBy{p7tQBzhmnV|@hm zqUiDjZDlGw^EnVVFXGd77Hj3Gfp&u(l={&1S)*Tl1e{@l+!_|2wayZe)&;?`zqyfI zSFmbe*4_aBd zCnu1jzm4cD>q0F1(zmId0Mhf7|AIbO+TkDI8xf26oIP`3FS&Z?V#%T3-Ze4;=d?C; zwtaHE>#4twFADS;xR#vi`i$i`d^rbOS*JjN|B%k0uFsgh7BssMSjLU@7A0;?J_mf} zi`_ZK8h!RgdM4wpcA2+~1y)N=0OmP|eVi==I}wXO+g$e=@#h(opao4rXQ`XhAoCva z)rkwX&qhEuCw4H`GtTFnN^IHRvzAAw>sQa`&HkcwsXPzPKfrE@2Vic(o^&~$!31n} z=p!Swo`rKY;I=hvPq5c6mge zotzeIgDsG2;FDW8YYlSG&PT^`#=~+34RRBl2P05Z-=;F#Xi;Xqg7h;_uEDVMxrDTL zznj<3S>w*uXUAvpD|-mo!K&XA{O}%S%b*}9W3{v9DzOv zT~3P>_Ns@tkz?ef)3x>I{^Y5*_;SAIf-l!E(1Guf@;CTn7gNtR{tyeaN5i7m`0OXt z1F}KZso1fzlRkO|{1LdH30V57CrIaN(C6rhyg9lxV_=Q@i};N-*jKUTNWkTOd+5qs zlZeIgn)36+J^KvGV=rSNl|2mbY1hP^LH$yv2lOlGoA?{@wHu(UMJjqkCUW#&CV$^r z$@lk5evUQx51}XI74qDtGdY8``Ex)Y%ly#g>Yw0mBKIAiwKe+9=)JY#_jBni{(!$p z2=t{d;gjFEzI-o{eQf;$^lkKneh%GxQ=@wp)|eASEIEDr>X(TnbmLo{UgP(!m=qhrY|m z2J}IWYr}J8So-&nOMQ#jOV-N%Cl||X&bJTUp7Pq)8gUNn=Jo!!aF57|shpE}3EjN} zP@`Kj0ojM!c*M5GScv7^)XlNCb#dwMY-j5ccb~>;kR@{jbfN8XjwY7RY&mQ65x)H1 zhitJE2%t?xMs#~O+Q@*^XHJ&bNgG*rz{>HQi%)6(wH#)P_^!Qnh_ygFEXtgj8?!c+ zkklmPk|)QtnaTZYH^8M{r_v+m>Fn}ULoB^&rL!QLgpO2CAm2xQ%w(*=7dtzkJ7<;% z&o$ypspRH7bkElN*8=}tXM(KF!T*wr<@xt)&*GmT$4F~xxb!th?H`b7Fa38})9b1q$}IK(Ax@w}k=qt~=Js zzg6m=kQ0!bh(}^I zP{v^>E0$XKp6`3-cJGOpb*?5B{u(Jq16=CnMc^zWkYm`+U_78Vpat5-Yjoog*&kvv zW5U0*b1t(7d<{7@XwjEvkZXuXeDWmXjd2j|D<3|W=b_KsHkPy3Z?3j0Mr;9RLb}ZN zT+A6Dd%%@vh~y;;+P)r_ee&~*ZExp{!1pfuHmUR*quWb<_cjOiG{=1Bov0JHe{Wb&6O*^@ z?1k`-ozW?t7o9I4U z&bn`{{B4&t&qHtMk%)WtzVG`!w}mcGAZDI5=K9^*o~f(3LXO`R?Qzw7U?2A%$hEFT zx2C10Py8Hcfi>>SzJZwg-2&Dm~3!GuJ2xT5* z>`)`k=?zQW7-_9sH5d~Q2@YVcy=!!roGI?~8Bk}9wYzJqwO>NI%xh!0r`F4(Pi|+l zk8$&z%Xouryul||zz*p7?9f|uQHy)Zvk>w3t?y!T%dw2B_dp`vfMASXa@PVL`y1~Q z6D_*7Ir2yJ8pu8Fa`!pE_j1u#;Bu5W8Z$pl- zwnn;|L~bk`qhaA|@%elw{P~|w7FFwguZ5l6N_ zDW7LngT9h@1ab=Jlh4_l^D^)?m_PuYv9)91`O&CAKIiB)u#YkCqp_yb@#fkVJLv}z$4KFf&TmW(XER=FyHe@l z`Sd}Jt#qb|HZUj0(3d&q(9LOxwLm`4*_rx7{MfBN?0g>h^y}M1I-Ahvd=Z!@Pux{2 zzy0O5xA)L~>dq_zsDZw5rz=OmRi~0CZx|L^eK}j(`F+leJCm=K2ETKSKtQj-(ub#< zVM4m^ZoZs{o_iB>Ci`a_UE8Ja3cPdU!0%VqZll|Of{fUnNkeXn?o!r*9swCs$&=qa z=T~?BR*qnq#Y3~Lb! zk2UrP_y)*?zr|;KS+hwj;tRm{pm4Fw(V%ZqGe$a_wdSQR&sp53{vc9$mKyX+#N@5< z3Fn>Tb5G7c+{YFC`MjXGgnbs<_qXv{Icwu1{P|g- z?kf5ipS|bkx8w8oP6=NhdmH$B`=X< z{v7zMOgg=f6#2}Mu3Qt5X=mSAAC~*z+|D}!-uD2UTb@8oKqjO;dZ0xY5nbO1)YxO> z07Ucv?Asrf*RcHB_OM?B#_Z+38YTP9I=1}sdQLrHFZYyyedM=SgEUurGA#UKq`XS^ z5s??v!C)5xF}V}9n@DqOmr6xH*0F2Dv#bKKEol`ITc}zdX+ZesV=EI0O5g z$KN334xNfLzX;xrn#;&5OjGPriiD`Uuopkn=mAGn+5+!tXLa*C!01 z?o0`3{W6x%5{|}rEOqv9HL>^_^7Ta!=p#oh;PXA*H9^V|L5St-?4xdNKyNxdA_v&D zeqhaZ}QC<#RBh*Z6C4Cg3XA1bzly z00UsW1zW@Ny|G2w)0tx`=VEOFW8lon5s1_^z&%>y9@UjS;PPCovDY?eLEZ!Y8uW*7 zgWJ~1?+E<-7S>sF`^ED8&Dt#>R~z9NhP+Sb_IsGNT%!eZfdm?0O&=%aRoFcLY z*3=O{F?&WZSWt_sh4(pz91lq2`VzYFR>}F#sZqB0%n8_IU`@`^2dvYVbCJ$ncfN@_ z+)e7v9ouESwbnJj9M3o5Go~-^2Y-z(=OOJ8k#W)Wl@GnaXRZ4*KjGJw$J!QXTNj9# z(?f5O4OsG~SYm-?ECK6B)F@*I^F!+D@tIqL23^1Lgzh}XT+5m$x5%hrP z;jA@~uLlO8>1_E&Ai(Jy@;c`TY=Rp2I+L#@7J-~{E&0aiax7op`pvOVOhuPB49kqp zoPa$hATOl(b?VL;Imh;OFN}>qA9&u*qU;X~N5YoN^EDPhuhW}U>Us-eoi3kTLftv7 za}9>2R!(OY5#(oc!5O8U*SI~jooxc_Yp(aUMi(I!e$S@{3426*(;VP3<{X(@KYLTR zcZh%$6Nn`i(9N$q=^R2{b6Wfj@IDB2_u3FkzbxoSiZq&KT(#J9CRI*BF>TK`x)AP4phHUd{$7^4^A}ul3F(`k=u!9+A%MZ2ILk zCVxT~atrs_VmpIz`|I~ytvwH%HBleHk{hecgf9>ugVBO`D)j?&`!qm(6S$AOKS~06 z9Ad{ZzxDDg`!$_kzPyn7+@HKY12xikjUK@SwB$q(fcYU6^oPh;YUFTFEw=Hdiy3#N zRQkxBZzcY@JJ$3!qO8M&edR>tL@5t;d5D^{mNL7kTr-$k!jSa}9b65^zPh={zoqns_c z4CGwSJV&p=aydsmEVb4r5P|&~Vm+js?mq$b^NVt(&Zj>j>sVxqU4wyjEdAeL=2fg1aj8E`39g50(L+;)4221 z$Yrj1#vAio=yKW1+FD@ia|V&noll>-xg*eae&_Wb33DL&V_WE#(Oo@s z_Y_mvZ;GXFtqt_|h&SZ5#EcKnjr%&Dpc`w9At(6kXFPN^TJCcmzx5#zM<%2*Me=eD zd(+qA>t8H;Pv`-?AwK}0|1nt3G%^k@;FGH*CxIH>JY$|+lgfETV$K}#tLv9*8()K+ z7rK1L%`?B>N%wRC-8uEE*TB3QJ0U%TO?B$!ceSycoAX-hY{uot_ZGd6Zl19kIUkmI zOog+DKl81X^K{10#hgLiGrkNWG4q`(#4>+PJagdB&l|A;x_fBI8KHX?@>uU#>r;>T z!m#w;MCQB*5B@%OLh5sF>+D+t>+}ymvj{kfKhHsIgg@a=_{G>9{CN?xr!(qnBDLJh zGop(GLIOzn^BJKpdlSi=xhCNU<}39jU~RUQ%Z5Ymoc zUEgwM_hU`AyP5_&>T`ais3US%a`RrO>tnb0t!u_i+C^&Jl}+DQ;9`%do;*V z$1~~jL${3OoK$M^9Lyv~&s=0)$TpI@C_S9hy^n!=U2<6Co)Y<@1r1o9K|psd>jF{^ z_Y(2Da&1JxR|EIA4ZPph`19~3r1n*?1rqrQpU8c%ea6jc(C7H&3Lwvde*kJ?5wyU( zmUw>8K|e?QGEnyLZ+A$ETztmnU<4+M^4)0P#j+p$`_{^LHTUIf)tGzRB<9)G^l%xE z5e3{TYu}Wdb!PUR_C(6b2w8C+-K~do;rRaxo(S`245de|E+&c&K+wZ zd>S}UEMb2Ze*cYsf*tW60?yhJyB+@+S;Jwze4bNF+<8O~xCeRV*u=gHYM|}>5#)t0 z^Km|MiWayA#8NEqd5`F>KK?qDJm(5P=^6Lo%^)X;2w28*PDBFv6281*vFn6HAHWgR z$?2iXtL{AdjW-dP16uTGtc%N$eZ*>f<|TYVBw|2cw&AkYxaVkmg6;X}OW5iydXAYR zv=@8%8+|NmguPp&2w-f!I`vCF?PXm)lh{39x28_`fVtLb2PHM`Yxz9XvDE69!x_{E$d;O1 zN2#2dmzeWJV&=+eUZahF1e`%xD;Fj1UOf|GO^7@18EHtpz&9agZ}%t6cP?jgpYryA z{q326x;dNp2OXTP54ONra2d4Zx%cs~{J#_a?$W>2v<AVBf{^|4t|a`SxnTQSz=J zFMzxjkUt^3J68aK>nt)vsbLDz+tVyXrzjc`hU*8&} ztByN+TlVn`<#VQnxXXS`Dl+%Oo)ecRqA%lk@;q4@k7cG2oVhL`(W4URpBQrui>R?v zD(BZIv71N?WJzAqX0;JFb|2(=WW;CO^S4heI1lsUBILgMTQI;@Pr!3=o*av1zxqT& zT)#Y?%K&($ftda#mgmbesk@kS2V(ZM-sQZx7M?Njl)8Q5SGQM-RFBG7cq3B1QKm9? zjdb4SY9kuGdY})?74Fg3!WB@1h|kxeR`JBJ};?Cb@pT|EU=jFinlt9jQ zmop-76TK$K&pq?*1QTp?#>9^U^+?^l#Kyqa&1G_|Jw*IY;J*o&a|!=G;{IE%?{&`P z8MPhMJ-Y~;Ie;dT$9+)t7wE{8$U$$BBV@#9zWZ_C<`|!Y2IPElTl_U~dq-gZ04C%J zx$-?HW^Z%l%HV#+E_YxadPB^6VeSY-ZDP)8j`{jCmwUk0?@Fm~80&)?v=K?45jpaV zJIIrW*=quv-(2mQnD>4TY7j*i+ZvYs2{{J#GtqvbHI0jn7Ma=9%!%wF`t*00PnZ6xDTIEIejj+(4EqQ)=xWe~{q`S5%P#1gpN$&iSX>$x^y6L6hx#e#d6PDWRE;+m#asP{7pg^Lxo{Q{fy>;`)kgFY`UE&jjuAG$0ZL~{$OvkXQrGhh|DS@yX@7748;ailyJawJs08d#LgI+I$`u zQKRp6kW%5eg0Ce<`vN@n4cKS#N09q7m;0>+K6{Lj`s8w6KWA6D#Ww)~-vrp(v&ysL4$%uDdgdny8W}Cd6%nAdkD5MkS=*=1xaTzi?$24q zz}e2@lP{5zefrS@K5LxSXD9%35_Zl<)&gHm%zZ_0X;_{k_m;3*I7P0^0x5@xUBO`Z@yUOx8^di zM}lKQtO4$=#Wp{Ivp{#EeIL&ebzc%+;pCrR_6PM?#>+mP2V@1mPIk3KujQHwAKC+K5067A7NTs)2dBxJpS)SoojJ> z-f}p5oeEbXUW2|dl#_dkk@i=0bi~qcX|FbJ~<*t==xivd5wrguNNiHxbx+i;ZZl&UJ0ol z(2e(rsR#0eaqR>e^cLA7BXVL6QfS-HJ!PQVXN+x3-MktkAjbq>?BX?c#I}E)kDQ3F zhwYMMuAacK)HnFd?O|KD$C?JROMi3IiTJTfhd56Y?cQ?_!ZDWFBJqGeAhz^5^R+z?Ye2nAw&>>6DU@dhTVDjrye1Vp z&w$^vy*B4!XMb{J9`o_{fp)8HjC-J8Tkh=R43wd><;{D;m(Lr2K(|(ZgWlkC?+Iy4 zJAxKnnd`_^x5ibc4zbkA7tlomY7l`Off>vnP^&H6i@eUK9`#|%-$v}t=lejmku^U1 z$TJ6lnh`$vtZR@LI$J%jSmrU$T0fWMlf$|iUEBTJGcW83x}Rm)haFPU%?;F!K|qhl zh}18q7%zyqBE0DK7BQ7)C!ZB|!ZzO?*2^Pr)9ImGec}Nm{5elO;S=Q7$ZW^*nzr7) zE^8;`*vCBQXpnPl&SP1tG%v>TSrM^obaUk!5pxFp`STZh6L@YdKK=P#A*~A_*Qlm4 zpR)vX^Hbc>Bf7ChADqU`t0NNInV0(v`0Qg%Oy$1U$cWG7JahESrJaheL0X$vD(BcD zYq*y&^)U$g1?Qt4@g-z$xRaYAagaHY5lo0{r%vx{vj)`r{d@rOa~4aEJz8K7^*R=* zoq*?Tey%f)9+7QC5|h_n`92aG!>=zGLvP8o#SZO&!z+3>kph~qqsB!g>m4W2J&Mm^~!;yK4mlK0R1jw5hB+w9}IX0kb!!LeM! zKrj*bxoX6z1Pg(e-p_Wg%_O-V@7YAinMbGEayXZ3X?B=NTHmM0uOr>rk9>qQ!Wn{t zzwPdEtq_89imlEWk_-zWjlgw62t1>OYZ3T$=0y3MNb+Z1ws|%a`Ku;#j1q=(VKJ&@yzk-f#diL^eI%A!1>8`$>+l7$TMpM-V1-8WXbioKF^?~ z*gjiilVm;0F3sc*^0R@@fT1VwOuPqYiqjDsjuUT364@3zZ=rmiQAasm1m1rd+02=_ zHpRse_jemRn3d7ozDIfj>}+2-UvI)aI8M$Y28)_F*BUkl+Yf#=}!H4ykYaZZ)vw*Mz^ zI{A3t$pk(}Gf5qRdvX6~cEpi15T8xpJIi-fC63R9&zR54x>Vx*Y_VKdr`c+mNm>ZJ zFFnPY7_?um$1`z`g*XqIi))6F&CizaB**cb49?RNr>VBNgGlFPeO8*Swn~!w=?Es( z7SH#JV>RM#Ph>+`ewlZj^^&%*g! zpKB+P0n}cl&#qt^Po=jw$D6Z*oTR*-Rd=}a4Bc9{T6t7WSw%z9XJfF&S z?7T{LR0`STGjJT~?AMX*OghiVejR~hxu%7n5mYiMCd~B<%(;R%Z7ScJ# zIs@svxLzF9un=^r-DlT>NLuqrat=QORihcn)+p9mk0jS%q!Di-o1Z_=%je0lyeF=s zC*7=349&y2XUOJxtTU73+M9?Ap?J>ay~Pn6{B>m5YMTEiALnwcg(TPKp4RiD`az_d zi8GVWnPQyD<{0k5oQ^>}OFc=Rlk>SgOMd>`pW}HyX{4u-pYz}eh)rFGK4sbYU{<{b8;`fi)P}INar}7iRll!pEG5q@AI1_W^Q%ToIW|QQ)?Bh5c#ajs6=PTJJ ziZK(;kgodM?q?Y3T}XF6ZkvO3{=QbDJbukTBfpt)crV$cI}_qaw~*vHxkm`;{2GlT z$+HBJ)RW{n8gebl=NcTFL_XFzm(M+lq?veLn@HyxJTK?7bP)T=AvDGuP zO|UNFOqwl*d-EQ6|9l_KBsquYu~3YTz;$vY*Wf&NpkxDVIU2prFPat=?nW(U`1o#$t1Vm-xnHac3I$GudNT#I|^2|QN_!J^rG z%rjK7bt-{;49+tVRPyDVDc9p3oNFN;FP@EUGlA=J4AsreKa9ZdHD-!ogb~lL3q67N zWFb!HXL~mM>^O(>*{36@EH&F4>@%rk5_mrL^IW`7XO1P_Oklr`plY`IIW>s2-9`2> zG+UlVlIw)nYXngre%ucr^FBB}j-ZmwH8+u;XVsCe5e(!rsrIuWn-}*nX`~RG+0W-qI!mt4 zd0f-X9D!@IkDmiCuE{>0i)9koainuyo-2%X;`Ai>`I|`c8St6s^-+vQ z9QU!CJ;aSs0u!12xm>ny}`j7oeqah#)(ts`9{ zUB@6F^SlSHuaeFPB3<(%osctQ80lt$bBe9LNwxJ0A)CQ=I!Ug>x%_N*n+jW|9t z14(DQ&d-!*;H9ecAJWZ)Faq0g>|@)HdJs5|=S{XtuE#!&IPRknbYv%`*!Ir(X9%3j z_tj(v*U)h+forg>lCEbzJp&!v1kUH*L(HU`2+pLN2pZ`Il5BGv&&V9xESa-PRcD>Y z;AguJDoL)-Ht#*BC)>`1Bm(=@6w0Su2KUM7#rrU6w)57JkNd=tpZ8!Q+f3km?#*#3 z+vMZ9*sl>R1f6Ez6YB(yWll%+*;Yws1d*+hZ6TfaXd>OrI)NqMapu@&sj(#JNVlFN zN#<07j$l^pwOHD9>+Cc$OKTj@pxJblV$7s#Bzb8X|A)Xayi|J*sb{m z^JbD^DYkp7lFmM^!F5d>OOoT_NN35sP9vV?;;9tuulM z+>fCng!$R(aX#xj^Cmk&Nanl-bn1W3muKTTX#{>xGZXlHESl|E8~kkh$2lps_W;i4KGu85v7|G{u{@KBZ05`a3qjLtdJy?EvYkmXud_oXnFFrL zz4a;fXQln8KXbfH8vTd;ED5|vYdq;J85YvnA4WR&G1xgXaTc-}u~ zG>S13I98|GB>Qr>9G&AByf+g`UVNAMY`MRVB(wVcK2}#c5x=Ey;A;~cu%W-_yxd%TB3vqgX+a2S* zg%Hp8mg7tW{{4tQgEC)*^lau+bk+&_$LuB#H_$mV`roA0=lCut%d z=jljKBFS-jf|+;wNjleLXsYcwP?l{^WShy)vjvgld`3<^m2}>Vfh6baX#Oyk z{`T)qN#)lt&&F}i?4!l!qOq>o_E&}Cc@G8x>v1I6XI5>q>PYe&Z1Xd*`pM?!$7j#` z(a6Vfym&^A=UKQn_sx+e)fUUS2D^`Q`T2*C&AA-IGnzEp-ubN9mqxxg0?(==aD6lR zOr&e}7(Nr7T~8<7!QXx#R9g(!;#ls@Gw4ZL$j37>ZzgHI=fv^iUhL<&_;+Nk&2t11 z&pG_et+hyUJzgsLcz&H~``(3T<>z1~+j>_?T4yI|Jwx*I9?S%OPCntg@y~L5^Ivme?;(B~eIe9gxuAX8- zh;t^HO-Le`$uCYcwc6+xexo;wy4yH zB=1Wl-NZWiIge-4k>t8ugE`JqN$0)sjGW)nUuV4+{LD16<4BrG=Jd?D3;N^a_p`-t z9e#!!rz0$-9P4=!Zy}y%;^)pYYQ(7&qa&SbhLJRp&76sB9f9p+;;pl@P0xns|WKtwS-)aqj=TV1kN?9w!S8^4R%|{e&Th0_8O$;_&A>P`E{Ttzsc@n z$un{;*VU1o1C2QAJ+sek%_l#{TXo_&k84`@NYX^MnP9C+ZniCUTO-N+a?VR*-JWB8 zhCByxR-S6}>?+y3xCZBOUCv>fp9$~Pq}qCtWjz<>tue$0aSSc9YCB6+v-zwX>CE%F zb8XhGxg<@R&2}cAwI|s@e#DW@F}%{qW|+u!J#KqW$z=2U3Ez)2;`yGMHQV_ZhzlWb z9gR4CAJNFqoCD?X`!3h!9G=^((mn|~yEKtBQ4G(MPS~E zI6c9v+J4@ArFgSFUQe-n2K;=N67Njly{Ht+XTW>m_trDS@w_VK@od~jM=|^?IFC6# zCky#G#zZ~~gLI8SI?t;k-3k)cMS1Kq6L>k3Z6K(cjbq-#pj@7lYXyvu}v{tkLOa!wt{S4s%CozteaJv zZao9iIgal(*JqvU^BJ+u9G?ku7P2)v$LGSfiFmHf=gzS@;&ODJne8x^6wixuG|hJY zA;fbm&%rah`rF0FELGo*7o*W>&U)=64OTHk*tj?XZexFnJ~vQ?61l3bVTs{|c!T$j0Y(z!P; z6Y19ZRjX->C6(t)qZn%)(w#{+kTg*o&%%A!&s-S6!QUp?XCU60IG$6bIDSTYvMr=@ zUXBbRj?bEXTwf!pv7dCFk>}^nn}Ii6i3$&1gJd-C1X5x5T5G!thbm9a7-NeEyOYJOuCsQ-vREeqZrOpN$SaFj&<&#QXJRg zc&kpFfqWLStvO_y?6yiebL=ydYc?LdT6X|@8{4-0X95d<8#ItT7J%qr|Ac!Q}7OJ5VOyp-@8u^k) z^4aJ}T7Bf_S)9q%6Zl!gk>uwTM7oZAJOjtDk84_E*+<~o);e@=HPUnLFmpUV*D$A0 zEcpzis|42B&v(y6lJ{h`^LmojIO17%Ca5X)S#-A8AhLDj=e?Q;8tHt;xDU_9b8?Lk z0`H0U5=O94oJJh`IgSxR2(s(UJNVn)#|*^PI8JQ{9K*j`R3T(jOis-+#9L>ly4)+7 z5Js?&ts@_wn?|tSMdI0SW*>ojb4(b4b=A-IIx>-DooD9yK?JjAI}1Kf?!)zYe{tl? zCh*>wcisS!Ar=i)p*lQ@Ezc%IYRi)@}FguwNph)*J& zYXlL@#BmP`<#0`vbj}ST=m<#!?!z&h&ogjd&KxTF!pPPUA7|%vB)KQwU-t9-+=pQy zTSw4{3n7>YDuL^>&!m!%z_BK>xex2s9;D~^bLw-fj=;RC*JbPTeI(#B>OESt+AA=k*+dSo34>GPz=w+&yRHm=b6dJv3x(dhP4Or`V`U$R*rby zC(p&m*|(W^uB{_*Z4-fK3n2vAB{A08);QugpJRAdO|#kjZ21myJ%%X zxCZC&V#%>8f%ADjvudA%@;H`f;Cz)N_q4{7ZPEx7!|}}VV$MR~c;0&$fqgtL`^`NzYklIFR|#C7c^$>DZ6cW2W~jD23rUr2is4$!TcL5Upn&|2 zCC-gE>&_5M2nxvKK)(k$lJ&o=61qmZ-Xr^l4jS9vvtRGgy*-5xP>cpcbARZ%{kMyn zSZ9b?xwJDi)-nEfGvNO}gD`s04Br3U41A`127?E$-uB=2R~i`qpY|8x|7QPu-uWHN zXEv?)-|nAo!T-Dc!%)xLyZ(ROKj)(J8Qe`hy`OEy|LzWY|F3rthWbW>XzpL{|NIX2 zY*bq2@qdNL4$DwSEEXwG9Z+@S3}h+AWb4L9ky>RsmX&>rhLaZH*63;QjCaS<$SJrw ze=4?Kd4`)eGV!}Y5`sQy2wCEfI;Vyp@#PCltNakp3MHdu#9Zm^^|xH}qOL48siJK6 zvY-sOkV_UT@EjxdRFET$t>lyb?$Fo$sI1PMF8)eukAK%pL6?o45Sp(k&K{j2E?;Vc zmi-!|*434wPt(3|xbBUW$zDj9?Sb&kp7Lf!0a;_nSA-mEC0Bfxaw7a>t^JQhqmpmM zkL3%3X_l%5@3B^0L02taH1DEV%s!;x9@$xL$xP-z!jNZvBq$r?w+c&`S7( zoHI8@T$%J6`lj!2p`n!X``46>rsk2g&$qyymq&&7 zR$n}-&_Ygb<}Z^>t>r%#$H^7l+sG4vLe}k+jH?4%AuzKuqP!wR`IkoF`~9mpmT#c^ z)wrfyb?`HKL>-fpzD|~-@7I;J_Xo@3i!$WNd&PVrdv=tk_76k;rdO5I9z~^oe^=ST zy|-*PxuX0XYCz4j>B_gt*HI=qw=DB~g6wm8A5KO@;N|q`Kqj_|moGQf zkgeM7L*rRbaIt(@+4V$Ax!9qkT$eweGza&RJ|`Q=;jMei292exQ}zVLT4sq2NS2S+ z=JN^9SH!2%9F8!yY`J0kbqT_XGT2$A<)%F1qex1i(E<>J%08lrZk zx}xp*?Q-4Ju5v=b{it8ymFWF_rWju5i-`Ji3S;I+;nJmelz%o)hTTh&Qy*NE>n7ip zcXyZcNu2NEGo#9U*{c0wnLJ^Z^qO%UMJ|WP2HTg)sYN1X%y=)^Bk>8kggMJ=*RSH< z>bs~lv7-#kJzU-$v{X)bHdA(NyHOT67AGqf_ruob&y_}JKGnZ^EL#4lKT7J)&6ZPE z7L=JaMqpBMcWm+9E**a^mZDe{85rJK`aipae8G-l`xQqpzw%0`Tg%BG5ued8`zZdMu?ZJCbwGM%adg^yK=!<_Pv&p5 zTn4v4f|HLrz@YaP^{*9>`P~lSr~4EX9h!+odNcCfnw479ZXkbjt0sG1bCGe=!{8mhSIoGeDWUDoiO#pzD#u$c7auMTMDfaT80CKoo>v`Y z7oRKeT0I*1GmoH9Z~_$Mht|;n?H<=wu0#|xCL4upA6HTCeOg4$^*$^X+`KB%+)_kQ zwUL}(-zYObl#`QhE)X$Y&M1{GMJuzr>E+CmmF3B}N^ zx9d(onQzNcdP_9ot8T^N-kor6suy}Jb;J9cJyCf>5rjNnBZB@~hzC=O;jKdjbi9*W zzU;FBdzaL~!C!hAUZI#=H~kJo0afmM;vru&`-T30CnERnSE5wO%i`IJI&j~;RY{o^ zYCKo2iF}))$l@nlWI3mYGBI+Q{58f`R@!tM^Y6x^)X2_g6Z%P1-hW= zkMJCHL%0l>f%l7-!KK}DIE~n#gl`z6ynU`HM#l^==6<$L+_~^V{B^51 zGS~Xa*aPKdeq&MD_1kTXueKQ@B3t6ax{Kmm{J)}Xty?13_K#xP#tg+}{By;zQGHw) zATa2wAD;g6FP6M{3caVZEKq5Ja=H9WWo)#U2<@AOs<;2a>c0zMQlaxo!*`2>a+2&lp>x#Yr&=Www}58U171>yEi#P9P&u0!v|{=h7e|Nbf^^T$^u z-;+B^%{v8@%sJPUy1i(hZ_43#-aq1S<=jXJa)ddvt8hDcRJl5)hOD90m8VLVl&0Ie zF~64w{!F+o-mDlVXND;z!fj+kU0Dd-*!vuktpOpH4KBekN6F^EV;% zb}UYx>WtBfGdg&Nip8VHi_YQI#hU~B#pfwcM11?wDC-y_-ZktcnznU7*OTwXW34#; znG}P}u=}V`v$%Ax<0!-XJ;aG$_0ZKrC)}O0M7IvnNG>%+%&i`Pm6w-dMv*GwU7K9O z_hdmNChQZIS+B%Ib91!odJWlerDdUB&+xl(GB!?qj3RBb@OQq!nEzoq>W!F*cSW6$ zI{z}B6ssolFPJHBAHR>+<=Tr!iM8=*_DfW|o=3i4wgSt>Pr|dW!8rDHI!<=-g-k6k z>MxB&dYU^1)heqD>RwZ<{&C8?%d_gTqt1TI6(4k@D8_SvF;N)e^#jNW3!g%2WL*A)|fo5j=#CB%bTV}wpr z!b=RrfHNi(%r!ucUpz;C>p54RzHwG;88HH7hkbzVbyN9xikmbxeTeNVF5*c{L1}4_ zTdE%%a4I1eS}f@;BG;zi=+(S(%=Oap;-m|x-Nr>WYFt}ho^%}V)4b&JfJ2Y9@7x9)j`hBV_-v2uf-w$M9%)|icI;fLuTk|Pq+}(&NJ<_o7;SBtB zMUVJ^4LFzo8$#r4-0AKsdyWW|9?vVwdIh^kENdxOH(rCRQLAu!n+boV{Ed5qdtqO2 zAYKHP5aTw?7I~(}h`<+1gywuz)QmB}?O6}S|3pWTdqM|zu%>+a zx~nX))JG0}(n47qbyq1IYY_DdF2dy^?{IHt9+~Q$EJJ=Qm#Js;vih8*;%U!O;^n2{ zLbtXk&UJ2y@u3}IDx8Xd-R1GvX^!HOYZ-QJOcy(J-IS))3|Md4FHWfSl-z-TOQl6c zSt=_9KP)%Joi%Hb>)9*ZTVEU#7rLNmiOVAIt;;@KAtRW+6 zwUhg5f5MU@B~eWOO)1>JveNn6edE#_3q-9#wQ+0bP^@WlLTCr)CSJVR~2GYrh9m#_U9%&vV452aZUrF&GcO^}xVK%SA@oQ*X~D zwZyVx1(imn+8H<*G!b;ECt{+e!DCQU>>1)Bw`9$d#l#)?Wz<~U*nS?j zMts7dcZbmGWF^_^R&zNu^(}7Qvfx^Wwa7A+L6CcH%=Ek?nr~i)ckRAo_p$~uJv|JG zUUyJVXOLCOwUe8KG3Cy7 z>^!$n&iQGQM<*1Or*wZ|WQFn=eEo;;9PEsa*P`(G<|T+xWu)IQqm0=23_Uyc5lhax zqT+!S=-=#Lv$m3|*JphoT-|E53D` zfx&;xN3lU+@NBmd?Ou;Zp*+jQ$YM*y^<3rAw#z7d6I)OwvONrmc~Pu|Kc-jOj0PD+ zR z_{*Y4NQ2E{!P9=iXi31=;s2m*i6)qQ?yG2a-VtwOBSec?ruqHj*s?O3b`w~#B@k~< zxZ_ye#==s3pb!h6p?I0I6yMi4dGch*M3!~@U z7Y}Z(!LL^r@Y1uYEU>~6%^z09!!wOByvI|q)!0sqs!&LAy+248i~bRr4I9Aqs}C}p z&>i{XiGf3Fh|oAG#_x_6Pd(>|XRh(0N!AA8<5L-tqg9k%S_+lv^Vp+8`QdW^q439b zF|mF*G~QhTo=yp<|7rYQ+ugMMic)iba%d1jyqVV5b2J@mS zVRGMbBH2|Ze)ig@luGc#uz?B+H21&}kDVy>*K{BVh&($(yeacdOb8!{5qCVXXW}mr zz5lM*oOdWDJ&HuU(|CN>r5Z>58*2<$^I7rfIUmU_79!woAibs=q17)hRKDndai6C{ z*)tqlv*~xjhYm0d%oGo*j1z0TqlNR4-6HSYonoWsaGdxu0hR{+P&1$lj+Ays>cDK# zHMTa2T0V*THH>1y;KQQE^ZfYWToQ$6okiLGAMmx-4diWI0y8q3AaU_O7ywEZ^W5rgT&ML zj^b?7#YivJ246l!3d?vQKBhDlm)k{%dgmu2qC+XHeq2CQ9li<$+P1>v3blmOg!W$d zGh2Fj2fz0EF(w?db49^WB^nj$oxt*s=b$s}M(LMdlmp{hC`V6~H6|Zhh_uej5K(tO zBL3{bzP4KsNU!(3KRm>t5r>r2;Ks^nH6EinM&aJ3u2@xU7@9YUM`7q#x3n2clS?-@@>)Eb{6zMeFE& z;^MwYad=*cGBaa6cI{k*s~!46xe{b7_vEB<`RG~Y#}WFu6SxZ>?`*`t-acZb(_E3Y z=cE|@*-6F*RhH|0YRaNRS1FEt^J7|#_BioA5W^RYhO9FMokvQvIj>-Shi1risy5m_ zsf}x|8bVift{6~agg9Jduh_CVH-;1{gI3-x(6(-4wA<4N`8~ggNlTnirJ4~NZ_U7% zmlu#4@)q|tzeMM@DI(vf9m0RXP4PT@oCx1O!I-u8k#~;=xuh{qSs5_@9V&fjEqfjD zlZ`f)mbI&yG0S-aUbWwWU~iLueA#2(Lk;_1EHBCmNZ{0_vTN9}QFy(AC%KP-YpMXJ$jy%1^ztr8F8 zMvHs*d%^8>7d%|k2endn;L4EosGchVOUqP(Pry%6WkEieztLyj@l~;hJ|i9IT}6y) z)KuY_?;t6$0kmww9k19=pmh^1ImI2N~-MC03pe{jub1eQck z!R{KtA~N%XavEiYw^3qa8%H#By&+D%9E6QuhGO&nad@_-2&SL8Ccf=mA{uoG5Y2A< zs~kIATl^a^0P~!?qkoe=C|o2Tru8d;a-}Naz~f~&^n42rW*$WGKMT<{av!D-OvW5T zOBBl82Ir@Yz}t4N*jK#(evfcP<+Q!Zy$frN4V;8=<$5>NiMlBky}T@p2MUPKam7S~ z?;*mqY9Um(pC65fJ`&v;jmOx6F*r450|wtbf#f}@*ngky`@-2+RfV{oi_m9D2i!>; zgmOv4P$8`Vs@JseD@PkpW3WG7#`J{$g<_bqCpUZ#mcY~AQHaQ1fQgOg!o5Ok zk$BWaY+LkQnX0;@#f)B}{IXRda%>T7?D1G!sQpoFIW-^oe#ha^_ejhR_#~Fj`zpp3 zsEEaRjv{vA0bI?$8Lv|u<)JZ;Q8F|eA>RV!R+mb0&AFE-9qy~B@`v%=7w_e1zNZU!vs48lv6v>S9@y3z#^1Ira`Jj56C#izWg6#Mz+H!XfmQhzQ>*9@pxJ z6W`O3WW0?PWm1qgkBeM5a6Ot0k4Esta28L#E2m&dV5?XfAaBeK%# z$P(W>Nck*625kw2%Z}~BH}qeT{2(7%ExIdS^*tj#_Kw3pgI|hPi>iym`Z@SAek5kz z2|?h{QgYaFDNV)e%TtlB@F#W|KK6Me45KQ^KciEzw|xote_13xe}18ijk%$89{d&S z*Nwr5>Xq-+|pdrLvN zF!C54R9zzWe3>mOtG0H<(V$v5*uYVw&hBMQ&1{4&Wxb{M)*~1-AXU7qQ$)U<}$%8*i$u5I((RZrR^3C%~(rv~!obt#k z^1W#!(!&a(#rAmdt-xpDJM|K7T`etFU$>w|X)_w{>nVTE(o5F?JJ9#UQ3!Pz7SxDF zfmI)|DC#u29^a3Y?xPT2?FOoi_=;G!4B@cuxp*D!jXM25$QW@&4sRSGBiCQXgrLEw zThI}c1M|o_$^GU0PHW|A^AdRXUoVn3oKO-zW{T$?lksgwGTPl6FTM@fr@X6P1L=n% z#k}}d%FObGP`_gwTwf(1E_o1MRX8W2&a}t*nI|ySwYy9_+F14sH_E<~c8Sa<$3%MZ zH)6=uNpk4kk<#O_ubkW^1b&M)qd5Hz+H`w2DHlY@yirZ$#PyYA#-0PH9O;Lod0KSTFmW$0$4kJ^l)!(L#>+hN#U zZl@TvrHvG7MHySuS0?PdSAJ8!z#JitAm`>#vuQ_J^-&{O!#{FFb*3w;X&6#tST5IoR&`zeO~28?$!%2 z^50*0Jc3@U3-!|N;8P`{NxX4Q+*jqz;yRcV9*BUc5twizpM0>ep^OVDEsMVoMf#V+ zB4Jl~ad+Nt40~BjPB4^^_ckm>>b7y{(x4LdReO$%#VVHc>7~5?=Q_HSD<#Lc>15~6 z+wpYNPRuS?AK_s)Fh1ijG8~*SKUXqlf8L1~Cx42T2OlV79S0h{o<8$(pLhm^{*6J? z7&qL_y&hY0g<<=Cfv%IYg{j~H;nKIS_|knTdJi~{wb!zdc6gKg6W0wkV%# zhG_3rS}C~8ENZmQj~&4ZaPFk2aPovOofv@4rH^6U!gJ`eJ{0{+xL{^@b20tlK_%R6mU1HM`bKz^ag=5h zl7XE8-@INZHTRAf`gREld(A}VodiTr3qVZlD)I4Pkk^SHW6^HmVH`c~EDIW{$kcT| z(Yk6VW;JPx=SN22mwqaKOdc#IR%#%&osAIPA9q0OvSm@UvAbxp$PZf%IOAHi`XY78 zU$`3OjzY(W;z!LDFb(a8m7TwdnZ3(nT(Mwr?0Z>dWqLhK*0ROKy(dL^-|r~v^c$@@ z96;=?HTaov6SV^CBWuStti1gQ^%oSDD|$xD;H``0hHnex+asQ`W77_Dwqvl|vf2b+ zmq?th{#tYz)mHAG7%2T`s&eZsUm2C)EpIhzCDeb3H{_B1 zGrG##W7^8=d5m)U!XBuxBODhhMdHx<0Wy2rG+ChGEIDr3GhFPFM@G*pB+Xh2>E`wb zZ#Pax#xxbr9=F7itetpuvv`=oCFMS>Et*NRk ztvkXPox7Bxe^*XMxRsV`66(tJBO~R?*U_@b{Sk86wir=*;|j4lrn7KpEo9`}hVqFTNbM7Czn`M9K85P%pUq!Jjq^!e4937ubyvA z-VaSKb;7C+dO6}wC0T#$E9|=Xm#i~vf-Jjpw7k_YkKCK;BqyjfWN_=n=yEO|G4)e% ze7S<=Gd)qf&nGc>SzdXtM*-RYyo)saP7rfl`-|>J=zIIswdI(Pjb&F~FIlzjddy2( zht=DT;OFA8(tY<#S^avX)YIRhRv!F_twW2-jcfGs(eE-cc+nG_8?sWYSzj0@F3d;t z-QVc-zKq=A?<8C1--`}!FT(TWC;D58uiV+Iopi2Zkn48d!>l^zFf3&X;Nvc1CmE%; z>sZ;dWQ45qC`Mj%T_kIr@|LN?%S-?ByRc|~8O*3P86^g$iow-_Wb(T+gx&rcae%b8Z0`#gq3Vy+6`Pp(u${%{4){diu;q@2gSxQTkf3O$FXXcgtrn$%} z-y-q7-5QjBmtVd(Hv;uqtU^AQ$M`bALHNEbso>Z}CC`zK(!;n`b~l^l@1S$oQg|bd zjjJlx&fhQcU+N`>RK6spyB9&d&x=L)+Myz2$6$0=+Z?YgJy9@e7ra(X$ETY|k;`#4 z78X7sO1*xn6dmdyr|pfxiHABApE6!-9}qA6ryN4&u{1m~ra<;AE=R?_#_LJ9(WKBb z`29}AqRdC4N5k9VM}sLa6fTI<<#MA}z$3B1%@r5x*21fgOK{o2B22XsL|o-USn{e6 z4n$N#$2L#GCR(i-Lfo~BxCZ8Ptx}f~@h(1q@d`Q1DH1sa7 zwL$u$B#7YRGH~W}gqM7QzQ$V!TwcZ43|_|99tJUg(K97DI7!^7Rtpua&F`s~8Kr$3a?9-BmwS0R{^@g1$6XX1@tdGrly z1*ffth1R@)SiYgYaNoRH=$lQ!Z1a5Nt$GjFoA1Geg&jMwF@kCit6w84axgOI*#7{+G*Ev7VUrOZ8kQ^@H*#kLu1 z#M3K{MdYS%p=gywlzW^Ke8h|&2_`Il*Am@392E5`pAlYD{KfU8>&DPw-pU&Poyz2Y z9*RTv{BhGc9F?mVmK%p}M_HvB76+Zj$(UB?cHTwl+5aN6-Ie5?lXYd%mN9bR%~djB z>P(q2I=2j8mtSUj{074JiEE}ABDnD^v1q{~@!;+!k*Dw}G4D^d_qka|yem6wHCC>Y zjMr1m`1x0B^tygcN!cN z-aDM7(zSIisT*mK19w%CvE8c6ZF3J|*OOkjUT2gT^mBn&d&Df-r4++-`o3;nZx0mT zwn{vB!QUIF(|2my6qkHXVpq^I3>dW>r|ad$;#GNtj4g?CZ;L`J*%FJVEK>@+idNRQ zPgHi!tBo3o-k7$-9krdtdpDQoz2EF@W;`)CTUlIry7<)drYPy#Uk;n}6uo>0A-2|h z(Pr#6(dnc~>z9%v?ET={#BP$|yNxe12KESs^*5 zaZx$;*Kou?8UpXrf1yhC%fhW>RfLVIg1r@@#G2RZj58a(6+^t=dl#+VRJjo##P>hf zMC8j7@QS!A?iPI_<~}(jdZwp}$xhAW;nm$`$R8=+rQ8>DE^QV)njaGPTjZBp*OZc1 z|IQ`1j!HtVURTlKPzpx9`=~gs7-6inHClPGXbviENWhmaD`0+q6E7p&POmwulvNhD5`y$qzi9Sx^>zengB9bw~VE7lb7J!n4>rC=z)E zd5U}$gWYt$vp-G___k1Ec7h8|{4*wdf!WMnl}TGnVKT;KcSagmnr$ryzu z2NOk!&6V-2@m@5aQCKt|?IHGFS*48KwHLYmI7r{qJjp?l5ho=(6b_;Zo+L2zv1eFJf!SciH~(bTL(Vu=9j+%vTlL2gZrZeXF8> zr+8FwDI`39mPgj;B`CDLm9nCllX4y5BG_}Y2v2-1N^De+;CVs3b*O@^krQyV|50(^ z+I)<^dv0zS0Xrb{$8E zC@br|A1Qks>mr9P>M4tqq<=>fRSNYI+(dGfspy!emrP#OUmjfSD<@nHkiPMsFrx7c z49ZmmRRa>CZ?PNwZ#rV>8&A}(9e~s8o5}c>)JMkhoQ4|_xRQ6YXE=oH1 zA=rJpVktXA^mqLsd=lp=5mP&fb)7DIfBJOBI9fkcdGxM`eBY~w%&I>|CM9pg$J+^b zdGHEqPN{%Pj}I#1u9NaPY`VPvZmleKc!q2-)r9Wh^G#hX3FF@2IH+lBL zAUUF$m#o;*h5misYFy608nRt}gW!-Wpj&+onX#+&rpoQqNeT=-)ahJU0x?e8o)KdBAyAkWw z&XhHZddi5Zj&kYAdoV8cLrlIeDC$>KKJM>^+fFmYgt4o{imU{YHs-N1a$_O*lpce7 zZ;y$gX{n+~&tRB$t`s4qzA8n=trer4reXbV7n$)^$}tH((sN}^xooYgoKp2TzFjFS zMcPtS-@M=0sAF^SDY3F#JfemiI@w)rKWC9ceniPd^m#)6q%KB0uZR+JH_~?u6R>t_ zURld81M~c58^45aP#U=35l{3n#t*+XiF3miVA8MG;>XUN%G&RRMReW>WmE1vV&7qB z)Nb1X@xPCv-=QlAIlKcq2F2q*kF_|ubhwDl)m)79J|{ftW{EW^kHl8}Su6`(Ab+?W zmsjdemp^BeQx?B&hJCvn`1u-COyJ@Dr^>{Rh)+x zZf*}?V)frel-Un|D*9a4#i}a{5!^RFJ)^tQ6B(z53e)dc*E;M(ZO;r5IBd0e@7Nq36K9FTg|;Byq1v)!r9@?GoqAaFCIj1lwSe!* z4~o7_X*jrlR|Uiyg`Z>x{kgJQT2!k z%Y!1hipW&oMY3HMgnO-@qTH?x`1t7-V$SE4Z5l2^=fQr;!>~)@&xi!f`JE}2_ci0( z)7#aBmf4r^wdEF(XF@%^eV$M14hP8DcOp@F^K86* z+aE=8w}i*$JE&LX3&c=oe5rp7)jC{&+qjCxf3sG3&H7o`sPk-w#|zt|r|*13mMJ1# zD>as@8`P1mvqp~%*pJ~vxbd)^OEwqHh` zGJkn*I2`L;s^>N(tL9u`sT71^b6z3Ev7{`!<_j!tH4vP%Md|BaPb3x}hz3Ji$~|YZ zad+!{d`d{hpwSg%bFb!d&?;9E+rOuB?amz~;aU=2T`VRuy^G1~C)T0GQr-VY(Rnyh z{r*v0c2-E)yHqHVd!P4tv`cBHL`!MWpsl`1rKm(?Z%Rfsm6nzyg$5!?DJj|-lG6Bn zet&@La_@bf_xqgJIhX%(dxICL-E{Yy58O@a5$rVzw=x@$ry$4AiA!;4NeiTYpGEEp zH>e5o+%$~^{EKY|g1blXKW8RzDXj=t1#0o6&Oi7xJcO>34%)O%qRnzWQqHdC<7FQ5 z+HOC-_Q?@GecfvA5&9qA4!OZrADGTuua4$2E2eQp14}+_gCjz$(viNt3a13`rtSH4 znz~{+mKwem^<+(@3lR%3BK0|nA4v(bcstTBokmikKiZ|GQ5rZHFB(rFt5cru+;kgT zCoYG^nIh7f97VrpREzKIKFh{^4&;Bo4d!ZdMq*5#2D1DxiT47ApU zpz1&{Idw#_uaS?~$C^a`+V}(C<|=8|sQik#9oo+hZFo!NQMx=f^e(K8CL(3AEXndR zs@W8RSDP3-{oKgM?jc(;c><=cE}@)!X-Zvl91A~r!)xtP#H%NgoW&%P4H`*)V+YYw zNh7jI=nEnNJ_k?8h=ez04$62|+bcuD;xqh>boq~8 z7W__JF-z>$C*w6k1g&K_UtThuXRdvYifTzrc<`F0X**!!TsJb(aHY#V3+S8oA=GuI zV8-AI7(Yxwz=%j}`&^6zJ~piKh9ujfTEWJpkLBC*WO?xUWQ-Yei-vh`M(EHc6y7__ zcHDo%UW{Hw-X~SK(IjKuPtS;F|ND&$f%A2(?!(uO`9l_hFV^63mHv92rMOw1G-p6N zm6q%x-!3iuS`-4U(DzJqHh?blw}p0*HLY7e2ytCGxE*B=y%mY@x=@RXp!IagLf}Pq znPKnBnM|wbDKiPHp?No-Q)%QI3j5nahfnxn^M9LA{V4=4@9(qB_ob}d#FA2fyHJ3H z6e0yr_rbf#l(@@}ELIp|{uL+wWSc8*71{G{cViSLekQe`Jo=lJMf;xWW51-}sTN8g zbcrQgQ|H2|vsb+8t^~cFE5be3WhCXUO;hA2Qefp|GP^WUxT~vC^C+H7?b9jv<^Wt2 zSUeA^k zt)ZOY|48D+UOTug^+)jggE(XMj7$tiQpuK8$Q^M=;8zDC_v?h;cn3|I82#`4)ZneyLk#FvKMgRTpNve z_>^e!5Av=0N(a@BqU_6QSWno*o=HcF`~I8DRQ$DI;bn@U7p6k$gDQ4-&V`@BS}Z+$ zfd-8piItx+v9G+HCasU4H7#vyZkj))IDNx^%eA?&V>E3%ya-(zThXHOk8W*VgcZvp zu=hkPiO=QJovHPd?;?fCpIYgZTolE+A3`k8Mu>biHa)*AR{7hH9jcY1jg2kv$~NG? z&YANCd$&=1)hn|2=7?!tm26u~EzMgy3$HJN6jR1h`Tmhu==~O5HG}xs@wWWx=u$@S z7E(}13vF%?A+BD~|0J7Hob5_0y(3A_=_8euY2jdM0Je?40d$t&Xzn9O1s*|5|7uqK zs)+4s&!cFwc(~nt3*}O6{+;z)OXX0HCJib^>?H#{v~KovGn-Q5$Fn^_mIjWoZKkIf30{1{Jw;~ zKMEkSXrx@(^~8NMaR1#wTnu%=(eyCbsRd)GLnw9)4Pdh09GLFim+W+M2w%FA@uT-t z`LlvA^h0|l{HA-O>fj?PT^vfcD~_Sz=Q~QcG8dXv8`1Pshv;D%9oN}MUp;1%(c&}~ z!xXW?`6v~foWK<2YN&eGDp3y?i!$Dg6?La7lB`oFE53J%+|^}xo!vKh1>c0GpCQw5 z{K4$9PO`6YGW?Z?u)CYZLREG(PUd@H{#i>*yd%X%NW2iwiFRUY4wn34`3W?}3Bu1p zbKbi)2i6__aQyIt#%}aOw&yThDx8ge8_eJ%`I~-y6T@dr7V#&?=xEdlRIhS`Wa&b* z4qAxv4@)5OpM$2j4EFS>vAF27pyN$)<74R)qVAl=n{p$Td8S9)p0S%5yBl!RcWo$s zb_H(+&qm&LQ)pK{BQu9a1Zi!@PyaPIJK7SCwW|1b%N&8bnrL>OF20``j1`xxxaigi zoHm&T-EKcjn4W>^wc*fspGodkB{cS`0!8NE7Buv8_#K^sn4u%#o;wJR0`oXaUYe`i z4}!;QY1oIE^4@^&*e3lBM;<8R`Aj)PU77{&j!QIGZ!jX?7tru_6KWeG_Ab3jF_Sco=du5)8aU~FwBkcN}c)_E@WP%0pcTr z$1sZy2lk;yl8x)%!unT**~*-d#|ja z;R&Af!|)c{@$L!xqNB;i&0j=0KOLdHG!}=o3$WHDnLU{=%?g4pve)xs>6l9})enA7 z8to-`vN{M;VebXyuwN8g^0| z^|P9B)U*sTO4m@N{{{p5e}~h@*I3b8i&GDeLGPI+ioIu`T+oT<56YrBh7W11yebBs z@kD9NWArgr=Hprocu}qhinU`9=_zmxJ8NiksUD0A`(xznTyfOgdtz;s*UYpkUR;0e zGh05TR(QV4Nqx!?+V`H5xu9R>-ymp~onz@wPGZJy}2eIbK2>fywh=P9oF*AP= zDbG1fjnZvo^hg~Ozv>{%%1$^>>tXZeEX)r|@ekpfSYhWC8ZluwZd$LU(^ZM|_V|4& z9q@*}%$SRERed~oUr*itmZ0!}A+DXgC;nr#o!zMEVLu8BG4^{o9IP84-jPnnxDpay zPC(H|6T02AlD(9p0+2ZQHMo7#yt}Jt)y`5etw*G-Et=hf%EwGpcMKt`N-RL9sU~6$#8)NycU_l z&3zHQ2z4M!{h4(1(Iq0AHj*e(LHq-6*bdr+%5n4Ho|a4sH+3L7We=mmDDk6ZOR3!J z2(4F2$FIXjvBYv0RQGez+`gQCSX`yCDeIASCqdW&<5A!dOdD?xfzIO}bSyD};(mwI z;EnP){<(_&Z1=@xqtlq|zniZY$#R!%#P3Y(p#FNH{J^kczA0H1!!8M2z!C?T?pcb% zsnKY@Z_e-DJHS7;*Wr<@8AhDQ!Sy6Fbo3vKmcMsMrrVBZx)pPkrCa&i{4sojjsx%d zqRFKmsqxmW#Jhei=Bd}dLSM(6%cj5Pe@rLY%?}!9XCmwzAq75=Tb?fDxcu?fQ-xof zc@1`&61Z~9k57w~u=BTVqxq2m{DgKVPhX_Qm;I_n;=e=a>HCsjcUH4IpP^>AT5~op z_1EGf+=So%-!bHztittce+4G-Jzh^8fvWL&*mOsq_dPGn4`##o>BhUJP~qv(-1!pWS=aTuieM5OdY~iru)G_ zA|2mL=kWa2Y}9`r#QpLH@UR2t&|f&8tQV!wir9SQkM76qSN_AoCsO>?e}lQQsurx4 z>qDeD3=Ny^W1rX&X-*%pux=OM@KlFue-J$K>x_G-pFolJ893Wj@`Ph<&6y`_;&a>{ALw`=jL+nDI@G0PMv4tZiDyj zzQzaLn#R+P^yO`C!}$Rnf46`d@IvVBK zx{&FdNlje_G<10qg&PZ;>3lah1x6ufjuuLNwXxvd1|-z_QD2k&baGuLtuz|V?<nt38vvFNOQ%}?`shz zcM69DP3i8%e55Z)z_yz{*e~BsKW*3HPwg2jIBAXoaVec1YDeB>w^1W7c>U)o@s-<) z@JZr3>b>N6=!R3+oBa@f)l2bh+e13C=Ny>?<!Ak8-f(_Zbvsm=2- zsd%UOk##7`zrT$V0+gU{It732EU|He397ywK=YInf}b{>=UjB=hW8!#qkWaMXOjes zBKO0;`Z8v>yuh}Ud&n&t1ZkT=xT!D}zw@+kc7+wjoz+8?!b_TT#2(8!gdJsX7|JSd zAvw1kKd<+vX)i2U(?U0KXuCG@Wwfzo&Q(0JOM?8r43xi2M$8cjKKNuOE(~kK^Z-NH zl`3Jl^#(e9DFpun_Bg;tc#S4ZJn)!H78i^Vk;QJ`ngO zWj&NXyGpD2o8v!iADkL@1R>gK*fvARSSxIX*PLjCX8eH6kUI9_MI-Cv&CGK8R&F!T zpO4H8;mda#BC2p8R*SW9Vd6CMt`owo3B&__V60ZxPu!|qriKVqwhjxbgH)w@9t!XrqrJe@+zc9Y3bU+T#5 zLm#6UBn-KL7QtV9dqK!GoC%}uiW4OLct4@YoW56dQ?I8S20ED_a$q2y3%RR3$P zu0^5Zi&UD^O6+4C4PCdG)M6Z|NU4ZImA{dbz%sts{|~G6f8+dB8UFg{MPwSN@vLGC ze$`hWOB=Ou$5{!s@jg&%5^$-kG02V@OeH(inTN|FwkvcnJnaNtd(J3)f4zWO_Vuzp zo;A!)*O0Pi-(*9a;^>S0S;!=u!QZ=&aQ;6vj2$)@-)ppSIl>gHPAKEe^Q|aN`9PA7 zBgmjCoPzvx_q# z)^938&*}n1>gM1@b0$l=7{b;zzhMyz-Lc);3&vB=BQvlK_XO^C$DaVWKWk!nL8hd1 zK?0k?uTfRPPwJ`uNJsaD(V4*z2HY>;c+U^h9i9@KJOttC6*{AuurK_~|zalBphSZN*w@tuz)i8U=pzjwPQU za^UmxPeW=CUB-xPFlboO>kQ#i|}8{TAnzQofMojSIvCX$VNBWe4r=`hRl zv|_5Od)Z!--84qun&wChB(a++9M73xVaX^seoer3$5$}a8_0Ki{-OHmQz&U`2D>uL z6?gWhV&>gcoXZM_*@#PMDSn6rLXI{2)(+~GIm(*Df%Kwv<5mWKz%UEO!S)$g+ z4YX?i9J-*rp6#fOU`H%ZvEZrF2&p!}^cEXb*U!R2;{zC#=!6_Q8Dwk6GUm9CjSad= z!Q0Q!^RtQ6JG@nF{_2yskC7F-uuL1y(#L6bfDUpen?UiFi@=M@qB6t7cEN<>wtuBP z*(4oD_OVKp9Xr;}BA>Ovvhppajd}`~v{05fz?*dx)Qk4VY@!UEXgcZ@MDse|vzhzX z(v2>`Z#OFxe@$}|9|=#P6a_E(v?`a<@05|D#u^l?jK#Qucu1^`rv>l!(}-JJ=s#B* zfj4nP!1rk|lnkLEYo3!*n*ye&8{>$>OnAmP!6NJqCB|vsr?NTb&OXH+uQ)BXv^q`v zV+oa>yH9K%>eYz%w@8~7Rx-K#na)NiwpOBBh zHJ$6;iyRGWe)H@+e#vJAk9Pcu!ICGik@?|Mq8iSR9t@F+vA{G>#o{gw!)?p3^q3S6 zJEg=&Z0r>Hp#q#M8qT9<%;W#r9z#{+ae-C!$E6tqct*4oZ0Lj zbvR#Hv5cSoHjL}_uHT0wlLzeQom!R$z6I<;Yy-|ODVlZjW^zFgoL^)Rz23{ z@xmFVoX~@x(xdnryT06hW)H#|)TuoEvbbYug4ow+25)^oo6Ej5;MW%kchQ9hkf=?= z?7QxK+f~8;8llEhUwubFRVK#n8;t~^Zz$h54{IN{<4<(~KctY%rCA`~E9gjjJsr8_ zkM;aNA3Mrds3P6X{b>33DSX5g5pP%fhXz|2F88engG9@b+}8}_WOrc4wZqu)Rq$M= zCZIJ;1gBe5`Iytn+;fL7#yY=dQ$CsF+nnR*l-j_(FSTO8xH%vZ=h>oaYA&^=Phb)&Y|yiK3b_RK z5!t`r$kZamEPZo^_>#RA*ZXY5Q@%ey?s{F+>4rk3^(NfrZD7K!Ouj0+Xk+|HK33^6 zyeCYCd(pkJ*~PCN_(PBC)t$3hxANN z5BL{J>8(e^MrYUa*@==oXiuEb$>kurzBZbjy!e1QHuqwm;LnC0Q|DF-jM#xon@HQp z3VTl_@(Ih2@iM2yTx-ZFc0%DMvHG8sGUF0n%F6Pbce32Wbp}6hTboC_9>CvK+o^c? z96b8(47QGtq-Cn($ffEf{m}@+`QrvWXQ>7MFz_cOKU+f5ZFbZ!%#Px;X3(=C9O-Wz zvFcYCl1DY++VxUQpQp|@Z+wZ3`kido+j>foK1up=ro3_OQ}|>@A?fH69{XVvSJHFm z+7_3@N~z`aEm4@clf9tVPnDmn=*HS5ci8TI72>psc64{96Ro^Dm?3%*@2&_i>n; z@|n^$OvJBWd9+se8~B|7eDcx|ddaR}YpEt*r=Z3S+pdzEp&oQT3tb^mhnT0ZvusV< zhpnkw@u+wlHRxuufg_598E-jR^q&FwzdQN)E)TvUCKB2iO)Pa} z6piUPgrxXi6!LyImOH+nU$K6We%(WQL)D-?L4q?@3YJ44#d+a#8CvV}uS%GlYey6GeBks3&7x#K`o$P}Ggnl|ddh%&CZSR|k z>Rwm=XU9z5H>Vfk4SqsrP9)75ZbthznA2D7Hj3`HA(xm3Y(sUlc!e5csfKcNXlw0FT5SW z6FP0r4wy@KE@orip=az({{>>_QyFZX!*R?hS_b9i%OIcHMXaighS@!$h5l9a^ghwq z0x7chpFjbt-?CBvHL<}fg3vYU41(paqvC}WdR&i@W|0)67tDa8tpWZ`tfbq=sH9@QAG-wTtU$bBUPGT-SuXdOj#<>&Gp%uR?p$VqBfwhnu*R;K##N!u#DB zN0Ue3-!mn=ta@)-d2_RP>*q6MAlX9hF-2r`S_{v|RnxL8NeX@y&GM&})4_<(Y=N>4 zdu1ZU&t9v=N1r@M z`!=l!lETc*iUMz72-)HmEcwr#2d&fL5!=-Gv%B|Eb$C4<)|FG4XC?a-?*iSZdyw@q z;A7olVWwY)!oJB5O7fi>KMs14ib z$$dTXtk*7xe&|SXlT^hck1Mlmu@u`A7Am@OqXCnD_+wS;EU0@*@FCYM_!_@CT%)R< z#Qq61WN0%p*q?>xV|4g}iecR5?Iy1Ap9imPvgMW@Qds7Co^)grgq)fbmJD8mmoC;& zD4v23&o)3kJQ*`jWI!v=4u|5F)2_H~%vz8EVN^hVz9PR?{|{B!?+{^8hdVouA??i% znkZz-^L~1vRecAhNMEL-4<@k1Z;pu{*Joq)^E4!A>_@uN3$}gheDSpKK(_4XSLQS0 z4(pe8osLc#%*S0d;dbjZ_zRz@d~&TXH#@S1H*_SyG*23fhZ&&x)j!g+-bq`Zc#&Jr zbb9-DI$bQVL%sEOXs`DN+t>h|!`i%2$gWC%ZbFvm0&dL|L_nt?+_EZUtLy66sS$l? zs+KlycKHUGQ%^BrUKY#@WO)1jsXXvOvyi2dMq>B?fm0Sb4~%}0VnCC)-F*yS+L6c~ z^pE43Hw9ibd=~%D<`;InGX{IIo>Y!TGP?^AqW=c`5tUgkVWqu&VJ)vf^7BWrQN7=U zzO`QFEYl(m8h;xH9d*!<@_-&*9S$QQBR6-zCzd=)l7?+ODcqqSky|mEp6f4TqVRL#`vWYKBX`kR%x(XV>&`seu)#;9uY9qw# z+eQc57O@EXTcS^WTSa>6e&R{R&)M@^mtnIhf}F25if&OPtyRfok{>L@oeBkbC^w7$ z8IsBi4Yc`&v;|NM-9<_5&v2pO6TO8dOL-bA(q9)XN*E$1&Qy9YPTi9$_E70yvy#>6 z+`wqD&IwboZ@L^>8fut-#%q!4i$7$!#-HR?C$i{_-i@+}DNEvoW+?rb}d5yiwF(zJMuxRpyZ^1rK)2FZT3=7vv_L zgmwIM1O;nTU5|*xINTL)PMyS4s}^8&-Uqg&{Rd0;`zdzSO%)gXS;kg--xr_j+$M7F zr%l{?99F_p?DAZI9k9l^8mejJPLKU)A2&;B&n>9p@wxawD~~<+hxC+ZlCBT<+K=7 zxcX@wfgMNld(2g~~=)+Ebj7VCDDfj&G z`(6l~lZW#KaU#Ae!<4W0t6<;KGRaJp;I+_^>CNs>GL_Y2v{W9h@(vC7qZXp_Dnez08d zGqSLc(WI`ehH!bbfD(mlkovG)FfbB)w(;e3NzV+Hr=PRHvdwJP6%kEXno6m9yQxG* zOvfCrQQ-zVT!`O>F&>xbNIz4o-Dimzx^*;3YY=PeAw?VK#EChtJFd(1~HZ>{0-K6)fCQ`$C{5^$!caRntJj zx5(bsizU6)Y_fwn`MuPGwXr(g3aFyc*l=7mPbH6y26Q#@G_}qA$Aa!IqTjzbdM6FV z^ts*Wr#y^buDF5557)6P=^cu{SmNWdy-*SExyPLcD6IDy6}&tRd+$uXG4UH8u)K$_ zwv?qM8x`QbGY8EZj?-{qN075tL#uHip8Y4!cUdX%w`mzzRA!y?Lfp3nMjU7CfqoY^0yFfOAT8UUWkxN;-f2Zle}y00;-vz4yM^eiNkDJ?A^fTz zNeOPBXpZ1tc=D$dY-&l9{;i{79|Q)0siV>@1;>BHGMnpXSb6_d(DV344Qqc;|N5uzY%#9f9XB9^p-~G*^21QRp`5%d`88!hv){zFg`ESAAuNTd6(N`Rf8D&J?m2 zhrTjVeF;gA>!gN~YO>P5LWy+}sM_|J4AM@J)r4O(T(6qWJNQwIOFVtOmP9n;Fg^EH z!`CulhLAT$@g`mTo}>V#SwW)U6-aXQfVc4ysLvUQN5|zbeR?HXI~~Iu<5Q^q^&8># z)pW#7gVJsZ3tXTH8JBG(*_ody&eD&r)~%w$A?-BK$(5$28&Q~nB98xap)Yx$Rk|WF z4ylBR(<3Yq`m&B)nuo!fLr`=?3Qv|P(TzKO>Bt8eQcRP@pLzWdV)vU)ztDxkOl{N* zRlw+5o`{PKK*@ARe7~-O%snq@oX~;&&&L6kAEnS}c%0&cEol3&t<3jjHZ#Akf$niR zLN}2&O)^V_%fcW$j9mxqQ~lBRb3FN%ePHLKd(oYcWa9dPAo@l4wn_uxs7hkH^#Y;e6y=3?9D}y8B&lI%XnV z`(DDq?sDvtdxG@}0i>5QfIRCM)lG_^+?H^fwd)q8btcf}VkHuF{l}gx7zfG2(~zz` z0g{W)(aihm^!LXD_G;5`l61VsZj}urt?o7$IJaP;Ko~6ac0sZGXebMQ^;m(Eb3K;D zzG|tXWquO7Q5+*)yxNq`ZJ&S%f=_N=VT0fsIr!(g3YxYPv1&^hp6XvilwS<28$@JY z5-x6tJuY^pC|KT%McR%)Y?A2%Kh8x9!M2mk=z15en=N#{X^$4&v=WH2;bUp& z_bn8wGX=GUbA>*QSuhSs!oC5akTMro+`U=!$)tK#F39nlcxOuLS4(02R-p0#t zc%6)j9qTafR4^LWuY{}5czpYL602s%V5{s-)Sj@#(0zT8rB_IKPpoK@%1PSABiQ^E zelVZKVd|rVha+}SnlLv@SSnKeka(fDu>cFyPNG1~4Ukj7Q>_}>VO0)`Df=;~#0pET z1pncq2=PwiQ74~)G@C@B^JyzS<(Ja3NNpPQ>JjrV+8?rt_0%#q12^2l@%W4rc6(mM<$I6twW1l}UdQlb>m=0A z{6w*1+iCJeD{_9DB>w%&2<<|ba3A^$>de{4aE(jw3N`sleK9g52d#WIWE3^0KN?925x;dl!6r4GeuP=sof}xW4QR z{Cd366K0On@A{$1xVA`%H~zN4yeCKL>iW@qWL6utBaqNOP&Eh-dpO2OotIgtvTb6DMj z)#89Bk1_hcS~7KTWme0Tnf+sTl1izfuGVri3w-XNWF`8e?1NcKwMfd6;MJW$JaOVq zelo$FOSY$B-3JwJQ#hPo61WCUlZk?!(-(Q7;k-;r=pWa1#B_u07|84}u|Wkz0$(NP zyckD!WJ9&Vhm@ytTJwFNP(1X2-nn0A7ngX5u3WO@p5g(#Rf+iC_(1;eVn6OEe;ZQ= z?x8|6SN75`f*q(I${O6xi&k7{6#LyB$y*N%=YNm?!rTFU`4Zi!c(OK2(9|S&-R&^E z8Q*~Sqvv7VercXq_W|?%9mTDvIx=_&bcLJY?uqRP6Ebp%^ZW8`W_q}rs7e3132CH}fY}%$&&kYL7>En>u>j1CgE~#jpNr#u>{m*b%xF_wUR_bdb=ueQG}+ zA@qP;FP_9tw(Vy;e;_o*|G>R5I=JLKA8f7`0=5-Edyxfq96p_oO23JI`J6AV9?e-! z4EynQ0n1)-g7uCv=E*ZO_+9S-Jl|M{cgif~PG6(Ae_$Nj?H)_@XRcGB{5@QJUrMjf z*<;3@)7X~YiK3@M-v8Z5bd6q#n@?RZsPZx%tdqn_At&lH;RNrU_62@TOW;*g!&)9Y zLX>?GO6`vjBG%$2?LFHB+TQB%ZR>%39--SHa$e%Cxb%hNOwBV~Y(YW)t z2@+yczFBDyzrAV{cRJLMcf2pd{qTIkGl3Ntxq^-Q+Qe%U{&Ksv_x#l3kr;onkt8G~ zsL4|r$2S~jC31nRoa^$*SMxDG$(LecOqs)&T+#DTKW5zV7Yl`)_?M1@Eaagn&$_=B z2Q6&rQ*VlG>-j;We9NZThR1o)neCIP+w>6i_Nm3Q32kti-GWHhG4!PZse{I00A2<#zcXfVI z$DT_Gds@hVd?A-HmYv_1FAh+AJ<4tRMNy}<91YG;gvYRGjQOz|?`IE0gWpFAU$p@p z3k7ca+#WoZ`b+vkkHHCfHSy_#F8s%`B|KZxk;iIiiTrLv+4fhx#Cm3qhpaYzyrG1_XwGy0CD0V4RJ_-3R~VDj6R=YQ8LR9Tb>$mle}((1#Lx=cDi^F z=Fn+@LwOXh#QkI^au*vXu9;*l^lbaHDK;-f&Qr9|q`HW$%61nS9P7t7f0)Flt1RND zPnPj{;&iSvb2qPFWe)cf+ECPaiaUiP;c7>0H^D-X?PEW&L z8Cm{}Zlk$k6t;wg(T>RBBp(sQBF5HIIA2T~JoKsKJ!n*dh&`C3!0uVv16PK@wsAN8 zQd@vCJ+^4=dO=Yg`$#!Dk|ty?fZ5+JY${JdzwzT?Rw{{r8yfJLt_0us0sKNu9o{b9 zjgL}aSYz*T)=>FFy!LGz0*wSeZ-zJ7p1zOZt-bK~`iQ)Yoj5d35vH4Nk^iCsG<;u$ z?QJi~<=Hq4)fdjsKnprG{}FTYZeyMkqnNaUJv0^H(#rFvNjY>htjr{lT@p+B2N%P6 zh$k+~%47B}V?OKW5BLb@-5|ePbp63r3K@PMfzFa7^CnTOQXxskYFpuRM%c5m+~{Mu zCqGs>m`~QM!cFIo*fBL4PZmnx?ABK5TM|M4=GQQt{*ruoXEAJw*Wm4&FC=jzPUz5i zAapi_V>FM(^?Xn0K2N4?(Tea)nt+=d-6$$n=xN)fik|s8e9a3LzQo`pek$#Pg0~D` z={Sy0iyqITo2K(eLch$aH-V_laixJ379=~_60hopKwI%U@e6~{U+C^`IhjK*+5lAx zT}Y;Mk{@@&(8}Gyj4ua0Ia9J(*^k-!dW!tVL}0#FDP7Z6rOA=P-BO^=N8GxNxvPy~ z@L38i_g_-qP-{}1-A)RMM=`0cne}*Eko^dF3w!W%wmNr8SK+fQ|DwG3CVI}~p+I*& ze*G4{SI5C!_}mHEhLFK@@&AnWV;LIRPX&ulTjQ_AF1%FDq_H>d)0~{8L=MCGwoQ|H z)GR&Tvtk%8cs-Uca2v*FPBB5vz&T`obsY=W%0aS@Dqqkb$)(o{vuoc0e9pi_82`M8 zx<5{!`l|wa6=}&kg`HYW>Y>mlQ;5(zKk&v|nQy9*=l9k%!}-G+9R7M9n}pZaKmLekc{02zJpBi157>h5^$1Pu2rPmr{ z{CYtnds9gvYA@K)78F-4#f7oP$STuC_Fp;l7r3W2^EYD49aW6Gu^S(LHbFtfWypvy?t;+MuGH8{~RFt&^z<+!om0cZ2@pHsN z(P|R1*9<|vY&{hg210&(HMRA;r}#I5S7G#re(g}EYwj`paPUk%J?}e;Vsq$&^bBNQ zoPdv$guYF+@3c!Mnp}-Opj=Yu7B+InsI#3Uk)i?bP5uZ?8O(!s3EwewQv+L~U4R*B zm@j(*yO*b8aIFd-(%$H&uS12jb1rp(#BHzx&1FP{%1;I(7g?)moU` zm4;;x5-_}R54|uNjEW8JXcXj#z0RNq-eKZ|-ZN+p%f``7fiSxz$6QzVlFKNmqip$~sMW;_BlWH4#oR*L@8 zhCcB(u}t8o4=iYA8;y1lEw!SGX+e}x^MO1MF2upUyD)iCD)vs2MRWHkT330Py}j+j zl0A$uHl-T38#?i7xhX!G6(Bz_0b^a?3k;Ev9XvM`4jWA2{MQ3dbTi3N<|M|K$DwSI zKk6k!+@yUwU#ZcbD{V`nae^M-?`Jq0a{3Uvek)L%Z+KFyr3_v?-GLj}sPc@aAILIX z2KUM%wAoCV@0LA?G2a(s=#TMy`>;~@30<`Y*?u%Z{SNcJAkQwkI%3MBE_&CTNT0eg zdB=PUzM-`Q%W_um*Wrw-r|R=r+gr%bCZDuwPtcZXPdYbiJFENrk3~P)L}kyb>APkM z_0jqw{^fW`?Aj-r<(HJQrajli;Uo}ev~C{D0-S!qNw`hFrBPNUyjbXKpGX;$fA@ZQ5wtpU+#w+w%5utFKRxz2OMLF4$6h(IA@U zU`&(eMDlN8u6)iIEuI(Vz~j|rxyhg+eB5${o_POYVwqf~zI{5MDXGEj^n^C}<8Amd zw+$t}_h9{O5qDgf3N`NmD80I$x4dxWOZuDg0X~v^Yu$Uy9McS=S?|$(WG6-~W?1*7 z9bs>y;PXoaL(9wD^mzb3^nM@LEmybO^Yb;IJGFr4`95R$W&OmPPtOo26&I>+B5@S-7?pQp^dMvAy<&LST1 zA&Abj=TOA`l{n)N%x~%^aC6@zUgRCm$ow#~y(wfdM<3;^s)Z+QkhI(O^)03D(!gNV z^Vrj;kqtUnMsd+gP(0%=su!)}>2;aBZnG}0$lAhZXNK}V)38pdHUbvV) zoMs=BCH3Biq%3&F_jmuH>-xaM8;9|IZxwzzs`B!FHqMG?~f0xXJ##xz8%5=<~Pn(|NDOZoaiBnXR%m#juC3arw$l>=F7#&D73elfEf* z`$%w2yS3b3@cRd6G$O%f9G~;f2PbT+Fs-xK^16P{|_nemhhc*0(Wp}DSwf7 zlW!Yb$cz^KpfrUtT<#-ei+!Vn@4|TqlL~h$7k^B-`_-5(M2IYQ`Q9S*6G1NivE zlcor4N=kz>PMig|p>YxmaoE|A>3fpeUm5-4{ea zf|5i)q7sy#BFuE}wHpvgf|yVd0TDqFL@?(_MnHl{P6CR65lJ(>8_Wb1BPxOc6vdnX zQ8C?qZ`J*BPt~cq@A+5VshSV-W%lm0yL+wY`8|=?@6%rVI~QZ+Y;|0sJqmjVw1LN| zJbL62EqcVBWptZ|0|eL11>f=p&ed7?=>A3slHgfZ^ERW3p3g9|Hy@I8 zW8jK$2#)DXz@gdk_vt0S5}=I%(dbfvI%WutI`f-!zk zs)a|48iKo4t;n#(3OyG-p?&%MjoblO7@?U6oym&WD?=LljN~=*uQy~90y}V7R|1dktKyZV z#gI8&8_wb5uwd#%lwTl)Rye2Q2}~KjvHk$|Zas>FOb+706Q1~AmL`68%>{q(M3^!d z3Ayw?DtJ*08t^g2y#>>-eep0*(-0o_Xc)d*tb#P}2>Q9lcHl;SgvF*((7nqIcKkUA z*XCfn`u8v>KP-Y)hYpn8J_RiIg@Mj27ksK-4;P_oNM!fG3+)w9A(I6C6e;mb~-|usvZbu%t5}Q ziIn;Re@bquF)rCL9^YLgk89+nTHP4`LHJZHl^)|X2%Ck^;cR6gC|sI_))aq3_8-zA z=1nxV^5*bfjWgIus#940sR(8B-VM{X3m|=~3Z{+S3Gqe!NMusL*DCJ7r%yaf*`ES> zDGj67YNHqCfbvTR(V2J|{Mm|Uf!oSqx8N9q6;mQdYc(#ilsM)*1>jh!7&dPiAz%3fjJ-i0n)-T1E^1Jb;J5R8ZQWe%* z%;yL#afLs7hQKhJXX_ASZy&6Y80Wm#VD5k&ox$A|IX&`*JbG*zy|9 z(RLsw9tZPp%7fCT1>mBUfS0tV;Il2W@PLv&l1etD`)`j&OS2!~C-2W-*T`L1diH#@ zNu~lFGy8%%IR+YgLr_V6Fk11o5t_L+P*DhmFozAO`$7#p_uNU;&$C;{+Ek;R_IIeb z(MMp==mDBmU`S1<$r4;XddNCHCs1I~nT)(F-=hsmr{JXBA9z2q3(7y_!n46@IwVhr z)_4&`{Yp>9OWKa$SkE+^vWb7Tk;uoI=@Gb8Ig1XrX{F?JWa%>pH-N^ANN^r?2KK&Y z;OY*(f5US+raeCkJ$o{t>6#pD-#!s4c11vk*%h#J9EpF7b>j=WQTY48W#F;$5xjEd znXFR+U`>4%gjvSHER}o|Q}mL)*A-43svM22_v&GbVjVo~Y@+qJ>bsP8yfpebbPGPd z+yv``Dd={X50chC5Epd_w!e48?ys}(hHF>wiOg`cDbg3|D)UT>t^B&Kev7`LI)DOB z&I6x|4d6MdAAT2`z@aJMP=mW zkiI|*x6Y8l#xCo@jb}}Q`2(c2DGte*W}^ckGa#z{49wAy!w;{R;=SWaAa?5nXh?KJ zItymuzvGke&hI6dS^gf}d7V|s`4|-S@jf)!X~TvOmtfW-1Ke&i3FF~tkQ^|DjPuvf zt1uyw{-T7FmfwZjy?-EicmmE@`xg95HL>=U%}9IyELz`q33{Q)^SX0o@vZfK*vhE^ z9zD(n8%KTU?3@RS^I7zJX+52P?;M02x5U@VgRtcDNnkWT50tX5g2&|}pvL=hoMv5z zug`b}VABZf9Qs?uQkcndgdF&7fP)j*_O7+BXXL#9>3=cBCe1!>@w#e0f1;~-Wu8czFsM?<$Q;pi4!P@8=aCjBzR=Pz307T%-!xN!rL zkore!^7&0$eHHLBN&+& z_DSLr;|8#h4~C;(q+#c#(eTHv2GnGoU@q^;oA6tXR!>QR=K4*rKXeqhX-e?-6~v{xLH(B!KFQ~%dU*4_&=Eo$lBkcx{f!XNQOWZiRbllRBUs;+1sg9L zLPp*R`YN{`*~Y~3{V86T_A>&%_S}ju-f0v;Ww)saa$OL zzZut|pObh!!0wR{ym~J-9b1N#Dp}n3MguLnaujKfcHwg%9Py_k2XU6vIlOcIPOuM} z4>O(iLTqmbTK*NG>pN7@RvlS<`N9;ue?lOBtvCgbYnzAlqc-E3(Qi=FY8_A*r4QN< z=iuR@`FL}+8y-F6+*0(3{@$JC9#_-A9O^%8ZM%?kQ4m}h7X}Yy)4_v(zp*|v7LuG3A@}A5sJ~l_er=fpF@j34{E~&XjQ>E} zSt=m+zXPZV%>palQ*EJg6`pMS1bYXK@VICRo&)?8y^Y=g@@rY}I5-~P85NAvRxbjj zTlEkzYdp>!szBKaHAw5bE*RyS;`8k`cum}F{IMt)-{E~bbLL#ZU)JeD*h(>SVIxrW z4jtf5*zh{{Na{GxNn9{O7him;fmcUFVVhe#v&%yg`^5~PwMRU_s)+)X8$&QMQBelF4*U z@+3j;hhNra_>c9Q`w>8B2gn#p!Jv5`pXu8I1p|B@#X32ZU6DjZpK%rJzmkT3-#?7k z+WX=(o6E5LSOW0>4*C&FwATUy>O}i_%Id%n?6FhD^M8GYF?^Q!L#6Lfxc(Ik{fCXS zR}VG?OQ7@`UD|E7s_;f_Ikn-&LnOJn47mkfK@BFe__@!7&!Zlr*Vb{6H8Bqi1XjHE zmFJqpUKGx@IwPDuZW11RZ;C$+$zzx2V)W{b9P}k9!#87o{~cGva!X|Kqg^Vf$=8?G z`KLu!nmFOU-70oJSW#j<44ORkYk_Q*V z2l#v;Yo6~Z!2Xd{=*XFT>bs*lm2_1KYOk{N2kx}6H_!=Rc*dhW?0J6nwqzV@AI5X3 zC*kEj#`ry0 ztrz^2bp(1|Am88I2G@lf;N;eWaHdBS4yeT8LCfJJB>XrJe+Tu^ zoSs=|xsN8g)GLAe$~o{i4S|10U%|l@H{o#a3Aj}hi=MWmBX_Gq$g7s;{kxumS6PAJ z$Y%`n5*a*jk=K@sZ=p{{lE_)Phn@)2@%K>%_$~17M}PVC*+UUO*e8XjdftIkE@iMq zCka+wm4GAb)It6FByf6BOu1gqq(XIyg!N0V!QCmj@FCP3PCQPfOe4O-t2A;fw=TA(&h4k_TpmuvXlt)&A)3|NmIxq#M z-Q}5BF)<*1G8H6^3t?-13GBU>1M(*q!m{jbuzP(Z90?eUZ&l0UG3P!*&Z|cHP`5d) zgW{+w>bv0!!}k>U8jSQ;o@=?|6J+MqfP!~E94>YO`;vY%zC#)vZxd0-Q-OXHrG@Qo zjl(C$%isrR4`3COK)i6H9u8HIz`iy&AXiWT59?Ct+~4MO+U6+Q#)Ic79Ms_7ofPqa z%_V3XyabeIIaHsE1^Kb^F!}LmH041AsL%9rHB8Pu7;KNNWH672akhLYUyP4M+DXC2k&D2Tpl2-l1ZpiIjg zO0KA3I$8oh+HoItFJ6j9@Qiq8uT0vC?;kDeZH27s1VYcZ!GyT`@UOcDW`4O$hfI1* zje1c@Me{w^e`Ob7y4ya8UMs*y_A1~%jXxoA>@@hU^a6=8Zlhzoe{ZRK1aMExp~+MP zS!r>wWN3a;F2q2IUuA{208_lBxX>EDnn4SR#XYK%D zb*Av!Mojg8?xt4r=ih}fAE0hVH%!puGwp&Mz}NON>bx(9-wtcx<2wxSwwI1*wN*d; z*xCb`FJgh-QVQmivOwRM@0Dx{K$|oEq0=uO!w&m;P@KfGmD6~gfvr4@P_;ur74mq; zxC@}sA%a!8o}j`_2Kz&!U~Tg&co}gNJWX`r_hfH;HG|^o?fg9}cn(ZVYvg%Kd~bvc zM1AhpDUB!Ebokh>u)5|eG!Krz-t*7k=CP@``&BqLlwAcf+n=Bt*`DBi?jxG6&Sz%4 zG{I?-#Z=?4DtdU@0;Jl%QA>4nP)buAXzKEQg@I2?uuEJkuimb znL8kL^97`KvKZ95N@2`KMQB!6LTcluLs!rtG&V*BHD;%R0;dP(J%Z6K-Ce@K(g3(U zc`+_Dd4f)liih7vZ^0sT7|tIth1ExHqBlJebl@2gU$=Mzzow4CL)E#ko!47l8)pks zHyuNJ^**5PV^uh6+XDPP<1Rd(K7r4F;P3VPed|v1VVaGNM_I~eLAYupEIJ`XMf+4x z&}visDD?iE4T6mH*@_h?YA9Q*5?j14&aX zmgf0smf4Wm$LDOc@_vUE@+fIcK0Fk=K;yk0WHWCctXCZchyTT)WBnIlq2+!^2YWcE z%xBSM7eVUB1en}20WuFOqAKZmbi=N8NLhJ>&k)Ol8~fkFkXr`S6bQj`P>7>8+hMOu z+Iak^6mVEP0VcWhqTQPsQFEjsnsYLPUR}u7t#U_0Mb=pqRHKCOw*kSUu03?~8;+jn zywh4d>_`XrlnOJ$cL3H4DmDpl$Ft#;Zjo-cCdvE6z@oJ|hK=C@n2c~5Z zSCJ>W^C;g}An9ESBT-TC2D9LED9LJ|n;lI^;j2P0L@RrI$zG88%np z%;)34Q!8*I{T2i(FTvO+3>f<WfwMk*ZqexoSSRU@`RI7;wPYqvvh~AGs(gOcveCG5RxLP+ios-h zJnS>k!sPNjXpzl_M?DE}V~itQ2v7i%k1QzVRl+c@TZ_=qMh}WBDS4qgo%Z_&dX^l4 zbZgJhWwDE}N*0CdU?z5(ydKwXbH%*^Bm7H79p>r`p@}#ZrPxe?aDM;zACllb%#*N8 zg(8lZeGmBU0=)Q%A&z({f!kzxR>3H~=A&c=X53jiq$P-6S96D6St1R8&0e8JKS#h^ zTRA@abq8{a;pj%^zery)6E%13Lo*h&p`e@T$Z3rZIyu!EmOPY%`BweNx?BsEPWg*6 z{N$i{yd)loI0sha*TciApGZ+Qm!4qtikhZ$77pLL4Idxb;|s5@qnLrEw5N%bFy_u| z_{n?!GVH3+$Ptl9bkOOW_jEove@*xGp+Ub}ud zHi$Wg%8HxlOjRR#L-Rh`%1D=rvfoQBL>dO=0kFr4qV$S1<|XcA&o7E zplKIDV4VU@zCsuqTaV_iw*(D+TeLaUnSK&B1O0Y+OI40cqu$*srs#-$!hLN^=+xE( zwIT~a> zA3}nwKhgD&%Gz`gj(>g(DqhdPXy-F{ zvUfk&&E+$c-DV)2QaSo>c{$}Y)fEgntl+t=11y;*pySm81vz3bf#Yi}IKO5UN^MW3 z6^5kfiBfZ@9ETigRXjo`Uh32PT!N|Y+3$eNSOw*Wo}#RSOJVfyc(k#zj86X>MsK?phF7NT%@)}hx zmxe_xLhwjj3WVpaZ5XqgcI|Vgov)5T7A97>z0(SR@IhE7DNpFUeg-8O-cH^4`UAzC z7zb}eqwyd2Lv+JDOFI0_d-~L;E~@oqF?H#=KCSkS0;{Yl(EeeCU!_FDE0sRD6feL= z!5V@ozPAOt3=3#*?)bu25V%9U3oAgRD0cOmXO=_pbOw%M3-MG1rz- zJ_)_l-GQr=XYeH90a#59m-9LFSOFz%;@t}}aY(;ol5pY`SxWY4Ep_XFKv*aY&N3wLD9&-PeGOC-M7_*EeiFl}m-} zyvJ*$3n(u5BG4XJfP8L&=hr8rKy3~p*-jvsb0_sLc9Q|%&pPW_O-{%X3zy{HURero`|JcSWAX+=8+scYtSNA;e5eg&j?QVb>2K zt~kDe_S|$1?Od1-r96}WC!hH_R_-K7mbCJoejWU(g~nr&tI?e9Lb}ZLAv!j85svK1 z!KtMLf92m>Ro65KFQ0ZrnZEO&(XS@PwhmM*J5R@)U5c!fFQPzg zS-j`aRBUrJ1phgSaCZ6vysL5*4v3tFlAW|cd3p(`siwj*)e11USqiq>RMFwGJk*;L z0>?ZY=`jJP=&8?+qku_IKy9%A2TY5=k90nu7&lEYy!XF%x}x z(u2my@qPJ#1+>RMGc?r6vm`I7pw6yo^uj%F=@)#_boJ>ybf*6(6yW-k-WO_vuF3J& z=|BrbH#G`}ZuZiRU!+jf@&vkR!Y(vvoeu=@T38sk3i+eoLt=w6KA#16(t`W6>)cA( zs51i{nUICf#=3%)oEV<=ThNO?Wl*L$2}0imQ=p-+4C1SMAlp*~t%{XERp%d~y~9Q* z&-5Ph4dQcvZd4*So2{@+&=0efq^UJg8GSM2;ULq1u6=GUMVY zJAVf>HEIsL{*ew^3+{kj9`AkHmklfC2H^$nvZM0z_YO#;Nl5(XyTDy)q=dw%QT(s} z*GFB#Rw7{2%8)hNf~Kts*b*=eArQcHOM#06NcjA>9|?(w8UL9>XlThPeqf3JeE)ZI zfd6FHpvN!T-?s)>i*H2LHwQ|L4aa39S&n?f>QR{~y|ZAx-~x+y6f_ zJ|+17_xMvF;Malw;`sj$Z9hz>_-pXL82|tL{DbNL_Zs+Lo`0i@lf-)-RATRegUs%d z2*%jTlNlRp&DbwY!XKM78T%)?+~wpz^7~>Co6M@B(7&eGNWBBA)@gEXeW7gV{XeiL z)`Ly?V#-8*VDZyCZG0c#GShUmf_YFog83e>1KQ?Wv7sX{y)Yl*F|@VwTQRz%TsBzdB*PONJ8khBpe&<-~TGVfe9 zIX$_7sQJs1_1XU9jKGd`_1O~Zzws!wc@rs?btH=;rlYu!d>Hx>&VGEb3*LLVi1y?k zX4idJ#u<(A_=)5O6573qt=Ke=alY{Z2dylG{hKnF$LW`toN5zh`qgB-?ve*{^>GYt z+}eVz_G{t&hKfws?Uy2xE?KDfn+y}8Mq^b^7CX4Mu-lZRx%gfUX79$?oLo;i>dZDH zluHXqW=`YWFm)#X`Upnf&Xu|MbvCoq%o*=LDuvY+Wr+W@mx2Bt5x(G_iS1^-z*hYu znP`K}jMlhRd~jbe=#2UaVTWgsUHAIg!JW?t_xLes{ASPi{Vis$ENEoAK8#^QYdUx? z1tD`UW-#Pr0>jDrFtvX*h{9rH)-_8&x-MifSJ6I3MSTdzRt;g@-z8YiZU^oxHx%oR zmc~5|YIvJij4wHu;kmX_%+&|EY`_;xs@$`%;njIKNm+$0v6aHpGWyw;{a^7j8v!}E z-LbyO$prt3W^i+z4bh1_f$wf$xW?x$7`xx%b80Pcg!g^882=0Roc853!sjtx;)C(A z3(s&=kR^GX7t8*r(IZ+1W-+CS)0oE-R^n;)LgsDaFDS`s#u8mMbm}JrQ+bBR_x6eG z%|C?Etlf?kI%q0ry5|(){+%9+0?Plyn6et)#7T$Lu8}TQoLyCCX#o5q3E39 zF|zHW8E2x93KogW@Cp48kWy<0r?bwuPi;DreX#*c=dEWFKOV*{J*qbQUt8GJeQsjZ z*2d!%-PnAuuskR~){`@YXmRn$U-z%(rX)J#Zm$J_#K7xnN6?{?oFFyS414LvjWOkoc zqt#CE+KBK5Hql&#adQ8HJ9VP?9=|%)I-*IWHumH9H=<#KW)kZjtxDW!CG5$&Z;}Qx z#Pgg>*yf!pnA|Q+PFmZN9c`zLXUFK^cY2y^Xk|I*{BtA*Nkg=D^Eq1eQzJd$?Q+qR z*spcEi&}-B9`30-tU8KrR{T|e{?9KfnR;`&Wk#{E;C6XUt&?$Ws(PArhRrp4&$Xp= z?VcW5+;O$8abGvB7?(+p{(YSC`fF3)WglK=aBhmI=u@i5uqH(m$fp>mXt#>AmtUyQ z)XuBD5{v6^n+FKWq5T|_ug zdxCPPl&sCD%&7lXEkkA4H&dH$MAHv%zM|GF4HayAuUv2Ok~f*o^c8GgxUN=L;$wZ< zD<4tM=9839-T8WzXDapb0^9oOzkUigbScszv)xqaBE8yQO;L4a>awC&B`@mV*8Hq# zjvQHQQsQ8>BT2PRHF;Wns!vbt+n4TiNx2VoRP$YJ;hLkO{P987f#So~^#(oSAGb8w zf`B#RBL?5C3S_qmTQwu=(>=!5@5;5LIMzX=ePS~0CHYSDxh7b+wfSKEk^4>chu=Q0 zpEh%#PBZ+nASxowy7ZWX&~;IXAo_iUU_@nW{gdjib;{?&bvZF(YYk)T>J-OGR6omd z7bQ=>F0#qhrU&y>si0afDr@;a;i_FP1<|VMRG-RND(2M~TDv4#^e~L3>hEq7&CD(k zrY%`4T8VXO66R9BeZsNYJp~%{;V4^byMB#zYpS9sYfFXDR}w@MGu~QFU$^<3?j@7j z6&9_cktg(pnT7SDJ2S0Ci4ux*(Y=4v4})0hK;IH!pZRiut^CotJJ*!x>oY#ob_S2A zH9a(mY6?vdJfxn8G7m2kX>ODd-ua$RAM@Q=_jFjjF1)Rm3a~Pud(M2Un|*nDjmvW_ z>U#Zdk?qVdLHUvUwDpo$D-F&0^!MV&_2k$G>X7biQ3UK3Y1%9mea}^$qhvY7crNdc6H?eL?V)v!1Pf(@MMM){nlRZ~bEWg_=d@HVN)ljS}7!#0ht}`PSLp z(x>hITF^;0&#EFXD2Pozn~B{onu}u1bSb+c<@(~EcPOh9hN7z-Hll-L1BH?jW_5Zi zv}muMY$2xn1%crTf{?hWwG-my=#7`(QPkRwdZqljdb2oX+9Q5ntr}HXo3@D(eeauI zZ|(crYT$jN(DYe;U3|ceS{t`7)LDrVYMIV?kzbA_?d#(p*e)eWPu}h)I@XaT(lCvp zeyVS+zj~mwUf+16sI4YXxY69H_S=uQ;y)qAY~d_v)CY-Q%v+%(4!*#CB z+pF(YOcL8xtP)9W8YiesHm_gwPf~=f{6(>W19gFS{6!jXdqmM8=W2dmt)f;)=G3`l zB-gT6pmxfZ*|l?J`I_F~py=!a{rYi_Cea?}QYd=Ut-7JUgy|K%veuTWx>mCBrXr0o zIk4`OAMD*0DJo0zV+G1ac!UQgd52TU*@LUuZS$hZli+L+l&ldeDTU&JBH-FY;Uv3| z&v!jHfHv#R0&a0Fd-G~PJEr>xVy$*D>IZ%h^Lh4o?`AQ(O(LI-yEub+Y*U0id>>({ zBj>SA?kza^>pQ&k-VQyK5i_xR85{}zg+oqUCz9T^+}GLwZtI2JO!c&K{Jh5(J0F@$ zZf5%s)q4o5uJFRyuNUDizp?nv(tdH^fn90 zNM__4A+vw|66Vp-#mwH8{fu$?I9z4TKQoLP%`-6vTZaT>pdSY2-abur)#sOJIjnp3L_UtnH6uw=FqLkhdDswAm29)HsBv zmXBu7eQ+dC?Ua~O;UT<_=Y$PUdCx4#Ao-m8PIWvdO=rMWXe)6F0 z83{7|C;m|7%~XAV3%i|~X+Ot7Vz6%&H=$3Go6{4@eF`?_j+B?cgYS`)MjX~{-aa4l$6)hu4A*&+;Sr%pcj#9yLMc# z>l4@H;exdr{$R_M$vAw)>;`mB*bsEjpdtN^8+%vn3u(Mu#vK*=aDOKhaGs!C*%HnV0xSg@= zys6{yP>?J`d!~{d-LuHdaY-aMD1*D7`he^B(oRO7nTqcV21VYHv$!fg>)W|{JMr35 zMV7axHaI)yHyrsM+fd>{H~eWm&@eN^rQy?r3^CQ%Pg+L)64ShfRdbPo*iEmN`(dfb zeeqHz>1z30cWf%r9llRixNA2Q^+)0gss(j?cVeo$-MHhW6F9S~hg>^YGCm46KO}AN3?+~8Ht-&69QjeF}oa3(F8X)go zmoYkTQW^1lSv=%Ju>W-v4%NtU+q3$Zojzw_d7TMy+Z{{R-R)#g*eW!9T%z3YWzHRL z+Rq;N zmP76tm@@XSLSSO1UV}kS8n^d+9w%@1k2E>%Bss?|@j zzbl(aR@r&NCGEwn>5h2#iclm>t%t2S2jS4%8BC7m53Kxom@A*_NnWkjVN%{xY`lyq z^V36*NsK>?&DPDY4?G@+PkS?j+@3^YozuZpeF5pbxSRVRQOeB^%*3uoCUJ}3&gVYp z#d8<0|0Rj9eu7}xSZcG*Gx$v(?6XwobsDEAe11TW)4PgnwqA&3Hj%2RCw)TI!hD z&4oBQlJB!jcI7r+uqQpv$_)F}5?tLlIAq2@H&ksVArT!Yu^@n$!~~%Wj5^-F_8j)U z(2f2*_F^5&%-D~LDR6Is0?u_PLItaB$#7f>K5=g!-lmMjc_}7PDP0BqXIJn+mzMa( zpEvkSZv>*`+N3V*s zkZJ!k*uQ%&vD@@~xJSmj*!1Eqwmi9zxO~>(_RDE=HbvL?XAWQHvW`B}5gp6?OSI&a z>-D(sxG`MFL22}5;RRAYtC*Zz`;#==J;C^1DU&JAWp(?E*_%6hsdxz|oW4tr6T03c zmJgIf2|)z7!3=C9_mB3PdlXsgcN4{oIyCcgicmJ{t}3N7o``OHx59 zG!i}>aVC2;0@xR=L)c=B86G*OhF#VB7q78gMg1{M!G2H8nWH<0#QQ%DioHh5v#Pe5 zZ1^>QGM6&ol8l?g{$uKy^WjGsn_XGhY zl$BspK20Rau`{v8f}iAVo(4DN#4r~go{yf-f5l!LwkB@(PLqez>iGy6 zN3;?f$k4Gi*6X4xNm`}O1^=8(@T*ZQeOZ;L@A(dK=k)2DerY6Q!$M*432kXa*9Tu6e zE2q^Xw;hMk9X(?-=we7p-!~wOX_JYJzafzftK_bon#37h$pZ0OX|BY79H+I)fD=r( zMo_33%>Ne3a#@jFQg=T4)~pZazYub9Kf1{Am0Q?>d*|41*7j^c^He1FU6$l@-Nvs> zJF1T<2(8yi(Q!82X*!0#s6Gjlw1X-g>n7n(_K zoeCx{!joi(ydrCZ#*>&`li4e?mAPB{j*(Bc7s%!tPq~ccSVQ|L z{ITvM_h8FM?r_pYvVXXozBg|gyXs2>J6ksqKhF4w*Y9y=UM(~fJ35?X?>ot|e|ukv z%}Z$ZS8@tTV)D2tVfiG$SVVqK*~r?>UJTBEy4kSRO`KzSHm6fMo|8AJAWp7tNc^2! z#Nuf>`}2o2>)JV$-lCUIxXAb9iJT)jn^Ym58C6WAm;4|bttXNqi%aD0_G9FvtB6}r zqt1C!p(IernaKS#kX>ICBwe zx*!LCJMk2UeOD&Cu4|H(eK)Y?KNY5V!)mVTnl9&`b&2dl`s}jw^=w=J9ad?@d=z)C zgCyAOCJI+$S+~ShY?;hy(Vd^#ESxvwc6~JDv`X%hftQzw?Bp6g@3EN6GScVdh3lYX zg%h`Pi9b0K-VT=!$8ed)IU?;QBu(#EabN5e#HmkT;h;~vXRLcPpCheAELYqkn%Q|| z?PCYbymu$Y7o6EoDUoP;s~i&(=*7%mCjvdoKTPfOVdiqbl#TXg2l#rt8@aUZWR+2J zy?M22-S6Y0>QhD)vtCkG?5x~8czWChxlfk^Z_5J6lGTD|Ws0!>jU{`qg0hJekGDA; zGTNr=o`4(toJ;PnJ5Iu0STF@vp-lNqDH|P~m+UAFdp^I#68dj1=QNie;bzPfu}9x7 z!JQ$oc!s_%@f|Y3u`#PrhCrR1-J3}d3>4CLJ804TkrBATu80&!6u`uXEp(sPNu)4x z5b12c40E?WWw+ef11(V<>>QnMRPT!=M5XZ#TyaQ;D>gEO9Y}+7i%p5w1SMw0-t&y# zJIp*Xk!3RL-@(n{E_SE;7!;s)kSmp~A@8dB%#O*Fjbn?kjk}k(O{U^%WYADSZ%9j} zD>Fu5sznA)1k1vnA1lap_4i`+&$;5FiXzk{<;0HZmtylqzZD6*gD~JfMD4jY8!U=O z^!!l#_?K+M#)2wNZh;*MI5q)Ka@QhG?iu2j($bKZEla=uya!j<_la8?BUo$a_v}#S zOwMUfH@T>-$#t3ekUJNHNlthyrll*%-@A!CTlgkq_^08ZcjHKFtQpr4+fUy$T?iAL zMB;5ZJFt>4k-7Zi5EHiII99pc!e&n1fQNs{k)f7Uer@q12@_&K;SUdiHl4w#y~$*P zwuCU#uZ_gURgrk1iaB21HUsu0xQlVuOR@2u32eE$f~agI&9)T}Q$wOdxG{1DcxSs1 zzszEMhN*__hf>jq_C!95J{?ZEPla~VbatcTV)j~355AnJ!OrfN5sQ15p$&=&99e6{ z1xwQ0@)zv{429&Os1#3sGZ)7fjv$d=C5f?OJpNK6i*4^@L3h_>w(EW>HW(boTr;$; zYq5MI*e@$dT|8hyj0wlu$6v#HFpvdzBI$SgEJ3l0_cX7YhI6jIWq&}oa_PStm$?pd)q6fd((a9SkX(Sve*c5t)Ym@GbN7&3NnTZ@ue<{bTnD?G_l_~&pbyeI>2n?;|kjHLccDUfS zpvYNE-0@~94EMLOl1(3A?V<5x-f?s6>v=@{dfqy5U;9M#U3V-y{Ou^4fBYW1HBto1 zZ`Kgsf>tQ&=w#o0yMcsPU$M{f*5aFg4`H<@@%YejHJ<%lz?w}_v{{~{Zu8lFsmymkP)5RO5zoGT+g!XMbN+REVCNDAu zano2e=F~53#^;#}u{I#A%!NbbtMqz&Lt-;_msi5}Dbu*F7k1o%;VfM4U@X?(VvJIw z4aM&I&0@a@j2(LRu^-?4VrNMoAl6b@_4^CsxZ%tm?!k{T&cUP@sl4hYL3J`DK5#xa zU+Fm4HE7HE*SU$ZT}s4nFF%LD2bgX9x0~F(I*<5Inap)kY22djoA}uZ^@bCaa>M*7 z;~N$^vs`~}E_b&gm9w>I;}*WpCJ!GKb2s1Fa8){kWRcGl?(bxd(;Sz_T^WkuWNyTB zH=@>a8@4az#vbQu4%Z^cX>up`(>;QFx+j&ZFZJZcNhxrxF0~+Ba-8?`A0bWW^0_}%vD~CB&N$px0|!>< z;fxo#$h(MPeqzW^Z=aI z4S>|%XzrH8HcoFyl6&s{g-JTW~V0dYkwpvP#(>7SmttT19Thm zpX)cQHP&u8rS*-gQMTo-oE#v^QLip*Xym;1FMfD6q_CktGy;CqN0w&`^ukM~dGOmC%-%0DS2GF+j- zW8&C`4IeM7bgx(-^VI#=pm5JQL%R|KIKqy!87 z7Zc^T>G;{WWHA2OhB9ug;11STatl3AaxyCd$VHFS#Bw&n{j3%?q#Rdku>D)jUGqCY zyLa{?lT#Mr1#bhma*Zr*mUA#?dtR-<;e}4ak?d|x_F@A5=ogDcPbQ){s(PGS(++N) z0_OTbgN zXmENr(nQBpLISl6i;>rFkSYkf^9ggogW`eJ&+sN=l^+B^r=~ z$Q*w6_kXL^TCMe-y`TL&p9j?M9s>#N zr5}T=8(eXI{2;w_+X?gnFM>+t7TCW|2g{uONRwx;hCq!XBx`>QzAg5Lq-Ba^eal{M z_NdwT?xXF*P);6ptlNq{b{xiwCR~8v7rWTqAp!7rUq4-cUD9@R=L9TFw}4^$CGXuR zWxOJA2zsejOg%!+(IG)n?8{4OMrb8!6iB+4H@OM|MnuqIbFNY8i>@fYQlEF|zbE-N zO~JK;iP%kQAHOGk0#2TG7OaXkV;5sx_+DKNwcA&d+G}fwresF7E68F!f2Xrl`$+f~ z?g@7J2Ka?b9NNF=1b^gH29&4A0k18&6)lgaO?Pe4`w4Y;tY;x!JnuDs9GeqI|0!7N z#Q}^&N0H+43RLO3iU03%2Xr=d!n1>ktf*9ml~u~a@n&U6J0Z#2j_n7pv>0MtEk~X0 zb=ay&aoF4YDGX^Y!iSr~;C#YeoN*$7Qj^=fSG)pzJ5Ywp2IrDFl@ToJK`JY)>BG)f z-cX^NWS@F(PG`-Myd9=P#r2y`vY(|_!T$JBR4{Bi6)gULy+8#R@AJZ0@;~vLuGws5 zj|VEf7zFC(3GjQW3fr)v7fvbVu>s|&?4{f+W>b^|Xk;nWI;=-M4I!k^(1o1*^c`!i z-iSB0H&e%h=B!w05**OThGkcdV}Va7(mybjnknvM`D2gcuAX8DnwW!wrrANe-eP#_ z`WGC!#4LLTK)>EGxV7@RV2f9TC}yt&pEI@#d!a@!UlI>5oQxn*(~R6XF9+vx4nbaa z6LgiTqFuFm5G^|$51(*>)$^}m#;bVzWIl*m)s}*#`xSmsvjS+;OQHGuFsuylU}Ix~ z!Rp`yxE&FWe-B@Xwbn{{D}#6r7A)d^YgE9ex)@3?_JaQ~Q&wV(Pgp>o^-}tMlLibup@Hwda3m#h*=+VjZE=Wl5-XbJ09Q{+I9FTu z!6=6twDb8$tTMHd2JTS9qb%2;Q?KOU;i_?>eOgX%tvwamAD-o(mzBY~{z2NYEeSt# z%LSj?iNI@(q33Rm!KME_;pUiHv#@woc2ixS#7TWbRrgfc`Q~c)wDul$-oAmj28||a z9t%nM&1W<-*bM*W?qa3*IYcXUA#tm`1S=;#MhV32)xw=;_htz%=Jjbp}2wj#O*9=<8YXXfrg`L>$%NnoAvrR|+i44D)TFo25F6Sr-2b^NbA;K+pQuxB5KTT^1o1a@A?SWCWZaZ=Ze%5v)r#)o zJiFUeJK+U3Z8a6{n|_}3#K+_A+ZouH8p55x@$kLTT-e>wNLte5go;xO;EwzTJj8fC z?dy;iuFW1UbRKn+tPWbmY9Dzr?5Pc?V+dhN3CPG#F~;Lk=g+hW*Pq zUVUi{8C$K!UU}{&Z?2vtLDDnG&5*M+`c)jDaSuS%HV(YHkHB)lSmuyZ!9`SDMt^mt zu*WY(^1Ev_(TAOrk!f8Oovm5S6`h!j>V&R*_9B150AcG71KKL!8H>$o+u?xSmlo!GhY0#se^ z=H{#UVdSxczveCEzC{~w*|EiTZ>+x&*Z2?E^yhf;b?$3=aaS9BCtE;jXB1m<<{n7P ztHN~MNucHbnT|9(L;qBy;XuV?{=kUY?DAkg*}kousPwgw2ayrvO2rIl_ERD`KQ5uT z{6=bMZbQd5PD3$bXPE5Uh4#gDfcEDCmV7E28fI>naL+r~Wp|F{9ytfkvIam|R*5N9 zyP*9>p`7*2mHcEEd17i~j%PK#feCA@Sk1H5Oi@>b&A6h)kVh;hC7JU- z35BrB@HsY8evFT_-)GU=U06?PD-^2Cq{|)8LD#Bcw9-O~j6Yz(cRysZ`{K!={BvaW^9hVv$m55-OR@W_0=gwm9tV}>iO!A`GQEg!=6m7_ zoX(D54KIIyN?QXxymcfFyiMRiX(T@1D8n4TAA|aEe_T*Mf`60HhNtWt1*%s|VNt<4 zX0;}dy*8iC7A=p&vA0g)e-nnXU)Lr3%yKcLJ|t|=;5xif3Il%CXZme*66^Z$mOkF@ z2Al3rlia@LS(x=A@}{E`uim3dc5Ho5l2Z?%OqnVWjw%tRow-C-UO&w4Ej=P|D7PSQ zf{O5=;>X}tvj>(~N_6032XTEJhQO?Y^jnudI(qpu;ywQe-cOGs^B1q;PW0-s^`wVr z{t%M}gB#@A+z3$BQHG!^`mn_F0(a)YWO8mqJ_$^ah2{40k+|Cy7-h;IL~a3%su-Of+>Q>*`}r$F_EAt!{%< zqS8dCFX-chvnA+8)LFW&?HF!8atTd)_5wNIjNr%4N~C{oEr*DZQ*83+TDV~Ui_RDp zfzNgtvKj0wezrCfn|;{Dh8SF6-|vlNz3(~NvbY6?yq7>;W&3eTY86@TYK}&AJ}(XF zUP4oB9+cH~)3&dh>96$kURITNT09k6r?ok~hKb zj(N1KUX~tzH<288b%gnKMoG+{(ZX|WWu$sh68Umv1c`j2!s@3wv)Du7Y(w`a`2FJy zj9RJ4R&?c()I+bRoZ}EOef%B#%IQ6A-sg@7wt7+W`aktiha zCc3#J9QTE~L(G9Oth*x^w#H_2eXX`o{5yFQ-+UkM{=5}OzAS|@kCSxSfIQ3h-^UW(exuHdrh;Jm zFgzeH!m~A@t(>*51M!wqKr z<1WP0rT>p^m^EJYo0CvfYF4fx3e2XcGuHLSP48% z`W#IySiz)PnqjW^AYP&I4r^>Iz=j`&;r5;Ubi6zWy=D7+N3B7(mR5;nzhp#s(Zk-Y!0+{%GiEVPUKgnui{&yp7Ccc zo`QrYjqsx86ju1NliywUn76-~#0@Xhgaqy$6?6j%J*7?_elxmSKwOML6-MB|L6jgSGDu#jkv(K+K&{=)Rma+W{#|nA*x5 zC7m%XPJoX5JOI)PT`0}Io{p$9gx70y*#^Z~ENSR_P(+Kc<@t4N@ z?(qY7=&K7T^pF^=o9@EinfKw9emr*Cc|*d7I|u)oVsU-WBy9TeD&|zC<22u3RF-`S zQr%T?VXzgycG6jVr)T4}E>m%9b3eD<-;D39wC23$ z#?l16?+8Bag$5)ITNQJtZhE5RTkg(2wU5Rp#u;F@T$0{v{1?TZ$F3JK+Kh zXI;{%G~vN$Xc}70U1dwy-ZCd!y7@m21! zZ53yyUCCZNiDErDg>03CPuvpR0>ReTz{`*|cg>O2>_ z)ttuHWtCvhgnts&r!BId+Q*+;^q5{>KM}S&7}0q1m-y3;Ec_-AvD4#p`Md8A^WL$q zcwd_m7~O@rwJYP%gTDC$=9t12Z+Q~FIs%VVyU8>*ZG*a@=TVl$GO%2)%Ii4#2cvF8jm1 z;p%WB8_qU2IiW_i1DwGaM2vQ-ldE4sL1YD#l0Ff`r+CK9k0AA6C+`0dj?-y8+Ff>2!qdJ(FHby(->0^tYxch(_QyUJuP_H6 zl(2?BZJCK@@Go(g-&g$NdYrEoefC0LA%0yQb()6|JiUd!{sRSYTSb_>H&Du)DqGfa)Ru=TSyE%rjjjZ&Z3w6 zJ}5n9Nwy|N;^DfQtl?%mmUp^?&3yHk;OGn@og0YvESKCiEaibmZgjF$I$iSASaegu zc>Dcw9hN_EjQ!Jn&Q@fpidX)8M6*`K;R~m{ukk zSaS*eObvwndRow}V~P`=3}Hv4beZu_D>NvdK$R!h=(gpJcJ$$$E zK02mWhb+s!LTBs@CDYnc@e-#m^rvVaO`dxM;zr1T>ib+=>E_BFwu&V^!B_Csv;p3u zU?-a=xP(5P@TX5U?7?3~O<`}+JfPAu7MiV31DMH@Pu&&t-3nQ@>hw4`x499Gb85k# zcS$lwmA+VCzZ|EUOaeLUR7?%L@Q1<_^5w=Xy5OrHb5WFqrwKENOZjNhG_DKuZ7kTd zb353WGlAGM(u?Vo%^(XbX5sOtztc+b8~(@G&E(2d3+$Tzjk_Lo9jD$sheuwwKxaLs zMfn8r(WwuFB#uvzQ`UbgWPKc6k}UX+P9h{CA3`7<=LdLI2(3z->Qcr*Zc+dR3**l)mS)Q3j)k;Nev0TE2v}t3={=B@)doPX=Gz5kn9g zfqGLXVV#RIWfzjaLtyX?_H`@Ot@lE+Jzr zQjAUjv78O=fA$W0ls5?q&h6n`a$X|~ZFw^OVJLtVd(f-XufS(w2ESA0 zD1Ioj4eicZ&TjiX6_}Skgcg%W5dT;X9G=KATPzUw48O_Z)ftHMGlYNF2!&d740*LT zgcN8i6X&x(FmHBF!cjg9OXI8Qw}5yYV{(I2l^3ERZ&E4I`HFpOpYjnV-Y|55JsDl2 z1K$<2#BR%+SpUZ5IJZ9<)s^Tm%kg)S^P2<0)e+-`UAc3F&5Jx(Uic3%{N~Aw3MLA* zG$^tDF_XL;8if~}KY{0Mx`{W*?}dt-RA$*)3DaNJkcw&%G#$IZ{J&|kOf?UZzCIjg zx%IK^RhQVbT^a0AfFW5HC}Hfy{sMP}Sa=yE;o5(%#&_TFg!!H!JzuQw?irhy-KuU_ zeWa4F(MjPI3r6EFg44J{I7m{X^@SE+w1q9zZxL$X=|!t1Dh#{@AyFPgOL-IWa+4FY zm&3*Ntrl#ho|qO6A5L_0l4-EV72@;h5b1PrWOLQ)QSG^rB+M)anemb2TqRF_we?8& zTn)g7xYO1jr>Omh>1sW5$pP zpN)u;$c=oQv6=*Dt|jJ1_mRHRMd}zGM6a|mvhLPBGUR9odaz!TxEim3H1&7j>dolf zvYU`6J(6XM?TGO6We{yrVhc8&hNpEkxT86peo|^H8!*Y{Q2r#&X@#7yc3Le7d!j1* zdg3-tcy$=DP+MgFaG|i@Q%QJc;0sY0dYmL#6~I-G@yyfn2!3_uKXPKfEZMoFh51eK zVrj|U(D$yNJMu1+%b*UTviCeLsa%S`X_fG~a=GN-a4Qn(beJ@Tm%zoyR_wKD0a#f~ zB%2G4LtwlPBrKsUar}Q!R`drnqh0X1%M;0#X@8(f#z}m}N?NS+?hae?cP9N)tj^Tp z^{C#FLQ=7&nmp0mOq9cC2;XM83E3h8;lrdNLb<7AXJQ(@l^KLjtPIC(Atn$M*8sa$ zt%Koj>rlFMG}$=$uE;%Ao*f#o0~<@7Kwn~lcaMGC*z$#^GsWEg;Own0=hn58o3s@PQEtkYyPw?9UGs zF8xzZ7M{}~MOKk;q0kUS&sWkVVfX3X5(iMtO;-#@^LPshMrtPejf&xD%T zHAv)kivR4V0kZ2SK+f^Q8~ru#N1nrxZ>mItx#gn67libmYcKapG8>25 z-4_i0%i_K6UZ;20y``b&+mPt)1bAzu1Ro4uQiHj2NSHbx`sZJV@1|#Rmwgc|*`~n) z>r=oetPXO6)Y(;!i~RHQ>wN6TUhdY?kAf|aexQ1}V<2y$B=KcQ7DOX?roPLQ)}(gO z`sTHGOfrwW9A{GF#ghF*z86c6zK&WKKLoE^QuyAj4jMe!06u+|;pfWLg4VO!px4oX zEK0*@j{H#2$T)*+Qf=@U6&LhoZk0$o7mNHRFW??LEuc%Y$Dy^xab>$#TH&tCnP~n4 zT_${-f}YGZlk5bEXwn29l(Y2?m*$oSw>zR~)anUvxBmlje%b}`r+>o0S_NjM_MXow zBw&6kSn@lLV{I*cTq4nOD(1WMac?id&_X59EPs!rDz5Ol`wzk2jGeGt#}3>MEr-Od z?OfF5RrJm9kzDfY_w<*MC1k~`;7yX*Ce!~j8Xfl&{nxV^aO)9?E?Pr-pc8&M{NT6V zsHWf3OZj8tRj8xj0F9`w=8rwm=h|%x=sx4yaQwwu7;bb4)(1bNSw_a}@2H_HzB-P6 z=N51UhsUG4qq^y>nx`-$TbZ3q8pXMM-j8a_uA=%2rZCC*9UR*D1;R6Dqj4|8@U)Rz zvFL&szg0$s&bzFF8peqr@l6EV_B8;mKY1c3AGBZ-WYa(~iYq~b#Nng078`Paj0KAjp+#kWwjl&VSOBzyd#S*)jZ3s z8>2~GtUsWxQUB4$!IPl<*9eK<#ELz?9nI3?eA%kepHbuj3nXLmmJf>>4#F&Bn2mF& zp?|d?;pb4)Z@U{BBFypE4P&`dukrBZqYA{&*Z|V|K11`*323#wkPC7$rWrfuqbhAF ze6VmI%A>iW?Q3_TSF1j8H@Bz}8LJ3va<>W>Tug>diE8Xn-alxaM(LgxPf)$*VQeR= zEP)dCZ&3)EBhowzFLK@%sz>a*#=OlIV#-j!9Y4bEE$FC4sgNEKwW(b zV3}2lNa2qXn41}}fMFu+Yu1GQbZy8*VGe4MI)hsyf8zTO595b1!`MKMH|xJW1lIgW z04>K^tZ+F8rTrr4cWYp~U$oGt@tqQ1X*FCKKbc+%e#>n!{e)bWbwGYkA&6!M;iRSg z)L(5aDD6LPS6bwZy7n5wQj-9-cYi8pn&`&`>_36bCH{$nCZk|c4wlN;I)4v|9w{n%XqVs(|>t_w~!P78$}G^ zL%(RklJWKYw}FAO#et&*^XJY$mTyOsm=~W>&mU>(e0M2 zhkq>&gHY4`a69%L7d>1K-+!BjTNYdgJri%1F|~}XSs`L7wJT}YobUXVY!kM`w2HRr z-lQ6rS25$iJ7LC^KQQE=ir8CoI4H~PhmbXhnO{8*$M#ym8=J2X@YI1*E!obQ9vFhY zeS9q1)>qGGIj*DeXD-3h1yyM3A9eQmat$@u5kbEXQWP?G1=fDIf*rJ0Vp^k0Xs+N2 zr(QV?@l6kS51ACyn;po$yt9I5(*e#Y$TRaNfp9=kGCxgK7E8O#qfX6Z!F1IH*xi)~ zR{I8E>I*fdf1J`Yd4}AqVGHru@nUc*ISGf*2NpIbn|{*Lg|W95(zm{nY{>n)aBM{w z7&)APKeCfp$gEfF2QS&}HUcj1)faDeqx7=!7TC1w63Dl-!*{1B7@9a1J`@6uNa~@t zgptVi{vbqoeS@;NyKI!7q`%~v2HdmGVCOa31>V!=p>rQEVnxeuSl%ZVV&nhvOzJte ztb7zq3cL;lw>zPwID?H99%k|jqS3KgH|a^mX7-NR|GJVh^*e$qb5OUI32 z35y<54aZrq>RSeLX;NXkC&;qjWDFa8H%OI07xR(JSX#A!hD!1nKO?f4>n~eA>+F2a zAUB@Bos>-vomm6NrFX#FZ8mIm&rFs}ez5Zf?pP3_%GO@$Mt<=p@Jdw!uF4{j_pv(( zZw=?LrsY*EzB32b9!ZA5nvwYEc?0zEWE`5XE|1k8bcC$T`9L4FvDOtwsA8LdFZ*#9 z8X6vuy~`^|mh?KZJ^Ukk1S=RRif32M)LCO@5K~a;#mDP^qh}!s@bLEny!Pp6SWI3+ zrt&7D={tqCM_t6WhQ;_^aTM#eknsEeGX{TyOmQ%zij9Ml#Ilp(sZYv12-=|`^zE6* z#!no8v5v>t)w)95H^PtvWd)J-lbi5o<7nu)ox${qD%foER(8*&n#n8VL!8Pcy4fWL zR%k91%&s_tA3dHUxn0-7RM$3<^YKg=(RZ4EQWgZRu9|3Ak~UVA^cBk~6{B;P<8a_K zNwvd*NhENyJO0|D1PSXsU`=NkjDPbT;=R?`^p|VJv(Gt;uWoPm=B{u* zHIK}=KY(Sjp5n!J&g6#LE(p6Zl%x)J;@U}>kfADtb6YI&)eU2a^Rla0_SQY76r0ah ze2Pae7g>;}f3?W_wQI0VS|_Y3dkk->6gwHdpT>Kqh?PPXh?hM+%!-#-k_j6_iH3R; zPWta5^{G???d9`W@-rQ>erh&(abq{SU2_HJ8jT=V0}=>q^k@5C)}UB56QbRc%{9$D zOXnHAA#a1WiWl#45HDzR6Bqn!;~QLtqWKRO!0DS$$dg%BSiIK|ymhsRWn~_obu*FN zaXy9Z^IqeaF}diSyu|A^VjGl19DsxC7K`6Dsf$+~2j=AUl3qFdgL~I>lv)N}BO_}K z$cUw*h4Vg!OKzA-WVzyaQW~&^4mKL$@PJ5?xy^@2AB-R)F2!QKh<-fYG7H6xDZ{a+ zOxWJ0*LZP|0$IK6qWI+{FY%Oh_TnMK_T&7A-$i-v`l-&bA>xAEU+idyym;`Gv$!?U zS-kN0B?(`^oxRFaCX^0vP#>Lyw0-c^e)$u`Ye#Exr|K>Oq3hS5{R*^Nk?I&`I2p!S+gtq6nDyDqU~ zqRTAcL^PYYZ6xjLbfA_)bg_%wc{Zuu0c-ayg7=AAA=vQ(p7*^IeMv7OPjCm3m3~QT z%`@=lQH$8hH@eJuwiS^vtH)V4GKv0-*EryhEbC8Ev@A4#bNLV5dc`gY9mdk+qgnU09M)B*3<+;~sVkktifm>wweSq!Mm!-Z z?!LtE`2{kgS%thWcq-`-Jx0<}Dp01eWIi+WK-o}AQ~%te@^-NjKD8ybH134tb*uOU z`5x}m^{IIN+FpwGs)$deDv33GZ?N?;tBLdd1Y*(ELK44wiHAkTihs4w6JM~N!Mv|K zvd>NeCjX<6RhOHHC#1@WlO$ez)4&n<;r(iwyXPpx1l6+1Gg27;^&Sjw=w(AR!kPcj zEGBC>j@505U^^o{#2b^pu{qxQ_?K!rIO$2V?suzLUEN&qe`yoNGoy{gzf?Z4Ui;Im z&%K=0eUZhNa^9kKFf{V)cx0ybd2pS|#kWqvLXS=;7e;>ZIl*yp?VX|H5H?cKuI z#GQt$DrOTlEgvV|!%Ywujsli1y$>v2`9RtpBiQBh5&8xrn7Zs=xYwTwbx)_!NjW=l z@`-=oC-a029Vd{;4T-F2ZzH&!9L2Vdh=HrOjJWCV+p(8LCiT~u z%xt{lpnZ%q>kyx$6%ml7Hiuos7L7Su}C_jXU zZyA6K6f%k5{C{lPfkx&vVkDFr=^pPX1|N6?sEh{>3P8 z`1^(3u*hO2mv69HwNX^1p`4C8zmG1zB1I}21Vp-d8+mwrJ6kwv2D?+V2c3S>#^Qgf zihte6VHYli;gMBc^u^VyxZ_GH-G597O#?D`+OHJbH5DA&f9E$S&R@wAUae+VAL=k= zJqg>b=rKOyGZVTF{^Aovlej(nChkbGWEaZ3z~W|Eh#ytTv)vlgm~4;`K?(JH{4y)LRdoFm2yEcAf%Svyt*$bwM``1T{)hx?JtMYgl^VXY2 zTs6dI3vb|M-(rz?_cYeL_$V7`rU&=$)IsIm*8(p-7aP4hk27zMCAk(ivDy6`tVM?M zdpAu+mPc#Y^=->oPK72jn|1(>mYXrL*ERMx^$asIy~qmxD`yXEvf-ZVQRrM`34hl4 zu^{grru8U*84M)jbIrrp{IesOx^Oldimx%>bh{fr+~AiQNZJ=l-Moje{hRBMVwXCYb!SrSCWAW$)mZD$-BkSHE~IJ4K+0E& zLWwkPS+W(o-1HH#?x*~x)VF-*nDw+=yOH|s%;jd!YC=-qhNH3mdT5(Q3x9Kk2u=HL zz{2vbz>%=E+zy8#{6r1IZ!aH6^4x&+G#_O@^JJNw(r38v*9$yaov2N+7E!_< zX!e2?P#?>&w$qv@bqb@;5;vo9g;Gr0`T%>q$_p}+AJR3YpJBV-alE#A5?>Ui0~d>3 zVUwIK+v2%^4NN`2{>)s%JQ`Hk#_R%6;wzxZE)dV$YYTOMM7(y3CA2lx0MjrcjSauB zc#8?q`p*Ddrnd+$ ztW99K(yLf`#X`uRA&vTdBsb`7MR0PrKg)1S0EKJYkhJY1ZrA$|7<=vx1jUMAY()aJ znn~kbmkf|knKa9;2lO%L1M-<93td<0_~NZofqXnkpLpCvs&|~3ky!FGYv$v|>O`y& zszJtlltcV%LvU^~q|e^jgL+IDynLX}S_ZSYA8*x|=IB;T41Jhhk}7NW;6OB`7|&Cm zg8kp0=NB)sXSJ>Cn9OhshSGzX%A0n2&&ZPH8%NTjC3282ng$vrj_BXuQfO7xLS9|o z;GH=fTb4HSAHTVRq^*K~c5WMAG<`Vn-+737E%XAIQB0Q(@fNY#)3|>?nOR)<0#}GO z$Yk6E*M(nUWUd|@Ha*0AijTA2dQY};)(2R~+cU+^Im~BeCRT9NWYUY1@R)^P>6{7u z^i}vFI=i=*4tE%bWya5emf*QCLLrhVBQbaEd=K=|>C9sHAlw@}hvDlR*!s)~?6Ktj z+bQum+iRD?t=|}i3+73@Nc(uBV_WdW@`v>B>z$zbV-b$9eu93l9k8q4cpf@GO0u^q zf9WiXVDxIx0V>U|1D4sv_l=NZRhOLTQkO<%u-2d12qM{(W$|dv)&yAdKppE$72#7C z$AG(<9{00J15ZDxN#2d}B&JgcjQaPOejLUSy>$$J4qFKuUMKOtZPM{U86V&x4noJ4 za&ECCyEWVJ&LDw@K2mbQ0u(l5km6BEDvK40{{S!^zsm(5=_EB)O$7q=X~EpL@gwingN3 zq37_s`kDB>?|A--(rR#gJsK1&c9EWKh9r!cM!p-*Cwn!OahvHRoU*zdM{hij zqNBr*b?7I)C8m(8n&gDXAM}Nj?oD)qSea}G6*BbKbn-(eB+vLTqB)REPW4~IqvB@a z16G#ERx=SFFts7m_M70BUYGEdOD-UpFz~j=k3=gTE#|B;4EQ4z?bQ79L#!v+MN14c z%6`XxL^Gcl!^9B+tl%5PvCtmc_HsOT>eGH&c%z<9jCJ8==sWSxKh5D?5@rc*9Atu7 zzDAr{WGA=s@-&{^St7dgU=VkF4Zu&o1)+u`-_b#x6f_XnLL=sRAdk0)>6ZtG@HWXF zyTi!=UIiAzuj%%9p4D%3%7dZ0)pPN`4^QzUFM0Ig+IA8;XBoLQT#=|gT}~eCbii3# zTX0}*u*f61h-x>ER9z+gw zzsI*#+VHNE1;lCFb`p|dg_hb*#Erim;FMC2GOHe204?**j#bBe87V9MpZg1V8f` zf}JnK@RthjYYdi#i%3b)6K~t$g)J-0puVN2?6iIfZaRKml1D#`pXV<}mkgYN zoj*XA+D_%?{@N?>sy@Nxs_w)SPMhGg?o1T4?jJhw-5y;E+<+ic!W$dT<7;1rKznr_ zzW-_?>Mbrniux<@X7^(J*j>zRULYVT?z&{Q+ZR09?jw?(z7myB_>MNti6wVVUnVuD zQgQB{T)OeQ7EbpyqvgYvfxnz0_WgE=eyFNL4_i*+6Dm?%v}OU`_Fn|5|D%OshW|h} zT&s||YZNk^_#MfAyNJ@S#&RnLuHf}S^*FTkkm$nSdeUZrNV;nhHaN45&UAK%sUNUjI_Tu7_4D24X625nY;#zSIezfZ$&TV?iOHbU# zUrrP8CVK|Zu6%9m8#Eb`40Om5*GpJny8}<3zm=O6F`gS2Tu$Q+N(2@QchN<={CTxv zHDbADKPgsCGj=o0%#>c8m=p0iGmv^IsJr0Y*n?{qml=T@d@;qO%>T=ynd z@Arp4ExtwOy7`i)6VtHrgJiU^?-2T1`4~;O7=)LPlEeBRBC%a!B3@@zg-x~N=$0ww zc-XqdsJ(!Ddn?+K0C)OZ@eX6Y#p+W!$7)=du3n zi6m%m3BH>l&)vVtIKw#x^rxRZt-SP=KB_y%^Rui$>yZi!^g2RUh$n6gd_bEev+1ey zKFNE!6KdP8qn*ZS)IvEDN`9V(?jNe~=k8Z}k7aWq6945r3ww6{xGbCTJ%@jvwZS&@ z;24^jeUm$C_Jc}?q`~tCRmj9AjJ{a?p3d)|BG~pTrcC%}pAC$5z?*|}(GstJR8(_< zN-$1z&7m>u!5TdtX5|S2LdFQLz4=V1oR5RP_mAkpQ(CnCeG00b_?D~r7cTgtkx>>< zY>Qm<%TTs!82lT)3~e-(;q~0Sk^a7LKIldUpPr#e8)uAy*`HhJt$1GW+It2!%|DvA z3?Y1S(E#T+UZSV1mc=*zbkU6;s+rg|fq9?3#y(obv1Ne^*wDf0?10XDnru+Q75a{% zM3igI{$FbC|c@j;#a~Z8nJjc%_b7;tc49$~zmaTT+*>FLYM{@8{pjXkP3Y_X zP7ha<(aEt}kzG~|jq#WQj;g1zO59PDZa+%!+28?H>d)u8GaKp2mtRG zo)1s8W^RXJ`wZw(vrfUnPZ=VQ;0T)2DEX!h3*x6pGF8Xp+tIT7ryx9T6eM)Vmg&v~ zy2D9oQx9z96xMzcJpD6+w_P`l`QCm8v(&~gxg$%!#l4treZPWV|J#u1 zjF4k}lDk7xf*&4QGy^SfI!TjVF3|dl-MDPOg~a!JlWy!=2r~P6L9j0!=}gu_2{$J2 zQb~Tuc)mZ?lTDWRvWKyWkOLd--UHV(1J%_>qGJVzarR$p+|xgr{wy3r`&}-9_E$n1 zPa3i<7rr7Dw=k-ap-Qa{jKI7(7&9L&x+7&YEZwdTHdmkVf9&4V{qX~3L!uL@vS1BU z7+%7TRZ&(vEfH>R_M*=;PC$j@Y)rPsl*PEe=W5t=l(z9Xa(EGjTwMOq^B0uhq^k-& zulx#$bhPN?DL2rUh83W8;W$;>-6v`qnE-PHDoO^Wyt|Pl05Vyxa()e z#y-CTc?%txw8Bc3eu zNXJZDt=MDK{Ir7YQfDG4R`_MtvZvTqZmxlyMw*q#xgn`+`15X&RMSaLq*5jrd$fk% z8DLCfWAeA*zW9a(BK%IllWwb913A)l4WWsfaQvAtKe?C%QOy6#Au z!GuiC@{>R9$W5m1t^4?4Hzo@v4%XPUR3w-E7=1|a`*)haMbBKY??Sk3jO7)(^>ric zUMAnM3$r&Xb6HtqbBj8cR-IXHQxsrSHY_R2CT0K6(j40ZW#?wOmzpi#W_?s!j;=S4 zqG^tQ?c(C52n6d2i;pI@6bEed;3M3xiPE2G3y%E!!I#<`wX?BbCQ8~mk{XEC^Xez0 zxEs&ziT<_ZmX2I9&gRLVZ*=;@NI}lYBV1{F0WaenFDUbLrD>)f0<+Z8sZWgpt{KVp z^56gQ)Ft{dKjm&m*)>ydJAZzk-52Y8y0yxOZhqY$TB9|aPhEf2?st>Fz-|2PvPNxP zQSXpwySI~iMUm-4By6#A!4J&^fexmLC!4I(&;lE zmrColb7bmr!I9PB#a|lxN;b7!E>#kpDsekn#-E(M-tO$f;k;4%CTiWHQ#M^|weWw*e(S*s5Bv%N?(QRM}%Ic2S2-G!5OrYAGYT>8)3Ts-=FN_>P1U8>?_`!8ri znbz~iypf9|zjJfCUE3H{LAB0syF3vR{HNk&doU=>R`KAVZBs}(-?4B|kiYg2AF#W< z^rp0GafFJz-HaEhX#4zL+R^XJNn7qm;iyc2yMy^}e;Ui`f8+}i3+n8u3Ul~t3yej_ zyd7-XENqa%*=o^&swhFe#u3i`Tw1Yg>TR1lM`~@SS)18q^)%a7PLs2%YxroJy7s7T zB=@!Kf#Yc#{R6}JtYT$ue_W)<{=ZO~lRHji+2CyJX*P^s^H2}S(-VTrqayj`chpgS zV{e&@$0EVzXJa`NkCXh0e1D$z_d_RoO}Lm1V@0DwGDPm@Rp>uuP5xtuGMa5fOR@r< z*-cbkY3C85ZX*e0TCZ>#DtxeXJn4y+C2Q*D3BM095iV69O6F)(pp5}dxX|D(ZvF6z z8x_!lMvo0aJ5MPPyyPi1wjD+CE8}UKNdstA^y1YIN3-XzeQ|ee3(ixOWNk762v>F- zKN$NNZu{MYp=n=XLFxpA{m#-)!8|s1nT_WbV7yKCF50~#fwVt4Px7yNlBdoAkkN4+ z`Z5;Mtgj_p{cS5eO!g28I-G*v8m))>>eujW)6@Lfrm>)yeu=-RxRHNk|IIG_#~abF zzlYK5<=0`J(*=k=#kg3m81+6YK*34Z>90mVbUZ&A4gchbJ5)#Tg5p*@Wu*nuS)7cM zUXCT}>LN({t0QEwZw}ev6~(5z$1>I3zu@y5qrZ5;n>(IXioVAa0^_EU+ic^ z=VtzoqVw>l`upR!J+fEG9+{O2_kPYfXj4L^s8AV|mLf$DHJKs$PZNUUE5 zg^OgkrP48^V7I_ov9}(ExE{mb-#~OgEr{;;aRi%PTM90U+rmH5+R*EXI(A4P@cW1~ zayL;yvRlrhsp~XhUeYN1m*WLHX2-+fs6)inObBXeCXgrFO2Azkb76L} zQ4^Htl1~Jq;QmOQ#y~VYO=xmvuEj> zFWNBq;Q+ksrw>ooa!~V9ZL~2(2o=w*BSFcsD8M2Ojz>*JEgBurXmTCAbG4Y9R=tSx zEIyJ^C3QQ69(UBfaxIY#UjeAsMdE&14bA=8hpp^RkQ=R!sBCpFximhTEZFiJmnz4> z_&EuPTW}6}tDYj^4kxg4h!h&g?16SMGUV$zWvX>P0xd2%h{TL{qqDCnu&AvUlw8ck zdD=6e=1N1LD`4fF(-d?eW9P%z5e>Lze=G@-m_hU%fruGgb2s9d5HFIe!1R0hQ($IaJJj=HR!ilCnx<13Ati%&ym@*F?gvI2%_jpLv4 z65R4gIe35(VM|3ng7&`E?;XF8Bz#%m)yc za|UJ2*@k`(>7Z$EVo>{`0BXA6EdBP^kP0}QP)B7gfZtV7q{Ak-`g1kRlyF2R#<}>P z=6Zr_zQPjo=j4^S9AkK^5%7F^q1zrcNCoF{^TXalW2%7eCMuxmXY;W5;`8W@O);um zSpk;nlzLp&1rr@OR2$cb!_id`AxJ_jMS)xMoi!3-oAt zyC{+m459CvDp@(J9%$jdmuOv|N=CMyBz~J3h+#q$^xM@*G-mBa$#QLIx>zMr8y>~o zyH((9pSgIO{}o_971?@PJ>)_bJlEc0=&_wh@v zT8%Zbs+`1Ew9BxBh!TvP?o8(|F{UMM>v2dJ7acyl9Err8M?-a1RQBx}`nXM-4y-u_ zRK3-CidVOg(@OGacjHp1wILA2oXTScSuw%heKFnmVl!;}p#mJoM>tI$Q<HM??C-(}Nm>T{E4i&k7(7HVM$++AHvUu@foe zSCK-eZS9%WdGg^DYF#d5Cv{ZNTf5Q&pMJj$kBrZv@6D6x zrVFEVjzSUL=G{(r|KreT4SD$DlZaqD+sY;$5qPHN+oKvY74&3-z)7I>oD^Emq)E;d z0>LmBvA;fmKkcu8{pLaJ^;41Ne>(*>ZFDBDhqtpPwWaXEZ7$j@H4cSV^GJtk6nuIR zA)CeXsM+TobZ%1~u>75kZf*)dS_`64YU3bT)0;?EZw^B`Pe#dD_&?%;Vrk&?baZgW zZsfP^HmtdEi!Bs>hYyR^VDZiaVDptR=+xwle@dFe&5HGKNW7BP=dOm_JSk}MQ4F4% zvzyBadB!~Eui_Z)yT@KhFa!MoRiL@t1;-fnZVg=Edm`jIAPHSPxgS#8tkfd5~_$_h1ceMV6uF^ zkz)oSmC|f7gwf+EdLuI@AbvM9xKa7Wd%au8+6_coW*11W#0{ z0~M_}=&F(muc~*z2wfH6F8Pr4+d;|wPD|>yF^RrOeMO>E*U?$!8Kh|7Ex7$*HcW#T z1TN8hHYh%t8NVAK@SUp=#js}Xxq1oMO1FVaOYGsvw@*oQy*Sj;(gSzK7l9W>H=xaf zSQa*3!`b@1aMKhEXm>b=9NcCA?-UQgDVtYdzvo)8;ueFaZ?T8{i%;MSD1_H^1F~jX zf^3g;1v9mMFK^(gFnFqd0&LIl1PN!S!q>XhVA;=f{B9%-FFT#gUj5y~Yj~rA=Y<;s z)yY{biZ_O~rEiF};JGUkZ~`Q99)ruPyTKvN^XO3P5>kGK^r)vy66!rp>jBsETA_t1iIB>Io2hFXz4d9rDKPU)23T*sl?n5g06lt} zV9!%U==z@pQ<OkfUP& zqa`*%+^G(1hi1+v7gcg;7U0TlAYYWBw^1qpeADHE= zZeoJX;@EPHBiK(|30r9tFdvZ!$ai?k^&cAm+ioV}%Jk=Wxv?VN6l4qMbv?s%u7{xv zy$&O8h|%LOy1?zdDKMwvKUVX`UO0W|IwNM<%iN9V0uQay;mOid)Vc3s!TM#aNLgQf5kgj*MThA z3vf-p6g1Dr%%f08l6}M)ifcb68q0);tPXQWE^qgvu1Oux@;!=;E?5ukhV5bU z;$oOp{0t8849WKDeljL^l|&m)kmrWCxKn=Vl4BR5X_V(a$`(u0iPme>qh6YSdV3B1 z=bM9Cx*nk;A~uN3I7hz+?4{f@hv`3_EMJ6sf%@1EL!MhYaIzBc;0B}M{iVZXhNvnT zv7Ufi4&Q`YLt2z=S%&u~2N9WSN$3zS30x1Uq05H{pmc0CdO1@Oak6G1?e&t#DYpfu zj|9@bXDk^$GLx+J^@WV)BbfF_mp{4Q1>N!22X3p(z~?(f;Ml`VxVxQ056x^Ro6=_S z=BaDKS8F@bR|AT?^Sn^ll5&)GQU^8F{6=~D26nRw=GdLc&4Y$23*pag{qX19{rtoJ z?)=44Gx_*>A`QRqLJb9+;-bt^a{k^FxKPFn8zwFx>O!~3?};R|cYZwWxN(WfJ-kVu z-8CZ9&0+;GqjplDCqmb!i=bm6ibyqPC+fT^X%~M?+s@x8m~4{QCR<~qDRFvEj+tm7 zA%(L@?CuZ*;q64;GnJ$XI4t?gH0_cNMD2uX$8pDwQV7S&KvhdHRpg4&hrvQ*yPppl zR<%O4Vi(}k;xXpR3ndg*y%j~um{a~vWny!zfxJI*7Q*}b$Yxy;0*5~`hsK@*@i_>_ z-P%Xt_ywBFX@Z&IA+X=;7gTjoMUQ{CphrSW?9yl1BR{Ud|C8(r;+Y1hr39F-p}p|y zG9#4ZJ4~n7DDh2}DDba6kE6O9bHT1pmQeP&FWgX?Cg=id!~LPp>G09>bh%AClU^Z* zBztPm8%1-w>4Uq_!gyIbt?nTtJaG?x-MbZi2(3aQH?rUwHvo1yrNF~JTUox#6z=YY zjQ}T}g}1e*p$A7B1sywc^tL(%hOJzJzO1u_qQC=^5qVVjuou3(`;<6uEJCLxCb8{e zLB}@pB6oxG5h3x=0dPlqI|cHInveLe7;{W1`^{vUo``HO`4R^S~XjXc%wnj|GTh212qQ7@ub z4{FDgu)>~jd}qcJUT$M5t7$odeG+0xYjzw3UU}Q##fb<6wSEFyhr7Iz-BC#EKm=Kj zKJ)&KRk52s3g$w-3o8=Th+|Xhcz<<|*JkDaLf*2^(A5)%*z36xV5e~p%2*_Vau56j zKKicg?24oHZ6a=9*i#B~lr*>}+jfFkjsftaqXh5v>fgMcGG{ho&HaYIDKEgu`xW5H z?`obKa{&!)QNd5^y?C1oLP*<-PiUP&0+XS3g!m73!iA;T;73^~^4wmAMfcc~@mq1k zGxQi5sLo>VT$l>?ZI6NPLjuv3HLHm;(m*^J5%hNM5bK_^1DXv7fSszX4J{Bep7F_? ze`?{lK;ITVU|ewbp-$X)XA|6ItB%!+LC`5 zA72A5s@8K~H6}nY^$7U>-Z4IfRtiFJg4#^Vn7T!gdLkgGjOc zFwaCy7WmFi1B*Qr8Q=D1G_X|(Sx>HKU$jah^$Wvz^Lr6?=1?hvQWl{dq6=WF^&+(J z?ljzRz89+&_VFT=)R6=CF`l{N6`6D65}GAl0<}&B0E_byXoac+cgA=uJp)dn;QEhT zVde^!v&e>bwT(%X+zEI|(U!M=U<%jk@+r3Gu`70zas(lXC!oy96XY#oxk|i~AR+4< zcl)APMx{m+cGu;Q zhwB|=9^nUh{y5m@DC6Z4h+9jlz+c}H*7%<>%3a)qDmQchZOd28D#n#l>=n}TAX+!_Ij)TT*pk;^P3+^RgP5 zOu34L75Z55(;m$JC#4Pf0Y0#`b64xn}OBj_vv1G6Z66}z)~Bezb#DE{zm3c7!{8wSq20R!hMBePAr zP~?I`XtSvo>b`Ip8n_*Tt-lLVnS}$|aP=CyYsNk>C}d4I*0nURa5s|sC54t6 zs^KMRA#6IciFpZ{>n49<)N+;>TN-_zqtq((kMecUdGepqTNh8#g zKNmg8KLosjzQ8!EL}af#8@<#M%p^zD?f%;!==2;a;_g~1P5e*S(x@jFaJkWWq+4ph zZ`oB&pQ$fJ|3#1Bj@7f_oOj_sQgbn~nsx=5rWxBks=Gv-**k*VwJ9h{R)$iE=FOs0*i*i864r{2SXAA?Urg63R*|Usk zGmSeNL!ZbLz{qz+$oyqGI(Pg8O73|KX3qM`_-EJQdA}E-Q@niejH^eqBxa&*4^2^> ziwmmUasw^#Txgd&+t2Pok0Lvw~gXW&?ixqi8DZ=!bS!t)&0@kC5DuZDjZBUR=0A z0Tm{Qpu=0dxKEQtI9H8?xRN@T;l1E8bX$Bk>RDO_`um1q_}D1?K4>KPy-SfPS?}4- zbY~#hrj7oa-i(x1bwkIelVG=L8{DV5ixxjQK-H_))6MgK(w@R9YI(U9Yz|xpAGgZF z8R2!L{qz?$)$|BFlD!_9?aD=o(aX`5(FG{dHVt0ZRVV)rttK8TrI3QtDYRt2irvg} zwPelTPt3W@o+(7*pw=$$Pm$cyrYBnY|Sue)lQq2X{Q^jZf`NneuCNtR3v zu7)oQ#fgnW3OmyO8su!0;D4GiPKE2NKI}R~-S@TaEa|Zz3z7>XXhHeMHYJ6Tb@4 zBfj=5_PT5hOHE?f#b+Xjf0!@SE3}8hI}9lImcV~9ogtzRgze%}SHK_D!L&|gJNbCx z9j=S}hEFMPpsH7c=<<0S>d$cisWL^p*o3oSuyz5_Y1<9GJ~)9&^8zAgzl}=WUn}U> z7lCwrEu`v_K8+M!itXMwfFQ*p(D+~>QQK%ri}#)r^g@e> zfq?CtW_g2*?w(J*r^_RsKsT7lNh3$6HIndPYm$`P1^%wCV$Me#Czr~l;E3ux=<&*g z7@fuRTly91TGvA_-jrgteGdjr>Kh=NkVxGdHq+zo6_oc>hw0zDgJy-w(AOS!a7~2~ zsYBWF0hKs)iaW6_DI-4S_nD_h#z>-L z0{xOBLtD!A(V=vXDOm`m`Fz#+xFv;En+aRS+ZVPRUW@D@|moNivC+jRpEaLU0QA z8c8}?4&&SeJBF!ogk%aBFh#3qo_z%=4Y!AeF#%-d7!Nxc>m$no`)X`v9!B z`~dWCN04N*ANW<`b|`F6K?bR(b=;t4RmhTzZQ9TG9$4ba0 zBQavBvjgZaeFYmnoq$@dfzaqk4ouvBAAA|ggKi_C4a4ibxy$AiV}>We%vG#|{sz6g zj`~edP!Wgyj}2gAw?0Hy;_za}0pMRa2z1+};US91NWyo}?3_d8pSQl%a?7&C#@tuD}OryeO@%L?AVyMdCHE!h^j z8YtI|f$#Dc!IjOo1%8=2SpCi($`$0nJli3hv3eF3aj3>FNfPMozE}A3uWs0=DGD1l zMlen%!$7O|BKR5U!TBT_QKAL@RdK8G&2!Nl0H2;qa>>rgs;No^C zyjiFRH>J%b!hi2D7nyo?XH7922s{bDO7)N#w&%gpYEd;U=Ou_*~8ndCPXx@ z0mlii2Tyw(!GMY(oZ(RbPkr72^^OcQv_wac$peeXnavVtc6=7rJ@Nvd?{|g|6H1{M zj0DcGw@ll=qS zaE;>n$eMvNXFV`~t|ClbtPH0uS`4QNuJf0-6nb8J%pC5#jSFh5;dGn7uzXY=c>fT1 z@j4fRt!Yib-*pAxNjF2gZ5hx9{{*QiSGetZVo1JJ6PA<<_;okLNJ`LVsP}g>3{BDl z+|{OV)JvU6Gk<|Z#9Wy9?itR@k)T%_F2X1pO=1E^)XL4}u`=K)S`vh^|_-g5J&APyf^VN9}+J|K`0lFzOhG zq`T@;!=qPW_tz6}vwIk4@Lm(Kb&w!C7oVX!)8$apY6l2=?m!k?KvpkSp_J$;h+&RX zuU}Pk;o%%wGer~i#TAejnoR`lpF-D7ork8YR*~%uEhKl*Lg-W8M8xL&AnWdi)3&eq-W;F!QKUzK}kIyT!~OclRs@>PrWy=;e6tr_j~|)9~zLEXL8`J`D!rdWiA-j zyhtSmXHdp94*mr1ZYZK1Mcsh^a0B4= zCgI4I5+dE2gU2+KDW^G)EqNeB+Ws6RE+mm0^l-wTvR;!Udx-h6BZPv!M^OrAUuvk#(HNvYOA3_?x6ykkX#%9jGibSc zHVJLZKvIK?k>J=Jgg-sSqyz?$dXeMgPgWJ&xpM~E*_i~&&m^F!uowKtiGlah%HX|c z8bm`s0lk{s0knUeqe5fOH1Wj=5-H_`TiX}VYv%;6_zmlj(n=R1^z=0lx`FYC$UVZ{ znTj)4^O43cNo+As3FS7IK(}B~+S$L3SUj{5u=IG~K$br2Dn7*o{3&NAW*;Luo{yoM zs}z+wtw0a#9UyBqWe|mxpP*!a1AOV84wf}*z(wDx@craFz<6a3Io($SxphXo{lV#2 zQLKV1pIpVgbMFCp8tY3(Yg?(-d{xvlcNjJam@-*c#qr;1UF6f@e%d2n0uOjLvcV0q zg5B*j6w$Fz@L4us!bFqrdC-b)mpO%hwATgMu8BcgU3J0h-lxRE;S1ifRUJN6-ARpn zykN_jI=JFzEPCPP0+;`aCku+6!JR$XL`3;I-BTTa?4vc%f4oMfu8gWotEBuep@*A~4X2vIhn1g7@Q4_Z~xff7p@WfrCJ=M}~Omejo?C(jmud6`kcV2wmc*LR+zN z60np@mnf>xHDWtyLghPB+iHzAkM0F}Ae8+6PaM5G`;e@RHAa)c#;Ey5Hfmq?AGWkN zr}x!<6Jr#O)Lw+6LTHW{tCi?Pkp!WGVfN#`R$J3c zhBd=^EMp_Lv%&dMJ@@+eqa0=BGTym{N*mtvON@@$AZumxgU3I#k0~)0YEbX)w>_Ah zNd`S~cooUQyzAGW3V;V1;IHu@6DjS)^D$Iyc;KbcU{=}BZuZd!lAUT`_OLw(4>V$< zg^Ibm%9D7!aT(h`3ybRq9~QAZ6(vqr&-D7_oyR$Ag~WIg+I_X%2^l=Ms4I1hPMb~C zlnu<=>^9r`VnX%qErGl@#}l|qT~#^HMFji%-)-QosUvUZ>!-Zj-uBwc8p5$x72}>$ z%ID^8K4|M8Tfx1klwxbrlFVrycL0XdxU70;5m%xy1#dcZk-aWS8yp^n@Z!TO>+i&* z@NNeg4AxXO1N|)IHkqh;u^Gi&^(-gRNJC7H8Mot+;btEwj1y9(UlJ zF#ctf&TG5q#CxtX4QowSvO)?Cy!!f|SWkQ-D`NAe;iPCG%k6&3jf#~49Y<5R6g zYiz(en+@0ReqV+c7;M1-I*4cfUXE*SC}s1$DU~Od70SK(P=*utE}A!2e=&&)OytU5 zpWm>y;W6)|w=u_itvzS?j8k~_Cl^l1>M!*{?Gd1}_Z%x1z-J-`PKBcPJd@sJsWVf;Z(r8`l&nHxt1}f+1n|bcuJb3 z^;>NoF=xRmZuiM%&VE5rb7O06{hO~ZInlrUh~=s4%uMM|w!cQxaJJn7-m-6+4fn6j zVGoDB4AD>8T_)=44UB$kLO%p=aN~MoBI}75%(h9}AgtAKCh6?P3jA z13f^tNfKAaU6cE5!_ zvu~DW{mvKd^)FL}8&2=cw?5gKS|53(nEZ24=0#0@;;pib;|}C?bN4RRWxskG6G>}# z-mg{H>W34hNVGvar!Qj`*RI$NUmAAd1lMlk>;W zhV&HqpKgrO*QVF0cH4Bm=aLP4bHhM>X3Qx0+}Q@&f{(N73}dm>$yv~7`WeB@s6ghI z{Q>^XjSM*`0TvZTF;nXsL21xT9J)mjcAU<_Q)V0T9sk<%sfsT@CRdt1$(%!5@-|TG z{?mN;v67$Ee1R_&r%Q{C&%jy1X8^SSPG78&;$x)3|LYx&GuW+Y&||6)BVP9$_jy zx|S?fL6~~aA;s@x$nuH%xNfZ$|Mmq_{{6%*>JhGpFXex*d9y|m1T20{br~i8_8}vh zB|3xr*f4YwAi74Ih)e$#w!~MG{da8|vTYE7-=st-e~}h> zrxQkml0w*^R~FRFcP>2+{^H-8EWx321}h)eWtt`;NQLb$yu8+ilswd=t3qtyFW-N- zraqDl%`3$#L}i(-gK^}(_d25Z%M%C5j*y%B&NRC7IQ_7^n%#F=jyLN?I@c~qo*$xk zkSYfh(4MQa$i^x|+}S0JQ>zy8MKrU?-<4-!`L<(Z$#i#8Q*o8dC5wohs|@pVrxh_D z_>VTv?x(uL=KS;x0el-vRerUPDnCz~VGdb-;y%cJ&d9mT(iIt=WTZvF9T2)oR~W|A zzenxqjBWA!8(uk7sXvS!*xkyw$Ho%5Sti6)z#^Pz(&czR*T7nWS?`pM5%<80C1mS5 zF)DPxoXV{IO!yNjgtOPXepO)yJY+3GYTD+pHzTHkkbCQ4*Ub?!Wv&HPJMx=^33#>l zTVvt6#~C!+sgA~-bfAm>M9}cB($pdC5!qyQpKcai!dFi8;2-{Yhraf|PL=9gsM1SC zl52LHta%qiZXJ&Y>HIMq-!#Ni-!P2|MJZ6JuUd2wM-GiYK0q{9WI_?!W~d{`c#F?! zCvvLi$@Gz_Ov&GJsBBb9|Fh}A^?pyF{OfpZ-o73;=M{ zlsOuU<{s(B1JM%X;ft+cvbqs2YDpzuwDutfFJZf28jCV^`=dKeOF$_%oQVyQhi=ZN zL9tK?@-lBi51bX?${Wq+n7u$$$m_%L>_yt zqTHTKTxn%x0YB{;*KN^So>qDWaHpa8P5#4rr{Zbc88r)RqmHa(YP1irWNsj1-^svC z`6|-5-j6jnvWuvWm%$(33=ro;6kbp<3J=F|$?r|CiR6~8XxQcrdUs3%uD|O+{!06k zuDkEy$`o}p{Q3dhFqNS92bEFgl~jbe&*0S;@!;qVU8p1fx(3Mkw>aEVA#J0V!Ue1bYHQaVkHKT!<0Er``6T==T>ObQVWhVPnL>;}Q8{`Mn|h z(iM^rup9~1rJ)REbGU3#JbdfPg|4UkNlVEVr29Y-eOq}FbAuLu$`B=RZgnfPeBy-V4&AdH|Jk05+d4lIq(J@kQ0Q*jun0bp2vSo=U}Hyx|5s z{I(bK@{U1mrN>NE@?Tc#%{lf*(jLScd;;Ugj-f`!P^Q^=KTkgBAgiY1L<9Yfp}8xj z+ijW@0a7Oh7|V05+_I->NHY-Hr3cv9H6-eiw!;fyRoQEtC~*YF3fqz>DM07fkE7N~ zSH!7*iuNRHAhQ*MOII|S-)?Yzke+H1UmNq(k#Ro0E;)e8G z&mxZtjNMfK19qES){_&amq|lME14GUfG>J2A=d@`#ZNXr!L>UC$h2UAOGMD=xHti; zp9#ZXCf^v#w|791r#t*_G=}5-y0@WqTQ1)8YeK+A--A|_SEA2D72N(nWx@TcpApKi zN6n@uP;2M_x|Df>N#T5l-Ybfcl%@sx=l2dNzmG>f8AY&YC>aadf z8eS{V0FU45u_G^(*hkKdpx^cY%6Ot?S8$6mgRft5tAyr**+)c?b;crW9&!=-$jZXm zQhm_vwklkc*9+U!9ncr^Dm1+6F|6wy1is?R@Jh`mbX(A2-7jZHgVc0D+!F@spKgI* zNj?7ac{jBEZy3&V(GoDPa*=uFC&Y&}j6sh-7*jTa%e!_9yfatvX0beU_rzgrAM}H` z1@@2-?jT(eJ(Zu(CdNN598MQ03ZWZMu96=bw}^|v6QZInP7Mo^vB|SHr1Q~h=&`vE zj(NDjI8SvLT(=*2Mt7n(zYMfzS`SUWf1S==ahOhx8zwV7GfdrGZTbEIVPC6N8Kd%uNDfvp=rN^(D2fy z=-+x-R3SGVvEKwSmwVBi{McHy+VTpc6c9iX{;R~H+tc9kEsw}O^nh4>Z021X)y4rw zY+$NyA^_8d>1rtpesInpU8WoepY!tJxs`J0#@SUU8!Dm!g*ar|6@~{`K1q>EqZ_4v zqBOC5cF{8`?(f2%*zHZ{s|fME5;yW6Yb)?&Ic9vGtT|uh zj5$A7Nt3T45y0PZcLsl-+&9>Fxdi$(1(B>8eY)l68@vpDrkZ~%VM$dqNgrs(iO2Mr zo{80P!QN?Tp@$>?Hl4}0UuDbZ4Yj~I>HX*r8bei!73fc^-y}VBKdpT*LcrJUl(RF6 zoc`b@n6om$;$>gKK;mq?bwm!wX9?pHpCNk6>J~kEfTcb+CZX)EXT;frp~}1V&>R#) zC-Npq(YjT1OA3MYZtcWc;RKDNS7|M2qHX%?=$!p;uz%T7Dt65l)hwR{P0nA19g@0~ zwOEJy6w09Vp9M7ETLL=X`vD$2%;EY4Kjc>3(!u*Q@@ar%I8lB56<(Qnk(Buhqy4`N zkf%}@H9IhqKc{CZEvH;+_jWc-I=qxboE7Eo*jh`UYsk{OQs?QjQ;qb?MqPUIt^~>} zvW63@^HJgKi@4mW6Budy;<%PQKu-l(SIhe!P~}S8~2-p~igb>eEUe zvPtCVcENYnwHH1xvLKPxW~eF8)NZWtC+dlrW~UzG1#?yjI#{vJz)KI2M9Wn2NXC`? z_fVV`$_n_KO?}WhwTwV_Q+N(fMJM6~{@bk?^nUUMTK2o0EX%H>1zXIhqEQpwcKjw? zbKQt0-4i7X=bwSx{ajRjJO(yQJAg(7IhbloOIj&D6ZfYkk%p!$y7Idy|CE;vz5GiK zoqs!%o?aA#!omlchMjc~p8p4)KCh(_-Xi?GzMHg4H;bC5exzru;^RlDZKB;e(gpv;!@;E}0>1X&2c`p?0)Qxq$p^=z$|{%kjgbj_}kz*))EDmnESPI|Vsm29*a?~K` z5LIAj@kfr-(((C zd<3tah=8l-a#2g?PcSg$5Eee7B$!DBd(&B7h;9<(-(z&pY{<4A`(Nx-VGfnV5vXaI{l?m>3SExlMhx(S3 zf>o0p%&eJx?994lP*L}`3*SGCJ-%{R_)~)_zk60J}{NoCEl4pe64F=)I)zS29 z++UVE?I505@S1MRk)}4QTHp_vbm+QA6MdeSKvd3^!8Hrr$YjnqzWm=UV6LzRMjjW( z9b@0%r)SP+e2NIYqVkMn7ev9a{U*esavH_IbIGcLaQf}A9{umO8hz~`PMpqHg7h;y ztS5VzYIvQcob^tm=t(6#>h3~w?jAt?voC;S8wK})9{OZpl-?v)AZ5H~}d!8t=>%NZ) zV@+sF+i}c>v_m5s57@T%9CWIxhNjc^QrEJ08rJid-jf-jvx_XLiDfl8GuIX5aehJ zLsF1tP$E(3K0&1IMd{+e9AdH!&<#JX6T^Rd@FKHsoXqiIP?4j7&My|G3;q+K|HWit zS*dftxuFsdaiYL;3_krxERWJeb@Vi$AH0b*}njLb8B%T$aDZ;U& ztGfsetMS*;?LGW`1Bh8TA?CjfFQePO}-*xmKtlr6(B4261zFP}eyO`_JjaA_XZp%>D!pu`6Nsd_Pa(Or1CnIqz|}6S{IKt z4&txgvmq}l3SL%M=a@PjXHHF*gSS&pvX@TB)K|y0HB^6%!eYxeG9^nVVPvflboYG# zcm>jM^WaP3lG6vxw~iB;9Vs+cDMc|D>Bw2oSgltw< zC)bu;z~-CQlKqKU-1I|E%))=y!J+qj(y}WQ#LP(nm$F@WYT`8@+Afn8Ti+o%RTsh0 z)4s%-vzocSJsx~r^_EHAwh!c7^r!mOO7ym`H`(4A0P+`Hh8ce=frk1Bc`m$!KK^o< zw{}2=UGyh~mRQUuI>wF^&>FJv%`mIgp-7#qOsTzT85nOH2fS;s;O3>;hC3Phpg$*r zzI-Chd{Vhi(v_z0_Yxg=dwCY@j~^!In)mZ==;eVG#ik%3>IPG=Z3-mO`=P$29#{VR zY|iYpmu=OBl0b3RfAr^M8ckb9*h$?4*!AuONgObQ8uR8;+lnaS=_<|MS+pF-KPn-q zdlT8mxr$K7;{zV4{K-_+3A$8*+D*K`d13Q5jvC*3N42sxkTVM|6OUyP;7i32x!G++ zvbt^nD~=S-nI?d_guJJgcP+{1{W~bs9{?>`ru4f52LIgnWcB5}%*07u7}C6f`SHe* zd)9Ros{mgzyk#YLS!NL#ca@`w?oe%`+M9Omw5mxluIW z5F-1dQfb8*Df}#~24)9LhxI|?`SXM;jUFox|BhvlgDth_m0JRz^Qi=G)E-9}uE%Uf zZ&`%DZgI!I78>HG^esvMYl+-n?uBeq2i{rC)-tSbqfw)_Kv=LHK5@qzztJ$Fl?VOl zy4GIsU!VknUMX zm6P+xj(sg$=5{?YsRIj!SO0`znk~3TiQqc5#dwYWJhCaOlXp`~qgf62$<(#cBxg-2 z3D3GK-2Le+zrkp?b=p)R*baE3TM1i1JJ=8MN>7qr{XDq1Sr06?s>7UtO)&KNA?WOR z1{!-N;T$(De86!OGgHHl=)IpFy!e+px4{wz{=_)u#!YcoIt)7Rq(H}}YW}$C zS-R1g&;*GPI`YhJda1jf?mWPOs+g%*_v-}Rde<8a-Y$S6F`pn`VKHP|TVZSU;dmL; zgUa?8=rwo@nXUn_j;})XGZjH}(20g`NhKeOzYq(t#Um4n`Kzn}(RKf00GzvDc{a+(dNHSp#bK%kpzW`nZp>MR2Cp4Zoef3jaAXACDbn zh#W+kXhygVy1G>w8%WQBq{*I0r=Y>6BD4`{Wz|7%$}oIB_y!!+zr+m-;`r{_mm%j~ zGvC;>5*#Ytq1N8}@Kv-9-Nidm+n6ad9im8QLkB8J>n6BI2St1CfYkRFkVRBER7Ve> zVon!+)GguXUHAr)VLPCI`Xwa0$phZXk3+q0^{|Z6130|Ajut!-FzQ3Hg7-Po4gHQwvof9J~RqO?I_@@ibT9}a61g$Y!`m;@`kB@rO9wUiTAMB z!`%-XN$ZcL5X-A-q;y;qzIxOX?-(@>f0Vez=iDtKm-_4I(M1dKt6Py+Eo?kI9Ib<| zx{DphwaH*9A1(a)qaWriwZpsi6w{+#e^C4rN!*)yjhJJKOb*@xHQNx>zQ6-rub<3^ zZhivoOMCglQH$t>J1b#ty&QZGnus+zlJN7<>+z)iXv|f%aK6)3sn+=vqHOw>zi*O@ z)Px1RzRO7zF%VCKW5s*?NvR~%uA5lKr@`DYyYW3mFRWnS2q{(v`D}wsQtuUvCc7V_ zlA0}otmFA~m-8Vs?b{ggIHBxcg#lzY=P*#-(j135jOP}p@&_Q=#t$d zsi0>CpI_WZycS3B>-rBu^$|aArrbmETC3PM5KPC1g2Ni zVO;SXpj9g53N|MzE9B^<*?T$fvzKVqyEtUi#^F#_0qzrqL2mnS>RS;7zk+U|ExQ(w zPhWh=w;VU2w?Q@C6&ymm@01XJq&D>?-^urg1~PHJ0t=gEfm#=v^Dlk+ans%fpl^JR zUsxdS=gG<8Vx2i$b2J?+i%QoQ;0>i&3fVv%U;1!RcYSq;YAx^p~RJr z)06ln>1U)edPw+no3X7@yc{TMMDeROOXGP4m+|N$r}?0YBHCy=8~%I zeXc!)SgT&7r<1~{$E54L(tqb^*!A7KRfd&t`!_S%vUD%XJ7rzB}xEGCW6nDPwxsa5%576eOR%l-4#XqDeoc-E#;nNKQUbM%= zHZ^<|LNiwJwc<{Mw9$Js-%A62?R-c6JQ!cuDR8s#g2!aZ{STaizdJf~N`{`!SW73Q z81lzY#^TI_W5I8UGZhuuf#LKVeBJQ?=dybh3;eJQi^7%=qjjdXiF1^&rmh5+DWPhc za%3g$-76)$(XL{9;r1K;wsR=|efCzmed=}c!Tb)_Vq8Tho=&Bu2cHSw9ag7qKh^od zAErchwIxZY+bx`;rfJhojfkV?exfqwDKe^>1$t2@h2NJ-a(z`5RCY!%&CMT1Uajnd zIIA$O+JK19UnQi_sEZrkDdH3~Y^meuV7|!O30=GS2o%K(&k;$x`RF&|4CVo6JV|{a zck^&BNfGyS=DrN#))Zxt?@wI>cBf8OEV*V6nF*JPP2;Vqpy16^`;$De(U%76KueOV zl}r9yyvxN6xMGKHCCIZV6mATjq!z&&$@P=2lhWS*PApBYvySYw(`&A)!3slO7) z3Y)7)ePc4P)>Fu|@C;2)n*-@hCRMXnnIY|eEBI^2Hlc3fPE*b|(hWOL(EG-QTz|be zT)h2S@F#0O-G^?Hx1X1xyXkLeN2wBwOK9MF<4>W0)*U3S86PZp0jtL+75U!M~EW%AKm0dHH#T`1iRgup6y|p6KNWUmA`C>0^V!+uUwC z*|v`#>bS?Pn7RVDoJ_WvZ@;lJXPYS-*nA$TCx{(;IUCuW4K6I}f+{<7ESFYIt3a2; zJfOjadQ7QlAsgsZU|I4T#h#3Idg#n2Bwf@@Ub(G={=<^cHNOu9ZnUDO@{iM-jq2>z zG#SuT$z&6qv+-p&8HNjrnBPM2_oOU!wAbq@nkV&)@%Ihb6zyA#d#V9Pug&N7 ze>=o{e2)o@qrdU}tL#XJj1Keln!)J%diud8ffF7Lq|>Kp)0T)vS|k%oeG0b`xA+On z!_JUyd{e?XpS#a2@@}$pB6elEN#J}j!*O@!5485gU3~89G8{N+IsBVijlSw^5spgM z7u+`27H8K7xDjWj@nJiD(DKkE>QS_dUo=UR`CZ#gx?;ZzUOl?R!aS#ns`aqwM%V&2 z&*n58KQN4Kdv=*k4!+C=@BLxX9|Yijb1Yl@Ns|56oGNPj;wLivEGbgPf2hp8VrHM? zL@er_vNcCf(3tl}*n>k!Oxa#elsoG+lUpzsJe;Fg^S(o5DbLFG-W+T4oTtJKItC*XPJ{ z_F1y(-$XE&n?q}(Z<6KRDP()mR?zB;1MW`@`Kaj+{?qcwBH2}9S93Z0UH^gAs%En# zJ~~8X@|CwhvSgH)U1l?JIa?q8o_ZeGg9`mR$*{%6=vi77lf80?#k3{Sg%)w-NPju) z9yX7b-*-e~%9?5Ixez*B=*H4^8DZ{nI2{BK9d1<+dFb6^ou)QC%Nve{?vEDxySCvy zXFTx!|01A9LtnJjPD|90Wg>bjf3|W>o*&CTQOkk?U(kl*bGIzw^iAd)Xv_(ZCDkI<#= z_gLj#Gqkz-Kc?t3o?B9Vgwe=yQE_IuDC2FR$RwqMs>L2-fo7U)ruJ!eN@gb2P&UG= zon=LLDkMc->i1~$twYr9*d`XyVhjnsKg7HbN!XbsmX@!&N@Ledfz?fEy!ra=@aoZX z^x@HVzD0E;82SV9zwnKWQdgw`?arto$_18YuVZ$liR^RdZuan5Es62deBCw<3mB$|taYp?abXX5Gwos2$SJXyF&)JK=Se#W4GC20J@y zKa<*YmzQs~0>ft|FvdB9FMK)-mFwQ+y=%6Uo8%+kcqfiL{9euVPCg=N9&Qewjk5%u zo3{uCSC8j{#Z2bnrf?FgWk@@|pAvNMH0E~R-pDsDTF-wEbwxvEiQM5~Yx#p$?~uvU zG-<7qK;Zu^m;3iQlDoLBmamv^LgsRF$eIBYo7-21RdK3*0=--Q{AjiD{O-G!f?;}F z1XE*M`Jzf?y8f48WxL`LF8khivh2E-Ab5Bkr)fKu-)^-!bQfV{Q9hMf)_Es6&R|MqYtG7Rp(!l*8SV4^qKvf)bTcM{wgo7 zM)OP6klPsQRlw5``^F0!YOVSC$p)Own>zAwo<9Hdg%y{d)yY4;f17vPBj#*%c?beC zkC1@R$EdkvP36~(rTplqNd94yBlmazX_DF&ObZ(}3;zsA(oMf%j}>Zj0kAtJ8Z!iFN%sGEBF*>XPR~j@z+EKm>!T zU}DadA&C{-Xp8rPKE(t4@mmjU;`KKRiXNY-`hDSm;758QsS_TxX;RrjI%*uKTWWme z$~BIB9qQq2&W3Xf*WM$K3@m7gX^~KFUL|M#ZB)a7{Q@OOS#=f&xy{l zvz0%6N^D|hpB8MGHdzoV8OOt@V1A#z8rSOi97&XYp6Nr{R zRZR{=Sf53jM;q{;H`kNQamUG@ zIT?au@^iV*FLMPWY@beZt*a*T>P6H!+nyx$-{P;AS`w{FbN*>56KAm5u8T2-MYk`K@*~T$7l~nYu%sO#Zi&d%IUnAiw-9 zDNEW}mA3M|AUfEC80}anyeQF3Mt-!SiA{%ux6UpQ&Qex~XWA0nI@1??RbeYB_8QNv z-)PCt9=@K)_j(CODx?c`#e5N%_9ax^9I{273mt^bDw{agHo&>2g5ci#CPCI27rxZG ziGM6jY-G1A70T{gQ01p=%c->8;`ELd2-G$osT!%5&Q}(+@q^!{Rkd)lkzZ}Jpj&;3 z)iN1%L0Q-@n_J)aTK(>x&pqkysXFRwEId2IiHlF|LAigTtFA1za91&gcA zc@5PT;rnsh1P5HjCCDvG{Nhix!U+@RpzAM;Z5%&4b0@W?ae=-=+}fWGmHt2MgnyQ4 z+eGSnL9+c2e^T%c)h_k0jl82`EBYH^yXw0}nx* z#@L|s8~tccMh}Ygeumt3Od|V!uE7^%Zo_r)jJA&#_$;GKJZT`{z$O zCc)^t0^1C;rMC0^M%yO5I7zknP4vgvbx`ML3t#7L7w6)qz=br}%h$g&=jwKmhhD2L65T%(qvAts^P?@8Y zRvpO|Gd|17Jk-i*t!#z1*)>oTpor~!BI$GSi}5P=lYcATqeWJR(a7Bz%(+H}Jv_I8 zEtLvn%Q7%qSU!s7{P2KnTi381p-<=urE5%1zZ{t^|I3?%DiXWtahzUd5_VXagd@dX z3lb8C-*mQM^UGS?nOaZ!c!4&~?lA)^vF}80d>$P2IgOPIj^epjIow!MKuabMBPtE~f9(#0;4MeV$u;gKkSY zMLK`q=Ra()z(!dGxa?#$p1NWmFO1vCfyW%a;;bEHEmy{)wok&T;fMKs(fc`fl{EA@ z-kEp?+rp*fV|4eYouKmi2x^~_N9~@}QM)Ex7Grh?ADo_nl2aEUiMTm}ORp|6xl@iT z{>ccMt#OM?jqc(NSuVMLDI9zx5Aa4BhA`e#m9FYsP2LF7kVGQl{#?~Z9@~15_R9md zZ`v@PulxxI$4QVQkuF#dhG-xe4m~%VP};;<(Inf!xbm0-_L?j80bR&?Wm_nT4CIZOY#YI&d~our6H4>)$Vh ziW`fW^E?*wGn$ns&a@>}{Vih70S519i~26 z1C?(aNWeuUn`w6XRT|Z9_|ec&C}|VWOE>QFfuGKz-Y4!hIxlx%@cRI2*3BT{CWQyj z718B&?fkRX7I0?rU$AofMIWTRqjln)R(^O6w^O_Wk7x@P-@(V>M`hkr;QD}W3){-% zf3(x)SY`G%GX!rg-o*?@ZevjcR;=3oF4;OojGCAUtnO6~R{f@C8`q*?d%xods7}xV zJqMmnfuFoWj|%f8+0QuGF` zMiSU!!+U5EpRe9`C8NC1JoI$L8uE2em1X{tL?NmrSV>YI@4Jlg;DbCY^(zivD>zGT zJ(~b^-fQXg9&0GG5z-iqF|7EEI<+#W;kRd6)06kU(lrkR^hMzVa(}EHLgU4Ajonom z;Gsyz`agxRhKrzFZwK4_+_>UeRT!u>AeU?g;8bdW_`Fs?(p7)ax$d$c_uw8mIwv28 zH2+2a1J56-H zKxdCgrYq9D@$6O^Y%?K_PL#=|qxN|+#iudsh1N)1eyR&!wA+Q>SWLkVMpwvltF>@K zF^0ViHG+d1_XDl-V!QonKx6bu68hPMecjlGdLw7ky0~7LweTn??*9zer=8$Ze6vYb z+HW*=#9Bs9im0{vF?Qndb~-KDpG{6%$1chz&|fZPY|@*PB<jPs)xr^k-BvmuZrv;**VIfGLS(Yo|gmPq<(Jw)bF&(egaVzJ4^!S*fBk^N4re$ zkS_V)g(+BPCa5ZlTwY;>U{$JdQ&(R62I@lSOPhEztR}^yvw_lOG)e&rk z+6Z>6JckL(k|1&EWqLSzKPVW;vVg>~qU?!^W3O()v7H{cNPio$coa!FgQ+a@awTc}I}CL% zPhpPRjG*J#dlc&;?gLhWK~Zk>gZ@nzgz_lH(hB(NJ-*25)pvE%u^9;+F#90xtC zz%}10u#wAhJY!V@9qyLSbn|QI?g`6a$pa$hDfr{I~eBmjgzWO+Qy0(@`JrHyHlwU%A z#9P`vaTu!_HH?Z=Q{?KNX{>e84i>qhi{7-1VHNXx+0G;5M5;kTmTO#0H#(f)i+8J% zxkYQ3|MY*(esrjgd@gC63*Nr` z8(OM(1bq)3gPPVJ75qN^fM*Udc(`8^s2n`Vx(D~LYd@QrN!@)IjP64gRG;gcAjjUX zNg;txE9kST5lrH36#HWD&1k9=JMdzQxZBoD46`SYhj+b5c=>RWyV{LyiYQ8HJyBVH9#oR!_$etuadUTuxR-jCx!x^eH)=1ljINDjVJd=f z^=8NlT8-CFV{lM7ovt33L%&InrW)D0xO8F)Kkx4rPHwt7U8&qe2Im^W>Pw7X@SKKE zZuO>TJ2F_ghZ=Wk%|q^d>ulkK3mZVDTN;bfbn)QG70jkRi!E9p$r_7{vCg~`Xhr8Q znw`EGWv%!GGdv!lBjP!H+SGD*xl};(6o$d?y6@z)Ru7Z@?vBSkOoNF*E7+26>v@^k z=h07)B>_M5n17lEecEuEWGE}J@}e93Hj6a+d3iW1Z(hXq`aWecOCG_|=NkAP$p!h0 zv2;{F(4d?bC_`C%Coi8*ub+KEH&`5FNeld#*eTB>O4kyj(T~u)DN~S2*lD4A_)Xzi z(V*bX@i^GqI>cXmRLe#VJRz4ob)YTmA6U1J6vtkcLCgV3`eDFcsAHgr=M_vsDj#06 z(yN-H?oBgArtA(YSSlw939DqseoSMbuN7G5yd04EVTjy@*P{HddE`y*6PiA|gkHJ0 zhAFPjVzMLFu!u)KO#QYt9(r$p2XeB=RE^uPDPu0MuTG>&cP?Z-Dkb;&;~@E)n6VVN z9G{3B;N+$+$kk8{QZjFogwJomQSlW2H|ZO3@yMX^As<*odNpf~?q$u<4KZj2HErOP{+Mrhk16|HAj;k;Xb$u5}yx zUZ%#L4}50N?P?+X_)L7?VITe}qf6I%ju73%N(aLvM!SKX!ntdaCJ7&)2N-WGD#{zM`4|O33@05+u$Vi33st$i3ua?0ZEcdbi>~ zZvGQ#dN4+S{_Qe>-RI*WUEQ3{f9yo}O|@h)t9A?0oJ?VIquAZFR{^plrqiY(f4Vtr z9Sh#|mR;!-vC%PxEX{E;JFVshWlpNtJof>x8O!k2U0(FCp)FgUa=7aH?j-WYRFD3) z^aRWHB1FZ0IGv9NS)pR(8N|P1S2ufqdZc)_u=otL#QV06|coD;{ zZ?XF7RQ5y4LbT!CA{?GjMT(;bsGm4%*VyAj8#KS60x@AyMJbQY$&&}|^(Op8MB&(D zOL+dc9C0Sy)cEQV)*SZ`xq3h05_V{Ss!$80<}Ifex=MK2u_<`&%sa4Z_G&hSJ>bW~ zCX$z*iZ9Aw`XO`-^ITEPw!93aZ=LdapZ7=c)7P2oz`XgKf$S};v~L1ghr_|v&VkK) zaR$fxL}S}gN+|A7v^d9_25Ji0(6-SYX7X>)m{u2v&rYLP$E3r{4@=M&$fIw9WU$Pb zIQTqA5#QDk`wmYup^)^+@X!CbAY0cDe|#~U&T*`vN?C<;R=h1+<*x)c>U8nT>Czy3 z;|l*iHW)_!NrQtb7X|QzrB!Pr`DeOCNj79mg=j6QkfRa`i-&g z^I!Oq?`0gHV1Va&%JFZ^3n5f;tmx0x&s;!G3iGH@q_eaQaM93Pnq~iqM6O@J-Q7C| zWM9-Ega}ixO`g;`?Tq>JN=`X$-`6Vd8HJ-o92vf+|Rp+;b{4*T(OvKZ5LF6TQ|S z!p`247wN5(qI}m8m^9yvDa+hoD>4h2){0@QBVPwb&W__cKRzOg*(3O4+qT&JeKtjy zVzHR^P11lDqm${HQ2?_%CBb!M35trEhO6}padIf&vfEms+&CZda`pjk_fP}K)*C9CGw4U3qnkJn2`XFyP9k}bO+mZaM zk7SDe2YB>78eW=MW8GU%;rF>8fF4a?YibVA2VPB7;mRcT-KCd(>D6POvdXAYqanUJ zTm#;p_W|Fx6JhNxX}ED*8CA+}z#~;!`Cn?+>1g{qWRK1%UQ_6TS`zb-V0<*^b?^_L z9I&1p_*+gJ_J*J^jX*ZvBbr>E8pnTWHpjx$P2gg^4zBedXM*g?V5pnhfQ~_c8ke*${F}f*u*#N_6EuL48^~9NO5(K4?_a3G!u3!K;yZOpap} z?Mn2@umt+tKuvTZOp3O=9D~ABCc)FCa&T4Kl{o+Iqumk8w0FT{F#9_W4j#IS9Q>xi z^tjFBnd1+Bl(-A?JW`6ZHXpV*&?SXG?7PZ)ofGdPBFc!vPy$-AZ5yuD%fy#v%q0(I z9;E)<9=@z8feYGZPU{^z$wZ}kG~iK*Y8U=QZG&=n(N-V4CObuZ22rdk=m@GDW?m^Y zK4Z(wR+fN#@);(yk>!&Irjk|dGGZo-xS#TR2w$Z~FOrT^qFV5iq7&>m_hSefIEQp1d>!q6C|Ca;6hxLtU| z+HCB1rU>gtzh|qPm!Qx0yz$PCS$R6$Y)&g~>Jpz4+L-6DNBw%Z8>7%)7STfa)WJ;@uzIExM+a~+b z%_+JdrkdE+M;O?Sa*?*3Vn0?idzGu0)vE}*tRFC`MUzB{asSb>(dWp=(oUA9AtM@R zl?X25%wg$$JBY4%%Wdo#g(4SQpyf7!M0m@Ab^LpdEOgfMm%K&XuJ=n(!@mNOiuTdF zV*=0!WgB=Cf0A!g`pNh6x7ij|ccS!r1{pt;fNw4S1>1u}XkJAnEO0kx-af_5@O>`o z%NZtSjvK<+#j|nK;RR6oB9`q~TFmA>2}U!#lu&1&5+5+M3ffFA!`4_MwrR~Q_P8-b3HMpTc~U168#thwXcm@Po_F zaJ1GL>lj|a<(xKN0y;SFz;HbEc0R8^GM4Z9PZm9&ah0}g-i(Z9Zb9053RuDG5pIr_ zv_08%l$U5zp##edXj4Kg%`SdG$DhANbS}D}jC-CaCddocU0p#s?~SC>H~5q1Uk6an z*8&vHO^1LMQ*Q1O3-0=p;{w4GS3oshggDGb!S54c(Vr1`wN?;LkDSD&>&CF3<8#@i zd9ie`@*&^#@c`ZX#}5vR=heK90O}jO6KOnpOwODuN1M1bYqQoVFwyz~H~i~SFiBRz zT63$Ag-ac{Tt5vC#*1lNh8|lrMO6$<+=ZjwjiP^xMAUIhKP|f&MQiF?;9XS`ET|pF zlC#IM*}lGP$&?GM-R2Wjo1TEBP8>m!sviVXYU9w(7|g~BS7N(g3$SbNd3ZasmrNF) zYm`QQh42sRIMz|zul;(Ps?R+M#=2q9CYeQN4D+QYeVVw{6Q-~+|Ba>Y{jzwrjttqb zKpHMyj)5k(Y~&i!WOG(;99cWJjXn-s1F0$}anZFEOw~LR6c?G{FB^u)wZk{i33CO! zB}#mM`rl0aCw~B*6%1$YPQFY_Jd>`qQpE=1z~U?QH6Xv_Giop`fO9P`(9FrN(5^dr zxHiTg;;PN)k*-l}$T^!@SejTzbd47%xh3(5U5m*0pf{Z9#EZynb~lX}E6#T`?x!2R zrqG`kh9i~mvk*N+0G>W&kl6T7yrZw6w+o6WHJ!<>OJ-U-U7cI?;p#N>d-_ZqA^#lD z_vyk<1KyM9f(u-B?@JoCUkkr~_y%vVEy0u9I5K+kWa{yJ9u>-;p&K_CvR$!V)Oc$t zI&WwU-(IL<3(WxP7$J@S%VXfx9zeG5szSdxd)g(`q4}93QQhx3w0kCIX}_OSxv3kO zYri$#5I@59xB7*0~1;C#1U+#?_+ekb22S@{G7hC z)PTRgmy??v&&h`tIeMHMB$=hQ*xBg9_`h|*MgbTS_){68UZ-z$5bFgslApAWz z3cGaA!m8JM;j(8BguSjp{pWVkOE!o=12!^y7uHwu=9Ba8RzB9JC zhe|#T5=o1d^r53UIXQYX*1dfT#BO_;d~b3jMtX(+@nWCZHpGu%{juBTQ;+k z0(aWKEF51rn1iNgtB{djGO7O1H7YBpre-M)PRrgwhgw2W;G$G{?4Bw}Wlf!{7EN#X-#lJCD%U{*a6vSdpULtWH{xaJX{na>6Esr3W(lLcMOg;;W>hHN(!{3va ze!6h+l_$O5`IHSU99X_sF@ndH%jG_mg!V=RBuB$yI+%C|K!N`WvX~> zqz7xn<}9f&PiXWyg`a%w05RRKE0U5v&%Rxi7O8$&M9*n&p=D#s(fR^se)Y%i{PrjH z+?p5e^p3X?J-mg?!JoyQm0EokOrL*fQ6yLt zgvfu#kBZqiChW`>OZwG06E~-f;P$zd@){BS0BphX1 z6n2HI7SFxG;|t-V!7hr*iH5Mx?#ikCoRsL!aR{w9WA}Ds7;=q2W2O(*id7 zu8_th)}iA+>S^`vW8%9r4dSplj!uewPyUPjgS5oHy+m=hA1Sr+I;J&r>SASKQb{41 z_HaKC%|N#P+;2*nbC|mAe#UEKc6`o$mh9C{Gc2dE#HFjq7!z^sKhqbSOv-7Zr}%6p zX53o*Jj~3S&r^2;U9g&F#`Ne1Y9a0>=Xjh(%~oIefwYBi%(6kSqq-N}x><*cJ#*+N z9VsY!H46^S&=WS9X&`$81OD9N86?0;lkQk#12=kaRDC`f2dPjDL$k%t%Ktku8m0v) zn;dCUt9Xa)bCV1O)zE^BXH~iBpXig5+0-ZXBz+{&Otmfdv&bofbWX@tru4WL9b4oB z*Zw83HzN|@sg?k)EiQxk*2YYw{VN%>b5B)Jb2BpTt3@5PX-vr?OzJDq1PD;BR}c zBdT53AY+LGJ?Fih9W(ny=RGuKJxX%W5HcFd3$)ONh@1SnC6lPls#Y}D;W$j%Y7UZ# z0(@V)0j*9sYUAW{o#@z0u)o8jsq!o-oa5<5CtF7$4-Hj1;pr8axo9FT*|{AH6Ap1> zOk_z{jSLiiCggO#8RWznz$$+kc(*4D>NV`(Pf;F!RyGNmAICvlG$Z>;?cvO|6tryc z1oO5rW?l2#;9rO@y5_r{&xx9fGE2?S*3CD0mx)>2!ojgD@#0gm%z% zb{oN}LA<}a(TN71Jpk1+;w+ut3D)P3!jyL&r&U}4?Qs3+)Y9MbO5{= zk|mk`CX)(|Q) z?lYZMkVGW{rvm@+69vHw^3o}Y&KrG_-faqGH(O$GV% z>mUr36tYkIFH=Pyj;;3)@{VF3im&*5oo9cWwa0v=YOa?!fznd?eUUrC5>X_yg+Pi( z1e10~!s!X)v1xTJl#NJ%b1&7|X3v?VZ2J*`-6~glZN?;x+A7)+QI58GJJ507!#Izzfxrt8Iru>eeumc&`*AtM>bNoKo&J$^ewAiL zI*Zt>3LhpE+@l?0X7qbM!~#sssr%l1m?a;`rmXj8ZMhzFN_r7U-nmY1?u`aBCw;bQ zK@^M!>e4fXRy2Px`X?9P-|8h8p%)z=(JL=g==2e%xo}TOn!DbZO1)OYduNQo z2MiYBQ;6Jf_BB~GK(7u`N3 zNe(~QNqagE!k0Q1h`X&$4IE7Hqw;HnOHw0XyBzWRT*yzEL_TU0NA|RrP#ujlG>~8f z;h|qpoUs8c`GF93#tI)ScuD)t?PTg1rfk96+muU9qNvM-`Q105k3}lfry_v=WbZ4O zK3*IAM*Hv&uR7A_CEDz-Qw6Cu)n{=X9Fv$J%hpB2f%W-O{OhD#H0EqAO&2@1ti}$< zQF~s|N&j5{lNf!xvdoxMx#$Y#oDPsfm*-P+Wks%kt$1!3_kus^e-ize_kmA4ahUv? zT*wZ%ucw>t#G>bpZNiyM8a6Im3#FBQ5cu#g?3#vY#1>=v!esz;%ZPWt4VG+PY#z1o zNQcP-zU)#!0$m+81fO<{gzlVR`0MXLW@!t!S*cIy=ttt*+YJc_EIb9`eB@@ z`17|W+TnhV8~@@l9Y1wE>x*)vdY0*Q)E+x{bVU_zC)H6ekACQNmBV#z;av264Va*l z##wHj#ENXC*@MjebVl!W@~$SDo>eQRf`vUoIVll!P(BXF!iR(>noq$%%q18*BN8q8 zun5w(ABSO$98TOWfgjK5grY?w@tDXk>R)z^9@zen#H}39epCc8?{Sw%~24WK9MMRMixu*fr_rlu5kEZQB@MP3!}iUJVi{J2+}l?#!C`_4 zo;za;mH)Dw#hV9^IMoMqtM4IHe#RLWJyeCgXRe@`i=9#1hiho@+(znmX(#Htdx|3^ znJDZtf>&paa9&6j+WP(=xNZtzt352)!PaS1p=2q5h9XoM_fWgjmPT+)|7mqob_5z@H^*uE8gF%}4B03j5Qs z6n6O+fZoI^ZdXDaRlL2FHW+V5F~!HhW7;F&9vLuA8GrcEuFOVEv|%A^h~h|VU|p{u zVnH0PnsJIhua>}9Up_+)PV&aJTLSQ-9S&5&pTb__UEm?+@I*GL(D_qG(V6`x$iQ1n zpY3`hU=0kSE^d zJPBWttrh1mWI<*1Y%pt7|*r+$RGxb<;1&+am+g6C)x1z4Vh)LOkh=XuIj*Ab&mfcjng8k;pw6k z*u3)#S6!6KDXgB$wZ!gcq00L>Wy?}Bdh{4_V{o>TIo?Dn* zz-kyJ8;i3_|D&Os5omY1;3Z#1^U2)js*tEV=fAD^{5uEi z?RSCZW+y|5IL|t+ZvpZ5T1H3i8O6etcUJz~lZO%p?P*`tGqiM946t_=^zemQFnN%p zNjAQ8g`Ww0>(Pe4Vy~5y*EV!T_Y86w_mLmAG?WqXKELzX7WN_j8;S4V2(gD8;JhFW zn|Q5*$7T7{>3j&eFua)TUSo_i<%ctmsb0Xl?8NFX3R&AuF-NPli5PJO z+{Jkc1@#g%?oT+o9QG5p~_RR1xKGQxRF* zk0pEE-oZhsA8<*iOD*~%*|XUK_U7Dy_V!WZG# zL061_n2B@`IMesj_M&g1Hc}FzE_&A6$d1>(V*d9k$hu@%eD>2Mye2*ej?S2jw;@Xw zVtZ1&GxB7XSniLWhMKViNijqv>?7h}8lt?0J;!i3HO{9p@^q%+g z`|aJ=bv~c#I>UbUdY<2N-#VEt*dkhAxU}VkOap6)!}mM{OVx>!vN*}o{0;(@4#3mC z!utH-B3TN501we3Sd~1C6uSt#V^&SYzI8@-ti+*a`9`J>zuQWijycN+JCKg&5%onuW=EEoO0&10Y6s*~LhR7R3ffe@!$!{{i>TmnW zJD2%_xvvvpT)Z`D*rjY$ZZXlpd6IlkQp_lJ`5PQBEzEr#lMbo?SnVL=< zu;BsLSN0*MeG258+ZR0A8!pVRWJ8PVd8DW#2Z;)MWYI0lQSLQeq!;@Eb?i%GH^;su zb6T^}=NTvAz>GLFZ@mB?oScE>*F=&F?c>-b)Q_Z>DB`lhU3k^8N@Q`~h0IEz;qe4D z(qnTO3o^Ws_UY~Dqmc8lOV>gm?^K4T34Nt+eiG0YoG9GQ4n$vdhS2=%8N7sh3<@*} zgcs?XM6^d5Pjlpw=CKrPV84|py4YY}lbgh;@B=BCaTd9~6z;*l?gDq8D4~z3eYir& zkL*4YOV(-e7d2xiTL*l5rC>Ms= zl;R`q(L~{X0*Mpr#TyQN!+!00z&;Sd_umWf{6-ffUNr)tYYvn=JrA~hloR~3K0#Kh z#j-9wCUCy3kSKiQ5HI?g2G4YRKWcoD2Ssbu1Stb0*kGFrG1hPuoc%3Eaw^nN_k~h2 zwnR(t;)fIYAkqm#>r3$K`{tOQlaEiXvcb#!+py?gmThU%BO8PV64xx2e(>`#_HB!T z8NLfpY)m$~EOHSVjg~==Q)O&T;3?F8?j2m*G=`;SEg_d|Q&GjGBe=xG1yv-b;%&}_ zM5*#W{JVOqkgczRV{_*7?tD3l#w>rb8w=(MJ{mqJqKl7_8=cu?c+nr6gi1-#l#S#t zd!A71Rfw?rDr{pvVUGkULGP2wq{C2)oL+Jjp5fhN6)e=iYq6E&KkK_}Kteh0=}uwh zIcMNO@l?FKGK9=nrUY_Kc0rfJsaSE;fZgw{#a?=O6Mojc2dg%XfyeJLw!J?McCPJ) z;`w?|G$S5#Ut3C>M0En4mZ|Kt21nc{KO34QjYA!wmn{6x5WBefA)Fzc)6HXR;Ed&n z-0QI-!+Ww}&}=>{7u1aZxoe@*TB0Q2*e~XT=|Ym6sLQ@Ci^K>2%p!~P5ur5nU(9D^THUkyTl_R>Q3Q-gpG%xh!!Z8^m5=$wLg$p2w2R@bs7 znPA8K_Mu?K!C8$p=Gcv%k_N5r2l9KTmL9v_SGr8?0X(A z|D{d-E*itZry8)G1WB6IlEDBfL4$lK5snguy>VAbWTgk=K%9 zyRUVCu9g+#$G%qlI${IqJN_0YP0nHutx_iEY?Fy&cOTeTJP*x}wuPbG^$e`~4J;nb zCPrR?^bOxbv}r0I-rw_>Ej%%s%`@i0Y9A{i8=*;_XdfX`NH*fbA8m;F z?09VU={Qy$GA4IFI0LP#HTaGEDbVyt1Xv?wxU50=S;p! zL6cc~#p_s2unPO_*Fk&rFT)l2nOH5x0Q>dnG28Y|0Y7f*!^lhd;7ZzTpqD{d=VlRN zCY(LT;YNJ#Y9%>fRn122e2n)7ErtK(x3e{$N`dS_O*(7T9;(!LvEMc=XCB=v2MMc} zqT|cN=G;PGTwddGo6^k+u;V?X#%)Qh*6ssf)? zvV>=CNwx(F^95`?dGtL2MhJU;N4sv2Pet2+R#+1La?Tw6HI@KV9SI(I(T8OMZ?SxD zGcuWFMbsbPffae?SZB{v^i9Q$o%_p*tWf0em!>>mCmDvoGsDSO#F7emRNWRCi?D9*efXs?n*R50Xh_R5b_e;Lxg<0K^M}|aut?- zoB^*-^I%m!h%nCwb}Vuh-&=;CW-fq5VRm5t z^E#}W=2yvo@|W2Qggo=Bb77mi4ZPI&4Q*6t#6#5#`je86xEf{rx(pQps7CwzYL#8 zb-VAQlj8#9xUU4=9Wg@rM~|}>Q8UQ!k4l_0yn^0oY>ZEZ+$DR;ykP&ldC(L+C1L9c z)T{ggbtPIzpPUpt`TjDDjLt#tEsXFAqs#D};&$}v&@DWC?+bpqXC(?s*PwHPQfaxk z)zIe8Zn)=n8%*Ceou?pM&5sRKhvB-)prPafi*H`UHgBgOyOzx?_t_k15;Yxt(d98p z33E_+#3W!ebogb@bKnO5d^DEj4W6`!phl~=(4|?7%6KV*9~&NlvzFE4$K%P=`Qcb* zkHrYqC3nC+$2%-s?#f3i@6ekrR70Z|!6?yZ1}gQQi8hchzLA?a{ITN99~l$fCM6%+BhZf*R%j!Td{USkpEd9u~1i`}XT1ZLv|H z8h#n=8BbzmxiVmV%yB03<_=gp!2?{&bI>$RgH8BmK(FgQ$BG`8gHgN+C>CW#a$RqM zv1>Qrh~-9@_4N>(5pRK_zR4kX-I^*jPc!K3po82h<3M`(s|vZn81UZv9s3)lL!0#L zF!RR=G!Ze5Mw3tRn#at92y9(E7UD|exjN|J9%&@`=^Gj`(PpM}s-oV{chCx#P-eo) zgJ+fN!AhrI1r^1f@M5bde&X*0s{PeiWjPV{q52-S)9?VhJujHebT`5&H^Pu__#b5X zeT11kPXSx&O+jJ8^XAk11$bop8GNuk7Ib}5fr=6lz#z*MyC06jL3Ou@=KBd4@$3ej zFc^qSpGg4YZLi_3WM`yuZ#7(2cZZee)W^H)tDs=@C|*+}fWz&=GaSh7iq{(EaDtK^r$u1z^tr9Y~M+UsVryB5mR zwZ-o!2c0+2@B{$;0VQ-d9%E`dQUU+cNw)2*2nl)h4}N)n78ZXz0(GwF!hd({;IkA- zFdSdI3M=88ls{%n&_s6J5<(LhZp!g;?KOvMNrNWEx#X%9(X1&Chuq9su`kS z?0gbDb7eE$psWZBJhsBNt#`o`uPMlRsv2yc=7|zR!_eG16{NrTI#R#kjgB0XL0f{O zkbgfyiwd5@0h)_;NbW`fn+}7NTSoZJZBJ18K@l2@uAy6=*|1>}LKn@O8185cL&5z9^zm>mIvPG7g{!ZpA06Hd!*m?r^eG9j;J;M3 zX#s#bJF{VCk6$gtOOEbXW_u@h3s4LcG~}18Pqx?kH(ea(3Fk~tdU46R=E@l zYcDUN+e}>9o&9Z0Ly;4VPqRQShDXpd?8~gYx`+MVu#;_N=b;sU7oqda2IRZt3&iyy zu;?iVof}z%rr$b;ju!=>&BGkn|3d~A#61GM?FZn+*mRgMZ4uf!xDKfG3Qx8LZ`kvc z82T@70B%W+MNbk-xeP)#mcD{Jr`)GM zunlmoj1p4)t_6+HEJTJ*7m-d`0x*$BP+2ntzN=pkUdI%`A0HrkcK-)8Rsj&qy#Uo> zr$e2;73`5aA!zE>XrV{B92U0s!E*=HD<`x1faeuEX2;)HY$+l`iiZ|KnZh&J@o4~g ze7%V@OoqabV#V;iWGceX)}xCy%4o0uUPP5_h6nboK_VJ|QLV^Hc5z-CG%gmp&O&nV z7Y`fwV5pQmUbq}B&Hun&uw&sNA@^*TiY0pSQ3hrlJdEN*U7^ds6HKAUFeUsxEjnC7 z`?uVL(+(%YTkei1W4iEHq)~&r{?+2SF3b44&hBUOZqI^II~Fm+!Gl1>cOi=G_{x-h zyUTLF2pwq;orJSeDXzSMLEgb zYER-n_2adr3aoV@AD6x~0KA@7Cgb@5IMw_oo@4Kc_b-z}mIphrAY=_BoJ-g&FbXw1 zEkj-J)6h~oUvNNv4coEC7;W{`M|N7}u)?)Qs4(&ZtyA^kz6*0<#+h5tYSDMpq5lcZ zeIW|PWKY0+NfVSbZ8e!G-@yLL(}u}Wy7W`mDmKC75;Bj7_&$QfX-mS%#hfA4|ePRKU>io*Nzqu_#h>#(8n7H;ZW zEmWK&hO8P3u3v-p8v@Ph|yMkYbO-z7?YHEordJ zClrb+w6dG5DIB^;4xXx=35&OP;#=F=ATPiYWxf6Ymh8KU3kDRic-RB#+JFHP$O&1G z5o>AvetBfOycsSxjih(4I!}3<7BEAjFt^^&n9ZlEhp5h|pc4h3?v?A|1?w z+Wa1{OQ8sR=-3gnoNxk+97##Z85r+7TgYt}?$mW?5`#>Pqqp)=H9U`o1Uq5hN)uGL zD;>`JTndvH%!6Jz>sb9Y;$U?5Vme7r7iK%GL@_s4q5hgeD81$n2(^GCJt2FU|7H<%BtDd6Y8!C%y5D(BWR)hWsCh@z$oMpnpHLkbV6e(Ax75#yV!< zSD9US=HqKbTC1C7lnRl$@i@A%*b!K)ipPr_*T5$YNzib2H#8Qq82)&C!pm2@#Cm%m zl92yK$EI&V=VdD4(O)I(omGGM(x)m|(V3Vw`BBKEj?6(Xw91A1a>Gibwekm+QI>}vmI`~u zu$*~OGyuQNQRE%0cSasz3h>U`TA=%P09{KE;mI~h@!HoN zWW?9$p$^UY@TR>NP#QznJMTQG_-Tb+SH6Uy>eq4jxnX=Kzgc*;>_sm>97CQDrFg<| z6z|ZinY@o@Kca?VW0Vs1jtwYV4j&J2c;{x!=J_`NK(7?nVk(G$r-jFG!{0l|a3leR z&8bAr`#zyHuYRJ$nawCs+zqPxu7t4{@8T`4+AQG;doHfRea0h8$rmF-vQH^ABZ^gJKFgmiHVv`PQOWx0ds&>Syr^gFd3IIj%hM>`lCq9}&EbFHd5B zBTKe@&2oI;&umcMkWiJ)`@`;+=AwjJeN@mk10C1NWlwyMhi8wyg!GRAp@01-{7lY4 z)wAv>&wd7K|7e8v@T|zT;qQ2^+)w!3APpWg*omt*N3#4`PtfE0tw{WZIcj!c@c4US z%SZJHmL*YW{MKT$f7B7}esLQ}FL;diM?}HA?cMm?vFUh;^l5fc??I&FEP*JGY^eD5 z4p_%O2F0KH!69i?o+7m8>Gb;WdORo4oPQ%I0Y62To|mKefuC?wfGlqt@kY;g=TZ)h z#jyOR0`E%IDkNTdUK!?uZjNVOQFA(96l92jSg2e0Gt=g>KW!_dZTFFO)p3gSQ=ym_vhy{(eZz~Wq%ny0|7Q`Vr^N5A>(83|(iBvPFkyl7T5VF<0yc%H8CIPM0{N3p3xr`mzrCpIW$((>w%3 z(iV}S^auE!K_=>F?a-iq7F=p{7A^1PB4%+ST0f;0Sr48^UpKskDw^VOe4YZCvB#EN zm#@I!zX*2G@|7gt+8dVneZYT(SyI(vNs#nlIb3iGV$HrH+*{8Taz5-|PwDgnmXxE~8>}z3jyZi&o;&-P zIJx4l#m2Sl#INUW1wc~_wJzWT^{0XGd6+q#Vk-32N4x{2qN~x~^N+C98%rD)G6B5r z--Z515YCS}12wmV!5iaVxYF8D&}S+mun8X`1*h}b3*R;o&w202)CY&)KhaPo@1O~+ zEa(7bCPuL7S0ervwi;D^LTp|MA1hwi1n)ZSfcx}a>4;5ExQn+R`mZ~JeY(!$+;PHg z`XuCA5JYx3Ss}9eA#{5wLY)7e$JVKF^t`T%WNYJFf^QEIw@^Lkux&FyO*gPD*bd+P z?WLbQvBuX%L&+5v1yUmCOSW%OfHU+LvrNNI=u}k1o(w+Cxa{h~J_ie7O?eBruX!J5 zI#i%G6LDT&b{A56G968NxuB?X&Zw$Pjl@q?C3&xtv9@3{TdK1FoEn+}vi#(Tx=1mp zC@?3h6&GN!b`@50XFNOa`vm=1w}P+gS_0E&DG-gOU1V>J2flqF9Wi$Kc?0hJLWW zlm0U4fH!D|2XE7-#q1yS>6?qzEt>@mE0qQF?Jtn8 zl#o{-et>8VB;ZqaHN?bpvU2!~3i+=)gEa54C-hfg@?`OSOrG$m#D;S2VWkxIsc$Vie;0h6UnY+ z#B-Afo9|ame{`jB;;Mzjzit?}skM^>&39qr330Ng$ABnKO7ncs36vkW7p*n_K~`B` zBwf9a$V2&D{CssRds%M|smL{lYDb=fpN;}Bd*)F%^h)SZc_PUk98e{8bDQy}Co6IO zmP{dMx(U9zxD{L+QNVU0`{zZGB-UZ7P;R@xmq8- z+;9|M#fi+3p^N+@7DIIO<~vkC>RO_Oe8|K58bo!60giaPjIXTZ0)xkIF?_lZvp58V zEv>{u^{XcEtB@I_)N>TS7!o66euzAMsz{t;hj6xL9Cr8l z#vV;;XFn+(V8{QO?0i z*nSNM1?dn+6$fOIz773(Jr50e6oTWLGax7YG}z`-%xyZjjxn?`Wke_$%rw+84skB* zm%uOlj|~I#$s>i}&E-VT)e8<|q#56-8ZWF>s2v4;a2K%D44%$tyg4oeopKJz6oaibI2a3`K{RegxS zx)6Hs{9yO~Ot`=_5w4$FN#8i^h2Q!5;_5Xe>|oDvw(i+l%>BEcwO^aUY)qr8`YWs9 zm)4#5)t#BBsc-?D6&uU`ci|ka{#nW@m=A+zbD}|OU?_xVWRStTC{Q`8hjM=Dnq0?uTK+owy{{DxBO#RjANTj0%?pM79iQ+eG z6lqO%*}cL>uZE%Z(Kr0ba|_T?LtPxY;yq~pDvs@(ZnDjv>sbA?_R7nb)^aVgx3QVy zakwR{0a}z1cqQ8bt%^H@UmkVDh3})`SKJ32wSDP^pAj&BqXqk|S^*qg^M`%?%aSp8 ztw)4FVA4D74fc$9f;7msZ5>IoU96b|mbxOT*G*E7@sBv+&v}jqLg< zTY%jy5%$)ZDMZBY2^$;u738&@g8$YVp#CyZa={MbT{YgsldFLKI_jXm6Q7`OeKG6w zRu&wNt%LjjUcz-Xo^axr4cjwa1f2`$gR625Bfg;wY3#JY{@NW-H9!OuW}47X?{(0! z(%1PDZ$2}6u8-L6o_uOwnHC%2($0>?Y@s)DCFq3tF?7apWjX}Cpkm^zKtXdF6}!`b zwXWcSm5!gQmbGH~eDq2R)x4nQ@b}V+mrQ99cX4|8VIS(>sUrG^Y9g&EnLs;9r_)w) zUi2@!FHBu+Ex2T#O!3lBFml31^249lQt?|&RYQR$`^(m<>bP1P)0`CxEGi5s7xhO~ zv)^x{AB`B&%a<;#Iyy}qES++Ns`d}17OG3oqP$qj*+-k|Pk2OS_&|DubDugWhLBFEV5de##z1G$=map>D!M~Iqa*aKQsh^>YEbWtyMAH$akh-{qot2eY__n8n}v@ zvwP5TgV$>6dgO6>ZrV+b@L5x-oUJW=yK6 z64PO;4HgB;(|Qs1lwEKFsNc~+%?Jyn8 z@!N}Bi5o8b;4EA4D()U%QBjW-amof3R+@Z0>pn_f_8t>C-H1Lma})pAZ+)2TmP!5e zsHdIWm+)ta3VF~Wv$(J4-e$VSr72Ap54vx84sbh(=m8TOAZV)O+q}==b3aD14Svs= z1WAi3-rYJTR&pczVDLEK*XkbMRQ?`Qpf1IqwHwp(H_u|81<27YQc=_|_jA;rx|RGP z^KbM_ECI{B5AuIoDXqM=nvH5?h$e(E=*yNhBlc zUCPW`?8X>W!OG>_qAI_7eb%~rI^DCXiv41c!DRM5;5TkgXX?+EQ%0Y7)Ui@us)8}4 z&VCZm@fI9lc2x$Hzqn27j-91O+-8Xr)uyY%CqAXb3AAl zUr+XS<;*XE^oq4-ne$bvxMJ?VD=iX2_|He?QuifND!muYr!8u>P=!kp!Tzp9M*F`* z)OKq}ko5U9eMPt1qIk@budz3hzT?Ya`o<~f)4dCRVOZE51W<=mDg5^-8@t|xte(&< z65*PJd4a&*lUxZCL!Cx#uxW&^MSJO!$AAMeiY+36ScT#qlGj3pr3ak ztah-4b~gEB$(*YsXJ8I|yYer)`zP)I6Xp?%53BFSKbFoUmTAA}LtotBCd(?=*E9qr=k}my zoy&wyLJerYxt&bFYJ5$$4m3y^kux&4808^*bPXK<=RH*UJ{EB>Eb1%&Oq>&A_dtjB z2$Un?%OptB+$^TMRgv(*jxx2segXaa7w`+}B~5hnVVUbJ!9}T0B+Oh>U^`JqtoOVk z8~u#QbBE953b;bLR|_+>w62FG8?s4~ zSSh|B)e9m>Mr$64nps4;s=knvE%%7zz6B&O+lFlMR1mnO=?gv?H;_B`G+<1`GuDa` z=B_R^k?A>>q$WzA7>EfueRvtM8-GAgaEGz}r9^1t{}i-_6qDDhr3DI;EXgmuOHTD` zLA^KI!ITk8ydii7DSXk3k0#sUXuUM{k^5O3ZWTc+g)FnF!Q%LyA;X+h<QPpBplk&7-QV^I?M-%OkQ%QIM+dM)GusFTQL()i#% zU3_Jl2zgYPiQR-A)Ulxi7@M>j_`jcI-sU^v*sw7+JxvF+?NuUPHgm`$MF|p>{D|@E z3S0#gNV&v(9H!Bar8`UTlYuD0E9a4+HSh6By$l?-v7Y@=mCRlh z+ei9`{=#L}e&C4OY@9Li43@pP1kHCvVL6>A?Cs?WWR+GQw<|M&NqlykzGQrs|Ksrw zx*}U2<^J^~$EEmvces*(Grq0{|A}-co=6a(<0gXx%iN16*z5g zP4;AJ;nO`se7j|5VTm4#gP!XX!?ioN*8`V$@&Qq-C%M*k4^ha_1Ja2YV`@oWfxe(5p+{$qOQi40n63Ec&4$__|Cvfmwj!Shf;n1}+@b_P#Y;x!+&|_1} zS}xf^F3m8&%T^nb#a_8E>R~KwP(F>X&WI%DZad)T*-4~HW-0b)j)Y>m3GnE?LMRom zALlH&i=DNn3hv}xA#eHmL_Aw~ti0 zyaBzcYr%y{ebDrC6dUd8#q^8@h8o+3XA2#m%WjGx!Tvc_cC+tLtIqtSX20yln@o-H z&*^W0?b;%6B}5t&E*@hCN2`$Hy$;53&rI0f8AbRp|FJ9Ojhv zp`X7E;72bJQX_m8e`q8@-+%j%G;hC4BI~CJ4&0d`xKsIqyMFE=+_L^TyW#W#AY#o2 zJ#AHBtA_#}-?0|iDaNDYa({7W&>8qM@e!LCZ-s`Og)=5$e(&v)8%S?U7J4LJfFGwy z3+(@>2u$}E5x8dxZzTF63~aUqUTM3~htX5usJMZ7zo+3? z_=bgzM_`HbTnN;J`Jw&tB=<=HylXuj*Dh9tor$MFy1TH(33SOV(X+(3Q$RlEKf{WP zHo$h}`NX1H6$gby;%Tq)U|a1XC@@aP)uHRLmc=Zdu0#?DXg!1f+tUdYL#G3iiXisj z1uTcC9|_x?ENwr|Fjh)s|+vTm*Meb`q5@GxGjP74+KGl z;+>FPd>plYSV7dnrh?!#>)C2`Q=Y|02AVWHiBe`8qlNdandtrHpgph&O**{?^0MXp zqpEpW-vP{hX$ z1W5>IZo=7Oz58+0yJ`gfK+!Pc`5vg7a2cjJDG4|~MFf@uQw1~s#bKV1opAlieWIo; zg+`A=!g`^{dZx@x8u`@I?{}-wy4`(bZmlB@5r%_*tx&=>HYbShN;Sb{%_ z&Ly){jswwG7imXVMLZl92dh5Uf&kpE#t=<-ZJGr>mrjtl2^pi}jA^3RF zd)9Tr7;k#qh{O^H!SY8{c*HN8=2g$oks`q@rAfLL=*N&3G*HEr)O|OC@O+ z)tyM8#TmV&e$z)DhO_FMhS}{u2ci5@ReYlMHu(JCFR*^t68ZnGffXYU;k~Q?)T3s? z#;H|6|5qL`n6nLb=7unCRtsQS-5u~DE}7oO^o3>+R@04JSii`C-s z7x2sz+wq`!A^vjqD5<)34v*_uVb2>vFX5uwteO36Qj*z;-z8~)>34+BQ;uTDylYIz z-8qFXo>@l+=byv@-xtI44_%-~_6NH7g)n2|`$EWpkOADe*-+)#No?YC3agdg5L#zm z!A_eJ7<%>=TJdTgdi39C#&2Rd6!lgmSFWyrLDx=z)UzkqmcC(pyvBrtoGxO2AuFu# zHJP31nGeN1w~=OIgPY>dV(OPH%n_`G_g>$FYaR#Skz*A?XRj3c8b1q#bYkH3EFXp| zT%_Jd4!}j#I%uj&8K$ilK(CEDXhwuN-uBm$sOdkzT@^ekIM*KjsqA9iOdCNx`x)P; zTtm(X=Z15ie}R1#WiTUOl7P=CY*W=%_D7@`ySHN_cK#HEckbhOYzE@dPVnTR80=WsjDyN-u`NEpJfA56sEoGwc5~miznU_dmeQp z<)h}x*(CU=3Gi4yj33;|g$F)vV57F|z@kE4b&AI}(7zZX9Y0Czbx8-w*iIvPXO+>f z^Mh#i$v$#HKAGIwm4G+viIGe4A6dPGE$o?~Njhbf0eyP49a}6O0d`D|V9n|=)_i0; zbXl?*Y?g0@w>oP<&%Ac(Mafz)R@=({vbl(Re&yoTl5X&HR0L=}r;3e!%^~XJo&;z{ zldZ}AY>C?{oIQ0ldiXd4Y4r;4w0fmz)r1-ntayXZdb=Z&4Ht#~-%5}ddImgr+YI%x z0Un?An6+Eh2zU#3vVE@~Aloc2v?SpTyL#jXtMlzQma)=hH(M8h^zWKTEA}Rv{-_nL)P^3 zJo2}14dEU#h3BP}1#$Uri0+wi(zPayty|d)cO9>UcRbIqPki^mvrIX(g@>U04smcb zT#vcZ$bh>)i=dhKb?hdSg0B_$;TKF9?wG2L?;9523iqe@8NP^{zJ?GbjxovlJd1=@ zn_*MOKqCLO9oMZphMm7(6=vJxg^oT*29zv-eRVVR^n3to*CxU|lFNpLx3SOn?8AOh zLM9*+L+npDlIl|}boJXT_Li|GepvgJ{qOx0cDI@c{h_4=PT6e7FG|=7RDR5-bC)lN z^{FaYX7rbEujvzs`jtlJAAbQYXN#b-2@4@t(n)aWkCs5;(OXhC#fli5-b^ld?ji2B zCZsfT1vxN36VDR%`ZvUgGrjJkZ1tmUB-~hyESp(QzTy)kA*+lC>JZthRKsrDx{%x* zS_;PxXOQxmJTe-#o$%^f*=OAu_>r0}-n^!g3}jUhDT{4n-n|K8lKF~6KCK`Rgg)$< z15PCPlLmP=MN)8Pfwn+&?l)4*N(o+ln=gnJKTTGe*pPSMbqU{~6>GbUF-{&Fw%xXp z(P-?&Z8k0Vo`@}M>|$`YMJyio?PCoy-r}Oe$=F6LhSdH?fys&0bh3^!zIVA6+Mb(6 z47`LMX{8QU!?%^4tWm)Muj0s4-9!?jT0utRo3ZIi71E&zaP7_#_TB3_Fw@W;7L|yQ z$_6JE1{Y(s)+!=|5)iLzl7DW9JWkN%B=W%Go+ zCf;tGaVQS2msMm{{Xejnk|XbA6oB)}EQ+VIlel<#;n7#e1q&p^1?jK4$u$Q#vOZXs zY!Z=1F3ycrOO2G+nM-D}kHgCG&dQ&7RDodEN-=o3AK|v3NLYHXnp}vZiC^t9%q)}z zTfTf|56NMeQ@;}*GRwpVe$T=8(0_1jj}{48GLvX*c4KC9wNSp`FRV*Y#)4pR*z@)} zHt@U-!^7vmfOj#__1_5+kY_81qMws5oEk9ddJem7Y@x^BT7x+iy?Fbd8vLqFgT%~; zAs{9li*n`&IITIPq;@_Dz5Ic=HSZ&Lo#V09pe2rZ?LdwM3-`#T9;|xn8i|zzmF$Z| zIkMS(8VP!)fP?Rw;lzw6sJ`VJ_3ZjiBI9tD+*WfXxYQjVsnEoI3-8ko)`yveFYwzxMUeKg8G7*&@aK&lAgMB!%zoHUeB>bs95coH zzGdK31vl`lIC;G9MkJQFu>o8PEyph}+{8OAc-Y-pO8W7Sy^11pL`DAh;Au&yp17PGZGJmG+KbKIuFKu zjS_l~@?etAKDy*Z039K>gEp2 z54uIWl-{N{g-C#3wIv`hunbJ57lHSq7pa1SU#SYUe%f+jGUcFQLEC51;J6!)F8bz6 z_Y_yr6N>I&gGMA+YI7Ala+IV@9do&Z%NA2!H{Q}dM+WH!YUA|!t77!7u7}h;dKq<{ zN9d^@w)EFeE%YVrXlnfKNoum#mhvxBqGDFaP3a!~s^>}=vUMx)o-oYop>xDbO^Ny%cU)>uhN@lmRlar2;`P?>=9+;yGgN@K@^oN7A_79j&VsIZWytofqI)T0dW6!Zy>7XH zP47l%Z{wHr>L^Y6)Iu$KU40b&ha+V89Y_U_y6=O{Bp%!<$6(c9JMgaD1V-UCP|JA% zLSIaQ@|SerMTNyM5%yEApLbKCJDe%mcm_lmJ^_mlH3R=AGIZk>4SIpsO8Rtj47Fv~ z6bkV#QggMB0L7hIK;&o=&|0BJ&mP)G&r}x$>dP}}y$_4&l+CEMOl#;` z(0PJIbj_o^Ad`0!NN4u|(X2%J{oqqNv8s+fWH=KfEfKn-_Rj+wj|+J!Mb|-3L@jtS znFjPKu7Kc$6(D`#h-FGco~76NUz~CwqbA{FFYVuK2}W<2LA8V>P_cxDZh_^1r=Sc= zQ{17h@cb@sYz)4969aeO9j8jH4pDMbE4a}ex#rC0xtw($94X_0wRFYWaJoRhhF-LK zD_uh+({*8$wA+bSl#;&)JvQ={dOWd-in*0awJGP&!{YntzXnD0YQOKa>dG9b_^1Tl ziY|d045tChMpxjJ7y_t&ToA0b2y9v!0ut`eqNVCj(cc~Rg0=8I`_qZQ_+^6r>-=LdfVrb5^+uW4+DsFpZ zDmOmc-11P~6V9|fJ1r~oPMaND_saamt{Tpt*Gi_|mA}m1j9)YR9cy6rzE;+}{b{Fp zpxZ7^GCaX4g!?Q#4P3e20*YJ6`(-&bN7npW$3lzp=QqrKym~BecrD^a8^v2TC8^Mk z*X8Ie%WkTna;n9&Dr+-uMH#cS$H`ouF&#?AQHqLqJ%@QqOO~HH|3fyo=7r($}X_@(^_l(^y?XupDwAUO1+t;nIqFp6|Xj#+ndfe zYfU(5VKG<6e2e`FvytouvwxMI=DF;5bGvGmdql;L8&jde-Qu*u6z*vyN>1mA&j>MD!&YG-Cocf2-mi5O{Ie+*MIM4e+%wwdaIjGvqe8H(Pvk^Z> zi_mYH97X2=vmXs}%u~(VILA%$%vRPMveY?G+r&ib@>Jc1=2%Zz~sAK2}e)oV{ng*OIfc=W}+ok=IS4Bn*Y4`!NULBIrG(2 zkvV1zxd-1^aX+Lab6+X{sm%X2vay$YP;Z{#@lldCO`Z_(h)|0rC4%(k}JS~b<; z)TM3a>|#gG_Stth^-j$k<5zRdPOiMoc^(~M*4Z?hqd?wp);~$S2 z@3t-H6hE%zY%!PO?3nr3qUdRlh1l6hj*nZ)qDoz8qFS77vh_b~@n?L(wD>PJo zXDq}2E8#TG^R~G3PTE{_fXmT7;A2{gludt6S!NdFoo%KvtJysLNuODaQLV-Fb@MoP zTr15jX1JI+zdm3tnev}`%|SVfpUA~LLORhxP5q3;wbpl>pT~wd{YvM}sv;~c-fcEA zizt3-=@g)0Im|y{X*aHBY1!CdVN(1*6rFcm&hHn-8`?!_O45=dX+58F&K=rGc0^=L zB_mryY40?og+hzA@oi9%G8zaeTOx{*QC8&l{Ql_QUeD{Z&biNdU)L3C+IlLJcOpN6 zb2j_N`SgpRL(o*b(r^;;o!=SP_pdZ56tgj|wwExr7b!E|WTR@^s@H^ zhTk{Fe~XMkgck(9>ehJ7r30eP*TX-DKJ>d`iatI3aIo?zC|D|jqf6@G-D5-C(f9$% z9yEYd-4$~Dv5|h(3Q$S0F74oF`&q0~L(x;QSoKZ;|6E*&W5;tilrMlnBYQ!m){Wj= z)u+zpZ@BA{DY;~*(zAyJILlj_Zagi(6DA=jBk=(pj^?7^y45HP*=V_AE()Ey1GH=m z{*=!KeF>HE*ym|N_| z_cj-1akUkczjFoc`XfsB)zdIQ`38gyOHx+%dEE4C3?>^Zu*|==Xl~^$S~UPP@e|(? zKFtqV+FmKas{25k8ru;T6w*{ z!p$Z)7vu5S%2@O97yJwO4-fI3gXHyPaeEd2e>SVaTHQU$0LEoyypIB#?WLB z16uCifLrU_&`Y)&4#f(?-q%aXA><+k^;hCC=P8sPF^oFngYm;94%yN%=r~;m%dd+u z^T;|1x*JQ~b9zu9UzPt}=%5GRt4^kHAB7)QrGC{Rc+J}ms~f*_ZL<$xY~Obnj1qv$ zmycl~8sY7ZfAFzs30_=z5f+wR1nvE`@Zp6S*qYzwd`7QR(y2(g_o*H~ZkvUIRyX)| zi&3EYG@N;?uwwe9Vyrnwo)%pSMdQ`lcr}a7ys@d=`O>Ws`BVM{XI0OM?hXmJ6ARC z5x_C&C;a0h)3R85ApH5FL3=$JbeuJq%Zq4 zX?WKTiZY)_OAQn7R_P4fyG|NnTb^(?CcnX>`B|t?y9oEauqELG@-%Jm6P8I&XCCe* z>{H`1mU{OS+mrO3>3!qt2wXnGkW)L>22G&91}ZRgHyy^kQ^c8e%kk0+KHGk|fkN~M z&m2j^k){d^p;gSSeF;l(*w0$FUBj-B6L{2Cmliadl9Z+!O+Fb(!{67l0KXlqOE!l6 z)ied4sYl`S6Fux6H$)*mU&!&XID5F}AHLr(o)Vi3XtLC4v^ZRgvn%~6%t;8A{Ah%B z*#-D>Pc}sUQoxrZCvnG4H_qx_3a1Nc@YZ<@s@h6m>Si8By0^odo%yi({xnS9I2$@v zUWaGfv~lP1ShB7RC6lG;)N@suPHYyXz?%v5OqU}|&VYgqV(EO73hQ;^d(rW^)%{vd zEal+__OxRQ^HftuL4JKXXb^$^L(b54!WxfG&BMbzC(tjzoVX>q^s!3<*9M%xjiVam zEgnG?$*CkH`;n}KL}``Jeadjwg~*9mN9oF)G7amMx8(Kxd;0GI9_!m`O*=)jdpB)3|Lyi_*P z(5NY1-z`W@DQjrW)J@3u#>I2vW{~R=75q75LZKTHsXFH+Gsm>)S)-<-^hEA zbGN`8vlG{k9mopQl&bs{i;p2mGjvCR|nZLn-lE1N;J3Tg(sv0YoeRI z7i(P~z+4u@uzQn^lav2fa@)!=X|;(=V(C_PD!P$Hm)*nb`qDHq=tAj!rjTu_4zH$# z!>x-qc}FRaL0WRlX6wbVB$LZw`>O#M=V7}wFd0TlEc?~%&^{c34We4g;r(* z{RnWU=#66V7_ISH&+DzY+g_w-_cvf~! zfeG%dM5&1aG`EA%q3P}z@$4d&{E(+(BcJJthB%vjz>=kGSc@@Xxfo%hK{q=*(DYaW zb{Gz$+0x0ls&*0dN$K)WqbXP{)(2ss2l3_IH=MK^U?JZ(&Gk(@xDP*r>xF^%<@E&A zA5Or{UO}{_Tpk-XcEi6^LEN0yg-8B*p_WJ&kpF)epwouS>i%L@S`CUBPoka%XBz1p zp=;Pe-2zrQPGGz^1E?ooD_gY|n`3O#J zxrG7QM=?LahYW73((R(Fm}5T^ey>-CXJ+#`;mrnI-NUsIknD@o9Bbj0FAp8v?C4eV zOVB-C2B}+oQU8EGou0UmMl+JAr|klq+WrBCqcd>U!A11-p*d~YZ%_Uv>!6X&LA>I9 zs0nVTN&nu_k4c|trHl?F*Hpl|M{by+oPvt;XVR!eHdXna!(XQ~Xx@$-+G>yn=~IKD zKS_{MmP=FDO)om|te?hQ4Zzwj88~)bH7aXM(%Jb}v8XN;y}YjBs#Hs~a8-n_FXJHe zf&d;~Ed-A^1N03E#k4ISa8p4#dFS1tGvg{q{e1^i78bzdtC!$V1EA4@38+-jiS>t$ z;j5&xXy+|R4W+Z7-){gmobJPe#j?2V+YY>aN07q#)8<;k7@VWEiX^_q;gIx36y5$4 zpZIxj=(vr$|2l-TpKQu|sHJaqu5c>vpT#6@@sdNFk*F$mmPTQ&Y6iB9OF{!vB^+TD zkn16h{eE4jt}9HIn2v9rSaBlUM9!+_FwcMIcRYGmp9WozQS5~X$Q{q8S%&n{N8vXP z@gsZOEo4~EgX3Jy+ZOJeMm+Abpq~zye;fmo1i4i! zzH*(vPD4gY0m4H?`u!~w=g+kEzNsz(ojlLJ3@{^2kDJ}g9bCtavx<#{|c zVF3Fkh>-}Br#6*u*!wpGm!yoPubGy#;K6#5hBDl;z8d2;KE=`cr+8!1D->)Z8q89n z$z)4MY(44qD|tMT@djdt4Dp?SA8)By6YqLaC%5&n4jj1O!Iia3!A)~-=;}%a$z>W~ zA=iW-1WsW5Q+u=?b>{ranmB1yZFG6!0Y%+5(8$@r>gzV}@>4!=tIt4`&n!}HR3#Nf zVX~R#K`k-es4EtYzl3(c_Iew*SbG&Fe5r<>$^e+LTMG6}<563JJo!s}!CjK`ab5T; zNDCyqv$zCitXvL{dv`&<#0@n3n}?N34{^Q_qIk^{n9s)J6Mq-<2vWw%AYJ@?1xk3M|s&xv^awv zAMRvGlbN91MQ-a+XEeAH$vY7&2=zkQxbR&wr8&jX3d5Booj9JHrJthzi7WWA!2|bY zHgU^zl)26Cmr#1W9R2zI2Oo5mQO@DjG__reevOv^>tHj^b$SLjCo7Cro2yXfemNQ% z$R>;QZ8Szni@pYB;Y+!FxW>T>oi2VR`G&K^+Z#&lkNU8=ycNxQe_`rN5A3=shi5z2 zVuIl|$hX026!FQGk3;j|+_%k+&$22bnVVrX~+bglj&j@y`nogEWl z=_+0LojVP_%M{R=$!lof<`jywIu38U(qY%%BG|q)1l2CDMvu{HxG}K}BTgQ})@nWM zt*nQ=6K_GE!EOvt4Mm0AN%%+UDLhl%L&aB>$!1mquG$@Mrl&8vGI&oKO7I5z?wP{zA0I7xZ z8TEL69ZgCAAK6$KI@1ZaAB;sU@sE)C-$s(Y-H%x{{;2w4DsG=7iG@pr(Me+uYh7-` zW~3{!+Qo0N`}l5DS?PcU@3f%ltvhECIt@xz%%&hEKMb3m13}y3Q0etl{G?P5d#>HW zy!?wewS=GR_Mwhfupxk3y{VduII|I##eIaQua(hZdJ$)P<_2%8-6GzZ#!wuM+KzU? zYG~WB7=KzC;XT`VxTVI2wnk{s+HO(0X`F;+(Uq|EsxVBFxlG?P_mPMskG8}+(%56& z=p|o-Ti0cRtC$dkOmE@uWTL1lWj?+4n?=%hI5hMX#PN6dHXNVkurb3%?1zjx3yb)` ztZ#QS*>CME-#L<-+T72*c%RN~-@3^(@YXhyFMchiwppR@`NeZi+UOOhtb3l9R9MRU z^<0gcwfqVFmb*_Ew%((e?Po}QToOIq5lvdZjA@?0RO-x{LZ|OBcGGSaTNa?fqSwBI z-`7haI@<*XErghtvk>b`_)AK5t{5e-6z>b}!c^4*7@lc|R%YYyO41dx>-UmPFR#8} zy64m$G*Os>r(*syMzt=G88>PJ9N_B0%xavYSkoH+$G zC2r-Eb}sHmKdv{WUO*8W*OA|8 zF*c+p#GcOnL;5eb@q!9|nMwX=Gn=5?!5#Zv!aWLG#4Ru~f`I&O+!Bv4PU^We@0fSG z*=N^W-c#3GbgQhI#?)q!)ZV2a5+e`R{B)Q+kv6tG`vH4V_<*?_U&Or(aO56FmU2b; z`^f&*CIXpQnq>c+JJoN>wZ#^3qtEZ*tlyD1GjSKz9WB90cZv~Oyl{Ma437J-94{M* zqgGiAt$plAM^-H&t6y=vDg1BD@vx6PQ#W&_TW7|$@WUY^R)3<$JKmDtu@~gH`4-nE z;Q}E~&%;cAJ-W7+N4vLkbYk5%@`;jSD-D&I`$RR&?iE1!Uq3)~-!bkJxbj4dQ&zh?I)Jk z7vLDYga2NIvbPdJ?7ts=>`MMlHZ5=$tFGI}ig!n0pMf|2jy;I(+8#!_Jz~=vPS?!z zoEij?J=y%{oCEQk_rUp|fc#;C&ELmUOuG~fM~|Z8J^?&6_cjQ+^}{|DVTjH=!bQ0A zxZJOsQB_U@e|wsuP0w?B%Jopj>{nFdpAKs-r2yO21hXzwVr5`CW}JDB8ky%QbOt|1 zQaO(UcKhzkZU>x=ItoRA5{U zf2Lw$1Lsdlqn=PFtO~gWHy(t-xDX>axJ((cqISX9?P2id+iBnyr_IYD##9eAypGu+pY0==gshB)o4!fI0QDVYklJi(h8X|6V z%A^qU-BNMo_^s$;ehB+NF2_bYD@;GF1~M&D(5hydm9Vb79LRZ7ACGm!s+sa>(|i zZ~ZoO`Q9-a(hnz*_87Vg30OBi0L8Ec^wJq6SHv zbx>YJ1?`SHOY42($m7Ct(wLKs#s$8pn79w!<)72%BezLXs*>tkrC8JW-$YWk$+$fV zMeq6IHSP5n*LTKje>?IbOfT{p5+>8mVol2Qo<^mG(%gyHYMjuWD?GFH(JfyP3{7$y;jcjj5( zL|Y}?q%|IUpN_|xmE91^JaEuF3WNR&NA~mxZ5WKAd24pj{K(5@pM~;yCvuv3L(*o{ z#Ho|PR1uOh4nV6dPIx*FQ2w?Ac-nsAj?Ce}#wP}?4UJIp;}kqz-muVAJ z({|l8_-p1L@H}%LB6HrL<*F9^wDBGOtWp8}vJo!!q9JT{ItpubGvU6VKgd42g<6Bh zaVYa79#n6D^`kZLuD=B&Hnu^s!yO10dkQO(m(YpJTWD_WbQ*JNJZf7PfJSX0WDQ5N z_|8ZcS{lZl4e)zZ15*n1F{UjS^qJB+LAG%5Bhs4|&fL@&FtZvRCVk}wzIvaF={;#U zcaaRt${Y*({>_Kud$r*E_F#DPP!Sd_4Mn%dR#7>iDY;DH5*=2Crd$+-yU=9Hk+ zyj-04!2plR&%)W}W;jrxPVH}X==ItOBz}zlOuarrBI;qZ_+cOo4mi{A1*SA3#DS{* zF|r$-Mf>XmC@ysi$;Q|cz7eK-dfhnoN*$g$BS0pv1*uA0kP0q-#vP-U^pC$+cqNg< zBGZyt;?E?ud(Jy-zGsgH?hnD>XDP1fjlxE^6}bFnBB_m=M#aUQ&^CJm4$hUu#QJBD zp&AU;4-wB-k0GZXaTa(`iOp(OW7DJg@3>Yrt$klcK8FTrL5~ts?-O9Q!KYC%_$oes z#86XiEE9-SVw3idV`4S-D9W<1#pn|5ZQoDYX3=!6#h$X}m0(PW8}8ry228_+=^TE* z+Y?UVpzaXoIA;v3C^QGTaZ6Cl0+FBQ4zK-<(e+9K9_Uzw*Y4VI3)FMqg@Fog*!~-) z^ld}q-MQ$H(no3*y%bo~Me8CS(SPlY^wIhZNlz`q3EK7ixtA!lTW`euhqH0(M}1P3 zoDY|MgSod?j`7ADSmC!^HJtJEDCa0|MMh&jV&#r2ke+-DPxC!-Hg`{j9{K%DaJ?^s zPIq>`*p3SCjpE8}K^Sf0z|LsTXTP);vx@U?aBKNQ@(}C95<_H~MGM%tpoOfS-9wko z0Vr4^3G>$dpsDv2*lV2^G^8uQW|jXUoNBDP$LQPd5_zX3CveBhNK89sm4)8>`Ll)HzGxkGz?pN4YJ;gvwCYW_RX`4o-YZ32MG&d#hC%_*U3m! zzL<~uXNRKWeiIUnzJ>M1n{h)l;`3rryteB*yxA0wdRJ1hu*e%2|C&h03O|?z+9~`RkeN`CMyuEyse5#9pBX{)x25D2j>{7f~@h1m`mTj`8(=SZ=(F zyJp@CfBCb?)O0->O&d#9wz;_KR~c%##38p;4~OMDp!A9>w2oA87q4%i)bp9N<7Wx! z?HfkRBUkwM$Q`ettogPnV&+c*`dPVO9nA}VLcLEelS}+nde$SxVoJ4`R?rl7&q|v$ zl}%-hBi+cJf5(#t#7N62m-RbR=NU{^&P9>`yDRy%hNxjIYzR@Cj zxulL%7H3nlZz(zbuBD5IYH7}bMKrC*lkSEX(ev`x7@gmRb}0?`ZNh0)C5!F`B$X7SA%VPbBd)!Uu4tU`5M>?N9L&1%l=^a3P>F16qybBHlcJIIf0(WG9X-|*(!8&S>7(9k zGL2dcZ_77uUehBvr;1@bcrOF3I#;1lXg-#on~8VjZ-H)7FxrI5prB)s?`fD z`jCt>wD#bp)k^sCt0A7*QW^8K3BOu#Vod9RTOhr;lw0gdc#_`zue;PcaS10Xud&C(EPgVA9juBj^VfX z^}jj+>NlQYfAUYU0~pN`frrmp7*vQX=GOUmGRaxetTwueHs4iXaaZrs3}IIat*D|M zIrAy{!F5d2PiMgkRx`^^RVLwb1ONR0+RWkb$^G-JOPX&NnBTyf-87j}(M)!$Z7~Z9 z>4y8hQ!sO)H|~z}VKp!9n2($yGyU-&OK+@ThA)fRMxEPKpx8n!4#hNKRE5TeEkn7) z^SoZQWfaH%W|pjbh}?m-D6oDxnTh@FKzIdW+blZspd{gedAuQ`$OaZ zd?X{^0Xmg=i{~12keeFg0QbZ`G#^n`%;sAo4f676^2Q?=?IDAum)5~Yg>Nis+coCWkj@TD?_~=P z&Sn$prP)W%_e@Ae$~@?cjQN4`7N*QKFaycU?DkY1O_Z0R@&_Wc&8>rb7JZJZyHLti z_gB-N4+(U8AyL}hTUg`09Fw!eu;t`seEK;DzwB(qXFOpl%WB4Sw7 zEOzT;IxG0QoGpr0XNIBX|VB-N+a$zw_tg??+G=Xq<`+is4DglRC}xOs^C zD#2n@17@jQfsEwE5cqoles=fa>qql2Lp%^%f8b>`97t3R>Cwi-7|o}`@{7tkheq?(uqT*c`n zoZqDg-T}*erX?56;O!2k^wpF3Dw(kQ9A#E#$>@rOJ$21cqa62>BznPv>Id~`Sw4@| zdJ3?Z{~l2F9XX2m*ny!e8I^9Nq12Nn@Uv<=H!^K65r1}jM_Zh}C5NDN#}N2O7D9me zT{!qJ1&sXEAzx$`dKK=2?)PTU`cs%a{`{JLt-eoA@kZ!7Ll6^%a$uFkZwgXqrW`s$ zGhZ#Fjr+dgvQHl{z^(z#q6|njjD@NeexE9}9aRr6$G7f=Y;UR<+aC6SR79gWA?1_Y z^(&irr%NL7N7@t|-&Y6U`MbFH{`H`hU<{SrC3K=~BLBBe#<(prDWXxD-an9}CC2|y z;b11tTWy77rqnU-3&G4H6j>SP$lkA+&Sqv!WU*e0xNKc9Zq1?|GnL0$%(qjPB^pSw z;%Y_K@6}HN`t4MF&VaAsucXKPnZw&dJ7{^W1KBD`Q-s8POe!9a8hSbK?twEd-)?~h z$6i5vxFah6dkbG&=fM6|3UE_lA8*{-VV?T-?byiY-7Rbkuz8*)d(zoXqf$}Sxac+) zT`-;N3Jm5wxF}*C!S_#HI9$&(8)C8NNFp4XUIT^30_NuBx0&A?Y%{aTC{yd0_d{TPZ4yN32_WSEcfW2zSUkCt_Ha+~G{ahg_nT;k+a)YmqHw1l)s zFa0#8@6yABKv|@baB|-mNbaT_-OYUt@ktt}%{|3GzDm$s_=tnMN{}gPLUHFU>6tR$ zM`_x9%xsTF%}Bz|xC~nR*OQVW8Tr_mVV;sPX11E6hqD^&Ym$WTVKaaZoWP~SH$c)W z6yC3ss zls$c|#pYa}$`W;SnXr8hrm4zcTZ=g!35n(2{A}a}b@#!fXd&{{u10oZI_;isOI^)# z$&2q}S1!62k6zr2yTXHTy#F40Ie#K~s`ca51Z@a&Vq1TVgWE^)4I(p{38cqdH4TNZ5sf z?ky*Y)oQq)Y6`Ah&0k=}8X{(F#KgVF(K(?FE;$?q<$*-_+*yZj`q!aH;3nt@< ztH{>ViTbo-Q1@d1WUYzBrkXi0C$yTg7(WpTg;nw6y<4E)bpzLZ`U^&9rEzgd4K#0i zi{EbNqj5(f9(4E!T*Y~KJ|Pm`c>d(?VUEC{{X4k!HUZxrI)p3u=W{8M8cuw^4JRa5 z#`*h};|Hk(j8M#^|8Wjo3o=btTiE?xM^#X1>+)7O0L+3Ace<$mbKj|v$4;Kc=G$MWUJui(6BKm45` zi;{McctG|A{)tm2ZJr$Z{43}0i~qs4*=w0Wp*`y;bzymT#$w2^cMyH559aPXfC<)3 zFt@}C45xeJoX`rmCOQwNRh@$7=Py7+QX8!ws^hL3<)GU78+KV(qscWxxcX5Dw!JI? zx1@_uKRp&)+6%a!Re@ZIRWFzFLI47G_j5T}6Cua)B5FNMMV{g(?EXWxO0>1^c5HzV=$C3tpBhlv2MjWkn!z1ghadoa0 zHYj?ars!jQ_xmZvyer1gp-?|CU2|{Y@~6CR^dZIWuuaAfLB8v>e6c=irfe z{@J5oJIv!>1jWB<&B|vF!2QbcjGTQGJ#G8dWsH1I4SzeP#`=u(4FARX7>=I5Vb`Ip& zhS4yer-|DcP1adTSoyaYZa&@)_g2NwuYD=>t~8uVge6&zh%^&>EW`%iP9*0^>U8+9 z5Z!u{L7s~M(RW>VvDZ5Y4~+3oCm6;2Hb9m?pm$ z57s%O?)xt2Gy4hx1*7m#Gm}mj^MCijDy-UHfSE;QsIxc|H;kTOA?XQBeJF|PU6@7e zlolnf976_OtuU;&8&Y3|!0Kn$=z|)cUux&;*i=(7QR6gLZ;!;oIy#sqtdC>&O+nw3 zL(tag1U2@Ou)nSXCwWw0_J#!Pe>)ztQpVw4or(Nwm4~N!MX0{30%e-7U{3fO{C=wn z+yA>vk()}%Z}2z`=gHD4-!Qa1=!Cygj4Awp3N3%AN-iV1WN~=}TL%UCz4LXvxvc@| zeg;1Nejk5%H=v^T4Sc#p5)G$Hq4SpUn7gO}XI{3Ynq_6QM5zNO-3XyBo-{iYbO8lh z|G@6h!;rrB7)JB=!J9qd)qseuZdK-K$F&Ar219K z^ly4J{fmyEv+g=<^8sU~pgD&%yZfoRtw>mEMbdmEQTUc)Q=d0YASDq8F% z%v5UlRVMWw9?KA=-Lb6K$uu5|l>!eJ@8JX3+bX-}plF6LCo zXQE2JOs2!9#DJCJzD=dn=FWASvKC-0k3n%i= z#8tR7GoRfnf55y7huEEY()3NFhu=r$;X$9ZOuNL5t#n(#?g^a4SNt7dT2wLYCk>85@5;I1!P~Fz_Rt?ndH7rtni5x{oT-l8H)l? z<4z>4Q#?bPsGU|mUPMEB79@W%fO2aKa3?PU+h!)>->=h{Kg(Oinm1^GRp>OvLSk@rGCIt0Sf9MN}^H;>wuO;BLXbnsfn}?xuU%&^0b?99n zNS(@EC^V%7XD!}^7N0$_{NWTlcJDSOFYdx&qcZIG*a!b=jdsz!~Zg}Clt0;<08$H^a~NnzO> z>Xwuw3!7^&%exgS%;#fC8?v|W3|RYPzQ$qY0JMMNb7wCEA@;!)y0-fd{jrc{PTW4W zG;bmER#9P#@A|Oj(0y#b><~S=$oJduSVqCi4d_RtDV`s?4ho*puy4mH&ZEs9ZrqB7 zs-a|hCzVWncKkDh*K_P_?m$W$96LNRo+;-4qeVj(NImBr-TW%YYRx*>*1-{Gp*qHVTd5+g zUibukBc9;cNJo5d`8mA)egfL!G-*(O9{+qN!=GLMKm#>-l3!bZ4<-R>uuv?KE=ChR zTWD7}fqHk2;1Z)cs^50~;Yv96H1%Q+IU%_p(LE!6hrJhcdy@ww?T zikp6z+=lj2P(T=2_DQf8YbLQfl0xi9)h622=uEb1j4p)Tz_SHyxW>B#jh2LPC2hXk zHi0n^vM>s6AFPI}J7myq`cC+xFN2Htoa8epOHey}1e|$O@PRR(U93vM%YH_r8uNEyB!hXLNa$h;x?=a-ptu++GQJh;SF+8jT`&@UNEZ zon{Uty*!u_egr1SN@I4JE?&937YlBlh7Fo`V0}#+e5*`@?Khu-X3P@Q4KTs{!~-ZD zQHLsp{9ZFQ2okzGL9i?cc8`q!H^KXmJ2)R2F0X^U-_@`ta;KTo>pZi*ZPvVKy>Vs{ zx!=ut;`o0**YhSXlHdtYVO)cOlPN8JSJ zPii2WRLiAJOyUHVHgj`7hI6J_qg>W=eeUm!X)rtA62}~DhU!ZjQGA&k{ndI2&5!xo zckgXDwYLN(9ID4P$0TTAbTYn(--!(+eOP?3oVWB!G_OR*nESG909xKD;r!fr=-8)? z`U%r;Zud!?)%23PzAy+JI?d4QwId3uXW+3h{Q5C;2#1OiFr!?S&JO#iB;j3Yeg8kDP@v_4a0<>DXG-jU7i0hWp`7 z*)Gt~ZG}hH$)MtK8LowVhVz$h0Gp_Zo=$OiT0Rv%mp*{D(Z8UrdW*YpMiMke+`w~0 z2aL6Bp>AstEPi^_bX8lH>C~I&c*1MvLM2ZWbdSrUEx)$XSXKW1*blASn;_}nUi6OM zM)wx0a>;QqT&}<}xEL3T=UURR;xS)e^s(@fp0cS!0sR6oTj}!cfe&LHC3&m!Taaw+B^@?Z*{Y-Jbh;YeAF`yyB?L{abNzoY3&O>e`SGv&MG+QYKV6- z%rQzNgL~z9jv);P+_>0>y-=Vf=@bvU5pegHI#x$4?JOML@J1+R=}KUam=lkfb$LXguwoI zIM)yXQa*YZU*p8x9J_$~9JGx)YxM&*?Hz{4n$z(}OB`(d(#F@b*5Gw%CA>DMfUmSn z(PW7d@4xf}vzotlyv#q-&^O&0B@ZRx*VatX=-$Ui*h<0ZQ$26pyIGv}+%vG~TRtq& zcmU<1w)nH2Kxz9r7<~2$KCB%96SG#BwRR#-zjhZa|7F6TXj4?2ISN9#K4?E#g7`B` zF1m}ag|xp1ZWg8}^q>(18?y`x(yjml~K0hGVUt4%US@;o9|uIAZXZx2!*hv(Qn%bHU!cM-s}s;Er){Cn6Sx zMYEy5^A;3;*~Hl#)8u-BUUO3q7s7wCN+4f3!bOU!eds?g_N2|jGY3#GlXc7AWtSw_cpSY~c*R@iKf;+UNXBPP+At840RAQAcxN;o z#5Z4p4=w}vL-jHQwFsk4O(X8SdI3@f9B`KWb`l*+hzee~ec&N3GPr~@WiFxh(eG&U z^)okOY{Qubo#9PdMYTSBuNVV9 z)~*n$5&}2=v_j(QetdoSB?c%d5xZ|o$`aPNM`;VXu2i6g2{PP=t*1EOIgHxTG3GTo`>8ThLFpT5;Khylv2RPS~tK6Aur{P~sJ``m%!`=&X(NunfdtmI% zjk(C@V*C@ZQuPxCVjkt)7sp9+CZo}=-FUMr3Ketgaqd6>-sIQFVns2!cdC-!-RASJ zvXwZM&(6L)GJ+eXXi#m`B0f*n09SSlz~xit;lRH`;B&Hx+x)JPdsXleW(PX(cUnM~ zP9nF_@GIBw?I>p4|;%!wEv>-&G$m>-s`wclR;Qyz>M% zrT;-;u^L8Ymzf1CVCJjzNJz|t<`!$yHJ?w^BGpEnTkvDvy z`g@XL{9;3guo>zE>8((-hhz8p*bw=Ic%NDcS2d{fwz#>$VnOpavR_%d))f3( zx*FH0$xy=DgUnrL6;tSlf@8r80`lBqZZ~pZlEiC?$%9B#Edr zXlj$vvR9Feiew~8Bq=E+qL3&G4QWpeCH-2e-}4Xnd|tfweeQEz*Y|pVVlm%-ct7t0 zC&_02VZmGYfULfj@u@M@eAJ{D{Px5_wD+B$^#n|!{vHFH+rI|c%<1{UTi2fAU0kF2 zw!rCpc-|Fa; z%RF2;*%4Az3h=d(O!`{l^E zshH;OjNm&iY~%MTNAbUot9_X}-y45e4mRzR-`9(izCt=LYjG z?nXjx^#b;0$dmZ;QVK(V+P8fSpW`x`MyAM<)8A&SnlXsF_bsNsP3aVoR?U~TM)NzR zocP(P?|J{E8~pWEb^MQk@i6JT4z#?^gXt?a@@+r$_!S>JsVX|0x9K^|7n`*5TB#2F z+K~hK5wBM9bpqqT*lG?1jx46h^Ug7?Ia|Qg?KuRUG=&@IRI$}b8#j&_M=^aoDI_M) zqv#xAd0mSa)%1wfe}|q6TXEO3U_9D0nvd-=v^hh?d(Xlg3pAin}gwC7y9~T`!Ri>|RI?a-%4=)qwU{trqmD z6O_2khZfjPr$fetv`lzsy?XS7d{v}*S^3lav~&A-RkKZe{s!X?^|H;WCi-GpM-1bYhl2V;dpS| zD{%O#h$=dt>7P>%eOoHe51c%Nt}h9~fup?8Q~4@&jQ>V?x25^_%{F{`=_G#LxoLb! zUJ$C7cniJ^QyemPI^ReVzEYvV~MUe^Nb; z;wRTy;GBhHa6z~NcI6aN1YMvHqtoc*h|iRJ{ymwBUeGJ9i3W!prM4Xj^!D6*-mJ`- zuiZV7PY^QrCpSo;eUCi4yaYbqZ8YD{$@1sd6_VQMa(YnkkR(UhVt(}<{BPa}y7c5M zJ>Pemo|tq|VYD(YsWP5VZ=KFtoeqaw%NUT#w}ADp9-*s$367a9Fv<4*quKLic-hTA z=+O}ajA`v*{?iMYv1<~AolT&3rB$qlkn9Y{xXG@h%knQ1k8k!MI0!Re)vp?gEKi|>M_6miSrW3$YQVRQp z1EXK!s(}vH26^(CQ|$Q$NppVnJ`G-{LYh~Zb&+mdE2V|*M=7>#EY4Uo z6!i_;;hfn3k^MwX^nW%VUpvkv>AVHh@?jgr{XT=bA4;$!crg~8UW!SF7NFg$QFwE_ zC$-iOw+krqrUCyXUz^|V zI+|ZGJ&SkhZsLue{NhjLHKNJkHq39kjoSOeQK8Zw$5@WXf<^lDo9WUyEkpX!eGE%R zo<=->0@n;$MHz+*Xo!&)$y8V1?W7Lu)=Crn0CptS98Vg)V@a)j7flYWq@D8Z)XOSq zV%jGPxvtFb*`Yw+XXw$m!kMI%r$cx5@21zjd#N_}FRjy7=KH^y@CPsCp{>se;STQ$CKP+%BK6g&Q+2?JiZax%34T;>l$8{${!fZ{A$Z0^~F7P}I#Caw%;EW3xH`@Z17?lXe+cA9i; zi)pL8K6*Z$hShCTG3xn99MS87H?@55<(oJ#t=SKfD!QOv-38aT--6B`Rbb~;hWz<2 zcvm?O!;>`7RaG0U{{9DfPAOD-Rmf>}7SaZVQS|XCysqH*X)=^z zp-qP#xzfvv@ucsXOy_GZ(0tEY+IZ+QL?ze4sPH|o_u?*E_9TI<{^ruS#}ZI?NFR#r zy1;jo95x@b#b%9-cCL>uvLzYCi%~8c8j#O_Lab`#NUuKLdtfZ|E$1 ze>?yyV^5){g&a+Db)@dlCUkBeME?@DlfP{}csqB3-~B4-iqJ*$8fYx+Pcb~^g+ zuEiwxdl(qH9-LjD!sTU`pzl3;TAFOTy6TkP^zvmvFwK@fc zQitKY`5zX<>to%{F?jm=UQnMk6{nfTg2E0fxXh<H zr!S5drR>V(GA**W4Ox%4G2NBy*RrwDV}FYMC+s-yOAf_k@9*GH-wNaApM!j%r)>PI zXh>As3107CvFDNk-}C7c_#LPq_%l~w-w8Xc`tTMqItuZltl+D-VuO@;c9t#ZRJ?&C^+({~Ux*GN%kV+cGx+JcfW?i~WCJWMIrTl( zV05U4$;=$gCYYAP(<8}nt#%LN&Mbzx=IdIE-?-2?Z_(^>J53|Q?{0NyQH z&^-)*zS($pn3B$oAdq@uKSvT z*H%6RLn9ZMwtp`u)gxp#x5DEI!R_Fnx>wFXLgk^ZjB;56nq{R zVw8kAe*Pf%S?OiAoYO^hIs6S1*cDiapm=fV+1K*haO>_`XSvE)C2@$0{T6tO&>wnNnpDPQL1nk7i`D0F5fs4}f%Rf$>fJ1qh9U-_Zy12HaI;R8pv zCV;Ebe_&g-7xonj+T4#w{G(mPL_v}`%*_VFzO2N++R?21;BK~m@kzERZ3GN9l*Pu2 zDPUi2i#7Q&sCHoye7iA&yA%G1o096riFb}>tEXIM(_hbI2R(<8(V6L_`1KpQjOb(5 zE^%yCY^Qjb@c-hmzY80B?qI>S&9FgNRyZF9vFeRhxN~#@^bK%h4L?h;Q^*2Nt}FmG zV+lxYSOfQV3HR{3D@-PF3`l*`6nqS?Am@oWd`cO@o^;NG`0>%8`QDUGFzMx7O;gzC zo<43z)hagY-wifgwvqD~sDbI@t+4NvBKi#o#?pVsQO|F=pb{BHpqbNR>lXWrso#fEZS zrd2Q`&>gkx0FQ2{1A`196etsf(fZfets}45yJ?5T0px<@?!2m+Mg7FN^UO7Wl4fG<39= zz$Wu>crv6y>~#Bpc$xV!F6NUPEG_C{n*2eylM{tC#xpQNUIUdM-(#+(y4;}efk%t(qsh|d4pjp)Uh*{?YZ*jTI~0! zS}v*}nY)!N!_xQdW=pIbVaCf#(E8W_ybldP?@DK=oK?nTc2B}GLup)N_zh+{5_NgH z(!8HDDKY;oN=~gr<%sS0@m>?-8>c~v_#d}ZX^PlSY|CAkaT}yP_29R-nI&`Dc<_TN zF06=#F=ofv*)xd{qoe^lMs9~s4TCY;V?7or$0Gdxhm9V)DP~7GVV4QsPpXDb0^@D( z$0b1ev%mHU<9$^n=#d1oXqs1M&tzg>6%`k1}AxNF; zk7wRY!%e2jSSX&xv<%O4CZZlN4G=t60YBl`(P?-#_a|fA4HopS6usLrpjey@%SXxM zr$rgCp|ccbjvE4vBTlk|6|$gm{{nllcRk4Hy@I6&hGY2|DUAGh2e<)7u=ar@3zi+l zg6`_G8d(Q?5-)*c%s9OIekqgJjAB|!_U!H7CG1MfM^>9t0}cff(QLgjigw##uSO7D z+L3@$;T_UzcQE8$K%U+?DA=C_^yH}EdcQ1Er$`rQhm<=iMS2drcz=gI zbD9d*9@j9XJEl~vb%Ks&U7{H4E@tRwCfw!2z#)AzC=UeYXFY?v+?mgf>m3FYPhVku z3LLz-DTcaZcVI_n5Bp=32aDD>fUoBv_T#rFn=QJx7$q0_;hIMVm>BRH&R8eIP!&(;-?aty_-#PVt0QUMn}4|TLkxb> z0W3KC0nTcC5uXU#Dyq%cE3!QOpY`_PH>{m*R#{7oek+bUB_-ZgdsdWkc&d2)p-fS) zf4RsnL|(ke`?z)Ki3F=L-5mE$d%P$!YKB$6eY7Y%daBheha5{g$6k?Ic(=&UTvp_G zU^z4V@mhS=sl#fJelRrm8nN*;YHZ@AU!t=nl0vJiz@GhgRh(ClBX&wLvtB#ImCar{ zk4s#$OZ>-u24nyHxEc2GoKlIh*ih%9Wpc(yt0Qw{L>76Itj3!hwoJRao~f?gWBszL z%6f;Wh&!BZBJ!R;&9W-ZQMAwQiDX_OR^#hbiVZFx#YAU{zEm_~`UF!%6j4a;v7m z2?_e*hu^9G5V?0cC>%z3ff_uC@V##ZqTrC95z z>mJq%EtDW3!B@zF*NcO0sBs1Fbgi5kvn_46oa0hEqr`V^q>J8aSVuPAYX1gJWD>n({8Zum%w)lZ{*qrf}I(IEa&0ed- zmz_UZA0I5t5F3w)k6Xxz)2)_UM}}Hj|6D1uKDK)flS{hJy*j>4oIir#EVqW6%S`dUVOt=yUmnHV9~OAWGGM&G!JnNs2< zBU44TPwlz+`IE%|Hkyfc4V^EN^{)_1R_(P?ew-k#`Kl|9SUE`S@!=;6YZ=ZN1z1=b z&Ymb*D^9c4m2tOR`pZcyKhc_Nvoa80OOoJZ`qx@p{tK~EnEOSv?DsxV&heGv=8_aP z_kbGrWtIKnJPB+EpE>&>)eHLOXYaQzipBhD)eJ?D>{v)drA@i5(8Np4p z+hZL*UW3WxxN&wfw^*knDzdPO46bxWoY?2=EfLd-w!U-S!TQ>XGMERzCbt!d`A7Ri zfBr>Se|8PE3T>GV6}zoi#;<)`)VU_1ORGyHy==OO9w&%Yr!E&;+~^e5zm0^<;z$@e z$P0Gp26NqKy2Q!T`$b&`CUFicbF5YKS477jMshl-=G?U)6Y-J86!FZ#lf;kiEwbKv$R19}U*^P%FNrf6jfWp;UjCfBcVO!QyEZ>BM2JeRj~i1qauIc!brIHvklncJgwgnd8t$hr*^ ztQz0$5u1(QYUR8j(Q@g@6jH8qp!bT=)DalV&;JB`^yATdNI@Ll(iPbF!rrR&-D#LH zuL#us*1-0hEqG4J3k5D4jaJ?RHT?r{lad+!Hr!5bYrQF7V8Ip~aUlH!MVfK38~aaW zVqks)ScLosmrQPep8I|Hp???s;zkIb(=%*48wrZ@ztK;@^Y>CJmTYUBu*9Mb`t#qw zl@)`aebxXd*<1=;*X9Y0D9h0?YaOO-QH0=*2dv|j z>tdSV^dm1Aw*Cp5T$01uN2p?P`A2qF`6@KeTZ1vCLHOh65p+|Sj_VSwaItwfPI)cM z&+(GtKMa0I{j(0y{pV#Am03@DOWxqsk}AX$F$Ju8BtDW}z-FtPqP_fSFizBlSI2gN z-WyXq`9~hLlHcOh>Q#_s5DJGA67eNi(bym>8ZBg2R=7{2eiK(RV1e{-RVV2jkD`*{ zSyZt16z*@x!qAUf@ldWd^^^^z72hUP&BA(ys|K=T!NJhGW-7EePJ*}hCZLae1jhEd z;!-F=TOns^`KcL?=_u2WiPLG@bHV?1Glqsfegztat?>2JB$Rj9LetvZ=!e+?TD`Ri zCkgL~scls-f7eKI6?9CGyT8zH%T~1CD2v=#Sv0S#TBMz4oNL7)5tSodo_uH0V@mUJA}C99)RjIcXSpN!=J zb?AHCo>+m1s?tZ1y8mFOO`0^=IQz}CHnqPmh_U^@aCd}m~!Njz$b}^qirW3 zGQkVasZ~M8nE8+yxEw~@Y$Ip05W*xkA$OfF&P|=ed3pq~2iL0D)WZXz)^9R~>lC8r z(^M>alOT9lJ<+nV95+sRK|cE&(e0Ho9{WC0V2+qkzxrV6^zRWo{^g{TF_lbxv`I$D zPimbAqh+aAslo6Y^U7~wotFng^MBvbG4>PIAM&J#l+U#3(IdL4`$ZENb=)27elg^mqv|+^%)o%w*Hfdn#nTc7=g0DGcC=G~^0fU1Q zu;6(!D18qW@)~o{(RwV}Sr*ct1y2QbW*6Rn5ddettOmc-pD_O68@|=Ep5NJ3!e6Ui z#6*o}*z--lKw4nwq3!^qo>f?enmL>O$W+`q=L9D5r^dLN4Myi zIH95l9k;4LuEy#3}fr^sCU@qJ>3|Jz>$Blkob@d)Tto3P0?0 z$DK24al#j#E`HRY=V7l%~e3d&yI19c(*n>(ayK{(IGB(g& zdoP-(dYx9>O{3+Dwv)S&9hsUbQp&PZ7%%vnm7hMOGqRPWm92_*i=)8ld?~~pK2Ou^ z(rI#2F!d;(L+^@mEZUcWOZ`@0+aY6ovcdt=C5PeL7CH2v^aFHcOW_iE!5p{%qGfN` zRh4M=Iqm>^6;%!nN#|grI33zL??a%;74W7+@J&<#t^fAId4(j1Ei1(nPf~C@mf{0H zH#)!DoF+tRk=j~c8hmRp{rlrUw~ofsf3`d5X8$JoW^$b3j)zcRmhjo+&ohnoHEe72 zOm=OOA4uE2XX}R#1>ZPbocBx~r~3Z|g=d-IVHysYVhUc)ZgkUMi~NoaqQXOuaT|7G zvDY2kqOyr!@ZS_(?eQ4ipkgO3{4x_a8{1;j_LCHHIgK{ljw5oO0A+g9K%964Jg0rc zqxva=pQRah4wom5=V#G!VG|brtHwod(lEC6IGQb4;g$}qfNKLqDXo<>czBe>hp z7cXW!qT~4mG;-4c3RXy@4H5D5^>R9$SapRB^lM-y3!buUQ5wx$pF-Z7lF0GxKH6Lp zNKa?(rY*(slsPhj##}u{qw~yYR^b>DKh_|%6Q|&(bSkv34TZQBiL8IQHyl6h3}e?E z#YHWruqUqq%gkcoU4=cU*?xfGDtBP{^KS6Sxe4;`!-!_grWfyq(zcq5LY^)HuYOWYTwZ8*>o1tgc1XeHOgjq5=Hr^MC2Ry(X`-MquS_{X@x;R%jVH z44*Y2pt%x~-P>?C?<0e|Max+H9cA!0@f39CaH0RQ zO#I$y8*}%W4pRr|!Qgc#U}T<>N>b{#hpF6zl58(Rg$u^Bxse3 z2^~s$g5O7uAbGni`eSYkmA@L`Venb3HXX*L>P55F2lsIE#A$4Nek(`~{EWpCEv&Y& z6@p|Fv8-_lOudlK7S!9ZlZ_KV!hIqXtP}QYVN(S5kf0YF-c5Tgqu|glDNLv<;>Mnt z4LM~MOnQ+NwokBQ4Oe!E!(W?l+ZJg;{GUcnvEEMXCU=6(ohrj@Hs0aRYsYZoJ2%3| z!0nj4PXz|vyU9&)&tsR)k6`=W3uoj3PMo@UE=D@&<8-4I{4`t_hF^+=pM%$;^D!Y? zZ0d?0rUh(p%_(LsF$spfzQ<)W>f)ACWn8fI2s@#z#jUsW;O1MM1dp zIU)tu4O+rfYsD;oY7{ptI*yZFxC0pEApTTgevU;;K|+;Hs|*+rWsSu)%UDnQvRtdSbZeh)m99nHaZDj4H=e`K8HJy zatywAjH7~&19^A#Yh05lv7zY|!W}w*ZC`MSeJs8r_8*|Y{CgB3Z=)1>oGOG}>7H2h z+kplLRWQBS4#o!nm*1a)IvsEE?ZR|AKX)2*WM{E5If1)*NCES|CepL0MZCnEH!yP0 zC)iXI%&aaf<1(ABi=}7ih#qGZ!I%%TP~@A-HPt4vvtEv%;ID{Q9}jRf8<8uRp~gLY zxt@(mKFB<3W!SP!f^Pm^i&#`IYYO$}JY5con>7nq!@3oi;wHkJ_1ZANX&+OXepVbQ z=g+njjD;ar>SmG?5gM>$3{xy=ytU`@kAB9EWpj13Sc;c9X2T*3D$W z)gv(fxjlJpi)E*h4|6}S)^is^9Jm3NySbxFXTxB{$6TgeJ9le|26G%O?ENp~GyA#Q zK$72qCdUt;+M*if_CX3Wg$(46)uG^(GZ{>taWJ^~8>i;}6x#kaL%WF!cXH2f)}~vF z8!~TDvYa&gJHno`yS7{`E#4+_I6nxa1sy{+A&fs3w3(0jOnfl2wb{Gh+2%!^z0Jjf z`=}cw_?oW^nd1q$a7fM`|9d$BXWtWYcqZ$qVZJj(L|(ynp&ckPXv2(NU0(I19RDMs zm{#99z{@Va%16XM<%^nB_(RQR{Gto1`9uBkHadPXHd_PV@%KNdlH)TOnkVEjr6nfv z*{eMG;gy?sZ*OB7yx}|Ua512oj#Cu0BZd+`gws%WW8_08;0gOf81@KpYRe;qW zO+|l*s{a^MMUkYnQaz{Gh?GHfsJ&~%comu}|MHpHOZ zls{-@qeqF$4)6_*=Xtl8P5jMHflnS*PdnUx|>v!?q8$R-y<@2e^B#ZKd+39;{Anl!KNL!Cd)5)ALdYPC=BL+UCc{63`P5g5F z7+r?rd<@8y+eS*|#S~vXmQ3Hi!)$@)^k8ict$$cWHTkb;WwkmQbTz^(6CFH$Jc&y9 z1vEudo0eQp!h8MCF?~flDp|ck?`Om4>B}xO9qEHxmZjp3rZf2JZ46mAuB6EZO|+i# zVhc1cuxXjsScddOKJM3S-u>r7e(*zMtOznj^*zaW(|sa;s#BNu(HX$+IuL@ELSFKv zz*D5x94Z_)#T@NNnBbJpJJ?n8p`q2hkL*pp=U^K@G@_AjNFIRuCjgH7brZ{bU3d%O zXWe}@iZ6IB$nVAGc%~CCAxP`_VZ1-rWYn1?}$3s2rHq zlOk5NW#VCS@@OdIMn7BD(^cE~RQlc@zfC)j2_{$2C|C#Ua;KmucMdKwP$b9VO7vW^ zpT@k2L%Gc@*l%zLFQ>fVKdCF*ygOlOGs$TwpWCsKpBR$N53jz@ml${OykZsq&$5Dd z|8tq2<#C8FWNhj<3~J;;p_(@j(@z$!_dfEX>oUVW;-f+;+BAUR_$t6ydgQ%b+G9rh==d;0RMJ5q||fBu1WH?~8cN>=;g>P>b{05CikVnmLd2bHyqv<}sGKwMny8@1W>=2EyxIuny7xbBy;LH$JuYJhULNOm z?pC0V0sA0z=Nxu2uoIVvJJ93gRg~@SLi3+{aIfo9RGwJKEqa&A&O4`b@*44QE9W;i ztlX3v)pu0bhi_)aFFpw@*7J;9bl9E?>De}1t!27*fNj-;_m7jYnMFr#syJ_NR2U zW(PWJN(!7%1>WlJO>7mnbJ_2;dFiKWboROpC`k|G^W?K|*+?&O&kh^@mx>e(p8t(~ zw7-Z|Lf-295DU?<>s#P-+jdqzZxMG_LYuD4t%r~MyUuq`BK!<&nYghxK8DFr&0I8gQYs2R7DEW&^$)=B{5Z!BlMvXlf4^dzS8p ziFbyvj4~zW@aHRMq&I8C>esg`W!#bCqX2 zMCuF8QR9yUs~9ke(-2m5{dUHXJ(S~;f0w|RU~MRo>0nRvKf?TzYTWfH`dqE^YRcr+%xswwcW{$G+O1m&@p=GH^o#NCl80=tXEIY5wh*-oHNM4VQ-TweDED6@_HiLwsd~W2wYHP*REFp)z8?{S~ zS!dEgu#$bxeS3a{J91SIgc2TZ;ul+SvEao$H&Y&blH<9t^Q~>(y1t`Effu4}dYti7 zOu71O87@g?2ETOqU9@kT3WWkY_Yx^{v(txiGbX=6_vd?=T&b9S%5M-l$U5Qe`5YF& z4Dp%8V3=RDf_t)62gkd*vc8=m)`1~$ptx@{=bAQ_3n~@EqI=5huJ;vgbN@~p@3oB$ z-IpXTG?v9WlK}4b$W@|alZQiodycr@vr_E+PKU1%vZ$g3HuQn*J0bKAFr zR;moe($BtJ_Q&D;>Xv%2y{N+Ad?Kh_;9*hFJeED@Jeo__u#mz~G1g|n^8IJH=|`2U zteVTHQQ;sNTd$&JUo-i$tC#Tp;hwzX0U4Ai^o1c2@h~pt5m(?r>`9Ox{8%YQ&F492 zpRGc?ojknw9RYp{ui(ZOUmCbUkN&C~Q}5hnYL&~Sq^%d}*6=u5Z{SUH+*eazu?#;% zN`rUaX3t-`t&WjJn=#~a1it!~PwVD6Q2haK$|z|>m;XX=*orjFuTMg^6TgIwm^#f+ ze#%wnU*N(=-(Wk0-M|;Ox3K(~37%+?1L7lIFB5b9}rmBT1@lld+ziFS2($uc&8sj_-Yjs{>Y^?Sl^q(Mzbik zd`T|l>m8 zf^O}1qGPAS=*jKN^w;?@$QYKv!POdAc3g(97}-j{ceIeXcLS|qLA3GhN>bTShZCPF z;bhw}U{NFdeOfK4b=zXH3+gAY=Qqf7O$Y7gpP}*dOL+18XWU)mi>jY4;AAyPdL_3E zJ`Gv}@;fr&>(nP8ze@`5J)MR5jjK>HG7Fc>s8QhZJg5{KV5|Nie7G?KpQs^vt`@j| zT|2q=b7Z+uzaqsgrVlXX`dB;=@E+{rUb7_=l?A`cNJu$(8obpn!%j&Nc3vNa9~$Lx zw}m+#8yH6I+z0IVuNE)Oe$B1x_ven70c(}sL-x-bDE&|$)m3J3?R%p+rLcH5dnHif z<74zWwS`UzZ-UvImO!6R0^H3kgux*j7?izbW7hm2)1vEC`njCC=10>;-3&_c6&Nf- z*1+bYV%XCu4?)Y0)4@ks^f@k#k`8?bwJA@at81SyzdfKOF^2ri{Du6PGYmEwTC;bN zc5IjTU2#l8y(l|QS~T&vHMRyWgazNTSVfUJZcukY-PSDR=VvosD+FYQYv9`#^PvCI zK$bluo2v|{p@Zc_c1jAgODY}q#0hy3`k2daU+khSHgGl`1h;?$bwD`Wh5(Q=J(6fT& z)LbL@fgB>>`tj-br9TH9TYb@AHv<2SQiVJ2`j92`d!|2jMcutoc=SLYp8IP+Zr;_H z_2CZM-+D}50~6`a$XFU#5sk%n7hnQKVARuh_`4_zbvIwcp){Vvg^TIq@MN+q7P#!b zW=N&8P`B9a{VYP-k zJ7hC~^UAK|^2+z4Y{5xv9utRSZWM#htW3~68U;s#oM@RZl1nH@iE^_k*VB;7V)bay zEumli%_92#gQIPa{mIRfgxy#WNqK2t{riDGR{5%np4xk z{Y4d61O&nf>8)UE6$!F`27u{*Q(#i^H1N$>k8w8cI6%b@r%jfoQ8kayqxc{GH+?$I zF5qc)p^#sG|A5plzM_96J#>C{D8)Gjk#n!mODNyXHdT*@@!!Hh)^jq14_^t&YHF}h z@cb+n_zs3AJaKJaIXrm$2Of0Gq2#O_GLXxsp~Ftn%Xb2g#xn~)$6dtqbB4Un${!@3 z)I)h%r(pB$4WQ(A38v@or_g2FNj4>nVvAE@Rqa_=8$1>!Z;2wCyW6O$(u?|(!pXQ~ z8$FDiL*qQ;D8KR(@+||&V`nq{yM3G1%x$7IFb2Qo*ML(|3+xhQQAK+MMOjABNB*p9-3+5u^F8TZ-*8OL?FB2_VQ8dh zgk^K|Fh{`vrGqqS>N9!T;QS9uXC47P-7|3Axg3hSTX1Nj&>t{zufVRIfGXyIxlNLI zyi|pJqvQl;*%0#oXbG=Q3i)T+4(l|gV{~9Q{1!ZG`#;*^QDq*Nz2R`|j8@1tIt~4` z_nT<3;k45xbDGhvNqJG~N%e7iW;Wp9cl}c7_{e zT5wl&KkNxmrg%?dYJX=+7BwR=%HR#ur+2H?l77x8n7i*OExqg{6j zwZ7Uzw>l0}$)qMYqa6+EZ*!rrFqZtSB1rE^3hf?wiozmt>BxXQO7{#P1JxWF8+U_V ztP&O<{v@QuC4k#n4sLH;PaX0`bi?Qq?lf1%eY2XN=le%+>rLi6%VK%A|3Z16==o&z z)`jF%=h2?JXQ;Bl7Y8;};*9Rc6n6bKsn#|UwJ*U*i8HZg@&weetzk=F&0(6Rn^{w& z86KX=p-*xY779IF!Jikr8b5sn^!Bz9dZw1*)hX{lKV>y!pQr<$pZnmf_Cgq}>x?P+LQm(zoftFM zl=KJd(usTRn7H!*-f7y4D+z zw71)c(t!*e{&otzfm^UK(h>7w6w$@!7ABhH;m`5AP~n>;rpKxaK5#|+nf_jk!7&qf9tde{Ha?b@1d~zh4y(fk%8nBbI zOnbr>gzaZamwC3t?*VLgPGzeF&b?z=YdIc!SSd}93 zT2XmT99u-Y!1Lk**coa}s=YsO)z3?~?WGHq3=uk%6i3mD=Lcc!xbaXzV))!Rm3-z| z(5OGdX?k@e>Hp_Ws*mic{fi0h-)Ka}v3D`hsF9jYi|AtDeo7irfx73jQ6+3AR=m+B z3vFpiHMxmegRXI3hj(+Q=X!I0OMP&4_Y~Y&VvfEVk+jm-nFhQ;Y6@M3{aOpL^=t}G zS}7d<27+c{eGDE8I^2`B_OQbJ6FXSmMhg8!0?#*{HrB>4ljBv~wxA(QW&BwBdTA7m z){>_j;oj;|N`;Z9=YpI49@=_n8m;agL6>DDaq9MNP!`{V?(hXTrPmPA^*vaHK0%k( z47}AIfU8!O!Jj_|;lE|;;a$NuczW|GA3d zS6!+RV9KhyEKw$dZFr+5u$bCWW>%|kK1@cl^S$5~y8)^jw}5egkSqTx%mb4Yc-#M; zQ<`}ZJ#<-*-xl@4itumHExt{)6OK}}!FGBp413`(K7&`tRj4{@0B%Y5*zMLW4BU?6 z{fmX@{QM$@>Y1Za?ND@j{}nP?Vxg>b3(Q?L1Cp*4qqSEC77X2pe#)8nRbU9-)f|Jq zB{94qzm%W;1Np7VCvZ_jI3B4xgi#XetZJAdyX9@b=KHClbHQ7988;6fkMqLqjcRyu zpB0jjGkURPE++i4#b~z!%)`Zv)vo@)eR!sdMq4Fu^4OucsC_xyK4$`z!am~9-z;=6 zEJHd*xOPG|F7i#q3)}PX=$gs2^A$%jiUWlHl}WJl+)Rjxc7P9Ch#pI5k;;1|D)fJW zNlib-Qi>y0Dlkv-HTE_Z;-;a=c-_YzHC7i<=&T^hJi3r( zS2XZe%l7cowYKuHS0ivjUL;1X72Zge5;!g93Vi*2M)24_!IWQ5aayy8-n4{arMoqz zwoJxnS2Aha(Y;jkd?(GTm*CIsDI@d#bh_eJNzr^94NS8pofM)u9YuItMV-bugkfX? zha-xdQQ2LczrOS>HMd-(mX23=O7bX{P7t!dYcJx1EvabI+J%d=Io$b724B{HgAWr^ z(WxRAMThs|JSP(z)~<+iHZoLfb;iZw5%@S+P=S~IgPB|+Y`=35%H~?J)()XVWXxp< z7V>Um4G*I3;9%5Pdw?!}6?z?mHj{1jISLDQri`TmBV4zfLi4VYrTI%*eEp@shl&z% zFFrWcWjQQzc*4AU+gOC_4Hz4H6?RXrhr2^k@c55toYj60#YH;!&_xeFUx~&vT{XV$ z%wLLq^qz+C3S=?q9p=_IVC05feD4x--e~qfe#_klc*Xbtir25ei=!m@^|>u14!%r( zt`5dCucqU7sVR8kMKiIkV-%)XOgRltDB9-=NzSUJ#mjck?QNc9Efqi&+G0K;Sc8vJ zR_CiCj^h=_1a#5L#0io@Zo8&m=rQP|LOp-F;G{w>HoNeyloA;W9Kiw34{?IC3SXl8 zP~a~Y)BN>!>8(o~HCoIdiPNXCe)Mv56W&#Oq~uU)-T>^LD}1kQ^U0)dFZm2#O}8(O z#KegMF;zASwNEdmE#GXZa5mDaZKrV1yBu7l`4^j{t7yg6gTi}pIdy9c#p;ijA>d9i z^vfHN@0V`$oFGGgU8Lz|!A-22P=oa=jWBYlHfDY>z?&|?$h{UcyeoQ`@$@fg-Dsfx zQETXEn>3q~X~1MIhq2J5bLiSGecJR_;ChSG(Z5KHBN_z;owq${Px7WIQ7>pcW@DgN z3eLNH5aVv^k|OHYhwk5*~~_ z%v|qXh9p&8gy zof93XaC0asc1NP-+*tJasb{mjLB__Y<~3ibwTo_L*^rx4FkRktfp*5ZlH$Ob^zgj_ zKezZj?KBY>sgK&RZ%hrIT2X?ZPF{vH#VSyk`;=MVzk#QI9>f(#yD-@%d8hQm8?$uri|+T&#V9SbJ3htIW719`@T+6hJfZ z+gr>3DLM~FEWbC5Tgl2QWrS!L5#ING&bjZSC`oA|X-Y^#duvcB8b(=_8OjWaibTW6 zYFG`0FPfyHG-&uezkk4dp7)${pX>U3xT&o(MH;cAM6Rtvq7~sG+^J%DZfHZl$hqm1 zMIO}fZeknG(<@JO^rZl_8j891QyoRq)OK>_5BvquflMgV&(C%TY6ZTtQ!JETIB~sy zyoFgWPMgV=pLGFO^cy3*pkLVbzuqd;?W?|d%kwB}L;Oi_IiyZ-5$g_pj!kc?93&ycl9;)Un zy04-n)XzB1&n5!6gHb<3vB|AmS%8h`(h5&uQ1uu(DbJJZ{IZmT=LX!|LuaD z-}?O`^WAnN>E~1uYVuw*@Zbv(Rym3Ga@PcHC(qMcE0%ImhL<>6^NL1{dnj`5Qsxd? z+kpCzE_s+)C(KrhrsRkWIep@Qh<);g#I3e`ChZ(2*exK{+8q{C?)RE^TwKn*n3FF0 zTQ@_P;Q5kBCMt`P-|GnNUp8|U|213OzveB7m7Yn8(j&PwkE{eAfBzxZ*LrhGra{8R zIors_Uo~92)*{j3K_4>NQD0=S|DQ!*P8spmvExQAZ52kgJ}2A>O>XwsC4#`Gfufe| zAi>sbapBzX4pO&mx+raol8}vh0y>MNiRISqBA>n$T-zxvi)*>P=2lvzq8SpGM9(zF zin`{8lKrhi!iM1Y=4zh`NHy;n-1pxDppgrP0T!D_w@Pnzd(M*9iLz^JoXGZ zHd|8EI2?p2#{0g}^F+7We7I?Gb46o6 zB?+od@a`PdR^oiCmwQnW#BD!%Uo z{9oGzovY_^a#J%zl}-Odvb%($9nUYC7Y3w~o91=&mSs8Y4T_^@;X;Efo!2xJVwPpA>S|WUV=iF0ecvj0p=QnwC@*57=(X(EvbSieIFenL z)MqQDr?VZCHn8(6mobzyMdjyesPdu=nwCqmJG^tCz+yaG(tnP+TS$Y{tr-wmFa;ME z@UB;l%W(XhB)DeGqY{dv$kTyWP?sGGSDqb!|D0z+e(^18v1KkTe%nWq{B`JEg<7)u zSPcAdh=IzP_dxf=4w4@;hL|`0q~(PI$i3(cCC*o%JEIUce67XMH_tI=?<~B+=a(HQ z&xDleq3T}q=%{)RYQRhPS|$#G(zgj%t$vJXDBl%Mnq)1UTr(Py8|10qJvsU>;2oql zb5$Nq9wm>Q228VXD;u@qH#*>O2Rg%o-YpZCXHnF2s1Y0`$T4(sl zbGjsPgmNsD)NcXhO+{o#w@CDC)hA(+>OA;XJ`TEeIDpF43<$iY3EM}GrK%fwj$yzA z=7XU&70dAC5X>O1T(U4|LUPH-0+I=Ob0ZCuep0oz(WkueD^rqQ7f z!OJ&O|8xzatS`@i&Bw7176NRT3ZE1GsNMN<)B>9z?^-0dzR!hVZzFb5aVFEB;K(YQ z*09JdU`xe~*!Dfs=-JFM^m-ko#s!I3{we?uUE72il2h2m_W#fTZ{VTkFcxfQ#4^&W z&~Z^LxZgNNmpVy$T?$c(mPmuvH-U6voweQo-mnHxvf$NP?D>~XSQ=b_P07R9oV7fU$IxvW1UmSz8J%ON+ZVDIF{iwzaGqP-< z5_j(II^4g|4xg==ijDGb;J#EkxM_@r6D95NaM1wVKKBsVt549F+yI5AOQ31NCj7UJ z=PbxRh5(~zbeFz_>PoTr?CO2kAG{o0*f}&d{s->bO8B!}2^D*N(eF+cT6EsT6gD0c zi%USvaW8c7e!OEJn_%YQ1XwtJIzD{18a4O0h?>_|Ajw^3@sqtdEs!#Zwi}OeuS;_k7uirT;cV-LOA^65R4w7 z&rar#XX4XmGC#f}sMUR$#^yYtAr{Uke)K46yo=)bbsveiSsyWtQX~tP_!A40C|dP4 zk^cCyP83`vPozHRaTkMblGpD|h{CA|E{W~~^Vtnxn^q2<=P%KDzj|rh#NlxH-W1%z z=Q7n_97QJ5W@S&OF^BxctgPlOCu+Z0a z)T?TU=7(!Q;LAK}=Ozt%pWUZTVdHq8<7#vbn2C-G>0mc@7`81*gZ0`X+_ZZOhN$1c zeZO1apHVruq=Z4srX`Gi_F(ZQk?gVF8ggt|AxW>)p<~r6c*cDP23t*H{>@WKj7mCz zU{|u!{}@fDM_|N?Vc0U#l#VtV2^qb!;3hwV-a6(2@<2vxwXi`|mt?f{EJm3}rnJdm zh!pS+_Jt76mU)LU#U4ABORH||Dv)y;V2=qa3Y=)9=W zFqC_{-j4X#&cUmmXK_>m&)Vvhx9*{tThY}s=G>-&)pJG@K5;Y%Jw zDSW39bt52pWiPcFUZfd4dkrt`tb74z znJhze!ne}GL_4B<+l}>q31K~V;@NZm_pL4`4xdZB!xqELwQ_>)e z{}hMj2KTAvh;Q_=qdh;zUP@X!HApP?f)mOl;jOpt(5dM^OpF^x?2hV@QG7;rqnHU! zE6~MP`orS82`Uq+`KRbB3?9&i4flDj+O}1&^~E<#S<{Ul56H3; zA3hU-;t}d{#~NyNjA5PlD)=IJ2>tIaqTAxvD7k4E6O(+&joeU8E)2UsD@!Rxopi&e zh37E(M-p14zQSCwk!&YEWNptsu(kaIEG@VXtSYvH&VzXnn4be_@3SC>pDTU(5QY!_ zn4@lE24t2eLierRkl($Y_sK@WO3fGunKu~|9Val8;5p1)u#AnY;GK**%b9w}Ay%BH z%3I9TEcq^Ri)4)tqZ_T)!R|B#sCc`bnxzZzz|L$K>#G50 zEcU|coq_bN#4#GsRLvY^o-xN|C2ZEFjd*L51OD(;!G7uY?3B|mOLW(;)Kqx|#`|@! z<=#|0dVdcaZWGDmGLzUqUotqChlAysRiJNjoC%zw*mu7}to?QvGp>qcqeY=i^0FSY z6;5Qg_YPx=j=r8tZd{*(bk(ga1*?Y>7l8 zzRPABP(}Z^{U6mDz?KY#vf+o$Fa^D3%<{oe=93rBCV8o$*BKks~+AOrv zn7yB9$j;r~OF;>#j_U?G&{GDMyqm&gsW=+>FUF=kAG~|l0Z;zoog`;#;B5RYNVxM3 z@;7{e*L!;5o1P86md(OolWP1=VLC||d6PR6eUCdc)0r`MS61X@#eR9JVc=(N-j85} zD<9jlZO4>Z%(Q&Wu&f95W4+Myp$-&dw&BN}XHh2N44z0ojaQe};hX;kP%$))RXGH* z)4@mCi!>S9UAvh2n1|6H^S)xvcR8jtVj_E-S;7?Or7^q9;k;Y&7JK_Nhjq>OW=oQf zVMnDu#_(&iL#zT_5v)(^c2$r^nO8~MPz`N;@QjXD=AD8+bFuYA3f88L#($~&J)X>C z(Cg8FF*i=onJ&xd;h_q!9D4}e#Lr{@kQ5HAk-}$UeV{mZqGe;ZuH~;>HOutO#mxQY zHuf=eArss>gde_Dp?CIJ7H?R_&Rp+jo06m~?d0a5$rj#YqUVAg+1d2;#ag;{b~o+Z zuoERF+amij8zYu!qr$DpsFI_CPaBtET=G`TAN1#cu1%Qlu@XJg8NPM6g4$W<@bX(9 zG#*!nIbQv^acDTZFtiDyw%x@~U7yj3r;I$S(r07uNckIpFrXqh0z9<=DOIo`W{yk$Kc4#SJ8GL8*A_9W4Y7~+;v8h<#))lSG=dfX3Q8%ql3zpYYogSo7$S; zt(QLLO&gDAE-KT;_8=P8@_~MSPI1hSJN$a~0~ZflLrbdO&_#vcXz$V-jP*K=R{7^J zFxrrLd^TfiTiw~Tr)#l7ITc3ebimlZmU#O8Y3%2jO(**35N;X8Q6o(Au>;}AwQut2yBXM0}7Y@SEF`-V8!S2)0iwL!4lvjQZKOhNOY;c(`$ z1ssXu`8Hb;>8S11wDQ#qo|Cd3u81$;!8Lav-03m0p3TV2R^Z{#YgpnEg0m`5;(q0$bXE6pD*r+X z3_u?OZl}SFF;C%5q8w@|ox}7?x6yr~J6IJZ!N;uUkdt44xB9;0a4U|5{+NurEM}oi zzONPJ$6Z!jiQ@p%D15Ia+ z!}c-9vFC>vcJqcVbK_LhOiRY<+FR(GB+mY;387UTne>%J72T(pifm;BUUE&sg;Pi4 zUF|8jPiqlAQ$0g^Z`)BB&20L_(h$5Pl|kbN&wE+=6vtdzf+LIeq5IVX+zH^go0B>@ zoq%YZ7tx9j<|?w?SB${ee;T-a+5&A0UgDu1SN13)oQ<7U1kSbjz^XsPvtxQ_e~r&v zOU%IU!K<0mz&iG!DVS+h4&c9UbJ^yIgKW`^yF}aOA@|<7mn^$?7Y}4d;B(Dv)H*kT zEw=Y!zT+>je19`G0W#S3MW0xqgE5XSnuD?zkD_h$33`2PJr%c+fW943kQ)?5A360? z^CzuTIDZ`&i0Z+YX9CBn-G<_$lkmfAZP?T04Gi2sPiHuD$en@%X71QfVFZCD>2TS% z7s|e+uqn?<*g((#`#BJVvrOvoYs4rq-V#4;8^m6oYv=eS7Kdsqp!e1ZuoNhv zpn)H)tm}h{X>)MRwxy_A7J;L~o?*?H{bbD%4f^d_JPo|9hf>p*px^R1^ghftva8F8 zRKXFjja*MZo@}MAk5*udR0xf;dk!;=lThgJR^gD0+API;;*Fsw%O2-1OCt;GCO~0ni}nXW^s0C3Ne4b0m(#S*W0yuvdPN z@NYZ*IkJ}fw@VxBT-W2Z3QZd3;|ukNocX&-LhojWQZbLwxHkA6S*BG(WL2W5%==vO zTuqC2BCW*FD|y$nY&=+fUVt?|mgHh4zt^1HhF_oFB^R4AL8`FGF`y^Sl&a<8#O@0Q5P!XR!*iI7y*XHBe=g+MV$Z8D;mk(aWgIy!N&GN9QSG? z*}C&9-PuzOU$1YbuEMtvrk;;7&jsAn%?)%`oHgRhXeyI23%NsDEb@mei4eDegIZ>& zyL2KsxU-4KZ@)~ZC?tu-8A=jU-gmlj8d4~eJ z;dC8}tte%ZyN+|dbGOiynd9I}x+&L*Wh6vwfVTR2a7&ylNQj9VoXqc~vje;_X6gba z?Xr^|IW>n~EtR0{(R*R9@E-O3dJ*)uM^L#D>+#fPYZiaShYItaW9$58rs$tQPh6e` zw7CwNv&%)PQ+&vw>M67#UXlB>a)PjW`VCRS=#^w-u^&IzyM%7CM>wP7>hP}B7T0cD zNc>t)(x=BK!Qr{XX?tQOcjfOXg06gCBz-Q-KR*^9$c+FMRZonyCamF58F#7U98@UW z#-;bxk&cgNsTH4#>iV;X<~%B=Ue+GiFv*V#Ox{6aMB|_c?~0}!SU|X~$LZawo1Duu zUl`tb4BNgYa);Kvpj)QupjUN*NIz>5H@hI2uI5a*rm;8a=SlooHEcQA;h6x5!CCx_ zSPcG+ok~N+|Iu-`kC5e4lsSXGBJRC)87Y?|biC(om>v_cz_)%0|55LLC#of(0P~*tF|Y?y`h8jN5>UH<>lFF?Ppox*i^yRYYwDNl{y?aOt)_!(|i`G6+{PPH`@5!an-calsJd)1vv-uUmv7I-#T z4yRsM!qZ!Ubivj!^iSz*>bhSEysS6#ZiO?D5G+CuodtMr`T}ge5kU0~VU_l^U+so(27e>IG zL-%Q{nI!0?y2HD|Z(Qqw0us3DEpd`9B6E%)m3*2;M>-}z+~)W2*=`ETFS!e!`5trF zyA)_xafH4%oJpI9w$XcQm(gjSC#d^|WV%nJh*3d&-}~GQbh+k8muw%PF5-@~%Yb(@ ze<*`>lO*uqdQ({Qgl8P3M*&8C_)9;uJ)kx#F4CmAd348*E7Ux03TV&R4Pvh|;8#Zyr_Kq< zv{mysi?Hor_vZ*Cc`t{%?_Iz%JP|H;T>*NIcPpw6g1_u{KD*1WOIuGp2cM0@`d`0D#V!rH(DgU@zVo|KXH_NH zRpUo9hULND_DVRgeh55uDyiG2C9o+r9R7RILV9w&>CFd>9-R_FJF^X7gV6zy#`pAH z|7;koxetDyiN~V8CVV64!Tq})!PrliAv8V)%$MCHho$|lhG>snpk`^{T zg5uqKX&&!+?EWmyEN0C_#iK_k7&*a5rC@ko|DBttT~6y>x{(>WSA@q-Z-ob8UF1{1 z3!>+>nx+lJk^}C2^oZLC{NZ^JEK@c4+(!srI2{jb{>YJGUoWFn{cH@`(;`yf_fwxA zUL?ydb#VVyD#D!6N<{e2RN%4l5H9y{#p9Y+ID=LHc%O|ns<{3~%_}mgt8zSfxKM!$ z&(?tv8PRm#HyJX^I*FbPQX~&<9_5CWd?v!@s^n8~uZ7mq3Yy^WNp!|N68+^J$p`nS zQ%Cn=&VKY5G(R>R1KLVBtABw|Gii8+sKiK2N`6l+L^u(@3z`_D6-WJV9p?^t zY=M`qZ$$1@(?u8mnqkAB9}S&wmdkQk1~cO>Pzl?0Bzf*5{QYP-o-wJVZ(K_7=1W(c zE)gZt;j>!SI^K9VHIt;~RpUy#v1py(%w=1>Mf>gk*lBr(NZ5abZ3)Tr()I1!9Tgdv zyd;@4x!RMXV!C+i{w#RUSrGZd{q#s-6z9E>zsG(P4N^UmsrTuNBIPBc;E(HN@_@E+ z`Xj=(%z@% zh~3f(vN1mjRz-%<56UY!%}7(mZVqRv^Mgob)m6GJ#GP)nJR%xZ!gsa$4^W%wrCh(p z4yd!9ORbHyIdh)-G8Cjl(yV3Z!AN^}y!tTB+!)E(nvKWU@Ki`~a1h-)Z;n?t7Qu&Q zn?%}4JioklkZdpaq%#dqLZh}V%>OMx1?xY+zgTg&SnLbemzm8N1djU6K%jPuV8Bx!% z`gJtwM+lj)g*fi`i}B#v0AE=()->bYhH<PomLsXbU#l-lEQ+%3dYrqJ5_`(LFpLo4zK&vmiBM>+uxN^ci68 z-`(7b_2TGKW=q~qi6?1|kHKI1G}V*}=8Ao7LGN28sp@pMNE~d%`8J*S{#Pf4BYh6&5>yL5exNe0l`tPM3x0=kL+82czjQS9==0^%V8pa2hfl zRzlLe`A}WGhrB$UM^f5O(}^+~bg_*poz#_2#uuLe$F8Md`C$??sXQc-)iqSn^B(P$ zvA|MwYjjx8&|q#ibbV_CxxIDp#$h{(N$khxuLIGpFAMfG9D~2d|3SA~F|OQ}j`BP6 z(Q3*Ds;is9J+|r*J-cy(W}VieIuZ8d!|i{tK)H=NxLzPWd1qnd31eE6wwQjc4P;v# zLfM1CBP=i>9bR0sfa9~*16L$x`R2h0%k_o-So#=o8lLu?wm!80ZM_xD*vyOVj9bZ6 z&jS2PRf3Ib*(Ay) zr|!(urWA$^hQo)9I7so4Wr1Sqti5n7v$f2E{5SGg@4@iSg%(Bej zgSFiYvU2iA^2$FQJtpv74{=ZYr9Bcwrc&^Ea}&LDu>o8sJp;{-$>=q=fpGr%G$(5z zopbdP9b0Mu%__lgakv<`g#^=@V}6$d(^5%6t-3-zZKF!7)V=ql+$+SdsDVxEuf z(Zwi?Q6y!t-egaIC#M&r!+hhyJbJK|LVpT++z$6m61d8l63t2$uvYo zl9@~pV`Yt>@zPQ$p0C5_LDzMluVDmh*saHcluGe+gcn>@IRIOh-UFGi9^%zwOdThC z(^G{3q~A>(u7!tCOSxzq#b*WYJez=V{M>H~f42}Z?>p7`zJj(t=Xx3 zg`Pw&Sk>mwGr2;bU`8ag-dGOe0~f&U_+7O3jKVwF)A4q~1zd5s5NjOj(Uo_gUT|o} zk4d$t9cO{BqCIgKJBuYt0`c^x3;5sMWGru0VY&xj;!InfpCmT|Q`cDG${=H`IU|ie zcY9%&_Z?`Mc}95UU=G=H?FM-;>WApz_BUjKDpG@yZv;WDYlI)_Eoi%$x!|j_m?(DF zBJ$()HKFA7Boe1o%rnC+gy?dPcq~`q{v~RPhCV;#B6VtrwQ7h^cyKlM+I}Wy?Vl#t z;~_&j{>5=JNn1sezBrM>=i|Bh8$IS{B6f-l*1Qzv1b!E`@qA0pzlG}>9WVTue2J6X z%xG+8k%j#3Y;JqDn?~W_?QaA!ev|%|xx$eOmQ+P~5uJ3=OEAYJQ?R>GOl0L~AapFO zAOl+4NP4oWXkV2V_o~g3T$afdMCa21xM{TCa+ogb0 zdiL7l>mPaU=MPU#;P+CHFsn>xnpwh)YRoX7c3+FbxR0XI5sL&xujIG|yR1a{r8$Bl z2Tzl>>*BPz_^Z&POGs?a{UO#{u3Bt-)+C54up#@8%n=Tq925?-`Al+uZ4{g+Iw#!o z3Ak_nCJ6V7`3fhFd_i{M4~x!*C*1tlQ{4RJ$AyB^7WAQ$7>z6n6?nV)a>uQz$>L2L zh0?EO$^5&<6oeV8P_*&3zv#=oLBZS`0ir)ABZ=NgThZXZ zO>|G67WLn)EwI`q#wjhh%&k!VVqqzW5+>>Oapw~ci>R^`X#rbu>&AAW;&=%nZDmJ3 zmpTexDGNyE`Z=_xM@!nzcgymRaYB~#P7rXrof|pFR-!2&Go> z9?X%|WQ58&bHlQeqCe5KB44v{Lf?8L&Sc1o_!!9xpN-a_7straHw#7xgd2VeYGP8j z>9j|Xd-$Vhm*;RI_bgI)L`7TpNqi=i_qWItBdiH$h~x;#)vP(ki z;yhxfcTqUXbT|F^*^&CDCJSn>l#;pb4J5htoS9Qjgm9jowBX$uMMmaYF+<(ud}fDQ zKI)ogY1cB(^08VE&An_-3I$&D!gURLN70erYfmQIUT9+VuH&G$p$Y;W&(hN4rCjsu zzuZvFT9g)?$L}pesNqw`hE_k|KZmcg^%JJBNkb!fH~9nHHPpe*o_xuQ2cEK*Nz-tC zR1M_wJ}H;MZ}@Yq7W>&~#74B;g^jE3!=s1!u(jko+O1xLXMO}>K-_s^n8x!S=Bsfc zH#=C@W=W?i?V$V2(r907rx4zs5WQa4L+1O>AQcCelRuAsKx^?wm@>Bv{HAtM>9B23 z8NMFM=Q94Bx&{TBarkS4AKP=rgNe7VVbP(x>C*F$xfN=f#CX>|2-5V2_nyCC^ha;@ zdh-c3_e40;9J7#~s5Pd05B;P^od2T&lS=OX-LG8RqCwc(whg|B)xfalry)m95`3mc z!Mq(?S*pT3_W6}5OL-ni`-ZM?Ab+05+{$2y+jE%Jr8`W_B8KKZ_NVo=yJ)+9G*i47 z!@BxIn9apz`q5aHlTCZab#MKRQ&)sw;3{`K&TgRGR1MVGoC$`a3JmX=iN}jgF$gW$ zZRb}wHu(XXzwm>B@*?V|xSty4U1NoPl}v99&xrXrjQ!m*o@sf^V1Z-O=)3XH>0@hs z*z2UijupSgNnevOMyi0NXG?4*ry++j7^#n#B|0^Q&h5-EIMmtS9ifc^@<% z4!}g!5=4nxsJ+sejGx=V1*&FnKg&7x{Fw?{lPk+c4cXCa#qLD7Ifbj-DbDOio=5e^ z;_SquOq$kqm#dXXr1=ZtsBX{|nx=47!=K@`kp6f04l zoKBAaca9s=X~;SrO=0`GXS3an`WX01KsRTJgTtga*79u?OG)x$I+JG5OtUFuQ}0ev zTC+vuy0T7mhR@4u%xa~Pk42zrB@f4*d<5+?BfwQ#3>I8}Neg~!lAVc{i1xWBq}%Wo z3EkmGxPL`h{I3qbe!qraCi8RCH*N&}`yd)U-IKZ9U&1)Yg>15;J+p}#&L*r=WRHh` zXN9ZeEKen>TZ-HIGb@|ZY-?u(J1rXy6IWGyS9ql_Iks83I8E%sr!ICRBxlr zlV-GjGZ((c#n1)q4@jTYN4PZNChX`g2CI?lKw(fE+UJDQ`doV)wXF`UmfeS(q#5Ys zuYx;tJiw}JG8?*U&#tXn$7b$4hRv&8anl_`3~x=vHKnJ~gx?2mYpvpQ7_~S_(S+DDycd0{0gUh- z1qM^Ie;2+qKLxL>INZbsY( zou)J>RDOWTU(TU(J)gDEyoh&<=V8M88Mw+@5?m(Vr&2Z-sk5#neiR=7*~CtO70I|i zJp`xy*?{7ve$;u3JN;8)Mhy(J!1rnt=w)04M@?Ihe7p>NUXOyk_T^Nnow9cWeHyMvTZ;oA&byEvj`yYSHLvhJ>t!%oYcGuaQ3}O|bx=E0 z405Bx(e}btoLRFT6|=^(__>nt^$vGFUwxomiihOMTnSxsTqp|4NTfEBywyyKt{ek3oI`bz# zQ(KWoha@dQyYwktd@7V`)MR1m;{?p93&&>*FVTG^i>RU~oIaWCh0W_`qxWCLP@RpC zU?Ky0rdlv_xG#3MEyqFrzI&~s8q|9l!pYu=;L{RDJ%W6wNB?Fj^*#ZXSe%0&3E5EE z>H$}GECm}w6VQpD3emEXJOfV|-cG%O^S5q7a|s))49g?BoIR<1JBBP9oQ-|C+Ndr) z0dptGja~7uI&LhKm^aaN(Z6Wg^%gqyp)5|(Xott%^$-wdNfkEg zQ{UKcq-$mxetZ#)@_SFC&CD8b_DliEvk7qdL=#HCdWst}9^$vc@A!PE1+UC{h*#IH zz?yX`cul(QivWTvJ;1BoLCDFjcbLhZXQ|uvt*}o@Y z)WIE~yVeox+_a(p`zQzweM6I0m(l=*MpQiJkE;$Fp~VG12yWyZ5K1-Q z)%|dGZv%w*B|=umYwS%fMy`wZpoW~pj58GD#Q%WeBWM2qVUIGw_9z`}fTKPS!}XIs zLzQ(PZu-3kcWyq6(T1(C^7m(0AHe6|{u@IBo(O2-gQ;{z;}~|XScPrati;4X!}9h@ zY0Eq<3CrxF)_kBwC1DOAM+Edg7If9vE=a9O9B2 zsqgDB`YzHN_tlB;*9>*sasCFlpB)Z?J&|;W;xAO5(uB0845bUhX*}PLP`SI3%CP`E z(7OZ+tlhAtM-6?HAA`$@5|Gl_P465XMg8-D-p&i*Ig*=U@xO&&yrULVw&tMv{xqze zWXL7o_`!*6fS`OTzn-Vyr&s>i&{GPx6eD3=$5b#pkdE_wH{j*1w_&KUo6j=bhlvxj zAT?AL?{yWxW$Pk1J{DQVBms+!6S8Pu3-lm3KS9dDTDaTJGfE8u=$R&QI`_G&7NtVFMNa&~9lp zBr!KA{g4dfuAAfahlbeIWP*}zuMslqaY1kiPPT5NVf$67^T13x%SjE??u5|qR^njz zdpWQ@d(dLMCX;h*!S0`SEK;tD9X%$2-jz!Dw?Gxk|8~QvyD`u>v=UT>kH@L*<^1ei8XEa| zS!_ovOPn_aFURpYBjI?QF!?@y{uhW*r*5HZF3%*~ehO^fr@$HOb~sb-hZUs(Xr?HJ zdbiwR&AAx*?U5XAoc07%zyvo%Qyk87t;R)`(>ouoz;N?8;02|irGJ;67bJs!(Q!C^ zJcXO(Gm_eQ`BO0~1vuqa0yQN?u;7^!w%_o_DOzzjp0i``6+YsPC^`J4FaewMc@MCS zBc6L10OO7*V(Qoj2t(aO;M_n1&hCTN`3t}*O$2+2vS3HX1MsQWLw7zKK4q>L%((Fo zOlAvFgAum(S?v2Q~=Tg9gyk}hTVKL_dh zfplPp2J@5Y=jX6;%uXT;%1>Sc?)@!15#k5)?PtQg|DxgIQ$1jcS8=5#U*udjj=gyH zA09Gah#9ZP!k>Ln(8=@TCYw}X{Bl`_>#I=XuL&O9bOhTM9LHf+YV1tj8#L$L=p%og zf}Uq;AY(10DEU^XZ6sOR{F>VrdyrN?IfmWZuW;{AUAD<#2Ch)~02-nex>2SaM<>Ox z{*)B9i_eYW>BXR_ag+c4MW{)1FmG)!U;H=eZy5 z&rJcFd{vM=-B0|xj5%4IX|&NMi(Hz!mERb?!w*G0__09;Q)O)Ny=^Zfq$I$Wdp>X_ zY7*^z$bTO24wzF3E})>Vh)q|V`8x+&%=;p1c`&=3&F@HH(fM03(*6mb2{LExl%wnR zxuazqe~06&gYQ4tGXJ|RtVf`MM#&vu%xB(%_7sCk*%&yt=^|G%M*t~7CHQ1xJr<4% z#3f2&@xw7AWb@+zsxH8J1$|`F@@T(j5$dM4z?CDL@n=deI${$x1+BxY`z29HR)~*k z)#y)!9GX=j#IyWbRG;z{LJDM1#qJwCAK3{m!5&yi^T0b*k$uzI$KHMKhgH`#Fz}BT z-sC+AZyU@o=!Yd*ZJWoQYMo}*W5QWah>vC6SU<~|%hy_t?yRBL=B}j+>PA9kf`aJ5 z(Mdv<;6+5K6)CB#WtR?)IIAee7d3xVMS^z_q2d{j+)Q5Ra~d*PijEd ze~Q2!@5K6|`>-Tc0Tu@M@?C0cZo~YEGzOJn@_SwSfIrJ;mOUrJ7D;l<=RCQ%Y(Hsq z4I?UppUBgz{EYq1M_4^R3lv&Zp<3@OEMHGxY4?A8-=QBS&J)AWr`(xgy(T;Gy$bh- zOhEriJ_CBv9xFzt)7YoB;3PgA@x3&=Ic@~|HEtBkX&NG8mFG$M2?erh@nvdoc_vNM zR-9@x>#y8k=V|^qub(8$Lk;lS&6~3G%1*RXeabcPeD#F zfxSI@h=pg4qd^mL>G2&QblmS+{8+sYD>xne<}6LU>nZnA*PX0)$-sSs<9WVTE4*F5 z6UTLpL|xupjt$iq+qVVvZyKUw=OR*4UML+~_xHS{;e}~;{e_-A49d3@ZBp-a&5u>L0l=S7{3%~E^b*>*%UDP4cQ5=kG=0aGx zGR-`+l*FIBN`~x9=(?H>G^XV#VSg`Uvqmnil1fBfiF_f{h;vK(+c=$tZCL&*jAv`l z#s1@tqWR;FaTC-9^nFe}u9^1>GRv02`C?aqtQ|nA_?*s|Vo1DJ3=etE?*?rpOaCZE zOC$65Z06DltgCDoyME>;cBhZxR>be)a%Vr}43+obSfxCC^tcJH%A7(IM^}82yZ}vm zHnO97-Ml;eG%`&Ati9|6d-Zvq@fBY>>_H$k2@9p!8&qJ-8$-As8%dQ4u2UVqb0lrD zFKt|Pl!i;`km23IWRbZdD6Eu%Ip^hpSS6rOT0TAt3dPgv^J&NbC_3*)D&IGb6UoZR zCX|_!b$*=$N>OMhX;9jq(o`Bk_MRa-R76xlB%`u2Qbs~)k+zhU_&(ns`~jTD zbME`P-tX6ISeZF$oWh(^h@-zf{AqshF8ZKcgi2UlKz+s*d(9ux$-UR9-G(IkOeK;y zUziPH?piojudl34XTh`K7oUT%bQVHu~T$ zwO;tyZV4hnLCnV{S>idu_U1Ak!7N!_M1B!|wUGe+pb0cCS_J%MFUYw{9;`l=0jpA4 zq3OdYNJh!x6)#Itt$diYD_tUs-2I5gQDoWI5;3Sb7l7@_FMeuw0IcN(OflMju+4RjPm)w=% z>iQtq{PPpu`zJ~JZZq`micz9@UlGKsRiWg23o6?SQgEA0Q+w{=wZRq?z4sYQR*K-y zMe2xO7UPL&N_6Zj+mpX-PE)lD;rs4B7+EEZ!t0J9$21!|otm+JcLQ1CXaXx6fbeuL6}x zNuU@bjlbnJa7o^7RDbt@$ZS=H-LpKvOimD!vsBPT(-beROoF9M2b7)b0iP5Z+}do* zGRq@TYoed)Q2Bz_k$s!t_}ySyp3f)q!gU})cReXe7$uiHT;co~VYpDY793v2LTfq$ z{^M?7w>t!+DprB*lvGeq9)s?vP~I)cAfDBect(ADFlHQ2#1O|4{Gb<2bngj+gp?ob zf4&eOZHdJ?UM2p@^<+4Ee=zsF4v}whnh;$Z04`ylV0COgspT6(m~$@(?94?AwhQTT zP?FZ4&u79^r;wQF5>itc0H5caf%5&$@cv~UzR7!q*XPL4BEws#%zEGt`l?fEGC_t* z4uNO+7ieD{!h7N(!yKzGW(+5m^Mq3Li1e@H#6GqTRwoPLlfC9F3u!I<4UU1D&Hm7) z=S{AtIg=S{B?xLwg5YX@n4LEW&pBq$xal;kpB@fR9BrxK7kBz5IEqR>x`USb1Gs8` z2a0&hq2%4Y2$AJ%Z`1+#EjD;w=@c%zw+*KJyapFPe}wAKFL)mm?=nIcV#)7|A@IAq z0m`Lh&{2LL@ahvm=x7a0EMxPsH|iLoI+aUz4v{%sOF~6_Dsu zJ)-njnBymO$;`b0xX0sJ&V_h35qEAQe~#5~9cTXFnkRf_O0q@h{rgoo)#VNvD($D` zvaVD!*MSPLjwFeolc=S(2Lt+7(ms*p^mM5$b(+;i4n(IBGS)*5>(EdN8C zBm0wiXXlsmiY(N~?;>&L(8e1Ox-kkY6U!jN{4%Hw$H0586UvsB)A5qfekm(^Ncr_(%Tp!Y;Tt#l}G-1~7-b8Zba#;rVMO?Qo2E81D@U4Lm`KdaG z*?ya0W}NP!sj3+?Rl|$!INL(ZpI4C1NqfmnzAetzJAwr-WAJC|GT!4e)3~$r=ktQn z{sXBPfQIEOA)xaDc@W`6oIiSzv}{M1xqbyS7MKI)-8wXFVsKib91gMGmBQ1L7}p4C zk|T4GmeqyOXX74pfyYv+bk2m%NfW2h+nZrDI}2Qxi!j$_8cG$*U`Vte-t0d{b!uIy zn9CBH^6eMFPx+vbJ_yR<)5uqT7b*1_Au-LZOhWAz5;IrQ*W5cX`L3 z7x1bMOl6)`%*9YoIh^cdf@YClfB1C-rmdz>IjhCdmr(l=5FDx8uH`Ld2^m5jqO3yaszhOaJ7q_=;iDMRC2Z))mOK{{t_2dFW!XRW!3aURX+QSUZMJfAJP5A zBh;5>83tRPkwJs4#AQk*QIJXCYdE4%X_PBGetp?RAyX4MzoI>v>u=-tRW zA9D@5_>bk~ShyWJC@;&>}@QPT{QMR@k_Yk42Wp@#$}0d`IJOc(o5)EjbB} z6&9epw2Wn$Tm~cUQjlt04R151!nweg#4kvnw0r9@56EUlW7%{t9g~9K2SPBRYe8B& zL`dl05XMKz4j#}AlCW6PS7h;)v#xi2tF)( z1zs^Dz}ma;>snpx8qvm%O;w~vB!^rm+efsIi{Q#5)v&DiJPh`&#`jxv(FCXAvx5@o zv*0Z>jTXU~?8i9Qvb_a9BRRN6y`e65}O4u^f4CWnFW%pkuP_0@_ z)}9Y#y#noI#P=i$ely1<6-)8nyCkMB*qxX8>kThe!iNOSS0j;|M9Gh6ahziP6|^7M z1GF}Ph2MNgGj)L>+k@cJU<9`_#X+rhH_~s`Y_HJ>?_b;t*QMn^kIndB{m$~#&3!R7 z!5fvv!(ji$Iymk;3;FA>q1XQFYCNYqI@ zJ$)m-RNR6+^PhmKNgzPCFFcM_!1q2PXtL-Fh}NrOQ=%lUm+J>lp$ky$9tH!8RYA;$ zLesbg{0N^Qo-E_yw_P|W2}OZ(2+OTGc^rlF4`I5B9aU;_r?x?d=;e!5kmb1w7K@q? zea9wpZ$%?9Uh;sLhs2=%dIQ`&O$f?A#S=Ez2+qB=gu8?N-HyK_V_);&)hubWkBCIC zdslFW@h(hntiX`?0(`94X4Bi{GF|Awz z7rjgb0b_OSD%Qp9MLAgL--0>U#!(}GF>Ze4jLVfXFzm`4IK6{)hoq129tWvH)r)-c z>dQx7sJ;_wocaZNj2bzmG6@gHR>KTqDQMWyhU@u z_2HhQ_p!y?0VSH_QTA&nX21Im7VU=g#070?r+NebN|(WpYvXWr>?!NI4M)zRzih|+ z3)Y(E5J~mZ@J6i$JkQoMFV`D_CGY~9$(l8^x*=v?UqN7ju#5)gW`tM-* z0Y~A0>~`36Kp(!8G?A@0pA!G~lfdbM4~#rNh##k)#Vf}Kne8=0Op1;duQK`$HgMC~ zeWeg`Jaj>;B^0Ei4?*9anN&|0tM&sZoVtVJ6F2Z-Mm`3a3DYDgaq56qQ6G*_ z6HPICJ46p-S6+s6%lWuhii@R(FY}Uik1%Qa)5xS#*Et^=U#}ffsDA)^5KL6#m{upUg?1 zC~=Nn-d%tq%9qf%T@533WN>z`Cgy}~L={UBs`W*Ure=yz6JrWI{Z_l!9L*LKjynZXLt`*#B7nKKref#ReP||U ziQNs?(Wle~^Ly_A|Jr$I4XFoFsb{csRDwhwWu13M56F*>X6j#2N(+Q<(5tti`3m<# z`G3}U^DTCMAwMqrkrEL>=$n+qn=4q%Q`elu=)}*#$K*G;9QTbJu&QEixHyxeF;Z|+ zY!HvOe?_^WQ8e?pOb=J5(`o<3&>x?2@#P6&dTW<36<^&?o!4BUcK!aeOHKfd7Y%?) z<_IiUo=JrTuh2h#<7oc34pVUlO%QyFImwHo&`4m-v<-xX?&2$(S#H7 z8&GqF4Vtj5Uh%d$^vb{K^q05-O-XcRZr}e*>{M?&Z9X7R z)%7&MaKM-x9Y|)+KZ3NkUVv)oJi$QLZGEGq8&Ai-#E(r4*isXKwML%!$@d)HGww+D z7cZx;ujde1kKg3!Vo^9`@tnGu_EDGoR=U=91zno6l$JE`snFbGB<$!}$nClWzH^nS z;rkhsZjh%Xyesrg-%Z+kCYMTv^YNbAMwIfhM4w}6;A zI(y;7R3uJTo@8;;A1Gx1ANH+dJ!UyEV6(WINnYSf2I5}hJl$4YlHGxoCr@!Y)*RrK zJlA9#Ec&n;62I03TRxJfTCU1r( zvd>@=dBJ9G#-LMkEs4-O0<(*D!R%)(bhU2-_4lr#o0q51rY*>&}T}CEHmjc7KY~I$l+N`S!EDOxU&PC7S84B zbv3b`x(sIetQ02hjs%=O^onSRECQp~Gl=ER-Q+#%p(y#Yx~Ql7V;aZIrM^)T^ygba8tgWO@B8XC zH9Gx_PQMaKWtMKI)h?^(xU~>kMX~4m2O4f+{Sa>50A4g~5gJmgbIC=}}ex-tvc^e!d0idlP{Q8-12Uu^S~eoyYC1vN(elmFh)h3Y}H zyKR)7{B@J&c3aVcn8(!hloS2WY6&%atcBZz3gK>PD%hA<@?~aE=8qR9(N(H?^s${8 z<=-)+OzSOF>C(dN3`6|euRz1C-eXQ*7e+K+MH^u`)MZ^hWK|F>IO7jzO3ty|wMO*b z6oHc$M&bIabMWq>D`0$-4_mIzKq^xP#bgW??o^?p=Y;6bMhV*CRzues?WHzrkt#=P z@--sF`MtbnRC9V6HJ$u~Mg~vf^JQD<$@T(j65B{kC7r;ua~)_d7R0sZ^e|Ib1U@Ui zBF-8fSY|T?rS}k2U0i|n^C~brsR{4C*WvFo?q&Og=V+Xo1FhOCPwUNRP^0Y2{Fy>K z`ERHm-)94t%3N2V$$wgK+Xmn#X}_f*7)nhX+M&vr1K*wsvfjje+LB;FeHwn^F~1JF zzr&xlxz3?>XIQ3N%L00Miy8f^smB+nZKm8k@pRjT>6D62ra$=o*spnA0FFJ zFBOc^t&)lKl${ZU!xr>o@)CM(oJZwvD)HZc>7g%CZ{; zk#DgPp9NT;?(tu6=+Jz;C8o%49vr22)$h^zZ?|D>sT=EzN(ZguyXfTVne=SObKE(? zaxynp-r!Y7*PylL;qj1(nr& zfsX{TFfDWzHO~BvY9tqP6OvJhvjlwaC*ub7ub9UzK|RI;5m?6CtE04Smn@ZiF+u08 zE};%OXQ=>*r^d_&4Bn+lxdEd%ZAgsDnTgYxMNu@+_AHI5yh4l4Tku|+l0!OshHp^pEx;CV~7lNh;_uV(m(j%DT1z;k=~>!~vT;P)Z=zUU2>4923(Np=o= zD;;@H0MAJ)RjuMF|^l`zP#@n(04X`o}10mou4 zLtf@%@a-QUa^W*UHfAoQIR}FY%L~|if)Dc#&moE;2Z?CLQ?g)eoR^-g&jh{cV76Dy zA;~d|Nbx~M@I7}3mux$a_ZQ?N@A?DG@hQWr8I`DYdH^(IvcPK#>xpuffUZJCs8+6q zy0CYQ`)NC}H@2MgHg^&?);;9s{g5>Ko(K1#XqcQ{0NRGK@Iq!9RC(FLxZxXI_fVWx zX=u`EiMlvjOAi+;wZK#ZduXl|#Gsv7NO#`=>t$sSf8iGV{gaPvf1csz7m`%ZeK%Zh z@`H;Zzu{m_KCZh}kH0QW;F|Fa9BpmF?We|Z;P76Q97)8s-eSD#uSQQ#UqHPZkJBZC zop1}qP}y1-Pd#LLaUZ=*_KQJ#_AcV>W5@_L-(_B^1!BS6ySR9? zA7gEUVNL%_u&thr(LWZ#g?n*ew6qZ3w4Q@w{k0I@{Rft_J=0>Xd_2&586Q0qfu;JN z$&?%=C~qZrGGG_()%C*H_gG$VLLR8|@4?cyap?9umt`Lpp(7hAxFBq}W`aXl7bJBr z;bnCnX1e@$^8OwYMCm)OsMTpIJG*x74<=?h8da$KA9U{+R)QWmsq4SpvJ9NRT{BZP{-w8#53|5jF zd?L3NTvjJQ%ZVm9Yvcf{V~nBt_hN|uVF6W~WN5pjkDps(K{dM$HgdZ_W?(WY{T4#v zr)`F!ib3#D=z^`!e}dR_iahoiN*P;&dy8Iy(Xm1JF<%@*v}>7z>f_9h1-j6FVkJ&^ zd*S9oQFv^AGfcZGh#tQIH#MJutVQB@Y&0K_Ys#R-E)x_pvO p>x|3au?1@$4-f zIHxU!LILb9V#_VKR2GT4H5ce-D-W{%QyY-$A^6xRo9nRNnI{zOMmE(qlWVis998db zYWpYKn=4}JUw=ktElV-hVEflp0R|C zWEU_O8Y5v*tO7IOq)fyE+4ot&2D8Cm6--{hWpZlaF2?yt10$&Qm{?kU;`QA%B~Jyo zkSWeEAC&Wm%9dYbTulm|emV_O`R4dCB8!RmcM|HZVihJX_u`f(JSL)Rq)@_Hot$~H6y?5|(Qh*U zkw}gbzGAb4Z&k*aj;+hc%2F5H!e| z@pzmNWtwZ4JH0w+^kOOXNIgRmcTa;I^DOap)h9@h?1LS1ig4bRUyOa{WYV3>Ixh7a z!D!}5u;_7w()z2sSzG^cT@(F*yQ%;x<#OQ6-5ofb#OE5H^I|OQWZ;0zLbB`NHq>;I zr%XpZG3+`(=89-Tk%b~rsg_4k(L304Dv;SUfv}fvfaQhiWV0!QTNvKX)rF_P3wE%&Y{-ef{&Xc= z-?a+K#@5^$4D-aZG9hF3%4-~f0A%Fy1J8YUx9 z5sxTOr^;V&7&(z7V;6maf#~w)am-1Zm zeVA4u8|2@Sp#L3RO>89m$nKy2;GnD}bLY%q=D=@TFw;`uyv#18{&z zy+zp5?m@RXA0}TSZ6HNn8aw7s;mkR;n3#*Cz;Yd5CRjh0Y5ko+Z0CxyeuCYgzMvZ- zQUln$#W7Im4uJxe({S5z2Yi3EANFVzg4#Ys6nU};rFY#YITkNTF}sg{c({k`(2^hv zmOSI>J3J<5udjgcmBp|t=`G{uqX6A=enWn3D04W8f!Qx+vN_Q^jNXJ4bg;dH%SkUimRV5MN_Kyk16z#u!@FpA&`a3^{BD1E78gT)sc(Vl`;I}4 z_yMNiP9MotttE2vGRW6!TZx^97a5Rv!aEx!LEc#;lHcPicpQyT-u#dlU;BAr=X3<_X{5p`44Kx%@4-@o*1!PHHo!n%!FutOzFyD)mY3jE9Uf%d?f4qKQ`aZWz8l zf*J9uBT~!KK(0B9cerK>xmcV@%3B_g*lC&Y=Gb58BMO+gQlIU#G5B`JXEMWi7WBIw z1nHz!h>G8aHhST>qwG2q>c4||Y#)f>YOriXeQa%-igEjj$t=5X%!a-B%mdXfL=wz=?MG%10eEuW70%jaFq8e_$oBdn@^jJ;B9aqLmYg9Fqp)cSpOrFI@$ANawNV>xdgMv(qW`)7s$TVLX{}i0cq8Y5qrx>8o5Lkry9cU z9i16qqFvT)H_(o?A?$?RTRMt`)?;DDOZ;oAj(2X|#dGr|=+G-3YRr?tGqoMCZ{{f?+mp>~_v+-$cM9e8 zer)D`+^0*9y7|JJ0wMfqNb$?Xd7$_>2e!ME!cdYMiPD-v&V|G>zm72EZpAdxBPGG^ zj+|gL>^{s&Xkt0z?Z{dC1J@akqqk-sNmhH!KKo%9WueblN5(UiGv|^w3mswG;~w~( zyBjUqEXndUY;NzdIns>Ss)yI9J)hZB5lHF- zH-SuEHYg|zz;o^(xxIBBG;h5QuJgZ;Gj_3{^;a10D%~Q!UstjF@(R#79Z#OQ&xFb= zR=^3q1R@FT@MV)IsucvHVelnfwEF?3jn0H>h54+5qZo48Ir!B?0T??c0c*dDVxEpM zCLf5!c8_xCJYNE_Tb{#_r3dg``T<NN0?0j*3sKy!3z>VyH{d+T^ zXc5Z?FZIC1AvvUV<{RSO{*+X2i-w2+*6CAs2YAQ)Fd(oI!{a`qjj<`X2qMTR8iU%# zba?f-0@kYyftYd+sg@ljV(}DCjqN3o+KR-QD@C@f7$!?g9*{TdYDnyH5uzZzm}Cph zB7#4~iHplurcOefbUt4Iq5u8^pQ-FF`m`)|OG#sLogy|0&p@ZaS0HAb4nI$Bpf6q> zqVwN}(5!Vb^uJhr+P4Df)3@_MT;BtJnnr@Ksx|rJ!6jwF?ZnRN9_%vf1&{H)IHU0i z1n-r`B{ri*b3?badSOE1L1@;KD3QpGEpfaW&?(5EFF{xfKR(o#X( zpmqQ=z0y(7?lG?0HU={`O-EOSEjU5Ku}JU|p848?$~{@c=?xc*1R_AfI1rv54FR>; z89>;4l3?%x@Cm*3L+3V z%w~vQ_LJ8Vzet>DIk|eihdiIy2VyL*an+qjc6&&L_QK4bj>D%p4# zr_##@Uy)fKr^0Ktf8eY8AH0;21)6xD_{h$Nw1XzFr)vq=f2u^`ir2W;SB3szKU1$0 z>ezNR1cmj-a8kAw^?9&}^%wdrZt;yo#+oaS{8@vxLftqTk2?Z)5cynP3DBv5YFp$9cZ41EFItHGZ)WD_JIw-3VjTx0yNLyb)I!_#Dv03A! z_WMNfSr)ml#DI+cz5vO>X>e_7E7bS-(#E`%ROGWQ&CXSYPpovguv!|DrYn;zi&U7^ zqH;{w^oMYxARf*O#KV50EcmiA4zzo3fJJu;e*AnNQ%0}jUw1oNz|5jzo8{^D$1XTr zVu>5}O+hZ(S9>%14)1?1!Cip|!1I$8KisU@EHvpObNmvB7-u?;kXLY6D-Y%-)Z&Aobo^0%8FknE z!G=}O@Q6YSE`GfNJ}n7`bk>v%7*0=H*x z#7nGWdKc>@U8Kg&zIyMIH!l)MX#8=qbgmv$Hn5pT#Tn$P&3WhyI0}($#PsMr7`~YZye3yrSJlHL17Wm# z-31B%dT{vqYxLXw6c1InAg0bmix-kOV0sYOWUt3lpUlzohAp07qmC(e)Y0_W8z1t2@t6zo>8Z9w)g*I+|)r?csI&hZlFzP(ZpqLp$r@c+41LexF zl&l7;MoSTscVn#!P>37PcmEou8}h}u&X(3(+4 zJ=lXS_B)Wfc@@%2UhwhjbrAOJg1CBC+9n?lQ;QMQP_z&^@LZKhq`Q;2xD--*shSk1Yr_)u?EU=h6C!0U1@|pP@aq&2M4lR!uzkQ$ z<`{17NubxIk#CUano-z+#ytKh&U_Q$viy4oeN zNJ)eazYwQKKWI~lb2jfElR1n+|E#k|6qB4kN^-vol#kl(Ex-)Q!e) z`%4IXKVJazM03E#qY3Wsok`~1wjCXszxWRVbG8MzX{P1PQ_?a)W?P)Hf z@Z&Sa`3>S$$G3RCA`P`QuVM`&MSH9!V130nOtF0mGA;u2daw-D*{)3==xM?#5(4|A zwxd8zA`VHv#cw

79<{c$LFP#W7t}saGVsE@Ux|wi%4?+!)3uZ3!uBnn`x-$bcyE zOCa51igKSQ6C!k-!IOMwwL48tAKgssm0pt1nqMKVMH=6$vbmn7O+?CdJ305=1}vZW zqXOF(Jz4t@Tl{Oue{WPlccv`%*UhKK3l~z)=bPx&kS4S(AHt4CO)7bB9_Y!ghe-CU z`+0Le5I^u{2hu;zEe}lLMu7;p586cM)2S1m_!@{a!W?#23 zW0-J?_%1&P0#m}FSw|P0*1Mvm>`^w0;ez%8v3PMH6#egn;k~)XL7>VX40p%C(Csmj zZf6E=)(c_V#p9S2n9p{rWa-l4L}*@p4Mg?3K|RU<{;{3nq$}6J-gyQXD+MxHi9F`d zN?V-du7p31G(oaHA0AoiLFwru#I{idT-xoS{LUH3^U8swo>S1dQf-NGzJH{T@bS!k&`t)*5i}avy1*5CYYG z77*vea%89PAU6&P!RsBXA=Ym{WLbv7)(F;_CAonJd{QH)`@@)-gMPfnhmsk>@@D+^ zjgcq3ShC}W3Mrq>=Ds&P25F}$xZ`^^@Iijy6>(kcUjjczLQN_&$6dm;->bfl)u64)Zu@Y=0O*rv= z72MA8f)T4cc({23OezzD4QD1uPS6;17}vmc(Qv3eT*u7nx11@tcaLFP7aLAcW&zDPAfX@?FLPiJt? zD4ZeqLro6dwfdmnR0&F0&?sl9 zkdZEbO3cT$fX%re@RRc<4H5<9$k!pVs9;f{YH*c^8f%N{Jk-|SxJ%ds9vRv&^5!YT0A$AL^v3L(;$G|19( zI=o+FlgN(gY$x^SPuOCriGiEdQS5CE8Ma>!3%s>JX{3uhvP&S$nGoXr3!!ms4-8vM zV}-Lb4ED^1GuPcAvvC-cKPu2~T}sqYpa7N`X(MIbu30u-*fO>UH!z4RF1KT+Mh-5S z?2X?>5j#o-!(D*kSq{YZ4;%g~=RVO6pQ_^lhc(ymIRU23!0^C|PVUuU1?DT&p2;s`YHuj>#(oNt`27RKRjAD@%KaUut@$%gEGd|?sa}H{ry9hh zXvTA*$!ww+yajSrSeS17c!wjz;v8~Qk~q4R<{Y0{TX>50vb@u_-@GIY5n__^X*(Y5uaB^#BBeVy?E@w<%fhag%a5u zX$4_!skJ2UecMxB?89B8hPw>hU!?FXGC0I#wJI1y@wiDN17uV;1ZH&HF`d0GkN0F& zB+pahBd1HSj(GTLvOA?uoaE77vO>cb=)Wejg$*6t#Ssi+>idebWZPsSls81qm9FDj zX{GRjEEa;)4tLI#rKh;1bTTttyp1#XY%_1_Z%^_>A=|7#G=lkbT8RnaMRT0H_HlWl4MnG?1liuKAjaMa79IQH=s zq{hRGW7}IwBwxvrM@h>#7NVZKyzlAcdy5%IeEWNz^SKnJ|J4TW3HH4+Yw;%hHLJ?( zn(icSr^jxl!`spHre>hoIU7l?efvk#JlAWSB{J8Uufu1#dUuaA|Ib~?=|*vsoJvUO zcrY>h>B=)0X&}~9qF_w~a@uy^=j3+FGL^q}bF*SKpfGPIrXF9({q)U*(Pr;T1|^DS z4omWQk^jVbriTYO`U%b4-|pSKp8`D`+q#b=I_M+Oj5lzX)0ss0TBVstToX^eR*Ccn zz2YQ?`!QvGbK&9aN8A=`2j(z;A3XaI&bgx#z-_JF&9oaFH03W6;rPk$x#u+xa5j#{ zGP8fKhHoGLm>r*dlXUl4lE2!{W+Il_jQgp-ywmcTron$H=d|5x?yH5vX6^T?iQ;e@ z`6P9bJM)M#uVRxIv$FC4CtEX&W8brZiQ8q!nRJNH9gwqTd~_Mk&d2L``68B#cf&Jo zg8Ue{$9CEsCMI$01N?b|CVWDfSYwuj&q>=R%bm?U2DOJAsE&C~l#hnfJUeGvd_sp- zRPTk;Mhwgxw}HU9*Puzn55|6k!Ommn!Bj;X3_TZugt!5`DEvSQO`ekJA-y1XWdM|x z3*$(?Ej7^O)AdbCw5@(E3jJ}!8~cs$<@^?qkjw=~>sH8Tu7jfn%e@Ib0K*@?W3^&4 zy4}se;Y}CF;@WPqvqc*6wrb$S6(#iYl}Bxd7MOFX2sB29;JKP~J{*$Nw!DL$X^DFw5;_UY~yJx~|`K9OIt% zy;u8KU-kLFSI@X~Vf5trH;$e?efnshzrM7(lV3MEe&QL`AqRh@KJgJ()%)`O#|u95 z+a+5s`|#4IeRpE%^u~ME2QMGjC;VfJ`ur`PSYLa`=Jl0_-CaHJoVDxEKjoBa|Lf1K zzPb~CPU%){WnzOc=!MxQzL(e-)T z{;E3r+xJy(T7FHn8PERn9`+yAwGaAKb(^)Sdx*dD)Hl`tefnGLi>GcTj~w}@qp^kTT~jxBAp8j;jCor+3!-e0{Fo z@9(2=_1!JUe^|2V_~s3+AFuP7W#eZ)|NQaqzWUy3om*d3@lRt{Yu^{9*ZPmwPJiw7 zCoKK$YtE{^@uEws4WDpnwac-;sV@DwuGd@d?&+(q*>w7hGxnXn>jyXS z=K+`3pS|yb`ddqGtxvk(5B1BQ!=I<;->UiBpHH9O@P>N&lq>%@dhOvCj2>{&^`mP) zed*|{KYYRH8?%Ru-gEc;mi+QR-@4QCJ=R|GnD4)7^wx{+o&MPG9x~eX>UFE+OWF!XOFMn{jZ1DkG|lc^;PRWsXF)_*N1h4JzyA3Of-kKa39c-v*81OEBN(E|^(jQcHu(7N)!v`jrdr_V5}*EvM^t~mVCyR0vunL{w~gwno*L@U z9rDO}`{#bAdfkSftbXv)tH#@Z|Ge=nN4{-*2me3x{Pmty{d|MHtLNYO^!l8Kyr@2Z zmwoC7pR?uY3#&F8?YZQEqjjo1s%KujLG}6HZ(pt2;9u2epL;=d`P-Z7Gmrj4^@`2c zuP)kWz3PEmu2mnh^2+LppIl#E`QSsUx8Jm+`r}vUtB*W-`|(vzd*b-Mk3DvL-BBN} z-*eH&>OVa2g!;z|C)U5-zpfv5&ZhNd^-0xJzVOoO{ENRi8lTLcyZ!w}iU0dRlRnb zr&Leg>3-F^#~(Yo_H$1jJ?mkk>a2H6ReyTeWc82}H>|(5{EyWqesOtq(SEm&uD>{q zzI4Y|N8f$$KJ_t=7}ckJX}|iuQ$JUI^h@Vf&m5mrz2dPOS5tRyTphpDW2)P({KRPQ zOTRk0WrxE@*In?p`ci&R>qo}#uRr?5bw*Er;m?--_U&6vA8^@Mt6RTyQWYm&SM4>k zQ}x3yzjt&{I&$==(|=JNb^Ap1@H-}}r(Stp^{X2`T)lBgTV4LG4XY!5{`k@7e{kpY z?z?`l-uAMi>h980>RtBReEi7f{^K2ge@FeWAHJ`;|E4Rd>eshaPk79G#=HN=Y2(iy zco1u8*B|zrzT`FYd{6kj z(LaCjm(hAR+%)>+iGQkh-t9~Eq1&8L-}UpijF#^9q|r|w*o=OERE5twL0sZld5aK@c8jlU$Wu&%isKA{o;qdX?)_lcOQTIqdSg|+5Jt`1*aTb zo%oVXs>^@1&h$$@@am;=-`aBe$8TJ#zW2AEtnOIv*y??sy=`>)#W#$;u}3>v=YiYR zM}J|H`YZR|zy87t{ycr|(PvJdeawGPk0x&%o%FTsN54y78|}CCTH}2`^zC}LOTShx z`QoF;OXmMopK{~3>+ze9s}8;V=<1yN>goeqPS=0haP4}74O4Z*9#_-{|K#xcGfSuV zz1~yRy^ncl^@LL%SuH*Dxl4CAV7z4OkFULS)6EX9w|?8odZ(NA9RKW)U)IOGeQx}- zH~nt>;*Z@qy5^RB>o@%QY4!Bx6V=b(`^swi)gQ0kcFhZ`Pu}p8dYzvL(S-&gJa zwhO8ce)sljkH?)~KV`cM>c7A2oAs+Ee^i}+z@CVdcUXh_YMB&_tokDbzZ&e+Rv>Y|LpbagFgPd z>XI8iGTQCzZ}NA>PuHWZZ{S~7_+<6#Kdw`MYUA}r2Oo6Z=;+^myE^1ukDEU9?e`nq z^n#75uRZTt-a&RtL1m?Sv}-yr`Gp>&bjroUi0Sa zrr$iRe(@`gsi!xcsJD3LzV+LGv(b2$J)c>R|MS!JcHh0RK6;;v>x~cEVtma9uN&{p zuO)AM{0~N7|JKFT$zS8wr{44a`to%)8}ISg^QK?^*Dw6rhsLkl_*>PX|G1%g z*l0<8{fc+gXZ+$n>Svz$!}_jemyZ4xZWulK`fclD-Y}}KIpO%~?jsheD~>v)x@4W@ zqoa?TufO}*J9(zJsJDIkN9#j=wPQ7U@QI^y|M86KLx;{+$A0X>`sBm@Re$!|U#w4h z;mP%v|MJOt;UCwHc7DrR_37*0Ro(v7B~|?AMEzR+u7HnkyiN7}2cBMk?y(=K-}#78 z@A8qI>rF4cy*}r&XH`F6ckAkn`#!3A>DBjC=RW@j)mD#uM*a2!-%~yDuUpplo^Vk8 zgZ1~UE_}gns((J=m34LA2S(?=;qcMBZ#b}e&z6VP?|;`ntMTg(;O}NXcKZ60&YOP8 zGnS1ueA5T2pFi&6`k6n8<1g=eP4%@;T~lp)|IO>0Z+>Y#yY$-e&0n~Cy#IP9k50Xh ze{1Fi6Lo#(hV|bM`rT;DjXz%9dfD5n*MI9N^)1KVT0h~Rht;Q?`t0#&{|m=ccfYT? zuC1$gJ@U%x#-lE)_uAoK^|#-5;CS|c6YF)}vGI6^r@dqR%L_kIzihkpcaH17YUe*4 z^Sye{KfJwep7POp+_u%eAO8C2)b+0%r5C(^dgeE0Oy3>rYG$Lg>JRL2O?}z1j~YMh z#rzv+&s$j?b`rbnaY1$6qh4KomcM7{2}`!Em%QuO)wl0_WBr~d>@a%cc|WXfe9JTI z!~Sye_@7UD_js3^elh;#=U!NCwDAY(2fpNK;}iaHPj$lWAFglx`o80-N1RyicjtEF ztM}-}5Bbfi(KnCSvwHo9L;d_$TwWde#-~)fUNT+%?9;ne;Zdj8D_{Tn`Zt?DXY}WH zeyUn>OVhJU2)-F^^`*}^ zpnlI`$M3xVPJ8V9l1(?>`6JJMXZ6x<$Li_#-8cID9uKdc^N$sy1GhMGw0HPyb@oO_ zR7;j^ULE<@ZAJ?Zf68d)g;$Qg`>2Q4r%rsiTKM8M)hGY9V*H+0pFQ4c`LW|Y?#Fxi z(Z8-Izx@4r*(UeYTRiT(`iJ~o6gNF1)O#OrU3Kw|msMx|<6YxlZ2jc%bME_7z4E)) zR^R;fU#bni@TKa0TRyBlgulb~p4Y6Z4&CJ2qxBwh+UQ-IymNHNlfPaq+v~&CS{uHn z`qa6bS8x2vrq%hA>sBv$)+W`_P{OFfQ)s2rIAARXY_yZW5xeWU(5KOyz9CD&CapZwVR z=8Z@E+W8a5uRZ5i^@=Bay598bM=U+=p7$^P6aS{&9#?*UG&}Rg(ca4s9Xg5MaUv~Lu`bDq2WO|!N9y8wKA+H{v@x&L64?pd;>aM3<&hHa%t1Di9 z*?8s|&l*4I!nMa&{q>samw)+Qb^AGAt#*9my5p1Pep27J-FNCY+_E{ppXU70vB z(NUXjHJ+_E9zS-yRrQnBe#-d#2W>LGX~zeR_uA!>`lF{@T>tI$C)8UUb^7?sGfo~q zcAwXbpZTpLt9xHHUH$pojjP9Q|AFy;f9}7=N1wL;cYM&PS%2e< zFOSwbd#~v$UUba#!u6l5AM(U=>rcP(xca;Ne(byU-GB6(qrW$O^zHAczjo+>_2z%v zvp(;q&#JCDj=u-@_&<#vyun%Zj}AG#{@7XXuXlU(wbP$?$LFX2xfXvP(y#fuyiS^} zH(G1+`k#laUqA5KcUI>g`h{xq-Jd=B%~PwQz4eUxr4P7a>2Gh^Z2G7>zr6J5r>#4F*SGJecR2dX^`CD(qq_N_{M^Q6{2O}T z5A{jsyrY_X|C6ehzUl1hvVXn1dc?Y~t8V_}i^)BpK7GA|>gCOE$KP1_$?-224jxa$ zoyJG+zwY=|2i{SiJ^P^1Cj0+j`jAV$J$=r1mev2Y{?Ya6hizYJ4YTtO~a*t?s`5N27h3Mptq*v| zMb)LxIHG#h@h`7#;QN`K-~YvWqwdQ3#z(zy{KxZF)hFC^VZHbF_N?!|ZOi(``>k8Q zeAh43&)w^S`u2PIb+H3qR$aK+U89d)e9`FjXJ1#{!rygv+*!}7ezyMJ^-r(Zss8IW zPp)V8U2px1j@ss&Fu7I`v-_^SP6+F-AO1IaSlBfjc=#({bI4l`-u|Ei-+bWqRaMtJ z?pRfO?#{ok{qM>!$UT1gwQqc0U;Ll;So;672mh$!p1bexf4|57qxqv9cAWaR=7-Si z=GkBS+Vd~v;)D1|ApOiM^eBKu(a;M)Gpu9a;D?zlu`G$NAVp#zl9TJ9&6%XOOT^*L%k`MU zzO7s=d|V@*uotDbjSqZ-ZC%^Pau57~Z%iFQirDNB(NpZOn5WGhD~p;t)HXK&>-F8$ zTDk1Y0VSU!mYN(JY--#if!qf^*U-oQ-qX;-`?Aj>`rgh9spsMHtiG-zmW%`Y$6hzs zxi7x4!MC>4N1wQ&6kJIqYLzOH7>vc7s-KCf~Xa%&jQTq`Ah; zb&j!ohJa6BV-1}BJM0L|8MJ)96Kc(s55%*?UZZ*D_8 z0!3{+rIJ7Cd~9<&uz(+8$!YPml_r(^A(nY_;jNkK`}le*x9P{y+j-{OKlX9+ob4Uu zUc`o+i9QD=_iti9Kj9Ng_z`p>72o;xv#&k1@6iU{i#6se`89wGvDC{&Z2kPKQseU# zQ{m)2XV&n`h_?d11M1kLK`Z(RSl9NpI*YZb-^07XUd+S3f_Q$7BK%Cr%UVCJ-fQT$ zsEOq~dJfNBh;Z_V_v-yL2}GG9*^cEdx$0D9lBI_`z8?a z?Kxpj;AglyS!>PWXAowPhqEsC!M2Y(}H3O012ApQnemeRGXA;`xROQt6e@+O4|Ji~5oDp3@H4Lfc;U@6g_T2QU0w4{+|NuMB$FqfPxT5sjU9 zq%LX0C19?3`a$gsZ26!j-&hNbM<8!6x#2m<9JRC5EdcqvBk4`z3(}^n%Kc!*So%zv zPrsZu? zyz5O?^xwNBcEa8T+zfAB0@s4|;oP^4WoAom7C7Il^B^Q%u@eVyp-kSH6`;%+FXoRz=4G|zI zl^*WdVSDy~cBVbOKl@nM!kcex1UaUi)_6ZtZ%i?6zO`B7Ydg#R^J_!g`rHdW2j8Oe z4D|5Y*fW6!bUmzTQh_z**xMW>#eOb5?Ry&t#64S7a1>}O7dHV9g&%s&HU}rv- z-HnC5N8TIWxHG+niE6L8s7jar}DlvQ6Ka!K+8Z~Hvcc!g4%Np~$2!SoMjoG&q_z~V% z1Lonp4|A+ltZ7r}-C&C;a;($t&@I~fh|WSE{(+beC-x=V2b)>_8`ZeVh+DS8?%3mJ>cY%nmdTS9nivg_RRB$ z>pN%g&2I#}ux|qM%$NleAQ2C0ctu~^+$=L?uJe^H@`e1o4|eu@-6eQp0Y73Vw6@6g z^y;vktIjcdrIMfP*;8p#sreA`{CORVZ?1JMh}ib=y`U|pSi2|@E0tJ|V<#XNBR!dC z{v30z0G>^M;N4qTn?Og*cmPf8H9fTDjoZiE7Bu+YNvBO8?>)i=U{1E-^8G+tlXu7)bSex$7S#Y7}hj~#B(YH=SpzrwucAHSzNJ^!*bt(SOGnktN-6bqw zj<)%cJ=E^mqRn?si#8{SK_ao;iZkqHja))^#;ujZ?$GY59`3;bP>-^Ox=CP84&UX2C`KjJIKyl3xd4x}gxPhOT-dbzJPO^-J9JUReSd9*mJO3{5BT9!!~YA+kLznkZB( z@-$oR2+(u%Z2SE(*Y}9Tf;ph=wY|NoNg%Hb_h8Q$OV5A~_!-za2RBeR1yU+`5p8aQ zn*-XJ!wfT}U+#hJO#91ssk}aTxyCwoYf|ZLUc#0)H&W|+GrvFKBAl_%>sFhbVNYiz zInbuk*Bs-%mW?*B?>sPnR^D9qAQpgUwMH)2VYl$sB=lf+eV+S7_#r>yFLL$)IWFJN z(>G78m@`MrI_(zwwmvt(g$NJaZ5CK7r^qYnHkBNEHP`{9h(==MODZ|$#Xcv`Bc}EY zU2mJ)^cY^(mRN!d=$ucDVx3$}g>TW>hr8A}#I4Q!tkFhCv|?T6V&RPG`#uc`(8HWA zmOA}T9@r!2<7zAVDHYwuGTU71GEgZJ7pd=o?PqJm57?8~Ep`Ih=C$Y!9b+hc(AvZK zVat7|lAn3FA%Cc~R}is}q1)bXu-!5D^^W8bdu61;wQ==*nB!eF=#|!@7u295&t3`6 zcndsd&Zqwb_Pn{qV<>x}?C&$E4a8 z;Ov)q>1`IU)Yf?;jGrF>@!;37u=deB)Wr_Uux54GLMlnqy5u=Njx*E)~8-2VlGbt^4E4ksJQ*o*MbDJMVG>2{(PV3 zom_ko`DhGe%{+Gs_`>-ql^v{^n(5<&TjEnI z`hM8vnV)wi&-etr?AH*pHo-+~&*UBX`H^jS^NrbO);LEN7a zUEU=YuGPmD0Ys6?Uc-Io7~$+2TaFV;9#$56+Peddh#}|54QJO*K%Iqs+St#dK4K@} zXF!`uo!pQ=+|3Ychd(XmF5$b^&`TYuX^C~c?K$=HdF|KR z`5rtkG6A3WO6Zx=(|C%-4)6&KHMv(PJK5X%NW8;#zUOiG{62+iW=bw>c@`5!EOv_> zd%MHVwV}jvj&aZcduE&6)Yf$%3p#-Ye2Ba@1&IDUxC!iler%kU#8-Fne-nj1_pZUoZtABi^D$g=E-)$;0gFezEYJs~3 zv}hs<9YOwFm1FM-`nSY#uRsoV=zTf;Mx8>I#dE`HPcWk)rhPNEdFIP$Ti^C@R({>d zYl#Km4ju7agPnCBi|EDuu&qg;16k0{Rhm?4W9-l93=n}et#SIhuXj4^9pvp1%YB*C zV%uln^{1$F_xx;;?>u>P%=P&j-V2R9edMhf&S;N`zDJir7SETkhkW@wClV%infZvH zKXV{~2F$Ayek^lZ>`6Fh25k3AXl2+jrc%F>yw-Yh%$=V>MSOM+@^B4)3o_sLl5<8Q zjHi(6?Gf?qC)`zOWAOtzCIH`|^%H17t7dMKLcUw)o6~{_j7``t@*hYb3)&p{43x8H z$u*|~5m-Cq$t#{ArP8~N{jI9oU; zMhK8!e{{|vZy9igdCKfex!*bXlprENth0w`1ipPcZ1-!?Z7O^~XIsCYoqcnFare)= zpgVkfdd7(D?}H1yjkVanNu^(it7{^8ITtPqzBA{o*M`@3=H$%k`Ob;N;p zhqLK7#Ldfy=mD6{&b+;o1a9ODXw(THIhOeO_1(*Ru_6Bo;G8y=du)hD;GP{^!nTJw zYGW-3V2Ydtpq>EvH86%)7b4Fme6Gj$K00||zlhd1Zmz!XX?4VJ&>7nFaVKNuC-fA# z&KK4TV-d^|%aY3TG*8YXTpP>(7YBHAGd$b!(%V{h%jdz4)J=grmm0+!=eOtp+)G&J z8FO!KxLJ@TmYN}d4&FY#cEfX=b-za<)_?`_hUcQ~xnZ_(xlZ0Z^aTB)zMz*oL}Mvr ziD-Fiv=!s-qfGWXL>!R-;@!nq`pPBH;J4@wZA`7-fPfu)+t*1y3pzhb>=071<%hH9 zJop%U&bj!FzM48=&zu9F;9KKjH+-};i*<13DtUIF6H=)$PqF@gNbhGJd$>32gg+MN z+CDbeM9X^w?R^fsdpSSvLTx?^zWs+gX;RNGr&vk(D@8>w#^J5cy|7yldoJ{nXTe!J z#Ldrh;f)XC_f+=I!pXhE*0&gaZN_8vsvoPm$nDHgQo24Bv2hd*%oe%{W~ zGr&(Vf0?uOfq7X{c@76Z_kr_s+`2~4%NZhoi5cuz><(WT%V&>$4cdC)`(W`sLoc~T zq%zxm?6Jt31Lxe~HOzNIj&aZb(3h<-T(MZn*gFCK_!pyhg*h+9BT0>&Q{^ ze7M`_Z%&G3U$uS*8ti~>)xesBpJU9I3o|9poFAA$KwiY2c>%XL7BsX*R*-$U9BygRhm=F4SarqU+9Xvc!#PCS#+MZWYHY9qcF z>bg{(nHDbR7y~ht`NLe}O1_g+;^uk&+zZ{|3!g#j?9qY*6rXczatz&Q3nk)nkJdB- zTQNt@84253WBIJuekREoZ_z$mlR(|A!8h2kNA8WSHm8jt&o>@`_43-z%x6Y>t_CjG z4SeWd2lSdX_Lv2s&$EYI1ipXURCdp2!OlC8>)sJxozS7z?l|n5U*DOskNZWSI78by zd*nOfjLx*5{=EIjX^7|7DIoa96?bf7!EB!!l95u=rP9BNtLGTc=b*oR?PpxcJEUxu z*T^}s@JTxrKcdaey|7zgUf08Z5y+d@p{J-H{{Jl?<``dqA3hi42F$}p;@)2e9|U$n zi;w`an3Hg*Y0xc5z(03npIM*`by>&qjJN2J3L0%@Opw0->@nO`AeM89Yl|kW?%m;= zAAvf6d`FnM$8Jy&2Mqj>FPC{W;=cB=&gWqcG(f(EkFozO<8I9h!W?u0#wYcu&%htz zvFFWkruEht_uMVEbMzJSy_394>ifYBeR6Fi2k8=DKoij*fjMd?$@lr4L!0CIr+Pc| zG3(6Fv-{izPTo3qv(7$4ZA*S4KJ<(D0b~u|g28U!hu1dW3B3}Wabq!sEctzg?+ooB z*I@qqLbt?40#T6b9s&4_Uj!CF({rA2@&1@Qi*K%VLmzv2PQTZgH;yl+W~c>S&PK@waX9NJ>%bD>*pc52X}$1q1eCV0e}1mw)?QrV?J+e^{T63c7u zIT~Sqa@6kUZYh=293$VJ_Lm=KEcT@?uECqD#F$n7;_MC#wE^3B?t!f|vGlQ*u^hKI zw)Hta*zo#6e>Z@y47v8`V(}B&XQf3uci0=R)W@QYDfUVU4f3-QS935qZca?t%vN&m zxH_*(klbT_Z^F0ulkAXvYS-L(iadWDQ|ErXLv8p z9nSCTnDdC+FR00xCr3%K)Ee`-n}W|X5}FyoxPVV^xi9r{c@{N`HaRVP=pjGEQqK)% zaE}g5P-DKm&lr!~^+C6H9(}hA#;Ygjjl*+xC7d$BYg?@6qoP;RAET!23FS z_fzuN5pAtK6lYjBM=YWB-FFh?et~>=XIrPO=m*e7IJB_0k|I7kWlfDcOb|<{{P@+O zgE_|F7kf~n$eZ7$(#PB^sl=Ly2eKwFB!FZ;5dvK-KbGZt!1fuPCHF)S@y$_&`PyCL zODu7B9%|)>e9tIv4|B{6_cP?*@IU@GI9ztG&t-q|V&$}jgkkp(jpku`D``W`@=YToJx;}2Md5u0W z)@s9b=pYh4eG_&BExs{1d$!v4z!%wvTm1QsP-=!b#wV%s?iB6(-ip*GW`LpJFgx@y zpMlWl%MI}k+q#xm!X9Rkk;T1kBi6)z_XLfd_tgds$hGK;%;ikZlDD^h8{x3C#FDS? z8SE9|Jd1VK*sn$B7+TK!Mva&sXG-s&og@Le1KyFG&#Bt{{9Ln3qYpZ`1X_J`=6XA! z?UC;T+xuzym_H6RvBdKJ@HwwP$I#b&_cxaBpPJBPIJ3`}(x#FZ(SvP$kf(<_&!U%h zmRNR{kEySR%lUHHZSQ9;^go6Xn+FZBmwE1Gj=Lspv^_g8%+2?KuQp~~uAxrbx)vzL zLj>fu{j3IVh!e@Pt?y!~wPtvJbMD^YpPh&;a|)&GbHft!d#?9}8-bXJMM3 z`;r^s-PIW_b_dpsHTdRu=UK2*D!mua;7oY?C~YcdZ~ML-{^Hr<>N)L9c|9l5!*1qf za0cgkcay*#0j$J!&IHI3i^}_@)vt@S2|w?SZGGzNqBf9I+Q`fVhYsin+^6leG2b_G zIfieIGTfE9c~+kj$XA9w)|t~puzdCW4nKj;7<|C?Y}ye!!5gVh;G4ft~QnTqP^s^q@sqqNwMS% zpC8ULKfo*czQ#(E`nzV&DYzz<&$3x;X9X}XkM=Bzb5kmQJ|Dh1V>c3vq3z)_Am71R zW6askQes?P7gIk|9y`yacBspDN4`C*o3J-pJD=5_ss9{68&kSiUKe!>+$YWuK-|;* zF&2Axef-=r$In|idn2v(iPr!1?}|w}0w)gJ1LY&ppU%;Fpn?>#^N^3RoBL&C9v^NQ^z#?-G+_;aAb8 z!=6Gb#@fiUhM#+2PZ)!b=vgp-H8Eh{1X?)n$UY6ol8VYZnjg#E4(G|Q7dah>#9Cl1 zpgS-}Pj@h8t~;budKhc5jhU~O^Y86$Qt6rd!y8v7uP(YFXCABsxgL76-AjLSoFQ+$ zJsa#Vz?@+p8mKVJG=1a$2CwCUuBLTT=r}8>lqi?Sco%dyKs|{NC21GS}0@=pbo}l+iFfjv<_Ith-KH!@# zKipSJg&TTB;&~pt_7JyT-@Kg5&N&axyHdKumst8WX!Dem%Fga(P0)}1`xe_B8+DI+ z@V8nKa$Z2|d$;qTizUzf?PZ^Q7HT_SpB#f%^1O%!34~rpwDa733bgq7JoMC`1ByE< zLoD~0DfubE3u-drYI`Mhc`x)Tu=xI@*4Y7W0kqaz$9a@&M|8pwvCqdG#2ooSN9^1u zl|JsCusxeGbG=9HyeG4~OLIg3H(LXYFVlvLpp7Bd%Xj!6GKYHo+!s#W0JU`k8f>4n zSP4i!9&IercesDD zNZ4npsnW8C}FHa~+L`G_yCXs@|<&fhJE7Go)r1DERvC2Bgj{JNmM zkHsFuls3Ud3}V*#p3qhsvp#~Rx6KvaXBO<#V|X4yB0_wMK7m*ZGKY?!5uDMzay*si z*jSF^H^k;Z1f4uw3ycSmO3o_c^C00n(>(1~UmZ*DmY6g38!*-Tc}HyXJ78Qa${Kb| zWv0Dex}y9)y|Wd2M$n`Js4UrMt}^rwsqENLdl4A!Y4Nq;HI(lny|5$sE!d8k9lUk} z0?wrurg;bA*0x}HzYjYnxRm**zeren&XVz6Ky}ys1rcC)W3)5-D3O9FM)HRgDBIT4A+)X(XCYwVSC z(5*g_UxS1Z%bwTvJ&i{&^c{AVPoNR*kU%e)TVCRsw^!_E!Z)cTTGzt2y&usXx$2+| z68;pmXU#^@1&(nh`{{sZzp@$T9PY`TQz!%DnZd-ve!qJ>)|Kh&{4iUh)Rt9z%Yk4Pq*_ z>O2b_fajGnuS09Ec~5vn+=qF?j1-~aB3jrh*J3ukopbOr57&@C1^n!51)TTU#PZ`n zz`p`M;5(;9d&f$bvdYg_(vGFh9!Z;XOi~-M?bDzmI>+^icWU}3wDkcUorhNBhu1&A zyQ4qfK8LoSH3{T9K*!#fx5q@^tHZXB=hQZ4Kl2)GV6A=3GiGfA7m2mT;I*wWe-W78 zSqqnUjs2a$=Un^_q*Rdm6B=e2Yx@%D(^O4pd44V8Fd+0qiQ+iIq zD>*m8aX#$^WDXs`JbA0Y&E}`_Ji62DOt~|E{zP6*amO}c^cc>aV&9jzcjAmzn1{~$ zVGny4*EZL9N6enNANOs{yYTba*idKBs1MYl1u2%CAzr7g z{k%iZTa)jS*aEiuMCu~G&q|Kr+e1!tK+$%NVvT+C-B33VnrNiNLM-QO(dH$z+gsyuW>IqkY|t=iFIJ8v(}h3xd%E+;%hATxz&okH(Z2s zjx{aXc^A2pfU|EG`}@GoyE9{wxNx4Ie{BzEIwyg7xG9i#qnEF#u{=i&^p#=nlvc+B zwfQV?K_KRJkuqSe`)Fq#KVj=DOt{h{pB!g&AU~Vrl zwy8@!AC2|#6hoHWA1=rG9&6g}H_M*pul)D_}*emyOF1vWYYIE&zThD7}L3gRZ9K{|1MBq&0 z+KP3~Pv`(v68F#W$_ew#Z$MZB#@x}GMM;j_AN7{NNMHJLWv=iMwY634~aBVrA)4$#G7L9jP%#_epa=6c^-}lbF@a?ChSnR=<%RSHyu@*F`^a{T=$4YRI$i>`ed6;LC+UV`@(l zmv4Z6mrx*Khh7_#>(H$nXa>zpbIoblp@R?DN*fFB*<+Gf?N7OH&kyHS$4C;G>i1sE z>2rs8o66_(WZ$Rh>smG9e&*`uzW4#yFQJWj_JGz`{Op?t=4CFHdhfdd5xxUKpI!4a z-0x++@u8pj0pD7=77Q`ZuGphV(6PK%t#?1`8aQLQ4r-Q~4$gbA&b}S-)WaHm>#dF0 za*DabzHOo>Q0(2M5{uaS5kK^>M$Xv_KsynecQ$s;LpQ>?sldDl0_cD-{UN7K^rDBo zvY>R5ws%>3c*b3ZuqhTI${#okk38Bm6t5Os>5XQg6yaQb3+COo4(B52g+ zng=>;`*@zD?rrDpx1z5hX6-Bp36PV}&WPCdle5<(IjugDBCpp>HuHTKi#WVB?zIdw z=D-Cw>wOk;Ew;XBh&8d)=DY9rB|GmmVGjG)$KFaxS(P(t$Fk#a))vn1xAG3eay`DY z730>JAFw-Z!JL*n_v(7yc)-swbEt{x-VW#lGDm$%{cB=ez*%oUwf&kN*W%Cikc*&G z!?nFXhb?dID)8nN&pPRv0EVB}}kaM37op(T+lk?a=bL`c@nWMy1Uen<;8 zT>DP3gW5g`|rGLQCb&hSi)VAmYOHm0;Os}f7}CVvXdf{5P%b6>ck{JVoKwmDnjx9FRJ zKX>IEYy7hfOMtc31-KB)Y<=r}4H|S@lgDm*xMyeFS|sOjHsixRrBr%&4+C#+C8n%$ zowhYYj((!Azn1BrUv7-kTYetC)dxySS>@+ZYhsAgFU8_7z~wWVkP5Fq1+33==^tV_ zN7mL56Zt%`e@4-bp#Lm6zNShSOD}7O*VucX1o`zO$GT{I4c(=XCFk_HEjq%dSpNUS z9syg>lphGM@vC5s|=a_eaSfg+M27Acw(rUZK zZe!ouJ@L)8e?&*LIi4Zm2Ow{rSezI89&q#aLGt@2uIBBdSYx03|0e7z-%!)S8H@Oz zUlDTuBxerVd10o{!M~!Pli!#4N|X2!OAqrK>^u)!UdVYK`?{lCO68u0{?VtTD61J|Cw|#Fk^gDQ(Ou^O;K}C!xiV-)P4&CtwdfvepNKG!b7gC)Z$GlO>j2 zV(Me);rzxII718Pd)zbISKIssoqt^D&l@6bf%P4FkcjQ<2Ho}u=#)Y}x4m*d_mi6e zNWMonb3;NgPu+k7eBBpw2x-SyFywdCTnkQKK@G8dKjh}Tgx_=O+?yWuo}(tBAzk!uOt9v~hdPVn^cEN1*P2XP+XbbZL$IG9Q0{=g@9q zsk29G94!)d-Vv>x;5xLu8|(%<-yOEGw)fRaOvITn#XemucFuvf$Ao=|2Yl-Tn1t)( zv4@!PAdilq0rTc?f96Hd0!7<+3x@vQ*=^>}l$t~=fC%#WaMjj&pGuQT&pGP+9?Ltm z-z@MmXmLmCg_Q+=Pj{FPX6-=@Y-3heFjs6!*1ZGKmsR%$(gd7x#qOw zKmcjK(7-fM@gmYXyOwAjOW8te(U zIr-H8I&>)g%(KS33dA~aBL48_6^*@ZMd_LQIh!2oB08cwZII7{PATMFLeKf0bH2Xf z{oBjl=BlkVFU11;c^}S-#)0=Ke=~ZE75#k>SKKAU(l_rxzZPHK{ipic7Ci?7$gib2 zK)*{RzrlB3Ytq1>t+S{7XR!mEXK~L2@B3*|9y{l8#)#eZutvX4AxpxJ*qf{ z;koo9VVUNjddK0_B7PdIBJMtqQZ-JmlEJ-i;?p}aX=Dn0!(FiE@LwZ#_JsDpN| zLo9rUpXjv!0?2)-S4Uzkc|)Ccf)j(C^Xc!|t#63QwO~pP-}w>UfuxSJ`>wbC-;R3g zjbR}$nkQx7wb$YQIoHSgU*ut7*KpwBuYAoRZ#j7TgARQ2f!kO04)u;xJMFmV?sa(a zzboZFn#5l24BYMJnX;ev7x9}|&guS(wl(kp9jMDTrmu4sp@H?S8YtT4+sC|C8+2-N zjVaki`+gDD4R<(i|3nbj*O>XkZkeb5Bzf5;PuqF+h~(ST=W+pGS%s~{lvVDVV(A~r znFDL~bCz?1K)TqUKjB~-&zikjxWUi)#5+0QKI$PyUnp`B(4>;%IjnJq;f%&3`HHa!+|ye5Az#12pOR1XBgY;&pLA_VC(v`1Uboj=tx$PJUi{ru-R-e3$SO_}*hfOlcDL zv7*$QpL@c0kr;QYq*Pub=Xy8x4zcvP7CV7G=wUDCt0Qsm<2d*hKY$K2vCknk0iqo6 z%v&ttjvN{CdVR+kPSd?{N~$0(0ijG0_6O0PHt6!*dHBw0rT5Tc;LXEYEd^ zZGDzlc5m=!sg3e*=BqQX-Lt`Ofw9!XUJ1^4qF#T(9s)nt$C86ro>zQ(hz{E`tpX9- ze7PnSOwuD^D{bWci$4K|XD{c%hgABxhtj1|7t!|i#~XL@d{e-lF_xYUI)Hpe&(Zs^ zxo?w7uKDK9QNIAx^T5v%wJ~S-`K|Ac4G5|5Q@x#HIYWL{2-&}bYeAeTyBUk*n`?c8 z&hUKtYU1{{K7tl_H^N?iPwsom*VTP|U-|wKb6^&j=PYX%@K*vq3w`YpF?-oN0d)iC z?5LJQXPzE|ETcz`lQ)U&JveLF)!Ft_=4belN?d=+JbW>sP2Vn-*E8UoJ4cQ82W zze{C*_q_-|zm~DoTjM_VxrdnZv)~Wf*UXv)>QjP^eGj+>I|4a-<-5n!7AvW7rv`*I zftzz929oWR&2ne<#$UJ^%{lYuAYsov3#LGd5Ww18510S!{Y?LPg!asBDn0UZqfNd$ zM{N5!r^UCQ`3c{cds(@^cM48CT@9#PXU(blxe!_IA`p?AHS0H`6PmGRyM} zb=JralIM{>OU?pF)~%pNENIaprI719LxUaRMBWAM9s2tFtmGX18RDG9ynJ@F5))s5 zu_l&hF~5%3_Vykofw7#!et8c0)+zQd)}rN{VUM=g0j-_utwXHoz`%KC>#g;5$a`b! zdyc6-KFprL4}Bh1&X4#h7Wlrm#yPnMIoZY^&Oc9lm~XD1kAB~`3Oht%oK;b~+XQ?< zE4dEd-VN~ANY(b~Qph_m-amQPHDJ)@v_M&)#`mx9%RDocGrXRjbHML+vP9y$ftfOQ z3T+QDM{Vfi5kKD(I%!kmeW*KprHKVm&3^Xku$5c`mt(Qi4R<~1Ot?981nOaieY`t& zX;RtCd9g=}_D=M*8+8ne;=GW`Jzq=T1<-&1-w66=O^#=2>D6I-f7Tgm!7yV2-aO+S zd`u;-ZSCZW^5ej~d2l%gJ7CYLl~|ttS@S`Q@2nQMk3Bo!%qA92-@7r_cmU=q?&Rw% z%(EtemA&r_dl^&on;va~i1?f-f4*Zo!)Ml<1(3ix_*3Yd%UtWM`TsaM^Dw)nEbMQD z1VI@>5M0EpDI}@7=VsVd=J}en@y0wWS{q_IrWSLVwV@qDBIY8b>Yh{Bsi0#n+7L~1 zh#{JgM(gM{eD81Xa}(e9=eM7Jv(|doyVhQNpL1_j(*3lR7F$!UqfL1M4&+$HPC!G> zhy%n~X;MOVz3$nTjRw7hp-uM#YG$)cyR;qM!KSJ1yhFaYvHBiY>9k0 z|GaFl+en`)*C`D>tblbwNqj8p&sx;Oan?@RCV_o8M>DfR3e?C+mnhB!bj-<}(|5RX7T$$D*v~1p#Paz)sVc3{i!(O#N#r++ zZ&Hs$TvfJFM@-%*pUXOdxM6uNoSVN3!%=HxW5L*<^|5?r#pmxh{?0WMdpdDeJ^ZrD z39Q@(J=PRHTJEogUOsarR1{ZNvcA?o zFBf_{X{rK?zT|Sn9dL^2H?J)`lSHu)CmbLmr8gkS$DMua8F^1EE%bj9bua}P& zpSx=o&zx2qcbLEnx#0WAl7nYC7xay@I2-h_YHREvP-}}`iQ>lnE$E{jR+>bXJ`GE6 z182j`^lIe1RgV2kfoAQIi%Vpu2=M)?;Y^)^Roo%V98mRzM{7@^!9q?n71ad ztP8bRBaZtjwS{}a`n)*y<#`)nIM#*z;L&i8a3B|~4g2|?@)L1RV0EZ(VigNkK4-!H}9N*AM% zZB{Ok&s@wwz9fzwXxN87Mm{akU9 zp>bICLrV{0?yy5JPGH%PgQJ((0aXiORk8~_rH#LD}6#b^MFJI3cutqnf+Rug!}R=JvTv2xs7qL@1&Z-iWYF1}0fd&ts< zTr3#*u?w8f!z*?`IlhaGeE)AM%&5gou?=~0iG2E4`OLziSK@+=Wyx6^=KpPMw48W) zkZ-Kt-iLExCL@b32D&F^N;Dszn?$*OStm4$^XHSFi+AV+d?U-^j;s|Y(2IKV{M-j{ zeg+fHE91{#q@lJX#PX@5rxUm#;LgzH;*4wpv{vvl$~Fn;tU%vp#q_GSg*lm(yP(!c(^%GpnR%D+sd2WbpKkne;;izGyO=0G z5tj@8ZWK4pLXLA<bS2PXqeGbMSd_eSBy`OM_+R~qiPSzzTh%;@`!6QMYE+D0%! zV#9r-Bldaujk|%TH$8LZW5tO8k0rLwEL`&8Mz@Pn4VpxpU$Wdi;LdsXV{z5Ch4;=L zoWt4Bw@E&Itl}7O23OpHtP_kCiXQaskWXLw_*(hYplyKXmRRvx^kg4Do0V10VFgxA zG4*_pT7kav#o6|trM7WLR=T7B^~9iY=0=XW@MFOV_zi2cavSDLWch413-}$XMXy9Yc_$bl7pl^93e3jO z4fZN2W?2__tGHkAb>h)Yk&b?e4g7*wJavudVv0QHh~+aIHOyZX*O;|QJTqA3n8)Ee zX=Pb8*;vSJ0KLdJi<=`fw}pE}GhaXt=ZK1|GNq}@La`fOVeAjRmklQd%O*|T6 zdUh(NXD&Y$_%7wYqhxU&YGbRQ&zK_wo?nwt5*&OH_sf#Y#gQ{kvCuR<^c5o_^g@S* z=SmOy;Tvf!8d+vSQ$`UM%KeUOo>4goy(023l6^!W`BvoC&TfO{c)$ zTN1^&;Gd~D`t#Y~P5^u6s^=W&Zc%KDeTeCWhCPgs2;9#Rg8x2L?1b2vFjpg1ZzJCd z+(A|76msy;S|#YEpvGg3yC=2 zPrUb4HD@2xy5aiIq~f4_Rp&GH+4J!jQRK#TRu z=}#N>;rWyVBM$xW_VDZ*GoqtsqgSW2jd!2BBIfM$fVb+C$anNpEpelE9=bN;_wJI- zmGfsTcq@Igd;?f6)YFSO5&_GeP8L`J9jMWVTC4vkrY;u};x-dL%YEVds8g{M&wk8D zJ^Dn5m76Kd6PkqB8eeNx8@xAL4lR*a>FdiHj*rZ4=PGVe4sMjXA3lfF$lVsu*U#bUN*<6__uVhVOBJZ*1VL zEOjjZb6R@x%){KyhVQ;+#q^sAFOJVO=d=Z=Y4`*Bg4T*NHuUFQM)4HsnF*eL%ytVo zVj*_2LP6D)foo(U!utWClzHlE#zLln=II*FYejrya7O3Mct@1`l zT-XOa@b8qkTt2gtzql{>^yVJWu`ibYJ&uD@jyajHrW{BF^t(tiU)D(#Uobn^2jg|8{`Sjp!`R`|9=tV#3tn&Pf;Ah|z<#>ip zFywjGvQ|j!!oPp`ya}HJ@}Nh3{QXGBjDR{LaCYjOWOK#r=@eJn!d=2s>)#*QSQ_fl zfJ8p^)Rt5e!^yXT6PlH$R}5!Ejn~NUQV!n9XC*FxnDaIXzD7D0_YKdpV6RT`xijWW z@lMt#wz4a=E0)jRepYfqRWW-4K37(j?*)3yQ%oJYPVqcLXT$7?Y%D&Oby)6R{zU2e zg$^MRZw0hy;En7NeX=(2@)mr*(%;{yY%Hy96Qtu@e1~<#d97d~gD zzM@XG)L4NY#>&b!0-Uw-SvK@wekY$Z_?-L=+rj)kddEQblFK^5 zw1xMg(TlofA$H0MR`7cKf-kA2QQt2vR}NTf{EFo_o~0+O{)E#rAjiImVDSYkysxnf z@56{ogsS%N8lIsRKJOdvSZqUGEROw&{gQ1`K9@!B1g}vq?vFDj;;1nNh|$oaQ~4%= znRtFIVkcN9UY6x-sDX0={piu9eB%xh#he{$iaMSRdiF{L>X?_lxp*tYLP;@w@SV`1 zm>hlJ{fvsyS;5Fh3-}%}HuPr?ComVf#2KF>Zv|o}SfO9}T$Y~v?C{?+vekk&TD$|X z#S3?t+Yn>vXPh9LNSDZRUgqH*$oabw@9=?gJkwYlu;Ps3L>4|5M<0J)^&s}|lYA$i zXN0A%EpkSyHeG)RokF7*I$z^L4}AL0D|ALYt4pyh>YOWPl0}OqwgFbwFIn~=wz4el z39S=Ks`Jm%iqnEi1Y^V5(9zRkbq;*a&UuW<^6E`ZFu}3SbMPF*Ep_aL;fY$DJTpPBrAis^x-S>WvC5^J6WOJl`R$2&rw*oB!m7qhVv7w&*_7%PbH6bAH#y_k_6%uAl% zDQn_7;T8KimFL+up%L(N7kW>TE=MO+l|$3mi&}EVEu4q{Uxkut*n{QYJu541;$XtP z(T_STITm$BU}iU6v3e16C)E19tnVYv$+xnDr5i4AHfN3RYi0e}WW8VT%+E~os26e< z{!YTPvA^hy{OA>1fjt3p zSXpLwvh?>>F{>oYzPSyvHYui_8Uvj2WQfVEa$?xeGT`d z7%mskCI#?g7jj+dTNQEvEweb~xw8&|&%F^}lFh}(ptwWMhEG5ICV{;M^o5x?EBjF2 zu-?}mSS#>Zr9Kf@HN_qDm8P@kNj`UyDP~z4Y8&^OD@McTksAN~ODvx_ak>}fID=8_ z6vJ)Pr?_3;$KsnCybZI)%3GkA+FX`;_LwQqgM7m>8>kBGkt>(jaE=(OdqPh?LtNZ} zlb^%22`=o7u1P&?pJLghfWK?y(4miI(dNWj_(o_vAA0XX4{;sR8^u1a7;U4UpG`T= z*{G!^E;wp96S+hh|1%v<%i7}ocHvHKp<`}x{jyG=hf_?C0e!OK*pHvXa|G((=gIna zP_{$y4(;LZq9@22ak=`$>cirWc*fQiTwAD36k8`>Xjb0G;+UbPnCD0?5lo+GtoSbR zwt%tW-&Z;nze=o)w!M|g24-&`!-`d`Sv0*MW!`qFyP8cR|HeV+R z&t;0xA)p8BPrq0kmb3Is5C4v27vhKi^Y&W=jg*6}*vUzgW2r`L!aRsCcFTPfa3zP+zbXy#SZUvP$x6%CllCaAqM< zJw1Hvf;F-Z6gq0HV$Nm7aet-4CpH2(Y%JtLB2a^l-cGR*ycI`(Vt-fibAdbJbHuV% ze@F+R&J#gn4ei#P0Gbi(458wKUWO=8R0YopF2=h z9WnFRK4Y#F*dvk7*B2ITEFV4h;N&xt5f=1TAGl6wnVEW!V6|eaP(Yu{`WfZZ0}c5O z^+rQIu@Mp=-p^X(sb!9N((#_6&!uyMMekIvd49VGF?*8-HPt7wU}$@oy+eKIYs6U> zfbY@BDenr-$~qSi%kqqfnajvpfqhx{xj=1WHn>I)qXqB8C00I=Po8s7i{FTA+M}@_ zT3hUCq%(@0z_KoY^EU2-=ZW4!yd`HPmbOVez)yr+umUTw;j=@X*Qt(Y=Jm?^vr2=m zv7EGMa&f>EDD{QEOY|$>=x^kcXU^Ek#V_v9^M&s@)w7%n+C&!30-bvKxn$|pm@Svj z?^DACp0!mCwO!IN%S<5`7YnSC;+5LN_aiYpeXXn!sHJy8tmU7T4ZS@f-?)YS@T+Y> z@8E^TVjgD|3EnD(qc^Yu^?XK~g${up#xDGRK<}TkIIA2!wJfx`!2HxyWjUXZ#Ty}R z*y2p=$@v^sIcj)cjKE%1p;=%bdZFd_73x^c;!28nwuuWf;iKcbjDEQdvEQprI&o+k zHL>#SnY+-J{j7W^_}(U*fqHUAe6D)-U@mXP(}$mRaA4wPGh;;GFEiT^Pl=z~6n*4sQ>iDg2taCTXZ^l5ebh%~utd z$S=tUh&?=t` z2|DHb1uUFj()TE?$u?s4FamvYp{9IEd3wZxD=ablQqP?=3+RE7UlPbsOHD4I1F>Kf zS7phesR|teeW-OV>{pZT^|H)9OQ`h)-=GobMIZb`umX$zd={Nxh1`j>`j3zc++COW zTLkXY->EYy5Nyz48;jmNC+ZtodM4@v;H_fnbNLINseE!yJh@m%1fHW+ouS??oCz&w zrM{#bdo?TO3E(!-)rDusUBvQ@0N<>>4WE0@ZSf3csilrSMnLNn zQ`fLY{kS_8`{quLl`nYfLJxej^uu!2+$xq$#98_DgpcJDCql!=x`KmeAD$PB`@kRG zzVIF5ndUakL>)*3|Cy1UN39czHJ~s2+sbLO)LIwjBAv)*=Ge*>XxuS(2Iq5fS?IYV z_*}jde2>C56??~u{XDui|Mz8U{Iat5_8Ie;P-9>2TU;(sgWW|r{8)MZPS4CvycL(o z#z1{Ll!s&fT$~Zml4B(|~jTIhh?h zQK$-5Xu?WIToR~7rcN*U5MCeNJiLgSeP6vEk?2 z6zQ02mXIsYdFg8fZx#DH(B8~QPnMP6Ay7}xE?LeNw4*N70KJh;w*q1@9|wm18D<*W~1LeqTQ>%vw|2ELg#)hmmzcRcPF=KUKW2`P9+V&nlZMrjMUN^D^Jy_JwDn7qyAi2gOW7SZX%V$6OI^}&I`H6Uzw@!|<__2Kd4Aq~-e)J%3T*x)6 zex|_pP?z98pKak?VJ0U%HM#UwppJcGS>GGZ>OZPsMtaXvp7T>@1->WI<0rDzb-*ch z0$TbS`K+`r==}Xi*QI<(ur}ze3wO}iGnR(>TpI3%bL5JLsn1OU=Sa{9^xz#k2VNTV ztXP(r$TyygbH!Q3Q^VZ!_5Gxyrb(bLdm3?0;10<33yBM#G3rW+xeM;wpIKUZp&QX2 z-sR@DFav$jGY=ZhPYym7#~!&XJl{u+XUoj=Pn2^4n&!UZ_a^z&CjvSaIX{mqv$6-v zy09<%3{cJ}CQofmI03zENth$Jwy;+&YZY@&C)^@T5jxvK9?N;0dNIp3>O+rap-DNv zua#E5*U4uF_GZOa&~q@dEc`^ZH6a%0MO{yuIDxnWPP%!r)WP%F?}V2&QJjcpRpt8` z#P`dOg$|)jH9mIX3_OP}Ay(}BYu~DBx`Zadpi`bjKfd?ayU~+*xv$)X z_t(l#$8a#W_i<}kc&*E<28kYGRI=Bn_Bodgvq#B;rOo1740WCH7Xz0to3FJZoryjX2 ze*cip$Oo=KB5MUWw48-|&*9`Z^(}li*~_veXMv_01bX~(F^TNdKHL^Jok^T z))w}(vQFB5SoEboC6*o$*N(uW|->-QJ$3M8^M zX-YyOjZtjmTUq9|ZQ;MmN#yf6GVS5B#N6blr+0_)hP)6LmL9%_ocx)x=!f@(|DIB; z-<$WKM{vsRuUu1Gn3bOui4FU76dbd~;!29?1-PSFycN%4Ru=C7X8?TX9ie{wophr7 z+=e;)-?3cShxw)m^yYNLDC!e*cc=Cx2y>nUS;SBWc zS3SM_l10b4It6-jkK8X>bj(YB8}azu9W~@;2_0?9Yd`AwH&ghD!b8Z5$6PQyi43&hb;5t;*5aSFIj4gVCDN|!yHZ=cT|&g z;-(O5#sTWb9UG?_@$|#zj@XBLY!cv|z^sW73*@K)e( zoPcix?wx%S7h?Sx-eVk=4YTAf)Yw8xZ#ecT$>-gnFM7`_M^By!bFojXoDoNC<-5YC z2CXF)JTJ>0vEXCvQ=^A?^mD}fSUk(w03Rr~VPE(eQj^b&zD~(RJl_@cV2&ny7p#@_ zx`Im-b1r`d^_VBV@vNAwDxh%!IlkLsYkY4jn-o6#qDPYm-lvbyC4Gm`EK~&}EtXmq zT5H3*=VWt%d1I_#1@^E4{hVTQiNNgiTfV%H={$eM-tT75T>Vf@6O`Ulu)i z|M{GgpC!;Ucg0?9(r8u|bD-nSsYiot>_e}i?Z*G^kMDaZZ|SGl$&U*#6Ln~e4SP&c z%sb{3d(H)YsV&ULT-;HjJm;sM@2R<%ojK7`@1F%$IU`*na87EiY^<1tmd~Y;g`-a{ zYX#p!_0%}U1Nw~DaUP?%LzdqQxD$RyMn|t06#FtW{SsMbZp?3_Evc6K^1nkX?o{4? zFUqH;@!YNAM8I-B=C294VrrPBNj8y<3!l4$V}G7$Y(uVDKJ`v<0@s++h4*~JcAZmT zk@Itm3+LzioaNMG|Gtp7&QpGGw8S0COCaOBODdHBQj-&&|pa ze2w;Ce`a?=A~2WlVM9Il1K5vzF2sVvsz1JIK(61C#p9R6 z*RTS6B?8ZUR$BmZRdKUGZ(?cz&o@`h&odT&VvD(?jTKwLiK|M(b7!%abKYIh#}F5J za*cCY8+zp8nA@p#W>L>Oz!|J^O|r2pKC=?%*7%}6XkRHh`De%-=)dNO;Hv{*hrvGn*gaY;ifOaBIv z!k;G{?~)Z7{rN04wHs{`?g<-9?}S)B)(U?8#xgf&HUhn@3%W#>m^(?N$HzAA$B`#i zJ})%6VkYO(N7B-p zJ1!?cp4WxZmC@8or;n3-)Ah-$OOAay&n9PG9j|ES@>7V3cS1^I)x@eD1>g z0Y8O)vPK-Mvxw0+ArZ)9JCw6<+HW3E>};V?-;%h7)@Y$^#LUgl^r|d;le8>r0~XAn zm_GFFkgZA2`C`RJK-Vb_9t}BcB9Lbvz9Vuctt@xfERc^uoAJxa_w#9Pw8S0K8Tm#& z^Euf>Acxi}&%Pyrd-Tve)h=~P!zL%N>bjB`xN2p;>`a1FajIrXZ{6z3M<#TZz zDiG|XWX8(I1=Bj@z7EYFXhMf2f)v;1d4obMyc z9I*|cpT*ainTf?dP4Wk*7w@7|%x4s!sR``8pr3pLv~nAsvA;{jL)C*>=d^{M)Kk|Z z%oEUcwMi%)zqbq%n8n!8hrM$7{+?XmCyHZn4r>d&;#mBAaYj8j137BZGlLQ4$Ojz_ zuCV^I)E7Q$)Fy(ju{v{GcosETqnP;`&y1KgQMqQJbNa&ne!Ek#6JmjLus>&?f-jWX zLQRvbv+5~srD6tVW}ifStUAynjy=#fEWYofTw`81dZCYtdiH0JyZ|%MGZ)yy3Rva^ z#)X`d#itiq=59P6|BMx*iHn>RUd#M5iyHP$$ZK`UN6Y+1u<|=);Y!LmS^9d|aIb!5 z@ks${=$XsLg0aD(GcKINpGCftj^DW~IW20f_A^${^1PGIhCLJ623IZSNtDk8e+SB= z1@!XorFA_^fjFhCWt}k)AtG` zzU*ZL@|=O3zkgZw<1A=9g?R!x^iIf?^K;9$g0X@$_*OnMIyiy2CcWpKI4ykU_qD2{ zwxNw(v4=EC0qR-IXPjUc-akF~UZRdOQbRw^WE8{4;{4g~1PtH{6q zwKmNu(2pEiTUaB@8M*}KqbI!+7iRPOYF@vuY%ZS9IyL?rZN~SF)u%%oJ*a1|TwJrP zUsxNS2O1+daZVhw#fnQpV;?LtP>%-R2v*1i=0t;aih+e27xb|W`9zi)LoBem)RVdn z^e%vUR-+%j&o%nV(u3#0zE-{y=t=)X!1@|u8{T=I4`*fNHt9WqA9TP1ztSpfW9B}G|o;s(PyE6jw@LqKaBicipC~pM% zl*AKT!E@>ZM{n*e5n{m!B>^3KQPX&?u~n^rWgiebX%u_CEX(U~ly^>_fAwQdbWSkx z`Cg-jJj>T>KXMD|Z4oN33gVjtwCs(ai|0A8e9kHVOxUklXxP)k z%4#Lfn+3Db+s~ys7N5n~75hL~!G*rma`dswT_)P8^nn9<32_`AyQX$OFD>(DQzgFDdr=KJnt5EAC%8 zbexm>xkWjnnDfW7%z&1=?GWhEDQoam!+hxJ8TW;r>{U{nxDZp1)=I~m^d+7r4IJ>j z73WSlfjw%{`WpFuKk10cGbd+@6*tSW{}jO}#~!)(T(K4KnJJNN?C*Z#8!gTri%VSi zoyp2K0)H1OsYfD|`oy`QmFH{NgIXt^Ir;@7SjFs_Tjh0DR<4*{u?^=)WR2hiID8g9 z^?>E?&Ux3LJ(2GUplSFu^=5`#Tp}Y*w@I$g2h@B zzCYwALZeqBZ^Y4;eeg^2=Lr_x{swaWvU(=U=dS3FRZJZwuS)K(n(Huou( zE^$HY_Yg&FYfnJFfL5vlwd{%4&5~tqU$Ml6~$aC}>@rg9Kz+&#Iz+6={ zHuNVSJ3%=#ymR!3#nXqgk#j|C#L*`a&?n;2GOJUrahDcOdl3VAB?8OyR*rzKx#&fG zL&F@zti*b=@CljTz4*@R5XiBYlg$Ns#xP{dXiP|G}C zCru)befU{3N9a(lahJJr=o(t@OT&IuSr*#F$`SGc_@0vn+ohh)8L#2l#zGfXwd~y~ zP|N*v2rSO)?^&@GYSRO6ZI~I(2-F$nVke6)+oYJEEu0l_ABlM8sJ8v)UbH{^P(vSl ztSu1BR@J(m%osEN)WeTC_Vgn+KVtl8<2P?D zZ#iPCmJuViYcHDv{(sliuYFT~ddecvhgS|N|8(}s(TUkogFZCvC8*<`7G+X<;u#CV-Jm<7+bFFGVF%RcGt9L?wp;{ z-FE&Z-D1)n-S5vbbu;kj==Mdg$(m06bNbOn4@7_eWJK0yhGeh&ctkeks>9NkpLi*q z{@88l#pe2++Kfwkc3bV;=<&CY$UeM%k;;T+R;nEQ&-z$@;aME9U4*F%0o~|8tkKSJN z^7PjB&(aaQou9sV!7i<@{51UuEzWuIx&r2Jm z=YFzR{eU;dMs5E&JbU3oTbX}Co*w$Ut?L^u{(SW8sjkwo!VTF{d*7F>opfgxuKH5; zLiB3ZvsRofGGXbFYdmy+&*1gOP5XGW6)PuICP%xEUbS+@gzxLamftE`X6_y3{SQCA z>~6A^jc$9k{PIR0lppBar`63|w)K$@mWkef`snh^39GlBJYaCP$rfLvOV2w#JL7~G z)7Sp7MfA@#TGP|}2Sh_Y`!d>N$mf+IzYd%>{nb6Tyls&;N4_%d*=Vu#YUPzzd^4JS z@|^5}sgtwIFTO3?=%+0!%Pw|J<@V>Ft_*x?c;$xAuZRY(vupXwO|EKPDog5vhL5dp z{q(-+ij$_LTYOl~K7Vgqw#Dc#TX(wm*!p1y|F6FNEhk1_FTZr9wLT!5aA8Y&-_+~N z59YThYzjXJoVS|#anz6RgYhye8951 zMf*JWaMbni{^`lDzti&dYXd7&X3R>rKXqvK(w1%6x?5eG?R45D>DA9)RykwX%9Xb& zBPu_Ss${RcIko=pMIPw+;Nz*?LvLHVd$*mho0cy9OWM5W!1@wnH&5R`(5ACTC(%CV ztXEm*rwQpb6XuqGm~d0{#q9U8OU}L^`?BW^R8pT5@gSvqv7ZAKpb_xrc_ zscGP}=bBIK`Et!0>#ME&de0AL*RuKa;Pl*1x6VoPo4+qlo_KV$-rO-g zS3Gy@s1-k6qwcvWk{7Sl`_kU? z(@W2tlwNw>`sGQV_hqXr{zkUy-q&Os41RgkUVr+z=cbzu>ptq*i+hgyX~Xn^t=qEF z9qVTQZ2o(C+UZZ!=bUnDy6p|?WQ)HxJA3PtyRzY@uGD+VpEv9M{Dt?@gZ}r=G`;+T z^vpMs`ip<5Hw7;%CX5^-a zTs`vi>ZA2BZn~Ns;u(JW|hZ3dLo_h>Ymw` z>u;Q0zrv>FGd8}byw=zIq?1Oil6`u_UD=Z759saN=h)UsXaBS3mph;DnS11*bpBf_ zrz6HZ6K!$+p!(u_Z4+HJ^s;p7c~jD!m$ps!pU|EyKX0pO##*=34}0LIp8V^tvx~N-v4_IpevO)pq>-w2xQ%XxbI`ToSEv z!8g$+2i#x2?eS&OF zstm4weD1h(sh74N^Wm~%zUUach{!TcfK>d(tYxam5nZ+P&s7X3(Ld)v0V3?9p}}T zNES&~U;Wl}mtU4CA3k<$H2K>o8h^--NcTKC8rsV@~c)$BE+w@Fp zzvYB1d;0WjkMnNqzGKvo?5D5#)6=Icm+s`wD^EG*_WB=>Ov-0IIj}tP_%Zb#X6>F% z{p6bNPd@soKI4HKd)_>JNXr(_EYq^$21iZ%-Qux*ONWJ{lVF-=Y3Tv zU-4#Tky|HKo@%A_e^J;c47MID*s7ezUF}PBVSzJGi>ab)*Day zC5nak$jggnk6c$uZ#uBE{_}2MbWd9E(v~0I8(TU1=qoEPPF{8N)bZDpKRfT-=!5~A zX7AqKQ`upIIh9p!T5I&w3F~A-kC>aDJbc&ek82$rz4grabosVD(=~rxY4nWezpb45 z*6`7nwBA*p_vR|)&)w6}`YwrDT2D)#{o|t14!t|1&7=QazxHoew;c6NrT#;FNNcv? zyFCYwy0ZM(8arhZ?)^Gl$vz%^S^c~`cCXLshu2mg{qo`ajlS!Haii~lewpm5Rrbmz zEH^Y;VUw1Y$z2COerwD1Tl8P}dHVY;K1y$TwzK|D%b@xhmu{II`1G{w;#1a4_t^K@ z`r9iWRKMxiEwfuceWCvDfj^h+flF2KD&wJ$#}j~Ksb>jrsOw02_B4c5HA`{*YZ&o<4^ zj(U!o7%g^u7HzuP25J1?N_pgpi`O4|_LlOR@9&rG|Brh1O*M;d`{wDiZ|EJ_k=M^I z|MSVC(s_q25*_u)m+2mV`Kmr?yQw|Lwtd?2>Vzvs9e>1bj+t=$XBullS>kv|Ps)m4DyyNc7AJTSou=X?S_=&rd`jUb=j`$VR(mGj^F2y}k2O zmFFM7V(Kdwj!B#Uc2M-M-L8o4+~~6UoPl>_OJ6s(vdK3;Rc^m=SY_XXwypngU^&`& z@#bj2?z1Z04}VnI{y)2ozUAwwE#rGPX<29M^;^aceIp$)A<9}_T&L$BA3Pr2mG)Gw zd;W>a9!K1tjX3_mY|T&l)BCO;&^qbvM^v6&eACKJPfd;9cMn8=xuh-I zdEl>UHulNrwEtGK3p&osF8-U%?tEo>dd0j;)5Ct;AbV~7m9rz)**H4-kyq23qW);{ z-50BD^Ywm}>+boqe#}!pq#vAjL;a?+e@MSFE2iJi>Wr3{acO1Iw|j(Ym-wXe z@BSg9S3Brmm5G1+b9w&01Iqb}r}n)1QGd&GpFcM0?p1FZapZd!S5_G{u=2+TmaCjS z;-t!ef0Zg99du_U8+JhYuh&dv`SEvF&PX??ygKF&m8SErscg0E9s3vzCZf7-O=#KV2ib=m8h4+u)pCKWtO3Y%q4e%IzQgwX*J@2cl=oM@08M zyL$SUeY(<}4lJel8uvyAx4#t4byr1~-Ed3!)VuoXmp%4)bmx(iTbH=?@Sal+J}0`Q zy?Ji(cyaw*GMWws*!xA78d{w9LeE>s#%cm0vpb;qrh@-$`$HdvN{7BacegUF({b z3tzvvzU*H|wVZa+Y4w-lr}R6qOSc_0EBolI)w6>K@0EV=_3&ul$L>wn*yq5OhjzWU zeETJbMDM(~b#&G%o7Z>T>u~6mTmR&kZ8Zl{@cCpDLt)gj(@9X@AJ;?zHH^^>*J1lq;l@ijiS9Do1JzX z(;am`cXH*Wr#sUJZ`&gJ@w3aaL&x8dt@ZK+*{9=QN$>mHndt`qxTgCL4|lhma{s}t z`=5MH`r+e$k8WvcN{1izMf$%v=cL&)`$QYh9#FpE&X)As&rgm9J%4w!ZF^UJ-_GT- z^zJ31%U7sI1E;k`m3^1*y6&%QZob{>7j5_}l=Z9`tzUg;w zMHdb~tK583OL^8Nae3C?^Z>H(#C|_wA%;${ANg@uTZk zhO`ZcHroH|boiFG{M5U@&wiXgG~IaA*3sRwm+F1vjJ12$KVsG1+b1rU9(d?e0 z({sobJ>?Mtu5Vd6TcUE;^QHPVZ&agoPX4fULan8|{-j~)vE?VD*+=h?{_CD)(vObY zCcS;r!Ic9KdN#ePd#UJ{LobN_a({pMvx5$ZcAA|;GY;IazQ>|FR=(e~HU0hfvs*9x z^1<%v@MT+0d35=f<9=-EPB;J4lY`D0GHS*>^QX&k%2bLYL7etO-V=|jK&eYD^Chh*J% zy^_i;+0V6Xx%BBRca48MJ?WHr=?PcG>2vFs({ooFT0ZN5 zH_N9U`k(CLvls2{-FTJWEB9V1{ol>~<;tl3@^!G_jqWKW&5b@s{RrRwJ{@$d4$CEF{TP{8#sn zo{0l@?cV&sRni;(`%KTq+a8pDdD(X9E?Zwz`DKM0D|^0lTJIy*jqQCf8`*or;4RA8 zi`SPgoqACD^K6gSPxrcI9|VwNjp{I*MExc*&&ZsJ!zR~cQy|gb>6hEM%=#q zwP|Vb!QBTu_ImxIuO5qzUA-%sfBtgm^#|M(9sSO%p7|FY620)s%F}jQchB;+dv7ys z;4%+q6CS=GTdd{AY|O)N)?1FNr5lbOoE>}ddG(9ed@}v^!MD=q7k?-H-P!A9dw(`C zJMuIA`@{I-vTr)q&CcwMTW;Iv*w$a?Jkh%TRohp_?6iDkxd+azuFx}Nn@773 z>Ug1L^0cAp)Q!%MCcZi&opJJ6*`}L(9$k6NxM+nt2S#1TpB$~U=cv}o)TOf(4%sq0 z`IF1Dbw6CM<+=0M?s@;qUwdxb@x%1%6IRJ4?s0nd#+oblezO14y&HY-aW?h7C(93> ze0I+&2d*9M)_zyzwHm+z}am4BSDaplX&YgTq2abmRozK^ypy5rR7m0yRZ zXP&fa>(+;FA3Z)|a?gr$_iwp=zl+j4KOUI9bHJ|I)GH@u=X^Av<&&o`?_O^IQ@S@; z{-Nm9caBO2F26-KWBFU6|J}S!G-6CodSB(u?BxNwWWOAGS2|<0jVfKk{@An4JF8Z< z=o{IZHviPJ@g_^u*EwQ^%12Errh}U1MazBgR#sW+@7Xy^o|&!sR21F$#W&@@j9W3< zdg_NQXO^nb_@g(ee4Gr+#{A1=OHI2l`)Rcs(_`0vv}a(~y6FX%$PW@9Ta-*Z7_ zjc<0ZJh#oe(bFrhSfB9zJ84sGgQ)&#f7CO&KV5bWQ|Z28c;&i0PhVc?zI54pZ|J$^ z&dajtE!FIz{hPA)&RVW@iSp`)u{>rVY={Zv6V9$_`glE1O04MXSxaEyKOZT-vv4`GA+(>X)a3qj>aH zt=|nfGTqU=U;o2p7u7E--`4uW-3LbHqxz!F{yizOhs})sbi@+Tc6+W|see7D(zE`< z<(+Okp|Z;(D^>1Ud!&)#3{rEJXR zcV{<^_%_;ZnZK4t-oAFvz+=uxZ##Ii?3U-gO251GlB{d#)3bBRCu9e%yK=Vm;0NmG zo!^;OcfBk-vv0j@+}>}eV<(=N&3O2{Y}Y&9&nB&1iMF2cQuNI8mz2-C?AyYQu{q}|sK_b<5jo^#&s_v@S`{Pl=fvAz5h_Pn34ZMzvWPAi1; zI|R?z;sv`d4Mz{X7K=lhcO$8c^Fn&lm;N@;VcH3$*f)43_P5kW#-p#}Mvx;RFW2Je zC6Q#BM;aMkdYhCs*|4DJN$mRmQ>?Wph0T!v!!(rzHM{1W70iAn#rc|MiQl$7#_t-d zvC6}5=)Hw9Zft$P-5=(T#o?vk8YzYIT62~_Xd zSfXchnL0PM@lzW=RCcGE;>OV9sC*N_@)uNypjU=q79-xdPfGOl$Z%f$yb7r`d4jsG zCLx(SpRmEom5_JVmPPD(ju&NI#ohItaMvOfPrvqn@2vdBMeNQ)N*4R^=#MXXGG{vq z`|gAP`x1dWC&}YS{XR%S_cr2;mociB43no4HZsHsI&mH?iyaNiKK`MmTUzj~uYYji zopzyH@Cf|b+Z$|GhvHeE-LT@oAYIX;4O_g=(1xFLNGjVXm=7-qu>;zmlDj$*$#L?y1 z575onD_p9zDm?gk4j0T3!6VO)+^@1M*loTGYn|{Rm9L!{XK)j1?+GORbQV-YF7RY@+~X(tm5Hbb4{G9t4{nv`^MSYc^1NdYmTrOD{* zG#k29VKhz~If7qjW{C`yli;|<58ErY_r*&l^a;*O&A@Y>s-Wqc6!DfwH*zdo)!r=a zD_K={1ihG)iC=Md@nQ2VY$UykZY?ZiuimxNx{-fT0q!wtpsv+m(o&|04FyA_I%tX0q`Qe^Q$bCV1oXd0065E_o!cWWV6aB=S$L zldN@;CnJ4)$Q6f$Nm6?P zPT{XGYyK^}Fwo4Bx2fB&o}*?zRZi7@b7P5E>B(`CS3nZ_DO1mlvK@^l&<1`@n54ye0E??Z9mc@~mFwC5+mfib8s`QS^^X@O%7MVqBzUZ~vf+n3}A@`oprorg#jq zxX~y$Gs{}=ZP94Esq?hhgl~?V$!2e8u14h4Kq;R4PK9h4^0+RjwPWjhrFizZ1iB+g zj^s+3keA?&{6a#(?A{M*LAA+LMH{lyMV1_75uAk<;4*7(GWUfFVun$$qV*&4*N??l zrWE4f(=B*w&J`4!?@W<~3VPdF2~U+)!}Qz=$foYZ_1G@5=(Rp6;CuK#jn?o`ya;rQ zg?L@%Z2WAcJ)Z0sLN2Y_isMS{aZ8de;U7DZ`j>&^vOxhpq?Cj=59$%^cZ(p)`ZAIq zcbN`Gckg zFa{=Ew1`-F95iETK&OEvk7#$3+e@(S4C5{NsZP4UQv+Z!{scc&n7`?Z4o zm9TtL{%I_WSzZk~vm;^n7$wM#S&A>#FT{J@e!{&q=jgp#v0$F*#I9|-2%=Ap?A~!D z2n%^bQyWsjZKoC+{;5AQz@jr!px~Wq~w5j&hP)`lRAh92JHm^W>2{Up2$|4+?R)F248M|(G ziMft@#A=T}qF3YN;K}YGZQpYn`$|X94>ez~_v1lPph=spX?p;Fz;QUC;Wp&7b^|(b z?F4weGGwNa)@+k}5^mf(AEob3;5Ho#X73Wluupfs!5otzZ)CeQ1Ss!>f^E)%f}1-9 zC0fG;s#Uj;fo44Zy0!zq-gFK+pPj+Y+HquecQAw(S~J)EC)jIzk>>)3^LC%Eh}?2? z$?w4}RPSmtAA5Z(`MN)h+@I@APR-RLhu9P%ec%CJUJ{6j$}w(4$vc{oRfbQWY{i48 zH$vT*MR>~YEUf!Zf!w%w1eP3a!+&?U;G|vqsL3&RNc%Gmj!ZCzB_2Zd$#pFDb?c=e zvH$pn965aEtqa+*K}eDo7}2g%MtE$wB-t7B1K*MNWxwkqAzD3w2K?EMbFISZV2J~o zUwRrXntK@g-4DRe51gw`(pD7ANNp4LY^i4*9p=nqt2}eqtHt!@1Txcajx{mdSbRWI z2fACAplD}3CO1Eg@oTG?;nWzAKG(zfZMT4)rVf0gVKGx5pU(s`-^go~eDZOB7L{1N z7i6UxsrJVKc=0)bm7a28_wSsidB2x|$vrpmi-B4ySgehf4wi9_Q}f_LTPxhS{hQBO zTL_P;r{bp_@$hx3Kl}Vif!UqZ1wZvz_RrIcjY`|doI^)3iA+0Y5m5oEAC|GB+m3R_ zl-Bcu?d9}mcN9Kgl+Ryni^b`$1_gNTOzw_xmiP>g#@$;Z= z%f^xT<>@vIU zy_5v$a$ijwU`eS7dzlsiM<%9#i?b8UksZdGP&kAy|47}6Ch_)5{rQs$=f$Ukm(jA` zQ^N5oro2i>iS6C@m&Gql)opdQNw5+(B~~$fIm=G=ViNxx*@Fp(*@s;nbRhm6ZQP?Q zzCO1fHr(%mAo_`ZulZ9w@!_RP-7D36PL~n(lKRZ;d^e0cp8b^Rk2A0N(=)zibLnE# zxA-R7x(MMPFSfE6Jzu8o{{;qQcM5usG};|JIM+7I`6H@bV*=i%G3>u73$FtTX|PfS z7##kK9#6anWdlYmswI{??|lK1+ZIBWayQgZrq`xf#|)X7C!l9j%e%& zN1L3$1$hFjI)1(@8t;;jMxHh0yyczI!V%haw8STk z6%0HDho;-meo1{OCw|P?&iX~i+UCQEa$m65xXne8!%VsV5F7I+lR>Es+x*Xu$se|5 zF430kXYmdAs}aqX*9}KJF~EzZLr{xBqHuJG5uXy8O;yZaiN}xF!dHJx;&ii(@gOIG zcWPgzP9w76_Rc?|#t&iizs>Qmq#%jzJ5FHqgQ1=`QIn?i-{5Fdm)*t>es(ct#&&tG z#zgG(3I`=_MLv^mp&n_BquNxFgI^e3W$BIHYQ_osN_k|r?3w6xTOrcv--+i>9?J`& z{IF@)iJ_9?jn8jX!WL?_mzzf7n#5f3&%IVK!K_0(+shDt zdl!On0N_`}R&1rr0=Cx8lI@zTh!>|n;D%dlhRT`RFzS{!w(Z@)op|2^`IC1;?5?TU z!=wj4Xu&vk+ZY^UFpO!Z9c1~ji7aYIJv-RQmeA)gVTF?b;Di! z%92LmQ>{>;%z9^0vQ101c&~_h7aydPhWMM`BU0ghZ9JR#^fpTl$mTm*U(oUo+xe*R zSD;Yh988LwgsbCavF^E#SU+$wzH-qJgvnQgt}T~o#0?&QO`eCo?3_qer+kCye`8qv zEIn3qqKmz}uUwNl{3&Z+`dvJ?%$ytid6g>NbKxLDj^;KOP{*`ToIJ}6FIG$D74M{C zE6b_)!1*6=H$Mc+Udp4N?&QLR`+s5LZgp7vZ#RaYzv<00XTj~^Cluv#8J8rs@s~Ct z95U)J*XiJf<5EmO%T!C)^FW7t*AORMof(O$bDoR4da&??;Tc{Wn8F`hT>@?2dc_xE z3Me!@gst|g;Ji{Q6iH05A|OcM!C8;=)SiFCXZ==4aFMx^`#-E>`OO3eZGzE`z-{M z1BW??Gr~~^7UR7)p3^CNAEJn-)m)!LIOnOkOL%;?2JQ;^0%srJ<^QJ1K33znOeEw*fP7*U1w39 zPYR|9t&o_Q!*+d(f%`4K)YGa5+Dx68X3GR}s|)idG_zN|1FcZ-Qxt?YsRzPT8u_j4C&s4L8=-ufZtf%#Su%5 zU|nA#cKOuIZ7rB7{`zAJq|cg-ky|RvH5jD6;Dp!yQim1qX5zuYemwQ$6I^)n9*loF zgZ;g(3_fi``p0ZMdR0CO9-5z}chc8jTRj)7IwO%9d!FM|XB@&c!{daiMRj~w;1u+s zva_lzB-9S}B-_bt`iCQL6H@D}PI?We!<38BY;SlZ%eM0YgEL0dzAIO#kdeavQ)wQ` zuXnOqhuNrUdxzN3@f$a7#!mRza~^I-s!|H9rAY=4V zdcUL#Ybos|0;NZSibf4CsFaG8%g)irf|t~7QylBte;t4D&f%*E=Rtd546Hxw#M)+B zv%;raQPHG@$n%W`UNl0Hy=*-MKmRl!H_=X!@y9~g8?FH^t7DL1pCu|jBY>5)&*;@} zMMLjNm3=*=OeH{`x;(a|eG&s8+OZJ)(jGzNa!I=R;%IDSuEbozieU6lJJ|GOKmKpP z6FwePVEf~|aoau})H1~sYu43clWESdaQZlQ)a?mZR(~2NS?$7ik9$Dm&}?Hh`2!qn ztcR>~ZoJoI5$qKs&wTXJ_8zgGaW}<$m|M4S(OD9&!RbKWu`>&J3ZS zP%Fyp`G5nCCga}7P zXt|9o?=^ZN8YY!PA4m4$aZ^7+tQ{nEi7<%sgIJ~Bm z;#RYeDg%kvg7npDeA~x^G|_%O-hXQ}t2iKot~NY^LiPtjJtOH{Ckwn?T1fw0GDXGu zHPBj|gqwnr@VeL{8s#2?Vvi_uD~~M^8IOzMG@Yb4)yztf+I($sVoX05hGI~SwHL2& zREOWYq)2RYHdmMsZG^k!#$v;7yHR}p2%2ziKDH^a!z~BagYO7`7@w6f)CFuIF>VH0 zQ#hRLTaD4aRbp;{+T&}NYG`>%HEwwE1Mjt5M(cbpb4I%3QR4CnJn=^z_1qUtBR5xb zSLdEXvTNj_Y2Pt)XJwC=HNV7p<{yyCLJ_szzmM|yR(N|%42(G}4I+g&kc?KszWc9Y zB?S-A)Ot$4`2FOK^;Bp?u`c0GKO$3IM-l0Lhzwp;BEQZVu<-J!Y$wV@K{qADr#2t9 ztI1!8JcQ}A#Z`f>zIKGaH^vWLot0HRNhv{`ApC_Z15&u#{WY9_Kn=_o?hpI@MvzZ$ z!+HG=*Tf&1KZ@@ZjzBk8sfpZQzQCH#k5#U@cH<(}J}WAG_lZue&Ouv4w_;>6lNm(J zXXCHQGV=#TFmb-(P);+C`R)s)7hYC!!@4%ol2fU|Z6m7rY2t0w>8%;?USc`?Hh>}b zaXO%%$><1{mS(iF5at^;*l(ZC^#|8XIU3|ZPcH8yi^6Q|jt%eFpmg!s{~s#_~2+c-=q zvpd{B8NFZs2;~@rqxGtzK|CKZ&%!<^OL)vTWy+PNf^BTtEzW({Je9s$jRyYO|Uc{p74ra9Sz(ahu zuSt|NR$dSr^T}>Y@>hY@lnlFDQ-s2>T5A-yX(rl|xr?eVxJ#W+sG>=0?~7~mYU!GF zL%!q?tCP5>fIoB|q1E$b;nkd_RLXP=1h=o^e}tq8UrMdw4?bClevZzd4dhf$$M18)6x49ZqEu9LT}o(a znT4`qPeVpgo2dI>JKFKBM%3tW0p4Xr@h1P7f?dD@@dw91@uq5Jfz$db$V<{^XaAmo z16>t3^R)+-S~VM=i;PAUv&NxsE^F!X2@9}zW+pnRDT~!~9ni*6PpJF+v3O4HC3q(< z3GqGSSO&7@BPSn5?{7qc+TFM8q(TIndgwB6zm%BNqv1l`3Kh0hGl{KL8-QDHQ`x=V zG-!=;rHMv2_zF+N(HJvmSSVO`QzpYNOEx z`x*vlI*He|eqnw7Uj}BBhzR)JVu1SAd@4|>gW|R4 z-`QdwO(Pi~?MGiRXGFnRXGzYdb@ zjf7JymT_O@4fwtZjWnQP1VlPZuqQ|2@MMLjoLgcBxu3iPrb)e}N2mF-iSLd>p36-f zD`iL5kISWni5b*=Tp`5F{RQ6pegZdh4*K*u4=!(6Kr?q*;!ipj%;R+=JMhSzSrzv~ z_|40Z92^8~Is4G$(q2d~n$7fT%3$NQQ&8Vm0iCOS@XEsqMDjr+o*_LNXhI*3-T0Yq z?A(V7$DPEIjtN7owN_0{z)`w;rvrX*+P|jevtCW-6q%Z_?=5R0Lal4s4==CziesT^ z(;C>QdJIn)e8p{#ROF}D&l38D?#D4PGFWGcA?z;fWlm0GYR=74t+{!1Ir|m(8EFgi z!0BKHp1<@v67Y+$uI3c7S!O3Ltg#>A30_d3Z~iUpt=UcWL0q;7U6JxlOc(14g!5lHyF^3M%dyeuBk7NZqh7q0A zuXu3fO`K*QOVVYQ;XJil)RhE)!ni0Lx$_w6E$;_u`#-dWjw78v^yv4gA#_sTW!#~> z6wB)lqCE*xM9O0ZKJ+LBOnj%JqN$DIv4<0pQ(3(@y-Edj7q{5;^XKMekW*UHp80OzSd>1)J-39 zmLJE<>Ow(c%zM!2cjb<%8`bD;QLjnbqh5p2D?DH%#uELOw8-`@39wKk#v{*=f056) zKV$T{Z&`o2%7oQ)^A=Y!`=J!BH7R7QEsxFKZq1Yx1gv3#Ap|W^VuS0(z?qsd_F|Me z+c3Z z(FG^1a!2?G5IvX__#j;`|l3ORcZi>O$Kr0iaL5@%Vl0CI~l%i znh14YcY{i1GFHetg{^OxVC5Z8v4Ybm;$xh^2Cpw++eSyjiauYgXuX1}$9upI&WYWh z{u>Nk<2a{TA#}1&0(IZ}m`<(_=Y!UD@}_q&8~f0b=xsX!_5qz>yyq0%eDoyVemWZ+ zJ)un6Lno8z7vJNfn^f_xQy#pV;0{i3-ODdFPa5)Vx1-0YMWU%?*Lk<^FGPoT_>hlP z*KuvbNHRKO1YEv$0SdQe3E9JD@v2MFe7NQXe#4t|R=4X2vmEBb!lTu=HszmuY5RHn z&iyFOQ#KHvl{P>}zSh80BP%j*i57Y{l=*9^oX24X(Zp8`kwu3#@|w#1(8a8(rS+<- zs!y76CAV{^z1|NrcmJ579x)b~5hXnLmp_ULHYBfmIlOgUDVVQ3M>E2_VAV7qddQ&y zDm7<-W-jrqumj<&`VvYm2rwo>vqQ4i-SRSO*0y=qEGVsbHMY29=y82!x7ge z0dFo%rN4YYBTa9^%8O%g!Z%~;p26eYZ%bjgMgVnvJQ1BqU5H|Hd-&804xNpBjaOQn zH6rbuk+>bB5~$Ltcn*B);)jiTh7@ zOZY8(p76^|Om=yOk)yBt$Y8xU{kK93_RV|6PyL(^_3I*GLAn&{zg|qe`hU{h`Zw_X zfm6uRigISL*|bW2vOxXKpj`+Q+I{oz$8Uzqu-V66;{p90xH9fCoVhcGJ&}^a(EFIb zcrFufOg<@yI(&swG<8L}m;SJXmOl1pmQGDZff*}vm50XSOW3V>G_(jdVEsw$ux4#I z)0AJz3O{%;*NMKc+2k<0Ab7)8_#Wp6cP!#czGlP75+SZxtw#da|KR)Um%wf{TlN#4 zi}R};>8J`jIytm}9}ElU6V-6yWoS~ZWVG1*xFrQvI<3t-Ws&IUI3R}3VmWAe>2Cbw3(Fn;Z9Qi07u4t(+$?`1b zq;!_e`sxB>7C&Rx_f4y*)mw_j?}$VCKbp|BXAjA}5;3`}Wn`aS-pdcD+(cw-Mb+ui z$5_OWw(1=07BjQ|*g%ZBCfj<&-(H&GjYcTGdgeejb^qJ%Y`CuEQ2D&|q?Hd&IpD3`E&c`^7zT zr?G0jpWSy&V9u83Kz+{{uy;2ki{7th_VLHq@q0C(QLoFM*zAGQwsJaZ)=2WiAqM8f zMvz~sTEuLL9Qi4m1@F)gm?~Yx;{1QJW3eiDuB;&uJyjsx%jB5cOJ#I<_y^?q=QP`t z9LDylUS#oKu8_ZOLh^F$d7`)?1s#ihMBVD1)2;D2Xp+GUS}(5we`nuAaSoFF{a9_m zMl(Hj>Vy<~_DvpNIEKh0oP%Vfy6?nIWtN!94x>|>)6XF+vs5xulDg5Ro8Zg;4m zRQwNHinlE;Ci@q^!aW`qq~(DfoBf~k(B765YhCskHz{4?R;0~GIXm1){0A@cULu^V z+oTTTG%WB3xlQov$l>bVQ5OZ#3pEAHC1d!>O%32xti_r?XtQ||w|UQZD~X2DO)C6k z2DJSO4mke{tw_y9KWZ+T2ZkKquki{xuR!_9l1%vtRqyJD8XiRL0$(5E!;ixf(|s)lw2 zdqT$_Q*=upMI|>s%x?yYndonMH#{Tmw3L&yE=_m;g8aN#kEZ=~&NE zfU41GxEOvDl>5uTabY~&-?0n2e~jkKPxgZIGY{I{R0($;T&90CHq$Q&F<8Fs9~3S< z41&cGOzY81xKdsQPh=m%oC~jU(Fav*zcYs(dpHsg+C8S@WEp}81zRa&Ws$vokwz2o>biS5XQq_XY? zUsgRbltp-}lFL?VILPZU@9nS$x+?YYum#zq<$ym^+p~word_rb0t&ln=Ir_qEw13soa2^Gkuv6nVKfhykuU&|zVOUi&LFTBWoDtf`S zE7~KH_>CVEdLJddbiwxyW`ORUe0b^bj>f)JB~==7%XH^nlrFJAWt2_#)^MxQ-D8{w9g~-q|1izn>0&;&m!OTJ$ zXDdmN86{t!@v$2y4vB6qBUn&aIL~fT(@1V*)FwR7rj1%FonV{#Es5@;rQ~_RSYqM2 zQ&{(b307uuYFbG&Q2R4aM!kol1(r3jKCwg0 zZ3MWyna1vX{|9OQZ-rWibkOQA)@=xk8n^u+sZsd(&B(>X*&CRT^{J9^YBJ zvsBII?OT}1#3R_nNsc@%Hz00v?jz-4V{lgg8k~6j1IqGWh&KG%&FAg@ffg(+K;hQ| zVB($E+)Cq1obuVboU6c&eXCbu(&KHwD!K^t)Gcs&a5h$r%|U*by^y!_1)Q%l3OtG< z;Pc~XNcGCbW@F`uc0m!eS1IA=3pLo&fK_aS*Acq@k1|e{If8=|#MEMz5wWo!!}RQ* zbDc?dz*?-2C(4`D%+{8#xicHs)ebM}n0l5j8|97c-X1{%3wCo=<>pj&RtYtmvV>nB zJPO{;7{=0TvSE*W9;lC#B9^a1a8~UcCKhU-KQe3i)@8Pk-}VMJhb)G&oz_TZ^$fwc zC6@#WcILJLKO@1u?ih>tFrA%C{EFjrBglsxRz$I719#xpUFeBQ0=n`jM2wlnMCN1R z`<^@)KZ?*x4-}abE`u9OQt`H+NA&O6)7WRtd^Fen9C~Ej&I0=Lm}rczW z?FAdy((ymI$@(YwvIDO0M{EuGv(`aHL=9NtpHSPqjEnY;!2FuwaA%7S)qmPdqs#rM z%!en~Ah3zQR=o)ax`o1<-_j^+^GSSgF)wym9*_6@$cFfH8f>KOTbT879s23_3Ry{= zfom_TIa`NNKJ=h8IT7B^UEUobbj44Eohf-}zuIf~oE3$STym$=9`xeF2}v*_*IHP( z;voNN*=SMVKP!^pr%Uc6yOM#`BjDiL5_s3P2$CyBxTe(xvPL9gNts4G{_R2{{5ixb z9#CSPYdE&~-C?NCw1?}{oZ!_g0ZER&DRk@05LLZt5{nmH;hGkhV|Tj@de%okb_V(5 z&3{6$-)aRI@kp7NIXK{`o+GILMl^i5??7@V7x1xn4~Zu(>=)=RaE4ohnQ-gsJ$lVp z9!=cs3bWq-g!P>YxU1t7dbZUc->^;-f9}^2&Ir50dpSBW!})Vq=;%9mR4D^Ka3njs zL>|eE>cim`rbBbHFL@Pg36|6P&<*d!XnmF>llkq-QbXOzD4h~4WnU$h?>qzb@^+*m z`#1KzsVe?2M~=QM%cXu@y>$AhShV{;ZM3G-oL4+q&VtoOgJq63c2rygOCmhkC{Y5e zcOOow)?dJ_&HiZcT{=IfQJNGi9j*xQAk zXt{@vNy*zcn#GU+2dLmr~n;6f;w;RB`vQs`lhg!qKMr*NVDMB07J8v>tfr5vFT`HOC!o2HjjS!^ucbx`Qel? zIlHUxYs)~8sj@J9ra5sq6Tqr|hqE6`SF`3x|6nAuqVGIZ!R|pi4jMiZOtq)riI+TB z%*i@gTwe}~2MX9aNu=H+xCLmw}eyvw27p>L#PDZM)W zEt>vZ)qc#|E^8u(R7sxmaTi5rq-d8vEH2sx;L?ZzsSy%N#Pc!)RgiW=gN z(^fIJ(-xo+9f?(@$6_x%K%1*h;luxu*w@i-VCO&tp1o=kS#!gbbgZx?FSr=k-f|eF zELlh$X8~C>n~|V%ouoZ+F}#0fNW*UJ!m0gx*~RAlXoFrgz97AwajScH_vaT#^PXW) z_k+Qa73bmp^_wVaYZjUu_!y~%$kWR$!>FF_DLyqIlRWY+p_2cikeZXSed4fJq2tT3uVN{S}bxgcQt5p1K$cyX_n?O$Lq##kj9`@w? zVKV31bngkE|tE)#B z*!AdjKupvl82QT+*8i!+7pnwxms5erCaoN;mpcI|bGOp84_%z4(Pz#uOo_h~{fhSf zvZM1C8-u@F7tNW`%#EMbgbt}U!t1g=pmvd zkav@*5$9ZSq&vSPvs>4LnB(RIIMZ_)CR;Ya&*((F{Mt5diJE{?*;Aaoo;y;{+Jvm- zhjOR4@hD}gIqpuq1^d1~2B*N+P^?%7i)+S$fAB)Qw894a4Lqlg71~^FFE8AkZcWWU zz70mD%ze1GAOWWPCSrx$P_FjwQ08;~Gdj>^3h_acSa81MtJ3djieTv7u5?To98mK zfBy}!Mc2g-B?*llx-$@c=^IK}kp>d3X;AX}0X}7Afm0Lgaf(?m6WvN>sqsZ@*B9xb zO#LeS-mA+RhZ*4MKU!&hW;9I7y(kD=VP#h#IajD*?gGie$b3A8g&^0M<1n zaBKY+wDJBRwNCZJmxA&ccPR(r<93xx+^SU-B%xn& z6GW%x)2T7nNs~-6+0YwH#FO%fHUE_COE@>Qr+5qG3pMd{Hv?$jnucfSW#I zdx*fGi0FC^;IK;(#P8SzVq^OPj9bsak&mLLqJBe;{coMe;q1$@64Od*QuMev7d)CsHpGx;MoU7C1=4 zygSKr&&i_p6g_-l#3fXEas_M-<;f?n57^3RFI~>-km7Hbq3gT`J9g;@-SkPDH55Cb z@3P&vj5a{Qt2kR3&q0CBz)rEm#G7cQdAW%5(;xzE9qd*8PpEOaOf-1Zm=yk+Po^v1 zCSEV~U`FlzA%->|`(^py5zcwUU*aJaddp!cNfp?08i-!SU!-aFk-L(nOfA$DP!dB|2-@Yo?a7>()A5QqDj?$ zN{JEO6Pk<9>A&D>9^Az_Uk{W0mc?Z0F1E;M)T*((>iYYLaJ@4?|HAzl1s+6e6T??Mc((FNG8Mn25+irQJc-lD_`x2|b$v84IBGt0$zdNhbl-+g znM}f_hTYgR`9D13?ryB~(2$Jj!^FsE6WJQ-gl9(VMIY%BQuQjA40`s_bK5-V*K3cc zim^457h1yc?p${2X%HE#Lrsk_&A%p8|*SC*@P{rz(s}jp!lUTvY7a+&t|U>{ulH5G*gavTq!3e$hf2t6fB==BkqX9IR2T?HEb*9ovz!yFT9K_6si#RtIxKRq{YnkzC}< za8Ugq7MyRyPwUpf^44-fWD;_(;%4)-Sz*=w(dc-MCHw!|GBbQDp! zx5@IZCL$AH4i3abBSF=b!g;P+vR$tRa3WQxr=whP(Alo2^pcicJl?|>CE z9eodq(v_k4tSJ$WETwx`x}G6;Re(W0HABoutT>6X~gmATjY3e)FF! zEM>G5^yt6{CRrtf2UrtnEwp86(uH82;Vvv% z*v*aeUB^# z(9q~cnX5Y3b*Ur5qVK;UccDK!rW(mgRu59kBY7a1un3|rcEk7K_xa<0p3iL=T)}x%g_rclvfx4*N31(tTA@!RyzX<6Nf}Wd1{w+#Z~a zeAhIf`-d!W|Hen$p*149Q9Tl#$mPJHLso2dk`%b98?g(;u}GL5%Ln{QN2;5y;|KqZ zAbuH6cxzWAwMo1{Z-0$v?sKG>|J-M2M*dHBp{tXzbM{SZ$?M!WLLIuyTJzw)I6i ze&2ZnZmgFk&XNz1S)B@uNg5O`s7rwza#HNZySupm(cGSB6 zGTuJ109M6(fm8Ad^lIoqmebtKj17OmW3xe&jdnswo(hf_bA$$>H84Z?7tM3q3XU;p zAZ#AaD(1Oh8M9m*cVaQ}gWJ$DkN^!0GofPO9-!w#J6XFbU|HW&_Cs@!^*7fu6GI8| zY;^$1dnQ93)i}TtaQ;hG;<6e@fs zUXdq_3u4<*($`{|@$478dl;dsv_xd$meII&!KUzhvwI0FSw_i2y5d0gTOmd z>|ikk(H>xSnzd}~{5n?Nw7$kxG^NI6r*e&Y)_nH;YXW<&HG|C;E`??(O}eX6iDvbi z(r`4J=6}Fh623Yn467-pK4^`Hgo_du|eYW*b9W`hLLYEq1tjk0v;1%%cu5 zpV>aW5>_o^$@-)0;oj>4wEosoLH<2S=DOhxn6JAC^}ip&_5C`mPF97z7ha>jC!6tx zQB`oG#kHpPT0Tqb2x9*gTxB=T%hbq>Gh)x?Okul%3|MOV8EA=Kg`%2rQQ_ui{PNpj zFzDTjo*cD9#?=HP^FKIbFoo)M{$lCg&)KIF(QMr9cM$SFXXh2w)DwSuniLUGL8MES zqSDDZGZPhQf(T-xTM)5g!-^o%kq&|&O`1qiP(YG%W`YRz1}IoT#Da=HP*g-wl$-zS zy=&cvf7Ut=c}OxVYn@EKd-ne95*lxaSPQcadRe(16Q-lc$^7# z7tlU&CP1mo96!o<#Wvn)U`vG%NACkZ9Av3VR}1nOvSkx%AM=)9s-+C%4kpljV}{^U zwlj{HITvP@OfuCC8BkT@G~CrpLH|b$Xr$Ab5Mh=5U0`4 zFRA<`#iJng(+wCaZ4Y%-cEQ_q>S%SW0g_7@p+oA#q0!oO;C9vkkFUtjh%1Rr(ylfktT!D?kJ|UB5M-Js&rUdK%9)RbkOowmI zr|`%7r;|^K6e$xkCM#Pp{On#tRLpcqmbEUp*O+Z{W?d%?TP6uFrANZfyC-nUc1b}@ z^&h-eLQb%4`)+31$aTiaHkdv7GXq{yMexN&bDS`~8*r*;HbwJIDY>y6o>~LT?hvw# zCaNN-mHS+YKQE4)kR0QSH&uYMd6(ghS1ZAQjXaStKTm2-C9{XW{e~Bb9-bfT!&gyJ z1+O>|rIsaen!7S+@mwu7_J}wY>wT8pewc;*d1<`WgF>vpdLuTXs~fYmy2QZAik#3y zAZAlDyGZE$;;0Bf()kQ{?b&n|u)_06ISX?3OcEKmYes)kNM+xxDr2Ks#L3ZSV`6r! z2u75d;WcX?vqR10?Agdb7>r`TH8*3dxq2HA_pO7rrT1Z2vklP7c7yiQQ}Mic`Oqio zGEj8>#_n$m`JbRQz9YV$7*($zVKb5mTzCinCvloSE4~W(T8N^KWgp;`mfOfC^(mUx zYr}ruI0QYT>*1x%`9yu(2H(q{OTtX8fRg7q;8QgK(@tqKzaOmt4_2%Nr&cd!oO|-D zm$eSkFeePX7-B)*P6mFtcZGK8`iK3*DoN0wEMuYnn%~vF4Jd>x0NdoZAfvX`@Nmd{ zPRtTw<6<~&1wO@Mr`ZK$Y1wV|SxAYs?AbiHiRiwgF_hA z8XZm1Ge6FsLyI{-s(;}%yC&HklPUOe##5M(qlq`7h}rE_5_aDV?SmWnwQJAeANw_N@bENJ65c@O{|Y4?_v=XK;$p&S zpMj@7(ZR-5rua+iMpiC6nIzr!CRD?3)+0BQ*Y-w@SN8orzX>NZD#`2E{D@RIIHMTH zjbI$?cLcu95_%zRKH~Cowb)1S5h*nNK|5WwxIJgzfVgiN@Mf_#IGhuYljG|dqec1j z#TOY&c2^*~vgHnI6B@?M5VORTt0C5(txN3|^m8tEH&SJWF@!xCNXk_Qae%!Od9quE zG_4OMCp5L|7edWX;i$D-8>MUcAn7w?WsCp_?3T*>gNAkddRDWQkeHtu?hy$B63FwMS0h2bPj82s; zb??~%TK7aYUn_R1P`5t^o*O);R*koVcnFFh+4L%+(5pypCn#njlltur# z_rRWiA7FM@4b0zt6y!Lo12dWdIC?S2SK-m3!uOay?>1Q7&Cphhx5G8XouKZ57juJ= z=HI%~M(NNmXEgQeVw7}W z9{E(HgE8(2kUtyJ>J@uHk$)$pJ?RN96)cDDZEtM&OWhe;K>a#pkSCBn+6|s;SwaULa^YRgP38qH@}{?DHdBS$4srIL_kt^*FN62T@6$C0 zp8$_d6>!|N4#MMe*n?ab;2-La%~GC0`)B2_r!|D_eqV;WT(e=}dp)#qD6%m?=eYI2 z(psAlB^`40tsJh8mL?CIieO~x5t#nb0`HW`fYXESfEQ;**^Rf2$=aOhNPJcv)NApE zy}Mj6x7(lPUp-1x|8j}>jBNI(Qz)tZU`6Z-&4|u)7n~7g0DT-3;C&w-bhpz7X79__AMs6|0v;-w=yV2h>dWkF2Yj|;&4*Pp?Tmt37$C< z>P49b@ZFh7d|q`Bp9M8!u0R&;i(>Is>r0rDmWFd~n}R6+L&DsDB0DF<0a!c!M_b;1 z!(Y<%g*DtZL?2fkzsz;Fpl&57bH?E_ze4Z-dxPG4XQAg0iqV6wMqHP_ zib!m!E;_LPAo5IT$9s%Z;17~dx=a+nDUl<*`}6W>FJ>;BICl~)%Q%X@Z!|`SDkh=# zk*&zL&=)-l?m@HUf1qDSgzU?;oyfE;1Qn0`g0F4&!L#8!;Yk*z=!Ei^v{lTR$6 zhq9%yxiJ5DY~UNa#gl_>_FRX{rp|1-q?^LmDyg!zpuS@N5FxXE+Z3i^;ark#eg(Gv z&4YDQCE@4DOsLWk1vyVAXyPJ?E-e*>9H>He@hwTq;b|nrMjtI|N{3`&EIRdj9a6U} zf^Ul?QNe2tNtW4%?-yj?fJdL$lk1hS?zAq<@pmM*N90MJr6k;}ok}#K=aAj$=SlRq z3b*#qZ?vLVoO|V~IHQ|k3LA$-SXYrK^iz1Y{4Mr^5|m$>D(KE=DFKJu2qweDXomOJVTIb>pZk&^F16@WCp5 z{~TXA*TarG%%B&6S*YXLB0T0d3`Tk3P}AfTQVJ5`hSnQH=l5yY&fOHO2*>V3-YtvG|{95iKQOKoggpVx8>yyzaSrTBfh;38NVxt@`V;hwUEZLBYt1{=o=Vq88OD=Lk7QCV(dxG#IzqR;G zz#x6h?Y%JjzM6dYTuc7$(??$3G%9L~6#6M%usf5ckXPGIfm+kIP|j@x{IRYC3wLQF zm#;gZR7^2AXm|ymsuN-TW}C1RQ>LRhi*S_t+8^Dvtw(f788q83WQy}7Nv+-@JfAs7 zqK;VO7iF`+tG828Q@s-UH&_n;^fkhZ4vzH37>ZJeIKWCWyU0qZ8vLvDDOgi*9#?GW z0_^r${MuiQB)!~%&qcH{qw!hf)!pp|Cs|HG zN+o@R{s7K*4Pdp}^(6gDExG$@8|g?Jg0{KK$y~8s{Ndt7yvuxqJ-9Uuzt7%6Zr!>= zwx6#dpF>}gUa7O>+qyxv?0O>H#EWA7?o0$*_jHrc1U~Ur)F*NcA_CJkYw||+Dhao_ z!Psdo10KJYf`1JcN&f9DvP&|7SiamuJPMtN#Opca@G2cTY*!gj>z%-l4HUs_TQm0j zWkYsBL^qQd+J|?@&m@isS4q+EGV)~aO!7Qi45{vQ1GgM2v54axc43VXUNc=2zl}Tr zV;?H8g)6j4)m}|z-pRStw}rp?8S=qI5aU2<)%|eP@|l9atDFUNorS=nJD05;u3!%Z z-(vkY^GP5MAxHOKA`5eNfRlq&0DfJ^eod1TSge*4XtV7E=j;&N-_I2=8{7pim!812 zY9Z{tJ{E`mJ4wb@Cy;o)F|m7=PnF;}8jQ#=UVVw!&G7)8vY``iiq{j&->WT{qpc}8 z8Ci!bicitqUxglf%XE|}3EwAoEag0 z&f{dnIFn`1SE>u%$J-~GU}X9=3+K6 zvJ~1PSBN)j`NLmvYtciW_dvW+0t9L1(DAe6ki9Sm&AIylCeO`=ai?P8>iKbCV{SGl z*gpk4N;?W+ktOQ=Tfx_8nJVnM$;8|b+4z-l3_Q`&4|wz`xbiK;apFr@uVF3X`mPTb z%*moNrp-i0hJ_yW#Bvrps4c{! zpo>1z8V|~J6ItDp5?JKSeeASHjcD|^kZ6AmqG*#t8t&AkZ-+%_QLvcXSJmUT!gS{s7DVfMS11V`| zf(%OGc}Z1Zw?PRVhvrPRHY3z>p0PjwB=J=<<+0ky`D9Pj5z5BHQ#wvBt$VENgxQrW8mbmt%2Y_FYBDD!jx0Ug^_1L%&(~ zOfOPtn?n5mdm#kaoPCYTu%8AAum)>y> z{~{e>WDnHh=CXN0#&fx|54H*OAZs6;2OA_N;hpbM*rxLj z8!jtgd+#;E3ZI2&fnf{-gslFVa&e64Z*BIiZ3)P)(&OxK+{i&x0XVtse_0`fTS6FQ;KB}zwSZ;q?_lBazaYf0gnW;; zCr_MunDo|pWazmeYTmg5!R=r1OGh27a&HNWoH$HwmxkcU%m-j#Q6O{4mj%sTldRV6 zJt(KQ5ijZvAsu5B9WB%pPq&sp6HhbtS?YCEx9cY(CfA31zH-sMhi}>Ke|M3Vk9^pp ze+Si!Unc8B*5HDcZhWgJk;o-x;fyV((Ft=Wy!Yb?mZB4&Mt2h{xArCIPaS3ZGqN~I zv;m_&b0cND=eE#uB8z56T|$yC<Eu^c?`9kvO}}%4lMxiasn@#}Bri?4ebiQb=ke#_3w){sUMUOKeD9jM8(@lV9I}^|<^&p&5bOLQ! z<3fKHU4WgW-hzi};^fo>L1|-RXvL8s7~#4MUpI?Fhn5r)_fJD4CT$9~V&37UvT0aB zB@|76lFeTjeVTuw!HDV_F2!de=AziJ6(}Y=9&QSDBs21gQTL8d{2#Y^Sm}Hv5Z4qA z??tRY@53vhtA8yH!Tlh%y#>Z}uBSt_?!xVy0Q3N@g*JPm0cm4lr#uBK^glxfmF;jM zA`@@Po(8vBSu!_sCcvoUGPFb?8x^m=N%kv`lj=nWan#rKNIm)ys`_Qlz4KBXDzgZc zCKtj3MXQn3_AI2VU4llv2>M)6g!-GJQUAJT6yh=)1#FQ+aod+5d6&n?udW$h@j5~Z z{L+B>ye3q9`ZfR9^L)NH2xrfyCL$*d6Xf^N9_7gEkb}GhMD1rDd!pAE-7OhI2{K*q zj4*#()f+)_7oWm@zJJ+=-GAVNJ^Co7EVW54poBMx3V1$I<*?yM3_WMk9;cPvAiQT> z($}4c+)|E%fj`q=be%alGG#e<)UV7KsD5LWe;1R@tA^N1&mfYJFCwcW$B15WBMEGi z!&BD|O8#7H;C2QL+p6p3;I)4 z811*@FFRarLT0C~C-bXmQWoMRaFYlTO#ViKkaw$Lx2QM&>xJiRc1R%fe4>IHTP`AV zkNNZ?_c9K4xye5gC7c5Bd;DdjpLM;YhqN=jP%ZB;lCb)M|8uQ}uRVX@ zn+i|y!?+;c$MpSR;5#1$KE6vYIKYJ^`tqpe%QrT2dm-!9r$P3e)h4Mr9^_4lH97pa znQsDiQ3qw;P>(-0;(dd`U`(|HZkp?Xvvvg$vjx}5qGhg3;)E*NJg^bfChFlE*&_JW zZw_?Kzm9_jeetrEmvHwH2A~(BSofX*4A%EV;)AWw{*xyAm2Sr`e}$1l2fT>+l@wtA z_${LwzK?vDy@t<6S&@NmdlJ8PG3#Mo3ELJ;Cm-7jq3F8bIIeLec~yUuG~_6eW1mc6 z?(jH&a`7vuGTCehoG{9Eyt;J^;lBl?4G)Kait*%^DQZoboPi;fU0wqmOu+$!Ys=(l%(3X?pT zuwWzj{_Ea={$va&RRSbcDjf+vY^%)M?got~tA~Jqo$5(LzsqzQT`vyHWUr zJ4(yWfo{?1B%yyZahu+W6`C%vtxXNI_IMZn%FZ%oMZZBZV)ia0hUwMt!I=qwf zsH4UD{qN~yIG~bN`|uu}*xZ4tK{c94&f)oekl{;#X13zu9#k2>2F>0W4P-CGP)!5D z>_|%}wunE5t}BY7;5jqV;lOSrnEe^azcJ@l9!^BxKI}wGniSDH{qL+jyN)d4?Ez+Y z4m80>Q~3L%iulW2Ujgp^uW(>W3*2ee4Mo2K;I%%HQ`2^iHUF>%CCJC339bzJeGtOy zcdb#^EMXRV_HJM+En4s9=z^@q7oe6OgHZ6}G5=}!b6gF-;*}H8h^9V3ZH;fxzI#D~ z@?Nvc_O(q`PTxz>Pi`Cvo_Y~!UWr0WGZK*Bjn|Ns0=Vv-Ii8;GMh*oIu#+|S*sGq( z`2G$t7@1xNJr|!t>NjM#U#&t>u(p^Ijw;IEa_eJc)Jq6gc`ULh@3cx|L zp24`%Z_w?=C48^?41BUJ0?ltr2gP7IG0u@B4wYii!6A`N>K|l2yo!YmHV4_g*QU}( zZAFovYXhG0PYJx{b)d|sNI3ibCA3eDhvr3>B6)Ne**P0A5<{(=I=S<_fd?6oC+h_( z7z?)6c+4ttY2oF9M7jF3;b+i!=PuaXX$xiJMTFpHFZP^B6&v8@08G-(v(^fW=!kdW zC}HFQOsYDG{>H3<;V*u`6FLU$%|?Nhs+UQV^YjuMf6J{{{$mR$AMhdH7T*R-j(VW{ z;;ZQ0yDU&R77F!BoX`f7KDPE*CUc@y$c%-FZ1E~S2#ylwIp1Up{hKd@{F@`NyEBI9 z8c&5ABl6gWsS|imn5%b{ISbo0Uazh)-;gaCSAYSA@7xkIeWTkER6igzQCeJm!^3MxI|5 zdU_wQc^Mw$wcZ_)H@b&}#bi(`ZdU@?J(2i&@+1gTd`-(}FJSZZYHeDr4pJKWanz=2 zVQ=8WwfMnfWfK3NI63ho9^NS5h_-%tLG6y2f)_Ip?AOoHF!rMeF)@onrPt%xNV!LN z>h45h5PP4WJFykS%S3^2?1-M{9!1N~T5`9XRskx{e95M#`-x~#26=vYJ@H&HLr|CS zkrQ}mgq}Bgi;YNj#B2E%*b6d3wszb&o*8*rm~|QjyWUolm<}f_={o~!-LXR1j;nD% zr6q|IL)bCqBz138gI>_pj(6)F$9r}D;VJtrfc*0Z$uY``6tAC6@SJG!qcxW`4)wX5 zJ!R`x!yQ|EThHs5~8R;5xLg#7X{w5cE+FlW_*|G$-yNJL=^b?`ME1G%n`U@+#qK8I>?B)Lw zbl^W>9xh^d3?{`t$4^Uz{GaedOdI`TM>grRwR;@Nt8OlFyIX`EO^k)VX)Dm+!g=tG z%zajC_D9H32}7OqIcfoOlRB>=4kPC$W0!^${KrO&G#|<4f7<*V3>DRbSS@+l`CxY{0gzyvytyoz5yVfWP#)WjMQFuN_CXKEJO54a z=NW!)y!-19XGVY=E|cnDd;XTdZJ7t~lt&-1vd3{+WaBuvTQAgo;7#=PbsT#5SPov~ z`oR7x?Lf4|5=KS4QV$ytdqu68U7|D#8^8U+tTHKP4p}lZh~q%(ufO;wFZ^I{UmR`X zsO)OI?I>Blee^&0appnxZ^(b(d2Anc>Lv(Xl1UD<~MqH|&Y?l>qed?uJ~y3B&WPsXO#x=WuH*zR zHnJo4_Jp#tlcZrO5$?gWNFLnu7x7qCq|T?t4BC*|>@Xubs_zJFdy! z-QULdzMBdcNc%tuH!+~ML=EK6h^K;e9`Zh%ZDln}v&qh9RpgVkCGq}Ij(JV(@CJU2 zzo&TP=PF5X=dv=?zWEp|cg%-v9R?spxc3E$bFk&wGMm&t0sKGP0UY=$i7RtLaEpQ_ zqxccxt?(T_^(X<~EbgI4lj?z(SR45Acog5+C63ndk0KdACA9b1Qlwm)4@;{?;FRxO zw4R6zmJs$gXUyJ??dPblR_#tGUw1y5y)p}(wo5{vEuzrb&&SYqCu4Z6<~dwYDV$}? z2I!K|GkjXHgSoqkk5%5wlDzL)a9VORP=7F!J^!c_{)n!Cw>Zay{$3|!E#yW=9!>-U z>ND`*<~U5o?4WTo1ww9@S(o2QX!@@A?Xp*BJzE|*0gcD9p^d#H?0q^FD>I3>y|tMY z*{;aO?j)?)$Os4*`p4Pm5xB#{5Eh>wD zta--Tu4x7~K|)4nhoy}d1ox~=AO zRTAza6X7+Y=nN9~BDnvf1&j{s1X=ei!AiOHbT2c+zoC{#=}AS?#=n=KEc6*^c*$`8 z(}CQAI9u*+x*he*JIQ~@eZ^c=a|LDhvyu8sC${;Z4qT<-&Cir*>AyCvtyB`fzm~@ENHDdFF26L+Zs) zJp7jQ0Nl1?;0~hhW%fz7~qb8K&?ub%`{6M#F z?B!QrKB`?60H5A6q}>kP;B^}=0P|O?v+w>1JMQrybWLwzTK>MH%JuG1t7k;>W#>2H zHj5~>PVOY$BsAkWZg>Y@)(=zeT>|90Ul(n*EI|_6z1hHeCA@yIKRWkNo17V8u(lwN zef!!DU6GJS(nA|@G`}XkBxm=h$^B<`9-iM^h0CDf+{Z^|$1Qr{)@XAy`e`=uQno<#p$+Kf`!1wGJqLrudcusx zHs~>=3q=E?Vb(krK4}!PZN}!IhbR$wuP1O_T0R@t6A!h@bKoDdPWD}B094lNLOnul ztE%@Y^j?&~WlE27XK9#oljln!@x5NOSQKUv_W){iybKd+4r@!!5}9X9)^S%}3=A%F#sMMYQwBbF?K{1-WdRhFXe4 zP^CWs)vkf;Gm~jR(&jT2yrGNMEIcjD{Yvu6Ue-~)^&V8r@)G#uuL8`vR*%M%^Q{Ki?Y~Kh9-`EM{Od^CWaxupMr`br3!9 zwBYWuh~d6Lo4L2DS@isIFFG*G4pFubkzCwQG^RF;jDM_W>P!~#A4)#q)5o`=#HH8q z?blcFxsl6g_-iuCiY-D`>&=mpFgs>{bOk4!p4GZS|=|Rrt8!hkRg^Xx95bkPBH+Q%O~ zu4qGxj{ZPXi=xpjFLxwr;Eh7}u}ox#8SGuH0%ukKwQ+8{!+x0I0+;W73^!b|MFDqu zm>NSFTw5dPO=u); z+CXx-O}H=L73J~Xmsz(QT8CZNsN#dd4(q-Lqv-AMeWW*M3(B2dg`On|Uw2&rr`?-| zmn}rNVp}raG^Z1;PB;OodwPJ=*>}v@AWWO~2cq!65&W7zfK`)(#_7&$Q2o+eH2R)G zF~*D7>fJ|~1<#(cZ^syXlyep{842LxN(Sv&&qp4wrf@sG)6k)Rkx1udJQCgF1vFQ@ zVAlq<<3pKhWOPpxe%)F?9-F+tR|aj7_*zvQXf^_Gs=A_kfBVtLlqsbC-2mQ}YK1Mf z*Ye*qQrKk`2an8sDrAV8qMtgm(D|oXO+ES(O{;A0@p2qC;pmEwEI0yi*;EF_Y-vX| zeL`2o$A$E)M?dK&@65rl=Pa~m;27>YrAfFw2#jW|2F!|D5Uz;ORr6rzW?T=a2t>Fp zeGTYj>~3_gQ5nYjeu5XXk3jF;k?gb1d*p9`fZSGSr!(Deb54!TV`sach5hkT_^U$~ zen+3fq1SWqf60%jvTeF#Ragxkja6oUU6_RxtL|bE^SQ)uxij=G{Ai zT#o1?`zJrB^tLE4w|!EQbF67+cS zp-0eixZ}-B_++>mp1b)9s{1t1i$m{W_EkOkgg(--%LeHURhQV4V>Ph-W(HJ#R3qK7$SDtVFZ_l%sC{w@CY^6SQNsuu3==1YM}4(iMC8 zD%~RR^-Ev+ZUGmJ;Wyx)MhR;rcNZx5WWn)?JVv$~;i3cSSX*W*yrbg@s}-KWc#{Uu z(pm>R*6-vc&{Awo!AZVH%L|&eO@?##twH_CBJ3tq54U>SAj9qPDCI zscbbuvMW5;x3^^Q$}lnPG&K_~6Nsa=T8XrUyfpjbzyJS{}r%4spUt4Gqln6VG9GsNGa+zTZQdqICD=>6$fvb44_*V_m8b?#P7I*JO@Q=@Y&-M@P$2JSy zgaupX>lI2&HjEC<)b^g)nu4Lmt_CiJq;XU!-F962iZFL{Vt z#5;MzyZ|fyV6JtJMiQwz?m<>usKB@WJ4Nz?28q#ANx_};Sm-uZ&3fjlf%n4M<)XU* z8z3EzFF)uclg$m}!%9aY6EEaMwwMylj#T_wl)#b8rJ&YI9$NLf5@(oC&MJP#hO%A) zgFy=cZ=Hw0T{jYP_r7O*KX1iu^&`+@Zw9IKR3=L|DdX&lMFebGLDsm=Wq+hH;H{mk zP!B#sIs8_^1~o_Fy`Sfa!NT7J?u{mi4l3m4FDb!dnO;&@DB#bWKfyL^Uy6qItKnK_ zRs4%9V7FM@1v3xNWgDV|XJO1BZ}3(u@7n2F&cwws9H`MlOzQr_pCrWuys_nE==O9$ zePtRxb7UpHA1H^f9{q^L)HKO&&qXA3?lx#$lZOpP1i0#67xr1-O={NGkXdR^1c6h| z32w#T6@; zRubF^?lQAkr#zWYzd0GVU=Gi{##=ZM3I?qDy;hnk7H; zPt49{j^8T7GJ4Z+wvQ+t{k)Orl01^Oa4j)f-pGrIDdk@rEn)}F?Ac$bc0g3voz+R{$Y0_Kl%rp@AQE#*bpSoloGVE6@>dwh76gnCCbmA zkte^)iG5Q5Gz$m;fq&1#@L%%q+J_k0bWS3+(fI@gpD(}zw;$o`TfecMG4 z#4*ZH;Zz(v)x_QF%~X_`;ui-Nle5CJugl4@aI5w_eE6CS+Le5P_RcQ>2j>35)eZ3w z4{5;XJ`A2G1QE1u7bjG+Ix(_LgF1(rq4tMWa7uAKJihoHe35$*A6b8ilr;Z`mjv4p zht(U1TtGYa-YP~4EB2GU#o=Uvx=HT$zks%LM93jIH)uMy4F1zig`?X&u=!hcd?r;C zFZC87Q2iW^P5X^M$N*-BlQr4#LWx9OcnTwyegHO)*OFCmHVNA-!nTcgyTV*-ciw zU74lIj)Awajr4%fANpGvlCX{17>+E)Gk*D_3!}TRVeMnGet$Ii{m7k6(eEH!xew%{ zUL{ecGl+0`aTDq5T3PX3As{Q)2f8MZT93n8Nq3qd&t2R z9su&3prJ$xNxx)Dqz})62g8MZ1?qdT$!SUWA|jQ}Um3-J@Ftm_mUsoKXlg=>Gl4kR z+KS~>X2JH%wP4nDIrf%AEIXWFf&YahlNP_Dc#32vMqO{k>8Y-$)m%8=G$vRo(u(d~Do%P2e3!3ko zN=sY{r_FV>;n-&lIE9`GbI$bAPCHooMOp(LDq%tEF=q62eSNyk=_J@wnh54590rrB zCqTxBWMH!@0qDJS0b9ShfuM_qAgrwo99UBen&zDcJ8vd|{f{%jucB1oHTr{YI;07j z{S`oi)Kd^V+z&SNeFmJ{&%lZMAHnq>cfhOiFzV;CLTZkzD)n*V1?^rwLVrD>1sdWo zs5Cwcdi_hm2AN)vwD*CK6?YFDJba#BS$>MXYnVn~DL6~X3>H#K@*&ihqra$~bqciU z-nq1d*BNVjwP5R^NGa=Z-bs4Rt)2AW|FUVg;=(JO`{hrwj%T zX@QHb07`Uo;TLaL_`_kbb=LYq8-+B z)6~I}B_MLV42X)CgNmi0mKoA(%#349EL)28!B0sY;5DKMJpYY?+Bc$*xhMkbG;|>t zGk|+vSix%JWO{wzH7f4lTq-tCkrU(A$VvED%n4PV3E(D0aE&)arw#q1)u(@_D|CkF z__ufH^mB87zO6Qp82?B=y?=);%eY8O{4fCPud0Ammh(YdpbN10z66|Ybpw4h+dy)r zC3xs_lb$Z>P7kX-reMH3s#VmFesy~~-PIvWSxLPFnI;0D9Dfu%daDS=>tw;`9X;T> zcrN^9tO|?2tHS23g|v;vKKikL0KL6#ruFh)@)q*@YpgQP6wupai)r5l0=hNj8TiEQ z1xx4lfm1S{=#E5Tm**wn_q8k@eA*ldHXV%vU(}w{8@`AF>rrKp@yOHih+dlMU6a}7 zy#8h?`J5?dv&cASy3#aY&};zKiim)lvjXYptIO%LpPzHuZp2e%t~EBU6Wvx*(y};< z61zCMybfynjU{xCX$ke+w}84;yN_Dm;y~MdV(2nuTk4q6d1~&lJo=NwB`T65OId2R zQU2H7a5PSRw(-9h#aXVH0F0-s1cyt#z=v2#`qbk>+ICn3__#Ju(U)ZD7mKy%f1-T) zCEH84+oaJ6b3{N!L@r%UXVZ6wgXm$oH*}=T3SgKy1%#Yd1E;P!19#evx^v{8&G_Fx zHt}>Um7_PvStX}R{k|1%zH?TWMZg(7YmN3HI;wn-t{c8fhe!Me8V8Nw{1SiI-jroC z)?UPE9$3uL_;Uixx5B_bYZtIvw%cZbV+E)FN{VGr;vBlecM4q}D@~7gW`d>-e8Af; z18sXP>Db;zYB1tEQ$ppv^UZ-uC`I4 zZPIk+1qJ#lh^43PRsh{%iNNx1?0M8?0!J3M96!-w><=o};6X|PIBz#6G?|MgB z+v`w2jPtGQNAzu;zr02zl>1O4p&_*YRSOF>)ib7M=3!Q#%8lbW?H2W-dLL~Zbd%1R z+(GU*Jf{o3DS^83C{PfsMW0++MN=DM>EC<}s%a{sG;U^6_ZBJ8?$V`n&s0JO z9!%!Qq}=Cx^Vr2v2uq|rtaj5sgzGh3R>ZNme40ATh4jGfd2|x5io(n!WfOdco?dc_ zGfTICb2Gpe$h^!2l|Rpda5YW(@1`tzx-(1VpN^sUv@sn!HjPd^97X$@_0jgP)PbQ! zC9UKfMPD#DPxm~R1gpM1pq&prqP+O>l$4;BvifL5KPz;#ws{?B^C;n>W$j@jT2jb@ ze^g{mpY$`LeZ`EZ87}v2^y8<2<@Z+r`Q{$_`ImM&{@N&2ZasxwQx{JeeN3Zba?em6 z5wEN|%LUfq|D3Fg|Ax^!BqQm3trGg<#qS`(;v4w+^a%)|YH4M)Dtfnj4ZXluisBzw zN_o7?r?!6{03GMMfZtFjSi%bhsdw^0!ZrqUcO!5w#}QEV4&ZTk1<>j%2ak4N0ykzY zfXYVtaC5yHyb?HrTFNQsWSc2-Y=qE(uCF`kK$l9o?};bS+qfO5|8oNNvywR~ow=NO z7p`#9y7H*l>(bV@yl5+LwRljPTn&N@`oK%kXdzEO7+h*y0j@j{p`8pmD4BqNl-c2I z&XTs}9PxFlsm?EQHrp8)o7s1qY$kvBSsuKw-^$G_%PPZXDK$^~5H*N?aJHxmxdZVz zuxMQcjJwrt9dSp}x^o}m?0k|2j7E0@@YWF=GmNkfJ0?!Wx!G~1noflxq2lo5G*Q^~ zO9ch_$e{^2aWtf|n|3idMk~yWpx-<(1NoO+LFbr?V9{pry$ zx~JNhj(ph*R(!k-o|FskpUG6P!afp6SndIv2Y!L6i(Z2@GB-fy5`C&S?2Og!x|i$9p=En;;+^H82pOwWSLuUXtmQQ2zIPH)2dSIP48L*jBitGl3OqjVg z72?IOz{Emh!`9Un1`@{Ox<_Psz^@($`6K;< z5yP6YdH7qKMclL;LvAm}v*Rr(NOozcu2;Yo23PZ|I|MXU!y^z|CYA>1=T(iQx~4bqf~}aPtmB@=g+01s!bg=gxF(Ja#HgfVS-lG}(U~ z{}|r~`~Fza61xNqRolsh?nvTrMiiWjKGCl4%&9e@5%i0MAj6-tZtDA8onffHG8E>7 z--2a7GeKAN6)Tw7&Bk^-XIG50)Bo|VGP7=PBP;gyU{q~7R$L3k-_5m#`_2c_Egzx@ zYhNkN)?a!- z;;)s_v2Nd)_&^)_-R*hg=grAXX7DHe<)!B2W5YK1U22D~Jc96`dL*jc45Ol77*HqI zi_(sd{qTlc1nwA#LLvTtq&-pxR@H~W-bR|OaJOKsb=R_nABv&!%zSWDn*;GV6q+U) zVdP_zk?>l9%YEjO88UogU;PN5ujGDfSQti`+2HpGbCg`JgNf_>vEM5IW8D4l zUDa(|IhKp#a31xv2nrOa;WHyu?A<2L7C+)jh+G2EqEmx#)1*3~BpKSd)L3`1LP= zZ)X|Uv?q-;7w3=^gGQ46Gn{{EFo$2IKEh9Tx(NQo0%*2|kJWm0tm_+2N!lLD7At7b zmNt!a(Bcq!{z*st=pcy=LII#sQ$V>N+rTWiu0UL;G@)if6bj3)z`GTf*uP7Uut7S; z?6kW_Q2K2IzVwL2#S<~?KDRV>`Rz1zDA&}Gp;al<%!QPavNx=@Y9Rs58D#8uC>E{O zz>TB>UT;dqabqu>=V*yf#rNaN06U!Ru@X10(4bQ^zVnwpyTmU)&;|)1_hDjDGwk8| z1g}YkGm9;wNPY8qv~OO9UN0r^^y&cWP5Wgk$>ky4)p-@BYgj_iUO|}P`ae4Yh1qnM z9!&V~8m0bSz+4qKT;O8N)GV1zK3>uwrte-r7^fjU**6ArgHIz{?SXNI8fa>3jNebr z#jEz!pq?3vt+Vdp!hjdpHZF+1Nl(CP_6_(kNr0W`_<#*jf%xB9F6*z<`&nXmhX) zKj{gu9>H~JAz6acie7=P*;(*Er3Tl>Z0VibD(H5>5?ac;2nAP_pk8YR7QPO~FgYiz zwl&8aT1&{QdqJc=e=k|Q=?z{nkz}jOh1sHOjZom71%Jci;aE$w{`uSgw7dQm>Xn$f zgJ)O@JpPjkBZddaxzqcZDqT&|6%xvfO=5}a<5Uuwp~6I#PbRO=UuJl_q;R%y8zg-w zfdKUy5|;Ix=-XMrzE7g)<pT<@QbXl8lhEL=5PO>A3hC9w=z5&1&rWWZT3 z;Y9cnu>!KrD^cseSMW5S#~GH#@-P^U@ny>*Y(77OJnA-P%+AP>(Uq#~pSV8c#a+Tx zE~cR0@_{t|*+T>?Oz^-;77Gi^aKp@MILXE428z4|r(XqVY9E32o_1Jx4N)a<4O(7X zikI^f!P?#!-dwssj5WsaT<;fDGi=ANN_KdPHb;#=GqEsoDVCfO#}n1x;gf?O^j($% zd%F!pbaNlmt$T)!JLbq}4n&b7ijyE~qJ!vVIG~_^7Or2^fPU}lvFp`s?7n{iH%&`q zcAdY+_&P5nccS9?FFxHd=+F~1G~XSF2UVkRXe0#-k6U8Q&ew4CuQ`kzE`TBS7<62* z1}{#7?s)YtBJ~_&TEtL)y)Ztwlm(JM&hy_y{N@iXui<~~tzmZR(6oTu3=*f8hvzl! z;Ouh^`0c4av|Cz&&aqG!{JfH)E?iq~zD{;SB9r_GCryeS4GSU6@BqO>CTv~2%9@Ck4 zbY?jm-onj4_DblPf0MD@t-x5c%_OSd-jX|Ro0&HEIFcdv4xIOE;-;=({2&*AStqXG zNNXX+{v#OTD~e6YA3-j90nwg3%D7#zChG*k$ZKvElE~gcrkGsf>;3p-SUdZ!VGunI zzcg!MV8LY=dwYp3%(%lUS#uc97()VWlu6a^qrKM??1w$PVxv;}l6fc@ z3J=8c#4~u{>s2gB>|l#`_pos$4Qwc1mE0SRq8cwMF{N$F>_0sYcfPN~nzK68VAuwJ zZ}wlxYW4)k`^>`H*2O3gas_Stl5oA)d31`@#o}FCu$m6WBU&49;*mABJl==t%lb&_ zA&R6g^da5}T6A*8d8WlCjfht@&@V4UkxSF^$vgL%_y7#C>&`YjdhIKJ`lJiA)m;rn zH0CeV-Fy#1fg`Xx>IwPwafCG4Rlr!2DSL3lmhHN;l$8?C2WI~nxNZB1wBK%r*-KNQ z`=tj|UeE#)V*~I`xdxFr&mrsD6x?5^jR%A`02FM1DM$R^_em{i&e{h(Z!SUWp0%jO z%}ugqhcHm}B{{kG1!?fsfQ{G9Frs8Jn*UJ1CdvkAcV#%Gqt6JmSwduFFi6&{hr@BB z(D+>u#S5n4BG28xv#%tL-m(yu-->7b%hAKU0FB?a)8S%MNR_Q3oU8syztVibRFpe0 zGsbF3+Wim2>D6@D*P)5QPi#@h<0zV?Dq+wuX%sVPf=D(Ojt(UPE)m4h)FWhVYYB0X z`bqLOhw*qie+H$V3J~nxrY6)>3t@6^UFbgf)BG>++m^JG8lNe z0E$jYgJz~Cna%q^WL7u<1?q(g`Ho zX&#;$90Wu8A0RLL4(`47g|&5yp`qa>*cNobNtu4Ql=={0Xkt_>WPv z?xA#sRx$RJ56-F(!?0X0coz{%@?`vo#mv9VEV(Io!L0%cw$^j_X%_ymyM-r@y~erm z3aBZliTZQo@ojl0KKWjY8 zI@-XPm$ESX+FBTkJq`8V$LaF)gS<8B)qE=%Z^*KL2FG7@!$4R)PTG) zo1R0Ox;FY%`@^?Rb27eu8q+(-F#FE(aT!ZKZrsfzXtyx( z(=D-5aX-plam3AU*2DJ?=i&8W2r#41iA>Z9fVd_u%SQmW{(Au@tm47PXf^tUXyV*^ zA+XlW2V!NtNtyQ@(xy^IvRjcGBn<2^l(_Q0*6WxaZ>FLJUg+KZ3?$$`Ns^{Gq;T4k)8#ZZ;pex>q|kQ z=M{)QnTwSi?rCHh0d^mfNG%&r3RE|f%CxgY_j?U&ZZ?5#B?I`jpbW3s+(d!i5>!c6x>p0!JU5buaHyN*bPQyTLNnAa(6|+=c;{3BO@TI2-DjChelwxsQe9MT9IJ}DU z=&oW{R*J#5>jmV7h!EK-oz6@*%%xg&DMr~l20p!UfDM5Tpg~2U-7^L&{{q(Ri>8lE zTgU8rlRz((+kwURy^!6w7k|DFC4t?o^v2NLgg)QVDtq(6(IU35Vui>qr0{NS~2EnS}4Z!Sr(2yzw_gW38n^yo*QxI;SV4X!Nw<<^B z*tSU2a()7*OMAgoX)cOSgaV@`26l$p_~lmZJ~Gnk%uomIMfmi>Ot zmu+|@Os{0@$m&!bdfk@;=A;_hhtUP_oAE2 zPTC4r`eb3-(aSI^F$!XDMu3FuUNFtM%#To#;Ek|gfJ@P`{zRP7d-t-5H*W5m^MFnmjp9UZ8MPP;fRMxxt_D7>Z=i8q$72I+{UV7=NN-gzYx^&?(HQ{90a9#-Ma z$amwZ+9&aXmovEh@B}E*44`!uK53_>94L1JlLAMG?c2D;t zR(9WhLY7>jmkQ+(D zVh?dSml~#3%zlqizC$0)^gDZ#cchIpU)uyjXWziw<)e_g^d&fRcD_(o%(~^QT;F8!ERceX z`Ki!UqRIUAK1NIMG?+=r+C+72G*RGiwbKr>Slv-M_OAILh;|>qj1Mo7WFqqpd_`E_BvQi`|01MH&pM0v%1x!&-pmnpkR#a zy-d*~Oc&pCaKEiKkBOVB`3QoN7Vj>J0XnN=>9r%1QvnOgjeaya@iTBFnHU5+4KX`i0AUsVM!m9g; z`HXBrJ8xsH-t1wkCfl&b7pIYjmV3yn!#A1IncFb>PA7Cb&Ihp+Z9})rjrz$FN%XR^ zXW+hf2%c%mV~+GfmhRlgc|xMtn@Os0ck?g_!5@U0QHTbsPh)d!C=LuQ&K#%dn(oKDHhi1w1ee)AtMD{G0Ny$0nD! zTzt(dj;Lpjv~43*swzbG$4!21Mk3GF;Vl#X{vABk6~>V5N@#I9!(cG*G_S|v2tTT# z9~AD1tXbBv^k9oGgrk z`@Lf@|K%6h{{ADJ3=br)hRR9Xg4Hl%{UyRn`a=$$`9u;N7Qhq%N3f&hU^qku_FaAh zW#7$F%Qc0$f8id}INgkR!BPBGEQ~(eRzl7|6v(}aB^CR%nT+X{@M-RM(k22VXOR}Z zTcU!JzeN!bpeKv)Wd!h8ZyZf2#f;x=W zGB=PnTnYN?7QvZ^VdQ=iAAZz4fZJihc)*Q<#rNWvxu+|E;x&VPlOWD=i-FgjP0+Y1 z0D2#|87Q5;VOTSIk6JV=4L{QAiAmQZay@?^bk)ni(Y2?cs?vfmQ`V3ZU*m|C8=oXK zUS^Vxyk(?Cwlbow|52gSQpou8V50QEhxpzhEr9 z8%@?6w<3#Pek49w@gOnN8eCFL;Ol8VE*GDSnFpouX{s}PZu|=G>~!(6juZ@S|48%_ zC7?1$1{DQA!^2~>pr*YUzKn`NuFwpKzEcLFw?r_flE!pO9bIgt@Yv9Q5MgLU9u}yO ztQZ~AvV0jM`81Sql(@(Uy!lBj?O)IPry)ii8jRtx=7Ql^*hW~z>G58+aNVib8nDB9 zGh^1A!E{|-$n5(kh)!Jl-Q-RG;J->ucpk*fW3z_II;A%9t4NP{1l}N8w)tQe$j4Xy zXYf_n8jy3f0G2loxQP`fG#_BWjZ|!&7E7nnIdsvfN;>(V7ujm;50i=oarS?!h|@hG zl6B3Gq>78+yeBviD9Y(<3De*JNACCUpS_N_)7r9N!*;#bJgjPMgR5}4`-*2fEB+xbGMNz3Fg zHW%P`R)P2~8Z9?;lh}(HAY&j2cVk273X_Wr-?y1`+}Q%BKhA>OXELyPO+EQgkWXfv zTu1g)2ZQYf3Vs}#1LYN7Czs>^BQ?c;i_aGcMkU-7Nj*#;26s)~|9A@cD zK-uVJVt-~ox$=tRITAz3H_!Vd2kr>lQ?G?V5W)p|m-i>ghM+;9)m%vw^Vr1FIyG%DV%EUc(XE^sJy|rhI6y&5Z z`9!}BWkx%nPVZ!Yq+#|*igqUQLjoDV{sjv!#8v=0S{7%wgz6n-( zcYvyh0`w=mgxYi~)ah2od`lTTTfpG2z+Cd+#sXOQkZ7C&hesTUc#Z0y&!1m3Dr$^uyeX0E1uxZ4!TK^=|1|T#Pu{O@fT-T#Hp~F?o(M` zkYO}#Po^E$`Y_E&XJFcrKA=vjBEPW!CANvN|Ew7{XRw@{lz&MI^R3~eQ3`%FOU8!} zn@}(yjk3t`r>qQe`16gp?w8WJxDz&EfI=ALNXNp3hozv)pAX-w5R2wzU{W{*;mLP^ z`8yROrnHmBMVNTq%Awo-K;7@cf?eo2ZkvgITZ;C9K+|NfL`bj5@gbCBH zpZ!NIad}J4KJtk;+vUR>;lB`isghJB+kx&v1^k-s35zu-^bLzfvyGi_V$NT1;!*~U zWlflOxwpuXdket6I0QCsYJhoVG$xmu5;Y%V@+Y{P>~T2`4|Swr%XtwvY>`YpT+f7E zUnk*6avc38JO-?v3gDrdmn6u@hUCvMfNcTM$gjPNZ@KunwBB6kUs?;d22;SHYZzVz zDr1uBJglD|fFfH@nl=gE|}x@nLFb|P;Fca zR6PT-`ZCP*nVXnu7n(Gx$rAhHkzNAnelaaCq zUMbeV+-+N7@yYAZl>Y=iey^mrtM`(~fjXk~A4nNo6Lf{~PIbn5lM^J$iq_c9pG;^ z40DHv2rQmUew05V-|fO6de3KQJ>!DU`~HAbR|{u zqDy~2hv=@YB&%Q{`LipPe1Crr)C)LXKrftttPCk!T}-rvBjCxsHtN`>3_=~qg)Fgn zBC8xjf8Y9rNk1!%0k9YEjqb#yH%lS*@om^i7lT#MJ#u$*JWMGsflaIa!86NHlyu6& zK2`49t0Ik5nG7lyHR9H~2GrYHjqKt~{?xjBMs>jHTBPWa|@^NrScL^M?WH52JGrDy|<83`1=C<@7{`ba4x>tFC@vVE!SSL+_Y3{xT z9T(E|2foebPkH>9sk)d?Y<7PkZpmqo;^_dD#ke2NHGN-qYNS3FSDI8F9YR^&g9r&u933rvdx z!1`(}6gCCo>*ZODsiH2q;2sI4ogN$?a)9)ltYU6WtR>wcGf8Y_8ObyiCp)GrC;yi0 zCh1$%Ncal@Flc>4S|daZ5A_KeB>Yz9aX=Sro8?VTCPk5Wy|s9Q!w?Gp&H*Qd5A?x* z-IRmpX(m|S7=j~JA#cxg*3Wq-ooT8>E#hU<-OLu!yL~0eDJg*P#X7h-LlPC_=EBV} zNw|3(z`orAH0# zU2G?RE}Ic6FNP@Fcv9VldZc{T4l;h$64qQ*V7^+0@sC-{!#459O!V3kha zYwxF-wJB~)VSO+;Yfs$;7C+n(Wza1a4`QY5PO7 z>HX!a_#ZDm2iLG+@LAXby1xg>iktc{NxhgTK7GfOTX0Qm#F);~-3tfr%E67y8DKe% zL?LIWC^i(%c?)K^$kb*pQzek{IzNN5S*vVHCC9gdTp1nC)8zyM|6crM4)n zTR?DpWHYMUiLfjF>tf2P?=gX!lIbyOHWTcLjKJACxeY(NMt2~$Zt7z{`YyTb|0S=a4H3#qHY+8ltr;&DPonwBVE&+$#N%4ls{{LlVnY? z<);UhCU3yII-5}N$TO%Ic>@z$g)pd>#d%T&*l|%7C5At6v-Nj$#1`yl3qb5}EeM?n zg8Z@pj3^aln=)pxnUnqT?t*X}2#P|5)(yn)vm5>QTpw-x^d?FL-N%UBS}aW6kFLJk z@Z`}$cu6S`y_2@#)ehXGeweKkf8PmLjC|sZ6(xMkZ9U0eH}PkeKK5t5KRqFGH?qlN-8NDFOM^MrZb3?W1n138 zV9_U))p&W1|L2hj6Yb7tMwiPF!)K)=R9+p9iH1W(nLj+PiUEl~QJ~%K4wmO1L7lNC zUi-^nQyiZ?^2geB3{$G20 zHJ=vD-9q0#p264u?g0S@%t7AL8zK@;;+2pv{5WtPTl*j4jj>9c=xIaYp<>7}&x6lP z8X#IZnwhsul$=z1!endx1#6BEuG}^NdS}-_ory0TUqrzQ>vwp`>IGKVyv09#%HY1A z^RxX6hSwjS!bH~{h?J?|M4QubY6F8pZn}u)rej$ji%q5n(JL|w!o5}F{m8@eHsE!lXI-FG06RgT*MJncUT)d{jocMh(oD zyaajPPvMif5I&!`7wbk1@x+|P=-oDzY@aJj`qa7^|6fKZXu!w$C+sme+!s!Z`a{Ks zU~bQ^Aoq)JlbVEjV!1LB7UzePRX#MNR)nCYw;ek8@bK=PM?}nD4(x<4L(?+@{PEog zm(&@c^S*DKu1bJ>tbfG#LoxMg+8wI>;#w~h12O7B53|MFq{FzSzMj-JKW!Ye>+hY2`HmXLFuKjBw?KRhujhAGmiv>?(rnv#X9h?KnsKyUIN`a zg?Q4g1V6?TKZEqQ#ZveqTm%pG zis0tqTd-*Db;zE~Wq*%2;IXs;pw8KWJ*U%B)H{s@cY<)d-VticXM(hr5gD<~HC$`0 z!rLSv$3MYk@)tQ4g6_5_(0BAGoGWKwPqh#XU0VeQ!>pnBr7HMwbA7p&9T+aq<@kU~ zzT?O=p5#|C;w_Lvjy*%tAm7TgRf|EiKqPck$3V7;0-3-WU@{{a`qHK{DJV#q$WtQ3 z-+?Q~RWKKeAW=92G*0KkqS0$$>GS};6wgAoVjNiO-(*Ud2y9EwC6!#p$&uyK_=lYJ;j8W!jiSNT+Ca_uF; zch?~@YuXLwHhA)tB_#NM`Fp@TJck^gUr3HEj=^K!>hMfV13nsaq_=v_q37;TpnY~Y zqFT8s%0zPYM&%{^823(k){zlLuTFzFR6QqJA?6_65DK31T4=K+7_Z(N1@+rg5zlh( zjj$7b(pE(6cqQx{lH%!m4p5(KHZqA_f^c``HCS`$1DuVRMNV;8rcKdGvP8xe3tIvD-AJ|k33+6L|+iev%O@>*f&(AB9 z1_38{bEXdD_G>bSpM9nryv|SpZa0}uaWxQ}dK{L{m&b<|nwaWujO1cG`EWb|g69h0 zshj7a(8mm(O;u8l1)d6rBWg#8U&-_VBP6w za6`BOI_jU2<=1#n>!SjnV(O?gx`z2En?@2hbJ-JycOkPr5e{uyfvLh2hVS|ffp#tA zk3|A`F*Zn>-42H{@3rAfcO9eS&!llss26mk zUl1*=izK~pCF!z%MC(uL=kFGKNWJOHfVO?na6-x-8fT@^Yh%O6qUTERC+9J|qEg|C z9zv6Z4&11uA>Z8^hMMKEuHyvkjD1X0tdCO{Rt)ejEZ3t1zm=2qUyb0yxH|kC6Jjs^ z{ed6!|KQbC6~t?370K9IO!$hscrhkPjBB$iIn#3)w@g2enVi1g#c35wuj$FwWk@_Ifr5&&f&BV!(2||&s;jaREK5@ipVvUW@deO6&T#L z#ekSSc;gQr%YAzwESJ#5cdMv{`B9WfoG|62_=CFJ{gn{)PV|l+z^RWOVltzZPS7^8xc>%|7b7#TB|XSCjd&Ad;Wde2rJQIhD86 z>@p0R_JPqkF`O8!Vb>1aVh`;}V=dIzkn<4AT!?XEqCa28nPQ9ZU(qbwmUEF>u>R8Ze}$mgs0)@ z2?_GAdxqhpj7(}(&^cxw{g;^-)Mr*NJO#5Ot-wPl5*#!`InQY%mdTyLx;+^@MY&&w z!tVoleQ$T;=ame42G@X8&jHwYHwNrn3g8IWfjxQ5m9=UeD5IG#jYZxBt7tpems$j;zIH%a(<7*!_5>C$=eVGB zDR$8oQC7hAIT}Bp@$$%NbbEOk!&^4u{8D|;+4Pi_seVB0^$t=(*DIOrqe*bUIG=ou z7GWlz=i!x4X&`Xt?*I9Etk=mw-f$6468?klLq{-T)eod^IFWA`SYq{7iI^s;Q;Rof zP&!$+s8_oOAuX>5Rt9~ARkr!4R&oQET&Y3Xzz`TMJIScdyh3hqGkU~|6l`^GKv5%G zJSMAwF0G%yL$||lz|EPO`plGie?ACCbCMuBVi&AY5@Pk+1-Q&!&TDNp5B^L|B*9Dj z$j|i+Wc#5nod3>+=={h;k+X$pQhgmg{Yr5-BohrAvvJmD1zdKS!m+&zP;l@XqD=-e z{L84zmZIzMP1Moiaw1EYknY>hna_{rF+1Wv;r7^WEbnQ@McIuQTHb(+at=QDQR_7p>p)Zm*<@Lnn_^OxEOfm{)7)A6LIqB0-HL{-^>I zt7Xh+DOnPYkqGErz6}Bkf576a(zs7m5$$sY4N+R1%9|`hPn*6T0xl)N*f&vJ_$eQa zt>Q7im(#9YSjWDL*~3bZ?QHXraOg`s2T$L90mG$DQ1R;~*I(}qTcY=yfT(uK0 ziEYIDk0zYs=F?*dF(~isj|P`7V(+sCus9eFQYII{`BW7C46){VtqjmpBp9`Z6L8=7 zUZiD@;`BjX9P|=F(bOWcO-m@#~6>q`0D;e5=sMCHmb=f3_JDH%XG&-1dn1a#D@x zjO`Ib3E!!IvW# zsCNX$Y{lV+VkA6F2>|cdK$!pfAY}4YF~hPS=y!t11*S3yu~FneOgWi1@`em@+5o+t z-{gV+D9`uM9s|1zYk3X>g2?||3Kw+aK{(+QYP2QezwfK_Gg2^kTqW9jff zZ8OS$R$-MloTDsiB`E&149eC!5_+wklFVO1uwQK*I|~fi+YX;lR(36T>Q90Uu{VSX z_z&kV{fmPZCTz8q6kow}f?py%lTud8W<&z7Gdsith_^dS!jFZM%)SZ21YIPjnOmej zL;+5J=7Yy-*NkJTokqOhp+F{L1G2B?Y5T37{gcBAL zteQ|Iv9qY5t;HpXg{T7XB`e5RuM(2VoTQ@p4%9u{m;6qrV(6;O1HsC0IMVtHeWpxi zEe8eIZMMTi;9MSg@O34etJn++p9X-(t4VOEHyxd=^Rec2K6;0xqQOi*936JR&2PDM zj1oa?;&`xI1!I`8tsf6@`VRa2m8?g%F>4+|vDKWX>i7a9yj{2l&vIB=<=-JF*Z2b> zXM^y%xFcFds^GERkBCBcGO;$jL5!Z}L(s4|zE@j|LOxYEcI`A;bb8=Cn+;f-u7;s8 zDj4zO8p!7c!p|FA7Jy?XX>Jq*oB7j0X!RT>B>e#6oa|3xMSA&Y{e;I3_u<2FX`|X|QKP8v->l?|V3?h-5orE$XfpQ49X9iDr|&Qn^>p%&uN~z} ztbWDQ8(fXA^ThDb)HtwSkcjuPlkin~4yXAy#)Yj{Apf8&w5C*Js9y=D6cnJFn>#&o z4a@ts)sZiL{Vmy*yPi1jvSLbFMVQM!<7k@&Pe?=gKHPV6IjSvC$BAGqs_1?TFFUP? zI&YSaja`vAX1ES@>Kw7&%@nggPs0FbifzyK)o zShg^3o+OS&<4>T|&x7+1?8ViK)?&QqQXD+?8g@?-L*M0-F)3#ZoOf)2q)Fk#$m%8c ze^`Z?MwJ-1?iSu1zJSvwd*JZV)v$r%uX7C3$dAM;w4@A=5!Ll0OS_cFyXp;0y^=Z^ zZ~F-ePu_#ithZnbuQ}~<2Oj18FFgSaTXuLYtG(nfyJ})F{QUNel!f;YwZfIOdBkZ- zQSCZaaAXpfZSjkn?ek$_1ULH}sls#dFR*0kA~LitnuzzALbG%-PTPMHGhZ&kr*9IV zd}IUoYKOxuyBfZ4?sdcOzA}c(xqWOGr>*aIya&?xd$A;$<8TY^!0+5>Q0&|aE`1{K zb&VcK&fWkWhh~CEwIV*aErEvz1@QcpEw~HpaN1oJJa~U6u5mNM^9FNpW_&0%cAmn$ z!P`+~VHWbbLr^Zw8E2?`A{#Wlh|Aeb(pXuA$?tnPLEuk}yt)VLPtL*S!H+QZ3_-+0 z91NM?WYgS@^rNd0OkRgGQU3J{TSza8)D)rdxB`B;{SaIZKIZOiH+(eDfzGTu(5{k6 zQjesPnaW#8olqpx*Hb`u+UhgJY(M|Y>fd~C-$?#PPVXA_WCRS87hp!(6eC@SK2~c| z6RUjOkxJbX&fGri&J63^LXC-yD0$HuABBt3t}k1d@@@@MulfKrhR@^KMcYwv{#2uH zBwzuK zjgvuMUV+5#*g)LkR7j*`06yS+pF0_@#{Gr^FE(~DHL=&2-r+G9t@RXeEaegS+_{3| z(_(Oc@LJqZZ@})GufqPA@)JKe-6b~w46_~hs)tSJnFvyK0-VDDdK=~p;(bgyZrPS_-~VL8sL7A0vB1^TGDREi&TT-x(kc{q z@&TJK)M9;e7WxJSg4t|u;+GsnnjTExHg7KX<@ZNC^Zpks@oI(T+kIi`4i8?*y>+}J zce42+x7I_^>#2~iWRN6&;Cv)on_;c07#{9g3wfT|TUc|$R!A(#DJ;$A3%K5h3Q3ST}9@^(Hg% zQJ%R|mP*geo7_xH zbdkhzToYIY>M53x`o|d3PoxpC-RH?rQ3c!@nvAcLmT)uoKWx4{pH&Lk!p_*f6c*af zfi{aaJR?(1@29c<=6-hQ3Y1m|j zq$rh9l9EvhB}G!7N=Txi;dg(3fBDbjoX1($^%~Dt4(k(q+xi{&N~~yVibEgT-N(`2 zI}8iM+zq^q%iUh1!lM@S{;?dR3f5xJbYrxi#`a2A%%%6d zTwx^UuM9tPpOUzbxy1Fsd{BShMSMmb2zwrd2W24;+fW2|^9#8Ht89$)9>p7{OM3$y z^aX4dOW+knIZWOyfqVGA16ME|pVZ$$hmsbwI;jjIsvpS6$0wxjhc`Jbl1)Ae|0T=o z$3ft{7+T4xu`Gp?Wb^AO@-x&8grElaYD0l)5e2Ox8F&(80#DlsRP?5k4dVvz>v9Rm zwrRk@9zl4X`jb>%55%TBi_kr^37qoziHKi5WtYaGS2yp29VreV#db438p$KqW)yZW z?}M85MmTLb37lo>*t}H-PO7n5j&C}gXTFo;oiE9<3r9(B#D1>r%2~!AO3JvQ;TypE z(0Vu$#f3K-1hRe$!Emr2d>svg>21ow5KU$~ew81FRPW!#{yzFsiDLM^ye}wZ(g&6<7>WnQ7osb_zri?vY06Tf|YO ziKtk_l4F-YkVRse(BRh%lTl-EsdO3^>=MRzj4qBpLtGZ04kcyNuq=(i>Xaj7=k4os zwf$OZYd(kk+VY4D4>0g9@htKmx`@}$k|@`qhbvvw@z3UY=%`|h4j_UO+n<7G)oe2RniIKVTSgBi zIFpfo2E1mx{ z{`Mh+us!725+026iFM5TRU4U3)eg=<6PkN>qKTXJeGyDOxe6IgJy5q(kx0dCqJ3WF z@D61rV)2q9{U%v}Fz!Od!35J+dDOeKUO^Oqy$iXjzM4)IkfWc?__0BD% z=148szUUrID93{_DTgTI1)M*Bw(%A9vH}!ZSG}e+_B6wfM)&A%TkcS9=hW$k zaUxjNIuCmXSe?pgIe0Tk;Ny4!Q0pR@!TA6zd)VWWc=>ZJfC z&#-}+GHK-CWCXqcP$pRIiiggkU=VDzK)bUP>xE3m#;W-!_(mNEwynY%p;E}&qyh=& zB1r9fP4J1G1AkoAA&oWHUKg8DVy@j>Z-E-PsyPm`Pp}@!mifRxRZ0E{o+C-);#eM^ z33IrcAjOi0%l3O`{k(QY>)L{ zS|B?3Aw=ai!gCo7jDLO;CM{Fp`N_MO{x1c8?LCNQpJu{Ybt9;B34;AW<%o90NayFH zs4u~578)4vTn3W_^&me)2h@TrV5dkQydL@t(hJ(a{M-=wy>4fH*(|)gK7Ky!yH1?;7`w=`&;8CRpEHZMd5;oK zA2-0~9oBf{>MMqf^e_`iZOoC6d7NXPqIlUi9eD3zEO6axWxV-B5Es15;BEaX$sKDb z=6sk}jb&qA*esxqmsRgETh5$gI5ZXw#}jygC~V;O#~OJO`;lri_e++aj2xI7Ul zD_)|gELYE^1nxM50W%m6PSzTDI{YPk*JJZfPb1N&B@(yyg`nr}y?79O(fo=XPK&-o zE012HK+TXV_{WR&29`qCIRuSHdCpo-Vq7Zpf;RSx#+8$S*lE2H1FyFluH`p1_}wwx zaC(y%_;!htiW@hm!=l%5n^gf8=AK6RdLHx2PnF3!Ii2Y;T}c1ZJ3;rn5+X{0gK#@O z4jc|00g)p$uvGUss2A43R^=GLJQ~tgDZ`Trb&L!$$B8p5@QsWyT+3Vq`!^T?bhTmU zR206xW{33b7w}$o1S-$Y!|NMIK((U?JPj|wo|PUnFG!46O~rD>HIC51;dQ*8&26-_ z6zl(d-32Q*I{~o27seBgnC9e%dP-9GuXz{foOU6i9Tw1^zZk~DY~f=c3%sWq3PdAHM8~LhdqWY#LF<$ku!G58h8I??egppI;&h$fsawXdzbaya%U? z7eRk0K<~9)-s5MXJp1v*yaOh}FnV7O>XH_N)db*Jj5Q|91Y(1l9H9PZvh&9bSSsI0 z^5V@PtiS{I1ri+ZTZ8;p*`83qZqzCd#WTmw&OP6q#>sPlI$dC-GT zH6>BWbpuZAjKf_P&D5D|dGs@u+wl8B7%pU)8PRb;`1r?tj26Fxi)W-``40`+SH6i_ zKbK4Asaw(^XXS`{Ni$isIR^PVL$M_7Fgi!~pDa^hvc`^QHc4M1)geum>}Us#`j(> z^LZMf_U#_w4K}{xh4<^CqS;;`R2nQCRppvx%pxyFqDXGnevF7*jScMmJ$hR<^Pksc zM&nZnbMja(lSys&s`JE6~c&oVvjNPV72k=Q3m|&JwcHL zf#tZV5k6OSvMsEHOsy5^s&?X4HVB#MNarR zHd`By$)5<5kwKVW=Vh3rAD_t$yKK7kt|4I>&w$-PF}N)(f&C53aHt9Kh~GR+jr4`u z9YG)vQv<(arC9%M1w4)231OWq|L}Jwglg;qH3=WQ^l=&9_tV5k^XVA3*9^;kFTtop zCp7-f`dUf?;C&~Lo{{v5=Tq!MpL0C|n-W$Lzm*#Fa|a20us9jMK68e^1t~b%P>g%d z)u6|Ja`f7PCfZ|90&LhG1Y6ZkfUWiccvv&RrKi?Y7jBJEYyZwe5oIGRXY&AwX{TYh zp#&TrOu#auInd0%oLUiHML+0oz@y9VqS)Ps=x}-r%0~~w&h9Ak&BxhTef@Us)CM8m zWuXD6V$ZrSjnu(*)&iJx^8p*#9(cDng#2rNNErqmd+cB?>xJV8qExH}?iHMmZE@ZxP`MB57wpD{o-+97Q4e_r;`og1uDH~N zL(hVTP}{iz=hSJkI|wsT>d!1}=`csVi>!Bk!HZOQTat`{0CKII0hWpl0c(B0V3{P= z@xOuiAa$&42Q*Ihz@@TB@XU!fq+}k;GD{pMV_gcP*pfV!^~488e3&TRM96Eo#*VEjn?&66(mb z!RgCYAbZ^wa=2Y&wNWOin$`;sN`FIJl@%rmx-i-QyqVC}T}+2SBjgO^L*{rOjGWKJ z@4g|}mwW)NJkRpdcN@_mugoaR0#lrEa6QUq=;PI0c_3ER&EBJ)LGOb?)I5F~dmgRC z&pBaev=woNlmu#rd6WN?gz2f>d$@nsRgkzfA*4;jkx&N}AuanRNqu>fJk8Gs)`!RQ zS!@r&IF1wJV#!s}`^$Crn1uccDZKDb0u5N7@5<17keqlASC>6OYVa>+*}cbil_p?w zaES0viIcfJL$v=OhY827!R&Y~D6lLhEfrsh`K&WgJrxL@f3LxV?L81YYX~-UltOqP zo2z(m5mdi9LG>1v*SLaU#efyQ-06r8_OtMO?-+zFoQ?VciBRHriuCpg5?k*B^y?Xe zl-!n!;IV3itoRf{zNpFJBU3qiy|oYSac7`pfB@?w4MFYBG+0^o6qMQXZtsKBB-d^$ zx%9V$y#A|9ZEn${R2{N7X{VHs375p%Q`2zozuUx<2O#?NE_fxIz>B5FiA7K*?J=2* z3)s%PUP=fiZc+u&MW?|+Jre$;7#S*dEH|DlV9sFTnY6pa zYdTj{4jxO0(_#>+=Nf!s-OkNf~zGcEA(#})8u{bOPkwH=O| z%_Mq#;p{zY2b7k*q|F4klMlPc$mK1wVYxv9n-LbmShb~WR$8C@hX*B?1eX8z14n+y`jf}mh>x9_K6{VK}Xw)48NpyEjg z`>YQgM|z1&i8$`Mo`J2$CQ&xjipcV;$wJtabot|;Rj^v$Ev zY}nUehZw9i>m+HD;_Pm;79)Im5z}?JiYM?ni85qyWvU6KP;#mVCPU2e@w`qBZca6d zJ|M>(6%wMFLGGMu%U8R(NZxk(m?e7}KfO_QPFM9?Sx-cfP=_s)fJN+mYLog7WZA%UqLi z@M|X4q_7@sLphpM=doNNnxxhWkX5#MBx6B1EX#XAtWOAlEt+Gl#D4sb<<9Qa@g_dO zk@R}6Eb0TR9bIpVA*KgTk+kBosNlaBb1iI9Wn>8+N%Fv;g^_qGUhz!xp%e|cLicxaq&0pRyL42;6q|L0ZH^$|Wyh3uIWkJ+ zdl-eiE~MuiMRZ%2lb)s!ta8f1B-=uCTR4j+n^sD5!VQS*$aOS6!1iF1`|--XE+%46 z2O}TV%9tJgLkzzjAeldliD1k#)Vs3gv(X%m__`wF9#x7v?I$08V14S5Cma-Z zIEHiX1t6UB#vETIJoHZ}J5+f6hl=83_-WQ8mR3K5{o{OCJDv}_PATF|@9$9T+Xaip?m$j-4ZJDgv8;Ac zlz1tOF0oVKilXq*_4L2q#4C@=(TB}3 zYznHx)oaSwPG1_jE#)JwX=~`Y;T6QgN0tuXuE=vvPo!_Bm%*yI<*+JzHpm`Q!L5eV za7W8b46t|wmdSr0OyLLHz~mt>i|rwp{e#8VD!AFNFLK+SEZ`nuB8(WFbmOy^h_U;{ zd^SHS0daa~N$z|OX>N(2ZyfnUZ=mLprIqotN9#E{ZNUlhRj`Tvz3~FgS^j`4S-hOP zsOv3ffq)#D`K*xodLoKv@I`fq}naWc>o zp$i`}V%hBP8v38ndMf|(ZE8u>TV9GqJooae6k3(~MWyFRQ`&RpkkTa&=-glaw5Clq z?O~`vKhzzg`s zPSDcV>geH!#kBA6L3;S>8ZeA$CtKS4>Gv90WM7#(`J&)VKWNd0D)0SdYLx@|8K4CX zl23@R&JA+HN)W>ThLW*OJ4lt?6cuKEm1>!lL+J+8QyyM#dCxBG;aR?0O-JAS!81tC z;^jrprf1g~^Tz2|s!Pq2ik2C&>(t% zTrWMxOX3djB!YUVin(=E#rkdZ7UM1C`zdv*vbmMZy!NEN?6V*{$7IO-dtHPc+y%qY z6J*o6r9|<~XCQ5v&~CE_Oz!(rHT7z|?AgnC&!!vlJ_`QgZkT?8=4~n2)KfrYz4=r-2sVX+@<3 z3(%Fz_ESrCNbQz`E2iZngXq^otz@*OiQ(>j#`s7!Ge4To)6L~TN6ky8 zA6^$=)aJLK*zXJIlBo&@?}))=z8vzZp&NyY%5k8-1o^gAungsU%(9U>=5us6EG2@t z_?sDiun}N%Yc-gIHMR`@y#&h1+>qxux|ug_t&I+!jd5AoJZz3J#zm|TKJj)fej9m) zQyDKYYC|uoyuOQHOP-^nK?|1V#=$B7DIyrVfs!%Jf*sl&aLryE^_N?McThKx{cxG> z&>-?j0P@KiprAo5XOoK)S1CTwsI$ltbq5)gW;4vYO!lD^a}wz(mW8XMh<&fz@Md5X zCS77*WB)ZUW34)|nAZujf@IKPF2TI^0=)k43i>q`;bhVzyxepXe(Y`p({8pK-x2~B z>z;$Cn;DFrSpuqJEU3Ao0L|x=vMi$pysO>>4$m{dVMi(ytf|Ai``56%B?_~t1cB!+EIjPc9X*{n{LM$X6mM~<%4Le=#naK7CE z9Cl}8x?(;)EXlYuFGGhw0nP|DXf)Md3`?Gj!@C+0bhhpX^Y#SL z961MuGksuIsx!O}JOoxYTCg=;9_;_BfKl06V${BqE?*eRvT_9R@`g8X{&Oq1o#n&V zy{wdcb|G##@R$6YV*9NuLqO$PFZnsD2#1O^psL*jgp^gmp{cq;3F5*OvMFzPc5OL*YU9#M4Pn+>&A0+@aEEhr(| zC%O@iTkoI2TApX16w@6CRDg5E zhdljKK~IJ*AtIq6X!OYgd*_Cu)w@#)ksAVPl-kRel(1|(D(IX=o1b~;zr6?-6~$rFIv*T7Pzyu1 z=RlL~B5G$$G~8bi0>vx~C+gZ|F6WLcM4ZwG?Pq=6-q{juwGz%rwP(hCRb`y+ zLJi(_zw=Z-zX=JlJcy0V+x?|N|5_;JgZ^&-IkJtQBSfvKMHs(V=r3$Mn5Q${9vg)@X(?#>DG_yhBk8LJSybuNX~JPO$nVuYknEHVS49V) zoYKOrAN{d&*a1890&#OeF{WyKCvvZHsFD^wGGnBy`k8!cMPHsJrk4TeKhJnu?p4GAR!om2yy>U zUx)xt8Cu_@Ve0BUlnF{iM{q~E94mH<+Xd%pEkO+p9Xuqff^Vi$@xZnHC|>A*w-y^Q zE3R^w={M#x+L6+iDhPR+W+>bu!^^iw(>w=PBUAOr^;olw@34pEuGqZAt zAveFwqfG=Z(_bwwV#n{}c;n$ReEmEN8ZNMR(!fsQfA|rObw6eA;;df#gx{pb`wuhC z=_At?N5Nt7I&$o)6UeC9GbSg@7+py-hCmZ_z}1C*vc`tjsml6022R-U9Z_T30E&eV zp=)?Qj%<^-3`~l=QbHSYty122G4#vUfLb$L0iJ^DrO}OfP3;J)K#uk;`IAb6f zYrh6F=Au4~zJ?;Lc$q^{bcG43*JcYa&SK!R4I2bW%!~>&EC@J{>-~M7fuXT$t-5k-xH<{(T zc9Flm(d6=84vBwQK&M*S(_d3G!Ns+T9E{8&7w0{NiZcXwn-svx`6on$%A-MoG=6&{ z!5F?4VQ%sXFthlPy@R^LrAc47=#yo<^4Jp2t>03db&3x`EA1V8TgqVC*ABcJrD$e%$Do??ccN zPlE6j%HZu~ih=t%==snSMNaC$=N-#o!ZDfUIXlBD!}Xx_Y>@OagxULoKw2(* z-t-D*%)X0voUY?KigwlmTDyLg;x(An@a1I8|teiz*MHjea&#Pfp;_9V1L@ zd;|JxSAbo*F_^xKgxrr#*s;YAYlGtP-3xu_T0cM^pY@8&6U?M@8o$zq+;!<9(_Op> zJ&GrMe-<~}kIgQ%5uC2)iq+x^nBSj;nRr13#-Hh?#m~jiySB_Axj9aF;dUk-brE7x z6t5GhO=aXu+&$vIW*zW{dXf66R{Bn!2xHb)fd1MgxOk-~UbOuRHkZDGXXG(v{6GqO zyv=5=>;I)vTRSQC@l$;IyUDgG9$l?eN~f5}fq`lYIlt7C?ftWzRHbXA;QAn`(cwe; ziJ!nHAc;Xo?7`3FDm=^(z>7~r;bta3IM&I->OM_S{jm!8U)PamtOi*Wv<8~0`QWHl z2^Bl@C0A^uGz>gVKu>y!+BWqplLQ_;jH<{LShyc3j% zZAeU04KZ`egx@Obz-R4wI74aUL|`~p=Y*jD-gQ`^6Nyv_+d1IO!CjTR$aigja=dIi zaZHWkol2LWwarh^OBJM1_4pq!WjVKv(|2Q}loa}JdkXbo!Wi!Ei)zEwSfr7S$1+^7 zFR&VpcvX{KGqcF!1_mCOFXx`zvyRtwYmlpvdyo8Hq7I=?*1}14D{0O9{dkIx-Rk;w z2fmgHvTTG2*w;`%yqC|1kP#E`UVH=I7hVGX>@;{Qumc+(Z^3L=2W)yd7qj;oVp+px zy#CA`FI_l|?ZN!m>&5uJ4`DhoZ-J-$3(%^bi5ISGGQDe6nAKNgnaiHDIZ-!jIJzH3c@NFX z+04gv*w;G=v$wo})&1SD_zBCQt1)8vcx&H&-UUmHT=mo7=3cfENVx zv1^tO@}1d;N7$^%$p`6Z=6?WE_9(%4umxOdbOFix(V#k(4tGZt$yQgEb9YSy9Ar*m zt6DA^Ti-x;5q+RO?1GxK2yi{115Ts!p~dVV*qXm0&fc>jasMo+T+hXvr`u6XVk1`1 z&ms2SfuO2%9?mA*M$gJSC^%7%^RBYGDhCgNRQbA=<%I_ zyR`JMGrkbs9M}O#J!VjLZx%KR@#7B3X~@gJ1f?7S{3>ID6;wYwJbH`eW!6LFS$^)Y zGFfgdSfipCEP)B$eJ=U|ly{ZA28*hX`tAgo8z5aBXf4)L(yT>?Q4IiO7Ee)KFBcZy{oC&+MmWfu2W)|;OAbfW|Q!YDJ zQLVz+n15NA33_P6*i1i8eas7|1q&qUp#_)lao7b6TAhg37CdK!zh7qLZDW~YxiR|Z zv@bks1uhYqdW@Bulo;pqb z9qR)riC)Ox4s4cFhJ(V5L>6q4PQkN7!#KC>35GKlu;IZsIRD@yq=<}y&z)eF$x;UztBzyS zK2=yNew5sIR|l*2K6Hm~J9#*A1-jCz;6UzQIDB(DYV6vGw+2_^%GnXP&utEKVABky z$7KYsk8VFZ9cG5`$HIq@obkaM z!W$5#JK~1mM{qAk6ALccvH9lJ;PiSro7;U#R^GCQI3G0&*L}w-5ZbT7!;}5KKjvGK$_MjOw=xrtgdh_spX{%F)b|;+~JhpkZlTn>!AzCY!jn z7yIdn`3mrGjxd_Z{s%h`e}uE)^O?il`b<_Omzh=20Y5_iz>#V(j4!rgs-sL8mUYjJ z-ihUOlr?c4zv|&OMg4|t{YseEG8Yp=)j*}Jm(c6?lZX>__(ixG59wUR#H0E6u`d7@ z207u7wmk%_*-q(Jd^U6vJPS{`p>R}k8>nf%Hx}0XNIu^;0GozWFt0urdiQ?@s%?Sc zd3b9yfA@cSvMwys{si06S|EsmU#6o{wl>S*UWyNG=b%=oJ|?`i#Sxa9=6r1-tT-hP zrmVK{`=}dy=B2`7KX){|Y>F3!O!1W23Q}dchNKj++lcuep!}E+wqFp&5l>CnQ%8Y# zCm$rJA;yVWqQ)X~mL0vCe0}QypMHOVG3SSPa8@EJ>RDk(@I!h+Uz*BVu#HncQ3GH6 zyx~zJfeihbSg}?HlU>zu#oYpAVrq~x*n&E5W9ZL^p7R3st>cxYo`8w-$KZ$FV(=@~ z{;A-M>g_v(II6JOYmj ze>(7rQ}v%dSJCo5yvVo>T#*deyJ0ha*~j`z8s513x*Vh} zm<0=Wxxm*GI&d^y3Em%iOXf9&(!MMs?PaMkq+ECe%7K@Ft8odUoDt7VGs37#iYV09 z55C8R@q~&vR&eHHxPmN>w@IUmr8IcDenFS=gt#TmK{xTCQWwH@DK#fRtk z>4pv@UmYZM8zac1Y!~!zJPBuS3&OWFbu?HP1C&Atw8s~s?t$xQWmbz*5e;aTkc8{L zo8n$IAKI~CjK*P2Vl?hV_Ch`}dU}QIn_Xo#hImHHC2Ss=>EBOe_bBEE` zFIawKHnU88E%TDLz@MX9n8T4m9Zo$XS{B6|xNOIyKH9*mzm`dj8aGf``kByX`vh9< z%*MguaI*7s0y*kY%d%F|F|48jZzo?xr64Ixl+Fb6(=`xjp9TuKmk3!e!0q^u0G??S z1bZq%>$ z)_tsQyeA(o?MOxA%qQquE6YSSw?Xy&+n`Ts!1?}n;yg_bqSs6V&}TW|Y0KbU7a#n~ zFCq(7bf6>I8P<2j;O&VZwC)K*%>xnO{dW$?I?jRdfMl}3u!TOgJB|eVpTyI?8*rdY z55E<*;H6DzC^u+<<~|wZhx`E|9Uu+6`Ar~u*Iqb!-Wfu#WZ|{WOjI%rN9Vx`Jj|ZE z#P!4QZ zdKfCAk+viLJy-}aCq?1ZD z!6ujyHv`pPs^Si?!8~1llwC^U)L(b}HKv6{jb7NKn~#4wwb|X){j~YR^Td7b6>vZQ z04z6d0s?C=r|UQh?#)7(yYuk;OmF0KIf?Cgt)yBig}f*`O9J?2LsYdQGzm#U{JSDF zFHFVz#o-t+dj+xsqf8Xa2Om?A7oJ!~k%?>7T?=W+DYctc`WzVY^sNgQvG@!FG{-D&1 z3%K>gJQTKD%efz@M(W$bh(p^j-kn*9b`rj*Eu7B$iAZ9kH^wlpZwJB9-*ZqN`y9@l z;A5652{4{wO3W&7;mFX}ImaGdRHUzs;hW{-RQbuQH{=<|B>liuM*AdJM_IH?Dx-nO>o!H1DQA;B_F3m3J#`|TfR#% zL{1!;5=oq$F`Jj;-e;(}sNG1#>I{@l9s#S>e7JOt6&jIKXcCr%YM)-Ck53QYOa6kF z1q@-_vX+G1@1`u$PQvX?N8xQ%7J!F5+WpkR&0I~~w)82O1*gE3EvkSw_rTDTIB4{H z51h(8;0tPkSJMB1?cHo#Q(J%r(aG5Pbu~;FTm$n!IsB>{h^wY-aaeE!mV+@I6>uUN zTd(rW?qz{O{Be-!ze<+Yv0S3sRe0(Idsa_m`A)8ZcrhdcKfMV`7%><*qvQ;7JX65$R7Br0yYQD)-3C zU6W+L`dxDFZ~^h!Fi1Tb&Z5lM_(?$>G1$C5pc;^|tR8y2}+q;&0>8i*5J=u7H4#$-_z)R`aPtqOU z5uF}B%rw=-N^LLPcEFRd-4x4+Dr7K0Mti9iI{^w5ZD?gl7fejG!{>V2u_#oZJF`cS z^LeR1&uvT?HMP`n$c=}?+z8Y@mxCQDm++@p6sF1~p>Sgc9ySzVo_7l|Rc9u#!c7(A z$GNapPZe~FRq3{rNy?TR#hbR%flP}R2btA5WaZl>F8ZZ z!Ow%(*Rcbq|F;ugce>F%fh#G+_h+ap8)Wf-Y91OsdxOQTk2e|zw1t7j_#U|BXbB(BDzbm2K>oul z@O$hDl?yA_j!X=SQ;|3^5QVF!9B}JvJA7khgX1|RbZp@mty9uNe=StzB?sT)jajRa z%pbO}vN|33{Te`fnibj1ejn7%TnY2n>fo5#Z^%`f$?_GhfbpIYXw@;p#RkbBJ1YhB z9u&hLmn`F)whp6Jdb;t~zc5oH=53{||6PV}iiwZS|7!yx>W^cIxQ~FMs zdiuU_V&EetW@Jm8lGu@BS1|3WZ$RDOY? zZ?_@l)m}I|KLb33ev^Oig+RiH-7(#~6t}f&;Oo|%aLX#0cyUB{=0ai+X}*j+tCJ_g zvE9^zkwVI@^*XJ=Uk|c(DnLM~6h7+ZaViR=d8gF-xlif0Bq70=OkNQsmyTh?4Js66qMm&TnpjfXAZ*O z*z>O{s~L%J!B9gF+O(!e-X2QwpLtsAZ8T)=v#V>nWamkiv7?^Yy zl~_Km-4_ouij2pNRf(ir*o{2j&_`b%bL0lc?BSKNdlVkaOi?+P;L@G*uqu?*5{H)H zt@Ug1V%BD8;hiR3mh9cQ!xL9lufTUDTG+s5JJZKnV2wPxh2$Y;+-`Hu*uHqCVQPCf zGzzjj|M{H|{x}Smo1DQrdSy6gqb|s~s<8j-INANjk2tJ~ATdkM5}UXPEL(9Br6!~B z68qkMDd2}5rmooA5RZ>+qVP-mUYvS;59=!~qjzjP?weu2L=)31fS;|9uGfuR zHhVBFxe#}6KLv~BcffIhBv91KK$C-Ecqc9yyV>1&XQfo~!o7&Bak^-O-I3f?{sq8$(*^By0$680%;dc2WS+{@FcU4yh-i@vJ$bg2>ilk*W$*c$6!2iArpdH0Rg*atQYQRKPk`h69riGNu%P4ho!kQPRFke9N#wJrO^=)4`@BWz^JN1TRb_Y4LCR zl$=%~7}#q-K%NMhsi}Yy8=u3U@L14F{ee+h!c0Y$0CUWo-Nnq$!AJHNaDvYf)+Nnl zIe4+;pZa>vd%hFK%l9f8T``SCm$#=-dSf^aUX6vKNq-1^+y;i>viNRG1T4?)fgt%l z42K$wkiLe2r`*ArlL-mO1aPUv30|YZBktO2Ti)uYN8p&G1uWdm2N~xM;*XuWOeOa$VX{-e_I6w�o^hq>T*785R+iuh}$wtZDN@Ry4rNX`5RoldSj|R zWE!ewO+z&~Ic#z0A`&?q2UX>-*=APSIU63VzT(aU691Rx<=(s)zW4{ z8d$VoGwR=u!-uzXNN6*VpVNMlm1om{YikN2kIume9SyvRW>!bP4!s|fVf5uQsCWAc zj(dI(?fc5`%{&5RF2>?J%Uo=cd4xvoAt1H>5=5|CkDFOKOdN>JxjO+vjn#lXZyK6^b-Uct2aT#g7MVF~O$32wsCmG<^ zu?0Pchj6xSJ$!lhjLdV_g828_(E4=>cI|qN%D&QQQ{s#t&7$yy$Q97(8-V${2Vr#Y zHRC5K4-KuU^_-l3SK^)1Og^611^#5VzZRSZDp9#0&p8SShZn-=t{xI+bbwNr{+y>L zwVW2x{>|IqCP7b)ekZz;IwU{v3N1BloE&WWMdA-u!;?3BXufwE?bo;h`Zs)pWv&CX zMyWnM@vMoij4UOF=>p_RQU_h^tA~bTUtmHz7uI&SKTY!#BP z3vR+6bu-YuvV?N=3;=$WY?c$oZXxiWg&(Vaz^cE>C_d5*qrL^;PjleVKo{*(W<%Ri z<+OOD1AU~;jF_%ypzT*5;TE}N@Qf}e7)!FAJ+;&fCOd}7!68}Zz-)J>i|k|CPd(x# zKkDW_TRK9E-#CX$UDo4tp$I5{qC{tkc^NsEcJn@K^TGR$d9Yk20E!b8@XB}`c1Lux z*n;y0pg3KAUVF9{{H?Ssd!RL?kW8vTGhLF zC!-Z99hOzg7xS8y_I^x?^H+meZV*Q;WT%m?|0#p9VTw0mm`L=6*TZE;A6Q;`5o*Lm zv6YF3B}3{s_Ad~ySaaPOmV7FJZwU<){Vz;eJ>8M6(zYU9;i35D zYdq$#nJ{ks3@AN*mQ)#a(XPtLFn?8-LniQv48gbD zQdp|?o$Xr5lT~8V;r&xxnCG??{uPQ-+YUY9_FjzR*4e zt2!yaD~751DR6<~OeP*L1HK}5bAZ+Q=KR}&I$Pv0#k~MLPB-(O2w&w+8(GThx_1PM zmhEAg)nVYFK7jGMZ*ZRZCoCHMM}YnOUwaToq+ZV^ENzwUpkC3<_YdOak_bF`bsw%X zP6O+`so;6>BD7z71C@%6FnszVI8Pd(CaZ^Ds4>OKAbFBThS6~ah4kOj zwZy=?9P972V~KDVKAb)Sb2h()OgVPPT;V;X|NA$+`oT0}C9;W%Fy#3e0=GwuXhS1Y(OksXZ#zu9j;7Kiu9_C{%f%lvH{q>W z`sg^ipIov~g;%=L@YkRnMXNgSH}4kCtso|wXK9+a-&Zi1UxU<&#oi>?a0bMEV&5W* z44A?cnt9+j#XSMaG-bGhmaezOg-e#B8*d?&C*Q^LlMiug@ojX>E5gl#O;{h1hB50! z8Ke5?j6b_OW)*yaRKB^wYd^Jyp7gu-f2_R?u%+2mp0|Mkgb{EOWFyLoIS3(sjm|JX zea{%sI)lugWPC8-0u10S{?y0rAOBl_|B`qdyZ`*fr~J&1Ui=>qf5XMQeKdce={tX7 z|Lbr6zWtY9)cwt${L=lW+<4Xgpa0~m_b>lr|J%h6{kfmK@$R4X3pcL5|9fu4|9I=h zmw*2!U3~E;yyW8ZKl7zG-t@>z_qTuZ&+b3*&9B-&`SDl#H~-Fe?%%Zgsr_I1*-wAs zE&uhOe&U_apZ~;b->=+w)j#|@H}1aqy)QoIMNd6i{;PlZ$R|Adl^^+zKlk$edzX*i zf6d?c^!+E_-aK*os24x+U56Jw@sYpv;}@Uy*}r<@WuN;ky!&_eKljpK-+%gd|D*l) ze#~nw-tx5{fAL9Q_Td-DfAWiu{P1hO{*ixt>!C-#%K99fhryt(uvt|Fz-}etLKKS7`?7!n{-@tz>?sNB_ z`1beNf8RfT|NYlK@rwQ5`JxZpzy23Kc>i+#+r+>4%~;^+VRx8C^N7rpO| zhhOmyH@@w&&o{pN^{;>QyZ`mKKJwZB=|6hpsc-(Ei(4P|!581be^>wOF8=ltcRu91 zpZL6A{OTwE#kV~D_>b2=apOx4z_Z|8;-qxhF1P@sm${@vr{a6Yug*|M|P{}->j_%kp5 zzxV&@!+vxB^`H5_?7!hFfAx_Y@A=xt-u(k#_SonDKX1PHogch;@rHN!&Wm?@__OwJ z`NCK1U-TO<*za#VynplOJiNd2Wgol$>A!N`zvA+B`+xSXe{ui1AAI4(Q-AWeZv4z^ zKl#yr^@ZPmYFYe`d?qXe~AB0v7i0Q ze|qsQ{BIn5J^vlx@BMK8x8pvA|Bc0O{5KaL|A~Ke|37`sS6%$}|NJ*EzUzH|;Ns6e z{zLn(_^=<|=Of>}|Nj5$J@=pba~JzJ{^3{cKl*zf+JENTUbz2^-~Hl?H~r{;ck!k_ z^rDNu{ZEdMzx|#5(8VYG$U9&B)Q5i2jlcX3{7n1l7yDm-;b#8>f4A;0?|#evbH4aT z_FwY@zr6pA-}syRKf=$uFMIXhy76;&Z{B#nU;CyTH*S3J<8S%ZuX*gd-v6B*d-#|B z+{MecPhY&tKfAto_B;Rk{l~uRf4l$KAAJ4(=bw7_{p-H-?_T`SPyg_Z_y4&6X8+Q^ z|2Oup`=D>;zpeGIFZj{lz5JDOS&BHm`kfvs%QPG|L3oG z?dN^@=e^=$|B;XJe-|Ief8p)3{-CRr`buZ`!Y}-uA7ffAQGkyFbwW|Ka!l`0nuw z&i)6y|NjX4-`MXS{SVkbvy>L41lT^yqd!a^uk-|ggFTW9Y<=s@C%iSrMf=KGuN8wM zcSW6j)jorL4&l*pHU<0;?2Y58DnjiqZo#(6$RB&_g?cpBg3O3O-psrv$ zsG!9>pMJ-;M!l$@<=n?vyqAx<=BopgrB9-+w4p?;xlV1+4d^pFm#NPdgSh&R7M{qQ$g3V&`hWdCj}8c~{NJb!>@JP6sS` z>(0R1v9R_SF`AJh?;7(34hUdxwXT|Tzu$8v>N~tEW9r@S1g^l5ywBA*b878#B!1a7 zdhGfR7I^mzY{42gtlWq(6<-(!xTL->{+OM68FlblpOs6kH4mdwF#b4?!Y3VUbF zyw+aov<=9)ddwPq&oFxC)T(#(6TEv`SAoyyCD&zsZQvI;_ZMSfy&LZ{V%8crd~GG4 zz`pj{f+Xiz1#a}leTI00&bZ6tx8z*J zYj67%u=aKaxy!}sTPLr^`?$T>^=`&l;aqpL;0&zw?5>J6E`sHItIoMRN91Dht*yif zw)on>>OTXl^-RO!?|``zXlQqbxtD#5wOW?<@xmu@V!88VAD*W*E_^Bq=`F?%MQ_vStUzcB|FVEDwHoM+R<3S#flInCL_ zd<941mP;Svg72R8ly@KFGssx_AEqTUW-0KZ0rv><_F4A{dG_=^)Kzkq`0C`{Yx68t zkk?n!)#=OQiz9x4XcLO+Kg#mcs-OOpXAPHyo>P}$rJu42bsk}3g+?H&4wMPIRE zgKyq*si~ldN%@%hc93YZx3g-XT*K1dv&-od2QtEY2O|C#Jcy!dRB|sPkyYJGgqv&1*^ur5?gcL zuXk;1e$1i|SFz)MbIz>3g8~AW>w%9pu-04SyLklaTtgd^_bh774?Y(kZk)wBeR>Ce zCtnvdthHjyg=WmD&3O;a9B`>KHg|6Y-h%brGasOVPEOzc_U`xzSFk7M8T%5e8$2m< z*2vjY-!;xakFjFOVO?YI{JqwDbkB%+Cl$NiiSNVc4Of5~^IP;|R%q|N0>9>O0iW|& zF>CJKolSkx+xUb%f)`)kuaR#-EU4d-PkcG8_4Y1U&pFNk-+{f% z2Yv+w===OzPELL313A~mI#9q41ZV+=wqwTqy8)fu%FMEiV zWpEY$NH4AYgMQ}nPO!&L4Yr^emN~T6jC&yOyVP@@J&3(C<1@bdRq@KDCdycAj}zJ; z?^5IQ)|Q;_LyP4;h9tu_*lWq#=3 z1E{mUCOS}pIS=z*hL+e|2gXOVTj2AN4>;=*?BtevUiI>0KltrZGjh(lprLVJ>-~M1 zt7nN_v79mIISMx6o+qEc&tk1-40!8Z_U$oIu?0jJ88>)yA`^pM(eVqMy%OI$ z^6TglxZgw9bMI&HkD#Klhwq?sSmSH~w&uxoU>~_4FE{Q0zX9tSoa>|}XI`5N3bscf zIkon(hrB)A%iP*`Lr>=|K&`p;_ej)NVE(T8oOgQ0KY>o%)zDv^&(G&R>Nux+j;~*G zW1s7JM@KaB&tWft{Tf>JH_pSsS$*;jbSkb|4bbkGfIh=*w*4uAu54eI= zXP*XNvFlQQN$$8{tq(M#kHW*8^NeN(T$%G{Tdn6SG56Y#caITcJ1`dBk^2O`y~oV# zUx{CJJwKBxno2I=yl3yhTK64iDfj#B&^ENz56fATcmPhm5U+O=#MXqIAa-BpcebI2 z^StKA%6WHVe|&q4Ij|XXU+vS+&O1u@B4VE1x(%@Id91bVu=GIF;Qh0RPR`nZ8!H!! z_6V-XAJqi4+*!lA?CU*QTR;sgh{cYxqxT&!-+=kCur;T9N!~f_UQ|fdIV_>Xq`<2x<*{c+s|AF))uUL%NMLY)kR`i z8N-ZvYt*}6V%4g)jG{JMgcg}MM zzMmWZKHhq7?$?9*ocE-^LuYJ_u^Q(MxZz`uXzX_ex$^kV(TLamyoaX_Q`6zCS@*Q| zNY4GtZJEn?FM&PB-qysNEwSUhBwrRDZ#>^M*J8dm>b8g3*F87DbNln(t3aZ2j;+1$ z8;Hdo;kzpWNPdWO!ze|m^?y;p$#ynHS7GOTX8NHgeaD`>Op~jqb`s#(Xo--Eog(EZWOuct! zKefWSA{R@n#(3Np)^mQ^_41C;w~{aL*4-v<;34uZXEW#L@<}be`>D0>6AH(^~KO)_*4(vyv~@^oO^mqtRMpy_$yavvR+!Pz&#T>O}$Tdw;Z!cp&D_ zJa^$dMXaLFC+j+y$}w7ZCWJoYr&OGr?F`>ng~=F0=aHuXj4m zMebe=egg$~zK&IE4|(;8_5Lq`%Xio_-vK^X_f)H{fR;-wwi=^z{=oMPa%*2}gDJS+mY=9Psuzqc2#Od%DaympXg6YRs=+ z_gwk)Y}SujG+JjEJ&BW??_$L^c{uxba3XFw`&pB6XajETZ9nz)C<|YSgV>z+VO+4C z{1M*z2ISYf?fCKq>l*V?BPSl5CE(Sixkc0Q$4pvhf^TT-D;oS3?&(EuTo#NO@U0W6 zzaSP~Z38|}^&P()@PlW89XfdZB&Swgk}qK0gJ|w6$6hsw>FJEdzI}Y| z9bbDJSR>!S>cb2_}Pc*Zbv_L<34h+#O{&AiQO&v z^*ya{O0RZ>N}i$-EU~{<*z`RUvo=-oCnT6>;6860$1TK)dBmj=Q6jG zV?BdyfdkZ6tg(A0*41*+;BVlaOD^njSk8!VU5D@3d-sf_8iNGhWywWL9QYLkegm#C zXW)R{fwI$mi`t$V%9tFnb>>r_wkdwy_{Wbl5?L9*B0KqoHZ4V zJzSn&eYsxx!$(fu{nZ+aLc9geTVuxN7v|>Gv^Id{1u*M$tO7iv!pw74f&mcGY6UPE;)?9HB4+Q+$AHN9p&{{kC zEO8}1;dgV%O!y6JUu}aQJuxnMwet3{=1h%s33nIkyX_rZflmGwwyMV$4ancy3*VX& zv$xiL#+=Y4m;K}gvHjI0zEC6YS>zjd9>2pGo0C&l-~&7Ef7z$WEw%El2;`uGvRHFI zTYEG(`&g&-Yty;Sxq1+DN9qQTJ(Al3??&IgJM4OwPvgg;A2m78SF!G)KCx>*d%K@? zueqMNLSEfH=n}WYS@u$6++*&&flqzr_uU+4fD6#k`TH>AGYd@tno98h4zt zkG-E!A9G)44|3}aT? zoLKj^&(Oo^%U7&gd+N6+_j2CbgFU=2XISU#=nJtu*ST*$JNuXmczcYV@|?F}U86RK zJN`ZXCj8*A-pv(QpQCfixiaRw_H>p8j^wNr)>N!{QOO7P78>uVB|73QzI+8^M)+ck zZP+7qwKeEVWA5&3G8S8!1_+y1;)Pb>lw^h<6USy zk24goMQ7ik!8a#oZA*AL>^M)%9+lV{_kWK3CD7W_oIOs&YC5s^+^}i_o1oxppU3Wi zdFQEE_m&I%3fDk@s!sn<=8^-><{sMhebG9XeLEOA?0Sdh$~B;73rdj7skPR=a?X&z z)q-F0K7v?SXRc!l)_Ui0wOsO8SB=?I>p8t&@5*_0z+T=`Lz<6Pv&xtMX>OU`$9o!K7a??YhiJNCo3PX*@nuh_%B z4HV)8_RtRB7<5oU0OOi7->7fIO)w&=z}bH=|k&WXUT@A3xT zd=GNVG9N*J?Ldw9Z0ub)PZ6%e{qF(#BQEaU^BHI<5EC@LfG; z++*f~y~n*0d#CCe8fS53%+F9iOG~0<26ayX-zDoe_ywJJavR_J0NSxn%(|f|b6?Ep zrLKS)*kz4*a_Y>R)2_aK8u1l-?buV_d(*ml4m007?~L>N>~>&%PR{x?aiWc&7N|2W zpe?q?+}HZ!gZ294j)&zvH3mM*O6|#*y~94&a}Ry*F#-->tb2BR^R6*(V8I4`VwphIV_|}e?oV~})i`6&3?mPBCwpHtjnM>>yJ%%#@Vk}W~Rsb@nRo_O@QwqXO+E^Z8Dw(ca6m---BiASYZg z&!g)eJ@0?bc{Z^H6@Hz;eJ-wvpIyE`lDFTE*!yt?_q9e}>#}}W=5_}4CDDL8#1$ks z!`nOI{h9AJxC>gKJ*rEjy>d86l3ZNTrPaYs=o!!*ORyRk@ybi_yNZGoH5WH zu%O|Ox&Iw&&!CsHjlX}$nKt;N2J5p)Yp`m^Ou0NutUUw1fKX3d5(@^8bvf5~4)(@x z0q})&3FjG{!(L-Qd*SckoXMV}R(rq=(x4l&tg(IzA|b%I3wV1Nt8YMC;C6>O*AeU5 zVtb%T0$@)$b@obfqH3R=?^s8t@80A5Sf6vkk2-8aA8>(h?MW@Ytu^P0!_3Kzg{I_E zOHQ4*2`}8!egWsI#4WMRDZkFXg_ra1{CE%6-Nl#l-1e-%y{t>PB4R!h-?vM!_8fDQ zb6?Sudh!kX6!r>K{O$Gp`yb}T3Ep)&@C)cMzb>55{CZc3@6tX^F4Tv_x||lPx7pZFTs}BwZZq#IqbK^`g4B)`u43k z`xNpGm~(!=UT&L*>pAVlm^*01_78dP)xDnq>U=hd)pyT^EpYPcg4p>A*51~yy|f9h z);e?cFz#~r2~POFhxRxfCReq@C1yO%2j^b?`%VDwvI6Uh9NG$;tIL6IOwQ+NK7r>R zy~(@$dHV?LAzIG+46HT6d#)h0&(6Pp)X8^po-5#8HTVU1ChNCYvG!7%@Xn}C-)C?- z%zYD1KFB`-Wr$v%3;2n_V6sX!E?ZKj(lJf-4gG}dEPAu;0z6odSiP# zv$Yj|)DfGPZ;Kz~#&a`1%sCTIIEUO7xt6p4u=?N(Ux9W9?t<|=+0(gP6wU`C1e{xR{86*;_k3b~HHjaNMfHl@{=qi5V+siYpIq%jr^w{xUV0-WjI`6CJ{QiGwA2~1<{CI{j zL-%O1WQY@w�Ils@1dio5#v%}Yz^T*!aZYZ8nIgAx>)1LGW6cl=!>=9XMF-k;4;4h>JwC8 zFSYj3`W{>X^Bt~&v9S6HT-F-9pYxuU8gm8yNPGzbzXA7^caL@TnEaNUFsF9tFpb#$ z>H^;tG2anIKoP9Vh!~*+| zKF0V3m@5m+um0M@uc0o#Ec*@(Gej=+XvTiT_6T6zX3-e;oEml3*+X89HA!5t!^i69 z5^vGm0=Gd~a_U+V!`+l;IdjBB&Do<8t93=ro&{@N0Cna&Hn9ym^7#IHyz{A%8+XDU zo?)B=&B~z|ePYoy?C?GBVLsRA?M1!4T$%g=H}*j1ir|;Lue*0&(E84(?Tg0RBXEuR zgFJAx80P--Ph#NbSm?7xzr4R+1f*cqoV}WMncF%we)bi7Va|BvwGC_+jWvDY3RZoC z)7M&?_#MB3p&`~+Tk#8i0COGN^g-1wOP#qf3-^Wr_vm|?`1akB zQoxtROQ?X1J9vG>U5?AK$? z?=1S>*Yn^G@Oe1DID!i7)pGV4`k3eG3+EoLu^-wKID>U=XKzanbKZO8a!2IM+jrD6 zbKtMDnbR-W#Hx$K{qw1{&VJ76%mLiny{tEHz5d7%Z^?CIu)!DR+{0eh+p8?T^}8h= zd3vmUlbm;aMcx{9KJ!ZLs+AM=v3HVlzJlGTU1l=xGg!}L-?hIrmAsk&p5YQWxB8Jw zVrnY(2;Z>k6Lerd_iy;tnm4CDuvg^OCGo1O#IE%{c3!nRY)=yS)?0TM&b;*_SLXAs zaO&-KJ}h@Z+<#U*bNPE4CFf^p$8OOCb_cJ<_t`U9+lXtB2MxI2NnRgK$GVrZJOk>YvDSPN zyN7l5^?m}b0Ao?fyRUuJ7ByPntXIV5+_w<-SZ2mIws(coTAM(B+-olH2(dHSLu?Q8 z4DR6!=G?>m$M5}M>r(5it_f}VjufRKqoOK)cfV;#xhk5ru5jX6r^?uynIsCe?zQ@#deES#R zzG_O$9*MQibB-KwL*pH)bAOk)0wQKl=drJS)SGiY=dou8`dWL8dg3#1mhql2XHy4i zVg)(-MBuYLi@dQolMnJ+Fy0S1=N}8-nS5`bfm4^n318tVHi(_eeVoG{Y68DuE4d?( zb8l;}$hlg~d-7S9oVwA&8CmYTF?X2z=$pR;*1ErtJAs^h4rbJh^D)O$@!A$ayC9_Ey^io|QRatz}fAqkP+0fbF9@d{0O~o(p_K`2}F85U9pC{VOy*x_`Wcbz?pS2+16Iid_ zd+fwL;eox}dxtH-Kln#9JD^@o0pqN(%%>%9fAfAWRk*-9k9!62nhP|>m7ISz<8za< zrkNw|*a8yRcia<{dDq#RSf1Ho)*F`v&TvFiz}n*y-?^R3Jx*W;#&dvI=M3&8yf=F$ z@C+NY_SBxi^S~ay-#zE&sr#J{^G==BT6;N z@X`6~#=@6d;eMGD?j>(Lo@tqVjXlQ|2vCWeTCDI)1?wH0@YQVb1Joc#Y<=R7TD02X z!#hJEXI;o+oj=VHTidXK#(rC3{SH^~UGC+~8S@!AlV{GHvpD-1*ter8@b`SUa+q4J zP;SC>NPxIb`n)Mt7&iI)nXWSM|U#xrB%X2lX zerkCyMM39G=fjMH*dAxN5^z56Ye3wHT{TFdcP_O-+!t%knJ+;AYrHG#?QLFe>&kC%6Zm|(F z*U5V)-L|4h#`v_<3xvjPL z4!gZ3#&6_C3}?-~TKqz;VC#~r#7Qh-@jKQ%wGBiNb7w-GTmk0gR_oqd;LPEk}&T~|F`^t~pV$akReDm%Z*e%@7Uhwx;EZ?6JjWvg5UwG@?OL$)AbCwG1 zksv^Oz;`eCb>H>OqeqYfg}m_Hw6A-r_iN5tHO2uNoaku0C;u+hmbfe$=P|af!MVnH zbJicJb8WRb`x>jQpip}P-n+fLAA37fX!HJOmo<%e%!{pH^tT2)P0jPb8qX%2&2t5u z{W{RMhx@I2$Zg;%`3B@4y1xJK7UnA`@FJ+y+S{3X&d-uddYBtp{Ep@sVxOJMSiSj^ zmir&i0RL!A-uqKG_;UY#y(3rT)I`jCUC)=qg`BGeE;t@$-%Mgqg?lc0j2ilmS;+O6 z_flAaFF($N-+<52^Ld7Q3$EpyOKrz@AL~<}z06xL0!?axxh=iyr`G)ncB74Xr_QT> z+z+`z?tGZNII$d!5h~;&BFMbT0Wy{)p{Z&*l8*g>P7wbGTOt{=wQa zf!ZCp0yf|h_>LTbnsIK@6<@vmGa;}|AI^Gvx?1qdVa|SrU-#@ewPa^)%z4){2ihz3 z+=1E#Jlh6t3krHy=G@DiToAiz;KQAOJ*@Zllva&B5?E*6SdH}^ByeTSUL^_e?Kk!! z?9@CBtXXxVA3Ea(Z_QXqAMUndkJR)a532Rp^I`6}_9}};-kS9u0^Vhf+_=-FuX`u_ z5o^8is#|+FgM0L)-e(40W?9o9bw&sqO*j$rm9%Ji+oNJsp7M)zq z_hWq)YTak_A!ptd$uIO>XRCn^5^e`vnR73{uQ+Ec{FN4T^6rx{pKoACUu-3R3p9M6 z$67C65*@tSEwO!k&K2t!JzFR5-CJ+0KFJIJ+%@4UoNL?#@e7}v^`5Ea{2Uk>{EYeB z{5>%D7z_XM*?Hc8MA9a2I9tz8|+t2)Sz`fRIScqRt{&}qDIv?i$Z>D=ykTLfiGZBx4 z#`p|sOiju8O#B?Jpu=wu>8H{JA20yS3&FFzO z?=olq(TCb{n7w_b1>R*Ya?a-7#v_NVz@LA1U_;J(4Y&phzMT5Y!_>-qf8Oy^6SK%t-Y9XPA~+<%?LUUGHe?Ni{k zX!MWNSnvH&Pwv69_#6Yy+KwHy_(c#`(110@1Ne!bNeE8nTmcfW>>VX> z0Cn!Y?$PnBQSUNVE2pjjwVgOtz^Rj0CmJv>r!O9Qa?UF6sxkNPIeEFlY%cYlC$aiR z`Z#OEgj@i7`1Rv6Da1ZkHP#k5>(nKj%UlEMyvuP<^zAug|N9+VaEts{vDD|>%et!X zELz|hwau7)-NU^Lv3EUY!FS&Z3dme^SbKHs&Kg%NNGyoJ0?$~n_G~%p0_zz)OCwi_ z<<|LDzsM~}Y!d;;ykeZom|2X?7Z6}*0w1K7z5MTnx}3?+J(s*c<2Z}HIqzZA#Qf|W z+MMUD*wLRot?}LqI`w|01j$ND{TM_&IK zu*J7lR8YVX*uPREpV%|8cVb>nP087Rrw^RZeaGyF8LJ&Xqry8^kL9e)W#5EX?@Z>M z?-GoAC;$3u_5&_^S!2#Uy>n-3aAP)XgYP-}R$?{ovGUGmpDxH3;(~3Uq8+o7^KJ^T zuX>+Ts!@yYtOegW#;nZrDE0aQe+$0l^!>i&zSbtzTK91ea}_Slf!oAKxdS>@3w|M1 z*Mac~t5$vs?A<}Y`MlRzgPil*{}ZlfjoK|-!5*7Zho5pW&#K)m{07^Jh3|&ChVR{Jll<|zjPn@>cpjboy@%()dItAW z<9-dO9q&9;#ZPcXWA7u@-uhQqf8Mpm``Q7Qu{qc3N6y-_JbpRM9wpG|!w0^7{BIJ; zCkXi@4>pUY!Mnft0%xtuXK}*cVLe~OoN1hsICJ6-zWvVV)Z1sA7tX%s)apM?z5(|( zb`P~d-u~7HX!xQ4>qj4Z0XffW-q}Vh2dY4`fgg4B9(&|`KK6J1j_wM-fU!UESX$7) z=uK$6Lnmin(eb_K#EQY?JZD)r-|sE3);+WXOTBgLyHoK?&Y8#k5{}v6)D@94pV&_9 z?~y)7Pxl@gcy;d4@qPAz9nWIn3v1jDIklrsl2v2&ANw%x22IV$pAJ~=)taJDUcGT9 z7Bu_-b*)41Z1UE2?9diwgtt$Dvo@&ly!!5?&YDphQ|~-OhfLj>SdCmGzKyl7aRtwS z2Gm|ImT1ld=kgrZ+WV-6wSI5Nm^1s%?#vO}Yh5*<5@7AW1;+zDiS27_-u?*+*peSy zEVXdf`ZapSs%hlx(edxXeKxtoPkf(u#kY@V3w$BxyR`b9_=@-z=ut*rdp4l8-d^@R z0`D>53kcx+_Rt>->mD6zZNVSk$EcMvXAjR(1e%ls0o=oR)Edh-eBY1p4CH}5O3vD{ zaOzw=$b*)P&X_rkt#KLaH~Lziz$fhCgQ0x(C*M zLQ8(+7HggPp|dupFE6KcmMV{JAmDsf33oconvQ>7aslTax`{2uV3*u=qtYGIN_TgJ#hoS1A7lmK%!gyz+W|;+^9pJKtABSyIgEp_(Ht3C+AYv zKnI@5cVOHh)_K%;FJlI*wPOZyNlhmnXTWzKpUvohm^IdkB64c2srVJ^*SfIQzJV28 zPUM`g$l-6{^tWgmw!z8kS7Q+R%w4da#rXmxv5@mj1H|JVwC3~v&mnEe4QmhLGuGw4 z>-$*kO>F&ohe(E|z?Gbz5ktdV<4&>mOY^KA=$#9zzF_r_%YGpzmc1H&i+MIRo-v8l z$e$0?b4KU2)?PdO2JA6_G(5#wEg8y4Q&n1P(Wde7Ri#?IwA z0zVd8bN)W-C2@sUo4_^da-OBdJd3|4a0l%}*aA217posRd)R9W3UR>Kg&$h{2Inl+ zWRmw;xPoVPSbFB{(XeNuf=FHYw=Ztdw0p329+U-(PmUSd{zZ&=Z zth3r<-OJwY-Qdh$0e|+d)ONT_?3l-VPF-L(=t=-rfW%<^`mW{#%YDpuU>|Gi0`=Cp zW%ThGS+g!{Mqf1Rs&EA)=%AsO8}CET z{hi%j<9&s5PyJ!l0_O~1k8u~q`pfgs9}8`9cSk&W1v&xWv1O^#mpAS?_pqNcx+3P>`iZY5fOY27Hmo`0hn}48 zPhmNOTK6n5(TMF)iMP(cKGvCUAoW?+jaui3sk7Geo|e9a_-yaPto507xCGx*%J)Q(=n1=Ix84&T|x$D;8(jr__=cvR*-J%^loHLJwZy$Tx)7f2e&h4_s^K`6r<|6O^ z3~$_teeMOT?gVE)V^PV~#dj}j?!C|MNzUHFS~a5wzWt27qvsCu`E2B=sqhWox%IE) zu^V*SGuAuuy}1NCP(-2!M}1Z#zu<4d8Q43p8!+Zvto0FSSuKW{@1TCpIkXL^b!KOg zvo3-B9bjMgJA%_;)}9t#t^0Rv&e_#jD>~eH@$GG|VZDbnj;Sl?>{qc#&bTesx{hz} zRj1!p&0+2n=41A@)_8qa^tZry5;(JKgI{v?uhxA;R=&fzx91S<(cmh!16SnC&0u+^aNhZbTJ;62cabsomv^ROe`YN3F7s+f zjPHu@mxa&i^XufObyoY0vj#kV0KaDB)-wzZyzvfL=Y4dzBX-mUHNgB$;J zg}2r^wXPoga=3pN&M*M4weL6s);UHj$7(^G^PCN9PxJnLBG1+F)jYH8SJ4zOEbf>g zmcD3m&0)PqYh97cS?KHh+rv_8O-!h9PtWf8oZ0gVS76onwQr9EmAD2;Ewd-K$a`+N z;CZ9Bm-kwT*E1%unp@9e0XI?q^vcRu9^&4t;_=#VEOWqkO zxr)7`&ulLB_AwSc=U!TSYb&;3o0i&+9s0MMx_t5o>y>I8Yx42ENqO;zdzAIvWz4;6hza=DS_|}bmw8RByz0VC8 zcb@P-Z3I4hI?MAwPF-7bq}BM`{d44@CtlAIhh5e^GOs)59jj?;L_3l@K zf9SIX_CI3f2ESNmG_JM*J5z@rE9ckWC7koR#=DR> z#P(8ay-S^|#W17)@QK&^RewU~{^r*EsrY_&I>UGd9=D!yPexe=cX9o8c3G5~B{108vJM{d{ zXg~MxeBQZBy}UJ-a8Cib4p-FRmxKeZTjtSb5F}WvJ&kt<^6ckQ7cpzyzv8>B9Xz_8 z5Y)IL`32osEwx*034CJpVxb5{kpFa;?!k>Y?T&s0|mo;O4tp3`^_qK??H9j=3~xL&?aot34o0rwR8MIUQl z&((;pu!$}5Soiha8t`c%a#ybP-|;eMy>qviz2)q^_N@5hT-Gw9 zd`mQ-;hXb5U4u(3bB_C<#=Ldz9f6-qy}ob-d2ny(+f$pEN3n`>xWP1*O~n*{8;qL zxWAXWD(F4VkcADr+3tX zQBU6fPlFA(1^n!GW$?=ZXrE*ctt3OVmM{@Ra1~myfd==EXUxm9(-dS?V6&IXm@n^zOm-C#~%iRR_xs5H@ z$ocncn?A7Kd{K+<9Pa0zn~!%g=hvjqZ0L|Cdgo~5?B_b7kC^DxcYN~&*r&vd-6P?i zN3-5f!Zq^t^!xm{3*xOjT6+j%m;J1@=a>U){ti7GsKkL4Bc`wO@61uRHKztVbUpVN z_eX5)=ueIQ%gDR0^_L*k(Bm0;6p?%_!0$kia}R9+?%i|7_BZDmbL8Zm-(J=5i7b3SkD8nLS-dJ(+)2j1IjNUZCF@gs1Odw|=~N3CZl!9Vy%d|@o_-0pn>jruFB zI`a*9Kc2x}p5chC@TbGn1%8Daz3|OnssXoweb2_+SIxz8=J9oSnEeZN1Lz8z_i%}I z4(~F_?}$5gtmIb^aPD^t*jsL=4>rJDGiL83cQ(e>!<^rlTTISc{SBP^-CeXH58TW8 zOJLuM#&_WaydU>;Mthp;)VQxJ(_a9~S?%q!@@@)v3S5CzXRR|7V1H}XdM{!uxS=7x zCHFi?ptc32U}&)}dm4|Lz$Y$Pm%RDg)Yk>YUP1iUc`on8_&Ipzt@wTq^6c(ueS%o5 zoV9D6dw4(UR;Y3I0^efRpNYrZ(5?u7nZ2r(KAB+g<(fWw1-14bGwClH-!=EB2Rh!tYIc&hMTww)#f$urbRq>78H?ia1sEJ(e1nYc$j~Y7S1O<)v zPy?UUV?IyMVDH4LU-y^4lmq%f&f0od`uMzZa^_BGwqR(MoOw~qQam_FW9}($T*(bf+`xGD0f+COw{&bHzG+WZVEtTFCx4gv z$T{EZ1U{%y3u3F`wWcT%Ur{^Ch!wqw6zXxA$FF6?|W(7RwOaG&$ymF^!@q8&!~rf?=$}C)zpjQq^w4*Gfz z0q6Z#XK&|De6{APJh~mZBXE9UZ|fSk0$q;UQA5t>VXb|cT7%OL82_k&$ z(A)cgRb$TI61|{LgOfko3(nPZ?r;BbHu{8IkOcbUYXm;=eK#xidGzw0^9a;AyP8oK zle5prQGX)8p2M^HJWBEp*7@um_};~Bpl{xJ)pz_8R;~Ov6L#!(n4ErtyF_Pw;5$Rb z+DChZJpy$_9OnCncW+kt00rF^jB^CBU|E~Mczq|VQ9p9<+VzeS{_;Q#yfqzciCxZR z?RuVEeESBzS~-1dT@iEN5hTAZXxJLFmz;MTz&q9Mfc=E`eFUS%JB9OJGN}ivANc0R zS@0`<0}0l%tKU*9U*zrQO2W`##q6t<6D_eoYp+6&0p=P#uo-jCj@9>PyywwXk!9%5 z=$u!s!>Mz5MxXBq-sNYsoM%y|-u!K>zC8-Oy0H%TXEWD;a}*HxSLFR3=&ZxyTYttL z!PT=fZg4&WbsgTloM8v#TTq6^ytcsk^N_q6_Z|0x-^mBo{^o^sp23{Doj$y0t8jhE zsg)1l(pMYcIZ%o1;r`l7bBB2zxdJ!l!ET5fwqYwcFLsMPEq*Mz#QMzSt*gWZX#LD~ zX7?I(oW)r>@Hrg8xVKnp$vdYteqZ;DS7>#+sY(_HE?st&ImXCv?`xyU!VH z1ibxk+0wi*sH=> ze;eOkF7NWV_*yk;FCXANLkYmz!P~!qF=sMMO!%H{&7s+V6aE%!tmYZwQG<2)OwPc$ zL*P5NdpMWAxQRdBS!&!j}`$x)?7^!AkfN#ieItC82@Q(&pGpzJia+sB)?qG zXFYQ8?qR>Nu-0C}CDs`m*e)99^{m30ipKe@@i}{Dxsg-D^6qY<*|KlTi5=Y7Gp0Tk zYu**fFCb&-M|~1kV6H9Lz}J{@*Mf1@Yj|dL-y`r0-o166U+yVT4$L0doV_c)^Nw>; zUx>|n-iCF*yTHEoy2P&ck>mr|d-T9RMWjdow1;4IiHvMdwX!cp`pH$U-fDOs14u@z6<(lwBU_WKVNbc&Tu)tb<^9|^JMrLR>c`pX z?cBL?Sp5omPMv2N9P##g-ktNDz;j@|*C1bF#vMCmC$AkmeJXrQ_PoYg3pZA>3oGf8t?9+cE&0L(Hc!yNz|y1)#&@K# z&dQc;SsIdHkWpTFER7|Dlx5k5A)|&HHVjyB!)Cp9p@Ibu40zyy4gEYTB6iii&`bU9 z_4lryi0{j+pL4pqsqKpvzrp>FUorDXKFuM;PmK6s2lny3+Y9K-vR(}G6Ub5PpM4VB zeHyq@_XOq#exMb5{F%5MWd{8Z_THa+<*IGu6IgTZ+nybGU+?xq{@<_b4&U(Qdk^C6 z&N+a0rdTIu#F}4RVhvx+Tzi&J=mPgitvt`<4)#lUa=fRi!5*OQKi6~YoPmF0K8JIkO)mWc z+^~;tt+56Y6hGGUtux*CPEcOhNp8Xmw0*3t`S?c#oFRuJq>8Ila7B z?`V(@;wk)A7;=mDJbxbOb4dL~+4 zC+Ch0TWz1ueF+Eg>TQmh)-~X4b=Ri%6Ua3mJKm6)D{udEC-y6ebr1DCuuIUHYo;~U zKbhqm<0m)qoujymzg~OJo~x&Mn`o<4i!ho zhVyF<`9@5>dc@vA8_4NE+gzZHC3w(7G3MSg)^5&k+Mb1+yhNLK?!@-bk!t%ibQhd! zY%SL_R*pXMlAL}3KVK2lpaeN%i*M{6f?NXc%HB7$ ze#16zz3+*vH`hML?(t*Ks>N4c&Idk=2m&Gj zGGpzLoZk*`|p0G>guSo;Nj z2kw5=PQWhb*~p*7J_Wn}{BNJwcTTKk;5Tfg$9nhw_;u|u@a2fw0i9rEtC~|ed9HSP@iit1q>F!7sEsH^uTv4Op2taJMQzk}*U&V3rWj(x?>J3yCB z+`2H&jDXX%)%f;SGWY>zuDHJNoe9vfm8zk?+|2b|VK2Ep1N*txO?%78RcG9qI@j*0 zZH@gNJJ#_h{zRYeqRUS_`EmkI-2vwq+YaLPDh~5}kDc+!hdbd6U~Q;D^+L~B_jaQu z@e^M$?lY0AUBG^W=RKP{60!Qb^Gte^^BnfGUOm7BPvTXJ9iarsxY@0b0wPp<(_d3~&3qkZqLpa*3@Udh(Z zJvFrhzS8O0a?MM$aG!_0;JAyvweFg+V6I`Cb8pYBHtu|-_K=s0*U|RzzC*;CSADR} z*KT(-zVVV=%AFKBGttcU{5O#JPxOdoTVJ2?hF}NqjuOa=+;McxMa${r?3vh85O?nH zz`X~tU)7+IYxwSQ!MAQ019fLR*U?vUow)iP1a=ss)_bjUW6M<{HU~X()tVc?nx_6` z<<2~Pds^$+tM}%OWG(R9nTdVsus&_;e9vZ59}72Y=xyBZ9L1g8<>c%y9>k5A(|%!h zZ~;lI%R^Tjo%jvif$|RTY0Z6|<$*2N-kzt<#V!qh#_DI!3965J+v>Ul`^Dy>Od@~(^z^x-$+=e?NZZkyAv?U&Sg277pJ^`ypl z2i6z&Zz$|{Vj+?^((}g|oik$1t2%7Y^4NXs>1Wa&_VVwNyjT0%uX@JH4g7lc^zqa@$DJF8s)Fp{yW#6&;I5|u6N{#ZQgo!l%qJq`V91m!D;vbDxUDL6}2(@);(-r z+ju$yJw@)F$$ck&1>V&_2RZM@fA{eUDn|STd^RoU!yfv7*9ZNl_!Zk8?`ZKyd*D}J z==0dI_DOVr>wQ6W6k#L*LMsJNNo0d_?Yu^&Cg8?C34OB24}>=6Q29O(38CU1WN54 z>%A!(-hg4Oo9oXs%xrsKO#x~??k06I`pla#kp6*e3w}0&rKXNDD$R}{NcB5Xtd1LNq{>@&%cX|@9y*95MfoGlKymz=YJC|== z2kt(J$-Q8|^f&jarZ(iFhZ@MWuA%F`FdKgA!#g|sY+9_GHJ^O!0cI?|bv^^%J^vog z`Nbt~eE@qp(|^~?f4{`Mb)8)8sSW1=Cy5!a9>l!InQQMLK7iu0Q`~dxI`2~38td(6 zO@cbJVQl*~blng31>V3nr$q7t%vkl-26h9P|Gdt4g6i3pJmJeVS9hAsxZ|VU$sXq4 zv8@}}Oz~w(pwA9iP}8+up{V9d}?1e9j@I zAGz+xymwLejIEfv602Qe)*M5WJyZ=O=`D)J>aO8cl4?004He%1$uIbnbCUCbAD<5j$eu_L}a+Hh}^L*Aj z*E4x1K}^0e_dICC>|xCf1b*#K%zgt*Vgvk@ytco;$Q_%pddY3TxH>=wCAoaO8k{YRdj) z#Xa*2_?foWKK|b357eLkUfX}d{syOJ!5IM$f51)bBi8%=1+fl_r_Qyv&)}R7t9a&G ztohC^pHSTme)X{@cE*o+=Iw8+ldHM*$5+gqcc5?E`%~<7yRZK~-#fAT_jb|tsXM^_ z?1wFP20gK@4ea8wgY{p4cUVGB?#cK2vRm$5sWGoVz`0X@?PIq}4|b^!UY z{puOyh+XeX2fj0}a6X?KdF{D#?yvMHRrhBvX9Xxh+v`c+H+t1>_=ETaa_wnd!= zUWxPv{{iNXf96#kadR2;&_+gVu67c~Pc?HUc0Jp+`kGTp=e~QiJqv*rxpZg)wL6bo zeJfAuoCEVyPONk+eouM;#dw`1aKOx;d_%+r@ZM@Jw)Nc@xbUv&%dywtG-BFndFBr1 z?2^WR`DrG$e$}9AuN^DtZfb=FwAYOt*AM|MWPv+}fL*`tR`kt~c*pO>_hLex`g6?X7&Cjkr6VJ|oxt+Na0r z;ST0~J~tpg)Cpqs3iPL7S52;8N9;AR-Jz3vY;7NWL$AV*}beGnD%qW8*_rVv-Qg+H;L&_Y(<_N;{$yKa^LU7 zfWMbdeES#no?SjMpXWR|`daV#l9+uS_!DideDChSxjx_Ozd3F5u)6S*Md$yWK zyb-sb`prIq8P=4b&G)UV4+;Bx2hPr1{q1??bz{`L?1!GQ3=#5=Lu zeZ}109Woc}FTN0S*Qt;0XrX_(uRagr?&|x|h&v;|--!qQ#1?_P`!!`KR_9w(%?4-`m?q?j+Vh1KP%v z`uEV7p+9oxaJ`e8y@-4N_OrhBCT8D%_jP?2nG13?tF3?j=1y`e2jBkJ`|^!BOFxM_ z$Jz&Y0b`2aai>qj+T{!R4w{;KFy3e9s3HL`OQ2r?GCCQKelb1e?Ovr2cFk6C3&H5o&0mR zN$h+!tUGaeO>R;P6m#0v*muNF&k+0UX3dj)61#nE)yJIkeJ>S#d)3TXy{z-@E+B~M zs|V;?>w9bUkq2h2Ja`5H)J@WwHu@j!q zet~nh=iTsm`d)a4je5nHeR{0D8oFkqr?Fgn1-!sF-+}io??LSg`-b-UJG1&>U$C7s zfHlsY=uYm>z@GN=ZUXqvJ}j$;q1_w38U|H}I=`e0P^CuQ-@#9CR?u zA@-Qq^{%6Dz+GFedUq(fuRD5}Lq_rgRBv)U*1ZL>noGLwhj3Va`yV7&AHsk{3PE&z;ER2H)axdan4x7<*O@-SHJUb>>{N zFZ8rF!2nhBwN)KI*PXyQcj7s`kGhAHv382wJEJ{AsyBCHTi?+3@8~NW#eOvx-DBk& zT7IDA`Fp^Q_WUKe`=bYtZ*RGO2N(GIBGgDv?E9Kgd}1?JAMa$4m+y?)kz8r^O>Gb& zcTco3V!a3EPQ(80JmO7YyT=1W!b1<@_3m&F9ov0;c5mjeeLsCp);9brws&crzI^l622fk4 zMEYaJJWtu`Tz92E=sVH5Q9SbYFs3%&H$qKJG3R&12psGSz0?URhAZdge&`^71)kUQ z7(2fG?H$C73&ox@F=Evv_Kn!-ANT`2?)!Uj`__nexZgm-wudvc?fWzi+(A3QAonnK zU*|}C>*P$dy9Y7xBG%A1wEhTe>a-Q}Gcn8uwLr0tyz`D#4Kd>*S4~G-)6m+9eX;kh za}RgA07X4w&8Yh&R=#(gAeQg@U$V8cn}~G>ow$2-eB;hMyLe7#K0%N#4t8n$Olp8R z>+}OOIQkDzvxqrIu074aZQXawo(VjcJmaMUtpqWj1^RYhJLtdQk6bzVl_Q?mFL>pj z^HJ+Qr{+ejP@MmVuj~87{jMPL$DBbvfiot$1AEJH)?eWJU9InDq@$fR&?4f;*{YtM zF650@&Kmc|Zwb&7efe15DdyyN;%cGXjFGcOzVa{!GSLCh_Vqc>kJax=-de>ON<^vR z4eXIWdT3AJ+)3_9%>x86?=ZIA&`HerAHJ^dvj*ROaWpyeq62fD!JQI&g75UPuk#X) z+_v@k&N5!LvFh}F-(S?6c#yMpf-dkAG~mwqT|RJ!^3lbYlZ$UIu@&pyt^M1Nnuv8D z)_o!8yYK=7-+cl+w{K7T8M^`dcvkBMaJM=mrndZn?i&wk0y}{@=XA7reQ~%0`$GQw zx;4$=JGUh8JqNIldnoekx68Qu2AG?OsdWP^7&o+a(x%yZ?@Vy&9 z&kfk$yl0c^tcy8f9k@f)V>`=y`McL7Zl6R~EqVKN5X9_%AlDwwcjvbASS1&0bJab` zC${^#x4DKrh&Ajd_zryTLY{leb;di;f8F6zW1ap)%XttloBf<+Z}*fV0-oZ%{0{2x zvF>a4K-XUIesewy$oW}w?};zuO#CM>U;I6um^zZ0`Z^(R+a z-i~9fK6i916bF0gXXD4}>2s3f%s0>y;7_zQ@5Fp>lpg%}y3X%A^FLuf&;x`yqNbzhDpILU9M5;n;C!_~*bLfM?^J zxBGr>xG(IH7_sJ8A7bt)&pzk5+RO951OF`g-5fpTJb=HqQU|>{xZ!(d{U`AU_Q1Bj zp@p-nzy8)BoD69FIj3vq2{xQjbojDZ0Z zYiD5Oj#1z7&u7hff0s+%J9%N-x4|*j1v#IQeTKkyR$_aWiSG_R3u_avc2TicN#hS z)@O*gvvc$9`3CIi9bD1FKIjf7v8|PN0p6RnzoYA&Ku>&oI)7llnL{_U_7i*u?)lih zyxJ#ME!nCMIE|PHV9$CU+|P-3dKfRR;ot}IGtZefLBF~K>Wd!`Mee^-gmVFjAI6d5 zbKi1$to{T0MX&3IanXd5t4BxAAh(}agMTE64`3`3EszL6%xCUBJO5X--&4hhGjrMA z;~$B?ZJvEZ#;S9!{Yw+GF5rreFL!{7w~zWE<1?}_A6kY=PrNXU!5KF zr(dE^+*kwlZor%vK>tXO)!Tb`>|S-2T)$5J_g~lDSZmy8H_45SW*iI<+U)wIPf4y= z?M&Rc9oXa4Ta&;$w1@gd{`~qdbDx1+7xbCvBz}XZ4&&_B@Ef-1!1LJe#Er>6_i673 zHGXcxI6Vs3?)U|C?CbqUO+%_4Zp5CkISoCD)sEQub*Jnc^suM>2OMkdD<^?^fY_SI z=lL>zJD(VbWWFXmJXdw>aQ=k2#G?qg2LSRCt(4SaRL8R&-ZzSS4|wre}u znhxZstj?-}H*cisz_m+RfV!GD6fM`OU}E8ke^1J5>y1-7|% zCyy3Q4cBL5&YB7AV}4+(ANa=22i&G67U-*7dKfz>KUS?V@5r7>%zcdcUI)4b{{bcF z7x|e>R~>eOCvj)hJhc7f*+VVIUNxIs$JTF29&InVHJ`BdATEDmn@_ZT%?aaj%=rxT zwH4#~^*;RqA79)4);5qkme^0?&U&GZTQA@58zJ9(oe!<=F4hXU#++|2zo#dDn1f$- z&==qS?yD`=K0bH-9;?PZtx@c0jeQ!rYSBr}L=PZ0!BsBW?}`Tq@V;$(cDMz4b$rGC zH7oX}KDJJ44*Lb6{QMI=3@!;M8?$o;&8Qiur!W0L7!eddKQz-rbc9 zVxTmr5r!C-B`GS)uCN@FOG;H;4BAN|K+key?TRdDhy$&cAnR#FWgvgBk?r zK(XdZ{~kENe4X)cD>rwn5jU?CH*sS<6Kg8AeS0Ui`pRw-t=(e(o)z2p*){M_-q?iq zSwpUE*i-R6kQ2Z@-k;*{ZM!x)K#OJbv(9)qxg-$lPHXl)>mcS`R9&Js?ElFl7O`^0 z#Xmb&oRm9s;`ZsFBv+3_%lD2awwwf|bLGMvutQ(`f*oyr&!|E9=&mmxaqioF-GiLs z<+^j<5#j6yJnv}0aS#2(F1|X~O#OkaFSqu#5!8O zXj^OF?u&1|anIh-39LOmS(#lof|Rg144vCd&{ zYdXGr*w_Av7A;p_YXV<1aB{{I$SV%onLSp#~U1wzha^-!px4 z+V*RK4b{P8 zeEbAh>p5#6&hTBxD7T)ovFDyOleo}F_g^0W{y+bR|Kb1sZ~slZ{wqfPtM>T!|Jfh- z&i)5D|3d!D_TROC`ak}c|Mb88U;pdB{*V9kfBw_I{>Oj&-~5|@yxng9C6J=XwrM*1A7=aDOa`d7bLcbfEn z0BftB!a;6=!@H5c0(+hK8~%6j1vqo&s#k}0EML1}tA%||-hSHZ2R%H~z;D2~+y?BG zz&<^fPBkL0f$AH}b{}K%)i?WMcaXq%|B2&lC1P=^A2Ijxd;=&mR{iJmBJcNZCfvU* ze%(L5y#}~Y>u!VC*-_gWPii~bISr0IA??R&v|q% zzPSMczIR=-@#TjaalX&n{EWpPNgk+c7V)ZOU-y=u_7*(fM zI=Lrsj{&SP)?>XFK3{ET*}HZ@m&De|-*3Qkj|plw<{$3$gHkvdJ7+I!d(LD>YWy?X z(^|xNa(gaLof~@+v#$0?a?BX`_6aqe)qTQg(LGSzd{I?FxO%Fle?FHyhOh*cxU zcR+ODtg1;y!}a%^Gv{}`JqNJ=1z7tHytW_Kv)80Pz>za+5IeiNL(M?9 z1V<7MNKl;EHFy5*1`NSGcW|y^PkX3mV1fV@N1vR1Z`62>%dVAMdG;=eJr7^LeFoZd zspWUzT~=*_L%cY|<(qf5I|g_-i)4hZ{L1mJ4;^2c9v(=_6*i0u)p#cG_PFHa_sNk z_Ugn2+FtsBuTK1?1`{;m$5)>o=G{fUF?akkaCRp*fcw0N)y~+}3sw1GCXlDEM6Q0u z-N&4wc8)wb^-R&j*glUnZ_Wa85*V9kck@2Z-8ACD-INS}tiByvZlF7{0X&!cxPP6M z+!MQa$@4CR3XKdX0wO1SQxfK`twPX4<$6Ty4KC$ipfOErkwmkC< z?Pu8AGt6X%ZyqjygBV zyU)PBs_~T`t1jUxekNa_X2#0Pd6J*G&q3P0!kBlUMEGOH<^KxY&z%(eW~?*0pYU$% zaYMUX^}x2?+9YP&Gbzq=UeDEU#Nxb&O>BM9a-n0_v!>2-A9y}F=B!iutZQGm1G|BQ zljP+Dwx~LEmlJ4gIpRUA-goS{uR7xc+~gz9`i||aKYYb!UXPVGHge^BUdjz!Gji1i zTG-EA8V7Iq6YbuLGXo?Txp0Bsu*GliQm!C=cJN$(1@35FfRh`<+~0Ty&-+i$(Hv)S z<^blEE1Z@fR_r2pf=Uoo*xDx;m{Gm_IKK6Vg zZvPvQTeHyG1D^4QoxnQJ;Co_kd&+ObM=XBFR<7i=eYd{>Wh6gh^~U;qhvnMCUN7t$ zaJM98ZpQK(wsY)Jf|l=`b3e~JX<&~af5lcFXzh$0Lz}nGz2qy!Ni@Fk2^t7|?`+2E zVV@vw&X~K^4)@ig8v`%=fi~w3H{j0JykVdIa%u*#3n)9+JMo0yv0ubHysN&u5Np^2 z-O=tY%mqC8N~mMamEZMq&6&jCiJ5Q2&wGJaF@3omSbG6yb`TSl4>URFy2HtJ`~i$> z_guN!arbJ-RhwLkZF67VL+%)v&-dp>+@4Qr6z|h{4|MI1UGuhzy8!zS;t8B@k0)9T^5$;X z4{W*S>%GTT&b`>X!wJyg&qRX+#?0Hxy{lf^IDSd;6I+fo$`#)opBssGew{y<0ki`x zSJ=b8_He#udIQyyco1s>+kA^P)18bP>%^R6U-R~iB(04e=tP^JXyKj%_-oT7F2~s& z9poEYTeMvFd4+rSy}>n}psSOtJQkn;}ICBzd{AMWr){9=xnef90pcl>eR_enlB+;PZ9 zyJKXd>FR5bdPaog+jk@y=$|@u3H_^vUR8%5yXPCa=ej@n_8I1|<+$thz96sS_-(^} zDHqfXw0E4znbF>lXXwQ3qsX7&@v%NbH?(^+VgUwlmb+99xr=!vRxdgJ{=?qRcz~)S zF1MrQ+QU8c&3h*ES1`ABM!>UI!s+;>iODsmwomzTIJZ3;u{s~Vxt8nQu-@6$26h99 zt?p{1$L@hWw(VZls6T74J*(2uRmZO7Cp2OYkg@L7`0c*nZi>Ht$Z^gDazD?I*r3ja z_Dpgn_5gC5wb(awpc}A$#CrD!@fR@Wefm5)Is3Rvaj7|CPjlX>`Qo#yHTIsM1MLA4 zc$O#JbLW08yqE95{oG$aV!?WQ4}52xUgu8Cf5zBb8M(PS*p&}7s5ndCrHmNKe2y_3dfz z5ex3yuodsk`*ZJ`ae+io1FX@OXPrFf-ag3X2Kjmi(Fv5wMVV`E{;=&=J?Wty!2S*W zxdV0w?qgoL3}gxmzDeqbA$pn>X%8L{A7Z`j7wwR2*F4$k+@jT&Lheu_2D z7_sn1KkLjJOQ3Dduhc%R2XEj4&K?s^1McI!b^Rm+64bewQ@sM;U4P z3-}9(zMg>>j5xZ|n-1d&$sCNhZ1hE9l&#|7I8MIiRN42~Tf;#_<@&Z#I z9MO)12inJ~kthENlDK`H)nYwc_c*?F_NY6-mh0V`H=!T-O0nT*1qOT=sjZb zMy_Wv@#USF=Ji_=L)%wwsPWyufvzSdM}EC8`0|>*XF@6-7#niyT|;*;L4u*5Kk7n% z_ekTM_lSIF*Y6j#KNeS>yP5kL=Y9DC?|0xk^bP$f+wQUUc;d@_2R@HMe1ax&?V_LH zLOzHI^Tq~n-;wC!bC!zX2WTRJ!WKdXb8mNb$EsBmPvD)(O1N(w*v21hwj*raz zu|vgU)mr1MlH_L)XSc-m-U45#eTlbRy$0Ia$-EbEKWmM5!PzF--Y@!?^BgzqiQPfV zr4wCyqcc|Dhg=|U5HDa`Gl9HR19uzf7yL=QF{9yIYu?WQD2wG&+72fFeo{U>$xP6j@INBX}Z-I7B zU@H&w8?dJ^_x`c^*7J(BgIuNR5(C&b;8s6$@p)$M?7RfV6t%OBS*!HK1nS#;_4Yh+ z|AH2Ef9NJ+p`*oA!+nvPx#oI?ATB@6Z#;+(?2~g&?M_Ty^-p5JUf%by)wMs`9$gI@ zu%CJ34H!RnYtFjjF|T?UqqngM0yx_`6>Lfd8zf;|I2|U!5OYj=d60Z2b<5DH-_88Hs~^W%dooSB%RSN`ybw zUiHiq-}&AN@dS$X0ZO*M!t<`|X|97*$J))9J=6m|x9xqnqr4LIuov36`8QzinS{y7 zwU7P)=RVZuImsyyt2%x9#g2QfIwStEugu8Acb0J^a@AN@JaUa#i#5CW_z&X6!R{co zuWj8GOl6-OsL=LQ_>)YSv6oE@%kdJsTQ^Z@$88m0EjwSUIG59TS=H`ZN^ zBmiRjRp(A{A0W*E&o$A`7{DF$h3Bqw5Vy}8zPJI!GhFUVw^;Ur+|zun9#`xOsQx=f zw!WKR_chbG@)KC!(asC<6K&j?18rVEKo_3rXl(1+hMeNzCvaayegkUzYd7aof97BC zgSh*fSKIr{BU-h_JcoVkpSjMYmYc@4oxN>m3}S&U9`VY-8Q7j%G3UE1-+aS2ru~ou z2L6+LpzB#Mb7Jc^b*yJvIr#2j&jfA5J;vsGZu?qqEMv`g-VM|&Vpn(_Ew|2wc9!)M zEpqD<_s?zqxj^x5W-fk1C$Q(~XMGUsXfd|$eEHT7^p&1T&U_{f%7B#C}-E{aagn{1?1) zW_PcC-H`VQG<@%bL|@N<+?+n z2RNJuT>rdi+`qpYxol^gd$NyE+~)?|#hwk+Ka2E?N&U$ax87ORhuKejIqqn!`a&#; zDaFsVL)}-vgJX^RsO1je-hpnw9@^GCiQ8AvcfNlQ>+3y%dm}EAoW1(i+L8PD zW4re>M=Y=d+cV1<*!rI53bdbaoNw$~+?kMI8SFY2Hdw$C-S$&0E-bz))smkn+5 zBY(V4a_7cC+q-(8y{8ATz9jg`Ebmlq=HdpnJ)dyhD|Wo<$k|irvG>WPjVbmL4S3)0 z;0>H}Vby2T&!;)L9R!%!>P~9r&*wZgI^nsqHR>CXKk>~OS2F#9|2H%;a@XP49pVq> z-@w0=h{dV6`2?~0PxQ~vz;jLV&Iz=&YWe4`1J1EO?*WcA9X+%+Ud;1cBBQO(Ia;nU zd#NXU@73pIt^NhOq2)f|&iwguZv| z+zH(6SMUPgfO|VDfPLiIO9Z%)mt%hHSVPNu5?5c*_8r(0DB2HV9oVO8iCKHS6X$;1 z{{<9#IjiOo>sjG-kJtgM^Lzgd*wb0^{LJKreFM9}8RV>!Gh@xPzq^!g4!ib2D_8O} zSIqmc-xKXjed`+e0qpCn7x+6E;IANJ&6Q`pJ>^<+!EQkdpgi%lA9pxtdCG`VYkzvQ z?N7AtoPFi$+qdS}!+E<`qjvyn<)l1p(X<7<4oCY6-=6L>$p^M|ig(vy`{W0*xWL3;Yh;VZ`=1kK(yzuD<%YZF%l*t=~=VF^M(ujd-Aq z)qU9u|4IBw>;>G>+6Z#MzH;2v`~!62b?4lV+Gl7Fb(-6--g#Hzzk~Dty=7dEV$B=y zv*(pcfZ;LPGDOT*yfz!P7Pas z{V%EC9T5%_5*w-C17dqN`CD`_T zz){<`sdGOwTNZfnOt z6F>aYi5~#%oEv%P%nj!aa_%vRon8%b_62m}&N3hJiG^{&eb`e`%MtcVbO?A?@BlO0 z664s$?D4##iM^wfxVE(k%+*=Y&d`7RSl>C`bAuz-KF)ETb9^TgoHNv(#2c~0bH4MQ zX#1WsT(K{}xepM)UF_R{x%l9$*k^W+#kEF!0ey3=^={?*%)Fnj50n`T*2!@XzdxVA zTx8?NvW@%R4*WVZwpxzonP_3}L@VZuIr~6AW1kbP=o_EthW75>?u+BT!k9TBzdrlu zz;Eah^i6)Q_h_Ppd;Uh;-NyDKTHV)^nEQ{NlOMo7KIb>k@kgM!KkuqOzxY>dpNV3> zb01wBOtp8Oi+6hc;p=)Aj5ly*Uzp|07w~-UrvDAtXT<7j-PIW9dfnHYnnm87V)L{+ zw&I@lx5oPC=4GtD=N!&4*5E4N(E)D2nGZ0M#C^??V=rep_kn)b1`R)RKi6#&UH>0W zAFGdb@`{VD`(np8^5Q#)+~+_~Ih)t0jqNAend;iF?K{#3#&7hRXzP52#`MKgKj{I- zxgDI`jog3}KygQ(%L811`+Y|@bn(y~wD>W@d;iC;Yd2@uL%w>T6WGg`kYlfoo#->i z`@NBqudS9pW1Tm!1G}CxGlxD<+s8XHKR~0;1z*wkoL6FUJAMPT6K3&q%{H!N@*{qn z0k2{+cMo&}SNa~V^PFQpXPxJ5{Dtp4aRY*?Tns5e9w1gJJwYKA{Jx|@q&Rg#s_kWPnZpILIwU1)m2n=#^ z?VF%JZ#!n~xeskQK~9m^g4$er+s`{vEACW#CQdA#TxZ?jOkzsJzC-lLHP1b*Nl^Rc zno<4e+lkvpn4jdGXFPyAKk1=wZlb-H&<6kM>xvEH2^4?bZL#7Vt;mnq9PF+SUU2^k z>VC<;K3a)+dpLV;+p~1I_Vn%^K#soJUC-=HLd4?h`v1rN_2b8TXU{>d!Ku0E0dB;d z)q}dN?Kko3%;-)$&{f}loY{K!@g20r-#ggbUgiQjvFjOd*Y>cNIp+r2^S$n^#Jv4H zr&8}_u9}ASJK$K`pR{udQOC`?{{_p z8T(9V=hk_#19W`3$`h_N-jRJ~t{L*{Odmh~ez6Mp;mj*}_jD)s8Hq-`6Zg7;8EaPU zm&=xa1Ahfi@D8js-hqbt}CHS6C&Wu$PXwiwu zw@$QJ@dn>nH*#a=CwaMb=kUfxBUX~Dw>1sB$C_uK7cuu6#JwZWAmkXA^TdDbctbi| zd&_x{m*Y8{tGtu@2E5A)h}`;461;#t#2|kI)swXElbGb($A4GYUyJnzaJLqF7Ft{W z*sgRUx_>a%t<-it2@|7SYxer0{6B~Zo_x>P~)37H?XZS zUp_Uy123S+lV9%ue&7rF>ICL8*W8A_P%F$2U~B+=&mezd>rdi!rXBD23Ct-qmt0GL zeg{vGcit0l1z`@!ShaP>xoX|R{>33CN4s`FTmJyHFTT%9`vxXld&%wOtbGv=(6Q|^ z(DIz|^PG|RAYyS7ZO@8hn``I};sYBzfE;(Knfsn@#Oq9I_`^DKb(cvkawVsji(GvB zS|?Abc&r|6>-YQeGuLOh!}UAYnROoR`|_(Fb=FSM1$(^UT;Q0Wx!^2&z7a3Mwr&8$ z_tv@x`6s$^i3W1&b3slSNjS7tZ}aZw*=C|a0OQIGek7p&O&&Wcb*K1OWAWpj8@_!! zr?xqJ`5YQLbNAVI8E}&L6J2vq@;dMyYBy}_!#KPbwteicj9fdLcdmL8GoRRs{0=U} zJgYU8Gj7at#d>e7eJ9q?omg?TsS9+1LHuDI`t81Q#=NIV+&N+r)3?U+JV63?vsa6? ztGt0PPw~CE0(uVQgC-yK7+~3^o z>w2fBHhg<2LtDgpK2L0G{sVlUb@fF1Yl%C(V<+4fzP}%Fw)+&xvj3Z0>Xd-*`;zfv zo;8D5Bi4bwwzGe=cH2Qbv8y*~;2Y~;#=6&2ccpd`F9~e=xJM)I-h-IyaDe}`$fEZ5+3>Evnl@qrd+i9b?g8o_tiJ`bJYgAcH^Gh{XB1b=m&k2 z5$oNvR-FHT$$)c%V=Tb+E5PsA-q%EHzi*DS?3uauLM!&EGo!7yzq_;mx!wzVoxLyQ zeD?;qB>o+{V^Gl0I{*MB!Nu$}LI zl_zd(y<@}!*hhZ##I@!|EYU*otk$17_I!iWckH6g^H6%Oz7y@==e)qFw@zD*wTU*? zV%6(=CS&%C9Vb$;$g$C^4I?LX{d-HCL zn}5ePKk);<<9D=osC0DgA1m$*pYH=GPqgP3^30D|(C_%p93Zhvwl?Q%=O(c3b64yJ ztO@B8os&S>{F^Mze}?}r-e61)>P_YJgM@9`vl z?mn=c_h<40?fEa@3XIo%z-i>|>GL?x@;sa0&B1Lu?~cyfnAhtE$%73gom#>U~3TjTSxz4HC z_%Ha?7i~}LW~}@3xro>?bMigEvoGlCjjitt@7>;+#DFzV(8CT=sFT_O8u5;PpeOnreFHC0 zc_>QHeUGvD4Q;+S#@rQs_GyTIOEd^<`yR2*vz@-?t!-*}U%(rXGmInFcM&+?irsSA zi9f~1`u$A33#6(xsbN2BI&hBh37qAOFfNSk_+QCA%oA@wT|2|S7|+#Xpsi_O#M;AM z?S0}kAIiBWx$R>=N9+mu0lUh}?VHOW#$5Mkx%u*pA;k@i#dW^-qwd7c+&5iKJ zsOjN5;rr?P@eO#7qWZ#_vEGU2oqFsCxf`&TeU6xa+c-1L&s=A<#(DM|f_QAZW1F+i zxHT>Cwb2*)Cfax79W;BveWK+na^1mmm&EFCT#;vu=kr03=;CGq*M=r=S<~i*X#G45I1Jr$My!Ba5 zi~xw73CjOKgMnZ7p7=^G&XpOSMZWj=0u$8C|BtJ;RhI-uvMmJ)40JFL;9x+53mgnD z^F{+54)jOxM-aMSL5Bqe69!C{b`m8Fm{71_!h&}#lXT9h*4I1CY}>XU$^Aq`W>yPOt;@mBSC<(O&KQ+t<>}FKSv8{eu?}3p6BHL;b&rqr&n6SS{a(PH)r0+OH*&@Ucq$(_3GW^C zIp!l4=e>;kj@3BJ_ZI5TGfw;g_YaVu17`%VN6k}% zcB%(Ya?ZPAZ)oh(KpdN^?w{l*zBT4LQ16~|`&y5>^9JTKb`AIZJaA2L)V}Cn`=N24 z0qh@v0MzG;H5TR`pyi4auI?y@?_6UsiS1YX2L@mL4d0mqbm9Yb=lj$+)`}Te(C+l0 z#y9D}x3gz~yt)n&s27z(XCHTWzkqkX^DU3BGh)vnwuhR4f2fPqKjH0buX+z;PkgoZ@;%(eURu$C{j?K|Z3C-bd+y)2 zpW2OqRp*(xdf3msZoA%`J>08&zQ@O&p6*u&GWgE0hrMeCvAnp^=d$@3Q1ZsuAP%gY zJv?gUjj!n3Z|2fH@YQx8r&bKmLDNsHC+gRAckbgJ)>@}_fWVh);7VR@0`uM{v0=;` zXniB~Vi0@k=L`G!UI|Ap`?>oMFg9n6`o!9=%iH_D;5*3&HI1A{Z3B_J4|k~;ZU9f^ zV*UF$|9bpf>kFk7!@{>HXe47sEE?Q5fzw8;*TT^*C;I0+3kG+JNxpDkK z?0GZeFED`nI@1|Hus?xsm4Sux%%ne@dUN)xvt#@Eck=f8>TAv#d3$uM2+*{N<wFYSh(kaA*Dn&VJ`RJID><28zda5L!OpviP2=39S3t8&qF%r_bpdAM5WeerN4B zK@-2PJ^z5OI9HEOF7&Yz*n43847}S5$a^Q>#JS$p``4cM9`|U#T#J>{%5_0q?)X8U z(hv{14QJOO9=UL|59ax%_PAoLd7>BY;%-TeKL`H#T=R%M*1E^(Wo@E&|DWXSXK(xX z_sPII=Y{|IuEgHMpC7p=*4YC-;3w7{2EIDKFAZxSb?S|8TjPvB@h4ozI_m}3(VyBu z?9YVv@pG#@HmPxzJ72KQu(ulZpL5`yV{DE71RV_GhIJo#&l}%I^Z*VF;u&l-7ka?a`a?%av(6T}bf1Frq|(g(Reh|OPtb#nfGHFAGOvNRD{|N%r_Xe)r*KYsy``Sx=z(v7rr%b{78E8W}YWp#-5*RPwVSv8I61T&JE2A zzneohh~53nX!7Wt<#)qbH3K`*Ip+<21?KE^0d<03fjvjAp6;Oj3Ig6<<{rfQH+kaQ zofzL^Uy(@;>n=0a{`TJ^kE;y?sw@%Y6s(#`bxG39rW9^7gck_5lX^cHjNviJfV^IlmWA@>kIC z8=O4{=)m6M2I|gm-o;rFi?ep(+qXD&4>TRlUWs)d`&c`G)_%^o;y0{*f|^AS_n*Wg zmgaR|bKlm&H{gu2`08Hd-0dHrbj)!6An$p={|n9Oq29bbLJRy^4`a0E5?|i@024T) zIP3#`19CsHr?>xE9#4Iq-ZcPrqH&)Ith-Ss{|CPH#xJ06{x2=(0e8Vy`;rTGgG;Rb z6RvtD-&^=5@CUrsQ*}FbuPgoyjHEAdU=Qr4f4yVzozaN(6Wig{boeeWM-OwJk&A0b zgWfx;@h;U5J7Q_w_X4c1p1{iQ`koQ%_YUWK0)6D- zt@%N%xhH(>OwN94)Oxy_Sbtuex2(5@Q`f=R*4on3)9<6Z2RiSrrVhS!ekXsb30UyL zf5Kg{#;3NywFI1ub$55}39t>ly#t+A?E^os)nohB*F5?g==i0 z+`M}Ue_qsF;MHi|Wr{W@s5H126_z}eT|uiW_soFyOR zU-*eHr#^x2r+&aU;sFBax7c&Aa%uu3zPSWl-+FAl3AWyaUDdizlK1pj`wy)533%U0 zjqhVze-BHzs&SUqJ2$NN6PfUoxobof`ehJAsNtB>>i%>1+3 zT*p_J@C{gBKB2yMRX=0#4}5#ncOR?vdS5@wLC%_^ZPYyF{=PKMvDVY^XD+C7-vQjo z({sJmli2rBd&R!NskQHqkM$WutQy~}p>?jdddI4>HsAwZePSE&6%1l$bx^&q15_`$ z&8zLyy|MoL6pwFm12y{&0j|KFGpQvv1im}J@EcI0Hq->8zyu-B@%OMaZ~U?P^-hd4tnuz+2ODX(#sF{esTV zU;F;p_f>~)zXb9PYyAV?+Q9E%0CUDy_}Y!QtC0tjoP5W3FQG3tu}z@&SZANi1^c?& z1y0`@_xU&Q_$KPBKUPh|(wp~g1Fi|qK;s?$bko)EEaE z`>WTsjlfReei3M~#^wjUz15uAWUAi}_*4UCtX+Kvd-!{i-*u07Y~a@qY^<6KapJ4F z!FzxAYQS0Ua=~}@KxfYvzIEoE)3knm@6Fdc9$vfs z7P}9-*U7IPKHpr99X&nE)GzgJ#O9ov_yO##&iK5EpKHWdT>XeW*3_)P(yMCV+^u;0 zp)YpcTp0hIIXh?UTMc7%z+>FN%vDo5{0Ud@{!QyHXYaF9#cIgeX8@0V^qphf*;QTD zkV~NE<=6%druio$HiS^GN<-E&4qqc8UeOtQI)4qdzo#0ns zO=9ca{tmZK=IYh3N7uJ6SMQG<@l_l669jYI$)hfE_X-Gk;sGYGrg}L0WAO>!HXr!* zIQ!LGn}+xF*giJFc^B_$&OYZohk5ZEJbC#^?9ZEj_78COO?A2T_`dlx1`R|k$kpD& ziEnI9js1nSLEPm*&A>k3-J>TKUpr#Ky3~h@&6(Hlpy7+y@J^M(+JD5l*V7rmSlHLR z_5~8IfeAA4;R5{pSik@Oz&b--O~()H0735DrJ39O$X2_lZO3P?H+tXs21-xvc3-=w zv)9C4$X|i|)D4hef&kvzx{M4d3Iu3>>UAzJqi1JtNlaigVRY ztorkJ8{gy?;CSxaSpwA7x zIl0JHFJFD(ylcX{k0)Z!gFkoBzQ|XsE|&(c_R7#(KTB`w+`smU{gylK#I-kRYmBus z*KQAbyQ@2QxZ0V#I*;~Zj=d*-t`%~A>Y7=)#K#ry$S3)f%*UY zzgzo%fqyr)?!3WY#Qzt3$Bta^z6tE@F7noVxAHCaK3ZTL`nmecxyxrP59hJo`<-1% z8%y7E=@R>?LHh!}X~26vd&BvAho}0-wu#1fy6?{&{_H)-Ij4G{c~Wox0PbQRYu&-) z%o{mjkE-KtEtdYogSdVsOyVJ*aAN1w z1xUC6Y993V{od|CeC~d`ukXI!#er76b$(a=lmq|kzgzn+@CSZSTlL>|uAb!VG0lH` zy_Ih!{EXbsM#)%te!rA);p5Ou!u7h)&>ASyyx--kXeC31x zF5rLj{@~~Rf7G+r*`?kWZSCM)&<s%73IDd1<6g!c%@uTfV|O1|ce((rbL`XLJ>Fk^1OfiU z>f6)$hHtOhlf3 z37k8DJ&ijE^~4=qcCC6bfc>=Q-8t~jp4JU{IIY@_?>@d;>4;Mf#75(IpGhp-)!CVA zzl3uy`#4kE;p*htr)^tv(HMIo(TZ4g6aPjp=M8n7z-E5a(!X@@H4}|HxPN`0&>EW? z+xLF<33zL-$0r^@tvTm5VE&;d$pJOiJwI9%9XWy%VpTK|5 ztTlh;hFarqocAe?-8^cY+rf-4>#PCH*Iaz9v3!6GVxT{n{TJ}>a%$wOcIN{-(LKn! zf8sYd`7{OtxJ$hQwxb!qI(_xxf-Z9T{8}diEvT6?_#W#b0R8+7iYL~8+XvWBzV`lY z`TV-xiMZ->?eRkE&-T%32f4@H0(ejdYh7w{z0IH0{sM~omT&k4df3OmQ&>}ToaGM2 z>O8sMB$jV{dVJQfm;8uTH?db{yT84H*fVnNQvtuI^R2WV`(>8>!{YmZ(9-x2YC%$juPQKj_pw7?FoV@d{*g@RH&N;*U4XANv zHMKi)&mE{(dAO>_oL&*@{cOLw;jWK z>@7-9t|h*)ef6NmYm0aRfqk)Ym2qV3;h+?!@ASg zH};>++@+fXwbzNYyXTqd_i+OMzI}F(yMd~`e|=}W8cxDFD`V}TZ_NYy1b#QXyR*#M zXM&*~-nuLN3ux8J+arkeN3NaBsj;?$2Fz6ryj&93pAYyRbE475`~_WL>pgP)USkjI z)OdVD^X}Au`A)9tu$?%-i2Zx~j^%UnYU}$<-1OPuL9aio$3C>ondH)ZwwgJxLXG>F z3w@y8Tmw}@&x+9n+!gDL+tz9?V3Hd_y?SiZr`OMq^?Bd0nft!jAaC46tojR@+8NF{ zPcZN&dHd$(^e4Rg`aLQi+XQ?7Yy6B7-@b=CnipIbTYJTy_;sJ{yO;SO*7wxCbM5uM z(?1r>S1sIm!$kKlYJUNN?>z6<#C`R6VC&tv)A=*t2lHgdH4gr1AD6r z^Vr<+4>cd_&u4p@Q(rp}zTo=ie5>;L%m&Hfs55rf(-@8C2Iu|}Yp!=!_XJmS zxsc@DU!D1?Cnx87SZA+J-db(P`sb&ybKSvNZ=k-#*UZ2-b~k(4D~Yf8?(iV~1K7vh z6?FXSgU!vGcgBmnTmwB;jc;Vn39OU%$Tf2AS-TSVTpH^KabV9${0mxZJoa;@+N#IC zh}F2Ov%H7@&cVCcYiWjjhwoMZCHEV$f=XB{o!hVB<^X>pRbM$V&N{< zHn>S_kB;?ZtX`3O7TkGL`#jOAamN>^pIvJ}=9(iH#If_vs~&LXynDlcXu%uES#RI! zgU%f<@ZQ(DLEgF!1~4~s)u;TPtHwCU>1*A^Gh@xNx4!uSx}e_UoCv-3=4vtt1gE|$ z?h`BT&eqP{yJI_?ymQt2Zl%K?aL#b1JsxU^{e7i)dRa5gQJ2_`J^M9|@8$0==A7Gh z{_?Tjt;5?hu$9*bgIF%XAl8~U*YWLZ-ahWGrgU0(&-HhHF`w(Z=#XL;|57506AhOe$`nd|W`?&Q40cYpJ0LlDbd=i-U&S9g@- zj@H$TSTh5gpl_I)n{$r5$J)qpKkwb*i>r4guV#Q7^W5E7zh*~+SUVc~ezhC4J72)( zuDq$W@)PSG0qU*c%nf4uS=WiJt-C>MZ}8?CIZ-{}tqb6BM=^8t4y^eGCg{1|N8cWm z$9CcYjJ-?l*w5Cv6W-YaD+2%Q(v4%yYn-_I>(Xf}cWQ$;*B;slJoY&A2mTYiZ{r>w z@AUo6TR`T5`XA(lbzdDxjMxDeNX%ZkxQrEIqIwntamIQjkV{?)3{c!N0&DPS9d~_vKzdojG?m-^qu-m$#=i2Ws4Rz?I&ZT0b-EI*jj5EIJ*VD^F2M*zUHe} zE^ZJ{G=Hl5eb=7cw(dO9dF<`(?s$XG*n7h#d;pL8MBre3FRj`J?9s7f$Du_ta=pKN zf(bs4{sC^?G0}n8)*15lI&Wnk`UlCzk&`mhbT1 zapc;^{+^zTlXrH_#SeUYINLgP{yV>_Blo~pBY%$mh*ekI@i#w-JMgnoJ3#_{ttXP0 z-7;4HE3-XvGncjn*1hb5uXcdY=AOT;{d{Y4xjAbGwbmcLsXHeZ-vR=tf1}m6U+skL z#O6waUp;c~j;-Tt{0Sn^V@rpT(?4(O?<4wZsxRr6b2%dyuinqxctoqa@oV*&u{h_{ zXSU;+t2VZ+->>LvoS=B_QM0vMH?a1!pZ0S%_$0Q^kUzET-g55+Z(n2Izs0V>s=ra& zbI-O8je7gp*Zn84N9%0!4MbqUz?V;a_14OHW|XN}asejjpn>tt>PM{acj^Oov~RFW zo#fYm7vx$3eXMY2>x#Dqe*_#D_-g&WRV`sBt{k>b?)upMAl2CCvu4lr&YrVJ=GMYJ zKGwV&)|ms?M_ifjyFa%k>RO=26SkuA4 zx5m8o0(=Ae`&L@HF3>nfuHHSCA19`T8sAB%b(i|yVm;@(`uZRRo*;AG$v(yn*ryJh zJ#KK3#jIcS$+^cBn?Wo%5o=EMz;}OVJwW-n-;MakhJEZm&|kOCJ@oDOd1p1Cc7dPx zenvOY<>3x)cYKg=RllviySH4q2mU0d)?@5v@B~`5^@iG9f9LG*{#_t|v%QB}^SAq| zDJvIiC-)e@_=V;Ij(u3eG3Q?9oNLWhA8r8q%ga?S-mS&r+)22PZ}J3x0N>XA)z^QX z@xJbB@AEw}?=B&%$u-YA{u}%vr~bbF918Nzt)AqZH{pcme9zqV2JEX}_saFg&U1(2 zufW6Gw!=CmN{lvNW&$`Buv+lzHKlue%bKSk{>1XR_Si2(e zjM)3a>w6;0e0~q@HRMFBeYAD3>iwKv-~!BbvG0;Qc8A!gT&~*KT4%Ul?TByP1m>OH zftnGkcO2gs-|v}wOni@=IA`dC0P;NmJIW`%`}sD`n?PG{MbAuh*f;R^8uRVsan;#N z{y%cWLmj!k{p)@GyvzG$9o&HXH!$<-_doN2uHFut`r5>DA{XrCxuCCIV$Ey0YV6?- zTF;F2-Kf3XU66N=0M+N{b7^mP^1xZ2{m40^5UUSS?W?xlG*@r?TU&d;ozwB-#%TE`ht9&m_!)nP2Z`KjQ}PJ{?SaV-di8 zJl^Z-oHSa!tjSn>$C{Jtf?ZDkK`hk%$IeL9s$IiflAg|{0ec8uj-?}4q z#M(LW8@_Sn+$God(ff!_9KbVk`D%y0dHctT>oaDyX9hXY5^!optlg~bAc_6%2i9|T zZb6U>M@v*c56W(9UIG%EpEX3zV@<5;Lq*b!~Ftl?I(}UJ`Miduilcl#p1N~ z);_5f9p71U0gSCv=bkU{21)GBY68DB_}ca@R%Z|Y%<$bFzxfWEK$Ua|F_=Bf2qCjzu@s64TI$UA4~1K;ii?CnlhAa9Sr>VF=6fqr7`JMaho zpsr&RG{HH0^d!f-46N_q-&bGYc}EZT!2dkNGpGg5afWfrwX2*b!(Y~H`?MXax7K_8 z00aO1!7T6-IQs_bX9N4bue~Fe{z2?JoVmWo1<2Xs{2l&>9H+;c;T-R1opWDcBm&L_ ze+GL|BkbpV`#INRoxD2zj&GgEoioasyzDew? zCg3~vpxzVB`EIItqS2myS7LiON8cVN=RIyG;3kSDkT?0~5T7 zMF-X-uqKSP#3S%I?rBZkJ-_Z{jk+|4Z9t8&$Graa4v)QIeH&}ko45CkdG0*0wG+N| z-s{X9XiEdH{)J}X%iCXl9jv|0*<0(GQKr7T@=tnHfB3q0E`299@Aoj^f>^D#V}};F zgWSx;d#^fLVC@S{!?r+!^|Sa1uJG08@9^NvdcaS$%A-CVL!a3%hVVymV$M+mN{6TE*1ffM2U~Eim2Tk0UM&F*+HDF(B_0Oy)oRG7J z$DOJ+p}_z>05-63U-iY)UyXU+-MqbpexujG>Z`k=Y54VhAhw75cDRZEZ+jTq1J0e@ zuai^j_e9+lYfm-bdp2^V-8uWHO?d5uZy<=R*Qz<^wlyQD0WV_f>P>QS)|o$O9^(2g z-xojgTW{(;^Vb7ZPOUbgt@D9y->KdW z-kciWO|JCBwGVcZKX>#Vo)P?lf88}b*Y|f|(;VnPe*llU>Vx&TqnZbHV%5dRdbhv* zzBuR2+`X_(aO_onU^#UscCY>DVNHV45qEqw7yJayad#2XK7UT@&T#%7YA(r%2Cg5w z-_8v<_1Z$V(lu;+!;}YF?{8oGbz?A&h4aWo=&c&9ob!9EToO0p!0Kn91r2`yd1uL+ zpF7ryh{bpP131qJei6BP)Er`UavpvA+eZvy?F98%VATk7ZQH~OKZlN$7uFgNP@lOm z2x5CCw()N2)RdmLZ}`qr*F``9PkolTc=HJ+sDqVL->?IGKlT`)33kogdhYDKJFzF? ztG|8Zt-G=7*;l|&V(w7)AU{FC2YP!YzNaU@u#0`nwOD!e_6V$f=k}Wi zzkrcHyI<+kjX~w@XJ6xh68Kj*=XK%%%)6)l1f5tL8;#hyx}R2{xA(wrvErd+SC8L? z8&G%lQ}1m1>AQn-9v{Slld)^GpboZr(91n)CVKrw&K{l-`!4L@+{z!GoN?_9=ZRP~ zwR5g`VlV7f@3VJTcjhUX#a=bqNv$Io8tZZCas`q;}q2ZFqsPMknK(D^3T$p6>x>%YTj{Guik9@tCH z8vlP*{4X$xC-EfjF8=>jaQ{Dn@dMmIkazET3-t2tqsICG=1Pawn*7Gw|M9VV$rGDD z-=R0T3kd4$XWz=B3w4~bx#y4{vG(}-xVE)9qm)1V+_x*nMGrocjyrXr$ ze60VS<_+7ZZLwx0{y<}$Cu7~w9Xz%3?`w{A4|Gp3W9=Jo1BCGh_4w+9+z|M0Y7$?r za&QgqYK!Aaac=tr*OyN~^x5pYx9?*;Es_abKx z^}eOPeAN*TV(aX6L8~3aa+Cbw{kw1jle~Ru2EKEu?|s!bxF`96b!YD)*Ma@76Tbm#oNumS?Ri7f;oU<{{S|BM@4;%VaaNL>pu?%F`S9

9OiP@^Wg#6W+OJ9@#pj&(-6xYt`$!&$)|p zo%4bl=m&oF|NH8De*dbsK;wB(Q+3HZShs6^lQ;ehMEKrU&bZ@0@wMK!X=DH1fwiBz zxudxWI=nURCYQu&%nf{xyWD_v=Cvn3cigd&0G<1E`~r7uy`p!B!wQ|PxFt)~XY~-4ifd(|+hl7 zujhh%kLBCbnU!OfzV~Xm_H<5*HT(25U;ASRxhvLt)eNlP1#9HoORas?iO9A86F<=z ztC`lI_nf%FSC5ZXYhQ6AegThp_q5Lh9X^Cy5VMDQbH)Q`+lKxp5$L9X9vwepaSh+Ry*$^QKUQw$x|_aQ=iJa)>#4U4`fc=F8nqM7 zUTP;WCt7^@_OrJ7+w_HWLC&eBpb`c@OVhZ$j*e{DL}Rt-2R@g2Nv> z$a}{Qe784vZn#hBfu&JAfi)MkSo?=OxZw|d z_eH0s?Reyx?cO(FUG0wF%>gxS`=L+ny7Tt=OP}6;enH&P)GVxV0(WQv-&%XtTR_!5 za98v26L{=%YP9yWU%+eKLydg~v3dQ{n3vm{0q^~uz~i6w`WY2g?eo6p;GIcn_N z@Zb2JU$770okl_gn!hiOXx$a=FEsx1g*&RBa96A+;%lGt*)?)Ebx93*YX{c4+M#ih z{{b&_3C#O8TK8#KwTZP)#aMmyGgdvB(gb|Kdw2I30#>6u=nn3?!1XMgq-!U zY$iSNhrkc4oS48m`(1&)JNS2l8T)&V=3m72_MWFl!28ej_El?M-+U7Rtl>Ul!ufWI z)ptkr9lvYU;s^dqAFFO`UEqebj6Jf_s1~njq)N*!d(NXMJ(TvA^GN9{W}= zbk?hTm?O6TQ_DR%e9H~Y6<)17YiBOKn))o^YALy5_zJaG@#ymeI9Zgb;fnjd*mXQ-q`vc ztJXgDb;p69@V=#aKd%XA{n4A(4|r?4b->SKz?+}jmJZ+k>T6GS_PzXUe2aQRG}fv) zI#lbk+q}A(=iT7!T?gyW(-Szy&p=%QXS{%OuiHkfzo!|i^RsrgkPA3_bZo*kV9qmg z^-XN;qb}C>&bpcd=Vy0NJ=*G3_eWp-Ki1z91KX&zr?ae^SpAV$&Io=1&ise^eR}Ng zADlgJptR=k&AXTV1HaxI{sAgazxT&_JKsrd$2up`tFhKP>&&bhsMl9FCmDu}^-0y+4zd3pDV80}`W&&s2;LIid zLEu}bu4&234g8@Vzdq}><~dt!3UW>Cp3XLBy?yNAzbhNWKZrZ=1bz?X>ifW3+r!$* z5qGrC>7d2BW1{K6S?&|~>b2@SzCFwrhvsI?nYnVliM~Axn1)6!!4uy;o`i=qQSS$Iimh9gbi}%F7oxBaL$du!s*-BSWP4Lee7@FA0Xk}F=OS7 zh1!6h*e;^2?=D;r*9^F!PoK~JaQfB>=X+Q6&6u3`IiFX+d5oXrJl2fBG0ruk?gwvd zFZK31-=jvZ>gC8cV4uwO9Ua@g$eT0P?^x#yxHo!xh8#zGe1DHf<5=H;jxPd!!a2*G z&OTae2fTUvdrB9pSIhl*W7Xv0?M3}GkFQSufuGusuit}D-wq!j$=!f;S8Tm2ek0dn z^}(xq{DALp_EhH^+FPBo%y~bLychwO``>&bOyC z2eJ81JjC`JR*m=4dej@MwQpJG>#I-rz&5OJp!19(rd%zV_*{ zG*2}gzZ)ahySLc$iM_w|9n`t=D{u#U4B-B9iQkePfHm?n)=uZUck+{b^}y=OU$$QU z@v-(PeJtPpf$r!#euL9KsSW&r@4auJ6Key`JF9K@Vg}(IpW_|tGtE_#fdu@rYbLz4 z*3{=h=&{Z}uYLt;BOvc+_S*!UeKPmX#`u0t!hY7si4N>(uQ#y1=3zZOR_}%#z*^tT zcXs9^)_-g+&3}-~U8j%EzCXbe1i6Mztvk8@z;EQ-q3Wq=xsY0Tk2M|CowqMHa`X6p z?pJUF_I2MUP^WzZ-|z=;mkG`cdtb?0*MPh02U=_F`2^~%wRZqD`VDU4n>#zpC!CrN z@B0odz2((*Y}@tbjo%;peaJTvyH0=i#H(FKti9FQ*W+F{V6OyMe2<*^28tsNeDC-G zJ;`xKF3$Z+52wzf=J*vev*x0m!S7jEcU0p%d(`KrSv zu$TD`%==yR9rXQ~?Py|SV%^1gL;oA!;GFfKR*m@w^>6S8u;2Nf=zGT}aMo*EqI%be zl^-8#zbDqd6W*DBY+v24_|EIO-vQ2Dob6oiWzITx8Tiih?^pH{12l3M5OTSD5$iXQ zK@V!`4$dR?GYxX*-Zx^k@~@5S$GVIE|HFMx|Fa&Rhy{07Gl;GIiSN&C#c((L1n#Hr zjDb#gk1I(0NlwlGLw|dE{JzW}&b4EpZN%>9&ehu-`iZqq13{gz&i>}@HHa_R2Y3PJ z8ULwI&NGu=?yJ_=d=I_VD}tEW?qWZCxRak%Y2n?_=T@kC+z2U;sxaWIdqPDCDvZ%+)JH&`+e=&h=0m~q_zQPS$~c{ z1J0z&YxzEgs*vf!nT|Nlc#y_yBr)0v8n}`lxj}3n-=||sj~(C!7icrr?i0?t1o#!+Id2f~_ET5? z9DZM%y`0tI3;TOuA8Xuil272B4?oKS7UG^Enf(++XA_=5zhN zUq36xz_--atUEq$C+eWDV_qHm+G7Iu8By8woq;pw9a^lN6FY&rk*)ZS@6X-eeqVbv z_yq3vuqXZtYn}Z0yc+qNJe>W)yq35Fx!TVi+28!JgM1z8y@zjTo&B8I@LQ}JdwR4# z(7wP8*v~%pd=O7!>&M}-Y7*GLZ`-i;@XlIi*`s6aXZ{Lr{M>(mSbc8x+FZy9dQ7aEGxy?L;wRRlAE4*{{rbLY2C;SK{QKJv z>vqi(%`lg%&e{gl)_0w}?@-?-b_o1Q{?`w9H1%`N9n^Z<#ouph-~9G{#COKaocdUA zZ7<`2wV!v=%FCJa*e}TavGYg&y0162#yt!8k^8#_cLNj7oo~R}Aa+*AKb$k<(mb|d zKX-}M$NQgSALrR4WA&BOx0in1)fzbY7k$;&p7;Z~J@8_tneK_3S$E&G~zvea|~3wGX&P zUhe3q z7eBS|9`!8&c7mF7?C)Q1Wlr6|h86_A-*abJciq18%@@#^Z&;6goa{~ooCNPGdAq4f4-mI)tT0*>)?U^_*nb+SDd)J^y+))Uw_a1ZBNJAwpOf8zVIR(N}_&=&bQB z&aFG4RU_|i?(KYiIelyO#f+uV7LWBjtqX8pZ~4MD;tTvk4n(fcYrs3FEPNd;c)+W( z&U*~}H++2;u->t9@aCSxqGo@ry#t%pVI#dobqze=V{4e{Y@h%+*|S zH+jxu=bh>IWCY~4_L#&2>pW+9wB|0G^ShD(u&2K|=Uw6Ipl$dww}$xK;mmN(Qytdc zJ@);v_8nM{diw_UrVjXX;9l1Jysw=9_xAq*r^bE*-dzWp%w<2{g?PdT_IzK})Q<2E z>H|B?<2%nj*4*gn_hFE0@CoPn1?~dWR32;Z^A0UnPj`Ewv)0~qS8~-4-n-VGSZB3Z zxi$BO^=@9XoPaQ1rvXW2uz!<$<78TcaMoi{=Ga2+n>LG81@H`aGS?3+G;dtI@9 zhvkC&5A1_jy?rLI##-Y6v~9y(Ju}x%E!Hl^L7$GTgRMUFQWL=aypQvIKYOWvV9R%R z;`*7#di?yWHsJ$;cwp-@$4`CgwDKNjcryKE?b_l?t4{QfuetR(qg8)=H4}8;(c1fy z&-JF}{-Bq;PZ4WYVh3^c#CB@Uc?aJ@?bA9re2@3-vFdu_19fuFj$P-S2fXmJ9N9K)tIk>WvEDlC&B^I|j(+C4YjODYYO!h)Yh404 zkNcn9Z}=y#RnrB#dwK#i?s(40xtrRNtAF5IKaH{HGnD((kh9MF4|wexOnr_ywH^4T z=RMW6`>LsX$Bxn2=hTEe=Yjr$-ner6Z3Di`fhT@qecy(acmDagboNYg9QV1BbGEuh zTsdr=ShM7-pFKUn|Kxg~s`Xus;S*c?&{JJI{m~3^>eWtg-eCqDy#w3f?Q?47gf-{x z=Fi-O@5JVPpN4M_V|4*4h6*_MF=zcWX0G+Vk@^YD>04KKVRt|GK|J6aur~BLnTxkh zy{GD_DH}_3_IHP1?x}A{0o-l#)d$Yn*tw2geh^>rPfk04`^EOJ^pTsu-d!u;^@n}} zEl}Tc_r~|Lc9t48A&gozWT#y<+{)(7x-EsH?W?O z#PpT(vyND?`iZaJ9ekgbOYaWGomkBf+xOnJGd=C0AF;IBl-t%n=O6Tmt!d=^yxc{u zeC+2e^5afD+!G}JuUh&AzSf-k^<3Y-hJQG%c!6KQPcYGWtce7%`urg0@9*yL1XuhH zuK54o@2l>O@83h5RlTSkaNbL;yi)!+6Yk@zKfncC!9b&J)XPt7H_kO<;M+IuD~_G_t_^q( zXFJdR)h4kzd)H@*b%wrZ=%zWWbtCY%_4n2R@3F6YI7j^r-tVllulUaC@BxMv&N<%W z1o<|LNm@pWWH!t^XBG5bHnD1-9wK+t>Jl^$oqJ@9%xox{tk%rr~$stVx}- z?IUN6v(yjx#J7jN6YeM08JGL|@7${&^mxNH^74UgSo`}PKfnY3Us!WjY}MfhzVQqA zedx&?@Gtzp-r)US*{6XSOY_qhKe1{O-+XcRb+2E{;rp&L$^A}ktXUIlKEOcpz&iI2 zeD&26U;ThL-?8QbKJc|az!T2d`U7jvzy0@mANf;jANR4hz5a>sJIVXqIXlUBG_?b{ z2GmT@L5tLmc?^y}*{wJbCB3*Be|u7OECs{tD9CUDr4-a-}gK?|$~N-dJ022=6W< z;jmS=wTTV=zyFUnKwCZG2YqTbI<0Sc=G0x78Ap$A&RGpSnWyz+`U_g?^*M6x`6qCm zyLjITFYG_Gpzami1+Di9;K^Kht-S-=1m{U?j~V;^JFXwSL)zO5WK;I$$*cdWDQ z=bxod(8(Ly)BcUv*}~7`{CCpS)vWuvyKgQRTI$3Fy~i2;9KAut&gZiBP+RLZ zV?TG8)CI9Sh#;;x*fd7-!s?sX@4%n6ANuqjXxn{%Kdht1xqj~Y-r?Zz9q#eb>Nvln zvyb}6)*ILFk@CPE6YH!uHO~JDyi4(U-nZr6H+Bpa`sjmP z;@js+|Kjj{6YrpXZM?CUds_!P5;s<8Z6j~5E=R1k<`ZAYxxe=@=S~6tgufBj=fbYe z`rGeo_TkR{)sMQW#h&|4`gCIZb)X%qpW32z6$AEv}-|&03x_jUo2mF9jb9&~Ewd!wZ+@;}XU^!>rQ$E&x%?HqW zr>;+J$G67|B%B)O+CyK>>jTS~vEJGGEBwqg)43gAy?sg#cft2$uAb@}u-2W8Bi8RF z?rLqo)&9)S?GH50RIh*d8~Pi(=YelNKnwl{zT^9OyU#z=p&4+_b_ZwL*FK*!skN@D zgI8!7sUfk8=WL-#^8n4e;yBfcWeRF1N%Mo+2>8p-gR#zS9$KJrh}@} zhx0zxHz5ZqFQ3Z~ICsd6ASd@zjdk!BE!=ruW8sX)ogDtJSl=;>IX%{n>g?y|TmK#+ ziSf^$rOvkAS!y5P$F>#USFbCaJ9>Bb@SNEfIQs;6qO;za12nj*IWd&{-0(AU=X1A& zcb|rJrhB@3eLuOcd22l#-dWC!Si76oAHer9r{Clf4m*K+>Bq)2=3Z)iM>Se?-qXJq zxT`g%_WiMDy4wwStn>ILEs!P#f8cwnXTSsR*K=u{ZSDovfc@%C$lJ4nCO9+q-q?$J zj(1fTK+fLcAK>KdapsweTzlBhef^xgt3AvaAH6%-cc{hsx#YGDy|8Y8{N~HI+&DhA z>vOzgr_T*MfWAAwfpcEq21*-ihI`p}fb&KbCpm96n*VIqTZG{13!$t14nDVftrVF-#Nu%mEU3thL5n-6hQ7Ke67!*u6Wbe_m57IObpGV)dB0e0vp$ zORJ6bnP1RP{43VE`T;t=GoHXbyF9jfa`!=ecykkEuDjm0zIouTcP#LojO%m32Jno8 z4tV;CKxce7t#zlq5qDzyOnlFinhS8JnaDxJ$`AYost=s?RR`D5UxAv${tG;Tdt9(; zo!iOV;|2E%Q0JcZxnYCc6>A@RYVF(L>f|yJTSMr+&Ph1yJGR~i+w_4wwF%T`uDM+c zH-LTv`p$GexvmA)+E@6U5f3oo>my)CjzuwcH#?XoX^Oe z8r-QTsvhqTbl{(0By~Wm)*e-Z^>euZcT3{3xq1xZI*ATk(W$Xkzk||zth{_npauVc z&%_6vSkC_S4un0S;5;MN&t~cI>mKqp_R!G8gCv@Sg*& zpn;$C`O99wck;V3(ZqehTJzR8vzd#fF=u~cPtP*zZup!7m9sNsribe*3C@ zu*-YjrSfP3?EncTXdt%z>}||(2jAU?%r>2YiANM9sP`$CW zL#*2B$K2{g?t(AI_Umo7-#C5;p5*Njv1U2*3ZHNuKf@Oqd#i6LOB_pY-4MAnevTdA zxo2MO+V;DlZNT^OJ7JwY5@;uJ^};&y)CRG2)$_OI^J~@AOmfyX(DCh&z*u{B2{}&n z&aDjy;CzpodQ0vyWAUYDp9VMJ)E|soeG<-DH3MH8*w2%TQ$GWrUZo}WsI^DeW`kJq z#QJ@F0goC_&sB4JMC{zhsz1EFKKG~2qqfuUVb8?pP3z8tot(3*^HeTyfOR4_8pk@h znXC3ft#?y1L417GoEeStAD<;U&j@}=^gwG*zbEJO``iV(g};!?g`wphH8Z|`2JLz7 zuT69Tm-6`5c>WJ3?;U1Gac+ATMg(EPfIuW+wr!#@ut}1pOF)fG#+XgcumzZm*yLbB z10ts(XVW5zJb(aAw?H)_i7Yq1ivjr$#$_HUb`L41Ri8Y#+5tD@e%yJQ_G<;B22s5jF=;^&? zsYg?GCc8mQ+;ARv^4XuyTi_}OmpIarrmihrbLfH8=P*4Luj2AMlp`ka9-!L-eVxz| zm>+t0?JlGO{fRq#Y5F$>dXNLmrao|_>D?#rcPjAs{iGpuRoY+=7!?T8b-bRDDGsP3 z2S1h13{ZOIrP=8b5 z;+YD;ceCFmdcf-voItNqG2+GvRlHH@O!2fQK1(!n;M8}~lFf5OoR6q`lS>ujo#kwX zyfkx5*-dF|K#z_CDR((5(T5YF~NYdujl7=421VxtVlZv5rtG&i>Q` z_V((7jnJFE%!W%%Dvgb1zYO0;bSV#)p5)LCfj+4~FXLv{n93(dKQHu(&$DO?UhzyI zj&=floqS@U99^{NP5;!i-!Zz$Jgm~4zP0YnuCJmgAyYY-do&0u>Y~_q>?mXD?Q^BFtFO$x^ zY@sA3z~wX9q(*VXyT~UG&-rRDan3GHT@V*J*qLlEFpC=cp-Ta249FMIaOlnFw*Ip^ z=YdOqdQltZiHoK;NQq;k$;l4;&{wfcAQwdaspVOtjgNEbV}w$G8~Da+ax|DB# zgc?3`^s4z`zWS!haV|Jc`G&we=c9&?A7bJep;QbV_L500{S4Ym1NXDMd?&!kDw{m9 zFi&>fE9|0psu+^`pu)XUL%t()2{kt`dSVYRJ~P7kr18Cr_tmUwLLD*Xj1bO;#z(w* zIQfFZ7U)5o7#d-QbJDsWH1}(yYhLQ3?<Z5YtB!rVS4uk6Q6q6)Ula>-oIXq9_6l$_p%iAFTX}YJXzhjeRm`BhpuSlhGx1J% z-t^(;6?;$%H}sZHoqB0z3n)(8S(~^yLij%`m1*D!;x_BO*ixa=C zTCW&)fQ_(^@zp-A+E@I*6Q|yj9a81#!}Fo0DVu$e%vC+5$q{EC=LBhNdNCI+bHlk5 z=PX7sPzvy|;n#EFWU3|4&+qz-seI$bm3BfyoJ>A8aW5G6zjr1E->6So^`w8j8)u|H z^E2__gxwW)vgzF=yY5XN_AGrBlN~;H;_Kz^_IMmGFRHgkJr!>5i}jVjLP8oQ3OL-@>!*Cm@-^rSy=FPlF2%yqKK zX9BebP29_tl&+ekd*cqNMVL!%>g5Pc;?gyrKHOnjaVI+!GOw5fpWW&`4ICfuKiJB9 zq5izUXE&Z#S5NitNyP7!U2}+so{E>^vm13t>Z474DccD254L!jY~~`=5>J6_hgKbS z2M)TY;!KIp!dIM3Ic(4rC&Y;f^%-lsJ_~n1&lJe6pHWk@oR7Qp$%ao)&IKnEpZ-p| zr8xIc$_8!Ot%=MB~4%QMs_L~FCae72>ZxJTH?W{H{5U!n$5h1Vga&C<+`Nl z8Qu}?>fi(Mc@K!Q53$n6-I?hHbm_FSG)OD$&@T3N(%9rcDP%QV?Cs?A_sp(kF0<1P zi1(=0$3E-3aVXd;RDWgJYa#qO1Ar<>a#qg59LAlP!=p0y`7K zcLIX%1tX+F=4Gq4A*8a2m1t1!A?*}*inY{}+Dzb?b_r-NKNS#o^gufBeVkLecz4vg zfp~W2Qp?$h!*N0?;1jQ(54pB{W&%f!9sTSYYT-4MLz720#%D2RW&(2-4K~t#j zpnfMfUs}b(dBpcV?${cLb}1jE$~ox}myOR1;|29#Rwg@DtS#TkhmVGXrk0uTYY?Zd ztz4fF{yt0Efu6X#5eGis<2i>P(46X-&5%tG{(T4hR)wbG)TKUhjr@jk{QeDQ^$~}- z7mr#e9p(+hv#>LMCiqGlubgal%mixTl+tbLq(fXWggW+h(#&8t&e|o=gZhSake>Mc zik$IPKfM$5rM|7a5!jhNp@(?%U=I0Gb}BmqvZ;ehKjzo(Un-xuz2pRTq<7FR_Mjhk zMNfM0e3;EXMjU21<(LuPDi0D`XKc1F-DwohCH@%oPoJ;CP_2j$o4`h82K66RgeuQlTRNfAl}7#Y7ysy z3bpiQPxv7(oq9oMV$W20;(hR(YyrNjwK(2aF=_ISIw6ydPNlN}<-CBnh*Fv}dueRu zbqQ!k+()!!XUg}HlQ5sV33?!M>hBNJbV!*2{@sTbgCNoE!phpWqacIvQL>RE#Y3i8HhFex# zpbp!idjv@--@9OjFU)p|F%v#>OCc5D@!z&UGi#c7>I1fff@me3`ex$w!s zNd@-*`M$<0kBCD}L*V^m9y!j;eNt2ZHzAY>&yyO?HUH@NEHc>*fm(Lhc53WrT#PZl z(__pfiqjj;nF4v{@^cLid-I+%6B|x=rt)jaoGc%I2G9qe`z{shs(f}ciqj)i91e3h zrX1Y+^Zb@+S1gs8-3v$SMlY0*==dw0cOG*G&T@|Me7=c)yYM2vvF5Z1f zPWAK&y~GQ3@@p<}cwR87hY!cO*w+!Sd~`=UAl?TRcyQT`^OTB{3-_UU_@(mgD%b3Z zJ7)5U0VCAs3U?yD@ztGapGJ@R%V$m|gfq!E*wS7ae#l`9#Oi&R(PD;=9Qwg!N9_6? zCFTV7X$aoOS?H#G>e)Y&hQoa!?321+hQ9Di`A!->Tt0J7UQ~ApUJJFv!~TkMuVD|x+5)@Kj~#jic*GjAGcQ|`In^oFBqs^? z?PPdCF?i0&l`e%;2xmgOxGVc+0)61J2RtWmCr)v=U9#!nWv8+yDDLH_@|izLU?%5s zJ&|LShd&cvHhuWsn8|jsnUzTc#QDm0F~@BB!E30_0C7t4QzF-63vKZl0yXU8;R!}L zq^~FXAuk&t?*zE@yFrY3^rerHkBts@qq%5=npQQ7ymK+fU6ErJcBXjUOJ$)SF*xjO zuzmc#X-3OeO*wRHL7I>dwC!p(jg}};ACr7 zCXJ-7im8@*W*MOrQfvuzT|&F3(y}|U;iH|8eW-!QU9ey9rFmEBRqvO{WWH{LBypGLX8o~(}!Aw`NqfZhxPZwz)>C^ z`EX~_>_#2EICCb=Sz7{{xx{#9xF@5SQ;d5EJ(Z&;=lOXZ_0)!)s1cgtl!_UF+R#Uu z7(F;|hAl9Y9?a~O#-1*a>yq~Hw0G#=GrNv6(1#kY9?WDGd?U@AUd2k;@YpMr-9RfB zpuWs=e)hRC+4Kn_-B4{NO{a@zzxwhV!uLSM=tW)W;%r3xef6Hr{WPK$1Hl%C-jrg^XnbZ9Z6GT zgpT}hkK$27jNfy^PO?kIm{H131t-ACgi=TaKKsl?xPQ){y2v4(DaPDPcCTQRbJD~e z@UhNI8^Jkg#pw-*WlmZ#kjbY%@D(_NAy%J5Zt#1g1@GgW20dJ~IlEK5RmIS4fuD_y zo_IcH8G(I+DCY#u$*fGeiS1&XeiP(VW2Bv{?s)x}r(dQRJZjoK((vTd2Vqu6Hh0%1 zCh)z1{iyK*G3H~_(?}!ERWWHe-s>E)8?xb{jdZFQ{hL6sQZe$_nPS0~CeA)yx~*Ia zM?N-v(MGXUI+G8G!#Cm?*`;jiO1ScQ5AnP3v2`~HdKxjIRmErJe0*solknAZnKHkxPXWpj_=S;}S~@|lnIC0cpZy@tPwE=dMx1chic?G9KIsjnDyJAe0-y7z(iw17KiSMF1?JPs$PT?_BJ9b0b|K!9 zP2Vor%rLT59ivx-yCa*9)HT&%^e`SfQmFOHK< zFQl%`{Oq1^N*8%;^$)gU#6y4eOb!G^=x*?~GIw1`X7FZGozO)eAQd-)yJBJ}aD;>zY+a7|^Wvf&eR(skU( zr?w?NvzoHAngb85m$=DbE4(8CDuxm$!@&cz+tvf&4w?6Qi(3w?ZicAcMbh%<}1 z-ias;*uS>JEcxuitl-NIe8qb!JG>)4?k`rT5B*)Nr4FAxQ^f#1IeX|Wo1X4xae6i? zeoGo29BMMzVHZC;4xc+EPM?lAPJSwIPN%qW`ai`>>9%}&wxr?1NoA+9nG44#)|4H3 z%Ju>hxIXp_d&!1hqJ>O0?*_G@4?QZIT)Xnw*~=^dMmJmyY25}d6gi8%R(xn_ebBHmAvp9UASNU*v$}!U`#va7j z1MqtfJu-nF%^qnVV}Y+&cy|;_m1_vpmx{AnsW^4a?cfX4Q;Vc7_HxppxALi1LPCAm zr6>B{$(K#fFjINz!p_oN@{RJLUOx9he|*kDzf3k7p*OXqt85?7$?ugjD^8{yyBp=; zJ2Y@6JP-OA+4O-++zY{0EYuKJt&ye=LY)_!lOrKNL36y&5}a2|Fats*vwHd}U(NJR zT)_ZeL28zd=^}?YVHY2JWN78QkO`%b3J5X!If1*JCa@#oTtuBueXOhN(bVv~;KG9w zWFXq)d&yUvS$X;4`HJs^hJ4GvX7dhOGjlT~foG zQgE{25od15NfR>_oOBK1)KTwcbg=f1Rg+`OYaOVp`JS8oWo0dXQIfi?|^es6S(Z>;(H-0?C9ik4(^;;ZSm<{%I;E( zx>5-BijniuPPv9)lp_{Ic6fI*r{0ggA+H=gGilHmi1Ac0?l2Wf#aqM$U)h|EJ@BbF z0%wC;i<2EEacpeTY-s%xiaq2Vq>|lgcc6ffu!E@5YOWE89HJQK;2>HN4`)K=0 z(+4}P;o>}E%r(;V!^ci#mjZht-o?4}VFx&M9XTZR`XW9${B9u4y>h<#?wk`xaU+eL z$!@78+yPqqeDdM{u&)wUlm4lM9eS(<|JYa`pQ~@36=EG4I9kKJLw) zou1gA81pjemSW6FWwVoU0^%bVlUb#FCzRL~-^*uaN*z+g;bsDPdg3>F;!G#s$!C6; zXJSvdfc|i?GXY++>XB7BV#W)KBdJ#})xg7c0+KpGcR>z*DKxPKW@qw^t90di#ZrNq z`klk45AfnsgJ3%$+`stTF*^|BdtaRwHNI;;-dDIMaZCC28DxdK%oL*!cTn%?hzV%+ zPFQhCo;mbUU!E0x>N;jL zROh_nh-~kJ73LGC77p=D2eyk4zd~5I37ojS;*TSHO4D zsWiPEz!%7KR$@jr0zVUsfL#h6+gJO_MjEO~U3?!huYi1b2Kd;Q7&+Gye>U;lgD+0F zZ)qpMM~Kmb-a%xCoOC8QANQ_nQu)N-w}iUB6c;{cBhDVF{C1CQ!OI5W`6)zQ5Fhtr zW+^xq`xtcXOOr!VU-7Z^>_b0hNSd1IeT-*LT5$M4b}5bEhy7Gy@rj}zikv|_UPY02>9Pf^IWK9h8N$-MiZw;Djy)hRxJFz40&J8l1^Qm z<)k@lCbWe59P|q_#HG#)|2|daS z+>!GFF`a?4GoM~)&XM9vLJx5eYLLvuTI~8RjE~PVt8^(Hq|&L2vmCldcP@l}@)72k z{|j*G8#JbD(9mw|2#gT!MDfhUd?uSZ`Z?*evQ5wI_qCYM z_wws~Ll4E!a6s5WKC#eCd3vM*pFMWZ1fzIXL0-rkCqoE)?K4XhGwSUGc3?&+@Xpbf zSz&+G&@BlbYn8_zbEnH&Q9qCls z2%N3-)gB5N+4x=R0~$h8a4ybv$}!JN!%KmSIWLW$3DjUypY~`!fga&5E1gM~;uz`h z{Itsi@x5ZqM(Fd6&?hu|RIjsvmTc!^edsG2?FHiWz=mtk;^T9NP8v=o%^Ym*CY8^A z%tTzp4Rd8T1n!v{`j-N|sKsU$JyYdp2<+UdG&%Zs<9U9`A+Zxwb&i-@t|^;4;yuAP z#4E@PbBuBUQk;(-)C1y;sy584^7Np$_Z8%`Ln@zn8Q;#cK+DcA1q*hyUH2I1SkbNa)RMq$OULkUBX& z?tz9&oI39!zbT&_&pvRZs2emjzMqrCMZ!IK8S&0(HIec^A-Lv9!vyp}T~(01q?;cA!taXG5`$z`pEil*`Ch z_ZN1QkA~}gM3?NQV(@&$CkCe!JP_9iZP~p-pTOP0pLeusdt$v)>`Z|j0$1_rsY`+UnrB>`Ne{`0nI~&=7Q{7$Wt6JjfOHcBpw3lux#$1mU=tsO1 zLR{LA^OcWQY`S2dXfvzRhm4z5GUU&+X%#v(96m2f;jd2z`T}_D(;2SRgfLT>8n>yak_*- zQ`J)_73&>{b=0(FJE0>^Q<^+=sdAa%1bi=`K_;*lpW#KP#yaC-j5we5=|ZTHCRYk| z9p@vLRdMpz#Hn|(IVb0JKIY)3mE9|zk=;@L8-ZD~aC*S+o z&jUVOKrbiM!k_QBL---*qTi=p8BiP;fjl~M)iYI|ne85hg{DycUmSc#zKR=3CL18m zMUeNhr_-Z~!I@<%7Q{z>{f@NW8>w>G+;t|KTAmTT z$vNeC-h5_>!wWgtMmZ-A=MOcCdHJPa#1DJ8XdA`&8TIq$PCb50ak%J7(%~HuzwQV?$4_wnyk|*iRg<9Cy~Ja;5yxLmDnQbSB@_aEim1?}K5m&uAwN$1Bd>+!z1bUwWmAJH?%BBRi7~u-jEWoL6>BHKFG~ti$%=Iw9<( zvJ~IP9oWanp9Puh6Z-ksqp7~^LoYN!eW+0^bF=R`C9W8`noHdT)zOC>ao)oqPG1E$ zMsfD(MGN#z1@hSZZn~VXqtFr0D8{atkA9bGQpGZXdat;bT?)0hsxLlgEWHe3((v%v zg*uQ5x?vf2V5015_-w6r9&V2 zaIy;2BH{atkG0Gz9nsnSyooV4s-fjbH3RV+Lo)i_`6 zPF|pfzHM>I;fHzHh!?~o#?GaSbGggFCs%RE!3(pdC@+7S@-u}^rbZq+ktc_sOCi*$ zmcE&MFR(j)9Se0{@u~68Dg6HD<)==4#A^@4_n-T471%w9*H_u4G_%m9Q%s--z7aAP zKVuoT;N`Oiv%@`0XY^LQr8qSW`ScE4#nE1IbVr=}K2rI?)_iK5{7is{rdOTE=dO%2 zV3&?+5cc;j?%b+l+6$qESlu695NZDVEZmtBN+0Jt#o>Z5U-9~!ZS_i(XSSDSu9J-r zL#Tsa*U}H$pe31$oTfOXClcSaKKZGS{lj@=BdLoBw&qcvdBw!%Y@Mo(cLIM_Z56NY zrY)bDyt7{Q^e8J$trzNN%C258KjS)liSzm*w3KrK{o$vAQIAk3GsE|@XF7Zdv*4vZ z&ZmwZ{C(U-0yUXpPMWhZC-~G%iT%ixLPI&kJADySAA4al15U`nQ!IRsk*2<_Ji9P^ zrt-vk;R*C{Znkga*PnHA;SAE5IP658{%Gz5JM@-z#Dr2dv#4qGRJ$oo%uBbalcv@Q z=+wttQ~XpueP)OkJ`LZACRFBXfJC&v^a%s@)7y!L@h z-%Mx@L>R?54?WvLpX`n_XE9R+=}ep;X!)UsY&58)ljkFJ3j^U8A{PW{?KeJ9w|XUfB& zHWjZWaF$+yGnUFxJL{fg2Ve71aBeqBc|Y|a&)m26^}Kp*CWGpw(A0Xy@` zNlF(~d_O9iKPNeC)v^yW8?vbh_lj0NoJDyryCco}GEKZvAZIF0tMc38z%^cZ^{o3j zeBweTH0dE7diW}*T5{BtiaBxMgVaZ`13PyFr#QJ%$OK~T3dYBC5kFHgzCTdwWdqI( zzah{ApZkW->{ND(nEc)WX>70mlTR#@&F-aa_D!XUr*M4){X?uePlFyV=BTBnOPnxI zKF^i=HNcCid?}zqPx?txC-_*4jW{29_{^Ok^!3Ev%*d2)O2Z3+cJYki4)75lXOb&r zXVT$rq!GBOkA28_*$)t&uf|u=tOg85T;v$C} zU^eIzQer;#;dzB;D4#g}nuWFhc^v5P11nsNu1(hB6iXoX7U;RtzVB@P#>Z;z# zxeBRr^l!@#^YA65lh4kTZYWPbcFq)|FFrM{j!VPmd6*jT6^GYU%nQsiig^Le9dbwX zMW@oUW-E^muN3IPos=%#WwWZwl zc=_;aak4WfjC^WINg1 zIWffhI0Ks)cR(GSmf)Nc#07dVKk%f9b9OJC325UY&&dw=p>p}m z*F1X7RNqpVCZyy9c7qRZR-JrDOvup8LyN<1%;cRxcpu0aP|2(veLMp_8se4m`5l(` zls{V$r_L$vq)UMqJHcxUsr<03c4)}w`*8hxCN{m;f&QM_3bS_5&cyJUV`R_b$;ZwF z@Ivvryuc@ zocd*n*@0uEeLMsGG65cZV1T#?e8BSrrH_1gAt%2+BQxo1!~;C-TE}jETzrdI2@Fh)g!ZUJFXK2l7sSmxl3t#0zjqDQNn{WFb>C)W5b3f}R8_ATLF7R(qm`M*bLJfZC z<5dgf2M%*(BhJTrNu3<^BF|m0W9a2$EK{5sFCQIts_aY}5T8{?FU8u@L1Z^%dlzdn z>4s+Dm&%pWMo487qYv*OeF5{lbl6exa9;Vuu){uzJNcP2O6VN$q=|8l=u!xL7xP9w z(o!A{dvv7PtqW8({cAXJLY;Jo`RG%N@1?0_*R--T>2Md)>|msM)=st;xSIfCRXo(| zY)(1))Its7T)Z1JQo0E8DNrq+->mY91NxTI#E~ZbeXQl|^f3ZldIX+)Y~rbaggtvA zk2rVLmY>O{mO5;7OYvHqi+MCT?z*K~Bi{>XdYh_-xgqXjzf}252zhZz#hrZao*k$S z^JE+4Qh_rwt4Rz<=tGWLdNHRYCU_t1rs7UEF(c5Q-d;BI95MOi8nqS&pW4#@tgq}$ z@ltVWQrYYhc2+L<3Waw{yD--f7dL2Z#T%;WRV>U?jGow~w1=a;N@*h>ZiuO_Ee_&* zyo*#ih~iGX`ux-ZhbFFmAjEw<*Z<S*O?l!8{)uc zZYdr1P%QLhu54=Q4+kzi*@OFYLZ{MUU*${X;TXYt>7MxP$UE8WObzv273j;Cm%Z#N}(lwTXFm#mF?v_fjD+UaqnY&OEyxvc;^{4#N@+CrDyi&|8F4-8m+iu znPR}elW=d&#q${H%*R@|#KSuyJ7^Q<5~n|W_Q|Bl_sX6wa87cmP;*uQ4Zr_s4)aUpoxt;l zQ%mY(`Pt`WFL*{gr+DDXr@p?YOmQzz>xm2XJI_q!XNs{ac1JOGU_ZWBbp&#zs->S3 zn8^&MII)nIF7X9=5o-^~_Om%oF@zc9>hFJHUmu@iOYw%ljL=JSO63etypG8W`;yCq zy0?=JFO??e1#;o8)vK;g)xT@B;%Fgol;>W$gkB+2ZCw{My#%jZDSRWeg^tj`?up+U zc-OsVat7vS(!?fI@lrWvhC3V(Pf|ZK<0@M^6Y!0XbLfSg3fN86I_Z`)^Gn&>sS`Md zcjETQ7vSTUKsJ4;Wye&S`86j~jy`GB!VhXOo$7=XC&HT-#-KacWv8-s-UzOpKBq_Mq^yE_PCoq72}Lp8*)OCc4!KrGau#rKt;`N-jn%wrZc za7$^V-nUeaXT@FA((IW%$9P4ZAnl~7170?uZJbER%+h4$oe9jPJdXv^Jk--dip`a)y~Yho06P^5-0+XFD&s|1FD|bh-WuU3qC`HB7&r~WND{?9Mu$w-?I4EudC<+PKFi64DjytB+=liz>$ z+TzSPPb&7i@9W~E;rACeZFhaqI_Z*PyDt`={N~wfPd@hZ_Ux6*gT!57|8 zEU?0-#jrusicMa=v^en2fnuIdcQ2lN==%JR@2t?h@*R8Uw=Vkc?%|IvWsmy$M!VP0 zW$YWPo|MeH#MFG3kq728r#?~ac~^fiD!;$DEq^fIe(S@F;k%3}KHcVD`G0TPqF6XT zqZs|fj>RVfzDO?5ifh(>Isfy}n~Hl+{8MrDFSafI_R59DdgI0v7yo;w;(-~L6%W3? zMe)HJXBX=(I-$6!^GyEMeBE}j%P&gKUj7Wb=fi{Sw}-4_S6S{bd(N)A+JD`g+p|xY zXs`I}U-p*=mG-vfPPZp7f0mtg*j@IEEkCxez43+pV#F^-{kYEEc7fmCXjlKs4R-B` zL;II{ZI$AhtHveQb~o(cX~r&o#kO|MZCA3(CXMb1zuYkY%q?V({GVa=kb!OO33nc0 zZ~7s#AH8y-y>fxA?V*R9oLqZKqx;+~FV6q+))~nr+nt;I>D-z5LjN6XTT@QWr~YSc z`@-W7+pD+!DB1n2lkBZKoM_kDaZS7FGuQO*`_$gaCmX(zJp0pT$-Y0%GxC*}&&ban zzGN}sot=v%mgp3Fy!d_hwnu!CpZD$-`8qFtm;ZUq6N=HtEMzZU`;vU4!87uW><0Fe zTgTe*KfPr4%5Sy@?{lfWdc^z5IrD$oz38sz*jK(d-|l8cG_MiWLDmmZol^l1}HP#Lv zXZJcHx1C`(+fOH6ZrA_d2K(N(yV)-n{gqvLxpn$?+VvN9-I@weUg2L+&O9KT^1+?pS4xqJ;;st=auUeukEy6arfPqz4=gb zP`;0yw&2os#DwMTo3~!keazrnlaXt^W)J+I9qnTKTxh2+^@P1=N}pX~#T)GzLtjn4 zd+OcpC(fPI?$-Dr`R@_u*!%8T(QfhJ;`Xz(uE}>Cbb4~z>lY=h3FjwE4Ou4Hpy#Dz z|8I^=9{uR_4+1~xnMeLG&qwJyivi1*KuWNU|a;Sak zqwA6}Q+G`6>3_R>;m3~X{&2m6^BG&6YV!vMckg)QiT145FSIK^xv)Lpn{{pDlvV68 zhp%T>TCHVI8~g`*>+=sJgN_|$pMPf4qv%lKs`~CwKoYDW6#k>1=ns!|> z_V_Q83H$sj`S9`nj?}Mb~eTMu6Od`_swfhT7G`J-lKoF zhhBV{-Dmuj_P%jD*=43oN=|+JwdA!qHnfBOc7g4F=1F_Rgh%aHep-Ki=sC$reXmYh zHQ6<}?ZaQ%Kg`=>H(uxzyT~#J*iA2)+C6V`O}q84)UAD}UYp#v_C6cZ*+|EP4B|{>SgVsDIN3+sQM{ zi}T^Dk1Dq8JEwT~#0kZ!w=P^v_{YTl$zQHkEHnJV;sAT~gC9OVDL<`udj5H9@BHWy z+Z98{ewsi1-c`l-AAFKO@u&0h_I(@Xr>tX=Lw|ZX&%PL3tajT;#rKPRQ(QCirQ+4T zr;8JR{9AGFk!JEsi`-O9S!>Ybbr)TF^0^bgDb_h@>B*yxxu)3p?Cpv#w^*rI=bB~m z85>VbhAwtlGHIKglOYpUOZwLsQf!&cRlGZJc0T0cc0S~d^NTGOydghivpb8|*4d)i zVyV-MO&j0!FLwO<`8{La&)0c$rDVeOs}%EZvtV)Q_16~HFR)$li%U&$@WE5^x$bWl z6K~nN_+Z}Q`L$b~Tr9Qw7x^lqpH416^`FVGzbtDH>w2O4^EbZk{@PzTX~kDwn6&JD z-{xNq-8g^r`Y-auH<_n<#Dm-BTW)lH|FR1%VfQLFvYQ|9lwD;1_w3kxX4>}0bB~(Y zSb5Z=7frWokA2xLb8coY+wEDq(;?$+?~;4iv&Ni~96N`}pBnsZ{@B>v^KGAaH#u_K z#rvPSV>`R~NvqlQzdSL2WwWL2naezoPwBcme`DTz^Yt&EkUyE+ojkPXe8s#^te9N# z^cwb%sY9$g@9FL%|NDCP=xv_u9=Y_+NoSwyyYCpf>d2dKTEf0}+i&eN|GLaxFuhEE zIB6NX^aUp;Tc5U6|MSN^m;C;|SCjqM{~+l;dWC$BQ#T%Y!m5+=1s1!gn7rSL#e@G| zt9a@!{`5KkFhI`dhe$NM*Xzh!*=zDUbo5G zLr2a1#@F_@?|foEy7(eH&RuKwU3jAXyl(dGAdtr%6s|mc0M>ie6x}L%O8G3{=^MS7grA2v-qa#FU4^~ zk0?$avv2X@hu`JPz4~rGX1*eS@Ta+oM|Zm}Kjw|YyU+gUi;;8f{kMG0A4lX@-gHFw z4KF;GtTK9Fz$`Qg9%OR@i)1Nrmke3(D@RlC@D%X^B4hFo4`-~YWB zc;e;a_4}_Xb{KYb@%SG17k|8afypZ^HDt1XV9v>h{^rr*p^J7bZr{cgU+;f&v3zT_ zV$MlR73VCmOn&4^$B$h3x|8w;UYn5L^4_15UCuoud41yj`OXXPTAY8#Va3kt{`BCl z@0-WIw#L=n+irBKJ#yc1_PjeTvt$3Ur2VLUvOV*uPm}*%x3=AHsGi+|huL}Chb8At zU$K~D(#pj}C-0Eo;fCACR)0PD^z*~)AFnvfzW4D0{g3C5KX}($oqU_Szn?VsLW@p5 zzBy!a^5QQi|8d3NPJZOZr6=Dxs4;o$_WvzDzTolVn1eqnzS`lcV!?w?ES~ysx8mEC zZYmZUJJ00n9(}1ed(-oZZB{(G_~PDw=bKI+lxKHe+duD&E_?AKXV_JrzA5==?&a;! z-(6{^Z2NC}-{0S{U;Oe7JAB^1y`14S!`n z9Q&~S^T6OyuYdl8J!aaAcB6A%wRaqJl>N>8kJ_QP4%j)iemmJ`(*?V4ntY0V@+4!2 zOq{P6G3Ld5rLW&lUO(}Ir0<9>`^1*J*xrSQ+vH!5*fq|*(B8KFb#{l{ueJZZ`WCyy z^?M};Ep~B!{U*=nN8Yw*G31ARio3`DwRq<5=M;k$dbYTG$Bl|F_x-eJ9~k6DsDYwqhhtEMiu6OCyI&1<9XKt59F8aIJOx4&tb*n%-933eJR;+t)=YI zqgSxM`{1VTJ2!bX`O}k6^{=%;|D zztENT!gri)K6<-d?5Er9iiaOyH+l5ZWbnHi*#0rM**%}R!1fLqZI7LJto`ia+moda zT09wh-NwleTi(#W;RSc*GrwyTC+vMyvE8UUigLkgi`&m$u6X2;qIlpSS6ukcUd62& zK3@E448NQ97Hf`wsF>@=J&U96IiPrLo@3qh*SZw{-dMPB^)H#M=uMV|Q4l z`?>uVDCYXy6x+P=M*n$7ubHg3`)}=e@2#oN@22*T-6R{oS}^ju*BcdBo^*^6Prnne^HU zH|B?3yF+q`o0KfSO(R+T$@BY{yYR*SgPuD&8JO#pwO|8w;A#gKc>FV@_Ao?@><%lx>x_b(2A^QdC))=tITPtIez zciqXlD;BV?Jn?&b`L0{pkzZb6%dRc$bC*uH->n z-j>V%a@2XhxXJc(hTFdmyT#fy9zR((f<9?<%)d#kYw)RTiK;w_xVNlU6yQj-n04m-_K*; z-~Z0!vU@JKZ!Le9UF`d-?cl91vq$Z9mYw+Eb9UnuuCnE*3HH^aU$8GuoO{%6cF|Fr zoVM7g(=L78ZhX~S_RH4FqfWYi&QU|w`m5dP+9~$lVawZ729~yC|319?nfE{F{^-&F z<`-T*tXTP`Q~J06{PYLg*Zrng>cnsIXAWAr_~m}97sD3)V?JWaxZG~l$@iOkX#ccD z)=b)4J<>nl*QexThAdeuy1*v+b7uz!P7uk7FWuN&Aw3%4gN za^wZ}{!89VcK^eWWR-JIv8yfmZF2bX8z-mkQRF9|v98_aq(khePs(J`i*`@8yJ)oi z>iq@m)EW2nul3IdMm%=(5y_g%osvwt_lkVMqxAXw_Q?LPwmUC><%GZHE8a6_am{UU>cD^L2jTzwBF|=ckU^Dc||!$$9?xX2n;pE?%s8`pdc9V&~$t*VZmBxO%1H zx%rRmKl0I&i;u@|ocEkLta!ZZxctH6=F8W*>-7At(SIlw+x`0d{IC9b{HeKnksUR-`0>I&=F4UW z=i86pCcpiVox1mX&m=#7e`NO&lfLQS_|4xGzaPI<#Zu7W?Y@?k7fGne2bV2z%1GqwJus zd@^{o#9n^M)038*{zd;SFE^9rhCb81$JDXO=PSRKyq7eS4SzqVc=Yt+k|kfeKmW%~ zjpXnyW2bdiwFjNGxZVAr73>io?q`P|({1m&@Kn3_KFYyQ(8PPaEcc88s6@3p6#wwpb1t|je~k4~}vj9$C;U5d*t9>kMDl|qCa<^{-53QE!Mj#Y20#R{@7>B*u!7H z&ThTn=k}DxuCn(mcVY7K7vI=V_FZ_?$N#;-F7)-gcKSwd+d~$bXVe|f++=^1tZFAb z`}bslr>1qUeEzX^(CXLOU+;F1)t?;fxu-teeaSKdO`A00;dS$w&kZd$ zookDH+`y^*L&of>-{-c?Pu>2M;^afW%@??BkN%T~oSQ%N`?+&FVXWQdz&`ut?KAuD z-Qn}(+%vbg@BZz;?xp{9yj|qW^^z%H-(sI!zstUPW+Qp}u`g1;-|K`eNiivaV zUfkW8rx7_%k^}cElf!QL_sFA$-<+7WF3Hz!EZ99`%SG(EFJ}3;BmbW7zTap04U4T@ zZ1mmD`D456R4h4it75*Jmnf$0wPJDl)Bo&0?!Bq`L4V$;|J}}r{HZZ#JP(s2-#$W$jFMaDxCL~LXnhYwU>6Ow54J1J<2K( z5h{^!&-YbAlYXMDp`oNSG_*8+_wW19J@=mTJ@4~8pXV)R&hvh;U_EvAseT8my4TJX z&6>yOns^9`&nhtRR%dm6Q$g>_H!j?u3N{zT@HtPDS*wvGOI3@6g4G{j^YH^*p>Y!4 z-58CtS6bra^s@jO7Pv9q0~f^nfX17KAXak)+I`%h`=uGow`&I7nO*$g-8g(9k&Vy2 zPh;@q{b>HM7|Tp^(QQo`4(b~spELr~7FUAw`pHZzD-Fc&zT%Sq*~0zwUS3c89IP`` zW`55evPH?^w!S_)c=?RmltFf(h zb|#&OSyVaf5pz`C$22=^1Yf7Qaw`-Rsi)>H+hD$fM(v8G=^Z{~W5RUvOU%%jlHae@nhti7`nD8##HgB9EpKVvn z)SRX>m7m2d!Sn$iZ(>M4v%^@Bt^}P~c$0q+XwScq_l2Tb5j5n6fmrc)@M<~C@0^nW zI|md|l9Lnq8sk8sR~ehz9>M(0ZLm?znTtPvj+0Qj!cH_D0>8O3aB^T1fAiiv(tnW3 zytS_IFB-S->osLq$SHFy=Ud&cb z`^d~YF0tVPe=hUzbXK!x65Xy*r!cvleB-F)@al~MnAeS<_ihJin2iCUO9fkW{0sXX zD@$ST?y#%#-1&+KZ`%h&C)t@9C9L&8A@|v22mDrbhJmFcN#5fK_cftku+w)DY>=_X z{c*MI;#m(Eefa^j&z*s5TK7Qt8DsQUY{oec%W?kTI6U=J5x1-qLz&uTIN5U~?KH07 zrYsvlAMXc1fy_tdG-@rYl~TvOj}|c1x--n#_%idtl-1SvB|N1q)=fG;d zIrx}B(`O!A`bC4zExyLS+*wa8H_x!zN_i5OoXKu#T2pY>6dJgk&o`tQLgqFrHd@4< zzRC~PCx$-(iQ<{i`z?_RaR<_xr9)@e@3Oh%r_Vn7X;MUF7=`b=M%&Ihlj7+8lz(qL zOC5a6WY_CbRp2Q$n|k^67awzChM}}#=NDM}88pmy=BCGk-XPCr(U#sPIzmA4)f(mw4T8^gtIl+2G z{7E+T5S_Fvqd8YrGCzftbm-9^HZRGP*V3rqE$o|M+~7QnK7ACY9yp5^FoV+?nga!q z1~_kGI$juAgRYOy;F^qF^cndN@1*tNgyPru<3J!@Iw591X4yi_eZ3h|HeQFq@*i-* zw*gu;gHg*c7WIT)@$lW!*qASeBff><28;bD1CP1;hoew7$roSmy9`q%gu~3|#yGZX zEM93H0{7l%c4c-tlj$S&UAYrxu^Vh>#h$I})&e)hx_D*#TAN#Jx*j`_&w(AwBElGTo<#M(gG z4rZib>_W8{W2mIKmK1hG(;c4$q#vnEqv$iM+U3DSMcV{=T7!@-6~^uLE&vzR2AJLL z0T-srki1|bn>$OHwc{2zsFMbdOOw$Uw&1i{X^d)+6>8xRVCL?C4`T~qW}h`=zJ12! z&elX@1w;I_RSoS$HleO`G6wVqVdGaPl)ebKcuEnL4vWJrVc{6~bOM$g%*4X9&FrDo zZqDbR1w?6{XI2i2DbY`aZ1+v31~qR=H%*{Xy?UgZJ(|+S#KGU&Vl+1Y4^Pj}gBcvb zm7AyGqxrGue)BC%nPQFO5p#)QGVD-z^AU_qYFT2xuAM@701*bWWGENUe2q|GqQ{tyT+);i*+qdurP z>o0i7$a0x)#9_1Mbzbp;3-jBkMS0Fjg7*&(a8{Ypl#MnJ_rVH$J@lB7-)VRe)d4g6 zzY2Xq6|`zMK-Vlw47CWrzN~cIeMJuM8XiC?9Xq`8Ru@AiIN;X}N~mL+iIN$|aU%tR z?)q1tVzd#4YeygplJUCxOwzCJM(rC zr0-He`Gik!aD)dutImP3ukOJzuZs|WEeaO(EW$f~J#gPrYwSvj#aN*>f3}^&je$|% zIedrht2vAKC-LjaK7AOOJ}V@zZPE1BRi7lwTUhs7X*zI5f!;NpVi%_N@LkkvcfTu< zBvQVzgt$B8QX`;VDXI-F@x3%7NxWgl;fW1;3m#C-tgjUQvW`;SMJax> z4#iKFX%IxF(A|)eG%F~Uyy|7izUeCy=^RGIGEOMxkN}^35}@kIVf<}hkCXN!Vc&Wl zEuQZ{>y{>vj2Q%JvoSbz*I{-*M2B1=j#KgNA8fksbhb0Fhkb4=W2;t_@)G8zyrT3p zF#MYi*?HBx!+lLm%-@a4y7{OTeHNoyZs3B3w^*wtV}E?>Ow?0x6{dBc_!*Ux=x)(F^g6kj-k&ZzA(8>W7uq*Ku^RQS=63< z7VLkI_4W6`h9eh%j~mXd8TJHfr|iS{TYflLt%Qv&6)4i^j_W**;oEmJaZ%lLShCI& z9wjbeZZTmX*4zMX7dJC8MFDtOxM1G!d7Q|NC!p!ChWSf{^N492_WSI`lyMdK>w63? znExM|ja`MgE%~@!GYX3fu0m8$6yA`m<^rz|vEcS(Hpk{6+jP2=UaJR?%H#+1bMpXo zP1R};|7VW4ryG(rmcr>9QLuR7epJ@{!^f_A2eOM>IIZ9&HU?tZfUh;_+?hc3TJEH} z+kq8sddJMeN7C@4580#wcUo8Zltr-pq`-aVl@-+J%P~V*@t&~{iR)=a#R@taY(gI> zl+-;|)3c#esxXM9kljI)Ikt*cT{%Z;xx2~Ob3a`)$ zblO8kvP>XTw+nJTj>4i-#02+b$vMNDvP&k?=hAr;)Fe;*jc|e^qbXge+cbq|@2hWS zzsffXkBS0x2PKj`zS4t~sCdm@SM ztze>(#cax_AH0^=2ezj~o3{Jd3N_D3w#YyOBvpR0u%8F1zDk|OdJm(RLki@qtwXZ= zG-*Izhg>~o(fFt>q*A0sbG$R@!51g8E3u}^k>%{$ifFcE@En`gF#=9Dj76DQhuNXt z;qb*ymbqTs3v1=qvv^xY@^XDa7tJ@(-S_uM?b4VA`$1^%7c*{H_fw<6bf#m&lHD;4 z^2?GMthF~al)Ei#Xese-P?wwDkabA3A>`97+MwG)KV<@`dQCH%pO-}b@9o)2fdmY0 z=mpQFa#ZqI22-MR*-b|&8dQ~`exDGSb0CZ_Egyuv;s>E)q&h~`r$FZtGiE%oitYOr zL4Q7p@S=y#a2u?D^OeD_bfQa=LVh1*ErX|MSFi&OG-Wd8S;ls_9Oe$!SmS|fp9H1B zM$Fox6kvT9wC$P+QQ0o!L7(X1%oJNAtMT08HP$qK%^Gf#wkXYiy$X-@2Me>BOz3!_ zf|ea?$op;xr?>hTSGDyb#s8Bh_uCF&qu|8-T-?geM7@C;+hAOC@|&H`)d$>fc~h3| zJ(HY$mP6c;1X3H2rU#A7A${sZEcz)WQ2gjX`3H|++Jaa-t@j6ZCRU+Z0FSZrcwA7d ziFQ3gzdzwWuqfRPiVj!c_RZfAc`gQxdXM2*1s**t)$FY#wCyLn6=2|r3f#Rs6{RCe zFnL@jjCg zt1sf&s(5V6`VXx>hU3v&Yw-1Jb1Yg>hyt``xOOM|y*7cx>8<83^$9pJHA_^E9|VtQ z@mz+oBD?P@!THBU2)=Yjv7zskuwLwskNST^|v|w{5O`ex6LB6=L)Eu zEk$Erj6qlDMCPD+15`hyfMVoVE~H%s1@XQhp*#&cv#gMG8epVfCrh4RL%hl)ik`EB zPBpKm_gz->LDi9VU06i{!%JAfrbrqWn#T@hE~3iir&y~{&;M%PMyc-8jOAwyMX9&i>6}nltLy zd=*!^qL##3uMeZ@>x1mgiA^*`%YdwZH}E}sWthRs;UIZHiI*E*!x`N6g0LOpv|`Oo zwsg!0N{e=(FKrF%StCbkZ^dZFQb(BYHv(Q1|KfB%`m*VJ%(#qQy72Vvb5OGjz#7LC zEV$Q=%@&KX&uS&EJy3v>3*1o8Y6;U!bf93iiNd->>1DzmHrG>kg0Uka;86G!yP4GlgKjbc z-Tl|tlB|9%^{x%8pI8Ld7l%P`dK8zm$saxhPlIubUVvFl9g|l*%dTkO1E*GTdeP{` zuKl_Mzm!LU&W0xp7mCu+$rhx1a6cnL;X7TN)+9|Pll zT6TPOQ2pt65&C8^md-nWV`fdqsdmprsvAF^0zO2LTFOgy@No%UoOPJ@UAM0OvTY}C z&JMDE(Pb>F!U{v;wxja%2$=UI0mB}-2#Pw&1ZA)EDPFY%c554;zA)qIlzt7mUvrr1 z#{o9&-z}RGCq;O6zK5kXUuGWLr7(F=ijHd^VgqBRf^*3rJ2=yexs>*@?>9}L^41*s zt0+kcfwm+taAoCRfq50|VsC8BIJ>!N+{R^-N%Q0=I{dSd$ylb4TIf2u)^9{+wi75T z_y$|wv6qf(=8#ypC+&B>NL>y(^mzMX+A9%FuC>$XhW>WCA9)Eb*2F@PUmu*YR7DB5 zyHND17+e0f;hGU)!vAC+MjqOUl}nOQs`DK#Q#^yE6MkXjpfDrZBWJ(exCPrU*5OvE z0L%&5gqNF#GoyJt3l2*r{j2lo@|xQue^I4jV_zlhZR;bs>Ub)TNAm5lBnSQoHE6rg zFLhN4A79S7ylvpJg|q$Dn~!<4KgLuyKVw;bA8i#@9A&5HWwL*{$@J2GIvw{-p|2)k zr1J9*y^B9XZ;qZN^|^Jly5>0r7!A<7)Q40RbdqN7&7+YyqbRpx4MoLSPy&l2HSIEH zZ#u}-k8h%DU-ZfNt2{ptl?Cp4^>#zYA2E*)L%aqb%T`akFK`nzC8NNfoOFsDFKzyd zMf@@$k4@?n-8+ohi>Fc252RJMN7L2fhjt6XjL;oa8#J+6_?!O|0$%yTzP@n~;iJVuG9I$gyT7n+lPj6g z;oW58(N0})_b9i+loqRQpbs50>7epz78*RBYTi%g<);7Odv^+;{MHLD>?q^!=jX!c zQvoP9`a&VsxxWu46?K8@ za9Nzyvkn*AZ^9f2OO*Va3@V9ckd{!z-X1fsQ@h;)TgD#&k;b3!rRy%N9?t+bcEGFD zOxWt753iz6aW}Gu)A6td)~hB;m0kDP1&!})(kM%IWT^zxJ2#oBhoo|277gP|6%ts` zGi%;i&H=I#C1HHM9lWy4K@HW-81z#f)BOE0Tz?93BQ-I#FBH$tRL8W>=60{|WW#^C z6ZkD(_A#a(NKc=Pp;sH!X>GM%y}g(bC=+Lwg6)W>qW+kJD{aeBwje0ODU zX#_nTCrK-xzG3gP+n8*fEi;qY1f~1x1+Q+-VNDk{v+ci?xNFr0g1;XZz}6dLSY~O8 zMfP*JV=r%lXpj{`**L80P{-!ZQnvZobm*^Zwf)wq4PEo@@!u9?aYf=9RF&?{1P5cO z`Nul^ol-!X)Z(b4e+8xdc4ar*57XCKs4<5z##wE45_ zYzqxR*~nLtxT>EgN)wn=FW(RmpCm?_q@W6Ht(;`QpjM+R_)wqWiR+dThS)-CFfpX zNHw87u&(->ZM4!y=3fXcT#Z^3aI~a2f%5byfUOaBL z9?e^HFzmmjxU(w;C+?}lP>o*f3kk(1DbDz|D;M9*c?OS9<#D1K18lCL64lM_W{-b#*Nr~TsG}Y6<2j#^N;GhB3E56Jrw}eo z&>Y^%N(^MkBR_(9G0iMRBt@|Nt_Xe66{nfw!`Tv<9*Qn#V{KFagL}T}=(KDr zHdal?xrd2_AgAQ>KOxUrJ?JJ97u4|O(SGpc=y!Cc{wf-fxURs+qg$%Ic7xmeV z*RxrUhCVaiItTvnhr#8z9(-*UBdhnm?3H*I|KiROZ0wN1V)wJ?ksFCWCTn8NGz&D$ zj6^kaZ`jt^h3D)?+H0KphO?#&v!D1x#l8|+vCc!x{*Yj#{ixL6*e~yBZ?aFqe$;DY zd#fAI(5hd^a$bzay%yUcq+>mPm^BsKRL7&Zi6@@6=wb(kxlyl{2Gxx*qrzdwv7x97 zpGsz++w!LpHHJr&&jbLD3D!bQi0oMPd=-F*AlKmp|hId?I zLL7^>=e}XK%X~@9vsG~1ZZp3o@d693+QPy;6)4AK6ZHz`*`NDAGQFx8e$t^tcsMs6 z{|!CF7c*6mp6HUimqjfLw3h?dkXcl+NSkKq_A=Sz zM&`cJm-Q+ukjal&DoKy9&G9XQHDf%XY-S^9*S+V>6@I}-%`0G|Ss>W!?o5jwJz-0= z4$LPE9;bx|*Naz1SfZ{p%f!J2{<=*>Z=SIWIg9 z!=qVP(lJ&#?Ui7mn;KqFuH?OsnqY)m8a733M1Qdc91OgV!ykRdGu_7aGk5*Kq--(! z$A@pA|J*_hmTkttAK$UZzzQ1+4x!}u931Po8e{hC!t3EBxFpC8XSGko1?ygeU*>n{ zJ<`i2%=yd4#YVxn-vPYB36Ojw4M|SN;UBvR<~}X(SVj-)OvkZk(Xrs0nauw@ zpUMm+M}h6;e11g02vpsW$dygyYMdpJQj6vH(p2M zz`XtVP<0|M;f~{aAsa8*d<$vsmV52kra*xwWaZ-qL2s5 zIS$*d8k2;v8~yvcnSK>{(D(yC+5Km3R4uoZ4(GEgZL`>ks3&~S#>w29z2V>{dxGC(*$LmSJK)IraxT?X2MRw`!f0DFvN&-X zHvCdz+bc#x^}bod>M zhm*$RNUuHU?R5^8o0TxRK2dDE<;F)>e}SN}51HA(5^&KhWu_9tY&FG~vXfJE*)_>R z_@_D#e`OY9N4h*teRl=_RzF4Um)~)v-8y_Cu?Ek-oy(HujE2=ijm${-ER(#ROvaa; zDER}>+yy@L^n4>3)lH#Ae$TkekFDsQq#y0o5+kwH7@D9cNACiw*%+18bYjnC{=kti z&a5Y%O*Ps2R*k_*Za$jD$>ReZ{}jjaWRydcbiW- znF4B6=lR3q9U!nI-p=fW4Hy@VwvV556XT=qW7~r}cx+Y<#*3fA%HwxY15crI;dYeX zdjzWsa(ljm4Kw{BefK2f_Q(BVf)(#9g1ngmcAdSRFVa8;1pZ5&Tf7{L7@2tm{j<&dXRULHXPRGmB zqHwyru(RYlAEhKbFlty3MxIZ@o~vUpcx(>;+pk7YZ05u4tXtUDlG!Xgv6FxGOUS}I zhm%jl1^V}ck>06X3N!Vk+h+1YjkKA%RV2uA*&F8apBJ1F^@TE%Wh55T%AR~$!gg9+ zVfGq&xUzQ-7xz*U3v6O=Smt45Pj}#)6(nRZThag7dW@_}#x1h}M;+Y;4O7I)vN)V% zqf@ElLn|FA%O+djnKZ>d1|pysC()$59+rGN0Iun)Vf?H}@E%gbaj)I5PAUd>Hrion+%OE&Oo!eHGPu{d4a*O- zqIJMaw7d5johIGEqmdC zt<&-2pGrQ1Pi8CSr*X~ms_G|MOruG=-ms7#cUi`PA!fjh<6~WCg4cSRTLC3Xr1V{hAS-5LR<;fnkgn<}={;izER9z&9;Yh&j>IrHF}nep zf6PJ^1w{GVS)Bc-hy0k+#J(@oCCfK<_;Y3uynpEe^ZG3DP^1F>nlcOJE@dI4--OOP z8BjQ)f*Cb+z`?d8*tfb7p9MO>^q8qa7Eu$Un`?1hV+wwClfmjqrMSaX*WPc*Z_FPe z{60AdT?%&Nz6)mP+ops9BwWIrW|>67Yf@Q^)~>st+vJ@f=Kweg(MvB9pf8 zqP6*PB%u{U85ZXB{fIJU@6Vw8$V%oh5kTjXE-y7T25PrB;f|gAg)FTkMiwTalH?2M zORa*lkMBcQ_gQGndo1+!dLblIlcF}RU>=VxDD(S4c79X~v>wrfEhd^+J}?SZY=SZK zlo9^z-i8iw)v#Q#64d!`aPdnRdh8j8wMsT<>Yt7gh804uB>{Er=i=kF1-R`JkAmD2 zsA_i=cQkIoWYKjvb@XMZzOfuE)tvbEhXi(0GTJ!%xOn*YD-Ajp>TyF!54kbJgFr6! zAGh#VoQSDhWBdQvkeOjM+xH=evxr#1CjTnr-t9im-B@GI8A;1v%LiW^;?`sORU=Gp zehf+5H6TU$2q%8DgEy&J!BpOIylvhqxUo@HI9C_oj-bUTe|ZY38l<4jeQV+M31Hfa z8{jHG3iOBng~_YKevW%t#>Eo?Ov2|K97cGq>;MoF*2KYl;rmL)2+Fw zq#bmEjIv|c{%sB{Ggh2JXRe@@!kJY2b0Y0}(#44D!Ivmx? z&7GSE2|fWZdPELXW-MU4Bsw9deJqy06=fek(U)^Q)0K z;!+crzu_cEUAhZrgD3OnI>g!W1#xUto&r5IIZET)X3@UsN;EWK3duzVP>JUn^4wue z3ntE|{kj<}WxF!lZ}ScW{UH!qt3A8`Hzb6jxwQm3 zB;Mj1z8z)71qQ;tttvOEc?4y@{>rW>szXgzJ*-K72BKcW!Lj5q`zp$@;>JU4;rV;9TLPnV+%zWt*-^-4DsbgoxiqL$)B6^^+in6|Jp||H2*o7PmV}%c`*=DU& zmJu7urY4rq>)t73l^H_YYs2Y`m{>z!XAFICSVJR=lj)tsXtAxr_j%*DD7Yr@p=D|UBqc>3X-OT=#LOQ ze=Z&yqmQz+N740ql917B$4guh&VHJXCQ>hO?}&6fW4#$`j?Kl+{k43-S1;@-_rj)X zKb$iX@WQcr_+FL7p1-sMu}l!$o$!)-IB_~{i7i^}iRu}N_b8h9C_-iuZ^cliqZy6_gl+9Tni#b;=ka}*}5 zpN=c{BnYm{mBOB~Rq)S71tqs0X4N|hLY5I9;hO`CT67`D&YPbS;Yg2+Hq^A4b{=4{2_H2gPWeCxrvrGDMgOC(*V8FYDvI&~X5kmO7W+H!vkot7+TH=R$=bp4r>By|W>t!9C< z)p#5~Z65yTV~nTwJp>c|<+y!r81|&@$IT5@uv;kxPuIJnw)YR{&CkXi*F+#=*J#$< zJcUmTxzF3)zXr>MzR}Z*p}6rvD|9^i4@CE;@okE~SyIg*I5#Q=#`s6W(hJFO>bDry zM~YLb{B*kPFq}^QjHj2sGH6p#0u4<~r}BVk5^r`TBTdG>-#g9rsSdD`p&2x=mH2^8 zE`oh=Hy|MOJ-^^-0JL|lMT7Drd=an$4gNF1f=f>^tnnq{l>vP0G!@V95N4Iq0vxAv z61ipJ_(yjZ1|A%Xo{`1SE%f4Qa?kTUdFjk%>mk;xGzEToO2cWFolJR+3V9|2UER>c z_GkFf{z-q>%Snnfqp6CWsNO{8-o{id>?r=@zp_p|wv$d|onYQYCRA+XPEN+r1WIQqu1tb7M~INk zPeao9Yfgzj{HVg^05$zif@@A@{L?=&?D$40a*{J9QNKOZ6|{@xUh-rUw*^yIc_W2S zUqr`7pCJi#D{X(RVSXwJMN?6DQ(OvNd>yI zvbo4uPw*=H2~+RxVB%vOVc9%sZqapnP??$o|9PHZpT3K;Y3EL}Qs`&JcT<`C8&A0Q zF_RrI{0==?ez?L%68(cExXy|fOs?OZ=}esgUJuRqH2)*0uQ&kHpg8kr-uBxdY@h3bGn?Om;i(8LF|osPu@HQ6>l8}197nB=V>n@}6Ixn{ z*sG{8+}oCoEv>rPeSb9esaitc(oh&@K9X*Cj)anyVo=q61U2tmaKxAog147s@XuOL z4F6gNAO5`Y1G4{ zeWOt6-gw+-)eD2+#aK6TCk7>-KxI8IwA>YlB`*`uKrIAk_T{3CQwB1292@J`q1Mr{ zxGeA}E^P6^g(+STn=H!>j+g-!tA9ewgDk<8Z-1dkS^=$Y9fd0z${^P+in42SL3!gm z82@1tI{1DBy`SMYVfJH4pU$ED20v6Ub3+ZI7~G#Y2a{zMpi7`5h8ZLYe*H{gzc&rT ziOsL=;=~vj44%fEvKU;W5{OnJR-l>-*rRm{1N_p_x2g~uk{;ptH+Rrdr~!4w;_>e= z!iFcCFy@>*dS-@T{Cy7;q^v=^Du4@4E5IbboQqe|#^Fn9L9iuzSS2Bt;>80hz7{j$@Wt z+N@>BQh3dh5G33Okzpz1h_zJV-E=vs8+C^>^I+^o{7I&raDk0;R^Z;pn$g#=R+c~K z1Jm5y3Cm`Mk!6AkZJaTe6`R^J{{a_fl-~lo5jtIS7KtNG^VtO1PZz0zG z66fr%4#yjrN8s%lU#5~&4&TnYu@URW(D^`RdXW@E9ckIr=vYdJMkUbb@OWW=Q#b8I&=$D7i9 zVDQouwe~8Y=?)!qT5|~7r-@jOg$%>ovAMQ!O z-lZ`x&~_L$UlzySMTJmk8Hy)&YvFYAYROjxq>jck;QLXT$FCbNOu9$G|Pj2c^e0@Ln|phO5Vewd`5uy{@0Ho#PEHbIVxd z=MqRN`D}+iqhLa6CQC3p%dH#tn%^k!p{iV@uXX>iifyKBOGhnR@wtb+9-+@>>&<67 zX2{c&H+_7?f=!(CjO*Y!bcoq0eql1RPc!Q=`K-WZG#y+ZBrG}^fBo(%@Ky%SJ?tDW zzETm^-*p6?Y6TYgZ~=V%If|S1lmO#Pu&+MSeZO?D;WaM`iO3?)#uW+*laI zz5JnrW6n%Rtz>r$|$M0l2zHov9EkB7nv>GQQ_Lf4{t4n zMdB5_{Z2icqWufZ+DD?1>uU5|WP|<>CZgV&0IoPYm%C-=!v51Z$~N~Gv7JA6GmCed zIZ=b#+=Dr1*triEVYf#Gx7+v!)W%)oSWFr?JPkviGoB9wm6h{kE% zc&GgmhV|H?+o@7?P!V>ZZ_LB#ugdVxP#oeiZ!A#khf7x?ar2Wnu(O+ZMTh7! z$r=SFaG3*N$HwtR1G&6aK^qrzV=+k`-a=c2d_x}-p@xHUR6JY{mcHG<5;lae$ID$< z;D9CCZLR~^Mos*aWx&bas1npneF(2lxzW>|UF?y+B7N{NqV!!8xv5Npg&gr>b9034 zi{L;JfgI9OHgyNm2AlXPBwfozFHizOd!AfqHu?hxzmO_-P zHopG51lP^2#2wEAQ2nD9I_y3RvI8f$nF(xuvMRZnOrnuzxAXHxTQbw{3M^>4kWadAKnK$n(ZzUA${c-;zJ4kpzG^E? zcveDH*%9z=0_Pv=;6LF#6UZ*$saS*f7$O&S`JYp5;=lyRa(%D*Eg8t zm@{-NL=Tjc=knwK>xaXYc6gxeJhn=R;JQy6@ndBW8m&sjuPb_m`KK=)`4^3)KeFIK z+zi+=n+a(9e%$r#57x&W#L%G&Se0}d_ic^Cc~=+Wvg4WfXU#2qq^W}6j}OD8*Gp?M zN)1u_?F(=fSt7Wl?tn!ei75Ff3)R}@V6{+p<$xHL&mWGSu6k(tRu*rzj>66aQGEMF z8Kp;kgSZh9xXn5XUkdY=6E^2?Nwk!GrU*58^Zn>LAYqFaJDq%;kn^_;r=ZVbUM`uOQ!w;zpDYrdZclB z-$N!B^^G65@C{TZ@5FMoQxGb87Fq(7F*>yzMz%!5?DVnNDUr)LAGP2rUYD6umQS+3M32a-)%-bMpiv{d~Gz)J{R=!yCNzZzVs0D9sk% zPnro^*`$VK_Vv{azOb}_-7YF%Q|mkJMqk(BT~nms;$}@!h|i>DktK9zmVX3?0OVPwwT z+sQTZ3SDv_~VQvD$=+Vu;wuKmr<%Z5Q>!g1EMM1-z8 z7V{?$hO_sxWbu?RfAJoEj_a7*z|wO*u**?9*$6I`EM7;Fp@on^n-)faS9^G$wZn+t z@{AR3J;Z;y*9xPw)!BmcO6*BP7c|+cp#79kY;zX54+g*W(Ymd0?#v@_H!)*5a?*nMj z%eQ^{?+5Jf5Ox$-KZZWX1a8Tv+5FiF%Anw{PcNRRK<_hIkco|9M_4qI5i=w)FsAlB z+sXFM7P5gz3bnbzw3iIA36Ij*@F#Ih!rhNAmz1Ew5>0rZs*SEMZE)&Sb;x+8joO}( zc;M|J{5acJ=!L~%SeS6Mm=%VP0wPe|Hw*Fc5fn8^$LJw9eDo;-?`Ces>-9c(_>csK zSTDkG^}k^LMg~-z$HBqoJ}{Z5%USkL2l+#)m{*p8pFf4-nPw0Coo@>#=j4M-hzKrE zlcb4HjmctXH?uowWUJej!Hhbb=s=he4c@cj9mR*US=aq}YqtP??(f6Geya^jEw6=# zB~{QICk9#4`jA;)3VmiH@W{>woPOnX(A_?bJ9~P-w%}PapXu?Bi~Kecx|ZwXr8Dc` zv0MuD96JWu+l^tXOBu`&KMh-6E8%tpKjhla!$jqLe$DJ9aLai*{ub`A`PYpYQ!Rxp z!e@O-h%Q$5uZ2flHzEGRBG@@e8^VkA3yebfhZl~*6Kgk&7<>x%K7^vh%;Cb0 zv5T ztl9hh`s7%}voIfh8Zj_|ee&ExFP65m_va+YNyz+_Mq99_LajKw$c`U0PBNr)C#Z-%%#) zH?5?pGb$+L{wOL|_(U$fH%O$rhTIEdXjj}Ml6KofU%O-)LN$yVXktG7I;Pcd#CB2x zOO>YU-fQeOHfW$*NIb@U8)onOjTi0$YeSc^k67o{gfm}%!#d~lxXf-4_aC^2tA6dq zql)|gr|7)@xq9C?ZbfA`l~5|HgcR>{U+=aQnPrremK2FlM#IRA$X;2Q$;fu@>rGT7 zrTIZar8GoCX=s1X_XmH#`#hZcT-W`2y`C>{F**yvpX8xz{{v9r^vy6ej%DL2!r!s6 z1?pQS^F_UK;E~@?7`pcrf@3;h5q}R1oqq}K1{v`5c|5!RRui^on_`y4Z#t{x4PI_P zg;@{2(c<0GSoGlx8VVb8+2Klh7(UX3-aJ(J?<}1keGz}_uSXTe8i!m|k?%K)?CD!a zlzcLYXDkoCUg{-3lYE&8+wW7Jwh&&Hsi1-Gf#g^uGJF zOU;tC{#U^E9FfPQA7k|4KXr7;Q^rItr=g_skUn3oi~bL%kTNq^m!4Kb1beu>P9d(b2SydT0;g_ z)bNbsBI(}?+`L=)iWX+yWzT=>VS{H}C3#$4e=x3*eZM3MwjNPLT`G?|52n+i{#jI& z;{olumRFd^{oF47E8%+-@A__-9{cZM4uzQQ_|?B zXJJy*Ioo}z$Lt%lCvVI@6PrAq^ zX6W;BeAYtXTsK%F$c6sKX1F)C9#lY>&m4OU&0~}JvmaQ(Zq5g&Jof;3`^N#?w=V*( z=$-KXq&dixJS8u#8$;ciQxN2G1y)ujK>32N@X_WLC~F>oulSc3CS4_Cf(amcsDT{W z`GXBPF;X*{;Yx+*N#?XnAhy~UQ8~H!c(FMHU2UE+pFQ3&%ay;fBQDMaXHSMjF1BD? zcbHfeMU(X4^W@cyQRctCZDgzHbE;M&4`xEI8_6;;>>EsFAN*57h1%&TbNmW@XgnV^ z`W9iA$uivRa2X5EhT*XC8vJoQfy-59Q%MQrye-=3Frl)st7$fS=-wi{b6N>ua~X|R z^~Y=cT>34xmK>M-M78rb;LUaVBwb_)vGu5+EkZx&P1W(F-ERVE|24m{`1&!DyyF(R z?UF+V1RPx05M&g%Xc_=`F<4WQz zeT!Tj&|>Bn??RB@kH4iEJm=j{KaaD;_WK@K@oPMqnj2A(L~lH|?kr|+N=CCd1{HNv z=$FDJMEYA9GhI=@w1xQ7wiA=#S+xsYus#gSyZY(MdS^`5uEone{Wxs5h`Wc+qwho~ zpu)^Z6cV;I7>f*{W^57p7ShGeZ@5cm-TJ|J&6gmZZ!fZTX|B9YMYCBup&p8AnP?NV z7rzvQVu82>u5z*@jtPe$%W;JK`?Vd`CqzO^-~mYU$s+O!&q>Rzz3@nBF|@8)0jKOD z;G5eenBzZ4R@+R0OJ`m+jyDmkOIrAA;ONLG6d?bq7Dc71$KqzC`;u4(!5W*3-Z|B^E(@j(^Eeuq6CjIz)v_4-lClZ8GQI9I~R=he6G8 zm|yaoW*)akpYttfNqf+>H32KnG~n}=S9pLZ3!Y@_3eKy{7NkdvV$S>z_8ZwT&fJgf%CxW(?;+v%!0Jxn~=369Nho>g^caD z0Ip94lkMDoU~dQi)+NF9JOh{-Sw}?nDuTw#y=1LIAbsy1MN?gi*f}23P!ehZyzV(* z<6gkoi0OLtLekOaBwQN?&H0Gl!&VnV6skqG+=O8gFlboiUuh#C;oRTvvfr z33cR|Z3KDK-9zRdoB-Ry$C%f?YM}UQ931cqg`4uF(DOHd+go?Sq+%~Qo9zNu= z_6XT+?*a;g*3kVt9?md_VUO`FSiaF3+9l#)!38B~;!BW`%z9qm)Iao)=saZSdf?FS zY%a4}hel3T2ZyhGuZ|bt<&>n2|2NCO9A22umBe{DQ<>YA-s# z4oy5r><^uR{4I4PFJ}$o@!xN9wCNsga|oe)kNJ4kLlVX%CXgt{4EF+8*_(a;G}E~!jaM#>QQD4 zmzO@|^u)P9?v*fuhN>{3zzc@QPQq503Muo0;1SERR?K4}tELXATYCuU{Y5JK3FBP5 z3+&}i!+PNk^3U!yiF_T-tkCo#t;?gx;1f~W_NavseH=kVJYvbF)yCxYmRzD#Buhs{ zMesI$q`rF!>A$b_Y>l)ON_t0AspLIaWSN2=?iS#Y8d(g=5)sJW5)(Yq?MCx~bJ&n# zDv;W@LD13tkDH-n1^mY6I6g{JFz4wfbXV^{>$YpC{puO6krv<AZ(VSp82I^;T@d9n+-INMDye>)guQe#)HH62#DN zjs)F~8DgQAkgy^ov?C6F78@tm`OY8f&&e(}Nu(T8`7uY`$@K0qe=<%ZgC* z?l=h?5fHn_?Zh^|iu8r9fHomTI4jPB$$zxrr$jti?Y;tDUa=)5bq&nqDeFnOSqAa- zp+s(0EjhGH3?^Juhl%p?a5%vfW{7Tuc{=l;JI5NNTTI{!msPR+ZUQThvS4?z3A%1& zgCHgjJVOZNo^yj6(Wl{aN*lcKD}aq2_AuhD4c#*ep!vx<=oXCx>nolRcsqdGAHyJE zG#g4|H-KBKCR}qpOI|FnfcPMNhz|KoqVH6bGwyPnr#XQ2c$ZGER}`@p9pbnwL>qHN z{Ao+?Ns<(J(~#Ay&|<>c^;B&Jm+kMSF@rN4g)!^P!= z5VPzo=Uab50&T{F$U{$fEc=Z#L}!Bpr%SO)Q~B5ay@yx3#`ABMU4SY61(5dnJWLNe z4x`e`fS9*~#_``Of87Z?}A+3;XtEQF4=5IcEQ5@DP| zHVqX)n}a(U&7BMTbTS}3Iu^d=H-p9=77V1~!0WsSyy!6{GuxcO!AO{B%T9*MuqQ-# zl>x~bJIe@5Ekd&orZ_U?6gIvCd}er>J<}IO=XbBgI@H7f%m?qGkZCrln z?j!ogRT1xfoP;3-T6D_%U^Ka$ipSluaPVv-x?L~9>y0eNO|V33|Ff71SjNTx5Y$(P*xARFFTLHiq}wge;UT6$p~so>hZ&pWPGt~ z9~Nfj-r*j1WGKFXM75ua@9!^%>x} zaH%V(oH45FWw_h<$jX2QBO!j{0``?OaV3SZZOI2H0W9NlU28; z!Q>YQ!MSZIsW56H3M&CLCZ)i>^e~Xj+6ewn?V-}Tl`PF&P6P(yU}ez&`&eWz4f>&s z-gT=DK4-loeltUepKCCyJvIX-UI`*+bc}(gy9i1hRiM~)DLDTbVRSRDk%b+z;Z)Ic zvUt~NQZs8VIe(y>{q-xGv7RDK*{RK5?{kIX%9W&U= ziNU0Nt|6FCJx=J`jgY&wfqZoGhmZ<{D3U+EnRGf6L7Q?UXx_C%sx(fL+*N$b=noCj%>g>} z)9cx|ola?(6yDDEe*H$r4u%qYqhR9xt(SSR_8HMsd`VPf%%ROM7<`vz64m$Bq;OIJ zDJuHS5+NUY#a9t5@A!~Wl1bXTf04CKTZs+Fk_q735mR|n@~P@4>E#)a)8kgb>Yy+P zkMxC|r?0{tm0OTin!)*OPeD<9Irwha2(H;tAlkBzbZ6Wo)BWR#`wvOX$Pz}cyF-lq z7alyQnhsUR!^l@LOX|6=k*OY)CDq^3s1wz~0c%a%W}`r!)I<@rQkGGznMRHp&tNp) zOoe1AL(=fCjtRQnN5}X2;Leh5G}Cw!?&bBgcY+*U^wIb*@K$t?X({52z- zc#n-=xQ7m}NoJ+wuhi8Ib<#v^qQCB`Qs<59@!0us7`yl+p5uL^fz~%DBm9JJK7WwC z%)d$82aQoP>JsTVCQD|fClcTI2h{x!%k=FN!+*N&ROe$eUB&GjCx7>%$lxSAzugyS zJGi0MS{1zi)|y&%D&oU>6`Z=Q9$S|sV`JtxoLOdxLP}zIH*N(UU9FBg?iNymeec;r z&dS)o=P&K{-$@mhThj5uO+@@j0M+!?!?{k$SZZaAycap-bAL9Q(eKApY*isu-XECX zdpsGH2^_|v&d+3)FA?UJh8(!x)m*X~btg4X3(uHQ%`aXlZ22ytL;_$k1npSvJBaT5pz8^QHMI9&ak4E0ZU!^3?IuwSAbB5%fm z4SNf4LMPPG0}%Q4418Rz$XEPIK*uQ#Vmv!QVH-C`&X!=}BvWa5fCssJIfIJC%5i?3 zbtJUUgyZafVhRq8vAeu(k}S?U7rWF60+l*QZ-x`_{%wN#up**)^)gfb!~VjGXWBg3_emTFZfO!8-Mk#{pA|

H86h7ZBD_1;Fn?Ye-sF6Djc(ewO!O$mNXw&W+#kA1ErrhBR7#Z`f~ic* z8ESVinM~1*BG+Jql+V+Eiz|0=y&MzZN3S_~@!uQ>?OX@6{W^Im$tRn`R)X5U3q+`v zAy{9-po^qrO;Xv41;S#;N_{dhV6F6%&^%o^F`8ux?kB&=kEck zt`$qm23m;toM1M*uDsDkXc1{2dr1Di&nBn-BoW`quJAx=3qrr@$-M2F5PSkfvQNq0=ly$nL(gtfAC3`tL{@``93W-OUO^f~6x_u%fjwO>GI- zTzN*47k?+Qz{z)Y5jG@868=O}Nll)7H;1;9=^lmBn zpnREmQyNA-IfcWRP8>)LRs(Bw4o=^T0KEwc{E;_8{Iw00;1nRqufBI3qQhPgnO-xn z+4Gp0X?CV@g2)=s*BodNDEBj_f9Av5ap@d$pcoEjO#|CK3NSM~m=t0t^cto^UYb74 zUCjr*h%T_%au(DT(%@dxcJQ!w1dE^uIFOuE)!nUdZ-JE1I_jG_)>gNC@n!)cr^z+>SI9eU<2G<6bkpa?wB`s7ckGfTgkIM z*;L`}JdSp{x(MHSRgu)6 z60qg!5+Z0Yg_XNjLW;RB`1f*oDb@-0+%bVxhr{HKog8{dePK~^GoFjirZFoDY2EQt zMC`L1wR1Bd?_y`uwnJ*LFH0OsokT%%^c-vs&0${cXotJ6UK-DwOs>Iirk7C1wgAIL@1TuU9zI)Ek4e)W zqs1qRLo2zS2MGZ>{l1N}s!wC|-DmjYSvw9)eu#g)?x3WIxF9Jw3q{_V@FYaLNodk} zIO5K6pUy=<=+#cn510lH{4_|2&H|qW8F0cj0%jcQCevGniBdJ^r&3}d=D{Gj-r~jO zRL{Zey*Z$fTmb6(mVjerJrtL0g}hC(Aa-UIY}io8`C~3J+uYnhm6Rqp(qb4y&X z?r#^m3_0VeFSqcyY&kkEy+|vK=TjMna4fTs!n8RmEnQcz3kU%=1idHCi>{jQdHchiz7RuFk?97uB>KC!PGNdDb4CM0+j^JGpUt?K;D7=Ftk zyGM@F#4W>&DoT+%U%g0I2$%%;FibVxgi?QeU_u3dPasIlYV{Tq8QCbqEepfJq(euorYxX4PX-ef~E zB*?nbKdj@q7JBtb3bkXsX-;)L!*ocHxn`WVQooLj-r#fq15Z-BZygOzKEl>ooZ#6! z31wXBvWQ#h19E1oG3kD}mHl)_h&XB-W8|!FlV`?f$rDFMnjjfWKL;1kPrDCNpJ@`T zY`YD8ZrV&^9V#0Cc8)Lx*-7NGmo-e~vezqn?CIlEUF4qmO?t*)fJvymPe*ParwLq# z1YNI(=baKzq_YUOsIS3xj-#P@Fobe+2ev)HmFTRl;Ge_= zx`tyvP5OcCF&Q0_`Jk1FQNB$No&S%1t;^sYkt?7MJ4&hR7gM}*F`05`HQFi^%6`hY zNPcL~#)R9Cc(JP`NuYl(o4_c5f{$Q2$X;6nS8}U4U-4q_4GIFOVM?M^Ou)~ik~qE#VcM=gB20mI<4uoV z=8#e$Pbj{W%n?pyd~=^0COun6PD@k}{|}B({>vH?0zDyi#2I$(DS?f*Zagi3Q&Tv8lI>?QCB+GTHHO3b@HUbow;KA|_2AgvSrCvo8~&I*C%XTR(KTEb z7h90Vi0lraPhLgR%l|@g^8G`2TmArMjs>B~+a#<^J%Kg4yYc9ja8zhLhWKSUTB3!}lP&#z zhHiOVOeNQtGucVM$in?LAavdeu1_e3Sx3CUrEezped3;<&Lz^^XbkS>TR_TM50+L; zBsU7AuzC*k2fMt47cK11Qm~hUw?BFyQBW^n2@x z66$eOdzw1dzKTN8f+&1+Vi@Bka*-GG5EqD4;lJZ`T+V1MNNOB`W251)#9$m7uxlgl zS0@X_**joM{9&-`D1{;A>HH;cHTmn<%kWI{Cd38!5~*%+vfF2n^u4wv8VUu(CaRZa zc4d-Xzpn5m>g>Su4S!LyS4dE^?l-Qk9KimXNdm!ZT|t7^Z6v?n;cZ3*`!8^Ro_`JR zv?I~jphI?QH1H%k$mOb;Bgfv+slX_{_ zV_{T&XN`;c4^oB4E+Bg)2wOQ19Ndr>sDJo~qtjB6fAA%Knm2;xf;OCOUxXTsZkRr{ z9p~VAC&gOXVa0&YixsOWz!bawYjFv)D;ebDwqy&19i=lK+D`{ql{Y!q>U>^l;0{ub#zk_&~| zM&#j_VNz%qM@;l5g1*r$dV9YLdZp{*^UY&4#c&($Q{d*O4pH>_D2c11L~!HQf7DYg z1jn(%v~lA}^y}J+DsoA}w1=?`Pp_snFoRz=d{36!ilIY3g*?t&@r{bB599_;TE z=W7;+LhMChzGH_Uoa645#bh1n{^<&s^D+k2bGvZfI4!~VuD2MFoQkJY1vE}AhpMsX zS*?*wB%%9ItZgy&9$JJ_bsJHlB^QI9bTufRD`&J9m@?Xf_vv816TNj(g!R@NPb<<} zX~q&COx_knS2k?IZmm0bMLiXHA|@Ej@mU{51mo3-QnV?(m@XOo)M$A}gb0*p0q@-j z_~N@2q8}K7QQRpI6>ldu-h?$;zAj*XJ#r&o>*vC?3(iE$UW6)qJ5FV?dgv?@9c=me zlkp24WxgItqsLZ%q&qwaCLOV+H+XV1!#k5^rYg|3^zmT+NtUtLBxYD;Rzxh5&;$49Qs`hQBIbY1^!4^$4jJZU-x46Fe7pVS2bH#B|Mpu()uT_hKC^$jyc)_N%#_`#<9KCxdi;=pr+?XD(N~4!20Z zr~Ccy(cNcOunh+yN$R5S%=8J7B!Am8QvXkn3~kb*n{J#U=P!?gw%J9bGDni*5UhY1 zc|KrlD-0*O45r!6KvrdqIsMkzMeivk;FzTwCUJXJf7t`-<2?Z@Gg|2PD_w+CT9fAE z&)8RIIE`xUNnX&!Z;fR!dSKlFwCCGOm>Vn$At{pJWI4>faV;ZC7lgp>N+WqEJPi(b zUm)uq+S5zL4~cw~F*&(O3yjDQViVp;F6B1U=F1$)OfQZ7a=wZ>|F*mvh_;yVJxne{4r1j=k-A#rj!gDIX;Ky z!7M0Ni3Vp|F5lT1051k+gV=EgIQY^723TcyCU=BjXc%D*Fuec2An8z43$vH)Dn|`BDtES+s)fs5_JBAMTXkc{UGK_gK zL{H_bqvg&E7|<1tJXvS_SYp+fQ5%ltZ>8vM?-Hi1!j9S5E(Z;EpULR0Sdz8z3-h2? zha%g|n9jdT9G+N_oNo=($!QknAy^LcMk?vO&Q^MFrzsh-5vNODI?%IrmUy}(4fn_9 z;Xt?pny)Y6G;krof5YXN+42I9t4zd@!(|*ZRut8Cucyxf#L4`d8f5IoeRh$Z7<)xN zo2fR9CnzINM4wuL5M2P#t3U9aD}frX4F5#;Ld!^vB^N|Z?kZFQzWI! zBgxm-*VrALCv;QlVtiKh0PCLzpuW4PK*n1_ux$K$ENSS%;n<_t`O6lCE%R{2af&Ot z(otGG5M4_)NcR8J8{~rPMzf8XMQpo+> zLzyPSUusvt z)Tk6nTsp~1{Zz8JjN?DF&&OrXv8WL0fFC%IH&1$kfadmLq23I^1p^kn6GX7!s4K2- zG2;5dKGXMCo>DX4W)yW~(csoX!Ll$lf&KS6f@AOJ2-2&$j_96Q0;3cqL0-l;^!c+; za8T?6zW6T!--X|xGi)slru{9WxN$eLvq}gRpE}akfDz*GfZ@aqj`^;A5N1BegN6g= z;K<)UP=3iBb{|^@kN-`E7=x`;Cea!*pKrqFfibvBG3(e;B*0 z9V}REn-bsMWDq`od=2+~Mxo~8cbH19L!PxZUrTH} zf8QBler#wvJZKQ%?;I1r(Xjb2AeO{=pv&NW@EdS{PzSqR<@oV8Jo(420pCG%IiIU3 z;!kxw$d{V8ldm_?m_JK*4nLK54|t}vAfJ~3Ge2f>>|$-OsObdxdAHz2>p1?I$rJh4 zZC}CK%R`{rnGZ_Nv%z%ben@$n3SIj(U@*UjSaE8vz429iMuV@PHHZBv~9+1V0!;4X7AdWcmHq#wt>xolZ zAQYWwh0RVIVEfwXJn< zf1y8Clz+e~54zIMf%~e>uwzapnKGfC-5Mqe9~-9PsVCj^SNKY*TF57-o0nj4djjqk zs>Pl&T^PJR8DA~YME#x^qVDsDBq&tDMty6D{1ptxWmdrO_OGjNsVU4lEiDz_Z?q(NjhTzE=i8d~F7_9h}AA(c24}dDHpXk*{EQ zzY0IoPoA$VKaIce&;q{mKsL<3{t8NF4UikRT4~t4?YK+oHKw#n3G^fz&_g^8BbhI_ zX~R?ecte)%7_F&>mk&S1oekyaV>}T*@wp7? z0>VIF*WTHHD2m~ z(OO#Ar!fg1cW=biK3kB{b-^L2NQ_SwLG3O>d?9az&-HF%m0(*XbllX(Vr2sA-ZL-j)l(zfzom0qVKZ-B#UI>Zgr>O-4ioNnao5;Z#xMIQReWI>JYnt z)nFR2ibU+40bXn8Lh7axVmkJWtdF}y)QqRFJ3|bZ*MIAY^uKi|Z&yTL4v3<;kqOOo z_(|?;aUH~<1%8oUjkA44*7ob5MHWER2`0pc;oX@xC2{2ecle?Bg6< za!^=sUF1p(gkPvj}3DJ;8HJDu|z-m{Qm11{<--dEqlO1LW4R#=8z=+OKCAommP-FsLn6DvzdRWdm;bCA$|U<4deN? zZ*}-TbaG(o4g-F~M+3g!nWvDoW-0$-j5_~R+Ed8xSi*;~Rs2ThUHrE5%lU%*g?!P| zi};WKTJZD42OxgcRK8j76n>SdJNt1g0b^#mU@FmO(+(>OQr^l6^wt&NbH7S_RH-A_ zdiV#m*eyrIHeZ7W#*%#BWyLU4E(x+5%Zb0ZH}rblCyS03(sOg`sKVMgSeH>s4J;~9 zXKe~D%KL<0XeYi4Qx!g(8B(#OB$ z&~sue{ZSQ#>r3}ybS&2w5!{Zt`5pM}mH?Nnt3_#AhHnQ8aAv(H%4u@^08ba3pqPQ@ z&xs07zV5^GlX6k{;tSLqXhqXvVZjljzo;}`T(Fky$Qp#`w)01 zxt98waCZ7;c(6VeUdvUJ2kFxAPKL{!Z9f1@hgjmpWi5vPlOX4MyhvTl9GEhiOdO3E zxS_D0yxQ#m%fI>&3*!mEK9vF)eiHk6@^0F+*B*l=S09N;h3`wVH<2w#V<%O87N4iJ*YoL%{wUbm7Vp=R6BvvwbP zd91?tDg89+MFmP#=F(4JG~v;T2;f>fV8VxVQq&kvw)rVS2AAV}oG^t6e5Q_nM{C)@ zjAz_+ec=YbcaMe*#uo4mqCs5dMK+G!OrdysJoH}orjgcf$JFZxztQ@-Y-K*>VT(v zcXPiv1veCy;h~Wv+}mA+?S5rgA(V|K7Z2czrH4@^bvLFAncz{^WbC<-fNMD~TT@vC zzV|qbf5tW7)y<74Hti(VSm)s2mMv(e+D{eti{YU7cj|cZ7cr~c4Q4ipa9vXf_DNd6 zD`!2pZ_>uD==;p{~;f~ErEl_gV@Ij6KSi(2(|HDh<;M5u<2?d zwr_fcg%_{mjEC-c(QY2TxqATpc$RqTTxO%W+k7^KJDaCDM}d*}9C-UAfy#dhr|S>R zgw3B`5e2VF5EtwP-t{&_rQ{;CUGjuau2wL(AQns~{UAbVvTWzANocxxCNk>laC2G$ z&KvZ>nAW4XqtXdysO_TP3;XHtvgdTOr3E!RSV667B3A? z+RBCNJY9%~?%n1x8NW%~olj)2yNry^521H1_mYJ!`t+etJbe_Wh677=(Y$j$lQ>pH zN#-p&(K3YmBYvb!ZZVwK7$B$5-flE?bcL`rAIbRTQ@}(-7QAnVkp5tw#+qH0jmkOy zu>n;@Y$#F0Pu4X=XITdG%;g&yl5zsOQ6bQ%{nfZ7%97mhw55*>gNewNY8u_6LUO#Iy15X5u4rPU~Jio3c3x?`^2}CjR$O*%~ki9 zNjLYBw+l|tsqOM;IWLS(HT^}Wy&I#gJDu=-@q0Q!!Uexv$wjX;Q*5i_G$-ZrwEX8n zR`&Q9iJg6i3B4jk6Qp-A`>iA(!L*s&R&@c-Ir;EX{33|B3G+X6HNobRM5s-ThQ@W_ zFnC!TD&LHe9iB7jyU=>7HWondf6rn!I6Y(yY^xZNG+RcgaTnRXbdWYn@rc##SkhI^ zbuZUeFlxPr2y0VGo4>nLqXHAA=-)%CWmQ1!y}5oe6FwTe3&w|gwqfksY)b7%sHlhw zdYCyfnRb`yRvk8T^d6t4b$asEBg>YZH z7pi+N!RMjzAj5g{x$VhSABJ3uqQk6M4t2@NcLa3TwvG ziL5gL$yhRTmt(*@|pQP<@i!}#-2#F{A#S>}A$Pg2)v4u=M zF$-nipQfhL!n8zVkd)kwCH)hPN#%15(!ZmQ={z?X>#WS_hp?;kzRr32^P>)Js6Rt* z8h@jEWCL;hpUtQp$ocLqj$)y_7wUS?M|S5Fj-76gBB{2No~We0cP3!m;;n>=dyy?I z4d7OP9hx=EAyB^vW-2suJI@_BQKrvzz$U?~%t{yxFMz+V0$_T60$5bO1euCv&=T_m zDW6y%f*1(6od*>rr{Gxy=jRA6hvPc7u;zFH91kdk;DkeP=gcFr)?fn|It>u{xHT}| z(H~Cs6~a$_3W+akV9&-Vj>+?60i+S6&wi7+wJv?KW${v23XK3irM>@3jGNFH_fQR=8V}5@Y30iTJ z=&#hmaR7=b1_tx&YhkqTt4X*HC!#JggLLfodZsn6DoLAB(j? zuFnOYZaxRgAD)CAp~moEMK|FkhCxbu9hm)E0G0FW;E!A=Ecla0bdS6zT{}f!qGSfZ z&rmRPNQVG}O!#Ue!vC^&8<)+GgTls(FgN2ZWM7oz7dnaag^s?1lMzp0OaD>0)Km)f zb2v{;bv+y#ybgO`q;mP7=TLH?3e>EP!T)p+)ULYyJr$<0=xj(EM!W`hAB%1Z`ut>G6KjKyV6F8D~ z3HH>k1dHMCM7UCzPI2}HuXiCJmHv&~&hrIZ2YnE@M1fLzG1o;gM4pj%Wb1$k7MvtWC+I9a{@5X0zSqIVY>aeZyswBF7N|NF<_mSdJ!{e2~Nw<)2duRP|TjX~R& z3dr12!J)WXj*V~{Kb>2Q2R*%zRQ*S#pIg&1F@5&zDjB+AeGikI8pKoG8N{nNSHTnB zBMBALU_O37GoGc9hpXi(o5Kxde~X2LWlgHXQ;==XP$hufk_;MVJ8uk>r8 z!J7zadv0-@;z2^QUNZqb`DF103-VFsv_a9`RCd4Od-{&|fVH!AB6ISym{#F)L~cZk z{$e=*H40&#bM}MYz-oEdq83S4YH}t1i17mIk|imyuL3F zEoF7=O0JuxZoLN1f6p;KHKS3YIuu79?dBK@j`(GtJlZ62`}x~_sBNN-9igLiqNg~` zZb+lq!j5EU;}DHdlVQEzUtn+Ct)Tf;?sT8KGIdM-N>}CPQ2*Rsx^#gvy6iE*!t)k* z-%kn8OEDD38}Obyb|Z_I%G0M;Q>bp=R&r4?jM-?V&A7L;(e&Q0R7CX#{n4--kNkN{ zJ6dPZd)zs2xThbUj@^OnB~M`eytAOa@f2uo(StX~(qJn82V@^|1R+~_uvgH8TL!{# zV&!KtA$XKc3H4yCJ~TE)JQW2Ro5#wAXA#e?3aXvfPVHYNlKXP$L|f8=j2WlEp7a1n zUy=&fCi;Mx?FNno77Es03~ck>2$`|G;;X>dr9pRRq^{r{|!)j>u3#=acVQ~ zZQ~x8P;rKBx+KRjxGoUgM_jgL*cHzmi^iG@QuwIp4YO_UXCg6d3tgVMP@)kC_r)dQ z?l)ifYoiH~g*#yK#Zcf4#zSu0YtoPyLK+M8NltbHQO8eSBhwPVATo4l34A ztj~GX5(>cJaW1%tM}VO@=T|Ad4Et|nLhcK17;I$WL#;ib@g|`9*Nna$C}1l$)l#GR zs@(4i#~I$aTsC8-Ab)|Vppe&yllsmg)O|&>lqQrk`iA9Pp8Z6wF1mQlq2IQPVfh3u ze{I~)uA(m+Vye2>gOQF6+A|iC-q%H>SviB|>a{m6)1OVMw3V0?=_CUGJ`;`O%R%04 z86DlQ7@7|ik`sEdL~Fw`GVR_~a$!d}3D#5wdfN@Y--!U(kXm2yHHdCku371=pjaBQZwCXLB93 zS8HLea2E`FWrJ;62)xGkptsWpUOvr*)fU#PVJjso-VB2eS^nN(Z> z7!)=n0ipNF@T)1f%Q^z(#*X5e3$YmC&UFWDoQKwLj-vFd#hAHjDxRCwfqey`sPR&O zJwHz)=pMsJ^E`~4+ltp_AeV))Kto9zJklJBveyFeRJu8i*NCP!xZK>uD`PZ7NdhDM z4QSR~1$uR7Ak8~gj$sSkaHQfie&wykN@p%pls1U76>4zb@Fvvz8IQ+oTX2$PAI=S^ z<@VYJu7@cfe=Xs-0^%Atr*AUOyR{d+;xw@Dj|q;By~nbFRcw4dHk~DhOj^<59l02yH&D!VyzDd|{(Tq_*8%B3VC|H~?r(2>R7-*WWt;W|3gWGyzS zo*{|dxkNn90-BRm=~mYzRIObMX3p{>eN$3N+q@3qaOV;u*(?ci>mHKmeV4%ETOVvr zX@WUpbNCiVXY&_3IB!JB>q^%FVOh317yZO zfFENoA=9=7WUgL=duxRFCI6=J6)dkp?3W4<^-hL$Z>m5^p39Y7jfZPYDJ#~aNf+<& z$MC@;cyM_XzMBw$Z~w>HnTJ#LMSUEZ%amD$C_^MOcddO5Q6efd{z$V%QPHSmrYJ;; zl#&L8NaXB&6HOW<6{R^%Ql>^K={@hi@1O5|p8Gu4DQLEefEIuHXc`Q4@Jd6|A<&*bV z+ku5F_-rm)ELOsD6y`JfxsdH^oyKe*MX|4$9!z!I47S#9D!cWGuiK9FW4oqoXHTv5 z`CM%p`5u&wFO~ABa>Z`+-YP~ujEp0@ccc>el6tc4{T7lMT}P&!e@kR`<&f;+lcf4! zc8z^)5KDz3c^Y+NoIqq~;HX-r~XGv!#yJq5a8jVCg>08-BLBe>`;B`hMFVZ^5*w9 zF;m&cUma8>N{l`B`bwA2ZlF`8dEcdoJg~L8%vJUOrb=I4*PQ$2MT?IpGO15IGi~ig zmT*#$nTc%K=vpDWmN1zW@%jQmO%D4jyPXLVQdz-obL#filbJuA!^ZE{Vso}EV~aQB z(ankcToOp=1H*kza)K%7;2cQl0cFyv(FO~|w_NmZo=lvp_Mq(#Yp(0xNqW8O3_i-v z!G?@`css#_9($5=QCx2TSF?If`Mx(uY!#q;wHL|_O~tg{3G8*S3h79Ak5&tf*gDZ) z{2Df#OxWl}W>Ya@c;^>pZOOwk+nmX^#Wo~=_-d|Y{4Xv~Hb0V&Ckhf!M;Y)Q;<(=hF>7dyCTTe02@+x?-h>b`w`uD zT);w7i<1rH=|gQj>Zj+*BD%DgV7CQa+c6mpPaVM;a)t{}d(9~*Pk=kO7E;;U0_!DlsH0N1KSQO(rXq4arlH zEV+pF7Tmt7@{%3j^5~ zkK-(DVD#G#X z{0cF4q%VgqGCaj=QO{6ALlySdT8Gt{H&fk-Q|Oez9lXcj&x@tzDr`*Tc{*?Q5*A*# zl!Y1X19fVFug%WUbtZ?lDJZUxS+PQ_@T(q67oV1l?7iTk{`Mzw^_;b#_3$(#{72^_moUTVplCJL@y3dU!pNp;#HdB#Q@#Axf$RNtg zyUCJO@+3{Noct^qA$-2|5wT7fCfu^6oy12L5YYriG@RCwRria@VYe2tLbid-zs>V> zJ>rP};v~}dPm`pGU%Qz1tpsO3(85b?5~P1tIrYC3PdELsU~fVe(RG$5sX^v->a}Vb zyYsS>`cGWIjMZeI#z?1nO7<8w#`-w*dQ(Fc%6e%`%1SmmE`o9UZZhMK1uUTUFk627 zFCA?t$2tlh(J3o-q1`15E~upk4b5w@siX}Z9$S(u8w28Cu?5S@u3%GI_(kq-6NZXs z(_J5);kW76a5m4M+1!2$2a|@A&x5B)&!v3wcwrXls+~Z5o^Bu~7K@RtkWdo2=L3=p zA!M|dGN}mEAv3cdV8^8j)cCDTj`jwjb^k;felAx~zOINnpl}WcCpl8Ng_|I*b)aTg zqYm#IYKzw{Z@}QC+0>(OHdXrE&DB0lU^eE7>@KrrnsxlO)CgcE#beo5eF?T{m=qJY z?4y>G#<0$Nt7z|w0h*@2l4XajVa@A{S>&%QW;Esq+f^FOMkmJbudSHaacg$bwup@$ zn$2cJO=p+=c)rDSdlr&c#N7!>p@NW4@I3h-Es-^#ht8LAUOWR(l85;Q4Btu{!sF-y z4L!EpC6-21Hq(-(8{8GO{}uZC)2k^1bj|Z2`Xgx?o4BW+mK+JE*4@ci5T=K6l~s^9 z*PiHYJxD}44Y>7n5jl`@n#dihCXJGd$j=TT8Rzame0fgsbL~Z_J?;S(%wIq@T@Vr9 zY0+e1@>vqHt(JTUPb69=f{0qOGruqTiEjGp|fbh_Ix8kwWSm3$q^ z>3#T)!|wRN=^ehJ-x97O-AdFH#TVF$Rz|Cc>Q||V7ENkrWJVve zYETt5#Lg8>GxQY&e>M`y$_PbUBqxbvt=5RF-Q7ehm4=Jv4bv3)TZ-8Fn8(c9+lFb$ z?Wacv4nX86W%4{J6~i?6oNN^Xz2CgwWZ7(z{nCI8-H#=zJ_cmhqbK+#JDn`Awk3z| zY{H8}>UbZ|)1mFk?6CPDt|Y_RmBt))P;QJ!&%;zSGIzA-mg`*6xI^ZmWD{x8`2|uU z^*5tLO;SRU-H+eQ*QuLTUOmo?M?~|C@?aKl&xP%iDn{8ARXmIA0x9&75uTNACSR(C z3H<_jxxCMNa`>UPuxyTmaM>Pv;ZISlP+Q7K_*!P0Q1fnv@UCo*aPadgVcF)r!tdt( zLdjveLd#L7$j7b6N!cG`>dW)`9^JL4q3+8`rxYd5;<2RTLod4jcZ-}&5D|Zi2r||> zg^T`>61+g3h(;9S`1(X_zj7S6xBtZ!V|zUJ(U5u<`r#w{kvLU$FOIvA zNOmXpqUTl)13S`i+SJ9Qukt;{^jsiXW0i&H1`~+P`b1J6`jPbSOCq1g1P~&>3PZd; zQQLaU>ScKfY=8PgHge@vHh$wn>UcMtwPX~s@hUQSa^`3vKi7tsiw~ljt}_t@R$%*n z9pd8O#r>z+E9eg#MuVS?CPSKr#II))d~miUPN@o{f4Vk#{KXm%q-x>jc=5Jyj4xlhx@%-91B!um!evjy|j z*(;-^%{lCLQ5Mh7(PFyxTiN3BJ3NnZ2J@APVyWsU*-)GV6a7BJiZ<@2 z23mzQjNPY|P3?62{BNB6ntU#*%7gpU+)G21RoLpo+DvJGIeU~Hz@%4Y(evZD< z*0ub_1(h?2vXmIx-;yFo&be9h;`StxnP@{Q<{iN5U6S;A_IA2;a~P&ScR}s058OdZ zQ?l7Afn0XpNWN6BCZ#E5RITC_{Jp2l3D2QHz;B4U&6**^<@F$D%tBSWp?HF zV|x15T5grIAIci?9-U{ah}-#4V!h9W+%HuixvvUIK|&d+R9{L?Z9GSGmPe50vTLME za~G*!Vn8lW*2OBrleoZlH8Fd2keENpAOk%)WXtpgWSpcl2`CqlrsGnieO?Ml9=Dey zOtYC{&eviR{TE=x&Q6WsBsjh?Iflj zbB635`;5faA0ua@;)q8_FZRr}$34$KVEcb1nEdDqj$gTy9E|B8H=7=i_`s3mXu=Az zzw#5&DW6ZI=j4+1zZXbR{0s79Vg(8PSVW5I-;iy1gKU{}lo-7ILTm@WEa*zX*On*EF^k1%gMwsUic|YhU};t#JN@~ zIMPsw#JVI2+Q-W9eHYH`k6|rm(B2JWRW}NTJ)F$uC>&&?OpEB_w1upF$8PptfsoGe z^kH#T#_ZU>P}V3s&O{pLnab!KrarHb*;QAtDSv#J9IyYIyeozUmRw^(?F%e4?If$3 zWWl_np3rqyT?Iz({8>ftDHgUMovlBjN&AYovZhaxEKBJF-Tc9oz4MvJ{)vC0qrS|> zi;v$>mG0xT_D3T7EB1>E|7k^!nHVyG@g&wV-Hd(hNuyB@L#adM0q(QJ9n@L984c%{ zF!_SL*mv^{9gt6e(RH$H^Uaa8ZjA~iYVqDmK0WmF?Y*qGbq}rVcuy;eGFak|R;q8j zm7SKF%zk(dV{zK==(+_n*z`at_JqsDImXXXvR;GN;EI#QyAn|+{xy#9NyfVQ>Llf= z8a-4}OoyFKK`n(TY{8mzmY?RxG~@l*l#sjhh)W}z-5o^LemXHhe=M6-WyGqUyv1zS z`zQ!MML5Y`yd7~KL;L+Oe47Mwv*}k~_ z1;%N3TmOcjD0?hRnlA@exC7oR zYW%$)g3hjF*u8!Ns_InY<}^Ffb0r%$gtX$Y$}n#7P8qh&vj+$E#p26E-cNc<7*RN? zL%I&=6UTizoJ^Go4yH4DulF5iy^Ysd73?904wi5QcCt)kP?~ueAEk*8n&^bzuc`H~ zJ2mGvtPv>6jl#4x-n;(5d3gTXkeMn5GTTFUX|v66n(Nkpmq(nUBTq0a;5i6)|NP=E z*nYq!6Fn+>MVo{cJEGDxLuUD^i;uG_X&9ews_*BW1pkF`5wQ!XMUyGKDt$};6!g)Z z4@}wBN&E4!|04R!#Dmq;`7%4_CtS@50dqD>V!OADVRg@WUv^td?zGcU?$Jbf_HSc1 zjn>j;c`l~(+lT8kEan`Y?PgE1&dAc7H$!0WT!?#Q=Qu98bpfN-E74bzJ81v)M7*S_ zjTu?#p!V(reeF;OrfV8#!7Dwo&^De-Jyl3XZOtZmGZM(5eI;bpyCcL&x&pZuF64=X zD#;&MPE5>ruI}Mh=-QP44=1(KkepN0T3MMc(Uqntlj7){0bpC*U08VKWVYV%IBTn% z#nd0JWApnaG2;$PCb&6@O2lci4C(Lmm81nbwK9;I&Cz0|*In7TZ+guA$?F=+Sb0(# zs6nP$@2|NND1!@k1(A-@T>M_LjQmwnCwnGMAPZd&lBY>+*k0yA9v6t9LU%iNdcjs4 zZxo5zr`)l7sW0wZbqc!f>!4!VNF07f8>DZoqr-kaqccniEp-{mgw&8FDxPQaf|J;a z#yqz2S>d@Ig?SAVeY^6smmcXX6Nq3Cc0>`ybN*XxxSY!o^3}* zU%JgLKeZlv-+58v(jlDuZ4sw#ufx5en-LnE1vW-c@zJ&sq`hb}u2MdZiVH5#4{fFN zi@Y~|Ul|U6B=>NUZkAA+Siwavjpw{>sHo~KjRgCNrJdcPrD9@mt{1BSX*_w-b+53*p-bTlnpH4kIE0>ChKF z9PO6{bZ9R+M()AF_>VNghoiD?{%k-&gC%=4aVMrEQCH70?22EBai;rlu|^o0Sbv0C zsSIqRH*lNbB6O{oL=KyokaGnx_+Qr^zK)_mV_wSO2iL7d(vgi)~!imk6(3S{Ny%F zYmq|ft!o9-{)m$n&GU3mxH!EsWjyPiZ%iky3*b1N^>iLHqIUu(;IeJoaM-#_uvxT< z*1r6L=1b$T&3-sZf9A$jH7v&h%?tE)c^17;zMcL}HK7B)AE5I;FPa#*jBUG8N!RqA zpqFD?sr4dHy81&nZT443&!=4ic!O;C4O<%J8BaU&2Dy{2DKvF^27RoO$Bus3!OjgG zVDpd6VuzFy*rruBY~vj{=5()!rk2^)T#)!c^-J&3h^^6dxkDdpiTlIFpM8Uudw7k- zr`=@vJsomq&u=_5#RnUkLUGnx0iFD00KTn0g38`L^yN%F-Vb6P+vz-oiMM^H*xyMn zO8(&zr)tvZ;yW~Gz8IZ+Z#nnY!GPx8GeG-$^4z6U5}dLl5;6& z5NE+X)HQ?aq5(Q#rx>gJSV$fC@m)?E&Q=75v)?bI*ra4hlrDXQ%qAMR`5pA7??zmF z?f^|{jh_`5!HZgM0tG&u4vrDugM`KXl^C0d9fIGZIdNu#WQh6dNX(Qf;-ij zEy+H0bkHNcDHk2rO^05?OZa3p-}fkMf(v-R*J7UBM4_*$IF z>uS^f5vOY|8_Z*W2k*hWn9+FQ>`R>Y_YkEY|KhAud&!@lG33AZqe)Y#EpZ?F2ZvWA z5nuQFWLahz$#l9vJ|2}Ii^{i=L;v{tetjuPZ<|7#%4U({Jg=)VGLy5Pd4XGcX$BD` zZ6nFo4wJ&DE#y?oQIeZl#dGAUutY_lY|lPTn&&JglVA9g#hD*qROLU;SK0&ZwtvF* zNM-n`uEwunBiWqtU3ACz2AcTr61|X=M>X#AF;K4+`*nB>HFh4&y4Fr*UxHszQ?n1; z%V6FwcF{Z>K4T_CB+8M$S*OU^Z>LD!ia3(-BbnI!ts>Iudx)rd12JB^o0J4pkulG& zk~Kwl$XwMC!m-Q5glZ;*q*&C=`#qGCl7el-zB!-hbw`u7F~%fFRf>#P$s&0E2HEs6 zmTV|GO^TczWXOh!;u-g}k9xbU<}7+! zaaElJ&fTp;K5agLGIq7t91~0BM&1XT=2Db+Go7TA4Z}E(>BMb$Ir(z^9{IZN3#w$> zlCX^pWc#-3d33={p9ASWU_5VCowOU z6Ap%|3t{>w;k0`JLiSHf*eYQmbd{Se?3eQv`n_}(21)A*HS8{uK(h=oF7YhT&B@e7 z#)Z{e{-y_B13i;3$+LQx-G$zZF_7(UtX4_Ed#6R(MV|q@A=&NwJ`j#>L>ag zs6f&FvL#;vS@z)zoN;tB?ln9Ds@aaD6d&RIn3-lDV5MqulML98FRf|U-DBK4h^m}2kVA;_c{FPCFYtI`H+4rf~dtU*k`FP@uP)R}a$_(s2EyhD*pZIe<`{1pgRdJ2`spYWL23sKf22XL`8T1_d;@Oxv>fXFZ9DegkHeo%KQO?z z8p52U$fz)$rSkDG7L8irG$fEBJA3*u%U*&k5&Xh^4@VLE{GWKfWIay&Gl1iq2hr@y zAVyzxBJL@N>9BLQSiLuqn4)W+bmW0!^qylk9nrXqR-HaZ)r*F+XI45?fm{Tee{uBn!7du&b&{stZsN?& z-Nw{At=uLJ3);8i63VSmBvZ$}<3eYY)l6zGq4It9-0A%(q(QO-w-u(s!QQJsN|JjiC%7xIzVO=Squ}rU>vV8iB�U%f=-fVtrvV*>C3DUpV0~1_OLZK6X_Y~q2J3cb0=2H(D>E& z>A`b)to^JhE4yBbr(+M((Ro`LpDE@2aT#6UR>Jc+_oJWtdFs$Djy1Y{NF$Y!y%Y>DM% zR{S%Y4c4WypvP70{DKBHx%@Dj*QLhP&AX@%&s$#oV;X1Eyn**&dJTWKCt-ipD7N?c zL%P@?j>!-0r$6-@X@8s|n`Wg#W0aKW?jsKb@o$$^)qy5`wH$y-({L?(@Y=b zSuovOeEqvEmbooG#b%_aupfVQ>5T1l^uW?&HaBe%@5i9Y)?AlnSzEl=mr;F!Gp|>p zhuUfCajXs(|4hc=4$H{-!N<7h{21cc5ldWrCz9M7@9;^ED+$gu!}Lwpv0`o+c`Yy_ zHS#fJ#(*-3$>x1AK6+tMu??+be9y$YJT~pH7jt`-&LZ-*Fx2PupHrgP{?uJOZzhQ? ze=cNJKULU~lOx#1-8{eWx-8R=y+u#ErBmCz9&B$%5SxF)kULd!pN`|1n@ZMv&fo1I zyE<|;Kchyo)U8Eqv2;HBclr-^`b-LbeWX*9le3$;^99Blbh!~h_aC*8I#k1UXN4p`u8;a>v|KL)m^#90}HXOx{D@_k)sC`iZEqU20il4 zh@JavPR9=SX5EXP(jEWpVt?vm82jnYA`ax!DLc#QoIP7vWt$p%{yvIbQ^;rR{VsO2 zKbb8#F^k1(CbN;}53kx!@z&SaKaC2OJ3NVvA4p+4 zA%*!}*-oFo3Sz-xOIUDF9KAN?8`nBLfO#!(E8AZTx=H*nf?Dx+Wr|u$By(R1?R$H_7we4~Ue^G19B}h$zH=B~OQq z7AlsVB9fPmka`nsVdb2wq(0#RS!Q>b=s$=eD|j7W^3Um{eNGT5x8{8mA}1l?wa_j7 z;W(|>1>byG!NsZ?a#8=(9lAvcWWnaWIC%dsu6^Z!UA?^bqTx%NrksQOnvTG!yoa2_ z`)uATBbE(Fr?9wP!`UpuXj=AW9h3dNo*jDP$zsHP*xwn_>_`767#w2^{9H`df8C1# zK^w`<^vSqND;xd0rOC=iMx?K242j!y40R5Fz`3g&arf00dbY)zojo4M3U3!Ok6Xu> z=Jh%3>93vazb|&IMbDin{cQp1+k5E6Gsp2tQ!s9E&m{UwE|L99?h@C6>tu`67owCR zB2(7+5Z&X|WSC(WQSSRlBEr6r4TtrGl?(2WxXdVG6}6byY}rY+Kd~a}Z%mubc~cIm71(foF&fFLA4I(wcZWiMcTCbbT8eIe;IXjcW_xliA=6E zk!2)Bv9s4tv(?G}n7&mTdu;ujxvV?QmL`?5TlVYOk8#_Xk!>gKO5RKj{PQ^TWd%6p zKS$zZ(}E5B{oR_fk$enIAelwUj*;bGY2@4?I-eY(k-W?HT=91f{Aj4tk|C~YwH@1* z%KxrpCH7LYlxi-Dtm(f|L!0uRLBv#^Z&UREmTnKkE3Je0ulGL^eB6^XW(oK(DYwJN zn)&6H(EW?!=`s~t=60eJZ|GN{>6#`~t12bs6N-qLL<6}c8$$}&AimY{<=NGf$ihK6 z;(a~-%x>Gigv4&!ETWR9g|EuLa~t)|G06V~ z8Z>SrD>r;Vb)|V+&kalBYv@WoWDml)GEKgZ&Xbs2sKLh%rO0Qc9_|+lr(;)apefdp z?1hp&^A^tI%2V&b)kR5AmgvUO183>UdS%k!Hk^8jPh)i>H_>mZci`!O1Y2RK%`Tda zXCGsNSjYPnZ0dSH#{G(74TIa5^m|omZc%`*gcBp`O_U`>dA&I+wf4*;hML{OYq}+8(3s^lx%L;hLM{O zWA)VvT-vmb91opCgu<<4+yO_jus)19UPvKS(Guei4?`1GB_e5^fiCB6bFR*DFqm7& z{o9=hdi%$)_w{O2_OAgBjrHKUMQd<^U@tu(p-tD{ae@slvaAYf=^N8YbmMX_!Rzst zY0-qkf+DkY8Z_<~o)p*P_fY}(MSUcR%GZM}F`-=3R~=Gb)Zyf1DX_v5`CJso35l}}%w32WZKJ&z`o3Os@jZ2G9Fk~!0u z(M^B!+pt`32};kKu!l~2=&lX^O!vw+dQ!=sMjG4Ff`4P#=y&C8?9d9fou4PxUX*1+ z`W7sEv&7MH+zx; zHs<(eOAO32*X3%aSK&psPP$cN1`FMBl$L2stf>)2ksODKq<(BR+D%r(TQLSC(<+E8 zoB9%O&&t71x+9==-+S%|bkX#0e4pjhlXRcQak^5|pJhx_XW|Q;xVUvXct7xiK+*3j zJ*cva&h7k7`?Awn)t^KfA#;J6JmTPhu6H^~v+Y*%;n?1X|we zkp7Q1=zli3^j!T-`fs3;M)0+)otb~AvWW&gqW%q%X7Y@Ulg%tM-;mk{>GE924Xh)9 zKUaOr@paHo>=BLUS>Pq)o|h6?y>ccwP+&{8orxz!33cSdlBFcMtcJvOe#STNvvFhQ zQ|{-ZMw)3RgK^tu(uJ!>kRv@WFyX|lnt7=cvF0zs-=20vp;Mn6spNG576PIxvBGKI z2wyxWBf~U$_b{q|j!A7SVF?YB*xu4+X7xOieHf*~{>zxcNMIcMzU3l|bllCheD-5s z3XIv!LK$`~Mw!<3&SGv$yjgO86-~~4LAM5Q^qA3Pc*gXJx4AkTd3y~jjd#&A2c1#= z?>VY^TAfKo@*Is>Ewo)sj)gUr(0iJZ5TV$>JuUx1-%A~(v6k~G7qW_ECwTGwkhy%l zp%hXQv%u)G1nIDkA}zZ*u~%gs9xI(Gcp{@fY8N1gEqsb4(!E%E>k9_fY{ZM_hm#xe z7344%L28yB!!2Y7dDwN4v^lOJ7UnmI>9Hcd$Getz3~eF>WvfV8!Y~qiG>ANIPC zyvFw;AwjddX^8f54B=}qb=ga4h{JRyy+elOn;EfIUguta$CMSl+D|34C$Nik`Ft*W zAv4;M#a2pgXM-ax*dfK^)T||pr9ZG^ig7a;bMK=HLu1*T{Z4$pcnj9gxz8PJk|$G? zqQIUtgRdpeqYyqI6BEsawPrp-mvu{o88u^tpJyxGBeC73+oT zSF9B(rp**y`Y==Yd+{RS*k6Id9lqOzIrejfDj}E%;|S$j};{1ON!cG?MLk6Q_O zKDn^`hPJS0)JgIZrG>`Tp}b%AAI>GYob5lF#O9xv$6kADi7sB5EIQQph`lN5XU}iz zi#{jsXXUz5Y~kC3bjQ~|YH$*9VKMzoPV6X|K`AU&tF1wr^m5Ki*UN?qLLub z3+RcSG%h&blcoRsMT67)8SGDFsu8)&WJv|vf3}o;kIv(LWH}ZYa*7qYWU|ZlxommY z6sGtxf@yEQMRz)twx%_5iR3dJZ|Ftv?x*RRv$-v_vr(MxH?q5D{t z{}c{|@T{bzszfe)6q)t)B1wxhCME7NB(63W{Y&#m>YQRClbcSi<{l+BOAnE=eFxCh z%8+EgOUGL~Z$tFqVdPx0Io7AgP^-F58m1>t`%Ko+&chP~_38n%O#L*VrwnmHcX+?a z3PSVJX%p|Is<}g%7Cp&;+O2A2dCW50A)b%H!&*SuU5>lcI1^v+I-@(s)9I=`&#CkF@Yc&JbNU;UyRRxf4JFP^j4XU?+Y)r_@OH?SLv zc;3o|EzHgB5c7{($LiOuWDn$rsA@w5YpUmKafaI5@7H&@>I)m`IU{Rqn>~RoDnC`@ z?V?R!XBu&Mmq|K$9`YUrZ-`*qUUEL{0uF7~B$w$m9I@>&o#K)~JvYVBz$a2Hpz8|F z%FPjcNm@??@v78pz?WVzRRQSO4dsF6Jz@ zc*D|_t2TvLEn6J9*h&DlPS#F#j-Kuy7Wn^tDCg><+CqK&@Yn#mN9beHQeBiR*TmQb zdT8-a8@;P^@gnaFvd~W%lT6jHbDAplkJ84SwVLQPT?17njza#Q9b23<@ME(sYHZL( z(;`h=wNnlMTdjpFt+Y_>q#}CkR=}tb1(XYy=hrA@ln7D5F+OT2IHHDiE49(NM+-%+ z`uM&^3x8GVqxBsvtWMX%b@TKwOF|Ft=Z->~?>aat&H(Qp)xq7-qcE#Z3;%|U!jPG| z81z>c)3kI@Zo2`Nq-bDD=}0V(P{t`U6|vh^6}ulRpuQie+AV3Dv#a=M_|i(1(aCEzkWFaPqoWqorV%t zO&X5&f)QwHu7K(7DkzFl#sybZaQq}?TrZ`9`#x*n?R-si`m2dPzB;HPp^dG{niylG zf%!=qDCMq(`*hVYT3-pT{87a0KxG_Rs){3p3fS(bj9$ak@#}moOj6Rs^X;Qi`oU;q zc?KAM)d=SljKSo!W*C-Wjx#P;paWXq;HPn@cH11YBTaGGA|w3IhVNIpXoi-v%yG)) zv1n4QhdWwFVe(}?oZ>wiV>(7-x|JQAF|~n3@^isLHe>GXg#xl!&42QK~fjoyT-su)$x$G z*cz-SIKqxDTc}^_2q~N`c&k~#rU*xfyX*+dF9{(3h7fG;dP4jUA-EXWLrJJR{3EW= zebEJ$XxV|sk8z;=dpvlgSi>oEGdTZl4CqJeL)=?acw=q{{xj@h)Iw(%>50(PDgce? zPVh_820B+-!Q6Tqcp>8i2|KJ|T)Yh=6&k~cIVK?XP8Sp^bfH|R2P^ng1Yu z@c`t4~`ZN zg2#qG5VhkQ-1On|6Otd`%;*nrA*CJc-hKoH;|`E>`3C0df1on#Csbt(LCSndWR^=hD{FTM?Lo)dMz;Jw4ErBy{{exjAzrvo{K0vEa(4xZ^EBar7hAuxYqn9wr;Uk1a zeSvBD-y!?*KR9bDfj!^9!p9@OVA^#Fq+-(eXq+r=I4*-1N@Q__i!2^6l*M{`Y5dPz z3V+y%5;~>#c~+7d^qOJl*QVcJfr=s0w(a>Z=pZ|qwD3c)L9v) zpHanS+trbkE8zocMO>C3i+!?kXn#Zz%XY|Pp^PHt6)0gxraU$-ltJHQ{#x@|%;d2$ zSX?NMCl^hH_`UOCqkRZ?Tw4ml9Sgu$Z#I-!_`>4{?htA00)aDJ;e9c}OEDpg{W=Sl zJemit@BagdB}*aV`T{tvxDcS*AKYL0K~l#gsL7lP%D*Q<*(?vJ+98Cd22a=?C4?K1 z0DoRNLWl&wvf%{8V%*{5ZWr+Jape6EoFRKIz#}nVFo+@$c6JJw&YlkP#nYi(aXRS! z^oC~LNiaRo1+HJ61u#1ZzMT(+wv+yFMAirXeDs4e{r=$9;RQ}F0qzF4z$yNoHu~%e z?e}Ja*~LJZpEMUb9|eM;+YFFC#Rr+~li+m5RG9q33)WFr__&zI6s3BA$FAv6Yc&@} zZJi0*wgp4?=ULEYwE|9Vn+cI_9w3%D3kIhD2jlhU!R@WnVP3&h*xE83f+~ejY~%os zyX|1-q&}j=6iIy;Fjy22}?+LPh=RnS(IWY6yRJdvG0rA@f zFgwQ?{<%zr&fD$~s^$gao-RtYjZVrTd=E0B};1A0;g~H=| z+3?@4H0TUSf$%jc@a=ONRB7ym8F$j)_mLzRp_T>NSj);TsR(_m~3DDA+2m=peVUJuaJPz6pM>OMM z;+a^O$P(bn)i{{*B>^V#xe_;tSP-PfK;y3{*tRASA{Amlc0xSgjhP5-y2%juCl02X z#=r~37>LM@hp|yHFzZVUT#Sr`t7mqA#$6#(~E=bo(_ zeTK%1pW%k+4M_RF0~7ZSIDNMhs$D*S?WbNSU)BqzLGR#8NjpsW*$tmVKZ3SZKMdU+ z0Fwvec<7!uzLJs0*&T9t@ZoUmTCRZR5+g9`wGzfFNMrudVVIo%8$`bcLCpFOq{$3` z%hxWLzUdqIkNXWn&p$&|$XggncnlTOU%|4HXE3Sc1+3Wf9N%&_Qz%oShY zta%rB_x8Yqv)%CAp%=PDpTOMzBUJC`0qccbuzhh4gr|Ooedm6F^2p!N6d;C2ym*aD z_zzGRltAG%an$}IhBe=QLu}?>c=2u!oDTkgvs@SWJ?R6R)_!Om@eO9g^g+U|zaZ%O z1DXornDapltv~#O?WzA@>l-orv}G7Jj+8+C>YrfxXb|KV|Ad{T;z(Sj@WmE!93c|J zF#SQeF8&>G)mJ$Cx*tCM{0U#Si=nB>5M0|Li9@<_Snef@$;#rm?BH)m_ZGwNU;cv4 zlCQ9y_JE-6Gkl!-3+7+`4vz}If+l~ya?eWQi6g^MYpWC<;8|47ZE_e_%xjuAwSubs zJ@}D+6Mm>&2Op>Fz>Zyke9Nn_@!e$rt`6RW)xjN^YFI-rgSSEhbl<%LXT0vf@qoKf z5q}?6{JIXoem5b4KeoN|3OM;+hLd79q5tU}$UNTw=d-VX*3%o{cl{bXF1`U7zE?qd z<~2B-dmHpeG{WVbk6`}k2jF(K1wtM^1%vHRU{y>Ld=lLNyOj){06D7z<#UO&^G%maIZSyw0kFL7j#0@gAQnre+y@WI{29IEfiF~hlo>MaN|M`#1?;m z6Cs~LxaS3|`uPMlm$!oNmPe2q@&IymTETnmBUl#J0QD6&;rp*<(8y?odwx&AW5y#; zz3~7XM?QjW>P>t!48}t;6R!oOkxszf46E8Sh z>JMXOr$WgAA!I(92rAcv(DTL>Iy>Fqg+0Kvz;SSv+Q4cpD@ZU+@THK+~!yTIJC&3pBXBcy9GOP>`z=aG?(7p;#qaXm+la8R_LSV`>H&B#wg*blw zpItQxst-6ro46OeRq=zL4$d%M$^pKJT;PtT5DNBtgYO*|SWq(&&h!YO`|4!a6f+TC zH6pBT7s9^|KF*i*fN>Z2&-E9=pprFoueXBL5!RsJH4aK8Ea8QgCA3Z9W9INNu=J-X zRNWX4vqDWEsM!GI3Qb^IrxEP!HG-@urr;6o2%o|n!1Fu5-genQRHOh#Ugbagrz=Ev z*ubi{wh;Io!MfcMR_=9yiW(v8yNqytp97ekp8yRlPH=pR6I3)#fbAm@e!g>oKYKmk z_Rh&*deO-W%g4dK`(r^havaQ1;NyBNN9Z&g z4@Y%fp|E!{{F0soaa~hkJ@J4Yy55i!?FHF8=fcAHxzP8~2mH4CLe5%07;|nC?3gnZ z0v1n#5pTWWm$M&aDhI-?U#nrdP5{)XO$WJ^zAzv@3luK|gI3Z)i24=?6K>3c%FT;F zujN0um@^H&C`^Z7i^VW&+bY;%wiq5ioC{)gfe^iCDZGgdfh)UL!0hg&uxQ$UkTH1) zB*lb7Q0pQXj9Li?8hv2>jsGC&;SBITHV*;~Lt%w`03C$t*xlXj$HMMzMNw=J6eXYY{Qj85V%E%@xHI?c z{f@n9mokza9R@HvVI)I72GDoJXguA<;N{~_LDz9Of0bk8PNi=3IJTZ1jcdC=jy3T| zuWf(U%Q}7giu@Q5(2qMZJ~L`Pie)DT^8Q^{+87VQX!k(6J&eSybRY?VBiKG*1Z(m~ zFga)-@0Rt!On(@i=MN_3!vMZ+8-#hd%z4}SF*m3`Lxv9^Vz)23ivuW~H-Pv~BWbHP zj7$9jFzy(R`T8Lwobbct%LvYyhcK|w0vbICAoTh$ZpQZKj?Ms@8u#MJ^Fd6J>%ZjC zg{ZYY__r8<{j5fE&^a!cvX_Ch{}e=&?{LC??)g`!ol~cKtttx=D@D?L4Z`RD5fxjiMaUuz zo_*FK^MoFYhG>vWm$zL5y6JPeQFof2qQ(zZJs&r5b71{)yHFzs0Lpe?)878nNu| zH}R^bMsx_O6?dw&@!G2<^FB@9+o^GPwHh||m126R3VqjWu)I)(QN8QMfv#!<^igBY z_fj!5zFd4PDih~+tDs(3DdtbsqNYflEtPV-nsQO#sEU56J_Cmukr84@r(2p7?9k-# z#0Gq6Y=DNmUL$^L;1Fv>%N4pbdDawf zNEFsT6AS9Ui$n3hMf`X9n*QI!_mY0lsw@%NV?K-BGfKrV zrC2mhDHosqmWbJVKZ?ZsA7WE%neb{-BC1<|7E1@d7k)io3eCJX!uID^Vb%JFSbp)F zh%xymju-zBs!de+^s`({+)*zaCuy^!Z>2bH{aa|M{1%?Ue}!g=4mSE)%xbK|ziKr` zjnd%azJ|PgVaVVTO}0(fVnV$(N$xtVdZf+s6*>etXyMbMPFxyUCHn4Dp}k$b7(1y> z*mo%t)7{>SEXNX2<@-bQ>HI^S`dB9JJoqYlwR|tu1{Dj71D{0a9e+gV%sR38X{|U@ zq(X+TDyy&j6dg5HS=O*zR2uvi%imRr)1S-41Fhepy+N7K@~jlDX@A79d;i4UA>|_8 ztxBjL{VJU6zY62g8a!Y1Te$0237@KJ(c^}yaV+!iI9AjDv5Img zvJe++W|2QB~duux4obzVd9pQWnU%u`c#pVm}9FVs~0hp8#vdTNScP(4oX z{?XF(AGT>fnRe_uwI_->KA;HQRqwfP@Ii9PUnu`w&cy#zl<2#)%-m4Rk^eN5lXA?1 zdYVevxO#ej{mUr5zuc}Uqy5I8EK&Q*)`b-unf?p6gP$1C;xm5_|72H9IhKtou&k-2 z_^YZCxm)5*Lp7E3Ds3fXjH=RRgPL+kUqf+jtft&HQdj05)=<(mt0@DlWS`jUTFQn# z8p=d1ZRJ9Uw&L(gUs3JZKO zYH~R}R(@mW+;3R4{wnJmRdYLD#^UxG%7-#FC2O3P>?JNSpT`=??!UUqpG}&|kxgn! zhqKkF-BD8pACh}_yoRs6YEe%u$0MYUK6BNTqpxaM__c=puT>P)Qgvm@1}&v)w5~Ej z=qY7ib(HaSx=O+#1Lbe2zU)`0udHsaD;Xt2<+Y)n@-i<~ph> zBUfrE<0k4T-F;P*@CH&RI$2FQFsz!Ou2rm?_J=90k zbrcM%d3uSVbD15ZXGIU@VemUbMXEmPDEkYPv zG?bi>0LJ_sNoZgwLH^MM$v(b+++r!}KS6%C!kD!;l#5eBII|_1v0>qSu8%@HFq|L8 zVW>0oA_44}AmFcWqKqBUbMx5OwKS&YP8o|nyY#*kqZ zMDKs2$?zXYhR0BXjYo5B?Fg3aAIs|>V|Y+Ej=X~-Nm(w{RPq)p>b(If`+na0cNNo;x$Njt*mK5jI9jANle0Xiklm|An8g0eJu~zJ9ZqM8PmKX$C;quyye}Q(4 zoM=Ji!KPT=b7W^j3+%Pc`L*4SR_WHXyl2k%Z5CYIYQf6{2h^RMu+4X-qp3Ax+nP~6 z(VPSOP01Q+&E(GZY%#UN_KO8NAr`oLSaR;Z4RL{v^ig#t<(V^+wXIom)&h+gjuPXt zCoRN|ZwWS>=x0vh153JA$sYHi)|^kVAWM!J=3vT+AExN+vu4N^bM|buBDRwaW}}-^ zpYOz(?(S6Ax)Ze3m4hCx+-%g6fNW>_dAhN4l@-l2Ezxgoi`yhC271}jY>gc&FW3{X z$QF&Ma$RlhSkl-U<0E!Fd*#5Rz77&kYR0ll_B`6_Om(CS8q@9MZ_JucLu_PBVncP5 z6Kj9Ev(Vm&E9UMzT;oK`qb)E#-yCz{!0NV6jP38r{Em)f$2cH`yi8eSPf&pypZ|Ii zG_O5cd0vd(){c|v@cv0lhPG`^>vA_1M7t?!9?g_DJzW&{7Ea3dTQ-W%8Ed8ATRUZE z2PdWdZ3m^tcq^qk&q8Sy=A`J|cTlu~ER?_|cFLhZTjfHGy>erljWR&i9e-GBr_4O; zpltHBQ>NUpR6@>JDpyObl;-J8m8?Tn${gv@yg1TAnKaT$>Au!US(WLk1qSz%gQmkW|C_i?Z zDwk``l$E0_l;Bp5O5gAHir;NVWy4q}MZ>X~VmQND$-m{KI7hoGuOgil^$-WCC9+fg zQ+Jki*KHJaGY@4{LMz4lU~A=O50r-EP-4b-OMalWvSPNka7xAJW~X$FbWz+2?Umi}PKxhWM`dWZqtbhsi(;~+xiU_z zyA0oObb*eoSm#N*9X$h^BX~)_st9-qb^@F?=?~fiz z!o7w}n`%UhciI%E8X!h%qjyb}r7~Xqp-~|ogqDj{zkZ9kUn<0t8#UsNMx7YeU4?gl zDn;hT?_%M+uj1an3NgZ^R9x%-NsO=lE@nooy zqUpC9aX9Ih7?G$(=MCz7{H{izS*3XMxmtM5k}>AFUm`-KM6^F#EB?wD@oul5qS~rL zoH|t|Cf)ff4$soyQ!5SLw$YIHdMz52YOtZRF6X24F?sV(L|iBreVkQk(XvL2y75r^1buHNs%3-2Vfrr0%W}Ga@wQ*`dL1Zyk<%s}OCXiHVv%Wm#%e z+v-xeO_M?6G*FgmV4SJNyjXSGoKa`SU^PTTHDVsiYj>+oSY|ZjOR_GLLN!QRtxjCF zrc(P;RoUA|a(3Fabkb8%N^?~d_53Y|>;yyQu;j4% z%+^!3Y|vI(chge(j@MC?hZ;&B15IVEsg}}fmZoyKlbl1oM>jL6ze-EFY`^63*ikEjT6SohCn^#&Yng}_dy2Az*+iAqbs_y z#*~4gKSM`JKdGxMYOSsG6#7cagNDkr&xVS&nvv2^qk&?*+(@y%YOd&H*eQm-4vI#p zowCi`QPDTFS4P=eDI142RTlIxQ3^D2X>e;X#lAWCcgrG<9G=%NB5p`7DZ|z>XxB0l zRM&A|Z7E6V%gMf$L+6Zi^0iV4DNZ3zTXHY8sT^OhfT#Ty(APDUr`s2jGcScmn*~f( zOJmFYh4fjIN&CZ@^d7f>yZVVdc26L{K7pD$GckM?OOMhd96BbG*f){CEM$@V-*b67 zkM^ao+9DI@ar3!kkb%zUByPtgb7Eg26DFi`pge(Y0r8yKnM_4l8i%{3FepdzMiC2n zo}a{yC25>ln}NAn3R-<6S0j0;EyL5;(=>tLlS#~JlTD!Hxt>m4M5O03^6uu-vwt2v zUMy#FTpsh!EP-{|+%Cw$y<-lun=huRX9hmSi^<8%Wz5q&2G7f-%zP!+Tvu}I=5qQ? zT83NAVwSyL#*1A`IK4#bapn5=zFxxG6&YOGm_vumJdEC?bM1fmjBJ!c!+$yQew&5y zmn^R8SdYsr2UF9!LyptRVQC7Dj#w6v$;F+1KIbt1~mfv@Ry z)W|rjy}uh1q()+6oD-Y7wPN4`FLI1L=zgIE@i)Ek(QU=91E8+f0ecrunFqGw?@1?S zJ!nRw+~&B{IrFM>GgK-Zu=I3bg_k=vcD6L>X~RJWCr&?iC3Ax_`E#9lmt%?EYAYhG z8Z)nhDc%WoOe}L^&>A-i>zl^VbwL#g@nyC^K z&NA@lp2jnH;zaQxJkV`6&kI1)ubJh&$q*?wjGT^ThrIrlWp(Z z@%Hgxf3i0>dw5Z6=t=!9FKSdhS$yA(7{#7&RU0}@b;W#_3$0UK2oH9|znvo|-%8y2 zmjegAJV-aQMemCvDc3xB*v8y)vS1F4;vH?ln&&bKAyvBL1? ztvTw{UY;|8#!D4;eS)(cy{PNdi3fDRd|_)DT(&1$rz4@~1ncrzl9=d&-e!3Fvn_@r zIx{^V3?x@j_OAnjrG)Lq0EwfA_MkqY3-OaWk*U*#1ra_JCbVZ&!%j>pY$x@99k6Zf z&4{KQS@60Qu0aZoodsbA9mselx$(4i{OZ#h>#407{#!7^wKKhYw&r(K8_XxPWykZ@ z=%>1|RbI!Pe_G<~-JDI^-C6#*6{9jbGE2joL1SHM*0v3^Keog*t}7Aa{>Q^XJ-FfN zi|*tOr1b8BVU^%%Kxcm2d9x&1&O5ga-;6r2tw|@Qtns0`S5Leo)^6~yH@zt)EA`pi)DSCcT{@{~v95tGW?waNo~+7t9|QaYb+KAu$Pj&fp54%(X%7Q} zLUlPda@rpsGhdde|Y`s(5>W5*^dR2a5EL!JvNq%>5e|4}t6MrzV+lpfvg zYIAUdCQQ|)P2YO)bdnY?j_Anyr54TZ>a+ER5pEB)_^oZkm-_~MwKU-RUp?lj>Ch<3 zfN^_uSldmPMfJK&2+-kqxenv>w4~1CpRl-FDbAVx6&D)(6R#Rpi<_#NXu7Fz`gf%W zNiP?xx7LXJ?lod@r8>vksB=}rkP%&s2)m)pRJr#3S2X!sq>BlLe0ieAwb^<^yss55 zr?fFzt%dQWU*g>sE&f_*VYg3>8b4LZrK(9zQ4im#x`dt5VL`q&2a5D4P}9XNOPg6< zn#?KDM_20IXLZ!$#S1;QylB868RO}7Qf0rD3NsIDa;CZ5o5_0g>T1MtPj&pmb@B1h zrsFCt4Ck4kc3ZBPXvDyW`uMJpYqm1xzXC&A z_SKD?du=6y(X z3*xXcicUVE1a2S9jktj%pYuoiS|ky}<8Wy=9@7u=cyKqITauf3TiTDf!hsA<8zKEj z!x;Q|6or04%<%MQ)WS%%goN@uAsqi?iQR;aLwDjd&PLAQo5cjy`}?tOUod`lAvF9j zjFTn(3A*Wrae5%OCFA+tJ&aM0Bo_^$6qY8QVT4Z``lJ%*?^0YrZuNQBd9 zQu~eO-SI(eN*hCS5sb!;04jcr#;{`$tyV;H^=KGzC8H=^Gz!~|k$9DbV^JAHvpqqK zoi>{1$YJodz+}AONG5l3&|4hPd_PnNu~K1*)NplNjO_`5D2R zB-VB+oVo_%*-$@~kmb=7#l>OuY69^V(exb@gx1Lsw74C}gs)@KIuJ>rPawVJ`#Nrj zlI#R{Kk)Yiq*{KFJdt>bB-9!H$(B>sJw$(>eH zDS8nvxw0wre;3A#KcUQDA4S!d30zN>`Zo7y2G~vH?YXgRc@slr<#=9poJ?$b1ec7) z@iS*O6_=-STWu^CF9(ygKb(<2Cv!<7nrkt0c^Dc^_^c@2=0vl|CYI{ilbCvYH2>L$ zp#8rwoHEO#hFm(I&LA{2llh%Ih^y&U~^eE+iSA;;FL+y z`!q^|vrv^D^zd1k1f57@^6ng_U0O`%%3MZX%_XUL5zS<5+v8;#Hb$8YznhHa@r7jP zC*u>ifOUQ5k>r!iYN^AD>zl+6-6XnnTS#7c1`a2Zcs(kezo$|N(8!?1Ba;Wa<==)G zxE9IBpd7~M$(-Vi%r`XCFzT0s=CNg@ugzmkNjhh~WHWeNCNBbVS>9m@{-g6Tv0FvE zt-0(<%H(LLbaa+1%)| zx!Wd*2kp~&+b123j65=Gv-#ieOako}vC%l2V>gzuq1z(zvNHKEAcMC(7qCHU!?Zr8 zqWx9&g1L~)UDG9GolOI+RF=9VOFmla*ku1>qkYL(cq9;_oaq z>14@xJA;KKOBiOINA9a7c>1km<*i(XTX_)w$b-vkJoxJ3%*GuKxXiR+?*V(J*jP~Z z%#3+vP6WNN;ZCtNsofjX+N~i+$Cxv6lR1O)teCsljNuP0(OzK1@nuf(_u;@*iS3qZ zNzC=26=hFtY4oEp&Ac08(Yqu2=LJ{lSof5k~Z` zGU7s-iBmHH)O#(BkC3#aX#G$Gl@a_Tr;EXN>fU2H|4}f z6AE&zd70=+XirPLVr=;I(Sju*_6#+Z>sV_`RiUM<@nMC|CNpXV*)a3F359u0xu)Wd5>xjGffH+H!2J9dR<3xma$-unF$u z4RYf^e@FUlHs{4x3l1JN;mI>oPL64cRwoPA$6GNe(E{h$X54FG$ESvNRK?meyPGR} z8@ti3j~80QJ;)p^weolEp{qS1QBvRH(Tp9%F2rqhXN_-1&UN>|Ki8Q%mt0WIX+wEp zM@FZ%=lc2fT#fO<&(VV$52Q}#X&csF?#6b%_P9N2g-W0s`TINxsAy07=yqKH)SXU^ z+wjY@GgJ2{ERc+Yztjg6)eFkq+u-ltlAta!hBfkHK$w^0^qtYW< zPj)w>rC}F*WlqqVj$rBG--ruoc(ok2l*EFZcs2rGr9BQ42cG^u}YECns99=9H5cE-hMODsvjAK5bCh(}J|w9yDI& zAp5wxQvabH#WL@@^vs)@U$Te!_tvPM_2g2d8zZN*=HzpGN_R`$v7aZ|_N_T|(H*mb z=4AGk*UU|!V}cLy^SjXWM_WFnZo^rBGdjUr`4qE>*FIY**|e2k;}39RI?8&D z!!#;6z{-`!c%E~R0~-$G;IxmJ-+OtraW8@G4^jSbKjEkMVsUj3f7b2e`Pu!{J>0|0 z=)D{{yo)bp2iPVYr4Bo=j=3998awCswHsZ5qD=k-VBBS393T(HstY`~MA8uw%&3383*hJT^n+SWk z7T3dTxtOvEy9ev((r5#=8S4o+u!7weHp+U4o5+b;M~_pRP=B(8163RNUb#bhKDM!U z%XZvv?POD{y=+b1#`B{)Sm3&u_ZzmbZSNXP4@rFI>q>Gr6|nE#T6$}*=3V4z#Clr_rOW zsLtC&TJ07Fec43vtWCHL*+{#eTN#(Ji&u`jxazi#d+T;nXt$Tmjdsf4%RXA%*hBC3 zJ9!qkhsALRQ0;zz3xzweZnq1^@=bg+-b26ZdoX`^m=7F7mNH~V>m&GFI>OAsN73~= z%GcEs7}6~U-$xUuJ{?2x{Ask=Gn*+3=J9520%ikJI4sZI-AyvNnwm~Tc^aDxQ~5M3 zjdzRF7``)+5l>S|n4U=ZG^wvXp3FPPWb_x!qNha_?A|YQE@qJbrzdvR&C}1(!Ef;erGKfR+cDIFeoIa1HzvfWWHG#&JGnqbiCjTvoXX2{a3|cyybx)IcTsM#XIkD)Rj%D?m zSrkU4k~AQlWbHJx2QI*HaU!mM$z%nmb7Gv-z%NSSu4Xo_8flohWRvwiozYtsqE(qn zYr{0U*d=gw!%VtYPGp(&G>%qG!G1(E)gIwY{@+Y=^Jb9kl)$I8NsQ6UWKVqt&6Bbz zU!Tkx^9-I#jPjF34(7cwDf+UE_P^8UshLdUPs#j@Od)(oI$KtzNE{}en?~uh>ypm< zW4mK??!C9%gO9hI5M*cl{K;*iE5yF{`YXRz8Y zlb_A!ao#%tO^ZZy9cH7}bs`B<=i<3;4&M@Ge6wyUPb^c|{~?au4dR%3X9iD~Clcx` z_tj=H`G03{`|cFHA}7;r@f3V!O`z)g40iQTMD4#Q%Kax%*?1hk4iJ(M%6!K&kz~0?IoIcr@MI4Cs%LR#W&(@dl9?5o$l;}O zzgwnY5|qo&;7ktdq_8w7oqg+42`G@-vmn;S2D6xI}VE87MMF9U}4YJ@=FOcOaclE;6Rxn@L_o zCh4~_DeSlqHyMjpbW3NAXDWMrG8i~!5# z!hGu=(ddBW@b(vppyauirsD>#FqGC(P~t=I9gpJ&h7mow!bbF%a6z! zlXbPCY@s$A<3Ed2$z{T$!6&iT=$lv_`9Y+ly%HwVo{LTPFT~cud*ajfCt~caN5bOr zQ&H~sTBuij5}t>?i(#R)VsVWs4yr$RIPk=S+puE+>|BbIG`Bjyc!DfFE_2#x#S#LX6^ z;z7RT6oY;UCI734?(|OdGkGpng}e}ZzTXh4&7O!x!=8(iNpHk~kdMM{`X{mOdWopC zdM-*QKNItw7mMbr--x8tpCatlXVJ57iFhh?YKxhkqF=XC;VbVcqYo@2*(#U$IXR?k&!g|;d|Fp6r%lWSS-dKjlP~kwTbP4Svuu)%XR={%7P;=Z1oXto0@ESf6HRyj|}=1C9~(lLK4Yj>y;dyD9dpk zy@K#1D`;hs&$`0pgr^pekhF>~`fK=Jw33?r1^iECt;D*QQoLm)-_-L-TegxXLkq}U zyP99O3%H`bnrP!y6rL?0B6T?nZ!cwK-YS_h<@4BJ4dcV}8CQ_cN4En0spS)6xr_qm z#oUwoHYqEM^Me=393~Hw=uE_nG!kbm;@62BhE7>Z+M7J4yjaT4BFPDzSc&IR$>)^j zFvo8h*JAV7sk;Qd*hSKFn!$k&i@AItAE&f!Qs=&!h_DrC9bS!1FH2r&I1%>PlGod!hwYRog~O^ttKM@Q9W^gL+85L+v%G%We}#~Gh4Qg?IJ7@bCz zT)t$%PeU8ZmRQs6yCb%>4WuX6n1^{L?4Q~I-`J*zeTK}C`Q82aMr7AEWQ%(PnFs4q z)Iyh*76#0}XUNem@;x05SUJ`}Vk1VxD~-^*r_U8DLnL3!TwAHndES_7+l)A{-H7M0 z)~xzs$7w4&Qf^pcb=Hh(Gc#f)HD;mAO)qaVVaAUpWa!xv`P7_fUrXK}H^pp57 z?z-BT?I{h}^TwQpiaC+%?P$BglHqe??qqL9zx@r#)i+}HfX1kC?xC zCo`S+qbGGrF>bi1d69C+1+m!^=V}+mFLY$BWpn)ZxUoqz!}FIFv%@V&-)%yuV#)lm z*1XwgOJg4!JcDdGlx#!NMhhOx+|`1|0I~K&dk}=MM-6_qmkbZ`+SGEjTWrJ^DOa5K9qH&@L zr|L~vFy4X#{$65=x99`3vrq>qsGDWaRBqway#x(?KLbmE1!0F) zjZa!|-Q0o=-5zVn_>`wPZE-dKbNU^OgMHlSZ-_waLKU`U!=1g*kEvH=VSTw^4@BYoG z_jBd1og>2y99g`}iift=^8B-r8e3c1SGrK(V9BgXIo?nUTK;Fn5idK5CztQG@5Ng^W`!>&UIqsD7lx(E;yfeqI`f0!?${HyTpN}7u=a(?7-+9&itF< z!m9lq1fO+h?=5GFZ@cj8S2H|jx)Xg*@_p6PW3sdrW4k)j<)sbdQ=BN7B)P`_9B_N! zM)i29?fPFc&K0_{Bh3ki`wrwDm!9@GS4?B&ygVHly2MH9%bldBz=E5<&FFl{jN7A| z@bH&4yQ1uv<7CI{DktVw+w*&f9h3Jp<8E_X+V*qh!5t^&_4lN~f1cd3mA~a!cRHqc zaLA$!r#E}h+pHzaR=1!g%Y`i-ZO}R4!LNnh%(~u+{~fX;+s~4*D(0M3tm*UAiW$4* zwbHkf-WLlDT3Pb^xHXT4IRnjSk6q#5d(#Lle$Tw;;Bg9hFV3`F_Hj zTUPQjKH73XVh|@5S`$9ilyzgxxZBj6HE(QDzBpj~#1Zem&a{1OCGV5gYzuK?==A2~ zRyF5ALwmg9tZ=Wke}O+))bXGQx0}IqPCEE+l>3Q0k^sn~(keFQEKudB? zm|>V@O3g+q4&0D)yJJO8kTqA5?a_*{=6;$zPG4OK3U|a&;x1mJ97ynLN@<(}y%Mdd z_8!aIjIrD`j>J!L07-{JFjB%PsEEefF#(O#7#=2u5j8xP_EJ0g{laVtvtuQnHHYNd zSn08j!SBf!D(;M7=#NOcng`MNe}VjpiNH5&6sILuWDz}z?Jb6KZ_#MR-;}=U#87q& zh#=-=3>qGj$+MhHy4pl;OV3V^X=AbTpF;bu)0mVKA@v18+-Vm|>tM<0d<){2=Xic+ zjK}?C1c9lODR7eGXNEF-sGPT4!@J?*P$>`NuEj)FY!1im-8imFe^T<55LQVHy^Yk{ zR+f&WWz8tQO$ef6ZU{kd18Dte41rVqkw* z&?$;Vl80%~FoqDP2(FEuLifTbW-XFjiA^LA=7(aoB$PR+A?)^zBcjh3jO2B#Oq@t; z>;z`WeffPXin10FoH!f8z2MPQcb!ViDhSm7-r*&ch=WTq-Wd!u}^;G|IVHT87JvZ`}A)&NkiOL~O5O__YdNW&LFG zpb}2EE1|E(d**B}VSLRyf`-0kqIf|`?;@fbl`wUEF=rYV^We=lj$A6`!=9gX*!&Cg z$A4ISq>`qCYf!&m!xtl2N1^f$89ggV+*gW?&QB_r{b1&YpDYdeg-Yw6C+HGRw){RY_^}7gpc=CHa@%L@g+1<-T8>KmVPAC&ie~{mHiXe>p5FS!GyF{nhw?K9Z>hp;Oa;*%6{P>EVo%FIn6*-ob!*g>gN_=?#sQj2R;;@6Hcv~jFx8j+ z|8$fKlMNN;xeb*+DTd0me@=qN{g^p%}fdP-Y0 zJw+o*QO|L*|HP@fYB7KCKQXDO zR=iqQAtp($)0uab;%c8-5!Sd;bTcm%YVB)!=#N8~qU-E~!cUP7Oqgu$F~nnHL;SRx@J-W9V(KPbb~C{1t`6f1^vK&~$b4HvjMvEc zbDRM$3JhqmN|)sYCb%~?!f37msqc))d~U?UYHg;B)#porA;YR0V6{@lp5u%da#)9b z-fGn5OWuBoCL=SoDQ%_6&dqB4yQ7Xthk9|Zs~To2{)*0r%Y=XOAJM|8LKw+?*>_v| zd*W4*4Hq$%oO{19I;?-MPNwv+?T{X>-DlNMf2_*PyT8R$k806Aw?_19^G^)XR+qH_ zYDKqoH6nNKAF;_yg%R8S2~!7cdc684cFmM}&N4O5K2>AtQ|V1=@Pv**v z`Go9PKrIQlOp@4|%0i-hOC4QeDjO}6QIAWap=%QRpDe)t>q7buSwO=D$<&@rq}i!t z#@ z4T)W7pGoIzpKR%q%98j?79(QP8Psnfg&HZc_D(W&aZ*QEy^woDQ|KTye7$25*u6QC zRLcxvC7yTj!92$2FQlDsJSEZdnYU;*gY^<5zc?Fp)A<|^pGC8iv*>(t7Bh#;C$VM@ zE4sy#qdE`IwhL&}V?MvU7joS!iSF{TUE*K8YZl-XH51>;SvdEa#-g;jJoq}7YNNUG zI~m8HiYPk8#xZE+EZXf&;LG2I{3qw#S#qckx~5@1Jr%2*bgcTM(qu+Dtz~VBm4?YY z`I$h=Zn9>E#MbWbOyiTxTVJe9W4heOOIg|E^j}18y+!=@N9KKwl6zg6LvynwxJt}w zq*V?}_vh06U#^V57IP{+AGc*I82WfM(`zLoI_zp$9Q zvSviv{<&oAS$v5~=c$1}<9<#B{V+LP+jQNJgIG5i?XT)>v-FwN>q)$ve z^p-J?UNdr55er7YSxj4c^2>FLUuek&9ea~_QDJX+XD1yVPr}VSDfJ}Cbo&^^X*7{XGgHeNcxVNgyD23k_ArD+{+7K+WArZ+%}Xs{l`*c z7mB}j2zzycDHz~SgGD3hbZr=`Hmi<`#%A2HJZEWP3S zBPbjc#{E;{nQtZa5-KrREFaI4jnT~b7EQ=ysjH|S%{2ERY*{}X<@*St9*-ip!x)}l z_vc1v5a-^85>OPxkg8x34MuV0a1iP_BbhUN4A;ZLux=Q{u<1kCWe`Yh8$VQT`17aD zU`ki`u{PO{jW!W%tr$aBi!p?D590fzD6R#L!f(h(ZgvPE!cY3DgNE_1MIfWR<(S|7 z80plP;U{H%qg_K-ApQ1-FNN`@-xPA!#}a=yhI5f2JX;h=gX*yo4~S&y@-XxU1rwMM zOozo|2=@)(`oW>J>O5H11RF}%sliyb8A;l)L9(t)I0IvX7;#$0g06v>=E}V`43jau zj6r3LT9XjK+?|2wnglSZ^H>^5-`ab_v3%?t%IG6e6knajj=S;PD4v0d{Of2iNA?I0 zCvjZ}?;Ro${vrGd4&rk25e(k!&m$7l|nRAk>0SsVs}KH z_%rdBIQHU)7?GwTW4&rI??ScMrCB99Xw{0(nrif_`X_?bD@CKmrDBnHnQS{(Cfs^_ z5g#K;MdGQ?;^DI&qEYxKajW8|sH*!duAlfOEVaLjp=(M-*T|ouprslu{j0=|Q`N#^ zM7=myRxg@-s1$bJRJqLyE1XF?6>&S z_pi9D`%Bb}s1h9ue~O*0RVbOEF2Ad4Y}M3ezmE!v^qRZI){7I?RpMB5l~AoN7m+hn zNV`-cf;OpgcZWXbrfG6IvP#4y{1I!;*9qf(s*D)?Oh>)v)gqBf- z80lOiPF|PuyCQXkspVq+?{cx{bd`AT^i%kZuMl+~s>R_tRm|mcKXO%Zmb$}Uh1J4V zY7fUd{uK!;tHqAn--Ji6KjMBtt+=^Rm4zQQXt=RX>`YXp?wKa_J#}bzUYmD&wa^`+ z&2|}^r5kAC-$0%De^i*BtHR(lzs1$ne?`+^sbjpP%DXFS1V?Goeug%S+p5qmMuS;V zs(ikn#o~WDSWeUBbY??lbkgMDA$1~_)r&XD_IP+Z6ItlM)FMYLD;+VN<4)TT_MEA- zA+(*ubMD&HS>k_A9y0G7ZpZL2D+DNVDL{HqQ1XDsmdv*Wt24Kqetk@nbzto4p$ z`AO|Vp!B!y<){;ZHeVvF=OxS#%x|K<|@tvgjwRg*^!>U_S7D+rTbAktY$e?dZ1G0l$4MFna4rZJsB=OFU?Q&x!n%PAJ!%skrNm(bHyh=Nq~$XvXH5 zRt#HehmM;CiCav`mtMd_yB(PF+=2X`www!g#x+5D1?RNp?ziSdCp72OUOy+D6SK{^(RrG)iE8>`aPY z&7#wo>2zN;nF%MRN}bp&`b5pZ!NW2q)0`YKsG@#kEMJ$eFkoGHs_95R7HO+nyoackS$bAt1P5TzK7ey6rQ1G)v z)6Vty(7YZxZ7vAu-hu8P&UiNBfpqdo^YQM$y=gvhw>^Y++Ye$77mAC?As9_LgehDw zc6&ylmOqR*Ndb_YdH{J92eIs56xRL;K%uceG%S3PtQG{*njlEM^+8m%7oI-zgvrVv zNJjZ0WFQ!KVBVhAA99LfjV`^nM{?V+wKPwWVG~->Q8U+#3$tiD)he~!d zKJJY}{L&b>(ER<50oBz~Nf@#_LV6q0_pMDLCPV^03nsudClT?a`FvsOi;VLj`0(FB z92fAzThdAI_yNpYydOQgiR(go!`Vk%5%F{*?noWP*trGCnxShGil|p3Xl@JPrF$?m?2465 z)aM{BpNEMhY_+%GjEFtn?(%@x<(=>dwZd0hOMH>rjN2m4SWeqnTknLXxE(MX*#?Wx zdtsC7fT>Y#u>9eMJI!uLg9n~G*$>T0`|x*YJF(O}5uNIe3-@gBF=_)umTbf#yB%16 z-4(hLt{At+1yZ{muqb;QbQ7KMe%%)IA9cW98(UPQ+JMKw0pjr+z*D~gyJ8$6a&aSZ zWL%)Pfazr%+Jb~dc37CU4#E4^;6j86uKh8E-d_u7zPE>Tu`Q;%+v29PIZ|I4z*$cZ z6Jv~^ZfZc@rB!&4W(1ur9EvrJQ1i_k4<4H1y0bB6)RPCLb17CYF{ibhDgNb{5w~O& zj)*P8e)hR0<$_KT7x(<4*A^Y}vU2Kg{*f zEpG&((_$#o5EFDc9MCYpdNVx)zSqaZr7MuPzzBnXR^h*iW~jDZ4GDb%9MN9_fjoW0 zt|LE-=~7s|UWS$hE7A9UHPURBVC5a4KimL+zUhHt7ojU-3f_$6D3&!rcdRk)``N%I z#sssHS0E&A75Pj|@TzAe>R%BH%h?op&MR=+%Lta!Ot9{{F^*$jF0%?Q0=kCS2!C{EIa$&V#a zPyuYZ3~=%&!>ReY81B?Zw5ukzNvq=v&1Bw)h~ra|9F9Jahv7mE3~y4!m-7p8EK(C9 z9|3jpbe+~+2#WzZ#AZlgN0<_{+!XNCLJpU|so+tU79LcpA-aTSM!OZ@QY(w{ol@YV zd6fEdB^a0Mf&wYIr+hql;-_;OVaj`LONSki|1GF{9s)`nEe zA}sXQLvKCl688~LSx*~pi6yq>tr~m}s6a+b4b$q?acIQ?2)vfWs(Y%CJF5$Wsf#gn z4WO-N4x4HN@cdf>u3Zlr<45o=Ziw`&0~kN|DfFj5!RJoOQga)FS;Q#Rxrf-4GKhfo z4y^vvhb+naSh%te>*u#4qp1z2R(3&My9?V(+A#ThFPeJ?5T!ndlBeBx;xK>%Dgy|7 zbPo$^@4;+yFMf5m!AYVCk#Ae!c)S%O*E=8~Q4ja5W*k3Ri-;{vIKRFNYSDdYzc7fH z8-qCe-xKg@j6rPOb7=n>Mf|l9+#49c6tQjuzHNhaV;4StYeQ*72VSdmLp-Mwvk$l6 zUuHXoBkv)Bn8Joa-PFfyN8;B$@HX~9=IH<~hz;PP*f5kX4r4suOFYtj4Drra~WLcxt@F`>v-bpZ6M)>=W{+5B)vnBaAEG;(Wv>gj)Wh zIniHO&Hag2SH2^n_#4*H=Q*&yIN{S6J))E3Bn9 zoArFW##(~%*fog)R`{ug@qE6;;+B`PGp2W$ms>d-5iMe;4wN#ht@)%;y1~XylrtC8 zTqe4tl&ySF$zD9^F$fjUT~Q`iYQ|5-sLg_jce@0kqd0w z-}6i>H>-QtXnjb^-RxXDc^F~2yU`K z`73Pk;~UIQpqQo0kl2{#dz}|oWB(UVNYV}Dj!1j(*wxv^TvA1Am|PI z!hJ&+%J_T`vn>P*sis9|FkX&E;?}-UoC*lU-a3EGTOI)IU0&!A+=sc7cS9`U0G?7! z?klht^AvX=;qz|XcL;<2)(~i}4n$*~FSMwZu`LZls|c+-_5|Y3nGh(Z`Qv0%7!2gR zF`Rt>mWe*dJh&Gdj~*nh%RXe<`XbUJ2s5`GgvzN1%sLnh_1}j`H+cx&C%qtUypQ@( z!Pqt8MOyv<>^$O)l9X_mXhfpQEEE@Zhv3`gNO-;u#iJj=`0fyeB1@{(+amDiQ8?AV zF*r$D(4Lc|t4a*VdE3KCnQ{dCLylvU#xV%rPDF1?GFUnJvfmwn`P3tLwKWEtxCo?d z35Vj@7)+ZHi^nDLSaIbrHn|>#{I^8RZb*XT(c?J#o%}WdNwDEPO*4x#c)gC8z~t|E zu`mttp()VqK8sItPoU@932a|`7+!BrpjI>)Ee6z2CNFB!CXwIatb;6!@?;@&R~{uMN)XMQhe5*y$iF6!XwtTn9Gr)deRBA7PaZA*6fmb% z5tC%)vGB@5$jHiJ$9JkDMl`AC0URSALB1iaRW()dV1*inPpM#f@q7%ql5hN$Dsm-B z2SFN(RPq)SKbemoJ0zjlG#|ZTGO(Xap6|o5FrfAIA)2AvO_PNFVFf51kVE-7DWs7{ z^U*{{KvR@qD)ly&;)CVDbK*l5L zJv0#KUtfv1QyO^xRvjXApI-225hmwxh!E6-rLa19sJGDfiMDmGG@OSQ;+Bg7PED7` zi0wS2evzehngU|R6d`V)fUSLs;51ah!?f@`Ulac(EXJV@Bb=eSbVK<{jGtwMaH>g0 zlT9(}jUGN8q>KcryQXGI!Q-?fUJ1xT>%@Fm=_z9WKQ$D2X=1OCDx5C^Pmci8qKy!J zpTp`rJ!}fp#_FXyu(_*;RZ5Kd`wW|AsG(m{6MY_3=OwEnvRxT@^2(S>HO9x?{#2XOMg*kaN zG}ISCD^{C)Smd+UQ$=69D$+GoU{tP%hac6Ee}VKym*ykhMj3@`m9g$49b*j&c>70& zG&gh5^OIOjQ|95HtsM0BkmhT)G$usN!@F1WF`TRj&kstttw|oLVRbk?P(`GX4#M9s zG*i8(u}>RbwVOVJ_JMgeYnQ!N4@(1yxZUl zyK@1EvI~J}aWJuZgCOqVgHvw6#HKDe-ov?UV`;$GT8=v*Qn?IvG*TXzuRVx+_9qx(*I z7@Ayz5jQ@7GE~Ay+Y^ca0lH^i3Bl00C`{Feg%F}4cOV-2=ZSe@LOw!=Q2dk%fd4aZ z^7I75j{FWs_In{RDH7{kBXIt17<}!+ksKEUSG{P=$qC2qXQ3DtBA!WOC>mmX@nU@d z{z&>EG&B(9G<)o%*Lo_&qAoEGqRL0{=)WV-%q9QClLVxGh{6w<1gwlp#!iuAm@+L1 zx<8X3Pur#?K-~QY)DN+BflmQxc(N(~LB$cp)%L_S-hkc+3#=_z3!yeEV#TjVMdLd7 zzOcpjdzP59&mJ

kw^ck4D}N_-tl}3ew>e^gE*e!AAUTwMK`4E%GI;@W|7g5~5Av z{dfg&`;2f~%K(pO8{m@A64a|K#i)lN_Soo=W`@@2wTp3f$OOGlDa*ji99w6sgLUXe zT{@xn%$5cjC47o;*Kj!rMI9=W*eS#@5b!B-5B_}8x_(Au=L+hmmUx;%_8C zXdlh#e+7}>H3}n%u~@z#8quesu}+?jq2x$}J`09yZzydG`E%wbAZLF9e4`_A-#i?v zI>V59BOGGJ(MUTPgR~mr^!Ubr{f))QVe;)-#NpUzEMg)Oab#j5<-Qz&6!m2P^GSqB zG5JeYlGoSs2+UrG!L2D8qm-@C^3oSMq&E{7^ucTwZ}h$*uScXeGWnzMbbTnQenrAp zKMH4*#NQNnUVmx)L2Y^ z8jgt)A&9uS4}+%z(4ZWK?w#IPbHN|RE#64+2tb-=FwT*uNttTEXC0wf7*2W!;vV_X zyxx_rg+Hj)4O!*~AEzMf_~HZYd!cY5jf4*M%~PKSV*K$PkU8Lm<+(nPRu0A0m|z&{ z#pC4Q5tJpwqT+f4_V)$Dh&)UR^d<9JFt8~JQ^?;l-60byNs>OtzNH40vA$py z)c1&^Oiv6BzB4g>Iq7=aR5A8k78bAPVqGHXHD%@z=Uf6xL!^nXm%?y?3{3LGu{oK% zxrH)_i4{W*CkjEwSx77t#^O_g$aWFJ)U=t@-)(MIh)UjE!Nl;V&>7 z`Wt3a?v51k*Cf#$EeXFYS@ILlLjDgi+>M%xACF1HdRQDAUkO6{73s%sDL`TILhzN$ z$MLI*h$emL?^bC@>5Ac`vn1APh(g6f4E(?6;vdys)$uf!TSqnd1=3E}zhN;(AKBKv zmu&p@A+}Wi2^%~5hDnD#VQS;rnB{|hrjXRdPUv^CDP!&Ir_(T-Q9aDQn~bvZV~^SA z%3+o{`zc#@^*-C_`h;Eh@`ferzhFd&%E+_Sl#veY}Je& z7H!neLRFqJ-=&>_WJY!Zrl-veeuTW@j+S)OCm0wo6ah zkje<-YkSH*EPKjIyhfN?zVxjkq5O5d>6wm;aGrdP~+(tk|g!gF@x(=+z^ z^&|E-y@$2Y9VKsm{uU!8}Al51|vLQwggF= z4B=FwkEPo=nwOg4&<0B+zp{kUej|+OnnLcN4t93v!`(-iTi#9iJK~eLT2+2-zUp{x znGz3oVaGU5X!khod+T51CHmM9=^$V#*<2aEsKfx7zhwM5&&SCH;oE*O4r{QP( zb0rO}8R=|qe8QRgKTx>xH~RGc!RqrkF1Cn|o8IsfUX8q5k=_TqH5>K^9a|=H zzNUO!?x*qG#8uCp7!<%7jzZoufHbHof z2lAHEd+;5A(dX^>Nj2@RC|YZ+*bDi@UAW$P5Klz*xU2I3?~x{B+75X`_7{ z_J9huu=>d2k=wAY$uy-qNmw+nx|w^7e=1FpahWn(U=cxaDKvrRZ+;{iim z7t(5Z;+d5fr04lVFrKstkG)Ax<_@#G^%#9_52;)BD3y1_zue7GNZLg8f-}m#nqef) z8n3f0k>9=+PKRt@*69G$V>tXZ29(~8{uw`}M|7sJ&t+$Y8 z$`x5QyAdXI5a*6=!tIZnVf}6=rb=yrmgpg3Ds4yXs4K=Kh>`x)9$~~hSKDg^iA-C( z9kxQ>IyYLsI6;+s)>-abQ8Z;czMJocoXs|T);fciiv@cKJZdBG>IRuwbUEWRz)DzBofbwZLnV? z8nQ!CG)N8y->XP?2Sy=dA=S}SqcJ%)7Q*LaASV!o@yY-9WK&P~b~J81iH5gi49y$K zQ$9Nyx5LQ~FBFaU^WzYi5s%qsr!f+did?5OI0)sS6R@kC`YbJdeocv$(ZB z6DyKZ;c_DdQ*%?GkE3YXPCNzb1#F_VR7P4jq$PtfFdT;TXJN3U&&itoj}V9^2jeRXr}vD5!p0aB+)jW-XcSVfgoEd17zURH!r(q>UXpzw z-5!KLnsI0oJBotuDUkL}A*RuBVre8}S4$EePftSo)T1apa1XzYmKSc+?j;@gL8NE}Vs=M3W%xv) z;v~%%E=AzU-!RfBP|xLpAO4aqd>8d=bRB|_xIGkS+xvGb;k#&fz$c5(38^v^EotKwM2U z%nHMyN}m<`DF~L%A;@}2{VwVS1#b#N!(afa2T0G8n1m+1Bb1|Ym^{eGAhz@BJ%Eo+$UD zbP(po#AXoteLA=QpAZ)kGLze*A;y)Q731P=&gRxm7UiPv3v&mK2y=Ozv$!*X;#`aI zY;IU^1}7II$TbX3;|^XD;@VP#xV7VixZ&LC+-G7PkK7XAindJU4o6dFstjea9-qi< zcjM~8O^8Sc&LR&?+l?ftTvHvV?-gpu23hndPqjW#~Ig^W$o5htl&*2`n&gK3+nZrc| zN^$P8vYd(dJWioqh7+}u=0ZDWbIXX|{9{yvgaV%XJ&DkUbDCeWiiemSAy>E zb2+O?w2zSD{A!5z{B$N)-!qeQ=oRI9;zha4H?z43brM`;^c?O{p#*n`xc9kdB)OYy zv$(8(Gr5yHggD&! zlS_?M<9g3&aIa2tocnrxu4Ws@iC$mC-6dW<-z7axsfflH{;HhgJSA?TpbA$Qq{K-F zDRWYBs$AMpMJ|Ir*M6}or_`juJ@8cJavfE*+?sL$|;mT zps2t()KX@4wH&9lNr9X9R*CbTEYJNAQ{b))D07SHx$}=Hak|&Exmojdxi2R*x!^Cl zoI0;D*X**K`@P4KTf?)NbNIHElkauqd^$FA>rd_AT10nqvE{B@@{3(u?eksS)@wVs zy&^lf`cOCSy^1aO{veo8(=&N!qS(==#j5;T3qRTl8E#)ew zaop3F`rQ2}GcLZ`ihHonk-P1?i(9?KlWRM&jdN)8;T(hgIqz>?T+4BPF6DY4*Z13p zvkCR%B=mhbn>-)RI?b1x+~Ljn6gT6NSuLvi@-eHf3=)<3FqFEC=AP?NK9C8om+AO3 zauIXSUxfJ^x?i;Ckar>n3r=3bGeW*eXc}XWeLJ;%5l!N2)h0GFkW#Ptv@s2zbqa5 zjLwo?mvT@2lkj(1Jd~T_Az>AbPg9bR+MR-rQWwy3BpuJv({V&39Zjb$pm1k4a!WIC zd2a^R?MXwnB+VH9oCA;5IT%)-!z}ZY_#AQyvKLRmVE7tdoXR4N6fqkNt`Xnv0v@HF z!-JEVxUwx9nZ#sL>bQXP!3%iUah`k@Sul*r#*geA9M8OjX?59{+n+&t>2oL$pvU*- zz+5R4(Z-p0QJIYstyhTWdKtHRuM=DEIt+!2hzVANOX+zC&b|(d_17_Em5p6(u|Cp+DFFS*{VLT1}X4R7bhZHMp?84)VM2AbVO9v9TJU zeZCI99<`{R+ko+_+R!z#5lQvcSnt<{v0cq*X>G<)(RN6S)KEtH9n`;Rgi>)GCJi(o zq^lP9CpMrvq!~uNcktDz5dzXpxMg3D4YXd z#*u}cIM>xheY{R=c;A7=8^OSj?Y*@6p|4fyn@30H65!^Jn9=>FYBc`hvwn%9SE=kFtK z4`q1f_F~4CF1#pchu8f9SRLraZ>15)@QzSjHUQlZBgk6Xi_ock7_9AqXGas%el$R_ zu>n8i8{xmO5zpRrLgYUAvLaicF|Liyv3iWXZh-pUMttOJCPr~1Ca-9xHC{dX^DB^J zeh155m6Fc96h5a5V3TsgD!<@K2M zum*ZFYY?n_3(r()@Z6&mX+srwURjPUauwwFE=S?^N_a?C!KJkt)-%hnB)1HEua@BU z&O$ucaRZuZ1-M&Si0{WsQAx+QWbh3LOejS1#~YY)ArDaw1*AdEC(XhQ;(lGjaXKg3 z2C1eEx&e8s0?g6Ohj&sL<(22d<;_jxEWQTi(5twadI>p2R}dzVgNy~a2ye^7219zE zhebGMRDjlsVpwF>KaXp(<25({~|LiOs2ZAUn_qG0}RY_L2Y0trmCl>DWxYjlg|HIK6`2<9RuHMoXZ% zp$H>|xAEg@HI!1yknU3oaXOAOOsioMQGtKJe0z0zHVM4FpWmSPHg>ozzD8U1V z5}eX0#t{9z!66^_e&yq+a3TI!6(RBUP25n+$KJtm`ku-VVOEU(xDq@KE z>N|j{pgu%-4#N9aKd#vIz<1RE7LDtNfLRX&TD$T3StkO82M}vMfJcoznE0n3HDx`h zz2A+Mon2^_y$4P4HhB5cxjfv0{1x4}Mc1FCvIn@Y@&q&U#~}0VHNG8vg@d7^sGKo~ zxpZx+@Et~N1mzsRXeR%18|*xqA@`~YhiBDcZ*v0-JUj7Ia{xV={TN&L04~Z;@vHn9 zamid^n#2%o34&Lx)-2L4M3G;S%*xKOBNi#H?u7>#vW4xSW z4hffaSUqY3X$gDG6|^G84D~M3Heme%d-Se$!m$iXtogYa7NgG4(Oi%H0rt4kV2^|3 zk(>YC2IpRFfQ$m=v^}xI(V)%53EYYYqMKm9XCtm=Ib%jMJvPM|t0%g{$9D_Ps_a14 zDmO?y*$VF~_6V1tjMXt)$VfR--m*QK|C!_UYzut8Vu8jA(n{0k%LeO8{iW@i@p8wl2FcF@Ga0cdtQ9v?HeEnq!`n1s*rAL-rF#)VbIqcA*Qh z>WGmMw1F5&uCR1>0V{IFK5-B52DriNg$v&Ey2Eok^^aEVgrYyKUHx3Jdiy49R&qk% z&|Y|d*#rk-VBbu2!=o8(Ol)@_vpdzpBAdEtzSGBUd3Q4xsZRE)y@72$SkGQ6Hn9DB zI+Y|LMM#g)+jeVWpoovtMe%2`0L;6yhrKk+Bz>zNYUZ$O0Z6S_Z-2iK%SxBtQJGNe7l<6#b z%wAs`Vs^s=?004dn`_s}t_yav#Pkj}nd@e`7rGd4eK-5XU(1#+ZD7;mo7tnsO>Ag+ z9n+1jU}hJpn6y9(dHPz|MYm=qce|7Ei8is#Wi2c#;12V$Dq{*6ci6XtGFIPF#OkF= z+3qEUOvv#LgKH(bp4LtCq8!6}*TWhZZw>M6LTcEatFb~0!@B0(VVwIdW+B~50U6Cq(Y2NJoULcFCHI(? zLKi#cPZ_P+_t>K4U98fjhuzlXhp*36L0li~ZLCwov6ayVk`EcRgX~ z-kyW=cc$Z)^%T5qpbV@>ym+8C2}iufQFg*_)+IF_1{VbIBcBHolfSX57ytKe{$lNG z#z7`-BK}*%gFNwxXxq($R&72czu?7${O@d3fDigX{7^Lb%XUTa;jh$07!OS(Hoz2^ z)lPzG_XK#YcUmp(D&kZ(-m@INL;e1;_-f3d&(8i5m(R>AgR;Du_|VYDhi#4HaCg&B_AYT8V&%venllbX zc9b2=&j&3#UepYY$G0Y49G)$Nl$F!q^>-4iM<1-`Ur3|JWo$9t=A3Aiakd zmE`pdyf%SmJrgnW?F7utra7M*4~kO$u!-suk)1OI4oWkLCohD>=|Z?ZVm*aDJ|b7^VlQFN<+;ybk2Hsw4D;4u%hF zVa>(2#kDN&oIAJIbsRRY+4(VXjj3M{0A{2 z*3B#Hv5_f*-h<)wbw^x!=!o1k4iH&yfj*&S*b=h@ z&TCgfYSb8}hm4>>_uUOLcC=2m#`4zxXW1KJ6WrUaS&B&X(mhQdewHKt%lCeQ$fhF=X$yX+82B~u^usz!t z0Uax`?yni7>G?N*8ex`;9v0Oa;k`EbX{N3u_Vp52?l(pLxMi>?UyKHK6Wo|#j{18R zU@hiwf4K_vM=YQwvwCQdd-U4aHP*lc4&7rxHwT!o zWiN}`)5t0<+gQ`wb{5&$%(R-@nECA%_UUs!YwaCkz9;%w)Mzg&vF~EJH-_2U{7$xi zbpuP*Y+^ro?y?;2HX9gE_iyWatoB?VTQAhX&aN6}cb?OG)4ff+rHAp(Xk?4_++&`T z+E~QzR(4`REnA>l%_0*U*dK{r*6QBFE`04KzehXMDQRT#&Q0u8aU**^vyXLkx3j{A z4yHbRkfko^X8eYoOsB4o6_~WMD;+&dY-$d+n7XQ8?(s2%g!&V zW*K#LY_d%oJEYvq2K(=^4VB%@`2p2N6IxkqT{m-|GssStQeD;C!ko(MnDC?ymg0Pm zoel10!aNP^yc!;_BI~gDvcjQ7@ZtaEKY5pzqYCpP4M_V{@&#SVqk-yS!$U zUDkm8$k9jGGAbs8+n&p*{Ui7|tIKtXOaZc7BF5C7%YmX1#($!`4}&mt>P91* z=7CQ=5+PuCl-6ePC|Mr^rGJzKB1qXFUn6jlFA=L(Bp{o3<6^UtAnY3l`_^d6@ruS0 z!$_!a4aaxKD0rtuL(u;)dY{C@_BG}6T};G}k!0MSo(j+5Y#gq>h`VLy5ImTMpwAhU z1#=dM_MOMc4Qc3npN77dr=k7xI4++|LD~7^bWKTyMb$C9ayzie#2kdD>sGf=blGI*%xU_AE*%F-{xWzz+$|8@~)mR*JDrfh`J4QbBPbjl^p zM8@UgF#MPTqYEie{dfjZYfeC_>jZv~M`WwhIgHaKoLIaj9rN)Yv6w}3!Jz6GT+)dG z+Zu-ZEDX-&p@CIch(?iiEG`R&p++nMu0tV6CBKoA zYACGenDz4tMaj%iG~JEFWP`(`cD5)y$fw=nET55n=RXy`c6F`Z1FsfZ|C`V|3VC7Lbr&|EPk z8kdO&-_#O;VVYmMk&ZSzHWWX?!r?C)jo^)ubnK80c7U|oRm8&=ru@t4F_^V10u3`` z&?-mgkzEq_+(O}+7Dm3!7+gLW50jge2RZdH#7NWasuKqTm3XY|io@Q=DX4rG1AXx$ zh}B3&a?S~0`B~giJC2{pi4Zz?7xmo+|jgsl^$EyV%Mq zh}U$R^ij95L8BIi!FLe3wia#sD*yA@h$Q)S7S;`CGMA(<5yTU=G#&Qj8{UVy8?~7nc&a4g8iD8FoEVju@BE9w&^^&mgS&rQYKtz8@zNcQm$bZCM0D+ zK`0ltf9O3vUB#oE61*_Ejt@`spm+WnR`}al9c6hQuIJ_=u4!8>~X`bO$&Psl^`)qG65Q;dsyicscQLEPRl;x-hZ ziF(Yz6N+#(zX&OlZ^G2B1Y$o*A=gs?muA6$%E_P66l^OMw@3Deiv3i zd&4crsa1m8T?Uf}W!QJN95wmWPxmOo{uza6ep!H)MTIDlxCuu)`hD5Lk~nD9m{(ne5@}7$xXZ9nU@>}H4Y5s57l(^D%)McNN1592 zjnTurXkgr!CRQ!iM9(-4NTib9@%%Eph%v;b7y3}lQ75gx4!-28ke@~i36~i5{Zl8# zv^H^tDR2f$WO9!F{6mZ8;9quwq5aX+hRDLaln=Hl7 zJ{5ecQpC~(H8i|X#Y%4lV!$Zl`v(>3u@OIF*Cub1Ijj_v~_5H=4lq(mGsbE1^7Qgynw>aEmcK3Zmy3 zX%Rzx5nX%q&?>5fj~g`b_LwHFP0_&1CKW`^T#O2R2H}gk@F!l;R?<^HlT%043N@VZ zRK(&0C7ho_SxlY`JmJJ1_)nX1h4hhpgx0~?q+Jcw#-`nBaNp12tDzp!*BRk|oSk<( zR&V_O?L9)GghDB@*Yt<3CQb|{KUduo@I(H^K&nm#QJ#C=XuBB4b)(cU{G62I&F z`}6zfx5wjtJS5?q``qWguJ`->dOlmL93WL^1+`Qg#;P|(SfeRK*|u0W#U37$%y8ZLl9O!gl{at(J6dHj~-i$qcX*9Z2qfp#F0kTGHR_%2J1`drv#iKaPOHV*r zO9Wh6BH_a_JdSt0Va#fuQYk)2F7bh^UJ!Oqb%#kgfqg86{ug@`b&SNbHx?LYX^ZJ> zKblqNh?u?1`@4{$XYM%UPh-B(G2Za_7={sdW3c*pB<^nt#f-3U>rW`Z(ks^}w-8e?--LB0AL-bB#RFXFnEig1j(?<>PYqIODpp6L!Uj(38y=987bd z+GGdIqfU77(*eRaw(nyZqeCgohx*Nx&CWaG33C$P9peD^x%SMhNRS-MW_2%)hJMv( z#)YzmM7lL{Q_L`1a}-iLM#IyHu`8@>Q2w74<|-3xka5JUKi1IF6M<_2GP0aumFI!B z-!9;)U!H7=rnIgCXL17>m06pBq%z|o+lGKL@l#JHSY-fve&~eX^wd!ozco(51T8!m`iRPaL5Z92IFAj;f2Q6K{z}r z04hHNvGbol3}5(S_Ec}=dwF9!JKu`R<50NN34fQcYvCeeVodiz4CCM#?+QidGRDT+ z6o9dRDekaqFkngm+SFO@Div7u2;leHAmfQ8nt$73w<^1aI~)-I*BSR3T;Ub%2p?8U zk-zB$OAA}5%w+YAjpoed?S$ydHn^Y(JhgGf9>;NLjq<}YApo5-*m+R&#Ef4ohv-GH ze})rG4-;g4%%fn?Gg{>LZV0tcEyZDe>lcPuo_%24vlod^8|aW z|6z-77M8di2k5wo@bWE(oBO?Rd2kTgeL^rV$_su89ys+c0Bk3MciTLX{L=;ROhg#6 z-wsQ6jD_?$=E<$F1*d0&x>p2J!zt1hkHxCXuBbfY#d@%waA9nez9n|JI@by3{T*Rj zXN^7!8x)*$U~FqA^c?{lBOPG3-w}a>t(foF0?I>eU})fg;sz_IZnZ*EAiHl`o5Pk} zLkFK&<9)I@DleGh`BGbKJ;CZqw;i#pgkX)wIOwoh?Y>7YxLhQ{ST{%P`sRY}>;P1} z55&U3eh6rHWehJ4b&BrL*Yd`xa+b->_Cwz;A2^%~#ra}iHdo-!vZrkS|9CuF<2}Ju z2C{rw0IoSO=HZ4&R%eRBmVHquVl%3ir=yXt@Q!$AJR_0XkI0S#ugLD@FPIDF3F%_F zjH3fuNOIRlvfz3fSvB}IIlJruk!GF}S%v3h;ip!jwc;Up({h9C9)6Pq_1z&&7PVxD z>@#w$yn%$Ds3$(Rn#hd6YEt2Oi?|!rlOvUNWZcLHWJ2nFBKPMBc~$=C|2s)so<9RPgy7y0VXUZ3{YRE5=xZo2R)A@=xhyEZhSG^;XH?$JV zwC5zp>JfS7*h)$xo{(9rE+Eh_BvT@^{oj;&S{E zu`g#CluJ*EbmLPpp{|8&w`?NAS{{&Tb?ob9Ul9x2mn2{H1+icClk7eHmZZIXNmkBm zC5oPJh+6j>QpfiGQsxiIQuaHHvc5~m=Vn5D?h}Q;8zlMZE#hFwTFis~FPrd2HLkg7O}&bv2~W@e70*Yroy=T)Pr=?q)?bg3ns+~!2*xQeLj zbth_mhtk(qTSf{XliX_MN=d#Xoi_J z-8b2RrcH39zoiJBVdY4-{h(Ai%#pr->Ou?7jip;PJ!#P%H`=ntn}&va(1+hV=>6F7 zG`!1`{SJKK-$HVdRTk4>O!Uk1`V{a`wDq(4pQcc%*-2wh?7MlDl3 z>Ha-VG;oU(z27>HuE}?yM=M11T_;D!*o>v2yX@&ycUu~t#^&Mrttnq>N$GlfS~ONf z+iRTZCBsm2GL=?{xo@N5WV}=hkna(qgd!d=}i~vutO20 z3ly32Tmi-v(&$_-jS>H8U=8cN9BNa6ey%FZvMICu^B@du7>Lk&3P?y+K=Bqu@Xus$ zbf_!}ljT`|mi1;nsbb+96=e1(vUvst#@bQF%m`Hso-2p5N*dU|Vi3zk55jjn#^)~8 zKxeo*a>r}ne1$rz2M<8nVNG1_*MQ!T_ru^!D94P_Lb9)J(i z6_Ncx5eo{HkQu3r3&x5Vc~A*4pCmEZRu+a^Rgs%1g*DIR@rqN%7He6y&QwEsw+t*U z%fKsD72|KKuo)tC@HI-{z9`@b>se7lSy-pap!%u;=8cwy^ddO~tXD$xH96>3NWko% z6w5P8uzk%xl02z{+<4GIDtfz!t(*k-g|ZM2k;eD{Iiz>WGxn1ro>?g(qfrVMx65HC z>&G-CD&wH7JTz|0!)b#&5;E1%RL%S~3Cc)XtP1C!>Nxm71#eHN!*r-78lJ0R`gX=7 zf1?bgUKQA6sX${C`<}}baPgu7j_*@sYZ5uOPmsp?aw%34JM&|Y z!;^Kccp>cu(-pxOQ5y>TZ*1M>!s^uiQJB>gj4fsXc=^r)?_LB#@fl;@JZ!UTL4 zSdD$2J1oWrV4<}q49~DN-)BF_vG?Zhg90&1%^UxmoZ0*{Fs{G>7g*mW$JGhrKXO>< zI2Oat`l0Rpcs#o0&YT7=D81u_?4M>`-(a~phx z+<(E&(Lx_&%nN|LRuHDs2=u=QVtb;An6)PqJ|UqPR}=@WCdM*c8I0!g1Po1^hNkWm z@OhyKa+-*VzoM`>GlF^IJh0Y$BI2z4Pv>X=553B`pv8svI%NSc0e23FgI-j z9uCVvQb;yit1=GGzm0grvfM9|HY2=hE3Ag?z__71@aOvuEcM%gBC}o43d=wt%Z()8 z&PT@kooL>bhetWP&`opUkg*%4+nFPbWf`U~&xTuD4&HCeX52Z(yGhxE2~#p)n!5>= z*Ed3?I|pBXZ$jCaP0-BW4Cmk5uqkjecJyV#F@8OYOgFH7%3AC=kcPc1mmNQUJ`($AG%MCcaA`@i+ zj6E@B16$8!qh82jdo$*OK9Ixa*m7X7HwziE%o{dhC&rK3f|JEtn5$w3;$nAWip4I* z*ULx#*&gsU5KS1U5_ncPMU{GE+;%=M>wA`2l8)}yeDvF$sv z8ACh=pWQaYCVo3UJ=%iD{0_#)VVw86e0-m?AM!oj(AC<8j}~9iCC+;qEsccUs&a>F5bF+3~RZ7KFB;(YU+70|o^? zSQp@rZ5izK=jaC&ZMJ_|5{SM10ocLz;wau=XUKz2mCj0EcQ$$ zNF8bgwgj{qtY7zreXc1#04WMCh&OeG-B-qg*eHU^MthW7xMS2!Km3&o#X9R~__+sR zVQM%gCk5fGg&VeQn26FzY&I#v6W-t4;r-E@)k}Sma?lqmBitD$fN=w@#>0;t)8zeu zxYg_jzQqSK(nArqJ`9=c{*~bp4ui@_ESeR>xQKyBejWnN^C4JP90JdOlOTC~0>=Jk zebu-q%u&$}5PyGIWD zR>|SpT3L4g?jc^=B|-1V<3VT-IquL)gwQ{vB9?KvKm8+?gWAa`Hc!zY-A`P^zsNwD zza+{04_WE=jhM;*A~MVRNyq9=a^d1`s!M`ib1o`bLh_{2<0HgAv<10EUhl7?rMpfLH3!9ixf|Hd;7xb}*b> zb>Qha6dQ^MBhT7^%~=dYR{ua;O43G3^k7sz*T#>*gJG+rha8iEjLoEpe=!3Q^j8-@ znsgceN1x5G>oE3=4vriehz{>TxE!F3I0s$kgC2-&e+DqO6Js*?X`}J44%{SlaN0@( zM;x?~-l&6@=k-t+VuGRJL*XuIh_Ba;Fv>&^8*b`j%UUB0b~MBudqaq)7$WP$FudGp zfK!%3aCMSCD*K0E)5;+jX*>*-eWTE#IGlOIhv7>#bInN^;GpCvT$dh!5PMS;SQ;Vn zFmrqnW6YK@LeeZ##;MkY)~w+uu{FY42ga}%YkV{2annEV)y zxODdKO?o&r*Z|29I&2oj5NV}DP*tam-xqc9<<$smcsdk6tB1qk_E0Qfvr$^pb&%7c z&uSw>*nM3eT66W`V5E=6MZ@4zXNm58W-Pa6fzTpr#z|%jtq~TG+G&dE>^vN9`5%^* zA(R#wA#e9BQfSgZ`l73enPw$1zkG#cEx17zoNOWy!8gf+54Xr6+CU6#t4O$M75UzF znMA+6MEdIlQZq;(PdeD=!>*B0S9o&mA?q-EHIhe{FB5XIjC5*N5lgkpMCxNTarP95 z+H9V*`&1I$9-i3PR3DRgJZ&Ittu1nX5v+h+Q z=~+YU-5SVk`8slryFnIKHIl5x2C_A|iF^xfAWf%l67%d@a^@zRpX<0zj=EhWi8}Qp zL6;{PwU>zeovUQ@=1Ss3Yl+E}OGHUYAdB{0B5!|HlEB9|iJn0f3Aj~5UQW79P93~S zIQ@I1WyS-tQ>}?)eXAqBmi1)zz#5|XpoTc9)Du6(cF=qBlw8~JfNWv2gh_GD#Ovg3 zvh#2qIn#KP-2Quqyl=TpWFOuorh9IZg2+Zv>RUr<*)awDzC@n=I7`xODo9;pIiZUy zNz=nxaxfm^KhF7M}nL!su(6=RA+jCB@kr;pNRQn8yaeQBmL8dWt+_PDx z@w_Q;kGUt=bwA*{5oRA8iEG`A0hnWn9Z#97&}j^OPcdHJ1snLW+V-_3dz7*f#zB+q z;a;nP4Tk}4yfe(l5yqBe^=26xw6hu; z=jh1RV%Er1aYF}ymK!#Z$hXJP4oh@3m|^U~F^v0Uj!nPpFzdG!mQ6Nej3gV@AF;qR z##Pxr-2xV+HfU@%#^G`!-14-;ghpFTVEnb^abs|^-xkN%Y@D&-XvkeR!-=m(_%(~U zAO|uQVhVFr=9#emxg|PnEs^=k770eCFktS>nAPmO_1a_5NNW(zj`fD^pc?Li9?!A( zuYlE+)hRY**`aWNJq8?LJ&_1IX#BRsq2D6tC%PcuHRC$G26L;rA@BM)^c;3aj;AwX zXSkzqsta;8I1CyJ942_T)dNRT*vy*P5fS;$aII(TxPNMN=TRl9p2Hks z+f^vhRi+V73%83bC96#GbQO){Xb|Z z`i+hEBxtU-1noZAg=fS2nZLUismCPf`mIvb`l}RW;1eosB}ey$DbuaNax9xCOI2g# z=~Iycz1_f^Y>cZ=@%s--NDs@Twj(6=8}@SF8L#>yejfjU=(a9ad+)(vg?}i~?njxj z1pT5WO{>}@Y5&B2oH6M|*CGkp(a;Okx_?*^BuQrvXPks9zo8NL1s@&0VDi{@G$(bz z&Gs)&9Qg~`DH3$Gn-pE}>>m{CBq`mhNGrUyI$) zjpC#2*fjPJJZAM_^CSr>S1e7_)=E+d1x5Pms}@aNr$tv?Q>QDKbFD9K0Nv)PPK#L{ zb`#5FnzyRcQ}0yi%JJ&-!wf~bVT%%-ldDcQzE+{RMkwE+W6?uPZ1h)mg``u|HhD-4ID{F4zZwj)J1^)U>4*wS_A24QF%{h6)ZqA44fAKKVo{DXR36{TvLHgICBC!DZ-P@BG)dG zflRtGnue+%Eph6CEp6FzuQe^w`gLtI=dGUL9P$%pA%gn)vcU1NK)3Vq_6pC(YBvj0J-+ zcNcmAH?Xkyx#(W*ckHmy7Hn&^KI8&2`Lql30#sQ{`PiBcv`$xfE-Vo~3 z&0)}Pgd8IyyfCwf>rdPz?(pR z^-xI79ty#85cX6J!%Za}oXF6}*keP`C>jEV`J-?md^ir>7y2s1x)1F69vL!4;X8#zSH(>)kzL^Pr17 zp*GDEPM?^+d9(vgKV&?hV)p-iZ1Gde9mcU<$gpO!+$nDG{p5_)-QIXRi$L$Z6Z9XB z!Zc+S9Me_7+CA#9idMnZa!uygQiVj+07Ry!K-XUZ|B9q=u3DWr%LYJknj#)LsK7Eo z5#L%Ak+@R^P0OXR^LrQRVe{~I>mU&l~12x-ExK0B*G)Y+`tXF{T6-nf6lw|ctMGW*;#cKsstZ352hsEkFv#1H10VeX>bN)>CDsNqGDCWfXe;bMmrYIiZ`q?H_U2g%}`i7I9l zFrQMNB#K@rz)DF8srmA7?`A)b?S&uYE8!ZuSG^2U#WEvVsIt9WX3zkfyCTOj(7j}c zYd_g5mS&uFIf%;GKAP3r%MY{TyxU8Z6FSKqn+~$mq?gS4B?FsfUBoz19;=?qV%Tv> zL_O*v^7eh?+DL`Nu&myu#430$uEhS?Rd~MWBF1EuAiDJeezg`uRj&-LI;FT3 zegU&Tl;WGuCA4{zo(RrMzJ%f1pGx%_m<<#BIp{%(Cbw&T-(RG$R-~AsxXZ?qQS0&8XdmgIV zrEpwRfzh-S#fn#uX2E_(T^=_IYLF&z9hLuB&iPm^Hh0wGYV|E_NT|lE^HnG~T!yk` z<*-)bQSE;YZl_P+^vz;+48^dlEkKkbuJpVyH!z zV{yiHs0CKy;Iu2Oetrdm@-M;YYZ)>m1VlMrMv2B%yi2TP-=i9tqB`uE*$6$K8g{+i zMTlZO0@=KpwPz#bL+Wrt;u`Zc)+2t{H3U^&MOM*0mhZd{vmG@EJiumta<4(sxQekR z*>SL$w#r%cxLm+ygKDngRg5G)F(+B#(Oz|k&WuW)x7t#MF53`+;C^79J!;3q}z5l+GEkk~i>bl=#gK`&n^j8u;lVzFf zOcJFN`p7B`30%GZhm=OKTFo9QR8MRteuE^@%4#WduYMt2gFlgC=7gKt_?g^Q{y}uw z-;spV?ZjzTH(6T$i5yAzMea1Vljje6N!qnmf2`9ptO-A5!;I0)(x9G#AR?*%%q5<+A%mM<*HfLk2%a$YZDEKk`!M4~b#x zr@TFqkbd-+jG7{W*X|Ms&X$9#8LQc_tki?)lF)YgOzvD{=T_839Ikc~%hEqY=Tr}I zx%!9vXV6a4KmH~Y?*Af>J-f-~t34#;Zx5*_e@N8Qza-O34m}rs5<7NZZ95}@QPzLR z{E9zB<#`_o|Jh5byjX5(h9ox3oQxG8r{TqLmJQ5g&c-__=xI;H*TZuVt3De|C5+$Y zJRf7?X2D_eG?=6=M2q?__-0|B`U5BnF2pSzjV15vkWwkv3^M_}odX zrW%KZI}_lxI|aMfCBxrlI$objM1Ol6jP6EY!?^^;>P}@dV+nAUOv2!(1bnS!bDDe&=11Rpe+IX762 zG;szdXC@+NbqZz=oPym=%o}ACf#X~p(k?|K`9>@@e2+$esj?go1DD~`$HjPkG#M9Dr{PL1W0q+$?_R)0 zpPj^J+7nR9N8x;H90J+-;Eb4$-E#_LJCm`>g?YUeuE6V2OL0YS0aSM^U~J-9Fub0M z)}G1OcyJP&4PtSH31V)w27xpLz@fy{?7K9h=u{a_Zi*%c4jBcEWIg>&#^H&&}S#EaD^l3 z%5lG$>jG3EbvFTF?-d+nw>!~=bw~T=cn@#e_ z5672FF&M$>@eMnp(R+U;6#k^5R$wl$*f|*1FdsfybJ60n7^|x5$^V?LA@X9fVr5VDPTtaAUp7*(?vdaPuT|F($VjJD!^Y zyM|ZBBXVslT1HO5))C>@H9j6!@}{E1U@DyZQ!wEiyJkWnVfrrxD`KYN#Lj7$(U5}f zM=8kp#k`{Drel9>JhW0{VIvud0D(E`KS$z}S}bzLtwjH`d05!J2uqm)&E5B_Ekk+iN*H&|g}L`4#-d`(zp7N!1uVjq zS+il5F(32l7a-neCT#C5hhy|Sj8~q;dRy%Kcg7ctqAv>Z#SmSN+A)yNHB$9$}7KsGU_$nXs4b!OnyZN>%W?pT$W1^=0=;OEEK z+NvzKzaaB96JocVeS}=DIDvSOZY+G*$r#rgHX+SczG8CF_GojqJFXr zpu0B`i+pf5%N;lG1mjq!Kb$n#xxD3%CG2B;cE7QG?}_le#j6(U~Kq#aIqN$Aa!gSqnN^?9?7P#QBqKM@z#$h*`Jv`1b zuTAz|n7-K$vsqs7>)8l=T@!^>%i^HN=IETZOhM?2D8$FJdw**fOe_OXR_qUBg>Vck zi(*-xXhdF{ip&#

F_pVcr%yaZ6ZhHvv3x~LmEfQ>S))oux|C->!)O5J&_Lyg8dvUMcN$b7B^S3Z|%y2vyFuMBTYbR)tJQbay($ z=CMJeG%4=04~-%kCB-XIqDhlRP5jRL@9$c_KR)+b_gen3_Fm^c=j^?&=kt0#JogZ# z?88EK%reeO=^?df`A@XOf`@k`fT z=j*4dlG4e20=G$EkWTRzYH00bhvbH;1gRMEMe`KV9z_0wQ6S-dchT#jg|w)xoNj90 zMV0&NamPG6I&YYe^|sxPniy_Do$k#rdg(@TC{ggkyd2Iu{q`cf*J9FmLYnQ53?#wK zZ8AI{fu#IsCy%o%XWG0rJvxEvj0QQ@sw^z~R{Qu}x(&EFqRdVU%35miS>MCwiSEFI2^8;JAqt@7Nv zqEgypq{8BRf6{g}nosB~BhP+m@phsjBCF#__tsgl!;)RNLu2=l%9nxsiM6uySEn`o zvpSaS%nD+^JW`_XUnGb^X7l9x2|fOXMmXloKE*X;{iEj5&aC&v64B}t8r&GiWFaZh zmtVa)i@e=j!quMm%GLh9Ms#>1s@;N|SeXGYdN)9h$XfCnSv~5#F_`mGvEZYNOj%VA z9dbd7<>lM%h(2_zqpgF_$&G1&#BW~L9)+m1WE)#*_Gl(j| zp9$mem)_ghMLz^xj6W;z=wFe@NDqF`p?V_k)j>Xuj9}U4?)+TaAJp5-haIylp6(y}le=6T!(I(f zqgA~lNO#5*UiZ8VxwP*mFGMV`&4x$q$(U}OttZa;*#(dn79pZKK9S$vD6A2y4fwAe zMzm(>4t~lzBR*KMj2}lY@xzR-kfSfR5=mb_Iy||AOWR}1Z`h9fwPzvxm5=(g=jTrD z%-1Y#QJOB>t+A@aTJ zCS74%geBwhnNbt2(G3RM=%J)ou2bwdKU%(#X6?UAQ)amlOJVKm`?rK&)-L3Q=46wi zvx%HeuqHqEM+`sT_d4xOqm-HO4n8j&psv6F;hNGkUVr3azR*{;N`2yM{>OtPx?kW< ztZPkUUqxT%%GTY*%e@=v=F>y;zfA?S%SjQ_f|bbr#5R7KX$t>mmA2?)+AfY)Iz$Jj zDe{vfW{~csHl!ebGF@2Y!^atqb=yELUMr0|b88{pI6aF37q5X2<0MYD`@DDY4mf77S4H9CfzZ-hw`mZXjR3>k)MYN1--C z_;p|OiMMwy*|2{O8=$;~p2=>e^-adS@~9p{E?*TLIA$TRWs-2!R&5#}bC!1IZX_z4 z8GAf_B-;|wMh!Lu@Sz%`xzP7%)N2<{Zd})4GX_|qv2ZV$@pKBm;bJ>~@Z@xQ>DU%_ zP`7{|N>HKNlVsTCD%Pw)aUEH<@H=_>VibEwIIky5c2Igyk9S!!oF67DNi2$e`53N^ zm>X)a6RWRRXTJ-edlvu4llBo%r2dMkx_Xo1cM4p~XhVKpupebT7Vw@w3V42-0dKa+ zf{M(i@lGM8LQkRvZ2Tj^OF7hV1}n_icS1hf<_S{tRoh|mpY;UN&ap&7>j_Od+eZDZ zx6%ILllW+DMfz01ikvDfrd;80lKN>6*Y-r6*!BQ%{wYqjE8ZmkWX6-n*7-Crr;@66 zydd*S64-mcuaNNaUi>@jBw2R)ooFeuiC>vt%YPBpXveklaQu}KyzJpz>b~b8KGIrG ze_e|pm2%qXe_faNJt92!B4u{yVH=KB^uS_^2%;tSgOoLwlhUJt&;C}Q=*Y@WI_B~O z-mPDqm;G>s&W*{Z^Zj^wIzV6+2krDOWNY(_ z<~ESVo`a$z-Lq*=tsmR7JAwO9`H{4|_u%P^e9?v(lgMS$6xK?$lx%Wwq(ds2boo#o z;r0F_v_OgflqdKI>U61Hj{zICL4wXJ&)}c^n8jt(O(JG|j7UtulN7G(A^}wr+{Jx7 z*)i9Fo!A?S)_t-h`u!{7_{c-V%<lvSQt!6LYl;9PDfmnMV_T~r>%IK)1S!GuZ{HW zue;>%qCe!0#XLgN*JIO)AHbE(BO~1x^4qs&(p@pzV64F??9}jK`|dAg`#m1gurfU= z^XM+ljT^?+N&O}|XG%F`r*5+R&3p2Bn?Bz$AWbv$O3D5SI;671hAp}^oxEyJCIdq= z$l1@ zNN>`os?V3C_~nLIDPb~!&DWyXAiig%ov)>1^iFv>VDqd;C@5*o?+nk<~!B$JYD0M#hR=xwGI`KjXsUKp4ANDju?C*U@j4 z+N5wmfm;3X;VpwTNz2Pj!nwqdGeJ*LW0^QV_CGliFrdXwept(8tXxBnuXW?~S8DT* z9WT(!KmGZ+3oVK6X$gM3;Iqt3_(OkM9w*hsqlwe;UF7346}o#|2!D267~AwHhL^lF zuDa+-5b1PrAt`TyNn(aTV63@GXa1K?6=z7X9=FVi^fxQsr%j!|9VRE_OqKH!_8#T^ z3uC!~y$<|iq4(`5+e|7SZl)S5OXy#F7xrOAAx-`Cmdx5I^dzKI^23Cl>;2iTyiQvS zTe&cRj#pbD_~dr+*Nn=@w+}YF+>alemboi3`6)Tn>-9^3X>}ig`e0BII#(p(3EP0Dyi_##yimsCQfwy*ie!p^mWJ|G-adL-NPxDrm;;` z!`S@VTAG}lMK+WN^6QhPaFtSG`~YreE$gIM7dc<{?a!G)4}&;+wsNCrp@hH#FYe+! z49xlb&}_DEnBWV4rOBoaC9;dA+mTh9ib;`>^`V;LK=qHjC8mxuh}rxqqWIW}w|DpC zXItI?v6u3MPOGGfsr`blGlN(1`ArpWd-B^1vx)BI3GA&}3*PdIJedP8*~x!ck9qAZ@<{#!Fx%IpNUH*NdjIT66Qrsl!o}laS`eTU@2hAG3&D!)(s* z=W%+&L6%pXEX6O$`AN4Asj*^{ZWF^r6Ij^|Blx}6!Ww$Jky`WT$(Gvtq;zUD8>?{Vi9y z=%7w=f5@K=Q>`HrLfcT{pD8D{%ZMEjUP5Ht6j|TwWUeUDoW8yCkQlTM5NoaLbnc2N ztT71p^KB*8d+!}~Xuk>``>Di64P>wpZ>8ADgX8Go(VqOR6KNRSm&zJA4RTqUEtp<* zgq#sOAu`omMUpSSC&o8h2-oS0&ffP4BdyEII|Y$1&7o*{;0s;yshMau&LxoKL63dv zBL0QQmalLlp))c`YE-Id-x3*GYEnqv>cr8CEy?T}-vD}1onbq28%g11ExbY7;Mkjd zF4j(hy!#kK_OExx^ua^KZ>9*>$?hfXsU_s?KQUU#j)rA>%g{LJ4t-#hPSh;-5$*#e zX2IXd-tU$qMe8l+Te6CVXD$-FYI>}~>QSt=T`w8gc7`<5@`@+Y8lGb)o`YV;*Tk{;s&XD?_vVviI%Z+;fL zx^4=;`~4y^ZLL45&MOl3@}Fo%WeP2__hCKFlW{`)LP6ubMF&=UW2Air2^+hXHF17` z))6U!=S7uld0{@#JWnyhvcqUzTI1eu|}uM;?-$uEDIPLIN?$RG~*k zC&2`%crxna0k*@c5k1ac6Bt-4*pAaL*@>vd)^AP#`#c49%Bx?T;tFkgeU}WY()r&7G6rPW%YC1! z3L2LXtC`=(a6uOg;4|UBOcj=K)?wccj3)WRF5zOmW~@hZa%ZkBDVk71Dw8I251Mwf zD^Avu*NpN39_3IV+t19Q|!N1q&{600T(^R2`i@(zCRg?MP zzz58St_q_1>m+f~tR%B{`mj)=FBSrts=AG4)xb?YH`O*6Eyywng)XZfrzosac5BsRk&wsmspSepH4@&N! zOZWBBX+hJ;>dGVZY0X~#o@FT)p5Z|pTw{r;!~iE_WzJ`uTg*>NdP|(XUL*dlr|Dkz z3@W`tTm(YB@8cCoR_e$iA;0bdwORQS^J8Su#c2tBx?CjG>2Ht)g11I>A;CMc*8Ke% zRX#GTmiGP~&0kvO!w+>AP?z;P=m^slbjPq5&h5M$i@O)IagJ%U!7_-XIFS{|fgr8s+N6BXDxPTa_RX(H~`hL=1WxUgH2-0Cy)>F+tC_>!4QM8D?= z9Id%bSKYGUoo_uSs+*Nb$p%ZpTo62ImU4XbtD{sv08{fI3C#&)1dMP?}v@Rdzew?a&&!;O7&nNLVzeFj6 z>-om$3esL5!qo?~5&iNhRKS4K)|BaJVE&Yj7!yp&Z^cq#;wN}c?$es>!)cG>X!d)Z zBELWWKBwdpMcBplv?2>h%4@dJbh9djEt+QB*QRuGX8Q)NZ{>2*KU$1G<(o#w z?lPdBLLToIgJOP4?13hejR;| zeM=<-W_GH``)Ds(3A*-!(Jdq=VKgnXZKf8e#BY9bj!O1B)4lh-*lR+D{`=Jur+)i*a_DhN6CJ#sPJy*GF-s;>( z`Q@CKLM&C3^yKf!oAdUrFUjUd4>?I~c{WWgigULw=6*)xh*ZP(a?AJeT-q#qI#4Bs zTT|b29VaumI~G^yxzF9z_v%(sPo_yQvY6BB<7(*9-AS;OTZQ8H;h> zqz}hlq8FF2Jijaq3#_)#l>$$#C9swDH2vTLPaNWYDqIvLP6*@1FH3>$vL?E_B--pK25HP)x_(s^q3L)Zg@8^uB2qHx%Vw9lYxuchvkJ7j2{|)Na3s z9xce{D(?=6#Pgh=&G)iQl;ltO~weV&v^dZR64i5 zk4{h#%2oq;)ZAK(bv|*D=!o3Oiv-M+l$Q?Y*FKEZvv(mXeZ~N%{zv6EyK-sn<98QsR&OfZv`<9aN9NG(+KXgzjWR23K9cZlIh<+GUYY`5 zX%Cso51k#uw*?%^R{ec=V%jQfrv;~Siv<73f;zXo(_!Z+@P0cSZJQdW`00tLyL~m6 zb#@(hwZnxQvUB8)_uA26sSTt|@f8VNvX7gm(#B<0s3ra=Dlknk9o?bk&jRvpZAWFFMl?z%_N>zvL$7Ee6y5j5`E?46v7Eik( z%J$tyV7e+s!(_++ErY0O0zscd|%^oJ(cXMc?*mZHS|9$7pmXs0`la&?we?IDB4q_`p z0@tkFzG~vi6`?C80$MY+U<=coQEc)5_4B`B{x^p)T`psVfyMs!{r{T-{=ajuv$3_E z?risen}aZ%)BncbQ}#8mW*kb_tHeCqs{RA|~gB zUfA)pr;zdy4kOt2?^I_W7tI)Ns2#l-Z!1qBqw616Zm0ha1 zR67B0W$s7YgQ@tg@)$<>Wa7rGy|7>LCYY!;!ia6+=vr@x=W{HuKJpY!UmcB?N8}*z z(Ly$MEI7i=u?3hav?85S*a;)v`h#rUYgn-99-PsZ z!y8xS@us&bcJzLM1fyw?f9Wv9&Q6E5Gg6>xLML?pu7~3E9Ms8=$I8LWn7c9xl{_=? zMZgif^eGrmCdZ*Fmw-DJTQMo56V11Gpwi!VxN9Yg#|6)!OXdztO>#hadJwOv--4f; z`oT8fDHt#8ffD;PSiki*L@(vA?o<<+ZMuz|&l?y~3}|?NBxWZq!Ad0;luVC9$9^HB zr{)wUcpkyr{zR-_lZ1z#$D*@V7I@}`!R70F!RPvEB>K1UdvOIeYI(v3)lJ~OFc1DM zNyleR$@q961$WH#LnW;p7?!*eOQmbDukH&jI;Bls_%=Y=m%m`JOA#Y;rO|*R_=i!) zYHJ=graXk7Ki)yz#iQV{dj#qoGQ#T`{UGzK4ffT21Bl%<(x6-dt@md zR1Uz-fMl=_ErjQlhv2A!3jUii8^@gU!KRe0xbes|bgJ8h?IWV_;F3~YcC#G)b2DJ- zDG}7=*1+=d)vzgd2%e5J#{=`~p(XnlTt9sqhAC&mr@AdrSalVI^ePzTcpIK~x5Jn1 zCOCea3GRM25oOPRfP(cy5L_dL&(p8MuSfDIvqTKn42ax0Q?sFI0)bHh>UX(VP%D8ema(zwuW zGLAhz4l4wns{e@D#MfjNak)O6Foz9EvbGeNENCjTbe7=TP*2R>G7(2Pdf>|jz%2nI zagL21vThS_D<4MQHvsu5v{6234M0g{#5-G7;j|_>)N>QV&+A`+X>dMvS@__;vHNgQ z!dw#HJDI$EXG;DS&cT8qYdkZb;PE$Qc*iXd>+*6j_tPiX)!YmtHe7}?;?B4wGaf%C zZN-1u`6&KnCr+AFfj@720=3~^!KJwmoY>dkKS3KW{1V4qQX?_VQVJJl$YP_VC;1j@ zLVC}6kjvL1iR;EVV*j9!JYVqy1D@SO$L%V_O-*15dtbvM_7T1ru1Um`ECr21l_ZKq z;mxX4JZrKG_p_78TWtk$wc#Tw^YVlbpFrXdsF5R|UctPK|6t6AZ@^A-$9x!paeh{K zq@WpN3o21@dJRU_j>I#chT*8t-|(M)B^qTGqL0i0f!Do{JSr+CKW^lboQHY1sy-IA zJF_v+fyJ~k56piu5j7{j!2*LWoc>ptXjPb@)o69}FAu~jnLMHpoJO|BttZhg+UR^n z6_tVwu=DvpP`oXUX`Px_^xTG=5;{Xg)rU zd{SRZI=nUzy{P&4UNH~NjvvMm!RqK@=7Op1WAK+>6TaHhf$g_Daqz`MxZ5U;)ibrQ zef4!bF-DrKz5NVV%fCR={}jmkDs6IhwHhg&BSj3Sx{qoBj=N6q15!n zSZWxLr9f2riTGN1mjCT0GlX>E0o4y)YgTVTC)Hbyx)Z1 z^PMny{d!y#?SP$f&rsv`bri2SiEGB{lCS4(Vd6j!rnz-M$KltIC-WDytFlnVIv)?d zUV+c8PYV9f^JshPICiW$iXDfK;D`-J(K68(r~2CBKDEi{)p-KvUbu!T(;lGC@ed&O zV>r&6W`?J1$}w`ze)RA-jBId@~<$jK^?ak z3jFBYU#OU;E9m#aocvy(o76~hBo~RwGA%OpyEPdd+5usfl6cKR2}`a>W4o;+cAgW* zXtoOqQmdiqXgj=mcMF@%PvO|J)%fIs5#HErh67hjP*wH-)?X+<<;ZfJmo^;l>uaL< z$k8}+L?GISF2t_VnOHw}5B}86#5b?Yuzk}G)V{hJmzen=<}@-wZ3&UFX@aPPP{{ol2b*4;Wx4~qndL=#P^=dZ zE5Erw*jEQwL{39M{cz!$NTGm_h67h@VaBKixEFWO zN^rii2=rg)LEA4A_|9#GS%s#MzVjn9D&!5r+lRrQJK9h_WDZ?35!?@|0a^MV<90R~ z^t%4SH=z#tJhTQBYFeNvgy5P-V_=z-KdhN_5GI;R!fktJ7>Ef1Uq2xQ=?;S0WhsoC z@d@&ZCqr833^>u)#oTJ^Mt8ALNIaYt82vW~4TZxKzNDOk&{I$yg{0a|IV!Bm8p30A-(; z!5s}(c)%yYF_m;UdC3vxCgebRzbs@2HAC*k3J_yggI-NA)J;zWhecJ8vv(&>+5VoX zf3cUTULpn7`>bHzLldwul7J}JQYi4<1P^OW(I-g+-Njk(dfOkk@!u@SY2C-1zyFSD zT@i-q>=HcVu@Ym-4e+x{5=^x@0dtfK8Rt%0X4QO6s18X1C8hg}+|7;PmT3>Z-g*$D zs{yNRePK(Z1n9<%hgru=;jEtp>{wa`H=X3-2V*l-9eC)@|wJ?&66MhoYLnnIFn z8?(-TEa>`af&a>fOvX2DSUPnvocfapnog|{+m#J|b<4rCb0ti$GzPQi+srikOR!pi zs5v`K1McuOh*UR+W=(C-@5+M6z;1vK`=FyI6hcEwVcxQGc=THmLsle!+6G;)uyBWB z*E10ES`Y4DSOm2XE`zGXYLIoXgaB9y##t6HZK)S52p}6)G zwuE{&S7wZd4KvSTG%WUzhEaW9&_1CB4C{U~GcEQ(*w&LUVbM({*Eflo|0jl#o0rSb z;3tgOhj2KeR0u|k0^zk}3iMBU0eCk9+QU~1IX(3-DX0SC%F|$N_+}U-^NvxQ`IAw9 z84lWqM?&)Qad2kLL}-_ng53IDaJ|nQ{O@VP-IX#BY}o|g9(ls=TO1gMon;(UW-{TE z?lHf1Zik)&S}@vV5rp&&Pvt3OV1_tb5^Zj&a zo*4mOQ{6#-itrm|j0T-IR*<+?h|7F;3SP^aK^toX?@#<5-OL=F)iD3Z0;uYW zhxwPKVBs!V@ZLQU?(7=FRDbHU58oxS=Z86>`Up#W^}zxE3@&9jWedjdoiu1XN@wPT zq%qwicQKO#6_~aY5A1V%pV{9CJ_ntwE=s=$!2~CA> z;EpB_?V%+wf8Bi0&5}g5RAqGEq>9?>Kil6V;*4LqAyacFpOHC#ijm@6zA2 zs9F_tkJ4ebUmD5$U8FC_b+_!J4h(1H&;PbxF;EXrQ!3!|t3-$i_-gOeGiX2Q3(H6t zZ-I|KzBJU4_hJrdqhjRw|KaqG^Dhx40H9$0%mF9JNx2qij3mq zhfJJ#4Aj2d1O8EF`1hJ4=Ep6<5%*n~OjRndL+6>Pn`4+41(O-2jvnTm zzadmK+k==z7^I!p&kT9ZW(x1^XS|+>BWI+7i*yTd)J1FNvQ075>RiQ4bhKj1Qry>!8 zcE-XfDJ86`HV65wEYy!Q2S)`r=E^@EW@T3>9UqV=!r zHQy{}hT1NI_Rc0SwSNjypRPm1#aA$~+z_c+k-hQX0sAA{LK#`Zh0w*MgOO1zTpX{+ zoV+%ixvssLas51!nKARI{i)+u>{ra30e3Y=KzzF>RH3AA0Uuix(c+8+ysLyD)7v^ovBWf20b+!815JY`_!UAii~1n z;fa0Hq9CR`&K+b!Jwe@TET4C>4m+U>!Gy65G}_|gyp*m zA#d*|c$Z%W!BbCwq=zIL_Soar@v0coZ;wyj4?~%%Yv4KB4!<8$z`OQ=)+$s&wox6#Oe;^>y z6vZ}KVS<_=emRv4eX7=Ivi>l5hnGWlrWkG(lSQBBGjLemLLAn-24t#dg1ciG^v6}h zl^1WIOz{EqH>`uSo-F9?`vGQOBB75F>I~cJVC-53BJQk(*>`H-=4=xo19=s08k>c8 zr~HE30rjw8n+m>;2?6E*!eF7!LU{OB0*#&ez+dEq?~+PjqeeNj3fFdah##&sD@N`4 zS5a>DI(U2|8Fs`M!eHw|IN>C$2YZX5Qb7?VQ!TM{djUR8Ert5QAs9EPfgOVuFmrAW zw7)(BUDrooTJkjfFKIFUY}^j&zE-%h!W_p6_geakS|~C+1okUbQNwm0xLc3H<1<}h z`N{q8>8~n?ov47GQB6=Gn7>L3e!<6ADtIGrI?fFmhR?JnAh&S=%5@%uoCgY+RTzk~ zJ+eV8@Gx}NO9(ysMsQbbB3$b{4~9(zp!VvO5NEA}TTD(s^_M2lA9oraRPKVWV@|=F zL7^{sX*Aegk-&Xxe}VSXnb4!&33CP;!Sldr`1w%^Cw&})x77zBq~#>!rpe$XVUM6U zMGY^A>~QMCT$orL4HFA$pthtQOxj{0ySiNX`wSHMH$r{$3s{%=9Fz^u!IB3b;h4fM z$WSVUi-#VXdEt6W;mY^I z?>;>Pn%4h>{yAa@D+U?UPf@VX_!Qjo9|hfSykJUjJS>}f5xTs2h%}eQcMan}#b7lI z#D>F$xHV8>lL|etQmETULCJ3j%Ed?E&{uKH8)Js~%M8%#K{ou5SOH)Er931E^wQnj;g8|@csOE7>kEt$Khi@J*cxQN*Ls%W%-j9OU|@!B|~GIJ4pm!!#mOXl(?a zEzAIy1i_W*=Ae0}1k=A92b*~g*xXmn6m^F&7@op>Fwcg>Vo#8{b_7!3rt|ksjtlGZ{HZEES(8#dCbVII>oFjSBK*+f0-rs@|l8^X6EdmHniNVU>?}5 z68edh!Sb6NR8=@b&bRp>DIpJSkH^4|&9@<*Z)R58sDtMNE!dZ;2;B>8p-DJX?t0Mx z`kwu;IjRfR-2K25EP2J~*D)~tWHFQd>Nqpjcrmj$XaS6TX#{UfS3;lORc7$;YbMM~ zA3B9|_7|~5@Z7){&a|t-{Dj-gpfrKCp|ik4rj_}Xz`%dQEtx}RpP1=;Ml(M><01WB zFC(yC;rOJNOqabTMEu(VY-k7ung^j{_IlLVCk{_HoMCQ32Q#j5986km3dea(a1-V_ zLRt@J-TwvU+P5IT@fl21tB2?2PM}}j%ftv{_pQ?gBlTO%x(NtN_pSz+=msV(MjN)Z z=rYYFCz-XM{FrcSH}HD27}s7MhC}A#!7*rvvDWTlR^_`gUo;LfUANkpoWLqZK_Z-) zOT^&$hZ5%7eN|9={g4^#xyHz;B{PdAuV$n=k;!^+lNqYhg2np|G8a1yL2>?njH`_+ zEFl1EgD*hI=eJDb3^lmyqXu$MDwy@=ADQ%`dZwKP#yr*#Ie3L0rR)Xx0(!w+RiwUStU;->A!;A}VV7@66f^0LuXrBtaStk$G zZfjt{)pTZZQwNhAS;&|={$YNPRE00gRtx_x0UpP0Fb9*};P8Ky%$blbrt{8LSoJ;) zmaYB4{OL%5B?XU}Y0q1jZ)G|VGAW3uaNNVhzf5IdSv{ky5)CsOBf(_)d!|)1AC#VF z!1~Ekn5lZmOo>Wl7PQ=ElDw56e#KZYy4=P@S(gYXi+@bm$%BmEeOcJxDh0#!#(}Ml z0`#x?%Xt5g2k{@WpmO>&;~#vTkt+Ve)JE$-q>`Sn*Kf7IozloW`<@5Rij@#zZ~}g4 zu7Hs#<3M9hA}o(T#k~1>h;h8|k~uVTkQr6=fN^(7V!~7&b2Afq5@8_3;vT|FM%<`|JpF(I%VuGhf7P*g%;LvCWKw%MoF(t_dD}YS3dc z8N@ya&F=ylVFNZZbut&3Z@Ygp&oqxS9&a>3^WZNgyigH7G$@1DO=WoODhcGa4g`+< z!vqdGLs_yBkFO)JLLhmtB73z>T^%av*a+UF)`kVnZ}bTAhZ($UE$DdZw)o$qNkXmD315$Y0CI7)XDjTZYDc5N0z{)> zmN?#cE#?mG#@=rUxOMM+kX)A!DGRGW?e7efs&L2MOP6Dt`6^U12*z=id(cYz9<<23 zgyqkU!@WZlaN=FStjp@z;&5U2Qbf?Lx|m~SYCr!IYk?!Z1l&wBvRYg*xz+j><28!F&{-CI1C+67|U|zML!+zZi z_9c_St|}C~k6u6p*BGQ*Q}DNr0R0<#8LBtc!sW$MLd{DHLry88%(6~6XFT<=E&gl9x1iL3_!|AoMU@z1~Hy-v!=yky7 z;TcHU!_Z*=LJZp-htC$(<44s}?5{eFJ{nWe*k1+BCK;eyJcA4|#_jqh7@ux~rJqcY zOR~n78J}V4f*TN>-T*JRr9)Fe5iASX2g8pDA-+c*LEg!BNYV0xi}Ut@g;N?dtm%M1 z{s&;CMghFO-vpETiec!(AxK;Fhxu{C1Ok6f1Lq9|cyn7U_D;^kDDMxr^?V*`yBxs# zejd1Uls(G-wnc}AzaaPSFKE40!b9Vwah38gbn_E*nWf6G)zT2UE-i*u=(Ykw<)d*}Lo&9%=z%pN zc?^=e1*3G%z{b>KI3#HA(W);X@#j0C#wLjkc6qS#&3>r9zXP~Csh}%f2xktvLzrbK)@U`#l=IRR_TRi8}-^2!)Eo$MDps0hIrxfRjcDh~K>kA$rl+al#MVL-yf| zul3;OmjVSA*FmCC8UN{5fdFrWymwyc!Rz4p+13~6&eQE(l`=F#L3~SgoKB*WP zi45FiqXL>C46JpZ4Wl9_gZ60x(d{;%am^0q&M|=A3|Z*ayNt`NDUJ%MMw6LWfwj2| zZ;LA7<|zgid?%n_HW^lK7W&|NOi^6R1~q1t;*^DHxK?f_wzi64z;5AwzoLuxt$iWU zY%vV8pAVfveS79gTL@H=g-cIPLBiZC5OU}^OcQ)>d10P-+QAL44oZVVwiGOwqykg) zbReU}8Z5rffpdBu5IEc(=F|p3JPpT5qf$^|=?*Li>&G?knsNHgw-{omi-iwOu(eGK zGg`&*dg4!5$SLDflfCF$pMl|x5oo(@E1LO-pmoAB^cubg`~5Po)PD!(FoZ7`bE$ zOq{j>y3L*8LP{sbsNO@?>lC_b#X-*V6qvT^7>tY`4|<2xz}XfcG+LU3r#{DnCLi$E z^Od-IaRR#Rh`>N)U0h#ciq-OC@s(i{zVkVcj-5AA8+PoN+H6x3S~lLGyd&w(TIZW*`LbuFb-n4dH0!l7#+`|38Y(!>_0RkK(1Y zqh*vxDrwU2d7W3Jl*-B~E0RsJM@ndDX-G+0q$wqovmuHaIUHG<@ilftgQSRgPf$;rhfs7 zhMq)jdop&VNb>_189rbC2Wia*P**bs15&G?+UUYgPi*IRL$wbmJ!w=opQOAOKaata1r2;y^1oebyzb>cG%A>|#zseOd}Gb~5{11uD-<<7Wy&U4UD$$OBLm@Te*>1b zb@(*;0z`q$7_p-n*W_Ao{Esed>-Dkmoh|~Cc0oTN6;}7-u(!aLr7ya%$M0!erJT#F zWvcj8QaQKGjOVBSt>dR+i+JD}O?V58+u(mgq2gqQo^3lx zH;t+0h_e?ty*!TlU-Mi0j`_HoH(H8P7I{Xi#FJ|!}{WP{* zx1Y`3oj7iX0ZW7$@zHN$sPg?l>v~(sWO4@u&N0Ek6GDT_q;?b#``c>i*IazMZIEi~f=U^}PkzH0^;bhlXc7AsnHmcTWkQj>X31))dIs!9)w9}CF zzxl+L8~o3pjQcl)^6z&Oc|(^8d%pRCX{r$jRlYBa)B?5{E{V;sQ5_rf4^XV^pxuAF?Z}W@f$y=7O z&k18;myP9b8QZvKtq(gajNufwSk|>z%?E~!;QHD1u$RA&`i{pKt#JcCwl?9rbPuYs z&Ooxc8V%>fSYP5e1{GXH#6nk=-($r=Gc?#NP|#)K6}FXMPv>ljY<6};P-<&c>81&v=033Eva?!RzA-%nq#ar_kww-fI9 zsYmem$8iLYJpieO!AOtok85MZF=$vOJl-6`lCk?SdH!EmEW3r}@fYCSrNarcC57yh z9EV(B`tDX)zg8J z+utES_#;9C{|J7}X#|?hVx>XWyro&-QAiq!(}64d6k+<|6_!SPK+Cwx7-_s2+mi|3 z4Y%Qiz@LqrYs&Yelz2Alzp#-DL~mq%<4RdAH&4hdsZpB#^+)3xHN?B z$;58+xtLon?AUXnG5YmE)Cpa^b^E5Uy0#GyG+Dw^PdKtz_Fh)2Yi4OdXH~Y2XY~@{ zv65=eAH4_i^Gy=`J;s@9Pv~&Okk5Fd{vH=T4dYKfR#vx0J?C8Ct9 z&!_s80_)i|7jK7!W8m^Wym4y}w*9l=#z(99uTcuO6ddB3-Vy9Da0n|lsIkgnLms$x zHGjCVfmI!y5LR*;8_dFRO=&8RvL4Nn&CHW8Tl1$g6F6k3JGTx^LjI&|EYY}%Y?U2Y z<*^UEH4Wp-kKp98cziwBi0GREywY$z`^(1i3M%7>jWSk8Z~bP^_Wv;JT{!|x#`4J@ z9k5y_&i5;ZaMzV8t{D~2WF5>)7HM-`k+e z4u+u^=6V3z*GEIgG!w_VZsE6f2ZG0a#q2_1-tKOPTFWV6{w_qW@mUmX?avK;eh4hB zu%ir;u{zv$kX5hM2&=ncLwH^cvz@ygk1ANs3AIW*?yMQly?BZv(72hk(UbBx|hK2@JDP9 z>%`XnYq+_5BR4p~FCf&rFg2y<%LytqAsqzk!3LILi!yZBB5pGn4M>egfYRQ24 zoCmnN=PM>RR6=>tYn*hHW3vr<{PM3n_qjZXPp7qGsPl7ZxprZOs|ru7Du%RjBP<=e zP~a@X!|5^{qEEoJrwZ2tZ{o$;rEC~}gtrz4a)^yK&zBp{NA5fGqHbw^_d}bf>>bO& zTY_0~@>-s{Y7JXG3czcXO?cIqgL1E8xSfB6m9J`1De#*4b8@gm_YBUO81wAYAuOZq z&xgL>Mnzj1Za5Z0-`AOE%gA$+?iw~8lf#c^-r#S&XKZ#pn~`iMn;3Sl8JF^G_`pEF<_&_K&e|fS90<$n$3FLTE-` z!?fuoP;5)axQ=vW#Xi8&gK1bZBNd;99mMm>`JCvno*#d)tQS@dg?D;7h^u-S05sKL(M$(*RQpV#iGqONnFDCc!P zEnf3gq?Bnyp$nJ7Na2y_yJ>}JhMOt9Pie%u;oTV3FBz5Fgf;T#4f@Z@3~n-8Xu*`* zq6soTMPH*Pq5r}(j9k%;1x>qg>+~#^xKb}@r`e<&97E@icTlA7Wil)&qmO@|5@q}q znG~0a5@%V_E-g85-(uSI_%hCDdP7Af2BEuD`Q3mBk)zZH(FVFGs+V-9M?se4=ee4S zOw4J}On3VAV-bz@HHECXG5#2)BBW(3@Az)W`9J&e8|!MkAD2jlR{~Ia8@S>mS(Y7mZ`8QrtCW2E#HKtub@p@kY2^>xTR!30ctVYq)OInomX)8TFJdd4Q6u6{&B+s{vg6Y6NB2(!! zv~E>F`@(ek+Bp)rxA$Ue#C#MT9}1n9&oO3q500L>fuDY_iE>k^cb63XdHtOB4SR^v zl`n;z=ey{(@+Z-Xb_GhibQrA@CAoBKCQsT@fIj}3+~7EnU+1=Chi52W3lBxDQA+3# ze8uIl&v4V!3>W^bL{En{vRocfTh1?vTW>EgTQcx_`jG+xWwGGHSlVUaNv>|=C_E|{ z$9#O?_OJ}|wfmz(ayImOx8qKa4hkm$E*b~$OI(qPtmCLVxgQiALn&CUnG*975Tlod zZ(9OjUiS>k@+BzALV`l`%_!7!3e^g=N-kdC$ufB%dCpLy@H}JU(HWvH8bt#GE{ZZ! zW%0w?pMC^OLA>xAnVoRL%mRObpIC*6i>Vks@ge<<90`vRNr>;BQjlnYRfb4T_^#(ilQj@fh$BT}|fqtgm9UmQkC{~5sG-OcgBZBo)~C1l&|%!f#R7f)Np?io=%si!$U5JdQ-(|*MJFR z-Ybbk;*L}^`LSr*<0#Sn=C>l9_9l|3(#BvNYt)?iBl6SjOH$9rQ0F~Oil6tDs?H|U zrYl!PtJ{W=!Dv;m`A>`DF7Wc&(^q$!9;Q(ua6!t<2kt zCAcL#A5%2GA#l`1Y;P2<#rZ49(3Z6-sFSoBng5y{pKEYJuqL+$YH-u&ooH^@j~3hM zSl3m=K9?hT&ZZ2;uLr#OXDz>5RLl;=tNGQH`MmG389!F^X9wR5{@@kCJH{M9@bH~* z5Ll2=VugJ9Svo&D9l*nra`=crBx}i~@|q8GaDK^A41KT>pWj|YzoH7XkF~Yt^Q9JSpKs2K zw_jnMOG&(VYzceIq_D?T;n^=2$PVt^EI#QOKd61omT{%*|0+kA1CR4{fuTGn?hXyz zg%~wg*kz0wuxCL%ESFgeth6aFGB)G@wG>=hz7rO@yC7fX%B{mp`9EiA?%HBw^(<1? zO2N;}ij&UpMnz}5H|C-hzxyn`(MI<$@*eUkW;b|6>({($L)eOO07 z6>&=0>>TIKF(0D%_u)%$x44C)hJ|qWp~IO!dWADRhwu?WLu~q+h;72XhHJs>H(~-u zFR|uHW@60q8=-vX25uA`6+DF?E?u&PD^IFeJ>MhnUxH_|WqGjf2k7kjjBhTB@dvZf9S-$qm9R=U~%K9BREY770#7S#QJPATr1YcQ&V-E zOUfhLXVvuPR0!SqBgQY5)Zwi`J}!Oug8iW#c!m*Jac;Vh75qhiA34G$cnl!dy;br6#}a>M;Alk(9X=8d%dA-F!$j~lATu${m|{oX9bDN|qLjo}wO`Sbvq zA$HI)2!V{hB5XFlq`|`;P_X<6p1GwM9w`sd?y7{_<$GX-|`Um zW>%m~W)1AGufu`-5WN22fhm3aAZNx2BD2?Y@!)^7^T9Km7;+s2RfTvZSqJ^k-f&yJ z1vRG{aV?{68{LD+pxoHS`i|k9NNc=>akpB}iYeRSb z4cN_kfqy!E8-Kgt?=l^X zQ5b|39l_{t>LPd}fJyN=3`p|C-j`!gu>T<5r@A5RNnadRbb_QW12{bvCwmEW}MVjDe;P~?c1X2?1Hhf|L|a7}6mx+YD>E6>k} z8hH)c=MoV#6l}ZJlzZ=dKu>7})OV+0fBaNz`EMB%*35_V!Hsz276C7ZK+GLC4T;Nx zaMoxGCZC&vJ_=Lu<&8VK9A1#_j|+%1lIPHAmtg)@8Sce8xW9G~)|w_k6jB23)d@IT zvk_j}n&6P($XhGPPd%?=`<+?P9XSc64Z;bG;S;cQ?hQ=SNycQKeMqh@#YTZPbyh1t ziD(eN9sUfD*PVcQ=XCx!`y2Lm-NQP|qZAjJK^C3sFjwG%R%;zaSZXEixK5?jvv$(c zx^?9By$B;mJwWo3W*mJg&O`G?v)znIoRpG5S*Pby@9Y?IdA*hnKi);YSVvN^MHIX# zlAJ9{sd4ryHrr^$(IaQGRrYG!pBo67_vyg4%jDx>6sT}(IgVrd~O_O*b{Oi+cZb1atLS%b&_ zj^Xo3FDw=E!H>qQW!1YIg_MFmy}3A=PHxVizAN_f#gB=6{?%>Ho-5=;UWy^XRu_5t z@pLOw0h`-WDKBaWN4G8EiVYc@b#EFETJOjXPwn_uZ3N8f>T$L?AH|6(d^tz(=dS*P zXuUfs&UGMN$dQ#_6o*~#Xgof+7-#dWZ9l97`Yeeki=MAF$l#N&xd#+ zgj8#l6gXsVx5)jsC!LcECG*Z3^l)A>CtVX+lh;ERs@XY@mVXVU5#Rg4P%VXm?)y^hyP4446pYcAQ{b?33!?vJVujWP z6#GQt-@zMrBL4#CEuLdqaWASWq`3b55S%F2!}x5%k@$3c-Z{WB%vh6-Wb?daS({?^adAs58*2&zoWD5IC)8g9wUtnyh z!A~Nl@G%<1KNo57q_;Cz?&%roZ-17wre35XMSmIw1I)4kYW!av3oUrT9BU7hw z=+gkUJr~1+_ZIPt&HqU1vm8We;&@RMPQ@?((3b)s6J(MDXTMm~e2hTB+40=4a|!o~ z*|1A~0Z=l|^IR8ge{&mqMyI2xs0No}VyPr; zAB|1lMdL{U%k}QjY1#2eoS}xhGc7Sv$g1h~RYyO;4-GgegZs8Al=P&G?x*IH_|#Rn z+!G8JyPX(*atlIZ$K%Dy1<)#31A4Xy2mX#lYEw7HguF+x-ypW`uY|qhYp^Zy3hd^M zLEfWC$cGAAjdV5sZOOvp%D*_d{u?CL8gbqVE57ya7FBGv!*oqa7!GWuxV%WZo0m+h z4~en(m!TYZ+JJqmjWAnW0b4zt;JRZAcJAMe6^#xEaZSajn8P^#vjDB)5t!-|3e^jC z_;*|n858@#WZMM1Fp9)&-3h23<%_lTJv2sJ4*RD_!^v_sI%iLYqKPp^E;}JGyQTEy za5*VhXk%J@5JK&u@UCk%XKfnG)$&0+`j-Tr_R)ZMjS((9mk@N~0IZzsi?PA?sls_Y zmM5=n#buOFtxZ64%O9zwtMkI^dO*?n;H0F0Z` zNMD^xDWu^tbxQSMVze^bPkDei{Q(%+CBl1SMeOg$quvL1NG|9xiC*tUbNWt%4%mb9 zQ6bp3b06d;9fDE0EM|68QjeJ)PHlWZLCc?0!Il7$YAT`1UTb8nv&3(KL%f}}1U62_ zm?#oB=3(0SF40Ci+*0XkUJ+gSypK}m#G>hx;904qBFaM#kH`dDal{GMspXtBV81TpOGQB zJ4h2(Vy9xuB2i=QQQ34~yt7)#Kss1T;>Qf~H+N zO{ug+^X)%0O?4>pC*GjP_F?2*c#`TR=hBGUHdt9Ians*}_^CJqf7N_p*)o*T*I*=@MI%lybT1+WbUBL_ z-nFBH1NA9E#TedOZ1LfpKK{HZgW0Z&Ft{24?_FbPmsAY-{0Jol#YJTQ;yU$hOQa_y z`zYc=8YySZp&JQ1p!9SBCR;`z;*C5Xl~!V_pHg@xGK|{N&f$mc&$eZ&B@60Qa+!+rC?J$s#ce!P4K*QHhq6EoU z3O=EP%fe^&xvvA=Q8v)&I}qmtO}*r3AkDmCPqi1~iJvTnm$e&3DPDok0dQb)y3n5< z$S)oJ;Hi_3!H&B4QFEQVFW#kqlYP;CRe^nmYw+mrbMPul0^c@uQb*beikqv#ca`s8 zeuFKSxQatK35mW5zYELfYpH6Q@SekFParZbprHXTX{?2?Hx7w`{=fme`HVj8 zvim?E&mR-!ll_npvJ>B3zF<^l5X^L*(S&J3al&*0x5(lEbw_LT-5yDf9rHfs*ji zY@#nO%n^`cFJ$U!sA`EBbUGs`W_TCP$UcDhMU(M(OcVVZa+<`~T&Kiu15hwAcGy~7>jlm(a{>V93K|Rp|!!BeOjCX}X`|M`4`kjaQ%zJe9 zbUO9R@}m>$WiY^V8ybS9AoxTwU5Q*qvl8}`SFWm{@e52r%Li&)c7q&kx@c>>42ti> z;=(9#7)BWDogLA=4}Nu!u|l| zzZyd~WKBq6_8|&8Fc_Y*2cow&U3BoGCJw0^!)1gV%)aR%WOf-|1unz(RgH8ZsFIGB z`H*kHPjVi#mS&Gth0;lPEM8>?D?vAYQ^&ZJzZy07o2c*QZ!}JG3oRahp1f}6;N_j; zXm~ju5w1s3?rr78wG=aB$}v>Rc^w_*K!=`Kg|^)r}B3bq2!moG0DP zok!AlLdovo1(8{y6d5!nQ`;av%1^&UgFXtmTfc3TB$Y~~-FrxGr8BwQGp36Rhmxqp zLzKDxhRDZ8jM`a@=OllCxL!LR$*(0X|7v}xDRN}Ij znztMAvGX20)N>WbDO?xb)3_tL>a0P%VRjThS%coFNTd5uo=8PHS(G<8UetcZg<3p` z*t$(rzQCVU{jF*8-d-|QEfl4GFrvqQJ4GYC6KUW63Q<7NLs7}Gmm-acouU%HDmu2k z3~@^{F!}o$tm-sn<#VANHfalw9$kq=U(djFiWG0lYofEK{)+AwsL|De`P88}644H+ zFbXsyRT)jX?HfoD+s9(>lG!jWItIN&V=S^O#p~d^2ySQ;wPt(~1-`x@>dG>r5{F-+ zTW=((Qt8ao%OTrlCkp=&LMI*_p-<6^D4;2Z3W}2HgR(j4wue#bka*s)bPDHwO6CvB z*&=zFa8avAwCKgEX3?IjA4FyEdqflev%zzZ07Qv}W9zTEsM=7Ay=R&+c#S&+DQQ#l z(*&~EE+T)AR_dOVPF{$lN%KPJK+-%amvfpzTPBKUt&Tp-<>DPyY-az z(hzoYvPBVB6GWy-Wn1&Dch3V-)ewMVFdnZD@forw2`UC38tZBPjEwb7k4o(e6YJrj_KBtVnwn zji$uc*F-D7bWz>OC?NxC53hiE)bzlDR-6^l{9K~HQadSJBA5>Bz9*Xh{Ge#GwHn!Z zmm=@%18fW)!g~{9srGmvxvr4I>iEl$x>ks%qtfA{B}MhIFGM<7u9R01A{v+(D0=bl zm1zDOWl~%)gzT0}P^LdYp(6yA^97DD>Z?e8(qgJsA55*zsz_Ws6|1M2qq)9L^ek7K zepMKdVwE#x=>}18Z!Yc5Or#f;8)^SK4cdFLhjyK)CWDR2C@PYrm$H)NRj*5yK8y3N$j+Y5aq&LjBccX)aDVny3#IEVV;-REV{zCRiI6GPDZ z?F&7%7U7o7aC}VrgYXGfTyOXf)>abuqA(k4_6ET8%vOr={Z8j9L#VIwGz^=ifa3Gk zsP7()g)j8bBjg`aI#MWC|1sr1?V|T=jCySwc-K{;(Itc0-fyLM871^*;{eRM{E}=& zNJ9IVIvUpZ(Cesf`u9~`@Bv=XI;Y1p{Nx?l;dq+9{Fg_`C+jF7tBVr8N(kK_r|{vv zko6U^)}3Bs`0ukteB`S$UlER2+VFNQ?l&xfQp0(=pqNMhB}wCb^FW9@6Z&*_(i5A( z*ckSbwA)4^U_>@v34LBYdWP)&+Kep*YxBkfCcF@H$!vx`{{8HWdFstne!>$QU&>&g zMI?S+&qBq$cvycuOp&`J(A@NomcQx|?rUvqm6(Hmqs@8GyB*wW63@fEgciFw{28fMAtCbQ~o!*Wn1%WyD$wkS56xiGjz?`XdSUXJ{4^Hf-qnZmy z;lI8(?IDeNtN+M*XMePB4dR!5R`8ZzJGkF58Fo3|j29|~e0H}A?&%ppb9aK^g|DNb z5sehoKbubdsH7XM8t6G=fR2-|$og`B+&(=5#$tEr&A(>KF7p&TG-XI>^}%P&`RKc1 z8Xm0@KA)u<)=qkY=}t1d#ZZZ5BZXeI=o=XMq7rtsq0q|?$9SVO+$_t)=D+(9|7Rnn zgzCWmj-KGf1z^j`Q#dajjCBU@u>9r<_!aEM*AzGG8F&Prhb+OJhKD#AolSKwci=_P z1gHwnk?jFn(c%(=$nM>^d7us(&-=iAxiO~gkb-@P1&(^9B4p-v`oT`6IUSmc@qa#w?L{f7mD2(w9Y7$6m z;vdO!SP^uVpP}{v{g83Fny%#T!Lg^#c)n~UHc#`V8Ely9;hO7kv%*~NJ~4*XD@rMCX$y^1tDI`u5o)kQNmcn+QBkr7? z1DWd*C`>et}*e`?t4lR^iRr%_AEPI|HHn4n(-lf;}&6r|21f7pSB*(lQw=L{(A z_QxE({z5;KkORGXg&wCUZ-TCr;gNi@rnA+5xHVc-1KwFf)qM&ra}V~!gA6%z!X zEg&kF>Q_6E!JP^8&mw|8R>+{|^A(ZqWze4&l_JGU*JxzC&_5JB5Oqh^Ls@zi3Nt)l z<#${3R(1txy*Vg~oVkyd#t`+YFCoR@j+E`RpR8v_()@@`0uQktV`}4(DR@ba6`B~j za~k&A55NXhJv!0)l`0kp`uxks)MtY@=9Kr4VvaT3r#QepN60B>9>JUqN6>QrC`3CR z;pUoPC~v(;`X;mSXrDTkJyj-)Ir22TVjvAXD-LI~M3T?bgiOE9^fasw+K0Rno`XiD zx2a5|^{ra;Penq|3tf?IbckA8uVJ9TOJCa+&t=96@Jvj?_w#nRRg#TUVrLQgYymDO z+QG3;8nSIqVaT0Plb@`tgxns zB<@zuqJQJYka>0hc?dpGuCP}OvKpU_#ia#4P77o9a!K-~|fSA*10IbHZ}PWIB^ z!)xiwSK%`)7%#lbFVf)5B=T>m6<8}ZifibkAfEw3Ze0ZL@hj0#a~K&jj|sh!8!>5Y zp`hu+pm0MN8T>d)pEcqs)%G6AZ=QjE3x)2lz3DVq(5F(W_fYtyV)A<}=#UE1sPWQ7 z*Rm&6e5;m5&iF;{9etphAdCFQXxfxgPZKQG(KZ7!Jh9q^gE>(&-sU)zGWu{!@(Nsu zxCD=!AUG(e5ikFrM^{J$_r>e=qeFm<}&m(`G5*jf4 zCQk0HAaB=Z{EDa-I&=Qgf|hsGxJnN0`JseZC8(eKj~;dS(8EPrNIb?E$|{K>^Y^Q% zrR)I(giBDB`74TBm`q0Bwo=-NOkw_hhG{X~RKGhGex9x5)MYJNa;AXJ31^X)L; zk4h*>ae$gnDD6@=J1RdeOm(|p?s)=Hj?W2!*cPPj8>9sfFs`hDyDzF(_tuIO1B)>iB{|6iLBHA(c}|T z$xHhoeHt52GsOZ$KlD~o+qrtwy>_5JnXXXWeh$%lwn8gt2d*RxX3>J5a2r3E`#BbNv|}>W@WVJ7qfO6h(ty`k6&3TXKs}!0B{(%>3t$G@-BQn(T9G z+C3JZhRgD&YFWXrOo#8?W{SVHShO`mgGL56(~__?G$*Bi_HebP;yQS4Zi`hAgs=O(T2HYSBJLU6hW?!LkX`u;EZNIt9_u1A3z7a52ib zxENZ8ZqxV3T(bEahM5!N=z7i%S}w3t;g25Efa0a-yIB!8DwNRB`)M*$1L4_KWGa>i{g!XcAQ(mjx~Hqt)7n zMMD-2B-J0cXiuk`=;G7!qO7rR$z!M$)&4A^?7M-q?2(FSxo(@NpZ-;xim4|D8-aJ7 z(T5|IKhyHI99Sj{!Jg8ca9UcjvH3p6S5|sD@kpBhN#fOgbc(#&|9MjDmZkDbS1M;R(oIg9ZW%GQ!svuCs0kQ zFiQ-cgZDMZL}y0L!>WgVi1bN;W9d`e6uQE0Hs7H5f4kt3>WV`~i7*}>fi}h2JiTNq zOIzA<)h!2BlRAd>1byzWyc#~GA^5oW0F+cB5q5biY81k8F<=CY#D}3SUKQb&?qY_} zll4LVDXR7UQ0P%tR0Nn}!-ZNZKr@L?`$xz89>JpIIXn(t#fig~koxL`4UNj^OR4Y= zD@M_xqlg{*450@SF;`9Cb$SXY->H!9&QZk72?1!?)P%V$m+?`Q00SZ8e>F&opKXwX zu5UZNIje?rt*`W{djzVQ_90tZ@K`54p^+N>u{K_cos3J6_IEpOtJI>T%?s7f zGo-~wsiNx&HR)DEZ)+UNk1fIkqu=Oj+()pWNAdGwBi2}3j(-z{{-)?SBtJ<*Pyaw< z=jdVTlve60auzzA{?hG3pXglDTly=^FS=EeaQ~SzVh4@FoXZoD__KpNO#5M-yP}XM z{6%^r>Z$L{aI%aJq1#sPu&=?CEiRqH;)t{Om6HO0q2n(^%?112oM3%w0;ZVQvcltB z+}X5;JN zwQtDYq|Ey~+9_~(Up)O*Lc8Dzjct}5I1O$2X_$WNJPuA4 zL!aEUCRjxaZt#9_2tq2mnnC40c{h1OLu!M@x)9LelHC0OUS(4d-9$#{wX4>dLeem?5EQj zlIRh$U zI~tCfm)G!R?ps)Fy9UeKB3`=W8LH=KvFpgWSjQr4LoS~7od<{H`RM001+6pJKD^iL}B5G*l16p4@PFScoIEGMM)CQAJ5WE@6)t zR=Hdy1u<)UI5QO!+Kh0-*n=`QjH9BWjdUSLlUGe2#Bu>&k*1W4jknL^%fjD??X<$` zqELaijD(bsPxxwAhO(~VP!m{&^bToQCuj@V`B*9oZKI*iLG-6^IF32KC&Rj7SSRa_ zNSTu`n9~8TOQ*>#!4`I-*FpKc3>Le6CV!J`8a=ib19WF`(eD9#NlS!#ujDXg0^?EI z8_FrFrfV5@>CE;4cp)Q+QCD5?a%ML9^?gId4eu$i-$zpDupx`=leBF8HUu@Kqi%x- z=AXDpwcpC=`{`QRyZ15eo+F3mu1NwrSxrAA6Y1OIGD`2S&K|yRk#zDB4DUUnbyqCm z^=t)>?N!Hp!_OqXdLt_KeME)OWvDf#8N-Cmv#Pa^5FY;t7c2wtu1AFNar!uxDanT< zYM>t1its!se)m<9wVYI0>p&68KcpgW=qc#-CUSs8s?fDw$omeb37rO~xFqU0YmHxp z&=h@`H7>!S*tM9FU=0nIRrot{IdZB!U{JaZVO#oHX@9HZ=!^BN_*&>2`6KX{dp{vW zaV>T)wLw?n5)6=khe^|ZK=QI2M}E7`YbT!K`K3)fJ>nX^xwj*Gvyf|?rjK<^3(-Eu z6Q@kK3FjjS#YltQ5b@R%juPvkFlh$P&}Lrt*j~_(_VDIH5iaf6j(Ht|uXHdM+M_1p z#hU|=Ip~a?JvIomT82GN5&XsI4Etmi@Z>8JRhvD`mTi}$V$;=GeFyk~GE7xm5N4R5 zzP50Qu*buB%h58o42rW_aq!0%1RF2pDaVKL{^mifI&=dr+zo_s9~V66QsyPj2K+;D zFfST0z)GW9V34mUTj_K~;p)yDNXsd+?97uwFPWh8EI0;}+|yhxmBLY}6@1ZW53KI` zBS7$nG}ZL2tWD*uq92$1x`05qN8sv@%!h5yaFArL$W1yCKle-L_@sG>mR%3_e zGR;ln^o%E*BqJr9Z_$lS+B)2I?k0QAE8`^l*KFNa=#05DlUt29a#-6jtnxpMNusmR zQjX^*M-%yi(2vr3P79mEx~Xp35STAr#`hGr^3r2-`Qch$e(JM_r9{3w;CLW!pPI!f zY4!Y9tCUM)3t9g9X>N^a#nQ3&@LXA()v~K_{^BbbnKWR2?no}^YsQJX)w{Cr^4JfEPm$)#c6-)JHRY?4_FQ<)0$W?0aBkjWyn3|@ z<)O<_xo{2c&TwG6bHaTh9m@C(k^y~Q2V&jrEVR^V>)qJ%!QgSZ!W z7(eG7Li40l{M*xj<|SA0Ax=2-Uh@NXc**e6%I~mPX~O}(X7GC}Cw`*h&UKUh*fz$E zXQb>y?%+s#pK%D@FNg3s{eJvF;I6if9Lt@XqgZKRI8Sn1&U!{^JSZcUUDY;V&Y}NN zbRPa#zi$+m%*YB+gqB1k_1x!rXpn{wNhB#DsWiV05wbE;iBeffDx#8+(LgFn8nP;t zHVs5lTEF}E2lRU7c|M=#zV7Rs^FA36y*C?kHVdfO0+jy4l8UwNz;WMc2y_m>=(GYo zaBo8~Dbh!Q~{UO1ay*L~4S(owatrY$ISC5z9BayC@hC}bxprIla64vpUwkeKI^Y=XN z!Fh_vUWrd7v5?l>1HFDL&fngQfvjM>dshQh2`MUEFq95;?W9dEyGgMmo^rw*aO==q zq?B9Y^@(p#X%Z^C2B!5TXnQau25lpJL7Sq@~X?o!L9Y49huR%2qlLGd_SvnamogXojXM7sf zz9Uy-C+1S8Tlt|x$iUxlO{@%Wmhx4 zj?$xdHjC-od|Mj$<4cP+uf|>7$v8lf_;KkjjZ795mK+?Qd#lEhdXFSkrq7|{-{#_J z0?!ZK-GcA>UvZ{7A66e8VQ{z)R9`HD*BlG{3~+_@&E@!1a{{_;39!`6fNggOWFCHi zcXKu72mFFbrXMv0Z6~Rr`{-JL7AhytLh8sUY$|s~-|%s0+p-rUcy?=hrz%y(X;Mn* zHAq{uVZ@+vRJ39t3?JzEznUFC#yr(VCh$k#p&tL+Bt@9 zdwqxadrdN#_6oHbBDA(no9x93G2mYYPvbs#bzX=1r)n&k%rnFyO}K59h4JqjG2u-H z>MWyCHEJUQYjz^EYCY#b9f42WTAVummDRTX<~!J5?6i~uZRk`a*-NUVGS7g7x3uWd z)sZATF%|}~`mp`2h2`HLBH5z^-aOCaGx0Huxc|~eLxX&>N7BcWD&#ZuJ>MCWq4OHg zPbJ*M_O^rYD_w&BVh-YA-F5hyABM-#qmZn~#PZ~7^qzl)566UMq2D37XAq5-YlD9KCMtB?O9}tBl2;e! z&Wt0hyto>z=Y1gZ&mPVLsW|H_jaSo*v1ywfK5UU81dEf{=TErJ&(J44H+jGRJZxuu zn@CuuwzNAV6;H1_H|D&>6}10AJ3Pl)EH8DC11!nz zW^@;G6DrX0vm73=oaM40jB*@f=zB*pRIMU0sWln@)n!BHzyX}Ehy!z|M%>pS6lz+J z8DZl|Mtvf+UAE;}cwekEcg1M#1s&%)T`X+Do-|QPdiWVER+cUt`+Jg72q7sBoYZ zeKSg6E8U29e>tyl`%`GoDZ{3fap;mR$Ii1!$nn*r>04%!#MZGC71+!7)u||yi^tm5 zT4)Cx#)yaI_{Q@UcPC%MNXIkKf1iU9?ne;c6p7-dc@!k(PB-l4lC{HXs8&xwa~+4Q z4A-aAD&{0NemdFL4#$L}u1MJFj2k~fXzld9v|r1Sq>Fb@!J;&>y}ga(Ukd1VnGFRF z_)u>bquAps>6@h$`MQ7-ljcxR*+eoGbs@LJ@zi~976rJ<($n{nRJ!9gvTq$nSi>>Y zK8(OWll@S;QjS64N%)?33QrRD;q%`mIQ{z^ehfQ+)IsYob5{a>H58)uOf(EK!|_0M zHNK}>qx{||I#s!ncD{F^FBR+gj2J;;;_(#z%at}43FvRzV(RJa!iYMa)fBshGj{x) zsxYG6qXblc*$mDb6QL-#28Y5b86C>o^V*~Z(@EY+i=lo9lbGk)$oYRC1@x>U1D zo6?QIT@SUTAzxgndXo<6%4kx|jX+v6s}tXj zHDC`LLc;q?IUn8{DGOI2QnMV@J02lP<1NM=&%%o5QFzdM7a21HY2*qgn&lrrDox=K zKKIAT2MH)skfwdFWa(A1HkIW&(fBG8%FA0!VfK#^bfXXx*EQg2zBQ4%IqB)IrzH!v zQu;&(GAdh2>y_T(?2M1-`Phj@EB?HBijsCsI}VGZWclW~E zjx*NgYS4y=xwL!3FdTN9g{(?bEI$>Ap!Xj9F2&DTjR>3)3&oS#a2z%9AiMJ!WEme# z={q!OPNx<1Ty~?ck`u7bVjRzf7+{OTS?qdu5;OcSBBV!|%IvgB=e-WumkFS9zn?XX zAgoX8LVL&%Qq%a2fFsj!C}J%do<$-1#vY20*+ZgZGe~@RIGMdZL?4zNpqh6ElyTFL zGP1l$=5Q&kaLFK*z14Kib|~4lDU);f7%I4O0JFj~@raih`StT&zqfyIKS+#BJ^K;7 zq#GG$CCUH95xU&qLducnDKc>bjE0TFlXbH(cc?gdRw>i@9U5eql7{Isc2eic9NbAg zO|i@OP}JubT5)I${Hupykd_htlpLX&oa3bYHH}6+Kaa|jS25yn4ZpK{(8tg)5;?(n zH+$6Rv*;umw@g6ic3h@mF=hOIeUIKR6XL_p$q1aL22+Q>I2!!~OOMFXn4Op4rk#wY z`YW6TIRQ~S{o#=>0p*3p^rlIf_8J@U4z(Ke%`64=JV&vk4aK=mp)j=-<%qat-y@hk$rkAK~k3=JtI(&&?-rO(vq_u3)& zPpA)W4Z#gVNxc0y7_U|^n6Eko(i0LrgU5Z)TS_!E}qb!;g7}X+^3bCPcMmqsZ$Ze zJ3gPuH={i^2d(wXNI_&5d8}JZ6>qL#apWV!NamtRV;AZg4sqsOI^Ovtle*Rk%9s>O zqP7Z{Ico;isOsX_hX$-FJc%t=3UEbYDBYUy1p^;HA+tUeKMV>HJ1h^P%LgIb#SF$> zV{v@n7*cUiCq;hlYzTYBOg6k@I*V)Atju57;P(Q$^bLZwVc7RzFRq2gW|} z!p|mBh|2*q=*&lS(=o(N`GJ}nHSp~lL{5KgX`Qw?&pa=uvO9B0-OPv5mT)G~%D)JI zG>mGe=#t`Tb6gGP-t$mdI$EL2B6n!8=-z7Pdigg~UChwob&mH*x>7*oDr8^EqVqwA z+31%F%=)}OUau8kOp+-&hwOu%x)I!CG}AiUlW`>9k!e?w1&L zuRh7_+uyLp0zEA1D`Jn6<&d@+97gBaBSUQ+ zCQ2U`v<*HkaB@1p>P4@x&{j!^RE1)#`*?^Ps)MF!C9XC5(~0)WWRQ?S@;=|$bT3s* ziEf3_9C3)&YN49*&m0=U+4~`hY^J;+P8-U@BiaJPx159eKMk0TP)CeL0?$t*v!J>l z_PzW(>SbJ6-Gpwz+jFnk_C=1Yy|Rye&6B|B>p!9MdJP$8-(vphz3fO$ID(o#Bi!|w z;QNN_g3i#HtZGsY+hSA8w%=HVt4~idu|<+ljvfl>mAXvu<(0s!n4!hLQ?PWqIP=h& zfT?$b+3dfZ88f+wYt=Ud_CBI)*V%a1r&GjQY^s=u{6x%l*}$@r+*peCV7$D!l?9)y zVn>d9vHeRH!lF+LA6&DrVR(l?+D?+ih<;@g$|{+~ktn>kuE7t>G(qai4nd~;dp7o( z9Nt*>F$w=|(BPW#2+zkjQh6PUtBTo>Wqqu$un5;aOvESa%~<#)0ry%oAkwdhRhtaB zCr=H_-WlS_`mNij7;#&*X)VBSuRg8bATL4leLqXUnan+?xQ9O;3KwJD0e zC?Wg&LpbRTuzb%cc+Ynx?yuw6fq^o%!dDFsmesMD+&A&_a3)6Ob_gz6NwNNI$C>H- zRJi7SLwEiIELiIV`9(&!ZOj=%VrQBDD@m>y%;kLMD}t46R|WSQGnu-XAN%lA3DNv_ z?q1g-w)BgT-Ds(2=XbwlnOqlM5;7X*Z!Rzw1%1S-b5^y;W~P*!&CbT?;PL{-8pDpT zE!8p1Id>@b#KfR%S}=^thOkSrO@b<4Q)W11I*gM{Fg`s4T37b7myY7N7d-)OUth7m zJu@*fwFbYw-(oF`ZPCloe7VR0>U7@~u)RZgf0 z(qqT9giJPF8H=}tV(v0$BviN~#qT>)+jWy&d{EDZXZSJO&2<7ljSjZH<218OGJ`wM z)L3Zqvm^2WyUjCV;}*rT$ixsfYtA2L`ezu#ch6&o9saOX^>YHNWp#p$Q@;tOZ(o6B zyW(Nx{hRkG2Mb0;gfqP%pIGU0E9M*M%if<8vM(xX(D{3ijY^Z@%!R`&@Y@`yJy#^1 zuu7KpeG^kVd71?;nupd8n=v{%9OtU?1-|EU1xrJGnZjj#+>J1XNSGmf{G}nyc>*2g zo$QiBnLx!kSs+rfnn|2I!;~FN+3dtcEM=P*3;5+H(AWGaaO$^Wi|4LmbDm{kx${r( zmQ~!e8BY_8+}Vn?!=d7O9LJ((F+c8qj?&V=zYTBLjg0fSWE+p6#)t6n@pu$gN5l2c zQ9(e*YqoV!42o8}KrXT#hdL$c*M!>wHNQ5&G~R`@Bj*g;G&zWgI)!l^oOf0&$YsaN z)}uUg7)xo35L|NZ7HH0WA*jD-&s?%?u;;lJVmQm|bFw~X13Kc;`H9F`xBy9F_N;lp zfH_7LFsBSXEM5}Jy^@zO&3Psp_`p~X_gR-F7@{ii0H2l3sak9j^Y_+bVUBCq>ODrR zTm6RM_TNDEOUWIxlc&SZs2@L#=dgoDYH+Xog|BB_nfih8Y?Ejj+j#K{JJ@Ui{enPT z5$D;fpbS>zaTqRBi&<4|4RaH{#~N?##*iX!)U1k!!h)NE?fX};2mL%N@o5g5$iHjp zD<-gdza7`Jjd02|6S~6N?3HvP6AaSDWD7~?ZP|hp{|;DgPGy?Mr?9TXX^;zjBq)0_ zlo@g+>+H&^})37{`gR6|Kdr>r^v)aQ22^sdXW%54wWxYxZc`)Xs{e2`zih zu(_Vv$k}xWrwo*McY`#7LRaGIm-+Az?_>cB`5uo(;DP&D=Gny0$#z**IirOAU1-R5 zy&8-`1~u&LhCS@znPDtdHj!-X;IE*Hcfn~_++hogMDexN7#H?gv(?l8 z3GBz7W}lZ0!Jn9^uo<@&l|LV|g}-mH{de^+X1FCQ%=cx_lH!=maNav1n9o+3jK!NN zJ0a!S!bIY@$92JQ^e_6t?oZ{Ky8UMs8v0AnpzO>hzo=qnlYN*}#YlAej$;2vlD+>D z&UUBF79_UZWuYTCBPBM7%}!8f>8)XG%#Pda#^V9@_joY!zH6`!@#9Q(aVZ-TWWu6# zxfdxr1`V5bvT&}QmbZjsqOUC*XR3?|bC2Sr(`UiEwS(E#Kdaf!uo}U^Pl_zG&UsS7i`#7IY}3st|IDcR^cyQ*c0_xwGz zIduZds#M8f&|99Dw!DJzfmJJ~-yXACAG?&kmhme}ug-q2!WNu_dH+ET)LH8{5hMh&8a~0g~yn^=z zb$a&Gm_|x`!1AMWX{5lDB;zxYZm3PWc4*PZlm-~6ok5OzJMJ$jhd~$TcC2`bD1R}k z8?l=1xx~|0qYhk68c#>QxR6DIfM;Y2Vav1E^Ek(JYYgWIR+&@tK4tRFFei!72=23; zL`%G<(d7sFlp^#+vgUP&RDMI@R2fn|ZAz;Dg;HwIafDoTe>I&dDdBY>nBdEV+lRj5y9C{lW0t=1*xfQ zC6AM*c%O_6CDApsblgNyK{zc+-A0>i!{G05AEn)Y@ZX1Put|%cO4(SNb!`K*l1^g$ zZV8f_l8c=dDzvS(4dZ6sh26=C)U$IP)!b{r*2>gW*kgVC)>e} zwB~LcIk(;CbE+JT+dY-;Dz&l|Q@!{z5(E939gz9M`QrjF+MqHH3wa0aDU%@1b&TQq zRTGx%7(yQEi=dSF632U*;k@H8b@DEO-w6qnA9WQbyyLq{hW8}jJB&My7jQH2JuK3e z!NcDT_FXv`yE7V|uVtvWeIy+UKLO2eukeNIfa`B+Lvw2}=IS)!7w>*L`1cPsEbBvF zkOqG9J<`+-#i%)5gavcDG0s_ynq3X(Z~k=pJH&;4KO0YCqsLK++h$s7%I}|!;`HUx z7>eC^3Qeb4k^Ja2w#b}?j>a{Za_!}>s01lxzlUPncVzMGdXwrlG|xK;UzN|Sbb0|k zZaa??r^8?_(}TUk24K}4jx`Do(N(WQ3%q6`S1JKHS1;mC=51Eyx(6%vrX%6`P%_vP zLhiRCsAWSJ=6^LJ^R!snEza+E{X;2O?Y-$cgtQ4pz(!rkp|xW3{KYMYhFHGLCFbOqz_nU48Of3{!1I*-~TNaW2Ld&M*%5 z@x=bZctp5&WAFG9bOXTl`#ksm(vmK(T+Fw>e1~YPNgrNBjVNpw4NzJ)4k^~ z&=|~jIu`)FNo|TVPJ-=|798yrCC!XHwtihPlNu1><9AmaHxEKxF@Mf#t;yqpDS0pT zq4=NCEd17aI5~cX&d8%k(#^+>6a4QIy5ir9C`f+K#@Xapj2FFu?KLgvdMZvKGv(<` z>tyoaIf?cd&SCj;8|Hj2YOLA;y^w7DOx+IKm9FT{YQ^Z8t&H6Y!lI4$14_ zrMDBjlicyW=O|2lyOCnL9lyt1fWN^xyp`FDi{p=@zd8r%%oL^1OhJ|b@OX6{eLj!) z|D%C>qV~gKcsZ`xzedNPZ5Wz;4_l7szLSXJGU^A`gwgYFaD3Uu^u>3>rdf)2P=}-G z`f)sZ{|0GSV^Aj*hFf+~h_cJT^N4%+Dlb8s|L|_L>?m|QT*l_!GgzIK1yk+QFqsg; z0^Ayyx7K#p)#`CZRycf`1IcIW8jRTb02??rHSEkF8X2!ai(OaK=rOZVKP3xEyzAib z&9*YM{DqA4zl&k571 zYB2xYF2_TCTpMSs8Q`wjC3ed07#41+qIEF=)KpYXy-rK8yVVa5?|#By58h#ADUG6h zE!2yOz`p1_b}YM%_|H=OogIy+oQZfZ%QGCajWAc`0YCq0*)QH7w)S){oVafBt4SB( z$?9}Y))+g>7|IJfu{zouq0A2_hLvLD6g3$C9f_Gaeu$LtL42keBqWc)Z{1!P{^K1j zD{{GJ{fh;g&4BN@7W`Q>5m)3^;1%ypcU`iLoolTFXnS*ENR^in85>zRy$C`)hF=Np&nC-|z z^PeK@=>Cj;&p!5Z%T0D|L=c=ld$UzjhT-<?b3xPIa{dvrY;)BNXSs@gct;7NdO zNj2jmJqD*v#{MxcS*iF)YB+AIAI3xDSU9v@!?J^c_-FqRqAw~D zaNscfuI|CuWpXs}AJ^f-3h}9@9JK>qn3v8{m;Zat>!yyH~hD(4XclD#k8k~;hu5_L2G$;Z=VB< zJCjitrh*9pythQY8f}Y5BYLheDgw4a&s-N$kqzvSIR8Gz`Qz2+Qds;uffPqsyp84@ zjKu+{c$@|6CTa4!{tj!;z!LCeC!SlVuanOkCcKipjiR;^;WlQN{wKbW{+DW?v>*=hFDJ}kT zhrZ>UB*ni)Wcogqcc@>b{XAP-=5~WTA3mdy6%xXIjl~ovoIKy6DnYW*ckm*k1|5Ia&;svKG<=vPv3+{NUrP;y z|0Zh*`>dTQ=CKPc^-LuD7*S!y_V4trvzG#t6@)SB!-TsBD|5oMEcxACBBrOd%NG1!T=y*jb<@nyCZ8MuFWL+UWii)Pt7x7d(s1sQ;`Ait!i_%0Hp?~rV zTH>WEoEG#QU*_>1p@Md}dEccXi&9W0~DV=NFU4TNkqKBl1+|ny!=vweaVuD2jJstXWQ2K8-&5gA^^sQ}*aw zS|1)lpI7go!f|QTG;bdb+zg=!BNtK5JS&plOBDHK7p;B~N()MC=+b(3diFb*`qw9t z(Y{6!=`5p{%PQ&BYy*0`&w{FJ>?w4Y2W{W(NPnY4==w-&s?*e^Lk&}DcJUp!t!ahB zN$!D~Of)P+g1)qkBJHnEblWU|4ASCAKXNwxx-*4-`&v?k^FbO_<4cM|7E_4TB-%bf zgmNw@Qf`kDC6w__aRqH!s$WE-B%jku-x_+9no9YHvZ;vP(2395^tjBOCck!}?ynEX zspAagNEOoAze7pv?NEx`B|-~Cc|S{~I6Yqd30u!>AWyYDWU^)xrJV7gjFD0FwkDc} zPqZet^Kw+Bv4Sd0GHFHJ0m=`EC%wz#g$i@3$?J)<(84{3mP(!=#fN8T&<@VzQho%N z)3K?qHN1ybRgK>66(QRQ zaSVx=fV0D$(A#o`cSWB;(${j7oqR+^pPQ)pT`hSf^8K#7Jr?tx)P!sGcu;*8$A>+| zoJgJzYPyS5$7k^1cSiA8N4SLAU}WrG+BiRg9(|3ZLAw{=!gW=QNZE^b8+YJc_B6!H z&4SOsS!@x?(y_xz5P*)ioYUeVPQmvdLiwf+yjwQ% zPCsXa%C}hgV*FzS&&~v);dm3%JU29}AFw!v{yE%|d{J6m*YI;%A%+vcFu!?LCdyxm1(B zHLFp|^AhM!M}AIMC~r{{DPOpTsXw0~Sg{(XeYBuE)sV9h=0bnx9+*fS$F=qKyt|QS+!yQN z>a`;GG}t$ZdSC z&&9UqMQBocj!3C|Y!A4DU!(5fWNJFx{ZF7r@dFwrMHncrxd(IKD3Ntm9M%_}huyN_lodW2j~+WA+ba|*CRzv|=7Epz zq7W-L7PVux;)7cu^44BM?$tX`m%allnO>M}RwM1l|ByCwI&K#!VXvJjeC5(GA~*+c zUR^}X?DZIbC7pEoNZ(9{T{zAnbb?&Szs=82m%2N9Rc*}hym zK4mI|_3%`<&)*G!btihOFGH^I7a}>Mv}>*npHm~TEc zxjxjk1!b>OvD7mI8+LM^V~Qu4b$Zi+`!=K>>`Bc*OX$u53v!QdMWJ2;jJ@+=ulkjm zS1SpP-u|KA{=dmN?GrU7i3%G$3@F`;?*wj-C6BFQwAW(_%@G%;qi$K~yn6@(vqNxX zWFK9)-%X!H#e`=s4yVM;T4eWChT;_uVq9A)hR)cBJmy918g{(@ZU&v|>!xM9|5DO6 zG2wqr>&Vo1B6&27qu%UQH1Ev zBr{)+R*g8uvxGdWW3&WyZ;s&di$JKq%*1ZRbnaigfclN;h|_vUsjr`sOU(e?Dqk+V zwQYt_>aDrZb7wL|CMVNq&zto6*$Nt)Ye>HC_I#Igk-{7sNTT*UIaOOwciR-&cz!lH z9=T8JPY)8F{nSY-e=7)oTUV2C>@#u<5O&>@_=K(!( z{zoV4Y)PJTUp0G8s72*HF0N|EQOQ?ueceSyzsjk7ekr*&oI~ri9Js9U#zxu8)Tf?D zQVAvGdG`sahF_xd!LP|xPe@lS}BqiHNjFUjyTYpZYf98Y(m?`PgadgxWOzAZ(d-57dwU=CT&Co*_^fIdlu z(}@HZsvFoyD=aL@BG8ojD@5plcn?BjH3{{{>705fNti~^Hj@k-7@358-m@{SHkx`j zt)qLl9BA|ATnyK0P!bXTBfQpPmfY{na?kcc?jEKX5Ji8clPoqA@jX zbm)CMY45#433ES^=Mr;a*NHCJ0Ih8tOY7!4QF%f?LQa^`u4M)kGUX9$5=xVnCL8)=ONy9Q$c$1|w4uY;CqG22kM9Y@-|aY--)zh5O| zWeL|I{iW&F$X$5g<$>*;!{Bx-3Llv<>cwL4a_>%j_pn7s<9w{E(8Ad)ZS3RR<`-*q zc=xb5t;i@wl@;$P(X@qKj4@1-rSSHqA!kM2WThFY=vUX-*!hf(mjB{Xr{9)g@UZea|6UE{RSh@Bs zbAP0VidVq-E3&*(ZU*Lw2$8k?7mFVvz`b2+C|)=nc}B{VWp9d-g(W=yIEfZ)lOuKB zzo&es6)sw5@X7liCNAZ>j3Q;e*PnuqcMf8C(g5GhaP~;eJj~i9hr-O^D3}DU({lFR z$az@rBhKgRQbdf+#?{tEFjvcA=9-=C;?odZ2cS2qpAA>f#Ednw5IxcnlLo?YKmRui zHj+S6$Zlv_M>CV%nG7TA(Yx{wo_u%?)y39$bBk-gk7TiI@;xL)UFM&CD5gD#BpL$44cx2?z9GG*ulKVgnD zha+jBHgsZBSi_)b?rqja&y{g-doO`c--aW&=NYq?GQ&E#URDeRd<=}iq326bJpKtY z2}!`$X$w%;+|P=?jH8Spzme>%KsS>OaO0~Ex-Fa8py&GN(QRUe6GPeMT~8t1FbUpA z*Wk;`SUC4=!nk%v44-X}ZFY4`a^rng8MK4*`JS`e`%BofzCAd3mvhlouR`p76fW2^ zygRiLJJYgZd^7}%Yf>u?TZ8gOK%??8yLNa7j3)#`@y}Y^)C$1seIwAVA^s+41iVOk6M^up3Wa6 zw_JwS!h;aMc!xyidG5%~Np(uS4g1*vMwlwS-URZ2JwA6Zh9h`v^|K{TK zy(ruicyTU$4g~WeAdzT~6J&zhYuCZIEe2A9Ww7v+1`-dr6bPzg33K9Pn7j*Z4h_HZXdBT(@ zlSaHAnfOnqf8)o|2laVWl`w&7^pYX26p4<$JWSyG51aZ|v|Q#oRmTq@)x6=fFH)8C z=sb>T9>nyBy|~Nuzkmd3vNidG^A}slA*h+6yei4lR!GsjlTXLUkj|(IC`m+v&n08% zkH;e1H$9BU`Z0L@?m6D{wm_f%+sV~O=*6AA^d$TkRW=`iV@M1lhx=ja$OP0*y^3Gf z=ds267}iu2Atmqxh9qXux7nH8$D2ncQB#DAa%6-vkE;vkZgQuL0iHqp?nKkN&B>x~ zIxR_CNXtK8rkKDSQr5XZ7chZ7yqQl*Zr1esPbvA0e@0g2*J#(xGIGp4O+&NNX?1@L znC($)IJ*Id?LOdiaRCmhec*GHkaRzQwwO+)_@G1hOT}=PyM!|d7WCt>CdIEZqd!Iy zh3cPY3kSVY5ngS&i7g5}==bYKrZl7DM;6h+m1cy0QbO%p|H!GXoihDpDcV$?3KyGE z$21M0i<_G8QkR^td*&B(B~~MR?K4bpEy9b!RBSJfMB*J1PTC+XY|tDotd+{8O4k~` z)5;)SGk5Zku%*bB-Te&`_+I~h?q#g0dP#yy zw@Bf8F2yb=LD}sz+<%^mz59;S`>0EF%s!QNdY>bKYz95cJV_@dw1frE2MOO@RTJL0 zSV2b*<xGIISmk3OFnMg5wFr0lIsr8mUrx%p5!`o^5y4zuLdc}NK=M{IK|ACpyi5OSJg?y z9Em2(5Q@?{u6x(~CqdR}x;!&#LC35nkfI-dRwkR%y6lzI@3)P#`gvAvTqx1vhtQ1W z`|QlyXt+O`j{F@#HcJi2_*fGC*4;;szZ@WQ=|o!kmd{Lc{Ym@zaC&w|loWrMklUw3 za(EI=ulx2>PEj&4LT;c>n1WrX|+n)eZ}ajWVPH4SnKn zal9$2f#}&fthyP^cQJ=)-syw%_3j6n9@IeTr|;7+nJkjcEv39;*Qn)G0<2P!v1V=z zMEW+Pm7k;CjnVkfu$5~dUg#;|p02FPRQpkrGAwQBTgXpbQIepMk%LL1PlF7uXw%5H zarELX?;hQq4Qt0l_$`;Gu`@WwfQi$(1rZe4pGe6fF|;s{|E_FNqpidFEUR`GN_^*J zx90=)JD1bSpI7L`%1abe@R00Q*3takH^|aXR=DA17tL<^Lw0>n+5Q$8Y_yX=-Sr;= zqZ8WfmCSc`kTTgh$0+tr)sNlNwSj1NJaPj~@JOi7tQHotq~=4&v)_isU*f1c*UQG7 zOJv^?f3kuJi7dk^oqgC)!gf7YXW0*BFf4Tsw?DsP-=3VVaCS# zEGPRedm`2(=sdj`dKKF^qw_lZcu)mmldV`G=M4Sf9>QU%fvoqWBimuN5>ASoC)f1^ zBTsnIrzQWGTzU#yU89O3Cl^-MdY#Q`YhW9>|Ex0TG#k+4XEsw}-8GW9KJo=?GQEP~ z+`F-Nd=mM$cL*l0cVpoZL2N}#Ae%At4m&brE%S=iW>*tL*^g;)xV>yEPF=Z=vV}Z5 z(L9+|#ExT|WBD#KhWohY$wL0+dX(sDv5R&p5PrDG3Tj5P+;ze14d-pYNJ(H@1AUlh z|8>@>uf^^dk7m)*`&pQ~B#T(MpB*X9Vb#y)vD?@CSn$zX%;{4dn`*F^`8-xdhq^Z2 z+9}~*_fd>m_)efF*DR>IK9yaxt`#I{bB>q!F?Kz-lAXGo!4i~5;*CcDE3rPo-f$Li zmr59G_pB0VCtIOZ{wOo86kT8_wg!A-%hamECP*drf%`JJkXb`#YJ+ z+z#fT?!bm^{38$@EH|&7 zO_~wL%9Tx-QNTVn)^j-8#qY2uCl3oG*Tk_^@0MU^b1g%}S2o4-AcRwHGcoQFw%g6S zQYs`_<{@PkFiOB?YL>9!TuatopNyZPi`ot@R4%7~q9g-b&gv7euU*@yJotgdS#=J_JC8 z_Ch9`{hvGLP^pG{vw|Vnx{A%IaAfZu^8T0KAxu@{65BNHAM?{3g5lnx<+%>qS7b)_iZ>OZ<3_3ZakkH&3n}E zTfslGpRJu)&2B7S2%qI&^XN0|avWWVkh3r;OjH#G` zft>B!ha!VBZZ6O>almRZFWgjEgW95ae7GEqzud?0{pUFhGGBv$?GsU;xCK719C#=8 zCl=Jd1y0x1aDnWQadHgA_{;C_l9w_3LyX z#=obvXD34QwISADSc5)CFO$Bi5vbo+fl5XG?6Ihb1J8y+(Cq7b=2rw zgAUdhN8*B9G^QWYr2Z`ylqf!)S{k_)p;m&Fp~J|3X9Zhd?#9-5DI-1d76$T*anouf z?Mu<#P7W;FJxn*@-lo5aKrnSe7wJK7Jn=+ zLh@||wsSqEzt0n&c+XsJZ6Hn_;Qi=U-ssBM4d2abXv!19S;+U#pN=Em&K6${IHxaP z3D=AmK1Pp2PlglfdQzabcpH-5K46nIZFv4SmiO0R!;+I@p{FziALohU!OvB=FTWPX znmP!njzgX5DdeqAN7R|W$oUqI$uZS<^e+>>8Mok(>4V0`BM5a)!Rk{H_>iqk@0J^3 z(wDKAc5gHi#w|dAO&r`rXFzV%082f6lbul-j;d)f5Sw-g(-K0VtJ%tKEwsRs6iIx& z5{>f>-Iy3Mls;Y`Lej@d@Za;FkX33#c2X+$B3;GsSO53UtHZEP0T-6dMV77sW^G&w zmoz!FAOFTQv~}QU=Z1Us=9n>k5_96ZLGMIe-;4C)mDo2Z2&Un~@cV)S9?H#ume*-$iLQm? zMP*u*8;=y8>)SZzE@CeYBa3I!Wc1vVvRmGxeeFy->A_ysBW7SwutU1yT2W;Zgb;O^ag{T#sZvCg5(Id47Z@v&2HjgjpX0FOVVfTeLc!+EM zE9B){mQ;LDCvwSMXnr&r`wT^gQeV;U@G*^r8!K5Ca)29MteICbh}%s9xZ~$F-2N~E zo@Ig^CAz^KgPoY(c$&Ar3XjFS$#f9n@2;h>C|(-{LmMxos1)LWZ71G&Wy?pm;&G?+ z98Sk(!`9>*B2PVns_1`vqijGQ!(O~LMwga%SalKFmeqXCv;PX2M<1dNStM790$m4wr^7$oDa#Uvx*N zkLbzF5uaf-eHFe*;Dui8V>|&6lX> zBS>`>HVV&ZcB~8MsjcR?kV2?WAB}&8$FcX{S;6q$i9>Dn;X|bh8%oXD=hQDK&OV30 zdlgu)K!x$cwovV4BE1gpX6nB~$ZC3u51Za%fz?*%SD(h!>CdsLHd0#IoFNT(TrM3H zx5GU@(`fFtkN>UC!-`eL3|%sr1+$+@^ET*U=HK;#-K~wsl?E8NaU{w-#!H*40;F|g z5`|y>FfVJDbKvl7c5WYymusS-s#$@GTMwmk(;rJeT`!34u_CztIfEni?Re&88b4l2 z=9mLH-1%NXbmg?c`fU?YFgys^b9158=`@^r_mM1?2uAX-c~a)FO=2FNj+XHKxGnba zZ$V-vZeEC+-+N2zi}a+SuO>>{PK8SokH<)3qjyR%N|DmOtC^DW?OG{($#`~jo4}w4 zNpw{&;?g1IbbE1ut~0E-QN^4YqVVCI|7$3Zz6PhBKM^GhV%Ymg)?M+VmPbC;IG3WN zYb&m(81aRX8?AP1Y4TsSnIiTy&EJ5eSDu3IGSTed;+`*lugcU^|2>++Z^=K9h8 zI&7TUj;%GlrF8>*C7<+(Qo*M(_^zmd?%eyh>Rm#;&ZV@Qk;d@u`TDEUP7>cR$1(?X09{ zHb7FZ+bHhIIg-z&h5UFoj`PYYSUh7A%`+qEF?Ku0d>3p-4`G;u(R_!l(FF7K6~u!t=50H5MzG;E+m+;9v_MaB-v*mlP`v7x&tx zTUzkn`B%!{XoIz15~QJXH%nh;*GQf_$4i$LJ*7XZ2TPaxtwd4mL7X?OK)aVZIM-$t zYW9?1fr|k({d-XFN@uFx^5q!8k=UZ@#gyPo$)kIvbV#NoxQh=Fvh^W$y1hVDel~6d zCnIP3ZR{SGhQFpiVc+o;_T-+Ijx~sW$E)TTVRTXSFV$gJ)=$yfxlgLzULlqDyeKIj zuf&8UjhNx4!kP9)xTtRr2jT5dTssxh&uqqtzefanyF$__&5}}_{z=;X%xJmZg{ie8 zDfeJ6zS<PP_hN(Ybn*)W$wW zI@D{OG;P{{c;u_ZpX0RYninoOep@81uQ`&W_6%B|nvptA^y3Z~DXnwdASLz7k=h^J zh2o#t;@vKU%|1{3ryq&@pYJhLcy6vn7DN8dJ~)&GNXa_urL(3*(%E`dj&-r8npIa8 z7GK82r^>wj#GRo*^XSlRBjYY-@$sEJ-s+IcQ+LkN*rx+mFOtzW#)>1ACo`^m8Ebpb zp|1TgDh9T)^~q%p(ml^vivNkVx{jM(#)u5ZOnyua;G!?Cth+OkXGYH#oC7)Lb*vV5 zP*a|MQ9;#FSGeWZY5tdMK*KL%#q3^(6;Vp;b>Jy%m3|^1B!~ghHrj4~M3vdYxj86{ zGb4nLb)5!Zd{O0IcYS7t%;N0eOx|%UoTJ?$gDX?%oL#TU`VVnwS|Z4N6@=kik>xKqcUvn>~}O-=+e`n{)z z`ahOUyi57>-9*2P86%Rr@$m%%;qU8C=O@;jp1zfv^AB_K^-Km8lu%doI5mQbY5F;x zo%f!nsedU07R1ovRTQuLg>mkMYBYA-2c^&&p=bWUwuA08h;yKeUpM+!7<1HSBW680 ziuI2TXm4jkCA)7hv;2ykH*6Rc*u>uX8nTR=+Oi*hV$GMYq>Wk`_c(Q>>1tEX*wdZ^ z+wG)F_IU<j9S;$l3i!>Ro97vl_q-T#dC9IVJcm{togP3G6qM-vqO_w$S)z5uf;PV?;ynHb!7i;j#Lk^`lJXG*js zL&L}Np_PmaC!E3Aq(;;PpMl(kd=@Un$)m zIWeE3{KpkOO?{)r`_;_RMpMx(lRymCi@8SUq@ zajg$)`u)X-O??9%+v&Y-Co{VpVtQl@&4wSOyW3uRP-r$Shd4y_5x*IZM%wdO z3``z|wp}xk^I$mqmqcLa{%sf-wHsZh?#Hmw$FQ|lagRD1W#XVfm3b8#4!|*V}UR6lE$5pU9$V2{`@e1YSgDVn};& zCo@=#>DN{XhKVWs`aG4|Om%@-h8Q({RWr26DTK@Yy06CfTBgA!--4+@FbiCYNA&wO(Xp#`2SV zERTjq(CD2x>~;&T`VoCtoQ%R;_nX-MvKFqRtk85nzH zv`KvKe}nx*1@(%L7 zCUdhz4hq(d$9t_TbaaTuz{QJEv@9KiCvC?b%YE<-T!I-YshFsoi&M7+Ys%vYemd>L z$Y)*I_uXG<@OfRd^Rh-KITc>Ac>y_j;pcyOM09gIq9(=<78Q!PC^)nyj29yQ%anv)(V99-Hi)NGb z(S0R`7w?5-_(vql8Hj$0A5z>R8Gh)hVlo`?tWhwjg=bG`R3^fbQ!u%p5`*Ouv5qHj z0hW-sV!aY)FYgU6v17Fvy37+jPU8Y`#;K#=^tOY--|<+r=#`Wjs)yQ* zqC4)cBG-+w;;3v3s;IkQrRYqVwJ8C|1{{ESiVZz2wFvb=I3zb2yEY0Q?QmU`y=#>U zt#lCeup>59jKSqQld#6wfX>wvsI4l%dc*(Fp>GN7Lf7JP#3qro%|*k>wRrR= z1Bp%DF>$-#rZ3-s#DsM89Bl&iuXf0s?T=4G=VJW4G@M*>NU+Wp2p@zMR{c)HMBB5N zV(SgRHd}E%&=v?iSQAI{|H9#g6IB%af+PT}+M z-n@I#jiH8vxGJ$9XWTZFCAiwlY~4&{?x8E#a(6nDMW>#H*fTQ{jG3{@fr`iXQ>jND z`&8vIPTyQMq0mJ3Yf)QS?2ZWGTMy!pX~DF~KFDB;y-cq=#ORCd+3s0)zI|*VrWm$sIm>zHPjL73E1aj)oxvUZb42b$ei-D#j<*MKcUOOwyz=G`f3g2=aO5Bt zWm)p?cjDf6j#Kvq$vWo`ldV!6F00AC!JkWRuuM^AHfgHTEd+-Hpo`586k`}6pHM1 z+8jB-<=IsXyp~PxY{4E|kbFkB`V*;lq&_+tme*~+@FQI;KR)4~}jV5hu9 zcK)=Mt$N_fOdtn zbj_o8Xd-JG6WDwuoimQraA-~*2bE@0fAMCH=pz0(WOG@=4o1D-P0tS5Y@MnoYY1zm z!Md-En>~{9UA(w8=dL-HJImoY=sv&lw&Pt za$DhW=B|w9IlUb`^&e zrLy7uA~w$vf0May>Fj)o|FtN{O5E;?-{U3qcF$#8>;)StzhwhI>YVxE&`a_RV?9Gx?n zYVVz>`7fT+f@jiT;(V^l_{Luc>luIw9veD^OM+%{%z zhqbY9*mI+zZ1VAcEZ2R*kW?pGk25ARry>hkvEWwz>oJChr}=V(s|IbvTyDs(Yj~4( z8r@t^pk%^AWX8@y*GKI^mq291zGTR>qip$_P0fj3Ty@l(T@#grPr{3boyA=BvpV|( z%|}(A;gD-L1HGS4;FKc+dE`wWPMfGmi{yMPuDgk_^+d-I6=bX!43&Z??%gnnvyTtp zrBgxToo&y$?7uLz7QgFFcV4>a#>OWj+4(~U&hIsfy8=Yl-LZIFzu=2!x9q68d@S9{ z+wtP*n{3J7%Q(|K>Tc@?cft2?_VdJ9-@UNwuokW+ewgh&1$8Pj;VLwNcX43lD~Z!1 zUtmV)CQiIs$Xe@(oM?3stDo=3F~17bXRkx2;!Ut_(M6BFTI>>g4llR%;s%fNjBmG# z`jM+yK72bp#LW0XuPNM=YQX*eO6)=>!8lxkdn(RI^OwQ*w+ZjJyu#QxMS83}jg-xT zU)-`DUB#ZGd8s#EOkav$>2j1F)QgVlvv^$ijtd<(@$8OB>dEb3AE6WO@EFNFzspb- z48oj$#kgsxA=|O)2UmLk$6^Zw-d(TAyXQXRLB)Rdu^r8Cr9=35YZz!k^V;r^hpF#RV( zM0o*&bJOV6J&}4Pn=tPD7Q|MEZ{n=djQctd1RJiCSQcw!$mp7r92gFj(j zb_NPh_QIrbE-^u{GJJY7bn6KA-Qmi`F(&MySkK|Q#hf{zia&j-QMEM%6I5dHcbX+v z6=^eMr#$b~d_`D6DXPzuV~mayf_^PV?r&>!w7CVpx(HMcJcP-@k5=?iQ_L;wC^sb) zeG|iRN+lZI-7h1~Vh4uG?ZzZ;Iqu2&juEk!aJ1qCGM{=@UZNyBbi9ru z4rQ?O204+Xyo!;c3$W!}KRnPh$Ayx9aC&eL{Z+Ft|JE5~tJz&^l!I>zSf?E-ZSlsvmDPJ$b$M8Bd6B5}FnnbI6b9wD|F;lK=<;9xA40*nl zpIle6%2=>P5B1?)1x2QKXs~s`G?7D7#(;%aq(3dn3>ftoT7QkW%CLrkbJM6kFM>WU z$6y~*jpr8Z-ZVfvWBugttDzg-$Q=;_OlEOAqBUg%LdNEx3qt zHpAp&9M;$`MQDdSJTMVF!%qHPZe6bI8SheX3;XeL_1An^xoR>3Gas;wYHe)*Rif3Q5z%n)2T(yvo~+`Npd`tjdaeeKHX) za${4T-$eVBTHIAESk-3BP*~`U`#v|3^Imv!CxxNUB{?cb79sal2_~fdhxsR`qpoBg zTGnsDgnPq~A+&F)qu_#z8G5RJC%89ea%tvjF6(Var|TP8Go%NnbQNC39j<72+#VmY zpGXsjJFe4+E=QybO!MeSa6|i+gjsgJYq1!wg&9N8!4sBC6D0NWNmWle;hy zA9`e>ai-{(tLzKiHL9rHR)VljK1h7-hW=S|V7;IVz8<%LrQq!hX^h1AIyryt6s@$++~#d}Av;B7OOpT7o$kU5lpYtFb8#(Yzv&BK}R;l1M= z0;(#Ya#VQao7bVj^%GKrW*R=@6!!S5V{6StN#X7->2KvGm~tx8(|j>^z8NcQ4Ebx4 z0vo-G@Swp56TZt}7B6AW&@oV$QX`%1)JgDDb5Qar94|5y_;FkmN|p>kRI-@;?1{rZ z@lGBHsDgjHG!*UihiqvAl8@#=zs3`do3ru!l<0U`@m*3}X-xB*!sC=wiG#jlkUaRd z6msU6WO}|?^b;J$ou)cW6z_U{WEk+;LvU6naIv8sFHTV4k-Rg|OpHKeaW6d9-v`YD zu~5p+MEv*)$p1G5v92B{wAhYm@!r_qrBwAjIGpEejeE0v+x=CAZ}tOfki$K&nYKA4`U1=+)K zn3p#QkM;;o_9f5LeO~eEr955`^ZRu(jo3N=i}3WcVti5#zKGeyp0rSm-S!%e z4vKtyyBbM8dJJ7E^c*!K>I94I%da?=X~uBCX={$yXiC$h7Tif1fR3Bm!rJS#^zq#m zM0Yrd5W$Ffz33Sl|16(A3PG z-)gY%-A&JFWp;9*}kjjNSUyTlZK4p;+|yvJvOaYVZ=Ifo;hSF&g}X;` zqqOY~7bGiy`H{FH2*Jq5f5j+B;13aa$i@`n< zFl4D8E*)}#MZ{j5H7my9iGsO$Vm)RJ6@7`P459pY6dZ@WmvTl8!G9&mP+t&(lBQ&A z)IEj&S}q|f<39Sti_V3H3M9|Hjg=QQxO{Xg?&$o(@yz20y%>&E&z1P(U4x&0ZsCCV ze?=ubSyDNb$8Sb+kQcL&~YB8)$8)mLQgtlihp!qrhKg7LuRhh&u z&4T&;IGVff*wJj+RBCw(J?)z(=X4yzmOJAhJu$}6LDpCmU1&JB)2SauZ}Z3U z+iRh8aRCyo#4{k)XpM^wTnX2hT1Cj zvcj~!ve5y?vY@qV>Dpr@$2@f9+Hle3wXZu}W2`xJR2&uOg>kxmG=pv^%dB6um5oQ8 z=-6ms=BySjp7wzE_J|BcO&YteT}JKn-{{}+66a>#M1PUh@G=hOuhmX$Ioy|DLi;f_ zvop2C9?|nnXRcl+w3Ma@)()CQ*&=U_S!gfY(ZfTQy%|g2J_$vec8;#Iz%T9U8pFahXnk9mV+SkB*0t8s zv(t6vjd{uUS57iRp`7oo_m^Gs^^`T_>B>HZ7jX2`8$7Lcl<7-0Q(yE=RV9k9!yu8ns(9v-Q zFE>eSy*7y7zsR_4xR2~b9%pO&YbiX8pFWpV7QGP;ZTo#s zX9N-UPE>4IL^GY)^er4n|A!9j<7Ld3#g??`)tSMQOnBTxj&GYZMW?S7V`t0r&7GHQ z7QBx=J#Mpu$m%{{tHYb=JvmdSJ8Kg?_%XQ)=hwgDKfimz7kHc&FJAG-rB+@NU2@^w z)MS?|JID+YKJmkaO;UcB3Td7F2kEfuA*3xlfxBJATE1(y)I4#Tbm849Ng6qT(Uz9H zwQ(TR4@IzUv<-(Y?8tF0k`#7Juz8Pmm&}(KOKR2zQuoc8lFPE5(yox6(*CJY(t>Y> zjLa1IH{WMi8NLx2djpZY_=j{N+Dm#VJ$0(eiIP0-#Yx}vcT2ZU*GWT94rg_ZiC|jY z$3L;gp9yor??Eo`{bMi9sQ={juii@1KyPW*mKLYNW7DL-p2m`e{&2~ovzgR5U6Wtp zbZP6Z%C5akrTKoE(l6zlcgZ@ zWd>Iaj^&K4i|L`T33`8BF?iNSu>K&hW(iiT?E=5|Z!vtD7B8Of%Fk2%;oveG50!7C zs7E-b>xmwjA6oSGD@5t(pD++Qa&>{Xq#oU0is@l5*(}w?cBOu>R#QO!sP(W;9)Vy- z7aUCLCMExFFC9&HlML1NLgVv#sHc`dTksCfq-OKM0#E8BShkCeKkyaJxo{9vgs$qEE&Zn)keg`dU9#VZYdyhhx8;nQd(%| zC{byat{7<8n)e9>K@RUkvb3Y zNz5FBl8spxcn?-xyKqAAP}4HWN>CU4-X?2`|G(c{wWYuw?m> zN8rQV**HHx4576O*uORgPa*CA5nb`-iVSK8^{{s2MM+w5O{z>u#*J>yUXwR0?M`-@}8;YbWM-E7*`W}}? zw@gEqew)!T?VL1ex;|ci(1CBTIkM|jr0K`zNbhflOOYkvsNFJ0d3GU0xWJ-x|VT;}YqsMWA%#^jPWY zF-2^cQ6*(X{FCG#zrhRF8<;G7jlUD^ur&CVwAQZ+PPm#&fnQXmn%KG0>6VV{S**bI zTQ%6CVactJ`ttfX;mMe~h-uoqRTz>^A_W#iQZ3H5s zZJ<9b3n~{QXuEGCJr$hAelO+|Zkv(%@CppntWa;&AKNWwp;tc9k|sWv&epscSB=dZ zdT?(>3jL06K`Eb*B zX{YeOzWHs>UgqI=6=uYpKjOI?{iwIfliRvYV}6e|7;+;RtIBUc<>LdXSx>^kxv$X> z(I~n)4q{u)H)NgFLdBX8^jY}`QSkr<_|3+}7)E=5f`FQov=x=Cl(SSy0!Fo%a&+lO( zL-cPCGDfuG)e|SIRjWop=i6{RtcSrOCvwu~Co~UQqr;~#>@Uto{hIlRYE8m}nlmuX z83Bi;p*Yt#1m}YDp=I8R1TzEf+1&;ey@fv_rXLo~j77JHb(nD9p7SyXqZFFdpD>Z0 z{|0lA;vlY0_M>^eK9)AGg3paL!qZ@mn~GleDz^xqPHAFhdk5hGaY6dg`-peSg8Tfp zP+hnki*#P2rS>%X=xjpcxm{4p`;O%C<8a|`0`{zkM`4%M*p#sc@mWXF?vWkbyo({% z`Vl_cPhphv1>A|)jgBUIxDY!Kd1Y&H;{9Q%zI`4}KD6eGOLwFY!H}N3Y9{@s3=$mG zzBJW5L)YC4Fj^8U>H&?YHb0HAyN|*%W()d{yMeYjyK(QFIx7c0Vp6-`bc+4NFBi<1 z(yk3#ZYc4-%#4exbI|RL4(~QA@#b%N>gu&+liM_S^_qcWK@+g}j$o2rJkE1-x3TNC zG&DyZ72TEx@!^%2rCi#Gk>?Ci-+UB#BK3?F)7 z)TAcNnW@aE^gmd(-xHH89AVcu98QL-aALzQ^tRfMnKpy4=4u~I3>t~F((d@x>JKkn zJ4hYI<78xC7KL`e%VaPHrIpT$c9^3HBCl-0N;_k4_jU|1C2sZjGmVFAO z=hB7Tb$cqtn)k;&gCO)BmXGUA@u;{o4%gf0(DS|)8-%yfbz&6y|J{tTl^c-$K#s1G z;8}3+u<{-{R{C*Edai)WiSnUhr@DJa2dZJi}H-wue1n#dzE2}ZW;z2 zZ-T1Xe)JAGM#u0hzR}*pvcaVoUYdiaDMc9F6oD?sW(q&{6nq>U&LIh7=~FzA@nOp- zy_(7}<&9MOH3eQz1>bO)A7(t(lwHexLgT!8zEqxw&nrhGbH!LZJF*;qX7@+?Dem~v zq2x@wc4|!RbxwrtiJvL#;khYXMX|b%kF~5s`AET4^u_(POE9&22=+i5Z z&uPmmLsBtLCkdb4_lB}V8M5ByV9oVC*qd|=e$Kh5e3uKa*4s$zcNRW@IcPXC3ZtzS zLp41CGNBWxJU<{l$0F!>uM_jFu}JDQ2wihuL8+fHl_rb3(`XAy`@FcVLoZg8eUqxM zwZo@*f2C#V^*E?|5{aefVNg3Ax~jtw)w~pM-k0M=Yb*mcEX?g3-@kxrh2j* zTyEW#jyJ18Yv2G{$2c>8pWu=TzLj-{GB|4f!bwfOn+5=JbQqsRVFxX@Q< z-yi4W{G?^r^e!HUe&%BS@WW8n7ul2X_ShHgiLX~eP`LRq(%hcoZ`uJ^@7RM)8ZN-a zI80dCnMVd_vT>UsXN$beuzuoew|jw)PFXy3DUxGv2`>F&W&TM2fE%OCcqn9vbj2w{ zf~N`29x%tpwNp?XP>RqvGoHNfz>{sJ66SK?T7DDH@RVQHr)@H6`+ zX2QdnQj&xgInmE0W>EUZjZ*lU6v=*sGB@mdCS~6lj->BRm}XZm%{!}%eRF&;_p=8k zZ0d`>*UB;Kzzs=Ba0_H&4*R{s4{6fHgOXOd2R^2XoYDC$(nj%hyBm(d);kliNOVr` zKe-%TLdV0cV=bc6D={qZD8`TP$yN1ZiEKlr+hwwGwKED24Sr$ok4NO= zPf(lp2WKb!LQU!khGt#Cd%;!vBIb_v6oi%zDbfMwR_S(m0e+jQ5XzI-vUCKZdYd5a z%5bqppOXT_ex)%m6LV??NawzqAVK8*98OA{<C??}wqQ7l9#nW6&*O9ORQuNrMjz!>o~QIAMP} z`n+{VV#Qr-Q!J4N6_!aagHo{~XBk>A2BT;G5Zs#BhGRvpt9#WC_yt0+lC?RwlRH&< zCqTh42nrq^kPgnorE}d7p&btQ)3e~TJpo;k=HSp2M|cNI=v<|T$@W^f)bLtLFw4h? zdhz@&oVdPV$$LOb!EE{{j^6JOC)bQF!=y>OoTxSqQc2hW&cnMn_X z$4$ZQ^Q)w_O0T6i!dKE$S)7F)5T&z@qCpaV;fyG>i`xOU_r@r%?Tmtf8PHyqBv~0O zlzs{g;kR=UuB$YOdzk~b_R5ydgcxJ}2_;UdX#>rUO4#wAH#)!Rh-&Ay$V<`_a}h&m z?QRRdsDIL($Rg?8hXaz^kOb)M4#Lv#E(ndik1HwGl49aX>0qcoemtIo0^>JQ+ZQ%i zAh%c29rHxWc-j|fdru>)z8U-DtQgq+t~AwE6XOd@5PYTwE<|0Iw(3aeFS2#XgLfb< zRdhLv?Dm&g#u#^4leAICNEcOjk2R-BkTO-?x(IF3nbN&2qEGP)zNfchOxJnnYMF<3 zZZYUJSYpp(x{TV^kyTwpZ{yb|=(}?O?XOmg$5e;AC$wc@ey!lIe8a?Mbq=u~gkODB z;lJ1%exet6kzlJg$wK(iP;|uHd5Ip2L|*!@CBBX{fJckCSJ5{*o9x-eu2qVw)I#0JPEZ%Rs-H^(eqJ(QPt5?V z@G!y1<_(zA%?gUMsw7*TCh3sF5L7HMz^iULaQtL~vIhY;bt)L1)+*G!*PA7$B_{v& zPNLWNEo^G_W;s^VqoEg#eteZ&6aGj!qDN}-6hrJ^xCjf6Z9!qmO&l`(g{$p8 z!u0rgDX~wzR2yx8wJQ2NyZ9Y+9n5HJ9>c5y5&SZso;J#Ae525V$2tlA>VTJM@lj*$ zpDTFYn2O6)+kp5z_`T~K_FlS%C(5ETMXjFK>Z)k$cAS1=wP4#@8Ott+b7!;?nudGe z+PgL|@b1qk-S@Ju-B<2DKZ)LEYdCjD0(0}WL%qi$EPs;&z5QflSHWYN@6K4ma&%l$ zg-2hyu#2t>zU!{XX0Ig7uy@D#oO~Qf{e=%+;aD4QgS@vugJ9DHoJhd-cWS(;bpwm{ z8uHeI>D=X0ij7}8a%$@gR?S_)@zK#d_GK2E2Du@z{w(T!eqrk4-B9svQ&Yp%SVcU{?vB6pN<2Ksf(hartu|-}5(Z7ikcDD?RTzr&xSP;C zl#gLOaL>FvqcN>?Z)p*2NgHp$% z_#@W8_LXMbBAr3c3F=&SsW1Ho-IFdqyCRun1>xFK1@7#47D-k!p&liCoYM#6h1yjt zt&c=#@kE@f%tMU-NNmq_hRj^_Vn^iR$R|zOb?e0)g7Gyq&w-_*LitWLjaJ#|{3lbC zNwcnEPxwh}Uw#ww9=?}`<(EiR-}*vfdJX=I-hmx|^6@Rxmii4FS@+~5o#Yd-XVX8a z#_=~^tEuwqygTT8yYoE)5-@%H*STx@Yd%2;K_EE`O(xpQfk&>aI6 zR8W+#0{ZE;IJ#d4yPMk!zUEHcR6mZY>Gz-|`aWu^+!^I*&WZhoi|l4++!p8Y8yg*| ztyO`YyQ^`7?#TP`Myh$ChWFOZSRsE9C9<<{uT94;BVw4CCAh7QM2~HLa1dvmrZ;gS zuhSnB1_`fE-Xz?bYy;yix=?HQiN)e~=xZ~CbMDvS`n>tb*}0S<>pTSqPzwu^&tb0R z0(|d14UwCkVL@{VE>^x1UGu#e_vJMF?gXLi)+O{7vz_l9=fKllj;1#ZQD+y0rf({= zvarOj+(v29P7|y#TP-|Jdf22r8rsgHuy>yedOhBY)AcTbDdG?9!8X{kQp|1gt)Y21 z4;y;b!7WCQtF8qwbCU)G{@p<9$WpOJK852$Jr@2~0i(DeoT?51&-}*kkgJG)GXdug z3m;I1J`byIpxmEbG!uEE_p9V^>G3$s6g=3cZ*1Xs(j4zVHFnW3Jq5e90S)YPKVt=0K&J)4;!o=NvAd~hRK@@nP+aPr`lhE~H#vg}sKIVbz>SD@L9~ql}@*_VmM>-EZ(rWigfs#*#^y zH!repLdNMzsEVC`ui0xb-82plT;s9H^bE#q%|=GlHMCl2@Z*l2T-`%-ax2%fY(O$c z$gE<`kIxV@R-QZ3v^ly$mS5g|3&;IGak5NsTm>h_IAjnmH~4esvg!PzE|?pmTyRa4 zd`v#>MroN~WF64PRaa|FQ#I$aZwy#>tsQrZUCd96rm|eQH`^K{AY!f}&(d#0r$|FS zbNUOqnf}Kk>Q!;uQvvR?dZ4xF1I#~Yam#>*f@QmlkF1`=-jjs4I;4cQI;%h?b3T5) z6nBMW!M3S64C@&aP;}iBiO(j(R8N!h&#z!Nb8Aj5?aqfTY~uAATY2O4_i%I8X3taV z96dV>`JLb5a@=KPP94EL1RwigQExsXJC%c9kK|tl8ob)TfP3o=HrbaU^jR$~`MTo5@BUD;GJss-E1YYo zM)S^W;Se3giHa+^#r6=tx6tF3SwC^bT#;=})9~s_nsE7LLTa5F$E`AnVMn&%SS<2r#wSw#QIB}V|=pI|KSD^*_^;v~2)7`PVF%0Qb zQn*{;VXixo$-~qm#NWYE;adnmgM$*wM7XnM+)!TCQ}_>q>#=vGGEXe?XUWTWKCN2K z;=j)pJJzwuM1QtbAAp5!-QnvJkKH5O;4sG>A7$5I?xQF8aY~*;M!tsUrEJ0OO2gb6 z*KwpyQFQZ@vFx)ff9!k_xwe%UXe-C(VozXLfD9)rYlO0{3R~2DfsU3wKZ#S}Fp}ji z>pF8=(P9pL;K%xjyZG0g?KmdhA*HhpLuqss2m-pl8p6WwWsFNiEs}1$*X@=*Z4)#luTlTpeW021~X0@`)k*LN5rd zVfk>J{1mGq@8Hs-Q&5X65m`eC7xhnN4euFzsqUDYC*=yHVkks zg4d#K6l~Gq(?y!b0^VZ?1!_0lk#7C0p5&!hcnvGF{to1QhbAuoso*X@&q_%q(JpX z5q5q3NpTybc&6MPe+46Jc4??!UY>L@f>lZ3R@~;A;0Jjx_HKl`Qub< zUm6LE^hzXJJ%d`T5-*6?U?+(aH(7OI*`31uvuQ5dMY@QYxh$41P=@Sa6HwX{N&Ipp z`n$ea((Y)BJEit8+9UkDk7}^qHI<*=@5ljq>!YsU1&D!7+_iyEe0 zrn=S?dN22xyqxW^R+q`qd_8sV?nv>v7O=82#fi4h^hP-7@Nys~l`5c%uM~@RR!~-# zQbOhu+?akCe?%tznQjdI{JVz2KU||;Uq;iLQ!(UMZAfxcCCFI03W>kPY~Z{2uD+zi z-+Mg4?u^@%YPy?LB5u&PnI6>tmnt>SSVXY_D^MX7eT$aGi0h+=3#C<5-?bC^UwbTg z=)DD}zKwihx?rY;6()7F6ingk^z1-aT*?@a$|XIitm{Wfoufa6c$-j9%M&zzsxJLg zl)+p7W-_T%M4y>EvF<-PG)^4_$H(EczN<5BFWpKu%RTWm-$uLxGwDX!OA;u1Yqb_wUYWYsZ)HMDOpdDC$qoyw6r;o=C&G;L62fu6f_V!J_OQ?R}t_Xx`d+7 zq*Jfa6R2{|Ir=MpAC?KZG=I(%B%M%0nAu^FU{%gKD2w(VQbcW1rHf@ol;09aLo?ec z`c`keTc?0=2S!t{+DiI*P?vO+?a)c=sf|Ts$hI(%TEiaGb6$*%<%=jP{5)MBKa}Rg zEyaX0V-etB46Pr1k<-0`<}EQ6yG&VJ+fhtQSsBaU4ngk_CE?ob56A3fw6=N{eR;Qo zF0Bwh&+a*N!%LB-M#~~GJ(#**9RrnHec|rd86qX2J;jWQdN>dnOr&#* zifLBpUbw5 zkn__H9l}?X85PSC9~^YVdC?d_8qW9zlv*=GV=u0o>yyF^Qx)ydG_R&^lF9&*{rT5-vMfZAyz|OCOTrm=z0qI{*LZd4}#Ok z5o|q6m)EH+;IOC)GBapp4p{8@AeY& zI?qjPE%w2at(0KJO^Eu6$`2}iu!kO6l3jAida_jGb zd^vM8ZyF}NC94J^!C&lxc`Ntb62|FAlKA`EMCv{%k*eMX)7XP0+%RN1tM~Yi57~9b z6OmErTcZo3_MQmuQ$_JV1b21EFdk|l_PK|@i2H**Uq8};tGoNMj#oK8TNR*s;(d(I zoX>|Jn{cO_&TRU@lMBk3S4rpa(a;SrQ|bqu*Q&y`JCF7BU08QvGWRmtPE&N((5+Wt zv{>}9eyury%zf9T~Ql&0lZ<;m($LK5#sgHTAek zp$#4C-;frsizyAx=)7b#T(1OS@M#YS8ZdHtC~?fTVnn1Xacu4`HVIn5FE0uHdd(hs zGWIfUzWtT_vSjgF`k5Z@>57e=%jy1zD4Mb3Ha+ZJN^57z!!P#=4GisxPj0<&dzB5W zqa6_OG8i`>dEm9|OUMkVMO|$=Ud@%^<%7HPyGP|1b6j*eCflGvbXx{_??>pwx!8JQ z02+&Zc;)HV04c-3?WMz2rfe!l|hrFMx%9llC6AH$J4U=;Lci?iO@6tWzqjC;L* z(i^`+Xf*YP;#4oB&^lHww&Ku`nS66YAQG_*D|07dbkbS!E#E`02S(G8wi+1EXvKlq9)DurwMD|E{1Xsw(Btrvb;_Xc_BvkT5Y zF~gE_3m&&?B6kY6;R7K(5v6>Ih6N4Aj`LPHe(fMeBqd{;TN*+|=b&R@5>#}bQ|y~7 zbjIgA1>Ng_%he7Pc4w~0F@#mXVbapoLh;tCRAqLa?B?vEw8l&XZhD9jx%G&7=Z|qC zf1~%U0V2PyfD1od$ZXhT_|D0NpXWoon*SJMqE6D9kX7V7=nVN~?1nd=h4$ahydrNh zlnemhun0WU*-PPnV@T^po;br@B$;K|q&20G?Atz2Lu~}jn_feA6uPpe*uh2=$*}cb zd!D&z2&!dlx#{(Fq@BAZ*d8G`p0Wk853Zr`~>-9zK(F%j_!X_dVj~8CX zvP@k(Y`4PMX*Sp*y35w)YFyUbEPQII)V?s9Tn}EMxH*?-kG?9V`nzDzGE0#|xqu0N zsc_#FPeU}0)7As=WRiY|tP-MOW8nzbiDwa@@>^sq&G^%o{n(*?2B`)`m>E0`F&WFz zx41uS&AVft@GX9?^upb-nwa*pgL<~OA-#Mszi{2d)ii)dW=}`w2QEXy~0`zGwv_g zc<1jW;i=*#X!@L>o7)GWb3qxoZK|Qjw7FOirH$>oODKO(18LWlQT&RLqJ#F4#yGo+ zti~|hGx$PYKiA>Ngbj$T@`iPCIL5zvfD!R$5bCs+4Y#eK{L<@`HToC*9ak=vMHv*| zu8L#cBL6XG9X_rai3&Y!bgG|#S>=Lh?Ga4%V#j~ma3JIs2f}9S1q>J@vPlo`;<}i9 zY3#X&pz#ag;U0x&x^K}T9GOb`&8Qh)i8af_uI8sBhEE6)9284P9bVA0hk}8~vN(_k z1kQEDB|ll{)hojKbrTs)JWcJ>9@C?jH>l^?@lZcy0ymrScz-T~T1-=ER^m-sEc$y7 zTzBJUWPxA;r(%cdBsezzMpv11Iyg9#k{9GrP_rXy4g|n>4zP{C+N2{9tp>~ zV%yhrtk~0xPZ=sO6aJhP;}+r4ojcUy^-P#mnhRf{n0@A^P;*=$In!(V)u409 z8_h~X+5FK>M4P6fxhk7vyx!2z_z*g~-wFBa4Uo4^2PtWpbj$iE$!i2s*SG}ycQ1_W z?c%UQVVGdb|E9zwd6--pE!^juv9(y_AeIHwug$^qZ_8myY~4+h#(JR6$qU_{hd^aQ z8qVI1LtD}eB#XXkGl?$F_}^q2)=I|S$~Xo&R80|k?MdqqH^B@g>6a+o#RwaTONC!Bqr~vzU>xek1L?Xb=OH|m2iWpMAE*$b(EDXIM+&sxZNl^b2FR} zb>IrUbNNFW_N}zA;XO4h*TL_IUO3%88udqnV>49DGP4wr9xU=qZr$0)Q5ko|>~KJY z=%G8ek^9nT)GMwFel}TB+s&_Z@aGBa{@Fw^^6m6Ty}w}6hhyg2U9cMa1!ol_Nnynv zj0mlO?Yomyxv7LCgTIj6S@DjL7dd2cSC*!C#p5ZNg3~_+x|g#l(C;h-?$Z`7(_5so z66!DNyL_wQ0i`;7mP_ghWj_>t$gi~r$_cL%n_OXCAF8^xgPu(^ z#)K*hVApqg_Wcy9=LojAf_TP@bg{j68U}=%MW~A$hut?p=ix_@ap@CwnjOc=5h`pO zrNVvnqT%;r05@&w#S?$`!181%rgI#oR>jiExpnmB{zDr7D~^&}lQ7Y}7dllz^ajVF z=<`Y8W3xl(Fh|S_3_$3Y0LpZ}K|0Z+glql*#i~cc($WVr6z`I5>KE!0a2Q8^w9?NT z4fJr`6r}EWMZQ5|UKV7I@=bqePZ=R?++J9n>yE)!>>wL47SCGx;Pj#esNc3taC$#b zLB9^tu^5Vgn*ZpyZ8=%In}X{L8MQ;4MGkEZMlbwJI5+}#jjrKm{cMrn9E_x#6Yx*o zL(b7&R5861Trx~xxPC9?in;t&(|>gR&NMWh^gzzVFiIM#i%qV*@#n@D>MSq*{^!Qw zXy61CbR2~7k=xk3xijjO|4{lsM@(~7hx$C=OT-KW*4O}%O9gj)jCc-vB1N3t?ESkT z;KMFTjaWf{F8`o^1GQ1HY&#Z;e)&6jTO?WUN9oY7NGy&Jc?mzHir& zbyIP3*91INh@*^S3dptiBs?fd;`1!WD_svVxFhDMzVljW9F`1E%fVv)=YymrDj03Kn(Y4A(8rf6 zDYI7-S-GvnsI|hs(h-fjsk@+bLm~9EN?yT&Xp~D&_#IE^4=LB%49j+aR{U{>rJpLEgL4~ zt1zSslSYdo9*1^C{_cTjnQbln^7^dUP$JI%y97fc4X>rbBX-Y?tah8z{riKdmtf)^ z9JiBJr$y43a#QgvU!{Pi6`oI%&!#Lsk@l{ej}&%apAf?udsbW59~fN5o=QiL%PJAcea@FfIe!x zS?Mgb=S-zL%KGFrZ#h+{Poz!bOlj776P_~Hi@WdK#*Rr3;7~5^f&OW@RilQ)7t8Q! zWF~!>^$3fciXEYX-WNx2DBO8;*|CH!^ z!81n#bpN9X$__*6+ReBt=6mOltj7QiC)8A}$D%DGVE9T8B}3lOg(JxI6tk;dB z!zF_2H%A+{ZhfQMmg{g@5(4Ad_Ly_-He}Zv#1-5p;+5|3_7RJ;Lr9VotQ!T@SJDcvWRY(@M604r zX~e=|@Z?FU`=Y@}7)JurE+3j_li8>Q`zl?NTH!Rb2LsXV_oo?Mj>3z%YpYE>7aIF^f5_b|1SMj^h;LGiI=yWZEzx~+7 zBXleFo--~caEjm`^fy?+-Lk&$FnTOHpcgnW zM^PGZ=P(!aEaKFY=eg^erL33k%2{{cbKeWMc$8xi*Dv^rjlFKb^I;M^9jEi6HbGUTr8GUJc=O65B z)}3*!D=RNh=F}o1cBm4W>D~EoiT=eZ%RX~M(Q)1pSFHn6edRbHM~C{D5woV{P0 z)%y+MXJQ}u+@mK4wszxD?SF9DK9Y<6d2{z?{|SbgJOBI7on=)Vx$}WBeA0%wPNxsQ z^NHolm*ctH4QHOnbNN8=Aoh@|^Pk<}oHss@(}o4`xw$DE=o-fd9`xZ8YG3&Di$;$9 zd7kA@y7T?3ZhZ5nHb=MzbCL)RXaqR(n7(tEWG%Vqv^5_}e+A7uMff53A{|i^d6?ku zmzZ|uoCSgWO)xF>sz2dMN)p>G59gEL+*xgYF5fpO=0&Q)%N*H*&j`5TpH?not1C6E^I*KBqjm*Z{c@8e=AR=BYofJ_y=iyge-xS{ zaz>Ark>~mZI-()`LT4UIo*w)`#s>#smHiePHhKrWdFD-CJFiLX^J*l=?j_T#f60=8 zi!^D*jSqBV(mjb*qdJ)>sgq*#K*{Fl(Gp`bM|$aLOy;S(DAwH;eImC|vAmtcc=vni z`sKbv!F3kp3FgpCr9o&}cTA%F@R;P*z}=FwsUIXCh8rX3$~qeRR)O3~Zd0ep)smN< zgXw#!inwRX(Vl^WX#BKXnwAzS$xxk4VRow6E$K?@w%w5Q+%k^br&>t7Egi|&HIsan zos#T5vW(W83!qKiV^gggOV+4o(tgTTjnpRdW#YC9*I^80|~)d+Xmw#J#-clk@Q7S5)! zhbA~ZZ!X2>9itt$hLiG$SV>Tg39U`uN$(T8QEfLDiqiZ;D}Q{DG)DQ8Tu?4O(7r7h zb|zSo>iU8XzZfN{G<1=eY<8xr>mnuVhWgVGZ5!MP(4nhichjwI)ztlI7gCS>Px9oN z67_O^ZD-S*DyjJ|U(A~JN_s46lt^6<(Nj-zGB@od3EaP$-po&wd{I0jIV&qFzdJ@s zu2&_{m6=7<6eY+J7NL^TkH@I!$~MWK16}D!r(HC&ZH8pZZsAg#x0xop>CiaIaLJvM z5*oB+fn@d8>EyOX2Wm+hNV_RY;;`zTq}V|s`I}=wiMrkKIXG7`e?^u=*k*#} z4vm-Gm;CTmlGxNH(DUWHX>Au7`k-znnbf@(*%p4Y>oL7n@>P;5DJeQ?7d~w?bqN$6 z6vbdk!qai&SlF9ZYrd2?9_UQ%CgZXEi~~jc-jKj7pJv?tDCv9e1PvMYk`yhUNZ9g- zMD1*&b*gik^5tqLyAK z(L1x3QcC_xp1x0(j&E0UH!z39Z5;WTh}xWsb;lf#KiWPD*81x?>8 zvGAEjrCu8)h1t&}4L$tnMT~=Ft&<5oEgnZNza5cO9=wLK$g4=6-GZ+H@q(l84TX6j zXn(W~Yt3?y@IDCj(P4bt!IPcl?B&7pD==zvEnYPTuuA4;jI6B0%k!6DZu=eu{jZ>h z%v03b4Z$Z1U-WjLfa4ueD44FywtxEbs>;C_`#6D;%rXSm{UtS35gG&^sy4xb-HfbR zT~mdZ+iS4oax$EzH;Y`Nr(ijT;oqMpSiL+SuZ*i<-`a*>Pwn9_zz|_8O;Ij-%ln_c zz@{1HklU4lIy=EZdfSa9J6$kAob5m69>mII#_ys?e5wh?RnL)vjlTkG9hXAquNJ>H z(B|i%-MPtTEGLQcTI^o$ABS$!_uMsj6BDQCLPNsFkGD=oyoP$pxDbpfQk>GjAf{jXl@(xCoEGw^078 z4^LTQ$05Uq@xkCmw0~D%ojh$$an<9Dm&?#uFB6>;zvAvM3Ev+W!q1-kfaK(Ugl}Jt zY}|<;_r2KMMf45MZ|B*ei&;2N@uu>kxW6f~tG_oiELMpuz6Wy4?Rc4a zA{TDl$Cl&tdHzDt?QUtu+OnR|IwMcv{oh{2F>$}ujT_GNJqE3NOYn5OE4pgg z!OMRO-p;&$;J-%@b7VIri+iH|jy=e)4ufz15>&ZmBf_i{!BuZD`(_H9#NF?NZz<%e z77PB`WK33j0T<;pXsk>_ieirFGCshVrJGQsFXrtcqvUcxo~KtD@Nmte$mzF6bVxp- zU)&@p4%~^S%M-ExLOy;?4Z`F*`7qffi%qkAP`G{`M(InD)yqj_AXnq}i-{<1A?$I; zLB4SUlEltvcVRe|Pnip&3#VWm^#LDzbTHy}F}EBcgjPm#FW4k7N!kV&0@11nd;GN%s7Yqfb)$15xj z7Yx>WUHH-Xf5>|M7uQtZqlcLPX2j}a@~2MN6EX#M^}BHY+(KA7{iQ|s_3=Ae8_II2 zIJUJAN}u0|=U^WKoT5>7ApnntPeqPrIn9WDO$U6ZA?d*)sQMhl^PFYasObw0KVN*+ zxr`+m4+Td|mu>cn*?5I967J)zM*Mz}m1@v^NAgSF@=)c=5Tx-v z+sWRu4$|@E8PruqWGk*!!07B7Tz@FbGCB7!+Vuu5a4|;S(Z}SG+ZxjFI>M;^y{G2;UTsF0b`@Y>qL@4xY(QE;jhPLJ6NVeDTcnD?K=Jg=7_P zQP_XlB3Eh3uFbu8V?hU595cpaJ7WwK9&GoH?r6#RN^?E(zzGlFr@ag*0s0UvWq7p; z#{bU`lrqv6!dOXH?;b|<%N^p(@Bn5Qi^;E)uzPPPX3yP&+P{0@;%fzqst)RwC^!Y1 zbE%u?gq%Mla@w8mq2b6e4xn&8AG(g)ovdNd+Xoj$Ou&g6bHSz4#H(NfoRgme>D7g> z9X%fACv5rdzJa{tK`VATzoyt*F=RaaG`+Ka!Yz;e*xmFx>&5A!he{po3J{L#3{!-? zR)*?1P55jVE`_j5G-hcgy{>nI(q0J$^qGp2_X5a&cojuiX43lfcl0&*4mFxv<3sPi zG(GPWbzPN9=D+?4#()+NDDq^x1MQ@!-W^wZzb4rURrGrG2a&rgrD1D?Gjo#QAG~P6 z!LLz>RnA0vTqWKIKBO1hZACWf1NHgmj~@0j@ZMQmqi2QVMTaJ;`$u8*WEK7+7?LNi zKf=jg?Kt~PX=d(A~Wv;a^;U z%TqI<^2-KRi=`N{c?it^iOjc&9CrF>AfabCb(t4L4my{~GyW>*?R7jXt;Ohn@34R8 zGsKHL?ilhTvmwEB_e2DZuna}6@&*KC?}SSFN_v#}j2v?h(a1TSkf0%td-1oZuj~U1 zGEBm`AqD8}UO}%f)YGVeau}yN5mTF5=;*y($QK!}Qg4ElRRBzGx#0CLT@)@^p z>ley${q)CpJui^l`)1ITLqn-;W&tucT|mXNKrHQ%MU8#d(F51>)LF`m6StICr5 zK)9b@&&F#Dkum@EjoOQElD?LfaM5(ar%-ux%FCdQIwEsFFa%PITUhk00IFqu5tAN; zncG4!{@Zd4Y7It;oEdHxoS-g|Qoz_5hJq7nQm_Mih8ED=vV3~~`836<1;g}C9BOj| z5&L2sX5S5fx`feqs*J8uHEr+^&xrhO>h>Q~qEa|@6)u*^?N!iuKLTS9hqG1@Z z7&m7!PwlMvch(&4 z+&7&Ad$w}g8eQpK4#T^=LUb?xff2qT7~+wShZ>r^+Ehm4VK4tFWrxgG0PakaSm-Ki)Ut%yw5kaegWn`t0Vq3k&#{ z*J0lFIft#1!tm>#;HHienWZ^-XcB!CGb>GQ*$>v9F^1pksk5bz$aJ+{#LWj|xc2J@ zktuJ&%@>2Y%K9nHVakge!ui+Gi-^eR#%6wl*v{sM;1BiSNeqvS;N&|1zRA+mYuzObLsU)ILC@k&#l=k4VU4e2OBZGLxa0mWI(=eIr{!I z6d92i$ZaXa)%i>3j4`?#w;E8L_eWkm{eV;&1j~xg9gOaPod$e=7)OzI)+N z@)U9pkE2=S@tx$HdBg#G9Nb!3kf8XTix$_XFRc(vhjK6qSs^v2y}tF1er+_nd@mM2jq z7~VrnPQzRD`nMY$z?a5FDD{niva=G0ZPVgIOGa{6wI;mWXU-+pda+Vzz361iv0;e; zA8)yhJ}Rf7zwk32D7`}c{vmwwhben@b>Nf$AHEwM&$D|8m-Y7!R8Jkr8b!efTkslH zQdNGQm53uNwYff`54$RTKxTn8+YXq=*)sz9P~(3*+$ol&&mZ8i)M2^<_Ggo=#cy zX%u`{_**wc;PJbI&_8z6@0*eDd$PiupA!^JuuLW@%;B!Qd8*#7Y|n~ z$i6D_UQJk{*_V6oX~i9a|GT&U9Z$hWV>3RrCl1NQ8_3@I2mM{! zLbp3Tr08C+smRh8g_ajcGiV-88T7@EhzdHN`Ge}j-sy3%Dqi%6gXGsvWQiGtf_4IO zdY0jQjWNaz8V$c}PxNx#B{Haj5!G-EWVi<>8TK z+g)_=xlm8$)2er!~W+e!KoiMh#6$0 ziVCHSX(UzQ!KyDFfWqmAFy)e9LIn=O^PT$;wCy?V$h=PtzA?1y>u`up0D2y>M(C3Y zyy^W5@x2b?Vh;sgU!cihhJAU1d5DN-SA*v&H@` z&**917A&6hk3JSWqQO^eu-4%ktt;M*`r-HRMEfvKw$@T$zcg|jl266T!(mo6ACi=6 z)I{4L=8-?P)vv)@rvhA^*cYb{G@#SkL*zBghotvosjrtMhc#8VI{SXQ^<01x>vYO%L*{@v>M05>kcXYz2JM z{6#mCWHGBY1~+XI@XgJZz2n7t?^#Fb zd7nJ~(ld;A_YUW8omaz5Fc;f?^hC@-akl+&nlxX2poi2=xV9pYBAt&e&P!NB?360^ zJ9D?WiqM#Fk}Mli$Z%U9lo-omerYKcYsXXBf?bqz%AcG@jKlNAHqae+kMcWxN6fMy z?0S9}8jmhgkLp13OFc|ZX(HG3>$rHvHqwC5PxQz-h4kVB>GXxUywKW@yNO;?e0wP$ zZ}#R?*=3w^-o_ z!74tA_D=^;eqtCex(m+SV8RY>ipfH{kgN@YX|91KhTT_&Ny`MZ1}N}L$r*XK$ookkZQkfe_qU#)KFcF% z#GO!j=4QuV1!rlmXRhcs`tpa3qj-=}Pj;K?iKgcJG|u`oIcb>i3af7Ht!B*6r~2?N zxAPcmVuadN^YOIKR`^1CVs$_cMjkW6suCH@o*7R@0dqrglRwK{`a{x2vitUo?>Sh0VXPuNiO z9f5vY*!QY}6iTBg;d?S(?D%gJkx=%i#9Q2(xX=$ZarXn!1mIYV4fEW8svJTK8Iqa5m2-GZ%$ z>+w?KJwE17Vw*4hI40hfJtMUt`~D`mkK0a}ldmAesT9+m7ejI8ZQN{J1Nr9L$bYH8 zv*&8_=NWQ5JUyH4nO-2r=mN_1%tO?0TU`Cz6~*%fE2~ZpgZd{?!r@dn$cgOcfe1Wb z@r(Lyc_^5oFG$O?n4DGikjaJ|DxR82Ck>ZSmZL8n(sbo-F+Y(L@Df{Bg<@{&KYBRw zHD#Fjv(H!55QG1O(ek>z-<-KG%ZYWt)6q367ZYob7OReRKg5Up2 zu+Q$G?eHC%{zVRYxu?k^GX*b~o`&P9G|b<7f-ZH6qQ@P{RO{!B9@FPzoW>Z;60Wf5 zX|YHeJ|E>;*|@bm44nj5WOV6OnwT|<492adQHvBgTI7h|U?}6#ktFW77bPuZiKIo*Q{h zbv&CN*JU|>O`aB|&6i*F!|r?o439O&uK)bmAtZ8h816omFPxdlGxtv52p1!%RsKgs&YN7og2Q?apS^ZNIzwaH&F5YLjw{-z`Ig!pQ*1q6a+fLGB zY43UUs7c&q)kMB{b13InnPI_C!jsEps18@)VO1);DMW=`ExYm>-Mg4^yA*FKdg4!r zIJ;j{qh06z9fdfOshgtkUAf@_zQ|9 zy`|3+a;!ybg~_kl-v z7_4_1NM);q2k^9+bk8q${#Y5w0~6MA?2DZ|A!`ADw_42d{x7(nO&bs1E4bRKCwaob zQ^NnA!jIgRV$Yr}IB9kOgA;4`o2#-k&q-10{;7g`P3fS+>BcbWaZR|tt9jXoJl?YH z0Y6%o&Q{7dxy8*`I`+N1)VX@B^mn&F(M8$BF`L49aECpAo2|>A+=lXpsoH#|w;ET; zbfBB`JnY_f=ES$n=w!GY!(#UePgWG-hA-v+_J^^?w{`4MJc3t`x8T;%E_@)zjTdU$ zvfqJu{P6Nz#J%xD!F3UC6V8;`!5mrCfUS5rT%tBY4g=C<_+X z6p`IipS=+eO9HUq{&v((UxuxNM~IH{0;J6&)by=G`?YI0;-tu-#*^9W@(|XTXT$3i zTcO`7vao|Xu(iaDhkW>j@~PMzzL~VvtfSpB%h9s;pt(C!b$cMgG;ku zJf#VS!beo19RvRxnOHkE8b4NC$NLV!@%y33)2|9v@CS1&%(Db7+JSAdzUbV2IRe~H zbGdkijB>8>+`jh&>**|R1m8!VJR|Yx8jOz_j}4pt@aWgq*)*Vn|IKRV@affD+q0I} z>U?7BRAp)Iz=vGz_Luia-tv_q1?kdYA5Q3hl;i65bN_v_dDhGAJR>oT{dCQx{u`X6 zT`SF`(`pQ)bx8xHIl2<*r>r18oD|MqBa*pya}fUv3ue!@K$cr%!nMe4p@HKfx2WTjumeBtoZZT$A< zA9g&E$3~yiIqhB!H(M9;uI5W@`u75Be^TN9+)Y^Rr5g7fb%Fcj=CQfeU0&5gQJVA9 zOq!f;BULW%!T+MI*{{2JC#yK{nZL$7Gkz$W^wj3<>cjb6#7NdT^^LbS)w9E|F49g7 zU8RdPCQEy+pDrEmzL(G63Fj-zj&Pl2B+DxX^XtfH9+lU~FO|ATbG6l^uFuq@(sW&^ z=_fnsw8NdG@_YYswW5)9<*d8>Jx4fHcRc0a8}-@L|Ci{s_v4P#aDL(>{GKO|u)png z4)F`&&AqmBYQNt6`hG9A*05zA$u727w}bcJ+{x05F4DIV)1(vi=S#=zlacl~rYP0_ zt{`3H;m?bO`_pj1cAi{1Tv}rlES2tCB+bkSWR3TMoNcs*Us;nhH%uzslL6_SPrlqo z&6gM53*bX{Ect%71w3Js4KF!2OqxGxvGnCwFRA_iI6CimEdTG1BO@~j5uxl+G+gJL zyEIUgb{djWw1<*ZWXlYhi6VO?qeT>vQ6d?A6554AyR7Q>{{H^%;nD5BuiN#$&Uw9_ z&+3uJIK8N7d*D!eytYj_3Dz+3lQi?O@)C7SS|%bkT*(FYTgZ7A`DH7e-AVB1y_H zhN?rIsW#k)_UF3M8I>lw)cK4Ym*vsE^Oe-N_zi6cFQHJ$({yQL6rGzFM=3SYl*Amn zol{aNU~wt^e3?l_KCejTJn&{OQ|F1VQQ`glu%A>``%&u$<{j}qO9I*cw3^KxQ`>fv zDPug_Zi@c@9vUjFp>tmPBp!W?_O4k%?W-K9O(>d1+`!KHjLqYgDsWLetXGbqq+AH9e%A=`I@_~q1Jd@@~{UVU9p3MT?dc1bWz z&v0U1i4gkHca#=iv8Ge!oaj`|Dk@Zvr?OjfsKmvT3Pjjzi|3KluI02fUEB+U-uq z?HU5#`*{#G!5%({ux_8zd8~1Xz#~Kd;r12LIHM{BHTKm*)wD$T{m>hNqA#J$g~yn? zIEOL4CgP*#Vpt=`!+~Mu4~_YR+7a9FOR5cCW<8knF+2Qel!6jH>exAO1kKsK?8ggD zv?9O@+GkN$^8^-NipD<{jA?g!KW@%v9{ZY0sF6^OV{8VctyqqlORivwLk{}HjVDzD z#;C!3lwaNkMo)90;>cVy4k(8s!O@^v=nF?gnV&AG7@NM2Bh|}`&|O9s4IP$Y_nHf! z{3`%*e(M1QYT@hW6VNSp5We=Q;b}H!^4=zniz&yvy(+}edhse4O@gq3t88U zmxQM$$6;9IUM!F}iWmFBu+fBN>_e)tIJgZhKE`5nB|l!eJ_i0Lp1>M^1PUTL)_u4(s z`#27&AEctwi2_vKe;T(s?!xu+m`|f%0T!AZ#UU$u9F@1jtM)VS%q(48+SdYEQde;6 z_YgewI|GknA^2w$gP?>v>^vq+SA;t7%C^V2w{HVJV||vThZXS2aYIba7ed~-hwhkbKD(K2AMCUu=~bq{53BhU%nYn zPKmyBTf~h*e)y8q{CrGx>&C{QB2@Pr!OW}$v|7)c_8Yv$YbERHR#P`d$Id1z%LVja z(TnbXwjila6MC7tm$J_pFkaLG(iUG!Q-$B-PH8cgPnMxSRs5*rT_s<&D^C(}?5$y}|5m#>UY;Kpu&E$wGW7J<)KWknC+FKJWr914L-C0w0~4 ztW0g&zM<2QB<*6{$^h-(_+h6E{hfFkU3$;pHU0C5_bthGTnt&~x|62qS(JGifd*$Y z@W_o5cuo8q=AO8Oi?ScHz3CJDC!2w(pSRI+?=uwH9z(f`B6Mz{K82;MplHWqSREFK zU%JC_-l1}o&M3w+=UZ`fmm&onTR^WQfOdE(qP3S94rjAI`SE^?J=uY)v?h@|#^Caq z#h9XZ8Ltd)plrJ+a!E<0@Lqq4E%aed_Go(elt(iD5_Ed;T9R4lL_znCk^RFk%6I0G zVU8JNrCQL*U{f-`!;#V&9a0V%#l&dFe$sxAN)IK-uTzxVWtgY>;x78cA49u60?E28 z31|H-$JF{i__}-owYv7A&HQfM|8Wg%+N(@D+xC#AlLNUV?57Kd4pUm*B9s!@gjI@W z_-uta-5B0aapvyy93HTK%yWFuQH!SEMCnp9%Ws7)qS+e-XotrX`kA9mHQW;1KJ4H;?l}kurUHQs_-J&ppXNK9JNq0%&N7FA3>!bZO-r zvbeI60@S3)KXNwp4Jgo+CDQbk?H1pcE3sM43_RWHfFEat<0QshP+OgiAzVLh8+9aw zu^{@?wUc(3KERwUw^5@1E^5>*mZgo^Q0Eg=A(i1-em)=A5Os?if7SaTLA|C zy@PjK8_+YdoiQYO7*}gNDYE-ipFjzYN)=I?cjqe_-=9REO-1N+Rst?@OhT()*?61v1Wc~3rk8EZb;DfUZ$%?8 zX!J6=TxZ^|9U?UUvnF-Q>X5|@qHk#&Eo`4l3wG+Vd$uWUyUp0gu}jEivLzXFQS|qy z3l(Xdr9MMp>c7ZOhLY?VC6%A#B)(!x$Y&e~E~cjPyY%G!Z5o^8Nt!N)De}&8k~#7c zGdd(luy!6r<*g!T=5CtHeBMishhn8~8uHBuV2+q$0(iiDvC)<6_uc5zgcd+HQ6WSzYV=EVk8VB-F zr?n6>Uq$1Mp=MlqA`&gyUC7EYke(Kvqmd~^w8FoOQryS!Dn}>N;kUMQ;-n49y3C*_ ztJT?ir$*BMIgx*Z2ZgM2p(Bw&wB%VR=@&-RIq9vG^?5sa&tF8R-54`n+M6;$+^M!Z z1p63+e9>jr!)MPYU5TQvXX63xF3xj|#l92#6u3s2>O)j$ zy2Dr6usV-wyuOp|r&_W%^q@3{UG!j!5;gGtp#ND(>iT+#9@l46`HD`OkZwR-Dhf0w z`!f!ntET<_FR9s*kEf`14)Yu@qUCqy^lhrdK!FsN+0Dix{zCdFnN3f=i|{TcJCe(t zOQbe6j(P^#FgKK+JkE%cb>4N#xt2#embFnrx*eS~@Tc5+%;8s-hY#l^qqux5&YF`% zvgx}?Cn}7z_-fGSdI)-lX5+!{lWFHEmID^pPOwszHg7Yb!ROJWXK{l*jzv(_X6Dq} z7f2pKe&j2WK>Y7(QN!#7{4USKYP)2d%F&u9Y4ZOlNQ zPmGVQnv7Z3i_mkPAgxQ4r^<0}(Jon$E^l~>3#Ei<{(vg28`h!PDJsz{p=9x8uy$U#s$&CEfLHMFps{R6r(Sk z2<2F)(lN&oWSl3GGw;Cmh0LLIx*dBu|XgkX=_y)G2#()O( zD!#$9ta}tImxrznx9~!pJq|>a;Pqu4_^9XsPL?}M<9BbMngA%f z_`XJAbtzhSe;?&L9-z#+Kz4SzLr%wD(Cxww`n_6&s)`Nh^kj~tX2jAYt z@uh>DCdO8VF zt^r|wE6TRXQ1-J6QOnntgsd{gUrz zXEiCFcBu%j;lUDm?iEL8^KxkR8ga7C)T00X?j@(tB65tlLtoRr(7wJgoOE< zK7V_~-0~7+c-(;T-n(+`*RgPD85OrCQHTD05^cDH?{@N0bkYFQBws52 zkV}&+TPXBEDYkC^hkmh*_(y^D$0ft*&!2QU3YIL77D8=BX;kp}BAFC7lE#Gh)W58T zp3nF~;Zp>88i(KEcQp+{BGgOY6vi?RFD|4$@EKLzJ>*4<*(wA!*?w6yP692MTsk$llY` z6`DfocE_o$mHnNZoK6#H6yJ%;(<#Okbx&8LGxJwbY@#*gF8hZ|Wf?2lZXF%zzJRf& zf;2&Si0y6P;Ytr7I&zl%Oak-WFi%SB!TI!f;tVp%iKU<0E>Lzc(brc6bXWNz$ughh zT;qrMUr#MQdSF0hYaD1p{2r40r%INizEskgL`4}}sUYDbtv1LbUJCncNsD4vZKSA; zdaNJlOaJa!(_qPKtWi>=o(t>@`o)c&zMDgbma{xu$X$$HS%TY{i@PLYE%h$;Bzx`_ z1afGuimvZU!7BesoH}+3tsJ%J3BNt9&-JHq2J&>jPl=XktR$z2w)FSe zLYio4MCE+%(cJk4)(MZJ(vAUKRi#d!R+&(}14q$UW9W%)7(H|Ofl>?!epOAA;%|50 zq$qyY`81%r56o%MC5#3ZoTu}HezermpXPWZkwwaFGF#e7Cz=#^BUM1VD}^Zd+iP5t zS&jJzV{zWXGkEf)A-PV|C2191@<|(`ny?H~n$k{kd}&x=Sd0cI@8XLBT{@|3PUo5y z(xba}6nLu(w_Icl;YCV}2R+LCjZ-KsJePJ%tE7?8Y|qS*tS z_pcYZT?SNlM4m)`O45U}jbz0-C`De2DVH%_hh5Jgu};RQr5bflm_Z_XTGUzn9DSRH z@S=MIF7{QW%Q_OYX;hf(=e@>;-Mx73ML9nE(ueE%gsAR&JqEID^H9$ZT>0!JR>+Hy z?mvF|7uJQ}?p+|hPv_{+*-)yhw;&hcJvOQ)-Ti0x*TX(lnC(Eq2 z$VSi($5`@Rdx=i=>e7g>IQiI5p_CbCaiaATe6;WbzEtp|yKMHI@y?s7>b_z1U*<8K zP=XHMgsCN7kQ9U&6KRY`SDFnOpizpp_S%Dr0xzPFT4=nd+ou}NmbaX{+ne%hw#hMa@^X@ zx?R!5Xjff`^9y2;i%-L%acp1pJ_ZeM24U5t1dNN3Cy6`XaOl|p-n`~SqwH@<`3x5t z9Y2a5mY%4z;WqZBMC0bs*O<@FCEij%H9{H`y}_9BI;<(;6653cO{IZ{>>3PX`G7gs zQFP-dZm|<0IP@5GYu_?ooG|&R2~alMG4TG$(1VNew8uw`lwPUPF?T)sS|miggU#4} z_csA!GOCs| zV*dCpTy)hQIp+Rr|L2Ae6(^EIj56_xrqfu!8hYcuf_}QqCy~%_d=Y7e&1#47AiTk) za%C7CC`dgmKhej$8Q-mLLOvsQrk0bT%j1XecRhRF>zPTftu!gE<}JRhuf%V>dYt}K zh(aDU;%jFC=IQ36b818AZ7_fWzqH8DQJT(~GWKBcbUO4|nckc+rfX1u0W*D(n|l;J zOEt;Yav{4nsZ)Yc1>-wc!Fgs9b@k5>Be zQ{|BntY_C*G0TSgiAgdx9Uqlm`+zsLD$otqht6vIiYv8L$^U&SdcST&KE6TJ%XrS7 zp9AoaY#W^0Op^6 z%pIv{;7TONL4zdo?O1ejMX~u@_g@oe=KsQVFA@idscT{5O*OEHi-z$wZCumI*IqY6w4KLJ^@YSIzh_$&0>GGGj zp6RwAUabV~(TOn6+Xe==t%34cS8=r~AN*VOj@!Rd6oO9w<^KKo%vGc_PHFE6tZG_` z7K?uX|AP;(%DM%_;}Ar)NkFqmAiT1W!=GulxdnYG@K4$Wb^A0hq-q~t`;Z5d1r_0s z|3NP7MmnBiEC)Ul2R!h&1+q8I1aGIMFz;kC-0Sgxe)qF*-{dOnWz1&xzeAiR#c=c2 z`*Y9oUvS$^_2EBpIpEJ;1Uf5&VR%{xXWq{TQzA}6lM(C0-}=T$3G78(#WC3J{s}6w zWWe}Q2p5SOppv%+rl|CBQ_S>0agqh*KC{I%$3A%Kd<1s6UFK%D)o}*`)`Rt{M_k%~ z4#<{IMNfqx?wE=R$gu0lt#tux7?FmvCZe!)cM50oR}wZIl!h5|BDqf;=Qz7nccHES zC6_xx6r$HDz@QMqz14}_rVddyN9^Ew{Cc@Lkj-g`UgR#%R)9;U4e(HJGI%AZLRF*> zjQ`>cI{KDiGNTv@e(=L1Re@3{$6bHCP^hB^-J68LQ6z+0_tpXmOpL*# zpp=u5YvM|eC}Fuo7c}b&;G|r6*g5kPm!T&CoeR_O{*~kSR%{i^k}JU`H+|T4e=bE^cS-6qv57fCtov;8a8q=&l!lDFW}fzPc&ky;T`s*v7+% zWfGw04KUlR3|+3$P?fKUiC(XudHyF5@YF#&F+sQ=J`wbkrl49p^M{2mLjLLb+_j{u zoYebA+&V=SBTY5s)!E*m0v@NLoFn*`-S}bH(>YZDmQ1@ zJWTuN0K1|RjQQrz#C)*?5cEFIK9}>@W;6~@t92XqEcxHA#EQA{1Uc=_xyUUWWxRz= zFg2fLdF>1_drK4dT!``ALPB7?#u32j&+w4&KPa6&!?@;|99pZWLGfKZPUG|tWJ*~< zi$n?+`s5ONT$_m=K}}rm%Mp!006ac4bquzRx($+ZLq<|#y_b8GNT#baYBohURb6-0{$3*%z{ zuW-Dx8cy2Sa}(CD#qXn8&{WjO)gS%>wJBwgb3opBqJsnTsWiaOeH%e%f)E;HXTidJ zLu0pQF*;Iy0ejaq8{d#>#oFoCIJd%^OS-FpGW?t2#RnBGNkxTH3&PMS(49M^GaX%U zD`1BjtQ!9eE7i+LsBbnzxM=SG>aQ2?KER*>q0pdM!5m9t2gxOzx+`HJG?Z9`5RJ z#&Pn0@#@HI%*g39wx1x6U)I~e^2;rp#Mp8m#LZtTVSfyI!1#LIY=f&`_m zJIEZwBHV)sVQ6-#0mQ1DjWd}Oj_vg{+sdiiR#SDqYZZjqK#d; zdDvh05roqWL43CvZhWEuLi(M?Io&%@KPnuKOikq`{{fu4 zlDUbSw&1(qQdoZd4;Mj$c#_{APwq42uA&CkC6t24+6%BNPYXTT;$Y;b6j#aIq~H6Z zz@%*!Je@iW{ZF02dTuymLMhliQw75-DcsodX()YM0sP|j8|$t83}20uLE?so@qbd$ zI7uc7DuxA(PhRSU!Ru=wrY_dFtBDT_Yv;q?ARpt|#ktH8T?Rpc1;&ZuU3frr9{%21 z%*|o^yB~J;?B_!{zGYEZS+0q;db!-4C%R~xwgS#NJm5+_%e0CLi) z+=WHe#yu`)P(L#o)avA6#hg0W8#Et2Xiw&@E{H?h``Xy=*KQoU?gi>(d1J1XFn8%& z4$HD$1CuCsW8-dNyx`yfcMYdQ#PA?2NIeKgO3!e0m+~RNR~n3JH6eeMH9C7oz|X+d z#>JRGsu z%lu$#xcK2wI9)XdxDW---B=s{2+xMatJlH5V+H76ErhoY#~O#pWaHla7HB)T!T2+K zf6`TW7*OtIJosnDJGrmD`x;Ab-t&N%KM(A@pl`)q^k#FWB(DM<)$8Ggs z=OT%3cm^o6ULS9D{eZLi3-RABBlNU*0AEK4Jrjs#sy;?;pDh)jA(gUR*YgcL+OYO2 z-v1kf?g&zN;(e;ojmR%?O5?qOIr=(!ofhMq#QozK` zyV3l=F4)&T6>ser!&$+h&lKbtau?835ajd-2u*J)HSQ2dyWM!%4fAqwJMSk*EaZBrz1+c$eOP3OZ)iBh z!#>3?=}Gt_IvF?IIFBti#o2ojhi|7=W0lBMT)A`$W=vrF;+g(<+b@=N#_nMI(_^^5 zgLS1_8Rxk|23N|tqV2~#lnZa(8Ac|e%mkON#V^X)m(`_ zi}Mi1wc^^zB4p&sdM6(3u(9+Z?2|Z#(xF?>)bA=9iM2xMTm_UlvKp(UC7ByjqpI`YJBQkh}$-#;={kp-@IN3 zN6zoV!=8S)Unm(942rPQA{O|2(_nbdW3c^r0tIuP;v2S0Gn+Gxxq=PZcjy{oG}4Q; z66AQso%pv4Z>m1HKl#C(8+GwJ*7foXUR~0P5w7-IQH<^zvGQJ{y@WGa3#v@(8_ygL5jBT-m zUX=`j(*ZT+$>7J%oZno?nJr-bCW=cG%>lvTAb4qZ5rp3=;(_%X+TPuVJ5>5$XKD(V z%{9P1?NX?@sS|u^3H|<4!S5wmz-{aYt;}?g4@`xGO|QUr_h~#6d>B_w%)-knF2bVr z+i-GfAY3jEg}&pGAPEa0I7<~n_xONuQ!6a<=fk5Gsrad~9L9G^!1@KMP`dC0tSVMT zd7oUEE}@GRwZRbGune9)4g$XwAHXuN6s$KC!h=%;SB?h5K>tKMni>v=_VB~7l`g0% zRzUYbeb5j1%>8ck1D(}zARc)QBID#S+ z#|xM@qy&Do+rfx*KR2y158Sl;p~WW@e&)Y`#m`vZMRhs8>wN{?QRi6q#2A(f-GRG# z|G4oMPPoZ$3p%nnd?w2;f1A|?0`sp!wV)g(>e@rD%|CAKTL&=Qn+<$3*Mj@m%_uCv z7~-y1pzTQ}=t*wGO)$ zOpH4Y!&2#EL|4 zek_Zb4f$NdqD|nsX*Ty?*D7#3qgy7Q)nP zy;e+=<}nkaEiih2GA_oWK`&Y5sON$QZJnZzS+#p*?y_xZxHCgA%Bg5*5H~}QRUS~dG7b7lC*zsMxwwJ#I4v6_Y1%>$ z{K~R(q2Z0N<+>Gg)UAMUIdzCLWlo6c=iy9yEa)U&hO-%)(BC@)B88^0Y+XGZ{1y%C z`&uDCKmvOh*ZL>p=bUU}_mUIi@c#2Y*cz#ZC9QkmE9?D*M-9UCr#fUjxezu!(uS1c zTTtk=3pTS(vA%3Lc+|#W=z&kL;Ou>%PtO=Dtp?UC_rlDGJ0KTe4TCcNuwcI_#tLS_ zkkJ!JJ`@4ZrYr`DWDnSxw+=taHG$Z^FYt8qA2_5%fX>0|ptIhZdB;LQW-K3K%N^iP zOBqBL$3pDrb;x@80M?hVdqC&|Z26bQ9nbQ`e#`M_*9vI;^%d}ivgd#@_8uO(4Tn0K z!Lvu56K6A$C$7~X{`CuL4TXT%&c|TW&zvGx_)xs`4a_bNfgv5nIqIH>tyL1X))dq)uCd>ykAno7)V>?7zd1a{@daw-Fkq{^ove zb%Mf4{J6sI3Ggh!@s>z0@Yak&$(FwmdAtq$(@(&!j60y9Er%z~g23o*7kt>uylgx3 z;6JGzhza$?z%6I+;*=0vVG;w2^)|zCqZMHJ`379o`VEu#&qGJ81YU|4#lKG9!2jVw z+Y!Fhj^urV8~|zeS#lgySo7F zx0;9%CH0V#Yz=|?H-P5jsc_?|Fw`}OV)uh3pg3R;dmH6peKH46C8xo(RogM<*##K2 z*GIvuZ}2s4B2Hbq8^6?BfY#s%c=xIsjMv72{Zwyo%a?%ot4C0@vJ8HuOofCsKj7k$ zn{f2ePg6laM7#NC%-Sx^itnClq`1&C-F#)FhjDd+MnQ%}}0)=_~@Oio}uGwsW zAFc}F4!bf)l~{;(E;m4jSS?(b5(gXW?tyAXJNz*1gWYKpaq$RqyXi)Qv%@*~n$F&{ zlo@E($=qscz0Ap=3p48L;cXd#Mu`BVxM;vk$xR?x5CN?Z#Bpw#9~3+)fboA`aV3H) zaU=7eEx)V>KO0B6V#gHdJ0T0dmivI;evXwSL> z6$ctwA3q$jfA?{VE#08EbOxsE$_2U9Sh&G?f=q@s%o*PaUl_9y@;kYubAqAi(L!7q zErR<#z699@ec)GJ3*FI^A%uBt)rRBYaQ;NJn#sHx%$s4-kOQ7^h>v;yahsz+Ny@uCSA_E2{D7CKIRr1-IceoQ(`dr#ZZ z@nuT%(WMkG|9FLkk+F3Ai4ScvT11(i-Q=?>ot7MkptBls^hx0@F1Rv`)z_Ai$$1fa zu=NAl4N247R#B>0o`)@)Z`1NCX>^^|Zy0QtTwj@!&){R%=E6 zjFJ+vK9DvmuEr!T~Noz+WcU#3%Fmk{qj zbtKhvB+w1B5SEWzPiaF!q;^)2XSn7O&6w9k)9=>MyAalg0S}UPEyvFE8>qN96(5IA z;(ca3-N$qM$TowImdL+HpL3-cH++S>g`9~?RUn7Q7wPkfg>-Te%j@_|;r%(L!;51* zc(F;9q~}^o8Zp77CTz?cnQ}Dq{2D409@Icx)6iCsXYiYDHlgvN) zk(t9vGQ3nuGj#6KlKqBsVT&!TZCp&sX;Ku!a)p*|i)eJ_0n!tbK=YC+L<|ML-=*E$E6&M&Xley9-Q}KdyjQ3wnf%l#1w#sdqQDFznlh$LUL_rP*yag+;EL#xz7J_^eMNslo2a>iWf&O0nO{4GHrGni zYfL7KInh)m%cDEfrcn={0=e8AMZ9TGi}nf8r;<0gX+WQD)<}^?0)OU+5hL#&CEn4~HH04fNi|4-Cz+H(ih)J+UYF%Vlj_i3 z`w~h>X^}_58+Dj{NH2hAOcaN`x^{6(HV1YW=LgUyItxR`xGk{@>)h|*KPkpNYLTFLH#Y-VJL zn_igAr#-Q~Fk+ns5v%%v!NI|SpO5#$U~LHWgb{4XjK4W&!j?5G`$ zy?fwEtU123*pF6~`QV+efvOO%|A=yZLoRW;QEd}tkk|3^IDTnF@7h%QB19;&3 zQ{2C~62g{6!@K}Dj9IrCYugR5NGJ=Z7xh6pr;H!P4q?`m8Cb+z_qvB1aZF|mber5D z;*$a1+t34l&u_)mW*g8s@EcGnAD%QEgyTCdg3<{Ybna8d4UbjOE+h&{dP3ph!KV3Y{UgXwN#+S7w)BZQ%~wZEzHo-{xY1t2ziK&xDGo%kXx- z0y;IxVEfxmIB{7yr2LXVnHw$Oq&xw&K_Aag`wqWFZi4-&BEC4a2*WMjK>E&Fkk7vi z_4n)HOZh{XQTYU>DJY_jViIhKor}^rPVi{-DY&V|!FMxJ=6OGh7bFtUe!(ngfy1y) z+X!CjxnOu#E~@>C#(8x$@LujR9IFq&3YiG>%qqcV?M2k0nWw^ z9(T$ze@HZh=bnM$w`|ssrGm@*LvdS1J=V(B!)$}eIIFA{Cfci`ciUUY%-Mryo3=rj zWF4H=uY;YBzd@6?8)`W8Ls?A~oO$*TT6QhMnuq-$_S+d&?-9m8?O$;JsyGH#O5yjt zb5S68ExKv1!Qp-zuohkizo6me9VhZ9A8SmS1e+EZYBD2wTO;ez2$*Sup zB^EuS-s@SEFu=kJOYYF@osrajDTpR6Gor%%2CO?Xi?*@MT%q|e`l|fFv1n(itgxpS z_2y)7<~b_6HRI+dwb;Fdom1mkc5s0ft+^vi#=6sKzwS@QY0=V6WQvVl%rb)`R^tQ7G= zmkvjX@Jh`7=g#;^LIStRFgA_u?+K@6I%(*4FCK$#v98seF_K%5N;|!7(|gs=*jw9& zOK*R|h-aB}tSgcNPFo2B!ev4@qI;42{56W|5NR4AW;13d-6507(X5l4MA~OU_Fe|m@bwlA94ecoo6V(feT4y(<_uwl{)%1Yi%X~z$+obf$8bMYFU^k_ldE+yKy z$%H~ysnCaeBN#L90ov);V&UEb@?Dln#=A~a_-sKstSmww-JW6ks4^8FlA!M?Qz^is zmTpH~B!|WjYT9;>9#uxtIL4l}Fndhl@}*=kFM-ZK7oll(x3MBcin>?ipyic|__=l} zg&TEa^09cft+ zHhx{0n6Q-E&S_DC=X5%D(v=ESZs6Xod+5^7ys8F#Ypb8>X0}hja$UT=B!}Ul~77h4uem-J_PoJgQ!HA8mNT ztV1DB7evd@xcUJKFBYey)mF6XpAyM`@*#ubB)pV%hs|iu;;~={64kY$exno;nX(R} zofGh!&?-z{X^#&lT4DL{S`1d+LlX1Xk<5ceQVbf!z~A#J%ee&8MiSAw>MQ>1?!g;J zl<3y16>YT+R3~3+cq0eN-frMi*yJW!J3&bbLPPFG-k065~g&qRzBiDB*h$1&S4M*7*u_eq@eP*RNuyi7OuO zx`;VvFXMyU>sY-z3?n0UV0vIZR?4TL_{1VK5WI@H`Zjpbdn*QKEWoArgV0v%EH*#* zfR`2u(xwSoG;7^LN;OW!P<3INXS{;lt6$(F#eDQwc@n)OK4F^P8&p@EPSe*aV~dwA zdiO-3P%QHR$Xv(0>EZZUas{?Foxr-zL^L>3jY038;(*gFboyt23JsB{Jm!QxvJ!Mi zL66NmnDd~6og>b%zNG(TvewT<#UL?~J|Ih@*}^#ci~!0`W1ez$Ub^DNc*`@rQPyiW zPWdB&+wd_|n=`g$-g#V{eI2D{1R_`Z029Xh+%F7v;&pF2@w7k%}#esgsDpSdXKNyrZ4=0MFajD`M z(&a*-9Qh(0mJ>SrcfMu&l(Vz&6LpD-ElQGTuSU_?4muQxHIUU(GmEM$? zko2oNj>gQl4gNW zadE?7+i++U?4g}adueXb2D)}|0crkFr6DcOv~G+ljk#__BX@FNQHvIhdTmSNUM;7W z0lahfZy!4Kd3Rb+rfIW{$R}bN1#S(Z2qBT|dmUg0COp+WHp;Z4>Fx1$R1GdxBZ16@fc&w#o)k6q&|_hp@vr9K}S%H3T}h38u7AA3sIQb5N4Ig_chCcSj^ zqLOR00hY>F>aB!#(lwEk%w&d2nljRXE5`Li3D>JvRzol8<5O=;RbMmu}v(YUkc z=vGE7{YptADepdb6dTYW-UIJ4pH41A7SXPuvb0-#EDf&Eq^WzCQ-Rb4oUiLg+4nx< z?bS^D@_2#Dnil-AR-hfTwW(51hh)Vs!cg}!E-d{IyF~Joxb+o&TaBWsBcCDb$iHFY zPw`SY9!dqt7|{IyPG=)2d-h365mLx?-YHTuI7T1pa;TfHOB%BksrCL0`Ytn@L<#H2 z(`XspS=o*)lG0=}QI_88m118)B5F?FLFud$us9Ts7U#?Oyu^@(kLG@oHBRK`ln<%2 zYIL|i!&lp}6dcou=L1GkbM+pTr|N+#@&^#|padUy?^$-^SBzdIMMvH=BFAbHoy{6g+x_x5x48(?8V_*$ z?Hz1TZ-)W*dE5^PL+jyGB*eW(*vDx8^N*zbxCm-;_zT_MYp~Jz0CCOv@!Su$ zV%=QMe~Tb}A8&dca)4ZPqbN1~1VzWXk+$1HdS9PJ?#mX@6sg7ZGSHT~?y8V>raGyA zGo$`17~SuuPbw#ji9!k~G4Tsc&^}JO)2gUAu$iRdFOczWKa#g{Aphny)D&e+0Rfh@ z{PI-#Jf$Aj<8nFE>=a~Xi;JG#t0vPcilU`Q|M5BD3);W-r=w%8;iTdfh|f=k_+gXbFcZDMM@D8EeDw-){L@q`;w90G+Ewtu)@QeIta4hc!4=%zN&Xbv} zs7;&rOxyp-MjEztA<0bKOy6R5(A}?RAiF;nvi;JL^!gl+B?}D)OR@Icf7ti* zHf}sWL1R<6fA4xWN!cifa!ZDaN;3wFo<7}9-lu)Y#?FI2JR3&eu4~Zl3MG=#ImG); z!KAImopGyl>ClwrG(MC_CNqorm1fYX)ExTJ$=!wVB{ZGy#fLA6fqh^k95!(l#V2VR ze`N^Sn>8Zy!(=MZHKx)9y5y9?*ev=;kj_aeiyHcWTeN~a} z?pahDWlC-?W9jL%(PZbv_kY``(HhRgif)4}F~4PxSQMLz*|Bn4FTEY1;cKbRbuq zdOC;_d@HHh@C4}&eLx3SFCqRIQ`@Q)G=1nXWLDImi1*gI`1*UBNfCYBkxuHo7yMY` z4Xh_##R{#%v{j7v*5Y_;bx#=0nsA!-JUdLMEcz%>qnW;p=qK76+erJ4v{L@rI@14U z%-6S!OxJSX^E^rFI9iVx*ErXK^8|91EhJ-&_2d#dlpLms=+3eU3!4~8oGN+;W+bB}ty~0&Kq%gsmULM*(rN!Gx_r^{N`WsG=rvzzRHK0ZzjxLE3xSBKx(S% z=Dw|Wh#F={YwDNKRRuFTKXWF1RM<+s>5PW0I7;u-g6ML22pyJhL%ntZT2~IGX6L`q zTA@ze&MGAO8A|D9)9HG0D%}d4M!WrX(VFj0BoiT`_L4Q!8Z(Zt%!9gHchiX(hv}T* z3%D5G#htAmaldOUXOu~Ej>9Zkcjz4|_1p02P8%%Exo0tO87a{QS~mC`dImg$FgpXE zzn-Uym5F4EDB7WYj<)8VqCM8R^k7&P4R*>P<6W(E>UTUw4n2xEk2ly{*^W=PdC04& zMrrpoY);F?k5lK78NzuN2S$^!-dtLyCqbti#3+HkpL6DNens0~43kYmdM^CeN@o{U>Bwd+^;Vg;bt! z$7oJ6?R%9^3qswZx^C>+uDnj{}bObM}MA}bFoNBs&;E|Id9U=+4 z%fn$gR2TPGN`UwpOL1`otIfy&vk8M(e;ep7@Q0MuZb*$U$Dj>qkTwWGy!9C{p0!UM z_=x?<$YuIQpV3->4IR1nA)1uMMui!`WT!lOP8?%*joMlDuygFa;#m~OK1a*MAiP;4 zK{-*6F-g&m+U|s6_oeAjw+qK}u6H-Tr%Cl|BkP!G5oM4*tj zK&<*U(7gUPsPfxDI^7hAmzJ0D<6{gQybogbBvtS)6rP!kN7sKrxb?6<(nqIZTdXg< z_f#N7cRI#)m+(ABB<~3hrm~=OkUshtdp8V*u}KNDpLT=!%gsehivfzJ&%>Ugc!X}A zhNRg6=-8;j-8GeLuYVnD?Mp)#&yQO6MnH{sFded$=zEO|X%E~FqkYlH^{T=2-!7PT zb_r*1so-Uc5gjhuM{m_kNkugdcJqSK^?nZy92TgMuS>p;+D9)2N#NbAKBjJ?iA}L> zEMwg-<~~6l-%~3wDE>a;rk=;BeRuK6;3778Rlw^_7@v<)&~Iiw+IO$uEX3oGd$NGK ze@an#O$BTwHshw_Z?wK2uK281P8f^Ha0>^Ft*ejmhSoumFvo8#S zSbhTwmW)8bcHs8Ocu4WR2_=m!D0{C2uK?~NKbp;UmxrP7zwv>g}q zFQK&RC|-B&1bjzegyLFs7|P?h&Hz|MWHX~2Rm|NK0F?{Mm=(VWGczVb?~VaJ&OVMk z$qrba?t-r`kKyO5+jy~HF|cnWWK<`@ReUH`@3-S?C3!5LcM-RKbU`w%Kh<2yM@IiH zjFP{Mt&e3WzH}iESxNZ@|$CDwsZ;l4A)}xQDFAW z6|h`h#B$R0;P=2Ec4xL>$p>$w8a-rb$Hrp(IbFyE)#9~FFhv)TBrRF0u9 z<*;hF2FheKFzBl`)-~v2kG~EoB0}(o{l~n1RI^c~Ijo{)HCngyv2=~$xO;s(y?xe< z{-&*Pzg*4otjvLob|%fwDJ&f&5E0Q24qBW;wdN1%uiixO6)lWZ=DxBt&H$Iur%O85 z;c6jIU4aXP{B$p&se7F;SJ#I*rVe2vF3U5sMPaOXyD2-j*ouiy7t!KhjwCLx9GaQAH#j%2@aiWm!H&4*d3lP%uW(!@7j>1BVEWvPtqfq<9 zP)IaiDd^}}3t~w+f_334?wgz{Xm^W*@M#r-XIrLVXq+o}-Q)Ye zBzE?Bqp(Twt>AP_mK1brki)qzlip+t6?@!-&cp=4df!RGeUXDu(6~-GJ9sW!bhaUU zk|Cs}d<1pv2qAclpI|#-0CSC>z-Bj%Wf2o6A?)l!cIz4MFzTlX<#RR)ixaGbcMpLZ zJAni@ zSDJ9>YostzEL!MlX=KaK9A^btoh-$>1nWN4W3^uklvSmed{v?#=aMf>-ZF)PGxUhv z6iDqrys+LbK}hObD6Ff^WXUsDGEzCpR%HbW>eGS+J$oOah&wp#x1JF!mL>~-B%Fnb zIgf=czRLy4rhd%&%>&`fpbNt0*^+F@jaq?ONHc|*jqn$X#u5K4%${5+NCf>8G>uLO zO7g|R3h8r#Y13yRd|fog4);Uw&EpVh$FdP&pM_zTQ(4eKV(C6wY;?N=`)IR9D0SE@ zl&tU+9zEn{z=|xkYHq4PgHH;F?|TW^HI>2Xb{5ZiZT5!CBo{6 zc%fKso*=O+QmB!R5$fc-*`FabOke#SV=G^<*?FJYrk)GTLcCZwd2^ny_+N=o{wP-1 zH#14dRPz$PnoB~dYb?f^Xy8^#iV(fCQkax}P?+-W22`hT_GRU9{H|#bOr0MJfg4{6 z-Z4Hx-ZWd`uV}6yeLhrZu}&7O54s86{tF@NXN*R>(P)ee=GKLo?9-1I!jo^~5pq)= zy}rYca63atu}c!(?ARtOUMdSWp2sp9_LG&CZWdmhohMBDvQ+pHlflk5zhT}{yIJGR zW+7@xoUn3qn4r6JAagxaDI5}lg;x?wnfsyvEGL1nw#rau_^q6!k8Wq9e9Gwi$b0ne zSRq~1x1*ct5p;jnUUC_l#aVzKpcm4g{<__zM9#7`wD^zqZ&4792-Fpg_@^p*uiOo_ z8$+l#c`!|TIgQGVC29CxW3uh%4|Q{%kMh`trH>AhZHf<_-Qq=}nETkFe*rIox>38) z3|s|)ziB&gaMES!>ODc)H=9XG)|{&Q@qOU@T+&)=M~3^-NMU!TPza$NwxyP_#`573y z@R_0GBs{MA;dRFiKFiG^cS~d1``w1}+(YT@(*Nk8t%9iF#8-5#FF~b*0*#NKO0ybd zscLQ+KK=EDRP_07zmsIrA zkQ2GIX^Qf%3=>VhcM*3l2BY2VB2Ij)=ic?d^q#Yv>wOH#s%Qe~?#iVx4S%R8Y=%f( z%UAUNtRx-m-b~Bn_tWDECD_?AlvGW}(1bfZkh8F)l3ntYt8fp$6g#lee-Igrd4LzE zUZSq2mwRj{kV0S|wBqgQ>8vz*Kjbt&@8c}x*WOTfeS(BZ8Wf z5SxkgpEKZkHU+CPEC@03l>E6CTN+F_FMK}jZ1$pocQhzYtpRSqV`%M1Bg*_cjovQt zr-Gkj=t8p|x!%{M?XtJ9Tax>IFB#C$vn|wi{Q@P?A3C|Kkmf&@5bYDji$uRaL+y|v z&sqtz)We_V%rB#XZ|mv5hl6N8XMs4r_aNm(1-LUa32QAY@whk+(+@`AD|gf-d#~b$S}2suAPqb*~nUyL_9T z2N|L~Hx)ZJJADx5odeQ{t_2(F&K4e6v@oV^tWo9&;eZUk+AngITwIGni7W*JuUl)_dcd)mY7Glca)7W8_g;A1WNWb!z>836~nbHaj*VBjU zF%V8n;d^~v4+rC+-!<+xSq{lX zN7;boQ&CGdSXG)L_XCXwmJi2-{0q=4yv8z}CR5GYTP(xuD4XLT0r#u}xVUx;_C+Wm z^HeUotW&|}y-Z;BmP6q#JYp+mj70RAcbq}{lC5d)VP^A6+0oSxS;3cE%x&u{HsJLj zxZWva_X;AJIG=+Zb~Z8Ok4Emj$L!IbS$HoV4vD3yP*M5`iNSI(@2kOo{_j~*X+JDj zF$!A~b+OoVE214vq1{0b3eJOJ(y57zGVV^!(8LsF8QgORY__c7YZ7m~nLilfVdvPN zI&seKC}zuw-Z2L?dE}_gf?0qz#+o|tepo-u-1C5?#dNVNYz1Ctona?s*C0nekm1oq z_^qDLU38kX%G(%E&rO4UC0|pn&fyNLW|UjM;V$idc%Bx3i<{(d?Z^n+cx#E9b^YPL zvJ78x_fa>0A18hMz`x}hXl(9}V_VbM<=z);)_^va7qATvkD6m+X)cVp_wVfT01W(D z$TC!l*z%dui2Jsgm5;xKgOA#A?}{$`%zTF>$bs<`f(^-VS#}&4G`@=6YJ$(u*e? z$(h5@Hk2@X*abwH$>EuSF)TjY!o*4qXXroHc%+Yo^ZkLHmsZ32lo|H>{$jlwR=~Gn z0NmW(u}jVg`5bs38Gyr|W@2E67B(3@VMl8mvFpw_ zTt67kT|?ejc2FBV^Cw~H;;(Gt!u?nzXN}u(KM}S}3eh8`(AyUS=qb;Ti^MpGv%?!P zaiPfkGKwvoNbJlhZ8mz=E{yl*?7Xj_i|ITgp3Ge$!6p=aZ5!Oi2EoZH1&WDknB*0S zHIJh($JT>QeYK9IghaBaJrSgQ@Hcu}KcIB{6n=hWf|_wsd~Mjks_phN+ofS_--y33 zk@|o_S2E#54)Y=8GprMmVOV?=ViUxet7VrkOJ@R;y?UQ*@*;M6atibRF_Gyz`?LO! zLRsFfL<}4J1rN{P$DDzOXy5%ny4ARbmK>0y-Ot1+xV;*8!`0C=_yZfXLkv}h<=CMy zhQ7y|k!Py``Rjdx^XYc*k`ifDe1yVyDe@b10YjqOadpl(a&%wFc1^KjI<0e=YQ{|1 zaj$a3|L)u9BiL!Dg~FXHvB|-N%{f1a4OY@+pT{;p!#@%(qo2TxGaJY4N#^s3FKsU_ zLT%u1`eOM40g?%Ly88Hx<6yeV&kbbHiqriF4UCyN7Z2-0kkkJ+RrX|&QvYlk zaK4Up-95#I>m-7rV-05dNpUUS>qZ4VHXx#7-;{pG9uf7I?gqp@lLUG>9|NPHGyUSH=Xhe<g%MwP# zFqu{5?C$b>ERVg6+_N>Xk$wT|-xc`qDiL1Y9T;UlguK;{B6Ex-^=sKm?(NCkk1!dh zJIPP3?!ssbtG+;;;W;WeH(gs;79#M~t6`VJ2T*(@aEriXTUfyecj_K{i z82sWgj3+pvc|CXDtZ{`Icdt9d>C)$&LutFZ1Z+CSVZ@6g@R8pS@6Zs0J+tII`942oj;j))IfyqdVuuCW_;9Xq*x_r-I6|OJy&0|2BjD5(#F5cCwT$ty?hWl=LNI5euE7< z{);tQeP?(0bG`BLL(CGmubVrz$JKhniD&Zc@4aLjr`0m2QxWX<#0B{E*MhlS})!nsYq6bdAQ#%A(JtcAR z*l+}N7@Y?=&jz zG~u~bJWgyYf@k{}8opDT7Orq)$=**)mrvP%B@P=9GGQ!IAGfibk|kIm_~CK-BN%5N zL`T3>lt)kJdyO9GJnx99x7rYz84sUM8^nIjL^aO_%hqmV3k2>y?5$O}7lF)v{EOJth?Q*eMP0Ty@;FYuDMzGE9DX}!vwIHaEZt-aa~wJz zA7hW=p~EXAa@J1fkx;xG9)j)B?@%RGjUkH46gvAItlrJ03YU>|Ro5Ot&fYlEGX}>G z$w3`I7^|F!4J|q-YuN_>=wWca{fYJ8Qo_kUk(-x_uBsS>1tei}y(9ajeUJHy zUt=G6XP|LaH){@h%FL=Rvd!Fa`Fh$_)_X0Rtq(cEI?e4_&fx=W*^r+?&-X{{vod!* zOwDAf4J$EK#5pC4rr^a}K4%^{jDu5x5O-ufBv$`o>$roFO!^`7`C#Nul*0FGbC{K9 z4!m8g@yJ<=Jo2Q`SuhEs)?I;-x*b~Ul#rnIh83A-f>})J1r2p{5jh> zsXvYUBOpXlnHn~p;rXna%wYK%H1udYW2~Rhq%-_MK*<&DXMxzwfZa^S86^ zEefnUBa79)GQ>x}6bva_&igC@Y#i^Qo_Qt(Py3PV&ws;kUvUu&o4bQ4>aAnFIu95R znqWm=F*~95iaE`YB&&l(u$z+slac#a`af%SGbxY#*7ati#zwM-vd7tk&-&&e7gMn}RE>9zcurDPh3}2)Ac->xdR|CCE@cV6Dhu#w@nF_0 zi_f+hEL&EcLTBpI>a4jm&Tk0RLr0-B&IxjL(QF6riR46GW%o3dS+(X7w!Y>NBfi#I z*$t|aT|no)rZZ;tn=N}h27`K1S={9xY~#nJxMsbK4Oigvf<-d>J8dnW&vRHy&{=*j zZDxz5ig>440p@-|>`2Wirek-AeK;S)CZA4Yv9V{FX>}Ug|09rH6+K`FYR8c6an4~q zr^|W1M=4fvEj=HUPsZ{2uy093!jk*=eEbrOI?q5(hvz7RWO!ya((S{U ztII_u?V1Mhr*dR;9kh4TISldpi>j~Ml%N^MMo!$w4sXw4gO^QZYxkUB!EJjO{g3A^ z8-+aiNZh7a=C2sU0-M75_hL7)R;FO_+H*Mn-I})T+C$x1(R4fdGP5zuVoR#evRfA- zFyZJa%o&^nnRDK#-1`|h4Kfr|E05@dCYai8j;gUUXyqp#dQfwezI`;o!*_LT^Xf5( zcpA!#)Qg$;#T3?Wx)O=kttSsN7ZS8v@uqnojrJQsKi9uyZi5tXx_2Gy-LFw&)*Mp4 zc!L%snz2t#XW5JWVz|U-fVs7OjI+GqwL1`pujOKG@F2=NR>amO#e++~yeoWvF_yD8EN!$pZfshN?L$?NE2V;cNh(lO zQh~hkMhsUlgWQUts1MC&dZ|X(5`U0sYiBUwK{j(=uY;9`1mwD|#fZ=S(Rp_%lt=Bt zTKPb{?YWG;b+@r}?J?dbEQQ0UY7AbmiY=-RVT!7^SYq%BY`GnZ<#tDr^|X@tZ_&i! z_)W;w;taO^Tlm@gC8nD{1|5GDp`yJUt1{0*Dv_@h;tR39W-aqN62@LigfYX@`k2+~ zhX|PzoIA*{uKyT#IGVvZ_${-aUCEltQxF&;uvIDRSo^HY%!7A5PtF`j^QC0Td36%J ze_w>Lq7k)f52Tsz#OT0Ac`{n$&wNWVSpUNFY(AfjMrH3{Ub};s*5d%Ck-mrhry0P? z&z;78?MrA)e1Ob*S6RV8&db|65oOJG(24SZn_ku2^%dS0TvGYDp*17Tl%5EI8cV-e4~wdcx`fy59xZPSB>ll~NbWI3JgP@-ji zr?Hzwz<>LG)Vc^{%hxcuSCq(TYABiu!}#nopXWM;(%3;-q*%x^u48V&e{m%04(vl# z(s@XXjzibHBn0lwz`kqeu|a~n5O{WD$naDcs2)Uk!XR?5oJPi}x|A7qf#fPq(cVkz zNlQVN?xYW-eqHU@I9t=?#Ny$5R;giQwNcXuS<3S;od@V-YjN=&6^%}=I)3CttBA7`UMmh&! zl13oLwMAp)jU&86$vbh`m#O}547nUNp_J|=WV=R*eAW-4M~}zS&|OA!<&-8Z<1WR~ z4=Zv1S}Yu{{fC(^tKq%#Ih;b0>HDQ)ba!DG?G-D-pa=2r-YT6|Bq|<-kB}zyB@$9dj2ioOx!@{X1gkmp~LblHTMF z;@ymHh&S2)|MfH4`wP!Je&Bd6=YIdkS(QPZcrx%e8s;>@NV5UYme)h1d>+m+PjGJ5 zIW)Caqv14nMQB|{lPO>C?Ja;$<`Mkfu@gtklkq-J6Dcy*_|VbO$_LXiSvlx&1+b7&I>oL!BF8`0Nxm`|-X zlXUpC+@VIXR{F>f*21z|TIkf!p^KhP7|1<4yJc$dXG;ji`$a(Ez$_ZH)Rr60VWKn!rJqfkyF-%p-;GzW_>EE zZ6BjV#G>hI4sP#H19+u&7b%>bwAj>z_I(kj@Yd0^uebvd8|$GjaS4f?{M=#P zPR<$*!hz0}H1z65+8ydd-PIv*$Ie;{HUplx*6=&mhfqO{S){3%H^54JVe@!q2z>f5Z~;=kF4P z@BfXgY;pSIBta6UnWTIwo}!sIMJntfCBNwu$Yzjs*?R~~9^bL_)$BgFL zQmyx=0fX-F%=lfr>I#O{_+xN;n~6%NPK-KwgY<&0)AComuXS@6jhD2hdsm(#yl^ZX z@El9ydWMqA%f;leKZ8E^(naLo%dW%J_Rb zLrjihbND`Q%5z9ezCf49@caC9AW8UkV+YUZ7nE_{!Y)6UR`0>?cYBchobznm?qUx= zk4>KGjgApESnRR@{^i%<^6wCCS#F0!x+%FQ{lW*6pZI$)3bs4XVPz)oaC^Hi@+} z-=(i9wYrfK4@J`D@p<^N{s9JlsKAl8iAc<~L21Z3e($Qo%3hXxCi`Paln+1u8H%CD z6tL*VSL};U#nIb4v2MdMET2CfBfJ=z7VN=1cSRgE6^E5KXDEniAya-D)Q==lZ%hQ8 zzIT$E^29{9CcLLL4R>kS<&|_n+@5+{oXGNqA~{SSMBT6C$ZdHreRm6{GgY2+rehIm zR;xqyvlPCq4x@K(gDJmp1?@9PrE?2H=y8EBSvKD$YEDMk%Va>iGA6Mh4+?A(Pg=)aTjfTDZTCgT%OI`Zee# z@1AGTn|b=UcR&?cdnNcR?T=oU09;pchQa3ynq<6-W?bA)nV09`W2XX&=bK_=&s2(@ zI*IJB%%BTHX3+XDBl2=ECM@O7<0&^NM&cCZk1C`$f3K4C<9vGf^AVn`D}~GT7wxNso`!LI&z?qY$>Y%l&O86ozWCU+Ner)BK~$_&ax@$*c)?Kpv< z6%Dv8e-bO4PVzmr5X?-h#9a*^I8HXl>%zr2B#cJKzGT$=j>N;(3$S;}Md-HU*kBrq zYaeIR#UI?`ttC$%I+oF+Ggg$%zu|}ZUh_op6==Jwj!l{4>4(y2s&%WzK-VlBlju*+ zitj>iKF>vNv_yCdKUa(trx>^6m}r)cAYLXKxk8HWpgJhE7Wm0xyatK%>nFH=Cq@CbsnA6c&SA-Z~u-W`pi z?dj|2nokBjv`Zt&l0cH)Yk@IjjM{C62rJZpeE$Zvq~<*Psjd#g_u??FlEM@F33M)L zI7vqiBFSl2kUZ}cH0_*W?5IGIm&IxM=GWXEBnQz!EgU*1pwQ|SZt9*#_xcMkS4n_n z?FQ&7E8)Sy4vhQz3di~SO|4}3^}uC% zC2uULKa35-qT!iy9HDjl@tV8Q6vPW5`NtnQkAul$Y&c2V&7sm`14VZIxs!hAZE8E^ zMGjuOseIsc3NdD6?4d_i;RY=jHg>0~6~GBQ@$i_S+efa`aXBzFS51k<&@{)Hup-3Wd_Kz3ATU zfi3N})N1TVR*$u4ZI7%dS6fQt?01*WJPD+;=CdiZY97Bf!zus6D9U}zf9)tE(fQXD zsF6dnpXid)7@k%2tAfs)5uyvDq(#byOUdzYJ@OrH;mqx;sG4&Qr4K?-+_@i(Z#jFx z}Q=Wepi1;i#^BSY;-vOgC>2^&o4(>`4a`c+ODydyy&A1P8hmh30(ppx^>q^H(` z#N?}(Rap(k#zK0cmrLiruA=)*6_~7@v9>js^EBf;&6y_q8NF^+R`1_F|^9DjnB5cGk9||(FaqiE1y9MO%rJ_ zceg6bt5VYBx7buUfOf13Bd3Q66xFhuR&m$Poc^PzW$07DM?`Dyk0RToYP31Y(X$Rs zy6XB4GD^8v(RcxelDP-hbu3*86QkO+4a4qtxb-)dej&c8+&U z6P-z=VG(yZJ;qheE)H8A#ycDns7=Y7UXJGO)2C->mfbvR)zqgSMki=eMj}Zz>?X$x zGNgTMAT_@4gvI7?T5I4>k*eGm#ouu?*KVV5ass;7h)ArT0hxa4!|2#I_4p^E?fp4BoU`8saF-MRw_I+7{L}e7kG7S1jtf-Pp-wUe z(iFiv&8;2Wqjji?b8s%9XMGYa%}=J_yY`f>R!4N9i1fK1aKMRHG8rx z`ouU2m^+bbbOyrpj23>pjli#4AxQ4Hgpt3q(RyG85a$akHDB}^m-E=te<*n}n!1u2 zv1P$1n(sH9e(2Xh_XOum6f|P9>L~0f3&ZTOTT$Ef4~raU(c_!q#Qtm|%R`qRR7KQPg#1N$=!>GuB7m^|$V z+nQj5!{Jfb9vY1IHnr&U+JS#87&%5h$k!W>pANeqelr03_Nd|M!ZG-6>w}c3GIT6( zDLswYM)%h}M8gvm$}2G>yR)54(Zm5U(YoAMKAswO+fY@z9Syj~J@r1Xu&K5QgE)UA zGj9psTR1|hhyO4dIR`&o^-$GsBjy~Kj>WrHA*1&_F74ANn;G-z!Hl=?>Mq-4cIk*iR>a9~~cZecs zw`pUsxGPLc!cfwb!r4hj5a{2G=K5fyFABlD3H;j@JDEOcET(upHwr&=0~%jmVWGo% za_F%{<>p23SmA?D@1!|nVl1v0ZiHmtcKByH;$CP34m>`FhQvk)wRiBP&K_BN(otH@ zvw_>5+m`-Gc`fqv+>GEjs$vV_NbQ_;U|S?l8&#yq#-s_3`TBV1ZI1V1v!NQb1X6yBpfLOb zrbA>^4Umct!4ejI_<^<)e+c!ayv)d-g# zK?^E<(ERuo9Id74QCf*)mUbg<<0DNwKg5_p47`Bu1soLM;@NVuy znBR(rmNB&FR1!^{S4l?=qVO&CA!=KtDL%G_uG|$e zJP}WFU!PNhLJ}Ks>M8Tb2pIfPLG05ISScF9Sa%4-8~WI2eR){(*;To`hdomkNBlU~;zbnw;nFqI-tRXvR%ry7Zpsp1B6GvvLL<_1#X(uO1?KzP6O_>0&Ld zYjJt$5ZV{piHPIV$mnJo#h**&v+*5T-m1yHEyFPH6;xDi&c!T*e^=S8uAXUqoROiXREk}N^t{c|Kx;YtIiuw4| zBuTU1{laIfNWS)+j92_@``)`K%r`lU((ftAJTq3bL>MhPSU*;zIn)jAFE#LE)IO-D zEyTO`ThObd12Jnydb4&i>7;O0YFZBMtxurK5tXE;{(|KNai(>UH0lCkY0Je|bmV!^ z|8aKZQ8B%7zfG%3dn)Z{ktogF`#hD3in5D{k|KnJiYV>7_K;R-^`nKTxzG1f+OrfX zOCn2_Y#~cZdFTE6{qMc!%sKPloH=vnd%w?TrF9ylsUa;L%2Q^5J2iHMqMG6Q6()Fu z%G-RM(&fofDUBjjvyBKfoIl8wUowqigmtO$1=-ZntTO67>7yj}@}H0Av#BFjw^FeO z`T0e8%~T)1>%Gc+CH2cbjPml^ONDhMQ?cL9QNF#mDBC7Msw!NI`liXD%HMg>zOVgg zT=2i$y~pX~TPx`98~XHOQp)uSzRLAiWVv~lq-Y5@LHc8wHm%CnwYYo{rTY9#Dd$DE zxw%#9l#iVRWxh0ql4`G^-uIoNUUB$YNC9=!o9GP6q05}I%AZ5^9#f>A)QHn(;$Kl4 z9X08#ovWzIIYm^9Svn=sJLl_dJ69-D)jcn`l@ISwNBO<$(nS}jqU#=%>}fInZxTWMup6bUlYUXE zCeNtRb>Y<8xNX!t)0b30P$6wIahe`DT0$phnRB%v$$4qHoikdJ$m#cVTP#u0iJ#R*?+&Uw5kgR}FdB3<{dmy$7SrP^nI;p!Frr^<6zaxZc%x$bY( zxTn4Za;1kBa$|!NxYubp`ZuG*_s`qXu2JID3w;f0Bu$xG{k4P>6qw5S8+ekFZM~Ee z{3n8=TpGh!d@O-;E-rzyVND(j&nK9-#HxJd0`yuhD1(K&H!h_ z=_Za^ZVRWQe}Gfq^@1}nl}iO)JWZvoEu#v?l<7L3#q*9uDXlJ+3JvYxdt$zGC9>vHB71zeyUU+a+d>+sUt$vUYm`ZK7`9S- zgL|m_i&>m6(Pudp756!Rl4H1MUMX|MZM?Y`|E;EWlyNBa5e>?(&Ve(Y?#3x=OyY1B zI8kqh9H}?Awo!g^F7!_q8+x*N2VKzjn-hJoU5i=Iec*?eN5SzZb;rk?~`IUfpYgb{l`W*^LjHlb#o&*$N z(4++Tp5R%Z-f#^PK6Cxu?sDf?KcW6>tfLgaG*eH5S8<3T;s_lN;?!yK&%z4Xlybl& z>gv!1%FMHi+U?Rzr9Bd-FMSlFU-&4~74d(lI|9?`s{auEsretbKH7*{DW*r=k$B2o zQ`N?e_7$Tv`Z!eAawBRl--DGEYQuRsPzjQ71&}DEBZ+PAl5L8EQYx`EYbK$E_)tqoSP1(fW3nik_231!bP5JndF+ zR1$nS8NoX_EjjBs;-6M>3P11UjBH8gq%Dr(c;+T^F3xc0cw9;4Yz@ib5TWUOf4Bw} zG;2D4-@{b0kSt{;W>0~9B=z`UHr1X{Ou3%mXT8*9Qr`~LQqS$Gxq8>*xiJ;pT(?`E z)Xo{ZD1}M>8EaZBwN?HY^+2YIakgfv zignVd(en`;1w6pnG_;v}Y@RtK;k}$ne4j~Oly2i|=`M3yEG}|Wt|W2fE&Vy;O+_5X zg%RA8_zLbQKeK--=s(K0w33Q_P((eSdP4b!|Df(2=%o7VWN5>MJ=C$ALzJGqBzMG6 znyWV|z}@Z?!1=a5j`Og$jPp+^m2=@qHfM%f4JYNbFU70%rpDrus96zfIG<(Pmt( z*&>RcPAK5^|GjQ@IGuhDfpXfV*Td;UuH&PpX2>+#~!trgDjyB zk7Y}<{|%WTD<=t<@hg;Ae-8rf|2~7Bi#wdWtBiNGDq~9xaU#I`KuR_lLgCk!RFcqd z2(SJEHq1Gw^&bR@?g4n)p}>}ePX{@t8v0cBb~3e{W%?({#FQEvH>*G@4$yZMbJ(sbzZKIiSn%gc3?=4rGqV4`yzy5?h{5s zsEG)PY-0bti)O|Dc(E-%y;$0KC7b(h4J!`wSeY||th_)M82pPyf+Bg)U98HE%{E}& z4_1NevoTOgl4FP7T!Kw=q}cuU75dRC{B|juc;zG&Esf__x5o zI(sVGvK;mrNU;)^ig0N2Rno9S8x|Fb;sk*MkQp!x+CkM|F+&!gJYGauKCA(g&>E2R zjDV`2#(2JnD=O+-2t9+T#O2#@*k~vMDd!b<#yXuOuaaSuBckBTyA-H>+)Ml?*TSB3 z5w_7Nn!R{ffjv4-Lucm}z#aV4RZ24K{x1o76VJ1Y%oej}^&?^OLppjpvksjwl_jl` z_n=^~9Cq|~!R~Q=HqKm(J)n9Ty1nFCXQd@<`kher%xEeSF&(3Cy>}vc#rL3Pum%F- z9)rl9wd`zvC3eNyC^olvJ4t<-i&CC-0d@;OcN+WPpvef?f6fGw?B_xJYb8iScf%RA9@w;JBb!*8%=Y^UTJqv1;b3b!7`%TCKMadu+Uf;tyRI>)wZ?*x zvk=%nX#_hEVn-BI*tf$;SST?Y+1X|Aj_-a7vv-sNxA6q>ZJD6GGv?9dXN)0BT^-t6 z3@i{~9~`QNmjm0`m^(gf$J}M?3QGYff6>Ske^ekB{qyMq zGd5zw$WKUdXBxaWiJ}W_tc~GVr7|VaJv^rUggatI?3svQ*4jo#x<;%-`nS{7|s_=It4BU@i0nsn` z+Y8|k7?1)ltGC5Swp0yiHEf>Tcl+m?{XX0GsM zE7R@CwY!n9D5Q=l7fB_G3&SCMa*%K(M~LH`SIk=J!*rK%F&Y2xiJY>2No^}%E#XcFUreZ0Y83?8CfkPSKr0$Q5jAQejLJrRt& z2tb<;8iA>BFvJwlXxTk3`Q@oe#I$AL>|Re|S~vmjCk0r&tqrg(Wq`4_wkA_5f#i#0 zI(b?=lc?wC*IPE%ARU7%X!P4wv_@?%gsiKF`ml1CZ*m$&7FGe@WeF%w65=f<>CGNH2i9X!n_Zqc0T;|fVOl{wQF5{)8l_u7 z!A+Jdb6f=Zho$P|SGXc4ulLM4-XCs(IsXixZa@zQYr%gDr%0!cAndpp3LJy=U`PER zbfAGY!b-<7Te1=QmSw={l5;4>vKk#MZ9w78*3^iYC24rZ zfg2$zWco&7>?9+G(Gecm{b&{lKXrq8fknLSFL#hk2TziADG{c`+_ADlGcMgPX8GQ> z5azAwfKq|~V82>5^1N>Y^I!FXu0t(6YBz_(!#8P>wUNxRXVZb}m4RM=oz6bXLXfTb zllMGA3RV`YV*!)L%mbY?@<<~Y+|^x}S!&-=f^02i-C)c6fJ<c< zC}_PkK(;*ifbRRo`bYDduYO0az;TZHw zBZTgkb0%p^Ov!Sk)96#eT$I}IjmSqNL0cmaMvVN>qxuR6UACFLJ#C3SloyhFyVt_T z>`pNH=m=fg;$Tj54Mg~lGe-(zVd5`KEIpIak+w+6I9H2KD=4BCUtEF&ue?Tc91k%~ zI(1|qLIVrz`_0Q)uZ+CWd4lFf&<`R~kcF->_E=X-7Tji$-rH}?}YQ6&u7B#{0HBRvOs4VM| zCdrme`NE;yLoi&<|8M?X2xD^gU>o-U-o5T-x|40l`MPm<=Q2o%tcu5*g^%LR3wvOD ztON{t>(S5OKPJKLL6Fv|2EH$rL$`k#h*aDJeYJO}l(~hsuiuZa`(_ajnHjMAjWGzN z^pNlemvQlNz&E#6Qr`-X6K_#MdIrzIG?@T!%(jCR_AsxS6?R+sL}aawM&_59#8I z=(~e0+*`Ag*%7k`S(cw+J_|pk*L^ghKYmAK@X%TEDB~2#uX%u7PsCs`?|8iM%v%yu zXNQ8P_agg`?$Gd6gzZf>VK>C}@@J?Wx--&?wnYZvmOxGXtVEaRbte;U3O16k*3}f( zL<&v0?V|sMexU_2#)-`F10Dn2$A>jB3Xy|b89O*dZ=_8EOS5z_253FHE@Gd>P>Q$UrT)c zRSX^;i$bs5b?akB$5B^D7>eBMMcfptiGI@|vdr%SnY7fy!s1WJfGfj|e;|w7$S?A5 zxe%Pvya~SqO+iq9h}?aJnPxF{hO_ho{U%iuf5~cL^xss|M~n>MzG(Hywug%-Yd2nxN6N8tr-9!K|uRA%C|{P!TLb1{<&PyM_*uo;B;x zx+`YH{@e^a(6|{b-s1#wI|SJ)rZ2%NMU~CCjNz2eM;O@F1JeDgNJEAoiSlxW{j2Rj zOF;@&$E?S$%UODjm?Ej7)-%YA$E*~*!8>5mgKnzcq!r8KiL50J*J|x}KW)N)=N@W8!qw6=r%@#-}h6AUUE`)GStR%j^s!G1oiLHTHf1&05A;G z1*a{iQQWetXw*IhwKmtGeD8B;%Lza`X11V5vE{_F){eW6yeCI5sgv6G2WV_L#?u+0 zaT{=Plady8aMXdjGsOY@ksv;*xky|~4Endvf;CZj@WIUnoR_R5PZyX$>9sL(NGKF~ zgH&MT^%9cCY=#|pF|@4%_;_M7v07(eucjHvdp5I*H_zx6y*0CnWRg6FXRsSQB_u#) zxQDjrT1Nu5B%;f{y7)E|3St{ylN&V)Kw8v-A&aU>zWqwr7V;c~#0p!GpamO9{e%$oTW^EegZbou-(T8gmn_;PW(+&~ zw}WBhc@SD02f|x^kwvcrVeRc2#LHUF-Gq>b($n8oydasNmnYVKT6HwMeA2REqPZ@`106jFSJyR4f-`OJ;rEv=l5 zyfqc@q_;4x-qb-}?8{*Uml)%r_6j1TaflRLs-WT^Nt*a`D@FWXeb$s930wXV zb(#I43&rz!il%p&I-6l8XUh-Lu~-vw=|pt&*dQ~yX*u$EZUSfAYe<>$MpV4`J;T#K z%PbPIg3~J4ivxt}RDig#Vlrz?3bKp!`F-%>m~x-N+d34A zxAL)}=C}Riw%ZA2%YY5KUXg~ZgXSP1{bI85(<0_zMKW0#eT9h6IY|GcM^LMp9?aM- z1P3;5r@K3@1@4Wrj|+pS4b=wA&FLSw=E zNCcc45hjFHg!1!8K~8c3Zx}LYPEab6(|JYY-U*?^(Q0yc(R|P=Gliw!moZxF8Pb|P zhXi@drdJ7-AjygID1W;;9u#rK##{K=`ttGgj`#v%Y6!dwlZ)Wh7bOTjc!Ef5nU8*D z8v)j* ztzEPkbC$6xte|rapQ2r!TtpSW;!w}S2-xsqBgj8KL!8%Tppv=eNahg#UMf)nMwA9w zUw>vvpe3oB^MUN%5d_CSZ6X(>wxJ_eTA2oM2lT@J5o#Z7pwm?zk+ka;@Va3Wx@0iH zJLRB`E-34Qy3{myxA+w7)#-uW`CeqSZIqk2`9FJH^`0ENkq0>uimq3 zJ&dkD4gwL`@Z^ySy8BTOXI0*%)z9caeRwH7I`n{4d>4i+(-LBtQOy)@SHTaWE%Bm) z-H7h6g~pS8(3riOj4xG(M48WIXkHYNYMvwyBbp&>&sX~TPd!@NPKviLIvW)qp9|y3 zLa;kzjKn$)p>dOZq++!S^&LNp_I&roobXy&S<3;+`!(P{>t6I)HP+wt>{WcG}7U%W(^Y@*ojD*ty()kHjnxepR93EPxB@sIoVA8kQ-57+{*kAYhv!c zI!>GA&jY215KP?5} z3(~NA&t~$09VYixw}Z5a1qj;BVwY>L0--;$AhnkR(L1BqErA)Vwa8`m^=ErfzNt;L zf9-^OmomwxqB`csmtHb;(umHOr-I(2wS;3f4-97d!#sm(xII^c6%uq|9ltxX)0M74 z?&C4);f6QNwM+3};kkkAYAGkxQd-Ohr`ep45Pjsc{vaZcf7zP`b6F+%Na(?eM z=4|f*FwKl0D&{4m`qoWSvSJ;`8JE#s69o|D;t4-4-T*hdZ6MPs4?Ffnk@M#b;Euu& zIkP2>__n-EbD$rfZ z9NxyqAFojJM`yvz`j4dk>02VVsFKv`in1ySqVVW!HuE#65?n^(O!pWbV>}(F!9=1rVmo_iBP(af+}i`| zPV9sYr#SFe`yMPmsK};F_W|!*E3Bt}iMA9Bgsh{&yx*$_(W~)=;QebEDo#6(u3Wc= z{@-8F=OI(9q?w&Oi=2I-QU zcg>ORvP&c;t)AFNM&ZipL~6dc2WlP-hnxM&X`WgxukhM#u>DbpTrDH;RFpMV_KpDG zp9V0{Zwtz=-@v;IE6LoFw~gpRFPi%yRhL!N#+EUz5^r*;c?H++B`oa2U<{Y%2_H7R6;s1uyf z{KMO9V1#JDLpZEt0*qx=fbFXU*dV(UH#sHZ;-5V5P`w4p)1Jd<$8D&qy3S}F>?GUW zCdp7m6YAD3Li{_E+?^VQn13-q8Ap(j#gO#ez<%h9iERL1{t4h@RZhty@#bq z-al~&tS>=o-$!||UV6;pM;B>p4P&?&tPjqo#n^cR)7k4i+AO(tn;foeAu_lF7S)x~ z5w7;6@%2^CwB`AXNK78YrM-l`TGDu(+&TRgwZ5lE$U4UOW?!fLT2VhFF1H~3*leJT=5E#E6 zK5OlvH{PqFM{n$*`V~*3x0jmX-4jz{I=T_*oe+cr|9ygvmo~78+#|L(J}~{}k!Y6$ zpPQJO!CR2YKUv82G4Hjf!y#&r`SEoIUfcfAudE3GC23{we|@h-ys)7v}>| zbg_iISlUfGZ|IR-v&F#Y8%whba8+Z#{g)Nj)vp+o`JNgESvb=mHMK49-fP_uxj~QGQ$SY^;0S6@oy3Kh6cqx zdgTL(@5}Ks$sqhvkgTZR8jWFeU!RY=cy1scl~2J*KI3VQf+l&8w7n7@b0$GN3KL{{$((jYXA@V_o_U9p0RC-aP`4|z!T2G1wNPX?~u)SNwfyLG|HHre{C-N z_vcX}>@5dj&R%GK(-G2pOa@CWUy1LtX-w7pCgK;f4YXVq!r<3sVEudrynI)J`ewP} z_HPSd=dK52%WHSMB!PdXE%f4j40%hfTI$Ml{tn0Q?QY|uaa+q@0#8Bv#6GZnp=TSG*I8tAd$}( z=bZI}6WSkf*6#{3e(oW(Y!QZYjUpUfm3RoTyaAGO(^>UABlZ(cL*Ga!sS~_TrxYgv z>9uCNc7K6I^Dgt;!p+G>%Ut5H^E2r&mB$9vIe1T)05jwy}$b&}e9Tqw^6HDsYqC`U;S~V*Kw)m&=@*Eq1-yH}p zw|7HCxh`J#K8N)4{UIM*?nBY$W)Rxl2Ya}yn9yA{NrO2FPeIImW-ZJL9z{kkl!(rP;dIl)_+flVxu4^UH%VL5r;jIbRBd*a3MbR z7g-WvjA!q9g#=ZN$$cj!2(>y(axV7}y3`%Fc3r|F#g9=ziXYiUlHkRi0m9XjwS2Qa zlkqZGkA_CX@eTPq)R~D0xKE-XK>HwkFFpc7Vn@hdUKhH(t(*v)QO22(3fSYk6TWxk z6IqrROLSC}K*K%-Tn-^mIMkm!ta z$dMjEI)U}bG-#YBGuMhd+pvh2ceV(^n{L1%!+v;q`5&?BUJM+5@0VDyJ~GZehmLJq zK`ZBKQeUok!pDtgm~NvdbnAL)Ix2Y!k}_V8-s@LWI)2&={qq5{?O7!)eW8t;P(PD= zCM*@HAxWL(>S39+828Gjqxg-JjAjOn7Py=ygZ)k9olF+QAucR)PlRt zt>n=!GcvAn8jh}vB^vj1Q5ip1;`a%CSoy>flpkyW?zs@=!p#A$$?Qw$aty%p=4`Mr z5rOVJHBeJ5AT1ZyQX8u(NLlF*GVTHbiU4FE_AHW zfz8)g5+8q&7kK&tl6Ibs?##G`Hb)tvylws{Nv@DS5okhV{W{WcyPf`)qR)#p^C05& zYY8uNDXbsQhEp9Aq^VM$nEBsF#;aC<^yW(HzYDoMIh*$+QtK(oRJlmH26NzbQXfhD z?gOtC1uccHh+Bqzoo@Ll)gKl;w?;>(GCmrWg$6zDqPYehD7x8!`n~TGPvU~iLS-_l8j`N>5top!V*l|;u!2I7u82k;V|?_`fUkGXu{K5x>Z zl>7bRHKJH61sz_q*u=sM@cdK?D0k}Ng167mc#Q$JSi_*koDlR~<0N_@`k3hp9iX=N zs)E+S+x4$ya?tV3>cquG9+Iyml2;}DWP1K-vP4T3GunK)hF>doo)%8KmfWL@yRw;w z?{kuhqlUY>n099{3VG?&-vg;A+U1$~Ayg6@71MqH~nxasc;s&dO;P zb~!rzgCJN;y(bEOPZ{5*<0xer7a#8ULH4cM3BegRVb9`RQna}TY`6b{*elVH8vGSa zj!5HUPp?uQ7WU9fT8XU9OxP=skI4N%qFOo^zV3L#mC#+lXxRQ_md1S`ZvXz{sk})5 z`R4&huDAuwd?1Hw6Qo&-E*JJn$OCBDwi#zMOX1cX9Gs{>1q;*uz|!*)tc}!5Sx8WXYcu z@a6CjbKux3dh6U=eh&C7%efZjmKTM^EJF^g1=oj5;66>!MLMQ%;+8&y^K;nWY%D^4 z0VnZ&DL@k^4eL)RO2C+E5)<;k1iQPQLC@{uVP0?~mhIxd-=zzSXSMU*wumz}`VZyi;uFlz6#5o~g$f6->e*>Hy!aD(IiJN1s>(>c>?9_)otQ$i z4(g!QWw>T?6{0hQah!qyS~6=1=!?}5&#Pu+^0OQs7s}(J?o!G+<0@VHSQV}1NutK$ zFZAZrqgbZ13g6pXivKGOx41e(hW_Mxg_-^ND2`lq4@XPLTF&uIL;+p*cGaY zQZ2>NpGE_8VvZo1yWNmX3p$IJhUwvBKZ5b;Bc>>@Z!y|sB8VaaW@5e$4_8;cMsTPI zy*@gNTn`aN&txv5du!a$cisO`P5ghzt1k}UU-JrU%*?{SBxC6v?IH9TyQ9SSw=Wzp z;`8~_*WrB?^NHh`{p6v^c32XjhrisY$AfnIF!x?Nx|n$%J@q+<50u=)qZU8$lFAvD zgB($C5t$7WGI>z{)17A-nn-l7H2}O3!+JBSkR%^ZoqpH`Pkfk;=Gxigc`rX>p{lpo z|0Ij;^!MN}e{byHql7b`w=!dGC2)M{Oi0+gjRcjZGmSc>#4+#@P8y{wGv_b0R8&0z zIok4YK|>6daX&CeH#F0tpIID#*%IV5H^DEX0OI&jl<{-yW&EN!=vUhx)S)PXj67d5 zMe}L2L@Ss4*!P`mPQK4f9Fk=EZ_Z*~bw7jjTrZ$N8xmUG!J)z$q_2M_S8s}Ao6wg$ z*bbQ4y7E`rj^ zJUDr+oRt4ChY0sLI><^6xu-=kBh5z{+fQI0u^hN=g;bp2UGlvk1%4d; zhH9k*U})2K5DHoa2Kw@_C*UY}#;JnzRtYeTipDQ=kC4t4-tdb%kFj2%2QJluxc%l1 zRINF{U}ORt53PcqDuP&OsVMRDTnm3=3Q>DYHf`3{P3YItu+mp!w7GmM3Y~QxtT>@i zWO0wt+Lwo%`hPItz6LPiEQYMxMCsuM8mdb^Gl#x)qN>@yQGnnR=KS(ExOPGb=l_z% zLTg=^OQV|D>a;4kTB<|}$EC5l{v-5i{|He%JxqiL4Vd{gQ^Y!ACQ+mHknyl2Xi7yx zcbEx&vR4wjDg2-;cYH@_qx>AKZL^?zsRb5w*vjJ_48*rQ?l3b{7lW+CHuA^B30o^| zLCtT>LEO0nnde*KCrca2oL>lA-y-CYk|23uP)JhKB52`1bK!C41n+=%BolDa7x%eO zl7>N7n0S2-F1_DR+~3|JR*CiXi!)-0M9&ZtqnZR8{0q_Wf>k8b&lpaniQur|S2Qti zg`7=+jJMKXVs^iZYV#~)v=t30rAaNY&@~~U?MYysb%o5HSxRri56I`4eB9VxgB?sx z1t+u*-z##j-{7+sJhY6#lJ9p;usDiF7I4X7`+dxvHCvHFo^5?v-cu4kzmL5C!{;TQ zT!alM9m?lrqZ1}KNO0mIqB6u1rniAHynmZ4$VvfC|D|Zv$uX#LY=zRd6nmT!gzai} zsM^qt{(0S;1Rg$5=6}fs(>aQ)yJ8w$>a!ZI?vf-KhqvSJrW?o#_b8m-xCw7-`ovY# z?xUO~RA_%!XFRi83G?a9Vmb`=BHvP3xa$&U^G zc8XoNY9E;vk4d8FeB`8do!nK)VvZZ1Bb7E3yU^5xWzI{n0prK$^E(dWt~Fz5c11r? zTss9uH}1r{Q$*&-IjC6#kZOLFD2hx&OYb~Fj|%o;pUPqgHebTdC_2WT z_1c6+j?4hvqrsqZB8t&kSjgN7RYN~Z&tvy91z6t9fZ1}z09RPg#|FG*& z7=C8xh@vVBT;7gVn;$MWN`1#4Tzh61y>rcA=}lJN!QgL`q8``#`Tc{(c|9}j)G6gj)mcoYamd%@-u@i@b zk|k&2&Dw2P?By)%^Kl*e6*L#^6Ld$fHUuF5m2Mc{7QiMdN3o!z5t=-sgnk-tKu?o@ zFy}@EP(|+^yuSNA6MRC4!2BomCgQRryBE_o|1xl`oifwDS`m@G)nuQPu;mONLCY)| z3UeE6kZ4jfSvu51*Cd7DuT#HJZ5kNrxuAs8<;ZhN zo>txMh`u+SL~@_M(9*Z0dDGr(#YM&*xLQW~TEcZGwzkryLV& zUxu#VF2%xAukp6EDkvc757G-0$J+{?p*5XZX#39+^2MTrC#coP{e5eM4vuJnB%8kw zykZCJyWb4Ft5tBnH3C*V?C0%zC=E$hWvFQw{yetNiN|%t*t`y$-+u7uCUNp?i*o-`a`soIr6ms@r7buLK zh9yt#gS>1SG1_yC(K?$*8vK)J^;KoOy|yDrffoVuv`Rqtt{22kx5N3<^N5{G9+7(b z620lN#^aBL(RP1Bm`X1up?8#FB4dms@VVIa+aB|^K5~?g`!)3K&s$`r`2_`-^r4%F zBFI3fE>K2oaQm$g8w`dX%2 zq=sp#ldKPQ|4eAFNW`{u5vzwj%$yrv(VaR9xCJdt+Vw7`tbQYkyq}8{OKh1%frKbT z658tQ45qHohrU`LN&a1YL5|&90q1v*kV8plnWtB7z`2knuvw?Vey|}3k^OjvzB`U@ zvf$;rn4?>lU(qJNw2;WNcwB7Ygy)(iGN;Bqp{Uv3c>UTJbar9}@>N)a?GuyHpGiwp z6Eqt~-O41?d~;YByaEp4C9(yhu!L{1o@tpW*v_ShV+Ms?L zYb12xr|lKU;c6$nQze*46^Oyh!w1NR1+Jt~YXXVBOTwFvm|?>Qd|oip7;`$cuygQq zywtb>XGiWtIwYDH@j3HKZZCGAhOlU$qNR$M4>O*#7|m=kLk%Uj@!e)aOaDcxmKo{M zWa9EY@@!cLxxY~m^)^N0hbDD6^uk8;S=JT({VxcwolL;926J(;K`OqUT7cAMIpUMP zZde2Tz_*_XTb?QR$4_Stqwn>a=$H1lI_||L{N zzge2hn2r}Y|63T}s4m7j{k!necc%F8`VLfFc8?J`eur$#&tRUa3-i)T`8i!tF?92V zWS)`8C~i|-WjUs3W%ikUH(4T_puS#G63t>z1Vo5C1ZHebFNw{gF6VLDT zJEr&XS&}z49T(lKMenOl;%lEGar2W7RO|8&8;+<`>B-Az?~6&)EnNy%)~e&OQ7#s- z568>*+Y)E9_h?VvVYIzo1J{Y^;fYo0kTHG+UcPxqc`rUno(7*p`5oubpVLXqw(I%4 zrTZ_?$}*j>K;N9ft}|eOl)<9beJ~{B4gdZV!{_o{k=L5LsIO!fwj6fHXJ$>I@4{zk z5sO2#%bSP1!(Sxu)btQE_pum$?70oCUZ4UE_iNEkwL|FW)|0%G-9J#)uL+1wQR90o zPQlilhxwY3gN&TkJffv$z%FrUgcUtAh{<(#dc_89T3eTUuSLL2JGWj ziJKO_!VA_3V5L!2{G~w~|By_BgVuaqNt`kWJqU(sW(#UmvSa@3?kDNfcVUBfvvJJF zQS_7V-B4*VfKT>4Xik$EF3?$x`@a^G&$)9z+CLtij`|SM=s2?bZ6G~OayDpKD8u>w zL6YCMl}x_wWQ-O(CyHw-v6o8$cG+HpGoKj3`(5QQSNb{xWmQ1O`((ILYD^3w8<15l zgQ~mp$O688{u*xuR6NikHw%4`@~}JvWypi3Wd%<^)|a-ZQb3}vNo0!OWi9?Bml4`0 z&C9u+TUY*IoO|`yN?K#gmU#h!WK~EH6nvZyE6Jb!6`+R?e_`M5o7kv$4e!ev$<9Q2|3lH4I8xPhVZ4xpBt&M(JcSJRtaUYzRFb4bWoV*Qid0e} zWXwE7$`}$V6!+|Xm7)}dcq^fKph%@O58wU%gM04TXYaMwdY<1y-PB{Kbk-5xf6cn& zLg6;@wpbio?^@E-OXBpCm>oO3&xU;cvky`81Py80M~$avk%nCn@XKo-9juO}|Dqa* z>e2)V2+Aj&G3K12-6A|C0px_1haze)`#Z#LdG@@=J(#_Y*cKAle|7=f$&*o=M+oUKR(xbXO1g59qC2& zj+N4gQ&Wh;oNaLN-+8Jqy#;>y?j{w(zEEU!3@-P60>j}cI9c%k=cr>o*%Fb6vYX<_ z`*H~w=9h!M+QEi`q;xp0AB#$NI3)VI9ev8y(v=H@;N%8nnxu4$SgMsUS8oe?>;7}B^Thw?@9!Lb$cY$6EORO;GM2`ZYu7;Lr!pf7_uZr1T%k1;MVr3SlW!3 zpOuHQmk55$@j|n`Ol>CeLf0 z(tqYHZ0}$cxDL&MY1%d9<%CAoW}h+b`ku-}wKbE}krxdPzTFrrvIX7qB?L9UH?BIL z48mbnbVbcsvM@KA_CI>f3{_sFX%&i8GU65;cS$4dyLm)5cAUQI^}?w+3fMjUIo&l$ z9*pj5@E_f-WzNI){2GP)Ohq=l?tA@DR|v> zEw11flF8kAwEE5-(%UneKPOt5Z?mjS;I|FNxmg9Q+<&EDf6xra7W$Ib(Ls7h#~*wx zwWJ~(?#VW9)*k7#9uT#iOYq~WMP%A& z8*vzXLgxo)QOm^}=`#sYTB9%p6Gn{S_oII##CJPM!&K@lU=-^Glrv|%PQdlNsqBA$ zwzD#u+jv}i5mYwRWNi-;o?>bW(aybw+2z@2GE)~@iXXBU%a4(Hf$ePm`?oylBhs+< zLL!;+<~C&Y|6rah)go8BO*va0Gc-57kses$k7s@>5$on!;t{_Yx~;3&9qp<_B%?|I zlK9SKyq^qvs52l=bQ%S1U@` z>ObNn*w&ULHBF+GlU&H9yx)wF`YJL{#)Dm)dzmasJWsCF2EcDWDeCc4FxPrmOp7MC zW8UI{%K@%>RBruVOms5=-FxW}rxMAY9X>|?5G}s(?pShf{}1NC=^S{Mt4pL+CX=VG zV*;&I9<=0$p_uq`*tqWkZ)0fzE}s~P35p-+9|0E({z@`t&y$$Do%$s3?0MMw(Sk}V zrO`KG1vLNp8S-H`o`e=pMZdYpf_`Na6T&gaj!%iWcTWzUOi5#oeIH_@HZnwO`Ylpz zVaG%?iV@e}w;K$18DLdr9h^TaN}`KZiIMkr&bfI{tmi%a139kJG(JIw^lDELoO{#g z{%Hm9#LxlzioZk7*FuuFd(neG3&gHpAdB{Gq>rlB(bbl2G{vTp9(~nBN9C-jZHzbe zF8D)Me|bnI+s2dFX*>C8gE{=8RZIAd>`k&k@+YxXmB+`vspR)W4V3>ij^Scr)W2&T z{rWrz>$jRhz1RX$EbQFQWTDR?+usR-zBiM}jv4Qus?e4gOU$PMH)qq+JMY02zai4E zm`v&(Xki-hK@U|22#7sOt3My4OSfmyYgKZL_bV0JMW)b+=9<)7G?QLznn&B?%E^mZ zFM3%vl6FtkfVp>&ENi;Tyg#8qxU(-(xm7iE<_UM2cj_Dw%e2Q;PtK!x&k4lw`Sey= z4-H-*jZQwJu*xxkSs;;<8l#>m(7WF2Nlc008{d8m9;J$qR>A%*a#cuB&(}~Lrw0P>f-n_cev(`@EoZDR zs8IbJ5sZImijhZ-)1B9^kyA58@$;4Gbo-l^6kiGRUrm@pFRsj@ZK(@MEq-P98;6h; zD;tP&O)Ggnwg4-NmZ9aGVq(AJE}7xhFTotIzDR@B5fvr)`<5{cfz{rAUt6YJ{fhubhUh8{qkcDki*1jG9`qVN1X zq3rn(1l`Gip^nE)Sd%PcXMVxbM01NYH~Po6P%ib~n!o*w3grVDwTtwFVF zj*tVt^Kh8mh_(@Hf{WhlrmHYD+a zp8QpDN#>2}4wzb5Xx(yeC*|=Zsf=MK*X zzlGh`T}^Di1<=G;Q=;B&1zJ1)V{Q#Qfx_AWGG=;^v-rab5cP2+?VdG6)60;Qm$wk@ zkv~NAqX8AalLcRR=SgV&5%S~cYqG0ijI2yGrVrwZNco22WV%8NStH=Eh8aGx*7&6h zLI)J-x+MkV%{*lq+H)9XrQ=9VL}kPVyqfOI8&AlFVBk~gwT$kQvjkSz6xb<{Cp zcYIQzr?d^(s1ugNFY`2!jN63z;Ytvy{ev{9H!;4~-Pvx7b3lVvV1Bs-x>((T{WFXq zENcRK8g6Euiwr}^mqF;7%ZGPaO}wRN>UdW-P9wskh3wvcl^l5%L1U90c-Fz&K-{+n zF8-}=aC^Uqw2NI~s}5$vRxVU00m@@iZ*ie+WF)BiMV6*$t5r zr(lW;L!Xm9XrQBp^wmAk$(_jt2#1rSv$aXx;7W{+Gr<9`xp=JFh_|ii0dxJydiX98 zgb_-PV8l1U>**bG*LK_2RZFOc`b?| zka21zxmeoGUaP5veNS?kjydCGK_C+U->bm*+I#X(cQy^#t_=6cx}@@PNS?>o($%jgDpvx66_b^(v~#^GDn1u*HucDk5rmG>ZSBdYBs$uyA*?U)d>pBeM?OUbTZe8#jZKz()F{ z8>vp$+DGv|(oFn5tr$1@D*<`vbt50=L8PP8lF0^~{TXuuRQ|3fl6=eFhQC3!l9^N~^ z7MAs3+l6yjTU!VbS)a(qOj|bl#C3M%JU?>WPJ#I<$j>_d#ScacKH$3RQ@NUP3^iFi zPvBvcV*~CrVcNet5I+AYtQ)$6!A3#&l6G)A0~k#ci2(oUBrf~4Vy~;Kojo|ThlpS-)N}8ykC3_Z7ernfp;(S z#Zv_rCUwFgqj~l6LeofKhc`?yB5;_|rwa#w9ukp)@rYfh<^2f96i%XbjtOxbHz(^a ztKivA4gT)WderQQ0z6)>O=m@>f{*N8u=IY+L_F(+y|pngM9%!O~Jc^64(*E2dg*B0k^slBK;PS2e%uE`om?n(tTnR)V#h9&+=vP+g44~v*E&Nhl6-uHwIP0tC=Ie_;B?1 z8oYgN8VT<(hR}utTx4WM$BSlAKlUVB{<#o-@h5Saf9rAj;ZLxBw+>e4+p&XZt4YBn zBhJ9+Ba&fcPrc{al7suUvDa?SXA>oi@w3D~G(V|~^-8U%p?e;_oti*AUv)6KCYqQc zox?h`Ol0MMhO-l_hRK|GF)X?;4<1`{fk_Z0mnZ*ZQ}PWUNZ5<8^+r%t^@tA?l!cv1K{F&#B2P%!RZs8#QtI0;g z*bN|$weT;QfKjV1TK3|V=!fHwU`6zZ@Y5=}x1kdSa z0UEhOaHlK}WXyY+f0@%+M>~NYX6p&i?h2>r^V3Q1(NtX?=`vbb}$C-s9vMOUJ3N9j}`s0Vkww}yIC95s?zxJ zKg@nrO}H^l1^1*}W}f*OV{w-@SgrXgnC+cq^Z#(!k3o&rD=l}xKwCV%@onJUuq$JZ z$19K{{lD-JBh6h#dePIal+@&_Le2D>FdU*ntJrk1DS87{NL)Z(i3xH#*QLO|r^n#S zoFm9=OU6=@EojpwNyFI9~9CPIb1Y_UUb;!$KWglC@#)CNntn zOPzmB(t!VZ=`dZ$Sn&M=mheXwoAQeX^!Psd4RnH6B(*V;!2>g7&@1^9cn*JVI2b0u zoOo#i`r$=nyYx6YGxrYD7i$Gqz5QUr{pnC*kV7*a?vk2`RZL=sfSZt~$(OsV%8$QP zPEO~AkZBin@$-H$zPF_}ebtmnvwx59e)jBPpA;=8rpJTGDz6Ibv+l7V+G$P4B^d%a zx5zZ_2h``yOunzW2*2XNX*$X80?i4a)F9ioe$c`Ze9ih{lCyvn6#kT6)as+`bqivl z@scTSsfOTuRVrV!jH9MCnLbkYVti*ugIP^8@2{r6z&tz1NL72G()w=N@3w-P96T?` zcAG+S8z$ql_4goV-5k=>RKs~A$R?OPYl>hmnMdAVDnd81374F@gjb{;ap=eyo{U5! z%>AW;&nGN|H7CpQo?|J#T$YBn2kUUPP!*blPr;i(zTo@wJd=`k5GQxt#g)FN(N9K- zMn#?=n)h!pnmKDgyzDykNxp#6!YaJvQ;lDZ=d=3nZWA}r$7G3DIJR`zEueqXHulb+|3-QNsw!@V@j>TkrhPyUQ- zGM~I?-AESvjKChJQrvISh<63Nl(7yD?fWP2IZpA#(XYv{#=nL=KGy{ne|p1;q)IcV zmz;o2?Uyk~;wCOWqlRhE+Bj=>my(g)ZE)iDQGo^}92J8o_J%a$!bgKx=OROV8q9I2 zsW0PO+>HtW;dn0UJX^vFxPA*-VV~trX#X?~+gdB}-Me$RAliu;yP?b4&rTz|2Rk^I zMk5(ePzU!L()iw{4Nb~_qC(9R-kb6ftbhL+YYCt3;L6hft{`&s7Bg`h1-(Vhi?H;E z5A*7Gob~&tS1{*~VD@)8n=DRTjGHnY(f!6r)V)xI%7xSDbIEMdv@(rMu9(0m+o?cU zu{U&mETMgK%;>1)b#Qz-5wy8^II^(`Uxw~Q?tlrZ=$AmJsTX!cOaa5)w;8!tchT?D z7&=?M!!#8yyc`^hwWs58vZyqsR!&6;v1OQf&YH7Tc@fWlp$hriFQ`cm3g*5U`Ovjv zh@?EPg$u?)u=3U^I5%Sx=FHc?Fsq9jGn@2Yuh3{SnipQ^Nn~oEc+5;t^2_4@-uig5=}A} zHt3(u1j{~Eu8^=QxBe&!*bJ1c*fWc%Uwnm?dF+No{|eYc*^_u{ zjLg7ae-un_>cHJK3#sJe2;MRuLB{(BpmhFh+Ufh*LnS!C4Ki0;Qkai&{K9Cg?e7Ibgt?P+0F(f_EG8@s)=&X_A%#xg>q^caC9UKHc^Y8dYh?J=Wf3?H`TJS=XKS{Wj z+{zeXs;YFrlskcF9s0p&d|OQio$kVAS!a4tb~Rmpp^~Y7?@P3{wh&vNVz6#?;!LwK zp_$c!>4D8HIQ3eYB|&e)N+DR?eN|{MM)Ij84)#CqoiG!-3IN7$vSTx4_7w znhoBjM4xYt;RWnF3tgs!lQF~>tVuU^mEdmJTIkN4c<=;#E|}uk&$D>5*Prw`e8rqiWGku+Fk8GU$Ix50JGSH?8-5-46dOq6tEXy)8u%C4FW(yf6! zZKrz>Rv<=xm^+fkbDiPa^BbVH#T1nK*v!wWtKvcc|JUQxd1(A#n=M25M?;S~9rcY%$ZnC?}a>?#lN2$G#1y3$^ z6Fs3eL^EGCGr@s3DPB*ecjwyB?K|ygSFiFrUZ*=e|*T0G357iPK9JVzfYd()gw&=%p}DAne-_I;rqd1Z`C zz*6FUZv(le@`ep*v4MrtZnNV`DeUC?f52}+HIdM&ATI@L*lCvxyJMRwfsz|c{A?L$ z4!2~UHK&nhnak*)hAc^Ilz=?PAu{W`G=D7Dn7=^Fm1b!O*bdv;$h_GCwp)=Wbngy^ zDIHey-10hdsv@4MD6baimILAHRsoA5fkCfX1w?sGGF7dAK=uC!BL=;P3aNbJ9C?#` zBIlvRvJ3nUNzzV}OsXOz#J6AS!`tp2Lq=89P{z!i>~OveQy$#Fzl*y#Ev`M({>TU& z6fidDF$c)KiW0`|;eKYm?{SXFseCp~VkT)3_)J6myICjM&Ghq;9)OFyDR9rE<1+-mkC$B}WIGz(vkAj~{uPP>dbq&die zE_!r?`MDqi?%oV0C1-QkxtX3!FE5tXD-V*%%pK@>E=vaUCWH2!7p!pTRO}4D!5n)g zh7~U}2xI9_H6FZW{q|Rph~%TN-v2PqbK`ER%j+Nm_ivEHffuNI=5xB{R2L0$iD9_u z+Cwj!A1kL42` zhYUK<-$0djW>Jd zztqN_@)UYtofefZ7o0z)+PF+NnOQzQ!YE3-AW;m^*uIO4hl(cR zB%T&Nnz##ps}(R-TP`z;%=WX|<-ItRE5r>?%*0S>A#U6LHgx~n&IXBgvafGNl2>1N z_-5H(TvT%m9|;Y^7JeA}vY`x?+%3nM;d*%P(<3-NbO{spX>qS5E#Ss=3hHFdH|QeD z!~4l~ST=H%anb(>&c9cq8(##C*S!PNd0qG`wF*R%qM&D|DbBGnXL{nx$d|0UD88m0 zZW;#jc*YE~abPW7d&ZP_+u6`<|I~={l~_T&ZpamNw&5c08qVK31GR4kqWPx-pk8*A zWXDfpOvxhL`{O2vNR~1W{|fSgro_Uj*eHPxsFj`9yqu_Cse(&u3Q+piF*rHuf`|S7 z;NhD$(c{f|jQ(YcKNgf=^jQf!_MaOB@z-FDz8b#0HU%Y4p2mf9?qRRueY`ppju)FW zxqtf^ZnSz7(RYof^6R)X+dmL~XS_wjMltR(k!&{9^%QI$dIfK69k_cx?B#})#d0?d zKEodWNuWKR%3C$Fu_3lkhd4%fft>3ZOg$-w5#IXTODP=ga`|aonZ!2K&y=SBswNVg zKFmz}y@MOfN#Y8RuI2je9mXFqDOe`Z4M-kN#v3oX@PME$HPze9m71H*ooca{%U7=@ z{d?ua z{RvfgRq6-KwG@KY*5V*9e;QQZ&){}h_o23H8Tww{#q2G)2^9kG%cY76@-`xj_=Tl{ z_@PqP=9mUfrr(&EA+31lnHYCbT_n0Ebc1P!u9ZP|B4{*EM{+d@)f`Skc)l0@`|=MX zTPJfr|2&GHq+~I5gEShfFT?Kx6S=j{KA6K?gv*;88P}2u&Yz}`4C@wdK;q@8hNB9aZ-mSoRi#RMeq0b$bJclnLib3Pg0&=@iobI5iD6Tx8c@}LB zadkOxyF;1#Iy4n8`Tk;-$9F?m;zM|dJsdqrz4Teu$zU28>F&lH!Kmc5oz0%NS9v>p}$X<@&BZAaNp1y zHtx$x=(ZXGzu<1F_W3C73Y4c^X&ofYp^8}d1+(|$#F4v6py5oFr1xtkfxK1#M`OJi z7+k%{8Y(|QjgnF{TinZAeWj78-JHQM`H`>Q8`5wldbw7}6J9!It{LWPB zwj7(sLW;w6qf(JAC&IIPmu12ELdz=)0oWpLL;K{ky^_9^VaCSOF=h0sa z1T)|#A$0yZEm|yLPah@yZqSr2Y={BPhCI3jdSrx2>Z?}rTEN_|tC~aG?_8&s|A~>( zb1jV3n&ivxSvhWu>qPFtqi1jfD$p#~aiUr9mP|Zi-SADVnY~z6%icKMiaMz$Q8oQG zrYL?Tv&J&XkxNq4|9~vzN#1KXFJR#B=sdw1Pt3I*9x3OzF7+l`0~i6bNe8Nvl8Ekt zAEcyZ583?bKN7gxg)C2dK%_Z;U|_#6D|||v_hxV$$Mxm8a;?wtymC=}i%X?-lvNb# ztb4m*@M|=-XFfqS$^6UGj4)s1hQAqb7N0FBwGgR zz+vBeP_j(G)xKBon)GC@ksyolh;SMSQJ18JRXVIuJBk;-Z6WRrTT#d)0u>h5Ls0Zf#^(MU+|?WdTA8tyY36Z|w1taf zzG7JXV;3piRYm@eD6sSYn8D;%)o^7*9@p8NhfP+;NR8+t;HZ^DOhp1z>~{o3T1b@c z76ET>9^4yo(IA)aIGx{7{B8v?9jp@Swkvbu!#TJPy#dg{t);# z8V24NqP|)geC2G$wY&^sn7x;EJ_#acWj7ke6w=VY?aTwRS+V#CbOA=T>-UCk(SPrn7@q ztEiM~5#hXaCo!eb9KV;&#M5>s6Ja2X?Y{dM*?)+hZu;ndAE?kzQ(Eiwle!#UiCOMI zg09y?{E@pbyKpw{P-=np-UH;2uO?Y{Mi@iV9>ChLW+te4Ci6Ex0=F&ShV||YRzDdh zpT9kV_8)(sez04h@w>dFZ@gZ4QRYDX)Lt&@UxtsU5VPBvEsgha@PxyUMp0gMFCRLG+vMkoAQjN&8t5CI)x!Aqi zLr`~FkQ-jXtVhZ{l4N&-)D%Xe_uf=&*lCB))=!~rneC8KAPTw_J7_pIF_$bA=#zHA zjBiwlF8KR^MSy4tSZ2bkk_B7i6?HXPeAI8Q0_BbJ-gYefMM&-dS z@N{0u1U!qj8l5a9=#kt7FNKrPr`pUI{H?*YUml`mf-{ZUoIpR!9Hn7=34V7}C|O># z25kc6h^>r*pS=&My`T8Q;w=@FlJV%)2BhlHG8tL%ly=jo-yL$e)*+|tob^fAXPA_`SUp79=!D$*#^K05bvF@32~Lzlge zC7Qnlo{huBM0Z0q?98^I!w0xjb2X2tZb~Fq*^dqFj~(E7=Si~FFOz*MsLMp{WKn1H z11K?!fNr@0_WGf_FtPbHG2XG5wU{0VpBH43U7<;w`#uiLJKIQx-1`KHMcTM&e<75u zy~q*do{(X^yJR$?kSbKorj5AE(*;m=3vhTlLG3(3m!-#`$Vi7?yt$CQZLWaZ+f+Kfy83sd1!=8vK@7%P18(NVYEs#*6FvIo1k4$ef&b0-ju2HaDE4 zTMP1YW~`)3#)auot8|cAbc@sqM-#Ef51C`vPLaO1dm+nIOoR@!3aLj{`9m@e>BE{8?gvY=YIlbJ53&2Tr6h9aF(LQkE9b!~#| zpg#)iyoMRLU8zK3^MA|-@a1Yku`o85@ z_TK|=o}NXyCw1^hl^#~VevJJ2=G@Ekl(?%yvhjYzOK{Jhh=e;6B;Dq**S0O;#2F6r z{{3f+{Za}^ciWLndq7KZm@feQgCIPeXv;kgP zl#C_c{J0(xGr1dQUc^9Mc|2?qj~wPISnSb&!fnNHBT@+K6}~~`f-5jGB8}UEYdM9( zNsw`)vf<%7S-f;G4UZf-2WE<=@c%xmh4mun;dq`*5NgH_~Q$~nkk*Q^W1@upPH0lTJ#YxS==PMmSNFslQ0 zOKRYBb$xPqUo`1Hcab^HB$M8aiG*A4$$OyT%;jI-!<93c!*!a`%>=I82Q~^}klc3# zbH#+X)~i=@MdP?wHF5z1hTX7Lb2~rq$UN$x($6T$yhqvEbGRjPCyeBna3c zT$$g7c#X*dFy2FAjX#m*a3k!QUIRYI1G(Qt)3`?SL%3ztrrfV1wp`QxWK=2p&Pc1L zz_494uXinl@JqR{G3X@c+iHRKJavc_`MifY6Ke=w)E;Gb$&e+BB^jYvMl5f@k9gO~ z!}C?OB&$*t<60weZ@e&0s#;Dzl)Io=ixSS5EQRu4--1K(E#7cU5$tNJ!3uYE)SL2^ zb^m&rqhzCtT?1X1{`V=Aba>+&Q6+q?tHo_QCd5teFy+du$Ye&1>zRyqlIVJb4|=bn zX@SjJPEN2Xt?S=G{HxV4CqS?i_1a*D(L@{=m!Ns?Yw2;v4dCW+6hr2TaT+$}VZOXB zYt?>$IXoQAT7^mzx%55wc=-?vz33Fo>Y5pUdXi=?Ud~z#+~)l|+W}idkK^1LMV{)0 zQfyXKhL1BQLFKD)_GW21i658`Q`_x9!g~{`-2Mf$M$OqD%WcszVhmOla&hDB?a;2a z77G{5M=6th&f^eIh?kVZ!aJVWcfuPJX8EI}UK#Z3O=O~2Sv(v3mUOk);d5)kdnI`l zv>d}0lV_ZnwCxgjxULj0 zM$3Vy&rb37iS=lH|DmmwtRwIqZhg&ZYK z%nUQ4hknc>l|dGuR;a{16?TJZ-d`!G6`F|ZABsoc$6)F~2NH6)5q#1=l78P*Fg!E^ zOIQ0~tmrFN^sO&k&b`hAO^_rV%RJHY;50h&E0`YX3FWNS)PmrJ3DB%O&k=TS0t{IbLp&NBG(tCTUwY-eW4iLw0m zJIS2YtC_c!$OL+&SP4b^vUYm=p65}@!-+p%!g#^{K11{O;w~V`SWDhrE+ClE_F?C8CWtlHy zni#cG4l_+7;cwXsaM@%|R-OvJyr@-#iOhS&nN>3d!(OdneZ*(-BpXk1e3=vER-_7} z^IZ#~&3NFrVLJCfbu;#gDswGPwc?@^%TPh~NrUBaO>RtHDAq1NjSi-K4Awe}7Bi;e zjgMTa?6IBlRSW3mPyYB-B^D>GKZ9Q+)VLbUo3Y#{2g9YGk*0_B5OC-^oXpH38S|qd zY(g0+54s1(X9S^pKoaWD--Y`hrJ~f#bX+l`7kq6tVk}W)Z-jKio{3lCPWwT!=u@Cz z4;1t)=6=Tv&p6DEwncaImH6vUKk2xeMwbki(YLb{>8{%mkRU6#YrGnyTh$NJzyvdB z@C~PiI)B(bRq4>=y%m10+>Kwe-EnDLJJZ+L-5~Q{5QrVKp#PLMSeN`Mr4Iw7$-Qlw zbZJ=`@BN2nGIXE=uAaEfzVH&{5p10f)*&{a{%97};yJML8>NYO{v%psGxdiVRCY>amTvGsSwme6c+kxF{Ep?_Ii1EZcF~M)$|hdqf)r3` z69cPSiy5x2`4f!xi$07vzhF(f!6I%y!Vgm6%=R&sW z;sCFF*B@q+AZPLTLWX`17vdDR3i#+gSLo%&MO19+AYEn^PCh=ZCB9pg&^&cARdx)d zM%@jBac`xffiw6&m2&9DJ6EXXnG|YtL5Q!_6HZe~jfv4|X`G^?!}#ZC@?H(h=4(_v zpf){6=-)$kEIG?(an4we@aCy_)9H`PX<{7EyIUAER}^sLB2w|}qw_TJQ!A}nKZT#E zG{kB6*G69M-^H4$S5og^Ih1>%nwCDh!SrecAoo)mEd6_p&XFyp;&ttUe)=l%JU^0I z8N7p87v4m~)pE$>tO|N0HHtHFxxPcbz8o)Y=7UV$(>RuF!P9D&Ny4BoNi{}bWzO)skRr)jQ)jr-ri2|+Jj zPc8-;$wZjuCQcSizfHxP1U|q^2~_5%0)L{`YZ|*WoBGEXTLgUBa2{5k z$;0=oBjWm0&gf%3qJ2o8Y~Hb+H7!{Wf3r_Ag>TL=J!BHxH!6UM8h~G#4H-~&>kTRTW=Lke zUPr22W5MiQ9eh!qK`$0wrhzq!X(V2UPzyUanLdj>_Uj&zIF!%CN%xYoBaImKYYtS+ zx=D2Hmr}#VX0ok?u!7G3Hk5vajCHfXO~edSl|pg0zb=}!4iF`^Z)}=*4N+262a6^* zHbs02ws}q>)xMdm#t#+psO%;bX?_5|ogJk9xdYRst;TbhLD(djsnmH|A?b4OgtG%J zB(hGF*`hwe`AmeVSL7bLd|?_rF}j}W_asnZznNsqv5B$0OyIadJK8~ zBg;xLJDHjL)KieN*Fp=KEA;fSLYmOLgM@FPZ1EpeT&85g-}AJW>I-0nBR@mw&KwdgwgI=h%^3i|19|J8jLWa)zj13LBZWU4wp zmss&nP$N$R`aoBdhAg|yJ~*$$A1r=O%V*nBHs~kyclkhMiw~mx@oJjUEJV-W`bmN` zi|BGppsy{$Sv%)BRPIL|<-aJU&yR5VJO9(;|L{$xb;%p3mtQfxxmcUuuvL?Wy|TtR zN4~($AWbU7wV=F%&Zsb!4u3KQKIpfUhb__Y`Q`=YOoT4yv`8ke;L|(UbKnTmRlgef zo~5KDdj-90(n3~=uA(M?#|}v#CJjU_IK>^X$f?a$4a)n^12{*PKM@qwGiijCm3yU zO1jUpivh3@Op4QNHB7hb)~tFaV;b z1M%TiDL8!bC1a5%O4}vVft(s8VX=`kw5yzq?I~hw9iA`~b&Y7o;)P`1`2?b{^#$|3 zxQ^U+nId3gu+(Hsz*GnrXOFz}LfHh);GT){ZtKGKk{y{34sn!ny!L%7ItihYb$!B zRfFBx%CXa(*O0sQUcgTb;|c=^x2UYgS_NPJE}F4FI|kjKtMemk3|Y(M zLGldAF8RylGv$l#tIBT&8_R_e75vW87~HbF3k!Ni^McGN@|Ejrc!cpC{&z(iyXx4R z&wH;5HdVV3uijxv?Oay5xK!S`x>errwn9E>jXhVrE#U@J_i~N*-T5~Cc(e}xO?rfN z=LY}g@rqa7;l}rTSfex_f|hQjy+*#p-G`5&yUlnuSpO#e=`DJ0_AbWJJBo^bPMKyG z9AzvlitP)N%a0SEpC7QyBboajs^u@QyYXKU&PYD3mL9f^k@dD#mUrrX?$yu&4vm8I2ROP28ijKyS$iMVS;5{o_ z*buJ|!u&G_WdZ5mS@cX}uI?zu`V|euDsR$xz}y_Z*R8+kw@Ai03aTt&<42r1q7yCa z`?Ea{BH7o1_k6^(1#sCn<>FX0-T_n;l4(-GdV{6ZnB`E2A0{rmIL<>Ps|iU(J^Ybw8g?gtOJM5)%d6L95oK2MvO z#e;H&$a@MyLl4!Z} zJ{0 z=Q{e(*!7fusEc6^XXl9*yFYxn$clH#JBjfNPh6R(K*iJq6Y@|Jjx zDY98uTkA@HE?h%Xq(j+E^A+SnpCU*}&}W0Z7vt)M`taM=1Aacd1%VB1tf`(TCGZDos{c;6dWQGK5kzXpe~HtLFSI$=BXQc>cYR_2gCeWze13Bm3K zSK(P|6iYajDC@V}mQQT&fJfHdVAzApcqM8)#D8Cb>6wplmQy4ydN_*@xmeG&H}s)b zp3TOhZVOQ7YYI57R^zuy+_86P3yU&P5?y>|0^4?h|2sRFt9w3VieKYlQZ&NlR27UV z6 zWg<&u*wjk+ZfC{m-`%_+x&R*7*^p4n4y=9gfG@L-<*7?4o#G!!^Xv7fO5QIVuC7SW z56dCnClt&4G#`Si>V0T@yp*1?HAeL%H4r-7iC>vlLAo9!q3$dbzFXB(>j8FgcBjNrwLVaSP?82yg+V+uB zO~?6k?z27gV#jdWe|sW>-L#D){SJ<@d$hR^>0CKkaM`+2=5zb3eb!EK zuX5!l4Dk=bjK2x|-j8ee!=aR9j4{ST%T{CE&}MRD(=Z+t^p|Wke1of3T;h`PQCx9a z7mnCh!ba=M7~j@OdU`AKyx-XXMFbWq>hrC!udv(3FQ}~X6XhF*@vAQlFiO_R#_zhv z+#<_B`fMxrzqb*){Tam9l}3m=fC&D@6Zoa13A{Y^1Kah;hzsfmqyuWF%clM;6K31( z#jh5RpeE|Qpg(dI^SLJ>v!;lCvdt+}Wkfxl{6b%{;;JzpvvdP4zkLz<{15|W<>z_9 zFW>_lcVhIP5%_xgROBm;Q`6if^x;iqT6JQs5D~<&&)s0I7wE&ac6#GK%foj586Btr z!DzcRkWAbZhi6s~;v3)ngwA6-QRAR4RZs3ir>8XF&J#0uuMG-BXYm+5W#C9`zxfH{ zhv(uFu?x0Uba1k+He9l^oyF_yC$klk@svjh%P8vuQHRU;v*#XMLAl<}#jqb0e^}|V z%WtU1fQOiQN1)w)=CN@U*pWA?_+i{({(g%Q%WHXu*_R9X=)Nj=WTzcymZsu|`7wOY z$9z6yyPjNr$DjPgyV$&4Lj8|C0<$hk<%5#xgS%qSToJLCnlArjRgCei(=cJc5sVdm z6g@Nl<43&5$uEZtp++BWiIIA3dgNpRcpnJjRcb!8PvbgXIR6h{^yxHJx5(yci}IL6 zWGYss9c7kThuFCU1X}PLAMT2TUOlEr{%NSAY@!2}e3kOq>#o4}&`Wq|uNG|>t;k8} ze4%)?7mM-ShPi38Y5#d6sIJpEy4YHiT3_;z^nCCP_da!kP{&Nv>gLFm;?5!YS%vk1 z*CER83a%<<*cg({V>V6UJ+!`I)Un$rkNv_b>MSrYrze-|pT?^;%`C=FpJzT zcgLGMYU+M0Oke;fwxFylLIb8(-Cjohkn3j6!MmeYsY@-K~s@@@Zm$h$V2 zq!ip`>=@N*X%v1x#4Ak#4&44hEl?1qagi!laE>^lwKg6!}JoJFk%} zDGo^R{cEJAW)3#pb+u32^;%YUbiCl|>taVtt}*IRM|UmQM+J7DZaaK|o?W0!&%9Oz zwbNd7v0D?hh}lOk`$~C4&IW#5Zh~i)^n;s*S3$v}A01qjOD3KCj}60%tTlTk>^x%0 z6P`@x+~oq3U)ai5f4YM^iw09^y&9cXYzWofxsuk-JrYx;l@b;7m0bI}QnFbgL^2}z zuE^^de`len0s|`+)!4p?W!VU={!Aa^KyCl2i9$sQG_3G;C=- z?6>Mc&GtWqB_EdaSLHFh=Jsap6W*U|o>|Bnf9TMx-qBd9)rA|**5fSMA8?&IkX$uS zl>P1O%@;`G(SO=Vn)@e@IzTG@Qs)9Ab^p+Y$Lf-!o6peXgf3d9c8u=0_Z4;?%BFp0 znbD1x6RDRVC-dzsnbC#)@L!m+gMJW`>^L`BQWBLz#^2vg^Gol*tbhibH=#d&|H+Eq z`EiQ=S!hi|8?V9Ru38%R{tDfAyq;ET<&kYC3yHr*xU8r^i?csRFz;o5);wc88SAji zK4+~SJJYihcDP)`pxAVH`~Dvt5fn=YA6P{v_%H~b5lTZ$cf*+>vn0XoW)k!B6?CJ0 z5MBOJbk=P+Dta>#sNw7sW;8Ec+|BQj{7iL_42`-@4@AV!mKhgl^n((5Z+<=1eUL#5 zgaDZObRYfcu!UayF@`S8w3a+iRFkOmaFndNtwy!&-@@)oTc~yIQ@VOV7#+DRA7))u zBtPp0NGC3_8JW^SzuBLo zfhL0_wXIiZ#ELZfw*MHq%ep)4j*rI6M{d#^;eGi{XMGlKP)j?^AJeFXQFNo*8&Whh zf-d~em4+744I#ii!!EC1)F!}R%48Ls3>oVgwEM_p=)Aa-vUwOvt5EvNsb52wxLYD3#` z+z(IgXk`PTSI)A!iH8Z_p(*(+OsC1xe0u%Haco=N3*Ougk*&LF$92YLu!;A)sZKb- z#sSJ?wE7Ej%LTAU`EYDppoL)`9pLAE0gi2228%WtN%H&mkz5)$L?Rb6*)CHI&`zZX z3mD%Ud@W~@D8sR!-lc=FKQck@?^QfEM~`X^?MB};MA2N$sr2xL$8>tIj4m7KkLRln zvaW&sAZA4to(cX1Yv-%d^)W`^IVM^b3<75E zpt=ozvHR2;c(-sNJJoM7xzQZR&NgiomPS6MVb&$I)VYcF=uG2jjf(v3i2}N|Lq?kC z){}4Lr==xT)2L6~RhnD0nnq+RU{(AwSwU!t%(<}-y^(qqgol2BQHJcy7zOC*lR?Vw z>+*0%H*(lf$}X)s0$Fqj-tF%TVcL_qL0dFhg+`Ot!D5a=S%3si$gx>hbpl+}~hDD`NW7L3_kb z;5H+8aNALMYIFzho!%tD%rHsvoXL`3efp5r$9q{Xrwziv$EQfT^I4g7P^9DsF@(7h zA~!Fx4&WLzdclP`|YoaGgGPQwU&9r_5=A@njYlmrAov=GmY3lae!fBS`c+ z8#-d83G6aZrc?CGd&oZcBIGv+EB{VZZ4Z5v$%1{2Uz4&ZOtJ zC(>I_y0Og`b7^DC8JhCFmd0mQ)0pXBq3H4vX7RBf46W9M*}K<~0UE!^sJB+6x%`a) zcN~R87)ST7IS1j@rC@b(DD!dMPh7Nb${x+=L1KjB+@+DS-t=%D>WBLsSuakN$Bek+|?2dx7^p!T|c#;d2>9b-#kt9^>fe)WFn)mf>00Jt7xc z1C;5O#1&-n(&O+tW}YObt5o*yPy@Mmx&i%$Hqn3c^dzHf3?;X__VaO3383#d8R?Xs z%((i5&@p5tD|B6n$G-cJEBiO`m$&X4Ph7enZkQ zitBn!L5HgWyy`^{{w{s8=+CjD=lKTGp8NyH-cdu@FKtq+Rml|;v&8%CVLn{wkLo>a zgv7}^@!f%`EA>?Qb1G$g#y)*^&oz-< zcFkkSj1v&ayp|_JC@-CB3IGk-3V#n`2*Ba1Yf${9g70JGP(XwOM;* zwML(17yIgQwn?8#lry3E<{scpVS?qaTr#4UC;oE!LXuZ!l762~Az!c zVULfb$Mzmd4==b1?Kr& z#&XN{!KqLq@$S@OH%Az2oDRBQCg8eCU#zUlK#v=_eDi`-ZfSpt z|8;nU2gcW9L;QC1b4X;%O9#jbhAn0iaW7jQJ{AZ6(co_LP7{-DTK4OX4uRQo&yy@; z7pB|YA8$N~6c*2nBfpm2#C3W5$n+UcrR(SKXDMcKY2=@gB=*@FT+nPq6?JFOEf;`Z zUUmm&PoF`XZni+*pPBsd%^BQ0O@SI_e30$iH@caT3*m~gKToHSnVpfswRwS9lNu+p{MX9LN2trZJfqMw+nR+9ctE0; zO_W)y_h+&&bsQk12pvoPG1M*`0=o7JuU--%x$O#Z@-&ydduJphVJ*A5dJ-#~YK{%s z8Yl}l0jJ9xrkVtT7do(sOZvjLT_MmmC&At>qnViuAA?@5n=s=@0tSoDy>zE;vXPUu z$WNti?2?KPbJ%1G7YZcccD(>6+&hE6O+?@I&oY*#mVmDHTX^axCo)*Ah^fW+Ft;WL zu#g3T5c$g9d-Z0b^{P!cw_!aRlwAb7(|cjq#SA*Gz71TglAyc390sTv(nMtqI?`tg zl)rR@%^_JJpBe$7xnc%s=p6de<|Vu;-$-BoNv9WUoav~$)DM@a{^_A=kw!&*)ZV5lWl@VU%35w!CBKw*F%-dy4avS@=bKfbD zGHN6BJ{Lw$u9->yDE^Y2lefU#`=8*MXAv}wPayes_Y2Ll9>Rd|d9b3#WBa*LCk4UF zfZnU_O+%U%Q_bS95L9vmUS!+`>$%-&z;6#aadQdX-Y(EbLIdb8^q~2RhS7KFS73v2 zzL*KSAQ>lQN`_4jAP`h67(GmtIj&zum)FW^{mi3O`PY4R?%qAJe?%p$Z_uG@s!odh zr)NZ=QWb6`?4dG~Wwd>#9yNS3l@4Ns)cJJ?tvo!KY9!}_fd!JU(m3+g;ibJqZ3>Iq zG9C7fUrKvsUt-Hg9b^+%jvym!)C%96nJ1*M)w0Iq@pMeAG1Ka;DV?0C3yCUD^n_-`mQhkSsVzdo?upo2W!5d~2z zu7lqoF?+GaoP^CgM!J6-Mh~ubq?LDPP?vdo==(Rpuwlwp_GjH8GVW~@Gk&3hJ_pkX z%o6CQLAvyl=<7K7W;-nE(*{?CMfCfFe_*|1Hg#RQj#hrT2=8uL(5^n}lq~TFv+M=1 z)Xu5kTUF_>)HLnnM}rfW{5Gd0nD-QZgbp8}7;hoD^9fWdYU{@;B1thXCg zcM@pkEj6)Y^O5Aqr?52}N8=bjf3Ww|p}(9B;FYEqY|K1I_WSmdU42oB-z}W5TtiQA z(Ow4OCQXoSaRhGpABP32X98*c0O|dU;eJ^gX_+{R{nd9P%O0DNJEqp8UG^Am?X8Af zJ0&_}@etbGz@P40JcT~;J_G00E79{;e3<*glkDRnBeHgE2{G`nArbjO@XX&0OSbJ| z9ld*@dwLFLT8Vx8U>$z#-yb&i=v8K=d5l~gs}CV!Cd2b|s_dKDK?s_c3zL2P zH^GQ#y48`$IWh2Y6Yt!({-5zwp3`U5V_%?;2eGz!xt&pJ!c(D_M`#5+-es3x{{5lbc;OFz#st zyD&P1Zx?eO;~urKvG1+;x|vN_y7~x?8RW_C-#d;g(-yJC<77-jzm;iK_+iEwIk~## z4O@R^y-zZrxm!?LtZtF@|kyb2mndAb;gEPsK&5tl| z7Pq%GEx`(TIVx>_CL~srkfGPQ%k-^YGR@CFWbVte*>L5BGVjG^f^_^|!OXvl@XnK@ zY}6|ZX8p;xfDw2+O@om)%h`I79gw^66myZTV@q}>3NeHC;`P>}EbN~y>6kDI6j%jj zY~Kq7UcO}Vn@c1%atS(W{U(i_6X>(MN^q}x3^aA?f+<&9c^=e1NAHroc3%NUK4r7ZzfSSrvIg}0v`zLxUjwyQE+hVyqoKQ7h`>V&NKJD; z^1bC8uX~}xD<)J4=r{{(Jnfj)s7U6s^fp!)l*%fmrOE;}muyk_5KmGTjODKz&15}wgCKvK zJzJQ`kn?=|OMVQPWC%qI8s*^yv$=-P|x8Y+>`D4i-L4i+!EZAGFvg zI8S!K$hn&k8(<4OC=8CfJu%#y4+>COJjh#@Ug7Gv%9A^5`d6zqBt z!l&=A!r!N(S^qN~Ts~kwPr7A>tE(=cW6=m$YGvVj|(~YjAU1FY2=7D%`m1Aq!16p_^Kc$iDu5NtCZmr+1Z>P%>~MELiW) zvSa>|-5LYvjo&q}qgqC+&A+hGPjskPjz;(EOD|774#i8$AaIEt z)wmxHn{!IRVB`imRWS#TJ|7jWPKPLpjvg_4k*1YH zW~l9g=0%T);re?_Vdo-pVy7PADtet#(3x}3gINTuJtZ>J{=UPI=m$HeDE8T55P zun&7r=4Kvb3%`!0$3I$7V^<5R-B|#~yZKVDUUTW=#ev{=Y!aQH<3+P&1L-+;W9r%{$0H!Uy>BD0x zRDFCK=ndORwQr4~KXU8nxW>Ws`QLu<^?V{M`Xgq&|0>aoFWX`0#_oLG_X6fH_&kx^ zy-wOLy2H~nS84iV(WkQ}oiyDrCZ9%XptH&e@?gVaRe)n%f^{o_B(>*2DB-)_BROuqBd+zMhhBw@{p{911-^k&biN z52eHYfrEV|d_KOC_7mPgLi-!hbKRfbU-c6HERw@SH;L?x#dcI$Tn^9DAO7JpF z1((orGHdY>_e61EgnpCG~m9=MR2t_2v^do(W~csc2xT)dEczUN4_m%Hn+_2 zMVke@saznQH}(r(bY(Knok5T@QJE?@h6@v{#})5dG#E5Deii-J&v{i%D*yhxCm&X5 z%~#Em@Et!U;RyFbIBJ+2b?XYbaQPt*5%*Zb^;P)zrLpiDeCXgCijpUyv@nn1D# z?IV{rd4OK>aGZO%k_A+J!mrZ~;c~AMQsXNpeeJVw#6mT8I4z3x>-}0bj&38*e(aMq z4492I<4`!eUCPXyYf0+j3X-*X1US!^!IR=;kY^D`CXL@Gd*Ak&eO)>n zmDuB2*;VFTx0Wk*8_&0ew4wInYz&lkvD6!(7&7Pue*3VEPu`ctH7zyyIj_5He_s<8 zn|d4%E3IOF{?RzG(u13pS@YtVYTW1EWNf?|fHw<**^in#(iw~GWbYGQWC_!b$mV>} zgz$z8VaC`#Fe#`+fV5_IdY_(r`-N|OPERS%7a3F+_uKFtIgO|k*q7UXK7`h;kC-HG zB;K`&#T%zi;nc7|6h3FkX0*!5Kj$I#)}b3Px4S8?eXWLJUmUR}N0FV^VyN_O9t%17 zPS_-#c`~%VDR5tMAYG7e17LS?wU$jj_X65GGZFWY`IPNd_R6=PX_)Ra2^-c z-j|vj-9s*)Y7%Ja9@dbc#{b1Q@B_O-c+t}cE_ICKwn^&vv0^IkeQFms@AP1C*Yn5T z58RGPovsk`eh_5OZzX%i^<)7H-pkG>rU;eOta;wi-ng=3I&XFOi2shh!O_~aIKU^F zdn`T1N6h=ntp{eH<73Kq`>o>z*CSwLxe|Ms>J522y=8|#8uOTsd%4l>UbyLJEBj== zmxY%9M(;~%p!EDY{B?-rmx_zIja3EDFw?;6Vt(1d<}Yrozrp|Q*}%i5Idj|dO5A+5 z8*hz@=HYZ0mWv(x>If@-wss!BB(dSPWktBQ;4qK*yqvc`iQ@j-GO)DMj5Yce3ePP* z2}1f|v^460v)j{b%A{e1O>Re~MFnnW95i zF68XqPUC4k9-`Hsro+#) zn!XnBcxE?oFMNbcx8-pccW-`l`U+lV@5;YV*vRv?er6hmFY)~TTiE$KgN$piA@f?s z!r{L8`04N)rkHBPw7&MID!Se1JKb8*n zb0JdK3OdAG;KPNEY}*qTOj_53xhb>B$me&++`A#HN2L$WDeH$GD-!Toz9*brz5q7w zm<4Js8~8PRfyE)Vyz-bt=$q7zErC^d(ozXlil0^HcZ6xAzaqo{P$&2t{xq2ae&6k& zv#LK-I!t9{V?$xC*%eUhHxRl;lt9WBMS5u5Dwa9ej%A(hVkVg#!nR3=#F^I(_H&u7 z=zXu3u6Zm#sMkU!TlS8Xnz`bL!gK6uYd2PCqsy$W&H?$T@#MPQI_96zTTu9H%pQto zyJ>k6IDKX(JAX8VImNGp1B3R$wm->WEDvKBx(2h}0R#EijC~~Xx*mXcFwuWCSgsx8 zCEq@LrTj)`B>eE33xStfNnxu2+mMsV3~k+H0hvBLv_p@l&o4&%HFx2_gPEkzMHSbN z>qeLUw~kJ~)&-Aa{U9zO9!AW147FR2O1;->-yf>``J8hNZ5@dB#`S z(Ayuu^}k5aD;$r-YN|A_bGxj3_E>UpQ4Co-Y%NZ-5qqOb)gZ2mq0{Rg87cLW%`lg; z_8}_5k?Rp`yLTotU8^qZG}7gM0WG-N-W*@g4MN|3)u{bck>BaY$+3IW$;ubXvgn+r zsCND;Mr$Z?ZV-#X1p|5A;QKgS?+TdgAgFU@Fh3)mkKXR7*fP|P`Q|nf{gS>AFLvqQ zTTY<01K&c(La{hpr$&`-e1h>gFQM(>Z_)Fa0=f6rlb5r0qTl#XvHBiQ)`(pCYu_?t zf0O@7Hw`eP*XzO|*kwG6^|T{HYFpT z5JMKdy;0CIHW#wj_(`|;Ux(Xy_OzsDDHP;C2H!!Nf?L~PS?lc4)Ma2FsG{C{T8Jvn zu3s!$P`3uQZB)Vk9tRInyQ{9a=Cnu3pT2-G(Oyw!+bj3%Km#D`IbV zl;vOCCpg>Rl#NR|Kvrk{gb{ukXrC+xN8K}G@4FG!s*j*s%2q*tt5~KQ(*X;Lu8TSS z-DLHovf{(0v5@m>7#$vV0rtH}B+u)D;m~y#*i+aY9z`{iyqf1MZR9l;G_nR>2eh-b zV#d;INDdlX?2vAajATa5EduE^79MXvvC}jSOj5I8=i)G^U3CgZR!GEvV+5O1Fc@44 zj3I3>@MD8+GSz@q5ZanxY;X|#a%zQ5XSc%*XEjLf(1fSU9UeMXz1M8a89OL9n=gxTyAItxC~&Wi#k}fK zB>(YTn-~lUhBsweK)T(520Lr)h*rl(Sqs>in(IWP)Cmt?T0<_TM+)WnqsauX?L=X2 zHd)^13&spTg6j`Ofc)ff*qJQvXi={+cL%jl^BSs-jOrhgJt2AgYqlWcTt+F$C zWm5c?{+O{<7LeDu2GZT%vOimrNzXcUs`}wEq?e}QmN{9Z==o(RF7#uk4{Tw2%M>ss zVm2;4`UHP2?vMA*ZIl&yF6Nc1KjE{#r`hFRLwS|R)zx?+`z8xFL1^g#CW^mSp{0u$W1 zx)hJw&IH3yIkU*SLL%CtU}jOCbd%i*_EA*_pQy*NS1ad>+j+4BKDXP22VmYVRc>(R^|?al=gn~ zMcONHXb}6#{bVG~r-FF*-A!C<7&%}|NT&0a;-039xwnMLgUfGafxvFW)yL_Zb;H=Mncl;i|p8o zD?&iQW})YVMcDLnv3MU$6$V@WAn6}%u<43A#=JbjKFC|pH4=pHmnEoaeu4?jF9ff; z67pb>20yt_Nm}Flnq9HHl5&oeQztGk?<>@lDs{R>uq zlVH)&QG{125tHugVXO8K$d8=_y_J*%CB;VRnU)X;8)?S+ck2yLZmU8-?lt(E+a2}} z>j{D543V0T$8BpL;=vb{!bGheApbm+D(oM`8V~BS*k&!%iRFTt&KP2H_5jh_sZ2Y< z6Tz5G!t0wKgUj%_c;HY7zqV!JQ=K<2;`RNjX0bPr<~S9+u0&WO1LMu zjZZ)eqQ(czT}IsAcEdqmzc6PjMYv#hjE#I92`QguFr%X1q^Pyt{!9O8aLT9z(q|Ao z;V@4a<bnc`YMaQ{t-}QSv#aq^PK7X_&I_iO1VW#SQ8JtN#de*K4RGofasDx( z!2U$9Ub4^2Zo`xfbJ?`5hghkBH$0wp3`&2=z~$sUc>N|&I26^21MHXMuB~%%zti?$0CU>95(rq#~XU>=EgPL-8d`W% zc^L1-pGyy5)U0K!z|NA-v{L3L?ussj)+lZ&-ircj&lCH-s%-5x0T0Y5Wo0{tk`MC~ zpncPi;;jyM*udF^c;uW1kB!UV?H5OLd&?(+|6W(&jJ69~av*_w6pX@wb9YPeXd<8F zXU6~Qt;pl%jw4^}eejgpFf3WF!PS~eN#=`W#(OD3_vR^NvWpJ3hz`%+o};mTb|SOc zC1bJHntXV->8Sht7g04{#0MmN;YQOT2zu>%+L{$^Gvj2MKT zGt=!ajTHDmr|+!n<9&8rs)FGLFJ!aQzT=QnRqWltxA^Vg0`50Z4YsVbftZ1NWVIP4 z+$gSy_1=uQbXv=OEIQ@6g)h=b(NsDk{MxHuq zwcjJv;+)e}D@QnWHB%^1`o+8a+W51@-?`?D-(;@boYx$RC1i;X^6$JW7`$K0>aKdQFW(}d_`mg-l2n2UDmwg*h=*HbH-~*| zHNw!~?kuFWSg?4s5Hbg8b4#o9%)i_jb)36Fwe=g~Tr3sx+wKeN_U}d0>jolQQVGMf znpuZa6Ep0i$@V@nz&Xar*rV1DV_y!!$v3@mMBjzNr|#3ugiZf`h^Qt^}tOFeb9PIPx`G#Dx7_%&DCQpWVIeWp?0AU*&5f241elL z7VPk3lMG^IwTrCn(~?#S1%nn92l!X88TSo|r`lH>6PL_rpb;)lixK9{jUu<#T0mp# zXu)NhIl1c_!*=#~Pcl=!6E&B2nDpVih@|p`S1L+$&2A?g>iR=EM(kc3Ibn#|J~w60 z;xCcns15k)YMOWk*+V9&ekScd_rcDYZNzAWHp@OZA8*_pKy55VrhJVB_S#h|JMq;9 z7TFm=Y< zA%(XwC06uZ?481g%MNg#9YfK`>?Aat*TL{@Bww9y^{J!?l14R^=GRWX&n!4CRe|_QxwS{lE3n1b@-tZh4;#EZ@VV z8&?TNXO`meyE;tqzY4bU%P2hXx`-SYep1-~(j6B^jKfh$g=|;L4_WT=W74$KVKAX8 z4!lKY+ftP+?6l@o3^?(RJwLc!X1+8;x>UbM(Q+>}G^YJm9O2LUgzRGrBX2^X({vcM;S3w2Jpq5D$njZ*1OL8g1wMROi%sHZf3U#>2If|* zIyH%yv+cI?=LvzTi44fdlvjc{N6ZFD;F4ksM7 zWmk;POYO$5VC7RC$%8vVc4eWiY)P#r8f|Jvzu`mqP5bMD-0=^)yWt^g)UC7MKgC!o z?O9{LXXa}5iTSftZIfBWAP?5hQ;l^i^<$2g&OvkkCvf@69q2tiL6)ibiL}#`q$Ipl zP%pR*ZWoi;MTNJdC4DCtToNEF@Gxlps<*#7{46=~SD7i9bjfygpTi~;Pa~z|G0D<9 z0d)my;Cyv0+`KmocI|Hf*JaCTRB1D;IB=6JTW}g8<{pROKRem2|B`U~l>=nmhAZ%2 zPlPOY6M96w2}k;s*njmT=XX%(YMOzg8|pwQaRns#X@Ey+gK+1|IyThlpy=8t zgI=ZP8>Qyfzhr!fBfM@Oj-93h zix0H9V>YX1m3QSh`}cP;R#AlGomrLk1akMI z8u+d2K`hq3gENK0nbQ|1aBmsH!n}Rhp^pJ9xR*L@{Z$B;t3n{+do&qj_6RPYiiB2^ zd)V`h$alAFVr2`EZ4G+_y?U#`oP{Dg<6jOd|Dp!-!H*mX3%QGjR`FDL6IX~??>%u~mxjG| zlM84+{0liH>11lG7SZn=2W~?H=#{_vH0am`Jo|eaJ{nRD1D%~_yp;X<_?U;${l*t)nEwSyYDtRs^2Cn54VEVdym17(EV7`Is%{8$Fpfpms#uJI(EG> zhin{p5W0$^;a9{&mlhw2imE2pDo)g&YO1EcjsN>>e$ZyN$f$MH_YlWh8p=(6w*IK>xme@Zi)K8g|tdUbj`i zf~M!N{#tLIbwP@mH+xeWS43xwP?a22*ONpXlf!H08!*Qq9;%myL2{BYBwT)Jr!W1B zrt3|J_jQrE#D-E+jZ3s~{b{K=~&LUeY*^4Y8ta{bK3a4Zv~-B;dZz2$;2Y1r66+M0-6POVx)3;wpt4 zaqi%PrzWnK9UI;QFGU37ME44^JpH-d+6xM@iH8@EQJy!Uy)UDJUVnPSY%|q=Foc$E zTOolVrzK0~RY>&uro!MM*7WUW70OP>v%3vupe*-*Vc(}ydCGRW^Ft&xwKZZTca&Ig zLJk@D(~P+Vi06S(ne6cjV?0*{V6Vc}FfJ-&Di6n+ovmb5Cr4rvo9zz2S>fJ#5a`rQ5X? zAYRN>{SQUw9njPF#&Hc=+Itv9sgRJm=Q)YUsB9^OkX=SHvr;KqL<3Q>N<)hJ-1D4D z8X6)QMRrCa>nnTx?(grzE39j7f`vndDJ+gz1%kFeMxa0x6o%#aC&d)}@ePhVh(dn#o_a0fcOC^rmpb3YZ@1XpC zUwqzYH#H7dqA|nz!(p9v@lf4!IAghpTuRa57tH5C&q2D>BS)1*yY6Ee)D#2e5{!9r zM5sTP&(0_BWF%)TDzYr}N!Df)zpZBK;sqS_WO#o@Jq%g7}+O*p2vf=n2b#;ko? zNyy7=DB{-W={N{~uByhb*EV2GAt%MTx6tlX3QtZRj;Ajt-&(|Rd!upO@R#I8=4Q;f5|3Tp$Yt7d_%(~kxGdh47wT@s zu#&mB`9L`Cntv8=ovoDEP7$C|91U(W0D=eo6F1j;@w;Did7a53r#<(yG3<9XJHFe3 zu!qi|JMx7v#_S;bZ?OSCVZDqOkIEqP&Y8ob10Pw|yf-j%LQ7b7YPSL}kJREy7w7T)hcmb^BmzhIiMZr#84gbQ4wBIn zd@}S=cgtbuI_CsgYYm}sXisYMyN)#-I46|X=&}VHH2KO&)1@4s0IZWT%cCdlz>KB$ zpj}^^=G~w`hcBVaH>?-Fw%%ac`G3JNDwrh9dM@Q41W7s;bCkE8&iqPw_jD!wJXOQ1#HcyRT#k1CFfm3I9elB-B-~YFSwNI5aTZh$o#W6jc zwWA0BE*r*;-taRH$f?3x z`hi%wy@+qWL-;hUGq~a40ElYu2`|QZz{d{Bd)c&#?JAZB1v2f~voovX~R#CI(RUy?sMHmxn!xYXefCtUD zge1*$r*FgJVStM>J0Kq`j(xoY-?6zkCCZ1**nXAQN$lP9#!BqFrxzr;?Ugj%1l3Q) z^DE;@_|L}c{By!Xd~l*K|CjAUE_>c2230|1k8xk}W}+j^YSt%rbhKfTmpZ-kdJSny zDJO3N?x4=kJ?zqkSd1&m#Ey=6toX`uXih$g6FTo=Pd_JIUHOTuS~q}gnAj>VD>+Q^ zCIvyu+1Ko6%y!JTUxG6SA7M2cHuA4a*6@s@AK1%;4KVD(A<}84LLByc!*j2`PO3-c zpz2i*ni%{Qj;Wkua(c@2W@=01;N zQ$Dit3_D!2|0l-XJAng?XS2wwYng6EI60oZWyIN@*Tn|kJx*CIBT4x78Zz2b8y-|= z3ckAsV{&mG8TM*0ZdT3537KAO+4ENNV3P*jYCT^}vARRr@8~hPGCMplDH?z6nTX0W z5ep3k{PJo5pIO?6|F~e!4k|aW);e=Ow!n(KjXS}Ay^P{n^)|S5#!57LV2^vgZez#V z28%u#8cxdTa)MXGdn`L;&GmharF`EuTF1o*2yi<4--JeQa=A{aFYq62U-`NR2UhAP=dp!QQ6@b^%7jT=u zsl6 zH#zMZf}?n&`UI-$BrxwO73|>5!$kgArP!?BOB|@vxcXRM-i!ZCr%fyyBnpmsefey_adwcJL zm)hzuFfR(#3m)UkfXUc-cQ>1t_J##taOEsY#@!C~M6zMv{uMgils1Ur)d>}CwGD-N>(Oj0h z62~n+i7??FyD_5}4T`Vgj2(l;9$L}j506nW742ZA;R>?$@@$rC-zmn?Ka9m45}z1` zlkCcMaN9Qr8A(IT$Q9l#yQG@h$eK{hK@!v>#j z_*M0pm}#2=;}$8wa=RyJ*gTxC2z`&kwx0r7On2IPJ{z8vHi#n+sA52y8L$3s$-6FV z^NSX7sQOxwsXea1(Rpb?{P`%%S~iU*^&5!ZOZMQk(F-xC?LT~k+Sp5J4(h+Xh6&wg zQLdc|yRE;0!_5P*D$)$r>zdJ?66@RY^?dyJ>?Hg4Rs**WzRtQ#qgnfx^WwFy!$_fj zKh!#G3Cls5DpV-b*vM?L_rFoBKw~dN+GU7Kwnni(BW&@{rf9livJKdr&0yzD9H?hT z1|4Vqni}bgFl)~es(N`gox0hM?)_>zB{!1nnnl^0K zx$b<>Uj;h$V+cL!rwOsEx6n4$T<4vPhn)HYLIBRf#MHBD;qT8H+X(VaREF`+ec7Sf#BEZW>=2(4H5!I>?)VDaFOY~j;b;rWhI0zIqI z^i3-y_lkqhanay;*&e37wuiq1wls2zu#@J1fGk0zYGFr4ubL0{*z%IC^YI;k}LZnNW8DS^S2?2&esxeQ~Nvrpn2+g zS}A$NmRdf>iWQU4{Fwc-%7%rnk5ze5hC$Zv)4VT@>YiEhaNYF9nTk zb=J~+koi8jEm<9msU%Nuikc`X{eB~KJKcO^41;^QY1WRjkcxqAxz3olm=pSdC z5pbKOc>0rovi_KptbpMS)uKUkPxgCm2(yU}BR;#OJK?RZ!lRBUByZ|v*tL2unfqa% zXb_`>Zyr1%;x1$Ow4t47O;hDhd_}C8+=s+ZsKzUxhp)ERurEBCxf|EAMMi@$WTzi| zmhVqa--$u%euil8a)M=AOvl-2I?y~?nuC}OhTj_QaJD*#R7?6Gr{i6y8ymzkozmgS zyFXx_aR?S|OJoPCeaW;ln}yhk$ylf6j5-;QSPm=46D^;y&rlU?Klek_Sz95d+;}BC z{ks#aZl=TIqU$8J$&-}(?IFXHXM;&264`(xSpT>fe!H~@?vp4!GBx4`gBIbWc01@3 zd_%Zx>klK37m^otjo|Q09SomeCp%ZS3iU&5=zE++pWr`Z5~( zl;=T(e4neNU$3Ag(mV2Ad7A$f!hsYS5d_523Tm9I~QSA!Wg3^qy_Pr)RE|vR}K= zN-3vK?o@%WX3;^CI3}LGygQOTe0dN@7|q~3`U#tNZWLadFcD{K$xHdO6Hw>gPJGny zkW}{DjVm9$7Q-i`3lHWl5UcY(I~KTkkhV)j#JuXdu%_idFnhKWXVxAB&z^})@AG6a zDeDHgS0GP&dK`z-Hrc>02a)dGEa2PBP=I$w=#lnN`l8)T%2_cI+`qL8Khqge-qKx) zBFh2mIiE?ugk9*fr4ah>J4^e`y-s!PztO)VyQuv;H%`iXai34BIN*3DNmMgnN)3pH&5ZQFH$9#=`H zP8Ho~+gmmICAJf$Kg=b*7cHCvH3mDM3>@ODk~tEZ^b=XXTgv!2&>RQ-K1hmn#}kbO zgJ^pGIheBIC$V`GM}Npu5<5y6z|P-s-kBuo@bxu~GY`NmM^dpk@tdea4w9&2Ct>tC zY5f>I7FHd(CajSQCE~RtOsv-9omp3~G2}ekm34|3M{FV=9p{jcpuVKvk3LYkcM`od z{VxO z*-dK55{N0PW(DtZ*_Hp=VcD`4;K!}t(mG#gdr$!}1M{F@@ge5CsT*{jQUY1`6J*;; z2XRE#DUx0I2^KGlCyPv-X=<^F^QDSk^u+WnwC2Gtp)T|t8ME>glzh`9!{!Y}UFJ_< zn>j*>uh@Ni6j^p|I^>PoOQL4&!vVrL{5O)yoPXBS`)d01X_G!p*Gq(cKexi}b(_eg zuVrN7Jc&oNY&Gq9GJ!f-%TdkGCN!vL2FU(@CH{MU6W(l^j)QOvn;Y98hM$zl*p6~S z>Qli}v798%p*;J5xAUK~lbqG}TR3+Zb+R3^Jn>JxE&t%%o0{PWs&(0wzBrLZ-4>pr z&FT{=)sLj3TuF&a~sy&v_YM{>M&T@8jcoJvcpsB#Am5f$OkVyG?kwU zQBCEf`N?yU?mkbP?U!Py>lCRMHUg%_DiX41F#7*%#khoOacy=ynz(0vFVkUU-97k$~je!ZcG$~CZRw}FNIu9Jlmi|MNN7Wyro(?NZ5DG4ti?76#n?Xa@rQF!iy8vbA{kbWZ@$>FbV~Bv{wzbW#1Kd zoj(k3Y;D-Xj6-B`i# zu3>cMnh~_y6&?D#;vgMxW(Xb8ca9K0ULAaP_7Jr%j9}F01+4kuKxu;~vn@ z^sHS*=1E$gBgJQ-R>PYuPFaQnw{J$FVu<8tT1n^sLc$+>r#WwZobw-OI7eNxbH21$ z-`Uw%+1cw+tZ>6Tl5F3a&NiL)qEmkNpf8{3QA=$b^n2$EId?9BexJFxCpH>4g>(w? zKKrDC@dVtSaSDF9bR&VC7vS}}d$7yBQ1BHbPq3vvomZ#LPAK;QrT+SYW@j79PRp@8 zw_Kv@^$7&eZ)8%m20F#RF3OFTqZf@9fo0n(qN($_IA=+$*p?uG)0bLUndOZ|h5pzy z$w20OREPO~Bv>V$VR%^6u+ru=lVy9B~R4F zJ*r@Db*i{ntA#Y2j=`D*mU!W{E^NAcTX^#-5F%QtAkbFYYd1@5IMb8hKTaKtJ$lI| z+Q%`6-K8)T4}y!ZJ4fo_UsSLTH15)7x7rMCn8yhWplD-J89UFC*PLZ=oR@ zNrjgQw04OcHOStHat`;|oQ`b#;~@DYQY`2kRmsa~n~K53+amWMgXMhQ68 z*O>lEOQ5H$U$WG~3(}sdfd0=iAnox!cqH-g{&_tiv&;kF$Gy|AE?=4cU8PJ%ytoE^ zqlSocm7Kx&O(CeucG9A?o=`q|0dx8h3C%MriRgWsjS7_J_!pYUVTWtbJOA2( zJc_a>`GX>vJ6(?7j(F4F{@&CrU^Eu5Z4g?2nUSYBl(yDV`l;7k_~X;S20C7Zv(LN$ zr$1)Rs`GG**FEw)A{90z8bE`Y0Txxq!KoZ~c2)C{L~=gK)>TEbO~Y(p!5Ix)*58di zdmBa^+>f%a)wheM+_fV8EZm{LtPsq;UxxwOxio5WBk22m5Ed#$QJc?kG}$1TT8!4f z<4bl5{!zL#vgsJJ9a+Q{*NzdC^jm>1cVlC`YVbo~GB|9oq@}M%(e-QJvGe{iLG5m{ zQ;N=1^hSJ^T(sSbtT%^32}4oPy3l$DpA0P!z-(x$J2sZgd${XcuK z#&0Lslf5ROzN{NR`o03Ut@$qT;=IADtPUQ(>P9vCo&b~S>ck1$#ly2cV9v}&oETRL zIoA*1(GRbg`ZHJj@YqVW_-}XF;j%KKe_I=FIYkk(z>Tsvl>+kPc!N+B-XzA355wiB zjOg#72k0C6SxA#=q7loF!H%GNu+G$&u75F&YSU5l;+x}8xuFMa{4tNtesdbkc14hF zYx_DH*QXKB${{4jO&$LDrvZmO;M#?-xoIN(qc?(TBstSbKOE_Q8@r2(rnRx9VO(~1 zfuiI~_Z8nxeghk-C(yLj3REt840RdkOGDPE!pno_S%}LCxE1FML-+~WpGr)S`$^)M zp}wHfZxgKWv|!`wD@Eg>5wd@g_hkWl&$1RN&u;2rSNiChM?{0-}|`!0K8I-F0QV?wLy z8{pBqLeW`u1YIAgMc28PgNBML)mC$*X(4h{3l=!1D%d#heP!-k`gS1JC;H*rXR4$@ zA%$$1Ws5WW2TP1{U21xu7gf8qlfL=h1gm>`kdKcnKxtkmw13+|ZGTFhVrlk$^4Tcr zoYR9&$SjiCbuAFAOV0`M#%aRKMIkUMWf3U-`pE8{R{`|}qu_Sa0tomy62epx;K_A^`oKllxV(i}ua1Hhrpll&F%inK3ZkapWU-4KX#e05bgkt9dVKjWSh7)_ zcK<$@4h>2aem{E2wr!GwODiWp%Z7DOKYb>Z)m|41TF>H;o3*&Ost=ho;}}W09SH?v z(upilBnux{;Za#XaCn~yM@F9`3xo#9FUx_}k3T?nwH!@-(37gC+`)n&gNq-T9u`aL zs=@l^Y;360#&ONZncMy>yc@rXb?)d5|CY@~n|@~eWq}6DhnmyDKf~$dM{8;76@SQ) zbES6c#?nDr*NBlqKkB46pT?Dhf{(--I}&q?{dxbHRR`Y2H7}z1f{zsZ8-=k8hhgbystfHO;Jq5s_!$@2?1Q@R=&d%Kc3f zk}9(Wt$AZHzp;*_SImQ+`(l`_;w^&CWza3*7pw{#3xP)dm^1DYB;LA-Q;Ka__PQwc zFXcR`6ArU1A(2ef4;5Nx*~zx+YKta0De!&YR66n0A($$OVFpy36!&-;pjpjt_RnN7 zX7}Gq?5u{P(MUb0Nj&W|uiO?kI!q)ExBd#v3LBvInFhVFDhFJpZ<4iY(u1LR&O}-N(IXv55NQcV{q%pDdgxBIXe2x5on5=3bW*YgWbDd@HVzgm=|gd>kB2V z%$8Z$-tq)idOl`lUwVMb?Gs{NSqS8HI!T#@dx?8biGQ}{CMladkUDG&1{Zs0;my`@ zSYzIZv#(!geG|O|8wr(cH+>_tM26!v_gXYa=tG8Xs3sr3tsr|9OyE<8E__+$!%W<{ zkdf@p3g_FC7w_fCypCsVL%NT!WdB6ymi!I$lkDl~tXj6VTmzM*{OWS&X|QGcPEg{K z&TIZ2czV+j-KX7@@=wfQOMWxF@RqUo9xYHmc^JL&%81g-EwZ4sKbft`f9#8-{hsmW z1(`l77N41jWbt(moD&+)f`^VHDS6@Ik<&k!>1=x(lo*XSf&=&juP9jh!J89Up7;floU$sH4L^@;zv^Y~qYA zV*GoWtn784qN9Jm2>ioQvdU2j6*Y2*%g=?v%HrRoxbGNHvRg`4eGdX31w|B1pNd`| zTu7IW5|#bhP5nN0!3wQ|bV1!4df@75sy+58F;y4XF3{ltYaXL^`*sq& zeH?g&9fzRM4gB5DAMEG=H>S~TIXrgKr9;^QYV*Q`Dh5VEj`Jn*RoqSGGh%4vv9q-O z#Bj9JRK#!D!|=aDN66NW6lnRG%vK(BNod*yN*)??y+1!v`yB z^lB4t`1w;}i2LD&D++YVAR`)|eIAdeU1ljgdV|9K!_c#r9nRk}oM#N%!`p^vVy{O& zLTcMTHlw3>4i{s zI2g7pNQ0pJ&M+i;JA68`lqNV!wRqT6USFDplb=TNZ=bI43;Ghfdy`cOlkZug{i>Z$ zJrvBBo%;geCR$S7T_Am9??^BGIS29=Cb5LFEaDM00y3t{;0{Qh9Oo*wSLz9ETlSqz zlDw-mfrjAlBN5+plYzhXb7#PD_0@i}8oWhSmBcUcpd^ z(S9ZD+_8koJO5!RcEefs%%xE4=K`(LI&|qK`PkD8c=4GiC;6&2reb^*_4mrbyXPJd z)z|^sHXDP!RRI{METhT>zVuewJNR&-7I*0XAw5DvN#lQ+!jd0(ti2);zic>-s0)1@rr)NNoT@x?v*Zv=xwYV@hZQpamb?2|MjRJRBM z^Ws8b)1g%~+;cN^Rx5{-w{uy2cYnCJrUUw{8iBVnZy^h|X#N>u9*v_`#i`XEhPv^yFg~2KZ3S17O=;o^2k7&QM7O63(}X^ zldG#;AoTNg;&^l`@l4AUHEVXWn|e=K-4#99v*QA6q}R!s@?7fBTtn}U`39qZ)zXGx zYR>;MtLTF}1E`-?JN(&J2sby3WdlFU@sPL2@U_b-n)yDGzBlPXJBADul|5=Nx$NkG zVSH)P?1)$5w%*D3wA%%u5Ua#uGfuOP>ea-h&WBvcJH}4i&tm7=jM!VN&*IVp<}`VU z7R@><(5$Y*u;GU_gdVOUjXQ=p>J(_ej%qtHpwAgru9NO|ZzA)uoQ`tML=%w-_C)-VG)Ed;vmq87YK z`N|YJ0q(rJO}Z7fkno@G%zsNCyzd_>?MBH^XyF5kW+js$(Wk{Mr=6mm;#e#fZc3i> zrDzuTU1pzf2#($DN0X!W(y#3YWcS0g;pQGWI*+fXa@`#0g(GEDafmjoH9H99aV60F zP?7eXn*c?t`_Y+p3t@cs3oyRSoK95TN7iq)AY~_g#YH7@^x3su^hy6qB;aZ{=c+nG zXG0w|=NJ=-0r36|xa4i3Z}-Vy*C8kB;c8At-N}UZtWC7(=}G$2={-biDAT18C!lLl zEWJLgJqNzMnoS zyH4}RCW7|3Hz4O1K!?rnhhtH5si*pE+8ij|bFWOLrKft(pA*yI>Y%wm=IVn_b`V|f zkVl(TlBLj=YxJ1(GRnt*orUjg_iO9;pAfHAovQFT@*5R0Y`!w9VvX;i?h0?eXQ@Y@} zCS4Y+N9E;z5Sn%XG`78BztVccZ>w=|;=)R}yia1F=?tRT?OGJqz7uM``-y4$WK5%a z9==Pc$0^tE;XzL?@b!5Qck1TQ0Rvn?=lD6mM@-0$JV{MF&(hob+-X7oCC<0gS2=6y zEpoPSJRr^K-RX?<8MJa(z9_ja$O7k&P-fJFF5WYb?tWe&aga{aSNpHg9y=oFgu$lN z{^uDwBrc7fT@pzr?2n^ndoI#@jS7%$eNPw<5@*QcnNavC9}I~hD6&IzflUs*FU`xp zsaJ}(_d8&FtScT^Di8KKw~2hJB^rGj3-eVLz&N>hSe!deVs7rB*|GEKnBaDLj2)zF zX0N67+e2ZRvm(73t3>ZC+6NcaPQwVVRPZue4RhzjNqToC=;x~k!v6(#8Y@dbwVIu04>Ic)ZL1L>M>>U2?g zJb$v`F0RsQ5R+0R?#Zj~7=0-j`SWMYX98zKz#WHI`jQ9x6UDiMXF&CwGw8e`9LMGZ zIjJO4~?%HPjYH^3wi4#=b@+QLszP@$_Z}~hAyMG*p zHR((7tKuGh=qfMPPz6YE+uMeKNxEH4zedn83XL9-b6S-e|2>$aq zjlSo%;7+Yh9Br^32UcEYfvJk*pK>l4te7dbKHNkW=NJ%`w|S0MY8&x{Q6c#~Fp_+7 zRU;YO>RHygzpQKNOPn@Q50BA-D4{x0>(G8ywkQepmMz9>FMu?Vc=2DnDmtCe16jaR zqTlbnXuXEuE6tTyK13PDZr>;vq(?bkch<+4L*=YTpHryaHysTk@-bZ&jm<$RtdGTG zrZx1j@a=CLU*GE+26lAkmVFFK>6CcVuyqt!-TxsOA90KXIfoFIsK|%!yMfM!hw+et zcJ}V$K8g496l<-1;Y2ZVL z0`kjcGJ#kEns*(sLt6!XSI>aU6%M#Y>7%fstN|N0^p`a+{0w%R95GqaIo~SZ14Z4| zz{(ecvFpz%(RjYC#1_xSx-0`I*{z7n@5T$C$HoiKUlzhHZ#C)}>?hUU5|izNHQv5s z%$rKLF%Ny2V`Gz>Q;5Ae%vhX=#?wWbubKV&5#i$FyouZqzOKeDoif{b^xC9`DDmvwGt5FHNj{*lgIjW0+{@na(DT znZ-@IFJ$7tUaUA|fy7~cB^JA^AZ|&q@X6>DA-W#?`V@cOv}M0gs!=8!40+8gcS>yK z($_fR;yE;_Pr}gSgRx*xo8Y%o(jbq&%e^kA@;UCDe+u5qKYaMZhRObr*<<6tB!3jy zKpwIIruCRUn=|A40-wkfSlLTYXsIlQU;c-%C;yISq2u^Pbq-n^Da1?HephWJ+}03^ zB<3JY9YXMC+aw$+b|Y_B`H{w1hlOqG1!U#y>#W{qF&ruM#K|3P;)J37h4(jOSVyZF zKasVR4^PnKo7~s(_|r4F*0f{7+7G4t{hlj4`{)+ly81qrX6NGBnLBXVy0>iH<`rbg zuR>YuC4WBOA`;J+M+tsu_rxRqXGri@4eSc8!tIHzIICRZVJzN)mAbWPzTScT%_$)* zvKs6*Z$FNEr+^PC`!H|augtP%Hahn6CBuSMNCj&V9b#*Q{bxrKE7N7|sfk zs9%GJ0rI3+@+H{5H-#V14B+L&xlW@qLh;d7Ej)EtBENVUqN$BN`>WGL68a{Q#Ge-Y z?VlZ@Lzo=?{QQ==y*!A$;Rz6TbBB0sid13+*UVqmh&Tz&v?Xtg4)Yo73z-EuaQ1 z3w!X9P>h4N8u7hBD%`QV0e@nA1E0y{xS?AQuHtf?_3((rv%za|{~<|R|F#B`dq!bx zpA_NZ!35DT@xAO+t~_5fQaTK#k{DqZ77b`-t@nmwYsDUnzWa^+t7#}6yz>(anZ42>)u0$B z4L^++d|QMHwG5JfMuV+hnI!I;Wl3~Zl+c#9vieKm*qOhV+*r+s(KcfqQhE}*Dv!Zq z1A*_mew@Wd1XKBL&#|VvF`ndOXg`;|#I|Mu3;Z+?hgT_z&n6WTPbXD0A8-`A)K{=i zQa4@E!icx5l5&pR=JIE{x7o`X31q#NCH&JIf~gZC(7kwqxMIyU^qczqD&+YiO%$g5a5Y8|D8_Flg81jCYie1ypxx>wNcBA=`@LEj6`H>0Yp+HT%?oz-P zFVW}sKD;8uQ!nv&xidIfR00>{<^0Nx4|u2K3Z^wjW7m?S;s@skL;h5ZK5ty6{gN(BCU~7C2hOb_SlLnWfs!DJ4-CXOmb7Q*jBXBUh z8%>FgRWG7j_JQm=c8`a-@8kas`iL1}6J^VkV{x!%J^pvmoSljpCF8kLDp-x!cj-tR>3Zy7?|cWKT&QgU;i8U+J}hp{zRo5W_DB=%(C zFIc+rsyJfO6ZRu$D^_g2f@jnWaEx0b4)kA!Uz7~+#@!j%|7a2$D)rLNYkVQuxzQvo z_$Zn7cs}MH@DObKv=QTNo7nVeGnw53>r2Wt@lHm8u59OwUSy4ZAcx5+d|H6Fq&FHS zJZ!PT3$1NwcJWk-)}c=Em%vQ z7v$oV_oC#Kf@?`#qC9iW8IA&B7m$SJjXHtt-X7FU@F7=T-Rg#~2bH zrqi)b-$6(5A32nH*0Drsti)fDkx{8qUtrvLRwh@(rmYJTjq+cKVGX@mp5%X2)H)4s z(yl>J9}QUeMxOpl4Toi43|XRqfm3wOBiV=@tA(ogSEAwdE8?%sDl~ZXLYlF9JN%Qf zK4XVQ(ak%O=yS3g=HGY$cFH2GpQ0)C{7#c~QHz=BwnMP-Tekn_=fmMG~HWf$YE1Nv`;Ahc312f_Bwfr(*5Lvf$?XWXb3@;kL^I za_=#~f`!I(3_buI&=XhadVt~Dz2sR5?IJTz>oyJYIPg-rwy`y{5;JkAaIwjl^43+`XT~T3Nu%bN9$3r!bsIhQn~( z-{8+}X`cggXr+a=HF0aK0fN~Z0>hb@%Jj3l^}wrO{-`dF&wFJxkRPXvP;PQ;EiQfn9>-IA_$5BAGqzlGdorMJD zHPivO(W$RAsFT(dvMJ;gyzTly27TQJwF4aCjoK6vFv*zB=`tnNpNGJOY$MSr)D4Eq zx>McljpWC-1{qu~z_n7XdE=iSJR3htc$Sp{zON&NAjeMjam`+MA}8_q43g-S-Ldqj z%Sw9c`B-Wn)|Wbc{RK6lR`lSJr>x{t9&GO%jniiRghlgnV4(INp>9bLMh&D;e6K$( z7RON^`3dxx%V?^W`-D7IQUvjm3%ZTcq-(-^PzN&+dX6rFpv-!y*W-sCdC`(a7hvR_ zQS@oI19Ziv=ir^a1*W=BC+^+f2}|UH*rg-3bo$g0^mEBn`Z=x+F5Rhw&%5TqZM9i+ zK#d%IH~%zQ+OU@<51mN+Oe=LdG|QYmN?V8nYgNebqJ^aYpgpAKM+0$Z*5uf!dHBY( z6u$0Bg>01^nDovT4hH^YaedN?vqz3)GZlKkR@0j#g%y%}23s+6cqomT0@OG}pGF_~ zC=@+AhEcb6!Kz*!^vA&SQnp)16r6B<@&gTaxyxQ#etLJm! zbxp9$uQQIUHdP|OH`+j>lL&?LJt4$@FKIJQp|(GY;pEEzSp>a6D}B2;m+#Scwp}$J z{w?Z)`3)|#<<2lPQB>OIa)AF6a(hFH2|sloaV~?ar4pKER&U&tTlW zZK%Jm0qX|6!d1(+L$frWjW1Y;zu(Bg7vuY+__!DN$!8I3x`jRLUFf*rY5+HFD&*@F z`jt!^x{h~ST!Uu!!f^eGPpFcFXx}>=HBZ0h{r4l+EquuJOwRLrmV0=_>SpX}@s(Jk zE%@YOBooK=$Dnm4EX&dm`%jI-9mj8BTJ#$Jc2QF5nRkR2 z)j097*W7rl#%`Xl=qi>S*^4on@wm{=0y;ek#dSkn*j`;T?lt;9Y`JO9mkcn%z^Nhp z+?fn+oH+>1+6|al$p%^Z5JO(=>cS64AK}%{?8K3^<-!a5S1c>P8pquIjwAc|kqP5! zFnWeIDQX*syCyA0&Al7NwpD|P(ltfCFg$`A^^e9!AAR`hMGX?aGzC4Y`a#9(1$Z!U z8IJedESQ(M^AT3t@QJo6mUJiF&&q~}&6!d%Y+mpMgC;*pEuaeY(a_X-hh(ks!=6gJHF9A2TG5@V!T;oz6VaX6ZF);*ExI?6y8h7@gdozx_82dY$Xe zw#3Eq8+C?!Ktu@cdbx*HYz}1O-z=0^tW$`6-AvKrO&zE$a)arTufboY#*2hpeBk58 zdvs3})8g(D6SqLLc{dbS?C`~|37Pm?u?X)@Pr#*{a{2up6M0#68t>!(RpIJ%|dNh8~0*x#aZa z^LMeabpp|Ln*%X_GO@t-G>*?)C(VwPdBE6{RQ8}6dKKyM1%7q1<_GghWtS1Z930Aw zo$Am$^CMfmZ#4ZBqYdjv>G7~P73AFthEw(};!&P;sNEIDb7oxT=We}aw^qv7m$IFL zMwtnzbZrqw?GI&+)1t+5$#EoxtQJ=eo61MkXYw(wzTE5DMW>Ip!9wPfcX&;^0p-T( z@aEl0-1J3XKDpZ*&Zjx?!4FDF)s%d;SCI6O1^+RFd>gJ|eSr73>r2KB&c#%X->7Mr zj;V5t^z9ffyOq$5XWIENQ_I1-Pi$HC82 z2B%wIv%-Z^Hq+E4ICyFI+ltmg{3w+@GGpxhpjL! zARELGiS_HRML%!VqUBl{5Igx23ohG-vHD-xfy2sd(%TrO=x9&mr9Ct};y^VAd3* z_ge4!Joo+FrHEw3CqmIIBaB&|1H)HVk-G0=@vlb_)b&c`FArPrwY{aetnz5?I_)_= zyl9SR+e+9Ym(9|%!3$NEe<=_xr&Bi8N5DpJl6?!k}3t{$$*|KqS@xgpAHuCTc zQe2XcD}U?ICkn}AL-`0cWMD6P@meU&9Y2y8PT363!Mo%Z!|T|?>e0fBmiKJdkFUhq zvkUXu_8-2TE$~%)9$-ZHbo?>tDjCr&nI9i^9*uvNqNV%^F0xEuCqGrQ%?d^0agPP^ z3li@j>(&NnD<3Po+@%OU**);n1P44g@ffnbn&QaCG3@P${urY6lAoMj$A7+B#NS2r zTf2%13{&HTh@EV-nBUXqQ+ZCUA2Justb8*$_ zzS!Zy$#Cy1I1xd?XLSK{Q>`JD+T-Bl;jg4Rxsd$29L7#MrD1K(e5BDc_*R8LzLs`H zkF^GQ*SK=dk3q8j0S1yU zW)=5Tu;yybhcJI?IG(rb&MqEFU|Jz{cqXhbM{0gyJo7D#x1R{(OD`njzA;{W!?uxp zWc37oIIqe+8-SE>|?HnR*7d8b>Xw_8N$b2%G`X2 z8Sk7pnMbMpW4&kUaOHk;P*zrm*ITPezI`(D`K1oCSDhp=N}GlMw5(vlw;h;~Fq=PL zJeQk9`hmjP7}0lQhP*)9D`y_;kA^=8ew!q5FaloCwrzK4TFYP9sPl$+^c+s=Co`~^ zq(M{tdePi@)(`|*uy30>_CL2(Chx8SYKJZh!Dota($iz;Y(5%Bm7j&KN1m~%GYqK7 z>zO3>?{eniYyiiHn1X!82qAP{Bw0{bN(>Wc5%Y8>Hu_uxPBK&D{$2aN7 zMm7u(=j0^HH@m0d=Q|ULifRez>U#tqw~vu_K~-eS5jPw*NgrR`yNeqvZ{Xh%)5uAM zc!&?_3NB+tLXB((Q@GLq?UP%WuhlqQNmjyY*Gg2Yn2YA`w~@zp$FsA2NAV?_8il!c zUlR@20iYVFKv$3Buy;x%{r2P|YTPlBX31aC$tad>svXSeeRWyBdX9KM?J(Os{TIYU zj$#+29L=)VQv}sDc4)Wmq1@j3r?_R7G71xw_~;8h+<5IwSa40sBYv}jmS4l<^)}~4 zi$5R4SzB62@)-m8Vr^6KwpQn!B^cByS}ba&J!pN~2;WEbfezDla<8igSOgnFro93Lob?2U_d`HG@Gedr zcL;|*xr~*!yTI}xGU&JFD8$yPVNjwY3!Ai0uzq+A#)g*=rv_VevnsWpzao|_8a@{E z)=lL<{!YQZi{6m|T};rWVK6J(zFuy*c?x;3R2vpctBKC^7_usE8f0o{(RowfNj!tD zTxfp+g|}PfKNQ@s{!KMD{}}=A8|B0-z#3-kGKOuhrjui=JBIyx1A5!$<9GjEP`zXj z8~^YW>~kFt2d1y4+RuB?jf?uA%YWWjWA>6H-+wQt6K^Y zF$_DmK^)tAIaCksLv-Av&mM4y?8*lk5Nb^)lzs;{uQphI<2n?kRk0wg{>Xf8K<@!} zK`C03E-`P1>z~Z&+Tlv{a`G$k`s`v>`0*-9I2Q^%mW_nKpGNek%0iGkD5K4sHnOy> zR#yH)CflJmkSO$iOCE=6NFA+uaLp+Nm#wc^@3=KWSN-J>Td)F5NfR-zp!DaW2uk!+ z>Au}j5S4ZwjQ)i{`%M?(k7Yu0(JR>?*#kInGo2Pj+#zv(A+Rfa3f;edBJKG-gq%yZ zrk=O{0~bSU>>bp>&Mxl4cb@U4&Hd$+AJcSbX$*sYVcImmIuwR#%*LTJw$Zv)H>%I$ z5ftBGutqjMI!6~6bhkB-u||q?)O0nv zvm}gq{?wpqS;b^p;ANr5_3>U$#f; z`06O#g~0eGYV)tSjm9wxq-Ic0N<0Lo?V8i%o`Ijdf zskP63YWTSV#_9R6ll!j1%c0MqF+5gCM|<3^R*Cza2D3B%t1z^)3Bt>Lq0xA$)YD|b zvVA((>(E!$+<8l!M6~FS1&LI>MDiOnwF&og<4M@*BSLDD7n{%}3X-g&;je`T?7i;? zC&+99$%jed>kaazujR6!N(1UVBaB+y?oQWL<$~grpX_4)81Znv3R?H>L;AnWfQmiO z#q3CFuW(Zx4nwmjKG(m+wUushc%R98uT0_Feof}7$2KAluu5WGd~{pzWOfzPx!~a9<9W%bH{Me+5hl* zvL5KFD1xzV1Wp@Z%Zdh6FfeUqe>R!m-Vbs3`-BCasoyVo0vg1|h5GpN2}`dng>4f3&j+*uKe)W;o^!If2OB0P7F)>PCV=VWok!?*qE<_nA(;? zW;N4=n`RXA@9!4z@ple#F+qb@{+q$qzZuC5r^k`~A*K~!P7m3|y7O2+aw4M>eem#7 z9s3iVcBcj#uawVG>xOMgm$0lw312(j$L+rF@&5HLeDBjfXk+Eb4yN1Vl^etOVznN8 z$8;mU;y(S1fs(<69W!k7RIR>Xn_usMIirMJ7g3zwsP1u!PM$AmdkcAyqY=7Zd z(Nh+KdY?Dr_>jH!Q7Y@$B5?&ex*ftyM|&2Fi|k$Am%zB|k~8h|OlY3o%%JBEbRU$C zu4~Nsi?Rfs%AEN_7Qt0x5Au%0x!gb3`>?=%@kR$oKE-j0MVyf2Rq zaTn%#og^0DmdkwhDxxX(aP(|ZK>j%#E2@Oi{<%T{om z`ilj;8%(-rnDN-b2eIgb!1N}lqpfWwa}OzpBTM~J`9KsKBBrzQcsq=!ox&>Gjd{Pf zqjB4~7BW83M0k42mS;%&LC^J9FvDU!g|qo!`zgu3I&M4+GA#K&qlx^y9gD#}cS*(D zuWa3vcbMleg{yl>@8BCD6pwtzzd?Fzie@s-*jvh0R}gY|=>oLt2*vWG!z_Gr1MB&Dl!015JLInaXqik2}&_fTT@RS$ODydf5@It6(vqJ_FWr&#UDG&VUW zjGTINhuw~gWO;*>u;=j;Vzc&5c6Z@^RyX_@9DTbTbB>gw`NpZ}ZuSHsvwFZ**Oj=T zVlR3R&cva!GvM#o&A6=RSD_%#n62M0eV0QVF(*C@-S(z_6ft+E9#!oh8 zQW0ytSjlQ_&XdQNg3Twb}i=< z#;(I|laI03{_(I;at+Yl-uV7@1O{{|#dq@8pc6Kp9LW=jvvj}Jwe}|?TK!1e{f+G4 z#lQ9sJ9^VpA5XYBbAYfTSPN>e)=MrFH8d^H6CInhxXBB5UXVSSFHo<>;Z8^Sky2&; zN$(04{k?>an$G-w&O6K=Zp=rX494Q4`*6?nKBQK+AOHPnG_M_=3VI*+kY4N0@;0aK zP}*2RbhTq~&Eyo4U*rzTCv@mx)mk*H%OHe1Ni2 zrJK4R{}C~OFOBa)c4RXE?XLZ|Zd`1zIDHX~TV zw|dmMbO+b^YV+s$f6=z*Anr=~^143L`6S83(UdDP@7oLTMwiuCfMf8WX%YWEJf3T0 zo`9{F`|^W3IDZy3n_rx2jVVR7tmwTiFVqag9SOb3lWcunT5*FfxH7%6dCx>%K5Pd% zgCDr%mO{wVcKA5w4TRd*5OU3*xxPAy-5YH9<9~&?RBZ{433jPW=tr3Up2oNJ4>#}!M+URvPN3bT#@Cs4Hes@o>-?N&9UW=yS@)SG#{-QsQ z_?-@^!_I@}_!D3=jYHn=0g$-ENjifD@~ZqT{Nn30=*WI?(v9Vj?|mAa+r0vtNffW9{eaH9)NcxU!Cd#j^TuVA<% z)?HjJ>~EV0BThXbErZMDQ>I6dte|V`-s5$cHZ~A7I@|I45L52HGnL2Kt5r_u+E-`` zdBJvV3&W9%{fPG(7h+c&1gf**#fh#L(Rt-d)EcdT-Uo(qcU>|`*U|8b*psdzp;q)D7V9%MpLmTPQ{;>7E|@A z2s-i7O&D*q1A2`r70&f^C0QzmVWUM5p1s~f^c!>?eh%J6t9$IE3G%hjE9f6as|=^S z*Ks&_%N1{b@P{qF1EJbNkIqvWfh{}#3MCB`_PMVV9v;ZG3tqJm)mI*&45De($3rw_ zWGR_(-xU_k^Ru7ZPZ5i2z6hswx-;kB8{w7YQ+A%&$u`;aqZbx#rQ>vh1pU@nuyG#7 zzYI6!4~Of+EdMv;d(nI16<7tM{+ZHeVGF6Qrvru+S@GD`e%xpB9AV+uMp;0$AnP$gtmhaW8{um+JK<22Gim4Dyr1)HuP%!2GyQw)Ripfv#(KJ86)xj!f(Os#Yr^C>?k#YD)wfB19&WY z42dr`%k4Z~lgH^FSz@OdewUs-tr{_GaQg(rU0%5IoW!i%70RER6c8E7MUfU^*1ktL zlXir^{3N-zpGePm-UmN^Ho)I?p%}Yn6gK@efvY1{kTp?@uq0+Iif04x=b>iO za92i-PfTVd<3v1jW)o{3-6Wox9K#wHE@4X^6p4Q_zM+f17q1)i8hw&h@ugp#@m9i0 z+%`uGTlS6v@9=DPS^o~oXSw5rtF7YiX>Isj{|vgt72w&R%iLcpl@H%`fsebsjNMw) zj)PZwpkhHDvv_Qa`|MQN{BO~4@RTNm-;IK|^Xnmd&R*LExG|kS3(SZgfakRsf^Q%usn zB??_!F0cnIpM6xw6z4Ps@RC!EKaRM?%XD}1aDPX%4^o8n(^ulEh*bXiT{-tV8pSi~ z%lM+qY_7jj4_ntLm^t*ZFYu z{>S;|EE}#mW<4Jix`XTNapGSI(XAEksN6Y zg(W(XsD5h>Uy*FdjmtG~-TTKF@ns=T3j8EGhHGH8i#I#_?*_I?d#UI1?qQD|f!N$t z1=bYrg5U|qa88UlzjVGo8f!3Em!(Uuom@|>j&0&&b}*7_D{-|OcW|F|12E$G6f7y6 zj%|DM#exBjXmc@-OwiE9jq|lhK*m43>XM1y@9%^Y*^(nP-Fvx(-&Ama^+8Om(1Yh6wLr|&AcHRbB+2*EB{yFsUUW-f z)9w4=*6>J)=Wk8!#@&*3DIVyo|C8B#+m6H}nK?XkA*LQzkuEWYjo-bXY0?H*Ge26K z+W%2S!lM8T>Jv|vyNx3?PwesR)!V{pucK(X@*&!rFJo5SWU>?6+y#w2Yh;Hs?hB>r zZbW^RAC&$6Di(d!B{P+c*xg5l1hb~WtFmx`*&0H<#9Vf<=r3D+L6sf)@4nzyZo%%U znc+YifpyUhh00NLDtgAaKxX?{W@f5}PrcTQy`z^2<{i3tB&#b7j5X(O(j9;4S2KRD zPY;Y4z7Kv5l!5=li!3$PS8Q@UDwe26lZUYvS-9#{R505`eVk{~|K>`W8M}LI+8PrQ zy3Pc4%+bRh!Y{T;SDP%Zu;$AmEAfj`15PdMLWVCrLwdRRF^d-?$a{Z3OlTMZc4^Ug zKHn2#BJZ=bnnr=012pUTRu~s+hYkT(@Nm33&%1qt{e5(ljq>fmNBd{P>!Gj3vn8th zbKoSN*?j;nxR-&SC-24C5fS)Zr3A~5FN5}ToA9UBY38H8M||&>VfWT%E@a4)p!l>q ze0#16p__HET0ar@hje3cjhC72T!{m|Ig_=v?k3z|BWx+)O?v$JNAZoJ5%Jue1~&rs!vWog;N9;Ji!d%gqsv+tb;AYk{5wFN zt?ti%Xq)q|ZND+rq>R~KJHd?pJi~KJ_fX0DGmIS|z#56W6qznL@)j-!-+)_UlWM12 z$K3%h2PzYlb^T$1(jTF}(P*-8`hL_{tB)>cgbFt6H|(u{31)hu=sb^UyjC2{b(NaX z!EP%Fn9~VfY6IwkmET}?zqinF(Tct;RV4FQ_lMWXGho03X9$140t$s5bVBziw!P*J z>le}ujt7mVwyUGzAzA9 zkS)I46E$-*p~1k1Wsufp>joWzE5gG%n>-K|Hfh`<2iy)0BP7~u?YZ&fs#2!SN zlC1DXut;(S)saqcc4EDFK~`Z&GySef1pcuEE=%cC=W9@DR-$EZ&4 zAaTdh1UOL@0rmS+NN8cO^qXH4Z}icoGirgjZ?=SvTpKn!Ldrc$eO9FpCx9OB2LX+J znW$w$GJ9`lYb@`RCL>Fh(a1@*ECl|WV~>)%MLx9ty7*dsCjIkcHSJMvOy{{J3nB5E zFzf6y82TvPesy&T(NI!>6|y1+ChT$O<~S(0tWUSf>((* zSzKSt(wr`^+W-2&;gOzj#Uda2YQ7Pa|0T=robVxE9`6LB3)N)8-61S*>2o&1xrOas zA~ATPI`Q6NXFg0tffu_>go1*G}T{cb#%J4CV~*IZor zDUS~~`OC}g-TAGe;r#x_b2!ANg@^nc$GvL3`DUk!=(g=X+^2fK`oFuua&I(eJF96#-zM%JvQ7-G}RV)I|~(dRDVvh6vzN7(?v zyxxjhN|O7*-Gy`OyX?gWW9aeUIOz@$1~=zdfby0hsF?W}Z_9h|6>YJ+U1H|uoK3*U zM<-yxVyS0hRsvZ=DOnMy2;13B@w}lu*B$f+FRypvvmgJ*3hniI^~Zy}c&fT|Huz7z z%vgZxZXl*5r?T{v!93_dDW*)&#gBiUgH57aWysInm1-LUEBSXHIC3zPO(_|~Ht-GH zAmk$s?y0~7U)Ar2dD ziq@e+VAJk-e9V*_-WGEU7Cv|+4o&I762BQS<*P}kde{hE?muPy`C}6PW;p5Lv{(H8 zo2{N{J(p#IL?u8n+Lou z@*YO|IP#>!gvZOmNz}ET^h@u0K`ODczDH_gCxvr@;`RV$$WyR=MH{MGO0GP!b-227 z4Ier=l6ciM<1DQ*EUH}~_6gO*OHT~2@32$g*5rb9P{YEjR)Wa}sTWnYmOnjR$&ZYv z;t{9+;EFmM{#|-@zj!u>J&9h;*J>VTRfT87=$m2ukH%-d_B!%@-_tnkci?WfebH*P zJ^uOG$(QLHSK5tQ&ugFEz<4@Xl%LGQcJd!ntXJnpeM~V%{S)(!xWof16EJmt5kB8v zhmG^j5%<;}JZyV5-v9>u(?CP6E;;XOP4w{8ijmlP_!xHYVgeU^ufrZaW$;$Ii)%A_ zG1Gj1W>(@4eLri`-?wJco0?|KTg_1Dn5_bzQ|4h5l#>xvDe#|tEZLAhl(o*4{Faxe za=)I>@c3GDoVa!*szs%U-Cyn|<);m>(U_7Ylh#wyQCl!YJprxoIxci51Ovy5QU);t zwp5S8nJ3P%QEU6tZ`)k)G?6&4KR3X?rU`uRu>#E39ZCxh_|j+NH%g7>FgpFxe*D^S z2=~A5g*$wWgoV)okg>5DUj5iX*F88*hqvU>p1#+a*WMH?dD$Ino#zWWhPmw1l)B}vQB-bu9pzg6?G+uUunhsKO=-_emf>$h*xGth)?dRx%8{Y-*){`(w zu8C9ilStrTH!|{|fX3k_l~>FC`G?%6P?2U%jSuyuBMW=c^nYDx2mcBUyX3Ivx#Yes zna8`_+`)U+&lk7veodI{G5ONB3di?)iJy&*LFtu1=5@P6F!GR`o(qnWnBM+y@_`rm zD|wYBtbG6j=a+)|k{e8?=TP~RwC;Fon=)Q;nuwpK{E^E8=a5-Dr$I!{5wb6QAe7yz zkyy!Y^w0Vvnp<{(@{c#@{L2~e;6*4k@2~CfCvm>~z_5PsVq_S+8Q+%%&btTil}zXe ziJPKlei_c>WJA^DHR9=?@gVyiMC(H{$lhNX^v3nR)LE$tOd`h8`Y0P3P$~H|Emp!s zmk^wq_Z@rn9L(KE9>(nEl{8_cKE2?XM~f`BL+9$#VCz@S@~%H%|LHNj!w(BhALg;k zSBz2r`!n`yiwRM)1NQmTATVhupoi67(}fT2(scK3Xm#Nq>vv2WwKkrItF2Wq`|f#G z#X4BQ5EEX0#F!g;axjoi)DfFZX_~aZ$Q11G)tA0-kLhsRN1i`F^dD4hwA7~NB|TCn4RE;M*Phx^A~lA|k?Fs#QYVl-|ObS^p#%_`D; z!+t2UFz?0sn?4pC`ky1F7p2-uk{#L(Y%db}5UDzf^%QrP+4(nUYt0 z*GTNRyp8R0ekxy6VS_4`&xF*y8u-@q6t(ocNn>?mY2v?HwDbNzGg_XFhie|QVb|{=?D|bLe$v$+>z14G@l#L1-*H~Vtow2-G0MPh+ZhmrUj(xcaV?Cb0gD@l7 zE{ld?FCXJUe^uC!7XjVIU8OpUwH;c^I_VuR4Ts12dJa)*-_eL;wOrAr6b5DO0v||{ zUvAz-l2yioxbr`;d*73y%cohaMXt*WT=Q_&f-LZS`jHgvsbGuR7?$mLj(RWb`L)HC z{7`%j$TrB>?`uQg&%qTW{PsSk`K3C@jr&W#|T2Y%0Se3eG&)U z(%_5d#^GqYZg~983y5h6gA<|}Mn*Q`l#fc>%CrHu=+vdC5xSdoi9?&8>r3VB(TA#*601Gitsk)p1VM0TYR+XII2vh{^*l0qByk8s3e55mbY zVS|)!I{?!w_sC`rwibIfHj&eLrD(9#3Gw+t-1ORz_o~t3jn*dItdHbi?b(A@`lw=s z&Ix|zQ5D}Yqm7418Hp>h1on8m3_xOQ6}_E}7xj)~ZZD~;dBGhs{uQt*txxb&y)K6{ zefeUe7E&~RK6x|g08BGGO@|(i|RDUre<6U(@VVn=# zcoxcc6(#ax!3mtC*JDPL0_J$u;D{q}Z26Z5uuSnJ)Ea#wC0lZZm+EUU$Mnk87< zR0xOaxGZD6)M6ZEjXF(x&|kS#h@ZWoz)o*7mZ+#ofR!pMm23S{A?HSAM+ ztLUm{#l5_uF#Fh3`+v&Y+1!P(5UMi(*H7t&J9DE*%CA(iS3X4ehTVx{N-107)_@9~ zL&^SW9jx@gN#?%zJ(lg2`qsZ&eZ7;CqKv!O~R_X zF_@HjNZgXv7mvG?lP8)FNx}Tx-2Y-EFW*$mk4T;#73~CUOnJqKO9ab$_)RdneL#%g zJ`Jr;1&h`D1t{D)UHo_1SNyqpEV-fofMo=$;J6vy%(*@VzoehUp}#*dAKPxQB=VDh zhWE&@A}4X%tcEi;+ww%MYGp`PxPrmu&sp`EBRFxm2HGy%$R2qy9z9HAkHrl`OP|T8 zc(5C02WY|Oq)a?!UXIx%M-isP;J`EgP(IyO*cqUNWy{LZ&-)~j(aK};yJh5`JF!HR4IS=i($tm{ZA7xmDW`SzXychC?n4Kbuk z%7#;`j&EdUUJksfT}7N%9Uy8KpNKT>3fpSFlhhqD=O@;^!TH%YQMt8{HC&pAGc5yz z7cy5HWSvIG2%T>Oqpn8?f8d^cqD2oJ-_;Ek z$Iq86PEXPO;R~EHb1TXhtK#^1(HJ&RfemfWXDQDs*vHHBVP&H)gjP0$Wsuz?_?`~#)v@~?_}D&hhYuIW2Q%cJgno5EAyY= zX8mzsEoYH79Ns9bnfJ#*4i0FB{rNX zr`%^b&63Zy=^MV3=3qzu4aVj6QN(BFN!Gkb@^icWk%^ptrD~1tP^QIA-QRi7x2qV!}&Pe0{K2#^32C7!t?+gIDhjR z)Q#1~-htBJerLtKpXu;02NyQErzYJnTAya`jTPVBnuCgGAF+EL@wlpKFO$z+1;bC~ zz}s(Y4$dR~OQC?Qi#dR4N`&AI7+w;iNLAyZNte!-@osXBf68E0az^)tAu;_R> z2|p>r%{``}`F(q6(0<3N)*I1!m;d1BjGjEtzW{9t&fuc^?u55brsI#A(x|vB{83kh z#j2g?=$a05W538$pME8MXC;xPU+%E@Hgz`w9L;l4yO&xbXRNrhW+7ohP)HJQ=3h;({&!KqT#{l~5Y5bqKOW(Tg~ zJ^wNie%u1}M_pzfTdt$C&1w9x;t(n7bqt~hxWMtQq0GtO59S|Hgb`2jSjCGy;O$!l zS>uvPp9K-HLCP2nidhWek{-|~<-oplID>nc!1QNchZDn{sd5hmOqvseZ!DDgrRy^x zN;@2;Rk*;2QUOXwXOP_OF~qj|1+jA5LX_sFknHbSWTnKx>E<0z29LTcZ|J{^h^Ad5@p1(4kbQ-6Lqq8~j_p5{{S5oaZDs&g~4?e~LN@}!hY!A?Ft?}5IO z`bxQI-(jYQL2^}ItziHj12yhz;pks+^?W)5%D*|Pf4!jNx6Wx7tqb~RSF z<#E%yFyl)%g_wk{Jb0fEcePo|R`uG!Qm>DI!1I0N+s9PdUsyHL^B9}OV8>P(W+gU2yJ7q3EZ`O~e6h+CB9tf;vPLo*Lg=pz}NRx#Y zZh+|EPb9u=GVYi!@wQVE;KMv!vTMDJjcc!DhE39L8IGBHBXtl$1&nKOaSF6?X1$o0Hg^qIdDc(s90vj zG{d?P#mzZHZQglQdNx8h?^p&~a_5t2Wj%N~&%)ptI`CPHA_WS2S+_CT7&h&Nkab^I zkSA>y&MmtQo?-Cz5jm17ABG{3LffWvJ-^jJn zYr>lUCg9-E-gMr{*RaXI6tvpIBp14jwdt#eQDELpB3)RhAV4XT=!KX$O%i3<}aL59!JFLWWHx1!shbJ6X zHG@N;E)b*G9lBl4Cx;q>(D(0Yocpz(Xgn|l&h}VN*9;jW_2PDuT18iA>X*ts%}*q| z=MBQaomsF#-I7jTzMmDT?k7>_G{Gaa3mTJ3`I440Ht>EVv}!*iQ!M(yj@WV-{5275 zm;8ksqbPQxZog=7Y9vcktpSr=o`BC(fqo4p2GZH)&IC6+m9&&aZCe8Ey^ERsjtF!r z>`PZ~aHEUvsDo9b5xulTiI$yNLVLt1LzhBjs+zZm&enCO%dKPSH|0E<^tLP2P(KB) zLRuhxV5NHTSigKCo!}vy#X%t8BV7-?W1uo zZb0{;=5)T95uIJp04`14X_nF!np6Bu>YyhP>#$DZw#is@?QIIfp3H^T3ID-_eMPWm zq!r~ES1J_u?1Kx-wP>VU8Dxl|b*Z=L0kycH6BZcIXcO=|)4%+^Z~kcpeP; z&j-#wct<*S9D(NKJTP%|f(@0uA*t_usFPfO!?qgGV_(+7yad5 zdAYnjwTb0>uZQi2zr(EX4WRn)1OzU>1alMyQte6ip!uxituM5uQ3t?bt-Y1Qiht`I z=2-WGN6V)`Y=AbkuD=Gqf>smr?g^mt;yNtKY6f)Q14lnCqhX3M^l8trRGd`-k%O)1 zAN~U#7|OvV=p4x?Z;_Y#e<1EdEa5^$EUcNROolYh1I6q(n0lofy;EUBqXb{t*>fNX zn{k>AED3tUH`Su0Dl_Tig8ne+jpRE%b04;9M~XRirntIjFAOX%mZdGyhHtGwpfIYFO&J+U zjMu8*wqC{5^}~Fc^7{*Px~!qf-s`E$nK=5pJd0j)vY-V^qG;du9^i3bi&a1MX7L7x z=xP3G@%!LTzeme!u49Gb+eA{TZ--Z9Girr$g21V?bOQ&7P?ju$ah7 z*{;S6n7Oi;cuqbgsMb%z*n+WQ@OvA?f*NvV*e-Uk`2*W=dp@ylnvJ6U5}dlJzc?|$ z9By`uVL6*8QaSGj=0E!4o78F&`NSCGln$^%x6N?oyGpiX%zFGXJqvU@3P{4j`7nO? zR626FCiR|0V07FGaoLYdan1g6ai&yjNlwtgPYr48>(gKCt>r{kX}KEzJS!BI-%EjX zwOlC8za+dmbB26eehVMQ>neKhB zfcn;we_`w`R@WyLLJz(abmW=Px^@7akx~eSH#v-2V2qUo=2*V7KX|lg!l&dpbaYJ< zxGWw_gABW3!`oc+7!i+W(wxDcCxT+hXL+AN1L+B&5)$WqA(M}cf)T;@$q&so((|At z%0BeMho?5OydHt1FzPY$vmGVPrh5~Xxrvw#s$-7Zrcq<9Av9#fSiWl2dRi3IA8fy@ z!?&aABtAqUT#q1ZREY)Pe?8z{TO@kz_JzZ)5yGA5B*tb&pyw(z!T(T}lrd5U-tm?d zUG9g2zo)V9pG{z;pDJO!{h-Aqj`Zv74iC~rxa09o81XmSFoM^y<4nBWGdIvFzS*ab&~|QQxRPulhZU2Mr9u zadCIBVU04sJ!~E-9KVYGNnJ6dc^}54?_wU#$K3ad5MAR+>?Q zc2AV$KG)RPG;6FQgGC#z$p zn>waT3^aA~tE~6DUBdIM27Y5?x5{zJ7L{pUSA@u|FG=x-CH&{fZfJLOD1Y`Xhd0|K z@LpG{c*3mre93k#(yhy4^qrcC3#1&|si3Xm^0IC6d%-8UUN^bKq_~RRgAcIYJC*Uf zV^3aW;*EV$YRPd+1CsoADR1ePA0I*mTP!?8rSA@p-k3e4N=8#&4a6i^K}7 zESL70L!DTO#BGrG`^r{nt>Ns8UZv^jPkh$Z1}-~kEA<>1cJ3I$pL}qX5B%YVoeG`! z%&ZMlG=7Qo_9O7rmU2dh^}+|fMJ)69Of=log$+?k#9n@yTz_>U^P7K^9N7L7BYIen zA2Fe<&!ev}vh@@US#VoWNwh;G+l5&hGtuW+Z@zobK`cEq7JuH_jU!!up}DySE*&?C zWWRb&_D(e?o+%^w$7`LKbHx@zCuHNP7F8S@K8BA77V)V2F|;jRkEwY#@Zq14{H#p{ zuX(3Q&fmGqK3%F}8#UB9%pHa^Ear-hs!{MS%s_rzbrw=fFKoZNp7ER_rk z=MGThr|GYXARUQ4Kdc_}JR8{2qZ3I=lT5&p^=zK2A7*{5gT>Zkq1{Otbjl5Ja*^ao zA;u7Gat7Xg>`4ntS?-2eq>xT{N-wQEq=lmSzZ^@D3Y4wCz&Cb0Ks43w|^BxEmJC)WA&LnYs( z_~==h5TRQxD8(EUZ>9o)v6^TXvjLX5Y$Z=l9VZ>Znvfk_NhFFnW=^(+j=Teq8&ymS zW2KJK>}7K8<7#LV_!Shl?Izte9mn_PcLZki-v0e^Gq!nN0M^{UNB$jeWGWgT(D>_6 zw4I&9Mvkhc9zW+$>q&#@_J)nJ6ZWUk!tntMm0zkjaa+pj5A-89PHEBKAE(nB7O!AU ztHih&qzT7!eF&+{5dVu)B)2O^$i<$;=o|Ql`QOl{Z8yfkE2&4mF(#S$ovD?1+4d7+ zEPt~GtwJ23WCfdydcYSgMXcMahV8H7gvH=Sn$$uu>E#}|*7goT_ndR=DTIvHXnOTFsb_3Y07C^{2&s@^UPOQw<` zk||?RArv|LU7IK(71AIYRA~6oOiH7o5-E|Qq7)@flwrT?Rf>ve5|uQ_l!`=Y(%{?Q zAK-GGa~;lJYdz0>7d95PYPNuTEx_H9AJlx|cgXSG01xKRqesu?(A?i~xH%*Or@ycQ zhtmo~PkBs{r9}b>m}x@Vzs}$;#8l%LQPnkL?KN7dqQ;x5J3)f>E+VC@gEQjGxtnD+ zf|=8CIPf1FPs$B|GR;Sq39AhQfYrX*ihV1(9(h z__n!r{Neqh_>^IaU}JoR{*FIJ*Bq?i>?(J2fA*wt>*cP{q?prC#tZG zr<;hz=h+xPXEjE=3*+;{5!N|Y5$^qMVb~W17MtSBu9n6M4!7QLH;$hbXHO+)BLhf; z#$xbOe?nWHr_pa7QDl+B3wmo(CYr?WM(f&Pd}#Uz-rM^KlsX!)7rhi!nk-0DqWFJf z^*oBp4hiG8ZRbZU>Cc^6>0K=SLvg`UJ<@GC8hbYhwCb!LNDsUZQquV_OXT)FTm6J& zpIin0y|=l+!}f9W8%_%O+j1~uln)xnbA+|d7S`=Ji=G$8qsof!WM1ez80lb058l(` zgN;6L>wLH2Zw-02ai|;nw{kc;_+T1;MZOq5b!lQ;i3Z(v*i?^9WPmT%( z!)Bt^A7g$>{W4O#F_8ZGo&|p-mf)yJ$j_u0aEm`emiWG>_BddlDuHdi-*&wd6ftv}q2 zQG@Wjn*u8B;Yjq}a*1wv zYf1*>bd!GS%lN3-tx#*E&U&Ui#|?cmnOlJ;bPfon-Tm_<^=Bi|vgQH4-PnWHKdXfU zKL^rpfn}1Us6n72)kMpR!ohvnWVF?H1ec;tVz74@@2H~B`&gRur<$ASz!{G??La2j z=H91!9zK9sDjC%O_Go_Obvqci?q>_8yEEe&|$59++Q+@7FsabXh;isZlb}ug8J^CGOHTjHc+VUxZ66V+EfV zK%46PXu8`%oNrl+tsfGxy>bkwlHKrT`Y`@_u=s3p!HCX$)5P5*`?+(vUxl04PVYR> zWmB}fC2c`F@lI|u*}3Wz3iI_*KOhT|0^SP^dPcB##?d0@US&QJUdiB}goMU)Q{CU^)R?KG4 zaHjL>1zsuPVM((R?dp3Xjwwg5c91$hbX_d@KHP!6H@}9?%9BxX&uokyw*@MDEWyg` zB~H7y3mcjXuz6uTS2y_vDUL7`eyc>|s8?seJ?%KW`ne3&iEe=v`%lsh6SUDRHj^4J zoe|5{5hdat{r@spr4Ihn+zPWBM&jFFBk;@i zd)PPkz1=L8-Bj4`!m<%(xG_7k=~u-t)L(la)e2PD;`vkvnPi9;H=jmh zD=ji=LL8B-IRgsS6b-zjF@V@rPo@`nUutveDlA@*^V1Z9o(aF)%|^FklsuG zXiXtWdg}CcrJuMzT1-YQQKvbs#$;=sB{_d(JsmdY34Misuu785rfh0qGn39T>6EkV z->NTc&00&gBBGN$(f+~|l8WduuM41FyN`s}AX&cpK3Q^VFZ}B?z}^ZKc67pMHsb3l z$;>e|Ldzk2wAmMd9kpU^@TNS@PIxO3E;f-?m6W2yP-XgDCRfNieU_9@Z|8Q`Zzh*^ zuOkX#Ci9tp9~C51AV%#T`7Q3P(u+2c4H>t%X(A)3B2j+RKRh=?!rjzyBDXPHdAzVHF9B&+*&gP2#3m3D=EN9|2Hg;DuA@Mp?^GN_a+o*~Wm0N{?r5i|w z_`19+m$ylU$gMRVgx9gwvhAY z&5+oUMJ5bOl$_Vp!iryK*r&ystSa!W=$NbJl1GUQlNFE28UG^T+7fv@7-)=(j*TJ1 zY~r|~0c)^PX)>%TO(whdl#)^V_MvqDwYWry$E${aImLB`*z{;Agl66aqpTxTe@=z) z%PI|O)~L~*R2O!-pEW&~dJUvZMuXC%nRq_!DY@iajaCCR(D6qy?fWyT==QfvmcMHR z-{7myUN7&)$5Fjhwa6Ctv^rvbj+!vi(;WwyPlZj5K^XL>QuyFp087U_1($-M{GZ=S zR98*RHrHRJ6^?yHbIV*%LDxf2+&NJ6tL`G(9+<%GD z`Op;HdOQskBbUQXYK715bF8dN0p&an(K(m4bFt$0cWuxZ9GvHlb^F?i%q|vC`Sv_m z6R;Ky9S?C6vvcU8(A|()Yz9vog23i>mk?a0iPB?N6W$|_s;j=IQS+p*zRDhaugBm9 zh(Z6lB`Djp4)ewPEZ=V^^ZF@*dIo7iH2;OJ)2_uo`Twvyd0DY^seN&wuYU0*lQlT{ zn?JfQN`=GV&hCx1XLIC}&{6#;(|F^~5{(VnUa$YKXzLGr{*&T{IWf3o`BO+VcI3@k zf1&2T(dD~8-1-F!DdPwRhTBxhBu;y`0TiAx}Iq#{_ zY#;115-`cUKV}Chuv3<@FsyAcSD$njcV^YXRKFK=FFQ=+Ce34hH&3w@h8vmLYb#a~ zpwEI1xv>&qIQyx339YzrY=0jDEem{v0*#l%FyaxJ?syIAbk>l=`*fMNpC5Y8)n$=t z{-{$ngw64YWEQ@v$jfd*S$jqHrRh0Ue2|4#HrT_xF&E*l=5I_q{Tzo5S7Pqr-*Fzj zP2CEWm`j~4lUsE{Q1Bjt3(t3v!rp3n%3KX!v@YPLUbQ7Q_7$kzr^o8{|HA!`Pvi0r zd(i&YBvNk^#uWQbG19b$-41HTf#Uqg(qy4fMFUpYAHoA>BWQ0_C>wNdI6G>j&oXzu z!Qa(&IBr`69(rxYf^H09PYh0D+8#T$_GAbS7k9k}BFnJni4}XiD-ECjd5>=$<4`q1 z2_8K6Bvp4CLDt<8<|=<^J`4LA%ymTtC;DEU5RPi#08ZB7Q2f1}*`Gfa$1t+)MPXyR!XTt3>u> z4>kUjTi^rRNzH9Z;KEdp`l0cWa_Z%BS!7)_p~ zyW*Fi_u|}SCbDr_>=s53g(!V*lJsyFn*8*@)&o^^l=%hBzt5qt;scHU{)FD%*$2N} zPr;VKm$7|_G&^BAo{c+l6~;EMC3A>4a*Pl(Wh6kl#?Xcq z3C_CH4~OUe7WS;v!)ZI0VqWepT>h&cT1=C|5&Ny+_swGHXIn@fy^5yeP15Mv%R>dt z#&nAN2$ly0(xIYnA2FsRD-*;Fb`@&x8bv$kq|M_ z47jBe>DY_5sQgCktak0Uvr3;YygsFfoiX=t$oW<1)uKk{yxb@p{C$nQIR1zD^!CGE z>lrxnvoUUYcY!PIF(tJ>%j~vSq+w~9I+HmSiM>_-gfCmhfMxV_c&OWtUa`sLf~!80 z3C3#$lUp^Mq7iasXX@#~!jJT@!y#(aExM?-WMFp60=)A|ykU$jpbpPOHn2Mt?zx`f ziUUd@CqkcoKM^h|lXj*T&K*JB3V)m%Q%Ro{$%wt1R2U{^o#&oD1CJd-;Doap>2(Ok z(}whPj1Z3D1e!;(G=7zHQKwF$4PZcC;IPlR@HMqXTV8Q-1= zf@6UQMi*a@??3hNfKM*$SXo6}W9>+NrxE6EY+~nCoYCG+R`@Jag54?|)H!YsPI`3* zuPlBp(T!Qm=j;&YSFs;$N;dFYwx#mpSUhj5aF;oMe#T5{kF$=ce~ClQWjd+u2R;33 zAspzdhRc%^LH3j{KTWldEESh_11{?FXJ(|(M_m)yLlrd|@tc!qEpep#C-lNG$7tdH z>_P0@$3D9+4<3-fcZm>qA`yaZPGPP4Pn@$-70x~%&Tkx02gZYgu1{Xu7w0A%FdvMcHT}UOZaBtmFT|C?Ggx`86|`jD!8cPSzE^KFzpCGI-d6oTc>SUX z!Z+O^(&16Gd6ERD#55tiSHvGfG+5)Iqo}rHFj%f~UvC8`*SEC7crccZk!B;JBxoeE3+@hV!%aJCh~w15g(Z{ZiPy5tU=$ro9yw`1 z_qe4Xx7LNP-mC?yj|a1$s6kxbtgE2&$O0Sor=oaK!&{pbY0SfL>cCCLHfd*ecl9lN z;b8~U{yc}kMdM+(>MdHtXOnjpm0}wZMpkRm66C zCii`C2CTSV2$P5Gfu0?STzPmASD|Wv0~V?9KmAML{?T$;6w*iK&4$y?jd#iOjboug zQIqWdc%N3h4=05%PxSZvvn#Rbryd<0 z2l9QbeK2|91Ne7d4Z^ISkSm6(;o#~-b}e%ybCuo8D)rsiP>&XL(rTqjRdtBgr>M)E z(e#e~1$=3rA$ETUu+6>pH07cR{X2IYp1+iYioa`-tgNItPP%w$Nf@<$dR5q9vl%C3 zm10AlA>JQt#uo26jhC#pp~axx_@&MktH&H5(RI66S-dnedeVV=3{SGLnPsS=K}kGoVM}(pAh2x#ELeL5eKm z^%nd$=@4C0f0t{Vhj`W77*lEzxR|M_+%CNW^!*o&0fTR1`PpD@#^=%Gg~u+c`rsO}jh_E7Tw_1YFt$U7a~AAPS15bsWye&n{-jGwl-NAi1Z-Jp3d55P zajyJERBMc7hv)er5ub@tTQqQ6?@p%MSc+>cvZ#-iGF!HhVrJttx?;W}!%5}Xw(lJ- z`k2f-r`}}C&Q>yuowjVYczb-|ejXj17O_D4R~YqZ4yH}`gvPKIpI1*OBb4Q6+y2&~ z%Z<|9mIopux9BO?WcLgmWk%q$_|rIg_8U|yS|jqSKA_|GDb%b^Wbve?;6*XdIqZp; z3z^*yoy%TO1@Z5HM(`dCJvofl%O;5~mI}e<$Sc9Zw_NZmXcnsC573Yy2%}Y#+;It=+Er*TwF z4%xa;o;kG2HILQyh&43HLA5e{HyNF_5HVx{!MEzqI;&jiMIO|&p&OVgP^>@!K+TX50ajvmo z8m7k`_B6fGKK=rJA(Z5uW^UbQ+xrhjs z^yh`pe7}gKE-&N~lW$O;Z!6HbZ!S8`evZ?&iVOWYjN+Rjd2TJ7yVhIj#Z9N z2@g&<_?imdYcrx>%DZUGl6%zPSAU#uc$Dtcp%UpInxt;kO=_G}i2DEira=i$gx2{E zIQ?}1?p~f)G(XZ2Z)EH&Dp{R?1!CTPfl??&R@bAN?`EiOoQtg<3T!K#!Mp~BvpH`2 z(fIrbsyH(aY zj&^GY!L3H|_ipvXCXYTk*2;ng*S-*M6RO}zEAZUvV7%-Uhq>{9O4av?vq32&FN}nQ zg)?zs+dqi6vZ7C~0{Bw1LH97a&;-&~R~B7jBc#WBx- z_3Y)kIp`LZL$Z%ehL_a?aOnmg(b*7BP4cdA(^ss)T8D#Nquf9;wdNwHH#r(#ZH#3% zhhAnH!)@8P&zYRIQXF;v)*p7Y7m!E0oMBE<5ZS67i*p7yf#$9P*!HuX-lhBL<;U^d zd%1RquBsOHri^CEDJ9hXt^q{J1mMNC4laM*NAA914_O|25UP!mp}aH`4^(u~w7Zl# zpL@nFYn6w4^L^lN;doFsmjdwm3hf?)`EgCr{Bc~wuW$9?H7h2dMvpUgNbM#GrDp7E zt`rLo{fG%Uy2Mk#g-@xP%r{>W=bwd@Q0yIt8C5IM`L!doL?%l7s}9q?V{#~Gp+ole zspAdb{&Y)T0@YMGPnw=vz?^_7Ec(V65*(_}y}0R!=Qbokl*=4`z=ZSA)R6(JttUZv zNC#J&>CYzZT+QrD2nUx&><49Ei#7L)dS(86?D29;`q7C4~VE;6CUy zZT{seW=FnpLu|?geeW&YzkZ?Q!;D1Co1%ldT5|B}Qm(j<6`kQZpmWNZS82yUr8*)!IRs#`rE7F&b8sJaf1@_vrbdFRR`I+XZoz=?HOziGD1Or zKb$kqCy!<&qn++OIymDo$yp!Hn&&F8{d;<_h^D~-#c(?|Xg` z3m3H4bKO3%v^M+`nK>pIPDIXw&X{v_|G)`+Mi$|J1)ak0d!ukq?ge!8wPOVX71*Vi zM7sCR8EiDu$DeI>xE21v!<1&QyCDnLVt`*C8_wVQuukN6h#i(wnoPCOgmo2Ng@|`Q z$fBY0u>V6ZnS@#xV?0_IGE9LV`)3n0Mn};Fm(Ag1cq(XCbwg*`W{6bP;%8{Bh0y7? zyuXoxZG#W@#71Jm9i({IQ)J9!hDaCB6$Gs_u%g#_);|JoS|1)4^X5 z>^WKVYDe=+%F=0KIuh44(zs%JzA)pukwnjJACVsui$-$qxKTIXlVMNB(gHPQ$u;pt zrT;O6T)Zdtvv*X2-!d=si74Z&{53@XWe~X`r3%r;)8WI-={QMaHvW`bg4Qq7AkgIk zM1DyjPtFyY;ql3D z$hg}v1fD#AX=!%w_GC7UezF(-8_BWb(b<^vd^(zh_~Fv9nJB-&9d9g7hUKf1$kAUX zfoyF@dwF+s-K>t2EBo=63+6%Bdo^B4M~NS`*@E9+Ig6Kzn+Qb*UT~%TT)^)^3_aI$ z0Px`qkb3wUa(<5l^?v7RZTltRKPQpZ<3EU>OXs1&sy32E#msVz48Ga40ER~)RF`&O z_3i$cUY0`Kzjni*={8jIY8R5ip-e5%SA6Ze$+gOHY;N=n999-bE;i~jrTSkO?(>nB zj*$rRzK-zd<3Xr-7)Jhy{qOtg@mxoBDwJl6?3%$Q%&AhAy?=ZRlET{{f2a&@eA)zd zx6Zws}Dn9py5^u6Z6D=RrP@JlT$zQ|ht7?77`mi1LJ9X3N0n#|@LlDk0zhn1%?=@TJ zanX|ZO_yPVp*>0Qvn0Vs_hGf^NqWcfE_rt07#$JWCWIX`rn7|IBqBG7PFOXAo(!9g z=}&g!TCK5o`Pd1Pm0ONKOy($E3c-+8JGp zrDco8;U_O;zTwX_CC z#4VLIU_#9$)-}(Qy-G?#`Sme)`=uNHiy4nCL#M*5v&V?*7Hyo;`H6db{R5g_9w)M* zvM@6-jp{G&gyda*LUBqa?uwJ*OQJr}%&!OWQp#R>@?fjzGZ@95rcNhEE}nz5=lfvJ zsclsAs1}*%WD9D23!$Rhg4lgO$@xFiW>s?5!mP?1_%>7uc8xm*LFYoCtU;Nd*yjd0 z=BBW_W*+m(O2W{FdKxfSkA0r7SIi17KttCIOz@h|2C5BbJ{}+Otm}Kk7ln9bzE#oo zdtKzr;ZUgDR}D>m5`NM7UJT0&AZg9N$iJs$aQ47<{L<3^izK2~VO2i$Qg}p5G=I~P zGm7cuzuQQA;!n=r(2H}u?Zi^FCNaIM-teY&52sbN6^9-92|905K)u5V@6J~cFWo~> zN<)etw0$Q|a!4U2tBi2Lz-e@wwJao7sZ-i<09RW5;QX%~q}<|0YT9*(O#39}3uHp@ z#1Ln!zBiuLZ46lU;6YtM1YatmnBkfR+UgRFoarc<`qVAku;nDWz4SP*)R-VE@> zlH*IU#!8M&h>~WtW{*K-=r&N-Jxfxb)^o#?tl->eDxBS+MBgg5&_>yZ)N@TQYTO;c z?)#+D*r6Z5VbgW+aSMX_+zjwMTnB$_ufpcgsnp8Y5|>Ade!IO#abf>^f_=UlJ#b(s zd-+}v^yF1Z%yIF??rcM{Blm*Cje-39gK1d5WgwetSwZ8TII+~Dv)Dq9Qqa=VWyXW| zvJ)vKuy~z09IQTpDkkb!b7~jSs(nr5Y73z}J&&w08Uvfvq@)b;TkS{12; zk9!XYaTUAKUt~5eD3t++dyT>Z#}xAF_V=RQnP==YRb=>3F9rVdykT%zYbNV+7hR%p zVm594R$Q?#nXcL`g&!Xtf`zZe`)&JatoxV>syDo7r^zn*-*{8@Hf8}@uIv(u*LK4D z=3)|iN|t>JD8X?F+35fLFK#_~jL6olp-t(d=&oxK@Vjv?e2w^t+AHUv^{Ybix=NW3 zR2{(lJ4$ht!xq8i*k_Cyok<@*x1gT8AJgo-BeZ0$K;KDE5%&6v8OVfj*qXEt-2%qo zX~lSIK0@3z{%FHFAI`vzE-4nc{yXOOKSJ;Dx^$;rEM3$zTik20^Nu8XkiWDd@I@`en&ZcIWGoXO>ns$9*Dt|G@L z$ArT0{w!us2wUE-h;@rc@tjZT^zWw;RPAaWbM}15`jw}#7vJxZ_sRVvMR%?e-K%GX zWzs;*6L)bsLmXg_h7;H1)Itv*+d?hI4oC07ZM6MA1-zA`fot>SxD_E%#nV;3vD=A? z%4(XM$Aj6huU{cjy&4=2 zWzbh8%BWmt%6deN)iM{+zfjl%Js?w^Uon&ol8kMv^ro zLkMp6!;255!egIG8sa?^GV0?Ys@R4v*Y1JN?~TINj2>F6O}Uffi|DqaeYCms7>)At z5T6M|x8}EnpJ0XgCVy+oFyx!3B z%&qt!`!l}>5AMr@XTNi>K2Mf;Kql@!vx#o}FcVE~`GDKs=P-U&J!riDQb_KO#Dpoa zOtO3*_0qV3dQZ=yAntR1vJ4{07|V1&Jb?I<*1Tj5gI5hH$j|r)gQXSuo_%uA`tO<0 z|9mt(svnN=(ekX&E{&+mcL@iazTwJpHCV0a!C$tX!)M?64(Rwvh?luTZ_IDx#wa5! zP?&{3zjJKnS1I27s}tA0WgB@O+*0J@?L^nhSU{b83OhNJV>d39i5l8yDx1OMLPbjmhX$H7F=q#xa3^=8c7wMFDCOGeC zEM|{LCq7#PU}jzojQR7JrV2{1;#Cf)M*N3LV@+7=)E@Hw_$6$A=#QgHzi}g0$6%Z2 z$-_1q_WjKkX4sL$yx*K;_Vu~Yn*EQis$GNm;%m(sm_XtShm(a$tr(iA#zv(*p#v3m z(zUUb=plv$7w`ES%Myf`X0i7N9Sgn# zD196Xk35y3b6>BJ8>R!sZ;QF!LA_WPlE6<7pUYnh+Xtx=(qWg%Y1E$?$mNT(;_YMS zX>>pg&Dbi7qfO+o*;fti^E$|K<7vb}Z;i;}UrU2OPX({IQM~0z1DZ9}7`~LdC`-@RI(AU`oVWSgczIKI^3EezAYsdV3puZFx`QxlYNS#6?23C&SHlQ5YC# zNPjPHkc@xfO_sH0Qc|~(H}wkSFArMGTe|O|7fv=qk~nLYBo3nTqGMS-Jr)LT9E?$3 z_er&}7pc3}pC8g)2CYXQ36t9k!F$~=7%_Yxuh^0+e6NfYXO~wn(xq3>8#k$_V7V+f z4~Y@ob`LlmLubhw-H)`|EFWI)sz>wiQWD?Ljm~cA@E}ZzEL}4IzrNa!)fF|u*tr+z z?w13}^|T~>$l~b2IZA@->_@~Y%Mu>Hal%47UtzIcJsIS25(-Tx@!IQ>dAU0o`~p)` z{>->Ne9E9TpfGW`ef_&(_B$KY?3Z;7E|l{b%nxYI;A_>H!8^kYR)GEAg{j;`!(D5qy(oEj%<_M*mB~@xm7B<0U8|xQzcv&_;{z4C_yw~%9?S{NWK8l*t2#4fP? zi4KyzT859>;|DM2?1tkt8hlWMGK-yC$L&zs4r3kuk`Z1ED)wEX9s?=pWgiBki8b_{ z-%t`I&0~D!YCP1}0W74Jb9A3c`wtkvPcm#Lrw>oZhGippz2Up*_dE{H3x5+<`t?2e+ zg~0ncgF*I3Zc5KF5E33i>)s$<#YLR89Ywz2&zFR2y9Uz@op^cZPm78cV8bXL`%Mx# z`Ckoq`NLv*Q7!~*jZ{Ex*F^kbC8009c7fiXQ*iq2Le6N~1TMOHJrrfiK~n!Hq9OGQ zOndu7TG$ktb7Q68l@-R($cpFVa%b^XX-j zU9kzL+%FZ@4T%N6B`&D=K84$N+6w%27eVeW5scn-f&NQdFJzwyqmz5P7owsNafRaMk<(kzIF(4Epzh z4Dy^rR%aBUhB%Ww-XJo}Tn`ng+)t&2HBscoxHCfY2@BZu<_(uCTPtX6v@A+h`&Ce@ zrf&DU(iE)HPeN@6C0{?C0$D3Bcry4ZZ1S58AtLv^IOr>Vx8@*SJRt&~DjX-_U88Y> z+kR>`1;F~?ESMiNAI47fBc-8lV7}>k&`xr|g=cQqjn&#hlLE(Zsr52Az{r;rERGZz zCyChcvxWu_NGa;sxea~9N>;0x!;7#IL1Oa? zoDn6jWA+K_2P;q!|BUTHakw}_fJ57a&)_4O^ud+eJU3Br zANQ&#Xi$+PKGWY z-1NbM6l?pkc;lNGvpx|O>3h0qlsDR31lpeAk4meD;Zey^RQwZ54PyqfBo|#ab#D3I^=J%W1c^pTLg*@Bl&9o^)-6yNsOz`Fs*ByWDFb5|^FIfGRL+1`jA zbavEXIy*b)AXrL6`IFp@U^U))PzkZ+xY5FV45@R(cUpbU1>sR3# zOD%q&c__A7oux3y8D2iVRB(IlV9C=ZH!(JTByTnUA`~`hvr}i}$@Yh`pn`^clIC_g z&iW3mzH^zq^U6V;^bmBt{Sni_1mDd~5q`UHm~xHSjr?gqE&mYA_k9OteooZ-p;BRb z>k{hsIFB3FIvR%E4k!N_50U_VU3BvwirIe0@s4>6_Vmnwg(MIX8mGgtbWe6yPlFvU zJAt3lreTV+4P3u<5c#ZWMbe>aFyWsi?w(yp|2?^hZ#TEma3xJhONpfqe3sz#*C#li z&QPLjx0$)gcajRJk3_D)5_Bz$=+II9am&F=Xu9YOrkwtaACi4B_vitL^`8o=*Sfgw zc0We%-o;zf52DlpWj-QTWO&LHfk})PE6|_As;@SpPV*5W_0$X3t6ql#PDh}#%8}pp zO=Kdd%oE%LWcZiKZoEGp#l|F4cD~ONyDknSnKxg7x?n)}Mtx&lA6ByTq+X0{>V!7w z!Tf^V4*ZE*0#04@6g~O|Fk)uTLL*(6&g~rb>F9p?=vW@vkUEVd#Hx$=8PRE#qQgI4 z6~?O!Sj&4&cn&uF6=FYmkI=vmf?2^F=vubJnK^O%x@L0>UM-J0GVYkHVoP*f6`^KF zf-uqR8(df)!&~2V=gWpY09i#n;d%BtA@1c|a441V%5(PfC#E^`vm@*H@Ml~3Chcgj z2sP)&RD1{T(Bo|XK@DbiYYEFd-bF?!0X;6c0jHiS7k_zM#blBbnEqFJUXQc@jTC#M z3YYl5HtqbI6<_$k&KHoPe2hfd)Z)XG(fqWCO?<$OY`!whpDG7$rE>o*f!Hg9__*?7 zcr|A{e{4uAMBX_90j=YqGbMng-jL#T_e{n)+c{!&-vCmJeE2Wbvh>2^%{1g;HVGSe z2zAbRqePct%L<-%KR=#-){p16hCkqI?6UYd>)W~1f)D(xv*YaL?40dyuWzKj7TSDE zw=4g-CKO*>z9zZzZv)!zcnBjZc>ec?2WaBEioKHPvG*skFgbB4_01c~9BbY1>pN9a zSF6Kcn(V=MxH#aGpHs1HyfHuBZxW6_qJgd}IoSN?0|aI@adYO5hiLC{_-y%naI5Qr zdo2?%tE-foFd&ro-=tcYxO4?qJp8brwp0sFo82V&dfG74^d{+)KL;{e*I?dQ!Wv8` zvIpCpY3RZQFny&9KQYOiUq4J9m?Ia&{UXVB66WJvDrRmGA9ej{MPBAEB2roIBuMN+&bp<^_ZG|XpW_qYsdyt^^*kOjObO`T zdJcuk1#qS@3u+~*;I!x#eLwXhh0JLsm_xw%aY~N5icZ~_=nfuW0%H?oRqly?5mJwU;G%n>*1y$%< zjt>lMVO_)5Iv?yLi5Iirc=Sm4;+;-LddacJao6!)zY|bvXNa5qJyCY9zwk5a zGfeK610LH-A!z$Rp1-q;M7@mU*2Q!7P16~zfc*l?ykT2iEy)YF$4K{D7 zylWqw@U97bQrl_6w0-oMb02kB^qf53G6UL{i`|%^b=?Aao&xGSX^SPuYGO*mX zlh&;r2~G3kN%HKkRB7{WtbA|?Q^cEK_#AC;aNG@J*M{;J26q!*Pkjh(C?i6H9AAVN zxV3yR%hq$SF36sOx`eE~r^$W4lMm%`9b8Ma9BEkFPLygD;ecf(9d$7T zUVTe}bFrdFTsDBjNwP@458YztAPLTCOo6vk$3ov^2`dytj$yyo*p}9Yv!kzJ^|}B& zpdC&pzu8MRb?Jgd#X4-t>I3J#B>0`J$98=*XZMTE@%}tzm}Wf;Z|j}nmag27lWUIC zxU_j_bPDsU}M|<(dEJ8SxL7&XsW#?%;_o)`jRi% z6ZM0hl57yx7iEjt04p5wcQ@Yun@j&$N3rVJx@=2i0{tFkO;(EhvHDJ5s=I6=TA#d6 zBU7TVSbZZ-n==vbY+FYHRpm%i-%g^n&=MLQ_rk8bF|_z{AudU7;F3RDir%!vlzdRa z#ie^`Yn>)vscz154h3Y)tZ`yel_N7$72(UQVz?F9#GT+{x!1!|(eU0@35Xn)8xv-; z2mfB1g|AU|O9>v#oeVoK-^TFY(#)Lm=f~&S^Pz2; z{2PUMd^rC9>AP{BdOWe{yMrU{h2T}wT$Jp(O`n~dMzbGWAvee63g;IE31(kPDfhcxxc|DI zoJDOCcrsnG)?yC7(o}=57p_BSpcPE?2;)i@kAzDjClSx5*`!wYB3Zg{D0?6!gT~7b z;t8b$^5R?>Cp^_6KgaIG=;Ooa?2xPE&ZBGGr`!OlZL=McGzF4XFjLH+O5j#SHh%kc z9m_h}Y1+PINq|~9N!V0L9NorZ{H$>}`C}O!ut*J#cdMVy!&K96+?p&B0)98<90zL$>~P=MQNw zgbb+!@_D6%@6EO2EjGuAgt*1zJ}V^O9YaVbbpR8S7812@0Xm&MM`!%koHACnTdu7dsyr}mM8#=e5A2A5t0Ri_z z>7KjuuideQSZ?<#+F`yyvaJ6LTq{1)SlQ*ku?kzhcXb%A`1n6Q^`a|=OYOrv{X$d= za;AR9CxoS&bm`KXSh8uH2~_{6fa9trczI+GUHp3p#74-&uxmGq{v^2<#`#_&UJi%J z1dk@_BC-$Oi2aO`i+hDzqonN~T8y?-t^Hs=e8yO#(Gjz8e)^HLZi z?*S&G)X21sp~Cq8adw{JSiW%|M=F_FSy@S}Q@q?b zm8ewKb)J_ZA)`X2DWxQ#U0TL-J+Gg4&wU(T@dn3z-RJrHeZQa23QIOOcN?Go`3I#c z#*qI`{%;$UgA8g*(9#v_Veso5_S}YA_JP$wwDH6RY?yTpHz}Nj)0w3xTc#T|uTMa0 z=DovP7w=?T=wSYw-n3x2V>DnD{NL-ie+6r8E-Bc3*IlpyZ`0l zFn&*1I4Z>b;4^B{&z_Oz?o~9|u!M$eV9}KP5cpr`B(%0hoF*^%i?+9>QMce2EGj97 zm+#!dcf7A-okjV$+?C&f7A^*VM*Og+)-4>qy{~fptVARhEx^mXuF_&_WjZc9#J0Zf zr5#;z)CT^nI9oqNQ^MELr}y}~ZqqS(CE*ue2zW=+j=dn|+e7)}HOB5IvtZ+0G4fYI zi^gnG;LiS;MzaGa(L;LsY2HgEDite)HJZcds;V#$9ra!?C-f5y&VNakx)jpOw%5^n ze>3#=NjT9>EulA@<*9_EFb&fjrr$?X>6_kqqUw}QZ|~Kj3*@5FL1v@)*UHMC+UYg4rJ&jfhsM|EX-`!G~`ynyXI5y3VdZe`1hbx^7S26ek+ zNb?lF!z%e2t=eV`qkJYJpIOI{;gjby=Hm^z>S#W>Q>%-%MYocq>Pg6N=lM#1mtf@K zDS?_&%Bb+MYIYae+qL|8ssfeh2Fv7RScEOXfYjBO!7gYF767Dp*TSa;=!SZ(Ns@4}b=*6%KBFMCb z8R9u?o~RaDW>SOVnqHvkpLBV@G4DG13h}n^U3{eY0d7=1fc&&3l9Q*h1Pk<%XrICx z_Tio5?EI{Sw0zqO&R@K&;s`;sMQ>;JH11ApAaq&@vn9&`M7Aco)43W`X{GaeP?vB8&$`u)GvZGn>(P3$Tv;4)9YO3JomCZ<-<2vO zhGVK+8{~=Oj7YdN`*Qi1jUwr_y+AGxuLaA@BFWC7yTmit0R8#oN2gSz(;2e0M0NNy z@|#-Eu6{2sP>`O2Uh3$B%PGO;u1UNLDxAMBo0EAY~I0NqFcZ6k~mWy)NwpJ#H$)D+71orAr^Ex2hf;aNAD%d9U!> zKqus-a0@x8@8OzL|Iksszoeb=irU`21kIXdaram`8c51Ur_=VL1aE|!WrEmWjW_Vn zsZ;dTI(K}&^*HulD_41Y*aa?IJWOU!8m7b1Dm3R8pYz%_h4X0dKpmm)k;#@?LKjPr zC%_30>!wpftpJ*ONT24-h@na@>S&=8#zaGv8?cV0cZSROAFV--j~6i#bEYy6xoTFY zehkF9>%o^6BKZ7jIhwsliiS%5MP)*72~1PMdCM?;bIgalw>KqvR|K%`x+!{7+JV5{ z8)(a^ljyFqF#hN_gQ}XT(|I=iWOTnAR!X>q#2rN7S7~`}t!Xflv(_eQ4;RtTXQp$j z9KO?66@4Tit_pR&%%Xq1E^!0G>$xI3ju!hEW7FcTSmg8rYO|&oN+$oq!-bM~)U^>G z9i7R1IcC6^NF1cf$Ig*}+!f?reGYjpki_%7`JLVFWU}MfB9gjfA-!RBhFCt$Bzq(r z(a-Bcq~TK_y1zLVD_Ue!#Hg9lO3g+(+*gLjC2vFf4drZH)F5fIKTMNTmLk8o6t5q( zVh-d3=2W>Fv#Rnq<7FwwygS=LXVgO4XxfZ3Uq;i*0w;RZC5qbJzE3>EpCS9YL9nTG zH;|q$Pgl+2Gsk76G+Jc>9Sk?3`iq)qm5@G4{ThXf$Iu-_y1CjtJAo2OIoEg z+mFuoeNQ@X=+a*CDf|w33@-`WM;9&sOREP(xtAejRU^7ewBh*=^l2=ZI*%VG_59r5 z;=>#K-(nShHgFY-{>&mjW2$iMp|`kEmm>L9yo0#D6rVVYn21|*?6GepAgd%xn$wk>GVltT7^{&6Uf|An3o#?pr|kz~v5*|f~a1no_q zh5cJ|dH0|&^bmXocT^KrUR<8)&eh|o`drug(;epgBtdHLPFjT@VAZe zc$WpAK`m{ihTUOwu=6I~VtNeuWm$6R)%kSef~`os#v2b7Ze)Db;^4F(J(~ESgE+4; z!8?qK*pYJythcczeS4shddUosHP@r?GW-Cw+OFsQC7whvNC)k0^G8i4zsS8^W>8xY zS;hYo@Z%{>MD@22t311yD(d;5zKeM%pTF--n|GZq?KPlcUwEfQZY=%p%@_)QqKf2a zS<&j2WWp-(Q4^zjuDb0iXEgr``R(3?g=K`9a?3i@+hfi1S0qqS>sNr|qpe3{*FfZy4_swQ7_Q;CJ- zdNdad;r*zLY&xY%<7_8j2c=ltmhyqE$o~PmPVHtDqbHDjn?Ca5XEHR_lS0ykW9-M) zD00?y8=8AspW4MY(YGr#*vUT5*rGd+KMm)Rt=>}j(9kG8sFH$N?P%ojq8WwqTw04Z z1^U5jj8-2#1qX!VNv5ADU2lJm`Zyiqvu0DdsrMDgZLtSv^o(iMtH0rBaitf!F?TZS z40G`2PZBtLqJV5tJWn5IO3@3?lWEDXOVmEj9-ldGjd$OggUQ-T@?Tg5GXIf}{A$-? z>wl4OiE8nzhM#Ztx%?1apHSn=6alyl+? zZILyg4sRoAtD))MjoJ?Pg zw|}x?-*i7f+0)vQvT!UNg<(|u!RxBTwSDt?M%VVmt6HkLJ^1l+l3#@RDhy9GylONQ+%^r zm;`^c!RcaARAu*Ux-fS-4!y+Tr|ew(JfslqJDY@DXGNnb{d8jCzZ1M%x`X`v69K=R z-i4dy=3(cIQ)I!-RBWiVAKP6|WaXu|&{e1Us}6g#Qa9x*bRb}u6!vYVngfmaO{o={ z^*55s+hoNJuTkd~jN9SCr#wUK$7xo0?r|v7_k*qQUPN2oFQId1oFr@mhnBTCqNztV z;ZhX?y#1*viG6pG9CB+Tr?(H9&oy3!rxyN3?mpU;-TCPx)5Q+{laMA+!?8TxMhI&! zzDTG1cZYp(PMS4UKY{isCE>nxJpZWw74~rnAO(I^aDRXvT9v3ygC|Y`3zv10i?1fI zUzO`9kQV1IF;nTX>(;1D{2KlJ$qe0mo=gsmn4+PuFSsG(KAy4p66w4AlJrMK(@Kv% zGDUlkB_R*k?B5d1&+&=)_?#FlsVsw9rmUpvfeg3#jt!WSwVFPWy+L1x$>QZG5gIPG zC&%yKAmMEnklhuYXP1$KH|c9L^#l4$zkCS&@8L^giiN3rOe$X0aSLXveePUpuqqKg9vj4a1Y@+^p`NyO zcv3~pEV9?=1RZ@C$o)>fil{t276i=8-1EW-GZox<<%Pi- zS1RFUODy_2*onU$MxG6z&=<$?v1O-d=et@~=Q~rC2DGq_$s@rN zV`1)*%4GB=)s&n)If;8#H^kmGP^S;kQ#d<)A}hTa!kZ%Df=ye3v4qSYv||1Rve0$| z>sNT6uoq6CQY&j}SZYkC#m-~zO|!!OIy11qqYfXfn?hUllF;3H7i_U9iv91YGi|ND zO84IOpf>lmki5%Bam~t7HrV|+D!p|GwPs2(>l8nd_w^G|#QX0g_~r@Z`JW@{gj!q~ zpMM+JlSDuH%45|nVkl-_Fv-hUfcxc}kg<(7+;>P1i`AvW`V=>u9Gge>Ur8sv-Y;nE z@PEw4>`9DB$QVkAe2nj&5~2$ocaSd`cks-sQ<=cqsn~1J3DWPx(b1xx^xe#6VCOu6 zKA&pNiJtyYDci3_7smIn`+w}f&3g`lpsk^F@9~+;)l?PMr9qgue)GpypLY`>HlHdS z;qQ?aQ|Nrp05&whh#Z=iO1XKWT>4H$^mtc4I(V3&o^XhE(`LFqUWR+NGnc(NsfQK3 zb%Nr5g2{#>N#t#_HkMmqPsewEWv4Xxv6^qJk%h!lG!kKtvqqi(yZ-&~+geww_bHz1 zYrM|w$}Q*inHf{P>{2wR%ZP;c>(XqWX;g2?8LDoRN(~%LIputN?%=H`ddc-adU)wA zRP>GEkfS>1QC3cs%8nCuNgbJWYX>^~ZyotL5Y6gknUU)ThIFFn6!!NgVOqqyTZNN) zX!H5aXus`J++FgKG#Q9deEl%dRegoPrG-dxW-$sYAm~&{4Spim#~$WCdz&9flMLB4 zT)>_-x>w^XyU1%VN6zYT5BW2Y{EB~QdfIMUR|vVUrhN1&{Rz_C%x4&aI=J&^JGc_V zA#R8{#Ijjm+1kJh#5dw8J1J_E_+6QWyv+g6feoaW17FfvDkaEp_7PmMeH)sl{U7n5 z-)P+=Y3!G40N1_vRe5wn7e%LoQ2dZ4Zrr~E-D%Awg)!Ig?M+8$ohR~Vk0UW-McSi$lDolP;_?$bxtxD586`@B9oY!LoWqV3P#ksWNyYz9@RMgG8mu^kc$FJ*t(UyP09K-K| z4h4@R|Ai)aiCY1(v|d2DYYo&?B^RzeR7>9eoKACEKF}*o7iiYbNV;pa2(6kM!+lxK z^HeO1x$gQfde%FiCO%K2Vo%g)-;4~J`1l6Rn3YG=BL_+74ov4YIkVQ5h&4SXfh7FB z@vleCR6bptI}!Y)N;}&ZTc~ZM%6mr0YhD$#kAJpy-|wN;O%pg3u?BixmFK0kD$xAP z9aSod?$l~e6%jKwVEX)rFe)}=uDYa=8~fU*>v~JBwLpoxpnsX#EtaKbHjD9ngBLUa zw9?o(Cr(0tBO_{3goCG@Ah9B5sQS7vox$&M3>F)cOyoobE_1kxnd`W)Qa|p^Vxel^ z$F=-DkLMh&cm;)$Uc=?@ZqOp;2@)JVN|XLQrwY#=(~Mg~)c%VnC#u45MpaV@vpRtK z`^Is5a@?u>j2tR5!=3(JDTKd?KOwi&TIjzk{Ps*UlEm%2$vajwxQGkMoPWX@u35v5 zeoOABJrRgI825=h5j#vO4p-39dvVlM%9@jC^yAKETT#skH>f$Q%=vYMaSfO4x$e<# zT$iyF7uDy)-9Bc@DLLMylKnbd=+(be$F7qsOlzXXm&@rwTQ{!Zhj4ZMn4miO<{he* zokZU@>2b40WSQ4@6LDN#9lriWfwTIBIhDh+xF_EWs9N81*35kw4NiK`KAF}_9&g=C zmx#<^ziOu8!^_s=FxlU%Q_4&(WJxxccla_9OSXnorZK<#qk5 z=kFLJ5-L1HVeDq*gwhmNcEK4`@hKhEeVoOOym&@4owhXnlpmsr(eE(OH<0lT}R`Po`FF^`<)R}Q5+PgX9ngZaMro``UwsG0d zGdWjd5#skdfR3Hcr783Nz&YP+*#qUuWZUq0^tW9Ph0g3iU8Pg0xSKdUW&4agx+hA( zs@b#yJg)lEW=Yuo8Kn3yfdS9d(dG4Z$aSVS8C9z$-5<`8bw8wW`Rb2sRGlArwkCpT z6ffuQJo!TB4l$ft$})Vyc_JI~dJ$S=BtxzYpD{)_l;!3)a_(GKmSl`cuS^T6h=b(m*DXB7;GN57DxA7 zL`N@X!P{mt@$!4q(V~lTRWq{Iq4}3*;v-Llarm-(DD6r{RY_(j`RDD*LaXP)>{<f^w#5-Xohz$ zdEQz=?y09j2b*cE!KrK%*4K&Vw4Weq{ERTcc?GK;nu_?l4c@PSNYCHXa8$`1{a(^Q zW6a zL%xqrqE)5#q$6V<4fcy9zL9ctO5sFQ;+oIeXRXH5M-E}@$E!$R;+jgmAtkDM@ELGj zvVnN{NFwhCGwHzqjs%NA8nS*C+VSK98#(wz-L(Zo;>RbldWH_=$a-`y zMS8h$0j-TvWd0>>W3C5LCT>bO znkS{sctq$j8_VCJ(_;Hj2<#=F1H_5plyPhYW0^b{N%!xP%F}{S^6aAEo`WaC$186)#C0guk{#X(LC&X$_Y%zL+s zaZ38dNLH`qmRfJ)f@SnM*&-Dta^-m@rm&4^Y`MW6zF$N1eZI2wB|pj9s4T2(*@sWD zKe4HTG4sZo_oHmoWTG}MWe(q*i~HA%L(w^sc-i)k7|-X~y$++S?3fyvb?_H7U$+US zyxCi|d`z8DjPk&Nm7D30$92rmNi`;4ryu2iIKedLpJmvJbf!vrB5odD%*@U_%NSIf za?)3Ja>?4N+&&czMl1e5#^Cx-{MEJ=M}9hsh1TvN%lQ3F*&Jgtr+5&M0Kwwo0GbTAxOnjZd(5RZn2Sqaw85I)gqK*ur_Q+DJXmC1KBS4d%esNyI%{ zlgt+_M$4Ya)A7eIkn^0`=xfnMTBMP{+DCrCiAQT_%gY$9?YBOs6up7|&{{)#n=YW4 z`LpTk;tb;SJCcrCPa*d{giu5NoD_X(A3c1S&tX-^&;u)WlPtYTvc<=f?{9v^7CkkU zCs%DIzrrQB+h0q`H=8yz|J@XN_v(7OJI;y@PO_!aiHU4L1r#?vrz&HMB&d z6Ip&@&`QY>(zQN?ipVC@Z*yeu{56ZXf@Cf3#~&H4zE_Uk*>Rb66r>Ov&lU8^jt1EE z#|DRUNqAVI6ur^6g39Z~nPYR*8IwXUd@OMUtK?98zd(`kl{~pZ9G1Bb68Y$yoeFw)tKf4bpx}r&yh*WNR(8{a8!eMKY;MXgizQ*$n+Fr19HE z>x$7#2jY3Nkc5`Y;)*0dmGAVD^G*T|EI!8FQ1IsZ(z9{V{LiRmM2AlATS-!$wxwud#k7BypG=kcH&7&(UV<@~ELuchh(22Jva3($?xRmD5PPtj!Ga!mn%TFNn z-Qi^3ii>!{m0h@ncOuSk*+gDS1(LZ_e~_PzywCO7b}W8VmI>ZfOJ$GzpqYg~X!?Fb zoF%J{)PJ;M*KwVyB_%4v{zM#+h|?pn0WR#y=QFBy#X_FhUV=YwoPe)Bu%=C`YpBx~ z8~k{EDbaP%LYuZ?6010h&g*re#4dB%-M0nVxtzjAJ1$~@Lj@_TSx8=~p5#BH)Ugp9 zh0X0Pq()qt?~SJ-o$piW#w`(G*FOpB`Pft-=Jbwj+k2Z1b&k-9wgpJ-MIN$@8Nyo= zx8v{3VywXPxZc^bY(m~0l2sFc{;FrdRKEgj#`9a!)Oe1Ak17-5aulDmdq_q%Li1Ga zHXia6rDn6k>AWsa`YCS!tNwe&yPIUN>}Bza^;U)G!ldcAKs1WRCCD)6Y8{#QO>I^0 zJ|toF2`-G`Pkz?5SdMp=DA8wDducd7mrkB=8!P>grt@20QQ5Wabl-0`sOurk*cjB} zp13O96w3Rn79Hh%X~h)vR(qW6imYF*Sn*zZ5C+yLu(p3R~sK9r*W{#1;UkS6B$t_ zx;da6waTe7Ei;p_%4%%%)+SgXcrwcpM+HQ6q2*q)_6;YD%N0K52|x!mc5WVNI&xbDW0vg`!0gsrdU6 zf>o}s#WSsX$gW*~aozJ0Wcd?fG5^CvC#H%uw;6%0uf1@1XfM`||BSc(cZ!-=CBy7H zNw{nD3cg(Z1b1b;LZTCnQKWB+WunCxMfN394V+DVUtT0?aTM7u>#j=Do(LDrx2M}C zB_NqU1~j_pExEE!8`tHYK$R!22{v?yGb8?TKF_T2K#Bri71~?chzny>ab4eKhB6Y^zZ8nq_8xFv}KQzbmIcSp?g6T??V?MOL!9#8y*xXbcEo+~^)Xw;hUp;ihH9w5V+NH8c zUpxdA4Zg#!ve7tcLLzROUxeHa8ZO0UqTYT`T0C9ZOuYozf_=@eGYiPejYBabwXxN$BFssFGwvT z8)-?qVzs_(eB#v(xWhS|{2lM4PSTO|_U>TXKl>)TFHo1B2{nhewZxgU&;P+?i}>BP zS2&*XjGtLMi!l3(s>lfoV{|q09o{866<=`KfG6a=L$z@|_|?5zxb0XP&I{X*Bj%k( z_QgurB~gwhY=4gKy?e&{+Jo@Ho|Cw`Z~$vW+v6J#T=7MIN3Ih+j>o*S`E$p2%~>wo;2Zn(fGtqhXBa zD*hty!-ARhFGx_hG?FYW>_F^KU#{@@75d!uANe4ci5eEYgX;5b&D4igsBFh2JU?`h z%_~)y5|h+7}v}+Tm%OvQ7o-Q&9lr>vQNa4MY4Dzk^Av0Pi2z$-?a) z${>29Yj%kGB^@KBlJmK%#ntqFV+u7p#m`ipETdU^i;#|x z7v8_Rnbv%L%yUT^*lpLsv0dV1JUG(C1)h=RCWd8ELAXD=qbH8)Yn-IQCxUr4;CJf& zw}oyn^1+U#(~$1G-*oA*L|S;nfQHvUp_7+baWgJ0;$)x8aYL*x8FP3|-A=VouQvoW zHi+YxEFTn)PGatZ8Qtjqi4L-l@XtD9#%uaaCg<2Y62LQ4{=IoXz4YhPGNMjb4vmmJ z;hEgho3`lQy&2pc2@wuYy+(s}FQTU!hS1TksZ`BBj{FwgMW2a_bJzGfSw_Hop5a_Y zht{=HkHD7*7R|(!y?=1X!5VQ)b&F$ z-SKH2+BWKq8=Lrk=++F}{cI%R}OQh(zgFmT5%}S)R zooB!NU4z34f76L6^4zaSWyFNf_5Wuv3oFY0M_pb7kfdq3B%12*%)Zm)^36R&&maSZ zbjnchs*FyS(4hO4?REWF}J9cHeh zsi*hT@ZErN=ROeKHh)^7@}u(5-lg=yYkexH)kGFvd^Y&32_C#v00IvfVA~Cms3Q3` zS-GX1v|P=h>GKbxtCywmiOt#QfZ!sX>^e#||6YVV?XXy1F5w>oAA;TO=U2Iy+I^y$5){n>i)-=aG9y)$l>}G2Hvr zA9K>lRG!ajSA0w$Vn<$+o$JJz{d*~%vqX``5ML*I-A`fy(xrW6#(HGJQav`SBuvhWaABO+ z`{ct!ri@_5QGr|52bdT2mb@djgMQb0i8QlzVl8D=X3O4QTJG~0drIgtO}~ejThCLN z$y?pr zGncmKGe@Qm;Ha-!_-YKslO9E|b*?ImWT_hS-|Hj9DXAE*+mek}kE!5)Nd`(0`))Tgw5K`y-$*@510 zMof6CO z;94T1Xh#+Xe`16GJA|($QgY?(N_^l$pCH!03C_G;MS_CEm<>h?8I|xL=C46F-839W z&wLXnJJb%)NniB1WS<6lElUC2d9O<@yuLyF1P7?(?l&ZKjybg);Ac24v+!ZHl{9|G z2S(~zG4smVjalz;ZKN` zY$#WIQl38jU`fTtgR!eu9SzObqxZi>Q=gy_w%~;U-PJXfwna=~ugzOnA$>m;%Kd4@ zJ)j#Ok4~q1KU-k=+8yMDQ!ksEl}lA_Oys&gThkS7y|hw;?*a`h;ar%tXdRIt&vK^G zOaIQ$qVL9>+vi=}kpuFKYVATMYEc({CG>@vccqGP-t5dM1RFD7z2D($mo*qE*~u3E zd~e-es0T-h57VHuLR!aX^SSqdqTd<(%_5bWiip!;o^@Dfs6_84$I-I)*J;`ueU3f(p8ir}nQXlpW@&E@ zBfTJ>y=M6rng5+h56#+%l5FQ=;7E{KZ#Jtva|m%~o{$`Qz9Z(VPWNa;k@{KlnA{g7 zc$fbG4sO2Gq;2B&mu5#8_*R;Av)hE?msOFL^YM&Pk{0voQ5KGMzDR`*6wu;i zNbk2fVx<&)W-i~gIHB`oy6U+b(Vhzbdbi3$Cmm!i-DAy?lW9=i{^8SQjE;A(`7 zO$7kXH?v%dgdG0Aua!zpf~CdPg24(CSm`MRLX@|HqWX03PP)=u<;P`#+luLMf+T{n z^NZkH*OzeK&_y_-Sp=OEc?r8ew}A2gR>Fd(i7?`3Bxsv-1$=C51>_L_f`YFvPJ9KTX0s|K9phe*i(9lT%UE`m?-)4Jwb%i>* z_$Li%pZ$)OEZ2f^H^fg-XU;JK+8uum=> zZo8@j%G+YVx0gcjhSL>bxLN>T4d=l=mI}IsCxQtNr-4W2bK%lytw5-u48**BV>aC6 zE_gx9fuFRS8T?x%crG3V%%n=pwj9wA{1LSRx%P^hrcbh1f{ zS$or6ftvj#LDt?ufxmqy7}{fKeob~*upzk@*j=y$uit+H8?OsNwbpo3s=CVT>d^y& z+?H;5`9M89UK$2Nz9yUh@~Rcw`BnrBb{U(GUw>%6;FB0QdbL<^%V<3qjI0IIF2;k6 zP2T{0s0l;gWPxR<1Z=o>7kutdGCSIOQE<CE>_ZD;Q8+2t}W3qtHdSfY@YNIE~SUFDn+qTb@f{+J${E zJ8%f>n6L)AqCz+poCLqg8$orGZ=h>N8vL=_1Wx>TAGD}dfL4oZ!2fLJ9E3an$Ef-j4IgX!HGFzBx| zJR{K#e9ZQOL!I78O2!rC?z2ED6AcB$im%PX>M8^o_R{dj){~%D(imtjmUv8FzlI}4u;BqW1dZi6prUpUD=Q;4)#!8sqp9$ZZ2f^Il zI2f`v4hL7p;@a3SoW;Hucv|(EN3|FWdOJ#?vs*3HK9mDHd&JSNhZ9g@crOf+Re%pf zrJ=lwJiM~o7sk&?g$;v&@T5A=cI@8?{YNgsm|K^Dtj{Ch6#EfKer`3t7If8Yvqrbs z{n|FL%tZ=bc@_efY%2ieMR&pPnp#kwQY@I|ILq9!`;B?fEH@ab`~q|>dm+$WI}sKz z8w4#6?wEB{F2m8GJTK&)A~t`M4#i0oypmN3OWi&Jxgs@~zJ3KP3AG3AQ#}MS&;14U zyj#!e-gy|eCmZ5+8-aSZf}poJ3gq_m0?)kX02QmlIl=t%@^7kOpJgCuy7)#gXRkRZ zGZul^^BCM5b`UE3S_obDUV-YqzhLj}L8upD45y#Vg!V-nD!BSR7Q9WdgRP#0 zU|xO!xZ-gU!3(@!swkiD*OtIHxznM2-AS;w*b>@3+YOg%EQTjOF93#xBH*x@EKobJ z76zPI4($gvz@S!l@Z@18XsDNlBOM8Ff<`;g&{7a|&pQZj=+1^};YA?VC>Dyo@r7*m zQh4lF8+=PLVB)wL^qF!7m?YMN@Z}QlkO75;`GYWC`z`cI;j@SYlXbwy*o(82xP2de$0MnidK^0L8VDm>5rpp63ylg+T za=a$+ioPVcxoaBSwsbR4J!1=&%`OGv6+&>ILo3+ZCIO6mb>Yk)0qhKqhL8RcApBSq zY;jo!D(2QfCFw^HSJgutx*AMdw+@u=KMC%Or-S8^$H19id0^9>Lg4Fj3nbe;1Amx_ z=zh>|cthed%;CE+8+0RKpcaKWDm_ry?K9-IjKhJNspy0CG^DXv2?b^BfEfpEfO@?> zNS6{qhAuMbj1NF>-=0M0^W)IO;0_M-4 z`H|~R0y?CGs%<~Oscs_ZTZ=fnzElhv8w;Ro>{0k)*b7$ZtcSXSNrJbF@&#M%)PeB* zEiiM~5JE2pvgvOtpfawkW`Oqf21tmova?>s7EN<*UHTb*{i{kred&F%6!cFAFye%fm8> zGO%!eDfnkg!Qo}6z+-kDsCMMLV*#0P3l|4pAAAh@L0Mw z$eE@Cw4&_6^*SZkH;BPNstxcuS^9^(2*ur9h5>`43s1~-?2 z54q~_%7u1;S@2y!YnnW$lz$0qCp-i)nT6n8-6U}KMZRE@g(*M_TEOr0Vj$=o1AB}y zG)>zA*DqWJKX)3z^!;<8_SM-i_=zrzIzJtzmV5v{?e{=meIhVgpabro41{~KouTm) z5BTn>C7k!w1=jZ4!S>hBVVCh+c<}pVbd?Q+rrj%`z!gLJR9>a6p(r>}Ya;kAuL5U% z2e9XS97L)7xuHQH%#qXsHQK#!X<0Jd_S^*ynAySU1uD?`$QU^Cai00-?w4kd#Z=9W zXJ$d!iSOa*&zk7Z#w(z1Xfas6T^lI5VVKeB1$EzkfYCCeFf+ItdMs#ya{~%tn(K1t zNQGg2&Qz!&^#$0kJq#=jgLC`%^21snRkH-j?PB0WH$Eq8VgaGQCfuwv z8*bE&1xr;$!7Rh;K-EJ8{;6#Qu`++a+z3-B+7%2}cP@a3tp5U^-=^?c)gWBfvmCn3 zod!$Vbzz28JQRId1RZm7;UTUKe(mpoKmV3N6`tGmu<9y)x~BvOua*L4S{-0X+#WdR zT`|nE4uydUs<2q27;SRDgYr)`Bjt(K@U!dVyrMZ1j;)l17fx`{Q?w5X zuec7qiq8r5bsrGezN{83w|OXt+O$a^HX%!}X^jEwHFShkDm?#x{c|80AO>yzegh&N znPz`)?ltc>ekF+2YX&Y0Zh`eHY+ylnF_1iA0zVo_q4n)Mu!-$L{LmKSXA?d`?}A=f zkbNEAQJ)EeH(A0BmzP4ZVPPn&-U%G9ZiP{@!(fA~5Oio0figLp!Pl%Mu$_iOCg2CS z`EMD-NxJYy!)CDP$QE!(V;P9qwi8%NI)H7l_Mr8)Imn*89E`bahks3l;j;!q_^l=i zYR~Tm{ts)xPm@BBes?i^=`|wiI-8P=^NdL}0$xTKIQXDO~mFH8`qb z2Q@`bLj$oExZw9F{Cn^REZpA?jo!b9zEdAT4V^K8k>YpIp_~r$OYg$qStnpizbQPj z#TuPhv>knZ>x`D=nZpwsUEo6ASskJ-35^5}&}-s(XyH9>-Yw|?CMM{>s+f7GV2UG> zT(kw1f6;`=8$;mSuCvf8Z zj0u#5!8y@DXut#R{dfkxiU^0UmkPm@fY~tZ)d{%t_bimJCW{Wce}*+3mr&J$5+s@! zf>JNn3%1!T2R_p!;IN1?_@}P~9{2wQmV*PJUSk!!9cKfh^i5IB(b=dV*#c$W7l9ij z3PFS2T&VbQ7f@}#A^0f12{av#0}F?Az~H4H;F6adBq7pp;X83?y-*%LH7)~5hOOYj z7iV};O9=jSC=+bzPXr<9QZV+b05)fc!>OU)1rcgy0^0-E;hM81a9m{z_*A}0P^bSv z&=S}RzBxRApY4^<`X$3~T;T?k!UHf{<1rkavk{-?r>UwxoN!Id3~%q=VcJQ?m3m+Oz zK#jtSpr*AooHc&|eB&zwdd__nc=+T9)-?OU$W!*P3n{|Cb5)^F>m6`!Q7UkgIc2(| z+|2xVzy$DfMmZQ+l_E&gcq{nce-o^{ogon0uL|F$uET~K7UCwJ_txHj61wlHhd-6y zLuES=I4eL2ZW~(yE#v~>5!nbBdu9x1R?dbZj4X_wdkwVhIV(75*$wPx`M?qHB!S5B zw}Kfh1l+mO4McWK0c9s$z$c&k;K!jzctKXmAWT@Z37Q)?K^Zki{&yZ1G)>(oI4n95tj&D__x9Jq4MpMbk#z}N^7Z+(+PXnF2wFEsiU4jaSXQKl> zHgJNcGAz3`5l;Ld1TRD@!vXWn@LoePaHuH^4Qmu9U+3gh&`q zyrF982O#QL2qxV=3>N-12gg+(2o@dQ0W=#n!^uz+9yJ@j!;YqC zIPSj>=Bhwc3KyUk#Y<7E$z)hh4dCx-Z^2$Kb+`bugZ|L}q3FEhv3|cWo;^bpDaj~G znR)JW-A@Z8QHW?!nrI7AX&9ANMktbKh(wa4R4RoO6{Tdf_xKhKN#pnV{q3LU_3}LT zIp;d>>pG{YQLzTmLsBtUU}fJ-Xd-!uuk`8WADVJx44b+thdIhyve{$Iu)I+XORUTU z%`K6PzwM)>v03D}MW51jRmnQCht*_i!`P&e=F~rBJsW?pqS|;S(_X@2?M%dThMS1{ z=MNDNYA%E*G7cviL-AzEXYx$bhlSz{tWwg(`tt8IL#mRl?;Q+0SHc4MXXNTyM-ne8 zNN)E#@+zvMC$psCnxhWw+EnW2u!5pJuaKX~E9zFQq|=WJ$vBNe#l{#{59;CBY+EeI zng;WCi@`4RgOjN~PML5Rh7Z7??IYlEW-gW|3oNv28Wqlw!Ja`G@H!BSKhxI2;O#)X z%lbzL1GqIIf)x$01g52^9h72SMyCfyRo*Of<`pnaqk6dG$``S9%ule=`Dj zg~JfsIFo^Y`(yFX;5(%3rMao?4TLL5!))bF^xKz6Y!b)7j^%5BHKP8<4q#{i+z_9438B+|{0 zqpb^cn1tyU7BrC4m`6|PMN15f$H}9!b1OXzR7AebC`b*IfrE(-4y?N!kKWi&*olaDuEGra2*-&90n7 zmdnH<-w|BMYS0!xoBEy~Z9KzuH5Wq|RgtyG0ZWQ(VL6f`JzNv--^pVvDk*SuCE0mw zf$vBXYG@VxDBLCFN9@6!SBr7_-80IXe2Of7AERdlhv@cXKZ4d97FO;Db({4F@$y9N z#T3*Io(|o8!*JkQ2CeH_M%{&nsYy!~ZZ{X<(FHHOd*B0;ra&04)y3GC)^PpWOl501 zog0*iBUcAPch_6WpXv;)w~aLLS|Syd5(dsO!%0EUT>Nu3O6Lggio6Zp#kw%R#)&jd z;xwtPaOQ#67jvC@FMdY(KST|Z;8qn)z^G8vr|-cziH(p@T0^^TY17MXo;1GkJ9B-V zC=Q(4WUh48zz!B{apEky1?vDUDPUvR8%rC^L;#n^lpeh9#teh1!v4#Z|s+ z;fH7^4fU<0l^^9vucR*(zbK)?L`?+P?4z=IPgupCV=Obi5`IlzFZ;lEh=$MbYzhdjnd+d%obJEWuf;^X_P6sLa$fn<$^g60!{)Cjk? znt{GsL1F7sD2^$@T>EI$CB1<4?{8Q&=_1&`9pZ4)IC09=IPv?fQIy*9j}rQ&(AS21 zYS-#uH)=<~Xx&FxD(Z8?1ws77nnVQqsY9f@l*UGmN52KZxMS-Co3H-(-Q|OyQ&u8; zxGiaA7c!eCYP8|@AyHjnxJWvBI1AF=iHbQk=zria`FWn9yZ&cr$cYE^XXkrS*n>@? z>q305nOJ$@VDU)@WwEOL1tcAv4P$i!Jekr& zZ?gKrN@BUNw_Jlo%IWYut&PBGlTqvBiMh$9(4Cz`>W7tS;Fbx_x;Lka0Yq`xH* zuQqR{pyXUA2P^Y$QTJ(V{RBvSCPaH?W24LtR0YMsSop1vSCXQK7xHNBPz}z@wxYsI zjP2`NVfE-bbe0~1S=D}~G+;1&a+jg<^bYiScoH`>s?g9d2rH!-vWySmaP}=`)+dLw zMrYIVeVVA#7NPs*SQt0`WTQ2}O!XGCV}9X$#k_;u>V6nsXw)D5Y|St~G!oA&YUq^g zMEw4pgO|d7YvJ^0T)dS4t)br$?V!Qe2FUW$n)h+!VlLL-v_{OqpG>a%3eB_i!OEM1 zpgC0;F^P8g{k4h?F(+EEM*}yla`RNfMQvynW_{XE(#Dg*~@=L1o`L?yb{BN%VUp&#D2EG`E&c7${b&?qQ%AYBJ zNj^=K8j1_qzvkdiMfC6hNlXv+o}KFdOvFHe7lgdR=kX)eUi{Tra5dV&V8 znnmv0UelD4_sCQ4f|c!4L7ymPU!02B6XCp-!gE!8KN^OB?+0;q@&_FI_8*qtxP+qW z>EfB|J;g^(%oHyidmewLKEceQO5j8}8eXN~)9YUN72UuW){H-YzCiVIBNa{Vq_xR? zpr`+o{>2xP?lcMfiL&C_LcVBc@-RNyB7rV;R5;HXGw=ru$qiRwqr7TMo-gajY?>f|Nd5x7E?!qGTv*CqkjLXQ`(UoNQ@F{iK0J(yKL#rMOTX0qDcOk)i7;HgYkVE zeCjGe(aLsQQISzb%& z9-god9Zgs3(n;2%4_@tY;GrK>c&&XW9_wl1U6K_Zj8cHrAdV5Lv*2}i9InJ`@b>!8 z*uV7!-rcdm-oLxhS9%vj1=pDNB|%r*t03r`?QG1@JSOE-OD&2oS;41ZvQp4T=z<}5 z@gpABe^%kbCpCQEkU)_gDP(qg51k1rprk2@xStfpHolLe@CSQorNlkV4F>m-YQi$_ zcGjA?n)G{xx$n(mnzz^zz53H}K=zdN4nH5d)n9nqj<*w^Y(n>Ut;OVky$Ebd#bT?y z@TpP6kuE0`j`c=Hdtc-_{Ux8P{qc66H=8eQ0R2_TNE&j2KKk|L%bgGLeK{f?b;*Rc zL=NVEio98nXaY>my@%ZS@k~7B5^IX%w96+I$6dGMh|DgGT&Y7bUrgE2Fg1FaH;8YG zxsM{_!!YmCK-T6hMh($*1r=X(;UT_edisrqoK z&+GVDnK3-_{&KGPa0`F%wU-C{HWDA|H&v{hr_1V$nrVCMVqu=Gpt6U@=*s&mWHLgJ zPuMEY_okPkWYr{*SWcZ{I;Z09G)J-yyhSH+wDDy8Iwq34L>lK#k;=I{l(l0a_IQ}# z#|r5&u}@x!dM0ipx14>LTWE^+;U19cy~0*MI6!aK2Ek?Q2xzYyOZ)9B*uS|x6ytK9 zG_q%5naW~%snr-NIqmLq?i04`AdtSRb&crpQY9& z!wgxe_fs-E)J1X*7f8Xcjd~;p(bcShm@{GyvP-3CXivolpX!psOA(!2yN{|g zGw6isIoj$SM|zF6=(M)O=1V)#-q8-vpA%@(`hy5AUq)-$Z3_4`5`%5p=()>WNF2P0 zbwa;s5cQBM`Y!}s(dW)mkFaE4K1Ga?69q?GQrzDfGGmS8zi}VU35uXWxiRo8zJRjX z5%3G`hyBuxxVC-*w|;ULXO_N36{+$dGofCac(LG?E#$CjAXTnTpv+6E@ZX+-!|G@7 zd~zff{pGy1vwW3+ z@n?=w$3ZC!5IN)8XLDNmu!1y0Khlw3iR@*nKPzfdptl_}5hMCRdNT$g@5e2&mR5wU zQzEu()@0-Fo1j?k76P~vef8GH>`fw6wcaN=D}TIbScRnvY#^`X2-!O>n3pTT_6)zz zHb`Bg$H#Wio8&wi+TMVlgU`}*?*Vvc9FC8I&O*^InR!|`GyC?5xlNbEjQ$_dw5(L{ z>fOemz{^Nn^qD4&^ActEEoUvh)7W0i9?|^cJK2@vJ}lJSnfWJni0lW{()m-PuwMQV zLLP@xNAXg+D7L0a!W)$m)t53%7L(+;ht$?*4wmK~MfLJr42azi@2^3SzZ-zL@jJ0F zDhR3fbV+U8NP4t*8`VVG(C+g^lwTl?7`H~2W7f*by+=^`8fk86XUKzM4SD}trBv{4 zCiP3IA-BbTbaQ1HD_yryRUmK8blPp1bx71)xn9+}%t zQ^2&4b!|kvpJ{dEQMIo^I6&X*G06kI1kqJ+!x@^1Xnr)Zp z$H5P*X|J%azu|yTuYIn@*Uf*1Ey_t$ z;CGF30(wcYdlSaIID*z;NjMkugLPPKCkwhm=KK1gTzMU&&NSeeaxNvVT}Op6QWW%O zIPxbqlk-}23MdVu6j^_`S)7Kan=E@eaXKw2{Y;ITdCXn(TO<=;&CJ`fKp8PODH?*G zYI6L{uD!_JoKJP4H+;@h39+;HCoUQ^7}IwZqr1;OW18i-TQW6NXcL*UNEE+8OCgF{Q>&?LD;37Iz_PyRbZodkBr_}iDrk4VhZOc zu_^tV*^^9ndb?hYPU9#YF-xK9&@B2MIMph5jidGItpg~*X%hY`Igh>5BzW=|Ia*V` zjb5=%p^lEH%QAOa(6C$a(;~k>8G4*4^oVAuqHDjmSq+goF48I; zOt*v1v0wUkt?%~sCAs-hlo~2SvGcoNazTxUTFP*R+9mASihb;hoHGradji`A%5!Of z)#le12y@~bSlnHVIg5s4jq_2;oxh2xdsNYjhI6EE+RpMOc(MC6+gO;dEq(L0!{8_t zyiDIm38Rl-N5*KbuezBx#RgHjd?jtJ?+fjzld*f4ErzM;!Bx=eeXVBW?Z0rm%8G`4 z%1)eAU4)~1*Yg79ByP5960b7&iF09hV72)W@&-*uwZJ}9>@A}j`-za-u8lW&&BPkk z;&p)VH;3C~adj+y?OBbJQ{ACyY={Txw)FJZU^Z-@CGw(Yk#KE^nHtO$X5d(CdTIt4 zJ2k%GKO5MWI#Ttmas;nRhl7p{?6wC{NJlBTUa~=nk0hK@C9rWtF0Jpn$OCJC@Z+kT z++%|dN_~#ef5+2l-Hf9+)n^()cncjFzD~IBPzf7_hcKxd485FhG=BRbK_|?kMwjG zN3Y2ANU4zB)I!GV&rq<`aeCh{h^u$)!m@5(dcx=7mGmd(@i$9UE98yi)ImN($SUvg zF2aWLJ;+!w3+K~Y$msrLXsT_;e_O0!^50)U-*07QZ{4tbKI)3e~- zdm7d|D$yr!I&CzUK!1fW$Xsn_OI+qtr*ks}?hv$Sp9yqy>OOY*do$wXW%vv?5%)Q3 zN0nC>2|LRDG_64sUK@jP)JKYkd>8WvO42;wZw`XzX>*AfU0$Lpym4tm@GHm_L65Uh z^y@ImRB58~KoaWIEl|EY2g&6b7<(rQWm{!9^>O7vGiCT-Lo5EWnRA(l2C_J81?>^S zjG$4?&fOf#$_FGc$j8Ds_BcLGzKcaNeYwchlaH9QnYU!P^Eitfu2WgewWp-g<=#eO zl|$faSd6y&hj2eX8?{$fK;^?SNNiq?d7D5t^%t;x@w>@;s~dDe)}dp;9b71E!l%LK zF(NSnQzDjQZq0fua9$xp!WhlRI70)AGmqBIoG2}FS#@5%j$komlYr~%4!J0)# z?kdBpyoY$U?wXK(Ao!5)JBeTE2z%+Geb_zlB$h-spv6FvcdWMNt(WciurhmI96yiO z*_}g$`*mzqdV_7w5;XloHI3ZrjE}E}b3=PuzVE9cS2j|l53Wb)`xkYDRk>jBi#gCV zF~q79$w<&SgdOXmv4I#K#w5cpBox0FNwS&NUZQDfKFoc21$eYG7ua3??8XAVX~Zr* z<5CdUTRKg=(`TnR^6+u-U!AE?5IERB3WK?K^cObg{R#HBw3w-TU51}j4aRNzh~89N zYG`>wr-gZBQR{K$u;~FaN>oAWYL|@%`a)cKX(L7BSnEc5dFo)K}kQofE5=1u9(INI zVyes+taLhx{BB`K_$dkZ zL$GGf4Cq}-!TNI%WI$S!J^c;S3U8wUSDm3Nusp%Cscin|)8zSa5zO;)`P{{vZ}7Yc z5-`H_)T4ruaK?EWSvdL;Llq_`UQKU2Gr8Vixw0hWbggZF__r%Pes+ zUzHE~Hi5Ps{7298z3|^Ydp4u}zO~E1H!Swiae6xc9vQ@4q0+9)bgQ2wI)(<}-Lhkd z$dyMyw>!x+htk=MB)SF-;pS8gTk>5K_fvC(9zT|5_1F_Wdeg7}WKpr_KU#3`IxU(x z6ZehwQ|gdJT6NPDN8)avYkwO`n-mR~aC21tyeMSSir4b(_gFg}XVEX`cL)+# zOnc>daMu1x4wa8c`AaS4Mwh}*YcuqWDhX0ENb}e@TxxLysmkI&eFiD-S3}U5q4*{e zp;xXi7QM5jB#G}ND{TtZ?IO6V)uCXfBgTXk34KfmPCku-^reMRQTz$tGIcn=cca`i zXQ=oDlPu@-@X~HlIrEqqPtKwvD=)*eB^ZwVjj`*-IP4xg6BEK(Y0d#xUM>~D1HWnV z`Kwc4p%aH8Eo%{({Y2D%m^aPYG6kRR%W~rj`n=8O9}djw!iZHyye459e=tJ@8T%#gkjV$vQ9teQEA^(9~|Wk{vdOxWeyhG<^41pktQaiN4` z<9~C|-^CL8fr`8^CmZAL_tLF4JX8^%1iroXRBLvRhW+BAbJ9*c4wx5j##&Pq)CgYc?~L z>wc_JEmFw8ae?A5C0sf$%afgyc;%@cSj+ri7u@%<9?Ksh`fW?S`K>I+CxV6P>=T$> zPlPvWV8r_uv}Vvd^2lr;bv;!SUOPp%_4<+CtV}xTQZG8?ds{f_c9)IqT4h~0(UP4@ zea^-6_6i)8E8iA*0+9lnL$`m^sl{92xzP(>M#>>>ml5ppI>~%z5W(#YweIw%`!i!$ zVO$9$8#kljoElo`7rB%lq)ih>(em29sC@FC4T@!=AlpPF)yE-(Md0GVV%nIfOiyDg zMfKBuP8M~eA=1bGN;#See2{yEB%{SKDY{TAhF_HUuR&!xDo3CYyl_wI( zQ$~U2&l!o^9s2OleoP(@B>2}I`S{deif49VXm~LeD`ZtM;*%R$j#fgK++)6{veC5~G35K0 zA)ik3<$=q&#z`Tcxl$X?4jrHs)xq@SX*Mcu_0sB;E?PLG5eXjkaPB>YqAybDXlkQV zDFRpAM;FJs)zS6eVAM@MK!0Uy>GpMDu8WeyipBS-`00Im_vj?v*aTrBWC22NgyA6ZS3r?krO_s%rU+N{}$)82oygd^yHYxH| znG*bQmNL(YO5)NZFYwLAkNC)$-|41tB^k_qM6QQ>=;rYTDj0r<-Z2BN5cUms3s1t% z&4o<#HR#9~H$qATd3VdvxO3~sA+-ptLej%#?uZ3+lIH5P&97!CVM?H)DNqS%%9ZkN=mYw^+uFt$d>n5HcmyJT^9rVdw zTb8bliK8Kx%jsWYRhiXGh@SlaCoBA;PXIw}}a>}Sv_aF7=-|5ePk4eS+7@c^u9`0*n zuuL(I|Wzbje89O*_Klp_T{EB^!9RlIJAmkcmE!5%xcD85?T8_?nSz~0!GySX142>kcF8cE$*nNkzb2N z#ZM*KSaWmswOfk&PrXZlIKei}h!T2sB|6o%ktOsOGaV^McI(h7k<2j#apS8W-0u2o z?(yGStXTFGOI@TntqzCQM|}jgYeCUNfhQjPCHx<+vBu6zq#wDPeqRb8^^`7QH~x^a z2OXqwGV++dv7Gue)KRF*SSk+fU`KnynM;*CG#)pyLTM43F<6qMPWv&L+o#$3dO<=n z-b-Qmiuif<8x7#bcrP%Gi{iE;b6^VYon3>J*E-nxMiR3XONm#d(!T5|bR%asUG{fl zsp9L*>p%oKi=Lss{W~nz8_Yi+kmr7%o?_9-o7iM@Sk$~AOEe*LBFpl;hH7kD3r+2*USbZ` z9~&mJ?aF&+k;p=asX{a1MVH3oe@%VX2WXfE}7GSBpQhOz$6++AfbKRo{r zj$Pc#CdQs(OOotazM~`7tdqypfid*-Z#v0`db3EqG$y(igBCdl3}{$S;e~}Xqg0;K zvYc6r9LQ1NJVFx-MK2uBQheh<@?ATQ)W&}ygM^o?Q+^o>-D`^8Kudb_e6mP#Tqcbw zNhb*xRYW;AAbIt9c)qO1z}vxMsn%)Y7en>Lt1gVhEGVFJiv^CJK89c;Ef`$YBDL?Q zuuYT)iL1L&$px;>wuEM-MbNSkGi-aU43km$wANmOkJ_ioo7y`uRAv{&Y*|4yE1TIe z+d(ua zQV)Nm^RL{=yZAF(Q>TO13$khW`19<*f?uRN>k}=Tl|Z{AVWLBEdojW8V0WiWHHNg+{1r99o!F1L|MF+E z*H7VVHrB)N^)phOo`X004#PWh8+N9~;K`oj$enu*#+I(wTYLzKFMBaNalCkTfrohY zfQ91c;$}MG(vLZcw8=?e(9_Ldllq=R)IVhOsjdbAY@3SPplX(vsadlAQn7L)6YTI*wyPU1K=Mm+vX zA8}1#Jl3n9!?yb^7_c>gdO|ZuYEdPb%NfJHvP$raifQm;2>L-7dXFdK_7xje{cZsn z>KVgkZC`9TFcNZJ<8i`qC)<)XlKMZt!e$S1rtXh#s94Smo6pxlR1<;3oDoR+;EcEc zP2BD2r282a6yt74s}^g~lfctVx_AQh`{quc+7D8@u^wss8c1O)`|-Dbd-3(MpnGJ; zli2(wb@i1;-p`M4^}dPFfl1i2Ko*L@dMK~kj%f9xJYxC>R zUHQJAbo!P)1%FI)v1Ft!zZpG>d-YM}E#WDMSk#DC|2j zjXMmbN1v!k@HVT3O~d``YsKS>0>p*WmWaJ(uVpnFHDvr@H9TibBH35PRPN`A!EwuZ z+VFK;Z$b#an|O$veD1P?ixrv9bY)Dvvxh>&Qgqrw8uj_R@Wu5!rl_o7YQC4)v@gp@ zr?Q^5FOq=H@1JzwRI|wQ^Ko|G*^J6Q6wuPYwNxRY4Nc8Bc6{#x%A7F{%1o&DBSGq8 zKG5wi+vs@a7>b;C1v|qY^W>H@TrKV%~>_!CYrbKTK&M;wL5>~pvrP>Z*w_Sj>3 z06N=JvF5lWpFATTm$(0>xms%w8J-FK2Ep&_lR_JAwX;8_Peg50LMnp)y}?}xy9#A+ zOic#Y@7GcCaZ|o%$0dADSckS@;}BFLhHR=UvWhs&mrX~&z_}Q5Y&lGIH{s^h6)^v5 z3dw%bFnT)(Q_l6{XJre7o@_f7Eoi|Z<5Fxh*nx~rHH7V+4VPU<_Ma?$k8Zs%7)yO3OqDPLt&LZx2svqU*B5JU*`#0i9-E>60}BEd}wkDV8@(#-{|rE~{( z{$#_OUT5O;xFQ_=ZNlg5oXizkUb6!4T9&Xu(DL0oVAKBr5;`W}cljPtl+0%z>L<{Y zG$HqMdO7-;PsN%=GW@Ua4DLSsIL!RwskuZ#RNB3Tp03Lktu#NtT6bTh72npNeZC^y z3YbRv+jgO;Zw8E(>mwkmNR+xF5YN-@!F7rk$$!{DDno8lWVaQ@tG_0du3_}H(*XSq z;t6!S2#skYDl<5fyQe|04XJZz++#l{Z>TZlaIsG5q+v6ffqh z!dg>!bBx+a=Sc+84-G(rT`V@Nh{Oy(2|6aW!JhE~cik4i@>FH$SExD;*;}%>Z&GyQ z&Rp91cM?sUZ%PwY4CvM}9kfv_ZrDY@;p4@D+%`w9C z4VCs^g?rt*aPmPdqTSD8{773GRFZ-A>bD5^IS2X+j#Jiw8un}OC6=6GNkv`p6jF7c ztoQ68R`88}x+pPc@pUpPj)mjZ;p8+gjh3ZV3C!m$oHeP$-*@4-{==U3DHW5;7Z*fp z)FE!p7fi`v*gGc!8=L#!v&C#SIWQf!wsDeE7%pNdcVWNb9$FoD;;r`zdQxahDu)Zn z*0Ub(UfAj)hp?af+RaNq=4q{ai>`)!rA6eC8(B><<(vj_%&fowJ1D< z{=G^dr}sBWvleI>pM<4$!UQ0zf-_Z-Xfa-oL*c3LI%tN!p@t~k5(;HUb+nkpkwMu$ zy7#r76#AZ`(<2|zk^OqODd?R0C!d8V=(3=>%kogu{#-|-!OsNjAzQN(l&aTBGo_YL z!t7m?{saLcuY55Gvf6$gU|7t#$9Dgpde9(rnA*3O4vjFnj%U4k2$^1 zO{N83`=jE)e3*m{Ma&-^e9QSk{enNTIE`rPNWU%|t~`eZC1bc=ILI!0G*Z^H2Xshd zD&5n(LaW_()ASNM`eh>IKYa3m`yE@Fw@O9eEJommtroJkj}tsOUbx*H30iNBx7Gb| z(%BpxcQnQAp_bwkvg5?H{UnffWh=>lDj*BhP;>~3Q3d7A&`unN0&f!>aj=7Atl+Cq zIF0zo1f&iSyz%dbWByDRyx?YF&zB)|+!5sZED=0~Yqx5PC^~lvz!fAP- z_Yu6UDRW4>xr6e2bRc@@PoI|EA^J8H!#1BL-^c%v?wv%kn3+S16&{hv^T9BmdrHD2z zxQMg8%1SJ3;01Qm=&NYpdlTl=`j`BK*?C!;1QKNr(hT1PG~P>_5_&?p{f*VUVEPJf za;Ak6y55nm(L-{vxK2x4!_nE6iiXfOvUHe%@3Z3Jrx#6TC!{g;*=F>7UCO;y`18cG zGkJINTXGG%jb#g5d70z}HqGU>$n(HT_Ch(;I^<53Xh=UfHeWv#-GifX?e|VhUYS67 zx?0#5J_n;4Tv?x)I;>;LFZ$poD|mC$De`e8tyz1Pxh)$^XLaV$jof-1Gd_cU`X>C& zwh{zu?MK0Tp{|;E!grfAbVlE!6_<>FA+zu@Wj3}nB#86-B#Tp>e#J+zWAWwh&Ho+ShUU!KZ~nT z@ACx*Z{1;HyQM7CF_cZ1)+XX1dF*e&bgEnXo#h0D(7f_Pq%WFAJJ)oxm$NzfI@}XI z%~vL=87Y(+or3NMIr#C;f;-iX(8{NMCr1(_sEFPS9PYu}OIEz6=|M z<8~tIdHau%+I#lVIu=1myJ6N54P=X8^L8bAN;@$~aWK_wf5!Hyj-oO33bY}_fF|f{ zrGK-J)0+^XcJytb5u4O-@wBkBcTIv}%uxO?N`|-2(&NQ0UunBR7b!7GybB)=!$xlu z{#XF7%jH;9d>dJPFJau;ofKzi1ka3MJUyw&<9`n4zE1`3mstY$S3SWa-kju3@;T)7 zW+$?4oyFgsv8eoY1=}CjAgD+g2R2$`oOcFX{J+!n+@)wapu%sfgi`2CDg1Qgu*&U$ zS&bae%U9=h0~|&6s|ByU-B0#zgbogVI82-8U!*Zz(X3_kEzyZa71}tmfZhn+jJRI| zP&Z{F#!bCI7GDO_*X$tfzB!ZkEz0D^cQp~Za6BYBj-Yx-HCY8F)2a+5ND6FB&G9+# zut>yU<+TXSn}K)Mfw(X%ls|eD#l6@X?!9*!o|Ia`FJlZw-uA|?k<}#qtC;x(E)iMY z9bjFWV`crH+e2*XUxKEJ!5UK?K&9V8>$cUYRI3rKoE?$nD71}tmGnbMN#89lE zD&=i&WmOvM+1c!oJl1>!UuvPsM!Zp8B!RK)<>ne+#zJa8j9dW?yBAc7;FR;H7u+K_kH>WIQZ&tffLH|>t z8h>?W`*;U!Ut7seXk8b%H=YsIZm}11x?S`js*&Cb_4Vqt`Dh67^WH5JyV4R|tl5o!al$6Nz?=;~C`-LZODL*E5d)n1!1a1RjJ)?2qnux0 z=H|N?>vKfNCKJPZ?M<8>))K`^W7rnx{dtMk zSBCO}a&7J)WMpV8mPf#tdU8^$qc?|E;$&uAvFb^_R7qUK71n;Tf!)-T?!R3L+P^Ec-&K|O-`_~kh zjPYpP*mV_?tu?vQ5e;lG4#1326_}apPLYc)(Gf34R9OkT>%BlP+v5_Xgw@8 zi19;HjB96=@bPIRsrHVf&&p4z!S4$x+#Z1KA^UJ&@S!hMxJC*m+W6P{hg>56E|MM)^I*@J}(9g$eCB4{J~`Lyw?c}2G| zpLsSPj8`Dx`g|z-T!jV?b=0mF_NV4$wBnZwi+eZBdSrYt9X(k=CQ)4!|7<;WsOw?D z(Nc0QXr#TXTIlDb1xQJ;r?k+U%y(P{y-_`lbwg)km2v~^$d%%LGYzFiM%WBQm7IeA1cAOLx+&kCV1ad zCgA&iU7DZvl(pg}*ANN!66#WToQ*fFJgrEx6a6Sf(GaTh{juFG39Xf%q1fexvqdG8 z)OMMU_?@FCE0ke&!JhQ*d9zP7v;HAh7+f)N};-ylXq_KE2@5*3F_y)TgSUkuu}=)rMv1dog!$5Wey z0K+;%ek{?LzYI|4(|7bi_yjrJysd-1%Qo|GqlfdrpKmcy_B#C*U8W~?=jhjzizrRq z37JQS5qobH%$l^Id`Ag+lV*{N(2MkVm*Gq3YBpu)Q(9CWjZKkxEa3cmHfy{*lm@4> zie(pAZBq-AwI6~fZ*P+j8bsgMw4$y1Gt7Stv!-`aMNe#$l&_AbSNY5E9z*+n5z=h zs~BZLlAgbqufo3B}j zceQL4KR8<~RgW;o!N-UxgVasqt+#p2@6d&n9pc;M6?TdRJ`unve` z&fXUY+# zVnn@JM8!qTbm7Vpel~U#-(1**pF?8M`+hy1`3=CS-=);2R*m{?U%=+CyGC{bH)7|M z&J@1}(|<-c>8#Z{1bXJMcQ3zOThU zO#Xs>yX)xV{?9b4j|dsMgP<6q33b8eV5l(_LvR#dMR$-rRm{tV`tkNhLiTsvJB;Wn zXi&QhdF6(8?8LJTL>*z2>)gq1RJ4(y)pGcWHIcMq04}|VLd3;aI2q&2Gp!PMze!Ql zImd>+hz)7^wKn#JBB`n9BTYKfL%qkeU?cPo5f!R@v85glnbnV*KbNGd{}of?OomhA z_uy@lGkTwRLoV_FR1O@*t}*f0yGWM?o_b4G;Vzgz=QqaWOY@dzn!JC-Y3Nt9p~Xv@ zkG)n3pPY+`Yi>ZW#X2M(D1_GW6687z5SP<5anj;#V#(U2bPO3(dAx!|abM_@uMJKu zbA##9xn$DQ4?Eg35%sMA{GgEic_|0G;%DID*I?|(Ou|5izDVnsg}3jr@IQ*qJFMsT zjp7X{McRl4GD9EITs;P~l z3#;4dduSiLuJ&Y(l7DPLA1A6+T1WbJRditA0NiMLF7inl{9Kco*tgqBKOJ!)mu8$)lHoPQYP`a0 z06hqeqz!&%&_4@mi#|((4efy2ao9Vp0Nd;CVW>qjHB4=w%2Q1=b$<_xowpd~Cdbir zas-y>6~WlN0_qQbNUYQ!ux^@_WOL08Q#%AhM&gJGYELO{q72+4W%0Vch9_>z=Brd9 z_&?hy9I{kI{3b@P7b@~ILvQ0*mH4w21qgO(N7}8Q@cU|x9fRdCX2dg^B z%p0{+VrcW!i%f`_?BnSQ>L@)(C33lRb4DZi{VroZ-zu&AQik)gW>u~w{}LB#HMzUZ zH)ID_;2dhHYhD6fEA^loA?hyI@l=JQJfJ*)|H?3- z)qm9Ja_=$B=k`=gPMM5J!`9+W<@u3bQ&(Cgzg@)kueZd#!e%CcK`*(s9Rn5JeRNCV#K#~Rp!mF_tCVo=A=1e zD>Lh%~l0?q@H`Px|q_MLXkaf=poPKPM zr+zIYTQeJF`esN=ZbWRhq4eRMPJU_SW3G|xz~4R4biwF4sDu4qWT6mpk>-dT#CX?28qygpxM-n>RCU1q ziz>7)WC%4(8c(0g%xHuEWA<0yhLnnJ@p;2L`W4?qf^La*H}Z)qb+F|J8d>d+)mCb) z{YNhUJ(2PT8B@^NrF1=B30>cvpg|`rspj-=s;ym4jyv;M&8$^iBX=ZUcSw$R-`f?d zw94pp)OKR&8FZyyurN&x@U#3Xs(M$UQlkRT=Jn-pA#4NvN{? zNX?@>X~&+`q-K>zr6HeKWz|@w`r|z$UkO_;{zB&%Wn8!@^i#oDu~$l=(54v>OoSw<^Aiw}AsNZ>9;RTREc1NCD4XOmX^@4|Klw#j*Mf47JLG zSy41)lm{jqtq3O>|_z(4sgVkgETZOl5s zIx#}rz=cfi^%-()nSwE9UHSVZ2l1o-F_OleM%T-0G2q-;icip>-fw@?DJMUyny79F31uT}C`|5_he|LIC~Y$r;!f&-{_oj!^>>w;;vRydi64EKPa!?6FA4(S-H(T8=Ol-b1%BQHp(Y-B0b z+8rmq&F|R3ngFZs@0ujFJ71yK=qgMPn~&b+=KTESObq!nmtLG-C$(~okhaZqk+KV# z_^{;yLX0)}x1uuAn&wBHcW%?32^xHK@=URNxx%t+e@hbcvdLt}T=;)}KvR|#<_U9yK`-L}oJxUUD}ezG_;D~tv?y<(GB?_uv3)zOR};}CaD8QlC9&ApLJWA}?( z%ii(ueYTMn>j^&5vsjE2XXx)r#<(+CWO{a|lLDWLHLHy=+Hw!#MULd^oKR@Ve_=Y! z3YcE79z~Ts@ukfbNjV?z>Yp+S(p7QUZUrol7Epm@7{$MnhuV$T6z#SRi))qnmc=P} zeJ&7}LbC8w<{Evld}Q^^cr?wE_hR`lo~bZynahBbq+ z!A6gkFI`2xn=jM%f+0BcWf>w)hvL!GM-(jb*GkqcH0d>=78=;(mPKL(Kx3y%XsG)z z+Bemd=ET^OkJz&gdFsgSv>H&~Bk2^mWG;8wx14wN594Wr8&Re64y}Jap_9InukcG0 z_q4*4!ngER?jwyWlfeMr4w}&YEzQXfr)=H7n5t;OmuwiuUm%_PGEn+!V1RVRdljfe)sVv)2kgA4i(}eO$ST@}+Mw?|sZ?HS zE&6k`DsI!igFC5eRRPU$y+NtogYh6~3TDi{&8!_W*vW~Btk8kcwugV2@~^>^KT9x7 z^f=v{5=R-$-fY5znJoAGXcpAVo(?Pw5v=bF3N1WGYo43a{UNW|(y|X2GpQbv5^h3i zK{dXo%J9ag{#4_WASZY`JdE@y)x(y?jf$sl?)|CmcP*O{Awx~8zEVU*3pK4jKvxDEOOnUA zvi7Ctna&vv+?IV$AK6ws%20>;sJZA+4Z$0SxwPnuc&9SHlFstk(9BvZxbM@^cXoeF z9A$8}=bizW{F* zM$>y6u_t=-iZq&v*!K0=yvS?6e~DG`%ofN1ie$G zh%I}XIsGj?S99Qnxjp&U?<08C5hpYVFPfoO39OI1^KP0}JnU92u4mO?P5+6I z8)uEO@?Ee!JCUldhDuWXA4v`rlt=kNSUD25J5*uXwGYn`*{6^54pW0> z1%IZriD%t-i5;J{__26JKJ3{~bg5j5FZJ5OGhjkdLnT;#B7;`S_Ms&qLygIJE^-kKO)U-`N`tDwE zb6HFJs{bUF`>tB~m-eHs$%23Wt%?@AeWb5LC(>?dB7LsAiB+;Cd~8bu-+VEOJ6uc0 zhOBV7EDVI#`JeRm#3E=k<-lgI;MO(|W{G_b$xU3hDPIIzW^XptortAlDcxwF&SD(B z{*WF+fqYZksb-ftjU2U*{FQ!?Ev>}J&%@pU$9Q=w-!pBEP;2L`r zeP|c;6@O>n%&vUCjywL$azpTN8{E{aL&;-roY>e-sXv|3B6p9bwGD+2OG3$#OUuIp7gnrLaJ@Y?|=jybrv@5%JbSqtvd5PtrX1rg~3jQ@sg+7d$ zOSiiVuWQm{u$^sC*6GQ+Y`#rX&y))-Rwn})T^L`OVMUdI~3a%Ag!;hPk^Wf~4c;MTBkuR^HOf`U_b3$pVZU`L@ zNrcPdAWRUunYc@W{pK#t+g*C2Gvqe!9FWUBi`Me#{=L}*Q$zMu^$lZsz#p?bI-Pf( z`3H@m9m3JvO(urzc$SBC$F~a2grT)~D|D+pvGuABoQ|i`?FFkXWfBgIY}9*7XHRCc z9DT6;l?PCf5(UfFo^U+#09nHC5!XBtmmam@&q$$rc~278AROA`O7PPA1D4dz!2S=b z(77xKi+;pPSFPP4jd6^YMtJpRiuqgEkXz<7HP}(i%Gap9TMcVn+mny~*^S4{e}nd* zcXYWhm1O4ap+!flQ2OvTsxwpIj-EI+=?bv`TO2L{jcW0tv$=6(q@61nm(*glxFT0# zw{H&B+ij=J?$H#regt~f>maq?eUhJ4$R0FXv9$1WEM}q5B;zMj&|595uv|B)dsbpK zSH?wh_PazHnxQHEI^!)L9H`D)Qhwvd+B1Sv=#MGO6Hx1u2**j2=zjV(Hev8sD&q1y zyikFE4z5DA=P(RJcLa-Uo|l0X@{TgNG_{^iKZz230sY8mk3DI2%cOhZ`gmh^5cb7M zcziz$<13aRasGWwyV`-6Jy)R=+7~;s^zrMr3X|I4~$(_d(b$O`7m}}S1ss1aGHAj30W6qk02c5W=?S;+asI84lEFUT*1UXn1#P< zmcZzuIH%4%Nm)&D@V*y}j1r-T&fJZopXN*Xw@~RtqixbDB@by#L69Wd*9d*G4r9Pp z1Mat|7ysRRDLYrwuN=jzDFXv{ zuPVX4vKo&+`l0OMTrsL{i-((P9Nwe^!8AFNlJ<)8g|#P=45y(=HU_Ig*Fv^b?5%Ts z(KX)&eCbKeF4YA_8)+`Y{sKuy?_>2sv+5C6iXjI5p;P$%= z*g7iVKea7&hWDBs!Bbv);@coioDw}T8@^0NT-17; z{FaFm@71}w{A51;{u=)9S|)nRoWY*QX?S5)f}6+xL*Y~zZY&=}Rsm+zdRU!IXQk0c z`&Sh9TAa`A%y2nyD_QCdWu`sS_!QY=d@Tz2!AW8G;g*I;M=C`}&uQ|UTtqF8yJG$I z6!vVnGa2{$KnLz8qE~A!rbiz{!sW_a?vb=XdxYS}z zjI>2NOnQBT*oo?NZ`XPrh6#PX&x7OR&j2fo=@sdyR8=gdnbwX&||xf zSCYbkzUbR^Ch}yq(=Klvks}fNTWdo&zZi#!d?Rk3*ep?R?ZWg=s8aZ=ldu}R9Fw1k zc%8(M-p!jp_iQWZ$dkF4{qX->KjC?8E2A-Ex3Rd=scg0HUb1~|OrB@nGpoS8DE^WI zZNF@2zpR&7u+i-4Yz?-fk0Q5=mhjYZ6SzUv2D+8BoobT)l7Yq#8ffE6KPs|m>p^|! zKRb&0fB(>WISO~>%f;U8IZpNsK>El4Sl?bD`hn_cYd2N;;5vZjx89}CRd=blyqGrq zZh)@PS=Ky?M$w4hl%sqQGQs%>`ytCGkNb`V7wVAhoJlslo0vn69t~9$&yr0#k|c4+ zT^7t{mCll^n8~O|mp*Kw^#FRK5Km9Ol+nMnQbdaVUQ?H8+`nWbAL3%cD}Vn-`@c$1 z)|3b9lZhhVdlySLKseMvO$*&`}{XMrwn$0K{iQRh(>{qh_p#g|Yydx$oXQPs&vOmUfBbU-v2<u#W+naHBO#}Z(MZ{oPJ2F= z(v#pCny8jU7QXM;U_>qL#k){!8Yzj#zd+6*>Nb&GKe1 zF6>W&vFc{XW*0+d>OpAT90ToS@vYWfiu>KZ)8M2K>VD@HeYiA;PA|%$)gu3(x+I8# z=t8x7JEfc%jc3b5KB?$2B|lyW#hLknU9uXtCgfmeTwngH*A?Up zm=C{lk%u}s4IPUL>D-|wq@mqJ?PsphK$U|it{%wqx)UTyE7AR|4(x5Vfw3KEQ~v>* zi(dFEpMq&O_hO048J2$H8Xe!b5x;s3A^(PLff@~jd>EI z*jKE-symNfW5S0ne}=N#U#WNhAY2xF8LwGWNM3yg?LTsb?7mFGf(=@Ta*v{3;jih+ zNpaszPvdFLA@r_|K;@S@H0&yXUfd%LJ*~jU?bf9E2Knr7qAR6Jw5ao4J=3>6!3qU) zcued;TH0}k7D|?5>#kXbv&ygosad*=_!_J70Dx?e*2nOrE(--krOY8hSo7pC^w+-%EE{wJ=KKmXXq zZ5E7VtNMz3cQ;V`im|oG|(Uk2m3tktJkn%)P{Qoyo{gxSI8+(aHwoeo6 z>zR18Gl!i&mCds4LuqHEEj?^fK~0PeZrN2Z)j?Sl*e-|1{2wM;y}{u1#kikbLXjd1 z@=$2osW-#1ewyfa^8HQU($}C@Xc5J)lq1E+9=LW+?Bln`Qq|bKP^mm9*s6($-+768 z2oKPB(*O+8+D}~;u0sD0r6{_fLG59B)Nh>Vd}uo+Iwxc?^q$DwPZ^A^Tb5x^a~#Z% zu0yEjO$^zs!xaL%VomfE>@(YrMXMH}&xA?nZDlR=+g1vZDI&!u2Plcla!NMgf$nSh zA& z2g~RPTvs?n*G-0?b-oQ2+goAHrOEKo+KYknDslbP5?quC!0hxD*nPf^=AGC^I~I%I z>aa^l2)qVKl??B?ce?bsi@&rbAwp_?q?9ynMpCBt7utQ$nBVze$RmPedEqf7u0DJc z-~20pw+j#A;nozao4yIjeR`0GLJS*pMvqDV^y0CeOL(J9EI%mrYy*}g(BuR6>5kDR zs`U+Hn>}_(G+p9RAif{nl~s7wv~qIUAVH_!PN+%^Mmjq6wA!}si9})hR4i49g@c%B zA9oKy_XFboh)#yXAA_aREAGm#JwdSu{Vjqoak-owfs#@KU(qi5L!EMAa+5bg71 zeQG?}nC@j+??1B{@xhdsHiES&4`iX^zp`QG=51&Zzrr!UMblBy-{B~3TZNCxH{1b z(_f^ZSg?Lp__VRv`7dd(+E!Sr)<}dzN%k+HkD(| zBnGYBOK-Q!V(U3wv^S;mx4l;Jgl!JIdi`qDJ>G(ISAr0w+?U(+et~`J=g>P$Xo&q!XT-D-8_L(a$ztIL^5i#h&evrn9-nBhsdQ*sx>Ow+Mn<2FnE?>`#t*&iEbS4%X8Pn1~CvS2YW z3y@N@49Uws6I_X`byeX-%M` zG%RvFsxlR@I{P)9_-u%b=WbXf&Ul+{sPS#r&S3UodsIi+Me!({MzGOeG zFY1Qhp>ou(+L-iqj)#@D2R@98MUw0c3^BFF`LnYTRAPpWAAec>>&TSswm3?MdPOk5 zQ{9+%M=1+DzX;Y>G%;xWUwXMDlDupDN%MpuT^d*o=cBvPVB11RYkN!QT%9TnIkG_N zF)o`HuFR)(y2J2Ul7>Zsl~g|09Lv?>u{5{{U9+O_tvVOuZjQi`kVM+f_v1`+I&SSs z#eDrS?10T^nleifeh1XBPRvN-CYp+z!#5fj&;!yoU2Jx@=4&(u@#tguP-^Z0t%3b$ z^{Y%asBkt$?hx8 zW{<1C{V3^23eJnO$G!2FvC1%)9%}Wc*mvRd-pzzYn~o-F;|zK+%>f_WPh**$5kDxM z#e>WodG2&4KL3s(zYrkLpN5=9U7y2PT_(kY56R>+aR5E;f0Fr4Va)a&afQ~Qysch` zZ?@@*TR-yYbm@5V?z%|up=@#Tg*$fZ#`0X}<-FT)37`5q2knmOsMgtue_~%AWb%^! zm}tPLNf&Qj4~u@Hb)>c41l^q{W4`cF-)YWZKV-g0qIP^?vu9iI+c87At?+(Mx&49K z#)~T3r{WrZ?j?E{G@!mQf-c!c;&nkFR7-~8UC&SqUT_Ku zKS?%P_9mI zx1??8ws9L~uGu6yP+Sp|vK?>6++f#7O=rDjm$6^UCfr2Xp8I_VpDS4VzA`7+yrVvh z3(r=UNrImrIucE;39yK|0UmrB1BSd|D-Va$-@DIgq}p9tdHXiIegEjl(25sSQk_V# zyKKm)dp5ltoQU^9uOa{V7ymItRjO#~Aiec!htPTQ;O%Q`l+;L&sk*1J-4c=u0!;>+_ZE zdY(wWidQJ4jbV$nEBbWIfZuc#?mXo=lzL|4R&+P9+t%k5y-fJ_v#nItIf(B4wxsAM zs*<*VohV#!w40tG>x zl)Y>qrtv+HPxD1~P7bmbmg4NFD%^M{$7c>s!}_0Ah*BE|#ZUfXr<_5ncCM$O;zTif z*kQJrc&5$g@E^+@_{_c=xQml1KPP&Vj@Uc#wk>UR%Vi|At36Q7WUx3clpNh+*p{ba zcX?4C$wC`7T_Hu5)fd_;mBY1cJ9u|BMq+ne_=Xx{o5y@~*KokFTQ)T2U1gqwO$4hB>0!)*A~n=fl*qj<4-(zjtG$K4em|Z1=}eAG z-_p97!XM|IO201UN(9$X;yCXb+78R`v|vR(UgZn|lalbnE*65MCw*1hPdZaFL~6Xu z1*3+TqmSw|yfZY%sQe<5mpegm@9(2#Vk4#)d?nH#xTmahS9DqaK3tp>R0qabkt_#wwu$Mq>VJq z@;%KLT^X@eqNnbX2__Y03GVqXlJ?R@uC*iP-nW9w-)Y$T(1^cOcE>3592Tk@#UBq1 z;k7Ai_-wOEM7qA9W&MWY#W{J(4w^xl30+{dIh7T~%G0?E;@!L3A2a1kL5n}as=kCJ z7h2Mkb>^^HJP@aCKEQXGlt)=xaG$B8d9>~{-q9%bHm_=+^`{oUr$*9U{T*cg!kt|6 zbLfu50NaPe3T@bnr7TuN^}=bQvv4e}ElNjxq&1(JFUJ+H4Cl$N3wheZ3@Um$lU50q z74vGN#sQwtlRt&d(((Mu(tX`M~*rwp2M>#&6iNwkF z49s#BOt3f0SnFY;#Z`8c)o#ZRb|pS-rab?yQ$rUYgkViZ9uBwfBkv4ZxV;{ZMfWb# zgRe%I*r|;t+OwrI?Bb*chv!P`SM^7`PCrO8BMQBVE9Xk&uxAY9gai7##%{GD$*rVWkZ zfq&-kjR}+Zs99^J!}mu^2Yc?2KCOGjRyYUJ@4UTq`K1DU{!DIg22TovU*x#(t68*ALc9#yQdPh# zx5SZsK_`ZeQRX8KtMkfrYP4$i0UGtI8(!RZmIOFHXL_@{V2h-P)E}?Lu&grC`_c_z z&YQ7kObqt6cjp!ny?N5t{@i+4JkvRpN^(Iecyrg3jVuUfyA}+iN6BfDD^_jnblhL+ zI!YUAQ*CjtdkQX9A=LBMGxoYDiACl= z!=MDlEe6iuXBN!G-kSZWKdH~%6Gu_yu6U}Hb)mev@o@XK1r0|d5qfJsIfQ(rk9~|q zui*uWUk^9tY&D9GY)-^I2QklY+$-tYTMNDCFQ(t)FG@5o_Mv$a8(JheK>X<~>|3`Q zsh$04$~ix=dl&+vTq`lF#?a75hv})E0yF}@AY4y}`(#+~0F^raC`DCzPEKE1->eSV z2N%g#p)0<$=i&QMCI0x@KweN`M&k~dQlQ&mDtnp$$D|m1d@j!RhT72Y77w#6f(dZA z7IU`s;Wc-_RkEJ3_L7}U?(KI;zl*nM;omtltyzVv-mgXL#w@{Msl&_o!_@0!Iho%) zNB>crWNFh(Hb}pNowGH<<{qK2Z_Y%umJILxWDPPz|JixV0n(?F^`*ml^p+-q8m&JB+iR_9*?&%C7#t$GbgY&24PHnyZ=^|FEQhkO!pk;r zawa`G`;nYh*`fNnElF>?P|L`0a`b&c|J0O_CV1NoYP%@iK9{+Kda#3Mo{{bt(PQ%0 z9MuC=p}SagAq3acm%Ip)i}Rr9*YafiVGHR+j!Kg|;UC9emGoHZKqa=T}k zWTB6DDHmynN;hOgYx8u|kIy(H$_+5J>H^tmrpG}%$*Kj<#Ig=p>o_*Xu#3r>YRpSm-b3BA3^h)y;!$tJD6k90PH_6i^)YF>F7;Q zD*qZSxtMIt{_|aJ_utRj=R^4^s=PcUr{bV`L47)=T zu_x0J`PdsOz3{^ClhAA@qG)Osf7=wvr_c4}F`xgVTk|_<#d>A<9QclfKbztFrVthgI_~p*`jzC_SE{91|%5c7dCvy_?6Wt9`_IjW;Nhk;C=K=c#15INk~thiX4~82yDn_W6$YR zc9+Yf-Xy~h#pgq-zd4>c`trIb>-i|HG`{jjPjYv-$kt68!5shlNuOulq^%7z=&p<4 zq1Y;uef>>lxp)_%87kbotxWw^Sven_55Ub39yYw(;Hw{rU zq7O!YsXS{lJ^2(#B@Vl(&xti;^kx8E*}5CQd(XzhAH(r;^Z_U-?n39JK|;q#mtHv$ zB7IP2FFk2C3p1Q;p+0RoTwe#l;jf9nagfWjge`@mcp~j1AsWB%7=6*wwsj!Qr z$=fL_ER=e@y}>-zWRrXKVLE>JE9E6-pl@9`D*A7Sn*4G+ke`A-!vB@>q?$k5u#^wX z`G-ACw{fF13$sFIqvgwb+%m9(&y9YNnRu5Tr=^noz1Ji+usghZoTP;Ov81Lv6JuFC zK9>pqUFThz?XZWwo>M1oW{;1uws4jx!0}uoRr7KxdoHe9p(2L&Os6QRKY59o!=Q=P zZK6CHS1fQAllI9MvdWp)_K zFHaSE!3TO?r^MZr^07S58!r>1ap1m_U|#-5V_ijd(`X-c3XO00r~-6f?SZG;TF5(1 znV(N-fr&w^@RzIe%z+ww*uOq}n0W$TUh)up0}Y&9w2bbq>P0o@r&4fH0=~Wy*R9)m z6bzd{V|CZls`{6-^;|SIrI%rSc{{$Y?8)`#_T{~rCUXCW#B06`n%kN7t;pyfG>6<>WLSUcD8X=(?WK91x4c8Bn@%Yp6zP^J1{`3fOKMK({o6cgC|XT6Fxdd%{oLxV0DICIahS z@dNi|uHwcLH7NeOLPN?Uu->W$8^gQcu#Vv71`HK_B`qX#sULKLEf558&QfS(S4%ER zl$?g+R&5~8?V63~e`b8z{X8DqQ%<^7-V!Q`@2F619DOt$$-in$B6KF-3wRezL{cM$r% z60={WV6=Wx!~1{DEZwPG_zvb`xP2lXc$DDCjaPWGe=y!9SmUOzFV+|?BU8=Ew6Ek2 z`S(|W-mVYSPqTv*OC}?8jv1Hjk;Ao{I*_=v6~84uaQ_pE&Rx?GDwyLP7bEeitP``( zFW?`yUc}x3?@<%ihg(gRLUCaO{k`5!<;O=s`}gqNgF29iTvwz5>tUI>M7>snS!HD*oKric6$jnQYjj7P1fTbbG#506c zPR0)ZRLp62;qC`E@c5XyyzcZlDh|`9wD-C6ud0)-scwUx(3RfW9wS-($zAMD5;Z}RDgqW?A`QbRMV^&D_>-kELM9h;KJikwB1N(^_ ziveUahx3nrt$B`4f9_O&jaScz=YP)kB8uCX3%zJG{)0|)R+b}HO(IG1h-^pi@*BuI4vBgi$;7qJEk z+-1yFXo=r~)$7A?(fBhB^*Ri{KKrmdsEMteWF(m$Va{a6otLzTKhTkRM^UpR8Ugx?@TIbV_UL?}gQ6>Ntq{<@ zUCGCbP+bfh`3zazc&nes|BOH*c$~^f6=mIDi2xv7vJkY3mx+w z2G*^I#B>`RdlqBD$7>kh5`(xqdr2=yN}DuJ(O`uUc%IIbz@=0sn3I9nC8~!+F$hG>YD=q&JLqPEaPj zf8k7$?#TXnrLnaO!i2v)01hMe;o=^S!_A@>$v|`j?cGghJ}rdOb3^V^zm+uaBvD+r z6TVrG#q^M0bbs&&s#=_mqZ!J4-lqg!G%{55J}yRg^EGg8d{uoxhD2I~aNG!Dz*^gJPw5=nZn>PpXCx0gJMXiR^ z{B;`{nRwuve6r{R^~cJ%4I-NtPk942P_JG+&^Wdyd5_bh3j?HdGwLwuCMv`0hZ}Bw z*JRT|r?3v+Zd5zxE-f{)p{f!eR%o*fw@-6CIP;oLKC(s14=b@cZ$rIaFD`F(kcSUG z#tl89d8+&+zUP{ev_WV$hIa=cv~~$}o#$h(`dW;aav;8sI-zB=(8{AZkpVp4CKDKZJk!#@e$ zV<~JNmBR1rX&5@b$JuiiVZHVeqUQ|cZ50MwuJ%3DWt(yAQzN<)?mSeJ-~XsC0$tbWfKCQA{xajI3zYcYe?O6xtigRQ3NL7@6^bu>$Gck@ z=xXE-;}}NI4~H%%pw zRCxb@cTmboK~T2%UdG<174D@twMrLoZx{35AFMz*lJk=>pluUZeig%e4By75wOqmH z2`5qVv@2f4UZsIG&uNEh1bws_NjFUj+3JTE+2!M**wv|s>mf(T?5!J3u3bXOv#Loy zsg9O4>?675CG=ra7QU6w#L+pgY5k-HBpDe@ucAX~PlF~}`f4NWt{N)E9&gmE-IVK8 zLwiq){rey}k=MD7B*iX#X5?#}{Fj5*mhMukhohx_ox`Qk54Yh(k3_5;_ldd}%w|{T z$Wze3H2PQZgrKkE&n*a7^CrwZ>Fd;QeNG6b3s}W72T?GifaT#rorRJ4eTfNf>3hl9q-w zF#AU_6x`9o8ZJfB`_@=Vw>)L`-qe&PPf(M-{VgkHW=;ItK5J-KIbl=P3eoL11ih;o z>C5<9%2#`gfi_8~j4h$6)<@Lv=oWPz`$s+^@scUHKFiai(3*abRR8m&d$H-Fr*|LC zu$|9tXtcAEo^O!2;wBcE6yji2IUP~n#J(>5z#^3@;IO3{Z9x~IS-yjfUX{!C|M|-b zUk#vfA+8khJBo}>ZlvYcr<2T3Tk>mN2>}7b@wKziAUHDKE)zvh)fBAtIEIKprsy(W z+_Q~`A$^++{m~()+nov7$>UIR@*e3n+4IFeRk_R!ZN5zm19s=t)n*Zj=4< zYbf1Oh`CQ>_?$ikMENCTEO^~YfBs|V)sjj3bWb#lv!M&?PqDP1a55`eU| zc3u0SWaBJ#_Uy_Y=IRl_+dNlrapdGz0y}6^*TbThE1#5JRgiI~FCBgAg(qcisf*GK z-Gc(x!yhWN#N9c;7J=u*4pld$a^rvPYY1{QcmqWkl;@5GMeKCXF&2lIx z{1Th6_!et)dBk+?9K?d44RAFcij^Wepj9#-l}jH{`lLkM@Qa4y-4k$99>e7D{8}Xn;wte_MP!T|#h%$s z8P3lIW<)8z zdT+uhpDUOyUStV012MyZ$`@r1-^5a zEXRr*F-O^>BW^SfI2fT#zB|5OTp)Ps|LBOb1<6+xvk2Q$v7VCi`lc}p!sqaZt5OJ%+38EeGt_Z;Z3UH|nMB@I%Jefq z@b(VXqVV!Co_W|A@(E!$ar`dE&Fu=SXMN~`?KMhqwHG}FivQ#2yu)(--Z@PwzB+1N*WR#U%viY6s_kY*b)$?3^ zKIh!``~7<1t6eIYb6f1)qyvk2*%T}GVSQCo(78jFo9kpS?*(eKLa>iby(U0fc|~|I zy7C{Z1!Kl_Kh03CBENAh&@-+?)!;8={A>f>3qFlo!&?g8tV2T{JY|I+R7F^d@X4-7hPU zF1pLh{H|hFYYaEjZR43q^3sfl8hq~4P+S;ric)ulpz>D~whxZP-Q%gWdh1q7Kc&V7 zy`F$$l2u}E-yhQ!{iP)1`Pe&lHC`(HLMF4rzZ=up58Wg<{z-w0^bIXDi$KK6gLvms zgh}Ivb6GtN{!Bjtj?Wju-E9+g9?iniuE!~6bx$}9o{mWkKS}v!Kh&HN%=Z;{xT(QN zY3I#2>Gql#NZryw=}j{9;-Cy~_Wq1k&3ABd_vY{B28d|=T0SyB3F&7mNd4yux-!y) zT{$$6#)SW($%7Z+*R?oYS2_l3g&@i({=o_s^(VPu17Lbqj>__XN!FEYpiMo#l2mml zO4S4tpWh&}r+=tu$tFqe%vQ(Bk3r1x`&2&VyEeBI=Y=WZ{kg}jjVS9KO-@@kNIQKN zO0{>|N)P>=gHeukFltfede2m;(`GlxS1MtnMF&&_qc`b|DXn{uK_43>xDlL+p^qn0 zlD027313>_oiMt$#R?Vq?X>dP2AmZ+;=M*@Sax)gJzJ_GnfEG}RV+(|%PeEzjT1ia zc_s+AI*ZzsTKwDX4b;!j7I()-W8sNWq(7~e?r#=6Ka;U|rIUb{*+tMVOh;$MF-(Zc zMwxj8WlXaay{K(i_TM$?S(}D}UrOB0Y#KXuHIWQ8`an~>KfI^+rJu$2v_QYcQ8L+x zUNtk+t$a#fKg_@njZzc`&8F-xqQe@v3I~1m)9K^}I_hqOTMLZvA)*V`ZwsZ7zD9h3 zU>_E%&*FYYa@f{hLsTt~zh%+l-XL-;88=~lxs{44`(S6}Uf3-d439Y-)btWK=G_xc zPlrRM&s_W+eT^MiV^8Y0^Jz@$6xu)h9y{i+l_?ura`PHn!PYh6AB=yacY_@7G2|QG ziL?I3NddTXA_dE5>e9oM2wJH8iSEqO!_6lu)aWputtbiN$Nx^}!@X3++*8VPe_8WO z{Y-fO4RbJ3rY|C0vq{zWKD$Ii$?4xY3as7CUVZDyl)JSv-Mu66&ch3u_jNGGT6lHr z594t45_~)mOGzh+B|Ao+5j@94^xD4&rEly|-v1jWcs#_@gLUFLGYXccp3uMTo@Dax zF3IgYOx5kCP!k@*)NPjRyT%5}jG7@0cN;Br9x_y_veXiC*9u5${sTJStiX5ZEAhm{ zKj^VfUpl&8O)5Ru$ZtRU#=eMIItzBdt;z;=@!Tusn65(eyGrnKRbR{-HX5cTBD)tY zI?1ahQ;XU*axHpB`3k4#>xLNexwV%H_b4K%%#SMK*3lfjOzO@7;D6}_5cU&%|eKP}(TM(0jFrfHMb;u_6lFNZ#13x1{09XAgctzIp< zbH@ZXF`ijxUX%PcsFvwBM#6qvI%?W4(%b)D(T?uZF)lg;(hLWzNmz^e_DKk}{Y&j- zU2yp7Z7MrC7+?Fv(2r9t>_p3FM6Q(OR<6xx8fJpC=eI!fPbH?-tMDC>mVDkNcOJ1( z%=cczL&ql{Y2OM+{`5Z9T$jX-F0rJ*v5oX9EggfiHo)_qO3VNM$j^b#laKYOHfyQ`p3v7b=7RFwkV;OH$@prXUo#DAULIu6^-pt_|EqSmMco&(t-kg}!v;Qc;&k zwDuQ$mDYaf{9wkP2Ice7T2b7{ZU(>U>d0H`4XEw8r#M$8V%d=1Fi-6d7h^rVF03Y} zEKSH&Y{N(kYqol@4JjWUfV-+6up{;=q-D2JUzd*&$9AAGCKfGW*%Gr~1 z@BAu&akUp7`n%v7*5UA$-I)912D+(~(GcMlIOi3Mm0cZC>7oujk`*~OH{412_8L#KjlY@B8opZcHOH119<{_1i zjw7e}v+>97GNwcdztr2!v^3a+oW>~gL5m*XQB4KrCf7RrtlL6L7Fke9?21L^FR4Jc z5AH5k7CwzeTD|%wMGV?SK?jX6rqo#UBz8D@PuMC^&M0MiacR)~AYAt6WVqY)L^k8v zD2fm5iPfR29edZMN@lNJO3P(8OZ;P9S!bLK?bi1fc`;+OZQF`CJ%4Od3dLdl&G=pI zF4^XAmpP35LTl|uJKkFHz|r}t4@uK=Sl;u2wD7@4dOX$|?J7pl4$_6S+84)~u<`7* zbt)ZrlELx>d@I&)BVBnRN5>4~S=8C1tfe7>`b#HLpV~w!TQU}1u4%w2xQWu#b?BPj zEb0}t4E|h!ZRtCdsckKzfYk}~Ph`uE$_p=vk~TBD;l)l~6|)(^M7mm$g+;-?5kEy8 zs=b3yw>JboOk!wmUMpqv7=>7?G{;6u4R%{FiuAAO^QQ6zT={DTSHAFzc09@v{4Q0# zPBDs1H(zO$%g=SM0vB$QHi zbvQrcyqRCiir{mMW>fIR4qD#ekF`NB*lU{{8l}?@YB4t@e>cZTK9r^~^ZALab9NUx ze_#N`MTXPtjw|%5`ZGyO;#hp@PKos|H2$zJfz@zLuj_Bwnu zx0^GRTi-c?hnDHQ=Y~=I#@FLmrD-VbXYVOprs*O58d1qccy3^pe=1qILk0V_Jf0k% z%ERE_BJ5qi2lt$!vE|tpI#lkAmC;L~(5gYNt zgFe&G7I>GDm*{&&9JV^B^TzRoOnX@_eVMly zP5v(^|NaozewYfqn_%lbJef-3Up8^`5bRRgNFSnnQNOVj1q-~$E5w@~ zEVx9reI8)T;0_oCwIE0E>^gk{F}h9}cb=NzPsudwcF9MBa2SlP77XAFQ?%^aDcqc^ z#hfWnqFv^h&TIplKb3qRU#ExirfvSDoDSL+KZdTK{y|5A2I5|IZ#%aUDf9|~<x3u&_s!CY$#&OQ>^_<@^QN6rK~<a!~UM4+;d8Bf1&}uTB0YDPm^n-9C97HN_Hy!ao9C0 zm91->O_@JLZz$S@MsxV1409PO9OYi zORMK?l(xw|rR1&zw|i-b*`I{v`3YE9C~|>sGpJPB6|eV9#;#$hh?*g0ms52w@O4!pJA0`_eM%Eg4>CzZAyb4st?-z=^TbMd83h2bW zhq}=BD5NZ%^JLgx33VQX9Sy#Cf)w;Mj==iI5}e$`>D9}DWVlC>9tMx#HkWnyccTuJ zxW}S1H4qQ=NqC!=3CGqp{Oi9O$A0X@;h6vMJwBW=cV|*|!ajO+NC&+Stbl`V z8)TL^^MVh~{JH5cE+ao3ec#Q1*ARES>ZZe0PU-N6``VD$Hj9jkp0c;=hEe?Ho4j*X z1)nBe%_q*80!dgPwx%Kd5`IIT*{CgpHR66=S^W_y*Kd*v!6;+M_PA(KEZ^vxWi+U*077Be7)n_BEXyn__-6X@)(8Q=I*rO_F^{KX{{+R3d5?UvC!`qTNtb0RDdeRi zo{CQQi>b5G=I)Gp4}xLxEEsEsOqN=vnMl{p{>i(njD>#bRZNZQ!T02rq4+x&+?|bR z^8;UJH-LW$w&TwRCX;^ui_~gmjJeAOK&`lrP93PDeMc1d@Y1LF9Cim;w&@}tyqVUm z?oIW5KT>980bTXIK<2ueu|s??S>Ycj~M*9Pjlh||lnMU<|f zfNy=Z`Q2I5sH%%P*R+|+!yF&;7vsM37tKD>ThA2vF2l1}qZ9@?)yw#>=_p*DgrHYT zCH;sj0qf_&O&t%C;mvf`#mJDU_9(_B&23OIQiW2x8p<29$)+KdGWNLPgr6GzYj35E z-JRhuQx;R3a>%K~5Ci4RQQg)656yH`3Qo+@8%B7Sx(CyT{-SG#Tv_S+-GZrPOE#6t z)Y#)LE%a!kCx?}=cJTlN$$HS4n6B_U9g3~5f7ARH9s0RpI=8JEz)xjX!tB)~7}k!& zgzGZcpQwryMIS^BUWjAMt@#X9EuQqT9{mQH^Ce4#BXaZ&l*@bv-+T>o*KWp&zoM76 z;uYOYcr7?STUpZ07ZyEt>I7~%*nkhd zdI3*5uR!}!6VyIsihGZ*$g|CW%*S~ce)2gfdgV}|+9*UcszYz^R9u$YLo*_dzW%qJ z{d>KTf3_dYO)h;#?usOSaMvats2aklwFxm7&Y{Wo9OivuxO>?KQ=iBp|CG=^>^hCE zqpMN)IF~;h+Ly0S+J;Zc3wX#{39qyez0UMj8gaCS$}A^Rf~GDlpIOZo7$2dLmqe9w zodT;XsAmNQ>rhY>jh%a4;H7K=#0#HzcOM+PZHWw5b9UhKC&$C#YB=)d3i0rxrg^=DZS9}L zx-dJ**9`-?QQta*hNlYl`(e62<^?GaRY$0c3ROP4N+~tBsC!2gat8jSfm$!u}6d+cSmUY?@5Yez`?_bIye*&p8ZHZ=DYXXh4rAmw8eJt$8WZk~HQDex5U+NFwz z1$AJ@jC+WkatY^Kv*DDJ1!cjG96QzqTLUezYmYUC&;Cc(ayHS=UUGCpu+Q$93%1Cw z5*$9AOp9hFvJpc!kk<5IXrwE{GP#&$`1Z$!yGgX5(wsipyMY#Hz@%OWlCQ!ZFQ#4v@GN=5kTU3^2w$MP;~bT1G)X$FohxQt`~t18O8(V zXz+#)S@_d8fIn)S%qx0Z@Q4Y?sB~V7DCJR@*YuZdzd2i?Tvm%T-jOLl6oZ5%89m&YQ+=rN1b7{Cr$bg(V!H z2^MM(D6jiP9vh!f->)3SqI)O1GM2gY>4|HVuSlysi@aAvQ%6D;W7*XXRnw9v!KDXX zN*~BV`yL?KBck(`_?8V+48xS7FQh3OK?`!6u%*uwyjV94i>gZS!k3Gg{zwXZ^_%9F z#?qTSZMswKf%;NqY&5mO`|o=&B4r&e-nGWQ&IlgRE0Wh5uH}U-Pw-*qO(f-9MA{d@ zF-n$_%F`VpYw?`D-!hrC3AfaUqDfpfX$<$W_T+nxm_lKf=twzntWe6rIdRwcD6*gh z{@u8p8{^@M-aPxf8C?HOMa|r`7<||f#@<6vrDY36M@|;UhmmvrJNn_-OzIUK6cYW5 z`m`Im2{zctFNq= z)k;#fIxQGnnRsNh4fiKpgx%UGI4p=ngPJFs|0O|^@T^gy>*a!=;a2dI7{WE|Cb{m8 zCTAZ<>auYpS_bvTya^d3P?p%Bnh3q`Z)o&wZMIy-gt<6b=6yk$4f-O4I9vAFWXa z=adfMlgv3JPkWBpSFY2Y^eQ@1eU!SJ&8GZ-CGWu-%7`K_Q7uj!Jmpg%f9NmlAQ-*)BBF1 z?Dzj@M_~swm_4A!pVR4%*b^9;-KW__qJy)w3%b9zK#0vwgf1<@{++|2A);y?y?en| zO=Nf5EFoQD2nB^6NM3D*hrR}gcq%#p-^WwCxjL03oni+LZ3ptaQR8@zCM?J#{iQX` z*LfNi47(qgRQJWS5>MdS9;}1^foSaWLV2{c2DApdb9cH zuTu0~(m)kgH&DaH-dtx)0ix&kLx`s?uhlo@Hv3HY{YVB2PkSoqDMzi3mI{800sbw$ zL>j>{$SjiL!CV#iT-}W8N6j#L$P@ZxoWvdetm8b%n)g%N#^+uLil;HH zEbmTdeGk$|!OfL5c_lKmGJL=WPxSnKgKA2baf5&6eA|srs8byWhh9~r{b3WGNZv=o zeLZN6d9H3d2TC?j1u>sp-D?ofy{f67H_+EC5Rbqv|pLg5>i*@=NkLHKj{A5Ish zVMX8W&=G9;5|zye_!kO^Z46F_doJX9r}bEMKiee9*9KIll_eADHur8^%j?i(k%F2lKjQ}O7k9+Eb$p$%D| znNhR?&8l(2+}+wxn0T2sclcncd_P#Jnd7a8wjg$?Lr%UwT7pGasU(ZJeb%K4;qJSt zxfXUFvrv7-34ddVG+$|Y8CR$R7NLjm zZ09nVv|ppqg@OUw(T5M%F^!KqAI*OzcI6a$4^g^`oO^B)?pEPxJXc6HV&B#?(h~c+ zm5}MTqcpG1n#^y5eGSp%HxB0@O6D*=-a1T0A7UvZ;x8L_${cIW$6%mzH_p8u1&zw_ zu;@DvYUv92H%%M+5*NTkHI%xo>w&XB?C~w6Typc`9f^(ibJi>CA)66pO^BXtzywB4cUgvZ@f?*EJ5d=K1ir+q184=DQsFXjlHFeVP3|#o<0`K z|!3!-m8L6(g9K^E1+7}YwWqgMRrg0 z+V0f7WApN^ki6;u;qab=0>NYWXYv94jfY{w$3l8K`#lYtdxrHZxIin!tiR484*EM^ zAZzMK-tY8wx;NSgTXbzvHFOp$9Dl&EGw&>O+1Zb+EPEg+Z+9k>xBy=4HjgWvKEj{- zmZE*%1~_!AgllFJY{E`qj`Sq1ZQX*zUso|@_;Wai?x5hTHmdS4!LF0OtTkpnQ!eOA zGv;pWJ|C#JG4-G;IM)Vr|n)eThg-wp~W`$gaPxzo!mE!rFHN!OmWN#?$E zqN>kt=~?+$>iy+4JG0J$E(V(MW&aK3hIeCD;QK{m>NRc` zT^}y)TeI@mlg>Ey%lQGz_xk2&xh7q*Vn$CwY9-A=Ke$*rBX-1Z(it{aI0w%{QOjCV zJ;;yE+LTIrmsilNNDo?_dVvZoHcEWW%Oy`|Xp_@uC5)QcO8rXGDDuQPx~{2Bo{z4u zf*=VN73$$Z;r|h<5a^De8b7Uqmv(l|p9{7fOZ&QT2Oa-4W@6W>` zo}(q1fa6N^O;SKt3X?GO(tpfeu7EyeUdHckT3mLUJ3sa`i$-*ae8oyz(c^W4^20Ic zdVUc+yeHD*$@%1eN6bpS#5p^mfj<4}joHHa({n%z3pgMVuA@k_i@o#~!)7eYH0N(i z`~@39i5Z6%VEwxzlvkWASvbstmbh-Ei|>~p;wlg;(;Y6Y!jqxDo7c4yzt%qnTPn&V zHJ$!6sM}LY5WI%G4Y@{#Dt zh|a9YlPb!%qj+Wmq^$#`{}wm#BE3I6H?kIs-isXj0At)=c?u~*oH6^&emL)p<9Iub z+fF4uk?({}QX3jh|AgduDznq*#(qD}qCdtbFef1skMHb;j*C8Noc3fA-#QksOq`bn z&A{pBE=bxu9FFRuFCo02t84zU_2UIME=UcNo?n0kx1$?3FG%zrABKXMEly44h)dto0_>i+vpX|C-BsURIK9Yn<_{JNWZ72FY$ny~ zO}8Jdqj4z*N$ylnF530@w(K0*mk930YeP8PK7)((qj~bsZqQnGN^n&iFH zC~SEMIf!}Xu8;N9er-L!@H~oNl@I0HP6T3<7x1M#kG73!!-RKRuvRJ$yKGJi|8>Lg zb0TMSB%10XYUtPLG^+TlhVkQ+P^>MBr4!#$>p3lCadV7rNu=dGk)A~=!^V3fA6je6 z=h-WBg+?ilJmbu_Nldx*!lN+UnG4m~iCFqJ0!K{-;-OXt?G4>VD!Yp5h>GBp?J?lu z9|)@I-!S5RKfJ0F{y_EbkXDT$)0&a2>6JZuF!u(ve!We}!)htf>IVhR6=$d6#h4R3 z2Y-(n;YqzX%UL(jp!ThFw6>Pk+LklVb4S>t%OIN@E^x4VLO%z@lG?fxNXfs8uusN( zT{ocbvLRw7HyhV)-b3}mW_UZ_hIZx;YS)+nRSm(&X^W9|IkHgdo8u>);V6UWp(8Oy zMf3%|8=&rR8Fq7SXcO*h-$Ra#gI|#F;ucFxR{gL%S z0hd=jqh4<_X<&I*I1ayo)CouNsINM=z*t_f(o}ji-&&g0WF@_{Po1AWXwDNCGCtff zjS^?Cr}yKkS@<#?k(o8a!473eZ%@RE?N$h1Jy39gtDsvf-qn{^{XZ+(+s77Z&2upB zupzwWJfgozB{XNdc-M;g`@Rq(On>E#{1!bVt{H*upV#2-vJk{X`XhW^CXTdl_;)=^ zmEKoK;p9lNw@G7H*$|R1_{dbEw^Gv8kFZVN&+sHQBl^e>3#Ru`o z3B*Sn{fiZ6WqA9fw#K~~*6=&X|weE-}uHC`g!Y{G z7(o~8kkgPL+~d>X)MpGbCr+i$efl9`e;VTKVsJ2Y0-m1F7T(CytiDo)5~H$d?z3Qu z^>U)YvNhx+IRVA-$FSu2eT-Xw1p%gI*sQw?;n7NnHB!gNtJ(-2n?jm)u{1`Zjk-o$ zq3@o8vE)Auj}^AFAF~`Od$pd(ZmRN<8WWyaXT!@#%-FL-=vTHoJ6s`nSF2Jee|K+E zTeg&Ly{e{*O_q4!>Wd@88_+Rz051{!;pQDTAa5Lv`|^xh4CKTdQ5FAvG{rNq_xh4; zjV(6&X@%1R3bgo238o$tAMQt|Wct$QjR{zEA_5ndiFwERPqgZXDqMSM309#dO;RbM zO~c-jQdBbDuBpQ`kIPs&znx`gj-flDM=ATHIK!omqaFhuk)y$6w2SvaWcx8p5_8-o zug>x`TSaLydEs@9IVO&MN)mhxEqeURwlv!Wiaa}^hrsYZL{S0q5-sItms|@CSp(U zB!q0fM$zBLK+9k!M(e9cE@r5+xIrM-Yz=1ZT*bl^lW0V8H=3RjLaUGMq4D{j>2;$G zGDmr0EK(g$iTgpziu0tsI-dnuxKjMo?{vv%g2c`}i;cUNK*9P=)VX#W>4*5!l4u{a z%OoOAunMc}M_~VRN8A?~n!8qJ^sHT$waH#|uz%jc7M;&w(lMHj@|&j6D(@%ke$gIQ zWz#MRQtBmgR9+}5l7-#M$r$228HzcJadG2WjO(eu&H6s334e5G%ErfZY72PYZwsz3 zy^Rqz;=C>&go1?guvzs|I0IqbvN^Ii?-xSqpn-gB2TzxySF z%Uaa%orgzKS*8kA=_}FYahGWO=@hzAc9?!^Yw_KpLq2fa3WWPa^YFa){Ko7qQmxLh zIPviYjjW%|&TRCXB_R~buV+(i>1auVZ7vjE z8}QLb!}*$!$y^~}EAMno;`xoEp+2Gz84vE@{fXmb>oXN6Ctb(Mv1c)4(konee+sSz z;(jo|7q=HL6VF{2^b6^O;?q{NBIKOp%&X6mxE^n)tF(j+`>NoM`v)r8^v^+Z_`1Ux z(w7((59N1{oAV!QmO{-ajK}0u2p5hTO$k{qHR~TIbvqCto%vYgFEsKo;gTwUce|PL z9vUJk)EQO1q`XngFAfyEXF-#eB5G|d{$=&x7KT?PGV3}W|2sKFcMFRqwveaN zL8xBvMS(rC&IDP3{{BB97=>%f!oxfCJ#L8%zH{X_jIt0NaFb*>W~HEnofPbZTNq|h2jyB_*h%(@Xa6R0 z%aX^-N^|5k1A9K`V7YnMcT4I2cEL6 z=hP{TCeVRdQMA=k3+fjApzRWX!1=ey=(RU3UsK2uMNaitvKqQ*B$Mj;HY|Rj!#l1! z@&QR&(vmoquyTd)OidR#pFOc{UxqF z=_}#rU%jKRGS}JnAJ-(N8;5ZJa#?=BvkIr4*+chtU+mZDhi-CFRBGHwg@S*x z_t`i;i~IARVsNiNDjw0r+3-Jf1QSP9BjBM6a=fg?_nE;;4I?CaSFX@~&+Sw)WjJ%* zD=XY%4OA+2;7gt>a@`|FR6R-?Grl;Y@19`oF^;dDaOmMzFoFN27kr-< z1m)y~aCsI=zvIhkgheCO3buz^U@WObC?hdSjSZN+pWTWS_sGI8xLc~oS53W#1d(Yq zc=i*dY0P)1+f(eR6#CDuD+21iQ<&m!Iu)yk*luk!;fd%`9g0ERug5s!rN#R?ne)jZ ziri$Q5qFCB;QmttQ^zHqWG)<_qOK zlkN^}q*3p$k=K(PG8ii|)l)L4CB}-GiY_Uu9Zkm1JDJO5cgpKe^sXq2IoR7##-VLg zUUUu8F^h_TE=@F(X-L4}tR~MeqOIyJ>?x{}s>L=*3 zr|?lfFhi%2KH>)|;&9?3YQJ7gSzFWSzo!=Dlc@>kF${l3FtQofkH!xTr2ozo(eAc) zwD?fH*zte`>Z-EzWJkJ^BJ%S4R^x*3#f&`No!nDy(2k9LFe3c|o!?`OO=3p7_ER@( zKd%V$lg0F;w3bGP-XP7IT569j6|>JYJlMPu#T|0YHZG4W--(~Qsw4b9S}4UqtT?ui z$xj|cz7x;Uhxh~v3{4WOsZYeub)`k?f05$4RTwn+2@aXGV_c#f@767p)qAUxL%!)p_Yxgdan|R+z zZA@|GUsvouy#-EJ8sIAYuTOeDg8By~&V60r8t#r;U&mp3!+i3x+DTK6Nw9VxXO>HM z(UhkmH>U6(loS>UrcX8%Rx0u?`5hSGZNW7JU+()G1H5~oh55~|sn1#i1P|#CKUH_u zu%%y#(;8q`@cggBehOCB<7r$5d*nDpBNmG;f zL^oA_@y|&*n=Q_%Jb9nP8s6IpN#7wH>u z^+V0r?HLWvz(4d~z(3M#lM~%y9e%Z_FOOVt2ahJC@h^|I@OLhr++%(P-@5iRpZ6ew z$1PSz)5GB~^&W}rVH5D6`*BiAHKBK*gJDoThjhLVpzu58w5aetjqag~YFk~wwHO0w zc_561zauV5LvT764Z1uH9j?>FoL3D?+EZ}QG#Bf3WWzPS4`rz;)5MXolqI7KkFlS~ zs9-*h?#;ua#=$5pOQ2qF21!Fy)TMhyw}{)~PmGFuii5=~U~XB6UztPkcVjsAzs=$< zZaesF$rgTl{su}v*`I3O^rO;=yCfebjbtX1lxVBk3LZ84J+`X0(u7GTna;s{Nn87Q z_PsuiJO}2}rDG~$E^`+<3L|k~(E>6#W{k`417Lf;2abEy(k3U-)9BSnKA-O4Z_mHD z_oxoPwho4IYzQ_R*1%-;Z@z!^Vcz&4iLZ&*=Mhh?$w* zr=8s?@n=^u){)0ROFK%P^_W5%2E)27LpUz$QK+R$LlSy3t6~cp7x;#48*rIaZ|R~+ znu7Md%W-$f2by6qhh7IS20d?pl937m#BOr`n)5i_EW`KA48f-g!Pd+jDf(iz(AhEw z(wtpbs&!i2J7Y1VX%+1MTTZ*8&6tPP3@&}H#&bi$aL@H4!Z!a!&)~A4R;#x;rJH{byP!E@k?415DCZETafu+8%E4HM02$U zW1-_TEQ>Qlu<~gtdVilSuCif!KKv(HD$0tb+W$%GjTGoFw{l>3y^|t+(jM zUtTca-p1NoCL{~r*33msZx3AD9zk`}0?9X~gdTdxqIdmUs=V4k#udW^dt3wCr=B7Q z6AQlS$3R{kVZehrbjkSk0yeBFkWvr?Jm2i>cp>(zWI~cB0^*dQ zsd|rIURK2Vg&*nZIe9Ev8%hhWs!Q^A%yIm!(nJGuHq)>2-qhNhKz|wq!>`X#8sn0I zqM&9Ba#2W9uBV?h5N7?#pNPff8i+EY~$2aHJbcy^hKBNHt^l6 zhcS8*@*h)7V@`=#iO7q~c}G#w>OgulKAag;U18_S7;}GbC)seclvRI9qfdfid42UI z_TRN1?6PYctBBO0O(-O{UPPX@3eQw z0y37JM8!8xk?#EW?3uw}!I_^&;~u357RMvuqZ)*9m;$9QrL;6qu=!$zdv5+v*jL-) zieEL|5-b$|#hoP6(M9-qT4?6ief0C;1!?04uhRiF8qE!6tlt`XdCti zNj`)4xF#`ku_Q#{~0ZVlHwldpXAS zQ-h258!}84{Hkz6B&La8*Sq_;dT1bTyflNKe>E4fx_2>TpBgu~`GxiDBTKjb-Dllw zCW&t7EJWR10kg9nWM2K4mi~%G-HNT;#h39L`K|c1R{R}lAJOBHGIw=+Nl9yC=s|2Q z-TqhtU(<1T^G{ve9p*U>`|^$Tb}_{79@JP zoPxsp#hBHjo&}vAE~&Ks$vo!2qy+~I$#$s;9XA<-k{z4q+2v0x^I?U^Jy+swQZ{nV z1)^^4b>W)t$_q8)rB6eHrPEEONMql=rT&Visc=U&ty=hiLgZ8EUP>0(eqF#9{29wN z_B(J}T~8DbSc2B+dT_1}6+Jj_JQba_%l99VrrZ*kzE$KiS3jnoR-&U7<;=#4GkJTL zz6jDfK?P2mXm8y_`tZG!X6`>ohh}8ZJ-ISkuIxr9!>m}*xi1ny9Fsh1*(Vv~{YR2( zXAjRqrF8SpGCG=UDee3wk$Q!gOLru?;r@7ejM*^`W0$>w^Qu}ji*EVhu2mRWb03a* zCqcq{=kxY74IE=l*WPr+^TToU!t)bFf1XdH1QTPAY$I*YdPUkzan!iDguUv!1)r+o z@uqnL6l~4-%QY|Hx1%dgU)YU|^LA){`h)_OSfct|f7&-{HhXbDo!V|bWAbW>H0kep z7TmX)e~w>}G&~eB^J__C!9-lIzd_+KoGNGc;<-&9Fne4wsyC$}#61TQw{u{q zk|OD{Lh#GKgwxaBj3@n=%5?=NHs`UFYjp(itZ!+2Q>)k)RBpi?(f2*(c8m@mPoe7d zdV-l9E;%%})gf<=lB0r>E!^ZYQ1MxwXV!Peqgr>+vgL3~$j4~eeVB0kFl3kbG1Hg6 zJbX`uA>BHJp^s2Wj_vQkjn9`By-k|z3OVsI-aMnu!&JT zVMGew8D53gyeOry1`SNZIFimx7w5bygJ5+}Jl}qb z_j|P`?g@5|Nr)HT)NUr7uV$DmtAd4hKM?t5QPa=fq`U1dwY)w29g5bn^ZV zl7viq&(jfK{Xp!L575W(Npz}ko1^K}qf{EP4UZpPAT~+{ z6VKbgY|=*9Xl+B#q+(pSS%+n6RVZ_PyR?QqgL`~v)s5rx+y=~DqQeS zo5=IQUaH8`=Pxu~BfRM>vX3YVeu)~7*=fpay<6Gl+wWOs$#<65RTb-cJ|Wd_Tj+<2 z2Yre@&Cb4%vga0Sm{#x|N&FpUwrTGj=x6T5yGgMa<2H!zUD=yUHz@IcFHT`sis1Q) zcgAMJO$dC*k!q%Zs`y$urku#uJV|23$$DIGTO$Vhorh+19oJZz&!-1%=J!@CAo~fP z)H+L-A{6^$)Ce_nHIsv3bg|@T;#A3mVlPQ}wLZKY>S)r+19ZI09D2S%LYvfd$jw=W zjd~s}NsY*t?2*w#Zxq?YubG}FG5 zq`6LHCirWe`J2eKIYWAUWs>x#d#qG{l5p}qm4jOJc`CT|2^a5IV!)F$gr6KM?XhZ% z^w1?Y=~^dKct!%t@=S19vyH`m`^p9mYhoz_@8uA!||Dk&uj zrJ`|P=W*SL&>{(yjFA1cSD7JAg@!^}C?pLe6%AXFC>lbxP)7FdfBhc+yI>}`ZBAcUnxL*H-%(A{ww00ydDW5?7{6|w$ z9}}caK0(cghSTx(66CuKLBRX{R1|*z;#&iCrEr||vyi5^8B4#vSCih~at2)or=d_! zaK_5c;I`3T6kOc^50e{|n43)pzpo_~+aPwJY8i`8&t&av9!9Hwrtue^QBeL~LdbS{ zzj!A-wQxuDRXx4;Djhxb(HqVM-&DC>}^F%P(jtQpHc@uHp z-vzoP*r1M|NhLaq9(~#&d=JGm&~`MP-sM1do>gGkrh|a&Cc*M8rVU;VRPE9m+fwEE zhzXrY9N&hGN44N;^^#b4A!Q%RCIx3hD!TTKZLpt$!R?{&f02w$;}`MadZT#0V84$$ ztIrSaypP}dC1Cq>&~u(1Y7}H3C-My^Dh%i$V%W0om$5n8nfEPN$cr}ar7$N23|gyz zsO9N6RZ)b0e{LYq+77vQBzSBsJSPnmlxD1mx_oyy9#W%`=IZQGMi=`K?d4=SdO4~5 zxgzEuPg%3gOm?@eg^~l(_>q&_`2VOYpT>J)+bMmF_0J=J(`MGs%LK{21F_*}9UToM z=vvQ2_RU>X2jWX#-eEK3?C zaZ;Hhv5DI+S?fFm`;W}P#O75f3hjrQIyq>n2$t)Fll0}=R?56`gI*LqV!sV{Q`)l~ zSa33vy!K9FkC*OX#;$I-Khm2LjmMxcd>71S4ua2*>$Jfm6dn;3P;+{O_rq3F(^q@6 zPYc4LISZvXC;Lj1XWB^BmU&9DM4s8s{0X|>cubGSO(B&_d&s@!1Z>_<#MD)lw07-0 zvZyR#Esl((e;$RiZzXti;3>7w?xcvA6vVk-$5zE?;S8Teh3jW33VmIl{ zH^+mo#>g%>LNnflP>szUruIOOyaRf$IOlED@7yg?)e516yWwogohMY`GL=NG3I=<= zV#UyAZep3ri~kmNO!dQ|o}SR1U<#cL@<i3xTZTF+@^%seq67NboBI|@vG~uTTrTga4*?!tw zWBM7GX9Ppzb2t)rtb@)+4Xn?T;eEE3qHK>u@Qn3g9_o%golEdvuQH~&XDr2JJ7UD? zH>gWe;dA>}qub)kV(u>T%40@w*{>R~Zd-}1NkzEhyA7uzys>)Re^7-Ht=Fq#8!zNg z*qu$nFCEWw8WwTIgy;0sxq$j~z9aWj3z6{c2(~t7BQxX}Dh5YDzTZ0Z&tsU`XiX#g zOl2RBY^3>B*VvjxquHS!H|YH@7iwSVPus>tVfVHfxY_!c43_Q}nfXFI{7?g3kD0hX zRf_WpR(P{PI9T+PQLZjRoGbdFQtYi(?F)m|nPfWq_$4i9X&?kGO#wt+_9uw@=~gR$~kAJh(Ic={t8;U)v&s3wCyF@giPpg;01>S1J9 zACiyEl$ciRWrq!&@Z+jCb#sqlm;0&f)?ZW%1ZqinQ-mphcvOS0JCcLBzaJ8stx-O5D*lLl zirwpUD&?9~)_H;2O-krQ-7z|Qeh#+J3BdJH-jLlBiJg*I)QCBs;<~q_AQMPUMYG69 zeKSr(JE3y85%yUn(XYUJ?A5YC6gX!KRlQ4R=O5QebQ5)9+}uJh=IFwtM=hI}P)GL^ zdm!MNEdrhhfAx`{f**c@CcM-jn}bKFvdo`q=5(W>7v|A-!46HVDu>6caKxA=;^3)M zXf78HDKQ6^ce+nSxj9sQTf(>H^yPCKe!+3S3-_^@z!mQs(epI@pFfHO^Jir0PDD-6 z;&(;f^q9XTsl3@o*&ADNwzeD3I`bA|%gw1%qZch2kjs|-=^*!^YQkq3Bpd=}crrH> z@1Lh&BP&JF5_R6y%$Em!Pr%!i{7iIv(8H(w=c)3F{qkS*7ksrn8gi9 zIEW>|6`0nd!V|W~@#$Bu@Y{y6xcF))uD_S#o{Jum`TqI1CBGKSN?uE@skyNJb5q%{ z$@hhavKC*`Z{d^ec}ftBU<(`JgD7o>hjKPl?)As9+-vxp+Jq7B52I|sO_u2HMcsCn zk<8!TyzIy|bgG@AA^q-BpQ}&Ee0g7d_!kZJxSbe1AsPGPm9c;EC@f3#M}2Y@|2AUI(k*1~V{SyC%( zr>OXz2p0^cCdU-=`~8>ZRlDNAH^#Jehp+_SwRCTB6RnspgX$d_l)*1byzLZNUX>oZ z9G1@XyTO9}7=cl_C9L7gZN~Px3ienB4)wnW=j_?Y@RO3R%QedQB|5&A`E2XE zI9lrwjev$>q&&KUdb&k^)IdjU1bd?LM_XR!Wa zx?rKq<>$@_hnmPc9If%hi_DvluS!GzZdYOSL4(%mpCR>#>9`T#37s3Ve2ddB{HQD^ zFL!+m7whV@h-^xd7vFCkdt}U0mvm`vXL2^e^*rMu@2P)`tFH0ldo48h9zzFy$jX72 zVwUhBwveP{6qd}-z=z4pk$pH6L4^tkOUfZ#Fj){W)aRI^m$c(GW=0Tj+2h#aotzc zEVRSDp{v9@zmU!kC?)qV_eklbSSPmHK+(pUZ_4V;4~z`Of&bRR(@X~j$2L&co5xUH zmWz<7O$aaiCb8M+LYl3nSl4r+)L^K)bW!PisnOFg?DXlR)vYFEtM&jXpH=y+{jC@> z30#Y}!Pfj1wqJcqD-TS<)GU8&9hF66mK~r~e@lp&7La57EBaKEfx1Praq4>(@}na0 z_9MkDQq|}&(%DC|nbPUiG^=JiRtaXqxWG1YRemVCGNs6? zEyDg;gAh8t1Wxi%2;3osBtD(4A9+bdX8HIwZ4*Jch~4792P}DvQmw&yF7+psq8DV7x9$% zDvz}Olf@YIR2Go8L-H&8BjbG{kUrI(2Xwtl1LoeK!I1;-YUn!n4)&x;XBsH_{3Cq5 z;Ddw(`)K(XH;h*7q+RFdkm+h&{$YXWd3yB5os1Akm#3c0$k&`6Y}P~1OU+bpXByb} zS;DvXj}=Zk!eVp=U~#w+JhrT)E7jE!X^%uYd0PsZUEkRfo8xTBRb!f+ah84lY|hHA zcBd%`Cg|NI1;?2(x7uxq3Qsi{hX!H4c%PPyxJELe7sx=I_j>F5u=~*yXvEItP-!B!r98toQ9-!pVryGAjaH{&ba zwA9A%=UVtU+BQFSbl``Yz+lJE=xFdeo{f; zb8Pzg3^9=%Xl|Cnp9KTqeRLi)zvhsMdmDLb>q4)m3+BcfkcRL$EG|CJldjeAL7HmP zK-|U6Ee?oyCVF;P;;18gk=Wx@;iu;$TrfO}YTH{#x7|qhO_q`LQUR?Pd!8k%*i032 zPsq+tWUK_o&N(QQt~@!x078^bKArnY2 z(mv@&8t48>E*599dbd(~d-pA(c{hH<%!qIGROWr|hil2D2Rs*_@x5y= zXr1e4N-7^mepP))+Tu=zXC^~qeg&21O{WQ0hH{I(1GrkLEdQFD49Tb%^j;l}X`gPA z{^<~!J=>LHrlwL;&1p()s-m91C)4W~1BwXhMXevj%-g+~7W>7}Kc_aTeD#Vo|Mg~( z%BLy&fHHMXoXI*4n1S@=F*N)xxkPor8O;hRJ{m!)kDn7Moun1jJ7|2cHhQ|L@8yW>F{KAVuTKO222Yh8%C#nAxJu;p_{i>#+=%*_N-_*jfBu~*u0d!Y8 zz+F)as#ZnL0zq&8)B|PpYRDb?P4F6uank!VB#R3WXy!qaFY3{lbsA*8ZyViuH--K1 z_$>)|noK?)%c;OhA8q2ynR7XZ-oM#N7stgh)3sxm_NFZ?Iq@p{sk({{8+f02B*`<@ ze?9Z`iDg>qGSc7n&qe;?5;u8r5V1*#$W}axqDPrDP23rp2E>!fDP6%@ji+o|FS@bp z3k$v7L#&$;3OJ{Q+YfW-qGC3Euo2GPcX{;q*HgMPa1{;FkkVkLP0wp!4s!Y!Lf$tI2rv$qp%k0dl}1 zpIS;s(-zZQygYLmM9J3Q5eA!W>X^f zeSC~uf4NxH5-hllu1b-O3AKM&-fG_@NW-NTc4kFu4jbo7_X4DBb4D+tBh z8DhRsB>F)EITNKUZOCFlqJ%tFvB(d}Y`f+^a&>8-2QS=_os=qO*m6h< z1&*2AqhmMXG5PIJ3|pFni~kCt*LyzN(#N1fYZUsJpQej^BQ?A*hV`tSh|w*eYjZ!7 zjjbXKoGNL<>u?MkV#5FBj1fG9b#S#^h^NezExRczI0zPO>7z6@ATF2&tzU-1X+5!Z zsUNud0l2E#Q-)456}kx~L&{H9?01X$HK-wc(;0h}j#m3k+5wWiy`I2v=|j?u)MtZM zg%Ug-v5&FCBu#SXG3D?E#AYdoJIj0&#n!;u?G)N37}3q*UZkjSg?4`U#qJwklGON0 zn5(!GO|h${=bHx5 zbxI#4Zqt-$l;TL_NQsSFS!q9eywomo!&^yQQoba(ErC8JPnB$#I9OsDnQT8gdo0z| zZfEk=AX0oJ*yIJkv)$IAC*=)&|pHy*rGDTjUNmA7dlFrUQ z_G6tR?2CQEn786j7B(i0TEEVwh-KN#D#cAQzIdOcP-C>ch4yT_^vF^4{<;nOl=;WL z!~D1;ecF4;+!?`?Wza#%TZ7@RE;vrQhDf@15SPNn^Ba+}I5zS*)09t^n16Z6ZfVbx z><->389rwusr(D3kH>;A*C$x=M*oGKneiT5^ByZvsWk>uz6ox*=p;$>Lqm33Lybw-|CMafenJM`8>wPRBddxD zppQ%UNi+*w>{ORE*;{U?v46Alto?Sff%OX)%)h0EkZ%RFb*BwA&g@1``?c|RTnQ!R zjv_uY4f`FmVUl1(^J@-C4x9>SKXYbFmN#eEHy@oSQCS$zdd-{1k4F1(BVEDFu1ls; zzZ5!{c7)Uuo}z4G2O_ibgzx4STwcUt`%6#k%o#-OiX9ZOXArV$pE9wDr8z2lNq1g8 zRjqwUzs31XHhvXt>N1>`Z#E@2;ct{*T7u-yQ*o*O7_@~G?9kc6IGr^G8y_zbJ@;%> z)yJXY{b=a_(L-$2cqmkig>mTvDlPqvwNwt*>NpHIw4DEQ^Wd2`$MO+RVkE)yPe|ei zma>v3O1QrMBXztfA-R56Ski=B5#=JZ3w}EW6{oFH>u3TS z<4BkrCLra{6eQl#l|Cr@&F84s^U`ol3|F;={NL%QouP}Vg6%T=H6cGniz_Wr;v;r< z;~Ug+iI!cYQUC3smn$;pYK{%{Pj6xE7lfzr@NJQ$+=+kz6WIvYo^0W(x$N?(M>I}j z2+nBN(zTWt!MAaNQ9~AG?vc>WQ)8*i3`ZI(6Ga=(t)Z3sp3}*T&*{LXfAsv; z*^_hW!RN6gE&Y$Oh3ClR>Nx&+%p`7i?+a<~5E&_*nMiKD0kc)+JX5n5A2KAB*7wV$ zx%QekvfC6Blk?c()(es+0S)wRtGei>52Gg$5mZ!pnsUa(lj`(f6eSwK>S90oY;VRF z zN^6cT6^junW?8|)F%;iyELn1M39Co|*4{HBpLI7S8>gFLZ1gBRd!r7Yb4e)dyOCsf zh>W?&5Nw#>iPY+U^y_LG_H5aNT{ncQ<$W+D!*9~QzbBYUPPlYWL74QR{50vTQFrOz zWh?ARmEz?f;XD&vr-)mJ5m)#R1D+4zl2K#$iRspC`6N^JVS)$KJ2MwA@5(@JyA=hB z{!X3fE847!$Ms*E+3LMfOx0VC&VBG@2F}wZet#_3*U?s(Wb}>Ha-ISwI)pmh8Nkm_IF_LcB&`v6$}-cwI)jqb=$BlQ=r?a{^B8o{3{k12ONMGxA>h!f)LS=>2xb z@Kg0PQQ3k1Ug(0_mb0Xh6;9oodLpZ3AR;wdN#$!UJsL5I4z74cky``t^YJE3C|r+= z_xqqNz!$#?7YoPUG&HYWf`0=EPQDAU^@)RcC!VCUnP%v@qc`r>in-+$Q*_MnLQ}O1 z%pD6*_4olc-c;lBn-5@@Um6sS1i({w2fbQynQYu|)9DTcp3rd?fqnNO*-V}{{prt* zduZ|D8EMR|V;YW~{tl~dqF?y^2NfK*#qPjRD!MC8b`7QC0mEo|h&$US z?sXCm8e(3cH7sOq&=(7R)EbJ6hvN+lo%$VEXu)lx^KkyM3!Z!%43C>eG^N7^-)aP$ z%HN5?{C|)<7tXGUy@Y!u7_M`6iU_wm?^+bgWt9{8PlqlTR9Hj3LT2J>&<`5C*$Pez zv~V+LD=93ArrIxS$V+K0xiyqco7GV8}aNpYh$%v$<@ zl;_d(k@4geeuD-bkj3gkGv2Ct7^XM-((G>EMelVc;x5fXPVi~;`u0fV{fY(i9dHqS z@b-?QSSw)T9Y#{=m#d?t5m1ClSpNPPSnoW3c^d;2=?SWjM)1ZGD6tw9CQ<^-A-RWvYx()4;gKFw* z$<1ndF>pF8Q_Ru5ZZ?fH$)$h~tz=%AVK1wd%@XSTXn4Oy^gMhV!&Dk6t>h&dl&dLY zygE}c86+(z{lH~cC-N4pZD=@Cj1PTEu*|`VvGqi(rIrRphhbWo4!o5Fv%0P?N)9c= zhI3H}YRwfuRIM>rUGY=$G?}bj6?J(~s zrD?kHl(lxPnAzGx>8$V?$M(T`cOR;W-$>?s9u@V^CfPfxD9{mi21!nE^PYyaz3wBe zWV>+x7(l0e1st40V0kDMnHfKzJ>wke?g>ATyVxV_-Gg%v&*RCUKpIlDo9)&LrVNud zT)&{keP=(%x(A|5*}4*n{1PIDG}=$S_>j(fXF|5eMCxnyhsj>uPiKSQQSgx4OkwE~ z$+l7ru5D?eJV(SCtkQ}>`^{s9d2YCv?tOH`E^^RFX~xNH|=oDuJa z(Ne)EzIunQ4BkTj1tu`pKELRKl_nC_IHRlZbuM1)K8_Y6y6YEkcS`3@w{-dX z@n2wjr~uMOp|ttLc)IS{i#ChraQ?oxq`h7nua0?<@zqc|eb8Fub4|E%wIRPNXU-4Y zsiO?l5HdVvg9I#~+9OTu^SXTIetU^{ego(TTSCo&!*Hi07a7`$Lp?q{$@)}QWv*haZGMaKGxAJ%VaqCYl0Xw>{h0UoEqqcH^O|Ba(PZ?sjnggWGqTe(p zp62PVq?*3z+wb9!{gV-#Km#k>~ zT$X*HFDc(}x4=D4>wJ!XjX!%aJB)gyc9+`meACzDR`)@iil#tGB~ z2>0!k?WhoZk_y3r|1U}0S@|9lY{T8~bB@HfTt+E6Ewp>w8%oc*L;9j08KC-Bu($S5 z_yQ}kdG?*%`umki9P`tbZ+F?m|An9BV|hC6sYhW# zpN$x#Q;dZnIanRC4}Vq+qIAU~8X(vuEcP~S8y<(KpTZf^@|3REgdu14KKxaZBJbN* zdKn^%++H#4wc!FX)GMRt*Ils0b2A(_Y(w>S8SJ*4M=LObjd#%_#hGU*+vOvzefLu` z+RT7i9?--3#yI}_>vDd_O^b8?uee_G3NGW~P(|-pW92|n$httEKQ2L!FL6lUb^tSz zeCU3{64rHn5xrQFg139tL%Bxq7Bw2#sjh;XaBm;wwUx0mJ2Kf&=@VvrCstB4d4^== zx+~1=vNMWCgdo>&E9Qn3)1cX1(52fy>KE*aUYENdD9(b$t}2vt*%U(Yi>C_K)i>sv z6UQ1RYA}NzyKrpdc~UMI$eh*~qU1?4+2sb(PcJVF)^)&v%Lk!1I#KxL&d`Ye#v*Nl zhIFo$zO*gNK)S_!G5htaN}}YFOS2O5&ZJN+1v9^^ZgLx;ZtJ_zmd$0>rx+jkxzXvkehIjT~ zyXNzQf6}=}Q4nvpsbH(GsL+;Fk;#~1AehqOSSOc;XYz@3HPKlp4Wh7L^y2DbmWl3C z9hnW#VjqOVM>YKv4fttCZRgih!^_2V+Qx?siWN@YuxGU8Mj10qm!Wxl2Tc}UAFT;D zaI|m`+_DFw^-50!`~RbnD`T;t&mByQ+73&r^O)DN8&8{^F}sU|UKMsBtJb0PBd#yk zpWcM_mR>j-*hIJd)990N1sN1)E=711vR=?Bzm!(|<9iG8xwU^+d_;r=`s0 zrs&~>7QtoqUtEnGLKVBV&?TmVe=dSqdHpn5WSAjh&qjSPuhxgrk%LoDCIR#A27W8 zXg=!EGX8j87<*kcoCbfqM0+lZy;+b3IxT-tNSQzEFC>FaE5yRr>U{ac#k{t5Gnajo zM)IHckYx91T0LF~vvbD5Fv}B-kJhpD`El%Ttp{B-F2b_dJZx~igki&kXKM~%v3)br zOf&JY&me=0`PLme7e z-$aJi&WOuAOP?GWjo-VA?zZ=1ojP~Yzl5gr(;N?41c*nlkb{5moLq+kzSfPQR;VMyVN3T2pNd{YW(79uasD8Dmk=kCKa=s5PACTAw z6WtN>9>vIG_U1E^?R`Q_SZ>)Hc=y-m3MZU+<<3wP(JVJ&F$ht$E~U?S&mZ%kf0j z6O&a=BGd5*x*GI{fyf8$@$R6Yh(2(gJ`-O1ZowwqkNy$h1t8xlgwx+lO0_ z=3mA9A8)2Gqc+-c;4DovS3#7#C0?jIvc+3%s4UqKk3<(uXYM;X(9svK)Z*E>vv#y8 z_5$h6_(@iN1_14`TPnpZ~Di)wo&3@&}P;qA~ zTfA6_`ahSUxZcg|fRG=!xs*_>xG#M3-BdWD%%JHZOT!CdSoZvAmVSE;n-jsj3l~=$r16)|q zu1NOgWg81qc}Jc1F49+xH*EZg3dy|hOW8KL4mN#>87U7opaYt#DCo>;O8T~)g4)-R zckO(B+-p8h3R%JRtmDY!uM*WPILIbcYSHtP$4KSs6`PHNN< zcZj{bFUNH9H!;89Y__oDD76?Jrl(CuX!@1}iN5Al$&ha+C3g8atl$1jcA!p%&8Y5V z^^xD$e8I$eFz5h_x%-g4y84ls#`csL%$AYdS#!=_PHQ(!GdL^mxxO;llTZ<3+@{eN$=bjQx~&UwBPtQv+F2kOD5i6@DR?CSvzU@$k%M>{yO_H zzg*eG#Fs3*;1$bS^_DH~(Sx3u-(`-k6)4zCiSFLp&6-Cm(zd6!SV8>_HYw*QD}8c< zt+C%mXI$0Eyu5_9eC&~7fCkvSEh$J@Z+avaJ zy*^FrlFT*@ZesnX3Mb(7NZ~&TrFzTBu)Q%18>IFaIwX+FX7&{sWmS@YzeeKNbBX=? z57#8+9uZ8>AV)%5*GLZf=X54GUT;6S=I{vD<2tJmJSnCZ$4@?PbWA4&-F{2;p zFUuD;4d*rvNk}V-!0aoFP%hRg-{)3XQ9A~^jvb@T;W4zy@E~pLmQDXj@6hXkhOjId z&OgHTTdlci##6fgWEZ(FtrgwocHt`eNRDz3NOeaHnTnY@?Ug}e zmLU&wC;lPciN|)G4IdMEsNDTP@yq3Kd#F4zny=AKapzFgMKE7E1y83mP-VODo_^__Io?XfzBR zH_g$_Jd!4x%ksX(jPGCR$^#2B@!8B7NjG|9zsR&yhVFsU>H@eKw+biNS9D*O~Cm5%aE128QFotDLh^m-TUWIUROqKQ#`1;AtMHG>w|33`XzmJqF$8gG212P}RkG$wn7V zZt=H2kLn!3RWcpGM21VU)*M^PcGB~4$0#Fp20feKj)9evdF}de?&DN~Y?-?lGESQx z?LVGYKieva@zG&r?ef^NCYAP&Rv?GbTge}v$+yb}3VnVRUj%3O%9SV#`y0s$KVPHq zyW$~r5X?*UWw>ddhKJ8y(1=aXsQClKL!$xw$^{c1a$AkZ3ipTHN=tPAJ_KVvC}2qH zVA#L%L%%yYxZCA2<~JY0&sZHyORW%oVI%(Zxevendj-D~w1O8*?~R5p1L(jVtF2skAvybh!PuV#S_g!iABFWtD0eC3g_f51r98 zP#$$(Mj~%x6xPKqLHAEC2>2(((ijKGTMeeQs>KvEO9@N1p2VzIOK|DP9CXvain{h< zL?onPuGa@%w^T){ea=$)W40C21{lL}$5^Jh$qt*fLXno&fR>0Rz=b)x-awJ>WG2DeLbQ}E}e~$!Pz4t;8L=K zaxYs6{?HbWa(L zO@BR*e_70S7nRV-Ts?f+Is>yGif8Z1X9`Vh!KdYeuvWSPMqz3AH^qr8bt755o+TS< zbf1D!dg8a!JQOC<>-by#WY zF#E=zHISS#KP&TgD;kb&%w*HyHWDM8^;udDlYvK9k`W4;bT4` zwD}r7MjeOVjv{hQ9E9V&5^zi~x-=iIqzejz$k8E@wyihEe2*kt|M44_dQ9PMZaaDY zjznl^4C1ca9r^XaN61!Fc*+!gF*C3Ws%Ji>n||G~aIA1Axpu*3)AiKNx`{Hv3@}@q zL9_%{z+}kvmg^+opR2+qRaE>$$%c{>eCj;D)#yYM;rBY1B2N&L=~ z>tr#okya<4rccY;1Pja>)+v72rV@qwQ(|%DaU8ZLhmxw`OtSHtX>*L3BzLAKo0=K6h!75x{)sp*sS;f2B%J6Hkiqe@|O{59a9HkAx*?7HLj-T40&0h>TPVc2@ zWHas}O;tJxs}-%VZT^MRgWBu|tTJJbx_MCjU1gg4Gn2+xw$g#e^>jb~0$sUMPfg#e zDLSYt>_gV0vHv;B?xPNcvs!p;QUI^XPta@o1q6zYmxFlL#yn3HZo3({pd5(HLCSE} z9E*ToO87F_gf5#OAsL?sDaRx)C-ki+GZYQN5 z3fMPqG?XUw!;+g$H2uf`YAU75I?>%a~C!^>sdth7yZ(jU?I=O0O@mjd6o zvj^wmJ%6|RYc{-L4V}x~N^Td$ce5pwQu^6ZVE zKg`AaDuWIj^<>-64&kxK2lGCb{rS^3*GM_@3;9m&Dj20Pa9^B_v8p@Z5WOC%S6AcY z=Xv-!&jAx1jFCRPlEPl4^Z!i4`B^1L?xbCZ_#X;(OFy@~GW3WlCU z9<#rlr|`sV>b!A{5;|VcqS`v@JGY!JM4g~+TlS)Qqv%ZKI%3kXL^3-&o)&gJMGrb3 z@az61{EkIBKUaAX&t}G8v-$vx^BD;Tnv6XQH)Hl&FI;LJiFw^#Q>WW9XohUY+27u1 z?dgHB$A@EsaWKA?Hw4?8nqFbN@+&J=3Atw+tP=VENZJmhC7LifXBgw0%s zC)3v9&Fh7@JY@&|AquM={*Nj24RV9Or`2)XUyloL$-f-|_wr#JDbMA8 znsM9v%Mj*f3f(zBY1YdbbR^w?W@P+F*IKplJXt}^{^mf|aX!ipcjNy0uKZx%ef-dO z;%}Nu5T&<>3A!VyX1j`6STiMTHATF7Z@4`4#HEVmSl#sq<_?yjarV!dgW7Id5GBJa z7u>=1^xarrKNCL1V$Fa3luA*^UQG<5ep_^4{MvzAsA}@k-uKa3KMSg58({NM@Z&en zq1Sf31`J`|)^c2>O#H z{2kNq)Tz7HPmP=^PiUX9sw3Q^;S0t@YJtfc2n@VJ2o%nAx zWnQ~rD6B>k-!rxnF@mG~xnj7~f1$Iq;Rcsxc#4kG=^)tcDM3irPpq^xjk$QWu-4)h zoUl$pWqkB-aOjC>I*N>fWQ@4-84-y>}@;C8a9;H1ubX$KFg6) zU=W3@tDtxtACh;tOY<7o*bpEZ!|>l5r) z!Vf#*BzV5U+-!q{Ix@r#mZfqzIJ(|Z~pEyT<4nJZi|3G5Yaf#U%jU^4YIaIfz zlXTN2Q{n)78f29zI@3F`z}FZ0xpMS=-6xhkOZ2YREFjt7yOMP~W{}S~k^R(Cz$lY2 zO6~|D&&^&!QRPlapAR5oixS@KFr#4)>d44GlwS8%M8>0$G;?GKt&k}uwW?I<)0$;c z@2*3nlQ-R@&Z}SO*yG(u&eY`=&Vu=o-r(>fN> z)Q|I-OP!;ntVb#`{*>Ub*L57}SxwaUGIg_0rYW{3Xyl)#bf~8>y4DCD#PV$zwrU&t zg-v7!V`bQyZF|||gr!i+S&eL$a9lUn!;6}(m@BUhhXF>Utr$WZcFo1#FVFbvC3!r) zDua6qmW$2dETsHAh!KBl(Q)SnJ`TEwwLZT{@xLcz;+2G3rcvl$u8A$K!IW68%#C*` z@-FQ)II=|!nWCpFIl7u|-RcePsjG1|X%3VJ2GF*lp=7sWH2p5zDcE;`K&LAb$Hq}_ zk#YHRb})wXVl1#O$E|&a@Lg<)dCQ_nuG?fb_FNZ{X8?VZk)!7$Wcf^m%g` z?h1)~s}3Ts$)L!52nrX;LM8ey`BZD-;))2qZE-4}H8qci%^3q;2iU!|M7Dwh>HQl{ zxy>R+zIZ62F1n*-;waQVv%=z&eUUOR5d9{8!l2{U{0sBvHjf72>97NIsbU~>hIge& zg?7|g7e+P{220kvj<&b|-AhswT!07W1(2^zfzyNiB#(Qf^}8>+Ll;_tZJ+CtK`kwM z__bJYnLZ4{Aa6VRW4eaaG8fUTdC#!(QWxGDqQKj;0E%?Pa0IJ~8%gD_uTb&6J-8 zvZIIWC6|ZBu+*$L=KZCOby$bf+{|?P5lz%}%@g+Uv;*5~FJ&j~`e5Fg3AiyV5n-8e z_%HMbwtcTdioBZiNQ9m=sH=`t0FtrUS-82P7Xa6%kgK!cV0zhN(F=}^ex9i4+Y9p- zVmyC)!K&stPXHNXhb1>acf zj^|f2c=zLGT<#&`yRH_}>SJ0M?Y|2J*Spf*59L%n(+z6|=HjmW8T5!dj6uKd)7w8c z>Gx_46uX(>WAi2)s=1EEjRCkkBN>@)O=5<9iC$gUPmf!#P~9oPb5O`b)Q4hxIq{D& zCViv*KJUnLoTAeC>s5Wb^pWZEzaP(Wa`IieC@FRMfB$s44F5mQ&ch$e_lx5;Sy>HK zNV22Bb6@8sO^T8b4J4x>l?LsMghIBgl1fHVS_S5w3=pVX+MXEt^9v~r|+!>8dS%uJ5i0X zU-PhXye|Iue4@tnb~&x?<8V~sX6$j$!)58eIHRO2Iy5^L@3k%BjLa!n?XE$u7fiy3 z>yB}2XY53U-Dlvp-~b#azvyE9lPG_z7wx|fauY6&e_LAN&eqifgLd_$c$+WatczTjQa>lzz z8ibN8iG83+Z$a{pYO1`v5?2}6L$zZk6Q$HZKz;)}SbG%jx4fk%r50dxR0ut-c8GB+ z3`5f)(T>CVmvqHPb9^u6jrk!5xY%N8nzg8ev7Rjk!Or*bTWK8T7(XZ55*LUVni3SO zsN?<(uHE}ZH8*(A94Bl%F3 zC#>M~Ny8c~NGyI&3#~E<^V$%ySWj}RJg{|OG3{D-0-t!=pybv{ zIKs{1pS)M%m1k|IF{_>Vk~bgF=J5f%Fn0xY-rtK;myX7OEvdNQ+>rO3ev#(P+(Lu1 zV}vtL`!i4WW^+rlBbe`-jBv%4N6g$UevEb8Awj#hBPgXlCzdC7W8Dp3p-c2=a{ERm zxnenh^TTfA_LCC$!PuO;Gvffr%B~}^F|H`H^dI(}l|=9PNBOPe>v34(7QJRIjf<%> zcYf?8{Nt|zsSB#FQgT{sbqi4I(q%H zIX2&ah#JPzv2yllI&tGeDzjjW@JkjYpYr2ifx&s4nGj5_q8%}uo5yG!O5^3*DtPy^ z4uYxAB8g792E;2Hk!P-%ylVer{-v`&s>F5jpQn7~(}SksRh_jMqVe!`;d3~e#{Z7qP_d+(E_S>FW+?R}x>P(Rvr@)85SU|ryWN{Kcl-gZ!!xvlDlWD16 zsKwlEH0XmW>sz@OwIB^6?z^JhlGl{=QiDnN?AU~k4gxBSmD86Q^nh+Ep?e~*>BIw= z8g`E=tMu2zc&T&b+GS35+a-G9&=K^xvl7~37vuCdQS_8~0P#qgLK7Az(f7;7S-Tcp z7sTyL#=k%Ean3+9=&$ymeGO-zQo;Z}Smuzms|HR{rW1bJ^x^@XDUb_YG){F~{xm`nyc637!a5k{+pVjU+u6Fq!jvttlAt>}Kq~i8D7s#uBUYt)Sj?4{oK}v!2hI z(Cq9HI&G+cqfo-OTgS4^NeBme>O|Rar^ugkzsc6;UYHkj4xn0{(Dnivy7!7#x_=KnorZ1S0OaI`4TanO)+WI5zsqm2lp143Tt*WS?A~;6fRY?p)*@X zlUMFh5Mq=7(hnoKoyq!)k;qp)qR)(0JUho+e)pRGo)d%{Hl~q;@eA?myRRgry%ORL zw~{&Q?$cR!cHz#kx}XUbY3f0Ie;poOy_PiSi*~&##-NS!Y0wJmf(Vj^kECm8jNJ(?Q?iQN zG!#e9*Vc(L9wu`?u3FJH-OY@-u$g+G0fez5Kyv;~Qn&D+aQOUDvaL=5&Rk~j?YLBK z7;n<2C31B3ums(_H;4LkE~5L^Os1s+{osDjn)Olt0?Q{fk<-2Tl|_Y1xj$hCQTFO< zY&>g#4IS1rBcYtBt(;2|*WZVv`43^`5l!NE+C}&{Q&O<}wkOkL?F!7;&0sURoOyk7 zHf{c!hU>cf$dF1Pv{Yu&F=Hdi*bR&6OdB7PqRCVXSSF;9Wj}-4nANW+biLI{k<5j@B>AS8h~N%}41pwS6Af z<62CH$9%^5U$5dAog?u1W*}bP`31E$d2&hbR?x4h{!~f(6c;1n125j`rl(V$klhcY z(6rE%nPw+V#4n}N`=8d4t3SG+-{?5;v@>Kjs9s_o9XSmu@#dJkFq|2aAj#g!OaZTn zGDvTGK~eu!v?{9<8qd0@s-=ruix6U~ zxjF3)Pa-Ds?~wqHa;hz^gYmsu=qJ#?E0((WTPKR%c6=ZN<3m(jUINAkuH_WXMuSG{ zJ#L;vJS+|jht$M0`mm^uO3i&P)cC#y*4Q2)_tT6isS!Z<4bcw8)JVod{1Z7=zmBj$ z4;d3PZzjton{51UP0g-+C2gJmAT!)ZRxr??N9=ZXHSzqIaZ7-I|4U6 zl!a4=dh-}PJYKk*3`IH6f6$X=#+cdO;G@u5BmjQ&PviQl4iSG$Oq&P`HrYX|8)vxkJ; zZeULM)ME3K`P_2DG$trTj_dtH;hal2_bu)YI2rg^={kCHUMoCMi;+VUxn$w!m5+(e z{XA6qqDbCtGvy>b*9yKCDS+CQ*&=_qIMUxPq#?2fL{B@5EbTf>GiAopamzL7AI&Wg zd;1ghnX@0|Hbmee%MGHQz8`GAtFVcO%0Wr41~<1nBa1ASUWh8u_l|r zRooN{4JVT=3cH9vUev?*P1j3;a8_05#QShf=D>0n;ggl86BRO)z%<)B%w0+$_ znll*2N*u^z$sA+Wec(Ei^y($!pc_McOm}m~Hu=yP+YunR7eiXVuH@V!GnvwGRhnC# zz_?z%X}!V31267ajE^u*5L={$?_S;pHs=+0WUnN?*(XL!i$vbyn!AX`I9;fJxs#xq z82X+y1u>Nn%-azG&0UhzQ7}Z!4sA2S)eUoHWLZ6zY_^yf^FZ3bpp7L$=JrF4qyZtCNw1jp8H zvpRTE1NY1xAd1lnP%fqhd%Mi=jnN>PvNVUg7?1!azqQDkzesQR9VIsdJmKan3oux} z3NH*+qave-+qRAq{4u?bo!56@)8sIEQZ2U8P>!=4EOE1x&CV|9SFVYb8%E^utUXwsZ8On92^eY zk9R`zAaKwO($8u$t{)C^!b@rNv4auSsF4yXjh2Qojf32X7ZyZ+h94d}CXTBICyCzB z7ll$oiO9{+LcK%zR4Gb`;}53ez^93%R;~^$20BUeb5`W&igG>&t0H|At~s<@gz zp2#-Fh&-np1OqmU2v;M6Z)hP7$KDlA&0omxo6?3uyE*Ketwc9&KW$wg(iTs=&w%dN z=47DPimXxi2Ldvv!(?_3Nd7 zX~&yMv@b2h{1H`_sez8$(MNTJq>teT?^`MB@i8hoxV z4sy|wXsUIUrp;;N#_I>r6uVVuw`d>Tso_qYUq|4>j$BYO`GR)Zk&v2VAe51OLCYJr zVpH}vys_m5{$9D4dcI!>OBC(U-y@EGx*ZGcN3#igPaJ0cvqC|kEI(CJowr~1kqNUb zBv#wjLBX<<^qcQ?DrX=Lr#nh9W$z;N-`7GtE{dUl_5g0#G7S@<+l zsY(XF3P#I%aS~-kbj|OlbfK*rcs$?9WPB{9v&NJQHa?5Pn+1SkN+)YtPRfFJ@-<@b zl+O8>{T23c54lk#gJgZA7mqx4&-PP<`TUaZ z=sbrLeYfxlz72w)i~Fekskyi;LIO$T25NWG7Hvg!+*I!@3is?q%8jq$9bnD9u95wx#zpo?<7IqkHOY>!?P87q|Ir7nnF z-h657<=B&XFnoRlRlgd^e3^Qe$S-h036E6#uh$SxRw8NTjj->+5+ZXTo%7nao_cJZ zh;hwfxWl@Utn+Y0k%t$03`x=U=H;|wb_x}|2__`P0Cj9sU}5ZXvSU^*`Ek>b`S3NC zEWYuHQ`S6IW9#yR?%SOr_!BCHA2Q-G$ub9Xf?1rZs|=MFuUR|mOrR#F-9m|;_1po` zl+n5Q8~L+S1sx>@nWimgi1ff|8dN9+f{7fRD%4^F6qDhdZWMf<7(&vu%4nWr0=H|O zDP7$)g{=S8%`6!!MfMG5f;TsTsCuj5M$VLdoG-<$@0$vvK7|prDqAM)uK+v7u+%+)}9x$>GJ$z6yIrZ-uprY{GBdNrA~u8(i0^(xeqUEY_xLk(TeqyJx4Q<5wz#6Ddpzmf_frtI+nxrj zF()ey?IVAaHo)D;5TMc7)N8?Zc$6l~s$Z%jsV^(Iyl3$^z}nE&yAFW3ejeP~n9WA? zs<2<1%AsQ6XzG8zRS@572$$gu;xCv;peLu=hHspdHPD2!4B@YHx;>&%>uy&3C=!&>{zs2R8 z*XwAZdZ#kHS3Soa8X1DOY&F>|zg&8MWR9RfN}qIHO~It?E|C3i9Mja`33*QH7*Qi9 z(ked^=ak1}!&40$WnD$C9(_e4cdmf*&L5%ci8=dy)pFshyfI8F?uA|@SE{i|4|8lS zX`ithv++|jk+aGt@3%$@w%Cg}I$c=^y`0TC@66#w{=CLmcGn5XhtH5Gz7U>X^TJQ5 zC&7a5q17kES$my3kUnmhS(;*ww(|;UR@hBZhEpAQY-t5+4O2GbkOrqV;RH!=+D(2} zEkb7G5$mtLM$DoSqae#r9YzgpV?)=CCtX|5!*JUgHrHY_9#NXij*IT0MSca`-*t5) zK%$dueh|ks7qN6q#@D05_tRfqKgyOhk_AuFP68+*Zj*JLx z2cHw;=`GtE+@!^0*z}1%A;D@1910$yb6iTvsct29!_W%mc%L7cx%oSHK`)c>jc|hP za<-)V(jojg>;iKG4sy>HKEq~Co{Zuaa%Q8(lMl*D)ak-AI<;W~hITH-we1AI>`S7` zxmTI-ZFOMlG!3TQdPxVh0?;rqlQetS3f#|c;vRbz@j=RV5I%7-d$DmTTUJql$~XU! zx&@AOM@%~xl2t|;_tn983l+M3hAdZnW-iki??ayDbTE>I6X5K^{+fTKdxg6;|Dh>y z8eFyg7GcTgCNj6VoKmq=uC!4fm850S{+}}U%Sr=UH5x^hZw1W$07cg8cNT=l#xk$J zIq^jWGx=T2VeA`y80PExgM4uWl%~f*)h&@8`pcfHo1hCX1G}x8Mt`i)-c`;H3xzP7 zNrE-=-Y`y&q^xp1BLw~$hk(EFkuLhQm%3GZ)0Kx`(f-gN;bF-o^?-1fR7)W z>CXp?@WQkz95i>t-l0X1ZB|Yy%xBYdNqNlS%g7|j8}w@VGOBb$mf3yFkh`{V4Zmi% zh&(qLLf_;#cDP2CT@S&0O2uNnEhH3ec4!LX?y7)nj{@iU@haOb+|TOWY$LI2`fx^9 zHYCf7Q{y|^siWf?t}c8o_dqX(k!(()bEjC+F?+>9>{1U64L(Lc?q5MtkL3{mbVssd zdzR4lSAj6UG*g&%dNgjeOyRs9We{D7GW?Sq$yOw2Qua$4nlSd+ID#)FIN^kugH=z7cVRe zng`{7E|Q#;-PXFfobbKrfbgViDeVtB&J_*z3jf>jk?i}(lh?st1aq%3bhgx8;y&>Z z6S<(C%g&U<-F`pl_z?{IMEezfe)5IPa!O?NRI1p&^83k^*MGRlHS-xov2jGVe?Qx^ zXapOz#fx)m$l!^r-sL*SWZvn*LuT)48zT6b&aE1hU~j~{!{$dKR=veYAbsMTwbkOf)5N*Vpj_v&Zv$P0Ded=M*wx_6V#@oWMyR%VaMJ z#8`s{MLaE~21kyo*=XEpgVfkI)_uuCDjns|==mF3P;;`! zYE*>3P{-mtdH6b-IW;jzkj7?{@P>AH{k@o2s*a->`!hLC-kvswZfEbzGD4H{26(oy z4$1{JRPvH2mvQZ8bi1Cw=$&u}VNfhSVQzD6CY|)iqbi!bP8vI{v+=xI79IX^5ZjGQ zQM)UPOZIssj0)1H$rCkh^~vPNKTG2H^PB z3Jw%VVM13XozVKB=FKJ+G;GtMZGsz$rVg|zUK;OsjbNJwUve`x-M~uG<937HKh7r6 zjwU+wQVRXTS4E)QV#P5$PsDfZ4 z7HyJbUrX1L4ZB+Lqe(G!@H#|p3nlohbx&xFXm&Di?KB4b+6VzPjmQMIQQHIU)b*Mn zHhRpUW^v1C)FydyQ^yIIJ?^5sS7}^PrGT?Iag;safnU{|si|`?d<#m44+C$o<&ij_ zG)WeZzi|PxY<;eCPYkD-whF#*1KdH8AKrl53%)2kj^Cd1fO&N|10>3{nWHXYkU8}+ z8L}M)BQLZr-np35>r3OF{kcq5U$LjJBOg(jSZN?d0%QZ1V@1n#deqg3zkl-#85#DM znC()+5zqa=yeS?)QX8WLf2qpDGO{f6DLRih%1@WPPj8vEh`6*+Smb7ffBv2$UyVld zeN!UH->P1+r_7mX&7O?XIqtM6(HqStyryUV0tW0#qJG*}>5?01Toaaa-4iNl?DE|> z_r3%j+mu7K)zxt66E(hgZ8-{Cq3EhE?*f(J{)2cHEal&ipsk zX?hP|);M9Y%xdeT@GWrmuQ$`bWe9h+)?-A^Jx<>zjdQtt9BX>7a)zp^q|GRu{FcZ= ziNB(r*!>8{r3dnXr7M`p`PnopLK~b$jlke%g>>$f5&Yvy4IN4VaiU%j_rhDJKpovEY3OVwO*!Awr5-b+=_B;kq%WnyknL@v!gM_W%( z%0Dzl&*NDz>Ch8&7(WsemEvi5g1T_DGAnF)W5X1R=2|ZCgS6kJlGlB;gO6P^hTkV4 zAT%}z=O|fGXaCu-!a6~nb1+;5p>N`H@dZ=nkIGj z;-7>&beEeAoLbj`;l)wt>sdqjT7u)eqEPR&8U)ubro`6=pShTk&M$3r?s_S{r8owA zBUDhZB%Th(TVnl@%iId^W8`sScFOttEvl7v=bAF@$w`+smVwVl`|j^-BR5mJ2A4F}tjvC!!l-Z`F3Rf7!3 zfn7ZDkUT|Py*|${t`Ts>MjW}xC`0Yq=hX4VUAijdbJeGNmuLLZ-b$5WCY`eMBXeS) z2WE=v;?CXaf(NtBNnPs<;B=y<%Pe;xdsgv`-Z)JzY{oav_fG+pEsv%?6w{dzBkoq(M()}+ZR}aL5abU3#|0-$rrjkk$c&OD`23DLQ28d#rS~E|slAif zB`%@YMCWXaTpW6oZR6Ch0$K867jE1A9e&C#IP9%{fh9u>f3;1t)RI^vonfN{M#x!Lm*zI{7=6te2^|RyY ztK?YX7}-RZKPZGBal=rxuol`6pW=!X7gK+|c{FQcKg}IKiyX-a!Kg!O?CZa#?D3hg zQ0bLN$3;%0v!_OK3&JjQllHpN@;#Z@B71>S)M328c5|n*M?iPw1{{=A;n(%p30#kv z!e|M2Ht@_L0ednL{jxM@|Q{%0qdyW=IP9Wj@LFEOJ5 zrgv$FXfIcG{B(S@!vRNS>>)0xs@VT`1772+iG!ySJJ0wUsXC`)t#QMR_G#Eau2njI zaJHp+>H}oQJDz+GQ^$W34#To}(8AXT(W<qxOej?$e7irhY$wZvSqaZ8~6#^ysQ6opA(DM^2C=PR8!ngF_ zkvcG|`^V|fGNvJR0_JQthM6w%Ftg$atur`J-_ES0-i{TNZ+75p25f0V=Oy8*_Eb94 z-5JE*4$%X8gvn7Y0|!9~tM9bdy1I0yu=4W~`kt)i_9s5&4qWoXK&k*I%3`tds2p2x zT~jEdCV`s{E`Vj?7lofgM!{Fr2~cjRh3ZOeWU+WHZm+sqRUkEo?)Yye+TAX|{UTm7 zs9lDCFD>$1X^*06hn1-cX%xn|#}Hn8H;8c`sl&$rzJn5cK3fN!b~nn4Qn`W z(o4Ghoq!YDPvTr&aP8ywqMz9xPU`t+$W1(q&GuKgPYaAu(^d3c^d#AyrIt{@Zot+> z8FbILr<60cqc?n9aZ6Y=-EQ#}{fZ7V;sLMVqS#oN6yr%<<$p8L@0JmlEAu%EHEnn@ zwU*i!Ul44TZ6J}!Q@OOEGu%-(3rLl7hM#Yv=#weB@HjY}F1*=E0xV;=S*=#+&^l81 z@NpLDko`xtO)!ARX-BEfxCdPAKWEz1aF8T)ZN(3xEXX6T{X}9(5-dAbM?O4bKo!EV z2u5M5#vEK3F@(F`gp=1%F4UsYm@+97gkR6?6g+-A1{xnK;;;CVc;lo2WBf`MPrp^5 zb4T8$eJ#m^S#8cb=p5$GW{nj(H7#O-iu!2x+8OA6F_x5GJ4>ey|6-)yCKCV5gLKcm zKlq=kH8yU0h+Zl3tmY9Hh(8zypOh8Q%;F{8<^K&Ai3bYYpEU_hLqC$DmP+oEzce&1 zI0H8y|D{`_HR--dy4ZBifc*XBj@#Bgp>lZf`5S)IEMA<3fG}!M>A!1T_P{ivZ<5) zJG$*AA&Vkh8Tq(eZrf{724uxVoa+0Wlm)tAZrU-Lad`vdnHgD~vFI8utaT#GrpA#& zsl8Zbca!>rq|k$!>U8x|YuMfv4SQS%$>x9o>Z+iEgNNS`dDE%%M5jA>zwZJzaEsb~Bo?C^(VJU0G1O3UOGS)(!)P`w;U|245hqEds$VGrV3vkLYj!d&ke z+*9i+7#T8`e{}o+w#{^=hqJ?oSW-9h#n+f5_a36Dy~}C!0vB3+o1-5aOUSm1bI>_T zpIcSlM}k9+Qr1X^&QjF_)w*o*T0RqZY@dcfJ8$BW_6$06n^8OnYCQF z6cO&Nm&BH4Nt7BZ#lG`%1NAc#xR9>5jD==4F*$5X{Z3_IXvz-~GxHP^7g|nwUhE*` zX(mlJu;UX|O!%`Es~}>nCKlFn#9=@iYlHfj!rzyeV>$k)TfCo|_(qa~oP4Tm@RREt zswW{0&xK|gpQ*jKBhEIBg^9Z>agy44I?gek6qtJOhDj=X+kOok-W^F#**vF)etM*& zcZl5GaGn&kyMU4SK0JG#C)G>yxja#>y@Tv{dhy*RVy$_PNZKEvpY8p)Xsvu`xn4ug z9e1O5t{)>&p>~YHp9Iu=9)-soeltgG&r)ceg?gnaWJHT3J0RVPlRo?6_z@cT#W9wg zE3G0wMf2|8?~we`nL-+Vwvdh9*~kkXq4E<$qWidx<{!_;npc@%`DHoXb3hC?B%kF< z(@qF_D)%s>Tn+i`KP>ZO%T-!oQ9^V)7(8IO7{@Pf5^`;O$d=d#%uxvk`oRzA!S2QQ z?#c;n!RO=nV2vztO-1AsrwT8AzNJQ&RrsVsNH+a>K~xtf(wO6_q$74JHZ&$O3okTs z`tm-Qv)-FdCrM=Nm1nelY(A#%pNlOo#?rGL$FXmD4ehxs&i7#O&YhOE|wzPdK5Vv~=+O0Z)8bpeOIWc3dXX7y|{{Pau27PxwnH5#_$WqtvAZvM-?UQ^FK?4`oa+tcvQgd-%eGM1jrQRN3OX!3J4 z{v-W+)i6LImTsF?kIF4Ws5t2gcd%vxCs@%X82@V>nnpN-xP=jX419_!RouAfqHEk9 z%)p)nqFM0od*oQ|ZsI;o98&%?(-{d*Vd2yxkQ{0PNh`Gkwx!YZy+8#E&*h@FQZn@l zyU1zZTZ>lyCj6YLuVhNrU9R)sep)EKlJ0zDiXXMlkWqVXkP*k1kmsdGxcoIr%s}ya zOw=01t9+Tx`x~qy+a}%QoFr86R)jC3zFL*Y{I?9=y%``g%^7Zjnk*JZxWT*$zT830 zSB#3#m5$2{qGf40+~2GLoMF+ABVAmP*;yeN*}9y2SQG$nbAqUt?sR_7;7EScR7%G+ zD`I774PIR2MzrkDVDQEoF2Yj{T^8+vs*gM9=i1fuQ&IuG&z0rhzg)r%daE&~NFcp> zr-jrECKI`sYmANFCd@e##b3!;&WpeB0UNV<@VVrcU~*t7{2OkfD`O?`b+Rn1Gf`)L zd+I>m$4ut%=rwfV!^g~^tOpVsGvTT^S#-^Td|_{(FR!MHCvK#X3fVSZmH_4}R5 z8s0rA(#Vf8;RTGfW4jm86eX*)F3|%US0UVAI85W@7NPX`&7eFh8B%t-;rG5$@XTG! z&I~poaeK}{Rjm(X2bFMQTMse3(m425(I(;vrD6H*`|#rUCQ?#<2JZOXf@{lJ{U-g$5p-T8Tk2H|xBGC0$DFBq?R^oFK{qT$jN-(bi*^y_^H)uA(zN z{w9p^IK&cv3wF~iJ$hx@0=BPAm94+81SM_r*nL)4Va`c&NQf#RE2Ix|yV}*M#yu05 zAYB8+uX1Zj&nJ;ay^-wfrAw&ziCZwW<~i)&84KKrQS3aQBiy6jb0F4v0#aLIVXSNr ziz(-Ea8Dgd=-gsX-k0WmyLouuTFOqT(_tTY=aaY(0`}0YnZ!i04f7mx_E4YGJ4P zXL!&MDg5%M0yci}f?Z3K(Rbc@w&O+#|5<&y8L4wV&Kv^&2K_Y@ngB5iA)$ktM6Q5a!)txbS!pbmW}C4Q+c^c}oqJ z_d1M+o!@h{XRC#kD;pqUiw{{pFG6tT_c;=f6wj2oI}7e?^%w1Arm!W!M(jHoQTQoN zqdSrm`SCL0OrL27_=ZeiCyBHulfzosuWdqXdsfL{%E0!He{z%)?EgK}z zI;fNW@RtIwxmfeLt5^8IVhW5u+XaDp{NTALH>N#yGdbFqhuU6pyxd70I@mOtgxe-i z|KN)xXb%G?$DU&3u1Z_AHORx+nkz6_RR$coblB0wyC5U$7pRFe?%9K4d_$9jFj*>r z9xB;Rv_y4Ka^^3v{n^5%EVD($!R7GF_7;=|j^o_I9x@MqMqqHdE2RCi2LEOQNPCmZ zy{pZDy39T7Bac)OGqjquN*m8!ey|hdXU<@Yzb<1RKU~2@bV|a?J~;|K!-Bn@^RRvX zL>QH9!Ahvqu)%o+?CGKpAnt#VKHHv*N@w1Idc7O#$f~m&uYCtG-=E;JBo+Sa8X|jT zW5`1T4R#b1!3XCcR&PNwxi_tmWQtfK>C5WWMU%ywdFpK2?Tu_lv^tfz@P&44TFiRs zo)F!qMUdX~Q?xTM5ie#R#3iYJ>7{Lfn3Bzr5D~k&<%T1FSj4i<^WM*YzWSNO+(^MR zMIp@jahKQMcZzq&iQ;$J?%=Pk&*!(SF695y2!-yRJ{rBpi^$J80q6c354|r$tm3#y zwCTAN@l|)@-dlUa`W=UePSXo+v``ybtenZBbw325clt<_>^ZW-=Rf9i&scKVa3+1e z;SfDAYAQrb-bpfxr_p6w$FWnBK7f;WH57gO53kmah5kEl=+obinVY*^VbkPHTBd3V z8ytq91Q-bZor~Gw-?>dIYe-tpBHk=09@oDJM}gBDk@t-e$5|W`G%pilN3f$XC4M7l zWS@qj0uwei!wMEmHY9JHq^9S3@asHZ@C+w=@#>}`#%AV(w;lwvYe}yurY!xGGJI|4Ikug+V zXD�Cqw-ElVJA54PMo!)8Eq$Gxy86eEk z2Jm{i{lE&GpWh&MZ%t>{b=W2@#Vx7gZMgmiYA)~UYcJDx;y>Y z;-aTSBH$fH$2XwfgD}`EkxTOJqP)dQNau|K2g#03G(c$mq5e_zj!op&4)5(UuA-z2}eOma7~jh8Af;-E+d`f zx?L9hJJyYti=Fr_6I|%47(_&0Vil`RlzSyuF&VNYQX7EH@QDH z0tF=o{7DmcI$vCp%X;5Mk{Y6jH@!|r^&rM@i!eC3iZPn{n#l>xCoBAabJeXBno?(i zj?)b+NcG@#cORrB{=MYcL1pGEd(V1a;C$ZKeGDI(831Ev0q?lm5l7pNh8aIf_!E-i zBoZ;oI?Ao*4?2ly}d-3^QcBj!1rZ9CZ+%Xe}GiUA4#ax<~HkZe7@0=n0tTWwn z`yajj(v3NFas)Fv;w>bvvZt^t^_jTi5Gi)uorIbAItCDeir(MPcYgxTXDOWHrf`C zflcBjTyj(lF8eQ%%X5t+ORn8S=2teBX>CBKr;q4`MT<$lmK+rRvVr8vQ{ch*;FIqq z)Uq0}=9xS$(%+8$#g@G9?o>SLaE#SCbDUjnZ2{FbxtwjtYGjl~f!Gdle7!50y!w8S zrh6#jJ>H4B=oiqN*LTCsjs+}x`Y)?L(v-Cgihv5+OuBp375aC$7tiUhpmHK0^0dAK zKViBwFX8@8F#1y@|5i8;ye*Ey8TrK`KV&Vo=)7o$@sk-|wBAX74MpRm);e<2+>m~b zKM7GvO z$m0RQphP0OU;H(kp&G1x^&yD(cppxwR*}!Q??UPI@9=M>KHDE|fxOaOKGvla&jc1x zeJgV~?f!@e`mQiibFzv2Ko4VZryql_hG0jlGN|RoQM>$Z`l8+s-|nBmNCrRSf}3jS zwALs(tN#!xOh^9VCPmD+KaT(SE`hJHb_5G@oqqo&r1~qrVz$z9?C72W@cR$>T>cp} ztg>L;g~McGj{!}a7)5^<>2ROl*n^*_zD<9yjXG>R#eX_*g`Zuu0nYx|1{aSF5|@}h z?si}^)2mWTl$}1&o}Y#o_}Y(b>`bR`s~T`}-h4j$r!haU<_XO?QUvQiit2Hn27kRy z9&C9F`bbKL&6TQP*1U8=a(yN4Fb)9s-gqLc{Yx!=&m*gw72#*T5+ZW3@t5+K2=!bA}RA5McAC$&fZZTYCzXEmJtc6qNU!?UVCRFYF z9AYLoORj7!qM{rXvdsD=aTEHo5pWAoq2G5vxzQSNLHY~LDA997q2ZP+d{M@!!-%b*ONCO6WDckiL9Y&JfK zZsC5o1yDViwdBwG7<#u@9}-s9lZ$wUBwZflnCbC!{=jeYhc$(%6a660a0V#eS0`H< zzfj%1S8>Mue4_5rAYu?(1d4xndd=aj@acg==(;osD^KJyI=|h7%VztMhRrFobKe1M zk?zIR$NKy|k-mBCl{v_@oTO!kZRz5t2KXuJBsslmEq8d*FS;pMlJR(%&Udz&^TQ$W z)X0e?Lc8@i@No_9`j02~CQk!aISbni^dV=i7uG$o$N4HMsPJBsIy`Y9XI7ZQ^9?5C z;1MfQ(mD~FTP~7~r;G8BK_<3t%E0m3Z&3f~V_4XqNKHpbfTGqNl0Wh{y>94Ans*Cm z9BBo;yoK;6#)>}|F^!MDGlk!INs6fNG$fw8131Yqx{x^ZPv~6~Nj#p82bp#tQt1!L zx5}~P{uOno&ZuXe7uM6WdHXT+K}b!__&~DJD3&I!yh5Z)cHo+OhSWZygV|=o#o3W1RJWBACik1c z#*hT+dsyU^e`*=HHC1pKr{w9=gF*N}at@~{DNlPw#S@h~8__aGmEPa9240-q2?Jh= zAmVlKlei12nj}GQ+AA(mCjx|To5`Zb<=`Se2jubrADSfN$|nuj6gEu0xc$T>b81ma zaUSg#E256&2kE?+nJDIv2CLdd8dtjjPZfyt!q^CWvqOw*T6lo$-8=?K&0bvbxCdP; zD7xi@FspTYN#K7XK7QL1`c%P()-TK=iMq+c9XqR-xz65%ZhcLE(N#EGr>MGg?r4bU z)j^Mtdt6PnG#v1-0JJ|0FO9@Sy?hnD5Z1*UOg1FZCcgy1TUfe-|IN9qPKUR*#;~hD zCxTyS4X2gaLXLZUpq+}rP^&3P_C1}(m_^Cb-DXR%z)MMZXxNWa{IQOf^q--h#9kB6 zd^;w!bqrQ-@?odM8L(2F31B?Go}|>iqlMe0z$GPH5GZkw1c}L#fNfoHz0etJ<~lnlLjuk})u|!q8QJh|Y-ya&ssH zQ&NZVz~T({ScfTnUsZ_Tjnx2G`GT?iVQ%4as7E4vlmt%`H61b2MPI zM|#r5W?9Tii*`D4S|wGTNMTouF&O1%f|Rx^=W<~cmAkix>=o^sxIOkINnL$Z7+{TS z%B1*@djhcO#twE=|4dN*U?9Bg5CG@i{-RrCvkCLOhx^=RK^x|;CTWdftX1h=wncrG zu&Fc`&!#64UBOtm6vwgpn8@yW|A36zFTiJ_{kGV*q3mrnU;G`k0ad0?L;dv65F4|W zt8UOFThu<%pb#ILtap?QUS9-99&E-xKkHFrVmcjZ%?bCc{7B1e%!pjz1juNaN0^Wf zx};4N27movY{o0Vqvt-P>R&4PY%Gt}%AaVu)*X6f|0N>%>kM_4RilwU0)CZ>n zqqZ9tF`tfVFvbaq|6}Mp{Hc25IBsMnGE<0BDkD0NlC0{U9CquM0=7TGd?i?zLIl~ai?p`UZ(z;Bs`u&4P1zI)X$)=QC`UNn`R{(B(?ZXXBEo~f`R$p+&T($M5WEe6_p zQ7c6YnDea?jJ|4!b-FLmST{2=VMP)yzaET_PmV#pa0({v+RlBuwVHH=UBLCvOj+-N zlWdR6Q98~%C6^7QFy`(PW?AeHx@(sU z?Dh~tZUp%Z`6V(b0Nhw;6-EzLZ6l=R6f(i_Xwag5Up^fS`N zper`Cw!n;U(^wCW%I323+h*d64g)@|Je;zt*1=qJ3p1z7{yanfVr5+2l>@&ge1|C;Ls%mRMRvEjC6s#a#^YJ%B88*JH=_$?MYlw$pLje=Yc0lVE$IDZ5p2Dfy5p$@;D@ z0oA2VL_Ynk_{5wO%xRMmgxzLMm#S58-KFFBso6(Kz=S{C2Kje&o;D*eylJ**<p^d?i= z%2K8Bi)8nVQvxp{9@BSwf@$3q(s2Bk_~-5ql-b}26OyJuO#B7hyEPam=TtMLank(o z;m=Sjcq%{nyA00O&qjyW%`|96E4});g*+UTN1LbLh`OaR>{NeC4vkO5<+t20`Km3J ztn0>j>tu9pnvbh$?ojn3lxTghq`#*ca?7^K<4w;G+_e^#_`F;~MhrKFh_eoY$1@I> zSWO4ry;I1WPm8gx#u+7?Vf+)!Zn4}|tN^7&ZYR=QBGpKH52imR=APklCvxa&L8Xx!EWSXdAQGBI6vM)L>l82gvb z4#?!jZSBD70AF0VHkq91n?Ti6^GJS|B7RKwr6cvcsQcapOjo`;*LY4(Y&f$T=E$ia z>#fcto7%%Yx#e(Ur7RXbF2o-nKhsEh5EsAsOiaE>u|XRhVD^+CP;=>n$5%6=Cj7(O^R)Ph_7z;vVHJp2Y|pHshj7!YFLZ6SC9}z^k-B*H(P2Uqoq<_`cx8_? z*A{3M2>fJvf{V@YD35E1iY_}hd(yz{I3Bo zeAM2EH?LdJpBuWV^`|0Ydm|2KjE>u;kpa!42v_ZKHK|>>CA03VL>*1 zEBTQ2zmJ5*_UrWM)q40L*28riUePm}x;Pb4}wA8hXJkmRe z9V<>k0-k4&!5MZ~tR$A1<)FW|E*6aIqBVN=NKj)ecE-k1wdbrzN5=~q?yBLq^QURq z69WX>gE+@c6&3Vmki>*IZuHgJs36pmv#ze-KR!0*D?9C}!l*tN($EA&zDlU1bxG_z zdMmUanFgP`wuq*k9?m|Gu@=p%RplB=J#cBHI`jsO0&n4*S#|6ueOo<(+nU=$W2d&l z)9<$loZcgRcOitu>(M{;PB;*y!fUs>L#h02+DIhG<6$}ERJ9qgC7fvgi0N2ZF$Py_ zHE~`Gl4!@vJ@jATaBg2;G;NfUB6HOm1v=tTIziC!-#M1d+<$=Z{c0u{ta66cqBxTO zGXf))ZbCP64|0CQCniIsmJF*o3U0p_f%KYL&@Ho1lvMJ9sj@S`DBIbze1SZ%ns633 zB?b$3@nfvrJ_`8z@|ZtU5~?~L;)tF+;*qZmZTxWZShb(#u787*k2G*mv6GR{dryY{ zJ%yTbt;}O%SCm{i48xZ8f@-QTqwrB@7mrth18Y)=|E$a0^}Tvvz2FTg|FsFnF{xl3 z9|*FZ(fGY*1TNGaLH_Ppi7#IcXYKc?vX&c-*c}ZT^uwoJqFfpUkGfOHwCLAFnz=}3 z9UY`!x^i(}?RI?gOOd6faWLUtK3%nlr$zULET!YR;B{~#4e=Yp_A3l!w{O1)|8(nN z2DHM*Ur=sxYsR#YhJT)(q|PW zBxMy9uQRB_ZXZ|}+zuh%1L<^^Anvx(Xqqwg5|o@M#XH|mkhcn6SaNz5wpaIYdxHzC zrtd#0(;=}@vjyxF-GFS(L-^mK1+;Oi z;I*vPK#SE`_|l{gCM?Ouw7}oYozyXcCoU2OVy?l3F>AqQ!F|Hn+c@ZUowc&rN}sQe%TAgg{~Gof`$1}D4Q#(?fe%hc!Gp~XcvsSwzQ{{wJQ4<| zM_nEH`~5srpVK0HwJwl@+vM3m@13ymmJhbxQsW2KCX(^*LqShMioJK-38brlRyOaV z0ga8MdR`n|@m~@5$;^UlatWtFf;V_Sqm1__?V&S%FQKZjCDeJamsaLvP}9F&s26uk zYN4w3CD0!cy&9Kw48{Z`2=IH8nN5g$+N$6!}#I}JUulz4lAN; z;r0T`tow6{q^q5QO~HYbv2%nUcgjS+1+U+ex&vt0HVI$2jzqOGp^tjfjCxxw>0qp} zu!pKgnG{RWpjR4d?<~SsN5 zyv>V2>^Rthkq=k$|EMP4=^2UFM+%=2Q$%mxtt6YT=kTqqp1h-KB=5D9w|cQo5ce34U^@T5y}%WZ8l2z71F})S4kXo#^?kviR>mDe>8# z+^zbM*o@f9<){56{;%Ir6`LP)9_t0GjT+AVlrST+WB!rfvF_lNZB6bSdP#h))S$io z8!Q&MAa?1Qr1iEb^UZ1`(-L@(^a=LWvkmG*x^^aAKM&yL-NjhZM$xd?mV{Osl0}*d zeC_Uk$m;CH^to=lahf^*d1g3nb;-u;iyuhfd{b(($^*YUZNd{7!)U6&f6z&8V!WvW zA2>3aPP?jxV^XB)`OkmJ%A%)Grws5b@d8|uPa%PuCgGIN``~6m8qA2up+Pb?L@jAI zabMp_Y?jbv4vTvk-!J`i(WsM5;_~}-$CsRD-1fi39V?aj^D4io99ajO8`qQ26hR}7 z&zO~4GiYm;D}ITMgz=+l$i#E~qN#7r!0lsMVy%!RpyG7_U;Ug6c3nlZe%<18Uu-9X ze$Xs9TCx$Yt2}1jgxqJ6vi$Jt=?<~QUJWuKA{D%b*w8~Bip--rWB8}p#_U~_WR0mK6iYCT4TMycY-r5d0DE@H!M*P%sEl4Lr@3nx%Y3gVd4H6M%b7HK zd7l*hy4->u7?Vt)Rf{gLG@{oFBI(o73wSoH4@iJCt@&<1W#?YSB#Ymr6?nHFZm|BCp<>GVw@-ygq3!$((0kn#i3Dak?l$a**i&ixO!yt+muFHc03 zEh}K6uP$%6>L|2Ss?wXiu6$%?5DqvGG-d7R8u@%iRb_ z{Njk$TGhDN3@woAm*Ki?E9u2$vEb66&D!?`v(t_3*m1EOIZ%+tmF1^H#v&l=v(FJX zhkl~2y^8ELP9mGP4ssF)&etig7y$~6CZPX)7Y%S}j0dqwAkURJTZ2ZYEv!E|nnh&;_?ce`MAa0uJyjt6*A!+aZ_ns$PXNu; zUrF_~5GHoJ5mB!gq}x&z*fSURfcnxph@1WpRAp3HZHIjj{UceFwN8a#W-ZF=R|5A%isFIzRL1Sb8M-;^ z2fUY*gBczIQ|nelotsZFci48B$kF03UWzsayUfSJy>=B=>RGT?8~nvaimS<_yl(n< zX$d|o%OfuXD?vwo6rDTMl>9uH3-V338TIp#G*QS00|_gJbqs)!u9;9}-$pJcmT`w} ztblnp2N;i-=cIReA{8%6M9+0QsrqAoy0lXozQ#mTqsu&bW@^gptg^$8?+pY^&0*^H zOcOVhr850=JuUi?!9?vE1t;Igk{{D%66KQd0_$ieJr}TtJQ$pggHnGX`HCYA@L@=3 zb`Uu>Ld>n7pb5E$3y96mUJ~Ogg-_m2fwQMQ;PIJLMDJ@T={;NvkIQ;R_A|Y3Ff);J z^VgwqnHT84KVQgFi{+%L%@VdR&!CPm{dIn7f2_YuJ_XnAakNfRor`TXK-*_x2>R_arxs9Mnm=u{MwWPtK_z{Y;SHM|s>r?%4N!Z570`JR-1ppr z;CcQxefP7NtDH2zxzx;OUN07O`k~LEJ6;tnO0w}!YzitWMBvTf*UZFSVRZMaO4^!H zPEPF~$p(Ge&R%%2md#jwiZtc%^rB0#DDsUByLR0y*4!P~j6P4sZu4ls=M$)0Q7J6z zl|q#mHTab!!A@NB1jc%YLHd#=D)Id+4y(V+bUU79K9xJ$zR>T`9Py*tR# zhN%LJ%!Ul9mSaCU2%gG~L)f4D55uJC6VAxPx3t^4X5Qr!i#;SOsKvsS(`E&QqQ=P%l-|dDnwu`esBbz z&|}sPqYp4Ae&%rxwrjwQGM-d!HiSoxKgrteB%aZvvcSzW7Od{|2VnZnk~5M!MyAv$i^=b7^19FxL_5>Ot-n^&D^fb_GL+{eij=r58c{T$ zru*dBldGT19z5nO|Q#W%c z3sF}Y|1LFZ`*sIc$~NQONM%987J>bu5N_N2MdII=3$5SzYy|yuTX1t4APp5O;k{}C z5o_wg5rY~+73VS0dnSMirvxR=u4JHb9%DYKP4HO75Shcy&{^S*A;~Rhu3(4H%df-A zEERC;TLO!xX2Xos4InvAhq^D+r!TaQF`oBA&{U=gMqgCq|L%yyC+_ATH}x*rzTpn} z8e#zro9@$*ovXM_#uN%>jsPeB7jQ#T7M>1WMk411k+;#0xzPgO^~$)ZBRx+g+xh_UbN)^}I*czONyXK{+t~hz-cV5O!P79}rImZeGz?^2p{U zv)@b-@?~DY1HEdPXlp`8oL)x5N>pGx<3m+gHF%*jnk=@M0`+#2*w80&>=DyyNYc$G z)V~@`rN0VuJ6E{*%z(Y-;s8233H#Jco4s`C2W)syK$4Eh!rx6=tVD`4^Ik%ai|KCS zYTB2Q=+T`dM8}(uYh^Gfwt$QoaYSL`Q7AU>f#wPa(lPQsaCkbNb*?jD4Iik}1>dVV z#a$`PBAR{f^#`-MWkb$bPh7DXHe_Wn`)~CVv;8N?#-hzIX>kf@2TX#(gsI^CI{*}2 z+}S0iqgWE_4BuoY5j#Que}cGR)vi`DRLzF`svpY0pSQ4bTQ>PJaxz=LG7Y5i)meKf z4n~EafE#ur$hdJDWP$$_`aM|&YQE_}M8SRXTdk>Q(3tbdePT4^p(85u@4y$v7_1%LaD=1pY8I6rz@btsCp z^+|4#9yOc6!il4k;il|J_I!;a`X_y7TKi^*!@k{PvY$%P{gJIC&Qh5kaCs-1t}f*1 z?Z0k)%A!!rwr{4|GgFyksaj|vX^Y9~M&Niohyg!o{BuK$+_Cv6HvdH{PrN2m1jcj_ z>7ts_EzI^a?OX>xgEJ8Yh&yD{q2PxWUzCte&&W4np{y1deH_iRT5A02TRL>+ult0t z31L^1k&AeDdrich=mXsn0~rOC@{QHZuvo3j9vz zphBEBI0vVwD?_K$ASo%I3IXl&U|6*!O-gC!<{j~2(+9Rwo5Js;-b@;TTiii*ts!Uc zKNFa$VQlTRNszs|0%uyC!3XWLN&JNs=xcbGhN%t1yI#vhiGxUMpBK}d*Pg--CIpV3 zzsq$wIf){M7E)c=R?_n*jXP7Jhcm_u1r^a6$XGTEpXvRfLzcehJk=_h8(x-}GHV0z ze>ap|$`>??8Y-lHQX_Y=dmjBe>!i5RUzWSwA4CIU6~RrIIaHBj*mS#^n!GZiM&+b# z{~S9mJy(etwf6*CI*!9C6K}A2t;uSGC3`DIOch)vaA{lqiR?|6K=BWO`@1?o^iiCN zox#d<$sBc+qQw$!=Y^TaCmim&1rMj$awcxkq6sPM#Y@6ZVyBEFIA42A zc5Qn`xJSlxK*nyW5$8|GI@yXo+59BVGbMQbQx5KLG~nMm4Kf}6ZnUq26G>>s;0b-Ak`!?G|6sE4K1` zI$h*vu!@j%k3=V2a~W$vCz&z-FPA@N94qcz3I3O6vX|)-(jcT`=6tB8zH8S}x$r#t zu0<026i#4@gfonsR!FZ+4a223sJxsb_|NZ<&0lqxx)OnRDQH*@ z#(yKzTf@k<(luoL$w;F7{Vey=u@|1YEW(>(8<~eHouXf5kHp-mBrF>L3`cAt{IU@e zyjz1T^y<%o?{0_0ZkoPy#(*WcAu#?vEbirAo$kO{djoLMv;B;fr9LDKexZ5~)0wO7 z0tabvEb;C0$Bsuicqrcl6((9zm%KFa+F{Ih?^Nf1JOfm{WW-9>pY;czpq+`+nn;f5?HOQ31EWc^vyG5tzgPhHZF}4DSRMT1}WA7Q7n)Q`T)_Qmz)$ z1>--Fi(8H&{!wSA&KpOZ9v>sKqRWZvPHp^pxr8!8ZSOxfiOKwI$Y}&`0sMZBao(YV zhbtDtsW~0svowQg+i{u0gAeG3GxDTLl)-sy+JWH}I~g631RCa1=KXIu>i)|Ric|K( zj^}TQRoHkm3il_9lkPBidzxt8kR-A|^oL7Wpe+{$!vJR#}l1`Z*cK|I~(-q zKNvmH8l83Tal6ldCg*%d!}M=s$f`q^xkj&5kkAQm_wykb%{QTrN<3tIi~@D5;b2-& zKyG>$(7F4BS~xJ1-rOb4O|!hq?C|rWVe$uv$HB8?b5h$Insud<5p=vM&q zB5i}fP?i);F}X*UrMMFZ2|Mf8!zZI1J%yGTPpoAW2Z{F6dio;zJdQaS3-;MUZ($YB zxHFk3ZbrD`(Aq@d4CCu!$Gj)po5j6Rh2T_|N@BvMf zp=|9urgVM^7kuUv+HFZCkK!kbF8SOh{y$kd^@le-dT2f0t7k>GbuDLYG@gM((;RkN zN(y^wwliqnHef$RAH}zQg#9-DAhw(~z}?e}(I;dd?o->xWe&KIZ;91(+`RelS?Scce#FQ24`EKVJA3L>8O_l-g0<3_jE!C#Sgd?3mLIW@&K+0_Cl9qUztjE` zI|Pn}rZgFFdn}3Fx%Y{VbOX2TLN51UQZ^0Bc>yB}_d$8#X_#U^6l*MsiRr6({EFyu zhP(M3t&gpy2d@>PZs-fNcx?pUSF})3qXYH&K2eRVbwVCMDZZ{7MVj>{(L$wkY@CyT zW0LInn0f2ydGA`zH^2tBtIt8hgJYmkFO~TrPN(&MZ*x1fj-%%qP5dOe4x@Jc6YBji zsxnZ*jj$R`p8TAKd;KmjTXRjRU5F!S8Wm!EuoaeAG?EpEqnIOIljysr+Nd5-LN$|* zQyTUhcTQgp^%|pi-l+=B@3b;2{M=!D)m3iEc_-}E4xzo7t~5unmAcK6#NfncRCXEQ zii}R6v9dRR+S&^Ge)p3r8Vj*&gA|3|rJ$eeL+%vxa~JcHX<39kYRWg$GyTb^qofv7FYw9`Nq=WnxbcnXP5_qyGnRHGz!<9;^;QQSVHk9q*ChRYSg9r2BF-+m; zB{x#ECW+1zse;Pp^YzhK^;kS%`P*3b=DvrY*`J8 zL*(G8{6q5RT>}<=-H*TCC!?(4QL1s|si5K9f`6)W`6mHCMLMzoqmz^H!{eJYY#Glj z^4bfDEwZ%WzyaOEIjR*#qEkJp@~t3qVw`9v`_GjN7UZu zM#-MWlo5^O;?arpa%KP;ZJkO}BzlR$U_E^59Rk{GLeX$%Ce2>mf_uBmMQtN(z*g4* z^LHfUn3Ixxc8)H&tQgM>d#a9#w_kC4M_TbCmJa7F?#R*9AKI8Y%9dw>kK+4(Ih<2M zH_h8nNVQcGLD%k#a0z#lxp9w}%rY&Jj4+o!^)nhXf^*Sp_#;$#m?`eM?S(#y=W%A{ z6`C3(g=Yj#Rn`beSeW^PPU|iwUrV2hb@e9Vj+ftQ{j6cUcaJvj{dxo4<@SJ%$vi=| z1wP;TAHmE6iF9E$6hVEr?czH--T5UsIV3Vzi(e6c88{c!u3cFx35A1AI|faXV{P&0f#-_J(k-zs&yE_I53 zcCQcRzNq8H$<0_*exD9}JxeThBH7=W#u&Z$MvWJ?lXHPf$fpx0#rK$KZsKGYTxu^R zRxF#(pHewaBX=;ILZyfdy=8`rB2%GZ`Es~*Cznpw6!ePUZD_!sY{tE2Gg+q{p$c%_O!;$GIn0HJM4j zj?j{~iQ+124{F`_l=w_fAo){F&TZVo3fXo?N%q=!y6$+FsB_3}qP1-ZN=9_hj0B*Q zl_cR>5~1m*)FDspFko&n{@Syds(lE@{?)(8yIrx&*c~|{dwn6q;=wiMICGL*n3>3F zt1pKt*|k=Ia?MRmbFu`QbONvG1lvTa5Nw;k$Y)ZNTOxG(+8(|=v4c$q74U+ zQ2%>}Y24@6bm>rcqI_2qSKP|y%Nw$Jt+&Cvs_@%P%JimgKOd8tmp_?NI$6Z5KZ*-( z)nvX*xhg&(WyDySFN24*`|`Mm`}+o@V=f5?-Xsw5B^GE&fhek*MG~BgVnR? zwdco~X^RscNm9%x0plY ztBJ-lV^}@J08Li_Oh`_oyXFghXJ0vbeO)K)%?ro{y$K{ovs9d?kN|Ig@@RNH9s0C0 zX;F_I9k0>Eoq6DnzAIi5d6Ohuqd!Qjd~4uRNE|9X(Z+{=`l$EmEY!FyVw{^)2-AF& z@frV>1h>y6w{J}o_s$FxZ8`jqtnFcfSX+3sm6z zbU>X;yYZ6I7<`kXMQz6(VFI5HkVeVhb!BR^iOJs*Zl9JKt7^}(6IX|@2R4hT-w`vY z-S~*C?m9*tjAwJlmZ~sWOE=PvkX)MVx1ITvZcTNAEJ(tO^{~q`jOc!TMttMF$?Ljh z+_z*YCdqdV=v|1wW3&{vp1aJnkUXmVbR~%oO&8CZyqRDBER9#V7=i244$#9}7?ca~ zAg**Szi!O{Dtu|8mU}*vs@YFSl)67ozI2!|*i?zNG80fwQ<1kE_6+}af5Y>U8902~ zErFZ-5HH9U3K|X-JR&t29P7OJg!(aj;42;ee)N3$q|1cvWc@|?hlXMQtwD-Hwvh7e zP#p8BmTM`GacXHKmWK>%qNCRvr+=PGjGXu^!=>J z``qlI@>%ml{kR3cPfO*_UC$?mtw#82o+N+pk_N96;0O9f7vR6D z{|H^%1V+KEPhiR&fi5A}U2%OQ*+0DnCNCU-8QZVY=@(*fshkA%h>iJ{1*+)y#{icZ z$?%?Dk>v88QDF9KJ2}JdA_ZA*MBVl}bjIi@P~D|OzML$=k3%-_GwqA8KKqPShJy(d zbWDcbn!fn9T$vZ?tl`J(oDF$bQs@`fl>aZcf?Bd0N&DnL;;iI`HxI<~ulpABBlPFP z2rdgp&v+wv0%eHSQw`EMVjN_SHzjwPC(-gxN2$l{@3^OVkh7*TSN z7->9UlwHQtggj?ZUfsu>Gsxva9 z5@cmZ9O;Zq1DW}&c#T!Z_&d}4(es@G8|*w1Lp&Zao)@R#zR-`%l?pHNM#PW>;v`CX zBJlFK7tBju5qX!JxY5#*e;xcE4R^gRD!Z3PuIFmQVdt^%UHUM6Yu$v7Rg?H1u0M$S z^d%(txQJSu)kh_R_l)xfzK(Lv@J8Mq3VRXS2JLClAARxGY!Q~+ z4Zfi{Cto&OBy?_mj>o=h70jC*g{b~I8sUyKyF=VXvK zOoD~qrD*G~@7(UYPp~3uG#*5r9Egkrxfwc`+qRuPD`R=@94~Y|m&7Qxe8vu~?^qI% zPp#L=^P5Jh@P|ErG3Sc{KvQ?vVH3@y2C65b=qe0vOassY5!rQtG^s&0)NwCUr(ae zbX{IaZWI<*$8cip!?d^U2$IxX#%jU>-1N1U@$jC&@U0GD+3JJD<}_W^{vR{)>=V+} zt4M9+M_@pSI=@~V!yW%Shx{GWPttAQ)85)sc&+0ZE>`_cBrfI?mC@?FPV)nF=~Qt1AI#U!O^{CK~5+@bk}@hp|)q1ymU{cEw; za3l5?9D;O{EsRK~5EGAP;H~5iQRdGm_;&UTG0^#kHU6KF`DQ3&U_7wC)0>O#?LP&d z*=6$BHHEg<7UF%Ecl2r3R{pWxBV;#NqL;wlJ@REJ+DuyvrL|{ha77NCpBT%}*b)JY zoNnUt8Bb8!cRA{hG@?IyE6K1UR`8^ABQ^^66rFbtGgTMxX7}HV)2Hf*8|Pi8Nkxr# zDo-6J>!=}`+rn-t8=I8S@HSRl>XIqp^sW8l8O6r@Qq$K>GaIB`jer_zHNyE z@@HEbDQY1Y)%TVxWp65sB z-X-A=O2IqKlm9p=lppiZ2p7yqBUS4czw>lMdUxvleiZE@=VSUhlc}opYA1K=ZDFB zrRO5tV1AW`PmyPPUXQ`9XC3$fL)IJ66~1DSX3^vv`ItXk>G z-4t2Z9ZKCp>&H%~?s=!cVof9T+`h=;iTzw>X$l>h6NP~e?XdLN7ks!>k7v|Mt)s?H zX1+$vp^FEqnC6N;5*Rsw-=(1>u5WW7{}!{T`SB`dNbaPwhuPBV01Z4AdJ25U8$owM zGBZj2DSjO;Fk5zM(9l#ld|N@a=-CJtesqrvI>;WTTTVWw;p_I$l@Vugrs-KY zUI1`RFP+}+oxrzay_G_NOPh66#-*t6Lf&1OyHUzfMzUw0qT>z&KN#iEo(JZd8~ z@pCXDt`T>PzmI-PZsX#McKnkctH3<@0+cA+A(A&V>0r$%)Jr&q{rb1bmGDaPtL+Y5 zrS*f^=Nt|xH>B8~>s9c{RmA3#E9mn&HM)1EqQH?LeC8`x-sF-KKj!E%{5(1eoarIh zInMxH+-r#0=?wBdO@_Rj(g;2|?Of0+7m}VUXlo;6LDBXH%wL)X8oK#ZN$(Vfb@qaH z$YnCGrjjZNnG@w*r_hy|#5YJv^UW6?;vw~Q;>Q(Hs4!20x8G7m(~l=W#?URCQRX4C zY}-2_(`gDbMvYDv>eW*a5;JJqJ;O{4pVym6n9Zqp1hr)1&hmu zK;Q9^Oy%V~!MDw#!yBLu_{!-+Zqr)Cu00<~19atF_TolQT^G{j#QI*HAfIAXCShIpKw#M(pvEBvpqgTVvk5eJ1`xomuVjZ8~4SN9nZwf)!x%=yJjitd#RSaQ%ZPs)hFDX6N-4-rUH|d^SO5i z3`jxxLuRRO1=VUdAVJ>Sz}h{OE*5e{I|pV$z3hCZbc_nwJ#{nQSN~0QkL1&$R6{mv z)fI>iS8IrHvh4ojHptHA{dH-KIT;a3@o1PYtFZU%orVL@b4@QHQ=(R9g z)Th#m(}}dMK5#wKSg5oTl9RU6jIikfS1Af-|I8t#m#)zm`CL#~a}Erbdy|2~*N9!a z6p8+u4eK|IfvbBp1#eFvv3rn6A|KQt55XT{7@u_&1!gN(1dwN3fIB zv_R2+DR=AITCz5$2x|r;aooxAtoz3e(9)*Grsx+Cl}|s3qKXJ)W-#*A^|C#Vw{)hXmf#9-b6XgAPaZ?qxy%GsR25o?%m&XU+_xyM7Ir_ zc9_E44QH6seg+`5Y6%DsZr*RqUpniJCYfF{l}#(^1k;g8%=Sm>AkD8NZ0C6R zxT=bd|8bKXohiw_87(hpLK#Pz~+MM?1qn(ti3@Y zyVB#2`0%Tt;3E7_T+MG02ST}*;};SBOaRv_;z@Xy0x{V=kzHlv+ne}|S=aa- z|4MlUCYrbFbfB_(RQYo)YNCbjmf{*EW0Yx^()=AD4U5^u#gbBl8!O+K%GH zF~QvNA1fIN5z;W{W|}#77pHxI=L(4*p6)+P59&(uPG?Qf$x4eq_3;H~wW||H`hTN= z#(AwR=ytMz_T(z!@8vsbbi)`v_l^x;r7(hzJ+Y5} zB=AZ69sSX`+Lf8MDUE5;iDr`j9HK`@DTp3=o#&VT8^zmL&gR+0v2{Clju7897jutN zFA;Br88o;!O+0g(z~Sywf*HF-C?g+F&lwmIQ~Sdh6D7^Nc(l>DSbgz=0d4AhUz_)j z0RCq9dsJ9sMEPJ%(HHBJjLJ(N=D*H<#j7u#mh+hmRmp(l;Zp2!IUCa889`ePP|~l;V@FRDCiT^jytX5C z`_gLb&RgkFom;iE^Hn&#dwM*r9h`^GHvQ!J^^t5=V>^?%s+4&~)15}$Nx3cAeit-JiGjw}ya z0%z8`K~wTml$7YEO0(a?lGGHsD*im<^vIkV9_c4xek!mvRgb^@!-@azC5L}YdYOG| zWQozWxA=URvPjvo57%5OKsld7jH%COjQx|%mz{fvW4iuwE4G_c35C0q*;+`YMoaT< zGfa5bQ@2I`X+$!r6*X|f`64ymq>lSfG>c~Rw4#77W!CEL0Q=>a`A4>o@kHnoUcB)s z`QUexnQr%(SsmhxW+?b}9}k0eHvMfLv17;#+18#Xl!6;`c&x9C&NW zD?D~XN86`l=uizlOJGvFRlj5AU5~Je+BArt_UoN_vG{~&mBtOcAFah+*tb>S=C8wc zgK7+ICn%+P5$B%sz?&2IldVFw=&D64(S3y^|NVUq$q$@_CQ9mbjfudTUgSM( z-)H&(??qVG5;<_`eoaR48-TaeXa9|ghC8EvLBsCnkpJKOx+6(np!r@BT&M~X{Ae27 zhCGgJQ29k2^J_?j;9oj0kA*X5Gz4i*A;g+nuv>EVn93bh$Z z_fhc<93F+Fs>l^KP;+*C?++s1Mq$F|RfIj-NPp;DM2fO>Ldaiy<^F>{VD<3hGCktt z^^`b1xB=bE>d0fKuf%P{a_f=bLUF~=WiZ7-=v8SH9kH$jN-SmA+E^_*LG+HfnXAJ) z41Y?}&8NY%&x-7|%535(8$sEMEhKF{6<7$v*uIt_gsDA?JwxlEMDr$oirfW}cdvno zlRDdJA%%Bq^g$wL4|`~#KEz#kP3tFAFkg1+k`uzrbj#xs?$)Q1B&9Be z+uk^sWv#|ppSul#0}eFZ1|jX@ED#UQ#kOQCmZ5K$2QY+QSTPaLxEQea-4Day!vZpE z_&F;36X{uM2?{A`BF!JUVEpVfBs(UM1vhTMQf>gw8)QN05it{*H-@AxyCE)(Kgq54 z|3(@oPJ~JGikWjmQ_w3qlU|75gDW2n!-Dr5{#fb-V|((s?F)mL8O(UxVRQi_zs+V| znQIb_5If;rO0Z2MrP#Tn`dNMBV4`}YorFYoL(t3#aI@nU5p}$v``1liExsw!sFpm) z2}yt??IAm&*z6STgJoHsBgcTT#K zy3=y3ib^K667*(?CUc?eN(h-!Je$02vxU2BHE;(fz)S>>Zr5v1&oE$FHnIcYl+dheLPv;3c< zGYzEbd)u%=rYJ)yLK+MuNjQ5wyP^!4N~1;%RGQ=0s3eq-l**Vnlu#tj+3VSqBBdlD zG%7`DBqEyL{eQooPiNR?t>?M#>w3+slzswc(I?ov>enPtI-l4yOy`WdG*GSk0cV+5 zPc{`^CPOC5)6|f++|b9q#9-+lG#`RoT}vH#lz$hj6J9gj17E46ZZsU8WJmS|kA@F( z$MB!pH_%5z7h!GFM%d>NN@ zT*vSCyfFXaOmO+A%!fxLLAPF_D6*rDD6U^fMlSDQ9t@igt5jZ*%?hKix_S?nbES}8 z3-rg030py0Sxn~a6?{Vzm!U;RFyNsM?lG0&whdaun0<-jM*Q=j>7r@a-TxoAVzo6) zIoeD{_g=%P&7U{8{pmyK(XY z(G#tNw*ohp{5J(}+V{|a&w7mF#w|?M-)gdbi4|A3;W6=r<>Wx-EY4|}IUiMN&JPKi zO>HmNaVM=ksnZ)HC@}eeFYa!{DIZ@GT}2aI(tC$o&X`HQIYgt{hU4(>y&pXKvj)t~ ztTDGIjl6-SP+e2X?ArK8@})K$SFY%zzjXc4KQe`R@u-pDrmvzi&VpAVd?Dz6kPvRo zDbB>yhVHo%FB&s*95-|RMXv4HLAqqqPg41-fa{8V2gjy4lcMDN#HDf{eV9E0#nUtc3pVmv=OS^4VlnsJ7CG!O8EO_DP&2F0VngF5Zj|mPH$HPtuX^6`SeLxS6a<9 zZXw(W@{P1uo3c@Beb{Md9occ`jX}~#;lAJzniImog=Z1yZ}EWq9uo>Gj(uP?WHQ?# zbskPA09pOefL<{ABe|uwMBi?2YqV%r$6jXB(&?U?z&UR`g8 zx7&8JnWB53^fwuXN{uIHdZx3z%9|mZ=_dh}FX8W;k03rWo}GK5n*>EV}Dt&T-|;$MBG5ievO9d#*^4l+U~HV)D{BP z?*=_-UwoHk1xLRgC;gwc!PxKsQuuW)T%DoEZl5uTy}R!`*enQvDGz%gm=Qdt4tngC zS>xGb7L!FTwdQKR_HHCb%&J?9DU5^ZpcA z6FHroKKh$vc$PgjPVgZ!GbTV%$5giT!VVnfsm12&Z(_4Awu90xS$6XB5m+NDbb~8? zih7qH5M1fiFj=#a9JyQx(PN*2c2+&Pq<<1_3%uP8DZ@}U{4T7Tnkzi34vE%AP6yqP z6RcMFP{y#T1-e$!fbxL7=sn&8m zL<^$6kEe#+-K3rSA$&hSvX2)fvj!s(aw#rbR5(Po-J&>%g%kKBiWp{iy$C{)o!lB#&vQ8=vmi#$M zuP}q9lDSiu^Kq=T$ow7Oc-}zHi30pKY*W-D#?z^pO@R$)U5&0BX53 z;a}gaICS?VeAzPsgfJFY+b1oVn30aBze#A%qRTjC&U8L|pYT1Y{y_Sts#0){MWg-Q zq`+YqS^p%3`!qisYde9z|4$lTRr$iFm3B;cZnWgU^lx-Zsxm5%FO-NXTj|e5byRBV zAl#d;MdZ`}BY(Bais#1+=GpqGxNMmjkRu6n`}JWM_U;n%>+*6wO?Europ~RpzOx56 z+pSr4V;l2;axH>lpxh?Zx^vn06H}dDv z__~;Gs#nH+!+((F?fs}JHyFoNUd781MRd@re(ub4FYJpgpfQE>$&8IA7#}+pmj+9t z)Ywtb=x&A+TrBy8y*5Pd`2mtwp^X=Z-J_d+xZu#Gk%*3UxaW8(nKVX(QYJa{p7(0} zB$X~2Z!E&mPk)djUoQ!qkqF!Y1~j724zJnVq@O~9sk+=W?oX>1V?JgQQ+n?j_70&((yTyDGS)7=Yg49vpK-mwx@D%=q}#(do(kc>22n{z#3a4H-LV ztCu0yQxT5ogNCBqK?j)iS_5U%^2A@|gnq|h4>DrCz^*8oz*Mbz&EQ1^thn)!S+K?n zR+Yq~_JlL|w!M>7-JebZ<;RkZLtl_9?)%9DT`~QeV+_xQoadmyBe5nUG&mW5%#AP(Z$<3(dguJ=4t&& z;#Ji^S1C+H}R;xZi&-iH22qaHX=UyzPIl&@{RNHTK02p#&m1Co`7uYB~lXerj;~ z!z@f-g*@W-0$N)Bf=uX6;6EN;$ZO6gI9>dW%hQad_63RJ8|opnXI2~zQF)6O{N7?~ z)h%AVL5{DAwa4VCD*Vz*%6QE*7AGqmr)3HWv|{^3?9r?yoUIzYp}PmS&42d z{zzNAUojof!ATo-qa2auQy~{07Y5V%D`&7}!fX2Ga5NphM2t<>FOcX)A3jz)8Pm#S zAgVGOZjNm(*8>@<#A_rR)gD7|qRjdzyCc?Bt4LWjIuuMx7V#1Q;9!i=O=^vCA}BFO?yn`AZe%<*Xw~ zKccu>I03Z&%CG_5)!enhHsbuIoLOf#hEbjPi2mCcPfuM8CEV-JzfW5!mcj#Xd!Ki9>RpxRMH#!j#Jm3AQ*P}JfyA3g1}ZG(`_*g zn+@tv67mB*?r1>v^zo3r%Z-WaN)b4=1Ck4GKEm|A3hwrMHOq>CDG>i|eo@r3_jG=0 zDZ<(aahTssj7V3ZOMLGTW5-xf&2;9LMlIt0E=z*7&PCiVi*|Z)e-d66e8xY$J8VVuf!(7WUu2m_OCzW%&~?;JMTfS*K$tYMg$t}D@gvq zOzM+%R>))9^8q8*OL{6)sKymX=&4n~KE--kk#-YCPWwU+yBdS??swE6UjZ&_A^n^& zj`sD3t1f_Ql+3L`aX{l1zPlyr8^nAT<#jJqY}}9+txIA7enU$?=Cj>H=S_8 z4VT>agDj>)EQq^4plT{?5@G-^kDdHct^oa3cfHe6!_BrR*iYPri2ERF- zg`8`bxR|wesB}-7mR$^ChQB7#IK92lB#ML`75AWE0f48$Nvo*og~iin z&!I#za-<=O4#SOo5*YXLo+EM(~iipn*$o@I;$2@^utUHdi7mW z`X-XRPff&s9$KX4`*V_>znsmR8v;@_9m0L}Bn)aC#1rXYJdhK}8$Spp%QFX&);*q3 zohrl3?2CmTC8ns6Z-rlc&(X7jyEpM>5;bV6qg^jX z_Vo*(>)t44z&V_13QmGv$3Yl5QkM^F9gI0jais2?KDZ2RCttI?=}V~q^7G3==zd^A zXZ&5w752U-e#b6ydDgc0_^K4@{%YVB9-2#E2$`CHzrx^twkH^j97}c=i@D-Ou~_b# zO;+#Ep`BjVy#Mgc_{Tayvi_Vd`iIDY(Z@V`Dfoe8-jkVFpISl}dB4VNGYkIf2!Z7w zeHd!jXtT_X7~-_;H0d;sg~t=kz)tEF_4`kPW@}7Pr_U67o{h)6@1}5$zNF0#6X>K5 zbMV}aRk(cRRJ@flhuX+B(m)wR^N4gtbI4{KZF3Dfwok@;vyJ&fmz4NvuL7{;^*-F& zd4gHf_*}^2H>2I_HGKYp6kNcqdJ&IIA}I-#>w{+d*dE|PRLKcQYvPobUoKKG!w2z}PE*zWh6UYcM*gF^?n z?`OjKveMBs$ZP_-T{=bw8S99?DYVi1^Os=`j=<~#lQG#t@Kw8ZQO9>5vBcs4l{s*R zzb^b$%#Ly6=B9I)nP#WO_wfO~mn()^m zOSG;zk0DOsJn{TXq$9^dyJ$V?4|LG6!t=5yBZRhYNk?vvC0||a!uN-6!aKP`@t4q- z8rx^WZ#9?UwU&=V!{F=GYCw*!Cpt)ed}SUEEaBG0is{R=dHlsrA5`p`#1Edt)8xHn zSYd6ACA=Q5zwjh%zdi+{TnmYz(_nb`ji;A1>-g;_GSF2aTr^p<069K_cl$SySBbcc zUao4GxcmuajGm#>Ll^X(w2=$D_5gSJ?WccEEhg+1EA*Qk&WAU9K}Pm;o-5fYIdO0= zvHKp0+CMMhq0CtrpS+#Ye~;;FTM@)`7hu8!3no1kqel86j*zNO*Q|MzrHAFe0b4&F4y{yQEXmrGj2&&dT_ z2hMuf9cs5Xk!jcJq;Z}F^z`>B7<=yzRe74moL5dk3w=$2WA=}pbMnR=-W8Z+?TI6H z9zfC?#@w5$jhaH<-mInzXO#S=j~#9JH#f%f-yalXQh^h_uv(peIP3(U*`SP2RZCv- z^|))tDae+D!wcENBfPkpbzKg&BovL37OoRx5X^(*GPCk$Yqlr!@4hB8&R9Z)Gz2bshftl|WLO{~Ool4lQ{ht;Ronm%@x3I}%SAsq&91iUdYpC_d|#@DDy2@K0YK zN0VoH7#2Mhrrn4~y~P^%zP5ogXqKaGyFN?G46`xr*HsJ~9?vIFn82^o_QiyzAnNnV zk_?{n1Kby%f#Q3LlGcv1v@+@?-{UXGtA9_YyXy)tRpte5cD#t9CwaWizz|IG+rmE^ zm&H##Gz#C@H4%y1JNh9YlAkTG8>`h%;^Ek3cxqe+M8&4^Z$Dqh%7kbaghMy&CTOHA> zb_w3HQ$#DfT%4GsO_kI2MX75VxVL6ztVWK;49^2Y*#jo4&~M6c-nINWu32dgS7w@k zS=VSX!O4_ea%nrbhpNK$-Vzdfcz{g4;!hS#OoZ3tI93+#urbr7k$W2_!wSh}*1S%U zd?^}BT7`Y6ZpKVlz$d$4L)t4JFxxi^vw!BNRpl^c(7v@Eq@F!u#W&&t3Evxw{8K z%$6)s&7*o!bwC>Uzh(H=-VNp+3}J__muFYXZDrE7oCf<{l{j*}kd6CS%MMLTXFUoy zu-T)<9eQOUNtK;UBM&XWUj`#t9h)$=?_D2skD9`bu;LK1s`*Z2 z!?;0Eu9Cw9m2P3KY}(2B{O+bfF>k5S^HXqn*Ld1iKyYi+NSL&A0@~`G!lcPBYA=+je+fdlgbv1mjJaty^5te6hi&Zr^#kuX`hH+-NKFx(yvkOJ0U2qAk z%t1rDR;)Pwl0E4!%g*OEvA^Ai2b-`&zd9$m2~okw@lC#OzxD_3-j zaOiC^eVREfDo+HN4|b3gVGDkBvTUcYqjKB6lbIs(l9^iLNhaLzVEootkg3D8slKuz z6kq*Iqk|@5j#@R|m{LM7QClviBZqZ$7km}xyWsT-L#o)3$=u%2fg6vnXAh2SWAPY; z6{aVOMm~0(Vlx{keXK`T4#40FV;7+opmxZwBpTY z{4UwUCNKL4stH=KX#8xZbx=K3KHfz;GK>Z8>l4r^Qe$7xAbRAQKj&)ZkBcgVUWbJh zUOPPkqTaj&a_tIM2iS7SJ~qG|Rby9A9L?_gaFE@qvWS|S{Gp#`1c}UCdKhP2B-7@4 zP zo!X5-3YL=I1&bv8rB1{>C8hYcek#$-%!D@OqvR~LAjLg5aIaW_k8slGH5RAAsTHFk zb%`b#@1Kj`{S(nsm{naI@6FFQk>N``3Q^&mk&ypxV^&5lf;(pq;)b|$bc)ps(q1=; zHFnItf{@ zYX{x%kxUK|S0#Ya!Vsu)b*2wp)Ff3IJv7zw1OyF#%+&X%5s`)loO^19&*BxS;qr2j z6L`jRJX>LGLNfmPH;i}44F>&H3vt*lj?M_FCZh1U{Lxq)zH?Xs|Lmj{UijwEISRZ{ z$=RW>y5JZFc(p*BXfsrOvZOBi&2dW8MMSMgY`tnko8<eyH+{1{P>`bS( zSABpVe-wqz>KYvP>HzI@c>+~aqwx979%l0W`DE)MZM;9{EG`@ITi~F#;OV44==tgi z*_Sk)wb>=6)#4efahN01^HxMp^|euP(F$_=i8|a%E+&Tx&XCVn&EPomn982)MiqSt z#%;0U&-}~45oAB_Ss#z57u!wXY)pr2y0eWjX(S*@)KPD6S(NN z5#@H%uhpg`dh=U?3O*#APv`#5wc)Cw1uyyVgSf-&Aa~s(imUxm!z>G3Muso@K}Og~ zQxBnQfAxVK=ol!mp5nbU`szu}qU<1?9u`1n3M>-eRf6lzHwI@_-bU3evKZU27&0o) z!Z*Ps*)8VC1*?AqUl@RE_e|*X|Hl3I$r5aqm|&RXG`?3jg}ZvEWBxTm==_*WX7|s9 zVm_2?9eNalCkF~1ky*4hez&h!>5)$4AM>tfBt zK3|3vZwK=aXGTz(dKWfmtDx!HMBJ7mix>Vp7aa9pxD?|qXefUO(ji7LWOo?z=z|8> zKNhsfRU{yiX zYsxwGsuYqmYajU(m5Qe=C-MXM6KpgKVOmowcsv*klKE5d^X~?FIJ^U56vD4J__v@& zYc&2;E~PQc<%z?d7ffg5G0C}`UMP*yaL^^5%D!rnn8fL!<(+Qgu}hcVFkFR6@fLPr zQpIF(W(4d{TLtAO{K&@342=7)2bM2OhDGLgq2!$r?6wJj=2Ufbef*qzrPPhvD~;K| zvkKtuQZ3k^b&D!K97?`c%)@zdFNxM+4c05k4-#B%kZS)mw6o|4F>Fb~#r8ij)$1Pi ztqLXZUIeR!$<^nkbx`aV$N5ElgjzK%FduUg=eDiFRpS;h&Go9#miLA9f3+iLm&Y+0 z3nNI)AzL&x|4Eu29zn-NN=#V1CSNswG}i7M2Oc#?A^hS{eyzYE`zIrKyqwmfv{Meu z?exU5=$9Chv<$Y&g%G#8TAdmsiwz z(g~_~br5!5m`lFR|F8IB{$g?`<^sxUW&+&rgWWqPLy7iyqGT|@)C``6TDt{a?fuhG z>rzTeOEPF`#%0ucR7rQ7Jwe2O^>OXxd|Ebn0@r1e0B7$Hq0+sxXnpiVSXtIfMmJAJ z#$pwT9rcI2DxC*Pn({dK#!*P;_G9RgaHFByF}jlbH8` z-fNNI10i1|deaHFEY#VY2b0jxTN)HJM zd6Q1k)}gBW2FGeT%XNU#0yAGPy!l|FSEnFlr zkatamoifIN9?o?J8N=_K{1O?aZ;db-$d4mC!akDEnk zjmU0Vk4w&!qQR6_5v-7ui=E(&7v+L$uUOL z2*1qPPj^dCBtfapXyvN}L3adS>-8Vhc2pmCaG;+4n|g~mH_eTf9h-t3b1Sj=s{vTN zij&m0pCx}hvrr$_9b@joLRSy*(3OwqUdEjsmij9mzN$pH^IwJF z5(mj9Q6H12FYL}6WF8Fw%n0U-4B!f^=y_VG_)F6#=uNlqG2uvT#zX9r~Rk#iEEeIA?r0l=@nN zev}TL(JKV!6SpNl($tu{Lk6*~0xSM*W-Bp2bekCMxkxn^lybK!i|MiyBazXMNb0tI z5q~~Io-qh(C5=8cWR9KhGy}mWeNCql=X5>cmYE1HSE9m4$n=Pd{VOoX;WX}jSVGOm zf1zWhtmM57mPyt+g-cAQr_clHb0`fo;8nE|_BJ*!rcttZC^iq}KOAPRue?bAa3j!G zyNt3e%&iVZ+5WOijTTdYj#koiMNHR%;B`rLaI$UcZS9O1OgYXR3+eEG2Ss zu)q-Y^CovrG*gGPQLKI0X8I&q4VQl2O>WgIu@?ECsG-|zJh)PrBaf`4{rV={%c+`F zD`_9ICMX#6#);{vb(`^s!Ab5+!7*4rGaM_NWqC{aaJaCC6HWf9KwGytoTWLkbE zGuG0c#3>jGeOfmXG(3w@(^!EQe#_9zEJe8Kwt_KykR{R80+~ zlQydIL5?<9Y|V0xM@E2scp$sGcPZw+%p;ez(%JUBQmY&tS&scNg?@-XOrA6^VdFfs z1n+YV9Xq5351q}YZ2}+qgug4CS1u%4)=^yL$cdz*p2ar`>D(bNc{KFBMb)N`;LaIW zFe6Nr@o7LB@vq;FSTmB!&d{VIJk-TfijM?`)KROvDu$f5&W79Tl9`EDWoXNlcrvKx zA@^VReY(fq7&oQ8rU!;ML*ak!@Xvn0@~-0v)QCKWiU%Z;sd^^x>bdxhTAN8JQG zynMDqZ*V!fT!<$g{f8wc8>$5Mcs2dL(hGMCQRcl=H_=Nw)%jOy$Js;k`kCqbv(UD) z9Cc+Ua@ig&xO>}2>M7hC+FODlyLuH~P8FQz9{c&qx(ICSdMmVe?TK9fC^!(zGe$mb zxI^d?`Dum_M(6>L7;Z)wHT&U_>o>+=$#dw^&Vm`|RWQduj33f9 z+2Tt%Xm_X;H8&3+&X!=$4Q+mWVm{q*XD>Ed-^IxyV?KLB2wu7!Dq!ygw}kr-ykQwc zzdn0S|Nc2fe0IxFbAj>fbytrKj9UYCIuGdAd6O{t`Z|&}coDfjyC0SY<#6lmG*EGd zDVcE0iHy;_M3?%H0i!1)@RZ<;TX;DEl{N39yjB<{8xQB^j5=qyBI3bWdL|?n7&DbU!%6n{AyB9n!x+hhz>D3&d3~P_`K6;rN!n%cmm%S#HpCY0 z#2!WQl4g{5h(jA;rx20si*4xwkHz~J^-11EH@<%jX4ij`%-avGmQFRK`d=1+_^R;P zCO^ReBT6PYNrUvBb##?WA_<*rfh%lZb5D-zNYKTbEH-{h%tp^e)!Gc^p1`z>64(&? zTIMnbiae>!mz#_}i2>DTuHY@%h!f9Ok|GCHaw0bfZbZI?^eft;3b_{4z3>t~>rN7# z3}1&6_J`m~dqvJZDg>JUq~IraTi(K^fs8U5iN1G3@#vj+a{l{p<|m%QG5fzUvdT#` zQq726888)J1<1k*sk>yX;uPR^KZW~oviQTKgZ%lZ$>teVF&0K2p#6R;W+iE3{F-7k zOi>p0r2#O@sE2md=yN{&PwF?YmCrs#IA|Tt={~(lFS&b?g#|h^qURRI zszjnYvl%yL6FhpJ5U<80S^@J!^ApG5RHGsskQpjC=e20xK5tHbRw!*h^^8o|V$AkR zSCJ}DN3J)Z8n5ptK|F29$4^(}ZNKW{nj{N6Y~{oiB<|-vm%3n;^H5eLRgOF1K7gI; z#Q5^MIaa!^!gj;!=pOZ+yzRY6%Q_x0Teo*oeenob7o>(N9#df6dR_3IGl?HDOox}g zepJ$1a1qIzDEjcM4XirE@ULIq#M-3yNYL{ZSHD!`ME8?ojgW>b(3sAaxA{!ax1M|;0(jucS z+UTnZd-q)>F)nB5`bj{oBA-15B);CEDaxh?Wszeh@Jhkf^qI1ghThP5Ic>#gwZ=y`Mu*`ky%U%_6$qF zC9e|5;qp}O2K$ca=E?w!P2~)eND?^I1hZPEF}Uo{4tT1R2}xVKMdpffY|gJp_^OsbPc^5&LF1vc z&)~h4)HZt@GAI#B_g{pm2fgUhCFbm{vnN3Rlp-JU%aq-VedRhc*2tR_u|RpKXyW9S_N8fW-|wDzoO~Wdm?!=4Q_ax4L2gh z4w^ib>G3MTCDZa6pDj|w7gHuNXZDJ4ro@D#yQ=fvGMa3~^|9=yAI-o;Ph;=aNHO^f z&M_sbqsU6HCiran7TjWt+3pW9w8?U{;G+AEZl|lsoakcsv(^lzMr)8A_5K{Ln8n~*LTZW;%HVS19a!wf+w_8HCFn#N_D zN`sSj6|5a%MT*Ufh^yHzVqOpnTD@;@T(1(}dkyHGc)>$kYCx09gGsX74m|GY4yz8W z0L9v4&@OkFX3p#;W9=ePFHvCVJ(-A}{XFFi4pHJ3Mu4t`1NyPBG4&a?e30SGcRnSx zFH6C-vx~dEt(|Gw!!aETra;p;Lw4mrA`_$`jn{vap;|;TkwG00?UF#*l|M}ES97>~ z`3&s-D^E;*`eA8bCf@hc2Bq3`@_m^S@BZ&Gk#Q3|8zFlz>H86!6}p(rxMD0YKz@>W zaq`@4wFRVc`Wp-`K1sjbNGDIa&ZC3-NXdu&k8wtwIc!?%37U@KRgh|hGfG?iwI^>&U28PJe&>BQHQ?n0$TG}pD!{`WY*(L zVXyI!OpaX*<)bI@_mXbHG7VjlckUOJud>9|U$1bh+pWb@O!Z*&y8@VOI*d&J1{~Xa znZkYrc)$K8V>f9Cd(^EO;sg&z$%%AosC$qzxh>$**&zHElnHw8M}pkG67hD=X7SUi zTryKr=rBf(B5YS8?f(XH6_r|WMd}piH8z7dPZ~yVE;eEWrw7+wo`r*yBDvRo z>7?+IHJRx1f-aCZrbj1@d z$E(Ej*&BGIKY=>#Gp7Fq7QobnR~fgV`rMxHx!lGU13otE9zGHF^I{DSOu~B^%cZf@ zYUW#k1#Sz55mW%=Y`T z^ig*uW8ZO~THB2xd-ARE=Y%3UBFdUS5Jkzr2y=GADGxq-(GoJ_zvs-1LVd{oxC@7- zh2!q)C#l_|&lqjJi|#y-N=_-5(u?DaF!uI+lAUKn!$ZSp+s7&Y4-h-k1bTGSJMM^wJ8oKT19*KRafyy6g>A#Z@^v;|_Z~;A-4~KP%fY5s6a z!DYDRG)BMY!^h0FpU27tRc2n^SK2(Z!@>L)iFPG9SvI-Ojq1^NquJhM9wiD zTdpo9j-fkn^b#q~HtZGM5`F^JpH^V~rO~)*@L6m)cAb%LP(rmHX*`mZPPLual6#pE z^yr>bqVdQx2jU)3b5n22!4n17`;M7(g})_N1z&OsnIZ|Sz`%w^Rj}%bI zZB+B}b*etp6z*S~!~QRqrJ~Mg3HR7=zlMd%GMmjBFQ%rxIU3QFftZM7wfP?ZM^8Yh2{q>EYZp7RH8_N!6U%)U(Y-NtcS_ zv_pKj48;iXqLM{q)LTFn&`uxeGHg3|?zGxR5=LX{!k4yA>*C*~>_$4N4rr^-itfPDS zqww%GBOE{L7}v1=C~DV-NLq7t!>U~oM8(z+5{VRbe7}Ppel?RmbQ_JP_1o~<2xF?T zz#NB!*V4i%q4fR59P0bvJc*iL!np;;kab7Xu<=EyFfY48PHei3e!r&T@$B35j?q*) zyOra5QpGgep^W%jX3^E9!)S=H5_fEtKL1XMzLbgT8Nxk2WH&k{tI)q%+FYCFCbWG2mAjEThio^S ziW!Rf%;oDR$jKc;anj3C&|Wl-zLCk|^qQsd>WVZpHVdY=yK?bE=n9N{I|Oe(0Nm?m z0#|;hh@Pofh-OQNq2BQuXz!~{Z~2Xr^eQLPMiXbOFAXC*dJodK`5Ps;_K`)&pEilk zg_l;KzXvN~7!3M(k=v|vliNRG0__plOH-X~}xb ziCl{w%E`!tHHmioy-Xusj>f<>Z)inn44Jo);e4J4<1ODJTICZgnKAP)^$c*9+?ib< z5_!3k4enuhWn>XO@$f4??Cn8c+X*ODxd|?INRhE$456~^6rLUUfhP}cN3XKqbk4A5 zwDb+4A-`ntpo~7ApZXI27(Ap>GblG&b^<$V*ld=b=Oit#jyE!>9Q;{}F28H(EM>uG1M!05`pj2bqAC)Z^gK3uc~3|9>!OXtedJrzQNUEKg5 zkN+SFbTGj#`ynKt?+3Hb!U^U4^!WRBigd9rPwNEsYSfJRv^ab$JZ_bP8MW~kx-$>` zTnotG&I0izWe?FtnJ>s_I^sL09ICZHTkx}9LY;3tu;$MpFw7WB*6+)t`=lqL&Wb=5d!J_5Q!K%NJy6Xip0a>x) zc!NctawroH4l{-UA-ko%@iV@($U{BfKe#bg8tB`S>AqKI2Q9#ctp)i6&cwvqwOY1Ih8Q!Z(99NE$VU zliG^?sPk?f7=Ki^`Yt>VT++wk*a8bW*76x?9f+i@K^JKK;!u&%_Fl=;;h&kGdQXXz z3}vqUekl1p&kfgK-H4Uj1lIKL6_)yIym0V*b*f?~%s0kV(4QKkP~t0s(lP&WFAW~i zT_J^dQS~?^FC7I-?ke&Rx8`BDbP6qh6NrVe@5LVzf@y8mboBN%5Hjyd%<$K1=-<`B z_~A$iy!9;-nS6dC%I@It`7~k1{k54}YW0&Dx9E29@zh3gYM~0(R!+$heg=J@a}U=! z_)xBn`-WQ&~eAfwnmo%lGzvppN z=KW>jx98FY!j6An$_Lo-mggE&r19IcWNfTYflkgDIwnsFH3O%Mr~DTBtwSb2 z#p>atKR6WM+w0+ezvEor$w`n^8izmjDPw>1GZa5arK+krc=dc!@ga`T-!O}+$52u9ZnDEu<-Gy(3(j0z1YhX zdHHgV?ol{+r6WHxz*WdiD)BOg=J4XMF({C4k{c!oBIWhVaOATgR>qzWt!zgX(RYJ3 z;LolHpju>uU*?U%ZZxD>-=0z@XCSvcQZU!Lnv{*uCi`o8tt2`c-12*>{Cgh_6pM!t zy9#NjdVGd1|E0&>(rKg`VU4uS^&QH*je#EJ^LS_QInFmw1&^IHqUWc-oA7v~9_M z^TVvlTa?+sT^-PP>>>^g^<)j(ipUTXMcg3sl^m45%+TaPtTO_XP0vzoh51%fcFsjZwMg`_Q4=pQaHRPyzcKc&5d2CEfD7T5P@S1e zD(+N(xJyhLj+@cW&So$(AI~QLdJau0YEU*t7s!IOL_PinTw!DB(FYp1+bI|pw|>L* zv&-oE{0Df_UWI-7qL5iUaF~r5`<|xVx%8AtoXG8CW0z@T?RCn?thE2kB~_jYwS zWBkG@mb(CtpLI}g&}9sKCczS?g-{f-nMsqAV`r$VLU6=KI&XFrG`568W2O&LmyZJV zjq~8&-%esaj3;lD&cH~85bV@HioLDp1y0#ec)8F|;{MYMu0Glhoa1Rw9y&-;wjxj* zYsL{5l|tesWC1hAj)RiEqcHKK4*vUH&#aRTMXzZ?z~Lf82L3Ci0ZM_)uJhWWPz4rh z9_f&Cncfm9cSSItKZ(Rm4kB7NRzsk`ZaXllfQ)%|g)n(}#6N!$Og6qlQhjD}n)VqO zb^Rz8+Qc|;yv5}IbZPg@ zB6d4(DqC>KhL?BJlpYI{791Z*#DC3wq1t-~@TD@)>K7^Kdh}`XYL^-Q9^DT!Ha|dX zzn_J-&S~)S>uvgNf-G&VwLptsC82!=mY~O*kFU2}q-*SiXq=5Dk{3I;(CK0!Gj3%= zCiyNw>{Dg3!)6!T$&-bZ^=g8^?8(&VkcI$@m(jFMC()60|1taaZ7{pvD+`M@E`%GO z<`5H~t6=g+o7!(`;rWwF+Ip>tzxGEDwYanlpRU;kIsdH&=XB2L^*xTUXe&Y2jMUA{ z@0apSCe4HKj{QU-ShA|rq$g+31@wp3rtQuQ$1T4 zx|%YO3_ih5X`YHkHRT{xe*tYD;@ay9hv0Hd8W~OtVM=V1h|M7lXnNAlQ1Sgp+s%d9 zx2A&BXD_3n>n6Yl?Gx<9ELHG`SPz|jlbGFB-li2q4=E*;@^!nrxE=p1JL+42F1t*D z8BbXxSDQzc>W{F7(&x!;ry3^csVv!+CLk3D${{<}nbJA)V4MB9YD@1}lzAo$NxfD> zhs|Tz5}h@i!}|rF4y`7d2BoO{$19TcO`Ha7dBhY%&*$I0lt=y>UO=bF-C&2eO{6_w zPdoI*!O*vaH)mZmT=IW|R(M;oY1ba0g&#_olPet{=t37f>DQ#QtWD_?jRULm@Y9Q1mk?w3WHdK6K~!L7&uV$OeRBt@TOX-xIWxqwKp#`-4nhJ+v#k zg7d;~&S{v;|6w+Vy*hjYrFi!->~&cR!+K~+7{Dp7)8wF@ELHG|WA@-0w65bFeW{W_ z9|Q>tR%QqCmo5Fk&N8_IE7nW0zG}P3;`@^XS7IKsPO|yrkgE+9+b2QUrxAknaaiya z&m%V)t4RL81UiuPhqy_uA$CD0h?|y!*~FVQq^np+p!_2isxIoH9dYUKckxM-C|%41 z#@?h|ohz6Qp9&_h=n3ijy^rhPv?1l72;`8(2yBjAC+Q0wv+ciH$eCTv9RF<}uZ-b% zmUpC96;}-B6YI$bkj2?cwbx6WtT# z!3C2ad|&%oXp5 zUfUQO8#q8u4XE+SFky03kCL<-cIbe?D`?bO4UewJBad&!DEz}L>bY$`v6`NOTto7a z`HlGyKGMb{Ijf?@j`8S#%ttaKq!is&l!l7~IP~jr?kMlUp$y&7XiJnn%HFx-ULlQwW#K+VF*e0h~1MiN#Jn<^|W)l1Y)$SpHxq zK6!ANxzCq++>^KhLUTsR<&OuMv{wc2`?xwh?^8o3en`Wj71r=a%$#wV6is&QaI4DI zbwXp;wBWpI7Vm&+KPysDN8Y;&A-&&q%tqrW+@0oPr0TAUQZr1T{C+Xi#2x@L`3o;6 z=^x{uc^w|+i!*gSfe5u+B%96|;fU1}@RCYT&>If|mwgJPS9$_<+162e4+{QB-$-JzP1W&ECER{8jp!NP=DmvWgU@Lr2xIdwKy%d?CY( ze9=Ht794_IrT?*RYyUyQt-WwS?sUrC9QZgS^eT_5Mq~dj79C6d7N!WFS2j`bghcfMr^t@jZ za*;1aGjdO$^QN`@7v1+^>TWUouAZ9}t(<^qKc5)3do$u1e6)9x15!_T0>3$K+)z;w z+*1f*v{x@_S}1uG2_)tL{mj~?@P`>ByT+a_VL_15TZha5Q+BTn;9aJh{O!q|O8 z4Pww-;&UT~eZ}>*E@-cXV$SV;L-`o!bfsYBgC#gk)Dzx!`6GM(2K1jx4{{$c1<%M# zC>#BOOye1n{N^s{-;;oZ&c;{=V*EI>J^k)c+uR(-yOXXFfcW zX#-4w`VjfNwHS7Z3Nf#iRl>&7WvKSsEL4=`0fEb;;6L$DT&=bT>_0!oDL=EYY^?=8 zzHu-6#Zi(6L;=mA1SgZF&WT;PA?QGJOOVnb!H64xC5|bKy?{!5Yh|}VjJyMaAAcm zzR{D0((96{t&`)cKd;h2^!I7*caqI--1ZEnUz?0Cec^aWT;_MmV|f_QUIs3gceDLp zXCVECr>LmZ6bt$eg9K589+NGEcS;xsCq4u%gGr2Yh6Sq6c?5%^pAi+3!a9wA(T1ZF z(X+GniDQ*4tlO1BOs5|(-H{pw{t`xzFkO-C7FK}k-^*#xY|assD$jL)HjSW3p4XxFu`0fOI}MJ1RUmdL2hgm; z2hqcXR5r@x7aLjhiddcfi%z+%V*BnSu(_JHs9~`zQfpKuA57HX^qKb%ZCHg`%Y;bk zt~yq`X&gCh-h(`{6e04MFS*0bobK{_&^_sTsCU+B5{r|}6rRuHTs28xdtwaT zaFl{0Yh%DdX9oOj5kkzrXnMAB2nqB5f_KVQ6xXl{jj4!`hn?q1+?5%m`Lrp09NUGi zY<`ZuOt}KThV3BJ(hFKo#<7B4P2T-e+H}P&C3^6#HmNe<)_|Z$yuh;-6!+(`t1N8E z==e_BX>SWRldmv^2V1Hu7Z83%a}?Ug&j9N^??55u2n_0Fz;A=wF#FYMW}0LeBxy=x zEz2&@Jg}Fjd%Q%;7G7Yd{5K7lZ(1~XJc6Am^#;|u7o+8OeeqXyah!fH9@a_SN3q{U z$W8N4?B|tXSc$`R@s(~PBat|^xp0twHe-OVVl73c&Q2iZ+vPZ>X9T&kYZ6$0Izm2p zUV*K*7Qq5G9#)>OCGDACi2RBqmhtB@2WQMQCLtE=p$WL^qbDL(-oi*<0(%j#uIBhn_uQ&+7J=HM(+#4=~%QXYd%$Tp+}y_R8#QI^V(syVVjjfE zPNOHiqlnrMOZc<%BPu)+58(l4;kL^SbPDZfjcm$EO?NKlGOPM4#v(nPReS95c<%`nY+55zVgxVaQfqa&JBJoIja4V>66wcU(p?#aB?@ zqd@pnybcUmJFxaG;6ANX72vVmQ7s0bO|v_#W)`w+!FJUAIyjnYGxK%dK1 z2z4rD4)3)@f~!Kz+=cmYDIM;sq+TY_KlC2?-#a(re-HJk`p4>RZFqB-uFjK`)q@N4h~Gx?Jd z{GPpz<2G1;(!46p(b=-#r1-}8@{9#edeltmEOo>xb_e{GyI_dVBf*^i%S zx`M+6Wx6(?zWSWe3MSyI9hCeN1tm2_y3sA17KWdq-hbuD;U8H9|7MB8t}pCO$#+mD z_m-*m%0!V%6{&`#ICbvck4$fdf>o?IdUjtMe;e5fTRM+maXWE5pwNMimE0rGimnmg z`h~2W@_F)h#&MFExtYW;=g5DVhD0nz2tSZrOb?wu$K<@0rUloF*a{z6ev|GIx>IgH zHLpERZrr?wWTRyGWzL~!(IR&u2FFm;&^hS)c?Ibmn2J~N!(p#&BB)O~30l!&uqAeC zRhhE^L|P&AaROy0-zh+kwN`gW6rJF<@!~rTI3h|?^KHq5$&F<4*cv7?GM!XMl~sGZ3FCiykxK7P3}&S5&zjje zR3n9%E#S661Is@h;h%qRijV0y;B|}6qk*ocP!sDGH=rONf$P!L;bwkc7M>x8Ik0hUGp{=)$ zqV8p;Oq5y^Y6;3>2h{RR@y7<(C?kd3YR{6rcjchD@iHkQVJJ!U4|6HnhO9VP%|@)Y zz=PT|pyiS~s@Hmh+H!cX{oN>O`BsYb0za@l!5ySHFq~IUBGDsLKAft6YeVJ6_<8bI zD5_fr>HD`*vao^QWFSF%M4f2ruO8l9IcM166~;=gC}1bCTLHH!@P?|27qbWz;1O$;^4|AEd zLy|=L^*9+?6Af}}6F{)K2EK|_p@6@bR4*@p5X!?5H;nK##}qW?S_3@B5w0g61HS52 zTJSj?b)zPj(JUeG-W^Tz=GfyM6)R!aj^m`@g$n-V*~`BzYl}>0O(lC;jnSvI`HUd% zHmYh1G$TR=L`rZ4_G*N|c*cH|BEOdLz2Cx|?|qFn*%XrR3G(!K>?!76c*%>zJFs;A|XIFNCj_zc^oQuFmEcP@3}#E%Zr7l3)0+pwkmY3BHn`4@*gmOaoivGf=(1;uIfVP6MsE}5ocL$5`gtsN|Wkk1TX)oFpby-5C17bQ@^onTyuc6R*Pim#n4!dVN!V0O$z}9`Ode|yV zUR3R*SKccSjTsBLH6(`qwi_mr3X(`S;R4(`bQ*+wLx7wrha{U)R=dNQS+XzExT=}qU-f#M)=RTDe)`~+TZ)LzYjkjb7;cYV zhtC*^vSsJ>td&ZSRL@>NsX(ffcl%>Fxn(bingz*;~lKONQ0Jgx%6^x zGGV??^|V2McE-&N+AXw_`2JL4a^jvK`}fZnQ#i+_t&1I9`8FG!DHJBW zgw1$;+9^1cjzOwZ9mdczUU`=?FXL`BGv**mZcflZXZ~P^7B*RknuD$EpihTev?5~ySH$@_)!$Q&Xb4^=zvtK zF1(7&gN(3okk<$Xl{Gn#olyYzd?4}H5@z-u9_Do#NU%q?PvK_j({Raw2701sB4p0J z%#>!jfQ{1xtdZLc`8u8;cCd=;A%@|q$O7zQmdAW+N3`IwDXrM7V7|<@6PH*e;q=$3 z=m--AuNTTfF)@bMvo51*!`0}*{aW&}^Z-2(^_D&1cbT-zjpJ!<*-kFo=h43s=Rs`I z|JTz(WKpinm*7dV@6C&8=GqKYzdo7Os4s`zCB5k94Z?LfzN494L7XdIjoQVE@|KYV z_P4SjUoI>Go-7+hf{g|^>D(-4)~tM@6!Z|yoasXE++9iJ*S4{*zRKgpEtBxb${mvHp3U?ZFvlB(D}&tvI@AWZWGn`z;O-l8n9Qt2eWImpU}xOjp={wyBYHgbMkO| z62VL*@wlEuew}?-J=R6#;M{cjYMDDJ5_M(Y`gxFmHcnK0NRo5EJ|&uN z_0^$b%Q=6^1>`QOjD$Zbp;aHV@RlG;{QiqDZj#x;F002R+HoD%->4%I)~Cr0l?agj zB?2$EE`&y=m<(#nA~y?lk+EVqy{>BmOOK}VH~NX8w~1#-O=LgobNw-2HP4%E2%JpL z8$_eoF$YNbnYYAMVw4gu4JP&1HnQd5Z+2woPRJBbAg)wdP+iu~H@g_d4is!BXD%Kl zpnD9h%e@8p#?j31o6n3|hCgj`dQLaG-eRQ!?l2mw{XkEA1rC#Q0~u8>v}{TWnJ{4@ zU32^(evvT(%+bkMGNJ=yE@;4u<_Tmw{R|OnweeZgS@`o3S!AO1ofAR1s)D{s{*M)X*vr1~E~k}_Z-rqr&YF5j z3jD)2GTo~K(4kykCP{?rTgW^?6W>eG@yEsVmwP#gKfH>b84R%Zt736xI=~!>8H#y^G0M^n3EA{s%c`6o#&>D}wv+ zT+g!lA?h=Xr!foqiNvyK=BSe)D2tV&gMPb+y%U8$XzT8y#^=`a-&ob6c?!|h z+xQx0&3)XO^Xxtf@4jg^t~UdA#^~W`ekZ`8Y8gCJDM1}6DP*x&CcfBPj2-Qs0BdiH zWLG%C`=iQGx<#L8%Pj_n`am*zZV4)AT8qb>&%yl4jj%s*C7ctn=XuR`q7_bt>{g@M z$dD+ZInDQ2m(W;J>39wdUW~xJ5l!TiG>Dcj-UXjl4Wo|BEc{xm3yV}NnTEb-CT7%^ zGYKpr2R6r}n;QZ_rq=~4Y?MMxwpWQtvsd;0U{6}zqeu(yajeD-im{+)S09FQ9gAvMX~aV9t3qN~I+-1_l_X2iHoEx9B&tzghm2z*pxyBd z9DZpIt*xO*PRj??|ILD*AumzdT5USueTR77594_@G?7m3{BZi~H266p1LwP)pu)ch zjP$p`y3n_5CL;}_UW=gb&`{N&t{9$qXBV2d)EtisrlYNI)zL35|I+w#6})(YK;uaV zxKCSv9j`3I^S52#X%rdY>CLP0#%H%t;I>%iYx5v0s+kP!LhGUHv^Y$xy~@^$SisY_ zZ7?!#0aWHMK)#!!(ezihIdA785>~#O+3cSK-|#~u@0^6ZKBe(yY*|!o>6>d>R z^jxc#`&={YoRP~G&do&MgZH5|e@h@;{UAo=_W1Gs??~*nBEG8o8{D%D=*DCEcvk;Q zT%o-HZ@d&q=E@1v3kkDv`Af`sFJ^6H8PhBhtR?BQ6`?dlgpLM z0`ZOZ%+?QzRJS>mFFEI;nP*!ga&$k>o3ZjVe8ar=swW`$*IjCvvW9Cs-A6neCEC@bKvxbo`?s?a$mw+QSKQv5}=e{8quDfi=io zr5L@LaTR|4sYkN6mV#G#K8P($1$B#VCR_6%3VW%FkGDjEW-=zOwXM9>J>u0W33Fgh z>r_UgU^i0V7R>Fx?R@WB7x@RXZb99Fbj({2gB@OB>NK5U_`gDVTV(net9>)8Q{)Bk z`+7VUcRq{5z5n6Xof{E*hk!TdHZ`q13>)8khPR7vSINC@gZn)vVM5?f5}BUJ|NQYW zY1}iJefLEV=3M{-yWvVny z4aj4D#rQ$}=2B>{lqdD~v_az1eboKa1n(FL0E?LQSov!`^k}VMCH8SSf<|eu*(r}F zP%G@aRF4cE6+=Y@2>m)b37@R{O$L9C!M30I__ELy6n{z5f}=K{GXBF zT@T5$Wud%H9ulabk`Pe~Bcih?hUks-BGr2uV8J&>g?gWfx5r8d6V`?FA$cOqgp;qi z=TUq9QMmXz8Ny~qK+(@!CM04CEnQKJrm|-se)CzPO)@Ib`}|FhadE;+A9&M!|uNhqK5(;4G7e@U#q;u6~NU4*H8Ct@}x zwG5i_65#H{4=^{k7t{lufbWVF@Tpw~`l71v=l26Wt-Nif=stC6g=9E23=eDc4F4Faf^sxNYNfdlzL%g)^a@aoHDH0 zJ`MZb)xk{$!fg09eOBFlGLkCY1PKGt%$&q1u={lpMVd*$xJd=@*0z(gZjY8GU?~dx zD*`H$)X>*P!|L{v&S+m%F4GoP0P8$nAZ6o5QdkvAciy~>vKzBu@NXe=wR#J1=2XN8 zbeU#>E0o?k|A@<@VINp)R|LeGh{SX|>Bs)H@sz|=dl)n6Awytj@38$4PFYz^z zE}KsX$HkzV!>Sdn4JgM@XElWDL1@uMJm2Cu%1ZoKeP1*erN2D@YOeF(X|5ftFs_0v z;+I+3f73{-(o>|NKM@b-x#N^fu4k$6fgfjh6Pei;poXd9G*jj=ak>AEOtdRw-vsHA zkIsHb;cgK)|7k)WnRX_kHVvJeBZP*gX2F5HBcLqR1v7WA}W z#L~;aY`j>HuDpH$$wjVaQj#LUBkd8$Hmw5BYE}BXwV8J}ZWhO2RtGa~ue`GAfZ312 z$C%D{!qW^=@Y0L-VfvbQ92qGf+qP4l$|Ef%?fG>612Yk3f7BylaHy-=yYdZlsb()3 z*S$an1D)xeH)ZVQt_S?puMN?Z{*P743qnn-U1p-@WyaJd

e1z7dTzJT!|vI04PP ztBy}|?=}Hk#sH3G!W8Q(*lnT~Hh6T8J#%pdU35+hWmT_a-L3kNfxH-c@1jP>q)Y*G z`_Ko^*?4GB1V=`1XV&f7h7H~yfy8O!C_%0mS-&_?Wt z&QTCWgVFPSkHCWs1-Ft#Br(YmZY<;Y46&R~r- z*;MqWHvM?*6O-FH7ZrDk^M!PcnJDKqIBel5*t1v@>Rv7fhsmeFIk=0{I~{@ShaZ@0 zDN<}l!fUj{=n1)RJA?f^uM5qp4JD4FlVQie0-SfK9;!LsLYaml9)>;an%0j*?5i0l z{oIGF*U9p=gQuGRK_BtpwG2pEny@P;h*7(voU6W)bM2{))KAWWEyC=l)Ddb zMBTMIpmV7oLSEP5H5Wv16xZK7#S_BP=Y=rt_(ulCmm|f5VfL+bJ&{jW%OE%WZNo5x|2tY8eZQD@D0@37YFD73oslqgD3dVo zi=rdrG`Qt;{pOlRT4P zF0c2+47^Im8B?ySR}ezkv4?tWdY}}_t!qQCc;?KrgVOAqIrVfgvC$;dgVHKX8EpHAvr?JGJ892*r0ha8|Bs(0Wu&0+c zp7v^-C$7oDr{rkl>Xpdr9FynvuWqL4QYjOVA6)(Z?0MeD$Qx!BX5H*&X*Z}%J3)N3 zvS3yK*JTnK-xU}D=U;d+8<&nTZbm}*U!e=Gn@Zr3xem?#FO28Y z5yw7`*TY}Gd!vr^?zGdj6f$D!(I)3LjF>b-%6?^oV}cz%wmlNnoRtzpofd@!RU)w2 za58%|GX&z)w=*vlqF8T#39P+!h+Q0$PL-nE5&fQuMoeF#**ELhgU_Z5p5KwfakjI_ zP&S2WhxE|WDd2Tx5%zRXAn}*vSWW8{WTkaGT6MY=mc?11evLXN-U^_#FP%=T zZYPf%by4P}I@1aVebOH-LZ%!hXywXGB(_Nn&Si8U`E&+%Sj02av~}4IJy+J{*$aA) zmq))uxIo{YU#M`QBpp%G<68?#m<#PD=szhVvu%H}*&TW{U?=mLG(S1WjxV!hCx=WR zGrufht|&Gzy{6Xi;>1N(R8*DLCKQpmX7Si%mjcK=oXY%dttCw>hhVHwn~qxgkd|Qu zy5DmvALyA;`W=c*IN<09}p8n^3TJ#izsRDKFipwTFv`z@5APLzsgK} zV#_X8>}PZ@OTlkyO^w-PEVJ(<1S}WA-wn8#=janKZBE55<4>W;-jw()X{D~=q9o!$ zEDY+XV#h5xc&4{0$htfOhb3~jX<941uzCW#D{D#XvI!um_6s~VKLh;?ZJ?XY(A&8> zaA0#1YJMgJT1`v9x9A`3{FO;(TpL1NcGF?#VJKTxx|iMl@-y1L(Ex^}=b$NEzM*H@ zU$ng2hH6w!g(A0Aq{872pZnos?Oo1z|8iYyJhYb8)>=Y5Y@%VMN*Q_PIY!0%j^Us4 zBcXOlIR1TGAD_SThwt=iE9b_^;N0OVoIflHpViO86YBp``?DP=G3EjL(a8uNxyGZ( zmb=Mz;dD^ZT?H>)!(p+7EByFExO=?G=2u^d(C_;bscX+I-p!{@u-d4TO8j_(EYwm_ zeY_-f(+#Ay!ux2;=wX2&*X=%Jmrl-aUWR{fl`gaTC4zp#bh`myg zk94N0Q2OEwQ!&4X2uc6o?{VIWC0@4C2Z17j)XfRZ{?=94pg zA=KbC3IF*wN?+7Vq0LI_yeoJIDFyQB@YPpE(zs zURGr{t<5zhV#~>|b}<~Dcn`GmoUn@BYi8-c+pG|6qic;DXi`rz=<{F0{*zW{D{mRf zxK%_pojFN;0-OY*4Vkp(YyucHNswQTmw3nZ5=qM(9l_We15`fhOx@16(1RKB0;T-- zbiDWr`{U&ddPk&@Mm|cQO0Fh?BVU&D%^Q^IBCP_NIOr~@X&M%6Evpvz@K4g~x1JG+ zkxV+RWD?@w)YQ#;5j|lUM;aacsd``-x_r?UISEZg^OC2dLf2l@m@*3A4J zR4-4VPo5m}FNX`Z>hNNvE~IHE(z!0D;ortj5XGI---%iATFaYgaDFgf%us|xxnE~X zxr~V4J_pjc!;S7vF`*^3S?tfGR+6{+A5VN#4gPCV;=3pb30{1jBiM3OQ?N9294XIv z2S@Mc)6`i`%r>zmk`3|fh+r<6`k{il^lYPV{2S;Z!v)kcWjQ_LK1jAtOQmzq-laFz z4N-?lH>uS-eX5!`iN>Bd&$G0fi=?VGA<*kM9S-?J-{u@Oo6~ZF2Gu_TTNN#uI+{ph z7OB#MYa;~d_O}GVSr-Mmm8n$AFOS;%X`&yLli-){PP1H%WIA)d87TUBFm0mS$@hxe zH2=5-UD$0#1rK>3VXA@e|oC!2c1h{HM`QmT@GB7El2 z;pZZ7s?V0xn%Kd5$CXfDH&yWTwI21dox|>vIfUw-E?o(V|!3e|R#q;sf4Qpt1m@(;1vnMS!cbWVV8FG`a$&SCj%ZBuL zGiP^t!Al1z+8ox%7QLG;u$G)H`0IC15Whe{Fjk?(lYZ6;HSJ$$@kkiS-+GwJ!eR1Z zWieAzeT?i1n8|KvT1+nqvZ?CiRrHYE7L>pL9x7t`INrl^!Z#S=&lRz!CE>B8Q?Lf@ z`zT4Jm(PTAsUl>sseol)Y-WeT53w_zZlZT*H50$1hQx3C1ehKZ%t{=zXWAA{5adjW zXVh|}ixk^`W@t=SW&G|-WhCL&(%4ulLk5s~! zHKOXk%e={(V-8=G?7>P^MfBMU*S8O>Gm*h@^dA&%^V+!DqVcstFWw1V|6qcC0Vpp!P zA{H}TX@OBT)emYX$6U3^t&0UDYSIrR7Ar*EUM-<9p($w4Z6YaG9AFavEhDG;z96C{ zPp5J3Ra)-fnC8c!tmW~kv{NsX^b}vC{a;kbg5;gVvp18v1kYl%mDM4wOoh7TZ6V&7 zm_r{taViW zuM&y6QtM8!Yk0>|&c?s| zA89|?9f>iYE;-oanCY(feK1y_T$tCi9@Dm-pGDge)QZg=xM{n;wqPd@* z(E)C@5hoKxBCqjjVvixnzCtbuxJC zCSR`cG_uUx#IeCu=$6-BBCGxl3)>Y+W@e9n0vM|KGJkOviMv}?Hn*WpSdrt)qy@3kz< z{q_PincbltdAsT3v{F-l15;Xd$%#g%%8@ibA!>8Z91dJogNLJ*yfY5Q^d&w)VWAYU z*p-XS_`~SLE-}WYO$YXkNYE+WWpvf15@O?lC|&4GcNe4+PvdxUxG;|2__K(qwdzEV zQ)bd984`kd^|$HphUtQkzEh}SvN(PBcpJSc%||DSt}+bVV?LD4V4dq+pr&LmGwHV} zTTmd$-;<+H%)cujsS<5gHiqLON0kXqyB-&uuagtBxrzzwPZZN81z{R|S(fgr-$J(U ze}=kpB;ej^6{__#pZ1>bCl3GgC~b!z6@U{nnVJ10lwFk~0#c>R*=kNn74&K`JewLx&8?&7c;93oT z4F7^i}x^Y#@)GZ%~(l0$Nnh~4u-)*#>_ zEOdKBt|Yu+Cike*e(T$0e`yWhQjm*=&o?mjMKAa>(=yrIz-p#s$!+q$WgEpkrYJ1j z9343JkB#~}$dIMRBrJ!L`X}#siX)PQJu(|!A3V8bX-A1UJ zKb-_m{LAhr?_ksvlE7PUFZdasWY(USKsp<8$uH>^r18=mnT1p%+o4J%emt-`@J0m* zJ>`p*j)pM*H5qaab4}n2_khvpYnUGTn4R%F7d^Sx!k%-M1`Dxw2o`pe&aPVKzsV9b z=9mT8n{Efcld?$8sp7h`%VXxoSOF>)vtz`IpV9(dBhvOp4pnk@zlX*&vHm|r?AarV z59ci*$EQAKmVJ+AXU1%+vj2RB^^o3)giRV)+5TJXx=TXLhK6FK?sF3IUBm@Hod(z> z&171puMVv+6aD!rLw}St5R%6+lRdQ2dpSSUKl?o|RR1y!ex?V$$BvNTMGNS3u}sF{ z$A4^ScRTZ6J?F7hil+mH^8`+7g#_-YK4kbo7jtJ)Dp8uZglcJiWOR}TkaW#uknLB( zin?>zJ8NtNgWTLzZoRLdKg5Tz5uU^>S$c)IT4cs;3?{p|Vb*lu9Z0tn-a!DvB zr^swV%m{4XHXTcUT8dNB_aML6a8x}mO(rZ`fwT;aVC+l=6`FAy$m;>vsow zU)Jo^pItQn(Q%>_H31#2HYfZm=5$Ttbf$GuDBE`?l6-BgWkPp_kSTv0=!rcG(XSN| z$i{9vT3DLFn>DQ#Rqx0rk9|JyZywAhx<4b3=X)z0QndnCE=+=~(#uG2Tb22=%@G|n zkf+L`K`2Tz29Ee=A&22BV5mGqGW!oB+c*`f{xyo)xrOrFbQKv-=}0Q(d4#-PElxAm zJt6Wr6ZyCG{^J#_pMqM#!#KxmE9p67LDjFdG4U$O=v#3%dPF{wp_0Q?GEk8}>A9%D zYSw=c0yGDdv(s#rQU7NXsd+#Zttt=&%mZNO)T5B?J`+vxe2<2+Pr+C99L_P=z>Gf1p+9cs@E&yAfxMzPK5;6V zPCxgWt{LACv!u*Xw{k9W4NhWTj(+7oX^~`iH*W?X_qD{XxVP$aV;t2ft>ASkuSeTn zb(0k~Ygq@5rRLP1OI0VeqO2Qgg4Uy5WSd zvRwsaOEFCDxDG8x_-LDRH9dA;iB4bmi7KBOr9QNZifs|cKIyF>>c-HTKPlYW5=-o& z4)H}jlsI>V1~!nH$W$hrV8>df&;@bopPaE*$TeGU8{%O&Z%x835>?4YoiecjQx6+TBE~1*KATx*ADx`9xn2L0z z(D{WQ$fpS$!{m$=FeoH7iNXg?4_sRs}3GHlaJ*)(Pt8 zrPCu3mymz*52W$vEl*!QgT1Iek7y)1z=8BcIx%%CnOjrIA3CZ=Uwzp|FI+ppG%Ck3 z6I{lbQ!}(Vj&wTOlP&{E)mi9f%nvGh{UqsX?nU3c5@>!#7SqD5SG{-T=?wD$Q2lF% zrk}gO&4!=Sec=zNN_YU{zkN2FofAra3SPi=p;_4Mc@|zWc#9@fhts%XUT#hh-u@za6KV=$j*qDr_N(T{}AP-u1vJyVHa9vbM zV=UF%!^&>2=C2I-O9k^}>9a(0dO!C8GVC&7lXDA+qoy-5&fAXiUzszv{iE2iKvPDI zH-S_$w;`#{8~rVQ4f?C9@wf7EGRxCMuxCjj-54{URR3H^;_Q|Kyq^p_$#YP#^da=z zb`dBz#?s8^zgXh`j;Jg*fvlmcq$EcQF^PewW=|3bFK#9>HFYGFYH+%wfzv13;vilij!~F^zl`dl<~6gCw&F?@ zz3VpobN9!l5+}GB*gsP4kxH30%fZBC7Bw?`0ZJz}z>zjBy!oLbxbIWL>o`Y%j_FUn z)Iuc^^LjPCK5W7K+0sn3X0$>|#2M-lPx#H3C%~$c2a(;^dkk+bpy%ab$oHxixW3p! ztRo1L4(j25k4QkP(tvEg5f0v`C2@D5DbBuUgiCcl0Pmd=zTm}maV=!=NZu5z!M_Sr z#+bE^dVo}%OL$!?H^KY6CmGe^C1~(S7}8p!MlTX^_Qn4wIun1Yx;6}(XHqIkr6>}W zOlPlWn^O`csYHY{QKEsOgpeshicn;%43+8ZwYG>%FDglC6wx3w4;1?L_a8Vv`|PvU zb3gZWjXGk-@h4#S{Q}x;yc*t<%ecKJjaoH77-nxYCXYdY=&Z93XJcW>}=NYfHJ888=e)K&y zmiBW~g=V#O{+&68K+!^=ZslQTuCuzs8UoCVm=?8_sPUgV61PquzZ|C%E2&dhytfEV zRH*Uvdv8(YEz9vlo*oQUYoN4X5!zJYENVN@PVL&l^(oR2-lqr`kS*L@TMFkfjO*V2{LBrj&oOrK>YOv=EaI=6g1Tml9c#}3MoeclpM1^ ze=GA}>Dj7pLY?rz;W-+29YH3GF2Pb47s%{8fb!QwfLOF7m}mIoZQM@JiIe&KpTD#t7(DyD!xH_w6&$+Ll0rSIsg0@n*{8?Ta8mL_u}eDU9qQ>aEd5e-Zgqc<8IhD8c` zbc^SCdX8NUUS{4xDNH{}4am*`w}MQNjLJt=QVJ--SAgh<3GnyaZ^OOn4NMW};k~^_ zz(40BO8n-Bz2gk%rx%;)E2f8NWB(oW#jDm3wb!58^l+H^v}=sAz1IoL>qS9*hbsME zH4}E0d`1%sEm+k{0`&AaQxcJL90ZLFm>7Xs{DUMC)h305MtHH(CDoKj$ z85=|=Qdcr5OGyrPUO+ptU70}q!}(-Z zhb(ExXd!hc1ZYJ)U!u985$nW1AiA$^P`%r8;Rv@|U9-rCWPL9}x-0bfcMd)!h8lyU zW?>b{^im^6`*ex%kT|HOtS0Anoo5P9TObX!%V_1bugu@GT=#Pm2NJ*H0M?Jhkz=d~ z+mkuMywD25L$N|I?aNCX{iU2)bkCj)Q7hn==@l?OsR^NhH(`o~4V-$V!PejR0`{L8 z*k(JBY06C`wcQKF*Lk8}dix;SdMWrXRD$-SV#vm~iSq))kvpbxqSUi+|h$S4UJyb#xaL_#{HAwG>m6xt5rWg_GB=;^dL`RjgDM z33o+a8k#0*!_*zcV5`23ib>W%@s|=YHT4}~($%ri%P2Czv2yVBGt|ctTb%K}8cHYC z;l^xC9iAV{{~jO6`io&)L>M z6`XS?ibl$8&MA9hJ>&<%>HRSDWrXi-tOyh@l0Lf?qo&GuJnqoNv8E)*+s9IDVSf^N zU?z!rw@iWeV)8JzX#>(cdxO!q{R}@GSK#ejIhEb+(nGF4T#nBlGJz3Ed3f>Nls$Wc zkbRQ9c>SO(Rh)~+?RjT$!qkt#rPw|F*EJo4ZV=U9!LsM@4K zkshOk><6AVHjR=X8FRj(-QA7+H}e%Kt&n8UADs?^Qgg_o_#E;+coi({KEW*7U&ACV zIcbn&cbS~B=CbiS)7cjm_9Q299Pt;YSLJ`!Av)og@%n8_c(NS9fDE6Kjx)kfg?mWI z4`X)6?gaK~LI-RzOd`Wu5(v+0C#BjFOFFL~VNfmAYSi_0P@N7j*$vpGb>OoNEFoF-Mb2%L#s4y}`+g;@`#8`Ej1(U1gf}~iH?jvCd*!yk%sbDD1@>heHBUMV$>6) z>9f0X#h^1ve*TC=7w00cj}dmN6v6%%$=o`nc@Y4(TfSfRod6^rct`<)&|<`ZM+E z`mr#&`&$w!7!84=$GEwPOdaxhX95*p>XGY)L8w{T0uuI#kQ;44Z|L3$HzyUyom0|~ z-{FG3Gg4>=tw2Y;EXJl6RzmL7PiR%Y6MXy628{%dA=|fA%%i`aW|CW`IVREdJx2MoHgX2DSx< z>7$zRbjCglu-_((luwp}kFPpDoK{R7SWe^C2x|)XFL^dTzL+&S0j{6}_HFryv}C}U z`&ZDiEtBNjFMTRe&RTO(;f)3-2I zEkZ&#PQb3r&Cs&W5U6XzkTZQ17%$N#69;B-8P`%+aB&}%JUAP*xvj$$zUJ`4Y6N<- z)L<~*4rd$ta-MG=xVc|~Hd;7`EZTmW7@yvao&1hN?XD+yU1A034(EVeaSqtHL{s1N z&%jRqb~IPZomLxd0|?98XXzN}>g1o6x>NNf55S4V8Nx z$p!VzNN>$+*dX#2^_{H6eRs?toE1j;lH3gJ+I#r4f)DS?*Wd*Qv+%M?50F`*4u=&7 z;FZxiYQNMnO1npc&d{XjG}Ca(MCT4NwfF;~o{Bt)vzlnV?EoXE=zy!eK=W5;@JysdT?G;kOQ=qI3RKaE=7%3>Bj5OV5C+{xZ;z`i|XD51NQN4SP7B znXQpJcsifO1^6vqpHszLm-d9Qa|_X*@*41p?&ReROJVdr4ZbJthI?g2_=0vMBdm6j zdA;ufq~4gru_z=F&wmv59J~#ydbq9CPsvbRWJ;Ua1=CLY9pE$J0L{{hNNe&k3a~3; zL{nNIr$LOaD4avLz4D>8f)~Qgd;-)An9%D+@56k*QR*M(77aKgO}_JHz-7k<#MAo%i+^r|IQ3Z2 zw>8HmNFNgPZ%}i;nV`C_(fET)GViOi6YQX^@wqD_q^I1UJhCXklkvsGYj_6SDC2f5 z|8g_lCs#>XnH+vB7)1_y#*ob5EL`woH6_jE!Yi(PV`A>ck|L=La@ue?&NW-d4ts}^ z$lkY%MRYaeTO^CtA3sa39UdYQ0b#`GLL2W$m=vpUK7j-x9+>`BC##oeq3T)hc!^+( zpX^*jVn&YQ*53WdQoR*D>l?sx-Au5+m7V+u_}|Fc`Hj?@$O%j$=7`fxY1LQrC-gY_w4E7x%WTC1IvCBi_!T+BB%ua z_EIL3WA_+4X)pG(!%-Hr*5R52cS=ml97WvCCMxSJ$@{ukeCC)CoBvmfoCwZDQXD&W zU~CTQ50z(jt&75&E?*`V%QwONd7r5Fr-%5d53R_h-?ikoVG=uSc{DNpB}^U$d?I@5 z8i?)f1IQqK7L*3PC8evB@bb_&#EX+gRn%X880YcLdZ^B7j%rcg5*yL_IxnQOESI!S z#*-r}X?9Dz75mmCg^Z2M5%I;RsDF>2g3XDKl%zp}!OR~Dq+9(nH9Rf{Nh(LE{xd<$ z$&AC4Na-?Yxc3MvI1(~io=I}0Q%SeQY--u{>Fm*_NOC`UCEr&2Dm|{L@%@GQFkE;}#)25h%hFumVOVIp}dSlYyb;Pi*cbK3+j zi{rZ8*we)Sb$KRL%=qCiD;=4*yFjk$FTrxv+9V|W9Dd_FACfaNh&Uu*_19VCTdpJf zS6_(zXjOrKZtr8J-N^w@ZC5h;%Ug7HlPz<@*#lc}KGaV#8SLi!nrz9^A$-VK4cm(b z8@34g;crvN7?YX0Xjg`eVR2JDN^qS^O8=^%Y*r8Uw_PON5etw+c0O$Tuo|wq`J!AM zCwR*78p?hC;nmj+@wt`KbaUe8h{Hp#t|mOlLAu>(L^u zF0?y#KlAf9$7Gf+qu*rK!39Ypv_R}8eQ(&1zIr!?+Fu!mrv0--`8&+e^3W|r_iPUp z`?8bp?q47k+l?Il;-+PBTcDluI#O0Xbr@-@74o7YpBpDQg;IXr4bW)+} zgxga3Ub_=2iGPm|w8RiOjXzY8dk#ud$)@Wrnqg&gWoWRXplaSs)Y83`S;;X=D(wQ{ z%hC|Twg+#(Lr{=5o*vB1$}C6!No6Co%rdko(URWOV2ZT{{!nG_%o$&`m&iG&j*@W0 zSa;rHXmgpMm|d!%ny!aa=iElUznmcSLm_opFOqkS>+6sBATpB)@J4j3Pkx zXcGA4f8~3NUxAjwE)*Ag2R#eii|A)zRFHBdFJbL2d^6w%in!VZEA~vl0Tl)Co_Nl) zpRdK}9vMO2?oB9ul{EAXO~H+Cqwt!~J7~S%!_cObN$E)303~V_Bo`*|E-chSuRr#q zSIMDFR=^|hwy*fpmn>X5ek8M*1RQ zX!@l%2p@S4PXE>5e|yDoZ|xE)NIsPQv~d<)6tRMt@mYhG-yTBd6$(S(EFQVl`jj$! z!RH+9^O4yB6-wb!0_F8&26Jdmui?Ac>1gphTeSD2A|=n?Oz~apnSXJj=BnA~no=!QsG|qpha@0#u{I@;mjT$P2<0u)pd?-nR;CGi zQff*nhDsmq!kT0m!)0+V_+ygeoLg}=Tq$RG5oZe{ zKR_0K391cRXwHNLaMdrcSkgc-h6Csh*YQ{XD++5C%pjfntdLvjX+wwbNHn*j0CYC@ zBCkq)I!90y=1p@pM1?wZrM4M-C~2+w6eJ?BafS;>0xh(kq7xrnMN565~iX;9j?&(c>j~wAE2P z?9=QAQdeA2X81TWvDXn6D6PTQ(q15asROWll?f8E7eO}Gm&k==ZFqJqt?H`qWBy0y zt%j>vD}$K|H|ePJ8)$2zh4kl9f3$?1iAugrL&BR?(TQU-_^+0h8m9IM(x*CFd5_L} z;stM3&}t)7I7Z+xzQK{Xl+||?y!W^tHF8*tp5Gq|9^pC6(pyR-xN;X-5h9ENO+4vc z`<5ZOmTXFNpoxjDR;Ny1xP-KVVkyH;bG+oiGT!4!1aGT$(F*FLs3q$y_3D%ZDq8## zmdeDTy340g^v)mrpnmJhPS17d{a>viTG4{pZc(Vej8 zM=5WPvN>w=YDGyKHj)iSI<-^z3(d!|n4dKfG;Brp_haNIHc@rrNIZz87vT?&#n7*^ z#c+ClH!hpP1E~xLRFcup_rfY1I)&x_?<1J=)JhhulU9d8QFB(+a_z?cVb?Af|eUNDVTH@kbgaV9WapY5XIxcZ1+_Kd{|7nFX z<+EFm^5#gCP$lJag^t@lm14V(F=d560QaUpUSy=&Q3Y@<1ta(_wOE#v>NA_ zf{XBO@5AVN4GVj2ErdrqpWyeMm(YrXlMMb8OO*?mKvd)+$}1v=XD5A-_?~wpFTYLU z+>1+K$+I3D^xrEqTFC7Lbw@zt4^Oahw_s059AkAWdoX?fJ(~RJNS5t>%jjK6BsaFk zkvZPIsC~UC*yzcz*=;2V{Wr*W@9IErll>?$eH#$>jiN-IllXd4QY3zrDx0=&9y_(* z99j0{GnViEisa&ta_p^0GQxKz8iy}I|Kby*({>trsdpJse^p05-WA#&Sz(0cfyI$0h{cB9& zbo&py@bODtNciJQkw4iyjf)1<4W(OPxug!cZoggS;y;ZPDIFwhtt)sX{Bd;fDVM#x zI7`ci?&ae$Ln=&DgeO5EMj}a&$To7hXkH@H{(hai*G*#{%8Jnm`*)D!3>&y2=zuL| z9i+S#&w~@(+|6sN4ec;QXrY^DP-xu+`sB+h819~eggZ}DDnYAwQtiKSb;f_h`4h)Q z*ICK`eSHq2zc&yzjb5Qv`e;!ap?>83#C`62z89NCnG+dWpNdUpDF?BpCEs43y|GahXg%}vDXD9x?tB7 z_EPOd60}qkeX<$koF>WS_~y^#=5Yy}-WY|pRF1+Cz0>GI&R_mj_iF0;?r_4scO>6$ zwGf?rQNtah^Qz9~Dzfu&so~YSuS`qD8ggjOHO5gmUVrG|Ms`)AEOkNi0$!r_k0}#4 zO<5;T$Cm^)FvjgVc$?ArN=b?HjMC~KRVhhcyqfKc@#LLI{U;5(4Yvgd<2N%7qP6gr z+=CqU&Pg}+dZQ|v)xVBCwrN)7zD$Y{v6aW4SHB~>d!|rwOLfsUQ#sy!hlfn>_GZrEyZdwA&BK$7%B&2W z!>?y6@b(kYWYL0c?4sJ@D#07&Oz`zC9FT3!xAre#f=+S0`fq#S%d8Yi@<1eY^^+eG z|J8~IMEzK;a(Tm9XR8b%!&I5?b5vP($ye+S&41+ZA3@kSQF!K{GQ3HcaX-e374sDpJwySCnL+2u3g7F9>X#EZ+7sfUUEX$4JCLsGAkKF zhkLjoY8Ufk+jnxaz7kI!t;W4YvE-CyJjsr+plEGV#@p9|iK-UBGo~b087+-LO1_5F zvZ+fDFEogqePb6ZbuNgEs}_^rZ@%#ZTLh{e8$F=t85&jdGh_G<+@_Gmt_Z3>J{EmF ztc^w<*AuU@4P@_ajt3WCYw&^V!aee*aKf<#)bUeJOrfb3>AD`z-~I3aHqMwufBfUi zKfh249Ub3KL{)T|b3Rc7#ouQClRCh-zcEC2M>W`t7!`iAVgWN%$(^lTD+nroeUR#d zPiU){I-|9Ef=Zv|#734a!H0YGNPf&ivI&I~Ikjo5vvv^-f8LB!uYSS@#tlg8r6tT1 z-_6w9z-wsTR&laSzM20`shWCm%@c)=)u0UHY{wMAd#FP=1puOHH=U6r4qf1WG#>)j$;s^896 znYNw?mwDno#arBNvMYY|{~mmQ5wkt?9ZzY|5~eQLvPxzjjifk+P0`OsP~2zF9=}_J zTZQ&AKNlw8T-{rYF|EwZ^Lhkw=eRwYDWq#Og;c{9$CV`%knO*6*kY@P2rSxLu z5`&5<2lzM4oteO8N2o*3yTj(V4i=LZe}gXhtO6W za4m_PGCYlUTsL6ii*BGzH-k_|_*M98B~ANQ#-J5>G;B0XW3s>R#9|WN%#FjE)Y7+& z#JwiPaGPZ;S$(0svhscu-_EfeyGWUkpq1{Vq?ZrXSIw!Hixl|v`XMMyE)}we7qUWS z0`Mru9PM7Zhg#G&1)FEAr*)%5>7*+UDf1Q!{-*UCsNH=Z8TKoO40BM%k^Yl7P$rAF zLpzRl@{JNtwfhIr82O8po;>H<(Lzib8i8Sb&iSU;!pyhH1RJRyChEf&nY&z{wLCAx zR6H~#w$~4_Yror&%jXKPXW9>ba)c7u#M34RCy!zuMR~k5#~-Q9m!%c9bmAyG0qhYu zOyz9-NzwP>;Hjhc)v58(0-r8M{)!dYk>|(A*^M}UDk7Cf!+-jUWb{^Af1(2Be4cr+e z4JMNzIIB?=>E2iec5~YK`uABX_sBYG>fu)WW6wIa<@EyQ?22Zdr$8>6a?^v_t&jme z7pKB$55(kHoIwVEX243J7~bhQcMTIw3Sr(HNt|LJPF4IZ$6tmS{72D}4Ycv3=drV} znL-y;y-?hs=(8xuw3HB~{sdC&unI@jFF`9(S7N7*d+_MWY^r8qHWS$Gie_6cp#1#$ zk(!(#jIS+1TD>&!agbmXeW|VbygG8cj5+PckM7P$O+)h2~MbbeBo-9{tJpy zO08w4z(4$C-~h3+Zo=>8N)farg5+skrhYrOkhnM(w10CHq{UCeCgECe^Sc{7F>SB< zbsregZp^f8mPQ+X$MWbMJ(cmL4`5_X8VE6yxO%-TdX>G1=Iq8)+Ef)%ABW90TVb*<9B_R<;BY$GZQB z(aC=BD;C02MH_HyT>xq8kRT1`1X#I8Gsybc;-vQTXG&%RqQl>$Hnhv6d@RP0NI@+}`cURL*reJq4yT6fo<4{|2e^!(i(e1kx>a*m3wd6?ShX z8Gd6zj&fZ^?FF4+I$sF;{m`cVxbH+ocP~@I9~6j6LOO&V{Es@DQiEfoq^WkNB_v?s z4NApa1@?y}pxb8q`C-cpa7|PdGD+`+bcdtx)uEsI{lkoVwkwEja0g00GYt|7F|&4^ zAG&j>7wp6oY0;4g=3e4ooZ%@07psMkkJ2gh#eNZ^{3D4s%P|$rc~eB`Z_wwhy8Qqm zU!Q_kt!k*RRfXJndICFD*PwIc6u)TS%c><_oU3&BBkum?PG(4b#9wZj5~l_~L{-V- zA9$SVvvQ#P2OpAI&IT0!pA+i!tWt zitD(&YBBtK_6}t$H2~I}0x?p#V4wR9Iay8t#on{n?!E*ge7zPm4@5xfvKU4gPX(94 zL<%;=G7G1zVRe+{NYO|nyr~z#?;@j!2c=66o6klyA1_t8+|5Q~;R|>UvnH^va62sV z^v74;c*Dj`x8PRWQhL83LaB@>isyVxD<6d5niW1s@77-YI5ZI6C91-<3#SeHHpsw# zb&r`vT({3bbuQbm^c|`xID>b}9l}8&v(S#)6*&HsKMWrLlwT$TF70*9;452L(#5er zodlq9@d!%od=DDKO;pU=3Tj+s3NKMPnA>rF1X7b5=_Eye+Oe1i+a$vAUcFT6&6D|5 zp43VbcA*cY1kZx_aVKUKmth{usztIpH}D2p`tiHy0iM#~7^=x?DVVU`6tj~dDV2!q z$vp&4EpD*tTqxNW#myt$C$lFsHzPCKQrK>?7R;Nb(mOu;@H!NgVDPa4(bY0#Hvam^ zr0g9;&-`r|>)dc^?!QxT2}gptz-zqw_Xw}vl&&&Jy@6Z~QxNF5h!WPwBk7yG@wTvG z>YlX(q^yra(Tios7A<{tn?y2B=>7?kH92VNB#Zh@b#Wr+UmcT@LKT+?xOJU_)w_RT z-3Jk%<`D|QE*Id_neXW2Oe_1@Ej6WO~VD^Eli{EA^cCemFk;6mF^cCgFm6$ z$nxS8Y#OT$BC~q&;?27m2>3!Bn&Agc32Pu}@2jc{%@|wlT#pOy{N$_XTxGO5hrrR< z*O}jb_UObDb0nfON;&=QU~0T#p)=1Lg?df{$BFA``%*JZCA1(fFeS>Rnbeabm&wm# z;ppO{``~}CjoNx-Ciphnz=6>m$lE>}S*=k*>vBz4#|@Fx^v+gz`7|5v-n0U*WX2g$ z?9#oOiSKe9oZ}?}(D?H*x(>Al-G9~xoQRTy7 z%+?h1r3F;cF4~Uxc+Wx0E*wKQ8e~D_jtTi9X$d7Wi>NP6@l?ohY2K!W9%jl}j^|_& z$18QH;WESH=&80c$NcX?Px1ti{fS(Dj`9WMHDe}v{l1DSmPy1HdK{6aJjJMZR1rHd z5u%&if@`^6W7(La+&O9hF}62}Dx zhKO1sjTfx^HJHz@96_HM^+A*O+U=vkhzBWTOEh~ zG+MJc*AtM<{HN4o-y+84*ayCO!)0F7(P(zIR|e`_Va^Vj?_jfZ)X7r$SYo+w8UFWs zDm^pm3hsnvu-dEx;%$;RP&pemjOId*?gGBTLwhJ2%0q{3^6@9XWjOn#68c>JffW2c z$EGMmGUs=iB5C6sgtdhrV`?We6qpR+jx-og-2lJ3R+C1`3klyOFmZn~7^mbzTVWw( zmwy3;tq?(9JNB}Xa{9o!Sb?H*8-LH)a~!Yv4cAMV0yk`eVK76R^o1`))}>$hUmIRC z4iE1@`#Wh`p>rEDRFh|%CT7E-RezxSu_p-frqRj5_j#$G2e830u9Ma=&TKhxo9zBxYHFUgs-QYh|@?bLMIM>DyKO^x{J71bW0MpO3fp za{CACjKSsZC4{c8hGkjxh_;Kzb~%ISi(3nJ=zoVpIexh05-Xf5{1Hj!xzR#9;!qps zMKN&RO(Gih5i#YH$VYP(DSbgx>8-UqiAUo+FJZ3ZP<0d*eP}_oAquo+=PfXKQwbNJ z7NFIKjzilbUD%znhtk|(K(ynzci%}wM<$KJg>*?!Y{K}=@O$*U@-g;Yc9Yn?G=-I? z?!dMB9dPA*9{BwIg1H!sK8F_5YtQ?TGQn|lDl`d?1nt9St0lo=x&jQXQl)Iy>GGVy z5AfFx-NaTqPD1)92NZvs1OB~5M2R&+zYo<=9fq}_7hnh)9bcITr3)}Cc8<%@MChp} zTZvfobn;Qe6%E_22N~@)SV|2eTiGyXy3T3hozF7vfl2gp{m(E(=>XbvjitmIO5j|# z2lQ%8LHjFK5cMWmq~tD-Ba#K+PecwfzbA_R9j-v_Y12u${+S((8u>b4yL3J^v6MUWWD5hml*QTMJ0O#AT#ilE zQ1C(z436h8EoeXX%UBETLy3Ix$h~z1q&j-wqc4p-6*{?$ZypNzOObTJ^AMX@jNHEp!L{p! zxM#{F-#sM*=X;A#pM~QdK6GspI$=h;%}igUjg*{glVtIAKWg` zJ-84$jd{_vk7={+rA&%8(YZ~c^u+W++FNxWQO#NmA?yuCtY|r$E!m7jhi<@M3+}hh z`z*D5UN~uMTL~&0yDN+HzW?022)*okfHoOvkg}j^xFb1-PP}%4UeoSJ6?9Lb6V9Ze z+0HiDY}+d|^Lj3tQ7O+(JV_$F@)zeVQa;LZXnx{Qu0#IjO0f|$#902V4eZ}!Q2Db3!!q{5a6|G=5XDfo5pPH1esNp=^s zF=?BPaKdXfENF8E>-{SxGOr&axj%i#`1~94_)iSpyzn7kIcxzry22eNR$Spl-k1Yf z>S4_78I`>F+^yvK!HXy;rw&a1Ay}ac- zsp3DxgXt;MwRDbYwy%l=s|TSS!3!X-I}Ki&C!iLG8@OVRG>AOE$!v~GrgqsS5v4yu zY{I=eRI$rCdVZ)9c#rBbC9zpB`~#yet~9zb$Cu>nx5Ls#yI6s>*6gR?v$*_r2RUV? z#ZF3`Bfbtt$f%YV+I6Uu8nNU7b@&GPG{Ei8rzVgNgHNdQodBC=4eaRir7)@|#LZ-^ zn5!Pi5K=BdqSxy}N)N{!HouKV+|7}en?AUwgpsJ&=cG8K678{$LZ0__qJQ_7vmvTm zAU0qvHWTFTulLPRd@6<3*zHDA@18Q-_f=5Se%R15-cTr|HhU*4{uMi-~Tw!OmC+0;YS-`PEQ%&-j7 zZ3p1=GYnDaGU-W4W#3LcO^WvQk`6C7w(O%coYk1hgz8_w&QjiRap6kja$uC8kJcpb z-z>6+SAm>YoK3p!pQ-N<$Lmqg)WMX#&(}NSCAoR%*EL3})T z!vfns3k5ZqmuSgHRh(VVpyZ#D_>t{#=EBx>P&ZJ`3%vXoFWGeq`PV(Aws*@`#%_EJ zC%u(${qvd3V6+{Ql9T64+GI0ouWH%Co(9%5DFCXS3z>}q+mY8iWy<`^71SI!fu;1P z;%~dXpuo$6VNLw$fhoU`g!*HmEH=n9UVRI~H~++s#z(PD`DXUYy36Fzq7kybxfu>E zT1yOn=2P-p574hO!%$t~3o6J#sk$vpsT!=#vSEpWbnadgH25Hhgy|TNw~i?&Rv`c{ zdbOT>i|QbaifftWH?`;r@j|dSkV3(IWl*)D3xCT*Q2N$|{-id7n6*tPGr0_`f`Ik`V`k{4r#(c^as%Q)DWdzN0y6e|dX0tYsbzBvN)pSCF4YHa4F;$vrCD zap|?0up@0H)=+&8SC9P%0c*z4`c@_KSW};j9b5#oym2Oz_BMvpxbTB^F zQZ)AJhEnGhcym@ciP~$ycDOHKD|cN-CZBmQa?qY`n;{O9A6FrDe@XDVSuGGmTS-ux z{DHKuMIph{65#EI;eyRvx@)Z}-QgKWUTPZ?^(ZrD<%%EVxws~Fc_2t!mQ*2?;vDKg zTp~L8U=)h+W0>Y508bL8(weEGXse|#QFs5uguYb3wQU(h#9R$R6L{og+yq}N)CY10 zW@DY>`P3Vkuh_A8Ct;gZNUDSrRAm{!YlGEX?sS#3tzS(ppRYwz*|ijG_ks}D1+a;~ zn)SV{&j_7ak0mW$Lf`Ig=(=x5x(ud5R?I9&Iq5{+OJ||3(d~?4i8Q_T_%d{1dpJ~d zVR#|lg>J=Lz#;KL%G-@1VxxidpAI#8@<@@}5PA#{0?fgfh=e<@Oi*(caBwX#M7Kh%-1xGUw`n^_B&|ySR+0_vwe0Uz)Vj zQ#-8SmkL9#YT@VZ4~Byy|ENx_ozQ+e8y&Qvkn;0Je)A749DY%m&f6LSbuybd2Jjo0 z7pjXiPR}HioeBMMz74HZo`oAVq>0d*CCo&kB6HC;q-w{3aM*OPoU!qFhqg%OqI{20 z6g=PuM(ZY#z2XV9^v)(&?EM3-@QmR|cM$zL`~dwRDTcwT3UP*UBznOc1=ss0i1aQ6 zW{0N;ouiQm^FB$i!;3v=y>AI%*)pHF%4Wl@*y#}QI|((^by6mUI(+A~Tk+u~(RiKU zJ&1ELq$?IBk!$ZyAan@v_0!w&k=-^ZEXJIPnq~~c=`_?gE5UwUEy`#p05OhdE9iQ6 z;#2k!n!1bO()w(?tfvL9$~QpEMB{Lm>P#?CeTRJC&Swi=@bM=DXO6QRK}~+rro3C9 zBC%FIHnRRbBpNTI=ZlY0$IJ`J%>%7OK4ggE_ejFb@fa*)tBhSQneeMWUqxHsP(zm^{9OAC&aX>Ehi-D4?E5LxbA#G#$9=f@+xp2gEXt2vblNl4dq_HUUHpT)z7R_V@Rj-4+dmj92Tmzev z=F(+dZ6LcQ1R594VLkr}qK14Suo9w3wQT`DvRMVaU$u?WD~&=onOgEBKMl{uvN(Bd zBK1PNgGWoY!z~vzqR}ORCD%%$d;88%wr}2|38x!ik(P@EeLf&%Zf1K({XFRtjw9{2 z17LN13n?++Tr&OIXuMznHb}FFne$hnJOH)xI)qS`7Okk0`@vLWyj6Lfw2`A{7BcM)Rl!>NJ^O?8g3j zcAiQ-o=QB_9m$x;MpWP@PwMi8(2hV&xRj+rhkcW!#a>>8%N{#nYFG$4|Dqnde=S1F z9a4-{`v-Jr#%pvxT7@VLYr$ukR@m5S$2Og==AN|~P&|-Lar18aYCj*Iz2e?K9pcDh zLWXf0k0m~vw$LR-Hgx|^J$kC7H#!g(iB76&R84bJqWs^j#`pKHg|xhbkk};+_bp}U zqAh)u4?AvShXXHBz>W~eT+RJY#+7K1i6exW(1q-AJ6hDMIc9^`9ZL4J&!8mkOfhaRX4zIu2;c3i`f|5JCU6i~Ii zjq5`#d67$f7}bIy>MUwRTmBfb-(*kW%Yp&;7e9%e@#Z%XBy-r4XNI`G2gOdl(t!RL z7qmR>6kIP7M{{RXBXKQ7VTJEkq{n0E0&vV?!$Ompk3a*RjXYL7f=RFUmJS~A#%V(3x z(L^T0Vi#I$g8`W=0h8chbhGX}8IPPr^8Yx%;;jL=Ahm*NQ1gK^DPQp-K24sMeFO7o zGx{GjpAOg3$H*g&*T%2HS!Sc?n`toDGuVw%%C=J)t91EwSwn`>vsaNZ`9{=`F`fVC z$ZD944N&6se&&W6mvQ`yVV+)V;yf7Z=(6&J`W&z-G}>o;QQpPc{H@DqBn zCk51$%XsIJD>`K+N!kadBeT>3l=f5~B}6ZV>AnTveEbgRc)q8D7PrzL@@=uAQWDm* z*#gtz-|3RB1!+Z2kr% ze*KUz&5bZ~OVAo~8(g|co;W-|!RRcKr@wj1(3c)-lj&D)qaz~QKyIf3eWx`YC#H)d z*LR%^edR1&^7@e>FPC$tqpKuA?G6^+dl|ltT_Dy0KXCodUudO8Dw$@SgB{kC@D!_^ zArZ}hq^MD}LNSQ6pZ-S)R2#rH0|k1HSw6ZFsDXdg-G(Q=U65Nlj)%eq;L!J8=7+oR{}dZ_nHT>SH4AlIJn@_KvndxN`+bTQU*Wtu*Zhy zqp|xff9l0)+G2(7w_wRB><| zPh?)4pTzSMG)P<>zcF`W@`Q|BLX-0+69)oO!1_niy7=>|

n(MO%)}?0O+Uqr7Ae#T=MHM`$yLm@6DC+t;0F@2 zTuOP1hJfXVK1P~5e{&)Y{1B@$UlW{owdpCeyYU_BvHd%a=OPN$H)SE=_&glpJeS`3 z{36H2Dm*o(a1}6$ zKNug=04(o(f%h@|3TQ3tAxGAoq8lCM>BVBcv~i~|6a3qgs`LI$`S-2H7cFCnG$RQ7 zPB(N*c@i~lDq>Crm!MUBNeneCf>q9Ou4Tn3xNXJ`v?}}rDv6WC3mUS?tB{-26K+o& zQ?E}uwBE<9{o&;H=W}S{S`>fJOasEce$HUW0-hGvTiNYlgwmbzk)x>^Hh+@9yes}p z{kWct##4_Gu;Umw^F?v!yac>(YY|RaZH{#$-y{CfO1`Y`9Q6M_ZPm_x^jJEH+%Yo8 zrej;t+*wbNZqs>^V10vWG>Rq~rHSNm)^?QnLyY`;nMF-iFehcbns6t?pWIGR!Xo45 zjP*zy980etEAc-^B#I5=7&-_66A$rlCtI7Wkoc z2>Q@CL3z@?cum$4ENbn5?@c6AlmTE%=kSFewCtVt)zbz`9oUJldoIS~#oYhe)P=;? z)}zMkY<$(E20IkgM`h}uiZpM>-SEz|^cd(s}IejB>6nQsr+!oHMe$VX_zMXBz zx!h&(`fW96iB&ERc%gu!VsCRVQ+xCTexW5>1kvyFg?N2=FE*#6_@N&6snMbDl+jT_ zX?!cEDt|hn`P{}s=3fmWgCRJ{$OlP$J&Pv(TZ5^D3=~9>CMPHrP z;9DI+DCu50>i7JBWFu-wz|DN><+jDV5srJc>t`G)u*yP8%ZrgH=Un7%H6?S^zJu6} z8%%100}d&9i8?p(Nz|Mvljbvci;WqMlyDNI1Q^N?){=qRb2e&6^tdjj>c$z~-?@V*Ab&MAuVFg%ZzbpZ zJSBnNU46z+ZNGy(E=J?On=R?}d-kG?YEc-T>>@mEZ_xg<7E3)FhKoE2I3~)uQo3V^ z@E;rgx8y`hM#KQTlHQ=DCJlV~1-l3qi$xQ=T=0fj4OBt)LrA+-Oia1i&^&1g)LuWr zOnFwbP*t!1-H zHa;S)_9^6o!cm-`6a{4+LHPSRT_UHMg{oU8iFT_6xcxbf4|HchgZnP*f3KE2$+dvb z2VHQW*a%wtXDU`~%Z0P)=IEpTQ>?3_jpKzxaPU3}dSrhTi7qN4rL$ak%!np9UE)z` zN`H~{;8hU%{F!V&a-XQ$f2Y*ar;(|Xv*{ndwonn}-z_volPTI@CT7kGK{)#aPfydt zpE_qVbyKs^MhP>tQ*<9%fU#9H!QUW8!*bF5Z6bFus-Ig3ky2A{TN_ zLf;Avc8TRB(62l~8vh8Aw*m#^ZV#V&xAPVVDldg&C$58g^G$ra`7W7uOonz|-@^FM z7GZX0eB>z~UP|mv3=zSP;&k`mequ##kZ#8@@;hvV>YF#f(fQ&iiAv6CA+-sPnkeEM z?s9m*5C~ucfZc|H0^KY0*lD^0AeH4_$YC5Z3*$AUAtvV{MMDXZs_M+t*1G!Clk% z78PfiGw|&j7_Pu>w2@|2B3s2zkaNa`8AZp1cl$DI(p-oiuFOARY&x06;gYp53$gu@rNkkuc& z(YC%z*m)kuW#FHI_`pWU{-_5Zf)|jcD_ILiw`&2>j|Dlu4d`Y7H_M&KB6+hgA|HO^ z`bih`@}&VdbF(AYlYEkMd>UDyA_AV}p?I%T9;tXG%{!PiM9u15gkNo|!uNc{Vf_AG z%4w(&WEP|mv*T0nGJQFE`kOc`^DUeF85UH>_CO!i5VB`qbg$hq$a*9gG&%Bq|tL7H7^llTLbyXOz{O61kH6)PF z&3aVeaTo~%%;I{L(}{1$P4dPk5T=-Gk-wY+^~BFBaB;?4RH7kDysLL%rvpaxuYq%z zKM{*7uj^nHe+i~ns*UlhIn7A&9x?HC9&o;WBW2v$!r1p};#DOSz8$;+YcG``m)Ez! z*|)Rl$L%uo{!0&VTibbj^JoWNI8B*uzAi>g&rF3aulk6-fDNUyToaB=(I-h7ZA9Pp z78#hs<=MI_$(cC~94~q{7RmoddLj=ZSx)V}b;T((Pgor2COsT=%Mb4klg7F6O-Rls zoY;zgCVBskQ~wN8(C;rj%%rX{(tAD3W!gB;Q%ne6?cGMAIY+Ub(-b0N?}qrzL#PEaX-stANG!mE93_svMYI@63tU=jh+qq#xt|6a9rZs|jpDH$Hix^}15^9F& z2D~8UHaJF9lA!Vte(L~BsXBLK-p%8r!Q&=x#n@lw#-(_4Sy!1pkoy+Rml5Ix|4!p4 z1`SYBD>P8KBE~fI&iZ-5`EjoN5@6D>*X59ve3Ck1fS1D$`@6! zWw{A{dn*OG`f5TqErZ8r9Rb6H2t3aroJ7y?hJB|!$q~y`W=CK=f9+8j+NRbHe*O|< z-REiI#lf?1!G>B=o}f)O_!RO5zHx5Q2RpFY$#I^?U=y`0rc@xGfD8 zLcv7*T{bn#`6=rAPY|tmZoqXk?}A|MFCv**P2P(o6MauiT?|{q3@N=OslA`DX`n9& z-lRvr?Qunai>D)}=?lO?O$4d!dCO1koL4a?Z#U9iq)c@`h{20=YmvFYQ|g+S4q5O* zf}a`~z%+i3AtxL9dE%l4)EYq}>a(8_eC#v>h+hO-l?HKY!C@N*pWY# z`z-1Pu7ExpPga<8BlY9TXzKwl^sv1aYigFGRJ(cT{*iRv-gUS630oiX;s>k9sGcU) zxhsiNqaUH)#pW>A%^I(sYK8pwbNuWbrIgxFD@tST8g}j1TBdMyDlz$V8MnN=jQyU? zq{U)SQVh?P2)o;mu3cTQyRQJZb2FA9yLQy`(*PYb`Oc$+&B+dhEc9vJIg(;}l6l8p zO*AempRDN!4_T;laXl*^u6qXslo$bsukQ9E`)n zbA)LatOfPdgYeDP8O?P_r4l9S0p26X4dTH=$`r1BA?iv&w7X z8)}(D@d%hU73AGh&W~H(z}GI{WZ^!#ndD!6ihSlC#a?dP$Q^ZMJhe=Vesi_~Nt)ym zw$kGqA3#_j6hcLI zvK<-}`>5m@mX3VK81GIeH_o;a*F!15L4>KA2nFnTNP#Kg&Mq6J8LYjtQ5|$F!tSZ{Hin+ENE?l#ztm$z@bqku_T* zbebQ2Adk}NG{)`M<2i1s6;=&;hkQPFlA}Z#OAY(Nr>ZlInUy>|@98JUErWPdI#tQn z>pj$bWCJD!f~>-9$|BjmlKJNO0>!*(=l}ksjrAiu$xGvn$kJ&99mDpV=Pu1p)?k|B#ePLc(RWfgr zT_}FkkwIK{y+vm2RX8+c0e^bxdh)cvnP=R5hMKE$3N;l-NR!0O3ZPt6BcXcp0!Vzc3_Z&br6W17r{psm=A)Sto)@n{*Q%|g zz6h72@ZuEc`XNGZp5{O+?H%R#4hiVw&Ce*i*%c<6#8KG;MQpc{<2+H_DCbi$xql}I zPP^FRgIPdz#qFdHmvm4!+i4PZ&Oo0%Nq`Jx8{CH+XZXO{u`g6i>wW6X&)e{Ig+59(WpS3K2;N{l8+F;W^Y%IjfVx%< z(T*O*hdSrd%HvX0x7b-;^F%7h7UwZb2e**hlN*2u&qGr(TzJEL1UR6%B7l1&P0Ojo zaxs-kxmPhWG5F#}53OxPMLY!Rw z@*k+lF!Fuw{I%5w7|EHxz;EM2?97v7+KC7Bsbkm?6HPkma=|G12M%d2rtb5u^K-X2 zKvB(i>TuO5@cnrnuBg>9cM2kT+v5_L=7~yX@cle~{GvJ_W!99)qNm7jxieOgb0VGy z&_M$UI^&EOdJ**V_SBsj_)tt4)BW}kx^jCTRMo#^E(rfc-~N-qIZLE*{Y66kxZyxD z(u85(xiwTt*bU}ccO1|6pgY$?Pa_|%<)B`_>v)E11U@P^jemN#3sq$S#BHA!e>2mE zZYX4tlJQ<@%qAVlE9St$c~8moEm>Fx264NMMR`QyG*pXQy2`&GU+ueS zV6QQ&-+uz6|C_>}J$411d$R$5S(?JA9AAu2w1r}8^&+G(#So&~n_%0u5OC)mLmFG| zGsQmTRPMPrysY2UqDXHQ_YE10M-^O=I$Dlz48*|8k<0L~ED$PmMwl+G4*mjH8C20F zfW5BmMP6wVAolYC)EKNnk9zb-wEs!G!_@_T^UGmVR|Uyju@Ex12}9DvOjd28ls^Ws z;1L=|Uh=yw9J@*#!oY%lFaT_)>MvdKjXoPdB+kpRmHF7t9E3>-B0R1gA z!qd%0IKIkyvS@!fB|Fm`?W}$Po3+P4d5DHFuPin~_AR^jnLMl0FNIP+-$mPx%AkpN zJJHmaJ}|>I67tIh!M`F2Ud$=vxSWnm;kSjn^)UlHA8s0I=_v%eHd=zub9mTX2q?Ngm{-na9DZ z00cjd|77HUm6DW49WWGqo03xrAt{23UX)`pOmk`*tPhH^X-j_=JT5NkFJ2DkM=w&`E^qr4ZK z>lLx{h8k38uVc*58U{RY3Nsm#!?SR}GWlP%=j^tYrUkn|wuMU+UPqGEW|-laW| zwzQtCqjKT#6;0mVl^USppGeyKZlG+J+vKOrDVX@`MZW1sl1tlT$%gvrWVPgJ%u4rQ z`_gj!HvS&s&DKG7TQ|esJ6Y`cor>%XF8_YV$B#Ml?-dcc8bYjhiNG?B?@t&0;&<+L zL59YD(3qgWDix{VwfYO7cBV8NH(5^FlJEtXINyVYAU7u3<>RZJ>f~zzd<7>Lu=wGSMftd>%;Q%7HVXF~E&1qi&HLVDYtgG5^$bXKv?=L4wGccthaYKE`k|gBgv9B7qqZP^LSD4O$_~UKqaz zeucA1((Lohk~imJN}W3^7V{O|y7vN#rUG0(d;m!f&mf+*qT~wKDgL;>1>YFF&x~c+ z;|zxud|+f3nWExOb$qkJTO`#{$F-A~_ecZeFFpZdsRs1)+A5XnSIRFO3@g|OxC zZgz*5G3m;4$95l`>8$*-$R#nDHWxC&p?@2oGE{{v*EQda&kUAUKE+xXSs{m+ zvh>V`QM~tN58_Nf4c$Y8}c^o9bumOPRH$63oLAd9UP={7CefEf<6t7JW zWvF9|UO68mg95ckF*O`{MF~TdKr+m#@SvQMev=0}8OX3vfX;kWYw_{ay9(sA0<4#X zfnuNqn5kO9qt}V(;HOg9-LjvZYqEh|(I*DePiKMcUM03qOB4#zGtt6@&S=jw8=M{+ zjzvYe@9B-*yap>{+Js|r?0I9x8a&;>dhHNq-&)l(qN&>Og5%Q<_@=`2FZ0Cra^Eq*gY%H){XHP=n+E?LzC%^NOxdBIP4KOyhzjD)wD+%G!MiUTFs|p;F?Ie` zRQf(WYE*X_)4EHL^ncNVfAXR15?%$Xar7P))n!KtIS$mcs1EAO<6l^$yOX^B`T%$4 zUBWXG!eD)$9MAenKj*d<1?SiG*vqUJ>3+M452{>bG&xq3*3=Qks7DZu_1=Kj!P;;; z_zZO3+Z@JQj2~z`t3y3}4o;L_FzIG_0hKx0hv-yd;+L*;GUMJQE@RrN83WuCE|R)b5jHS8nz+ zrI8dv`dTcO8-2)#RKA9y)q!ySzdo{S@(Mb=TaNjXpUA9y`G`5V zDZS#c`5yGT?JlalDMZiNBa8pc%BDL06rjIli?P7*`($kUFy0$Ffm{DM6B`X5tmv|b zEPcQg&|jX$FR9%)?durhW~fQ;i8P>%WA{;Mh70J75EdsoMDkrY-_eUb9%$@q9uu8- znwYtXa8NQ0{v7qeg$o=N$Rph_G_)pa%5DU^}Hwd>I^H|JTAZjJB9 zPQ!UuLU^;DMW8N&Hx%s=g|1kxW{he+kd3ucEO^4)g(ECl;NHq%m zcy*B)T9}IZ(zhVv`To@7iZ|%%>lXBTrxNAf%LGYD0zRmgOuLx&TJ#RRstL#%OmK+Jkb1QgzcX*-j=(Wrs zn~(R}&BO1s50N&D1$dFUFYlyh2K93CEAeBW;krX7(4rerykwob5K$`2e&<5C#dT7@$zr9gH+a|P6SA$N9t-!; zB(P1C-0@RI$GUgp9szCU%~@ll5>iXuPx#5~-09B*!tk4nl2WZp-_Q$3rmK+a0Cpj%|S!PeV;eQ1NMe z)%*jqp0^9lGosLi5c1e7|?P{LB(T=H9@1 zB)Yr?HK%j$jqTfDa)%16UAPKcA79B}v;AoAkS-K;tiT84kC1Pi&n-wm5DcOZaxA$* zJZ~Fp3Yn)&KP8Qd=_unvinY5P|zD)y8+#E+8Jr>}4=n~w^t)!YIN65_^jie`pb3{<; z`0@r}*iw)}Z0xMCz2z0W{YfnU+x`qlXe@ylqJo@TD+FZipO9eZv*d+=JS=yVK=ZYn znKz6+_+P9ihZ}nt7sF$4$UhBJ_OhU*d>pM?l|~-wE&$zgf2qQk;Y`q{2$+a6U^d@L zhZhc2a3`N*NZ*|TJNo0uy=kc=a_1rvHns>lUWY>5&nF-(?L)lyIoSDwH%Yn|NwnTN zBm6H3!rQ;XigqP%Gt&nX>t6)M9`aTsok5k4?-CtZBXnbTIhjxmMLRp@vJHz(;Sa}v z5Hb%Zvd2?FG1ZAREtrZNz6-EcGVkExmVWZt?JH%*?TMu_H(ZPA3KUWA@ow*1ylOG7%a~Fz^%)#p3)#T5!RQwV^5bF?f6|Ju}}l6EVY#!SII)Te;hz%I2`txoyD>>1;o?8n|U1T1ylbhv-5LqkSM=> zjJc*e>(CsDGe6(ur|bwPU(-I25SK4tCl;AkM3WSrz5! zY{UE<5UZ>PgD?Mxcs7FVAC@3Sa*SBLIgi(Jxs<)%kHWU^lO%*aM~wUr!{U&6P#kv< zd_7`O*bz~Pxf#P63%HXNgBg57D?XHahQT+dKT!ISMXIdAPI$uF;U}#&M)$(8;O?)bN!! znBBh=eBO3a7m_Wh8HFFnm%Z9Btp1T)U2+F1l!@lpERit(csSPbY(&&E$9$ zOH5m|Xp@6n-e;g1#$yX1e59S)v>=7NmA`|ljk$UEG8w!}T$ft;BAB||C=LZ{Q^0up z6L@Vm4s{KQ)Z){2Xf?;0QOGMI+KTCDpwyfAwpEf1S3aPck2_%wxyE$6Jz*Bq8ZtJs zIS-1nDg;a@v*SOHlD^Wp*kX@3I&8Fxn-6Y=?k}f-KKqJv-oHXVarc6ScsmJc?T5ML z3*p{;KH2is0qBob8WBaQxOMa($a9kSC*db zQiKXT`^ldknyiUTh3M{lM%c2Cq~x51UdSUUiaWu4hAJ6R?Is5W-HG!HCurU&!wT-^ z`0q>a6E*cRv}^HCCgb=DRFX0u#^%`J8%;H6R3F3LT}m)_*cgS*^+%DuD}?xIhvI&5OjhbI-#sdt30Ru)(EvTj0`PCn9z=hQBV~ z98Daw;C;^$XQX^|8K>$NRC=+5A9PC+HVxLm;>MFuKD~rlu*4Gc&zh10T>|j&`gA<+ z&?@v>PYXWSZbR8)K1|c$WC)QrC6W*I_#x(8M=d82Bs;XZdCOF?kiQ5{G6zVJv?DyS zkV00YYBV!ThtB><<4?=Q@Lkm>=pYor`+X-s^jSRFdg&<1;Vt9IkI$#v%1%J^Q%&&k zxQ4vK5ZUVCi3V+VQPc9O(ZrH^bhZYQxffbUO->bZ^1Vfl`YeO6cj;u~Rsm?ZxE@K? zzoMSx9LKLj6+mdcHOOw-g>7hO*nC$Bw*BSu^8bYBxVjXM`=tjL*Th5M-5$8YG*^g* zUWC7q{iIZ-3c8{Sz-GBR$2}Fs^%iR&u&|kUN?s!4-;GIp&0!|Ja0eN6yo)^dq@mwo zFNiTW|6kD_&BVw0;3ZvL7x(#P;#TsQUuxP&toDC`mV~LS+IAUSXY-aAhKqrNqdwEv zrGu5%mJ=tn1^gu!^r@#3cG$xs8vjlcA_mhvh`vk*3~MQ(pKevmW~Fr?ae)sP=HJIN zj;};6XD2CxiIsSr%2^^KlMCJkTDa`yd-ON`IHk4hJreld#LRY(gIu#rJh6U1{_x;9 zQ+CE0a;aii9Oejd-XYB4=u7;{az32wu*K;XpLqXON0RkM4DM05K-4{@{6vj<_lN2eFdd#anVnmRDN; z8Rdy|p}@Z}OhDsfB0eq(HZxz7=Ibl)7n3q(rh^(DcpX5BinoIHVHXs*XBgeP6N`76 zmE(E&MU?-SSInH(=gDN}7@M2d}f=_h68rr1@~4kfMI2?snI$Q+$luuEnKem~s1- zqtiPT+3UGONdFc%HHRkcITn<^Z7+Pbq~Y1SKFl;4fY3$Gq0^AfZ@;sSC^|(EowU=a zO~)7B-F%EjW*$N3m2XkoF6V(qs|#e!CTa1r%$wBp0RVZa1FejI4L7zexoLiGY&ZNF4Z)|KiMt>l-*!z_)^D7)27Zigh zBMbuXX}m)23)yF1!SNrz5RZoSq)fL8B4lI`1dr3z6wRfS`U$aMjNgO?j(L2*CIt*bG%=9h&OHBX3~ovp-bsp-y)W=IF_UY zdG6{UD7unp&C@`6%SKQy$8wz~*F-g}6DCdEOfgu50(qI2aOJ=n5K5}TXx&^$8$5}c zj{QVi3kzX;m>IaujXj?n{uzVAwM_yV*RwkJU@=Dn#gfRl&sF;zo8a*}A3+t{)z$2ySq+>=bs+(N~ zzEf=BQt~zUc3XzAFjIu3y#F7JFNC7M^O#aPk6EQE#JUcLu)h*-fQQc! zP-*7;8yhFdOt(j9v$_ep=4u*VxL%!1U+)2iwB%A z>H4aI>ek7j*N$Z{Ao&p97#)MRZky4$#VWY)jw*`hzIE62&%;?_%}l%S8)RZ~6z3Km z!dh^F-EiVA=slZ3Ze%}1>3*j$v)c;q*%3s<)7_cq9y`pNjKaT`D6$$-hhav-2CA-V zHMP;$7-i+jv!Y{{VEdjfqHFkr?-6zriRQC#ro|p#J^TXH9)@8xmfOV#dEs@ZZz5aH zq4KUfgQ>SaMdI#+5dFzJWO?{$EanlymTw9nM$^Zs6??Mre?e_z*Xp;(Wl{!qKAA?z zZ(y0%fA{0UI&pSZj1n{~(I>{qVNhC-&aU5@0?c46ob%&h%_B*0G@Vnk73W~6`9m^~ zW#R5;e009j966uaj+ThL!7WK+#6bTp8Pa>hWuKQ|B=EcIy$t8?x-` z=rG8#P-K@gDvV*)YRsIhV14bCG1N|_yDD0U`STe}NV5v$Sw{n_DaGzO5d*q0JXZYF z6_Ae+g}(Xw*^JQ`HsypXLFF9R=y?rOd-)xCEicY4letFz4SKNBvZqPy)F?*a{2}NX zd<6v-LDcvw0i0hZ!G>8?lV#2C!AW8xb@8YMI$^7c9k;z81B=FB=08oef;)>mbm=7H zqg5zihC0^gcuOu|O0Fd9j7nuTo&=b?V}j0CD#0&W%9CBY^A{CI9CEXul|+l-?=w zBmD9!?xiL$9c$a6Up^m_yO|b+D~6?$!kN#vlbO=+8xXW~5pf(lz`oY; zXZb$bFzES<($cXcAD(2w!R%?QclH}{yDE-}etI3W&+8CB^ad<-o#6e(G?de^o8#uM zL1m6p!2Fj!ytO%r-*+jH6E?q?n!TCK_Gd*<)-wZLGC2*A;{^~hJR6ej&oddeyFn!4 zI4o3^!#NXbz?>cB$6u{R$znR}qFXeXcTJ!1a}=PQDw3fZEg?e|=SX?XN~)8~_Kh!C zhE{pGu|^x0ZI(Dm- z90UN@J>7Jj8ICx~)=txAF9|(D@;L&uB>xFmL|lagxy@kmbQV6wF-CYpZg^#-CGtEd z%{MJK#HHzC*tm3(ypwN6T6F^8?|YMPzqN}Q)BZ)AJ-JNj#*ai-v7D^@SOjJ}?-Rq> zX{6ITiQL>#MG^#UcoSPV9Y%0Bdn#steWf#xteHOsnq_-of%O^i7Po;l`RCb_wwa)I z!4^iem$7!il^lPPb0NjoLrml`YGch~((3esoA2=0&64TR=qblOzsVA+bQ*qbl z|RaL8sg{(`_)h`E;O#T|`^(=MDE9(z+y4sZyzuQ1b-V{aV zigMi5Fc;@UPBZXGtY^i0NcnV&IL6<_Z{B#}$CKLlmzE9o zcIu$K{Ay8HkOX~xLnAG{F_I)(-^3%jE@Wk}GOg-(6b}qq;}g4_zAFvon%kdHFPAuJ5)O5}!f``zOMR8)@Y3myLLz`bPRRj^{FqELqx{4&_Q= z_+lwW=GP6mxsCzbJ0!#k%Ps>wQy<*uE{#H%5*$9lWwE4AW5*{m=xsd|YMSxcA~=*s zue;?+$8TrJY2I|KKYs~jvN;}~b8@9+mTjQZ0^I2r%6HK*mn7^O@D%gPV#(>lHuRlk za`chXhxpqub-FiJ5(!)EV9KofD-6#ElKY-VnGfnx!kmyaF zQ~t_)Y06`CB)cfH_M;#=;e+Ecim=K)L%x0a6$=Z)L~1;(h}-A%p_jjX$+N>#Y3=hs z`%IKDN;)I>opUGLH#Ns9gXh4>r5X!ZCo^j}?)CHO4)o>?m*|I^<>>trv+2ZP2lO9b zh*mh3NgOp^L#k^rzw!GNVlAYBk1UL`2#w9c@$7pPV7Z(?VmvN9r$E2z|Bvb4@R&ct zQjFQT@CDv4_8WD(YawInW}bJF8yQp-Axh4RXp8UW^ahG!9t-ZHI!%6~dap5F$-)>i zRQL?dE<{*(GJ~G{zL4JcYY%N2Vt|CwR8j8V5Lm*P(ZzPl=`RHk4u$r z?sr|yNVpy%jul#XXpu9$VUHsWmDy48w-!<}1J+=%-4P_JdOG#WKa1Y~SAm`uCPq)& zlmhD|&y$094x^6CtH8itftT{G2RS7^!C&2D;Plcooc;JWbO)_OC5$zGZ!?olniWpP z?D9;QG>+sxlUHI#aw3z^fv4w?B3uwD% z-yu^=3xAn&5qIhL^0FRhQC@G(kRR!Z*utO(*{V}`p;|6E8uOTne=|YF0*mW4FJrqq zhjFUnJnC?mBovDpa`P<-w03YEmQqopc~zUx!99&&c~%g&`+lOrVyh_Am$M*W;uHEk zHil{+JK%wPhOlsFzC~UA0JhEK`WLoR{O0?7G;DZ)@m=Xd^-q6?GPh0x*K#@b0mt0J zwqGHkH5;Pt%Az@ss>y|@73i5{Hcn6prM}g3SrW}#*gPZ%AcY4-i_P$|A4X)tz?eO# zWz35C3RBKy?!?71A2vISqWcboL@GENUh10y$#;cSZCkP9ztz;NWnR3Me;-oI9wZY5 z@gcbXIF$9>Txv1A^BS4*WDO#BPoPiJIUm{%CG@AMllrhG0xd3?hA;F_GJWx5j8eJ) z9k9~{e6maMwV8oTz~*M&Ow%GHpLhfPT-VJMZKI()tO318e(aqFDb`I^fqnXX7upz{ zk2AY2fu3GCyN9)8-A6yMm9Kx0x==Cp^1e~<3f~8h4)O4>;r;kjLLlYawG#b)nt)~Y z2vHurQ53F8Mi!1O+Cq zL@VNdq8Pmg5=bDw1iA`eqXi}t;KrTN{|j?KLXkXH&C&^uOeCXwXEUIA(`WujTO2H% z+y*7PXJD3dH|xLO28vH>slMev1Rq`_^)q5n{VL9X^*x-GRk#mZE$72i{onlioPSk5 znIN5^0C2d>1Cc=k_Ey4fmVdjB#mbIg%d-H}`J0*Rom=sMX)3&~Ghpo>D(ZJHc?4 z30oUG6=r=r2YYMwqrWoQWc&Ig($}gFU743*B5oJo>unPo{H2*)X>pgYv}-SFpCOEg zYhqcUus*ocq741|ULbfT4L)yBU=6Kju@_PwKoHl-{*d+5a*%9_q&$;Kw8g%=93 z^7rsMlOJ$dV-7^TBy6uhCd_flXK$L$z#|Du=xf#j^s2QnuqgH`lfR+_z8p#-K|15i zyGcb#y6Pfz_}NvI?+>{3;5VjUJ%*CZFu@ZWLW!Zs2Y6c$2~8&%$T+x)(QLKA7x-R(0sKR4uy%(DnO;x? z{9HlScr=Cb3XLMmTP-;z@j@K9<}9FxHB3&uFFT&Ng%!Ho14q9k5cFAu`XIIw1%6wO zZrg-$%uqi(NgFYXej31$^)J9q^f2$Ze+;QUUyE|aG1XA^1K!20r@Ssx=xftiJXPid zWbBV*7Uk%&YvCf>-!O(YUvq|K9Gm9Ji&P|REXe98{Q|>+AU3BZ4-}8LlOJ_jFxaRL zbKgHAfiWtudjZ#fO#esh7R#epr3<)Ne-2$^>xvuo)bSZR4XhOxL?sN@TD)sFVH`Rw z7B0+hu&B8@mrS!iO11^=K-+)MM5bRunWi6Mct4IYZ`MGkLobUgL1Wg4}NNk%39?~%RbApZ6DKBDFN@u5l!`ap3L zU-!XvUdQ4WD17}qGUlm-##PqfKVjGSD^M5`iIXFT6PSF#|E*+L_0wqI zp~F}iQt&Bpp;6Hgdh4Et}~I?zaKvmwER6 z|KsdT!>Rh>JuYOX44H?D6p;pJuVEWXgJ?hlg*2zZlm;aV35gUkXDFczmFeuYHqxk6 z>Yt>Pp+S>KljhugU*C84JkNP^-kj$-XYak%`u@J(&-WEtm!L^Z-xrbg)jDkWu3$#$ z=)~G$Epy~><1t=wx`e&EX$^aSW(ACyriNVFjxpmj=i__!LSLHYB5X8ljY+mLTR5Q4 zmM=XF702hG()6GBmBI&-V7eP6j2Mj%TOjegPy5lrp}%BcEeFaR8_?#^<#@pEGaBCR zk2T(=q4lp;!Biz<{C)c&I6Wm54ksSrLZk~oRq`qPa-tOb?#xzp%Bl*IDpknrzHWva z;tpdMzu)-SP#oU4p@fSMaU&D_0e-ROhuG{>0^C~C&2{a3fPU#m;*Q&DSj5NSE_G+N zJh_1V9Grl`lMMW5=^1=%P8?_4Ueg8vS_Lw0wZkilY0+9$IM+l4(s=f6TmBlVj2m$`#Dsm~OUxrW?)8x6efR46X* zXcYJXS=@h%W^i%O;_(%a&uEvk8XIxWpAC}Ag9}ox1Jm^}NcoQ{S}^Se+E(>B@a($~xEK#;n{tnK zI&s@A)#z*R7#`iKM~#+h(-fuI;_EZ}&@$hRc&w-!z6QU9{Jb?@GLVMSjw|EG0aJ1R z-mB=_us!~rehqgmTZ2_UeZ?=vNMZ?}_S)IswW(S|Hre9q1fpBIILrEtcy{9o6r^a4 z`_9gQK+chFSbB>78r4I`bo?N3p~9J}`x|_Fp$x6gs7D7*EJ1p80zcwP99Y&k3%~tm zhRPRsa9QIfA+6n$IX8_;?q6>yZS_4%B7)or|JHyybcUcFYemu$>5sMa3mEm_bhO>g z3;!)TkHo1C#IAKc`XrrvWS9g?Fg!j&W%FigZTnryZUC zZVq&s=7P%4%M(987ntpyM^wd!>5`nYSmjVYxuE%pQ}6i8HIx|ARRP%~d(l}CdC8@= zz0Vjr8t|z6;~H+}k(;QzLXB=G_lfJXOSo(2WudQwpeNVwpfZc!2>q*~$F|F3W&3UT zLqsp0v9t~vr1S~;dLAr`%0MGtKNnAL3Mb`-EAfbXl3cNR5sBKDO+I?*6Ay(%eEQ@R zy#AR!x^vA7|LYSL*!Abo+Uh8}_|SYV)W#JmymNDB?wH4Io@$TZt-b|SgdRoZk0;4H zt#e4rp&q5xe&Qw?d*I&^i}BiNchTLx&!Divg}l|d#XZXoCmnaM2rR(aa9N85Y7M)@ zO$aE$n(Gy@*|kzqH$IN;>TX3F*1SSVmp(z#tb`g9X5-1tCc^h=V;O!BdM7-CZ~N=W zJoN)uZYhf{rD~8$gB0BSU<7;FXc?*0v!FlxHR$}3lj0}Jr%=(4t@xJBb=uj~B-S(g zNBUdma@hy=VihS{yvV-^`1`e!onNk_RVQ?j+1Qi#VSg#t_CANmkfXRPW+wH^;kbF{ z^yoE}?X=?8XVkqT8mDB(#Du0>x?>9T9@J>)WHNRTylDkRci9bUF;1YSJ! z4~4Il$AhKE$<2g|XzuO`@v(WD^xcQ6r1Ppa!R3F@Z;47ASyM?0oX2A$nM-7^LyYK6 z=?NjHejL>YKBH`J7VX7Z`U|Rf*Q87&FKLX<3fwL5 zA`iqCWs+$vr?|g(A?>nOr;gQew1Q*-mzT+$IzG+?xW^*5g#_7#AEUzt2KY|lC|o=0 zIW-xdK=(PGqNWOmiTFP^{^iD(s77x-Jh;CGneA;N?viQL*y1rA*Wt-8{BI_&e?ngW`K-ujYsoj){^?;x5dAld0HjRRDJHA zz}Zr_kfXCJ_wdDIWYf|HIqzBM)}U|>3;z#3(z{%fRh&fUI%(nA-`-H?>Dy?}_4(+J z^dwlGmrWxYdeNWCD6;JOIC^%#fwqs+;^q`Z@zDX>`S{JT`~h!6TJ%NeLB8li0*Xgs-$d!jVuWbv^wx^S3-}}GrWhxEMCh~k9<8>Oy4(jqj!^6kz*6C(JiIE)G0!N z?yuNE9rWYr3*&R-e0u|_sa;NnLw&K^PJ(0{vuWHe6}oA|HPW{ECr$3P<|oda&A0mM z(?2hhYbWo@Ao_jEytL&c3No&Mi>(a;Q#e%U!(58S*r@PkFNF1dWhrR-6i3&Y>XIJ2 z<5W85D7t+u8NZZl6LR)<_;sa;eB;+SyzJ=lbXYEpMETUC48QaIhF%k1zx)(k;<}7~ zlV3qUyj#fMmzc=koOhh}ifW{J#|CgJDMU7zt!UDYM!0kIL2%kxif(wBNR7Vir{~TJ zvx?MwdTD(F@?Iw^bX&}1O<(09|H>EieA7s(*lNt_zJFL7o5}JsC&$x{b=6{pdM$J@ z+<}8Z)9r)}WR_{ia((znnwVV-U_WXKPP=28!*$E3GKhCg_1FHvtQA~n`eCQl5uCih zRr&ce3B#K@*m=1%LdM(JNy-24Qi+r7D~E3O;vF9E8y|#??_`jyN(=I9wj;MvvH`nW ztR_z`#9@(LJ8D!iAc4#koYU?te&d<}ZNND+)guK@txDk@1Pzg-hnK*z`s4UQ<_h?) zxC)P4V}-Xa?*-%3TgcA@b3A2tG)|uIgZ!Aj2HU+a!&_|TpvjA$lFQ+v*>N5V@M?Ss zc|Qn;-fimi%Pey?Fxr`oSR&7EhCL{yTLq6heo){qAHYRhkCNQmUbwf|3U}!i!%Y%8 zEE)PP|KWf#`LE&0%K)y#s$kMY>j1MFA#=Xm)xU$*D5F}v&D z9QH?lCXTc1?otPqXA`YIPq5Cyuifqm@|A z+H{m$HjcGPy1==s402c8&DiKFUA7=E0?S#Zv5$QA;1@fy#GBTg#E!}`Br)j@l&Ee* z#=q+Dlj1&fNAP>7eQm`(FB_1J^fCfWx{&Jae9qUV1Z#;c@d(A=!1sPGjn5CDl_pbJ z$$o-+au4C#ZE;9R%aVq8<>HfLB-k|@7ZFRV{Mu0mDTao-SpNQRd?W4zy!_u!?(trI zlKlG>n)9O-2i8g1XpV?tKfdio_ZNRgyN7I;cf0>1BVy;VCS`4C-EzzpDV$_KfhFwx ziePGf^%yQXqJuU3w&9pB+L%8)%$@0+idDg67%U%+Rf-1D_s{Z3xyljl-PlWPTjnF) z+J-CA9%&Qty^S3@V#^N5>aYecZP-556KM6180?F!A zB}Ut*`Jsd7JmH2B?^3{^A_|mga@nc_+cqx$az{C^WKaO zpL0WwU%NM%`<;7%esB5)6}^<`Td8mKXP`7+)?#a&P+5q>M9|pD&U)OyVZ7 zIIGErncc7$|E}9e_ZGDh>BOJ#v~MDJ`c%Q>nfRW6BCB)v-&?{f`&d56I)ndRlFh#rPolRbd(&zLk*8+H z{A=|xYBX3ui`3@|Y&%(;9Fj_V`gYZR)%6jZn?6R?@8rqB`+`OUQn0M{IL>oUAl0^Q z$u%IapEsYWQ|Bt4V?Z17|I|F(8bBY3$ZtSD)UPST3+Yl$avYESJ@&vh9 z_><|~@$hzU6^$T@{Ck^^$mwb<55iPvfvKK&ets)eN&8Hp&QbCCGYb5eZ~FYKhy$?o z`cg6zUlQNnl*LI%Mq|&Ek_~iOpesgUF&n%bY18yxsg*FRu z;=(W*=ahrQMn=>-#f?k+_nv0{7^KFf8e;QN3AAK%J=w8uHBL<8c?b3;|0Mf_;4@97 z86cx7&iuXO4Q|;%_OzGrJ{H#COgp5k!Q!byztIH!% zf!cDqR3V>dcdg`q&4{L=gNuk?8z2{r50fwX?$ADCBjv2~U`EDxY?#wSed-ly&lx|e z>TsVfGJ1~cH5hv4w4<__IQnML1p|Qp zaJ7rSx8^0kZ?Q6;k+zb5+P0f2tx@6ibuwI`DUS$XKg zSW9vae>BElV+L!juq}QW`j z+$o;6T2H3a<}*C?KP9jXoN=V)Rceqn7YA;=fi#OEnZKt?(UOPr@bZO6p^3O0NgFA1 zuj5Zr)h*Xhq1AdcY2p`19A=Z}TyD)=rPZ)wlUV3vItOcfzQ7TOHi>(*RH$f-60xYZ z$8`f&p{{eBIQQoR?Acm@*DE`azY9OowK9{bjK+6bKdKRjq$J^j{xLXOeZG5x18S={#+={z5Je-P(s#{;nf# zwW)&l?>4=*yqKyKYob2aKQK7QiTjh9OtkEnJ_ToWrgE2x=kOEn+4BK`#Z=mfL!&e_Fy6O@ToP|ZnKk3_gG(KBUxz(j zk(tbyZObI(B6o80R;S=qAIUGjd5qIG_2JC&_M%zYXHavR4lN6bA-tji4P}wE`lfXL!cscqyOh4X)I#QKej|Aa%_wBG4B6^GhNxaj z zji_#3H#dH>@cnyqN%6@TVjahj%}@N{wRBUG^ZPrQ#R)k5k!56HNMOG|O~lvj$FX;j zF&0f1p`oc2G($U<1{JI$0qZ%MS1jTqW|`o3r`&Pd@erDF7n0U(_P9SE!W;Qpk?Olw zP~UGVZ97{{nx^Lv(PSkG(y!v5<$)B1O9;8OJTzGTQ+%<&6*fe;kRR_2;aA>^D2Ub~ zSFdO~<<&oObz(5>x$4Z1f9T7f(6yjL`EoS0GJ@0j^aFXN=iqTOFOk*GztG zs=X=}``;6jKc`%X&R{OFf%nOn*^8+}c>-S*6ipM{PSBuyA=d&&@Cl2Xi1{#f#|ub9r=3e=!-pxtwla{eXNwr-B+}y@=d|1MJ2CW%k*S4EyKP zb#9Z4E7z&o1YH_*!O0de4tjC`-);JVZKdVexzYrmPw>aNPDS{!*Jl#_ZVpR#P2(<~ zKSY%F9N|$sx*%4kYeGB-!%8J8V-5Y33)00!1Q5PRDgunOT%@Ta2^tY7IP zsCB~*ttnR{W@_j0!wGe$#_2Z~RLh}+>Pe{i{S){kr50P|{-Hh%6VT5zD`L{5PYSDL z$x6Lq>R!@~OOYmSiI7L3KdZqNw;JLYFp|n{7I^Eyx;%>CgZ8|82@YJkM2Flw2vhKk zGg;+NU&Ralw&o^UvuiV5GdYKznSO@II_T2HKGR96^ECSWWHN2P8i{5Pdg839E9fzo zW8~*=DeV0>7b+PU3;JFYvTeAIwj?s-;@@X*;#UW(?22fV)IqA4q=2LoEm2lgIld{g z9fhf{BpY8TCH)Z@Szs{DEojjJu^%7?d7?`(T&7pBK2cMv!#W(AVo%F~}Mzia)6 z9iiyKe6I5De-!pq@{PjWlC|4NKm3&<8Mfb$*;`+(Omi{Z|5}wd$$C$GPA^9z)MEKG zqx-Zi{}OE*4&aWz-wkK{T*!4jEQITS4WWV`U+~M2H2Q3c3GEuOitkz8h7RW#v2t2U zIBv2njT2kag0Z3a?v_{BdoYgPNQ&cQ&E)w+G97Qz(xSa{_4on(aYV7;9?`jSi|Phg zQnh(!(W2-NG{tN^yqJCxqpEnE0<&qi%4mL5ggUJpQz3>&M%U_FoZ`0FTT$`K0vLb& zDcSJo1`+)kq>t_2@k2+t=-Jm&e5o?#vlB3NpYVteNYUh9_$Gq5ghcWAb!BM#kzLfM z_6o{frNFP6zK)K1@SJO^Or!N1HAu5Qp`#mX=&6U3u;pqIeQ(`Bs}fXExv4E$a&3SX zzL-uXM33O}%~kn4=QeJ5$y$1?CWYn(=1|Y6^XQ^ywX|zSC^8cKzy>p&`J?y4ajD-0 z@!_t!G)-DYyt_^gP1c)-u5}%xzF!^dEWlfSr)V#yBM-k0FHJgtg zqstF|gM7**TfR-1#T84FXr;j+a&?70r{zcZ?2%JAHj$DDdTOXnkM zte%ILl;u*j;aOzWTRHyhR7rkt#0<=6ZKJ;B@w}XS6g+w>2#@)4f^>!)fze~`g7`%- z+{{`#IgZ-!{Y zd!f@&^%@s@^EcVt^Nn`3J)z$V578XYTV(a>RebP@X((HJ18)|+gOqQ~5pT=714js2 zxPyfc!5TAYLqaCHJXMC&Ou0tx=QokVoB~OHJ(Aa{8CiGqKrHv7SqCY>HC*h4ShAvA zk3N4b$2iiFoQ#YH9ZoZOhtr04C$Fh-> zt{&t^UK#`S`qyyd)i=i{s($>8nxv{ih(Z*os~aaYTD|g`#0~1+K(HhQDo6 ziAq17g2Q41&6ZY_eJdn&=Xfp&Q6U=q|+JINi{ zR8Lo}_(VPn{Ux{8p2Fedt2w`z-$a}(`1RrzK`)!VaJ0h&nWTuNAy+6;4FG)Q_db>u_$k&s(1EwPEF=USxD7#>#Xn{qy_89s+CzaV1ep5WT zg2ksM?ndi}4&rBnGl{zKR4z(s9c&n-h{a34qd8}zQB9sdo}SbXcTEr&!@Z}-3H2NB zi(3bG_eBMkTH?egJA48HG!*a3$i`D{T*0dG1H^J-HmZDgfwMU>o9<7urJn*vaMIQ* zh}o$~@fMBw7@U#A!CIYATjeb-j9SB;lL~}A-(bzT(7VLlD+;VHQzWC_hoPuPI*{DR zWr}9}qq=*KVMQdz{|V1S<@tK#v#?LLxmX5$pA3lZ-IhWh-y73qMNu?PK7&StWA=K$ zF81e4Pwe#jBv;)OMbB)s$Iju==tj&oq-@TSw>H=DL!+a(aG5F|N}q%lZ~q1v_X7Ax zI)bJzoQwv;0hP{5P5!)>MM6TaDj8v{%5!RxY5I6Q9|0Fl=&nSw=W;n28 z6}}*~@I~B|vV}0tXd)@dZNyu`^zdMiKmL)qiF$YEaJNq$M@(@I3fEO-%|$bb?c1H` z*PcQgq@InL6gBSl)WhiYw@B`VRscS+Pzr@qAHv7$XVU|xQ!&lhgG?L8vPJ&K(YzHE z!GQ{DtZ``3(w8bKSd`zI4DJ4IHtn`J7b{wbanbH@C~}%x}M!6c@O&^F-Nyc z>d=*>(KtftG8$(mi?!NLaX-$ygX71`1JCuA_)kv@gzutJR+A>Q_h!f#5fUFuOyJx~ zIGm?*7|X1b!OYr8nE!kl<-JkjujlE}M27biJ3&)bpxVYrq(+Oh`O&tD@F2akj< z7bh@bmt5Gbi!b1z`Z8{>tt&tAtuL?V(@2drAH|HvB8-=}igA2!o#`vYl9$!MB(9@TWM zYZv!jxeA}`z6s0iY=msCIXnH~UbbxW5MEtziVHnlMZVS#0ngSQS8cnR`G!cVLy;vx*X0Cy0;ul1@3gubZoeH7Fcbvxi(Si1+#F| z0L<+c^vxe;#Hr>tdgVF`2L))7_A{x(*h=W1cyI-Gt{90O+rDxGhl=4$*J5M<@8Ytz zsTh7AhtHmnqJw*W!*?HR#JjIFqs9;oGGC{W%v3*#jB{+r*ZUSE^;10Vy>%Wce3r&D z^(glLr;SVUCGo0nhdIp+M{$vq(BE9I88&$-m=V>6Lv&qEvY$+%^BZdoW{z2AJnk(xYs9nHYk=mh=BsMSv1s88c z@wsP6G>jw-Hc4DZ!fGLttPIAjN+9DsO;G#f!(?v!!rHb@b>ten8R@@2!ugj}Bg>%( zS_3Hf|!Jv3!ISlF*sfJJPkhP zLpfHEziiknfIxIWGb<(Qle(7+EJmyMKn{lfpe?8fs0cE@bP;`xX`Yt zs3k}ajh=pr>zP}IE)Hvw<@6k~T|O6=O?SkO`wwHx#u44?`CR*#MOH<@^KpoiQY%e|;i=8soi4`eEbS83pQj= zw5(#MA{nw&awK+{r%wG_-lOiI*Tf`i6!|L~LrQIya?$uJ=R8DE;O`a$gJvNaLHm_0 zpNktlD-h;tF0{8&A+_Zhbk(z0(8QvZ3!b7zt=tcyZzi&Eoy|6G+wBH=wfP*yf~LG@ zV-@u9TSNQ%e5l$bL$bs3JnHudQMt7X zkW&>T<4ztYmWK%4CWooe_rbk;aSXXE*Wq3@>XGa~DKcS>4!6o{5;wfMn5M{_qVYH6 zsov8Gc>S%3L}q*zF;|%bB^0dSR109#-KpU2T1QyHOi5lMxLB}i^Oj|$qtYMxj&DP&GkH(eMg<` zo&6E*PqN30KK=*to?JpE^Giwp#wu=O+Bczt39%ApgyL%v^x~fgykzAYvdLBok3F1- zX4jrT+UqCLl*@NX!?$#F+TsBhZ+QVp{(eZl4Uea7^A6CKxZ`BO_^(K_Qks5xpGAFy z|2#qVGz0?pMd#YjV)`_-b4h zq=5I=%adadMfAamCA8hcp2{U{;78^^B1YTeph50kao8nQKCtLL-E%jH$PDo4W5s#6e$=)6gHOXr1nM z^s#9hO?Y$&mDF7%1F8MwY7;=@(kZC3N0+L85c91S(aXP7Vr-=ex-|0g)N4%w< zrzo<`K~OZ%?Oqc-{5@EU83#Wt4c!-N0u z)9d#5%+@8;5M*C;J zYjXWj^m+*o6^ehb~NIZRAT#$v6t|H1bL{@68G zikGO$;p=BErhBKTupcLi;dV7=x*E$9g;NQnF0l#bFV=XNg?BLMCt=Jrv>`T}_R8i{*&iD8!D&t2L}27>iaw4$ zE=i-J{y8*%j~exzok4d`6z26(AMk;4Vc$|UA202TBj-(zkyW3k(U(GZ;<)+ek@gcc zJYq)`n!2@%SS>Zh#MqFRCnDar;0<{-hodYv1$O%lbK%EZImt~fr1_K~IutjBU%<7| znQt11-vC% zGNdkvM&53N)lU@o{)QKH^7r|)+F}aQkBmar!aefxLKItsaqb#VH|H{u)Kx^#rVvJ;HAgJlh+L6Y0t7wZX;;*T5 ze@iT5Q7 z&t;

(+JmB{V5;K3|(!o!{&oV2Y1`%2JKRT2x3=l1FN-ex;;W1kxP z{mlw&nkLHz221iE=6UkCUxrunoQR8a+(`Z9WpqmZ0`@@A9VGSrHr)Gl8QvEobSPOr zBSEV6)cvUu=hD0fy)Kr;%e9^&wf2+T_wLAQAliaV#GBCD$!)ki3c|3E*nUP8KiiUzY~-JkwUnmbemugq}J zzAvP9_IdJs;y?5TY(vLIVR-cS5p4J6Xv~D`(b@67oeq}N$g)I ziDy4GLXQ{=lvsNe)dUn%vy53d#qZIs?V%{q;VEt>bE(WsLz?dPo%${A zCN1xa@iO3mv}gVYYrpO#e5xzy%TnX=PVc8*=wkMc>LE5{*D&6%V2yfrTBDqDj`l5C zg0sJ=VSC+rGVaVX@rz6+a^mk97Uub2YgIq|zF;4T5WM37eWr9f=Zg}~Eoc9Hvtb=( zWsn&rEFOQvfV}olC56o~;Pjhi*e%1EO__6=e#>xWjkKonnJ${x>|TYSAEgnG_7)ib zN}D>F7_%OOe}w&5FwPDgk7X?deSdfzljyjOD2$aMK1MFo=VvFoR=B?H_a)-6W-HF} zPb9hW>mjguu7-w|-XsP68)Zpuk2B=x z?%Bj8k zS_CwT;FnH4lz8tbPEKgJ>wm@IUn1z*+9#39p^h`r1a9AF`aMM{mNd?>%x zY%7D4#?+#~wJWfO{0Lf?;Z5H-%FtjcQ)oj3o;u;Ji=)KB`bXr>&2i|F-4SI*kyBI%oUt%q( zxG@P$>%NXPQa9qwUu5x`AQP&@uf<0DW&yithsaY_8!zGikg%jV+|Fw;#IQJmo;Y*{ zzWX{K`T1Xkkx~jcW{w%%Y3)r0kPrQ@w4YP2XHdrbz1(CaPgM3)nup&XA;e3N)r!5e z?^-unT)vC==w^|eX0Y!=wTr?N>f%#O&vdM!^8*|3&BXCS4|NW+c26dLcPxW$i<9CXuO2}?=J?PP&f~@D zGQY&w=^Qt5B~PYRUBnybj>30*=hF#8GUR$#G(8fVNPat6ATC^<))nkWyVeZCpBrDH zf=U}Ye8L2Yvz8(keOc7gIgZ{w@s?D@@2jyoI)m+eErmzeDNvif4`_SYL)5-Uky^=C zBh9N%!QYCz;BjaPxp6j~Gfc>)a*IBr)VtOs%rTxG8QUaeLiUjjH=?0GHQ`?ReIRkK z{bG<5G>v)cQHE~@aD>tY=v_s;`2#NoFyDpHDmXMH>llVWcF%q^CYlGKs-m-h6 z)rPe@*IJ>e*4EQSfSKojBs~AWuh+~DC|$G*<~C|TeUTe8W7i$g@hc|a?yMVNxlA9c0(s- zx345K_opK0w|vcPwmksiOQOK=ZVJ9+%0S7VpBeKj`3JfP-WIm1UU^K(@z+Gi;aDT~txH}^WnlRE(HDDx|h-Jb4%4LG4 zJ_200HII3}ybv@+7&3F8$1`=mrZUZkEC3vz3nnR>!g~uUnR`lmMdqq|K=L&SMkOkn zQE`k1S(~E4i~~PF*I+7mv&sW}ZSMw4eyM==zVo2+!~@VeRS_PkmWE03aiCn%0|fp! z%sdg7g4WOd0{=Lb(M4h4NWlT%eXSilI1mqXCPf1yy+WpHa2jBPZh?ZYi$H$(LB=4) z9GLn)1NDK5pvU(XQyASQS~2S@Q+HGoT)939maUS8(^{`Gha@H8YSS!`kvbOaE;t6v z#Pi_b=5{d6|0JAtaXY-I^bUN|lz@FB$HHpkiNNI83+VVI1HKA<2KslY!BNjXfY&iq zK)y*V+M(dZJkdJMoOMuvol#do-EJk|HRe3LA#k?F^7DbYTs5;bd<^JgM}s8}E=+Ix zd#38xYM@Q+Ku{tJZ>{JAR@dUej|I~3E7}My-~A@)dAc9$ANmfumYXn_H7|(9FcHk@ z$76x<%5LDUdk%>APG^oT90RuTzZng!Z_K+N(Lno0DCl*=K(qQ7s56-iX6B55(uG%n ziJAm3QtW_kZwTCWGZj!_Kk(Tp3bZ^e1k<0{0rSx}!Nf8j&}+I07{?q2s+}#2kwrE0 zj}v~IFrMiNwG%z%ri+I1kBYdyENK4X1)T6H6-HOJTX%$pQU_#1fpx^LPG*@yo0Kas>%~l(9gcnS#-%i369hmuXO&_cX zO9Ibx0^mAw6S5hBV7la3)E26bLh=^Ee@E@WDbGL5Z^bEa^GF?FCUFv&WVt}k>HN^1-3SRe}vte1kxGA+zy4@L0Ayb!!~>tM2rE`S*A60qXc640Ty zp7|=C&8+%7?A2^8ukK=aZa@Z+susC3%}$p1?b+4ZF{cib;9t+VF>pN)=i)AZf& z~o?QnT+KO;S{Bh70-~z**N5MDG>lx`s<)Dfi3tKy8FftYnz+XxWhJLDM(!4p* z7U#8$>nK;|Sa%-KnYjzhe`o?8YgT~+N`;_4Wdn1e>?{+urCC%xR)tAAenn)x>NE3B zW)+k(k$~HrOPQ1qQ&`|z0mO$_Gi5tjXx?iM@3tp^ZQDJW`!-cfc!C=<%Fq#1wk`q( zEWd)Xz$M_qc4>G=s|yUP3G;+tQ*i33KSsAb$|2Z5vFQKDh`dB%(imibnqBASz-Wc|iH zhH2&^m=SBsfdAAM5ZTrZ3d+@)?fUzeguqtNwSEEwZLuIg_7?mQ69CuBECYwpIq)BY z!DO8a!10^C$a0oQq@kz;UV7aJR-=``++zT2Z4isjEHnmMmD%8Tiz4VMD`W1&+Atz% z4>&2|A&3?$1MIm}1nk*i0{wn%2j@iBnPJ<*pi*`W?EC5p9fMTi zwRfKxMvh_pV~@h$GX6jzwnnt?-yxX2)(HM|P5^(93mCioE~9JfCAwXiCwgzeS&uAV z##ouj!sBs)AV|3%n3uSM*RKqiovPU)wI2L0UbN()5N}gR}Eado(v5Eb%=S>Akz zaa+^HyxCO(hEWV~ZW7kA>I48!ZWVQX*vRx=SR=aoP)#(U&=O@%RYP`A8O`rG!pJ*l zfh-N-+CDY|9pA7WSPgbC=TZ+K`x<+=o&UhBe>xu6w)QenUu(den=#B?Qptpkivc@K zW5N89w}8*w9-#9)*V^0jsAzKvV|{2d$K;c6(2k6N?%_YExyTd6eI1KZhLRYU5k;c0 zb9zJ{KU@ZHxQX!G0ZkZotr*-1O9Wv~R{%dL7nFZ`C>lL-2{V^;fzkVu1sTPhso$*( za(9kkLj3F)FL`Z|l-pKjgXMZ=n@bXO{2zv{!;z{sj3ax6jEpj}o4DuwJx7t0Qb^L$ zE>Y4{(Jq?^8A*vE5*ekPq!1~jq-dzbS4z^>{LVjcUH6{*KF|A`>SOWl=3s8Y6$NZy z3Z&+(MWMP;(6Df%jyv|$^i-Za{$-NK;bqicC50hFR?(ckskGu&GLz8>#`x-|xZ5$5 zz1ZwcQ!-@v-WYdQlIw{by5sPvreAbRUJE|4!nu|6AFl7MhPXp6)~<`k+;16Nef$cM z_xU8ubG*dnSy!-rR~x!;trNK3!M>O-fp^vod>1K|m@KBs}?-KbfkwfWe z1Dtwyh}|9~MxFY4^y7OxJOAJUHU;!z_q2R=E-OjsAUKTWH!bP>g(B*jHIXd(Z(~iq z0$tR<0SPNHJkI{e-VdFGbvK<@YSUbHw08kJ9u-6tdwl5d`O$P!2+UP+-6LcbpFo?& zYvk0&!8c_No9L|s>2Z5-UFrrdc&XCKGyTlNR{FF@i@J>RyyAitP-NzV22>ud#>g zEpbUVpH^3Dkluz;{Jy;$hnFdFtub=kO9ex6JaU+`TJxLh9g~c~&nD1=H!5Ue`2$IF zhO$)A3tYg*Hp}Amke(?GH&f0uruWC9W9U{B$rH2X=h)5Z zTAX)99n{th!u{h5*t+@>>`hOB)9RUM!f94~`x->QN|3lRgMAH7!%iK4G}rE8$%;1U zSgc52o>)=BLk~JvVoaeaD!44KN@nr>7~fDtkH0*J?-wyvbLJqVIC~@(H?oLMeW>{* zz*@Qjp#~ikl%PU#(>jDO-$HEuIT<@GHn9z=ZkU{21?79~IL|HTV(8XH16G#op63UfS<5Kl0VVihg^4M%1RrBqxa%CCDVPlc{&*7CbLj^ zRGvPbkrDb3Zb0>#ov;J%!MvIWP}ubuif^yrvL z-UuxD$Euy3SVnLQIpqDK@a0ncwh=!u)%G{u3VYz-@JO6hortk|lW;EM98_A?U>z^-HxqBeRCxhhfKE)qj33e*alYQK8fYGA@&}2mzJt!Cvm-4Y+y#)=4_T&~) zh0h9E+=0r|_-kNIdh5P&At$UMsTGIjSKrysJ5?w;*NLpgaZGx_N4B``JKV%(vCe5( zu%FP2I-?0}(VqguNPtfJ+~JOg&qtBZ4pkx{EGyp}hx9c=Xhd@_cPri#^P`tx!!uK+cd?l186RM-KUHYp1{1PV)h9crNM>)4 z%)J_Lkn0RTz@mHhF^R4`?%7E~!lcK-8Fz?9nkCTKWHGAJ>|jm^;GEV=V25oSBy#oW zQh7LC+gCzY{CMpBv53@WSkUE=PE<+SqHDA&ntVMtrSY!porw;%yx~QEf4mXBvz*3W z+r4J9jUpg_O&(^U{iNnoL{X0xkwfl#wk$n^m2C*&e&#q*=f!STxEidfXg?>nFN7J7 zz7CtSYsot`k1pul#IOG2I9?+~NpXkh!`iECZSo2nnm+;`+Jh+FP=(%1nS_%mDa`P} zX=HL+X!sIGs(05T_e3M~%^AyQ4J@GWMg0iUE}_(x8>I2`3j#jRz^kIqFx@3je(J)1 zte3KG?((aE~UN%5ge)oDp}I$L|ok9FPjfRoA`ma=ye-TWX=!`jc_ zj#n~id+nfwm09q~`w3Axy3yRfleoMQ z{LjkgbUNrOtS_x5EiX;VPpV;eO^4yXyll+Ap9%f)$s&3!%)kFE(f@5ZCpuBV)Ghp3 zRU^-0o#%6dtJiTUYVK5^nM#i4Gf6vFmpxoKj9XZ=lKbtNz`mZ;XU9ip!_8taI~#4v ztgbb2kIR$rL|=`>4yB{$iinJYgAu@6v)8?V}pDtO{^_>@4gG)%n|J8cF-VO_Xb?$Wo`RWgkxo7{Kxx<|wtF$^JeDo$;?Yxd*G!ckdSp z)S5&~bwZfR)L^EUstd!_TC{ah9I8TQl9_@me|Nb%{~=Qua@Rq%;ydY}-Ea0l{Ve=* z(vURa37f*nLw(l}@)_hpVf__Ykn)Nd9`m5sx;-?$`2rSI*;D+}^^kmO%Iq(Cv4994 zcIB-NbGA3)jc+E?fSQq9;=DF4+iEgerkT>kJ;FKgA%ogYGic#|35aUQq39{yG}HdR z@I6+LdvY$F@mGUt_bg_){Q<-W%;e&?Ph=taYbnzriM@+G53jm|P;%c2xhsI|Iw9Lj zzy>V;DO1SHSQfl@AdOKLX2+1T4F2UX6Sty`&j(@K#uRM(-i_>0kJ-;x58&lmMB$(3 zk+ZWE8O@WS)V;$|oD__^esZkzq!W_PUEx%e#WD!N4-m)`HBE$?8jrVYW zJjA2{+R23frpq~tsf$#r2UT%$(=B=xB5^HKKr!!Mk76Yf{9 zd)bSuHB>o4fzQzQgZ{u*@JNZoyY%Z^U%m&|F>5*dwJ3wMe$1xz!k*zTAPljqop`CI*%2f{g0wvbkb}Y4`e>9rQVbuFp8HT)#`B4?|*^92~v1b ze4g{W8P2A$92j^-!tWEL!y}B0PrrPl*Di6{adW*}w0d)4`WM2B^2kPiL z!d^F=L!a4FnqG2`zKu?#zjFto=D$Ul>$w{%Ui@Y96Xl@3;xJc|vKlu|1|xp&ZQS*F zkB<6R-0BT8@U-zPjMoQq6WqPHfyS;dy?%$&pC8Yr`sQPSLz+!~ggN{9O^LoWZl+Z8 zleELi4;~wiu&i2r=4i8zb1gr@#`8P5n9>|{DVyk&UQmHD z&t$kv>OoV98Ffyw$Mr9#Xsnt)l{n7C%nS`$_hc1aZ{9@nwe6^8ffo6Hy~vV3*1*~> z5xaa;s8v&)x}_YcG}WFAF6A?ctOBm!WsPWVsNH$=%>Lf35#6o;EK6;TYk-^S=k z#78vWIE~)$(YPw3hJTkT+2R4y;3NA)r%9Ifaq@TkavY z{#c34E0-{RZV%Ty{3Rmwy_v>1KTh2?m~}VRv4FWp*}qUj*5E#bE32rmIXvkn_o2|6 z(^x0L%|9}nPR6w(qU1DGhhAdNOO`U-tSqMg^#F^COctI`XYNe@QT~)&5!GENXZiul z=*18b*wpVV7x%Em$y>QsOAp~l_83ZQ z4W&(8rRXghjnFhd?&Xl%O!fPI*5Vh{~TOE4;U?=CP>%^XYU%-tkIl{rXgA0_ABir5(Xn0P>?ZM&f%w!4ZeV2s8-pQO_ zXts5+&So2>3&Ko_E99OXmBD8N72B+xsr8P?_cx;WHfihZ9M!j6oK=Vs=O;mu^@ zY0M~V6q{FxUYs0glXPRZ&B2MXkZw_9i|6j*HvRMwcDIdq?pMS6kIvTA5BQ1&`$zx?wy4r$7>iS<#88OZ7AT(dX}&u zoI7`1oNrpwNzY0i(5g{$c)eel@GUsVjDrPTRQWPe5Q_EAE+5799~{ADhfEa7{HKi^ z-`AXjhY-pWP|Z$x6mvSi-`R|CtmI|<57My>`|)ylC?{KI#im@X=RA6O{;5?G4)3+V zvD7%*jloj3L+n~Yy^xG%bQ#~7JkFEFDa>)7^%ds*bC z9FecD3s+yioLg;rhVw7WvB`8hl%kwXbyI_Pb~^Qvs7(_Wd+h=x!of z8zfKf`mS)z)01sB8A;f3dsXEy#zu$#9MbxMpK zSmMc!X9S}$XCDe&npoqoM0}LZ<4-m^^4IQv#^JbM6g54Zetum8j~H2uA8&{SJ^|2) zpFtbichWf}d-Anggz+a;Fj-{*j*pgrq^uma+7HFt05x=-NJGx&;dJ4R9LWhj;IE-O zg^aaW^q9`V%oz<>SSQR1ZY0tNn?P~AB>#P0FnvE(&D=6BVBD=iw0N``EH4d*TGA+3 z6>H<+2{Hb^ba~P`_YvkZ)F~ykmg!77%)A$C(CI7jY*ODZR{eS*M31ytjQKvc|MPKj zF6_sr&lged=|Tbjqy*g3jul-h#GIrm&OS4Z<+y&w_QicD5I>4=lUdaJxQT8~dXK8s zWVT&Ql5LVIM~>SFeroY$a>{k2?$NqfRUeK|89G$6;tiK*yPhpPm4rb`yRkX`1@n#b zW6|xYSor-gH*RerbJz97(c!hMJ!1j(aYJZ)c0GF94e7^}61Fi|9{n*Vp(f1!N-0MS z)e|^L#Y~J;FXFDgFhILo8hP4nCkwmtSX6fo7Y?06&*8u5Ol`*+%%mT)z99Kw9CvJx zvash_k@NaGT+vX&THhRec9=@rD`(QWJYklq?dLsjGJeCZ`7~OG@M4HM7A_%zQXHPQ zx3Gg9(U4pwXpEB_X!v|{Y*l`UYfl>xS;pb?L1hxJ+eV6E7b)RwKg$_>jtwii456@r zg6=BO4PR;6Z|aTNqZ%O5^$<$i%-PV!Y&OjN60UEygM;Zo7TGZ$eu4+W#ygf@8lz9E z@4O-n*BGi!iABtqZ^REb>nCw6b23Kb~UBAmZXa*yYbtXVJ12^Vl#i6GieE%ZxX7K=NIY?2+0GEQa`{dZH6 zs>|o#WK}n&`@O{9RgHA2`6eejoMFdUS7=zSC6jR>WZd!q%Nlk0Rhe6GIk=0tos5H5 zs5QhcykddzPEeJ=N1!R!U;jvib zWh@$CwNw=G*$Ta>Ey(d#LF}6l@bAqre?cbfi`y{RJdBUopu~3$QKy7(ZD!k~BQj7d zW(Ha%Y_pA=z+X(qikegW0EzSb-Bn4nw0Hu|Rjq>8#U5t+Y!d~p+yRe`QJ5KY9~CnV zSbNqeoDT59gelj#G1E@5F=J~uhf|f@xbWw&TPVjDwQKRC+*Ejvf&EaE?_x8VFnhOM zV*PglVcqlx^OObOaNHS$-;cqT7q0w`xGHkfPsPDUUeK}CqtD87Xwt-Yl&CJEH z-8dQUp=#h2y7BV-eUj4M&u6}UOa__rMW0n{xbGtb4&*v9Ewh6Zc>55)TlNmSxpg@8 zZLy}G?58!?4q}egUK(x&Xr0EM4;9Z=;$ryb?jd?0(w5@|vu3wDWbXQSp zbPk(*u!PClZ-a?g3CSd#ByX)*_`N!gZmIZ?!nGWc)Za=bH!BONvVrhy*+L)0yD4P; z7?Eq{G{_m6(8GLJ+TmVBwV7+_*So#k;r=4_LD!%C**;B_|GSg57A_%AA4wYcB?8f7 z3t8z%DbeYYS0cM(#~3X<&E1h}#D11RllNCLi}ohSX&xh|#MuHiVgR|qSL~ksOs1S= zj_$%2?1*as8$ZIHwkK%vhvweJfR_rWJpB#5ns>OWP%p$AY$B_cXe7BrlfU5Oc@WbN zi3fS)<@f^U@>Qm3RnyM$HRS zPJYMDX;xwtZwwI=8ErFeKomLytLb>=88q$|r`Mg+G0YCMXIne9{?wuqXPyW+YzSRd zT8gw~`n;b`3?KVvJ`FFQ$;~eR%bMmHlKInC9FQ(0#Rem~qd1H0vp0w5QFk=F`i%)j zLwTL8UunoUG5)E;O#Zdraw_o7W}DuGp~f|W3fLrOXA*I)teyx?|3Gc)6Da4^Eoc`R^Xm_lP}(3Ra(b5q z=kFt_>cuiP=+8;EJT{&w-|rRSaU?hJ-wwz;n2He|D`>{(T$-^g6He*2bWxaXk`cK! zQvDN{gv=L=OK8HJ-Tx>wG@T|JEM&jzhf#TgkUwW}8*hFt2fxS=hGmZ_t-XV}HEj0Ccg!Ix22$T9^84F=@l!V+;R`t*KK!OVpLAq0`Ijj$`CE6; zCHQSmxEgXra<-W4u$mVB>SZ4-_M^tA4bh&5+2X`^oY{x(B8|?WtpC(9wtAK}9Sw~@ zX~i@g403_&dp+*Q%@>^J)&Y2vHVqTz4yXE`)}--jBF5bGB6p2h^zQXqT9nv8_daHm zeOxYguQ?tS?q6tOTPLR-)&djpeRRQP9vxQj#Het2d?H)6RJO0!WsM;GS)B<0=n;Xx~eJ3}iCJNS;2A8-s!2)*i^%Y@F%{DTEO=@}C)(tw z&c6p8>gCuBIdTknkCP`xP@b5Mz;CT13*Gs&RMm?9%IrnXyL)(0zZ%>7-jddy2lSxu z3_S~|hELRKyeiWtCk0$+3{#p6*V!VJ zOjo5PDY9Jp?F%fesF)=UeaVUX#jvfwl*G9+Y=g2JG_4jQPGSXS3oeTn)9xdt?VU&_ zV;H`Q6Q%10>Z3a{^HcsHF#G!iJhEyJ8(R|r5 zJkQ*WvK&hqn=ulfo))6@)mNtV_z(^l1ya6G4Q=#!g8yR1k%^`iUd0UmV@;UN7q{j{VIEpR2wG28tB(YrUJTna9v8hUdtY@mx zE7ga{A69_$m=iQ}_5uoeT25*X8~IsR=acu^GwioX2HxHqLHV6^FkLZ|#w`lKsqK}_ zG({g3jsb8~(7Thkwm@7n_0ZO`l=mx}Ubi zSAyo8Bk`@PiP=fe*>TS~-`n;GJ~V>*w`tMDS<-YtX(Z{^hhoMY#&%kH!R4MI6%|)d zwDv@b{(1uL+n+(!I1BGi?Z*u3JqR0^fQ9caF_jq=sCEcv+Q#>=#jS>9NZ1SPcT#n_ zB45>aA6f_MA=~^Ay`jHhJNOir+$8MVh1zs-W&?G+KEX@&g!9vfy3)9}`ZTZNJ2RCG z<}&Qk*%#?w$hQq6yBo2zEGCP#$jxJ7f4;HU%?H`9{}kxnu65+MV;7CP*ot@er_;OE zF|_?>2Z9TKLo8`B4u>YP9~KE{{yU$Fr6p;5fq>!4t)n$v4_Ts29*nl9G55d2Fs(O% zB-+Jz*}1RSTes=;8eb-02A6g*-G2jNcq)^9Sr$&+|H7!qW(lpo zaE5Yg#rP{_PQ3OPc`7;ilEkch>CyPLNch`;tqC8oH%|=e8}#VO$~n~UZq1VJ{9u2l zh|vIU4Gz5XqS>#^$>N10ehevqiq`?~?*#43w|8{I=PHF*CQ-fLIs6@MfLRw8(aLUd zidQ;;)awOQq;rS%+z27|Hj1wB6X@0gG1{qqlX^0%N%O!Fx;1(UDTXA{?FGZg^|C#G zdqNODYgz!W)$@b>7Essp_Z+sbK7!U6XE4%uHr@!B?iBwewEN8-&=d(~IO(M5`t{AC z3|wdHubpLI>wmH>Uc%qIFK|UuWexO%{&^~WXTKZnZ=e~txSu9$U9kHMP%60{fK{Il z#GUWpz~n0EI**|4mMd@wF=N<2jyqN)&Q#{jqm%{%vaXS)^^Zf?*vu%jr+mN$gZZMi zi;}Ef;JupL4nfUdLilYJ!2kYXPJM|RHh+AJ`G@prLaBgjx2}b|yEOk|n*~*ElEas@ z9E?8eid&yU;V_%$!ZJM2GuMRPuIXmCoR`wO#tF2}n&{>|Px8O|1Ieuuk-GgTi?&_G zavO#Hz+gEZ<^5udylydZ^G-Hx+CV1BonRxBA8{Kp?=hd=@3hGHI;~VxA~F9+G)^{P zXVqQceege$e^f~Cqol~spV7_pQ507%OSZ<}aQdtnexwOy=bY$zrP)1-!&s*RH#YDkclu!t|I7V8e<5TUpO!ig6V=5rO>a03tXRN04>rT(uO_{+HNefS zf63sKDL+iLnthNSOX@;XM&tI0RNP`mzA;^pe%ePzTq`JK#8Mixc@NXo`^-J+6LAB{ z8M~7$uxsxX_TS1zev`YQ?fK$gyu-G|)L4I+wLU$}k}V%#&7K>$=CccJB}8KuSJI?{ z9!QLh=AQ1@g&qOFzPwF|hR0^GIo~F*Hxmvq?_I=7oMhSA>*>tG(jI@>elW>ZJyh{z zFt2^J4@cWonM1%!{Q9t(PUX%+-RVf0ATRI~57&cIc2JB$G5s8S3ddvCQJ8xSjnbI} zp{gEA!aTcGZ_oXAPz?7J;^~y*U~b$h1$^-6LARSFbKcv)TxGYw@SrvJjeG~K&pm=J zOPZ8UUBKg}5O!eDSr&NX4VJpqpvvtJiib7Ohpoos)R@K1#(IoeD?tZ73LIE%5Y;(; zfVePgJ%0x;v_{X9Z+?c}Y-3ZdX$QJm-!Tnk;-rFUJ4y;Ot z2)&f?_bp3l4x-6+!IVE{33kjmf~d8(pze8%$;h8$hEtBP_UduyxPOJV9eYSqg5&Tz z_69QM+ELQ48qC=2OUCzQY3a2r3={mS7uOfC18OX0Y)G;>T4dK=bI_M$RGj}g(?ZS%-vQ4tws2LB3B_QzT7s#tb zBLAocwmumM+Mb5a@GflWN#XDG>hfyS6IkaCLrm!Qqkd^#4{Z|_#)Yt$xsh;r zoz0dmj^@_O?S=AENm{)3E?FuTlG&RQX7{5E@&_1=UO1cNrA4$=GY?iXiqQRAirg+N zK*h%MSh7D8?5M!oTTkVWNpI)tji=KywHeSTn?h<&e_-wg8FnvQ6LOg&Npi$=UhL5+ zKJdma(%SKXemxJPSeu#97_yS9&6&;Ae%?m>6((9m zzFE4|nDY>tLHF2w?+`p#Jc0JF^1?Z_R<_{R8f09}5gom3g#%ls(B(V(v23LlJvKUm znmcLm-^OU~^)2{&HG+nY_(b3ALUFa|6YFc(N_`_;dEJyYW+6L+^nwMgt}MgQxmn!b zYk%^Sx-%agg(e1`T8buzLV3z@!j8ui46ZxAvezuUZp+2XbI$}5Q41}wm=Ks_qddxDG| zJqWipB_pk2H0Qw>jAK1a|C$s@jWMQyI(Nyf)s2>gFND(C+pO|S54;Y`lD+u^3Uuwf7l^ALlOY?%p(6~=p6ksLB*Q+*AT>E-jH10N0Mle0h zOCqbOiWE^UMv6PHVwta;t(>K}t@g$Q{+;(NvWR|7K1-KVh59t%EDXd9F$d(uUcp!8 zz0e%vieFPBm`tiHH_T9xZLx3?=H_`;di*jg`|yYz%Hlb{Ws#!1M@jfyw~F0*w~=Lz z`@`CHh_U(a53`tRz~bgM94ZOLvCskh?k)Y)_hJVnH5t&`21EK<(+ba_DcBMbfthkP z=-OFCXSM~>i{haai7>L-tVB&8>!8>nOBMdLP&yEctg9jHR8%z!&JSjN>ng}VCzHO^ zY0@`kL-NQHJ}cKR2t46OgV=lg8YXZy)hnsgT0}ciPr|2AhE(MS(YIJh$_%Tv9v#bOIu^1%E{B~+DrLDpM9616TzIJAt6Qh}N=}C<0 zQ>8_kCj>pc3{I{dg3Gx_*{Hnjc&|x zO%2QNxZT2t)}D*e4e8MFJB)yHLn%o72W!at!FD~LLwhH= z3A*bH_HuRPFDbKxvJafQWyc!iOL zFHmml3Gc(w^j7~M9QctmY=AWj@ED7tS!dAM+lpz$BLT|{$|w$tmAgQ)$f z5}r;R$+{aZGHV?_>|E}MPnX(R%i;|9e$C<5iuW_lHIAQBHIUD*$)sm<44m~n(|AIihc&d$1stt5KZQGH4LGMj@wwPX@y+N zsr0i=pH?1dz_SnO@D^8~v|?BKJt2)ehD6f5)^n&<+f7pIJ!$;w8D!J-7))D=+x?+K(efDztB7fzSgR%IgW;ATsw7c&|2VDmM}5FOI!Ib~9JtR^%>N z&n?74&%u~vZ$_�?GTy7?{V_biz2($08P~e;P+7y=E4Z zZn4KFVzI8c8t%@)G-jb29lU3Yfq$h0ZG8rMUiaYJ&gYacx0L)BSF!B_odiTpnOa-z z;b<=Sk%~faJ997Fw%$;9W?d*hK9@|6hmetuCfymfkX%-tqSWmXwEwv>?o2rZ|4k++ z-+P-DYAW%YJnzzVi~SV0NuNJCI)VHroS>T?5fpjan5=5k=>3QQD%SO*Gjcz3&SDYv!8E4zme@<}GSzG0Ki5;_3<+nmB(EUzuylH_sO1H)iBw z{ZW6s-nR?ic8AiNcTRM1z7pN#??BD-4T77K5oQ#QCkw-IV{|gTD!9c)r`JL6?M$kk zwu8piY49&|h3?=y;eH^gjD6+VcyeMNJZG3t{SV#HyjFifwYFz zk+Nhg`Tq{3uM-OpCFuh5)@aCli^buF`E1Ung%}uVK}}nvgkX*YI$KyrdfQd$g+7n| zb?fQkhd?a3GLQ-aY7n$5ijG{}MnyZ_N%?Uu%*I;Ly0}Sn+wctSc@oYi9JJ#btw&?H zrZe+*vZ2*|)$A`f1KS37!mIy0d+J?+k@0(Ii=_cYebb=Ajixl?;VmjjNT+rwWvcfX z#ID&V(#$RY=xMwnX-?cvy80;;CMiwjr^CqaXD63i#KH3p5Sk;-ubZ%v->^D{FKqpX&w^JbKzk&N%AHwxa5O}-DxlsN&OdG{ z;5UA-;p-ON!?hru#J(BQlmVUe>C;)VStCzNmP*l(af8Ujdkj_lmrdQagZXvx6X|9D zG)OO4PDy&MG~S?nINAQi?9AVisp8QIciP@Nus#_L) z60%k$wjQI0fw`oYK8p%Z8lh0{1nynG!*rVra7%QZMIL&|Y`aQnz>f+1?lJMa+}1~E zZ`7r7UWy*lR(!Q{MX6>MM_+3)mT{);n1@WNB%FL?^C(K;J-*xPQ^)ia3OTQY z#dYgw|7$IRkRz#)Eynu|-bexB>u}6$1!4uicFqwTX@~7Oc&i3kk&hAipp(aeEKiuYSGRdE4BHXqh$EN$;$l7lYYo~d4WBi zZjJ$yPO_Nr7&a#-lC2q_3+wlJY`UPMmzg|)g-uE4C_|HJC>~|Q-WRcfi>%q#vgd45 zhcat5bK_`EKf8CMg=snEu+kW7{H;=jf|>_Dj`C*G)w@`a&>L56^%-9`uE*2mA&_!9 zh9SpxwEpNwpXmEr*5{fGJBLH3A5%x5Suji}6Hk z9(N~>IQf-Nxkauym}VJ=NeYN*qtRz4ySipuhQb6 zOvwm)M! z`%mpNyZLCipt*FzBDbHYeGv->=a~o_Fbl_i15axeFxS`{ZJ9@)pcn($flFa{;4u7t zu7_{U1iZGkC6A{m)P8wCxu2SevR&U;s7Wq6^mPb+_shcKUl;S5FqhP2^yzG^z)9?@ zBr%TlY_W(Op&`%Mp6TaO~94t~o%c)Q}a>qP8wJ<7hOe_D<4`C@qnK=B_w4eeFbizhQ#DpGwHVCAQ6eJ=4k)JTB4A?BgM47P~Ey1&vF@N&!PR z^smQ4`&l$RF9N=H^U&YwfOFd?;edY`<~nZYU*!1l%Xhi*?Xz~`_o+}Umr21nn*kVn zqJ@3As|+k(h{^;`&~Mmdcink5`9B+$nq|%IjJCv#7hOzx$sFj+$VQUs0JM#ti7lrj zFunLUs~gnFrahd4PTdrCwn6a6shUvKBymEsEx8vQg@JiIdb@%VI`sfN?@L4Rj6T{Z z37LXlEdQQ3Y28=Aiu;+&Y06V()j9~vmfH$=s~bM$B~b9Vi)8fm3YEnRy1NNxs4w$G z?OJIHcaulxbwx}tC}R%RQS9?zCpIi`Gn=1i&RX5Uehu6JqljMCXE7N`^Xuu7jw5Y; zr9&luUNg`2kJ;z0`An)e9IhYsGUMF`nTlm6DGL=s?x(G2u4XtSghY}7OKtF7-Upk$ zo5H%Nor#%dAnQmEiU+je^bSMrR@QcIM$`e$zU?M^yfaC_8BZ}CwK%@{#~S{FY9PN> zt(aL%In7?>{9rAswlNKl6jo^6$=%&PP3WxYV3!)z@hv$Ee|%ijlljbz`r(2_pI$+Jm^>c-mBK0SZ_M4goKhZDlHagf^nJPt zjj-NA6DF$Cuj}e;gZKoLew_i;_q|NQ=rXcSufpf^laZPEhN-%}Vpp}EvAYq*NM17# zQQH!5TyH-PY0qZ8F6OLhqC6_=o7wn-7RU`wXTig7aek|oqpdbV@QeRsLqv|qmO6ni zR_++%E5~%L?J<}y$NJkoY~L6;R9?@=M&a|GxbXqry!V39znNrOxSbZ(-p9xvCZu2c z81oO$M&?He*o~89@}&{1F=8H#Uvr%v1O!v^Ku;34a3OnXbNU{#g;pyG9Lux{bV{Uy zz41brUJ-r%T4TFeYJw7y-;BaJ+JKf{dIFBJ z3FAMgV*cMq=Bw;a<6OQl{WxIWZ(|mxwvtQm-^l)2i&3)U5;~Lg4hugBTFHQE?CH0i z?4($^;OT6F+kZRpyde&D`DxI$YG>~+3BR?s!mLR##Hllm>{-ki*1R_hg9e3R_HuzI zGv7_)ELL+}i(N#6qY_z=Mjn#9#UW=q5VM~K(zW&vkQVr@0O>u9AJ)c1@s1eV`;_KH zp2IecDm3nQVX|rZ?Ba1v*7P!5&`gO@g;@*&kIe?RwuD`^3Sci}<8Z}S1|6%T*s1S! zbTLbTmep6`m3aY^em;+8tvo_)@sfxUIK0ybyl^Z=z+bw?Q1QBvG*4QN2B-XHFRI)zwwfxme9&Gfr~28^e@9sHEeBefsZK6mRbl_m ziTOC|(ejV`$k5G+De6`-<%kc={K{5Ntn8d9KTVU9x-ktJ#z*PZymFebIgROAPeRB$ zZxjh0uoXXr+^8$L5N-Pm&3_e4qGE=CSJ_~GY#w`3b)L&Kt7S#KI~%N+lVzdwKsKAS8&*@PuL|A0WKJ4!E?96RgFkeom49A) zj4zharjhII=)sO{G&IKvpN{@OcV8$XSDIpuply7{ePiu~UpQ^yPHr9WRTMgLF@AkH zh5OGJLVA`umaH4dd9T`HF@Cv^%{`LS3re;!`RWkb%|8U=S}LDxl|`fg@XN_rBtC>)^jkr{OGeiBv& z?xsg=tLWCy{dD?hG`weOA}GuSga0K`z5IDWS3jB*r|+jslFNd*R`!MJt3oc*zJWT5-UH{A&aZJ#-i5=`BUmd|mpx zS@4%QsNGGBjVMV%6u8mqwe9lOoS ztUI_?jcjhVp5U=IUCG7A{1H`LDi7F?uU<#Bvy+{&`((zbS_`Wu75PaO80Bn zaq|RZmza~gaUaWZh-O_j_Nbb54bFq5@$KvuymflXa>pEIyNw<*f6LeGzIq#1<}@A| zieYS5>t@b(r#K&;vyKMO(WmqGheOrk0VLOz;``A{D6pOe%l+vH>=})zPkboK@e2)W z{XunKztXxlS>)4on7;p*K&1)-(&y~TtapW%M^XB$e|R6T3mqN_Y~leG%)JaKZk|eV4*xZAA_~kKE`SZaleB4-V`jE1Zrdai&E4-XMP4`knzQBz> z>YyXndTI1JFH)RlPtET~)6kzgFyO@oym1$N6(==lDc%`D2OvzPUJ?@5@c>u;dby+C)KHXN&4) zHgOBs0FoO%9es_{AS1l~B=H&8cS?!&!-Bi-&@i+mrS`|te~-BEb{cMa6B*p zk?}@=uy0B!M9}r^Q~A+9(|E0AduiX<*|hOuJ*KD{V)o1wtak9k_AT#N|K3)1S79d_ z)fF-HtP{ivP9WlUH%=|zfkipHIq%y`Avsl{y|40vG9SAaGc99k7yKPZO1nvd-aA zxPJC3mJCloyV5Sac0NRPFI8DY*m0Iq-HqUhtJ!=*B@9`;6K~qyV$8ut?4CH7w2!9> zdYe{U&EAcEg)kx-*Ot%7!TF0Kzfg%7z`;q9NvN+n7Wa7l(@3@*XT)`ILkRH*W# zB;F2Q0E31AXdmn3#;o|xwl4h#Tb1odOMl6}HH?Sl>lBfW$dk2P)We7&z<}p}k@jXd z$@Xq!|0SPCtE4T32EAb(&FSbntU)8UDKh#k#n%0M&KlIh5w%j_!e1oPc{v_R*Bnq7 zs7tOL$o9hrPNJBNb13S z5^Dv`yrGV9zwe{L$cSEcjHanu-KoWNBh?jY3YdEUb9cFjPh14|xi=dbY4X%=D~sU8 z3K%zc34)|IP(kek8hE4)17^pdCefC*dTCQejezSoFh2-mXMPiMNMz|ls}ywydeR-uI8yo`_!3Ny(VqC(WFnmD$4(rEQ(_4q znu0A=R}j_)M^^+GM#jnnY_y&@e|!@=z8Rauu?n-bVCw zr5Hs8r6IXQjH-KzQUBWs()F+K;fw?A?Kh?DoB7aeKZi2n;qvVzQ@9a^z;gp=?ATN$ zGky~O=jc5AYX09huDxhzP&5>Y3hBJ>>wXIv5h-LV8D(UIjBJWZJCc-2R#HSlvQzu-I`=e*8qT=#W7uk)`eam>ep)0Xw-G@Iv&mwCf+qvLV3w~gbR z4`(^gV?EEY+W`B=JJ2&?20A~viV~v($QqQ7%Fu8=YV(FY%eHancqarNuEh5hZF&BT zbENz*7l*{%*2MS~&vKh9`-cpb)klPLMqnSVShx`{cVuIE-43)_Q;erx($R*Lrg#C9C&{o^qH-#e_FZjN!$+K8F5 z7g<%u>9}ef*0;ZlJ>4h3v*mWkKclfWt39u~5zUVcBe|R}3;(A&FRhB<@Ef|)(xPUn zF5Zvd(l}hW+YW8tq zoj9V}oSgz4$tZC#<^DKBV_b&8UE>+H%I!I+<8s~{uPw&}43kTP0&(B#7v@H_<+!aU z$$rXwcpeMIQ|D;Z_s@pof+GB$UXNdwRCr~ZS8y45759}!^VH-~JayPb3=179*hA)M zRV$$+-ia-RpD8J+EARf9BW<{@qtN9H?vQC9?-07NiKV*y-m{GR$Bg4wkJ`wIT`lFg z!e8=RRY`u`PMg!#w?scvfu}qi!vl5K@wq>)^3VP`a>v4}@{-sJ5?q%!aHJ&13`%P_LY4ZVDWsE^tUNx$VZw6*?68c8oC4TFi|E`10&4RN%i zJO?Ss1~h$TB2JknAVvKh9T}E^NDVz4KB~b}-}mGk=OVgQ?7=%6wB!bP5A2)@;m%?X zH0mPyMv47y`UZR+*$FAz?U;6H$?i%HJnl>i_tdQ6qNoHW)w%q-zYhKjxk*{I@ldvx z;rcU$TQ9ER`77?Rd4Ms0`DDc7{!Qe4iw5wE?PK}SX7OLii)QUX>-oLjUo^I#fqsH# zbAI;$n&zP`JdTN&eNzXc16#ws%NqW@_!=W~4$C79d3nZh8XeQb$(ttev)vU4_%oPq z#4Y9rCsR>7HjhtV&*U{4?ra>}5v`qv!JvEvsci1Rxf)Bv**=t2{1ag$?#2nj9H_k_ z9zC|^;AxIG+Pdem`coHv>$sesY%W7y^IfdZi^8`_BL$E^TX@%(vJ}0Lm#;ia4nt

^F6J%*iObbLNUid%-x;G$&>oH%AOhTV0f;Q zyTAFss}qYc>UBDuxm-gw-v@%#G!Y*-A2-X7qNL3>)_tAAk?$)x%gciE)GpE7LgaZdje}4w$T2|AJA7LiTC_H!pE#mVbPifBCAb!-2ar6w$n~pZIO*nx?$2wkwJ2t zJ4{{~k}l7^G)z8lRtZT>_h6Uj%_g-6*suB^2fJIayGIy1x3iVcPIQ+2R620|eh*&W z=JArTp^akWw=CW} zMNMwZ7%o@5Tgi{Im3UzLb7T>euizou75lnWNy(MQq#xu+mfjUe?KY4P#T~(gA5Y0) zO-FXkRpoXj<2e7&Kyk=^XHUmA^1H5@(2{-8yeyMK_i1tazp)5z;Y=5|TFbhJx3J2M z2OPVl6Hl2^#d>e|@~M3e>~S<74Og`($YLyZj@=5M2y5EQrMv)`c}8n#c_;QaE>Z4DSE+z@zL3bliC=Y-bzb)w7{Um|0HyS09r)FWapcvU@4=_0D0g zrzQ%HC(yOJ;bbNHjRU9Iz-zr9cGa|m`uoS|vh5O8tT9mR-?Itt_Vj`7l0Eb|WibLz zG@;MnK)wcjJP^=!S`;LwM6_~Z>fO=+yOp!hf2Cj-E%oK*Qx+I9%80#Q zg>cY>m0X_ZNQ>Y5k)F>=D%9UD^r55NVyF+R6|Lf5z0&xDaWHo?T*WuKjOI6Ou435J z4_H508^--x;P2gRYR(TKmsblBsJ4h-l#8y#5h4TmEnRVV!)-;_b!8MeRAc;z?Pv;l zNjmnayrb6$R<~LUr#yY+jd+7dx0}c~{|0`^GQatt!DiKG$=2B(9qR3H{6Z^QdG@p7 z>$4o`lEw~fmU^PwXH%??5jlHLXPj88iH@#M;goV6|Gf?u>@5ZMj@gJK$ByFl!hx7D zsFb!f7%Q@`4`!bOx}4p~634EM<%IMa%(y#^HvX9^{PqK&;a)|yQ?$spdoevo)JLbh z*5tbT7*1uQIFY^pwybf=WVtf(iJ zR}7;qrXy)m<~Qli?bfu+vX19{nk;h?+DGlRTWgD(I=Pfm= z4Z?CW4W6>8H=k5ZW}EMU9CSJtrhQE~;@$|}D&Di-h8pm+z)U*eJ&9Vxslp{VhqU)6 zayRSqxPN<}VobZXyr6a(Z|^XVogAA9;~wLHIOo467fb%{!zicpjI?CRBRc2Zl7COW z$j2A&$Jh>|`JUxS{_~<6Z`~W9@@HRF)gVX#?MWm*|LUi9yv;$u7%WAXs#Rg_U3@W zY1CG|4z&+f@xm2Ol#ae+dc}{N&YMy@UllwusiD@h?D&>PKHq+Qn!V;8lma+}{04`? z*ii#INmqcdP$WCVvae9J(>{3vYfL~dZmDSQrE0{wK{=3(=encwhKm@503WjOk-WCoFPbn8=pQ9IC$FKpMr!^(YRv8<4nDu?ohW9qnbq#OIc%4frFp6vVCm6zVp zLzm7!dYZ>PM#_<1_-FA^dKZ{Zo#&ke-@l2iW1eAi(qXX3J$P=qM+F+L(CAc+4Wi)r zVdoyKquG)TPJbf##v7#Djp5O!H^L;{i`%@}$x1#h{C9H)M9G!XxYrwSFE5uS#5GWR z^RZB$rOtEff;l4e4v!T7m!DR1No(33I5#V?VVezzR9Ph{Tk6s4pg812SMXXpU%vUw z4F;X{p>R~D)O&lWUuG7(?kiC7W4Fjn{lGV+%iO)9k|(4lQsP4^TyE6mT_>heqk#^T z2VJFv!NuGoxw|~~hogMS8E9XThEs!Jhjy2EUO*8qY^llJJ~Xgtg@N2EPe;D!VkECSU?vZm zu!9?3>GJl@*RZ0E3v{EEWgqh?@`&0XIXFEK?=JKfe(a%aUTDA`we9%VgZ|K|sKai} zhX`%)p0<+}&ia>8|C?n}{C~Z%$^Sc4+*_mSrafx6Tv4c(HY&UpT4LB%GsM>Q!?mla z&`_zC9`1VP+&5t}N zVZ2@>Qp&$k;d{b~ihc;q8&9)-^b~L84LJ936SX&Y8W6it-@( zprQi1&&2Dio?yPB923NSBKmR#Hl55zdF^a68`hFf))AdAA3@bpj$oN-v48GNcr|aQ zYj1v1nb@OFmn_5E7&TOU`7S97(||`mTO|L?kXCK)$UW2|IDA(oZ(nRf^F;Oh$}u%G zcaIr7j}PIp>Ro(p>k@_Pk=Ar4dO8$a_ft~oW3g@?a%89_j+DgG%IP;T@xWZ^V&Glp zqVZXZ%W4U<@YYCaacQBVs?Zqr%LLP@q7$E=yqD9L^uwh+&eHNmL+)Yr1AqG{xZCju z(oWrLiihXlDK^I+=VgDi*&^W^ty$3nXp}} z#+E5(FlJ(3zWl+9&8~Q3_tMLxpC!|z|0W}8avR=gKZVZOxhk$-p2y?zr}CQ@!}z3* z7Yc>{+3~R*53dm%-rgIX%UzqD|La~UtyyeO?}OnB~por0k-*+;hMp4emr6rw!}p%y7ihQ zMbElN9VXnt_qOkNom&I9y0ej&cKd;L5A!ACusc%d8dsj6wH{9`_DQ;-o;;2&;+3Tb z0wc26rprjaYI_F@XV+qt6p8QiRdMQ5e~u}*4Mkx*b@QAh7n~22uX=Qox7sVoqrUy* z^LtGA(FawL6L}%2T0fC0ghn+Y-V(*D=8>yGqSU>>TI$w5LQ=^HhlRMSu73hb@jF8+ z(tjvadc-=fN!kU|#?csa#tt)<1fZ<<9_jTW1-%F#fXU^pP$0Qb%+x9l3c10Py?S%Y z9}lHtfzD{Pa|7BgONFc&h!`=`*__-JL*woE@(E{F{u0B15qI(XeIV_q+{y8?yK>AC zWenKyK}wx6gzOh|m24ir=B1(KJm0q!59@uLUsfFE*(JokMs4O7S8H&5b`jZySn`SJ zjR=tUkcaINQnb_IebekkK4k!pFPaKZ>P_>WXT$oU74Lr5pZE2D56!ZlWaii;Rg}7+ zTZxm@v&%6$X`ss0!aJP$IfH&ZkKnO>4RAjFjy8UJL{FlB3ctlSsd#dVR2#E`8plk+ zynsryUV4vQw9iU|59_er^<28=`$TdsP~nm-V$Phm83_w5IV#+S4RKocN%-%S#p@t9jmQ|(|eW8blGPzjouT_^!y)jHF(ILv_%!LJD_Y zC?zWk*6Aa`ZGMwYV}<6}*Uv*PAODBnrSz7!g&E6ZmfCWlaT>2Y^N`;Ll=C{JA>=oF z3$08DrSI)evWr7H$35{x{gXE2YrY644;Nth*HMzi?U8if(~`PY4C4iFrt|%V8u;9u zj@G)pq^Oi5iVp_HTv-h?KT($Ewt7fL!7(^kq6ejE9a!Uz1)b2JiR9CXVmN@WxJ*`+^X?2=6<;%xF*@O-NT2M9@3ZfM>Ig~{1e`JEKDx<>LG(g zZb>jYcPiCji_uul>h8pCQUdWncR!nr(&DKT1badFdcuDi(X9;`(#0X_tg4trD@|M# zijQV;nR^e}JN*n>s2!!$kXICJV@YS-Vi2lfKns1eB@eYLu$}f0KO-Fw;#14tTA0Z< zvre#@LkPOG>@W5At(JUJ_VCw$t6beum+OXwaaQI&k@a0fS(T^xfs-#6MwZd%>0_j? zjj?DNWXc9s@pv$0Hh*#fx7k@tsab#ElB~r~E}a2p&gIn+-MEVDXru7kPTzY9zux?k ziXZI8m%aPhm-9yE^f9P0IK~4c-q+FSXZv&mswB{&2_IJZk<1y$o_c>W!>5j+$ zx+8OaytGs~7Dcx^^Q@`o=vY{i*bg)ilHQK)IO)RjunWHIFQ(z|jkr`LgkAo#hC$jv z+%7nUS0XRAzndD*v2MY+!55_iRcW+KaNqyiS_9R^-h3`;JomJ@O;r~sAoQFPyVist zKI}WrpS_CxfBT@D@Q%OtI*aPt8K|;NC-YH7G_P(vY7ST7^}%kKaQzG|uBB>nyKOT5WKnB z6%h%WIk;GzRTg}t5!dqI=c^EY!!mL;%jc2V#2@1hVv1fcbu&DGzTT~QZP&MWZJLh! zf;`$Lx;7NTgPz~543;xHaStUIo^WkB@;Wy2?yM?y9G;1(W3p*pQw-XK)#Br{Wb#{` zD(>N@Xv$Gb`0W}(tM+)4S6&O*>sS%@x~qrhS|f0)CI}Oz$HAccEqd=~i~m+d;_sAD zin@@14#S_}R*Q>#GIbA<#d#M}bq&95gShIzDVSxih25az?Z84AQd}Ro3cepD# zw`$G17Bt}WfLv-v%P?qE4E7BAgh6L|i|?l`=B?g>;zA=B4rtGu(3MjLZs#dtA5*E# zfo{v!Fdgf}_W$N^f9E}lY2Q3#7kyX23Vx9EbZIC!hS5v@;-? zO8%|Kb#I~L{?6gg54|zhb}4xqr{ZYFF6`Gdp`fGxXiI839u2pbBmX|(sHvaOeRdjq zn7oJS?Y+{tUsf1kVaJkn0iJ5ON*d!ed2aDhmOGl`=$27FSsV%wT>ooeNzD2rot~HiE)`O1P4aC*A;-0UMz`ld%5$Tyux82@J zEh=rH6?qO_V+vt4@;e46-iK37DEc(C)DBGMv8KbgO_n|<4DWL zQ4KnY5BR34x_q@pPmWrt4X-;r@M3cTz6BqlmdjUA*0-JFJ<%5D2R^04O(o>hNv6@8 z1EeM4?-buwzMyB*Drs-C@cv!$q4p`K$Ypk;wC39(+N$}4qBCPCL=4fl`Fx|UJLAcu zcR3ZFy-2+SepAjlPx8H`0{ikZJb4*_{nOhb@zn~%{5?wb9Zu6VhdiQwB3JsBq3BeG zk9*Xyd&6aTY~6!vNy8DkP58b~20>-jJ!#ceYcjk$o8);8xZzj}!{1L4Jf#p3Ep9@) zWhEl_&BLzYhv26>1mA>Sl{+sK?AsHm!xOQobUf0gEW-L-rg(nE2bs$a(KJ>;*LJo< z;+-WJdUzOY{|&*D@yTTPpp4V2#J=Axn2S8r@mpj!Rob7W-m8Yg+>TYF0{H$K7Z9a?sG>Vh*yIT4(9ev%SJ!vSJxZ?|h?+vwd(?be5QH8;Nsq z`E<6s4!rKUA^P_TYI^>Wdfe&5?4Q$q5Yg(DSzQ` z(vFuYT)7u2UmBt1x7Os~X(rz_Yvj(GP36z(o;VYqh-GPOFl|g8jZ@X3|JtsXG&;SJ z^xUlJ-liyO^&*ZYIlrO#eTBxfqXqog3YFg+$hV?I`f+?8P4W)I7132UaY((*nIYsVwzW;g}6)aIyb zT3i+*x(7aYq&IGLir;Orr4O@Pz&3X`ozHwD4L!e@)o*{KphJF&bCJ2?yIF>Vy|$r< z{P87m0p@=`hPUxM(fWQmYChy4=FU)#?QbBjD*D1>ZLRsvEiJa)ydQt!lAtI}MoxV@ zWau5{CpF*MaNaF`E_mv$*Mhi<^C8SD)exK&7Y;4b<|``d_%pXH_W7A&)Y79oqA7>n zBKmTtg9>&ZI-Q>`QsNVP-@vnZ12+5$rQPO^-0*ZW2I@{B(-WE${_z&AtuDu>hHoOD zu@0WM1L)|xrJPsn%^Q8(*}msqJkU(W#=67wc(pa_1qH$5hm#cA=RB?GR4dk#DoZvA z>}V!&`J6x$JPkr7MWJihF&dQ9i}&5!j%AvqFgdM5&z`mx{D&&~^y)3So$M(xe70C7 zejkUXF8uITKK8@|K|^B^*W`y(CujCvF_`x!IdNj78l8P(jqj^sVWzl1o7P{Dy6Cvk zri^%=aW$QX4^v^QgHy4Lg-q+A;jl zIR-y%F>ALk#UmZzKOVFY|7D#(T+JES^xIDhmk*({O*w-VN5F%hbUf%4wSOC zM`*(x6<8kgqZ5whit!(eCF2hXn0MNNDyQvG^o(;!r!8n z4p2@VDE3}mJU+Rf@>iR(U%3I7YCpw~A;J80xadK-XCj&I&PJ7PA${5vNOw=_u~(Hp zyB)ISPil`KG!0T)|5BPF`ZzpW6-fqdvZQ@=tzov_lT1H9f8BDHh)PGJ+C zb(8sytz3OHM7^hK_n$Iq-}dn=SKZ^S(|zK;#ae-MWGgFY0moWRM_b9};+u zFzR<-iHvVQRGj{uf%RAR)Ax^0rIBlb;jVQ`Vfefq?Qd%%%X=rveQA_zRh>$$v??fe zD(I*4C}*<{ofYk)(-bieA4_MqTd5Ti6@uI^~zg>7A#nktCaY= zMViP%b%u>av!v(O1uwVH;MYBR3*KFCd~DF-NfiP7e$Oi?-%gWG-x6%yvI1(~YZhAC zmBYq(1Glp1!UyVyB3kfTo=j=Q<(adnXm~e#>NbR*2KaKvUtv(btHm`Z8Ie&DFs<6e zC+jD3mcC$|ZC9}GA}s`rokq)to`dP5?+A$BAPq~|z$=s8<#M;Z)K_G|*M%?Ot!Y|# z-P9JNjT-51=o?fY6s)KaRjyj0$zeW9*y*=jIy-SDhW$=P2k9EM8Q6%`ngdzcw+kY} z^I&17#WOpup}1cfoN+Q8-RIk5t;+!_D0m_?)dxuKyj{^#w7xL}5Tj=QUT(U{~ zM*WX%l5QU^rDqR1!R?PRrP*rn>x;1*eK7*_YB!)uj2eHupMep#mvZUWHoWiHVCj>7 zBejD)9tGv0&)fhG$eqm^!B+gyJq&vePk`Y9W9~mKjDLUphs!@T`Mawd*9pzCd3g-} zTF#@p3)dpoI}00f-_wt~-N~!+rzE{lpmhFVZZZD_1|M=~wXd;w`QSf}3qQgaKaYdS z?02;FuQuOp@e0${3pCh<@jNi)72IxWi%x>M$mn;3QYt1R&GsoRP4vU3vWa|SXCfB= zdPBK0TgcJZXK=IAPef1c&(`gXIJ`O%2cGH+o%0c@Tt@MmB}}bH-;`c>1>sZEbne;l zIieN2an37@ulBX(dj-xEnRuE$v?xbE$7`bRWid`B&qL34L8NQ@M%rQ(g-^-hJXhI( zTZJyie->{QQd}MN@O6R7;b*uea=@#}n9XPzZxb`th;i#t7Ez1#VeYceM?L;}wE)kX z<9N}CEquts3X^_k?Re&o#bY0D} zoA&V7fkil^u%N{sRO#@F&6qSPTW&fi`o61Mq4Zo5w-v0{``go`#;*~$6Lt(iD+gi0 z=0e1*g?xOsudLqvJdK}SM}c4Vpyk%Bh*QRw+pN330pNJQJ8Xb5p?9F#_ zPV?9?{iK-N#nRfHg8Ost5oWucLPxqQwVM--4?o@@dgT`US7av#bzH(O8(Yx#9Twyf zs>(0tU*_M-L@%aKJ^E|dWA41MxW4lko+kNmFU?-C{4fT!CvWgxyK0<>Si`A{jv>5Q ziz^Rx;uk5(yduE?O}BJ;;dhZu9kP@TbpMaGxYto()KkPQR^mBN+M=?0G4ERah4pT} z;nvP2IA`39IXsm8ZiwELV_VU8mL@j&O~O-?b$A|BgKm|hFyyflclJNWSK|Bd@ZX)fSf2dzn53{fnl|X${iAkFJ>7 z*BYi1w9s`(6pfFWMLX0>C~d_!$?m&EWkGHjc)Sg+MD(OLDWeckt%EhsAJgQZ#e8SC zmdF$K$B=+X9_;DKp+_Zd6myX9akVsmod^8Yy79rDz_k1*3{41zeV#vBZykis^G0B1 zW+yD0@CV7spQP#%CEBSxfTn#vg%2J(@#@xneA&4d4uf+rqA%F$=}$VlGlp&oMo(3l zI=hvs@Ymg&G4hi+EBZU~E{6@crQI8&-!4JVfXO&G_$xwcn`x)kP$UhFrPnGYn3nJe zZ?;*<`>aFd=8F+>59uVScDqW)u9o8YV|5s*=aIWkf58EmL5Y86Q9!!rJ>9=ec!(W1 z$RdCnqUxykuIp4V<~l7BU1Jxv31*kqHuQ8^2}9rE@VY$>>E$hi&)6EpiNmSYvJ%8U zbA(yrQF`e76Ei-_9RK(zvbN3So#_)fptTxL9lH~cHD^P~cLJuw?WS`d$J38!;ZxOK z$2XS7<5FZe=H~lgx&8rqw*3=Tyy!{A&C{vxWN&B+=ItGue-zoRGe)-?z$cHD;HGM4 zC~IHFr|?{?OMZz5;jPJ2%=ykLj}ke!)~L)f#a{hA=p`A^pNNlA&+_L|tp9yN+E^Uv znnyzteYk8}H-4p9${x$!amC(Rqz*8k3v+t#CFeT2lkf|>)rxuj*n8r=X34p8ZSjw7 zuxIXL>>Amb_kNs;VIPBpZ|VZQcs`we-c2T_(OPI%@`chab)_|r^@QL3FfG_6 z{&Cirx7=tCbI&B{-52o|{@N2y8;{~>{a(5lWJa~!TJmF$ee5PrXUe^cfcc_NV@Wt} zxvs{7Z7=AFqv(qrqROs)rl26E6MyZ$8dX{9*m?FJ4Z4&oWsEOZjFPG-J>&(S_BzEC z2`%JXJvQQ!-2!SgT@8lIuOs7J4Gp*dOm)@%l;FJslGjG`DN>N*lz1|B68&L4{z;>( zACPzBKB_YM&Yv-o*P1J_W;2mVpcC3DilBdX2P>?n;cKid4Xp1b+lt?|^4W9A>*)&` zy0|UW4n9DupF#M)u7dQG^*9>bns=ln;p(hfOcWlo!PUAnG-xzkY~Kp=OAf*#=rE%1 zofTa7mGZ!sr}*eB9dhsbk(?hEWB9Po;@%&J)3PW23o?eU$7LEL@+mW8$8zW3Z4~g! zmiG4Pg4p`;u)k48ecS9u`+d))h|Rr7Wo8VfSv273lvn6k_?4cATyoZOf2mmUxQMlc z4^?HL8+EM_Jcp2}RI7cPX5PKY>(hwiSH!_8)g@P6CUf2syxA&&;opmv+qdr<{mDBpK?RbH~Bz||{2~ut-QCU}as_&+U zNJD4Dd1Xrl%M5A$J{z`Hv_n;svSjdRGCI1igon~vv_G~1gX7O4cG@LG-4uOeyOsFh z_GG$tCXiEpE6X!(7w}uF=P*z;;Nn&7+4SEH)af;ob(#yk&`^_qu5**G?Rm+8Dht`g zzb`Kw8H=pe*RZLkUeYTmkdn71!g5vz+?kWD`1gJT+YBDV-%|_l;=2V(>bH_hh!klkZlMj%FL74s`cQf_Q=2;l`-%>TWaLNPz=6Z|NUF`CuLF)_ z-trL?K6#=v$Y25=Kkyh9U6Zlcs*z^B?xfJ(Z0T%p;|Nsm1EZAH=$Na&=z*BXPP<;< z_JTcv6?2-36UwEq4@)@4e=BcxklE1A8jD1SaeVbP4EG*R)<$}YmfJQ-rY@V=XKoZ9 z>8i@%U$UrUha_p~rpFX=GXnL;BDwbGEAE!!Ctbf*tr#EDk4!Ebv;W2vT>bf3a;zIJ z&9J=T>_2Xxw4!vfw7lgoQW5XJzK>fVE=xt|FvYm_)|)Z|z9}-g4?8{?G}-&qYjl$OL*PFpZwf{aX#U^=aU&H2zJ`&I~!{b!$szUXRhfIhC>sjTE^?gL(D+Y<|zK zg45ELJ^LyPZ(Rc%589K8SesuwvK0k~Ybdk4E#Iz}*uC@;8y$WQwYdYS^DyBxK_;4Z zh_0>J6m(y#)rALS)<0dkQ?^{H zoqZfix00m3axj%u{lSGp&)DEN%S*bHAymf|eGYC%uQmCINa?{xwXS26vNvnJ4(49g z*YMfAJ$#}zNWMQ;C=8u);WH~!5j@h++1^)Aejc$wp1UqpZV0%;jnn*ici4T3F0n@c zH_eLH?nk6XX%G(@c#J*wx^j!`M0!O)V2cDSI%$ND&8noCq=gX^1p6y#4BRWnP{iO( zm}yfkrF?wqym;h3SQMATH{Vrc7!E>Z^dw|27qg6hHs~>HBW~_|i{cF8-tLQp#wb|X ziQ4@5{&Eg}SIRG+iY#~9EKyY|dOt^P-~z6}#19P=TCj!6uO(3V0HNKPI8(ogv$4Z{ zK2qP0#|YId+%>4d@)<=^SMz*2_<1$T15Qd!oeW9wV+)SAe~(20$(ZuA50-^ICk-FL zVV_ok<;go>TRf7VU5jSduimh4D1&vLH}vOxB(3Suh&WP&NBiS3?)C#%*WW^UXh&!T zlv9DN5q%MQ`r`p(M1SRAE>?*_w=S*N+2k<4mqW<>@Gm;@-VAf@?x${zTcjMn>(Y#c zr%>@B1C3HY-den$z74kF<=XMQVS5{@86E?-t__^|w-tH_Z_{pCz$<==bzP~8V5{#m zWxEY7PF3OBfH1bGcjW-T1A;Ln_+HPGuy2j%m29rVxGZCymyyPuY?b8elZ(jK<^^^6 zQGg4{L#g3xyyEdO1H^yups(Wpy7ky-iv5|uwoCQ7uYWxC+ctsjM&E+Y-b`NlwH^xs zP36|t6Xd_oUA1j1B!sciQG!?_xk)dj;Br?%fGr<$+iC(qK|PKywbO0 zUxO6L*}_k@$cAV9?8f(o>d9YBU&tf=oTdjE3cUXQ9Q`jGmwwyDqhHE9@Slm$c1xfo z;+**t+lm%Vec@yr}<9_7R*+O&o5 z`!-xPwHO9#%|v!Jl3SX0|Q53?&!ynqe|I+ z?h-znwhSiwH)77Bk+kl{80nv*IpzHuLw^SRkYXx^E9RzeQfQf^iP_{F^qg>6a%weJ ziZ0ITIeC?SkKZJ6ws9-4Z&Vu1Z8I4=`Wz8#>}EwR&BH2tcl=5^gTZ3$Zyb}t71uLZ zJAV;UcXh)#l^OKV&K?yy>#?U%9ntB2Fxc=LD@VEUs-j5V)zXY@+DTZR;v%|$v$?z5 zP^J!x>CPHWyoq(d(y~2l?7fbS=e)*a;|hLTcZI(%p2Lyf)`)I#6JBXvP5amWgN=(n zV%bRA5jl#S#!Vy1N*h&vGuf<(_=t50jXNghrnW{>@r3sDV3aPF?!H51;Umf4FP2}Q z{fJ92NA$3+X!3puYvG9;Y8E3pNCcPFVH4K1)`R=1&8WA~#q0oGY3TEVivRwdp`O=m z5MvyU@1HxvsOk)CFGb>NeFP3)szpJkIY^Tx!)A6RoPBO$Rf!eiAKjPsrA0`A{&Q(o z<_N)aEv2!AyXn${21KL|g3=^2k_D3>Z(1xp_wXPO-MeJ)X%33r^2smiEc8Fbiaq`q zUNn!TWnVR=Ukk2F$B*dJ$Zox8zuH!`zvILQ-r95T9nLU5Scsc{R5>K#8gy-Sd3nMB z^xByxS>HEhbM*+EdhZ~a4^%>jWxk{{{39#J}wCF-maV!e0=Z8+s~im)2sKvHGTq1qjq!O z4y`!OMx9T++l$F{Ey%#hMv-1+t@tHoZL{7bNj+RUNh5M^2`<87Jn_9ktAFJoyZD*Z z?@?RlSwA}Q%UI#5KaxYIyo1Th@fp>&xl2E-LisAiG+Kp^yp+uYCBOG z1>=X(J&n(jr~gd(Rlfka+0jDQ8&e7McMp*~^f?Op#Bk4NquF}RIUG9QfDs9%T;es9 zOX~Kq-rbd~pK}tuKi-9%Mt^>~{T}{JjmFL{j%d#HmV&3*LNCZ1qMnYfhZE}^?giz` z{yq22b!Jtq->|xN1EH!5@V)Q>8P*M>Wpi&RTAj_OJmLEqe(E0d?S1+3)2(d!>nIw$ z4Y}ds9xQ%8h&?Xbu*a02l=^lb*Jr2l?N@!->3s{~(Qv@@3Cc)$(?A=<`z5>2VhX$+ z!9yna^Ww(``DW!z^v^yD)e2LdG<-59_8pDC7GI<(f6AneCqmKOKrZeZu_baUAxk8g!QL4pv&adHG zN0naoipO%5$p?Aw%NE>qk~-`qF(bLM0xdd-bMtW<#18b}$=4QRw%SD8^xTNfjp=kU z>?OsHJb~5fV-P?|l$Ul39bc=^-sML{zGoU63H|qW$`CAw+JZFKC|LWrBl(vm0tT+8 zxR#?3xUDB6eJHB`e#NA%n^^0>7H+*$!8gxUz;jh9J|8ed_=@YIm**1dhpfl;a={Bo zItarj{g_^g@1mQ)&QBKIR4!Mf#kG!fxzUtHUR=R`1$WTga*<$0=HjvQF&ypEQeD%f8C9M}F>z|CRH*)i1^*-10;HPjOU>u<4@-gIuLnT@RF z@z}ZG4^>!-Tzmh&Txew>KRw5C@Jd&xFTYMcIRh!cP;}MVp5qQ8lUO%85a+`}$jq`k zZQ3${2Py64c7xZl>NQRL8-GDc`r1Z%ELgjj9et%0b~Eu>Ue8VH%SAV!J0{m$MYOr# zgRiw=^I=c%AnBZ7M4I9I712|fvkeIkZ&3FUnrvz!GWXL&7u)21@ab+tvuOqCZxdPN zq*bB|E(cqBc91g*?sNak2YI^reonXV%v*mRfy3^(^zY*%x;S|reARsMY?Y?)2Dank zBi~S$>44`&&gdH+3X9}j^fs(JmwPpGzCi}xE><9J*nD0U+lfbaDMiwL~B&Dc;f%3o&xlJl<;@%sPvLF@Ezknj-e(W%CY7Nh3Z=_F^6pF;4${D( zS14m;1m!N8gp!Onx&E1_Jo(ofes<&{%2XbZ{}m&p3M--MsY@a3ee#j}~;$TiYHu<*;+Qh6>L=iNrDa&5kBr;0nx`nVnT0Ig5G zLpPNGbRIuf?h~IP-yPCbRy*ID2UUWH)d_vJi!KKC?g}^2hcRKa8b2D)gF|POK<9ij z?eObFN`|ZcUq>1IrjoWLs334vK6F>Sr7@A+ap=H1NCul=+-;Wd`4z*hI37!bcED!X zA-sHb4*d;nFfscOzm-1lPW^TK`;uVNIvhj)wQ=~+F91D6KSTb*Dq4_Q!~^QjGW{9B zrOSKsLftB8S33YV+Vb}l4_H>U#!(FoH0ST2M71#LGcS>|V;1n|LpLz4Qvgg~exYrf z6G^F8Db9M=Qguip8E$`#Z21Xv1G13y-W)dyJowB*cRsJAk1+~u+HvOrb(EqxQDm$~ z4|V0Skzy=YQPzyVNC{8H5Y_Q8UZ=~&F}-2hw1*Tu{E@AA zO$QfUB*PT}P(Nu2M~m|~)5#pCyDCwx{eHUX-;)(vJTak@8IHUaJ$xzu;a5%!a=KdL znwX_8ExU&!GDk5ot<0eM`Hfp^%+X zw#X=DtBfQX3YC-!g?4D6sFW5;zx($Ocpc|;&Ux%#7>#b~~soww$1hlA1?(6O}`PiFRUPK|H5)`moE5|Y79+Y_Mmj}x^Y zR^yN7{3B7@!&E2vm!|pcq567ddbn7Uje(^2RI5zJ^FKYYn+Mwp|0o>r$xG!YsvVC1SLs&V8}rO%sI?<RcG^LV{-A%3c( z4n;fWEJ z=RZc`^&PCk6Jvy$m)vmNk?9m}Fr60vxQ2Z(r^#6SD+Tm=&}ca`^!TZcHn+mjctDRv z>CdJjSk}2nLk@s7?e*G0Vib!L6uR3}$hO764aFBA^4gYxz!L>un z@b4O7YWgL{%^&3lGo&`+mM5w(=UNK1Id$Sle;H0!=o2g%nF0+`g%B$*MFsx$^r-S4 z_s($?XmvAoCtZW3|CDLaBV~ z^JRp&@8a)3{9iR*)>RP`0ow6GQRwgVZZA3bk5 z?^~>gtCIq#4URDGAlun4k)~@~R*_<{4DFVhO8aydP;Xo)V`H?V*OwSHJ?Df@>FJp6 z{Sv2Cy5Qix(YW{bdGwU)<1UI0z&V)?oO<^#j`JJC_FjxdU_1wb9q-zhleZ6%(Wt4dnij$-&0Ey|X7 zfmJrg7=t<*#cMlptz8!Snro0n;u|#IHy_o%ULv(BA%4~OESBfLz*yi9F(`}qK1ckA zy=g>cjH|lRKaEzd6VT^s#%5tG+VI6<^kthbR)*(d)=7C9aMI@63KRMMWvRS^a50V9 zH$WbBs`=oSPq9@+B2SU6`t4(A%k z;{6wnTxs}dm~&l(dR3#a&sH7Z>>Uui{Ok^?9<3P3?!57WI8y3fM-E|M;N#p4lw?#( zpFXssdwM?BM+s4?jv0~Y73_SJg1)sZub2Id-u~B4Hy5${Mu;&DM^2;r3&eQUt^$%N z?ZcL8Wc&|V+Q)YC{vI_H(9l8kXL70M?NfSZ7Dv-oyv5nCZqT5*IW5#TfWi4HIL$8= z=daJl@OR6Ylo!;smA*tK3R59W)`wYe6=&ygt z@~kL*v5&!yGfnhm#ddnQ{4snSl}D>sP6OO>aCXUb(i$))r5t(m0}CoxbDQ>;Wm3gL zeG+54gv*~gxTP6o=(_(XJsdcWca0ukkndQuJ*7n#bo0Psp93vFFGpK?C1Bd2WbCST zqLIHesNXCfB@29LVx&4n4Aqm*s#8?l^9Y04++J#DAs#IHh<(m%kMb5^hx20m_BMlM z;Fr>Ij~M*vw+s6P0yIAN9H$ipk`S9E3mZkEU|kSSZ|wrZq%}y!+ws5ybIMp2idCyw z7k+~wWFI!5Jj0ui{BSF}&l9J>6pmVZKB4X^VdOS0fFo=+R#Bme7?z2rZkyq$X({9$ zo=6kL-rzeSTbSNmPM?XE#kNi#!XO`27aXP$E*)4LN z$}$zz$LZn20IC@^5q2#Sz*_qQ@Oe)k_eN3?uRKx25MGpC?Yl=GZ8niYT^Q!eq>*Vu zG&%jUpea4f$>Y`lzt%~>uE;DD@0Fp4C*3If^A+5De=o{KCexGrQT)e+28<)N0NBGC z84f~8a z>^r8=na*C1BSqfzAceM2Y9F?wrq+|VlYM?6?i`$J3qr3%L);>D2BqzkxPg0Ka6lkM z=N1)`#v%(o;yb&fos7Y`jEUhm?+jkJ8co^f+zAYpbGr{tp?@Db>FK#_GCa!qP$jx} zuxTfq9Vo-(unly_*c$Jog;KxMKiYQ17GECCWvuc$@SFVTVFmNCFOEQ+6+3CFXDAu8 zp5xtO?diQ#DGiOAN=tOCsdSeZ+J-#A*p)%J{+#i$RRcdv2M>Y<3@$;?s6}lje?0 zQr7>1=7K%Avo??X%wy@(uKPImP!Lu8zyHHHMILOHpm6s$9z0%xDkT{-Y3YBwa$zKI zxBf5GkMpL!^gfDuEd>#bXTD?lI9ya|4@w&f;B&_*+613n_T76>RZE8}dl3hhucYEA-)JiL-$?cokK)|72jFk=AWF$flgo!7%)Gst z(_MO>GyBs;wYux5^tBkSjJ*o#%yT+cdj{4XDkIVJ7sx5Yh|c~tg{39#SnjQlI+l+i z`mQNrs5sheRKkEhJ8-!X30_$?bpLiRS|z4nwx}uX+I0=>_Y|YlvO927*@#+SBcaR% z*eJOHPsT@Lvb-BkDk|d24d&x%je{)HU<}*7lydT;WpH$9IgHNOg50(^NXtCUm4`T> z(CiJ6-1rX5PleIMGAAsaC1Kg{J&11&yxx%yTBh?WsJc=( z;0YG6!>o6Yf35;l3!wE_+w8r7tNrJ3HgO<||r{TY=S~Mv9 z5`8~0wwC^UipxC1nK15mMbl;4GUX3>)kxCcp7D5amyzJNc`tXj?Tlp%w*%g3T;#I- z-!sRg5WjP!I2B!-hvVDhxB<-{Ioth{E$!I~2Wp z5i2)c#kBGSx-jW8mCGA5N5wImGN1=h`+Xr>V>}gYVwrDYeZFsw9zSAaD7F0PCaaVG zQRGwGY?*A3Gl+?4D=CU8Q0gYpuW-#$9`CZ>u$$@ z(3sQEJS5Mug&_h=i-j}a3Ndw-3VKd<#hW6DFg#%gXStPi9`v=S_t6hL^0*E-)i{nNow7qXeV;Nq${hu>O(S8=T2X#W z#&depv6Chnn_~E>Mr_>1xaxI_=*tmnmb2f1&s3zaX473k(ytkuokJX2Fb9F(wdZ){ z51qwrjw5MW={qbl7(!kUX!7;pw$rkLJD@-QbUlgp6q;;c@ zkoejP^0qER$tkBHCQOU#zp|S?T~#ErjSkcrFpaXJSSEYkFpN|DL0RK(kZOS(x$G^X zU)%cWgT5@8wg%$ZWEt?j9>co6EVpjIkS4Ji<_OCY(z+tWH?6(RX6Od|Qit;-J3Ss% z&uHRoJAW{ra2mwd@4>k6TX;42J{E3L!PVHom~6~>c3qgiTmOO<)JgCoj+mnRAqBXe zC5$=ZK6pl91Vz;>Cx6GkAoW(9o*Kx}<`64>;Ug*j!L4if%_EemkBs1Vd`qTtj}Bwp z3~6d|7)vgz51^@dHq75>!|Aon$IlAID6@mP1CN&AKv4`eS>{JD{=eT zR7e}qf?Fwux*Cz+ee)Jy6myF=UD{8=%=I$k;A+(IbwJhFQS`utag<&aVAF5b{gE)n zp3q~sm31E-|BK>O>%wu;JO_+fa0iE$vi+2&2*t2F+veg-@<>mW}0bot>kU`qQ!Qg*Z45D_X7Y_O?2h^@gAG#*RVf5(`g5U9+j9xuYxIlhczu7?Y!)WG|b_1J<{ef-uYw`D_0qFa87KS!I1dTsY6uR;ZT@b6Ji5)`N+gO6T6C@dX zSe#O;y{KfyMie#m!`g*c@yEd)bg{b5X1QjhU>MB$p8;SvClU`O{6_zPeyIJ>jq$8M z?lP(!%|Feha_4tgXz~^gv;?Ed3%1X{UJEbQu3@v!NZj5hhdwhIKXRrM>E1nn0Xdhj zMshKAGcN7~je8igsP+Gh2G+&5!0)QLT(7zzw{e^=M2%9%m1V5&UZBcv9ep0%-MiRl zS_Kno$I-MHA-X*4J?4ii(Veykm^nKWtv61?rQJW_+vc%Y!17jwdcL$K;{ZkWhvMF2 zcOdvwAS&hD!c%MGQS){ru6ZyS%{LuG2cL8B;N?Xu$grdQ`T}}sC!py`=aAbuj5mYg za6DL|>!=cpy1tYelzQM<(0_PLO%6MiAK{qUqsXE83`#PunQ_WPl0RccUrwGzSx-Cc zt;~jRrn6wr9c}b1xQR1=vCqOiJ>GX+GriPTptfQYI(D!X*SvS>Lt(@76WUUAoR1@hQH)T;vB_icr* zsD{q}cBAG;JKVNB1+PmT#@9l@n6hF3=j#-a<~o*fw%BQzviG;;FRg#Dw?anHFI9}Q zD}1ObcMq2pWCtRePjOyAH5Vsn!6k|@H10|%S7)G!67$$h&C)~QliXmrLLn0#waN*MaSEA5nd-I^1vkjRoT!X<62P@IHPBwAlMPPAY<17vKaf zQ?xK~QMo|$$v!X-yUX0Xmn_Ws^Ei(R3+&9^D!3jR3=4znaje!@K|`N5Sa=L#_1O?; zjN66&O%-T;C_<26Si@;dGDcIC;fKI==1ldY4ST{zNd=pdPsznO!oE<=d zSyHh0yf?m{B2I^nghBTAvDmdB71v51ho2sz%q_8w21e#^?N85hTU$cGMO2PE8L5LJ z(K6&)HkN&Vt8m9pj!Y~S1f|<%G2cWO_IbBiRG)3(CSUu8rQ@SO-E0W9(?=hUcmHa}?>SyC2=G{nO5sxEzF->ajr)ZA?=;}O zJ==g|ebJ`dMc7yS946;arKU%b^x~{J1+95WtA5o|s>e9X?Cmn+h z{Xfi~y9_NOwxQ5#DYSEov~0A|63qU(5D#69<@PhC!Q_DxxVk?N&J_28WL+|N*6P5q zJxB4(dR+`#KZORijN+diHQ~J@E(%r^4uYR%KYk9J%ykTefoJ_yE>R)|HD9n?|BW)f z&pw8F&V^Hi%qv_fAtPvC{u#vWOL5<;r2@UVd7Q?aYSho;sF@pbsRTuU|sL+GpKRaC`vbXfc$~d8Os54rdaJW8jK3;j4!TnoZh@^gy){i-f(%U>yZF~;4v6Zfc^Ost1}dCv3v*0+2L(50ibaz%*aR*qc&HKij6! zoW5j83atn22V&T3rQ-gB)_8CFOPo-03Tx6rnfpVQw3^~kMc)^X-Ec(X&8&A` zdmFkJ7T}Y`>!{-SK`OBwL(2rV)b%L}{j*--jg$z?S#boWzw8F*JT4q+Kf%T8^-G!LFWDl zZ1Q}LIf2yX<*GGs!-h57 zzrPnTWy&}*WzWI0mvJj-Ljzc+rux^VN^T|44_Z}M~xxyGY;it%1?v+3{WmUAbDo0N0G`a+HblJ+8!cxD3^VZcc z`J*oW-C)b@d@BtVN7q1by#zgub*1Ogr%A3j3EJ0}gMo!VTD)om+q@%qezO$)-dlpN z+)klgSth(+coGYGLh=3_UC3L@vMZME7*K5rTjh^ak(wBv6KKnCXa9TZT{D{T`6b>N zn#E~dao_?~7NeYgBu?JFhwYU$smZ1uD~c_taE&R6cU{3eu~D?oNR&p5zXOi*jnM96 zBmVg032PEgf$}*wNZ>^1&!`T3JyC!oPn1xBMLoT}^Nw~@G_$P61X8jdM6dD1EQ8mL zQZ9Npvb+&x-`8Nj^efEBRK;VAUz79rCOTFvq(qS$DDm|aXRJ5?yDqH29{asm!Ooqj z=aTU8))asB`)f$h`y%eU`Wu4gZ^U?ma?~BS9cC`whR@#3CgbKPz#s+O zUL21Z$}hn;zM1V$flf$#g_a&w%v1em22~G2J4x>bZ7&mpTe}sdm7n3U(JaqUIC*5S5#p9 z@rR4Man**Sbi9Eu@bpCL4R6Mzzu&>ZT!P9yQCH6*{B+p zfn*saxp8EF$eQ#wzvkY@#DNeb!o==T0y^S5Bjf#xvvyDscACdklY| zPW=xPxT7oFao|w}Mm;Ws-FHLr@R}-AJ3fWBy8S?Xg(@`4s(^!0|53_+MYPgPnTk$X z@ohS@_-TH#;8$G|^m%3?|N2=FpKF|s!gci+ zHL#bWYmDJT^+G5pPQ}?p3b8%!k?SLHIIDg4{Ap`FpOHc%8q_RA%>%_R5q~^5!ULzMTL{Pe)_sb17_4 zQzjdW^Yk?02z1|(!|>h5aLCpKp1x?n^)4Zd^%{nbR}G>5>_ki%{Q~@|Z{qlIzW6>? z055mh;Tw+d(~>@HFIJ~li3Y?N0Pwom|Ns1I6grU$g@5&FhUPa)l-dfWEH_&=UyR1g zdIZ&{<8iiSAF5sy#hlbEyl!ocR$D6IKz=v3eQ7!fJ(b3bdqgNchxq|=E0}NfIiBkD zBsY5t%#*4`o7|1`y>|>GZMFo-H9a8rJst-%83Uy>kk$=dq~51JwAfLfZ!U@Fj$Cg? z6}MQrtInf-R3GXv2Kl;AsSy4=6Av6wr0Q@pE|@XS+jbaXZC(_{k~oF5BvEtpB$&71 z5$s~G#hc}K`dv?wbVUcnf+psuXtF9t13fF3NXlbjC_sLZ&u=b^SR{M?MUA%U13?~ML2Jjf-~+Ul4oW*yxouvUU6EW{isav=y4TT zsy_gA*G8x)z6T$&j^Y9>Ha~vM_;nUpI5?R(bmm*q)Q(G>mX10o9?8W&LW$fbjX)@w zSBJLD$GKu*DHOROl@(av0NbhBTn6oiCc~unH)`xBl7HQ5tzM5$6M|pq_g%pJjt2Eb|ucV ziJfc4Iv%5%-Dk-D@jgyMcrQ-dra-f~0OBt)KcCzXmKUGrdOqr7;6i6w_d1?Bx*w8A zh9=+szMZ=wEe(417vXyJ8$nCQ60U1qGRnMQnUQ;A$H}Lj7j{H0yBT!*IDbZ3F+}0g{;S+=K@l7q-6z7L~odPhYY8FoF zt46tLh4f`#3fYf5L{CpN(VAIrsJ&<;c)U0c0`Xlqd7~*kSCFRg<|insGl~zM{q7C>4|juaVeE7m;@fGpKxzYD*b1A z5so-YQOTs0(c0$j`GP;Ip+o2;j2k}Qn!}X*yua>R@W6D+C|~t3A501 zWFX#PZm!A}0r#$J1m{zI0T;UWaO2-x=5C0X0>9Uo)9+^aa@XBt`(Gv5vn;yrjkoyT zy9DlykHPPYCGldw7`SHHz&Rfjg-9)L?i9<`n0Cw48YMS4;`SB#7yd<)?oS{(C!4<8 z*z+67l(gTQLx95?n11C5*L?h}pzF?5G}*ZylLkk?cw35B{)Rc)rjV`1|b!?wj-p_gog{j)tw|^f!cX`70lC>vtETfrJ!| z&XMHDT)050^VN7)rIFm&{2+|5mZFn8FAF-4#(?eAY@8<)2FhJYXuVmFOb#hh*N+DL zB9x8a*37{*JH=?pwm+oFde80-Lg*)L46#*p@bOO^`Rw~ay<39lwkP8}kNp9H%L`EA zO(*Wqioi9m86VIv3Ck{=XB@+4*y_K65=|`VD1QkJIxVOJl$q!57?xP4AkWT;v&}}} z<QHyHpB79?+ef1-rLQnJ?T4K?i%1~6tcNp zRf_mw^${3O7J<)|4Op`z5jUi%V0?2rtYqD$g8mYosaUV$hkPIV+KXCqQT5wOL6!|1(;4jToPW51dAb-a=?x%%2 zY#LO7aYFMsw`c93x3L!rdz^9U_7FUOc{w{XFT#@EeK>8OIZB-U1!s@{MK30mFM73? zUcY;Xjekq9p>HjI`}P4YZnA>qk9e>=HpJB#7Qvw+0ai+^;4E)9!P!+jjP%L?ju*$3 zEg!fvz47!ZA85&N8Jf!yXR4EAIpo1QO~L@PW?c|mnzKkS+u0G%hSlR>^?AH9S%M}E zgy7sS)nIUc9JSi)po7htB+&u*E9^ZiY?+7Bmxs}8_6?}hxW%n>uRw!wBj{PXmVpLa8o^;RW=#D+aJMXhXb%~b};-&DS{sTc&O2vj=C3fu~BmmPLN-P7Qc8@ z`*jZ_PkaT(;0xTfzDe*^|2=l?wZK_tzJr}tEbQKY9rpHB;^;SjV3&_HF6LcPl->OX z&Wk}|Wf#G3p&s-bv^D{T@6n~)^hO)x+q!1vJ8AaR;VeG^f<)DE<-Ti zbs<_jS_%pNV(=}tn5%yg!ur#9@j9%5){AVOlwS#+AxbFmbT!O!H3y{|QQZ8b4V-o7 z96=_V|H>Rvp>HO-H1;Y84*QG&f6WPywXp%Fh4^9Kv{IJOW0|Yx%8-@h!~J;XhV9%%l~N0qfxx2SQeghe}WzfN_hTAt7X(wZ@9mnVk7F+Dpn2C{*xI_7Ymt4xWu3gj z$zHmF$&AI9z2_dd>U5LBTNOS`T7x&fR)jJK@^K31hNd=ocy@X=F7f8^KUjdS2BM_< z$y+U&Y;)~sp#B({>`~->XVKc9 z)^PXkK|J@)j!I5mMT-d%{3@LozPoH5R`2b{?yerZBs&YQUL1|Fzg~mg;wH>l{tL42 zX=4`St7&|&7HBUo;+oQ!FZ7>0x@Z>*{%!BXJ+cxgeC{JR%GQXaI1{p8kPkb=5>W7f z#~VkLvBF0YYV6IBhIQ!9#-Emlbqc_;Q-$VjiGjx#s&PV62~JvY8m)g-VM-s{DW%@# zxTHu_ik74#&msJ9jq$+;+^KZ2B|ZIhnr^-tPi}kmb75O-&|}MFgh|Y4;;BaF<5{lJ z#hSbo7gG2~4YGMUpG2;$p<~PMV=sFSgxW1+PSH?0Bpd<*|MA?X{93NdnmwD2G~%wC zmbjID*At6&pqkP#T%T!7trxyR^K=nV!Y*=tc8m6Z$)nrdk1^@sEK)dAhvVvm$g1@P z71lA4Q(=JF~~2#cLJZ*cd=7+%$06%Jm-u0WioFfQkH(}!HdT3L7#z5}Pl z-b1PFGjZDiG4k!-LUk6~aCfK(?pFB^{g-ZK?61X~gF_d5S@@6YX690Lh8N=sbYoAk z5Ur43&gMULxMD#Yx9+gGAn8FWK8s~6D-$#L=Wz?&zV9ag{lzq7^O!rZ=oDm?1Yy-0 zHq)>-r<;nZbh;Y_%H)AQdPSJavMmaXnfU!vJ)MnIh7KA@YHiiDo~i zop&MNYnkBhgNs}i>O%hW7r3!@J-N1OQR&LNoQ}^rs8mKp(TA{adnI>TbPzS4g;4nG%_OAe!MI$~BrX{U zUuMRz-|j?i*?diO+_+m1niRpUug`*;3)fR^(rWVDmjJI0uOu_OQ+VXXbo7i^3cp8; zrt(>JG?#OLEZsJ4y;~;iPnDTAbAS@txs7Sk_X!QP|K@PN4kIarml z=j~YBttJa&_WNUSo(-%i)`P8M{P31jHCn_f;w>%aKvCBK57Uc+4VmA#1sR%n`k)R< z_ZCu(`7zq&Ye;!lmFctO7FhH1sl`;`MjB*Ygem`)!I!f#B)RMeoa+&S53bsL@0Xuc znm3xH1&MTRK?!P{*N1g_OY!H`E$CqHjMrFYuxEGGRB_}| z`$18Jd4)5#(lqAS>EXJ$9o;!pM)G{j&ST^q@*g!A&&2e6=HjyXi~EF!z};dsB&^LO zT_ro(*;j*3tF}^|QxcX;&;xxJITE3DICX|FC-vN#?ik9^jmKS3$g;cd%jaQ1{17+& zpcTpKJw_2TIgrx|BjJA~)GiW?!tu)Vt&Z`N8>&%q+H_Kh^P&w_tPed`hI$U3!lDm} z*ke8iRkV)?F7G+c&8m?h7hFYEMLGCp-7ehsJRF;ivTj~A+pJEO`6SS;n6vumj(JPe}Bi|F~!CDfar z%bpLRaacQDTkvy30yVI# z%DN4z++VSgG{Ev=&q_wnZ^>lxb6txCV?J}6tL9+c^FhHFQsF`^W&?1ni>|kx25u$s zU9%ne^*3MP#fBH~X8AHm`jN`TJZ_|iIGHTFufd(%Ub=cfmH!}8LyaWJBBOnh`^FL%&Y6a3W{5fLruFpUtn{V z7y22;FLO+$XoqI3Z~VsjxeK_?)&P9f{|e{UOu-9mzOq>46kk+)j6O`gL>1xEq^UTT zUIr{7x2u6T>@7r|`o@%7dY@a7t%e#8?tqd=5ZA_>7~+mfC~kPh;v?h7jNyKh$#qv= z<2+;0dj*iNhYHkRm80!H6Jd*sG1r`;MQ3I=(Je3LRPHjQ36^iD+4cs_jLyZ5QzqQP zBLUpT{(rbL?F7Ac)TCMW?$f_hrTB2?G%mbiEY)>IQ&gckuQ}xcm#cjbK3&?1>gB;$ z)BOWij+{mD&kDE@yYJw2h;hU3OOT|nKKp;}q3P|B_@Z8b$BS!m*M`qfvSb{ZYfNNp z-rErCH3;p|sbKJI8UEgw2RFd94(2HR@Oq5PCqydOu$25 zrBH2(7epJCLiO|>EEc{5zirRKQ8vqaJH`(8u8T+G{w|!r*dq=)zu>-K0QzTWK>ed| zu3CBvb}Gign(}2JWH=Xk=eNME6CLodqXSY)4etvk#e*Czg7QEcm@z9ENJYIPS?)lo`Amgre z2Gw()hb+-|R~7EG2;mBX9r1){G~V0&5p#_SakfPSYPGJ1R)cabqLZC<>k8oTmeshh z-Us*mVO4{KG(@&1%Vgf^9WjNt*Gw3xW$w`V+223kb%;N-r{EUI>v@Hfz1&waA2=Damxbn z`l^LO5RqsS;<+!CuxBUM&|Lgr^UoH+xf>ncI4xrw)6p>5lO@3NF8uRTl~ zoONGMU0DPxMsCJaQzFpREFA_v%G3C(^Xb|5Mo^l%3;Q@>l71xtL9>OhXe;1&=HR(% z7!Q9Mvry&Wbr7&!!%3}rTrFA)&m)(EvVlC@`EwIxY8!CO*FO9gSt#gRGX&jnO-z~z_Nd#ODSqGdJU;&y*%+sk;u1;yB6 z&$vtm%r`p!2IZS7^9M}41mmaXbG+J2>?i_ER5^^kr78I9Q6t>H>44{LdvU|qVlH*} z5jeB^IfM(R!2XE8+*+r29Kkqh>Z}tv7`GDbZLeYUk8rFyoPrti3h{)bq2SDvf0kpO z^O(oFChw(Z;N*x1=8#-K>YnX5ma$(isa>O2+KDaw@l?aT!g|zmKObKY$1Gx1d2$KV(#x!xFb+ z0wGlgP-+mx4ueTtV(@NUDy>0JMyIj&|0r(R8x8G$SgyBoHk_||!R=mu8(cj^P|lvA z8-w4$!~w(>R=yCbIZmh57qTi^wJXeQ& zych#dEgKil4a3nDmq5VSmU3P}B;CPsZ2IM#VC4*K&OVH0VQrXLF#@Gnu2C&y1YMnA zOLB$G10$7zOT-%l8S9L}yK@Y`)8Qpa-wCISH?3GVnaza~^Wm_NKh@->lJKIZ*t$~> z3r9G?rR+1@!N4?JdfpUMeo5iLJ_k&RZs1DZJA(Q0VyGCQh=zNDP|n(gp19q?myMn9 zcEJ%Czfg$Cb|+cZ?!_5&kz3*4B3OLqKkAWBg4urUT&4dDDCmg+DZ>bC9bX5l1tHww zzL|mtdHQs1-ylvL?}iHuQgOZdEz;WBL94`FXtXeQ9FS{1{Zd?YL78Q~W zN+N5ORoh@d}(>GYB9A+8Sj z;1|DDg0bP5V0~j2R@6AanUXTjod1MvS)*uiy*f1<6^2hn7NnzepYUccM)r*4S4(`Q zz07&A{%0*|h$!(BZ&nJ5xfO6XwV3f((xAQa3&sJGlm6+>4`QJT6^ zh8rwd2B-V~aeLg>a>-vyaK2F^#%%nA^Xm_zwbl$?Pk9qRx896bIk^SKhP-C8%Wf<> zf0a5rswmXmlza^rP-^}QQV*!0(s%>vYkxvM`Z3h-fw`zQEJEk1hmerej7pP@xZpKQ zVQq>Y*6&z^sd)zIRhN%0SCH$-z9cx97EOm7qv)7P0>aZo+LTs|FMJ+w9tVC{D4lWy z!AMQ=ZkNYf-&2|APnbU)%$!`&wlvFr6z+;KBwiyCf>ot)lFt{I(-}&!(~@aKZ7%ot z$4sahD1q7pTl5p#Etu{V!BHvmVk+5T_0pLXG5Z7D)OUw1Nv1fl>J~dE*Mo$W4usxf zui0D*#V#G?ZafX9NbLbinRJmRMb(1r;p>8>FGhkvrYsHTou`j)*AwdRrrT@YVpRGf z^yh-Ge3vs`Gyeh4Esn!Wv74-iBu8RjUZRSjJT*lu!q+VUxJ~U13h$Jr%Rlq+YE(JK z@3cXyZpQUWcnUx3ol)&$D;2m*rtkhq)Elq}o{cWUvE9pPUXUw(&bLCZr9a^QQyW_J zxDwxd62ZW}dUX5r9U3+caLv(+&>~%uq*i_b@0wA#u{#+%r87ZIO9J<^PPAvS9iEvI zji)7;FYICaGucx?99xFS<1k-Me#c5kl!k@64@>(4F>UGN9*R_oK1 z7AxE&tAV4h`QV(AA0Y3VhHv*9;A&14S9DLOt+p9x8F>M%JN?k7U=1F-yc~?&#c|q| zYf$UR=sG!rIJo719G!_bReu+U&4o}hHV}mrihIxA`$XmlsWecDN~Br8q)A06Q;JL} zGAE%>k)b3>1Bp;kNu^0r(mcK2_5K4{t98%4XYc*_JWmT8&E^# z88tjwNh94#Vc=pH8h&@DJ?$2_HX{NDE&t%{HwEb7B1O|JBC*6>hGTyN?i_pv&&vIv zxWE9bPco--{0~%U?<57AzgXaFNNX;%;?;{TIBA6n%&I=Oyk{MeF03{V_F zcV>OW#NeCw*4UB~)RWNl!3zA|9){}ackrxM8`o)g5TdPbfZ3Fb5OE*@b@qhe?porH zMos1SzPLf{EKi{E^)4p7{fJvT!*T1UN>o&CfH229I5J!Xjn)#*J2IcRh6jil(-|vX zo(!*yrYUdC!2U)N#>~*dS?lEKFLNl2xp;>9jJCQTiMf5#|eN(aksI2;8tMGcTR$#~65 zFOk2z3G^khLE}G9?$cIJF8!4TRcf2imWs15FeeeC|71gLo+J1Sdj=xe0mwg%p)qn- z;M;$b=;>)`I%>jhrGA9}k~wVJV8?D5ohUW44%bV>LAbpUeP({TN*^0Kkn#zKo5te7 z3mjc(8^=F3eMCMVtvFNG*WU7|2j9<&riBLENVY_dsy^eLTNzg`{_OUF#pCUie_&p7K+J!#{=TJ~V9g5n8fD#?-@L- z6UgNTRSI_|&IIXyo9T>m2bOJcX8hJBT7P6FRc&{GR>rU0=c@!qW5;rT)||)HffFg; zNDViw=-_4*mw=Y_IGo;Nj};9|;fY2hY;-;eX=n2BTwgW?UHpb83f!@>BoJq{*x`ft z0F)ZN97iP?@s__%qlBLx+;kD+4`+vx(W*Oe%j7s1PCbbBq9@oLAsstZ;-Td~ePQ`i zSz5Kr8q?w(A^nsip0?ddSM7%JdV)&syTF>RmqxG-#eDR-_>de~r?cX#w$R|W6P`7) zp${CuuXA1`Tr`Qlu<1Idy|D#5b>>p_mZvyRH4fC)lyh4T>_F#{S#W22Iyd3gB-~J4 zkH2=Mlffc6dbH#;ryMX>7_Byl8ljPl-*=FvjO`a*JmSeMi*ZA@+AKbC<7j?vQ8X@n zbO&~q0CWuQf%X-{QBoEmD|9%T7n#wpV-frd-2y&3l7sD$Z6G{Zjkm%Isg2j+C6DX! zjttIXV-37Vj=h6J86-=70kEX_} zu;toB(zkLXt(Lu*wndKb(S1ry?2g;Iqya|;8If>oG0`${*6H7bETGJ(k{UQZtA&M2 zb>;9Q53M2LeS`c=KHv#)-rs$DZ@8B!#9IQz3rsJhV7{ z0}^JQ!gV7}H4la}IfJKZS37qJ^C< zMpQf@n#9}K{AzX$FRi+k@)igXbjV*YBW5(mQBV&=qVg>*hmA)0@{I7 z!QSvX_qHie_{~rnyk^ei?vAe(c=YTbh&V|1FUl~+{Vw`(;yP*RY4iBrmF^XOCeP)4 zG@vAoJ74$F@m+5G0k?4a_ezx4TUbM{H1n`MOoaAqnns3#$r$(|1J(R*p5Xcmers=RtWO0E%b%Lc@(PIC*>lo;q1obfamjXk^E%fuNbf7BL_Wbif$h!9}B=ybvI~Q;#K_7q=KyYWm=p;DcQ8uIc>t<{A9_R!?3*N0}6vn{kik8<1zY#G~RK zcr?$9V)Ru}d9OcBnrO-Q-7li#m$9^Z!CD$zCByE$Gci(!F3GI^`oq$s#`Nxu486 z2%qB8TlUmr^B#Rqan$=LoR)?!r$w0!Sl-nGBiK8uzWyXi9h*<8mWU%>pvIY=0k;#Qt~WCCeYF!tgRoOUt~a`iensi)&$UqXkl)jdr3=uHfsIFt_W zZ*1T;pUFXeWlPvF?=~a`g~Fa^39zQ{F`k{R&nGF43$I@6RTCV`hF1?1B zMN_c0cr1SP@kC*NB~E+#7gtRbA*q2wxHTvq|Cy>`)X6J2E4B}3S$#miAs^U@KVjz& zWvoxvA&Y93AKG*nLpQSIv*`tHpCAmY%_8vWzR75F=NkBY5b$YRt@&ccLD|E2Jh~H= z>2W#h4<;#6XRZt#vUNoNu7j|rn`Hv;-R3T)Ou&bKkD}Z@%|NgZ0QRH zwXGN6?S+MOI{FA1RP{68y%jcFg>g3%B=GslW=!l~jz7(ZfwZd>nBJU66BfS%TaA@q zIHm|CcZb7D;d7ARvKpG~KXNTUCd02x0p^>8psHmvcjo0z_u||mJtVbe!p?X=iQ`VH3i!p6LFfA z78!MhgOg5zu(IhtuE9BuQ|GGWRi5Q9XpZ0QQ2S$Rth#T3i z7)}0{V$rB$GglsBDD<1|4IzvfRT;7#bx-ACv*b3sj~%cx{}iUA)L~_>GIQ?aLtMaY z61Yn8R?-surQjYAS2hNfS?6Jqo)}Ht=|#SKU1?{lKFxQjqzu7pYKYCoYC};H+c<}O zrEB5T<}~P?rUxdUZlL;+M_l(3PdMk^!m0m7khOmYSk{NmCq?mO<7S+prA?(Z-_W$? zA^uzU80X(iM@z;uEmSNODEhkz+Lp2zsxDe+=VlL?$vbeveR<|qYeGAZn~PnUOYqxAn$X?z!3)S^rczJRv==fgZOr9Zk zi?tC#T=@GuY-l=*zuzn23;(BFPQz|~Xr~MRbxjX-t)5SP z+KcG)%NS(JdDIa%0FQa!g_%kDc=_i<%o!XH#zBKz?!EJZ?J-`0fm;huQfwn^Zad6* z+^K~tg@(c#y?KKCm#1;YI!osGH3r2ZJC-+%#o5C?Kh8nurWM-6{pNt*Woksr-q|?{w_$!SS~z1?<;ySHb>(QGkp1HBF5dE4HBCl zaC$z+A*}Zo*OsS(d;U9(mS|kzXb!e!>J@FNNXBl|7)4HJLoiZs3IPe<9fCGHxl}hI6E!!=ay# zuyJ=d{;C*>+|_g}8Fdy<{4vI3H?QI1xA)P=dVo3drjYT2K%6^=hlaUUxapBM{@&RP zyOe?n3>dqr#+V>yIQL_X2ix^5p~TIqu+u3If&K?i|l2z2iW59pQR;(wtaSf>j@kL3_hd01e-zD*z*zYDI+ zjuBdiycC!`^+1+>VthVtXk`1(VZk5BqcE6E)khLY9j1~0WdA?EYPcQebtXsh@7!7M zOI(`Lh9|S0Z3lY4J%VOxk?67oK~;Yqs3+dzaz>p8{h!6$n5)y^z`nnDc3v!wy>|)v z>MKC(<80yiEA7H>9~&TR$wltcg!!D2oH3bMKEb_;;%Loy306|c`0;!t$oGnY>~cjY zJb0C}(Md!n0|{C*s~^jf8z^5lhpd0r;BxPa_(v`p?OheHF`o73v!1}U=3kr@oS-Jn z2)dBtz-}3n=!j7v#twSI@Zty>-Qq|hiUrtqPZ>iVmvM9KW((Bn4#R`OaU9bbaBq(| zqIh3I^M$7|VY(e$Ivqk`OU3BFG7nVR&m8pzR-mFWg46c93}bVi zbFvB-1>!*wy!zrte7R&kpMJL)MN8+QCF@aXT(rO|OVT*^$#aA^3q9z7lsc6b8KS9I zF11fs%(|@ynOkEJcBOU*_x-IDI;|W*`q!T@uJ(8o$@+qF*XH6E|9-CS^hc8XHJhxC zE91?-XCW)+9lrNEK$Ep|X_&=Sew#PTUB{fD4#S^RLI23I^d{X;ApZHOPI}_X?rNE% zDaz$At=%R{I`2NxwP%U^jh*^%iE-k8pDDqfs*f<`$}BwlD;rO-+xf)q^VpZ0j(gK{ z@$1h~SbXyyj4hdk&6_u(PR>&NqNPh2z4<8Amcz)uwzxWd7v6JN0rg8{$iZp^HQisu zJc%FSb&(m$Y;nWdYq22KmV@K)mo*g?tmA;qV=HDBI$V&xbEZ&1V+)Z>1Pr z{WXk64w#b4R~tIAHj(2_jLou!?)PZgrFvQ! zwh9e>;$f&u6FWV3)15rlk;;z4$f*Wso|nYM%}EhlT&+h^Qtxnb1LGHeV_jLZeUy=S%D;sM0F7NVQ(9!!!5!+T~H zn0i-{_K_nkomPgw)+>f zU;V-91}2z1E*xb1nU|C8*z7Yya8BZM3OL|SAqiG=a`|vhRi+4fl*-_R;Vo|J<1lXS z^n8$sEJ76pJ1qa!j@K@EQ^fX>G+%iO+1M_{P6a1gXO~6uE2DAGLPdyKV#GP6MdFjn zt1!iV2!6Y%<8v|g89b{+1wK0b&X^_|9~4b7iy3Q-`7<)}uHx?MVW`&V$ht{;@W#k5 zq_{VdIG+?c%ACyegiaWsnh!z8GVtw7#sgmU3zm=AgZhW#=~MV8(q!3{hr4%SWz`Mb z`H%xe?G7$WJAhj_L5((VkD>5gz0~Mx0CVL=;N)CY=J5{Z_K3~Gv~I>WwW?}kk<;O_wRi%)o~}-KD5B_ zHCMQ&QinOcwVJfplCl5W+5}E#Uh(tJ)$%1f?^BCXEX9|~@m-8FR98I**6M!Zd_pb? zu4;-=Amd~HoPAVydP*t`wQdEUEl#-1(h9ByMhmy!b`!ixnGKrT>;+%17?@f=nQgIT0_o zkAxx%W%Q`bgxKaj+L*0GUnIn^wl0$1$Zx{uzb^_7UQna*iyG7=riSI^_Rwj*P8jDT zDID+qo)Q;J(SidRD)){rsEj*|%U(>%WT8RGI)xYp+l+?jQ@IQ7{_ zI5@eBvz4iXQ%C3Ff{E{O)u0jG$(6%;Ut7VCzG;qcxvS78uh35nnR->p($$W>|RlHC9UM;ylJv z(b!Od=YvOZ@BYLJI{hb7MOgt_TdLz&=g-`OzZ`ee>I*z!`!|parVtuNOav=ey;f|%1|<-?}t87>hT>Uks-m0KTYKY8ngJh^Tlc9ITe_1ZpgL9 z&Y+v07~{;)4$h7kgICMM$dA3l+T><&mup{h^K#=kZ@ELX#$1FX_$03s}3 z3P%@;Q~1VyPHOi9lhr9vB>(Xec4_T|j}D=z^eBaU%l{V6HYkVM{kHh`W-jMGbRK6_ z&nB-`p?FFE1pnXZseDK@PirrJ!p0>>ap8SUcsB5o(q>JiozEw;`RxqNpJ77#Ug=Sx zsUh>$4^Y9C{dh9?2W@hHMT=urpt9m=ZbZa6?()tp+^YXBprQI`%(%Z6Ue*qx#3BW_ zvB_F^WzkW7qrDj~b-e-uJx+j<`z4dI3s+6<-+3U^*wp~b4mzPksUKgmXn=n-><}-{ zc%Fju2T8TN1RjVrlD_jK^7=9yX1_c^GRLFv$PE?l-BfndYZ>H(dk+cLMWS%_Jb7W| z{7ylqh%>*yU>jd8)r4H`Y$% z*GcgFH0Eo+LDwLUA_2RYd9RUNT009ZjCfZU7Dwg=!~5A#kY?C9OV9 z?epJobHpuhy^d#t`u5+nc4jK3Tem~FLm|!=QvgA;I^)~^;@(-_89 zJ+$M^6vjFB=@9zv9}BM1{_x7zA3DAD=sf>}q@(k2|Il)--rEk-G=X&=uF>x1^|V>} z97t$aa5KYSz|f0ivVU}yyQQ{*%|iyHoHmcXS&E`az6w-2BnkKH?T6|E!}#c0W6bk# zBNLD7;HR>aacit7gLQ&R4o||e@5UfsY9{0j^gwm=H~5h>1x2>!qF~=Sir&(U;}+Rc zSX4NknpMv^-W0NIVJ&(oZ^bPC1~#W`LB%C47#mnlmm~v7{MuF6vGXRL+@6OatH#re z*vo9Meu2tc4snuR1EA$Oh|lkyqi-c8^yO$UML--CJwAbno#k+Kqmi&?oup8`pdOCN zj^}QBd2mObOX0(R+sS_Z00pXZywA}{H1Bt&k0Xujc%Q20q|j#XZS-<)15MVQ1`8xx8JA-$2^U7- z89_33)hi&D>>z~$X7I_dmwP{5gD=w&ouQbtigz9w$;~;u9%jvGLGq~J?3M~?f8%#r z{mKr5ms*l(c|84L)X?UCf6;q&C8@aG$An!8)RnP~QWM^h^w4j77McoO%WUX{<9_(* zy#ZQwtmgkR)|}XgG?a}DhQ3s`SDh}+2O1>PxRX38wf~jPJ)~`m7UQBhHt@0siuiLbk7;M3X-zknzSGx>g#B&rHOr^I!yB&{x*2d8{%_!XbIOaTy!LQpAvETGJ|Hne3mtf00(`C=3f9Im54t9; z`f`pIZBU@B&`u10u^qzxeimF0&>@v(<~Lk5g`!M!@fsJ3KSVt6^S*r2)_FoxTx;m= z)qHYx%%rVyLON2{14gNC_*CeO-a##-7`TLw3yS9dBwawcg*uvY=dsP8VyX+p<>5FrU<-y1E&!!fBk)$;O3-}#nCo_3iAAx; zFjz7Mq9&){K83v)CfN@ueCZjOU4JY zTrM&HYBi3%^A<-(p;=K7zgSv#QFMPlKcvbSaPx$O9pl$ zDa|4uokVl-T7M|!{XLEW^c?zh+ZlrVAuN;oi1!umvhSOE)EuWqql{~?yW%xSr|zT5 z^9$%~=m%7}B0~?29a)Y&0neXZ2hwpiSTg$|HqCqtV|?6UM)m-=Ey|8&J#4|-V?FUb z>$t64SdAOphPdemZgam+tc7`wrR*kl7dwVW;i0Ov_*T0H7HB=gF%Drg?8YFP1|*_Z z&qb*1Ux=gq52M(YD|ox6jD1cnq3M;k$h&`o8}60x>!b*-{Pd3MKUh%z9%sC+n@g!# z5wt{ZA#!i#Q#qGJ1ABH;E4$wrd>hN%kXP9n9aX=K=c(jsapA8(Gn2ndCo$2$?JUS z>F5nq3z0+1K&(04wO&Up zPju;==p$}WLj@*2Qb+T4BYwl{9=ctgfqa-8F4;B@My^c*tr+I8Zc{`#AI6znEzcL7 z0p7P?n~#@^hHw{g{CZ#>-Z~sZ^skCsmIqSqz16fr<^$EA72{_N-N0qR6DVuEJC(@B zK_hd2$CaBh=8hu{HciHXnTt8YnQXtS8i8{soxqB@AyihmiWXODpt_BK6ygul$=Vy_ zdBU2i#`NG;6?3d_G9_-cF&8D0fiATHc)4)}V;yIZ_RYm4Hf1io`hFSb*sI}_Pi?5b z(F4=Dn}F+6QAzRk42KuFM-_+%!+Zw z`zSDTSt?Z9p^EoUso=(V2VA>sABt}j#|LKy;LE=@JaqdFEbo2?0mn@Vnt$W!2lMD^ zLIg!kv!OrBp5VwN1v2BjFwLf%^)n=SmuXV;{0;kEtv!o+W+`Cq5eu3YD=A1&in|*g zLVw+KsFB;mg(=o@*Oyw;0h_&~GHwHH626DR6{>XRqYCAnR>Xviby%NPj7NGc=!2>y z?RQ^FZ;q`eRrVWxVn#LTZM?+%(u&x2;T~8#Go}r{-HGcvNT$0}=+3|clySd~QNeHM z(=RFhOXX<(`P^s{eL0O5$r|Ib*A;k3OM;ZPe?l)kHGDbL2(wSmhfU^N@tekO{8}B2 za|-fs;Y%IJ-*AQ7ux1$Muh+m6u8a}jc9td#Od-Eb5?Cx}h937)p}_S6Op;I{2X%2g zZe1X(u2A6JFGf(M&2ZYIAVYdGRd`NJi4t08l6IX8eR?_ylU(b-`hpq$Csu&{CO$As zU7gZ<-q539TU_~hJVaWkK-vaN__!kv&Yi1+d&_r0!1Msz6ZICiFHfN>v48RE$+wtM zRERe;MW~iN=g0PPv~SK|Jmy!4f#-(Mx9K2-sVtyFYir<79=pvwyv|Kr=LiQ9kAP~H z4lEstLRZ-|te6ps=dT{2igI`At&rhWZT}Bg#JgS|(0{$yKVT7~=L}#`_k?K}lvEG&@9cjmj zr(*DM$bHV|pE5P8)Z_bCZ$O{r)zdVb5PvdfVzmpk=Oj_U?q?)9@d)0MIE*cw$6>G5 z8!Y>F73+RZr@7Nk(WU?7c-^z_==R@hSa629^Lm3|PYL_Za|}cAwpFC+=9<4V4giGEv%U!&8tay(}&)Xd~aSe z$-llv$4d55d*LE7v1i_bcoXVa^bO6vs8eWFFRC~Uhwavhoc)M%T>2X3v5aFJ{Ioi- zSvO(E^|Iys(eD%a)VFy!4V%Hq^)ReWHl>yOV@Lzqau?ycVj56j_E^$1HGG>s%UB+d_*@hjL>a z%7p5pl0f_CAdK2S6?@NoM78@1P-K<}dg@5fj`R?^)9OLj^bMq!FFH(`W$DVv3r$Z=SfkpOr8yc*w6h7`V0Y?r_pH5bKG?lNo&B6 z`c2hgG~@ofEOW#=9_OGr*&PF%>#%5TNJ(YgK+tr@#wkb zAtpC8!5fTrNg+pIkO@0N~aUa(9fR(~A5?i&FPirvb>-LZ57Uw2`$RPp!^YOwaKRq-~ zt;E|ShGFkkagtia98Eb>V8pt3d@5s4>sxt()J*Vvz~KkMTD>VDwll=j1}+S~Bw z*>RZK@C)|m=JK!4UgTF~j^S5~z6%-KHsg9-F$!o@<@TEG;k1(HLh|=$>{n5w;Jx+~ zn<)Uz;>Te4;ygZ2o`8B)##qwb!FlBE#1jJv^y9ZO|764=q19zc$Zx%c--fBebUja) zrNla3H8aTWoHp;|9?Y+Jt3szg29mE2^N_b66b5fe;JzAIVc@9`-0Na>m_6bL+|5&g zk4D3A;mQUWGesO`z4;F%es|%rGq&{e=WxfilWV+0{-k#K%is5o|Nn4z}8 zdx16UyI=epgmu=#04_5|cg#KfvCRPP$cKT0O%`0~G6BzHvG}OllEjJz;qg&*49YtT z_fz)boF78+P!6Htn`e^uo;%Fhy_kwTf57znjd<80l4Pon;*}U5dN!?@0$jWB(Rxdo zCDl*=)MijlUJ|Xa7vn3=O~b>rr_qZ!DoR!aQ`m|qQd{(m4%v7?@`SlKU6jqz$D3i1 z^e$8%6rq{*nz+`_gr-D}r6=!KqsQ77G!v-c*!^FH2a?a=)g?NV|BK~G!pG2clWsKe zPZKOkalqm!XJg2p zNId)g75Dukhqc-jcsOVU_wZ5;j1PW_s>$pIwJ(f?B+gE75@oE5Af+>fdvisP-P#AS zR_PC(I(is`53-Kk%OiN&U@3kdAx&8$Md^0y3*4Z4iM!KfgwlzOAM-YmZ%o+5M_v=B zU6!SgBO6Um-j-6}Hs%rVT#I;6om9I1VNWRYv?y0fkQt-;gU-xBpXIhu$Ts zg22^eUl>QfT!(|^*qfYO%N9=diyc2d^$xG3WX;>|iJ*w#t4Om(3w1}Y=cL+|E3`yCzM=`qn4Vl(L-Tm8}%ngz|!mw{fvLHOhJ5~f_bjn^j6q@#b6>FAnBniKENzh0Kf z>pz-=^1>dxb(!^EMV~;lQW%b!Do$JUm*IlZk5RQP9*1ort}cEWI4C6JP6kPl7+*&5 z(d<2}Aq7I*Cvu;vc>HuIo5p_`#=mh>=B00S;E@Owa9I+_eP`Z^Cv2D4(7GDKyBly# z_6&UPkcoPy8tFRAmX_qqC;5wMq%rjj9qgS?y`6in8-ULHW zQ!e(8D)mf{fH&EDgwF#MNsIO1v9J&x8#&;z++~b4_7)~3Zh}ioEij3>uyT~oKv-(T*O1TLb2h* zFC0Z_+VwD1DHS%BW`L)U6=i?I{$*f!MI%TrP=!fN ziunG&y{vvNMy>jX(Wu-A4>At*ndcEG?)#AQi#R0AnCM2@HhDB&ph=(PnfGdkHm(~` z2Vu}5+B!9!g8kFTM|B9zN9a>poH%otHDT=ChnSdfnsE=WBW8`l)(@Asf2k?xyitu5 zo{p!jAaBYRrFWZz=MB~1^I1Ri zKO~30(@UXnzktp;yu{9nTkzDo9?U-w1tJ|k0n{ea`=aBF8RZ6U6ocO`zr_`z%(t2C zjlQdbso<6^?a8S@`_vGOP*KO=sskXiNC!7CNA&3XL2$_63k1GfkCk_{FmI(H_7x6+ zSg9&W?ec>2FDK(SH-FNmQS?xjb%e?ml2A!NUOkdDV!>5Ba&#Sc_QNWn$ov}a>@t6D zm#h`!zWV~-eATc`B>{V*{MfUUM-M7PX!Nl#Jl3pD;Tt%**k{FcU(3T8M;u6F;wTiA z-G#7!FSZ5rL%L-&o;`jKr<)F;dY1>wI{IQrOD!m@)(0I+3ApM}iMj)q@qqJT942}a zUq==SLJ!6YrM~I0-Pdf~Ki&?vMrPvNdmMH}X~Hjs+f;re3vV9`!=>~7f`|kU*>#My z3@5NSZWr;BKA?50C_R-4r0V`zH1AX@C!WUG#s*j5HFKQmC`X}=`Zk<@I2wy%3+U;R z@ATec70mWbh9#fG5nY^V-hXx!&E|-AJKOQfD1Sc@IdVYb&O^in3_CV%6tO8mrvq4yHc+3(I6MHq#5HgWoY2?2+V zycCnj<3BarT^UJo0l8>$sGl1*K8N!=-U%Tw67(|Y2)t0#fn92*Q2+53%8H~iC+lmW zkaZdMbT(6EZUIg89?m?8yD?+P9-Y)rp>gR7cK;!%OEj!QrXxjPn$RR%2Zu;+H)65AVZ;*Y~1Y-%PrGg7pm}WwCn)bKH%u z#+&BvaAepz&Z@l{UGK4c`SJtUv8xzc*mp<6FA;jT;Tev+W`u{hQrO{jlKjqOQrp@u z5O+rwi(h8rQ9jf4j;v0oQ^@k+bkxF@J|>m1 zoZx%(?Ej8SF6Ka(gek6gsfkxe1a+1kh6zd5*t12GgcB<1_x3@|=`{kYGv&B_LNje$ zo=e%+-RWo4CEPLZDiq(DOf79jv?D(iE5A*l7v_N!{;mlXyFEx`+j;q2+_xKmArRQW$BNU=sG|18kiD26k&GSTf^FjnZkp}g*P3O=`nGRF?1 zf-sg9p86f_+2?IMyWx26FF?VSb2M!KEfil7gpQR5NGALM_!OyPQ+EYs?wC&1>`q>7 zxF4-0CsN#`dl*(!&kZe|K~0~T|G;)7Nvmg~Sla~DVO_rl-*UWT-hg7&%%hbVhvWS1 z(ZF*It*J;xy92-A{KF8e%a{dEPs)SYo!caKeH0z@dkA{}p5n!>B<5bVC9{TljQt^v z^E#%%2++ci1^;mxSRhQ;@dh=oPo)o`rgXnFf*ZDQCIvr}=No(~u)nc`B;XebHk80M zHA~`dGjFf|aZI{)oC;^RQ&zn^Hmgh`ZIdme-c<_Duhw9Uj}(14DF!KZ(cHzpXvXC< zgl#rjG}_jePJF#UMFo@aj;R?e5j#p@2|hIL$tC6}Py~y!8n{X>3)hT^p|u-F;`cCn z7`WezyMDDnmE?WqH7P=csumodVnVCRTd3b`9q%Y1$#-O5#1=z;jEZZ(DW>`~YWoUu z(G{WdYbtRG%MtW_n!>-`7lS*-DbOLtQ~uG9!-G^)gd?SfaK)a9c>i^e4U zZ<->;CFMeiWjwn#4&#O?6>(*QjWo^9gcjK6#)i&n_)re%3v%u_j=tMYX~Uzr!E zEpXs0tqmyNLYF_&l|!!+uT!$J81-G(qn%%iF#BC7-r6t^O?O;H^JUg_?w1G#$uC8j zSGAC@U`p>@67XM(4a{YtVJ6Ew%YB z#KℑCPcJdi^OyzXl!dO63$078^1k>;FWFNxN|CIL5y%If!fYuR{0{ zGvR#wD(u3s*6|`cMTRl-j7~u!|7pMF5WySP2#RzR9}~j8U`&8!2Gyd zd)Lszk<%&lwlzlFY_hOvlZ3+L$SwXi} z9p-M=q`g7fG@^PuNJZ(P(Bv5UFOlb`skqVQo(x*8%yz9#`B>mkfA zN`=g5XSA_>0=N2^Gg;Y^3>J*yBV}~xUU(PAe2&94!3h{MvKXd8DE#OB70p{ll6TB{ zUaIpji8))4t9%01HHtyHiw$&_A41cxYdAWSr(e&gsr(N!+H6SnuPB`v+X|02 z9T!%2E0f)vcvR12zVkElspMug;vDw*kuAp?E#E1xaR#5glc%6{+vsGzA}Wo1D_mc= zOxScg8=b@k=*rtMeAGEZepUHDoTHLUFTy^f&NVHvIlY)=|MckR+3g?*kVYfsD4i>R zi%Y%Q$4TALgKNI`@vWsMk|D=s3~gZZ=rR%!Jx@(9rD?=>Ihr<{?KQ%?@qA%AW|!)r z`eh~hsAa&I4mxCzvWgx^yrFxoMU=)ej(<+pqQPHdj0u}bAvUK;|I1REv(N&~yE36O zqJ~?0`#jkgpQnK%({Y_bELWSLho*&*6!nMgoL!2jq2m$tz5K{Ma~?(e-(}I(HN&aw zbTQq(dYf9DMKG_#8+TqGM6v(MATrq;uYHr{FV=nHDsM&L=?8(-;h=|yL@!~;*8n`H zRDyx!lj)$c8{Qk&h3}1}N$D}+^RjJ_%JRXb*DvEgWdV9gSkcRvG&~a!3Bup9bn5;P zqOlseoLfvY=V)TGeK>wjGC)DrHE4XJ2$yTsaHJ%WewP-uwnspCycOe&tr5N-C5<1h zT*r@@hF~D~0(#h-KCAl(WIQmS@hs0Q`qu%2U+Pg$1lwoDY=+T;4RE(^JkA|wgt|88 zNGztCCd?*snl_0v7RTh^<57q3vg0MjOG&}B-!4FH2mYlIhxPKa>kS2oEJVJ^{ z2qm(UvNAHVl6Gki4H5~VBub(oMK(p0Q7K!c@^M~sypU?B$_c`zP>xH_Z zh!Y3NGvB{c&?~(W^X|yeg#0}Goav8;XI#UTAAQhXbrfwrE=_;>D$~IEQ4sm!847N_ z=&MjDYEQ5dJEe|eEYHYJS<*n0f_VR4GYMWs^`Pz|H#%oKm)0uS(p8@QHJhVPC13K$ z(XpIozyq-F{%bH*V-Q)&&O^xy-ykD}XJ>Yx@b;JvPE6iHvEN3LVhukZm#ibL(Hdyi zT*=Bt$k{B(rbu8lYkMJd(DondLLx%oG(0u_X*A3CE?+zWP3#JBqNOn)DYtI5Ka=nN`;cZ*5qyyQL!WtySa4I__BgUsEDXlPK3i)?SQYut%* z+v6FVlxUaDzTaM-& z@I^^8O;PH^UNK^Tj#yf75^K-?;!K4ybeMDu945$7T2UWOZVPAexpRfP`(tqXv2#$j#|9qu8sHkG^?bJ$ z%dWaz785{2^y}(FA8Q7P=jDS**)y%UDBx99OZay|hnZSwpmo|xXkI(ayghr6ICbY# z+Gpd1$GktH(Kac36C(>>z=@=X6=6F+H`!gP;XVy7IIWOru4+G!c6h0yf71gt?BWiZ z{xS|d5~?B3unZwM0a8<=1f~DF;LDj6tN{-S*KVuRtUF=2eDDLN$(f%wRO(=m_a(vB zQKU%W82p=bL$Fv{3gl!LElX=21fxKzZP0Nak$Sfn%I-b<`YaEwU48U+3cX+S9yapoG7MTH{#x zuW;*RGZ^+COR@i@VTz1oN&9i=y3=4~$-Rp~{c+y;G#t1(hdXGM zaZ~PLR^EJvZLA8!!Yvoz_0}?B|LIX;T-{LezpxFI=6gZrO*I;MU^e}JR0TtpZH14{ z!Gg&57hjuJV-I%&TEDiTe^Q)_c~Ox&rjx;V#CBmx)kJouToz=To7ZWrG@IIW0YUW;G)LSPn#(qeAr;>!ACUwhrd<&ZEWiNFiiq1RIcDVU|?Z2v340F`rkrgk!P> z)a+t}+a%7z{zHy9dUY{duq%f329?7f{Wv)7WF@?AI!jNozhJ4d9d10QCQcs`LGFFs z(S76_OensDW_1xTe~doueWHmg!v%KLG?Z$8TtqkJr!XA%&?Sn+pyh*MAMZ6ualVYQ z@+;7I>dZMPF2aETLqRGs6vl3eWuCcz%=)~QMo*szc5#`O`HsEEM3;}}=;`EouG9z(;n%C+US^IcKddwRE7F38ipsKpqnVo6$Q}1?)Z@ zVB19*T0U5ko*L+4nA|gntGdjRDn^*?&lwC}qu!Y3+t!$leKB@IwEqP$-NstH6;y)z zUyeYBBum!oBo9fOy#?(@jpL_T@?P}aL!kFd5@%X}1Zk>9L1cgkZf`vF%Rl%c_y+Bp9z z@8+w70Bhq;C`ToD_`HYxsxAYWhyyrlup-^x@ee#FH?vm*$Xz@f7u^}Ri^XPZc?GaK` zb<>OvF20DptCP{)EMACs?GFB5&tRSB(eY2*EZ||EK-BN(WUNseuUSn;chdx#Ne$$C zw22lL{YLu@M{u9o0@fyzNX1Kr(c2S-fLqtW<0X=G`C$w>eRiWkb&E)Gi#@ivR^YH{ z8>rPZhaC3El0}Lm4K=@yb$j>I%xXXEJ)Og@J`ix3{$Pq0%4mj06B^d5@Z7K<f*MjQr5F!1V$|HMZ;n_@s?91eOg|DR+-1JY1&+3VLWrWLm3~1YGTytTi9?> zQw*NCNgTfFGhMu%1sOMQ;Juy$nDAT{L$|1c}i{Pz!h*C?PjKfey( zpUDL;W>bQt1e!*j5yEerW7BQtvyAF?`1_w6O)Q*>@1IXcbv;$+E-%Jv?zyttn1iRD zcH^T{L#V&vZO(WP0;~}^V``pwVIk*$W=Ye$x4Y@!x})s%H7#N8*;+wi@d~>AqY7R5 zOfPrPF!n`aEI#uzBd6b-RH+!o-4>^@`+70cmQ#X+ohf)INtwkPUJ`z0slf4b4ycyh zi8i_?@rf4SMX&UvlOES7wA~sH=ue{z3Z{%#3(+rP6lve6!Z8tw6x!@e`{ay?_oeZo z3HPcG$wA=vF<<#B)a!%%?0X8gO&$tPJ%_RI;6GMh-7dJ?bYMHKX5-$`6Y%SuB-);T zLvZh3D+I0_gz6{fvfPnDtZL>pOiJf>lHcXajz7u>t!#w(c91M@uV4AiK!UeOhZ27p$=FO?u<~sh1Sk$>&%$ENyFZdm&pEpWy zw|oS-wj|L`Peb#7n2k)?tPZLbE5K8-ADvn(A$BeafzF`GsBR@m)gD?nT2%%$-45gS z5=r4rjEn%2@>tpaYi9nlEtynnAQp9h!i(iQu;J!@kZRFmvt1-%wweZ*S-CMg7YA_q zrbx|u{t0f|S_RYd=WyivQ7nGIU*TbdHQb4t%PP7X*u&fntYWJ*WcB~cRw--KffYAF zqecm5hU#MG?LeU^q6?y1chkIkcZEpbQi$g4rnJ0Rmng551j`@POkH4_bawPm4S!F&s#Ls-d z=pXQ=YA(Hf&om8YU3^Fs)QLjLW?oBUcyS$g;kKMZ|A0|?ceK>78>r1+U z>EM)d8619dN9>Es+z~ULoab|w&y|n#BUn>3Gv7rWMh~ghtee(I2UAtkCFo6`gln#< z;*d%Uu%7T3_j4;%%SRtG-986}hNnB=xkNIpz7QoY9iTWt>rbet9@lKKa*OyahYoidiPtL4*_Cfgh>;%>RXD?EI zFJ06*%dBMk&~U9$H2ThA>`-^Z>$a(QCXu^AjOE1YvG>Keh8lBd`DGrmLkACg41?T% zDR|OWnI`Q%LNPVk;#tp1u^}@6lV1gs#-&X1&)X{8Tv!T46R+XMM?L0Vnr7zjWbC0O z?woMt+ql0P#&itCBc2P`w2gz= z(44VQ9bN;iDqfgh>Wj?}`@wJLNVEzZNU39@kTyi(3i%y0NNF1UE84|68~)&2nUidz z)dAM>YZ@-BAIvP9Rba|}e>5IqPsbd-(xb8hs(083C3iJp@qqdlNRb`rJ5GqdqKE5>SK;#E+*=$znp~b4Q0aqPn06)= z>tqgLV}CvT_BIKhh~;>0>@hgpa+Iy;Jp{vlwy~8jdtlgWGwL(Mk1p$Hu<--9|1C_G z7A?LZtY0Jv`@GkKY_2}0$TqOj*W1j~+OFXC*G8!RyDxZr-V9;4+t|96Sjbjmm^uF< z$hyvCuE{p^@>V8(h(6Exz!n&BDG;LfY+?hShYQjEi<$gwOBQi$Aan<;#?z9i_{{Ye z3^q9q8Ob#)tKV}WXYCQD|4I?4)}N+FG~)J$5-|2b4r)2_e=|sq%=GM7+{A@&NJ+#? z4HwNPwp|pKM{*9<%>(2f?@O2Nuc0fyX7RbyON?70uGqnlCvXc7eJX0rY5a4Mh1R})Bw_UmH}v;sS`%DQYUeN30OWT0c`%!B<@4Vu+8| zMZ@r{X0}@`lr8hFM{D&El$n2lM@&vR)C|i9z0B~V&Heu ze0}6jjChmA{wk#~r5nbWYxfm0CZ|Jz$y-=5ZU|0#DB$v}36wn86Xkb$3$sElQIdOK z`Z8t{Ui=wNM!ww3)s=y7!i{jk#Stjq$C&My$mh7u^C{wWBX>+Jfp&#Q@Yrz-?k`B+Wh7DI3+EX zl_-m0nbR=`Zs4AOCfIr9KZy5`$7|cCu(TVcux-&Z=)Qji-)wB;Tv%C(j9drvP8+}= zhe#-%GmTy7$^^IF4e;JWQq_ zOOn^@6u7rC9^_i2L6djb)*B?^O1ICTdpeLwoqk7y{}X9mm>&jCa)j+>1?)}tTBfF{ zPCu16^Zh^~^O*J$Cnu+p|E%j2^E-$&`VN2>`-|aX#Z~sFSO@*brepdSp5^`%P49x$ z@lMJhoODfv?$wPWLm6pUu6+oM7rEkc&O^wIDPms&?=lI=GRT~@maXURPLsbzG|Va{2Y)~0~rO`%{R6$MwcQy>XVpxv#DRlV~UzS`#sEeT(7s=7Ch zO5toXon39JnN%u3_~!GH8IJYuavtFGIVS)H_aGkpZDyKsit^UU3) zibJ&Fz9p@j-d8jbKhbctNZe8IoT4_Hi&Ogi;p}R8QN4}(c;?&EibyQ+sNJGon5%%qb@}*nMhs5eQDW_OK8C7_?-i-=-&QE zyc4>Q;=af8JBB{>F`i0a7B3{3)#GsCuR(Zd$!&<6Gg-WS)ly91ZqC6cD@bQ`552#Y zMrqX#Xztlv6kGh0^Qw`yJmU=MXf@n9!W6B3X`#utK_ofw6RqxhlA4Fyz>{J=PQZnv ziD#(d{&n2tkcV<5%ejO4J9e4B2dnmFI5DIh+Wk2vbU``p`cOmHbZ*f|-g6?w818j6 zq>6!8@Wi$x-0XZ0A{tjg$<~p0@Vx^Vy!r%JYi{C=UFTUvZviZBTS@QYqselAFD7ht zqqoJP_Q=Fyfj zJ$Se{6^d7_zyLjM@YJdltRs@Cqp=phv{rz#-Wu-xjK>%38&w`n5;kiu#;mkt8lj%Y zGj)ehbM`5+^16gq9r@=}-Ujo>Pb4oXO)`31ikl^*Fu>ml$LuyCtt=f}cteM&Pd~w4 z^ceFy6j8^_wRHNMiWvP!in;>M3hMcP@ZsWF7*(l;9-b=rbfN{eYwX4Q*D}cOY8xyp zY6A75KdAD&2(E-4M;+x3coX1<6FkPzm=9*+e7z~+j%I0bUh4`_mi>rg-p}fSoHfXruQTn(VxsMka?+j;1sD>5rmNskij){XfcZ^QG_2uQ9p!CAb{? zilKu(q5)re*35r^{706TG{c=(ytXV%dR9?m{r zsDoy;a%{q@yKJRoHu~FU<4?;gcF)#?z>+ zoV&OE5!Ah!23IR&g%?Mpg}{;za3`7PdMZlb-GCgd`&1wVtu$nn3j1jD6~z#vIF=Q*BC zW=^7k7s7FWvo7|pXu&@<+%xXi3!`7(hVU`#jdPZ6x^yoPuel z_u+U~IvrkHhN?xJ{m%d7j>^R;_aEgt_q%v;@x&(&O#ZEt1x5fn5f8+c5 z1${BTz5w=*slmeWL$av8jv@W>?J4^4e`coauB`k_UMsCME2U#f8 z@{d`Ij>9jzGn z3qNhWNPTns=%ceIetdMCQVV*iCcli*=dUHlXZBQgb|V={@1p4wZ_(3{8oaxjg%hp} zVw&X#*v5I;@F^f4BwlB-_GoE3@E{i*%e2t-?Pk2SstZo))G(*h3HWWU3ErCVAM7@^ zp^|=Q@%DN%Y;+$_BQv6<#*6Z zG7F969EV8!6Zsl@h8Ez8>O>rNVhi3?eMs8&6?E&l5p^#s!(meW$;?2CBwwB(y>5BB z`=|)_d@u$5o5jMNez5{}{Rfl#RfF36Hh6PA5GR z_6sv%X;c|HXf zR2z~>^cCv(d6fF7xU$POtDqyN4&LuHg>ie6SbTN?yShPTLj0{i;;RTzTxi;a!?)Dq z>qKojyg(L*8vv?|-w(Yb0_l2BA{O(Rpk=>gI!I|a)Ylb#xQFYxc$z&t@<_P6*oL%v zhU1{Sd#sa5Rm#$wpM+^?%c;qf8Y;0;HJJ1(TaB9sE z9F!#mS-qDb$nOVR?fw-e+F0?fbTFhWion0m*W>7~7EIH?0W$Bb#-x}4)cNTS2D6KC zY3WuxvFsq$&rU;|C6lq{Phaf0umy*^ZAFPVMLh3ih!Jkdpb<2mlvT3Ot)>V!1^I(w z{!P$#&o1HtM|516`S!BgI2Iac}}N1ac!qadpJ3~6KATrTH0O7L z(V3k6e2!-p?{MFqUJI;^;I6&h$8h;0O=^iR#|>+4p;Y=aP?)bwxpUnyvE&lI&f}#3 ztyp@s?ih>S^ad{cHy`yPI>F_P3hyx$aL(8UHn3CypSrdKJG_{DE^s#9vKp-MRzoY^ ze+b^!%X0OS@#*6l>^x%4S#tfrKVTkC+4>uAEsv#;1A5fsvj^H#9^>#8EtnM(k1~yW zFkp@mt=xDV_c|P)?TNdoX^JN8Y!72{&w{XXc{&CqNYg9NU3holWysj5B!=`ILKWg0 zXgocFG~emcl40~gXr7M$Ji1c4_2mcv24Ok&OHdgH*dF)TU#_goQ>I=lLLt!GibsW zKYk|LNeXW(X;Fa_HOb0g_$FPt^)?n%$^*dou8Fw&yry`|MMd1=I1lBGp5Pz{71Cam z0=92jV70?xY@BrF6Z&pR%mar7O=c9wpp{Kw-ZxRiteoJ$7bIJCJ5}7Q>!ryo1fWsGEe6_$7 zyTBf2e>SD$@1|7RzM5LUcz+7sta1P~o}0fN9gRQPejKc* zMxKcS$v`rc%GII-HCltebd_kp<>Mf~={($7cN<~`JBgRy{UK?}#gh**;riGCI3%V& z4V#;QKZEw5;RgP_e_%p4-XzjV+k0far-r4)|6|G)BiJa(Jb1(=Ogg=NyYw_ze7UN8}(>Xu5$JWsNxEjyh&UY@h{QGNL!<%^%e>qIW-r(9hUm^tW*d zo#T!so8CpRZoxPVl$StowT7E-HN={y(WGyejBi7qp!b*?FnjiQfg;x8f*+wUEw7M? zoe@}Cq5>PWON7#VEB>s{MS~qzaFO+9>|H(tZ;d+)7PI^DIZq>v`u>tqT~g_IHg_-W zt%ROi>-pWlm>ElGv%adCX!-LqE>*b6_et_}zpWZX zr#m0h=;fsl8?q1)6!_3Aq?-4fDnf zo~_xac$L2XJwTc|)_B%fo*ljEiW6TM(1`;MWL*-1w@)_U`>trLxf%y)yn{85Kd&zR z8jC7_4)N~UA138f3){v&N7a_S;BOl!Ebiw|dROkFg0mdjj8(*;+O_P}_j>5{_zxGG ztcU#dH_V%+kD(Q7-SMHHFMOma)ZO}>=T(UC)X2?9mKp{>K%2gGxbCPVnSN+RNmV)ekd%vN zQBvYCg<&+j&;yVDa>pSS7BC^=r;sr83Eo}1pR#gG=;mKlarRde3{R?J1GdejC%&=Z zoOl+>z7)V+!@l&$tOm!96`%Ta_CI}|ko`O^D4kTM$ zN+EUr^l8*c8u#lK?s6IrPiYcaRc@!V_3J1k{V4tlIRwil8w({??@?{w5PaZtgMHWJ zIf-YxndGJ-q1EIv$w-eAi#*nd!O3^gtWgpd4-8?=Z@1u%!gDnF@mvbKvzc_dW{ApF z{l&+7?~!TOHmZD33ikq!(xQeEERCsv2Q^d3({n1B)F6%1+K4h5Utn*dACvkwjW!o< z#5%31?5EaZag$!9__4P{4AbYkK>bscIsPr`w*Nx!7wd3RWIeNtKZ-sFRCsQ63BB=H zKz(le(&c4`(5HDTDO~;p?W00LO0g1AtrKrtKaL%W{mF-CG`s&~2)mLOQNsCl*gAW% zut=qZg1$MuW?J#4&$*sAJkjbQq~i12)8v za2-Ai*V(yC`=rzC@!D7Q2bHEeS41I@$F6pG$_}m+(nmh^sV18cjp8?Yq&^1 zC3VDO%F(!HtPw4`y^N0ScA=!v-_iB@9sE0O53K3^58M;2QBJQLub)3iRZ`)k5%>>3 zehokwGNFp;{&;eNB`x*;jjJ42@ZZ%bvg{hp-N5~D<=9RbWBL$R^q)(C3lOIs>xRtJ zsr2`4BvtJ?M5^1YY2X)G8eZEEr|Zua{WE^j>dERfQDHXb#6E?)4`<;K^+S+XE+Q$%JtwL-4XDx_#}H=nMHe|H_}k24$gP56_t!6 z#4dhTbL4Y&eLv1AyS#&j^lYM$Gg@)u*aCc?T0q_3lQCe@L(E&8hXwh32DLo|zPXH| z9o?&`CEc1%M3154Z(^}CI~EK_-(V(LeJS$c9hy9?l-B-ELyOjUVHNif#!ah&qdzy| zl;iE#rJ6?Z4*z%0&m{YmJve#iIQr<*OgT{&)G=TxuDUuxJYSnkFP}-{hi4ba{K9)u z9p^z~)x+`j+{y4Mccytr-(b4*b_RK;-G>WX#^JJ+7W6FeAYGcK1RFHEF|T^Cc;Qzf z8eTa<7Dl7QRU3xDETa$_)aot{&TAsO&0|H2YJJhTP=V$iub_Fl(&C;!uJlhv9(^Lh zVfBLdxMtWYl!qEf^wq!=3k$N;D56Uy@2S1Si|X|cfq|Z!xb*uSDllosKgF$>AZt%u zsgZn_<0&Rhze`uEV$oxOE0vyGhkMinc*@UiU;4{YZO2<^|X#2o})Yt4bohh0r+60Xi->Qtl1v@KfL(LOX3Y-CsEpx$mPk+*T z@Cx|%}&P8-*RGtW+^x#xIh;5h$TP3q}du{~{Sy3cba8<|_G48FWdw6otH zigJylr8&3o*zPJ^I3pfs>Gi{?rBW=r-)S7OxeKvDnwBr*`-OTV&T|ci^cBvya(6Or zcvXdA2L#H?QzpC4q3F7242<=e3u9K;fNq&Jjj{Pg4wkxN*2e~@9T7~|JYz|y_Ji+5 zTOn~rJZxHEjdTBAgs?BaA#YYNZk`dw{lO#Qxmn2No#4 zg^+_r?EQ&dta@<-e3I-VB)6;M=nrP3pe9c-2aPE9jUp|sDMp><^^~w&pUV1977E`u z!lY;CVUy8Ee0|WLBK)RsZ@L8?tGG`GPlVB(DXtV#ok)`t2T`l04kg-mpw70vs5)u~ z?@YdjMkP|~N6^)rbB8Mo1V?`uel z_U^*+$fvaN>_^g_KZxH`_Mll_EtbBkz!Q=xuwuYNTxXeu#XdDGAlOK_f)evet zs6>7S2Qc1x1u`vf7L1|r`D`|5o3EhyULUIb)0aM~Xu_jUhp~F?KYSHxBo6E+D-M0! zfo6B+(rS5oviVoZ=Lf&Q)9EL+aE8v(ZG69Vss{E~t!Lu#kC-0T%5%^0=+^iE*BEQt3QW@inVCOc^%un^rIJYd=9vwKkJ>{ zjawYe=-C+!lD3e-zq8dra@1Qmx_FE@C}ffNs^A#S)aOjAYzaKnvxvPP9*U9Sy!&%0 zm~#Ls>F@N#;@%a8V(2YjP#A56)$;jZaORB_O^ehb zxvdSDlQxU?-B%~Y%{eGHXEZr|JB!K>RKWHg=gM>Dm~7=7+>|7XeMCoaoxh1*wN=sS z`2)pI4WZbkS%C%8p*U(|601GX7ko*Ysa9WxqMmPT!-pQu0$Gj;J$uM2J(W(YKf}&b zFEBMlAJ3?6f)C#eDPUVSdi1|c?VL|HN5hT|#kfGsU4P+Nvw@)a>@1zlM)G-74@Q~T`1_`n zc-VG4oyhg2uvkBm>bU~w;STPvCWFFcU5vk&gkReHak_U4dmZ?Yby!|tM-p97k?(f4 zPMFN%{u>LaQ@IbNLl(y;O~hM;3b{Zmq@6rjF zem;t_Uqz6i+dB40DgumULvhfciFo{QI(*k?g7p6P(K|1jDo+`ayU`Q6xQeq!P7T2? z_oh()aNa*M-GML0%|owDNi2&}fXd!nN-6TE-Y2EFZb%pz@Y$<%I`2R8y+zNiLG(zw z24)l}Q&;&7oYSU+O*5Cl`|d6H!J-3gueIQ|VLMT0Q5Q?HsDkgZb7}iqZJJij-z%TB zp}_SE%dadYFU`F;{Js=SuvVhFz_S>p7mXGG|?lrU` z(@S`Ms2SF89fh;`etL7qH2nEF5$`Mxz{W|Fz*S!nAARP`qlsrAWMngp%DE0RJg>ow zwzcTuI26XIN0X_(tms=a%iL;RKl24Ex57K&DB9TU!z~62z-goBK-z5gOl z-!JYY<2Q%tnuj|*lD|OJ_WbX?Txt7ZXDm;2qY)eJ>Fd-Wc(PC(s%+F)f4L}B?t1~x zcFEyE+f*9#&Kq=>$Af$HP3#zahX$VQpzc{au`qWrRNWC-Tz)^nKg=7x2Ztk+X|arV zX*Ahpn|L*p_b|IhlXG)E_V`txPygYxkaNmA$Bw55PQxi{`9-{8V2x?(`r)R9wov<$ zb2{r!LS>VZ5dO{%8`io~T}3AiTfdGZ&W)zX;pTY#ZVd$N$`pE9y@bZ4Z4jkZhl%`5 zvzNa=mOXz+X;a?Q-Qm)wlw#K9)%pQz>y{Mu_m}LUcK}M zUM4+2){=ml7b{76;S{kT^&b^y>!6}tGkBYOFo_{qKp%!;3xEF%4BU#7hniF1;CCo@ zM1_{+{D(6Rmx09fm8fJ^4L7r*_*v%>{;s`+Ci)g+vtb{4yQjju#64jA^gMf#v4!2> zj^(E<<8eT&JUQjrq3o6(2$b(jlU-I*{hpcF5pRR4J>Kw|`}Yb8+~7>#o#3Yyg|T(h zXpQM2+Fw5mP2~?@*!Y2{-XuYrVv^B*zBW!OT!bqYj$p<6-yNNf^c1dB*ONI5K-7I_{ZBM-8RvYGemfKl5EMYt;k^ zs}Q`v`FIv?t>Cj&k@;}?AnEX!-J`FNGxs& zKZ?fJ`MbD#7tdUmb52h^%+#+GM%^)EUg7;{Ti{&ub^i?cs+u^mXagxLs*s+023}Cz z%9$iqXc&?UosGWOwrDBumLBCiX$f)uuR=PsFAJl1mfr2^|9AR_=%w>ks_W#l>Gua= zr&JSO9ls6#LDx(eRtD|x{DI0mZ)@A{I<8`J z!dl)Z^t26SgDwP8;WiiAuU~*p!6T{gT>{4TPr=64P%!gZ1&xN^p>%TtR-f5GZEgXy z)k+43%5%1>3`gW@NuaLfSa_2thb|>H6t0;{Ccn9}u6sCjg?m#o-&X|-oJY8E9`^R# zf)AGFL&?RhxU2jZD%g&uXBo3-oNWls)Efs=aS9||)y4;BW^tasGM!1TfV!bUyq~xQ zE+sEuo43rt!M>kalf_;35r3m<#2PSakY_Va@y=|YUHCe6D6ae#Oee)-SnV3d&&XZ4 zPSTj_!p4y0(m~v}ypF%e#=$PL4i`Cja!4aAhK1C{qraP&ky@Q9iKE!LV$^&w{#KQMy>ntj>k^%rrkqZL{i z>0q2P?{&w#!57e~*n z2dAS!P;kvgbnpLy-p3e_CmTl32L@p2LtEyoH3^@LS%)KJ9>bcPD7cg6Cb$ijX0=h7 zbY*gNiqWxM0shGP^jC zYiWPL_s=;nZ1Wkmbf2_1T_S}HGRv?vFaSH8E8y(}?y~$U1)qK@Qu^f(N}Y2UPx~ce z+}X|eyq&Yrc6dWh&~;|NgZsI8|E|y6Kz1X9=l8Cp;^E2}R9@dK z`)`0y$0smgKF>@wjpE0-a#mqw}KRAdiKQz$GKo!wr)j~XEz;JT^ z-Q*EzgMaT2hLK-$!SRO#`sb&@$?LP=o{u)U>pZ~AYaDRpfa`!1k7jRf;Hxq@REo}J zKX*BU^MnaBFPgjjQjiulHiM<=5H`lk3E1GZ&}bJWxYwq!nGduv$aW(nyZr~F_Fg8z z{wsNWGN8ubdz`bPOo1zYpk`qg<#cQoRa-v{H=uE8)-W;iY5y_b*Dd$9gl z8@h~0$FB|N$p;Pb&5~*`3JGA&Y!pd8--1iLKS1c^CY0B`%qoWEu-n!<;L_jwZ1BB_ zY>tH{X_?%>7q3TR^x(eau%rjAmfF%~?suG#U=2Rg&%%ohrevY&M=vhdg3nu1aq`K-BL7yGfZ3ye1}!-AmUs50=VXs~gs=%D85UB1K0#C~Nr(LR)O&3ZCG z=fOX;aomh2K!q$4f*{@eIdo-@fLlBdKlS)z>iD(`dwy5pZ@XC1t}f(!Xa{VaF^1NT zGo;SF`?1$zF4cPYQni0G?rPAXA9JPXjvJUgc9f&7iKppZ*I9a@`3Bn_%aFWl zFyZiQ?$oeD&mWN(o0p4G<4@3;ZN>CiY9iV04HksX>jU&?A@6hs%2C-a8B*I(3+F0-V21xrDt@{j!Y)m~7ve8G=$e3kjJV6YOrD|? zR+DdF1x{(P!Ow$q#A){N6hFV1Gvjhlt+j*qq_t=X-*rEzH=%b!qtIE+4X&MBgjQL` zq!9LncV!A$)82Kk4A#+S9L@O{_u=TaQJ6BR7X7{n^zYhweDLQaOi@b#i3&HAPhZD8 zJLlu=T`ru>*9JZ>PlDyz^Vq6jLpR&hX-}XMt!iz<^~SB(d%iDQbvhmVT$%ow?Zeh##0tvZR;GN0C$IG(7bt ziSTYECF>=T^ja^-w~z#ljV`#{?=7BtE=RLRsnT}NHY@AQ5Pbg@!;Ce0^lp7CNj_8) zk4q?E+!Pg(xXZt1Cj@$-^aRT;o`*FK;V9jE0Z)d0hL@YZ&AY9gvzDPt; zIlm)s`d*ClO#G^9c-rCE2Txd zxF)gd0PCD76@7*}tK$Lqor&=~ihuyutO!ECcgKarrV{K}P#7%UK9;T0{9yj2wgKZG4a2sR!O-2j2!swxbSi5{ ztv@<2(sd^bdm$n4fgTK-lgxTYA7DNP_c8WhD6Y*|h+p47#O9x8ai?A;O3W#PGhcGd zl|rk`wa=u`iQRFWC(;3P9u{K{XJ`g4X4oD5gf(6r%>rVzg$4a|#Hk(w$@Pj9HOdXg z*C$`GII}Fl`Emsd^wwuCt{;Va<7Nnxa`y;+^0~sXiXFJr^8>g!8{@R*@!`V7ii^drmqH~%v_ZH!$n<0>yBhAjp%L_SGpUt-G ztHRp-RqTj)20NIoCXU--E&f$LB#!>B&*sip$3A+xu;D}N!B8h3g&8MMad`CqC^`?n z9N#aFEA6Q*4Jk@d8qa;sxhV}~Rzjq{N@$?SDpXV|Erc>!v=JIAk)k3R%4*mO3E7cV ze%J3G;8i`(bzk>6pU?Y^qvw5si9=%G)u@O|l()eC@HQwux`axp1OI+9#)q8>$;aIh6LDIb0X$zio1WI6#*A^z=rL{{zpLseeaKViJx7@G zZ;LYdlzokS)e0#z89E9L+it=PM+p?|4C3B8kA_V(k8!r>T)1y^kQ)=3iISsBXlzjw zNiFOHq5Udw%6kg*P7I3ejJe!$x#gmp-CprHTM+=KH;OzF4o9-n?E50(e{DWuU5I2|0(K^GO$Q|Z} z*121Ve{mAn-r8sJS7r@8J#zu4bkE0oN#dla{+w>C>m`pwJ*e9r040By(DcL|;2b6XLjeW!Bv5toz)N5*LBrwGM&GWaP!(m(Y)wg%oNP z1kaXl=dKz%a>eqYwAB0*E~!a^U9KAie|a7z6-`FP%0lkGiWHoRUj*UlR*dCZjGJOt z<6+|@-sM0#O*__zx5qS_i>b(1V7HpZn6<*Z+Pl4YQT#bN2R-0-ZdB&2M&BignQxd= zfXLh#PIYFt2?E3QswVyY{DZHx9LM73q_?Xn9o-X#11ZpBps( zx~^cJ^**iLkqXZQwaQ2_JQo28{Dr8 zg)rw<3-m8P&8=YXI%0CS@tT&4MaP8<^vepQN22dJ1L4C^@hBGEJs-iWf5NnDPaM{r zSO>=(ro+Y$dU(Sno;#K40anMh;Tf@u5GwqY)4sS8&V5!8I8~OgZuC`1Ne&17K4Dn3 z`v=Yy`Y3p0^9T3F6B2EbM0WbN_sV(Qw6cLAml1^y|DutEMpa z#t}I(QBkAB(if06k@5FgM(U62GBjUs7OnbcW89V&7#YxtcXw;zyaz_Cr*hyBSWGGf* z43TP8GN>(u4c?dW$TU61WSa^{e2r;ocpEHc_2lfKVbN`I1&&XT!B?dP&}8xzeBV^z z(cQV6%8zi|KiLOEzK7t4S84cz`DE?%d%^L%3g7CX#xMTi#6A1+48_Z*Q^At8K#^%& zalSTYCCibiM*`n3yO2+L{}|6dY9y~clc?!tFgMgOSHP(mbDKYp6?l!PN404uDRYvL zplm}hgcl(OSYCt+=Q&Dl_aciQ7eQJ?9sYH7;kvWhIKFTNew=s@y~BSAqR%FR)dj@4 z7xiG#yl1dB`5YGiykMcN#<+AxZ!=DGJVyQ3Lw~PEki$83E}=z^lem?S?;HME2rbFv zG*t}ootX*5{kZ@S@7crBa%JoY*axd$5Z17~TG4B3#=$>>pE4eE*XOLl=b4_AcbKre zvKP5z9dZo%XufULNzUNg3HbkA_ky_?-RiiC4JqpovRyD_pp(k`lA&><99?Ad!@1sF z0&Obd+=NsyO|6HM$vlabzGE@qU9^R9<}kb}V7bG~6CrSwCN1$kOs$z~U_e3*KYlPn zlkZY=r?QY{v_#W8js4U*P8Ho%YRoRP948*Mr;G<&2hZ$JM4bqFn`q_WpE zZg~}w>c_bPlzw}7g~-9)9ChP3a>TH2POL+>)`NOazA{?q0I+_{4<@xrJmQu$ed z<|f0Sr)~{i5zEQWsT6;y&w!fPLF^3l#w>ATZeMEzUTdqsv;qULtNVak8rdu)@wvdZ zVJuny+C{yLvF0gq3VkP5L(r@ch>VrPlK+CZ?=%)ezKBCK>nN?+{*ct0QqX761{WQ? z0WW=7u4VEF{91M$D)ujcqu-9hn(E{DDB%-MxvNCYWsHSf_Z*_`i&M{gI}(@ogmjr+ zP9b9)_9na2(7iso;#)<#D`qlAeJ9RPv!V+2u9~8Lg~T&AP(pzpmH$p7%`uT=c;W=k zc*J^(Zu?-IT^>p?u3%-%W{rO@dNZ0D;f{!1hMg}xc zksSs(=Xx@kv;k*f?B4|$Hlcgv z9Ybe~kKv7~7_R0b>yub~=2T;p@v+em7AD5hiilE@GgY8&Up@K}>OzGY27IF97hLRm z6*-}C{Hx}%{AJdS`MPi&UNc%vdtHR-)2ewi=S?{F>#q~sby-dg%P!#UZZGh1ALI^R zaztbMNwnE^KFWwO&dER&y%fz#?~^2}FUYe)q*zD<+bJ8C%DtPLP!WPFTV-k5 zHDNluIULd-oJD(2S^iJiQC_L5j$c!}pNZ+2>sMKR+DWU`HKo zcFfVah$=nX>He|n-1Y=xbSbV!tG5doLtzY=d;R6qr|!d?6IwXR^EB2hzD3)`m(Y`k zoFILP4OJa-rcwXRCbPk3DATowcz-~hpR+hCjU2%x=N#PhjfXmmN3b|8k9;31QNVTu ze(l}y^xHpz#(Zd}s-=g>@tHJE&YDIiUGqt_^9?6{^$~aEV-#ezCu6_kLHu>X1LC$R z@!^%pRL=&`)PZi@Bgs)Wiu)qT<7F2q+M+)eZib}e zkf4+@Uv&#`aRkOhDA2r%je>^K4^Y#(9It+2F4gCKSYM$`hTmnLIk8_HAQ2s!R# zYY`X(YM|PcW<04DL^9q{^w|2e;0wiILd+1F{OLv8@m=U3(t<p zwuMcPHXdSsH)Jl0a~hG@Fv1zP3@L+O2g{%C<*;KyC0G7coYqfvqdd(*FbbSZu@yV1 z?Rg?MZoVy66tRr6o(MPN-X7S=XTZ^Y;Sd?LiAD|;lbcHkUbokv*oh{Tt}p=0R}t-; zH;cH;nV35D9oA|+!o<Nfi1S~M7&NvAO?+C{Yu`hA0 zV*$KgrHQ_c{`fJ!8PWnwLFdUZXx;M1%2<}C9{P>iu7i-cI}Y3P*v~Lw7CMwX#J8J9 z(d!r4EHn6!dL|{%*^{Z5Xh_83V$9Vx3X9iOV4|rV)>(RD`7#N*JNy8vid_ZrA@QK- znuLwJ*73=UBKgR$M$(xU&gpAkz(rRIQFEsrrRd$FsB{aOS>J}|E6nK$`&(3xD8l_> zhiO~Vcv2J@gK{~iv1O7O^&~ZbQ|KZHHEP9)CtNti7Fawt#$*5kw9UmRU<9 zcVYr9*}R@^Y>r?I#v}M5{vAd|)scmb74IT=LyM$ZV7_q^&RyPtt5ZLraJUw!hl|sx z2by#(=reOVKO)sYA^e*95*MEsM^>74Xg%u=?mY4wZ_Jp*E4!ETjn{4Y3`a-QEs-aG zQ)_CQ`h&}n4S|bpEKk*Om>a=7C0~veQTRe>i|heWNO0MKqpEcAf|LTCIp#<=Zbw75 z#2%cc;!VN76!3dh2>uIi!28V)1(9o+SG_75gRWR(^N$Uv`cjl}KG;2i@&5LV@xt!v z6pCZ)=*3GisZC`H_$+Njvvu1@*WHSn&p0`zE>RGsP%jw%R3Rw6FafvUP@w3>L!=j7 zL{*nhaG`Cx(Y@yt?mz!pkPs(FPJ0f}W0zv!=eMK3-C}Z2D7VmJyzchH$Ka|!1}ow( zVybj4rpQcz2%A6*8J@=2a_ShmQHFNfJ;e;?PVUNQ0hZ_0;IA*f&@6lqR{W4a70+H+ z8~7DDgDGS<^%YK7cAFmEJWp}mMKI-N4gP0dhi6=V;+&Xpil23ybV8yTPyP;G>X#K* zZ)~-=ad$qgejv(U{&;UlUQcfyBKO33{- zW6s=U(66Hx2noJk!zLMhoKFIOzv8K33LkZK4N5GqcTU5nZqATVJV}i{b?oY{K zcra!>#He@q8d^BroHyGvg)dXI zCfEE+7!wqQx-vp!G?+t+awMsfShx1zIbQkYA&TEL8Wiiz_{$L$sAM7qAC^ggaz`{; z^-aPBUk+lAs1Eu3&4fQcR|-z=BXFB}7v!BJ_`y}N^mI)%>fgM@?f>46AN3!QZ2{}t zHbu~u;YNbR;iOSlMO73)qR)CMd-EQ!Q@hUHl#Syrou140ZBwQz{$toZ&zVlBoS=2G zchJ${HT=E((fr}xPZ*#72KE2D59NXuh!l#XvpU2_h)sjA{ONf3fEf*qY{2=qzfrKU z6kk&|9d~XKP{{0sbZGk@Zg1s5y89uRcQmQzg0}4=nacUR(XCgQCh-O6qqjwLn=L+i zdqt4Jzv5mjzHc#Pieze~&D$%=@Hd__=byv_*t;r?`ACS*8!jW?xiJPJCq1o6c z8o_JJ7t;FmNtn|bOZMa4IPWG#8rRuLH$AVCN?aD3 zW6AT47Q9P6PSX}z@YA!TVOD)5=|2eN^Zg?+oXvB>KV;Jw#nEV1Zi9vDNAa?#7N3&u zK}HYKxoyo$u)Jj{K6QVCH4-=Qg61?7?hzIYd|F5||LF0H+Aa8y^P$9zsHM}hMd?;W zKL}OLM(fE@w5WTVz$Y~eUz#{lxOf7YjjAWH=(WsOvXFd)6iH0YiBhV>XyN`%AV$K=CnrIh0RbcZ?`t>25qOBscY3>2JPhtpy)lWC?4Y8x>QN2}zgf~sYtPf9c_R>`RZh01C-1BkS1T{AAjey;g-wJ-m@8qP%_ks4r0gUlp zM=ok|RNJjcL6RA)cNmAiO0#fOUn{Q9WA|moG7uVg310^4;OES-G`!7?lw8M9ooqcu z9@NCM+mqnFUMW}^Iikk;>!2C48!@E}zci1dpA&eJllcZAjB7Mp>Bz<9-Qeyx>cem5 z6>>;v$2m(aNke%hWe68zHQTSRc-DZgWyLWh>p%D@CCN=Yr9(>;dA4UBPen>1c$;y0 zP8&7iCCLYPZoyBqREWiXJ!O2&_`?lm#qjsWQn<;^h`P6mq;S~}o!%Zo9hV^H0{I4` z+l;VmXdB!#oPbN8rP0IW{gj|%f>$p$A|E15g(eFzNka@Ws>}s-cRzvTqY+rE>A`J0 zHh?*qJVpCBlR?dYU{_njEgW$kB2SJ*^ZIqrQDKg@-W+~f8jPX)WGOX$1TAybAuAtG z^xxJ5Gd>;!@uD!87idWdAyRz7>J$p)q^UM|3(jQ@1gBMH*f_lrzMBis+V47^aovVr z*=MJzES`08Rzl2hGK;_$zN+O+8a zb83E`L@7H$@#@qzoU-pQDrmN1IlHIUtsH@iL{{OX3^58V>&CPN0Vuq27e00=MPdve zdtXhkkD7*-p-nh(!(aUTU6?B6qEI;d8z#Ni##hpDj5n4?-Dfl@N^Li`dHe=T$LAn^ zt_?k`U!h3kXIT1Z02T+Z49rVyl!3F{;@w?>1seNcm)S)!@%=&X{NpHf_fHs{Siqb~ zKIn6Xv6bdH5*ak$wbxJa)^#@uS|>`5PSxO;oPeT9jL&9<eu!Q{_E=HL6RfMX) zjG#WnW~eCK&sndkgR*ToxPq_6yk%l|(ps2I8sG-dZ z4Bl#B?0wcfY?%%tUA!RpaxC{FtO*OI5ZZD-!SL-?^kZyKt8XVso1aVBlCx>u*&Gyn z)WFH~3>sEwq3D{?bX(1mR6Fv}>aGNhOdUyQ+a}|1AG3S-;Xmk}okr)6MPtg}^XTpO4(;9x zp?zm6%aTY@N7^=2Pf(*%!73CO+XY310yKE!0TZ%jW8KGS95Y{&w6YR$m!B*>jFlyO z14OMv3B22V2Rqbc>D*tUy%CDE(%TSquieHgyGt=~eJAV^s)fc4S5RnV5jMa1fioEQ z(duA29AjML|5~!p=wB>q^dG>s*CC8!coU6BRH0)T^Ae7$z`4TxcuUHNevVBAg?b)! z^BiEYnO&`YyJ;dg|SjzwL4?wTG^BBW%_J`2PvfPuQRWCP)0;#71>RIVg`MkzlgZ2HElGD( z({Z>?b_w4|V{Zb@xH=QB_Nt)fJz1I~bO)oiMq)zSeiS+x#QHK1IoH*G+*YLxWVZY` zy>wVa&s#2oU;96ZJ~JC@H`Ics>MyV^-NNep0-SQZ536>&l8;Uo8QqfPull8-jQ<>T zh!3RI3U6rRzCLhk8-;^iLwIX!1^Q{fK#%@WIGObZ#P^J+OW~Jrime`PI1SV}UxfUZ z9>(5%r!XvS1Zp3<3T2xQfKJaH^u5uF|EiO~VR0BY()}yyZkkEhVL>%{nkX5@oNe&E?Cme5(w}+UujNp)Ka-=`#-c5SU)cB+F5&`L_HgR1+13)r}4K zvO|V8Y7IcBxhdZ7kVd_{eKa**o|nGZN!b%06L{XI)M;J#xHSl)^9ymdK!=t)j-VVV zOIonQf|BPJNvT{|ks6wdsp4ptPTXd&moWnFVy%2M_Difr z)Ap(8=y3|Qd)3hP#{e2fnvy#6LHKcx`2@A(h!J?9zEUKZWAtwPtBpJ>p!gtQ#%Q2Ck$ zt~*r-qt6_LC(T6q_5w<>-NpKgPcZrNMI4R6Xn!!8&C3Kd`sPUR@U^6)Mai;>63WFtOuOc;N2#!lXUVJV&xq|mIu zGt@e9G~`6aav^qsj5BDBeRut^GbalV;8{pnE`r9wSt#A*jUDx~Kzb#@*V;UcidlxT zajo$F>o#m-+`G}47U->T3UdW3S^jt|wtNZ3o~>QjV!aUimDRD;<_w)Y@r0ZgYogHA zZV1?zhdi;>7vUsYUPLCFH&gh^0Nj!K1mhNJQq!q_7jbQ;CD%*-Bz@xp*q8`t!J1CIt7d|- ztV90mR}cCgj-^p=ncrjuOXHlb8NNnQOvm(y4*-r#+>-r|)om zoF|-mdI_~W$D{4UX(-gTn4)f}(9Du&&{JlBXQq$C!PLhf`%Ion4>*#KnjXDUWNefz zZ@9(33!zr?I7ywCqOe9&l-%EecP?aN-k5F}akdT>b)4z0#Z>Z)jfVxMh@wlPQ0SQ@ zdRb`Ux&>$O=zIsVJnl}jm;FW`-Aep-xEJ}V64pIThFKpcF~<5y?Dmz#>_^E|cf^$6 zW8I41R=8nLJ9861smIuRjmSMy#1r3mvKXyR&reoic&ZIGH5_dOe)pAWaqjrnA1rUG9Eq5e-d3 z+FSMsyjsq4S!uTP@w_Dem)h~>&?X8_IK*aCBWTx|Z*(DZC7%(xhjM$jF$e8A{BAOh z0`lzHUeFGoc}igRjZ^sO^$~Kt-wv*OQ=o7BE$BP`2+KDwC(<#dk_LO4Np1LGFde^X zW|B)3yYHMb2bb@QIpNo7XmtEBdlp&nU+PcLoWpK(Tvm<_>)244+adItJpyM*?SpUk zucMX1D*i)oFyB#gk*-ImVoLZwlz4d!tsnK1m;7!jVx0)XSU=kT>>{nrKT0>W_R)OS zJ-x;u*&ADclM+vQ|D9y>&@a$N{>DV`;R>*IXD%3>FXW2jRC(*ZNqjU+ z=Z(cS(2<)9Xwh>k^l~-85a+cJ()j?i_AbSr7sW{Wzb|+-b1F_*T}+==bkN9YKk!HH zRah_E2ctKy#EQyjO1-g>zPx6?6aN)7;FS*&*UWg|OD`zMNfoWK-a!1i-%$O0825iX zg?pD@p(leX+jMt@)U$%?V;?{4Z zZzqdlnX~cwUpxH&zDwmM~ntg*a~V|mQ52^TVqL(Io|O0#VRFr zvMGt7r$vgid9E*n^yblO9Q71=b-HCL- zSe$kL3NWceoc?~%rV&$3@ben>ol;JwHQA|xnvuy!qPywEgADjNRu8W#*MkDf6VB}4 zg3C4}U@Y_4t+VPyhl)rPQ_aVinb%BNye5E>uDZD9GsI-6XQ9g-(!%Py|K{xHsU)uW)>j3XJ8fM2KRz%fNR%3(9B z(^X?Z(zP9@s4|E4f^j50$B8RgngL7y`jEk*EKti8P(j~*>exOC8sgPq)Zk7`o}<7z zl(rZ=^alr;Rzj9+EZ!UAM?HFBFzG*S&~f%XvrX?YVH&TzCaTR+-_?2^~;i9}m0r zOyJ0}E!=CK5q}OC^2>IvWs+q(`In_poZHE&r)rO zJziDMr<=c%NcqD{Trz=q#`;FkpgChQ=|4u(k};Gpw+=eg4ngpW3!LPc#bnPX(oM!W z^8aDX-&T6SpWeQP|Jd9DuVq;OV4xPSRgIu6({@pQq5+L^KZ+TrU1_biHLa_AFL<=# zEDn7!q<8AGD1GE&EIl)e`+Y=+({VqJ3$mF{sG%3b0;bSbl{U=yt^t)R2pq#RIJe`o z$zqE(9XfO#N4B2F!^fCoZx-W~31-oR)Nyp_*jaRNv7_?v$tb$C5$4}Iic1d#qyB&r zx=Hu*_0=7G?8adJk@FRr^B|rK+~d*t#uR*3y#}7aEO=b34#iuqgJP5xx>!CCOw-nd zFKd5ruFWIRQxxHu(`Rl~v7Uu#e~)0Nfjc++qy#w)Ey(q+f!N4MNSJ1Z&oKxG?Kk7~ zry4B7T!Eq=!dbs3iJ!>U4M9vXSs1Wx%IZy7asEFim=Rll4vu z@OMWie&5lCo9s_eS)VM;J+6cmZJY7Qvbi{I+#q~?p--EhD$@|_M2dH9p*dfE;h}yP zG>iL;>eKJT9}z`?oWxIbtkYs=Btco<2@KdRL?U-tchO-qI*e`Q4ipv&)_PjO!5wPU z^KKJOysd-3(-iRn?E|xqDYSdOFx?HQ$Mj$eD$`=SByAgfSIGLXTjO!*?$J1nonJFK z11!m#%2->X^zmLboe0b(XExUhyYUR>Z*<^_cihHDzQ1wB{0Vr;s~NwGG~xrfKX`f9 zM8>;Lr#?ERlwOuMdw;#tBJQgMqecnrZ)J4=WX)r74Zk8mEF zKH>IxA8|`(86|%BMo0d>qA5G<$f@}jH7!l+lgyg+s8E^BsI%<-eu_uk>XBe!6B&v&jJ z?g%*Vf840eE;upm6ec~%BcHVtZ zglFYP!Q{aleDn1(IlkCNcz?PeZ)PDV-7G`DBrTB0--eGX**ix5CD_r=_Bp3+GFDqL z>=yXo48};AG4&~@e8&XyT2EkgxDdrYyan^*rReIpLgth2#JL522GM@e&M<*PcE%g1;!+a;cbnZu=Nh(xlIwHv2Sefz?Rpz(o~5)Y3p-d zflXX!;0G{yE&rtP|8~QG$!E}b_M0;f z`HtJm7m}iWASunV5F8dVLGS&N^oz~Td>a|3;mmHFZNwZkRh2m89zL@p-gyzX z8?T}KNqgzx9aVm^QVNa?y@p$D;xQ#U75BZrfvThWQ2%f)`MDVJI`Iv3>Tt8b`KuEQ z)~e$E|D>6>Fd3sC8xSpCPHLSt{Gx`9e8!g3RA_UAG>gCD!J(V5;@<%z4;=maEpw)t@j*LTlY7TCWR;Qi8Zgkb)3S8muU_e?8 z{az~}*u23TR^;wMgS;5_EIJG0T-9;#eLA^Ui;`8}ZW{C0021VtpyS*`)R=UalV0eL zdP{opjGFNME5U06r+DE`+LgHIPUke)O0r%P3# zvZDak#mi7aR5vJ&kH)|L9$5CU7#0g&;dfV2`V=Nh9gEI`!&O5xAJ5MBsmv4kIEEA? zo6$Y=8N_V&fN639P}utb;$8-#?v)PQoV<@@S=S~cRgpWSQ)5v(C)c7wSQ{F*ws605 zR^x)uCU9Ai%pGV6#M5sbP|&#+?`#Rcr0TJ7Z<7TWLnL$r&S$@85w?K_4^9a{8?ao%My7Bg5*O_8(Y7HDTcsPcLYnOefF_@m9OP49+{oD^8j zTZ0LUOz^Iop@4VH;6l|;f|uJ!EPJeo_Qns;`=lFM>eivl#4T8O>j;3=8rBb;hc2hS za{0}YcsG>sP9vt`1-}6tGnerNuIz>}pWR^9&sNx?Sk37dIpWw+>>Nm2&n-B6oZHa) z5M9|`BJxZct>Ko@x4#^B@qRX^w5N#sE|OyYQKH3s+|Y3@XX7O(QSrkgizlLLff~Qf zP@6x0JDsv+E4VQ)A0xjwow$z%`1jiv67}D}zq_n0SfXGEpPN}HwuZ+}?LRn&?S~hg ze!*2=wZJ@!$%OK^INf`f@sK?89zCg~oO%;p!Bm~UGT|>*zB!%ia}5I}=AAHD6^ucb zKA?TiKI*Jd=GV=!;LmyL@UiNa{GhiKzcYFs2yGL^gA+<|?0+fTZRykaTgsGnI*qdk zwj1CC3X%Bkq8>fJ@59f(_=w+Xb4<`!y_r+c9Y>YU%pK=r!}ufiG-^r~c(L#CTEAw} zogV>8-tACbU_x^QbI|<0E#BE)foDhW7c84J4Vsgg%g8j0YqD9-c^79x?j#%j*jQ8E zXMHao8sUXr2AA;bSrNRtbd(_B%LR+|=_k3Ak@Ltx_7&|7TgLA#oD2sop23E38}U`e zFGwsZ$II#^_(+1e7#1z1VsAR|$9dr0q*`wC(UIsc?9AMaCXAaS!ar8K z#6LWHniMliY3Fo(wo?oeWV_~bYV(@#n`eul>C78J^^hso5x4_pyjX}d`2H!=>tqomf?Cu4hbyR8=-D;F57z@!NP}= z(MDaA7H&L4FMU;c@5k!UA*llX)tYGN>(1%>y%YE_=kl#~WtXDKTenD5{6-K=8#g?O=ZBxQ+w<*c?3!MrGKyj^IH z^-0_D(1%`}xm^RgtQ1h8eir(A`U^%1W^tdjbn)9vWpq366584BIcKY|V19BLmnhr^ z>n6)`CuH7n<-%(*^TI^f@Olmg3U=X##R`!9eI>5uq$qKv0#~mx%+&=fgQxRrx$Qf4 zV(NiE_}XU*o{WpYG>2C>cV9QG@bX8Ib(ioD&lph;*W$sZM$C#R!|jaM^)w(2kJjD- zr+;hk*fo9HI9vgzl{K-#(if%PuZH6KG@Q8k9!8d_lkS;Bx@DI`|NmY}IkrNH`3~%w z{Mo##`w+Z4c^CtRML_p&8t^^xXc=rr5oSuH=kOAxR{CPV*+Sf~^e#3RPobl|0%}^7 zOW%rSz~qOL=qb^Qs&@VG$y=I!J$S|TK=I(7$XIqis_=WoLHzSgkCJq^kZq#_y*|kP z|ILS}eb-jbbW9c$rum|Mnh>6pd;}Kh-(iWzeH7g-j*V}dLE@Mj6|g(i+BP%Vp%DQ= zi^TEtVa5b76GqqQiRe-3h6d_&T=X$F_`M+l2k%!>KteU`Qy9;Geh>|Z{xqTKvq>bD zt;CG6)3INy5N~&|HT?KaNKRV3+pYj}Ru5tKGv zfYqW#oRab)II%T`_3-V;jQfs$)C`gPEWzW}4NURUr6mU*2&y#uIPSF?DhdtbZo^;r zCod822@)W}E*MjvtiX*!zED@30Iz$Nqw*$UvQKJy#q3_8|idgJ87?#^1XAN${3iURwpb_2;0}t6}^sJh5do{xO*@G|AubEi~>hom9-ETO*spLzRq~qDi1}LRtwTwRRlYq zW_A?Z8QxRgBsAo(-}4rY(S_4Tz_raFLs?_S20TGQy=&|S)nUIpcPij0X9 zkK*?$IbWycDDk8NN4Nim>E=FosO17?C(7fLlJ~&%8e@H@B(Avl97mkHg(*kgLaahK zRHom66R$R+hfFJqmK)Oe_Uo({If6S`c#V7VW(|ejEJE|^wp>Lh zxFp^Rj2j{Z`U%fyRo7t>m;8sz{3nnyn^hdSJQgk2%x3wtH#px|7G2zx=`Z6@d&;mm zbl40Us-8tDD;Cg`DitwVAJ-VqxTRVCl;tBrUOm28(wPcL@ka2xG=hSPwqwS4 zeN1s5O;TG%V}-j9bY>@EbelOwi2vc_Y(>z7!jxXgB;!1bgYcEzAE(f6iin;_+r$mY z8rsq!rbxg~e`&$5H~C83i!uDc;clX(LA0s`c{$#SUs@~4_a9__km_X^oNfz!&atGF zBTXZ3C8Cj$B2_QcCdYX(j3-$K(yseKC-Nvvc@{_a%57<`fenrSdIY1+o#*TZdjt}@ znKMgZg*r()1TTlx=)|Ss>ie=1HUo8!jxDqjZQR!@ZR=H9y1=_`#w6MyWF$hHQnn*O? zP(GVa-H}Vl5`?$KzY9vf$qV*0e5G+iQPgiXj*j`{@hAEZ^Fx=`@x9}d2vY^*_Fn|f zm5(Jw%^9@h_yr8Od=G7p9YHO|o4>YN0%JF2LSgK5LDnj59PK!P`>BvGShKB>?>v8i zH?kYgZ-1gdCnStfa4?s9W4r)5Vt#Y6x6-*VvLoA?YOLsEXKJ}3-#zagCG{jx`NffB zd8HYu+^eA3I+1qCSkSV=XYi)#Oj;Av1wHlp+>`aIa40Mfo;L^!QgRY#K<6i8e)fXp z!xy0XVk2mMzs1}Y>>1&A3XP_2g09|6Af3v1mRI(`k)3P@8;58leh!YfpX574;`tlh z6L^D2F}|O9fHU@1_)x+?xXyd^H6U* zl5Z{?!>2yZAb){8gheRe&b}I)_1{`FVE6Gq`ub47d`xMx#CYG{B+6mk2H(L^^!|`8 zHtEIE>82 z@BuaM+xqkP3R|Js{2!PaZbS{QC>$KJqA|IQ`@i@DT<%ThzPvP{`7-zDtlfGND-xkM zA0ts}!YueFKNqumZ*%hL!kn=B5L*56L0$1OT&1f+SMMsLv(X{&l|7078n)ouy`vyl z>@=?S-j4@OPr{0f81$^^SehO|l=fTEg)(2lbiu?Rcj%LmlqUb04ao>4o4C3a)t{GY| z=spD>tG^e#QgIN-ikOh5ofMu^2o_XHiSsjUBzai#o96b*^5dFv$oG37dEZvxkImEM z{oWbzQvJhp;aWC{=G;T|=egAV)s)g7>!5>CJifeRhiU8UQG87d+Fr0l$3ZF9XLli^ z@rJZn_CI{O^8kIaE5?^WUqMb&ik_S+g@!*VT&fl0hGfK0md658`$9C|e=KUN`#?m~ z4CMJiPR!}C;3IS5Wa;GK%_a8Kv*S45wcsT5Qq^=aROX(5uq9 z7?|^gODP$HHv)yA%|973y5FJy#u>bo1n_m5OVBOH8&EA#z@7epoy{lM`=JodV$6i* z%(=Wo?ubLE3--FFb{Jj_o8%E%SNv+V46A9lTOG*S)du!&vOrlw3E68|M1^!(rN{R19QQTXcbv+8{F7wQM zSYpi0ZTbjv7tWx8CDZ7H=ua#T$)Hg&E;P{ThS^*stqs`7c+&IGNsl?67M-$BKdKAiE^CA1`H6yGJ824-qL_+WM?1f|>rV~JXLoY4Y1YHp!!ialw}HKeWE zVlb*Xp9>iB6}&i-48}*i@zL>@+}P*$sHiNL794WHtxvUKxyUUT(Y*~*UuWT>IG&to(EowF0$<|X3(hmS#Lv<=FKwqSE@BaRAX?_C2G z*i*BYrWePN<{ep*kdh{Ujj>Sfng?N<(%EPHg~dZl56-Gp8|{X}(dvJkooQT6U)0B& zCsCRyQ=yWC?%8Ya6RE$DR7g|^MWzr%p(2{2XpjbpW~oTSJ^MtaN{C28WJ;M*p=2u0 zd0s#7p8NUq=Dxb0)4k`cz1DAizoUd5#PJ>QGRGV%Etmx9n=U}|#D3=Ck#=~~KOA?S zD~E3d&cg3Lf)SrepWb3A@F*vY2d&e)nW(+lkp3hF|2yu4C+{`KimW8qUH>XF75GIX zR|(wWom(+HxeeLZ6NUOV1Ae{v1Cv!x!o{%`$k<#3pYZVZxwgaR%i$2W?I-hgz!3C>owD7}7NY4cpU}>MF!Xg{JZkX^5N4~A zc(}l@?z(>(+q)>iuivRmXG$%4=@W&H`rTkcUR?o?b`u;>u}a{XVk|d89X=2Gin`nv z<9!MxQ1kT-Dn2NUO+6k#xywCdraK!=cyj`MQ!>Is{Yi||%y<-HyAEmgPepswp2NM1 zRe1g1%h*HiG4@#yh2`&6BIfEiY&l^aemT+*d)#M*d6hAWQXGN5+3W{qXFb~f;0kgu zOoYgHb&xcBC5gOOgYPD|3D;vOC|t{f8`7FculyP4?X$s;R|MkB%I|o0rY%l`2DI@_ zBA)*u8@x+Ym>=O*Oyf=ASz)At{R9rfyd*D}W?>8;lAl2C>N=?0`5nwJA44t{zRbzV z)#la?Vy3b~8Me%P39V6^QBclSq|q}G{xh7%{0S(9%1_R?c(D_B3NF@leNN`rH++U= z*-7wXMK1GV+7FocwFj9T9)mJPvN-d^6w>`Iii~fT1f3bebK=kq^nK$;y?0Z1)8$QNEQlzBozh4aD&A*F+pCGKB_V7r~#VLeJ+))IU25 z@{qr3C*WAk&Uxlj={ug!&5a{{nI@qm!0{lI9PkVkJUY$5rBKZHJ;h$QU-kiFDJ zCPr}Q6u2HiyUeyS;ulk~w&DgbP2y2zsRA~s(*uv$mCW(=)tH^+gxk0F!`n^5{hM0i z4+7&>?$K^&?{|RG+I>vShI6E{Fq13A+~rHi1DuHaQB%Xtk}OAoHy-8Gp+@I z$RQG1d_SQ*J{QrVtS+>=#Tk0Wb)spZr_h*%5@`42FQ_uPSu{U!8FOe>I5e5)3cM3_ z7`s0LsSm$~ijNMVsnd@L{oruu`{aafw;ALA-p6Czdu`Bv&kTq4s0n_8DzJ_fc0YB@ zVa|V25Vv?PR+2En+BSti9F_2fN$cS&6TzTQfM?tphJC9K30ye^LC`l7rTlJVHZ70C zRYz9fjv_M%ia3nZA2L|?Y1nu?bCNDD zdsYj}?rs-qH7)eSM;3>4Yh!QyCZIIgOP z_+v6yXVD!{SW*B%t76bq&)JM@@(J+ymkS*_#jt+fNi5Yij2s_+8y7E8g*%UPA?ej5 zoNE{kY%xIk;8<``6y}W&uYt9eJ}3xf{^qz$7*jGIEKU`{AK`qs-gXCmclJ5V`O}K@ z{TIO1hhy>1s#-7>{M)NHVw4%iBab)hfNOn;JfZ;6<~1CiC8SDH%P^VB z;BSwuL0fQQBxQvl>(WZ|Lrp`%4wnv|7P%L1-J?is+%*Z^W`UD7Yh#0(IWSFe2G&;V z1;t&zkx|1%;fLXvRmABw+E8%6$0 ztcSL_Dq!;I1oQHv6jVjMMbAn$K)F+J7k$w~(bH#R^m9C3oplSI+5|&yuQn3E z!l-Z9J2;qZh%1-X!-J6?cutWDcHglKrn*<6Uq=+NzTQpju&feXME7CngE5Xi*N(dL zQ()Sg5UlJ)cxG)hj*>r(?RLd8gVOr&bEi6%iLJ#q&N~u`pkri0 zeuap%K+NV}qr*;;KJs8NVkJWN8% zq7vcczBNc+{1PtB-i1#!D`I;HN9D(q;mvIYNPTDs+qS2`P4+rN)~$n|zvtlRM-Pb> zZ(1uV_!fc&cSQ=^!)drHM+Lurz5^M=Uq|;=E&yARCi5@&G~zv^alZ8`kPkWvqr>xX z#pfk>M`k3J{`Us<)=4lQXN&_AWDQHS+~C~sk@!D9!Nu}}!M(#Hv7uBDu6wVJ%bLA0 z=huL%zbitB;1=B^?T)PtKR|>`s8D~o;Px@0&~U;O9JV2RoU1W%dL(GT;V17aCyv`IT6FW)^<-QrN>Q`-kuPEJum|i9l!V3s~}43jf?R z2njKlz+dk_{9j5AydAwCme;A_!;zUlD=a|%(oLkO`w!%FLh!`HmtasS0snV59=jz) ziuR?)F?Y1YXkp?29{5&F0>4C)CtqFgN+nIWJjw)@2sJ^;&R!V*#{?%8o`mhTX86eH zC$LM}AG9q;VKlN3Z(pE{-Fsf+fy_}P(z=T2`?v{=K3l_gVXrA~i#2XZSdagPo`X1_ z`>0;nX9zvnj!Zuu#-put$fBJqm_~;%_;qp`UN~nxmR1fGa@KbE$Ndmw)?NdNuG)fs z>?zv&o8U01OW2n_fS0XG>Sj_DmtrhMa>T8&IfCBU9ARvJKT#``9_fo zj_u@z;VRIKIe~vkrem81ee`(BXVIo!2^99@g-~-^lEg9k@v=|nsCL-JX|c8kQ%9r<-=+!uVq3 z^J^n&`QDBkx*51%HV<3h@WYK$gCM|oDsVwOK9=_oP4Cgg2cEyft5tHCdrRM=htI{r zjAjHf8XtkA_PN6x{Z44hje~uA&!bz;-%;qQIk?c{8Rn<|1iQ9EuuKyce|z(AT-RT%pP&N49% z44DN>Z;F`YNC>G3fjQbI(61X=u;cY##e_|^6n4l?urz&(C!K|d~G@tUm*u~ z>X+k1KiC?0 zx3MxD3sMLFlcSJ|^Ht{1ktgVM)&b_!kqnXa{v&9^e_n{U`oetv(Th_D6v=wu6?pE8 zHs*xzoa*>%2X5(3;32+;%qz68x}X%w+@b^?fj6TaK2mhIV;kf3 z-BZ||8-OY9_4w(;<0$8#7UZlO1x@LacouBJ6(g!)*W5U0Gl_vU*@pN9U5oCVRYm@9 zb|Skc02c<0nGfHjp!Bv8?$fD)8$#wFRo4{9)d&ohT@~nZb`kEnkbqA)8smknO^l7< z0hA-K7pI-bWyH}D%+!l#m}T~<`1rC;Ja$eLE_zYHd~=(IRLMiMKwvA+<`;m++jw+U z|0Hsj`M|WEZvwN4Cz1K|$;>9#RM9gpEx0yIPheDiL-q^K;oUPFvDYkJtnxV#1!aaK zjq&Q(;&vhIs4s!BQUINIwqw!OYasKYoH75nMYJhFBr-kxMfB{c1|D!*3`97~MsLl= zGH1SEjk;`nbBrW@ej*Pf+Io@I(u1&jPzxqK)I&)fLY`>TapX4cj&SA_&L#9VDpdJT z@T7FWv~N>j$%t0;^o|YQ@-z_k$*cw0i04eS(X*LdoxG@qSuT3%k+@$lPZnnG|N9p`5_0_eWuAyBYmy8%3F7BGNLK zU}~E~&F??yN6qe;IO4uP-nixn7CXOXruj`|My?5F4j64^zW)^FksT7qbip-Lu(yv1 zIdTFWnYI|uYjDI5_DN&!+iM}&p9}YdbJLiM4$$$s7~KDrg07IkzbVv`O3(u;|8+ub z%W35CDh@?P6ro@DtMJO!5`5=OnlPVS2g)G~bcfD_324l?)DMKRYJnd|8UJ@~N*)jgeyK6FbVxj91-8v&cVX{ilT(S zNl0rpgTx7w(D$A$X60R>w0pG%ww#{~t)qUSiH|fOY||^W>Vz`fubhQ^cML-fi^m}I zxIJ)9IG2BM6+8tUKT%@-1o-#09(fA2_4ej>$aR$}Y>`%FhTbYO{tkf!xT7pR>yYSu`w)QW+YecQoWcO~SnsX@Vg zHArVz44V9Y8t8w~fJx4;k^Aq}=q;CkjQ^ydY1xCq99{yi_PhexQ-7j}??GsJc{z&P zB@JpDTTy4rPn1GC8RwaeOrU}WqC!T)ZK;qoe?ArxKXoB~MAY9!J~-Vb+-_rQZa z;pntU9hy6BFFLa?5H-9W3x5u(f?w!uWahgM6{iW@hhy&;QMxHC**yYErvFAZ2jby` z|7Qp*iUXHj)@bhaTa0l*0kiecUDRwn5fusN3z=v4;OTf*SZ1&lq}mQcLH;3-o?V0H z8k(Wlo8ysg)FL#q)(M70pHP}lIvO)dm@!R1ix#;rMZ+$>1f9XtfJNCT#%KftbXq{| zrSIsY`@K77~=70Wk(*RYLnTyzcQ zeTW3+KrnPFpB8u>UQm8!o8-hIVX`&5E>L_3RHuF5!5Kc~41K*K_0^_6&ou8u#5tEI;uAmte6_=o_ zW6>!8WdS;P$pEd?ddsXdkjHzXg?i;pKD-uqFB859djW51(3w;T`0VuqnUp?4Gk!?o zmR-$o4;2xN4!Y?cPYUOZZ<%J`;A&QqafjnbSG)9t>G$FDen zqdel#jf5D~wRb-%I;xAlp8Uql&=+c-*`Lvs7kAK3cT*%^8x93puF(1#f$^_X=<~&9 zl#`c>j4zFcz;}Zvzq|_lRCa?Md*=(i938k&`4o@4b{~5$)Wko~FkGTwfY&!n0P^ZL zTE!nj{(Upi0X;>$pl1rMvRZfID01^V2-!+y;NP@R^6$CuUOhjU(G z{e=!-wps>{843p0_6Q^l3M{gHuTT}1!%sr<38W@v zhc8F$$3JiX1?5bNHt$q|eD`3cbhIqKvfdRJDDH;JDU(40T}SD&l~HMU9-KED3$7ON zsHgam&}+1W~}i0qx)*Flbwb7Q*dWm%#V!JiHs6@u$jXkep|X zvt}NKud+vArs`hUs&X8;*ttXT(ngS+{}$E>ELh8N&rw7|EGm)fKq-}}FtX7KZ656f z35kcHt&c@(>bk)5feeljvay{L*I@BJE1Y&jR`72B1s_R&{CtToR&kzy+Z)C3>|`?f z7%I4o+@$d_2MHXmzY>>+L}J(L-uOj)1U7S;i)#aWQ2pFvFnq5j-q!jL{(QWOEN)JO zU4O^Ir65b8f3_3WfB6UJXKew;5J~7O(?qMEilATY0S3t;w9g_Q!i2rrS)1;o3?>xN zt7w7w^aP@2y}&U}fmrg6K-Da%1P!OZ(C?A~HA}5QQBxjgl0ta*=?COYTnV=QrecZ@4%DG96@aZu4-gwCO4=G+2T(UU>rJcMKZFarj`@UHr&Q9&~C8 z@r+%Yv45{G%-m6onh*P+bJ>N^aBDjLFf|T$BqEIMqTu=Vd~mK`4bSom;DFCXuuZN* z9p}zMd{rmNGo!J14-Zv)WbuQM2Zf%_6ZE-aBGle}373~XhLD#EIPyj^^f~e{FX29{ z8yb#6p2}kF5q+?2ojqzwu*6!`B^YBlAIuG(Ek|5;wBgE~I$1}@3ur|LB zJ5KO`CvATqaFP+eY$)(yzX{x;E1QrrHh^1e&I%ln_YmjcgjYNqhpXbEQGx9>u=SpZ zOX8m6=8z7i_taDL)KPF28swt&ANAnz+Tr;20~IJR-2$DpviP1_6r3FI0@a=vvbu9o z^ixMzu2}=Wd4W%Sehm}|T-_E?INH9Z5v`c?AC%1pEweR(EbR*P<@i*vG%#$M29>j0sND_Z5h2K}&~0Sl)+K<2%wg4^#fHu9N@f@hS8OoiX+Nx_9>Rvrqq zHcy~#VkeVfv=`aMD?q@pJkfpUR3_}+QCRM{1NOPrL1DyK^!ApB2`dnx4MWpmw7?ek z_g;*=zCMGp-J^wj9z{MmMmTJC2-@1J1Q*mCU~iul6Li@U@uUoeSg6A?xx26{EfaUY z)Q92Yn<27v3an|njDmd+!(o9RkQc56jI6Lr;kgki2ZemtG;ch#aSJY*{u}fNb$XwI;qND5yGwTXk&*;;+S&qC zW^^8=44;Bc;$Na&%8f8|tqj&pt1wqQ2WU-WIjU2?f&W)6+2PsyaTOrU#nePQkXL+rg)J0=5)b)k!mwiL`hdllSv9x?-gb zV-ic4$lGFc^6zmFr(MTMo-8>ZpHB89!>6?pqcb@HL-y6BSc1*ROyqnZAVjO8a2 zq~$&gqTb(RZd}YqB?nGHmbWIFxkMJdZ8(F3#R1T%PlDEv3vl^Y4H|ET;P-_J#5qJW z5pUxeue_h=UuPKle90PK{do*?BsQX4zhJO)Zv=EO3*_PlVBei^{NY9mBlm9}N(vA- zLQ5Ajju$0SQ@AC(>DY#h4F00#)MN0xij)Q9uC`VzC{PBj3OcnLrf&zyoG(nSxy%=Y!O{E%0|o0c!F+0w=2&{QFiF ze)BRG?8kb5*OdTt(y|);`@I?@UOj?Y@DGT53%nop2AVRButMiB2>WpyF@}dhb^i!# z@oowZi&w*1qgP^OClf3wa5&=+=wO$?`M8i5!HIkiP|2x-Q-^q*ldDa#pQ(^d2R7hV z{b+pkaUOgvPeW!6EF2qW53K=`Sm*f&>=CGdZ?8HEbwk}KXJ-L&-sS}6)e30O14$Tb zq=D^59f5~snW((M3M)-tfHSp+!^UH3c=O+8_?=P#75sSUQ8)pT)#G6GlY>}2{scA| zxDMx%^*}UT881qngip za9OBBD2AvQqBLjR52Z<)*O*iVmYfL0Ad?I|_%*mk9V}u`?*+X&T4EQ@J z?AIyF;no+HnD_b#T03gMc*DtSK}w4v*5{qu-DdQj>Gf1!A(~g-@kMU#=eh$fJO3n=vE|T z%N)X&XTQg6LjWH6u>&rDG{Bd(uf?I8H$teT9a_<`6Xn?n{-($BxV~^WzE(U3zKwqm zLx(2e(&qhm$A5lc8ytjlw+oD2p)6Rb`v_d*hU2WAK>$JxwPEo>eEr4)@Y`e$T4{?w zTZ2Utrfx!$I?4qm{TXB~H3G&tYT##rB*W!=9?pFyhmYx5VxQdtH>K1Gyr$OTUx82Y z@WmN;@M0Ep^vgkT!DZn8Ie?2q4tx^s+o!jU#VymKRanW zNvOrTJYulb$~M8jEQ@ni?1yol6VYbpeuScT;+}U~u!DtgE_+-l@CaRn`MBV|Eqw=P z#tgyPrkmhhc?23ah2V@0v+%mFs^p_p3J#y=j9Z6n;qdbuSo`7-44>?PH!rvi59`A~ zWq$}fls3Ze8w`Zb zt>#x~bU-I8ylfBkmK)*9_6B_P*gSG*Z96^`Z3f878Gdz~N9jK~B=e>Tq`u4IE}3lX zUFuD!mJ;y@X@nio48Hcr0%i;O%#`m-aQB_nF#3wQP(;MSI8+bt@+B%6k%HEj3z>@- zuTa|tVGs1$Qe-C1!q>y+laiq=ggc{(iO{pOn7jpTb%=*~nrgyK=n#x}W{nS{b$HNQ z43}EY;UD#-*r0X*7Cat+%|Au((ykBL%pQ-Ct2|y48VviRZ17}OAN&@K0=K@!_?>%( z;BE_HVl!8woW=w=H7*bCN4kJR*>h<5XA4rX(zxZV2iAKs2X9y_iw!!L;yqCh!2S3H z=-c!LHLaNnxv4K9Yo;3xdew!k8pa~uCgHmoy9h3P&>D=d^UZv; z&}Jl#y)F2{8Y|JfmpTwOcpshc^#l7yC!u%GW?<&(GsVGC%s_=LT2-3^^5Ijk{)B4~ zAH54S?d8BeL<-X1kAG{e?xy5}`@X5*R=QBgwmD#fO(!7tr84N+ zq{6R9-B3S08Qz^0_+EQIF=wl#fXp<9f^;bee36g#w=TkQ=iA|t{8gm$PXQyhAlT3` z2CbQ(4y^thNDz8(+n%0BmCBKj@<0SJ1!9<}(Fb)dG8p|m1G9d=fDMjU(2njzWM97; z;lz(9X~Gn^YknWS@>mLwdbD9Hi=fIy0@_PE@vPky(5tu|eSLQUZu?EZD^81`{j=bi znZ@F-O}VfosSxB%O~A6)Ug!}uLT|@B0I`?}P#*`{X{pFratVAobysj~>Vj&@czjo8 zB68^y?kl0T_<_GV^4)y^eV!zblYYb_+fpqM4GCVd#*55+D=|7JBLkC2OtK^iWnLFp z;L0ILxkKRRf4_=d{bEqr`qQxdvlse>MxZHfgXqtmqX2KCq0DkN*g9L|;KgBh-^~H! z768tKTK?HDM=2{@XP4 z&7Y!6#!;}qtCz_-v6#uZ-o=dimk9B1SA(R@KJ?Z&9UF{pz@C=H_)zH#;PkBFu~2iA zzczs8i}Rtx$PevPvH**L^`N@_F>+sH1W8Mp!FOm7mUh&@wKsyhX6!QNf?hXsdBZX| zwIu_N&y>WM2d6>BC6m-XQ2825s8)m-`BI25Zn`JOPTI^?&tGDn&VYTzv-VPPVp_`&INvz zH|EYI@w<$uQfCKED40aMOcm$~lT}=i-*>Jzl+X`95Y>9riv2rAiOPD8(X`vrbWYJ3 z?$_C4T!V5t-g){f7bJ0q($i|t(UXg%Oq95krzg4GnGLW%MS;y%^`L6=CQ#+_;dIqK zB_8(164M8-$#zznc%FTQXIGU_8=qo4ZLT!g{WXM}(17sH_-E8pZ|cv{kIC z7eVK*&!7SLDc3eL7wmi<)iPNG3Xl^NaT$B+Aa?c)5ygeE7KYe9B?ls-n3`e1?&7)zRN) zNa0_3YP8sa9-C~%S@{4znhOV2?hkh3Mi4VIdwSLE3Z1t?pBV4@fVSUL;AJa(__t{> zylr0&uli8A>aTMce_lBmXRnhd`vX64{u|q9o`Mapu;v0MZfvBN6qfJ>O_Z1KTfsLJ z#8%Al=;Kb;nqghdX?&`MA%A_vd49~-r+i}gXx`wzCh}(6b`n)IivE*Pp@(BH;`4>~ z=|aO#q-9DHVM`e5de4%dmU^9^;ZVix#ESfk^F4Ho;|Ds!d5~&!P2%;>cTyWSUwpcC z2^klXfUgEM(8HkQ6U|TuFPREcjQ8j#8cJdUT)rY(9C_TE5%m1n+cU6y1r=igm^&lWW-wZxrQ8 zSI_Rk7rNCz>V%lP1C#KJ33K`3amI90?++Mn;>*X9r=)bRA0MF_#Bch0j4yPhv~H9G zZCVsT4@5a|^FNexyO*CLoUbm)@JbNoqzu*997zYSMC0s4X>uaqJs!G#9nb3!9>ni9 zP{$9B^w#?c^sb#2Z^WyUW}W%sw$>fop)N^UV7`u@ZncCa*N*4aOQMNM%}joDauZcr z>&TlN+eNhvN0YY;Wa$IfG5pH%9lTHIE;^%KfuE->&n|glNrOU)@O;@jbn_h%95&fX zo8R{$9Q};iS^uFo2Cq?xJBv7_vJ2F5)lYF&Y6^AS_K?OE&!it$>Qc{qE845-%wBDZ zz*l|U`2@>@jhb}TH-u%CVe_jtrn5EU&HzK`NXd z%JMJ%_R)p6c7XH4R#Gvro!FsQblw9qzCWj&*HxOz-}^8~Cp8IPHSs5!y)B>LB5TNP zRqEk}i5n{CXK3)pkDuX}{g$oL)lKJ*&k5$=jh2Fmfwf%IGk4Omh~g#7tB6&u5qDyb z8(-PFhaXU!&U-#A<*Q=!`Jbi_=n5@GUQu!>pP~4a7S;UY$JVan?dv+lMGJ@HygNZu z=214Sbp62Pu2W+FxPQjZ+tta@Uxsvg;2@o1_XPiQv7~8P3H*ik_T-hUD~X)rB{u!6 zN7F*5VAC1L=-6^qn!Y%OKW4X_I;8f|&^TL~>3@hmF1Dtg%Lhp3Ml&i}yog_^dWz?B z5x+i;p|d7t(pSlbRHAh@H}sm2E3=D8=l1JdZL>M*nQkZ^c;Q7| zH#KtaG>dT1A~8UKBGr_su8hq%E9OfF=&E=B(Msh(PD0`#&3RR%p>#pl7T#vVZzwqw%e5EZrqVvE_{*v;RKoBJr*&#LKY#OAEQQC= zTP{7~*jhz$NPL9W#cI%&9SVGutTF%nbt-)=AIW)39pwLMr_Ke?R{xwU+)U`i8yM%;0Ob%kl1mzj??jrn)JLe9)zI!gU-a(P9O1W6~_% zxF(DKo&1&dxW&`&`~Hb*rp^R81g=lVq= z?LVBo=UB%59UaaOofC8OyATat^pNV=KcWLAbEwe;b;C+ z>3FMnK6PCRUtaD<{|SuT{qs_3>D=|CJL@*tF<~U1z!uWbgq>s>bs-Kvm+~WCm-B7^ z{P>5JS$yoLYAWu{q4@0+E-bEqPF63Z$B%L}Bj5tJ#JZVYT0Dgx`FMaVeYKv%U)e{$ z{<^}qpMA&w`uU6hedPlMhGgh-k`)1JS0Q^<)iN4bxe z!-&H+9p17anwNZ;&Yv#7L7l>mV9CH35>;{mZ>|0&p187yo8KZy=loG-Gj}t@TD+5w zkc#J%5B%T<*Z^Mbb{zc@)k4;KT%~S*w0XnTONrd5v3$n*beK3)PHIzlnw54Hf1kLN zoP8l8wRwuv&#{1SF43woGe25op@OSsPJhI2UfjURDjuWb-s~aIGpvcHVi@(1dqq>8 z=<@$^9+0pH$EZSZ2e)>Z8vPMjz!?aT=1KNS)aFns&sfgl59xOCQb#ZF>>qV%(>;?) zyH2Biu@hOH{W7$;`X8?7(Wm=vI@9@eGGvvDEr|>eI7%Zc=+f=4Is1~yv@b9P8}62) zYi^9@_l#Id6LY%A^n=^+HqR{5bG)1!ze4DfA{Ckt=! zzITaAMLehXbHnHrRfgpy*dxK+#3GWr3pUfA}y~Sq+jbbN$R$vdttFUJ} zR)(@?nAvcZ(*FRjCDJ}`7G4qXB~^e32!WjW9w(q zyvvJ}*C?^Qa`jk}@2wQFv2>}{PI~Q3F>C)_iXENM#a%mIL^_#utZVar*48nSoEg28 zw(fr?_PI2HL}=Z>Ppz)tam8a;=LR1(T|$Sok@!NJUDo69#nE_ZSTH%CDbyi(t7t%{ zIyvc|Nu^6n$u0v8cKMoEY%+tAfu?(6vB0`eNxn(;Zzv-}uBGTu`X};n#3Aal$dD~+ z(PPj4GqU*pG=pWHx0AnPH}LZ%uhJP?x6n`*M|RmGbv8mjnteCj8Vw6DXJ5WIV{cCT zQMpVagDAYsBU4@&ivxx-aCLeKr`VT6_P6G<;)CqwtYFfV$mxubXqFE z;FJ`?U1?*-F8A&yy3!nJ4SvdH_`l&wTSR!ajT;R> zGNiXR7T1S{kp9Pkt3RBhqZ9Ab%33M@`odzo^HU!8&GQN!jTV#a zid3>keY1FMqdgTr@uXWzlIirrBl!jU`neJ77STPGJ9w2eUtVgm58voBlUK4(_#k#Jixa=95Ilrft)X?Gw?tY;|&l7i2wBQj*UGGk88-I{{=Z|m?B}Y=zVTyEN z>vyunr-JsLe?)!^$?@!cC;sGt(83Qi=;#La>StVFt4bR$bT)I&hE9+vM4SJWCJWiNJ44_ zHEiF?c_bu}iS-v)s|)$;g5)Wznl~aZ1cmeT{+GZ%iKkckZ0I`Yd7O)B2{Avujfids z;cz`Mduqf{_D;Fy53c16@{BqRM-Fy7RXVJ*EZEq)2RM9k(zioqBx> zVUG?yC&sICv2|%I=iIW8b-D6|xKEDdUJO6UZ7klvRkW7l`NP!N(l-_4f%Q+^I5Ugz zP9N##>+|@_Z|`!~X&C!vx)DESqZ)tNs}Ef{ctk8CnMY2{d4#u#1?JkNY9@W!TvAg# zw$g6y{mOl7e^JlASpMbQv;5k6DR%tX%j`RZ*~io0l85*Ak~sBublsB;ymZVXK9fZA zH%z9Rc8GdT4zCF$LTQe>yL5j$Y$ z&KmB`XVZEy{ZBzMO*&MDKBmFnNX{=H|)HFP=i0aoRF$$PCTE7j*b zYki*GyJ`cQU3Z98z4jYAo(*Tunii8qZY#M%JLePos70W9;}f0btq)vNKD|9Yo0F<* zrJ6d5e6{>?Vt!{PyZQYNUP98JH!zqHny>%9xj{ z5wSa*^;xq!Hl$ajm3y%zn(jVk!kh0>;&mNn@|)QqI-;tIgd|E@R2F*)n5GdVwB-Ud zIRA%xC6h#-m2lkl@Ht$R;RM=fIf|COi=zLs3b?Fa`^9s_=Hy552?#i!CLX7(#D#qA zsa(9fmMT7Wq#ijcY*<1Jm2Xkyn$Ra=E>@sH3(r92q8hPmtt$7?pq~?8mu77yI&v3& zwE}UQ&pORLLB^bqAshYm$@`wEw9)nuH~Y6M5!Kw`)P7FJ{arbD@W)OXHO!sbmMy^2 zw3jB>#bN8mp*VNvD8BcHo!}{%%C=A6O2X`~VY=Flb_a!1Kf6$BK4}YoCw&pGCR53a zjJ$cn9z#CLwv6ueEyW0ughzfUw)GWebDoy$&nK4T>Au%kHR%%PU7xv6GU4Nu=%^<#Si(YqJIZF_Kzrg}P&^ydtJ;LddZ{=N|MOz#=& z+t)*`{xe_;-lt)TTEZ+P!KSp0{oqeQwq)oJBX0 zOKIcy_L(uX;Q3MB_xeXVuw7aB{vYJ66HCdfs-vVPHjg;!{3Tr%RM=tj?vXaT1MGao zC3v~vH*)c>h4C7+w7|`m(daCnsIn^95%??eLB?@UuE8E~k!MdS=8|x6M--aAQ3c65D*fRC zy;l2<>l-nOj&Ml9X{+Ms2)2mgQKPWo&Je2KWJ+_~ocUX*it4w?@H&3xwB$i5RY;r1 zJI{0FpKAH>#tB(;-Y$QkH`|QYJuk%bY%Y`Hx;9c6qf5tUe&9@o9VLU&2gy|F-&}uc zHky5EHFwa)|b~P_b)fDEl|(3+o>zYr)>G zW^+5r*z3Ea*$a=WsI;;g8<;G^u6sK~MyZ@45@)CI&U6NQxnKf~ToHx;%1LrKUXyx_ zr8Ia(2B#BVguhG9W4&*EC!B^hX*o=2JFVkxMl7aM$7EOq-{bT}Z$J0e|0QQBo=rPG zyU<}~epqUfB|qi&d%EVVyhfe9$;!ZoS<#a@roR_)*^+V|-R$qs1nU_wD-|F#w z+m7%KPI-J-oD+Wwi-?0m3_USJfe#ushF@@R32)Q2g4aHJ1{*BSp?O#9sGoHf-6^3( zSBjs~)IXS$9y&>wZeL<2^hrLw9!4u_%jkj)fq0znA1r@o4Q*cXh8wZXh=yI9LG3Dz zvwo#foMM4Cwe>y7MIG2dZ74;l; zE2Ein?ovL$h`5|-A`cX4!igiK@d3^bKSFOzP3B$4P2|Tlw9)AW&Ai&q*OcDrqkH}~ z(vPtQTzTAZe%+kWeCC60nrE?`tru#UTRYy8xffp$elSP;x8;X8YlJaAQM|v*^UN-~ z^RPahD%RuY%zn$MoNdOlmR%y*Q<^K=VnR5x$93Yft4qalkBx9mo`|Ywy`{%>WY{E! z3A`2@BNtxQ5QP`h=*5*HuG;w@x5}-6dTXDgOx6y1`QS}%cjcC;^Xit?Qsrx=BS}LK8QphU8Psqp?S{k&Mww59(qV9Q4N-0qxRFserA)Bno@BaUO z?jP^xz2`j7cUCWM^kBUm-*L4^JCo~w%(~ecR(bF){n(sMw;j?We^VWC0;Fb83-ZLtQoQQBLhfjn=zhBDwHZR8X?r;=MSHh{9($=$7#xRMX0`_#13kw?EkC$c6qWE?@9D8>R zJ>O))KIav)gijsJjQnE>Zyz)5;;pq~6Xw*0jcH_I&(33|;N0ETa1)PBG+}j)2dMq= zJ+v(DAtyGbgcAzv;kM|QF&(9iOuMoG?FM4W1-%CNX*9r#zH|#4JuV-| zSv>Uzhi8)n_nR&L>=J{ywYt!<*8%o~1#>p>Mx^wOHy&AO1efRVc+2MKW7RpU+Jyct zwr4%ZMt>+_#m_AKyCMS|f35?!ab3_L$PEg}y1_r~eh4zmQg}Sqb`Y(shXpSIZ)TOy zArWQpDtCsupTB{M;7vm`I9mm3hD=`_JSiQXl7k}+Gty0Q_?Y}~y?MMVT z3(kS@W=Emq*bGQJu?Gf>`r&iM0wS_C1R|fS@`W!w;8qo=)5ytTxbjacee~ryuQm7& zYHJGc)mf+E4R?pRZhOHBIEh;Mf(+(toWhQ5*#}+Yim8GYu!mMo?55#2q~_z%Qhg6z zkv&Fcgij=IWGBIDn}?iV6?WNh3-bs5QiFLgJ$-)-Z}bGA6mhKh^ATmpV|47C&8DiK#Xvry zf>Rt*G~-yy{1}?t(!_ZNC839wALTB;;#4j-(4pN))^)bWdEXRONavvm#M5&+-C}OxiJFK+HdB`~@nMr?Q7rQ_-0^XPK+EcVnIQ0-kdI1+9Gi}Ww@SNAM{gx$8>^Uy2Y{)q(aO4pO`>q9ub zISRnJ$1+nJS(cc`a7xbr&N(W?_G(_h8Q=K0QaFvTcCiuFKF6ZPtOz_QbQaxT^XX}q zE*>8Zz~ZnDzb|eD=PSrgtCv~BKdxNHzoZ;aeYaNemh`A%%#k{JEh7pFzew^mB1GV3 z(K%@NRE1rqeQgfLAup?euAVqFRI`pKWoVv z6#8O{t!tLDTdu*Z=kXZkVyVaMr&U7A6*bPI^ey+VU>008P2r@ID>&&SL3iZSn=onS zDEDg0DLUQoJbT!u$)*f-fx?TQa9*|oM!(haKE7VhD?IyOjdYe6Ofl%kJrmTy$y@<< zqJdSak|#Z+D22b7DPC|CfrTeBVb{Kc)F*yB^^YC^d#gwg&faA$-kHNKUOmLAznlQ8 z>^Jcnz6#fL7D~d}gVRv-haVkm6Xsr(9^ekF9f1F~iL-Y<4T!3*IVt#K1=?#1XurG$ zzwVC#{&Ky-7ia@%c-Tii-&_L}?_LAx+bO8HIvKyLx(1bLx?r-(l1#Qz<)oi#f`5t` zs=s&%8Y}{o#|_}rap7p17Ke3%4)n}-ClHYxg5}c}FyV}uSbkWF1+FRKUt08!cRf~_ zvss+PX$Ew1U#z!5xR)iFc)t;Ij()XTkPoQGf=IiQCAqKc#h+Yr15=)gQQc$Xcsjx& zWa=qVSfHVYp+Ae^;La*;*Kg!aELjRI*L$$uEdZ-mmEd=?qg;Ob01ibifW%qnxW%{* z^PqqTc5Ai6L-DHc}mxpogn-Mfvtbmo}2YF%i9an!^oJF}v zG7;v7wE+pJlUl(Ku+;?fbw}ADku6{^63Dhx>v5v7U3hjd4Qs+KJ}?OwtOQUUsxx=yZgboM3l6}$J2-&d#-6sFh^`d=o0}YD!i35PY4%A8^6iHtmVX?Dk25dgUTIh6;TeKD z_XWFA`v!Cu6((xCE8vyWWVXak9%q#o;hPDPs5`k8&Rv&8OTTq&mv=pH&BH?SZ(0m_ zrZi-|*haS2HCBk_YR%><&A0OV>J{1K9irH?_B%at$eNb=EyUkdVj#<3jy8@~#GxY# zWHJ_^kM$_zFA9bu|4krSTn{+r+A_(-lB8^Ke$DX4Q2t|=E7ZGMo=mhgBISY6C) z6xUWGg$ZxqhKC&&ZBWjV3%1~`38}oLzRtK{{eIpq^#~{^8bdNh4x@bWF`Vn}#n*I1 z>i1m{c8QmR{9jcz^ldDQnIw4szXu`w=6JGXv;lIrUgwG##rpB9Y3bH_h!))qH(zyN z@@gI~HB?}0%+gqeKu?f#$4c?UPGWl6V22{aX7z)#h~G{nz={$95fZjO0~k6Q*I z`g$a^NDb4I*S~R7FU7J49V#s2({K3PCqc&M?IqpA?@6J^3^HAi^BUiN3IfC^hNOj~ zOOPr!|0e~m!THp;TNd8B$-uh2@nmj(j1&)_?J5^A5K)C&g2{|Lh;?ENSc`XA9*^vk&DW2q-Hj$X#LV0j3Y~UL2=pe zygP|F4)?)o_j5Sk#*TFqRKv!VSK!TBQJ(HwCw`3VZf>jO2~fFLY`yl#B)Ij)03Di_ zfw{K|x%{A(`}eX3Ti6-;aKTw<+cgX1M-GvHH)Qa|BVSliEJXNoF48{r@AP}`0y5w? z43@fr@Ud5-c2bK5+%fy8<(*eWy$bwk}9UM(|9-_1+NxoVP%sJ zahe`U*y67w`;!m;UJ%dz*{|U?37zHV#40nfA1leqkXI-)z%g+#d91YfhjWZplJdG? zpoI${=0ON2WF*Mw+&#*-2=juS4GXvo@k3DXn$P=Pa}pbKD={zp3_O0dl(2t&-0nX& z;Z=}4TmOC|tgBKbdHtKo68%|l!OfHvy!NL)D;`)a-2VbbB~!@zN>4Iu?F9FB*w9Ux zpQ*)QvC&6|;9LWr)BCCj8<&3uH=k)_jLv2h&D7^cREqJF>NO&=1`r#Hsu=?2zq&pOuKh)6|$W*F>AYZ$hU6V=`m$ z0`hC|G^%_e6xa1cgYm#D5?B{QBu92b{y%e+Dm{kLW5qd_o%4y4(KC2x8wZDKr_&xI z5xAaZ2p9U}>CJR`7=F#e7yYp~Zr)ftxVeYxuyZ5v-n&T5G{pC-&SBVWIka)_-sOb`j>99=e|$ORLS{AWt;$~$vzC0dwk7EYZM=g>K3_0}Ke-{RJKYk~I>6I4r+ zK=&jczV!HFbZ_dVzuyM(GDkGA_17shO4)$Y+Ilp3tpb!=uBC~VQBZAR!0C5r(&-y6 z;2q5+Oz2=QJ$vdp&g`@?&Jd2xrD=VL+cWqcEB0IFTd^p*NuTD1(= z6<>Js6*Swv5rgyAVYkW;AWIJ8r_WJ9@;qT1Z#^^mu1X~r?f|E_bC4ZX1$KA;a*2Bq zF;Jz5&hdK23;FwxT2Dtjru%~HdwQ7doUt3fZ8?cK5-V`7ssWzakPd-T$8ciF6<%oW zCX6op#hHXD(bppv@Vwksl)HC|TY9$|zWbSx4MUk^v)XKYp{2;yg*vf=KgPl@OB;G( zjli?jqkz{VOh908M3)c~?37N27|$v2`0r(Sabp7eSS(8h*LKjtArmH8IsF3 zgfy2dL}QN=SaS7K4bM>?lRcl&uNNgT`Nw{)`@(HlS5b`n{e5xZpc#%@Kco3PTNe0z zGkvjqImp=9kW|qf_(7?M%^z>eZuZ)-vP=DRWtcaajc2iFk0iXV-U`2GX_3h7N5SvE z2xM+1MA!5_yuBNY2mHHnV_^s{?b#w4b=DHEF4W_XpLh?mt({Trf-MxQPb7lQdfaw0 z9xie@=;dBd8-K6I1=&h8f0`V#Y?NZRjhexsPZHd&%phG+2_UrjDF!EY;EHK6EI=`z z>8?%1K;>+DzF`dfE+4@|vsd$Lrklfiqj*?}*XY>?P9R9BVI9w2Vb_QrEi#MA7Q=XQ(n=76VxTu zm$gN!f$?xVaTy;25#NrI2NcLX*QcntRD-*maS{%MwN`8N{Q!@9XQ*h9C_VD;6SQD1 z_?xSci9`+Tvc@rmv|$wcEHSdI`44ilMy3bsT<~jjOhCtT@<hSpb2z1>2k&e=C`onbwdXD`J)8{NFJ_f18diF!Q#b1=wH120Y*>=n+ zWfe0S^MG;Nk5k>{$|UUE32x&(QFbj~l$n&uFuM)OtiAs?s>t~<=gT{4W1VNRhd-yW zbqd~0T7b=5eY2hYFkZy8-CpBW$(Nk|loV*UOr(CQr}+iHrN~$@6{59>kj|u?`0k=K zED>-j9}JS2uZ;v+*gCKT=f{w2-p~1(D-ai>o3JFH5zlj?EGse&tQ5>q{Ks}CvA2n7 zCaJO;eT6jXq!zT#xWik$q?}v2RSb+R(x~psCAjmzG1?&1iVYz(_{LitjU3iP;@xgk zFmr~)qt9T)+9H%Mzld7|c$Dv;JEM*I%zEM&wj*@`x{iB8@5|eq!+^aq<fZv+Y5w=w2%ey*LEP)0={a3B#^kD0^R)iRXD5BkE~fWo;cd2&~cS3;KSxTd|;u= z37wgTkxyD+TTXXmeH#?%^s#Fv|6O9%f8Cb8Ui5s8o##b#S9REYC z^)tf+@;h0RxVRSJeTh);|L-_iIq@mc+nWj2p53tQ%rK1zTTg3+8)#0R2D^XJnfqzyK&Y&Sr!o*iOE~FAlTCy zP2L2f`?GAwY?feW$~`drG>;AECy~`XHX!D@iY;*8#!j`Hkw>}8@Zz&E%b0fpovM>* z&fzND#a|4O&HVy?TAy4qPoqVg@S6~scd?4Iy!e;< z5*3SDkA;X%MH%UJbRcLF2L0P6a~jtY@webVd;Ur=^YbuZ1;=)x^=JvcQCvfRzZYWr zq~p(Qg^Bi;OA0~U|7XRkXWSOjZ@+Y2fLo{C0z$F0G)1Fx}IcRK~4dvwhP zF(%pDz}-Es&8nZ75xG?#NS0F=@f*00P5wi;(btTf)$_vCsP)97r3hYTreS8#GWb4e z1GMN^v-ew5cnL3`z@krn#O~QfOb9UMD*Pp(RaBBZajGUtYBF@VO@RIO@%ePF1!v>j zf?n;yq-{|wtvr*%MdgG+(EL8W#8Fvd#=i}}-%X_#($B(o!yz#3`OaBQ{fgPIib&5q zSCTxWMPhoFbI&5j(b?twJUHA1`a$_nbbGmt`c`o%A09G;KETYqTN>eaB;ABjrr0>H;&)kK+1Fe$We_Z{Wkh znJj+)Bdps$hh09n7L3<*(S0qhTv=^1H`FwiCPu{KO_ggL@9-+ZH-AUQHwAN@g_`hw z+$?hEuq-pVK9x!|2>clG*YQvEAal6BovG}iXs(!uhbQHdw{EE|udn zDSN1*Sv%WsCrb;rek^IRw7m~4V3rTw3TBcBuho;v~m45t+x>-<*JjbOH^vZf=1Y_EvmIa($&oF+c?&4 zkU`6{3(-;a3N2L>XPfd8`2|bQ!dj15;_sP8NL3TY8!cw4e}vgjn~fyfH-&%S`8)n+ zBTdvR2BBH>KHc|i9+}E7<_>O{K%a%ig77s-c0PX}`EuhDb-E+i_g8Pg(9JXHrY=HG zRj$HYcDAH_Rx%lip2S{lXs2WJC)S?Xe1^s5zTy7`b@H%19H$nPfp5ADGm^?6dSOD) zG`|wF#JX|DV@p(eY)E=0C3p4z%^#7u)CF4pxhEdA|9ZTszbzQy76U$Y2XHa#9r9eqgy-Sn<+N0-qgX)oR>moj?r(;0Ay_7?aI zPLcihLfMeacJ|D}ifyS=W1>5E;ikLo^laZVFdES0myha!UT!p4Ot0tMm&(A01*LR2 zFP3{Qy9x&`&cr7v59pzSQ*>(E9Bec>gRL<;p>5qbre^wGuv>+)HD8t4i?MfD(?AH@ zu-uesO#Z?iK?p`P%QM&34xCfDoK-Dd2SM(J?C7ut^^6R{go*Jq(_S32%=Ux!>Z3Hx zMht&{x{3`p&+*;~DRv>9(oZuGxIt{7y3iR9XC-miOvEiwQ;b%xYxO%#y!Gv&Zbx>sEF5IOy>+aywzzXch z?8I|l1L5S?7~HaD7ih(tfut)nG*eXs_C1`z)c3|>?^+j@zT*Ro`P7Ls$A^Rd{m0nY zpT(NG&fx7?)0w38G02xQVv6@Bv5Y!C7L31y?e31`#F9;HNVAkhwwSPrFJsxzE=AUK zy97VqJ%w9~joH;(9jIBT4Q@N)AZqD2m?-g-Cw0|{JAdLh+i^^v&FaVp>G}5TcFQC9 zt9b_K(pebZEEOvDhb~T+kiK(9qp(6L$8TP(cWkp3!IRG@hf+c z4gG2;RKx?x#IM}k0DqXulY+Td_mNDMP|RJp1CM+#s##rD%Kv#T6GYBW;c63xx&NN> zX}@3wCaLSW9M4j8{CbmnyWdq%2(H35y)h_AaX96-}4rFU1h8 z-RRu@6DQSKGwCstaPt=trtN);9=`VmBQ_CO>-QEX>*}MLuL(Q$C7qk_Etbm5J7O<4 zA8TEobIN5pye}_>m{GtN-uTTW(480u4b=zW{ozPV5_wI}mw0ng%Aacvb7Qe*%U8Os z%^Y7gchft|ZgK6qe^C}P7A#-9<4;Vlhk-p~;o#{qt}#{MFSXi>$9_ApK4}s3E|h^% zx!b6Jb%2XFODfoD%*0rs3uNB6%+vAT01(cM@DgLM?1lo`T| z&!1tP?VAQmfO&72i%I zuiFRl#_=lnHvbOvm*|oD;wfaFcoaImGay!rGI?9p-N1Qs0@*3Y8q{F%C@Ii4#d!?Z z%vMss&9V+V$tTOLH7Z<8LX9kc>l<`vGY2R3D=quGak9o5er>Dg`Q>7~JMd5HTMsBDHWF zdoSe7yLj9mb-p#?^u%2B-?NfNB;4ScvU6U4kw9F2vH_EMP*5R9SdOlYmFeLC>%DX05lv?$Oi{=D=b^KTy zyD15O9C?Z=9}=+Z{#Stp+X$j%260Tn9sDNh!^BUfvkmiiWBQs4__b>m%neLHt5++y zGv`9VD(V+l-?Zauo7QkAVk7xS4)tT|0atAGn~r3^GS-f5!-m{f=sn2;5A~kpYpq(y7-c3&7oAxR} zk#QeVGdFa6q0e8uz8h-S`jBF&5<2Vl0^Y7hJGjs|hnxITlXT%EvQbP4_ap~X+2^rb z;fbwud|VGLOGBQ42Om8g@^E)TC~fptB3@(nbElpvAW9d|^S#mZqE;^M6WPr__EU|w zdr1K(^j*e*ng*P6ZlGpx!~|Ns9)f3+pOx6&r-IywY<%=>Dy#CIi=Ul)Im>rDAX?xH ze=HY)QyQDl4<#3-}d9*VUJt*y-Z)h1?tKa5l_;X}vK4P#N1NcqvC^?&sd}IuPiW|l?8G2i72~v1aD3hV@C4G^~#%(A6H*OZ0T++oMpmo zeRYwqm>fmTg)^v~#e$lO!@|Hb-iE~^_c-@oj`(9&Bs>=E`U?VQ;{$JN_$whR@OB8{ zw0rrSs_j;iy-1rZeU;3X&h%!vQ~%(RtpZ$iW)`{)-^S+=LDV9NBfNnGvMP56@zt$@ zIXTnGziKzE_?(2ci(646%LOb{s<|rxID@!&H8 zEe^rQNjk8hLLXwP+n}>+AK6&dSF>wLEt)nZ<05sA4SAZ8V4n!^2@_|ho+~psBV&4M zUkFa=(1(teI<8xa#{`inu-bN%j#!3)_wpVHQZquQn~B`K;9LCBck(daI~}+0O+%R~ zJzibmR>Al47$=q~0oUxon%zpcg$70VMgA2HiMC}ErAOiG)}3VVT`UQ>LOGMj0m?u1 zl0I9Qi@b0H?uz>d&{XljP?`0(s?`#gwjaZWAyPAV?l|7QybbUBim}H@o`Nia)zm_O zJ6&>BV1{{#I9=5Y)1CA2+>usX-E|y(EpkbS?KTJPB;0h zu{+(qP^u-2-U~mVt6*MzwQM}qzb=k{2UYN4;ci};ek;U@yFxY*#^?7<`b#2dL&ca7N3ktK~-r9N0b_X8X16E$183r zzl+=4VFPm+$`#Z8L;POnyAIxMs>OZ}WuB zzNL6!f*iRO@CcM{ujSv9Qe?#gt$4;E8K1u~2iB`ajgG`%n^`i>ziWi`?&ned%mz4N zHi=mL{t1^SsIob`wV2`H2dwP6h*}4XA)smpwm;vF&CQRnKh=T1a%zjWq1PST_<@;Qu>RG18td@_(;|hL z=s7q1Hb)1`0#DH>iN!3Wdo1BQ0HhwY;BB)yMz6g1hptccnY7$ayd<53?iFR+ya_p+ za_c30yIq{vt z`=JHv7CVA#>Phb5S1mG6vV)rJCrtfyAvLJ7#L$+5@UG+q2D&G4R-c=By+1;+H%}5Y z8w4t*kd6bv@GXuPK&oJHhD{vZl}^NGe5L*6K46j;q2fx7rZH!2URK$v3!L+ zvlpL)^DIwOPA@Xx9EUIQxdb8JmfaiC<8T%}SH4AGe;CKEewQK_vnU$K=JM65 z=W^e-iIFvNZLl;qnJY;Rr9MIiG&Jpk}XtP z&`E3cGlDnME0PP34uO~xm0*0N5}!!?;*=(i!ryU!AiX63&c03O`IQz@$4nit4)f(5 z&^lW!=V6bTI-em%fNx!%cN)49mFU!`C#b^YQ!F_%lC3K5fSQMQVdVlDmL<@ZDlgQ; z$b?c1X&k}@$+m1>MI%@5dK9@5UJ-@9TbI-KI{g8Z>L(?E$>KH;$Hk%%ux;uhG{(Bk|3Wc<5TQ zl5SgR&zl+|0*}LWc|+HYK(2W^RX(MF>67^!Jtqs<$+h@A_zS1^&=_YM`+;0T1E&*m ziyM9N4tAat!8<2)=myhTuHXI+|MI1cw8&>Mmp1(~)pVNz(_Va}y0cZW^nwXV}Zjw=_$^clS+DBEa$1s|(AX>HF3Zdl$1zo<^`PK ze;FV5lwvKjn*%$u~q4y zU2ke(rO9D<5$OlkYE$srXb>k8)sOxj@2q8`{c!qOXLjULHGYkVV2(-inMbDwy^tMB zTqdT{AlO75d;s#f#8i~Kly z^>F^J`@5*mVj+_LSON2AY~gtd_V8DB^4v`AB^Z!=3kU9p;C0&*_<4;!zLG5E3#~Z= zC2IdL=GJ_+-#VPeD+~512Oo^%3$g!h?tmiQsVJecl{Tczz|p*8ELuAkjrvP2wOwCC zJ}HFo4sJSy=O%Y!PLnhX>x)mgikwf}QtWo!kT zi|5$K%){)}^IP1tb|p}2RKdF2Ss0&E28-QoNJqm&+IFprO)S{U0)I{*qsa|WwSGMN z(=-)F<=5eabQ@4#be8{P-xOND@Eg5-*ASG7laV`f54$`YxGvM@yvyYq@ysQE2(39w zMSo49H%`pLExzk%gyny5t2UGFG2M-EYU}t+br#F@4TdjH#~^f0Jxymd%rNZ&WAm=y z_V#t`S%Do3cVgVH77@Hr*TXezn?z?ycGEQxCu!&VP8y>xj4s!-L0sJf;%2Gha;tNk zZB!gJzNk$Nd*gVkZ>n+|`TxP!p%pY?%`tdoKNVlz0f~ zAor`%x=H8;{gChs1^xpZyCj!hk~8G2+mF(Pp4#l7V23`Rlz`9dd{M4`IjFq}g}yr> zAX(4}Pm3?3ZL}C{9qFPe7FW30s~Ad3TA=XsAq=)YkKgB1P`lDM0w1O-xojChd{1xU zEEH~8kyVd4e~l_SSL7&lEBrhE>8Z4u4nu` z_3VD&9k#DEgh|~wjW;=0rjXHYUAJWuJAXZqT{%0KNq2;@1p8N*S$YmkB+KZEy}JC( zaD=bk@)+HUZeY>e6z-p?9*&iYVP|GZvw@|`+|Jb!%vR(!Z$n@jH|Ope{OEp{ZhSU_ zZ@F$ad|ouuT2Ee}fm!ItKI$*WjS`!1#gswZGNTN3PPf1ZewO%Xt{k3>JAq$esLd zhsP|W@vxpV{cw3EPuy$+c&Hk3H}fm0OnDf_i1@?2uv@$#7fHnh~^@2(OrQJ28N)sMJQLC`50^JQ@N2V93HG+Oh+ES<<9MS z%e|am&HdgyOwCK5_ETmb}9s_K?N3uBfK>O1~t#_eFe8VRXSza6)Yw@ zn6cYlCN0^A`(sZdz7=B`hjcLLjxy}ah$o4{#^m@(aWX|}F;=&{MSk-Q9L~IlGGD`? z|BRrpV`?z=9=!^86J23%%|w#1V>Q0quglG;$>pumJzG)0(-14UatnWUCC=_sao)qGuy`gAO>4xL@ z^;i*BPNMzn$fjITZu+b3%==p!3)hZj=O22ouzG=xM*Sg8xwnPaGW>@QoS4fz?tG@E zPsCBY^B%X&H3R#DPcWoU@Ibv5v#j@FF}H-^Mu$I4(whZkD;I#IgdQ7P@5_{KAI0-| z^Vzjz8LM@?e{gqR4}5Q4%n9ECUgwhc^mUXW%zV0q{O~h}X?dCal72b%WX@JL{Yeoc zP01u^M-d5-EQ6d#FKEnJgy-BtIg>+q%#S9rbcdsG@4T5{KThIbd(%O6%8oNZ(li)E zWYd`?dj&Z;sn{HSf?97VYy=^QgU~cWt?d7n%<{B+`KFA*UPiAgyK(ssi!FTdoY_rp2gO`Lb-sL#y+98-{ zlH2(T7bD1=d!pU=HuRk(ewJz3de2;<(w!eP-ng8n&whCdGA#17y$hp!+~CC*s1 z=_x(d(#TCZ^Bf-wI+PgaL$)=$@>=7nC~t2voaT?iU;}B^(w2t54*a2R8;dY3Gy%IV0! z6W+j+KlI(@&8Swj8Ep$J(Id4KtN$&-PmfFJyKCNDN0%0efDJExzJQls8-u=m?YJS~ z0hVP=U>m%jWA*o2xUOmyo9VELjhnfasoz*fr@uQ$BDyc8kfo9 zTx;3ED?Fy{nZ=2oSc@$peYo_7K%ZN*7NdSz;d6LTy{_(KSxvg^y22-JOmhksnYNT| zS=EhSmNUArBn-P1LU6A6W-N97h|eWsah8}ZvoBnTpO&Uj@cGGYnp95LsNJI5r{3V_ zj-89Y$MMkSu{x96EyK)(Y*@Q*HEO(F$X1FUV~ssYSb2B{(;Fv_iK8`i8Sg&lGiw1; z3X)+rN7jI&ofq+pp8z{oeZm4>G2YTxkJ-bIQ1@RpX3jp1Q%~P~s z-+mv7#$(VD@J*~NgKM)`1kvQkr+Xh{g_XPu)ve~!o4rfKX$ zk}?VL^JX1BQ&~sxDbTqeNyg894!r>aFRYpk%bVMT>F+$bBfs;&KTVH~jvT|*N(H7Z zoyb1aZeV63=1ePIk;G}_3x0bn8}si3yYF|A6`1+6mnj!9aq(8RsaK0Nx9(vV_c<_~ zy=D;b=^(!C=%h0K#Rxa|aJ=#RpsXVjLM={Hucd2vTf;VD<1JOTkg2n)F#>+d_6$q7 zAefWK%CXmiYc~<{o=*PMh&^cu>+HG0N_@&icy`v1vD9m9)gOk`8r#duzrqAT>XkqH? z|5#Ybb6(r@Ac2=!mX=@J#1393tm*ecx_J2q-V6y({3$$*tqKxP8&b;R+~|Cc~5Tt!xYK|9A@{PfFrx)!Y2mem80-i|pOQ zRyKHg4|D(evYJ;X%bDNS<`%1GLZ*^1TR-Cx2FFhm%)UbywtNVGg;=4lZVOI$ITPLA z2{E436TE3yinmt?blBc=7`J65yV?=OLfZ71@_aiM|8fe;Su-0|t<(g%h+i1(;n&z3dJ_M&P1aa7SWq&c5kxW@9E+=OwC7+`pl8uh-Rs-}Oa`QvmdrnT z3`6V9CHO*VB8vz_{^+QH+g)Bw$1Kj`a@39p^5670o@fQWSQ^4!+st8wV~Uw*jRRYA zWG74M^<&~qJK295PvY-aV_E$@9d>NglqvgOz=Mvi?BiEsh`AIEw>Dj3cdjgBU+qO$ zvS>So87r_-o?xym4aX0cYN2OID}VUKJ6?$P1kP!k9{-@!DZFe|hdNP=S+NS-u~mzj z2Ax4;#g~{FVFvqpf6|@(zL+rSI{%%47p;p?;vSes@s1pbf)^cu=(D>K&AfZ@w3@(E zR;a~d^)o@&J%IjQI)q!Jj-ai}Z9t{9q^1hF2-&Hq=#;=+iBn)tt}*7|v5FiMUhaV^W4t~ zQObyrm6Z}QLrCWD^ZTp6`s2LL>wG@*w93mt7k*H{j z1=rhi0HfB}^851Zc%{ZV-jij+kE|X>JHI~SOLjiP0H^C%{&5E{3QXeTv*x2#O9W5V z3ZRFNPh@9j>G5@jnmp=|6jys_2ERUY9{sW&Cycj{BO73V)#_%Nyy!HNOzJAz0em-xsWR(71USC~7Zte~S z#enIoZr&=M+jN0n_*BGl5-0Ktj>=@l!{kH#3`{El=!Yl1fSPah;%HuxTI>p73l-(1ImBc=GzU?)EHa61bOv#ma@ zBSSigz^~abj@xXS!Pf};`uQCf`O8gNywBZ(=dVuWEgMGib;jFx(P$aIfDWz^6!f!rcG^JfWdjx!**vvbPy7`(> z7J%QS&>{0bvQj4yhR@v#vTs&{aiJz&ysE~@^!s>WYBZ)61oE-Rp77>3S%TJ?gHOX; zaPQFJV79j%a%L0(rdtsecft%3vmo5NAD?S2i#Lj%VT4CA8c=(5oY#b=n-9_Xe=^YC zN`}|0{)V(fhL1}f4DM@cAyT!!5tda)No4qJEnN?3Rjxt%+FM~i)A8h$UgmeGC}GT`mP+w7tNNII7!9xuM1~#`<0_5 zPm-l2Kj)c97A9N6h`HNweCS+mq$$lS-+AEkTy5~XdWH<%8_qtQ3FE)-dU8rdVs6C zWx+UeOFkn{TC$uCmmK!Bmb8VQ;Kv7)^HU}_arcw~ydl<{Q(*?!YLzY;dPiH_SJod} zK%KwVltUHzm}wUjk;QXEK+ohd{@fA5fA7sBZxiE*<_kSMSP?3oFcWd5&0W<0(T7Qr z5HvAU;lJ+w5WP}54SnwnptRBg7y5p}Nk!N3&X5)4zTR#K{Pqhp-{-=z=TRVi?t#z; z(#&@MlH!vT7b40v@(XiDi_Y1fVu__c*nq5ID3R}E-~Tj;?j-85P5#@Rl>JX)ne|ir zGgR;nU9#hScaHIr+aLIs6N~x9z9D?s)PtBMAII-v0=IaW!>?S~E?zAzL*s;OJajr0 zzZD$eYofBa-MMsBoHPsfyxGLBc{F3;p%~cTv=xU|P_Rr&gsZ8~*pR11D7|I{rX?O= zXT1C2-#I00`m-pUmvjSWf6D|g@PQ#)u840QLN@E+P|2}3CVbqnV3HY|PGUdV^2+Id zFi>{{l^?&JM{Wif^XHP2&(HO!eW#{cZ_+~|xtWd&$KJ%Cx&#1fuF z2Ym+uA?xA^h%RWw`@ylSh=0Rh*F(9>z7_o7&Un$>h*q*+p%5(>PvRXz(z#Jw0U6P0 zO#(26L`=L3Rgve!I#1H6;qgiI_;w$ztvi|jDEW-e`W1YmP8ZKtf6M-9+{6h3=Catf zxj4d3UZU&1iRaZ%!B11$NPp9rFdP3D5?R_HPki2H0-BEp@<#^MJV1Xi*AsS*!OQyb=iMzBlzg90Hyo|a;f zzxP0R=}(dMFMCmHn>ic!k&Db9{v{^|M8ms71*GZpd?K`3k@-fYEF;3oX};TiwnS$e zH@r**M)qVjrrBM*xXXlmPq~4sjMBN3bsXPfIZbjuI!NL#F5;%E8GbW-!fr@)@KZz6 zc!BN}vhCP!QAWa1@V{2ae%Q)TmuP|aly4|WGCjpD^lm}F-Iv62zh!w@rv{n+RZSGu zoJqJ_Hom_v&8>|0^U~jrTzh5${%Kw%dMNBm>m$x^zy9*v!$g)1mdL_`PlKTLxVqEj z?cstm&W#*XEMT2KqR1W}2XPk`kR`pQ)g>wQ+_PT{|9jbx1YArc2{|pTK9W7{n#&3H%Wz!v{Y#I!wJ2F=z@)vCjbAbZ@YTPiJ!H z(?sW#8?mHf3px0>3om$ohaW`~p-iiRHNi(NdGdg_gu6;^Xy zlO-~G9Cam5EU%{%j=N++#Ku(Kyhv7Zspl(~6`O$N-#S>F5C_{9_@d-UH^z_Mgj(-x z_~+9BJXEFwzj@cdr=Lr?j!`+^JHlDw5q+2IFMcF0ynX{sm93&w~b;>j15zt{w;N?+j} zRZqUwJ(z1QkKtCjH{gSZ9j|DYlEipZiH&ihBxvhpN$=xhlBQFB61%Do*i*fLcV(CI zxteu+-_c22ci#gx^ja%Fcq~AYSNKlItfe}3=&xp9dLENMmTKJMu(M?Lv&H=y>`1J#)h63(9N3<0q07ZSQS^3TbhY=C zIpptBQ`qX=LMDuj#JwZJ$uSpq>{bnNI(pI?qC@s!duER~`BVvnZ%JpowvU`^iDNPM zxafiY9@wKH$ELU2;YeG1_$hQ^B!3)**L^bt_UQ$eRHq=rY)nV*kA&WkW8Gk5{5Hzu)BtE}-|^JdsHQXaotYX{S! zvm|l-OI989QJh|OpW1ARqx-Y^JL~HAQpuWI^mDr^oHLYx)iJ^F_Tq0gaPtw+{5lwS z*BiCRKPID;6ye96)RKb%TO$7Tf9P3TDOw`S3^a;=`|2(cU}#=^G4!kikXb zt$VIxdq*^08WkhtB;&Aje;#hqCm1j~2k#p#5-m3Fz+R_5P_>=|3ja8;7+0~>_8{@w zFv=3GWO1I%MkbfE0kn(PRu5OKWA{JT!qHLfEPqA`9<(|D3)@D*f?LVeqK`{hM!P~D?{(SkfCy$Y+>Fh z8)24OiEG}S1A(!Fj-IxdD9@{NdOW$A${Qh9k_7$6?7>ODitw<_6LxD=1Qx1>kbl25 z=$Q>h{Ph(>@twbJc*bl!Dy^|$Bf1uW*VrleTWnNW&UE~u$nA$#w6-aW=DqrV0MHzMlqkZ{ z&NJc-hM_ni=sjDQJ^@c^8B(Xk`{Db%UhEwbihr|I>EyD5IKVy_hVRZ1_pcFlvs0Sb zyv+q0L^%*pSOn@btY|}vJ#$LG8J`GF~+qR$7e$nUN9TRon@y*S;nfPNtCs z=XHRke-ihfw4PK)_vbd39$`ir!!x%iWmRpC2g+LEt%-!3cX%hdBxfS{32uXa{Z%&a zbsm;jz94SLb3|)4JrfVmC3G=xO7rxb_ zW7tK^)==gJvOIfcj-52ba(6Ty7uf8z0n-#~}L42a&}3x6O3j{NlmvA0-s zeU=uplm5y|j@LTXWuy}e)mnCFTtDU(xCCn}%Ejfo6Uc#6{pi5qLuo$Q2XoJcKq#nU z*ylG)-Jyo%P&*tZ<&Q`A900qWkIBwcE^JzlJu0}Au`#AcMS~o=pxq%A0;-JZbk^T$lFt)2!<(v!m>=Z4Vv2NmcDb7?p|Gns{rCvZDag>HQ*Egq8Kg{7i%@O9S> zcBOGC+?e7GixpI`^H~`ipwy35DO|&mgEVQ{t8nlc5<_Mk@MpdYg#YVS^?}9oN92p9_(T8N9PIOzT{ zgkjIZV1(=)k#vm#*rcWs(aV!;(GY7M(|sN*<<~+-!ay{>AO|mZy3p$0dbn7y0**LU zGSce`Yr>8&#g9YT*X~#B<&ZrN2HG!}=P+&TzeLCyUGd|cHI*cNwjJGHZ$cl`lr!a} zYjNIBU${1I0M&Wj04@4kFy2*(M$RcC$)5(ItZxt7^x{4nzp01RS18kwE1$#nX`Fc; z$rVRtz6863Fno2%mzv2-)9Cy_AScFxoskjW*z*|k4|>7Z+iC3QuLc;c_!T}Jwxh$B z2>v8hL4VE}hxIQ3N*_;y4P$D_#`IdYNK3d#uXWVwMRaJcVU z24kNsr_)8Nn6jTD3``yhuO_r(UXe0y@K(V>!JBhe-vT9%g)YCQgCIA1JE`fmVE#ke ziOa@Z$Wa;!EqiX0^662Sx3`n2Y#+yB_1&EvwC$bysXI7x&ArZ@5ayg%^@NVRC`Zq~ z7z$?tQB*uwo<^LkrDru|>4dNhwn(vxtk^e*?hh3{)BP~|%`B6sEr`aqk#&N zkNXMyI@z~nZ2HRvI5KHFzLa9%D(#19cjmLa<_LT)>Boa24#VhU_eHafN1(o%HSQ_6 zgT?Y+M9xtho)jKp`_2Rk-ID=qb$K*SyVZ~X45-5&I-O*yK`tufAHn^tA6VAl_3%&8 zj+=#sb23wjSLi9=oHSb;Xgo(OwpZZYtBQH6*P-goa)qpBP6SFH7en!?e=Nj&CB}&7 zFtef(JiN$>?|F6w8*59&w>|!$>->>?@Eqv_CL;UGhPw<&x$G7f%#fq+< zb!xjIFiTCR@E9tG2|AN_&f6^*vg{3p9}VW)KYZrPLUg!O^h=ajYoKqrD(GML6S$OS zeAna4cx${duWAm%OMeIAY$=9+dj!7l7A4M%E`Zzq!`O9N6=Qo&!S${G$Qi9F-tSu& zpE3L~KUOdVOLW5Eu1_WAO!x(Ip;P$ZYFqy9PC2N4xguoRZ}Z+HYl-4%MTvtK;nzC* z;nk4sn0Psje;!2nksj7rf$c|_(a>;lpkkRL{yK_o2Lhnd@c?(bP5D-BL;m`y4cVEO zg(LFT^TC~!#INfp=}xfc5i+0o{CMEEON*JZ!D6)XO+sm-&$uE>hSqp(K*!LFY)ynd z8+TNS`!>j8)n+k%`*9uXV#26==m%E(>Mjncf6Tp)58)wF#-!(`GQV@wp2XVq=R05S z!(^{SK40rDlEf-D&T$v3*e%CBOn349FO~S_eUs6vDwRl==kSNxCwOAvC4NJ-Kc5k? zA1V}%!=)B!e7rH6Pu1Qf?8tSzdh|tp=;B~*l~;hj*EVvc<1e{5&x4QAP~(d|e_)P*^53`Yzj`m0ri{tp%7__;cI_>E_(@%+S7wZ2X=A$^7B|W+feYB z)nM0xcrrIv$W?Y!b0wQ%*q<5Lej;iGE1*jYzu-ae#|Kgg`$pEA_=^F_&6xIqTT z4EJT}3JR?CWjH_c#)Z!uw}N*>4Z`%)=cw3N&tCtR$V=Vj`O>~9ocvjlxBh*M`|20t zOHqG5S9>e>P`iT@BVw>KEfIH3wC72-5~k<1jGr<8!y+7qW9l?h$hf*29z9-yYe%T@ z6cx&A=JtX`OjGHZO=JSknc)i0m{C8{^pJdgGOLxfeJueDy@i0dy z=ukmX_n6EUix>u~lhMD!E$@3fd(f1c?To3@o4Ip$Y zwDPx)h1<{Ug{<1egNI6u#`P0MLP%mXZQ8gHD?VrO8*0IP^Y7s>|M3sD#=IOdPk2Dx z+-Ictt*KLEgS}{hnGMHN=Sgd7B0N4f21l5^CS^~8#jf!pc=2W(#&RpBJZB|I89kaz z)(^!M>jbu_>H+p~M1P*1&_PxX4@0U|DcUOvqz>6pfYXV+dci~!_Sn{ymj1Q{_62BzNLGmOLmS6sYi_c4O z>(#RS`Nzqyjv4T*BQAXJ-_zWG@GpKh*_Tfgd@o)ZH(6@mZj>9pHWT8i66L-!pD6p;j$AR4L zkovnz1%=Oy2{DG{VPepr#v=Ie-&O9IDF?{T_c5-%kWWeKk>DO2JCnq z&nxbo^m3X^7c4XdTGOFG=9cY{(~e%`q|uQ!6WQ_b`*nyJ;W_GH(8S~pY3~7iQzlU znfxbTESqJ``+d5K-3QI_vF|i~Tiyjec&bYN4gSn!H=N-~{vA*|MV)t?6Fiz8h$Ga? z@Ys`m$ZF)c&GkAyKUN1e$8Nzd%}=pCWi3-YQy5C%emRS>8LzP4HxAeMXzr+$)G_}{7Q2?ok{DNUyrk8Gq*wtS*F%PpNNY~7r7HxG7Rx8VSNo?8jmudDFt8?_EP#TCz+Op+dDhB zj&@F~Or#}O?4ejM7-NoZgxj8f;I0h8i%Dxx<+ln^s!D>|ZSM3&YJaHRT?Kv74mhc_ z5J&9J5+8SMU_c(?f}Go^WOf4Y3Hs2X_)^+0@Ez@+mP|)xtI>;ndi2GS$xLbE7SeTL z9F^7?hacZg6MpA{KBHRg^waz_Y$;qtt=GQ+SKmi4qeufBQV+4Ub;jWTXcU}D(xiO! zEP-8chJBya!wS`u@Yro>s=odyrkFe+hx#9c+4Af0^r9I0r)C({y-vW^RfalTxXpag z)^|ijGzSztevoUo_mh+5Au#BTBfh#m1@dyf5Xb3)b^)o(&@~qAFT4l0XHThod?d9L zn43er8CAX(1ruyMxJTU%p4Vr~YxIg})%P5_eo!#b8lnQ{#9(gtS!M>4P2;w3IEmf;Ll&` zv_L%q8B7hMck4UZ{IUeL*us-o zvN1UJRXA1!m$1$YWh6c1DU->&Oe8`suGLHqZQ|FV#mQlm*#(HFo{ZP9N5q!ie`LKr>?}^o~ewZujVwo zzd5!a>_MSD9r^Wp_;KuBjJ_3vezs*8mQsnom)675=M}Un;V^v2ekEF75<@DJbK!^8 zLfVv>gTs$V(V(mGFyQZRRy=JJ+w!?TOgyGab2aYJQ|~-joqi*XR!XNb%?U6!VggQI z5C`S?>Y%N5ja99X#}yBELdH!8HhOJ8P_$f&eYRI{)7BPRK^o}lg5A`kZ4CV=n?(kl z90X!=ldS)6Q#`jN1`+~3kjHM@h_Sps^uL#eJ%+k?6P_5752jKDVEtSL)E+J;c0%{?>)tC+TR5MVy?hLt zTQ{+A7iH`>$CbOD%z}Fg*J#BT73TnDMd$3MP8xbspFW92c;Dj;70~yb2sWA(v28mS@fmf)Y0vI^)GI{Rxo_A^p$Ac&kByMv<@9Kj znbLrdx2p4LTk?_UsPUYNrF5%)1axbU;_9*s_{yE7II}w&M`m8Z*(LTYJIRI0FV3I| zGY-(Fbxl-ef#h9p-&_k_I*zfnO5#)i#7micIn|jtNU;$>D?YVD!S~%%KYhg$@Th{Ivck ziD^z_hNXV&LFN(`+;*gTM1U6g@IZ%^yc|cEM+&~))rf}=%hNfj!2$189v<`?7Se)mD~Tm)R(W9Aj zd_B{n*T6jI6uEL~2wDF?2$?Bn!vn2<&3@ou;k zY$SH-xG$FP)5r9kCt!2qJ(;>qN_Bwy|SNG%)eT8kmbd?2KIrDHF@{mpvRag)I7n z2a%|mJ_LM!9u;QdA@q7y53Gx90|Te3YOAti;E6=pXvo0K_Fb>ILToqLk` z&3u4QJwkc>95ETN*cm=GTp&+X%f&x$?hyT$SqGV`EaB|Bc+kGC2u3WnnO*Obm=AGiO z-&0_0rW&!hf;{$omOhvG>_s)>fuMe79ClwH0O>mOaOd7=uzPR=%YXe7{qi|X!q+9S z-I6-8&?<(A#VO?CGr>>vUKRR(UoUVHZ#j7qS=M?ZgG61|A+kP(0{^&-%#*v%?3Ros zfmIvGF2{xB?AC?ECJCuTD! zm$k>l;;?!4#J{19R93{Z=ll9WvB7APk^hBk_3k8_SKSgfjqlHwr>-O~BsSDzj3Qmy zo(wnGZxYczQnY3LCzd**lqq?4Le?}b9P%*?rDhwWmX#hMTkEkqQkCCbbc8(HyPGcz zX3%PzDLQF$LlpbZ80J>w!SI?>kk^v~_xl_2BWKk4gaMssdglnLjS=o~dg)}gL>;|u z9c1&DpBDX6b0oE2IMH3<0Tr(aEL^dV>`}Gh51ejc?mQD-|Kk!qmAA*G;XdfKJ{OiZ zTH*lF04Opm6-`cdn^i_#TyYqGy`FNS=x9$S5TCFlEH47u9 z6%Rx^-!2h_{O-qIE$$TMUe=`1N8-qoIyX>=i-7`1Th`X!j6Vo2L%)O8{EB2QFYoBb z-7in(yVRzUW7{lwjnV=1ohR(ls}+RYp{6LrV1n3qst!)Fct95Y)f7z^>#*_pX(*Df z!QUIQ!RT$Xcx|LKH?c}3D|c9tZ3_>GgU&|7g{-BbasO=v?G4gS>uww%ldDxk2DZ;x z+2+AcDwW5p4Q$GAb@YC0v$;(o%dGj`0w*RGX;xo2;mQLzDB zzADf|_bP~L*f{VF(HB{%4i*K4L^$<06|*ZtYXv`mmH1?EjrhgP_u_>uO0d@NBse^L zz~&h#!sdQ=#LLXbz>5wAP@m%lPm?mp4wbp^c96L!%zv!aIeXOtOT2MGBCzO?Z}n9xEzK&x5zK`}1=_^I%)!11djx zFa4b}j868`a`wJcieX30_+zgkX#4X8{o4MKN8!gwchw4>A1y@-E!v?^gL1VK64qc> zL)NBck{SN>*gWDg^EAlGx5`Q zHR3*YyVz^KKPaQRZo4j6f*0x@%Sl8_0wx%;$P*VaL3mOW89yUlP`8) z>n|4^ZXSt2ou`PQsy3Pq+6Tj41>>&4MWVhPcj5SqF#=Pxjft`XajnpK(6aa?@PlkLzlJJ{^hc7Il*i(>|fXm3Xnt!tOBI7&J#LcWyqm4$463mr-QzRb8f4I|Z}ud?T@b8fUL58sV)! zj@Z6vHqRXL3_G3`kp7?M!Nzy%;6Qp223WjjDxnHIwrC)AKc?h#bXW;#KVZadt>*D9 zx*jZQpA@~lIF_IJZo{3n58(0|UrFAUDbhDrYR<7(bz{P2dUPM02hz?Ktsl6G}3 zIJ~3;*X^qleEd$>&@Yf3%$0(v|L%z!Be&o_Gg)37wO+Wt%EO%T0%K=UI$QUJ@R?g8 zIiKOleOi)v7mViM6HLRdC-Ad=Q+cBOGQQsEC@c}zv3VEnVu2eC3YE28`w14`<_g;r24Y{tEltEMzbG`(W+rBs_h;06$uV;=Q4+V0^j&T-E-= zf8mL!%LV?;e}{18^DaDRH=Wimu7pcz`h=z>B6`B(F5$_<%rrKB{yCV9hxw zeql+2OXQs2+5Cb?&njXaC*;?YOu4-S-T`>cq9eO{#fQ7M{s8;NBM#*xk(I z;uY^Fiqz|Nk-Df^Xcs2*W#&ZUDgA#KY@q|So;uVkz8YiX9^t?YDItTF?DS`hEIs7m zfjf`O@Fm8YeAR}0q!n4z@|`#TDLoB`RcCcxdcnY{(zTQ$8M{Jq<0Qa}VyJPDvfsS;%s4FW@ze_Tt%OKOeT_9v>QX8ixj+ z!wg?}{;4&Z&XKm{uir@X4fPr{S11G+IPWQD^-7pac#$Z%_zwIWJd7q+<}hX3JP;N# zTwN`}6Ar7zuga(6D@hS{4dZA@g*(-Bd2YUaJodF~q1Ro3?_X02@AGzx0!8cj{or&y zL&bn=i>LE8+{7BTAEmh!w=n395^K8lAMQUp0M!hQsrUIt$2IjAvGBenwhh4MW7LOHPP< z1f6&2Y=LLqG?dR*67yX4k%|19Xj`_H;AM{B-9Hrgw%SIh3v}dl=YGS+JbR)Oc!aE# z_hwNZ^Ks34;Z;9uCrbTZi#tB62_Dn^-IE@%?^0{ z1;c&s9x-zsgOy?1up)RMulG;_<@}c@>)$Zuu=jxM) zm0Ls$+OCq|W;5~bG#R$>%uRCFc&d0-##lTSTY!DF?P8-ZQMedIY{Y{o_T)wo-{A5C zKR6F#HI@Nu9@zhuiF>D4wbmAYX6$vFX^^`-{Kiop^7afFnd(JL zuk7b>G3TLY$$NC1@5R@uMDnM_Y5dlV3A}dWXe#S*c_+3*f77rQ|6RM$&#&vL6| zRVbDCrD^eVYx1$=o)UjI>l^AVcfv0>-{7v3MbP?t6u;GM$z24^-mjk~S5NN6_ikEr z)ZlD>wtuD2iTWK!$F-ol=?4fqPwBP#8KPCavYdGv^XnTNU~Fv^`U*i`3zb^YvCVUw z{O>xDhW4)zFJV#s*nfZ>u@Vii;k<3=mO zoWqO3D{cksc$-AdFAasPjiVucMjDa(uN=Bkbc77`Ru(d`6JpO@Vj11P;f#}zDJ{-I zrIb!KdGHywY!4Co_#VUj9A#8IZ9tFqAB^+uwQy7SDVY0j8!?F4N3sKbMAc<>%z2YE z-g-7l^nR;3Y-^baZ^DYnzpcN|?z1ul?aRSnGNViUG0X&&CM^^HysAO&S4+e#!X3C{ zl?u)6uSK(L3gJ`6NJzJHqmQCH$ai=uTKUA7Tc#BgIiU-bg+65OUaRA(QNzi_s|Ud` z!jisfl%vM8_Cjj4IdR{TMC7M8urX`gfLu?eZo+fa&99RBZgQeS&bLCp9)H>!HH3L@ zvlq81r8@0AD`?3Aqw1{MX%-~i3}pkCQ1=tlVUMm3-Ml;mZ%yw2ECUg)@_rzbjS}?zXpZKUAgJI`y4vCztlhfqO(Hn{vN{1gM`wZ8K?R_VEHv zO7es&{mq$JD-XG^K6-BW#thEiae6qmj#Vt#z{CziNb6n+e)%2FB$d(Vom3;-2fD@b zJHTm_dl>n!eG#f81AnY1b{8Yz z!^}K(MPO7Z+b7~0DP`g{Z$0?sD^az0CF;G#LA1K0n~iEY$iCVtk(%F^*p!tk&}!W? zy!2?hD4}LAySzD!4cV29=|blDrRF-&lYIz-6jtF^$ARMSv>SfgTmeV_8!T2jT2!&; zCL3+o$mVUhOY)t@v&*|iuy3C&q0@K}Q=Cjig$1Kook2Ek+y4lU4ca359+)I_Tb81n zrl9}EtJ962CXzjKX2bR0S$N~!YF6oGM0d~rEEWeY!pfRCm=XAo1$tf+H(TiPr3%S- zaC-?l%$P1VE^dZPIq^PEP-1`Jkg(qjtF*3h;`L~Ex_ z0qcXYaHW1Fxbl2_)vS#CN&VRyKjYxP=Nw$0%Rs8~K*)PLlCG*ALH)N(hM2bk1MW{g z-pRFsxX@~Q_IhCFp@`?Qc9KQUTFF6EGXsVQu;C`zo>`wbpNoP|2ob*x$4l3F~} zfw?+TsP^AC7Hj!Yv?q8Jnx1xG*Y1nX`ac?s>&8uBE^j-btWKKip7+FJkvu!3Jc0%Y zei4mkLz>cNh9_TT6P1g#FtaV09h;boMs%gSfFq!k*_^W6Gt0?DL4>WZkVs78TTAlEuc)di%y&DieJeRL)z# zYNvtJB0!5ShUy){s*g{?q!>gjwL#ouVC!ZZt?xS$6=@1ViM^mMYGL<;fQA} zge=#Bl|OcnVPFCp-Z9wiGYgNk7PAcH0c0(o344amfSDI&F#ezrMh$3%qyci&ZsIul zX`D1|8XZC>nNEkbD}I2_#Z){pxrH1!-2&eJ+SF%V6%1TAk*U5Yfgv`rkaTM;xqdzq z+b5330|OQj8OsFG(e1)sS)Yc_cdo&$$8Bj*_fu#+c7>Ii`a{|OZ@>*=X0WXgjP^Ve zWqSva_6rZNCTIp3nKMicPrugDsbx&o-@a!^Ks{ zKyPk7SWFFvwcnqK*XDMKmzqa9^%nhLxn>VwuAvOMs%}s1a?UaH?k_NPLNWP0HW!1Z zma|0{$5W4kru6Go1GHZ6#`4_sX-Tg($c1ZDz4R3JJk<@Y!w<2%_GU7orJYzzHiQ!O zWR!_s08Ijabh_0vtnkfXtu%u892zC)>Y*fZlOtYftr0&7G{$~a)A3$X74tYZQyi`% zbTf9lF=avjV25pSOW6_OHrOd@sXECzPlZ*FIH8PhJWjC0j7;*o<2O5XY5^0~ePYqK zWogeP8#?O!A*hY(V2vBp;nN;>sLHN@j-;vhEz1Z0`k|Q5Hp|f& zQ0+Kb=tF*SDGWPj4`!xM{IKtab@i^|-J(@5DbXBum|5StD#{RlW5*ILNXWNd_A{i5 z-LE}J7IZDf2YSCSdfHta@1sfmHI~9R#RagTBMO2JtH9CPFNB~b^Q^hVmiLpQike+G zW@HS=G^aUUv$SP)iT&x@oNzMHRF;wYiSNw)%<`{zuVy_~rP$alAd1kd)CYDMU>1aiR?mTX4BLXn$jSp zA`(&E=Q<%&Rw~ITp{y@uXQj~Z{{08fc-`kd=Q^Ly`+Yi|D6QC!I`bY0qet`t_l4B~EJV^A6U z_Wv2|b}jsK>l~DrGraTgI$l~Wo@pj!LCEmkrLVHYInMoSOsOzc61i|HbpN;t-dmDE zNH)Qf{U<=ipm;Ji>$crmDLKC^eUR;$>cQfE*=l`cnyB-9f+I49!Vxzi2b3o6QE?f z8cdF7;FQnKAbY`#FEE=9rgv)4Y1uY#(+C%=o$r@s<~|la=SJbyvS}#s_{rwHULrZR zB%X~q^@AL_Ih`G=_QbsfDI({phS}7X{AJZw>MQSGXkB~Lo9jaZsO|dj*EJ4#rlMmY<~ZC zh`!@1<}d`>WRw8y>GkCI=|Q;srv;`2(HVD;~X$WKxxkJedBLF*p}W z`;b0dalum3er6VU#oVK}0zbg2$47YE=2QT~4$?et3C(LdM>hwRQ;Vt;`i5GA(w~|Z?KhaTI#$04_ zSj(nTX1%ryOt*y!^UqGj@g@{*E3RdKgaC5y$t&pGc@aD=U8F_rJ3+gP(UlVz4O*at zzaRF{Nyk)Zq{C9Gv8xZ0?TjR4r$pwN(8?c9G~!wYWmJ7^33hBWL$!O!c*{D2IG=b# z-rl^!$S5aT^&pE)yf74m?i#RE*@{ALOwZ2Z7)R7CFLKndrN|t zFIh|P45e&kh7P>BHXdZH)3wI?;XQZp+6FZ~x$PTvo_G!eUhjld9KwdYS_{-q3m?A-Ks8@2ZZ-D_y#3NecI{W- zlYceikJ4Z~rn`bKKAtMkylTV+D-W@lIaT!J$?=~DC*kFRr^qPBbO>6u94pS4;VJK{ z_+idf{Fe7wQsMbZnC1JP6cHtIbAP6B_@=%EqR6WU4F1zDk|7d=Q`W|BL>O_ zihSp|yI7kujHlN=7L5NjLg?}m)-Yoajkuh`S5CL*F4>RJ{i+rX@A`o^K6ul6M!k5@ z;qxeOro=Ulp2AN5TjZ2@e^YttAx`rh&cBTqXLrl}GH4y|BI}m?z#$)W=;iuU5f)P{ z{Fj?83Hi5?__RyG#5E0uuZ^W{Szp=q;n|Yd%X6S$wjahx_n@nP6&q$Zgk5@>2+L+X zCS^xoVC)blq7pru`j7HQrW(wW#XXGT=>kc~<2uM~Rp1ZvyKw0nPZAjvPlH82d9kb! zD6g)=+EHh5#QINQZR1YULOS56+fi(89soL~`|Xa-*up#=)%gs~wfw|OE2x#tqI#i5 z;2?~q=95NKs|PXklUxb(r=Lh|v3zNYelFOxiygqcY`Vu{B~^4TLL}LE|LI%urT;s) za(#d#v&xkn@KdGVeIyd2gR$UY_lTIC_XYh+2Uz}>7(6n33mZA73h4PjxFwy2@xdxm z?W^5%W5E!5^3yy?z2bY}Xj!byv`gLW%wGj`J3LZ!8|1*9S$6bqfI#{=Z|7sDhbm?=kTXA>^v&>pR!;BWtTID#}=_h&~|E-PY+9u{|FJBNxHlgshbA0l<6iB-0df0Sul27ne#KnWR1kfPBch>CD`y| zb-jrYl_oJPe$Mv4(nIIKTctDpIFXg&J(ORhHIe!6jYL@{Q>b@bz%umYaeAf;vD@-l zqW8{|8oPSY>0kbX=SwHkBw0JCOYneeH$thb-a2^G69>b*{2--a4j=Y%g5Osa!obOCxTzzV-n?T+>+Ji`y4)~ozCo31+sO04mxu7uSkVK|MQ(gc6dYes2bKdK zly>)yf|j&Z#P)C)zbyb5++a&5>s4bw!mf+!Td6sY<*}f@-yB|y zSq~4);_0}!S^VqI21)O~m!aTuJ}LI;OJy%)GRN9NSn*~D?~t2e~4 zov{*yJ7Y-B{e_b38VRJVZnBGwSY>x*RD^^^^Z`~d5Nt2JXI)n~?ry7RXSZb$*SHp_ zuH8Y;=AXvCwv#Z{#(*`CT!!h(2rf6$g%EpHUXs^9w`xW}Oc4;(<09WgyvzLc*qcon zCd)NuCkxR!@o=Jb3Y-@m;6YkH**Z5l^ie3GSEimLov&5ksr*K~^l3WJJ}kk%S?f{9 zr5RHv9!CGP=b{4z_?v)ll4&jX$w9q2bh?r)eP(Gb^;y*g_Z}qkgsLxCQ*{ptmYxLb zIvJ6$}Ss8?~Ki+Z$tJ%!4+9>bebZ%CEj4l5F0#}980pk zr{k?aYqopvY+;q_Vq)OnC^B}VNUn)9RhagWnpqdnpNdhq`ELXFxZcJkzdI$N4;4wf z{(c@j<37%;J;*n=S+ZkfJ3M_?#hmuV;Y5Qn?!4v;ZXM%FZac<9@T@&V-py3(WVK?i z|9;b(H_y@|FWQMT&>qe2CScIe+iZElZeHvd3i9WAV?}xkds)$hO4r*U#eOc;?sJJ) zk5J@o1;>Q&{!|Fwa|8d@=)m7^jS^wcb;<6YTQH_=1U(b*7k9nXgv8Sp&|$S73O@XT zuGCz3ANLnGP7VUa^&2IfE7B!9qZoS~Ih8+1I!Shy?1C4450-}QQkK}2DPZZMW|XlA zNB8mb`MY0gM9x5!B$})7^Bcn1aNo!L^9~QN|01Q&P38+z2d6>&T~~b6E(@+pm%-nS zM@1*ve!F6}g1`Fg&PRT?hHHP{5Y>iaJY$u^o-OT1GriVO`G8~e-OE=Pnl=SfwkTuz zWlMTBHw9As(K{8CT^rq9&}(W=8t()r>fo-&{) zGi;W)M=oH;SAS=NwYB+{D?f4O9#^jG(VKP-O@-Urov6zn|58s2J$x0`AkKTw;gI2P zm@r!n)o0eg>EmUv`>qd-h$(}Zpe$^du0*qCsZ=v+rgZeY>C(|df6~R(W2NcQ2kD9L zYUJc$Wmp-ph+b+H-P1=!|Il|6{G>5f68~Q~o2fRCzn^N#PX>&@2;FDczTbelO_OE) z`dP8xCr`5XZdI5smt;5A<`>CUy~yq)InWicHOw_CkWaZBjKe<6p-+ReVY$-{`ub@e zJ-@vlotD@N^W3LPn#J`?I14QO#8f^Yp#!W`w$M-ar_mjAW9g#Ai&QA8rvti7sols} zD5m?cEhL&P-^ytNiG)>Wda-u(0aPMy2ZPrS;n8z;a+Q@~U?>>}1NXdT`AIwQ->~D5 zIcXqQm_vEN(fu&|P8>WqITZ&CE5$wY=7~AR*EnzQKk!_e3ezIJ*xeLEs&TN0j;L*> zZr-jC@$fYE4jauol6%na=u_t2b%?LKqD&osDDz#ZPOyB2AxA}ft|H%wE>8t~{3Z}n zT^ixz*a&OpnZ(!nmW*pPUlvGj7KN?_hUR6H24D6#wGG2tC!>MG&eHhuNLOnTqGLC za%|ijb*`D1DtLD!iZ0@8_z;kc@7!jwlJBzICt4jBo*O_kms(-YGBxRfE2T7}xrDZF zt);UERnq<+Y^m7MU}0Cb;LgbDFs4wM2d0FAv!w!+_?wZyHXS-&^)7KP?;tQc4sGjq zO3VzN33k#@Z1H^oFU%ZpT!0LB^xlI~u_I7&vIFi8Q030mOK{-fV7kxhn;`psGIjD#Jz!JIdU!Y{31;11 z3FEpCvI3nyx3odZ+5OlP0JY4c>One=#{Y96{m2V??Z4%oeo!`^x#&h zA^G-g8-LS17%rOh;6#}NXsLwqp&#l13F16y{ z9w_m(TPq+&TMt5x0hHMd!#sO!u?r}R?&b%;QnRV_%cv3TNl=Zj_^S^48$20hu8xJB z;vI)xr!_ZocR|au<~&L>9QB;-;LpwPVDT)lW4PIjZqgvcT-QdTykb0UKbsGIlEe(@5?R`Qcm{4Bxc~!K55?CV zxAB8gB(54-gbR%S+9fOKb2s%UzI$Om9=<_5qYaIrW@kjG%Xu|c-mSxLeBRBsO|zy} z%GaR!pPaBf;58Jhk3_4Rz0kYb4x+za6ToXK1dY-KpU3m5pLsNW*QW@NX*=PmmR?+g zPNOF;t%9{HeNWu?An&SSO3NbZC0@XV9~+;#Q<|0Wj(Cl`9~_`*ZfWQG?VEA}CC3)AtMbp=`5 zp#e6p-r_tpHGW!8j{6Q=PF@~X=dG0Uj$H|OrR*-rxHb&dt~|lsxv0}jy&>>-#c<5B zE+;Lw3!!c8KJIkGhAZTY&aU7dqBDOP?h0=uFVYnF=(iit|IZN`ZJ13DofqemFjMdw zvWD*_drJMPXQR@}OxCql6%vOLmX(u z{dBhTPcT!v)eA07GvwFJ-{DWg7#gvA6Fgn~72dCMl%7)fN6%#)hPJ)MaO9UPt^Br$ z_+~$W)Py&1w7d$w8PA0u?I!Haxj5RsCzwh)>OmoM63z)op#2ilFzR0k%uX`m~6vD|1j4aQDk1L1s z%%LJP-DEIZ=U#!SU5n7qBnQ%}lFBq)=+9a_NXEg>H!Hi_bVfvjwv|DstC~_Ty)3VO6p$5q?KRp~$ z-vgKIyn#k)Z-o3E*CaM>P&{w}$7dG3IGSC6S$_LZQmNU5^?Yqn3ErDGl&+UsLA-8_;*Lx6 z;bYc)*0qJ;YV-T>ZlgS1xXhbw!YEjzsX*hOEybxL4x_i96WwE?!C%7^=&)6zX1-C- z_lyF~xT-H4IGruQz!XXO+IP5a>mKZ@UpWZ(znILIKGwr{;U$i1RYtlc3FIS!iFryk zjtMrSZ;SiU@G~>eb&hz(^`D1^-r_sCsem1QHy_8}&?egb?+fc4gKWAw zNSLD91O9`RP}eDfOWrH-?qz%N+({*V>}@hrIW&cLZx`^KOby=p2>77?Uuc^; zoo`(H3=KlH`I&`l_yU!Bl>F(zQ$NKy_W5f3eK!}!Gz8G2-)yL5fGS%XV~7((UZ{Jk z0^2tzoDDIUjZfy>AZfHjusF4mob25PEqb1^@Aj98R^S&wfBOTWPEQVXQgYG1>m920 zEM@);OG)?m6&Mu~iA(ifVb2Ozvawy4y`Gdv!{z!i`>MZ~WSouBE3N3gjx@BAuHhc$ z7Feg_4(oNzxv$P=29krIUa$?``Vl@lY6m}=sYFd057J(mL!stUU+#OxgIw*EEm>mb zio3$LVDhLGcEoWEu^%>u@AUgZZuf7;#C3CV#Fmlpvu`3%&5mW?^tB*+62P6E_3Yg0 zPH2u;0J8!n^V`)$c4uPM_{#SZZgGG^R2;&MzZ(3&p#)wLvw>P}w502j+~C*=7r}aO zwO!^4DgHJ(h>v_+a9Yo7=6S9S_7^D8;&Z$4=}|epdQk~_CckGcRUEg4wZYN5(Ga*I zg$?%KPSPH7@ONO)?thgi9ykuFI=Q51Xa=g9X5vJX6>uZG03ADgxL@-Io|C?kcXu0z z43f#v6&gD4Bffj;X>wtE1Kt?BTFjdFqs125ET=M* z{9YaxiTHkF0-AiA$x;d;aMgEL zeCz52`T4zQ&f7G)FYN<5dU;AR?4x1fstNc$q8#nM0xs)0%lnP-;i=|UQ1R81E{%;B zdoO+YN0AfLtT%#R`96@pxN(Wnl zeER#@q@jB)Z*tbc1qtzVg8u_Jon^~^L_8%+KR?IxBe$62*=`}%>U*iVvnlO3phl_= z$*}mb|6oM=AM!YX>W|E9mJ1v&bMr4QzH@#~V%m<3T$txb+`BUf4X4 ze@RZ_`Zn_5H|sjC`+XTdUb29neJttqGmWHQ@)1d!eHbQtoj}{)?jSsm!M(;R633-C z$@0=r!k+1Xob@Pt-ciGT?@tqR@1L=%DTACz@5df#HxtA8xy&rh4o7FdWhb|4q5eP< zI(_@Gu-CcV7I|nkyd4eprO%9Q0l3JYveC>JdNa^#`to%g=IK=tt&rp(H(P4 zCX%VPQ_*zpU{d4jPM+?u#MXZ0Z1%x%Y)P+HGVw(e@@*=lZ@lO+`k5noApKDxBTtf? z_)>VgG?p|bPQ=+fk}b25!#s}@&=?^y`RX<>yRiz;yscVjSfRpKI7VUd%Fn>#qH)id zA$($^7kPCtl|K1y!lDrT-GZTnGu|IEpnu_^{t#{yTX%pY0fLH(PY}IP~kmf*pmdMXw(}a&sRX_cg&+ojEwazYL!& zT*WSjMeJXb4MvF!j$uFNfcve5EThPP@4T`KWae4HQIWwk)B6jis6EEcK4#>B)o`o} zFNbljwP@+!VYJU!BkFa!QW&c&p6|Zpi!+pMWo1L2@VsYEWgF)_=db$w;mu(Z9%K@N z%?WeJz35a_>J~c^nJMs~S3b@$R^TpWyZOD%C9LVL8`Q2H&d;21!I&1q(wz$&>8;da z66Y~c_b;mpqvCjFZf5O%$>Ud!uEN4_-fO%JOVonet;Qx$$u{vm1%* zLB1@=^z#DcWftVS%~JGoyUAL17Lx0S$KeqFnFya&GO6DxFmlpnvqKalW<&BAS04kb zKQ@WXr(E)6$ZH(Nqn**m)y{YZmCA9Ky1>8NhlysHVgKRUmTU*;CbZH{2w;lll z&syU(-5qT4qw7$9ZZCd|p2D{O=>rxYeTlx`3)k0X>=%L z=BGktsWZH`x`0;vR52lz`T+<@k~AKu8~T!A^6rHk%XFiR$*p=;kX2g#nds;m`JX`26@T zdDVXq4Z1TDeO6m8BihBshm-MRsQ& zl-nXtA69i>*)*dy({Ws1YLf+kuy=3g-0o zpgt~&g5G8g^5*Utw7h?v=6F7)gAAhSl+~R=>TX%;F#kR^nJO>sMNd<5~0y){axKV}?I*n5e_;hx}q%U40;Mu@Tj{ z5k?PmX-OAEU!~bq*)Z+DtNx8A`x-euPZQSk(n>T`MWExEq-}kZxGoWPjRD3LU<5wKB z#jGiNnU%qRcsx*JomlTCdMva2}yIYy>7ZvLD_!zySlM5@3 zzQ!hH@o3;U9*wu$!_^A0{E}N4n|#NX8{5>=sn(S=Ab365_q&4m+Pl$!plGsF+}CgR zx+wOO9*}=Kvxx1V4Dgy)0*P7M;a^=m#K^Y7K*Ji6@x&ZH`ou!Hc+Q^WV@msb&%jxC zhR_IUDusvo4Gv)tRdkUSzxP7phbXwKrG(_sVW_kIB{^|$ zCzgy^LatPtRyX61w z6DG|2RBDHx!KKRv`x=JBec2Uoug9M`2R~r9%NJtE#U~KH&I+uG-r&KsW~@jU4lZp2 z=#3ODTC}bfg`jZowljwTFZx5aVa$cz><&iVM6&P^s;J}1fA$6 zZ|rQylIk9k>Gr2?3YBl5r--gvBW( zu^ zh9yH#^}r3%acmEH74eIhZi%N4`#ON^wRc!?GDP$*N8<50pUBBoaYFO`WcU$V3AXu1 z$*%lMWWeHccAKQl_;cU4H$zCtgYy=SAei3;motT2EsHb zZtSsx@b#%s5m8YL2Zo|cg#sBHCnsqvccP0VsgPHFpS6g&fTw*$vDko5!V8^kLiQeG z(8=(n&OddydqOx$gM96*=10=s3yf&`!!+VGOow_J=D`^rjgC>S%=()q&fV4*A1pp< zcU8=~TQ`*or{)^c2$SKZpP(L&RQ-Wn9~DsP@1s(GxwFJ(K_z{kbBXr5I1=S}K8D~z^QtrZgmz^$7clJHt7mBrZ$mtx8h491{uM9 zsUFqcSPeUr^o1Eo)e!q}3&gziC&l)I!M1lLRnd{5)kc{l@JlnsUnszx8@zE+;sKa+ zTg+pPltJ7qo{?57(y=9p*o}Wn9?l&D>You#xxJ%RKlf9YQ?WR8ViNn{b`Ls3YnZ0e zR+QH~MZU^hfP%i|@L5-hZj`Bo+le}och`op)^}Do;$9Xs4gE~?SLH~qPo2k%2Df2t z?{IW+s$rwN4IwvrGd@U3pai z-$Iz8T?9Y#G`PpQ^RQ6mv&329F7bMsO@@m7jXcF{zWCJ*sb7(ty|qV!OddT0 zCVOUx{3{n`w&skmdC&;jw($zRc2$8F`G^^vf458KE=eLytHy%Ac_zo_kl{F67&odA-PA_Ww(`;NtDg!VR;I!mPRLFLXCX>42Npk4mTc%-4VrzDaO9qe%qO@Lp2@agW7B8bQ7!|>xDzsTwyHAtExE!} zKRhP?O}frz)*iszp63{GEd(Y1O!>;)<&v8XKS|1;wJ>e*WD=->!n}#elEN*cu;7^k zj#hN1VH;nP{Gi+HmTMdwlab@jw-u=A^0m#^&bF=`jGx|$j?5@2A2p{;@UhV2 z0l%H$;Usm4C?7#B;V8X!-5%n!6mhFle@yaz38&;_>1#1NJ8;Yy_!xZ>dWN`i_1Y*n zZEOZPR`1~LC@s-B_X8KC>XNA~Z;8%z8|v}2FWmwM_@S21yy?~{T;H7pP51Lbt;v9% z8@mc#-q$3Brpu|_{50xxr3&7Oow;pUv-p|JNjz-jA(AphmtLK&2JiJ{xaXTR?&zq& zn|m^-^%FU2xg!DsE;K{^d_9Tow4WGSS&fOSuEOKxhv>bN>Tt$+FU{`w!S0gBWbVfY z!r?8a!9FGwE*j~xUThLOJNX2?^m+&UJ)TH%7rA5lcSElF)P&zGUCehav%&OgHyAM9 zi(YPbesJqf8mBuL)rXMQwn_l4+Fk;Y!jCZ zRx1lOF~wDTK9U&&ZBWY*=$>1Y&YyS*T>)sQ+-Z1oGXwO0XTi748FcT~18{P}A#S{1 zok!l)z(+xZNDLe?9{XbPqwmqGG00a1_m@ z-RK@xMYQx!QKu|9dhg|YI_q6Dl}QX0?%!TO)^B;vyvEm&Iir)rj*%xXxG%ahqF2%6 z?+s8qN&^*3)tH~W4t~jv#*B4iV7{gbu^Y}IZsBga>Y$9s@Ov(dYPF*mb1u^Ax+OHM zxdEKIdcp2qgSqdt3|i=|Dot2Dg)SQP73P}H zl0m3su?vTf&A`epvoSJg8s@c|;I*Zb@XgmF!d56D{#l0Lr>e*Y-SOo9?;9{lX%P;+ zbeT=^nI_T6-;7zo1GvbB#?#*)vEYbHcy~FM%ZIhC@;Oe z^pl-rdL-T&6oG!xcDT~NKixOB1#Vs3OXZHF)06c#z)R~W6k4S)r`9}?TeqsjZPZhU z-Z7Y)eyb4vYy;G6+erJFnX=4j?&L#Y9qTTzf;*ezARx;e9LE^oZiV?`*RdJ*?7RYf zMxBDfnU1Il3WVzR#v1QasMWj(^rEW4PdyhC4lI#G$19*_rhwj;f03Q~gXw<*mV<^t zv2a4t_|?us)y2; z+r{4KlJFXW0teC6pM4}tbU%QN;sv;$G>%&R*n~%CE~jVuXu#aLQJ8*82@x{svaqZ4 za9uThHgFLgD>s_9Ni6wCyMu)FypgOLGYzWb9N3C!`Ai5k#HpLM(>1%o$kwJXP^leE z0s|i5Pmyuh6Q5xlK7A$)Ic$L1nFds0X&e4|mWzvJ#^cK|uZT^+Y*_TCiiJ7s!#~j) zD0}-9RrqhRxQ8^r@eAVd>waYv*6Q+w8~RCRO}i|t5IIUZFUFFtUIX#$`vs`6Nu7BZ z5L%g>NgRIsg9FldSpBey>5Vt#H&c(Wz=&Xa*nSZ@{g_VuD4}axIQwqbN(P=*p-Q&L zz-56kY7D4HC8B^q527UphLG1og*En9 zn2dHT{`9nhM4t#!QKv~ZT^F-vL-%6M(OTv{emeD05ZOsrm*KWWJyaoNB0cde8aIAd z<&!gv;pO2kToviM(;mEqwiZb+2$*>x+KtHX=M^8ofz6N z!IhqhdI95A-Ke3l$eh<`m&7M2@EtA^3}5a`mmTv!8>b=kQt}fR>lh$BzI~WnnN$X| zU^sM+)y3kMIxNJbg*EqB6Z513QaNlg7L=Q!?)q49uKx>UvPaNYYNC(CU>m)WF^U-4 zZ^U`6(J(WtkgORmddOFnimtcw;I+;RAL0>Ea^45yQ~J?$c(b2bUN3=&oEVT=JY%yu z2XOnDo54Lc7d6#NXw9fyG}=mmj&ew(FApz8#U1&A|G;cAVWhR>n>eF9^n40m7;V8# z20g?E#~5Mw+OpDFW%{r_ElRR_a1%cN(T%Bh3+bREQFLnE2iTiiD(WFuGMzi>U?{^AS^TwDRI#!q0u!Q-%M zD^J&q)Dk4!O7AM{I30^W~K^I z@$fLn^wxu$$D6RX$|GXAs~Qeit;VkwZ|)1scq5Wr>xXq zlgV&8WsL{T+u%r-d>aP$l1{P*I2t#+JRyv5XuzGjH`66gR*U)HdRCHGg=N;$P&IKr zn;zJKv+CL;?jNHua^)Vj$>=d=jZwxeV>HA}aE;(Hr-OOCye|-)Uu5BzN;p!XOylO< z6Au0rxvuA)CXU_~=MK9I@z6Uc|(3&s6(xZRsgx+MJC zdLrK}GW|aG;z=qJ2x1Rm#(%zKUP=zMCIVbGzlpXzNwiK|22H9v=;1!eRDVtk|KnZ5 zLsTNUi&ZG>u^A~nx8?$^`Bor4cPFXp4JE1E#UWDvZ$Ids#8ByM`WdZ4GVzq^V%E9; z3e0=Flg>VS0TlM_LxnGue5l7V{P8yfg2#_#-5NenDfW)Dd)4sUiT&t^Xg@KGcnmD~ za_RoyQ1W`-QjD)e($O#(W~DWP+zVNJ9nwT3X=gCqIT1e^&*Pzq zG0bV+8EUbjf_9bk<`>Os`ETv5#V==Mpkh%ucia}lFAtMQ1LG!26Jz#C*L<V$CP zT1`-qF7fz`2CugAn?YJUqG2OBxBmpzU$?-ZSdlm4)xiE3sL_BHU1*ydj>eKpEb{7C znELD*488N3KL3+M(|^T)CM}~&Qrd*4!8e3eD=y%x?-x;V&3-|0@(OEC{taGeO8B?$?F$zed0?o8z=sS}i;dt)MF%7U4Rz0c?DQHdgDnvAr|XvAVj7ZE79I z_TMXI>RFzcyrGnB)sRT~?di>~U_b1q{{==&8o*VTh2e(VC#h#%9G;!#2N%@haIo$M z9F;InGVh9km?O=>>m#$#W^ymqGAj=ht#{J?e{86E*m=7>`#4uP@`s=PHw`;<$H5J+ zsdVuD0Fi~7$3|7&=cZS8laqtP1ee?v@Z1wGM zau3;$clT-7PD&#>rcwWAt<*8b6;?hO$2FU*%J%LsDDz26;nEL-fJtK6Pv>K38ZUN0 zYm3mfZ&s<%u9f(>{~Z<{kA#M*V?aba=`NZoM5nEx_KkXUU$qLP|HbRQ;>%gc@v-(8Q0 zG*KOFzKx_s59`S><7Jq)ARLFUROansc1I&356*_A()50ZXy>j47->5M>O$?&f6ivE z^COn;UOf@wmXD!t4dPhsl{swe5p8}iW;W*jC}VLl#@OfgaD4N$7B|2U{5Y{rQq?~L z_A9*-BKA26p0bqw^&zeR58m30-Cm|14pUVl!sM%;6@xK5)B$9A;E>6z66K zqseg}TvJ%d;xCr7D_Nq8rEN4inQ5|ccL_edqDA@~F_qlCbc4N9H)5TS6wo{V6B}9{ zK|TkLCmFw(5H?~ZBp+TzaudVREb9P2Vcdwn-pVk$K0|S6_6A|us3qL*`~&`CL?Jd- zDO2T&S~OU0g5!uPcHOk&FU_^6(clKGO69n(*o^i`PQr$rWBH(C=lR>s@~m%Q38Yu{ z1*3rDT{`$F=FXL&W3oi9hm|^SSs%mSei@CAoWJ5~ z<#+6Fj|JbF`3VOZs!_QoGwAG?@wg=JBJF;91s|?($EOc$FjXUq{EVP*=vW-IjQxwo zRuft1yGZ)$n=_Rz7{Yh8#?xba93cn>;h}|!ENc8JNS6F3B;sG&f46&~nSzw%+@Ce_=er4(A2e2Tf5E?~aOMOELe!H8Kx?@iyH8vQyLvw0r*1fBZW%8;(XD4+w^=eSVwoj z*?cz;geWp>{6X7<6mz_mzJa``vBNW&d)ZBQU%q;jIcE;Pz(j2`QwUumEhu}-7#w+V5(VJ?n2-oZRGudRW#JAD5 z*C}|?xB~BvSKx;wO*o)!Cx3Kg3f_#rBawD~g_!RhaJ-*3AGh~5CO$n$oJ~b8;M)ea zueph-lsKUkzl*MNCqZ^t3%hxKpmg=rtCY_9jYfYqW9FebM9XXpB;*_wyH;8(XKEju zxKEYM|5gDj6(8Fj-1HXyz0jn4Z^;8J&mf~`B}2@Bi6C5zhNo{U;LMVNuwcw-kc;dj zm$VIVk;t()KIs?>R+`AROz31O)e-0>ONo5BH#;=qk)X8WgWZ}(uY^P5jQiwQXSUNL zO;|VMf!*M3Uj?rU6*9vq6<&88fMMmLlkWU5)c<({6lxQ;yGT^Lo z{$%2KwU{{U7tcvGxy-3U?D!lRBRo@IibeyTkvDDW@XqWp8zb@+vy9J*jyW&MuJZ!u zo5|9c-^wiRxGPLPodUO->&UVphV+NQ8OZo7-V<8x$LG_FnZxj0N$YOCylH3O8EMZ$ygGwepFUT3;}UJJVq4}!Qal%&SWz_Ncb@G@H$X4Rj7 zPv&YcWw8e{F$j}Pi8Z6+lf!Y(=Y!b$;v6wsb^vNV%%DrIyr61@F|=~aUG}UvRBDuv zB8{y)BRzQ|M`~7~D?MsrCw)ICMf7x~vi&a}L8SE;Sg6s+P9zNx{!1dTik*YFb8mq8 zi)@|qXPH~q5$1MBR$9Nap7y(7BdvU;EUjB^EA?&tLeJQ4lic2OmwoMffrT7Wk-X79 zN*X5_2#?Q%z)FKj^jPNxdcCOytK=?{Mepx}(pix^kUj@KTXsONF*%@WJ$S;@7j$3YsaMlIq7f~vX(b@^Ni0ZN;N9lZ&B(>+>xDqfuREgObYmu>`I^HdoB zZxqf?h+*y}V0(G;0gNj60Vf(~0DWKsb4=q|+^55^IC;0o?XrNG;+lVN&m}n0a1-wI z_e1+9mRq{w{Z!$F5*3#pg0%Y?*D|+YFEDe$!!b@D~4?|GfHb@kCK=uB^LfM z5xed!1jR3R?6!rCgM>zrsolScxjkFJ2B}YiFu!jl+5S|>ztqgkOgqWTO{awTdxoGi z;tUHGdt6>|Ma=!*5OOwA3FH@qOV-`{#fr~LaYLRTy4_t=GNErV(QL_+%pLZHton0- zc1RRySNsC<-BW?KWZQFt*jExm*QxZ!?JumxXcjwgr!Q4FFo!?%aO1yOa`KvT{zR-cY$10?v(U5Uc|obsfPaZdn7aL+KG90QR%sE9b$PX zo+|6N(a1tAsdiKj?OI+TnXDQqF)drdk{=tBe{}<3<=4Zdzb_StT>0BVv44cH!DyUt zP5+)Cq#q=k=nJC$b1>-aJOP6~ZGyb1dU$wZj@_=*E;wDAM#^0>!E9JCIe8)yMd}Ae z)DDAnk~Xs4B9^8_=hLJchiO+t2c5Fqn9j94L}wNjz(QXsRJLv;FIR0OgP%?Yk3se9 znfFMr9p4GV`bOd3&DU|u&E3pICK~bw?59qFt&sHJIOg{Ks}O%mS^8(jP}b+;2Kcuo zl!j=xuy-4;K$%|`EdSuaM(eJCtA1O=&lygdZkC}z@Mwts5CW52b4zdj98Eru6k(W8 ztgtFAUNUX`ZE#Dx26D&Vi1#5b;M!+Dd9)%9ghiQ9H!>gIJG8@TizBEUX2nw%s?aSH z9cgCzGU}rEANWm+qva~j0@ocr0rqAWWiXRX63!((9n z&qB$9>s@Ss$7)HzK9O6xP=zgwzW|qu!r+w6a31gDNuG2>vG)8KOg-AV)GAk*TzWZ` z%(X4DJEm65oID5N8U3BYi{^XG_wH*!qb#^IW>ynBCI3noN<)~Zu?!=1-%3nfrjrE`?Se~N77dRi z|KsSq<8uDqKi-a{LZPH&rHInF&vkB5lu9;*G>uTQLwr!uP$CTtrA1q4-RC+LrI1}I zvLZ5*m6gKp{r&x$|9a%S&$+JG>-nUYcOIZ4Pqo3onG>khpJ#a2=so-t8`45=eNsB_ z9hM%y1#u+>ENR|;7+XFP^+-Q7RGE(2Lkh?S%VBUuw+I~k2Z#snrQ*)($HZaR_lRYG z2hk}>b>QeIyd9oI3x2F77FgVm_G{>7MDw)3uxcmouCh#(VGtmyN`p;T@AKt8z26`C@~(}~MPwD+Ma3@Hr6e-B1Ly5&d=YWIR(<1y6K zSwj15drfYsWr9{!DCwUfO-}rG6Tw1_I#~_m_go*qMe}DcVfFyJZq6K9sTf7yE<6R! zXYa%5jsbLscQqV8a3B2o$CF3n!r+3&03sBYVeleTc<-qO=iGJ)|7QmC)whgjXZIZP z&bppd)^Ea)!S49uw$O*RSw<^&9e~jGNXSq74qn#jDAN0nYL%3;XL{RV*zpkPzIg$j zSgs~=w*A3i&s607t^=Kt4r?{4Ad$_7)Q36HRI0`nIJuTb9JQqTn#a?PE;jVu8x5L$ z$bufsk%q4x7GVFeO7Ka~qx~g`P7Lx^iR#K2XquoyW@Q**W*=zegu&Z(Xr) zCU=8-a42utvYjq^Tu8lSOsNVOQJc-@iN^!M%Q-z#G)eXgkt0tfxkK(Shn013C@m8_ z@6Mwd@<%D#lMPwF4C$4n+3@lAc4p;c0w?!(!x6`Hc>8-M2uUPS!7ql^_Xgs``2ssj z=(Hwoo`I^b_tUP$o^;mN7ZA5;Jjq;dO|P%lq&L$h(2bw{pj7`X9TU+{yxb*$#@y1R zA5|xd&EoIVH~qK5q0j4Rf`2Amx%n8~7amIQOq8ef@1MexnVAr|bO4!qbtyd~Q${yy zbwKShb$Z@?DE=y5N}F}A)4qs%bg1Sq%HAJ=h1q4$o2WufDz{OuxHdR;d<=b=6A%4M zL%@0e1m>8tjC8)-1xmADFk6H9DC!C!I^s>F4iBpxnxc~;}Yog+7Wbq>?Kghv7{Dzo#~o5X|e4SU#cQ^lB%5i3jT31UuGyT10o9M?J^o;5JhYYS~)qv@~FGx4=|7X4+aPd#$xQI@$LOnmRc z?RgE5auG5F&$@J2F~msa(WI$Xw8Q@|_<#Ek znn(9S_{R}cr=k}$-?S3@p6A4IR3Z5Ow?z2XX2WcUdiLpO1+|_oFKWGNPA@)KMn66Q z+V~YPC)bHx7&8qsPws~L-2buP<3y>S##8w@^XSW`g5P(z@V&MQ1x=HSz{j6s_hXlW zaa25g`NmAreR3LEk)nq!o&yDLssh*4H-g8tUr6ZkII^ea9{je5A|7H9WUM?3^-)*B z>ys@8I4bjFmrF=XSTQpb+@8uuZNSnx7Uu7J4ax5fsm7=>_^e|_d#5W>>pfNUZ>kAR zY6+%&B$}PL5)L~~9A#l6Qbp=cirCmzhr_nyLapiy(V~9+>0;N##3aNRBL0(*pe&&e z>SQaCTzn0sok1`vHv<-lM-yMWX>{HSZQM4<9cpt4)V?|)(v*rrFs>!#p9CkO_?c+Z zK0o?rSQ2eoXGY&mPo?j>By{atb2@tPBI0#89Mqm{pgT=N`Rq3~w zt=Y*|db@(Xa{?T&Xojtq3+epQFtpYlNzbMAl9Q(ExntpA)O`33F4y@{>%)`i0p}{R z;baATyDdwd7c7IhOJ>jw&70srrwL@;E&-JZK5*~lei#&R4ZIIum&_ez2Tl*auzr8q zBgkg z%QG;2pN-(R38T?dU(xUTPJ-TCAtMlU1M-7Xs3_tDYfSBr3VZLs;h;P+&h8z{9XbH) zHE+XOK9s7=?1CG&qUrBd>!`hPHG_r@aH`&e!Q%c11^uvQVKiM=WK7#8*AO!qHJp0s z7&v}+U^o4i^LAx7h;hl5+#WC!+CD48JH^rD#-0#3*?%+paD6N}ac>rFU3!M5=iZ@z z+AFB)?3eV({aTQZ-b6hd)am*UMeITM47lT+g6`tP0KdA1s_{yj+=9lXJ$S4^Y7RMKZD1E_3>KOO7y7EU)7((q@F)T>jG7P+S3 zbh+$yLFTF{NY*Z<^mH|1^f55-V0ZxQR!u-ZXbm8JS_&s9*+#1~<{67?dZg?U}6|53=1ksSM z+CrM%51?9eHi`1?>Y~<$A#~*rfh{qh820*QvcDy}(fI2`7Bq2&*v!~pP_GAxz1O@I z_sn$@zyDq&){&|Kz5FIPwfP0)oXV0+4ZA~MZ|fi>?MvvI(xLRe+C7@1umb#d_NS9N zH{!{?Hy~p39O@F>19QW)=&*eKjGMo6V{1Y$u4;{f7~QoJdOCI<)vEft}U^Sx7@V zMz?;#CyHCR{q|=#q9&0owDDv&9D7({n#AI5cLGLlDJCgyEzBqD71?rGiyx0}DM?L#NO`gQ|qpV=v>L&%j0=m zqc^|4{waiv4i=p>?1h@+2C!f)N9V;gVqL9)h)gx;k^rA zWqt*3_sim*gRQvAcso9Py))UeR~7dRUW4CcwXx?(FKQK}qj-oajXKf};*%OUC4RTa zd;SR6c0G_@s}me>a}%+#Sd+W?bDlNsIGncjJZm&WaIUPs}+C#kr^3SJ`ApT(d4ZKxt!d4GZW(G%8>7VlpX(=`l z$rT%F);d{o<**Ta9k_(fE?h(GmAAmthsU6CP6AH*P{UgG?BvNliM&k6J!D?6hRKfv zra*zPtJ2Kp)%lFSx#q%4e*MJ_fvs?DlRieQ5c-Hb{K7XaF z!k4ZI;dP(-@#BH=)FI^ocm!Nv?st0y|I{fwTY6ve!XTRYD~~1_`X`|4uNT!Z@}o8} zGVs_olMFhk$zpIky*w&Vl<63XPd@x5+xz|imeT6#jo3b4bA8mo>c9W^?&WEgW>LI4s6Up;l63o9I#XKWT zxT9+sE?+!?zje$)b$uT!UsNyhy(Vye-+o2zr43KF7h=!s!??1v8x;n>LYuc@?h(^~ zMy+b(sEQQII=vwuw#oCOeG&ZF86*Di>RMR+Z6cksJVSK%+Dw*UlTHT6*}+SznDYK& zXQpo1zF{Ay z0kgdEk}WMuWIB%<(RIjE);>U!e)FgS-Gjf;r+z4$HT{B@Gp>__u@{JwY&o$Jb|DGA zzHsd8GPeHUPV)G%0X*v6gV|FERFdlXMxNp~8%HZB!5wyDFPlgpWw+627Odz7p7 zjK+uERgzfoSq$jjj|)0>!m0coiJ8+voT0h~d`eUJsnapst*jZVb(HCLnO!tR@aJy# zy$bv09bv!LOyiyBKBM;BGR&EHk*sdn1{ID{prSb*^9JrAJ*T&e=7|EqBUqZ+&)9?J zCM)^29iPxqPLthS87z@1|AnbyJASTp9(Q|~1X*RrFh0o-j)z#%Z#$;a?Z%-J?dmR! z$co_f+GDaN!W;cGv-qLq&(QPOW0AUlghj=@Qa_kFmG!!M159fp;njh+55dhaK8$s`KfGmDw-k)zmO=LqDcmq-@;r$iIC z&!*%0Tj1b6;eNQz4yV0#;$xL-aIe7PW1q)!ji=-3tXvrgP7Q~Dx4y#t?T$RvJ&!MU zDufEvN92;z0IKLDaFiEBkUXTgw%;op{b~v=_%MM&p(TBo(oS@jxp2wb{#<)^7k(No zcsWN}V^pJ6ia{6g%ZP=jGo};*`tA$wfi2u5?=3&)Wz84=NFhJ{ zhT_#JH%V5?aK16>F!yd&f|{Z#luFa5>V+}%zoK@M<@mD$C&>yPo@Z;oSY zRV0siA6W5X(=A@SZVq>N848~_=8;1&doV}7mx(GAcz)6(s!;d~w+D>mQ`GCxCb=K) z`6}ehPxq17)veI>cr2ulMO?XG5Wjy=gFByB#>0~7v_*FbeB36-cP;mXO`F@9rPNyb z#8Z_`-x`ju=Ldr5pbU)~F7P_9$kUEh%A%N!VUn5bJNO-IhT&d5pjYz|6#rz2Dl#g` zsZybgci=rMIg&+|i$+1%8eLpF=nYw+tOB7qW5^^aH7K1P#$Gt8G7raUJebxCZe^#* z`4Cwm+0ZH)@ZV74SYV5vDrdm=bLO=3=P(GI+9YWT+$FkuU6!sNkOvE*n&F|$ISSfEd#0hXffTTcmtisck%4@PHy}0 z4v(8Y00MrDgu)~-)Fut(aUU(YX^|z4ua8EJRm1ts*j!j|eUBVa*b4(UZHJG0hjQ}) z=KRi>D9C?58_!o?M6IKNe8$vN5^sM_;`Y0i4Kx}B+n?l-JtH*miTrSGbK^0jy4=Mw znb)j9eKaqfF^eC1-~@Vq*MYZJJEk8P1|GXC@!X+(OlwpT`sWsl3I+dw);;&K_AiFu z*2&4J)&=B(WJUQcH^G?|ABi2w4t&*sDco~j2D*mr#YO7AbX8u2sP7Ab|GX{Ppn&x> zVzIWkqiQH*DNm|_1vaVG&65FEb zBxtcV+31u@tM>>V>ZXM};-4p;H>Pgo&qJ$W#Q{KX-3%PDq>g-eqD-sq5xTN60rib$@uRz5v+4Iv;mUV4?ApH} z;H0t@6h6h0$x#8gQu32r({ZIAp8Q89AJB&TDj{%stY>e$f)biFrpH)Bb*&;E&7qkF!3LX-d@xr%yVgx#0I!aZNhm*at30~3L#4>6Y zN<s2iuZxsAU?yz8Hu%W;cP)g<65dQ4jOpZ-now&%p1BjqG3F zY<%GPAHUT`4$EK6J=>r3AD41!IG zABe|?T+-BUGHf|87_$5R;FAp(V9fqN`0H?m+@By#V!ieW>Fz2>-IoQ^toCzF>oCb_ z%?|d|_yUW(8jb5dPT-4cU!Z153Z8RbOQdt#SaJ1L^Yq6mB7f_d(70U}^t}6V5e%o1 z#vaVSD3c|Z9%I|QKPnK&PRDjT} zRia&2hrmzc-;g^l3W^fKpx<>>Xfn+)?_6Rb+TCM{s@>XL=AA!Q{cRGyVa4F* z!I%3u4feg$vhc0@PTqf`csSRFk9DrXGI25bh23GJ!=l(pm$xFLsmA>0ttxVQd@sgl zjKcj9sd(acJFBUTC7aD<*aBxac=7Bc^V)5RAFF4gN7rd0sy`x$9;=BvpY=)>>db`9 ziYkx{9Vf7L9+d5FeS*s_C*#`wN3k&|O7J`$lNc_@Wh#obm{EU5;DxlKZd?iO3`wGs zemugpRs(3rx_fM~U980Zfi>uwWReQ=$p{*w%|Fz>A)oZrNoS0{q~W&|RK+B~*M9M& zzs)mJ-7}^9<0l_xHClOJG=3zKnWwF4=eofs!or19w)5riHAHKaj0r&in=82C_;&Ycm zEY!)vtJgIwWX}WauNwxl3ua;L*dmNEvBYUF%*|C^*TcR$_vwLMV{mTae!g*S0Phqs zv6qg!k;TvQ;B7$DnUGFqpQZqpRP@>R`>ncEtT$(@Cghq{7>C*cqor8NLeJ(m-=FOwj_?i}cz z%VViKg&8@ugINxnh2DP>X20}>N4F`znbBB2&bUBul`g_Tw#cq22%e#sT;|apUcTAG zo#YC=w_%^JibmDuV0Gdu=6&F`Fq1hcIW4f!e7s!P3ZXB)?9^RMKhMFefYZbM-Kcv_ z5ICL+=eE-Fd_rR}*}kNlz0dsurWL_Ne@ZYrSW$(;lvvm;R7RAqL(m}-MAqeyjpGOC&Bf7=uEWaX+bh8`A>L*?KeT) zAA|Yv#%2uf7lP{|@3A$3W+b7b8}gga5Fe{0IJmuxZ7j87{*TjW-mXf@3f-VDIEmlU z4#d@ai(%g%Pb^tuhrWMbVB@lP?8?<_p+^t^t8Z|4J?aencXuX?%J~5iQXA0U=q_Vt zM(``2Y-sm(3!2vBPV4K-XwltV!J&H`8k}=s^6$T7w)AnlHLMB;)z;&hRAbtiYsmtx zO((-A2|b|aYFOMo0k1cdz*)Tuv|qV944>CuWco#o>&9)wkZ51dyIOxV~BcY!RgB0uX!;$@#5*yQ#?f(kPDq`y8|ZCxPDPXyPh-5E4n zavc72Bv9wv74)&jZ#;V;pJc=WJnFkZYt(8{v)-DosT)MU2A>n9^w);8i2-O(atkE; z%xLb}cngn@hco=XPbNeX1hKp*&aF&3Qt zU(h>3_s(l`6M59x07ZuY8sq)(U4Rq1)l`>{3Q^);6_!%%pjS|>-HAK&E`e$DaXM8= zg*psX0_O`K$@O+?QF_2Xb}qk&IgL0Vx!5K6pQr1vF+E=~e#r-XZf3;GXHJHOI$t2e z?ubs-N0PKn(tKTHFwVXhL;~L?5n^137Xl8G^@G$o&fP*SYH#4hJ(gfO&J=Ilj;7C8d!2sR62J7G^gp%bK`_?v%4B^Tdf7R=Q7ATQ9#A_E4$CjV}69Nje7Q(zaV=>Apn@kwF z)x1-V1On&{!BxbZy2A{+XDGoqRw5~xQ&Hg*ns zMk;13riyPB@L2XIRP9(ys>)4}m>W=(kCH@eZD(C~j^nU+6U_F8^Um?E}t& zr9}oRszrhIJRz@Z)`$=HyaxM0hjI7nQFyH_6Mw`lM<;TPESTpD>Ox*C@ns`=hIo>3 zwnN#KPj-+WGZ^WIU{HJ_4VN93fw|)~i}#)9NkaZ}`rahTa|Ng<;x>{ZtV~% zZfU@b2UlTSQ47g9IvIvM8O{Tb{UUq!mxF>vqv&ag4|*>)7CLHbv?j0vru=Lc^>!^4 z_;~{Nc-MMVJNArp8XsYQ+E0S~Iz#w1Fg`|M6Obbb z@OO_C-g`2Nj_LOWt|rQ$be%k@kQ45KZgZ&HG;~ELB88M z4_ci)RykKGP}PwsecqWO}4}zQCLiT;M#KoIds$FN;#)UBXcDEJZEa zD>oBAJozKaAI*rg{syi%K#m{!odEr0r?S-fk&x-}7rAKzI_|rUF-g84c57yPw*6x_ zM|iUpB&{s?)^>{-851FDs1hN{BQ55|$&n{EE_iu~EOoE-2M5P7Ahr!FE3Ua+p1F7@ z6pWltL;hT0lkZxSzfTUs3r|h>pg0K(Blpn&n=O>jl_fs6n#o9$Bjob~ZyGf5EJPjB zhvuGY(J2otr@XL_`d$+i+jt(4_6?yTCt&{k`d|evukVu~` zQs>@X{ppcx7h1hmL3l4OfJ=I9&`_L%y|aqoS3@LgoEAe8Znp`2B&G5b_n(oek0Z!a z%`-U1?I@QzewGAX`%K(D>Ttx{QT$H8Enr1Ssr_+SJG1_bTOkB;^wPTH1pUz zHaqb!ng~qVCsTi5c=ZC>weK)p)iR8lcg8Z;IbG!J$vRYRNW}H`=keQJH?U0DJ8y}3 z0R11%5OV2j(M2U1!j2fygrNsv*1bFAVMGV49jd^k<96bOMmfYlT^2gbot#U&3igs) zbfuRQ&aqAsIeh)W4$LhQU9@;fie-E7U{fpASR6)QR>$He`)pCf6ghP^l z8E-Li&8i;__@tCXd+Dq)}BNLuHu1N-(5rVkaz ziigQtiRI%hX|wGDay9x6M329QI(|}O?+kyq&~G#@5@zvMUuVGjn|XAOV;VjDyMql< z62k!lJ7yHIoaGf<7oGkX$TDl)@bd#ZTr$mA*q3a@D~YX`y#6Wm(aEFxCtC48h5mT) zvMyuu6d+3}iUeGk1G$b_uv<}F$LicVGiYAz%TiGb~?r9wC-)aY{a~yC!2H}d{iEO`a1X=hz2Nd@mCikD5 zffr8)!;KqfNZ-{kFibO~hg_E7;sVkCdo`zz-Gfi(4P!;0_Q1SBzwqpPdy({!8Rc76 zUBdZ`=Al>q3^=yLjvWftX4z8~;VMrfxVkq1Z&x(pmb^V|NN+ScA00^ob_~Hhq2rU> z3GhAWBHJmq4Zb{k$bRY-u!3F#!E1ZI{G{AvFnnzdhLRJ+cfKC|a3~NRyL#|_MG>Yr zbif0vUD&y#p>3H{O+V^vj3|FtJHi-l5=#a-^D}tWcwtt?q(5AveD)LWv!vE z9@puX2j%QX>oi=k^p`~I`3(WMwHOzb$e`XEC63b~QD&o%EuQoOEJmnNRlA`Q=QTO( zQCtkl*4{_O@S9jyJ_CPtPNU>(Ikq~*pM^pYe~qN{v2dO8y&_q6Ydwtk>xsRI!4fu81N3ZvlHMz$xN7`&qF0~- z)#cOa>b{GtV%-#)@+b?sE-;8NlSOCaF?h;HlckK8#X(MoSm=}=%+`t7aB93H5v>1YqUnrwYq3#DD3C51yZ ziT(2fB;0HTSWi_(pU5C^df5u=itmxaMTs*D{rGBuKHpKG zj626Xg~I2NbgX(CYTUYwfja8wTzr+$zUL5XFqJ-OeF$$KYtqXAs1kTz{bt`F`Kxq>_z=HH9qEv3h8%TmaiTc z2vfAj@w(;NsQi92YK>5a@8OCV`+6s9l6b&st1C?M)B(g*()dvOEHT!NMEBi>ys9u5 zWPdpdchn?^7%?4siY#zh<{Mm+ZV2mg>PXVX6hO<5tm5!{TJnU0amUD_P^NNc3AFt`p8+YLp_G zvuPqUIz0iKp~2+BnzB;s&UBF3KcDuzxJ)e6P2s`0X{7H-9MRl+r2J&%b24$DJZssj zBgrp6g_+9g?CP!_QOtEevZ2Y1xsK0aCE=5p*`a3Gw&51*cPL8I6rRcsj@$?qe|Nz1 z)0PmP`;RE?RKy=+bJ*N}*#gf?K@v^}W8tL~P7^oCyNPe7Dm$)8TgFl4~)64-wAZXY)SU;r#-m((h zvhoim*AC&DD=y=i4_cDP^Uq^=)EXSwGXuP4rjvf#9N5YSEueYx6wyvJCP|+{NY1DE ze7E!x>Z~ECjAHTRj1k{f6}zsQsHl6>aIO#5B-Sk1|x|2Gm_eTONH*LLW%U< zFtm@)!;N0+Fk;Pd{7+*%J{+vWY-$x@l6g2ip*s+_$<6?!+sm2Xp9}E7%w4i`>}9gG z^ATCKcM4w`6^YWT@3Ld7&0ubS!AHC10w!rJA@a}MEUaIs^ReiQ8KX)iCx?DTjs6UG zs- zqq^rkEQ{r)cmRR~mlP0+SBU!2snY?)y*>=_yNNpf!iEfEbt8eCqy8oi8 z@@_88Bl)FcA=Laq6)kRgCefUH2iFSDgZZPDqTEVLo-lX}-?ujc`?=}y&D-SQ-V`gG zHv1~tpLv1j{6@k4OGPLnO1@`tpm$Bck4WZvVhNKHH#TE%AQMxbtoz!pf<;VZ_v z;lbK4wy3H`$eaC;)Z1&)^rJws1)hedofc^qSl0J+V$sNc7|0nI;WJwYoO$~T$JhMCHI-mZM%_c8<&{>vtZC$!En}O9&~@T_JFQ`qO37@5!d|wPf6w5`ORO zMl#__7FZ3`0kwg(pCO*o}Um?u9J<|KP8>Ga2f+u&#)tg!4W-y+%tTl!?d z_>2o*ELlgNt!6a%_gQecH9**Nf5VULFK&`Apu=Tv15d4?HGgcxEBiKxby6zD`nqSt znK_SX^^>{eoR$Q?wm9KtrQ_I_J{Mk_>QP@sfCs@^u*Tnz+ZWtpHysRkhj=?i_1=&y zc2MAQ&Z)eh$$(y#e54LO{kn<>A=sCEP0c6?vm8aNA3)CC?R1AgIL- z4!%tQr)#D3gRch6_^v78qqXt&31vo+8-DsKXo+IZ;{IdU7v#Lix(hp>VjY zNMs@K!Hrf6S=Lr-HvG*S949w~r*v>yKjH)3;d~v86w~Q}!Dmr_ixV&EHKbfQoXX`Y z(`$npsod9}^oMAqc*ddY)W7!tjea3l64!c{jh}yq`RzE$bvGFBO{xmCvO0u#DelHW znl&tJ{4+cnaRm+?TufaSUWNpV0rW()2p9GS`5tTfVI0@?lJx(9FDj%|KhN zn(lpm9=cc_6pcPW(-vrnt&}fPnO8sI>hd%aIdu$XC7u->)pzIj+_XUc&rtD&YM@&* zHi+5-c5=ywUT(T~0Uf^0j#_U04wV&lL`Wg9jN9_u_lXZvJ~I|%CB5(|YyrM$%;)iA z1b%hI2YSj-TfFHeWe(~(Xwa@9pt^5kZMP2C9rVYq_=zQWK4pimJdl)sngZ@oQvna< zh~9Ozl$Td}!qT`IQ0Nj)28^r5Rl_4hy81(4m{fLIqJ9GI^HheA|6byc!S86~=@3cV zm)Gq1hN1k&ocm55b zF4wmkhdYeE!lnG5;47i9X#P`DuvdyYt)Eiv#|2JM%LbC#|GH?9`+1AgvLj%7gbJyx zeZ$^xC2TvU2|HFD^0O98TFp;`?L}`Ubyd4;sr^INy(u3u ziiPgPyhkMZcP@?W`9+H!%agLu_ZF{&4%VOR?)0@*1D$jKJF)6JQiMUtd>vmQNa z?L*wxY{fGPO<1od%O+(mV~s!Wf!RSNa1GBE@nPBAYH}2e=sOBqjK@+7m5HEr!&r3C zNS8i((|}(e$MN*EEHs^>hBCX7BtE~=*uvb2bmP1ubjhRv(PEpIYcf~qS9_(={e8WY)w|E zgul8=URV2~p|DR^j%X3srPs^rjotCzPbt)R#qo-k8l5Y!UXNFArG>&7MI$w+y5Km8 zw7O4c&8h;M#U_%NE<>~~j)7lxQ|aefYBWeM8fv%riDsreCIilt!dBx&P!7{s-XC=k zk27Zlom0qywjB`s#1-ExmF44|cMHAxnM657n>N~=huSzbZ17t`Q?}%hpw?hgy=jH$ z9hE@yH9NGC1Y?i*1ddrT8{NX!fU<5o`MA}G5YZ00F0725*}0u1irOW;8)tLxbR#}s zq8@vD%#i+OW9i*_8FY!M7k%QZNT=*qp~I%?5~=H#B+YuglE;k)@bawyP(N_SqBbpp zc-uOVq6Im`DCcvzx9VU#SYAr24o!iga2GaP_a#nBSEku}XG7@XMKmmT11Req6os6N zWw%wQQ{(AY)Od0=HOg?NH5<3$y3gA1G0}$ajnJ3O)=;7KDzSo(Xgz${yANtdO3{H8 zM)bm=So+g5kVZw>z}DbPlBqo|XjN-ZOLm{3J|0Sbp8j=zQ)7- z>g|%TN)6<8l{$5Np9km1ZH8?ZD<#42j8Sg00?E`0!S;_qlk-^I1%ZFUW*|(&`yK+nE%vVujkBg_pzh7S_e-w_v%-Ry@sT6^cl(hKifkW(- z^kdQFQA+&!f@iqwQVp&8I!G+@V;Rv;=`5FWQsu9Q&nB-9-iAMX6E!*>O)qa{Y|yVW z@Zh+>qK$Y8&$6$>@zdj|u~#*FctJw50#3r-OLZ`&+5OPLeymA9yC; zhtF?Dfn4KeN$G@m?0KoBzyM+-MsVD}7=ADR&c= zy9DNVufuoKQY2;$RqTRuKNzg4MJsmb(~b?vBzM+-XsQ*2GlPGVqw$)OjF!brX;zew zuU`&p3+6C2&H2nleH|S40Ajavk$jewZZBkqu^ji4*vK2p78GrY2Aqi2x(gfCC5uiyxA6J-ROtWcjk)y zZfA*$vL}k)?DU2nPbKcZY$W7Py)8+omI0d^zgWkSYBp-Ako{NdB!{~wF^>>D3muO| z;XY=xIY*P8vh&7M9+C8NLkwNA;xEbHTEZqT74qg|SK$4wvpDK|KgsfwHteT-5~}u8 z;;*HuNW697x$HL!rYHmQ@2g;eiZ#qrv<25xU9J?fgvV|gEWz1^P{f1T>B3;LKy5t6 zE{#IZMN6?kdm-5EG=<^McVX*2cbGgdjb!erB~cI0k$+j+NV(jwD$4_$9H{z3M(6I87N2MvBVMz>g=AWtpxrgQ=~xSOT0d$mHH;C%H@_}+ zWWptU6}bzo|Mn5pzeAzq&RVFwcOGk<6R~ZjtH51rrYD;$X>f!weURot&0O!0zR-oV zN$?u=SsF$v%>y{KQ&YFfn+I{p(J_ZR2V5>M;cte%KOip6?S? zScgN{jAZg*_$rZ_g$mXt7UIN`c)IWEIO=llG)sTrfT|~rsekNgP+Rl@Gz*r}lV5g# z-S$DCesT=;e6W_zYbuA@D~{yt8a;L{^9E@C$R;Pxn8TN|m%!un2J{O!4Bz_=<|E-V zdvPrU6Px7OtS@GC*6s$Z?%0oVm$Y!yM`PS|XEhC!dQ59yhQp<^C8GUim&l0yAHeAN zBUZ3(Houp$ieLF$gB8XTY2v6+&~hUgZmh3l~9DU@|(Z3L--VuBZQI(et7xwoBtPxz)M` zx47M4a-}0sW1$=VO&o^frN>i)=^NSh;yPlJugKi+9GMl?F7jwU!OG&J$f+~Sq4MY& zy0pcUBzyfOHTrUxE*Ha^2AE=5R1+KEBF_`l2Ept%eUgQDQ`w6f+HAp(MR2yp41CRB zv+O~#?EIWck}SN3_LK`8lj7HKV?!J{uUrQDS;s~5)03fLjuri~&y>1WnbFJF1K`7` zX0{+c0&_#!V9iENnp$imx@ed{HyVzi-fr=5SE5Z%Z5N}RZ5*q5e$|=`7+*2M$ zBWTU(g*yQgAbPhq&fR&NT$YO>ng4c^0Igr-L6;3$mXr~njxjW)J^@Cy{bfTxAIFzh z^4PlE?;;zocVtSaF?rg-px@nh^z-58*tK}2q;UH%sww0gPHdh*q+vMmFsLWlb@@d6 zX8;r%r?OotQsrxRZf4~NggeNEbcToOnR&!F_S|484)ICHvEF}~vuBKCj7JQau~!w8 z%LkGMgXr?*2BUDR(R4I>LLl7wa%JX~}K>KD!@n;qkEVy7~XpKugA9jmZ)wi_OwqQSf@|FY^kt&;AR zSER{ao;IeXv%kuT)bW@leU+aHE0a3GYlJzy5Fmr;qiZBdMt3AG3#aqQLF=(?**~GD zm<07F_mGZb+VDqg4I$%m$j+@R>7~j>iIG+dwCwl6Ey1ST0c3tAw&+n&k& zE+(e`Wx>3pgGAh%%`U|R;_@qVMXpY@_+#-Ov~WEQmJ^QP!YM;&@{QSecTXTQj+!p< z9@oc4RSd$9M}{%=HbDRR4tR3P2ll<#6%Tri!|_I^*gs8m2>5da!dWC0#g@X3PfqCC zrGOucoZ(OMS9JVb#)2MZi);-E2KLNC-k3ye+K&>urw7T6kW~#j@@=eKv#L8A9mA{i|HKZ?%7ovZha<0_J5C7ViRWsjWaeyCLHn^B}4 zEh;zK7;r@fG(Y1C6--xNAp!G@g zRG>3FKiUgYdly1j-5};jDDz9^XQPb9T>5)&F-*wQq{}+$us^UKo2PVI>-kW(jqu-sRBZkl2unM( zs4ZrZtZ9v+)ejx{SI?UG5zhtRz_DI%=s1ZE&K{_tIf>6tuVtIwoyV(6vpD_ba`+>x zk9_DVCmV)K3f&kHDYXk>iTykH5rGUv6=R4=+-0&z!P)x6x25ps*<|jeUY@9K$tV`8 zIT`D_PZ2}ivuxs}i&!5pijNS^lNKpk$%U)()F*#8L^yhp1C!ElsNl+*RkH=wxy3>0 zzspz@tF*1dS4c*u|8 zZkeHUL-G@dyH`zKXBrd7tTp7yBn_e0+DtsQ{a~(2`n-6yIA)}20atkh1eufsKow1kSgd=lfKvm}^3B1g}czPx?c`HR{Ulw9U zoHE{ix*s>}|BT>d%(qz*}4lJgpbY1DRHf7jm@rKD? zaw0dsPsRDql&H4lYG{=Ar7N4Vpkk6CzVJ+8vvXw+kMKMa3k%zX=Q1eZ|UFFHWMS0F3X(LSv3Qto=6|Mx{(aizbd75YQ0cM`Y4e z_2YDn<#$o1IKZztiR90O3b61o=D!cIB44{E@Ow8n!|;;L)VMX6=)eDuubJ%0%a6JS z51Sk)zx@y7oyoU$d|yxE&UTde?HtLT+8Dt3XO>_^LeRNa^!iLoipWmMEP*1bw+% zp=ZGZGT~+&JFt8gF(XsS>YYzuN3=f7{o4;au5mc=$S5e5Zx%fjVtHFE95CyB6D;0; z4DQ5Sr(ud>RAYJ=8QyOPUPpq#vuP}j=ElLGel)}?D^TszMdXuzA1-VB$C|>8F???* zx_74HoR3>Mvmy!py}^I{;?x=V-)O;N0KC}D z#jNCr1@rrS5we@S$sDECM9E2?)9Jp?+@b}>ki7xQH`}w3a&f5rB%b)by^f3A9Wf(y zKok}kMvTEXKO#=ox~>NcD+M?%y9V25Yw^bgC)}&KFYx|(HTqiV zvB=nK6Bv2DgPYr~P{YP-s#kmpZfg%C5!I0xu<|vq-2|k%l4>(>WT@n z!1oSEubC4e*{+048_7e>yl)t$J)8IWiFjgR11dP{S|9JVXEG+!tu0M8*w@RqIE$5z z(AN47p0C~NmL_rNsz@S(RULitkcNk^|Co$M%eRZ2=j(I_fyot*sNr76!d^ zn_gV5gnY(VA~F73`kN&VUIKwqBR>d%09 zl}b2MSOxb2e8K2N1hjt?GHvx)Y)yD4`Eu(y9?)^?VvMd9?!fL!CzI zEn%59?=kSw2o|J%LgeEj2yo;au1O>_zu69Y=Il0iP);@!+RG;;4pSmw4mx3J2NbeT()&0r{sbl zXs@u(e_X?)mZYGh*>spr%SCS;&Tf~Le{;3 zPT}iuZn*_o9k1mMtxaUwxesAq@CP;{C7FE4tcF%0pM5ER8Q4aP!G*(*L`IVtI_4qL^Wv8m&n&0oOZo`0rdCQPI^G zZ5}TNMXBf6_@oC+E20Y5{mTN~(+@zR@&tJ@g+af1I5@o4qsy%3K;Ni?tfb9~+*Fvv z(9#N)CJm=UVgn#>s2>=H`+<|D3N@cJ29Ldogty^pRC{LxJTV?p_E3Co*_3-D%A`LA zVukrM8|S^F;Fwz#(Xdy+>(^D;>Wr1pkh%~QPLAdO+Su?vJrY=gbt8ai88vf~rbg|r zAYzRNPlrGxEc{p_9{$;zd z%V`15@~&dZqYto5i9t3)eTwMU?D5zqwv5esno9D*moxX?>!qR#^Uz_k9+{LU&0bb@ ziG0#tz%P?hZXuroS!&eFJNx%dqkQX~)hD+vi2`;zUny=UDN5>IX}$rso!$)Yda zO4u{e08hB7!FHt*w)5Rpvh3XukTr_M$4#?gf+jU-Qc&d zN_0f%9UNKmjD6oSm6Q#uv`+4Mf;LZYVMKo}8q})Lf4X8c;N&!D6W})Jj0`B1?dC7@HJ_5E+&tok zy-#M~+e#&t^!+LL=J6_>0!a%HOzakpG;8BW=^+;(I?mM;KR2ne8KL= zBtGpnG)d?|-=-VHf8QdeqcD?AJFFpaG7X9L5h`*Wkc8tn6l@0SM3Q02=sGHZJCZe$ ziiTVu{ZawM^U45Yx7wz3YA0FyJBC6~>Ot1n?#I9B-^h>kTE%}cH-PRL2{dDY6O_14 z;k`_q`E{Z~8-4_6(-gjp^R9cynX32Tz)2^pbPhz7J@=ULNJ(hXbmr2JerK9B(Nkn!=U}aH z-%N-sK)Z#J_~>3CMpk@5t0O~UdEj1AE-WW@%TBVMvxNdvFc!RX0aE`-!zZ&+JT`td z)-IOi`#X(zjT0}3eMBI$n)4epD}IU`R(Ilk)em^8t&gaQWpd&^zIdSd1&&^E0C0U4 zgja+Xv*=au=u{u6>DbJ&#Om3N$Wt)m>NZin=|Nm_PJ{b2%79r)UB#En<506P7w^fP zg}>q+xcXi!wnj|mAFMry;rtg=J@uJc6_}w>j3ICMbs`@yDFC}!acR&KO(Zt;*eL%K zjXJZj(>Gi+byEgDdY%G*K(gNV53yJ#uAE8v0(d2EXkGMWuo>$G8JDuHGp66-zTc3=e=Cug=FTAVvJy(gO}3$uyDC~q zXQ5~F|NYxZyxJ-+zWmN7(fWl#E>%7l9N&iU)`tJ_*Tcg2&-tgh^`Y@ZEUkjAO&o>S zM~!5SmP(*z_a8*;kA(+cmok4}WqdNm3He$@Q10{~QF40ZkbgM0^2t7on{p6eJe-3U z!lsdyl`dqomoIyx@Ce6Fc|m%rFXApMV1B|H_tCQPP}QXj;xD~Xb$&8Ty&|WtUS`3KouXu!Hc_^khlR6ZA@^tJZtmsw>Qe6-M{?M>nAr&Iu{OQ{ ztG+3Yn(|TPO`JHqQ0eDvKmX&#A514pbP`zQiE?)6@Ca&`-VHN6uRsB7&)HN?hH*6lxpi4Oio1 za7oAz1TGf)cRXix=fkI*n(V+TXG`IqNrf0>xQ;y?hyjt*e%esc1jaL`qx1U|bSz3|>i4!lw%=74CjFL8%ZR}oi(5o5wGWJL zRElU0DQ$=<#^>iWc#E%Qm|HgvzS`un^tY8nDf$7^sq06Be+$U(H-bmYCrk9HE)5O# zX!8>K?ZW?kOQO$}LHB-X@QxZne#ibH9@pLRgsmJ6bT~!ZMvS78^ zKpE_6>!3N;3>KaXU{$)|EOLf4mI}OznY)vqOz4a|Ri>a?Z=>k-re!2KK~D5Z+Kn{^ zasu~R$TAGJGP~yM++?q7?1=dpQr3_|7A4v6+hQl6TEIi-Jdp^!chzBcgTQk07)E7e z8p-=F^TE(55!mnd*gShaf3Mdc4jeoNGuD=}4?E>VZ`@39uIo3^q`s3-z3&zI+q?}f zvf1t<#)uUA;}X zHy)^C_Cd2mR#FP2<%l2oerW?39aW@n7bnBp>?G)Dngy+ScZK|;JV+cq3P+oTbMS$0 zSh#T#r!#*I+$;G5Jz_UlO=TrKc$Q6OW^4kb)gOrbg3;8?U^+DndO-6FM%tub&!H8y zDd0b5E}m3v;dIN-(M9F+XxYGUdi&c`>Jxp6=EQ{4LN1p&s-2>}*G_=kXnFXXoC=GI z|3K`515|vPC*4~VKC;yy1s7ceOJOwyird^FCa8yb_gAo`;Zg-~Dnkj@2PA{+8g=67bB{^?y$M zueB@aUHh5zjC(AJc&)>2c{>@7+|}YmOJ;($NRRSa0nj?VmuM|Bfckg7SRWM*iPm%B z&Sh)ZQ1*pP@lt~9V0jv($xYXh>=Y3<~q{$EOS6Yk?grmsmp6 z^R%eQZXtb|FG0_g3FrC#bP(I02R|2Wq1!_YVA#8z^m*O}+8OTxF3#rg+}IpUdW0;s zPIT!k(^24WTL=ZNPheN{W)``1H0)|uprK&8AGb!izwzrWZ?s}Jc?ow0Rr*en^e z0+#`uL)ezHvUK3F6fHgY6pC;9;FPoKQ19l=igMFo=&99Y^8Yg<;ZI>$axwetl1{oJ z!wZXupnj*)^1mzmotxo@$ieV<+}?_dKp00 z6>Wl$?}lW}YCW(GU(6h|tjL9cFgTQ+!UjsE;m#y2aLN<9IpUXL;NC~rzr>CH%x#B< zs;ZE#pGK;7SrT@?1tfQ!fQ5Z6WVhrB?!8YDyZPcQB-SX?5Zy`iSJg23fjyw!XAv_B^`(=NQ<(LJofae1s|STOjs;J6P`t zWozYiaDV;N(%*tF^5u>dRAYlVbvdjHYk%D$r%^huKGEM-)INKaVni^ zV?#d{+JT<8IGlYv4d(QXrCC9*iJJEWdT>}cHMV=f-Lo?S1+^e3iF0QINy8xSl?B-| z{|2{p`d3(VBoGc*pTW%A^`etI9l*5y4z$<>fN!%gFd-X#`l}z2uGA9l@k!vZy@XKg z0pq{VON|PwV8N0BY_}GHt?)Nu<7ShiS}{O09&zuhQ<#Bqi>S!s6TI7!0*8iL!CHC} zWX@i|j#<}m3@GF7rzOPlW(7NXQkzZRYRTUYzbZO%*OjdI)PfZ`+sV1;F7h$r6gMON z1nD2zNcQ}>OP*N95*_PknJ z$Ns`BuXHljg2AD`QAB5I7Iusf_?^;U$%)TOP`*H%bP8Fw#HEVZGxHU>YcRo%EMyO4ObJf+LGg$_@ zJtd*^P!;?YTo>OXBCO?)l#%u-B|?t7kOz-zNy%+f;`db(SWE`g4Q;|D4%+-7=TJPc zpb#|X>;R`EY08^@1CMG0T4$O8ZSV65G+t#h+fKpQe?Fpn;U{T#aVXXOE={w#_7k}4 zO}6g}C#M6a)63Iz$O&gF65ri|$|k-TT_X2h2Sb#hp-#hJ9le@EdM4i4@m=Ci&ea&`BA&HfT+?6?ag>*8(rJ)Sk{I|IvV~ z)#TTTB&ZboO4fTfa7}wBL;Ajn^ttgu>M=%#^~v64Ychl3?Ii^qyZ9rUxoIX9<5Tc* z!c2G-Jb^Xuj7Pnb`kYH=K5-wL_k;SaCSX+@o+R&yIzHfBMu z>=#(yQw1)bBjM6vP0(re1}8sjrX|ca_n%1wNzo%XZX8NQbzaoPz?b?(4bs4e>9kME z#3pF|2yAhDKysFhpeoPL!E4Vc^pbuyL{E!_04aMga(Rho=jVXkTV#>A=U zBJ6Q4fWf_zbW-RpV)xG!l2&YnVCvhxO;&0q- z;Oz36NQja>)3a~^zO849=I=c8RQgRe!X#Mk5=?3%j`0t!v}5JEHMlxMk<{`*Tv+r` zc++r@wpb|e(Sbwwp5-b0ZvHSdT#1M8`?I(etvz_*ax=;m2k>Y64ncVHGt8+`<)!Z@ z;f=M*@MKOUx%KA{Tbuuu-TN4fWd>?|`!0&TSr_o)i|@ot9Ogyd1U%Jnn<>`o8wHv2BRY5hk>Dth9P1v2dN`%0W@ zmX4Da>hVq&Q*i5;lcL>fvUJ6dSoEyhBU*541dBAPf{PP0sBxqiU-a_;FQy&F@9#_E zgRD}S2aoMH9chi1Yu52RwCA~A_sGlFjaR0^jdupE%!UjH(6JpI)~TZN4mVmZcrE5iO7Iryl;0L+&#QDxqSk-K zsFuBe4n(BDibn_e_Mpvt-F|W2UuhRv=;8%!7Pny2zlFr3b|vnWbAnaU*RkN&7M$?@ zI3E+<&3~HIi4{%C=p@JC*gr}%JMAcH*sbQBr@xp>6R2KTb zVY2GHfv_W=D=Oi?yqv)TasoOTqmllYpocRm>8_{|Aai|<4^v3h%(^ak6RIW#iw`h>#&)k+P+u@gZ?xcp7FP3@dYX9S{7k|Xmc!p`wIYk1yP^8q8(8C+ z2|Yg}aMHeBOgnuW4dwv9eCv8XJ=y}J%xbuG?tj7Xtu0R1JxYQWrNVK;L~iym8CvLH zLoQrt#^iOuLeKvTGzaR_iThjO(JL`x)iH?e$|}5p-zRqFHV6FVabRO5OnDB@P0ph_wB~VT6=_@9EX_aLvCmshlS<&q^vv-6@Iwjsgg?=y)m54 zv!B%@tJ>qCx`z}awC6ybsxL5a~ov;+eGI%+0wFGVqj$wkBiGo`GL|XNNPQS z{*Fmld{Yk8jf>d&amV07LS>a2qYnV$(ffc}p35 zc8I~_6(J$U)pQ}Sr88?3cqz*{b&1~Vvqtne6~q`CvUuY^^d zy|7X;9i?R>;LgdH+?YeA=qJ1@7CwSkkS~JT`mt2^@NfJ*M~eTK{2jeDN7B-hDl|0u zIPU!z%U+cWv!|*Q9MN`%czN`J;gTu*E#+X|V6uFf+A>f6w9is7v7QbW{g1KjS&CS5 zzlz_doy*s|JK%xb!}xUnU8WiEn<+l6E?so3mx(2-(nV*@(Bg_68F;2nlU`hfjfM-r zGUO~oY*_@Ms-`%?VjI6ba1QU>I}t+brRZyMb!sTDOyeYglGh>N`!Ew6GLshZX4}Gf zy{+ZEPWNS^X50k*skx#RrWq{Wx3y&T86`$nt8<0IE+@1}i;SN=f?dnVz!e<=7qR{| zvtOu7uE&e>s&NmPsp|nO`X!I625)e8@?VnFsX)WrZICMyoX_h$`88X9;Vk>na8Q^i z?p-I`6J4Ze@}lqTs?BkhSg@P==^B$G7uR#qu7{4HhpuIjfi-;F z<72Q_T>{+7d}-V2ar9G80~zO5i??Ky`QSIrL=+ZJqkAS}^pc0TEm4=BdiNeb-F}dC zsiu=zt^=^Q{t_tf&4cE;qu`^O0OMv?(q~acEcNmeeo@BIvMu4u%35FOL)S4GIP$## zvz>hLx4aAlD|i$0S+|&jWCnb=PzH}Xgm-kqBks@HRosX!ahs?6ZEdtx{ia(cD%yO= z&7ifp{h&7bJejVoN&@fPA}Uk2p!(T9(IKgQ=| zs)oSDe1^4~pi9RGL^ zS>|_+M7u?j{U0xLZ5a>Q-2FBpt%w*n@k$Y%c03Y2sZ1hbcQQy#%4YCd{G1ET^aHeg z0~fu=gY$v?WI^;_(T4N8agLx?`6##_mzOEh^~3KI-|{qwR28E$Rh`MQjudh}U>syk z6}qW2O5pYPZ)D^E0cXQmO|?dFkSw^NE5KWT!~g?iv8 z^I;0FBm3=91@&nY#eUCpj?9eZXw$CGh(FwX8w+qG;c45y{>pL$t4KAgX(V$eFg!VEjH3 z4*f=v`SCGubbK7wscS5pS*yXZClSW@#=~fpE9if81HG$dizhG5;PdvD@$Flz*)MTz z-hatNqWNtcBudA_v6;u=uaSar91B5sU%&r7kENys8yoMOU z=NcP)SXa$?rrjj*N3MvZcTHo9w~k zbaO=}B#z30!s#RFopO}$78yPp_U!1R{Bfg`mVy-wL;!?<3k62DFg$h}07ruG$FBAo$}t9|MF_Zpyacq3e}k)X}{ z%BV@yWi~9r9yYqzByM))D`fd)(6ZdBj(^^XDcslJnUaT-4uK_~#ZwE?(LMva56j-r;IY zoal#xbP6#`{N! zO6pDa;bRh;vay&cs(vLgw>#mmb1ygd+eujb-jOLhi0A(59w*ms27xx4%tnF%D$xk9A~r?H1YMI$W(Nf3Sg*-s=#DWU>#Z8B^Qv@-pQAQxY*4c3?^V$NgC&J z&K#C~y$F(XN=bNjE?clLfP6glj-5{k1h31DB+`&U>pN)@b20=w&f2oMWvkJ~{Tq83 z{)rXde?UGg42AxdU9iYJ2xGP1bGN_8z@+(>Fs@MeeL@CcZ_Y42*?AjpUl{?HK5XL( z7H6`5y&-dsSF2iEru0Ma%5J~449ttk~IC%;(r+4M(4ke zuqEmoIW^T5J6@Ebaa;-BELB1hbDYTBy^3FS#IZt{;ibtmkRN7I(ET?Y7L7b3`jz+> zY+eW+EICzh^C=?V+%=(4cuv%ZWU@K?`*5N3eDGRTRNB?726Gzk!S4cDF!R>M2vQl`4Bgy%cPEsj565?*ngUuG7NORUEA|bYnyl&SZTY`(oxe$BOFu8ynoH~I9 zzNvv-a&06z>LoY%og=9 zB^_tBVy@c_c7IMNv$z)w-%5Q&C#y#=&jWEx>ADR1aXJD_g`-VB7ts#WAlggC=z=hH zy7ccEdYKcvrXUMtqO+wz&&pA?!5Xh+89{@_c`$6d1U@5kVMbOk%$c(REc@d{p0P8D zti(_f`lZwQheSAO{y7@Dcejaj=RJb~@!MED{4&Wseum8M)Wuh?>tLhjAc>2YqQ8f= z6RngNL?+@Qa(T<4%l|GK&6i`3f9kUn(MGs_)>}OJTnWz(y$zpd-A4Vh@8J3JZ%p4% z9sY!Q@+o<**uujrFgo=#*o#+`&ibx_ujp~Ssn`qhzvt5zi>ER}#av?VC^+3-&1Pa{ z6>!_>9aMl={XF=W?K%≠IGz#4AV;5D*xwkI?B)lO%8V#s{g+e_&tV2hqi2*ykn3FzmEEpOM;hp4&NEve=WB>{dkduAIYcASL2soRpaNxadhT25nc7T7p}i|=E=7{)VR-}&tn{0HChZR z9g<<@CK0?lH=N6Oc^@ObHIawgr?R1Qo)Fh6XXxGU2pNa9N;`vQ^RhqZ;=9mBa=Pv| z37Qpz_96R4otKkP)h7>?&q?6jnSwVsUW+A7ZV|bhd=Ht4xsc24fd7{El@4o6XV>z@ z@L3}d<35aN1)f4G{&>{^TmSra2b-B%gu#l~THS#tDc$ zdz@;H8vw`GWuR`hR8+cjHCpJVVb0nTxVb(AJlvMS9GBT7Gu9p^HV&a@t<0$>};bZ^aJzYP*d?T4qB5VRJzhZGPsC4dK3M?q|7 zA{buy!?YyYIO4trrzxi3#=Z-DgDQnyRS6pQem)q{Ay_@n6%NlBNi90VV3*%~Tp3*spP96Z0`5jByPVJXe_M;yL0PrpKWE$VO4VCXaD7fZSA57BuG!ntCMTdR_U_sjGxJh4EMNP4P6hZ-gN< z)4$}O?>Q*GzMa^8iDTamn6m7#?(BA9jkV+vV_58a6kcUtf`_M7*fZ7JSmmsTD;{ex zRjZ)VquMs2pZ>a}h$#!KTW`>nu7ivF6@}iwTC(ZsY4&qS4wH|P1vRIiQwpP!*Bm(4&2iq;^P+UgRdB(1B}@?ZfT?TkMDM#)nC`osh-W3ZFV{`s zuhnwYI4Sf^s}hLE31hhE{t_@x8ZK|$!uCJWht;oB;qB#*BBLLEFvP`?)@>Y4C5^^l z>KI$JKX#rQ{lW-7h18SM>#fWn(oC*xYB!AEdL_PV5y@;S!jjl>Az z+S^EucaJACMsHz{{&OMTG9^=Xj8}y(Z3)nT#c)Q}hI#$;WCx9=kYt~Y*q|Xnl{d_! zU1vwpqxNCA^^3xkOW7Vwv&;y~N6v)N>C*T?FNjR#|1hc2a$GdI9v9E&*pL6NkqJK! ziCXPdL3+*}(aaB$^ym#uT6kRtEa&Y4J&Pofy3=|Zs~|`3M(l#g0V8nxJ{$6}Gq!Yi zyb+B4T>;%v-H@kv8@4}PjQ4eR;zRifIP`E(>5J}YfgQ9}bU(++YL3Dh^nPH6GiR;F zi5sQ(zt8W1_KmBcmQjeYIl-b~?KCO$*3-#=g(=>w%4gdh z`ypew5-o|apwg#5;~ERf*UDJJ@+{ZV*9pr=MAJkT5)w>G+Es|%2nTDmpQflK%-R*j z7QpJSn$$TpbIMuK%hGsZw&8PVI!XT>4Ilmgl4R*4#7$u#*O1`AS(T5adk+~xsA)B* z-=4wRq({Kq>3UqgZ7N%GG=dl?B(w3MPpv15o#GDe&KH^5@l#&j@`i*i2kNs;g^tc2v@Nzhe9aBbV%WQh@O23eOISYFRpR<~| z_qZ^-hd3_c2;R71j8zq9*$WqWzF%r=>0X!PY(ujezL7~H%kwvr+7Vq?kV`}}Cbi<7 z6Y=owVhiousBBY}6-5J-vT#b!P@ZRtWnvTyh7CukkNXu6n(E&>p{-&M`|KaQzJg%!oUz81}k2e;V zX3xwfzaPaj=eFa}B^E{k0?vc%NhNxHtR{6ds$d4ek?gTC$L18Dhro@}G-sm(UA*Z5 zKH9JeDy%cej+PPB(-ItfJDT#IZF?|}}rqg0`2p1@T$fmL_x_}OE`P{+cG zRA*I*MsXr?*1TQRxpO7@UzNkosdsU5VK})|RuBF?=V`;n2vqbN1#y1+OLbLOUUMagqTW0xDzxQ z9=5!O&)U*-!^_J=w?>EdeR>ZLC&$8R(Q+`^z7ng)ZJ_>^QY9TS^#Wt-8$|s(PVdZm z30FO4Ldi})I_~Wv>ekYTgG)!#%vulV*Vdx@Tx;m#_W|@-+DZQXw;u3Q9b(f|wZO)_ zzKGtDm!}HB??}?}xuBKyMD+OkBx>ST$XqY@!lR3#DQk_rVV=Wn-1YG$F6cf__C47I zTOSeT{JR?buTG+_cZb+~TN**L$0m^>Ckw!5z?R)}e2SmF^k8^{4z+i6$LwK#Kzw~z zt)Cfm{7wf06>GZMs1dH^{^H(=S)yXG6JcEeBJ-yKFsg*&#F%>6daD`FC;tB%nn@lx zy@I^|0JexfBnGOI)OSoHryO&N4DPN5xjsMo)kOpjtzvXTUybPWM8V7D9Yvjv+#xP^ zA418w7*tzY1+!AkV2$5S>ObWc&W*YXaZ%%`b*ut?(dG<0dw)P6TH$byar~pqZd`MG zGb!2o2YQdWq9(tBC4X`Ot>6&YS~yH_lp2BfU>7VI_5}R)nS*QoDEQ|S2K^Re(9irC z)2|)}Aq_Ke+Jt#9%tanDei_iFrd&3l6$Yhz3UrQAhi4O1v0z6yyf!*R!#s;=?!^i$ znbn62#C*spz2l-+is|56a2~c*1(Jm?enW%{{G(Ej&Ox~EAAQxZ>tRZArO zw4{zkJo*V6BDA4V6iTkG`wVk04&&=4jAexnf<$Ht&h)N{1l@S&GI~S{XNS#EI8r

~7 zHU`gcUT|nx9)xrbWBrL`Ve~^bX1<~-oq1cHj!s{PLUg9_L_8>lT2PXXDbV7+1iG(> zF`H#dnNEQ*tlN?Z8RIih*|cCJuAczAUml~3$0ne8Q!`PD&rG7wSdZq6m#4cuW{@%R zHoOB{TRAMmhKV8bfe5PTiS5p}X_?A;T?eG}z-kwpPZh44qx2mOPbWbkuk-Z!a&8h zL}ThYv?-)RaPFBG*Y4U=`?&t|+9i=vX6KR8Z)l@l%k{8gh#Fj3x}0mTX~<%D^K>-* ztMdeE4If8l)gC8H9=hV`?0Sia=S~*|Nc%b&cK3Or|GInz$E^X*-%6~G3`K`kWMNyK zFY6l>z>E>P<2@Dr|74B{e9s&RupuHkyT<3h2Gv1-qF6STbwsQ$N8dqU4eK?-@b{Ja z#J%JKBXjKowJWfWd23z>3w)=+RT2jm#BK>B{A_UNE-g~EqZPHeD&cPhLil`c51QW? zfOCf~VuBsMp!$kr7~*yg%;P$l&s)Yo|H?$F$25t&4E7=tyz|(0sxRtUx(t<+g>Y?) z`0_lINxk=9z{E^yCmQajxa%_>pF;I+UHCzx7%w zTsJu1FKRoN&DGVo@F-anI+Tr<=1CSJg5yo2s5;__FBwPR;g_F^^v8E7JwRs(N>-#vA;yM9uRB$9uucmWjS;zJ{z++>xEMenFOtUo-{*F=-ibszTHbh@S zJAddBGy7^HsdEjs?`HF{Cl8=cY0-iu5k0uY|1AC!5hj2+>WuxJD4v3@0J#`OA^oNj zT$yYIquW|h{h2Tz__{3!t&Px z3D}hL=r#JJr;GE26Pw~&iHwdztQxtOJtH=JyCxqBh1PHQsK5yq;vky zi@H>^mJYMoEfkxZOQ37RLO8yU_gLcL$7(n|c)najD)UtkGHJ&pPm3)N8ZxeOq z-cMe6<7*LpF@ug+!Ed<)(OAZRxHi{n&s$^sYXrmWioS$@?Pb?UeFlAjL}gDt+E;qv z?cm-H(f{>BMxo};6Xf{MY%+4pC)8~>4Mxnbrv%@sxjFUQ*A3CQCO&HFmKMpy7uETD zIPdDm#vkUu$~jt~(DxBd>>1okC^y3mIaW+W7iwjY^;caE`?^>9@V0amY2Wmex!#h_ zSokNQC!S4|&-aP{bknOG!u5F(ef=}pXiudy9NIVde9?BkLH0SE<*;3+r3>~AVU$<4 zB~w0Q&^Kn1c_4nLF*wGG>c37sh$Mo#sgE7i&ObP23O>^>MZAnoFrVLZJ{bJj z8P+Il;D=?g`6?``d(%3R!*zwz7_fQL{f|HC4}DDUgl=BcqS45sZ*VPV^fPU^_S+wB zdwqmiM2CA;Rz?_rd%y%JGYG_mZ^k0kge6clI~^DgHwZrM0me=aCAT*|VQ#Q-*Sdvy z=ncCUB>RmVHhY{vjobg7Ik2mP(KsB69(3QOI=YT>I^Qb43b{Qgr}p&gGo?|{sCJSQ zjNkDVX^VeAyv&Wr@uVJ^Y9j`9KEpxJx0Q@vAi#x=Um25=pPA`nFTngIS{cd>(G%o|T*AJV8b<9yS`k zq%?oK<7C%buKw=b(_nb`L}>I1N250}C~QX%3E%l|Er4sYIdeE{_d9?-%~wNNo?JiS zyUxv}7Pn`MFy>DgE{2WoD#1#JWRcALFEW!LUDyYE*L>$N__gReivRcsU6;y2g~w(x zzN2N3^3m~3Z+;zhrZ$SXEV~6S>JG*4s+I73bpbWGbQX${%0}nQ*>S7o|9n?Um7h?Q zT@=z-*-y<$%ECj3f5zr7exQ%6or^FK9k03rx!-HAyHJa(-0LDZJK0#x@EUI&W=M`q6(0|U6mDdl>!>>F4oG5L*g||;n<@6vb zD>`Nw%5t{K{r*%;6b-|vy*CR`F#LoCJG9`C>MPvaS7i0KJYv5~JWO9eN*(v^AvR&H zCTHJ^v({mwS%oNF&5mk$YyQvX*?J-#`C&^RQ=M6}dHLQ|gLZI0fX>TbM2T&t(tJ*>dAO&9#VDk>=@F2M=)GZ2I8L2fx5M~ zQ1@vm+VFrfhlj?9R6MQr4!Sn(E_ZaJ?D zuRFxK>reJZf(JU8OQW#;K}h|>snY*^W^&_c`;S;g?xz@d ze|b^5s09lyM~=kh{}tgY;EK1YhjMoES68C*&W|RgX+s++t@TMLKjk@Ebf%c&E%3lR zVN-nw(c2$F%FC{x*=7r^!WzskKxbU1M5!;}MpO2D4>a~*u{Psvj)Gkk0fsWPFu)$6lLQ7qlQMc65>;ql+ z_N(u>tL+EQ@z_klo%4hrCRPXoH^vFOpB58sd1v?{u;p~{R|b=^*Tj1zPx#6vSD2UF zg%`B^z=|&uL~Zcb*&0xtimwDsqkHf{-4mqeylTk=5`S)*@SNjI5+79b7Yi{_ zd6S4W|NifJ7K7=t`%#jR-9qv^liMrJ_fsh()m>s}Plr;e6BUd4B)5Rz!99o=ZAI@3TuQ`^gUErd zSTgjQJn2)@Lyx6%fhvEDCoIfILBd>gIK&-1GZ87K){({_pev9Zowq=e)~<3U`jR6! zJ!ial!Swto0m^{Ja*T(4xOO{FhljWVbGctBBKs7F_&~TR-M4?-al<6P9 zC+)(8<5bQH-{cz!^KMAf)pmXGU&KjxA#H{O+oA=UnlyPlcQGT?w+m~pi^4VugX>fS z>+YZy>-9We$8?xIO`aC);DH(&!?z>20>u1vk-*MMH2#Qrz*p)$DWznjbYb-Vcs4Fp z5WMh;2m7vGlrQ;(w=-ib%9`Q=rvi@xu~A2g_H&`6Y4DwA@44|1KA;S$7p=2d z!Mf#|sB7g&HunfczYX%Ibz>uGE4Skicen)U&brUr=9hr&tUsWpwqK}b^>Hk-p#}Zi zG>c4}uMWZLe6)~Pi(KAcLz~BO&t|b#ZZ#_gUA{m7ju@cJ{SwA(+N%C^=al9{u4*35 zU-ur(?Y0N;uLWqC)eZFPeK)ex)}rRsNkhe^4IIB@#7XFXrb_3eiqpr9(qPW*a;U$d z3VO3%QT(3F%Pyif_jSO3 z4}#F=;jWnUGDX)aoP0iixmDH563VjOOW_xs8^6{ehH2VtDxr z_o73Mr%^qa{bwAdv)bxnR6Exf)LZ@Vi=e$^+0R^#?~Waa*D*W5Dpu285P%iY1`TKQL8})H zEstmOHJAbK*$81#Qv>-OpTp)^UB@)y!F|A$NoR2Nh>zG@HwC9INJ4Lwl;C^eacbE~ zIdU#s6-PXOEjamMa4qw}D9*Q)IZYhiZ~Heki0&CLQ6~weBiZ%L z{gh&R5R+1sh?aGpN3D0$aqX8GRQ947oE?bDb}NrVtyw!n&oD9RW%dAdE1`R9BKQRb zLg?AkoUB!Qtr)NSg>b^tgcdJfNPn(Yfw!K{=*$>%G=6#r^Yuds-YhePTqzZ!^KPoq zzAvo7cV#rW@#zR@fHOp6UKeWezQEy>8nuUd?JEM7ei{SIqItAc*;Hru?yF4kA^ zwbz#Lev3cxbN^1%*>yir9nrBoT8|{B)bef&6mWIFFEOQKLOkhUb1^#V#Z>e$tp{($ zi;ziG8?$}&Fi=-82TN$iTYB_38a%;SD8~`7_rIqg8@U|@)}LdZYYZW0@9f1Z$!N-S z#vwtYP#iJ8Kcjp5bI`&2UX8XqPOI9;Z&qGl7=p*=kb{>35D24s|1B5$u*!Z?T6gZk* zN0bD&>cffsg9=n&xdL6wJBD%%gUE;r@h~@}1@SdTkkeKX`0Mb&_{>(nwS`Z7){q-1 zR^)VuIJtRcpH+X=8gO@Pfi8`~xk8&+wrJ*-KX}HBNieG=9~M}JaBXSjX|VIjz4+Ep zRkZ7(0Z!L5;;w)5uO^Ag_lQXT{U5ENDmfb@r8nWHeiCHs($mlr4uy4QVAlXSM;qk;9dLq-GDf)iwq%ULOq!)0*+RZ#(ghhyWZa zt$^+{IiW{-J*YlcO89ux2FPf(f_c$fk;msTaPEgR>{DUm^3}@=N9f84OV;1CGMd?L zm0^bPf%0hR{gMsMLk4ZMYR_JD#pM)dr*%g9u&IL&M!JQhqC<~tO+F>komk_Zc>=8k zj$9enav7{(vkXtJI!CrIcP2$iYv7akSQICF5NB$=j( z)D6V)yqf{1s|f8QqIURqRZ6+`B%1>`T<=FtFo`1*$hwMf=si1~;!UE- zlCxdd)8CG3Pee;QB!dL3F{EOf2j`IUR$Ibyw-u0cJRRwM2*wW{e!zLc2Lh!TzGQCt zR*skE(_*BcVjhz8wM2>alSS7!`DAz@*d*vYhxZQta-Nk5#upBa#`f+U$NO`%db_WCm|TE_^8*eC;7tR6W8J{E49&?w;iIvLxU#o)BDw zG#36w?j>qu(yUjY8`#KJ8QpIgkIVBBQnTyEQE~I}sCQx1$D*LLU=H@Y zc}_%2WFP#FdxZ7twIQFM9wnPb)nY+}G|4u(%H8+UJxSdAhLNYu?{LN%TOuPKN-BdZ z@ssA;C`Kkgq(gq`T_fIqT}h1XW)&2p`-L~R?ir=I`W9+kJ4I-*VW}|n>K@^%1KFS< zcMOF`eCK7y9RtI8Qgp`~ahRX(2r*OHy;Vo&pt-Z6n2Pn9s6&d+Yf-*|lXptM@|g%E zzUq;;y>mEh+-nzd_yv?s_-8+Yg$leksnTHJ`X3mxKKWzwF5^qiMlfD&66~AzPT)Ol zII7rUNwwu%V9d6vQ(iL%=Z0ld=HZ*E4Jh(f1+uNT0&oq1!v{~2W3jWz;!U39{)#fn zOt zlB+k4aV`l&a?d~W9$fIipBqb2+g)!sZ77S%5AQ;2nit{bp7U(Z(%DRqvm<6#fXT&x zAQJE~j4)ygNXXvrIIuMi-#2UFu^5Yq4ii~6@6f3X>PO)PwC7Sf(p_H-ie3pYn~h;~ zs9_~J%3sZuJ+X(4o8!m&XI$?WpmkfZ{yb$7b{q}4|0W^3GbdoJbSGLkd@QFwZ_f(+ zNXvpaO(J;HgHX=KJgSc2`u$$iz|sS_`obG>^=u`H4m?K$T_@Q6J|%G6Ka8tu^^J>P z!Y6UnC_i`{I(SJ8t(iVS_}*1Z`01!NsWsCB*`#KobZ`+4A$RrXsIvQ%D%iv9P zJB(t#$3yd#QN^5zIQ>Q@_1v%l^?8I6m#D|&PDZZ4O-qc!LB#XOUzf?U{PSd*>?h_| z`v+X!9gn2mu18TT<4Cqy94_T$V#_;S+}M=)I0_aOi;-mqX<`uQL@kqB1~R#y*>$+h zL^Co6%XkJe;sz4JGs?EYk72in{H=4G3WIA#iSREXGY96Ox4cU9$j=wfnni=z?*i=Jt|j7K%ph_InU{Ya|5u*~ z)EjSdYVRd7e0>y0t2fk`?E9EUVt+_;&mr}BGnpDQA972a!a)PCDaQvFu;~UjYWDCN zYGp+@$5Zm*IwYd|zk{Dzu#%%C`Y#@AAzb5cB)mF&mT*PQ*~^nGr19x{Bgrn(2r^Y~ zU|+rlU9vraS>dUSHB6pUPp%9l`^W7f)R%ZTcg|XPcMRuiAg0LjvxmhCik zkekP1e9Ukr6C1+Ly??(!u0$!piXEpwLdBLSj4(jc+&zU4UYQGXqJ|6IMjm1|iMivf zO#=L@Yzy)7v>-$$2wMe4V6oDR)Q;vOl&gCxR?1DIjFTth%1A>{2t$6?WI;#AJ4`cI2iyWG*fUekPqZ)Q! ziPGt_sMq)^&*H*+D=K_G7T?=~kI72GaRp6O3E$Af6la0!&6~(j<~pS}?hu^!T~8I= zOhzu%MTn}3L@!J}a@cAqWMZ2$x3FQX7svWBip?XXS0|6F7b>FphhLM6(+1;uMJL?gY@(_ziON?#B8a_}hZ4X6 zDV$vl)LmqHryjup z7TNgDG6P~a;{jpA!%}m0gy2h&Kk);}Ih46ivoe9(TVpbZNlh*o)q$;49vrCn;lgOlLGf4Dvgs5|fFDc`o+NpaoI2vk@g`TP$c=9*+2&Z|t`z0>VKllLd=x+We^h;UJXhcU zI8jKb%#f8tWhLBmZY3cUC246P(x9ZFLA{h2kU5wwR;-BkL7YXXCIN z>g4k6u~=uuWb(Tt3RYg1{RhjxIVm>L=!A%Of(_Tt;Mv^-@BGn8)D#@q6^|TP&nZdl z>HC}bZv~#t9NnjMWnww8TlhxhA;9;d(3(9NSX7qW?~cbdii!Fcyq|kPXyL8E%f!jr z!FbQ5Si<}+Vybtk1K!~XDQkAI0p^;l0yCeL63#;T_IoL_{tJ{=We%%Y70eHCVp2X5nkt+mb%gC^4bWJejorw_4vZ2O&PE#|1HTGoIo`4x~b}r z$=2X35AW?CqmLWP@C!W+`mr#ks!VrxRhC>(m3Z)6eDJz4cDCU1GB#D<#K52U;?+Rz z+o+$w#IHy!W;g1TFUHN2cpFZ^I6c*>|sNHw&Bh5|73ITaB=kG#xwHj-B1`WRE3K} z9O)Wvx`LXWhjogKM$XcHWWS>_7L=vO5k$BDZVce z4iSOsLWi@iV5K^n#2_gmSKiChSX362NWrbbb71e4$i?-0ie$CZsT_@a%)O-vpqLzn zw?eJ*R~qT-8|LJzF3$YBb~?zePqhl~dbxI`eA^`8kiBZ5GbnvbigQ zQy>&R5Uw2)Y_(&dBWp6XAJ!ezU`JK5kk@kw1gQa_HCh48#6BP&=WgmG-AqZY&*I@T zd#%Pc+FgUPv*Iv5J(o>?oyXn{+|TaLs06W};e4H{2X$maMnBMQm*8Ne7WrfzO5%Ec z{fD0j?!G)NtR7WMPPL^`&(?;bCnqp`JU56vxzU{J->JmIJeb8# z~Rc zL;IdV?2TdNe4hbMgRmzPY@IN)nasS93Rm_UkvXH5pv2KFXhF6-JvmT-!)l{I#U&nE z_KspdEqw%O>VR0%;0H^$5S0h^4SEx z5`V$D9RfHJ=mc9=5bz0gVM*#twsn^T>sT%eRr7X(Lcl?2wvuK$0u!PAd@hflnVJfG z8sAUkzK&)y?;5e2_ZzYDp=GFR)iM$z9uBXcW|6KBJ76wz0-Swy*Z@s`wtA))Ki{dE zbEM#bCzLdKv!PzI*!ZHi5SW__$~)%4M(s`lH3<)kmGG>b^-;ZV$AoR!ByY0Qn4(cl9eT zTYY8{?7h0ltl&Ju4mn@Us%0mz6W8x$KkxHola>tUVS6!cIWPb2t@_yIu{mg_eIu)6 zBw*6Ixs@8yYGiH6Ybfc(@KctH*?z_tHm7WdLf0g?XSf>dh1_0~v%@Ph(PRHqD6s37 zNU*;9`ry3TXtt{88t7G-viWv5;PZ$)XuWPq6x<&(83CR|t@$q6=*;;`3=0OdPz=sU zoPbB-oiMzd@UZBfjfAVk7kIiPy@&_@sq@+8{<1v07p{KeaZ43AkQL`1LTdgQSTZOL ze6(Vz@6Uz`r`FDboBPg@ulp*2`Ch>HA);yds>{sKpY6=|h8N`Y!vy~S%U^SuhbwPb zD;1Qo(^`(OSt*h1v5#}valwd{Tp-Q9+u291_Uwg#S#eN!>NNyM^~2y(4RA`j4K{7x z$M-Q8zlGrjtC8SkGnU=$k;=yRZe%yLQLN{~JaXB8CW#9?OlDK^tjC6DfBO%=h>JHaBFoP{CCj6= z$-LSnsHcb#_N{skqt(0kIwsC7_(o0#R#@+YsTen$xL1;{e*R2EOM&0y0A%HPfO>VV z1wtHOkGV2IgQ^d@qDQ)q(gCW z5C;*YDt`+tx{-n|{E?uy9&VKLb#IrE0TAt3nDh&3WY6nzuMEXE2>v*@|Xt z9*--!v~f?3IUB!r9;=f77S^u2$-`wk;tGB|)0c#Ol%m$`en2eB%>J`k8j}Vc=3u=<^h1HWu$pvvVRO4@~hiK)e)?tIykpDYznl&Y&c4E(7FGGcn>(PjqQOGFx z9CM~Fk^FpJ!H?~9@i!^28^za~v``o24d|g1K`9w{dxz={qcl1G=rW*RRIfsIkSa|JM0zO%*Oq-ku<=!nbnNSH!vc1@V42+DT z$5ngL7tAARW4q_pVK=QrF;D~#&)eci!6qzieG5q@^pbN924tbSB;V$ydhhyHCShO(OUs#zYl@Z!y@lYrf#4tpDX{367AhYH`B}rCG;s4*;P*GXq_kc8X?I3w={CBNjY~T=4ncyGZ)m*Nf<sUxUHE~Jxt}>>KwRCdk)CIfYkS=)w3%)}lf4IQ@1|3~Ec3!?}q=&_5hYU7bng1_P8lPKac^+R!d3 zQ!=A=Drqm}d_~VCAnl6Pp!8LOdaN+ca>4j0rmIJc%#Bt+OM|`$^AlppR$&s8#>SG$ zZ(h*dY|qB#?q!#@jfQdGU0L-R1+1j=J~Hux4{WBitZX)JV(&|jW9`q;?2heWXlos& zl+vc*2X2$um-l|d`aN!7oY2BN4gspfdKfX7aEeT|93YA-EYz;19|pR*;V-Y{U#M8w-6~T*nn&HSFkrNTG;H#9LyX)to_TWzBl10e@-ouF&xVI zT#f>J^-3K1>M(Z9tVT0Uv+=2r{a7=n8QplSjVfwPNM$o&{&=*a1xxdho#YjAc6=L* z;dCI9(G-z14kO(whLD|^X~;GB9Gk9jnx)_@+gaw01g>*oLb)8PG<7PBYRiI(D?PAB zu$27+G~4?*2q-y4pdJ+w2j!tWuK#$qtXhGGwXLPsI`PRHo~Mx+&HR05xt*vkFUy32aMes@FBL6fckQsn zL)W@MC1;zwiNrA_A-A#h@pU+`^AZx&55t*WV~Om$SY~ydE`8&gG@YIP60h5Q5=obvBQ_Ko)JZ4-=#%(j;$1qe-XvRZm(fxSEYi4=>G@9`!^?Ky+4eyFyYQawsgUvtkLYG z#|wdRTM63Lzp3;4>#?>&IZ4DC?2;=>*=?zP{9Mskk^~>>M$*6WHV`xl*yR^fp+?ge zlqK&$FRMY9R;GY#y&7CroJ(Yk2SIbN9LQ~wXL~e)Ae}q=8S%J=Jo0hp##jd>k1oQ_ ztG}Tn~HW|%1&H!TvyhM}P4tvxhi{mNL)7EwhJ>wX{>j6P9fR)8<<ma!9vtqt&bN)c=uMxDQlV{GIXeFJN&I?S z3Qm2z8`~%T!OmwNVs#@$ZUL~Q4R-9qYJJ71!P5mVn9BK@?dRhC#BWB#IT0(TEamrX znXAQV33V}Ap-%_ew?BlOTM}^e*b6Sk$0Fj#429BNp0KC?E+j1U1lz+-_{<+ey1!kY zPV>A)ENfLEws}0!nC!{)*>f@NIT>uI7)4LfPr+x>f8y&}oydTEMSkh#_|(=jxbx>s zx@SW){p{Ov5uTzoM&Pw{IEl(xOw`}ozBLi z>_MOI{$h$3*ok;1)^O+V9w|;~2R_~vNxV3}&5~`VlvQmV+A_-@>1Y=Uw-|5Z>*mb+ zj1(4q=gy)_7+LHGXWqWD?x>lI`*X!<|IS$0s_cy7V)vnM{@GaR#7q(HV*jqeyS77} zi62gp^gBW0VOh!8+>*fgtY3JeAhfIjL z!SVim5|78jtJ4wlVkh%_RXQ2*(_j9oOLOFbzEZ$2q31U>ljVR0ZaI*2OA9RT9u| zLVBf?Vd?nWMB$quluTa-QCGP2(Fj%xDx# z4mQpsTiyUsy6HvUi64VA>m01Z*^J4UX2g7Sf;)0~B*Q+084-RT z+gF{$1>xnWJ+S^i`FCrrqM8Z>cnQ55yUwy@Pig)HITe5j2jbw)5Cb^bYY&&U9w(`- z)#%t8WuER}P)n?}+{pBu*}(2xL!J&fOe&nsMDG)^d27;;xYs_WUaf(#ix0M5Y?BR( zO7Dus6EjzMh3;sH@qK9S5{Gu@-=y#J8LG0ontB@hg|c^OBHt2^#MLZGc<{T%ZeYKi zn484hyqvu+Ud#$=&#=z!;#A11;pBv+9rGu1B1w-J#MV`mfv8;q*g*emEs%Xyh$4^- zp1I~UCHuUNSs5{$%RlFZC&zj-65_6+HRHeeA$tC~w4cmlev$`rF=S^De zgtg4cpEmdy?M3>A9Yh+#=fiLo$UY}Y>{!)7P-z;m{w9vRuNgu}$4jO^kzYIi!>6VF z136GIo_SkS$>UjS_X6#gP(iys??s1)kH?>xCHS8EV`M!q7hU^)1if{ej>p}+P4z#O zWB11d!sJ+gbo5vZKHM3Dzx22xkEN?|%%hDW8t`L;EMEe*4O3ChH5;7VlmT8}%gEOu z=dlTww|JC~818u=jpkk6Po%DEfzid0pcVJ`TwK)t;0BC$Z~Mw*kzs_13;@s76JXU{ zKX~gRM(pQH^SnQ7c!-&O{WMRjP`TUqVm)W;esDBBL-sv3JtV+99CObhE=2(vn(oT4 zO``vbA6kI)s?Rv=>j1v*`d4p%-e)TF#9fcvv57}_#x#ri6r9v5CfG8YTwbyT-I?-t z{v3=Phepr-gLW@VqkbGYhH_-D;#sScuvF&jG1P|nGZnd){OZrv0TneIU_ zKB$O!wN4w)t5`{b$_>F@2k|&{<;%k__jc3^^C;o|Q6gH1l?X?O?t+WkQt3FUi%B}0;V6WdGTK$JJEx1k*E4@lm;JBJ^`zr$T^S`NYD&0S194)IF zNZU_RrW41#Ly_ovbN z4M3^NgbTZZcwLD=Jo*~B!dl@kFBNF1R0Xr6T)pO+QEk##Ek z_)58tz_y|nwir%EdFkfFu3#MxiwK^GVR<;?2BL+LKD4jjHkvx1LkGNAjE@E`!x!ge zprXU$>1o|u{5ywdlxe_e>=PqRj^CR^LPn0F)gMh0^)L9)cAOeGdYx$!4V@fj??L$jcRnmwmNGCS$}jd`xsiX z@&(eKvzhm0(;jf4BOd?X_m>D((UngP1;un-2%LSB%+}@fEm|L_JK5P(mO~NFpP9hZ zX`HViWAal6-Td5%^t2yxxi>Chg9n?qLwq?py2})|^$wyVa%#{G8BWi{_4zU46;>Fg zfu1bdNjzi=0Yc@w^`TI|P;aakkofCgV)| zI;(mvc6qan2{ULE{=D4EGPbjkZvIJlLuP?!(gFL)}<1O&0&`SP! z{Hwov-QBLIkz$1^UQH+Amp?ETu3kZozg9&r)Av(S7SU9F*+tU)+X5$OIPmi7E*^<= z2WjE^<>TomwKw^3T>c!y^)b`%;}Bz%Xd6o(zm|ZX-uI;r4bMV_fp%y@wzAMfHJct< za0q>En+D5gp678T=OjVwE=7}jEI((9w%g%rO`+(|RL{z<6=!kh$}iMeztMQEZ3XI+ zB2Qz$MCp(xwHdGC+%it70n{U zC(WR{EdTEJo-2OA*QO4q56|g9o>rwu=XVS`l&bO{os~6wh|Ht~bcC)v&B}3k9vrnH zbjS#z@pKJ!^Hw0^Xgm{lH9o>oJ}tY}uTNSt)`Sdq-lI&9-=MJ%#Ptl9Ag>TckE*Bu%pi(Dat4 zsM^Vb311YCy;m#IZO2ULybW9F(1sHFQfD3Avp9qfUo)3}bwq$Qo+r?Qp4HH2jx^K5 zvsPifFC;zO6?&wVJnosKk4?K$c^pqP%oWj7peRiHPfp#h?Zk2Hl)iA=2UPiUU4Yv)F!oyroSeGW>u)a6R9j)*ULU(S+aCuGZ zsk4SnOq7-dTDaAdKU2W6!Px)xRJ=DK{vTZhH)qMxKaMVC{bRx*>_k|N3NgrD`k|Gkrs~-cIA?=V{DbT5sey9v;oC zcvLV8gtx||Q%&wHHtScUEgl}BeR7+bhQU&ldchE!GqaxBC(%a!H}?dx{ZAQOPh4};Pyv)M6x#*SpvGU`Lt25 zjfm}*BHp+Q0fZu|Et`LTNA#EA#FdZnSBmBPG^JA7#$YRX0^d@vN2E>7m+jMZVUH|Oyl z{nuz~$3fz^G@l%_?M9=bKNHD*3~x5fq+f(gqBV|x!4viKuykA;WByte&7UQUqh~+j z`TIiuF?-c-MNO{ zR9dl3VJ%)X#s!~fO+JRtIFPdue{BaYJC)$d$3GFRF_xECi6ab7U!cv)!>8XnaN(rgsK4d{ z(LVE;I9`=Qfj$D1R`|}^$sv!q5}QYbMJ`3-yzk*DGVAc-DHYgY#bZp3n~F_d_fjgB z>L8+tp#Ls@yoLKld~twyz1QdamU1Z~i%wkN>Ce>>6J6snhLM(#B&M+O8u?JpQ0KXM z8+u=fR+A3F&pc@yyig3^9Y{eZo1QQyrca>tyGPK&^#A%#cNfc`#ycit)7?ZkLdij4 zlp(&c%nbZrM#IBZPtf~>Cd$Sv5JqUfXI`ye!jJpmTtAUne*h}-{@}1V5lsK25|Uub z>8@HO;C|>0>Yhyqx$$0YzPm!!y(wWYiSSBrME@@it2As%&F+$;c7ph?OSt ziJXKiD>SXQyl8;BS4Oz@+bNO!@pK%}8HLL?$79on?daFrL(nF(8@*&R@JkNHvNtDi z>-uNHDHqb9Vo?Gc+I$i2&yIl|Gxm~%_3~I^zCPJCTpE-!HQ~ygRJhYK9WK~DCPtS2 zaM|)C{CxTb!tcn?q=hDhPg)quiKofVOSegfNe0=hu7LepGC?N)goy8AI-k4HitZ8E z*4z@Gk+mb=JqJPRux6~i;{o#LZ1C74EY|tx-CefCb0q395_^pf!Z6bb^k&jAa z-X9vmohcuKhJj*uuA~5l)eeYMPNT&hl6iRoPKy}v?^{}w1)HCh-XH9+}EzM{G}Xe0ll_aMCBB}{MbfI|yX zU?A2PhK&;E>o(*V!=rPPN%r#jQ0{I4KF_=H;4|W^+UpVQM`|fj`96%FbHd;fTySR@ z8aB0n|NW1T*Kf9>k@Qt~c4!bBsv66#neq+-r@Daox|_(gC>Z`6$ims1Tan7occMMN zK-6wznm9}q#_{xDJ+n+mUTNUEFXqJH-3}o$@*|};B8qg&{vumX8iT=DSBSgtjEoHU zL_W%xlfzFgh{hMZFj`844^~su3=d}7wy&hnd=@I|U(C;2(HKJ`ZTaWzHyn`b)0)bP zcQdWU2R1X_SqQB)+l}HnzoQ2u^01-YPc+SK7}jGYinZOtzHxQ$`AL3`;S_IHjHD;nqU-YC@k;trDd*j{;A{tgrD(aX&HdWaE| z8;mDNy5VOZ$B;f4YCWQwi~oPY8||r7#w|;Gkf^P{(s5F(&iUJO&(3Xgu*I_o+-@EV zmThAognPeDjFN#u-O*s)e2&zd4kgcagcDuPHbGS1b#Mz&P?qFr=wEY(xpI68<9pYQ z|5g;0ulXJ?Die%%$RfKyfw|@wPF@~xVRnDuVq={zBu}wDi9L3WNSvNR1O-*7C~*q2 zG%+1HMBNwGC)=WfTi>F)yH4W@*=lmX%Y%!#u!y=5auc z2oGK!*40&z^!9IX2?we1%|CEr5ym}#&&a&Pl$Ut5hQPyYg_#@3PEAWoh%Ngf6 z+QNatXnx-iE)h>ULbB0>Q$n(~qgQx5Ef&92)S*J+UQxpQG=9B_{IHAIeR5}9>yxZQ z^`8o>G^XJhGZe9HXchk8%WvDnwE_~wTMc;7LOr}4H8@nq?g zQWA1spY*@~K>S;b;kL31INo1|qQ+c72R)N<-`w+**p+JQu+CTfxMFtY)_IG=QSz}v zWC~~d=wm@sYYnoPt;Ib^?9>rlm3FnisB@G8UA&gsRfyC27a$0>XKSuv0L;kL7^LNZn z#Zh?cG{C4`1?}VFr~RDH=@bs!!E0)N;all7Se^6pJb|Pt6@PzVWVWZ`>UE`fVcI8t z-c5A20_Whpq?C)ZE|P<46%%TpRvR^)oI(PoSnzP`KAKETkTt;KW;Hlzy(E31Qi5NT zcD{8*JFDx^xaAYE-FRgLO+cuyxeKUc2m>ao3T%E9-epT zEOJmDM4kSwfakX8TBrC8VrE>G#-ldO;M6g(sX?ILl7P?2xm@m@wtA?S%~N5pC>_jdI-z>{U)G>`%+d}A(a2hzpxy7xe*;J zTZp4>HZjTP_mXvsTgYph6JT3&9JSgwkY<}q6fQJDh;Q*+WGHk zF{hxhSYk+F<+PvsQR22Xo<|}$rLrQCiM>2-^}dHnWVpOn>~h={ArI@X*-|F@T+Zjj zDR_>U^M80P*g?=zwTb9N)MoV2+=g)*ZOL5!p@3${2BKxRYS5k|8vm_HPftjI|DN}F zlV26KD%y%Kq3>{8(~iB@F%LdxZG}dEdE%e>9nJc5muYvO3|o%&!95uT`r05(e6DX9 zxwN+d2U;w`5noa<@-QZsU)Wn;${0aDy@}&|a{PJT)u$?8v6H`ezQ-;~B*V5`B9iiR zNT2k6eol&Tbg54U(cRPXuf3t@S#+^qwfR2i<9A`nG(X(2MG8L(Ilvq{c@+(}w@0a! zolIe!KfUtH1-f<11A6QgO>!W54GDO(i?LPjLfKYayl?4C+}?5v_CEWVr)krLAILmo zCMjilm@DJWt+)Fp@naOVTjG`5xLiQVSFqSU54Oh6jYW&Hp?p{&M=fMzTpv zEJVi0)A=>?;mP=e(DG#wihY|w6o>A?eq3J5;(0948}Gs?mq_S->4|NmW6`;WVCrg9 zK4s#U!tYy7#7eQM(%a#a&lQLSn50Ho5 zF3*sPU_a`AI-gmc8v$EeZonPWc|=h;2`YDAJ3sb|3+h~XiDY?urjo zk&}BLIdHRX?(swNZ-5V~?<*?sHrF;}awr6zyD-)I1C*)Li-qL5% z^$1T}LB%#`j%frnR6{~sRoUI76pYP;NZH8=-WYM;iejNed0{rJydlN(IbxFLZ>{JBR*iMW-M2qw?Js!65q?vQwIXo^hO6Kq_XAKpDxwxak!aYwOisDD*Vr z_a4x3f{Bn^!9+%Rk<+;buxA(-A6M!w*to=j+Wa8CPZ53ocM1MIOPkBWVug5;W+KOG zOv(Xa%^%Paa`|uz(nx=uD!K7)I!P?y^0^aDh#5D9@5gSVCArY$hx*04;fVY`uwPgU zU$e}??o=`?dSeBPn8fqtbFcIACcHA7WsJwO{rX?|wsiS280p;!t@&?2DeN4`P1gqB z9VfxYUz6RixeJ!$?qZL|sTx&T~Y2xcxN+|0{s{WlJ6r3F{eg-a15 z*146J58GrjJ{ElcH7@_;>XX!;UlY;7)(|pv?OhVq^Y`01JvR$|IjDsH(QJPe?{g;> zl6BEq%2OLEE?^QT;|%U0LwR0^>g{Q*B@nJaikHlzPNvP_#}dI&zpn4-|J#mF&bFyqy98`9fquxDp4n(;6iy7d+CkI*4J@7#Tt zqjkF`A-mrjLC$g#3Nv*%X>G^;rGYu{Pz+Lu=3eM$Qv* zk@>Ell#7NNK6i8%t`{5kAHBa4cVzg>pK=z?6VX-Rds{%yRvtxP46~zkWo@Y90|NNO z%)sA1+tcPgk@(HqOuVu6F3NA3K(bO#Fvp873ZFdhMGl4zXxyGjNW?#HhjZxRPA}2? z;Q4n(btI3=5NR(YXwyLn)rHtHbTYUmr67+@<;dbk1e&z#KK^xCi_ToGPJazphqWfG zKr3&lA!Yfg=t%Dar1NVbYB~OeRCx8FY}Y!}Hqwqt&L|TGEWSq)G@fId{ye-R{Q?@Z zLWT;;yHBu623Q%tMe=DB()Z}W)w`XsLH~3Vy;PNGg$<#@4k@CAFJ9sNLRZvKG6ltr zeub6u$^j?D4 zA~!Vo#&dozCMs9ld;^XuQuqgb3ojQpLWOqw(c$cELg(gUBGYvSwOw7v(@9kK$&WQc zyFf#F+0x;3QSl>unVrvsill10cxgEO}(yGrx*YL&o zK-Md))uqStQ{|OEUoNrhC$U;kVm-kjf*N_coW%J0^Rl>Ht_V+2{EQ!oYvQmM!*GJ_ zM0jhnmx%D*=7O!QxG@FKoU^judYK~I;nq*G7SVE0vh)<%nHY>{vh zZGY`e-Lvhb=2Th|yyZQ$l(Ug~cy~J+zv>Wa^Z9%3Z?CG(9`QWMo&AkruO5CS{BiFj z4`=GqU&8cz-|=kUFr3}%DH==gzl$GVH13Mw8>x0<41sHofMT8p@$h|uex>J-&Wmw$ ziUv(DOYg%MgC8K3R6jg+!b_}fk%3Z6su)5wfOFF*I@E3=8R_MZ`#!$ma<`;|?sGqW z?A-N#b>qk0U*Uf@q!wAltUe7++?uxDDS@feYlHz023&*!i1#i)x9)~x>D2Gm5ikV& z<{rTh!sF4^n2kiEmrzrOE~H1>@y9(}d}>pf1z z+D8YlT3J5scm6>gUYkl%^be9LqdiGI8j2dnM3aWqQldF8R$RD|L~-`Z^_(r^pmkd8 zyTQ}hrY~D@obL*}HEFi>@m9#TpCWpnV;nj=(;MFTJI-*Poc&L|tOS-e2> zTSQNumTE|!Sum}fOuO;d9(=<6QtCWA0ZRuBC6gX-u}-!2L;9E;nDw}b`0B4AziBaE zrq1CV)XLB4#BZ1^wkpTeqOk3#=|CbZ{w;(Y{TdS8dmAgf(1o0Ut6V(ma^kwjhphI$ zf?QHe(eW)8aB;LI?!I}ENcp(n2I+D<$)p8^E{vi(UmT#{WzC~sbMZMG3YH@I+zg%% zR!YwBeA{B|C_WkPux9KzLmTTO;jP#_cmR2Rx53>xdI6 zWL--qGiCICo_0T7!w8BHL!n09XueMcu91BKzD`MCwQ~>};P&{>Qyi(uHN4HDnCzSO>SO*GADC>g+Ys;*XS)0zs=>7$!o`BS6@YLFQQ45#}0fch|3xMAOyGmEJRY% z6L`K9I8P=uoNuR5pP5MR#ilo<6ID5i(0fYBVCzB>Gkqbhad*HWx{K-M<@;%kH;3t_ zHUDSd)(IOyXm|*g#t5O#ej73C9tro4tK*#RSQ5P^igcuUF|N<_>CHcz(3_5zcX+uI z(Ii?%0A&%684V+ZQuU)4Z!QkA=>N8}Bha7XY@(LBlyps0r>Mmp!Zivf(NjfT`e2_r zopXDas1HHyoO-lp@)&g4_cpqcD=o~Jk&H%7Hbc{w&qmfqZ}9SVaT?uc?_ zp*F7*;^*p_iORe@q&#wlKy`1J_)3qc-PlG2sC*zvp<=M({3hYbmYtNl>R(;Mgw;36 zVx|bLVN;-P+w!nzw7ZbWsSVt}6VSo$lOaZp|DW&QY}0<|-L)B3`mTTnk3IPMUhmBy zYg!C(TIh}~myCl7h4~P?QX3k6Dl*A!retKN7^+lmqz2SR5^re>7*Uvs#!xjV?D-z+ zm)#>kKCGHtFEAyej9nmWAQ=w*tcT_E#mL&1L(vV+mimM7F*MU}2tK;F6cJ@-s$*6$ zDQMV4cI>%8T_6Xjk?tj&4}Akse_KeRE$)&jra8=l@<+tYe?PTBcpd1yimUPeNsJ``hotzEeM?BDfNROiE-ct+*J zTE^uW4YMO&Gg?Qb$?e`E(zfC;Ll*sg=k*;EM>e+mDY1pukt9AtG_Ps%d*pb#FC=a0 zT#`{|hpc9Q5%EWGX1)%NopKyhfY&F8ofcQb^ej2Dwrz-Yp~eW(sAdXf^AGbd?sRfs z=J{H~{kYC&NYhj%hx{+ARk0T&R2nix(CxuoUq@OSrZRdo+%a6gRHy2^D<=_3z@s~%4 z;dyBy@mdw^htyJsUq0dIo#3bztF~zwj9u}bV%H{B4lh-JwcFpogdi(cN9rVN*R+xU zraPIO26Ij2piA!e|CS}PnL#xuyzrP*toAJ*yhc=Kcd*5N!nl|^_aTknMPtC&Ee%levbmH zl553Mqc?)X(R;*U`a-1rCJ$<4xLg>6&rk*9(x@Sk@0lCP3h;T%d-!>;jmwKO55E1# z0K+NQz$VcS-U(Yssd67ux9B2!X9e>8Vy1+g4%TLUzxBbYq%$CNT>s77BOSUu4aQIlzqH$QH(rVfR+b zu)_x(hY1DsFs3XKV!j`Tp*dx+wr~*pUYLrs>@r9!I}#@J`Z208f8%^xN2j1)zLH3( zXdSh#eE%u9!bN`GCQS?F`F!+IIXPWp#CFeK%T_lZ zV{LO3SSh#BY{UXxR$*Q^o2xOK{l*19&Ar`0qSgM|3@2V3%P#v8%l9{??lP2Yn#s=F zD9LIzDw4-O@4(|*0yMfxfe{x|y7Znnc*oeYPEWkq$iYTz(Oi9UsD2ajKKvAQOw(a+ zX}Ghjk|Nuv{+&t?Br`j&rxWJ_8@_K5yc-_gU@dIUvYXA1!0l0kNXQ!ldPDaOd{v^2 zGN>%)>AB<4aa`IjNsrDN&o=yOJMXhH2#MNqm*A??F=3@mBK2#Z6Sy{Ue6`~AsC}V$ zn1TyCYVu7kTu2-QU7kT4YRYlsn<1>p6mfR<*oojj--x<>_YT~YxDHw2LR>qh7^kIk zxk@f^xmh)g(YVva@bf@CPs2}}P9nuah%L>20OvXs@vt+W;rBo(EXinuEQN5Gk>5>Z z<`*+f>k8q)?K}9u&(n-bS7J;c*~hSJN7EAg=<0We(u7@x7UR@sAJ8hrO`N}q z9V}BRA*-DtNT|jD>|JmGhIyXA`jRe0Y)U5no->>NAC1K5jqQ-%BSG#A-HyWbT_Akp zPBbHADy=x{3EsTrBy?K}!JLboV=n(3;)=)M`11v5Se6GI|1=77*6Vod^$Be6t+!Bp zd?M31#~S}o)uYP|i*c#673bsc0?+f7;+yX4X#HVs^x=p$%B8`Rw$S~Cjh^3zAEuF{ z`G_*O+^{4A+A=Jwc!+gaQOIVj;B@dqmrxI1%%=)gtmo-$ETaUEMvbHqlt@{b69lSl$m)A6^#{p8G-p{y>wk(|DU!Op}B6ZKQX>TU{kU)B~RE^D$XpQ2fr zNm1~zqZPLPkmcn}{f;?1;CT^KQ-9;Ss?B7mcn9@BKLGz8yN-p%SCDq{CR!XZ8Y;Y- zNYr{0{B|A%6AmRqK&Jv+ORdLU>APUilNz`#<&CCJc88HGjUjWK7gU-?z&m?uIDEnv zAG$J@J-2iUtG7WN>jk?)_S`JezdC|G@>G^CI#Nw+cXF|Dnqx(H3iewy;G9o??WX3O zsij+@8fm-iZZ4;w7rBxeP3(KzVEX>gC{}DAl8m~9VuBCQcZc5R`B{*01;ecjeAVI* z_781GJ=sb`Twc*~!yX6vO~V-M6TTJu&-y`b7AvBz!OOY$gvn%gK4+sO!=1&3Iq-BA z$%UaS=L0>}6;$5_L1x`PdVJIY0>3%^eVTy&rV~Q1o>hk1S3N->cA3P6=UQ7;PsBe0 zpWwZF1PBG5gALOjQ$~x;XalRw!UInmNnJrEu~W%IhtDbFC%ax#JxbFu`9F8Pj?<%5Y~6d3R<8FFziuH)HzI zV*IVQ107DQ;>Xn4d>qbvjbeVEc}|krSMv3Sx^wm@scFGLKPS!N8-sX<&52{%~5^nP^KVp)?G)!wND@eiHo9PH5zbp%sEVCN7LTgpHTE`5vtFeMtONvA%*6nI6O3kZoTY@qWqH? zO%rQo&DRdB(Gbq_?RowRl8GZpw^atUC?N<zMdp#mXMFqrP>Nr`~(Z}RjZbb4AN0FF_tBmOzYq7jayuSLAycSfE{PC^4zM44C zyj}JJXhmVO?(;aM{2@}A=x(We(9?4B(Ycm!V=ggKcfK((@1%*bqOoPu0vF5jSx%OQ z`Ju$==`mCQX}D3X1W)@%C%Mwm-ZEv?1j|Ql9dwv<195nAfpWZ9g+98D=jZ#migZ-* znTrRQ#>HUZ=1)h=2)O#Jf&_DZvh-Tq@t8q#F!kgH+O=AZd>-xq_rt%EpXJIBvuYj6 zsEj2t-u0Yb-5|PIG?q3fUQDZhO`zpJx6_%3GTvc>Wz6w^wazPfOpur{lSd>B~dXgSiSoQ`oe3rFEuqzt`(9D~E`{Xyt4 zlgHOOzmmG0!r5dmP(U#~r_t)936{^UX;`lEQnK9WbDzFE;2_<;!hznl$P7PkID^bz zhu|?b9|a@(OES0TPDX89jQA}#3vqbwd^#*8n_f8RGJT8963eLYXm}%aXYFLl@PQV1 z{^tHgaQ-J}J@umB$Y{_nRzJrxB*XA_`F7NFLVy*=8CV|O?QAJ+Gtn|EAeEZEQ9~^E z!Wq|mIa^D6e!ebP?#r^eQEdIxP3+SeY4%(Ba}caehEJb;boOn#FnN;ZW#@GXh?__;}Bqn^Rjz)UZTi)+Wg$2+rG6GM6S zWS=%WGQkB74BQS&oTo$Za((vkyKyY|xUyV765{LSGnza$vt)k@tXRvdPl44MPo4$d zMj}UoqK@9C>f#@Y>@-@?!b9g#A@{#;b6hnKo8lc+KAoy@jOn;En-adOAmI~MGuwU~ zL%K~#s5U+pX?@**RATxlsX0M{8C#8*_eu3UoZ@k<-&m1rks4*KeM8DJ!Zgc_V?D{; zPfkSj(mDu#BgZBOaQnowBUpzb?#wUDnj}4W&BatHV)o~lGe(_FXm@Eedm&4}TCSYK zj_TE6y#n%~M&%N^_UZxkVF|aVpI}D@Ssr83vj>n@SIQ`xaoK3`o-KHdgAaaMWrc~i|Fq$ro!z0{}A{8_g|*7<$#OEDVgvR>kT`3qWPUqo8I z=fn7wpFw^7Qt)vc3eS8C$cjq_ygXDC$U=O$Bg(OM3VXA4^55pkD;p& zLzH(Ivn#p(hIi2jc9VPy%qmI+!IBSr`y1f{;Qh>U=uk3cCP?^kdzTy{IX8%nAMC)c zePzmClTcvQ!^g8b<_%$Awf-3N4J*9*ya5p{0Ig8 z=s;9OF8e+rm+f4*g*BMD8>te@lepkh68gL^7#qx4 z1@-xJS*Nq3+2^|tf@s4%nDg<^|I=L8k*GjJ8;k?%NLKoE*lN@WGNWukxKWbTgoBV7 zXG?ayfKpCl|BMx{M_HB|_k=49J>Yj6N@|$I1s{u^9JZ_Y?T}Qrp>p z6y86@lL{lTe*b0k=%Ryko~RE$jh~5A3yNUXGfQ|_y#Swk_Y`-ZjDd~wXArNfDUAD= z0OFxwLyEID|IZu}&LrhcT%5>{UdZRud@LhM!0w!{*d4B4h_TfU)*v&R#NEw?iJU4^ z##Lq9TI_~P(jF0-$|gVhougJ)^y6WgJT#O_+FZ`_s6k*)Za=<4R%>RH`9m$p3REH* z{@t8;!XCg-{T}zpDw13DGiDWMD{x;6J_U6(oIqHhPsOgRs-R(kgE^#P%2$-E$i<;KF^a8r zP+)yWq~m-Q3#xI48oYcJi{D6;asN|R;zqM##xOub%rD{h<*}6X`;!*y_qGTQ9rfnN zZP@kae0AQt{#f096#1E;MeZBz5;?S77JZPoMO1MX3AITe)2(+whzl2I*h5V$Z^B8r z8lp8tZz-K?@2Qs&7tpgITKLzrG~{n|0Y6^ej~@Nmoc7z#*)Dw0hI#KWyz_h~5>i&o z9Z?tOKkP1aMCYRu17<_dM9v|pw;5^tq{WVTcNlJFOh-x^q_MxQA<}vigoFjt81?&$ zn7Nv((MitXU)3E~ye~G4r`0^$&&=u1vB*^O42&JJ9M&&Tf=YKS)>b=%hf!BQ3QmT* zk|!Tl!^r6&u(|z%Kv&5QSx(MIyDNLC(2^P?h;U_Zb3V|^I^M&jn1y8SwUwySKa;ba z;e0F2Y=kqN2mgIf_03rUsi#Z%Z^%XYQ;Bo?6H3aHiIa3NJdhg;i z+7ZMh(Woi6KVE+E3HKjJk!{;OoMmln!04hTth?iZ6L;Rj2c}5Dv?sDWZ0SD7si5EQ z__?!s+!D%1Ux-e)dQ-g@_7TmYlfZX!E__=P3hgD1F!ARkRK4~Ix;*q4qkl*rU!A3i z56oT2)AO$TO(b0|BZm{}pr+{x(3pV7X-W1;YXZ6B(8{kNV;fj# zSHA=~b0)y3*eqh1{DWT)_6N#<#?yRAu*?Dp)p;npyqgI(%ONI5Eg{WKi-*y3Y#XS* zK7eX()ZnT?hvDh{6J$u%cT)N>1{V3tv4<*OQO8#<;`~_s*V3*uU!tQwUV_#k) zggR3%XnCdu6HYG@_bHUkX%fxpw~LwbmYaZiTTt73BQm%6H&R^ml6*VcOE#M1Q>K?+ z@cy*k-F(30%@d+^kGcJT-C=6k3>$RuK?7N}^cE?u>f^_l9F|M`^(K&q^%czB2Zu_t z<;s}3^NY!|w09)&o)6YA^5ALKQ4)(hS#xx&IR?Ei%tQ0D&Y{76CwZQV>6zfz&EJ1X zmF3|S=l#9*5ny%<_{tn;SSm_w6=P3QMH~NxwRiG7;Dr%fl z$RXn~8Yc)}q`4=RezMj!aJVMTE4%keS^k2^)&ksRX}$q^jvMkK?$0>!Y zGo;V`pOdEsQjTB#=$SWfn+9c@6nQv@EtyE|p3=#j^>if4dR=7EgCQ`jOn@cq58&?J zB6K_J5Yg~bhG8Ec{=3H&<8%C43ia~C7BOzZ5RYcmZhW7aW)uy(4lSYfG-i<#s~h3a zv?2InUL4Oq`(f_5ciR+Z(eh4F*s3t{Jw-|U{UVIZoPqayhGWGYhY)S8Lrwd52CbNK zm+BqajZR#Rz^vajEOAK{zrGlZJmtni-6dI_oYy1VvXKDP87?puxyGJR7Dkl ztCkL^aO{K1WN^**UJ|$`5iL!a4c8lu zShb5RbW36q-emy8O*DyWkS7jtGGuN?pB40vGh@?_=&>#KQ`s2XS3Lb(>lOaU)g=i= z?ra`zzM+d9dSqd>aRy$gGZI~zC5@&zOX6}XegJEc3;W|>0{fYDrwFm2vD%djV2;A72$cBeZV7G_RY+{cST9%N7+U9Cv zF`b${hLk=({eYP|to?Ko!dVHMms!J7?N+QfRbShLStzCj!14x(8@ zS@m2BYb|ldXH;^r-_}xODA6BVNFAh_YWL$&X~*#BH{)Q{hI#DN&jsYtJrmG8#K*^$ zfI(Ks+-)Z_Mq3TNOerMN%QtX)R-w&jxQ+ zMnSTQi1O9v=_8iGs*a47QA9EFO^`L?XKc+Z(gEEPA=Bpjlj7qZ{%1XVXwo_;&0QVuTJ`m5F42(PU>(8X6el z!M8tgRvtO8Nkb-&T`}=vO*CchoM&Z#W%2*k_8{a}DvEYc7BBm$0g+K{vNHGi&<3P)$FE;R9ES@X9yk z|CW!uxd}UW`7_PKdeCjNSJ-vyE2{W{6l}8!0&5$Ji;qJ^BDFY~+<^e(Ot)exfFLAmM`JB&1W1JjNgv+H(=mSs3(d{oY zsl*IN+;w<9F5A0<#9XXJ=i7sk(ws!_neb;XY!zam`;iEe|JcpiwLPF#4!9xswlhUA=w~)AoSumf z%;bDVfeIaGV@tqkDGndH9qZH&5YtfTawM3j<_Ch>VFp{@zlg6_D$-FV!NlyNHy#;V zgJXU65T_thI6W?jgr}E+|Aqu!M)oyuwk?{mjA^tax-9&JT_j>@i#LgMn4jOjYkFG8 zSv1Xsi_d;e6$Lck#cvc}{mVCW-Hn=BuF6GOjmK+mAHuO3Rk&-1HS_2MuRkjk^G?=L z4{uLb#%IqI;VU-}(C67~I|Jt3@SqABmn`_@g6o$^91U?wv@dj!2`Q9zRJ-3J1X0BUZ5Z_9GM} zh^00!)+ABA`pmddD{xVjAwJK=^So`P$n$W?SeBXm$pt$rn9$F?AM^BAAfH6%Wyg}i z5!2|9smXL`f;yd@=Z9CgIHop~Hn^KkUn#puhjMXV z#>_i`3##_gl4~TXenT`wdS8H;izM*vG6x*;L6^EUYb0&cFO%LsYYWY6pN+R{9Lw1R z@j8dXn4&}2yP->#ee z=^e)pqRmkrbfWGdT3@pPcWVQkvT!mz=|UxG;e6*@x|~7Y>s}!SZpQTG)xS_?bO%{< zbU3tV)`3o#Dtl;}2})VNiMT8|X)(h~g`Qt=jc!`qPW$Y-$kX%hLYc;q#!i~1sAhGV}!M@MgD($5o= zEhSCHS*GrZ!(%lBlm-4{J1k}N7?s>i!2?Hxp>EwbjDcSU9)K<&^`w`m`P(Jj@4;ny z{R(ATG0F%ZFe#$GM&2S*PP`;cfH!luA{@PZ@e@@Vr{cv^rs3pylH>-rPrarwkA!bK zPI*;)r!Ss+KpVX~PS2@~E4EGhA$V6LARq6qK#Lo+vFaxaw6~^$SQ%y@uXQ;{#_9<6 zxR;Mp(l22(vma=1#U==RtPVl()|{UJ8ZSMZ0s>kS3}XU8cVPiD{DC2{vdBa4tB*1< zTX)0CuSRH3V-7hNc^)Zrt;G}H&f@36x*f++^+z`p^1Ohjr%-znWq&n_pI=3T0JYyp zWrWKIfXRk=@Y3rns$%NFMUDPs+ju_86|4v2w zs>+bWh*dZsI~4z%S%tmF7h5ElZKcU>o;sav&1@$(5@|S* z5l%{v`JN2#xs8gpYOVP z02jkM3zz2%p;NZY(6a+CqTXf-d@b&hTDxqtC1M;_e*1-|zj%D{CAV$_`dZA{gEcGS zYo|6+4K9p8b|!a@aPu`f=~<7Qs>?~={m0Y~`(-F@;V7P0H(OFDqsA)c#h73`y&(*H zdSSfm-AUBrv=x6AnBotU9B`_|Bpk}b;)e@DP~`qo_(o#~ew!zc=}t}5cUXSIm#amD?GQL17( z2p5}QqpCQ&QE#`K==vZ5HuNyZEu8PA$GLy>r~ZyTj2tKL-rh%sU00CVwH{O|xI$)F zj76al4kV~(FgI~S*BKyNaMrW?4MbYHjB-7y(EQq^GvRvh~`y)FYh)pHg9i^;mHop}olqsFBqM8LOSlUpO8)^(Q`)hT-`B zzF4K@9O|dOf~tI1L{9I&jU6kO;<`KwXqoes$Mf@66`p?kU$}AKvSy-|Dgl_T90}C^ z2*z&BV)9#I8d*N6gz1g(14E0~#Mhyca_=ui7CV`f#E93#?M)UGkC9^QKF?w&aJs%f z3L}`kR2a0*SO`NUvB+j-9L!G)LS_B~iDkSR3>!NFbq$S2^4&H9uhG$r#5YG0AGPEU zf3-hQhOd8;E%z5PTe}Vu5$U9o<)zVdOC_?f{WS8h>q5G>xD`?k7$+Z$m{y4lLg&fs#f&7wxGU z$kY4jUUO#mnDM}{rDUQB{Vxyx;e}AQ<^Xth6_XmOhtT)JL`j$9iQ}DcsGRKx&!sY9 z+TfLN<~4!MIz!nPS<6JsTwl^0J|5I`INvx;W{~+<0LPUO-|wf5ay4`d-=eTW6+8B9X=7?JKZJr)v(Q{rr**d6tg$Q7?19=|o;{@8De z;X6LGjXYSk0ZNn4!xOE#Qu&~)(whDy%nj*3YmK;$m&Iu05^X3RU-)o*55`4lSb3=& zPPLu@^=BtQ8>dU28@wK0im=AB9xK5a#~m>E>kW`aEpYhi86LKOXd!XRSq;Wpj^q0} zS(MEji$2PP+&N5b8!nAaM9Z<^Kzrz%f0;~EHG~r%?J$(wM^C)f;9k{NQu%QVYppWf$$i0$j_)u0HZhv$Mj;0>s^kQOBhw%(NLoEio ztewx=6nn5POUJP@PDPQ{E0d6L>@d7(N+!}w8IAYv+CjEy2vGI7PiW)DM!x@uL(|AO zY>dLnABuAJc`?6syh62V2yZi2MX@B0n3@h`FInFNN|2PE`WNSg9 zaTvQ`ycs*4^N}%OqYN~Z6N1Z9alFEK2(q`<09O(u>s>NGvsP88ha}T4YSE;prxKb`xIi^&8o(8X&kt#qszJ&^tqwQ$_rkrB}v6 zM646}S=B&guT{d4W|#1E?Kk*<`+9h*a0sUwZNrHLAJB5nZg#>Kbv$<*r>7?;!Jna; zE*gPOO#?1=vIbmSQHS>Bcw>i}MEo^%m}uH|LuzQpKtcZ@p^Vy|2GaZJ8Z)S12$fto z2;-0nWMDK3eJ%XVw6%PP8IOY~U5Fs>1+w_ny#Qvy%=Pd!ITrS_E->O)nkatlIizOP zi!Hdbf$kxe;nR7_EhZ8t+D#BhSQvpbw)e6 zp7oQQUm3;Y9ICdSct(Zc*3qWe&$JLNjK0albZa4%Cp~1>t_vdFi#=5TK`T-6b3fF7 zZ|GQV9O2&!^(e{xb1#NLoG(uQ-v50Tr~mts_)i{x?{qBs%Y!A zZ1_)}@c*d8=f%C>6e%ua9iPP6`>3A&Pd&b$f7*-h`TGB)|Gk%3uzu`0r%|1Q{((j4 zrFV1Wf9i?r3MX9a8px#*|6KoH`+w?6WNJ!<{EC;a|F-=n|G#DY{a^UyeL8dhSDn9g|F86Kp2UB~6X*TS|M!0I3NBASO)P5? z{QW<5{*Ehd^H09SKkfOpJ|f)=ahv}|Ly7;$j<~+WKeEHcwfi4BD}`u`7MK0g-``aD zPagmL)>fDN+?I{|R;lx6nxh9yqy=Y8=}j@Z^skU%bk5g0oEkSCznMIoi_hhNa~5pF zvlwkMo!QUq8*&>=uRnohF9%aMhhC?Y-ydOaZ>eS;P56vr-LEpM&L{Hkil-s`8GuCp zskul}wT7GtdkafnDPqkR*=SwkH>9$27A$UGz<>Kh`kpRzoc2U?KO&l$t-BhF?xtc< z3W=S2QQ_9`$QPE}~CT^u2f) zS~I!=Z_LWX4xAUQ)bJK&P--IcDSQ&Ci~Yj*ow?3zaaKSr1BRj3*Jm<<`s3)`(RFBy z&nub@sH7)aWYVcJ)9?zd0%9&F1&^YB;Wd-4;W;}WG9%K{@QC39kxRr^w9icwZ*821 z&l--vz8inxdL)aUR}W>uPC!RP7nWDm$3u>4iQ$#lYrKzit;<6*I**ZAOK6z1K@sQ8 zDM6u1ab)boOL(Dj8#14!fK7hw#s+qO>#}{oq z0=3gu!PJU!@Yzv_s?06P=EHyXAjM8d;-mbr;+#dV#ej93eB>{+vy}_}zn6U5w*#-l?<}g?nODx<7;3TYDLq z#Li)ii~6$+*DsQgWDS&Z;UO_av))xw*Xv8L;C?!uf4>d)zn6j!y6S)%=Yp%~On-dG zWg*yZS78Uw=ig7lK+ac4#QZUMVOcuv>rBJ8SsStDP6@VZ=L7!Rlh$93Cu~`W_KXN( zmh4qw?+(DM?R^b4G<*_ze4&n)pI@J3K`#F?SaJW4G9 zxHy@`3w410^_`hcr4gB+V))#Z2RUb4;ZBz_K5^$dGw_@iTzOsvmV4CT9P1C~gI^HU zTnQ50)ek%;=7XmCBDP;z2&>J-4@(H;^qj(0{0qa>JbxlJV<+2@X3oE_4?Y{r)BD`_ zP(rViU`MCOv)5J@z_e?h;M(SB@@;PmGsopOF;Tcmn#pcKfX7^<>ZF4nTjx{0tDl2H zpB#K#ppKp$&@78h9$EIRX?U4lPz#y#_y{R?>|<_EIK)ob7(30qeGQ(6|r7=)7f~=AC;PcH2e7EDt4f}8V{eCZcA+!LZ4GAIs9ogRo8fq z9Ck1yt<^Tjv;PE;cy^kXQ*k}P#$nXc{jbsKAL(S${;gz|i-@%E91J}>t5ERDolrY5 z1tOKdl4jjLxcWqn?>}US8w@qx1u14rA!gD(Vl?pOzx^4XKfriy@P{wkiy(UCDU$p9 z0L)msgfB<;;=tgt4t$s$3vxp*!nTd6V6N-OTtBdnd<#6u=YO+^$%Cf#*B4(o802njD3hJfOk%gu-665msqK>iXy{9{y z?(#o2`p55GgikZgS<$irVli7C_Q(une;GFe9%RJkJKBKHk|@Z#V#l&Gq}UCsbfDwx z1~N~*Qxw;n2Orc1u{mEJ^78O*_*(vqEyjc26T$UB$d>JBxw zge+iv&1Q>fC6xcDL&0@*v^drAUwLYjI)F27-^OlP2XNXWU!0TbOkAJjP)l|cWTFRhTqTWI#qr2(W)#u3F(=70-tsai#(8_R9X2f;NK(J{ z!+NDL#35`w(LMb~-^TUcAy}a{8F?AMC#~6=VfLa$q-V+h%&(JojZs{r5q*5dW4vYU zA3w>#s~__GI{hXGw)FA)bYTwuj9X`B%g=iF z$Eh7WA2PY`{Qc$rNI_(ZnzcMg?al$PvSTtl)zu;^=f5Ya0v*QZQ9pe7@>9A_#LyX4 zEX~@SV{Bz2(Ivek^!<1!ZFBG{O8@AL-QN7Baw>Yn{R(47T*PAwWw03M-sF9FsZ0<~ z(eS`yf9B#}&WZSV-C?Bgc{cs5{2aCT(QMMW@fhl=&ER30zq=b7G~P#8%?7)icL9+r zf&{C@$Z>}rma0?5eXI9C@8?=FI*Eb ziQsvuV!4D7=k08iz&46cMaqQ_id0Ui;^yRk(5E{FIB=BcHQV} zVALKGbmAl_x?y?r{{0m)hj_WC${$dL*dUTeOaSNlR85w%`_F+8Svg^H&uR<&Y4;9c7 z$6oV!8|u=D0X`;}BeRJ}AHKo#aPN!Lzs#WO$#X29l7edI6i{~4?eIW)DD#vV$my6| z6vHVLUp_7Ad>>+%BOWpMiK87J{ZotSij{>oW$ECwt_AE&2IFH(eR%rBRUaUSUwjlS zZ<8mvnSSJZ?NOq=I2ayM{2o{$bLlr`R+k;{$mD-|M zmTD9RM=l%(F?~M$4#SbU`e@}83d$_{$n0@FOxd1dvP<&M_ucm)TdA!l_7m(mkff`O z7u|{33$Fd=!pZgVe0y;pzp6D+>hse?|JD;;MiY9D5v_7>v%%RQq3KZC^DJ_8x>(>qTJwp3|p! zr_S0vZ6(Vm3egQZikuvJnB22^%&fcih2Xg|)B(Lw*tYsTl^1cE$X4!wUB)B9q*Rja z=6oKkoAn1zJn%S`v*b4OpygSq#rb1oz0D3%5PYW)g z>u~0z~cDXa55zU7H4`drPd}+QR!oC&;(<`B*+g8_yy8N$yo0sH4C0FnpTc3|E`( z0`@I|#i8G*ZVP{uKbt}H≶H12?iSKOHi^XwaGIXYi8E8^K$D9KJo-8HP+g1jA*9vb+P~%J$_xJ;_=~kXrE>WJ zQmUy&$6sF)&ku>BedTyayDK&i8HY!oT8}jp9pD6Kk13uz-OehoYYN4d)P8}5Ma|@) z%`-~tVFq`m=m1CPIJl}N0^f7p%(?wv$>FkYK}CEfWm3hU<%P9q`tLmQZPTB#VcVXU zM0f6S_#BZ%NN6W+RF|YPyn1k`{6jR*Y%ETpVo=0;A(s861!~d)DD3dX`mG-L_{#P8 zn7J|b2pIyMzQ>6CfSL4;1*&vc#6G-Y(l;#gX$`g2uKjeGrDm|A5ww3Y$B<;; zNFM$Loz_Te`Y>2^=NdINF$Ql-U5FR@^UijnLVEk;qC3fAXjV6ALL<_FS;Rkco zX%FcK*mnPFd}EzE*<{K2jf?gpqt{!)iWS!x*Vmk%SHJ$?v{)WJD$alvIyu-|z8b%? zFTyF;_racxzA&IG3=)4W!J5PYv-eWbn>IN-yZ0)t^&LSsRpg7@1}cf?qEPK*7FkKM znA#aC@TGMvt2?TV^D%V^4$U{_VxeRqZHstn`*UCN{MZ#T$IB4?ekMbj`aMN?ej8!I zaF%o|JVzAHCy4PB{=MQn?WwtU@1V2e9Z=jOVCJg@;+f|JNWC@;?CN)d>8i7k92X2; z7FUR{Q4LIgH9$tMJoIi0AyG}dj~R)F`o<72>^wNP?P6IsJ+^*!Hff%|g!~-Y&itqt zi?581Ly?;MpgT{t;wwg##q0 z=rG~ugNUHnP_%WRE=(8Ag@*xC;QHZEP)-#GBsc?EmSf*>>V zB18u^5w%SnaAE#&bc~DnQEo2@FSk@NJJL#dIJy^mi=qssu*OMxF#2pbj9Z+Gy6uJ& z`l}9K{w`}Y3H)*e-F`nG>J0gBDxt;QZ{+!cvoO;k8v6BilL4nMl8K*35`~sxqIpaJ z*F9uFum5{gxMi9^<%SLkcbd(3Tb5Mou}wW;zQ*?hq1cQ*4>Xi^-!B#W3dhPBxiu(oGf(9tQpA~ zU}S#xFmn2=jiUC|GjqR6kn;Pr=;y6ZaQ~q(J2YbELEaE7wo4m&n?A653A9)k0{ZBQ;(3@ zsg6vw-2%bZIBi;cXcAtwhVujU%N=`^a(bIy~eWcKXLbgy@Eq$i_y;lH9TWl zDE-Kci)qy}gckQR{6Y)*xTl7EPzV%{$B#AA$C?h`q)QL~wVb}Sd+x3Xsi21N2b2ihI*u>PxPb9P4RA~F9zC0X- zbkc=)lUJd{6f}E0{iq3d{m`Zt*KNfY8wcUy*{V#~O9^`5&t!aW6TwRt{W*{GT5e5l$ZOGx z6r!cxY0;lr5^#FoBYe;Q7(SJ-9}ktc$KPJg!#`jTy(&ANb_n=}-%cDr;va>P@rRoJKh|A%YuOxph9E0!~f(`&X+M>)EkkdHM@TPU|J=<#G%&9(cGT&twt2Ne7BMPVvvP zlKgl&NS;<^qpO3d{yILW&MO?$K8%5_8D{@IA( z6JMj+w5V~%Trjoz0N#B?0yS1fqSE?(OmFlbJAQE;N1dT!8u0ahFVlxKm6_~Br1X&!TtsA zDC%4lN=O(3?yp_JFSUW3-_{35S90fOQ{JITarXGP^ELA7LpiEbO@-9<dYDt)bU@#uy*p38# z#!Nqtqo8S;1`(IM;MtBfWcv##UIzc-w6OaqDd=5*_gA+fw~_#~Wc+LhGuQylw26nS!{ojijc*69jv`M6PbRr|^F_RVU{tYVYjTqw&- zeJw+sYMqTgmWz-DV*)x(-!J$q9Y{6~kY%W_PiW7s1}0N;6ev6CV8g9@EeiI|M}5L# zu$V9yZhUqJRJRs(4UB@-M{?oa;BDY#qXR>}O0uQDFq^HR%r3IkBnh>dXv|$pJW19E z7HB5Jq`hZAt__2xqyQ=nqes%LV3`>!v6 zrz7Q9w=oLrC0|FDxgrFUiRDyF(I8|_rm@D>TiLu7W^6#K4puGS$WZ-HA$FKQ*)7fK z_HE7JRSR)0O!*}sActo=ms!I#tW1SmB zMmz%lZKK#xg%UfGjM+h@6WiiJ?WP<<OnZ9HKXOFu9mEmhpVqjihHRgUuD`ICQ5Lej*%A(G41uMS2%$ zSD%4)SnA^Af{Ea1G7ODADir9L_`w=l1dCSJgW<$nh>=o5Z+HA)nCngGhcbpKY8n=) zVY~5w{oml-w}T+JTtKQ%N%QpmHb)j58V*sKYDMUx{Wa7RnTSWMd4`*3PRFOuwUFj2 z&TmJI6Aby@iaaETPz#5v3qFl}PXhPx{zipXd;7zD_YC;41oQogaa?UO7Kzimz#O7@ z*#vHXbcYMRZ^bka&ie2k9TxaN*XKgivF8vvPovy*2Qw>cx7(ODMC>l=CdjilamQWE(R{roL}uw#C({HrU} zAMPW0Q{&L$G5t_^$QK@-_5p^lMAHLixt%2IZ#gr?yN$`0ona6&EEdS8Y##oMC$VJv zIYJgMBh2Jp-baIk{hT^N`r1jF1u}FmF4JPc;4;e6Pq5(91$peX*i{Y$HF}xpOij3vf!o9^b47NK%lH%VJ zC66H@jhjWJGfr%B}+bWdY9sI@fDyn0$24ah}(&uh2nIpKEJ?F>ZBjTK-?ar+?s13p>T!@$^hcN3!fGH>o25#QX83Ti68?Pq-`Kob zkosmY{$~t+T&gaS5_XM6(>=EHFhl8N9)?e5tH8-fjT9#H`xqV`d)ec}R?mhE4xd9f z=MOk=wGFh*oB-Il0e*Z@LKW(Dl&4ug&L5#W+E#mpnK}QDe(_(N0-bE}X~}H-wLJl! zH6j1v?%b(P9iKdy^Gy)T%T<}j8a#Mr4wjI+3MZz>v$8+__;dMsE{r(TyOPf%ZPBup zbS9=w>EAxYV}xw?hszmZNPK@UUJ5_mV2irVro(=?t-O6gKO`RF#6SSd-<_jHYMes~@= z^i2jy^6jK1DPLsHrba{L^lRkZyl%3)M2;OEI}0}L|43?wjDzZwlOR)A4CTi=Nmk!3 zvbFyuMq$`hk}>iE8Tv?$DNxpCH=2)OKUb@=#c6G@`<@Ufj@^WxN9oe)q3-nHBxSnx z@GEqSNkq+K>rkTFDLkoLiZ=RTNn1|f{IaedK<9=3z_L$Hz<|(WuqOMD-J7-kT9Lt% zIKg$fQoM7+35;sZaAjo&Q{CT(NzsmElnm0~okJ)@sy&CmT4&~t{W_kOTRO5)-}_mm z27ysz;j*)6%bU$iy{WQjl6fRbQ?)~%GGjUUNXNs~Wg&lu9{cqqZ5G*d3>qiJK&#$X z*yQp6Tn``Q`MB6d0cBRzQ}2!$u(_66tm_XMII*IV3gphS#XOprU4R?s`CzkP1Ju6a zGP60v4hBj;;mf#kvCIjT-@k*MYhqxN>~coMXaqFcZin2GMG!qXA36hsP{u5Rg-h>% zeRvCUYZozUR-Vk;sEwq=;w{NluEd7>ys`F^deqR>jkpNag879P$&>OQ!s$^$2{eIkoD^y`2XPT??U_#09cwG;Y(OeTT9FX0cHjB&<# zC2aM`h%}5o1qF|8^ZfLlFGY?Xi?&2;uDt&yVzZ1L|#4Jur-E$wC^GQ zS@RBujDC!VZ@i73jvUDPPi%)nj){1R#uFU7K|l}F_=TUyHXwDK0iu88Lt^v5i}-sU_Uj3czz)mK;0u<8U+ zIk}YWTdj^e=^^xXeNYe&0Wb3iW#f3lAN{Hs>zD zhv4&|nz$Y4>ftEeyA3UmEd&O2ugO;QgqlG-Y~s9) z@t(Xqt0#S@RP1f|`_5G-5o#aL`^x`4A%MuwKSxZrPvG?YN+5k;4EuOtIzlhqv8$0L z7|SXlt0GF(qu|SN~ts;O;Sn|C89Ef zN})MvkSX`lU`WxN2925rjVjUn-g~|7^YnY(e_o%@`Rrq#z1OtYUVHDg);YIUl{+?K zGVemH4)4&TFT#1r$04v~Li~vvma?C3is7hV9K{(^kj#~e`N*-psqSi{FTuH0T*LEm zzQfaATF5Ot@`jW0=RC({$ZO6>y=Y#9+&6B%)gAVamrI583E(sB7^W+SpQ zlJS}B^V1%<*k+GruO2myo!pkf|1GBm_K}ZAxNS8$T*+GtcoX=!8uJo$sxlS^@s9k~ z=Q@)G&tj1^d&q>IcNPV(8_f9r;w+iU#A-E{wk3W-Sqkd2C(Niaee#|)ecTrj&bueW z*w^=u=FKcP%XK~$$+f&|D3Gm~`~4YJ=Ua5R%J=8+wjV0yxj*o!KD2X>3&&t5H*rK` zmDBcTE+HE?3iqXnYos{lWP z<60&U;$4^}=Hl#jlARu%%2~K>11GmOl|6CMW{v{8j(re6+2^l?vjf`YMLNcZyT&Q# zC`ZkxoZa?7t4b;HrR%Fy2~NI58)swTInMs+9*RH#RauIqVh z%MhMt1&jZ!TS>R@S@LeL!-`yw=U58y2yn?Q59X|^eapKowZ7W(M-ErZZ2{MQZUXPj zw=3Ljg(ku=tU6p!t#l>3I^Z{t)S4*xUp9Qi8#`yqu-T>pGibgQ(M<2?5+()<0ahg4@z=PNM>=G!aF!|p7%I!|6J$XRH}SoB_YSa3_wu z!A(6F!?BI3Wrw{52JR@kUd5trEuUYoKb)y!RWvtPS9wI6q#mo}(+ z^&xrQ%Gh@9t^=<*2c{dl&YZT;^;l;P+xW)`?ttf4xqrKrT>UR+x-M9e%l~%!kt#EX zGH%q_5ZnUA&rKtm^wA zO8j_}UiDFh{?)%m4yzW7eebm|y5}3v6SlEklxJ7*tYbQOff?I*v-WjYneOv)IWXX{ zi-W^Uw$$cSu5XO4P`B#+&X~h8mlx75=xaIV8F!KR!;_EJ zMzCw7GODb$f!FY05bu8ALEem7mayGsZ~nWX4Qaf+1Js27k7P%(HKTjJHL-h*lmMSt z_K|Gv@|GYrGiL*P?Q9;eXlx-bFmxx+^06{k*;#^@pli*09$LhfDB8i^wDb@AVJQE% z42#@(ciT_6WDVtVv&0ATGR~y(?u;2$-SWVqdZ&Ls;dpJ%T+dR?8RW7yy)V1%!Y`K8 z!9(nMRZmV%%jy#H`E2(QUWLAW^)s#9>R_wM)k$Wbc{hS{d1v#ka}V8}$~)<~o7)lK zz}x#ngFA4B49hR2hI|M*jp8Ikii#Me18<%ID z%3C}lr)uG7C63L-Q``|A-t2dQ<~*q4=K+Vg= zZ?vr3=MR~RVkY8z=a#sb-)KpGGdVea)Bp2lAjT34TQpB)pJ}`+N!WW$7Cxw&okNY49-0J~Rt;=OeHz&kGAZbYcDbqQnKe z5}Zh=!P^gqFlIwKo_QX^550OU*DuCl-dm;@esIiq+WTWHu08RVk-L*FxA~3?-ez8e zSN3TvSJRFUhwb44T=CV&o#j0PI zw72iq)6Qq6sJ)=gm>m-Ipv@Qim$~BHDQk2x%~T*;g$2V~;8<`S#VgN4)@~HykFG_+ z>I{4tzX^Ye`eH1{ACmgXSXP>g*Yo^ovu^}7%*iC<*xPiu+KAC>8AxNjQb^kNI!T_s zg21J>@$_mV_IA|6<83q+#HAxZ^%yoBI0|H3r&{r231N4@OnoK6e{nj{gq9ktw-YR}n-C-3uvOqKrO%@1kl6PxysQ1G{1x_hr3csW>C6+@-fhLO zv^KmLx|Xsgpc#Ur$UAG6=Kim`xqnI zoD@i$uPcQxW`%KZWl=Q4^#WHK6S6 zSxjG-j}aMMSZg+5?)YXnn`Baz;wBn!VGWtN&!e>aH!XVk*ZI>c!>G~FhH?*z(Zo*vIV@j^d{*euT3vGrITuMQ1}(+Y@-*~u-heMF7h}YW2h=)2 z22q(mD8@|%%N zG6;)T4}ll!D;=Bdj;4=p@Gu&U*vv|m`yrh|UbfmOqQdD7C3UMnZzT((=S&si6W~&l z)6uWpv6Rm2IE3ZzwnFiIB7$a&#rxbosQ>DYg(3cE?Bj>MZ*s)NKUJAEuBoTPs}9k! zA#r%0vk~tPWn;zO4dAjSBhYdQN`G&L+qFa?e0kHYRC;n79bVZ;Y$I39Qg+9i4tp%g zlosI+ad9R~juDBMUZl})N~zDI^Fo@-THC>dxyHx|a*b zWqVW^V<{;ncG&UNRv1KjXq@sZk`&pJv0h%V7%&qFb7vyJ%1iis%1J$}ZB@WQ8lWtI z``=efzZ>M-?=D&ND^>TpyqPAd8De~;wTKSG${=JeoQL+25il|tg$bjhkb5`{!SS2X z%Gx1nCxRK1DU{1t>8%tPCMh=k$RVhVJ&DuKJj~m544LBFQEyy`*!S6ZHf$O8c@KkT zvIi>Sz3^hXHDtmVXv)sS{(H;N?KS-$`0I2uN&VAc$g2%S^%D!2v}r4W0wotocGN$n}wMFX)ov_`?E{`tMYc)akF+R z{jB*zvMWbo&%Ge{h56&|Bo(Z@YK7rxZn%GAj2 z@S}qio?F0YeP3LUSH|IZO>C6*K>amWe7!sjUhkwJWqLt~PXJ$hS~lY!XZ8!yq|gP9Nj@PN;sgo{_xxftf!`la?&A z`G@vi9Ocv1?w}X6!tS8>GwDWwP|kZ zI}Q#(ntA{#eyPBA+7K9(df@j*A0bVe{lrN%Q}zE7SFw}_Ge08`Wudsz5G~PTU*aO=Y1}Xg1Hto)QF@qmUSEiJdCgIBIiLci zllq7a(7@Z#rPP0He{4JJh=Vp7uy?#ew@nNnQ8W}%hXN5c*B=jE6XAQI5H0s|(bBOI zzjxQtAm>7|6sx39jVWYjHWp{wVp0DgScv1;fI`$6Nh@Usm0(Q$9{l!9!qwqk7b12N_}zgoh>~3L zA#amfomCDk>KU_DDgm$`I{^nT%)+yImiQ7l3gcVD(G;`@>vztEX7C`Gx-CV-`wdV_ z*oe<}chUFXXQ}^yS2RntoAMlc;QW4(LpNC`$Yb~&=b?d$6mvC<>dQ;%$9R9lukyjT za$TgHct8cC`(Q@6FJ4U_2l+Lj@N;javO(Q+AhN*X)sZAM?;vMr&I^J@SOD4{4Z{wr zJ>*cgOEe~u!v_(!b0^hkoTV1s7P9Ma0s7e=+!soiy6S_l-ID=xiI!ATpu$a(^6n;6 z>l!jScZw$N^AW-6m637KG`cX`fu1`Ui=OTID1!@qWnna`opfj=?ha0d?Z)L8g<#xY zkwE?j#L)qJ6tDBb!zyk37(N#H3iEJq*G4E7?7_YF$FY4)2_7HbjY{{8u&dh$cI!%b z2POZHzHW^i3El6}P(K@kYwKK4`zn`wCUEKKm2~BM=AZLy}HRiB6t_T&`ji@ThhgNtVDwVQu zDe?`S_kKVU3KjGqX${HOtcS02Hl%-T!^`d4V7)F2^+h@O;JJ;ze@REgto3+Tmxw=8 zWASv-I8@%T!`86_@!C-v%iG07{FqdIcIcJq+rcB~XGkVZIDLisOnpXaQDS)i`wJ~$ z&XV}t?Nq*|ilVdrJ(u;*RTIj;(;~wmCTKOq|LHjkTrpM0weKsHoQsBGL5nMNoKMiQ ztMydTuYsB)?h*U(465%hA;$QPA&F*p(HIJ7`ToFyPVOE`f4(oJX#P4q@#h z)c*SyW1IJgnJjH#dds3`-5fGf@t^(|4MtJuRdw9hYK^-p{n4Uhg{n7wVdm^q#YffTlARZO$s1dcyf1 zI%+sGx^yINZH>gv`WUQV9uC7#TG%VCgAyfWD9>yX(Id8MFdbU5gml~Yk=%ucq;~lh z#o3k!pT{fcGDqfEG3kaQnBxu=LfHMD52Jy};dFk)O{Py^yA(WyV|Y z6xpBo{WbAqtTu*>R7L*&pOj{-h5@fE5mVrVp-s9tc=w=u#`g%T-m#UI9Zki@x09A; zCEeZhUPNE-^WHM+Tm!?~=}3pSjKsUFF|gk-3<>R~xV685wp{s0Tt*rfH4NaasQQnO z#%sJB_CFtpo4*I**few0cAFsaA%m}TCSjWYaBM1^Bcwe>H=T*U_?o$8dY#tKeMXCR z-lMnC&qef@xwKI|r<@!mSCUm_EY*)|Q4`SSIm`)nlKt?x(2MVzu@dR8QIsx8Eqo`m zr3ZbrXqS-;8GLAAg0tV*8EA2oud`z*S#lyhZC0Xl9(gojeltINKvmQ~YQ+>XDBMdj z?h4p+Rt?t&Zx?glqX_96)37%+3Klx^F=Es*T)V!Rp4>l9@tZbNSz!p3R1U!IC=YlX z8i@zBju4+~ML|{*D1F3O8o}O8FC(__&)d0_6`VoS&dniP@v}y~G8U9BFC8&pu&mgP z(F~$r>cV+$D;>D@n9dZvrqiq&B6$ef8u!npJD=AI{Q$i%HcgKv55u1Q({PEsWcS4h z`(_Zb0~~S1z!0t{RS|Yv2iH!EqoC~>-zMqO`wE7_q-PSNHbd_3`yTuMIdw?C^0}Hq zd34n*7U@@KM++@}l_#Y6Xx3hOu|}6>Y1#7kAftq|6=WO|mf0?M$jb65#nfD*Id0vQ za!n3R^9`{@p+5#%aj;wqP%2_@{g<5G+0_r|z4m9Cx>*^Ynt##KhaDn5UAUV_te+k( zJ_9x(2;rATVeYo&$Pr(HD`ysBm-#k2^E{V+e2S$P(PrdfB1iQjN6_zuHgv8aoxUh; z#gXazaqMZHkQb>ZUbuaFIyPL;Q8+zq4o-KM;%fa_^q0JaZ8r7z?0S*LTS~)Ixr3s= z>BHOJ7n%jL@y&1=Y}fdT^rKhag7QVWnovIW7d@HZLs~E~^fd;1enRomC-`x@FP4t{ zk6q2XSWbT*R8g023#qJpKnLdgBjI=&D(zOH{dfqD8Wc(iVCKKqriVw9L~{7?2~g7; z{SW@j-`5D|fS~d>`4iJ0!Z6*Gc1!9s31WIqqV=hdSD{!96jfA^e0{+ssj z3QO!C=Y~#CFMQZA4r3T~_(+b#a@I^R9SK5wryb;|H7|h1p1n*t?;9|4)M?BN*#|SX zMYuo37jyO{z+!$nestyFr_)4~bo*gao+pB?S)%k=5kARsQ9JY)H2)SsEjAs$&*UQP z!aj`LTY&9_%gH{RW&dnJBKh!bVW0Oyv0BX$ZL2LXXUh{2{{or={KO>rx&IiyJP2C@ z$6{B5SDzA<8q8k$5Sg2wWB;F1BKf$_Rn!sm8MOVd>`wlVaWG+p6nS6~t#EzftZ;2M z>^H??s^56LPJ2XU6&|?$dLG2#LcF#m)MlK%bZbx{00Uib? zqHkjwCRvrCL;E14J{^U(1D~%`v6Sr-A(EB6>VM8iUuImw1KxeO?S2IIo@vDHdR2yI)1Dz{^XhIiKBuT>uW59K3uMh^1dDUq)1NSbwh4M~-3 zBF@S|NFHH@!ZXr1WqzM7O<*G-YaGOfh9m8L&-w4c%RxB%eGE3fiGxLaDvrI_hy7+{ zC_8=@Imgx`bo@574oXJXwB^|RhXt1|Pqa4sA~awHT0P#;rI+uh+vPj4w|uA8%I&o8 z&Qn_RBTMFQxu<=v?#d0iE#g;Dr!ha8S(fRnQ?yL)e;!K>kKAe1x~KE8@P6r1)qF{Me@TOW z&M%RExc4g)X);gL zrdL+~UCszdrS!Y?^u~5CIqld-dQQt3o3ULQ((c@TA zrj=(zOEM-3VIOpl=+g_6edrNcpOh0bKOuwrFXWYa>krF{iDi_ZVJ^)tqT_{46z$kZ zp}Gf1p7(~XCch(#d57gA6Zs}Aq zayt(EQ|Gm^TJ*pAy8U#Xd)`?7h>xbF&U*BxbQ>+ss-V$N&x+{km5HDnD`igHxz_aQofs|A+9Q&8Z=1fu zj#GQVIr8=GSx3h?9ik~ao5>^R7VT-jLk|iRQ8`!>rxbNCAwdz#j-;E#^}qiQoL5Ws zl5)y=3T)j)j}?sIvD*YgLd~Il*8+>?s>3nB5N%)k_RzjH@1MTCbo_U@EGa|&JO2!D z=BTRS=NAv@{;22FFzh*9YGv6vdIT}&M-?zi^OiGjcZ4ySHuGpGC6Sn|56ufnqTNOr zw29eAn#Or_cxTVvq8E1hi(Oi?9l~j9Rs{Kt4x>)btu*VX8QKQ|va5{|rZ0}cHdkmw zunm;L{g7fg8#C?r=Si>8FgQF47vlm^s4gZp{dOBOY@j-YE+}G@J^C`W!zYM%+WMt0 z-ggYe({~=YIour8;eScKT^zV2fzMT+D1KXm&CQR;F!yQ|G(GqKqYnZ)2m0C}U2`bx zx3PrJY~5Q(*6X-%Y==KvN^#Q_n0~pD>aen5LaW8a)I4etI?sZR@4&hL-Tj5AnQ<%zgT(8Vm3co^|utlO1@x=E*gw) zTRl)Zbesr&=Zdcm%8~!s@Al$#@sdFIT?2SpTOw%haO}UL#%NR;3gK!WE+k`<pgffcaVAY zq)>hmUY2;u5f_fr#WWc#($Pe&fdr1FeWvF9K4^+siGfQaQGa_B{1PmX{o4kTccpMY znt%3O;ZKfT;$m_`?G1y!9%LkBG%;fIUn*RGg_MuSQ_vY5e3iF=x63P1dRaxc`Ee49 zbNu!HTrR9Gr9k0rAwOnZgDa}1@T`@G;Mu8Wy4Q>jn zU1D*_vDl8~U;Kpi-5w>98hVwqX{#$WeHkFq6*tvV=N(J?ist|1ngYl#Zo;{eMx1Qg zhmvWwX1DZ~BqayhT`m#F5ka&J>X**6j2g0_#d zOr*DV#-Z!QQZyO)VZ%vvh)?7qY;6;!wckQb%>#J2c|f8z3;I-m@864s^tavTd%_mR zpu2e|aw{cqG@&2L!fepdxZofD-W7Y%hsrZhUvUpEGx_^sK29sA@tC-86XYjtM(*b% zOuNvt2bK?0Qyz2pIHeR0#Z9wt^q)Tm(`pj=F}K<3f9tq*tJ`AQ%cGz?Wc-53 z+h>dTw{m!}RRb!^^GLN?gK#;H4BmCnf@QMUS*ecsrxZ}-tAb%p2dQuWCWP$&2K(eL zI4QouxBO_5zUoC*Po~p>_Z#V4P%dpBol5}^{3&;a_doIy;ADljGdHX2$Va9x#AAKI zzP^c!2Ue1GbsH`Gtb!w7wxhP}0Gg$C;^XpcOgAaR;F^Qh{@%6q;Ke;{Jl&F|E1(tdLwm}j!I-xKb;@+k-sM> zuC_zVH-MY$1eRnlw56xvf%_xtyoAB@D6WMjBxxWw^nc@fKhNo4GcyvmbtmKPu5?^y zy%Ns7^$W9@Ek2i-?#L+O#V?@uxA)WCfOZPFQs&S0%}iRQ2C=&8kUqQ(vbFW_t4Ie^a1N)By}}TcB(sbIchI$b zH>~#`!)2RhM9DQ`>1Q{(Ib#kP`bE=yxlcN&q2Ul4We4ln(}3AmWaJWcOWvX8XcZ>$h>y|J7n%N7WKv?8)qrx^YG1{J@9tjD64C}$NBNs zr!@Y88?01AA=k1A3d{O29&HJXzD2gKW$r?z`i8jJx7^uq^y$YxKUWaXRfcuQK!+8& zMTeDdslpl}myBZNa@eb0hx^8Pijufsc>f-aTYex zh~0*ki>K_=sd9MAs2=Kx5eLkzWwiUdlRbYQkQcrLw){M$Ufc(UON$NTT@cN^-ue%J zKVpu86l-kRSBPeA5HyU#ePrycg9OQ;& zHu(h2p*_#0Q|xyqD)D_vn`c?W>CUc!6PFD^k(NDJ6V>5$NFF9@50I{!IvVvMKV3qAk*0Q?^RZGo4Kdo<~XZRR`TwwtFK`yf$Ltqnj6$&Fr8y`zx()*Tm=f zeGvBWJ2f~jCF}a*q8ucn4s8tZwnpLrWBgelhj|yrFh_ojWJ=6Egt!Y%^7lp>kA*(G zK}bWSpAE4$}3dDibBo!c>{L?5N6xtGl}H*uzG0|lm|476jj1aC=8mU zJJMKto|p@zyKSO?;YIY5PSCK$duUsy41AmVVA^a=G)KP@_Sc(uOtZl*n>^Yd@zXIs zQ1yTT@LA@An8rmIGkiUc`DY_*N*;<{uEDAuzUasg#25=}v`hWJd6<0M0{jnM=1GP= zHN^_YVNkd~1lJ0jQEl%})>Fq@Ojuwm#HDsrRyXTb&slQz6gjc4)}!&~PAKlLn+h}I zp{RJqkE0w{M*F1?6xC5jqmAsr6xqV^Dp25LgfiU#++ClD>v4%tcAAN%d_(NLr-?No zKWOI9ePo$ufQTWkIIzkUzQ1iT*FGJChUQ}G;BAO-PQ%)=NmzX_4mTneqJhUjbVxh> zYLUnK;{F=NZ~x<`y1u-W&dVPo%_c{60dKd)ey80>3~|`Q>JtSeSft>B!BJP zWGUZ_c-qU)78bPkF;5ZZ)Tu4mN75?`=+@FxG~vAlB!Ze~?~ZOtUe`vCe%&M2mT$EB zxD>7|(1ui?8oK1RVPonxG`!En%aBz#@a-*=y=gAB#~mh(w-+g3@hSQqe2wfnF44Ag zrzzNYuZRY3l~>G!?sH7H%40@ZOOAAHK`+jzVZXN-PI#Ik(MMS{2S2?x6ZUHf%|u@9 zPJHRg!}uvH;M)3IIFIMHzZT|8ZkgLj_w_nSHolcMX#S*~uUBAOQ4*d<1w%2>R0vPm z`mI#|VF^qZonXZGg}{ir7Tb!~VE&9nc%K?fj=RRt6P^E@Q3UcicEMQumC0%v7HcDf zk+gP;R{yX5=>Ae8uQa5`Cv)?VEY+{&UaxV3QsE8iy!VxU#EU`hpbFM)9R#B@5lHD?j7jUq;B=-AHXI!Q zg&m%_n?DNORS8s*BdL`&Yyc%EPNae@y4ZPM34<2MBJo8}UZQ}`7=GON9e;jK@Scy% z=Tl%>F%nB3E92R%%}}VzL%G6U#2rh(Lkk%yQBkGY3r*vsIBWEkyB!lzxoD0)X1`NiL%i<~^mk>~D-e4qkY6TB-)?s+@0${Oi?>T{Cj z?V^3#j?r7K20Al8N2I%gwtf2zpnikj(uBnhXdA4Aiq+x>-@bs}agWfo`ok1ha*q_} zhN0=q1{8OuBS~)uTH7?SXM_Tltm~pN^DF3St0JoRSfM~p6<;dU!!(m_;@T;k2J6 zZ_zpA4ijK1E2Q(!LJ4x_$K4nK_i0jz*I3m4;k~Dqek75XnjazrU9bbwsq&h-(X^DkAN1QOQTF z!f+4P2b@4wXgQ|a?f$2|^>!=mZL2xjlrO!Zi)O9DTs8p>kAE_dwDLx`M*wKDNHl$8CD5)-+vtQ#2wiNvNr9m^NT;Nle$8#Bw4g++mQ2Bwngm?U3rEGI zi_D0oD3VS(MyEUPkXD~*6jNPH$7b>9c5nwt{yu@Q{aox@T?H-meORu205Y~)u_z`Ns#2sYV)l`Z^y>wEk&+%kVC>>LWQ?BN(F2Bf1q(wvM6Qseahb^58rS}RK;}*`H}dxgW>FY!bB%I{||rq5r0VGsum8U zD&dHp6w=3x#f`f&U_3ei9~vgYWq25U_sgUh$s;sg;SEKFYSSuD{(A#_UX1GwkbA;@ zx__X7#`nKUTPFM@-mm+l&0k5EY;Pl;fsbSB3`L~9YSvmKKNlTZOBF+Bs}bk~Q64RSh_T_v?~=7Ss7;xUQI?b8xzPy+XJ}!?2@hNu zHxWOk_`!4FN(_0N0=L&Yv1?;CqU2M^g72%?@#g}SKI-Jt+DRIQ-E@WXhfZrrp=(zv z`pizp`fbTrY8}n@WsG9ZX$3M>+z@7Jf)V4WmyCmdcjAx6Zhj74A)cO$1*ObH^7@Iw zx#WB!nC7foOtFij=wSIC5#7DxAt-;Il4n$<1~Y>nN{OEPViG+Y6GP2I+L?uFL8KKr zofJQZQtrwIW}>AUJ&gRyV5b5NSv3O^gW{nY8wXkIF+!Y%yOJ?rP6~2nY=ZxkC$yHw zLa^>oJUrwF)4y#rYPBM|8tri5@I<&jeoeAQM(BIq1CNi-`^SEadMzpD#GiL#XCxx( zQwT~PnL%yjM>24;K;7nHFr49srt=aQ{`o#-F1<@)ledugYrc>DTCGOz5;II=`s4P8 zEA)HXd(yj~Z%kKJF)lctIC1epJ}ndts6yRXWd~SLyD%?aO9FV8q^VydKZS!I}WD#w|kH<9P7+x#H_m2`;bbK1J3CkX)V$j*tf5uawYyOv_=(OxU>d+GtJLy$Rlfw+LdaxtL9Ucu{LQg)f zpT=D!o|YJP{V@KY{$3wnkjFqZ=me`mu~r)esvGINEK;dE*!CZ+jf@HY-F>NsNAsVVq=-u6|^gKC|EbsYI zg-J zVU{1A#0(mi$P8U7N5?0urmrtA(yRmtr2bLCsji{a;+{%l!dKH)y?o(VhwT|8N#ReJ zZ#93JYlejUNDc=0xMIfYcO>Ds5+0GsnAVnoE5RGEY`r~l8|Q&FVl65MCSh>S9~#Eb zb5vZzwG$LkF|R8=tOx9 zwWeIBx26BdlkAnBpsZRnRQJFC7u=hhs0#PZ!(+W*>F$ip{hfq4*TW;$(X#;$>6pC? zVmy@5Z)*ZIdM_90_-$ohOc=O;?}uN5H+}dqltF>?8aDJz%$%Oh7($0VkC9V3m+pPv zOP_M{>1n(aE_i5R4S&4~<^1I58V;tt&oXtsJvYa#lCdZazCner)%g3Z0f>|tf~fv$ z5wbB8hi-3y@s|y_b|?i#xhYuSxD0P+uKh<|IO{3SyWbNt>9pU-Uw2pX^PFzd$#LR% zll7Ly$8PL9?(`SxlVSkf0)05}Vu_hrvn;f(7vX#P&ad-*J!w@<;y4lz8ERzjx!U;3+iib|z=))^(4L@e8| z0h9VJhQh6Kb47(;7RX#n(2^!PCaFg`ch ziE>WedzRYc{-oEotZxD3X^iK4;@~uNrUqYXsoW7Sngs2rG~miG}(L=y-IZ? zyODFKo41l8`EjA!sdhpd9A}jDDO2dlkMp=xCbS&_JggHHG^uF^Y5&+syK1?VE$)m} zTRHeL+6IHfJIOd}6BE^Rk9l&soM{P56!vS%6c<}^TcplRL?Pc|W};r?^S!5#uxeCG+WO3$tcQ7gHG3 zA);+<(?Cl9?$125FJkr=$*R4%WX(TA^OT#cSny99hNmlg-bK<=93;Y>HK#b+&LrR6XhrJ>#9nt z(-`Z90Y~m>2+wZsJ!FJ@*pAo2wVe!p%2>fHM+d{&LU7qK8pCd{wH!68=X}%@P%m!r zB#mSrruK2!5J`95=g^hIndJ8{k#@I|XuJe;oY+AB$Y;#CF;qy2bgp7KRg4=*M(6%g zu~pAH>+`3M2E31Difz8=`_7Od{jU?~M|Uozs-GdA;sBk7ek}Et>BY3H{}l2u_SK7VgHKm%%VwI^HZqHB!Z;2G%Y5d3kWBnzEZcmD){N+kyaY&vHr#xib zFaIb1t5;XA24ygBPaR-f=5#VUoF~zP3wK2P9KN=n8O4vGROT7d2N^>$Z*U>TxS3Y+ z-zQz*_K6&g{!%Re-CysJHrpSLN9oj$E*iq!L&ZlMNw2PgUaT)9r3Pnuan_gb&l^Kh z`!|x~d}*A$r~>&ZdU$Cm1=EY>B76dxUhB#U<#)FxfV`EqP`uX+nxDp@iUo@N-0DWA zY<(NEd9NuYzWB>L4dIXB=MgQdq-j&FaqgNE0>64I6na0QK}W$HADqjWa4s<0 zUMymoX7YVu*26G?GaSv{Zg|Y!OCGkpM_2fM*6C@*^j)`2GGMcurgwH1Q*>SFA6@8` zxuD!zG?3Xosf01lup*85{`B(Br+#OOmxuL~I)&a)K{P+-wBnSxa31Zw{h#+F5;LMG zZOt~Cd!mlKRw=++tH+md=Yut+eGf9@=<1`Y))Mo%>ZqOnZw}_PA3Qo+w3k-UY&!Rd z|DKZ3DBAVsGA(dc!MIL+sJvE1a=~XB)UFPr!hy&%w8JXaY0>!l+!;nkrnHm#pT8tA zR0(^wXbb5}=jV6{WHM6!IqjOWpDecIiR6%wcZCY4zc2#322RFD>6!fZ_$To90OQbMS4$0bUBdUV1vp~mtPKnJKK$4i zdH6eYQQiAK^hZ6IzIcr#iT67xWimepF<=taF8X8S-J(Owm4{Q@fu7v?zdU7{^~MC9 zvi!WOs{3)@hUHBie z{Hcy@9sD@WfWDD>vVusYg^PYK$sv8=^rQHpr6ATR+OD z&2;!i3MDzE)7^T3$)GsBq3)(4DN?Q>>Qh!(v(l}MLLODNFrIE^Up$#Y(u z)RQamt0dJ?@k27R-6fy#$;)ESUzMXP$J=S~J25n0?4-s+FN8EZmaJCj)rXM>!FCpxEnq#QRav^ERCsEbp zd|}anyIj`qq=>uLLTBbj2Grl zt=6`o=8AE$-ja$~o~47fDeBmo+A}vh+jtbB^=}UB-#Mv#zd$LCn|g!%{l)keOd1Cc zk03@aoS$#;pFPSmt;Iq-y?p;aoV|H8mQfctYEEV%vr@=ZiTCV%UX&CPMJlC4X`VGm zBr+8lLNZ2*(x40tij<0^NE#@KBr27VG?2QF=Uw@}Z{6R$Yu)?Dv!3akefHUBpJ`9$ zzwa~c z{wxHGd@gi*+Jd{d4OFMuLG+qgaJEsJoQaw;jIZFM1QNQnm`t_1NG`9im(9mcCADH+vO38_&eS9=o=V3wCypjh^2M)B`kA?Wa z|NDBpHvb}tn01XDntP0tha?lP=3`_|(J^v#-T@+bP@669r?UxJBdJRsZ2;mX`<=C? z-k5Wo9pgHRTR2U{#hjnj@hpD|eX${{xn87P-jg)cE+@&QDMaVsS>iD#nY?aF8>Wj~ z!AnVQ`47WR>w~0s?M<@Z>jv5Rt%1DQyoZcViX^MfC6b1+i7C&Is}jX0+~ zBO#x|iS*6SWK_Hq6m#U^r1=EMU1taO&9)F*zZinQ-sNmFIl(FFdd1P1zJ^mGpvSo< zxQtWLt-<;9Scq@>{BdfV?HY;N3|`+4TCtskRIMaU?+=kS*C;Z<(GaL6VaEemZxfhp zDhq4ZjRTXF;;>&&1f2bkljs$^ab}1wVmwk=nZVHV-qvK|y;7c|7^cF~;#_(d=gNz- zkfK%yd84Xe*^jeuQ}!9z!6_p;pYn)>$VOt4{DZj0^Fvkf08tVV1G5!O?2PQ)5W{y` z=A8-8ulc|EUDkS)X#JXS`TBLrP?4z(*B&a7>nGL_ja!x^uF!={41Gz8l7EvgaT4&i zrGcgQy+YnvTri!9#q((_WHNCm+)mAhkO$M@=^Ilh`SFB*Lc%J|!Gah1|6Nvoo;a~@ z5QOowi%E=pAJISjAKlEqbdmijutN4A%>TXu?t6&BF79;buUZVp*ZD(5=NcIEyMtip zC8Ba7m&lE~M)Id`fkjc_(6B57+BDWf`TD3~p7;lI=>GoCXW|Km`CWk^F)sCCMnM@CKG>D4#RLYc7l~tVqkZ+H*gLy`ldZReat`D;rrvrUF&S} z<3k!b%-90{{2B=8upX`|t%8mj6TnC{loYQJg^J**(38ga-Ci06Q^FlVV&+Q7m>md) zCf9&dT?t3F-U3bCEAW_!Bk~XS(B0HFh{Yvh>Q=H?e>d6a>;@k$1;OU*5O_azDKs8y zC%5i@CpwRAlY6n3$-Q^M!?+LSpYd=uytdg1eD^$Lt*vFi%ux`G85@814=rT=ha%$A ze#CI|>$w0=7^iVRL)dsJ1q@vCK=M%zh^jOEwORl=H}w#U)#c=;sz33L4uJ2|4}q>~ zIGmau1SiW2!E;I(r0>5D7ZzWId5I|?eJzzpxG5ETvg0&-df67-KV4#=`tGHBnGEt&|ju8zA5*3JdJ`q5ZB5 z9NB9IkG_eY)DdQ3o($#rDgHSAjKl2t z#Y+k`$c3gGWWM)j;xzF$sonjN)CO6Tyld{n-eM-{=-)nBtYkG9W`^?SgAZP?C|;K2 z?J*_-7Us-e%9SLg?;~EDBgxC~Br@&%4f0}{GDtS-!Mz+w@XmTlCd=;O*mNJ^*xp^n z)^*)f8|d!Z2=^}d!&YSML_RQa8D^d*1ELqmuA8x>PkkAwS?WZxcj}N1Ga2$d>p96j z!T2GM9S7fKRp3Xp84G7;vn=QQiT&*Jqdl|6`49GS&TP(L$KX@^|B+8Do4%b?E1w`0 zW?4kTr-1wx6J=ulC^B=@XmGnP3qh;ZS=y{p@*)LWBFJpl6{LexOoY~*By(r-_Cx>m@Z{`CJzqx)b22-weo%7eC$ z+@uefT}x~nB%rlY2Xu8OLE@W4wvKHcb6EMG>_|aSTg>>mIJJ|lp1H%Yj`wwuvu7KK z^EycgjQvIybV-!UY5hc2of!kaGNVD}Vj|!i zOHv%Y29|`ofw&hlmPXsa9!Y!9-u#Lhu0eplE7`>+h zj10AfzvJh_v%jM7V)38hvTNg~6Hf0wlA5!d{dV1&3$@PHu=CLjSmrnzE?rRqJ~V-1 zQ7(K9;@O`3!{bf=PXXbp=SY3)lkpGZRv6D)|6&;KP@30&OGt;_I<0ZfBD8z%){%O( z|ICa3V2HUllb(*>Wc`=FWbi?Uyx+ET@>?gDMD|q@sq>7l~2geWPSM1j1Ti#X;dsn}JdDC7)UD;KT5}OYjLo%WN>jMxu{SI6#)4^$TDJ)ND zgH7V?urTrodHD7qX|h{MQi}hh2mFUm!lQeT=+Omv^IpNq@S8CA@=2J}aRuBD+ybxA zCy+oVAagD`eh9F`Lji!_(}RJNp9OL*XTj?36>Gtyu|P9a1F)q$*dAZRf<@A>J^!ROBi)*ka;-G=UgA)5G6W(=mR`vd16 zm%vT66A;C{0p~5=LVD6je0FI$i=(NR1;qOrKtZw?t`btgI2BcNKCFPPu{p4KV+B0Y zZGxE7A7PuFA&h@A3(6$!Oc61n9!mb0>-{asiUrTiHx5M>8=6Gtr z7$cIV;3o@h^r}!nfoz^%`M)w$>l@*`=wJA7LkQRUHT+L|4W(&(5)5UfA&}U#8M0dD z!oH(_Ak0+;Rca+MGvg(s92tc8mEt&RsvthQ-3%vnN}(&@89X(ugN&vkNLee72fOsJ zM`%0-wd>(m+5~z!Pe9PB63nyoWUWk}gY4UHkU8@kEV{vuy&p}`dp%-C83*k~>EetR zLik{kFltWjf_`oVe7C#|Uf1tInD2QL4m>yl zZsjHLdrT{Q+#!Qmy+zQutqC?ieE`1=U%}59nEd}W25MtO z0lVM&$4^hTw7~NdAvmynGH&;>L(dVL@I;v{sZ#yU3G3O%i9abt+y!2e&kK%$hMX>% z&RK&WjS{iFEsi|AGQc^sniuoZ@ku^;aIYHzo{50mI)1D*zJ`lSh|u2Sxu7jC#`bei z;n^_X3-5sZwZ9==GaN16O~$DXM{rDBHy&Ikhyr6&Fm<#cKKL0yj))#17mOIaD#O=h z92*!E{tS%1EGIOj9jvSX0k_IwVZ0e9hD^8q}$ z^Z+CV3c=-t%CJ%W5IMGg6s&nWj>TvFQ$d1G!Vu3_MV;*X>BK}`?oe668+d-VgNbJ3 z-pLYj`GN@WCFhvrywBonn{t*MZrwNKz$iYxzC1A#uJK+-b*X|jD{XYi`+}M0I`QK2 zA{M@qq01DHesg2L5uQf5TtA~124|;ehYG{ZeNK?punE>nJtse1!^s&(4^nyZ6-QsW zl)Syg#C~IN5WV06bJjVt&!zLS$?Gx~IJ^HCI6Sxk(-&7mN^36Kj(dpBTRKr$*OSau z%HmuwS0npRNRUevo5_)JnIPAv4@%D&znRUkpwuw~Cq=JEqf1;GeBmj+6X2un&E2Ty zF3WK>%;Y>{;yXybJi)o>MoGob4bVO&#`tMHa0Vg{bN-f$hETo`kXz`6mLp40r}hQT zcW6PA*9Dl`n~T*uuCw*^tu(@n4i2sAs4-cxVgVgO8qV#}(q#M6c%t=90&>sQ zqnGh7JkZ^acZ_xu=w+QOal)x}05Bw5*l&Pyfi}&VPo!~G} z7oUGI2hqkSqEE>W3{kNonqIwSvY#vD0*LS108CM zt=soOoL2)V9~DN=$+JimCzCwiwiPaTXMt058SLMsh&$x_*)_4?ojGuPqX3b!9>v!G z@JuGY%FNH;zf}-B`rgA=iAi|F$^my@X@e(c^iiV#zi~oU{B}+Xm1lHMb$PWN!1OC_B%nr2|g5+(Lo2nnIQ6v4@;4CdY3tccI4*AOXRa{B2-U10_Une zL6MLgjx5;A<{hg0ct>%ZwEh>Yb@>c83l>B7x0}$iQ5Y4S55hMiUKz$uuVX{;4K$j2 z9~=ALkR*@?YgK2kuQA4?!5I89p`UH%)iqociK=v$kU#4a(HG#F@s z(cRNHVo6CeRWD)z(1i{2Y?@;uo12k?t=Sgn`2WY!E|E@ zP#%AdJW35G4vk$fpfnXz27=i-HU_wZ*1m4wi~#oC{#r&W)0>~ZpwZ7nEbdPs>OXtgzW+1&AU?}Fj#|YLsMs0A!n{9bB7Rl; z0)ZyUFlCPc==HwA$Qf~%XfKH^#y?5I9RZg2)<5{q-o;?_1XwU@^l-ZlwZ+iYGAEf! z%vN(7UXJCCGAJQ2VnV>#{t<4k3Z@3?Ms(`VUX19PhiyMJ(Po7hj%&+<6P9{#H~l*~ zD|!RZHE+lA_FkCnz6U2>lfd~lYvE=2SU5ZKCuAVQYlde@!rVkMGVc&c|NRr1q#lB3 zMk2&#>f(azk+|r{Z&iO zYrKB%7(82VOV3p*Qo+V!EIm-O!sO(PvFzM_XmC3CB@2UJ2Fs5=U7n3f27< z5SP)9IvT>%rnQEJu}EYsG;Q{Vn7UDP@S`?;@K%8xcZSODGaQMoD|FGsNC`hQWP`hd z6pGK|`Et$6E5{O#n`r293vI{bz>=dppCU*5k#sQe8ZMGG!a(WKAY{P!UL}-*ftLln z&>KSiAMB?S%ED>s;$m9xq@M2cK1J`I&ZORbQFPMHV01m?kJDd;V4IybTi-Q-nRr3? zFx3?drUi;FEN*cdJvsJM`S>=-c(Q4voX^78x_g+Vtw1$sFYYO9NACALtX=2I*3mzB zocMKXX8vxFqyY=$sG-v+>bTMbW9?Eft*;wbS}dkdFD#^!*CF-!n}#jc4;bFq!REbl z_97|tt$<(Y(p>w~f?Nfc540&L0OrzYh;V-cw+BnHDSs8dGcdqeZ8B8av>UbVmg80& z%jO^GUCG`(4pa?!buIMKt|; zDLoSXhU(lgVblKWGZnFK$Y!tKjQzPAeLHg_RXyQJ#ebwx?q44owqPtvpQs#joOw7E z{~V#Pvg0Ud`aXg`ZQd-c45<%1SjEL($MsQQ{Boq%u7HF4d+?vq4G%BN!l5Z^;hT>s zc>e6i8TrPvuX`qSo@_vMx((?oD`~ouIX}>)xsmeqzoz|Tzr%*DDR8&yG^`im&9isX z=HPZ6KRosF0-jU3f!>-c>BYXOv@K!-9cQIOvz27&9LFaZ6%mLMBB`i!=nY#x&0qWB z$}4qjHk=HN-U~_S znuTG_c39v32_*{EXnt`gx<(7qC_5dh+-ZREHv4hY!UFVvunO1zoq+l~6|rl61*Q0k zZVfD>?r|mP-Q~cR|GLAE7S)_&+o5&DCQj!YDdIHH&f+th*23$eHTZ715|+rnfoUD5 zFk$>u^zaD7{E4-wC3hb>r*(mxss;iao zH}$zv@G_|YJ}$b2BZNL;aFZwvHg87n%tqYdl8U3AI^%WO6*#VS23|ed069H#Kw>us zK3s@Ftp|@#sP`OBQ3~QddY!=ipqa>>aOO1T{?x}k2c*$Im(UFgJE^u+1YIn03b?oB z`0NgCWnz=Ev1Xburq# z6Y^hmvoz_AJB%9BBXCpELd-lm2_>GmpwZC)>w zX5wAKfb4?BkCy8M#<~tO>V0Duyq8Q`q+D`@J4LzJ#M@>{|A{ zF!-kl&fQG}^3<`4sb_tJ1gPr2&7@b;C&lV=TZV7nA-66F9*&IVz*2yAT6MU~(l7H5}H z!lO5Dt@krM4TWZJPRzs1!XzlK+XLkV>%nidDfgYJ0{7Pad-ThV zeBd-jz`Cr{EZw@!0?Mi6;Sy%um}c}8H_xxfNnU~&ExiYOe;q)X#e;BruKP!ClMU-CTrh|hA5V5d_CXi2(2bfhSjuB``=L_zEn zQNkNKi&4{Y5sKIG)3Yf?^m~6WI}SKby9|3cg)AM;Joo@(Z@+-+HSJI_XoxO)yYXcI z9=hA*INfmd8l4vR9KTr!QI~3dHgDVAR;-*=j0?8QV5XS?Ua^eD%pGbbsUCbLB9%b& zaTJcZWQcuYhFD`RK|Mt|)N!>n75w}ge}$URb(yyGj;bEU$}_(CJ}EfAwv!0?Um!p9 zFB1Q}OE~ML7^T-1(PED>_$Hu_zf1jbwO`cZbOH0w9PzWp@GO>J4J%KjeMi6k5 z#%)jc;hS1}W=+ruTZ})#@Ar#P)Aufx@B4#>`={aHo)2&^@eq{FXv3xJb!cPtbSkk@ zm{|j>q96Yk2pF>+XGC1ZXGH^e&1Vj0;-V}QexF^i&>k64MN9*FL(ukhfr#$xBX3|x6S36oB{;6A+;NSG}Tg^5#9`E@=1SaleG3%O&wwKg8S@q^^uEF~Mm z4ie5=R}9t&#>P%j`sU0hY?YU!(s3#*%po2by54_mA_sJYV1G{l+eQnnNnzN5TiC(K zRXGo8aBHLhn|{fYi8L?Sna1Br!Ti&&a1=AX9BoL&Hwh>4Pr)~Aacrl1EN{`i`Up1t zzAcpy*}{`443%@gqXn*NF2q**3_P*;4KmSj$fRZ?)E}RZr)5vT;-BBa`Sc`ARxm_U zF*D4hTKM(M6z;@lUfdGFRIb*CQdDochznveQJ|&?-5<8$nNJE-pkfmkuDK3DmM$o@ zR3FxspCP%mGGz6bbs%QV^SK);-`kIou@8)A`($+rAf^Zcc@A0f=}C(z%7=-h1pDul+z8Mb|VO04W_bb8k~gj z!|G_fVVcE*+4K^M`6L+|EFskJL44CAhtgg?c#9dMWsOAf;uI5%^YBC4OTBpf;7XdY zJC(wsLl|*17GLviz=bvaESw?SH}s1_Uc`1-yjzpSUt01lS@t*@>QxTFn7xMRRN#XW zcW(p0a+K2L?K`Ave?s`cVRXulMZqI2xUrE#m406(GIfW^(FjX03_gTLCM5{cVsy)= zLQD)UyInjJ*89g9932bE3P1<3PYZClauOpJr3sK){1YP|_o6dZsjNyS@ES-kx<7JbN8X4bNc&F9| z66-66!1o6~<8N64$KKw?CA#}@#8Z9>Qv#^pqO+)5EkkwQt5V%*Es(QP87oFuppJ_u z6}>G>_bqZpA!dyz_nKKRtct>)I#N*9rp?K7j5L|I)&#x$FR^soo|J&+1teG;LK1TD zk@`tyzx@bho<*?#$>>wCd&xJPGd~j*=i1|@{iVbB7TljpO&4sY+mqdCqC*67>Hw8C z$l>kcY2dN)DdU2%h=gg?(|c_%XvgXbnzZu{xojuL$f6ZEmJ_d|+%s|dC7eq~9y&V= zb4zP5Zd|bhEp2$Zg6O~`lZrdzjB_2$Sh@_+NyAIj^qZ}OsU-I=x3>2fnx^TXt&SPU zHc4_vfj#$~gEQAh(F>)=T*B+&rEI?UKKCK9GZH@^UrL?1aa79gG&OI!iLpDQ@L_^E z7L2kr9)dk|DahI5XSr(#{8?DCeksm2D8Tmhg)AIXCne0*aKwfcdhqCo9dONfd&*yG zZ;Ab;R?_l52iMpO(BpDSG&Zvfqw7Ry-Jb{8RAEgoUGb(Saf|6_k1Wg>e;e~vci}Hp zRXS~7D?Sx^3!~B}fUDIQ7#}(XLcjk-JqtgaCi@ITt~f!~w8<=u4t(d}Et5nHIL=S+ z1i0falX$FHUdG~Pz!%HT?M+_%!?Y^Aehq{6PGH}IMlL}ax6dq`S)odp8^w#I!+FkS z&IBrk#V4wvGq{PaS}VhqH#X(YS}jd$vXtp+lTq}S-6LYyVgp5fA7JKcCZ5kL9-JY1 zS(apwO|C+?|40x$puLHTL`l(FvKV3l1YoJ~Z1_GP3CkXbf!kpv^x5Bn-p+?`_C0wV zBV&oC*M7k0YJMDasHY!HS}2jqqS`Z746jjUS58C0u~C@N@c@e}6o%_HRPHwS0*Df| z!{w$yIQTvk72Yky6R}2^zLU4-SMuhgVlHFp!p+8Xioz~%yf=Y`J7PP*QyX|`x>oPQ z9geBkyv&G&edXS7xZ+Lzr<{kkv+=`;(^xHd5wB^8VqJ(1DnE3?Z%t>Q=I}Y3=qFBB z5(5;ni^Q9QL3s0g7bLFW*&db8?WAq8gVe{kftD0z!OP?YEd33g*Wt0~6KK`;3gb3* zqyN@k3{C4}%her!5;mS=_73w!@yh`by!+xixHi<`2_Fe6CO?36jXSu(OPO;~r7v+` z+=;>N%Nx*o-a*K?T1mM!AE=vGE4{XLJ9sm)N&E0H*t)SE&VG`^MVG4Bvd>2tVy&nn zMrh#+{TdplE|wdQo?(s7hZq(L_mAIwSW z6O?-;;o-z0+*rYLwu|E@(C`OAw}2jN=6rhcSv;yS^7&VjKmLOY5&7}Z=Ka}&;D3+JT6R|4Ikng zVMY3E^;Y^b`WI%mDbw}0Wazl5{EU354eB@Y;k>d__-n;_yr;~QH@Pm3$Cu+ip~FTa zVkzax`Eq-|$;xy|H2oFG)}=aZKVCTe9&UzsL;EHf`kX4#a_≷yj-HKZLL2h_Co- zVGX8goy6ew|HLdD5Ky7}vRl!upczNE7vP`tSPYoK#hvnj@OZ*?VirG*ERrsvDK$^% zfsL=|G_6*0cZ>*mz2lIQi*zz53 znzHkd^|#CD;&KOFRyN_(`xm+GVb{5R2X1kNCPibDw+8Mt4*~P5-|4_2ajssA6xTnj z1#U9urw!r1>@8x5*dmn6v_fAEBkGU7w8m^R-8Xpo6A*Yb2mx|@QDPu3JN zuv#0hR{Nr5cP>udpUs1ZOM1*d;nm5eD!t3_JwRaDoWu_#m{2GO!-@jt+(vkF@rUDgnIEJpv zV{utZ7}C@QRCC%4dbo^Nuc3Aux<)k$QEIJ57r#}b*L6nIkAZ6ROuQmn_YCnM=-sjn zbUo`(^v8ERD=I)eZ#~D27D;$SD1^n~v49trI`-nS_DzrtWyH0>k;L@oBiWyVzc*jS zU#q*I;38wgGDj2Nt_dM}(b4cQmp7+>lR5+{^RB?B;zSUK6YzI?73gS$;;nT)IC+U4 z7RD_>6%EAnG8vq&tb_3@98vGee9X`&!(}!_sAf}%8}6T?ecQ_!S>RVn=G+1a&k8Vl z{1)1e`~>k>CPXH4HfB%dNZixW54(-t!`LhXIP#;OsBCyl)Iwz7FJs&Fm+?QWmY9OM z23OcRZFqN;etPDQ^%vI?5xRYtdDj>7G`Tfo5WE*$+{ ziE64PC>x)KLAE*Y?13;suo^}xyy6@kFGm)Bl_&UQF-~x^MV+=5*x{y+Nv3ibc%=uv z=zE}C89}S-#uzhyIzE4_if@m}<4yG#5WKVz4v+Q%l|p;4P#gs|6M9M0l|N9n_R26n z4Ygr}f(P07{wZw86GoqD6Yx6Yf2$cjl}&RgW;Q7GPgpYRy$d4#-dgRVX zIp?Hlm?&M6sg=+a^>h#N`IBF@R zMplzt<7O5=-$+xaGvLW5hj7hEV`AUl<-;2vaxpfe1pED^sGh+H+$mwU!uqB>sU7Z5@NFtI{0_Mra2tc42wnO$@f_J z{Eq2lh0|2<%c#QJ_7Cv(f_jYZ(xuz2W2v0qakkG6!5g|F-s_{=m4g^<+=08}+tK%8 z3rcZ$a;HOgpTIQX5%^Vq8Kd)7!`-`bVRBs-1S#lY^MVan{!yGOzGe|uN_{_QDOh|UtG>BVQN1jDE z38sq2vt_hKx3G7!%JXorQX69~CE%r8A*$bJM7tA?L7BcDj>+;xcM)4G>I=ss{_D|s z@o!SUem9hUeF-mTL{n9V3lsxZ(pO#&Ks)F%n5#EJ>J2M+lynaq4ot?(z`0=Ko&=A^ zzlQpzy-@mNDIA%119nf@jQTa}(Q4Z<_~$p<{*t&i#=%oPMZ1Nj>nj#+u8Dl(r8|^qOKEcXyL1g)U$6bWI_T*5;cj=yCgD{V>3( z2fuMQe(mAjGwa}v`)md`ZkjQ%iJ5p7=2i6O%VxUmM;C2B7z9eD{1_wWI7~OgaA7Ra zlcrGz)u`is1KP7%oqkC)rWY#4Q~Q`ZCh{j4d*B60@KR9<55=iq)DC6zd3p}!Et>`! z7Mozj+3zs<=OBC!{_)_(#D1Ib)CzL-+)(VWCz@>!#G~5|;u)1P zjNaRdYJV-LYw~RRLw_DU$HY37wl4s;?xk?{x*1w}_re%oKkyRkgqKmx(0xJmHBL-5fM!-LT~ zuy|t~F1h^Y7-%dQq25?nxp%TpkDS3Mhu(%0$hoU35=CJ}sPIN$Wo9(h1L` zXhX+KHvJI4k93G5zfX&>@6{CtNQ)seSF~Qj*N&&zZ{bN}OiJ9t{>PqfDBsX^Wz-YV zCFCT&yyuBGXU{+pwfC_6;zWG2dlT--5u~4%SD@UgwU~Qv-Ee35N9lP(e6v5j0S=Q~{Alx*&l(leX`g`-)R zL+Mu-`Oy^71RCk0jOYEVQDp+)E&saPXwZ#-xbH`1)@ujxU;pXAAm4BKbI#_s*nM!7FL>ES{crLb|TT zw?*1e_?2f*va{PBHS(;`v2zmkeO9DB3yN`PkRNt8UdFJ506c$7g{9>{j0#R^^T1cn zO!3^wCiw0c4z-as;K$gJDn+TFp2!4vIOi9OQ;_;uTv*1_G50O&M(gO)_-7_%;au3n zWoaEGz{gkH^Av9LKV)Lah108ZVrgm7Uivxf4W4d#i5Ha-=gsnTqAU=^8dovaVZiKQp z=d5JL`oN#4SSp5>?gioB*C){D%vsd>8ip?om!R(2IjDU`6vwV-=EaW#__nVE2Sfbv z2U~)kk%W3XRQ!q9(zgZOm25ELR51EZ&Onb3%X zL-wu;Fwh%`4Nq*5e`ho9{`Lt?RC%(~1KfAmrXxjLjaHz-2o?Ox*mM7* zxF7%f{(6!T{F%rH1zIQB_CxzSxMO<*hW3BMk?Vy~Ye5J`Sy!TCrrxlx>b;JGr14Dv z*+!2N6W=Y!DYU`uTX=1E`*$*N%B{$83`P&Y#>)JSB89;419?y9Ta1)RU!&L2$1p5%$ck2Z4?2sA8Ka zeI%QSR)P7jVCfT>eys_XpXA4*1~uTvoKr3OwFy^_I*EVd7hzM*HhfoViL1#fYztja z52OWB+rMsf*7k=c{HiyH<#(=@zBu)^r+T4I`8)`swH&{+v~ki$YK_*kADHS z93ixSJ^}yU@5lQmve7SS51yQS8SYv?gWGWq=r-{N@aKgRGkZ-EVS5*U1RcUD)50J= z(E_j4DdOgeC!p%^8qR09V0`rv?6w_?llx5Zk?0QG)*?WyWk*v#iSPJk;z5|RMg}%J zEhA<1@u=Cn3AgFTvAkM2Xa$`O88A1Z1ShsFz{M^+IrO^M6R2rrDE)n|2}=B@ux#_t2MaO4qe4*CD+ldC>$O3y2EiZ zDGa>63vY;C#V0B6ainoKi}MhlMJ>Dyj$MghYhl3RCVAHx8|ECvTGeA{4JoKmP=Zza zuH#a2AF~XfW5kyaI9pH^H@^M>5>;m)xxWE!R2&ECyCty1R+^sGZ^K=Wa&bwH93Heh z4dGj&ARncu`zKSX_`_b zcmxF!;~{lIKPlF+!15|poaxyKHa(W~bb2S|IOnis`Weq->G8EWiIT;2)K9U8#xFUA zvt@&DYQ8IGs3xG$6G!}2tUXLOmyU;=i@o(wVWNT#3&PRl)dqTMX#|b9u$j&{(L>A> zM}y~!BQR?EV_0hW6|$T)@#)k|w!ASPquIN~OK*_e>4={sr0LdIS*T_^24ASpAEtwr zXB=2I3Sge;E>zy&fv=D9)^6fb|B21qcl!~Bt?NNshrhTjs{;!xPvDz{ygl!qJ%?eg zj&6_#+=@V~%GQ@R&P$G~|Lq3eg#-I1Wd~pv6j`6}x^CR)SVJxO^3jl$n61IK6UTK6$o1!6& z@%21y!SnkZ!mV)33cQ+077@M~fsIVV8DR%|y2Ywv_})mLG|SVphmS_vhEOzdq{ zOY~VPfE?XUXmVUR-0ovPaZtO6x4-;17sxMiB}*p1gTv+CSkI|NYY}f;5M78y;T3FI zc7bO}T>_W$_eC10Y+VFtjPK9*;ye=6qg%Mv8u!oz>=9{qD zjq!;yfysWJa14Ctn%h})=Jf)q_0kmIoUlS={Z$z6T!*(t9LJRpCg285=4?W}^Q6+M zf;p3o*|#y{)DVu`onG{IojF}6Z$%U1LWn*;kSaB8_FrF316kfC%RZkPA;RA6zOIFn z>z{$(@;~gm$DD0A%Krow%U*^D#dE;i{RjCvLy@L!SE1hfm1x=>ZMv$#l7{O}q5gMV zaZ0`_#+$u_=nv&Mf6GUdzw#U%KPzE_dIF}VXQMX17?t(@j-MSeaH3K(iF5Je=pK&X ztV>VFDNo(;&U<5&*%yJ5Uz2d!h;1w_hU&3-q&CUk?jR)hZ5KH;<}^rKS|Pui1GbuJ z;rY>r*mNSU)hy0W`~9&0jSMy3ybj+*WuxV>V%*4_ukrNcXK=dl1tbQQxb=&bxSh9u z(#<2bz)tZZme0daoQ9eUk3hUz2A_;uh-x}K-P8jo2eL+UGPL<<;%LkN$ZvLUiH0k= z4`Fu(2YZzyxLX<~bK6CxaZ|p%!fmH+qOpB2s)jD7pRVQ6ZjpLwouG!&X`y(S#NdV% z&e-~uk=u5}_p%4477fD3oe~`2)#iApgB*U zeyX*n@pBpZMx+s5o>&b1raK^CjPboYuYno2wb*u$YITP9oeH?sQ;&qkrn7swsuyCY z?Uo76a!WwiO9YF?Fn$$F7URA8NZ>L)?s@&Q0M`@`w|C9wVsgf5kj#`fX7N?f;G^bs zi|NDc4)Aw*0J1)MAS~ryNf%G|R~n%D&GBW9y5cuO5&3p88ZTz=E!=;?j?I zr0EH99o*)hgWnT>p@1Hs$M-r2@r#5|S9{DndjQo<+u8C18pI*i^b7=>is3R>ij^^! zhVk;vj|0zo396y4MbmU;sfvXTRd>^&tqBTLC2u{8YbB_{_?<_wWZEw+Z?-Iv^ zu&J1PVI=(=FdfH8&SBx1Dau0l#WM0XhmXJ7#Ee>wSD|$S_t}5V`aQU7(d1#7m3uWm z3)_HC7g9!#g?;kjxaw~Kevm6cx@`uSBp-ot?QmT9aTobA;V`rWWg=Idq17}4Tx-MA z9WSl@fO``BP;zV*9UWFmL(h7XdTs-f!75xBS784c>* z57#~B?OQnWSOO)rUcuP0T=X&F`MTO?MzCqddFRr+IwNkQ0w3LbPKTbFtw`b5Pb}=2 zPPe8e&@UVKQA!0cLUt3*y(P!U8AWjKEpx_ZECo2B5oF-PR}wyAE=Oa|F;3a1z2Iie zi#gl;XEW&j9SdpIBk||?V`vvC#L!RG}^>iGvn@W z&KqtpJ6?o+c*W=)HPO>yFCOomL8bNVspIEabmq!JX6>+%h0zo?5v|R6IuYfoHz94i zD=yo18lT80l5r28k-CQKEWi2HRzrzeIJP%i0X!0h@#jatamt(lBH~JPvObeRA4724 ztb*faUc`cttI3)KIgkmS0)KikIBTOVS$viRkAbf3J4jEgGf5m3%Z>*D->QfLw~m$j zgd6y>`wFM!D~Yv*Dn9fc$x+zYP4u$Oaf;SkGD9+s>^8ken(pxY<%eh%EXdQrYDoq_ z^hX2IIVXiIKClnu=T@_0xlreOvMlupNttF&0$#*$lD3OLP|8-Q{uIO3YY2v{_D$$3 zi{)%70T}Jx0vu+q9cr}*4@~S~eXt@mpPpRm7Fo+^GVDJ9lo;DiCATN4R3}=vp7s{6G7Y0$tW&!8P9*u##WtR(lU62 z?1-973K#Lg95pRi^(PWuOc%tE%c{8Wniy{Ta-UPEl*HLQ@Q0<{LTd$>`+GSj#UcoP zF`{R?=xEFI1S^K|9s(oTRlbLQZ`v480PTRs?eEi2@3E&sAGHtdZd80kcW@AG1P z45dli`%2Q0Ia6Snyb@T3r;vzOF2eZu1}9(skMX)I2ApU`-sR_7OKm_R5i62&>M98e3$4V{!P0--#B@VO*b8 zj2(vg@4wjPx9opgJrU9+mQM7%M!=gZxu`p$02Atx@q-O-?J|@;^w1e{hw{dVfBA>* zTfdBC&Lx$H<%K*ycQ&mSFMi0-v$sVu=T90xi&yGJUGjR}VqJJdv}HkpVAl&&EwNB`pd2wm;xglM-Kc(d4z# zbKnzQ0&j~)5zdYPFj%cm?l?sd!wN0x7B-rui?^Y`+(ZsNFHF`lvVxe2#?(N}l-9M# zQf=yvpAVIieUCy+j`4kE+eZFlFe>|M0B15!_BmAVbWK9GpKs@giY=v=pQKSyn{t{^ z8_CjP1ZXZ@JlnXUHs~3D1@}HecMfT}HY|UcM(d9f@tBtH_4 z`YE=59hS@!dBWJ`J^h@8uPfqNnCG{uLtN-4cxl`#pu56*SW6ioco;XT-Wnr#m{BLpEj$B4G;Y2 zc+bjC4d)^^yrJH1($v(Y3n%w38KM;ruu!fGtCD=JRuGld-`)VFUJs);WaXv`UQk{i`p$>-RJA*W8OxE<iUuF(N1|b)>{+V_UDD?@(iyC-}#XRBz4aiL!QmY#lHo3SKfn}y|HGFYmJ%L zG9b2O7Noar0cRYJG(MI5wN0ZDZ| zPuT&3kOy!6j5C_m=?5`x-6EFe@Q6<5pQn`Ir{v`)%YGhLW9JS3bktcrdlX%AizT3@WNSUtKyw?9a1v+cCC ze_pyQ7w)YJStIJxRwL+_e$YF5959sbytS!WK05cg!qJl0y>+G$^lj@ib}AILhU5 zBW=q2p`fc|&tWaslZKqurLd=2Y{}znc4^&b_HhFrxf&>q*-j&bI-lGx#*ArWp%6Ts zPHD{K?%s z6xue8O}DOAsucYh*N4kWlE|o?j*U)b+}~4{wDPw+SKdvMTcZ7c-Rf3OF{cNn{&Gjo z?*VN0^h4B5`w7`Lbi>xvT8>Q3XyjE_Zb{HMvhLQCrtvMF-Qkh8$Ov%#=T^+u((#2>SrcJfmbc;&p_#mIGl z5rsT=6RvZnZ=tZCu|l7E_8P;+IZk11FK4i6P0h#|BhRwkB(Zz8V zdKwMDvJ6YuUiko@<6E)A&If_xiiPrh<{cpQ?h}}UXak!$*qp_8OQE;k{e*i?ql#$7 z96bcy+<_rO0$|d|7U^Fcv9#aL;q=Q$kO+K(-|qQoxcXtKmBp_qAPtuS5( zPwQFwk}E52&(Kwh&9oBurUtFersU%6r~SCqEK9X&5-f3V zyB&1NOd9)v+t6)K{_>@>@2PS~G)a(*Ah*$z#e4 zJE?%8gzsp&{TQ9!WkSmSAo?Elz_hiEf^$G69O?W$#o9#1&OY|eUGiCtf zWdh-F_%X(HbBEQ8(@<7f#)`tNF{Wk@G=0Be-TW*ZFS&qT`7bfX;wIGg<)Z(ePDp)b zgkQy$uzwl@?p~=dp5l6(8vH8#N{7yC&WggM;>*JI0q@J$JK+a39%!&U`xKGl>R6F8 zk6|X2ZHBBv<1trmH;!K01&1j_h}sZoYIx05ktDt~rYbOc!|c|8|xriMww zxowm8iW(g?#aF_$#Jjq75r2%_!;L$bEAp065IL{AMopjf@WfyQu6_2$nsePzu46zR ze+KjO-WHru(}(sz9`}LQ$M%nxF7E>ymkrb8y-*sKDpp7=5?dw|i*;&N33?>$pTueW zJI|)Pml98n(h(b&Xro7N9~cdA#0IUI5HCxCLd{Jax725Le#@Aje+YY%7E0P1tmy9d z`Q-G5&r7%5aPZKyd)B>_Y|%~H5vTTRqtn;1*w<(PZGP>)x+Iqz-L});ZqB4SdO0^Y zM_u5lXwN||-(C_|XWeRmJL?7_VD|#l@VFcL4bySMeJ!fQgIUOt45l^hE=#PF#e;?> zp>L>6!|$!LU^4qMJq%P3>g!QzL+=b2v;HuiMMm{u_pRR0x3lR&`HZZQXz(q;npyqW zGSe>XM!gnGUH=M+Ng5sN`=yE^j32|>**B|kH9BTYKFx}qI8=k>{JLkmmM-ttS%$NN zR0O`Y^EM~@FjpP$lgl~UpMJeQ#%eY_X46BTv4(reXd7|_F1OF)z_FLyL+fg;nLjt} zJ4J(ZCQEh5ro9g4rUZrRxvbTKV-759p9^_asGk(Ty}{Q9_1S2wcf zUvr_Vdlk(u_E6UkgQ@z%CYmlC#{z2(vhTMyuq$f1qTiYsLVxnR`e5V495naUVT*4s zWv?n$vP|;`#ub&~`q{+@9_)Z68V;C#KMV~HSxB1yi{9?kK(q94bn&0UKG(RinIi`< zdOcH$()Nk=-M%TR>>mKnDZxnYB8B@+i|{rs2*-anQ9sYOD43i8M}r}7csh>el&T1Q zT+qjc{od=$oZk0i1HX9EPZuTO|CKS4bM^NAug%G)9**)jyTJv6eTU;#jV#~K-IO&W zpRGP!&HA2@5g+99T+VgRHFW%xZ>?879V=Iy!Ri7FVeajEVFLb+^Fzq+0*VNpMQu+; z(C9a|G(>SN?d+yRPQ67i=AlG?Z5oF07plZVPL+#0E0+lPKQY|`a5NwiQ-iLsUO%M8 z7vyZjius=SHhLGHZ;<$t`=0KNkum8w&&PtM87i`sC0cB_Q7JB26+vy{6)4VW>d+tY zlba;vV2?Zc!ASVAo@#fdQGAdR`tKWqJ`I@|NbfOn*?qdJEb#~aZCA==of^k|^>X2~ zhPPl#aCbJ@;SH8@;h1z|CdyS8V$P}ktmf@rHuTkIwsC$CZA>^$Z6EZpe)J7eH-AhA z?svj^aVT|HlH@r0C|7~!wTJni5af{jbqgh* z&86e(?5X=?14?&}qnac8Nvv|56zmeAW@8H**GrU`kb{!zYhe5A6$O2`44+G1P(H~5 z{%%#&CrB5)?UZ2qUwi_ms+K7+Z!H73U)I0`A6eulKE_Z0tm|5ZAUR#cCg;%J^c%EE97!)$%;1jimh9)+ z8aT8)vO>10WIfYPXH%Ix##uzbJ138~&FHhTT3w;<+wV@rDyi#8=ZA zh50mP^I>+c#D}RiHt}l~2kbc8OtBT!q$qxdE^qf@@5^9(xV;BE20B63SK_1F-cI}H zp~7rtt2m4$eOF*p7jA;V!_oL)UQc!t!!d++NABx=l`hOJrYVzjsF~m2=}weI?Vzq` zo|jJzg%`yBI(x+Z`+JFfFF-S7(05Yb$j6(nJ;3fwwqwJWy}>JGEyQ1c&ExPyQT5Ui znm(1E5BYWj$JB#%SYOQX<}*u!ng{EPfA7#E7*DIIu#AoZb2L&j07hq5DG`VI!# zh2m1wP<+eG#Jo}0P+8W@W1;M#;?ezRuFG=z_ck2HiaW4g#Gj!^XTg1+E2`WTVKluP zC1hA}(H08?J-TLOggMt&A#C~z=q1jD@2aWTb)c2rsY(1j+uNVA_W&oQa-3`PJ4jUhIlO{OiLJPuIu zNh}<;5RQlS5MJj=XGA;}`;B-CDP4#UtwH#AwHbd#*s}yiY`o%1c6o3QR&Ys~Wg7m) z+Eu~0#m^H%b4Oyv5+36(VjowuaKGs110Ck{$&O7-o4~xEL6E?{X@6ZzHGph9xOEWEWMDD*bRJ_-0zT` zet3T@ycR^hCId+SdIW})%He=|19x_O7<+bh4eNh&8|xG9NOiY&3-dx&Y%F?H1gfQq z5wK{hKueL15&dyx{8}oE)=jSE`lapVR(IDI%F8<3#_StO=oIaY%74se`9%wxCwKeV z{71h`;U|&L{;N#dyox>Zdc>B78=xfI5`X2?Agk2{!(=|uk8|1xI~m5#nQv#wALlZk zeyKbr-%j+}#K#(M7%a5ow^D(6m3UK_$9wLHb`^JGq&4ECwv*XEBkB|Lhcb6C7_SRK zWzBYaI)4zInOo0&5QWhrl|$UrTE;0Koyu;_vu59V@_sYvp7_d~c#KUkYBTeh-j)hh zCV%b!>^16mjF=$1R<0<2689`<4nMccL;1yL@~$ywPgUNazgaR?Zk1wFA1Sf8OnG+a z{VjOd_hx~whcks$a;!A*e1{A^Zn5RG)BFD)xn|FnXOTuFuAC7WNN%tq7^81n5 zk0Z#qZ6yua(}RY0K1N!T|IzG=ucu=5e=5uZxm17m-?1D;A{wg5{nr z?8y9;%*x7}Jzn6&Zr!NIT(1{6(W??QvF?yQza96V`eNB*IXYL(r0zp zqc{Bf-t`X`BAUlz_#P1Cd}{w3*bjDr{q3{VH7$dDH13mZiY$)GpQ2}t>nSv{3)U2V zB7N;UbUoJyF&cGrxjKfHjjJM6<>_QvKa)mSoI;EARdl_113kXKpn%Y zQ>{p>JHHbOgD7D&joDQvHhe!o}a(?d@+qvF5ye`F*sKzVBi-eEIcgnn{U_W z_yUPNH{j?8PWOupt#dBoj$Yv7j1+cJe)w5BGxe_6NBO$=?ygf}`N!ULv}HRvyJb>h z@>X{Bstb#1W$dTSVLEp-jvT)H?+p3Pl-Hc?B@);jZzj+MVG(X-){hTTe96U-@;dx|jdVs#X zm`i6%SJFRPMFSi6qI~8~oQ~Lp_!DDrb-NX0^RLiZD@Epe^ajjh*1<+*5Uoj%=iUW& zqOx#z(%GY9X>FS{`RSUwg`-HU*e zi-fYhQq-l~dbMzM=DWzJK7j_Vy+YCRKXLg=9$dpTDXxv*SNu0e1pR(M?+QYx=Ehih zw)Yk(zR9Ii1&Y`)?;O{1@UNg3?Pb$lgV9r!w`pH9#mDbo$R_s%EesJcqlGr?cor3s6eN;qyAo;Hd$&H;5_xm(=c}y)g#d;d`P0(a>R_iiy z=*yC~Irx94| zwh1ZB(R7xTH z_Vaq@g1++VnCE_p{FeB1@N>_6101!kpw$hAP-?S>Z2C-SYVF0frCX_Kt_j^KRHs+Z z9p0{#A)}snTOiYOBD5VKI#(9vf+ zu6kz;K_;FqJw>Xi=D7XZ5Y7hO;TZ8mOy8@-1@5KdrV(1uzuX7)irw%o<`TPPag({c zyTY1p+M<{DJ95|`McsL<9ewXem|QEs3i$;5uuH+g*255Oc0$8p1!x6J=BCK1Y;Mhp z8g6?{C;W|6#N5%3iR){OUm03ZJk+0$d&@`f^G9&;T>!-26gv1|6?%}0W<2H2%|67n zZu@|iy9ePtW;ph3``MvWQ)dPu;E}GNv&EOraN~2{aS?Wz9rxW=s&lTq{ia$^a`vHg zGz>mGK61N0dLH$~8u51gZCQ(bilq?c-a^KaT-G%AZ`9YKbig90Ys6vTULVkci3sFzsAv1-Q{7=n?3=%*3f|T|pfi6vqgu`s z&Lhf5(eVb!&ngn#Z0#cGifptVi_X_#&7FQ@^UHJ=_Bxi0yuFqki201zg-7wybOTgx ziI950635MXoWkx;xm*7IxpV#daNIb37V)VXH-5Pad}JPbIM;7H2I$r`kbC2ZWSWb* zoBy-1bmDzc|C=kq?;gjT@+9HfVA&JQ3`>G+uta{5f0SwNIb~|*_m3$NSy;bu2Nc%w zIbi+fP_5n)T7K>zOh3oM{8A#Ew_JlsVgjt{V|fhGu@rPpjmN>(r5=;tz=B^R^mCr9 z>?^uPa+g=r)Nvzdew{o0_<5I`T(q6rqb#D}lX4U=kI$vkvV!98Cki~zQHd0<;FHCE zn5iocmcGYb@_!`!Z!;s1^%!8oPROXS@2-72^mBMWC;F%>kyqQHW@=E*pxx7W|Codx zY+;BplX>+HbRZq&7GM(~lf*ni&;^Sh?f4d>PL{mvQqPTkM1 zA$CFE_%hx)Z%0d%8Xn1oagCx6qPDbqBGsHF2%ksrm9r6KCaqi}=+AeDD!LIKNk4S# z=+Il<&eTUACN2#EZxX(Z6TY7q^XB&~$vd}Q{*C6HP%rL;dlBU*{(TH@jh3K~{X!P% zzlJrKZDmgU`J{AZFgZ-vOSgj-u*K=xNE_V;F}pX= zC_YDqxaUebKU>74b<1&o_*^)Aj7Re7Sd3f04sRkJQSPF}^kYOFr<637w4d~(z;CIf z_izfF-5qel%pcEM9%IqS&g_D_H7i)0fIF(6@%G(d7ALKYRpN(an^#Q-hwVTPXNwu!cP1g#GXmzal_ttipJ%>7Tp^Cl->+i$DNgDpe`MOQB46*(LDmgw^v}H zCBet;k(j18Fk$D~CqnqDDirUzNwUmfi!g(<(8@LZcVpm{U z7i&}+Z=$SdGh})cVZ4)yxT#e}ta0cgo0@$RM`5mg)J{FrOcMjL@ewf^6UP+Cf^+TS`(g7ktmfbQo4v?) za6DaTQN{8ep#ra*2DWr+Z(~{PV#>YL4{f&mezQw1_KbRpf~^msVyeI{yA5IbnPX7= zax`Mh+bC-6ag+qaqJdn{@I9GUe1Avv%NC=a^=Bcs%~;J81J=jpI6ggog$B1D$ZFaL zzn)uhIdB0sn%;+T@+*{#NkG=)(;^G8RY-Vc$!8v>?dj7*YVE6OQ_9xD9G-? z?GqGRVucb}Gng#sfi95+U~finAs5xS!3%m*Q`tdg{&g=O$FYfdK5)Z~t0VAcM+A~) z#A9{l13Z{7$4XA!pv{Jk_%+6-LmyJl4u#+F94PiSz_w+ZQE|Q>J`P*}?F)}uXx=%DZf{>=T8t(O&1k}p?0Wp(SdQ$wvTW~oGq(SO-T&DivBk11?pI%Sclapw zSZOErUwnve1%24I1QiU_0LJt2=7B1^us9+O@gpSutpCm7VcaiFIR6^MmX<+2sfC+g zV$QX;ZW82NVrES7E={!Xi?txz_GfGpr=c-h3;Q!}k&MD~QB&PVq3pwDygojX#!1)T z)bre+j`ANf<*?9lI5Osn_;?%0|s^JHVVOe>*Z!`6#;2P_dE6g!LGzpJ64y(+1{TOMUJtBGr>I*XM)H8cCw<#f@w zmbklC`1r#g=`WAto97MaTx~#1@(5ft z7K2<4jqTzb%l}FccI=_Qx@@gu>Hy7{sk^8<~n!p6wwa@Tl&{|mT-USvA(RR z&4>-pVXVA*TYwQ z1fHJ`qcAaxjHeo*#x}fz&Ou&kl<`0sbi0}~te?`Vgfqyxy$i#3&qTADBlHXwV(`_~ z9p&5mC6_5?(OoUrE{{I!;ExdY^2HEge5f6=L;4s6w0cZJ!;!^!Y_6T@>SE$Bj^rh3k{;P+T}0p?;O(x5x9u{XQoN_WzWhS=_JY z3*3Oh?qav$yZuveDT1Cht5V*Xr(B;~=jqA^#3|0xZY1aVRU{u`4>vi+y1M-I{~vht81E8Ogbqo9cMjzlcIXXLKaK*ZRIZ= z`nU8G^-x<%y>4x!6SJ}eTI;NR`8;l&Q8IclpGU2QPG?NP`UExV&W>?D?^3z*|1!bn zBS5-_ABvNeaOlwtoLWB{-LC1;xowxpIh~I$oxDw`tG(^9I%$}3@eg8>yRq>ax{y9) zjr>JMJVxhA3i|Sh3gwmyIx$?7Me*|I$lm=fzO2eZ)E^6YItAdztYX|8pw5&vmC&km zk(6r}(YO7DvTytXLM<^%l%oqsS6Ro+76pWw06jux8~jjM%0m)cwI(3J(k> zgZ{2Y;`g&`y+e1gajl!U+zy!U6^5=u)3A8dUSwRo4etg+mbhMn>ysbHU3zUuj^8(O zBX8Stig_+vWX45q0^@P`{)W<_UzsF*a4be1D8pu%OC58#%d85nE=!HFebPzIgU<(2 zmPlXHZjz0qCF*z?>8`fNYW0((v?ZQ=>xWW&)DC*RBAj-aPo>^HO`tR7QU~wcVy|(& z>A9qzTSHq6?2&Dgiff%0(|X5jT2~;!TTa*$KutBra%2~2OqT4N zz=!r{{`+KS{eg_{MR+XW_q3(cXtayO?# z%ytb$UicyIUR9=OPN-yVXs`3ch>Nh)RUX*>+iBZ*L_a>x9o*|Bw7cQwGDyFYyff_r zW|B@~Hno&H;OPAVfd^OYt7){RB#yfM`2c%s(h_&4h%gE7{-ZgxC>XlWUZKXym36mS zB0L+jN(%`&IuxrHCA77pyTsS&jKUa7Tlv3S&Vd^uxMRlyxOo=`QIv^;@T}{G<&^O0 zI{W(|j8#mI66UGAomEt7zZcinyhY=Qlknzww0BiJ<=x&z*LLJlj;0)CYnF1#K5}GL z!skg_a)C>@p2Qg(Rii#Kpc>^{})E9ElQ|r=2kkFUPTwHT|oVGHJt>ji zep^Sc>*oqQ`!L#)mOgN&c`aSZVE=nAcVZRoUb0?z_CMaI+)bjqp<}q(E33I*8>@xq zjy)R6-AIt+!CLq;0M!$%*iN^x!gDS*6DY!I24*Pf3vJ6;)N`-hdr`u2En2ps8cH{^ zaPs#ybom;>?YEi5rMLxhy^iKUsaOyGTbk);;uDge6AkA~U8X&GNCz!0zdvv)EtBa^ z%WO)i-$)Aw*iro)C4A@c^8Tas{0E62VNrk#&A*h$)t?^3-MgtpY5}WQpt`Fd=gnu! zn3sDHd*P?cT;4vw_{ot7$?OTABtH0VY7oK#Elo|d%pJ@t^1ZO{62M2hqTu6=lHx_h8}r>d&~C2&~9$WSZ@FS;OfO4 zG?!b-Gquc9*eV{2`t&W>A2l8JxqZ?9y+KFc>xUjWIyc-C=dr6dW&bmDZn-=4?#qxL{GR3o#Zlzv z2{1boCD8XSgU1cmcu3JcuEKR7kD)trWwy}Yhs$Mzem?4{gCkRK({bfsGP|vg!a1Qr z*$m&_Q28gxZ z-y={)WV%&})R?wqAI+Lk4zgy>P@RPpE&~=18oZ&7je( zksjZDN&!V#WTtkWdX^jE^V?T+HgFpGjZm_Fd@Eew+v%`ivf9Ft*<2N>vwO+)+I@mE z@IA%JR0nbieI;j$|LI`+w`alzW|}*IJ?!xv7Z%G?m7hf3$8C5lXX6s|Y`BQfuoI-= zHJ1LBJ>Z7i2w*TD%H;h!3-Y}gEm_kITGf?hP5X!fwMuMQrpDX9QfQX8G+C77I#xj`mdkBmD zyO6zD>c_W$VT$VEfoIO8XAg<9yT@eb!Is-UBth1@-?Z9gT!Z@lN*%4M-V+=Gd=bl2CiqwfAn|Ff-q?omxwE)IgL zc0X2Rqrqx8MRsF*1#Wm&Kvum4%*PZvXGru)%Jdse^^k|5(;#rmXVAB47Niu}gRJdq z#bYx{#nzWkigVAK@cBe%l7H7^S}Lc&>^v11isjgg>5Hk)U^A-!C&7)4Fq=RpOor2W z>s+L1pGLY<5$bn)GT${b*mundg6uZr@z`BgFY;Kz_eFoMR&f=vdycmrGQzFuhSKb^aBv6GX|3P>+Y65~MF4rp6CjX?kW0DL#3JUivTa zz_=cY&pWe%H}b6CKW(<-Qx^H|>c;!*pTU*^JU+JbSPC=?qNa@=Y~8wP?EX&?3+eS3 zM@HX)eNhS2V?6Pf-|My0eRQG%nn!A510UmjpqXvrA^EVFTf)9@3tW@9u)T1c=GgSA?r zHNPi2Rj194|J=*PCE9W)Tno8J%UxN@xG<(_xRR~$cfz(#19;ySXW8_;JXDw-#KUeU z@w{#n{JxIHpnDd4zP%x|HKvps5|Y9hd_Tr!-Mh&mKb>bT$CuDRI>31=UlXZLo`X&& zN)UIbA3L2Q#rppK2z!rQco)saztJ`*+cyAzUR6=!YRUQSx=DNRzDo(t`dz{9p>g7A zosWpKvrmf+f5)L>b|lo&f-w4@3S0AU8cY9K#FX``DNyu^294p@y}UexPuU4BuyYTX zw*?8!J8TZM{s*RWvpt-8wT-nnZk@?lljAd{@^?5!ud#!r@gNN1bA6oYtOrLg4Kz0Np|?kl2y$HD?9A%knXnAC zGiY^bK+&N)@Tn_?{+%q!oIC*Cb{C^jwVCGUz9ucH5;|A7k4yc#L!>C%ogCLm!!+?a z`PUeerdW+xn5eUNk&Wma{txAas%+`=_n7W>4~w(10l5Nl>={K`a;K@`<$2ovNt>NL zJe#c?xPy&oEr)BM2J_AA%lr+`QdEEh(|F0|c{Dc1hdlLe)8S)Fxoo>l-1G(_NBM6U z7@)y^9}NqrcbIHV9_wbL&;Bm06lB}V zW0YU>I!q-holxRwNn?yJ3q0ID^&{^SYR5hX1`2b|iKAVu7l(Bbm9C4W4-58SaZDNx zoZ|DoYG$I~#2Xw|v0z^ww&A-?96X;rr^-1o0_{)jzu@6q9#fr}vch6xaGe|w6LpnV zuL#E-|F2|XI~KV^z6)*ctK!d#|G&Pid*@9vio)5B@)9=bk&JnCD8p7%lV}-t@{@lTpS{86CL_4|GSFh;kL-&rcI_ZuE8-BS9ySw@YJfA;A$dzQOuEN~!Z^rT+7li&-ja^5hUz70rSC=$SCAo-uoNPx1=MUihlIc*o zV}eQLQtYTs3?_w3F#FqiyLRF{^u%aZegMp9&^OFY=*o^w`pTcv zA4JXLPSCvnn>Ib;ohu!Oz~#hZp{>KeHQBc_t?){?iz>GW1kBN4z4>#pA|GuG{F^~z zGDG+n9Y+j{8H4>Khuky=HvN+od#Gs3Oc%vtX?!%?qu!$LyStdWe2CD0>A7D;+lQ~# zZl|xS@!`S6`dZe0IbPi5r(5uAtTX;BvKPv!A5h{<#ToQ;!Dmi9Hk3@@mM@}k3p}qu~?t97)~BDggWP) z=+F7hj;9l&xQs$4&{T#Us zJCx~bLMDxvy%=eO!r?I{U8qw#sJkBLNO{3!%baA6T z6i0a+iuOLwU$&d3%-0fr+cw^!)~tH!@2)^?(rSFH#B8K}RAV_aqS=f!uh=;yFLb$- zfRzR*Lit4XJbW8t$7)Y$(c)#*h2<;!SU`V%2+>5s6YB)JY~$(LhWP8+@KJ%ozC|5 zpi@WP$i-s|dD!`pq5FF3n_5m~WhR22KY6#EXreTwi!YMtZUSNqIhAQIXvYBuyy}@x z&y##n4--VADimnhsG*!? z_9zPVZ>9mU>*>uvi7oW(_F5`G+Z&G_^+bP+!LA5N=w(2iz}Z$)4CR_AQ0AISF;XlS+U^ ziv`oV-%9S(4Kpu&C8MwPbT;V^8R_ie;|%r+JkJUkK}V%3FlowjR1CT-kzI0(K+Cx) z&xP;qs+z+8xpQ_6*v6j@R#azWTy^;XHM8mv}Dou9(L~8cVyYe7S_y zTK2}SlKD+MDZJNWYvO79<6x|R@fiP#YWTB|%XqIpgYG}Ma- z3M6yhc575OPOcN0vwD0@9l_PE3@Kg?Z|i@|1cOx(>q zbez#apWD%E+=Au&Sujs~?Vn;UDVE0w-RB{+^=al1awJ~X=0JMta42o>#O4fphajuH zIFqDF-v=M#q&&Zh7Dh~?n+nH7dwY!G5~oP=XRh{-qZyUQu`{ZSYGZTA#?TTkYNW70 zp^9R7TgwwKH~dZLi5L0H$&b4wx;RZ$kf&eK1@67}fe!j+?JeZW`bg$!r{)P{Vj>wI z?ew;Pb}c+ay({aHwFUyswtshW0oPOLWm_D%AK>{jP8wSs{RR1`@t73$CQs?o_EB(b z+AYc-znv=%m+<1Ww}((>zFY`J+D4Ga)^RYAdcZCC@JOJg@X-NorNTqbWx#W;g~#dE zkv;{@q_cR@JfCix$V0wFVn>ua?#1=n`ko6J<|VSt3ljRImY*Wj(N9-`H~OsR3}kBc z*p|u9kYv}BO*yR2ZnR9H;Y%g)T{ldlklK9&V%N%HM)5V|tKUGa++Cb1T&?0NZzj}P z;bq9@6p z?2an_J!?)Ip8VxhhuE;3@jP_9yE2QI%VQ@lH^Qt%$2$1$Qqc=T+(TKp#wZpX-G^<- zOsCdQNy2?q86JQ0lpgEh-^+?K=(PzYFnqY7pjIIFje`>9l3&cN%wlKW!O5pN88?Y^OGp_hOdKZHz9x z1;f;86xN-AuIe!MrpleERxD#418iVkoH>1on}pug6* zrTYroKBZ%vM=;7lW1(i5h)v_7vFleI4L+{|Y0dF?DHno>lpTB?=T+$XS%b^#Ud*l8 zxt&5s$dk-ndv1kNv#8+qSLhhN$F&OwaPdeM-mXc(-%kn9EAZ@?6O>ZLWODg1^?#F0 zonmrW^RWvo?)5o#KC1#3g10flzwIa zQa-9dd7Z0JzSw*KZk_3l;X$wHxVOZX-lz2*oYl|qIeGWt=92`>Mi}&OigCWco@LnB zFtd4vY|h>{U8+x63G}te9>)n$GkRZs4XlLaFG;>F8EFm+1X{LeM$v;&h1|+#5uEm#B{;03i|b$hPz{bFk;ffw zpAQZc|DcCcPSYe4$@!)Am$&4nXbG)*$saph@Hb8 zSZ!yXllgZ{UOjH!OvTL+@pv@X59UV(;s_t7ezE6Vwq)lAyjQr&MVZ=l@O?}6Ff?vo zfTT~p7*IV4Yrp@a`oHCLOSBk&+dL8B+?^Ud&j@-svVo7`{L06ZY_dn`yFKFh{?XzU zu4}}b_aworn$JIQ){QpZ`NA~ker6?YY0Pf`zgy+|G&BE@&^K$lZAfmi!q0bM_;uMx zxYvHa->M<-pOB07ygzbKkKy>R>neJ48aR7%BwoBwzzen#`Ip0>x|qX1C3MO*7YT5cCZg|f9&mpBtE3VzVAxi5uw-{*{iX5nSz%0z+A-V^?d zAJRbQ3sxa}wgYOs4WG35Q}j3qn{-YHf|OT9bcNabx9o*O98Iby6H_P^_jW9#g} z(s_+}A%kJs(L!B%^-=U%bdBtHR|tG{*w~*`u1mqCb|lTL4CMy6EJMhxr8sHP8#}&9 z;`yX{Oc$PQ?|=IzdzDegbwlE4X4iQp%NsF^eLCPzv7wV_L}F*yW+C=hQW+o18%=6lL{-59uwv*f%H#3adeyE0g68qLBKRCE7Ad6L zwS`(Y%HUeK4YRN7!VGIpV%e8&uvSr{do`fqh1W4~L<`cS71{Z;WRk7>Ny*+4eUpFa z%@x-xP=;a=`{H|%#Wv=$^1q)&UCX1cn2?P#Um4Ds zbVJs}KG@^;gigBic$&Kw339#mWGznciH7G{>Os%1P;5T-u(*F!gxJ<#6nuIYQB%eb z?(nADtXtxJ_RajFFsHw{DtRYP>i3pRx<99;1Q{s&4MweLG!{IGg+_!eTVy_g{p&i8 zm2X*&V}tj>w4B!IU7@6*BcX4ah>$6gm>kQ|U%3`F3$|t32=+;S2vd$fgJ-^# zIA>PdK})-yu9=`Ao8%?Ig9w>+m)5V0!Pl*(tYWYQtJGc2J~{7ZvEe*l|Nr=H|8~!8 zzy}?B)~~TM+v8jhx3Ll&&Vg^%At5irIeKfGd`9n6)s2KiBX=amPYX}qq-x{GH23@gHZVZDu!zod52Vh;ZxDBb z&r|wE%vgW^4B=I8W;|^OtFIY_-9}cB{@Vvm`?U{WgA*}1{4#6<)#$^?&mwC-Z_aS28BQ?A5ro#$hsz~`gBEz#L_-j82qNTqH4DU}n^%j`0^XVl@yygCGjPelBMx1=ER zo}OCqHrcgRZ26Tc_VCVEQX2G-URTVcWT~r^m$8bP9z|2dft6fHbFd(vGdlfPOPo53 zXiil8P@)Ze9~GGOnv8P?sxV~$$3A-pGI=*eOv!mgn$u*le@49U+!SL=PEJLHoqaXo zHPZtBj>@sg1E1l+(kwn_8^@-&nzKA-MP_jE1x8*{VYN#?VWqYX`d#jVkJ2w`-ku7! z^^%m>HBVD~$3K`^ehg$58Im)LV^1aV7;f{bs9x2gW1ROMrG?Q0ir{Z6CH`D0XH)zR9xgqa^2LZ>DCG)T3z_Cbxh&O?6>LX*_;Y zkAT}U=2lMT@t)s2ozxHw$`TcH! z1@DXADOPlKQBQmMUVSpzn8~s1q^T3zx`dDOPA;eW;&0rk zAT#=Hr69~H%htv*Zj3WCj8+w%oiY7@^~tC1gSoMTng4owHlVLPvx@4(e7e-cL1=M48B9%3jvh*(i^)^#NqqQZbRo?Cv`A4~mzb#lS;6 z{)6Q^jL>`q!@b2AWul8@c;M!ZX?RxdjBk}C^tYgp%UYo=a!^RZk;5S{R*8Uu<_Q$M z%14D$E}qS+=-`X_O)05C->z_rx4z_5d~dO&q;t%8U?jVvcojXDy@Q)UFIG`M3rEZ% z@!25>OEVqNJ$VS^jydD-Hft6>XBcxy9>7Aw#RC6sMkZ1oKc;d_x?*4PGg^_WjOm6} z@H;jIR}%8sjgMK(@83?g@ToG3F;V<~T)lZbRo@pkZjO*pq0A{2rE<@D?<)~0DUy^1 zi3XLVc`k)QiOgezA&ryRw-pU+cX?uP>T+SB>lT;6B}2 z`jplO_+Vt}4EVM7#e_R1_+T^}#l5Yd8!;X!uS;OEayKHDtbxt-`&b>j8^(Vo!?uyR0cr%_yS6pDFgjmW8m0+nH`EvW3T3p734BA zI>a4ckxRLL!A#@Y0_NCt9D5(<1GV-$6g@+N9s0`ii07+G zCc`%Wm1T;JmP~vwAESLk@p;sHx;4FyetC6AbKX6=Q>l*L4$V{^KL_R)h9LNgI=bsf zFe!!)r{m{ui4SRZGN0E)`95mKY!>?A-EPU;?z3+cnFL1QLEcEz+|xp?#|#Shz00Y5 zxyLE!jmHtU{@B?hEj-gPE;cBvAm@ZlfgEIgNE?{Y^JY4?Qv!$4NKEi%?9_v)*{Md4bqftu;yTYH2+l)WRDnDDSD!1$*qpMOHJDr5T{ot+Grdq z{5H%roa8)%s4=h;+BR5Y^E?R-ZftR;FeXZ0_@Zl<#c+ygpmPVZC`xMwm0u}B%bnMl zaG)6@CfTBGf+Lo#=m+D-+qB&wghWwwIyv7Kut^K~eRirEv#Jdi<|t>bMAk}jSiH{* z?(+HAmeU_uTb=QzWe5yHifHnTFLd^)CVJ*Cp{JMDP@u&hF6)>VO0gK_i^m~!;zaQu zoh@PqizM-yMHArnT@h(F9LX~=jCG&0gT0j6#fIGpC1bZ!bSC8!-HrcEsY8Cyah?mg zTW<*utlSfO98!?SEgnk~|jioW)>S^uAk5u|~Jo+dH2s*ZL zl0|3_Nle{;Yn9R1nP`~qi?EfRd>sB1F1B_*r;%$-OZwb_*RGm>WA^TGHBwC~CZ}%( zD4k+~VLIk0i#{vpvsFPCm)qBvc9gjSQKAbYzfjxskK|N8kA@GarkX{Xh<|1!lyz&} z6Iv254n2)lp-0hY;raahGhF}dUKFc3pWgVYu;2hQW;)D67}HiGf6$#jvxR5Zyjn&x zH^fs=lO1iU-$^s8C3xdmN)p^%V0JRb4cLZfGIMZB>n=Bf&nxWR?<##6<3yFFI=lm6 z44K-w!u|CE?AhmrVOzh_Y_Gp$-rP#0o5r65K4QxiCAPF>x+qviQP}@n`Q46VH=iMT zSrT^7x{K+}VYoeDBixSOp<#SHdREPJikNW`KVGM!Th;)48r@Dmc~0UEUFi3F`|n&+ zX4;E7B}uSG@1Bf9{|IIDdGwH$@fdxQRijT!I%3Ysxci+$I2E6_qURZ5$jvi=Si6$+ zRD!snd42ztb$h{hlDjsIx@u>@?tK6*b$&_v>|@Y!)E!Av9+CHr!)OX$gKAN4=gQrh}Sxj9 zDA#DpNogjXJ`#I$yiz6=bqyo6BlX<#j0U>>^M+7XkMu#v9r23p`8(2x2uYL+S&q)RD;%e4* z7d3Q#$eG^1#l1?Y zWpQ*MQ{&{>l@%qp7(Imiuc%S@kbc~#DofUyqRz4xKS6$i!#{myP2Ga_q$Gq655uIu z?v!G(k-M1EB6>Gw3rgfy!OPeW=9A<};m%IZblY>0p^_EVkL^#p8`g07Dn01fR|WFk zwwGJ?(16X|?7=Mgb~@K_>fw5V51?+YnNuqZG#aK zw}re%yr*4Pefah!vQu`s>{74QbRzn$sEvAZ4LmOR25*P&Yd;OGjX_9ASb%XMW8rzw zkInWAV)Zso%y;u`L6)!sH|YC#-p13Nv5_&O+1#z+aQ1ltF5)+A(oL9Lu`08ykz!q1 z>M(J06n543!%Us)oXGJob?0OJon>vr+m2|7ZTwn<_2l8P1+-?4J?_6cO!Yl)(E7=~ zF!no*Lsi*GTUmga?m;+mqAwJGy`!ypyXd3HpDa~AlVVpB=3XkpM)t^n*&B1K`_-0K zjX#H2S#B?-jT#f_N!nEUcGpa-{LVz&^RIzeW^prVe_Ks=7QdzyHP^6%U4Q|fpZJ~s zB(`at5&*m zAnacs4=vQ7hsT~!XReE}@@+BQIjTenRXqOd@ivl+4WN|~Av8HTlOClq6of}Jy$K6gOU5)7X0OAt z%B)#Uvl&~pG9M?Z>(o^kHSbx=h@&kmPqiq@YLsP|1A2^**{p z--H?7c`4|Avmko>(4%xdoX9e z0AXzZ*UrQJ&SUJ@D~Q_sTd4PWZ6KUa*J4xnDD>qquAl2?LZxpliW;QY=3$DgxIvA% zW$7?8p1b+I-BmuO${wNZQ!$E*$CtHJamqFfpZwp_p71 z<1(>GH2`1tPsTeQ*C;bAi1}28vqk*A#qYT^mNc9c^pf)RGG};NgQoBQOKV3I(&y2Z zxV+;Wo~YeKua{l_pZ0giVcF(KigU)&twd>rrH({T9-|a2*2A9-1lDO4=1+W0OD9xN z=SEZH{^auuXdET;>CfofNFM+1K-fPd0|_HPP~rIj$me7HYU-x4p)#(FTj0dn_78$o z@jf`NOu;qZ3aDDBvFDzrVE-!t8!N*Eo%Kj~%r53tvtAo6GT$}^me>yu+GYvAZEd%q zaf{E1A0Z(oMlsk?CLtv-`CU6xf>C=!IV9= zuwW_gMl!E=n`pxNMtYUC5@#lGtnh#g`?04F`>>bifDV}jU(eCFTRRw&Kdq*sz*vfE zE~Of+zd}E(yid&dy(?3Wox$)_4-0dY5jSZ$B8G3lC)NF^GWvu0JwE>%``fDzQuMRU zB-i?vE)2fN1`oc%^hUjBSzYJy_$WUn|6>MYAzMl6+CXa$iI3%q z6!hY1mco_XtQBpj{Z2o=CT1=!abu}`dY) z)+!dmI_Kb_bR2VST*b!cd$FzYoteQy9tY%c9ixrfxGV2KvO$L^RBtuHqkW)J>xqJR zCyY}Xfs4-%pl4TUOt0BTGICKgrM3;jULL?i>tM)!a$$30G?+%nLs-jbP{#@VpT_a|=sXjNMm^xceWprRZ@L(>h#Kwv z=u1i;%INw8)^Y)Sej6)%;_Xwvzd2$Z?@yh-&KTDwIb)_p3@+ zWi0TcDwAl6%)WzrOX}f2{w0Pf9OmvU&g66|lX=X260JQW@g?mkli(zaQlE4CvtQF# z=}lDKHJipAOc1aA6d*QiWMbtqSsc1wOXtSkq8Hg;S*+n#cC7Z1uy(C0oq%x(7LevK z6}N-paPnpbLM~-tgGC9rp=V&PvX24}{V&JEe{d09cOAdE zHB7^C6G^P_w~a-#Ifg$U49TL4R%Qqtd!D5DeZkxsS=LYM0p@={hQ+%sQ`x8?VX>Ld0v!#e?<8!Oy zt#Fe&dRwB)!gVw6HnGH=xY4)GE8`P6WIcUDXwiQR&}w#QJ$mxFP~?`uqBbARve{xFOGsVB{7SYdAgXtO#_A( z7xKQ7U^HFTfZ``r^5=bH16@U|I4)EQ9J@ma>GXCfkkiqGp zLY>*6PIw-c&ST|TkjL{lb{T&_xPDG$2u2N+L70u2P|kQiRc`b>ZTj%@GKI{%P4Po( z1vyWb@i_zH^F`N!|iZGqNSB^KAXhwDB^B14D#CyoQd`~f0 z7;9CQJviT}D`+WYbOdNrh3cRv386?X-So8;IRr&)Mh zJOOuCr{b5I8fSZf$Kr8M>Db8&IC@sXKQlR!=TE9^7Uce~pE^G8*UaGkmxJ))2#+iI ze3BO3u%iN*jc6F62=VDwL2o~X&ZZA~PMqVwQ(X6=L}U&NMk4P=@wuS~&Bn6`XjNkG zw@u)4aVQJ-J7~+YrqL$Yvv>y?s$QT$#(b{LiZom*N+aol5Pb3bf(kAdP5V8uuqFkM zoZnKC?_vJzaE9FLpTPBVI2MmiMZ?SP$b6=Ri%Eg72=fujcFydN9znsR-jE^mRSr+& zn`Wy?MVFj#>5VxZZ8^s!NuT1b8dY)ueJ0SOUfp2%Xg_V(A&K#s;I~}tdStTLKVz-< zo_G@NayLVaWj4OK^kW}u4&p=8c9edyq$6$l+<>xE68U*Ob?8dH^LP5*y)WEqjIizU zMp_yp;XN3XM}qVXJm4HPPua`$$vy;UyG@7i7$D(y`pEYpV8#dr8GEvJLT|s ze&-bh(SXir6xw?d{G+>g4 zJGdlBxc9{_E4)12Z1j9diq#*H3bo*JdyF+5o@DS3!r4rkfg)#glYy3*_i> z{%@SxvO%~$dqyp$96JEx5ek^}WuM4u;7+cz@P8QEuzjDnwgvZ*sRBab`*REW@?5{M@lR>H*Bf@EP({4}A5SRKW>%EG$Wx=giqX?Gi%oO$cx(3#KQLH&8Y$p_y>eQZ=|!g7i2`rAZ_tfGZ4xSTo9_a>z61ecQI9LzDdm{-T$>;q!mUreVyR+{1%@(wTT|i znuJts4_32a8f)Lw2=T=hZcFMTa<{&N@DXyzC~YP)-3ki)$aC!Zu0gpd8b&FJ)O^B7 z$Xnf^lOZQssJopc{yJcZlF-g&R_957m}E_kIq@4=r zmnB@1@9-Ylu4VXPB-y9l$>4Bk(Q>?v^PrFS`f|~p!J>I~VHm7`lF|ZSsF3rA zdGZu$j~)-DkFj|8z7kzzk`Yj*3^EU-DbGT%sn!L{#=D@Y+!lQ|^uY;lI}|#)GL!K6 z%uZ(w>t7-b`L#W8vwi5R~hH2eBC>K)pH2DUpv ze2T$7+Y{NRpCYdIEIw<`0w0=;tDkJp_q-EMDk;J4_zg3bEPDeSPJH6TZ7KiH}W+82OVzyZs-mzqtznV;DlD z9}DAkJkPgwe6a+(GCSrR!cyefA{AUxAcKxwflhOSh`%|{c2@prTDRFu4q!KScOSZZks*BlhnknbYF5_gQ} zj=v?oD@mnI9*R6Zx*?v=C$y*$X;FPy3$&TzuWsm@_L9!^wWWRy$>?ME2TQjZGf%%0 zLSGs0swb@@z1Zk@b+)QplZk70v6Q?Otg6o(miVX|J@&kUNTnQ!BaC1!BiZLnnVf){ zRl#uQacar#liBWZA{OyjhSk4VOEOspMOXXkie`7V$K$=8^sN3k_uRb@GG~Y4?e^L9 zijUVD*rgoX^M0e4j}hCYI2bb;`r>s@e>T8rG7D(v&2GJUO}a*nTzBg_?)ws3c51g5 zYi*pzQj((J*ucMCmJNn`stY-M97s|@gJ}B711#@ZB>Oyf6r1|Yku=WC=5D3m7e#Ke zVmcWc+3>OP?D3CCw0_}vk$;TD!O$*9PS!_r7i(0sNMfNn^!W5oC0gdZ!HY9RkTts@ zp1tgz*e~yf*qS9_SCSp>Eb}3|W9DLqFp+p@vw`@0Q-35beNW0)Bk3w1m*-*?gU3Ur z;g;E2WQmu+uWBy>jeg=ygD#ukK87XTI8J7Ej&#sJl&Y5}LR%>cFXV?}`|G7hs|v$z z@)G*E_oFn}+wX%@!6gd5dzQY1X<+A)E;#C2MIM=V$!NShlKs{T<1W4I4$=<)!P&(; z$EbEFyZLH4JAd0vm}?*3MIpd}uV-pL;_DvZ_o!vyhTo<0dPk`G8lUg9%#Jx3{X*uG zH3)cpnUveRV4W=obF062=W`cl0&_V| z-o~FjYHG=I^*L(ac=g}ft=cJ#)Vsx@`St_hn&!Y|w9~6M*;L-(nhvgKN6=S=`c_ZQPN34QO`f{VuCb_}n8E#O>Zh z%^ydQ-hOQ=jgr{d{Fj~Nid}3)DP_srmBBmdTTdfubszDsZLSLogzwwm&EuZlc+Q3D zPvZ`Zj^nO}*pi@B-49#G%jgK4EHk6)oXaHt$jz16(P;^`||Jy)5mTm2|B z;T<3Ax16?pxGw0nVz(UQHOppiUCR8^hH4Mpt4TdQ24ID1u;q{?D>(Zv4V?FR8Tw;% zf;8qNV*i>{t}pULOZ>ZYZys#mmad7R?{g*l(XZo9(V|ibZtGmyaTMP5U~TDkxc_AX z&+)qtbB8$7oK8Ja?A;5+6~DM;*WPg4%d3JOr)PqAyS9l6~8h zp%u56(Knt4%l)`CW+bb@jwhy@&k{uHR?b|Vuc@G)jq#TBY+5Ak>5?wW_&$WIaEasI zFW3Zio)=Ij#Fpq{I>@gBJ3`YDtm1?1#p^KPOe9PT z*I;MlNO%tIipMED2a4_u^!3za@(+Kb;cggQZrv30w7Iv$9}}=sf+2bm{5p_x_MSk{Up3fi}6BUo}v7Wh@lsZzVtD|h! zqpMJN-b|<`E87xo=iPBnmgk2$+nL8+&%kQ`6Fk?42M*?i;OUd?ICJ`%NU)&ft zk__KYrZYdCY2Be#ges*A<;<<<0ax3(LLGJ%mMqH3o^|TmpJkhW;eN(6a#E3_@VI0c z#jUW%1gT<3k1`gX>u68O01O|h)cx^W>@!zo z!D`u9W@Q3f9(%hCcWLI4bXMthniae}!{(GFVi2F(v7VA> z>qYtnh>AXrdyBQ%K2KL>rW(pr8;3AOjeHt0FPvsII^ocO|M_4-3Vo3s;euT+pJ8Fx z8tznLH_{A#NpeBQvCiW&b?1FQvZr_(VF%S{X2E7YOCaKXQ^p`TwYe2-Hd{qI_b>=$@@ zX2)ly_ZaMtQbx<+&-8n(9?r_XpxA?IRE{%TcDl6KGE`k$H?On!VaZXhW|#u!oLgfV z|MWc~xyP_Sl!tHnE51V=^LUPfI1I`^ zj17-G*%ME5=28C>PS+H0cuN@Nc@vc!vf)hBKUq4Tc&n`VD-fMsQZbj0TmCkq8?L{9 zLKQE*(V_%zJU=1A2~WuyGtzY+wN-ND^-z|aFAnE6W{u>s_P;<-aUI60?Z3*q;rmOOXE zG)wFr&`f=|^u(Bzju2b(@g0SdSoPM_1k|p*h1g0RcKe72Q);(mv12~K>R1K&nHpnG zZXm9I?oYLvLA3drI?ti8hi1nI(<6&VbjLgs^(pZ<&Cl<9$E29f>tFcOy8*ureHJMV zFA(0%P6z&no%-IH(3+TmO*{E@jn~EXnBL^h;{bCOrBd#(cv=^jNAm`{(h~7ZD37Vb zh|ktI|7AC}U$`gSZ#(UfsO4xzu56l*@W4!74@5T z$4+Fw-sZ5H>c7m}Jd*Z_?~_|#cbLCiDymXV6J+!oc?Y%=e&UI|iqMAYfDtni_YvlDq%2 zq4zU;euX@_dhB_^SU3hu%D+sdoN2;hu}!Nj;fC!&Ek>#GiG&m&-}%%9FmG zgaf*x9gqCS|7%Z=a|z_0Hdc`HzczM!Mt|1$cTdOvd%TZg)QUEkJ(OTa%^yc{VlM}3 zSInV_ch-=5-yE_}Pl8+WJFxNQtgHVGN?2|T$J^sD`&%bc{pAIZUhV(N3|$SnWm}4; zs_PM_Ut9PFFlm(R^h^xK5T-?lG7D4)}${m^6RtR9CMeBQKCqg_CU zx5Aa(#ZMQ*R3Kd`$+dJJhF4LUBJeOhPCsf$tl`G+WHE@_SL)1!0apDKsI+M47 zmmTB+FZZF>GXwwWdv}j`uJdVmXuXhNW0nr1hFCeA(GU+i;kCoF=GL*M*9+Ey=N zzB3Y;RNYOsbk!~5k%_eCj4nE@Oefh}rDSzagqCE<*>qFnHo9OC`ftswO==YCYj__` zm3obwjrUcmTM&%oPDVWU4TCt;8h$UG@Yux-YlgaE;GfB;+Ixj&-W@?r9$#pX)qFbN zErC0EU?(aRz0gVO9BEu1!hPA4BTA7!%=OXuM0b+pP&{NHsssCT@_Rbd;_Q4f32CCf z$~i6t3OseT}-VSo}}kaolgz$TvcR&mp!t zM}uBODADFt6+TyTC0x(V=lxO=+i1tQ)>qIa&7OVOIpcd6S2h!x&Od~5#+Ps6lMO|_h*&gx$U40rcs<{b6KQ^&0f!A2|+GLg` z!{_Lgk!IER<%Bh}%w`%z9xWyZeSLOxZaJntmaH>3$4T&C^2-9KNbS66Si&mN+=lOj zB~gN2=4#Euj_a+okjG5-avz0-ZcCB1Fj$xiF{wu}fXhJf#s2^1QF1-;pz!Lg*bV1m$|y5aV)3Z9P_i{-GiNc~@b{@?4)^{uU9yEFCT-#1FdaecPn z?0~Ceb#N?o;d+a``>2R_%xz#pZ%>8kbX{!Z_hb9R^ilS?H<}72p?ax08!)shJ3p!i zo4R2qT01Sk$fntXe7)mm3ibUqli2gBU887Qb~$caJ;mjcEY!+oVDH!SNc1}n$Ie;^ z<8w&N`^CqaZ@fh#X3OGSX&&A2y2ImR%8+J#455o>!AjE=%I|*$yk=b`@mdcFu!gF8e~Qwh2M)%d@^PnJT2%&ui+PW)^Ztb zdT)W=N-nVCe|N9@4v%tUwq&XWvmXA42K`Zj?ZOEtty)RpE9>aMwmy80=uF{x{$z~m zA8+Vzk9tAw5pDM=hv%cvi>aZAafZ-8wE!b*20^VkL2Uc}q_}ENzW9ZXoNMHL0VRHI&9Ma6_ZwMtx`0z4u$c0Vm8J;9;gJs+ljE=Vf8*xtH|flP}Np zxeJ%&IJ|l7fW8kTHnYlS%SlCJ3ZF}@5$ZqQ;@!q`LYuzcl*EkeyJ*QBy7Qdds2W8R z=YGTDkAL`FYHIB2cAitH_c1DT=?qJsP@Ky;haM{*!oj;mbnx+%fA1=-v#rAS9t)(H zxvdI>m$K8nwa6NkgjN;voN4i!Y^2<`9I^p0wKMc_icJ{rbJ;t2Jo+~hpY z?&f?>?chqfI3q5~@}EvM-|oglp3~O6dICn`icm)0b`{vKJwtmAHIUh-VQ|_u7O&SW zK-c>ZNxOb7N_>j2Qp+1Fznw+u6*JcAn@q!Xt$AC(6yEn2fsDF!a5=vU3p>5Xi~~Qh ztmO%&uX;`=1KY_kMFGpsdctVKPio9~Pj`NWQSAP?kkdPkr!0q?6#1CbaUX|06}!kK zWDpOa@)YOSsv||9C7)<5b3#2%mLFc#)F3}>J z%{zqnqcwKH^wFh?yfgpO45xWC?X<*(s&)4O9ZYkf=5OizJ6B>KkW1Z1lYLF#XmUxY zgBx!Q=XrlA<->dWqQ=KXZaI!aLnDMXs`~Vy$a*DQDx6AIdPA^|$878^@dEeI8rFP1 z%4+T$`uT9IIC=-0v9c5UIOr|%WcOfik3=r?0?!W~n8U^D6mYDzH{ElRyd#ufhYIz| zr@GSFtNeaZRvQKU`DRqJ43-CkVrzT|_F1oi)!2AE?8W19uNPr#&_`?KJ2(8h z94)vc!6fOuF~^R#=V*PKtx)Ezq`qv%pdM`YtgdWHRsb!L>qUjxg>E?F}p-nVybBrw5&OBY8j(j%YXHsjD`teqqdR2RJZ$=OR41dADUS@MB{1v78SGEQ5ta}WC&;wFiRVJDu;lmp(KKyVXShEe z)orf5A=`FQh57dWj5CfrFVTu-@F`_5N85#NH_LNh|L*0c-;d!Ms~qXfm1iWcSwSQ8 zF45FuLFmmQVLfjk`E2$_N(m0hN29}MDIk&r7cJ;-Hi1{ zTaa1d!R#Yuv$|t0>`;s`g%9-;WNYoVg!fHNri?xcBy)c!J<>Wu=9i=}XRS8#DgTar zt(9=|%SKySCRVAP!=FAW@EU#s$t%y{r==dQu4|+am3B(Fc8r@*zeFf=Sy2N`RJW(j zwKupcl`pX1@h9{@e;F6HmcXKIH!fGL66T<%$uU7k>!P)U&*f^p$@kq%7PHii+2%U2 z(Ie_maq|%}x7~pIq$!B`>4EfwxmYxS=Y!|-gm$bUIq!@hl`rwV<{r<-NyBJ-y08qt zFo|q^6DWM~W4d5`mVPwyX92CZv}6ATblRAU5<$VoHlb-K*KIG<+TY7}I@ zH+uy8d0&dDc9mv@$HX+HP*Z4U$MfU9rgQE_>$x-@&%0DZ|RRzyGQVgYCyW#MIAKx|mu(Q`=r0_ZD&V4_K%~iFyZt)4jno`)-x6$lm zpFmd6^R#!=VRCFYHMvzHH*g-FLkiyKhqAA-UvSpw9Fj+N$H9dvu*{92;N8QR=9M8V zb8$D;b4?g(70Ll?HMZ+c4uAewj|sLr(QVjhC|FL#@5$c(*7X9B2w&Ehi4O+c7MJa- z6OTRm2BYs@!gAA-h-=7W)0{T3(82&Vd`|?v25!Pjg>cyH)xqnwTuShsO#vo*VUe1S zK{KncW286Sa^^vL-~VEAM~n~SW78*N`<;6fz9NBo`RlNCHZpANu|^)VwFaB>x8fUb zbMz^j0oPt5AiqU~sAnq?+H(rV%l5*L^(lBbH3CU&8D=D!;ntKvIM4gDt;Wwmh)xFF zx5}`7hDo^g)Sr(*C9Hj@z+PNEjDc0}kh4hk$GMi-*U+a|7uMxx7j|{xXHr)kN;mm)V0+n1S~|rNOONw@j!X}< zPuqe4r}J=6<38FQrP=SSCGYYT_GG zwurBpj=Grn*xFExeqZeQvz8d%Jm$OKjWVk9siNv(8m!C6cc{b>T$9;?yF6#eccbfA zsbPq#t)D4!$Rf&mxC>DarelP^Ek6HJVE39;S=%^WwszJQI(;gF_Bp#!{sen=-C-a* zce+1w3RR-n%Oo+X@3N%WDJzMe=|F=YlbQbldOlNexS}Vz{#L}G>{!V6FCyKKXJ{Lr z3#xm!!_?>UKw-S3WQbUxizDk+H-=5_agS4ul;Tb=vg1aa)JN%JJ&gNqfFa{}u7YRx zk##2x?ay@~HD&@Fb{@fneCK~-?pUA~3_a2zyXHKGY9--Wc{mQmhT>wWER)S_L20l) zTNW#at0{IkEW-z^l;lx4zm7{SX(IC{xh%i$G3Hd7&Q6~lO{%xH^Jl(jO8;m8**q^g z`fwvvs#&6B`F!;MxmT!jbxoQMkJu5uPwG6A7B@xFJnX}@O^+bs^$SbRDq9`grEWEJ zXfLZ_TN5dKawg4tWb?py^ zm|ew;!Jfznn1Hth8!*1WkW{koldgMzOl%lJ7u>FMYD3yN-8(kuQKFCT(t3y}?SuNp za+r74yN$ihK`MePb;>Gy)!LAPD7-nyUn zcb-E+nF2QNkDjSxxE=bFky06J~`k$0OT3 z`l|0C$f<1Rgl@{v1c=;Qi9=i*fqX#h8uZcdT*3t}T73|>e z(K1LRl0Hm&4_DF035C?Av6VjbyM%8ApCI0%#L58&bzc=(_ix?VqJ~j? ztfM@hp1DcWXWpfk_2JO6a>u~UqwzEIHeu%nQaC6D-wk1CuJ%KHb0{vIkj9ovikO)- z3|`NZ$Z*|cL09T$(2xU;!4bHcI2nE%zr=#WisL6Z5tk_THR-q`~)a46qoKa#W)3jN}***A{ z#OG&NU%=;D_)LQiHBs=^^OSGC1z|k5&)SO()V}8fzSmvEw!H@h-NiMH$HjXOXuv5) zl#lI=F+87wqHYb(*UiU@Yd*t~vq4Bd?TBYLl_3g?WDge}WTSodu_q!X+Sd{(_&D0D zLop#ERw!>y>>2hzrHa`fDrffs2e5u${~|j066#;h!^UPK>{l!oWGa~-z}<-qr*^xY zXjzen@ir1n>6S~oL`N3oiz4%s=rJGj(~Y+^cIfPn!go?qD}tp}F|?C2QMY_Ee(4NB z)3Xg&DW8gA!RjnyTqkD6=T`Yo$LD*_z%ZT*vSd#;cC$>6UH-x6lzbCOP4m)u|DFs@ z%-l-y7qyWl`wBq^iV(L{VmsScr!9;_@v;V7Xw8RrCq4$E6PbIuF+@hr(loJqvrn*sWA4_OgX1BdlFxU`PScXIRw_Q&9W!1GMTVg{Om=j|lo~mls?q}ZGfDPGYkX_$ zNp==j$yY%U4gS?Ii2n}#B6(JC@ETXAUE{HLX$X}$#mrI@SwZ|NcGOUZwg=7R>dqYE z3WgidtyM{aPUhxLM{I#DZdfQHlgHFwS!IrSi4jovSc647Q=+|p9R@odSvf6q)t+oWB03Vv1Fln7++vcDsr-?2TzvBd~4#1oZqo z6pyBRu+_!tv~I_jlFsVa}sUfkvSi!7fSH|k@X(nSiW!kuu)M)viD3z(%?Qn z*Q1D%)e=ctLklf!B`Z54BBUfqDKfqqS}ILcGMd_@(xNn^dT-D3{`w#P_jMfi;qkcd zYn<14UH5gJv8J!QaGI}<ImMrkZBBr{QZBBI;jwHAl97BbSlH)VzXd|wcx!YU zCmYpyPG$|JD`hPx(1{Vak1rNI zH)6!QU2v>ThMrMh$Vui?lJ`1cSsnVPxht}ZgrBELOHuM*EececMU&53(6`A~C~VqO z&Zoox8GWN6?j@ii-3Dr!zY5{sB@bf6)f(p^tJ z8b)l_eK)ovN<|~CHbJ z9?OQbs58ZB4`DaEgd*Yw<9PCF$V&5mzNw}3^-&A;FWrU<+LvH3Pl|Q$G1|U7zt)AS z1Vp?KLNYrCv)Y$9eZYv5GhHjJV@=*)T79_>>71WMZoNf&rB43?Lf!D(%f|<1@5K6! z)l}H)D7W0%k-If*6&027wr>wTlH1S%y%I6@?zkwfV6T;zwda;H>!-iG=!s)2xBlKd za@%EvfiEo4%ExR;6kI^=s~XJt^aNSsQ)!D;H5X!@z`eez%GzTU*vkcm>}Km+jDI#4 zS1Z$TZhS5}rH|tJm=pY7Pa7r^2Vh-`6smVzto~VTu6~e_tE}PF$~&R#P0qmV3T4>|GlRM?bW5ga8X?U zAaxD$sZe7BzX{k!ukp-EAkE@1Nx!$lN;53m^&w5 zOihQE33c35Vjw1Im_ahZ9LvKixw{iHxY{Q+^q@?U_#3EB%7^M$1itm;2#<(Go2wT( zjCUZT@gl_4c&zwlZ}y+=CT5c2uloFZHaV;bpwl^$VnfdU;@6A!DfG<^f6^iqu}S^XazOUd2g!g6}%%~1^2a+^1t6_k}XN^-!0&tx#iP(ZkuLp*In5=mqk-tB|8skKX}L zkXOTV7EP8A+Mavb)N)9Prj*&!WQef33b(2uuh0Wq4rNiJmAtU+&n^!@0gt6p?mrfG z{7|gfu1{*~bt&yWYrkcA`D$N6uzWOONvqjwdYUK9Cnh|7oM?dt1m^c{=>J%@)?%_TwkLc-TsX&lGTX zeLj6}&8Lf64~6~gX?x!zqBy}v&f5R z@!)3Nyv=Fo)^j$|7isU?@l+q6B`kB~;yc{INyBM=W*$qwdYQ+9>|zTRji7kmi)oo1 zfqM^@Qc>JQDw!;e_X#(-L7Nrm>{3yTUhJGN)_IkpdEj)*j0{%T(Q{i-PPkE4DGfiS zL|f8F&{8`EtSxF)bD z2b1MA*>wnYPai-(M~9Jw})K2&X+= zM_8^&S2w3UNSYhqe9<~=gd=x*R0J7(=;Cw=wP=m!N$%T*d}>KFfqYjBjV>64v8($b zc4sjDYMp`7_4{}=ppmQFx{VvXi1+Wc9HpY>PvIB^DE7zjdDn#PG+!o5r`IkfX%|CU(|w9! z)!XO`zR>wor)Zpq0c=<5aC+Wjc4BsVQJ_B@;V~?zPsZs*?Z*DaK~-roBopP zoiE0f&FHYswCkeaD&B`VRSJJ?S3&-^JN6uq#DplGyOVRpa(^Wr^I$S&9+nlB6U*n6 zHBOXf9m=P0U6!}={l1K&x8pGXWi4C>pT@05`a1MZ zopm4OO$-7DR-hNgtz2b9E5v@&IYmoW|6Pa0m#DJX)$i!oq}Q~x`7T8)527hk&(f^m zFJvA_G{i%L5-0!S_8W~N7rAL<>;8s}#+?=+i=F;y zBzWaKQ6O=A9rO!#!(TTO*AF>i=YxZYy?Pnm;lWHIKZMPHw2&427TKG;9U_G7Ra)PR zyLw|6wN3JZFXmujIFEfga6R%gV(^SV|0XA#%<{rfd+8J(xBQN3`CLe&u`Uq5 zC685k=V?*o4a9CdgBg-7pl zk!m3Ap=vbYTROLX&^oLx3`UHR1XJBVfxY+kXU}S9v*Q{e!oE1t9#RO;r$xRow7WBz zJk3sH+>t;yULS-BceSzqj0B8t4no`>JGOhWGIOnZiAAq0xy9-Z+z9nw!g1-D7mL?P z!eDC?e#HEQ+$njcv}qo!itVAj)f9fxVv@rR)55KG_gczfBS9z0}AUTHN)H z+8iFhEU-6=KVFCRMtt1W{)m?PODL0f zg}BWvXu3$TA(CU+__4pJGVniPpXLW`qfx7dVCV}zhVGgT&Lr_!FYh0bGmm3`@v90p zd^k?CR*By0J^fGZx02lGAhjqwq9u9D1$!h{3!Zl@`RD(7uGK-0Gum;K=W=ZQwU3#O zQ-Sx~mDn~!gga#Fw}R&L`@HPZIl{J1Ni9X|xZc9gE7RZ8y*_ifm0cp6v!tXZEgijq z*u&Gpwr1xxP|{^{;k`-SLvh9<0eLx9xr8F4ydjFKzeJNXxuMR z&YRRLO}4{UiZ%E4MMGQ6*e zVfNWRtVDJaOaHus)xCO-Nkt#<*rf`m3euqVVk5?H9f;>OwdDR>gtId4W;hM+QY7xs zblNP{f+b%^u>nWz*y(pt|K54~1DnWe*-e^ZC5H4feUP9j@=f-P>)YS^xwg4S$@+^U z>;FcZ&AM;_Hs>Z$?~x)*o}PQ^)34Cih&}K;Ie=B1FJooj2MO9guV(ivO4+v7Lu^pk zePLY=?bEn>U$3Ea>moGt|HK_UdK0)(iQ@wff?m^s$+IQBmcPhxp zRkX&}Y~MjyPIBy`izfTPG(|WgU z5^b%rU>j=|GQoUL;ofNR+5YGk5KFFaOYop(Ee^cePQ9Ky=M1a;t>$I!wsvTliGV+? zG{J8)-a1iun&8&)qUyz!hggZFY_zJ9c~sE9RvzsLrfUt|Ge!(i=KiXSE-Tb#d5 z673l?3Vj{Nu*dtNSe@HmrnfGY#axPI!$-QZNfMi_(jCW=%!cb!&GX@CJt?P_p9(Z} zQwD0s&x6_la|Ft*gh6r!TF2B2b>_@HV~kw87w=|mqcNKHWZme@bEPNap}}t?+#seC zHILQShq9m)Z#H~c6+7(rj>+zPz@m8nV&98NP>6raxizY?gllV%yugoB;Ln)x5)}($;W}hR&@A=|rYK}anC8yze{RK>@>d*RJX+dL*p-=}# zs?0;dq;>Ql;;wbW)yLMp3MTC0#Y=c$w;7(2d`xLv9a^6i;LDDo!g1s|aIwU-oK{(n z<}oRUVT8t8)_6=s@F++_a7*8v-II4^B?-pN&O1|1=H^Vgxpp|u(VYp0^;tmq6yy%q zrt0m+C`szXGqxM=#BU;Xg*5vuZ%xPha+J1h2?fbMr8l2PV&959C=(+i$Q@t~ zFRb>ypSvTNB_!HrnMxN^A(w2wQ{aO>DJQwxS;nofD7yfV9 zmUJjk5Li(*e(-jMy(4+NJSlxhXo}9`!uOO%4aNCfhUCdIoH03#R)aeBx>rBJ$k>4b z^P6vJvNXS6@EwW+nHlu?hz&FcCL(+DVMI0anfu${V9YU549m#jx%9O701f>8hPM68 zrg|j_++XRB=9TBMebYd}B)wUJ|E9+aj%+@FtLbHkZ{ThDCC|{fZ!s&ndV=R9-$p41 z%gAI}C$-`ww`^=UU6}hqIOcn8X3<%xcDi&`o?l#?7N=l_!E<3c|07_i+Q9Z2>bQFw(5H2(2&IoAh}kJ=jC-|RYzS| z@H9=tcdo_H^(SER`Uqu+U!@|I-Bk8&6{P3=hr>-TVZ1twD=BuPLLWZn<^S!s>R6FY z-cZoW`OOTZgAG4vi}^-KO+SUR-4juCQREB$-~DpfL$hQw@t}e5oW~b4|CEMP4-R0M zY60HZ4#UHRmoa0oIvf7-FK4l?$=W2#lJn&;O4o>$z};mhRt~!e#dulv`-o_t&{OWR zLeW|MWvL|Ryr6&^oz0)mrdmqcBgc=k*3;Ye7{|cv5He+ z=IQ~vTON@9bDWKFeaoJ;^%bnE6!A{Ym6s4c*Y`v>JuW(o+i$Nz%Ay-z&P%Y9{lsu| ztTHsiw6P;x62WsCXl(Z_?!n>S2n(e&HOLS@`;1MXvsFg^}4g2{nQYLsN+y42AI|}Fi9Sf(Czap;G7tDmwjXCfeDxjfu^Z9ZF;7KvVeWnk_V&bw z(iGCN+DQ#rU7V(ECfRQOK@T=7qdI}ZpV}o5`xy;BgtUIO>#7jf#-`?xcH3G^3r3}yh@ZuR^Mwc zbxaJwPWx<*J7!ERYV&CJ6UMGr&tzXx^qI}5mB>A{7K2}U3g<|C^K@E0MH1}vH?EBn z#l8GbA9{WVxft^poTG6hRu493tltVr<+sSGhPu~e=Gp)Ny(WBtG@kxGu6!==drdq;8T`)UO5-yS3E>#&V7G}Cqw zIUf(9LAwG-yGD() zK0Fc4UIYH!(<~*&ty=fN`rWEjoo;KB|7PR{A)eLvJ=VD zuOB`Y-9!FEYg*{spJe)Fa*t#^vF%1E?goqY_C4h}-SQQ-QO(JN_D3w;l_~)TtIP;7I?m)aby%Q*td*(3e2&thb?qPsiXeL=4fx_$r2zOt;AcE`)}m?broRGUZ>bj&f}iwPok_CxBO zXgH>i#>TQ=bi4g81<9yDF{GUoj>=J;!%kWlzn#j?uY+_QpYytN7}7p%=&e4aA9=QE zkTcJdx=d%TP;V-OcW_SY-w5p+EA{8luuMt%Rn?D{O?H61-b9>PIs|89%CV*H6ylWn zFjYSG%x!%L+_QGTv3@5qT9ymv!R6TVH1O(HijWDQN{Je(S@WE}MqZ=3K4n}*u`;L9 z5W`K(K4|Uixt-H3QlN;AX3p;QB`)t%EoUlSOxK#<&^O2T7<4n1*3SLIS;-V~?Z>vU zhNCl>MIUt*9pFjj&2Ks7XD7IxN=MY{yg3!~$ zc^*4mG`2WmQ%oLS+n>QsfgWUej^_>a$wE0NpW|&mZ3)P*E287hr^vmwn2rv2^?{sPM*8IyQhp5oi`Eo`>t*U!lE;UiNyLH9j{?ZujRkH`G&5tq>n^q>F+JBCx zj<|iZ;@IJNhN@wyYK|xRDH+eumhH2;mzt`+nrg#OGmXVI_BzwXm#G`e>8SR!yKMR< zUE3gQtt3^6feVy)?pn=V{dBA>21csd7)mY+R#CsJp{;VE&!AyPN(@$XJn0oaYPZbz zvS`huW-G(~^3(P66mF;`MphX%4p?A%XxuEBzefgX@*E<1>|0{5Ut{}g7R1do?Wt$w zpk>-GT|Vn6X?e>vw%*pF&m}seFZ5Fycf8!N|D8BfpSDwm=Ptb&=roGIH(Nou=NY-( zCWiiwvt)WceM42)Da zPKeR!DYNHNx%N)#9IaJ7n?J1Y&()^VJ>RMNZ&%CW$H?r{Km*O@tp@!X%ybR+|5WX{ zMh|(e-F`h-eM57%@tqlk+O=xU#)8;q8ZxKL3~R^9YE7H_N8^4)qSWnh)_;?+pKeci z$13`1EKJmq>-l`G%Rwvo^M}5P5B%i#^3;MB59_1(;ic+U#rS^fr9Ns1?%bA*8r@$q z`E!82*AN(Ij|x|R_a;EUTPxl0*Enk8fVV@*Z1#iMU9@ngFigeFqQcL zmlP|eI(PvaDFrd{v=1!c&K)-9mmxEZ>c#f-y#|YFe>gYpMCrI)D4XR(!xWu~`t79` zuX{1g8Bg(QH6Qo7`zdFADUr)nT*?iVosKB|8SpWfD_lq9IOp%w zhf~1hhsN+TWmsps9@}s6c459v_gAijB?8YlR3`nGPI-Qq0n(Q+d)Rdb})AxB8jG*i%}79;pMZKc5Sl@`yX zH3ip97ofMy9ai?}KAY`!kNN!e!j$QXJYG;WWqpjpr6aqcI3gd14^3eHvK$-x`v>&A zJ(<&_P>lhW>bKL4RF!V8CDy=N2pFRQ}4A(?pl z@G;p(i1H1e{`twN6ee@a5?Z-sZl>(8F__{KHI{p_7D)>{(U5U~R;%PO_4ZZlxr?N5 zU-8Qq++W%4);?E9(E%rs{I47O;N-agWSyOlQ|}Z}TJMB~KQ_R0ZYfNc2V=xa6{LEq zuvu!Nc=n$02ppJ9-=l&l{aP;Nn}*Y^=ZnaF%PLOxmyGaT(-l4rvoDTlW3$@U6izOO@>-u<5x6ma8ORUwg#H-Iv zICH}o*_EM*{8#~Q$N_BUATABA7nI`j!PqA{U;gl&6OD$WLszRL5u zgkse2cy{l;Ia~bs9Jmg9_*Tz=&Hf#T`zeF1v1e$3;u7IAJ#v|u+8gE)+u$1)hIQ8y z;IY%3Hch&0{Wc4n+ThvrX-GY{+~0_dC-1nd97 zB-bIVd11G3E?hnNRQS%{tgA#e84n?oauGS=`)~ty@XWUY2l&{%UlU(b5FZ;_ z=4S-c#XJxzXGPzn3b+Sl(=bn7bQT;RaE;dcJtHqZ$8l@vZwd+7P21M&xJW5CQoF~5lpXkkDZ=Nf6KeS&-Q*tWLx2VSFG%J6gymC=~ zMNj=bSMtx9TxyLg(>@@=&=bmI+8a7|)fR%AxhNR17ps39go8vDd3+gy+jICBk+4;4 z)1+l=!c{HiViKAN7kjAegJ6!u(#A%(mVBIiU#LMVvQ9t~} z9a|_(6vJNI`3Ro65xw?i(vy{2@NzSSEqY6G#X8uq zVHZVoMbfFmC3H!T=RfrQ&F715q?i{?Bop?69(3oEkHlfDc1T8VjU`y^arfVR`Fk-4 zo1H}a-_UrGzjK>dJy*`N8C}?~MGoqnu1Z1evA_*x3CHN`fMCa0(Qb z6;L#aJcIIueK`_YMzgnawBcPmE$yX=?wYT3YD6xH;}W-8Z48CHxJpUmv&n-;{E>+ZP-e-w+FQQD-Hp+e>LUX(h)nrXB?LjRlj*qJ<%Y@&nc z;fDya`7xSh1WjQ}FMt_cOrVtS%jug>9akAYj}7=bn6*iXeACY!Rnp`wGg0okmA4Cv zY+=Lr{4nt~U&tzOAEor(0?&E=I5yE74$)O)^KckF*NNn+@)yBJWiSSvQH9$FkskH5 zGjx~IKb?D0x}D_&EN5ry{aMzb+3axA682y0GIr_iD>D4LQz)Nvp+6|;%u|Zr@QY@6 zn6p3&88-CHap=^EY;=i|O4jD{F4OJFDzt2?q*PtqHXMzOgL=P{l)F9-&7PYuXH-0v zb|qr6mnbi!VfJr0_u=z3cPg+g<6hIe;G1;TW+}~?l+PaRDPiLVWw3#_wxFS4ACCtu zvbPVj_--vZwwkMbc!;T6r?99e6Irm+NN(C^Gj6AO5~sz>JydxW8Y_?DyZ(MCr-?yt zpd&pqZ6ueA#Y2!8n^DLs6)wY=L=>n9%ky^R8xG)v-$xVlZ_FJ&9+u3k1q(MkAS zJWVL0X;lN5a$+0KK9y!iAKsdgGw)qgV9D;>u3_}Xxb?~8EQC;5FPNDkrmK6=7?3VerC`#dEaf7BcK)e5j!I}qEg zdSh|(14>NHC6nE4SQ~B30*&n1lww_$>|w^vr0cNHykG9U*%&ruJ;jA@}2kbEN!Cy%U7U*#k1|Jhp;d=z_kvn-#snggy^e1({<8vjpiXr3=j~BW5 zC^qoi(rifrd)PLgWpB}DK4abS)?BnsMwEuL*E8M0t^Q%1&Lb`VfyZ7*U0l z#nO!C4rhK&x$N=Oi|hv%h8bjd50Re2Gf-5 z56XARs*vezJNs&_QjOtC>oC)q25@yRu^@j2cfGX6n42h zuIi9b;WOT=KCq`uQxN#gR1gw-jl@SUp?T6XD4_fW=l8Fv|*muo9zkR-XTc1KZ>m=R%H1W4{<2z4kQ)s;6ueFSXHIN zHR?779Zpdn}hqZkaMkZBIpv^`4B=>=?*(fuopd0YG*^2RAG2A3O zKjFB4-Q zVuIQ_OsclWD4(71D7}C#ll?g4;)QB0* zIF)qs?5LsoJfyiM+)lWKh3Dq8b#+ec`MD81@AOyNo}5hr+gQ5Tm&X?S69Ji1Nw8O6 zNHecn)1Y2EC?~Ce-v0NVvR%Y*$5G@jh;Rz$0$1=@z!y|`?kZj6B|a8B_o@*%r4$H? zW*OshTO|b*2auw)wBUz-A3^D#diL8jk@vI3U{DCpNwZ}P0@BSfVs;08|Eb4}r`Ey4 zE??M>@$>u8#Q)NTdT`IDj*p|C2#a(Lbk@t_7|)9(zuyA$PjZkcGR3BOZM1of8Wy}7 zf~YV)w=rcVI%aLe!~a@H%=s!=|G3FRpFYM9>1)vAF+hBahT_AT*|4VdP*cfeo^8un z$uR{s%daoyluzeiK%vL(d4!tnsuLPm|h+&nYo7!sGg=XETZF2MvL`fU7v{pngw4R=d!K22KNo9ASl#&*38XWMF9FhDaKMisnX4b_aE6bU}w)eAVg$m`QEL%lO7Vo3z>tOr8TC!e#YRtX1 zgG=<~?cuzhy_i^tCF?uM$DQ9x8A`FmjytjU>^U@@+>XCI*Sn%lKUP=}hCj>tqjW+& zC7p~zyMzsvu6aqza}J|@=n?2nsr$e5-}a*=9L{&srWz6FR`uUpOyJ|)ECwlnJvjyI z`YwpKYY3kioBE6k+wzsxb!EdVQH{Ol;|*FHdc!~S7Oiuf$$}qGVbkkGT=pwy0fv7s zqh5KUxbU9#PqBZJRwWH-(AEd49fQ z?1TR$)({|uP7PCx^w`E@IZEOEzOD55lRcSUNJHzc1gyVQfvwI*$Sg8fSVzXOFBt28 z5VuQi;rp6LbUrA9cJ7I$y%Ut+HponPkAF-*TBLBF`}AE7J}WfPR8>PZ9?A4{>M@#> zzk$1w7im35dKc+vP9ftC4$m$Ue6>#C;_owX`F@^G-zhAWSJ^Omz*Rlso687T7K=|TK^9>j|>QMEwi-+Wx4042>GaQhkw z{qiW@UMS+~dUt&u&F@ymHoYXKAC$_DOb=()Nntc~1cyyBJZGBRQ#zRIA#9&oLJH06 zok$4{8|k9VbF3dY7LZzne3u0{Ww{>b4*x~Xh(x?LnS?ETUg7y((&!#w^G)rh~3B&5ygaS8dp)T0!P2ujv^`j$e71$mxS(Y(D#D5gK z-NE|r$13Z~0oEixNU`t1-BCRL{~>CPlSiV{8hl=phvr#){(jX~gmxyQ`nu@+a?P^{ z9rC_7HcS-zyK{rc4&vn8pW*mpOyJk3s6nClpfVa`MjBw3>OC%P@J#Lkk1yN5ZzWs! zV;{>aPXG7b@0k~`vrX9^ZEUWMNpLW z5TOi4?Jnm|mK~*a&c?#OkEXM z<`X7E1Hbo0McOg)uK$k;PAAcc=OWyRp8x-Uuldi+xw6$6%wa$T+u}TpiKV{9@E_0c z@Xk5(dRan4c^_M7oEiEp`AdcB;zC_>80^RD!pvFnn|8Q9y2E1)#Nph`RT%Un1peDs z{9BvM$`*4cmhRw^cL3YbuI8L87>LyxFFdB=*QVWPBoXuz?cn>)0Ws}p$Ca`j^rQbK(f{HFtm2w47 zg31J+n+vfceKF=*g~03A0rpG1mOWZ?la&wLh=KlE=+sq#%#?E!;P8ObbVTp--yxZZ zUvvUmb#A&zUo)scN5rMo7_*hDVp$|3C8Ek^!hMvzsBI}QC32~B2H?rNhH?>-|C zoI0YXzC$|?!?5@$ny$XVsyXY?QuLY3ba;;EBp&Z*;A48h$Nco%mub=;7kU*#cPbyh zLh`g?_f~rEQ_TLWmJ-Y+rd>nb zd-}sJvyuAnHiS5n3*2SXYupgOsqhZnh2arFs65~%(ANqUOuCsa7=B?kzH)Kc@F*GP zT0@!F<^*OnD3uMHIEl>i-6?(U4RU=VP1Yp^6t$%J+3?M~BXW+7kjefl;Uf9?)-UeLzog(90xkG#*Fm%#2tox*w! z4!vTT!R&T>PcV+%Ox*wkvRZz5k z5}(JRD6BK!t~uwURX|5q*eRDGb}UWBl&$!4OZaThz2mF($$8`sy3lo)!hVjSyL0nVI=v4|9cRMavunB7A1`RQ z$yOXYbc5_~J0W#O&A(@Ro_B1!LMatb>2aYgbhrE;9L}$;k11nqe@whJOL)I+7>`L& z-9qnDi*U2gSCkmv5Pm=DzyoT!upF=V?4Z2pC)QbkG5_?xr~IFX`F!^pkqyeT?gD}e z_%pa_JtCLMQ{OMwg>q@#)P_l~-@(l93Pvu=#I$THc<^y$hnB^|DJ=x`PoiMb*dLRM z<)AwKr*J&p9#W>>?{;vmo?`6w8D-XCZp(b5PEzxeDYSa(RIZ?R4bSy{7vhp)%qjdB z#!R_|c$Hk7xNwv<@VQ5?s-*Gy20_8u7>aF+DB8jaZ#PZBp5NOrJ-AY!_w=6N*^2vu z&fr*B^Wo43d7S;B--_82t^3TiP)0D>Mu|UHSfa0|K8#lx;?98S@K~G!t20I{Flhm+ zjEP}C!uh>$u_}h$a}mmFOzB`w>B178>(K<4r)%(SFY0NFnFqJ_FI#{9V{YscHqp@f^~p?A~JP0~dDY zaXg!#yNxDZyh>1iPVdYP)3_5BSjfj1^^C`9^m;`dOLuK#J6ayHv3ZxEn0^Db zuPf0t>j71C-=n6LJkHR7TU?XZ5SqW|0BMM)(C^~q&uX}+hx;jDIv-oOcOOj^TZ%L9qaYT<+k>WBQu^5l>U+0t{I&xGY%@ zlg@f%?C;CA#TqmHTP+xU`z6PGPkI}i$9;f!SUDdnyBGJyRUkOyKgf2!WFx&>nX%ewrszMM z#Rr+Ohx1LC1CvIB-dSN?7Q7Fo_DcqKUrvL``E>f!af2F)ztP4DH^c^qVDx1dM0K07 z)^-clSZu&<{553#K5DV5vxY3=uRN;?uR(NyKIC48(F-nuo;*26(=Gkk&emAAIc_hD z%RM#$Bk@O+Nh3YFaPj4u!ST@AIWr` zSFpva!-aY>}6kYCn|bqr6Wh67ckr}u*FA9I-c zCJFxD5vclgm)7(W6Z@Vm2Tu18G=H_DVE$8~KEF~m;QIMHaN==uk!q5N%0(jESyf*t zmQrBP>`$5r*ZaV?y@bDebfxE7%X3P7osxn3d;gCeN$JET;TTNe?V;o5S_s>GNPIi2 zG(>wX&vkWN%|Is-mwL;5<~jAuUjHMH8nKbRj zOZqT-1_s`p|9`!odg@<)^Brf_&gUE{PlMh0y)c^lgn2Eo68uO=5KPu+;qlgovKcSz zm{OS-YjaU&_j3K%W1Us>)1i*vOC>{W-uHj~Ss0TRH~!^`Y{-vthsKw=;DIOY23 z))OC2r$Natx#|0Y=wz`9wC$7l9N1x)$;Z3iG_0kB^T%mUul;1Zs+23U)~2yWBD~3- zcD{EP<8!mNb2A<<;ZzTH{OiX|2T7Q&9Yd3=kNm6iprjIAmgcb>2Qk=R3dW;dXN655M!MBT z-1d2rT#C<4Ydx1b>Ka%@bFF@pO653u=cB;e0ds`q|8GnKloi;`GsF~==LyGd`|}Uh zZf*f|fzMyYmA^DJ zB<3>X=#|1URflA8j>`Q|?Gg%SJ~#eO;RMK4`qTcQUie+|9z`yy^wCU>$7-H~?cN>S z-dA#n*>ZVJ*R=xedUy)oJ*TB-N;iq zoTRfa(Wk+F_!M&(dbdyEyW4W4y!OR_VcVhA7=y)+qEX=ygq4*(lr=V#f_^5E`hg&} zp`R0*vPPRNAGion^4fTs(m)$#uf?}GQJmF%N@kb;o5)54t1~&7!!+RXAMT!XC0Fbl z&z`Pd!}52H<^x`%>D?Z4I{ICMUcRuy>ZT&PclkZ1|N9hoqg13vj&nk=g+KQD=}AL( zoQ%zxq1}SRx1R{OIr;y_z^2(iu;9G7plJU!wn5wtaZD@%x*u4>aXtDBeu{LDdt3$zjK57&;wB(oz>l zj>^MX<+*sS%E$J6Fof7$SG0XWmy1jVgelUQW55|7TU9h z^7BWDy=37n2!9gUdAO{PISw z#!i-`r>#Ij$Ve1gI6~$>ab$hc#(Fh9xVAqd$B^Mzgt&6CBsU2-eQWr@5 z(iHgg>n(WMc@2&{c9C?+I!x>4d0k>Av$lof*q=9X^me8zt(h=@3>p9(H;-(e1G3mLeL*oUHCJRa^} zH>m8lU?Ith%woSZQ|wNM)VptJ&hPGpwaun@zWG zht$E0|7%{KZx``=CO6xX$D?tGY2+yhW-UR~foSgU=eyRE4Ht7ka!sV@or3Z>OW|DW z!}C9`;qjlo4_bp`=M;qHT{9nscAlH|^Nvv5`#2i`Ugp>xfAC*j?MHQa-&HBr)fOP1 z$I|YZs{xlM3FllmkIh)TCt^A=dvY70zn zkEc{QK4z@!AojkT_^*8rE=Q306)Q?H*iQ>*_o9(irCdK#C$4V%Fvxxs*;sXA+pstx z25~393hVqlGo8l#Cpzneo^lbMg$t@3Dd)juE@z$;9l4`|^3Hgg&@YijA5g}xs@ae@ zKMm!_zi=Ttz36RS61^}8fOEn?7=3qwLsKLw8+q&}ryAPUPoEVzJ;Y190AU;3+O4cB zd|q4KDL3P~JBgd~_a4dRH==Eq$d^)Sei^&88*p!7U)DWPjzz@Y!<3*f#A(&h-JJ21 zclQBT>aw3Um%QW(E$z6l_s+;2Ey4z@G9E1)2lc3Mnv<&xwfWrk*Nq_rXnNZ^S%mhHbh$8&Um1STF58jg5Zp znu0L#J|FWq;Vo~+eaKZ7@8u@{%H~G&7x|2?){dc6v7_AO84tO`kDrnIcqwTc-*bz_ z^!b73FZc11{O(3tTI~P#qa_Bh|A(@-fXb@r!iMSYlC};0xBRX{hj-qS9txO_xsoS?zPUEnLRt_%sFRf_MW-N%?q0{ zHGHDWgmDYTxxk~7hr64*8*xFD9X)(~+{K3fzyG)hBrlN;)22mV-~@|_uKzzB-2bmS zbne*E*T?1mRfj*{!Y*sX5tu8E8xFroIk%nyO~uhz@{1M>lf%WOQvBM|q*qc{rl^3D zUUG2QDvQa7l#xDD4I84gu!Z}hqWqQGh{%)0xLOVTGE=|?aaFw5mBh6}N-#PjgSRpg z$aWA%?;FijpkGTa8O`)%U_GgG=Ssx%siRAhQrJ&Qc;c@JqjYJE*rNmQA`MI%p@g&d zRB-E_IF7o>z{OV*xgHYGcqxHzk=i(5s)W?Fs&FPn1cyoC+Qb&}3s-^DR(TY4l*FL< zvN%v8gRZ+(u;IBP*g1K$o)LrNXBjvYsv|I51843^!Fsnm67I^P(sJ zi!$mcHq*wQng;rQP8?e-q+uqZjORJ>C^)VPhldR`AyWa8Ob@Gl6``4+2#l-NHb&^%cJcS-urO>RC zDLnku!E`=NSeij+Z@(n#ga_oPkW32$Z_|?V_sQ_kLkhg{l0Lk8MpmgQG$1^UUTSk| z>XTDw-{@o-WSl~2BaTYtiX+ zbA2XdPJ2$t-CmIOo2T^M`zaYoh>4j@&!7j{8T8RTg;c(z(o&0LDoIGE?b^i@)GveX z-N>W+=_#~x{xf=^l}Yag<T0Pf4;ejVy;{^7uz} zGwAEg=TvjHfX2zZ=3$@e(@FcH1RHTfmUXw4XQFpO>Tr2hX{E&HMdesA*|z-f9ErC4 zuah#&XO{w-Bq_yu9hYSjjg|Sj`9bZp=gK5me7;y4P5-nCzJH7O-q}!#e2F@~JjeD2 zcFVWpSwu*+(X1KT4xeP%o==LC;c;fxH)24fB%cqv2bpiDyYM#c7gk@2GZ}arc|s^t^7<+Lr(Gu9VXe z@R6&e)pUhs_}`?G@3*P%>g#m9=^9VBq|0^k*nX2PKe|i0JMQzaomDrf zX_c&msmB!xd3l$jHLuYXj#i{wl{zf5m#vIuj`bzA%S+; z-lEJ`Hz?wtdvr|e4vjFo#M8~vzDAa@33TM+Q@UcDNT2s7Q{2XfeBI<}DdeH|oC+IX z(%z;dDw=eUZgcOV#$4wY5B212S?-6ZT7EdQWCy>YQjf%sL-6GqS59w6j6zZvTpWHV; zpf7_HX`|;uI&tnk9hmTps^TA$Piq22UrXe5yD~e8B3Gu6@y`OP?(v+5Er`qDQwc+% zeD%dWYAN8(a_8D-?y-~O)>Q>ijcdxG6N_J|BYZ;#I7jMXn4ux|Ei=T-)h6hzrH_ZX`fxaB zg!zXpaPhbqj=EZ6wyp(2XPCjX)fgJ$CeYesj0e#sh)XttwYeEq-!(yTZ#^7PFu?K! z`p6p80oQJr!u`B4vh>VgRd0rKxdu2s&k)T!^wHPF5}u1J_~#%u-;hsRQnm41+XR6> z3=w_I95(*CIBa2nd{0xXy=T()%zP{|z_4?=*lgbcA*ZylCQ2WV)pW6(`vyksR>n7d z6*yi~#da-q{8Ot9`k;kfUG?zTLXb1&vmQ2$HH6eB9gJS4gUOcq*!{u;PrDeQhgSzQ zxf$b#i?D_j{?UM!E3@4a3octhue%{kEd@S*orCRl1zY3qH{5;5XD&>N6M9 z{*H=k-&5oD0_s@#hVnXm<@-)%zhW9Q=^NSPy`g!X3TVAUIbDnTNOg5tyZ@TLBo%)virlqs1Y4p`9I%-fwV^}%mZmA;OvtMcaks4AOQcs~jDrwrf zpWIpL&GahyH%)$0Mm4SVJimndAE}pJB^8C1Q*>o9eM~K)vbQC)SK=#$7u3_?gA%y< zrja`Q{!Q1{w4nQh!rv zsU*IKkH)AwLos&4XudB<7Y=UgZ_du+kX1bhk3NnRWZTa-OecSTfKYHSMKp-+Z1R&y3Zyp}pu@6?Q3&!C20d08svwQG)llUH3 z8q<@{Ppt#q3v=S+hKq1$?EFE35TXRps$1@#@V|<`Kk-X=@RcV6@5LAD5uvcR5M7VsNyfyJXO@Xs|{%pPKf)J9{>9Akt-Sr*v( z#1w~p3}N%u3TM(S@nf_(77nq*tzsJ_ud_s^tUV+qSz%?A6~=dVgn5|*PTh0Fz9TLu z-0OnN*WJ)pvJ=!ZI%DA=;P{wMc&*(Hc^BQ`x6BU{ZAh^9cK^6P;0EX?tqFaJN)D4fZpHj5FF}+sA(QpBHOVIca*y=3_VFWsMp0%!sJS+6G|E~LAdHenz zC>YV+zyBAGSVC{GRH5H5o$A`A>(V%3>@K45+AP6Lx(Rau+AYTpZctJt}0*;8rv_m`n)ihP+;2_KaoOUR)`~Y>f|EUVAvQT0V5~NtlCj~Y@ zU5Tx+lVt;|75VY#yp?~CSxpDEkHtm&vnI8VZ~wn`Wld|>LsWMvK^Wuy!ShmR=XLyL zD`rjj*=F~iIqh{Ke5Lp8yk1!;FbVVaxy-OfYD{TfyPf2(#CE<{Z5wN+?BwjbCfFn5 zHDtX4OFAfw<3(k=?P9D{T8gcCBEhQ1OR_tQmD!`m?eet0s|~ zFtH9AEMTn?JDH`%-ejsXmj*dz_Ecz#c^Ac4|NUZ2uXhWEyllp4J#l8MA<4o6B$(b7 zLGA<}S@!5zyA5-S)Z6rJD$-#!()w-X;BJm+>|q1XuMUWKWx=f}SYgu&du%VTgMzL- zoVdEWcxyZuASN~_oqNW<+d|LC98pci_^`qlUmeX6lWdP~-%Vk1!4zL7yWm5JJ?3!p z@Td3e`8MAdV2Lev9kI#O5jGaq2oJYH={gIXG&eR64D9i4gbDn)^I6h3{2i-Iu&vk*l244V$k&v0j6=bXJZ z=&vH^FQOS=XxG+%>Nc1m=awn_h8bX+r3JQ>Tk~}`T@2A^-ac+~uC~E%Zj39UU%cH2 zC(G<`)z=bJ&vWm?**4hp!4N%Om}7Q`9wY|}bjzF#G3|>nf_J#^bj4=Exc2u%u5FF2 z&{5k4xm&GKe!>c4wi+QO+X=e=IAW!-1)lp`!|b^&?$q0&CDj>f+szRB#SX8nZDAki zipU!-sP^uLk)2%OwEj3k)1-$MD`g8s@hyS-|PsoCvxo?u^TsacJZ)(*6-xgep*}5V94bIHnq{gb3fj0-;bxgj$*` zjow{P;cKp-m&A!kNQ^m+u?|OIG(G~$jgP>`Qe4bq!%nAKPpK40*(-Tv!kd)n&fG;V9le>B^JWut_;itK%mf4&!c4UWTin!r2d1hN~p6uswd&vv7iRgXP+I7nw)!q;M7o2F*M})Va@NipsM0|UHdbgq7 zo*!vPuyOZ6Xm;I;$H5W!w&WP@g&)T4Cx@}g^Eh@y9Kn^lf{fXhF1GP|^64Co#(nG{ zk|X{m5^?^=ab}qyf1cw(*e?*;rbXcp>c0qjJdrp6mGQ@5e(NaC)kdT9y6853UHS>n zo`|2`zB6s`NnGC$!3MrQfzSES(h7koM4Fc?~6OjgONLr!L8O03G;k$e^DS- z*aai@s~0}4@WX&x-J!9x8=DB5zl6IgV9CC+mZjcGhTW4x7lajy$j5HaPQYi zrtqKB8I>)<7)#&64ez-3Uzu|cIQ{a#fT6DN_V&X*dP#7)B_>YFdVCQ_D7GRKvYx@=KK`Hu+lsbSto)J(b^MRs{254 zz8``H^x^3n%mqD1zZn2iz_cJyA+Dq@?pqu z>xTgE!HBv&0fw%Fp|*1%OuLUl@YaF28aoMFI|m?j-WV)V3d9?keo$irvHL}4^nYC^0BEl10i??sarsHcdwPG!7 zeb@Y}-e>)aHdx%K>A=w0c=m1%a=y)Oga6_6C(R6;iF0OQs6H_k?~=#-U-3lv@tzAX zX6RhJ6`u#v3FmRA6-`38&zyF;aRrukgBHQiG#tTw)*!xY4Svh5hBBA_k@+Wev0aD4 zaci;U-3nY7yBxRNmmqD!D$G;f0Qnd5Fyz%gJWXPrpnurxDM+6X#=~N@7vmqP`M78s z1{c3^*gh%@J1ZySnfVkrdQahTnk{Cdf9(`>SuvL{`_#|Ih19t?#ulPGErL(L8dS_$ zgA(mk5dX3cXQkKU+xS(G@3I(jXO?1u-U3*@TZBVv){uni*H-7r` zhHI8T0u(|~{xJlLKMcgI*WCO##Rnhz1!3<$L*ee-2MJEl8U z#|FTwM;|2a@JHVIK%6M+h1+sLynTB&^@iuZ0BGq2;A2@oe3t8r#IT`wy~-EV+#l0w zW@6T?fiSu~9WP_8dqCnH;q28AOupucEE(Y$U(mT5wv6k-^HiMP z89TW5)}MP{J#Hkt4ZGpgz<$`AIU1dk!|>@t5Du>l!2&UFY}wisn}+v9&IMreWnUcO z-YsVy^?{3|H_szYx(7V3bm#M==8cNq-Z+-X@#{*+o7WAGR(Qc~wJ#w z&-cLqj9=6bUFV0w#M%c+1^y_x7J^9gq3}383|}S3LU!z6$fgD(@yA&7ygdvrhK|Ql znZCGa<&UxFxi)g}My^3m42uiEty?}YuqjCK5#7S%@Z-L z3pVYv!wd%}jJRckJ-4l}@qjbtUAF~=c1BuvS8Ts#jpNVl5w2zizZwf2o%s%UYvF(-YkN!_XNgrgPFNek$+GE; zL;GxSwvRLSjdjMLo5J@gG@3B*Q8!q{63=V&+}^O*)dyyr+?^b3+;$&a3h&DKyL?f0 zm|@apcRccT!patF1UgzlUD*Mf);PmCNtnBc`fy2}6AX5m;8lY#hY786hElmDOmYdO zQ@CR-Ih-J`ay}L1kB_V4^Gof1qSCn)Lu4B8*6J7XF0>+ZLp=+&Z&w|Ik3 zZlCb+>ucDrC_#a74(>O6#;lT3DBbD0z)8T} zQBV#;-OniR{uYZ%-(iqd364Jcj;+(n+VG{ue1n8ayWaa;f3?+#cppq?Xe%GOF4)7j zkMDP^32MiEd%L;~PgFl8x7|k>qc6^0)JibvW0Gw7vQ|V#wZJq@oLS9ogt_-`{McWE z*K;cn*IJG)v2WnW-n8+0yx?71ewj)=J|xw&m32oqA!$py9GchKRwvS7(sit_(x zenF}BSHAvEdIj`X|Hf9IdK^h^gvOncNfX9u%CagCyDCP>XcZ?dbNMqHlRe2nFRbBE~0tmqRm`F+t*)2FwwQ^krs=% ztjY3+sWCIJ_O%*O-P0|KZ0sCqcIJc>Gxd;T8^*M+7fF@MF>ebMcG*UnNw~H9=tXp= zPsp;23PtAFqkTQfBS4kaJ1MZs*W_4Me?>Opi7K02ug(s-$U;6s1qo_O@QzTz>7|NT zN#ZztQ5v`Ia{06-`n9f!lwLH`rXP)TG+TgD|YGnu)CqumKV~B+_xwxJ0+9+G5icWKs;A*IVgPQ7yd!UB2HA)zlq6nvCbySyd zzQ{$I7}Q4#j$c(V^t>9na#!_u1&mQq;`SV=@%+CA$w4+<8D;t!_?@hX>xn9;+oymt zKb84G(AB~a7c~r;tb&LRf}S7zbdciT0V6hB{U;r{cap7y+uz?|LUn7!jCyMCwqP(@pVppk-R zl(zY?!sl1gxSzi$B)W{$I!R&i3rWmOYN98DYRK_%HMf^W@Ok;H`p(nzZ4}~d)*P!O zruLd%tA8ga@lSL*s)9P6{77PFzS9y}K?kX{FT}c6(Im|;dR9u+;MQK&72y=&D_I87agHRYgY!=1n4rV1VIIom1C zox^g4iy@vP$IZuNkQ$``6-OmtfeMl&RI%JZ6>Bf6a_jv@*!Qdh1_o$Ab%zwn>|}80 ziv)MRg$%f$EWQs^K1(I1nup}vIg>PZB~e3WHW}S`K&kT|QHQvDG%xZQ=}&*gxA&p( ziKN*58r3yiBTey#H2>T^+NyGkir?I!Q=K1D&b@m)UX(Q#Gqy63%z}m3kx4s~=wQq} z>V5bTZGDhLd*dI_!k|=2nJ&aLO-&SH6Ba3EQhn_+>T3CnWcw#mw(D~mOsO?U0z{RJHuXsTTQAw2UbeDdq#*;-_JMGJoyA&Q2PY)e#(ig33R2zGN zo?pF6URQ3A<&o=DXq!mQ>JP}m@B!uf$5Yk*hm;!rjIyq#lKbRT3NjIRo#OoOiJ=*^ zJvg00N53Giv}B6>^qf08ESWw>J*DbT=|p?7iEccn@jjVkFwV_zVq`jnl?C8a=ny;+ zABJVkfq0lV65}ibG2?bm7*6YfJDe*#K*tiWw?Ck-l-9HG&`v?C|ToIjRdVlDL3_<`KjKsPS+#cEwpEiWT z_v!#Vk_v63t4bHUotX$6J*#+={R6va2KG4rw|5~f(=eHG`6mAzw%EkC?(7J?n*M#(|6FnI&$>(}{RV zj@HANiTWs7!hJh$8bDWEA5X4#XsZ|D{JE~`3i6j-H0IW^Ihy7UNQpLp$0LLPiJ!fj z^X)7!L*zyiyzFFyg~P0Iez6r+d)mUxk(jA>s0P#f@vkwx|oR3>`kr<-Tiw zaJ+VC^Z6CcYOuYdh6RzTcpjtg$|f)poyg8TG-8f*A`?dVsxy8$XTL*d}&#jM~Lyfn%#wQ&Ts`;SngBCxGze$cS9EI->BhHu{Mf+>fn66V2kcv zLB8EB70$0DjAcc-P1`T%7%e7?VHNH0{y$XE@TuL#xe+qFOka5g?39wio7qitWSHUb5c zh12RC&||kgFYEm+EqHBKN5)M}1oqWH#tsd*7-%7=T#y&2s)1Q~kD2qv{A7C&;1xbPgozY%NF{3k9tc?>TAXGi3qjCu>-=hCc+=etq=X^ z8n0M;#1%IWxN-68)|k`99(%iSeoaMJ*p<8C?jbjz-Fb$81G!nOsjFpzcQ_C|Gv%ru6P=I(f%;tC~B2a)XI!#Uq9w-z+6 zE9c*BG=g-60Y)hb?~l*7oq3#vV@=TQu`zyfeplrSmbl6Jru{h``;IWjS|?$Q{>SdH zjt=OQYt5bYYm0OZS47`f!k_C8E|z46steCA&$<(zb``fqLa`&C7twEBB-kg)CtR|@ zl5!io8exU9Gd4KaLl|HGkt0g)jIoAR1Ltet)~lNI9FTISotLOSXsi|Xov_7iE=I56 zJr@gEtB>xlb)a!Vom<1vhR081xJMfzYqcp7pK<=?MGlZ(WrIQarYKu$h9_Sv(U@q7 zd7B;JxW^H%e|waRo`$+b+(#F@76fx-|Z9DpvqPx}5*w5dorKFr( zyh~_I<6BBnE2rKr&iIr6F^e0ugmeSq5HMHnq zEv=OML<#qExpPW?Q*f6bv^2DihLu;4{HY3hd9{oNJo!Kee^gM+>naMUFCn>2CA2y$ ziw>$v9wl{A!5S?L=t_PCP9d?+VfZeF2P_>u0~f1|SHRkWu|1+6cxpeHKt z=~;dS4M=`V2LnIS7cL&J{A(qR53eCtR!wt_s>yR$HKm;WLR%NK(ye!&X+_~rS~s(f zq7T-Pq{$b$#m!-^t*xQ=$17;;ni}e{uZ)ULRFLo2FSKfS1=$_`N{iOjl2?=@MnD?J zb}GU~k&8W?CkK5431~ZT_i{`#S?m=@>FpL;HKTzZHPw--pLAPa-lQ}Jx89Dy;j*E~ z&K-cUoZoJPRwyk9(1fgdJL*e^i92wCc+r|!Qv*Yr}U_3}11ob^5 zQL?%ZvWEs?^W=Vreiw+PKEVik=nvWK9@sIvH$o*mvoRdl&m6e4Q`1-YSe+_i*E5(X)GdsvlNz@qO1h|6V=k zQ!$<6g-a89z~D;|-aiP$yPt#cVL>01zxBm7=f3zjuRm^ax*s&}j}BTvoG%~{O;>`r zZ*(vgOzMNUa{3N}+8n|qrhN7J0!Gi8Xxj+S`I#jRxTFvhtKHW;y* zTg%?V;oddt?f?aSVZ11kd-9zPZm3&h+9_w`U9rK3 zJTnx+5Pi8gmz`dwc;IOc1ukZ0m$WgaPt!r4!$xS15WWRF_i1CmB~=`bSAk8P63!H< zpmUNI#&i48_MTTq_ER-%E)d2ZB3Ta)sNzGi8j{DW;dz8QUfx#3!8NLAxZOUMxxmHq z^nS>VA8mE;V5lBER~TY-sWFnAOmK)B({0T)$CBaPSa_5H7xQ5Nw*}m|!bb;Pdvg1* zxELh;m40ofE4*b)(@o2vAJ&v$cgV|Z(Tij=f9@O8e;iY*8Y#6SA3v<;F z(@6&o`_*CbSQ*6%I#_+j46iJ z>BpKp1^%LYbmMa`mM+zcZG7Agn~?6suEcw>MKip4ULw4o6ML~|n*~{-@}Fy6oo5@) zA6QOwcUIHAomK=-|K`P7Oa6_Uzp4v6u<37H5q{XofPeA5Dbs}=?L_~APx>yj%OClo zbhcmrf8qI7hp^7o{r-Z7r}kzxr~anjGtrlw_Zq^oCks3-j2+C5cN@ZrY=$#6j}h!v z#xUkC*`GDHhA^?DAhx0h8}zjgd)$JX_y(-SS3U#fFHBNSrIsh;TctIfl#IqA_Fn85rlB!yspYPE>a+ z^KaYZ`?i-ORZsGAW%6!f(B7*!b^QWPbrR||?qB8WmT)n7664$LI+A#bhl_Y8Rtov@ zwU-bTd#(+xa3QWO?|tbOpFf>{0l#Y6^>RNMgL?x`pvP7T{4-M$@_wz9tKLYD8zk^) z=r5{h6ochYag6b9r0mO*crm<@U+h;q%-1fBmB1d=X4*Qxg`OCRK})Wg&JJ$o`IoCn z8_E2jFkabB^4Pvw3Rjaj-@)Sr_=Ik(m+ zYNP+Y@{$F)BA7>uV8<`hdMa7h%FBG@_l>so_(7X;xv^};Pi{>0gNGl@Z=wXJU!)iL zi(KDIVLasUWxG6ftd`+|3?$lQbl~(qEXXZGNDHfm#3gh8&c|D;Z%9yfRe+Rfg?t z{nGRYIV(yb`CvOw5ns_|$N5S`X-c93=A3S?KQSqd#_UZemkAGP_Vedd#OU26{l|_3OX3^xwuV{A9BD!r^MC)SS(yGuxdf2IuT(&+VahVK?mq;c3pfv9P zHeZn2)Et_LRJye}jha+*Xgc@5)ppYgNoGwEMO`eULiN`)?(|#wy*P(T;*03e^E^6q z;Un$6nN2@OeI~8Ld9-9t1|963MrQ(_Qb^%*it6x!7BoJmrL`IK_@5l!mX4;`d^%}p z9tCmxm__y(7-v!pXG?DhVZZi+5?St>uf*z}D6#}uMdo>5f$3^0vIq$|HuR$k3scu* zZ}c=-FB45B^+1E|uT^In2lbedff3sR1Gcqy2X;O{hmF~y#|Ck}Op~>mEO3hkI~}gU zhQCl{`s!M2{XlhA>aNO$7b~-tF`8`K8a1}NNQKQ`ufZ&zaQG5RY(ytHrhZ4AO?6ga zy{^i${>myW`>HB?HCC179Z_dCk5t&&3~gp2uE)kC=(0*r4K}H_CaX->V_T1zv5%ik zS>bsD{(EY0ONSlr&H1(U_1TzVI_yD{HV-#@B*Tum$gzf*-2PY2$Lk{dw}1DbvkFUi z@V6gU)MjV8>#%*_)!2^l?LJiz{Li)gt1R=4`#0U_L^bAlNtZ1(|F@4^M0;$TDl>cW zZ#q$(#^!(fi~q=L4Nzw8F@O8lw{KEpcBV?)J(6QLixgS7qo8ZCqdI$3rNb1stJqbC z?Ts*ELvl@7{6!Ns>w*~@Zfec!WgXf6Cx?E@M^u&LUfh_NOva;d-9$$&{kDOxj7oR^iP4zHfS=UVdXWhnlOQu=y z@49hc^_bUGo&QNQBiw}PN}K(w?tqFRd%vCQGb0U|r?m+)6gOqxD~*|CmH`VHXUeQP zTClT$)~xiWE!!6Bz$Wi`d<8_SbRGQGvWM^U5~mX_U0B*}jA3OexHq867uglc!j) zh#Gqq^VyCy^tIw~w%j&h*^!oPUw2d1bFmS7$US?PCK|ASj;1U)!-UPKwPPn|+cK#{ zdv+k-o;_XV%$`-auzv5{Six;iCU4k<4X<=%z7i+-cfydvSLxy1Sh7hyPZFI@P|mXB zwDfBfy>&f9$`a?OQ+^cv$T>zU(f$*(r(YyFJv&7W=ObxjcqBdAb(|hM zN09f61GJ{)Fiq`ug2xx#W9x!r^s(a!a^HWR#z>#$;pf9HP(<^2x^Edv2`ZQAy=5G& zT6K=g= z!qbuT?eu;!^g2K%ZIAH0E+ibK5>6+@!U$Tn{16>C5Ok9_J3(siPf*RXSh{%kBK44r zCgbc#n(P!sN#g{0iw~ctwW;T*MCLSQy^bc;8OO6LNm^YlQ)ELNEu4Rj8Z~06_*yK*YQ@sRUYDs*ESAFiU*PF7$}W*! z^;LTR@-pS_h@+GNu{?a@T){q(&6~Fh`fMI?mP(&{yG~NvfA-l;xy;vH`VvP?!;VrxLL@ny686O}7;cTj2`*Sv=MESXI%R>E@}~H{)f5}K zy(hmpez^yL>KIp?sdB`t-S*ImXkT}|aGyIztedd*DAK>!j#~$=cE|T14@~OGtzF)7 z$7vr2OyTwuP8rx4-AB5^B8uC~_t+I~=?<`9PViCg4y*4@sO;m2h5BX)TV{r<11(V# zZI1Ewf|%t(cfy+a;8(U- zZfl35+?s~&0Y}&_w#HT!Qs;}b+iO{sWRI2xCye4; z_t7&uAu+=a+vHu)D$^0qnG42*dm}c+5gT>57|Ht_&$;GM<@RE&oo5Y+SVshKdl9or z9dYY}4XT#gV@sSTyoWm@ZKgG(LmaTa+6iyx+vEBs7esLTed3GVU}e$?1D|(ru?Lh@e1rU#U8Kgjm+8CoWy&nLO7og8ko(?~bTl-EKF>Z&vI&tAh2E}Av<^mjV zlGw*9)Vli;&07*jL5HqzXPe#O`=;BGEA-g^DxbgIe1rA}3w>6-CXpf@Um@j)Yh=9r z3LWr_qogI5Ncs5%y7u!j8F7C41a^f^G$xQ)+$9<_;2LeIzDSx|Zcw7uZJwW)>Q(Ol zQ7_YA#Y-eza+#_l&eHF~i?pZuELnVtCIjjA_td%~;dvaf;v@y^h^ENcSUTt~oTnh_ zC!$M?dp5JCoTZghg=bC#%TAA`r7upB)tu9GG5iF@C7k1Bi|Q8tI!pSUW4Qm#iJ_13 zF;o{W(4Bm8j?`D4AVcmy9Mt1ZQ(*8JI^TMZJY>&MXU!PuIO8;hEE3Mi5bt!6J{ZQ) zNas^D_Sh+!sC1Twgq-3Ic)UR7@n@;e#VhodyAG|pL_@}$rwF+VbhhF=?d!pvogo=V zFX%cQTzP?tqRvy&f^%fJ{~{UaHDk`Vdd!;Gh#Q*qAlp_nZvD(Z52NQ)!M;~Dl!jLD zWsx0w8_V(A{A*iT1WUk|zsmk|oABDI`7fCL&0l|&?>{WZm*L<3g1JP0LB}8Ef5D#m zOa1j7U7{$)YF4(#l#1j}-`VsRO@P&R44KsO7i?Q(?O$buISpJ)dpljtY$;~2LXyqn zQV~zlb*WnmqPn)n3a-xhg{5o6*zwguJm?DVUwEkc6T6lQcDzfj#A?l2ZcVq{F5}uC znEd-U#(xm>`6(vJa_5S(+dYMN!!2`U*cTTG=0Cp`FIWruY-qv5<#l*jUW?%L@9=Z3 z#oT4Jn0fXmR-20Px^*u6jTK#6Fn)M@4CUk4*1zl>&e1AL2j}z&->FJqvvMReWh0Lq z^E2&>Xy)M8l%JJPeW!6}M!4kB#DSSKZf+JS`X$rMtQ49M#?9Xx9+K9}C$!k}IW5mk z=4ty>Kcj;qpV9gJ=X68m1+D3jL8n*_wIt_};Rd?Pv$8qAQZ7BW$)RPdvnXaM zcjm;tZ2C4Niy}Q!DeO`T_1gM`$}Q3;N;ZiMZ64G6$_(nRBiJ#|@-?|umDA^=?01bv@-oYHrb749fhO zK`KqD^yxx6B^joZ&+8Pjk$6r=R%DUMrVL^l>D+nG!r56hY1!0oWieI$dP7Y|KT+k` z_cXrb6W>;Km%fnt;_q}bu!i1Ql~F_a7usLI`Q+|ZkR0XH$9KhaVp$PcJ}RQ=Z{N|& zig#q~@PU+9)X~!D3K}=6l4OjFN!{=>&#&IZ5QD=FutCQFMdyrRqQ#wY(w95Gh4V+2 zaqIA#y4Ygg2t5>g8(|B#2UwOn6F^G5Bi6m*V(3?J-)l1`+|6=F{xxUJ&vC=fA`grk z)DfLp-EpGA5spdj*fZS;-6n8*Fw|Vo>7x@?6*zNyZtRfNix9op6O$}G(R`3_{YOWH zX1ZbSLuX8GaKmyxM{X~KD|FUa;l^yv@2O^j<27bj%YC!oU$RH+ek(lNqK0_u4tVia z9rN8gU`CcY`ZlRUWxFPBaqH;S7xiGn#WWn+Y5>*yR+ur(7*m>TQ90NUqnHCmaIuia zvH}n zV|0`-L|%jmoZbm?MbDl4PfpHESM2ZSgaZ#f;1uVC4D!Iz)!pD8=7}%60l%S6c+nZi zHQ~k#dN#Zb`3GzfX5a+xe;lyKp7Z|>;`VpDdLTH`6;sPBk#xk7#}nV955GY|EI>tr z5NG3U(H=kJHOZ34b2zKd?HdyOB)+>&%7}FPtu5dUbkoQE&VR=fEXmPDpSaaZN}O-v z_D(HKeb63*Afg?}9dx*Iun@2Dl{;7G&pSqxCSMfR^!^(*nrH`ZK7b)e(rpnowo&fA|7)eX#OQr1oJNYJI=`SRD0Zu!&MF3GSon& zNjr}}x=5WjIMdRTH*AI_Qr?K5>j$5}@eQSYRJ7hHc7(HYwb?ctOEx2>J^ z|3l}XhwK;w9OB|(n$!h<{CXBB6|JR>aQ_IikTNc9r#jVQm{uswg|iZK3lk+Oj} zcaC7qf(|%W*#SD64N$9W48wo)G5e|sGTxe_Fs+t6HN?3%5iz9Ckieu&DHPhb(1N$$ z>5btJQXlq%-lS9#xV2iDtY381@f%H#{z(rWNuy~~Bl-K)&_)Rfbouau&ZhjLvL0n5 zKI0Qja{WScbbryWopqFM^_B8&C__u4ku;t*kbHU*S+&;Dlj(I7H?@&`7XPB0VlkYF zZlt_>lFz28dPx=G>KQfb7?Dd5*NNi62hRCEp%L+Hs;x}VcX zikrkSGfx~c6=h^Q>l@X@G?J%cBgqdFL;fdmq_xOlt&0SL<~PyXhCj51N|7Kh;>|Su=P%9 zoEiL^Xn-^xeiDQIC2@>MP=n`fSq$tc51n%EY#^5=8re-2^Tx_Ty@MP+%@TvzAZfU4 z6V{=31WF+Gr4(Ymw9>Jt7Aicbfc+Wjz#LVqJSU46x^mnKp(LU>n(G7WuW39z; zd@*-k=FL`;9M2teznnW4sK-yrE8yCryELx&NMpug8EjQ=ATtQDOlj(p_!Xc6O;OCa zl`>wRRlsUXVU7Bk*c$YaS%ocT;W(nR2$7nr`Mzb>do2z)tw)^GG8j~?LCk{X2!B5x z`v1&_f9_mNx0;Q&x7Ogi)hZr;`lz+ID767aD(kWJ=oUyCY=m9c%~!RJQI8Jry;ukECfc)`=9*$>lX6xSKGsJ(SH%Nte2sBPWb=C74h}g zor?uumg8wsIELBJ`JZ^7<}L#FufoXdIvZcNEQ6x?-+p=#pUq+`aVK0>Z2RGjSoUNI zhHqWL^P1bdh)-$jI&KgD2B_@c0Ll0@2n<~VX-#gg=ZGzQy?5Vj7+bjs8~1H$)8Wkk z7vysL`_6N*iWL{kG3&iuTb)6gHHzK1Sk7Eq%-YL+KV@CTX1iH)v7;_L&Tlzu_**#s zAG+Q=F30ElA8!>YO8dU=rCsNFr4&*2ec!T#EZKKSNcJs6kwVH=2-#m`kC1&0m7>z3 zMQQuZeO=$~`_AxCT* zWr{P8**SWZ0Up;g9i}r(Cu^Yzj_noju2+~~HuINOmL-Z`(okyIhpAli0B5)GQ~UH&UnOp1MRbRM4D-5XeTm%M^Bvae5@^x zvnjwD!<8J+)87VrkC-uCPZz`vu|UI7YdFT5VziDKBF7tGVyYpc*qy%lWD`W*Wb1gD z9@0M;F&z#g*t7l3)e>VY-f4lI2drTI!wOq&8R6D3Bb>0d#F{)Nhw15n`;VF3r9n@; za&kk-MsF-ja)aX&Krz$4#EL5zX~XF#J8&~{Ex6rpEctrV0!wbRo;kPb|Hz9i?>V=k zE#l*8s6exYUr~BuQft}oh+ekb{T;UAwrg#zxCB3YPPtSvu9I0d+~NVuQdwy*vJg>IDTKAW}#+h53>cZ8Oy7F{QHg@K^OzF&Z`)bdvyJo}P zb^At_DhsK{=3MeC$)&LyK9TzL4>WmH5mh|PqmNq(X-7f<{j>Kw$$a}tszYk%U2Z95 z9w?zb3})ewM>*%ekoza5=a%q`O3!|#C0+Aq+TwgFS7F~EvvP{esia}D`Q&2og*MFj zKr?JUkh=b3n(X(C7IlA4r^daaPX3>1@XLG}JTHfO&HO{145q{zJNoC1h9d zlg=IcM8}SPpp7#>lAB!~y)kEZf44u=xfh?Q_tX#cXhbo2j<2Fadq0!W!ZP~9>}`5< zmqKko4aI!?O$r^W>0;(jI`H}jZDhUx_k7ByS2`mVGk$-ka)VD4qWYcY)#Om^i9C8~ z`;IRE{!GRSIdtB(l)Tne&?}81I%fTy^wt+qtlU>pUYJY138l1RO$llJ`A+i3zSHf( zGIG)>q_Dj|=)-^_x-XYU8fM=~sp%)(aVw_&S(UW!;vXW-?-UqPK+^IbNONBBC z8kyfjZM8oX8CpgKr+@Qjx))n3$f8p-8K0}9{S!(^WlJN~ZT?Gl&(`tvu0cQebOOw< z>4`ILd79(MY(3_0$dF${9=FrMnlvr!Vs^`~te73RfdS?_8zX9l9u&Qq-Pto^1hR9J z#VRX2cxi--2Mu6%#R#+93=q+%1JfY}c<_$_0)tI)slPGOr)VQ+n>7~SvP30YgD!Yk zL#w?x&RendHPISI&K5{NY=sqJrbzR(!I&CbJY(|F&>_w^&g7`wUG$mH3N`q2R>q-Z z9XQz=qd)VXX}820+vn)R&x*;a*}3YAya~pISwi|Ov%ixwN5EWLm^&Jw-2)vQ-)xAB zopgEquB>6Yur4x0+Bai%7Bb<-YZT6keL5E%J`!rk^ z3tt#w*K1=;bg_bIk_lFRbAUoBvrUh4K$ae}voov!yBSp4W%WWgC=ivc_^IpRKC5hWc_VJPT!h;f5JvREPm~`#xt1t7!4-%F2GG+yv)2?e^_5=gvK}<`y@(akpK- zv8xO&U#T1Sw#0)|+s|>amYuof-|V;;D{HPtswr1dWx>VB3G3#R`-a>{MKez6ml=0= zqk_~z6=SYomj!oqqtIuVfh|{&?7$_vNY+k+X~tYXU&%UuEJ3*cZCR85S9(58zttKm z!udN>xO;2CU%pxxS4(;PMpG`cyX0H_Z9ONf~SISC&90N8O6M zm2b+;)3oGDk6Li&=NWT{olLk%>&&?xhW4D3i)8O@^;_6;i~0^}*vh{jqP*{sNX*UJ zi0qD=V5Yu93_rhk9r|C{1l8ff9$&9^9bfmN?M5hOtm5+t8It{Qf!RtPmi>DXpZZ*h zWWGh#F#kNl8KU|A5~jN;)P3_?&f~PqL&TwleA!t_*z1bmX3?Roa73`33m2ni#v;se z686F^?QQo9?OV$KPSr!!p<>)R*#BOIWQ1b=*42m$S%LQ|^RRA51dPlgP--CTl{epA zkFXB{?aQ94;Xj78V|)G~asPWi-#1)1BP{r_3cH$?qg|bZ|9|t)l2^71MNZOEW?Ll8 zNejFs{b;@#@Lwv^oDUbaImr`ymni|2c;BR}&F&AOSZLd#C7{lX)bW7mV#fi)}R#qH_8b!hD$9~+AiNeYG`^9v=I{CL1 z_4Rzme)w6$V_~~v_?#usZsDONbuLXr{5Iy#OGlWe7WltYrG;^r-aQtN3MGB~yIz!E z6uKY#=l=iK|4BN4G0XO2ZAmQdyBrYn`fobJ$L&Ml*4;Q7wg`|_bIXX^B@LcZxYZcM?2aM;-W3CR zoK-yo;FLZL2T%7yjN{q;qfyL_rfVvKMax| zgu7ctAUKTqv7u48TQU%e21Ak8JOT=agK_8hP#DGxhFbbaL`)lkFF%Iz?XNLgM7f_M zFlyu^EdOURVjhja$MX|0;psS_<0QDd1*0NtEJnsn!iC&mL=Tt(YsY1%*&hN`#|g;Y zI|ci$1~VTsQxM#CEbdnaGT#D|v1V5g@`C2Fv*1)@cuz;>-ymoln~X1eCLw3RB*ZsO z!h&UiJP+>`L(px^Bq%5aW8&xOXqXm)m747T%X~S-W=uo#tYGG|X&(GFmY}3L1k*0A z#uVkb_@*!uUwtO=G=`s?j`Z?iOmmx#*F(l5YuQ-rdBu47I2J|km_OsQqY&Iu$3DO69F#?5?bEG#c2`TAnoH@uAQhBEmflW`TElST;1 zq4ahe<^!jJl9_C8_cAFaf0V(Nl~pv7*&SW!fkMKpLjqX zlj$G>_xDoBdG(hHn2x1swiE*I*U`sBCUcu9jm|@wsE@7;Mr2CC^^Xi>**GuFk;CO< za>zU=@H1DQO_Ro> zgKcoMe;Z8u&U{T=Zi_XqlyHN|6?=4&!<&V2Jinqfz$b={mDxc0{cfh@r*as`^lv|e zx5s${=Eq~PGA1xN;Lzntn2;lnmF{hz=ckC2%PQE>LkV(qg8t|)ZAF|vmrwgn=F_#0 zpJEu<^^ zztW7DFSJXskoI@Yp*2RI>Ar0aZMg7-R&*_)bml`o_){S%U;0M&zrK^o!;ds2f$47E z{!K&smryYC=RPd^J1r{wNw@RAlFznWn*A`BKb!n)lS3<|im7yQF|C|jMA0!tG|s4u z9&IY8@GaH!WnLYHwIb=39CPa3ULM_2lm(aSld^jbTQY|eb7EeG;x&`stW zKkf_PP9^0%J)ZfA?oSSo+P@`_f{niNaM8Qqrd>#SS4!yl!D@2Ym#c=XV|V>ZBZ%=3&S8$)NXgDFm>x!4EmOWXRyjod&wpzL_-ERno(P zdiqr(gCeHWyot%ZpEFtXay5B`B{$KWV5TEp)<8~PP{r*3>WO)gpm$M~6uhUB`fHSvYQi6S8Bk0rn+qv>h{O+wNp%rj zDlVe+4&> z?tjZ^Zc`b(llsHM^#@fjyFD3L)hIwOv7X-FkwMKuK?c70_Al{Vt{5xmf{W(0lKJv) z`QBa{&SVB#t7soPzh5-3rVpEh|F5XN^tzy@-uzd%Po4R=io*4OklU(jWT0k^Q$hIgFVhjSCCpVb?|);b#>P!p1bdteGa+$--5+p7faC zn^_GC2r6u(A5CmsiEoQ@KGHbdLl)-je!%r`Jw4Ffh*1s zImq)iVvNxDuL6PYPA>`WBi2^b=N#=lh@T-Di}jOev^$-Ez3-0m^*1I+_N?av5}Eww zC=Z`vEA%0vu|sM9ePa+!S zEdJa+zY3Z>sgx-Q{HDKM+4p~IDOE45rb)+Y={CF1lYU-9t9JdSqxHf%D{8l{`4^cT zX8wp4Rg-V-4>E4Y_zRQ#r=x00==!7I)UUpXIf+GGoBb6i#_8M*wR&fQ9A zUEc~yU(igiit4ENJ+sR;YNX(v(&%%gkt)k2<9N!})x=5VB>S_RzW*$x2X&2PrBXvF ziyO%IPBU4>G}1YLX=Gh!pdD<_HTi%bBQzaWOASws(Y)8mWZNf^YC})Z%aDZUMIAt<`7&K%_Ix@T$J2#yL7!MH;V3n! zAEmh1Sn6VOlwvvva1ow+te^*L)8iEN$~wmE&5zPajT5AI{4{k^O`*e5iL^!eI2jH~ zq~O&jX#B6!q|zpxVn(I&{iHWtqTNR4s4V3ac}-+|&0#vt#!2+G>@+>|OeV|n<8(9W zG<8fmLw1u>N$q|*H4bOxbtzPQ;1Vrga*6b&q*2PabgKU+jY5W=WHE2f@w838o~N@< zQfU9XvsB}d%HqPEBxNw&+*#*HCoG+whaIPXJ_tIEW7Z1#j1_B+lSbGHTHZ_0zZH#9 z$6ug#>8yZ5&XPJ&{me^Abhy`9+VS@c(bY7%wCNO8D4w9_&8Nv`RT2%;KSD8Cf==}( z3uafVnNC4FFOpy5W$JM5B<;9ynYOMwL!MIS$jD028y3;Mo@2{oslvToOt-Hy-|>*cvYi9Wju}PEU4sLSSE8OjfnQpy~E#%lvSP z`1;+|0^`V<&ri768P@SaANn7=7E~ z1Z!;*Y_nndFYK=FcbzT1-M2zUv@5cY*dw8o>BG#ohSWuKbpL6JWz0+0>vuMI{K^*h zU$K}SajeZe2b{BG_n!40aCYp13>NbvDSxUFpJft2vxzzD4udeV^4e;m+@#UKf5Y z&Uvv|Cw89D`sj&(=Y+9qy+CsuV8(o4-z;>+bEYpeF4YZko!#*HHt=qhD%XFe68El5 zkog|YQRSY!ZObkE(w3K9O4lfGFN#|GM!cvh&)4`6qfV(KhJAWURB~$|4RHiw&0cwR^pWA{}=8T*p5?MuFz^8 zPTy%n{oh*XZfO$FK?gH=&hw}uXQv?85XlE=aUBXYxq2;C?tZNjw~)nS>1C?GMVL%S z-tWoqgq0OXK5WSrg0-f zA{YIVzf252R5}fhH%fFRe03y!)TS?hDw6?CV>%bFyQ~$r@v4r%^<@z_KYT5;tR?%_ z$YNnGM7*iSPlDItV65pZ(EcqY=z@%hnu1Y@ix6!-2g}2zLHhYrG2Ka;bNPIQ^dc6| zVm5x|F(2U{vh)hq&w|#FsfaTWcze-*8O9KsKh1?0$7H^P*RB-P7(a8VI6wK6WX#7K zXG7}MWWH{=^9;P*EXaY?j!)-d5jl&nf2zQvi2wil>zfpcr#FS?BdYUUqlLB##;DkD zghwk4VY1Qy&()bm609|{V;%AH*2KN?p9uBcsgt=ST{aB+J&x5RtIsR@ncnhW zCM{-q5eZhH9CrW5V(bNOw#KQGW{^rYfy!M=Y+GiC4!s@3ZKg8c4We-tFy9nse*Hh1 zth{)Q8O&|9Fh52gFFG1x-4$K97HhJ&QJRo?Zo;=2%Y6C#wX?)WbyLh^v8zOM95Su= z^nv5n_R-}6ENHDQE_!svlpkh@DdR9I)f8MGTc!tPh5?4g$TK&A=L}=M-HQjBFx+W? zdG1EgJFANkw+(r?rK$mxmKa0hj-aQ=O*O(y<~u^v=Wf^Sn4Gqg&x^`W^LC?hd=yRv z?-IjAaMAC{X31IU=+9VOT@j7Clo%wwIEY@Rafm;%4~8rj%8JZxbAHg)?7mwXj~i8M6iD!IH7HjN|m3Lr7%zB~J!O?h;z+vXjsj2j*|}9jx4MJm zEZWjfORBHF3j=3F!^Yyj^WB0YN^`pHL6-@;5fCdmi~qY`l&}A}6Sr-*Ak=;@0#`=i z&%sz6GG;Mh?jC~5>mxibk=ZA4XGaW@RSsZ?`vDAF--S!aVs@Ni9-LjMGq>lK17{E} z=+gAWKe;g#R7G6eea^mJ*ZLK>bsx#=|&1rr0ZuQ-2 zY13xie|2|6b@4O0ak>@%ZPTW^8>bN2+UAtXIqu^NckZOFV54d*vAUoC3~y zKDg0Q9^6KyuH5I}K3vA2uH5YWH!lj9B_ zB2M!badZ0kaFY!^x#8nHIVTyWGdW=<6dz27ee)EYc^Qn0p7WSb_GQTHv)?+q6J0u*ni`HO% z`!&oD(LCg?Sb+PL!gua88$-_Ec^LF?3Cg=K!{jPqzj7pN4fb3YzMY@euE8dy2*}j0 z#R(l2=fieA7Vl^Jk&8D&K|PY^N&CfGn3b%CUio5Z7Or4EL6_s&op4O=Akc6dBgn8s z{F^Obk7M~8*!O=e0+@d6s3jY)OIb1(A{>Vf;V5@sjZvQA%y*Ea?Z4lJV>)Z`^UV^h z%MQl%Kf)g4-}Rz=hUIePM}{D)dk9uAT~`sze9=-qZO3$WN4qXTN9(1q-YwMG#m__M zCv$P5XgNl`4aN2YD-g~0J#TM@Gk@1?-UHU**^*5#WOBXj=lzlG-vbX9^g*dhKXe@I z4TUA1JZ<$Y-T1WPMo(Oi^XBuSzAb#ZV!;Y#m#8Szi(m_i20%I|0GX-%5xTDrA}&kn z^uP3mbA=ynw(o`f;cm!oE6_URCiEj}|NeS_n7*hi`aMsU_)p!vzAyG9GTOWP;%qOW z{b9@Qe4ADGd+_PR_yJ-XTlWiXOUL&^`j3J5^JyR^gmuTBntrftH;Bix+bEdR0TuV|QZ1X|6Vxx?%SnKeX2s-j`p`yF&dnVW7D$ zrqp|*`$9jSkIu@H_o>8381LKq+s|}Ee??*LTE-B0t`8Eo`iRF^vegSY zt9+1GCcHlhnRU~%su;&MxWqO{(h-IpI9 zfVH{(u_|FWk0T$_8wK+OTGxt)+I`&6Dnp(Mtt-u{VVOg9sjEExcPZx{OA-wLrhP zci;9cJRb;D9_W`!L@J0 zU0%~#ws4I5FTbS|Bz=l_+sR}jx>qIb{>n;8y=|k)bu4SoHMvOUsP>u~w?YNfG-2bqQG-H z?vI_+icbB!nNZ&l2%AB(5EwTTesiXyWXWs{SvZ@=TVD}^f;aOJd}9t4wGTr7^>Z+3 zX)vbF4d&}ln@xcC^=Z%<5{$>zGx6uiJXjA5$7S{PQ12Lt-6z7a=F?(y8!`_=low*= zucf$oZ86T;gupf~6y7&NakTpy^tm01I_pT7+*pSomq-N5Mj&Bs1m+H3hn;7cujcAD zZ0(lDod+@)6CsPmYo##Z3$x2AZzgT|Ci>Y{8b3Vbpthg+QZ8wxbur8b*6BuCY0^Xq zi`pT&N(zVWO5-5&7uHr;77sP$c>Z72)zjk9_2fEE8b_Vvfa+%27F188*d6854@?JQ zi~>F%m&JI^Hi*w}$G7q7&f+dew?PkPs}<3{JxHM)K8=#Y&>$uJUeXqyzB2!0^|ENv zZ37!-v(}aQ6f0o1QT@ECDJJ|crT(lVzxC{%#<`JHgBmD1y_xpUl7`GjW`|}k++T^F z-IMs=6!fKN@%3AAM!jOZSC41c!?4gaBa}>x(w>snCh5Mw0o}< z;VvbIv=2qNn>B z$i`e6sch^b9wLmTSPZAXG>zH5#f)sE^&XA%tX#O8DKru8%q-3cb1A~tTGB+hxeavt z1B*{`T^`}grcDGF{ieQ`=JOqy&Dx)27CXg*`GR#+64&c?5%RLnJ7D9c4$Kd%P%g+G zhgSi@p5|V!xwt)a7W#S#@2CFQ)li!z>?uWkba*um1674RPfNcoY1o^|IPN_e5ARNb zo99fNaGQpTSqot9#`e*R=cD$;T%MPOKj-qcS}pBF>BgfW$WWVwIe8M9p9t=DE)br} zrsB<;OoM%1lf`2Z)g3)N3XQUXXy-i%2`_~IXaAr%sJ=E9CaS~nGi5wxWKF>vCW8~< zk2pUa?|V*1U5@bI62U~jPKwJBH$1!*?UuTh)Kq3MoWITkT@hrT4m0NCWXcRU&!2{~ ze_O8&UukA5T9ygI`c-&J7=H^*%iF^KT~sy;5Y87C*MgxkeFiihN^tjonu%IpVSRnt zF9`R%gW$dNjGU_64BX8OLb&%lbPHa^+-l6i`Ve^Ay{LjnK0d+CDqhxi3~b0 z9h6@hO#h0}I=H5ZVr?3U>-~qM*;+1ozvG5VBQHS~vD)P{ZG0KYF@2rPnM^MR#WZt% zE%h>#M)5FN1TB(cIv>rn%SHx%H<@12v9<^eZU?`)s_<%HF{#EhQ#^}Hw5#baZD9V& zS1*vleQnm}M=gcFkw+gVX8)>RNzvIwl;fLA85eV@i(w%d-2F~vkG|5@FW>3Qq8gID zTTA-P?$}GKmg*-8`axQ+8fYNX>AP3cNEhAf$fTf(UJq%c>Aj`!AX*OVloa4{ocWz( z{b}d?B8A_~7MP9gR)92`*!($>B%g<-;xlFI8La)as!uhFE&UvDd5qI5A3|||247cy_7d(+y2j`CPrL=0h;)pzzKoMb z*YH~IIue&(N6yv=Lj=B8iMEp+W8j3-#-E&N}97SpQe@KKykosx^*=W_Y-g$pnF z^r6jjEGT(`!%@%C^yi%zw`I(tG^0%DUzC4da94~|87PeXhRsbhZN7q}h#N>y&Va)A z2Uywb8Ab-ZgtF&*q$|8-&-yjOI(|Wr@jJxVzQy=YEY4;=FlD_rloM@XeZ>){dblIR z-W>-|IikVO3Wi=zn9T0DyZ5p}<^~(oS=d5f-3;~(`jB$4#MtF#P-S=Z;Vx#doo|7* zY@HGDxOJxudQ@3KPl?$mu{-&vec4^SF1x>G_x&mD-Qc;RGmp30&K1MdopImDi}B=& zqd~6t65_(-@EoKH`mu9H4;23o`~~amVcAU11j1gT=D-#2x7XSTfnw z{dJyj^Z;hCG05+7M7vaH{I&JOavu&y%)Br_!xIfz0Z{$f9cE?$=(=VA`V0s_<(+{j z4P-KaDSZ*H-v>ip^}}|}A+TZci}bC6OyKy1kr+8^I`dCI8U1vJV923CP?qk)cHsBQ7ZS#uvWKl0?qTH(%Yj=Oipn@lH!Gk;&pt`fdI z^F$c)BABv-lM?Rj^MX8ab_5)v6r#ooOTmgSFY#yz3;s4EVw@P^??eygabw6+wEnBl8{NhTLh+P`L!2mkqLA@WaR% zxjXDp#$rjdylehAAm=gLlrKB6csKJ-3HgW%_AtKZAjZ=|7d{`_;EE+(-TA!VE?2Z= zaXhjwyJOKGPjqbP0+Zct__fys0oiWiwxYgd77Oj2FFIkZx{w#a+bIaNTKLib?TN8t zJs|g&1G|nu3VYryb-S7V$eY2Ewo4+N@P+xM(Pn;F0&2}+YG@K@c<4G^j5|yAUoTMmqHC1>E0uO_zfK_$*J#j*3v_DKZ7MClK+_d&(DNQ? z{I|((!)1DNGllw$NTdGqE|8f+3X92>Mjtx~^*cLWrIl~5Q-@>eq$YEjy!u|HYbm#B z*p3IJdh-^^d*7!pm#egF@D=jzkxrv-T%=Gh!N=U{ZK>2QGL3RfFH(|i27Nkqlhk%x zrxkWLnE$KuRPg5#&G5NM(HBygPGbu79es{Y4!uB$uWr&{rgwE={B`_3wimHjZ9U$Trp$Ai(e^cE-)DZaIy|K_ zm)=s_%lB!+qr}QK4DMdYfMB3|KP~*dU6n!9vhQz<1jR&7m z*4uo#|M@PBKk|g!XFR0IJMWUp(T8+;ZUzPUUZ)!$uF%iF_vm@fRl3~u61isHpnaxy zh&pD_mh*QAL60bXbT*|0-KSL@@9}fp#m`pLVfZi9x|dX2N^DXrgU`tl$mf^;Y@I*PZ*+gsC=6--wOz-@mZk^(;nDYu|RFdPVbQ zCxkhCfZn3lh8SWe8uJlSl-lzhbAg}HE2&BNspc}Q;jg%ZUIEZ@2X zS4KzhV~{S~CLU|x>-}&Td5ACHUwe>G2lj{&*JlMD#4xAb*!^KI=8lWPkb6-Wmn>=T ze=-VldodfwzOj6pl4ViiwmOm9pj*3LT(+LS1G!^_d*OXmyTx#|N4v!Nvhr=X;2nio zPxiwgR6@&AGmgiZIA||q79SAf&yC&<|A~9W@PDT5MH`tI+;@t@ zn+x$`I-<7A8U?&X#lpOYXdMyPzgQ-jd#Ncg$g$pqr)s;={#gw3Hzwc*eGqUoUK|4R z+6v#iSe$!s7(I&;;6CdxHqT5D(;Ig!2E%*oflh$19})F``*lxoy{H_W-d|kq%WO7V z#%FuZhtE$)_Cw$$Z)n)`hOAFN9yZ|MC^24X({you&Z{6iStQ&~jqJ#=eXAF$Q(aMe z*8znd!k(s^k2{ZVUgiOdGReNDrQghT{><((0FhG!o^O2&V7fK}&sK(gVHn;EuMY&k zt6u2Gd6BT^QLtihO9R<{%ds0WS9|e&d`Rocrz_%y;BsnrK7V9P0Q09SjAL;^AM8u& zg;z!XsOZuMd4?=T(;FZ3oaKYn2fg5K&7sK z?ZfuKLOzV$n~L!BzIY=#Pj^L45pSc7NXgzxZ; zkWyrys>DA^%m?eLKQO#p!sBN<{t>qe?b=$ORw`Vgck59o(NXSLC;4XAp02{ccwx>i zMagkBPXvAG&gRUI?7!y1;ht=gj5{o;21EM_^Nd2q@1R=bm;6T2 zoNC-UP{sUUS7Yfq$vhOTXsv6V+*jydG=F0{OLKj1N^_%`|JrA^b(oV?kIr|bxr7li z+{pdSh?14(_I+)~#m-XVLVKHIv6~$>Pq)Rr2`=!q^MJ`D2Yl0V#0hN&SQ|KFwX+js zSbc}nw%9w&25*~~&*>u;SZHd4M0pGB?_~pv$1d1Az!6Tjozbz<1qJIJ@Ze@=Sg^RH zGdHprj>)dL(Zv-rirvv`k{fp3bVO>ZH)Ppaf610Ei1+eB$N+cXbZ5vV10C9VK+2SG zr=%O!Ztj62+knkXzyAF6p77Y+jpOx#C(6VZF+$_Q2U-M^q=<bFh-9(`8`j^Wt87hs z!(p<6BEw+!a2zwnO*#Fstr5yS8!PG=8f`&d*vE#3$$DQNBYWL;bUfxTz1ZU zGm&tM>FV5*?~P|?Td(mg_*xt7QPIyC=f=4q@eT7Q+Ku_C8Q2-6%x2>bJ9GN==Ai$R z#f_|Rgu*Hg8X4}mmEns08BXYR+8pKfmdrn+BRqml&^MgJtUO21SUY5?*u(vY1tOV_ z?D;$+B*ZeGl8J41Sp^$5bC1sANTQv^86C6?2 z;EYd;-IzSX22$VcF|D&Ri-T^7Ay3T3|K*rrEM6{?N!e`C;bGB@26(p36gB7Uq2XqW zwWN#gJ6Md(izbkJ=LC}t*2r9Hj|KkBJ|ol$*Hx{tZzi+n=wyXX=iD&msbF`Kb;}-u zm>tORk)6?hV;A^8aR+OP21c{Lo-?YSIYRG>7knl${gQ885ZKlkM;v`Iy~zRbzD_84 z=!}))-LUqR0}2A|k^GyzI|m3;d>t`C%@t>3*tioN(7BTnmJe~kjZ0pz{6(-#CM0!r z#J;110R?_&)@3rX3J=H(_QLYtUGd$k2lKjSyRQeRS)u~9cPDA8}6uxf$ zfn%thbU}=F^hp|@|Iu&;jun^iuJ!*hl9P zYIa6UM+6JEIV&#TIGT*`uq3|h`Bvy#gd_hX6-g`8&}oMRCW1#!N1iUA=AFUaYZsBRQs76)Q<$$#|Bt_4#|54m2Bu(XZ89dd6UuYm zp1~9i$rwgu2xpofL6Uh}q$K3El1{^D+__d`*)}r`&(bd=JM=8J&1BDx)!%Tsgw!_| zdA!Dcms;U#eiqI=v4gMRLAUdmbCvm9+GfVZbTH#=U+HrD*6DH^dNJJ{D^o7`o1mlO z*T<0S@=T9QGLiU#iZjvRE=af5&Dpa?k2|4mBL0?p8rX4@m952P(cCx6nQ)Jnwf3!~ z_`Wrrh~B!~xrq|}7*V?*79YhjPU3GV&d`kebx!E3I95oHOfcm1y(D@^1tWF2u^m*o zX&Mr}o#0Vw+`SH(+}XPlou$p+jJS5E47o|y47ojV#$tXiUNhzDLe07GsSs{8qTzo0$jf9?Qv*&Ry?CwJ07<(<^?&Tc9j zyq&NAd^nooKklY2S7Rvh#X(|`PpF&k5xVi>2nEZ<(eBa%6zp(-#|>Y;kIt{&OD{GE zw9^qoZSwci)U|Q+jpg%HV`)>K5c_26+eC7^eTc>y#nIYT$&~1Pj9eETr0>Z`$>e(! zegAuqR*ZvuYxxNZt2;wu-ltRfgA3$SbCovR zULbeV>*Ql}lTJmPU@=+F5y_sX+`=^S8+Dq#&pAOyS0wTDEZ3i+1}Q&lM|_wQX<%>>eMDvnyUc6!gx8!;^BzquX1Ms9q4|7YPKGsA)69ND>jaEf5?RKDgFMHFW;o9{vj*(LZ8>G?H>e0B=H4&N6Q z(VTXr6mKlt$@bk-PHN2dTQqLFg+Hi!MG0+YK1v1+{7JFVMdUH>6D1^kq=74P=&^q`Py;$P)^TkYsjpylFl;!mC@bm=>Ez|DjHfz2S(RX zxPLYMiLIjRE>$EqU+_B-kSzzF@tv?rQ5CUMJL1b$dHxyB_|yiAAGL+aSOwU;lZ9cf za0hSh+CVRuUydG3$190rF&IzR)26@Ww7b}kYm)WlcK7PdEzR}ezV+?NHLM%VJ=!;l z+dgj$*SB;6cV}53clN+k&hfzn?s?xpt~fl93-2(2Undp&jOGR{9nTH&8O@ht`%mKB z3In;$AA&e(?_eIrbYwVn>4BWg!vHR0Q~)RcD}aah8_|#3uyhc&w#Nv*EP4*<8dJEu z?1|iyq2tA{qsK>aJ3ox%Hd_r9!$i1udyeKzxAt!ZTNGMkdPli$E0_rP|L-^J*MDs{ zlwJ(v>b(YX*ZPfWMW^}EaIRcLZESi?bLGq|$gKX5hAY5##`ejU%9*ec9-OMjyD)|dhR!F^pggqz_c@ZI^-biQ6^(tOU~ z=Mt{QE`)PUpUaJ&7sNHb8_yYB6Y%n5EO6wDIgI>lp?uc>)gMiu&U|{>F?|=GQI>dO z+y$M_d*M5?YaaaA0om8xa9-b!$xnR{v)_*SL+b*MORo5y(G}m=ndBt1WnMI_2R?Tk zfV3IF#O+>~{y^BH>rZom^cO)c5xJ6`5pO%fd7~M6>sw>T8EafU(-}YPIrL$AHl@xk zaGm82!|^sSJZgf+Da?+qj|mFq8uIV@wGXCPW~%aYkDvM;cgZi@$4xJFXK3 zga_~0d9a%~oT9BT#K4tr_jo+B!(Qf!2`s*ZPG7)2!w%0XZI~USHKga*VER*cyc}qb zmCL*0j;%W`GMl}#9XlgnvIRmry2FFnU_CX(i zcLZ$qfZMJv*!|H1*`+qvR&0aZA#ND?%pET8x?#bSZrC}+3kn5Y5Wc4uE*rLn=n~~ zc2h>i+IFasZi}sN6j8zQ@ha`0;vkFSSQaZFSOLAg+haP5PdZ`@^P9zVghcP~Q0BwF zoP95Y>yI00Gt&+F`A`lyTGAM^o9USjRbbCv3JZqF zVACKOthuR(lX}Wfs%i_H$8DidrGRcr1wNPiDB!0w8^bMSI2tNogl;>wPnN-6X=S|a z-x1?t6(MuF6FbwYqGDeMoOq*(399WdXGa@&Pw4=!z3Rvrq=FYLe)xzk%2<}Ej(OkJ z5!9@Lb4JY0bC&@>AKBifh#z9cbh#`LW@8E2-8y)8SsyOuYRo?li{o%Y1G1w!p>U@P zE?!Z=he#z9v%p^#HcX#%Zad%u(=Sr*fKkfra3hPwi|D5c8a7R44 ztO~ogDj1@q!gLSYe*OxeA_``n=+zSYM|z}CLaGZ$CWEuxb3O|y&8SE z&oRQuue#73p^1DJ*M++2qqnsb_nPU;M|Ww%ZA)px^)i;@l+Lu_?mrXcNv20+xdXR^ zdn@1m^@s=(?gQ6NugBJ{)%>`(e=5aVO(8#gcQL+I72<3+$=%$sjpaO^=z0GLD!{gd z1$^ALC5HP9`5E>fHp;I+Mh|v!CmS1JnYH*&$z$n8=p5& zE)=)-yCdO!=Se~4W*#G?qP_;&OXPs+u@bqds7~|?YAzPjIyU!Td`OzthnAAhiTIc+Po@jgZsd$U#W`i&Y->-b(R%uD(INP81-s=DxT^q?q} zAycN1Dbhf~TJO$OArj3+sXmR;JkLi+38~PWL=q|uhO^$?phA)~NAo;rPV;!r-aepB^{#hb!`XXHTTe{iS!hDGs+tnhb0#EVoiRDYzia zx9H06IMpZP$Gea^q z!GUCVFd$Zs+LM4Srexp>V{%u%5B6%bF=;0sE5dsklXqGc#Q$ywlC5n^hEDH9Vt&f^ zUp;pq5zo5F=h6sK9om_!mVZOwmHaM~XLs9^xIlZdHqV^oD|H}aEe%MdfdTn8xdRy( zX-Zlsn~^j7&58e8Q{q0SJ$ZPiJ=wlYes4=l2ciyUWZ*ATa_H+J(LeLc_mRA;9AbU1 zfEb*~mG47qoNKw6yPs^={WBM_dTe7l{Mdf7q-Y;}s|-Wc zv(bC|BU;owz~@dtD(1Y|oO!eHTfyLr2B3OV%>7|2iQ> z^0Q{kirvXA`5qzm%#no29Ley-K+MzJ$j<~1GDZIUar28iY3S`smi(4KqkQuqcW3k> z?=9u&IULE7&kb7|`;yM1J&EN7Z!+VV9N*fD)Kv8*?`QiG&jViM+*)taM?Ux55Y>yE z(Dx_4cHSg*useB|BEp;UMgS^SL6c6Gpzk8@5m5^nFfn?QmBPxA* z$=5NQNr*im8;*1#T`GDL?A(oHH*_a~)%Ha7sQh`fOXHlgN0})}kL^fG>zzdZOgY|@ z*vaQ-FWNea`5gIv$k%s`WjI zx4RoL{gn&OW5V#F;$W)a|GI{saN;e4&y4Q@_UeE7u4Y%G_fc^(qXBCELeG6VH*2GGA?s z7HKbqu9lx&NFWFJWScYGDsnk3p6vV@M-n^7k=&YvHAKcRmto+^0GHe6#U6Ve~3>TED6Bj3m?1rM*74q1qqYP?Lk|^wdLD zI%Pu(8h1pE&Ksyihc;8D&w`uL{@WYRS*A7odDfUcNkj2|I!1oCQyejxu2{IfQ6;QQZ1`bw3KoABp(%g*`Fn(1BnT}(|p z@8ci%f&evY-0-K&`tzSP+;rbR&vE6q|D+>_%g=yrQ>9gt)Tp2Q9O}_H`FHPqRO#@5 zRbK;xYDa!*}4zJ)%Wl4?vd`x(-)yG*FP zfe~F;VL(%+>e1&mn)LMiKW9tBBmbP&6l@uFITD8}je*D@vAHHm)C7u1&g04AkC6*0-;y|6|H>Nj^l;01s%}InO%D)Bg365g= zZaSfF7ja@*ux+Z1wIVI#O^g4`s|)#_Yb^gN5AOc=-1zC?f9JQ2`ZI?w(1)J+d+xkw zjr^MynSZ_qvBUZAwTPJe_Vj3C~UY4`;f+J4ucJ3c0<|GE2Q zc~5hi-2cyBi46drxdZJxvMcq^u%T1V+0q>uwlw#KHQjs4lFoW#MNg%6qBp1mJzm?H z_Hl71WB=(zY_|H5VZMIk*_Hs3zrvp!s}CSmxBHUU*`6eRWuQ1#2sWut_aGt9T*=rM z9>ln(AF+<}B%|g3SwS1!i6ik+;5GT=Ye6rPR_I5nf;~xF`Tm(4!ysaDPFL~$wkA9w z4$$jIN+rEWh=Vs-Hooz{^1tQ(O$W^7h`D@ibW(K>((XXxT+DyVh4j))XA=3_gA8(^ z`8{{a6}g9As@@- zdkn(+kW{I+NR#k7nDlX#ujg+YL3TYHOCqELNx{p(c^*{sK@`E9rJ5HBA&(o)eecMrAb3>Xf zKQG>VoDqFE|4*BY@i(LMM;g<-1%I!3{d(4sMo0ZwgKDaa`Z`ebk0w}wza_mJ{CD3M z%I=N%yMGEW4Hql=G3f7pKTF=9^j16k#WzyZoHkqZr>`G>phtH;Go+>dMsxw+NWXKL z1zkShl6J|mqR+RO(n4($x>LuD9@x~LYQ1ew$pRC4W_o+Na{3?JZmqMSyIlV0dOP2e zD$CbO?`i$5t9rx#w)3-F<9b_@-L(^Y&`iHSeP4hti0n&iD!u=*MS%ONG>)fDv`uk9 z@*nuq&z)(wwF9jRYaFiy9OsGcX|#bYExlz;f7o@WkNY^&;;ZttzELhzZ>uA17wbe1 zRCl5e9qp(|h9#ZimPMA8?z+rnW;Cu}H_7#yR@W-2q*%ncmfQHg z)Ku2e>=+q;@+g@u-;>(}`=7BNyBIX7>aqM@L*cld*;WgH~4Sq@W->Gc_e)3Vd5lz=M0|T=y(6Ed;hZMQTlFTd-QEq&nUeJ1|IJT}$+)KiG=8@ei@`-X*w)`x(u~wg!jqkVVrTa+gphjN70ctH; zJ5~&koRNRiULJx2RODagZP7xW`yW4JiK}Gntnf*b=Z)_=ZtR?~U5RrK=k8A3yF0o% zcI)Bd*nL#_5ZC7a&caQe53v8 z=Kp`>yE;0%x&42TFPAA5f|-*wc-fqXNp?Dr>0}DyH)=z#Fg;uvmJI{a7Jz^g;Vc2u z%@N?ytB&pr-3Sxc{DM`n$wYaFCWjHu-0+ru+|0@5Tuaw*?$&i@F2i98*Y8GG?&|BS zP?Fh$v$MJfhko6L>VFP{YJ)4d_t1jTA(yFj>SD@%Y^C#~4Cwj$7BEupGhLv67<=aG zgSL(#RGvJ_Du0I1_8tMW;+HE}cSHK{qX($W`9cThJpw+2gRSE#p!xe6Xtrh@Om4FY zx>P;{$BqqP+B^ho0!8GVs4&1xw4N=i^kTew0SxH~% z*v&7fj_n;P*t_X)6nJ(24IftPvg5g0+}^IbbXLa;!0yXneMk?^>#_;=B`lab>)8O) zl)7@=KhEan?y%$nr``aUi#K53`a6(TtIVCMvgUL@N5S#cFX2O{%Mj}R0L<)P!1&F> z;OKezH#crSg>m@Y&XuoPM;gxF9!OpGv<1BQ|F=%r^05k5W4Hu z!>{4NZWptUf!CrJpq-ToF4cBi>9!lN{aqIH4*Cd%+j1d&wi;Kh;Lm(+jgi!UHaPQ$EsK0zdgxm1VIe}NooR?5cw!`KP`ZC87?*8^8viUv$Yqc|{3N*e=RtjDf~bcy zEd^pGY=eIDJwW~TOnSbJK5@Ksp=3Cgf8RaX!iq;m`3`1%+*;#?+tju!f+$y7KSpz@&yR`E$|fI9IelN zXZ|{i+0q|}n8&6TGHHMLyYJq|uIDcJ!SR{`xV5qz_GG2Qvy1lNG=GD@U!r7N3~Jg{ zaN^=Mc>S`HTyWBY?cQfdU9K-vO`i@+Rz`!$ksDNR;V02&L-I{v->?x7b!sQ3THA;` zYE_ItmT7^UXCPjCcOC`4BLbWA<-XU1v{VR(zIEpXxDcWXU4^s=e^oY*Pw=zmt4%HW z;tOs0S{Ee|?^>quADqK#m~&&e*scWnDd&v@xg!0^wszdN9ULdXxrI7hK-~>RS#1d~ zq$NVwu9_$6FVLy4lL~r?@T1R93Tf$+GZiTP_KyhPsJx3U)98;`(qT;cGoO9ww~I|r zT`usF_V#=aFBaVu(vlsoExGE-4cv#+EnM$=Te%klQ#pr=Q$<~b^7lih3u&nkayWAZ zyb1Q*JVPen2_abJji`rER=(gqPTN?GZNntI+Aw3jR_!S&t84Jt<6H49Kd(Ry-|y(~ zd@~l*9mD*8cpT{)PAA8f(^K&&(8(zpwBuBuzjlJ`*F#59=O>>oiTVn1;_e<5VS*g5 z+>YFbTU6BR!Mji{sqUMitgq*G1&u%_;5UVTB5K95uNNJ;*Iin2cMazOubIiVg=gXN zx#Q8Ob~`rVmiD@UuPrl>&pC}JeC`Tx$rG9(#;M~Ev5?$_cyHNi%=r?-4qCPp`{)#x zo}@T>v|HrjFSO^p3fYy#6S4fWFP193VqMXGiFj*{9)0~yH#pbt?w$KTsqW=k>G-Aw+=%Ov5(>B zMcmw65+P>v!ZUEA=YGGoI$OWmO>!nAVpaP3ngJ&1d4k zfvfP?gBZH_=0RG$ZV?7sbJ(%og)JFa#X_{dvQGOM?KIj6>hpD=y8bJhdC~yPwk&|( zgNwoYYEQU7>b5wB$Z88guzC2Ye?YT-5_Edg9RmG_$d)YY3Z9o*vSRt#a!PkU(Px6~ zH(rfqOXlU`8aq`!_tO~+)BA)QBC2q}qGG(YKNGhuL;0eKjOAYIgNH*RFyH7G>(eh8 zuV zMouXZ@*=-UXT&}%-9FTh|2d-tFTid3s-W@wtt@0zPkgmG3O$||V9E!?#r@*Z3R76Y zv`(_MF`q@9&gC+8hCfJcRbpLrLPoHczw99T*BrLz!ZqqoLoqmKHGN7aihfBxltT^P z?xC6CJ7ps$+9D3BVLEe8Vy3?mQ>&>Gbd{z_*3bzP^-GjZpOeDHeMGuBKhl`A-9Bc} z(h}dD+lfjZv3UPYC=Q;jk3&k_F{|M%i?Q#3gWTh>di8b;4%~&Ib*#S3<4g%3{Lw+=%xyl{?@2X0qwi&wwcq0Nm#mXKJ>zISm#H*YU? z=l42vo^61|y-m?r#hOLmD@4QGa#ZPlL97$??qH{R{6{&v>Yax?Snj^JC-0Z#O?qJ#PeLprw!9I|X+!SDxcRryuyo^u%U z3v%&uuU)9zIf13>CBti@o}x}$4s8_rnpEI9y+tCeBmbbO-74JtE}o_Jwa4TrhH3{! zHt8?o{i}oYis3BmJbDAV4Ss=j@jjS+Kadsou4j|a?S#5FU+C|Yq+-78Iu-C+e2RfZ zfeg;?HsC7VM}teAs|fGd>Vgo9d}GV*%MxAY$|^Zuyj?m)90P>170n*}CEvAgg^(9% z-a%{r7nb1dBP&JNkB^2z8;}U)xP6_1F4f6Mkye>uroai${f0Hq`tS{-Tkyl|G{yGQ zf3g<;2K)1S%Phsbz|SkhO-y%sb{bzAx`^rV72R0nm(`fpY=uIf3H|pLKbjgWJ9UV$ zCwq0}?QI%wi!X^3E^r2Z%MdSZ)!Q7QDhi;QXZjYQ*lY3y*|GTLk47uN0jGIX05fn^Qyixisg!b|f` zv%ViMlEe2(uy0m2D?2g+*V)9eLieY1M<(LSq3dzpk;CXM_fhWx`G2LBBgQokMdyjX z*z1AO%sM2Cst(vd3*~nhN94uA5|82o`SI0uzZpz?dJ`{A9FT-=`$?T3k!xc&U zv1)x7mdBN&n&%cAr_&3ko|QkB_pig0f_L~vFAL}Vti^1ryBOzwlD+vAhhCSoF{W-W z?k(|Ck4P`6U67aQ>I?MO^?S<}?tg!)p1jn6tkzq0=bsI){$?!@dvwhzLUZ)r6X60?Ma|_F1tlvHRlAKrM<(?(yuKZ zQFEKmw9;%nSU03W^!G;)a`hHWxah$J{$t7sc#r13|I5aMcD+ToP$mS|gIz^_|K=r4 za3S7wx`m*J02AR45?w^yoAS$g98&N;Hs&-8ZGK0QFAZ`O@Fhg;07?ng6=vsG_`A-&*u4nul$M zAH;w)y-}?^3q~x?X&TccyNjMu=~yq27_WdQpZuWIS`RKPS_$J%-lKt!-jK=Pqv)wF z$I!j2Z=()e0NrR$DToI&~2YCn0A@x%1S z34C3y$b1) z6Ap}YT@N*%WN<&e8vN?-!te1t!7{29*Y)Bru-dl>K1Y^OEb<5gq zhXbb+vW&jhnZ~&NDElg9EBuwwZgPOwepRjmHJrN|r!I*FIS*;~@+Ttv^~J61y`4Gh zbYTbCp6HDS-E^=_WiDGcc?h=jX@jTk*U3&>oWQn*o;c%RFKlr6f@(WD;^Th%vHXlB zV(ACwc;Pn-dP(Kax$$&fV6@C_;d(Z$&KC-Hl*5aWIbfW#1CG?bC1YmF&vSjqfbDS) zpwhOMHmr@KAG+Km?zg+Z(7i3-;mUn*O7ATMEwO^cj5rvyeH*oCTS_CAEE9DV^tP^> zDEe!VRSJpo`9uR(SGzU&r73=T$q=s4ZBTYj1I9(>15^Df+APpFg@A9%OHMp(ocm~^ z`)@pWq>=3KPjMVl&@|Bs@r&tQP~}poqE7xxo3J1Twv;8%jSq7Em)sbSju4iV1a-Ef$#TmFiGEHtOhCt1v9x%o9gHz+%L0BRVl=l21$ipri5R%G2TpzR9I0DtUTbtE zrWZcNuRq)K2Vy9{o8)5M`EVR|E?>m|=vjdl=50j2f-U{0i|t*q;-!&dY)dUWBjO7D zE4OzO(xTm!t@q&Y5x)dli6g%gZ%sZWi})a=FABp&e)%VcDd^fr?8LD|pw$nv0D(pb ziN{AO;2V-lXcPR@^B7Uik(z~Myrv!Nqovg}c8apEr#%pHlh&kw(Ki!;pLA?lDm_zh zQ^a!_It(ft!fEh=Bv!ff!4ZM)lz49uPr&WG?UJas0MD^*u7Lkg7yo-i-Gnkf`Gsc! zz7WQSO%l>#nR$d7Ho;Brw}GH9t7Tu42g@#Z9Z7u_B+BOk^!Uli_PpyS9$n2m`LCOm zMSrd?ZJaj~_>@2T^Z%XYd>P-Suf^8^jdK8tAG8CvA^xJgERR4a=+~&nXup#XSeXW2 z0!!g-z(CmM{6p}AsNd5SNrEp${pPE43iv1sPeE6S?t)L4^p6vp+n!M4X^toh$$&Az z9bnzELsYMPFigDknO?fyoGV=A%(c=y2G0j=hRBw@EO63H#)jEK^*9OFRu9|`ly}A> zMv0KK?h8u{Uq$=aOrTf$4=LGtq+aIh^0>t6i#1xsDxuwq<1~GuCx+iIqK`~pu~w5# zkmT!TqD`H;-GqhB8|Yc(SG4QsA+Sk)N4Z1ER+ivo3oT#Q&@G2>P*UYNh6je8VdU!RVqQ=fsgc`f_P@GkpwjZK&+cpmsR=1e4bUceuNL6 z+vB|>DR{ljMZ6Xkh*j0`Y}Exv(J!u=t7SoD@d9trcY*dgIK*>^sHXsHF|VGk%DTgb zwvGW44}Iusd>Ps`45f27DdD}z)_6;PNBpLq@_FUMW_V;OvPFCQvWe;Ug?1=)!fH`Z zwWU_jY+76JH_l~`%C(@!#WE6B9SzwL*--gF0|Pdcu?uR&5VXdWbxGL9y2NG4dTo74 zqt2OUjmQ zq27J$+2|fQf*q3duix+jeiGA7JZF2Qq18+$G+t8?~oF<$^@oYQ$0~f14)i?Vzz+j7=|Y0^RkwVmhXJHiS%0qMM55 zk`i!*0pHrw5~EI_WT8QGZiI<`SkQ8#%=%Oeq*SJhH0n?D!Rtd2gh-x~you)_oI3&) z?GBS!n?KU|^R~g7%O&i?y~*s=NFy8+sZydp^$Y7T?>M`7^evU&b5Fn6cV{K4=b5Wz z6h7;?1Mj`e!=u+U_<7!2(Rc6_Y-rV)@4q$~uR2s>`6qSAevt_4+b@UC_N};M=}O!L zS`X7q)>2oi+w|~>)|{5FA@|MHmb-DbFK4k$hs!JK4!@UqaPEEmIn$AaaQbdfj^EIM z^Neue=6-C(b>zEpGpwz+?|s#|#mfyi<@?T@?k6K|@e&o`d0JX=RE2voK!Xe4@c}xw zybQxfZKdh?tvG+BIbhZ~1wxVu*K@-KXj7C9hmW@5ZpznvhWQ?a19xTaP+NN(@N8T7N2iWH?3w}Ewp}4S`90m!P(W>`zNYRU=YqkH1fJ5BAuYM<$?Zfvh5U;#4qS&MTllA$HTTo^9B2m& z;4*B+L;UW$a3=f;Mw!)M-u zac0S^|LIcpA><&)#s$Ov(^tSbODg|T?MCQ+U>#h0FQ3y38v^A!TXTD6*O0t86K<8m zZSZb&3+9veU@@!<=qTH8MJ>O;n~@c;R^vQ)1s{|3K5!b+P@n-WCN~+?M(fqDekvK+F^2bN!J$vbaADvXj|aRUht&vTG;lx87RNWf~enj;Bc=) z@J~t*^sguuapJ}vg6vJLxMj2C>$hFM((ADn+>w~+VCQ2@$6aj46N+(V*hL)I_Kb^wV^b!|O6=E(ViIicZ>_yU)M%#FK<2uapT^b zayNVbfb~0$(eE}4RHt(==dc5(mvtSc51tDieI7t^+6}0H$Mn2UA=oAr zg30%8+~8L0AZnL|*rrwLIPP_(F4uAK6fS9IgeWJtptE8;ZdKKc3BLW7yBkxr*Rq|x z_G8#%&M2p}=3I81QgeyjULs-GPXFT-iTLEc<7kDHg5W zBcvrlxN#sw=CZj98`-Lic3)%+1*vMB>OE80>AmY9v{N6L{V)JdJMDme6Lq22ysI$E zsXtWa*b4ecpHNG@tL+CnWa9jw$S>m9EQOt8Jay5fsPXyoiTg^_&Pl_-3*GSM&i?p5 zB8awMa*;mSv_rH_&?O}E2(;?|Q$auBfx766CSCni8|y@ny=g>aeF*80nX_p^JMsBg z&_R0fbK`S%6P=B9gJ`=@ruS|bx{XUh7Ltcyt%C8*+ncOrWVTzg(2ukOHGm=J!LZ}) z9xw>q4&nMMVCr;;QgLdyLbvKmk74kh#{UL^Z~dU=+>8O@+=W!gcVDtYK_}4O zQTZvSrjZ7YcnjXBuZBSD z*AOCKv;V$ooyfD{ArG|<3E`!uiTF#$Ocd%@dOaZqKV1~(IwadQZE&V>E>70V#)8c*c$g;A8`Go; z{;i$M+^n8Hqf0us7GQ!dVkp+q)A2fU!C#dq{k3~_RnlI~KhaFi2Gkr9yi>=rvz}PH_64?7%Ea4| zX=v2h4yzuH<>x`Q$H!9@_KSQk+$hW87#m@7wPf`133PGLD)c+D3BBfb z!0j*HF>u*-tbLw==AZTWcDp0du*WC#y1D^ZygGy19JYu$P1D?q=dbQz8#a7peT_#@ zkDoI@Xh%2)PdyMj4N!jtR2jq(G?AdH5<@&8_h9IqN8IAp_ieWpo5gA714tBN?>V! zt|U3VH{0O1nzZw&qz{r-vYo0GVD)JmSWVdgm%Aj=HpjZtd*_haAA8JN&(#Lwbq-)2 zJqHeiw`TWuwu4)zwdkmEU7=w@f4XJsZTeyU3~D5~BU7JU%C4QdE&F2sh(=938=_fXGpfUz1#bEpCWAJCY`%K*>8+A6V7JauzqZt-T zd*X@DXX%F{{+P09H|lRQMg4BO@nPO}oHlqeE(a8Gzo#c)O|vw#v>$}0tNYMi5uNb( zDp$%MQi0WNTf)sn)5NyKO4C8dI04Kv>V>>yf}1Ja)Qn;CYF{$%W#?pL3e#B2@ou2e z*GS~s;l~`YZKpgRf$J}~L(3m~n5};gyjgaXeLkIlQ|ELM@z(d!5MpWAw?RUkN|KY$ z(zqNK?A@XzJG!{LAWQ0hZv)$#ZUZO0`qTXS<0aww5)oG@+xbx3PbL}o>Ll}8+Fg`c z(Dn$z;6AAS`#84mG7jgio*?2UzFLMu`)?52jvz}2<})nCyn||jsE@$&-yoI;*6(Dq z!}Hmx(kVDmevioZ>J{w6?I_mqTqbLGrVYM+vy0Wv%@S=8_|fjU*zE2h?6EWu7mt{U zmSwxKSG%jlLnQ@xRV5wMd@I?A&q=JbWqb79I)$>72(*1N7piQNuy9E&Gwsqr)V0;g z-gthr7p_Qf!}Ax{lgbYcm}Pzj{|OyP6ZMbq!BDd`}bnBq`6k>E@BN z71HE$bZ~PBI2q)_+KXMo&*;e*7&ZzSVtI{&%Sb9?Lz!+B`38@{c5PqQ9u4ATx091Q?gQ1M**8h5ira z??#92!N2}%@}n3ZsoP1K;LBcT;$HcBnVVf6M$0bXlS?Y%|Jw4dXBB(}-c2EE+IM_< zdnk6f6N~9BM&j}#yD{xpzI=UNT(9{{jv(X5Piv?xY%M#f=>n=I$7r6F8y+2DE$F4- z`QLPA?^2B4pFu-o2FpfFT*(G$@UrS=Zs_;HADw$t3Ob20;%9j%Z1^(q3%2e&50B0% zp%P^oYc^#)w%T1UJNiLitcz6jJUBE#pYAi&z`;SgsGCnbxPvp=y}AxLlY-!5<_Nr) z-4)i?Su+FqGwuAKEE%ey{KAl89Db^++QPAH*{uaJb%*} zmP)eD{Roc9d_ZxVCVX8J2R(I**6==3L0IGj!ab-D?Feko+i=O%2l#sfhnG&d^VdEp@mtbjaiDbt-X1w03+w~%)UPvSxc)nuFcjUo`QIQ_ zJ>!Vl<<|H@&mP$mD zXfpI2Wl{OzST|%YUZ1f6D_S)d>$Wg256$Frk!HR#XS zzrZI1KlV%!HQ)HEHN+RF2 zV=CF}ZGasNH-g*h8+7?o6{y#^2uGH;68K5cQ(3g_=2#;kmcD$p2VT!;&S_mAAeIaA zi}eqRY0K7|L>&Y?A-L53gF<=1_yc0RXr=~#CZ=(mb^F?kPrcBJZ^9krb4zq#LAQCqi&=|#jsHR0HN|&T>C@>Abcj>q-aUc7_Gqbs z*P&hS;OF^65UQ2|>tEHv$hCjw5Cwihc;j&u+PyU347xSIiVz)c(PcHUz64%EAc>ls zS>0Rk&2GtQk(HpAG>=aEITgZ|yTbHwt69BLXDr;@3_91$LzR}jng8%=Jic!QrrxQ* zMa`5DTclGGu#&!$HAnj&VYt+=)pDrXJiZf1wwD&-|^3xiM4Lj+J31=t-XVbCXMo`{ zLS$wR1b3rwVfaUye|#wIG2aMYcjKU@?;dC!G#nOrU#3fQmeKx8UeMtW$J5of9DWqDa`F%~vxwtk=t$#B*Ri}&v5M$`8 zq>d$@w~^biH`&aN^;C0Y5$v0t0vD@>LBE>SK)lw1-s6q*kbH0d+_X;MlW78{ub052 zOk=24c}+KZrqbzgo#5oGXY@v^00{Q>bh$Eb16+H&77BFM$iJhCT*_4 z-ah&8I1E%gd4EISa%;?fZ9Dg?bP-CU}^=>mKx`2P}RI!f> z{h+!umpM*yhR!)oGR<@Sz~t33alAaDURv^ObOhR;p9||$Qw4m<%S|$tc%eDlD4(+* zWadCCTkIrz>N}&$uOqBaNGUW}AA|^-=|Vde?YP{^5JD?+(Iw0PdhR_zO1`JkiA9cV zOQ(@o*E|g?!;i6plB=v_;8^VTzyP}&WV7yBRW5=o>&g+3my@O7d-A6dFx@Q3t7Uu!Y0RmsJ<4@0rt*nSv&#za>0Wg}G% z?g?Xuo}tmZIb8m9af#Erd8}~DgOU*+H!wA|a;#jVggzVmsb)`#YR76>`Sa`&K6Ep= zG|&rw99M?ZwO=u~A&^zrhQW#jr%=UFS@hwzbHNxnEe`WtefY9{v3R3X2*&(M!)Zeb znf5vrR&{0`8}v(!_PO$yPO*%}mqprG_pS{e_VW|F{cJR@`k9ECCnH6@oBX!%kqufl zFJ@P^UBx;2mvNM?2H$x}D?Z4?g!h<#3FmBT!AI2J!_hlV;tu{a&Kq_e6OydZ!8Qm> z>|5}YU!26=!B79<>Au;XkL-CH?-XnE1qm+LFlaCS+OZqeH^m5e5+P{JQN_~hZE)vz z6=v{kw@hQfMA^#)#*j9I1HZL{n@9?`w#s=Vkcb(AFaX%i)y1}S=ZwxAKjiyDnk<7n>!@4fV zsh6c_rgQ}-Jll_FzTL;8?d5wl1HYkt-F6&pT#n5`*Wu}&%KYb}r$l|VWEb(hXXE_Y z@%3H#VM)2T#s3igQ=ftVq!h8%Oyf3t0c+RMD)w_B~DzAh-*-}dMel8{07V&>fvn)`v_eTox(VYSneH)}PPr>E*szrVN-U96+h^}BFhb$497VYP@S)QQcj z0*F-7lBvI+2SQ$`&nb0!*nHRBroAE}tpMMY)Oc1!NH>MOWz&iEqKh(lUZfS`gJ(RU zOYe6AKLwtUcD;6*1|BU}j3@unNMApn$^Iv9Nc0ruc%YR+&n8?OmH7hfFTT3r7EO2x z`1|pOV5|POEXOCU-G+YA`Afd{oZW5_?zz}{LPwGB>ERD>hnp*&DVc=NLKoqsxBB?` zc`0fiJc7U25?nsYi3huQ-07vp52(M4?+$e2KLmE-(VF1e4pw~TdJDd&eH;GnX%~L1 zf8&18jIRUu>pccH^%bdJ+97^c#8Q6V{4G2#)D_^;Hyuu3*r?|GnLBn0ytgGcn%>={ z@n)Tvp3+I&cOwya zpNp~=?dZL%o5)vn?i__(H;YE{$=agblIh-q#BzaV-byWA?~)OpSFl&$C-Qlfr^}P8 zt$AIg8jNq4fc2B|6lEip-%z9l{DFH;ih2lXn`4F|-?MKF_$g1WVC?K#JQF33ZITLK zQ=z>{{f2g=4mIb+v_KPC?xCR18QffiCogGWu`_!s;9KX80)bY*Z~G-qOi#TT0|VuE zayP*bwdN_G_MTnue}TPNCBcoo9&2(!#tffl99aE!a? zOG$#arYI*nIRWox`=X8IMcHXB13dd77FR(hu}nU7F6brtCOpnxOt(Ql2;F?6WL$?D z`uKd~9Qw_v?ZDXUCtW(m1@0U>PM;U_q3e@RH`-Ke$gZC~E9fioc@#g2UA&blmQ5PD z977`}bNq>&=oN);1TO zr`M)Y*K{h_m&MR22Ila*zcDO7xQgaK zoC;TWuVr2tr)9UCY}svD06RV`oo%>Z%dT9OpMz+Bn9Vco!9MSO$bKhhuptk1>5P*r z*=!?VW%X5T{fAiD-Ebx765K`RrK}A{&v=v!Uoe{vdNC088_j~$#o6%u@g#_}*Z?Ew zG|;wepwo|Uf_VR*f{!IEQG9c;g|3CE>)C^G;qmTOOv%ialr3Z!0^(R6Q4Zd7_!nE+k95 zI*ILLobG6L%U4CPPkNwPXSz-Mi`$gg&tg8u*_uh4ZOdb4k`_v7Z1RimXWJ$@a$kgS01kMl@P=WT4HJI`|BweitG29@xb z?TMR#PKSG8YPWQJ>b4IjZYjX~zQ37E`zU-F*^Kv?^9#S69E=x7Cg9C7rF(N{ppRnr++POYa z^aa5;vtDTMH7CV!S#0Zq4+Z%0u-E8xJQSI?xTY!j$g!F(nIk2VI1 zp8|hajnm;@!s!*=P7*sr_Z*#_y;SCgsNygJca>mvP=LWf`qdm)*UVgKvj-s9Oc)ek3na#1f zO*HNvlI}PgM^`_KWAMg!Rl2jodvfDbmDnr%4*Bx=AsO$nA6`2E({s?UFN7C&1{(`c zIF@M%+t&|gaKQ&<<*P_&oh2LseFk%f?lir_9t+*K8)#HuZ)snJ(X?w{e`!^H0n;_q zE{L9~+CY;Fm8s6AUfA%Wxi7OS;Q;Lvt&DZjNGiS32Av+y#od#x(Rph{($eB0#yfG{ zC|Z<#j6V2QM>kBnLeI*V(4UTDNYAd_bMkg{1kGs~Qj2LpY>crzr3-dj>`Lvtw^?L{ z4x~SGR59XnTb!J<7QgNth_*etqU}y~JRPFI??}Bw<#n3($G=oE%UNFMDRz1h#^1NL zkNjRSEnCQ^k9vG-ADcU&hCEC{W*ecXM<7cFn0=uOr^i#T)GqA#&j2eNJKhAZUY?8| zZ5Cod!$Ax@vX|kas@!pVp##=cbfr~U#jKBH{v`y)x_9QHstEUrUssRzGA2Er{2&Em zD0Cm`13RZBLjMjW5H-yW23GWi+E{IPe$o$We2c-F|5M7km~PzL^N)DWuGk~j6 zPK9B&hH$e^XmMFO{kgrp3P5#i6bN<;VVQ0}ZufOfR@a4R!#<0^t3DZu+g*gxadV*O zpg8eod_ohAXGr(QOE4qnE(vhRAPEP|!R_`*nt!N?jH=5g<^0-SaG4x5PSSuUj&5*S za~|kBC~#$som$_o$?+%Sn#ezTp6u^P`Q7W{@}(OYoQu;f

vK~#e->;L z9qT`eocy5%)z9{jJ~qqoOY|xd>%JB9#2fI0vpMW~v>val48!?1-x0g*gJFq*1R|6R z$Zr4UcOv^zGD&j1LGm8F5WlI z5kv0mmJRU6`x!(S7;>5N?YU>O6}Y*Id-*$i-ht|(MmX$$7D9%LI9#a3InC}B(N05DRtv^$J(Rn*OeGBu8IcMj)MA8e`xkIO9*`C1BX2N!YZ#M zXyM2B7-bg*fMsAv&2}q*j5BaHDj;)WbWB@e~G+gMh9gRu*;|56J9F zwkJ!nxYPxG{Wjv@Y0=oVyD=WSXNYP0VsZ0Xiq{6`qn2_Njw|EOCBLr3pZ(w9=yw{j z=VR3^s zJiIgrwy!mk)?V^sx;>^Y1zodi{9o_)LO1CkFdnj;+P|ul-XGot4JO~Bc{4L`$M!(9 z`Ed~SpKL*m;vBs6C>#51OhUchXVIO-#RZHb% z=I3K^j(#%EI=Ki1n>8;f%UqB2mqq__5uQat_-v|#&wW`PVEGd0rmPy>iWg{l`G?>$ z%ezd5fER+gRlQKgqL#2!yap0%RiOnh#DhM)hMZXm0<4uz_OAK+)I!(zasc=oHvA7w zer@x(vxTOqI0Ziy776-W<=pk~7M>+7VfVg+5N6`oDkq|MCX{Cm<`&o6wYEzeU-0jg zn?-9|5*Rp`o8e=?xfSLKe3BPQ#@x4*MzDSLiOD&le+zCq@q0OkKVo=+CTN%eHz<7{ zJf6$yf+)V)1I)d=>DxtD7>%_>0pw{s0QoQH;Cp=z6zb){x-|~uL^1~l=g9#)^Owe6 z=>e;MC{n96C)s;q`9#L=wU*~5=lo>#l-DCN%)29`Tl1IHtNGpm0Vi9&5RW+m3iwW+ zQ>mO)Bt52e?my+>Ul&LoW$M73Edii?HlEqG>z=qZ9|CVc?Ja1xGZdl%>qt>qKI503 zPNs*>ma=qo>1|vbw2|LyaRyh!#^Wr7M3i%j!m{q(D2h3W#=i*~h0Vp~K|AQOonxiS z+9nV=GKuNv`^*|D@F=Jpctx@r_5=3a2nmU+pf>#^tn`=(G2Y>@>uVI;_;nR}9s@`k zw?TY)M1wf3?>HKHU>SJ!iH4I!e~9(?J>+`3ICAIxG)T3w!`Ee%#PwWzmiJ8MK+@K5 zwshaR^WvRpB_uwsf^^L~Oq}+QpkE{>q*|_Xh}o_)RW1@L}&A}t-H4mR34v_8ZWj&ir@ol7}6>I(=czUcx} z8~vnZXP(iI<{gngAcX^?i?eUVkHwV3@#wGilbUam$9dW{IB}F5hIb5>F12KHts>8p zx2VRlVPv1_5j^s~gnIE?;j%x<1V2mK?$E-Iy}r@f!5l30Dt75c{3n}&~8q-(fAXtZ!R*4!Iyu_h*v&ia^+bu))Dzqq$iqf3qU zNqa4lkYzXLp#9*1=rzL)8!V*w&1)hJTTmpOO|#Lnayk}fdSbvLRau|vacG&MjG@Z$ z*i~~K9lt7;v^!cs?wI+J6=lz1`sxjEL~RQwiN`{;&1LBN=?JtdJOVrQcEP@?u5jNz z2_^(}hpX>rLHh^`>5xBeAd@_xP6th>b;CO8U2i44>DUv8ckYdG8t2fmZ3Rk?iDY(L zCFls1SlfFS)|jQ>O6dnYn>0}N>U3vW{lPlskG9vk$W}?}afqzB4YPIf zz{+8nxM0T>R9N2z-v_5)L4O5wDB{R?ljp4NyBZ9`Ig@LowzrmAcx$nFYEh&59#U5O zjqqhAdM;f?%6lyl>bhu*?5y~qb}6;I@J*T=)(@|rIYbk-@#ncbZE@HzZQ@|myxuI7 zwY#PO6AAeJ{t@^O9i!bjEU?w47F?Lw!xlPGMu$|Eeo*!s zv!0#B_$Bj%XVF6L0fV=Fkc)qkPBECkU-=-bH7(EyuuqR9|ABqD(wp%L_-gkYN=9C# z4@Ni3YvB=MQ5l^q{J#SxTmYGnD?XV`o; ze!y^_fAb|87CemUG`Z!E9utqE*okjH*Y(ABzkg9n)wB5JX&hhwjxw4n?>d+Zb_+JV zoz4HZ_-}gadnwa-Vc}jHVq7PE^!gUP{Mrn{P5`Z5vzg|;Uk>%!b4ka8XT`f8B|)N! z7U%u$7F^m=1mzB|V4|cNoVTSy*WZ!wA*2&GSWX6=j&$KJ2G_t-$#&?lv;*vMse#u= zbUB;zu`uY9h?8DZ;I7(T1+w4Vo~m`;+1pSOJ>Ht!p#@@v5vkPX8={)Abgk1(4*Tbdiw7n z!2ApC`SXzpaAU<7n6^^~0v`_rxAQ9C>QY1&;!dzs2!{oa{NYiM8?67m0p2>fN-f8J zhYcaxT$fp28sxXX2L+co5cB6YT-Ut_bKML1maq_Z zd_?fF?ZJ7!&Vc34pWvzAM+lwGpUXeem-S_GlWV}}!bmQnaX8oabU)5ONs)W$eO<6Y zGF)|-%=vpqS+`GLaoD7My25IfP`5;<$TuSQxho#InG8pbZ$tc=0N8JM6#OsWg$8RE zSeySF)}8CfU6aT$9hb`$p%1_QJbttnCY~|Fh1P>*)rkf&=U>h5Twd~Jp`MHGJK3V9 z^&48ppXb$dJOa`oec{BU<*dF6_Pkb^24g(+xGnW;3@6#?uE(99_#UhtiXrp)4hZq< z!+a;m62jezoiyIKGv@VFlr49DfhEOlW$TD8YqJD-|Ar$`>M{%a2vjyp$KJ;+vG?;V zW>br-f8!U6FGAtOCt$YuBY4ZTXLp=8;A@+-(%5H*Iu^efMU z{j-Op!;)Q0R&V&g;+=a`WFNZZqyFg2_#(Lx2fqG*yWXqFiceb!ei4x;J!Rvk&zJQN z>f7q4@F*>DH%TDTvvdWkbT4@J_xm8we|Q@M1o=Y^rp7NET&w#@@_3{lNXH#?@XmR#j$is-_bNox{uxvyO7At z?r?9N4>4`;1Tzb|f$4>!8=2j(gZb7X41b+$P?r9(Fr{s=;jK44;mMlE0!0Oarm`I4T-0WpVg>2siqY>yv z4e`g~xc=09d@kYMR#SoQ=8T+aI8RRflu>QwF||3YcH#=g4sb{4c}m z^vYKI1-_xOZwxQU{ToEjwcXkGR=~G}6}7OEayN21b zdN0rjVR_+ru2MXevr+u-+;EDUJ|5f5#?FjRp!FHzgK2iY(EiVBsUS}XcW3kC4IZbw zuz7!{kQPEy-@&Li*^#PtZSHq`c9@D=_Q=UXjVU%HdE%Itk?U0NaONpyG0X z_O|IrOHK7L%>D@8bzO)~28VEh$#4wlJ^(p;OPm*Qkv2U@M#XJEsQv~83~lO+x0+(9 z-%0hHK8YD*)RH9VoX>&xLqm9`GaH_T=|~+mo5P&!F~s5%zg|_}g?xS24qbnz(#{LA z=~mw_#Q9Y}aq#5M@ZL5%r!xD4bg6bEIt3iZX^V=m%dgI|ZQ07QsU1S`<#uoEuXhHQ zM-ISKy>iiyKZAH`YcPsJuA#jCK3r5wahU6J?DpameSIO5n7WK26Z1;RBiq?fm}^8F zT*7JlqRw9Y5|BxsfvsR%Q4@rvN)Res({N)7>q-5)q5nwUY7P5MNZ|_lR9T z8IBJ>1f%im2qZu3kpGLQ_?YKRH2I{5^%>T*Q;!um!gvXtrlBV_;@8-EE%1?gedhPO zmnIQ2zg{pnXCd_cAcL&(InXY21MKe77W94ggYno)bl%{B zydo6A=(#m4G33`1)i%-zbB@#EF5_{C(i{Fi2q$oJ zxea#TJA~E_8-zbzq|zJTE$PF~E6{M91$G~-Ci~ntNVe@!C6?rWK=bAIFy){M=DBm| z?A!zEzov=TFDsGimH}?5siAWp28b`c4wOa?JV)+l6oZ__A<)YBM(%Tbzsz(t@jR#v zOAQi<)ev9mt}z)pIdp`V^Z7kyPJ1D^%_yNgl3Wj8POo(cCOhtDP>w%`w&{jDo}Zmh zE>7{qys5cF*`0Ztfu0yrO}DFt zLS2d;YMod}rPwmS0o=Eb0p)~e(1yR~ z<8WLsJX*0AVx#0Z=eR;xys|5o@ZuX7to#CJY7$|fwmTG-?1vxgf~mJ=I~2V!f4dbiorn%(RdETbe*r6`55g@5dB=9`_`)l zF3c~7+1s0#%y-4zIhQWVoc7GItZno#>d$@HGnz{uJxq8O#d)wU+=mnc3 z-=A#*7mFSYFVGzCk|)$(0Vnz!3+a|;0VeA2xDo#^oaAfRoE903Hfk_C8{fhwY1}fN zTbc8}b%pSQ&cKhCTkR0!_1U&Ykjr!k@!3nd|Isr$o9zo}#h2G>bN8<|_kRTb#W8)k zp9812*um&bbej9{0=^~uk%|99(<$H!xP;DXwdH%?0_lqdpM`P}ZK$|~gBQHVw|>Xz zn8g=q_{A_BUUG^A2fQGs1N*_W13Ox56kXT5-kQFaaj`Xh%H5h=c0)p;vrgYQn3`Er%G z;KX_vw{fxT&m}{dz_T*sHGVNQWqM^%Ui9hdo7m9!jXpfY<|jn|$Yp8nt16>SJanIq zJm-am7bmyaEZ|${B`xV=XLV&y=B1!k#h3pdo#@!Rkw^`GNPDf=AmBx+SFY2(BiSB? zR#~;tqwf{-ZuXZApbvZ)&FM0f9oGfF$o@~}TIhg@} zSMbO#HyGaw@Ws?+{M?iCpqyvs_vwHy+t7R2DmsWtMR8W4Jyp6v@ z70%u#mqs2Y6HjUjHVZys;m~~_cu;bX3^x7(HOmLL;{F6{vAn^qKBQ~l2)z9`97}5g zv1v#1+VVu*=K8$Qyav^&qU_kr?lRZ%OV}y22(M1sC*)J~Z?NUh{tb9WQS?0tP2L?4 z@J#=|aZYbvaWAngI23mj^7=2oK+B$0M?RJ&#m8`$^&u;M$;ssWzu{w3H&mS& ziuDuspjmt+bt8(ht+7%jC(8a0>hpK^t8eTh+g9@!%Ix=v$E^U10X-C8F8Z-DR9JbA z+9__K_K%%OIafe#?Ad~m#`2JRxQN`3I**wid!=KP_d>{`1Pj6L-KNJO<&h%9@cRUw z@$*a8=BaRf`$pJ);TX|hr34LqHN{nxQL|xt*Ll(odzzRcmjXwKEQQzn$SXSJ1Xc_;lU~Q(2rkEDRuaY zFJ}j#<=abiOZae7``Vaa@1G+0Ry69D6i;}|W8%`T79&e$;8ue`?5EzD8h2ZPL%Qo? z`m%|bzS5NNlVMoa(H`yZ>EV5TZr6CmAar}a3}2iIK)ER^i08LRsdU>6c)8^~_y;|J zK1xPl{&gMn%sK_h8x!H__atzibCs;TeTy96&XfJp7~1z}S8{+P(3t6-`24X3nV{Z- z?)oy1?uqrp-X;@p=(9K2*|G<&INTm>LhkbW$l6K=T+}2+af(ozXibk7_M->KPsPRE zr{Iu9^YGF3P>d8GCu4?2VtDLLqB2gM=1%NRerxaQ~ zu?|T1J6$uEcI8H%?Z%Y^9|g(pd@xC#10lzq_&E(%(0F=;s_2H1qe-`@{zo19`<$`# z(ZYK{#F#nei@H+%|K z=Ig^8Oqd=el%GhqNd&_)T*bO&gW$YNPjL9t3wKAm5>YRL8fndK#gD1opx+j6cw+n( zHddTu@83Y9d(8a0daI^Crwym z@>A+{B40Z9cab=azn6r!S@Jo~7k?d`iDQG^@IlB4+~z+EQzkFL1&Icz(=mabDt=7U z{4R+ryYqL7zbaw%RN&KEG=c^$;O~4pJQY@GZGq=L1K`oiZLriS3jF&P!Scc?m^XO^ zglWY>*YVk~`Bf&|YWI|cUyp|SliBxI^89@xIp1eH1oCqcHdz@`W0%e_!L&;1@ctBO zT&zt`&1i$OZAU^zlR8WgpA`Q%tO5fKHeV4O7-?wWU+}RWe=zCe5=i!mS@r} z@>faS6%B}s-)x~jF_>)VGn2;i_puGU?DWv|vMS5;ctT<`_B`SCJ5tZsSVDNPxIeqRURXrF=_h!#0Wj}YolixCN zm#Ye}&2%mC-Q6C0+&n}&nWoZf2ZoWjWjYwMwhPL=mQwwOgLKIw{#-y@8ueJSm(}N3 zZ3~B~mzG?E&G9*096& z6_Cbr;89EAwo-dg%l}238|%r&1qqOCs{#%lx5)QJ4&YqzQ2g{&6xpOwKyDU}goDa9 zq(bXz&PIJda&g)ialGCdn5iTWDe)8G`mT>KvtkkCh_axaOcQ*@T!c>1Y+g?!*tc0f z944;VL_Suf63^S7uw|_XW(=t!%cfn+ITdJ4S75&M+qnbe>O47YXq!maADzqmB={rE z=o95K--!cfzapPzH}}z%s{KIgWSrRRV>nC-(1%58VhHQ6Aox(Ut(`pEXC+rFc3Vyyk67l|d(xX6vB}Q7fzr#h8d@RJ|XH(ENZx2G87=1zpVMEkD z($l|!hzqjmpsrIec-{i+KP+E5%jFyf_e6d1D5#nd3sXb? zkTs^w`)%UJHuq(xESx}3Sd2ooktQwmM&h$N1J|fd$03!r7!e)C|I_Fs&N9s-BEyc1 z=7Wp95KDwS{S6Y^o^8R@xtRGzevuW!8-%HdUBkMvbmxmB8Lx8xq*h*~2!A^4o)?2B zT{(?vTlP_jRr7faLB7lF=DUq1?n=kqW@Ac`z-xFgOg!eO8E)^=z~o-2N^hx?qGz=q=+8cjL7>~T zlbvs1JlUFAP|Kf5{j)v;^j01h@S@%^dd%)`b;f zbr9#s&b8PtnfP!yKKi_w+1M_j9P`gVN2yO+*@)}@XtXF4HI@1cIMKnLy>WVna#l96 z<2w@##g|g=xWjO#_$1t}vg<%yS8Nel-Wv`8MR3`Y}jr(t*#V7f7+mE$W~+7!pnSwWLLDm|j8mgvQzU#I-HB zC~u_ERzc!jZQjvY9Yig85b15}35(A)k!SJhaN7PDnDTpxbC;yShR$hlePbQGQ%i-w z0(ZFjCkC<<+Co$LDdN^47?vH{3e{HDU`w0Zt5Jzr;H0(`)@gVXpC5kI>x&+ZdU}$~ zmiZDRzxKGIo<9p5A;yB6Zdj|vx6zf&^}fqXCtMRZkw07PidmPN=Pd@_H^jMv)=KlX zf2NWKU$Vo+OB|URCVlw)8(q6(7G7M!H!^z98* zmg)nsLAO3Q^h0OtFhm<&Jr84zb|Ge4UqrhntMIph593Ww9EABh4AG-zW6sWiF?bzL z;hzFc{`~4XOz#(lHolYbb!rjydtXntG^%0uTj_Mzrk>~;AQL3CaBlb~re(jV>Ase92N{8Puq@S+pxt~k!I zd5rchJqN2#^Wznlc)H_<21buc66%aZ$Y1=)m8?v68~I?x15ebQa}cME-h>A(AHr7~ z0`cUl85kte#&TZ`jF5cf?<|z3C66+xf1U=*=l#k(w9~4zoCsT$9L>A0r2Ss|lA9Mf zSm$*ZIxjJYeU)n9Aw5M(Q{zaG{cEC{hnVeb} z41EW0hxwkY>?L}6-Z16-SMuZdS{T^44A%S10r|BWFzJRFOi>;QC-8~*Xux{GU!vEw zq0Db3k}G6H`U>facaw!YviCAYtj+T1tTm41d)g?c797KW6ZwP7c0ZU`=Lsq!Zi31F z-B7zf20SkXLCTFKu;jy7m}uVyTI8jf_=0n#JyZs%z~q4U#B$Sf673a8iXD5B2pPM( zNbs$sZuA}asZuBKNn&ahx%Q3Uc)uvZlSfjJH&o?TR6c$=KDbJl@G9Y);JXO-gzL0)G&bVT`gwMA44aCZ2<*kM>VcQ&j-rgn)}$;_n?QP`W?>0Av89-7?g+#cNb z(N^5C{D)9ye;1UlJ%iy7PC&}lAm}i#J?HfBIC)o^1iez{gMsF9AiilZxtjx6D6S@+ z+Wfe77r%E%Zv=FmeGzObf(S`g#sI~O#L%QC^sW!gnN&IpE{-%LHi?f|845lX{8H&S zpSByz|FdW07>j{RU(u&?JMm{SFO%1Cr9@Bv30k=~(naYTVcvppsiOQ%^dBCBqXP~y zAJ*CRdQ% z>4B1LqvQ&s_X(ZW%C{jbm&py@dKwn*QRVo}P;jKZ7kJM(4Wd#t?&c^p&M~_$eEIVR zl(f`f3jc4!hRI9d=aEOWeMCQ4$Df~j-;jrLj<>NiD-)x>>dUU1_hWgU@w2-u|7{Qa zrD}@rb8ljeMJX-l&unhBtLy0;ET!hH_WaF@r27vS7`Oci^Rr;T5F+2@i+|_rgBF^W z`1YEy5czGK@I7ypE2Q@aFQRt**@YH5A@+#dgUY>+;fb|=0-VuLb-aU5Yn5eo5A~Vd zKL+W^($17&PN1pG_Kl{jv%^gsG^(3y!iFw6_jC{8`LWDi(TPKDEwV+fE3V?mDVNc* z(*!2>`JD>vY1f67F^oD&PdYkb^n#1D?n|%Ma<7T574I=_o}(+iFj_jn{_nS<*)GZW zkxytud)A+n{IylGJozI%(>I5e+1m|$n6C4)Ra@;?&?TWYt=V+7HJyS7TGPEFdg1n5 zJ>gl%za;wro9klfruq=UY`eDX!rvj6eF+16IJPjv(CnKq06C*7@9-UlWDt?2|W ziVBDJ;O{HVx`Mhjw{=(Ll^}Zf8gcZ;0G7+j#WwYF_d2gMo+bw(nczWU*{(OPonihTICJ9+W}TV5F}1o zDNU(Qh3DG>z;$y+IMw?ijQSo0p-=j9iwl-;eQ#Al>O>1pvj@M{^5`Lqvr*(`4fo+> z?pDz5U?=WMS{rV>xIdS*uM-#XzAY!{_m8sY0vD@rG)++D!ziOAo}V!@v6ldM?!m>)pLN)wp6E?s&(kpG`=Nij(n*9c3L zHNiZ-SiD=ElJmBgnC`Kai7>Ck3}W^S0pl;XY25PWcBK%~K*!+_IFIrc+7rpcC9dFI z&^#aZ;O8@Zp8697odT}y)dajzbq%$ea#~?6xTPr_T49?~yWxuqJ<)q;5LQ^`V26j9 zs1$M!yEgQetv}vH=5vz23vm1=M(?lM+<&|B`#ok#FSN?==kIV5WPe(mL4K=gq2cM* ztUPk-j0L!)M$d;Vm?Xkv<34y`a53%f(Rv3JzwXie&$aUFOIh2K=hm%Nq905JhoHjQ$^-$Z%mgs@)I3jSDX0@%E3H9Rj{Ermg@%Wk$FBe27+#=jXjGN?K|I0VkQF zGK_rc6(z`E&;Q0F(_3(xlsCa`GT2xu@4 zWl^;JJ6o0kgY#0q<^CoLlkj{W*&W!n}V)`?Xcv97b{~J^o7}=;lY8d@I_8jq&^+A zZ8fcZbFj6KVzxXS2aG?6RW=z^{@!aU<=37bt~)>v++V`_>4N=}pKoP#<7(b+=r;8N zvA^w3B8T1s*~Ac`yhI1vX@lkSa5y?d#2tTd6e1QB!NoJdK+Y}))fo*?m9EZpYFKaa zV|*2LO7mrYe{YaW=7_Iid_|3PcTEH>e#gd8q7_vqbBY(AVfLRfU4xe*a_P_!YtVPz zF`U-Nn+9_~sp@4@oUPh}$+@0VOFw>`FC8-DhxCnMGHx5RjT+RekPDY8SsqJXCJR0g z2|-#hO)M&^p-!4178w^U(fD{TxRKUf$b(RBEDUj40v*+_i7RFONVra2tDjnA2yy4s z9MBC6k{(T;1q)mc{fECr&gOdlKLIDwAX~8xeig8DlcJ5L&F@UNy3w3TdIdz)U4(pb z3gbW7bpjc(SQT`C6%*@%1>h5NkbbyxLpu3Qp){kQi)?ZmDGpfkhX046k?g%y5Q%DP zvZzR}rLP_PK^y^=25qzUp%XvD{wgH>?yZ)oWUZIpI1|{2bzWEYO zPaMTJ&PNz8PcOMUs(Iae*}*1yTLE) z&3Zy@c4$IEhc|+524A{sS&K~3WB=y4%G-L)@5v|g?y_FH&*G~=r4~o*ZqiR}bkKW| z9QfAhl7>6_AjoYTs362l*58<=S5G}a^nGSQQd`z<5(&7*9$}0}NPpYVI*t(S9akvS zPYEAHLj3z!5TiS}C=14(o(PN zL)nftvMH0;_fX=uu^1~C)nl4fCJJ)fIW^D62x*;3=a|hk{jP$_s~-Ssi`cVA%6(8$ z&jY=bC(vySe}1URjoC4?U%ys=eeTBYXOQea+Jp|L3}jEdN6QLq+sWphnIP1Q);zp2 zu$ApD%@=etm_C1}RODDQRDW;GZ12&)?h<9V?R^%KAg#%`=pCCg6*cp)b6zZ+T>gRx zb_;Nc-xnI;(XG{<7Pt^=`<=$sx^d_#9f_iXc0&D@fI}QEHv7u#n6m6Q$!;$vq(wcp zEE4i2k-kczdws`3b=?l(S!A0L0;AUjv-H@;Ekvimo7|u2&h)=~5XH)AzV=}<*fvUl zNd%sd*;d4%ws~Kzn^A^9BNA}tKiQgu=z(k@rYo#qdRuq|n)CKk(QHw2&a4PH$mJ-__i|M zJ(HJ%=t3FPb##Sjm(B2bq%PxGnh^!69xMJMv-f#HB&^&I@{YYWv<*}UnBQ|c4s9%-mV103`FTVMX|_teV&x z=IBR~PgMZ(o-1){Ug>ex2NNJSQHAR**5JB+DF^pAF>tc(4;*+r7sjtRLo)8PfvY<% zklT4if_%}Zv$;4U-as}*{0HTG%E^?2zTvO_h!fQApu@ZE=yks@v%~q+AL@zy;H2_A zu<6sh2gc5S3Cr7o=WIWm==r+-T=nZAT!6pffAT2gqa}oO+KcTfTyRI|S*BmG;omT2 zQ3f_ezs3G;DzX-yf8)y!W2x}mWOBacefbZ6i>#)eM*mOU75FR;oI|@FG3b0E6<-YT z#&eXJ$H92lOat4j3diDG z&v4UEPrMZugkGh(cz#tMbQWE}sd?dKxko4T`KF6^?b{MXjUHl)d4@~Lgj#o6<{D~wY#4bmkp96js91bI=BE;^G1Q+u% zlD~W{)zb@vs8UCG)}|8#?8$?j89TvNWi%*vzDZ^bjIoHfc|w*vJulteC6eZ8{i6E5 z{;0OQ68A-~LZh+oaqEm=p?oAl_<3nG4J_+});JEQ@&Ba_ycL3}#w%#5Zu56{?eQw{ zvq`Tk7EQJ$cPz(}Aivqf%Xknx86PNaD`!LZuIop)OnpyN$KT9xnVdl}6M8``lxjpW(w)2;1(2U3BI=%ljRE?53$ICwu?+1G6joRg6?E={nPq8L5o3?rB>nyqyGs zETi4uh>ea5OA9hS-Bf;DSoj&L}#T-Fm3lw zx?z|(%nnNw?2~Y-MnUbK1VN_2&v>)Ek3o~OKg^Hn*s4pw>DIp^J2X}@*sCtAO&53= z-^PI3C>^3H^Lp?W-^@uxf&O-AlUS*855HF4i+or<4n+fB)1!AH1)LQ`W^G@K4$*4QE3Ivg_Fz^P65`Vzk9P9JpEv3S9)VUg`#>7s zhg!q(XLiUG^EWn+xrc-r5`#xbj*L(epZFYzqHn{%&Pxs#7af3z?YfXr@Pc%Fb{g81 zT!SyO(!eZpJ-<(5m*7i@Aa`!;73h^wL?%SOk=mLK1g)w;FX#ih8OaaFM?(iwz6uOr*!2C@HT#uKxpkMqOc$c8Zz1XmVOIX#H z+c)2sn_AI>dl^W$&itD0)~7?b)60~&F#aFUUP>yQbC&{;-Kzx6iQmBd^BeHd%>>8% z5J(v01L2PD@Yd}P^gNOdKObfgb>j$_p|Tg_s+8%?1NZQXQWDLX))|$<2Z^V4nu<;l z2K2-W1@Wa9Jy|)gdB4&^w7xrd>3YG}e)@1$YczE*-B10NwS#?bfsmFmg!#a@$342s zgoxGNOvT;X$B5BM1GaW|q06mx=rXlT5;6D+hJ2OBhQnGAZ?G4Cygnm667E2I>t7J~ zMf%>e>AA$Y`0b!3d-ij5fP#b7WYxhLaMHQXLSy4ax(~AmePIX-2Fim^!z;1&5DU`o z)GpY}|LZvD*GjNCkqtdZe*G;I_{`1M7`HfD?b_bsm2f^vu^zT#yp{J66(+)NY*yflTVtyx*xY7}O7$lQN-+IGwg?;p_(o9{GCBP0G!TAh+HR zBUN+taacBgt|tB&wGGxLbv9~fba@t)xBN);cHY1~QVHHzosRdyFXQ714OxQxKHRt= z1gj&?p-`SPUaggmvgG%duHw(M|2U0>8>iv+SJ62CS2jL-Rfl;?HlR+ht2okiD{kcX zqT01R#`rewXb^`R_rn@~Zm#>CwzBmv)?v`mi|8D%9B;?v;jAmk0-a>t^lg~ndJZu$ z2j!)JpPnaTagW28AL7K{>pUF$ICjT1o*OW{w~O>w`zEnM(kj}aMnu%E)e_wg{P~~( z`!K2JcGhN$x-tej)GdcR#rg2lW3t#dW+E#~As<5cp?MsNf8_`;_O4rXeFof_csO?H z1Q`D~(*lY{!MV#xl7d{XT(YKx(U8r)_r_T*%MDi=nOX@&ivG z!TL5l)O-#T#>S8nPMcU;ZvN>gu^ZjQ%HsR5{;mF)xIzR0a~yK@}8{a%IX^Dnw0Zu(M~v&G5*9BkQ~r|9223$~qDJh@dy$;m#gvS&2~ zvbhtDIf1Z!QaH3fcnTgq&4ON$??A0*9AHxryyy3*#7F@u##Ta^rifdZ`;w&fI|>>j znuz770{X-?79Z~u!JnxIaLO`08JBnp6U!pdFKrkuw>XmH!ObAO(oBHfKP~l5If|>^ z`bekt+Y9v%>!qgGb0Kn;64ybd0<+I%LQs7>)@QFQm*+J3|8?#x$$~vL%fX|n2kiX% zE$2aA0NhBQ2cMs}WqCg?;{QX!ulEA52 z$y>XQFkSUH%(w3jE0*nmic$w?=py1i_SJ$3*1ixlB^k*M{S2uM@KdTJVAJYxyu`pfbEg8c%!U3EAUk&IlJn+5?*fppp@ zFR{W{9WwluI+`~)pu6)6+7Qz`)}7?`6m8EOr}K*YW4f~!KKyh6m%ciS-|vTF!MA+0 zG7iU0UDL7Xj6eEsHY1ty3+Vy=Kib2rm9S~69Y*AJg8pJnjzlNW2MR`@@+KJ`rt;?j z-Ijurwm0dpzk5A$=9w+s586p{^bI`|f%@)gk6HU!VXhq}$NAbiI%k2_f|T0tRLs-3^ z{$eh=$Bf2Qg-dul^eXC=7h}%V&6sjR2XBNl&npVDgm7p@^PM18?N36cPXV(@fV-Fd zZ=5UW(VHd5{T}uWPS1#g)qY9vPJfy}D-uHKE_=9Ocd7LqoUR#YalGm@KIxw zhHX@6kt=ePYlm7-Pf(L#XNh0dZ~EqP^SvWm%}ZJPR_CS!J_Y=}hdZW_r^yZH0Gp2+IXu)gXe6vUhXAsHU0UF4(h0hvcUERDS-tU#opu+pM8an^!{p zk<8mYh<2WLRNxnB87pCy$3baywVt$BpdVhpSHx&W%;DD@o9`wTB~A+^*T3Ix&D;5P zp+xGz+rjhSRGx`b<{ZTWes7r|drIv9w6ht30!|3OUsbj83HXLb)eQGGEkcMT-m4So zimJ0zK0t-vH@+1dvvTP9&(ka#56eM*l`{!ZyFg}{cZNSN*TAk1iC|)|8wSoh!1te` zVea1Ra4WklSM`9ul}@1?J_j^G<;Mo_aFxQb*vHUCaTKgSD}`bFnZvfZb)c6Q4cl}* zp+_ZupF(#li0tEzYs9 z1a`H#2s<9v!{X#qV54~-&JFtn^+mtIIFj2iV*J9s~x8&9aQ}FtE8eX*t{5%-$LIKuXe9* z)$x~Jpc9GSYr$0e8jII`+*@=Bb}$)YjKQVdR^h;hhC=-p;G!(2Q~1=wNY;OCD5|Hl zksS{hjsm>F;jR!%Zbx6nugRP&@5*4=%w>+U=S429^F`Th-CNS44jt<7pb?SX(B%KI za79`6=moo5P|z#r5n_qTq~`G;op)P0D6jDUQ1&L^R6TFt_(ig2jgkrpEwm8#%*@py zQBjgM6={MW+-P64Nh>LmRH5z;c;oM0$P;0+z}6{?J9X#jI9l^s*_o=sSN~gQfqrxf zy=7s=>&oGsEVt=Ozj0qguZ%OJlkL*O7yhh9C}i@6&-$Nh-hG;uR(6G;|x9@xBi#yEXu=`Lq3Bb$Gh&4ssa$7_H5FhjP{%a%F=54KMdxMdqt2 zh|g5Df6;PsJA+zI3cBFI&p-b!dXC?5{psj@;d{8aQTLxbI9#5`19kviys#Gb?ceo% z=U;qIJ&dNQ((ck%%b3o#<>@%ygI?dHBXzIP!i~=x8tbY00pVR9fqP0b78ON2C{XrHz~8^3ax!{ zSGM=dkv?yhkW7mnbmH4={P#;meC!Z#h{xmV`?32i@$-#Hmew?7|BeP;Y! zcVIiPr1NCievZlw9BMJJD(SEpA*-+XlPwDHGM^>r&@x{ zkMpBTj@OA_e0qSyhgvfHGa}&lw$#PYts571;e9CD(v|L%n>7DcCydjB3xeaJ_FVtlw|8B_f;mf*t+~MhFRT16>leVjK{gW@>GqpEo3m)%*IF*h) z^^RD}y&{PZm(opIEj-{Ctva4@t1h|PB{g}W?lAKK8WuF9IYUybvh4UrCmX9`dZT3j#Ky!^LNe5 z@*sfzBSX-3`aUO*(^))7cD6|Cr`~*758q^T{cScHeK8ZM^(jUlN+ZPDERrJUaaTXX`H-*kLGHvlAZaRk4F4zg4tQU(Av0eu(5nKiJw$L{C&;n=I)`C z(|g$M3AFUo5iTug^*KaGp2+6mYmzSVFb>WIwRKT9?A6$aYhw?{*2fl?Y^wCH@=Sv{bX49p zu1&;TuyRo#2jr5waE49A|I$Nw(O-MZa0OiG+nx70@U$F#gxf7N`AR(E(p-p@+vg3+-- zV&row13e$S5=kCE=kUc`m_i!5XgU6wnfhIM`MJaC#Yc)XYhDmsc~qZ%TcCk!mro)w zL(^!6D3vC!31F-*&Lf%FgN|~U0~c-yqHa&4V0lp}T3iN@+I87_X^TbZgg`#S~| zmmYw{mp%%d)bM8N%y55!AptjN(`M2Ccv-}#msyE8EEAN32Lj56WtG) zg!TvYM19N8qiEmN2q&nb=M!1hJFSo$-181DT)zp1r;H?4L#LotGugd*yRIUMmjj$H zJr1=`E~0Pz3{h5lEV&Tug?`rUhT$KTkXuMGdb4{jx>lKn-u6F=X2%alsiii^eD4Hg zbTbki{ZxmlH?@Ep17;!@_2oFZvKMM}w?-S^?T1Gc*3jf@m*JlK!%@}cQnc~phB-67|&hD z(TOWu?%=Jlg`~pPmHrsw4L8dC*oKD=QSYAoGbrGAcZOTus?|cxTi?-^*JuWB)9s1Ae{$3SSK9y@5HT_S0St zuZus2%LRoTGoJu)w|q52u9Fkdq0LXZXIj98B8wBK`Nej0-TAL?K^RkwWZ(VJ%$W@+ z1zkhCF4v-Uiz3iACJu{VK-7>o4?R|uf(Pc4k?)8I6_=|IKgyVCHSFRBJ8F+ zo%=uGY$xtpww?yGI7_`x4U|rqxj=gU`rrMQ$+o%lgpL9?7ZG#8ZaSeH&CM)Zy6&U~ zjX9o3y%XNi9Y9P!ALr+Y;x^g3;4gP%cOF%b;Bh&+mx5|?z1_OAp9(nG_ySFiFQ3Mv zb*0r!S2=vK`M$sR#&Km@DvWf}N++qETG#$e4qxyrCzD!zlJDZr!7h!dl-;w?wXRn% zW$Pf>jV1#-b?4d*{Jv1}cF|aAM3|zq-LXGket8#? zo_+(RMnh|;+npwQZlQ&=kR2o4bIw5e;>Qr_{W*Va)*k$KPL`_6`&$GCH~!l1v7Ze+ zy5%BOs(#Mb^FwSUbxSeq)J3c*Ka@|?ga7>xF|D7>*U>9{KYblPwo6a0zH79e)5)uJ zrKeBoN>!7NQrAV%)ITMI>dz|R`EM7W_?LYi-$~-(E1aLw=BXt#>68ne@sx0;T=3XcQlXdz2gVg{p3Y<_VMWCCGb1BfZAQ1%lVzfongyYLL%qy z=)%$eDg~{n{ZMB)7neDFhPtj19E@uVJ~YVonEnTc%ZrEglNZ_i&)Qkf{_-N%KlPn- z;?3T((W2Ye_`Z9vu9b4UhP8B)HrD(dV>v#iRZ;Yka}v5eR70zAjb;V}I)evMj|7#3=arWI8D59lE=9>FO|lhPM}JfT72H8EEBm3 z-*Wx~@uv_Y*;wU}%$KS|=B z_u=(kbz&g0-X%tAOJwUqD*m9)enn_bW-&@;!qHDW9ocUhgpS5PsF5v^vm$3^+$AuWzD4IrMb$m<>h)u?=z1~O9qUb9A9knRoWqITI(vBRz*VwMtBEAR zTyjr#hJ{+--}T@P_JgT=^GdpPm?CmluEHx8EC%IXEPgyw4NXfr2j}*9PVg{5|9lgp zE&VNNf~hi9GSbJcQ#T5q52n~Px0`Go5kbOR{PQQ^+GK|LU}U;%JbJ2n4c_mYB)hly zs6taEBIJjcQkq%xq@Z(% zk8^0R>lZk%-~nvC;YqAIl#$11e>Qgc7&>*EHS~QJ&ilN^JZppJ9-2ez4YeSr=Vs^~ z8HoB8%5gdfh99mYPgHvG?WdQnPS5Z5q8)F&`L?>_)?=TT~tCck@1bfO21&p=7+4t`2J=XL5VeosBT7FO@ny2W=!Ni@R zKv6+dskWP=)oJ$J^@ACcU|0$0*OROff>aQ$4y5qn9 zWgPB97=vDI-;D-_mUZdd$y>hnO=N@LqSI3=I`!fCd$jFCm1?~2i|5^mFXtLaFU$G` z$M>H>!SKI6lTKWpdyO>t*IW8eJTC3K@$dNl54iYF>00SYaX4S6WrcrZ8=FniL@g_6 zQ05!@q}@I5E-cb;V zn`Jv*)s7ESNJpHG`SR1{rvFG5kapH3Eqx=36oa(p=sYC4VL*x?S$*myr$=!-Z;4*P8*aKZFfRlx^c*TO6+I^6Px(mMCWj1q@B(E9q#;EZ$MF($ zR9#>mc-}+?h0$nWa|Sw>mWg)nibFva(P+u-cgW$0t!UeLGm&&zPf`8NLS)(OAkyqM zNfi8UzDWI}lE^GA3;9&aiIQ8pi+p|=i@xWXib5%z`(_4rFj>5jr?jX2)LX z2+Q{nDQ|NWO`QE1o!q!g6#nfC%0F-)eY&=lYX`B)BssKdbPbXBYbI`LAIKBkskF!H z2)a`?PmOM9AO~+BC4r-+5WQnoP|i<)zW)g0=Y2m%$j*I=cOnNA+sM7xb1-Gj1bDXG zfEn1|9m_Wsz;`N{=;ySB=;1bdWLD{op5How#4Fk( z$-YC^ry}2)GL#lt&etVmbSOG>qzyjPkw>@M-BJ27MP&Hn08$$E5mk%NApM3)#3)su zvKK#5;=(h?YkdMrRm+E3N7E5pdJ^s~NF`MlPLi$t2Sc4c#+10GLsqsfHC;0rK8qWS zgkGk!?@MQzKR<$`;?JO-<`1N{Zu$$^elA3Q+Bt|^KiJVata!`)(}1PT!8kC z7zKyGD=@I!4)#gg2aAvAqK;3wLaeAEn`bH0U#~urq0ePIsYBn>Y5DDB<>eB3)hL__ zCdl#j-)ZhBRc%JoBY1fGqOsDMa(~~eL^_pp$R20PJrDP8IZ&JBk#yu2 z2QDw>LcC{hzKko6e<&s?!oU7-?7trctcsv3hwkRf%K1BPWrF`{Ju{Cm%n^!E$xI`ZPk~qJIHqC1>VwsyyH41a&V4D5Vn0D-)}iO>w|Ot zTNm9b^f~J?Dj5}p5+83s&5LKFvWc^K`PE~5x@i6*kLShloBp_Gm(2Hj8+kj##mQqw z+1gL8{NDIK#5Bl`NaLcBo0l=F8R*69V%4OAuC(*_xe1(~EkKj+KSDp~735Ua1miZU zqtWA2(eAK$xXkerIyue&%~+3k9X_dP!Slbjcj;rM?8oUS{&IRT+H|`c(hQu6UT1ib zKLb_J_u@yy`n~MDLB$FztmgIObac;3g7QPMc-`xV*D*o` zSLC%hi}ZRH06n`WpgWTmqu^e$^=s9Ok@BiYbkp@7T$cC(E*yQ9m%Y>R15_@1gFmk| zr{2%U(BMoeuRmek0B?S-_me#1w5Vcgq>^_OnW6= zBZceyk+;o2xNL1AUG!`gP4#>W?Nok~u@j5P(rvS;*CI2r(DVV&J`{i^#>ZpP;RCSb zdl3vJ=Mh>>(VT7rQKF4Hx_0w8Qj6V-6fL@=VUJ4DX4#oFHu(YYw(R_(As=KuAX(c8 z?pM;a17!Ot0=i4(D)!QrSA+OIpf0zRP7(Y~b9gDL$9VXl)Cp*=^)c8yJ`stgha$b; zWLSUqJshvv<%br_C#v!7wcpGJ75&ygKbKAC^BkYtTaUWZvb?};E}_$RMe)43at=1K z{0*1q)6?Ehq5U57XITh1n8n1w(#IKJ=&}_4-Jiz^3e1K}5)M$sX-9dw=!}Feel|UZ zN$YkGmQHK5;Kwhi*9NJ(!bm=yG^gLq9UE8 z!uwyuKNB(i`TFm-9T|?`b(Fqz>*#WFFnK2}AK}p@cjmE0UFodu2lzCteMLLGA5z!- zy?Gfn_1k#)V7dvPcYjqvTh2Rm$>;bR7+mLQ`Fie~F|m`U;Ce{^E<5^Y&LxNAeG!NE zvuZgXbLq@kI;eJRGP-+GiQN1c3)|pdKZHPbPwDkD#tIYg9crio1Ig(B7IPf)8c86~US zMpFtdL&vo4BHzVFQC#T=bih-UF};AQosUs@2IxqlhT$=`&;0gdqO zf>Y>J*b8!bo-%5>EkdjEW6;DgN6{~{vuI;CHBrb5p{QTDhUiJ-0uik-LdhxSBJx&O zl<~P8O*W53YQEg6*VGS2o5F3e z$M-1QUffRR7iIAM%cHr4*VhjXppSIiIGEV_t`?nkdkFbh@pnIBnv)u8kD7=~F0bQo z1s?HA2v4nr?N1cA{wMfP5OcJ*=637slY$U6TLM3&1Fk%wa0k0~LllkuF_&5l`ooNR zqeXR_8Zb%y*L#y-@9o3X!GYh;$ng`@jL_tD`iBgG|95#F_BfP(|L5g?zPyu&ZZ4%- zkF{j$u?@L$L1(zrwwd<0G>G0a`|Bsov^d4nuQ1N%<4fzV@bUZKH@fH!wDG=K!OO7g z^c`xZ>cwAq9EU^NeB4+07k&GJ#}_N&$<)YX6cYZqOk%+;(pLI+F7!Lag4mBA&*OLM zKhZjwKFs*nnj%h5t`4&EAwx-eWeTr@l@%qwOMViCYe#sS{|AT1F6U zt9^;m;ck3hSJ#wjy}p;5y9=70orN6ktG5S9xtzRo>&l+erz?i>`t)lUB>k*9jURu; zq@+q`zQ(dU(N9R*Zs|+UJ=-pQ9kM}6aj-P)&_e0s`Zsj+r((Kq{@=3)je2S!jz1Sn zUkXH=3@%(L{qH@R{bHbK(hys|?8jhhQQdi=Xl}gh?3fv1pu@rNop@3{<5S^u}6 zjGVQWHmL6GlpziY8N%CZ?}$qjg*VW^jyl?BX&$XR_>7jszN8f;cG64xHKaF8lc|+< zZI`Y|sWLy`UwwF;bj<6zcs5*e=qi&~U+AiXo3x0lgMbS&bwym7qZJDt*YeL32e10N zq6@A$BaKEj?4lKmzR^nKO8TPtI!!odA-#gmP)4>V(eB&~bX2{XlPkDZp+pbm@oBNZ z=sul(!6MQq>8d9Wd3dDp zQhGS|W>TWwyP{X6sGG(+ou<^?}zJ8*e%XVB1n!PTh~Ur=D)(N#wd zXU>Jc^TSRY>7rcv+?((FJddb3g}h#!<-&nlqQKrmc^Q9RD2gt%FBCaR4Mi)K^8JFB zbKzptzjRoe%Fq3~blBD^-=#xwQ&&Uk$LD7r#62hE!>k@iaQ*U47 z7ti@2jnd7uU40XKcU5n?dzuz~WP~_ePRC9ixU^VKC7%S3gJ8^Ry{@|c14nR5_8qHS zt`>UTHl}ye(@=_9ChR@l2a0pbV7}4c_bb=Nw{Us=c6eg?Vzf0V9=+KfiY5i6qL*v+ zV9OB^&$o}<9b9M7hQmh7d^Q>}&^lHTUR8QXQX=Efj>3INTXu_=&*rme>7G`xgXBY$MX{o?E_u%*AZorJ$qael|9yNH4kzzKQ5osxio);wi3y;G;RQvP);>W?)T|d}~D{wk<06w_84-F4< z>%zTc@Q^OePNi7&4H*oN(_w30()ss$N;UgtfTf9sAp2xFO}*hr_gxQ{-R&@wc8m6= zxehz%&ziqu{^ILmcxb{wr1EKDr+o#So?I|G^q7~OrEU1HF|o~BLGe+Iwz%ZI|}fA{MDFa7yr z@pMo9P#UwQt3U9%b}Z48${T*AH8$4LTBEVjr~`AQ9POlOCel@})uj6r&QnQh6P1>< z(0qw(->mh1TD$5n(aBp+qi#N;Mu+axw#8$pUV9kbe7!e)a{D9=4>q6{u#tXRTSG?n z2&M52AE=s}7gf`Uq+36pAnhAucRd^~rjBc8adj2g&mBjEs!3El>P2TQiluW0 z`{Ahlze&eX72cMdPDvgLsPDF~;Nh<$$RbyhgA2xen9RpqnhWa#T~On(x-J@yt}`r- z-$Ct1zWe91BM=q6BzJUWdshY^`bp+f;c#7ZA@^P*zB_g|$=%N1;VziwqNIFt?2B%K zt1=<7FwI3lhN7%2%l*&KP#_d6T)WtB`G%##mn>Yra5xxcJIcn|#>Qq8V{6SgIoJs1 z{=ffZLdJ1A^eN0TmI)Sc;r~wuoBwYeY^_HMfwhyv@c&mG`mcV+?%h(yX5{^}8`*S$ zZGAErZdrbuX&8aQ3g2$z%$AkJzf28X$Sq|$w(emqlb(Xe==-o~lM@@8dXn)K$Jw%D z1mtmzGZ<)_1_Eq?F=O@z>S^Y}V^1_ z>4^jEAj!>eV{gQrD9ig)1TG!*$N3>+fL>KEaJok-jF>qH+%n#UvG)yI**RbA(&{{L z@9G06cyq3+&Hp5H$K!*XNWayIVBdhvSpI8&D5zM>x>i0fo9s1Ym5;Kv8zFNh9E|U)Pv)3uWN&X#<oGKKR!3#z4)c568DLy> z0e&tI1e+r6;Y-i^Ks?(3_jno)@7Z}lm8V&-Xq!9S{j`)>d@CM&Ubh1%YUsi8EH(T& z+=<-nvxycF9DU{@odG{HanNPK&Uh3 z56E1+xvZg46}&Bx!<%V2h05ga!39T{h=#jc&G&HP@x176xS z2u#+O9c(k=1M}>j9hvr^0SEs20SpJ`fr@H(M)*+yH%+%@;zr~$-8NW|x{elByW}R@ zP#Fv#@9>AcmcM1SRl5`8_#bTe_>*kQb_wXK)mymW`UUo(ekwC9wkKKHvCB4pKs)$s zsST7>o`abIBG|LfH`w~X6lN&YFfYFMAlbX~;j^g$ynkP}|4DY&NjbaItS)o;`6Ood z!=6ITK5ub#>WR+&FP7C$4BMwMv8g@CvPNa3Wv~TGpUq~kX}E*r)Ja|fKlm-6>|A$!Achb!uosL>*ig}U zcFx%wc>A>v?D^B<$VBhQIJnya@V^juVn^(OeUsk#T*Zvq*K&y@Mpip}V_DwzoUdk?^6P#gjn+XGs z>>h|2+jn-4TV67DaRT<%_=9-K-0e7frwe;++iyGJ*1MA7C(YSpjaPO*7Kqv77AJvx z@f5uO*LMbO7%dE?k8H{uG|8B}C}93rb~mWqE!=2fMS4z|Pu@0ofyYk-xH>wES<;@& zJQ*1U=GA1wm;Hu;v0+=W$A@F&r>GsfzcM0qHEv|`pcl+L+xH-FSRssZ5rc+`N~W9r zB(khfAKq*l$(->|2JuN7K-tUnFf=!u??2i(o7wD-`y`o#58%e0Q{eJN6%xDrc_6#r zP+?T`A$EsIk$Ir6%9g%9%Y@D<0Ay+t{xE+hI5SxruFTE^QF$&{_^b~O^*w|U96@^a zJqeeE9R_-Dr`T%zSXSmXcRaIETg-xx9;wX&^A_#J&fBUbhCS=>%%>?hT;79q(}-nOioQy|``bo}6X259$~OEx`ai*12y zG}0uylAM^a8x7g1PnAiK;Q{?;J6j|GxOy=%N0jDpG1ke8z zGE1f7g)KF1IKJ`$`^57n4q=Xv&zpSkjM0Up-`<%dVSpp?P13^4hIV6%Zx*w7e=YK<{pT`#ralQ8286IZYyN;AYH>{1L=$l4*(&D5 z#6*UjITR{Pal_`7-QgTE0hLEwVTL4S0@cD`^5;r8c({KsJW(VEB8z6R+g2>ZZx613 zbm?b_f{q#47WExpJbQ(Crrd^q%_%5*^6fp2@bu(;k;6@zgbmwUan{%=cx8J%D_7YI ztk?O@NVu5s)EJmEv!sKxV6vqJ;a$s)E^(r7=fH1HZE;*Sg2=Hk29X2%FTKShpU zk4enKeU`%T+bk$*bpRbxFP1$+0eIRjGt!ts zcGkojvVsoTS_uVJa#DX3LA^z+$?q@txT!u-%AblKE}}x^Z5^z|%YN9qx(9kXq!j*o za0r|~rY!uNlqvZ>V*6XVwDah^~(G_w)kWqlj?F-qW)ng zJHq1viB<6?Ms?vrpZWI~@6{MKKg$E-zf^5YlAwNc}91^*2OWBm{YPdUVba1?q~YL+Y?yu%Ig@MGG{IJ zC|-cyC5|dvJF+j9+qDeLn_?(5vN_D|z3T_UOY|Vq?16uum;wAZ6@jg9)JcNs4dLdM zdaTLOA1UDm)|XMqZ|-Gq-doa9_6*9-coGM#=h5mceu8+39xRHM%EzcSSK+ zv#UF9dGZL`z1WUT`|f9kIxF#g`9dp%%{rw}cY-RE%N&QFecFY`TPMS-$KJA1h1J+s zY{Dv;OvLksE+diMo4{Du56l=@%Pdd&h6j2t1luP$gJ09W@wSxNlbint=Dz9+v@;aJ z?+YI!`@aMMPq)WlTU8*_dLb5A=bwdI(xxEM=!%8xG#=To1L|JfVtrxUj5h2{Dyj-gw(#Et?f0hZ@5^FlggN z{9e_HOv$?sng{iTuVxZZIjI;|ztaUTdh~_Wm7m!Yh6ljF`r&xgv7T5>{E8jc(vE}D zb%@#IK}?%LAG@)~1-L%+mhGwo(YAFBKiD36gXyrJK6t&0fK*$AvvuJU2s62gUDJ}m zwDkIhKWfbcf{PGqpB#<*T{4uMPay2zvd;`!kzKaPR1KOhIn9>#bd^{Rab!27iE-_% zKV?i!3eKuJgJ1t#$}Z~F9ji^6jGLx5urKA!V8X}U&^I%brEfAwB8&hAr{cgnA5{{V zxtv5VlP6(=Gug?I8>=hWp4SZIDb+xt8{#b@L_Z|tFQ&y zr8Zswfg9C<=GV*kNlOwtsFyunlMshDKI%r9n0>rvLtUntcd;JAhFo@nPJ0Eu+o*|0F zk%<<#?#?%+#4``~n>-9V3~a-mJ*wI8Di@riw3%TVUgMEFmAd}Nekgwf^B%{8T_5Ja zvyQ+zHuUGHSmT5ABKWQQ=fvj17)uWamLtq;Q{FEa;y& zj9D787rGxQW9qLFc1f{>d>ErjKAOH{gehi{EOCU;V`~nZKV}!1b)^Ll@!r6gz;H09 z*Ktss7>;jDIfYLe*aO*;DxBiRk_p@7Xhzg>JSxS7f8HZx?XhjqHL|8QQP_QY4c6^b zf^Y2~&Me>O3R>d?WosQB!CgBokaAI-y?R5!+UezDFSF60$UuPSX!IZ(J%)g6rM-;z zZVjTbXc&2=oWnGJIe-tIyoSHN_=2ZS8AIrj8EpEEV!PNA3-GhI!%6@10`O|yET*TV zQ24OF0-kf}J9A^6DbsVLkEFETi8T&8!qS=)5aF~6+{xa{w#~2x!6qwk4Ll}naD2_~ z@wkas>B#lj=)*M{uR8VyFJ=i{*0RH8MFvJ1B>k~jU`$tu}7+UkvBQox*IX6h&4mu{;iLEfjx zSiL>Oj#VNd+f9MW-P8Ei%9r@}8haS~A`k~msKHm0mVoMC5jcE)0#LtS0(uyR!y&30 zanj=>I55(d;3cnEhl&Uw|7|mVd-yBcIJF#j%n{)3Y%#M*ybe5{34mKzF{5bx94=gb zo*6qtmEHPeKTyqEfkVUvtkUH)VWqAqxt^v&e%bB=!(uC$->qg$(Dq5pi5pE=eo0?& zw0}9+oBIpD*Bk}DJDde78a?ogGj>?(SrLg`;&2V>#gqo7=NCml{d0Ago4gBQ;) zGMeMP$Qu^p`uIe6)8B{OlaFMz_h*1Z_m;2??N9OXeZ{!fvlZm-gAAOvjU_F*8bnl} zPTu|$!^DGq>6kw6$lYm=%l;ftgcEZ7i2^)r_j_Cmwwm9aJi2fIyDhOGquOrbzKNw^ zo?#nqmfj={qs_>}_?}>#qikN-BqqZ%XOTUlkKp$|ud!=0A=V!=fh^0P#=4Y_0d{r~ z*lST0-c~!1HCyM8Z+v~psx~(NWpVp%*%FX~ASZ%PJd*g_DcsDqtqbg>Rt zFGLZ6te@gslH&gui$-iAzUZo?Z+pAY-TpmDJFyODU%JA2CXECel{1g1n1Sl;rw^uk>lCSkASc2$K(oTz4c_OF#)8M|-0=eC z{Y<~e1Q32qk=?j=1d+c#o*X!sgMIHTWo4T;u~&15ByPYD=3T`c*mJZUoV2AU6y8fA zzgO)f3ezuwh)0J&jN~hObDKIiZ+w)QT5E{|TlSHT9wib@%Pf$a`IrrOb_U<>(SaBL zPJlrN>tUq+84~N@04}aP4a%QsO4NP)N&dt{(ogOoa}2G*4?k&PQ)3NsbzEQgXzpkn zKl}u{>*ami5~+%ea?FvYSBh}D*;i0IHW9p!y9m@RmxEb5y5sK)(us371sJHWf>qw; zGYT=mL~qYisF|^ZIT3$_T|7Vze=n)UIosn%_@+{BjN#`#-1y^Dvz%0WKLU#0?s(3? zB$6313OZa>;s1A5*FV6YN6v(YhS(7hydBux+Rf^^-NZ$2E)d}`AMz;mHQq4eIw`h( zNP3%Cvu2_hY*CRGb0TaAQFc@%*QTnFjOXPT3SHUF!}j1m8zVtNsybncli zPCz|c${Osviy!RNgH0d%k+Tw(xnk9XT|MfUn<-g#vde98YeIkMxOfQjaH#|Q_y!We zlWyS5<{-AwdX_N$`z1KFtsYeEsKH~F2;s)Hk-+4HCH_oa9{hX|JnDkEr{&qe_Fg=Ct92bt7q( z>Q=I{`(2PVVLX(-p$$oP0?ZZ9V*2}Zur*4NI8g2zyTFto3onf&Ph*qV>th5U|4kjf z9l8--R7?Zs_y3UjR07~pl_hY1qdM%jP7S`URfOlKxiOb+eFjXovtZ3?12|#r8gMdh zM498b2F7#7ld|VdF%mho;UGNgA+{^ZVPUxdJyH&tc#diFx6)2{=^SS64f(w!Qa0)8&FeNe8%W?upD zUi^vq63`Q@kJhBHQUTjKY1{oA{1XHgD?d`e;0!m<(^=JX}H~t_3zmc zRvLKjwnF0m#89~JxG{XQ22oW-Ejlmj0E5yZS;J)k*mFiDesggdi7<}fzJUoImDIBp zMLY2F-=*+!&^I{!)=%iIYX~eoG~wY*i^;y2$t1KP2LzSO!z%6@VD7|k;Pqt{;AhqX zR>rq6L)$O$?_BRMje@iM{^frbr0tsz`}D<7vt*v^F8iL~{4WDWZiy>A;@J+yl*QZW z20Uf<7i9oHML#y@`&VKAUuj^{jaZPf=rsNrdKxy{Z3V`@^01XSlQp(YAX{X}7A;c& z+r7uK+ZF79Z;#0^w|Nc>8nu_1|B(TUz7>NDCibkiUnaa2^a{!=NJ;!feRyZsXnyV$ zHKte+ zd2&kz43;>K5a6ao{5}ssJ067@?_nhVZUp(M6-a)K--46gUS_XsD93S?J8%O%heN)+ z!tSVwng3QE+_No}yzJPGx6Qv%b}!l(JfGzXeFU;eDzS`{8kI}i;M&f z&t74t{4+qL;6bWuA{f8SNL)8mK$aGqX2MMz*xv?*M0t&vY?VB)v%I2C8Ybj{nM3qp z;Emh3uLcQNk z+-~%d2F6Iu7UbRO&lJweW{&Q^ilf^$uod?v088tUMJLLhb>(~GJZV8o=Laa zK$m22s;D1Zci=fx~8ElS&tZ()%1&@1g0HVwoX2V4t=+zcVbe+rb%iU*K zrKyQH2LzI>i&lYMS3i=8gLLrXN7}gd4P?jnJI@$4A!eIhiru+|2f)&ZR=aP$?UKW_ zPgv{hv%s+C22gbi0!kKA9F^)wtXs_Z|C>`Yity+f2zFJ9*~&SO*~-Hj%;*{=(&6#~ z5Bz)>#IJe6i0YN?`bLbykFFFk5407)nv;`Bz|r}vtzkb<_wuWe9eE>=05h>N2Ipi(!VIeOTHE%sjrR*NM7SDq!Oy=Id^ z!PCG;n`hvlc^V97_p+CMe`g~iewI~6ZfBlX>*39!KFm+!?_f@199id?2tGs#n5+}U zK>Laj*f8rWD4!CBzd7||(wwIN2ZhDV*?so#TX_UXELDOl+bSh%N6ip+G&-=f_#it= zP6zL9cPHC!VQ_Au37+<}8*@6dsLb}d0o;>iLa^Q@{PycnyfN0CIczVR|HVxu@IyW6 zE_EemA~fMFhg^Kv^Cj=ke>K+#$d;%;U&q04$?lOjnJfqIqIZ^6k$HT7jW`-cV%H^e zZN$xg_z-uwFQzDj8T0iGDen`8n{UOieS=@)<@371Yo;w=t)C&&b78q%_2+D+?L`~N zsSCo7$BG%Nl6t)6oi6dQ&&AG0E7@1(ML6buv+#%03^vzeq~ziO8|)XefP5ULL$u5@ z@eI$c;L@u^tgn3-r;crAS62wh!Xw+UQ`tRyTSwe^AOk}>#)ofeyyUM@6p8xAt-o-qFN z4@1LNHT-`673R_@MNqUNnbppggB$cpScyphiCc0Uu3E22p2VNRnR1qN$>$lczvUoi z`{~=D^|%dj9DI#@??@$yuX+>l%LhQmY&g5q+!)q=Ou_E`9NvG{0;{|{4N9!=#Jy&sWzNRcTT$WSVUd-l1CsFb2f8A>V9 zJW!eknPp5uGNcS4X}~@E+)6}qQc_8yqG%p8iGKIv_x*j>?+@#)Wm(pH-}k(GKl|Cw ze)bO>EsD951{clu-7kblIKKU{VkWva%4N9lG}+H;Z(C(`^; z>rCu$@#3xh47mTTjqp27$cnlIT>rKoulrHR?`$r@9;Ma%*g|bSbKo3kUbwHKaQ|^4 zquGZ?t-r@kuXo3`k8|O0x;b&y@Sq`SXFy-aO>%Eez^wu!t_eeNY29{qD)0-a`>OEe zHJMyECx(v{214?qPzm3k08p?19p`L-)ge~cE^Wd6_2+TD`~>R5)iBn79?>~+hIAdX zr*&UL`I!}4Fn6UsfBqmt)W>-Ov<@_YBS~~cFkJ;(_5Za-iybFPsWNzskZR}#cz3> zY=DGMi{u+q7Liod9h|&QuxDN+3ws~U)gKfR$D8te(iR~cJ*J0^Jvul-HymCD zFC;JKDATHKdDMLCBfN3!BIQ3qXm#u*x?QY|x?5fnQRX{{UFAsWn7%^mL}glOWD4$o zYslXGFvxeY7gxUAF79_}k=P~CoId%tpBAhg3B$+Fp$Qoy>Coh<7%k z88e4a!@gTt#=#6&^!{=Me^Eep&mBkCIkaN8X$3wwI-P1iL%@|Ky_jWAj=cEEL}y+; z(3~IrIv+0_XEx6!4&i*CkVMbaqN#Ibaa`3;hWXyUUNsRjN(STk zSNF)D4JHD+G*9#jmghbl>AgAPf8tLcm+la#Ow>a26OKH;emU+xxJ%LxRi#PZ?%<`T z!fWN~L1o@{U9gNiK;?Ci*}uQy@UC~5dM^!DhRf1Svx!tnR)Kp6T<8l=B?*J& zkDuNTgJ1(`dS0H;)APnqpJx)i+?Q!I>DUhZekKN_mc?RO<4RILUj}~pgxZ+@>yHN~ zt_O|$VvxrGL9i;p)Y_NCY{*-Xs!{>t%vfkH8IShvc~oko2W{BW$_k%FqUY-n$#bto zb;TpR!?FxJt<>rDTN_b&>k(*AjOU&I)cNr>syKh>6&Nu!j}Cp8P0b2@`AwfKIBlN{ z3|(7@rn^(1;Gh9{QqhdMat}~rPZh{H^65{<&L$V92Hw5O_p>2DAVHtGg(;O^96;71&NWi<>~ zWlfK7c+ZR%ZKE=yWP9ry>z?UMumeERb5u_xG=o+qGSYyt^CrSTuA z;H3$}KxS&1K>wf{X@OhW{{sx{@XSm;mgf{ZJM>S50WOM^?DQ)PZY7D~pdTEWz?K8}Y;4 z96Y%p3%}|{k^=|i;pGxlIA$WvOX?SK<@S@v4fo-)(g?Ql!d>bnnn(G7r}X4IHM)L= z7K^`nm8*{S7JL0Zj#~~VV2Wl6O`H;d&#ekz!|HvoN~Q^Re0)mZjQdJ&A1%Q4zrfmj zAL5_0QS^OOJoLS4gaedcGu@YIWcMBcG^b{8!PfqK_?n?08j6=-|Zg^9#0%F`RW8ZE@vxtJo$)g zo4omJ#j&UrQ;nJz=ds<@duaKA&!D?97D}S6>DI$Qf^$Z|&4^sGWMw5WN%_X+1s=v$ zb0ePWeFzR3&Z6SL)2KK3m??`w(q7%Qc>u@Xl7G1ArK>L5pfWWT%>~r`G$iFFuyOJDKsQU^BrK8cg+5+K$IU;K&~(xb{suU<4d&! z=QFLKJ*CR}vvC*;lTl;^Vg499c`JVv{0-MEI*!S4O`_yG8Pl3l|`T?g@3@j$kDuq-Q+c18U{ZJcmGfnPSc$xCNX z=0Tq&=f&_dehIutCm6W$UDv;h!*`o=!#STtV()6aBilmO8J-Z`I-kZzxQTJQU?iU# zy9r1B&V;-*W^nh-2doKm;@`Wy_}dqcukdJo!!W*~`3-Yk z`wMG5^Z4~0H`vt?Z52Y>J%S_24T68Fsy2TXPsRl&M)EyX1>AP}d9j_*46*5WPxN_v zg5Q{WhYUzl;lG!n{^8q}+^}^>l{%(|0b-NdP#61dzrvi>l|BSW55eD z>e%ajgs;4|lq=Kof@0ZS_`1mw`bpcCPteXL_e}1PWG~wa$C0;0+x_cEnH~}F6kB%S z@g~vdI!`8E*ucvxL-~5)2TVws3SJ-AvWg9c+;^a)|9r%p-<>Nbxvsk`T`8JL$EuMF z`adwEJ{7&cSR;}8&CW(2;BG6F~ulJZ$4Ep zqcehL#aQv~l6$zQtC6h>l6dhBs-iaCzOeN~I=bfs^2hovSi6cqVSs_nhtms1-UIV+ z?kGb(exNMA2|tZjp0M6Jy)P_4{2YA`O7WDTx;An*Z@}vZ0w~Qf;ZaFx6&viGd18$nUwXXBCTvFl|J9|! z#mP_kmDXsodszaQd-$Tqr*K{wCCh7e=d+|fMZ6w2@&)pz`H77#qA;68JeVWT2P`_v zPn&1(%&Lq0dgN&^$s5E^$E4%Irb?n`YDYxHl}m>%#4r&LQ8*YciucxOSsg71Q-{ikz9iPuxSCD2{ANpRS9E%$L> zhSLm=veLKuQ0j47boXCBA~U2LQ|A?<`qN*_4#?R} zhje@HSQIZVTxlu3{>BMYS{Cp@c0T-fgB16a&J)!>Hp0OJ&vT$#NkY>+`t8|LxE$Gn zi7tt_?adWpm*>Xg&X{+a74J(h=zn>< zYOfyKn|hSREt2$&UhU;27|iV+mP5YAWuQ5d9+>r6GEh4KzW&Jrtvp0kWAA~O814@~6aWaM;o{fIeA1 zM{xdd56n*fhbyQ12{Na~z+>Zg(Elqy!{cG((Uf2kd?=4aM~y}Im1C*5sGr~gaTYne zJ4UBxje?A*A7B!Bw7jF@J9Hm9Medz?M6h583w7O$3ZrvDR%SP2UWo56u_WX?p7NT_WhQp8C$eR%%dQ`G7xfo(KeA?nJX|D~ zlS0J-vVnDAd z+=1Qeq_~`i6ohB~hP@W0L@?@>Xn<4$v|Lx>n_M@7>v~`6c0>cBbkk^M)hp&R@|$Sn zgFIZX@Qkf7Q05kAb8%51M=zsWOz2oFXxo}Xum5Vr>XmzN;OH#&ecn6Ki-mFYeOT%w6j#|~V&BA&Njv*)wlD@u09{J>MDM!4s>CQPhZ#_qarMVMkH*%=lJs-n*F zh5`E_bd0yqbAvmXxL=7Lw=yH+J`BVlmBpyj?<96Q{3Mg-oCC`bCZu~;BAf4(&ZZpy z%0%0aQkl(Ce8S#N?Cgre^1X7xb26paV%I}bmdv58ZIc8$dunm6(=l}WYQc62=A&DY z0BY6k|HHTcaX)?j`+Q_p#F{shVDaT9{%T(vDL$x2hWxXl-8nw^x~v~GNjw^+^)nV- zi%{(4*>%sSLd4-8g8BJh@O6i+aKN2wFv!(VSQH$MXPXD2tjxLIy7AGTGMeeAE>uV` zqq4=8bjzR(=>1)p$SyF27fwF~Rnu;h6>$N){BsV!ve}9&#isBr)8EB5A zzCOaA2ITQY+ScMeO&eL=oTE5cY6d^koG9MvXv0Un1>SRG7MNz((fc`vgh~5$L%W_f z#uzzq#~{vy84(;SEXDIKRq|a{={zvlgn!z25(9GOcuedFK5zVr-utp#?>JnXcFkss z@%D;gFYgPaa@Vp&hmTQ@6D8E%HoEt3&bMEb;AM%gUdCrOs_~iN^@7FR;lX{J(|?C( z((PxUoc$cHsDW^B&Q)qy{}s3Gl4$R$7l_G}>Ac+fE_bzR<)slaeBfmpuCeD4^ZcBL z3r0T_jXrsg`(M%Njf+j~GsHU%)`_3Ii4s3Td+=D6g)5Z4vQYaGVhtr3as7tpcqdnl z-&L$Yf#@D?IF!n7Re7LO$P%0{wT;VI{KiuU3;2GEY!a^Cf~95JT-Y522gYmk#*nq` zMl@pQeOBEm#yJW>;>Oy;qK*$~Z0DGhxK`;e8uypx<7c1YDL>}&>+z8|?y(wkg&^E9 z<`T*-%Hr82GpJJdO6L2Q@roT0ymQJBUen;kk2GE2E$XAhA(6Mab;}E$Iqy90*>dX3-E|Bib#L&O_Lnu@btib=%6ft-x6;VvyLrCSsgHDLci_65rpIVEjcwZB|%9=>OPPb=v zMFD)ylOn#n@ew!3DrFa{yVy~+mt^~peZ;<|Suz_j8`2*xW913Y@loMB)=>3|M>xh{ zxoZrbpCrZ@iN{Mwi8gg?egE!gysR9PvR*GVn7W58u7QC!Bl9Q)j90XC;Q> zceNVg$j9xx{ihLl$5ZZ=wvdni5GCI3uP6Srf1ddK)-gOHWHT)^3=`!Dk=fTrk_|`zZ;JtocTIZby=1bo@D==4hda~ke9g}dCX6@W1^nk~LymU%;BhS{tU1?#x{Ase5jBG>K4yUDGad1B z)gu`4b_i*dBn@G+EcHn&3y0{AIr7O7UM!VZh4#Nh5Cgu*Sn>TVqdZLNmm z&#LH9xpBVCRcAu>q-){O#sO ztZ$yc`-ArLY+?q}^X+It{%LG&IKb~ro&+~HFF}hP6y%?UK}3!Y%(mB~UTQ{ge_b5y z=c~lk6)O4iq!rvTV;G#sDFa$10;Nb#==<{|S^MH98l97)#(p2+;&w}VHYQ%+oYYUC zVf=sh<7Jq+Fne`VGREyNs;)Q#(kUD1Ub|E>{beO&JH&yJ{5@Pff0bx%v?TtoFUGzX zr0~NMHL$7P4|=YF^ji4>>Z+7S5>6gq=8}`+b&m7tw_W+HRSxw2qAsGP5C*?bP2&Dh z@z`=CiBEoK^I!h^U;HDLTSnN)6hdX^~ zbN72({$Nang{C5vtt(_lZpPuPr!j1*y9FH7=tsMLXG2KHBl=+AKx(yLm)vcz#WCzX zGrOP0RWrpfC_D`Bi>u)2&wot4@e5j(hvA3`vVaK>h~4(>qU$^6i9dvl=VLCO;Gfq^ zYRJ3I*rUG?N59%l4IP&7^H&~$R^fj3@ST`9ycBSkl+U8ZhbNFoU&9y)A2$EUYj)fq z2A>UT#_@`+e21+A?P`#w9@R&A_&zajwbO-$)d75GpLCkp$BX|SR*#lxQ+WN)Onz6; z%Gb?^Mk(oaBDtFWEVJC5Gu6pF;a3@M_t(Vae*;;S_eiw7hw&Wd`J^1S@ryz9f zb?mDc!TP?@D5xw(tn2 zOVGLHDxdrJG?~1>nr-wtEr<_tz|Rp4aN+t2vLxM!y7;Z3=Vl7{pK-S07x!jx(+}PP zO~r$FE&K{*M`wus!&QQN1^Bj12Q2*zAhh2hvM2NlcD>=eEk9b^p?zL7vrtMlu%@>;;OGfPopY5<+$wt=MVcIRK7G-G+{IE2&h**?EASi9(Zg+^8mdJXC1 zhU*%5?(YLIVn-aaT3Eooq|AbdhG2ZX@(}bn&A5~AZJcaA1JClaJb8G1h2{HKyva<8 zn~i8E>&9gBhJ|fVc<;R6`11rvQXRtcZhprtTO6E^5mv4|625$3Ge$(_>s zh~u8uaF={bc5Szk`0G0fMwi^v`P25{^SVmDSRg{}W{Iz2$O{Z@jOTtv+aMxgGo+|* z=4y2>tOc3#q5qJ{_-kzi^s^fc0eyXWNRA~{br*1d)vaRL-M8rplQneW;&^gnw<&h) zS-?N7&cYi-Ey!L!AxAtZ&v_Ltm};XS&Ux_86Oz*2_!$Z~Tc?J1Dq}O1xlv z3;%xRx9Gcr27PzxJFI%1%qN9e@wt1<*mx6(5wr0gSbE=%J}=i`bM_2kwu97pHdW%* ziv~&jLR7hLv-=ZH_)cLfR?K%Q{{!;0HvmIyQCl8^`!#u7?kjf z`z1B$nlLzDzLgylYx6!mQQ)7n0BE8H4F919X8HRC1)dA>RYWRzAL>Sb1}4C<{(YEo zp$e)!`62j_I8YS5@m%>j=^VO$#$}l4a{+%nNa52q4T5DEQRozTjw<_VBT;oDi61x9 zu#019kV6E@XS?E(+=E~~Hij(vR4VGDF^U&emS9!b9I9E1RCT5iKJ25+cYl+G>!Zb> z=T*#CU#h36_XBC*=}Qn9hVbE5A*ruVO=CG*`QTgH|!~YOg?Wl z=ha#k5H#;9%V*KV)6kcsk5pwT3P-t=_Y~{RK0A44WIW#+s0QMn2xbBMF?Qr0k*bFg zh>aebjxv4LHW z8z7YO87utrrVHk|PvX)Ik8sT!1^#QSB2P>l#1*4*$-rTUL^%zcB=d_}IEF`}$#n)n zYc=TD{AN&o>(JX{=wD@A+Z0E0(_Nd^JX*% zs7vu(X4$s`7yo+CT23>VZF32R6!sxIf-aNwdp-&7$QkoujdwU-JeMUsd4|_kJMuLW znqRy72oBD+MfV?1XyHU%@mmXbe)#b^zDb9I*^dJDcc?5*-5*B3KC>jWKoJ`YUo+PS zWw>$7GTx|lfmU8lrpt!N!kqGK=DD(p7Fwo~i@(!Q$21mAmJgumPgldTvLbeMZArzJ zEqc6`Y2jRJ-~W9-{(DDS=L`{9dOrZO+0mfU?;EbrkLRJEb5JGh1TX6{<(p503J#q3 z%zk9cW2<;A_WxLi>S_$U&28x!iEc4)R|plx&7pH6Ch<{{`ZD9~BEHP=1ixM8iwC3L z)1&>&sMCdYv}H~rELB{?8+R3ATc{h^FgyX;6rpDT2P~#R&(pHry6wvF20_QoGXkxKJSKNnlWP1(dexbP?79D3^bDC z$@;2E#3w7B9NgoBUUlOnJn$8GH7p+H{hh?FuD?$%w+w~$no2UXF$iXgwWxx_9>~eb zlJLJ%dY`Y2<3&>HBS)MaSJOfi2J>?PtcLm7OBXwP+-DG7dQKR4z;p_NM1m-cdiLKEhzVVzSU>AiZ;8 zEYWXeM0?audf~nb4g2who;@Vw(KwinYc$5iLnhK+4?d6~PX6U_jzX&LdIJW{;Pj1f zI_k3#^xfb+R9@sr^K?C!`w9tnx@A8t-seu|G=@>p@9XsK*b9p3V z8{QinqRsz&sY-S3k{5K0KaS^n4*Magn`P?SFD~F#9l>{OmId zLwy%A**BeVD<%e3NBiUM%>!}5o;q08y^A)SnGUZi>)3U<5mc>kI6d@cBK6#|kp?|H z2QP|e3XN8nQNy=wX!H9N{d7{F&S{T@2JItMzqW;*QIe-mg8qS9T@3Y<`U7&08U!O{ zRKNnh6Z=);=%K(nSX(RNzmsy|hin@0@=K&LW%gieY>KYS2GF1_{L$h%sRN~Z^P37!Gr(VtPbZ$5K*XNi*OG5&R0 zIv;yIgnr&RhK{`30^2BdPU?s$fmext+7 zhk~dHclq%&Me$=_E6m+z0Q|y5{Aptj9pBGjcUwMaTpWS{ZK~pE!Y>Aa6snjj^M>OPU8izYYOq`4Z79@g{cp z#U(*|WF32^yOTLMJrm`u3t<(h4`DBr;WsVkqt338*zC|rH|y!Kn&Wdx{>&A8uU0>9 z;kZhoeSFDd|8?*T(>!jItxiVO9>$M;%R(qU-!WRR3&q$3>R~)pw2zM^a=#2|f!YsHp$cXD)T>>Y3 zBk1-?sgS!Uohh21B6p@g1``!yv=6MtKZCSI_`H zw!G(}*)o*ZxT<5{t>b#X%m2fu?GY!?@%(n)(&o>Xs+2I(kTKMKVhpCQ8YXsUB|Igq z7GtNW(EWifbmDYt9O`+86gNlU^X3_RW`+PFJRYF_n|hqI^d=c+d55Vdyl0nUzjK3- z9aPFX8LXU===2dn`u&s|{Q5eSz5L~g0lNYP={k}NA3Va(`PCACgR%Ven}a;ww68d|Iu=VxFR<+yadb+;d$M$!G+)1Y z8?Tf;$#qm!xv|YCzINd)jLUQ731bh7UjDH}IZ3`cGiwsp5q}_V8QWoSau>UH+6aQr zX#W5A{=XV5c+2s8b~dyU7tJc=y1#7sI9`SqA4Q^@l8P|GC_=DG-l1aB0y+NF^9r8P zbRn|O3-~s^3%{+5mF&m6$)8(I#oL}i{P*5N>`qrOUozQ^t@gjjjT*M|W_FBc4AjR@ z+iPHBW&qso(Z-N$Ngiz~Boq_EU95%LZk!vQ^7TsUU;g z7gX?=(Jy$wWH~-T>Il}13df;We5l{3A8d2fYY=xvh)Or-)6YjPl1cMJX~EDF73nM$ zvekCe@(2~WR40Vqy01>h+x+8d{gl|vhH&`fX~yoGsL_uP%HhCGiKoPwClWn}Ei7tI zqymq-R8P`((Px%)_H>{!TV(0B3nlz{dI9%*=8b>uOaqV0PasIi2r3fmU`k#jpLp;l z4*tDcf?o&1f)WAMFrNXZJ$6&_%Z6SUvm2|ce8Fk>2)guSD?IhNg>7jmpn2mgNnaXB zKkKGJe7_vlv2Gs@6cy2e7cnp%N73c)?vbL^%5Z)2QzEyoUGU7=k!DI_gpA85$WFKoC#{*TBv{l5HigC1#cGp4`(&8O~%B_1?R{(M&8WAH5; zOp1!FAau`Ep5Io;Y*Xs-=^-^f<5Lw>T~Q%+4lZD>9z>57&7vDzYr!c?GSjJV&#fXe z_$1p~AkZnGX^qMP<`GFJE-#0+*Y<3kW;|Wiu@hA;2y?qV%n%m>2P3@p@GoI#t z2?VVVec|?t{;>G8CSN!HsYC!f-a-ooL=1I0gA7bja={^kRewO>LY($K8^m`RYZmbmK#tB}T-vpf%@w8vK8vQiZl27-@hMH%8 zSX*8ayOh)}7%^0n2aeZ*+S?CdqvcpW?A~Eq@!lL-eOK`}w@&_t$AjIZmFa+6Q$WXi z7d-lR7NsnAa#43aH~HL6MhcZ^f4gmP?vxo_yfGTBva-n3ZPAcu6TycRkAw@ylgOZf zu5@FyJ*}!Ibh^@Ndg5g!Xq-9#t*i2IkK;wu@a*NPdSlOjIY5`Sl49$95I4Y$YUG>J ziEo{W!`vT&4u==e(efP}^FwLwnaxxqYYe^J^_u7$lA#;yZo$i332t^Q!uzv*14UcuXlLpLi4GrQqo;xDoFx(0DGr%*46W?lMh3M+Y|PhG|&;CVkeY8!bS{#1{K)+@5& zxkpyB6AM4tsL#?69+!JZJ2H<`^M{9Mqrx<3AD$;FUf&;^ml&c#UIp8x6b4@>s`uiD zKr1VuMavNyW|fEIOtfjSaW4EybFR>vq)G=2ORqS(O%LM?FVUc=dTPIFBDoWFkQVKJ z4*~aD1wC5k#9nntdL>k_Y0~nFd&Q?%P3AcIE7nOga6=NH zeGDJvbr~MSOZdenSLuqRzP<0o0f8JeI=E2j2m<*hezL8HrxW$gl{D&mHCD#N(iIB_ z3Y+AQ(l3h-(c{Ov!M*x`s9G(L6iNDPtB*&4&4M4~)LMC7^>~2r^qspjEcX|D){LMl zbkjj)L?)4+tO0+#!)%tyEEBkfN%WLl0gWF^aTQ<5yHOfJ6{n35?k?eAoHh$ww%S5L zk{3yTHkHh1$*Bn4*Ukd8Z<6PW?y?2*_rQy3T@^!K?Xb!F)1O6|kAmo>uZU0o`5;ht zfQaqh+-`}r@TQ4|@K(O8aFW(+DwjSUqIblJwnS^uZ$tCpPoR>ZSXGOSt!pB$FHOST zn~bToLMn!u74i$ZgQ+;^9Q@w?g?#_*DRB-l5e^z)FTB0PhE^Vb4mCULBzq8q@OMEt zwjAJ~eJq^+yEGH#o()G4mnBxuYv{jM`t;_?$uMWiI+8fyPsRPSli29j#jH8_J{w+c zAvD&jqRSH0sEKNY_1AL)L{|^(Bd*(P=^y!-bWdtN^(jb)l3gDOf0IqF+bIf)B|5Uz z_DV3bxIgbaW=ARUsS=a{`PC9~_FU@%OOlvMxRz~8y zDU3FJMFQ5Z6%|JfB6Bp?md|S#hR(64dof1&vSg9y;8MZ4`AZ#gd&cOLjrvpEcBC+kl;4+qmp5dpsvR!5pT&f&AH_@T=3DId;E=4Tb8Y z?s2XlBYCb+E!RLep{=s_T>kezd*ja(V`16(FpAyMwCu1SeG)fP_|ihMKeBiiDRuRs zZGL6s%IX;MW8Gn*c`bo0kJ*Arit~7e%)g4wqfYa~Z~pQ9K12Az*V@pi-T?;hlX0ZP zH>}^+1WML!rm0Me>qnU5ceysviLJ)e-KGfAmn+-cOba5i&-7{1^uciDavq7&m`YY` z4yH4tu99I1s)7M?%B@p_H}WR80!WbLx(lMsAe@Ng_TfA@B>k8*Zm%Pyo;L&|WoOb+ z1(r17;6!?SVLKVNVLa=n@e&rkTo3l~b>LDLOxJm7(MCZysKh^E6HXkbd&U)jd+Tx0 z-JWawk=v92o?vL^q2-9_>u*KsZtX!Q<+xvH+^0=kw zAkl)SZcJm>FU!-ZhxO=)ofZ=Pg|=X(#eMj5rjg{IwI}7~iFB2^4%KrzN>4`0<3UYZ zRM?@);*wLzujEWL?pjES)WacgQW(tp7fJJk^>Ayv`moIsh01@(r82EG+`b`D0l+n{IP6jA>@#pp5RH~koCNAKqBgt6;2 zS<1%;MCqI>Cd>HIw#sJOMU$yihbnz`=nRO4Z-%QsO~^xMfX-_jB*4CcUcT}U-b-H+ zNZ(7Lqf^RY!i|0~$LJ_5$z1>~>vVgu(DT62iAo5xAGO&vS^ zX&x*ZW-o|a<4OH$k7C9@gNl0phm1ETf?c%4XHH8-*c7EFe7&}jEHM25n{I6btpRQ1 z#P#b;>7kH~QGbDJJnt~QuY}IfULtDLyg;_ZZ^tEtH8gLi6y>i|;6&tGdgg609d`K- zeZ2W8K80Gr64QaWee4quIoAuWsjY!v?*?eD(!_}YS1>9e66?$tvo!-2(o0TWRAz<> zRa7;j+Yc7PrG$P$p9g2?^S5S{pDc!tk{woU`=m&ZuH^YCm{4b5YYdoHPqG&l(62$8 zQE|4!kM&izAj)Dqj$iYhxXkP%hR*TuCHEOqIS|9TG9S~Ms|~G1|nws1^Kp%HVv*QEb!SJAa^w>)QnppT%9Mjs48$F{@e&q#xwq>dy`suTZ zNV^bR`ZLz1?q4{Bc`j!H#U%PIJr5=gGK5K^`p_pGi{OaOCVJuFI=XFkf?!OQ#BVJ0 z8~CgoN~iTXM7_BQ{oEN1hWkUo_==feptTfr3yj9X5k{h+qc%dw-tT13A9Q%3Nd4r0 zG3`J6(;I)7aws_I?xhB&rqF<%JMeO~G8Ss*)3`+*!h*+}Dq<>^LDisY zrm6o1R$N*~P0|)(one{CBEJh_3Y3MYBu#%U{6x1(t^D8k);RGa3^u5Mmc=(1KK@9y z8KuK{HzWLLsKAGRea|vf-iR(lzlZ#bU&xgL2fA~40-f#G0cE=TU}Ci-o_%-@t?H|( z!KaIKU$8m&-ciD6+Yf?z#}Zb2i0bq_;ahJoRHKk$-Lr>C2&`LXlEQE{cD zj$SeYr)sscl$ra`vGyD+mpvy)vO0uXeL^ud#+1jWt`Jo2+XKs6f|y3-tqNi=T68VC z4Ru0}lk{OJkT-8W3GmRtI#0IqT2-#K&!=IFE^B?^^e0GVKQ^>>xc&y=fR>uE$DKv71sU8 zCVT4K@y3G*EXm<2iAs>8I!`;v8I4@Q;J@pcT>4T`?br8Y*Us%W54XKS`RRc~oV6N0 zHB~^;n-8p~j*DJ*df>Do%KV;BW!Xw0?8=lV=3si8jFQ(u#X~Y^f8qt2FDyd4Q~k)1 z6ZK^0lou>4@**@;NAvCaXZU5g>yUerq0+3w_)XCP9<~*cOYJ?V(~yM9?j|&I)e&(2 z^Mq)+KE;2FSBPf)i-DN?;XLHtO*UOi6_sz;5q{(~bnV!J?`1r}xb_}gdmvBNROFDw znUeUrtp&9zeEFr!3PDn)6kdd z<2SZATqp_<)K+Bw)}j_ct#Il^D@h*TN3hKMI2k%&9Dcq1+xl_p8+cAOVszOevU}kY zo8~{&XkWIRtByE=6P&!+C&Lh`J3@&Y|52hdJ|BcH&V9MDeV)kT)C9VxC! zEmV}G0e{tH>Dr3}=tA|S@Tv1N;mYSRq}3Y!oRaY8o0~-D*n}jhKWwNp!XbUn+l16! zB=6hy67wMYjm91g(A(M9;U zRF(PMt$}TY9y}(R@F3%jU=`;NDhITvzh)0xQfG_t=R#qFlP~BL2XHFrHOMTts{y=$w3^l=9Gc#7 ze5r z7kq74MU?6qpwD+TI=y=dGc_nAL8>y;{BQt~mpY4Oi^ocO<)P%6#6xP**`1_K;|Y)zD_Pt!UNUPPkB!fE@agM^&=|+EV&A^d2bM9D)z49>Z_lf1Tmk1!KLrM&`BY@E_)zUlL4lsa1{ z%Ig%+e;T2u23}jf(|?)o=&3`E)YYr5>5~>z_trgZloW1{jVyTSR|R3h+hl6)ZUv4GLxfkmU(yHaCOrC+ zA{wY9^kV*vYf5cSHcl12o!0zgPiNf2wJWCy3lu(r zTMTe%^0*onJf;^77hwjBo>_zO!4oWwre8PZlzV@BR5Mw+Psl(+pqjlf~`^a^ku5 z*RixZj9m%LgkMw6u)m=$>|0v~iL@RHA6Kp5m(|12yuk}xKMb*Hh{&~R&3H|&&RvBU zy&72LZy|lRRpJM_rVjokY-M4O9*ET1c94@>??Q(CGr`2Xg-obfPI0g*oiWoG9-G`` zCC{vJbdamf{K5&~*4MwcmRC0WO-+IfKq@c@T0h#7=Gv!%SMf8EZoP`Wfnwg&qeXlk zFQNNPrt;FgkJ#eA+vqiejp(4@#NyOEaBgud-;<8mI46Wp;nH0F`b990RizIl{u-xG z4uCBm<@l8*557s|1Igx|$mq~VUFf;S9r}wf3h?jWY>pc^sqrO5r?Y3ypcYR@+_&Jq+lk(rQ z`rr8T%6|hL`rQ&v2G4|#DaBBgbC7A(_PW=JO0!6l5=o;bsT8FVA<9%jnGzvGB%Hn0 zIw@mONh*yrE0v-&kKg_Detmw|_50&o=eiu%IeYK5*1gt!-;cqaZSzSWZ$B(#OF1)e zeOpY?dk$Ud)@E+Xp}qIX*0^C3-vF<-PiHgJAJN)%ns~m&TPB+EgdJNtR676rPPWBt z0dY6XfV+$*#r4tX#nnFGRcb`TDt_~iW%KAJP9KiO%Sfv#-r!}O*O9bKmRz~3$dcXO zv(AJ%Izp#k?>t|Y;sg%s58*7g1)wA7NcaSqOYfH#GD*!7`f}=WW}&RkM%}k$GO-D8 z^x_Kot-gcvw!H=qPUn}Gj@D=A?FKNH`uC(ydkpz-G>89jM@A%fdLUbO$p@b9c_49f zJS5UT<|WyETf}8pm~qpl%j2x6@uVZ{0c&m%YJ0;PPm?In@;{crX%H$jIW*+k^S&-33@d*$|(~ z*AsvGW{ADB_HqXIUZM4q#zEr2AKZ|dO1xR=A-CM3Nr32$CUP(Dqmxxn_zgRBq<6lm zN!@>5C)W&qqW9;tKz~jsczJP5@lGmv_x=uKXiUJB<7dI99cH3GV*>DpwvE)|8{oZX z-*aQcUi5hq#~S-DWLJ%q(ZX}>=#k8E(dAvfX!Q*X7*{C|8Tl$S@o**olorAnbtQIi z{cU8FaE`7Y_@sAk{Y$R2NqjI0rnhBEo(=fJTRh80-)y}_s33!=21k%ttD1YC`+xJf z-m{K#7O<)F9&E|E#(&!uUsh8t$GuRnBbJvBb3vnqviso< zWPwJ1^daOc`nO&QR|Su!eJ5{5P9L}Oil!=PT>39$s$GuWM#g}9R2$k?T8X7c64-yg z7d~H=dQEOZ``4P_Yx!f@rY?J!r|E({b>vukex$@Rs2T14Vax(jW0>)z3c4%Af{oPL z$_5-vU>c=DZ>U6v`Ak_3r*uE}_A3*|8*)nx221I(MZ{4G|rFpYcN55t~K(db0ND+mly!gzHuc-;;c zowdKg9n%SNjB>Slf>8g%qCoE6WorSCEg!>_ady5EvwUMW8?_H_# z1~*V<*(60nQK;!`WeOdmsq(QBPy0O#-9iO?4YW+MzcLfiiyUZrh(H&{{=A&YITH_tJ??wRmBlPBT znvJPK!3^XT6@;H15YiCJVybm)c<*jw&^Gne{czq2pN-a|fU z)186jZKgbH*s2V5Z=>j&;3xF;zdmg3(krB`Q%nbiY-a1sZE3v49kem(0c3@|q0t^S z%yr5xmbJDA7WLS%+yj^4a{oM#n)QKB$rF~ z2MRGfDDVy=>0d#==G3)lc%G8MExw^aKCPZY_xpT8MSDJ>As22F+5W23wOc}6pROYU z0*{3*H{yMlH}duZ=Wt4-A*t=3PJEA_U?#io(>`P7|dWzv-1`LbbOf$i-l z(R6n03c5yp73}&tmn{A@g&Nua=JTCKgT*ji_RwuTJ6$V}osc^E@@lgv`N&mR@!|>@ zdS^Jf<&j5>w6eh`a0H~AsgoG3GBRYrr1-dOA>gzFCj{W9n_>V z8&tytJl)6{WPSBE>i%&E=q^a0eFfc>w$0tU4Hwn} z4R*7ZC_YK29*zfLXUt`i{_g;FHTXg9SZ+ca2L$nQAA88`_|Gt8m@E6{)5)f_sN>d? ziEtv2~2kX|%d)o&fja=ea3jvC^z&D-%U zOM%sTwwaa=KS_1|oudY$g7C>dm+`PDIk@V353T$7oVVEakKTYr^l;Jj&{Dh*lDz-0PX_My*KhPyNOj%1}6OHlpflKyn zB9^KS@(rK(-+%9M?7BP(jT%cU@dWd>y2z<%sQ11{dn~Y>vNf2O9KXSh z*jY(GDh`&GWv^t10@u*iuT$CNF+R-GoF_x7Qs_Gq5liZ^$HNZCu@RngdiQnyx<6`k z`9T&mX##A?rtJ>vVcMaUBJqc1X!mMOcz(2kPg`=1`VJhy1jsI4Ilvh!Zq))K`6o1G z$u!n4`+e{C)XOWPGxzPt!3paHyP9&m&LSl=Tb`zQCO1I~sc%hk8xe|x!1Y?&MC zuX=~Pm_|vv6@%l+1IhWHF>qqkOIWkq1+KX~5j6e0X_fL2q23iD+Nxmy{(Cg3#*ig%e88DEe?}?Ku5p>mEcvY;_uHC^#$$BbR-E{wh*jsjbYA5EWT>FcB+@ISv`|)Dvs_;MVw}AAb~J;K!P_xZDB!HyW^De@Wpw7BIFdbF z4)=|`4XQb5c=X9vIJW0H-uERIZFC7iD;J$Z=eY{5?eh~nH|iM9UNsfhEmf!Yb}LHE zC-f5?STq!CRGHxNrVtpla5P?&rpY1}T_Uf{C(ymWQpnW#LA2N{0P9UJCpsc|MlY1m z?Pukgt&SfJ{4$4(HPjkml7@&_D1l z+HroCpyzmsOxWQ>H(Q)1Id{v*+|+0~B7G%3vg|$ArQXap28_jx9>Xci8BB*au0&C{ z5AtT$B5}y-cX(cUAlNNji+AU`!!R2w$_L020QXzo0gDP^?Ypo z+zI64#S}Tk^4^~}^A^ixiL4HfN4XS$f6g25Wk$HAK|?(Gm7) z7%ct}5h#t-4uj7l>u~PKbz<4$)?&H+f`-n|bO=4MA2zv6WakTzIOk57xF>F&NV$%% zcifN*d1Gwp6rv7t%K}C6+6MTE`w4ilWSY>!9t*mzv9P~I56_;T1{Ph$I3jom=wFh< z=>pcpBmOYhEIUs3`z8Ump@n;0ehyBZ_L9!gP?R31(#4BDHHk9X%Xo#+B$_v?AIJO_@56zS#0as$Zmxoqei6<-|JJprt4lO-T`JoE#+f+}na? zZdr!af7RpZKVylNtO34%at1UH@Tbf9Ransg@IKc%WG?74CC@0N4Z^irs20&47mMCC z+cPAHn{8tPW{+#2ImBALJJ}Pb^}K}*d2Zb7s9e|XLW8dB0kQ#!}MF2N8*RC*py=ve?Qixk+JpFh!|_txTuuVvC#N1UaKRbDXh z%?>R1Na6>#Y;o>r5hoy(_=GN5-2a!k)a=rEnl{E8?~;4ko1d@HFva0V9}DZ_B5c3? z3@)3LakotlLz7Jmnt3&d@9Vq3_Gaf&nl?cSldt*X@SA7h*Xb^SGpa9kpLImsRr(DZ zO<&4oPBz8PTaSafG>AK(x*6CwL)^T37~YexmRR;Y;*OusqQz!w@w3=a>DHN#Mdu`T zV6i2f7DV00WdS^)Q>fQ9+FKoyB&xw=g$xFdmu3;Z@_O zimir(;H-X=uyNl@sOZQH+!_*!m%n-o`=qHPI^BR4lc}_OjVr{dy1|9+Sls{EQG7=` z6DO=w#lN`+@KR4CV9Vvj!(C@`SzyrJ7T!+hwclCbm+S&)(5S&R z<*I?>@oBdZ2wtVdH_R4c#pC~QVMGriCM_5Zx+w5RtEAgIf^cIx>m5gh*#d|E+&hfw zx8w1e!}&QAl%-x9D@fUj=Yj`|HsLB3VC%*0TxDtyT`B%e@~zTPU3P$=upqa8aE=(F8IKThU1O)`qZod-1_>R1?iD|4U8 z*&5OIA6r@f(TcP};6VB7=3)6KKj{0b4zQziKcbuK(VKIY^k9B1zuBRiKb`%7%uxBj zRag#VZ}X3V+KybV>_afRy00J3cr*$u?Y6ho2StTMx2Bz2<2nLlPwxPfrHChOUrm2m5Z-_L1@dU5F(%H2zQ!?0fR=++m2MM<~f?nTm04GL!`kocV zeNRSb?+8=fT&Dsbdb9~T!#|)e#naL5^T$BLay7Y_JO!B_a-=nlJ85MeW1eQQ zY=n>(9ID6I+T%6Z61*GCf4!!+zSzJ}Nf6$XZ^Zho4yP%zlu%mZh4Sx49HR5bV4V}% z%rs~@UEH3DPhk`I{I(0Y;!n^vauH4$d<;Y8Gx#&^7&&jZge(xa+Uxhmu-WQLXsSmi znpzcOYxXbz4_8Zpd7CSUZ1NcHi~b&5+x#7t_%EW&4S*A0*AXpUB|34S zE)LV33bK(K!D65i8j)Gf8^_(|KM1`anfxQ%cAY%xy`mjz=a*9tb{iEWM?j(NSSXo! z7acXMriBByA=l4!G)LGcZ=M+lf3Bra72^(w&qzXDdv>u-h0kn1e~DZhHIRNOdMlZv z^c*$#SJSg^`{IgBBWA1N4^`nZc?|Tkr|YX=`I?H^<>Mc2RV0fTFZ& z%^PNQu#~=-D~1FuJFv38O8RatBW+VGL0#JqeKop4GlC=#p7aU&`uqcVu1Rw3(=k}y zWCho0EVg%zMIW91QPYx4su-!oCWWQZUqk=@`Cppd0$1~>h`ru65;wSQfvRDGkB?CA z6{kf>O3MX{GirK_DR%SogYd-EWN*z%?VmSx>7~7iz~#g(XY9&YE^K2 z-%@_MvkclhZ6Jw^zYCjV9^+$)#`uR{0bQN=o?QIvNY#2KQ;XbR{MDW=&2hQ zK4r0vx%9lwS%U8lK)TC?|4aM_DyRO?Ig_W-pHvyYd6q=Ia&1}KkVdfZG~-9FA1URs z(onv_DpA`!M|%71eA+(nG99&B2A{hc#}HVVgWo z(vDr2O*NG=9hb5Azj0Y>d7jtsi)V)MzSK{y1+)YYn%&ED>C#ol>GugGmNz9e> z;;jFde+7*=n96l!%@a;Kqo}IvI$FQ;C*AsaD4i#0Sq+%g%2vuoQ%ie}GvE1>xH~9N zm&@BgyYG1<^SlZ@yOqdG9vvs4%haU-_DyjAP%FeuUcfIu`GI8n_)=HRi3lAYK&wsk z!COye5!$WoRY;6SE9N3T}Rk6LGxth($k=ndY5f?2%`B;wbXalFPi!;6)j10 z#;r@Q(uaDtiTGtG?Vm6Jjq4kMif=p@Jad0Q=v8?bU8=>7+q)C-qg_y|Vq1d*5+5Om=6*g(pYvv{kNcF~c(QA_I-e;UMzo~Q z(du!NIj-_B);$^xBzzL#t4E5b2D^(V%(D{5EcENmU9FwhkS@mx*m+$a@>bm90!qi= zYYh|!m#xIBUYF5?d0BMxlVDsGZi*(zP3f(-t-^gtsM!yqKX43hsv(DKJa@p2#(pF` zqzJvR&p_R;t$XM3FnWWVbW?&P!RL6TxsklzsZo5Z@i=f2dU_|u%0YbGWVm6w_`flf zku&fUCpAsR->w)7c|ncnu@%7!3$|e8qwPrE2O$_UA9r;R7N=Cqzyrk= z_^0_=e4*eXS~9y_z(1T|8vHW4HslW(HmVVAJzObx^xmcWHBaEVfu8taMw-~>vp$x4 z|C-lOQHC;~|KnD?@#yAmZX8NKZjHdwo5J2aVKO)Fj4NuXo{a+YZz2EUR1&sVfhs8} zuw^!$xbONUc=5usc*^uPyz}`jd|BwPtvqi>5tyHgThj+!naE9rtsQKJkN*pEtn`n@n^j!CD;gI~$*A*2l9y2$(PaCOI?n zG-;bU3b)6+g|PwpeAk2(^s1T{DIAvZ-}v?BUjMzHx^at6INb*w+kTI`R;IL#|u3&yn`LKO~G8~ zUof&x!#C=RK!@7Fr5(>j+oHbO%o6968-BZnEA@lLsgbL&l2a`a4f}|=CH36b_(=5Y zqbXYYJP4I*qQTmi(RpIrp3Cy23P=R@RJ* zqZ;ty@~`~Xg#(yh)-4*IRR?=&+}Tj0`)t$K+2pFA+n;IrisY{GK&#FU!JD(3S+!$d zSaf#^8`IBD5XE<7m!!{_$9!!T;*mhFHP3^PwHIN~$m7tsa30qnl4Shsd)S#sA_=F$>0Y)UEEM{E@BxFO~~RxZJT zw`8eaMklQd4rfm9M5N^XHm?4SBATG8%|081XI6P{9JxBs96T-`2BY;#3=jMA4i675r;9L9!RxxM*V^oH@N7@}_>rr-~dQ zX!cv=_Opk&8|i@MkOrs>h+*GCUQ76hEiB%0GW|0zhkkNTr~TGXfOpu3%1+r!<5kvE zyMG_h0;QjTn)KM$$=Z^6(Hea5#%Q1c3#d(>INsm!Cwe_?B=Z%Xb)cYaE2#eC^?J=% zGi5OT-SQCjPK;&^qYP+C#|~)73uba*FUj#w&&i|CWi)uF1^DiK0dW@XFsdnx2Gkd# zSxF*%+1VPeljO78E@J98|0H|!emg};+xT%yHPMki&7`Kj73``R{nWi4xn+$ZbNXQT zB)W=k)tkWdgE8dH!r|E3kiqR^_ehQRsY2y($gz&VcTPVf z$X(C^aEig{%P95VZG+3d_ha{sW}r_yLM1UxMSw;gfX2uce#Mm_zDu@_l?xsOTcrnC z-0&!cS543dfsu_A?{Rk6I{`njV~p2KtZw+e7F6G zbkzp)vJt6tfX56P`qvnH?Kp^Wz%Fdla}S&Gs{jTZf~c3mJ#-DEwYSD$`_1}rGwVDM zGe;btWQA*TY~U)Nz;AwCNA|}JMv;LtSgO>AU2_ZrWc>n44ClkEsnM`Cr@6QObH8#Z zRAEi=Ia5z;nsl4W#I2>7Ydj@)hs)saU7?awTA|SI?tAj_;!miWFV}ltr<&D?-_d69 zIJKIK)vp6r%UtMqdKIU=IgWS!Y2jPfz5?f)`efTnRoYW?iO%{wm+Jt`R)nULtc~CeIv+{yR-4pPiOI*#nsqJ?HzI1L%I2;?tJ1hfbm2QD>$E{?h&=z z4(miv?uf^}N1X8E*-NhF zM(aW^f^6VHk&&hZ&ED{e1iGw{G})>1>O;GG_xa7%{fL5p63)Lj9kW|+X!<#!2PK?+ zd`5}*7bQorgSNm|*&hv&;Ss&h>|EUyD7+Oedb2nQrhGA#2y7R;bDTA7IX)RJerb$X z++o-td;y-;9)iA_IK!_IpRq@13lxs;hO1j#p}=S``X#G}b^hMO5&`GBrOjBJrqp%&Mzc*o_}_~4g+N5hCIZ{hvc8Kl@ZkefV6B9aY$ z);nHD&4gN@_baqeXAbrA%MmmL|Djgp;czHLf~*pclb6a4u<36!|KZ^{+y`|c2kjTM zWo-^t+Z|5UZqNtk$vmhkNugVP5I)%bh3KCz!QO6y|AxsfZ186|d$iy=IUf}+c+WcH zz;_$)-cbgiv12bEmvtC#+<%g&HuM1mYT^X9<5Vc)LqS`lXG3p#xDwA9 z!?2CtX7obn)92n$f-|L*(0^N-Ab@6%q`(Z@yI^v&>0dTGlO`bNMJOcbE833Q`)7AFC=b4nXbXudXgqC*f1KTLcQ)b z-hwSRx?_KT3!-sC69v%*(Q?15aJ0D@m)Py0HnMlQGuQs2zj+upnvLh-M<@S5wny^u z`AU4C_6tN*8^bZDd%bIRpUX1v`Rxp{=R;Bb$$@ySS{GcYDT3c~EV28{)$pg(1I2_) zLi2xz@xzbyLx%n5!6}^vJVn4DeO!`*k0^=o0>de^?%WG>yZt$~JY9j+hnrzkBWQ-X z6{D}Yg=Ead`7m9mrpvwwln19H%H0a+=vDN6Tbcba*K^crwF=;_^#{{&d}j5 zue#|3?B02W7Hrm|t?x>}@z5klD{ms!4mMysQ4^+wg=4m}3s(eR#`Xhd;GKH{u)(Pm zn5nIVk6(HR#mk3-fA(Z_-eIhGufM~Sc}q7eQGIxx#i3!%kvAO8MK4@1-$ zSRCI*8f)*_n)dw$%*Oz8g0Ixj%{~H$ZwA}7ubosi72w>OFL+^R5`L|EnV0xf!1FJ< z#O3=3wCrFC46ye^3yB-CsB!_Dq3=+8p)MPH;3~c7(hV1288{CaLAvq=Q2nv`xLTtG zUA{Y#Ny~bAn0)w~Aqc1|Kx&=Z<} zx=z5i45piZ&BOB+%tB}OK0?+S?Zj#3N#wnM9n{TjMzuYqkl->06c+CS%{LWbH(9_x z33K0fWnZ{(BI?u9>WM)U;X){+rh^U;NYi#ExZM#*L{bd^6UU`US zTYN*O^nMflipTg*oQ8N{C&jue2jh^hIk54a8opK@ht^mJfw6@Wa(h}Oep@w4+`eIk zxGH-a)T}dwVFOZ7&X9rPhFBNznF=SdcTpelAj?VO5uMiJ`u@gZ6Rs8Cwvym+*#^Ba zIV))|^plPjyCo=#Yo-nrTRY7~C!d``r5BXZ!!g?UPv{adFFlJtt)&c+2GKC!Xcn=Z z{s|o&a{;ye{)@C1eIUdBeMBQtx=`ca<8T$N;jAXu7Hyk62=7eB{v&kSD zV+qinj6b$`XYJp!kO84^r&QLN}dYOU^L#nJ3AMBAF7E@9Sp;Jj!efY zYQIp)j$qN#8hJSE)fazKy}^08R!HWp$iz<C z+0+x6=tN2+Ow!K6&pPU%+sPg;FK!Sm^3kK=Pd7-rSI(q%)`adU@?dx722xKKe>A>T zi^bV&XKaTa^x2gHAxku|B7X;RAN-OjkUhr*FT7Kkh0J^*{xO#azW~Sp+xJ;UZnH@)1I|XQ7zflQ80tI@W~2(z+wb zto(5#-o5Q1#LZEJYbUgc22McV%TG(jc^=~&w(Bva)@8i^j;-{#b1HGxJSqM6)LXj0 z!V;>NOCf0SYU*SEfF|k1l8Xz+3pyXwNK@co?Ax$J==G~(xoa=ch8lGms=Qh7&gr6t z3v$@{)Zb+9#3t0=&RM$P>@hr#DngvSBObo5PV_8!7WKQ3M|`@kQQ40ju{*wJ0 zKGM_LO46}W2(Mx8(0-->My5R_Kb$*Zbif3pJkt<{*D$ni<1L`Zi*fDkVWQBthO@W}P2_}ZW#5~JqBJ)c-dEzl5F%zr0)=AT5L9}IxV6R*GkO@pe^ zWO^IIxFu~Pp)fE9-9H-9xkpRpxq(`iai3i$nozqg=n zlNr|M&N1hY6S%M^pQtKrm0l>_FWoLSAcg*QXpMI~)Z51auxKnOc);8SZ?Ygah;bQV z{3cWn{5?4L#l3>SQ)Qcnmyo|~vDcp%hGcnrzX znb6)GisU-;x%eOZXy8^Y>4B|d@sG{f%$om5{#&D(l`*(e@cUM9d`)W}M4+W>m(h>0 zlUU%bLu}JR2lgde9W~i%NN1f7iUZh^J#q8%2U38)-1{O}sp_<#1 z;nDlmaQ2xiyC_zr!Mom}^b#@ZIUfuQJvG?-1*U)>_`tu8y-;47K>zFwXEIHiWcKQ- zuue1qe|!0fe*0|&a|{Q;k5b*<@m_1=1JQ;WxMIRby8dY;=`Rx}=%4tqkh=q<4zs+O zMxul}eM9h~yC|kD1&e5xbtcz$MIPOJ;~Xsi;(~jwrNd0?9gN>Mg{^Bx?73$OHCyDs z(5J_ATap$z|2zdPcX*6Yq8cmPqren%zKMpJBop0{%V742V(MO?OV+jvnp9OI(2i;6 zxW0-mp^hSV+O@Sx*VF@FRo{4l4l zJgi}_02PdhmJr?8gRs@>4k|f5f|?w9NQb}EVB=)l>6?O==qhnIq(3Z! z6W*Hm#E~%KFep(nEKVuf=*&zD zLG!?d`mUpFpW}P#_WK@f=oYgu*}m);KY|?#>VsuF6>;(LVCI=QjS2P#1Qkcng=@{( zDvR~Zw_zYtc&dvocde#QTQ5;naT)FGT*8aa@$6+A(bP!hP(iTqe_0 zE`kr&r@|*>Au3fE%wD}8OXihVaO(IH1WAPX)MUrHm=B&+5Cz)9W|G*NiR1zQO;T&) zO;^l$K(!CXlTDGac=GXZ=t*&e$iwEGx7R9m;+6^BK5;sV74)c9YDe_OfCstz(3c0Z z;O~(6?5$Y|SdOaT0u#TBmOlA_loO9(pA~1Z+;($1>+M;dGj!sJ;UHnH4`5F-E>rbW z5-8CPqxZ8lVBIQZNR}0@WfTt&3j*1Tzcz47NfxUYI-)&pAt>TmwCKzNj^5fCOuAk^ z`tSS`^uL!(^Dl2?x8^R@eG<&|$=rs{9WCXHXNIt~wS7_J({~U)Uj+hk&U1DJ1z=fF zNcG1H{?7v|N&fUy(2fwnV7arjt;?8=O+F1nH09{0IDhhZjVm4E>wx;@>|sxA)!5QY zCG?e_A+_&YgC-{nx{1RusC6HLsXs@ul~13+qE8p$YDop3*!&Y7q)$ihq)XAXnvJ3Z zIk%}=h8a^`K7{r(FQX66@1yxI`=|SN95CYmmpv}-t9P7Og=fx}X zcJUIDsd))ot5@(yH4i^dIe_(~=g^P%>Fi5=7~6MP8MD|dtTB3n*j8;HytqFcWM&Lv zuQvj8=H7<)lfod)vffrzz{Tl*UC%4$ZGtZyAz0li6p(;R*Rv`mzD3K?`_?>8G1&!b z-)q70v+L=u+E>`K?*;fZCm4C|jRwhrYv_RI1$5!7ywFQqDmnM#7E<+6qPo5ofSx^r ze;#^}G;yVPtJPFKR+hpFPtWOS>o8~vC9e>d{4qZ8@Q3^jP}Q4O-g zP4JRocj0Z;90+#R>p|Z8yR{b zSe?nl&jph)@`9G=?%sW2qSGWQr??*5-#7_YZnF4qR~dG+xIzZrO+!dloxSWQoZdcM zqqn^=`)^M@bvu+=9^>hqlY7ce4pzdKJuUE-A~BUoZ^cS~OvI}7NqAekA=Znr!e<|4 z!LzZ4@MM7(pSD{{<2H=Kd%^>#E%y?_{S(B~o*WYQ45}i_lopbr>Q{8ZiN|1At4`LB z`_6kWE5{4M9K}9&Hi9HZ(3iUM2L})*JZk43G}knSxZeoHQOh(?#1SuitLqDYu;nSt znz5O_OjwA@rj4NH$GS<%wqkmB!gE4v!{J6)AT~dzLl^lE7N=QF!yio&u~$(NzOU2J zo3pI@97zKwnUYV3XGxB=odS<`8Jrp$hE)c1<9jOy;G4r2QiXkb&~a0Zaq+3Ja6t!H zrwV=}zIVV&Pe**a&;hUAa2{n?lz?+;4PIQ3NO4Ls6^A>bR25ge_oSNmkjX6iLGvsa z$tFOSX@3#?~uM<4i$6kgmnbUN^)$8=o<_cP4HI=<>?&YQa7k^~_d+kgBtQfUf;#_nA z;Br4+TVp^3LcBzPCAHwjCf2-l>K6z}_ zH3B7;sX>RrSf&pPCC}&O(Ivuume4hhZ4pf&tA~aX+ms)?NH`box;>a|_1eNUs9m9c zhU?fACnxH5aTw&Ze<9O~c9Jiygr3vzBqd!4DK0k^Yxs!3@e+gUIay4tD*)EmJd#{( z$p*)_O5#s4WjM345OxmjhF6{2@TS}=wDr{x7+SWF+awZj5(fHGUl|?dJIN0MALO!R zDVNA4OEtGF=gZwvzBAw(|bT;D(A|68B`u7am3FU|RbnVkHL$etfg~mYVj+@-% zSJ`~TX%#r~P{J8<|9G2JMfzjW7UnYWH1#_oLs!f=fWCZ-B)J8S?SY`>Dge#k|8hi9L&hkVh`H?ku9IJ znOKPLQ{P!KxnwQYo;6l_R2%rArn|V>E<(+vR&2|IMfBLVNO;<#&dQxy>HN9Lw7u>j z-MaJ`dFQ{J=``5VZGZGZ{-7eNos}c{=vc(nc@N>I40D0qPrc#oDNX$D(`52_l0GYa zQBRLgb%Mb+J?X`P&LrYO6}N8hE76M=@5s?MNpGz2{9TQ-rYmtTzGg`zx1!0N5yfof z{X=Bt2|XCBqm0~^o52mEKjih4lWa}7AzL4CmS)*dtp8mL_YaoxDjNdPt+KsPy?G#W z@fY@=xwa&`NR@^h(?V8rGO$H|B`DF+=EkQ?MAw?CxR>>&?5TeuTDqYdJm#j6*V7z@ z3e^Exxh;uzG1*DAoDa~cJGJrL?T6?@IYXw=sNZ{bFpH58_p|d*U&CDT>p(Z}?6n=+ z?C^poO_ic4YVPc|ra85UnMbw4KT*p;vamF2CwDphhvb1wH)l~lmBd__1~>eT__~QI zpc{M{#jLwVs>8JKM%f*-eoYIUx*LE7ZFPmj>b2-sgBnR57z+o3%*58kgT-#apJ{r% z9KC3Nk;<%DNMjxhVJ3Yf}bri$Vo%A=^jQ=8H+( zwT--Ys|mK274-Fo+@bTYJ93g&gwD6(VVjO6^F{X&TB1U%hx{KlaRnGjXH=AE+H$#1vFws zD&2pzf$a8J#{9M|M^STQXo~w?SY*lQ_$|N4%eMY_`<%&iMVTL36n}?`^xjhXRs$Vx za}cLCB;e#X+PvY^74TKQ80NcPhkve@K+nEal<@r`-n!>J9w-f=$6xnh`i+>#%}`_W zheVL^|1#lZ=_Io3mn^fJ5J0n|X5&GXi$&M620-~tc{KdrX4t>U!RAoZo%1(x)un>g z8ChFo)Vuc`OV@!nu8YC;l^Lg!6AyJB8X*35n0ngIVKz(pkd~dK4%4$f8>S? zRxM&*ETb5{@E1LMe-yeC#?t}1$H1%XJiDl@C}Z}0 zE=DDbPrv>U#g(-3_{(jW?LP$9HALbq|LaZux8KF;52Sux3y6K*Kblgx4{jz0q4ze% z%y!K)*8iXde`v1@d06O-EFzDxHPxleJkDIQ>it4mda7PBIG`C${HqpvZRvcCN|c~M zI|m>5wo>qF3Wj%WP1JH_7rp#d3B4DKVPRMl_wAt{tZ7oGd%coK{E%oiyt9W48!89+ zaecT{ZZ*hEIs~okBD#2L3=1&Iq_;+EQS(hF%RaUj%SrjDYw=Y{e92Xd()t|DV@Etd{#7>)PG zX27nH24tP|iY_{AN!$uOsaj(8!d z2GQ3=TBK{mFl7E?7Z;nkgWeHoLZyi=DwK7g^)emAs8x%~*(V^y%42v(;gsHfnlKms zI|BuMotXiTbX@RlL8o@W@)-I;x{w-<7Iex-M6wrZKZROwC>ek5Abw_Lk-;W z=#gAg2()+yN1SEZE(=PHH{Kyu!>wtxVm{Fx6M+N`NaSCp&t@$i%Nj(>{#4 zN+=~>=XqYyrcgnJuMZl~qPmA|p{|CGk4vyxNJ3CZ&>63KgYYl)iU9*XMKn zet+E8KI3_w=XoCEHhN%^Cg?C!vZ}5)%tDNNm`L_nW*5Ai_oz6L=nfu1m%U@i zf~Qqb61)Y}KNpdnkQ40ErSk=?Digd`tw~hdh~Iy zxY>isj%_8&4=-e0c`JmoP%6=PxXgPrzX2_I-@+sw%dz6$i_tJslA{$0s71v#sN!i$w6Eo+rc^CO0cY>+3bj44GlVDC(Jof&Y~WF zVfG2Hc}9UT@OZB-o_AM-^*w(`C-~a3j~5P-G50c9Y*rL~C919SF8Xt>4+(5?846BMmEUFPL5B{&D(y8CeT$pjXubC9Oj%2K(&VCHtA4=QC2 zGUYy9;eVk`RB?Qyz}`lI*~!PSg?{^iX5XcIzj0d7E63B^(liL!ZwZ1;HG*=3(Ew^5 zILv4z9b@QO^?SKJvJH7krN_9yu5-24S1&Dx=daXp?hjpheUBt_vz*Q=-^QrHtt&ip z)=LV@l8MdCG;lOuhUfNn57(1g(?wWXM>1PHco9;k{f62iPUBW;KC3zZg$}K9rwf<9 zV69m>Y*OfT0e>fD(?947C$EhbMtJ^)xt~XZ{Yk))(>$4RzZ9+GzOAw+nx)RK1S_{B zHgU{ZCcE_!Q%QWpo`$0fdwU{i){JtPn`_Q0=9ID2=60r*IY9k9V(CY{WAtY20c2`3 z7Oz`liUX$}qHRObZ1oc{8qm^0ay2u^%bU*lbpLANeO8}EOA53|Y!kV|NWmQWGXpC$<9e=hBDT^pVaHc9A ztbaH>f90*37|J-zEG+*+qa5a0|4Nz#GL9Q@e)4hlsOKjElk-gD>~Hq{j+t;o`DM$i zsiM^O_#Yx2{F)qDU5A9LMOdP;BYanw%hnHVs($`HvD)56l&*g$q~leOL&o(KbWp~F zSzG&}pIJ-L-wXaUaQjS<>Iq^sC&vm!PJLy~yDzhNb5%C`-X#>RybZGUPafXG9K7d| zshfNSzfUF5KNIz^Xvaa8d6Dy2ZcxDWT^Zohzm6(C9bgTk7BN|4DVoUX5tzK+Sp8PH zn|2ylqi+MjJdIuj;r5|$7LzB74b0;OClZpN`ndzUs&bH7e>%+z<9q@p?uwuhIhy40 zit)^MeiA##>lBFWIz(zZdL*;1FwS>3<(RSDiw=-vxn1Jcn1TDIlR9MaV*94XHXP zEjZRB4J&L8GZX$bCewd_-Md!CzU6?M7(3tHmabV zikx%l-(CWUmqe=;E_3ET|YEIMF(R!q>tDq`z*F!iOcaJwI>LL2VyN3pj zD!}QfXJNDWDbk$37UhmnM$Yrse;FLYgcFZhhWM?Uq ziI^#z@4kwKc;(Q6G$rD{F#?^vbpxIMIu;+?cnIQ>A262{oUWF^Np^8h7PB=hq=LF} z?9a{F!}Bun+c{K6Prx8u1)t`>CsQi-!|x_-Ivvd;S3RD?_)mPewPifx&yS&w-Lmwz zWE@I#J47O2L@7#rYF;k(KDu`(+kTsVx6);!{iKPGNE{l|o5nW2*u-`u46&_S zwzFt|A$h$;f}9vJlG>dKXIeUgg43R*0^?LgR;*D+A3e7HH^(FX_xbf*oyz{wVnK&= z*)?4YChOM9Cb`Ejr5ZCT_*u?8@+w*C&A+_Om8S4HF_2{*e+?6io!RqOW$at)Z{!zq zjdVM380fK_j!eTe5Vdxdhs*+zLR@3sorxB4Kg%5EaMWu55MuEQwph9%Fv z@C-Y`)zZIPhYk1nZ=*t_YJ@>)L?M~+^d8SRa-`txbZz!YZyED4mL2YgRZlim-?noj zug>JrJD;+spWJdmI;STglI_IeMt-**v*;dg(Vz-#jC@HK6>oxtOYVZJSRfQ`KF^*` z^1>PSZjg5I2&{eh7Zsnxr?;m&2u_5_pgF21LAy_cwau7EHA5reHq`5)356yd{Pb`$@CFP7l4B;~Tbw>{@yQZei*@ix zcX>D!G@^RnrD)=vr$`ODlhJn(S3&42AKtR`5s)_hZS^(-IkHo&82ZASpi1&B%y_bZ zXa4y%b(QI0f(?J6TVWBFjo6JtzoiP2Eyd}d0AqTo^N)2!k2H*#C4#HF6QOGLSFE7> z8uOI_Ci>pM&-TB<^-CY&>{}K@Jzxfo>$r)sM5TeB*o2mdsnQK83pigjZEQM`Vf73o zSP;1#Hax3@jD`r_*OB$S&1spu#%Yqk=Uj|-64T)^-L=pa%V=-KPrh*;_5OL(TZOW* zx(nFpYkAcD*dAhfAqu7i#)H@mGki%4n(TYRWEpgq2umh>htCh-6XmS z|BK7(f0&Y4?k%u*c@fU^76*Nk2i!SZidIH!!d358@kG5VuzJG@C{MaVzgN``x1nnE zX2JeLJZkY|fSLLohJWuVn|i7G%Mn1O8#B;g`5Mp)*a6m$!s+5ug{*w86Wx*P#>%66 z*lS5YHX|#9<#6$Gk<=S}$!{9|bAKEZ{x?!GyD-Q zEax!Rhx4GjBZltTu%GHr?*uLF~=k;6(J=tsE`?v7C7 zd;EAzwdT)aUwtN$zUipS%4{wL_p31*Y$3Ykup3C`ZX5_P9W zP)F}Z8ujHKdEK+0zB!#pm1LK)1oLZT-$e&@VZvIL*c1<7KOV=7(WO7njAUah8ixC! zgZq|;WePZRw2pA^dmoUE`VH-KGtjFpf0EAm^F^pg!oV?c7N9LZy#6$Ya&X+;C3t>B z2(Hf0WBadv;Pnog!MKOnv`{a=sUrjBzwIX*7S?n6VgbB6waG{# zR}mXcBEw_(Zyf$TU3(`1Rs@_Jq|I;eFCpD($<@{-t{m>;1TC|>OzU%>Q^hn9PWwv% zO!pk5sn-f<+5=Z`NR?#!P72wYOBoQl@e!*(pF(w8=EKSBsl;(?8MtIdGNnu>R&utR z&eySE=Sw!w5@{cpbzK2#SUy2Y$2BDayZgVpG9kn!JJkxy8h*v zYUvtZjwdyqnVzC7LREz^d2e!U1)=q()o5hRYt*;j8GrqgL?cJ6x=2>$d14Uq zPEhhg6sEL&hk~_AAjbJ&d@qx-p6Mu!xgL?e4d}p=a<~}9d4Wn>(k^sNaKio&68CFB>6};K z47uaHanIzfj}~x#Mv7%{Qf>k^s@I`QTRiDNml|<06KBg3D#-hqSL8_jf7aUHd{9$C z2K=m(U?P(zLie|IWWucg_%M8gRU6>~>S$KBLB(k;OJqdK#3n$W7p{s58hxb;a)EUgq<2-Bs zJw+fkz@x#vLFBYRj`}Na;C)MfUcKu9rK_YWsnZ*dvw13$CUw@Kr$x%t{zV=gG0qRx zR_cM{lU80cr*BtR;f&53DzkT=JdoXYeR?-Tg5J`0q&5Lfl!J`0L6K;9vs#KSj?F?3 zZFFJs_5jTPRDtR?=A(z2Gie-9)_3iW;N7hB}Rx{fx%oI}Sb*^l@HLEjfIL)A()E4F|)JPO`0a4$Rp77IoX~V~cvO4ChM@!WIG;Eaphvvs&tiJU!LRVDr943Lyzx05J)gO2_wGkZK z6{#Z&fFRS4VDn6h^A6C4gXVQ8R`en1(w&8JO`17x5gj!5V+Z{g51RWZY0TGvwWq{! zjQK8TFZzHYgjWRH73FAaR5*l}?@()~8WJPcurG zSpiA|2g&oCS{n7$fo6_8FdVP!)YVC9VaaeC>(>U*P2p)A*Jcuxiv5iYjCQj!%hA-f z=9OS|t{OYGHXX)26~l8Sb*QvJf^9!tM_Pv}N#SG4TRbO%D2lkD({}|NPt6NfubhdI z>R1|eL7yJGI-cEGu%BLy2q!WJzB0-eEz{F%fi`{PEL6We#=yx#P6R3e3gjtXuf1ExO1@NE6=iR%KXnnoE z$m-vB9Z}hbevCTA>~byAmmkNFu_RR1h{)>vCSs516iz zhH#O?0vt&;G5(j&bQXtWmtEYC6!l(`HE1u>KdVkQ80;hR-wlN--#pn3zxhJd&|6@c zdttbpFLRqo%cha2_4!-0TTfs3Yl5;+#OMV*5t)IPYCM7`dE>zE(PsRjY}9aE`n@?B zDI2b&TfWthgDz{4t*fWtS>8nQ^G-Ab@3_FeY01M)U2nQ&qBZzO9^`%8Z$fW4Md69B zGMUo3TU2{NJ>3#x#uPb`g#Vq-&t_(U$j-U2eDfi2R;&VWqgb+M5z}3*$ih-msaJmzjG3PcO@$NSQBOXHWkqQE%A>sOyP;@I zOA7Nnb&Nb0xiNf3{`dcW22G^9Y|lam-%9vs>xIg*7ga zV;Xnm;ap1~oV%$fhJ>sarV(LkPmakZEKc7ob57vMZAJK zvjm<~*(cr>wNqGpQWvxo93`XnQtO|>E@a2uf6CPC#WkA)p>|g{8r9Z5+~;fL-O*M% zBle)`8ktvcjGVkxDX4xPU)?V^1KDTxf`ZPM;r03FY6PoZvzvA3B>Zc~zxXr!+!Hnu z^l1Tm{Lqg!b9id`BX8)eizX2JIU0G5`ADahCPI0|9hk4S4PUX{26-Yr%s;x1+7Ic1 zzjqo=5_3m4-KC+SI+eQB9w3g-ipig$o$Sv1+iX{L73!YTfJBGFd5hIc1YU)Qpmn1g z$(egiAnpDF-kMl~#?P9ns!cjn)89jo>k^z8mB1u4dWe3=X6n3NkrvIHhc0qjBqMXxnbRCjXZdY7y_4__i>#;F>rUO-dnkG;iUn_{yrm#X%i|X9IO5duS z0C7$e{A+9k-n@4iJvVy}GIf^0tvd$c`^+=6Qhy)*V7rF@;+F<&by@&DoPLmAcO~KP zYo^DLIA1Mm)b!hm^3(GzJ(&5~K-#;^lI5+P0$LI=wCqv?k-YkeOtQEpSQDBC=9+&f z&Wyuq^5F%=a!X3=;$x@s-|t;ZdH_}xOjp0_p?PIRMrvA*!PTaItBqX2pNs*#n`MQD%Z zWEkw|At#o2U@L=M5@0%&=W*1PAG`1e-{zJxzjc-5BVWsQ1lg|P-yRg1!naZhOLG}MA^{EAdwe0bcr(hc6_AaT?J`ATZSB)8;h3mWm-iA!3ju>PFEmOAJYGs27V= z=F^uW-&)@a_(;x0GIBBJCZ~sy%QU7~3ccpuU|Tk1;BWg^V83ax`0nw=*jGo3B-AG$ zR;7kr4CQD6$K?tS8N+_6i&Munhf%=$K|Jl`J-qVcb*!hY2|_L&{ELBEKFYYOJ{ik2 z{=~Ny`_rt#PTsO#GIXj=CN6uE2c!O^5ucG!*wCgJGJ=nDXOYpub4};q;!FuvVJtyc zPl%z;ztiDl{~2ocq=cUTWnN^MgB<&th_%AzVAw-8AMaJQuGLK7BUg&W$F9U@%)Oz`_!$}5--4tbO(Q8v!NL}k0{T+t9+HvK%reR*hf1DR+0sAQ;&NwnIl&lD6#zSjyOT|Sv{Awiov>;esIm!Q z{<5C5ZGj1{-*g5~P&|%@%opPHr~3TTk-y+^yma-VWQ4!@o&%%bF4)W`kCnw9r|Sd% zg0;368``f*kB2&-sT&+=fSx&5s)Z*z>lUV7uW|(rl(h9P8`IkCjh`>p%}u?hT(M1!>PA z>QXi-49SM+)xXKdflufiJxI-CE>vhAvN3VGixIoeoFR7gJID9Q=8rHk;pT%S4)%q9D#Idb0Uq zX1HgFp2#ZWuab%7w||+<$8EO-rpM2dE^P-Uqhv+<;-|2wE%K~745&hRKNOm^;vv-> zEN9h&Cz~~4h1NQ1SH7RR_Db*?H$?C^e1Cx5=KJB&_2@@pflP~ zzMiJWNaF?j`)P?@5Y64xh(@Y7LBO;ax~)EjI)t9*WuD)Oj$P+`NW!Xl1Fo9Gd)nS0 zd078Pb+`}yt*w7gazHz%ZOsPTSDNf2G}5?@%5}B+I}zEE@6asa&5vqW!1tJ` zAb6;wN}c0xlgf-n2yV?Kem7233+L+cznEv@d6mzg{G=`)mpsI8O>OXuq^Gpj?V@1r z__b_5>i@e@go5~xp{%u(_ zbc+G2!_KTaAcVZz{+s3mbADliPtY2Rmn14aUQq4tAt*VSf-?M9S_j>ew!ZduDQh!O zWtwZFU}9(s8qhyhZ7}8#nzpcvjqE?pYP2fR+WNzS8rcW1WMvtNwTh$}Wjq0hW`jYB znK0sZ6!ogbybqHO3zp0iVGF;EAFd~^eYc&(b>`3;z0KA`_f*M+sZWRJMq|uos?fn% zQohg<-q5vYU7zEK|DW@$qt^rF2tR!>QOS$u`A!N-O82&r-qh)yg{Tqj+ zay#hzDZAM>{s?B%{M7qL4hW;MEkiJ!_C;l0mV9bAnEZQ}m=$>Zi zyQs63{K!%gcItt(jQ$N0RU|FUve?8TTAr}NrsvFOYa$!(dW#Op9%8FYYv7crDXw5g z1n+Kjkeku9f>er2>L30Gg@*e z0RrH#}!?_p)*HN!A29_}B>L zS)M>mhc8;49As7IS4T1&@|l|MO=DBFEO}qUwqn(k8oG0;I-b=#5ns972+zcJW6>Rk z0+#?8EUjq7!oq(uv=Os0w)=_WGKJwb{A%!@=7#G(-bB&4W7zf`kyL%(W;T}d zW$%z{f_sV8;5#dUR>}Wla?VSIGxG=7UT1OPB{ws++-3&ND2Zf?t~*gx?^bZN%q3C* zbyW4U7)x#F5Zu0>0GEu%3!dNMw9B&ch<8~rt(dux`YeoO8XJ>P^JsbDTu)^-t@S8t zH5a9&$5KIQ>@8xneKSv?S#Nk>xn)(zh6sOnjs9!z|2_{Tm6*G(0sJPHXkF4FW+gg_ zRXZ5ds}e3GQEv+PFH~S@_b0LU*9p6^xQHG~Y-bBPZHdF{)AX;e1BaW9W}@;lnN@f+ z6OPMa0*!LovP+SzNX($mKK`JWR?=Mo=j(Z!cg)OP`Pcr347uoP;t4x^T^mLHtMaJB&G#MYUfY;MKO@qOi;j z32Kdqv*8Qm?x{e>_Nif$5*;d{cMh+dY{Rb{Z^;)wI)pped(Bw+UmduJsX= zw!ROZa{4+kRu8DlZUXnXaJzX*D|i&7!Pj0(_)#Q*HW$a>pZ+s>>3K=;Vz5Tw?$g5C zEw4z{Wxgh2Ivkch?GOt7>IW+Jig=qvBo2)IO}`Bkc$d;H@uH1lpnV_~?^M_0 z&ybmf&;B==?CI)*Wm?7PB5x0*Uu%Qrw9ByGUKX$3JOwLbebPgGQPF2Ym6X;p*DIZn z@907+4=d8A)4Rwzxodz&D6!*y`Yb2Z0QgfySZl35-*3AaEPE9JY3tVF%B+vjk&q3Q zi&mr0pJQmsZf$H9wH;c&&%&vjCsCI#>g4fxVUj85}b6(f0Niym^PBY0PJT z=zbGI4>icpZ)upY76}q`K}ZKg?ozw7N$lH-BnUpd1(vA4MKbl~IAP&Sbn3nyEl%tL zvoVr1M=yrxs?EczD+aOD-%2X7SBKqNasb4}-e4PL&B1Z$D9q&t*vaV@5&e9BxDPgw zsdT$&IpEA?*gtO>39g9%OLGqzKmG%9J#NZ-VY`X)qi;e%vIRaDyb>Etm`wbd2L#i$ z_tAM)WzeKRiP?oncqqP#Y<8RohMQ_}?u~2Mf~!a0D4dC(CRo!SpNinf9ZmQeGzz>L z#Q0afJi;{Mh;AS!|^hyQT+O#Bz)!lQQ8mr&@lru(e+J#?dBvci9ap z{M<)RaK01ych94J7bB5*Jm-cHSVxSeox&Z?N#ygo_tazdA*wP-5`6xX0ZKIg)sz3# ze*P~8oj&YWma({d!gvmrB&z9VDk>JzrKimlJti4M@2x& zOBSNdr@_omy=dc*C@6hfN9Qh80LM!KEJ)Ib9sKM>9Cx*mSD{wu%%bC{{j&#En*5Zy zKZ~GlTZ`$>569WKdmmUybPChu+QK{km-HBjvu9~7EJ}9|`%(M=h)20x_s+!PVjq=5zU<*O80!60QHBYhW7!NyYmT>_8wmA z$@yv3%VtEaFGYb0<@n+^$F(IxcwlUEgHxlXQD2=HbNg$)vB47OT6cQF1>7O6Ab}X?=7{z8=%{ z(*`xKo5Vr+9Jw^>AhUfnok@4?V~OsgpvB(<<=rk8Ng{^Ntu(6u8O!DCh`tX7h^GjVwM=Ir0vn7*xngJ?AuuvRYbrrF$ zcPw;qywJ@#et5l1pWsPm4V|AW4Ql^M!Cfs!P_3FrA5M*8CgsQIpR)~u?{>LFW~h}M zns%%aKuktBKbHCVd)sUT7IDD0RVODFfn3oMPMQgyv7tejAf zN7nh^vk{NsMwLHIx2>&??c0Yfk0@aEnFHw4jHSb4l5|lFjPK5)8KL8u?I}C3+aC{7 z>!q1%P9RHKxB@?ykj1x;wva_4oOg&r3>mvg8T1}-wVT#bL6`DOn9?0ZWVb9uH#~Pi zRS|{K4{K$p%?TfGH(u}oOHLIq z?1&6vM&Hj+y)P@U#B@zCdb*9J*Zzd2xOJ>H&5G@Q>cQm99Fb=P7QWbqg?YOI*z47o zVe~O=FrXUDXQ3DTP!>SWt44%=x53Nrba-A~d0Ef!fB_48)yAR~j0EUYKK&l^4W*s5 z=J3IGVEj>!#I?N2gF=?_3-l_8X_YE2o|bR`~b) zyO3_8hFv!_!O@$=Fne|%j0#f4X}3Eeecb?B(6$VmJNF4job*Mf7CED~5FHfn8ADf% zHbAZmgjR{0hUh-Mb%LECmw~?AN&kMlQ9WVio$8eM9q`}Ea&ol)89B|(sZcQ&0qV;o zi9fSu*LbY2z`NiR= z8}68Zug?qg>latBT;hU8$=rtT*Y2TJ7HQDi7>vutAIDwp&iH)bO5DBe1(dbV!Zr2r z0dX!0E*ff&KR$lzdDYhc15!Q@0z?ldn%gUdm|v z%FiB^=GVf-A9GRpfCbSV5`&2v>%eKQp4>woR(YjFY8pb3!u{70}FQt;o_x*&^gxx+twRE-bOwQ z)atS9{y#ju_$TmvEQ3#9Rj~OIKYZ3Y0zX(Bf|^zwfz`nZ5PG8%?d;hMtH&*aJ&yOG zFUuZ#tPsN!HpxI(`ZwrmJqUdn;;?&<3k^T;nEcTg4=X)9phdQr=Kc4Wl#4FIV`nbG z-(M@jq232@%w`54&99udyf)ZKt6=-|PPAW59Qp@S@ByFya9Y%O{;c_Lap{Nww8~ru ze{C6!g;S@Z8M1Nkxc>p@c$MIcG1u^%#|$=py$wqrHo+#3Vzk)88U6;8kat1xsOQCR z5~wquhNTJU@7*_fG6Tt|`cnodg{u(leV5>(^<3ON?GBp9-M{K&2GcPfKFGVLjNEnM z@cl#C);jAk^gDF%e8p2~SWG@ON#%5Bzyj`jGlBAMRc!Dp8*bhG0j;MGqW@3s6Ct)Y zk=ut}lh($ybe!=TthQJa<|;YEZ)%XrJ-u(oNqBzY9M>KX{>&O|hCAQUI>jzO?755CSk zPZVZXz?9sb=;gFRUd&bn*s}ivPw(?OP@8y)$*3v_10TrKfyuVe;9|@^%xogxYbUY4 zI$K$l)=?JoV+6dJbcVdJor0G=@+Z96Ds=nbQh0vMQ2429D)Ihc2ropnnRBzL@b-yG zFz2utbshha28*v3I;M28E8!}vDPRg)nmk%Kbh}C@nbD6rkDi3Z>FMmq?A`Qi@lNKV zE6pmk*TQXiXLjB}llh37lBAiswAFSBq>uQELeuSGNJLJEwn$mIKlep(zcN|aF+?r# zZ5CRtEL?1%OoY}ssQPRb6fJsc9Xi7eZVQum%w`wJ&CG+14W;nTRT71LD&QSAn^DBirQz`vr)Dq ziSqO#5ItllY$_fH$+N|(_nO7fk23l6Qd29ZQJ+mTljaHjvkGQ!O!m>~w+;w4WX)u< z?;oSHb$x7lUmdN;sD{afX|T8t(}ahkXy3&c_^?-O_}qHb$Q6VZkD)SPHA_*PPp+?z zrx73R=^DyuPggt0XqbKwHz;l8>u)RLKCoS33;7JtAu~sKDo|MsJGo7Gq z)=e1wErmYRjRK8SAD9>Y2`pl?;R|;!Ut?wmYX+^XlG{Y^9+m45b=pNZNh_8%yE(GE zzVjh<-CdxKxlDOeEHrP?6t<81OyiAYgzksZg`qWLgaPu~NpHL(Di)f7rC>H|{;elm zY&wcoE~sIiX}g(&iYPO%e#si*vG8EjWZ}@l0QPzDI9BQ`iM>W{VX2;VtjS`u@SW5Y zaxG+zaPv=nVbb2+RGzab_AYa$C7e#0WaI~!CEdZ?C*7l?9votiUxveb88uMaQp0=? z@P+*c&LD^VN_b8ELYT8_KXa>n!khCa_Fw$@zx-#_6ALGi27T`mLMG38LASKzGJg*1vd6HSo*VOrxQ(13Jo%Et^1Z2{ zX1Ipa_sD`X?-MzmQ4f&_qk?(G{%q)_oN($1M%#|oTR)emMqQHoX>9)}Jncg}9l7@- zo7gE2yVn~F?{viqAD=FzA(1QLl0Y{(;6@4g-f=(L+aj$+5#qip*fb z`)X4Kdva1?zF=?S6lbB4gy6%$Y}{$P6V&p>;Ch%L?f#-b#xAD3u00cA^HB8!)_=amtKssZkpo>UTJEo4eS56n-<~CId7tnCeeu!B7f@Z+ zMpU>ZAHUz7fnPp60ND*m=){ZZ9GA$EiYjkG{wZ2$%_ddY_~ryXdLtHE-yXo}v$S#B zg`cF&gY$`+!f@S2eLA%{6K@$Q#b4vUhs^))6Phcx9i4nEfjZ=GaCPmeIP6RcbseoC zoEX^*_nz^I?6fa5(#{lyRAb1wxk=EpXbw&(-T~cPJ#hSsG;F$23)lTD#>Fc?g2Z`! zp5dbFVhqYX>F532-10sZ2rnr2-n!@)bWC3c!U4e0H+cfESti z8XYM=21{Po4adckcRZ2n^QS2LKpCAN5=VC~mn07~>{#mACHQat5SeK-WWBdej>=5R zqzSicsbnt4^*<6ne3lt4=MD*1rqZKc4dhLr4EEp2s>fZGy0BM0mA8GqHkK`l!t)MZ zAo|e}w5IhCXues1w;XJT$KJEi`cX@X(fWRRW#c`+*P5Wucf4Sm%^KpVl39Jpo~tj+ ziD1u;SkTM&mXhKF@0s15*({*+AdQncK`oa>(hg4JdPX;&jV%pe8+#8~ul!4RS?r|M z?MqJZy4edlYl`r=HASGYX$L;g8q5nBA;WwHvdHkoRO)Y0g`QrHqRWvl$<=EmIX5d| z-;fSIx8O5foF;=r^oxi4<$rS^p933+_&Q_I|Im&VY)q`h)l!iy-v_b_vhn9ifav$i z@K^F2>H74J9;)vkz8xxT%71xuO-eq^{*VNXUe<6R?zP~K(F|tQl8SRjx6|$%U-IT? zBmDEVKVE5ohu{swuT`Tv-UbP3BT~?R!spPda~Wpw89rtqiLdzh;1Jq^TUXXVVc{J5 z_}BRV@rds7V4ojqRvRjm@P3}{q$}c#A@!jnugk?4M!$P6*riYld#tXroj*l!Tkjd( zN=~=OrEfHS^Zf$-GQJ9O+$8w@hA;7b+X-;na0H#(z7G$o19k22hUg+Cs6NHj`}Qg0 zfnT$%=G<(9apyYWZSEXg-uV`-K5oEDH@MM(U0u}bnmskCIm!HNj#K@2OObm5CQqu1 znDN32wqXCj@Ek?+U6Gwk554zchxO7lJv^|W4euL0m89*Rz^b>+r(SEP3l_XxjOHnZ z!?|o}?v8sB4OUo&>w}-rwLKzq=}9e^d-)nW^|h6pi?r9=nQg3Iv1=0 zw%|7-4RC_@7g+B9fTvPaCdgQ+fq!nz!xz8tv910zJVDhAZyB5pi$D9a{vAcUs?lOt zs$eAzH?sjH&2z)!?tF4Tt@ru_h6Z1(*$6xGD0cmD&U0sSC_U}s!Q9neS|{GTC5Tzn zM%U#x!B}NVwVRAt=#C`og!P)JYpmJu?~mIqYn{^2Li_t);i4ulJiYq4b@;8v^m9yv z!2esZfZSKF?%l6XY_D3;R@K93@-A0);;0ujnwAZwTh`EB`ieN`kQiji-9l$Y^l+ME zJdIn|hZhdbLSI(5Bb5n~v?hI!JkT}9n+}Vxi@qJSu`3TX<$tis{4*A8)x+VTSrXl_ z@f+>?Y>Xw&IN|5l-iuQ~sl=^x0D zRFp!+zmGR0>?T=I{zMT7lnQ zL2Xq=VuN=Q_)&f-#rj6fDS0(j*B9jt#Z1FVrJv#bDG3M-Y9JrB-TgNY|1bYJ8hcAH z_UUF`o|Fo9u=m4B7c=mW%U{W>zAI4OQV7l)h@k9U3{`EKg%wqRiWZpi-e^vQq3CF4 z^s^MZr_Bb%5K(^K)0bp$tqEJbZ7%HDaaa(R+zp{C@4)J3zX5K%f~ed%4_ zXZR#T-S+R`v;P?8_Y6R*LLl~Ow!*aS8ud&1NVQLK-t&`8$#zaZq3rY*I`}dPM7D1S zoy%!N*6l9cA?HF0BF0*aD#~1l?wLSmCY8g-=_`5PdahB`R3STKehRJXlN1gJZHL#K zM)@H&ray$8_nT2|yu@}mUi^!{dv1yk|J*aDiS$i&7v5g6LKtK9gq2M^AxJ)OOJE!u zhCahtwrIMcQ0%LsP`uNb*~=8sOU0eEp;eA{{&7Ts!lT41kaOVtID@q;tQU-!`H-<# zNTueC8d(UD)`18PNQaE%z$5L)hAzz%N zsE(JKaFNCi;cV`_n8ImVEm&p*Fo4AvI@qLXgAQ8)M1Y)jJ;wx}$WHMffi%ZCi`6YZ(2u3VF@{uNJr z>YQ2OJ}Y5&{i@+uWSh8wjDPDwejRaRTel>Mqw z5squt6+X-~ql>0GvdKo}uy^1loiL`Ij;fHO0mnxO^WU}5Pufvbdaowk`1CAQ{a#+x zcB+Zi&yJ^25G)8bkQCnYmZY9zB2jk4bDI68fNqV?WkqjN*^ZN^nVDZSU0`yMeJ)x+ zovJeM;``%-89Ht3a!(&Uly8PITSf7ME^`{7^o_Z^^5%4hjfiKX98aX^70EuSzy==P zA|m1PwEN+GI3cPgl<`kzYr79(>7M_X-*0omVtuafuE~^2` zwi!aD{0^RMMHw905ya(qitsYiQMh~ONA}2jg|MgL9p?ej1drBfu_Jr#qeY4mB;Y*9 zN50uWr=F=`r4QVMDz@Ij^;c#HBSZ~_eRm^i-MJlXb~@*e@MxLv)9Eq7RXj0aSc4>b zlvYQ63M5eV`q!X+Bo9s}TqRy_Dp*$OcCenetGaGmBn!MyPM4)s!+Gw0T8Q5wmE%{S zWVSY-8t#sv+7ulalSTU)Z_rAak^k1*|Hhw)s;%y1U56;i)89+%_G*!bOOt8bfg0%C zXu=#^oY^Y^@zDSd;jK-LZ;d`Ie|FNpt|Cr0=8`QUZE1hYwhC)C;4GD6fY=0?g7#&7J zSKh_Xhd6JIZ7VqMz(};zJ`XvLIPw3edh=+izW)!{7-fi*BvWNbAqr=|_Z1Bip`wuH zQW{kHl;(kyF_d{old(_;_wK!KL?NN1B9Z2jD9xklIr%-`XRYUtv#fRQaql^2@BNEq~gpgJxJ zNi3d;&K{kD-d+ktvLlzXp;Khp#^xNt${BXZ4Z~-bccSrYOCj-ZCC<=LWoqY_QfHw# zTY2LWiJWdlWWSBTGMa#gOfn(Qj}D>VyZL-3T0rOC?WC*c$>Xz;iR7wB1e}a9qyk>E zNcBS)e3nb4>jnE^W}Y!^vQniIFS^mbpP6)*$Kio>`Oi-72wVp{EGl5kSS=JW&6L`x z{2|+81K_03Mxc$Z)SsTQb`O$6ZgrFK>bgFd#`pB(Xb01xjDw^gDl}Up@yQ&iY1os=N{00NVI8)9 z&v8gx_!_kehGC1%rday&NusHA0pGo&$373%rln8(;Y!OWY(4QCIWbp>>~504mwAmp zwHh-zd|ePYZw{c&+_j>F7)Pwm4z~|uT&HY zHrs@u@uo@4a-@_g2sRZM4$83$wEwiEkXh4j7Jb8SVw-qX)npTTqQb~1c<1IjA)W#i_k zkjGoZ^v6Y>c|-d#Oy#-mCjb3Og>zq#yXSLYmCDEe`Tu`5D2RIAAsVA?;Ez)`9*m7? z#q>RFIlWE)C|#r#*`x6INe}Skv}BI_uIKJ$xX{mm7s+hFX}C6eD}Hd@4CJM!P^pG# zc+Uhz8|Nz1oN;?ep4v?IQPGgC<37=aEf)CGt0pq-^j17X<0hR|okYVn@*G`T%Roa` zOtof@qkD`ismc;JC^6qIyj{g0>S!B{z59Up`d>u16VqXR`X<_U-D<$ss_`7!QR?#8 zd){s0tlr2uN@XMWORu8ZVzRjxEXAgJYTIGFVZ_T4!G+D9ZndqRci^)*MdtZ=VLwmuF{Rab~KZ+{pVmmpm?Qas`@mp7Vn zb37hg8aj2J<;FG1+@-N^$hi7s98opS{J zpVkO|Z=Z#OpXB1Pb=A;YkiwP;E$G|mS)|tMHa-+@$#&0{5EP|fV0v#4!SOs^-@PIS zN`+B)HjBp^u|>E!$enEKc>>A0yWmXUDA@d{lO}0~U_0;ic&X(au+gJ9Z^?Lp%axVj z!4=W2zwZWmU0!$X;|81Oa7CX#BI%>2XtI+#=C)kIo3>8FH9pRG{^1lziZREPzmB50 zCl8^gQ_S)8YreQ4=m~m~bcNm$%5uY9dWCcM7sBtX93&yOM4gwm2r6W2v7W?5qHHWh zly$emf{%QSF?$AN{7l7v7IcE-@gm}Re}te}Y8n1BOP*C&VpdvIMaR}pV2&B}!tXaq zAY-T``tp5?=v{F&@f<&usdZ%2D!=hq>3AQ6EV%+#Bn%*6>=@cAa;D8jdBjz>4nJ13 zgq+(8!2g>Z*8hBgnin+CQT+|{&EyK)shcC3sCF1?wG7esUpLUK={~gItxS|^FpFr1 z+p>3NJjc(4Ui45(7Jr`-1=}_J$?KoKmam^AmOYWbZ=Emk2)3KZqL5GW$aKgJ*xDO_ z?{7T7;#@9~W9!Zi_~^8yF+#QUd{K4B8>E&uSCIKPLg2YtlEhx(dj-#2LULhtu(E6s z%>FI}%U&LN-+e{!eaSb$S%bNP&AX4%!VcBEyw5F<)^2~ zC!f`1Vqy)Hv@FEUcXs0^Rp-(81(@0)KeVOP5N#~bCN^c^#M7jitTr&k<8Ovr|N9Pc z;T*JM`XUHUy8CaQ|1bWL$dSLsb$Bktq91!;=eu|8-1RCtVbovY!i~8s_vb^=y6@7g z#@`paDrdlo|K#9l_dVJ?wMXRob03+uLWNZRbYiS z+azhx{$eumViEm2t5me3P=ppG6US#)u5lN73n<7H8gJiS@w|c6)rc}PikFGi;5pD9EhLx zJSs!ZQN!qp;WF%sP)yu;tlyKE#jJQ?3~S#OD1LG4oOsZL)97J#0aOK@q!LQQ=&)d0 zR$p7op1DVn>I42zQkHM^wutB0*u8e(JLQgNk;rS>c*=pVO#ODKb$a%HTweblVcffy zkZ!Mo*Xm&W)i#zG>ujV`clOZ0PrCF^gaZxj+RsF{{Fz!&BE7Y85qGQVGJW(~l6jx> zC5+eNc=2Tk4l<4b(-HQ>Vvjmo5qy@be18w_sO#d7hb`dU!hPh~o;fu6aRz;NrH}Lv zSO2pGQ)Y6*!A98nB+Ye{kLC4X{-k; ztocSSOkT<3lwTlE-*413KZR-C=zyCk2Owt3PU}i*I}i-kpegq!urBEXtVhK_{C?Ft zw6Q7xSn?#Y#m$Gl&OdIocbPZLo%Ef?wTCjT@AXJDtAR`ztOu3Dj?!|=at7t|iOT2_ zG}Y9Cg)I5YOgi}%57ikM9NRUz@+PNXAcO4`3D(6OtEnav&v7CJ?aIeflLysnue`4NGr+|rp`la#>ki$`LO zbZN+r*AV*zCZPw*qS)SES(e@Kl&CphB}S#UY2G6rhSQpugvl5dxY+{rqzjsi=L!mzzl6NgHuT%n zN+ep*DZ1_+!3|mb%=(VuO}aTy9_)C|SN}t4u=Rx#ER;M$R_qIbh#MV9^r{#CJaiZb z$!Q6W{0bH*YuCfXz-qW?tR|2dG*zH-V-9y_#2LCRyoT>BtAx3I`^eSO&D561Pg<@X zOa2=sg!IL&=(7GiI1)}kw$6@Cx_N~WSK+ts7X zvWGGDFj&O0_GTd82f)JER=VNXUo!XB7h2vcCU;sU)A0?`Y>PgRP%P1=x)bx6OoA~j z%o$C0SXWYUYCd~h`I**dB#@C$z1h5>fEt@!VI&1DfhPusoEcVlx)Q#60 zG4x7-EhEOU*Q%K`T1thDUTs4+96ZdXDCg6D&ygUje4ZA)AI^q&^iY_4fGv>!LXE9w z(S=?6(fOcLtb1dqpz`2!)@3(X>^-QFuJn+=Q#T&N#g|Wl>FUuiVXY;S4Dx1`^)|R@ z=GXzdW2o^L1#UTwdj7kO^7p&4mzK*}dSMmp>3NS!i&nFqTb|5d!8mr=XgxY)chPzy zcZG@`)fLj$7AaIUf}ojCFw)0@1GNq8%PhJ68kdfAFvjf-U)+m3_tujhD` zS|2&)WXB4v3(+pm3-qArEs}7spISHGKvN+dzP7x=UBAo7R-JNcd1N7vS-Xsns$azR zJNM(diP22CbvTxz%x4xts{ta;1`CU-=JT`qh(@E#Vwo{80ec`*al zkEWU2MRObwZtw^J$u4Ia`DxsM5Bq1Q|Jk5ozd>Z|7M}B7>L);{3jT4Z0sWq!&Yqa$ z3JQ!1@Tx&#CL7P|diCWCUta1W+m`UxshuYue~!hYjvCQRjvTA<;dxF}^Qr&uV3_E> zn_d0;3+bsuGM`PstWR>cPlCN2B|rO)did&FJQgBgdAlxLSn$E%c7^_&r^?lmM| z5AC2gLkOw!sDl=nHfo%#fpeV=>HCQ_0GZBs-q!g@b;E4dvvmPqtFd4jKQ$q7db}uY zTRU|M&bE4fxDswW{|X+1SK-EqSMmKwZNX=J6t#xP(_$%2{DaGf?ZaFd2=`H}8V8_$ z4%AI6Q2EP5`Xx1)bjlmxMHgN$U6TXI($WUMWwCs}(+n0I6G0{z_S28L#dPwmJ+Rbk zBECFh6P;N;4v&b@0a?FtH2l>l(4BUZ)6+6#SI#%_dI{0o>Kn1#=_kw4Z_Ot3Hi?KH zUhK0{2zLU>M;@?Z@?f~LE7m&j#uWN}ry&u$D$^tU!p3G*pnYI0zH#0XSKXHt6b>_E z8{&EVx27jvGgprOnvKz>i@fedQy%3xftapp63SWjjD0y_E-vs~Dn9n{6nmnlCtkW! zN&M>8RCr<+M?bDXWMP;jjZkZ*%4SLAX?+cK^K@eS#-w0udXJ8K`U@`f=CXnx-KYs~ z#VQj5K%qMwweXq@!Yjid^|CeHmE4a~Zhohhk(beexv7v;dVuC7PKSl5V{xF%O0X?4 z#b?KO;Oa}qAg*c(nf*5h4|8?E+B^r0Z}U_Vk5kF!w{`wUJBfH|}wNCXU_p4BCSo@%j_N za9Oy4Zu~wJ?_L`OJ|f%Io$=1+_4bovWHfsjfeE^qeT399cpo1PK)0M z(^D2j#BfMDxNmz#FT7oiI$9i2^q5QN5afX87f;-~#R`8+0DQ%Wg4ttLP_#;f6_s@$ zJ8eB^8JGN9&p}b%Q{af3j7U2H0;87=r_p)nHvUp+2%Ur?qn#Q zAlbo&N`C=6zv1lC+YG8htmu#2R#q_tGjcVG%}JBONuzD>u0=Zd)h|84qBuiAVf8mm z)@4$^*UB)DUwgxXmh)KnWE|Tmz<18>VV+2hIeL_W-WpAC*i-`}S`)GDGClb8=opUo zmjYV_d8)t8inHhU*tUnP6ZvgSpq{$gxMa#G(fEDuAd}ZW8#=5VByTlPp9Op+`~S5) z|NhoraAeBKhmluPF`l7f%TzW;f`jN4zdu?JZwzGM{-5JCx+z0sl|Npjf95EW|NN33 zu{CA-CE4^`+!_2%s~!YTmGISx_koOF3}$URiBaH8a5`=WRlNPmj8(%Pwb$T++$pjl z={pL2s|$sDcER)7AR!u;jXO7F<6zA`+TuN$O}|+FFRlpscl6J6f@90CqJtkUQqi{p zw(l#?cV?81)@W^|Gr}#XT$396n=pbE4d%5yjyU7TY3;ObafEeq**Eb2GmX|1_EGQU zQ(2q28XI-YmhJ6+M2zwZko;R9$}hF(_8@a=my^OOg@4eCrcK1Jx*t7ki=&Im?dfIB z>vZ}Q7xv5hBJ)@fObg;?FeQokTn}IG)Vl3To5We9!@-s*ow!RI90S-O8ct_tj$q49 zE@Xc9+(7r072Aq3Sy{|iQ~~i&(zB1gc2Q-I&Ch_esGA(lGvl@lIgJ*6NuVueUZS+! zn?$=GXi&QYk7?@{C6?^fOXN@2A+xy_w4*4j%-i!I{j%X8TW;`?JZzZAe1^%x>wV4E zl{;_H6va%Yz%hF3a;Jzc-GbyNUnIBRxUsqequ6H=-+R9GEDd)lp#>iL1AgcDN*nqt zEstc({Y2Kpq{CawGT7KP(YoMIHTv=A1iWz>OeS1>jY2|Fh{+v!xS$%w))+=pt=eG_ z6F7;-fb%%*yaZzWYBghV>EbtTuxCmR`ap}myjrOcA_%XSY;1u?@xoTjn*uPI|+mQF_IT&qLB8d zDA?x&m3X}q*!~w(Ioy?9da{>3U++ejyH93~CnAv2hC!_EhaDSX*3QP+cT;&eH>MqL z!ZP$#=+7JTX!KM|dQ(_LWRVy5qwNee)-quqH(D~yv%PfMFGH3%DUyEJe4gp8-NCkJ zX<#mAJ~MMW&UmOW_}9J{MXl%g#24@TU;Hz$&RzAA$p&#V)eTCayDK{B0f~Nk+GIAf ze4@falUvZwt%`JEdOVl4_b%PxWz4J(Z%1YIKCXB?bnP8S*z|Kb824-98!PEM+Q??b4jQ2y&2wv}(0f%=!0DJYDJ_d6!CoiP z80lYld7LEPl6DH}Yxg48&!f?QJiq--ZC>+8d=t%jm5Dbm9*JKq3&+js7tviB1Lg56 zu-l6DSmn(-l2|twozfqT@25J_NAE{Lz4SOd#5NKiapgJrN65p(A4!nBZUb&UpG{6L zRiW`lmNZ4S7VdSb(r@1$)0UmKTu_k@{_wL0-c@M6j8f`+g?>`|qA>aAF$#imIlL|I{sRzG@X`#?1XSf?byOE9D zHIW(LL)mzV$GKe3CpvvY@!!7+v@K+@sQHTmOkTg82J)K5)fUBAYWh>IZQC!<94$?& zV?yDK@4ff|pGXz0tov`_!G$ZgO&NLF8ulh+30N7dH&^Nc~v)*TL}a zS?EpbMX>!j2WF1;!!l+0U>yGng2G0N-n@z7T1RB@x^FY->$EBq^XsSR^PF7F3xs15 zl?~Sp6@qW<|M5GWTd^FU-zp?-tA~+GGJ5p#7)KK1(?idDRH8c1c<2l}HIPUAizE7b z%W>w`YKU&hB7L9s!ynnXB<%1JSlu-pSMt1S9-E6%k;P&B zTmb3t@B_&jKL{&|K&?xc(Fd+(_@1XU&OXJ*s;hhkeq-t939x@rCb~0N9YV&N;%L(o zl#}ZeIlO465tcG2_|#G?d6Iz9jL~rX@O5%=Jn8N+0L!o6HKz6IEj)@42O_RTbf^a2aJEdv%YuvGS%NCrugd= zakNYj+cT|_wm(p0XFsOVZGDc^p<6?IZtWMAI6Q%B2WPViiI4Q-oP6q?IROj{CXpq% z=gV~6W{bnucC%}G{j}-k7CJJngK9Xfr^h`nuuPA$Ea=$>O23;?`JO0L^(d9I9_>Qu zzHsupQHuW9mP+4#drKR-II>r#jK;)!Qva+4oY<<2d%Q`4tKJd7MgQ1Jw>^D ze22n@a?$0ET(ZtIm#O}@fPT|VCDZd%#j8tnSfUQ!TXUV`aUcOqb&(R?lo!GsUOOG7 z9+99S#_>!xw_ddI@J8mo*n;}^#xS2N4Q%JAn=EjkGd?zw zCg{DV7H$dbn&=L*=6jYFj4@>P8e?5!OT{&c9S>6dk94a z%c6yQB-81g`NP;;`{6XIV-M4ka~<&i#umveZjrgzR5_A`)oX|g%bzm;IeXc@>}fDK zpU0A(wL_VO(M-J}fHjEn=*F+@OeT8~d*htMj;?%3qXJ^+TVf&N9zqEdph~J z`Z${ce*5UQI)1E~SIFL=&5yD_nWT8QPpaLX;s)VBD-f^vy*t_Bc!izL(w* z&G|B8U@RV{tfL1-*{pxjHoC_$WZ=0*G}x2RQ#YeAy(ZLS=SjArY%QA=y`C9K+gMM3 z;Z+uwF;QrC{wSw?^#t1I5lzBvT*Mt=s^aQ6o_F2xDz|rqdfBuuv1RLATTuIs24Wtm zPn*~GG4U=lai@0`EngPExy&-6i*?q~X7l}MMpQU?blQxy1qX;fMC=o<*?Eo$_!>xO zFR}XM5KQ$IWZ12PaqObRdbZAfmUu&{wb;t$Exr3{C7bkPxOkj|oyhm70=#Y6KXCsS z#^{TcQqx$fq1(VS2E})Br*gtb#z!y8U^z86Lu}=q8EEtg#ewUG7YXQv%RlIG<>Tar zc@x@Yc~^uww-UK;&qZ&3-m<PTpOyU=_6zY>sP!csIl(0#K)r9|LUK6geM=y79B>{6_+Q=b9 z@0leDv{gY?lf`6velqB_yTfDPZKKQuI5$Qb&uG=h5C4`x$;zQ*`^cTNVP+L9icAp| z?dyVoH?3$|uq_gg_ki;ruJC%%Vo~g^K$zRsi(a1_1wl&^*ow2t@YZCKXuD<)&zp08 zU<}I3`p~l6O;mI=P`I#pFjOCq#3iAYu)!}8o_EKS3p)(hlGp?Y40+6{>^X`XH>JX} zVjlm?+j#fy2Sw(piRj8?ZTR&dAO4-|?Mww8D#yL}nGRN|7)8%EfKD|<;<@M*m(bt} zcBAWw;qJ4-l9Cnl>Dw$^F?2aDJDy8t>0XAhdFilatu8e`FbbFNy2ZT@zRr{;+KR+! z_eAb%3c;A?hrFT5$G;+FY<%_{-v0G8d563DLP^*&p_-XKXY!T@S)q+ZHi%l1O(SH}kof^%SnyrIPSJFY~JRQ5Z z-G*8rk9|FVgzPw2$8|5vg_S0n_~6%BaPx*3ANd%CZ7i37;?RD2UMQl2Z-tZ6tB>i2 z0wMbM4%CA9KGuD2V0b_bO#hq>A9*}^&Kd`}eD^h-?NTceXblmi-U(s;t20>k-xzQc zeuSNqRLR$;>p|7}2PEdNVv9D)aw#vD!YL}vk9h-3Fg1ptN*B;u_X|lryoDad3s^@A zuYGagFgagki`HE30=M>sc$C9kyyH|9Xh%q}CZ}jvV{#Ar{hmYRs$DSn_ztj|pHDA1 z?m&L6<~V6h7LI$$_xkJ}h5Op^fY{PA%=HbmRe+f(0oFxmiPFY`_ zP>0Gi524eW>vo@Vc)k?G-r$M94BqmIMLmRx$4Ag2WczzsRGb3!(QA85tM&Zlu z>+y?Ma`fB%LYTAJ0lVsi;j1S+*`u2w?AEl6T;Pid=xc5im@eOd`a<)$)15xB%F!A3 zZ_&pIg`WoA+rRPWk03=&Q&8s{CK7g-pZv zRq*?L0OWk$Mwf*?MnQfCxZivTdpEs=;EhkYE3*@5Lt7NwdOe73Y!E~95k2f~`V0Tc zI?1%=u7MX(UykyD{AIqIz{})DV!`QYs46OEDA+dq$sZD|oeR;8rZary4pHH|;UKPD1&eM<4 z11*n8VR;m6G17x3gEvUC=$W-R#*C^RN)ov?AAn^JkvJ@P3q+1phu@Q?LtEw)?4G30 z+KMr%yw)k28#V*9%-rFo^)!||t^|9R7qc5~R~Q*(E$FJ)jjiJXSWAX9PMVX(QuNOO zHrc__X0M07#_MQ~b^`lUbCWvqb$%;TDO%Uq2vYX>R9Q(0LPu#cX@3uV+p&(Y-qkd0 zY_aGAk5wC!wGf_6ml6DGY!F7?{0135Q@Nh`k6s7^&5 z?B}k>qc7#6>2Z7?sSH7TR>~opryppc#YmRgc$)@)xlb&38pr3_*FfRUc)mx_0tYOw zLx!HS1k*-xG~^oKkSj${r?^iv{-hFm@VTD8-=9ojn-eU~3PDnT0`P-4p0{*y6iX~K zCG%Ml{h2L<#ZNVbRqjQ!+R{+0Rg(d>UGm_wU@jd$JDqbpCPg2p)N#W095^?*0%}YG zz)$x9bg5<10*N|uhDzcb_g0#i{!Ju(yn!s+EsyVLDl$t2DZ0ACmTCDYh!^i!Lq=Tw zNIDOkrM2P`{4U}Pm%BZV-gV-+5A0Mhj@HLPH+fCvAwr&$BS!3f+aG7?J%$}t>*4fy zX}o=3G%9%^3$=BvFoap6XG>M^y^nG%v~@f<-xT0u&z1^ztyRG`{Vw=y#x(Zz&-Qtt0jMS&cg~3d$9A}X!v6B6y|gvVt!A$;RQidaoB1c>oNo#Z{5T5rPdO~GY~FG$Fl0pu-)WZMXQsy+?J{2qo^`77h2Y5V?{e+B)sKVP{o#GTJs z#tzFsNy+gnO0<&s-fe=-w@uiKf*7nC5Q86PW^o^r3i+5cPBc&VJpO3jM_T+P>3rMu zILYHKoNSV36BMtKy4-{KoV5&2>p#f~6e38|;v=|s&vqo+SdAV`Jpwj@T6m)LjeA|} zOvOuQ;n<`uh#Ydms^r5GUVfn+Pq;J`4$PTCi*Nr&@5fuPDNzsTUtUjf)|VE%(MX*} z1*gJym0?i4XdXH}^DSp3Gn`BPA%ml4YM`jeClQz1M--zc@b$MI;?LhxOWrAZpy8Hi zW_~OV^ghV5gm1>*gT_K-o&+_*Gf0Q{2Ta_fDoXX(N@Zjou*=;J)bGh|lK)eEAPzec ztVGi0N1`neTLgCKGrm#iBQ&n#`{r(nz;sgya+}&M^tCIpF6KGxe?PY)DGwKsE#H0F z7k>+8yZ8>?t(OOtnL$`-s~Y)!Qj(77TS0;je?jZF#zOJ#b!b0|LLFQSj_~rJGcBX7 z3vw;cB!647ZjU0G&i4(K(k^0lTR?19T_)8^W60hX8%`s-y3FL53|eAoi*74fQ9DOX z!K{l~1RqQRhacx5nb!k9l9)qU>|>B(s2NvwW)hzEQc^If;R-0%nX+ZP*2aseby!aY zME7sqp;w5G*ugv*Ez0~%9;;uc<~vN$jD42YZG9@#?z;zG;>Kg>W&{f2>iyUPZVcg0 zdK2A(vAEH3C<{ofYi z@+%z_Y(=%GHayMfMH4!k4T_(DIV+-e%cz~wO zGXN#KQFwAgJl5kRq5hW)8=r7kpnA$r@X&BPJ<@gtrM;~sI&N#2jp7$_+jKT;^7w%F zeGDPlGZOF?=Wu9Ix(x~*L4vc}ECqL-^|P;CGAv(h6s@yp#px`IOFlXtCHYU`YL^T_ zWwo{_W6@!3|KOgr#ERfDyW&=O@q3Y=;n7cgTj4Dpe&#fuu;U1A%c`;7s(yvU(F_TQRQ}Z;ENl7AQ^Jc@g$0ITK-IksU9nN~L%!iJ5HF(w;PnZ?* z9Gsg!ar$q0o?E?Q`u?vTuG?^w7LBU|anuABX+9O+sW-qA&wBW3Fh|t4;RBi!IfKj} z)Ho2Ud@NA2{xj;Pb@S_VCX9JVH-&A6lx^RM>Y!8-V=oO-F1pC~o{`}C!Fzc1BF?(e zVGc6o=E0A+L-@V1j_CF)InkQ&(Il{0mW^xtL;u;Gf6w^uFx)nT9uqI7AsrU<(~}6a zwN9N1%^ex+65{^y_3Yj6BAjWljy-$yA1iqEMASaXib+KurypukY3yPhh$&6LO9+p# zAKF5E#s=Zx^Tt5sf0g*~!eUbUj0dxIorP&~mV#gR*K!99!;x~G95?FXkSQPLE5Z7$ zm&j7x4tN>zls0ZV!n`l*Av(twzg{@r@{cgb+WBiKYm+~OcZVwrwiT~K!!kKkV0D_N zZK^~2CM+P;!ewQ>-FgCj`z2zxvc0rqMJAe&c8p7`4JEEio!N~$eZrkf-k>h;zf?A4 z8jY;;rPqTJ(4z^-Al0IRHa-4H_sI32Pc|WNHGTrim=?f%R#N&kSr7MLRL6~qWwa7a zhM)^t;t?17+4Vf0o2S(jw+Cq9P&Yln-cVlW?r|l=+IPU9M0F-vb#1_I{d=!|uG)0@ zCKEV$aXKDvuF)E0$!cB5A8LeNIBFKsOC@5$(O*l^5dYN-+P_1>;p(d zuZ5chlfmImkjUfoa$(h!iSXL%3$)g%(jQ71@Fw&a`4S;TuPbOo~I&cC0bxL=nytI7D<{q<#-M@38a`fLo~eWy44+ReovtN3B2;}hDxpV z*gW~Jl>wfPr@mAbyzbIKW8DVf{=FQ|UA39Zs_}*WJH3RBYZzG>c>)J?Tph@rLdO-L zvw{k`Xv`<9pdEoSm3z6Kmub*@he5ZG%fL0W0yJpeyEGKFr=L_d^7!8ONg~JAEL?wA z8?)Vikm2Vs1G#ONZx1o)n;?){+Rw&mAD~B@eiGGqg82&6!1sJNK1X_o=7=;)`IFmL z!GfXk^v|P9V0l(a@MwH0E@%k#6_OQ4?>GWC1cho&(D~U8*27}$_l86mWI9}^B=Yi8{V$y90-Zu;<1kDuNR<052 z?h9gO7yZHigE3t3QW2aO6o?~v+|z1RFVw$O3eSDB3O}oVShn@>9e5<2$N1_fT<7cj z8)LryTW1o6XNM8_vU$v6(kuAv$YYy4Ho+x<4z%z507G8=0c%-3e6!3Crli*6vch-l z`CePG?nLK-d1Rv|Eok(52oQS>dcwM?YQ-a3@ihZF4m&{ibRRsYT83>Htb*@c62lIg zDJ0^-ch2~D2K-vyh;Dhs(re1$bboFa&*6_~Wls~$eptb!G%X~V)&kg58^ms8t)bh$ zwpp_RBk}4D$ynk0LKu1a6A|WVaaW~{p{sfdg&!239rcvRuD>8!G^H7BRenw5hmPem z?PoKcY05ZzQwIE%VqDv4C9z@mSr$<@PS9>EC5Wrt4T7i)l>hQ1XEoQ3K7aEannr5l zPLK8f?Cbw$f1*d*LcsYK*zSBLuJ8CMI;LX|$Bgp8_=7up?D_*_PfWqGGu?^C8-H%7 zW-iga{2E*m+mO-N5C(PL;!}@av8O%lY<|OD+Vft6Ntx=>`D=A(oPHvkb!US3`7EAm zchY^fzvTtBbbbokp0+}1Wf-V0zCpK$!syHH+uZO&ebjH=RIo`_B-JU@$cmSkJZ9NZ=HvhWt(Nx3c>vSL0oEVQN9`QJ(U?+-hw ze44G5$s#4mE}ZPC2*Cb1DD=>Ax@1r#+PQ8#>9ik3YM!-|*zI})bKA=^j;x6_L>FH6 z&@7V-(qOO(on3kcIT#E@+boZxb=NCtSLALs{oONk-#HCw$o+uD-s}Los z){AtDj=qP`L+4RtZcf{d8NA&Tu1o@EM zz-;cVqCX{TN#n`eBp@(`i_z1BEuYHS6TXk=*`p{fq&jfGZ*gr8twk>v!eyCjuw_Xu zxo(igGFOhFnxpo?eEaKS**A*f$hI0$v9Cb9`1={=e&jIBy?2CRg>!_v@tTWRpa2g` z&%lrA)8K&q0=nU-!oaf}i8#ud59_gQi%N;(6(M)mI$e}sag~I=Rl?UMZ$dK~o|B(< zgV=$uVp>m9=%7_=*(1FWvh}{I8K`N5-QZsL+B;Ii^s}>)5Zdk)9S@E{Z$0%_2YTtE#3?~ zZ{bo6+SFM@+Z1)k#u@{%=J+Uhl#_va@-9;6<+=mc{Id!3!kQpWN}tE_jiF!89D%g0 zr65_I4|^OkSzSv#)!6oj-rsP7<^7k*9kwh7ZCnX)Zma39=Gol8|999K#f5zxN2>R| zCD#{Y>rDom>E?ZxxV5~rz{Rp8E~0n@)!n5Dk7os!EqgheE$meW_da##=jI?%c9i5P zZ3itrpI*D%m0oMBM|RiU=;g_3;BUZdurMQZ^jrt`d-fla@X?iZIa<-KT^CXGHaT(F z*UxO&1%x!(kD2kNInxzv0r(rxV z&_cKtJQ>~h_2q4$vGv{y-6Y!dBV=3u|MZt7^nWeJ-)qC+>%s+a1!=KqWrL{F)Ppqe zfxlQ!Ie`v8^O+p9ai!wE@#uY;O%gd;;iq}ZlF&r)TeoSxJY~{S-W^x%kH?K$FDdN3*#=x9#u{07- zYh2ATF4dqmUu|LAf_JFG}P}?9noN>SHtkZ?Ma! z1_$SGq0i5NZqWkq=R^~{?8-b4uwE>*tioD@Zo;PM7IJs0Jj}2wWWjIixWOX_GnboH zV)?!hHZ49${JU^G_>8fI#n--xQh$vS3ce4aiw@1h=Kn2)8l!gbQA{9@e!4@N@En== zxeaX^UrG(SBa!4R2V_u|4);E3fq8W_(l9P$IrdWG4)rp&AUB2V%uI*X6>-N*z_ukRx`TCIiD4s2UF%u=vOTz6=SD|A0D#3*D@odU5 zHL>sAdE%K0xvRhA z#Nd0v4C*nphLkJ5BAt__!2I+?Y&6!0j?4_EmmLhT@OvqY=1j$#FBvfDObdv!dQ6hP zPox%B1ga_#{kFad-Uc{vshii(*GYqL&~!xTWPNr4^~mv7t1v3~I5Ia6X2Yi<)-JUOhyHG%clPGupEQP$$=%<4iu69#92`b8UXz9+}bi?`zIzDD9z zQD<;j2*a1{&au#}EEXm^n7%kT9k?-eq;$h!)~;sHmOV70kKPT&aSx8;*E>S+2gfFn zfqnwGp0~lD51l6=2R5*~yjI}I7@h}zzAl`UkEFWG4+>QdmZPhax_Ao4S@g`GsSJm{ z(&Do8~>CKQhkIu_bwBqvCsDzWO7k{MZ0HKZ`&PhV#(j)NM51d?zjQ z_kxASFaGBT24bL8DS0sS@FZ2SrC93D20X#A19X~?fOWE)*l*+|v4ydk*yKYGQVI3I zVbAu1i8{ihmMsWAgp-xjiN>rJ^4BgB?HMeF2S3+S-O_6MLB|r4 z_^0fEyB@9#>_@Iw-9+ydF2i=i9B3PJ2l@Ty1BNME#jU|k;_+Tf#3zLIXoY8;=UxN`(}=91{ut^GxZu(#?XNYqN0GnU;Y${Cgg+*>qM9 zj~}T@X19z)4&)%tkw1+KH{JoIzr)et_CE4FUKXYKuK>XtdE8}g4sp)CBOearE|~|K3-m7e2~elDq-6|25(H+&rE{HVXx4KB5!{0#c)zO z_iUpJUOMy>1aCQwd&<=Xy5S($yh;mWaR*kk{eo{@nV=3}cX zsn9<2El)I4Pt~uvgG(UaJ3H|;)>_$37HPK9!_$nZeE5Dm;aUQoT??!;F@pcRkUpPw zn#LYD#maY@a9cf>if)B*Z1vd_@TIzu8y`|78euUAd+%5eY9Vg0{_HiDb9W7GJUW)# z*?JY_EjbO&u5X~KA_ZPGH4*c!E!a$Z298ioC#lEJV3Freyu4vO+MpQ*F^>=8-xnGD znzjSRji14$<{OZME|D-SJq-9n2Tog9;+FznXpz5;pS6b2o~L?Lzt57*=J6`?UvEIB zt;YDg<`BM?B`5H(nn|C$6fp0P%b22!A4_pOz*by&No8_9twsB^ko?&P`1RT(cAxTj zh;%5{;x*N>d@`$zks-AC$HL~k z`S`iM2zRek#Yfj%FvOG5-psZ33bf z>LBPDH&T#iS%5E37txNv;~=?w6tvW@6h{6RkMB(J#n*NxfYQ8H+2Q|(s`rk|`HTPmn<{Ce(2$fALUp~)^Msu z5~}NU&bbIo$%Bt9JH-pj*--c@oLGL8T98y(Erx^0HrSS!}>YzbVj*k?wABlHSQVr#bzj7K2!$YTM9fp zD23gJwsY1Rx%_9(AAIw9q31aBH@X{p71n)^$KCrU`OlbCvluOKy>pj%lc1#s`u!{;as9B^u`K`G~F-%_Ix?L!4#5JEZo! z&;zO;#1XLt4Y!+seqEz-yOA?1j@xQMkTp2suX6`0}<7n-ko0Y19Nv)`AMB{$BFX9j;Gh>zAZQ1Vj3 z>htAU?#=I3|BY8biY#1l_|dl)xh@Z6Zns2iTv9iGZ!Tq1@lDovZxXfmwTq~{9tRnR z42AlSHmM19Mk&Qr;`%=~V5z+vPPyg@UIS;?Bt{6{SCS#jI?ta4sRpyKth3~}$}oIB z)l`(PV}Q85F?8dca@Zi$%iXuXfqIunba~1Z!I$qbjoRtY$sLn|6V`3PgUtwQ=FQz1o&&kV#^W_US^U+1t3fJyAQ@qJ6%5X~GxfKH^tE9${on_zMBwk#Y&Zwr zFBRC@7+Kamw-vv*G@8D9+kx~KS(0H%N8rhhFQRQdd*FtX4SrkM@Bh932~>AJL>K5U zhZ`2U?8}Np4DB>y|F#UG)crag_|z8j(_?Iw*NlO^qpq^!TDtV-p+-DAI1(=lOlJBm zKJLvMz+||>VMMkOVAQ`*41JAgv01< z_W*7g{YyICs=@4+z!dLV$p-DyWE(HdvvI6+V%t+A={Q68K2P7H{cqvbwkvT;q$Vm5 zE25yVX8u)nCp@`0iuR$6<5aq1|1|1*aIbiOoNnLW2W~hjo}4<5nNHYAZO7k-yMGP?Qa7aa)|bH5T#gMk z+SYf6_}}YJHXYY@Z=hZI$vW60n)Lts1Py(^46X*0aU=T&irdcTapWoLLZgaU!YM!Pnz-M8z8c;qBo51Q{CtU zq%E-W?mt*Z*KNB)y=2_bDFs9O@}Pd&jBDw3LCMpM`o6<7*?Gc!zTnBUY&H6* zyBb1*@A1z^w_4Z6_VV|-KJ%vqzm3Z&xuUl3_vpp`5j5S*nd+=w#%8WO&q^{qSZ_=u zRn1+%_B@!*u9umSyIDVotH%^}`Ia|rxED>m%;)gMPwa?FhndJUCl^iee#{!5KWAU= zi`hP^LEI<&4^!?|Q6SaGxJt?sRO#WlO;j|}hgv>TXBqn%X|&A??&*O{deOd=KD2y7 zz*3ry-F1*U-9O7ob%pbOGNaMO@pEXWWe&Y;KY-~rL{jPJN^DTcQ#5DKJy>`NirHXEtyNIf(l4P@Vno9LC4 zB7NdMi(VYNnOZg_kdIo6*h)b!-g%~o-qBTK&YO2UT86&lua|fk0Gwb0+UX)6joiE#G0Q>VH@X6V##(#*b|E>Os(`R=Qzwj z^l(iqy0_*kjSe10XEpgCx1pcW7|29LS@vMIOT-O+t|QWle#U>={gP@;Q9?}6px1Z4BE!`K(cJ#p{4iT7zIsgZ<4R+frM z!gOC4h{u{Hxn!OWoL-IP1%FEdi9q-mOex@AwG@YpP+f_XJYzu$MZN7y%zTpA`Lh#huXi=gaQSrvLQyq0Ks+1b1yj z>$Zf6Prn@_>hM2>mK$oJcdvao8=oWKjbr*^en8uD_AgM8?o)aWPc9)kVtF*(;TnO) zO#304Vsw%pGf0Doua0CgUI(eP0noZCWjfn;0dk0rA#rgnjIUOLdw*lB3x+IVW>=2l zg=e<-8hSYneFB;rWtU_KSMbB_({=)%JvA=$N9eA%#? zyIL5{ul6z_T~Py|C2BA4a$FtBZAhcbo~Cn?q(*W52e~kF<91PNrwZTmT>&Qf&BtBt z?I=Ot4s2T*#K{&9h_15@$gG(I(OM@&Z=&sJW$!*@y6OyDt1a}N=L|&X+Bwk;kHhG3 ztQ5xS$H@-9l@*L0Ofo~Wq5lXSd=s+ynf6&k7q!q!E0jp|2}DM3XrpuAk7L#OGVGty z1CkqAK|>A|^0iN`uz^(&>RPcEW)90_a=`=p;*IEm6H^Rd&rIJBl)UqH0v9Af39*~G zjGDVt{*-;+ezU@1rC23*6s*mRMl*CQY0X#6|2MC+e6-k|2h!Y_qvwjArtTJ)3ks5# zZ%l}#o+AGJ{T|95YK&%$2|wdmOXx@2){05bZ7VDHvmy7M`f zJRCDl_p3ZMvNq5vX@k>+QlK1*L5Te|| z_ghv>MLA+L>p&(~_M#dGJP5&tQRC^&{D<7;MsHNxJ%KDT?4>DwZ<%z}OaApcTa`0;x1Jfv+}?4k_hO@wcByCU`gVoU@ooJAWmw2W8OX=EtaB=UVtJ`Aae;5ax4G zU>13Xk<+3IwlUV3#BZ2LbpNH%7rL%UZ~Q`b)YX^nxz~pOX%s@}?R`vk_9#-9QcudH zN7MJ}^|;}KG?u=aPltu@&^$d6jT`wMiB2lhn5lnhbbSx)tq{{6*RpBM)}zR%`!~Dc zn@d%84`AurIjDT0422z&VZTct8n$RKjJx?BxZQG0;dLvs(w9Q(S3TmEpFK@S4t>B5 z$NWW_Yb)416=O2@{2OG`^#@K}{DL;^5xkIA*wT8rP{H@JfLBla$37TrL7}6MlY0V3 zB3q@E8oMvDF;<;`jXIN9;(=&Zv;Pn}xzUE^B!8pV-DYxmzQ<_ko-2%Om1Ch%-%+;e zBsR-xsASrYfs*)t)9J*pm$Wo+9*b#IkSweVV%O($e81ajl9w4{X~(xWG+t)1s4O6e z6lyNAGWu|qeq1fg6L~$RRWp^&tu&ID)+;h7=jd(qc;;Q83AbOK>CI#?d*$b-ZzXS>zcRI6S{Ax?yo#H zAkv&ID$Ay#ovv)j=!0S!BB#HPGDg+|*&f#v#V;PhM8My1BurugVU ztTA{nb9>c~8m}JAwuQf-2a}cA=t14+;2cesqEGO}yPA?DE)6G?%tlGm4lwtBdQ@dp z2=l10Wh0eR@zC3CFnxLkHjLEARx4BSq7hfwnU^7aqu8EiJh?`6ckAKV!xyqZB}K{2 zyN#kM|GlVNT!W$yzM&)ZXCDcmLD?fi-{r5yHN6! z4PZ`(vv<*<_-~-Bq(fVQ{5x=s`z~-0edS8PJM|hjaz7yDxM+IEg|m^BYlFdB^{A{k z8TRy7WL2?lHtpA9u>Hzd+Ah?xeRhq4oy#|1B8o&KhTlb>qn;oW>G8P#5H)mym9ZqT zA&U8-!nEe3_kD)RAcKu}wNaOZ)i6K%5{`;#L7Cy9wEIvMl^LxCegy(E(s?K9`f(Ms z-uhyPFWYE;gKlux6wd96_yc}=O>n|@IkcDj!5&W!;Gf?Y;BL(xbiluWWbQZ*56iOq z@-^go00yY6qWmWnJT`qX-n{%5UKB24Gex+k_fHr6CWq8R*sgrsSt{I)3?E`oU7c= zn_uCFt}*Wa_!9gb{uJ)-RYr+xmVsq-3OO@;Hn=@_3CHi8rk>q};1=f1#V>Qk<;zpB zX7C;EVQvaPxC0*b{Epx}A}XS9`} zsI8CCqs3k5%x_g|`>bnp){<9f=ImW`N!3a6u_zk;2%hsZa5i|IxDS4^18DR7V!=~s z4oW*X1r~)Kg|}C(BUzV+Wcaw@xMb@Fy7`nD%6T<`UNhImhfECcxEgs@_dW(HcCCWo z+B&$DZ-c$<)M0dB2C8484_Z?rKtKHqdP^f{OkjN9_^uRQYt-<8c#q$6ys&8soAhxY zlW#Z#6DLmMLK@f8Uy{S*nVmgqJ{w6jiZ%IPN;c$p^+LY+Y%blVBm?ujPqO9*f{sG9 z0{ZRr!H=c{;IQNy^j`NszGFo;jPsjF!foQ<%cNlZ-Z&D*IZ!D7_y_)ZHNYbIukhkV zC#km3!x5`@!0b#78hfmhT0bljU;JW#&z#AF<IjwV>LI7k-6m5M#^SMu#=*QmGaM520QqfnfGG!b(7L=su=(+5th}-d z6)8W&YD@uJ^E)8q#W0#apcJL9t%V<_T(SSZMUYpM%GX7_g@h|Xc*5qVz*X;n+{d$R zaLzp(CosK78TDsh>egZBp@DY!t<8Jnc0(YE?hcoW3K z!V!6pKeY?oy>oG^`AmG*>J@BXZbDXMp5{l!B%y&DYmv_FXynMRA#bUmJWJi4&)~o@Kb&*%5^nwT44W-4!GmHypw*&yuorkSw(6ExTk8zG{M>*?jTsB= zKdaG==>m)MN+i4=8V+WgVzJqDH!KS7q(wvZ=+8oJI^n2O+Budu-+T*C)W>7br~x+9 z2KC_dvvpv!@d~Ow=>w}*ak%%&ViM46Zqe^kNOiS8aSr zyKw?k)E%gt|xVC)he`A@zJ^(HASDR}T^$d9TmV zN@d!^ETHnJp3PtLdEmP!8WkyqFav{|f|tN!nByi-EY8Hy*um+v`j`vM%#^03)6b&7 zBTu0Cz)8G#_YQKwrGqQpdw}28L-#9UjDI~#%C%sxg9&61f*OC^$NeZc*Ue9K7* z-a0dWBit`U3_Z!`cn!nvC`@_^UY>f4cn=zk!{$FOd3~dj%*mR{uiGc|WloLa7oC%( z-0nI2>cehSf1w-MKX5zft-QhY+9%^Z3$!45btD{Hw*gAqOW>`E2d9$v1l8`Tpk$;Y zaT$1z{yr4~PTQO4Wr34%+juLRx^Wn@lFX%_?2@f>ubm~P^RJ@vTXLxC>t8B=*HYrY z`6(M@wE`^hMttvP9F)x&B;K1D3sL>TkbRLBL0tchAQ;$`8 zeuwpMFM>q#A=+%##uo3~$TH+DkW&78`gT|{TfMU%F040ZTI)3+?sf*>a8-{T&=s@@ z*N+nGf|)F@>=BaJ3MY{=1vo(~9tWNJN>_L+gRsv*T>tW;DC5)#_Ui8cXg8}T@O|SL z9=nTQ=~GEwi#rJ}YUdW*a>6kej$miK7yoauQD=)$BuL+9CTphhlPesd&Vtn{&pH3oMz9qWy?^hi`(g6y#yBA z{sBIv$+6EXZZYq*1SJ)nC4h-dV81cRF#rw+CRnapQqE$D{sMAyMZ`-&Qj#_ zOdVz|(4?EzK0tfwM6A57kP752IG7zmnR&n+H2Av)TR2GI-He!DCVTAWT(Er&Q3Tif*5UQ={*I%Cuo@Em?@)ZwKkBn;PPm7#KY^Y7dA z!M0k^2nwY!m4arv){UQ4HHpS7OyhreMe^~Y5!6EH@fa$*gmt*Z(yZO>+~Mul;D@`= zyyR1aJaiS%+^|4y+tDP5*e0U-saCWjUf2slyWl~15o6p1>wr04boQuvzNzvh-Q(Cy z$TV+MdB>gq9_C;Qs-53+davSnV zQBEdx-*3vs+9=ZjI~+jgdL}rZ_F$SOXQADBDBk%i6oPy+sK!!7syfe_7u8i$mwX4d zY5E6r@#!*XI4*+^=t#pi%Q0;IQ73ArR>2Qkyc)_zYvAsTd*Y43TiLkO)p$k69a_29 zR5Y+4n%+H_h9-KPCu@2lNbmA1Xx@%om`T>7ijI2jy;D7sD-MIP8UrM)4-w4U;RcV7 z$@5aCfnXnL3fDlH-WT*{O8H*!YLgu2zJ5Mj?hX{c4i$KnCf2m=l^%_-E{3(~PJBY? zaO&_;7lXePR@)W$-~KDLcdu}6pS&4f&b-FeX7?Z>8-@@6O{UAXC(@*|>d>3z3^}h& zBp#ZXboO{g-fzfkK#H5WQG2I16heOhm(&YG91Gh*mPjdmE~vOgii7$Pe1Idqz{6YL9f?Vz-!MVqtj_@`prPelhVC-Av;bJVpL$a zlQXr?RHxeRlh~-+2LxXMJ>U=`kG$Jy3>tT+aI?RrlpMn$K2=Hz@z12hYENYgvVOz0MijIw6~z^?ZnAMyxR| zpVG#TZ$B+Dj&YF0Otj%T3JSPSGym`=XQub<+s_s;vPNcrD1OcGs1%&{G)|0pBCywN;=eSO(LYlY2e&XQK)BOw0MZuC6RkgC^zC^40T=eiQhOw z9k*tF0!PDI2UUn9WYXtV zAoI6?%B%>aT6U+Y^ksd?trO2!%Z)4S>lXhNqZB}i{|&dr;*`atPGvZ;@*KsszS_l%?o>myxr@YOKq%WXTJQs6 zflQ`r}WA^Oh zaTui%CE2FCRPvgi!A4F$M_R7svzzOLzOSs6Bry02f4o%{e5IyQx5C5h>sxhp{Zc4# znX?^+IxJ!zu9u*FHXd~Ftd+EJ@fW)Bcnb3yF+}42BoxeQBVgynaOCi!l`4JCU=jCt zboc0DaLX=%V9O|hZxKTGr#+*CuZ^Yw#or~94hKli$E}onZ{tzLj3NBFkwUN6zYy^; z$!=^uJsmzgAH^C49#B(?Ej8+1KoXAZ;YSa=LF<2~!8D{v3wJ%D8RuY?=vfyn03 zb8i3Q$Do`MNZ)PCq&pKV@V^Ih@!n^naBznL`*CI&7)`ej>%~0;n3BuQ&pC%4*}ddy zl#4iJt{Zu{Plba%h3xL&<1FHO867J%3{AX0spPcfT_{n_7JYVAg$Wx<+0JR10?@jH z?(WlN-RT3VS~nM(JOUt6Vu|N;C<)w*4j6lUN}sgF_CGbT#mf)$+rfU+!P|n9bzg{w7zqr=dZBM)O)2W> zXN!HWZO3U*1ay^O!R#6KaQm$#ZlNCq&q^!SUvdvpTnFGk8%B_Pw+=R?=`i`SLme+x z+6`sWnlRBF@U5RAqPvTu(II^m_}O@kqw8(R)$~L3X~jdy%~cyD{gd7Z3`jR>JLxxB zC4NJfjn2R&vs+;O^_95Gp$85h2!zEOOZmuu3fL%cCc3{o40!c7cwj>@Q?oT?#Xq01 z+NmLuC90+L_LNk1d0hv)TOBH~5}2tm+#yKb_fBAo6ymH~&uD{A7nD_v#uKeu!0u`& zGS};20h43cgf0bW_nL|JCmcgg!ABrSZ9e=%`#HI4Q~E?dm=3bC#J%4Q$$6RS_{A=a z`-y@uo%$Xwt;hJw6hCxe<{I$(Vkhk3)3AK(7>IVtBy#?dsK4}kL|T4`#@N4vH_tR} zvOa}kZtf}|yKZ8IMJaqrN-#!lj!b=1Gb#OHiFR){!Mhjh;k&;x@x_BD@m{n3)UVYH zyIh=M-84iW*Pc9yCj0i#$1i2r1&M^J?2*HVW3q77jcCaGl!XtE6k*LjbFk{NEUNK( z3cEG01}zfUM&7D|&qsGK#H>=mS&^SD$4(1o&EeSUC)8t3&%SkJy zw>pUK%Qu1~lMrf{nTV{%MUtz_f5Wk&G1x&E8`G1M*^X0PWcL0y^ycO3EOm{n|8l%)#(R9b-Hx7lH;{zP`inFquTf3p9N1Pl7#D|q=HHoS zaW%%#q;mEHC^hiI-=1{C=g*_D^=f63aw>ot*BWBiI8M-sZ$=TLf5F$oLFD3?`*5vk zJ9V6$yFq<+`!%Rhzq6ZqewY2ijZ{~r%Ga;M^2Ygs&Rei--nOAZ%4 z?y_Y3hD6DKaY{#aJbe?_M8~UMr@tcmv%)mYJ}InVzT57Dk>VN_u2lmmD-4+JC{sEL zZlD{#cH>6vBti2ViZ7Us$MRQS(5;7@nai$m)F3el#^@(O+hQkd`uh|Payx{NtMx0r zb5SH|8K_QAj~zx=HB3OVuR|g9;Cz-oDV6wYzGF8F3s8;1aj-p*1M>=AkzXH%AY`Zv zT4(L((5J&#UW`x=5$|M2?`>sj9@bRVGZ1Ar>EQE$sd%2ID(~idf;>w6L+6Ys=0bG> zcob&Rmj}N0X-8bo3?#4DLh^RpvbvM80$EBS64&az;)mo`I_};*s~40HITvbw@A8eGr$Xd z`qxglPZ0J7|GO_qXHf<+;R*Sw9Zu2yNaU7T2zPSMaUsly92L%rX~%f_xLzBDc?{&I zJ0yuBb9(5x&FyPD-30C1I>`i|b;zcl)RvNxD3`=%x97!8^av@9Vjlc!1uLM0Dm1uW7TEZZGn}o)15(1oPS7CZHD5S=K~p< zF!6u$pT79>gPRI_wrH}!)phigqq+cW|`1jv&;8XPyN!gu2pE5R)(Q6F(kU8tv=bM!zG<7)q4BH2Bk6Ve9ya~zZ zzQgt|EuhLHf`#?*o%RUzr_Ap=sJrZZXw?V--(q{Qdiijep74@>o2JB8Z#qjm42^K5 zn_yo&v>iumEx{Z9EQVV@+;I9kEijfF18eWNpnvLNsQXPmX^|TXj-wsO6E!t-A*Kr# zl}xs=?k9!&ZM_P9zRTdC!fo+PkKuTbb^$VV>W|wl+lk~?YqR4)1@w~B%>Rw!Qg81* zLg8lvK=HX8xKvn^{1RCkrd*9{gRc{pN$C~ zKCAyF_z&yKV>xv%cG=s3jYwxS-^P``eE$mkmaN1>uFR$32~*(2BZ+wTzHxZ?R!&^7 zTnEK1eT)^SpT(T(d@S#-hbG4^pbPHUL4)N`*dEr#J$^S|JXH{gYFB2$m-YuBuGeL{ zDO*@fZyA4%#zO9ZH-ue~5xKq@4-2+!1Ih0~G&(1gbAR&)lK)J=m)^dIeJ=)yaC9;@dMi zcy2=y9A0w+!=;e)w|dG+gPV zjW(tNjEO%BYA14$L$Eq@A6Fx5Y~y+Bdp%^~#a8s_)e9stFC`_iTRCS3j-Tjx4OTB5 z0yDJAQD(&(*tQ`C^Gz?PyNSRzFq=tiW)_M~hd)5f@;S#VhEj5@1F2>WVe;LYsOM3S zb=|^+Q2A}4;2GqKh9~bAH4dpD;VVrMbFL!wo)oF^0c7l0OgV8Ij8yvrH>E4dU{NFu z`60!msxKHQ8jYFLmlIhE9!OJkP+yNcUEO*7K{IT>rKS?y}v8{!jzGXXi_{l@=BMmtUq{v1XepV!-F91~*`F zzPPd11kH9Rc3orekm$D6Ah2kGSQv|akN_U7x);CtQ#))lyBVu z{tvZP=e>b1~kQK~*9Izx&(ueXP)XT<|}_`-{Oe`gianZ1sPO_vkLNlWN6vkq>qz~R~He2l7i2O(S^LghPN zz{fme55l9^*71|5 zBz-4}aW+6GeF4DVZ|GBXf8=UoMMpLEh&n5o%bQe%ji+mS*D`BYH55OGr(Pl2?B>VI z=z3-&Y*}at4$|jYi(CSCu&Ik%eoYVNKR8WINGml6zd$

u}Y*?`cKmZ1S`1FZ!t} z^k$x1Nzxz6lOy-txEbLA#Pe?{nR;RyvrT>Tn_KW+-u$=c%u48{1lJj!#c6 z=l(cO>C@(Lb5AgRe?k|%8-_bP%$VbC6ZZK@abK?Bucrfhx2zS-tt(PO-o0ub8TH53;$o40nn0$btkDwy0ATx5{5+b3C8-#U#;T3GID56R9-bhV221 zsdcFbGk&#!y|{LYwy&`x((c7%qrh?C<~PGkw>;p=QQ!OOB}?e?Fco|@rjEBx2Y9T6 zaN3Xx8hvjw@$j|d#+|L9?)*h+xp9g3Q)fK2U8dgmTO`aFN*8=lM%w!SP^9}BUcKuj z*|ftSzHK@}9~#uqVb3q1ovr!-6vt3 zZ3;8EpTm;XirDA^6{6u8%{}^_L(0=$_fDyGAISrj&_F63cEt`Gp4aE?*4QKxx z0+k!g(JvuK7AHJrgQ^EH%keiUF%P99yKbQOQ+>(J({YdxzZw4tTZkj)9TDd^^k=91 zt*EwxCK$~chadMmhY=39!F@*{UjAqTzTeYKhYieOqDv=m?Z%rW5p6s8B_oD%r}e`~ zM7l=u)@rp#g|^9E;C-ykga4;R#Fnry`3~7xtSj$v4zQ= z^{nJ(78~eP%A_*(GmC?k?DEvh%&Fg2_VtQuxR3^9>4{`?{B{j{WN-*Z zaa82`O2T$e4uYEUp;&e1PB>gRfZS^TgXBd2!;Jp#cNQLjVPe7ay=gc*WjY=_citj- zg<{lOtcKCOa#8-6=X8y<3f|(jl}otcNe?;9V4ruK0+kDc@Q6jNFe1YnVjq2EKJv9B zCLkY-Z~TI`@_72VYXI1^Ov1}M_4;CA#lR{ms(D0{J$BHizK393mMpvbkE6aWiR|}M zCAu_fAS;VK%@Ma$5`RDfETm7d zGEtd6^_j*+H|NkV-92RE;zGJXuNJz!dLiXoDplYrN%!T6IQ{xX6s?htYs<#l)HP0I zn-y)SS9Ts=KWHL~-zJSc|EWRLgG&0LQ5#w^YT@9iP&j|^gWy?jLw+T@gU{Iu#Lr&U z#^T<2?BP5H8uImtLRKjgRbS@1`h5hw%s{3SHw=$34#R469O}6+n>`&v+4i@XtqqEY zb*2Nv8q2at%b64Cc>h=`dTWDL%&>w%pVjG!7l+~PrX#HUyEO~hH-XB}5bgjK&fYG zd#@cnSv?Q-RH*f>8I`X-5IRX43=u&M)_;leh{^b@j}#P{K0y}(PSH>Hp5UMFOI8m# zOB;4%le5M#Q1-2pD5>A1cAk##@=hKc{5%}aUzCNtM<^M8tP0(m^NRk+I4Si1YXSP^ zN^T>8+v0MWw=Ybkrao$T@s~4n(XQ8AQ|AQn5##Gfeai*-+|dfFj(?^5epb`0`rA|# zF^!$~--hI03*KbmAL)ix8MaD(039Q<9t8Vru-~l+`b`U2i{)_`lo^lS92|l6=>9>b zC(gi)kI%@p@oxV&&i@yGF8j$-%TL{?BTEF^bGP92yDec;!d`Oak{_Gibw(0oHePaz zbEJ(2e9&U>rQcp2g1awn@)Ih(NzCJIs8e?l3v*S*zrLP=#E?$3z)=a8DXZd^Ip2_t zmjZMfNWt^=EM{Nm$#Nc=lyn#AinYddpqFat)ao8a-kYAH@)=^NcIsf}RUk1O??%?g zX_GxuC({*oG;yMx4AwSS$dq3vLuQsDD?C&Kx57(kgV2n0K7BDf6V`>}wF#16{Tf(q z;uWUZFp+&6-yf885$jj~7QI(f>eB(F$`50EjXvV>x3&-u)19bEzW_2e9)kNbHQBOf zm(hD&xe~QC@~qZtJNvTr1Bbl|>3l&@_VtPm#O$n~VZY=g3l)-R)s8%B6ETY>COstj zH4~Z6k;81i*hn;Ce)Ayb&RH=4V-KaxJ$r%1e9tibx(M4CUOi_5R#P)v?1 zTBAJ(@>TNjc<&*seW@;WR(vQn`u3TeZwbQkDVsQT&m4BBtB2`qTFRW{;}E|1ji{-4 zvw8hPsc)~418Ci)M=R~nKRlh-E030>TT4qeJE^dVBOW8^vqOmg4+|pJ(~z`Wwqi5* zAnSKHnfR<-LBnfJX=B*lzIh|lagzIMumKE8Q{m>dXNbQNhfUu+hA|0VaMG6GDVf)) zd~_gEx+!Q}W%C7fQWZ@OR+f};E^~6b*vtRpBdXy@3$>9s#=5+Lc3T@ zNIAahZbrlFby!ekm&on(Nn(EUH&^c9L3`e;#BZ<2vW(&=Hq+4v8Y5rQ!bi?*-p44g z+T4bu#z&)F<{~cS!zxxdDTVzzR{_RZNpvJJWR4S+kX&;N^JiwP=iG0+Zs!ZIn)MdF zTWrVvT_7yaO`h@Qe&VzBx_DgEEq1Z~15FCqfL*St|ZcnrPDU z3e@EifF0}W=pR!B$)ENprg$Zl{B8NnuQ{*~4BIErB`MzQz{i(;->qx6JM7!Mi~DB9 zxQB0F6W2^zh)HT9kr#Afzflft^)o_WZzvG;)B6Y!^3b`(+HAm}Bp9rFg>H#cWe0a! zv+3iGK*qfb*!yB7{hcxjH9YQu^8$;0ce;Ym5BUGre*L#*jBx;UJ)zn6{;FmRn8IK$ zW^$aN=z>N%>8Cs-{*us{la1KUtK;}p&-_F`uTKX#yC$yZ>lm^jb0uqjF-G!gNI7s@ zj?!UUXR#q|QLJntAs=Eh*vhv_EbQkeXnyFH7@4nWBEz1!dD3oSBsswH( z5p-n&A2MqDN;-U$K9l_M1P|kB?BDuh5GSb8R(u-;({?q8ujT2`pHedT&$U}*!v1fx z;_qIlo1n@6AD*r|9?R%`TUjZiLM0?*WMsV0IY&mNGE!7pN@-}IX{D0viWG@#p`>Vt z=iKLwhKMv(T9mX)la|)+$@lmD{_%d^&j+vf8TYxK``q_^UDwN!OwMSL3cWtcSnwT7 zgfH#6Xh3!u8WQ0I+naB4c2~y1gEK|g;nXPddHYG^)O(5^d_0P53YWDT_P&-x+`Wu8 zDMrKU2OF?&?hZ6@-E1^+_9*bux{Sr``%v46LHNrI;C{6Z#ASl_1=U`Fqugp>*xNT~ z(~_lxswCo`s0@;L?J*jx(}tQ4nZVLuJ>ouLHB9p_fEsHFe(Sw_y!fjGekJ{r#|77T z#pkxzKR0OzbS|47`Q-fPS@&zdGpcBeX@yp_owA{!K zr(U+lZBJ70(D9DMZH+3#>a}ui-O0$)KMKCqR-sGhzKP_A0d-!oo4#ygQ2(`x*x6{4 zZ=RdU%d@Uz_yaSlxXND8Q^aIu$!2n_QV?Mm^wi3&)k*C`L5o#30lm=0ME%u$ct7lp z;Q9C%%LcTN`vwi@kKui={WY5`opKJPf6apWiro;fOApUKbCS+|x}FPl@)jJWuF>DC zUV)XvXq>RM3jffrqOVW?K=vc9(_teL@uh>KNTSyuQr|sM=(9I+g>)AFOWtCO69N3b z#yNQGrWr)4Y6*NYPlYc+eD=2!Ja+H;fcCo4xU|jP*6mgtd1;+U^ruOYYK<4vhSx!l zT_-T7V_%?F&>)b%)-6ztwnL7`9vFJO5$_mNfeyU9OZUH)ho-XQwB%KPSYmD>U~(0| z@NN(>%(JDdPoLw>&I?{%ej{j=Q!@VfcOhBxSdXl<;aKCY;cVf=K)PVW9eDWHADpiz zz#ii(BC~CVfS<JGKlt!Qy@FtSKyuj zhcKV4!Je(%gNN10313J@?0=lnZN7xAialYV9y|c$5TjpEDSH zx&)Di;0ofh<-6_sE)Hi^z7u^^`)GSRDIZ6PvZzYsDlqq&z$Mpuv9mu~_+1Zl$c|ek z#3XDUj+^ume zycyCecY}+(0-IZ$%p^&nV z!rE``NaOTPF86{um;7N6U)7O^&Mi2KI{oKzTmRZH)uD~_# zEz+R;EtU1E?qkc-<3Z8zB7WJ_i~e0l^$C~x!*zl(noBOSxG6z5IHke;Nip!N9|hg= zkxXF(MdDQ>iDH&+Up;KRgBr6}UIMkN17UyiKoXe}Yzx$7?6fn3n{VN@4J9J%fbqRQh;sBUZW+jEy9Yf^U94s%-cuI>}umXD#=TGwtuN zV}&!Rlk&zRP8FiHkK?G}=X5S3?onS1UU%uH{mX@Z_xeHDX?F$uTKN+W9DRp9l-&8M zjY4j|=_0S#j&8g2rVk$EA)SeP2I+H0o)~ib@&iF> zxEc=p@c~~Ad4N1)^Kih1-5BiZL}$0TK=zgcU_R;!zPD5ooEFPrt&$X)nQDL|_C#^| zL+h)v+|JUD?N*SuWzN6#5cU5riG?$^z@ZX(=oV%yTc4VeaYh@+ilPa4GZ$Tb_^~&# z$Zer_`Tj)R>n!vu9K|lpx!8CA{>9s5pEYdU%1n50ViZ>up^9z4pSIoe`ZZnhq>DKo znnJb(Eg^<@Ayjq_!b%sOpfw%$d1K%A7UywMutf^4o9^q14X5J^0v}bI&JPe?GIOt zcHzqw&vAI2vE9)#&BSeZGIn}2koZd>EOC7V9QE>H%W4lpM%NAO7b9q!4SI^Tn|k|j zc=*;7{CAcS$(M1mBPAJxsjnlmF2_>oI&$`!S65}o^Gm|<=7fTV>j%8^Yb>5T)d-&xr&9}=eRzv# zB{*ocqiGescy4Jal3!_#4#{1^Cz`I~yzP$AFRveY+j#pncVHF4N zvIs*RR?A%@rgux}I^zMb>Q*cXlrA8}vU7<-L?by|CQxs6q1<AFMv zaAi$LB2BJNB$27ViC=gUVPO_Td1noFF$~4+YSkbSKHjGL-W3im$uMJ`7qq523?Eo~ zuWwEF+8c_VPL4-E%IoOyvl2-4(I9MNc8hxa8cMRCKEpvv2jix36W9r2B`(e(?9&%T zCO7j1NIx9Naypv0ZG~^)_WL3D*vue^9CD55m+v7@eh9}B9tlex=@MU;3nVBtkMuZJ z6Z3o7#J#SbW{z$~b9`-Z{e)~hvsMoxdgQC)O>4<(i$c6{$~>??NQqWv4T-Ow%s#vy z(1#KK;@`f%IYii5f~krj&M`TJWU^9)+~zAXk@-wyCRO36Q%2Zf*nFjBUxIo}Tydy~mr`{2v@7yx)uMrZ2%Qi%PJI_hey)EDW*|*D3}mnz4|pgah70ztkco3sYC6?YMjPC zQEyO&Sq77q_}cebKY>t}h6eD>Qxj35w+h)EbpVaaU*Gq;|6M?ldI6r*j;Aj%OVi_5|c0B2FolcS^Z^7>u#V{}qW9{4ic&3^fowj;E z4q5RFM^+8Ro8|l2U4A;t&iRjl-M{;4F!Ho*>asW7F~cPwcXt<&Lk0NA)B#v?^>UQ5 zZXGWBo<_OC9+X&^iB?g;JMFy&d8d)bSgQzbJmxuid=|IVL;)r*PolMVb+Es93-)-j z3iq7)4vqEkJP=JTQQ%kyS}Vc#NjLC?i;iTS*o+LZuY|HoCd|jrNIYqB z6x^G=1cI_!>n-KTC`XG16UOLDPx>)jAd>wA@Y`WIc^n=O#u?;Dl+i#3C z_L>Nq`2HY&b~Md-I)=P;%p$Mqu2Bg^EmS;iEnYSGKfM3u5IaGuMv5BMY(WtbkEnLB zPxUabPkfHYDE!+{3jW;#xnmw&?`wHBK3)T@xyph0vV0mkz=ir6#S2@(4UpM3M$}pQ zhlXo*fH3vV^lJ97T}Lj0Th3d~U%?b-ENtk*rE8To5bUWB%P$VZuHiat#;*NniIX`K z9Tj{^yJP5+-U6D}Q-w-6Ns#`1flk=gkEKRmK=UTuLendE_3htkJPWDwID!tnDdj#d zdQD$RJ;K|TJ)n-w@AzSoLcL(qTHgADIj)_3w=cHOy-J1U31isBkxWD%-NP%M3EntC z8K|~UlkRyF#tPJ=`_56Vs+TLTz5`CbEcnYe6G1vb1y>B-2c3fcqmyYjlnFB=i6iCM zGc9BIV6zuFz6i(e-cgW&G)c>zfsod`4_1zwf(v^NQdlYvs-*{T*Jnj$RkD(I3foKP z3|3;O{SLYvoB>B$FQJ&ri;-UUOYX~hExdR2PU`z(8-HbdF*dmojPvKolf0t6c!X52 z-~l2jmfWSn4n@DfMa2y?%JBduzki@T4!h99#sy&Xel1j~zo420vFy{60(f!#0L(uY zO|$Yg*qVKgrp|&^;fKl|aNQ^mkJsPCGC^V}5d7S|hU&6c_e+G?*@65I2^Dt7V~EIa zZ7W)_(T8uj;)u8BhOyb(omhZv1%<9txZ5VEdQ5UJ+*b6(J*mkwX1X3rNH@c?o@~cD zPd=bwT>*H0g$oT^BFwRDJwPvQdjP>lPebm#H*~qOJd1V9$N3k_@L+#OHa=r6=X*RH z^bVZ)FaG@7gGr3cmf(I{7Qxw;5~}%oSXIEAN+fyy2_AqxF#Y2eprmAo#ho*$>Mk)% zpC->jw(g_Z756~PS(!PYCD>Q_8va`&hn82qhiRP->_UlV-}B*l$VX1@m>1H0_=M}H za2T)dy@B$af730uQ`veA39+A58vP?XPvELOqN2)7*r9p}^8Xxy3SNdU%6>*yZN5i| z(+6JXS}@XHI~~dVoB)#RW}w5S2f*^eVq8^;U0%8KV0*yrLQ3NY5ADlc?u|R}W=Be-+~0d*k85 zF$WZ`IsmkGU!*4=CD4R|ojCgaN}5vsj-Hkbgr!3rV8yaxW^^?lmit5?OO<`1X7uDr&Hv!yh!8 zK}3)Lh9Bsi$!&dp!;n2v1!NjLpHj!(orqAovvP{-b@^yxAOJh*KcnS&&XNJu5v~JhqjWrH|GQcA$g6_}M{cr}(>G$<)j!c|g(aZ(b_8j#!XPk|b2c9HRYtZNY9PHzmm(Vh=x}*Y-jm-njmZfJHU?SVTW#v?2}(eo&85PO$p&rVFnn~Zh(_!D_O_47j&KfVtm7| z6?R@f4ZV}S`r>`I6Hj~0lUbu$KHKvm+BP*Zvk!k8edfSkMQqvMkJbM@)4YfK(&3ok_H zvv0#Z!<~KW-f^h{m+p~6)t*p%V~!R(nDQLDoMUKqd_O#7XD{Bdz6+LYaDtQf-O#w3 zt2i5*hoa|qvcZ129NQQh1!1vK%=_R`7I>^59_YUu_xo*&xkJk6Waj`p+50fslG99W zJic>|A+kiy={Xn~U4@S|JUiKKftnQ0&;t>X|H~i$;m-~EOZpkHsN z^DmPkp`}%l9Xg^&C0``LfDlR8>C!-RhDu<5s2*tN9k!j1rvm450&-lFc+KT^=qA}~ zkhJeDcU0>nW?z(HlfNym^HPi5&e)F@S0rH5Bfi-9)M$L}Zj#{NE=$r+6{Ee@W_W|d zG-fUI1;bU#=x6_@)O_^_GVh)`cr1`-hgV}LNq<{?q@f$EH#TGSbta^R`@^%fM%?>p zuIO5)7CRa;4~ALllkxMTux+8xgLO59AHQdb%$E$qo!E(b2KUmGbb-$==RyMJq>?NP z2b8&BH~71EKw3XV-m{<;4uAAwZ{#fnOO14TK|>q28k&H$hAR%8(uI4>WMGiyJUl|H z3*TN+P4nLb!lXuB(CHV#&v?As=8exLxUj|8PUfl{W*>F(Kcn$&ubKX zz8Hcg6u;sN$0}gYMO*moA@ZdB!5BQ#E|iZi&!I~+>Uoojg}7_nI4&%{u5S-AE6tfd z(&;Yx?)9>IVs!!pPRT%L;WB5v`V~q$V}&)o1D{$Yc(}MVpf52RIK`tFbC)hcyo(7$ zwC6%^Y6M=6UQ@L;H?k%*gI;w?r+y9+qJ?!9WK~NFX*U@Gi9zmmjS&`h?#DjB)#E?$ z@7D!nbBMZ~_A+Oz(Qhs{p*)5Ozg`1w8ELrxMTIkSSe`? z_)SZIklG;Fcqop#Y%C#n0>I_~=EV&{K!OXp#j? z*f19d9kGNj8#bbnrgwdqIpvBr;+=cYX}uhjIWdpEpTCEM3>(KC4R6973yN`CZAbd>SIspPhkTCYwk^*@Ga3)W(D4~a{ox2Sb4eV!r?7}A_r9&Z+^~!v zW4M)n)_t3NOA!&tu)$3Bu_pJ|PzHXj<6+C^k&yQ@A2){;;)ze4*tXAC+2!~xWbB$| zoPXmTI6jhNbB<+*rzrhq28uCM{N*(qUo~I6ySYERx_1=iZzX|8`F8x$;W^A|v&FA= zYC_uS)#9gHc8KSk?Br!V&8wAr?$SC7U$XFiFTImIT>NPAb4FwgnS)jh%B%^4uuI1w z1Iw|YG7_|*x}Dw7jAXXEmD$FF0>3%WL>yH2AAhkf;d*}Z*-mvEH>2MR)Wj=)&lkL8 z)1dg%EcW-_PG%IY!J=y?!1sDvoLB!%^kzc@KKHnxdfG-SJ}EQ^#pz!_KOI8^ueTM< zYf3%8=;~zJ+I$F3jDG+}7Y2x@`$&kTg3UOO*Q*(>?=SvoEiYDy4o8~WiLh~G8Yo*5YL`Xd%&!B5zkx9MzBO1eS0<4^nfMgMP%H2y z&P$c4^QW0C{Cf)X5cqjx^>+gu@C0j@mC{3ko=WKP31VBDVs_G^3`Z{NPbMsy&$cTZ zq_d?4v-Jb7(@9^(aoT39m`w)9YLg|wFna{uY55r~TcyUGsGkk{3lC7A#YHgL*q<>M zMX~wm7PxBIOFq5{MN!7*(8#nW_}rHhEODQVc;6!nG2gbCMm?ybRqeZR`{83?nm>lL zRGz?c{x(9Lx(;ehy#df%azQB$3g4*X{1#QF{VEirGd_6rv|@JWvx)e2_i^MWcf4<| zQ?hkBw3-aSe{S8vw<>Dc&4Y*7`s4RlgSsvWR~f=CXG@BAsBld7UI2|9Rl}rAfAA@( zV3szL{g5zP&nf}Ubrajexsl=Fyhga9L2j#k~!_J25&5WYi z2OUUHuc_FqdjT`Exz6NAt3jNgKd62}0sK$KFvqv0Od0#)_xIYML%)uVJ9`>;{a(WK z%MP<9y|eJLd;+_8k+7`ZJT&RG4|h+thMCH*W-dMfkms-t|1b#b!_|K{{_myo)1+#F z#w_-Asf0K$TbN%I^`HscTye>91M%8~Vs_T!IlH@flsNsYKJ!}^L%&IyvO~rXMz(HuLQN*qD<*9C09B5=IGQ&4r?EIDP z>TOdcsL$*@kUF4*d;7Hn*&fEw9TL!oYyZwA;phm^X-j4cB@_93_T|)gR}Jg$gP2>> zVW>W>gnw6zgoVeXSjYH9pmxOutH!D`=ar(q&yW5)0B+EUEVybFyIVUIUpUZ0cieqW z->O@ZkoDS#lM`?vJ&`x~-rQH)F^D&?VzI)7ekc6Wc*BCnKqdX;pr$}Fe*&u6tefBeIb6xKS2}MfgL|n9>d+aY3RL1 z8eiyiuxj_P^QiIV)4m+uSVIXWk5mB*vwm!M{~Ek&)CR2geh(i>+}K)~YWP;9!>s%N zfk@-aEGnspZLtw%+-EgG#m{9`1M?`3`8tEirKy5d>0fyKScWu-z6*7tUb0EzD;%3# zP1l8$fr;k@dfQ7!@Z_C}qw7D>cwyEixWJoy?=L|D+|PjA{%LHnMm8JXzXlv?>u8Zv z5?8S>hmTd+iSFmtv-?CV8;HsT8d*~YbO-_8c;Ji49C-#3924oF2wD=!Ok zFZZDHYc#RnK9ziLKT>ts>Ik}8rbsfwa*4%)g?Oa#I?1my;!0TL1C! zKM2FCIDwx4pUsen%3jSUnPNe=RNE<5~iZHs#cIzc!QxAIEuL zRG9CW$i5iX9UUZANEjrRbG^!X;(sBllJC%9z7TBA_0WQyE$HjTRnVv-D-Mep#C)Ze zktscYSmkk{e$foDc(E_(GS%$EhDGCsif!6XFp0^NSf)ueypk=)+hT)J%I@{(cKa#* zQ9Fg3AD7VX-4CHGtCq@(XOIvVK|iPaJgk@!%c)Kduf7~=$7Ph}_T_{hHRn~mdi4nL zv$yf*_h$B8um8uNsGO%}Ow!Piota$;-#(Uco*swcY?R>CIN%iCy4b1jczl=NqPk7N zeB|?|&^t_$-MH}*Hk*01 zjzrwph+C&*^7pNb`@Y-ciapv8V#*|q1KHR0pP;JfCdry8$$Y$LpxmFLzGMD(diq=^eOh;x+u+XdW@xlm&kUPpwCz{m660LwLKIA(=T%(M}~#@ck{3g>|29 z!R5*j7`c5Vt!2NA~L_&?`nJ?1!TgmNVX6 zy+TyM|Iyr!c16lSrgk)5J@F<^Kc`7b_7sxINic&7=o6oa|2jEj8Nens|pfE@Af zOvCRNF2qK`M~T#>C**a`02rkLWI}5haXPf0DA|~RNy#1l{+AB?Dl-bjg>a}jNv$gD zaS|HQzK={xmtkI|O_2R-D;@E@0wQ%&;PGi;mT-j*-B&KSdVHoLMl%lD%1?&|oFi#y5pN<>o1JNSl~ zlElL!6E^tX;_J^FLIXNPiq2J%$m3se@d*vskkEj&J&$r9Z*`EPih{?;i{;q&^k;bO zqs(5k&L^B?(m~yU=<<***kti=dFp=-oYK9=(7v# zujrHIai6J~Fh_h(+ZoR?^a%4BJ_VQMsc zbH0ESoJwaC9!rvSJI0V7Q^xXp$SC}HckX(!w96!CbH;XI1>cp8N~E|yctM}#TO+RtCsf5~66GUX;b zumEZ?2+j-|1E2Ow(iIn#gc5ldJ{ZNs_2wQVSjNWfNXhHVOX{h&G1=t;`>1&$DY! z0zq!?Nj&JtDmrPUI#vx<6F+DZILv$+ZHnr`ol^brvF=>7$uty13!_o_`QdP?K7c3* zebb$e7eSmC%ZBTEuqjd7iR+`kUNc3PM}@PK);xiG8$IXZUU6RdodidJrX#BpiHINmNB3@%Eu z=w(s7>f}LmYv}=2-x`IT#GR$a>oT^EZYDW%T=wVn!}1!V`71pgXt(mT z|2?zgCGVizVOR03E%8v8>}CT zv1(^kpNgQaF{#?Y`zEaZvX#vpo&dfxmVoV><@DW7PdfGN0+tweo@xK}Vu?>wL85`O z(B|9VmZ6U>T=~xວ_V!pj8~45x?SHoi#Dc$b^qz&#rYHh~6Z=5hP0ap_J^$V# zp45VsQ80_R{}8^emteaVUS(TK7O*oG2I8A*Q=svsIgMG-Ak1e;v(8i3ncnG4_O3&d zzq4G>hX4H)2Fo52H4mFYA8J>CO zL=w}bi{QgLdoEZ5y6zXlo5%@5|_l%EXk z-Dv?H(O+P2u_+m8KZaOW3Vi*pJKXNb4Ymr2+qnnYQGI*9iMhj(MZpQ;|2GoXpU$Dn zzpUV|Clt}3dC4LX#NDvQfwtRpB{MSK~t^g;GpV>?1`l? z8?`-?Zf-n_lmlb&ES*Gra(^0$K!nVndV_1oQK}xD@d3#NR^p8j_9Q^@GM_ZC1Lf5S zT3cp;aB}P*ma+RdYaJF&ChMlLNSV`gkYxf>Z_~tiFWzu^{*L&4T`Zn4a}0!~x6nY> z9NwZ$os9R8BVEe&EaK}AdS%;Ner!Y{JD>lY%$g!c5*90vqs{rq(Nv$^*>(h6bA_3h zKg03l=LfKo^f1s!R)zV}Iz)0+JQjU)LAfRrJUla4!mTxMZ|eno9^Jr8-{^D3%F)C! z;52G1S%K~+9fxZ_2a*jvabSJw9euWM9A3~7jo<5Lqt(ZrVW-)fNr_@9%%!??$f2F| zP1Y8O&&h<@X*@4BdJGG?Y}m?0f6(n&m$>d+W0WhSR$cPjg>4OZhDX>d^WG6JuVoI)O)e1NWxF~$;! zb8zTJKw7pZ(WzUhyqBRo%lhL~oqE&^R=-fBFK?T`I%7BLUY|+l|McS)^U^pZj4&NR zS3osV5~`Lu(A$SUbH~S5(wn%8Et#Om#-u*^hd=+TKW~VRXPjp;b5Z*MWlFK^qk$PW zU3vv*tjdDRrF!gvSu<bM@G>#%A_jDAmmk7V?MhT@)rB z~H^gXdP)ySGq}IgK;WYN>mJAyZ0a;_v0+*^D$5!9)%363K2}ohL^8&t9>8OrPJDz zI9d21k}cF_vJqvF^{p1Yp!fL$-wBl0YrwJH{>*z@5cBD1G>Y#s z`2$T%#x9>bxh5hVA@{JxA%?rJkJHkppaZxT`o?DoOmSR|VRhs9|e?7d%)h z9fD@9VZ&xQ;>SrZL1vQ+m3)>%OQ)CMRSQemZ-E1Aox2JxxHS+vsbt{|Z)C-XHz+V%8M&6xrlQX1 zXC1o(wiEWxM0vK^@NKO(ET3lz z*H*|u(VwGKZ^H(5;eIY_$P>Ju4}W0~-5)XgF2-j))@0`w9K#loYmtFEhGDH%=zXu% zzc~DdKNA05W@^Ln{GTS^J~tYts>E<3hW7(WQ3n@$Us9|x&#-Sjy-2x-B&>8Xw>OBo zuiV7I;s+?dkcOm9O^lwtz}j9ahz@YQ%6&74kG6O&jljLP2fstq;Ca zVV(siS*U&mOZ%Y)_a1EE$KRfU{Z^V_)fZQ|wV$Q&J+D1P#o-*h5GYb3n+}tHNejR+ zAyi(*@4{(+a&T?rKJ=*o(DfhbD6mtH`ix6}{^Qs6VSpg<1pQyeGX)J@ z&OWS$UT7}DSr1HDhSF8O!U$m4=ELAUPnphK70>PRcHyQEGR2LlGqA@3SycT-ktJNI zXOY$W1r2ls77@LgKRVL`&iIJA3!!bI*1cJxI0uGPv^vlR*&IAp?iCiNKY>|)4S=;T zB;RI+;5U;4NRUoF871&>ki&PZvY?6_DLjn3>{gNk#jl}yvmsk{y#%(L9zy1*)u5#r z^?2jcY^>-s8h`A4#NTYb4_n6#=N4>t>pLI!PciIi=p?lD?MQazS^-mCa0ar1%{i~) zjJ~`rXITo_b})y_y?GL@K3;_;cgOQ<3Z>!qm{YLzR(s#x;k{c73YqH9iBBD6gCCEE z5VvR;*x5`S1@4ne$t0}(JQ+Qy*5(XF=ZJCJeEhn45gr}kOm4et(5y-)(cLeS;MWk1 ze<)e7cM|Vu-m?P!{U&$f)}{w1r_9GM>egWMiE?DG`$6_vC6!rajbn=<^I*4kG@RU# zLEoknV5hb3SpUI-PtX<~dW~IJ`jhz)U}+%XPc4s!~dBdPyO^l6ma1<abg-- zYL7qnmWl4IS_w`j5Am)_FY`xJRRQ6qOs1~p(a_4%#Da9 ze|1m7C!yD7n7x=>*X>UXGxQJKHAk;gS}6;jGLOjEU%<-;s(}O-gZI2kqSaHzP!9n=R=I7&W9+o>qmNH< zcdZdkIPeW?cn=~ULFmoK#DbJvCH@!JO|LYt&xt8mW{m;$3-rKgYM-e@)lg*nHH_rk zDJQ$I0yuS5qaCaEklN~BuvYpMYTTR7MsC+3uso5h?U7)iU3Woe#R|JT&j347<1ey3 zqZ#Y@&xS7f`{1Q~pU!bOL5_Vm3M=kt+nqWw2iwcNLJw>$h*Rhyd}_2MQk7FA9>=1{ zY=NIMUGTE4(NZ9*K27BFLWg75LNk)CJD&cnQi3nTWN_u$HB`IEo{Hakv9MVSz_a2G zb=Q)#^L6Xyz4E2mH)l!Uiq!F!u6pjNQ#3m8Ru`@Y7Sc^U+AQExK6_|$9gVZpW=nG( z&~Zvhs9cEuOA=;OYeqOJim)It9eHT2pEpxqc80Z1O-JY3qv7n$XxLM^8@+$7$rbh{ z@s@+1a;LuyWs6)^uq1;4g6FV2I2e3`EwiiG<+Iw*aae~X7uchf^@niLz{}{=>%jZ#*U^>DwdkSYYOK=r3yKX+;EwH2XurKFyrJ!NC<3gX^P@^#t4EEoN2eS`}NV}{?&^#E#T9y`(;%&)x4~k{%`cFy}l^xfnXMR4T z+oHe1r`B>-?*mX$G@A5N_am=mWkJV?rwX+{Y}an_1=aB<@fVdTc>K8Uc($uG5nAA6 z*Y#7>W~>|-Jb4Ze*T%wtE`4snbbEU9{v^D8c?g8L6p)zK06V2CCvnwP0mEli;}1%4 z*r)Cau9>4w7H7A@*)$EJqMnRWH>=a-b6%o)r{Mj!8B1NK_u;rs40rCkYharsO7`IQfH*pf>B*iQhL>0{|Y_kR55rW?G` z8BO-q|E-|E1y(CIp9!D>= zF6~>Jvup(Ja*5DSN-TxNX#;V3wlKF7qlVtzeg=a+2C}2dcly4sSpPlsntiT%-Zy=A z)cJbf&%!h%*=N^87ylMR!Jat+=mJNjJ@(1zgyzf9d zdH*GD-G^*=r9YKzy!;Th$4&>0c6)a1?RjYWV#Tx*%mKbUr*kCSnK3_FEUVj&_YU95 zH|f5i0!k*!@I8uV_+hG`D^W0a-S!8TK7;ETr@?9^0d*Q--#y=;Gc;p`-4wRQ{(oV}Lqn0k<{JeACPGjE6%-CPQD zBfim&GmCMHs=j ze+#ee?x23E>2US$Y9xm;a81NG9G53)#JvTWDm&6_Am`5UnHVO$zzYLwyN+_5Lh! z%WEVtpL@^_P5Z&?rbC*NX5oVcvGnBbv812TVEmQpW6nLEl=zua z?PrH@^nm}P?9JnPdcOGaLYrt&DoKk}D2kTXotf-HvhNCMLzFDpskE0QY0;i6MT=;8 z-I;l%La7v$P$3dUB4y9^yRZBHetdtA-+#XzkNdc{`vHikLBJs!s5xyDWWJ4Krz;JD`|8)?0uL81UZpFu?M}1M<47y+9BwBp zpGffNpU1#X^^xT3>BVUGxaTl!ixPV*+8I0h_ha(|@nAyLB0z0Uf%eWzIO<#wNFVuzz?g!qtBz+pF>pe!|t!3*J&s!6w0EhCkDem(+jb(ND_W> zqrggf8Sa1N4_fk`gLMI?LB(tSth23~g@(z$M$0;rMTPmOhWi z!m@osm>b%sK+%kSc&kK;(B!E(ESb>--n}0Ij(JJ4b&jbp=#w5udvu;%4ko}3RSrAP zRR;VNJisq3=8_QaA;jn5DNvL2iQD59;bl6(L`*&;M z#f@*Feco$)@6c(5z-AbhkSLs^_(V zw)kpTexV&6&b|%i8;9T}udag1$U4}yHWaSYS48*Y?O}^=1l;E|l88JXLCvxDFlzA~ z_R^XKg2coZ;M~0ssG4kq8#pkM0C)vIUYsGwD_RZ?Tp9)+n@F+(l{PRjqzbA!t72P; z0(S1+SZo(kf%V391CNoB;O|9!w!ASE^j0au?<2Lq1bhll_^e8vthE8D@2;`8PCn!9 z^8c|xepS-e-5YBw=YkI*8A9((X^cc#GtQkJj6Q|=!R#Yb@u6Zlkp64}X1TT2$)01; zTN^F3*LDo@T2uz&H6)?>=UvRW9!;wmIwIjb&DZSD9bcH}m(k4WjNv$Uk2`bMD}WWu ztrxyK9EFqT`eEz6&Un7mAq+ZWarhGhViK>xV9HdXdbPlKL+1Qm>Jy@RY#7MHCwz7%4brtj)~J)&uuu{s0>z z!f@TidCZNUjTXy*!0LX>T5NDglH9t@;P9m%@j)*J*Kg2>KPz#<6A(R z!$7uZCSw{I!kCE1 z{H=OJ68Je?T*qHlL$(QAsBr@uOlIN5D zC1V$GRc-^+);$Y4KMi~G{W!4I$l7wao%%xR%R&`CBbzy!Bnk|ZTfZ}1z*<@nSV zBX;@hlla5KW?-GY6@-2|$!?T#V~>3t0zV8*VRxTsz~85ZlGz?;HOG$^JicN9 zE#iXN)A>?B$&ABh`t2py()fpMdN-5pR&He0B<^Ki3dS;rQrhs04L@+x!T-+7aXjO{ zzy3^BAZ{m)GYx`opyc3c5~cDNyDk@io$J+^sy2H^@>vMD)O6r+9=6P@qmdx$ zVFZ9#k8#T0D}tO2MS?4er?KbQ6nrOI6Skj^$L0^00!`@*uv+c_@bpPx{$$)^LIqJo ztBJ$%Xxk1NiVBPKZ}k92jrnj>XaQ(441}@L6XCl_Z6L8_6l_x)4dvCYfVUw5V8*9p zW~j9U9=6|wnUdVW9_x^0qc1YxzN-o=`V@fwB#eSTbkz}RDuu6KohjORo?4kTMB=NP z0&#KCO`P$O;Nnt-_$M8~&nu_GuhAlWhFhbo*DuG7+L_EU%`~vDT7e8dS0eE5nZ-_% z9)tfxO#q7F>!BexPE1&JQ=lG|#Dp!#13u1Ep_QgK?3Nsa*YzyGGmQ1%;y>xI<3b#; z3p@t;=llg_DT|@%>L{Y2L2&iCQnovKCWklF&Au)Qz@vR#$llu5_{jdD%n9X0wxHxB zD-p92xcnZ)n%Yidsy=$+H!7Ez$Z-bXerpyBdN_Z{8&b^DIYB^l1>*5RHt<(!1gol_ z!foEA;V7G#&{i&sY4%qj>e~v0!_Gtt`pnJ>T!*ExfyoPrk%kH9yVJ~AU8}o%}~^CS3$ zN9=TB27k8#Wa|}N-y{!))~~_O{#IcR-(bPKm05y3wSHWimPTY!oS@0wL2!9?ID75t zaxm|niXdfd4*amSh5fd#4~GoB2)>_oBHQRMt4H?lVdm}2u;ZBsgpxZj_SZVNXQDPt zr+xT5=X+z7TMr~kL!k5dCCIwt8SGh9!nFJB0FRgEgRxoj@%~xcu&Lz%{Oy(>9`xP? zWSk$zy#7*-XRVZjDHh{cEAISu#pLg7&qcM%B#3^R1O((o{)EUw#EG zXQJT!Bnjf@@{ym*c6yH?3P0b2?@du)AmAYLd+kQnvgH*P`bvQv^d7EMamOt=*BS6- z58#e6F?|-p$mpwA$?qe-hOWc%{5MRse26Ii3mF1fA|M?p&A&_v|acAkrf8WIt zf1bckmcw~065&)?X0%5c`(djY7&&Sx=&H+jgEHd;;7q`Q2&&jRmX5UJ*(l!b?cE$YrCi}Ar zqA@Z*;FXo1z_-jGMqHlwIy{?&8#G?R6y428MN$gRKN`x#?$IE3C;C8_bymdf&r?#H zw+9vEb;Gja@hH5s7g)qP5a;_N{_#y*j@s&G>JeW_gWhD(5_E^%@b*2t8UGd{Lw|PS zvjXrYvm1MHc#F1crhvpHqmk~8lZe$=i8>7WP}>A=QSf(5QRE628h*&oI_ZnN^^W}s z$c6>PZjv9ej4nW0!;@&F$1yrTBLhl2UIEsQG>3+c&*2=!H4t3845$8@f|hC;h?Zgr z(L2R;RCd`Mm^JG+JmYW*p3HQGslS@wh>aRZQ>_suH4`}h(0ZupFC)78VKz;%E+m6P z_rmGI4RC9FEL*= zL@uAoMH+_;=;)b`iJP%Cs=xA9R5qziwD?hk=;swj*b(*|t~*)_y$4?hpE*CbV2KcP zEykbjyrK?+_q%bv!Ds1}-e79fcbZ0cJ^(TYmO;j(0$s87gf@>9fyL1ZxbUt&nmRlH zJ-8T(&j0bF)uI{L*7qDq)p`v*Tsz^$i`w*YTr-_CEsVCke?pGUamHqiNx-c*5sn!y z3oi^k0OL07gbi~vY3pid>R@V28;;xuj^{Pt$vK@+!(g~*+bT>gmh=+e;X{z}{we5T z>swg!NJVtoD1e$=$fhx?u0qu#@~B;R0_ksdMZNW*Xw7$B>#o<)jxI=6$o%fAG5WVzVg zDRn4vu@yS~sRng*5TE3Q{FP({`c9AA+p5f2DMZy2- zkZU?g=Ge55xwo^?4x18G^?V86|MzWs#n!16fM`Q=_Fhdgh)dkXHhoXP4ugjlAKn## zJ-&n!sqJ&nowRePsy9xs&oPEs^*)+SZMuVYjv$9)~BPUt}cA@)@)q-&Dr|Q*?HDe7Ok?rz{P<1dHZTn zL-9*tIQsyN)G#N@EEnLo`T@oUek;^m_f&X!vlH4AHVSR2o&z5^_P~~-Hf(L@38rdq z0^i4VFYm;TMuueEMt2x7QWBlHIt3m09gUiGqmh3$3reE~(c+VV)J*9t>fBw9T>B%D zM)ygQQP)dW9+V~XBjW`(-=~Q8S0wDG$n$*B-nk2gCD_0_dG2IG0v9j%4Mw$|($??S z#G})3lF)Zd9FaUKOLtiqSsyf;Vr?}14Vw6|8%P|oCnsM>&=)x8|4zJ3_`b|u-^-$fj*54+VAcZtd`m37Ik{!#C_PkK!(iH;d^xEQ1&zbc6 z#Set6zQo66>3#j8c+Z0=#Q$3ko^tFOA-{ZpR{U0`YDg^9R#0Um`i}@6J@96RZgoS` zcjtloOIRZNc@^yU%)qfHhNCx;c_8{-G>J-dz;f{m;F-obg6dV9$z{ExFlDhnFm%ph z)0W(X_wMzH?YM+>{av;ybQu;uN7|^;(9Y!~y+vEN`%M=7aHxeWIiy0LI6uKosSI(r ze*s)ysS98HjDYu5l;NH}3HDcgH~*db*U1@KkDLzA_^g8OI%hC`b$V?0$|uaRwsfGn z>OOE7ag4;Oe<7z1w-Wc8(b!UQJ(5m1hNdMRK*KXMuuo6}Y0>1?*Eu@QtFVlpq=?a* z>PCE=FED#nU4uuyasDMj6;l1V5{w&P46`PW`=8DBtFW8SxIem6bZ>(NkuCBgP776N zzLStlWURnK%AafZ+pV(}ul(kSXMPWa?#yf$zVtQI7v4a%+C3>(r(bVl8adOO2X^+F zLVfy*=ZBb<+=U~d-_0GwYE%UAFE3~8`$w@iQgYZQNDf%No&((^N7I9&WT~NO5S8^m zL(XkZ<#6O$aMle)sD8zO01gQxkAs&Hm*ut(WVn0(J0LhQ2HpSJN>b%| z$fwiNA|>tf!nU`n@N?-2;2=ARu0Zju;PFN^V+&;*C4{Ky1+v*fHxlIZ?urVO{e` z%P(tsVNU|J-+qRwB(~Dh)B?&ZA0_Gw6Vp%gDP#U^N1*X<9~}Zm$y8?PvA2RzdN7QJc$~)*%L+)$h`Xes!wFbLb3PDUU7~_@9R6WHu$b-&qvD67>V}^T$uS{Z z)KPX*Z5*)Nr%b+zNbB9jSZ zvF9T!N-P6~%G%T=Xg*zSR0n?nO?q#75S3ngmbR98 z%;N&t>@pHDR*$awxQyx@KS`g4hEZLM7T|!2Q9tZMS!UOe=E0}%-UoAdBk&S)xM2v1 zed~|21{BC*Zhly@qMa0gvGg}hFDBFNapIvgSU;`?=2S$%g1?Di>-*|V5YIUR7$d4D+2 z*aM2qcF@_)<#1`G5IUWlkG`12LCJ7+6wwG!h=&@kTEC6h)Z8QYj3q#eg)uo^_W%Rm z68eixA~mrC-2T^Ts+-qDwqH>r*VYyzIO`0!zViY%>{ zcLVDimj;mjpj0&bZy8$u=*>Q0y7`FW+ zZ^tsIHc8-Oaq6^AhEiXmkmQgEWFS1n)}38UBNWtmxrpnwIiSLiLkTyI6NkT_`Mgaw zLx_!c8Z%$`m#J<$3+{MIFh4sA2nnA^8}DQjg{A&*O4n6r))>XxpOfpAV9;MlD*yeW zaLlmLP@|gjujI-evvMVwzy|ScnJybi+=KqZ2zUL~M`A|z^E8*-$i|j2Pl=9EKfCjf z896fezwf2{3zJ}-PYIkYlxNZstH^-aV?I4z6^MF<2BWLLRQWp8yGr?a$={r2B883k zGzx>4!+60X_{H@zlpApxrp*rm24`=R-lP}A(=e9Ym&hj<{TqmtS}j;tI#S$6{jwem z;maG4?IpIcaeO%!4^$G1$X}$b;|rs5afP5hVFeVoYsRgkPf7nks9lyU_`v2PR&+%=f3H*A5aP!iR=fzP?-@U(R}D1$i*J`3he9|>L> z`tjpJeVaAf)%y?zZkdd7ruvgJIalG113GZjVBSWLEfe>bYtW+L~^K;$((ce`*=*8^}s)En2v18ftL=N;X+; zCwEqNK(%xm*iujUK5ahB7*?lq<8aeqbg6w37lZ5~{82WD$h3?>7r1zV^-(@d)sBrI zbF2ieKckI}3lveutS~EWDM&XSoy1(de2WQvYC;E8xv^bZiVWTDjpq*t&?~KIo9`8K|P#3R`g^z+38o zVM;IvcIqWM8d|`@t_$2Ti6KpaFVMRO8<26IEDW11XZ=Yl<|`Y5!g1wIS)MXR(M zk@n;a^sY&cc;?1{Vb@#m`i+}drQS>E#diQN=0~z~Lgmq^_CBHL#{%f}@DkZxdJdw5 zJ$U+wDs*$mGW5xP6R!)M1{?VM-(BjcyRsB+cqoO8G{T@|@GDrobUM;$nTTX(YNLWD z`smd9`EZi+H7IQT31#+tg$X88(DNsSV9cH=Agy{A>_o5NrD|o=5t)g;m;{1y-vwx{ z+h|l8tcbdPOhfYCmtcK)0#b>xLo_>yw=L5Yy*Qj z_4B{yvyT^B`E*xV`R|Vvk6C`=-^JG^MwiU=yJPiy)DE8K|9daKA6l#l=IxB)=VbDe zIo$t|IdGif5|C>_aCzt!Mm>Cv&~yBLtS}GYOwn<4@zs4~*!msWA6*C^p1g$@j_ZXp zzDdzf2{U*)>OzCi(WRr%ytA*-=rfb~u~S^n*MuiX|Af4C`3)`WV;|J9Ui=x}J}XRl z$KSHSqlie4TZgvL9xL74{vI3li-w>jc8+io>kwRkXUzCpZA;#&s;c7+|}d2ri$|*@`WrG*Y&b84ss~NXjl10WN;{-WMcvk z#=b|=z5d7-2f*B51@eCBAo@B}?w>41)Jn5!qJ!Wf&7<(CmpwYAaRxOQ<)W`Q4bhmJ zk?6M3VC$z*FL1*Y3z$7l4%vn-hqgBtF#}hIz}_|L*V_#j3Z7p1Vfos)4(4!cPhxt+ z*RBs+d7d^aDzP;e)0x{!lR>QL9Ue6#n6G<;z8n)>wGwX2*~j?4nMQNQ7DA1UKft>6 zO8EDvN5F~O$3+!-;F78g8I-AsR|?{YbL=?2y$7yNWB0F+hpp{zE%(|=^Y6rUE^K;@ zJF7#_6H)r68ZJscCa}?>*G9=F9Aj8djODx+zDL<7|PI04$)Pl6{jTTtkwPV})i z61Ci(#;se`Lvdc6<>%mh&zb04e-dmBy~^I2{D|q)(%^p+&G8j9%WeRxN5vAe>P6&n z#%8`gaX!}D30J*Prb|o)z|Bwwnpc~nzQ+}OzPl$)`T3bRU6wzB={d{z_Z6Q17cc90 zyeiyw^dqyl#+lTP_zLc&o+b+{ju4;4W6|MMNnRf|JQ+mOHMPJIsKl;%xCw4;NQcf9 zws7^)493`hKii#_$Q~*8rJ+JcZf!IMiZ1F=V7m*PeDwsHT2OZXIys)++1&Z1x9}U^ z4>Qg5h?p+%HO#-CgP@!}gb~!no7WecV)H&&@yTH{8VThsU6YSD%tY0+tA| zD_yqWG@Ez2m@o51;t4p;smtLmk1|DMNW)3|guJ=b8^i3%mp2^i`)oT1qfu;aN-F!V;*(;X$H`X|XiD z^C~3|YN*Tc9FiLHlFV}Gronq4J;@zwuJu(F6&e`BiKF+Epkww{7ms|Wi-fYGUw)bN zf9*@DW}r&{FydB~Oc%`9z`uW=J`CJECyno17)~Uew4uLd1sqwZ0yFZb(6ZyJm@TX+ zxD-;xuF;`Xf=j;uCPp*q+9W*-v#el&H?H$K<8-0cN`BH8f6Vkml>#WT*UCx}xp^ov^SH z6z&nwtWHSZdEKY_W@Xf{;|iTqdx5+%&j)^9n^4)79^zAzN-SPXAv-?ogzygwoopV$ zt>a5T^4(Ncv)q%t6yie7-BYO($fuWr>qzUd`E)~T5#~-Tu&EcFvHJ-xnDl7`a{m(o z(z{21x7rA0%f`Wdcl~)ej=a4G+TW;#3BOLU@~_H>+-(W^Q)M`n(LWDv`TOAgDuqPa z$`a}KOtY@L(SfJj>S6a(1_GZgFYtION0&-1q94Og!wR2$@O-ffQHaaJ`L%M;zq8uv z=ez;dAcf#P-ttsuf)VkpIs;#sq+o-~*U9CMDDds>1UPVui+nkHJmw%+u~_b32lbML zWn!3bLukFz26(u+QX78`@8{ZjIHL3f+Sz9XukP*wuG@=2dBz<4W7x(v-YZy6byn^UF zoj`XFt)ttQ3ay+nUc;hs0N$%8hMPX+fR(Fv;UmT+^iA|#s+seYK3r2sHjbT1Tdx9M zXSTn-j;)vJQfse6SoQq^h$br`Wo{E_>s!S#1#5Dcz?cSm*MM*lCweeOj1Z0AY((4q-!+!~8Bb~%6%vlh_x z2B!4;BPo#-TLO3QkwMd?o$0PCo5(<*0-bf4J9!%wOP9~MLBEdlB5`yD&(oMdZvGTl z%0Ksx3xyf)a&i0)MfP}g3G?d15gK<#iKbg4vV$8uu0U7S zR?=-gi}o&ANLMXWqORY5^7{7U&>W(7dIt%Y7ZI)O!&q&70d6%;CuS<`s1A{eUoHN*3Ey;+v-DBEV#2( z4f!0Hhveuoq;{x{ew&>Sudi~a4_h_qo#`iN%y&iVxg?Ly`n3y=d9n)`ef|Xf{Dh4RgK~LwP z^*wWtSZ7Bi?t?uOr5M}w9R7Uq2Mn?-fkMkTa@$;AbV+0&`t7ha)#9sB@;_@ zd&Mqv;Pz6a^-C2cNL|KLYnCD1gE45nK_r{dF~H7SaRpoqvn5B&YH%Sb0)oY(s6@Cq zuWMs+`p7J$R&u*2g6PyvLRXh~qaBS~P`>>~m}$;^Pb~kzedAm}_c$n+O|KTT2``p1W@Y^_^m1s-!bCi zk6=$!0#y4wjKtaIz*lLKDDh7jT&<^tPbI7(+wTTq_Y!#&Bg^8bzlZ*fOJbfpv5G8K zoids+TEfK*IUb1Gc5XwvT$E6cOepeu+lDp-oj^XiQdryg0Jdx@B%{=_$w|{(&>Qst z=C5qx=TKW>()c;S^$SICVCe)BxV}b!TuyUwE(~$?)sdtqOO@BD&k+NtcC4cH8gmwv ze;CH=xtNC8263op)eKZ_-45qTaO3!AU3f!pFgUpF5iDOo;GW(XG`vm*o*j0}D(|F* zaA(tgiayPz_tosl=EcLHX~mr+o>YlQshVFmg&%O3Id*|;(9JP1u_oG z`EcE64(n*f1#L?&>N3Z{8Jsj@`3wILh7d1s?yp0JfX!AUH}B4YA7{bEZNf9N_5uEBJbHqwvAiko;=|spfSJs_6|JHWD1oeh*X!541N=aWilp{B)cVS z<-@czL`p?3^=)_WyJv+?PL|He3qb^m+&H_bVZzpMP;!>I^jOcPUzOy$7#( zxrba1uSDBwHDSflW#Bsy;l(ek;khFWEKF^My0=*{rT&P(UGoCWiC2cNrfb1Zwuxf; z{lqf4wVLOp_}Q=6bSQeVgIkNf^PDuFS^*>MuCgC8UEp!|VjN&RgsiK8SgQ0oP`(ia zebtY$Q^&+X>%t5;aC#L?`=tjBf<6e2nM0m`ub&(QlfFd>@<$}Xd;R)UQ@57s|0)Bs znlgyEuF4q;$%6OoKxRc9v{?THtsCrtqo>3(diG;UK^%ty$kDc{{2Nr9^qlDkhe$u$ zNvx}4Iwi#SJdM}He?KN+`H|xCBo+)WWm{|8@T7_DFz~t&syFO{Dh_;I;`9H<(qxY1BY1nxG1Q`Q0S1)0K%e&-e4A#QDI@x(70LA=9A#1kFTZl; z+aa!BC+sb7Uo;-BeKs5wBs(HgO-Wi?ZVVPQmarDx1}Jxj2oA~EjvV1Qd?hfHow+dv zgskyH4oa)gv%MMc#EX7#d^v$tg_Y1V+mU=AC43*4RCx=#{Wc&n!-fKNP!AV0e#B26 z%b@WsGmuqd6dHUo9PF%LNv^Bk!=1|Z@Qd|o+TqkCa8*`ByX>yu27w!?GuQ+5+BEq& zxYME-v3>FrUnl?k1qy4z=|OLGx<)&Mbat(T3Wv>j`5Io$2AiBBVZ4_aJTf)}Z#nvn zsX1DM^NZ)gpS{v_P{Vb))N3^{cYJI$p=%PJFx!h>{mtRSsprCveJ)f&B8eXUeVHzH z)`7mu!cov85qfi=h)VS3(rpDwbdi@P3XIc*KfE6UJ(LG^7uCasx77YGKNY5c?-j3! z*0!n0zB>hJHmO;@7`m5v)pv#Ya5{@>9ZsenmTaRIPcX=92t?=E1!zDmS8#Cf31;%^ zE-*jgD6co?WBZtS`zIs&m-oO(`%E(6^Ao%}J_MTV%pm%BKD^g`nOWeynjBxzK#p8% zCU$v8$jAJh;{GF%w6cKN<9iQBwoNCMAx(UFV!o~MI|I!shN9u^Z@|8WOtyUs=ME$) zhnqL}qZvl0xIM325dTgO#jsg$-x6)qcli=Y)E`ToBt)<n0jvIDB`T!WmJxWUhF zlIXjPx%6PXB^|T#9O@`AK}%10!$&>ip@6*vXAk>ErgeWKv%@n;bHhC_=7o}=^&Vwz zOxi7I?0C*NsQJT{yCl$=_p9;kORE{t+lTP<%5LF@;{rHTlkclySxamV=;h@veimO3 z7JtO4UL%NnkQRtBGX&FHjiB$M4mh?vksYe!4v*B`g{iN&bC3rk@Yj)5P-o&`5I2O2 zS>-Ox7d4aMF-0UVvV*T@d-ob_GbIeJk^jHr zSs@XNLL!c6B9M#nCU z5$$v=PI%=Hv&y`HnsO!3xu*#|vf|*p3on5qvmV|b7=kjL%Y?gG5jf24r~KX045D=| zvNK0a6VvIZHvTV|sjtJ2t^aGwx?=~}EjKfO&FctQE~CP@yd5vdjMBz4`;Bn5Z#~cZ zzr<5()}9BBf#omFJRCQTrc|lDqBbazNIy_-|^Q!^*@PqXerz@QrCbt~#+@usdXJv4E(7A5(&H=5I~5I`I=LwNk)T&r(9% zX9Q2PmeDPtui_&9`H7xC{`znm-$rqJ)&x3%tiTjfeYut7>^1)P-fi{<_O1DCSZ2N* zjqtKXTo^01!@>v-cPItP?e^5P_y^3WG(e5_rX%m(bhM4x`A>G$8 zZiN~XFF78~S?Ej7is~(vdc_Kq#!Rvp-;^rwyfpw#dQ4CR-4FXOQl3^ZeNFE|_q56EgQ45V&!0u)hj26gwD~sdeL9KSE#FG_T{fWg&lIRXhYL5}Asuq= z#B_2hL&tGvadkIODy~0K%``3!0zc%`$ljOdh-7XcNxtjLVRi+=JINRLevy%;Lmf0T zh?LcU@P@k|8t<_b{YxD$jFY#rwW;7}I$x!d0k+dfB7HBCB1o%b;(a*Y1R-Y>)Q@?7E zAk#PGMpr*>sMKYaI;Y_0y$`TAPNE3$o7(H^mWj7x}@8i%9sBnZ097-SBxd; zuLnrL-dQ-vaXhJ!_@S+Eoh^Kcd-}lDHk0kHzNaM;^hj7{a zrAR_OlyB3)^00sXcKe`AGI@d?wNr3rS0!=22x>tneDVgAvuQf@Ax1Q5rx{tl$e36d zh48Z6E}%HiDIZ%cJISmb{hxgi%dIXs94&O+jXvk@f_*iC|7^-*#ayT|S6(=?T^fAq z&VsGBh)sVH4nCA$hRSVnXt?ufC@Q7c`Z*U{hbzbN;{i?&a!Jw8opdp*AVY#&$>$9n z;IPd&)Ho}Pr~jL)I*ia;g?=oahPFSAAs;!M5OH3ea8Vhbj(J-6=E;u(WErseo+umk92U>Ln%{yCC~#9N$K+EWcW*a-N6i zZxyVW>HG>8H}0I2=_NCqD>8>sDb|<(qa;Zv)_tUOvn6JV*k4(xqA3b_pYw>~`aAcwhWTtv?a-;*<*a}m;u zK=pGTK?SGH^wo$oswG=P)f!Zh|ISa)sguL5_<9p9GRZ(HT>KaEeJM1}mln6tui*N5 z+I+N?Zb??5k8=WG|Eq=c*UUA1pAz#%d@YP%;peey(T!|xvHbnS@4Ot;`R{8Vg&Qz{ z!@B(`yBd!$nvIt(8%&{gyRa`q7mQeb8h`(j!Smz)l*Q-$KhJ#G9qC6%@7hzvo#!vJ za&f-6jYcpt`c~7Mrk8mdbo2JWsJr`l{yyuGfa>*8yv+Vr#=2;Oe>(GBVJB;}Ee(9= zyUBPjSP1W3NPtcXQ8?A(GKjPYh4FRQp{kBNz>|i+Twe`Xb#Wf+I;eoP`s&D+`B?ro z4iXs9?934`W!Qgy8RB}jhRfk?R|?4L6GnVn23)@5qG~PR`MHJp9~~17>_}$vb+Ye4 zF5aC|3yW7i|HFm1KlzdAxTpX%Iyh{3R)Pli*AlyU zJ?!D`t#n$*-TFF}I$J`7dF5zy=;V6U~=D!}}A{yRiU# zaC;2;4L$(_r%CX^BL}?n=48(sC*ZPFI$0M+kYfsBR%Nb*)Q0-I$b7m>vFR8(p5pKQw`BwsZY}i5^(>W z4Ww|vfBwWqd#983%etf|B$vJJ77J=Wj3T|F+3>5}ThQtFljl)(>1enp!5GLbJjdKV zJ{_#oXIS}(`siH6a>pXZ{WVkwQtl({~TPFMoxO(i<)sv$c|T6}qCmJEc*R z{Ty0ca-I6F>7>15HDJErJs^r7j3L%fZb@6;A`?(|#b92} zMUu**8BKe6`K*{e9wf=c3%bjdEysKcpf{?LNtr;C9Ir~DXQZz4Y5u2+;ydG|jk>}F zaNmnsz8=|iCaBM16?!^%JhVJ+376_!hVJ`5fXK!ws6K;>yT-Nmyy1Uy*dJNpjLe1> zK3`o)4%Ilah$`#!lek|}=(_z$?%Y>2QP`kLXT3a!cg{9Of5942v)2+)c4~j&4v2SG80)jZ=k2|uOeL^YROHrL=f25Lo)BILNDK$ffLJ~ z($kuma9idO^nTDaY0aO3~Yt^fHUuNpN5C(lk|>I0*u-mgpq1-ox> zxM|1!*M2{zeJ-TzQZjJ8wig6P{a_nxFVjQi$<$CriYlpdF@NH9=*sv-)Ia1a{%~p- z4Lsvbz8TA*BeC<*@lAH1l1LK{ z#O!)7AiMJa~R>_?t zt0OE)iJK&lkjO`ujZzXs$NKI01T)loyfhyPng z+VXH3ySlguoM@X%)!Kd#OD`X&9~wf;W~h>7iUV+?4!6Gg(hfba7>weBO3;XS2Xx@* zI5c~w0#pv11Mm2|!L&LJ8vFb;vHTM(Zl~V_D@oLU+W>;BMJT`~3s$>i!DB1{>3)if|!xJwF%JTZ4Z>|Nj{_V;HUbBM|os&_z(RGv+mV`>}3V0hJHQx`LsQrhf z(>UzD)xd{%sNJy<-u^Nj8bkx^RWy{mlzEO%bk~rd*?pvZ<8s=2HjwNod`NB%ar-B? z%MokIzSJ>T&9DQ!n3n?Q>HHu%m5XS|l0I_g_Bfgt5Jc`?n@Sr8jS=VZvs}aA_)*t6 z+}?RaE$tLpb3P55ZZ{zf+3}>UVjklZqXHvKSon0waH4Kv4L{+9@ciU&q`t?V%>I-D z$jErIP&a@M8sY%=1=*9!>QQ)MWd=##QHP2{bVN@$PRtFV zswLVKJnkp2&P79i+b**1ogx{dVGOn(OTuYKGs)J=(`nH*B|6gu(a}Fw5S!js658uQ zTjy=V+aiPAUBk&4c+QsOe>|MOM;cx@{G5zV?;~yPuXsL*^M1^IhNIR! zBw&>UQP4gFhwPb6m4cnu5pSO!{>0xaD`c33u(`eLl;TQaB zyBubv8N;TsZc^8B(qdWQ1RSF4i{JY`C#QM?$?+dcVcwoX7&qxWn)IRsZR;OP4QrC& z$C-^VW$iWET=|DOIK@%tHWyfxoz2V8?$SBj+7pA476ie4&!(}um2NQl>uDTqc9yKO zmV*r=WI^42f3i_?1($z`n^X3LK(V7cWybF_brXnf>+kzw5w(N?=6ny`G5=)!nK(fjj;^x8okG-vKP zBze_H6rQt8^zzOy(eKHj=$4cnT3b4bR7ahrf6b*RblXMkJBJ{j7fJ9&_&!`~vJm;5 zvq8_Mykcf^KAPnto)^#YsVm+;QVu0{J%dNh;yo+-C*44KE=JYiatfawUn$I89bXHj$N_Z+@LnmfCMrg%$ai z;h}vA*p@s8>ZhD=JuH1pUeGN{Cps*evReG=vzhVydx?sXj9;ABRkXMJZq7VZYwX?gz- zQ(qoe)ARqI676X(r4->UZz)ef#|WIFEDBxpQV-ujlJEbLPyv7BKIiK3eXmL%YW}({cJQ=pjprg8wW* zlfHyNrDfN#|2<`R!p8`T*Tr^!9{{qt?Z8=h5a&%V7rbXK3mP~6eb0+^{d=JXY}Iim zF0~TwIk*LVs>)*=VF2EpoxxgN+79uKhah;H z7iq{m245VlgrnZyg2ud)&`EJV3nh$U%9nUh6lY6Jt+H@%o)38Fb%oRWWp-OgMCC+c z*t8n0OZ6r{JVZHZjFd>y;}HJfGjsX1EA{z0j^6z3-?s7(@u%_UHbf$;^`rO-$6up6 z<-gI@P(}WMSyT9o?q=ATP2q$kEYz~t!tYx=j4$gcft>BEK=;WL_`~wk7k?S{x8l^lv`Lzf< z4?&lFRMF$Sq5K!uiV?qOD*EU-0f_MqeqxWO&RB@>`XKZ%#gZGB&VJ|xd2&MtxR;Cj zylMsb<8F4~F9MrbR5jC#}K&ZwP9bX_?)ADga z&)6cg&DI5td%Fg148H~^Z9TvQ42a$BRDuD&M>J z0UFY%#82x^=IHPm7>-ng1u*4eA^O-8$KBh#VgsB^1+c^A7Q4Pr@NZ6v!{ksW+YZ}D zy=H%h9%b)`200i@Ugy=p7d+z;fQ{8Zj0_!u(Z&Z;;n|&{IpJ%6)c$BG61l{3;{p5n zH__S<1xM;Vf|Kt{!olsLaaM$n_~`74q!U+67GH;puQ$ZaVcX)KGTWaGf-~Mtfl|)n zz;v}lG-#(Rzx-1pN2j=)hR8qnc>)UsI0BqdRAn04e&A!-*YQye4I(>SfHtSw!k1@i zVeg|1sEH#X4=KZgBO-}IIE5cCTO#Qby7-ptcd%*3Dd;wEo;au@5b@jz-9&fL;+e7N z-nUsOT5}b!la+wW_U^$dTZE!m*E-m7ZadUhse`HlDa0nMgJyI4$?l}(M1s0=G9R}Q za<=dH1D2!0ovO%qOb8mIJQ~cn@5C%vB84@+3?=P$6c5cZ!}(U%!O@e^;P%B)+<4kv zwo9}l^w&?p{YNouwD*KT9=RaviUo();V~i`<4K3n4qS-9-0_02*z6jUcZrw{8pNN; z`;6|b>PKzoGEi-|J96I=izk0oq;p(b#eFI~VpakAej*F$Z@BXhLotpU&lRK7iD}4T zj5eAa^^**`m&EbmvnvgMw-VX^-(Nt-*D8aJuDdxJZff$-^qacqm$W?^oZU>&5P!OE zsV;SVR74gfTEpqBq0mq0#9L^gC4N7LXHK_<3T>&l)BE52Swh_I+3%CF2Z+Wi`&7Xk zQ#Wwy^d}hmNgE9t?+K%ME8&_C!s`=HrS zEEkT(_TNc3>_jzhrB#*P3Inssw~t>5QX}Qi_YHykdpSc~Q7vI*=0}n!xbayyfle-&rt16R6^`l{x+h}0ETAf6{&x2N-wQPaD)rMls!;5O(vx6je z;gDPR1?fuhNMU9pT>APciN7@k$lLstQ_`hiaLbke)emX#(19By;lVw@@(D5I)O$tR zyL2$!JM95H;%$ghOtjHG3%wM zSar|8xuqEQnI;UnJj_EBL(T%_D|7opoZN(CTG-v+YcXg{!>*lgiTB)RSm9|T_Q*AY z=7M-K(P9wE*ci$6ZMUQoHTFgt{N4UMe%F@K{EMa+(V?w%=$=6~I=gNMGoC*O{;4}v z-IA%#|90ReTA4l*C0|Viqwd~kLss>(j`c~%b8;2R)p&rcI?l4$HEB3zzb45v)}j{1 zD@kBUH0X)hU-RSVQo+ocH+C}nPog$KIttIc2_15ZiMP>Wod2wzc~TpRmy!TBFk(I4 zXDmlbyEj6)aqD5PlN{}Qdz(2qxPs*L_LC^941DKJEU!{<3!&E(J=85lecGj@r@$Td znup_)-;d#y&vgVnHr>oQrzpJ0a})RwQCyR7gNKJl%0Y`Ko{YBIT5xa&g1txgK$kF) z?R#w-^R6zQ1b2*LipSo9m2Y;UUGGB38S5ilzf(DqNXl-?kd|mW@~gL*r+IH2R8=~K zf3KUuA10~9zjRlRKk}m!c_-(Itn>rm$1G3mCcPeh?oTIVQ95v+2;tP%3fwp)mRs2H zk@gFY?f=)$%I3;&?L2Z!;1J7zcC49_Rk+s9Q=7%yTAo;V$5H<5hM*>zlxJ z4;GAD)&jHwj6lh!Q9zU+SxiIt?g3-Ay}}s(q5)*Y<1ltPMzPzdVsPx- zPVo3qqIe7pKf2{5yVALosgqCwZ+_0ePvDvH|x0nw%xTH zk^N+dIT-7H&wd#%11GPV%#Ei@?k`|>zZ9_Iw%!YFGF@MW;n>c@O>5I9j^p zYN5tsDGqzbbyq--^K;;Jz?~5mI5Gy>f3Ylx*5}64`}cgAjJJWTRk4!bcRtJ6$CXPM z3FK$U^ZIKBleWu_Tw9~G)S*?HEN0ft6ud5+!nU9IyBAuKdxZO6OwWdKp?K|&AW*aQ zk)X3i6nAafUmv!6Pt{3qt~A`Gq|DJLE-OA>NqTTP@Za@?8;e2zK4mtvQ=UBUkj24f zsZ2(BHS^-ZFOK&SwqucRR2BD3^wkLVk+n3db8|l~bG}it?(|%yq~aJ)yWNLWS#BXr zxiMY!xdc0{J_?m?{#`dYD*JccSEsd|q&YfJ?dgBWpIKw^8JoZQ^sQEz_RZ+R6|<6n z=1v{1%^x$K+3V%3<*+K7WCUCmNK(ZvGw^#y3~iku`;Yh34U5=;@o%{6MfYF)yWe>+ zg^cv@q$y6%h;s8RU~pKQbv)~bgWSr9>!myr&|FM}y7$P;JvZ_Fo6E?Y_Il!@wwR79 zHKy|8?vb5GTj+*mg#L%ovZ?cU3e|bg-D48T2-`>c6BDsk`XRihcLGtgFTpL@2k~At zQEnPVWnSJ*p5S-b;z|!ERjOs31XsyNLsbiTs=P-M2G)4vdrps->D6QLn)E`BUbh_o ze{_r6w9acJPAA7%Z8<=_Dm)=rW zqaCZ(G6nsaw7l+9&1s_u(ib*_c+Arx7vvB_bco z2E6i4jOe=?qHocg(NU95(&qJ83_Qo_EW5NQnqE1uhs0mfrzcCesi5HYITKEQg7hEz%F9#cb=@=R6y*;dQwA2RXTsoO47E+503lv01g`w1e!&A z`f9Q9u+1tO>sjt44at(|^vn(D+)xdq+&BR(>eEKIZzq6Svk_QTS{|oEvmY!a{g?K$DFx~^ zE3S$BI=5qpKLaSP|F()Kws?3sNvM&g+HrH?RC#4MZH_rq8>vaIj9$*TJSgVs@cW*N z6N~s{&Z!HydE<6CZKV;e+&G1FkBtGFWye5US5f?d^OwjN-fR4zF&LhYD-_%vJ|9Yd z*T>HntR$}kh4}U6L-6+Oi!jF}3BNO!rN@KfX>jsLI!0m;r-yYJ=_Ek?E$mFFgzr4E z@FDdCsCg-#Id{AZ$Zs+KHx`&Ec@j6PAJBG*EPCoz4w~Y|!!zS7NbSKN%)l%q%r{hG zR9;&$l>H{k)i4eI4k{tMwe?{1hnqx5Rp_P#^0f108fkg)TjXQ(jtr#ul0ho^uqAp3 z9F{Xj%x`#Y!A2BTPy!cuS)o1!b)@j+FkIsqOcmg4dZ1OAqqk9TgY9nr!%G@E7|C8v zfu>d5`&8oY)wM)THvuoux(yXt4uIXpAz1R#Sb9{*r^ifY(3|Uj!Q2Cdu(P8cDk&8~ z1)XQmKw}le1GA{5nKi9i(?u2sEoFm(+Sse1XULRpJNRqTBqE_Miap-C5!p{ode|Kj>!dG-yk(MpAH&TCn|dFqS)&Cz$%pm^!wL#zS8oGtz$o_3Vk_bgJ&VAy=Yu&;Ogt)@S}SbduyBx%se2A#BJYQ zHVcw9JMj2@s#x008R_O-CO-u8$h*=G(%%va@^>wU7b6E474FNsLslCazRmv6qr4M#+T#& zVGV;sv46LXCg(Owl9Rs?DZif&3&QNNZ~h%RqTh!^3<*bfI#j5dOa}4w`~uQH=HQ=t zdC*Ah16C2)F8$%>!nDK)xYYSL&XbYi=t!2m04D96Oh!i)k)UIuyxm1|v>2Py?dLU_ zWRb7G%f<^FPGY>@*FLE7fA!aAPAn^%;*Ic}-7>D8fX+cQyJG?M8qCl+A{(8E$_g0eUd8gRkUm0M3%uEvuczyWNeJ_<`E48rqszLL73*Gc)d$+R`BfqeQQOEafWrb7L2a!b#dtd@@!<)F(W z-m!o8tfn=0GGh6g4RHe(;R_TPO&Es?qkEsg77>YS1ObI1rJC^|D&PN&Vc0K978T` z;&b;N>#ZYu!#&7!%aeGWu?bo{^)9)0RGS+9%n*34`H78I5o~?Zo!{ni9c?md1#^w7^b5k){rS}YNNPCUuOni$1)eq(#<+J8)CtYW#T&i@8F(!!xPy?897rL z^qo|%7TM=IjAS=ye+NCkjES%znB(br#Af{Ad@&hi)5nVwgp(`L+iF%%?8HtpW{`@E z2h4@4zvFb(gAf!tg%3N1erGPuUkBVR0LYc|ho433r;cU@V4LMBkYUscCr=wKvSA+r z=2cJqM^ob90{plgsBMoQYuf&D?8RV(@QZo9@6!cYf8dA7i zi1zR5V1FD<KE3`2WdV)BFmfaxICUwKt4!Y3s&!DiXz$P8dyo=k6rw(c0}-=Q^~{+=~7TJIyJ@AJX=_ zg1py#kJgPmjSjciqTeY96^>ek>|8qj>9V-JLCgY&0}G8A8(v6_)WkH@vGf9>UWbu& z;{_rw@@W>Akykl}O00U((PJU#(X;?6Cs@c~rmpFQ9=la=I;4GB7nr;$nF0m6OH`I3pibVJpaoIOa8ax$Iz$K5U%StLf@rIxVkoM z+`yjFiov^gy7Ic$y4P%rcu;eRrjgocOWGjgPV_wMae{;mhlQ4>IqNf?wTnzr;MO0; zWlbl?MBfT+cQeVt^NYwn|2&fWdkfv_xsICFJf`2Le!+L+8sHhH$;da?299g-5X}c6 z{66L-{IM$@sV}+$hf8f`AKq)KmKVhV)l^c0HA_ZtZ4F=842JLfg|Fn*k;>GA*hcFD zt1>x?_pR+5PVaGK#{4`0RyQgW?+_&ppFwdN@LFOGWAMEU=-nz}Ci#hC`Nk#VmHH(@ z)22g0y^)86FKVNNBXm-PeVbPcPwyT^B4mbQ)6s8XZVD92ST7dFc&-)Js>PAg`dl#C zFU_I6GEX=?rbIYPvQqfeWfVTKk&iE#H}ksu0)X)WOPCm80^( z$fWal?zMAxjldPuYX0PO@jp5Y+@-Nq{{(pceI#7hr;L`bQG+5w9%k8GjJYvg?sK-TV~Z82}>qD${f$^Kg~Qmxmi$KuZE(1-oY%9ZH{vL`+xL3 znDH4btFL3*J=J0K%n|Tb>oPc6RS}MQCWn@kDUe66b(y&e`p|Og640vmlxL7yjV#uV zh@gG`RqK2F{6!=7m{~@a#J{vw1P477aZ2OC#v~>@PbRWS#Y%I##{nU+3+H{N8 z>k-imk)4q110Ev#1l8!j zu};QJ7v)8{5{8z~IS1nU`EaT$fGck615Y(;*4ugy&v;QfJ5!L%j10|%@4JUV{}uXJ zEWhIIs%XBH0J0|~kgY-@lB-cimmOPS@(mTfg+>bsIUa>-W`2arqf}wft6SjDrImai z6;-}%$9_2X$lsWp;(FW@l}O-@EQF+O_%S}qL8rkHFy%=G2}+dXj%?SOD|@lm$!tsQ8BVVq6f%+q6BgsM4%qcf+o+EKosTv+iwM2XT zcA~|XVqm;tCXDy^3Rj0n@xSgF%r7cEhO}K|aGBFj?3S0pZa!2Dm$hzz_v|&$M+-e9 z(rOaDPa9E}9n52XCD!X3Mhf5=_-@n-covqbEQ<|)hSw=C3;fdT-PM5?o zG_!OzdH9DpxlJB!-=6~dQ7v*;$wgMmGtou0 z6yR0-_j_2bE05`%xF6l$eHpo4UC9jFYDR{Y+^=aEsR^{&2^e?E3@h*U63=-gDz2OY zqh(fz^&xy`gd}d6=uS@R4x{#-GpKs6EALRv4zApmmS$02HFLpH)1c~5Q4G|~u3|VJ zRKWB?0yP{5APN{qFZA2ocZ&OZ8#il#G1?91#96X|!4D$Yd z*cZ03lBIEA%!)8@A^14Eq4Xj+^YSLJQaj3cuQd8^Tn^8&cE|Do6Bw@}zN|(k22aQ7 zl49O*{NnflaKEw;pVk`(rzXn%!~J^`%TAjJ8DRMsfAvXZ7svgzzcmjxN8b9gk;7R% z6fWt9f_|KZKb?BmaQPn`jbbcrE!c#EMimPhya2mX6wf;?_ZcaiDn;HM?8b?4#n|ED z5>j5GL=+-(ICBn#thU;@R-@ zmK?JG*AaO4f)KVRZAPQ@p5RQgesFM#E2`L13w`oq*^NQg==`hpH`2*E=pnb112IJ!@qWIcWoNT z2T^Vrahty`^@w6hW)p*NQ^5(>m#|lv0YCGWU`4?Ve4$~HVDed8kUOiBn4gcOM$+v> zS!prROb=qW6@3MpOZ(Yb!yKUG;4x64{T`UQxQU&D6zGt-Vf41d4m#t&Uixg{I6K+( z3V6Jz5e`1+PL!jhXxv*F+FAVx4_Q8jtZqp`NxKl;pq5W(8!i>aN6~~AYkuO!bX~Y= zpaqW&98XI!Jg`reHtkt{f<9Q8Lgi(alg>STIQ0!BzvrYA-|QEdYMcNGuMWYct{C_! z+6MY5{lbxHkMZVULvp||8sr|sSidiU8zUW(XOdmzUS!6$jc|cf0vUHdka(qfLF=Y? z`Y6y)JRgvlo-Rig+&L+Dbla4!@eiTfx=hKMc1lc6iTqJ6XCpteX54b;0C~3SEl}Gz zm6n^`AeX<6Mdg7!9MrIae(cdiu1cyn_(B@lYbgtQw~nL=>tm>(I}Nfo^05RN3R-ar zEEkr5EbZ06xjGHoR~DnAnyn;vj5&-hcO$kVJ;X8!ukNWu3w*rLL4~2nbjmc6vEUUE ztmnQ9!=JuoIsJLGCzB{V`be}Z7SQgSB446MPe|E$eLUq$3eKNbiW8*F(U}q2v|YoB z+K)7(H`Njur6-?p+j>vbl?^)zPSocxo&4hr(OY03VWFsMms=W#0i)_X$V=O zPl#Jz7yPj!9^E)|0_~J@=Kpp(jjVPZgI&@mIQ>;$Fo@>u*+d$C4gp~UL8NW{QFerl z8?dQ5jg@tBcPAP-9>v!H;OUGu=J-`27x9^V8N=gL4uULP!K zWl+Am1zIXG0sWaE36_4{gGO%k71I$OVDJuYTH1rA>^zL-9*qTiM+LzB=DXqOU7lk8 z!%HtH5sB!-%qW(CQwx=Or>e4WfnFKA?zJA23K!||nqss%HV#EJE=P73&9T(b9%i`U zEZW%c5}jT$f?sX_04I5mg8nhO(4t@rSXx&NZWJkSea#!6$Gf7WY1j4>{K4}}`1jVV z=dT|{(9(@HXhUxkhf~gGkq_6#Rits{J*bn>0cBHiVah#McB^PlQ$Fk?UO7AhE!icH zR_$B~CMf*1vjEY`fmVJ&W!%lO&tI>2kW`aLY|lYdRT3%ys7Y{lDiV>Q+rd)!~mwReLBk zl0PSSW2^x-JPu)xrlUS) z4hoR4My$0SEc05$(fc2L;`6kLhVWF@7ViFnB?54wYc-UQOy_j}m0vh~_pk~MaykLZ z##Eyl$q~r?i!us4(*UEFJ%L{KgOKgK{EGkT8)kZt%>1wzgqjPw+hK0lHvM~kvE^iY+tAm+WDpc)#aT>M_$;HpY=kHrlifM z!JyeH1iVzFK12=qb!j4LT}>oBkdJbM*Px`8WpL)0u}H!;mt_o#nq3r}ESUXppZbK5VD$E%Ou1T^R{EXR&NkugClW1R9 zaXPvFvYEQO4A3(Vr_)_e9BDzTGCcp*4XqGGN?)va0>ynUL~O?f#A{`s)U{nGcXd2s zwX{X~gze$4#J}I{E1m9P-AiWhWP}$y=xc&T+#ZQM_L?Fex!urVaswFotc=-k+n5Fz zPvrP5-gyqZ`#cgdHr-&zm0apzH;Oh}vFLduAd{5*m`U=9JOkxg{7%UZjb}`dfuV`; z=2t7a$tVDgpSBOJ`Su7|@yZeJaTa|fcNgAj96)TRAJXo4C(!j*W>#*kA$lj>Y2dbM zsy%xTO+!LLJ*SA`MjP-~jw(SugA;f)uMRVZ@k0>%(3yF@q@U>chSTiUO6us4gc>)$ zhHd5};i+f3kRR*A>B*5vKS@J3Lq;k*CJR%gh>O_-!kcuQyxwt~lWpSm(V(--orV}a zBN|N)Ve0CURDWAL*Vn2g4B#<`KivE4$NDzhd~i8U(Vjy!z89jeHI_(_`3oG5%||(z zzIa&e6qJ9e2wko!L+Si3=-=}W&G;#L59hGVu%?UblLr-Kl=)z)>f*;o4~Ow32Nsgk z_mxnTLNQW)e;wrrM2W&8L@})%f8zQ|^7ke52EUz(_1E!nHZl^eJI&g#9C-Wd(5u1T zAggQy=((7}$;B_pkXA=X)3Y1SadeBHxiWtzP;+}(-BX=t@0%PXaJ%}2<5&FO?BH4d ze((7*8^$=5gBWQR&-0WbG+;7Tk+S7YeiK35LxY${4pu;-_5M0gRU&hr75PIT;;5suzH-#|gOPdw?-HA5)?@No97to)O%hg;&#)KF@p-dzg2i(xNYKX>!70(Ll8 z!pU&B^A_mh<({VOCKLCqy+-wp7?8r4%?M^M5IsYn$$57S{nw~F(4Wq`AwTvP$>aZbucR%6Y z7xGKK6QcNpWE@&Rwq)qiDduJr%(J38p`&5NR*aWyxqxls9}40XzOpVe9dYQG26pq7 z-=w~3IJH*urY8@d241Hl$h)j2R-!Bq$c*X+8-mVq7B)&dwNC3fvQi!b{P$ zz-z}%z{I6PyH&&SgR!5;pATc{AmxoTvb~IE_*~=Y)t}}}XXY5ucgCqT*_-~_mK}FB zVy1M-!P4GYASb>REbWW|`}(BuHGczg^~X6*HkzS54D)ymc&FaWVIlh?m&8WhCEs5q za{n6>k*)Jr88DtlCk}w{G00osQxGq?E26f4;T5BkDfA{&KJcjQ?c6&HGc)O zsi9|hCQcCeTKr)U}(7OC#PoaWw9G<^}VUZox^C z^>ACm4iccnaCG;dvVdNBYw+6ULpb(>9CuGl`&Px59G_?MTA6Nx0w7`JDENplJCKtL zmvvC+UmgK7vQGfJgbqO7Swf%tJy^k9f!b@|CmjmA2}Av9(VIE6Dfm4ZaNb1o$4{q4 z&XLgnNd#0gNPs-q3_5n<3+lCKJXNaH{Wo5S+qS=dg6J$1Whc`yAzRI^kl8;wiD96N zD1MP8G1|2dZ?anrR;cAMgP&D04QkQgbG0U$ao{)@A@zXdDgx@2vXu4@OCd)urf_2b z`Zy8hUC!YBxxE#vnkj=mDrbPOZN_+Tq89qhk0CW_6N&6Fk$-2^GWhg)D|kFh5nR_8 zOcw0@M8+RP^#18mqH{f;IO|xD5974KE9rQAr{gqGUDU(O8aXZ0?aqSVZZ6sW3aaum$ehkiOM0&xH5MsPIYhJ#+1E=T*NvP$0M$*$<- zvJmv^!UZ@#=>W;4TBKK`Ly{%wP&)1aS8jpAbZn862kf4AfKqZ3UfXa4J`5#b7n%f9 zo|~W*os+q8%r!MmUv~s4;?3IeAoHB=})c7WD)stczsQ=@N>JSXR6SmJ&(nn*AC!M^WbggZOSn8NPo zoQyQq#gm7>L_V*sEo4cuzR>jU6Z*%ogDNcZrB_-~=ogWXVf7+q;j154=r9phzpP71 z@EvVx7BUx36D|gw^0`d2Ts2phn6Cf-Xjc5e>{AiqnY;F3Jrx=9(AWg`BxmC{DVs_B z4N=U$fbZ<0*o9=h@pv-su?%p2s`ZaY<)N9tK)phEdu^-m)#KN~doqziOUEQ3yRunm zwM?DLkI0RW58M$5<>LgF>qL3H$aXrrR6>|!q(Lv$58>+mz5OV; zSlW***R3SEypa?p)`IdKdF0pWHtL^LMa}e|(v>R|sDpnk9X+LnK2oz62I^S}cWF%$ z-aP#Z$DV%&tO~2aG~Egyc|{+DTmRkP7vmIgP*$+I|LFNxhF(+H2VBNqgHg@lFp0eu?(%htp zjEd@vDZzZ3vKP9|ttI79s$+SM$ zm{xXc!>i4409>d8{ZHS(C5|~bis!+)8GR(JJ_htu-(HeysYMdnLU_&I6Un7_K3rc~ zpfLs3T%1VGJ{m&(MShvNhqR$sM-II=Bt?sPf;2aGVytRQtJl8-kJd}jeG&1r&`lEs zJo*KmHt!{tK@Y$>Lq8;T$RO653J;U$)jLI$yaQmqlSYUoM7bB zM$o-s2;A(k4JP)epb_e;&`7IFj?XoB($G7VD3leGh8BSvSV6-P4(zHRW7m&C*WOCu zY|;9Mx6%wy1dqX^HgmaW#C(a5f!yot@oSQ>HY$d@Kb29gnRTU|-LhySyr>=pW0uy# zU&h}6J7W^t*MA->td*ma$R{|bOzd`T;M4I)CdqU!H?OTY)eASiy8y4)Vx}m)0{*;S3_GPA zp#Si4P+Q!>4Em{#($viv7sU%0yv~8omUV%OZz}Nnx1pTOZbWSa8=H>9<P30Irr#`sAAZe-{D6r>jC)aL z1Z?*kj(1(R2ipTh`w1=U(2^8=x;u`L{`@kojPCb^D9u*1J}fG~$x;fb_pL$rkt|FI z`|FEka5|CPZAisKfvHiqahNHA>uyUyvturZ zUoU`ukFtsRbziXKh!QO6>?8x)2g#NFA#_SMLdq74x%yW9(jwa)+M|clV#r{bUVQ13 z6!{jDLOv|)f+OetwY@*5n}<55hNCyRktp92^O~w(;g+cRP`%a`9}nL|Rd)ayyg1b+iw994CR!XBqQ~KbE3X4Fw2A7Ng0shrxaS z3Bc_A6u9l04HU>|qYJs;z{U$awEL1F@9@<1xZz|SVlNAk^_02LIw%$$nAgeiyT!5; zn9s|CMjE306t@!bkJ}YkrEnqLDZLj?Y>_~@c`MP|x|PIahdX`ua{?Xv;4=N=cAR?c zEG9V_CyBbkQ8Li_mc%H8l299UV#%M&$`wT+?^=B%zsd|%?Ds{d`l{itU(VF*avVH8 zy#XegZ-gVJdZKFnX0*K&p*t7OA%ok}=#1w)I!-5%c(otIV`FEr#$!&RxTbtmzh8rB zzn%}9UxzY2bL`OXX}XBFeF*YMC?ey{SEFZ=KkoaRKA@Hwnnpl6*d0O1OO8Ka-KyKVR_GG{!s5a3D z^$C;Vr{qbbeNHOO9M%n=JU@i;tgA#hI_lx*>^eq%#5w%jorP+m_o29+vNaDIqK zgyQS)Bu%bflL|@hEWRf`G!y_jlPHR(%tj;e^_vHt*y+44{vnFxJh}Ed{INa+Zk0KN zU#)O;sM^Eg$3k2~0C zvzMc(0S)Fycc!AGj$!C(gd4DvGZBvk;g)++QTDGv{5wvHe8YBSG_cx5OiQ>!U?bR~ zdQ21_vmbmiu&Y^`P{Ne2w7^$HF2f}e)u{9Ad33H0{o_Zca|@GiH3&~Ix5A&x=An^y z9pMo7VbsnvN#GKC03288B;((C(fBk-4-`q!A-fmB;K`kAt7Qu8J#PgS%|-bj*8Kf8 z>e^K*ij|&@bzH*HqJ{BjV}B^>k*@(Wk5*xss%luEd=S4Y^MF%S`(Z;MqbY-($fhs$=#0V=0)7xY^|Lb4B1_=d!BQk;SpIQ(2D zcp1M{jHkrICESUe-PQLrA zH`?}0Qj`Ov2YyV}Lz+$s@a)`~=<+Rn{`#dSQPBrYt9&lXu^fRmm@Wo}%@J%yX9BKNVc?=!%1q%Q3)qh8p|kHG_-$_ltiCPs zwYggYvjYFOHf(<571$p&9Ia5+L?d1*i}da^T$Q&EDIcgn@ly@>sYr!ibmT1jcK1pV)?lZcxXd!x&JP{o^)LOlIsTVx5=Z(k~Bod!AJOD;E{)C^F7yiR>pS&fC zwKIpGD@{=43nl0=avikQv1RvLq;i;5y$^u_p|jw60beYi@a~)#G$fFRs5&j3EVtE1t(FwS{RTi%}brS6gwM8Ee%|i3kW%-uwgZVjYBao|5lB2oD zMh-vllSj5~GN?c^6~1XKL}MT7A}R0taMQyQbSb(L-JIx(?pPN5)2F4|h3IFpKMFpS zg|Zhbqbm^>aI8%f*m82Fh;Bo4%^-{8Eg)zn_{<+ow)D246T@;ImnykSod6#M$yg{`sDU+J=~o8lH88ahaTT9g5kwV?9V@@%+T3Gz`(u)kW>DgoiD4$ z&Evm)Uk>LTFsw0c41!rloB>Z(2lC_hiR=^=uwJyGdhwzNc_*D$ay*RjDOA-lY z8IUzd8n4`U6pVWz@_idn$4~UH;rZ#!uqVtCKJd_m#uHAWPr;|q^SmhJb+Hicoc9e% zy!nh1L!01$r7CIo5ka2om2tfF)(5j5>mAwqmSOnbP-#K?QBnSsVXfp~0Uth;eah^$ zeFoH}WZ04 zi+&UR9QA_sstehvt8Rj_Q8VFcQT`h-jvI)b`1>S$hw%s;x;6ngA1xM4oisz(DicnJ zseOTkX6fK@`zv$4T4~G`yvA<)ll7VtlSa;VG z2GR$-muIqgH_kaSr~7#1^VGd?S9k@?IkAqTb?WyU+&O0<2H^`P!lS&IkiSw6@7{Zf zq-e?t-|g+BJN-urb@Jyx|ItU_l?)ef)lHMO-yKd)N<1Nf-{N7I%UaNKCx%3NX9?z3 z`Vz@Y^T>)Q8G2>Fmd1F0B0jr(xc+unHG*tcO`>aN$su2#Kk%li2LH_GBWPTpG*Wr@ z2_ISMOge&B5`Qoakr}#Z!6qHJHTX4LH8=^}y*UZ~RA{1a&FbinlLkmllaP?jid0+Y z0jZCg0ZIct3U2>?j%Dsn5PmrPg|1F3aV;3yHjMjE zXZS?4f8I*;^x#ht?AL^^7Q~T+8(Z<_upDY3e})e8YNn5ZR??0QLxrgl#0Tc?nENKL zwm%66WbGXmit^10`tv#3ldjz$4f|Fz9|9TBo!Cy836EfuuNgFwRwGu|Vqjt0aOU@- zZenbj2!1cn;XRx%n3p^L36kqMfez+QL(PMPu-|k&t3KKmR0ax3WD$=(2w6gVZMBhZ z@lW8>?LzMyauiN7)fQfNyiRY~MWcG}9f);vLTW<|gu+`g!bjdw)Vo=R-qSOu`Db=g z#jEx7Xy0ervR_r$G6RDR^dmv(aa7@e1H##&{Hr_X(b+nku=V~c7`XQ-3EW!pZ>$Uq z6=A>aDp9hv>{dv;IJY$7}DEQb@bOOnIishD}$J$WW)hb^WnM$Q(+AR+gv-am9#7OmU%zTjXJP7UAqjaZ3>ww78;$BU+5sYgeTjpYQv9elx=B zzW)FFdiBoBkKgCV-|w8+4Np&b^u+w_X?Z8LU3bzc-5u5Ebf5j%S1n%mvWpjgxb0Jm zr{D6y#h*TZbaKzL|9JA1zx|5tC&s=q@wl%|*MDdE=7~Rj>BGv{-@f3hBI%8M`Q{^e)? z_z8ut@kA3M^y*=jWKK}Xgp$_!?o+l+beI471>N^7JimMTv!2*}-0yv= z{A)$-r2aqetyf?8vxSRZ{rLL5Z#!$^4S#jV{0~n(cVdFyrJi-uXBIYJ^`MFHhpX3| zG=JCl@Pl4hf7V+cI(gl?2TbmN@o!8%>!H_AZv4PmlfN_jaqmjHvoT*E{r~3E`>>aNXW_Gd`nvjKul(JG{DwQ7H~rike~-mE51N18=Vlk4H~!m;uXxF2V#vk_Vk1jm;A70(v{DwDmf9FA;sjoY;Ro>fMPrZ5aA>Sw_uetI| zlaD^{`pJX&#N^_YAFV(0A0JrS-~X6%$A5I}srA2paX!Ci_fr@C{rl6Czxa;}>f^6@ zboZq{|MK`FU&4QDX#Vnv>W9y*?|Eyp`>~VG={Db-Pn_|>;;NzFt1mq4#&7fATKxFL z3mbpj{pXLrYU1`wR=ay&{_gJYT=Uh1A3Wnr3x9mz*$W>%`N8$kZ#}7g&zqO(x1aaN zlfQA(bp779?Ck#j4bPkSn=egI%uG(tzxo~DtN-=mZmpmBkb?{V^3prTpI`a=?zUe) zt=qigtnwXp$;0L*PJF|^p7yDGX0Q4FFaLhwKVCD_{fEE%MR#WQv)$33e7U~qH=k3^ z<5{jXw@jSbedG9B*B>eG^Q%Ag@BH=V3+FGae`VnZZ`gO8D0lOyZ*~{v|LFAYe?INvh4WAS&#NDG?Cp~eTz=EU8`oXg{euUd zzwm44oHRf7_Kz<-HPbW|N?vcsUPkPtnM}PavlMj00#}~hG{f@;O zev7|$_OWv&|8C#^n!NEVkJ|k8GtOB2*aIHpyE8re+^Z&Uz2WZ3e|qZ2Cf_)6V*S}a z{Jr|Wo&J#S+>0JOdG)JrsNZ$y*B5_r`-c|JI`Qd?Z@TN+?hhV(S^Ymhad!Q+SDm?d z`|Dpj`PlR8lM|0Q>zWT9eEz~eC3i1;^ZKV>eOd97@*VN^y$`Ry>Yn${*Pi>v#fN^U zyYRFBdj8@Y-}ULmg9U%@<5kbAPd;N?ckZFLES?;>z?s1^@o4up88|ndg|i4ul(@DEib>a`%LGv84^=XFn+ ze9^X-PQL!KpLD-`?)&QRdhJP*)qj8N;&*<0NB3D<=DN>*%n6g%jQ;)P=r7Ki{PZ_o zQ-AJT-rW8A@Cl1MKK9R)F1f0?UT>> z@I~EgpFP>VX6i)~-|PISd=|!k@Xz(>vwv{SO=IVlIrn_zs)?_C@{;lAUi6`bznOmK z;-BvyUwq~5_x-m^>_PnG6~90Kfp2bJcyGS8@b=kfEqw3ne0=?n#=DeB^Px+@uY8eFFo+H`V;@+`h~x} z>_-cioK{S{V9!hVcQ$AB)9yXHe8yk@r41A3KcLh7$iIK7ySMSY#V74LVZ8Fw_muDG zM_hXC!aIKatorUV&bsDn=dPamv=^*+4<#*KRNL?x2E;qxaqYM-#h=si|^X;zV7hX z9#KE}uIF~&)p>jO-e3I7!utOsf85TeJ z!C!Qrd*(YQe*RmByPbOi2}V)nEL)&vqYo z-D?&f@VK$X7ytef7ON+`ZSnQHpSO793qRL=<9ToI-t&*u$b2Z>WFts~@d*{_)?&r+VM$e&^92 ztoPpk>B%ckeg4EJ9`eL;&j z<&Mc~cNCLX|LZxEFZsf&HeWga*7EPuoA_jW|6lzN-=DbVxldU9`zt=Sxbm|5{@I6j zzWUV37ku}O$v@ncUGrb{k!z;U|HUb9JL%O|f94f`JNeH4JZbX_AN#<~=U;T;Lp-4nk*u6pl*?#4fPRsE{F-rfD`)6Snf|4aM34}Z_=>aTdt_%-#p_%rr% z|Kib$x1GIXanHNI%D+EvQ~mU-MkYSI^JVkv*1c!qrBA-w=QI84(bspMTs?Ji<1hbk z;@&?y(*5*PZ(BI!@cxCFyZ>%+{T)9X|JI*8cH(32`uxNVqf3)t8~XC(GtPO>;!{8U zsl|tX`1^~OeEkm=|MqVmvv||O-*kWZ*H=s)SpA2|o8JHU$#>MwoILXA->*-vY?++B zg{l*i%(>?O-Bi#=??AC?aiR-(sc;)5Y zyPoyJg#*9wiiI2RIAh_~Upt}uod>O7d_eV2%iiC3-b)u}W`_N4N|w*ax@r9jW&YX? zm)1X1J-zg~{F{H;z4HM>3s2ZTTmRhW-!*x{So!bfPp*2zuPt71#c7MD|7fIs$*H5= zg&QAt&AFeS=>Brczb%}+`e)sL{N|YpUwZea7jBx}T)r!wa`Hbfocha0oc4*~3%ZZ~ z*bUvcod47AmEXIjJGK7Fi!Xmv{CmOQH=B0f^|o6l-uLaN)NlLrDT{Bu?O^xD6K^=} zRS&9kpYyv9D0^G|OaA%Hle2|ye(BZad;MEq&nC`#&5nuhpY@>noj1n6_x0`X*;7CN zhvzN4k^kmj?e1sR|Lr5=lXt!Lj{2j|`nAP(p1oSeU2w*U_21uh(P?{s{c#ib{_&S4 zR%V}2=0E*)@9jS6hj&l>_v^~ow+`RB@T%qa zmi+af{y_cCfBnsgxBtb>-LXTL)W84klNP4``VnQW)3)uHxc-dG#+!e3!tk4)^_MS8 zIxHGA|LlRRe0Uh?|M)yIIW2j~YhSke%_a#rB*mTOKlb>?($>XP;HomEL zdgJ8eFaP@=icHzVNtb`5>RF5bTMsAyfA=sk@kb|5JmvJplmELtaD|^{{a3u=qnG`^ z*WXmzbjoR`o%;V&zXGd&I zeg%6D(mu?U6PHXSNrBy%0?dFERQj|AU0ve8Zr6)+TD@_f;lGa8zzZ4mIu@R{>l<-)_Z0%YVygh_4wEm1 zx);x*18?4GU{3YBz+TXqN|IwBz$fI%fI4GBkJ{ik{EBiO&nn>K$399Po4cBWTDu1I z+)|K8nbnW7sDk6b-qtmEs)+H^Tw(@v&Sd;P_1Ntv2dzGJ5B9QAJNytn!o42^>V>`d zF;;A7AcR<_L!9Ryklkrn8{HOSrqNr=URuQHa{+V}xw;hlDWJRix^2Dt?vD4+GuVX9 zd$Jjtf;nPW(cgs5+%0Gd^#28RVQy{qYCqH1x1%2h0lotC7v*mHvG$hXZ+~0ir@{H) zVvwUR&@aO$z1W+%&2cHRbq6`6KI{7J26G^UxSw<3t>NDFb4NM040;6Hh6eqC8vG#N zigo&aH~00PtNv|h=?CxKts?szYz3Pd@oc;w-Uat!?k)&@-uWupy$dM8^knh%pZZOvz@k8jWxffc~)?*Ij zF=&e1lm>cgL|q$sBk-M&qi;h8`3KZ}$X@|6Z7=fRo;joYUk1mf;vICi zN0CQFgSq;)J@8H7e%xC@4;!G~W&P*S2dFjdGXtZI=x@w%kPve(diU&E^ssxMSFknF zT@HT`dmFwA?u5TX8T&457s9uo7eT!n9bhhJXyLO5YW=~i*CGdMmgnBUM$OC6`D`2j z<$S~sVY5GL-vCN~@@$7UpLZ@`H?KX|Yf3KtJbUlF*xUCrCV~Hygzp~ zhy&KtXBC@0?ooOkUP*(umbO0u%`a@CQY!lp}yDVpe;}U-(lX{1U`Y9=MdNc+{5v{T_2J6 z>keUa4sis1f<6SEJ@fSS?x>W>vC{X?u^0Hct1Z|Q=)k{#cVFA^-2zhhD)4#sPP!lW zsLnpukUfi4=y}NIDf`3jj5T$9GVDY6Ig>u_+_<}uJyUlb@_2UkY;T*;H}SK7XEom{ zXoEq&_n`MEn!xihK7=?!4z1xQb@y%H?u;|9{`2U@8mOTuel>lT;QKy4 z2LXMC&Y6llK1=S-_eTXdlQXQLb2guCYgIuLowMX%A3oOg+0fs2Z3`c3+sEbx&jnDY zUk6^j`K;$NlLTb#*aG|}XaywbtX1m)&4ju z_6#_0z%PL}*X77kk)m%AGu%j$@7_=Ut3I&Xi}gET5O)~eE@Hgjo=F8*I|O!yUkV-E zf9Ei6Q}4cb-%}&VyMXh0p8DL0-Z;OXJtOP3$-4}c2R&>fPQUF?^B6yZ>@!@1c$TZg z90T^YrSCHgHOFpf4ZX0h44u2w=2@A`nuo9rW;WM=AIF|5Q_If~ek~9}%;8M-o}k+R z=H;2MoCiL~&v|lq>l!=S;Clh=1as&LFbkU4{Mf7K;-?BV?=0{vb~NH0XrtE_a-7g- zOELc%I&0Zi8@NAfX876D3<$_GAPI2?>Qi`qrsezhcd&yoe7$Suf%Tv{6>A@D#2FIe z^tB&*uaUzZ%v%*x*!_gqqxl0mcV*rx7}S!^VM~E$GX&c39U#94UvnRV9z-tk`0Nzq zcHRQN4CKJ))-(40Y=t*>j$Z<=>~8jf4M3YUjBNtr%sql_Z3#QFyQ#?Y$pW*W3;o#5 zjd(XI&?@lfCF_>=hnSgOe12x3_n%dMhCJoHhswL3Sz=nm`##er&8da1z#iaR@Cj7k z4C-e+{`PIG{pvGHe5J4R?5iLFo{jf>1(Y@OIFmN}w&vBuh7e<{zYS`yHBiN_t)PxMbKndCIbDi&@7y@>yfy-Rm&o_hGt1~4~=WwSPK4oom>Q}%8y||wopJlKKImg#N_cq9J zeT*;At)jaJ-9cbaBfvY}L2v9T7{jJt6(k|v**1FTafS?dF4jnZc}p7NxwNotA+Kko zPYZvk@dapN&+!e=3O@UQXEKXyUC$%~*4T)C8fsnl>rOIa?bkkMpeebXF+k^#6KJA7 z#CO55CC-gl&(xlTe!I}^12yagG1jbsKwN^(bLrxH9dQ0Cwhi#Jz5efh(DLj&KVW^& zE5|p5cb+Q#!}vM7XPAO2u%0pU8?kr#y0xh|zy0O-1oS0^Z|MVu@Hf97n(6Iya z`3wz^z}xqB`Y50CJibdVC5QFhUk;25fgJDwUYk1`18ooldiS*%Wcbb@*ZQ!>`KdT_ zjxGT$P?4XCX9yR+U(wl92-F7FRPLzj)1gM(gLA33kLD7->{nUmagq}Yw41}b zqRr2suNtzpxuyQzn9(^)p!1W3Sj#in3+yYP%ktQdXPN@)39oR#LoS`>9 z1Ln_>?PFad{*F@+n*#5mAJ3u4d7hWzS=`o(XE+4a?>y)|U=%oqdY_xEU=-L-LTn58 z=^)#izV_li5A>qVeyzI&G(q`0B|d2$eZ42=fwH&$e%!I|h6+3BD=F^Z0XMg_t&a=URuad(xI7&w&D8_uIs# zUEbNG*=tT$l0P+m*zIfKo?$6%)!57i*ffndG&?Llq%I7oprN}k> z0@V2J=$s`-cN`2g;_qn56usi(xn$_a&gd+=FCw-KH5WaM_16?E=Pct=*joxm8OuH03R9Z%h-Q(r|l z3>5{ooSe?G8F)@>z}-5_Jp3GbYj17Dc?NeX!=4Zuz%({x&%yj1_%^WjOgnrIO`yu2 zrS=?bfZqkwXG?K6DZKUW0G>$`xPxuI*r&5uD+4WZB=~n4@n@4m=528GRc*g?gt)`*Nibf1kF#xi3Jke@Z@gDDUsU0lWMx zd=Ct2yW4A#E7%XHhk6Ew!9@*v7U=N-yqo4%wfb0(~TbSo(0s#yhwCo?bk& zbWKMTgL6-NPH9@QI<6hMd_jRl2ix74GZZ*`1{S^4^fuFK> zy!Z4NHft#FkmfPxmDtR$ej1-)&)o!Iedim3_w1_JoZI(^ zKAvTOzpN4O#x8W;m#dMT+t_n}b^KoR?(Ids80ya^DZJ-?1YV!=Y;*Lyla`6MH_y!e z+>f=apQCf%8{w}7TY>Q*a5ngC=*#Y51YWW<)Jo^?>~j=B~2 zfKFfY+E00pL+n+ZvsL>(*6ju;bZ^$R{B0EG5VG@;SU-U*grnz zclTQx@x5oPbLrcHj-hwA%JyxHa!OoT5B)T9=;zRqI_1SvA4nf<9%kl-yy_%yMlcix%9EW74u^2ptp|C ziMhMb9DNOa3iWyP9A~j5$WrT>V{n(;YX;m|xnub}?nd9yMy%ugvYtI0C1-|DK~86_ zA|C>mfn&(_--Oq<%n6-CU*O+{dX5!o9_Ku)Zv!5^efb{9L+r`@$&aBcK&HNrWo*L@ zV!$!@GWQar=JD+8ZGA7+NQkkfJ7|Hqspyl!SHUW>dvo>#IR|rKgm~}Y25Q*1JIO(c zpK<2P;gwVP45R_uRQ$f(WE}H5$9x|-c41y~C3&19S&DJ?xsNz&1^5codAwe3lL2ida>;1Dt9@_nX%6(pgb2fhvVG+s{y28;v@yHXlV>+{1e>}N=r;G|yP@)1LCEPn zke3|Ju?E%Od2?j>EAeY1%O8T@1=5d`+l`NBWd5=*cxP$#-(a!{uXqr&z3aqiT4 z-B}IV!OwHw1NGy6-JyB4Z$+Mmu7M`}Cj2}zdE?xNpY_OtdA|c+LGNq{KJLXBXD}uS zk^e4!>dwXY82Z7i&bv%Zxex3^eLrVd*XQ#*&ga}6bk=LZf3eu_;Gl3(kmMG^=?0zZRHw2JQv`W&dUm(A!yh(Cj= zo7axK1--H`r~SAm{X0|fPTF^Y?nooP9}YuX=(az4lgB+K`1uap4EBNjjX2NXjE%Ryjo1SI?s`42 z2Ys}+;XS7ejKI4`aVh+qK2YzFHM}2#IAg^@^yYJ(J=oo;KAyu)sCRf8+g5G(&-CJa zd5GT!!^orRz#M+QD~wma27dvxj9ZHRSY?93w*|DYFM|#6p50NP z-DgF*wJ+ZQ^$yLU?@YxVc(<%I=uiJ+eSdl3865ygh(9Bw(BWRp?|t!H%6C>C|30T2 zyZdwR&ZVt}kGohww*gw9A4TU}<()*X!e0bdm*O4wJ`3&kxT6ts+Rp&)eF%9Tq#@q> z6#o|f2hkPK!5n2T*mn2ug{IgB-(UJP(Hocaix{ZJI4rlGpKi94BoTzE~V(KmF4lf);(!&p-V!%cb?te9&-O|3giy< z4zvlZZ>~*XFq?M!>kzjIxYHG|9zKJ*?_;2XJk#f?P}ZH9ig_x~9GmB!!(0D6_$GV} zWFfvIQ*e4AVcUZ524H`?pw=61M1SwoJo49*IjFnt!28T(yU6qo<87>3%q3n6K zf%yVDpAGd}@oxd^)v%ohD%zmhN3faeYH$I(-?!?IA{Tj#-2$Hz@A;MWh+hNK$h*O9 zz?i{%Z$0u*BR=QuRX)Q%px*-K;O$Spxuv)Vd*6efdDP|D*J1YzuK?aT=TGsef*juc zW$@GZx~pm*X@JMShFrkg<2XpsRnX<=^dEde9{1<@i-1kP1eEs*+5zsvo-4qdJAirj zfNj8jyko*SoO>(SfP8!@-ZN{hk?T0{O!V_^6oI-R!#;)*`Bfx zb2QPb^G>a51Ix-&yf=CzfDS{e!1yv2T}pff>N$9K+_h&=px0)-jp#%HAJ7$`1D`-M zXa-GlWIpCi^0?a+oA=hUTZ86H@fm4BcOoai_#^FN zr^wD6dP(`cr<{ZGywJ}A_iC&?yQ3nH_r22R1JquI!A6iE&j4pKS8XbuQ4O1@B3H4S zrvg7Xw*+5nECZ=_pM};Eb>Xcu(u;Ey_$ud8V67Em)<9@Py*bvw56;RRdH)NruNU*r zqH~TK)bq=s<`AA~8+eCu@_WY$YPm1(f-{!)&NJhqEkRcW_GbUiYTXpsyWu@(BIo!z zcMA2stN?3wKml+4b|cDW^~nNM zD1#hbWhyE6#Q&p=HA`J1esB5iGDimN!Fh+UZ3E_A1D;=z6A!h1P9FX3!S{{l)d9zW zXVC<+@Wz|ldsl#RcBu7JF!YkNLL&d-GfadTsV1Zy$r$0yzizCB)iy zVw}Dq_HIAg-M?q)PMx{H-o!STKSMSq1xZdFV9(B=&OMYiWNVcls6C(8h`aR}aF5a? z@V~rIm&k{%1#0kX(Bq(kuX2vBeYfE+gxbG9m$ZSqu$KZjpE1Vfz**+dWyGg_lsvj_ z@4hoxMmFypG4|$JSt|z#dE1Tt?>&g#)xmy^OQzyo_We-z6`v;60QcgH}Pv z?~Ac-dlC8>@9q-(-T5upd*CuKj+~=+*TcxpbqHNSoO~0yuCKTL2(}gID7qmq-PhT( zIa1_=eBNnu=-0w#OoqJ$-GFTkYVR4o=~SHAJyy_}%RA*g^}GW*&u9a1wjBK|x}Cjv zH}p^Xxl?G4t%Ys~`*%P}jVg2_$l(L1VXpvt7Rm+ia|~LZFaGvq?&ea=={aiK+=%ep zGwk-_p4~@^JOiD_x10fa7<;E5+ref(+km;8#oexe6u5(=Pv@rM-Ef`^%=PQ|JQ!!a zBmim;&}>8-n(~f$2i%*roX1>cKJ--(w1EI`J!`uY=bnSNH_yyjeMbG{c}dFn^VCtl zg`fG-5bL>r^_3}foTE9NEkJX8oF@m`kD*_Ozk5r;DzNqq=9mKHS%n%W+-(yC_yS0M zmaKDKFW%wMNXj_-+XC-1=X@(mF}HgeLuU^)==L7|eFX9)po%TWuk0JyUfo9YS04Ta&~+?J9|t(4SNfF z65>vr%R8%VzxLvJTPH=QY+v@YHWhoyp&g)qhEEO>FpJy-<-J87Y7oy(!&0m{(u=(P zWau*FW6(KbTF?p@CMF@)-koC=BtX0Q2G5E<+R7SJ@s8Q!PJHacy`0}87M*7*J%Y~s z1#wNN{Wq~4hIeN3%^+v+N1^HhKE^h&?L&6}*uyli-(DmBz0XM=pJ{XHe$Mt)xKYm^In8^ZyCJjzZZB;erBP&z>Y@TfjjAd4d^O;o79}%ld=xPU3(u6LCZP% zKJHte&5ii`n)A$|Gv`rs`@yxqyQR&Z-2IjCo?AK<&&-@rD zgdCoyv3rm=VrxUEK@D9C8qm3~ozMch0jhro+d;6l6z^dRT^rk0V2&#AjvYdt!AIZo zfW7ONp*swp6KCFY&`F)S4azytxhv->-%-#CF)4ogp%v_*Z_fk$B9EXSf!6wZW50{8 ziq13lJe%0opb2P!h8_qeD;vTx1otZXKnN3!1&Ho)H%l(*a9}02V1Qd-*ftIhTjU@pO^vG z3FZLKXkPQ%>l`TW3qBbr`$1<+8e(59eUNi?+%;{&So7}d#XqkP&=q9+$Uv$;2q8Y- z`BdEHwfI@ncxSJI44Au(PHIeo&b?QF`aI{X;O)nEiFQBg6A|M1x@Yae-C4)Ew6CGt zh*k2_IL@sS!Qt%5}GDFXjDbzg7#j`!j}$KyQqQNedG!=Pt#n5)2UJ#(0= zN$e;nusLH2yraX7D9@?mynJ?7fVpeXybtGG2duk-%^L2fiB7cP&DjF>U_Uj`0q%Hh zieC7H967)s*NDrI*YNR9yJ)d)c7|Kc@qI>?go)@5=$d zCOUIw=%&H0UVJ}f=sYLCAKZ=i_W;-ee;Kp_Rd+RTZ|)Kst8`;<-kMnvC?zIcX!NuS>c@Fp35w=apbI`3&=d`}(B6W|_o00DXm-h8H_u_ZS z3U>Fp16e;IpBl&KCV2POo{BZCGu-#BLO1KnKD|%7z&hkI4|E0JfX-QJpjf)kK7@U6 z&yo;#=uFNz49q(Qb+7hXfe+Zt?TpT6-pjCg))xTv?q?aSVhiN8#+F|HySX244*jL( z^gd?L!M#|Xitm65RG%i)Gm+l`o$2!jd-O+~`K3M|1sK8Jf_9+Zs|vP#@J*;Q?uFNP zV=vC=KBqwyy*|dSL33l#IhUU?d`7@FFpGT@{RS`$?9KQD90nUfiqFm-6wJey_7H8Q z|5W_vK~kt^?EKy(>&Z(y==I%44ynDF*N^+R_e2@^&e)FJ1p2L@8-gEiU`O_QHVyF^ znn$+R&FGGT0@*p-nRV@TV~_m-YnwYI*8LdkJjZ}LDEmY=i=X^;U=7*Y8PvOA9sA#m zO&|GbbmnuOWuU+Nv{nTx0C0sq>8v*|B3BH!7I?@4=B*G4v{ z?<@N((7Ols+yu_E8`;n0@HJqba_=F&`+Nr0h;89F2X9~I+Kz1h&RzrdBflTLww=gn z&KXU`czZbxZ@mmZ&&OS*$Y+BsU}?F4eYpPs6%};O zUoe9^v$j;9Dt2wV&}{;DtQrw4g zcutfF`;Mkmh-$X^3~Oh2Qd2+^Ih9tia}{ir#0&8O#-+&K3ebd7LAG=IA`H zAuu}??Vjm4@+j!Q?`yMjED=4j(n+TfS_vF38e_U9bCK@&d5XAPeVupI`Q z(I+A9v4ZS9$$|abbAiA2rVY$>2>)SYuxZ~+j6K*_f;|iIc`+uycch^F{zdjVJdS)3 zv<0ncM_*lvwXD}v2hG3?ejT9fdC%ZyEo1c?Z^V15u8D07**;R>{);|V)dg(%eX+;T zIfp*(e+_wr_zu+EIlcvuH<$TmfqATNFZP`yyOWYf_B@OkC71pcfjuKv02OSj=(MR* zw&x_o^ES^x{O8cQ^A!-tH3LMM z(-maT;Qsvw{CuZPLpK5A^;tt#1)I_N9bqr-A|;n+H@E|3PmOpN^jRfl1Ad#(4eE>| zyN3#KQt$2kXY|bH(MgSWZ-k92xP?UhN=tj_D}+&hFgCct&&Z8?pJGS(T!9hCGk;%p>#**b)%%^SjV{(Nfp< znMO9QiSKrNa%{uU8Zdtb0&pi)FoWKF9Z+e+eOkLS6=R)e7Px7EH1O20Ozk9(3kmWI#d#fV1!4PowM;n|Euzm}BAl_Q; z(R%XEqrIjd2w)cM1nxk4`5lAouB>ODZEVZfRzM!mAs6V))kG&v(N*Np+iR5^!@!-a zgZlf$40e0ZLOj&7=Lj%Y3;5ZIehjphVy_8y_vJksLhsB&$N^-;m}efH_9BoVkGVYS%fW?U4L@hH zrvke_L)-|q^x`@1gYE$C$eH&x?z^jPVshZ#ylWXXtUH+3d(gtBybW5FUy6UGISX~S zYhVaJcfAWb3f$>&Wb3U1&Y-V8?kUe>+%Puxv$GLr$gmA+>*JmVxeIvjgF7fcQ}JE} zVmlxU@jO@Iw}OCd%>qAnRD-XA3Vez!2c;jf{ib=`t>-s@wPuh<8}WJhRe!{4GhaCq zd~@9r&im<~|0kB73hrfA^mQ z{a1jy3}7C<1u9e2>B|8*kfQ_jW4#(^V{fCg#{}6Ps-OuxNBchr_5SJStW_{Q74umq z0ZpK7ydRSVdTzuRYpsVL0txy#gun*pSd&#Nn^I=ZyJJ7jjcbN9|_LP%jZ7Jrz7N2{KC1*fagxJ^M=VATaX9~=} z8(({Lk1c4e5BF99*7y06SLcog{aZT$ZD5Zx{kp9_wTD^s?qwBfO=B}?0X4pWcb+x) zDsegT+){k@iV)9Qx)Yx!KI8ao1)l#dU|wS~sI}}PG!p(V#=^yVISjTv#3W#4iMSAR zN;BkDU~T)Z;jOzS7sp4m^_T#x_*f;fK-J$)q(3@vO zUv2#u_pII?&hEu$#hFJz4V(8!{y5Yb?AiRj57wsQc@J_ovB?)82cB`j*Yiw6oWG5a z`_MO_+uew>dpAUZZh0!6U8zHDZSQ=Bt)>mwgK`3I4f_gUAASXT>zBSyi|5@$e+YP% zGx&K2HUsCb;Ol;K^v1d84$vn2dCYs+1WAa`n*D49Eieu2)m_iwV?Xw%kLOxG3)pI4 zxY0jD@@m&txMSfA?n_#cpX%35;yeL4#s4yVH-KZx_-FXGL75L)1oAaFGwfC1z0ht= z_n%<*dr_VH$dJ9;zDIY14!pV=)P2~8=V1O6y*2#|`t{625m5H`=2=vUX##7^U|$0T zXdxH53CP76HPLwnDf$Xn0rsuU9;`2);%}Y~NS5#+W*cy>4)QA41n=F;^LQr1jkxzA z=soDorLXx0@4o`sPYs%D1Mb({yMZ&=V+**W9BTa}CkFelJ|WgK?=fX!c0l*^d1Ys` zM!Sz{Bfgi7aew2;o{zofs`$#*fZryP3$ zX0Q)xn#cR#o%T5j`1$PGa}__&vVgbW6xqGCq1pn-uqk^+_M&bT+=%Qf{>+`g-+zW- zW%bPCd%5hPFYm=R55E(t-M$VW+s~f9ZjftkDegUoCLz`{UjSu{{(O{uc3Pl`Z-)Lx zu&kXL_Iwfioeky!=J+b|InWkZ!G4)KP`;~sac3K#_NTvRcN1{7)@lUa1??PV9JVH?aE$g`;tzcY@ZAM8he>)!#L1Nz(JPIT(M!_owt z3!2Dn;xpj6PlGgosklpZbJ)zW4DT$?Wk2hn{>;z@p7B0tL2Lzn&}UU$9(V0sQSZI7 zj{-mW4z$^b`&nL!_jeq*+Se!Kv`^1$4Y()sWPNJ9Gr6a=rP$Xve1hH_>#_OVdw1R2 z3beG7V+Vfwu$ylTx&`?gQ2Q>Q3;g{v1M?1{UzwtAhh3fA?JdQ> zg54eF$fMZww~ulT?6=@MkF5oquZGRI6yE$Bh_RpPGK^ip*8 z-s#&`2QkPwby|IS26+rv(?2)#jvNB+FhyPuQf$75GUARPTVr!0o~Q4?b@0|Y4&0A- zz616AM)02n&Xx4_0lq@aCiKo|ZD(``hmnWD9K3n0SMf(JGpX^8U#=u9Kl zF~2)#fieC=~P@LUS? z#`8AWH`gY&vS_3SFxD~;VCrzW2{%e<}L8|3j~jHZkyb|MGjFRr~{}VXq^pbzkQ^mlK@H1W^W0dtwddc(wC z21Y;&ojHstLbNOU$wNFtcbprGe0&Pu20*ULqf6kmw}7!#sQD87iYeYhPdGB$x-o{?c;(n!twx-440#kZ!Ii-Ns=%B9+Jf3&W?f`^ zIM9ex-rhD5cMKo*wHM_0s%xWLA;w|X7C#}9$OPXb5x*f=sRE|e2Ogt_OY%P=ke~< zu)B`}>_*-LoYR&na9^XZbuGa z;GZ<2J;CN2)5t@>eUJCBv8!Qtd$4vM`t{`Vm@id_-#C5&T?=2&#aT0O2)%m|DRIYv zIqlz^8Sz6sZ1~oY2ffXq+Y8*+2=J`jt@B6=um*NxJJd(DuRG9)y|{-tFvwYfzYO1j zEVWnr^c*~oSx~`88qjZs_vZ}j>}oK_RJ;=v>;?9KY>u{oUxT`9V{_ncYUr!z;(yS`QaWs{jDIIrwQ%gpA)+Zbr06_4CkP&rKocUA_eLP@0T_5rAzn| z@5gxkjsf$z(=rc!)(uPX9wg{1KpFyoX26)^z*+6h`Wf~O#J53#Y>f)E+ym5|TBiey zRjvVR?IgxrwW<5gYzw+7az@@1Y8~_HZ;!jNiv&5x$KEQ+*hGfT9#ZTrP$1j8cPUc_ z+Os^yC)o6 zodePY>?6+Dl$_oJ{jBH5{cJ+#b8#HKd0QX@_LG2V^xl1ai#$Hl9eDf9L%hRn z8UNZ+>@C4Ihu%K58&d%RAM+dU`4ynO6wf0Nlb}z~m3t3RZ07N9`7YTF4)x;m?DLf8 z@m@RER&198@2T&#CNV4M&FOPL4bna~K>a--2i~1MjTmR_sxicFfIo=63ATa3U64AD zdnp2QVOxVAYeb#;D#$_UyA%^mrHy?RIHP$kN0$J1 zJOezpt-$);_Z+0;@y~Ge_h*nFHsV}9Bi@e;d&r}|yLI+i{Mx{L!g~66c7eE@m=sxB zf|x3hAAIz+XQ{B40(+51*mHv1nu@tP$nMN@9crLcPn`EYUBVS)_`8E7k9WtJJ%?eS zt%7Y?JF^ONS#Puv_vTJY8?tlByDRevYb1esPmBBreob=B0rxcna_ptwQv5Epw!H^v z6}!)fb-mX=0bd24^9E4h=Y9w0<$Ufl)dn?o2bjmF1?;H_wTCwJYHT6I+$+f5;T-um zdHT!~<`gyT=18R2#z6V$pRaKlc5|gU zfSNPt4=Nx7n;P-2FPUQ%c?LO5MVQ~%0(d9frL(M{x8D?5n8$f@YS_Oq-ltJwD#*gP z1bvE64S7Gd7T5>O>yG!rx3T#t;~H^C)99SvndF@#0c-Fz^n=;$CFF6&dx&iV&#(O8 zYmc72v(EwjrOr1Gwf0;u#yk6Vc=uwTYuNn^zBe{CV*LQM*ML3G@kcgK3;0a+|N#@HLOm$2Rnp9f3UrN0Wvc1j{ zCHIrTT^8gPfsWu6Xl8sdAe3cmrJJ=xzj;Jw({z@He; z-aDwDXQSOS4#b!F;A>zbeydZ|fxiKM6zl@VInOdSseQZa7TDN}xv~(?!2C7r`ZVDS z^a12ReyG=fUa*Ipvj*E>oiQOsy?M*Kj+}ykuB?H7j+`|56xsd*I_u|o z+;8cR&sMOmPd$?=dB&igwRhaPI?yg~mz%&&FaqpHT?J~-WgYVGZ}fl9gL($`IzJWt z3;J-UP2j!~ppUWA4M6=qXbK(Np|z>_Z0J`2`#TP8p>L~$7Swi!DZGDH?irV7j?USg zD-W@D2On*R&^glvc<-3~`24L$w;vpY_vZ!cc;1`gZvs+#*bTLh6>K%Ax&f?boe*Nq z!;ScJ*BXAF-PwJ+y}O?zkI#<#YpTP?`B#B433;-qc!ut&f-WVdhAzcsee)heo`GJB zk2?2Ig?EPCjkq6swq8kL19DS8eCDC*-9uqs=*7VE@m;%&EkQqr?+Um8o4#}CJsYXJ zw6`Yu!F~2l+mCevyg72H_1vd)2HBZ&e2&A8W7ypNvn%jD57{_%o_B_B6=<_|0E072^b2uU_CJjO zF5qsGCH#3da_rk0v9Bh!3i=K>41Xv35$Jlb4ph-6_)kNfQMuf~6#7OyBlm0mJdZvV z;#$bgqOArsejECLkGg`KyNMqKIlSjlfN^AF?cKZ61jd(hF@K8hR(z&`=V?qp7ob&8 zQO@J@;l1?kcwam}Ypj48x;F6a=I|Np+L=>gH$pdqY2fen2{%um&>f*n{WP#6J*c-a(He_$#kKTTp$B z5wlD23_9raAIHc1Gw78)mnOV-&RMN{40zWQWP7k~L9PnC`ONG5=CE!+HZBLo4H1{4 z3-B4xCkqJ*_J`a7-Xrbq!854BtG8AM+ZfO{C6BqP_}K6IMy%tEp5HL?8gvL+fwq?7 z^Sun6MNZI{KyD+O=Q#Y%UOXpzn*$a7^85O9c5N%5Kp%jz`F)oNKlbPx+KqP>bLki0 z_rRxM7sx`qW9F#}eB7J)SD>C_1!}CmDSGFqO~JF51e&84%57|+K~3Nc8=*c&75LkF z@w@66KKAIjq#z?!Uy&gH3VH0;p1mXP+FaVLWo!;I{F=}tME~X#GXD1Oeb7FGEM_j#;R2$_csF zpc&ZRKnCjPfchejJ5Q$K*>#}au@Qaw80YTnHx02?1-S`w;GD+Rz~CKLQO2e{#a{YC z)m5>zfcGRv=lKrx;$Ga_D0QpI0Y1g%*_E}Cm(_t4b>Y5TIeZ13DEq+2UVNrKOW#`w zKGsuj9|gP!U>JMZyLQkdeir2L?$UXzy{Qr3YjfIwu_4^&XIvY2A3Vnrkk%6YqW9yh z#+f${{ksB9ki9!!mu;1EL?6(SkGdaw22o!nr^Uk#UF;#pM;0)@f`+NZ%_8j|w*QbTv823|!-_?lE!ZNb)2Y`Mf&;t7~_GRc{Xn?8~sJ9nNLk9~VDeg~(-uS35ZG(k>|U>u(`@Z7`CgKa$&#HsU4 zeZSdbnJ*^}z5p%Yc^}oE9@>o<|Eu>uI%{hi)Y*fa;&&A2=Us3{?MIMvsApdUZ0Iv+ z6*)kEHQ!P^$2R&EaIhD9$?-`=P9NynU0;QUaspS zA(ylIac}zAOCC@#lR4z=J>X~W+RYKrXP^M?U=Do>+*^)LzW_~v^Hv(MKWk|B{1fOc zJ!E+E+n;mzEQhHWTZ1Mb!*>?CsezmWK6!}eW6hoD_kb*qvoz7QfajM%%{>E-gYqu- z_3qNX?WYZ`L7PzTgtc>OjF2~h4!(1}i#dL6a;Tewx~~p8>sZrySMfUp9YLN$?>TDQ z)ki_jDti0#=aB$5fZ+!B9qa>#k$tu=hT7j@bl#%^-t!LRa4+7q0^U9p zcjYY9-qz5K!kc#=DDX=IGh?&g1ik{brvM$~z7pRI+X&S9!(bL4drHtd#|S#>Z|=pP zX`ST);2iGb0Q}&4CdYpx^0$F8`kmj1ztaw>GjBiF?R_8jcD%25pU3*Vep!hA?jgnY zGku}PXiuTm+yo7Iyk}b)@w@&as6Okl*+Wwq-#kRUGtU5XchC>R>+kt*2N$Ei0puV7 z&Rqk0;ER3^`!Qy2D$VAhW{%1dE^cl>gxkJJxlx1PilOIexN&y z?Ce<{_cYKa_+DpbQ!QOdy>vz?~r>p`ZUBp6S*0^d)*HPJ?obflb|!NckW7PN{pYu`MIYY zTam}U?a>`bJ+~ypJ2-+a;5!e!8ETI$;{0Fh4Zx5$VoO*iW$^YFR;+c3Z<#QM~OZUph_rx^t z&R+q_v&-ZAZx`}5u!i0DwiZOzZcm%@42SH`{eVjZxcQxPX)RT90i_Lg3dF& z9NFK;Q*bWu3=8y|)P=}zfgc9mg>Asx=Rh|A`|w`Rk<<5yG!Qr1fKsD{t%hzH*jo*K z6I8(v(9hYOcMsSKhCxP-?a=+ey=;K?z&YT5^J(@Vk)`XtUx9&vK?Mc|1{D|>7*t?j zV6ZF<3=AqTFfgdVz`&pa0|SGFNK-V@6oU*!Bh8|bMw(_Qnr4uuX_kwok)au6Xa*UE zO|dy_n#18Dm&>KY`}@_q-t8LCoGbT#pYMaBNs9r|%5tLR-v3*9t)j;@LP8T5fVYZ<$TEdLTLf(>Nf59c%{?ln5|B&FoxjQRoVeFsqD?oqO?94nPGf?|1;C~0|Q|J_!1J?Pi3}?jJm&BVV zO+&58`B)3GGkK4#54A^oP-ky*=)1^c;uhdD=xjUmYK{WgGqQ}%+NYH*L$7=S&P(nP zRG%I(_A|!5RR-$J%%NXWgnuL1i@KlDna3Eh3e<~H`JneKkCA7=xjz2k^StKs2K^8@ z#(%4NG4Cb(lgf4sb>6Y&Q|vLOi!JJJh>7|pKI#`ho(BD7j>toBQ-*!@eCoFWQgp`e zK~v;2Z~?4$R6gYXfE;Vq`;WQ-_SlDe6!q?vVmKpakxh`MyDSk$Y?5 z-|gla0O=5&-wpdPr~b3}PI|yT1AH$xp!fJ3LbY3`VJtDL@aFQZPVpH*_t51PHRq-> z*g}`4;Y@hnM&>|Y5SM@!y#3p|c@lh+BJ8pCHql*T--9|M=eh-L8BZPe+Oxj`btbdG z_#B-zW-5D*&UM@OI1PPu@%0Xwr;A={O=*wVZS4A|mA`xgY=c|;iX!-WMvYsB8k=A* z(y&)Y=rZI>umJ3HOzsvm!Dp)s-`*6yOO7V8bN5^Ae7-?b;M?z-m)hvT4`30v*PBrF zJ+K41U`)PQ=wz;I;Vf9ocRHTGO=SDc3T6*Kz~-7v?*;0`{nZ!%wJT*38C zY9hPG=Ve%Xg5GtQ-|tw=yIteexxO)R8zsGgUx2v_Xp4CHL8Y_rC-C+g^SKZ5`nD?G zyLAljpFM5DPlHcjv8?VOzbTBfKmG5?xek8;5_tFS4(iPNP<{1jUWf0{GV&9!06hPm zKYcvoJ$PY0b?$iqy+M~i(=_N#;hX6CQ1?Pw;QJGF5!m~5i4MAt?4S8s=&Z8`HD4F_ z8|f8tk%oEZ(3D*FAT5HfhkcAZBV7yULA&qTA-WUbzUR;(_Jla^!4lX44Pbxn$y0dy zYa(Cii%z5|{h&*MKE}KPV|>khS;;wmEAYlIL*0YrcF19#47+`=VsmZ!Cio=K9NK}} zw|j5|zX;6V)3z31o@wmPOnw)i8+d2^LLDg4<;dEjU*un6x7K$sPFV}`C71=D$}sB$ zeXKLauS+h^`#DgbL1W%8Wq2Q)>v3h*X1}qfbhQ@qh8X*3LG_JyS-wHN7?&bP{vE!N zcdhQvGjx7ejLXxo5B8v}@5?gqOHk*jejfT6c)p|G5WRbF3j95wJ+4)s0lvSZlYb9= z3Z4A>w3=@do%Kre+G5@T`pA#b>*M)~x=1gu4XJT+9nP!zxOQhhK%bE-LGQiwTzr8a zp}$4;98Ck))IxuYzK6U6HjuUNLiO)L&G`n}f_Bh(#;olg4WXh7tl5J$(a(Ssz5u;7 z_>{eZ>gzq&!gqwcgx?Xc)&g`7^uR(%|M+<)X5_&V7!y-K@6dk-r|{2#`3~D*UIjkZ zPE&dUcgVgwSJ++i2Gp5}O<*4hXq2Ib=b(f93Tl4$LHG^~uEY1;dF_F;2=9`)F2Nlz zXUt!x4fLHd%rEBr|ItN!>%G=|GtHTyaI$C}+cSfKkI*n>0wxn`e5 z_b>1R=nL%5eprS#!*zI8UtqI;|9(W*!fydOz~&jqkQ-2aZa|je$V`Pf)r-CI;oo1{ zZn1e!J^Q=ZbLbRw0nCF2%u#dhq>W9ew_Hiz&@LRyQGDq&lt? z?7!6x-^&H)6u1D}z7Q{4w_rl-zIXDM*=voIjcBlZNJ=wGLv$0hQ=2K5+ud!C2xmVb>c zke!{galSv{6JsOqx5U}#I|9~xUeN?PPXGG;pS8ocvq{_){Ga%|0nXZ3=k5M^X6)6o zZ%y~j^N^9xdaW|t2V+*@zoVPQ?iw0UXSogr=sh=6@LNECYx!2iwd^CSzlBEKkT`WB zt@+N7xZ2d&Xvwe+B;prd@Jt3)mnH{Qs{I$yaV?34V!Cv27H&|`g|vP=q|t&`Zv%xE6==r zTFZSNQ_p#dRbVe?P~SRv;Tp^nWzV&JjevQUfomPF5toKpIoq@y{(Dvn+5z_9jHS+V z02-hV+}{>4`l{;ydq~TWy94jc){$qx9#{mM-~-60>)#N53%CdFk9-bw|8sCj4sqBH z|4wpEU-5Yfw$SZ^OZ=PAI8Uj2Z$9IkdjdZ$t8ZB~r#PQu^gn`Se9c$SBmcCfdj;=) zb<2ml8uQd?j@|dxIXv#|asC!Lw}E@I49s%|eOmD`?|bBZFx9U9T?0J;kMqr5M zMj2)?4fR~QR`+0%v!b{7B}LGs`0TdBtUJ(iU~lSQK&>@~UxXf&A!eM0e}4|h^AX)^ zFoe(15AiLa4dh-s?AagC6nPKo-dsVQlXqyN4EvN2W6vEh-45tNTgZKofEAG8*I(m& zgDK=wb?^mx??9s+?y&wTx&gLTbOUe!M##>^d1))aJ9y8JGjYAH%N}m9ot0rNO?0=& z&bUO^hq{(ad9V!*zz2|{ciuP95po0QzXM$dle4}S)-guE*$&^Kb!ZND!SBGH?WY4y zD&GF1UfXm@tu%Zu?RiALIds-Yps#@Y_8t8YYVY^(QhQv4&Q|<9)VMCRflm|a8r|Do zJH$D&JFtc9J=}oqfdu>pPC=HkHfWCByX1^rV@_`Szq<}OjGZajV`T56``7|G@fXPU zS73Afu4xu(PX&1re5Bfs(K&N_JcM`N>+nm!-wU3LW%wDek8C}2+1DlV6ufJiR!3gv zV2>kY|C~jtFVyd)b@%XbJ>StwyHMYW4ls|sZ$Zud{W{Fm^-cDsvbUhl^KBXWcAlOS z&&Y1&bC2AGH^->rr?I>49DQ#soV_df7ho3Kn115-*WUX0tYC}h#k!Mk64ZLmE5Yv- zvTu-kA~r#mhTdF{-|!Z&Pvs%8tH{24AHZBY%%%(PZ2ikq{y+ST`33s`zCdR$!t-Sg zW0fyJimnIl*NCM~R`YiT>U?+X`B(h?HhHIA`)}X`G?2~r7No=*Gs(r9GeW2B19(et zP4mZCYyX5U#;3)DO`olr?FoF0bB}s8oBJGX(ky+*`5D_!W%zIO-apscMQ2~`hyAXD zH}b%qV=nW?-iGjb&F0xDYW7ZTUB<_m89&7Kdll=QJ3!Y2&%qIT`$*-1=b?q}HSpVE zu7ANN#1*V-qy5m6xy>~X`sG94v48Uv#Kkq#=`k_Zh`y0dq4S)Xx9;!y+}}Oy{{Z^0 zKzBevUr9Uow<@ZxMT~yi<-`5ig+Ilo1Tp_Qyz|@#u@C7caz>t?+7G@lW**x#sQb%5 zLFXBZes!9lcOH*pj*0P&=wW}={%h)LwwuhpO?wxT~oJ& z+{MSW58?0Nm$4rK&+-P;fA8%;)py(IspI*{vF}3H(Ypp?yYSZ70$14Or_uQq$-5Wk zyu#O-Q}FM=7}-4Dr4pL7kwFW754s0^27LoA;dh`I=PrLSvo{g=r0b_hR& z_dc$GS>T?kdk#MXR)IEqI%tP;WG!dhgj%zJIyZUo0(?jAL#@>Z7x2z*u?*iq^Y!4R zlRbq#L$;PahhV#ny#Vm>J#-D;=U>4sd0#^3!4=po!~M0lTl{9wy+OAHQjv15+o7KM z@9?oF=>oF3)9Y|X8&KEanP0)D3w3s1@pEsjG3gEJ`kmVnHoxKS=K$U{>+7DaB2U4q zKZY8gqo34PhCjiU5wCv}et0-H~S&-hG*d zy2c%NXQbaEI{n5)kh|DLiEgj5Woej6ijR4xuo-KdGm?J-zYf&7PyT(w=AF+=1blkv zH!2_H2`c%UF;z};M7#Uv%tqMt*~gY5JNp#A2d>b^m>J{-aF1U=-7o)~UkAS)v3VNK z=n$Rf+;`LW%YI$6wGXh_>lAVq+9I}q*Z(DQQiL8l$dmI&%pCd+rR6Nz5Q6P4_r^2$xe-T*0;bJETQ-8#e8*|w;%NGqi3ZBo)fbV z%zaVOSSQjRIqe~Xj)~3MA-;gtYec?S3-`r6TEypuxFNcj?=hVxw`aut&4}qkJ?F20 zwPN0r8b40Mo~kqVtCIZ%&cgif;O%)6$X|j6NWnCioLh2iw!_`~1$tRw7wWs_S=lJV z+nhkf6}nS-YMetCK}sI;b=u4mB*xilr`_8=a%yZ5zAJ~=ThqI-hefcHME!{_*DOW-}1-gEbG1gt#*_wZR7-o)85#P8yNkL^qu z+Jv_7>!CNs*^jH-Lv$T%ThIae6X1F`!2!H^hR`M8EE>S|$?H4$Z{;=JGP=2%Y;J3w zRk|E%tn-R_{3gq9q1%Rc|C&(GrC5N!#AYv^w@Y+M`tUs&BfrGw6F%SYG5;LAIqloM z=%AZ|cQy(B8S%#HGmXxe#WLL6*w+R+`4;pL=%btw`?Mx|hqdJ;J;M82&~JyjyXz1W za}Ka?V2|=-e&jTVYhJw(f+SWnrM|oJuugjo+1I^OFzk_%F`ibR=I!pV13p}G~5$e4!AO78CjUSNB z*S!w+c@_UyQ{FtvC!hh2%WzhW=>ylC2y!)`(_p6^=HwhQc;lUg)cr^zO~d{Bu?%N@ z9%`P5q6l&3%ixU{A3>|)voz?8-2^fAj~d^h<{q+V#9jx`m@Cq0bd$ZO&N4m+Am)<# z+hG=+Z-Mj7@!tmpFrRD9)9~MrS4(CD@7t2Yca^aZp`M=<{d_z4#JCiB7bL_vj}|`G zwHNQ-5G=q?wX3^`>|T%7Lf;v(?^}WF{c^2R^#}0Q{Q;UoJuklD9bgXU(*wW58$a2< z>+oBpi!KM=8*7=%y|j-fz`R{-NyWPl_UVjVLldM0eIOU`lQ~02Un-RI<2nw2eRjbO_6)p2r>zAt z;NDF^6X-_!VP>x5B{&1-^S##p0{C0dIXIUF^cHGP_oV}pwXl|?9o{N)Ma;vyXU=98 zG(^cd$l<(&`?!j(2?oHto;m6Bc6ElKzXH#A2YLpq(FeXau4`PudD*vT{1ECqVy#nT z*JbV$zYVBw$U^&}XZLm=T!2Zx&@?3ne$yrKMc6angN)qv{|kHyCKTCz%xzw0wG6f2 zcVJNQ(@^V70qvmf(Ge?lG;U-xxR&fC3+K7Zgh z4`#5Pf<3Sc7C?($^)E|eitz6EHo4bA`#OG}sRU}yW2k3S-r7&fFyANef5NNVg*u;m z;-jDWA`ka({yl!)U+>2&c;lmA*Sb|5=@z!bnmxlm&YM4wc=@p9yzX`6?MMDs?BmMcJS*rXXA1iO zTZV1&20)Fmk8k*Rw;S*W_!zf=&V28XGi=5t#CT_B;I*a5`pUac@;&$ia8-tHp|(GP z>sf_1(3@iks{aBsO#`oQpIXjd{x@vS-+2pl_Ia;NOcQ$toq1v((hKzZ%AY{x57EV0 z+v5tdc{yABL z!N+|2@J}k+n|8Rj`t5@*c_%U0*MYq7uDOTD=whCKQ}pt;_?a)ppq>04-XmrYy>Gj7 zIsnemp3GaMA>JI_cHmuqjy^5I@0_cuwsZ4r-yzT8(}E6>6Zjtd?)|eT_hTO3xqpN2p_?zmZ2Xx{WL25 zE&Q<3xxPbud^=8|u4xHA1=i3vNy9yE!cS={LXUH0ux&%T_?1ivf|XF$96 z&vWhEo5(ZxyoT531Zp1*?7l;;P1&^?y8s>IXUu^*V6O?(b@}$XmOJ9y>kf1lI^j!t zD#CyJP;Ox>h;cu*zy(+W&b$kD+u^swF*FCB=P~j9l336ML!jy8s1O+U6ZlSAf=~g zB{FaxBWMGx0?+4u`{6E3W`S-FoxXLP;-6A`>5;!NVh{TzsK>}#%Qc_k6MgH{KHmRL z+$7Kc5M%s*5cBR)oO753AKPK=u3z6B;Mtfi!@Sm^PeGQ3e=pj@0BHXKIOjq8;s5si z0=|RJJ*%(pPkfz81G{zV7$Tdm?jL#c)O{oG`Plpa#C*XXeIs>^8S99->(n}}wf`P- zK+HlJYCVB|0%u?{lY;l3IM2;`JA4an$?*c-GyDd61jg{5MbAaOW)s~3xGTsE+IUhKbv-rk(CKF(tWOnlM# zroBMcZ_+pGt)bU%P^$e*Ccz&PK|1ej+5G}e$=r*~JXPpsiuj92%v4Bv({4YG00 zI>Da8t3PT7TMBgz&cy!BJ3=18??Dsn{te;XmmIx&xq{q9cE8-`J9zi21GPVE4#31l zj`TX*mwkABy*u_Q?y!y12YFnEJ=nrF3sUkukFOW}>r=i{<`_?}tuYG3NR?+Q4GwfsFOs}v_dWJ4y^RGxlKPSj}5#Hh+ z{?>Dz*3f4T*;x-k3O|A_LBB)oeHs4pI-IW**%{`=7Eo@zM1NcLzag&*%@8RQl zcCK%s&NIjEyI{YQvjlY(^JVxqhqX45?Q2|we?u-Ir@)x2HF#jmQxM5;_?$tz(63O}bXkJ5$=#PJ;J^SGHkB9ht7R>e|M0VL59CG_Pw2j`t~dV`PjET zPk|BfaosoA$K;9Yk}5~L^ea9i^mU)sBOmLG$y@h1CFV8ydaj#Cwd~1W>pt#fJ&*hB z@0hyJ0dn1^hfjmJ=<^x7Gp+Z1h|hEM(PtU{l79X~Z!YKVS&KeTvEP%!J&Con`}ZE3 zc`l$Wd{&6n=L<;FaGnp#FcWilcCF{y?CTD{?%G4$D{R{!_U1mvn$ne;Z67|ZVoPk( zpn)7?r9*0*;iulaWe$6<=j@_$&l>oRuo<@te}zuobr&hMki`h}!QEQ;H{J}sNfBz; zpJ%lTwT5Rh!LLBqN9X!o^A)-|0--p3TQK#t9Pi@-m7*n;=IZa@d%09_NgNW(g< zDU|bqp73!^#TtFW51<#w_Tntp;m7#iBd5@!9p-3VXF3NxDg!?Ye~YfQ#%$p`&;_V# z7y#$~5)6Q|*??}h=?g#e<-Eok*>(61oa&d76I%!P4kuD{8Gfl0o%8zu zd>fth4t$p!Ir2KN=Vhpv1J-su_M4HzIZVr=^Sx`8VGb{#SwSvzC-7bT=F~yoLsMd{ zvkSiry1=)1gl>$jhhG9e1O@zgJNyR7kq_a21=eyN_Tm|s1kRpG=rq20)uk6mZTjPHs&kk5ZzXII=>00RP|G-;UsMDSR=l2PCCY$84-zL0k zz6It=(olN@|Fj+E_%HZF@E-Y(ig!QFKMj0Og|8;y{=ug~}%9f|$oDM5_ud)woHT3_6 z-n?7*tpe>S@)^9jJqKfWeTs@2=UI#WELHj$Xacw9+;9G6(LZzFW$+hs#*t_vLxCP!rV>9?MdEQ~~6Wc*{ zX3Oy3;mz9t3H&BH-*RKEBkvmS!4z@_B-f#i?}AuIm!;tyGq>yhtPW}(b+gC?KIYnh zckP~0*JIrjTdoWWa8w5WZD=3dp?lH}cdl_A{`b$v=*RHR&iOchXTMlMTLS+DyZ|F| zxM$|`Z-~7p!YsSk#%Z{V?t6lMA3tZX03BfeK|Oc|UIEv%T!!`T!8>bxd^bE3ePU*y zuE&`j)@`!?^#_Fw~?MLL5m$Hy7=)*kw+=X;G!`5wD_)5HG~8go3R4Q$qSb^}n)75Oo_tm#~x)!*eS z(omyIy~Nu18lP6wd)>nJ5^A0nKEIdY9^2Caym2$wV%;wMEita61KbyFlQpDacAwFo z0rgF2?Cq9Z8M4x{)}9^$kXulVVwEA11re(t=xc~KA?9;hSIsUY~dH|Mn-h=c-6Uf7K=R2*hvgc(Fc^gZx&oYgza8$;OQ`Rf`Mbb%--3EC&N$YO)VHbbBkw*%yEF&! zebe42?-X!vqu)33ETH>KuHwP2Um?;*KIHQ+=85za-@oL-@1P9dfaoLr73_jf*y{eD zk+;y-eQ%KyP+*I`&nx~N`nvBr@&&dyL+9CGjR)v+Z~`W0q6ptr?}hJVl5(c-^Zw0) z9P_*VM+!Sgu_^}MCTx8S{(`j}UGbsgTTA^a|WJ>Z>iUpnySvYy`= z--H}LbzR_ltbZ4s-%)83nvi=D`VG5i0`HOEtWEU#{0JKDP-hF?b-CBh&vn_~KpC2Y z4eSN-JvMb2ymJ~uy|Y`|Q{v#~%J6MW;0xev)Q_My*Wtc8pAI^GoXHrRq07nf61o74 zwN@7@zmKj3b&k$RJ_pWbr3`g$u_YiSZh+4Mde>B>VV_r#$Hcg1=Q-&CyYtfT0_u#M z&n%s~sXGPjf?2Qw5@5d#&;{qn>005=->xS>U-xkt-aPV6_zXIk zZ*@*K+o8Ta8n*&G8#%eO#rNbEyScTEpgo|h{YM}z3hdpiAJ_5@y*-KV$m;8ewn(j6x6A)r%Q5jw|LWs6 z>9XeIeHlH<9s>_w0lOKw?60i zPWE3NKKd9ifj+M)n|zo2 z`h6$&q#o4Y0^8WM&%+P#k#`OH$V*$5>=}-_X2nNYs?Kwqk*gm68M*FP=eLNvB(4Fj z@Ebr~*9lm~whB$r8DnnGv-}ADE4=$UF8*3C*7ls}YmJr4FV>De(#VhSPssZM-{>ce zy!&Qvr}_~$##Z1TR z=U_`67!xys8fQ)IJ#f1gei!)fgr#=Kua7xAi!G zkL=GFu>+h*ivAAlBKM$&&{v=jQqV-F-aELC+`4|Ky`WDKa+qhc40mA}x`*ut;2K=x zE>!y}u%B7rn(vA8eZPax=y?F|Y@LDY$;fXHS7nH4q4Vtc*FvA**T?31tUExyLGK!W zs`N!l|LFIS=g~WhpP)O)Q_wN?5xjqs8Dl?1ANO_vTb?4LYk;T7uc7DA4!$|IU*M;} zE%H)3?61AwBPYN)ZK0R9{}=FGaE1I1>bhEGI7jAiw$8^M67+dm@z{EmzKPzs9YI}B zijVttfV=_RlMZ%gaRKrw$8YeiMZK^G*PUP=>w|3FCuP{LGpMqAGU*52`50?`=P?66 zgihuQbzhP+{04Gn2f%w_oeX{-Bt^Kt?%g5qyc~goID2{t?_8$KFw+Hief>K{XWcP= zc@fUpd*s+(5B)pj6x`CwGV~BE0oQK5bx;s{3H5sw=h1! zbgi|pR~dA%jUJo2OYB|bDd73_oIOY9IdQIi_zbG-{v99}@XjcKcEK;$JqNDC8BBu_ zx(jF%YR?6{Jz2XCtgWoAV=S2IO-7WczE`<2aEG|B&}lJD2T9nZ^$KJxT||A^lh*&3_x#>YO-Yy4{YppW*M zn$EMWzC*n{4Zj_nfAozr_=-;6vumGGKUL{-VvTV<+MJ2AEV7M=IV-!I1Ly9~zUIY4$M>ctQ6Hx>U4 zeryf=T3{M|oYgnv47>U5cLj{_ZKAjCYhc}{z&Ov-b8w4KtTn)Q8J%aq-nvlN>iUe? zDZ`rj*p85`H34KES$zv1>s=t5SG()d-o-wPZ3SN4A5wG)x^}SQ-GyJS>{rCt(>%5fboMow0duoIXJ#(@{DMycFZST$`Z8?J*;!13 zKd^abiZs0CuD=Nu)ZyEO{}ZeNbJ)Lgvj^|eGpIc~^D*)bd;^~z{0OR_xtx#ZV+j2U zJp`{nR)q6!?$2eI?MLVgvN^`|8#^_-La-;vvx0XBQ`4OOph4xPO9lv|Y??LBh1 zu5ZvOXdj;|`0t<~&ndFB?!S*tdxC!lc*jnmx9F@P{Bs`i%b*GGo-D#&LjMC=q*Z_T z>eoSkS%#V?$W8K1dPg^hya|*)A$uPZ?C$TR7xdPm~9lcw=*TXF2H+ zoBJy7Jp1s@Z?>|hMd*8s-hTcU(EoW!F8F)+r|`RA3i%G%9PYDwU_butU^8w6R^XkP z`4fa}Pi=9YI~J?E}_27lj1d!6*oTHL!UlF%p_kYTa(X6~%kbaE`W1C% z=066`_Z*mO3>qadwZ5Jp-=L5EMCv|I^4ENQPn}WpiPYKJPd)F1N9Q*>`bm4#a&|?{ ze?ZPO4ez)<&S#AL1N5*YFEOv7&cd_PsAAn4{jHVb<6e8`l;@ydDn5fc-xa95cH=yw zcjS_HWbyKl$~W-dLGzhw z40Y}Loj~jBX^}Vf>DnT{jqg`rzIEtP#h+BP5A9TZjFCqEpJn)lMcEt!YPlZYw4d6Fl-5zjOuG3s|&}ER5$DWSi-D_)_XBB>Q9pp`T`_#5wQTYvc z_hkrN>kQO>-AnI*XSo5rM79UvTPjtzTj`X42U!~KmNPBD{WUqDejBXkIU2zqfPSTW zS5fWOlg_uvPh1n5^YS}k-Mk2Q+5T36b)|h^Zui9bsJH%68G2B5CZCYa?>ggo)#jWo zK?7d9x=5$-jeHBA*VqoxeIuWD-kw|V?umSy&q@{d75`i8F~k?vpn6c?-zLIO(_uelOwmk2ZPxZ=&}MeJ&qjci|T* z`3*LGGvq9Vhbph2cg_dEoX$WxU&*mfq{a-fjquroDo2~N&MzX{mv2Gcug-f{Ps;G# z^x(}aJwrFDX7UW){qRhwFUZ$`s$W$`f6)%LozFQKvyKv*wZuim|AZ~}8R;oH^F-fB zzoPT}N548v=^^^nse5toH#M^Kzt{LD`E&Z*sQiDcsAtf#v4h?7Fb(D^Ox`w1J5%TH zo9FzJl>dh=BR2Nwce6*H1?|-wBK7=7zdG&yjju7gpk~+jU zq07J=vlW%^z%K#MNZw{%;5{%;#6G-d%vrg{Nl(Pq=>YvQKIYzm#@IS_-FMVk!RHtl z|GA>?q4M6_5q`$$e+&HrT#tRY7VG?4hWqTCr$JVb8~q-dif_4OYXT1)h?IZd#P@vlc-+;Ch#GI!!KI-P{I(qQi>Aa6C{R?FCxDUp} z-v{#h=(U|=vyMJf;EMPmRNEdj){>s1yQlvf^clSODB3@v+d%gYMBM;idEXk(ptJj} z3}-6-p3L#FmVT4jB4;T*6%YDZ?9u0lc=;i@jOoK~fvF30c*zj zTl1YfX&QWtIR?+blM<&YUTSXb5vTBzTs5CvVxm3L)kk^0{Eg2Nx$3n##!&C7R`VbD zbm?^i+gtFvKGe46Ui+_gH;@OFJ?3>@KRxn`vx+{_)ki+QH+RI=ecZQc^uO19SKxhj zqu(ifzV^_UGo3>B8A$c>-2RMy2%ML9$nzd^jNlu!9QqW_xlVsWKUkw>`uw|izs>6ZZm;Wk z^8X?JC9&?c=ihla?|NLE>mK?|{EopbSOnH>61M<#U*#{!u?4TrJl;RiQ|50OTLUza z8|u=qr#Z1R$kJKp4KcUCnfqs7$}geL*)!{Y4nP--Q)-mq8+nDz^SKTdfq8u!^mjci zc>C*s7r-;6PY>Dv?$@L){d*S%<-_~zdCQPp#}sG*XEDIf`8gA3RN$xX8`L^az)Xoh zI_F`3=Caovc>B5pd!PZ#uaD<02i}Pz=$II5h^%0Itb0nH_ise545TIb=-s_|jqMlY zMerPWM^4+J2hZ{jaPL1u?bF$Of_k6(z&X#CA)oQvpo1>fNbosCHw%&?>|2_K{^qgS z=W0dezmwlNWl&??&v@SA`;ue-5&dLc_!%ecQ|b(Mp!RC)c^U4yws-KZ**(zq0qVTv zwL9y{yenB+TLsXS_1C9IeO3`(BHc{g4SczZ353(%qLBY@b3@#r&UahdxPxT zV?SvXSEQkSgFeRa&35>v8#7(f4}SJhuW^G;L|x1;jlA=GiF{ebuEKkE6L0|^{iIj5 zT;`m@N4@o94XM4wxa7LxQ~HMgM-`)A55FrQKfvZ0Fz#o3hRECSo*&Pz>-KE)fU}d| z#de0zbLcJhSo0M+V>Y1M*jn%%_e(Cvdv<*@&?F7_az)*BsC|p=9$wkn z9q2rEW5@VUmEm6H%Gll^&tjhjJ!IF>0p{qUGky-)e4Zeznya&h)&;uLj%)n%M3HqT;PV7x&=huL4q5BA$U60mj&OYc2eH#rpb7mp~6ZE5n|+ zR(;*~0oX)uAJ){zb-5qXO{n^aKK?^=)>U^^hWmJm>>I0I`0mYME7Gw48|W^Ojd3o~ z?;L&1{jsKx`J?Rz{9=q}Cgxfs<{dV_iwD5<8}kZ%l7<-9d4&8LEVhI01@sg-=8w6* zqEE^1?>lX-Zx)^Q5pchE(M{GwQkH^U#oj-&%SvLe1=v07fb__ z(QDfV`rF?Od;&awcffV6RB{Xc0G|!80up5RViw-H&43(ThVC1%)(ARI1LoV|&9vqQ za7H=Q-n5O7J#)^x0gQKk=91dCZ=v;^X$x5IE7WsiymtHe26YWx_<2y!gTB^21#_Ug z7T%O6@T=ee90O}Ff()z!^ZL6_T~3~^GSuG`?$06AvpCfb=h8m2BK+PML#MGl2iw4& zd{>s>Cue&t{9bl_C+%?0HlWUA3Ai_1;M?h5-{aQ+t|L$3%P_|TTNi#8xXuCS!%JtN z1=s^KZDOIP;09X<*}m7oGP(rnes{1XYoV6C9EviW5qnh^@e_L2oI~}C{=PlxMOx{d zwSB0wr=ke&WPyGTS(|k(AJuTa#yG=4`=Pe$jy0q$`gVRJ;CF8FhhE!l8D<&#_MW+a z-$9PAGu5W=0AFW)f^1Ix)W1hB9jD9&-D~Xn4DfYc19Yy*+|JQ_e}aoL#C?Fq`m^X; zmEN2wvA!SL{{b46EkoadH?J|TfIj22`fd}uQ$F;stxrsx*$}>K4)of71l=<9W8Ega z`iM7`&HYg4jLp9XpHp)H?-_8-OX!?mf?l5kS@z_-cV>eG~50>2BePzFA~4)2Zr?%7jp9r#!9`ni82e11f} zf`1D2UBAI61t-YPy(!w^f4`<~9_V`vMuHp}y(R5%k3XOrqk9i^KdiqCe+l0K&OzUK zumisW%;VX(1>5M|6a6#vhsgKv%g`=*=P{WHwt4Uz48UZ*%weic|MV06T+e%S8~7Xo z&zbmN@aA_P_Sz5rF>VH({Y3vd?GP9J9@F%HhNfe(t5~i+y-LT!`jM(=ssp-#PalxXo3V}>4UC^UB66dz?!zAd?}+_7XV1?eylb4o=33+0rx-oo=bw1n3(f1qj^Ga^k{Om8rbkPs4!+ZN{E!H>QJR87$js83EIsUOmx5h_Z z@;5qX(ORqCVrFBW=r8p={9MzWpsU-hALnJB|HRkzMW0LjeeYAKvx)sm_1Evd(l=^x z&Nt>*$4C5LhMC`0etl%~yu!8%JU1KY+#lvM}OMLq)hscQlKJQpXx^V0#^eg)DE^dEqJLunE8U2@D~_pD_pYWQQYiR`)Z8|%KB zTiJPzp?-U`nOlGN(A=N!>%pu41*nVo8Ckus&nEGc_Yb*GOaq<1`%q`6UsC02qH}NE z$GF~@w^RAV+;{jy+Y~WFa7g^BfRBFT;=xaU*AQ(|*AeqbTgV@3@t4SZI%h3CCY5EXl_B z<7N08=q&Ue`wfschwHh*_B+(Nu4%R8{y>+2{ksRQGsbn0{To1?V-Gapcc98=!1qSG z{4(gmr)_W@?u~0QZwsAu-J=W8DS5Y{1JDH1Uukfjz9#YsvNPR;wm<@WYY!^9fZwik_GjPb^4u@N z&jDwY(v$nKh|Tk1U3Fiu`P=Fe`2g6XegonM@Eza`Q*7Vi?R(Q0c<1W;`pEJ*ap&;) zS~A&x`*a@b=u+$>_$&=_20x45?`jvD^%_up?xFs9KIIqjX~8=u*Vu!%-VypI$TvXS zKD3EG`oT?j&%JB9#C8ntY`iPxlDejO_!xJKZWd^dx+Xr6Z;>af?0e9tFX4YACdVFi z(gNLYpssVzR%^PAGQ9uwny&p6eO>3;V;+0_RN4PG)Y{8H`wp~E{8mNnxe32thW$(7 z_4B@%KlXBhUD>^P1%FY=_BT`U#uaJR7rNNbSv&A4J=W)?-TWKc`TGTGZ|+}Gggl?* z@r!GXwcM*7z6o?d+!TJN;0Le@7Te+XopJvg?4X6$=+}7U^A4ZhBOmYm75bRVJ#nqZdp@LnsQTzr!g~j|D!V?ep&-t8I{NkC@9Dui z^E1#-^w6DP`v9V^a~)T8qu&rY+9UOhMZ45lSSQ-04e%9x-LLz|N58sVK7Ca06rX3c zJe%^7HN2+%vUH^Ii)kmi@|jo6uJS-+a$j^e4mwlm&WiTu#0}OzvMTs-+={mN5C`d`Q1dX zUjj9UeS1IiD#m>=Zy)`Cp|^MA?4gUz9-H7Ldfz+Gb5?{pJLtENjd9P-a}yU6A)@EfA{yzis)3?3p+?mf0w z@Kfl`nIhX~4o#7rulzDPXX9M#{TSKVZX;X!8$PaK3E6p!k?mI;;QJ1}{kpHN?Jc@< zkP|zP+{N|@s;~VR=enc=^8Jj>{w}~-rT+}=U^BNi>w7+$mHkKP?RBU#i`_o0ouPA% z?qz{)6`g%JpC8a$(-~NE8k_*{yuR8J;$jWgG}K;(9uj;O;q`O>SKx(Qp~QW@6l*?0nPe=qS#;K%SM(7B3lptp}J zznVYi%&CzR1!1Fc~z4P`Qct>{8Sx5bA_&aRxK!dysP~QP-u1eeCymz53 z;N0{pvGu_nd*#)a;G3L*q%kA)H*-rz$3;l*|2VeIG42X#9@o$T`lV&S z5dIRoIS)VA-)9{XZhMbi> z{Q$M@0@OZT`;#*4k+JvKoQ?f{1Y>fz4rl1MBS&tQ;rrtGY69ozyv(oP7;5iR*i!lG zo|!WvzQDE$Z|~M}jR(jDvF0xH0hh=}$O&|@9jLQt!D}KlW{m^YN~GKVJa%@uw>878-L&V-4vSHbbcC-p zJ?Br#uuk{eImdOZBCi1DmXtis>>EDrVP1sZH`?L8pF=Zj-q%S_*jvaq=vTokm;-wdQ2S|=;oNLPXRxdDF6|=Qk2UYtLeG=mobaCAJ#FNX_Y6NpmYRPKU5=k| z>JxO^=x(8|LJ@_O5PK6r^@)>p}PT_$YX3Dv6-iXd=4(LnM2w{ zZ=5j=_yX#@osVmEf6eLooQtzIpZVqWTdedipwB>7hM9T)E|3#?h`z2>-nCkPSjpxY z)N<<6N9Vj^tn?NiedQbQ&MDe+c;`F^qEC7q_Qmz9%c0RfO~X9jpjX#}u2g(d2LGGN z{}uXom2Hx{;=iqhxvIOw*R{vm&elAx`89}jTUAc|iX!OEeFesqErHkWT8y)P3&b4K z6XaOyqGlV_@{h}Kcl7J6J;Wa)f5v7n8}RmM&MtHlz2_kV7xMUcHYRxsWOU;+oE@pN z_58ZtU2KUmey8xp&a{KSXG3`!z5BR;>>1cVww`Za6S%M1lyAj#c;BYr``GN&bLzZC zm9KZi-j{%DlIowr7vM=5@|f52+(pjl{go7a1bg5foAq3i^GGVYzay33Atz~Ak8+0J z&02Ujy3k|&(YdY+zb3TN4sVP(oPm2Wv14;?${lDA`VuUHRiJ!W@hg=)q4+unWAEj! z=@*((&-(V97R&*91ZIJC524!5fiq6POZX|U477Ja(x!K8_LN^UXLR=BdZ*#-cYw{j z@-xs5Gz0pqBmV|}0agAA-+{Lu{nbguN=>(l{0p-7aZ#Nw>iBQ^4EcBL>SL^ahv*6* zFAi#atkp*+pQjJ8^JVx==(;lyotH|`aX{!~$C^0=1kQ)fS(zaD)X;|!%M z;JysJb8%(|V9*ZpH?{?@O@C!+LHxoaU9^v>lL&kp_{@Qi#KhR>Cynn!Ox}l0P+(W*x}GB&a{_fe`@nDYk$f5M*?Xw#%F~D67|nKwGgd$IiBC2C zKhW=1dgt9^T~p`^;F(L2hv;2L2`1;9eCFIljyXLa>O?$S+WrabJKEs>9pgI__5Z>r z=31;`UZXQt4>0jQisn4K`%{%8gY=JSclW!F&&;B+#*Qj5s4F8=Zh0j0_q?Nr{ z*+%G+>u|PH;+*F$KAYez@V*@aeLOS!CB4J%Kx4n2A7%Tz0M2ZP?=;kY;#wkgCRf0@ zY?R^Onlaa7`kh+N+xqi=lPl(wHpwTB$muuZaV}-=ZbmL=Gt&;eX3%<`$UC#T%XMPzt7w_$@zG;;gxm$+!+Zk>H_eHz4jfrT<>)U1x z*FkUY9{R~Uj?QzT{{TPv-=IzO@1gp|3M{|M!IG*C6)lk#C27 z-lLB`&)_}ZJ^Z#RUuUy}ejfY|dcZrfRPw)f1n1xzF&%o2d9L7-;-N2nX24bZ!AIXF zy2riWJo0gN-A8`b9+N-Ti*)mmpSiW|f#~xBeigqn5Nj*DUq4{`6?m^_foIjd{C||Z zO=u+NzNZ;5g3&^YTxcN}En=aC7J|;1tM1l-U|t1c5VRJ#&_WDapV4=~yhHrl&v(`DIlJ5o==AvobUk$PD|D~1 zcfS8E{kZRsW8KL**dx}OKY@>9y@NOOw443wnq7___DANO;`4FbpV61-PoQVSI+HPP zL5}_t=n%RBO{(ZOBZqlLpoh(Q`|zvKYv68@5Z@(R8GQlw?7bU;gqR*arGA{(RmFQ0 zJOcsv&N$5CbGQucX1IX=3M_#A>i5`>=P`F#{vOlCO6@Dh_ObsXx?jolM4Wd0mVngR zteO0$!GHMg7kmHD%l|uhKmvpR@ZT-)E?-3YZ$AG+^6!(s|3CltlmGaC{U874`0vmD z{@?u1oBww6@BXL1+wA|l{^tMq@BY{9?~?!R|Nkf6&5!#&uP98v%?|0)XQKd9c+c=X z_8ImebPued^Es}-0R04=`CdR%V%+}*2wBv7rsmPdyW>85m#GgS{x14J+#B@A=zLDJ zHGN^Xwy^=96@0F+d1nK>XXstH1V2>q?=0W3>+sKjy!Yb_z9yIFxeUJrnw;1Qu!uau zcLvleOMip+n(R~4K z;q74s-rQH<1u&1_6ZZZIwk7Zl*uZ9vEilJ^3ia-};~w-I<@?y1J<7X>1N4hvA0*(b zD(=J{yuX{kobLHAP z-ri;Ld9we`zoX{tc&Xp9&fz_s5;w#?2j;y4Mo#*S$jwEo9PSs z==OkT;!dpd0!UM2??nOzRqW|Kv>g zs(AO#p~ql~&!^~};|%#Vd^gW3{4RLJXCIsK=CRH;vVL_LbG&Tz_P2rUN6-WAxy$Ew z;aZD6Ii0nKdtS=gu3qmxzS}0)?9=CH5qt?e?|v&Amm{wr&*7J|IGZ!L zzZz=Z0PesD=r;lOSp+Dto}F{tpf~>{i|>USsQwA`5c?{yM`OKv_SxObO>CZNlN)Nk z+rSw8Z(2GepE1theDan8nj(8&og+u*z4#4Ugm`ZH`~yDH?o8x8cjQJ~Z~xZr!FRn&|A77w{T%!eIClvy(En`~g>mw`t*RiXefLV;_tbsdieT2?x>F}j4#W0 zFWql}-G1yv-yvAf;`g9;$9>Fp;ymucT1V(}awqtDr*h;Py*s#oKLGAZzn~vAy>pAm zo|pCQ#doN-6KtFCYtR~aPBU_uuO?pmwl;X*!3}wG4|-$ z51?N`?brN8MUFDgJG!SQ>?ts22y*Z%ezW`N@B7^Qbqcl5WqGgX$7i9S$kI%5yIjJ=OvfzNyR*B}Y)GpP?YV-BDLLg^5~#D9&zjO^pV)4Qy9M@BBRh|MyYnQ(|DDDRoiPJ^PS9@x-NmuXd5QarWmKcs3r%uGU)O@pkGtQ=1$a`+Zyh21j-QmxIky`*3veGcgDLN`hxb*_28vBwiG1r0l52a z9=``pfX|8dHi3o^cjR;N3V3()AHqLo@S=?Gr^g+5J}I_t=k})0khnf}b6;UE;1|&E z!rQk#DKPg4TV6(f6siv(_P=Nx$_e(FJTnu#WB!tbjn=GCJ?p9n`ut@D81!o5I_hJ1od4&X8y7 zs`mdLmmWVso)BZc-_YCJ2|DM$f%oG(_}0A8A3=9Vqw(@s@gjVzl}bJPVVD<(Z7Iq7J1=aTfpxe*xzXt|8MJW;JrsF2>3rjhu|IdqKx%c z@hJm+Vt1w<^a&s1hRDvf22S8_!Ph|h8)#QQ&EWme%K*K7ylCysU58%-cR=4U&`*Es z@4^2FHozw!A^!tdmdEF-ES`n^Deo!Emofetz7FxZ`3~D9_#BML(cfVv_!L^}Q^j4H zM=1M@lt4L`FXP=l!*&Wva0r@b2A@Uv6|ex@#TL|fcQOLjT*%t9hE9>aOB*18cRp*` z;}rB-8Ex}@sxPPoMD|A9K8e)&X0~r}#Uk zv_M~ifX#WD-q5cg+k^K^{T!XOycefX&!>mJ*&#mm*znM)eA)gzKubC0^5d-xc0hL5^Aetpo~nYOOE z%(;wDiOuuV-~A5Zt$&oo^Y{tc^on1Ntq8oE(Om%JhUmovbUo}>u|I1J;a9-~f9>vV ziavk@-dgIFH<8EparQC(_SpwMyGz(ge6OJ!*mCp*Sb(2EOVHd2>@PrpuV-hx^WQF*E=2wfdf!*-o%iR~&)$9Ze3ll`dETDiYjidAtv2-b z<-K|Zrs%sG1$yHK>OdXh{d8~JK)bY;#U2anUm~aX@%cNDCvFRS3VbHb?Mw;$1e?AW zt-hm0r-CJ?)(XM&qN>hQG|GQ%AQZs+PXS9y1I;KR^z*! z#dH1&y?fd~wm<76_!ZEumUWt4L6`9RD`?J&JjNa(7hnkQ{;fH#m;wI<^oQ`CueO?) z3v~WgHtzxa2dHQAwu*h-;g`aXK!JZ=#u-Bv{|%~{8J(XAJ~zbcr{4W~hKu-`r;nV# zkMYrNjyrVjqmTV9va>jc_0NF4PX#(-ma{n971Z25$4yVrP{#i6?(qdZZTR^dy@xup zHqXc&>JV$$U#Sh+e5j+Y=`lb{azl^MdABY>LHtk9Mc_=Q;0)OJB{U~rbKl5iym41R zzbUxIZvWPB-Wr|$&7NBO0d#mDd!FO7RK;g^gwDMfEAl`;*v&TuHE8sl>qE;coBsuI z-UI9P?uo_6Jx%l@&m%hHy(7-&&dmD(nB#MFMHYY8N|%5=hSsiZpL;;Naee3}vhkv? zZyEQo`;p$>Pmt~B%ZxrkJkRMpJpj+Q6!`69A41)iIW~|tphv*Z1ph7gRS?4Oe}m7^ z+e4nk*axUS3Fwi_*-LzG^&>7Xx%=|a64_YmxTpUN&Cz-8?#TY_(a$xu1>ii3*o8Ut zyMVWk0{IcS4~h`)s9Py4_b#fm-{cv zc+U1Yy^r_V{GQ)9i}5d^6JU?tfv27L9P57rZ_WbV&j7v*QQn5vzu9d|<<09kCFthh z0a)`h{Ex9U{ydNF-5q%!)%CGAa|Cqws}t6A_ulInSVlJg7P00_;R|ByyLbQlcapL0 z=UEkNSZ@M8fM3yT{{owHsDF;me!d19-~i~8llv0>8*mGjfcp&S?WerQ57_@dfHUmf zcci($2CG1y*U;T8ot+__<&uTxySG$s59E*5V^+x0{$ALEoSJwQ|?A;jR3!f z{Rot;-o2VH$>L6|WsRToLpQfKGCdD%X{ z+V5je+gaR=F*R7hcNN_TY{T120`-_zy43y z?Zuf6h-=OYyXUfxJjLfv$Uavox&mKy3CPi}ffD^1z6Cm;ckkj9xvvh}0&*VWU9%ta zdJb2>b90WAm01w+x&;5A@sG^LzL*#--SdZ~B0G2lO+a_r-s6cnb!|htL3BCFTfNBSq)F zLumhwLuZe>-~tTpnLmp=>%m{4n?t`sXK$&oAOY9N-t9S9g0G>o`}p4r5AgRMjX|KV zfW3R~J>s*4-8nqpTYR)xe*#|uYcIDxzoEBYiX70NwR+uc7yT_^xsn>V-W# zM{kejDFbH$Z=M3)`4ed8FSVB;zROv>mj(7~c=L3AIkGkT)SE$HV9W6{E+Aime?-^$ z5AmB~b6?K=alZrP1ph@)zz>Pr1@5JXPI`-e0e%kku5|Ob&z!n`SBkI`_fVosv2EjP zkJh!9Q{;Vc3$%R$671^So%MvhYF`5>ygSXYuVULnXHQ?ick}uD`gwrPTG)$uCiu_p zV-ELTm+{{K4&bfdotZVJ_?go^I@=g{u6OFu^?=Xb1UgcN&qF**Yt4|&?>*Okp0O`5 zypKJY=QTb{z@0wf-8Gw92frcnEA?kRPh!TxLX6Z8qx zbCd4@d$YE=dSDCNYp@U8sry(5{}tK#DL(Gj-U6|2&==?zz%BX}WY3|G-C6WC#{7%u z3#c}Gw61wCpzF|pt|aHz@W)^R_P~IcvbA;RdtBiUl--HFcrGb^=5*iBpq>B6RQo=@ z2|fk>zGLj&o|fVBfQ&r_UEFhc`}M5$z%N<+w*l?$O=^tPnceROKFS}EM__<12li$@ z_1Zg};(re8&7D2rdyQ;NUPisMcK-IiiTxVg5_AgI;GHEG@Ck7v^gbu{xs7fH9U$v# zE_b;LdiU-3W6OU)w(cBi+;$d!BN|hqa~^YP8-WGz1(>5Ve;wkp)b*RK_1`d zUi7Z1{5Xt3f>%2}j?WYmSjUbM>p} zrxiRkj^XX2nH~BWdTSj?u`8$OjTu4RTYv`W9ykL&hjoa&zQ&8svH3ZH+K0P6$JckW zegSz1y}@3G_`H|M_SNN(_V7#4JG(n{t`h$>sC}7#0h*xe;r9dbkHEZBVm#wHdh-s@ zJ)m<});fe*!!zBJZ|%+=fcLP5YI6qbB_ZxqYTb=0J|F8)eGj0{YK|Hob>lYA3vAXq z&FDcJ=wmxYzm87%4g4Hk3PwBF(T_o{AM_h=17G9oUHSz%0ZmPEyQ@|7yC4N)o7n)%gJ zJ!}QM=b>F&4(}}1n_?4($SE{nvv==H01xPDsAoBVcmC!akdNSfzTV<*J$F7qKLzL5 z+^fAVYDaF)3;qT_>l7hA|K>N(_xL#PJ}9wUa{=Bym!R&YAV!^b_2$vf{O(AdGxy22 zg6#Kc9^!ZECA_`wqOZ{{K~vzfS(kBEXC8tlkXG@XJ%ZQnXTB5n?fD0=hEGC{89q&2 zsIm4T>}`PEI?7F48Q*>8w1*t}8MNL3&;pu*(~3D-`2y;0&eUOZ4nOX^nFCse`1g^r zKE~M76)~T{cRiWYGkL(KP9N!W{M2uOXYdDLf^83a2%7jT_P&7ZO!iU0dxrjg@f@$w z*CGDR;XU&lq|++S@ALDBzw?x3tmn?h$lbmC0bZXCXict!yuFO~9(aI1h5Gkh0QTVS zzOC?Mj;f{WP=9M!{}KCl$P=hO)>%VNO8N)C!JjF&duV2b_bih8_$=AOExfg~FXNvF z{#_}H?@H^quM~LBzK^>+r|^ESEaUSAI>N{Pobxp>#(O!3_JA|pfdb?^@%fy=+qXU< zfK_Dgk9+hywk!H;<$d@wbl&5p4~RW%0`EuHn{%b$4tOR>8PCLbu6M+pC6K1c7~yuaqmK{Eq=JhuE3v@GKta*jcp&z)^m z(SIAdjhuq!?_1;%2*i$|_EW>V;}6gwI_vlzdIg^cdctR^)z9z?z`Ea}?|Lfm^W4NH zzWPa>$2v!#4+gD|@y_u+i@g=_%kYobYjU08`yH}7{0(&TkJ0Ue9{!%i5L)7Y1YeUg zuadv@WN#bjd*~j>`!)OsxO3-q@A`a>JwaCxZ=deTyQrUMZcmE!teZ-}qCE}(bl?0pqX;G3Oe(^t8xo436A{uzWo zZ}*^#&;2Dj@2_Vv2gU}dXY(_#hW!Pw3!bofNAw#&J#YQo_Y9q}CAqh<-+QpPACTYR zlVW!-=B??)K4-?@bA&E|2J{zIthotI!AchGyHNN3DRdvatKwc~&{bgW3DmsSHGTvg z0%vj#d$iYn7QZJ($okwM8*l%{zk=7tJNX>y&fV1l{Fh)I=ywI(ukba;eLQRLi@Se~ z-aBy!4S{=)-5J#x>m9z%*gv)&d>{J}^Z;ECd;zTK#~tXu0X5e8`z@^PbGU{)BuBrB zbP8Xf*I$@xg08u%`hcX2wi>G59?cVo%h9dEdw!nT7Vr$sB|WR+`=ZH%dKtU6)7JMF zsI%-<(eHDpJp`!Vi6M(I`u_~RL-!*vhqeVUp$}!@EU#L7-u7V+hxlDUy^qddK6mVU zW`y3}#WB2jdgy!bIoJgsh*AF~dDMRu*MnGSkL$w{2|b8j{$o(lXQyCT^1qg7vZh*3ViIhhJJ*vcV7J*-h1YK2>9Ap z=kK06e!taQ*ZK>{?t1`2pcZ=fSVM=?avvlQ$m|(7`u1l4RQLV*a~CHD9`ZIHh?Z-o8eny zcee?;TjW*rK4#x0=dj;DB-74@deE|B|#{#rO9zvgiYjvOq@qPFye3Pe) z&-IYJOI6I@oGE_#HvORg1@yCc7ZUQ8#Mz^JYy9c)5ZT(B&_4PHu#YVN8r=cB&xyTt zGwj2gtCOAGUJ9VjJ2-(q`6y-+ALo4mHP$+3@SZ`9{7LouSOV| zuh5=$Y+X+G(*s9Y{ND0;Sio-)7%%@07%y)P=NJHc?(#@y*c*zzXl>W%n)Ac|J!tA< zyWNTJ@(DDj-zE=w=aye0&U;pa5&BK!0KW{i7yG>c?$x?)K#ty6_v+cn7iFX~Vjs}y z@1Bic16LI_LfnD=+sGr}`RV6=JP)6PE%eTAPs#y#1N6}u`>%lB*-Ly=s5WCiAiGUyz632Ae+WguA)}XXu=#)30Wn z0ea7=59Yw#4tC=2#(=MLTgx8Zy|Ekkd_}DM6uJgR_$dddXHv8=BlsCQ=Q*un&(?4U zHCTcl;$!X?&^rM?1_NwgfUe&>#6Fcv^xF48UXlZU=ddsDsApZ{zYae^-{rVO?)bAT zKEuj?LVwZOi1RE0)VTtBchSX3y`%2w8k@hBO6=D3e6_!W&XL`{z1c?>Z-0;Idmsh& z9m4PD_C5Ajt^af2es`~`T2%Mjl+_GNAFh3_8g_u*4` zd(+-SXI*n7_wm11A7a>^(AzR zzeXQ{U`B)vi zs6&kHc>D1jk}U4XUVI-0^aJ=CaEi?xn|}jX(=#j3Ig4lJ=LdMtZUWZPc?TBIc@God zxp~g!wSW1~GxC59VBW4zXB}d9ma%$1&Tb$2JYjQI&wGxp4p9Bj{UwVv3UoEP1LW7h zXZa2AyVrT0=g(k(PXY2W-plC@wUJY3ANwV6mK0d~9C-`;*DRi?Gy2||VP8e(T}wi| zYu+V4)-cwcnadp3RF}gq0?%O|fAcjtuSUc}A>z^KO5Oeifa$-FY(?v=7`z0zJU)?CB2u z0)5n#@B#Y>zd7^-YMgmy@cN18@PjPYHs2I1w7xmKJ6i-pc>UeuDVQNILVG)L2J_^YpRj0U00rwFiCsElsdLVDo%ZsB_P;c?V8`b56kw-Jg&Xa`o}C?*Z6E=f3x$ z-5mDU;aL^WG!SxO$MBx3dH3O+B?n!9%kUc+I^vd~?o;2DDxQ;m6XfRm51sS&$X(!{l=1$# zi|6P^=u@x&e~9l-psOR@M!y35oB~;$|2I!sFy0`G!@(|yf?#_AKx#!ylpCFg`ncI8k8Lz^7M()=$UPD%IjXARV zW8=tMw6b**AjBOA8yp`V~ zyW=I0;^!T4kM763HT+=~&&!_me}by-YJZPz1o9AT?ZPMUIdqC&DbFrOJIt5G=b;A; z=E0`^2?TiezYpALQpVpg>PvWQu3)on4*vju2DY;Jyt<<~d2(cTuHF4@!YAtCQ*`UV zkFiVGCfJIA0{Ir)0pmOi`v_$`&*vFA;GbddX76agwhY|ix6n;|j)3nW-@WVFkbeNi zEVfj>kI#fU8|aR+`1|%fy!tDseY*32+}tbjnd>x*&!{;(H*1xFcN%(Ubr;&C#+rK( z4B%Iw+QdQ?@4zzD9VghG!y3{O-4*`ctsMCV*?#u0C(z%dZmyY>EvZ525CKKijS_;x)@b8Pna96G@64nBZ2{O!|cZ3L$1 zjD3Lj<2fAwpI!IaD1Zv~U_I*}GI>k4|b_s^SGic7x zJAtpf=N8EK0(t_oPb6r+(H}+eC96x(~4%}Pkw+V0TD=C=Z ztGy3Z?);YFhrpWNiN!2_e<&ZqC+HG*@3sBjfHm+I{VDKn`+FnDK7=}(bK9FcNulPL zz^k{u&#`yOe%x77#+`1X|A1`WGQ>XhcQ32>egfRBKIRL^_TVfh&@1o_c*>%0L5?Hj z7eJpWF?;Ymunm0Q{298N#rt^zHLtthML$CB_Vk2Yf)(%@eTvU>XrM-dZUNY5g3f)6 z@8PLAMDGmNDxlUi_6>A|{0Q!lw~(7!$ZHuq@&-uZ*MaZyD`a;V%II$|?m1xV1Lt&y z&gp)e9?%u=d+0-m{aa^1T#a0lOaB@2gMRoY==Slq_A5}RBc_=LxsRW7nK#GRneyQuU}Qg=lWlu z-bH(V2XC$degW!xCWlYR^-~3H_h!D1_uNnMGj0lh3H82eH+D+g5O~%DXbA9a-|9Bi z-N!R>hp&+PS$sCNJp;Gs&3_ATj+z{wLlbC^8r})%W`>T>6o2dAqI1Wd!CbWZRs7uP zAv*Wi&Eh*f;II56djFR2zKrl40(auQaF37Z>z()vo}+t@Y@ZKM`|rWqvvK}r@N?G6k0y(->&bK1`qK4Wl+zQ*ounsb1@04Z=LcUoXS$mjvxI`G|R z-q+|7czuuM$-CZC^PPYZ@GjZgPuQGc0iC{{=SIsbzlJxi$(6->FiChox2UXnv4OBURTWlphgN)wM zy8~yjr?QMY2-tIQhJ6E=`xj_)Pw`V;%HsEdv1jnk;F%U>w7a(}NvbYobbAL5C^mlIi zvi=iv19}47UpMm)RXoQ7s5{eU|26)1K&tOM?CZeue1qLu-r)r5jC0WCbe1DfhxmT^ zwugbnyfD5%xW7Gx+Ac&pTTCBYO83uyy|0y@ONq&(VL`+VmTM z39wFA!`zi2AI#8uHqK^#`?!Mt6u5^T^6Z`-pna$_ zc%D9I^7a;>`mPak3Vs0H4&6nKuXoa32GA0mp})J2exAKM{1ezjZ+`cxuQjwCBb(D+ za-iLr{hQF3iDqvaY@*zRqOtN5B~O*7;k*9yY;` zAO%nOX#YzUpV>M7?)4Ga`zk)#Yvc)7L0-hyTwUHivUhy}c&-DeeplcTIU!Fs?-IIz zJW)^HHDv1;Z|~OFg&$P$E}6?5b-)iSpm(km+C5KFd+MQogx-OX$m09)hIroz?sN#R z{}Q+a+Wrf+1lu{XJ5JA=6ku%~_}>fEROjDT}U?aQ66XYsil z+~ZGwVuF4OF2Fy5rf=+?%{H<-l&XIXwePRN5?BXyh|k7SMtykohw%EX0N;OokbpG9 z2XwV_{I8MUg3qh?yxZp<=z)tYp1seYb35ZV=u+fU?A{OWpSw7Mui>TMbn^f^#d5!Fy_zJ88?^zwPHLY?$vlDkP{nw>Yd$8Qf z>*@0>{w7MGCtyGg`}+aZ@S)_{C07}EHlf-lU<$MabWLsiXYk!UGlzO-?0ov@_Nu6N z7w$)&8GMfIN6_SI`*hY$wr+7B=h{Obz}Mi1>USUCh1UP9)eZ4C-dcW~$MfxS+GEpa zi4Oj*wcQZsj4zeplMwgl8LnpWH^mja&x-TtI|M!8{k2!m-qes+W^KU=KXkEHCIl2555Na_la|M`}h6lJzD}BV3z&_W2vo?9ExF2UWUkYvJCDj_T`JBn;zaW=)-Pw{d+PZysALK>wy`3fo=+Q9{cb1$I5oz~y{ zA;h_Acw=+)-hlz~71W;e55VWpcc-5(r1+myvEKz~Ld+A`fOjA3!2Ozgm|-vD{zHhp zyF=@w*zBc5zYT9KXG^d@15IvfdDa0R`&&Uj!pHeabSd&G`Ui9e@SgDq*)wa-n0Wgg zW4i+-F(t9qazECpfOq08RN30jZd?j==7N|$w1?jm-FNV7z}W`S5&E+#)~K=j@yttX zL#XF=0AGjL{{-C}IE%HW@Dr#t-?x63Rosj3PxJSHcc+Bj;p@JQPqO$-_^xm-3A!b0 z1N;`y*{6G3f(C&;0gw1?B9|fV&F|`MpJ7sV7Xm0Bcq37Tm;4?TwXAbY_ zSr%*hTpnOoe+<74df2@qGh&Q$KKJ+pUB;e-_&htOu$LU_oaWhw7U=9}xx#)Q&w3Bu z=UM#`G~i<&%I+^`+S_clI9*xcnB)E-mpp2I76_h6j+_IqVW z%sR9P@ocqC(anJO!gEkJM-KS(ke|^1Ig5I20p6Mu^gcsH8FyrSO^kZurTY76qPFlNn&>i_+F}Jhz zq2_L8#P<$e3Ie)){9i)lo#P6++9D6cQ}eEh@f%Qo2bl9s7I)`6&>6SDPeA{q1R?In zxkg#!-RF>4ea6sD6Y?{jqCgG-=Y-hB>0Am6?`dh?riQpKF+*+4eWIl67I0IX>asr#IO88CJV z+-XVd=swoIBgW4ZpB#P#AHUOmhKmq$%<%mRo4YZWcVG`Bz&xS+{avd3qc-}|S4^^a zCR1$c2iUv!*b}-B!2I^Li|-cv2!s&*|1-RC>rnf>$f8_gH(s0bzD8eQ7k%{Ft?j$v zXLL!($GuI%j%_)=RL^W@0Zy#}A5H+C7`8s?K~I|A

Yh?k~2cqZl4o< z`J7`b%Xp6F`c7Yb{NCu!=_mMSt$vu%3wZ|S3UQYMa&2KRk-aN@WuDJ z-3O?@h4gs^ysxXs>fb^Oum;>y4{T(ymUpY$uT(hqD)t4?BhLit`Px?s{TzEu4*U0+ z-otKO4xL~x0>CCl$OH5xI`a;|BJdvD>jft7-}tR>Kn@Fqs`vF z0`9al7W@GObW30hyjRB8XZaiWZq6c5AKd}EXV8M&1-YH^5me~(9l$&LCVsoXeRes@ z`}qDf#{k~*dkZ}G1$4LQ^tI;{K1X+h?Es$y{y8*9wm)NwmU?&W{}yOFs_^GCg`T4u z0Q24=_dtzL3GClHv)|gC-?@!>2fRn_)A)@l=FsN6tF3GgNr>Om-lL?9c6Xd$(>^5T zCFo?VA++fUpEAT4_csDF?CVve?)5dib@X?hZTLHA z$k3I)|BZKuehLpLcePk=p0o%!>@t?e!I{`puc_QWu*2ILj3pT zAu%KDCH^m=#?;sns6D=_;vN11>iqlA=b+g&u|9hR^jD}gb8rOjE=uzBtGGM&wvXPt zWAKmi9{$dJp-;dLAAov&m(Wj%OW>0b z&twtZ7Ix!n^ry&=_}KF`_zk#YpT{4tB?8}~rScQ(8_3o^%wn%=$nMyF)o((R5Mw=G zYq_&NK6}7C0lV*chdvfHY;+#o2#Z?3Di%}+t&$t&!dKS z&OmQNcwvs&PMj}A)>nO5avty>ukg4iYM47GkNsW2 zFM)|Vkl^p#u-=?JpS3iB2J{nTf}bxzL^#N58x$m2UC1~M%={{ zzC-9`6?bp}^;4H|o-!a{7oPnY@qMUz`|Wn`yV5yQ5U{zQTXeq9dSD&B`*4=EEWVS> zcY=Nj)_}76cn5D@XW$o* zBENzD080F(U{J;VxVuwePc^k}v)Ge;+S?_vwr>8xPMp7w-n{O&vsuerW1ye&yC-+5 zPaod>Jt&uPjzfI5f%61VhL~4cgE9U$=zZ?GJmzxd8M+);qX#wj61=(exrTaQ6EFq) z^5}(o3&aedbK?4K>`KjGWd#J2~33RckTGsRba3!OeA;%nmWpeyJOz+r1^a^SO!oD&zw z=dPSBg(lEMijT1YImgF)`yRQApQE>r1$AY-Uju!CzS`W0zoX7Uf!^6B*z7MS=MA!V zv8bYL6Kehvn?5hGb>HFg*V6F-bMHEEVhZW zKR-X=GtgI=JUwjPd;y(z#%DqQGQ{_&XJ*W>wFmSA{Oo;(J_q))if$X!*aI|&dMD~C zKCkM3K>r)EcjAV)O{nwQ^C^0JnZuv7{1fpf_?TbY6j}NscI!^T7It?~fTkaw#b&-X zhyL!#o*Fw;ooC)7W{976(4NfWoB^6c?aMQJf~MG~zv%s=x1>c zL#TP?`j;_>cS!r8KIo>{qyy+PbT5nvNt551E~|L<`{=&|+Eet2XnU}Zv%G^YR`K6= ze3pEtcrKgh^!K^2wzfX9F`m&Ha(5TH-MbfipWx#int@y71h}VVV#ZK=4CoHAcY96X z%~yu@na<)I_O=Ue4tFvFe@5qS2I&70-ksWGy%TlrXN)ch{Qu@f@81)~mgp|v1F`Gy zo>76%G5j9-4Pc#@$o91Ywa*m4Z?QeJdTq{PedqfUy=Q$3+`Vxp@C7&k%BxV{_wLd7 zGj#4TklTGdZ}ofV`{?!Sun2FinqI$$<|StaeZ+6MiuugxzZIC%bJE@eqt?C)?P>?| zl=$V~2lVfOd`VvO43V8@A9U}`b#%__+#B%T|L@@SH?}z=_$zb+?1LTj;0RuyS1tA2 zX85PbtKbCu4E!6w9bE(adBVpYozL^E+qgUI`r7MTY|eDI6Mxqm>%2Z&17fUUed(JF z9=w6ygATFV*8;Ls1oXE+{y8)cvBn(!v}*qil(+-rJ7oKvQu{sBJCe3J?WrdAqKePK zrZTn%e44%Rj1}nIk7qQ2-r(Exf!y^Yt;utX-92>r9`XzPoYB}E>Rk5K>=AyB?t7>) z&fmw+XD=t;BJw#HW%2t*{U70_?)nRCLtyL$`mV0EYA`{!3%q}4_=OUne}G>_zX6&U z^mq95fc>wx`X0QsPM{~~+*bnMP-+@`i0%yJfp~n~mHGl&lsr?|LKUC;BlyvMe4Zbm z*T~jPh!vL=lvvNl*ndGbPt#*t>m@q(3JRzXQ&*09DS(_XLJj{2t8FJp<0= zKCR(BD?+TD!0%UahS%7dcys~ZL-g*&bGb#{!~VR*0BTQL;0jpNzF$}5MRtcb&;Xo! zfGq*m*H62#=2TxIy9akuhgka-`2jp)&w(>sBb!gT$l^QW7jywRAuiyfUtY$0yN=Hq z_Axs9@qA|B1^fhELCg}o@8<&5OyF=Zb znd;}6V{>kK-#gNp{M}vd`kUi_jo%Zt8*q&8w_pQ=GS<&S+|NoDckaAjVRuKKmv_N> zpMqV`oH;goaW5aB?q)Hg2KQN_FJ{I~FTP~Yzh@G15jxuA!V9G;uK9RU44 zhkgV6c;^#zUn9F0bA%Az8`@{Y&agMTQi1G@&xo_u2-~7@&;rJ)dJRz@~f#ypzrm zpe1nbCG^UhP<01Tb#K5La2EwVXiM(t0iF4#$hkUt==wjc;(6WS=kt~#uK{biPv`vt zBq2UuzGu|GY3Zj~%xldFx<4YXLkHjh=(&JNTSfcW<2!z`N&8)S2r6>P*u4PV~?5F`xU;C&24-jov*L$j)~Q zy+uDl?;f{+Iqdx$oJrA{dju_@p0o2z@#{nFF~#QR3R;GE|GaPSL5+|1WEpzQMKv{UdOGdm9rs1c7|+-_KX*t+SyWoPuuOQqQP| zeHWPLk+`PkGWKF^YbDs-*9)k75!V&{{V(*3ul4tUdwhp&0`*MXv%T1>_2hkzp80q zi!Z&SC#!{8yoW0*j!)&ogvRa~Lz&Y5O29g!Z=+`g8nez`9-T9x+2@ z=!iV(FQA@@^SNX5mw}mhAKG`WKYlejW#iq`68Hdr1}(6iL+di0p?mG$6Ng^tyNu3v zoV8O>z~6xl^zJaAOYn3519UxTLhLF00wF7L>ge4c@Qc&7G!2w%h7moxX^ z$LN~h&BQ;0*Jlg+0d$Vtoc8D*<@@lSOT81{0qUGdy?wd6W=H5J@b2>%y?%?(=fK_F z;a^fyeGj<^@vMFS^pWjx1h3t5yT#wx?e_|(7f<*EU|dngUftyoH17=DR`8!ACs6lt z4YmGS7VE4c7s%c3qZGMB7WQ(BpK~N2z)QE#%li!f3O>M}fNfx3zo0vUcaE-CX-ZD- zinZ+mk2lyGfD`e*n z8MC1`r+ypgC)jVH$6yM~V{Y%xP#aLLvHbxoXL08#e(u?GpX0L+Z_Y!g{^}?R3dq-NDZoFp>xM;a$<@9xmWFK!N-T^4FmA zvxoFP-h&d_!>0$`!pC~%v={X?vhS=BGy$HI_p=9v%0StB>+Je@r%LD(@XqdmZLo#D zkIy1}3f)8wA@0{c-SZfo^mDb>x*zYF)Xx~+y7qV0^3Sk6 zki+x$?7naHZ;|cAo{T9%#1py_{*UN4Gj>F-1YE)ou$jX%8N(;gmo0B^2NnH;Ij}$P zfM;jCXZH@@0$KRCjCbR2J^^2R*3fSQXYgDDDB-=!1$OKA;0M6Gb%3JxUim2lfcG6) zm+_u!dldI^FP_~DzS$MD$%W`FUt&}5opdH`XXxv|S+p^KMW;_^^E^C*=OEzc4u_zR z-t)A!XV8N`0;j;aQ=rXWn;7)gcWyr`$V=cH+=rYnS36SKn~~@I?r=-A7`+iAv$Bn+VJaiz5`dQ z_)hvB+bT$~pP?&Ccw(IU64`l$^&gObggS2@dk^a88G3tjHt&%8T*LMyn4`ZyKLqCS z4%z=#z_WCYJN(O%7K}yi-aBvM&EhHr`e2>uHE1nkK( z@ct}A3+&EnFZMA3YZZEQ?r#e@K=+~7=r@4AedL1J8MYi>_whY8@6QOG_h143Cy-+s zV{@?{@Dp+@K|c^zlGnLkVe=09J+cV~;1Jzn%bU-=257g-8|3EP$e~WY zkKJ0HS4}?WvDPkVewRV@{S-fQnd26EhkgZk7VbL-`q{tpHFpEKtG&?b(KL1IGTzCv zD&CRiJcyeE_hj79()-Bo;&+C;03D+%@w2}k^aP!IvPO#SEwayoc`kspzJ>0i>wzT1 zoYt5UFD;1s13u334tfB#!J8_^hmyTxTSqrhXAUqY?hbv4&UdAAP10>w?uaij2%P0@A9Wo>__OHi1kiK?O?aNozVR{c7L3S~0(YW)8~z5o&X^1OHFO_p?+@62_<1hIxKin61kqJpTZ=#x=ZX;np4MO9rHMc5*! z3aSQO>N%d{`Tu&)mAUPGpZ~ef`JH>tzR&0R@%z25nbaaT0iC7hNTMHr&w^&)k-HxH zfq2jm^Uom)Ujwf`QpBY_{}~9vdqA~y9{KywiSIhl3S7{JZDq#adjj7+QnhWboI#ESq(bf*B+qr%b!uL^!T{wJK#e=d325w(Tl1V?MXRXv1@+qLow17plbtZ z^ilM2PzajP1w6lg3%t%Nj6OfR?y}}dVrxUM+F^Y9(djJJt8%7X&JSA`Q16A%AU4gT zT#d-R;_>UkJ}&0>D4X_~z@{2$^pcgs!#9H53`#(5N6-+^PSwB8SZAUjd%@Zid&+RYSEW1K0DP zOU)WWKO-~v{|$xDf+Rj(w;!)Fi6i%;SH3Q26|#Cy;G_N%pclR!X#d)o>-v4V@T${+ zKHq;*zQojG%l9dtV6}h0Jg9ip)!Ar=DP+~i`zyW*sE6rL&-L%e0De8ldUxi1J7fO$ zujVd<*NmD&Go^^r{ubabn`V$4#?}e4#PmX!U@OAsd8nZIb@u8_b!5|wI(PLH1j-R0 zf4+}#d0%|;I#KzH>P~eK(*X^EBzze(;QGI{fKMR2Nl)|q01D^joB!DghMAg$WsEb{5 zMd5X(X?!Edsyz{%cA$PYL$%{xAe&HzpL*4qHNh_c%8~1F&_{hw1Q}3~@$W)DK1b2J zpbVR2mwX9w1jreO2GMIy?fho=tcwr4cAz>s%L;hyAb?H!Uh$s4n>fD8(S~2U0I!+# zOlm&;ewI$Pw6n>Qy9e)0%%4wtSN=H2_r4sRhhLUD(oI3u9@KlvMaL(OE{I$g^PjU^ z-!2CH>#)rL+KG0nv+oDWnZ`d~M>IDZRJ`nk#I%A25X83z1ks18{r5okU;S1u0Bz_cN8nQ~?*nK6svf7nr{VSdYlaqdiqovqkhR;T z&|dVqS8do_{58isWWC4bqrSrEBG7uEk8JsQYtVU~|2I(nT{K^Q^wo=>_NlYg+zEJ{ zxolx{Uk&<6PRRzHh;CfWzxyG0^&-8H_Q;1V2Q2vx&-b7^mETDkK7TfOyKK_u zWl@isLfv2@TC&TMJ%QODYtB~99^U!6X zDfGHS9=!7RKwZz@c@Q7fuOOxmXlCu%rJn#afo~jIoAGz3J*uwy(K}Rin&FpV&w@9B z`p(@Ud?WC(=X=$2-UmWJd=+-pNOt>B&LBGZzb6}d@%qkk;X6Tye5JBM^Zn%VW9zE^ z{hgy3l~*xo_$FdBe=a|<<3Iq}b^YHBs-+t0c@j|03DErgmOOx6R68obru@pIXH2z~ zTRRc29R;9oLZjkc|DDp7VII$a4^y_5gMPi}_1=?CcSJN>%^tA_ zVty^1kB3eD z&>nQRG>i72kM^j(6{Gw*hvmS7PobLx@4`1j)#ED}{|w{EA@l=4`VhXN>(IyHg@F6} zyH{uF0zErHYze6PQ_N(jc3U_o=XIij`jCGD+J;X*RC}Ek^Y5zkE!bV^bwG;-^@`Oy zD);-BGY@74-*M>Wuh|MfJd`;0}Wc8(4)VoLi5L9Q~jy{b(p7Eax@oDrae1iBW zzv3mAW&H1s5I)MQJ&P_zFRHT+W7C;*VhbT_UhPbBDO5eEW*fW*-y=PK>P2~r;FIVi zyU>ZqUHEFYCq);4-JK}5O+dYLh@uNZJ#y)u>g>bS{yRK@9D;8_HxHj6 za+;iq?T1&byeyqihhGA$0g`n$m9qqN64Qk2kyCSY0P(q4&yQ6Psu{tr6=)A(bhYr> zLjbB>t7qjJ?ZBQ5FKIH2}mjP3O=1hD387a5U zNO9UYGV(0P?37r_VMi{Nz*^4GJc&l+eNo93!2mThXUb)s z4DTS&3sk=q+DW`-sz=@gG*cE`5MK9i4$yuyR|g1?Cka*E7WAnYfL9I=nwy^<)}xoN z=lORxKjUO#R73UCm~u7cOzbp}~zRg649{&m=J05qTSOIFSvc+d5JYjh&_f&e=8qaAqohoFt< zmH_SU8KBPq`ehm36N88ovmCw!dm*R=T|j=?iE8%aBR&NUxPER|cFFWExY#G*I|uXw zoqGkk?0@-q@O|h5*dw3;ECikCzKRXzg4Yb1U;EGRv$xvUmExCx*DhP2F6bnOdQff6 z9fc}~<_Te!O?z%YZpOA5z5{z3R6Xe)DyAJ?yO&=Yn&>80ALJ&WJC%E;;8nk4kQSoX zN6$uzxYbbAjbjgxL+7XZF1%*0L!W{cL-W50CnBqtO;Gi#-ye;#1NBvpUiYvHS_jsG z=|H}Me2S2h)X=kC6!YI_iix97(?bb5$)(8h*O}FVK6uad_Yj7c&wwa-!ri`K3aTCT zpx627e&{Tl;8i~ygOVc%RbK5!J4k|X#_zcVnkA+kn(G_?6lg})USRqaHlf4}T@d<%k-}rohPBDrNR{PJ1Y&WA@2Q+hM%-=&X)FW2_WC!Ugbn3AYzsX<# zXm7d$jnI_q-`{rRFtYZrxto1v=mS2tU#N1Y$*CTa*t5_Su}$cUKnwDm%qH2*XOk_^{k!dKIr@5)xX|ZiYbDs-&|en1<12ND<}s6bgQ8CgS={K#^um3 zv|&wmf;jw`2j zP&^3Ti2-yipb9<-4aNL-U;;i{O}tASIkdwNvV4-zeyAXw-Z$!1eaDg4kV7-}qF;_q z@$JwekV2<(tAp2mJ^brq{+(_@7Xa#0IuEM64e**nv0?l>KpZ{^YVg-{nsWWQ`?0A% z^`#i?Mf2%-nE+oG^S@71P(7Q9FM&^BS4;!4`c)q(4x4^H>_&xVSJ$%$#9Nq=mUqW)VzZ>OKZ(*RiS!^y)|4C>+R6D5v zoya;L&6$F)K^F$)AOMs{J$Jz`1Z$-ye-FA|pmXib_&V_^`6DNw@~eX146mKn4e}m- z+D{WmDJJIkR0}Nxs^3eD=F@MQ1ag13UvC+FGrC5Qb^V>U;X4^Qw_EaMK__w&sD}Y! zLqK&y=z=l7jyve<5kD1unf&q7^B;sNPdm1+@J~Wr7rSbw3O3S-m$JFx|Ua)?*oZJPO!Tg)!eJ1g}0cN0^vSd|jw`<vjn}~Z`xZ5syIDUqS~R(PtQOSdltmuUGM?6AhOf=y@$cE{2z{_LR$mUR3p)v8%3P1LU3n--SL2+Cc}hdUQSd03NzF(30_U=Vl{L z{%Lg6Knr{doz5nNtU9^(0J8jbzS@U&wmjzlF4wMQ(|*J7I)@5u>LCD?t{2Eh{Yf_& zU+wv3cSmjaeN$BPSJzw(EjIw3%PXX z%A>CgdgjvI%y#>GN<&MKyP&G0yCk1w=tT8y%92-kvgkCQ?!`FdPN?S4o#}_yUIH%v zgI1x_E*qfABLvYG#Mn6)fY0Kqv#mqU*NDUSbo;S-e}~|k(1|wUo9hXx+~QN@2qLRy z5FhQxg>FWULp!i(wiHx5k)L*+L{=TmQ;0khxeE*+=jI}>e0BF#KZ~sM-;69?ceMn4 zDbN`R<1&7}1U5bUZTJec=)+LeR3D4MQZNxD@ofQ$&5~F7*28N*0ngvhEM(2qFC8=I z_n>pk?GwG{lB;|0&d2aWUJ9)MaqRL90NtG+e(EQUk90b#eq{9#K&M&CL2g#``FeV` zRA2jSf)9Z>{PWPsK+jqcRDPZ4Q=XqIj@%gYb=pM}d@uGiP=7&m;`Q-7;#~i~U)Et$ zPlZsS9xMkzeA|I)Dpwmm+L!K&=F>baz$1@#sGjAkT;(x;&t=Fis6p1ex-%~RDNvdj zywm6c(tA*R!U8$kK^;E2A64jTK@z`0bgCsN|EfVbOMJ@p@5{oN-)9P4koXkxa{R)e z9cbse8_KO5%9Vl-qmP3Cc{GO$?LpRe>oYO`Szikc0qI@rI`=mC;*8(7a!D5g;ybbF zquJGuVkY7n0HWoOeAa`z?BPI-;0bK$*;ftX17kayY z=NmKd*gHT4ya(0$IE;_y`8!ce3$`S>>DW|H&yndnA52mO3d1+SP8RQ;%zVyI@5pX#b7>FbdbP}lW)RZh*F zLU!>HKM8o~KhF4hb-rcj($IEj2r7HPBTjJr`(1-AUhU^7gjbvk6sL1me{E33wql=x ztiJWG$j_caryXU1?!ha_Z6F0NpBDHmIdrc#LpR3!@1?0w#mQ$QwleS{kgPM$dhlQg zx(;&bPHG=>plS5Fi`s1nT8B>WDebQhM9?Lm!EXQAS8NcSW)u3rL}d9VfpV<|>NO0M zw*;zQ3ZOba@w!(f>VaZ4llqwmCLotf2UK70wjy|ag7D=)=c{?--vqxBn(Ki)(~#3p&7gCT zZydYMy%5?#&a~^#lSbAooiYD+mwIhPr*l*8e&kkcL7@FtK(!B-+}ia#WDh=wO|s5F zIYfIw9D4|w%CKu_sM`0{KH}12F9SN`B(xByp8g(-fHZw};ivnhK0nR))PpKUK02o< z&;)+!qZZyHZdS(sZxO2Bg)RWvfP91Kw39N>4gwzE3D7Vo1nM)_H-1&f?}6N&kTuuz zLAfO6f0wU9R?YxWzLZNp&;T@yt^>XTq>*c4{;sQ_b?CC_O5rENzXDZ`XQ28Wsam;b zrW!qd`pwk})VJyt$NZcD?AnvgSiOw{VR+5j3>1?Bs!@bo4)ly@|7r4Sw(<-)K@n&M z+PUuOW}tm|P#51&%%3HJPBqkn2b4o_;dL)cKq%v%K`rzdpgTGfOa`(A$ftfXpcJ_R zya?Z~T2S2?^(CJv&@{-^K(9NQCSJ1oXhIg$Yd=tYVaC4?HSo&S2ISv?y#?K3p#9Fk zuMFt-n9e=z`M(q8t7kBdUsS&EnmYlndp`xL`6EESf@%-o7Y45D?@fEm+jUN=A*gns z8b8mkDIeVneP0JMeoPcvJjmBTJ^Vdl<(nl=ds{MS4`S23kxlbzFWPejx)f-G-vCO0 z=F$#pK_`Bl$o23k#lV-rtJm_3e=eFgj?G0EaDRWNY3JIf?q&d={If9?S^L)RwTmp6 zfKD^(Tsnd7LkqD%c2FgjKA+>H+Kg<>-y)k89A2=H~DE_+Ldz60n%x|$`>GS7=QKn3W%a}U3}3g zPPyf;vvsj)FFLyvKE=pkeB}2g(Dz^)D1lGoQvp@1bXm`T*Xo?qo9=eKVxa216rK2W z@Il~#Ng#wy_uhqH1vFbNbP3RX(OezqJo-sNwKvs_qi>G+yH@>1cn=!KM>CaT_lPMP zr0OpP)VJ~#;ok;zK?*{~QAFW$>=YUZ8V;=8HfZ zGCnPZSM7fEaiIJQu$63aRnKvF^&7xPwg|LWdg;hhg*^zW&?|opG>EMWbacamFnk;6g0I78C6KHh z)mIU6DRD_?%JXBDdjMM!9}m6iYfm9$*#zyg2fgx2_B`L7!bf=mgT5ui$wxKwv8Cw5 z>)h0%c-^N?e0o76d9yD0@YB0}0e`XRgE^I3ae8L@ zx_$rLzr(_-KlMETl`Tb%Ui`|5Pooc^)0yk{@G~G+AEKGXC-IS=i=X;;i7Np4K1KVn zC9vtN0xmU>Tj6UYqxXRHT^axR)qT*8BgpD~5;UK$5MJ*V&8}T%poPdYfX-LnPkCR_ zB(h6x)k(l>t`xj>HUT*duiYlWYVamd-YhYCXH|eTpd~{N@(Bfly!ujnDfX$L52&{d zKr`26evc`G*UsYjS7BFUNrKh+7MufiSzCD?T4S#&9Q`3FEhvhIU=kYA^G;GxfA&)pMb z?Y;+{YN==SnEo_ ze;Qe+fe)eE0L}I0`R~hgH96n~^(tEuKmG1<(TyWE<@&t#<-teL$uEqbsGvP*H=04b zd;_3r&{saq@O?nDbYUxmmVf{#&pY)001MLwVG;!CinJb!mZ_$$BmAp7(1NvQ5}6x&45qBv*)XahRu z3aI8%Tt~*AH%rc3E!Y3|izV2#JKdLFawsMOijlSZAkhA5(5bf0wiH_h&~rWy=xk;{ z=L6M>n~QpE0d*jNUh}oXYhQ6N3BDJ-Vic=7o6#%Y z^T_M^_g(REY?{Rb;`4qwmjrSHP`{c{{bixjt%h#|%BQ-ri|WkP&qQphorlgefKLFM zW>qcqIs;k+0-zqa_yi>rt2q>x#xDVqAmI8tei5I3^;yddhJ|$(mcA? zvMEM)OFha*aRun}d(#|xUwYV5vb+9oqRsd#pK>Wqc~vJqh%8jSDK-T!9}l#ln*s_! z+V$TT$|I{^jG__Y2vi6TJ()Ea=>>1<xF8b{P}jG z^RU-v{JWt&xt@OqMQhQ!pcvc?w5u%s>Nka5`_f*t5A~oqwKEqV4|^IufX;=MhzD`x z6wrKGVztMt>;K)SI=#qg4A=>AUj1th=|aSH0_}S$x?JD*wGyXZT^FEl!>-*bW+Jls*S}S0UhS(8+c;#|E1>E( zfnNDEhxBdeCj;?)=ye`x6%|YA2h!{T(d8rv2&Oq=0(R?!+sv z&`6B-s=J_^x_d#W`cSS8c;$)4@aLm{Tj(G3QH*w|oa$X4`S$?j%K4$!c?Vqox5f>$oqCo{Il|=DzFc&hi1k1KxcE0fb!MH|l{cB;dkWf%tbSBqy(xYf zat+YVwYLJO_MuNhhWU|o$Hu|u_5ux~OXIr$s+=j1?^)-YKtBbjcb%2?rdms&dB3d7 zy+$WYjiERD)Z+#!Jp8TNs zJ11I)ulf*8fi!mQLFd^5O5w{vAG~t+4pPmieuXZO0u%7lZ}^n!KWn=8E_?v0Uh_Q` zBWFGTSq@jDhu2QEhZKA-d-f$swJ^2-37qjJ57y%~7|P@OD@qtmVwUk&V#n0s|PN;YEFh%-mL3C zr+O!cL8{x|e-_@uFAL%seDO;WlZGac7l3x);Xe*4o$iJ5#(_R@_ThIb7JgMye}N4uy%u0d8mk_VtVw>dyFsVB{<9<=+( zpc%-X0lh%8>OS@Z*-K*nx#ng;*9&wXlz#@cbhn>RcTT-cLZ|!*c%7ThR{eB|hq|tR zj#JQw!2o*Q8=*7f+q84-B#x|l5%@N!cCs1-f$Axr?q@x+Y6kELcz*1P3@maXHtktw z(gW3OidU^DRB{r&9w@({UArLQ;)^~7%_6G@?YIT1`R2e+1Z#ly6aii6(?FkMWc3xs z9+nPX=c6;xzI8`~gL>VOi7`K~cH_DJe)R0<9JR|;@KeErj6ZV^x*~Y>DqVfOe$X%9F&dcVZG=yGkR6@SO!3fM!(AIM64H-)7(9+CVe1_7yC*L)l&ewJ_&sLfWEurubpdW$}V!Q=}fe{dgKr^fWPKZtqimR)PQEt8uM!^hjcCA zO`wnG`t>wVEeMjUi(I-tnxzPx3)SwFdk$1P6R+KCmnm|z;Xey~IVb{(od{fPd5!*3N(A7x+wT|Q3z=xqCm zNs&K-y#S=pC4utg{(TOc?AmJqs0Hd-ee@8k9z54SW0x51LU%)V%Ei`#ocB!=lK`4a zyI!8*osVBFd&S$vf%w_D8vue~Lq((9eN1SHXE z9?`;#@2_0%i6@7in*e+kT^mr(+F=Rx4Y~;^ z-o@7?UUyluCh!YkQ_tR@Ps;VrR`1DYKmZhEICH4(mvXHFi-G#foewksI??q(!_|K7 z2{Hd(>nxOK8PML|==mszmq_G97xwG)PFPhy0 z%B`HzEy?&fBulSdF2P^44J3f}qfY~R?IPQafNvB06wnJ2U>2wVZJ-QfUH>`hB!}`V zpW<{D>NPDHgn`bZWH1J8Znj$FnLu|+yDfsB57I8Zdw$=A=(UqI=!5V&`%-w-(H+*# z%8^A|u`5>^1VBO*zjpk>8UNgM9$6QEba9|rmH_$b96gVB%b;8w^Ydl#R}XE}s6wZi zlK5!uW~g@31F!w(Ybu|RM4uvFXQ{qBKreo(llPet`<>U_=_f{cw1al!E|5mn-P4)N z-V5r{X^t>Dmx4OrVhiHa1|}hAk<-*D&-ioIL-h^}k*@%%8MXg1 zpqMtO2lO7%EFST>GbM-m3W5RrYO&{M6a5iiosV`Dz~;g$ul!|K?=Ci-z2wd4HJ^5o zLZ|zr88xqRzlr?c`4yXW{pWoEzd87P{(t!Vx4-7qS?fMc6;FHtd?&H3k~!;sbW!*; zRClTanxcmGrT!bRCC~-If6uG8?#_zKf-F%>2?a&oq3RnlUJN4m`PiLVx<=2ebx8}@! zJZiwJrgkEupxTFO#DVH+ zXT8vVWaViBiuI7ElBWZjrB*R|?axKm32g;BU-cvO5Tm_lhDM;BgpgCjst3JS%RmWG zUe#?QcR9Qez|W2O&zPQP<Jr$*LZF6ut#i#r(fB7sBiLQf_@Wy7=X1?7&}q4SMx36?mTSzYv{j)}WgN zv`_VKKc-P`DuRj zn!rc*;XM$}_;)y;OZ#slz66>jhvLfdixZQlaqLO#?dZlq#cK~KXhJf)cC22NKLCnj z{+_ewa&vc+6a6^!E1>Ergf4)*5~PS%ZTTe8FM}=y(kX|0W$S|$#{74RcA(x{u=hig z$Q3{_Len7MkG=N4^l5zSWB#6%D}zn3uIKmcVppHaF)&EeuK&K$^WKQBhmUmXDPONB z=I=mq?yf+!n@03CpaZmm2m>srZZKa zx_{E?ISsfV=AYdHsPqlkHRD&v>P7S9bBLy0fL^@(da-vPt3Sm`KY>{Fr}wN2Uk44~ zCtuCF4yeB7N&v-#$RYniPy=+vL3El=-v=T5TEx4YA$-8|&oY5;3Vj)LCi=->At(nr zd)-IPAs?M}EBY>Ins~iObvBx@8DBj=agfAk9H;<_jYHQ0#nonhKkEqm6!e-a3NOf~ z5nUmC15|x#)~WD~=rpT($pY1Pq3S&V?Lk&tIS3)U#71NOz7*R6!uaTJYd^}R{4MyT zT;DfMToPLvU)l9ePGc*EhS2HVoWe&o-4{XgX*U74!=uAB10QqDqKrWKbL&rz&;~;VcNPGV8nYHjq`M7=#&7oQmFrZw-d<-wD zeact$wC~&=m}hy+?=edbokbyX%JcVmGrV>gkUt;g6qhDFT3l{g>OQ;ZgVI4&D+RAS6M%BJBYW5q zT532YA{5zo`XcI`eity*|mma@1sOS1KXdl6tuh(-Q$oRdwo_}|; z_-UpjXve0rY9}Uy90qR8zt@^$!XU4EktWUqx#tI)&Rz598CI_?Ks!!CpUL>&&(g)Q zYcG>Qu-o5{YRJ|C)ozM1gc4gI+4Xgk;4>hV@n=zQ`TGz?Uxr?C2UI=eY{)ssn)vjvuzAk$kOGgC|;<3BsfqmOcG zr_$@)qq(y9>bF!rkKQv$pkCLC4(6}LNAnGUysia%?#|$=`=9cBTgvs%L9^;zp8|70 z5F{jH?;6b0p7H;GKVjrv^0Yzcqtib6;U&j``b-m-uPv%O7q9;PZwL9PhD&@9eKE8} zI{2mV`B;}c>6kCO*zMKir50t{|GYJ|(4qy)g?OkW4e9GMkRKFgZ_K-lYy#>*0*LshnUE0M~ zi!B8_Z*bS}((7He0bY61$RXksn}yHfEC1p_yL`eiV%*>BkHgjr(&$Q{@(IQKcXt+D z7kOOt%AFvuo~0UO^{g`-0D7hhfCukF)1KdN8+;0#dQ?Bj4Ef2a9>d5BkkjzJATr1+ zR~6{T9w3jN>o|6gxMK9m(F_#79$E$(V${I54_&Sne5c}>MbC$NlD!GvXMlQfyMgP^ z7f03}HbYlIRa^O!G5=gb@baIDEdiBYHC^=b3j&>w`qsXbKY{E~vj(c)vhvexapVB# z1O4dQf#%bDO|!Hi>$wZUt6m$te3VPPV&$(IL8zX;<`_C;<#zGY4(6frTzKpi@Y+Ql z?f8k;9EwxDEJ(ulLnomRfoH&EAm2i06f6MRvvkU>z7oh~=rv~?ny-_}`0qe8fpYj< zobq#(TDtLWSvV3$jPpsSG;nIrtl9_JAtoy?t%6K?RPp@ z4r(%f?q+BJTL>Bj^4I%J^J{-a=rp79cA=A9?~OX3@01Q?oxSE%J)KFhkRb+I59R|G zn|7nUslRq&mO|SGW2Mjaj$b2u+V$mt=l{JB9JEi&`2DmHr`{t#{mV~hSUBj{i9Ur+ zb$jsDd~r~Py$0&J{*JU8)ontrv#1A>!^E_T2U!pVX>^;Q>RIQv4qm>6vU&J{U@U&@ zHihS%e9!PeXbZ}25r`kVWdS~g`2HWCJpwjx=EdV1&%N}lLnfR#@ytUGjX06YQI(Of z2|FW4jfqtTPW_*MR75?LzEiU|g#*I>t%u70{~kt;8g*zSGA4G&|Fj3)o7@i0yL-w1 zL;XrSGGf`m|G56@3)e=#$?E4#6n7V@>G#md^5SjMIQMkjD` zn?3cH3j5-eTC=41_)*i&zHQVG2ewqMpIvWfy>P0z^RboAn?s&7ufO#llWn@)EIN3T z`R%ozO=$T-^T0U=oASPw&0fnNGY9=J-kjZgfPMMcBkl0IgX}#eJ472V92R|}?~i6; z?Me2Nw%K-vf@_>JPKX+(>;d!3;eW8jOE)^(kGjNp>x86vsQyIr>aI7K^`Aaic~;+W zyS^%I4<9kZY54PFBVYLSmQiP(<3@HEzRD^8w8^B`J>op^moau;(QNzKx{zJe{7+kZ z&qa2x#$U{vN2Sb_uRi3Q*0jp``kg1tW5suwYc|#A_Hx8?`;0pI?Rz8l9(Q_Vm!*40 z+*fvqcI~ly^!$g0*+0DhvUz;#O(y)~8uNgi(me4)qnUehiMg!%T66NevG(czoN6OSZEa8e_fWeoxkYrBajWgf*d30u z^JwRkO@EDC{cBg`z>jWn-uct|!dY z$i-&##R&fho^buAl{M_x``dxG@91C6SKn>vEE+x69Q|Us{pg$T&EFRO z5~;Z6no+tZKmGAfk*2-_oYT*~-~6HFIs3t*AKP8de9sJcLru73R^**Po0+kAwNtgO zz`pv;jW)c;LD4CXS4M}meqcV|>oU9Qu4C=BdyO=Ap1qZ`=!+8PFQ2?^_u3umji8yRx(&yjN~Mmfi2 zr$9u z+44`$>Ep*bBmPuv9v!jJ*xS-h&*)do_m!8Kop&qG-7(eO?&(X+gr`@U^3ji){;!8d zcOG9HeR}==_Pv{LH@hA7(#X!ezaP2fyNkAszj=$@BmS*Da9D5U&=*S0gD*d1hTk19 zzijw}Ic4kPoNvAwZjb-ZRpyc59g%5!rX$L|bj+(x>swpcolm*iezx7w)_wdETlwf- z_K#zxIX|Cuv>kfm8|Lq)&vCXnV=H@IbG6wcba`aK_^-?@H!rf=y>yp-ecUDsOB zo!9*kxqs#n_Pvec?ROtn*t4GrNB(gC&d#=nT;tsK%C`27fk#Yp&FaW5`(I5}pK{inQE$)ruEp-Sd!N}oe!G47veWEI z-^?>tjOvJ-z4`$2UFA<^r&&{N^asp6zDN*(1!kw{Ea^kNCTdowe9`efpT(GqTsUtBY*m?Pq z9V5wqpC73{<}&lfwrS_CNykNgNc_bf|IFUD@W#zf{Kg)$?SWIx(@R38{fph5K)SN> zxOM4K`)=rT8V;#8Kh0ZY4t(T(^Pe9+GqaYCwwbG9_RKRIP17X@MjAf2)Lb^qab{ih zQ{|Jx&v2$*^s00G3kTQ-_dCU!eQWJ;8!k4xHFcW1Z@s}8xckM*3r`$r?_N7#*S&vM z^r@cb&AvxZHLrYmlXFB@oo)Z~9p>B3H=3Ho%k3@k)%MBHFSbSZj4*xg4Y7|d{j=@6 zFl7gZOf~;5m=LMEv%_xt_s6Zf>qvY3etSFfcUWIJuJaA&gq`Eg+}@ey`I=s{nw~pf z-^xBaYK3|K#%6QMN&DJ;E`HBsGegYY+u!F5yZheAFAslgPWb2m^W5CGD<3i)HY2{rel$dKTU%lGwGUQQn@8R{%PB-sw?!0)JbGkPW=^A*;R9@U* zH_zYR{{5T}%s$6{VXl6?++MM9Z_~W>XHN2itIZR)9%zToOq&b8I>PBXYpprq-J|WJ zb(h*{FI{Q|4m;Nw@$ciEuXmkozWn)eXZC}~I%C}(>`SkIYgRnehF2zeUvZnJ-*tl6@DL-~Kt$%-!_7 z{pp$GqZgd;fGw#`J8!-CLgkQ?Uy7`#xX=7;QNVus;}rYx?yH=%Q)0J%Bw_Ema=JOa z=?in&zqYV-=ceq^lg_fkYg$d{@;S~QH=kVwuiv)5sE z4a5n97qA47`wQb!h8+qjOGt7B^e!>(z-eGoY+uiOmajCg$Ws})? zt8LBxS2>Yi|2e{J_~5rliw@st_PMp%gtq0UY&9_c(Wtl zR1G-EfoW#N16!Ig_gwF6H}zUm)pDr0?YD!?w>!P(+;r)9XYL`}In$OsX;#0qz}~s^ zH9K(KY_oaylg-@TS53n!wWcPz!OR`@nzLX37+ZAkF?O4q4mJg!K54#ad&{}^{+lCv zZm7zA7i69p8rgQw7W2qozi_Vl=qU4#+cul}Sy$TCqwlldxJ%47%dR(NhrZ_QJbR8k zY|75j0_O{}{Fid)t!quBeeXXxMf1OLzI|m|Gw|kv_SX5?%C)b(Y1(hz(|-7DgZ*mt zSld5gM|;8%H<{GP^UbxFf9I5LeT?(O*=Lx#!*92*U$wpIyJknb#Y?qz^6PIqudW{+ z{W&l`syp=60e4sK6aSktV#oW<1snIU3)alA?;Txkj#>4#DLnH&6MN@BX6_FuyYIux z?6}|d%FXiX{%N!O_M7Y@og3{X`(9+`oHfmE@!UL9c*&y5{U3P5Ik>gRKK<}^rfKuG zj(5xHcHZIlTd%3oTy@X>_M_Vu+A&Y2&GV_do!f5QJnF30k3ak%L-@OrHTFRU8g-5`QerGolDZSW@Yo%_MCS< zw4Ypgmt8(&AG_+j>rD6T2b~|BqwL9_?`+Rc?P`C$d$GA-pLfk0n?mN|tH;=@4&smB zt#7gS?w@hms!Q#DWlQa`yFBNlFTKP$VNu4}bpEGi^>zQY?eQ7*g>O^#{9B*0<91Bj zjSoFzW@Y?$M$IFqn4hoy*i0Vsx!JSocBdkAxtXwVYkTja;ZE!A&pAhYGSdG2p_IL5 zcs%g)fCbPS_ z^dFUWe)K$h#?5b-H{Q9`zH{2M_K^b}JFVpcJ7&TJyXcH}%}eFi*!avko4h_{ip>Pu zbZMD==-Q*q3B3ne_uA#=y9@6(0~cLkGg~aPyZ!6mruvvq&BNENbq+iIrO0-B4s{;; zVuSf*&MH$rs@Cb)tEDnC>!-;6c8GJ}oc$tSy*k}4?YP^nExFyS9C5g5H4k9gXA@?opFdHQxXc)+eU7mT+EYiHe;jbFb4mNwW^?VT$o=OIGZ%&)H1~}!FxR-d z**BU$HxHj(Vji4-q}hGKU(L+Ff8j)yw3>B5Q3n!b(?WZ`yhJ5H$thmb5Ke4ZE8G4Sr@~f}y z(-&-TN;3!AFFu@Y%T6EZ%#9vo#;*C&?EcXSCb9MrGvwpxruO6aA|IW9qbaTXt1W!~ zNHgc_e>=lJ3i$<5~2J)Sk`u?y^ROLvU!vG}vf=>F}I((y)U*Eq6-SW6O_%GAU z=6IR?W5ZB;+PRa=RR>J5)pcz)+dI`{= z=gfu6&8~CqGuMtSbhbZoqw|-qUUqhW;SbKAKWQM#%{L!kUPw(7lL+=kCxlh9$jId zKJr|`z}opVdZnsW*cG(SGPlM|kQM`UAJ&^dJ0sph;Lvd%E~_R3igk2mG+ zHum-p-!jX0f5Tb3;%)Qd&}Ys2qbr?lr*Cb(i&UCyXgj;c%&-SM+ii~TFL!SEb&0dr zw|&lAd;h~+v&9hm_ty6R?eDMqN6vFL&JEe6FW+NlCy&VeF0FfTLuB5yN&BMnxP9n{ zgbiQ(Y~=p?Z*qpM+duN^&QF+b8Djg3JsYyU8r6)%~^vmbD_ z`nAOz`S@JtxhopYe%s798x9|CpZUJjeqG+|Jo@~$=D-#Ibgue-AE#iibDcSF#mpn; z9%Am^_X($=qr|+icDeKAtntpaOGA<1*c;3}n+|gpUp~W}c+1&l!)>3LVORHRZ;Y|*_YavFMh$9X0*R@;2) zO!;9)8+v}0v+|hR&6P>dUcNkP|8waM_TDd_G||Vlvqye>r#a`uM*GW)C3f1uH#+YH zi|mNuJKMc4aqK^b-(hO%u6OpmX0-Fd->iXvB;_v>%93tlXVepGd^z4(;(%!QeloxlI1cH+?A>?uzb+8qv>;Vj!? zU;EOrzuK!`TVdDld4ye>o@@_$pw#Yu8vl;b^f!CZAsx2<`Iw!3`O)^7cMI*2b2^+v z*@0%q(Ixi1GtV*2t(%B_izCX?xdikB^wkx-f9=dNR`nRH! z|F`$gu3vv`Z@l>yd!akU{B-E4re@FK_Ng^Xb3O-DJ!fm9%cCU&XPNuTo^Y-@wc6Mr zhueYGGwnUc&$c_C{;>I5_73~-j_=t2v<-9O&$XFi_ZIuvq;>pT(TM1YM=rFtANic~ z?KQWW#rNE43yy!@dGOP7&6o)*{+*7!9IEv|0ev%b#~+7|8~B-@)vlkMpATwYP@k=9u-bL+yu+mz%EjZO)(G4Vxw1-#Lr_w2LYE z%Vy`~Pp_^VQdZ!k=Cs(o=gzY4z4of990=IGZ~m)kE?VY%zVd!sa^?+ow+ptl)vY_3 zoo+tdxH~L2hWFEPcW&wQG>?vq`DCWO|I!t9`JYDH?RGQv_zgGMTTj^D?l82#+S31z zia(FW@{QX6ag?ECCQ2bwN+Z`k_DzHa4I&B+nkS9YoMw`csZ>OgDI`NelBr39XrNM| zIZdi}bE3Z2t6R-uIoO}bDw)3`*rMUoUM)kEL#zHGZ$`JJ<&3G0d&U( z)7i->LZyKMl&pcL{q`bi!xQ0`G709nLuk}nE8*&hVKi;uNve<4;(N>0Im2K%1{z+b zKMRHk4MQgh_a18UZ?_w0>LZT>-wvZ>!VwIt4u9Z6E?viQWpg#C%Paa_>cxK^6(PPt1qbOSGG3-*$At)(b=Bz1OMD#O3Rh#+6C5_t z!sUml$4Xg=v4>3!YiyN zW1QQ5LGxybuxO?s@1Jr8^S9KKu6`?wRwwbGpF7xF*Ajkq8>qJ54iv3h*jbnL?dL%@ zwFl|kAYTgpP)e^9(y;BxQ6UD#!WcIV&Q^Yg%IUS39px=-FK@O~{GBJ%Y@LR0eupvR z_gbi3?asO_;k?@_kK4y8fGqn9pFP6JviSw#O(tO8#c;aO@)cRFJNWhO1N?MO6HX`2 zhUbX@`qUEB`3$YOp~9jX8A+3&n%xw#&an|P|RHoGm_d*f`vvbKDTQ-qzFM8LxCT&`w!lD1t@vcuX zWHLjYic`wh>mG`PIx6Y0)-hSRG@V?o;=(pcz*w7&_0ccpXZ zGwN$>MpC5D{}X@mbx&fbx(6?B@@AKpqLv+yKsvIA;ka%ghAm2@hdEP_*dfo`rW`_X zm+3-!|0X=mZp7Z+i8!ELkGdbHu=2AM4{6k8`x%z_U=}NUwc5u+ww>nm#zVY(PzL;6 zPLST^U`HyvonI4N{MCFBLyrK_paV( zGu~Vs$AN*V0&Xu7+76tiaBpv<3&oV#`xL1)bdj7kGnUMd*XLo?M}_M*{5s$DyZ>Oc ze{nd$)34%u+a?f;}Qka9pi8;sT~(b z$nxPEfovDKh%b4_VD=6@KI*-f|3rVqz~|z*bbkpmm7Xlwv5P;xN`jL72SIC69Df~osrJ^KJ;nuxhH7%__FDXhjreTc2*=dRcx07A9%flc zp7I6fLT2Dpix28AO2qT@o9;a+wba`9XZQoP12lErb5uQ8$wlD@V0Ghf=Q}>piV{xV zeNJQU)uSdq5c7J<@=zmlULO>Ovf?S+nuBq#UkJ@-?V&w271!Fg zQr5|R^l-W>ihT8O&Mlnp`m0JtMo27&smaq-B?FxHjG^&PUii4E8(;ZrEVK{opesd- z_}=DJVbbexdV6{awj@RI&cH2vx+oXL+XnE?Lq)iIa#3dBuq9?66CxVQ;eN92Wy+~=07ue=YvNmUHuf9+dq?n z67eVP8uSw6VY_t?b@+xNBD*hd)$!$+@oRXYc>knr`i&8)>YeZEQ(=Nz+Qdn+20VN5 zC+fF;I=ML~(83mLvbmA~<70h!YUFI}y{ikaJL>o+w2_9>TFl(#hIz*YJdQEv!@B9* z)J4=N6~DvZ<`eCGp}~_Eo??v_d)AzMUWoJFBh0Ysid$7aJYJ@Zl~ww((m3YDmY-?s zSUn6szY6i&r*QI%_grV&n>&pBk?y=!konbzdu2-Vf=vxr_=8C`<1&5lz6^sSJB2R} zYBcfpP}E2C!RCR!JhCQ7n5}L5L&!H7#>{&oYh!~92*6gZQ72(kEcMT)s=Eq4@(Bop#>+MsY1h zh)yUGiUSl;aRrYCifjpv+f6y-2HE~(dh#B4*Sk>7S}lZqYqh8 zG%eK6!o7_GZy&mVS6zIKp~3dROgp4$$Iw`r+hlb#Nl?oFA*@sy#jc~Zc~s;P%s19T z$6OWc-S-)r>cxApGKh=1Iq{>3B{W<0BDLIB#T129HtBMj@6Vjb&!&e!?#x)okJNxK zZ0TL{Q(E738`bX3ghoLy8cKI?%Oq8ruI@~QxrcTAcEMAr(V94UnA9WaRxlgV&J@LC>02eNF;QYJ-+?o9pCer8eyU7&E*O;}l1E4MY z424~SaBI~CtlzdDr5-nMLG~=Ndq{9PQ|zg1AH&B6)j-}Tj9)z8!rhN#p+4D}cYL!Pm+`p(F&vHJBf*p|Ts|%8poi>*AQZQrPat+qo+LdQ- zGUZQ7J=yC@2CAd=k=*wrt-V%FStgda7Vv;*@eA5BbpYS6s;}F#Lzp?25dSAEZcflxcKa?mdaczqc z=S^2)x7gAppK?DN;P&g67_Q%&gHL56ygZFI>+PU(+7mE*;8#5ORSGTr zND7~9$)~;4`FzY>GzA67ctJDg0wSlZMd`1ZqJ9!C zc#nt{l&fNS)4E+yU7U|E=3185q>dwWlNMKhj7HO(A>4Yp4N5;!5i!VtyEneY<^}P% zy}JZCFPC8lT}G;*sFg38ghA21==bC!&f1LOON!Eb%v+wDCyhkH>3D3AbLXRdH*-_l zFT~p9;dIbAL9U6GSLTsS890)9s_Q zxUsU4O#7T8pAqr!oWB9-IX$s(dKZ{4YJ#(?2?ss5LHGnE>^_xGcg>AxymPd0|5KsG z+-D}1Yf3FK?EN$7Jj#cWca7jw|6Z^=6Hj@+UXZ-TeMtYghW-;yKtufn4j=w!w*LL1PyW20*o8Cp(no~`e{uFb9SpmD*n)9oZ-TB zL+GaZgXqpKSof)bi{1*p`DP*q6g$GM%>j)^^`NqGrSSIgKbmXvm7cY4N3f<3wUm7o z`cFDTJ~cbAd0GqxsMQL(psx6w zfAPUQOAl6Z^}sflz|MVNI{pzI<&ktdVm!Z-DPc{+JodEQjf0A_ctA)TJlDkF&HR_h zFy9QVN;&x6u%tf|&r|mC!?=B4hkZwD@{2upY4a2pbZPR%^w+UspY#n6E#A(&VjzV% z=Wv7GTQ=9JqOl)tp*}>BYYL38e$6)AR&vL(E?qg^)Qb;g7;)K}M7YLWg~_g=>^fGT zcOS^c0|QI!-KL2?s#4S|(f~U?iuclgDfqav8;-bC;h@tD)|9>smthBC+J`v*a2KxF zT|@rEweaYgA62irN`9Y*;=s|BaQj+~;2zREqxu}QTspA0$1#L2T1cCnRr%u1-?$~# z!=0zS!bSHJg2`)UYfg==){kXT!J#^OqI8hXD}LHO#USmZLC``_q;#EWlH z*Y6%)xKD%ai$GlMI#TR~IYG*&J1@S}lZ}gy)2#Q3tTbXAZ<(=ySDe|;I>+juX7m^u z#`m$)>=G7v_;PGg5tf@C5WL?O!pt!kKi7xB?~XTQE;}M_+g)6r_l^>Md$H!lZrnI5 zfUl~2hTnE|T*%k}nJ9CbcKfvOIpqho?hW8k6Qeu#;s3=So2!caqCFHJmmGuea4{+L zEPzk*Dm>pVBlgGTxytW24*VH`q+ng%q1}Vct?nXX{5#wlCe{&?15oQSo}WHx!*in( z=&40q`bdL6U9O`wcI_xRFV9QzcOpt-2y6Q0L0X{-g$}E67J4FV}drOu2aKF$9W|GzL{3_ zI7S!dCGw!V?{Ugw6|HP~C6tR?)~A-!==9T+;Yk*O$UCuiY3cwT*k7DT3OquI#_s zM(j6a!ee75P1cdnmGh_IogRx3*No7w!U^6_>x4wtL$uj#KWc55idFAh#CPc{q+A;* zB(@B*__F>98Oh&5L*D_c_INO_-P!}sl#g5Po1O>1N2}R(%`lD?bFobSQEa!a9DnCJ zZj5V*sK3Lm@|c7|&Rrv7Qv$9g{|i8;p81Dp7;=6TMGAJ4H0%ecL%3tP$eB!#of zA%E^A>Z`K3wIdeo3;iuDZNmh&`j>)DSQV{UQN&wD1WPuY^Ogkcljob3`{8$^oW64# z0w&w>&<#DfJ&kd(R0fjje)zTClkYp9;t88Q*{DDkA0=7X^6~^GKHDP6)H9bY)B>-=gr6D+LicG z+!YpDzhQJ;1G|Q7AnR~foY}dR%wvoNudGr0DYK~aY=!hzmITWwNG?Qv!=vCxZi_Ku zk7Gl{{!j`nNH{JuO$`}dtQSuzv+vTlZ&s*q8;;v|>_pw$1q19n`L}Bkw)Z+}Iek(p zYLd%nuE{B3g|8b&OyA5;wC-bR&L|$_SBzElZ)ngRKMv8ZM%TEVG^ZgL=CQfFq}oFA zTx3fW_b%s(Lub%%=?89P@8i{ip(&)3p3+&X1$fHI{!a`_cDzLg z_SBP0%}#jMF2%&RHay4d1e&!2_=(&jv`w6k)?fksKfOc#76VbUH0Fm-_ww?yvFxrJ z!Xv+Y!d|rxbf9JgzjbrqSiKO`y)nWt{r>zyGMJTpDp~H^CRQ510MGxmA*8nf9M5&SHGh9{oE=2r>$UhZ7=lRhtjFjhp9H?92ySkaCMhxaIiY>5OeUNX=lj7 z%#u>7DtBYEM)#k?Zz_kXdE7Upg-9_QjvTU4q z*R8Y8t#Jyb(fC2$pXK;*_9!kZcftCMMD&lo1w*xNs1mi~!Do}vY;>M<5_|EJ{wMJ! zqD)v2wOwd=*eJA&$snWo#lo?2|<~j>&jqTP&<*2!tI(}f9=@sb1-mzXH1hLsRKBm^>p{D? za*P#JZ{5I+cL#a;d3S!Kcbop`KcUBq9?pW)wLpqJZAk_5`oQ|b7va~wUT}Hji=GDyFrZ)sMcH>iB0?Hc|8B2&da5S^68^4JXXvPwv7uW8AT_S7BwZ4rbhbw_b|N- z*5FjlW85e22f9q}$D3X}fp*;-*suOgOOxl}W1ro$uru3pE)GhI2$${$bLWcP8(lCt=6Qpr}rhcW9W-o~x; zkz}XP3stK=V0-jobeZ!B9^Y~?t3i`aIKBl!1EDb|O4txlE7om2I5_W3=Ng}_wwL*z z_;bVk9`cJ$p~n6ua@7(9-<|PD>{9^Wp88@>x(3mn`g}dmm}FfeJM-d2lSGbUPB_i` zRzP3P*7C4vQ~CD2HfVY8gnK@tNG7-Qd*u#gD7tIGIxc6CAl!kvnGvo#OrfVPmUPJF zv(SJ44W#Q`#h>=0SQ0#huTGce?|R7y-EBt0`i>MD&rKJ0J9f}!O62^iT(N#L#{h#s zD#}YH*-Q0w=JIAV*M=niKgfo5tSmuQj5QaYH{jCg-MH?Ni)8An!@SP^ z4vJsQqf_%iM@nxAY8Tu(E_fe0W5X2c?LERCNYd#u{kVkf06CH{u-F@U= z2W(n7A5&vbkq|nQI!Xd~?IU@fTX+Nx(NFRF@lp(_(CF&w-qXrY{kdfH%psDMC%W;X zX-2$qyDnNaMo5}XTlquMI~*+%8I~8?$W;u5Ii3!bmRokSdPt+=K4VJ$1Ue-fK&Ep}Qb#`vHq6)NnYT4~aQ8jvr;>pk zA|qz?;x5N#Y~XJ3_u#vruSl2rK+f?S=FJ}`aSf4`1if(N0iyq7@ZG%}U8u^dchm{; zo|nnQ>@yb4{E9N^-Du2|L$bDqh7IKN%meQ$X1-WU5FlBKydh`K@ z?Xlp{1v30M(hfDESA^!i#K4Fu{HxT(q(O;PGEf$$?^?63-4pbdZlT%h_mf1yoYR$? z`12@r<@k3-uWF+UhzUL=#vXZlpH6;tZ-*fvo z3$Dph;`@Jd@z8lYYc0uTpC4fe%-_V9z8A9h$%&|KiNHdKR=DgLEwO9qfGb<^`T{LjkXb+Q9Tv5lnRZ;jQs{{ILFo!<()N*Iidr9S7ru zS0;w~gp<^vdg>dQO{Yw=$z}Rb{vH2V$S7|n&+7_o_b>rJ+tgrpzklcc_@CG_#z{)* zbVM-bk1L?`Xa_P;m6aI1eb2U9TX|(-A}4QDmZ-Q*lWen{$XB<&!`X=+5h#?vHhwQm z;u5fF;R61xKc0pCCTwoJl|q&}_7jzLh-;rg+f$f!rw>j1I}1SrR$-yG58DrU z%;TMO$bVZmm`hLOcIBx&^I;Ir_*}t*)Q1dk=N z$wa@1ey_Em4-s*2OETogzw2n4UJ(xbHsLP2r=mSE4-ckju-Pr7llC(ZOEBI7{m=EQcP(%a6P1@r?o0SCEj( zmHuK*97dXLw=i{dCNxc*_^DbJ%UY_zBw`O)4IV)OE}Q9xgACcUnbGiyO0IZ4kInRl z@;HaZJowmrUUagUH#^OS{h50dmoZl4%lvWezZRIPgPnF)2;XiD#jV!&*s!oKLJU;l zbwU&C?OI^eSi|cJzOvENbL{@#br!lk;d=>Iu;~r@yYPgt-_D$$7LQ~-qou62$qD5Z z<}52RFJW>c@xy04>L!IF%6%xy4sYYKu7ae#F_P+2Zcyz_1&kgLL1R4E(|?Wm$dy|{ zdD}G*Gu?hG}yK4bGz|m?)B*m|7|mroSMIthc8LvUc+PArbq_!em%kji{*T=-jyoWjE1Lg zA?jz>3c+rda58WYCtb#^XcExDQQ!|?c-tZ#S>nSrfP zz8=FlBV0M|L;?33(7`V?d(p=2p~7V!BU-TTm)Ij5%NLH9^ZU7G656CJaY@hS^YS*Z zeLWoBMm4zVI|P)I4BgS)Iiv0q(le)VhT(or2%b$N+(T%bSzTuwxU%>^AuHd2c1aoW zaaVtixpJDld@s_s=sk!G>(7_eS8(6kYW!OJFgsejqhD14*!*J%`#jrLB*!h!UXI=1G_J93f z->Qqy?T8f1mrrMh`{p!pz!)ky6i-(-bd$_+A14W%vs}{mS*qYv?jr=OwV+U6B|h$= z#&OayyiKD7H$S!G-u!(m{mT^Z<+SKcz0*P$Gh3dAve6S&}vaZ0`o5!J0*Ov#3$fGYh;p8)|fD)D# zVqx%Q?9xAvplp474NDaUJ{(CY|8sJMFJlc^}7*%wTvr$ zxgy_@qiw|Zb8KLr6IR&!?m0Ez%|(v(3#4Wr7P){M(Cn8^ar=^xany|$njFHpCkLtb zOATIR{vIw`ihM!rCln@IVa)mA9CKta+iN6Y@b7sToiAcU*}8bx8UhDdv1Zd2&qw!X zXm+fo==cg6uq4E?B4c)E-elrB9MW~nL1*P~Ps)MXzrCZ?k5oyoc{ub_-lBBpA>4B= z!_r!HjvaLYnr}Dq%Mbco(&~U~vNo7Exdpdm{`c;YyQPbhKWtgIxL(ZBS8~342lgZn z$kifKY7mF?%45Vuka|tpEA7q(;x+1nC%(F^>~n7!Bozp2ax z(W7|DBQ@L#?Z+p#2XfCJ4=CttAez>RoT$`c zOvy6uWZt4!j!Dbp*rop(c9l)Rr0yzwYS~MS>GJ~t1)>+jIs}443M@bDd6JFV5la2U{@Le3?-2bQoW`ncjJhJ$}c*Zm1H+e^$VreW#Ih?FgT18^TNX zkAdp`xtuERz`~1exEGm1g}YR#_{s!q33f!(k>!a0Hkt!(+p)Y}J};WGmRCEi;V(0j z$!f0}dybLif2mvPkpIfzm+_1y1Fp1;;bC`MJL9fopEb^G%ST?f zL_Rr5;IX&6@|Bz?uv+iPcmJEsZFQM+qUsJlYisg?!B>&EPm!1VeIdo7edv)G#rjEW zSuV>En^w)lSTzkUh#JD%&Nd=lWe()(mr!As@d)^ljhs)vFwvkJeolCW?f=PgyIuzg zuW!RW$BHfGdUI25ERWAHX8*tLNW9jU4+ToWJ94C`#p+_v`@fyfTKh1z(|=m5`HWUO z86l%djCBcer9M_Y6Vc&S>h=ei~)(v{}*5jC=dXqT<6YS=C5$c31!BUh{_LYyhauCJ@b=fe`ID*Whgd7_*{GB7^k~K>-{V+2 z^#J*ceG%ErL`+)TgRceL;R46A+*&8YhZ+@8W0xVEedxs@?Wr(NJjgat&oR9F0vZ#u znl9)Z6mCjvEH`P@4__a(l8?`y$K#*W;cLW7zOKF!!x9>WGwZKmMpzNt)26}CSr$)i z^J(3j65?!a+*i+~yqs|1$^$iS7`+$k4=p2$Ru8<vZOR*K4p(>nXmqWkX+N{Wg9YhKu2a$b2WV7$WcY-Mx^;hIw?`Mk#|n z#}#9y(Ih@~Wigw-&ZEo~%kg`#KQ4{X;>8=~Ib}{c>XNH)e9U_kd6we-wV}LohcP#Z zI!x5?>%#I3P0Op=_bsj;$*2Ck_2F};kPh54=PbvcVo&6+(0b(?)ojcq*}+#tkKR;h zJnYWh`zrA}qiFm%>&92bngVc1(kNxof^=(&r7iv4g&uV6U|zW=3v!Ef-};yrbruVeXm-YuFt zSNzzkLs)sk0ZK1dW8%gD;pd{RJV{UVn0#psSi4D5eg@;+L>s)VuB99oPjvKm=l%ZCFi@4D-3OWwo3;#p zy!wio_FJr9*o-FCWxTe&H+yaB%{l&NbR=>*Di`*G4o8vQ>^z)Ty^K{N59%=Z5Sgs` zBDf|73X4S6Yf<-R>R-Jc{XS=*_wAds-*hm!l@&qf=VZtgOvPG_d&urKh1WbtV)s=w z@VP3F@ZY0^$QypVJYg+|&JyPX9L(>Gb2C4OVrN4v^vw6*n$rvlnGntW)h2Qm-iwBm zEjTc37gW;kqr$p^R+Xxda+xy8&5edif*JBYWRRRkt>{CZ#pPZDd1jAY82V#2&Tk0D zQ>o?L`}kI#c|He~^X>6I&jo6R@#Jb%hOuA6P=_frYp(|0ArQ$E6$t=r8uMY_jDd0`vJEN?C|Ph3O0lp~5(w+n}4qMoh(*n2Q zrtU^|o!5pbU!TIET=FIw!ZD6wqW%3-4us85s_k8buvK(w)NIeRxR4>||qY1QbGo{J6V z`gcD29g{Ng{+LbPZ@N&xY$rAjK7{3shlo}mqgNX)V?&cN$Cw^NX}26CD;J?*iYdQ3 zp~C3{eX&J|gZ)Pf#11gP_A7ySx@irT}nLqPjt)Rbf%QT z`>g*7z1R2)%O7W2hW81<#+Xwu{tP`wO6-dahhR>2@jO;!Fl}?_J z44;6h$0DKPFbS4B_2GA9F|LkIN5tAtELj}`71ua?XsE^SX~WQ^pD}+e$mN*H<`_Lm z2ch=s(c-v?2HB28)rXS^Z%cRPy0O3nL%eIS?fU>e8}r$liS~%l=)>ia)(rHgpt1MSlC-l5ChR zYQ}s&HTJW7gk_OKFzI79DZ8zr3n6cC>ykE)nA(F6cGy6jM6Q3^J{;>A$oppba!5lJ z^hJipO~oD`{yW z<033Kr{d~E4Wv$xqO?n6adZ0)I`~BsRjSr7DVvUr_l30E_7}VgMZI0r_p=Yg;Fy@V zzhO^o8*5B;bDj$Afscj7*5~L?t|2s67~}W(&FFYp4*dz65%D__({xv(^hGTBR!+j! zagQm(us^w+x()?hCH~+b$340$^DeU#^a`(_h^{5DFRBy!(8u7Gdx9-4T<2Z$4so8z z1$37?gS)nQXn5$$KQ_B@%g?_UT)C3_{944cQEV}y#?8S`OxT^kI^rEu&il3)>t^8|1e$jkXnV;BI~%T zbQ&}|a-kGU$DJ-us~L9gf5!AKgg;KJcEEon(Z7HvBCgua(* z!Y#atdUvZLi(zZIG5a%Hr%OwA{JDY&haZud&2i|Y7b8MN%z*~w<3z+cY-yF}Qw9xK zabPpn-^xXrnmZl`wV>SZF1*Fu!Rfp&&MMu5o8fGB|I?HClMY4w} zc>}sdvYwIfsNS_wF?I zMVauYA(u8QYGS+79`YN~4`YL>Y2WD))HVMU&AW05+bz^s_G~V0ww|J=ua4o^oxh+s zYjGaQDyV9iuuZ-?wir$7jCH@i8}R+QD>(PmkQeS-$GR`fJNLp~%Ou|EGmH%*E0N^# zk(%r)Iz7sJtGuyd=L1+|=yulZT1v9Q(C2!{c@%}Cb7XO(;T+UVjB)F006n{Wl{6x2 zk+&uke)6%{@>s3&e;!Y~Y3cHEG;cX%%7gM1`FWcY-UWq%%7dWV#hnyac2Ls%%?NO` z;mIeAINu-);U??o{w;x&4vKm1p`D!V5XzemhVi+r8klxJ0Bh_gA+Nfg_N)G{UjKM; z7HPj3jMa&|@qKNH$f*6r``1%A<+=@;FYlxKMXyMqEP!=SOG{2&U-tj-x969v$HiAC z@y~8BKNh)&E295i=FnBNxUc2GeWWBG=XRAme7Fz4A0;8?;Aeck^$~M-iZg~@%d%Ii z8_V34=HPiL$axjoy9#qow<3MmZ#sYZ!KKW`J1pz1okfrLvYo2elKbcK2HMpmk5H7 zAA6YMmh?n%LLN?9=<~=zXMSyU3U)nTAlmL3?ucHdw7o&_?Jdn}PZvR_Vi31!Y~!-u z(x{IciO&tn*mCJRP3iukQ%m^2+@~}CJU1fMqh(Y+Dh+*_uVZz%A}$vmr?g!cSy)-k zS0`ukUE>a@*rnrASQumtdhxhVZIIsO%WE{h!*lzAJD|9@E=!J^hXz4K;Vv{@H)8Cb(b%=W0NSJVd4iS<^HN_d(L4>W6>H#i{56i=&d2T2 zVHhua9u6qvSrs0o}a&-BvA*8A{R(c(T8ZzACID0BVOp~$h6?||`+FPKR|JZ^TqPKGt;MRjqEE~z zge0|7#69F8GH;Ful@|LR#t+%Z{R}dnjpT!}v-$jSM@$!E=-k*UOb=XxRaFd&mnG7WL6?!0qs4^2l{FsdlC>NZ;HF3J{{D*7cV<2#upWP@+$YIkC8J; zUv~nZ%w~>SsK^!P>Zxi~ETq5Q#GIH8Xa;~QvNbtM-w_@2Ao@PnQquk2*b}%KG4}g# z+U_|Gbk@Uui91T-y}6@W{Qd+he!6?5aKh&d?Tgc4)tm+5Y=L7~*dnrnQjbNRzZlz3 zMxxt}`xtTLCEUt%_^DenCfJVTC12G!?&~xhjR?S|pqYqU7%1MyBRgZblJXN$d3=G^ zDvZIi!Q!kd8?g`5UW|;@KFGaPfP@K`;2fuky&>=EZ`N9RdS?KJ-1ma6??4Rn4-@Zs zZSFX3%pX7QLbcXq%z68j`qVp!?5hjP?%U&&%o6C0GUS00Wv)&Aj-^*mp?2agGP{m=61j-Sc#C`l?`g!t^p~hskAlnA6{vjd2j8-OnE4|QZ9CNYp>I6$b+e&2 z>j<(#Tgg9b08Cro(rb?xl#98LeEAHdI;LQ(?FLBR>hY#An`!v9lVoqHfHR^W+V3BNa#BL zIB_4Q`#r+CGjZ6yxtZQ%I?%7dlWEuw2!`5~aO|Z{AxSeV%l{Nwz8LdQkX`Qr>yg)} z&BPkd+T~#SY9z0lQh=MO14-`rNFivPgS6C>=yC|$Xw#LYgMQQP?SJWShYnVSOG*4+M{}Qc ze>AU+;9uLtd2y+4c+{3qnjhvVIA`9rbP(s+yyHAny>sAC#}j#+Y)?txnrnQq`Y?`n zH;1mX7lx&+LA%#vA$QD9i<$HN(IB5nyH?MkXSOV%Wg{*j)2@w{2c=n#ot8lvzs`_K zn<{;@6Z_0J^;qTRd&nj?qwb>-#s~HxFL9Qh*M=|LJVsS=zVgo11@YVdJD`sO9@UTiE`T>OI@1A_QVoGa2YZwf()mO_*7aN*_kQu=)&QS3Jn zJ6x;4vCyw{x+F^%;a!2U>na_zmxYUG~&^0Gg9v_D;W1^6mqso#6D&wq7Bbb z+Ld9F?{a-4LsbUwx&L48In%M&;%~P@kUb|)Qgd5reQ^=LOPa*WzkZO@n^b6f-z2+) zO_bo1MnA{yr41|JP{}kiKxLaVYnF;)v~M7LbR*2t2zsgvR&E7s_1+ zNkZ)xOEPc$`G4}yX8T@}7{5U79iEJFnP2g$_A*}{az*rIEycl1(brO3$$DdVP{|HQ z`hD{tsoso+QqU%PeEc6B^b~8X4Rd(V@;1avn_*2_165|crkjpsJfg=bemQm=KMB|k z%Wc89_U16&U+ss7mBXpFaRoUjbg0pEi9Z{ zL@S5)!pDSC(OYyIURy?SbI2-0*$?2J0~0tU{TRFkyudTNHrUL{$GEs@{8M!(FEEy) znD%Jg-8h<$y?>4kTQoT^svo!XOQFj#B^YJBi(kDfqaItw(0+-?#XXB9`+ZUfjW%YJ zbu;j&ISZ-xtD$YvpRRmWz*NbQ3IFU-fi_1y5R)aEJRPm*;Cpk`z5EiN4r#?SgAzNBUweK3K zmvJFol)EZito25`*FU^dkD=WsZIIhD8`iC=Sgm>pGutP#L2;pwrQ8k2TbE*iX*|vG zy+C?*J@7Zl0g*itpnL8xo_&{LkNtD;tVGNU11p5twLNI-$5vWbe-n}0oH#aWHpjIV z39IK34K7f_s#pCeD$J2)T7t%Se?s5!o7l@!>{Gk_p-)98XqS45P!TJO5rfVXJqqqT z4~@(1aiePs25X60RlF^q%?{%Cu~{@AG>H1rFvPU!GM@PIoy?E?E%Kyg-`Zneznm}n4emHhr=#)gudNCda9G)FtR&8n{$8;iTA;f zQGZ3A<_{ftvI(cZIO6FGPgb;2kR-k6B{4O~qNypx^!8FT?jLwb>#I`ni{9X8hXvVg z4y3kKKCysn7+ z%)56~8o8D<`^&>~-$=%>IecYVFe~dl#$3N+m}#*EW$m|ww3u^rd3gy`j@FXh>K3al8$=JF_z(?2VpqsrpDhdx!}!=ow#GH!zC}Z zusEoN_PjOftjoVi4cGpk`%!09{yinlq<?>h{w}wBY5b(ssI1@_G_3SJU2Ab z?zgfin%4lg4;Qdh{U~y>=AxU!8>;TEimtDQ@P{|L{8c+k)W^^A$jx<_Zu?@6x@uCt9^RwaqpD9Y~M`M4@b+S?jL;L?n)tkpt^?vXFGG)$K z#z+$~#Mx`DlSC>hQj{c-MvYQPN`qM;5ecEpsc29_iBw913?&Wbs5BR%@pr$y9^c>N z@%f)~_TJ~-_psJ=Jug<%HBzaD;_OHe{HlWcWoM&BpC(#3*<*f{9riU|g&1X546S~I zcXab{?TKr6Wv?YZTiyp8)AXEZlff=0BQPx=g`aYK5r@i<7ng#TL9;Q+{33iFZGa2= z)}p+FHgb$bCc`*QBTv^rZk8ODwaDP=8ZC6PR%P||*Wm4%1Lwsn;PS7#5bdY{+Ka+C z^QETTflGR@Wz{>ddO8N3eLKMBo&&ZnGr+M0tc)UE24nwygMbw?p{k(~9#xITTIQ)% z5L(1|b2j)-!v!Z=#iPN-ApF}O3L|}f?H_zNc!seQE@F&p4~D(IhEsN#;pWrZFv))x7Mh-cyPjKE zee(vGY8j%aTMA|fjmB+NJXfxZ}E~y14 zM^7B^UyaiA5j_2SU`ESb$jnWFWj^b$IA9MRQgVl?J0C%6mpr2CMZDBtOnWo>(c%7Q z*zEIx`8b5>_L4xH|KkAWR`}x5$vfCPoPm{wRyZm75!S~kQdmYhntwGRMHM-kuJaVb zuB74bV{7oV+6eqn8Hl{|h&5w3qj7~L+smy(4TT$UCrB2@T?xe!Q!AW)o{vGJj-isp zYLxrd3o);Bz(vZP`{XeNwF^dJ(3X>6%l4xK9$C1NaVX!M3&W#LyKst04J_;pM)Sp+ zQBc+c+xJ9bYkeiwimk(y3xaSyV_%qFS&lVAj}Y2lq2^zvYZ!>eB#CJJ#-7*cs_S^# zrVw?LEbvQI0D5HlmW@1_B$h!(}d znXzao9f&*5FT@|^qWJ2$Jl3h)hqmHFsOVgTmdPG?Ubw_ z!LYye&96>DT7(h=1wDjye%@HQe-Re$`39A5=Hkx62o!dYhCeC$ zco%QFphV75PT8P?8~t98>~4IU@E3Lr$N(#Cp7C%6UqJvr|B1C znQrsJ@IHOR(yeH7jzh&{dE$%Tro1Q8{H^KB=%ls}9C&JuV_9vpLRuWoPnpG=8d}WT zdiouBu{z8Rbt~9(yAXbtzv4_Mm_wgv8|?Wq8vKI8LAApW*Br{jD-u!ocPN+2?eyrZ ze*|_#j3bBH2eIvb0kmpv8GfDz^ovYNKP1A=x+~D3E(mFMm5}W#iA$vpz$xaZT#&aB zMvn0VstuzoQyop*x9 zt6P(+yApZ!9pu(D9p;THp3F^~77zNHSCC9h1huOqVs54e7@O_j=D)}S)x(Cks%Rp{ zTLe_Be`x=rQnSY+mbjD$-5wV zEdf@q^o4s1*z;iT>COFBbXs*h>BVZJag-E3RPBJq^sRV%lnA|K{J6_)W_-m5#RN_Y zWFPej*W^9L*G8|=M$`?pb{)p*)j>2%CWbnrLvUSZCE6=q##!~#>1skd7E2^U8!gAX zo%bQQe%+}<0TYp+6QRyj;J8_(&^odg52 zp4c4`!&NR;=MF?yp@VV(j*E{N-uDWm8H37~CN8vbEdTY&pJd&?`~Tmc{O`V!^PMQi zr3h{_cG4>5m-h6UK}zvwaNHvvS$_LW(g77@z&N58?;Hn{5Bbo%`wwVVW>JY#%Z zuv>Hr8aF9{9Df-nbIWV^b7r^>g5Z4#+Ihd0R>ypxp80&TxDrI$cA3$he`_f*>mtgV zeu2CjJd#^zj=MiS=c;t(LY2W$?xw3ew_1hOc->9u!VV3Z*eOp!{7G2+XB)RYQ;hta z^3k+W3_Cks5BG)oc1c6U_Ai)_cMbepqhPnDFu$(uEYlU1 zrg_Gh%>D40CzSodBscg3wYKJxykH0Z*%`{oZ%H)ioqLvdbxjoUTdPTi^|JWwk6>NY znc?dbKO#!Ezn+CPpU-nbb987{Ks>z^J3{-d+tGUZJD9$&1x(YV`3*gR{2QN7^2LXW zaL_av8gDa}s&^T!Ud0Oh;rUcQPm;XjMd{~)Q2d=PO>J?^pOnYAqo0B)w;`Ak?rTxQ zGscRsO~Wa*%COk!JMUMiJ;Vnxj@bTj8SX=S|5j2*@MDbo z5x}*coQ^BjRlwQ)B5q_rC(hVA8M=cNcq`6pVUyqmP;M&W`KF2T-6z!2kww4pf&S=G;lDYcuU4HRQ=1FwOoWs&nS^E(`|))%tpOWBZkLT z_s4nA6*CFPnLNaRz8MfZxDu8R{@}gcyqqgp!a=9xv*9`7fA7=6otiw6C3|q+>Z3T_ zF9y5jZ=qKvl4S4w7H;nz!&PHG@5i+|lQ(`M_-~#H7(ZPJg)R|rYyVQX6FeK7ZZ?6O z{d;ask_FG9Z$GamD4h3oe>0~zb`7HXEv$PkMtcL?$ou4fd~w0u!T2*+pT_KUp!kjc ze5EaVeD|mr5)O32TiZQwvi^E>&2Go%?Wa*R?kQYeGl$IXxzfAwy;zp#L^1C>X?I;5 z6$~92UPDeap67Kc%JQd1mbE(>G0>uh419-L$gTB@FGb*n@TgTh@+R^Gp^U~0w=K07F%ae zg1`Bp#zrzm#2fE|2OjCeF9#DUT-b)2f*8-QH^k)JxFnP1;`+R)VrH~y9n+OpUZsu$ z`%$#ZkWR15#l#2Y=^r! z+$)hyHL5xE$>a!ul`w7p7Do3J){;2ukEU6L;1w-v$P`i_UG3$Jv6)DlVvNvFgR$wj zI;;uv$C0lDDEEjR`Q()1L%T&Nwp0#RY>>s@>Z?%o-VE%RoI(u)S4q~mn0C24!PwBp z*dfwMwXPz#zWy8B9v_ZpmCEqRS$!JW9Z0I-_i;saI5qDzAnW!B=#fd_%-l9Xhea~H zecz+RJgQM{rD>!4XqkK| zSPsUaQ1jE_cl+6^7%m~-3oq`CN8Y$|klVcq?o>8H!$~#n@xxe1zuQgwNs+IrHAq5# z)2kk2p~prJn@(+kmkybD%TX1N%2mNoTp{?)nS~iXAt)*rL_$N6 zlq#-9Dgl>Jo9R>yKWXCDdqSwSVJ$Q+$lz6ux&qo=5*T?xokBZqfXkKBXnH=J&L&Dw z(vR<`GBX4H)ThCp#|>!kz8!1)ThMEc6+QWsNbdKi5GhQca~Hiy(>Rlsmpju>m&3Sf zK{D<5_l(eDJU{7r7`-cMC%)z_8Z>R7Im)l;#-1E{kX=E&r(aRG`c2X;S%NZq-(dg4 zP|E(84|f>*S8kRbirGif!GgzhYTY4vTs#|&P2q!<(n5InM3aO#5xVS_PtAf^2*1@S za;q%~>sX?JIO7?Z06my*j;qxsLrUr|80AtAHO4`Y*rXe{;|9 ze7z+d=QG`DPTmDfxU~egx<=s3lziCkuFAjh;}D%F)FMr$2w)Ty_NUdIGc`wBr)e><5*2F{|3g3q}3M{jdq&)()1mNAdH?+uceDZr0g zs02~>=D}FS**N%N7j9FOLlx^9Ztu!lv~wb#dZ#~w!W;>7F1m>6`oVN`S}4<8AID&Y zHvBC@MD8Pqw<8&1e??M+x&Zx63I}_)0dzYw8j98Uu;s4~)*Stea?#>+`??94FBnhT zN`9cfBH&sVqHo76=yhT;j53vmnQufn)f`)Tki41{6%XP&oCeG9e8RgmT(-agUovK3u0kuuoe_Zj=NfrKE>BHvoMYNlwx=;ZFO0fp6)|JaF*GWyz;n-! zP^f$m4lMHHhoO@gN>MEJ}B-e4yD1X(V6Nj8BUEDRSOb^h_%R zQ%gBYXv(2$pI(vqrbVPXuO7WacHlL`8oYXGB%*-B1KWPFTewT_M@dK4* zALBIxS!xs>MKU!Xu&4eKrVPZQ#<}-!<3~7;W4x2^P7_EWy9JxQ2GK#k3$`fdqQIFH z+#vZDDtE`wh}H=-VDE(Y#|!Z#4gb;azBt+)sl#~&TXI#`2RWM+1>79PE%zQNaoggs*z&*E7h^M zV%{rmRl{L2eA`K{=2w#4mzChvu?~K|(WZn~il~rtpIhGn+`T=;n7J^S@`uW3uXZAc z^(m9t$4)Z(Je^j*P$RDbT@1S%fzg}ukUOD62WO;VyRX+{Tz9-)^<16+=ufjyf&=<_^T8oO19gfo@!Q_?H&eX2?aSWeydgDJ43 zeF8aT=1}dJjVQfx310n#!|&`z%^di4;Vf<{%BK9>1ZwsR!WR#EVPcjESP0kQeXUQh zKw=I`jkhB0)%!quvo-gtN`w1zLKA1d*Ti`>%TVNb8`mo3h^=PsBwId!`qhu&&N^38AUpsaTSJ+%*2f;6_}qRN~HN>c>J}D(d6joW-u|h3&(Fy4nLe!giVliL zFloUxF7H<`|G<&4{BbGK_}bWpld)(;OT83&rJRO>TFc4$vK{HD_M`sl9Nh8N2qiwg zWcu0B{J*hbBr-*mW;TSGHAF9}DA_u}~LM$~G&g_c*nMPZ5mn1Q=K z{-C;hm*|n{8nh@@qWUC7dgl?1USm0sdj6PqsAU7U%BUA)m|jb1fe!UX+(nxcu~=a8 z5obil(Q(IhsCZMDSF)@cZGTKd%l2Y!MF3krkFOxpom}RFK+>7WJa~pa#VXFU0o#t2o6^m?p<1;QRfKIC?ze zO@%X!&EH38qI!h(n*2hKLTAeQqDlMfrKsC>9xmQHhAUcc2-zl|IJuIs;QYQA5_b22 z*AuD#f6o5j92D`8AkC1~0T)`4=#@^~o*Oef26l(CT)Dt3>e?+qx)a7x-0o>q-Q`0* z$2_UIBAO2SSkm<=ZZyX#ka#nd=)nCgbe3fo?YkL?dWs{+G5;Q}?~li!zJ&uZvgIf2I86Xx|?!DTOPsNqyO>fgLV2Wvbb^I;z>I(`w?tE+;v zZZSsxl%vn`_ef%f4*h(!o5o#=!k2n3)ZTK0!fln%aeO-YFMUq0Q1ekXCb9ZoqI40gRwSt1WX37DsN>)aVO+88A4WR* zkx=R_u#2q13pwHxqTPyLmqlQUQyn&S6yW;c>2Ut>6ZpvbrPTw3Au2q0FZC1TtSeOuO2Ovn>rt^fTk82`_FfKz zUgH@!@=oIkU1FI!tX9%#N;Q>>xiR~Lc~VueAh0@|5`r$%Mwvt!vJeH+i3g$Q?Ff8( zG7L^7<)QD_8KiMn3CgQ|aAeyd`ujBtuDP;x`j#TJt}}cEN&jGaNau3-kVo(QgFG{yUf44>&-T z+!#n=)|{rCZVI-yqpz&4wZ?KP+#kW?v{za|Y;7g}PAVnwZNFHyU?MgJ6@uN^sUkh7lvKk&yWTCE<6`aE&e_`C{bLUft$qnf^tEr$~ksuZlgiw3t{MVnuiH2tkA zosZZ@PbFSrdTj-Y<&C8Q-MN^1;S0VnbiqIQnzUi&zikKBXz7;MJXds35VB^0rN&AFZtZTt1I9R#b830ZJ(UIFV^aEm4H| z>!m|W@UxUWefr%%Cr8Qf=MCJ&$nTfXTy7q9#VkclT~~S{cZD*JFC^@rNxrKF@j~Kz z__8q{&wkM(3FBmnnfaS^e6dQYL3o$*YgHJ#4B zd-s3lAE}?O&|hvk`D;%iFSniW!&Ml^CWYekj5oM7ZUrSA>mq5>3hqtmZivwE0tv%# zO4)s%?g#B6WvkDa@;n~flDqI#vl?|~jUy9Q=d|}4i*?Sd2D*1H@mFWiT9()IF3Xne zl1*40>Qzxdw=STUso73%BlOE(;y&PTI(s@$*V>rVfVfZ??9xC~_ zvF4&7%`uatyS(*uG89R#@C|Z26*|K59U60_Xxv2$`jfPX<@;FkMoeu6^V>osF{%!q zd=8{>YqBx6p}OXH$I)MKwGyS$IWxwK_mDK&vu;*oLxT) zi|+qKSE)3RGEBlv`lYC6F3g=MbOrZ)d$CJB8J>I8!_=lp_{q%^LRzNbj2V^a=w}0a zhSq{J2(XMuIrP5J52-g_!!(U9o(pzxyl8)D%od@Xeq|Cl7K4J*uJYWyyt(WM7ntyL z4|nO<2d=8r0OCT_>9F@q5-n80w?+1NDODQh23te_lPnW&qoq728x_+o|wk zA|12~$M|pS$$I$`$~iR#WMq6`{XBWxS~Z?A*4?NKuZB*D`dK)i51y857FRq6He0 zXnuGz>{(sj4diVAgFa2*Sauk zxJM%9@L&BWjd8GlS1f?(*U!NM17Z9a7ll9j6>({?J9)oWrKwk+gN52AFmOyn=`0np z*BnWG(MatZS}?Fc0T<`$;rYf4jGG;XZ#D{3bV3r-LpO0&pCq94;B@@zBTgISKBBw0 zEjtHq#J3)|(D}V3eoqd@%XtH66!-wO?oOeNTZ>8Sa3LjqKZ6HORS4mW&GChM2WnbN zQmCFZR(EgZQoq^p_rAXQ@b1Nw&D+)fn)gS(!kuqy=GV62v?EK3$eb|GWp2{P`Kyey5N4SoclFZNW zbNE`G+K=VgD>rcgF>$E=$^A?Xlz7&;0fJIsz2K@_Nsj-@~xKM%`H%ceg;=5+Cm1j;zR;J#T5K~iM& z@cWh)wHWm$k0blPZTP6;JLK$1<;fRq=T0~tp|^}QI6JXuu~&8wvi4V~1n!GKhrO(BQh+wuI4r5Lk) zIp6Je8HFkL;%)A~nwHGsd+>czEEn2!fvc>0!gZ*O;^{0~&#lT#+@_)yxoF0xz2XpsUvrI`j(cI{JI4Pyp?JB!Bp})or13pN@1Jo zewenHkC|`pp{V#biV{1-|5X;q-`}&9&Nvzj|5h)8OR!ru4PV4b;!p)UWP2%+WJNg^ zv~5Azdt=!dJDfUf1W_|jo~yUYGI3zJt{tnQDAQMsR!n5&2buJuKNrgf)Ycg2=19`V^V6amP8h&00^`)Vl=5#GGEANAZe;GWnhOJ;d z))>Co{>Rq)fAyblt9)Vi!E#&^zK*KX93d#n2&UV%$#=I!(*9jP4`Fu|A6u2IawJO72V{Ah&M9Nt}K6G|YFX=H7pcqvu~+QEdMUXmNT$Up9xbK9?IAmQ689)w3`ead8qa z!Tmn@dxlb9aUHHY!?L!wMPuOS!#KM>oHugAV&m5-!Mq$rZGKTx2HCJW`gG=7t-Wvw zwT?}p3&+nuPwjc`P>mep{4b=e&~Qp*x=P7)&ftGT996z#!~B~IG5_%qs83FU>b6bP zY8OJYS9Rk}>lxU`bc;vAzJcMMllVQPnpOvYrSxE9s{9s8g`Rq3ExnaAd7^Y8a0yB> z|1X9v!98nBVfoA^URPi~4lOXDH@nbcO&&}dWTEf#c6WdWb!rZ z#Pi42Q1PghwBd3BTJBv+gXgqJ=C=%nJduRKKE@aGdkpg5rVqb!le9;0V>(~+M%*6B zn=?C|yK2=3T6N=aQRgMNKCl2|&U#U{#3I%Q&7^PI-B@>n!{o#@7-_9TjW1uK+|?*v zN99O-v+6!}S;|sJTrB2^Yh!I|C~j&qB7w=)lzMd+tO;x5_TI|oB<;T9TET@hf1Mkx ztyqqyf~L?LmIeF2u_N$dJ`_D4PvQa}(DQFHCacZka+-?h=YcUaq31s)&c54B(_z|y zTW5`+?6*6xZ%jDL1-y-V7D4F1w7-Knju2^>4o0h;FyHwR<3=ITgkbR1I>9OXnQ;qL zZu4^Lh8X|!G`%UEPoj@lHl2wZjXdW|WZweM7d?iaRjN3}@IJczV!X=j(a^4EopxqV(^BBpjIGh5J2tih*3G*F*(nDusX=oJ?X`=O+zs} zoKTBB_vN3yQiHPr{{gjc%OBdnO6r#;(g(YQ8+K=hmeW! z`$yEN`;PSF&Xe1qC^<3@x<;QpTuWx&hUTG#*;!FsALkf`F|GO)SoCHj zH)+pjoLISt%4aA1AN|jff$`Kk<{NgX<)f3TGWJ~d#2u4AFfY|8I>^r7y&t<#P4PG^ zDs15H*Wet3X!)0nzx8ee& zKUhn&Y&CJR>TE5wK)aid!R?4Y_@oA+Re(4x%#o%EO+0ijF-0rgh4_6RW6lKq8r}>o_aAKew~fM_)JWoHI;uzfLFrS+@Jx0Uuo@v+;xv?U zxCuMP7vPe8m6&to3XBk3LZQn0iT{VA#g}Y2Pmh;8Y2k8{(!fI)nxl=!BI3F0;}_Dc zRna8>-Gh0&N+B3tV|Bk|>jw}yh^x3{p-d>N?BKg!G8yO>*`kj+gtTEaQZ zu7$5DvT(o81k#rH!APDEH~(oK&wa~6o^?#hq-%R!OkRs;p!+m+TE0h(a6$L*ed^5- z;1w^CL8Uo)*d|eO%p_w|HDR zl$6pZ@vl7-=imE#gr=PIp}03EajHlw?Em~5?tNX0@2wWl>J@iU_sJrz>M$Slm!5{T z%%A!}_cYb`^wN#vMP#seG!`nv;bCAokQF89zr%~hwFc4Tns;dU!wu{6mw=J9I4#@f zMt_V$X|uX9em7o$obz7XrEr2~i>9$LIR$5;CU4QBaFHSZZ7A2Yjr>*sy_oU|=a>ZJ zo#sqD?=}m=7p|wo%S&jaW)oTtxIo{T^B|U(2OJ(no22b<+U5~=pi3Veb|}HiQE}Ya zU6&w-)%0K7{mB#JiO_^@8@l)<7o}ZA=vmQ3`uOZPMn#^4kiG}po`z$*QnL+Uuq1fo zyJ#oKI6o6?`cvSRswW;0E61a1AK_qp7LT{)HmJ8<_h$V2qY1?XpP^~87o8K8;@giB<$pdgnOb%{z`uKM!u>`G`mx25rm;Qy z*Q!0}(Q3qMM=Gd)W{_(?HW8iYyg~Dyf3Vlj5_fg>a|2E3)U-90Hq@w+^`ZI0xV8KZ44qic5qr*BQ^jZ>cZhMbic!RWEz@V82f1`3%J+ND z`jJh*%Nfh~k!Rt>z5n>U|8M`3Yw85aM*5f&nuB#7Bk}V0D1@G5Onh_*#j>YjZ%Y(d zG?u~Nb<=Si(}%xfc}6ixVx(+{7r1gS2}q!HF(u$S+~ zMeQx*X-qweo2r!QiOp!r>Xm{bTMn%y{n5)?46YSi^QobEv2B z9aRmMp$5x?@wu^$qAlK{#RQf|{Q4LfwEAGCSOqMJ%7?&yIp)7q#0hF@SfH_s8(+2+ z?(i?8UScX+2U)EqQxtjkM$y2(d#It_1U8|waBt-Sw7#Q(+N=J{WpRCx!aH+E0XEL# zz(p{E`)-p>tL%(PJy@Cs50-P*qL*Mg;{|P~}?!8qvaF!u8Um@i?5 zS%XU0FDHYgo2J1HEotP0;!)~rCax3a;gqPz;qhvi8OU3EXEQ9BuTB#0i&(~_7oN>? z#Vam)!@t4E!JnA8RuX*o#WoZm*bMAooD8%pO-Y7GA%<%UXN+|G>mPEivGcD9GK7#0knB92< zibUcJ{<|?D{dD&$+ zV`B~SMuAW)`WqhhXz)Mk%;rlsjN{MOWZsyjmyj_v4VK=ir+b-w#SXqDLSN z=0-E0=LsBhEFRw%^5wccZ!6f^;&>!=Fd+9X= zy)zCVcWLtf+5gy@gy6JbJJR?bK|^!b!;n%vY}loSmX?B4c_9@ZkGRJ4%2R(DO{vyL>B~7RH*5!e0*gcwsH*au=9SA&gz5*(>hYKg^i(`N3&ot}<( zXs$fj-BqNwoH{0({KSE!m(h3XV;tixMAsbr>FIzrP2O@HUp>pm-60hqz2pO;VK4L7 z3^A6*Q9QG&o`PLR@E>2}QH#-Ak|+TFrLGz%p1%Q8_g%uBL-QzQaWszBZREyn--Anw z0_nxr@h9(kL&*yaC_ZB@J_|X5ZF|1rk541W%7(EW&oU-;dp54uF2{Gr4KToOG0pYg z#<=6BFiy^lu`4*tUK=h8nz*%%MwPg!sQYhR9OYh~@iU!Vvr2m@>W;h0agKM42j>p|XM| z?O>s)7w)0JUc|=%9JNFoCA&%X)ZjM_6Bo}0l`=D0|AOT-?lwV@Ef3);jK^8qnQwkJ z%ZO>3OT}3$No*Tqlkj9|aqkNp&v<_hz51l*n1u-ru~c2fK7M{A)LNayQC+~E(SIHi zC!Qa6tWEwu^N*}RAm+KnP`gb9?Fiq8qeYS-tx_L0vA&x??tIFcHxFN1*wBMb5wy2V zla4(*K;H_INYQ}t?RQ^A+nyfQWBP?sE)w*!gVmpU;z32b9CmD+h~aJOsI@|cG+M5r zsYo-(2i(J3_R(0990uphm~Tuw4L!wAc`5c zGx{1lwq*R!V`BJoR12JX_lEo9s)bntC*VoDGPLLGl6Ox9sQk^~=JT>q!tyxXd7Mg# z?@Ms3vJq{5&axouuFykcaq9k;3M#&-R82hc(=Wua8ly>d+b3?-gEzzbrgz7@Fmu9e zj0}&V@X%>=R!tGTtjF@Z&ean)P=;Y}4ZfOQ2PgCA++}xlTH}+6l5M8&VOs&MuWP59 z)d#4(a}@=hd`;cy2Pv)KJ7bc!qwMo_D0$}#+-N-n7q8soBC=kEQ| z9MHqdw|r6A!WtYI^!#SQT`oFHj~DwhWa?1;apP=lA?jEXO=W3Q$ZBv7n8qyT1R8=N z*}4Oqc-7FiRv*7#+04!~*HL5oCJ=c!jra!-kVmHyZJ*c<28BnUMTvGK17>~FJfQ1 z0^OVT2@lUr!7D8z_&Y34Qf;0sDiuTwkBdx{+9wJfr|rS9 zHZS35#w+BBO{XjOwvHta3D- zH6)o?Fnp~!`t9iM9Z08nDfDGRCJD`H$5o{%+$t58Z@j^ge5^W9;X^cBui+_-wARId4(_-N6OJNQWhziz*9JUvH~`xiKe>4M zKe(F^gV)YSF%rm zx-nanHi+A!W5NZlHunZb`C8#u>#JOow#_=o>Dx8jVwC=Fs`}ZG3^1YV>gj(P4K{>aV**rN@=|c9v7A z?~^(gveyR9mQANr$w`>t9Yq_$-%?L#Bj`;%%JjF(`2OW7wDLqQoqM&0ny;lZz0fFn zq!r5a@)fY7gojz}K~(W2hI4<{%;h_shfn4#_cyZzZ?QGQD=84-wiv^nlF$COC4+-(+QeKkK~Ti3RWu#bC`#xeyyP! zhh4d?n?j)e(_<8fjH4-M7*l4^3mWCLg@#5mfBa~1%G=4uV~NYTh6Yjacm9n1`}$~f zoiu-Y-C>pqdXe*1Y(xQE52?{cxana!ma~0`N^Uoa)TU7S!bn`A-a=@R&!S4wBMtMT!)9_F>D(8E7M{8>yhS|^r?uWUy^|5FZc zTmC|yNvwap`a8ziHDmXpo47u~i+*>s(u))+y8OnR25pK+WbuEVO*#;as&)&>_r@6f zCg?+pL+?@2$`T5Uiz5H=B^cgR0C)Em!|O>+Sf)^g);|7p@x3$6HB+Xj3HRt#5l4pW zx4^pHdFXO*Jsn-bde(zcxJfnu{P-nQ?zf(A+bGEozL1LVT}DyppEWcPR7B6(ud@tm zCs1Gv!XCRos>9vn(vk@0KMG+*R6lZ(UYu{aC7!wFO2I5c)meBvzt`w59nnufhnjiR zH?WX~jHUSA3o>c2GyVU}KlywOs@8LZuVwDYU0Vd-cHO{9x{K+n{}O{9hwoe zZ*C!Melry}{t=-5e@DoQ?LS^BvHaa!KiK|tCj9-r2-1A>INjR}mvT@U=El~*{v0u$ z;WH;ZH8GmLD;}V&vzkc88A#>VacXQaAf*+ZQ1d{QT*m8Q)00(TusD!*JLDs=4D(0F z)2Y8)m@mCQkBo-4)AgZb49R|Ma&W;hdh8cMjT?D1)LqWYUn2v3S4S0NYk=#u*D-(4lVtOyRR)*2~#^n#YYfHx2v`7V{P^w}cptm0%CKn2{gC4vpe8PpOAn zvcR90lj+SpjkhM#JG)7*@hD#XIGZldaUk2(L1;M#aQ&fK%&s^BI~PXc{0B*xVj75D z|59*k_d5*xcOJtYFx~R681&nj#OklbsQJ@?@~W-qg3do4bG5e(4q+s zCGnz=0G&9ePJ18vGA-;~{26A5b4q5?|3}q(M|1uE|Nr*RYRM|npa{>$?RI-=5+$V} zR49cAsWdbdLPS!eRy)EySY8T26$6{jpi4gn#-?oJx?xU zJMk|a!h-0njHz9OKNufy#Jbn8ex^P&CmevRg^S_Yyx+J+-;|ziW@q8PS@h?t40)&t z;FkF&{GMZNWCfvkhWFI)+4r5?AK@!dBVvYo-Q1xfF%719l);XWpU{0`4sN`03?)2D z(PwT2<`o*y&kff|dPNe&iG=^p#_4*&LY^3$y;{XJya~rCvW$~R_0U_ni~A@ZMmKxb z(lqNS81beJ1Lv&4SRCNQ*G59HgaPkHXa}!BWfa(GOXIn%+i?BsV9W_ICcoqqDn0K_ z(JWu}K3ELGa^~O}hmTyuZd<5un?QoasgzedK+9N-(cp|E(^g9Yqbsw}t9m}%l}m%s z>!Mk|q#A!*?__@Qyd&tS^Z;ADS>K805|Z|_#3z+SuwGn{|E%;Ko2fG+FUQ)flG?=3+- z(?`=tmo+Fa(1Zsa!$?Yd8ENV&ljZSGaHc$xyrk_YdXx%PnvBQk$?rj-P71gE{fFB6 zOgk{-EYGJ*o&r8A(ZRUw*!TMmRkQrFM%qex-lt4Oqb5U#zAH>%tOeQ9NOGO`f*z%c zb1(Z$;k8p3_8R`;IScGGyqPPazwPly8t5v-sq1@(&q}9G5t!@b4_~z;aql>Be6l+f z?oV7xdFLzeL*^IA-sz5hn&0vAbk=z$eI7nd`2ZD z@hXMg=B$q=B`iTeV;>XbxqELfV?Z)HKnTx3Q$Q7rXwz8h_k+kN#3LT!vM_rp& zOygdhuURC+pSKqIiuc9&M^~A1%_D*tx2BTyn!Le82O0XbG7{hUH_+NTJ^s0cwY1Jw zgkJ6Fq^!@ve9Z5Gu2uKB+5OwOrZaai(5e{UW=P}UW@mi)gvInL9+Rlcy5VQ@pU=Im z)Cr51=;Om~UpVw+1iJF1FhsQ-ju}s*Iga{pf7N1cMB^gZF)|KB3bNp7kSdBS{0-7& zYr#aZ55@PzQG~}%nx`yEq4Sh!)#7~ou}6cxn^fUMnuFSdqe06h23|HQp@WhUmub|= z={wiL`-w~F*VD~txVD#@eWeXr_7;$L*=gG2X-jLS$IyuPJ8Aaqc8u>h4Id{fLd=^m zJh)~($}Ct7yQOZ@ru7x%SRY3Qg+-wKy=gceO9-9@Z#!1!Vm(7vnEDxc$mO_gB6-*A<jhw0*+3Y_(4~&H8HYmtQ#-8dXd07+?9R>Rrf`eT+Uip7?cBBBZx3CF9-W z_&@wb`TrMt{r9h0Yc_f9zexcb9ciZFRd~jwq206ls4#R7|C$%U_@CS14(;SJQX07) zPit<64Uy>M7r0N~4V|6;VfVyX+r3&< zY>r?36`Z_MfjK8c*zEIWICC?c!ddTT{3-eaoX_p`C!BO zS^7-dsg~YrF;b8BE7<9&^~tjs*I}zrt~^UVxL2A2g<)f?TgI5ZRrBH!d;$_h2d|WbUA4 z8)b&=bJ&+6Ui`UZpsE>y6rqE<*2nQ}tvk5Bsz*s*1A5mq$bGgw1Hpr;ShgUH4tLAY zLT_`NG_jGKrSCJXN*%K4cnsfQ3n=7&g64{JR6is`j~IJrikmx6*SeVT^m4H5&3?GP zfHACGPvPS*bM6T1p-VQgK;F4y+=Jc4oJHVl;5<_4O3fr1Og)M@m8)TflrA@EngK3e zS&k3AbTEmV3bqj&V4i0ke63uAXN7oF73NA(TNxYaa}#&}EYq&FnTV4+bf{BWk~}XR z#@DD%Pea{lVWt5I6`9kFl?hbao=w|&_h22-cRkHy<#dQhsXf+oc} zQ$yZOT*Bsq?T*{wSJrnhV`CuZGX~?6t+P;4zKMIiJqFjSYtX3v_mHtr1^22tz(Uz3 z`kF6IXQL5!ju^!yR=xyR#lvXHO+bxJR;-@thNDefae-G0rWn_v;y^j(%w2|oHJjViw!ApUy^E^2>? zw}MXK!cVG{#`I5WteFOp_i_53c!pxwj6v$yS#(Z&Iwg2_phFAe&)6SA-OEY5(D-=R zv7{eull39@ktb*H*aT-z)P`(+to5@0P>z%|vhI%yHinOLD#)O>g5j-RTtEb3!d^wjd{8@;#{y8m@px@s5cevi z4rH5I|1z^X|M#`^o*Y%QU8Ie(1L?gfpPRN-6!grzVV_ne?#Ds8n|_NvC_A$pxg70i zmqAI%w$JykL$RY{hW>%Y!hb{Sb_qy8$OQJwpGNtdE3!*6vGf z*stS8wf$vWYQ;%zXMzq6eHF&WyQL_s_W(D?X);F45sb*tLZ$H!xG$IV!EY!VY1*!44$Q6kMK_;HCtBtf}7tA53k) zd*e*b{%beqwK!;a&-E7s@EV+SVRp|4&P}k+(BhOb?z79mQ->Zh&51M~@4PE#xkirl zlqpax<5ur`#3$|7fhauQ7}Xkd;hwh6a8ws@ZIMZu$hkb)%n>KK-w!3mOo7Q8w&nhr~ zQ~<4+ui(1zVXCgafZ{%`$}oT|scg?}v=r6NyCRwGI2s~<+4FX!U# zs)71J!kOC)@!YQm*mqll#C90r^INO1^-DSEjz1F;aC#vs>i~uTjS7#&2p7ymS9cix{SP79 zK%R#&{BsjUt{=(&YuG`)FPniEz7~TuT(PiwAL)&)CWZTge6P+GWcf9nZnd5!{gJ1T zZ~PaZFm?tQ$6*~~ruoml1}oQf(dZXq{8XdTVOwGO$zphB;uEX??_T!=BNj#DmOdZ+ zcyBgT_s+(;1{oZg5{U`(nchsO1l5WvfJV9*%nP`QDURE@z<=eO{fr@4AMqUABX$k1 z+yB|0jyH3###sQxF6v{fm?&lWuOrVrQ|W;HXz1S&f%erCsR^XHj-8)C?sWw!m>tJg zf{bk@y_n4}NTJ1UO?*;XgWb~N7#hMjoePREcKI{%eKnnblHIR_elU*yD24bg3cMV~ zqFahEEX|+{qjl))`8xnf80~q`={I z@ND13n-HQ31yM}D`Fl2W^p^5cr%Uj~Z;Mj?o+_^2D+}tASCHU7Nq)f`bJDptopC?n zab{5-ZS-WiQMo2yyE>8rud0x1^bVX-Rf2ERcYww}Sv<5kj?NS&(Vl11ah+B$y4%-- z@|GAbBFKfbA1$B}imTAfY$?h-RHo$UW#q5EfINzC;V1Ud zghyuf;__y73Y0sB1rnK5wKJbq_?Dv4Y-bc<`e9NNJ|aU1P~^cmw6`}8Kh8ggrr+Yp zNbCTAYj`YWDgUDM@-uYhlqo#-zmI!Y9j2`1EAiu@l@QA~e4kz_W51LTNvxSbz0aCC z8OER4k@J>rS8JndS{^3P4^rO;04iJJVK@5k^*{tl;3 zu|!&Hp+`L(9Wd7IJc~>F@niXKJm)crHkfC^%X{m2F6H02apy0>kwt;<{?c@QZEC>q z^M4>eliSx>hYup0Xij7`auSPZT+3{_ynQ4kzih!+rXQf0Buf$(&B^oQ|8#Z)eVpK7 z#Z+`uh$W)~L&Mh<`E{=$uj&vSNn1fRd&W}6O%1xUrw(mhig4Q8tEiDzg%gy^X~v%8 zq_+PHR(fnBgSk@tRby&t$1=vQPLxKE8w+98<*MO#>5*_h)Ra`A;(igjbq7!*cnrO0 z?`No~Xb`-14qs}&#jQ+xMPc)GSnGZSCXedprt-Sc;OZ7y8zMvD3Ab@Zb^{$$5@6$! z;&fQl0HXCi8(4;}qbJUJc)ThbHYmulo|1C9I`sjqEI&{8@3_;&1FXkQ@DuLoK8!Mf z=W$H0F5lWQfbSvHMJYcH6E>Jr9~nbqo(8jojU^GI+^8R@47kj#q&GOb%g zT6={st(h@W{g=^G(;AEpI!1qumyla`1ZbS-$4{(BJ$2JOF1aZL=2|aiyq&?}XZe5j zCs`KhY-%<6Gsd37L{BJK9tv>|vQ(n4OZ8S|I8D)?OrNtl+8rT!>v#YcJ7v;`#l~cT zZunKN8t2;^(IT;6@?jaap}ezDdLj+9Hyy&=*+77Ak6D8WqM9( z_}pRw>Z_hd0kc>z4=#pnve)s&8aEVrD?^W^qv-t)3$%C=1kY*{V8Q7aSQY;ar8q5e zuRjYxW^4w1Mg>-kR;5V_T2%R##Wv4RLAGHj8h8nje3u;@+W!%*Hw9zM={&Fw-Ola) zN}%N$2mjQrar#~6oXC+-Jig>Ss5^zhTtQ2CWta-fkDrG$i3mK?!RCaE6`{eo5?1eO zg9Z0G@z;yfaBf;BL@u9#vm)*#WH<9maZ9mZKYR!I@VM;BMBu z^FkwD1v) zvVXP_WnTPmhR`K#BYar91a}M7qMNZeK2({GXX=`9x#~$Ys{Y5ps0vQv@JdR)I)g^f zXL@dL198_U9m*(LLfS6jcz(VE#b27j?^!jK`ZC;y-}fhRXEA@x7CJQk6V+B|aI&*h z8JFV&x_u7e8jjWs`yzTi7qIQzVd{lE${2r-vlU;-^;sx^#mN9H5^cu>7*9ji>u7}P z1vDE|2Ddvmz_KSv&^gW>uZJ9hFhNZe+fs!YlZ;@$?I8#rh{fu^8z8CR8-(q?47#V; zI{r9;F|jYv*5oHzKmUTaCX1ncF!K?<-AC^zL-JZ1h`}+T6twgM8Q-wNMHM%2)`L(I zstO{@Xb}>4)q;bom=412$DIEABy2q$i<6CHDCn6Qr6|{dnci{kndcdvRQh~8Z`pl*g8^gI=*;88ED!Mrs^+``{n>H!urLOv9MphVR;QR1 zF_Fx+oh2$a8umN9f3iHi>ngf0?TPk>0AenVK=TwN=?#}@>pw+0npZL@5w<_b2wi#CBO;B+vyDNr}7u77~m#C`vL`NM)617 zX}XOHbVMP4k`StN9m4C0Qnax}okW^tG5_xgT%>Rf18Yaq1UF@F-pmuQ-){xduSNJm zyBOECP9~E}vKYI6Jhix;qqOwDyeslW+!HH5u=_lYA7NEMJ&a%cB7Zb?RivP|jSkHj zHxnCIxx%5I0IoazIm{{&8;&QhC(H3a1l^~DDLlp)45O^^Hlz~Nfio@&QBM@(MqQSL zAIm0?k98y}va`MN+AdT$;E1~IE~qK64SG?XoIHgDtw8OG zI_SGI2u#N>!70jY?_|G&`xUM5Qe}Y0*$Qz-YJ}*+G&_uXu^Zi7G;xcrHf(VS0VjSi zcPRFo!GGt(%n^yyb2^Odq{foU=p0ORt%e?7Gw_{hMDz8Q(&gpXuynT`u5tXuJyx0k z(j5bgw{RT2ISEuR@t|KHx6=U+Y0_*qpfeg@k>`_&)=k1jkGzGAGP^JE<^7fUS%Y$X zE1_r-?)gqtYMrE_=|F`$iJBNQZ2ahQ+G}0uHB%hVBc`Psfb^$AXu5}SrmW;%8 zm+FRXNUo+dX8qH}&`-Yri^BQVETDhY^@fZc)%guIDKskn4XV0J(Wung;kC_IR3hcx zV7#;c4qQ<5holxCUWE1m-k{SXc%f1ZNz?tfEY*BUyK6&f-xHCyw2As=ic@>~5iDcc zLAlF9scaI{9NzxFI{klZB|cw>@78siPD))u+XQx^s=Wt$@`Es_;STuy`oZ}~ zhVw(yM)2eLVbsI8aVQ|HuAxwxm*d%Vy$^yu*r1-Atpt zno94Bk;dL`+^So^w32J_H=CnwHyB6BZ`H{$Z4rqm58@4rlc=28k2lS3lP5ea51(cwaxo>5 zyw_oi`Rj(t=;}@vI^KJlw_Q%fFx6v{=(s+P|Fx%)o}IU4Dc~2V9h8nAY!{&l(>-c+ zsOI`=^f{*wH}S?ndupqgOCy-}-+YNV7<5Ss?S2^09l!0kvik_|F9eeGuMApzY6DHQ zti}5UXYqMvIu4d@B!fUzR6-7irlM5(u=j`0nA`4!b$ z`Pp)=IP$Y1juiR=tCGG#ROJuuO?)vF?S7Bxys4C_a+e&ojAT5>Gg$sll%&gS=@xs3 z_dJim_SK2lU&u7|ZY$we7I)6fc*SM`@_-Xnr=ymcX!TGMI+#4ryKeJVWF3Hn zt7_ccS&CeOtN`UkPr(G|`;gpJ&Akq(9)91)R(g;`K^&Tz>w))yR?M2~NJ+->lp@LW z=4#HMqe=iBuJa_96@qj`O97RqWi_*hbqOxH~(vpaurL=&6e z@!CGTPp#r&uxM=&t@5187r#_ShKcPM8KZ%Hc2-!lOOvK*n37VgJUu)&4;1%$z%;oW zNMNk5^TL|=_nj)IzfKorr`lqA!BL#~^%92mNYeJOBjiz^&)D9#@$w2cY7tFGn~hnx z{#FMD9D0UJ-%8L*Wi5njf4G0UZjxN zpU}0qg0X7zFj-9tQ{UCn*0{C&`_1;4TA7B?licyOe=uEsC(obqQk<`{#)~YrZNm8~ zW>9FaNjuv+vD4-PG|HX94Q|x{S~1)criE_Yev|vX_B-$1jcepOqMVdW1?cv31ybGA z%Gf7Kl(NAP!!w@o6s)DVhc8E9iH!i>xw0KUPc$c6ujO=Tkq8;s%2E32Q&@VV3%2`w zN19Cdzq&6@?sZ$+x; zo%4LEGg(ZsCLiE@WfB%!2GBQ}M-8vcDQCxFS~F!ab!3bsR633s&oOdt2D^Sda!w)kGOF3vH#zG|7XK{s%+>}^Afzc zD%?fAIjHYd9{=yj4MlYnrl?O@8^bIO@ZYGVz zDP(d&p8~ZMSZ|R$e}nT`(hv=$@#eX-N{+|$OIUA=UI;0l8AN%teHd$+3!Kaurrq=i z_p#ofOWrJh?zM>Qq)(EOvj>$Lk05!*P+p@loA*(to01zu_$F^=(}@%V{4MkhoYEyQ zBy={8h>YiSeS3KpcOu~HfH9V(4#AI{R`Py(kpEt272ks2LLa{cQueoG3_oha7+ksB zqw%YG{C?S556f+ z4EtV4Va($pEWY1JmbYi{--_4Mlm441xl;{uCQpOxAagn>bP3;xmVmVTg4(3R#pa%1LQtbImu8CQ6r89?*9+rHUc}e7Qe;jEFiNo=4 zFL{1jIxw^(8XqqF3)2E~hiz)~&@|Znb{^9sw1>R!?|3Q?6Djmo6h2~cYqZuscvtI% zbI!Gbg@z8lzo(oYXe~oFyoLtVH+X6tqc}x7IevhsE`JMV(SvPU$%S#MG#>rL`%|wn z#*7-+UsVR-|5#~JTj^W!LYnR%ftoiYAYju6P9p~C#PkYUmKH&V8rGQ0X4ATj!||?} zAB@c!K{3T4l&!xVK0g)#g;&~KaZw#-d)^jC>x~+Y3!IG{4gY(mN{ddRt@ARp{b+=n zWhbD-&_xjYeg(3fzVOzdfIx%ZZ7#ml5K2bNV^F^*>b{P_8*@M4Wy^YeI@O;GOjF_Y zc4Qm2x7zTgr%ZzAcnetaRT(o}?&A0fJy_kRMICuDs2l7FA<-6ulTVUkaw;9w3*&!j zZz1Ix6`DI*2M(VQ#(vdMyy1H)Ygb?(Dmp+$B~^d~Fm6k$X;oeE4i!ob8G~ zRyD(~t^<_gBu$IMv@uzs3-+Bo4R_K#z8tWDhTjHsZ%w2YVFBYJjtt^Ql5 z|GxHrEWrp?W6n)Fk1 zC?397j9=U?Li!dJs(CSqUZXS|c%uVp!67iN|0dlO4JV(mcTi4pD))J-GN@!PhQ*yV zU_N~>7i}lT)y3H3Me%bGrC|UK+&q5FJU9N@^1sBF58_4K@ZpII6&kp;FwG+AUnu-& zBFzw+PL*Oo6cBy~on4DLrIc_kTffb47k>hZPT7jGFVctiuUhyzaP28TTg&%YC_SDk zhYE0Ya4nWC5kpZn5AP%(gO^ICW8$~TctdR+ZuZ}aSD9A8`wM|g(`+G)?XjRLv5B;N z!e2bJ+6h-5`Ar%YwUp!)#`GWZu%JE$mi#!%doJO^g&v*B&GeDw?&nXS?hC;*|Ic0; zU6jXp-kSjJFT7FoFyjyhYEtB#C)oUH8U$pNa;9^&vH9y_@?m^bBe7S!f19iDO5+1) z-4(|P>+AzJtqjh-!h@%@Oc4q{Iq`lw>GP_}rSZCmF0SurgUnB&AgmU_>AY#;CK={q zq**emv?!w8$K{l;bsGLTCW<~Ej=&pZU0m3C7n2|u-}!XHsE3R>vNsCDmN?*n?!MtY zke`%H zGzg#P zlZdG!dGyPZn&V-#3fI7hnN860j%6ONS82u5hxeBAKT16%3HqCy{Dp5)7~= z>mBV_JLVJ_{O+bEn}gIaTa8?J+t`_S80%Pk)H3S?34C!HjtlH_{&2OUgShBZM<9CI z6}0`lgs$ynO#igU(7G%a4^~^^wmCt#yKma?Jzx8Gkej*s9nbgdG2Gdx^W5Nz zd64Pw<-Nmic}gf4(~P@VonWkvGnikg!+)D9aeYw@+9>oxQ|4W;i1|Id=cC3;2v~_G zaCa`6Eiu@el|p)}{!;0-Oxm)k8Vl3* zv%Fp=%72KWWhYGNR_qnzU2vo*GseV-Da1`$4`JbyMJ(>(bK>BQNB&fCsV#XBW_yjB zxQ^*?)n?(s-MZMjy&8SW#mHN83f|l|9{m&}A->lMc3m7fyiU(5P3hSu%m2sr{I^eD z2fl(vk>gyni4;bjRj2GbLwIqYD_TyIVT|~HkhL-&s)k~51>+Rv-!dhM9t|p;Rf>N-G_(St?_lf;5)k^{c*AzqC>(4mz?sz)?uLh+4MPk3L6+1gh=;w(<#_pa%o664( zpRw^zUveh|{9#qZDX`Pp&1G+>LEm%5EZHl;|KswPYjUoDq`Cy|=UZ3IUa=lUeZGK% zju<+59LIfWQsngH5-er=eWhwL9MRi@bL9N7D9sc{PL;*4J4HxLkxv4*>$wANS3#>@ zlMXMo0>0fvNa3=ve4;;2+;R)ozVBuEZ8zo*Z9zd>0SaBrc!zUmqONZJ@LIQo%))m{ z+4!ScjFUer3maJN-)Ofb+za`DcK7ztd?jztEmudQIk7mU24QQr4%(cSCYkrrP;8)y zDcim9ssAdN%6){l1z$PyTVixKOk3pk%I`@FDjV}8KxL*ebct>#r zjHs*vyRl(dQuqcYtEked1tFxBHx^t{K5_3x9Rry)>6uOYHrC zd$As#C8|MduMO_zP);dMaL{?{iq@|PiMaB$=3gWpFQy}=S@GAhgnkDD4m(d>9woEweZEL6*7^| z{=3gfmsGB>Y*+B69mg+c6=>-$6Y!fYg5`{-Ior1lwzO}; z%Qvs%M&n&@Zq@l0>w z_e|{Sip3t*JD7C(1O)G&frlaw3^o%I7)%eba4zg3{UF&d^% z4aCY1!lXx4 zmt4BNR{@{?oil8c(%W0Od9U0-(AF8NZ>0_I&FvqJV0xOy@c+*?E#|KJ=D_-n3!J5< zFt+YG!#Q?0^AhUjavp5eouHOoh;fQ6f4uz;-k98m^4ApL#@ij}TpojiOfRNPehAI$CHQ(i zX}BzQGhVxX8Ta@;Cd@2e6<|)F3Tm+rev7!EJJ?L;di&M?50P6?aLD9gSLRvNWo(5C-s#Tw1 z*3>e{*=h=$mow_GSE1w0Yp8DUA=b7;BiAT`N5dyk-{tevvn7F6MGkQp=Ejg8SpvO5 zw@FqdiqtZdaO7y|;q`alpo*2#SJ4c9AT8UgjelMnQQh5CdZFw`L80gAZrW*r^bEZF zd;~Yaw-J{75g^G6{xI0V<8B)?a8G~w*hVIhVOHsZQB;&gP(cHH3-%jwp3FdH$2uA7H(Qzj_m1)oOve5zphGveyPg#NQX zJ0g8yZ{awUI4Z~b9pbTRAON4)egV(ZdPEaf5%;(Z|9pJ}f`KDQEc*y89DIW(wb=~D z7$KIc2}Q56hcNHv3kY+bKw`^su+!)?XC`C8`^9SC)aNiMYBwSs>jK}@z3)7cQ=tQ^7A#F{d|P+stA`XvNo)cO9Gje*Ps)69|~N@ z!FaEexG{AsEoc*hZ=u@o;p%(t-uwh?JZn!YN{y(}{WQ+=eG1c73!_c$SxEA|2z&K% z;qgj2usHyDH|aDM+{lO9@iJ_!WO3p0qnI3?$n;l!fSNQPmT!xOEhlE;lHJQ-!?{T~ z;IRYGZ=B6_NbiBmFIMB8+cR-zn;9nU&_&p11>2nk(S+%fcfXv-bU4M)FU=F>cB|v) zUmIb2{Xdw0K8)vMbDnqDIR&OiY~pT>DudC{!i>G>#nFX>Fc^2A&86Ig6^9xjOX4ZK z6^#R3xw9YvwV=MQg>zak3LM^!=3a$b!HQl_&|%=8xeLwk++aOS^kR&+)B2#~riw*( zJ3*y-7Weh62y~a!K^VV_%i1yu<;>hM zDdvC(Ia)5oqs!iaP`Mvkjar0v4H9A7`5W+lvJ)7b9f>REuV(Y^-?`_v!njt=AkfKr zgQCJ#2xNpmcjbVdv>LYX&9F6MEq=3aMw??z7^->$cZVJZooPkfvq)P^c;k&(M-Q?Y zxZ~VV!dKAQVU1%Pr-Oj`DVQrH0rQur<5Qsv8*Zp3}? zV>LdHUwM#p>?Axn<%$j3Kfrr_56DcD#R6+}*t^Xe_PfM_+(UJ&GzkLnJ42wbvI2B` zuY*!=7zkWVfTG`Btk;91!9Bn5$u>ljECCFV_zGuLEby3_4({7C5iiUQK*vAp@RG??%-uQhz+*Fip?$~7Rlf`8?Hr5A~ zs&>Jx368j--x1>?BjC7cA3QuU1p}>H!N92zW^7{q@1u!u;B_c)HJRYKtsR!GIts2c zUcu;~Q7CO3hEt6X;O{NjoFB8z9cCwR*~k~*7+we-f4)Ia%u`s^b{Jvq0i0HMn0A(J zqs6Nw=?!CDoDLgnpG0;d$|gfn&rWn8bMt5gmE3$XJAP6D{jrWK^UW!3)NcX=wu^>HXg5` zYr|M2Q#G~W{11Fz3jd;QA+@0sZiLOpuZEKNZB`WQ5n$TyjT>ltPCJTQNx+J&ao})8 z4?inBfkn)QsGdqjiRPtfvg$ORdJ@1mUpy!sY0mm{EO0}W5bl2Z8S@Zp7ZNh>;Pg*#(0Soo+_t2K4Tg)Ov6uuZ<{ZKM zQX_EiVj$K&e+qL%gmFho5JXE8ZkW->bi~$kmuI-5!?j%c&?>=~H1$VV;Di}h^YNwk zIgDDIjIEmU5GyQbx~396Xm=t_rl(jiV?608$6(2X22_6QOIw|@P@*yuFZ>F^OdnOc z-DC|eiUW{&?iOBakmq~kwa|Q(&GgTS!<%^~w8l<@Hnna70f$xa*&q|W7}Hp7l_0-o z=_?w!s~)qb9-viTq2%>m5`WxM!?Q8nYWeM;1X{#J?llr4XQNL_c(5@?*OesHn5ZF z1}Z+6NBN0eSg_rLGSjNy{>2bS56_L9c=4&&*aamMhOxX8ws&dC+hz|L6EXL{+qqqI?HpDaI4 zshNCD_p-UKNVpjfxOlP#ooK!a%b1Nj{&@>s+~SXkTLnno_yW#7?L}URjx=>}2?gxA z2xSj+?=C)9??i*hb~JbcxW+vLcUi2#zgp|aeYQIZ-CKkEuMMK{iYFLfScP7y zUdSJLZ`j}8eI`}uKmTW0X&82097&oZ{Ah2=2N;UAfF4%ApYuzOZoe|(Up#!3g62nK zlvo}#u%2gkjV!vbC5fUHMJO`rC5B0Q(Apwb5nH$9=a2UKbcP10~WUy{0l~fA|JA}-%MNEBWcsr1k?|lK@(!d zXsmWJ&62r=tg9FEci(`k`hD2LH1E4!W`GSS!^Pw0cuMP(DNgAJ3XS@SZj8TZX8r~- z`4kGXI(vV6HC^>NOut6WrfsXlXk+?bJa}j%b$*Y~F3X=b+<4*)j%W8#%pIeaZ@rv&nB=9j?$Fi3u2y7=g2YT!W2%@)+q` zf(M;Tuwar0x)|4yh*T391b2htqX1kKw~#I^ATECTYAl$bMrK$+JA^mk?blW$JIi=CF-zeT4pjh!v;1RP-8n>_AHgE{Wrl0rFN ze$+a6k!~*6Adl@2@x6^bZDLx&`; z>>Hs+^B^v)_=x)ogK4VVVH{U&!CeVH#i(hI(NixF=LVHQl+SI@o>7MGdpzg~<1RU! z@Wz!lhH&k^0MJ(Puco2P zvq(2z0l$aEL513F^c31m+D9^}f9Z6ROAdi~#X7jZ#0m>KE;5eDB&;zy%)x^@T)of| zZe#l#EH$zwLvsoI;lF~+?uL=&pGy4rbuYCSy~i}_$GB03cyVGpo*Ft1r>EV6z^)N^ zUwH~TStUS|wDSLx|Cy>e2LH;3V|GF*wYH?eL&sEDF?Rs&Zpp`zwGwO{ZBVC6n=bQi zko(B}H21|Asxr;Roc?h1+%kweZ}p+(tSB5g>m>(6+1ua>E9tYgW z^;;PI@^+$aJ-eu@{RGgY3_Lf~K;C?LzVzukV9t7@@3rg%$G41O?4Ln;FD4JK=er~U zRPLHXjT`!TLq(U+*qt%Z#Dbu{I1u{=HsWp9DCFI+q3${{ioN9nzY|1{*Gn;Z<1N=2l#Ck%X%lX&6X45+D;T#| zha{krN;UfMRY(jqB^i)GML67JHGA0;w~#-D>6Wj(M!(87f_jxHINtQ(PMeg&#JOWI zZ1ZoL5N5|;_(==@)(ewY_es_l5CLYbGE}2sNhcp1L80r{F>726zP>DlS(Bq_%eY&# zGx`s=eAyLc!q=HK?%I9?-Yl50HiRk;#j85FnpgEG0X`m{K z;MugZ zNNs~qW|)P?t$T5?yeL1W-j{@Wcfl7UGd7#OX82t=VW)^abC4?ikJGZvu_Wgw$Uoe3 z4%L62r(_EeviubfBTe&pa`q1~YON&yB>L0;N7b8$bJa!t|1xJ5LK;YtO5t*c+*m}2{^sb75xwU)-AmW>k1XphFkrH@hC=@32H;SYk0$$o6(eEQZGMU|tN z|KQkBP&`zCetmcF;b{|kTlpGZc$@~so;SRP5yIqS}^EH<3y~LK%q*TmQ3oT<}-ZUttJJ#y8eP& zumSgSZiYdO!4lpbc2*`Z;|dl(nF~2nLh+?-bPQ_s5nW7vSX@d&uia;ue+vgPqR{Fi+_tlrHEYjecE9S6Kn)uTQ|wuWzH6 zW-FfQ*QCCh^;mmb50<{Wf+jJa@mFgjTCCiI0*&lCiwU63wCgzQ^?q#7%!Vh$Hy|oQ z2z<5Xayr*$#p(e&M&K47mYF`53wO?MM)94s_;LRoJg{pOCYc{5 z*EkWn;NT5u5qeCw!q|V^dr0f<7Fy1{2luQ;VDV#9a#_KnJt_)(zkv@pZWZJ1N$*GwSMCl+xVKy! zoC_+b*V2;Y>!;A2U?JuYe2p^-+-df)RQ$Ly2QQvzX1v3v7`*HZew&;Ibpk)|c4_MH zcq|B*jZNnySdS0e!#$XY+Y?Km)6EnH-mE9ptI6npC>+!4OwijYl#GARqNQ8a$s`x) z<=!0Z7P*VnH+SKK0!av7$(YL3IXL#-X~w+!2^;U+!rJSCG^w`zf4e$r?xesl;lXg8?|YOCL*~&%pYU9xP|~t}<1U-0o?S z`A#!nZ2{aD>tJ51jTD*J=HodW3*sJ`IAYmix?L?wKTk3KL0t&t zR9{0xcM+|C_vAj$ibk7EL-8FwT)*Bbp1s#!%G&6s~Y~-bC}2c@qQAte}jEhdCWu0nCIhj(X$qy%_a!;%AaTT+Ga4<>Exu`chIhsP^w%$ zjqxu{@ws9hY}KEJnIRm7$N5uF!5ErWCPaY;e&e|vwWuWQ2e)qK4$ldTPukMB8Fz5@ z0Mi_@7IqF%xz9PBs$!4@NThb!{Ztu#8;EpIhLl zln+a8#=_Ys0e*<$LE3++2=fN);G(1jr}{D$#1AXtLdoT5^iqQ6Y`ui?P4VTy@}+${13_(dGxuq13U1sGfV(b@ zrl*5*$ol#eEVnv<`RfJPIrC^7RU3o{+H#=j%r}f)9l>0v#u#<|7;2cFgV>;p+}nwP zu&k_D?%(@VArb|_d|6Da{fb)A_rTE99KP$RaqjOOIF;v&wf-fUaWHu}J9HWud~d;b zf#P%{aS(&pnZSVyeq7AJ9Z27!j?Px*aC!Gc+%onXc01_drK6Ehy}ApN2ZKpqM;0ln zM$=rAJWhJiLU_GpAKsRFgg*_M$nanRjb6|Jogry>{(vRjz3>B)6zwo3^%QDM9!VE% z#*)HqQ%qz1QF6UD7?eK+qnjlTI*QDNMr)UgY6IVxWL*XJ_ ztnyUGNB3js)(%PfE-Hw(iegdYY$f-1gAi9{s!E0{RY^=$9i=ayz|CS2G~Q!0ZScB^ zo4dWK-Fpg2eKA7oTdW`FSska;;fA~OnUmdA0+&jA zlHR2>+-QDc_?=}U&G?V`n6Nm6CmyPU+65%i)mH)ZeOn4Hs%4zqvR;h0U5>&1jrcV3H}2D9UPLyg|L2$gs~@xP(FFV@df>C~JgTr7D80Wld@Yv*JA=%E zA(*n+8J<^KabxyO0J97B;Axu3op2dLlfPV~l(ckmb(+C<3mhOF(^2&E7ULPV=%G%C z0q6hWBNkPM!^;L^H2Av+Pxn3MJrV8&{~HZ-alQ~=Eq6JY*$fU}qd5Vec$~NtKcs6n z`N-!F|DEh3HR%(DQ7kK-f+7CGu=&9+Uaa$2zN`HiYGRx=Jp(@fk&HRtL_UW?*?PNN zO#=t=cY@^s9tvxUkkI`sG?3AU=j|ilx^OD5Kl}uDM3awB$|Hu~3$w00lQBNeS4I}TVgVrui6?(7CQfcmggnBJIw{B zjZ8wJD~svC9S7$7yp9@1m6U!7Xp)5ucI-%G9M17%ZKFyby=L>*>^sW;v#^>EPOkLk zjTm1ceLBC@;1ua9*HLs{2wi!2nt98QqfoCX?Cw8F2Q8Z@Ylu0iHDzd5t03R4WF3E` zAYyF3$8hfVePAZ9G-VFxyPX2dnK2~TQbd=;#P~};jo^EIeTGBb0(g1XWKz;4NQ!9X znseRB|IIvVUKx!GW-Wyd6f+oYudVOH{FEmSujHP5+Rdr3PB>>P4|>+}77{xd7wwS| zsQI15~yZe14GT z+|-Bk1v|OPihcC_Qaf6>Dqw8gcgkXu?^zGuM=h-@61(4lLpfzSKk9n(Z*ET8*1ksSE}ymXMz~;f7F0 z?vIop38y^-;hD8?^Z8Q)vEV@cGb`kI8}&TV&$G@?eufIDPi(o{3*N|c4iA;z_(afH*xrM_NLj9)?RCr zSY1rxj#-j-!D3wFl>$B+tl_?e0;Jjp!HSB~;h!bvfDydjwF^}n7)R;$mB}pVQQHCnhjo@*2B4~)71~c>Lu(Lx6{hr?9 zmPjkIO!Enh|1}%?^c49op~%3e>>+RY{X2L*L>rwV&!FWyRg8FK z$m&?jvE=b#l$vY?x%>6F7t1|q6O)-PiBU$4xth51m@zfZdP{w~k`OWj(I8h0jpyr8 zWoX>+wcO^@Mj`W*s9Yxvs=S%Qbl*nw;7dcBm@hmL&xR7_U$j*YLHI`qDkhHQ63pwN zMavqioZf)5)&{E7P@`EMTG;e>D}~LHAdSudw9?Lksb~FwC)o+2Ex9ou*xX{CgrX~lSMyZIyuVXgK0X13J?Eiok1;3ay?~p)#n!Nyp8*O+tEeqewU4W|-)LFiu49Yj>ab_bgqhgQj_V zS?~+xQ1QtLG<_NleR;<04-2>|%S2RLRRu|lr=!ErXPD3+J^b%a!UquT$hbYC2!;6v zA*4VTFR(m(rl<>EV*Q(&rWtaPmXn!xU^gy&>j8U=zp-?yId<E4Z>~c< z&*gA8WK~?^5>xBB!sp41wUvT*O&+2C^k0sl8}uRsp5u=O8%w7zQOBf{1O4Az(o}MvN!~n|eE#KYsvcbtU3C^C_sg zF$^{}i=)VtyO7glNG&Iri|j8zpUgPAz+4n399wv=&qQFJ#{ixmx;y+F{yPU52nh(B zudjt?{`Vm@*$&Nfc0$!5L(opRjW$!p@*l^i&{W+j%l0(x zOEX@(Fp-jsR*;6YByB$|3w;qEIir#puwsQh?6=c~QLDayV&pi|*?`n|aU58I`baQwZiWL_*{&LI;o9qgbaZ(< z`oEIJ)r^zATPB&Z>tBK0H!&LNJBxB9dYG4PHW?(?!=$LE7?m4Ldq=v`%H5;!?{~&1 zb^nM~Q%~Tw?ilva<5(k6!8b;fU*T@FtM4 z-&Z-1Q4b%(BCbMjo(RgeF)r;919EOSG(7*ST#?4FZ^cCB9gm^Ghf)UTuT z96#Fe>M~we2`0(XO1!yY4b;U}!jraD=o#~0ol$nmIvnx9ke1~{(c$AM6yVqa%T&fv z<#zL9p*Mg|qyq(76E8_4tel zHDk#pdnGmN9mfk2wHR`29NoN70mr7Q(ad7TPSkjT_0pN>GUX~>c&kEB+6vKY)_ClT z&Y-fDS1J1Y6`Ewd5&x+NoGE@0-+8CvZl;L7o!g70-!r*vs}}5<4ixsY6~8GRrz`ze zDdFQ~)X2ytt)hDBT(lRbvU8M$LfY_X+H9IHqCs+gg~Rhsd9O6mRw)$ST8YKM{n&L} zgznoak=5k47?eGV;_s&5jlt=d{o^ff)0cSWFcc=A2TCMsyog@!Y(})!CjrBk=_>8*N>uBMX{UC0gOF@n~wDBiLyZ^8}-o7>X^^O9L zXmBK*)CY7?;tGC|ilEy=`gCXIW1La6nm(%j#rO96nb&0--I#7g*9t`GQZCEQEuBUO zj+tY_i;M88co6Q{=+U_RaFVcELzZ_7z)YkG>`ymwUj#frWW`A`TM$h)My9Bw&GN}M zhbV8%dF*g_hMVBX@Z4h8^aSO*h%R4E;8x1X)5G05B$jj?TMAa1qx_|^GZ6GJccuTr0O`bv-*tpJ;cAr(HvBq_1Gj$EfM>paY!?n~Pl>sF& zVHm`H`fn|w;X^GB$EZ4aCrzt5NhMkl^u)~*LfLs~V1yhuiuFs|=6cXYxx-YjEDmm% zgoELpY#8mo1-_n%g0;Kuac89DNlIFi-uNoei0S3v{9_e{xJ2P#;2adNS;`zd%W!SS zHr(?1I66-b$3!p2{fM58jb}%p>NOp_*`I+U)&!#G+H~B!sF2lR!ZGUiXjCf z0*C9?YOHs8~Gk?xJ;al>&Q8H$;t=cb!Xw0h8s9z537BM zKES(o;$Y8CZ5*dN9`7e!V`oCI;9l-a);r=1H6~IxQrm=cXEPWymDz*Wc*ya66hBQbSFFxvQSWp2Z3z?BU`%FXK# ze?y3@S6;^V0|mol@3Z4boRGPg44lf*_)B%k z>_uM00@MuIj9cstvFcPOXKT0z+yohW{-+N_%-N0utsJh(t;OH-GSKhP0}NO@1Q*&B z@%b6TIF?KNcbzbEPaVi(4tk2*#u1Lc@Ll~TR7#A;F6$T+xc&&oG}q#>(qOdIN~Kip z3P~O;r&)Vg7BH*-6={RX5i-;uu1w#Rqw!>nFWz2Pg7+fpVd~9Ku$Q@uR2(YtT(cS;Jgb4qX@hXP zi8;VmZ9v|nFEIVFB^sPsjlACdIL(>|59XIc^Wjpspy`T7&c@502Q1DH=bxV` z=%9s1HO^}hqswwSM^`O+k4 zep&!gZXOtG?u#xrQc-tMh4uw9UXsjzoWp;wd7q*uWPKE;FMXeJXa0Cvd~-V--E#-F zKgom_OP6A$RS%qm0vNBplm?#!(QSS)ep6`T`i+j`+28XhpXCI`K8dHW(-EY=dR8XH z9;1~>CS<=$g1(+*4vxPXs6O^Hu1_3K7xbQ|JSeKH5 z7bcfcq;4(Eb}T^c?XeiZv0Ql1dQMVa$e^s7!} z)M+B&uHC zg|mh1(ed+j+;4UT&Geno$>%Z*Ojr%Pt^v5evIG`?T#lIr0weDFuHq7Kyt%VF0 zkg@ACxHWD!zWt(&%At+JbEZU)KB%_V+++zlzbFia_qgHgb(3ff z_k=zxit>L~W}mY8!e{!e9^$2AoHqntELFZ4q5P z7sFLvV+;y=C-juM4G|Upv0bHC3_`)Y8mzD$$s9#mX!c_pILu!KFBdhS!s-Sb)B=2| zbrYlFEa+L99hzyk(fn)Ld_ny;G+_4x{XK#(O|PHp>CER$c07RCgViW-Nd^lYS^vO+ z1_-bY+Q1mZMqrmn$nB9>X*?V{WBIGnc6I(>w2sd@t*j;NW z*FX`@yX1@=9sy8VDv7->BtX;s35pn;#R?%+s;r8jb6VBh{l(AWg}oddnY)vu;@NuK zrU-(-c90(c+qkYVW_GI@C*d?$L(pXmYb1xruFz(`c_sbN%R)(Ly50b_A`)~id)RQ*ZYbOy!h zjN=<@ImnM(P)?&-b;9eu9h`S&w97tcX%t$aK_Q09YeOCxL# z?nb$qlXPvK53Q(PP9tYNpnc!Y;3)U=yiMOC!Rg>G$TKUzEX%WGY8y_wOd_Co?j4xd zp+@%#lqh5y%c*d$NIk)bYCpEaJnO}3~bS#u;f4T$F=s3E2CJ6RSnFvj)R{#4fhxxEv!;E$G z@}Ex_&}mF5k1U`y=``l9UqtOSRamp97GLyDMSlr(`gT8wZWx%82J6`k+th+FI-;bf zXU>lkox=a7_X_JJM$p|tJqq5}hwk!;!*gG)$O_)c83)O5&q7-0Do5V+i=mV8kEhHQ zLUA1deA+mgk}qtcFJD@bCnSu1oi|{K;0g5oM-uWP#_&ErswCS>jg-OmGygO9{QG+q zM-dR>t#0y!lz}H zC>1psD*Ic=AZ!eo?05$+t^{IF_gvggS3!S&Huu)YhRdueHV{;LNJZ9>R3&}|r5||^ zcTSkpO#>j7<$f$m4`9_@dDJ?;4clI;gVfzH&il$toclEi;v$S7k@eoD(pzBA4tOK? zop*2960ZM(EO+yfJ@SqR5;w9f>? zY;WMC=k4OJH5y5pJn+j$opSQyPscoWF8)sHn$NPGz;FQhRrq>J-rXVTHlrNiG} zIq3)92)RM~S)Wdq$tlX`-yj2r(eyUk12dLovL2&gu59ECs_5{isTH?SasFdmt388W zn2M6CbQ;dn9ErPvcfm^z{lV{S z?yGkyhP!N4{or zsqf@pgWq|Hyuz8^4H_<|!vV>WaOcoRE~Oxb+W1xUx26%LelUMU#82>BD+&ci%y`wy zIQWoo0}XcU=f4|L;P;zht`dO)C|HmpkeJc(-Zdl;$cb)i(cL$Y5+=uXA zJILp6ICYLKLQ%1!pnm@6@cBaR!KHNZWFy?R@@_CSzmYnK4uQ=GWk zu5*z2y%qwJE4Z9UZQ3Cnk9ox(K~umKJp0;lOv7bd-A^K`8GlJESD5#*!-28}{!sA5 zOv;Xq=C(&F;)i!U^pse})n(@KY%8j`KfrQAf2W~l)oILs_JK5BcCtJ!<6VmX>$%H9 zRevjx(;ta@c+<#K!eR6^m?uMjB|Hhj;teZ4v9r@LTzf0E`!OF z`gxj{qqq{G=@6s;oSQlIJoqQPf$|VV()>LiuSFDdPTe^$=T|&TaA<%rit^ME_#BHa z=wp^^I4n?><@(lWK<=^Sc<6NkZaH}w=AB(f$~&BBd3rtWv3~=lJ1?S4q&5DGy~tnE zaGjqKG?kx}{0@B2`{Bi=hv>dNiG-%5l4oQKPJfaFN@M~uN!oB>jUvwSQH7Fo-r#Vy zfu2q|O~2QO&@uIFPXFgI-iD=iTuy>Lf8Qvxx?2^3l_+o!CTBtDPNbu6Sd#VjkjBf zPA!wsO+^OBaNpopbO~No7>Bp57ee#z2$mID1JhP7#89&{7&+6Ct%N z>Sa!GZI~1}0qiLdg%>JvS1p_2mu?b#e2_yTUxjIrvMw%8alz1;5m48!50!x@@$FX~ z(&$T}*+ws^;_qZK{doqj=fA;B$4ne>xCLKC1>lsx2#mNdiDQ3kf{WR6Ay8{7a7$iN zubK~?Eb|&3 zEwkbrw+V0|GR~mK_GFV5PGdO>50G3sgFiUGk}+(bV`;)EUhcH9aP&tT<1VT4_g!I3 zdi!=vFDyfeJEa&SqlMxAU%BmvL~zT(y?DfPIr<)-hfXu&am+g#bQ{P7x9j`y@sDkvAUmTL|virOI$&z-VOd{rNApA zZy4p9h;t(*({%gectm|E#x~SqQu|n%Tr`p*D%&t&$dmh88V$W;K4Q9a3w}2bK$@Du z93*GiS$+^S^rj>Iyn#DzpF@#$YwXM(2{PNRaCgl^z|5ixtop}++9C_i^o$TCz7{2& zE1CGPUK>4FzqxU8IoLG!a$~(%5A&GAFwfTj8ix#_UdkCWj2b~{WISFA|A6)CG0-0N zmotCN!}J*y7(TWFT^kZHZDlfe+#C&`Di%YeN+RkjBtzxG-6$1p&(3@X@JLo0>$$Vz zuL|(wAEC+oh!~D#t#sbtUlDEyvNDUg5`WKusnhbl~V>^6l)$;%ASb zZf7jrD-NR0xk|L#${#j-T}`2rF4N;$3$T6F#Em;t1QmPl(m3NkWYRD|i6)fASD2%Q>VbHx7Rk+;~j~N3SsDG0&t+n!H*(@Xek;xq2JtYC1H)+wk!$93*)u2rKI-F|z zfGN`yQS_Aos~?`F$5nf1pR5JR7JkI;O(k%~G)iQPll3S+nqB-I1$&-i zC(DfMO#A_h%kq(C&ZbbWOx!4tivBq!XsO_VZdv_UqcxhIg>J#emUWnE!uS}~SLhIP z9lb8~L}TUvj|*SO`f$&Z#|UX0x@7{3r`SRrt3jOk&KUYxBk*Y1KIqu=0u?mV$)a;3 zIbS&&ba)zq<$B}tH+heWYhM{>Y@XA0S ziffMq?+JqVRni?RUf{CuL{g-xF(s@G5gO zIOj_N`A%dVy_uT6RYBJJPmrb90_Bz3yhU?64WvArd6U;?K;W?f-p-|exVc6m%r~ga zU$;Y@ucd56sQnmR<)(16!Yk;@h9l&wdK5!vyvGsuZ^0cA7PETYi*8mb&^7iOgtT`N zS3Qo7jMw4h$L+_t`IA&$D+F?x{TsN^7LKHL>n8Oa+)bZ$m62j+737x*&GmvJ;~3zv7AU_84%rE}$gN`V~}HGiIDWni$1V z#DQk2T&qGCoX(KWB`5NjFG!)x5&P_x5f@^?80AL=@YOjrY^m-B?biDSIS#8~Ze1+* z<#9T9+-^G5iLc;qQjp;vR4}HvJ1a=3NSWl`#9)=+C6u}xf&Dkt*%;J<8L>^g!W)Np zf&-msBKsCC8`t8M9dqfdp95yMaQNx;EWTEo1cAn3_$MkF^R0*Q^<5sxh8=^u$BMx3 zh&pzf?*)IW!}ugio(^~GP~Ty9()A@uZxrB<8a0wX?Im+GjY@;AUIT;uC2qX%GkbYP zV#_&e30>SfcRVF+wgRO^{k-4TO=$M!R+MJh8Iioh^d`F)7p`F3koR}FCt49a=c^j9 z;PyeRG+Knkx@Tdl+bbL;YD+Y!4EHN+9bO+SxL{?_9gqd97ld&NI2|OJTOo|)S|1#_ zO$xhMWOS=0x3xHrr+Zckl+~U?gyUbV^OxekU*APq(K__#Z8%-es-Yv3w$hW=M}RlO z22`sC0S8K`L*P9M*pEc9dB-Vu)O+ge^QDgE*GcNwR@~!n#8WrTM~_)~aCDRBfBUPB zc!TnbjXcqD?qs^zg}rmq(eTfG3^<^GeqxDSA^#z5cPJreyCAHSTnkSpT;R2SwT3mT zPH{1JBMeUM9L;+bKZ=@8$5Fd)-2d5sJ&)dlM{hC=B5!u#`B}zTZxqGq-=(x-Um6+Mt*636 z`>~09^rJ)O8Kj-!@%5g@^3BSP_+u{@Ldl(Gv{))kQ%5F1uE#mvhweaw+EZeD-Qe+j ztZ2cYmIQb>XDv@|ryTB;oP)p3rlQ1iUueG?j1%n)=*@jGD!yTl;X7S$ii;M$G8jS$ zxs&)OeIKl>x5kl)3bgy73eAj2$F{X|>CazD`n@#*BlOnOzQaezwe1@|sxT+5avhQr znM!i=Hj?SAE$l2t7$bD@FgEtTbBcf0mJ=(Y;nIN+ysBbO1%s2xo%N9~UZYJl6CdCL zY(e8EqewaGAzH4hMXy>}+PUjC?qbfA^zp1eWZ7Ft_~j1`<~m%n+ZQe%`X@JZ=>P;) zH3NBiz-X2GydPsfb1u9_o<-XxF5|X2c$e4Wk;QwcQ*k0~@0v~uE}1k#dob<`#@g{QotiSygQ$CAmd zP@kiR9SzyIdW$u{)o-}#aTfgznk_CAA`ZH)Qsi}IBJG(}gyo0+;avZ&qa(h@(B8v* z_!$!Ta|fO?m%;qwi*c08Ls))B1NxLMphoz3(srJMZ32nI<9qt`@u1y(jslxb(cMki z5chroNKTPQ8S8}Muiv@w77AsjW4F>(?)AX}v@DuQ3+L;?Nx>V~#TXMN8oP0geI`~q zkH@aEsa%NUA9#Mu0#%zff=xv+q%37UWkd65i(eC+l~g3_amP4$*$J>|M+WAtJc!H2 zjwbtkODWVf5?yXFj)S-_YWohLq$JR>KqJ)li$oQ>HSDZDnetoGsQrBr{=QJo`}}$W z`35TET)BnZ!1I}O@|PLS%Jf0^<3jNBM*Q$+7x~=_p3h9F#$Y2gSsp~am?q#pg>tjJ zrl4?6I`YyNV@KU@v^!8rQin(IM~6z#U%Qau>lbHyigL`{X>fZEin*WV4((Wmx5Adw zvvrqHxor`}KQBSuyiaH=Isl8EWuaAh4wv&x0Y_$9QH8QK1~vYL4Sw^mHcbyDWinxr z_Hj=CyEw{$1x^W4L7whc?!D3w7+$EszdMf6(JwA!y7vae-t~b=<3eGBMFlK<%VU0? zWw0(enKRNArEOD#sP^Jaiq}iWYB2?}Ot3-ez&37>{~46CC*!IgSIKt9GupdEjW)6O ztm)IctnTd%{tw)+RzQRd`uCHp;Q#BtIP2PnIrSQpF32Ya&)~u>Qep3mEKEc%%5CT9 zx;gU^4NZiXF@s!xbpZGA(m$9legkjT<>2$ckGL;!BB`@_>+C%@QSHkgl>BIn)AwEH zt{AOEt6Y1sKcPUYnHy@G)L7ctnvJJ7xZ~CO53s*_5}woTL5btFxL{sBNX}D1%ak6l zRcXUnr?$Y(7khb@n~J&DyTeiCN;17HGQ-VLp4>kp1i8U3G0tpauz`(*GM6Lx5`TA> zz>_!H%ujHi>>T=OthhS4d@H~*)m|W14Zye3I7!|DX6Ya2{xk$b$>xc0F6!EFZN@Fp z81R@nS_ywZ$$>erTX7q&?06Vwx$On-@YWcvp_kS3ha6y~jWzGNU@qnk+M?`-D!lLV z79JU!aBH*`xXf?cu_i|eU#?hz22MYy<(>+ic%1;pbB@6Vx3b|nr4w5}wKlAVJ@3+Z zh2E2JQ%D@_F!SN8XBEMZT7qfGo4`!X3*{tj!Ef3EC|>Z2bGpcSbM|_}2Ip!VZ6*w3 zPORoFOB7(4Ddq@HWxROJ7Cisk7JvUf4EwHWqu~2ku0Cc0k4y0z9=~O-Ioy>iM`+Wh zne_C8Bi6k6gb}Z#@#(xb;B?29Zy&XVe|_XM{#7}~yRN;7k4CVZkK+k?JaaQ`V2+oc z-C;0Alm|8qvQTrCJk>j^DThir~V%)&pDxQ&PHdk{|kF?I7z-3Eb z;k*Z{Xj@nk$qtU9ITurDRJJ?`Fb1l$bSEb-7Rq~kRGK?tTZlGm#$$WyHW++xmiO}O zA#OqDby)fLASV`Q!+{ZFcx+*PFY5w8?MgEDoj->bhL53zBFukgH=hRNDo{=H7P!vO z9{yZ^qYiN=b7qrQ_gj2=;}`7E7bhj3Q#k7RSD15Jc=($9`)*JUPNze6XOfqk5dD>O zMQ&>!Y>m~##tb2fGR{L8vpg)d2%_lH9$am{4zsUB)4a(osP}R6@W1iGdpEDaX$eU9 z>fsCSE-IW+qoJAj4bFoU_i()!KK9c<$98|H&S=N&&D+pAC>C0;IrF9lj^z%m zeGkr~%kkQqLvZfb8hALcoKEFS(;f{UIDh&am()~g5Vh_pSL|j2roB1bCB|>kZ6BZ; zi-PI6iZ0UAK76LV0p{#l43o}uVCyDNdbY2G#%&(McQE>i^W&Ak>gsfQ9CsOitTjgY zdUtI2F$NEA*@lT$Y4FV(AbFA{d=)IC4(s(~A2)9J=W=arH7NWuf_t>M64ojT!5LSn z;p^%U9RM#cC*ih}tbcE2Ix6P}g2<2u*8kea{d_*eEqWzRN7cWA??z{?^0PBsK7Wgg zlF5U%unk;|*CU?9WK(X(EyD*SiZduCY=(1KLe!{bb1cm*m{L!z5?M@ z&M7#5B@NE|Y^74hP5Rj}2JeioMvp+oR~3B(>%_ykKH*%0kkxj)whn(>-#>r&=j&d+ zg;HcEP=pfm?wvYqkZ$77V-jjES!gt0s__?nbNq=8mT8!_w3{)Y4&!#Y6X1RQCs)^$ z3f(_0;tl31OX=H6%Od_lOp+mW9|)p`-=Wx+DnzFjox-HmmuY*}1w3*Nqp zb=pc>o^(@ui!x3Ycf|=4Zs7aNQ8eb)O-i%1C6@;m;Ko~Tn#l5;Wp07kxv`(q+a`x2 zKZ?Qp!_oMA@;LJJZ-E`*66nU9@?(t_U~SfM3|GI2sxzCw>RK+&snw*T8OM2t|LNiO z_q^kdNMA>XKnsG*v)Ie#RqG8_xbo^~Ix==5ov|Ol_=_8H^yt&ru4V+!_WgxYMQ@ls zT7#}lo=JiZXYraz2;PaPLkA&Idf9AGK0Z=3*EfoWo=oMhGnvQdF4j=v_G_eG_KbQa ziePS71$HeEAdi##X@I@Y)7^@2ZcrD@*cgHWkK(Z)?>yK?j6wC50k%)G2ED%b;PvGf zj=nUHnlh#N5B76>i)*7mZ_)jga%H8y1Y54Ks~q zVAGaC_-OBg_XJbXDs>5~eY-=&s9)eHBaQb17Vu&JRSMqRfD0DfgUVDsTKU|@I$v2l zzDop4D#qf&V||>jVkgX*8-n)_GLPvIT@2V`fmf$?z&B?BWgbsHqCTjR$ol?I z0`)|D()<9RvFcuQyRL;!tWKqS5ff-jd@+n%ra-{|jy3lT>3;qlSoO~aT(&)k?oC`k zfp=ACyFx73e%Axz*NR|SE{QQix!B({fzHRV+Ca}t@{i}yNmEhWyNQqUPK`s}E5@QO z98XHQG4yP+A1CPT!MO@cqDQYAX>Hg>i=J-})9K6PrJ{hYO*XRZ6be{V(2l0T2?dkdqI zK9bzv+sOT#B<5^Afb_{-$_p~X-Ze&CT&zqvtKu+wraXVzZ%gNNJMi}7gPD6x;z=tv zBLBJuUvJ5xG5Ov3T-kMgy3Uca*IDz{J&v4Q+J&{1Wq9AZ`hV-B?Q>VWxI36@#aYP2 zMGfdTWd`4IO5>aDoKY2>K^`(+Xqc184Hvc5>1Cx*t$TwGgEV;QaYKGRq$^9GM`E?Q zD)!D?hn3C-Jo>RRPMe*?$fri4AO9H%#fzx>bQaQ<7hz2G8hZOf3R~s3$WUvn zAvpurfdf#wdL(jo-N5p1J5hEg8$ZsCfYw-3p4KCt8%q`V`HP9@-n|1toMd?6k*ly+ zc@Mft{iF`fouxHbKJZ6M;U#_@q`BvxbAW&Q|2v=enJn@)@2Ya$&#_#5xFdTU-HpEY zEx5yRvDJUp~B<#P%od;+i0`+Vr z-C%djzR-nqWmBLzFM@wW{=xB)2l2>70UxL3ky~{HYwf+pJE;RzcIYpBSeQj>nkv|$ zmOx6iCfMv#fb?2lxY~+6nPXFU-;pFNEYjrfd-~(t_arRU{(;3uW~$CRLD z&>_g*Q^e(pi5RU`&($ZEbN_zp*&^Eu+oTg&d3S|es!`1Cd6E$Ll2 zxV9r2jSkYQN5AmB!+X>FJ?_>!0purB-(CmD4o+m8x(vc40wdH&>4Pmb}+ z)BwJrX#zX9I}~g8O^BVakLqVG!Lj|R6e4D0bq$Z@vm@gXB$FrU<7xn>wn#P!+`v7W z^4YgogAL8@pz5-bRKD;&?;AgeXKlF5w<r)7HbY9hm1(t;kbkfy_=~1^DnC2egfZay+du; z5%w?g5~_AJQj^s&PFS~<{Udaw9do*|nYAscoK1kXb0reI)=`s3Ao2stp)U4c+(d@X z-+S?BcsR6k_S2tJb8S^5I#jY}lTcQI^Jb{Ln+TCNKz)xH;96thSc92>#%qV{goE+^rbkqlj#sgJr9!{LTYPveu_`t9@2R&`qI>EBKJdNNd}5_>-DDNe7o*HuJoOHYnFQOzt6Q(<1toUB<>+{ zdfA1ZSG68nBn;uaSyueQ$&gQ8xr}}H#|aK&eo*Rb z55DZ;%SQ@Q#H{rY-hfO#`78*BTMlsJ)b0Gkb|J4nyal0Y+o^I;Fdb?7N~(9}V7}%S z7<@{gJhT58F@M+l-`C_GTkzD%8ojp^z-Ow+dAIAsE|y{R{ctb-x?wfn+GWjCi;l4{ zK~H+`-Vy#jwSW(IpU7A9g1LQ{HuT??$?->nSgm9~pU9pe4UWvmVIoK{*b;c=p6j ztf*6hX<4FX`nM&=j33MkUG2E8XD>bjz{_6qNS=eIrh-!S9Wk||Hb#ZlqF&6cY7Ov&S z2|b4{6aGxvCTcuWk@UWp#-^Q?G|gCN*S|DRQn))B$MpKJl1(%xr42&D&@;58N*lWV zP3RWl!m=J#T>QC}4)|)L<0gBO-Tq0axo?T9TN)91aITnd(32N;y^Cd;i{M_U#E)8( z`N;G@Y2=3z&fYnk4KkcDIXEBpotu%AyN_RJ{erKR;=kvV%G9Y;`X`?DNy7hK!wbDD zWPdFYE-9PHW4tx>-Li^59(;_u(U&N=F_MetcyoTzF?3qehp#O>LW9Ry(LjrC(z#Fc zq_Wi}Qk(LDtnMo2C;T4DLxf5?SmH)S9RuNCYRi2!s~|IFld#uisL;6lim-8~6Ae94 z%;m%6rPJSj<>4c@Qv5eVx?8-wN6A)gKEKnKRUgN)`@XMupsmF|@w>3wd6-z^9%EdT z9)G{^8ir?7u&$t$7IRI`O^LNLh~mB;Y8oV4ChMj-D?|bPE@40Cxt@&xqX6y z@mMJLI!nv_=3>D90+DvoNH8ohRV8-`_ zQK)&FhfBT19P|DE^#5Gcdfd!!@5Ba_;~`-r(@n!<_~(ROKVagz6aUSZrT6WV$*P^fqn zM1x`uV`}yeOf>d@yQe=qKQz+N5xuc+>;jrEpWJf;?G{QZ-38xC4nmoEHngg*(^98& zVSwy2GTo6OW)5ixxstA!@hI@okfjx5Kf8vrT6l^k@&6)Hbk0>&xdJ((u^PxPm{~^2ua!c7bJI)sBf2p65G#NlCVlW$+Oz`LTYLQb=fk39;YS? zJ5}0n@7qPBg>4~+^PTBS_z}rnopX|I{Yxpar>wX)_YiiR|4%Rc_Z~M~8G&7$U0}U% z134dVqZgO5DTu>_X%&aq^hqXvj2g?{AG*@pB{jm!euo9eeHQQ%b|R|fG8P-JL}Wo4 zZT0UeB)`uUWFSKx>r*AKWq0GeZ4+b{6V@+`7Pjf%wsWnHv5TIw9mji}q=VtU!a$$J zJZR$)jtKQ+y+w{BQ}B)yf=5VtFpZa0hJU8kn zbdJ?Vd%bo@!BJ{ix*z?Xo~2*cvxPZgCjJ6(?S;29=4tzjaq!4`d}!YtMai1bEPY9C z?>&SKO9$c9y>OwlvzuVJ;2hn4BKfzrUncn=sW2Lio8m!>_0aFf5Xx(RK*D1{6Gi>< z-}RHUT8oYPZb0}`W&GM;Z zf^T+)pyJq$wotg_otKKlCF=>zU+tGn3vwTS|@(yNTTFMf7lds4&7gMkucBz@Az=p`pHs*6uux z_`Gj))~F}#6TkbKWt))M?hCzcPQ!ko6aJ)rjxLu+<>Zc%aP%##n0k@(@Eu95;>6zb$(_7jdVXiUQPiN ze)+nvywiHT@Vd$^F$1Ji)%&q>{!%e_&YjwRo~I+7hQdPJ9|bm+qIg9|QXJl&7l>>) zt2@O)mCjnaqB{{)i_hYZs89U3XBl?&{7AJaAv__j53RkJPg%d-(3;x;{5(^cmziqt zyUD4zb6*Ri)7t3Z)K(bHGUMO=-C1GYN3>dsGqH1GCRE20Jho^mr+V9 zwmX1hJ(F-a_W)(qufSwo3qGgv1a`OM$SqG5)`#73VDUMe^RnW|ff2BYQ-|8<7x400 z0+rkQS*iIY>-esLFtP04I@|QMOc*IID{abiW6$DCq}$Mf;V<;jMVy0Px{->hK_A4N z^zXD`usZh}70)eYg?ubBQ&7+hrQC-Rfgb0+F&DOvN+YH2O+g~guZV7uB_)Q?39YrmaA z=sFd)*X+ffpQf|Mx1qc=M1k+u5IX(XhiQ6p(%a9Qcvw$!J~dZHdPpUcS9k8o%b?HY z_Dgu(P2v|5oUq)_j*AXzvD}@H&1d~C@WWNrta9ZR zO;&8i!%ow=#@Ucpya{9PP6Z-|g6`boaA$7bw4D6jZKV1YUj@12 z+H^lYhqnigd@0ylNUI-+(s62h*vb&|Mn6Kt&>VC$6@AXK5x8iQOYP&PBC37% zzqQw{HUYcdbVJ|nPw=hK6L&%!S;(2e{T01%_I6)vN&Q5v2m0_4kz<@y?1~P3oWz-0 zM{en4%caWw(2b4Bar|)#{U&B!3$om0n;8o$_R&+t^LS*H0=YYXaJz0WHrs52uE>J; zskR4dzVbXUq#B8L4q)4Hk+Em1jbGDtBT&qM)|fqw+?&mW`Ula}OQJ-ro!e;I4;kDO z^`FsW4#4oqsGJV-zm@$feTj zH*owz4!c=1jV{rKe%@^w8a5XjgHoZVCC|Cq8Zh}hhuSR?b5Bd(3+*)LBP%!(o4>8X z;HHDZ@#W6!U#Uj^7jlK^U93py`#kPpZ_e+x4M&eSe@vPY1oeKi(PxDMY^UF(sSUOy zY>1|e9s4QF*dLd^{Gn<3|7DSRbuq_9O;@yj7W*abPw^CmLzw5H$AhX?3MIqyFesoW z>Bx4%(PLZ4RP{C0jjtfrJ}QV`T!s7RM9-?IVW~TQ2%+sW(am=pZ~Aqf-6wTKsK^j7 zG<(m(chA8^|8Tg381v|*Nze(qLBm6sAKG7qR!lp@+ojO@A>oh~92L)AQgkRjLS4+n zJd@H`9BZx=kg%7^x+=2qj>~v;)Rq;lTXK)0C0JCgg2vG&VAXRDl$I*6^X_u0>@{vJ7HpoyV?5P3O52|A_CAGz<7t??t|J?H>+*dct9U84l(8@j+3SyG;F(*y9gl6PW>MeCP^nEkGhlWUE+WFk|ISoiBsC)0Vocu8NkVO)FdJXY>?rm1cEJSJx_ho!xO z(d5W~^*Z7GJKLVuHu3S_6X7oIzir&BSvl}NpT3cZY(?Vo)XPYI(h0TI9r*s9Jm~Md zPt%+iv?fGFsPF#=TJs%vbpK{ZKBmxqQDbj*uQM*J>_#TtTZF4)>x9eo6+B7P zg5R`g(bZ%P$+u+(X-vB-bb8TxUfaA0YU8bh1JyS$^XyS{%u~c~)t+3Uum|_!KhUdv z-TAe@J`WbPMH^3~LUT?8Mi}?SqM5qV*ignfZ0>_ql`g$*mj~8W&=2KNsF| zz>N=le2!D$8fYoqM~yKzc<#MQp6B4kuS04u^U+;VGpZ!rpOuhXMdt*cK zQjXvVN!aJIifw)zTKFz+rq20q}r^rfP{z?=d$ zPZs)dsbuJ@Uy@s!;%G^UE4-e_vDJtOG3QGM!JQ)Udv}(gZ+=4de?Z_wUf04<+G|BY-u5jA(T{!+T0Ubnh@NVs|Jkb0sZrVH} zN0Cn)ayuNms%z-&Y(;vrcn3v_tddlx2C^P#M2ZW-c-yIJem=;H|BUiN!~6yzZb+IW z>3Rz`U419q+>tKKRq|pF-@8KAxnC40pUFIaE`PA!4lBD8uvdvjR;VFCH;PBD|M1^E zx>@$`=R!M?SRUR1lHMK2KjV&M-HgtD zFkc-fu>Hd}SMy@1Wj_25h8L3FRJIaO}1UL59C9%3?3VyK>5pN+x5;}Xt!eU6jst};C#w zfPECZ(hy!ea1R>b&(l<5^_dI zgx`JhP%Mn&5}R0IS9v0>{h%F zo=i(F`H39ypMv$+0YYcfsggoBYjW^x73SS~Ov#15xVv}1ZI!PTX5MZT@*EwJJZR*TNJ^5qnt) z#g_E#W42IsV+V;=3y}#a=Hss!gyv48@%4BM9hRPicKSycRn_p?$xiUjOb~MqB?yiA zKvuGbc$n9OEgin#O=~(D_S}TV6(HwB4?H^NMMpE`aHm%(3>vgWAJHo^9h*ecob0Jw z)XnY=QG#CVSgwwkD`X#zqCWV5Me=ChZdC4~DI0gw(1GG?>fYPs<5`_XmB zQd*TX7(oLxpcrusT}C~^#OgBox=f0%3d4AN`x1QUJD>Z0*}%hmPD7>pSM2Ze9Dd&M z6gM~o%U6l4!|47dT1PgG z2QXRWjg42=;Moy% z#;;l#zAI|OXUa}O!7ODw*%ZZE!`1ml=O^^GTx2lo%JOuh_WX4JQk*`v5Up1RaeT~l zwp!bQNO8`0Za_zjDDmLn8Lk|j5KTcV=EFFl66N9y;#PZ+_}__zOA(CX6pZ~ZBhZ-=#a@)1PO zu&_%D-PLTRUu!}|z5FKZ2;a&>rm0HrjCA1xbNA!&a%*TEU<5x;#<@QYc#yLYgZ8xG z@{dUz_$~>R^IZ`VA&>0v?mTN|6P87PppnZ?2*UMI7}HaiC11q)zeZ5|bH+0=h+pNx#|EopNQ_rLB3)N_3iVNL}PJr)f zMYcZJ3oc{y@L~8gIG%T8t)6JEv&KTeio^>lG>8oh?Qm?rQbD?e5&wxiW==ZLyvqKNPB zg*~TC%vM$Iq(=(X5)<*P@^Qf~jMaZ6tc;I^$^%ty?KK<=E%v}WDHi4}CZt|=iW0xw zMB<|}l-9UOupXl-jQCff3AdgUA1 za4Qz`f-^)f>rQM4vS;sHah9U_l0J_g0NE>A5M~}F8};o-3SK5nG(N~K^SAQQ_pv;D zYXtvnP?DakOAy617qBx%^f*7iPHiGfF1_giwp3eVnyf!^M<{T~s9;t*(~8xjS|B8q z(CdV?P@mTe-JXxerCEu*VsbH$-Ox#T!MhZ-zF&mX8^fXT%8&=FPk>*aW7OI`3Gwfg zcyd!5zGy7R+QvF+3sAy$&-MIcJM*Op{s_BNPMsq}jrb5fLAum{Vw0s9?R^?&XINqG znj|q7)|q=aKF7yQEB0kP4aax8is@MN3WQ?S6`BwWrO zBpHKNDvyg`SH(k|UD}8Gsn*gWvl{w2;R}(08{1A-;<3|0r4wJwkefh6`-!|@vmdXFbfKY2?}dER7WypaT;-haC1#Ec7JQ84 z;NGzW9!etLv}gh%H2b5S#zk^0a^uh5o7qtI9uh+mF)37umqw;i?`R*v>RK!W`iANX zM>5l~5wtpa3BMZh@jBuq>ppJc zn%zA`pKO4z_w^~tdpen>>z(Jx?(Ol#$p(sJis`&|8P&MlrM`zd@u!X*aOdGKq_4HY zireeygN7>wIa&!npTr4T3j2iF2m5frm5FRLaStxoT!NR5r8rn}Lw$ZG9_ZOaeQYi! z`Ap|KMmbpA!4AXazoOUpNSsa{i*^b3q4^-2cF2a~*DP-qqw%<=qa&{|+lOb1?P2Oa z4#y58vab9B?)c|3jusx}Cqr%c)YX@GKl>2vJDV>|c@iy*Zd<@BUZ-=g$O>C0)~NS( zt=OytKRE{PnjmjO1wV%Tl`OwFL*wYOw#Whi2_2MBOAR$q<=AKIiG4)EcE_ zD!l0NK9S#`iHdM@R`Tx87JaPv-pLr~7pS7Dek8u@sB*Edn5Ap*O4P2sLevd^{?zRj z8!PEZ>8GLe%SKCS)bdDSI-&#~u@qSjXP_-1I_Dq@*i`8;w-GFt@q<859C zxmTYPeA(HEq$11yd2udvyLAP9CdJ`V?mE%;?8xKW@8kO^NnB>qiY9v*R=jc)Gb@{r zXuKRv-@eg#v5)gkB=dNG?G5vUR_eU{q+n-wn4;E%(&>Of^t01zFHG929jFK$1C=bg{kF!BOMO_qbDMad zduaggcF`OYUyc+xYkB?NBly(>;T-!*s_$wF`%@b6YUxg_UA!OD-o4?6v(EBUxn#~B zBLjKUSHkm2y>LRT*N^i{(UKy8ZsKe%(;3S)?rS*u)lzh}*$mZL1EKGjMeJ!S+#Mli z0o~fbgKB=;<1BhU$P}Z?+mJa@a7SwC+1z?S>4XcvJC+k5d0kxMqG;w?S+K7m^oZ(wiDk!&#AkrzA?d&|3;$?-@f zJvQqEgFZ1h8&nEw@mhvNmSfe(Q|y{!!gS+ zhwjQKDm6QZH6e4vbAJ~auH6mEpFRIwTa;joEzRawFzyyzzoRSWg_OciZwfjpw&MaS z!5;OAaIF)$c~2K`iOBnkN%6o;jYwe5H0+pJjJ@MiME-3Gwu;$nyXw2JO2ld&7O<0E ze0fTF4?Du-k}sdpE#SuBxv<#k2mD-!sl&>|dJ+Itk*n5gNEqy{zs7|)arRWd93*Qj zX3_TGld0Wlz{YsHqwji2wjS6f?*Grw>Z)1){^w0)&m>jNFNNKyp15_c00Zva!9^7_ znmkE{+NFlzPm6dTW`VO^Za{u;dmg`{2fy4Lz}7iAbZOE$If*bmnLCx{2C*H=Ggoh=bUT$0j?#*wljtk55I(_9*s_u@w7j zRWy4}4a!HD@uKBpuxiI;ZV9)Pc1=CckJyk`{g?>_^P{xkU@SM)YVgmw-w|y27PrK7 zn~ZNyQ6>6c-iJA1Y_%D6Z9kNip9QjS`UY(5W>0Yw(}ayT)hKg+3ed0$!E%wjHRU5c z8TtcXxA*1|dyV+QmUcY-`U`yZzDb2iU1@9KQ6XcuEqwcIM5EaUTq#-xlZ7MMyvmZB z$`4Uc-E$6k`-cq&4WU4hP>?*T9SiAys9=^N&L}U(_c`G_Cv_^ncW%LEbuY+x=96Nd zAsEB^@a zSa#qH994oTz`O~AZVljZ--78|=bu7vlLo=sQUN!2cZQomHs#K5!-+CWq@RdFtZ^S` zO>Lp1w}*w?1b@EWCVoHjH)EN*2Y*g;!ouX8aKAbnl`dhVBTXwI=)|+i3pvO6#Xd?ofGwW&e9qx+F}Y7k^iN6SDTGoMc>o;9q9GV zh3;4k;6W~?(rpK_x$sO8!Za1xZ1)Z9F#HHps}gDrUoE^1EXJU20$V3$PA!3vx zhQ<%V$9EA}T5b)8MPE@|mo1*3*P!NGI5G-tA=}p)g$q9m`>hlB`iE8g^_|iGt3O>& zjbkr!DM!xj!4oE@BlUMJ=2^(H?XAu1nbV#1)=6mIi!zdIkb~N&(I|E!c9a>+SHlaD zpS7JIY0TrwcRw(Ds+%|is>bj;RY;$D4q0W#px$;ED&y~qdYm#Oj$H@Gp__5jFAG20 z%Ze=d?ZUzpVxJ={m+sux#}BzLA}3f55)*6w)!&K}SC|P_Pm8#|!zl#KFz3phL-^z} zk@5C41XE`hk;%N8V`FUPB&A8Q{3KYgcr(TPfY|UZ?lcoId$|Sr# zsLf|=EqLI!MwFPk;j60(T}+%n>Ut|^~BuhP||gS)%3%8V_nQ~DCycQnwNs0bdWJb`8B_2*?bTWH^` zJX#yCf(l8x&@^@|@oz0E3@zo1^zD56QYpHe?Iom-kP77vxQ; z%q%XQ=pY>uut<8w~lGXf^+GFUCIhAEwLgkipCY2&#f&KWdJ+Oc|ubk&>R^vxum z4L^P7o~mP!@?*EqvEr9x!lU`@C}vphy%vS##Xh7l^gEPV^SSWNVP1Y^3XeHF97+R4 z&Z5d0tTnzwL*fl+NS9C?Id~O0Bh4{${AbC9r51J%pFbD2%M}apU!>$xYarFj_2=-{ zLovy51BG@P#_f#dr7EXJqh+0_f1f)9x!>f)el+t?J_$CA#5vDm%Cs^fkJMR6olrsN zTbhI?)0XkM%THmkF^!r^Dh*h%y+Jm;$R`=qO{ZI#{^CSs(iJWiCSNv=NoBk9yLgoU!g62f zwIdPIqmLJ1RbGy8=I$%I@a>oRNY*&+>!lIF=Jqx4zH)6++1Uhki0L|=rkM*J)L@$OV z)%c9yyZR|yY=4B)d=6sC;+llbA+N$lRCFIPX3<@x2-TrB2Bnr;6A>6HMyY^O&a0mpIrO&mIG60ckswIR&I?EP{i5aa26SikE{VtGo7l_UsVsIILQ>*rLnOV%yE zdGjg8sVno2`IcDY7Ro*a+4xk_nU?iw!}K1i|7yqf*nH9}(hvq%7*lo3!+KF9Fj-6J}8`?E0eaXjCw0e4h) z!tl64o-oZ+8ladWYC+}jt-(^LnZDBYoox=8l=zZ@PKL0*HcUt`6>K-JQMLVBzxG$F zh`rkjLfvy0$-Pkn?S^J#$4nHM+KER!Czj+Z9xdWcjpzxCqc8Z_-WgVyn}{PasKr`*lQCADx$wlfybH?mY% zaguHyZxAYHzNJNh#$x{1Ms_afhT(hm2=cv>?Q-MCNO~`d7ux0z#Ka>5Xm6=ljD*xv_yIUee}~!i5Hh zd5(`dr>{9d>kI;*6nhc_uDyfafaBtPf0;04T!X}<+aGitIhc#wJo$S;9^#gHQRk`E zXy>yP>pDbW(~1P{;o!($X6B*j$uLyL#NwCAa2!b(Bn-&(7Y@l62sY!D1oS^7oUOV{ z?#JyV0axT~{XbjT^&Hxs*TyN~>h_DI-dYHQ<_IiSZKU4*GbFF4$V=u8i=Zxp6;Nlr z0a12l|JGvRN;zt^Sw^!a{2~L#3cF0#97zY6-jWHzYsvT!JF2M`&m<%I^WhWUFmKyH zUOqPk+7FV1_gmwk@H7!?!c_UpFqRhR^pLh)Q|Fv#wtUp)2&m&+cs}Zjq6glTvG6P{ zd2hlgKI?#5H(J=WH+zfE9!<>P8_y)cw4&d~B`WSqF6^TDhEtu-sLDc7*M~looEJPi3 zx>XD&Cw@9t6VuO>W9kSi)_lDc0i?loPv-Eueg9>;WS^DdW8Wg7pq(1}E&yJ-HSnkd z^_*_oUaE6y0Civ2Mm|yNMO~>D9&}Dada^9L*$t8pd#MVk&jVbja3>?REwtR+0xCD1 zNp9tEyuWrC#&dd8m+DcJGu?z9M_(a_?u#VQcYz+)+D;j*i3W%Tqtt4{t@N=wrRL>;W!b3gfxWCxwVVQ9|LYD-;y=05@aW zvA(`NZ(p#TLd87k+-LjAu&foG4gFa)Q+!rzllfP7`(0ELhEEdn%#JCu{tsJTG;0~1 z$zDpX-%et+-C(F$#-VlCROla*qT#YJ5B72pbz-UX`>6wc3^bvz@I!VFF7K3Fu8R`)WPb02uoAB`I{?8k39)4B;CcnTkP?7~mWHwbfAcctK( zj$(f7G5D=j;^`N?d9Y`om@zBveWRTa{p=LU+P6pXvl^W56+sJAQ?cHz9{D}*k-Dh! zws@Egb#qyU1oVjaTy3yP6kRgY%Rb>dkgh7|wE z;a0>?Xga8IQ+*|kzT69r)!xu^=|Mtp10@_z!@ZyH>FFjXwDy%C;+cUk=|wa(wA&`` z8|1Okq=bYgWtckd2;#E!IpxS5EGy1{;?be(QE1Mg6ZZr)mLWjx@ zZN{$~c5EZgfA!^N!)NSMR9-kHoccVM-X0i+%NKjYPScJRJe|2Ntpkzy9-vc4o@aDW znB6;q{0jcl8UC$3Z%4Yx4E?Z|M&Ag=5p7$3_`aEP^>ksa6ob?5rsx)9&I%%*IYJ|Y zN<)CGQ6Z?S+Kl)+J$TY>f1bJ|oHD$pqf95ATz)!W#`KSPA6SS8xE7n1thD`Kvz9~A>L&yP8H?S46oZL7_P%jA~U=~MgspK8ff381JYkB z^Y29~kybewLB$X8GvyP?@B6^}bw50Eo<_Tz#tXl_hoRo{H~q3Hr|ElpLRq6XZhUpX zqA$%Wl_dY&HDS^F;dEGJh4?O5geCbyXsk~f`N+NJT8F=!+TET6pWDhIt7SP~LtEUXCcFXZ;!} z(ENq4>zNU^TR#KW?6=U+?sp*P-GL7zC&0<{7uqYl!0_(6SXgqOE_OaG+zK$Ir@t=X zxY{mUx;7Rz2Gdbt6r8rUmlgU3ws;cYFh>}tJ}{kr=w@7jY>=O9QI4MvBt)(DwckJ`C@m@V=eqYoWM zhB{&Dv#aDS^4!kqWg*Eg87&)YsmQY%H4NNKecmrd{%lv#_qR>d-<~Fu8+U1&dncq= z=i__zcvv}pB=7IGXo%>DTP_aRTJ4JcZqAr#?Mf9<`9Oya(7xk~D(yz<>Ge>E(TfvC zneT${S_x~J^!B6wEjDGmtrW35-~@iwbB zx)jvW)6-WmT(EC40Tw#QA!&gg2oBvBr8Y_Za^Z9j(ql)=39;TP$$NJCwWpyuvbc z>))D*?Qe*}p%>Z9xrCb)P6h$(nBqd6Q22mnJ8Vt?zq#{zk z=i{?JzqP)9S=PGuo_o)E&OZCud%s_AoRN7P_pD=dAh#Hwm5js4%PDx}t1`6@cNSdH z$^+lg`%z?^9Q>>v0m>2A;oxsSlH)y?e0ek4v}YuapJWVCy~jc8=sDrU`AdYSu8Ps@ zd=qGo0~qgY1+mA63sb!k=-Js0xX;;0K5+(KUTC-6P44cDXdF>R+Uxq^4F^_2TcWapNYrbGk@d@_mcO^UW#sS7AEhd#G zKY$}o@OVx!`dj7D`fs;Uy19eT46cK!Y_6Ag&w{N(C526H&h%LJ0bTg6N$v5|C~Nk5 zxLhZWUvJ0@pM3HXHbXA0m|jH3vo53C3R6^kvkQGr&!$?dE?T|f4yn)hN{^~yNX=)2 z(84J|xVc*oNB;WWpBI{pIDmsa#OadRXArwsN#BLRw9xD-qy$B?sw^ptc{d10>F-CC zWirCVqtk@7E&*uU(}?364$!1*f1LN!6fFdb;8E2JbH6?2db9y({6P!S<_?8CyCio1 zW<4g|_(lm)(|7#Io8!Bf?NsJb#0bNBB+RoP4E(3FG2G7r-HpdONLvlg~{jS_~M zI8bX{2cXP!c1Tj3Y354;0pa#A7c%paq3+t*>}bvzHtS<8Jh}Lg#(fZ> zD{V)>yLyz!ru+2}WmSrMigmC^eE{cqo`8S)6JY!HDX2G^_ots0@%bhnjJ;1F_2V?} zvo6H>w&O4yhN8pLYY_LT2gfDK;XbYPbmY<*8X2{V4Qv-;Z^}fp-Dg2VLWKXnX8OOq z*SRn+{8sm0ZU=6Pqm<_JRAe#={{=s!7e&7)Vj$ji|1~+E^*kL8H_=Xp0eS;JMqO-ZPMaCAGaerm``{$ zX95-CFr_u{^Scrq{PzM|G~dv!G+kjxtc`HtD>ry$B!kzrUBGvjIIY^eAGc&~L@T~m zT^=ceeqQ^*yfuP(sjcJtN*!1h8IMkh*|fw(S=g}p0%_u2(ynQNQEzU6uSzOwcs&}b z^8Ud=-mmzuFNXI-ZbH7^cBZGk9NK2BLKK;gYh~0(>nqn6#GPb0eK+8SXNllweK{Ea zoJJ10EwGGL!iMFIcze+%S`;yh{vJ@HyY{+x;P0Hd(YpeND$m2BL|gO|CtSOz4n6hWVe6f4>d?Cnwv7eMcfvY0Z}oCK_%;u!l8e|_ z!$2H$6>+U%H(12%V?T5inTO~WxVzE_X6T-SU*#)M^j0s-`?-T%F7bk{^;cor*#MZR z@r>1sH~>0U9w0rJ^Bv4e>Bd?s>def;o8SF$Z;mDIYQKVu0*_MijTJPtOp?yVk0SHr zVO&I_MjB0zNzD5gU9%RaI~I2EaT(|T7>No`>zu>MJ|@(r=t{1;Q~B&%70fBJrNF@f z@TT}ZJkH5O)kmsWe90T$+%Urdw`ZcMvpWp9Z~?lW_QJ9?0-SnsBmULUBbiCxarQ7( zI%~I;eX44ObbU7xxw;+HowUeBq>|J z;Ln;AqsO%Xqg7#7dK~M0^9()QgHDNqw^!wHig{K)x>{nR>gcmdTk-iakjXa6>%@oM0F#@w^-b0%r zA&wavjN4RJp{4If7$o!n<-j;5_QV~FHgP|n((5Gg=mv|4*$!z_^KrK_pRND0hi zm0~_NT~nq#2k+x3MFlMKSA~+DHwEF5Q`uCeNNr&e&?8p}=ePer3aKTDZ*k<6K8Gff z25~DJY=6>*mmI|LBc5aKiZ4LsX$k8SP3ZqF|B1sro$}!FwU8RWs?oAG3s}8!7VeTh z0FfeLf=4^o(Z=%2XtU`aQ#tK`aVC4PtJepUnugK}sQ{3RyDzY&9rP~r4~}ow$Gia^~zet7p#Fr1jP z9}`6da42aK+wmX@_WE%R=s#CLEi-D)ip6KiQ{nthIl=CZP{Bdr8c=@t05|*A<6Q-P zJe-h-Q$5f1A8YjOIdsSK5;`^3GL5`+%-du~*OrQtlTQ_TcWlKG@Q`~R`Op(n3*oCi z-p{fq1gTD0RLT*>vG?AhcGy0WDYT-@dY)@-HA9IVis0!~id&}SVw7Sl#IN8!i%O5k z*ZDqaZ*k(Ya}WE^<-l$2V6i#_dVglYL*2n}bekEyoRLg&A5+PE2(Lx$x{5pRXmKXx znEo|5a$^|xsX2^EPu_xhWFu-TjbW#YyaO94?=3fU5X}8Lx|{cRvCs>~JgQovlNo6(7*f>^L4) zx2EEu6EK0q`p?n0coQ_f!TToKO4R+~7(SO9g3l*iM6=Jqc)WBYe%Yc!Z}Lx~T&ORf zTfRt*VU;8pDMf;PVX*7|#QxuVb>~q*aHtfh$nhSeRuFcsI}A#y&rG%6GrVCw8XGQi z?dgbVG;BDp^Uk=5H*@a^@^73oeew0IVDjl#TrbdtnFX>~oVAh`DW{UsH)(47TPdDX z9c?OJZ6Vl|8wG|>mh~TN>!!n`cK-ua$Xbx=JPA57y&Mb=ScAba?s*-djWu&iP(NQ^ znEHr&(d{V5tv7=4PT){l?fMk=e{qASoAn_r$`1GD<)LeUEXv+L58K5v>GPfsWOUhq z0#olm^^*!V!AF9P@iCyXYF}*r@|kUZ_C@e>eY&aVHrhHY@4tTkxo>}p znNjrlC@Njija_qX(KWaby7mRZ!rSZV;V56KQZl4F+#5D?@Jzh)Kocjnu7wp={(``m zlcot{6sdM_B6LNQ=FSB&We{G=PMB+Tt;CX?jLua0%Y6~Nw+4djgFzzCN zhdZ9tNCRoNA@JAER&XYg3ATOaz5Vq$XxeFkWv{h_zDC9TSrvs2JufuPYeM=bljB;WN-7E(gO(%FtLX0H6DG3G#!21fP}}lcdLPT>X3q=8IbiJ5@rd zar6WfJ05`A=#FA>ZrI!AOg;*7F2hn4mh1Jy8>~a_y%yXM-kvPaKN%dTU_EM|o7x?|_P!(IloJ1JN!Ctf{pR zA5juMJYk0GuG-0U|XjG&^N9**C=D$Tj7{_)YJG@0Ujj zozE!=zdzq33_YPOY$+H`Sa_KB;up-<4?{JvHF!&QBW_-hOP<;VWMj3G%Eq3d;$B-} z$2Upg+`wK_L#hAvGR6*j*uORuTxCGy;t=7?4MD;sQbM6rygAoHw82Hi7!awh2JI4M zh+dV>3T{py^^pqXV{3?mriu5z_ha^4LAA@eA30LBGqp+T(|it-xW z&mrm9+A2%G$K9ZWMbm_Ka=cgFH4m%Oov^H+1NIJmj7k0nL3d|1D^QCjvrV!1qt7Q0{k^0iJF!1tU_EuO5BX3Pa&)ed(y)sxR{lZFk z$1sU5DsQ92+V}Y8i7Nie8-ne*Avh#K0T-s$pdwIITci}<*`Hv# zL*}!$Q+Z!ZH4+VF9s7^l_7l8ki}Ep^GGp7jAUlT;>hPj^LqpEX!byuH71@4 zHiTmK0>Z=cfp{=|YX5sJ+bh|@!bYYUBn2-PN0a_=b)jpJ@c;DRgS8Zd(u#7z#@9RP zljwL%Yi!5Gy8}?lemy1HETLneyEy-S3%S})M5#yKuJQ1v_lCaL zZI;Y}nj3pi(n_E9*GprA!b8l-ilfw)(R{sB!|rui7*xzPJi*&2`;)k^N^LcqdUXOP zcQeZL`go)JzP(+ZXe>F~(w0UTLrjKLL?u*&>8o|&ToT^6U$T35L5 zT@HI3GLe}QQ!3yyfVho3!5`+8~2~x+NDaj*PVgs9Vy)8YO~ON zB+v79Xeuuc^~WKP?t^`tJ|2m9gnr{EL9OyyD0s6PmD)C=l*0ymtzUyfllFLkGa$q-=b>{wn?an8Eqo zKG+?;hE6|{z(;}}wATs3?-_g!bn{3ol^KBN3_d~FGeq0?t=RNXk%nP9ZJGX*K3Zzg z{_|&eeeD_zGMUUK{wiRde2+0`vplW!>cNMjmr?C|(!N}8HNSm<7YezzA=^wJh4J$57$}2cw*B_Fy zT_~)#^u~3!RjI0G6-|%bNX{=WQ+%QZ!)>Q=hG+rSKkTdS+ZCp+76;mdYS&p z=djqY1>fksLACmQRJS;j-gCah-1EP=AABHYZ?1r#HHXkbp}PN?{;&Symd$ILUl-Aj zASuG87{+3Epus5%&RpR-0lBLv@w*E(FO6mgp8bFu$|BU4)`l;+4|QRz1Ks`1Ifq80 z@M%pwujdM}XH_*OEAA)H!6G6%{#1eMq;YsXtO4#HBySy;u`Oz^kkzERkR;K z4GV8*xh=r`YmVVJ$$z+{q?A^1*299cr^vY5pPrBU!nU8XVMg;8VDsuhIJ7SaFI$bo zNo~{MYkM*K>-X1WjafQP$(SrG;)Vg~*eAJu2DqJ&FIsJrn9-W>A?A6(#`EK{aaRoQCJV!MPsjYgFIM`_67HS#iXFY_z;YJE37ihf!trPGAx(wB>BSM?&|rtn za+0(|c`zE#Ny?jcjhwuqU?%4oOp192Uw=ijv3^!?wX_D(->;_&Q%6&-hChZTpF*Q? zQ|Mlmtk5~R4cup+0AKCrtjKC2Zj3yH?-L#S&rjCqu~<515-s`sg(dXOWiMuy!G+=J z_++3Neft-J+CwU-{7(`MOie&#$2n}g-Ej6VLmzL}8_~n4w~#+%gy|Ijdk~=5g;x$Y z;cl13`04Fw9FWn+)HMxh(34`i*x*TMCjuHo z*fWupoe)~3L3L`5xWxA-Xj`UX>D~u0U~C+g$VTDo?XO||l;ITA6-|>Y9r4QQcQ|Qv z0^UmCY~BqbLa#UXNl_{RG%QTO-}@RE9?He$A-_=~PlAR_FQMdGaq_2R*7W5l)D0L7 zZ0uKPN*sZ$i;Xc~s++ZTxZ(^;t^V&hZSb~7Hf&c9CR0;Oy1P=5YxK0)=0D$=VO0|B?-iqS@xC;tz=#~eZ=+lI zR~+bh9(N6sMx8hBQ8_ac&BD&ITP;Pf;x+d&DHGwd;mS1lO9YxO3fDn&%P z!ztYw$SGZhC~abbuk zd#|Pj{!1B*TV#Nk6M~O6ac#c4JVh7{p>jVzoLKk}Y?nX4&#RW<#u^P2U)Bw^TxZ)1 z=g?!+Z&vG<&iMmjm@--n=O4U>H@e2qz@P4Tu-6qI3Y1fL`iv=ZUJBG8jUu47Pot?}z`4!9a@`t^Gv-ops1R6KaLFs)B(AD7v zE53!m*rTVRZoycvwf_yT__leb&PLdDx&R{GtN1heGrmhzr!dVLlJR*+o@+xXW5hKQ ztj!~vdGR#P;ua|Hj>Fu*xtOhe80k=}{P;2~vHd@=%!!XG$z_8fYtY)sQ848**XchF*Z5hlI#ptatKOnlLH7WCl` zIA4DbSKmH^k}WGByEu`Z$kT(A=l+}(#{1Y#E$ry5>mY6-OAW8MZ`(MYhbgL1$0{qX zk<`Yo_2&31-VEyBmb2imO@g({jRl?caS&nGgOy?D@HA&>4|Yg}uMh8n9PiL%7d2su zaRFAx@qR65yT7-ZKrf1UJw9a#=cL47_R-;_F1!yzq&-3H`V+jMbQ7B^(s1JyacbH5 z5gZ&`ApezY|2_PC+DW!-?LTllHHfP2ZKl4*v&o~ZMG&1Ksj%qGTfwZXi>MvJxrbTd zD1F2UCA4b5PAna^zSt*-nJ$92JQK0c-WZlO*n#idYzSFD4U+_FC?c%_SxN5f%dt}s zCHh~+^s3Sd=Kk^s`&N7mj~w_0JJ#HXmvt3Dh7DkKFdJ3szfr_UGhwi%vT#ygFllM> z{>ZS|wB)%53eTzF#nuq~dfAb#{J24peG#O7U4$BrYm$MQ4!yIw!F_=zU{vV~@H@UA zw*HU-F{c2yRB##9t@UWaX1-2bN^m~U3CwI#v367vtQKo!>rJF^Rmc`La8n-Z^m_>t z$u7wH`3xje?z6Zm094S!YX)6{qjh#Hp)>(5YjMV;W)ali+0`GH|Cj&Vu@KtD3;M2~rTK z`v9UtyCG&~E9`WPr9-p5IIm~|S-jE{{B6l%ZEIxUjJ%pb9YhTk z?sQZug^oroq*gO^nm$>B_P)GaBI~`beUa`yJyyb^<^F0ZoeO#e^xQy0pp?R zC+}gN=@m#j3Rq9$8+N~I0lBS`rjL62FoM>yLM3tBAo(8rZm7}-qwDA{S&42V`QBII zBR)OJIRt#0aqa6Ix~)Bt>@_|@5e?$?$$Pj)U7kLc?WYzw-YZ!(A7iBgaeR?Kov&2q zj9n2l@oj@iDQ;Nu#t%Jr52xcO#OS9RQ5eU423L#VJ}=H33Yv*uZ+0+O?nUCP6h-6b zx8rDq2E08qpPV`zu_;@hUT>L1fi{M4=VrU$%hqW)Eq0V}sf)gFNSUaxHa3s;#B=?a zoE3@XX5;*wlWF48k0|RmkUahw(%YA=)N03Y_mvUo_H_h$9+eZGYs#P-E6$VTtP-|m zc@K1kL|}OA3d-wFA&tF;GVt~z=Wtsm1qg|`upXnK>t>K)Io?#?b+=Jn5@u2lM1^rFc$$R^J43`Q9 z%~iWmNhFVHZkdB;KRm_pcfW8Jn+kmh^dPJHU|POBf#RH9X|VR=UaZeCV1> zIZXjon+~YDB!lxh8|hhC8d<*CN1AuqnO$cE8<24WyqcA0!`KN_y5}QK<+)6-_zRp{ z7L8{@4xq5x1(iH`pZsb9+WFODZM8ez)wzy6%kSW&Pa9DGUJE?`xsT+vO8>O6!p1 zg&#P7xdS>awV~`756XQM&Ur^ZSihzQogW-TOTTS2P|S@6PwK@}dn?JzH=8mfw!*9* zQrPOe7M|?Wqn30Hvi;DF87Jq`)9P~E#+fZO7ss+Ej(IFe-;%93wH76$29rYfD(YT5 z8LTguW3xsEe!RH^?5lsXzh@3ICC#?~A5U}f8U>{fqa_nE(Wm?f3e^j6`|}2*Zvmug z){76Lx?msAqyC^c&X_O@v=VD^^m?$WEJw_&|w)Qh!Y&1 zaBHQjU=!B#Gx$X-`AndRFXpOT#zfr!cqy|5>&7&|3a<4!!0!dw;{)i##x(ph(gv%V z3P8!|Dh}k=;-%$OJY2PzUis1u zXkgu>6Oi^Rg!b^+(g4-p7-gVJg|9}Ezk4p{!PMl)Ye;o{h)lHuo{e|XdlDOG4ilurzCXe%jg}<%@(r$5c+LbMi z^;bpVe4UiAQMQt9+qu%fB4;dGs11X4#)7D8DAny;LqCRUQ~v7~oRskxkL>Kkn;kK< zd~+E7$k2lLg9otm?;h~qAOy#lFYH@(Co~PZ=Tfa$p#JjD3$^w%c%){y7xA`$Z5~y9Aq=3T(T77j9~cVf9eXce)+SeP7h* zsaq2o+q}bV;)}31S`;3d$gZum8oLvLM2S$@MV# zg*JZbTtmXMk5RMdH0QG%;d{n8{qHw$n}Owre}d_eczoLzi~F_)q0SgJjM*@hh8o|8 zI9^{cx!s0$=V#$*15L=EhZJ*->vDU!kAzkUev~D2X^$5eUYXzjd6!oEu!GU#(f5@U zhH?h(0ySNsyGJkK%HgE5t-Sv=|M`A7swe61xv6xk_#}?j^FX!elc3v@0HacL@!Wk0 zd_Mdy+hrai%++is`PZDWlYX1&1pe57NE5j3jv9X<+##n&oCqpEsLu zhKvvX$&JAEDIIuzs{*Y~cma0<8*rD|UE06;3D1$DX<t4JTDzg3Z4FR_J=OZ>{h|wmqclzjys>Nn$CMy z^QcnSlq~ZDs4ZIB@5>Vm+clgXwATyHb9UR)>tpdk@^?1AG8-rD@uEZRiTGN~fFjQ&Q|sc} z0-wWQ;B1UPj(aha{MWdW%IdQ?RjL^lCOMH!O9Ab_G8EsW@4{9KB{J>wC5uVLcqo4c z`zvouZ*6-(Me7OtF8OXU%z7yZs$6L2d>`EQF#$D3Pp9}JPBbCFgzhXBrPr;;VUUG5 zcoa%t>8f|s?eP}Z{2arA=Nw|sPxwPk^$A!BtuQvBkc|JS2>pxCq0Y2uQg2yBo21^L z@jfR~c&kcc6^6)4|3}!DYy^=)nj1i>deSQTm#kNPkyO#neTo(DIokE^ZD6+ucb}E@g;gGk4PU zS=Kmh!V&bZ{E0;-(e(KBNXoFUgd-=O;W3XhRCRd=RqY!f^7PbFe0fiUHnv{k44D|( zetiL+F;XXgaT8kqdJx$&dDOfU3|3Q43w#VZaq-cERM-{6HA?63jgbn)FC9lG=I()m zeY#}gI+GN*H*T8Pbed@u4uX__I3Q9QJ2nfz&<-Fx(-<2h{;&U+;@-PZbF2~PC*_du z?<{P+u0@BAeL?l+LzMPejx!iyaa=$XtWcRvn`(B^>RD2J=8&^xhqz*JkufFrEh3v> zdAcKJH=H%tPboum2gU`%r>&twJ##-w5`_O~4`hg`{dxNDKTW;e5nLfyB)Xkn+#Ze-*Gl#aXjsborH@TmEq@u&$u8$f!g$WA9~v*bmygvDbXSL ztT+b`l?CEl!4NpPCk-Vo)_~F0U{*XL2Om$LNUgzZsHH)PevbZtCBOEf<#6shsc{Og zb3fOfRK8zs9l-gxva~wcl6)3jL+O}Wp0|ufnQNPHnrbWhN)_XjFwQ_(E=>ih8d%=> z6-NyCgn@@Ta7wlfO$~U1b%(R@!r?ls7&Y}6@m&p;Mwhb>V@ILat=$+9aTtRyM}cefA^dcd z*J0&#sN~;XDy%@#6%nUdJvn%P?lAm0tq^*IlBg@Df_wINlh@|~ z-gtr>{4{B8yGDO(+;?9Y-&blPR@5Ra*n?}vE09F*LHZ#Z&V427;hDl^w3}+qy}Z&e zO5+ui4v3>odcF9dN)Er2%agg&YIKyJgg-!<6yyHuleJ>rHS}MvNHOiIc=9&aeEqS( zH)%_;FxrnofB(VtOLj78+aOqBaEg_cTo+`CZNWRHs`Pm`?+rAz!yuO(XtcTqN$Du& zoG-^KVyDqjA{9F_4Cv3-Iy{x~2Zdep@k@aWxQ{TMz8 zcB8Vqn`ogNjTU^C-T280+_~of{RV1FgFaI)+@920v5 zH+!bQ`q>kp>aYxFRd2!eXS^m=at%8_tidlYw8>n~i!RCzqIH=P_zRCR_wD0oRfY{6 z-M14@JX1!o1Ll}^{viITQzNfK&b$YmijQ_^(>9C$di<;xoh_Q)U5B4O2GZM-9(2mL z?Z1w*Muni#AHcFx)BFDqy|ftfg3Ivr#^)GSlZmU&D9{1z2Khz+7Si_TL8lwrKJDw^+n3Qos- zXmr&nw2C)@!1TME@57(7u0oR`Z*j*uIg%9B#{E5=AoA4!r)(UK-yA(r z!+a%-QJRbIrcIzz8Limkdk*hjF++K4Is7H(!S{;-ym+h*4F#(7?T$Ft&tIaYU2?SG z%v#6{m%|gbVKBDdmEC{*2R=r~qn}YGS|@s<^l)W}nqVmKNG}oOjWZZUiw_Fc#Z{A{ zq8^1@_rTI`yq0{(927@vhZfG{ddB@Gcg-n>Bfl1r=a_Ih5PTT@lSY!*>DSbXL>`?8rH1Ol!%c2h+L7q?#d}D0_r24(Q=V z<3x5|;7)hdh7((=k6x=5k=8;nTA^?Pll|1lz>i__K|VKf`iUT{`l!I<%2Gj|&7l5k zyJGBqSRl`}o6B|ZV)#;KQIf-^FX><*AtzbO$~AQ2r$5b_E=dn}s?k{k6)K!-LefDM zc<<#n;QUNx^&=c#&KN~kR#?(;(-ZW!!HABUj-^xelEUdOC@jvF5o%1!ag7=vdD9gfljyIT=d+XD}&`j{}u7ulH!*O!jGaPO? z3sn}KXFsJ(V2t-!aB7)|t6s`c)VK?@wss$_YgM2S`&;;Z#wpy|5Dzz{>S62tZfrj~ z0R^@n;LxyAc$eJ_NufDt@B0Ifrp4m1R3l8Q-j6-Ud#Fq5Bsn*%puRKF6uU=*q~%87 zka3pu=WP*2Rn-2|Jx%|JDZYZcz>Q&GwT{PANCSq+w@OGJSSQ!pBzML2h{y z98t-^T%&hb-FOO%28^VzSMx|PT%95^pF&T^X;4(M$B&Ue2+w;`s&^Sm9xjEx1u1C9 z>vDB#&a%j9@-SnHFRLh4rRMpC=rL4)#XiUIo))hu%ze)?9=L$_;!rr&K)4Xj(}xM) zFsNbYGIC7lRkmQqCPr`6-3#D>l?GM?T&r-khZQQ@vXyF+AaU6$oc}?h|2wUoIZvoJ zrKvxr-n9NM__gYmX}~a^&uiR*1i?gjpzMk6!`9HPbV;F(_jX#|7ma>;e;`I!fLBX7 z2XCl1;-hFmij^l^Ns_?kk+FDv)qYZUR1ng?0Gc7Uikkf+gsT+ugeTXGM(ZANe4#uH zzP;6^t#5Xa{3{jox-bW?w5QY7nNIz)Rq1~^*t&I}aC-hRh;kf5Hs%3z*~=0y81$lj zL>AoO^;KWj516(o1kXrw&zGbu&WfE1+oG)SY3wPecOA?YO9esWyGf?v1#8*xO z?j$PR4#1l_7vTQchXTV1)y(#BI80m@ic7urIENduLQyFO*6p_M| zlQ3`6SL_}7U)^z0VLGnlOpL%MOCkT(2-p(69X+pIf+dgC==Z;M^!)uY!5pD3jD2cK z-|KTIVD2zsdSE@}UQCm_>9G$l$DbqB*wX^=i)4xE;-qu3fQEEy(;0``?Ad@69FbHD zi~jcXUz`7|u{kTAu!s6f@r>PQvaGG9k~?V>T`v?Kd|6F}qIXczWiHzMJ%e55-ok?N z7o=rj1_^$vU}32X_Uu!quBdf5;!zK+liVn*Ip~ii_eRn#bsH2ojK|g7hi-xQZ<07@ zi;{--!1Q8eYOe+Qv$ z{eN}UX>&Db{>pUNVUa6vOK_xbqSfg1P@5GtY@yUdDPiRqO^W7>oz~mqgr>b}LW98Z znEG%AOuKYku(X?dvzs5sE0^6NT+gz$|=JJ0_Q z;6`t8tZL-`r=1eiJH?#}&)uS~4@0s1CD43&h?}K{ppz}1nLY6cdo%Tf^UAvMW&8!y z3$w@WDXA2?VKx=s3v7zxktn_VJG^kL1pN*_eEpR7F)u8D_gz2G_J}6+ zy)_fwm@`NiG1Lc3Ue711D0TYz?JAq3V}qi4`GPAE|Mi~vU;Ozx#td~+MbMCIJpT4} zuwc$!*WK2SZ{0S?)O{W&;^-lCbWIxNyG6HTrY^Gi{IN`iVhzan`j)42+ir#nUgqFmx^A z#4mVS&xBH9!-(aX(KTOndQe&k8fneIvS2aoPoVD>af~i0u&T7cJ9?TRp~ZPWra_ow5`yfKESd~(gvE=df_q{Dw9X1e>j%g2M$LA-b&oSc+-}AK>lPT4)|Q8VAR|gwQ*2?2@H9t$MnX_7CXlk28N= z{Ghc>1}|E6(xc>dT7?@)DQqwun(?@Q{cPlntx~v)N1lXZ(wZBB0JSrMU0Sx#cwT_p zZ-n)K#`J*S5a|^P&%fH>W6yfF(@GOA7M@`4-N(?Hvw#k_`M~CP2^e^F9X5Z8<9+fg z7@4{cp8n=BBm&v)?0SLX$0$gjXhhmOzeC=>9D$O3K7OkSKrN@39Lw6 z7u>#Rvt+JQ5x!o6>z=o;s2y5(dfXZuTOETZd_{14mOcFGGDX90qj1^F7_fcU%AX4W z@6H+sl1>~$6&=o=o>+vU!a161Z+neFT_&YeNy%(1TSmRj9Tkz|uGLBJGU=EUwf`iZ1K%jfE z|F;;|*uoW^B=%=gCD4Xd_+elvo1Qrp4kc{E<8OMH`7$9&aF&K_k^r}Gok7tFz}gpk z;hT{^v@HGs4lYv!kH+;1W@}u5=zm;?GgK6vmwV#TvN>$|-xBuojsn*eJ~e$fSqy>| zcj4TTDePJP7@SoTip!)k;rZT0U^es*B!|rd?V<64G;{Xq$q5X65)&=^G45 z4~6^EWpFif80Ieg#*R10f~uS;D6U+J*))v2&9zzTqSpeCb&g=j{R)p(iGuoub8Il@ zJ>Q)$g32>>D78$MY&p;W=Ex1SeMdITcors5(p|z1H6F%5#at9SF@j=h&$2shM{wa} z2})|6%VxeoD6Pta0A7nd&4S5sIK!DF}GpwZ06b7$*xcPh9-i;Fez^* zjL`^2$G)lTS7{w&8I_@?y*orMRKR17_IPdgX(?BSPl78}^H4@hpM44X$GX=1g0?yv zmO4O`68GGsPx43c;O_!xzAK5>FLpr7f-U%UT`oHCd1s4^NBHigA(iac#){$*ba2-u z>e}mys4)w+Unzw|i`_VM(_;EH-I02<7V&=f3VO}NsqvgFoSU7AH|kTcB4jd)d!vWz zP>f=i$wBpyLWn(k35V^oMqP_lsKIrszq7A1^T2EH$)ysyf*&!DcsnQ!T8n$p5l;-g zhkX}hXvNu6@b_U9mPrq$&hc(c#qkU?DXRd(c&^8lE5ZAMp(N}Q6~1z8pl@fdQ~kVP zrvCUgJ6!e-=8HywgTz*>%gn{%i$9<$s|q>Xmljz@Fso<3nEJT+m{cbM=H~NQ-Q@_j zm~pn8$6z|@aDbGH8enNc5Jv2Kix#K8px1RhlCM%GNtNL+Yo#YD-`|Txd}hr);UDz8 z4#bF$a%|276IMKO2>RCyL-~$@l(uyXeGT#f4Tt?$*cO9T$IqkH7)9K>&8$EE*ne;z z_MiAO;<_jXjT(doFHXVakGJ6NrzvdH&IIX!AxC%AWf2V1c09bDcf zimziG(a8D-&eQ(|OC*xew^)s|bJW0XwFOoVuEW-5b@;=5fsPmW1Kvx8M-wtI;@kxC zx-|ppy!3H&Fv@v_X1%9==1|{QCah=vgs+%;Mv)ViG$l#wi`Ra1qQk=pn zm7l^5Lklz;!Dne-`{33%b4+NRfO46qAV#f(=vg_vT>Jt|wPavo;}%x@#RN6xp2783 z^3hAN4!0L{V8y0N@@kqyHZ>p7Z2edgk-Ulxnj7iBU{PwhQ!9}5aKQ9rV=6fL4nt1; zLFfDs(%~9XH#b$7dSf(A`RGrIqL11A@iBPU|15^Yt5CM(K02xEi|*IdVRg1Ct`HND zhh`}oFx3O^eF5@wIR{ZiqfugBBCq`%!(G4Z=txB@&Yo=sg=cK>m>A-w#SG;q9LAeB zt6@l|2OgG+hQK8PHndF(>WY$Cedm99V9Sj;W7>EeW$GH>dgp^^^5isZO7LSJa!Wyk zCE%YE;ux^!FT#vLRJXqun+NI8_%tb+Iy;*wyAh^3jqkti&W|%-@TUw6zdIX#mM`Oe z&Bk=WEDLtelcLqvPLS=tYyB~E$H)0-|6?j1j}k+p2_yR78<&0vVdw&y7NrBPtaQLf zJP}{jogtZ8FBB{zyRZ%4IlWqz(Nzev0z` zt^u)=1!%Ko0$wi)hFRmeZ`F}lQq_(`CwFxkt=NQf!%JAxt2S)3`h+hQ7qPgfq7bFs z3~ob1Xo_9E~g`d+M}FL*%*HFOjR#!pYt9urS#jr+$gNi|Ojf^AY#ou9|{$p+7Cx zSS$MSTH#)wG&GN$Or8dBDEFllT4pii{n<{KXPhF;@zKV2ivhaYtqo=T7xtRjk~FOd z%J&(GxO)b4Ho%_lOfL{t|C)(3(dL%&eW;-3vRlZSCI~rO-wKxVMb3q={Z!{A2j?zD z^f>1Z6;&=3>f1aZpQ7zl9-}XOk6$m?XY3a=4=H2GsrD2w=@Qw$YfC*d94X-ZAu8*n zE%uEH>hp?_9z7I+`xfKS`sTEueNbJvlK2K*zK?^sI%9Zc!s z**IDvr$qb1F9~*|tA!z-dpjw--NbDbj5t4tHLn3 zY;+v+W=&t=i|A{}9I1+q zmMN6CN?rKzESiS(_Jqp(&8Y2o_pcAe+WoV@b}ixMV}`oUF_b!SD($`68bRTs=-STJ z^j2XQnToz+ z`!FX`kgmiW$98xyyk6*hGMcig-%#Q_2dq8sOr1T4(VIO6!sIJX7$12TH`ivO#y=XD z6<5*|5xeEkmuxYPb_$*k-Am^^CHVP80eU;#F|WfG@Rb>uQ`!@?W8xw2#;{5of)m3g z(4ujxgguci^lqNWVbLR>uIt{Tri^H~bjw9h=?USEK^OWuJ)K6B3Y5EYA4yI`;`zX? zLX_Ejif;6Q;t35J61zlD>60O}i<=@G-4IB}~j6o`4i?jz_UD^eeS zdu+{e!Xk~;h!g#?iLXU`%<^|+d!Q@1*L)+NHFwE)L?Co$HB)e$9qG(|DQvkIMDFjN z(S5HULg6oEDi!^v8)8IE`{R!&Dt9x+`Xyj`!$Q$+vLC%7RtsvzZnVMokz{wmL5$n_ zi*|WFpc5jmT;EsAXq04%?-$; z$_1Kd2f^e{0=?*`4^0CHYzpS^gvEP`sW;ED((>W7HSB0w8WVCnl@y+%TCz$>!Z`c4jQl% z{S>agbw+$@m2mlaSmbRK6MzZERz`h_$*ju`ckl6TVY?KE46>EPbW5u zcvhm%-n@@B%^qJN#Qb9>JR4RaaqQDsQr~Rsllx8VNcAvX$9Ndn{*E)!cRRpG7Ph&;%}{~e$^+5WkR$hZC8}= z@pG=Q<88GhSnnY+w>=hph^2U4Wi4oWO%)0~jA_aC>(r_E60KeKky2Nt3Q4c!=u6c% z!ACxeUI^CYDpxKvycOr`+qRUN)FeE&6LEPCi|fXEedHQ+cFGjw$CW)(vA4 zquHrS|C{W)yvyZI+3OPo9pQp-ukAjntdh`nF~(&5fha63*iWzJPtbg|*Fw{(|B0m= z{zVH5{q9mjcQGbmt^=J45OI>lxfc4aH@&ziV!)rAMBP~INp6WQCI`g|4Y9+88-Z?;%I3?oux&GauF0fUA4do;(timyKh4p$ z(?ua};u9f9vzO2?ajxLB^}NJy{~C!$sXc0na
H3wQpOFk-;2z&OoCXET9Lezr; z!b*eB!ix+I?7S>;C@Z}tZI5_b9H0d+EfdT&*@3Ko4#Q;Wc-&iRhkbVd`l5!*cb?GE zf-IU_QYpFL{kqWmTPe+(eU)z58q(#OmT(RcG2nEb3jrfdDfE;p`L3NwLk36Di#e4- zU3RXJ98oIF5ChZeC+pybL$`hh4q;ncdHOj0tkD17 zbK&&PB;kjJHbv>J5Zd^^rSCdsP$^2JC*IRhGAkEqgV(~bFi{fHlptuMf;M(sK`%=- z($h8<>F5S!^4+nDPHKyL`gezfO-0VssVR+`gAK?%T#*`9=u(MKl4zTaqa4~tsr&q( zJKP1Yd!D1yzrP8ym*vnXHiFBvH;x$sih z^yV39RKBEJ$6I6M%^*Q-o-Qq_tES2h?Z{oRl5|#nqpN!zL|pW*LP?PrTk%$xqUwJM zJ5K14UfBv^eXN3r^LSRo0|qThTuu?*X4pPpgHU(Sf>u{vr@Bj>XjbDsYVKH0Q>@gW z|7I)>B(Fe5x)q(d)1Ka^-j)=FX$Y4dEO9DJxhH8FuZPO3*_dbc3Y*KuBPA*trF(r5 z;Z!Txylc78zi)4GUr7l26%3P#=rfq>S5aseOP;$P`lP$$i?;vso#1j+`Q#Y7n7m)Pey*W&Atp z_jwRJy03g7Toqg0U!-H_fiCE(H7K&W^e7O}#Tkbh(|EJlt) z+ZW1+)!z-&?Q7Gv;641kTgDK5WH#!jsQJkI_=iQX5<=KLONmsu>Vt~)2p z7j2TC+FcQJGEWPNwv9s9!O6J(;114-JQ^39y=YzU@nn`-N;~?qm^aI0p&d`x`^S-! zYcy>&ctP4Wagz8I<-$<+v$XtqJ5)-Gsl(;Bbjk5LZS}08CWEsyertE3`=+;&E%TCP zYy9uEXTJ5z7J`AcP}}9S5WjFES&92Z>6Cc7sVZ_=rHA8Q5n%Pj4a+8769PQu)39oJ zEV7wKn@@+*@vyCw8Pu6pe`_J+Xl+)s?(87)AgI&DuKh{pdpWAVl8+= zBm>(!Q2#!5|EDti|8;*jNS*$_+<&Y2C)pl4qobPaS@!<(pIGkD_J8s-fA`b1>dmym z#DkKzv?jy&pIB-W%*$UG{0IBx#;XC;_`c)c{dlPLcdRiFhWQ!s46C%1s!WaO(}LaT zYG5V1Zt0c# z)EskI*J4b19vqu)lFj3h+P}-rsGmw57uI37umSGQ zm$2%L1^H)NBI?tCzp}>1F6L)Hr93fY1V?ph1G&Y6sLJ8K@cy+txf{Hs19J}0fxUmY zhF`bZ|M$Az^)Co^=g{x1SehM!*5RK?$eT$yst4)ufpw%Ko)g<@kHskGj#M#44LT!U z(A5%E+zCjbf_K|Um3mX=K3xo(@tp3A7SC3N1F%Kh+pkEmL9+2wO1^%EJR-W#rD2xZ zYn*Hl{BAOJsqHJ%gYFw6sD8ac-CWd3cc(c8i+n-9<>_$P2lIbqU`@LPxY%0`{vsaa z@9#UhsnLPgBIciWwCr<2yD5M5X)z)4Z<{jaoRz(=YFR|ZHv>id?EqTsI#l-EDOYz& zC>c(j61-%e$2O0rbK&0PFXlOQj@dzJBA?{SOD%C=u!x5fb_m|>9dXh}_+$jN8jS}WY>*8bBaDrttO3{ zowCn|mwcjj>(vox_Kfm-6;cOBaW60GXoGuqgso{JQ{TPxc6TPl^(mt%FXJfpMJ<`> zdm}c=4&DPLC{!9FlQ%SSA{}+?On1b1gkfj4)2=XUI<-c`|Lo>TQ`Qe8gYsr!N7yn- z%AZY}3*O4sydifN$nMos`xp83GQj>FPFOnEiY}i}r5~^QQNp2zbhY^%o0O|TtIauWy9AY2T4J=6-iZ0X2X&KW_dRJDC#%E9tSwfg zhEtClBG#2?->WTlLtu;pVk3?qxNizZk4r&FjTc^CdqzFhs=%19P?h|4I`H5Q4S9E$ z6r5V4!<=KZVr(uJ?!1kVY&lj{@ZqQHo%mvs8G8hZTohkUP+9a2>R-|f^~*zf`!r*j z-VWO&lbh~)>fPCkwU2JVlMeNGJ!d`a^W88`e>Ak^)`))gf9UYe_q0mPL#}#p3K=hN z;nlzpWPaL#SDh-*e}0Ip9aF>e@a18596RlTByAr={diA$CfzVHPYvd^!T7G!7X1`v zVSj}mej4_G_Z3g{3|)-jB6i68XT$Nuq?+#e?xCTdIzZv9iA=tE#K>;s(vsv%fpF%4R}0a3~M0|18R4} zZs$krjM0M2%syBlZ-&7?rL*K$&99zoTX?Ejwuovqupy{E~-0xiJXZ)tAuU=;6E!>rNun(dE`!_{B;G&-tT z*3T;={*0C1^|;6-;{MV^l-@cGkIv6&=jAARTvtZfiz7h638?5N2by;epFW?*tC;OD zq9Qu&`;iV*+$H6$$4Nib5J#dsptD7S#&mJb^q0d`4>4(AU=Sa?9>WuQJftlOm+0-7 zMk@V!j2Z_tk@D&9bZ63TEYn_w^cnG3Ib=76*DLamt$)@cBO^;T?fw{JwzrbC#W-pT zrZ;`We6{DO3>?LY&!T9b(3A6U}S;S z4O6gg$#~2RjU|^gt;O7 zURY4!g_)TXVK^j?Hg4EU`J1*;>*MA0HqDN{>t%>~ZiN;~3t@607M>+1C^YgInFq#F zb@?7l4Uy;5;~dzq{wSF}>W1l8P0=qj9}7)ap-+$P@GO#7_z)6~hK6bQ)>9FReflG7 zb0m^F?IDw?Tj*l)dfGQ$3BmoZP}iBtDB4|2%42>}oo`3c6aF3d#^1w^paGmXv$w3R zk3|eSkI_C9VpmL-X6tCk2PY`>Z;!rxTI1t2V8To%JZ?7tWoI?EomG#>Nrrgn z+y-{LrbG8|_+OpRbrJpYckjsb`CHf9dk52<=v#E(B^_@ISE2Spv#jhTF=pkX(%!%2 z8h(fbQA;GS(yXIo*Y=n%{Y#3e<`e zuzJ`VQhzs>&3szQ)`<0ZhSB%45WHy^^6$&Tbi8PDJ-wf*7SG^+JUVk`hx@oaX&8Hb zb>KJ8F5>IgiQNCbCHp+lW@qOc_++t2c8}4wH1WA-C3SwMi2PPdpx<)kR@n}>`2?RRMtytFl- zEc)Er)gPyyDWbnUX)C7HO+jt78@Jpzg7=1Y=fy+8w?iJtt{rCGfE%TourP1}cBD0s zR`5?*f7A>}rp()EG}B~0ov)2T_XqQ^Bf1TqeB6Vi!Bw~seF_Q9EAVW?FM3lr6nU>E zVcD7p4D-B8(@#gid|xk&7g202CD&o+rNOg(%dm4mFdo;p!;NbSsGJI%II93fi=Ma~ z?SsOoZcrLhMM*Y$VC}XPUX@~e=W{X6r+7R0xyO@MtT9(*wc_E=HQ9cEIjaoT;I&cL z(Z*JS+(0FKzWb3(Ld_9Z(MVpI28eRJhozqKe9`?N3jHs#{M*$Wa5-1je{nhncs=Dd zbvpKdG~+5L?ch;zUREn&?4J{P;;fLp?FwG{Zp7M-Nidj~%A0#tv1UXbvuzTL&u#DG zcv%h$o91)1hB;SWUxkqkMf9;m~xc`U&QShr|5gT5465uBk9f= z2Io&67(3At@mepib954_#?F9u+7x8g_goyeN&#k-Z9tZ9Xx!p| zpxJ91I=mW%xcC!D8`VwB$??HLrN>lPl1CdhWs7_=<+!#(j7NSw+4$oO337I*!_iX? zHsNQ<+50P8m=XooI{kcdPL(b2C~k-i(kjN6_#=htGKYz%9c$Xx$|a zdnbx({fB{=?<~N{rZc$tIps~2hxN8C)cEo$#;I39c-@9G?)PNVnpZqjFODfwwAlpI zi8)nWv3!LdcV1}A8zX)o{iOqT-WUlvl`dG`Vl_J!Y~pwA4sv?!EqMLd1SRWISj;uX zh=8uh8E$|NZA{s?s|}z1G)FcsS`F5e|2=+c)%Ced-Jhd3CGahs9E>r!gpVd_Y_xF@ zeCIkLHrxdlhb)Hnjss}xABoE`X;A2sf&-h25jLn5r`>JC51XICyt^VQETm}47V$II z_vK_uTXr5lku^UL<{@{6^M*IWIjhzb*=&x@a{{qQw0rfcNXG}|QP|L`5;_UB@OktT z7oMu}*iBOIer^Jv(maB#En9G7`*rY7bP#zn;^B7I0q5@hLsxyP=%U!Jq=kJjdZiok zLZYBicptrNw<5!^1&`~WA{!eHM@B(6ITw*LvoI^696mSNa1X8ieCe({>+cjfST4WB z`$=!nfB$mSc*h}EV>zbgOvj#8eW3Jdn#jpI27Ss6;cM)N(N4{1IM|I-JG-#LL~-vR zj)#f%9k8q8IheiA!nM>!95mX5hwbxa_ulf)60~g*jc)q!I9MN#1-ra)eCHMslVU0+ zpD2W>ZVVK~XZG`^0FF-&;#8Z-_!2q+7D~&}QB4!4a=y{QiHg8vzrXYS*_OVt{<|9G z$k|?fr7aUjvDFR}X{1vS&PEkr+&{M=h;>>f&tb!qbv(21HN}hZCPnq?XtJ)ra;=@v z*pY+Pl@I9Eh<=do?TT5yR^wdjC~UkIj~-nI;XvC(Xi6E4v_KL6=ja!IiPom*% zw+l5h^O5o}6Ypv_;n%x3h{12@6}J!?E1F1g%v+kfUlj@y?9gfWPz=gzrY}QnVK=uo zyt@T}r|WaN5d|f}d8N}h}j#-$~E(twE zJi4Ss=~!GK<*+ICT-h*`J4lnTwR8pguE<4qV-0R>u@9%e}z!L$?4QN9ZIO+&HnrVnl&UWj`UQ`s(fy2yL%%hzA_z-^;Hwv~v$^*kylk|o85 z`RMkeSbI*1C!`5n4}07#(S`cYDCoravE}>W-1*5A{&3!y12&Ffxta(Lsan9Uhs@Zz zy)PSWsYg!1PxJ{kW>v2)xI!~={->vG%;a@0MWnU@cQq2Zl|?M(@+)U>vR4&suBh|l z4fEJikx+1bC!y^q;as(YRM^A z#XfUz<+=E1zt6B1$)kIf?df!VyuOWY&=#6bceR>24twK;Oe;L zvhrawZlPTl`yc*Ir(*S~wiL zgEl};%&~vCuNrMtHsa?BEv~O^&n9_)&iEE>EYU_`1y04z#Twg%GTF!Kb>^<2`n)zU zjJvfp;=9ozN4Z-F{}M4EYo=X6t0NtG>q<4=wfQ2Ry;Wncj$Qf7NO#Vu*v$Rvdtg~V zJ#ik5hA`n24%gklkPWTaYOy$`D%=sKAoj`RbX=~w1a$f$yC2SmV0ZpHY>1x+jGGRJ zu+i|joy%v=MDp0<*Le52NGx;+m6iS7XQg(_FiPhd&bZY>CDD{yOjPEC;&IU5F6Q7I zSq|r^7qIM}c#drK=M4M%nu)D5(bo7T%A}w0=DQM)RJ{iK9vb{oN0UpJO7W<`N>+cW z;w5+=T?bbYBiPoy6#f>ckQcrMrEcmdUu=dePQKV>sK)$Ln`gi8!I5ITPsfQxnBMLz zQm*eqm0u@^SX=l#04*D!VKo7j9JKR?F!nZ&TOqL&qy`I51 z8mA8xV^`dMl8BnIM*PUenj1f6@vw3Bd}2-)zV&lH=XN!~x_t{$y^tACu z$om9X7fglz;!Cj3d5M-wABcGH;@L%P_wWBJT_aI>z7>azdnTUmhw+i0Q(*h*9KKfn zz+eS)3>rHG8;3^1r>#6c9;?b5$2)L9q=2p20kS&UU75^Y8QN^};xqOX7vQFEk?i|X z9$j&BrpTLAvIJiJ#auTr2KCxT2Uy3cvGuZ!{6T*h8&`T^`;6tVTCf&b-x@Jsyb51C z_!gFx|L}0zPSV>tdQwlX2rdf=-~!JWUf%uz=}Fd5_R$9FH|!=<><@`~RU#H2SE}DK zTlylfNNQ@nj^pn4;c(sleC~vZts|b39+#V8(6=$tjV;DYKl%@p&K}ZLdhD*JlqSSW zYi9aLH&lm8y9C8ZZ_1zKpK^8VC{dBsuREfWo?SmkL!)w_VNoxhcQ?R$QqZVtMa z&ilrEGJd4rI+hS-Qv){O%>+L@8H7cEu>8$deUB#RG|rqbCl2At^P$oGum z|5Im0_y`WD&EWpui`iBE6h9vDnS1nq@t-nhci$#CwI+&lRzZp3R$jQVm=8D|=hD`0 zQmJkaX~(cIY3|lsJ{nog6Sh5NpR#PWzjmDYWeI<2KUX^A$9id8VUkR5N7`SP>1tx( z9^S`!+^W-2K4h^QiMn4fdqhh_kiz{_u#wdrgG6MIS$J%!ofUs zZc#9kTe=6ZLHf%7$a7@iU6Es>BUa5nB-7Ww?OJtawscKGxYYAmnzZ)5x>T!4S1R>) z`#UC8T>t3z-|`MRGq|1m3)S_cbwIt@K!xP{;Drk7}Z@S zL&qBjVIu^~>i=oFnF^n*r;79cle5(6k&#rtTq^xF!04~czhxcYp)a@4U%>B+X7lrN z?&!J71A{J9Kt1*vJ)e6@rh`t!CuFZ{>~E6On6LESRSC=D+;I76Gj5a_Fy6H1E?*2d zaNiJk>OY0^%Z;>k`D58N_6E^1nYyJr@SuXrlafBXS1)tIN=NBjKT>AnT{O^DOy}ou~&61z#|P<5YK;Df+)G zc&60|eTc}^Vf;*zgKY7vYpyhach*dy&(WJ@_iO2wfv~X;k@Kns>2JkYn4O+he$VVN zF77Oob5X22FVi+WD{JB^Lj^MzE2kR!b(+#PE@5 z4zHo8&s#(sK88>ig~lJnn31p_hfl4=gNRG;xhKbMKd&MC7Po1fYcVZqOr>9DW2mEP z1I^FgNK4n}ldbs&x@2*XEI!?)j$#bdwoQvj?OZj{ToY{Y=zt&1{y3D@L*&b=rW3yA zcoAxheiaEKhMX8r8*mJpC&wXaU?rAqJ_7$vQPlO^UTXSzjoOWUO4)Tv2peVw@534h z8TpbVW6iN5pzU9s{;LDy*@n0`Y&(4$8$pgccT(ffpJWG3q>6EK|I9l;{p&B$0CBzU z`F0wbcdtipF-GQ8kRgn!`iOSmp2+|Dm^Nz8#EA7P@kT$EVlNz`O|@E(=;hKb5eH=E zZXMY@4fGFCzvx#o{g0p(RF+dls$Fcc=-C@Ou`&f$%0_}7ex@_~hvR@@!QXzWv}uj9 zR?ReOx+Y$|JA?M}di?!~!1`uYw~;fb1b(lhILgRXv@w`+-ovS) z{jwddRvEzL?ih+6rT@3yiYXJR=w(kA`jw6oRe$`WL)#Q+-$_kiUcDnu z3eCb+w$1k9(%AmoHoYa=O*x6ER;K7VGM1&?x3KBN9_&-n1dltl7~S#JfBNUqLI-$i zZKHhmHRN&18uvubq|4J2Y1EAx#4jv?8Qjq%@+~{;Z>F!ITSWZwf&9Rz3D-m%h4C{j zc=6T8unNpVpPc*Tu;pXm(iOiPn~3W5ISWM=k1F%rE_81t&9>b$%;8P6ZhWy@cItavbh&n-`Z zt+_wkAGx9Rq-8v$u?2^GRN&1=|0fn^^Qt&(#t3AOw!r*&5evgq54rK*==sSlV$RVV z1f~BuWBVQ|C#9eBDYEgqFeCMn;8&h4_jjGIIpZc+ai$}Z6M z%tV?!Yz=;DyqEo!TsfF}U6v!K?^;yV`Mb!`uG{ji)hd$6T_^kU2w9nx$xdi)mY4Od zvr0NNNDEyRnrVy{XzZ7(!ux?af^uC6)URyE6!+!ml9(W3GWWqVXEXe>>z&Zm?}qID z_J#e~J+ZZDn>~S(gQ{uHt97#an;h3+wCh?F&543wT><u}>0Yd&0Y7CqkW|2y8RLuRvk z19NvrS2hgFk(J#W|5a%E+>cg|RUxlkA}<23prl?}vftX*x8|U}5lG6PBdcF)$9C%T zdMa5aPC@L=23eU7`r+c*Y0BNlCSZ5t8uSndre0w)Nh(3(ierfn%b^tyf*Wuxmo9Jb68IPA*@<53l zH$+V4676~#H}M6H5OI|XV02o_mj0jy(#y31z%&Xd<7QoW#FQD)NKLUhG-p#sM)i&|S=p=#|$4 z$!9JJ*J~>U&6O{N?pk-S@ybiwzEz4*YK!nwWzOoO^ zMUmlarg}n?jF#RwVg(C?3y{;zWbE_jFZ$jWuEU|gW>rN=(T1a`;J?~1wS)5dQ}H< zi}@uq_c|xM7`O`?X1~UH^)FbvDuX_z9iw04(&*%nP`rFtfIdGbp)@oe)x!+1e3dh# z`g){UYD;XTMdv%@vZ3n<@mv+jS&H=(^vD>K#+K3ib72@-n=+3cT3NnEM6vpflTh(EwX5T5WI&!(vv!`C^L} zqgN*3aq>cJ`nKjjIvKw941H<-Mz#qemd%GTOtu=_GkPB9w2DXSwG{ZRi^k^L7trw^ zW!Cw25l$He$n9gmo#Z;PM8r;1_aA{e(MB9|^`&sq_nPqYoeHhp8%!z&6KVDGlN5b? z5B7N8#TLPqckRn!Q=Lj{jqS{C7P6?^mT!m1M9~4TuwQZo>-@ zZFd5erXmO1#srGGsS3ZxNeB-W^X(O<)0M_qw7WEtwmrB-N!Kkgt!OQxPt`+tk`s3+ zy{=d{_<#_z`5$4efj)p-yWAP3_4jR z1dS;XqI*83`IBPBJytBT9dqdLhBE4X={E5e!V;HMczydI-!I9L>|Mp!-=ab?Jy}|yO8C$4Tx`l9qFh2__=#E-HN$JVfQza{D0^2{~iy4>h#EIy+|?m7xo-C-H^=o3N({T~P&)5SdTL9X=tX%*^n zjk(q)h_ggKiQowxI7sx*9yoJT>1prF!UwZr;mCohA}`)7%E~@R>E6|p81RnnN6bV! z(>CxB4pY^BCo!L2o;;&_lGFHX3K`vmXV2{Te0~6ra`D1^uk}zlu^k6uTeR5RqaB5L z_M-gQT->Xx!@D|P_G+m~KJq(-uM28~yphMz+3vY$k1^$~pa0;!|87%=S6_OquS!Eo zdWm_c`9etVBSPAj7)q*+M@(`(wnmE2J@Fa)zuW)s`*YQa?3%oZ3ojn!q+xP=*6tfR z4f!A2uN|^5+x0L`-`)bFQ=*@B@J=pE8hGf2<;7Hp}R5~ zKNr5`^5Iwb^P;l9&({3Y`S|=yUfR^CDjnjfC*7(ynIB5P^Y`>+pBahB3toW-NeQ^q z;WVFof0C1%&vN*$tGrP-$_Xtl@zsn1j;p@NdOc3DPXFHAXWKA-F=fc#`lE$-cu(oT zz2AuZ2gc%>Qf9*CW6iiYK$S1djAHFiiTrJ;FYmT5#<}=i*tT(#NZ{H|YUt5Hn($ps zdhhHbmR$VB2~VEz6w4BHACre1F^)7duRjOQ3*xUW2lM>T3G6a;5j!;|v&-<)tW#0W zYKc{>Yr2U89k%e6LD{SsxecdHw<9EWEtWm-h2cRF=V5U?x=T9oc0c0RR#tp`N?UHd zK#N}m=&+O97u;}FV2v5ikn~BFwMYEKnM==+XE2G!ot@5pVUu}cZ+|vU8^i~4hVxHF zEq;=&$1Oe8*)O3GH^=Y64bj$N(Y68YY*cvqjz%2npUe(ta`@GeXl_`Lfvt~Han3Ud zZiD*rkwfFTEHRAxxzzK@!y8!p$S&@YEYD_F+wwu1_o(mGh||e0@TIrtQ+;B~-<`Vg z&T(V;<>oDnrT2M{@?(D9Tq=7;?$IegVc052RMw&N*9IKc6Zg~)Q`l$M6mDgvz4*y#_U~h_lUp)ka*MJebFyJI|5M7r9OBYrbonfNs5_v1ry9%zf;{ zH<#IQ+bm+`TZPcd$b*v8E(G*@4QKbZtQn%g4kC`IX7C2&Z#jzawACnWJs;-}r6Jn% zHXD4Cla9>OkXHZPg=oK{Sd@^D`Ne;HZokL9xBO5Z(R)1W7tG+M)9VmewH&K*ME}(t zJ*Hb~yz{Ld2YomMyVKRUzVtZUya)2Dv>816>}cL>Gn~(S@?`k~J=ko#h#_=Hj4e4A zh!ppGkgQW=hbDb?R1xj2G4Js_z7gYpCdhPE>Y9QZFY>rBE`@KXC$sy?pYY4LkKnx@ z(aTMPy$o!5rG^nZAMVJTd#mw{=2jw)&con7yL0BFS$HDi!tStJfh`L*BHMQ* zX5RP{3v=O@qK@XS^|?^k(_sPplxD%eLiAHx z9Ks%P&+>8ECoyik32ysqa8nw{T{J@Z@SbRHv*aXf)y~S~UV39LpZv9ozj+t2^s5?g z5@jqHfxq)EZR{HCH;m>TnsFRID4h2jje$Y$CAd`Ri?)lzvr6F_-hN>dyG^ObWpR8B z4f_+b_`4s5PyQqS@AosbR^suGVqQP6geNV^bM|LtMEb zWD%z*gtMM?0-txwgZhXA*xxb_O?3%>*SA;WLn$|!sdqtVa`}{v$aCc)cl%v92V3#j zFAp$sogC|#c~HAXZCW;dAbDnMah8=TZ;v+S2KUP%&UrpfSn`}^Xw>l2prafgc=T_3 zU;2kr>AQs#arQ`yMV%KZ|2h}UXJ0Pm^Jlkl>=Kb+W6@BV-fkXB=Q{OGysu~-f7oKq zsV7XiyvCTNF)K*2@)W7-Xa1G9-yRbjSlv!Xx0aO=vy>bzK|(vdGcapAYYA`bF#6|`M$0mrWU$y&@EDoxl%_jlXz zlrLS`w*MeLy4!}=yc^9v6;t@O`XuO=xgzlGpXcds{cq{!1eL5$O6}Lx3kvea)K;OM zrbhikvv+b#pCcKhJaKACEg+Q|G{n|#ib?&uQg*X00(*PF1Kd^i3b7t{K( zDIZ+fj}O`nV6!KFX#05yf}j1F2fMQ_(GypZ3&lVSOC&0Y@A8s{&K%5+Q%A6M@4@W< z$(&ru1nSzPLFt_}c=UZajubgKlkZx}^!i)Yl?5xHy*VA4>;L3FNdvp$;Sg^eGJb~J z`UZUEb|WrVn((iu%nLU6=Sg$A$YSDxEdIy7`%SaS=?=+~m>LB;q zc7f+P>}UNhD)c7pmY{j-52ow%Zh9~oafWuqulhSyZfP3Qwlz*9KiiUWeH9U%t%mH= ze_|Iir03Q2+D%4v!D+HMIfi^s22(*Nv5oWo_|U6;CHhlc!*+a7!GcYl&^1Tjl)Jv-m?Z) z;u-aKfBgF|HQwF&-}g@(Tx2p2cYh$5c34e%7b9tlX(r{jNuuKUpXtcA3-nlPF-M5= z_=-j#-)bDotrh39#mVu!zSSR_(l_t_*#m!^WQYDCqoKq;uy2!u7y0v0KR+2?wI|DD z2s~_r4Njx5#G@OxKI+I9rQ`UB#WU>6zmLq0O|mi}=0uUvK4d)3iW>(E=Wy$hvUwdj zM4JM`{t;3ldb6-RjCZfj;qNhX=*>Q{lj-J>Ymw$7xPZPD63nReha_|W5_OUgHh(?t z3^{;v$5MHUbrzTHE#?k_RP8p-iCbgaT zai3B=`=rkrDKGIVU%GIVeE$Qba>ZBn{8Ula?fk3T>5=_7d$~Xc(y^WtABvQbH&V zDilT9TLTqJd#HY=`~JRqAD`p+{c#`nbzk>&p4ao-=lL9u2X-wrhV!|_u%0uPC+r!| zBX2sfMXwp$UDbnk8*Jok*LcpFbAX*?BKe{9ULLr}hu0o;;o=Q(81X6xLpEKM)N_4l zzWh~_20E1AhDq-?2Ch)!L_1kN)2jhDE9H2!V_R<3sWrFQ)RV_;(%=>Ql-a6hCsyfs z4_$L|FmtOS`&9MjMVkHDw#{&kG5CThFDj6`_YGE91;HpIOSZW^hd%g*sySECeQGTx zCaSQ0?rdJP*Po+;G+~zamloQn!ZO5+A8#DYX__u<)p-xkT6Bs}ugqp=+gLnYl8l*N z*HQWV4!T#1v$jkv@)v$YT4G<8UEZ0WJiNhym$LY!b|uR<3f_z6cXsgk*O&TLtwYkp z@jOR+BkxS|=LJ{h@v-C}4qFt$D!KtIKO~F2FD0>j!&k_a%kZQQTHMktlXu^W=g9Lh zl79YZ?R&~!(F#)?Rk=^k0o-3jll`wd!AA6tAsaVgRaFTts=PwM-%`YOItufp#h96p zC6Pg4tBPctubZF5$^Mu4%k?aNDmM$Ep1!ztcmqlc-oo`zE1p0<@Zw8fuA9}5TYMPF zcfNOm*^&0xRBs5qdza8Eq#PE)hhq9hKj9;K2+GHjV3In8&3%dciT;}3+L#@Mzjxcj zQQUS`8{`)1LrY*X1a>Hc-`skP{~$QGR-O?YeyLcOn~y(3>`_0XKNh{~4c+!m?7Vdn zKRN5d2N$Iy|L}PzZM+Ig`=_vO{Ty4;pW>=qKfZ3?oljlv#Ah;`SufR!FAQN;?emtF zeSbu%uU}ICtB#oRa54EN9W+9So1!eckT7&oX{km+RuV54~*sle_u)6 zdbf4fv>q8@)FxNTd7QmE7Cov@Cd6R@nDI4;DV`;M8&XBxgr=AS{kYAMWb z1#7SI|1LwBn<9@I9R-h{8%rZe|^F7zTRoRE`SgiaA9l(fB+s_#`(uQ7t3 z&R_?&ep!Z}Nc{Tfi!qO;BEsDl zLrhmAr0RtEOP{Vv^1D6JH{1`=XV#pKYBZ;4ow!a%l$U z3>}RDRj+940Bu}7+6RR>D^Xy!8C?#~fmLP?-ZlCQUcI=6lAcX8Tg0AT^!!4F8*=de z`9XMSr{Mlz51wMVl54DnuxIO~2)@%3JGxDS|CJ7i>FSQ&BP`*3w;Vg{4x!d)3G&lY z(SA`IZZp2U@IiDy-n|wGZt;`S7W}0lf={AT`=fNw`YgVUvb1zi}#$ohv_|D z1G)2f`rl6c;Y%AnHa!xCXT70bwhX^#S#f>6DbHLgY}nopFJ4ZZGCP|uh+P^SiVg*XW>hV%95V9NZ3U0Vtv@|}J`kDpfl!(>K z&(cAzP9FrW-GGy8HPFQ?gZ$0oNIf8eY<_-5hs>^Aanlzjp$qYBDNv%Y62}^hQBvcI zznujGuB8w1q6EjH=yQknKF7H?_h2WViNw%s!4+IcOA2L?X!2fg5Iv88o1ab!N9ES#XviB4<<^4)*NU)> z*_@Tg<77d1`?;D^*mlP{h!A!{N7gG`7h|b;M;O%OrKWYFpmFw5CBXsl97xs9ESS`UzCv2AIW-6j!O%2Vw zRYLKTWHGn+1vUiaLDB3Fx(zu?4-C~1*z+e{7jd7>`J9;_NUNtxQU>Bzkrsm4;LWQI8w>ICl6f`WUOQ%4xAcT1AlJJZW6@gq#ZO$?L$! zRM*neZh|{I^xVJtYnCg)_J4j0EA;fS#nA*=!nf?{raf4e-b4KBU;GP&EBd}-^)p2M3C}SR6G?BKEI!n#VAm!iv_3Ti*W`y{<87(#yQp6l7MR^e zghd(RGw+~F=V=Ibb;35kJt%Cdq*;Y^g4?_Y(x$AG$Z&QYqF#s>N-ks7fR1G{;Lx-vpXUO|$HU-<2 zl6=Q&G>{(9suK_Cq{Vhg{QKvn$f~9lX>%rTwZtM}L8f9SP zc0&w19!+&`8p(Tofbw?R2K2!`s(_eidHBiKYt9cVV!+9LB5m$CKmTk-7DM=RJ>syKy-B7WSyd;mAm-oVjv2|MJo8 z--=ayQ_#-1Oj6gY)|6Acxzw-5yds9q4(~v@vmz`Kqf6v%Ax>F|HO7zUgGK%8FVl z6kD_mQo*Hh!hh0pmqpWy2Eij{AjwB>{Vq7!#Y(O(>ir=Fqk9rti}E!EE=)I&z$wxSMoc@C}vX+BE00Kt}-uhlkOV^U5^Bp(8df@R8)7 z#`~~(cD&?z)5)t;x%h)ceuoPdo&%O(l%AdB{`g7Jcy(Rslkmn}8SfmHOYV;vsf=yg zis-`oG7D1I{8!&jK~+dpl={$C4wb>yCMEJJJ7Tfu{eG%6*hr_B23Z&k55q1?>AKmo zWgHqUA4&2o@R$U-4XY&A0bWv{!e;p~RM^tQ_kW*?kChe%f*0OXV4S$*21}mz4GpD$ z^>UKyADd`s^BL@nd%u$(9_#f1_E~J|wG8np(mBrH&jtE<(@>J<#m0fyYAE#=?D@Bb_ADAm z_o@plhJRRrIjZ^+S!W#`iN%m&pES$8W6oWRjkb?03^xd0upuKQ`Me3Qru&5lB<&x+ zI)Y9*-mqA*F2y3att;l)Nc~=WRw?3bsJZ0%#+G< zU(#06Br|_14>&AI*D=A)B1--SxotKix7YV9;?%dJmlzMt&z{;Fjcd91^Ny$#qpr5mzk3lItI-LlL3oOar*lLP;eUH|D*R+#jGVydaqqbd7mp-c5gF<0NV zxa!@K+k`su@vpA@J=6=YcFx2G!S}6Zw~J0`6w#pU1(Z9+0grYLLefcDC?7e56~BVe zJGCQkz4#DSCluJaLnlrtN+ZinWyIFmbi(Kc8vnNBt@lzRlG69B+)Ae( zw@X>VNXSLX-!nM%Cl*8V4$+oFrzrC6QM&BegJrvJVaHDgxc%XexHq8>XKHt4wH~YZ za*+;qjEv#ZDQA%srpf)g=V8J|=@@$XJdTpq24b#5Cp_q_kAjZF5cu>3{Vle`GLJ*Z zkM%(1jC!~yHj;McUpif^0iR1B(Z~Eg%w~3D=yvCoCg*WJHUozeq-*=*LL+F1J<(Hn zfmJ;2CDt$g2vdzJbiex&YuDss&G9UB@;i*ZT9tT|nGP2tFC1>Agtgl){A;^4CdY;E zQx28H9Du&95e^k6fjUh^)s{W@-E}ceUR;Q?mR{%@ZHvV-k}%8OP52pGN!sji&u@}^ z+&xM#BTA0@&b^E!vYH5&(?;Uy85lKnKa~pI@v!}eXs(e8rYGJbxvK-Q&RUtLoY{{F zA=B9U`d(_;$-1MW(n;EGQ-x>eKEic=2Uaq2;8Tvw$8K74iH03M*{(#poM}kf*$!=f zO5eLaCJFPald&Pb4%=OCVMrcvQj{@U|B~*P90adWbN_tU;T;V(&mq~`j;v*+zzbTx z$Ls^=aPDV`z!G=jQvJ!CHMuXJA25-4d3%8uBVvZ%2Y~7h;&`dxp*&)XOV|73(7G(VJ#`$p_HmN_pc}G+ zd<$afuAU3tULc`ug z5(J*rn7*mx&_054epS%=1Fnd^)e}okD!?;BOXxeyF=e|4?6d?g@hx?@#A`#l!Ax?V ztUTw&$7_o*VRtc{e*M6(kNaqO<_YTab33I^_z8``d+0o^Etl5&l7{C2YMJgsi()JB z&aRJ$mz>BUJ?s(f<%rkH47+K&>9EdbiOkx+QYGKb{b+xy_UtmK6+dgQ#+4(Elcu-y zT(p*s;z%c<$GCS9p5M0d$EbB2k>kz_xB?@0?!eq+)*ngg9vr6$4$x5ImA!m%vU(35A6*E4p9)T{h+~aaZX8wKxn^Hm{ycgm7s)$uQ^+vh-=rtvd)4^jte%p# zm8>x)5$F|cjs24pC4FsJbPXPU{)2{P324!jhS{8jg1}T%Ra`;m?=g~R+TVGD?A=nl zou=cB(5#HZF8{MQ?qI?j_fBQ4p!uHqisaH75hf4nnAEc~x|N6YvE;NG z#<`K5ptxBLeVUBzKe5OS5bZO5?1ESKuP=2n3ZV=we<`6a`g(fdN`E( zzplV$-E7#sNW;P_GHl^_3nfw6sD5)B?;Y=>b&q6>G*@G-Gwu1YkFd|Ykop8S%hX`7 z5gxg&c=L-es>NC?3f&FolM`{MXdJJ8H-pC%nDT?oVLYUD5RdKe#ovW3c-qdM|H{;y zN6X)dc(>~$oc5o<3zK!och%vpRXzERmk~eAybZP&`-#R(s6_{H%9d_C_5A?ewoAk- z+^oa-A(vn`bUEJjGL+;w^xIT!_`ZS7(zU_RG!TOTk@$pQu^5Abj?zZdAf|PR`!>SquDvBNc7r=HTz9CXsQ~I z(|?HZMnmwu{w7K~M4)8wPBJwtCCj*7bf&n3mTW&py)Fs8Mo35awmL!?T~g`(e0Nw^ ztbuy(gP1#iI&U7P%KHZ&!8_G;l&|PZgGYzZryC=Y=28Zaf=h5;5J6_Iw$b}1&&hpp zI?_A~;8~`@BL*$Rr=!+TiQ0|3eIw}d&?eGV2%``4V{v&`2eu52M$n!WuzuYS74Kv5 zzB))!zGaK0`kq!3Z86GfE&@iB(ks_0GMsjk#%|1z%+>xoO+kk*(e&0WkR>p}{_A`1 z+FZo;br)&%Ug>$ZMXz$2^I;WyHd^rJA%a&=HURVgcw^f3)g{LtkN%dwCo?_qjq?BHN{6HZ`7UBN&_p~q10uPPXKu&3~@afx0 z>$~~TRi`8}wY@TvvIfRd>Tp%UGh?4Bl&((CqciV%EMf*;;NQJfEkuO+8bVKj7ZI@Nsl zk>p=caT;ejg$vx*42su{qX!2~P#2&Bjj+j>H!=(gx|L8D979oKBS>~hE{!^G4UT2u>t*`c8@dZmjU!AH{`ID9Ecq+!at=tt}0q*rP}Q@@sxwsjIY z*x6&iGb@b#Iv;wqjdWMwMY;WI&okm5;=79>%jy2$kK?6&6jLV7K%cwrC`fq(m3pZT zXP$@P`P;pduG*ZYLFu1q+5l^8Kj=*R@BF0GIw>TxB#{CNdLgq*Pq_b0qydwYXlRos zw%wXfmToSj6)Qatm)$SJw#vbLw2uRsra056gL?#z$9wu&5KRZZOvb&y*0Ao9K%GxA zbd-JQ+Nx5jtzAVi2Y1r#_JZ?BSsm$lMfCROK2m%4n$EX)PZkGqsO;h(L_fGf9TwKn zx}Z)7ICF(wcL=6dd*_kE1vO;s+e!y=t|=xyRz`VpC`Q>7QB+h6S#_zQ#2G#UL)Qf1 zjRm-%Q%~9V3rTm-ZtD8O1?uN6kk{e1*sc7TJQw%GN(Pl&8Z!-VP7D zA3%b^0(iDKBgwOC(;R4yU4uKZzsO^gKWWbop@(gHqm5%bD8=2PZsn0wr?HGmmoKJS z2Tmch%`)_gO~L3spUAFP4H?h9Ppq3qZAUlJBsB}Hf9*>{AH-1I@=Q9kv@<-8C?VHv zF=A?m3V-J_G|MTS9IHFSKjaVH`M3;ePxtU4&z1Z=&VqYa%W?XNFG%^O%S*b|g_cR&CM8TV&klZmg5RAk!1Rf%WNKXBDoO?|v;44ipIDU8%yKeL6 z`z29aSKoqr8b3hPMrohTLht)l=;BB!XpB(ZC-zQZf zz@#<%*(&h8A#z;1{VC?%%R@s{DZcKJWA%q0;1(e;_Jm)C-Z}$TeA1dj%Y~oM>4ms_ zVGG_okz?=5CKMHly`Lz1^XFrm1mXYik-P`~p!fG=k$j{N=bM_da)>5(P^@5=voY*^ z`97-;*eB_WK?{;_@xV98mfV5v6U#F#!$Vvm5ti$WUc#l?D4#s59sGG##ZnO;bUl(AY(L zpW{yG7rJ08IDhgU&|?{8%N}#Tu%5wno_A*p9@Pb*R?Q6;-#o)}1~6FK1-q{}lRrYSU4HSrmTT==?s*nyQ61_4fsJ9d`~g7^q2X?N4nl*HI+V+x*gJXHn+>wf?Q`! z$URiTw7fuca=IyE#1F#HI2oJvW}u&988(cWgjcVEFm=spfyt%GHzTVtY<@oO_8Z0h zgzukocej6S7w?jbxeF%a?Xq2Xsb>AJy)u2YI5h7OO6J#^elxwc&W7*B^{a?e_5u)jE)E#@}12 zDK_%KzkN<~pHOrzL5H|_ELd>|zkP0ybD$R5zk5$ZD!idQZ_K|s_;2j}_gz)I4gGXJ z(a`&E>Bm%a*w}ZJs}NBb(~}D6u17dp4zk3=c0Ksg`Z_v&=^VLlcuVUeE>YyrZ(^-e z!kLxV=yB8^njfx*{%(pY;Vkw2D6)|1%MH#BgPrD6`X=_1+vDC*cwjU%0)lWiL!3WG zx8UKC@@)0FJ4g2X@PEqFEZ^N7eV~0u6P-<;(8E7FarTaMU%XVlgo?Y!z^q&b{)^|q z$LR=4=Ptvc#6o26w-K0vV{uh44aXPH#;F2<9ke?b=ccY;CH)zk@@O6x6;H*6DfInz1-(k_hE^kg()nXcv9|ID`6bH0{zn`1&d(u**b4fpZYF!m!vM1R z&*{@T9enzpk2_N%u(VQ>)y8_l@W@Gojj@xo(Kw4>8o2)oJ*ydq4t3T@8*L(D3F9!V zA{8;)g5h{<1Gi8L<{ekUIiT-zl9}_7R?ZXqxY-@y-%1xfOiZvhEDlSWlHfPMtVPbD zmI%+5uH9xSDSS@Ok7J!qvXRunct?qKFh>D9-P}-i;0F3FiNNSZ6EU-8ITWjo(t}$+=#t>2kq-ufk1C;;&|L)X z?~Kic{lz%`LoXtN@X%oudValtCk?&v;JzDxQvGAPLCD_KPxA(Pwf(5co1^($g=USC~>fBMK@DTS_*Wt~+XLK$9Azc=I>*eq_WYD7>dY;gLsa=1Zye~LJ zG_@grME+m-`=p(f)a9P-%1AA^K$ly+qH!XQziXTZJYMMIaL0Mrwx)>2zc@nWd2-mi z%7!2Qvf?%I6FGFuT6o?aEcB$SV7UGbtdEAm=!^=>rb%sfXZq~Lu!;A{`duSc>g&R$ z>l-RheM8;6#NKyjH1%^Rqit9J(5~QOD4VNrKdmn)tMozsoDo>LVY$Fq6xe|*hxCch z1gAr};CGdgsWIw}6W>Q6`u8}TIB*186hBkdsLinYHVxzR{86X47S~d*^6CCpLh@5JJzxE2S|Jz7UlKb+ST{7I>_5(6<|7Wk0 zSuRlXl;lL*#2eDrB3`nzJFhkw$Q?Zmv39L7ijTF1bKzLt?xw`ojEC`>{uOv`a{yO9 z96;zp8>qB#%bAN4mID z*r;8Ple^9Ut`j))9j`xyX0#e-b<$z?NwRF%eFCPBor;X*JMpRY72*+T*)g9wD#Uk-q0Q|$nnZ3;` znx!g-QiH$r{$&nz`}WM;p!2d$111-tG(Ksbe8Oe<1pF?gE{DhiQ0{U88L9b_@;z9(T=K16v=J@lkAz0zG>j|i!XIvAUZ-PH_@@_2Yf01rge591Z+vIEE zj8=|OSYhXYfh~M-s!VEY-(tQ8em{0d^0+515i8&sh#bk^%y!w?nJBSr-$?j6PrwGonZGb(-hyJ-cXRYOmxX=}qW;5UuS z)5ZDC$*`VsA7gHwg51)5A|7QG6n4Z)(u_2Vq6eGrQEWgb*!CD8xi`7rM4TGg1tBW- z`06->i?*5YopvDzJFrXG85Seqrqmwa+{Xe^$|Z6>yfXqnJO(30?8%;J&eKXwm<@CD z794+|1zVN;mgKvBXgE|>m_xx==v0=rm)z@SB=t42w;-g$%OVhs&IkBt_1HO&mBm)fC2_ho?tLo;}) z)oS|bbBJE0UZHWd6ZpyOxjb9@07rQL&;GvTTRD%|kM8w>NQ!Pu=NeE814gTj{-SY6(Ly=}K~ zTGxRb^sWtOmj&~Td;9p@fa`doWyAQ@`rn?pxqtk5G>=mHPAC7gee}$>C!Sqcg&u+< zq;uGLT%Xj6588f`^n-_Ps`w&s_M6j8=;#LB9s=wA>K(coriUe7s%UlRHxf!5SS3}3 z%b)E-ud)b?)RLYjn$!5sI87zAwWU)$+C@cVUYa=1m8mLI21 z)1x4pWXgtn7P3a4p*%_0PDbV}=0TAGh-?5fw@SxFbDqtgvWW%G`@hFMzcfCTVUhQX>yO(&$xik)( zDfJV$yRKFu?|XYAeg=+ZSmjp$>;O2 zu)}L?+}(=Hq8eai5DSOv<9Jb;@Htiprpi;Ns9moUh@Nx+XKE8Lr&lC$e!HQ!+Znvd z`19}GZi72>-rB_+C~Q<7(;ZQy9*haS+z=6dNAkV{H=59Mw%~iedx}oZD8L4{({Qc` z$HD0){PtTKxfG73xcR;`m!8qBzS*>Cxh%e?cgC@Nf%O)x3r`I@4!rY}Z)^IQSOp58WE;0;%?PAMqrm-+4_I+7|rvJVDi50(0p_O z>wkb(=1k=dX8XCjZwPqIVt7KjCN3TSp4N)9#OCwk(6Xi-;(x9{XkaM2M33jVUOV~g zhcGVF5957d`K+h3pL{o*q-hJ&B+YXjN;IWSQo2UbimNPg?JCL`H znZ~K|J2C24I_AWsBYECe@()x*UVD919%{?wZ|87N-GjWk&WlwxrE}nod)(Pr_?`!e z*zB%**s9?nerRVwF|j*GzyBy{1CN@XuoQdKG{HmETn~lT5yHP)fg|_EqIa?^k1K1* zJ%p|Ao0u2p8SBB;cLVOb1abF~p*$aF_{>6kK5KG><+2m_#FP&-`HC_6R}8?{_2qae z)0$haGiLi=x*V}ljiY)m3u(meO|)sx z3R*QePMj$fxW};?^xd`>u2K0o;{8%0tNa}4S@gg0(fqyDNtV?eTJf1GBmR^wd= z8ZGoSpfcf_@wz zr)Ou#_MWVW6<6Z7uFRGLC-BTAZoEn-RbY}DW8PxI_l1X{c;pkA3f=GT0DC-CsX$F; z8`jyX%_F@h@EqSgT--C0?cUl#FHXK=vz(TT0%%Oz2#P-`colQxxO!SkPS(-qE~2cp zQKLDgY&+}M`11Vk$9R=&H2=uXhi7>i%%7FP>&Ygb`fdOxP1?(oW#dS3`#v&Xb%EA2 zE<)a}GK8fbfy*c@=5g#tu!7;Kbo;4T>&4x3#d5a1j#N>7u-Is z{2;U=cN`GO-?AHNp~hL7_E5V2`LBKtHR$l&udh+iOPl`+jxfK}yA*r(J+;0(7PI;Z zAEqHnJfhwJ4R1YR_GcCHMn$2twGOYf_<#{n+8m|Wf&+y;sX0Hh+gr)GLnuAl6@qbm z6Nc-q!}!rCxIH@ojSl%Rh`dQ@-k)gf8Nq+P?hOyZ#s8cBDj(Z@ZjVn zOgMB4_Ga~xJck>2OTOE=tV7q9X&C>o9mjTxpzixG2%E}A(#Z*;`Lb#V`S^zZJnw{H z1w;Hz9*99>UHM6*4Uftw5cQsmc{{!$!Fecek#nc6VJpeMBwnm3Cdj|O1#NUzq8Y0$V(5^`Ox~&yIUM|%)gq(d(ubm8F9w6!nm0d(v=JmqbtKI&k z+4*`Cc3a#Rx)5Ewm zNJDt%&-Bd^P}iwMu-17h>~x*33ta3$ErVs=9lK4ZrV3mANa?<~Sx)T_vgp2bCLSG3 zlgROVq9UBSi~=%JBzYFzzx*dPynMU7-KRuSGu z<8W+$1#GW$g`7?R=sybEAe%600v(5mb|yxLk_DZJHu*sBxzh-hw|GD`R;aDT(t*ZMBLH(;V$S9v=M9i zFGSB%fv8<1U4Qcq=a6DU5G>cdpsQ;KV#c8nSiEGXoXd{mq}q4fzrIpeQYpFa9NGEb zz2^2D-8d3o@9(5@6W`Npn_9ZvHI(XtN@-284xH*QP(Z9R22||BXkSfiI70AQwH!+i zCX$RnU%WG_#7`fwX0Mf% z>HQ0uSiSNf8b05l;ueCdD7X%G@8ytvzawtkxQeU{bzvito|XC^nn@ZXm(b?4Zs?#^ZOq4J7`JLp)BI_BY#_Dkltq;phw1kNhkyAUcREO6OZQO=y^*-s#}ryc z=92%On@Dl(+P7JQs)p9M`$_OlN9@PQ6}HeSkm49@z1A0}N4a5eH>tk{<=LTQQzNDt z3_)IBQ z^EbUov&NIJExChtI5y}-VcU-=$#e2gqEYX4gk1Uvzx-2q^q}w>V&*LSm&bqQHGcIT zzx$V>`qT$Zur5H(kSI)kC9ppB%vN|de1oKZofdfT09zmSHS*!^?fY<@ej&UZwqi!Z zXukGtjo51)XXO#M(QJpZ)wK37je@Ck5S(D2C~LpMw9jhSiGH$H*WGM9C(xFomD_X;U_GR zKO^E%gOO$1jV;@}#%p2EG5V2(93?MF8Qt%vz#=LP^K{4IPt`(UOSeR!QHpF~udS57 zcsYGrJf1>Z`$EH3aLSki8W;DYoy7`ld@Fn-qT&U=&kBa{`T4MdSCfXkOiZuU9u_rqY%c3hUI$+Qob$DrX1IwL7 z>EwDi9{!60^<6ao{SDqS<0+s1rQ7;?%l;d;&{SoZ6Vs28JQbj<-i2Fl!P=Lxva5POKx zCZzDVC&l704G^)%Z64kyMMDL2AEE~bO$D5N+!=WbeQ?@u3bsFxo}J~J4q|`ED%f{l zM(fLF(hUz#rL8MZU1Q2G)0Xg|ReJDt`a*G+q&k#T72kh-vaVemjB5h$V#gMW{aQjB z4!xvV3)^FhQ!Ry>s={Wg;pmb;q^+D;{ZAi+yMJ|u> z>C}gG>XzXEr5P8Iy!{UVUm~+U9?``&F*vj@$*;Dd87nu_^b;@X$*zxdbGS6t z>c9S(O;jR#I@m#)?HYB!DdpCSwafKx6^%VWr<85SLtHBi2ZCO^wz=p zVm%16f`@)*)D9XW`9G#61|5zk;Q~zAVxbcs9qZ2rza*mnoL_L*@jqQe^SgB~)MHiA z7yR_s;DGnOmztCr)IWCZKaBLGF^pzcy%<;n}U|$~wuoZ_dX- zS&LWguEp!@bd-lL!6F_4Cqo-_x6aQ~58H`*kZ`a!ypGJ_;eW>SGACPJcqarp#j{~CWf`nbXu)B+I{JASA;y$=jNS+y zt-q1&=SpLTIy)p|!FdPXG-@^fa7pEs+JE6_q`(5~h}{LQuI8Vq$d;dkStnDWoD`0X zqyP+adO>Zw)zT-Ek@)Vg8|LW;q5Gu)ldUw+;p<4O4-~ozA!G9=E@w2aSbX{4?=jKV z_~7P(2XemXxL`0=@Abj7--7=uYasu%>CMLZf;#ErXI$1*=UHu;ou|(P#zkSK)_g3< zKMlKX8L&UH3@cPpNU5`k2?>j(>2n?MM`-{WKlr2i(_T2q=c0JRZ{d@<5oS(X@g%_) zcUJF%^N;-)Rdx``CK{M3@SRU8GRg-2#!owWZb~-gTVBF$Co6ni+r-0Qi0~zF3`EVK zxo|Y=f#XA}DY@8M%mW|bH(vPSsyT6tScgq|Xrt8n1AQ%O%jYXKSoLH#F4-pCPc)dn z#mPlmP*&&zY)a&CBlc?jW+pPsg#o+wo@AG0ghW z5{CEc$}(omcYj`2>tEj@h0y8EdBD(dc;zMR|x*9?-4NG z+5vv|4q(NYtvEL41pSzOoTgTsr|dT!QIt(^3F!t8b$jIMwS&*01<(!E;Zp_EIXXRz zH*D(2!}}R>@r^m$#VCsEl%%m^v&UGY$qAT~8jk(SQ=n(*E^H35i0^b49;q+z`tEH6 z9mqh*@eu487b*OgjQR1bA;PutaZ4mC!p5uRa=lLyG zbJftFv?cc|-R}`bW|kwl#&srlNeSi&=f>go8Nq?;-d5OpTSGgbJsh6v!DjMeywNtq zuBef)Y7vWBR<0&%t=2lp)4N_# zyu9^EPV6Yv1<9=`z~^1EyizZL6`sd)_@Di}Nnm`oIIPbwbu{K#?}l;xFndn<`^myyU~1=%JY&IzkMX9>cN`z1#2MB9FP=2)HRMnF zLG|_q{At>YlYbA;8ufdmZu*Wo&3hp*SQ_Cn{3)8I~m7_34;V$X~ zw_E!EVdSRw3tbzi%|{D(Zs-kimk?#BcN zL%O7%6a2r?jQNJpIJ863f8L!rZZTulId1!^il>-7<&=P_w4>?(ZS=iOFIMQGlZhUV zd(DA=XG`RGFT(py;n21HBuUG8x{`9ZpQxpmkEcWPv?kG;O=PZol(r3BkK~wj@UG94 zy!YZTLu~#q7gp^$Qf#9V8T)o6$G(R#A}tEaHpftRYyp-W3Kg+P`y|i**FW_pI4=3Eg;_*?i?eW?HkkdB&3LurRF2!H3fs7G zX!EJ(|4I8_S%=#Fv3N1wT4_jIVM8q{!u6K+lCpKPtfF@UYj(=v7$oQCp{s6dZuB3* z%Vl2Ufk%Jbbnb^jk3&PLHRLr<(@PD77GhM$a52@Wck;`@L0g{UWtNcx_ccNqi(!re49QWyLrn_#O6L zkcUmZ#{X}<{c?H)>yn!kpo+ zN%J6;CQ?x4a$9v2MzVOI8c^uRrjLr(L-qHfO)9U z^+j9J5|rPP#PeYT@SJ-pIL3{JhxsmSG@p)M;$}3*KZ&8zrM+K6edI-y*(X8U|23LEjU`Ir{&{=2Mqmy1%ix+t?+y*bXH6w6;yAG0 z3X-($mIgJhW27Rrm%ZHt&F3}v_%(5k!4Qrd%!I*`I@ed3vvInn-ij^w>!BF%b#id~Jq&*pHvj=stVmQ?tQ zEjZuCYMp$LZpXbe`iW6XdP=qsEYHf4C{5463-Ux)n}1 zhvU!V@Ou=;or~|oaHk@yT{n{HevW6^zGql~=0NmO8UbM=*Q8QWgXU#dWP94`rXs<3C*GyEv+6CmCbY1OF~&_=pwf1(jK-(C6g`Nc!sTVNoV^WUS>D?sG)rBGgj>|ki^c6r>Yl& zD9Zah>Xa1dk*x-8EQn%lO)hMC-c+V)r3ll2?vM;>V8P9O(8%L&Tex@6fjOnj{Gkw| z-VDLlXB~deQpodFhx|unDqS{)o?COD^?(9676;=-)^rT;NM*fE-I=dhC3C2#!jG=< zG@{}UVuK8r+vJB#cieckGBOy2-G5-oxsmj{l*f$g`bs%fNup&!DBBo*g}GKfXJ0w* zTX)U{wxHz}JAAoQ!=ZsTFo{^%%D<6!!!s zs5Z!D%HQ1CZG|$HuDF-=H;iUdwkMd9u{Ya1K}>A32vJ0;srMSoj0r570LNz8knJ^ z0J}Mz@AXG6!#Otn2j)H=My)n;5WW5hw3XY@-te!EwEu3NKVCV2y_hkA-Raehozm|E z0mn%Na|r&Vr8>ygAB)z}GNO6Nk?W-#d@&p1*0Sg#zXz7va@ov_N~rM&W%swmvpo4C zoOkzyv;KCF#!S?H)Wsf|*_fnpiS-k7a*G~H*g!szbDkuJVnRqY_WejkKwK8?_Zh{0 z&zi@+_g={syyqHLiUA1WdZ^VZPjK}{I3nDXaQ(n+n62xL2eO^Hfu#`&?LCi9?W-EX(MFKNT8vfP z^`-KJPgqr542PAS@r(PZD_N9Y88ce+o@M$MG8OqT7?^2;D{tj6w{>Nu`j@Dh2Mo%_iRU#7EX4O7^{$3jut75f_@ ztxIRDKK3NfFUQ|uAJns1*uJC8*7zc8*mDit6B_W0*ZkgAUBPaDQz4CdDUzD?2~YA$QVa94bQC}2#=(3s0e4SL3kDD8BBDu7Z)sAXqZ1OQS zEk2Gl?ayHH<_WC1_BeZhD7L%IjYW=5#4y!NtTHc!*VunG1Fq!u#d{Td4C>z*$Ny(d zw!D;ciW;4emTn8}*eB5Qc!`~d?{J*$5+sf0T4}9Q5xTk$C9SxJ_=#DfHgDdu9<5f2 zV)HCcq3h#woHKEedxcw3zjeQGte_n4zCFWn+mo0zx*pd&gUDy1JMBp3y7YeC>C%aGBfRX_ba z79l;BqO7zOy-NykF1P|>x1Zs`?)&)Z`V?_4*_dbHhoQ;N+~eUa_V=hokLtZRH$M{k z^_$S^{C4b^=t12wr%}n!{xp+gM+eUn(68ucxNe?-v%O*=;U15l9LqkZ$3B!;oD^N} zKYgWoh^Wu)yXb!CAdZ`jr^-nWaq@XEOr}*KqWK(5>Q4ME|91NcIQ0IB*`rRtIpzQ= zxlfnDfquAIC<(7!9}#}^4Cc8eVLMup);j{D?w&ww{6*+_$8h}qNpzXE5xFJ?Xwn~u zgF`Ri@L4|;^B7Cw&`Pi)#YibHLx`#yq$+nIw!{=UX3}&?PK?T$|xrx3Z~~3n+oxmL~dsN09VBWb-(Gws8#X1AbuZ`VUYz*r`$X&-MP_<0YlDxHy)3 zG;5!Q_x*D0uFb{QwmjV6`)$37TtxKe9u*v?H?mQTJb%Xj{oRf_qE8)yp7bVoADxS8 zrOmY3wGQ&i)fhJQsHhyIuHuB-d;Baa!4QimIQ3ow3!dx#P&5|~f%`CLffv`7w0D~0~oZf9VvRy;iN7UPGRHI%;V@EBI0s6DH{=)q|H1xN;*#OKC4xHhUg>7@if z*2Elko{tfjd=oC&sj$`Wg*nE5M9+CgABBF-$3uQ!$H>Z)@sLh!hX33DU*qyP*75@S z_b$MYVReWW6DNbIx3FM+U;5T6P9IaQ;k#Tl*ZO-0@w~9VebZ4#myfS7tM7YU-rNkY zy?JQ&-wn6!OL3o%<1bAd%bXsJORN9va6QJ) z{)LQV`RMxT5#$TfF@DS9|NPdQg3i21$GN9_S77trI-Jg}K;IRQV1J|(Y*#Vjy-Q&H zvjDYORoK_S{TvJoXoXcWoZEiFsyH9N+-^cIT$}{H74Ux^i;)`IqCM6m+o9B5`6O0M zsrl6! zw665RaQ9%$eiusfTVpBZMge8t&BO<_SR`Hy;hynnbiX{EHk_-c=4XcVdVv|Gr%t7k zwlGx7$6(oyGtda_Lq-0hDSPw+3fhoNvIBxBGANOj&X`0a*A1lXkuzy>+j+EaI)#;& zD`Cjvf{NmqI2sd%y%C1g-&}?6X=~64K|LNSa&L|_MY{Y%mulB&)3tUZN=&tcf!P*R zTLfanvB9xB%ygIrw|E|hb9y9F{7T|0o>{k7>`H0SD%)F7-MLN)bTmln~}fivmC z?d=?=7fuWIq|?HP}6uG4*O{h>M)wD_Uuo zcDN+-9G=@VuHSl*b4&8TN?S31DbFeWHl^k>78H|iNN+Sus31#7Zx=bE-qQuU7x?1p z=m#(d0x_?Kulo3vqhs(S!8+4JQ>D?&}eC12d=p zLVxX>DB}3v&&rn}-$jk`EyTz#r7t}w6q4-Jfm9M>O65Ns5&I?%(JPiWYS21Dp zb!_+VPOe2ED9=0##qZ^~*5E{j`j}_Z z&AAh8{9Um&6qTp;W2Vks++QV4JC1$9kW=d9S+)>rXW=NxuzQ(-5UN z2{QG%LgfG>p;VBmFjo8|ehz5Hn8kHaQRsnfw+5l*@(Aq7&ZI)U{iI%8LdV|tQ(>1h z>Uq(Z7TEick$gUd-#bo)ep+-mWfc87FoV9w4Z#p)J?v8Lhf{eZ@wA8Vo|24kmeN)#PLHJ2)cC*8 zbVnPOWpn|H=_`=|W{aW@c>7L1FB zZeqVtGv=kM)2bh1s7LBo4EX&H&o0W4szt_p6v;v71Cd+g(ywT~tG;MUX2%*@`Je%`d8 z?3G(-enAn%g}tI2lb7^+sfy5Gx}0!8ilOk+QBAs3zk%X+*pbAMcXZWXLpZ!nQ7F5y zi7jgH)V95?BBa$#V`+!}LMlzLN7~s{_{On~*&jcnKA3YgPJI{km6lbQShBq$Jn1Jh z|Mrp9KoVCENeb3OJvnkJlyB=$Wq)>8vpzb>E)Us4;^aTOG;rU=fo@dd*rF z++wo>GNG999tyFS(cX3ysna>vvq6?tXo^G4eTeAy=g6f|!x}$I+Fig+;C5lq)_c(C zc^)(Te_~(?&w+%siptG+I}AleGcj5F04C^j9OsB?tS%Er7--Sje<|AQfr0^ zVP_pvta{D-SL(q{u#I}Za;2qZfs}mnDf?93$@%(T%Y99Arjx<^4OIN9AEp`4ItU^JQGe^et+S zH{g-1Db4P+fRb*+pif;Sf?mf#ckCd%zt+lhj~igCVdm~z)ZI%n=n zA6s(ib%GcvE>a?KS3Z}AX+t11KxwEB{H-qI)x8stds&Nwq6kVkvy>W^M$o}EuTf^r zHPCM;P$rLA_sa6dB!5Rh^%3iJw2?*Zo(RiXIXu^LQ*_V4Ig^z}NqBO)0aio*V5fl+ z$=n@9HP&y~cV9&uD;kaYvA4*x$8%bAzN>K0^KB%oj-X?aCuy=qU)0_yVHcNcaxHE} zs#cPv>hM7vW3iDAxeTHOTFSI#x(x-iJ5kOwAKD&2pB`0O(VYkX>JR+6W7%)7c`maF zmWQoZ9ZUZ549C7yVgCUcQd|-b=NV2A&q=`|&(4@w$N0$C?etFQT~q`6MayZWju{Q& z+BzGSuEXI+;qYp5L`9+>$JQOh;~qt#>-16U^eLWmCmti5?sH!1NBA87hRfv-P{rdz z$BMqO)@M(cwl>Ei<=fHmB~|G8NQoX9C)1Oi4pfl0Ni=6F^5>vvu_mV6_ZQXM(Vo(R zQyAR}DzlcMKAyuVdd~#vdp+>)9NST*_DoOo$(kWL$8*(Sa;q{Vul(8cBCHqn@A(bR zbElFb&%dsDcATzQ#L|-`jrcYAE0%wL3dyNPbpE;psqb;1;lIzbFp)tRO)lmcU zeDfGRH=W0t+v$J%pyL_``bd%H?f#^3Y&>mH>OpfOI8GpK5FO)QE*;-%<&H(;b{{WrAWEf70zz}NP@P^mEr7fSB1jbVN9V1OznQ368m?SoL~DfX$x!FhH7in~W4!;$Mq z??^;iQ4wk~J9Cg7^=|X{EvkR&*5A-CzRXN5ZnCDl?wBs@g{4K~(M64OCe==f<|5;g zW|lp7CHc3`qlDr}dhmpMF{EcAtSud8Z@O|h%IoZ%dZ(spN8KHN7LzvMfWQe>DQWQ6 z7N@fwKasnqQ}4aw`%gpM(0hv!$xg3?%jH*Sna_3GKHNfc?scp>GYDH2Du6Qy(9~p( zIN?T^ziP*>IUm_3&l5~n@g4JTxQn^3n$fO!9}+t_Z~6Wftoh^#-ES}1uP;y8u7YQ5 z?x$S#=v5~#@wx9T^4b$f6~d7Jj4_$-5_E)PV>;S&a?n(IDzlhg4)dVRdzIk&Q3^T! zrYKO@jiTW;xNzPY(>V57HtH2SnpVLcs1L*agz4y7XN=FWG1TyWA6as3HaB-0?ENzv z3W*+~KDJ4jD>{DvZ3*eTP1tJajo0;$MW1#2?gNQnL~GbX5KXYGaE9v>XH414b0J!4 zbYq4Jz5QlN^7B`U%9|wzW6vZNa-MBUvNJZ&)LeHsGzTMx>)d~F*$w4>i}C995`1)U z04?I&@&O#vv{?x{ZkjlBOb;jCj)l~b&V1msr5n*U(+OvWEkfEPzNYB0AN?Iek;O5; z&dcVDesAdZMR;kn0Rm4?R5%(!sn`-;_cb&+blN^g@a37)I;yT8$k)9Pm7jgIb5DP$ zK?h65x#D@(b@;_SFQ#9=A*OF>CMt7n?l4@rvIdQF+%SDk0M2o3janX${NX+UPwnSG z>fU;kU9!P&`PsPsYb;Wtr_=MBACdfR09{#j2}AC2Va7{i==YyQy&iHfk70R|&z+SgE85pmp`v>=trlc?gHFYW3TMYqn4 zz+edB{cJMc4DrWJ!AA5i`Zq@78?^wc*H&U^j}gfAw1e409%Ij#j^`N>@SW(69G(OI zrnMI)p{^)vw?*g|=gw>6W9pK$JNHj~3%RGSkoL9{>*8hHIffkL+yPq;tUkhX_p2x3 z<-!koXUO1z&5RyO7ApTh;LWXu?4PPIcjrm#exV&@^w7Bm5 z0yzU5QW=3mTZUpm#sL&w>%@W1oVgwj@8;ps!jmXi+=srXYLS`KQquLf3PJ5%VPi?qv?o$%f=tZcO!-_Oy4nE82eS!G5=u+}nFE z*2&nSrNIVaCiV!7nF$p|8&nVXM6HVz#=o8gDZckipBI7RA?|2fW&`80H8^tH8IzVM z2z!5)75XH9B97OjU{?WUJM^Qb*sj=G^Me%zOXJ(f(@43;y%0(kLoe|yy<1gIGXvk! z(J4ZX^aJp05B`TGjj zT<{bgeYs0m755v9ZC_$SzXw>8e2X1z9{~54_xAtIuv>Q6UX~7h?gfKW>TXFB z!}#E8bdy>}{lwN#aMvJmI+j717L}CuK9?$bDhZc3{i6Bh)ztp5jOyGgN&08Xf8GlR zxyRX>@Cx>B-8t5>&Wv)7=+o0W6YA|?OYSr0(MdJ4|CH}&_dkEyc9FE}K`uS;4WU`J zMWp!bDBUgcqaf~IdhnVt{n({PF8c~t_@IL<)s$<53wx2u>34{kEK31yzbR&TSE13J z2AW$@#TNDZ#@5~|X6wykXj;KRn*24HQm_1`;+?0daLjYsa6p01bC0duDV^F}|FmJT zbvR`u`%>YO)zmCAT9{NlPG~mVK%^CRaw+#SQjBGm?u~3mSP_egsbmM9h0;4k*Xw%YHXxPV8czThX@Z@k)VbPZ$((bpKgq#CgHG8yhvg8P1(r!iJxr6D< zJn@d`9u{u!pwK>BNdD(u`lykMoiRz!H%NoW(Xq6-c>-A_O#Dy3#p^dRG0i8e@9$eo z-poaKC)QN>VbpZt-TmF+?JR|N`~S`NbzED2&AoJv>*Bob>OnCJ6>-buIji*l%~E#! zhF;i5|stssZu3FS<@ zU0XQihoo@HT6v+#0IuV9wvCidHPD+(?rZYwENhqV#QFboe@9*GPf8Q|(*B8Z^foAf zR=Un7ul0WaX@gDPD|(y!nu=RqQD|d09g*@Ui8tP)QKE;aNgQjk`6aWR)Iy0;7wEya zLW)R@X94}f*aCwSOr@@#+ylQ;X}8ON>#Q*w0q)3&AW3nU7(8XK+rF|FeNM8+)*)LUfTr2Z}6(bf??60zrOFs!`4;)0Zt(EEB1;j}J1j$k~Ve6;7$_(S_dJx&jjK5Lg@x~Q}!9O8V^!<3HNNg=%EL`1l zop5fHhtT@-X}VMMlpI|Yh4;JK3il}Q7QW11FI+MDI%%6cr)2lHWZPtek|pc^j<0|2 zh0LIhbaI_BeOebomVQTQb@($HtyD|O-&znHB}u-TJ*mp!Yi`b>kDx&ZTC>*Q}xG2ewe}H&gzW|EJym z9L492rML;pN$Iy8DZLjH?vi~=(Nk(DOt}YTWG<&Eug%D-DhgR;M=*uwzj8NRK-kXT z2u_ox*sq$xX>WT9Q&*Y`D_?pc%#8b`Jl&0e;U(mfbA_VSuTa+UL)7GOfTrwzOnE+= zC~w4at5VmU!bjZU4-VNRfPGGU4(U^IdDJ3bp!mk zH`#^xxW%<>EQ5FARNZ9^FMJKtN6NI^vAeLSb*S)-V4yIee!B4BAXDMBMQXx2wJ5S$ zmq<; z=-+NanyMq{_M%ui`FIv$GsYrk*h=)f>rZn=_|rMXWZJS|DowW8OuBx)bgRx)bpHOC zV~k$&`RWZb7S`5{6B_8w=7+gz@hSHW~ESq?bj;@&@WCbR|re>kgqH(nF;F@8=;p>eN7%>S(idGo1 zItZjXTC`f7V7PPU#321q~6VdyRyf{ct3#bi^{FKEllxRfO+{^%2hRXO9Nc zRVdPS7xhP&QYWWdYftC1|1L2e7qEloU)@K;oo>;W_9G|}?nL^|kiY%k@mo85xE6BJ z85BEbpzmFscHuNR(fbY#%TYH}1>0I=Fx_ezWt$t2Ao3NOy%X9mA={<7nUJkFxh6(7G@idJDAaXZCd@hsWV6_bl4=qJZ@s635yuMswff z?W8TfkbI1Hkoc=j)Mw6kx-%`7d3;^R^cBO{(<_zO+mqw_mbb#&vK+m+f7rPTemHBf zoW%R~Cx0z<`sJocZlyg*YJMAX?YLi^rW7{Ml|xEpH#YmgVL@SIiXg7Vkz;YAaI$d+ zlrlMP-B$)99&01*`XOvz9>g{1)A7hT2&qw-IKZ*e35#B^Xs2QpUti8NSD%7b+A64+ zd%{QrD~!`j1WmFvdVe-&ph z23{7(n#5qCO&Y$A-HF?y0?6;TIqe$JhnlC(LVaQdqx~<~<>Q3ux23T3We-HzTjQ$? z*VQ)VanG_|)I3U+QmXrq>#r%ycKH@&Viw6*OcAr(5Xn>;4zW+g+^hDL6xHY{lef++ zQoYLg1n*v=y!;08HRB@^VU;U(3EP_{p?+EaP*UCF-^X;DnYZ)_;*B&#k>^)oRYiPuVBx zzlHlIvRQgNux8^Nyv)?aLFKQ^QYML&EzD$*3%I6b0@qO6UWx8mFCd0u)F2sil&qNs z_doh^)tY6C0jc>ONXin&r6cN&-(aA5xzX>UK=7$=+Rk=3)ufn2TPAh zpnLTrHd3Es@-)>&Wp{52gyN zM9co?xhs487H=Nk#KQ6@T%6G+7__%gka4_SplPQ_yH_S4*OBvoB^ToPkh%EA=gxoH z`~M!lSB#k<|!6ytb!Ef=3uqshZ$=)!Du zn()qsS_{X}y8I1f>DrHC%``~W+k}iw4I#hR1B>ONFk5S@Xl(SUUBy=HO@RK1C``3} z4Hbzq$SUx{FUN$x_f8P*4lkjM*sj}5^jXw>N&2(#Ju33Mk(hl15^|2i_(%;#*m`qq z0z1fVio+YO*F3S5?*~rIho#U9+F|P(fu zNhC&36S9A7O|kVi zMgN=qN?_jyEyA(}J$N5j3{pFToat{cL{o~azr?e}gS^;+n$xVr!3!gJU#eV*Ky#$X#goYCVtggooM^73HF_~pV+hv%%G+(2*ZxzMRx0rUEB#EQjWsI)iIK81^} z@3L%X%=2TaC$_uzVE74F{21)VO!ix`OP5oaPw!e**7r5bHGRs0`#xgn+w!!r7wC7U0)kF^EP$LWj}+?OdGS)Xiis7i>`RRWIv z7U0S^N67N`>Fp(R@p|xh3}?~Uw?T=cl ztU#;depvH$me$BS#HI#k(1(2-lOzYz*ml^P|1mEg0)Lgbsak!CI@G z@Nx>qs*9uO`eg&!YQuH%vblCyeJ&nhC%(Dvra$>MWX!d|UDp;vx#a)R<8#Ep!)nJ=k8GKfza$VM^NL@6D zwj}G2kNsepv{8nXE_}wU!ymEMvvUqAa^*PVPhqID4n-e-V*LvC&xsEJ#?P%`P0zBVv8#-~N{CNuH z*rbC92n|J^aulj&&`=7lziO9;qplNY*N3xe6KFnIlZi0|*}plpf7DNBdUfA}rJO)s*QZ$LmWS!~NBA5ZiOs~hxE$M1 zTDuC@xz~Wlx8~?BQ;zjz58$<`80VLs!;{~~QO9}LRugz`&P9ek6_-IySb)cgJ7H0D z8Mznsai8)Os2)f}anoH)=liYJ&oAJ~ywSJ+R+zi_qonOUj;<}k@y*7k2Y}@Z38T6{aS*FSZ|*UBqeQD@m$Q)g)s|#U}4FsLj6$xihCB!L``V z@8f>8Eu-l9=;gFT&4q$?W@2OP3yinv+*3{+(t`bg%}_E4f|bev1aJ&#-=TwP=gGt-t^7`#%1#paW{Hh_o(*nzS6<{!%HrpB?w~K+g>{qi6z|?{%dL z?vr+Hkp#(3?nCqbSC{kU%PQ=-qC$k- zNyuB=*>VBjPQ3hEcSrrJ-+aZ>urHWc{uV;l2YfeNffu3Q(9pFS6W4!6YxjL{8KKBA zOmdW~GJ+m>Pkb_=>a0eJ+yzJ*7X&HbL;S1d5qAoWU|o0TtV0;?mNeO ziqk5+o^*J0F4qmX2@|`E-215wY0VrnlkpRYcl;49R)7}2gAgx&k397U=|4`?Vo%G^M+tR`QGW~fERI!NrcKUqnl$9|Fq*?XBwx5Sz{Iv5 z`ufi(Twhu^_Hu8bOm+VTt4SW5)?;ZXBv;8HGo0P*it^$p#Zey@%D}GsVeBDwx z;Z3zP+P3jCJ;^S|jX(7m^Cu0bZy52Ape5~_*ojH(sH5XAyLT7 zrUY-*#VE;PDmk1RPtWEkkfDt@tvlF{WW>L~bj(qVIdJiB9UX1l{-#g&j^mxhx6oK2 z`}Z1`o$g7|X;!pI`yBSXufvs#RhXap=s)M4SUZwJB`4CjPYSfd{SfTWpW*({41c~3+qUA$=8lY_hUUQG=G->mn+#_a>KuCDdPa07(zEA=5hZzh$&7J9Sk%%F53yg=47(eQ=hh<9~Y4 zgh^j9{Zbo_bAQyB4j%~a{Wpi>7KLh=IDb!uV#Bi_cUu*)8WqezT^Bv;^4JN<>+JmX zC^mD#8qEE^2?r-RV&y!Jp_(^`5-g_CPB$G&dzC{e8)UiPiZ7dR!I(|(jbyV5U8vV{ zA$_@LN{i*?5jtxu7ELh1N2hKSf9I*F|5E>Sqb7B6%IBEQLfd2(HzkqDMIRKc-#gmh z@n>)=h9VdHkgrx}{QsZN|2ZC&QbEn(PCXru;R@L5-GY7=wOCg05GGuY!`W~dYNqPJ zF5(niuco5LA_qpNJy?3+3HEzhG|TCgKzCoPCAVJwv^&*`q93j&Z#QwuSslw75)4?= z_C)4%U46uvo|qYuYN0mG+NVjI^p{e;mnPkFG36S3 zY3N;j@b5So`YVw=o8FAKtHo)*=QRWr9ua-_E8r;JHif{*JQ|NjaXkW`hj?HW2<7r* zjOca@x~0iDbDrU6bva89|G|ouiQ)ANOX!)|qGd%b1fCp9+)ya2CaM_er2pN1!bwzig-Oy`*p`y;aA`L>~6~i zruFp-%fHu;SZ_nh?=^{>^Swo7d}oJ4Eay1YaVy;<`@59Mb!%(?sklw9)E+qbBPs=k+u!%F=b-_Moal% zwTdlskrT3S;hmiRj`4AO-T(SCstp8NA;8jXPc)y5g4UJ{#H<}hPvSn7u!<3?2c97D~$>oM_=EbTK@rf+w?qKDcWT(jKI z4AqjE`1NQO;k}W~yL(C0wvA@Rtm)EWwtW35w&8lG@3D#u=cWJexb8S=B*(#!V{(!d zqp&rdYh_DYYaGyr(yI?N|o|#S3t#`HJ#c!AMjWvVKxFqB>OWOTs})4;QL> zK>Ms3GM&0Yx0^b2-`?Z=l18k*rzx5fu5VW7F==VhHQiL($sC_;VDC?l#=Jj+W26#rs$}uASR|rqao|D3oW-oU@E%}zSb_RtmZM> z`=uua4mQI2>Sm@ErHPGi_E6pA733{zPFFoQ;afL8hF4@@o>3)!i#@`4zUEv1bvZ7@ zW`fN(MAViYEO6Qp_NaRV8~Qm*U|^dgm^3wBU^F+1=`1s0(rViUv&Q%2{A^DY{cynA z91RRvnZ~O3#fh%3<33pF+6ofo&U6d&n}t3lfp}8#LC`+!ouJ+EyI@C-C$u6euuc03 z{@fr;in+&HXX&!LaYcfpGJ8R;QJ$bm=@Q2HcE$iY>YLSkndQXv=01X>aFhGOq?Y!_ zQuT{Wtx^LDPUWn!>9auY@f%S*Q|-JeJ2Ehr>C_HpVm944rh{uKB`M=ps~w!f=VD&y z3iJ#UWA2yV2{!-uCb;OPNWyDBQMa}W&FoTv==>XKczq8?pF9!l&empHYo3ehb=|p% z$&a&Q;m7tPw{Q+KTg`bPTqobERfZm@o`8EdUyN1TjH{)Ja6CJO-E5B&G)_7#I(N!i z55b_=RzcE~F6`q7UpD?o29q4QiS3&7mEC>e&vIUt2~3te614e?vtDtQY(!QTdy!(n zB9l;q8K@t5C+ABd|yCzd`Yi2h##YPds*8XA9&v}e3%Mu~OWJLE(=hseV zqENvmu5)1aUnN-PWj!7jc_(UryiOMwoaFrPah( z_K05W+0)yi@;%nHvDF)6nDp3u^p2H-%6&JSwW~#XOsCJn$nYwG#s_sQd(y}Xhc3r3 zttL1h{8ztdO!P{@)(Z+q`q~FA?)|ZS@DwKelpxs2z1!VaN3(e=8(8@%YqmR4n)NyO zL7=y!T+sAtiD13OYk_s0Dk)^6FXWwger#g$ z9p+^e!VGm>*-^hv4g1RJv4V*|V+HP`6OyyIj?~?ad0qZGaED^V1 zrKk-?hGzs8%TI~=*rol2AVcGWKu2idKP>ui7iZvW4Z%-z@cXo`&syy`MX{bxDBwA zU2Lqz?$kALUr9%(hefh0+^+B(&ABXatt=G%4~dB= zm6k{1u|~nuVOK?;$q(tw$5~huvWk)=FmGvC9?bqH>xkA#Pz z8e|vfA<%aPbOx-$D&u;TTpvKTC&!59aM|{78ju}HbEHS$+E8T-Z8C;QU_UzaV*u63 z2h$|+{uoea2*2%1ptUQW^47=F=!7c12WZ5SDNnG>r3*Pn5@zX#z}zH^$20;+BP)zv zehcF|EnR6&gOK9H4XKE`Pk($8Oo4Yosp(f2iY!o~8O{1cD!NFOQiZIHJ!E%sKb5dp z#83YKhxvs#efuQh6u+Q>bD;VaE0d9uA@>LNLR!27=Gkq9pM)V*kD5;NZ^qE3&72>o zJeL9~hYlpTLrP$YSA9INU`!LnZBZbW{TNFXWoYHrL9{GyA|)PT)ZNyW($sg+*e&@u zH~AZ0X-%bF97pPSp*M6>$HLiYII6QbuON94N*#D?|1z7TN9%=^p;?>@Rm>oTeeixgQhRiY=w`>rs<7T66jxqEUTd?Gv3SHUK zmvSGBqDeWjJdZbt)<|Cx&0CFmX&4i-0<&K_zAjyBEuDS`-ZL7}HkA9O4-cSG3Sl&0S1Ora?Su9O`iM`TjIUETFY#(Fx>fg}$69t6 zFxwf;=RKjY@forPs%wV9JhdCbkEBpQ)_MB9yPP&X z-VJEZ#<W3A(#$!r`D!RDn;r!L1$eRru$a8`Nt;FUd zw#X`(Omh$WQ$Sl5bz5dj$I=&4W6u@T=qUr|{{7MQz`r=OmN&Ne^(OVDGq8D7x>%8~-u1<)Ra93YkJ@`gfy@lpcTpY zp>@cVx;`CAlUAj&i@B$n_^osNT^oes{Xgbd;41$(CbmRLPedAK* z*$c2r`^jDe8DQ?v2ds4eHbk~>#H{Kv2+JE;N&t!K>o>WIuNgR6ioX+j8XaVR%52$@ z9U1KA6wZg$orAT`FL(^76T{)u>my7J|FCgMQEYR}4HkS{6LuRIX7`In*wqv0Vy}WD zGfGjl;si{3%*HL_k2n&v74L>UWG@Z`ajjo1Y@R z_50(#y7#ot-fQo(&pCUq^{nS<-olRAUKhWQ%GJhW+);fD?sXKd!?vKCTp@Ea(#Efo zE=WwBgr!sa;_167m{l|Z=ifMD(k552y^CZ0(d}C>$`#Kur&05;O#34%^ti+_HvMKB zqi?gw(L9EE#tZvya$cvq_n5|#WN~}4gC<~a*fv-d9%9DF4zR4b66l;z4U6gKoHz6o z{0}U|iW^f+OW$9?4xx@2=Qgl2=Z~^OM}lGL5dtM86BHQgz+QKP*bYC9YMjw7!R$9N zSQ3Ac75=GWR}_1o-l>!=lqhD8&q?vU>tY<{yuPOIm*JL7K9j3V6`yM@af}9ReT1Uh zM)ccP0VcXBOtYUi)aM+4_mXZfk?qJ?OYNV_JnXNsKRb`HWd#=EbK9?3)ywa!Nb46X zsR?6ky`03ycdi@&i2(v?6AnU}on*vuvt==n?6~3ythzKA8Y_mN*sg}vfBC}JP5a0$ z#*D_TfpZY-x)SoqJbug9{_gzQHRSIFW~>s+8hm7+b<_c7^OD5lC(&IND&f6weg7Q1 z1m{z?4~H}7$f!`)L*<_oNLU?1_`G7a?#ltTD7Jf`JJ;+=?1AypGI+nCzxY_cv9|bIZh`&$nX*dL4ibSb&`jYm z39hqxw@?+koX??o(pEO|Tp^p)DU>lm3L49sS=}v57)G~p-D?@xy4WFh{abN;tID?F zOtzBvd*e81s2c^a^1y6X8A2MJfaT;y1ryyZ(hC#Wy zur0JiR)#s;*ZV`V&sY@J%%$IH8!7lK*T88r;8>4$NEvNKzgApiJvUXb!bC0f{54s; z_L{>cQil5-_Pb^j4oCJy#F`|!SaXARyYD3{z9+RW`G7BpSMYE>Q1Ge`WO+=z>uEhY z(lCI0FNtWe%4~}Hw1`%^x{`9T0p(q>r%!J)OldaLJo! zT9F^UxUiG{a{SKJts2<0-UcNyJg4-U6xnGi)7j8g4B4I!uW46c#N$H&t3AYhjDNj} z#uauaRo4-;pKB#ng&Cs3Z6*SHbnJij*W73M%OrU2d;u(ic^-!20Zh)Cg+1kuas0vw z#8kxM{)mf^%{+?9`AM8#wgIQQ@p$0*S{xb@iN5={<5{?Yho3`X}YcLJ%k;Aqtd|@h-|4(55+ zG1Bq|ibSI*W7QF|2cw187+ zsV*g>my;>ChlIrMk?(NgaU0C~2T-q!MCvzlCk=jiiRI?jvg_+B*tnQ7@jb!vO1V$Y z42erDAT48wz1jXySv_0U)WmwP^`#RBkC4OOFxtGP z7c3_@BOyr%vL`N}m*r_3z8eKY#T?k&l%stX+8p0053N6)v0;T0Hbih7&HY2j%S^xq)7tU951zC^2VP4Niuw3{6_W57&+Po9TAZp^>Y*l0hb3Lt@fmpTAgL9&; zfgO+CI&_M_TGL3JSfR+*&x#0E(?r&5MRF~)p)*d_;<;njKNCSEyMcuk*!5r|4$hmy zwQy3g+$tZ|H!3j1G#@)N9^>}UMy_2{%{)xvnXTQf$F757#Sl@7T{;&&uOENKIU=9`?JdXt*a^Z9LGs0r!X|<6zZjMdC_q9hLI@=ZY zzQd69-UewLdy}QEM;>G34rcVQ zhy1TW*tw?x+D z@8%#uS*b781rDXP^G8v6-7K~u&zVh&-ppd0KJojgkh(u^rjYZqNZ)fI-5Gt8f+~*C zyZ8*s^{b?MKNtG&)}Hi}m(ZXZ5%XW$L)^y6jD2YMoQjOV`#7*&gLdxHp|9yuw0=n% zWvZ;E@7<=7hoc79o{_}L9`iBrqpslZ^pnOsyhWAWu9N15N*YtpNMrUjFtwV_c zQ`;R#)0cCYEiq(%C5kcvP_>+eRR~FE*^8OydzAa?HIc{q!<1! zok0%`0_a=JG+JNLNTIGLXk^-1>a=?iy}BGgOKjZe$!rz6A8AhUN2k&KA0Oz_kjrE` z!b~{4o{8(*-g_eJeA<+zUa{tU+qz^Uu@sV}f!KFuFPuN9ah#h5%`@)$pK%^`!;eL- z>%=^C-53nJ2%$%{g=rH<{>!0AH|X`Bex*}y+OW!m`x2(mJ37a-GE-Pw=~Z^9ODF{Y zBWSAlzqnQVSj6ZX(v)}Rbj-?$UPq|Ym{|to^je*wXQ!|zjVa6_JcWHb-HAp!UBoC) zC0gb51YC8b5RrbZA!|1>bpFET4jnz$vFtmMS*w_a2y8GlB6&91>1&7 z(yD?y7%#1a!rnKiy>3Xo#wt@no)eAR;m!_+&lcA+Zq9nPz1W96UblfA76yxt?fEc- zIlDfmZf=K3{v+puUbdgAMi)?aSU4><9mLq_{%q-jjxqk<^_4lg9TS^2q2ITwxGu-{ zZ)^L(=4{9M^?zmh@mz@1>_~ESD4kcl++H;IlQdtot-E{w*O-n!5p)(JBY|6!&r zqcM3B-^08%hr?}8`t)=rX|E$%)+LP0x4P02JA3jr`oIpP8sbyGo9t=vByv*-p}U;_ zXvs<)I6nHwIxEWHh^CA%$mJPbKT%BudM&V!|AA_I&VkwNgu=<+m@(I;9{PAYJ+|xM zT3WHfj3(6UQ?g*eH8A(kb?qp+eR;Zgu0G#d3a8j~^u9cuV_9?|IZ_#4Yj-1kXeLU> zWnvTKSdKTfC_Vjv@6Sim%G${k?bQmK?Vbpz=z{~=0`yeQ(TJi9x_C2)G=?P6#CO58 zLwg5}w44CLZq^901T=5kQc?6RNLR~I?)Xk5|5BPpOIBcz$2PKEw1u8*h^5uvJ9vJ| z`@BKa!KLDRLsH2UukXEKRy9>jKl?CK>GKc;BRIFPVik_eI*Y|&yT#@1ImY3uMI63u zUj?<8Qq1YJ1FP;mgn^M1d7tS_`q%DYQR)b4TBSlQNe>Y__Xf{(&cwc*If!1Tj)=X# zSK3_JbtOLqCd@OFh#iZW7c&~cc>0;fr-s4y!xDt@_yZ^ z%d@p4sv1md)}O$Yl--cD9LDiCbD?*8tk|}j5+`t`Ujfha*^7R`OR4{BTlyX?P5Lob zxX>hpylD^FmuWV5)1U*7zltCuD=L1;`KTJY(H6sS8c?!{lBGwI#cU0p2OvYG%NlTY zNmtr3^*iSi{>lBJJIPCwM5jdYbm-wzwxx$OPL8kx{#-teg$8hahQ;tWG#kU8F-+fglO<+dWoidC#APP5#F7l>8yOZc1Ld`YvC_vJ zAFidtP-h)pHBCSZ*C*F^QJ_&#?_rwV6Oq{(xIdv2UL2i^T?5WSdvydn=1znA;(pL7 zvcR`vPf%?18unvl$tpyW!ZYsSy?q`g-(%wTd~D8&?>Q;%Ys@{p2Q=>hhWjpoo5fmm z56UK6ZHJ?sqW^pB)7Ej0bm?3u zJ?T4*Twc9J{I@Qss=vg_%nviQ9ux2;XDa&N(uA7LF0RkC3(BGdc#mqv>BMZ*KW#ve z2ahM|1j72kB&1B5Pc7#C=}F5&d|Pu7ZJTQ`Q6e7}%R2D*Y0)2;Fliz^t+XYp`I_WY z`GfWTTg0AO9bm8QX49mJ!>O#ZF)isEhVS0HP@d2OlO5$acBc!P<6Bsho;r%hOF}KX zm1$RVt{OiBx;VT$*)N`on@5dsv|bIf=60d+b9Sn)CQ1!lu&u@pPsS{V)m(F0F^OJxG(#;DO|5bk%BObrG zoeatJ2ItTIHjw&fc87O?zHGO}({V)g7al+DLs^$s(dnq?Fe*L?W+q9qd~Px2PKt0C z%-22hUcy(&kmNEVC{DtP+RTWq%?PIBu?M*REF+_hL{DEwLpRnP7o4l%-Xx3P(OnVx zcL-b+V({)`2;!#8iJxg7`r4zr_foEl*nyFKJTMJbH`lW7B!T6dxh~=`b@F>Ln2Zf0 zY4)}2m@=^|Nmq(U{@DrkV%IG;$?ZM6@bwwIH>uHyIBUB4fCP+E#G2;ziMw-?FS=F_3gkRwA-HG)fWXjB73XqsNX!RGtYG+Z*e* z2)4B{wB@WKT?y|>f95^G%ZL2k5oATdDhWIvC!Ia+dzH=j{gX}6a6|dytvIN~9jlFo zIMEu2DV^`L0M6O%eOCq}8^`0-*&`UpV^QXH{m?};1dhAmxW+POxqIF!Gs%0F0xh+HQ3?E$PJi@kp zIp>7c225WRh3zw1F@9Y&0=+p7D&3Ih^DX6m(Jd%wu4nz{o?)X@8oBS!61LU*v180n z4BZ{Yrn+oq(ppK(YECU)tb4+B>C`ECjw_zt(#Beka4dLG$Y^pSvuu#U#v(-+WOY2x z$A7XUy>II!b$Ny5S~;yOZYMdH5`&T3)T;O*k?Ut85-@}*;4TUz7BenzIq59sAP{+jom}LjQ=?oER;0X=RJJj;>2Lv#>#(5psDByY&1p8*(q&$|?H~ z8cX-=ZDI;btl8k1|A#AatSQ7-Jq4sYdSZ%h4x1kt z&U7Aq76lh;vGihMLx&l$jL;Znb=sD{chbatEmr8-$zkjf_~ajli+u(fzi46L8EwSZ z8M6X&?vu5NL0f!>uj<^|>!PO~*F~pO{6&EtPsHahzq*^fE*ijM=N5^7lihLVTHS@M zaOlA4tuxhUC6Z2TdDTvl*?>c$6qiWRAU5iQ5X>7Q^-rQ4oLsd0C%5BW)&WC(S~f zPKriZl(2OlI=CMy3U`Rh{3vl@NAu3HIl(X4B&*}B@WgLslyH}s&A-TWu?@dcq|x{9 zNNoEh#gelML2BM~exFlFedDV4hSvPNp@d~XQ=P{`f1H|Qyb($q^ zPp`~|-HV^YS|qzMa7lDZ9l-Yq;}DE$<3skyNkTJPvpk0PIT5#lb-AxjXNWz z!Q|J z={I92eo`(?)?bIUUfvL`S&7$1O|&pnML5*>hXxNPTGQK#Zm1j5uX8*%%b)B0>8wZc z(fOp5zmDp?4$@~)6?K+Lp=$xxDdS!Ym99#sw9H-9maa`cPqnGfM*})pG!C-8IbZ0B zKqN<4ky523{YvFracduCq)Rtd~Oxd;O7fYB#z&%|YGeY2Jsxw z>d^tRJ#&OgUvHy6-$SXeTOg%%JA%>^+fh``HCDGsA=aTMZ~)g#61z)x~HL-^XxhV+{WZf2eGNJ0B_fKrOgw~ zsE6Wk%Bnv>A8OCgjP$dlvvU}>^L)IO(qZU)<^U`=uEe{9*(g>DrQ~2glF$nwt6wXq z$Y&FUUC5--1zU&=ZcxI^AN1g?h_;oCr^&76WLJ8U^BkNdx2X*@;*LE=ADD~v8?5kY zh9o`9vn3O6figL^#4a)n{x(~1w|fRoa_nAmcmopd>;*emi$uj*m~9@<`N@_duiqpD zp1n+xZ=TXaoo13&<*|o{5g6>akM9@G;?=Gv=+we>_9o7w8PT>>s3k*}HNue{s*6se z7hwG7dh`!hBk8W%)V44le@(cK<(VuvPVY(vM&qfWa3N{!?nV)c6X>XmJ4NmskD!O9 zxRNjg)Av{5kggQnde;C)#NhT?p3nC81Zu0D(QW=p7(Ez_>>0LnBh8fR*Z7k}d@DBn zyoA?rXOK3$3vGVah>TPBxz?gFneSDl3qQJ0=4*bvb_}2^{;IU*p#^*!-4LQZ8NGx3 zFuuhd)@3U&iqGX>n?RgDvL0Q$>&Q5MKP_$hPM-HYXvCSRr0leqY}Q!fLf5Y7!Q<9y zjRSEwYcEc$4ukyTBM8(hfo$k`xY=sbW=9oz!Zl)gPAi3te>vFIY?$m=KwksGX#B~I zbpDGhrJHwP*q^*OgyMsod##fSGEWbnOY8d53k6S-neBv4$8_N#ghJkBI@R7-FP^ti zJ+CA9BgfI-$wXP(TI`;%0nOvKVf&+TkQqE5Za=JHYrTq!oiiv;Ba}Rwk})pe5RS#9 zKr2UrRC=4y17Y2Vkvs))CvXnOIZ5V1htydSl|4WX;u2;cYqhJc{ySJ z)OeJ0|H09M=~(SBkj^jt4wYBx^fF5yn}YhIlhhBkWMM2#|>mt$h zixT8Ut-_q=PB0ia8xwVgLy@nShwkhK=Skh^a9LN{Rs912JGMZfZV86yO%%U}v>#p& z*Kgb>7}77TvAJg-c#R}FKSq=Ef=tPA)>amnlgpH6WH9MH*7Q`~fx-r>)BC$0#OELD z&-vnX65&_pgDIc3;@y=YaNDkdhxa6rHdv14P1q{g$g4FT-V@R_V=&>?zhzn#99ttQ#w$`Vh5AF`2>aV#mu z8^7nwL`mo^nf7(X!BP`(|A#x4GUo^9nX&(AW?{xK`Hd88Ui`n9@C{AOCO4ny9`VD_ z587}^)Q95hiD=&8juYBk3%d3x8&bE2ZR}djzSna;@`G8the?=vFBAze;TSCGE^cF5 zeuq}6t>hRId}q_-6+FL$+pybi5MqnEW8tkFEPi+b5mPvah-cC|J(KI{Swt46l0HH{7iyHKK;K(al(=o!Zx z%ki*6rZ`a92h1e6PTE|FM+%>rTiPAAuZt?) z4xRwb@3GKM9E8qOr-=JKz~>#CwkHW1r#LpRb`>UmpDK)d<0ibgy-_$jW6%?%j z?)TIc=HIsvM*2GZi_t8J;Ml$aslry3WTEHkWFa8Qk03Xj?36c9p4${iT;qCMKLhdU zNPrObb*&J1bCaMvP(@gMwVQCgx|iT_YO@e)9U&A9juIx;H&MS_SwU^9)W7<*wr;}O zG@h3?s0aPoG?1j6mr=l_Eo8iT6BT%TB|)o}yt;g)l!Z!EB2uJA14*(@Unn@Z`w1T! z76=(01Gtc&vyhYNEUX><2VVtQS}OSyTUQ;XF;P3{_pc&a_C!^PyQnAlDk}+B6Z#5? zr}YJueU`$*s$ve?{XrKjWd7OHJ|9=4Yr*c7JYQ!uvw6|0==YQ`GAiXf`B##ufb-z1 z9@|E_9D6I(uRqz@3?}KR1F6tu0OgHUpgzxQXyV>4biD35j}NK~Qh8m4_{loL+jWyj zCwd84mxdBk%@ALo_O?P3M^Lct7+PF2jN=CE1yQq(uxK0;9O?z^y4(*hvw_wHziB~k z4?h^`t-QQL0|`ljx|2$NHE@6HACQ)r~A%vv~IqbuB6Z71X<_z5*3 z(}lA86NRBQ=SWTd2q`=|NI|n_3X^+J6OL8S6(Ya1vL&rQS-GH&#dFdqD~M~2?Ak|( zR+^M0YeI(~T66y4edM1WM|X!N(>0!}_`bBm&Iz{WP>ryFyt@ISos)#M=l2PH40a1v zOXJ|p_wV5bdr^8@n-+viP=76Rntw8na=IO(#*7pileG%+obN*E-ckg+2!!4vY5Am% zn9l#k!FFLPO`Cs!{OUK+m6SYUBV%Z&%`PfT;+k9`V^MK$EDqM>(Zb%jv_mnMZW=Ej z``l2H`nW-SAGO!xp*;sp-_?ZT@=?O;zk`MG=IYch;tNKMG9vk=?bPZRLknNT(Vk5` zC_cc7KHcs`Z(fG+80#XM(kqN^IXxEN=l2%K;tA(j^^#eLIKz&-ng7|jS4|Ld^PL2r zC}-i~S0};tSx;fq=CR^AshOjRk9Hltm;aT2Fv*(a9sKCQ+F7(v)pYw_$&ASy0&6BfN^!fM^V|K|U?Q*W74 z^Bf`5y1!7YVfpVk`n~@*XEhA|&;G-1zR<9g7bad-5yr1+`i~8^wf^MLYYU}|HqmCC z|MOAoUf!WA*KX(H3G11+Xrrj8bScYSG>bJfcM+9LA1iUIJJ;NvF3Y-qQe?RkUW&)C zz5JZ&x1s?{!^Qj2|BXkz&$$QM2PD*o*yH!+Q6}2>0d>4)0c~yYP&Io6Me<@Tx`lg`XjvO{FqtHYN?S3kOne^2hVo7Fu}WwZNC69r~%7Uk8AXLl?^(%O0*_4{zqO&MWgp&$^4)PMMDD)NcP6K2|--kBRG(ZSxnOZxj5uwJ5RYpQOaU zJ%4wMXZ$@~R{tcC1v*xX9+{}J2g{nob(|GV6n*U1k*}jDC1M_BkC=*bvgr4RQR4IM zau$or#cdzWW}gfdd2Vi$o2r*F?MSDI?Tiuh_Qe$}*9I1@{*oFcUo<(XrmCR*Ys% z?k4Q>>#?H4mo3@p-c#7;M_WW+qS8bnJp{^S)=+?wI+#2PzM zfmC-^ng5Nw)VMFQklY~}@?#AB;kw|rzs)1*1&6ubSPt0Qi#VBa4k=gnar|r|9IAK@ zL$?e(^T|N>Nxo#Dw3~7x5-7Z1GldP`N7|1rk^{&1y`FFo%0I4P_~CTw`5>8+w92V> zO$LTEW$@gTN+@@EKyt0;=}qDV+BkF))uaxkrV0^{uT{Y0!e#g^JdZkUFY(;>+U`T^ z)wa{>%FV>0PyKt|&bU2A{QXnE8a%k(jQ5-$qiOvOv_3iqwo#k>6T4AsNEaG0QklXz zzr$*dYk9NZl!BM_qLPD-bXk;!=6gwq^xuQBH8s?j+(0e`FX+{v+Z3`RnJ!D#(8kAm zaJbuE+?P#7tBet?ThDVZ+y+za8xaj{wWjV~BdCYVE*h?uLi78cqWR;4sH@j_a?9LL z{pyb+HL?jC{nRM*ND4+?h~```2e3%zHIhEOMW&)7rG|4o%L)~`YAZ|I1ZA?|G00y< zo#{g9cYF+2A+K}2s9JqA*+$urLAfp6H24I%{1KNlUf{#TN{rXKh0~#(sXX^7Om}b} zLmbzezIYMxiT5xt{Uv{e^`mBa3tIf$jx@(+ z<|a_5D?8{B=QR2;s~DRlieN-NX_;L&Qa6#K;DpzhJVuFQ&(z3wbvg~Z7Ef>YAEEs> z@)4bU4_|{)F*qO}3;Jf_SKLL|=a|#9WPJ*k8&1DRd2+N_Sr^~ACu_0^$}7p+lh#&XHY%!1kOE_=e!)sblCGJWaD<>_{&T@ zhzQ5{>H9G{E(Tk#WpGVcc?#6O1!L73)V(=b@Vn z*JvEBKvOfTp?|}WEZ=pf)UL*~*GQm0GtWTt;!b?|!1bFaH6Zs*6Mm;_QU0u7nC>c1 zKPEncjGrETA16!GY$ZtgL059A?M?NP=2Ug75L3S%$9dlj+}*Jk*2b}@Og)L=m61^U zcno_^$6%@AAvC`@iUEQ3cCWMx@<~$Lk)NQbN&4(spV9sv$I^1& zaLRepL}BNokvV4=*YX;IDWZkslC**Re|F#}pJ{ijnT+^4{8sUGy4SOYrVM=~ZUbB| zrhQz4)(^w5*JIGB(Tqj}45sq9Ikdjbla62GI$m+}Na99Ma@*-bwOX9Vg_p;#3RE3w zMfu=A%(agrOb(2IdqX;8l3RH`Of$YnmgCneuC-%#9a8Ii(6Am(v|CY|?&}F0Q|f@A z?$a^(oj?7oxkY{43P?RTk4)zsB5CD#as4CD4MDuL1Kfwp&@PVw6PNJ?69lYW(i#fm6`7Pr9xo60sYL^X+l+|G!!)=nVjz5+9qU7?YeMMFey(A6I~;Cz7K&|-%Z){%(=9%)FL8dy zOIU6#N0N2>=|$NF+MvE&JU+fZ)o{A23zB3BQ3amj^9FHT(UE9pdQ{tsPDf6qtCeY_ zt~Zthn@WmmzD`#fl4-`M{V+`^LjE2N3ipqo3g=X67+6Lr-*mCyroFhHu>$u&PvJU@ z*7q@UxE|8V2VvrXLD;`#5gsM%!tDF8*skx1$!04M{o@eU%P1kvUZ3;1ity8a5#^;H zrR7Vxrs07FK#-rf&aeOp3G>I<*weil85M7V%|2HrPz6fa%#HtnOKa<+0b{ug~2%2YK$` zQ^KdPVU*A~lAeu`rjp+(nD5$KY-eosY2<9+nouRW^f34T)@fbZlrc4b2y&%3zA;i8 z{@wbcN@^w|l^ux=EurpPmW%uCKFa|v&b#4f)G_{lTYA%(9pk8SpD#J&uO`p&C$ap) zRW#SPV%d!esF$39*Qvp1_z?n~liN^RvH^9UQ)%t+80yhKkqU>n(uF1SX!zXe^wF7f zygT!JdzsN-8u_$kYYhF+J4@fQx1r10ose?giP_S*Bsu#n9o{A@#BS?K<@bqJ2RCEQ zsSupc2}IB3ixJY-o-TN9q60yPsCmv2x~dsZwi?;AIWv+ZcTOYYz3b__`g7z@SEI2X zxenQzB}hq)!^t9V>?`|={NGU7DSZF&Aj!ok+o#Q%!<|~pN z=Qll|Eki}FMJVl32qF6!o~G8pXx2UW3C~eNPf_&dDwePNfH{_ zzJ4Cv%sZ3q)Jrh(=NctF+c^Jthd=k!fbF=t`XcsKwL&KNH7?DaOywnvdK8Z#|Ej6{ zS+o!pDhrYM;1-g+zu-ikF}--2M)KRD$o+LZy;I^i=L?lsDlZ|PC++rcE~vyE`x;nD z6u_$FJ$`A8p!?~ObUbG(#ZQjL7DrJE4U{1boCk8pdE^DW!rE9y#|*YnSob?{ zTqsK#OJ!&xw?#BF1ZodAA?Sh|o#|#k374<7~&@p@)G27O(W;&YjMkU3w}0iq}}OzxHdxMzj+wamV>k|Jny|( zn!d?(Cl8iOug*TEJD=-lxmq=T1k01}QO-@ZQenisKzLB;dFU(tgmgx*=W@&y6>}{8iJV&RR)bb^Tb)zIW*Y;f|pLUD& zweKbrn+y{s`}Yv)R!h?!+c|XbmMKl__m};WyUC9C>x$aBhiLG#eN-D-K|U&#RG@T+ zQbzruxY*k?BO#wu!=fliNsdf1eCYl2xfJs4Aaft{iz#QnW#PNI=AOz2)O8~AK_ayxd-d;w~rHazk48A%}NN*?=3hDR1(I%R1`YD?j-1TQ5Qz3 zb6u^_MY!iU6K&tsXh?h?vUk&@OOCrqYJMgOg$bmSeU*aof-0@5$Ufm1HMb;_-?L-? z#<_hyL~6cde-p~t_UC_CO!uBNG>%X=<7zY&nJ+PgfZnkQzkR*4&TrJ%IhVxA^U+crH*WXdFNf^=CF}k z%=rHHT^oDb^CpY$E{PS=b7_bAX1f10mI5j=sh4yHDH-Nbi|$G6==}n(zC8PnEq5on z(Btlm@|0InLW7jhdPiRH+#~zXciP_e;-2oPJ#8XBFK={j%Kz+2m*tJA|NCn4Rx2e> zpYycfK?|9keM53zU((mk*0iy|1En#J?Z~@8=U;xMf>n3PXU%gKySZLG-}2|UQrEst z)$xvE?EubmThcy!gsb=Sp}U>ou-0vML}nd zflz(O8Xw;}Kr1=G_w^0?Bt?vIBRo`EtB20q5-?)SQ z)A&AR8Q0uibdjweqX?76W;S+V1ljaiL%*~9C_(=)c3Ko;gvB+;tx=}OT74+>lobVV zUxr4T1Q~voCS{dw)DkLBWn2^bDbK5XZhwg_I8(;1)G6YLbUD+RBnPKEcUaXz6D(WT z8DFg$Y18vZ74)KfK9!^tuwHU9*z2Z>8pUv0Gj|R} z-7}zqANN_8Puc9@w0gE9GL$S<#^Ucw0y{91yO;iC zp3kzVe;wzflTZA2ecNqCx+$5yHl=O8E9m!_%Vg+yg<1kDIT!F=Y`=R9l?yK6PnG09 zyZ4?|;myS&=GxX>>B0U)>HIz>NS@Z=&mF%efGS$9&~f|! zyDt=fbRu(hiWKuI=#Y!9urPNp*Q=OHuP=Gfh8PoAxOC{N`MUfT&-ZMLNjQ--5aHXl zVTR8$xESK62y%@8+)M%u%51Bpvf&0G((9FwGjD;gcHsK9^W5-Nc@NoPHtpYhX`f>a1219_W>Iic zF@3zKD9F0}rq=r^g5%TsShDpHw0Cr)Mte2ty?iOD+^?jn!tSsVuCbU+!=YqZfzhju zp=3}!sw?)vW55|$r=(zGiX?ry_a4T(JDyjQmq=rLcW>m3l|f9)OLnhu1ca(OQunE& zm3O6u(V3cPwbH_%YyXFpUKG^Pw{{!5?CJ}L!d5o0xSlOEHbu^l=}?mN!g{scXxY&n zC%Wn(;^cH>Ep29dZc1V2;cnP}@;%cF))M!%;!`zSHd+PRXKL6ChZ|gT@-=)0@@IAe z$BX$t#r<146w;uDIWj%P<#%g6fZ64Bq&lXWCOM3S)!mLc@51>8QojD?`K$x}T^H{k zj{h0YiJqL#qeKefGYs+j*JpO@jTM4Nj>3FH{3}=gH!8Ui=ylH|K-+H2xGKZLXqkdM#W|bnLzS3UqjEd?}Jl zxTdaB5;YIGOBX(Ke@Cnr<*!`HHCZkGwd)@^gzAO_5w%{Ruen!P+c}1!p9*lhmrD`JqMV8AMl3$F2I}L>4j^(&DG7+;4QgQav9U5iP zN!a}5$G@>{udn;F4h(qv_gfnEl*4t_c4(UX=Ux1!Xr}P6Usta0H&R%v$gy>IR7n4> z6`3h?NAh2P%-J@I`{plD#r%&HQ@M$bsJ^8g`CWv;)jGnj3#Tbf=OLXgt0V17Z}R3j zQi`dY|6@aY+Y+l*qb+U==k6awqgH9rJg-r-{_aew>Uo$NYtGT^nDJOxF&2uM1K@gC zN0>QifZ%o6R#dY|=M8NQUd1=;ozNc5+=AQ~q>G{49UKpCM-Elc_QHCrwZJ zLfU)vg|T&4aDK)G%&htgmz0iYf4kjvu5)Rl({j!$Y)aR3rS(i(?P;d z8F^v0b1XA5jbZ9zLRjReMHISUk3=D}>DRigh>(dxnNk!suU4a%)lzh<$IE~HZntyD z-j0}B`}eP!7DC=#b)j@nUt!%RNvtze#r&=d{!@O1WDqG-X3}rBXmU8##CqhJW2%Z2 zT5g!p!QoO=>M%%bbIsblsA)`x-lc8i@nsC9@I3Lmxf5v<*BZ-t9887j5zIIwlqtz& zv!31`QKR)9kNN)Y;+moaY&0*x~M& z*EI7#{TjGVQmA*;7VJ#>2r3tMP<&}RRo6cKH~xLo_Od3IG(4!^2B;^)fBRcZb&;ci zo^K&~!TrHw3pl3o(tqld3OPwKKU1hWE|#iqbmRQ711Y_aEKMKOAM@^6Lbu!uy=p?q zzUeqMwT08u2hmLN@J{xjVgs`oe~L<68^|U&mCVDYz&OVarW2Myi*u^v$yveA*cXK_ z-2Sx}@vVXyr(Gn!{Uvn&)GDYbZ^5#_9Z(L_r@1X?9mcVW&8YqoPfu>9(4!M(w9sS(8F_W4wKiI` zC^8K(Gtc9qNg`&b5Dn$L+r2j$kZb%>%3B*u{z@b1$Ds+NV5UK-Wy;iF`U&g+fLn@>cvvDuRN>IUMzh^rSr}+{lq}FFA+%Iy?`>7Ea9ovJtE;gfD{}Xcn*+bW&fq3`niPX2@Z6T~OYkYK z3CUBRqMLR=L-!jV1RO!@E-wVm0@px$kEqo*v0nBt8gqF5>XT3?Ta9Mxc+BRBXFuk} zv4zKNr!et_4(yjk4W@HVntXq6oJxI#gHpdRQtJ$&f7-IthTiOxMXR7MA-`Ifl-QYZPimSFS^$KU|U5$v&7ez z*-X7VoTE7h&(vaZwQe?R7$o=a{j}SAElzE5MDH1r>|S9F{H$ZSty0nC917DaPfR&; z6%Q41kZO?FL+;rl{7yNFvYNf<@jU_khpoU%*`2^tPX5?)Jdf?GGWT8;uoT~fNn%M{ zulXz5SKUPGnBy=k9m9SdR$`_P+TnYf$M0`PK|Fd1YG&N2=6`}4S3gA}VXdIj+cRdzYm zh^dHI!~VPhTUKSolF}-%x*!6zN0!6hI0%wmXV9=W88TzzaX#}5zDjX!z)M-^n(oaW zDF1?^^J|>GJ&I{niZc7{+XZ81g|&Q-FaQI4zJ}KJWVi;F9xqv;dLfoz`nVh@27kX+j3+$D+-WeE3CfZtz;8^Ps_sG8{46^;s6&T zlVaIx?+LG?u^k(o*I~A+4@#YQTw?1?cI3!FHeBKzHk*~8vg{#7?R+xH|wVNo|esQW{ zN0t^dqwQL-QQm<;Y@VRqf8TZ0@4m9T{kI@%nKZ{ZgaKwfZGR0ivHiJbf$hxst@u<) z%>Aq{+aVo}^l2G*k&yu{1>qiBVrUUt9>V?ExGKSXK74r`ljPX)=Y!9)N*Q77rn-+L zYq3hfZGC@SACe7`y0?hUkz}#fa%{u;UR>)?j-^|S;rs(a9AtRr5jIFX5TBKIBdFpm zRLJ0P)?8kn`i3HgzJ5xk<4j$|D3e%i?%T z`B@6snHr9F+s+|hM3i+qHe*SQGLtsh&gy@rv9;UF*=2s;7_Ap!Mp<8Rc#{~L_Tm~| z85pr%eLK*>W841T(yW-Om__{;mTuX?7JnIvGM6O8FRaGKixaT_VHlPTJ%(Poy6oB{ z&c8D@j7^@kgXdLcvlri@ z3!hHG`v;s;?r6TC@8;UFw)s5&?ys%tL&CTQ#}|C_=)*d$>9Gfo<=L^~P z!~WW_$Uk@r=_`25n_uhatXj-S$rc>{{r|@$H>9@BLiD0&4DA}vju=ETb+;7u`AHqT z4}Rl#QEgUm=?=z(7h=50512b#Lc;LJs3Rk`M&~m&Jdj||AGXE3S82%8^pl96HAzXr4B!H(?g zsAKHWPdPSwogoulqs59^y`k7E0v78|px3%N0-Jx3&R`bz^^rP#GP;jx@?6!GEJcrN z@z!z;)M%dfwBa0DGe2T~zhyZ5dMm!piiYgn&Ddx(2fyK*s7)%DVJ{ zDstvivgRrBl1WCuT3y!r*bR()O-!Pt5>WPH&ty_5;krAGGu}#)v+v=xe-gIo@f?`A z(-ar_mZkZ1e)3zSUhc(u?GHuM&H|Jy8vrHy0Kqlw zr!~+eoj-I=(w^ofSW@B8qvUSM8hrjv-(?D@1S{%lRY?Z`abQpT>|4_C%~lI4w!erv-~|&|#N)dTcin zJ2xG{jFUn>{T-KwL&PAQe#|MNcVTf<&iChMxeUTA8Vc3gd{|iK(#mZUXzhp{925En>KDY3sJ4_HRmdf?YgyDIy_j?iYse%}3I_3}2$mWT z@ytocdXbOh;dAL~TM`vpB~fhiV0<^YPx}{fzMPNw^z(Z>dGYww@cG^_`w_#rO$;G= zSDpI`2azx<3F*qwq_jMoHtENcZ7+2}zkctK!_CN}WY}IuWheV_zG6?B^X(_uS#P6y z?ysor3W4Odaa_->fh7CCBjc~{Y3r#Px?m=OzV<_~w67fIJ`m@;kS}P6lqD)%qtRvT ziPG;MNcmj|4R;MCH=C2>W_O-!)_73mmJc-HbROk?tE1p^U9dTojsY(@e&_OT7`&+v zw5g$OFxL3?El@-IM5+vDjDEc)g$PfSq)kDm{9Htj`H7A;Ikw&XAU^Y$T;fPQ%uu;OUrqN@Xz^)U zb8t1K^!!dur_Pg=(g{kBpT=V#B081)Jt?ib^p!)!cjF-5eI6|NOs6WHeVpxqSz zWHOA;j=cXMHJdPZ!yW!iZG`@XVA%H@kNkdC zIAtS)$&NQ^__JKP8u@`fDvF4lc8{l;*+Z#v;BZn{F_6}TB-4S0TeRFNn+}x5AgWFm z30&i}Bl9CQs>`9op^Ti`7hz0I6>aergW9!O_@X)l6P`|h*Ouex(PuqY47`j=`!s5s zdYP7`?IqtM7ifOJ8QM3h;wk!F0ll?*M)Ew5@3>w%cAa#DyW>8%>r5rX+3ErtnFN>N zSJ{59i*+7@Ph5w}Zg;f%>%lls6;0(kY4Vs#in*>2-6fhhuvi|+E)jStmWLzzr-1HE zXJxHF5&2Suxo(N031>1WMP~{P|Gb=zJPD+YTo+~RTMLX@8jGx2UB;!=>0i zWIdvSq!&D;GqZ0H8{mnvtA0`Qyi*imD1zk0SLs|Y5j@M0qARmj&>4@-(ctHt`EmYU{CT%h$h~RM6;v zT7j){q}ph8&!x1Vk7@5)hw#gK9CY-5kV~E}E^D^YOD|Idj#)v!E{+$JKgx!|Y1vBr z)RMuEL$Wxw{Tx+H&mp&Y6*S=BQPPP>B$uvf(0;iAZ+`c{#K3miw*DL~ElHs7wzDy| z)ByVP2O;620)nk2(ciL#;%(iKRoqG^yjP%Mn>(3IT`9O$%*{Nq%DhW!?_U!cKa#hp z_!fDYzN0;2>agK@Ad|=aBCj7Q7#lhSKQ`y$_XlJ8cvh&%a&ABmkXb!_dM1l7lQYzO zUkb))`84nNB>cYVh&k)!@FK(YL}#8ortSqvw7ek}^Xho6Xh0`5>a^0_som6P*)4iBFb?bYnn8N4 z6GYy!1*rpcCTZD21X&e6#W(bTHgPK~dW(Vwr) z=Z^_xuLoJ6@G6i0hjb(Bbt27Po<*lm-W1dk{B#fOay8kWC1F?<_z<%$j2EmowQlkF zcj-XEv7ce4h+D`WSSoOv$t06ySx@9!Tf=JGB8X{M3))%ryoQqevvFGQI9)sXfixM% zPfzkE!+}ktsMS|cCW-s%XU|P%mq&f1t!`T}(ueC!p4d;D_T`| z2TY&Fac!W}g7a*z%HZmuH`IDqmX*#IL3Hy$sP6c{wWa$b%Z=X~{GAl`K8YemeiSuL zs}kv%%eU&yPjI4G4EFn5Kg zloEA~=I;I^uaSNmBFA>FzhWTV&-pw=5sC5=uv@bT!|JB+j1OzU7^%qz z!O-53ojZJlEt?|Dkv+IBlJ4zPz}m~XSYg;iiGyXK7TjM@{zudnTJLp(vR)r1k&GME zy<~;JPU`1tNX)D^EXQ(PvIpBJQP&jx;`Lx*XpeVY8?jn6o7`G*>DG%!fIF76oH)e0c=6CJTj(Lxi8?j`#`361P_lzo%6AsVi9CXj$>un7xe&h^D){qiKLy+8(3FsqRKMXSok;mX=NoHj{DkYYc5)sC zwg|cVbXwXF*D1tQ|Mf>(-#vo(^*eNTLn94dm_fnU|4^^7-)TkAD6o#%f;O41C0vkd zz}azTf^i5QnSvS(~_h{vNx>q59}G`ZyzlCfybPw@g`mF*Ggr z(YRf{f_nd!?Y_Sh*YEdbrHz%eu}%i#IsVq&a3A&SI!8)#N@<=$4(X<43(EgV4?z9) zD2&eO!#1Y#wdQ_DJm~UKNG39XO+70*j8Ud4JktZqg7qF z?~?mRf`+`)A>-M>6k{Ar8CN0%dW|#Wo?+8nCAKFXmhu*$j1Gy5^d^ML#VYX zqK~h`aj8Sp5Vno(ybPe<#uDt76xT@EkjOT?+D|JFFBP=&$ixk#5?4S=9lGhQjWk;@ zb}+lwF7%22_WAFhWUCFUdohDGKKEjG4eKzG<-u`48`5r%rKqL=n(mawHTJer&n8j$ zUVcupi2<}^#D2B%yo3&^9Wipu$gu%#8T3=b)>^{ z2S4oF$P#~VVriQfv7;~VA%A*1j1E14==MY8b}O519DGiqixTLp&s93|AcLMHN(uVo zZ(&9`&$xD~oHSE?F2`~&EAZTRSF$>%Oa{H5@x8XkSRDNLKkZ*u?v9SE$&mau8T0kx zsNuOJ#O8(2Zo4h8e-wiQ1{L_el(CK-16iK*V;BY%(xVmYsKw$8X$*Q{6FoGP)W6UD z*Pa93^hTs>EYg%}Vg09zb23S@P0Qb4l7cB~mK4J#8jmG>ZGT{X0aO!<5Fj#u4R?5J z^K;T8n?+A0sdt$;_L%j=(}Hm*UHXOErc1#$<^h#Rg`n!J7dj0uL8I_ENo)$EGvj`c z{-JW4bmwjxLw`m;!aUjaj%h5v@iL3u_ehY_c7Mhk%2{-ZRKJ~~`!cTt?R>jTM1;01 z5FBeSvLydiqU1GFm_v7FPC4lgyH2VDQ^;d|Pu$RpC&OdQD5rcoX~$X!&Usocyf-d= zzGHK#PK+A4R?ZNUm9Xd?1H0dr_}(-Imw8;Ztj{!ZsPm-tQc2{o^8v(Wd_mP2d3H-V zl5Abh(qpd%N_N|Xvtqv3%KgCrD<2A3*rUhwZ58xeJ&8`ml+ldN=M*;Hkh;=qY&^G1 z(d8pEDdqZ5l5}54Pmb@TSHGXrkBUoVJ#GzLuG%8kCLME&o?}<24tuhMa}rIkr?2JT zZSqnxZRl7ZxK?J7e4{ihblPci=XdfbdoQ#}$SzVz=Jf>#^Bi~oOr_k#k}N7!oB3%t zu$4FSu`=~0^8eIebx5e-9GA>w^2xc&^?};(Ra5AH{=Fv}L;I3ppH3__(PADS{$R)~ zMp2duG`VX%Jvvh%I9L8lE@i^o_S%bd+crATdkDT1i(<(JDa1&dqGMZsSpHbX z^v3bHe@7@QpXS7N6)a;DOBS%$Li#8{ws{gV z)^nWK-jhs(l9^sVVXoH`W1hdz-vvY7$iS6z7J6|#i;C)TtnGIvURUj46UC>nUUye9 z@hz<=e5S`NoBY`z-Afo$m4Vt~`y&Pf2IiIR}{YF0M19(?Bw# zL|}a?52i-~(9Si&)lFXDp?xc&#lB)1&%Iy0Xg4!y$z)fzd9wD;Q`wpID+TYTf1Z1r z9@f*Nx6za*QBGU9_JG!;1ne4~hwm&O#p!sURVtVXRFyN?{hjL2i?y-v7yUmKZCcLKy= zbZaeaf|g>hbTVpO*8Rs8{zDHUg7cj|J6MUGUE*A4Oq4lR)Y72!#gy)MoQn5YLau)( zwtNc1p92YqSrUr!Bui+ObkUGV8OT?u!!c$E%dncl{!Gwew|tY~6kEl0%edCVhA@1K zh{f`sYdOAU^M9^;=AyS$8ML#6w?9ytrTP^kNIxBaB$Nko`2zwf_G|&-w=Qe zVT)O_-2wL8>nK}uwh8x6i8HBm1%Vy3Zj{j0AB}=%!9R0u_p5l!laIrTnh-P_IHCU3 zFch5F4$sjA|F&UV`+GdRq|J5~)gyg-tl)domtky3HC`@m|41{`@l>rn3ZWON!c z$yKK*(m$V;M-`HJT3@bz`WkC*-^V5o_kZg=JVG5SG)F)t`wyioFoo=Dj)^dm$G-5s ztdMI@Px~m$5&qjwmq`ne6y^HwxfP!tVeQ3MWWSMN0c}n6E8ZEyb-d85B+s?Ggt5^z zW4PAy+;STDLrieq+TSJ-HNR@ zdoxZe%6|TEXXv+!jX1dh*V_ugc3g#o%{O}F%=vB&rr@dhASimz#hZD;XJLjU!<8!& zQQKb?hkJw~IkFK|$#R&ls*GuU$3joyDqP>GGmrhY?6oQ95f=T%xsSC3*EPT6$o;*s z*#9l8cdyP6tlg9etJoCGR4he@nlQFFS-u9JTsZGqB-g4NmeJdRn8Js0L-w~{gP&d)^XfH=5vpKoo< z9!!;Agr}!Y;C_A+W-T{juVY`+;a_H0H2Qznn2W{O;Ci2IlKZ3Y=aG1yF%5%KA7W{j z(04v|O9FaZ8KnN?VR_dj&butMPva4uuiVs$o;)6z?(PisD}cuZM`HHIXgr&jj_(IT z@ce)uf`6RE`h)rS?Gh{)+Zzr%?#k!V|Nj$hHA8Br9(Fk%#-!2TFPQydQaW?%swcPi?m6&Ci zFDC|<<5uu`Z-}ogHkeyC882j)WB$Iu2%b0{{q_eTKjaBTa-P)Z&xaspp(qC&Xb6k|DZ$OW#omjHHH~hDZ!r+@bF!9tax^31*wOfUJ68jH(pod)q z_oPYaxO?)uS3Sv=UzYSZbKc!I%-LcDPD{P!j(k7j1N;sg302Yk{d-h|q<}FMd z-AoHTjc{LdIjqinre#%ncsbq@z8+Ed*1G|(O@(u8dc}IUJ`UrW$*LGtt%P+`0R5mo zXh;r3Yeo!uOw)mi;cUnp+#;~Id*(grsMkX6(K^~y)DKw0brr_xz-84$lroNEsSd%6 z+_xy1I+(SDgyZyw_b7U7C9s3{^p})i_*3w_3Y@h>P+!b|$MiD_XjgNDYLPC|y9aSU zWHq+PbA3F{|Ma)-N58AHuLgtJ*qVVXMK=?2G8vFJPlT!*=Sp8@BPern+D@F*YJ!SG$6ddmh72;|CUT@nB>b2_jK&prmDdFFQ1!)Ru< zaW_N2dM0O9!>O2U{V{{%dw zjisiqV{XbTSi|aEcJS4Mg z8joTNv0wBRMz7_#fj(JyeDM(?#)cx;bQKOS@kgmv3|qN2hs}4;unqTJhcSU!Sih+V z4-$GXbKiG3_eze*9p%31?#*nT;#{6H zOXQjHTqCA=(VLAoox;KuyW#oMn{Dvj&boh(!ey?#7qVg}Hs4=@am%Y=&V*RCefwWw z_^w9G?l}cx?j+&OzP&hCsL50VT-o+(zN|zh4|mVK<7 z+W~eXcnGW9-HUBg?#p}(-C2FhJa}E6fH|8s&pe-adD|jkOD)cZ_Sh>|KZw z-7yC;1^@%;275^pp^*?2EJsJu>l(hkvTGYy;zm%srh4`_OPM#eN(=He2~^uwZ< zQKb&7SIFaph8NE5R>o2d?z^A;iNJtfFfP*O+nGzTYM(F`d8S|lE4TedSx=P_6cGpm zoo;Fwr-{1-WANkQI}B-1X5mZ2+3Whjg6HW8KdyNz(FIe^-B#k`#LS;N;;8FUWZrLs z^9*yia{q+c?!v32q4-{Ah(D_W1?TzNKBazYOHgJY=grGSC@aXJFthilurQ2y_m_?KBh2uH_-Hc{>;__8vJu~tF+``IJ&rnW<=n}Ka6?f6eFC@+ukCz{m+wI5 zGI=&ZNsq-YbHp%3UoOKe8eQ3dSCw?(S`R5U+W zD7aSrg**7-{)&chf2z#Z55M9#chMeQHpqJu>+@b4W&uVpyE~EVHU6NOb0hHaEZ57( z_d>S2EVlFcKYfEFc6bd&(xrpgP-TMT8^JJJ6N}P0ujtD2mDu1Nfs;yeSV*W78(6!9 zt^8g_d9C%dKJ_|j)e_XTHF2d<3q7+^$arNgc{iUX>+A=V(Z+RKzI>qOx2>$vsGB9e zdcZ{Y_QYiIae{XLvwr*MQ})RmXv}r__czB&=`cwY^3myaiz7PHh-**CK`m%B!~+yq z<1KZz%DkUoZvRt<+NbsGU|J8hN3#`{wQF(z%LRm__F^xms-R!Dp|njORaiRm{>pdL z=PPn}w(J>O2sKZao(K|4;q@|I^c74$cY&To2eB z$(d(q%MR|7FUg_fFHBhfq%mxn%Ygs1&AcI>+}~di)R&=WjQlg^c;?`ayNV{brQb}K14+do%9manP#dqQg@&g&-g*!o(;PYQzTi$r{% zEX5Yt-GOmuJ<7Wu^0nhzCb?16cFCGqn8xE4FSgEsQ$0h4tqcYk3As-G+Hr2fb8639 zDf;(#LfK7a$p=4}qswDfdt1(S$rd55Y++W!cJE!Pld&ubbmBdy_WSy;DK+bw#LU_e1K~)lfOG83v|(5uRz!xyE!c)pr06 zxJRMQrwXscMA-5RqJsM*^{y$R)(`GG_k7vE^Yr9B4QL)xgmc#)GHLlq+1^6T*RA0W zsP+}+5wh+Ng7PKBhq&HZDrRuMyQV_{Q_qXxvX>@8iw9%I0VlZ4T#Et=ag25DgX7P* zrtYWtSUu?grhM{%^U>+Bt{H|?R`bv=+mZDjs?Rz;Z)EHGb|Y`}EzB!=k3rM*5Eo>D zLAKNI`t3!W;d~LR*U7`^F~{Hx>x|yYz0n7+ z7L01am7-R4{kf$;h$+`__}* z;RxE!*B~7g>eMj=)Y#c&bAEjarML-s0B_icU@zy>8QLw2xbp4nbnjkF>6Rty<7;wTgB95hQaxs$K~zkiB>Wbd*Lz z3GFrxTS7@YO`pcx2qZtp4;<%KO3k&&B)+PNCeI9`!=HE4ZmxHK&bEN26;GqMLuO>M zOp{bse55rC#$vLW1!6+%;UhhT#~>`xaKW4=yLwa5aBW)jx{F?nQ-$nze(tjk0^79} zDy;HH-K{89BFB1DD{lxYwC~u8KQcF2Cp9siR7(mmyL#xdauiFHQiinJMa}w0A+FA7P z7%loSiH^1LvV5%OTb(1;oJ(;Aq9~P4p%)V$*qCmv5{&Osy{U9BBVADLY)L)+l9EHQ zeKk#2vZNN(&oG3vaYS*$R<1MZTd%6x^&rQO)Ic4PT znosBAN=RkLEqW`TO%KPM5Y&6mwb8~T?x4-yt^%85oi1d0(SU}H8ZG$EvZ60_Su5cY z$7@Re7TVpwQy$*AXX()HT{cxCx7mD*vZC?xJ(vIa-PZ#QcYY!O4}s`b^DAo#ZOK7CfZ`m{0gIq&HDXO;B=J;-67`EpUnM$Llj44|V3$*Por%h=A0b3f-fbn)(lVwrvRlJE9RcK*)>XnstGujj$)}1T(?8rJnx6 z`(vFce0F$3GV~@zW~O7_g<4wRm5t(Ah0rY#rJ$nzw4yeiZa9}?VY?b*t{d5{PpT}= z>J1*Z?P48;X0+SKiw>2!(cqkX91K#0Cg00VJvc{D|97)QuF=?nm&$iA^wf3qJ-roo z4)1~#=kfbHumAaLo^-%QqMr%<($FVABaZ*%zunz22{-q7qnLhj!}r9LQm zex2%r-cpRL1QJ%Q=Uhsm7?Gchr`f{$CH4x>ua`JaTlH$`@>D+*dyg0NdsW?Px)U2u z{<9TPo%n@@R0;7YU(e?wvwQ&E*e*&}ZI;n(31>kYukJVsmO*QqHk`r}1RC z1T}xrq#r9C$bWw)rN-Fdeb#w6oHizn1ItKG@hk=23#W6H=jZ`n6S7Nl1Z_xb51_Zd zJ!q+Ex6ObTp7iFJ@Lu&=xeg2SD#+Dr6d9~2q^8q8bQ&D*H25uzmiPo8r!}Y=bq-^v zJ+%4EoN2qX3oZN-E~xWUyPUvIf9+rr<3JzcI!S&e*MiBqjX0U|&|!n&F%?s*3K<(16J$M~($C#)FyrAk}yTGZ#*v_tqSOaYK(@)caD! zix6`7;70H75*?Edp|Zp*I(hmiHccAcIbQ_44Fnxj2OM=-@gAX=bs~q3o=b)X3p{2_v3~0z$}YGH|J98P@(2s zzxG4a?^rYa!FrspEwC}XZcc&6)hY0dI_|x=K#zhJW6lIaWVUcVkq<`j3gbz#-z;cVm4 z8?-6Z1$mbj!Tod?I@EB=oNKYp*(T^uW_S-KRUX2or9WeXvi1n> zQ@8#_R9UMi_`PZ37PLLqVojbpP$Ll}-J6F6r_SSYd|&4Ic@y&=nuuNg$>_8dXEe$i zh4l(JA8n1RN7b43wPDQRWdxI&a27qL@4=r1vh4OVU$&*4=Q$3FW3>lEal3=(Quu4J z!o`+wFSWohEq~yN^PG~w92D;NeF2m61= z#p1OnbBgs8{C{2L7ji_^F|LH+(+VGKv%iSDd1a_qI|=2`SBPsSCTior+&2o}MQ69a zgqndEGW_XkWhkD_Q)Hq7m0S!2==HlpnvXewVord&PtUg^*FsD7doyYG>&w=@=U9u{4B zJ$7!oGt;t5#Go7JaIK#z8!h4q-<5n_U&D2wJa*x%ML7~apNHkCn>cd(Hhgx=u#8oy zY@g*-!8m_xu|%VvI=*IjL2l%C5}&VzN9UEehEq0TA0x<7F(( z*Pp4IRx>i!!8bO6^KPdh@@ELzk7Xbs&yLkvUMI=WPU;v~K5tjd)f-Y^RBl?Q?e!CxAd3F_(G00#> z_j_XAYdti3@VJ=c23Tz>z})(4xMN!egTxHj@;uZbJzA;jj|t?C)X;`yT~xhyFfO!D z#}1D1aQngcA8ilej>$a3b+B&sS)tXYb`;(Xtks=AS^_ z+N0R~#TBVCYDoSV&fCuSA@`4f*C}a)gzSW6SPrsMELb?lY7dxa4E?dQVf*AW~uYbjaiwaDmcLd+3jKXo3LhOogWm|pMFrz=4SxP16Za)16BQLa~ zzt>aRnQ06eO;Zf}dLEg!pJ8We%_di8kkjNi3LmTr#mHW4*a=tmxob6(Ut^5~zrlPQ z_dTl?vC-vxfBO_7 z<@jE*`9yp!%ti*^cQ%hv!pvzS@J!ne8@1k%@*J*d4`ZBeT}eUfc2L8?`}FYeM4VfB z1b3_F3!h6{t{}u<9y?Wfmv%aFjh^o%beiwK`-Bg~!s)-E(PRXkGJ_cp>@jR?PbN{x z@h6YN+2EGL)N{BPR^JfEw;~w?Pf~!5dOu9O8qPJa`>;rRXSS<)5-z7$6^~)F3$V;d7RB#0;r>h$(&Fw|V7Hp*G|Yp6k0kopchjG4Pxw_1 z^Ai0J%%gG~efil>> zYC79^hUQ8M*TQ3Lw&I#5=Y>^>#LwD?Xqm7NFHSk(PJ|?`&mNDf!?dt;hd);Tj)a!( zL2^5pNd~rD!$Ct7DggtK!DAP?xqoQkE<=<(n+P3o1sIg*VzvAVL|g}+|I&hHRw(Kp zO0eR&cC7zy6?TI20?qeY3`3q9xKQdPed>~cM9CO5J{^w2dlookI0DTtYQJ0ntTRQ{RF}#i9zg{NY25oqjilgJ>R=DZD!NY`aR6NBJi+;_)x7ST%9Xbq^ zW9DJa#oH(%oJMV-q0lOFE2kQ z50jtm)I0AwtB#%Mz*wSQ;>Lu%uxuS{WeLo7; z3UeEp>A2fF(!UeNxmGG^(S|>i7&s7X3nsyV$Kb`Awxi<3b8;z?6trc}liNHOP#Am7 z@H|R$KRu!%t`n*%DvMe9!*JFr7c%8iC|>G?eLAzD{HUEqbq&PQbQ##G2>1VQWr(tC z-@j8~ktgo16y{lU-tMLscf+yrI=9^`7ks;CiIN8*SSPB6lJ`2Oxa_HI9D9M4!8DYSRZQnXtX;ogB3nEO4X zsfjIA8ZUw4t0bZE=asSgWb|Rg?6`YST(i_+VXcjR(w_qqt7JK*x?ye$@3l(8lv&- z%nt0Yo&`U{oy_;g4ipIwo>!P+M%IWOn zw?H;~?GZsef6u$Ze^QIj%2ro5k~;GfdfQ(Y-kP))NoK0x_R+ zdm2941C4PyOl#6$7QIt}c|YmJ=4x)kd^LZ_d#1qj>R}Qa^^mUSe4tUBJ1=FyFt*ch zCA09AVAu zg))A38DOzW6PZl@ME9%*aE-0w?3dmurgm^Iv%P4JRSFg`n4$^ecNuKA>24Nx#Df*y z*T%l9`UrV82#2i{*rC=Yta+mOuRZIIJ*Qru9a!T5&T~>alN~*>iCK4r{`>nNFIns> zzecln@O9f7S;%w_g2=AJf*8Bw#YXzwlFe$?k7XCWZfB|nQOvMHi$!hqVDsNUgtk!) zO1T!Rm-8OhN8=q^|Ga_uwEiae_s69uS)3YGjQut@FazbtuFr+R%Otp1T}0doeO5Gh zC_8(uFMF(z_piMIb!1_CTMVZ9+PIgYZksj2*jD$TtL>uRU#WZJbCQ2`|KGBI=SasF z;qylK&}s6YG!PQ;P1IAnH%1QpK^pC|`SW3-U>tR|?~`$jI>bzcajEPnPwDoIk5r<0 zk4%{;J{5FQrhg@A?36^%U|XhfbSYDi=2#T2-TtiY7haYA&z^s;o2hr24P9Tr=5rno z;(P6?r+si~fh~58?#uE9_h6i9gXQP;fQtMra$1;AO&J$xZ`c(<+xs~yVnTr&-l#pH zouf_BP$~s#|L=U8eZYg+#Cx!J7ZccSpHkK?yMQg6RmkdnTIi=+B@IhYp|%o9sI65+ z;HzFZvFS8^+$q8HTPH9ev;{-63H#CgpE+FTSGFZXn|ME=6 z`kZ4I6G9x=##Pa>HueVLr^L_sQuF`RVG7LXhvxHl%gjAk1`b}^fc^*C6th&8C5VScfd__;M1l09~z!#bB{TD+sdr{$p> zd=K9>DzGe0mbE8YA!K+YB$fwb{I)l^8gUCl3(n!bcNZ-w`$l?o=D4yahmBgC!fFQ| zX7;B?uvyQ;S;#6Qc6CJ?O-t57WDsFU1?RWb(_@>Csj-?Z)7Yd#Dy+6@0y`}!#ui;t zVr7>$Fm(kp7H#Imq6_-5R+FtLQkeoR9)I)B2`9U0S>(5O4-MN^1=`nwE~^$){vTc6 z9hc+#hOMcoj7X({lC-GjI?jiLmO_P)kYp#B8JU%e3JppciXthaBvBbow_4I!I-tQmpAJ3<8-_Lbl-Qbb8H4#bL67SBgH@jDD7E=>DtSHd2>Ve&2vJxfEO zaykC*|5G@()+o6M)?#&;RdD~0_kku1-qoFqk21QV6iAbky=cjTVmk3=1J)iMj0^6= zFynj>iVk0bvQ`b=h2`Og<7Yf5)FQ^U@r=%AP`1GZ`f;fX6KqqV1O4oQT;pUA4si_m zF<%oJ^4^pr)(xV85obm9*f{qLwkT@3S-35qH5)-g5Y3l}}#vvMg{#3!1V==Bj zx534%WQeyNgoplp4848@C;DE4w#+T`a;m}}wYzw}xE4Vpc#LCs0Sj6j$$SSNVk+?# z2;Zv#=}I;HbeF~LS3QvCeV=XWQ0<*4Xu$Jb#!Go4z7t)WrYlkvc6E-p5ngf5QpQ7ZSE%fT$6WYS{)MT7*vIoyjv7=+vIj-zEE(^72%Gk54 zdP4*oDfyD!^-!hK(hxd!>&oACx^teYZYg2^M1Eyo!=A7K>;UQXJWsJiO}8VVpgCAnAKjJBxbya+pSlOooYkD_~McBa33}thIiAj+${&QqE_O~ z^uDx4*O69l@}!X*J0Q;YMJ4N8VIcd8ZF%&9t&wCnz4`}J(aU0m;=SP;eG%v7tf^&)`q}l&`%?qcUg?gGX znV;g=tAsYHcYI2RO4BIcCY}ABsg7rt2cz%YcqZtGW+iHO*!%Nc{(29KIe2>IEZ(2c zq<)W=(!Dx4;o%dp1<6;In{#ItCzBsNJZnaS-A zMIg_&DnF19nFs2)9PSC7k5jO&Jp+~DS5R~CH5k)IpEp3ONivMahq4O^iL9vjHERpm zOP;BjG{X5k9hr89&9^FI#|_`G;r*Tej@62BU2An819fDI`Lps~aZC`)Ia+?mU|EF( z<~0vRZB#p(-l7ZD9$h%w&%wt;b|CkF>rYqZvU!_FB2wO-#hOO3J=GW3UFU4r_x}Qo zgWpiv8i&*$XJP(27RxwxuOn6o9Z&SpU@fFy{TFfV1s5tBc8kriY-M+sd2$?68MLRp zgLptWo`im7I4hhxm3^L3&Q^xbA)T|!XxoSrf9wL}3l30Y41|CR%Vohq1qCW0XvJp4WY2(b?v7#~SrJlr!6yCF?ZU3-jcX1kbLY6M+ zDN>GD9`|3A02}UNMWQPl>^7p{^m;hFiG;-5g~%VW31&$fsA$7nS`xd025#}80XZgg zb*c)bY1*TQ?J!8xx}*2-9dv3{Ak}|Nq@Z&vU~q0RrlzgOm)0CQGP0KTe^(T`UsIu) z%XOR|PLUk8uEuHawMaM|4U_MNTt9a)X%_lXVy?ZpNdxX(5%Zb#Sv7GMLZ==nB z>iD_*173;C()B$%aL;=Zf;|tz^ul#S3@S%ZUL!IFW#f+9WAwc#PG5&AP>G5P{d5{b z+I@9l=(v$5J}iUOrG4b%mQANV)zIqU2Wi3dOnT%{Ks86)FeZF9+WZq?^M`ZXtsF*< zH99<_gxY>dSUji$ zrzbXI$--R3Hq?m5F048Z!h~|Xno-No8=Qx}U^~VvjEBOJBWP7fMDCF*_%hX+?DkEh z+~eNl$74j2)qc35?T?dN@*pfK#x1TZvF4aBIX-r#cJCOP((fR$p84VAr>D5+e+-jz zE~DI}5RSE%Fje*jPJF+E2sf@Pm~T(YZf^Ab=X$us1mLL34z%1~37y`$*nGeV&kVKb z;>Vs;+N4YRcde+voaaBB-9`G*o5_5YHEp=>LpGfMwb$pL@INR{(xbMJ&)`F(VsVgj z;b~y;AssC7RfloRW!(E*kFjD})VRkSO(nJ{d}W2_;nDE$If8W^tw?&j8{xw<@aQg| z7uFBv`%y>CY}||iISJJAG@b(U-%)~(3-W#D!&%LP=f&rt!-(s_c5jk8>%H%e;@e2!j&d;yPrFMExBZvJSB&Br*)Bt)pzW3G=p*?Rp(_d0nJsD8te z5G6XE(yO@7J$Ens2d2PBgKGPFCGv1O0V(!5Z5)jpa|KE z^U%j&4ElFpfZVkU;U>Qm#aCBj4d>6fb2t^wzZOzd^gcR#{uEtF(kSMf%K2_`0_O$LzS&)buU}mIer>B-r0?A!c!RYvj#~I zUW@E*s-F?X%P zjY+jwG`%dwgo;WS9>0tEvhQ)3p7FiZD5@BuPV$C6uzs-sb$hw*WZVI<7dpXA=9fLo z?-tDt&naUWH@cp4+G!_b99l4TjtY^V2DLxeq>Uv}?D+;)c4zefR#6ne^9GVpaNC4- zbAFsfIa1hOTf;{BN709O|4iZh z%rSrG8J5LrULUn-eY0g&anGm`a^E#R@8jTMpNk- zoEM*r>{c(#m$`_uTDyQu&aDvs4!{1iv;6xKkXhc9FD4ztXz~UZdMgm86)oEIdaygq z9m(;ITzheqVJ>SZf5`glOCvH(3r$s>ntaEw%<2TPLVrV(%CO4L;BP&5M;t=<^AK*U z^Kn}u5zn5CVQUnl*yzCuEa}ck7RBS0Bg)j!t~~-@N}n(OXuPw9~A#u4<+z?ppvbzm;lOWCf-^XM2lA2YbWv~SNYAHloKBbd19 z7prN}z{RK;C>|Myc$su$#hr&{(=wDqzrwTNXn0qq;^~KE9P24Y9&vmR%h%rT8lTwk zjgf5g&}z}TxKk*NRKJ^S8t2`XdhrkE<DIT zW6HSSPgxCvl6&)o=P)Kz|GJ<2ul zZoN8gozTTij%yyWeK!g&#;|dN%GkOQnXFBg>)xjy$1bt+nA$)%JlO_+EJD$D(Pri& zx0oeu3}8jc3iLRjFD3k%PS!8GaI^N?YB^8kNXUhoL2~b8I9^S`on${;;5pZo8+mMg z!z5fDy$1e%8`!Mb{aI`9G3OA0`sPV|ZcrtOm0Ty} znliNo^`Ie_<;bJOfW-ukXUZ~WEJ=JS=g^*kGb!^SE@s7J6{eiSZ6(h1=td`Ot1-y_ z2j;pSMA)Dl_}$Eg$*6j~-H?Y~J1nTb(E&t%yN4rVIF6l82)n?;bQ?fVtg4 zu9r6mGY7k1VILXp=X1rpyUX!4P6-P&xxVG0t=vB`l@bq03x_Rw%=Le-(a!=M^!x6L zgo0s^o_>;&zu41VT1A_lg(7vI01x}B;%A(U(0ku)ni;o+T;Fx0(<;(j^DrA9YnI|$ zwGzj_4n$+e4r-okMJMD9s5bH#J^VS6LTke4-h-_)w%0PMcekTFrJ;0sh0F8Vlqy)xuhax?xY=J#p;W}J5WFxRY$qSPD{a}#LHOF7`tmE}<$Asf& z;Skx5lkeNH&;2BPRofsg^#ym74|2YvS8yFsh^V~txIZ$0-YATsUE-^#bmSz7+~Nv6D&abyv$K(MZVlIx<8zheKq`(K2;&xa z43#tHe7Aa(nxaMz#z>Q`!yFo+Vo%@erjTiCAx0;rV#lgQ+^2XNrYR>7(0wZcW2ETX z2?a90b_G2`Z_#I;ouvLFgl@bxL`<*kc*OnVGU;3s^yp%IpVkN0!ga}VnFdK(be)rR z`_7}|(GzI0?M3R-{)bHb&WUVn;yM|6@N6Fr)fSRN-(-3_WICy)aD1rpK(4R(h>gwF zBcp82?a#TwTvYs#x6U4JS2f`n_Yz*x**F**gn7HO@$KR{Na@ak^K&6WxA%jkx(w_; zexN^jMHF&%8PB5*Kswfl-mSYI?!t8WF^uQD(Vw=e(Cpdr)Xsec*;mHnMxio3TMWP! z{R#*IQqjh9>BF=3;OX*R@IO5SW0TsjNi`Qz+UFr$Wki+}2avYcGd#Vc2U|Y3MZ0LC zSn(_>XCJ@^qX0|{IwpGNz4e^Gug_R;MiM9+O5&*RM6x?QoHmrIlcP=ve3(C$B*#Ig zS&6n7y+O{V!;oE21udmEl(OqMKCJ}FQ4e4hei*}YIS2K8G0f!Hi__cYqN=?Me{))M zJkq_VVbLaS?1(x|4Hf&y{QD*nJXlA1GIPkiOq%X<|Hq2mjjW|q46poCa8G*?7B6>! zpJoB)|N4lxp?fiR-DKLl$dLxlG$m{4d}vEY;9V2P6uQkK`;Yo$HCvSezROXzPdiMK zn|a=oHrkxEkT6ghJCcXdviH+PV_U*)qHbvqs83FovN+>JPd6A**3#do8J`B5<;gHe zbcW0m6*RaK`j`kXZT$kS(LE5QAqJfP*q-L?4Wy1|nk4qm3G{y*gu>PPktg{9mufRP zXT}4_N4-I^*LCz5FHd3>#<;Alg&oDyvH71Ja5%RHmzRvk9$60@c)tAfKD{3_dXGg z+Hdz5EL zjHyDj_dd*7*B6-@fq2b*2vUb5@u77O3`)0R_0v-P*>D~9>9-MWG9K3DI{3h~XL{`q z#3bcd>Ma6eY%V@h+&%Be?MQyZl=(d=hp*S+|YH zi6uq%-f~ZYuH;EX$X~)!i)c}MOq?#E+@~6BH1|7>JjgXmGmxcu3fGPgL6&DfWbQCU z;&6`LoLj*01V@l8bmv+rYw>=WGR9WqLMAyIs>PSFr{60iK0bvD=5=T)-zh3v%PNuc z8z02GjvYu2+Jj}r+hKD$0^%J*(VHTWH8&j#7afFms5@x#7@XK^jX{TE5uxlM`hK&a z3z~}OVh;D$eQFHH>d7~dcw`A?+*YB9oNv91W4hGV>pZ0OgqWVevBWDl zeIpjn;{hcpezs9Erg@2)q`U1j*FVoeUSJSB?<7Fy+X_fquR@ZW6H4@?sgJWVZ3$8k z^;N~_eQ3uXxJPTymYYNA)omqmYIea!4M3(l&!@SzS6Zn_%&>EvwVgf^FGpA-{q#st zy*vN6+`|EdBQXzqWfOY4?m~HK6N39^B2_8~r_Y?l?HFrllQp8}PRC;< zYnc1LU}FlT;LkA%d#|WtjkFy;Z0edXKc4IwPfOqFXdUW=6gSRSyeJgyK8e^da49@z z1Yt?J1xgF+V6!0&*2N{Fe&{>m3fAAej_r=dIBxF(Ygt=Now5Tze{aIRYXN-D--9bD zaUka%=&N%Ydg(7j-+O%{2D$C=_`G_h=sh8uAT0Af4?6 z)$d2SZpI#bM>symsgUmW*HH7n2ZvWTu=x5N9Cf{fJLR85?;}&6D>h2)252~Y&w=Eg+|r<^%Sao2Z>?6%YPBr@EvbF`f>`0-4k$8&*j&WaXCrR)+ ziqz9?{4Kxp8UOERX@Mf${vt!E@2ham@H~az3#FDxlWC@65_-DE@%1?s*DU)92Pv2E z)$b~0wSJ-9SKgBA)gxqQyB9MO&LIEgK}fhyf%(;yuv}q=vHbL`+wq$A&w57JhA0YC zN>VuQz-4j>^PpzSHMCCLmWthjVIH&q$106bnQ)uQe6C^{o>$q6rGaoh?$3R&0a(N_ zbni?pNPns8|F`}Y@gYK^yku(K5wv7{2c&LU%NZ`mVFhEq}s$ zmA}l!Jidn(>IP$`Y$YAo-a*q!OF74ZDqKgaAk5_|TM>1X4xPD3gYI)p(btpcq22&m znUl|TH%62Afx)C8#(Djl*K(e*D7;iZj0@>|X_~@5YHeRmm&^DrR zi~E!OGiTvPsX8*WRS|xF*UBExIm6`RUNNJ`X0%wzigM=lrD6R}z_GCiQg_ecYVj=U znKzN@cZ{Q=o(6EdsfzsZws3fKh#5aEX1mM&VV2uh3V$zKBwQi8OE|!w8~pBn6%Ot@?K{=NU=gbaip;n>&7z0vx^ z3}0^!re&Y_`k|YGIfwpXH;(USA8hj3=LarSJ98nO|1^gCyGKyv_a5}mAO}+KJ{)eG zSL~~y4k`rFLb0DOXs|?Iq56j7_;e=`3g6Sv&~8UV<4vjOv@XuBf9<%Y>nyhmV1)qbkP}@yZpCP{8;J>dYo(8&- zH(wKvq&1<~gyHsEb0{A5g8Qy#?7Uhx&V!xD+;z)yvaf|LluiTFjDe+R3iP65AwE3?JxszdT(=RciqB(s zK^s1Nd%#Zkzh*{vWU+M9UOE=LiT2y?qv7#xRJq-a!Y0{KjOAU{!?u9!?2*rQ9eV<| z=9lQVL7on49YKlKHzbT0L1VZQkoRVFI=)$wynm}xbFGkuPaQ@Xy3^2O(r~=iorI*j z3z%K#5th+Wz-Gy(VdL_Pm_L=DaqPBYi04#9CrpLzv2{2ezZ1hEqVcmV6%XpKV$+Xg zj7oEdmT?&BzfXqw!^!COJ^}|`9>Wn?e@MRf$5Y90SRR^&XNQEaddu^gpYBHTKfAEY zDhanZADw@69u5TO;=E2NQ+MGUy5cw3l`*jhRJ@OuqjO+BV#S+o8$%7LGMOg3;c)5IfR?mTm1t^Uf^6;t{g+R(~Zl8sp(nzXQEj zNs#8%7L17yr`@$5*v)K?QD0fXF13Zhzm(%z2d5xo$~WezU&RzIy<_9QG>L5eV{$!P zFSZo1Htu+lHX0YavYES5CG#D&PIP^8AlJ&E5#;g0ES#7m6X7AsL0?ton@;eSa)8`@i@f_^Btcz{0Ij_uu`y6xr z40d&OJvtqI6h>0r5G@(O#~B$41%sHRcr0FNuY1U?q+j z7IS@)=SXD@Y`e!n7UsBuwHzJ>D}OcQ9Q(~E%oYVYL!k6a2g-_uv}u_-aS3%wH&~Bf z8~jA?`pci8OmAx-TX`pmMTR-!*J3poCJe!~lJ$riD}-E=J{IsZVBD{09-Gd`ojx&G z*n1}^+#g5Za=j5@6h=Ny#Or9zOT_1}ev1ZSzc|mmu9rfGV<}s=F%k0yhd?Zg=TP6O zWDPvV`${H}C2+1A&kY4os?njc3yYBTf-oU{DON6)!)Tv!=DhwE>wZxRpIsHWzg-OG z*~4kH(nt!fV|4wr3%1T#h*zP$n18nhi%g&3r&|G>%8+4tIVfv_5>hftHX z7!%FoECJGVD%PL9|9F|Pz-;DK%l&rx!=d~S&kJ_RfSC%9p^hj++t_`g?{}7wX``Xi z`7^_|jn%h`q3Fe9c08g#(gOV1%;25u%M4YP`lgI&UX?>xUnw-I3h`*)MNu0JM~+1p z=iu)AesiN7=d+7s+P8Y()TY-=uVX2uZ@bP4e$B$Nxv$vbcgk3_vYxHJ@{{9o#bA{$ z3qQqdmR`Sy?U`H7{HLC0Y3d2=dH<)FU|fi9$GVZW)NvM)<0mR#arzRN4N1kYnQ?gh zI|AFTTH`~4!guOEHIRD8U zvd2Ol}s0UzkG4v@lw`DYaRR4zMg$>xy=suOJLznRcw2_ zI1CmYXVc}xu%&$t-d=gj-b$^(P6-{H%hp5xZN})kIhlD*ThH3sW{LW%a}0mvRYGT7 z59DiQv0gt2+jnQOzFco5#A^vFcv=8ggX!2eQIWQqw6L1OGpuMR=Mc%g%e-An*`6c4 z5q5Ppz9`GUSnMgQ;F^Oa)>2%5>nkg`aF&HGjbK01C$OYdDQwzVzNZtr$Xw#SvVL=4 zG7GMMv?RwtWRnZtti$$zE{%ZBI*pzk&P-3mvNne!%zX4sRumxx6~hF!h_5ld^(xu& zMH3NSI}&}=pE56=6ZWp7CqjP7!GqV~Mo0&$TGT0UR0Fo~cc6W86pEb_a5?J@vtG~j zXtVNo{zL*6tu{k1uNK5iN@Vq7tJ%=F9V}g3jLe2q;7ADPkT*(X59@p3xsyC>?zgZT zX4$Oe@c~i4{l4*;b0j@w&#Kk2YOgBB7jv9ixft;I)?FxITeMOMWt~3)q|3T8ImE7&rc0%MntbJztfd*$`r9yNP#aV zU^5gfNyacsCMotoX{ z|Ll3i5@&Cn!k>l5uvV@YHF{5?%i|r$#Z;YEH}lxFLo8Y6ry+6meLOO0z`y~y5I4(3 z_<%2nD^;PaXer7{>`8++7?aL}DKzZ{=fjiac@LQh81*~@i}mMYHOIE<#+<}3&O2kF zFp0Wv>Eg5fAMBy)HU;7yKA`huB9{qQpBIR+O>(Av?9u(T`~ zb)OU9){#QJ+PZkQ%(7M?XJH(+W@lmj@+?e>tic|uA9%i<&z}-KXnn8^RX_0piULqK zITAN!WTE$iU$DBQMsw?~=?uBK43OMH&n0u}b+js@7>zrrbDC^m^1@ z-hsKBFMxLZM&VQ;RXs?+_WqBs=JpFDw(mjs9^@?^n0u)|Mpv zBgjKt#va_%Kf!g3^F{SsaVw47YfET|FX)G_I|=NAD7^zPDzZW4T5s%B<$9gPMd)@| zf;wh|VZAly$lleA_LOuiO{sxeY$a;THl#H*2I*?`+| zgGla+A?Kyti~2=p@Fre^ykAeGG|vzk8M6VyGgl&w#~k0;hSNFay|i2WENzeC^GL6J z2zq~qMk}n-RG-8k>F(qeIuX()3?z>8tA@G^H0~%Bf29G`)|w zdd{D4q(2!edy{|8R2pVDoVx9(XCLzNSyN%|-}R!iU;mW+BM6vVCAc7`z$P9uV>$*aI-v_Y;>SrpE3mcvwjGU_%#Uptd6nQs;Qzf#%!Izx=j;f=LO!31{oz)^WU7RP9zM=F2%v?EX_{Exw)W@wO+6^4!95TqdBrs4v8i*Rn9HTGlV> zIvcyVA1bx}=hxV&Sj~=Wmav1nZn4|_E7*f$643NG{kP2J?%xH1eL}k1UqHsUTLRKVg^MzhBsrD>Iw0nI!46^{*e zvya-=Okv7(HnWv6!Ov)>GW4rJD(9CV$4{DR`6{qC!%8s_6)?#xhO=a)K*cvwa6#OG zwf*Q%(nbAAv%~H`eJg*|6ZS@?IALimviZ)scYc&Dzc;OXInY;3=g3lY}7O(*l!6k%E`~o(Mek+u4GhM_5?EJvQ1@5}VDwur%u? z7U~#{%^Q}W+ZKN~#b*ez^`uyzfIPvs_;-RRKQUGidP;Egco1z|YER3Z&1s8Ii(tKK zpx|B3Rl$gyr2o{V01Rr`%$>Qnvkk4x9V7k?nqx!UQKSGv8Ip@SUcp{O>qAkD5qH z!*#gdFocHPZiU;GW-Lw5{*Rrc=w8Du^|w&Q1`1=kb1aM(7H^L^5?tL z)V`DHz?o+DB~cb1q+c~xxsYDuKoJrEX z(`aY7C5{dpia76Xf7|kLnF(GT?czP0aoiblHgWvDIE(4k25`UG!Mbxi*BOriqWe`w z4#TumeY`Dv4o&|$Jb$h7pLa@S{ZD*5(4TfROrtX;Dfl)1B0kTt!bRCGKFxqNyC8W- z5!3b$0+V^iy!$n=!*Urce`f_-U!{PChvJwiE{0kg1tg3ci0tI;f9v^gznr<%<-gqP zBLg`VS8xe31dBgov1aes+mltS+2#d)PQQtoefe-)dY6ssFDWV`V&P8q?9eIZ@UDt! zDrjOlSU#6;Zx zGars`wxEp1hyT5IP^%(d^~}PQUFis(9tMZr#-eM##wegQ+5)E}$6|ikY7AZ88)<{T zu=fsxC`V=TFW`JF{kpj2C2s>OXK=hq$UxfI%=J~gE+W7$9%{i{&#)y+R9=(6HOCaM z!=(u(=o1=_0k@SYX|)*0W)i}KtnkP}flhwvfWr7eB=d&rM21a=<6xdI?K~Q1X1YM? zodq;L&&1hRCisWrb35xV>-|#H&oUgV&}XX$%$^6}YLYB`ljY%9*BzUhBmTD2OMV;T zR_upvE5~CbWWvU~5V5PDqBpXj7cg>K>EL|Nv{0{ z+O8I3Ms^pL_TPTju-n6PZ^sEy4R)YsYZy{9oM}U|F4Yd6N>93n!p^ZP2JEQallICrpy8Sw$(anKN>@|b zqBDWoAOC}rYme}5K_42NeE{zrVz9qQJS<9k(USIl^mLas4KIl0aq$4W>#+!#j}+1ORizoj8uq}=eGk5g$6>kcWz21kgr7+^@~{0bkF;t`5{>&jm&SS> zCdHkcVX^ykoMCg3lsbg&7!0TH8GWhgtqGpXn2Vk}C6mXRlbS`ppL4p4$>#4c&F3}p zD<6Sn-G&9%<<$PTil%L`pmg~FdOx@pjj;)MxAHNrbyJWoD(*U;?rRRh5#lkB#N#}l zp#{@cKgIMoWwOtRhGOtV)TBhi{LdMrwH)U8E{SM1_QAK3RTxn?1t%U`z)sQ*L-t$Y zkV=Hv5dsKLWb;|8&I@dkES>BShQOLdU&73nmzX9velH%>>ETOPR_JBS&f>8 znvh|S40wGw2FnfmQ5lejwt=@X-SY|jENXB$w-M{FHsMilFY@5~=T(80r0>~h|S8SkegRbRUX6zMD=AT}3*P zGITJu2UV#5fI-kyzDF}8mC5tz-b_^j$IIkIe8!!nsqp<2g$2*YiJq0_IvuZmo8ywQ zBP>U{Aa}-ZmUeD7?Vt2pbbVW(DcuuDQC_b`NEBIPg}*VPw)KF5+&prWUPqS0ltk^F zVX_p7KTKe^rIzjlSK}`!#Xgm}jYJ5qMJDobN45XEJ z_S1rCvqWX=-MorERfy5_AEQX+;(G|}&x1PF;bZ9%C>J@PCSe%vpS*_X`g9}=@k5C5 zJvblBgiQanD05VxT$dMErFsho@AzWQ=ONJQsfO=2W2n0KEcz}nft+U@r)Yy9%3ox!aYa$+82QNlFQ-y@dq0FrSkPS9H*w0Vr|q0cxQM)r{*}mNp|fM z(>ZqS>QBx=A~784k^)*_+MgEYit}6n8~pBY0=w%5kM32d<1zet%RR^s8%Hk|6YV$p ziX~jD;+{`D^irnbJzd7mZ$}ZdJqvb;>F6=<3_Lk*-)G|>8o#j(QJNfwbZ!sCL%QwJ)Vt+N$*V$w2&BX9LR|Ik;Q2;Cb^KOl;ka`n6t=SSdj5Qv*zq%0aA53C@{s z!k(Q+k^L|jWw|r4qTgId{+Y@-$K6o-E1vU6q)}tLDXn|58}4sCpy{BB%2#)h$=jHG zItIlOO61+A3!`l&=82`7`$2JZckKFHjG)x6Ij`!*G4$al{Lc4Od}K8w&;Ns6l1C76 zpdN;oGGIP(6UwVCarLMR3Lh%r*Th4ZJ9;~=e6zw`xqUpA7L1GeW>_(DA#UHc#29%Q zxS7qMle-m2cBmZ9AAJsM`M$vQ2G13BaDgJ{5zbzqh|m?e&_9%l-X&Z^gU4gb)5=73 z@VMuQ)!qY8S^JLNRPf*!J|RxZcEyYya=r7;daldcMGIYq(yDR2sqDp2nx3aj$#29c z3!6 z(MC@RTKF&*dwIUw@;{}xvnL57C%Z!N*)CidOmKSTMEy=(h2HsYG=8=P`o59Fz(ooe z{aTWOLS-oBS^)-cFT#QT4(R>T9qp_VdXp@7=6%# z{`8NKre>(uJpt;ygbR0$7dlqV5k7ljAY7w$pLFNR3ww=HgO08R(l>h|aN}&@{d?<$ zHv5+fw{q^OX%;^Ckvk9zZp;&=W{(#Rn=)DWep)2WUmZs6Pj=A#j0nVeEW|I489bbI zo1_*WqQzfsQ{Tx#Vb01K!iL&$!Z*8*keFB~S+0Cb`@DP6WW~M2j@VQ2FJrvE%j5YL z|MR`Hq}!AH^#i0gJb=!q$J0Cc7Fy(3PV=U`q9!GtYs6-vWoS5>pT`PY??el?I3);e z;3d4be3j7T!75?W60RG3f@=bOc7>$c20Fj{1PztCMW{MJw_Tm+ftNo`s+%Mn_i?K5 zUb2PIE>ul8a_0nLUGXqs8au`{tghf{-vr2*8o}0kAf~FwL0e-T=kq#9jn9K9cakA( zc<4l$eYrnF^E2#3FrDqcg1p|j)B9>WvKz9177oxPqsBSZyU>$T z=PeP9>%aGJbn~aJdM;F=>P`RLyg?x*N2x@zgC<}OULMiLyhXOy_-7r-j&>&hM;l1u z%P67B32Whsm$t$h&iy%1UyMGvJi^G?6*zge5i3n@VEg_KELl06ijL_LNnW8=*Jpfh z{qaA~P38OPKcTVEDG5eqpJ*B$S49I<&rq3GFX7wfKeT+0y729RTQs)0uW+|icj0RN z6%?ZrNJRrz(uapy!i3`jq32L_;aELAWDMi{kdsD0D=ma>J)J|1?|0Fmv_0q_a2Y-1 zW8h)<0n_yEBebOwFOI)O_L3sR1>Hu;Xol)n%kXv{=hg9eM|a2Q3yTxG3u7n_3j$AI z?)$Fu%fEd!uFpw&-keI0d7l4FU1tn>yB^*;{A{jQ;WQC5FU`j0lb^8dS1cB+R_EF-id31eM5!Ti)br6=u-Aid z*UAQUzc11!&opvamP3aNIZl3t6%}nBPEodsq@8C$VaqM~oDR{mtOksy9-LP$bJcdT zz8ge4J$|q~F6acckDu4(tRXw)SwGWG=Ro;RPu%;(eP`Tye#x^49b3zlS( zr2Z~)S$&uu#YofJn;KNH*ORu`k3(2OICSPu908$#qIAM_-r6g`hE_$e@hz*0~8Rv?|<*@l29#S)BXWMecwKRF@B0q zcpl^7ZM5??$2Y!SN>Y)kLI-74;poppgmb^D)AgxlEx$ZMT^#>?Rj0^x0)C44<+E&l0?0smnFpzT6|Xp9SRA)G8`xyv$B0Zr+S_ zBaMW@_z}YEn$v}*oB!*d>g+?UlwDZ7aFB5DWJ_WD1QIS06p?p$0-c;!Mp>V9g>QO{ z7Ao`#6#9nTBFXYn64dSeThAfR7j@3|>xu<-mNiyhmJV>tsHKrU_3?_L(Mcin$!Y`X~?w{{mUzE?wU*K_`eoCf-XDl*?0OpmSwlX3D1>fl&{2Fvq!{864-2M3|D zI7qa17>|>r_4X=MBr8i{FJoxp^X1eiyNk{Q-4$K$tn>FHpa1@D={uXIa9-HSzV`H^ zrh~Q&ts~8DxzwSiLkrv#D1y&#oOuqBs(A=%48l#>Mg*;G!i3X5aJZ#`+D?^|L5rU7 z;nOgz;G89n=XPW1k}DK$QctxzuF<)tp>%P69!=hpPTwzBllQM-lxL$vJx8gTr9jeYQLzSVVf?YrZpP1-Mcspw$}`&HGfvn z+%e8Ha7+t1JR%HcVTF> z7aqp!2W{Gix=Ul|gsK^R&37VKvme;@UXFsER6@deJngkLq7X--a|bIRRF)(w$x0Y) z3`2uO0@h~k#vqOP!er}N!r{@D!ucN~kQAJRahEzV)fMn5?IW_6b&~cpT+EK@&rP8r z8@6J)Mk>^X@|=DfX*y6OOR~cU(*B|<{BFyGGUtH3ovlI@NA+pyegSQh&>=al%XV!d zBdrw+Aj7)&c59uDseIo&lC?WPZI4W-xQ8CK_LBQ+pPl2IR8&Ta#qz?jA6lqj*d4r^ z(S-KrRp?jhN97uusIRF5eOOle*9NkyKjWp1mZ*&T^AqvwWdI&d@#DE58r_=@9DnsD>A0?n~%rDa8u!YeYy!YvMKDgN~l zs`t+NPrC;9E1+%@Zd0JXjBxgyNwDo6DQbhgrIt|7xSjNON(rka9-uqt*($QD`P;tE zI(L4mWW!+dm1Ep}meC(KuAl1`LqifOvHD2^&SgC3T()ojengGbMUS5Fd_I>cp$_$26RWak$9Q+XP$+^?ih90|QNe2>EQfyEN)n3b?`?9U1C=o;R z)mvHH)S=)^xoo`25z1;hMgct6(9W#O-;+9736X+-$ZDdFuufZ3cqH&S{&{r{rrQeZ;6iBMwrrXwXrk z7L4wQ;rs^e>Mvu2Dpf|UTBYiURl88#C|#{k^;o+Yu|O690~Q=>tq}nOR;TyS$UVY4&U#UK^ zdP#NPeJj;-Z+(1qpK}MRQ^QXQ&w9jj!(|f#;oL+1+?nf?p zSHE=ZY*_lYTf+xF@UZH$|FI@qzwpYiXw8Y?qqn!he&Y#SpCbg{j06xo>GlpzMS7Vy_MfD zHuujS9viOB8R_l*C zG5q1v$A{fdjN!H|cZI=kd^haM?(5V!`l26pem~&8FWf6U?am*h1FQd>Ubp`*s<(gd zs%pCJzUoIm@RW|<4Ojoq--S4Q>}SL1aX$#FuQ@&ZU+cF(=H7^{I`q3F<(6`JoV`R4Er5AQ2p7`E5lnS_6nCg zg70_u9pK-+@;Bk%o_BY+&h=!;jfi{m|Dx9)A4VYB=~@ewW!%UkRUU{wkbt-7mu3&$&0eai8CY)BgCaPVa|2 zep}dh%H`o>Vehc{J2#|9zWE2~6TkWV@by#QA13yF8viZhps?Z1jqpDH_qY>py`>tD z{&Mx_7v9F#)^}DP_NTSg&%NLzo}-)7gFk(Jdg$?2rl;RQkY#;;is4 z{@%cSFWMa5wC7FX#;>0pp7w9W-%IShOFs44W2(>n%&V)n@_pCbGoMlY#k~)N>C>KC zU3&A=!_?A8gb!TvtZ-ZD<<<9%znkCb`qt|1^yc)@7ydL|&hL%+_J(&>-{1If^(kL| zOZA9fzbm}#>J!2NpF29decQg_mOINK{ob>~8r&+w#TqwwFIGT=dhQrUSn}INW^I8$$Z_;?L;Jd(cg%rSnHT zEIe!Zga77rTr?H#JA67V9bWvN`rZzt`(5_v@Z$IG6+ZF(+r#faf5~G;{`A%KF*g+d ztxWr2!}q4Q-nuM&qH$_CuRgMK4KTa^e%xPh{tGa-4ADr_$g3{l;|BrYq79 zZ+m`t2LD@_kt1H49(nXt>FdsZpyT7N=BhtU_ul8Wj{N>L`-FEMzAk;^j|Ri`hh39? z@PkhbyZKQ-L>m5+uurklebKlGOL?!W$5`nAXJ z6+ZI0QuyV;>%!eHj$ywC&Q16F`~B&yAKWiIb?)BuX&?V%dh(+lNbh^hA>nPe9v0sI z+}k>L${xf4Cl}w3O@IBM@XMEdHvL^VI^E~UxBb)S+~++$Jn7)j>GPUc{F%(f=V$3_ z^J~(dZ#X^ObKp7Y@HtnffBj}X{l>NLPJejNIq7%4_4M%g9q;Yrc>G7^(m$SaQ~LDN zew#jVPw{W4eELlXgxB12T{_nN{hZgo^7{0O=CR>}A2~5xfAzuPpiTU~<-dJCd@8*- zz4e0P_xcTve>bh3du@8m)oFO%tFBF7{=jIs;FHIM**hN>e)%f?4Eda|gmzrEA*vRn9_z7r4rTMnP*-`aV?A+JdPH1h7WUB0^G^RK@yeinPj+l&8g z>8roBCEf2)pHAPs@3+(2ZhLBHjz_juhc|5CcfWshZMgo&{CAqIj|ors+46AK&rS`G zUh$<)9{0I+=Gyf5D`wMg-ud^A?#RO*9!B1EVA%hNEgktT{Df=LJ@OCF{772< z#`g5w%Rbz(^?JDYQ+=U%*KO%%AMxGvfPIb)NrQin>A36CyH8n2?|tsY>6<=vZ+h3r z6T@xsh;Z~9&kVQU)?eMU>2cw;7d?m)yKP{9w79v{B2?Ni~8D>jFZ@o!Zeap!-Ackd{E zhHggt&RM&%x?$bn)$8AJ zboGk#)PMRvu+McNFP#uhn|XWqx4o`@cqGLfuZG7B+#QDJo?U$yzZ3Qc3kO!GpZ=NZ<Z>-N84mcsJHx+JJ`}$7KrP(zyElao{`AD~yXHc;;Kt8|3oj{t9{9wbQuV_Z ze=Yp_rwi3rKkc&WvA?^hdfc5Sgl%7ZeK`FCpAXgd?p>`u?}yddi0^;kZp^@ zDR(4c^NN+>cVAo?&i?4eFuLfZ@R!HEI^6k#^TMCgsc`MU1^n(N{;j@quMRKlKR-O| zpo_zjFFh;#@Iy}v`+w=F;k*xTuYTc_PgGBT{(SYX{F~uN{-74VbIH#hm^7(FxG^Wl@ijh{R_?6+cV_#ywU^VdJy2qzpe7QVD;|8T`d{w~S)ZwUFCYr{Er z{ywa`<+Sjc1E)jz4Lic~UdZ1ec+DXpzv8{&e16BvK>MKTWE`n}x%AO6{at=X)Z~Zw zJG_76@0mZm`s4HY^Q23^7tVdebhzb$tHP<5{v@38r5$1Y6<3AzE1wjm`CjfvV|$19 zZ#ynj`TH0*UHumRy}Y-E9S0oEzY%@bKkuV=`gJ!QRz3JhudW{ZmEr28&wpgN<*SRr zy8Rv(KC$waaO}z}!t%>52ygk?y6~f&uL-}boD}L89aG)=eJ`v2$9*rZzHZIEVdp8k z!t)+EAF8i=MfDZCKTtj8NvBl*6dzsv^L{@HA2|2W>RBKANchIxM~4rLzbf3i*E6dJ z{rln7tC|N_r;Z;CPkHJ6o%huh@BV0YBfsEq$W-#ahtdFxGK)#y#(hDXkXTTZ_#+`Dyq zICJjoF!a^c;punHg>Uk^JJ#G@{JU$9`j>$)ebuYd!4JMSUDsTl&P{Gg|H8lhb?^c2 zN@pLze;?()_H=l|5BYnxlOIdJ@%euoTE1)l^!C!H!k2HkGn~Ee7s3}_ zbxS&M&nMEC&3-fu`|ekbH{29X{^W-0Pfoi#J%rzh^|9N3knZ!j_36=9uSr*K9Z6U3 zb49rQcg4S%zW*(+3zhk0q4u+r!!y45z9R{GexgQ-243WaDKPx zDJx!-UiSCLgx9=&ztH#6gG2cjj}2ercN(ucYBs(2-G_w3-?2KJ_o^?4KW1m7SKN9^ z`r>CiI(^D7-_e=l>pU~do^fV4_gxr4?W~o$KU9sS4g>8U4OmEOLp5}N4)>9dc%KmAng z>hzw!=IJrh)9KiSyVGCq`eXX{gO23iUw(Rc!kN+Z#FDZT>(~I7IZFgRu7^6&qCXJ=3A-@PHM_|C>Kee(;$`+qVS;=1o2 znx~a?$IG9a9)9B!!^8GZ!*QQIF+4L)r5}6U$?1zGK9zpt)SsuD*1aHHbjN$b)MNGu zcYlcg9qD^ZVfG(mY3ue^r7yYuH;2Ca$Zv;N{B&dZ&-N#)SO50p^!b->Oz+tH)oJ+C zKsfWDo6>vU@!#oPkN#!%r!t_uiVP-+bSq zaR1l%`E<|D^jVAeeIVc7o}TySpQrWfKAryK>_>!a547SmQKKzO&rz>uLSo-4| zZ%x~;ygFU~ntusjzWX1YT-)Br-^YB+czW;O&r09A{~Ob%zIt0Y@0^RmQLkSdem2)k z$Demf`s;6=mLBw5zVBRjL|8q2XxMSmU(=U{FQo5z;OXJQ2Ns9NeQPE?^CLepq;`A77K6w{mOxmUTymFQ0x}x;KBH|7VY! z45#1yj_|h^e>Z$@>}~1yp7-qZH^1LEedWQs(&M)MF1_#%2ZTGm`pxvT$7h{hPd|4s zocq}RaQx+0rN4OH&h&FXxtYJud0YCur+hTL_z|1KfnT^eY`O4?bo}qd?=L>^arLx% zko%6?6X(;w@U`o$4~&50 zU*G##`ig!3lGcxWRR4i{o%NEWMKGuz`EdTWMSnkUqyP1JauSlaz3=U7{_Xwm_?Ojh zTlcnqIc((6kry2Df|0ax*z?os;k6eeZ~8y~Si~L%e?0ay3MbX$()wIbh2=$5YD+8H?#bS3CuSHFP1 z(&Z<#cYAB2FM&hQf9=3I==F8l|e%Ffr$ zlVmgUL+keHOU-8w+wd9g#;MP)VlUX_s}Iq;pFZR%Y$LT|f5TAsq5k_|9-m0vd3^M< z2lM3649uZhg#R9>F*SD@OqXwNy-d?wX9NzsoySbY9Oyau%o%eMTnAiNf(0gxI zwdf0Z0p$3XpzgJIx2CaAfg$~fv47+BX+rZ_arV|$XC3ox#9zNQI`6~;e1tzwe{AbO zuSOky0=?(F{2?~)Umd;igXoi3&@J$ufxB{#J4cIM%|kPMN;Q1>oQzKe`?&fX1+)y3 zSnR~uNqok@1`yHPyMEftpX9|H*672>{@mRe&_tl0K`w(;wW3Z2-z3iY7T_a(?a|^s zI7bt{4svXZkz*|Ui0GWT=j++ef%~za3e2^#OD~5yzqxkf8$oBk__xWGAWwtg8uuP- z#m^qyxp#F6zY_8ac+aQ->`!WMksO|*=iwYXXA7#YI~c%L0Y8QBoFV@9;oV$>&O2x= z=dKXrGmt^uVH;m#&6NOq(Z@ZN@pHGQL!H69qJBmh-ab0@Ao|LWqjMIqFv?DW^%sM2 zV6WD74;6e)h3-K%*1frpJi_NitUWc+&4EemOQ6Pi58C)SqkWX{%aGS#HqwPb@-bPv@Z5rva580e+mX9eM8}YoEv0ycuYr zGq>n z%q4AtUVUk-Of^2`DQCS&Z zW&6-)CGuwA{ha~x=snA2=yIq%t%k3lPmsMU_LsyW|AsEV1AQC&0)B055q%%XKpC{r zEr)vk_SD2T!8d_7rj8tqgI;o#DznAhIWZft&!Fp^hc?i_?wzg@vjSAWB)S3U(pq6l;QQfA=p%WwPZF~moPn*g zW7_KM9eFkId1~Xk2u0yL40w?>m(Q|SDh z-1T?%EHCbV16z)ty;^GqejdyLcVG`0)N>jDo$r_Mb^eI|O3=XW?&O>Bzk)6h*ALC1 z_GbPBYOZ1Qo>L#NzVq3u{uSr~ScPsEwhCxQ;@F=)+J^8eq2B<#96$AYkRyKf=nfKe zi{Y*5J2h(+-@h8;>`|ZCz5h`28JFWLwQd`I6FNZ5DsU3m1m@AXYjf-8nN>l1ws^;k zYhyq9sXqs-ukrmgRK53AxIg=K2J?**-=t;+yi=#6*MB#*h)sS8x)I>6tUm~KzB%}I zUij#DnRckQ-ZQwAIA$Xas%qKJO!-)zlP0WH#Wl8fpz+&*wwe-Bf5?cwBv)m z0q^eAZ@?#u%##;)Drpt^73kfWXZz2y!oQB}Jx+)*esR~|xy+$%H}*F6MdWY~_S3{K z$G#oeojOlWj5B&?>&V7d;nmA;gP%eFBYZrQPF?hOpr1$9M`ZYTmL>EtFLWjBy`7e@ z+xI2thViKY?@0-rdULqfVRZ7-=(NdKpgFd7UYx^F*T|+UW zGw{pdt-B6tOoX=cVlJt3R6q{DfL$N&i@mi$9hAjr@s8HdUi!(INBq!lC#O5Jw=M9- z*o$ZDoDDDyhOk9q+=X{x40{D?%_jQ!(cL4O4!D=BkMnbZ5-MM zwMX^i@Xj^_jriuo*^hhz->4P$)^jjcn&n0OOxHhyTC0h_`^fMcXcc$FT=uNqy_#21r08*5_|QIWKfMx(p=ftvg8UCSdFU)Vyn3#d~H1xlU{o{|NNih|Rma3*H&+ zVGLRR3~ZkN7}R;p-$dSnuXid(w-#TYuin>|9mw`Ff!}8I&bS-iGnxkp_I51pi@CQU z>yu+!4z-3@`w-r4vf6vMt`kso~(N*E?VH$oJy!qP5Qtys+Jaf<5GgChS^?dsAucKds zyiQwQ+@DGK7>l!CQ-cS+K38K~3UUz1;XWe1OOUPOE}F<0ScI;PJTpsusPEHTYDL@- z)VtQpy9}HBrO?g5Go2)++|?)K+l6jDvNPMae(LnI{+Kd+#|Qp$U`~B1=+`1QY7f?Q zZwb17FoWH_mBC_s)`1&APTdIXBO;gZOOXAzPiuO=S7U4D#rJf+&%Ar;)%C%qpYBM09y(Sl-Yetymf_WlrSKcDdpEpGowGt;#xFN`MJ$7Zb0$w^TCv~5S<^J~Wkzpm#G z(R+3!{5^{%bTzVPHwk|h)c6(X>}4m&$m2epbu)JF*B;Q$>8nc<=rnK_p5LZ!9?yMM zmp4bRw^%FAWFt1`XyBJY+thLv@754D&&hlhuo1iQZTJkmFm4I@K70qj09WA1vV8jLZga{RZ&Wr()gU!T`0q0oWDrU*h_wtvpHSu>R%ioi4%0`E~>IL}ce$2{yyepcm_qjrDHV(Kq2Mv&B4<@E4$S zC;nWj|0K2q-4wiW_O%UL1AYVgwO|;zgxm+sW3eCg^XSUJJCi`&Stf6PP-p7QjPDX+ z)!CPQl;Pcv_iQP;W?t-L7q$f6o?|S&m*{VygOBJF^zwE6fd1YgKjv=6;*8C)2-*6} zx_W!{Y!|nRUFP^M%oelh-;ZAdaz~TI=B3_&4#1Aj41IT{eN| zzZ2@L3BKmjuh*yh)5p5j)NXHn+||-<%o(j>9}%0o(no)5T!1_TtUHg+Pp^h`#3Xv3 zWqWWgcQlR8`)*tXK5GH)>C7m(L5AL$Z$xj7yh}HA{nkM3Pv0EhCOP!8FYi|~vU7ZT zJxrimz}Hy?hxLJ)qYs}tNZ{>9T|{05pWvTIXt#ztog>b(GheSx6JK|l1JA+u z2D*0s;P*JaUd`w8u56BoPcM(OjBYKl_BRgS2JWgK-4fz9f_e4ma&j575nIO(n=zjM zMey2IV6%>UbN`W;UC2$S{n>{#6YTD>0&gz&X)bFufqiB@nb=;>+LBgbv&H~?=M(w= z)ZdESsngA|d$#ylcr*HSU5XDMG3y5Eic{|&RE620(Bqu=CiN}U4-W7=Ai2R`9#=Dk6#Nvfsb=q+t^L$BDy)? zXBoCEcK;g}_Ly@9*o?{1PwK~Rj7^Z+Uv9&!X3-w2bWB*4Nj2G7Yw2+XQm-8B|*b_zxnx2k-NS7J7V?GhpA=wg-3WpW(J+ zVRx_EmJ+iCeGYGo-y2Rq6ZG~puAUjKJp-Q*->W}{kDqz;lf*1QV=SIwcim6CpXJbX zU~vn3&c2}I*vn93l)W$3F?Wo`J+@zGSq)tYtdk=*$>EOctzwS}{^Q6c(7-PB?Ct^P zb2q&`dxqYJ-N2dj>)c1{V#~<2qE^`F^#%Hs@gKz3_n$v2Mr8X~2H%ADyeEn4^|2GV z4Q7CO%UwR3?cO`6=aNvbx6fg6?ZMC88B;cUhj(GgyKyA$}>;{-&^h8@sVXU8>!CIflHltIMF< z!P?q``&t1%0y?|w_G6yQ@LM29=Usw4j-18fduMy^^?VZg2J!^7H^cH8JRfJSfF;;+ ze0uiP*xiwHxs!;^`+F8TceVyJkz*{LrJe9Oy!%~KE56Q*LEGp~ZWZs?VW@Y&bN}D7 ztio2uR)T692j;&Lojb8F&q(ISugv0jFc_sDZ}|84jc_)egoN9PRgbq?9xc}~XN1mEkq3SS25JAjYf z_!2pK`l+s7o%?jZNi2HJ(KV6X=h7N|KqKhIcJ5IxKhzy)Ge;A@e3tpq4FJC{F_)it z^diBhm#cx#7JSO+hM`$3=H7!}J1=^h#OBNybTK{+{dkX#qw9yaPUrh&{JksIb7ns? z$l5z+K#!i2ciKFwfW1`E+52+*2D>`7Tkz8eX z{hR1Vi1pm;V1c+b?HXYg~T%ot>S>*$7R^oFnZrU}0a+D31m8Pr*vcM`eN8+K=E0QY4tBT#cD z*cPdyhBpB!^iU|!5f=^JTH9bTONzPBQycVM$oY%FW~P!v^%38cc?s}4Zcz<_FK{3D*UbMY|bUW7QNK7njpqH z^4c%KZy5B08$s;qy(8L%xtkFlIGetHSl&A%4jT_)J4q3+Z>Yz}qaiOurRGJ5ynuC8nqwLBy9 zcxUaYj!%SlR%LBH*oS^O`rg+U`_|q>w*=m2)|xZGdlZ4aKJ=`6`5sE$b&SPxrN42W zR}-Cn+B`FP&-MS74_z;xeU?BHsa@lKfXZkg&!BH0x1k$a#rM@d*Pd7180XE4?>$!F zw;X*h@31^R_0d9SKN+&;-}A}Px&I1qCplExAi4a^BYS7O7wX#)iftEiHVX1$zxFVO z-a8^)-74NS?#A63Yu}x7hWC4lKK!a+F<6V-gxX^Xf4{ffif$afOpZFdIn3|9Y2&Bd zz-DcCQAY0$7T~R=%|6YMKsTXtju~KV1$q(wm!mI1%U})s1hgOM<41WDn83fURebLf z;iu45Mk$};gzr4OcejC$cKzExeH&QMIj!ZcCqNR5cf|_y?qnUPqj#3Y@F&;s$pQJt z;X8K>+cNkzv3uY*w~Cspq26<6z6ie4zcawkqw`~5?$W;9`v&+jx|Q$^bnQqTWcS>F zc4o+l!)Cv}yLx$tiB0f{#QKRj|BrelbXk`_2_JJ$=^G!MYbm@>txDZSbChxt%`;?rSwMemu)4h^;{FXEC-lpdE|8JxkBP{zVghXRG*m zdp&X+zYHYk?Z-QjAbTIoy9H{$_A!H;!Mp!9K0A^7vFGsW2jFjO6<CBEhn&T%=|4(zJ}^=FPG7JLS^pE5f4IS5_c zDq{6-BHMQY@3ZL{xhKzb2Kh#0ckj@SJvklWCC=r(lbTmj1X80(zo zwB`uN@blT6fqIUUU?&&`zK4c@_hh_P)ER)9Cj%SNyYGnZub{Jk=mnrePWNYxwZOcC z=yT#q@b)o<-97Zq&0RT*bJ}xfHtcQSCqrL{w$yX4%IM5d2MK5Zb=sHW+uP4b3m$n9RDbo&uCC`}?RI={x9+Lu zH)&36P4dpbSD@B%U*o|3HWEJtzlOMe_yOo<;0~Lk#g6pLkO$FsKJd;_#Xrl7ck4W| zHRgf-3&`$4zaiol;M?f!Z3bNh+1Ml&z1PuAqVwLlFYjoA-QE@<`(9Az9q~+>#HowG z`&xp!!#z-Ic;|K_ms-U+MdYpUOTaL2&J6onsC~MZe)Pt=LuG3>>A^bct+gH7B(RoW zE49zTPpii+oU8Y4cFu8h*4Thg1z0=B$NUNU1YUdR>=75Yik&sE&jW1{-KMU78r>ND z##%A&1k@dyN1u886J&dsz}KGgSk$z)@vglIZ!LFVj$z=vpMl!zsh|lK(79K8aYxR( z3s|d+y;BdH`Ie*W15ND4^<%F?+tjz#1iU?sLw7>GzZ0NMtnd09>b~0AL4s`(8o|n1 z_ne>uz%voro3rqImZRT|tgei13|U(S4Sw?G zOF)EgLmN=dJ7ydcSwFPNy!^ z+&w? zO?=H6;oI^cgBR-a*`(9k^~8(;{p!Ge2Wv%r`?(C8I^n*SVDouu!8f!C;)bElI1RtD zR^00tvEjQ-AL8wI1#s>ub$+_J^_v5|_px=GvA8SVgElelw+tNzy|{K>+yT#N3VQ=& z#3^Uk65wp=e2(1Zf-p^cnd$6hh5x&MI*qqmSb^-lX0Ba|)sMCg@#J35Z zz}5ic_&THShC~~332N*Nv`wsWe#+RA92uHp+XK8e-t9Wno_2#eC}ZnK?=xnf^XM+Z zr-JTskRY4O-Y4MKptI*#>*kNeyI>r9#LrzNU>>=*-`^u^@5xg8vSwwp_!+!svt|SR zNwuQCoSfe448KKC@5Bb6{}SjVdOwkvwa8gs%ow9e>L6EYMQ^^de6M!m@vQ@EdEc7& zW%$P~ogwcKcE83=;j;*;&bbEB?}A!$2mA(X75GVDPmR&yzRe-~Zj9vhIq-Q_wx2dS z_o*$1E`aIT;$2}5=Wu5;pdWcB@cf)X{ZjamS`n|^+}5?{41GepJF+Kd^ykNREb^_C z$F~F$Z1(1C?x2ah8`w)7^x<2E_Rhzzakqhpa>r3yz*6AJD*eS3PAAcrZ2J{)n z=H51UWp(;W{aI!U`oXSk8+3vC74&}3dosxL;!M`T-v=(M7594*S^_try9apBeO|r| zJTG@&f}Vuz-P;4!z|R0{Sigtmwc>o-&%N-Iqs1P5j>bXT9-!v*IWEH+V{iIc=jW~B z&x_Sh!~Y75qwmx}=RK02BHn%PLUum;Qr~)rkNc=&a|gZN^;wTB^?X;>iZfo`)#*0^ zKi|!-pJ$Mx&(SY|KLe`Xy-S5P>}3M<*Z5uos(vvvlIyOn-1Gn6yQ^Z~4i;*BjRfBB z)1eWaF-zrR!7qX@Vc!YdK^56}F@-z?b%xIU>c+SS^Q?gO_V3P|SN)b+ai2ZA&0Sy5 z%R8^#uPsgZEEYR+cj|>cy*#}->V5Xs){1?2p7wkyI(I1_(XB*|@J)0L^~f_&=eFKD ze0p_;(M@2>^5S`GVsi)1JPz%|Ve>4?pn)!rg^}uGY_?gjy=a8IB$fy*E|;QJXrSF}t@6AD;tj_3meHmO8$E2C(b18=pyZ z?%!GL*PhJR*&VX|*uQn`ryYyCl|Z#E#^1dC__m?$wFEV1f^K}Y*mJ~h0yvL5-HELV zwt(JDIr{Ng?l6A4;M?T#GY`KBoqd@*2cAV8XzR@%^CI`0x~qzJb|ybt^iYNlQqQ~O zIn7{;z+Fs|KSCpVbGkF_)^XqCU0TOadqv>yEcUO?{q@XHNeD>XmelueB!FNp= zS)2FHnY;^&z-lmuT*q!7>W6B@?DlOx&Y~`ZTHjqca|Ul*I~F@^Vps2{f!=fRZYJ={ z$(smjw9(m%b+tQp6S)`LyF=@#X{!8;1T-(52^4ae9RM?vE8BiaQ-Ng$0R|5BKzB<%9<&16k zW-Pwu+pn`^$Pu{TQmeR!oL<|W@SD-we;L}2#rf>P<{4Rg2z|X)oU3sycxfj$arS1e z3H&CZeNbilT?@Yj$ooF>%ra_BLXFq=9Q1xX7tdh_Xkgz1?7a;1UyOcp*T-1T*j>6` zeci_(u;0$k&^Lke8E=0jXutLv1L$_cNfOH zgCyb$+UTc&y;!pZKZaeKF$-WZavQnaN;)~c^G)PQY@WA!aE=6BL{{Ic-)a@_iJok& z^U*1LUNhhpVx4Imv~zNx+cR6#&CyMR&P;@LWKOdKoh9cM{m@lK7<)vaQd-VQT-+vqn!SAlVI%n;LrkLa!A z9MYvwdmIO6!0WpSy19lwygP8dEuizYi6yAO;|E*CJf6Ax^Zk*a zn}P0t4uBD`vR1tFz76j^*$6Ux+;8t$wud$NwZU4TJdeGJ-#Y!_^`8WbkQ?~;4zXwZ zT!pR-Z-2v}0zZwt^Xb}REPCtBYYsooVju3^9rf%|XI|Ul&tKH%MSbf$)W;Yxj(!jt zV{rxptzyUaRz+{j8hCqjE_dT>&a)KVc4er35q=2d_>Duo>uvISFWp5a2EFH0uaOU1 zucyx4#K*Y?@#*Os=zL%Jv)UZIXO=*XZ^M_-uZCA=e|^Z)@SD`}aYpy*IYjI+@9vm> zt>JU0{S4x)WxggfgYLp+Ui&RWy-yMS9(ecRecy_#?>IK^?M`gVk?mmx)O*vL<dp8EnpxVv74B1+3Py+VqE|ukj+rVt%QyC*e~XeK5uvljv3h zXX=9vfeP4#JPCSzdKV*b58LH)YQQhdKIk)zz6@4@l`VK|9Xq}`XvU&X^YmdK$L`+7 zkb8Bc#@MUR*lz6lHh{j~A$|NBR|OT|^HTxpo3+AUelh%d?D7-Dp4H_$-}B|<#O@wT z#M(m>+K%|6vwzQW0KM_1Zy|cJ;&)0jVPhDG$8nO6#K963XE%=x_N8S$av%Odyetu6rgHHzk&z~LOJH6<` zXJf5vN49P=FPiq#_RaHA88wEk0s*zKo0L+)wcvSkMJ3`ZWUzk zm*d;BNh7kem~R6syXTaN#dUQAxxbK?imxz7Y$4#wf#$2d8BM$|7tHxF-p z?Z#aIZEBCjxs_*&UAOUfu1p(#qK!|EtWOin0A*+QGYP-8R_tU4w1K?}O(KAA=EZs1 zi+j?(xJ5pEBC@^Bqg#v4-CHB;#ZbS3+!!tLuffMNJ_DM?q;nTb$mVlyb9Z`1S3xK5 z`?QR_6#Edo{S3itTLEoilOMyj1bGVCJPXh|bRBx*R$`09w$bIWxKGBJ?;`A;Ls=Vf zO{g>Zac1W-Z{DTNSmfz_Ph_4E;_LW~g8}#)AKz6GXm>v89P$h>{}g_n=`w8g+9oE+ zi!;%868$v3^1WKlKH1e-rw`b($l&d_3HtDBLw%~Jec^tUAGH~uXx%_zklj!Qees*K?%-yM9xgWPTd=$147mwk1) zbAO(N&%!u5{jBHnJ_)twRrq&m!0(zZW_1^J^v<&a>a6zA=^vf*P zomlLzAO8&aJweYu@A{d?zHaN;^5Q-_lk;WR>d=#VcIxdyw$65Rm0EEIW%w3opl?#w znv0M-_4G%VA?NVD9^Cik!1=9_;Ip#U?Hzdrc;4;2nB^jEv&Hv_W%wpG_qzyw1(5c& zie2?;^yoTlIlh~)XV8_<9C(h_@GQKeC37<4R(SUq@mWAO1XZuSNqikTf=`Tveh_|b z*T*{UKfyMDK02FSUx;1fBP18oPThVH?D@6hGx< z@SeN6F?60s1-L9H?!~k2s}(kDx79<3(0OO|*^Ey+7CpAHO@bzJ1G-1d;)m~2{F=!2IRTv28Jy=F zYzeSmeeKg7YfnJ$&q0h`gWoQ7v6!uiTn0YF<|yGa4LpZ=kOOTQI{P(#8eIfC)tLtj zqfe0CT^*W3J&TE1c64DHw5Ou=X5@AQq%x%9srl<_t9ICKbk4>W@= zuNB|BoD5wCjCChI?^95JHduwe;}125P@V)i^*rC6Ub+~+9NjR~IjyOF2EOz4BNp>{ zUJ>{V^y1Cg#y)N=eAe}iQ19wipszWt?cPoT+kvxAgI(C>q0ZSDE$;5NTG5YZ<$dD_Zxs2b|<rR(s?^5QN@J?8;nFZQ+-|4DMKfSNOc1)$AdJI_H<%-GA@xqG9< zj>`Bn$yWly`3ZcA4`0abQ?p5t!fa&ktf zXS@cT^_tjsBRlI(Yzfp}A3C$=T1Ic3h`)a3G>?95bi3fkK^Bpreb_g)ig|sXRDivx zGfxKAz&GW&8}_jjd3-kM%(WOk!Bzs(pdGpM*m^zdH&~+|^zI?zWBwe-+nfDFWaCaH zUj}tge#-Fu#OIOT;H@t|37_F3a^U9@Vkbe87ypf9tOko;8zhnc1KK^CA#8Pcchs2+ zn|(B))?R{r1F&an_?f|XtQ!+!_kT-;-C0(kpT_QfXNcPcZ>%|c_bnl&4W`gJuREK< z&$%LcvA#w<;3rRntNDUD{PCPNzVD=w-&w` zi)UBA41Q*`h+lziBe0j{&^17xRjuL<`Ya5B&Ii4Fb!XPe$ZgFssGxJV&bb9xcb9tT zTBtemNDsB5zuow&8^e~tPlJfvp3SAL32%+1_>_nn1p1bN&y4ra`ykEmwU-7qeWY!9 z^pnI@K!UxOJ4e3`dAzH$*VVwh?xG!wuPKYrEp+qPljpFkR(##?tkspZckiwKp38cC z)OF4QpH=84@h?G*xd8hZuutU?c%MV}(x$e$fmU&kjBBF56rXMAw3$!30UEK%@4{~r zas@a;vsTPB0NoCJCVf8Jz+LvWiW;3ca&pWTK27LiY(E0cSo9g;lUiZVu#YKI)6Zh` zQ5*V>4!wJqkLcR5_}QR7THGn=F5sU5n8*2?!94DIExIablEXaiBqxXGXI~5WC+O$U zXXrC%udcLnANV?vpq~WJE%hvN;`CSNobJzBZRB}izLJ!)%dwm9a^lVhp2vbVU|b*2 z$NuJ_)~kcgj>%&!`^-RHdo1q$41BXz{P)%wa;$~lgKh}k+UCvhjl>PYH_`Rx)8~>J zGB#`ZeBKAN+m|u&-U0VzZxzr0dw{l8U4MDME=^K#W$gOsfVx6R0FvTM&KuaHJV^KFjt1JbC1^wpC+>X*r)tB)P5p%cQgqtq3guv@GbN{ zW8`XMUyQv1?5_ewuxsnZ3?rBEbALWVeZUzuK;899_zB`RK{Mct=Cns``p<)Ya1rum zsPiSUxZ~bo{TjeLGav_-urJZ5*Eocx!HepTzfi zAl)@uht26Ra(U}Xs!X^eCBk2=8d!H zpqtS5qZ~|b2!fwt%XcH_2)@@U(jNN^h%N+X7z*{GW_daM}3_k+Y>pOu? z**z@4Z-+0zH)@5Cy(H-MufT5tZRFnGs;y$4CNcU5=Ww0`oiQ`;)A05q?^(FVUC8Fq z*E_ceB?9Q?W-qN_4&X;W~BQR~AVRwddVnVHRCp?5mFKSqzY41JHi~ z=v!?Sf9{r{x35m0%lP~%6?=Urae zy3$gY$_vk}9lK|YTqe$Y5z+gM`Rtgx39qmEY_vGPUJmE+j(8sKZV!BCKl1o%*LRRy z&Z}*eI_moM$~)(02wg)SyZR~Q2o?iBIl40OO^|>bzmv6fXK^o0px!-1Xa(4ham()6cceb@@t)QY>Heu9_`xNCEkpze4WA7gx8OVFiV zc@BOO_%6}E9g8zgTE$K(&^9&C!fu`!umQdweTJ_#XLcXv-39NQ6_9uN75KaB`O)HD zNVg*UT#kb#IeL5Wo_RLjW$Ed);^&+ZsIxbT)wUV<4%q=V*3d`xDi5}S88BHZ&d0lA ztzHfLU5U>Aa&r1)Im!2lbAeiq3u5|8jWuJdSLAKij)Hm)9dJ*EfSXyW50vM45PEw zHjs$!U2SC8a@2-SdkMWY%xRqcE}-89jM)vH1b*DHvD2f49AojbpU>(Zboy?A_uV!JO@RH3 zK%3ayUmL#-+UR2!ym4!ku`j^em*=t=o%^@eFnnpYI1Bsg?16Zx?@)aw@Eyn3U2KBy zd}6mBVsc;(&!HDTh5v4l5x*LJGcVrbk+>Orm!gl@ok1V_b+$fu&rtn1ynD<*lejWI z=06|(I67zb&pG;Qh0Xpm;v%s3Gq5cO_Fo~kiGB{BBm#K%Hi}?+OeqK>Fub&LweOWg~o|RT{ zwpD0?T*7a0t%!H-GFXo6o!Lq3Vq`xP&}n>2=+?mZ=npv?rgxW-N9vgrB+@;_P`0IRn%9h=^?jy*-%MJw*5^unf5$xMTgC-<_;P z=iS^|>)t=?`s~3z4pyR%z}lTysQGtdt3$mLli1sl-hlnq(d8h?i`|>wS~*xiKM(xa z&j1*#6)}VGOW>D4t(C>@Ip^@k5I=%#J21xuwW60!K5RqCIWfjIiCKl8HJwGB`!Yua zi|}ot&)|Iz_W^g(K)(tsN7i=@bOrK7z#TiYXWNWL-+9iST1BohdULp+8K`}(2J6vh z`1e68=$qK<&>X72_s057e0n>J==8C_b=bV`#`<;1SwsWbTH`?3ueUv$z7^zosP}Xp zxF2iJp!3`>dlO_}6S*L-k&>Vdk{S110v4;k-{@$x4p!ex~ABNpCE}?JYA zce-a)PyqudW)xA(pkTl(269G%pn!@50|Fv(r@KcK6mvom5K%w{#RM2J^UbWk_xF$}H*@e$LZ?EIPfB2aH6H7hM<2d}i)`kPW7i^umn7Dnw z_rd#D=OllBp2p!chsRvK69+ynOAa|){S6K_TwRCLzxH9AU5eACINNFto4IDooX-1I;GoXw<{V5p zJ%p?G<8z7 zba4E74(?n%A2-kOg84*%iD&d;qA$KgM`KZ%c;fBin@>=7JapF;?z zYx9x=A9s!ur*mtAA0Iz9-gf~f`CRaM;Pa8pA%d&t=cItsC3)k@yq%A8Gskt}eC2R` zCG%yjj=z5Kc) z4t%_P|2zWzN543k8w1Bn=K3?`z{@_a%+GNM2R`pL9QeKpI7roU{cvq44(1$6xpIxu zKfXGr^KCmggw*_7LwsCP96Go%zpnTg>bZJfPS4@^QcnNQZJwOY=lVbK@iK?2m3x;r%*sTwAV=_gBhEea>#n!Iaba^}x4t=gNFsPMq#r`>#J= zPA}ke-k$=;lj7>-INpEa;OEncvxRUxUa#Xgy!=mnA7_u^Y}2?lB2Mx#@ara-ll=Jp zlXE_1K7V|?9h}bZz0I7?$HkAGZ-aOnhwmKt{$sfQB$6xJa9rN6WDdAG-p>2t+w<4( z4i1GJ_&wX<#LZVNHx91Mk57s#^ZXi4^1gYUA1jYOt}Gc3S9gQ2G4X@3W6XJ!kXd+Vh+o zj$gycWX{e@zW)%8o5O*RZ8K+=ND zSonPJ_52?W&(Cq<7+gJX>;4~mU+us9L>kwogR8IM+VMX5dVXFy zINg)0^Wxyn!IHD_{+&3k6UVLRc#?c_JbwPuIQVil-d_}_^K;7Qf}bxwpE(?AIQ=wd z<8_|LkDng{-w)4M;OcmtA1j{^UN7bBd`!H~@6Z2Szonf1wc+hCT>Un#j&IM$%*Vs? z)HypJH*Z6n9>D25KZI+`+e>Tz$0w)rbKt~*&p*$p;W#y%9&!0!zWK48=5#&=$#t0H z)N{7)9Ea!eZEQK&!Ex$2nZWUQn>JVG$H4DbJ|-{D&c`Cf_0Px2b4)o|!lCjq*S_Z8 zHGsDpayBuCX`G$UNeBmip7?P|adw`QHqQ>CMA?w5wyzBf5)0}t5aWaW8m^JiTA&LeQxpp z3;#j~duqJ>|B?UyPsT5Fr0g99|Nr@?Z0#Hc)c=8hkB$Dqh>b&0R2lY^2hPW`(JXC?37NAm-wo+Qt2smD8EVSNoTPSUpLw)Ipr`lI_qH?8W z;9aX6+GHBb;C5T)Ck&wud&wY?coKTEX^FTa!xHVwyvwYcrNI1IXM>y7i>RKNfj~TR z4G7N`fwxmkLEiKYRKH*l7$M(Ash|3dZhWx^ug4Csh zQ1ABTplIt-CNygebs;iFn6~!;vo0})8dG5`;MbpzybczOU+xYi1$9fM;#D1F%K;^A*Xl6)B=S} zCVh<`lE2U{m~A}?*!s)`FI#R`t+C7%P6US-=ZbKu@4^W|@$Gcsz({p*KtnQe-r2;Q zxcyPkHt{xCW*bK_A9Y~&Dl7PFa}wxs>K0n>-2tMWW>HwY5lJud7W!BwQ8E?{jC(?W zFr{jdFhe<1JZNq@rTOTkaP))_@b1WAK>;v?Wq0wO>vbsgxZ;=to;~>U& zp_keF}7*HabZEoXhN9EfB0ZF+s3%w=1=+Elg-!`Vt?io5;+K zZ)G}m$_Zb5s0S|(NW++2`@m?kG1S>N^O1_)FY&!Ci>R4LLn(Hf9kW$%5EO;XLu2(y z1#eUrL-X;8Q2*FyxKBQmiVpoN41A`K;+Ut@r$x`1la>#d)Ys3&8$EW34aeMI9K(W; zddL~@_OKqP9k~Y$Fssn~-~Ln}qshEJVkrm{onf-xWeRS5w`62??hwv49*x3U?hD6O z|HkW2m{WUQgQ)(s=PAPpdN9vs0gV3sml`VWp(XKwL-N6HMr>zm;F*; zp#OzX?>B?B=|4;i-A$d7K;phA5z29Vo(RX%Y3`J6nRAGfCrwl;Qcd7@J>n_(DTu*>>6`JnC5p} z?0zhrkupdZM9U`$e|$J1yi+)a8TBzjJjihnIPyb@*(*I9d#s%-P8@2^Y;>q)l9tDy zA2&m*&N#JG6YlH6p86!mFJa!eO{L;C+oCz| z{-Wgm_aL}w9gMLm5}y0B2sq@+z@6{MAlvhI!LAjDfP*XqJ06rV;>qr4>e+dM(TB>Z z=;hWxCucO2^>c<>D;!B+!x8M+OJRi&8FG2HG0>bDLDQiLv~c?%`t-h9IJ0&K3=G;y zZoHfe8AyHm39ou$V5<&lVM6%2A zNNsieeRLw+*gOj>Mn}UxpMFEL5ku&M#Ts;Lj{+Q@oC9-oRmhF^_VA_OGwfb9jqF-e zjZMnzV0igC#P45kUx0j7tkBAITUe&s4~&_k5MA|$>oq2Um_9`qG}RV}t_>o~zDu#d zZ7PTiOd>)J9g^*o2hJQ$hMP*o@TGG%0&_xaZ{hKpK`{R9g5o(50e`wSDLLg@U=2-^PT8d|%rmb&-pBba(@ zEKIvCB9A%~$+XLZ*_h=car1O3SZ=WuWYh)$pUWX&OHwbG(ftGjep>>M*&ATx@ECYc zPKwyJS0G%FM9i}y;PnyDKxffTf*sq)yfna?Y#9xwCZ)p*TTjAat%2k-*hk=jVB*r! z49;q}f*!g9jlX&ao{w7(oi)9Hv|9%A{!}%--})GJYYwLCnzn)jw& z)yHjNE9ghI*0fUYV7f2K5RCenLQaQ2AhJ{Iz-+@O;DFg8oU{*1_5sDMm+2 zn>HSQpIp%n$EWT*B)d&Lzyhr-DD)mo>bzv2d+i^fb47s!7a9@wxl&~P$wS1y;yFIs zK8+aeufhAq5nT4;6CAY27cLFG2(Bb~K&c-`$nqKE@WCDGaM8mW5<6OjT+`JfDyrki zPc2tcw&*J*l`tI4+PfT%b^eIzT2I2zpAJ~|K_F%hOegPW1e3*ALMR`#GPodV0*b9y zfWgm;LCuD#)Qcm=plx3ub7PY>`Lv@4PYisE3-V5a54v@@s)OJywb?iv8bLg*g(RR}mX`j=|wD-vu4#0ETN)>k{s29; zCGtIS9JK^$Y3_heLUUkuyE*)3><(9b)nOM4|KM{TvB*ZR4JnPmK)ZG@eDqxw&O1;9 z3y$j1%|ixYyKF1ayln^3Bw5|2Z(iI~QJSt%lRTdc!?SKf(0u&DhN(UN|LYG1&j61}>4k z1H(*~(1~hC!1vuP!c3nq7?9y6`e`>&l&~^fWKd!Ors@uYrju_&-xDP`^YjsX+B6L| zKd%BmMmvKc`i3xLXBzYhJO-VvkD-4pj-Y2vs-xAC3+OGh2eh1&`CZ{gVql31Y(l?cV!*hdD;W+=ZP-nV7{eG)H{p6Dx z^apDAv_l^mq3%FBG*6J4(ohuAm=9koAH@%@eFd4rL~weuDJ;yh0?+NA!YM8I35>jO ziLQL*4)1C11)VSr%u>Axf&;EoR&CZ$jR9Nq zQ;GG`{XmXwxBzvH`@vTSX&88D3@N%0i610h0@Bwk>Ct;#>HB>r=}*jTTBi8`*01Y> z>Fx^jQx!3M_tu97#olyd#v)SGHYz%#}gQee6 z33P>qKs2LiuE@W6lW2lsvhZ=lWGX>;6c=>miM|9L5q+pTB`W?km^=G)gRChGx)V7^ zq_8ktRC_W+q+u&2+k6_x3X=hnxmb!;sOp0+e8e<=9oyk~1&)XWbo2KsV6epu;QMYa zvSO#w`m@&4Q>V&_WU?Osi;VMh{pF4Dj0*$P*P6jizlK3^jVT-&BEv@SxP($_FXLce zGm7tT<;F!KkMCPVyLN|*Ty(QxKAS{?gO})nd&baCe9I>|=sx>5 zwDiUpcshG8T#^_E)vHI-lVlFV!MY(ZbjvL`uf>VJblji5XFFfC&L~p!$Sq5hbg_n> z=rKTRKlw?|J&oym>l8#=26fQw`@Vp}{9f?-)G;Ew)CNyUsnUnsuhHmTChF24d{aWNvinA#H~WFzy$o34`DQ!sqt2X)pYoRSu#Z4$)7Ox1fjyHSF+kJ!=1)j4Z4|aeS~2 z8GZMV_|X<`^!2eSb>T!kBb^gKRYZz~>CzXlz;+%L^!O!-R~&})EH0wMg(I=|#a85< z`y5x77?T(7;Y`Ei{S5DG@u}e`dsrCWpa@ZslPdE(<(|+Y=op!ztqP|5EucDHxxmZ1 z=IDxliNJdOS!BQLA)4t{Lp@vLKmr1KnO#j2(25=D*l6==DhZAf-(Ql4i)9j-6IU(@ z8kD-K^rTV*a|#?8IM)Ly4K@Vb`EO98+ytchai#Ew?Op1@mq(!Nb0{_0_BDzaoPvu3 zYSD$ldE)DG8d(0!JgV*fMe0G~YVu{+1H|f=0nZ_2aDH?GYEuow5A&MQX7fl~J|Z69 z+U-mN+be`a%_9Uh)G1-+^~2QS(}S27lf4<67j-D*Nfl%0`k1t^Lzt0Y)xbR`Z#Z8i zoVmN?EH$UZjj1~r1w@a3Q%;7>l6mEG!z2EADH8a<7ki&of)+~o`1g?cINL3d=$Yw* z-cPQS^|nt)D%D9SWBg96AVxskcnTkjdxu6v=R!Z*GsuGbPbk`Sa1%;`6q#+b3vKO*Lnhb5@XjgP%!$f4@ORTU zp;~`!)!K_IfcSSlD)Cmt3W07|B$JBEU*19eXD^_~H)aq|p$21^a@R3`N;-0@zKBBX z#JDQehB(+v5H7#&jPAYLz(}`5OX31Cfy8EcxnwPL{&^{ks#e0kQ3(!u@QgI3o)YTk ztE2W)VN}kjb3&q-2PjGfkh8A>v+YT6(#3kM48yMQu-@PsYNqlsGARu3M~}n$kLX_)QWyv)?~&QYQf1c z#_^^Zak467PWFBiAH8TQxL>jyecpQky{0#l*W(8<`nT?jr*`^LnHm{b@XCWU2UxQE z&I?F;h6?%EDPU`s>f@l$n<#$(kd7Z=`0TC>;AK@s>ca_>-gE^0nRXRf8Y}QpWs`?fhwnLWm4$fISVNE|vg2(KOWLYUp=Kw{Tv5!tUl2wqkpoV8|rMe43>94C7W4T@E(QV!BZ zp^N>f<1=H>N~gzo_?aMatjQ$CLBN7<>EDq_%Ts**Knw{cSz#w8Sn3DVM9`v7>6yyXPHvtJz1YxUH5_ zjX8;aKG$VTTXTi)OKuCzSN&$nyhnl8l~X~m-fXZqmOJ17GKX5vH<4Y+zwqMc=YY+; zMP!G=bG-f5Eb>Cf1jy~03H|48gf|Z7lJ?0H(efk}sz_M}%ypRq?v_0TFCS6lskDe> zy&BB8#9^ZKJBwU3NG4BQ7n0hU(b!$i8d}d-P3mTqkdU4`*h=>pK3ycnBY)`N$E%Ep zYO)X6IHV7}Ezl-cT1Dh_T{Q^nRHkO!ib9HyBHpLcsUa7ssUHH-=O%th*e_I-pW(hIiLN~Yq?k@G_b|$KSFNfEE@58Sy4I@gk5Qw>UmRdNikP1(J%V7nn)5p zqMZD(P9%!kP7>YM(rmCs5vkwRi!<(yfK|#9;5fS;DoME+Red!h>$_fJi>C|WI|F;V z($EpSw>U~Y%KAaA4YC(cjW+{zW0MgWZv(DQRzx{7j-ow%J;Z+SNwR5NA1=LEL7feK zk2<+O)xw&THVk zarf|$;;B@7yfo|8TTetgt;jf&Sn}boDOs~U4zsVy$?mro$g~^#h~Ll>G~HK@xUCi9 zk{R3BoE~4c-pi7GQ#~2B-aG>CBq-1oi+>;k&s0rJhzEER>bzw=zWmx3&9o`&P54Ih-L5{|K6bF(K zVE@IDWcj^BTshnruXX#0Qv0~~7Bge{$TxwbVG8$ZDYnN=xzk8K#)3vZF4SYJ<=P z-Y`}*1)P(w#=U#%sfq(Wl)7vX-h5~THosj&)m9y%4(H_HA(vC|UFCj8yQCQRl?OwQ z{0Ka0iGk!gK1FdS(JB}!oF^X+Z|%<}9Y-m0uOpe7b8!XvV&0AS&JH7!3++f#{RII$%?i7|zCykl^^!f~wtfMc^s^4_+^$771elR`hIW)+Z8(*~Xu_O* zh1k(15$;s9p@S8UA|pl>Jdjq#VT1A|ZHI-9p!O)K5zWuxcyO8mZalCHw@T^De7J!IEg=&;Fe2zlix&!($vLa{}T zXzGZsM8DuRIjUVunv14VNq)-w8VorXeSqlwnas{!sK+LGKf#`jDzHC41*A+V0I|YS z)T60SG#Zr2!^_Kv)a>`5bKF=mw_lcoMbE*9><7sG{Zq)tDZ9zN`O)P4^btgOcsN=8 z)DUk`3WpO)q{ymp6;dXcfHx;#r1dlpOMhI8pEWH8TW7`LF^aPg_BntD=u`OahhYpQ z8xIHfNHfbKw8`Pzr6haGC%mh94G6s7j4#|DAdIVsI3-|WA2AFh$V^~+$_}vi8{Xjk z@z=;#_iie#co{p*au}QNb)Dp#+IDX$dAT@?oJ%}Oq`l4(t91!PJ7fXay5uk!rmfGK zben;*!V@=vBBG?1^S6JD~l`Ma~m9hSY4S!B1Jq5m`Z~JZ{ z+;oN1em{+S0uO>6C&ysDOOa&M*mq={b`}|fCy{y5d+?LTwxD{dEIaMm0unwlgG>`$ zCWW!{$p_{JuBsSEM!hqFTP}x?1G@9Tj`=a5!Pf}(Ezlr0-y~7F)=9{tH=MY*k74b{ z0@h`tH#sVf!(U&O~6^f8E-tv_1a|OJ<}gKuDmW(ZPR1#t$W0tXlP?E zMTN4PYnrfQfez^|QiWZZyT4YR!6CN(ZT2N&WTc6B zT>+V$cMzMT*a=fCC*gRTtK{mku_RqoCRBeDgreC)I*oi39}s^S%NLthofo+`&PK9#~BACf4})*V z{qe%LVm$D;o-F-aP0U(N;5;pwd|B`o2Q19PKeDHjw5pyP%QPU6z@v6 z1?zgn%$i9H=qc(gpfm6TxOyGJJ=rtide=6}`duoF+3JI0^GwOBQDMNN=NkI-Xd%9r zT?bH)73`}|gbBCwp~L-CAUZu5eC?El(RcHK)4EK-!eu!nJ@)S&stBp7=x4^>frg;BLzfbV`S zyl-PGu1fq3yq-PBPdf_XPnD7MJ^L4E{s%R1@nQ^&D7X)d;=-`a!)$orWDaVOsY1(g zc7XHIEnqg42QTbwLK<3?D72!PS~|TAtGLDkl_xiW`2#z;XVPJ~cV#Ku9bEwNel;?s zJOo~vxE`;L+X@${7h$`i=|B?{pxt*nkZs~7P^5YU?ab1|yJlk*_6`63bpVbW9Ehz-rD!zjH1SqwgpHknp!b41Sg1?E)fZZU)7b)Cd$Itp z(3wpZTGjx8-(a}mldI(2S;*#fB-Y_L(92H%qZ$T-QF5uUPJSgAx@8D`#bF`5M_QoN zbbH!$_(dXX`VgJ13&du*=1@9+B(d~r!0X6yxLqRbte5sbuHcLy#+3Q91g+vw^^Pi(&S8FkJ%j@bN}%E}CE zXLD5t!+;ab=xB}-esH-JZm)2lVeM9Wq`d{2c4r1Ty7e_l+apc(?>UQ~IPS(T)Haj$ z9ZF>Fl`ew5X+VdLPbl%THH`f802D57gC{RV(C!}$sMDVZfiXVq0+o#wxF#};rQ(;f zmdlTUuP*VhzFLbu6LJmD?5}{K-%sJL^;dCpv^T+HR}p*ba-eSX3jFdI2X~!$Cn&VF zW^R4T5Z?V2i*sd8VE8o^KWiLKov@20rN*OS_;-YM**!pmUPUwNqZMJlWiNZ#*oB?& zTAF-G7$aO3I})#~8cO63da;_@hrpJ(Z-svQbU@Hf2FWT$Fqd|$z?G-Al0yb#n9Box zsKdqvU)b$VyxQO3;OtRUe|HOSNODg?x#v;r%)clpG8cPjMhHJ`JO%D0Hdb{eF!-CMD(TJ`!cO^ipSt_djD5FN z59m~x(x1G8*`>)0`{qXw>-bS#a(?IU&CB(SV4?a}c(7^>9AceI|Ex*4~hTTY?y9 zq%a#AMUH@xOK;N6VzepEzUbvAz*E8V#g(uJ|6J0{}CKzPah{m8)yAO+FFM6{WpLa=|2l;w`|9) zd1~nL$PqAjZai$6dIats)C#({C1C40VQhnqBHPd)!@3P{cB}pL#!;*2%?}?y7n4%3 zwXO}$*nNZe}6Pq1*iuE2sRxtVwso}oY?)fcS<#jm`Z9FT5Hch+( z8$)AZ;(bNQIhkK0!6p-gCntL{^E2$wo2$RYvyKt0^$>z}s=6di!I_9Q9RL${ETH=@ zpI|$SKe6uT)7W(ZPuaq>GB(|7fW5dXkG7nY3_r~Ef%{gz23^l>aQ*Bgyk%b()p@2E z?D$#&u8)|`41W`fmsfhj;GdtFK94f$@0SH2rhObOKWY#>IV%zzIdYKl_AUbI_tS8z zqnyb8iH_(+^>?~7rkW1QI7&}%+yUdG5Y!l_Oj{r6gu(O~_# zPW3HHHTwd&I;4nB>(mq7m- zkW{=_@d4=@{utJ{WznCnhr^hcg}|)n2xaJU2A*DNLmw-vrIo9iKxI%CIO9GG4DWHq z6TT-AliA85%i5>#!0ZCJdX)^+|5yt5e_DdZtn!9A4g;X6MZ@vckVxuI>p5^UB>*q} zd6ZOET!mE=$4l-X^c$MKe03yTf9C@p=P;ih@zeu;)h?pa*5?yPz1_gj(*+0)3gw+k2ur6Q$=hZ-B%wxr%+D8hETy9N^&!w0w(z;~TSM#oJR&(K6=h0;eYOZKosrDEYuaIjGQVvFQm$B~s~_~T?*aznfWbS?WO zoHW6buGZ*+2fpeOvuB&>4IWFt&TD;Clj~qng!)<9KEf2$ISc|Olqz2S8 zwFEATT@6QO=!=~8$cPF$qoBP0S`jKt7OB)77pVrEqjx6Hp?{wWqYtKglE8~QDY-r= z(aMBNAj>cii6YzxtEW%LE|aEzL}kGcV?DZY?*e*m(Iz@?_;sM0JQUcz$wGJZH;eLy zwbS-bji8%7Ku>nXi1WWm<15OG~5`eL85l1LJ7BjKkpMi56<7u`GBe4wF2i4g%7+ z?L3XGw7*bRADpNNty(Si;F+UbC6f0BA6*X$2d@dmnyQE#H_yUb zQ%s1l%>l{TVeLQ!wu)9HVBRfkq?UtsJL-|($Eo;hnmpDS#Qj#!ViFN<5t69~x8jZG zhJ(z|SWsHK9TtBp!q4WefhLO`*@Edqu<@h|;%5Dy*zIu%wKC2D7oN(-XOA6|?CYYx zKhX33FdS*IgZk}vki78-6Px|o2W-x4B@W_ccvZ6uiLYMjsI%Y|aj=^$d9PzH7mQE1 zjfFuj9mM}j5~zL`2&TNShGM%tMAVx^Vz0H4DUr&=OZFjn_)eA#LM~$4 zG#I~Vj=A9K`#)@DQ$7^ns!kc|aZ)OD`T842zf~tKM(0Vfs~M0TDx|L0&XlZo%WXE; z_Gb%j&B!JmFLRjWsV_;2hBdqAks=G<8L`dzZp4N5AjRWqiPs|$N%|5iZ1!CY&nxJH z7&Hk?-#Qw!TL+UZhvpK?AEi*_IUO(8ssVSpBMH@@jA2V07@i$ZZZ#X>kf;6V*5-B4 zNjMy9DTc$Tz=<8GRm6dnH}UJReC+U16MHod!+XQGk*Mm$Ma&9}|Lhg&zh!sgua`kW}=k;x?f7ZpXIq z2rnr&fb(nh*pEG3c-h*wK=tAkvTA#l&@$(pxW}^sbU!d=0uK354xjrgw|hd&H)g$K zPv9x1d_c{*WT;mvMN~8^z$4w8YOkMk2iZnVwZ=*2)niHg@R{-zQ8`J z>tGm|D02r_M7M#-0Z+hFa~t--*2e@dIYGX*OOuzs9^&;24dM3|JK{DAlH!$_+&mPK z*SGpeO4JP2#wU?b>i4l1^AURvS0SFucY#3<;;0!JE6JhRq4>Z=e{7=n4*5JBkARH} z*-TCg6Fz)n&i))nJ^AR0K8#QXc=cgY_}Y$5oMOSQ?|(@uW2(U+>(^kw2H?ZSEu>R- z7jf(H!WU_h z2n#)wrSceGetZ}_Zr;xBGSXzfHdnD4nVzKK*GzKx_f~T2ga=D$?qx+4v20ZHOxSbp zAhhng2OBc&aOfKYaykrQw+p)jMzU(~^YSKO{xg?MkbMAh$Bcwpb2gBnD@w?z(v4)b zPA2Kp(!(=9UjV2e5I%PO2JfI_w+$h zc3D7t@kU~09!2#|J%Vn)-@vnE8hm*Mpk;$yaDDApu;$b=kaZ@Md9`gH_Igo_|3awq=#ev&`qvht(Z8|ukqimNpAD>+X^ zP!yUn6f!Fx8L{?`oy2m&Dxw%xf(zFsqjT&$NX8ez$iHvF^Wq9@^~0K!$sU2LHx5Cy z*PoHIcsj9WOGr}uQ()pX4|HX1fwzm!lh1X3aWeCr_=#eXXjKtv?QcXsH|cWcW^KA- z{R`MM>l-?y5DW|~9N`t+Fl>9f2`qZ>5)AS^50=e3h_&yIgZ4XZ*>P80ao9cp5raGn5>+t1Y`*}xDUoM;1<5*J+itCCVnlqFI-uaQW# zCFI7N_26YlESx+k9MqkVVKuB0$X%IM>@}Byy`Our!m$XlHN}-&RXq-0>~I8K3FCp! zgjOo7-H6ouvBoaUbRb5_z)*?-p*_z)!Ri8V`s*$z+LMe6wAz727x&w|ISKHi`E|Hz zN(PLww8ruv5NEEAqWTJ&zyv!7a&ut;Q}<35HrWo~*e8E**PcO?RG$tMa_`<26mw_6 z)-%9;wloaD8{qZ#CE#_0D(pAe0dLN!04NKD za1s1++6SoK(gU0J9;eiT;^3Zl#n2}>9bUV(1Fo{q0{Ddw9`1AqEb?Oj(`f~+?#%;E z-2Kk?)??g#LYqWi{Yyf)-@f^e?Sx4VL)gE8reN*^b=+PW#s0mXOTHYABBy%KkjBs> zWWs)J_$v7hI54>yT;EzM9QHO3XmRJ{oAUW+ma0DPQu zNtX>-y5S0ZAae@qL?~kWre3I8ITde_q(j`Ox$br~7_`rO*#1&bNfyatle% z{s&+~X&(sgX=V%O{QyU|_rvepcVa#(-obsv_T+sP_x}Kol!<-KRC@KvA)@h{CeWMA z4#On{4#epBad5ZVn${VXNE<1MN!#38B=*YyeS~gC=R-6}k+YZN+;G#RkJ_@U4SW-2 z2!yao7$+MBx$lkXyoYn>D2HN_BSk~$i?PV4T8|z(EL7NFyok1LDuiCKLO5t5_Zz<; zcbMt>5Pj?7t}$K8?8Jc-;Il1}Q9g8njGgBSfTx^b;B z0aKGr*efo_;e6XG5R6ME?xIU{uH{MklH4_T&&d#~x1S`*>zzpQx(o2`mQwotll>xP z5&}0JQ>IU!t>dnro)GF#)F^|cc(Q#So#HJgGCPn!XFhj<=f9|sg`a2PWAC~#w*Q5- zgT3&{cbMMVph2ge41?231IVCC9rAMDSlr&fmVRR&LwjCsrNiuB;Wz8e@rntJa8BwW zN}IU|M&6TSN6sn-JO8GU)TvI;t~(Q|?lFYb<;J9{cbicD;wR=+k3TRU#of2ZuRuH8 zMxoxi0=j+3B-#$BvZjtBV6`*%{o2zk@;Z7FyrZFq2iLokNx$uh+>lM=a(^nx=qRPe zUpPI@qVEQtG;X<(z~Mw}h2C&EKB=^ir?Tt9pTNi94GANxEcr@8+L z(c3Wz9#F}qZ;mh*P1SlO_R#!;mOUSbHjfk2(GN64yANfEiZ`@lD&#zwUv-EzczJ*> z$Qvr^`lTt_bF~jI6yGA!t{&vq<%@WK;#ruat1r5my`79&w1KdPc0vPnEz-7Y5gFmA zOxG&IV zg}9p1BR%)P!fiR&KQkPTRXd13%+bP*?9_ZWDXf(mF4CWP{gt-Q**p&;V=y}O$ zbmxvwbc5g)u)Z@2nCaV+rl7-coPjcQjaDF4pX;D-_X+y`TMJRkfmyWYjqmh`dviqI z^@m~m@0oOV)H3qq!z3dAHJz z4IFE_P&L>I(;vIjzedj^p;w&A^Rk6>a>8Iy$>+0hdz%!IKl}jOxjDf*OFD7MSRVe499WSei8)%+A+T4FcVl4Ue%;3PhMH-lhLh{sDV%8*tq@#Nn4rr2C+@~6g>Z6y&yof6@N<| z3X;z6Wjm@;BBz8@bc9RDp=IBbYH>)Cgh*qv;vS~&28Iki;bzwh_L zi`L*ni^gHc)8=emlLg!NcZ=}-qf*K_dq9%&cP*wMRn1d!-7{ac8}HUp2kT|AK>E-e z?tij^BssBAkR#&G)IXXbFeU&)#W|CHEFyUun3O zA4^Di^4GGxuydxM&8`_eV9eQti*x_mcH7KhfN#gcwCuRV-rS3+uE$am zz2lEPWwgwQZCmn?*>l1hhfdAFYAfanGDhb(&YE_Q^86Kv2YwG@Z>MIGbJQkg+e{yH zUf+NXJgPz6!AEdOewn<(C=IIo6M8J+&$xvoAy zla(7=9QMS-P?!2jB)OVlx}5~9lwm84as_ii9G1!wOJdtTf}tE10!crQN3`Pp4n|@x zI~9tP7g~}vn$5x~Q(Aso_&TzKN{21tI*e9HM$cJ1HJSUm=?$&&^W)~g8`-WMv3aNV~oP!7Ca9^;WaCEYqQ0CrQq&BLR zvUa&B{GjScEPU1h_4LDdGk3-<%$`C1Y6qiL8ym^Y*~v_pIV;$bG#j=1kEN#ckEJ#` zrsH!*Pg9dlX^A~oFC-mOCnf#Y8Kk1cZ*+eWC=-imgQB3H%-k(a9uRGqr>K&U-82Q0`=6AI31fQKcqk};lb)L@+(mXR;t zw8Sg!j3i4sEXe&8yP31j@z_@8f`r4*hnBkhf7eoI_yVD>^D$;w>2#`T=^M#9QPbN` z)ik|Ak4laT!{y^q?FBta+YttnsbRf;&?yrOJipM880I-(*WhZ$bsK6(jL#s<_UAG{ zHUwD|luP=z`clRHKSn;LDP@#&vJsurE5dglC83Uszi{(HSIHQ$w>#0M!q`zil$kAq zBCxkk6fPYdj4nNxkDD%5kSBpEIP$TCIk&eSA;dl zJo_WcPFRbsueamZUL3Mh;M-gz&X-R84V z`c?sX|7I?ud?pcl6m7#lcit8XcD@m+&9S4_4_GjX%kE;NBxa6f`cs9ktZ?y1S;@RN zm@49uQH#j8>ahY4epk5QlN>(gtuNHoc4Ow-&mz0E2Qizp3(4}A>)66qZcN`YW9rC# zd3^0hGcuJqTWPjsuHeBeExe_~Ncg67Cb&&~5XjE?i%r(P!p}ySBR$=5cv@36S}|!N zvgkXC1GALb`_hxJ^29%qeWs%Sz;VYhE38=>ibu)B3e19aLC>NQFk5&S-F%#k@X^WC zhPM`I!_9lx?5Q%T`1=!W8GqVDWjT4Yj10B07h==FfRcy2M7BwWq&tH`n)lF8<}}`IQzL{C91e8LYXP5;HHhYFf4aC zN&M}_?#sL?ZnzzTrpdgQ@Fyt$!MBTEN@7+ZgYfyQOW0$MmBPt%FTRp1V6R3ep`uL- zaIB0r+17K8`O#QH?6Rjw)+E0U;xuK^GM7Vm<+x&8tENHaUtEhUi>29fH>xC@J?UjA zzhny8Q};x2@7ZB;MF_UWFtRDDNrAdF`}z4&WbcL;m0B^H+!arCj1D3vA6T=~6c;k8 z#dnBoQ5t(?h$r{nD30W<-^VU7_{gN>3}zn(MY4Fk70QowqWCx-J(w?^+Cov=Chb7m zlS75k*Q~hjk{(L-qtZJ>CW_5iS^we8vc0}|%S3s0yVPmnm~ofHw`N|eGIl*Hj?j$5 zxo6|?!U|<{Wyy7X;fWQSWHRx8@pR?kR6YOS780^#$sS5do9)ibNwyHOB!o)Z5QQwI zQub8#M7Hcpk&xw{nIjTO3t6&MRNA#KDoXj?kMHyR{_wc>JagyFnfIKT_x5_du^>+( z*)}(T`7=?)7T1v{Q>ju+p~@o6m1m6Gq&LjUqD-d;TR*)7RyNN1N4Z`mHL^3D+vvQ% zr*Nx%B$+<*oW1pL6>Hza%-K1&U8%wR=T=Sw^Eoz^>vM0!K-nG zYB>>dX<^Uv=%#%a{l-E+9cT%{&?lG_P;W5G;$LGlJ0CAzkFc(uUd^_1xWUr5%B3&W zZe(|JDbSADTj=Ax(O_N3B>45D9tZHY(g{<_c-gu@uy|E7_Sq{;ZofOCua~-#l~eqV zmGDxAF5_ECnwc|elD5X;s4K0WDh2cOF0 zCKj;;BwA<+?C)Adu4nrIcx!26uTTY+tCl0QY81=;sxLU&ZU+tf#z403Aa*e9!}e%} z{_o-xcFxJ<1|=UgPQx$0+)5eDiCRcfKF=d24OMjHrQP`c?fvA(*ZcaiPshECdaS6OX%^j?4jRdO>;Mx7Z z_)IBY5cwXrL?_cGR@)jiAC$1omXtM~l1tQE@Qq@xGU6v@cSrFsm*TA4=j|py@~I8% zy%0yo@$leBoenVKs}fEXmLngdmyi?N04*2EZj3*$fW?ff!1dMsINFe(h?ncKVyW}2 zurD^Cr-2Jgmdvmn&uf6mnytX;m<+xvD?$YGTv?r5y4d2~7i>Dl%}P3Isn5^ovM!Kn z){KWO{k{7={jOVsZS>|Y?!9S2G@@JdPe=V>=dsm6fF(CrKCvGBg{QH83?F7o5A z>1w-jd^M<#Hi|pV%CeLOgH^2!Vm?CbGm|<%Hmm`3R)??yFP7rCeO>sT+FiPMogX&3 zGmogP*-w9VlBPS7U$E7=F;=#{hy99Alj9pUz$a@y;iB*j_@qbz4tG_+Z5Dsz{N!BY!uFT6Ggx3d53Mf)k(~Q309#v1E!jr^=irz3M~yD_pDDhDZMD?>V4=*LS7T@^6Re)Ps_^ z)mahyy?ZdbUu+M#PqT}c;49_YfU&)booBw|P@^T#&a#P?6s*L7i}XW$C5 zigy|hb}U3)FQrLq{43v&q4pH~!%--6zpv&;>^C?4W7}z9YW`KgN;!1hd;Y zmyx_COO&<*Ainy=)B@>OL{jk;aXr^VSi9Ci?M(~F#qeoZ9Z-au=4p_pWHbSmtDlpb z@4KBer6-Wsab3f0WyfT?>8_BxDe;xt%u3JJmfoP&-Y7+Cz=OD@%Zo(F3b$J6;8{jjbx9~zAO)jwYSkQVTaKzmavkb7f0EtV!v zU3(uxt~o`JcNvITF13O#H>?6#k7cO;^nif)QdDpa!49o9Z2q?s+?tZ5vfI3gntUG+ zes6`oh*ZKg0;8z43=ebi zM6yU1*7L|fCB1+QYL9~U`)vBX%6FEMY7%VuDoJJVL^De`e%XsVJdn*2A2e6#0x?uw zM0k&`X6DN{;Qc$!qNm~RvwGBKGii!7_K9&*4T9nSJ|iV`lG%@ykX_qjY|QrvH^eEy z1?em#TOo*MLY!bxk|W~n=_K|anqcVO1F*-&mr;zHLlUo6u_Y@sn4Jsj@FV`4^t4_S z;6LOI9`2RI#pk@qvsL#&9i4&d1GUjs;z>-*ufohH+ZcY&g?Pli272x0hRg4~#=X~l#84Nak7?Y!saWHB0PqhS7k!3SZQ=w_bFbxj+;=u%ZXuDKeW0Th?j7jrJ0L8 zahqrmHLxEuF1CVDb*~z#Hkgk-m)`(75vx#&$S|pq93mg$9GTkMDB}5E0v6w%N3T3; z%8pHlf?&gWa;21yvG^#)IuKGs{F}a$lxahJtYRtowAr4NUKPd9k4@s*=y7c5*-H+$ z^pXx%9=U$18JkPcc-%#Uq$yv(WotQDIOTl(ez}8$vD|^J(RzB$#{goe)HVBVSs_oN zwV(|Du0F+jc0LYY{$N9UduFo#-Mx*+&yLcMPuyk?zv!nsUOZ%*eNV*uxoM&!%rh&i z%quDQOPAklKVI@{!K?Xd>`#>eZ1_B$#6<8D*<_B5AU~h{K6!g~e?2avf?do5XFunj zxQUH*SK$s7JNEBeyjbvO0nN$wrB%}imNr^Vxbns56|uZz)3O|FlPH7pv(Ay1-{{$U zeQ){%Dm_2V_A?!F3UBL|!e`ch(ZiPo$n7%`w90WE&=NU{yIQr$-T*D^Xsddz>DjS_tvQ)x)|J>y{hF4*B>`P#`;CiP_NRP^#=$7Uv#b#FUv9!*3%X{<5GZVi zc}#m6tPV@D>ZGIb5^06m>)GEgl4d`R*?#uPB*Oa7g1R5)zhqC3LugA~pttmh=$(i> z!a1|=%#J6AozFh)oXD1HpU3u|O2qzS2WR`KE|0{byu73`Bct*9Rw1D#@{v;Ev_zkzui&*yA%hX)-80A2bCAlIznREQ|WxxcRy?=1EbALzSnA#$8p+S0fZl%=~$j*1aXWMJ>Hj#fx!uT<_7=8Wr zYLcA&5kIkrWS^QBfj560V`q#{u&b#Ua*17x!~SxEf;ss#*T&=Ygoh~IHFpbC4qOS# zzSXnRMJI94sRc;8?joHkD2*BuJIS_?6)2W)Q4enkqOCUj&{S1D8j3xMdam{%TbKekYY#(U6E?gmsRdjV~VRG_xVIHQF*(xlOG11ZP(oOr1qObS)Rbun?+VcpNhE`gO; z?b#@*TDq1xE>w?_+8oG!bwp~*t#N^0C^`D34|YuNr8FKMLW)ZR&@!zW#2=SL&rw|j zZzOL+jVm>P#ON=a_Mj0okj=+tNeXvTO&I-QklmdYxN^C|Jca!5I34fKS6Jsd+@6CbNJNCP0%L! zAu82eNO`rKp*mF@aha@0}hEG*?oZAajwSJ)4|Y~&ciX$8%g_$HrT?!DOF$f z!k=2>DX&3p>ddEA6j$snD(qe~r7V_0ZI6G{D0k>K9G_7~Az>e&mJLMq(ebD{z5ony z`@o#L5UM2`lJ=m7*!`6=N|Y~!w{qIyX`cc3!c3Ra2scEu%TAze;DSaX&Y-`Er_cb@ zMEzW5IKj~n`H4J&hCbX#&D{dE&dnqzypOi7^s_V8_z11MoZ3HsyfMcq;C}WxJ{@Cq{n@b+^7jGRzvgz9}ztUn%IbT|ju-R3YI zpJJ^ue27MoAu+ZWz!pwda3l~Q@d8q0vFR;3F(w>JN`67boBZ+b_5t`g^E9g{b0;U? z;W6x}3dO|Y4P7(B2Zp=qfa~xRbh2qLjG4TIj9v;LBjFWrh4DTF-D0o>uMFB~v6)_| z#sj_2+tC+;ym5QSE|kTuk2>D{fF+A>;(?7^SY%~1{JcJt;xjX)cD)Fu22&vxsJMgQ zb8?Ay=J``P<*N`LkD?DrR70UZJm{Vu2OqU;0UFpyu;GFU`oqt+DDd>_^D4El|r+ zaTGYziTI4X(L)J6G;a0@{W@VxE!TFYtX}Ox%j)M)2ZZ-i#wVAMy^RM!oQEi-e=r6< z@)STLUyiUmr%aK}qA1kre;AFJwP2rJ({#WGR~B2g2XDzVz>)^BaQLYaHIu+i?F-m} zf*OCIoOjhkP}P`;(%swC@x`|3-Y%D>4NFWJr4SM9b&a5kj?GBAE1JCO=_YRuR5lJ= zRKh)xhpCLmm5lJkF>>YmMTQv?V-B7TVD2_Fl9h=|h|jVwSnw-9lO24TB!=iwfjrc{{@QNj9&wUz4)~Hhx!aO4BTGf_QEutStWyEi?^6NEdp{RiKRXK6```eM(L0@FL1cEcW5@yd8dbuQ;_-;hC zhJVG#KN$`zdI;altzrJ0sbxA3)-w_c{>)p$OVF6(7fu=}Ayokz$(5DyXpVs=Gsd@> zZ2ss89C_c+GLrGgFzO6?G%vg@Q6t#4Zsn=W9c4B}bVq>2oBs-H|bsEu}_X!>RTP zDavRvkl30h5u5M>RBZbzxN2zt-r{`=t}Y5e#HIsxh>K9G=e8o#+Z%v$7Z3a24#Y%$ zTfsQAL^A;o)}XPImeewrhiKW3`%KoRHEHEH56a>i229wsM9wD8i=0_3GFJ;@NxD-CV{)LG=KaHv;0NcRRgnu?Fm)8dl#|Hh-g9!nQ2>S?XGrCu zS|-HvAJciEn|WQZjy&nA02>xdAQ<+Am8>m?4qlI8JR4$Ba_lg)h!mlQ=Qp70_FMFu zIBt|b@RTh1V~mo%(kR@;8*Sa!hWhXMKm)g8ICp*!w*0=BxnjABb-{<9`KqPHtUtN~ z3Oa5^xwYrOBgMt!?5`wrPSh1H_cSJ<5~75;kxS-Ytz?dd%xkinn%89crj;=W6=^#D zIh8Szt7SAm31eAsg1N^36Rio_qrZ5#ih9pflH5CPQ219NF|2-1f)30jdea061gya0 zV#4Ha=`m6hH%PzHRwmv5Hc<;We0Qp0^YLo;6SVcJ3{-P82`TOP3a7d+lEBkdkpD>l zW#7w*7wPafWxOq5zPr;5U#}a9oifBBT_cP`m=051oleEbt!~Q8E@TwWMl-vszrtT~ zGhpP|0;(heAk27 zc!}BFQ~GNa%D~OT8`{a2?ctyx=F$@|w%CdS=Z6^LW{h9>Yd^+0XR-+&{DSiHD3lJL}IwAaq*1 zvGe^k-D95yXYc#p8OU|+x&D6RSP*Ox2j-=O0pGVNtc<|v#&gbVfQOvFp4ZDvR(fPS zIM-&!)>$qCS3b~y|NU+W2R&<0zaEsuNwT;X*Rmp-=7HWHYVc!1I?__3}?xot;O4?T2f6-phWllcX~ny;KN@&1(iZwMDe4 z#&PzjTs+9Qs|T+6j_B8R7O-N1*Ri6XYc;<1K36XQ>ewrb^L71qh|pKoOw8)ekAIwG zB~%sYeK+NTOJC+e)opF?32znKK8Xt~Sy$XJ$43}!&vt-Xm2IF~#TP8;Q~_gh;@JB} zA@I3Y2fbP}k@>FetVWS-Ail5{FA!B&V3|eAO1b z{B#`r_e}r&1?ARm5~3w4S2C|5i5U3 z411~FqS^NU^)E^Bj0e&X)b?^dHcw z$pt;6vVq1qkp}(ay=+(h;n`>E^D74L1^eLWK^`Es^8@>D=xTUft(&zeI0ZQ81kc(G z{-QxF%5XW*;=aks{AfqBCQbsSHD-vT^Y4cQjjwbrYns)nE&jh4v$Nfg>dj55E_J!_7zZJ{#@V|I2rpow2I0 z@l>!sYt=s$mf2ks_O6h#?2x1aP+_PGGUxJv>dcFc63Lbz)3OYFn5mqV-8tt)>^&_Z zfczBzlHnQbrk7*-N}8Eq5*LB*ryBHc>KC#*V%M^kbGgm7PaNffDZ!B}i54kPYIwYH z(Ec8H9L?4nzbUG(_w5i{G{sy0@q3kqP1j3-akQGQf#oFN-|j?TesT*mZ<_}wn|HMR z67`12v}^VD1upbb^-ge!XAf*qu7X)|3t8vy`3ggzpkiT1G zneL;0sCZ%zEVs&pQ(yWSxxewu(-2)|PjLYw=orh~=sC~4supOHQW|6uYoi$p7twW#)Ig#RQ%4V8e6%Wlf^-W7>?MJo06GKaLIP&Kx8ws z>{u3#2#3s}$d$C)@JS}lLWcC~*fA$`b^uFxY5jA(dElz3B=gLpf}CM>kfc{Tv8C8Q zob@Ub8*tfU*8^Fk=-D|UrEs5>yT_bnM>^{I`f-!`AIFGj`#M6`>M=5cdzmn^TIS{* zW#XcoOD<0NkmJc;33tFC@vpaL9Il8lBQlcAnKB7Rr)sRhGvzL>C^(5%O}aPfD(W?D z)K+W~i8Lis0Rp5-@H4i$BT6ond2#qSGQiBZH|?|Hu3nW-D&%X^fS=m5iC|72=3sG% zGfRzW-eOL)i;A_xFfS(^af<`a5oo%wZm#wqlIil9+hEFCfxdku{SP z!g~2|@2u?ie&&WXOVr6r`BHpCc`eS!c4m4$TxP^qJR#Gc1@J@VZj88u6Z zCa0g5n2ra~Cr>%+lRMZ+OT z6RmM0=I5t1OoM4Ov*=(2b5QRZbMdQA)2X&@X3e24=IfIZ{W+VQ=uLZz@aI6f=}uUF zQ~BBGrvKLO&qGRhaHA{vVJX*Cn*|wmNCEbY}}GCDJHh>Zc`R+>b<_$bg@sm)BVkq(n> zJV{);M*n}`{BgSgW{Cq@*0LMth7K@NPY*Hk>Jk~(ByYl}I)-gS4&jJj(!`ek2G->2 z!8qXyJ>K>XzeBOu5kvHEbvrA@V;UOR>o6(vZxbsYJ(9Fh5nl<~hMWIBq{j?)!Byd9s&U5D-d zN;D;|dCKfjEoYuCYGts&oF*05X6B8cVw1xVZ&U5QX~x+mjUDhnxN(`scRDN$Fk=74 z$@8yLq~mTLmLAH*SXDfvjA%CBQK5AP%&(NSzzr%!aU))RSw%`EoETk!5MhU42Z zvB26yxIHcdSlu0)T{8t`Gwe&xQIJU+WQb(7(Rn^ zMbE+7T!c3-;=zjsMxewS5BRdG1Ri@EhFv*aQ(o#$C{kxVN_S7i!dn;NJ??jL=tFfF zSMN;;ak#PK78_IVp3EUr!jfpxK#*|d*`uNF+lj7XGx;L9h!PoaL%oZ~=*0mw;K~Cn zIOmx=j6H6EYjqCM#*cnrO3(x!S8zgCbZ;OJn`)#I8H2)1SV-ZY0%KuwhOE{pMUO6b zp$ge{+V!3_@fSZ2&05=$p`0RA`IZBZI6g<|)2UefQ7^dgVj7#T?7{aEI&kdBFcSNq zNN_8{6BweIUx~2jm@$1QK@BO74x_p5 z$Dmp9Zz8swpE6Au0f&ofaJuIS@T2!F?V)=MuPt)Hdt9W5wzwO$|3@R`tP>8kjx4J4 zIQ|qnUdzU!p10|&Y$vpMff>@>6^&SiYAEw501>`>;qvZa9JwP9B~;wN1~anMTXA`O z^!P8l?pKYl(0|MyPYWaq*BPAj z>p|9;)K&0XgFUP~ABtEaOQ=mD@zjeYr{SWt)wGe+dEo5rNXWDEbi7h1Qk<)TS9u#X zW~7^g%f<3|>x4Pn_hl>mvBVzT5-mnXgZpsX%T##ftP1mUbRK21)e}vJB~td!OQ`il zuh~51rEpxEhTv zH>*=Z3WJE`m{C@z3aPXUTkyrtE_Aa=1iHQI0tx$E23ZGET=30j{=z*t;P=U6@*4vl!v7^ z%6{?$2EEIHD31*{wkX29hFYledj)0Smqr})?xQg7eu_3UBKz#CsRa+CDRcWt_{H}C zwXZ{;YP~IoOrGhZ3S}!q9Sw$&7sYY%TNV)>_yq*=4xr_y2=tZ7MgR6+pv=l7sQS92 zM83I_E@tOawZ5l7K(9LuN5$Z=z+9m0_6+7EmgDco*J0UMAt+-LKnZ$rk&^>F)Vb3~ zsqNzXkw!}=dho6gO-rOSwC{QhE5CCvb%D#Nj2nh{$=O24y77s9_GtoHs~(O{o4o?O zk43=;n>@J3D;osv@<3HWyO3V#O(NNUj0)a1PEA{@!;W3UV8@!pP$Fs!Ycnr7ywEh; z!FnxqvAvfzkhg#`DXDPJ@^vWOc0Dq?%!4{pDeT}MjnW@$kwcIAQRW>(>Zj>TFgK2n zd78dRexV0C-<3o^m?^G17?611RY2>n|9zx=c3BOgG)LJ2BS zrV2g%qK{|v@1XUA>yV2LFV6g!1~pSfnD{&?rlMpOX@WXLQmB&UxKacK33DN^Jq?~( zTZOFRljw+-nY5~T8xoZH3p8nNx^Hzgtzoo~{7^j3L@Jk)9`jP7Ud|bt++#AZ<|Hxj zPh*y`qfzO@3v@ZRJR=a}L4DT!M>2P)kQ02%q6pg!2IJR=@=P-ncKHkj zlk?%VY&Y`SHHy>=x5N2qbIFAfO&$4u#7?bsg$3uUk=^nZ#J5_P?A)J1mW|zllGfKz zlT`$6(u^Q=(e~*3!B(i_Cr5^L>Je`@n_L_CO0Ju5>V;1$B3#74x*H z`6W@5laU_v=;>ujeb;g7Y06>fzxdef8FtVl5dM&|!VjP9M3Ywx(caa|Q5+OTb_X^w z{*x0#%rFX!48?lqYCiOLy`Pw^Ek>T5tVESZpOGsrH^{>3RCWu$6Lq_O5%qOck)(gQgO*oW zQ<@?_(3{j$Xd*uZhi#ug{i@Re+D02v9|rA&1AoL4#bD zSxf12%ae2VB~ZM+59bsKqN5yc&&Y4WSbWn>ta5Yxtl#@ep&q^VMl=-rwFZ5b^Pu<_ zMpI2cwNR4KWwIf52;JA$qU>@46U8*K<7Cn#^tCa+IG(x^*u9Uwq%=B= zD~jQMrW@&fFrn68v4VD9x1o4@1&nP|AbsE zXcm@l@Opj%W|xT~z3sQa?r2#~jAsnlzx=>Ty<7`N#pC|dex>E~F*63m17v;hBjnv9B8ej~xw1`pO~KPjhsr zdOH;Iy#VKoT}9g!e#6(=cF0Z9lj7Oc1C7lR@dk|o)QC2NRh!53j91@JXkJ?W- zE>Mny-k)Lr*`^D#Dov5UQweLaM*xjd0z;f`c=uOD97I$7KfD^qTzAX-O$oV z83~>}h@&5Abm#@a!#B$~)8y<%IrbU2ZxK*JW_< zn=51B^G%`1e?y=CKD+tQv0n(qFdS~S*hOe6Vhif}lm=^C<{&Zd4ix`dAE*tFvg59} z(>$;E>ESUSG?TNNIw5X>an3JbJn^AXC;BzmrVxOvgWnQLoF`kdI!DEj;CBifv+PVUTPu^U=y5f6I>J$GCW#s?v|`Z>f?Yae6Hm?hK^ zx%*(zDqZjj9;U<2zW_aM6OiJmqIc*wz?45i=!QlyEUk6KKSNxhZ_5Oo6zGokC)d-< zR|vuJoyK6bc^qmlKSHGp&Y9H{k5(F>rt8<)x1q%*jRcL zp1#T9MxeCutuhl&~^heMtA8eCI``B`RB0zeKd!YClT`NuH)czIh?HKuL)~w z6?~&7NLSV#Mzx2&!5>5o<%dPGujej<>oiZmkEhe1kkuZvvCtfyIMzs6eE6m>^ZveG ze@-PN9X)vE?VTi)%q5TeIe0bmLR7g*7Dd-95~eZ)hV$CsMLst0^?gqWeiq{M?+ze- z(;8&xBtZ!^y{FfWEQ6JkPvPO)E0Kp$4r}k09GD9b8Y?w}eF>T9RPi7P4;I54H`}8Q zvwZrf)C4_NdK7!tZ^xG}-i5!S)nHZbRdii18HJ0VLgG`$q3oInWV1C2%`rVq|2VjT z^zZD0eD_t5SQ`VMt=$F66H3W0y-jrG>p0r(i4W>2)GPVW28>fd9-dSp~XyBte^g;rg7ar9R_ZN)~1{iN`o znTud_YYM(QDoXxc{DCEvtKpZ1LWqv+hHk6M(9ObnAf$Q-6-~vXhJ%W9R&6ER75k75 zxhIROEOtN}-v{vB3J%AF1s_PS7b1t^-#}A#F+I9%6LE0!g0Bk#NTtUyEIedDy2d;~ zr+7ZN9=eT>99~1}pY6rI4zIDVvk?lm3_-^59W4{{1T1)yOOKl*AS-qfee@we ztMkGlNV=5K^UPqFm~ezP+ct!1w6?)e=bK0>l$ZHw^p+?ISP~D>VPyYz4H8e_hZ_b3 zk*TdDESV^S`Nw#v)06=EmwN?9#9f0)Rmmi5ZxT6-09tm>5_t@|pbfjr*{9EH5IX)J z7EQNBWgr;68DYTS6c`44wU>A*>e-L(3gQ+Qse zBJ49AfWr%4Lr>%7s5#>;u2bblyY-KCA9H9(1L zub^#)1&Vrh7zO>P02Tdp@Y#kDa8+mzR#Uma(ltkLQFa0}+oV7p_tr!|FYScI{*auL zc!qDsBti2F_UMR~HL;Pn3|2OLVx3ELM_VVZK*|5Gr79e~@_-|eh|C7p#OA}g{v2VerE{S8`xuat$V5~Bj-inhPRzVkjbutD!O>wE(7>rV5q>@&e^361HJ_Hk z#4{!+>6<^K&*Z~_S3ay!4I$ilP6r(d;36Lz6yVup4wtJ)30!D<7A{;e3evlL$>#g> z$hMnFosX~RWya@Zea_A~iQKV6J9v{X7;S8ez6k7 zAKe57huqMOZNXRZ9u zw`GtzA)}6rY$w^{>C%Yb{x6mmvBz5$38TC6t5KMtCa&^xge=8w=#J0R(UOv+tGWZ{ z`bE>X`QyN@%v30A{)6rHFc3~xX2J*G5=oWvX6URKg~pSkAY)t!W$UiMiED23Q{y}s zCV7IE`Mw_oyV`QNWN*Qe;Suck`W)H!r~qI8S_0*MJ%*je7VuwEFAVO9K^0{Y>_h#A zsN>)Qn1!ywk234%U&kxiz^#&YaWnw|%0Hn%;M_(d&j;{j?0l4@eILGau|;(~>)=MS z8aQ`*2N+B^0{=wL0VgJP5OXaJ@jSi=f2t)xk+ru_N84qzT+jkZN0p+sq$8;LvJjjj z`4{XHod6%cUx#0HT9NWdDQf5N?=RBi=#96AVfNV$yJhMj zyz&sPuA7GO%U7Tw&KkNmEsYZFyiwl2?I^cL0q$vvM7w|$LW*W|;!G23!5=fQUv3{~ zzbJ+zQx2ibNV@-1CSIMRK~gTQLEC?)z#vOwAleWItL_P-PtA|vQaK0*Zb`Aed6mGZ zj&9H#*V<4f>kcjrFQ>1K)8IfyBwReChGH(~qt1XnX#XJywF&zn?fYwyn~w>(rjbls zzpD^#j!nBz{}n90kPm}|{Wjz`hI$wwZ57x0ZH6 z{&7KU>8VZ7ym;(+*#rG}I1NrDCUfcw$skL`6tw^AB$(+rh@AGkgj+Uv;F7B!;GQQh zVH)S`5y`&*9VB|-wi-de#fifmK2d^<@k$h##f?1b)sUdpUKHV#h5~~5(XG3$+2s>O z&|D!6JsAy$0hhMnrDiALy1X_R7UY0jJgrcW^&%`bvIM2}YoQHP1n3#Bf}P2-n6FWR z-XLy@GT*o$+eu->x^#+243@)ybQ#2*WeKgP!qGw9HrV~(87^OZ3%mUE#{(gnFwxKm zmXtNpd;=xO^z|l${*(gW1_5}<+ZZ0X`W;-d(1+ij&P8w9&O_nrt5A08P58oS4eie1 z{yLK&j-r(kVD##%bh+;c2)(ZX9*A1P9j?Og=}RTlGx-7Lc*!Hy+RgOdPFd96(TBUv z*266lPjR^ZYYaE9rS*c+uwPOw)^1x{73HSuoL)(#)|lohV6q1>N;=JNma-osuc?#ouez!+vGNvxSc`vFTaAF2IO$NmI=Bv`3~M$B1Q$%(Nx(!Ip)pMMND2$Ff%+>N35nh z@YyB2l+;59CbC+Kd8;ZxaeW`6PgO?%sY*p85JOT zqQ0XVAZX$t2(}p{U02;0jc_GgyhntX+POh(^+NnwN|dQX`poiO>5#jikXU~6$I1Dx zVV_zvvF4Ixe%FhmxJ`SA_XZ)H=sq8BcP=HRuQK4acg?hn-EPQxK@*yr#Ix2faK%{< zIr-;~iyCbm%$b++!Q}okC1&qn5lMM?51VF|;-~A(naTaP38$76iuRUf!XHCsl2c#m zK{g+4-QAB%bvto@lx~xvkwueBJ6F@ToN65Jkxn-5I*GGm^1%V|2;@@cfNt)7NUy8P zL2nAyAa7qmQY7k+7jSSUyClzG&C3JG{ihzKdtQNYa+70%7Tm?AqP$G?fePXq=Z~*9 zHZr??I+>$~_?ZW9IeTKKBn*D7hD3A_S@LKR=APh#x7QEhX!X7L8-E*C($-*39#Vp4 ztCuj@-785!Wg0H(ok!a0y_x16kBEy?37O-)4PD@qLx$FADBVMv8HLB0Ri`AFYx%}V zbMZndIOh-NzZ#io;qXK8Yg3(-3e4&^Z}A{4$JitWGpAl!GlCUX%uj0v#_Zo_a%SHt zYEhIk`&i^E^3V4J{&&};$z!QylauunQ>-J%JgYy8``T4V?@k`FL)9NmO=M6m8z5?u z4u+evdC@9vQLOMQ23^(PLj9gw&ko!+hcp{Cpr^KeBzjX2Bab&Sdk;Ls8@?DI%lDqt zrJyxTM2;eLdfzpYDVtA5IsUVPep@pBSc5np{Q=!snsA{h7n#3t4vZ*(wYl{&okGL!Byk-q955Crg`zC%l6}Gj9=>b0xvSyNP3-I*1C0 zB;9j2Q@cwxa{TgrFiubf$=N^U@C&`gBRlUP%XMN8y&vLzEX8bp9>QG7 z9U_b0M>7w&4>0rR8<6LAx{U4Zeo`oZ4a=?8VssbQFrNfV$ip~&c(nBr{F})GJpvw+ zmL)SJ>tZ6g{auXluHeKce~;spr!DF5gWPm?i9Ar2U52Foo&{Iicrc%Y6VXSV;K+J! za-rCW_)RJ?YTJ9sjvyYQ$$CUU%oP5!?F%$07ePsTJkZ5{OOoZNLo)m$aagb~GI6OCn>{6%<~{P(+2#>oMC><&jq=9&S&rtZPFX-m<-K3*oi$`(dz z5}<2v28I35phh30pqt8<(W|I?u)?as?9W!4LFY-MnB-t{9~vtQHZiulA?mY1|U(D4OF#! z9m;TVq`6{O;*2PFC~Wb96(eg4)y%_SYo|K8cnBb1xg32yeG;v`w+Pi=z60W`qQQI7 z571oK6x~zc;oN08+fxo>x3!1qW4-gST>Vq*QQM9J9klM+HHCNz^<5uWCr!McP z#C6)LU=Li5I9!%v4QSKPFDPYSDRO&OiBzgj;Jn8*Xvdk|l&#VtO8M0*bTrfga;y0;(wI1`2DTTH-5Zw?^6XIIee6Widh+aa#ePvnv!nRs&@|oIe-Lqtj7`t zdPMA06kTA_L1%RBVLv^ZLALCZWz62v9$BJkYesM5jj} zyX=KTYJ(j*uv8oItqMl0h1t+eCWy#xx&kyo4r%PqK?^juqq6BUsKnk9@gMyN6mv;sRiVW=p8WmfTqiQOUd8ilf?F$ zJT=KgqXLCKbkHXSsBGqj-N*}4TMpx`d<;_V9zw^;hS6JM&xvDLqDs?RMx>K-e!NO# zwif22-Li*~u1IfGCs}C3!Z!^B5r#Np<>8uct78Zw9mT%tDm>h z_Mv`+UAz?bZh8yPWEB9_xt-|1nVVGYFH!31U5nZHgJOp-aOL2SxD_D5hs3DfK}8B* zEJH_Q`_O{iRP^MSahj@O^3ff1LoWh29G;g6rXG`s>srypVmpt=0*=hD3E%$dQ|m~ z2c6gD25Ra`DF0+DR03hR?O`*lNxzE3IR2XxS9C1KPRSZT0%{1*Q9ntuGbe3k_W$Du-WKIJrv#= zge?4TAt|FoC@}U3d!A3nKG!oEc_uI6#oi}rBtHqmzx-{CKR5za=L<4bhU-yVUkfAU zqe;1hJVyFwU!V&Q>xt!_F8EqD8T)cL<{}o1K)Uz~%Y5!7xUfwIIos#a+V{qhSDZGv zwQN2q%1=Y9402J}LTQxrJc04NqQk^|UCpd?{Q>(!MPT660ZtLqLPlki9n*FxgGnix zfHH6NC?9ztWS^PBtm-<-xNfs!dbC9;%g4HKEKz}osxBx0$J2SoWBI=S-<}B(Wn@KU zM(+DM&r8~aQW}!dP%5J&rJazGQQ6tbN>oO;uk$!XNr^OQiH0?qTd^MabC54WsxUrQHtnltu_Wm0ccA3oqyzO}@-ZLT+@A`2G+D5k0 zeY?}?%vVaagC+}MR(cE10=|o4qC$``=Mq{^xFvtwne4 z%%MK|JLs-^O2powTvT6H1IZ>CcF^G|jtm-!CceE3=XgeyVdO%(YW{q>yd|GT zhi+l*%Y2``NiQ|9ZCM7|@N@-Q zRS|}`$rI7yCwsx^A@AW0>JUv%RK$fUJ5Zxc6binwA6x#agjt6RnB#BWG0zS) zF=N90P(=S>=1$p4yz0dcbTiYC;nr9($}^RS?d~_s%tc}N{v9XmU2)pp=S#Zaql6FF zmS`yaD7Oj;1k%`cxF^?q-yE%4(v6(b&7fhDAyh57N_|#XDYLDjedt>+{ZVCBk@gyTS+$-_=r^{&RsK;M;djRsUJ7rpX>A(AWW) zzHj}fpVs)J#&h>>*LF9Sarc81QNR0jk=M)~&MzvPdGRhs6#Bk{TO&0O?OAb4pxM`q zUQ8K|WZL4mm5;7+_K&26ojX=D);>p=d#QcQtlgd{c-VV1s4xBB8X0OcT$eEz|Al#< z}4U*@j5C#BcaZV;yNa+RyEKUW|=}H$|E^Ui>?E%0wTn zes`WUqf&%w+%e?-2xBX||FOUA{I!wzGKtG~2aPIO&B7GSRCiqIHXw&CY%;xtSad%BGHoa^mR33H) z3$L59wpTYW5xI6~8d8V*pNzTBmyEEF`fR4#=Qa1(RTkw_d2IGuN)*4UL(u-t0u_FX zu3=V~aJkFm7}@oc$=woPv|_RXw%KrFlq`2lV*8iO+O8| zvO*c?pChO;-i5PLFXQ>!_P8cJO-L5mFkcqvA^$pE&RIB~i&uVulKal0KW7`Up{hDo z2;MBbH8P<#SK!G_lPlrUBYq>hRf_oJw_I|5pENQ#&@1{azg3{Inb+K3#5l@5klFgZ zKxC*~jbtU|k@pcTPHjadK2Y7xwB25WHeFXFUg@`l`tBp}`<5qYe81|ye(~C_9yG_z zPEaI}LRickIS2Znz8mj2-(*Ws>?}2qz>iUO)_mc)FHI1-ECp5{6QbWc{pt3a9va=* zM7f)1K#Gnb4f8d~t&%5#fYK;Y$#6X~+~_X!iyotoccjUYBuPA!n~PQsJBPNweA-?9 z4l2*&kYQmO^z=>>^dVRS-~Qc6Ya72(Mku6Tx7pCk$7WDXsE3@?WAx<&O7&l^qdJEd z(@`#GAw(^fOLNx)=2$NY8f(lw4)*3IOAat^e~Xyi)?s`$VkY!fna~!+v+%<{nHuKo zpqGxHK~IWG@RWJg@@)lr z!fG~h@$TYwg?TZSYOf(|wg%og|2(IiEQO2sJ!x<97x*Hb1^Yga9hCApk}DrNlP1}zjC8~`=G3q)WZr5=n))%62&ari8$Zt{{y!_xtnKn} z`BN%Yot#d|yFjG6G=Z2sWnopW3f8w+$)Ov==;5E=p}FEMY<(?;d!mL>y}n-VWSPPwR7 z_rX@O?%@5~&^T^4d3?_ZFIwHlYt}!|(PJM{?@NDC<m>;dd7lqw4q`eRZEs5!$8Yk+z?i-94bVO<2etofB}-TW@swaRvIaei^>%t_g~l5hU)o zMU>ax#QK>6dUW(pQ1^-!of@?s@{(onF!fNPA*Tz9@qa`z(oOjNHx>4&Q!{w)8A&sT z&%=xLb7;6&G5rx)L)Y&;i5sum;6#HVcu^}tLe&Pi8lp>Y7sunPm!{+2@2{dN8E5*+ zZa0m-#zEsQhB5~qz>(z-;p2EUs8n``^@28V?k%B8Ul0GQ`>x9#A@L-Ss8#*8_wFyo z%NFm)hi)H6i&CaDO>1i5u8JS*(dW4yS=&M3!6#V$j^+DIP7tK_4!B_#ao5jKVlqpa zE>zK?$0DXvw{(W@?L2^4JB2KLI%3{-+DS~^I1$L>9J+;sd&q7f~2o}B_WCh5F#gzD|&+aI$;Fc5a@Z-%H=vK)CpXO%7c4^|rg*ixRV*=Rk`#>fId`B}6 z-Xq1w=Hid7W0~XM4&t-w%h9E$i||~=h^?#$XXoiozy~@K@JU-F*#BD$i{%dz%MF@% zagqYq4%HDw>r&jcUW=UTUJ57on$WS)0=}+y2b+yHtf#mxd#cqO@4CXLuh`WJg>pxr zRZdfM*Q**$v(;cnz59+Q_ch{GGo{!iz5_U#pQ9Y8)yBmqr$V9H3+R|qi<^@2vCv2p zE~GBPF?POq>NaWgs^tKkyL=0^QtN^8%cf-QnR>GOk{29X^oHL84;Mc8UXEt%mcUc% z6>x9NI&|Z0D}J}R0qqYv45!g7a9AHozQ5Ru(`TeZ`YCnVQy)&tH)bK-H8xDMgF0Tn zQiMBnyzuTceR%TjDROj{fOD~JqL=%>aqEgNlDk9>J=y&pyEqxL7hXxT+24%Vaupj^ zW{nAZCh-;wnjS@6H`VC*{V;K#qR2+&w-SeuBWUvbX8fsWHB7yq3FdA& zRKIL5_06`W&W=8K5zl2fG?w6yKy6-Q9FAweNfchU1OHYxz$?e|oE?qXpuOcLI(6a@ zp8I$u%v{zEd-FFFm2pXMeZdq^zNkjLvyNi_WzyK_3(wo-Yp-7;Tp7b+S90@~G&@(> zhHW%*Wi#&e!T`_nb~|yAl(idR)u!_}w9gmf)9#|}%Pnx5oISDaiN^~j%;XZ{Y(QcB z7o5h<7_kE=)QC70F|9QVc(nw~NFVo~>X5l(Vol?t|^)3Tu@g z%%a8u4^lWP1TFV?z_7EAq7Drip2;7JV~5>9hc1r7@B1t;Z=a=W23%=_MJ*imZNO%& zKVkZ>V(^P8g{><0;eg6@=;<#brg`hJ`j#DFm3t0lIH_XIX}srpwLc!s&seOp6VRkv z9VqJb5R)+zP$g1hZ6gD5b%z74N{qqQiyQE_T{U>gAF(#+nIA)c?gzPS+)p znXz>@8K*UZZCsRD#1s_tV+Sd>dCId+ z3&BvO6*g|uf!NP#G(`41xE}3?5EBfaPZ{z#2O_vUBA+lTQ^@3jm*~?$MMyCbz{lwa z;Kbr*5Y=A|0`wbQ+$skHWA!-i?Z3%8|F>lPkUu0`5{IKNbEuuq724D_K(p@*C#9z@ zlID;Ch+9-nX9;f81>bMe5f4m6u})&xuq2SHkExV0SSLe&lF)OuNZ&mj`(Xd3+9Pz4FDN8#sYA)N3JgSh6!XhDKBEx-O5p86jF zrybd(N?{8ZbMqHj*;GRQ=-2~u<`dB@v7_hbPoP#ZhSY7a3@OiXBlkl4nZ@gB$<+y2 z;6*~=`MpilZOl(dn{*x~D&2-*+rl7nXE=N?F9PN7&!DwyGw%iKAuB!1z=kJ`{gSnU zuiK_W?fL}RXJ=m*IJl_p+ukX42?0qk*mM_uv_68W@K@a3Lo7VCH-Tgi4eI9H4|6<* z(dldT$fi6q(nrsd*VhUmVw(ZI=dDfs`xYbI<`4RPf$&q&oBoQ|p&8F}pmEb1FbE8$ z_rGP*zgFtt_hBAtn75m(_1yqte?KI49urYn)iyv58_3t!uIQ1!DLpucrEm6*Aw~Vs zaIz^4woa|(mL%peANaZFiXA+g)Gde`uO~;YTC-gCta}hwu?z0)tb*&w&gk(ZUA)`l zICE;@IeWK7gxJX{6O)oXbVlMnda-9AZ3vkS6@p|G;1tHaTY3eo`P#^l*OvdPOC>!J zK`at@by~|w>Tk@~*z14Obr+SN3EcrF#96#&Ify}D)aL{oG znx9Q0hptSa_?Z^%_!AEUTV-jG=PbyQy9g_Z9~2EO1yffONIf&2XD;PJ!qc_zKDvp- z9?&66oCf%y#1)jx5zxiAzQEdfe;{$U4jsQX6*eWFA?xE;lTDfia7>~Qgo6c8w@`+< z-h0D8o1b7e=KUYD9w5&o2BaYzPS(wZ_}FZ)b$?5`^2Di!GN$3CtLaZOOh2s*qL)0Z zXyB$y*kW)4wwuR+bcY(AyiuC7{hdX+Lgi?3_8s_r*_iH4l%_!=`DT{vNAUeQ9+f{n z#JCs~5KI1kz5gOk7`pE?(Ra)sQWwh5@7xmZ?Kd%MHp2iSHXFjhIhpX-VGxY&G{T9T zWLlPXj24{Fp~Hi-VV3Q1I(Wy3q9#jXGx1FAto?G}$;8prcdO~5J|BAIIloT{tb-D> zN;n@8T{{q#&LztdQQ$x!=}1^Zrtvv+E5u@v*@qxBa5n<-4&6eozgk4&bozy^=k=s}$ao|Bm$^GnUq_whuOdsYwK zbel$;CgzdHn!0p<=osp4Y6*w(B9O!FI2dX=2z#EL5OqY~A(d8|%w9KHqFT~Pt^y*c zFdOdfuLKFDd*Eo(Lljn|lcI+yM0w{jc&X09mG`CK=}^Y)au<-G_hX4(Pdqo}jvX(? z(I8HX6S(ZSqr`8=D6&0Z1lVTI1@DB_sK5-P%eQZmUz^S_I=a3jTW2n8`xK6TYdnRZ zv+v=olQJwhJ_j|;C;+KNK9FMhgj0|ifuE0WL9XSCMb78!`T5>W0x=87G`)jlOOAl) z9=uM5@*0TJP(I1iQim(^bh$|ft%?7=R#CRIn(&r&0=G45Iegk_4p)^k;FQb_@DrB8 z+-a-Higow7h>5b$-%%snUnEA~Rn7qA%|%?>>3VW`f+ob}P6hv8k;rdOFKVthDlCi{ zgSq5F81LH(NwW@fnk`+Z=}8S4-?suyPxC@1t$)bc=n%5%w-=coln7nRQn{jMag2M* zK2EGU;CU_!NpD9i9L;-$+B$%of0aXiWU9f3Nnc5aU_Y^& zaUK0yvx>w!=`k9IACeQbIiP$$yAkb$ltzz6MVf-8r_UR_1Jd?PII$yYJE~kW&hmx3xi>=JK`Bm_s(igt$ACKOc&&A_!%wvw<9tC>_nmAqW z6i6xgOhUiJb1$DpkQb?e#7RdBCSEp&($l}N>xvZOF!Wv^e|HIUDCBeP?vuMvbHgFB zG+hp!6;;B%eeUoxAd#;dmY~FyE1=K74i;?70kP3#VCGxGv!`UKmDCr2Bxk7M+012b zaU{6do4j?pFSOj7NEYSfB8Q1~Tt!O7`O_f}YwA-(vqL-_XW#yi87rI~9E>k|v zjIx#%I94c&>UJCw{k*b|@$QZiy=`}!_*4yx0(3isu~nudF}+6^Fif6#AUH$H%M)t; z>yytdor1Cq0{*cN`aX;x9s>>Xr;Gh)Fc+xPgTD)H{vv-8>&DP^W4pc?2CUtQs=7$9m zubf0pYUjEB`aovg;`Q8Q*Cv5$Y6x@3MN5>k>7>YXk14OF_{F(@u&LgI%!G5K{|YMG zdpKitf1#|OHTN~phHwGzYtC9{2wRC3dcRXzG%#JjZE3wAD(e|9(3ckkJErXos$)Yt}q!!EEdo)a7Ia?$cd&`m>ovJ9X-gv^^ zCQKUa&vjI%+Jx1%pNr=#>!%2G!sJ9vFQpl$w255J`#90YIj5L`lv#r3<&A>Vrd!B2 z;m(@#1t}9(-#sr<$Q#9cfArg4ygv@vwk@cQ(0(dvn|6&!Qq>g7Xq^+7PL>h9deJKi z%pWUS{yK^{{`exa6>H$0Z`P=_yt_;2B706C-tm@eUNcJ+t(CxqH$C8nQY1vI%bMEu zfosgqWvA^`<@;(LhO}^HpXLZV8+-)PgMURkjD!rkGELy`Bu}z>B$#|=dhMw%?p)o5 zQQRp$x8jNRMn+LWp?2ANNx`2Ob>ZferOe@RiK006rfA!WEuz88Hlo}2O#m)+3-zti zs-15y67BHO=H$9fnJ8vD$f#WqCd-y!Q`_f3&bTJwn6cF&&#=2( z*@a}$L5Ee`rs>X5uDez=_hba4mmN*QPd^eG%FO2U$Au8{wXwpL55UP&!Kn~^RNGX>w+#M#th>Yqw&H)w>RW@fPc+xyNNa32Hiqg`4&NL<+|D>{X0Z$ zw6prZX9Qau=Q+3cM3JtR%uJ3#?y10<&4<*4pjSqerzZ*`Lq7?AP=h=ULufk|Q?iYoG z-4)zf+6~rv6Kax2-kmt+OSQ-)M_VLuwN)@&`%7)s&U9wSGI3^^{7AC7Je8SIm%(M{ ze5!G*9?9(=p2WG&sH(lTFoQexy1g17e@ITBa}ky7s^MqkxDyVYax{^CT{S$kS`LYw$c_r_wQi{RQ$OV)k2E__ly59gjdhOSL-h(q0DLhht7 z>LEu+>DQB7so^frxUvrn<=x?A)&dC2NrI&77h&Yqxr~Og3mUE%gaSrNfxW?F66l;x zE{xhtH(EK+)#}sfBV#8%AA;vzJl;%V_;r^c<$Kt}6A z^EOEs_RE!S-E2qI-LvWOE1O~TV?)sY6AljtPBHt1+wpvv60~PSJSpJ4?JKv6VU^MA zxrLHepyjIpa!L(QGiNz1zKp2xk-v~-v6m_>JxZ_am1mdR#G#H&LOA5yMLI*9alTeK z3_YLAWfvs=o2f1xsvv@`iJaZS3hn~VASD(9qS$GrqQvS!5Kj}(tkP-J+9H+kvI+98 z`xwzwqEwJ>2?|B;vF|%Y(jMSTLi^Iu`FDl1;_GBzQ!uo~H)YUix2wQ>6) zTZV06A)-fz?sM#f;@kPK`Pynasp%B;o$O0>vg^1Pdq%@zQ3T;W)QSw12s3;^KGUQ0 zi;8yY)4@hl?B34_ye|6)JawytA5A)}t+^8| ziEV;{Pf~R71Md@fyBGZ|aYwO{uZW&91AkY&C2zD3F=2nKarFA#AbI~Kk!YHR7VkX? zzS%9X-}V}$8_a{;t70^bivo$QvSiHV7Sx`+l>Aoyh_dsqkhiNJa|6#~Ab5iX`1yO` z$$Jx+MdHqI;57z4qXPJRLzW)&u%R{CM;Yd336bAx0oNQ0;N%5K>~?oIZVhw8+lR+N zqJ$(6wq@Dnqx>+m9lkM2go&eDH${ zeqvas-512Z7XaE7PFD3;G5aT_qQLM-W^6yKg@>w zH--4xYy?yHjioyy&8VWHDXlU7$jq)D%RGC#1{vts(ob%qDSfU+cm7%ljuE4%bz1^; zaeu>!=NlrIGaAU!oWSk1SCGwjNjyJ7nk~pGVtp*fiIzLPWmo)7hu_B}aK+SNuxiU} z^uxQJC@)iot)dEOi<<^Pkx`spW+TaeHHUql^^*-Z8sU&?WzK54J!fv)4&nL}sxUTb z37+D2NEjw>Lz*Tr?8u#ic$85FNuA%oo=Q?-9olUOGfC4S&QPCy_&6JjDYSCa^%dCt zqB2HQWCtnPVJLk31XwsB924#N_Mi-Jzgjz>z^>5PvdaRmLW8I{&U=EGL?9?I}{? zQfF3^^~WvPlMbKIoy(pzcMpFP`JTui`_c&3lzfS=U6W!*8+WrleJ@x~o$IJL`4Bo% zE{7Gy<)AyIW!OI5flc$CfY;sH&gSK(v(tPeNx*@{Fl>wkoQqTk1JgX#FV(;yTD*?^ zcEJOuJeq;ESH44i%j?0eum@a}`yt}oEcUqWa~!d90Z4jFu|E%Z|2O{^TWMKM7AKer zE%R77{n~*IuDQY1U%P>y*R5yETZglruT!z=qsi#To+He@acZo(T7poxTnu)cy2e=U z^T74IYVy*PWq9fW-Y20^$Stf2MY9KXu%j7CBq1M$J$Xjcy?P1um3IJ^`}&m~nf;Qu zW=3))Mf%io{d@QbF5t7{3){B+6C1K21zoW0!7DCp!g2Tq_i)=slJjs36L>lf4@$~9 zxQdoB&Jw3kh4!rkKpZEBFk$rv*0bmp^iB%v-a6$Fx~0m62Zcl4Xb8XE#Wt zu;phyz>1FzL{=>C63I;Q=BtNxw!-x_@oPKilR7+6e&1ldkJ2h6ryak zOr#n(2EK^RWM4I=v2Jnm+3(?J$O6$)Mm$}d6f0@L>%DV9?MVrFC=(1T#md=`HG0T7 zDh%l!H%22LUPT6OAy_5j6gCL;5Vg<$L{1-X6Ukl5f@kN$*-Mm~}z zKg!}@HqSJ?^-UT^YdvJ=FZE(2u1;mm!p1nXJ!xUHA4l`NmO?BuG!Oq?u!QaVgHefv z4(h46OrBN#fQGS6=u`aza^q(M&XQ|q8i&4Uj23?_*J~ET0MfJ`rcP1mn+v>+z_Clh~K7mq@i=DURKJ0ecSng)dDXB!-K_ zh?kiNzpHO$of2lTD$zSxSCKTPvn`l#KHJM*YX@7rA)C!Dvt^48bijq8Aa-$cBU0B& z=jMJc;pA15Ys-ILLGJsB0eYzj^dZobIC81bj_AxB35sQZx zMvLUlu5k5!!OVZxe@eA1-lp)0>uV^$WN`$qxqm>dPC4j#f{LpJcFaU5P${~EO& zlE&{RssY-Oga$7T$Nth~+~Zdf%+LGBIEO15*t{R z;T#%vb?(3CL;B4bQTBqv_Lq5Ijh~t#UQy-BZK}Tk_f+qZmJ?<0GwK>n$vBE8^hqIY zwG>$Q(+D5EY7SdJj>k(&a_}0(90JnTs4}7uISm^?Cq^xVeF0@${MlpZ?AU9hoYyk$ zTAxCuHS^4Cxkq@%qxpEvjEBU}RGphH$!n{67UO$vVz@;|?cew5&}$^ZzhklK-AWj` zlZRW=Z=y45vheo9IY#lB2d7*t$!>9)#Jayr`S<%w*#R_W)(2=Oq89}4rTC{0h9XC4Q3G?{lWAx!KKkwu7piNTNmLiytY|7LY8P4S)xoyX+fG5pIh^GGLjh#MmE={VBzNmGB`n* zMG{BQnb{jbC&LgneA`I&nwLW0vqJRD)Ldj+^pJe)48bP#sbuOUMN;LY%H>7f7e1I% z49&l)A!BbQ4CB4g|J`SmVFqME#dD_1Qif~OT}#f~v}c+=wlZgqh@rz`CPV-wNbn`w}DgK6o+q(Fh^Fy7N9dJ=TXZ0aH4g89r-d?#cjTS34MRG z2&;6QheFO5eHiNkEk`(5r=Er5nDb=TuyVX1<^XPfyAy>}W}%C{)?{w^PqaS019L+G zSbf_DJUa5T$mx47m)vfT(qEcVONB?L@Z7HF7(5T(!>>l1#`99$nE*DP)j2wz4aqr&>Qz3$*<<2V zO{#>1hrg;RPhP-`7kkXi4##;WO)62 z?(m_**dQ(j>$r@99s45i_k(^YaB>){;?RIALo(Sr+bSU<+7P!CShDRe|B}sbKXcoM zdR*t&HHS4<1Wl3~)U%jG2O+|z;<=Lh3vA@5(GUoGr}5Omn^ zm*BwyQxs&<&3H_?1BSApY}@s7I4GxxbN3lwWcpNK!t(?iDUpUt8xOLNcwg7Tj0Cc! zR~}iEhq7j?c+L9Lh3sg9d$K-5QE1sW zf{LxhxC3)`L(A+1#LGAzX=oyJ#%MTe8Y;_XTrEYpX1eT2^Kkahdn3^f)x+2=dmihN zkqn`Q7UY%D8&TIzAr{>faB(#zWX+vRctOi8*6-F^R-~%Q_UXE?>mGl^!M9TBdVVIc zTvHdi=JRuPpP6hy#4mhzO$t69b_I7kPQxlQCgG~TBUo)iPu58N8(eeBg+(flh)mE- z_GsL6_Rm%$cCOPndc|xHUFg>Up7HJ|wCgaIZ@z<4KjdKXv7tC<<2iic(_Hkpr4q?U zCNf&ydsxjgzU)#djy-sBHjaNf1slf?;yuOzaBaF2tmXGC127ED5zXNWmms9u;fQVZ zT(FaG8ZQC*j$K}CMThRM#6}nAh*B$cu&Vn#oMROUj=L(^$blj@X(%73o$@21Xl>SL zVj9+*^%YA@?`JJ$Rxxi4?=wudBw{B{X2%~kVIK@AIo#PSh5y)?v7f}f*z0zM_}I2q z)T<7ysw@wq<;*)X z?H{WsuzKqESnno^ZO+z_(;s#Jt6jN|I@$Q5cokL`FNVxWUDm+&J~rN^&$OPnN={xk zhY3UW$nUHd+i^t3Vf*5*oP41LyVTLxVO!lJZZJ83{T`&_@VGh_M+aN57Um+>@8JbL zFXls;k5v3wVp?F4~1>C0@p*#zXAtkKTC0n1k#uX+4JpRbM$4 zcjDQN#_UmBG4|TsH@HTp2zx%$V%07mW;=UV;-$RzTM+Dub9%+ttcf)cT$wE}_gqL! z0*!E?AOkOyvEqEyoZy)KTt1d(oWn1X01K@rVVx)2QNU6|*itOcUfPt(F77(Yj@B5# zt`5J44d4Q#=poeH7lvG)Y@rhq-vO04gL6BAsma4DQ17;u-G62?Gp>3VoQRsvdi4!4 zZ^d|T!R#IQ)vFNpXLTHVc(WlZnei0;As5jfgFooYrbtd9&V&5E6-DG5USgLIU$NGo zlUOZqNYKb=!ww@??3=oVu2>dJ)#9GhVT%sIu$48Wm)9Mb9T7pu`)43+w1|2!G4M|F z7m`=#VBC+3pmyvLsB=w(EoqDK+yE)Es5KHXo6d9He2#6DYa4txR0i*o=FvOnw$q5Y zZ%DqD31}=XfiwPKVl0rGO{7`*z=Nw}kC2U`o~(M?4xZTYHC7j8QZ_95Nm zRbnUCqHzsbmOg>=wRMQ;bH;aj4}tM>Z4la*!vUi{2f~IAo zcTI0m`0J7Qsk$6L2=T=U$Ks%}Hi(vIPr&QNlLrLiUVG!*%i!FY>4##V#ZO#Gf z12sHrb}5s3Mh1m0JVA7t7%KMu7tQG2$sOsRk9*us09MnWDci$oC(pPV<xbiUyW-Hir5`}iA%SQmpAk&Qp-_EQoEi`KQ*DK4 znpAv}M2@q8q47H4{b)Zl&95a2pBE6}F5Uyk`}{)+?jm`TVIMTK5qp~Gkct)ikV6&V zU;X`9!ponn&AyDilD?Dr^6gMNVktaYWeB@Y{Xku5v3TI6Gfp`;iJt3fgX#Idq5oE* z$erhQOnILT$MX^d_DR#2OV5)SrGag1z-=Qo%)&{UMy{YgzV)D2?@g zmIO}>=W|Wj$soS&5vaAOkORl%Ahq%=bdOt)=jl|z!?H@4xa$hhw4MOwOE~yeBtk#E zt>FQyg_o*r0oQ z{pJCR6-&UTGw!3&d%^){IkhT~e(H1weOF@mhCq`%u77o|3%oV{tt z_}vI>K7JwU2@izq*a0SON-=rCHp9n+Y-E^vU37kq9hnnr$Ce7lv8#T4!1=>J;usTo zvMw+T8@#=S4L4Qey;1p?+wc;vjeUaWOy~F2b9A7~d^Ka&ydN)p6U&5q2asRsGH|c8 zll(MwVn=oPuq1$BRp(GFX;4G547^$80X4R0DaG}l#yU(Y)^#{FpzWYPaWr=E2*=gE zo>)=Ak7%0gXP&*9hW2Mu^3<%K*j%Y1f7f{+Bb6}x_1-w@q?rdQuXM1Kwl4k_cab&U zUd(>A*~m8Dn2*QokEkwJSN;t&V2HU(~Z+9ek$D1u;B)eF#3p>jjh#ULvPYd?62x zeZaCBrufztAevE)Nbn;b?N#3ZI-?TU`h6UhoKlVBUqv88OI`dr(*|BRtb*$$;-uO4 z8GM$M1WtGj+T)GsMYMx7t>Wk4ZqrA-Kv`@canQGsV)4IsA&wf zpS3&ID*lXJHb>&d=ku5>|A{akX-Uy_F-@Pizf^lbQ<31S zpScy^>d7~08SrXP#KrN1NS2$4u5Vhwtlr(u4UTgpM$BQ<_)8h)HG5*ulyP{%=Op5x zF+_@oSCK|Olp@!EM9In#rrl zm##$7rkiQRqQeD(#b0sj?`aA&CxLf@)Z zq4_EwnXQJYOyRYIoMp)d(qk|aHuVO>?OUO+{E~nl_PUV3!q1|fZl14MC%FikyVK0Za#CiamA_bo7|U6ESPu>?sm9XJ`b2KybDiF%c8 zqbIRpv~A}o>Tf#*CRwC%7P2GI&78v|(#s#Gk8DPpZtG$5xz70XR1+-YXO2pT#?W1< ziZuI>y)d|OI3B+w632%`;=!u@v_SC|6spHS&5re?U|upIb8CsVl_mDyx)HzHDZm#6 zB4YGO9Il3?Gx|-PoXP8QG(9s?G`c|!kCYxnjf;fb{MUcDx-WOQ%W-Q7Zf+(%T62(3 z+;fyRR)RV$wu9T{Rb*~KEKV$PLK&J1aL%K(wV!P>h~mg^T-@Ec;CM<8f&&Y<;|X(M zuube=KjqiP(`Z4GCQ=-cUmGtw_^*bU{?L%yyfTKySn1LVg|#p@trGk;Zvb>y6?eT{ zg0DL7=MKxsL*7pnDy6g>W{lhk`rID^sF}05C=pR6Xk5zDePzKFBRShy) zqv6rxnRx3$5gOE7i@!=+k?T24h<;T;Cp~oGOj$X1s_{M(o^?)WWjX>iX^BwT>s7=z zyNA2r_*QhxEe-20(8kYe#L@TiH_Y*OOPND5Y3OK@H<`WtGPk3)h_M~4XI90G=a%qu z5UuYOqTgrLQLtYo6R7l&n^2j_O_3CcR{2O{?MsPBCC40#z3vxXwEZajGOIvzcs%cC z>4*}B^=;s)?xrJ+yz@-(HVw2@+6)&&$-|_hYayxlitX*;3u+5b)`})9=Fa6MiI`CHeMYa6YRL9`r9eh?1{{kB-|Km(f0_y1=pPlk=- zPccJ3Oz`Z}WyIu$0!iJ^zsINk;l3U2=RU6%V#jZ0jKs|0|9Yzvd9Ks%#~IA#=zihv zsmW;cl`;4N&zXAha2|a#dnMc}Oe3a@5sfsGrQIFZp{_NYw(c^bhNI?!Lw7D&)|A8D zqgr5j-W_?0&qOO~50KAyN5dS~!;JS{F%+bxMDk8*{rg}2e2!aNk{{F3Rm14}J?H+b z)92?c1N}Fraiqp%++Oh))+uGuqDIxa(o%2QZ>d5a-OPu4{RVs~A_IT%$V6Me+~kUz z#?h^}XCh@$DEM@VvDT}~p;*t5y6kp`SBeB`Rl;bJsGSS$+J#LuwAq{Ij#B%ubgH*< zEzObZM8B+#FzmVY+~_Ugkg;zSh_5lG%nxlE1S6H2o{1imRKtZm!iACVAg25?AUo~ z_uG()skzaw62|n!s6#|5qk=2vd4xY_n1U3$0z9UA!uzuu(avpc%%spbq2QD;6{*%k z!>QRct-K0eF+-4(&a-m@_{=?xT0BQi2A|laz_Uuua(M%<$U3lMFOV4n6eu9zCD=f-Y%KW9(cK@xkXloXs6QHt5?&bip_R>xbUPCp6^osCO;k z;iw9`G@?P)k@t%@ZpT!96}d|TaNKEccI!b?ym3`KcY!wG^p6#s+SMA^C<Pmd}Sv%gp%R4-3cs%3&Q}W88n;kC{2D0bMHyCgWbMg-=(5@gK8H7~^lw&RZ15 zK1rN_CC*vl9hnKtkHqmvxpowdF^r|Sybc$dmH(Uhh5vG6tu9^0K}O$DZK?>pA8`u112=w7K^G@nZi6MgUP5tmC@AdFA$lzdo}>N4?>@a$u#bZ8o4QF!0t&oh0C(l_`y3fqJwVcV{mxLEWCdD#fD*76jbJ>CH=|L8y--aCX#rpU0rf8NK+_kZ97yGlHk zXK6?VSz=xHT>SB76+YPBfoIv3kXx#Dxb=HFK5|_hFA{vluBzQAq#=rp^F76iYg}Oc zJr81+Md!#=xk{Ay=r3`*U5*WZ@5YiB*5F48=ZKi}Gq4IAhaFefqpSaqr1Ore@qhn# zX(-x5OFJ}BQl0yK-5JS<5;7`9LbA!;+NB*0kxD~RMjGdSU+;{_3L!GH%BEpt`}o~o zzyHtUJkF_ezwhgQU9Z>csi+Spdn-uyrL}P_YcIg-Yf+dOT0vzGie0&x^YN!gD{;K4 z#qXa`%UTRRe5@kolkmvtw1je&$tB|Cs|v)hzJ5uLgcE-bC_t_vJ&=b6kRwH28br zlkm{fLHKBW6}diP30S)HmxT1WMzb{$N#hiL}|i=h7dh*UeKj_1erb3&o_y(kBsdrNW6!&}#Is!!o;kdR6FM8<#JZ!fyR8wpXWwA< zK0ki>k`>UWRTH1Z7s8a;>9qgcPGZ&A3tl^_L+@%&>W0I|#gAHq>1OKJX}V#u}w$RGl)v>Y5`zmZj3J_c0judmMgD_(l3E z%fe{hl6x>_E$qK-L|RSMpwq1xK0e`L+C?pxa_<1x91}o!R6c}X8A_gSH3H*XJ+NE; z23kg(#I>{Mfb681puR5-OwUDQ?H?^nO0LFU2j*dDpA=r|yo#^iugiPp4&w{@>F^3u zX7LmEpT^iUd$duoLLcc!Sad8B%zvqZa)|@3c{mhbRn-E!Va{KtXCZhr=9;J10T1-?(7n{ zWS9xAA%##=nFQC1AA;SOaa3trCf#!#=~elW?By$8_9>+gQwUkce%`r)Mh6d|hgyHe z*NYll-vR7-!xWaDcY-|#X=Ya0#%xuiA*(;Lm8H0!!qB-g?DlhO=DIS2-WsQYllv}! z;1?U{$mea`i|`kud~z&Vg~oQ=ym{DyL&@ zPKn;b>!>Vdx~vpBgk6(%()VKbd2Am$u*->}{$hSxc1q9xFzY3o_)-w5k z9QOK)LfwFD8`gW|O!nl%N>-!(nVn75ud5j8UN_IXi5|4NN9=!Sunz{isn56-RK3KJ z)IX{u*ZwNddii{6aXXjo*4n}@-&?}^?Bm!Zm1uUgpDSA|v8@wuVx8Lim33!)v&aI! zU?FSvI_w$TpN+iKpPA~vL^CDP+r3VkD6UaLkH90iYshBm8n+ty7PsLSnadcUTSm|C z>R?m%)U%SME$o=aL1sEVkBulk$}s9a-t9btQ_>&d@WN|i*JK#J94gD&#~r}s>yL4T zD6-WTokRO7UwqNO2)$N2?EnJt_*n8n;xV!H}%;7J8}_SvcfH^`ntPu(^2 z?e76>xP`cPkU2;_44y*Az+5oh+6V@6EwDBCIPpKI4~cJopxeP%)a|Fps;%pBz?;u_ zK5zx=mypf<8Jx*^wmqi4)9S#dB9abyYQcpcI)ti^%1On`M4~wRg%FeuXrgu!2Ybmf z2Q@`H!R#d0aNH8SP2FM3v>w7O$ix{5k>u3p4KO!pJ_G~?Gjq-P?3MmJY(6lDTyK*K zZ9e&EQlMMsn?9)Sbzjvwg>)qAz9YHZBX;x`hO<+46f3Jf)34*i4xmZVEaiI%+u(PM z=&n@=V39Myob5o8eX@{ zZem^V8o#;-&WJ5>QDJr|F0ArmICuWz5FFO&hM%2~#w*3c(u7YmW$PU-xpM%(BM)K} zQ7W|6nh+yhB{DWOp9*fnsJfQ{NbYN4_9G2!mG6XioAs#5bwf7h*Jy{#c)FHXh?% z-}XWKfM)5Z0ivfoY5>`&-iPE&&ZlZlPPoBn8(ObUrPb}3*fD-Qv3IP5Jf8wEa@hcj z#9VsH{Y}s?G#nIiv}mgEl2#1fOnRFSfl=ah>XzIKxVz!ltHg>KckiNg=`t{^WERw3 zwn9lo9;#jmgO6@vf9^a7s(ED{Zk?}y)$Cu~w0F~8Kzxy>0&--P>PxQiF|9*`w?-FFb~1&^ zwcLnAK@_wfTu+bLj_0RO?k1Kux01szn}z?BM}h9Zsc>S43!EEa3KQ=J;mO5bT+;Sj zs@&!e@M_Pz_lQ@rsSyhG{sb}sNt47n5ckPIsmd8KRKkSpQ!bVQ9cOpQ~7 z*2|mWfQk&(=JXZL$4rn+JhBpp{V2wn87cpp(ZlO(yoKQRD}}q`3rN=ei`XKbzpoys zp;1K;vS-fpS>!Y}w&NGQ{H31G+3iJl+~etoOC=mH3}(3#W-+b$L+qH+1!%tD1Pv-r zVEpBw?DMg8DACD5_k{=O{+bN1QELR1{|I~u))(2ZZS?zu_jqrOJeu8ga%po;gPEIV zf(P+|$t|;?bKVM+9d(qnUph+%1OSKr%H>ah8!IER5^6BHY^+dWQAi&;jiVK7 z)OmB?B&r?|N3G-&=#Jn%{O_nznB{i{<_X!H&Feng+U2cc*7++rcjq8{78x+1HS575 zOAq!7ZjhMfNpILt;p>OJ^ml#=DWv8Q{rEW7up*x3_Spbmy%hPPlWp|6q?=PO3Z*?^ zZ#n;7^<21IFzuy!5W}`qVoQW3JC<6)`H)sBJ$MtHkCY2P65mT*Pt;Jgqw$>4l1$;x zhb%n(^&;f@>hL!k%Ah1Fmsg*>i~pjN#+&X8pi|EpfcA&e&gIL^*u&es*y?myHcVl( z5G*GTyNs<^!vqQVn(h%Md52^0r4;;HwN~udKS^(TZ6aB@TKuI;_Cm|ci||aY6~?V| zfT$%)p{hoeJ#=otMTes}71;u)c&h-4C!^^|Z#UvL+Y<(GwSr8Zn-JnuLE8@5am%{) zacWjMw6WWt9u{ApI%+!&*q%va7O#O|#Uk=_Z5!Zq+L2lsxTPE4{Yl0*lNR8qweQfWEgD!%>xd?!nnA69}>ed0k$%&?y*OG1a0AMm1c z3!FW^yR;7o>l^zNYBuIIG(@@rgX%`z(6tB5)u4FJ`@ zx>V;@KJL6@L}tBShp%Iu=-0*V)xEP0(hYvS32U+uOhhh&5K>PIt=%zXkskIwtJ6y<&z)xci3ZHc$d<+SJ;qcwa8+AGXlLxA@S3ie> z#AyxakJ5u<4 z`-5P5DT6*+TS$!EUkW#ZvvG!ZKAucbq+^<&3LEW?P$QfDU^q+_SA~pY-_vAR;j>y| ze$tkh^#Y5-~7$5 zjIRL4aCJ8Rq!v4L^enykOM{%Uy)T>*yYq?;_vNqk%7+#GHp9=KT70p>Zb*{d;cBgJ z<64*H>3aC03pgDb17>nnFvXeUG&?J(%R^P+ovkbIg_``fpq2c0u{U2Q@+X{a$s|6L zFTi4>fqcu=Vf>fH8wE4F<%AAM7w&Aea{csF!`0fTpX+&_0lezaO>q74Rd~HSgFiP> z!e8!F%nL7e;O`ys{N6A_zWkNgA64`Ywl6;TFaH;Km`X0mXwi8kx51(9GH4Fp1vWeU z_-0X0+LiGikk@;tzrjJmejg)4PwnBE+BE*QuPIN~UWerA>-l*~A$;VPdcM{^fu9?5 zfWQ1}A30P#6vqA+LQ77eD7FsegEEhT{fWa|WsNTUn(>S4@m&CUn>FC7nloSIxr)E& zxRvjJXAR%?UJAs;PJ^wXN$}Y|0^SdehV5^bN^K|g;SZkpL-Ygx&>?RWvC{4yF{@q= z<0}naQ)OISzew2+Ig{1-H32E{}mVq$n%>DKEv0RX)xga5P0%CpA1>$E?nHO zLI^u`hRFIK;Qjy3@ z>QCSGbtik%*F)EvSER6LBusobne6qv&8a9yNi4#R$o|!BF!j-9-bm>T|9qt*oNSAB z*(;I5_luwLWlpR3sbjbCyvRa%roRZ7vm<<-u#lgyD4BP97|z?~^v67DKOEDg!duwh z6qat73R|A1KzyhyKV!Z%pL|xCcd8BMFL&|$m8*AP;jlRV?#d1P2O0B!XHk3Ya;Xii zm)x_CbiGp><$CK%v}=Yq>)ke2<%^_l{DEd0*Z2@W*F9C$Q_xE~|F(V!vk1vHEGFL$AtHwv_`oYsa{dnUme_{Q&vHWjEXMUOMZ?Wq&9{S9< zNUpVS0=MfPL`XeIK3nTiuTFV*vegA{uXo^<)U6@z6{D!7ax#B;dj%h6UCQ6zwu|~_ zR*)cP13uXHD&4WbnLJJRhs1#caj8|4*gdYr_dWR1WpwwvhK;kpMx>ya472(-0XNI`4rfTcUP6+Yb&S0{0}2|*=1%>7q*?(sVw2= zzRc&-1x-G;{tj%nbmIZS`FZQM@=M=e=39z#_?Y@?VdShZh|(&jI+A++rrdAdY*jzk z!j={MnUZ}_D%WXU%u#cT0MK)2gA=EXfS^#~g)_uD+qcpOLlf|!o+qfi*8(8h(lAZ!S=`o>s1*|G#kT(J*Pq^KJ?$(l9O*@0 z^glrd26nql?XL><&a*)_(-R-oJ*3HRlgVGlWWKU*Bf4K({!f#6KV}bFAG}60gAQO~ zpVfT8%M@Or+8g?J81QR#b)Ju_ml zwclHkqOJ&ois!-Bp>}oJHgB16s2@BTzMk#bNpRAR)u6kxfj(7GMK-sIn%+#S)4gF< zx7VeFMNYEfr^k6g&z;@$-uE14q}PP^jrT(4>pYaMoC#0Er@&2Kmw)p;ip*Xz09Dv= zx_(bHxGyp0*A_6=(>I>^zIcu56SMha1G_whTpfKM6f*GMZW}}p%;P`v8hX+#67UV{ffny+@XpKg6!zybG=D`O*DGn z{DF6(mg0+F58<=iWZqRSmF-$>!=_bFp&Rawpaq*n4w38-lsr_zCB|8(M6&rII+g!& z{?>~(qy6r+yz%H_ei|ud<)uMvUhRAwU>Q<(baNw{6(0X@4)F+iO#9aEL3(B;k#R4j zi_7jwjn!)DlbuoUMZ9bvPL`mdJ zEJaI;8UOUw>qRv*F3kqFKmIM=&Qsa-P3Ej@g&NkRyu*I}1-LiOlD!+3h1$7^xU~EO zJG#D_)g-t4`z_6NI4-TxY2*5?iDFvsEm`U#Mbzx4$~s-gv$;b?GR68j`epwaI_&mv z@_7;m7wkknz>72Bky(q649C)n2S>5*0&liBYY_W$dLz}l6F~kNIs)%hLjLo0fxzYi zoZ$*2o`xYL$72^vEOvth7wa%7`y{Fq#$j_%F2+B)g&RgY;m{0y>J(y&E3X_zmpy63 z+Cz`|>xh0B0~c(alZ)dd>NxZ4a!yS~jvrxm8CzzZ5Iz27bcs_uT_a8Bz9b(3^WS$s z=TWKjkWC7`Y@*7|y2a5YUvhE4BW29diAJ6A*Dx}BKDnRN%>BHcCA9w005`qM0BO&$ zv@3!1iysM-F_46uIU!6|iNW))b=i#9mMmo{B_Rqk$gpfTB6Dmyyx{jiu9-aDq~S}- zrFY2P2O^Iy;vOx#WQ>nxWw2A-f!yvoPfPY3r7QBJw6U)@uvrEmPi~Ok){m(FM_*hu z+!jTdSoqm#kJp!H3U+Ze#8ccK^bB?+5oS-ZpHUx{_;NHn-grZz_)XM=awXv5Q~-uX zeqg7eNM@U_q#sLXkW;g^fo60sz}ud5W!r4(`)dH4_--clToohvtH~B@bHamR`@sF` z4>&eLDoB=jF~uJnnUk^(ds!<-JLfCX=`V-ExwDs{VI<@Hh3^(mWo~4@$j7l(LL%TxIKl&?ud15g zlIAIEz_8QWbnTiEMB`o}`QkU3Ts__@iP!wWIoTg3k7_(oW#=u7w#`NRN-H$>xrio% z^w^d*B{n=Rn*1G>PtFW>rP0kmh!ZufSL5tuW$*m*`P7<14aHz_!fEaA}$uC>~ykt4K87GxlZ|p4=j3 zQPyni`H$qj$tu{ayAm{iCX+F@y|@dLxOG%)fmc03NCC%@-o?tmM@ zmrEzf+d6T4{ARYB>kdknA2xLYwCHJ8Epg6N=Y#AodKS!55?!_hU1L52bivZ z5x@1sKw)=37PH|msx&_&vuG*v`R+#LuF3H;dxkN`-YIOw#YlEi{tG?kFp${T6RhV} z;GoPPX7X2y6A15dk=u)x}w+M-Udv*o{r7|+o42Z zG)~YIecX>FFs@Dw{08)cSr8{_DXkEiuKlFm-Iapzkb~H8p&RqGsZjNOFf8;>0OeyV z$TyX4x}Ir}%vf>#YG#Yk_Xo0@_e{vUZxbLt>jv5Iyj#@Ck5b#p)AaPLZtQhDj)s~2 zqH(e|;3TKSa*`}qYHvc1JTqsDD&o+tb+-`Ie1Q6AnPTL|XcP-+QSonIyrMV`p4%fB zj%kD!3&h;rvg6_`6fNnv6iV9b&GC$0B!+f{5!nx(BreO7whnqFtQ=g5mwp+rX@_;O zc=R%E#zWwI#$2Utd(!AN&2xCyVlNy_8zz*k`9$B$--=E`E@}lX!54CS;G=E@Y&asv zUzt9TJ@h+{v)ruNmgXG#v4~Q=XNKUt;2rhW8-*+1$*_%@Sznq`W-cZ1mZ>Uosxi!Z_Kas&2dz8SqdI|c`Ye8%?&62W2qDAe?7LOJuz zP<`_VbVlp~FYA|7zI#0mR8t~rVsCSIG__FgvNugz{1m0_z2NoECP6UiKt0=Tn(CK^ zN2Y{gbwwGD-H?jM9cMzov)y3lbV~B;Nf`c6j>NB1i-^8Y3>U{g#`#HZtg-DNr-vU& z!Oq1ncxMZq(TbveGRKHd@EbgFy9nP~#ZkA0O4y#U9i7I?v&F`FL}guHZqe}R)G@pp zKe!#o$EVHM?`0X%H=0%dcXf8c=BMdDvH8Jd! z0XivqVuP^>x~eN~B5N2{|e^V3wByN9Na*Rix8{n>KMrD)?Q3z|;? z(NA{Z(C?!kPB>bI_dCwfS8w%1pGYFUky(uvSF*8rZagO1`j81tz!{mk;AO{ESgX>7 zBSz~ouU#!T%4`;Peu>9Sms2$8PcnBaNr7(HoI~$g=n+2UCEfYSf@w6bVjUv`aNw4C zcp-c%p1IhIy`9>FXMRmaI#+{@5ceMWrcXHSl;iX)yM+rQ|B?ak)!Be`D$FUSFJnhf z!Ox%#kn{BlS&$^2Ib(w`&^L~m6`Ns)Tq^x>=m8BMY{JHbm0{y>W2R&BnH*K=1784n0F6^!?|KNv4W_Bb`66&>(p>Ur2#ry znK1nWrR?;tN4O;YI2zU!qR?(ZJJ+43j*D%P_mX3~|6WJgiaV&4^oNVfpTs34Jw=<7 zrrdhRKaz(d&I@buL$J+eGfKsK(vm1`hJ!v}irD`XmZXMP{1Pz9qKs}I9E9D2rqaJw z8G`n1c~;wW0DHfU#<5Ln=?8HZQm&JgT(-YY3Vi0%C0CCN%7uL~=fE*~IHs1aQm_;j zX{ks(GRvh4ZxR-frNth6QDmzJg>y{>f4R14dfYVsb?D%(h_7w0(|Mn6!R1C5#aIa=Ii?vxK-Q_QKz);=QFJ8K>T|LO-8*)G>ZJ z^?WsoRyO(2^sg%9^xMgjC%%ts&)`=$=DHA;%2aWk9w+GVt8(;SbcxXV{u8yd)Dw1H zHi7Mh^)6#oGXBl2x~FS|_)2GxKO8B2aLGdKE6|4gclWuck9MPmRVByAZ}KpnP}aicS%D9tvbFGKosgPz86Bjz3w1YVE4N|?xITHCXJ;V)@Vw>qn~ zdXEjCJ<(K>KnH0a$ChCisEJs5YHu7XnDkmCtn}5yeQD>o@<@?q{_8rIW8RmJKYCK| zDvOoQJ6|S=QD~-N)8p~z!lVEGFXdMU*S^$orOFPaE>pgrM$^@e*b+4jU7M$~yd))- z^>-rL>2(Y4@v-#S;lA{4O))+iG6-)ji>K`?u5!PAdb>6XhzzPxsvNoQ?y*92+A6GmNR1L$BkMRFrGM-RMvkb&ymP$kta+~bpWcw!qWZsm zrp)%m|Iybsi-geUoOI+sU-ml1m|0y<#Dqv+g)Xr&M9jNP+ZR zPyuWyBwMB^bL_`CmtHa1wSFRzYSob>?vz#l*Ro!oI)5eqKEvOsdxgotv;X}bs@0d! zvh-o}Ov6;pe@g?rbJT_rZDRLSk2>FX!6Dx7Z#igbC8FK+J-9MI0Zfm@;nS(Najfwl zRPmIJAWNC0H;TCdI$wqam+b4tkb70_{I$;O;jOx*q(056%vS z!_o<+2T1!84{XoS$Jn|H&=~lLOi}C*U~v>YzB!rJWUL35v=nHqodZ$R>)^Fg6>R## zAn3W+gYP>E&i=jtd&NR1Gv`$JUh9vC_Pa{W9TmvN1!=fkyFs{;avYX_ivX)?EvV7g zhTRo+xto5maB+4Ix%1o#wlDO6Xpw<9rz{o6t!p51BI98Gu&30*K%39BsUeTmuam?d z>p7L_^Pt&}=LsWk*8AjByh2zj_PJ_np9dGYsgl(PyaCYbG1it;v>J zR=|yC;b8v88_LQXgxpDu)XZ-Uv=$vCUq1YWW!n1uHW_W?@~Y|6W2X3R&_Je@?ZCPe zO3`8Z30QK(25z`76^zj`Ww9R{ktN!Ze+%KEW9 zZ14=j4QL>lF^A{=>`dTJpD5;KTPgps;yk}1%K~GAB7`}cGI8>9YkqaL$YpZ~=jVJ< z!Ih2~c-J-sRUY5L5c4ave%f^CcUNRghE>su$O8I7!y0cNyg{BGQQ`NW;4mdrWCCx} z<(IZwf&H2rIN}n3L#o*6eb54w^^f5g6C?JcY9hJ3bQJhCTqif=ncym~1bn&?rEw*c-GT&4Z4)gr-$gqPWV7aYA)hGFZ1C0o*m50=E;-)&UkX?PhX%xqiv84 z98K1y^4M!xIrW-UfNy-x;Lv@SXiI!M@zgbC6E2;`5n;AuW7{jd`dkh|mrh0%D;3`T z{X_`rHuUSEF zs%Y^wOSDBFbQ^dS%!aKC{phOOxp;bp56 zw|hKK86tLM{1h(!sU@%H9pQdV=q)UrXh>c?N`ZT& z%ZRh_9HP1<04JW+ghAUslko{9WbK$eq%u`gYP7wB91^cL-;JR-@W1`EJ1UBNFC9*1 zj47o{MIS{(?r1bRmrE?mytttcw$UHPGr1Do7GcF6D?xuQd+%`nQfZ;iZ5Xsz5$|7B<*L@FNY}QW7w!uvbPO)# z5|8bYT#GWGOMAOhi%-UI?!5u*)zRRG-YKVLhx?FmYB_k}(0~8lt9sjcB2!pR8v<_# z``sRbMaMe)ds7G8ziJA*lC(kn+AC7}Ap+OD5cTRtF~`;v4i3NX*LJFhylP4VkTmmBCxMDZ|dyK2~`L z&a=Adp^j`}#ItZA)eM&i{b5&mt09&79G^-N!LG|h_&aN1l80{^rG>7dU#rdfLhVCul`|4 z?89*S+oPC9+wP%Xu5Bmlwc;?~(nMKyai6O?kE%`!v=`6gS7gEzK;ra^f%b>Pnzz zhCC9Q^VGNw?QpD_7lvjxl%)TyuExZPRj8>Uhnp@{fpf1pBx_alJm^Jko$6o7 z6psk#yzfgczu7CfP(B;>zuiuwdPfuKJs-05MrBP&xt-uE<{<}GjKfzI`n2J==!>4B zP4C7}rxuA%M9X0&Nr)OoCq(UY=^HtcK0R?5^lRTq_k56JWuh+hoOrSSQs=VCYNOa+ zjY5|!rC8E1={m8k`9!)yQc1UW49H#-+1AfxadvwhI^`XN8?Rhgz>;#b_tD`yFQmcj zvBi-8Edb;#6hNWW3J&Ot>~Nh3wopOL1r1BU5}i0U@N^F@ZN7)s_l2=Rxdk+}aWXRw zv}VJ)y6?leh`z`g8Ng@8PzE4W^2=fmj-O9>Yg2AtBGbiL1TH zX6>{kADVQqb!`;R=t&pmE9~Yze@NrrzFqHPYa{B&|EaKo2Wfb3xB{EHG9N8U;+XTt zMA*7x0;tuxiF}V1TH0xdCJ`EV>U|Q;I5`lPY#oLU>w~#rJ~8O3)g!HrFl0}TGpFzs}#1cXM6(7jVa_7++tV%nqw8Vuv42V5Kh{SXiqi3z{<=rq8XTv%Ji4?5bWQz~?vW@0DZ0FIKYQ zN8(6$p~yb&CyQ>q#l~mxN{hM*H?IGDr73&6K3WjD=dPkDb`$I5Yx-X2<^H zH|B{t{OdP=1)F?hocJjX^S2EqKJK;TRp95M_wY<3(VR7i^ktY@WTL`yo0C~9A zo&4{p7RaJ@U@vwg zP4pfojX;|n33&YI9?l_QFtzVj#Qi!b!|v=eWT$RULc6)STok{IW=QPu!ze4XPtbt* z`+fMt1{-EATL0fDQ7(NZtWGaJf*ji6Vc69$MiJdqfJcwt4uelRj zF!UK37K^O=H_xduc8gsg0{t&67&i{xh|N`YtSY}nC|&+eIPp2@-#IR$QzB7#kV`$6 zMG>c79^iesU0Nyn-t>}=GKVHx8hT_aZt+UO_v0Sn$9ss^He^vxj~^KCJeKuOuV4$_ zHL~`e9O<#bMH0i=JyNmF89NML;D@b2EPcfrbo$(mXd-zqoA&nY4Lx~s_L)(u3L+0oG1 zGpe6_okDu=O_EA)N07~<%7mki=`i%SGO=)Omki!2GRTiRb9EU{q<-y&#OL-z&I7W! zQ%4<_N za%Hz7k?A!+s)74!oyS&_FT?<%Y!pF#{Vs0&L>Vq*{zJj6X%}ahSS7dy4W~)>mJz!( zO7xZbYC3jCCKWcF#*&w;dBC)z=(F-sC4x`E|d!ZvGP|d-N*J$Qn%U-;)(RxY^P(vJqrL ziWOb0`cc?xIGpU%8%#T73Z%c*9g{BKn=SPh!nt?rt6ct*)^K0yLb+eOxbt)s{YHNx zxW1zo2?aX(LP+gR$&N2bHRJnBN^P=)t9L%s#$@&9R(<}%T?A1ZFYpIFPepPidOMZ3 z(SlL)vmp4S9*JvtLHu49ORsbkk(6gKq+;*jTCZLr2j-fRbW6#4y7co{+O|(TyR&2= zL)5mTH;xj93@+lL%5USd@!A6PL<|3DSgiEyUbl$e>ywv893Go-_$0Ic* zoR?Oci&sE2*(JA|zEXI{d5HdbeM4n>>)t_b#>!-L>)OpZv=0!pBl|P?3wudTxhfg| zN}Dw#_NQ}>4;5}6>P=qwf0n$c{8rPQkw{|qT;!~ijfD-XWw7yjk>Ebm20nP@ z`PHD=rl%zKV=j4jeWzgKaD!}g-$yQwS0e4*gW#&l21vcR8y4ko)T73aBwx0orE6>n zntyP)TXRnGe$f?9Ys)e^_vRhxkL~KPauF34T${^e)O}ojZQdr4S+vKwU${4CGJKY> z&mauWB-es_@C*3$-vlU-Pv?Bz?n1P0l4E zl1$lL4DU?)!Qaw3;B~W-yvaOFF0GN_i%q@oZuov-qgxjE{}~QYq6>0sdr3~6A50G1 z+eWKbIp9vo9a34O%!im(LG#3PlIk^%ds4a!Y#&gvYSVtQFnl@O`x(b+d!2yTpBcpE zi3_}{`o;D2Ou$)7d|@B#g~Rqb{1E$E=)3PR41Lg#f6@Csecc))b}wW>*RPFWyxbIS zh)l-$2Fk*NN=poQ^3BrUPwkH4uz|6F8pR$6XD)4Pdak)H+oEG z4_S3@4vE2+q?g52x?<4@@@`ZfaaG<%vWhZE(XTh;{Py=`_1^~R)}!yqj_hz+a_%PS zYO+LyC-S6d;w8w}JxeA%jU_8>BY^)g5K@EGsBc7jHMdX)zZi?pd!M9WTI&xQ@2+tM zn|>0Pr^>keoCxff?IdT^Cea%kG{`&{nQYox=w7eDA!AT+DW@#i`NA@aposWLsu z{RIce?N$M{X$X86lTNz)Drn%+P^x+)ovV8^Rg!7^i!4}i8s3sikeJ{F0mn{}aoWD1 zWp)&-QZAE~8r47+oq?q;%fVvfRS2EePwZ!u#rb|Y+?vS#`SfEVL)$@=Ys_;2$UkQYs^basI? z&8zR`?hc9|UkyINiRqPO%hNR2SA!%eayPlDS4h-E|HCzXk#`e$jGH)2^uov>x0oCv z^DaMz!DY&zF#Ij}2P^XHWUh0T&eJ&G#V0xUvRSmqZY0?pa1a(te4zTFF(2Ps!f&O$ zL0RQIv`4$~YQzj0rv4&zLzBpZNPrJT{rQd&)ld^Tm*l^w70MFgs3UZex7#9U!F5+Y zy&#>W$ToucomQ~ci{aLEU(B^ssP&=C6$=5#=4i zN|WA*RkOi7ZUVo|)0^Ki(+DOW@Zqak`|?4m@9F;jx}aK92f26asL3M28(fQn$kYr{ z^}T~MJ|OUD@liq5sxKw;0n7Hx;PAT+1oghe%~N_!y%(*gL$%)0{{~PvJ$*mXOPLC{ zej`M?YYGbW`Q+HDZjuzT3VPkmg7cU4KykuxxW8M0UpHVU7#-~oucqq3$xx1K42U43 zCk4aXEn{F^^KbG~!IA$ocRIdp-2vs&E~sav;AB1?mYmxSCx#}AI_Yqd9deQCJlad- z4_p@R>209x+OdMqU_1Uy|6>qs5h)2+Fqcbd^PgIGHWnPixPEjb_5N?pDc3h&2#qW(Th$bzNHu=KqS z9^CHC6r3&LMgodC10`nl=n>Vry^uYa5rDa3@1b6D5kC2`i@E1@;sU%&n>A86!>)K- zKGA}8+?a*jG+TD|#c1Xh*olWTYU#zH!FarVE|Rwc@MivG{8K7}>CM9_cmFqc`~GnJ z@F|qEO>7qTWK-BfuVC`wj0M|~u0bAksln3Ko<#fBIov)kgI^gi8AfUI7(BmRa4(n- zH%`c~{>2FI_x+-|-G9mTFSf$N6IW?3k(F6II}A@Ni}`Z4gWSRQoL_weH~yp#>L~X( zYg(NpUZ2b${9OuOG*DoElHufPoIOlipi1ZVx&~%?x;U^{4|{0^Vv9lyC^Q-4oUO)a zZnZ?Zz&DH5A9iEsUmwEr=JGg%gu<=~E4js5`(VJg8&cnwGce_#s1@XS;G!o_@#*gK z_~b<`*X!9j`aQ3TzN!8~?;P`BlLy!1Oz-K;|L+8xeRvQ)=L?w6P7h|Vgi;UVGpua> zIx=L#WEOO=>)-Q5VedR)_?I@!g(95rMD$9(drx|O^8g?9skE1!BOK(j$y;Mje%jJB z+-G%)CiY%Hs9gfAwB83wv;SXPaW}Mi#XFLwbSs+pAP=G0h@lg434^4C(0k5qhxrHM#*`Qf5Y}0ia7HY8>zOOn) z9^{8nhs_bx%=QU{>u6&CK3}21ZY?=EF&r|6XOj3?>*)5?y5QTmU9wZgOjznwOz)+) z3*~nD(4a8}My#^}-hD9L=DVNLj3oL^ehO7q))5Md){=(t?htV5D%Yd2k=|@7!Q;2q zVGLabFW#hZe^1{h={W^B?e`X3w?dZnd>zMUL~7uSTVf>*(mOJ?Jc> z$>z-MLKfCV6?g7IW0Yl%rAn;om_A!xTZYf;w@XNqFWGas$)&m@7}l-u!sl9-Y5pWF zVcm`iG$}_JljBZHx@1R#OS&hWEAkHo&rbB;sfTW+v)FHW1>VqCo;@DjMsow(p`pSQ zuKBgm*!jblnY#|&xtPZEl~i%ssmE|mrk$LfzeqZ;p^El55c>+&j^i~q$ncGoOB7WaHkjyUi_Ci0_@7HiF`N* za)gc0AuOp3g^6@VN~=_c!j#lrYfEbe8BdwxaS#?c8IFhXlsF!KulHUv-}AeE*Z=xo z*YAJ**X8Z={BVEnd#$}^rkP!OAsACye$=T>L&bacyw+r}aKA~V!5=+Pn)gD;=U$%c zw-2rNx8A*%ee_McI8T}X*h9{~)!yysE9_;_9DCqiUDFqDb5CG@yx!k-;!DmkPwnAv z`e{q&^4dQ3@UJfR|E#>v8S&};&fT-S2mJ;-RJ#4PX{F}B-{If#+lavW@Js*nXXZJ* z=10Z)D^4o)d}UN{+p2#Ampw8iIR4qW{?+MxuxrEQ^bgm3Wj}CGqkY%VKE(s0ot@RI z{;&sL$7e?L+&p;V!oLPX=FJS&XD$eaJbR}YKR@?r zu=c7cr7Lf;N~;!jD^(oxMeyA}-tv$Ba$fPnRg3K6HKT&l1{@rO8}119U3+eLgI zXbHc2aO%yaL$CWs>6`Ps(ji-I2>$4EN9p6Hg{9Zl-&*|co2%0!ep)ivo4aq2Trf3Q z@$HjAlHU`&I6Bt<_3gvbYj@tVG;QXo!Bdkj^EV!UU$AKTuJ*?4>*;@PJ+JiE+?PvF ze)@driGL>jYx&%Wx3^mpyzrM+XXWel&ei*U6dZEkApe#j{Jq+*me|W~Im13^(Qg%T#0|mE z-wY0R+F_^C(rLStUbyJzV26hWJ13cXH|eyRTdI*55974hlEh=MLV- zzu@7S{+p-Y7ZkQz8caHTR&c5LGgz_w13#*4P1io()wyZ=k)`CZp9ZTh_DTnyaJQek z=D1++oeu|X_AbH75C6(%2mLLW^KzX(!dhxqJ=raN`>S)E&-VS;ZrP`3-}S`w;E)Xm zl=iwXRXSnnn?YlrtxC=n<4R8tea{*6;Y0j9I0*I}{dszeen;8k&U~Y|uJyFiq>twZ z+Z^&pP<7wR^oM6(;4Inwgv|FR@8mzt_ndaPqPsJ!#}=7qzdJCv=b;9F`BS5UU%p=# zJbc5W&hg_$XSOdRn&aRq$_3X^QRunQPWG>2_`N#voai3ljymRhf)5mU{;!K-< zkG*KSq_c5IrZo590i|2h6P^CI?pvCDeeX>5-0L&deJe5}&pXXIXJ(ZCVVgN2(GK$&6(Tp!Ax+-&6#uC9?hKfO-tsd-Fba{y(WFz!$oKD3B#QCx^Izr zWBe!1UA#WGRU8!jaAK*r_^v(d`4{#YeA#6``PTXE!R>c^7!2EOZsxl|?*)%#&v72T zVwHW`mfKae?s|9Tj`asto!{-H^7o=zXH?o1ThGqaPr1PvJ>;;eXw&h|yWhN+{=Dz) z&UP0c-`A6a-KusdeYEut{=?_rU zrgtoD__kZ=vtOTde%WzO(Dk~y;tl5<;Qw;y+RP89EYFl)c_;I?7fx{=+ir9D-(p-f zbXoAhv3Cc3j?V{+i$|p=HcxckdFKP?#PcSXe*WdtVCJH0GWY+uE?CmAowIAk%0#oe zSKaRJT2(jb%&OBz5BAS$INpEJ-P6C}n&X_z8;=gU{`a}*`?vW^kY2M}aM-i^6}LKM zWah4Z_t{TgzbKgX{w#a%i}rU;`SC0N_V;$I+I`DmRj)leyXyCm{JnrtfA?Enxx&7# zG$U9za1D6XI5chK#%Pg($k-22+iJ9}0S5*yoY+zOCKWkw#ccV_Io^_d$!*f#URP3d5p*{@~Z>z2rD9P{_|x6@_?mv8vQ&mB6; zAM@Uf;OOtN&VROl*LcE{(*;oZSQ{8_RgK_cXukz>h4_Jey@G+6(ch{4gA^JX47_=QRc=fbL+iT{WsoM zb?l?-ol_cb%4~o0O_?t)^D=+(w^2_Vcwp5zSG7CCzaH=0wWcz&+i8cFUOS~Pf7@kY z`rdQ+mC%J#oeSn~k$G*u1A>G4zUCix+%3+^_Prg#T zp?9{l@ZHft&81y4g%dw1_8ZGb8>ls?!aQ5w2>x6+0GOa|ZIa;)=1!>*-w z*Pc@~YQ+{+y|zE4>im9Fg6`@0LBAP$J8LFB=^uRT(%^^t_Xw^FzbQ8HxlR8$g3rUQ z@0UL3u_yeO{(LTITX<;Z$ZZbI^ce7|^ZkX#JB3$g@$Vn64toA)ac13;Zua=D2L`qG zy_OkYxra0Vg9q$gNsasaqZzq__LKGp$Si%(MJ0`v>m$KG=Q64CmxMrUe^znHB7Q^l460|JBY}=Wj~? zH2hJ2&duY3iL*ohr1@X^51-yGc(>uj;DuS21p8llW-xj1a_5C^^ZYxWxWsvQ&Nuc= zzdRfK>x)&vdAE3e>+LIo=bPr+XEep%N54IM*S>es`_47Hzm)0Pbw+0Ix(kCYhx0pT zxkm;sUvRyB(oxsgyUchrxanVq1zk2h9bC|Ua`4d9A;EXYt|@N)@rq!>JKts8@a4=) z6^;H8yUud5$O_~HI5T{Z;ge0XmuT75)mtI~0$2Tt}%t53Q=xP9Io zrS^W;r@Qep_34*Q7<|e}jv*Rzptxvq^@Be9Yy4%Gy_SA>Jb8b5! zVgKXE)#(dXZyh9Velr~`TW`Pc+hV_XT z(Dy1nzp^o~cKt{2$e1cSz1!qq*&QwRF&W>#@LbC|d+mOux8D3cSo-fz>|WDu3bsD= z_u#QMue9vE9ZL@+dX-LpHeY^CyYlWw{6oI04IUl;RPnd-b`Kt0);~C6W}S2Vb4S_x zH{W7k_hQfB`ff9V+})cq_x!lrzyH^QJ%oJw{Cko=eiDCkbNVO#i~Z&XH}rfZ*u>A= zPCx5VXTo-w;47bh%d=pI($SCaTl(AA^PNK9(=!XM_A<#WPtLq?$PM=H^Iyp9wcY#t z?#rM4<)erA8|LpFOnTAE+`Z*B_K1hJ4el9maOr{iO+mwd9t-M6d{YdctFirtZ_^Kr z-o5nnwaZF}JhD^i_2Mo5fQRlH9Gv>5z5N-#`L%CNElypx!9RKDJ%eCk5L9?$f;Gtl zN;jPQP4KtnPHFn%xzaCNpI#bsR;F~}@@2t~`R=9BU3!$}-T9F7&}*+0zrArqaOvX7 z#cO{|2hU%Bnf>FalY{ELo1IY)e37{(Rmyz11D~z+XW;1Pho$HGm;Wu|c{7el4>@9) zy?ovF&KA=Tu*a@hX#ex3rS>709_%b==J#Qq-@|!grvrm|Pml5+y5?D@@3!Zs-<)|* z`m~=170;x2lo#j^+mJFVK&W2)b-$#lQk6k+W)K|jv zzwR9n^nK#8%$74han?WED>Hxk1?jJw5B48j_ObuM)n7REy=wWl<2z)&U*FB&yJe$) z&4jay%lc38&piG3^i74!?U#P7E%%%IqmZ6F@yqm(FQyd_8L(?`A_#k-p-%6-xU6S;Qhq~7i}tb|7BzG zl9`{T|90k@^dmR@>Yu(vQ2xE)%<*5?pC2}3Rohml`cG|rX?zj6(+rUh5&923+``lr_G?(8SpU`42&VFm}cjeZ_oxgo} zP*wlQ!HMtR?sr>LU)-`Zxcpw|edLxcVCg)^N6-~MPXo3^(2n%N`W zw0gId^H%e>ZBPGqdfUtTI9Knqr9JM%i_+JHyE*4=`eyLpyNFIFbEwNI_!`*-Uc``9@* zrr)qXN%yL}#M$@ifx-Ab)#ZF&RxYtWs(s8E%6r&{KlAMupBm*{@=BZCbNMdmFJ?WJ z?r*=8{`P_g(z_nI(doNm&c1&0s`P6UcJcii>g+=DP5ZpP4s^QJe_4F%nZ1I~Gj}*$ z`K*-wS9Rxe>$WcM;fVJ)*^N)HbH3bh%HVe0Bd+@%G+rzgz!nOw)=jltkIHv^D(g%KYvAxTLO~u^g0shHH{?(p!P>nyP z+vwo!x%2$4FHi7i?|X+|*?Nb)cK!-`-_gCDC%1alxpwgR{#VIGtKK_tak_ucM*F@U zHWY8VueR9tZ+H3aSCxui{Od#Ko7{y??$%z;tR8pTn|n_4pSkjh;$<86DfapE^OaL} zd$V}n)M@4Obo_BC$9(E$YLf$E`JnOs9gCmx&)a-xFk_bQzdo@`u()~K;IDgrZ6CVd z7Eb1oHvhCed)qH0M%#<{_l{2MTXv!Kb$ib5KiS94f7rQX@t)4)r{DFH*VLxJUU#Y= zeeuuY(Kr6Z|76v*PIkpa`=Zaz_Pr*-lx_J_YtC|w=QEIsz_;?kKno}E7U z?IWBGS1j{?dHUVKM_u`9`MGn?>MfnIr5Vn+BQJIOtbWj0w8MAKt-KE!@2+!(6=yg% zJpY?>#kG_D(Zip&f1bZG`0`XIef`?5fwT7}|IQ=tD{b-Z9i{#1w=K0lcB*~T?AwD~ zS8r8ner0m-YOjI(dx*aV-~T+NlwI~yu;B6Mf^myi4*2bxxq~kodfTe+zx>>OaF0iW zPaik-iggENj;hN#ckHvsId0Lf#fK*KE*}2gL;m-#%y!m%dy+HtzAx=AJFRySd{jrO> zu6*yn%HWMZpXLJu`#H03PiEfa?|klZLdp4|Z~s#Jwioz2-glON@v`Q0<>T7~zufeHaNwDme2JsJ9&XW z<(&S`qS@QqE1sI|+%Ugeso{!GoL_#Lxa$2YHx?f|<2L6Les}bN`=<`RZQCcD<1V<| z`EJvmnYFL;3xEATbg~Cdo`GFYyUfSM8QPdpKi1KEZxz?fbz;7Y+#yeCAER?`1m#SFPA4xa!M`>@&M1)Bir@Hv6Ol z&I$JQT<5{Y_tU>NcIEGBX6-4x=J9_03;*1Yw=GS2qpp#WL_Wq&R0WLy>aAuX=5%Zrmi}@{C;xp(q8FRJ0BU$-u7_+Z^!&4 zIOxnJf%a*is*Hc{j+X?79be<~w~*4+_AqDh8H@dU_TSw;ciB|?>m@_dJMDM6|3Z%w z%l;p}c;?`Z|Ee!8d;EL>zPY|n`sN2lu6%0eC(E_&eqywLQu~Acwcr2U zo^@a$c#yyIe)gSbq$fTwIqhA%T+bbM{0r+>F8R;k^ocK34t{6SiN$L!-*MH`l@Hm6 zuRYTrFyeq9d+OuG-Gc4?TyUU$-j{v+Uq5=ze|*Iq!SK?n>Fg<|+TUI^)t~z0qE&oe zytCnfgPpri>Na>)cx~~7ilu|I7wzv)Y&yXIbodVT$A7)u{_KJN_M73OdVO}kFYNFC zH1OwQ^~kh;$TdF=-2c^I(?@)|Jl*w$x9vSY9OjI@uGiqdUNS%Z!x}ydyt03K)QU0o zz*Vo5pOv>ww~K2Ao}1o!T+7ps-?pQDZ+6@CQs*K6<4rT11wXp>r#m&IKRfaY`{s=^ zikj`?yDVARz4zYE$xqy7uiWyJ^pac8a(3GJf^<%}Kcl81;wUUg#S0zY?5wSC0yyZY_d?O=cX zSiRrp$3s^wdZxGi<7ZRKJ=}Ns<^HnPpUe4Q`Af>VG~c((&)hZ~TwA@pzrI^Hd;ai$ z+C5+6y*oV*-R(>N@C{Ra@fR$)apmIN^x)4sA0G7J6PK@gxc-d68|OK}%;TJmgd2Q!^D%yD=>)%On|F&1BQ8!qH)&Sz!M%gZf9K)a zZ@2NM-FsJ1tet8v`^SyNjjueBes<$9-x{+n{YS%@rJwI#SloW!+TiBO*ZeaVoS%O1 zh(rD3KReBTbNoK5&VKe^!N}WQ4;rVv>F<8n-a+>82g-A~GW)ncXU#|DepZ~?JJ{~# z3rpLNtnoKjy=iwnr72kb^~-~M@Vk*Sem}i*$hF7WCv@A{-s-+{{2h0$DsH;`LjT}7 zb;0W^tBMP*{Yx<8=EI!nC(iTtytFs}cJDl*I+1Cx-Gkx+?>+Sm=e9QOyeV^`r`y%_dH}k>;VG@^gpuy0J~rRegls@HrwAi z=l}kqNVR&{^ND#$olW8Yt%v^qk39@b^&8NyA3YrLf7`?VU-btLN~co)pX#?&TUH); zAsYZqXDw?&6MF!nM$760ZGcXNz7N{*agh_E#xN)uHQ-t{&Sks6m&5AJf4b`0iP19Bo;{(N)5y&^-h%eIL+@TvHv#DrXOD>Q6P5 zs{)^perl1mwl2uZp}G@+YzEYzYX;VH9l7AgcGy+ll8@x-xUSY?iAR?Q9(K(^J!_87 zqDz5ifcmurF7;)FfI9LMuNd`G3%Y=JfQMf*$b%3cSH4^>zC*EZ1|d*w+0^@Hki@?Q zS_>+$DZg@z&&E9(c+Dk(YCY;JLiUVh-3zanM3$Uu*mHa=Y^pN=B*9{!defn*DZV00 zPS82Sykf{FG{CEe-c6R(6Q4YEJWzg}ha@_!N%1MD>huNm)&HG`^2uh_t$Dm&ib&)%LJnX%*=$O|S zY!iU$_K=QsYF?_*POR=@FO2t6JwJwCcFjw@SOWGWau1+*-8)*7&YA_(xAr!K_YD2v zA3_zAZ;bm=f9=4R-j8gNH-x1Z(@Jt4$%*7itjgr zZwjan%_IU1pioUs=y3cjceU6Rt+`3(nK-@y zf9J3N#7Xnp>n0*l}B|H@1cvx zAvr>>MYfQ&HW#@dicWhwj5y6fwNyjA1)lGKa>?HSj~u!`M3qam-45D48{a1`bR=j& zuh#_^-*WHFLHAU-HoWq+R>yWx&0Mp}L%R@{1f6>Ed3>y}BMu)A$X9th^y)=E(lvnc zbCZ0kFKG6fMF?-f3j?z8y!2ewrJjw6=b3=ld@N)as29bQ`+(J)k{$LIsPsAmnva1O zij6oDn}RA|0-gGlPW}d3K<`0?LR0*HPxF$jIPI67XSL$j0H}qnJGvZxf(IIai(ay3 zDW3=w@tXkd1hQ#{nuBI-+ z?LfY=McDH2Eua>CGpNk+Ob4dP`k&Xg2)))-FG^kYmV%eQ2+BDnqqj^vFFjASA7NuG zZ-(zp-o8M2R8MOfk8AjFs0}9Y4Q9d{x6m9-EE#Qu&q3>(Gl9h*OQB=rtGB&k-{Y*^r~5!!{LL zgdBn~@F99VKUBj)r%UTef;rfAZ|PZQW#jysjpot`^&6qnjB@ardni+- zuQlt8T88%_$eOo{O}P}SeqCriXaVwH*kPAmeH5Uz)vJ<#eQAJdrUTGvEiH}lTw0OEYv#IY zh)-a5fo1qFjN>$C)r-(+t+^&-Y)SO0FTXzM>X0LN?VZ+M2lbFcVrsGJJx(nc0s3U) zxg@aFAdA-HW8iDy)vxmFe5fa5;n+Q4H|#2|)o}9dM}~A=hKmn({zC z>c0ki3$oVU09Bo;tVII+uout?s#gnIKyU2hv9&eg2OS5>`*?IrLt+)5lpX&x{vN#B z;jfrsK=t)?YECx%QosPsr2*Ow^vu&dH2V~Kt^(9m^Q3I;Iod}-*T1#_vJd*;{AX~(#re@y-Ub!ow>fMlE^|Yoc z>;{@Zuimr<54(EOxzPDwes{j0X-*`+Ct)!g#L4uwtyBY^?x z!2+tOz9SRweP>v>%p<&#LtD*F4pW zL9cmBmqShwBb$2H9%=rre4v_*?CQ&tzB>N?T2FL+@HhBbz@T@b?fBGy?$|`NjzUv> z-qq)D^xDTtblMm7odnuftDGmo)+&E!E*sx_7V-rA)T?ID2i~AdKvU4V>iF5MnI+(B(P?j{f=|Iv&>hIG zHE7+#kp}>+ReIIWq4$8|D`ev;_lZq0nzQ1x7Y!fXH z5AOjz112JC{hEV(T&M+=Jt>*lKc9_zkzTcgKhRm^HU?i1qU+ui^Qx)Z9?+%v6oA$c znfN)ZbEbJmVQf<`HJ~yZ?^6-J0G|We+agd-tsQ|=Er)uGE($2mkjbOzMIk86t8 zIh|ZCvFalvemHUxl<(aRU7i^2+vKMBEL35y5zs4V1Acj+HTA(qYfi$;uKqO#LyZ(v zdE`Hb_%_MtBSR1HA+(_}?oo0AA6+3j`6!>(WnnA#1gkSPqKQ7xS@?KBm-a|=GU)q4 zwMO+Ky1XfVMr)sS?yW{(*aQ61=(LZj6{7P%k~q~%4{qmQJ6S6`ZQ zdDiec<8{D-S3WD-@f*_*WIUo8QuuOdJ4Ace9@|9Kiu@LK$%1B(0P0;mNBB>NHbBe08d`wQ0qHcSBvj9yzF;I! z4ns~0U#(qhr~%qvgHHYhkVL1S&1fFV(*;@}R=rP$Z-LjoXg^$_`%P!99i1VM`lyE& zx}*1ydk`mm`AmqSSA7dU1bwiF#A?q2pn8grfb^PU**CQ!rc z8T16o=~IcTQZ2knvSC-5onOBm12L#H#?01SCL^Gc^VYR)<%V}PlS*HQ~D zpG(=f)W7`It9bbj0NPhw>P`MyN4t1~5`;kO)?6ca>E-i&Q+#%{hnlz6T_7%%jeFD# zB76-tt*-`K=d5$sJo!Q^u}K~S_3(G$wVv{Qj8BA4d!_qId#5}N_-uq%zj^pc@NLlH z&?I)n8{%@v7F4o@Jra+7D0;;<%Z{v>yaC?;Uk{=%p5Xv=+6V1Ph+dcKSOy=c_Sk?V z=+vvfFpSTPVv>#VYgUpxT5B^p)!vNW%Eq>GKjc(C**vIns&5y)dRLEvhdh=%>eU6Z zwE)e>4Y9-5khcw60z~+$ALUbTnycy1mwRoD>v{Mne=QhCe$86-Mb($~R{0fM&H=Cf zErWgydx_sPO#0(S0I0?1(SGPeiA)F&?U!wxBcEL-+(xFQS^eWdO7m7@!_> zsm|<T^Z^}!toJE-Y!R@+xSt%hB(m1m1G_G*U3;foT7%+UsC+eZ zeQmGYlD!UHJ=Dr}oD&n*X-4-wzP-^Ws^fVpZ;0JsQ*O;7LKc-i3gg)F-otn53^vU| z`&?+WFi;=4Gp=t_{M^v2M*$a~&0uO4o;h_vrx^+L#CgPf&^Dk;Yjr~y6Ysg!CpjTI zIz3nQo*R2fDZ5AwVpOG6It_1VNW3Ep{gT47uo`}hB~O`XUJ{f<*PY2z^h;B z#cSX4=+&p<)u$z}juqG>%UAI(RI!R_1|G5rJJ!&#cB?wxyLRlFk9tu|&uskq;h|G5 z#j1{hCaXJg!7E=LIRe^054b?{ZE1{S%RLgV{i;QmgH~WKKqGP#&{^=3Wedrfglc{U zs(w8A$p*@Mht0*$La!dAO8^VrMW-3L_^NmLL_mH9+A2MWWCIcU6mlpTUOjaFJDOqH z*fz9@9mp>@SqF1c~dMg{x#Gq?Mw+>x%b-V`4kOFwf>LJ8uGqfFOomzvS zUbH5mDjWB#Gp&7AF8Rw(c`Wj&ZbUBS(7bbCJvtYvI)aD2+y{IE`V{bpEyqDykPXz} zCtukuWDC9kp96*JIL3mefP5>dBY%Tz!FS$Qig8W6J{NlkB6K~n@r-p|HAl^`o>=v$ zym@5#beNLVx4-5!nlWK`0`$1*GzQJSlDWiBm8yM$_5g$hX5KvwJxngGjg%3&q+-kJ($>6 zwG!AYXdY^yWAX9e)zeydlU>X?0m-UBW8H4_s7&8HpNg;x&^U}!de zwyBp0oBC7EJT{$UJ@=$*f^P(a@D=DDg4g?oKcMQd9(zQt6!!8SXa=mc4qF>K)lA`2 zt}Q=!i&*tt1ghtu&qGt_m8Z5ko>?t?*^X?HuL^q`y!KW#G!ykFxadY9>x?Df+tKAf z0!$^&pi|6L?21<&4}|cls~FWaU;sYN(7xzHLmaXTO`=mB#pKaxCX&m*-X%*sR6Q4} zE$btE#)4W9qIZF)t~z{D_$09zQ&yGpYxbJAayA3)YZ{*jDkx61HERoA>nnl+yq@iO z_%Tp}+yY+#FCVQjM5ndqBtzw6RGYbY_$h~STGjEMc+e2Li>?)@mrC;V#5a#z54yms zry6Vq-y(b%vSz5KSa`iZsRCL@`F#pDcs#19Z72uBTcNyySXp1t4BI z3^w&9D6V`zTJif6&0fChIp1NEy)Ux*)eHu}PsGnvZe$Otm}Yc+K!bE43DGyBpMWlj z&V{Cs+u(=7E59W@=U?`j=rot^(1>`&)`A7dvdLe0b5ILJCa#^x#^5ch-U1P#}Mmb zGoS+0qL;0{F|HrtqZ(CNC{Ry=_QJ)lju`c>pFPS?J*kHR)Bw#&d)NSKfcR8n9P1h= z(51QwAfJg@dW33!M3dqh+h& zTC2`!8#?VxIUl+fpt&WWmSIg$L2MFnT zYVJLoVw!+gzgmm-!a~-4tDM@Gh@8^NFNeGtg!t-GuUeyOHiHnGrC3oe<&;f+qB(p8 z`KlM!#A|4-j{8;KBzomZ0?kSD8%50XKz*1jdU(ai-hwUUS7?S;m$?j({O1LRLOYG6>PR9Z-%=KM$ReAGBSR%b56QYc779fqK(@q5N%4 z@yr|0xnaCs%|L6Bulkm+awX&gRL{aEgg4l0p%yyj)p{GC@>N|6yL`k8ZP-0z54&P? zZx!SVuYTk|0G-aM>Q0BRLtg+XbazT$jhlgDQ=IJOJw)e*fcoDe_$udAXcBpRHePcJyy|If zA++2(I+q*~P~8yP1OG_AK<7CG!_bXsitFTL!*4dS)})?m(Yax)ParG5;&bqxiSO+Q zYS1f2>uLc8o%E_LI;Y7hU+TXvd`%*C2Z%vqG1E_t2>y_0lREIU2A_uY0*2o$`ms^67z2`O7oGFNaQY z5?+0)jsY#8H@@Y*&})qmP#)DtU{4`yjhd6@?jhI0t7rK|`04&o%@jHd*+cJw5a|9> z%>cW4RbAz@pb4nHPp8}kpxHJ6eLuakzWUzW80ShtCu37RLv9lSbc$<*inq|~sudr` z?}Z!S$7JaX*{hE4@z7vKR&NuV;(x!TuV-|x36bKVJ&21;%qypIblNos^_@bm`=$tT z@JV!%hi2&mT0mC;)jG9Lit{>hTIjWA?Qa5Ffc6CP>zj>h>F0IY8_i0!(;a&8_recD z--g|U@jghOK-L^1>eRza*J;<8)BHVb(kp*1YgwN`E1{ZW2-K(5;giIcLe4?cAVlxs zqqCuNwiaDQ7Jqb8;YBTcq7XZJJu6j92#L`ed!x(4-wSWxm9N|OuO*;2*|kzZrckJ~jBXV;_nvA3?F5dxBv~O1UU;pxYI`SyL z{A=+Uj<0w<`&*HFBWn)2o`7EtB4pLpY(f*ySm(4$HtuySRQsW`yb)9(zY5iCI@jI> zxrQ8_{U_m{GrvD~1}Iq4oI4U$x4Y`VZr?Rn7x%nYh-p9@t!6UO(>xNQ9Omwd)y>%2!qBVyHa3vqpr6~7ef5?`JP zBmqAhpByv~euVFikNPe9kase=PF{Vx*m|QY?+5lAayzkl&S~wDiT#_=N9gs8D(@}o zzRJa=Im_NNOJ6|wBjnJ;@db41X#iB`p#mx&V8CWjhi(!uA>PQ7;a&L3Z2a|ITO;{^@{dFpVH<}}gl`gAultptJ5Z07 ziTfOhtUO6{c`yfA`8B&#BS2QaJ%G-h=52^o4f*OGa*xt zu5n-~dd*)mR;-})Y0cUntz|uO0$GsVKsSRL^pdr-3 zjdAW2`mxBO258L%FgzQ}ld<)MPoh&E*TlYx*DN&;3!en&f}i>w2d_0~hU#5$MP$8? z=z^{{kdI~;0^PTYSMTk}DR|4ov)A0Z1FcDORG&Gh>PkNXG{9?q5w<*hJ2X-*sCp}) zQ{7r*^`$!XP+eND!KOVO09CHg#5FX31D}K{pK7fHx_2z}2DqRGc_dJdI`WOrVk3_S z)Q{ftY0c_gHaEl$D&S?;xk+HN@X=fvz@5NCuUg8#5a`^fcg1M^!z6>q#ItvyiqDZN zCq2HZBYC>?Kr!X>KzutW?<+cwSQltdlw182ps5c36#Rf}d^QX=E2Ix>+SeK|7G8D6 z;!^>if>&CQ#9RRF`Msb7~Q% zb?1O)ovM!edzn~0CoE_KHW&X;fKQ=|;3N3n@X{yYz3O*~k$=UkhJ=*QGi`(UoVA zMTS>?3%xOMt|Yu>W1&w$C*U_3zgqY%@bW8P_<8a}7r_q`HSu#yy(y1!d%#66xo4AA z&a3z59y+a0`>*@Uf|t%f#kVvP%T>_ghE{g&xAHh_2d|i*)HSv4f6golk&g0voDZXFTyVjxpYq3v2uYJ=wD(}Df zZ%)NV_*wXf*VnjuZd(SQFz!=pQ7)}RJ|VUW`M|f5vkkm~oB$rSLN@MK{_P!5{Wf^5 zBM%fSIl^DObRN*`5>R~|t@CbS*SY%!emHQECquhu(Kp8LN$!L%;4=Wdu8GLXseD@V zO!Vmvo%CAI2>Eu@HTa0vEDe51Xb8G=_=I81YmTn@@62l(iNh~KXNB?pm-C>Xim%R; z_CY?(bgPWrON@ zpnluozbA*zjn?FvvTJ#E1#C(DG_%e#FTXsx$>=mg3t8u-7CHdh2GvuVz!$BRAutwMIZby|L-Jvl6}NLgYL)VJN(O zDzIr!+OTU)s-^ja=6|aNKO8^#YYiUqIAr-~UO&R?JgE=GYyXsEDEb7nM>akmE>zce zX!$&cv0Zag&lTt+bYq(0_c8L7pVs7|SKsn820hT8s9uUV(-`-rJ?laIW9Ved;o|`d zJ^}_<$ldWNK#d_6^G?D?@X|Fj#`~9mZ-qBdtAmF4R1sU-p{s|=KL_g356i~S$Ou2J zM|FEc)k6rcTpl!nmM^F=aA0Ute0JrVYK+(Dq1Tz11XAd=k4FAnT1y3b19YxjY&!q) z3r*aU=28F?Wk;_y6@iD{C0>_etuVf?Yp~^j{I&Kz&PxvBKn?cteIy+|+D8i*WceG@5s$CtFbn??*^YCmSnRcm zL8m@LWDmI#zHd`}f2dEb)eW&jWmnHSZ{K5^gzt2y`pkn6zX-e5sJ>n3N7$v;KIr+@ zj(!9*32Ly{Lv^p|%!g3bNjTp!?3Au z4_WiodMfa%gQ`x1PkAk(6jz_(rG^xXYkh^w8JOKA-^2F5E3UJ`Bo5@Y>fRq zd{o22Uwf~0Taw{bS8MKqO*-|dd1)_{ud^@NjEVQv#V(uT6Y#1ddo8?ydXl+96WgK? z0h{8~gU+dJg5p%GJS${Zu~5y$NCt{g4M8<5sID-K*DL>8bgH9SwZP})N1Yxh6CP>aa!Fv_~m;Q4haSP4WBZ0BWJvd{syN=>95S$nsZStu-|9 z%+!(k!YR(aS8_?B;tmmJ4h)g_x3+iGwKy%c*Qux%vhhe-o1|Rt;PJUXC!N)_F zN1gz$8d|#vEim#0 zHQ18KT2~%k`UKR1hCpXZdKX^3tG7-xbm(Q1@Axcd6#I&%IIr%z0zP`~wWHH>vj|*# zJm3#B5N&r#;5RCew{xL==+5uKr4`5(7Dl`)#7Wi z@vK7PN5QwCTMS(RdNfg={8mTKe23fuorPW30Qh$7+Os6ky4GS-UDfj7r^D<0u7MWt z84KTnpVnr8;_HEH;-5Y1etQES<*sdvXVU;3if*{{pmQB-@lowoV4;5>xx6pP31Upx zF-xfQI+qD-;zM-C#5r5gO(1pxInPLI!g|S|KdcUH6 z<)_)_$fvcG_X*u%bXr4qsP?G~Hp%7Z5^`h!)Iv7~n`WSX$KtPbYA=+#1)VOfUo&(K zLdbidY`m}KKG7$!cLBxfj<0?25ue1S-d(8Ht$y-AIVM7N-=%;qJ%`k*=s03jVk4K>1ZyxpMGdL6!RrQTkWz$nZ5%mij=s8n7(|O+fWL{FlSazdWOC z>|+sI2l7CBuCq1{Xf~>=-ZfXPt3{}eXJ^Q#*rE81z}^$+Qk-;~v8k4kjBNsR4p2XR zfs4JCT;n)X{k6!-)d$F@J5+Hav++HEHvZjs)d+!!`)yENpuKJfT6+qF zKzS3;3aI?XG;uEg#n++NInIg-sV6A@<2Az80!nr2-Kjr&}%)4*SyMp=oIH6 z%Pzfo(5zh(`zpWoD#GreQx64T!H11;|1USibC6#$8`l{RRbQ&t4h`{F4)suvzb;qQ z#6IKT4SqrnpT)@XHDRoiUmlz4Yev$i@d?qV(8;cPsYX~Z6k98D4YUeW0+$&1wqP$n zL*yjXf=ZSxf>)h}#`w9vl(-_ihhDu|@R5n%2dF>UwD-zcfp2qlyvL%-Sp#%ms3(^k z4al0W2aTXY0fgkV&}oK>(>z`UNo*+)VeiZ-fBEKtE1pX^EfdF%AWpLBfa29-GkW!_ z{IZGWq3T;R(Ee+`T=J-v_z0cWZlTjvi>w@ae&z7-OgzU3-T=i|%11pxYt{@@M`zXo zs_S*|noR?;fl8MIli;;}*TlK+g;z|7tgHMQ2UT2WZyr9{?*ueY4l5hWo6+?~mw>9T zFzo0NS#y!EX0F~PR}iOuwLXLV7BRZ7HBX&8t)~s%GVtgX(^lPazoYYz6{ot&6{2fL zE&%P9)~owPvr(_L=;R}N4!I6ly);V(%4Z?xp)JsQd}JS*WnI7`R_jyl1U||m|550Q zP=mb)Z-sF_^&g_|i%#>?Jd)WsJ~CE(zR~r-uKeY_Zj4{&$}=S9M{J#EMfF1BJYqt4 z?Z0}}3>%;kP@LOAQ^bx0dX^eu+OZ9TkKl9g%FzN+@S2r@FTl$;2d}+K0PTSXT07*m z__hF@o6dREV^_TPQqSZrAc1@*RBP@7dVm5tz3#|QGgJNv$e|$~K99X8eipPHXss4d z{S?@Uuj;7Bu)|i1tUYNFlna_f*9z+7kDvBP`;bHDni$HNL?>CaeBV~beaqGWv=-%( zO?}P5Pv_+gsC=5Mv56;#*0LE?APd?z%_W7d4y4iLk#*0hr|#J}CQ>Z=1oCvC{v}&b z1JydT*095`fUGP*$5 z%i)uQCP0Lr^5#2u`04x(hfg-e&!!&G26W}UCr1%}4E9R&8$lBo4%*Nw zcL83r&|c^qs88K{ij!TlkY4l9xlwNQR}V%aM<#xjwp7RWo%|BOAS+I@P>lAtb9ROp zD;uxJgmJv=^`ITNjq$9s1`FQ^ANdu~FKj}`9*==m0L^kJRIxdD&AJB2w@VgZ6F+lQ z+u&!=C*X%+OTeo~<=5398CxDC;d9W=JnBoaRv16~<)>MDQ2DPXpY}of5}J73nrj0# z)iKZry{HA0%LUp8)$(M6YHvcQi|kd$z9D>hACT2Yd7q%_O*M382V~>-ZJLGVqa4~t z-8b@)uVxsLOY2P%la~%^5u^Rqr86$R2XUc^zn3ry-a^uRc`M4dZjJ+y=eQ zMFcG5TBvK_xjgBrV_ANxsoJU+n)tm=gdB$Pyh3=%TATRtS{sSOo)-n8RqDaSwNl6i zzPSlEc%6v^d_KzrMi93ai~`c@Y#H>X5x{%! z1F+?gCm~zJ)xp;S-E-QL=};HBTo0=At=`n9a=O3}*MO|^8NwT46q|%j026`Y+8g6q z5k5W8Y0s4Fk8FIecSrXnP*3W!bAJ=qU19{yzWgjFzb8L*&B~LFbLe{$7IG_8b4$e{WXK3Zh71{Dlp#aLp%5ZzissT3O)?bCp(u)^DVodD zJd&Z9J@?WK$_jFtDS-{sjI-wunrFYN= zhAsV}7jqtT_G0he4{I0rugP%^%(Wy&-TV!M53%kl5v$%9cjRN<6?)Gq!bkWi*4VSY z)>xxA*19e99=I#dGef@x?}2%iz`1fkMmOIxXd>qX)aes|cf*-S_#^Z`0CV_vP-k<; z6EWs1;5+Oe!3AhR#O5r+ruIIvT;GISO3p)3d;LT^=B{u8KXVlkR=lD$I+@X*9 z$NH{Zh|pI>MX3pkG1lF~KX_n=ujGV(UQY>n>ufJ2=_a9y4*q9wP^A_SeWC zo;|i}pzWxyyNmG7Yd`M8yZvH|Zh>FH75-uOBRUa?v5yIwf&ID92yV3Zb=T&I*w;{d z*(-x_Q{{*FaLD0Vw%7{!&fvYg@6$bcbFPp_d1wLVPvo+< zs17&y@2{{@;z}M1v&xGJwfel1?m-{h?`PG}(GxTvU zQ{=7?Q{a`K!P{?K5C;PG$2~hkXWtQi34euc0PcN+dafPxm)PA|x2f|u-#ajo$Fs4{ z1eI^MwU-@n8NI#v?QxbBwj1~r{QNf{_WCPya}ILdqQAiIw@jZ?Z0`3ZFo%Bj?A;om z^Ip(d-@JcD{~6o@V-h-hj^uvb*MF~EEb2^wIP1K?&z&UDfC*j1=Ij%3zsLRxct+l* z4xb$e_!aCK+F<*P&A#k;g>DJnz;^=P*^H6u6M%6~dVJA6f=}4(;T!lLz<_LhXR+V8 zuY-^0q`zmehVJnB9rz0{uX{9pio^nY&d8qeEz~;}q1rP18Wb=UD6p>x3U+DShS}G1 za4!3Ngx)=!L(LnoH3v+{5lqCJU%y-`BD!H){~eRzJqPzS;5X;QZx4Tk&OG1uVXj^9 zb>2Jl-lxy-&XIt7ontrn@=USzRM5}8p&PO7fM@Qx?BQ3~-D`#hsIwRJ^5zV&o{x3x z+45(qrUdEgSXCOWWOcWcumX*+}SnsJ@A~JMf(Jd{{o%?YpQeZ zHGTu~S5W8hJj`tm+Ec81>EK&nY!rvQ(gwMJ0AC^$-g`MM>V0jX7a*aRUqWxe1$8c= z?jsUsUeC$ra$$_1Fm-3*I*cKWAS^&|WV{?4Y~ zIdRS~z#qfE#kR+m(VfC~=nJv-IKlrFpGTkpk3qmzz#5x5oGX=@J1wvSb2aE+^fmVi zoi(qCF9)C57oW3(PYW-+7uNP(OvJcXWzWbN?t%NUw;jka%+G7=*XZ4YwTyj(&Y#B} zdUfu41(fHVz)!K(mD;oK$34jS=S9U0@~ds_&6)KZuyy2Cww8N61Cc!L(;5kV0WEw1 zYY;&|H^usMV!uI#H%_}dFlGg>yaW@nJzRiyebwn_T+j}#kVohMmhg#~2Hw2(I0E-p zkUdZRI&@dk18-jmU5mU2)?Gq9lWX+y=CWtw3)H_ZgAbzcnU(S`C9bmn2;mb;lBjNt%19-M|D!`8snXCFY3&z?A=PG`nr4X>udB4 z`a3Y^!aiVgjupBVIGa7$^9`~&eRfda74ukA`y^t0|IE8QAeXuW-;-N(?!xm3_-fPF z_$%nMzVKlmH?ioe+fIgH8XLXJVzXIRE-+~U=SueJ=mpgd( zQJ~INV%^6Cy~5{FU%vsyJLd?Nz`Zn3>q)J*-_*ZT*k6M88=&4AKF(mRuZg`>2Mw{# zqMSi2P;^Ud6ZCywbD7WGS=$+IfO$vcYmk7Ce%}ytgKWR|$SDHuxG~0m!fuZpaL@MC zfFtw;-vBPR)Gx#V<05tukeB!wH-HO}WBuJQLMLE<_EF$-sd>!rGi=r&x2e$o@Qd33lk6(cUw#&jkG*oA>e>=fOhC6Y~HN|{}J6I(C7>1Tu}G+3cWSmjqgiK3(J#e1gL>iLqW=K=+kyF$KEQnT^BBE-4Dfqk zOxA}PSLm0(-8n;v6#?pVjei0K7~}jcu%8K8|9K{4W9)wgZ%yl5BWLvUeM4_e?~b(^ zknriC%T1W?zh~iHwpZ(p;FMfXfO+=F#u?win=eAmX&(unQ0kr37O<6rzQ8~1yM^~& zxuZ4q3>}~W%)Q21Z$ggP^mi8z@fXDUyLtz{>g)YEh0Zk(XW9~Le$UWaOOSv+;~r`h zd;+==4DgA%#@UZC9Z+|vABezS0+_%X>??FlkB-Ol*{K6_uh@*svEBu9e*0jf&prII zuRflazV3Vp?&LuN@nCbl0h>MCpzn+Y8UCDH{>|bQ_y+t5OtI#1=HCG03p&p*Lf6~F zUZKaJL%#&pm72e7s{BM>boL_tH*$l`Z;dtW>mK_T_!hW(_h!t$Apah?mk7K&6FD0A zBVg?i>n?Wqf9mVro!xw21ABKy_iArH%9r{bNb<41XZCDQ{yik4TSCp*fG2(Z@0Z+f z$oCKROcFVc(HHFh0ZrJvTM???EqZCePhaHWoAYs$* z5~@BzpF;;|#$VpPMsU&ByYE?8>))X7fIG8JDD^(rYiA5T=h&THzJ&&;v$w!^@zKY|C_KgkNSWDk`((&-J+g{dp6H2eSvq$`_Q5D4vTxBYz=o+BKwyI z1Nu2Hy3hk2cc#xZ_H%Xc37vk%A8o6BZ7e?S;W2*J@O+#@|L5>6e1Ko#V;@)066+4# zo$u-$I&(Krd$i_=?EDcr5x+uj-9)VSA)p`7?GN^XPr&BgwXYL=CghL(VTai5XTWAJ zJCK3BWzfRU_2IoY=9}jr-wWdGrz2O6hdz)aG3Iclh`&BJz#fb>uX^okARnOCcVDCW zP2EciwQh{~U<;-4>OBYb6S@K3KGo+fI;e8Sc1&EsZ-s0R1sbtGLvNiHx!`B+hqJYh z0oygO&lBWKjtgXS2lx*C+il(HoClwR-n(Tju>;Ctgtu-w@XdjC=--p)JK(#VO6?~? zBeo8|CAbH_!F~-jrs=U$&%7-%&~_?bP1ZE*0}-CiM!p@-af$Fw>j3>^t1n&N4@|Bxj-kNUj~=JewWI|ESL-a z3HkI&*RAzrk*P@1*BptbJwb`0UU%2i+1nKe2x6jC!Tjrf#Boynb>)4LkLxT$()cD8HayoFp#__v2x z@0z)`&;aaV4JQ1CMZE`D=6Az9d+q4W<=pn@-Y!7IX8nTBbAOI5L7k-qk3mK@#(4Lz z0}1&8*}5s#-CiNT+~Ui#gg0gfHTOAs?@oqqh>5YjoAx%rNBk041M7KC3BB{&gC~96 zx%Kt+z0el*K^NeC{z_adFzDQ$zSh$xp+6>%`Gw~`zc=u6PsDccJ}3CyE&krGdfx@} zc+Z?;0zS9MDZ(76d(bCr>ig(SOL%9Je?;zt>=}&6&XR%s*sr>~P2Ft^Z+@`{_V9r? zX+d6s0G#U{opIM-0-nzeIJF*BoB9TR2c+u1BA@--fW#bo=o@qse1Y#4_5ZH3ccTS6 zbjB`$KISM;>)6MqZLMuzKK2nKzw2Oo4tB_%MFex7oBDgu71aA?-08N`Bj^SCuh6g1 zyDNQK=n8!c5^{_+pE=g*7d1XZ_XmCrU4k?Gt?Qli-$5^+Z$ZK4UL(|)J1`;_a%T{b zoxz=YR_aBA+@d!wZEAkJ=ey!Pu^)Xxth3m+eg&I3uR)`XUHuB#ovCwA>gT6a{Tka2 z+buBvykB(YpOBaEBeoBH?P-9wPQ)g!k2O2syaD+h>i6iMf%}#}1Cu`B3Z!lQ`;Y(T zvd?SaS=g_&(jhMy=g0nYp)S-M3GCtB_bt4(25N2lkaw1Ox6l@-S2y9Cq45B1A<{bZ zh}!o+--urU0e!)Khfdk}b9^%L@s=Gy?J1xS$Rqd~Xm@tc*}JUWUEbikJorEK9%8+J z1NMwT|cJ{kC2V=@%yFi zCAu{_{dd|x1Kj0^PMdf987RokbA!(QCTJ?e0P_v#)*v4Ak(k+zi#{*=+V2XVj2vTq z_a~^i2K=qn!W(P94Ze^1+EXOP9()QuA=a8BygQrirS>>M3vkBA>w8$k^48&;f zz-N4(19cg@G0x)MuwUmhHUN7P*Bj)*9fG%a-{nuh{dq3p1liwF{tEii|DxD$6dQPVlxqkZ|=2Aa}4}L%u{oh!+pno119V@;61tg#!j)y^E}w?{}Fn78Gy6R z`QSTXomtnT=kZSYlu~DQH*0)a>~~URW9GeXYVY^Z43_9atam)%XMg6|LG8!g`cAyU z_8Po_cgD}~?(+)10q)T{fn1UJcks^Wo_>OV3BG_&z$anz97?S7dhb@ioO>{#cc~#&UKmne=d9N3>k0*US3wsIVh}6jVS;xmc+WRZ~GSnT- ze(>h;-fPd~^B%15Yr#`^Yv?n@x)0}b2QPsBKSSM<@3ZgW2tUsQKi5aT?C}H5rtW%4 zjCI_jdrt5JaAx;pkH$_w{U>04>plZV!2YMm?)!SS{{mI_4Yq6I=AA9-tj=Js_8y5j z$M+@pjIVPaLw(QhuzR0w;QwFX9Pgn&0%y#`+UxV4QpCN0KY<^Kdk$}|5!&HvzbEka z^R%x$zkohM4&-@zpzol@t6xFCg?``H|4!3Kc=xb7*oOr@ptIk{`1}v(&*1i8Z_ows z7~K~zpDVTo-W{xgbK1|3+OdU$Js{sM{~!7iYs{V)@79R^0(b{n@D;w!y8;>CBV@npZ^0#Y_t64@9o-_Gb^)a);&`;r#|%$K4Elr43Yj2JZPBdJVnv>o^K z4<_&em|NTO5bHZJKhT2SU1Vsps54B^*V}q8o|3i%RjW^Dih&@E=0%MCj^kT~_z}fde`>+W$2Kg1X2vXtRAn&l9AnTXW85f~X z`pU1cH{jv9gxAlU_PAbDzdLgH4tS4ScnljMWzKo8X=E9(@Cp zuaS*)*Us%c`W$a-uLb=I`N6+~FQv|G>~r$&54x)bHsF)c1^9dD5^4<}eLR;E>8;dQ zYd<{BaIIXZPi}2V9=aEsYO{w4S)UFXBG}e%hwqR%^dC3%9-Cu8Hct@9KK2>GQ(_AE z5oGL5U+;sxehY5VIlHkj9^MV8`3t%XU4tZ#-kyAaMq-`a8Z9WmbF=>e{eWyQ8D3qA zHOHqOJ^Ur{&JjUIZlT@-b4y3$C74Q$vxf2-+`t=S9^-b%&&d~|!g%L$Mv=(TArHWJ zE#dRv`vrMK@B2QHZ;5TrPfSF%p0S>RHE!S&dY=;UDV5IsV?XWd&#epcxch{^cf+0{ zbk2=Uy|p^vy=;)}s|EVI+W=O`_J0R8RzLFw_z67BYL9cDvBu?tPf{*5W{yJ~@coQE zL+xV)c3{}leOWie+J|)-WD(KL_eCDPvfmJW*U0)tY{vV3lu~(Tv`^<~ zc_hYoXVkw%{$^Y4`nKrIu|&UukLbh^IacV5Ujxrbe|;Zg6Xx{%F0sFd`hLq_1N#}E zOXvuWushQyXhxsO^$6bmpF*#o&gbKO@8Bay5e0O4;Is0Ea|A1N*0bLU?9g?fLHCiI zA=W(R7?1~S&TB8m*h7OHfj${b@CE4SuI<^{dt$x=Z-ITC_4ORQm)0GCaZBt!0C%$n z0jz-kS^pAhE}s_aE<1ES5$ZfsoW)tr@b;;n{u6NCf^L2`{LgX#`X_yzA;vm)K<8sG z1N@t9y=V4t4^pXT8)7}9hyF6M{^kjNopbKk6ckt4H-n;D`ipUAQHMHf3{lO3FxxK_MVKeU|;;ds_ z08+6=x5wX}v_I_7oj)VqJPq`Oc;lA%6kzWy)c7f4gRa0ae(TNQE+Q9f*HGu~(EoN@ zcVT{Y=E=|_;;gmBuFjp=mww8LJf+lJ=JzbEm(kDPLHHJZ0Z*_kq33|Nc2pLX40j1;0Y~5&8r03|iy?n|)hapi*iH_9%1_{z z==9a*4EA830p1^1JwJUhz}7!t#^Ze13IunmVbty{Wc5}_= zw;>7kfV}MM&wXn?!Y7p)=l${TH$H3pizxN(of2a%=@L4Ee4xtKu>T0}xrA8fnuu@F z*`Iz=&!^kg@4)@0?%Y|e6|k@12dF)KkKZXU{|Z{jn}In!PoDS}UNx1LiA{ zzX_4=B`}xw*}XYmIKp}fOC!Lp5oI2_i7)` zG?8m6b+!O+PJ2+N&Dqa^G0tXR+MaH!pZbX3HTo0%=-Iph-5EX^I)T-O^8?OpK670_ zBenpvnIjft0WTe)_7H$M?zZsE<~P=}xdS8UfH|x`EbxIZ#IK=qE_5rPZLWcOk8O&` z=%(a6Sd?zsf{)x?= zTJ+w(5tutkH+7Gm&jt9oXD09pn7aYyOyn{?9;iLJw*j3t^Xk_@2dFyF{ubUxeU7zm z=&S$N*dqww7`U?n+`%RO<_qvU_!Q~ykYf$K!d9TS;B8;$)jr|xel!01?ZFZ>K)ukX zLw-!0e#Uvn;|4#FfqvGIUjuCyz$eCP_bg68L4OH713!R-oJu{nw4fI9T>l^HY%6pb zm~RQS)>&U`TYp41K+Q9P6*&Xc=L>#5-rFg%NA$*c4mSs0S^k&!&i5EUXSP;>cefe& z2zv*A2g1R2j_&ie_8y_9z#jaaMc+ic{S~NuEcGm-Hn5V?oBY*=p9?)aKK5_THT2!~ zaCf2BHBXCvf>(aX5%4#^@6SE2`%V2l>nq~F0fk%-_wfnx z9d`GW1U(mG%roKV%+`~?MmN`n_pIE{Q{ee$e2h1b_sBd0)LJQG2L+o?EVaiFYhGuy z-VU_*$Y*@a`H0-kV@|0t0o@sx_e&r8F7eezKETIV>vr%bJ-vW0pwS1Nz6GBKUY&C| zlkcCqQCGlQV)R?VpJ21Lvb(UBJCD#HV(ndB+SD3r=!DH!cYV~?I!EvWi1^xXf*SX* zy9u2?|Nj;L3BB?qNaV6-f6kjj-ku_~gZ@cA*7|#NMSfB5;2s))IlPk>;0Aew4p430 zQ{&h0+U$3?sr%081Gqz$H-E-Q-_vc~(Fh;=y7OG>-*JwS#lzhQ#1{OzP2G#Lc}5Z4 z68bARUQ;4?bwx~JKW-aeJji5svrz&_lUc|23^%6zWK&Y#F5y~9sx{tIjY zILi`!jP<)2pvJG%;g_*r1M9l`?|}21K>d9s;%~kqWc!(aH|y)2vDOpx=9zO5=geXa ztdnEyeU0q{zP|U?>cHc^(g3gTCAN>?ISAO?aResP?T?7b=nI(hZE9c6qs@Md$@o9q2k(jZ%NTX~r3Lw*BYN{E z?O+G?z#V01ga)t&1=y2$67Y;#bnaN4zEeckAN=8c8tmp6K#6sZ3|%3c!#MNJclB_G z<^O{`?#kLN==9svUFts~n{NX4s%{C5%HUC7?`DGU(0exuyt~MYn)@9-KW^*X_vjP) z-N81WJA5R@ya^i1AqRT%*t^eCiu{>e&J>VKtg~MsufbG!WwA$i&&d4tVebQU0@v^n znu$%o``xI6_dRQ|T?usN`@ZKL1OsT${}Vp{9c26i@-y^O<5nP)TDw4_dVGfsJ!|8+ zR)=i7GdcGN?AiNo%rQCjxm7L%ff4?}U*Eklek&<*#>XC<+uDh`6V$su+wgJr0YvTW zMjEhB2U`d4?0e**4d_#1y&Dl~4+)*|`UdLQe+O=W{RsPLkj-zuLVsZnZDQ_&y!Mr# z`kOyOJJ6yZp#hqp>h0rEPd)f0`Vpuv=rgg_Zjlpub6kK4{o8Hr&G-kudB^A{d>i~a zsI|=J``^Ob;}QIzo;(va>3{Z*>>i9g>+Air=K==sdu+aM4K`zLK_I?Brw9+?fwu?w zi0%acKW^)tS69#_?BT#W!vxOoX@RjTY(8sv`|%x0rS=if#mMYH+W=k@_bnLVjqzE- z?}0wTda=}TK9Ncc>^Z>4oI_K(EQ z_c+!&>${loiO>PNJ>5Yw^a$U8?H=BE=UpyP0dtSo*XZ1(IgER@t-d+d`*VfP9HTM# z8gC7I&=#>pd;_>g)^2Qz&s1uDb3XKDU(Q+t_JnWN4wCYs=FjMt_$<*s>Fe1!Ylqz$ z*8UT=_aMNJU?N7lwf)^8K_l>vn(Jaw&%j}A47UaHL)`*VKx+`(Wb_g0Jo4%$Fe(#kUjff51ONKVT1~)>|Qu z*!3$=dwhyr{T?LHiA9aEuNdoH^1U|C9EYv|?T?9{d1Ujfh@IfQZ!MVLOX!H7@t#Xs z)EMmz$oM;(`5xYl7N3Ca0=ho_}FWV_4_|TGgxEG==Ym?FSQr=q}@2Ewi_@AblN=2 z_uKlPR~$hTwh4WU-I_ta|5|HA^s~L7hmYDebx#HAY%O%|AKux_^^jwUoRLKVr`!5F z!46;h_nUYO2K>z1fvMD-_LG2d0ll?4W$Y21cJo-newXm(FVKwLS?+5Ex(jlMq1Nu;^pXcG7n6P{Itk<9m$ljrZtiScW+jrP5zzB*y*sS>-whrF&GM_YR z>+6}%ImqXoa8`d_*xzU33VBBO47J}3Ki^w&Ea82e!5kshZ@2m#x~ZUow%`@HoXdO> z-k%#;{(uk2VGEFtps$e|a6_)|kS$szFv^$C4TDW zo(i$>+RW#^2KWd(L(k4T)xb~WI_nQVGvNP%E)t(W*Z`gw>u!F5-y@Jq{YKl@BmCzF z>e+-+e{SYjZI|dKV2pEL!@q#q*E6U+d-nD?!FSLL=n1-cK2pZM=4{bh^A746?V*u6 z5uIl$Zyx{tIsp5z?h2X~^=$U=`urOEuYh}){fJ$oPxvSJBWw-w1n<5&umkq^=WYGp zi*i@aFYIXr0{-sJ-X{1Zyx&WG3bMW4V_Rx3b+68Q+SmV_#~F|D&711)zlXO#pKI{& zE;xfZ&hTxZ1F-H5aONxM*Ti+en&#Ur>b`xqW*_7TLacG_+`Q5^z}SLKUvo_068#!_ z4*f0m-+>NU-GuKB`UASKt^XOw*HHVkrhBwkb2-NdABefY-+SUc_zHdr-@=;XPg z_o;gFTX)_ay15>9ANS~ec8A6<o>xA-SHh5z(Wt_P2x~%f;w+L=*(-*+fDe-c{})6&@=O=ZSBY2 z-OGgD82b#do_j=YKtc9*9{*m?_ye#9>-xMxwoZuTM0W@5#d`X<6MI=gt!14#9^GeR?dJ-eXL^PF zBWTsfx+i<+kiP;WeiM4@S;xGd#}jPcp;M62Tg$yUqu+03dkMfe@8186Ww0;4^l8jD5aE z4~~I(ecYe#`8AmPEck2V9Dj}9*B~#DWA*iX-2HLS4En*J$wSTg6#q+jd$9kf*q(s< zgY56&J}tTq#8~IEPkXZ;cj7r^=m&kUS>p$Eejn#Mi_VxFt8ICRxnG3&xshU1KhHp% zXV#*>f%^WWzV2{^k8`=tiI}O>d+uxz*(c)Toyf7yt#2yz-7Zjj+2P}E8tCb^)|Q&X zS@eI^*PWQ#XFzYfGv!$Ch`Tahi;rizCdS!46Ys0NKLg%7{m%RP`;6yU$Pu&=+ai0e zK2xmyC+O^hJp<22KkcXR?!Ex`Hs6Ou-G%pYK6~^HIv?kecZLWbBH&RU(Q8{{GpD*E z^6Qh(8#6$y;p{E8jE{Fz+kA(xweZ^Cqff*JbP+n?t6xTTFSqayvUNuN&=vgLVMJbn z7TA~Z<}*eg-_0+;eqIo7-=6ncU;GjX_(hLyk=Nh{dZH^GDue)>B3152;1@Lt6DOP$y%mm+Rhq@2%jeFFu!LCg{ zW4FIXJy^jPs5SNXPQM0^KtgA4OMDXOz;Mt7@_1$u=;!lbd$}d|=CE^e>Jv+?-J*A| zp1c0$&e-ftB=|t>FWB9+bNepM-%xD&&oS}va|yY?{{sJj-IxJ>wWvRfzJb?YoBMUf zd*lwRfc5+x-m|j4&jmEc`ZMzldjB*1BQT+hpa9P#;FsWMl$y)=+)Jws=oh!-gKsua z5P|vZ!CH}6?@xjH&e*FyYw#97b6Dfw;6EQ^chdH?&y0Ui2KxCd7uZVe!@Slqr@8)2 zj);Bs#a7@&gcm;Z?&HkPxy*M9R-4-A1fSrSz&yrj|48m;UhRN0Wyd_t$){kp&TL1!&n>(J@;oPfKHkC;EA?mgQD6H`_|LOq zGf#`{9{xw@FYsyMmF>yB*u&S@oHHU1&?V@=8rVaMb#{APZmM6Aq8CM9WOF;$7yLqB zXHVEx=*+W|0_*y?Z_hP=2zJs<^_!qC$fHjxgva`KR{yh0ck2!Zd;)rNv|#QRoqR&y z!h8SZeII;B8f+Of@b+ZC*0_VJH;4UoQ19ATJw8zHtoKGA{UbR~u$kjw4r^J{`W@6A zL}e#$?ACHY!>@1e$CfE{pGnKHa!t^;$gOr_(x%2n_yX_UwqJ8j_&?+h@GI;QebXJea)e52aR9_>@$P46kE5c`!wI22b!=gfqj@WfCm1g zhv#uNdoJLcZS_@W%@*AfzyI^@jOb3$T|qk#h;gS&s57j=Pw+8f!ydt>zW)4JVl#IH z<}T0#P3SYcd%FTbePj<%A9wt)ha2qAki7>X);bYt4*Q9v@}80PJM{LS@Udp-E3a-q zF4$k;yMq1zHLvILkn0YgC1|j_lR};mdCm*Jhxc0$;k5^7Mz-H4^@(MVA1OvD}$S;V^_*^e)?s*RQq#gRIuRoVt=nLRG7=X2W#!db2k=v&|TFc$K zg8~&7+K7F=t@EWAFjROyA~3)7DL&@^9=hDrJ`%DykMQ}3-kmCY@66-;N8m>g`l`?I zaEGzqk*6RB&~2f>?~K2rz_`7W) z?hnwbzV6W7Ms)UTt^}T9JI5y+YP^QmF8tfVYE#etpW&}TKt6^R{7<0vY27vc!rdSB zHSZJX3w-AOpe;~moOP`+9pb-3cLY4|57+`W&$3{jpb>QF&hWAKCs5CL1)m~*@cyj! zUPNL>=o@6;k+H{)tnWPgrk>F&^j~2Q#H(L}XUaSiy#Cf`q4wacU$F1s&wB0vh`?Fg z+laoz`rWx9<`kWOUvWR7uX)TF(RIjI(2W0*xKQeBH}LMuTIO&k&h_e`^9)CD2LjNq z+t5E)>etuYpP(6Ai#|l|5wf%W0GwIhg#Jli_mts{v0v@ZYmeW6ufZdb@p-Wg^E}@` z_xRY`QGb{X*>kXt=P;F;$9ZqTJMa=Lv5kGrXP!^sU(^%(7MiIuz(0a7@Wwm`0o{&R z&&uL4=FzO<}UPc zhaZ74Vv3wQK1b-*2YH9@8aVR`oxLuh1GM0uKq&N%EH%ExKE=9Yv~_pr0x-TkV2<6^ zJk}`K6Ub8J7CP51W~qC0PT!LOL@dX_ynx={kFcp z>W}(bOMOAMAN|Zz(2e*`=sT!0m~)4|K;2Ek7S&<Fi|9^~ z6aI5<=m;XV0JK}*Cr9%3^=E~7yG^~@^2YDM4!#B6pI?KYfxFZ<6Tio%tdDz`=R!`{ zX29Pm{{wa&<8~mHx--8g-U0tE))4RB{PwvUcWk~R%0LUzi;OH4R(J%xSNdq75pRMIX;7~(SL&aGbfTqT)}?^?|SUudvFHA7GM{~ zT|iIKoAVmJpp<E${{X9^D<(8P?dm@AL0P>?b2<^nN!V`f!#8{06@fy)cin*xL{A zKVY*T_gG@xkvp-bJr$^ZJq7ppJb}K&{v8;QmtaJ9ik$HEoQ<)rc^c4y`9DuVe-3Yd z9rpQQx5l|Y_naQ|$f>VC|3AX(@9vlAyboWX?~y0;BQWP4+CT&P06oXY`p=-|j2+-T# z4s`J5vA+ncGd~Zr_jQIPKGyXNj?ra!??VR0{Y0Ga`|qI6;GI|j^EMzycyqF=mwlBl z7xV$l>G$XhaYxV*{{XcweL8sW(H=~N9NM5uAP49QI+1G){TneqfjxYS&HNGDM4W#5 z*k3?a_g8%10P|j;SEtV&{t|TX5B;_9Bf2%vW{=wMp%3qfde1XQC^lp4V}u5D1=w@8 zCe&I1=)Xj^zJB&IppODxdn1oehEB)<{mHi4t*_186NuP5sCwreKtX4&c2W6OGb86vs{Ss0hlOWjj}n%_r0Y6B;I^$C$a(I~#&St-tTU0L(kZ`g3`L?-sSkGkh zdlT=lzX3;m{k>_zc8y;swMYAK&kfWWjrYFj=bp_u^!2;~yuJl$y!XcW3ci={>qYHv z1^pg*#I}T2{^#WU3B=k z(^ek(5$Y^{JMAF?^Lv+Xp+7+b*unn+>bsW6XCDO`z~jE&8+%%T0mOsP9sIbh{ijVm z8*?84eFo@L@GD?HKK5-dKGxaG$C_g;kFPn#MMZ3Th7D+YDdub%*Y+#WzTSda3Wy8Fus9&)4wI*1`Kee2?s%i}(lR(AWG^ z#1Gk+1aB_ycqsKQ+j~HNjeZT}PtZ-J?xyK$ocS8!?7=%`9%uMH`~|duI_o!3b3cb( zA}4ItZ{S~`a~CbLeXgPIC15My0?f6c3H!sWE8~z8`UqAFVtCBI0wX#83?}r)_}-#( z-Ycj*hFIsjMz;2ZY+uR~uowGMAAxtz*w=ma8Q}N8y%p$$zw<1C`)h%50@uZN;*K}Hv$)_IMu)V__63wp%Y zne2T9^?dAYfY)cH@Cm*BP9TFw9c|mb=KBe|x)EK0n$LM62z{;5VmpFw(0v5!L!NMm zOU9u~`e}#mkX!JJzV@R$;dc$y&lqdiyL^B)VAveqEp+y+kGnL^nV0zJuiSwFIhIx^0=>vafyTZr5zk&Z(u*RN|_19Nr8@)n#A*ZSMRd ze8RVsdVcnKfp13l1e@`9_*qArbtAR`-re0po$DFCzRT{!z5_n`7U%@kf3>Z7{hLEV zb{=#1ZcO<3Oz;i57HZGl$tCi?K)(mS0PZVCy=xxlw2v6;eA;%z%>BUUQfs+y-;X!I zT?At6wP2rjiTrGfFVr0NXyL^f{&wPr-8k8oylPv~d5sdv`jJ-vqsy*>CH2tC0664`t! z;Bx~lrM?H&FqbgL&&Gj>-u^~t2eK4d=)Lm*%+RV$7W1Y*M z?!XneUDPvjr@zO~xKe5jXY?%Qe$*ZI>Rc^&+LM&Wde*UDYskBM=ktB?88@}hDb~KU zJDanpOQ3+0ZKd*2JwE1jA3hWO0B@Z7sm$s*#}a)5q!GPm;$y8=9@tZYdMD3;^R0pP z^s!I-H;4SNsIt98ZBXN16YITy3^G0)x=dV#cF6Yao&$*RH_&1%eG791WaIU9jtSZR zmcSaGNr6Aw*54cLXWlh_0ljezHuWRCwd`SzhnllRe@cx3SaS((f$_$ip?^55Dc0|1 zi(S6k;wSL?*b`ssch)BHT-pM2UBL2^y?m3d*Jf5w+dZ!!Uy!KVL^aGv(<9>$vZx5eq ze0(Q5V2|3(^{B@f>wIf%=**Wvz$XD`wEl!&i+%vs*`W(yuMEs}g)B1gZ}xlW8B}}$ zPe34#d7r~O=Lz)TtgJPG2>vH}XDaaa=Bxooo4)QJ&3rcApDETo+`uRJsIJtSNBFz% z2HmiM-&WbV+{-cc6*S{-PYFJNGxU*M&Y}Jen&r{`7Ms+(q14!dJgGyT@C)1OXCD--?lYA*2w4!G@+kB0ZZf?U>#?fcZOd= z*MNAi1*kE~FSm70dE>+45C`kt?8md!{{lX5s?P}TKFyU%?e!_S=6Ghy@lXSLd+-~G zk++t;=`-g?Z$4|RwUscRsXI0Q_vCfgPkQbIem}z7tFa*h9`8Yn^&XW{cV!*-vBJM# z%jo_6b&an+?mJ;O)*a8kOW^lQ^!gTf{W3H{1GK^R9yn(M%w;d;iP)FO&MBSugR6el zD4WB(&VHO<{|sJ}uf^ui2Jb>ZXN?SXMyYo@mHPYgg*@>;Z4d7O`iyQQhc)f_5^CKH zukS-W&rp7aJ;7f?L*x#?Px#Kgz-M&&T!3@xL};MKgw0+*p<6?p)jHGEYTPA#II(GO+=eXM)jexOZi3JA6*iWqA7z&{lfTdoE>xy415d>FZfpb0AK?c`m4De=#M-<&=TRa9sDEm1PZ!P(8JI7 z1e(E!KI*@%{oLz^-af1^hDEL4LbW~J)>!N4@4Z@KyMw<3vmdh06zfhJ^uAy2VF@yN z^E!w3Vg=nH`>vVS9NMoz1nQhSz>mPbCTK!$tSGVmKIcBnHK0rAjQ6e{Z)Mpkb>%SG)~If9420(>X{pDPf7 z{`Mm5Z;$+HTVvMn1N?k%u>WT~d;5ak{rP^~F6#dLt_;B5on;MjtUdn|dV72cU4erB z9P0Pocjb4`4(dz+4Dbn`7QOGOeL2?-7}r53bP*rxyg*l=53@c&zTT3n;0yh$JpDTN zLmulYzrg0-^L_$r;9c-O`8$3DA=VgscLvYOoc7V+Ykh0$yTfh|U%`LENBM(#?C#jw z8QmJ#kF#0tH~6Ug8afJe&#?UtL@=Q9`*{lN*ZC&u6ajDT8{&;Sg8wB5$kuoT{~fTl zIUWIPdT!6*JNz1W?bpcmeF5CvG4vEn5g+{AxqX?t>8p?Y01CKY)OyDU+bueKGIj?( zp??j&1<$uNul-)a{|)*9T;Oki=6a88|DSsrbc`otGvECJH#Zu42*&YLR z_P2ssyMrct0`y{w4VsbNi}`*6$LIs}E8t#Jqz?Q$@N8RucQnS?8{`2b@+`4WQ0wUL z^B3qT@chkhz8u*XaMoAoo`E$8rRH$nh;0o%!Fxvbe*%92ABdfOpz@xje)eXa7Wqfy zKLK+%k9Wx(Cj8xjJ>4OD*S*)C**}9DZ0^T>%s)qwkKnCg|LOuc{%s589G*kHBMIFS zp9ud6*o*U7-`>0*5g&c$+0d=9W%Qo+9_+9=-%DrE!ww|8THfPV}k za))fM&iNg>J@f?n7W@dz=X~a|j&~=-dJi+Sl=`1JxR184y^r{~zoIXXGu>mKO1(RG z$nLz+AH6kB;74N4zSj!%|qPTExy9I1^NQL zHRtdT=q}-p!5Of2XLFA&L3Sn~jkyEo2H5)`0ZZ!f+s5Y(nhDNYYXR&p=gb(re)r1g zk2z;-?G<*<1!xWmV4b(XzF(K|yCy?+Cas%Ke%5cD(t3so=bcl1zKbmY^X1qAC~6eM z#O8hNf)3{rZ1&KH?|}+=0bBhJtmO&k%x&HrzYhEu*t@$|{+4qE)O&Fa-`$Ko7{8sy zI;?FGzT#XD{th2w4&XE3ytFx!dHtZC^~?m`e7E@7^V2fk*9=_~-e-1%%@|3Dw9tlc zj=rr8eHZGzTABHK*Ax8g&p6I*ET6{w1-x}s_(%uXosqd7fbpEc1v+!vuf9o$ISXh8 zn)oc^qpv-9pPYMwt%aPxdp?K0#O7=c@b6(Wmpj=)_Uup4Y27ou=L_O!H@CCP&@ z2Yx`mi*70t$Jt$D{{x*h+`(6z4^U@gt`>T8i#hz9V-K5l_u;$x7x;JJ1iSVSiWHyo35TwB5w##=esf|9d~vxy^jsX$QXx zU~G5iS$!*Tj{Z65;HTc&jeh{Ow|CG9w*JhXV>6ERwbFPl5a%Ad_bY>%%UP>8roCII z{3*5qS>C*3c+VNLJJZWoH_>MZ-vitDhLrt*5Ax^|K^%89?HOvnZE&Ga5%0FO4mLma z=>6(~U-9wWbm#B|emSx=hLB9-kKnz7=G#iS?@(jif(CMO1E|6KloydVP8K4@`1L@7 zZwFeTx3?YSq=}SfAjel$^Vf2!~gt09sc|O_22#7!9N`T!{7buwSTqtr@#NxzxtQ| z@=yPIeeHit|CIdO|MrhbtYPW@*!u5DoQVIw8vgYERm0lf|2;MQ-G6Hh&VZE6Z~8kF zdd$3(-GlEu`7yjZD}Xy*1$}sJHQc%H1ZB_t1-OqcGzk%J%TMRNg)h(#q0N*#_8<99 zo6#+F##w~+W_-X_=UPYSdutDzfbBHCTc`GdI;h1OM25Ws{RVhf&X8~651_RVk*7VW zUqGJ_V+Yy?IapE0S>qQW()#T4H1_E8^9%4@v4^e!=74qk-SxDLaa!=!zk}XBN6=0iTO9-f8c(d7J1*=#8!He$L?i4SNoLg|{Z}{g==Nx*ojmCH4BbvlDpFTZ2?s zyFT(cK2q}y_2JW+=`N9vKz9~1+l+Cx@x6k7iLC|vJcI54?~VP`xt7s+Zi3!iS7|(J z{ySh!Yj+My^3dtW+BkQD&RE+}`hA%|ce=${e8cjHDH-?;(0j?H(=F}$;|$9r%L@0m9<>K$`G zZTxcLI6Gt6^8vb_p$T?<=4Lv3DS>g+|6k~htt^(XIrHR^PJKYv2j0y#=lr-YeRops zJkZZwnN$7*T^IW$bP?Ksj<8qQ%~OTA;}&uT-0KFg#uX5XcvpUq$Jd_bfHN(yZNVP` z^BK=wnfJ>wQhluN1#l+XJ%7vjfcz^sL+&E4ONrxt{4J89TS4c}obNWU#+T3n-9j1t z+(8pL3vq|Wu!gtT6MXl<_wWJz2)+fcehd0J)OrTk8p!LQ3#_$G%nN88%bxX_=Hfk< z74cbjw$pn_4Brv<<~z_F?&6v=6ZG~~LGut}8h;TV{|3&`8PD^6>|uo6KA&RWFR{a0 zPYd}Bcn=a}spsCI%b|T_@0Gm%56~TG2m3iFLVO3gW04ope;59Nv-8jqXA*dOw`b=e zzYNX58t`lfs$Lu;JIf(wJDfejwt~D3vOuZG4fud<0N(`G`W*TU6h*8haDE554a_mX z$GTT%I%Ak)7wTsW^_g?lA;f)ep!2MATrBYi)>t8rL5}YU_Is$i`7`_}Xdyf64!RYn zbIm~AV;5Z)pQV}a9@IP)x&v&tP-B=^{hT^@&-)B)fg|u9bkjIj>##oev<{sIm*{eO zDL_)t7g#~(Zj2ewxAE6^3>LsJ<@Yam557ir9@gtUdW}x|ugFJGd)SkLJK|X1A^ak6 zMj7@%j%8%?4oZCZyoFx?1#%0&9ry`&lF&?d6aSquIhd=ldBI_T_u5BMDSfODw82XsA<6W4RnCbDx_gsuSV zF{g9f2W?Q`=RMMvAuoX{MEg;R54N0mJMccQ-(v5BL*#j=bvnzVna()Y=6y&u<39TM zdLQk-1w8;AY)_G=HNzV}0G~DIV;pN=!6!p*qMw6y(K#RWYsej_@6`>^!^e8+UNZVH zhJJ@Y{VsHj?hA;KHlSnlz!lkp)b)V#q4(j zwy)6*pabY<*qTstUVwsl_V^Smr*W>HJ%`UglbF|>b@p%Y*#PH2Im0i*rp~-nE@b%qZg@1~DKporACFDM^uN7eYW2n8JlyUb7xu$zXKiwfVef0M` z!1|o6XZm0R{cq^rKm&T?&w(X$JD`J4PE31jr*U3iz`MgW(C5q<)VnaocM<*pT|oY- zL{E)}&=ETKwg5lG=L9(040^}8JI}zs22J$J+6$;~wgEZ8o}~x=fF9-n_JFU zkM;pb;4@-eVRt^a!2Ox)3An~?k8R*R8T2cH&t^Be&AD^L-?%~{Wk zz!1MVs54lBeg(Yu1%4HnFXKKNPK&O6kFn&V9wpS22|HRiw>I1lGK zkIy>jpgRR)^vB5CpoiUBThOQ2_JDK#4tfimZAI=RL_d2yLElKJ9eR&^4xN+VjB`%# zb;lbi=dcHC4e%D!JNp@%eeHu2;NSa0sPot=IYW$lWZ$(JST5tcYX^Q0c#rJo5;#B4 ztwXPXA7{A$c0d*4UiRUw-TOTcu8HA2bC>q>E0_b;_-z^Ar7dhtbbasv-7#<;7tkei z6YMYHGjh12x<9G23iuRIJZxks5obB znLIu@{AwBRyt6z<-$hQaxz8^4JD`3IxbwGAW4X5iSi?pc{k$Vt5uf+T4LteWRU0e= z>+5@_jK2l9p#%JOq2@J@^9kq@{QiJ<-bYY(XHIA09P=A^;ulc&==a(M@NQm1t;0SJ zN+@T1znH5)KZ3r($GO;JhTMTx&MKSVdd)&bUOPu7${jc*^G z&p2bQTWmShJ(){g0^Xzd{#nOQ@ZSONRtFf*&luXncNg6XI77By>ol)>@;$H(^1uOl zU4S2@@mZ?Sd3U|r@;*a(5o7Ja``%dt#yNrh6`0>XT2N;Z$kPPJ@NKB~%REULZ42nn z;msT1wONm|`V9IFwC;U*#@mDQGl%}p$C~AhKfTj|nv2-0{XU>SEja@fNBBP=4}kY3 z$Nwq(5AfD%U)E%7pOd#B$LA8AHuKCuhiNkH#kn~Ddvs%9&ux4L&>iGCY(6Ki;jPo2 z7tu}d@g7;TF?;ZJJxlo5^B8>t-ZSr!^)vPsvfuez==8CMm+(V$*1Zqk+Kf3G@L!_$ z9t?mrcxDGa_6DpW>U_^Visqply8ISyjfrEr-Yn{>#WY)E>=Urty0v!9GG>K|g?6^Dex3 ztmyzgDAK|b@F?r666 z=?rG)jP3iTbrbWs{~gX{09x?w z#5vw?#y&InoS17V=b-jmL5<(j2e?0LbC%A&iM$RU&@GfuV4V|W>u+F7;Ei2D>s?FT z*_S|ihs|ABSHk%KZ*ATIchMXs7 z*5XXn_abNb}UK{wfpw?Z#uRZ%6c0u%J+?!Z@pe$!~z$M4b%oAK7b7+JkNN{`ST0N)24kOO;h z|Hjqs-kfP(#^=?0euZuo9D+Q=_n`N|y%xwmW0z%|ow=1W=q~7?uj}s5&Z#t3%rD12Nm`LYRo zMSMOgWNU45ej9xTJ_q_|=&rG!AUmHw!%xs<&!YyEW72$llpHrgVDWP%Y!*tgY`9x(2MX3w=|@+31tZ&bwP@cn_R`bvf67 zkG+}S_sJ9VE#v}BfPM?u@(_QMTC+CanY&=Uob|EE9N_&eybZ45d+4en?%y8uzv8UV z=>~X?&jUX0%G~?tz5_$xxkc#DW%6J3?+lF5#+E}H&_#Um5NBnaz*+CoSLltu1iuWQ z;j<3!ZoMnd;m^@~##x@g>wgSyo@dC;M&4SDlR;kspHXMM4{yFDsLz%#p7|Gn`*XJ5 z*B?sWUBFp5n+x=Qj-W+|^UKi>K?lF@(aW3PxYPT1gD*PsEduY$0=5m{T{wrf(RbjD zk=?|8jUn_+iulYplOzE2)+qgm&U(h2?{fAMzQE@d`Zw^qpgq&Cz*|clTe^krvW#=J zmz=Y{&nsvFuHg4TlExf!(1dsua3=oUTLkLF zKD>TUvDw2s@-k?lTR}JN8`&J;CVtn|XY(RHTVwoI@G0=Ih9bnfwt(I`zXg8J46q%+ zFQPZT^R(wSI%9r@UVrNeAcudAPMh%)aED&Ix-)DIbfcRnhY;=BZt?LB-U0XH{LIzC zUz@sT@Sz|ce2M)6c-EP?gI#3jm%*#=a<2Zqe2tIqode)~o9+{zRrDdmcewLB#xABc zbKW@S>jCSZOXGcMA{U(d4&4DdcjP?uw{K^!e|t0jj>zDhRS((E2K+q8L+sN#xLrmc zdpC|V9pYoX-I-3kx$3^EBKn?SAK>4FZUN_Z3Dy4`enp;-H8k-%MP7pcMLj;wV+VPN zY`zY(u0y+U2UqB)x#7F|ZAM$|vxmLzJp}x*Pr#2U=a{eib#@uh{sg@-n$UjvDb^e0 zyoi4G=q~R)gWj0dwgK-xJYV3mh3+lTS2=)p$o8UNh2DH;z+SD>GYjwwQ0K8agE7sM zgm}*y*jw1V2VLj}@+)W?xrg5XJVSQ@+@<&5`79x?f;PG_va{7cffi6}IR-6cV;EPx zJ@3Q&3{=!$Ea&z+I(KA`oiyHs0rY#|?;rOtSH^SJ`4nt{0PcYG*ZJk$hjZ~gU_XnX z2@>+3mhs)H%~|JA<1eC@?x8cjcd9+pWzfz|e4YaS0p4Ex4N~6`X@)Mro&le0e?K^* zX^wy%fBR^le?Z@bx9>lLG4c?7z|S3dcdf}8I`<*ec={yp0l#W?`CK;5o34TVPGE%XnY6pi96#wSniIm-`w)&1+qcdvD>>2gX|C>^2Am zJ_6qzy9RZ)&f*o+cjyMN<|cR#>fBptjMc-}zE_d$X%C+;^Q({(Y{sk4 zNhipD@ARP=KAoHB=bXY!pQEb+zg&xz{}?P(rV=-jL4&9BW~M1t=RWsI?ak9Aqk zOJLnkK@Z&$vN4SL2AwlKhBvSJdvF1N0eucyAmdzJdtLWq>a#cP!Fu33oVn%9U(2|= zYk1#N#(IIyxvhf{a5mPGQNQn$Idm;zn9ms`^3WaRx4?bKi=vFbWe?#W@N45|4)2Q3 z?@z!wwLLHK57?ovL0$hDynn-;gAwQVpbyYMjIZFG*=ZX8HV&L`!!M%$4m^OmZ+Fqe zHwWfF!S6R0j@1T!7A=VkT zI*&O-eQrwe&5QV4U0^!~-ly8PgWN?9ft+Q$BhJP6(q(k^>zwTG8eMxc?y?PE$LZpe zA$uR}v5g#Ney`vQaEm+w1JIh;@|$?yjH^wbK1f2GcO5@N7l^IjJ$OsJH86tjV!H_L&d3sl@c{z>mq$YA7djJLcCi)y|T?g<1m}4HQUmNIazUe+n zWblCO$Jq}yV?WkYg_tXWpPTuI2#_~RyFCcraD&WDCA{X#ueTXxHI->+?eMNv`UnpZg?!&w3^ISna z^Bg#E8sN?;(4=rwpi)|fV-wL_Tdyz)Dh3*PuAScE$ zez))w(A^{-xB@No^;!2&^E<>b#t0Z+9OLVLx3H~%WD`L5z!_-b zZN~n4#0tpv`;VOOBe$V-%yanh%y$9)33hw7*9qs`T|N61?A`%w&TJXJgWZ0Pk-XF|x)bmW41saY_d767J^#1p<*(7%uk}^LT7?GCLU)kHI_kQ2&|l2p%m&y`fi);M z;47$nhR*ob+JPF^`*8xzvG?Kiy$3!=Z_rDvWdO`m=QTzsVok=p0$+kf;7oe>xAEzL z*@mu&Bx5>FG#M$XDc7gT# z+v5i?g8vBhy!95q`WNt%=D;`)H~19sIm+SP#|ZrxYG3wgowvwMu)=xktLyrI(V4B! zxjyF_&_#4x&?D#_F`UC5u*Z93^B=+g2#mK4t^4wBIrA{{+d}ppS))Ef@E*9QHduqV z<}AcJX}n*N&2JCpy2nqwy*kev-<>r2SLmLh({7xA?G$>IeyXX9?DMX^uJaJPXKFbA zjI(X%Dz+S$SNThL@5UIuKxZ8tFaiUxUdH?Q0l)9zmw~j2O<#Sylji;j^pFEIfjWmd z_?+As*oGgg2hOJl{Mcs(w7KU17RnfFNy@nwz}mGhL%loJ=3eu_S!jZ84`>rf8S~_v zv7Qk=*1Uqg30_K-Q|=zNy51J@7HH0B3)>pJ_JUaR@E4!~>KT7t#`mOkFM@!t??3O0 zeV*d`92nbt(oPxQgT)O!5cdFG0<~_>WWaZ3z1J|aXV{+tYq6h0`0qdf=C;o&5C?k) zK9R?74Vy8uG{*P5&wzEAuMK|<{|gub^PYpcM`OFACcYD3%q{HO=$&T={s`E&vL9mx z&{F13zJopk1?PL|)Ma{S@*?iccR(9k6WaiO2VdjYHAv@>Lx}GS2JYzqWZ()v@5?dp9u=Iepgqp+Vz+Kl)x9a>zd65vZh`xo2%pbO{J#ZP=!{_v@}4iq z@pwMp;OFj^fV&vu<9w~vKE4Ah!2N4ChWVQ}abN2x@qqQ&&j=*=CV}4I{q17D_fYpW z+>G~ph}=ick(*#UjnC{`=o?^7_TtBU_rO_Apvg^~iF2~gg)-LYoTeXW51(`NOW*_Y zEpVoD$amPjhT4;TE(7z0fDUS(Ep+?f2`KR2h2BGx&G=5r@tseR3+lqw1^#|BUPxn( z3?IK|UjpZ1KaFzMFI1hoxdvP6!5J`40@V2&=_4v&uag;X&Ws%1Is3J*QXRAnR?_(2 z;jlO7dU2UEWy|ub4ZT91w-Ejr%TYejAoccZu;LX1S+-Y$` zU1{_&{wn$eo%PwXb3VZS2XMa$=%7DCUIuxH-)}wiBjhIV&Sb>8gntga1AVCXyIs!S z1Ip+FXu~_Nb8H#$ZsC7J_n;lQgT8_M7HXfy?STZ_MoIr*4%yl(&Mv_B;O%b_K7(#U z%`?U>9isE&-SM9H(1#H3_Y3SfcJUh963}LhOddGfP{ccFjyC6Cpnnd=zjUSTq4Q20r}5tdhEVT}^SuVWGVW^~YMgVZz1fTNS^@6a9(L1M`}BSg)Bf7%?RN>7 z?`cVm)Ow2C09(ijdO!AWjwN*B0Q({M68?8!of-BH(AT*<$G!}#A%J!Kno#Sipoh>` zpni7h7};k0?q5x*3*Px=#Ppu~zOhF8`DJ$2on#^Q_anMBWM|>od3f(q3)%Q@XLN5y z)g8ebe=enNczxf&8&~KfuF(I4{0t0;-9}!A|10zY>a(cq9AASz*amkX5BNZh;~erd z#=e9m#9Eoj&h7yFHSz+q&TY(MCVvHO0%L2xhJKjwSHx0Rgt$}p(FV?Jx+nas(|r0p zcn-Lidte`To3p-&F+k@XNubkDh`aPRY(1aP;7`B{{fH@?!&w>gjG-Igffzo^JL#vh z&RxGA7y<29`0YWB-`R}M&;k5(e(3Gz0sS$&v6q24yJ`G)CTq^%o%7@Uy+&8pY&|}Q z_I(HTu({Jq^v3F;8_sxTzki&qdS&6gZsF5MUxcWucjoS=b)esc*Z&s(MRrq2P-Ko9#8x-YQ%Y!7eZ`^UMpk?rvnXo4|vMGe;C$6Ahn zcVLYD8N4+(_a+#j_t|`k?HIWM)$Y8c*5qASMK+H$tN$9lo~`r>w80qP`%U@^akhtL zd}sWC?K7|otl$2eZx`5Gg@2B|K)waWFs^e6=o9Q6bl&TF|9d5Im`{Q25I<-43-Fxr z2k>j~)_YHkXUKsVUFep;_Bqfl?C}M%=e=*<6XUw~^O;^-5@vIPwy>$QZmGBie(#Ua z8Mj%+9_5$O4?qvxqVLaqR*nS?>H>Y}i%q71HzatNAfn%WFojFH&@32%%&=(={_FX^s z*7q0W@1Oy@vh&}Do|p0Y_qnX+YyDfmnEFcXL4S9rY~S@xt#OQ9y|rdVy#LnU!cTo0 zTKC??zm07lT5+zJ`Iu`1NUIQ^CC_HaBYYN1>PGLZcG0~;c4wFHNAPv73ZH;(jLrA) zZ{TSe{|)h5_#XPM=g@rt0{-TG18*+l?83Kz=j=z`y!Nr2#=o6>u3oDL&N~s*TS{+; z%efqXKi;Ffh{0eHE5&q8GyV1Of@4Hv<0oxE;2E77417Cw?8tZmhHqf=z4IG50X{QF&{mX={t{aPoO2KP0veF*y$|fYj?+NCEaUE$;jOs^ z)o%fKZUf{XD`SnueGS&oCB*7N3($hMNBgj5XVFD>3~z1oGkFW@4Dwknd-wVE8K?re zIpgn=qmumS@4x{*6?$vGgkJ~VY4@{O;s<>LRcG9H&|T!ufOTYGjE}iGQ0q)KW6lY( zeVfmH1bA~@tEUcQ>+3x>?*Q1-&tMLC&yx`S2gp7r_S3$Jb2w86Z*FU`rZec*po_kN z+y__iE$9O9&P_ld?ten|eR~ah;2ZFOZLefDz?!QfQu!7(<2$F@8|E?dQEzPnbOmSu zdwq@HJU+`{8@~3x!q+o7etC#FoLz>_niJ50H|I9I^;<`PuBP$+yg{#jg5JJY(7PjN z<6ej0HG2KcwG4Hh0aSr{k-ZnjJchRxpE(5vxG=7J>^RMtpp!PO`n$J151b!2!wVSJp?>ca|qKmB!bvDlBIs6XTMYeWl z+J+xM7oqwrVc!Ds*0`@87`Kaj4eUjK`>k`$r%~^jCi+uk=W6fU@YdoPXW(br1Nso6 z?IXOgoxd^tK5HYJa}8Zqa0pn3^Q!Z%;Acg8ZvUYJ7eE(@|=JxY|g`a z+^KbXzK?u|J?D&PPtduSQ+(eOg7`Ey<@kw`B2|ou`&~HOK=nGZkC+MD}ac0xqQKPvU=$uIppIdy!-~j##o%3)$ z3;4VOZEV)EgwC^dU-Q@kz5zOiO?e&IlYJy2#GM?L&?4@Bfc*%$gZ&+J+B-VuvZ@R{ zp6T^<)}x!)Bi89}p#f*d=sfFhC39_frii^av03Mr!1wMB)cd)DkNYw1GJFx@9q7Xw zX9aXQYrjL#g6{#(_^cmcpTM`#Ey7!`{WqW^Xa`#XjQt+joCP`Vq0Z_OowNU`jCnKa z^{%<293<#7WOr}~Z-4vhq3*+H!jEwe;kQc!Y!m#acQ(Y@)QKH@_JFfjeuBR_^AMlm zF>$WZEr7bOgA`w6YcRh3FQc30CPu~C=jeX}73cjvIfnP+-F4nWsOR?3A7P*FhO^FQ z9$9*T&6x=KJ9Ne?kmrz#f`IVV%+DUJ$yz+qobgN8o9OJ%IlIF%V2z8Q06q)$nt?-j zXJ8!P8P@-R{wwwD*m&FW=&aK`J{Q*I9N%Db&(13evwJf0Z=!RTr^vVPAK`j>S2-$ghp9Ats&_>S4>5LDNwbXlr137gGwW@EYxLG3oPPk` zrykUv7oh{t#c#Sd{08vzCGo&h;EeUN*EXpAGIY0K`kujW4r~DTFo2rJUOUjA(1k!9 zH}T(F%q`mJ&9wr34Q&JU4P@ilS0BCyJwO{wcc8+aVPCjB_1$?}%{|c;z+x z-N!Y2f{*)K1J2O-wbg~#i!rTjp^SdkWM1`e(V442w)S1*L>s*I7%MBHEf3K*pxYpZ zH3e**cVN8M*8+fG3_sw*`t2-vh=UaAp|UmTnO>sa6KEa-(kN8tuoHg{Lc9t9D)%(ZFD{0UH%O?L;G_FhcmwJd5{td zUui-v=b0gYQO5rbMeX;uRC^b{07gLj61sQ5+{@69DZkx;J9Zw~4Yi?v1MJzn-aF4) zr*wWse}bN3+kg)7^Y?~3D4~nM9Aoej``^~KgU)>qKow?pnAk-TcitsVFO7R}c0U8_ zalZC4tqGa{zZbkG+fe)X3HlD?_~(Ih*w&Ga8Ibj{&I0NzE`aqJ$C>oeiwW?q`0Ts0 z7IvSl&yYLl6L{l%i>%G}a8ksY50EQ-oYgi6K;Qa|HMD{8F0o~tvxY5XVP0qYJG^~o zftt$b(?f55&yJzy^ttygI>*OZRXMoQ6Jw;#VtM`(lH`bTf{O;?*SLF0g zA0VHgFUaG}0+>f$MjxR1v~S{F9wM)T70{f;xQ8wwSH!!5x31}|kUQmUm&EQ7;~x18 z`#SUuxC0HaoyK?59#p;$M)*5}5tswslOI8k9P?$oUp}vWd2}zZCGZKlZ>7vzR5RZD z-Ncrpv*(vHc@cY_z}uI*c>`X70&L9SchNprETi51S=(QMyYTMWUm)IFeD=U|aF6Z) zpBx|WoV;)b%VnH-9^(G$=ct3tb8YNn&LsNb(?W0UE#y<^5^&CQ$nMiRuEBHkWAum6 zYiJ+z@Nuu6S3XCs=aUriK2G1Q*saw*jG2cxd+YTM73e!qXYTCn#rig&`qw=qoGXym zvF(5^C_$dadmP}e;P-)d`&+OL{uA(iZnXhT8G8$U5e(p$px(RdGQM~3;8)RA z4Ic1WgBouM+5pz@mooluRZo!ZtHSRRo%LM7+t(a?2ioTB04yN;?0E;4-~;E3 z)xy?9=PW(zj&tm7bs!7z`$oV2vSZH4YwOH>?c)QG?xCB5dT)$*3w2JOanI5X?BX1J z4xX08!`6pte-7=Tch=TnUDlAndw<5*?~tu~f^Hq^GvwaX2XxzI{C#l;T?aY!>hl@s zg42>a1Dwqlpv$>t%KV|;edQ(O7B*}89{Smg&(OQ8pOB3yufKlQ+ePpGdZ5DQy*NQP zg8KZwgV)!y-+-@??ai7p>huo&g1#tX&U5rzz%%yZJom6K17mc-7I*<(0&8$~JMd>^ z{LSXsDIijx<0i;~F_IAP-WTw%fI4%`m$Q3MOwY}slU||U00rmT=rZ74clH&!1pHA( z>OJwyQ)nBzIh}*|egxe^wjTQ^uqiP?hgtK$wA=oixHcZ;uiTR^{k=pWz% zSScw1x(yoe8MzA1T3?3m1^RjP?qmgi3%-Xw!{-Xw{zXrJsOPrJ`2BW?+yY&okI$3$ z#=D+yz5)sU?x2EZ;y=ck)dkLY4|hR^ul@M0N#J*&J~xX{dvab2@I$EmJ7a5_#)WSI zdpN{L8jzpN_zUPuP!ZR>?kk5^o*;WiZ=vp{MuD7*BL3&F_GQc^&aS5DH>2;j$Sw4q z%i*oln%$+fSbGay2dYhK&N|OWeB6n%aF!LkHD+KL`4uR%hxl%_h6K5*uYU0BWqjw? zF{RVHh`ozVU(Y$aYh>-Op>M!DZ0fCP0Chjp`N22AHOR?v2Tg8boGPG$w;%H_5~Bxi z9Pg)nySs;*c>gBwW3Y(6fv+>Oh7opq-9q;gIGbx=U*1daRChD_*1Z+j^lz52h7#(% z`3dTC(8pf;N}K4`us=ZOpl$rz$3PzH9_>N>5aa6ehhM|)4CEWoE_8s-8l0IqPT=RztwE=G&^g^a{`=)wO)hWGRC%sePnCS zW;Hm^9qgh3j8TCAZyinSx4@WZ`lNB^?#NlTfWFqY4!;aCY}%WgaZe#2le@+O{zu>$ zxaOQaRrtR^?twn%^nJiqpQ&jTVxBM4Xt&1q@BzKO+#+Yd-Yc*GPQV?y9(Lzd0q55! zi4QH1N1O3(cj1k>i+&IL7I+CifPM$|f%n87GjcA&d*)dh@11$YBKj7{@Hs@jBIXd} zoRg{#z?r=Q2{Ah8oqdHq2cBcDYXN?b34V){AW<-d5rx5`(&oqwhrGy zZ)|^87*}Yki0ioq`W4V3ruQR9cc)COjr5bhe#Xy%bLnEMbDd+GLk`eI=m_cz%;jA( z-zxl<@cv%6cjceqpFp*@iMs@Uh_0^FnAVhF(?{Rw?20I7A<}@}ym=YrKK@PkHQ@c< z0OK^qJ(9=XLjD8(C9u~wP;>O5o}YluX6&mfKF#VNjd6_c9M+H%?AORSy!Y!6YLE6b zMDN@JzUHw02Gm^6N$Nf?(aTTkfUP*=tcU1-!tV!g2|Az%v4%T%&xm>T#4*-K_=Pn7 zJ#8;b$i9E=N#E%nvAZwla|Ar^Js&`ip%>tmSk~Kxdd^v$fHrc6SYN?U^M_B*`&33P zea_kXz9}1L$hlw9yMOb}LH`kWKfBl}eD>it(Am59!8>^fzYOoJjd=p3i@?~@^BG@- z_^hh`9$i3|KPcn9--4R|H87uhY=PwF)AM+S&n|e0&pH^w5AezGsgNJ{wS~Noy(pr- zh0irON4JZu1AK2}*zJ7;?>$?BH;(m~;{a^H`%XK7x8^3izmeu=vhSb-+1l$GR+0CS zSD@DC&hFtG(6_*So`EA^%o_SGl<^(b!e%_tMOOl667b1F)UTuSYzCFzNZAdx34R^m zS?`7S#$NTUpj~92i#})8@KKkAI4}L(?G<>yZwuN+SJ&7niHF>PIydiV0^bIAK));K zo0J`4KZkblzn@vthyd%bFr1TDqIl;$g@G0;a_y!c{eYTwAQW|Ava0I^tTHv&d?*;Sk ztHbvg-Bb7q-d$MF9J&B)f}9$x#k%LPKL;0J9$gL{0p}B-)^-V9Lbi{Bv%AQ?zmI_R z^t~V_#NPEiZu3NPdP~N8Q^0t z0UAQg?R@R$GhmPAY?QHwE`Ii>&6xJ5zcZDJ2l%-%?yRp2?7f9<1)PBdjPN^3C|80zXT4z2jJ%=bVDBB2K2Iw z@vOBX?h14O^?Z%D_{hJ4?qPSn3()Nue}~+~rq3<>HvDvEH`I>2kDory=Qr>x=piSd zgtz~EM(do?IdtanUfahIxGVV%D8g(O*td`$pZ!pNl09c%`{>ML-QsiLPQCL!uc3(V zs2AujK@*&pu?K5igZDnDdz>pnw~H=-EX41W^+!G(V$Y|u8nI7%=gg3p0sVCE_^fiK z0qpS_n;-c^Fc0JJ;=cjf*zZ9G@4b+pgC2ro@DdpRs*HEa8k@*%Wb@}YahL9=?qvnr zOHj-GH2!Bw&S?XEn={U757~JevxDB6<)wZ;NK$g)*M+AXmgXRc5wp z&>p%m2TB|dx#VirZr?QFe=EvWZ1gMWfw6YPREav%Q?;yp5Vg6wlDU+;Ge z?~Gd59^l__ZUsCo<4oGnF-Y)nR_5FR)48F`IXA?f!K*Jq?0*&B9L}I4E#vR>TjUJ+ z5PE{H-u2J$-k~P24u6k$_7=Ve`tU6 z_W+E4jch;XP~$tl0k$N>9S_P#pJF$s@Am2bpv&0Ic?s+^KumkF zfAih}?Z$Z=KVZ*+-w)R4-K+3%KNF~ZSl89frykqL#{C{2=k_)HbPni!hkXuC(HSEY zpKSKKi~bf^gR@-*@)>dTwVn$89Q_tpo1uZ;J78_|@E?FRFU+X?1iN);x90(H2Y)H! zjFmgs%wtbe8QVR45n_KM{G0eFH?Y~C{_0C)YxV9erLq13TMJtT?dyxayBXt`@Yb;f z)t12Pn?tQv*lQNZ4S!I^_s9kOA~*!rc!$pswrhCjk>MZUyYP)P?jeJ}#V^N}z?;`v zGO&Q$#%Ap=;qBM=<#g@^y&^l89zNdhDqsUSvS+*4tnEQNzAbpqIuq$@WdDZHcNOU8 zT&y9P`3%vk^I5w>*Tv>}`_N|zyESSvW)0_Hz4qS6$2$D1NsBnoK0ft4_4Te70XuQ* z$^N{P&w(-O*wT6A<&-%C&n8fxIjMHvOD$k+p@_SiQh4uBC}SUIQ2jIf+?TxuXcM0t zz61X_uetPTqW^*T-s2JcH88$&wny`~;UAx^`7P}7&!C6kbS78C=mK}L4hr~V=nD1# z-2=V{^zoV817qX{)ZH1=_fb8EBE+-$=GZf2_c?~2fQ>Z%mR^J&pud35L0i~fL-i?@ z(Yp)dPw##K54!63cz1rsN4xX+4!(_U0DlC34xbfqme$)t*7s}ZA~;EtC z0K7mq0N&d-zajfMSy*=4~6FI;;^DM;w z?xg-J*e*Z=-e<;m_N>3}+imp6UhK-qgI9bqqkyU92o(78+NpGS8D zybs2ilOn4(-xcy-!96g351;xT{D{v6@JvC>ir5KqfImRj!8e27LgyXTuL<^m_8nla zK9ghY2W5=Whqn)7S?^cy)}BCLf(AH5_6|8;{eMAT1uak^zk<%s@EP=7N?qUt48d|4 zXF8b0xP>mDE1(~s`(Q66Co^~h?+!ZHzJd2Q*_C^-Kc^1$=ZYbp7LQha>}d4K)9MScmKnfaZ~7B+p2 z?Ht^h^}hwHz*sptX~5sS>&Q!B9s4b3zj$Qtq8p(rp!V#YxdZNC8^257{M*nWC_?nt zrwOKYlWUa5oM-xN#(It02KM_7{XX;vEW=j?_eUN02=>zWcixZCd1U))5yw8P+up4& zkl+3e!35cw&eF(t;9mpx>N(+aZ0+qd-qj7L{@UE-F*fJx{dCszP1J1;-i_*z#MV$Vexea^9e z3GBazPCTB~8u|vhW#GHV=XeL+dXmjplY0{*umY^f9=s0$`W&jxKAON}T_1XJr8RsSS{njOxb9oB?8R&z${@==Y@7(bU{E0R`)>3dL3A5Qh zx*PRbX;s9#=AJ+2tntm$DdYc7m?6~pTw#AO-lI6~ri=X$Y~a%XP5g4Gv-$zZJ2QO) zHv1G?=#GFNcV_+8@(Nu&>k6IENDuqvOn(j4&zbtXXV~Sv=kEA-&;!OW_cpl4c86aJ zOzR7Azf0Jh@mE0qU%+!3@kzm1u!GLsw9{BieNI|nUxe-ebNKN)!k)~x3w=NzLd^XG za=pK^G(J{Gd<*IxMF+nN z@B+IzPJn04>8=JK1NzDjH)9U(*cEaURMhI84nP-Ky*ht)dA1EYV41j!K-nE-A>QLY zvgZ>tS93-@kLVV-sgEjS46V!OrPa|y^o>^opv0p_S`J+@>s z)|?m22;Mju)camxZ{EarK@PR{t2DloFQC`>cps!0zDs2?%{f-4#{yje@BUYTIjq6@ zUe0vROWy=rz(3$GpFkh5*^_6E(zv5{Q1!+Te&(^aHi;kNuFTZ~Uw}Dqtshu`_g&zw z1DFTy)LcDa@A}=tH?XPeLG?ceK5Oz~gr9k{K;Os}zSihIDs&ysf*g#1JKe#)0-lud zZ}$^u_a^SoIXjDlv(`C*FJ?MpTjN|B=jj=Jn(E=(&;d~IOq{dv>@C2%b9?I|7vKuC zfc+OCzB4P&5Ys%)%lQT9+Dz9&KfSl?YC2E+JJ`*253K|~yD2dOJbLegGrGgpRu6jc z3HGFjGupvkAUhX#;yGpeGj@lVq6<7Le~QiXP5g~{h&%w+lYxPhI;YPD{@zdPH(m={ zA3Ol}e8xF@p4Nq~n87-nrTW2U+<8eZbJ)una1Q3ku_y5E*qJ)twKCp&|7P&MJORcs zW(&JB%LC_uwOKBEjndq)y3}$dyd>k zzQku4yaW}xRvEvqHlP)L_S%OIvF`z6P3yzg`;uN*Yb%Aahd#)Y5 zyRhC7au=U_kbyC%zXzn&IRM7Ah7opa(r;SYk;~u@U4dDDI z=n_!ie~R1!BV=c7FHe!hI>?ZR_*s8|w|DpN^O7Tb?r{zFZM|LaJwE!`bA_+{n!5pg zfbNxXAKG_72FAb``_Kd6@31T605wJ*z6;+d<2;w4&Z-LWJ!nlC{5HBCzBzn$^Iy+- zsr>Xz`h57@6d*xwUDoIM$9;LPtYZNE2HOd6mcjTXzD1m;voemqArs)8)i(Wv_#5pW zU5=l1OlJOlc+UpTjG_A2mvfi{6}oTB$m{1_Xn_-S*7*z6cZs%c8t1$U-2%>iguVmq zL&uwOKh8(km-)=$`{Y4Nt>%9VmVx&n6yyR;>^XMxAD6TD12xq1dWp`ya$vq5xCZts zcGH+YBi0@Igjh@H%y(YKI?VAIe1VTM(eDC!yBYmnz+nn ztho;gc)!25!2>>jEr|{EF~0M6PM7GGL6XM!#=AhaPv_Rgul^mdhtB!fe*(V_tjF0I za{>Pe@*>n7x~mMEIVhst_%?p_>_>Up8?=Ysnzp2mb{#_ew;}VcVjJAV zJK8AYJJb3->u!b5h4dZR0X@!nC!Cq@m~WvS;Cbg^uD9^g9C;bu8trFJp7YMYGuC~A z-u~vG6?AkHd-MDWKHj4?wC=&@-5nhP^C!e|ueDto@c9nAbvvg4vd@Wr&hZ-AdtqK{ z-ovkex7GphjDF_MLwx6U;GJ#2*S%>Q0sCnp2mGDMA$Skm?@}6b*?SM!J}S=aBEJWo zX+qt%v&xC(Y`4Gy@ZPOq+W=kU$2~gB7wGHvVGF*;d7l;g{sS1-v-Z9KHUBBdLY%k0 z)7cSPzZ|`@az6I2oWO^h7}K~FI;rn@=TrcBb2#@u%J_GwwI0BC(G|oRL9N+$!~pnA zs5^o_?#tYZ*f&7VIpcQF)jnUMpMGxQd(7HI6a6$+8h=y$h_}wmvYYH5lJM>Kdfp-M5@`7I*MJLEVLU+?~JQ+^sqMhmv*50!1-Pq51nxS&(S{x-pvHL1)W14!25hS2Ynk*>uCVv z+=C><`@fE^S;m>0D=FgNUjh9ZPgpIvmF znf@I8Przrmh0a~;e=?&_p!VWS?5DnKQg_pV25tB)!Fz{_K>g4RETPZA1t^fupwro< z@%!K03*i1H{w}h|70^dkZ=7)%pFz)=$J~cNdH}5b1o(XRz_5(}d6hQxLv-Gm1@xE5 zM?jnV*oL}i^WJU7xxa&716|~HNj>17Ku(Nrp-12jUVA~WU3i}h+R0lo&-C+yoA zmx1|)z~{jG`3>lUj5uv*3%z%$K5H%Jdc0n4>zD(3!1>(*`TA^4Pt&-kiX7gN zHgLZG8DI1H3_If%@XQLh1AD-_iz2=sGWaFnXT&+5xy3U6cZ|P)dN199F*{&@U7s#z z_OaVnjx1e8uIn42b6z9hnT0g&u?WQEE{bQ^{sOG~3iv#G-Wdw-sC62Dz2pqo0{8fO zz6*Z_tS6wGKnH2u;VbAqa5gRN#5u(`3-R99b?)GkU|)t7Q2WY31%C~^Urk`nYv=_S z!Ut@=U;0paXXMPC(FyblU7g>1k&}N6{VI+3w*}S5y;Rr+*ls}vHqd#0)mzstFh?8P z9Z298(QTCR*>q0&S#MuE=k&jVzeJzGH$i8nYocqzyR#ltySzR*@Z2J?ryuxwUs7vm z;gbaF1Lho}lOI95oRsfzO`<{f&?--in=Na$+GRTPOnQi#T zHC(6h-;BJ+TVTwY0$+1EbMrm+f6VK1j87Zf!dt6%=dg_T#%F#X{ufY!0rocN;**3p zqZYh1&!^Fz;Af0B@c!vv6|sJw=Uwo)2IpJgQ$U?x!kGepeXj8R9z38Q0{gM%6XG`7*fOa1p@8?kTk`;fGUjeWr}ydqA>-^pBRThcU4a4x3OZ1r zKtTlx6e#FGfdU1u3Iz%jbf7?if({fYP|$$_1quowL}P?$%wU>f7>Z#S#4wm<7zSYm zZ>bmtVT@^vF~Tf|!W<4_IUE{uxrobhxzE#Y*Rk%sjeq<9{(Rr6ZrQs_Y*NPO_i3Li z#M}XYpW3fJ_SqrFBC+1#r~dY5pF8UD&+*%YFXY*)y*B@GJNw#$_JH@Ki;o|D&Eu@*vgf9*bU+Tz%Jq|*i#p~bBi9Z*Prk+rh{CQL!Eh? zF96Rxxs7+M#NO1+%Q#1Z-F$xfRlLW>I8#bo<1gLEE{w5fpJV6m0p&JuFZSenYmB}q zWBxXL4euwEF}8)j=d}+ufMS99AkI7;d_03=;C)#K`t`vGTjT%#(K7Zoz%L=jkG<8P46|C;JfC+U zN1v2&=bopt7`F)~_-#YyfM*b}*^}}CaMmF{&i1Z~-vRFb8o3Cu4{bHPx&8a%CGa`0 zA7|3{s)~P~w4n)l^ZRiR&TEh673=tf@E2~5P{|1Y&}3$%&Z1}}g)&pOAab33~e@b(k%nLtPA$KYvgci}vqm-!NW#VGr0F3+_DZR`_h zfWF4wgL>9obdR7_MZ0|;&Eoqr>OF4ZQA_1I=TQnV{wB$CDyak@1R1D-E&OJ z*}kF2=-t6QdV5Vmyoc`NX?~>Xy>(ZfZw_{ey+a8^Ze$2Cg&nmW6=rYiM7kmxeUke|1 zcLs0HF4P(u==Ns3y*$7VfX|&f=n!Xb1+c#7{h!$6AF=Jg_n>8ne}|u7G?au`@2S70QRz8#XA^~&8hzwv_T2fy~b7(+r0(w0lp;e0b326EdfW^ zcF~_fozGnR@Xq!LOlOAv2wK?mm3J=h-aBXraktV}$X)b)I?xn3Rfhfq`XC1vV4;dL zxj*|_LDvFrX8Kj==r+D{O7uPW6u5&HHfOj(|2>f3k%suYxC5`Bxjhr9wl(yN)nDiT z9NymMrTDB#w_=WS^!{dSBcGyo{!L{2@_d{8#)%_rbHIAeW3SGh;~z?Pgq#R`zo5(E z521I^Y2BI48j~&T#AxfmZ{z2ku=i=tvzq2QK);OM9s|BFpzqLa0Q2@i55B~<25tKN z|0wl_`1)>h2LWH7$z5n3qR!rSfcy1teRp?_zx7w)@6{1AfZF#7^bIi1d>?>+ADdS{ z&O0KP=b_w&&OvK>vYvC;&k-@xUBC~}wSo8G#2EMm?7m~1YaPA=Er2r|XuFNG zb!T#09rhofKY_H0b0pAXbT2g;>r;3Vt_Bu z-vE18womJLj|Wx!ermxV!`rJnaz1(Iv+out$+-d@SMd&1@MpliOU>{5)ZfX{P8I9c z#5H&B0G;&%x*YT}@|N)~c{lt`W{uS<-r+@fdw#f$@1$MyZIGbTPrf!6u^Y(hC-_?L z8e35k540D^pMW{F@4>V(d=TzhK2eb^aCwJ#>h$Uc-L(oH)lo2n~19M8jm<^fU_m&UV$C33!LW?^s{&t&OAVFfdR4C z=zj&%`+;pq9=-b)(>|$fE@vp=?e7r2hProq&{rSp%X=T?=YV|%boS>tyGMI;FUF6y z;yJdlb4-)hV zI?vgCq*a{v0b2@;Z9`A-3)o70?Zv*l_Y-U#d_orU7uW;v`D#~jraNekya!%^6u6gd zXa{Upyi-7nJTriK#hh777j=>}7pm)dK zUu$>%w0rWRlQt>+jm`fDs3VciGDthnX2>yQN zBlR5ZN&OPC^SH|b)WinR?4^fKlUv%u{{{FSge?C5!`Ou$fIwVYanIm;Wkb03)u29?nB<0+|{vqo?-J`?^ zt<~gvf&LCOva#-Di0&KU&YkyZtal^Hs0Y=)0hWP1c|M-mGkAM4zjwtx5_%fJyKnEq zziaZ`mDutU7wFSk%Ema$mn{BXdIfD_Z>qoeTAsY2jL&}#KLiQ#8aDeIfNOMXS?nnf z@q0K$p60@S0h;^U!uJ^492jnq102D7r#&xYd>&q)y8=FQ2he?R0ooOzA&ysU2@ zp0jv|EOnpO=z;>|=r*yH)JVX;kQ3wU1);eXQ=T7v<0kDK~r@5(41Zh;G8x7`t;z>pvx6G*!MB?Ie3lk4RjqE zpm)$;KuNrPjEJ-UKdSh=ICPYfOqhx=*;mia0E)QlyPss1fNA{4xA$(+nf7- zh5Q-xkkxNi@orc21dI{8L>NogLQ19b_d?VstBCi4OmUZ2cHv2dM%fRQy zKJ4kTiubOI?tA$E1?=D0HgpBqc|QXE-=lZ#T!eT(%pLIkW#%KFqqCPBpKsw)FaqXy zUPa3PoXXDIC7-qfZ0@sw2K1i2bLAoKN1eVa*zHN)9TZUG8+boFE9dM(mxvp~d#?Hv z$Q}3|^bTsgXR6KFd-!_Zp5xc(y;JI1$U#4R>JY!njCChp(0_yeEilhB=molMpj?;n zeX@uAd8RX`HhcB#^FUwlIlk8Df;sdFdgE3=E5nY@d-wz_NXfB?-g(-BypQ;JXO4mS zyeHQ4oo37gs?Ph}2EJR2wHIp_A-=yS$ou#&gFVm%`gQPk&ja`|J_Y_~*sjr=s|T-c zgw6eV$2Xz&bd|;L)+6XHP`3@7FCo^RHqieBu7P*9L}wlEm!CGgGbbU|E6e!r`LC-u zqdj{^?8!RrqsDg;xeGss&i>804zG{*Wejf6_3=9f?kp$Hy4KvB`5ChS_JF;&v3~`X zzk_$r)+(}?zeVf{yg97Ve2>`2s($D`!h8SrK?mI$P#>s2APl)&4JF$VT1ncskM_dB6zPoo11^$oF0Ih*{Y&VNJTj=e> zT^oOn-Wlv^bc-H;KgK)1d+xwH@AShn>vQ`p{`Tqq-CrBKwa3UeGyOfhJ?A0TYU8&6 zO7saZegyPWKESpEx*!m9QAJx1dW6m$dH0{g+kYSZ1986V9zaga0O~oG%FqGW%Obso zKZW<4oM8!Clka{DfB0WOL5_fYrwux+VqJaI=b!}p=&h~p6~5k0?}pzWDe?udhWB+9 zoqR`s>gAwM+?Scp8FYfJh2P=K*L@h1;9CIaU(=5`-_@SCasWSo58wuPx5w50o#zws z4OjzRkc9Z2Grv1Z$mKm8gAMHN$$cDybL1{^57HU-|^-5W0uXe(a%-{~>e$ z?02(@&qRuTANdk_0KW)-0Pmdk;~g@mv!(DARKEp$0vLc6@;dsaSL3?K&esC=af0k# zbNCSAyJqS~&O1<|JH&RC#U6?oe?xBlI_PtJ+Q=WUJp(oJ8FV>|cgj2X9NGTg0OvH; zJ$X*vLuXyW=BEttx5^h}ceDZQaT$3utE;{T+-V1$b58q#PvKX9)OYp|z`oKMwfA}} z?ysH2`>+H30LIwnp$R?gRq;9V{@J@VjSKJt{EYn(EUyZ{T{i%w=C%u z$Qx(Ag^Juj{!i!~z9-nE1^5|V!2dNmecZ)9x&kPNt@!NDW1rSSHm7^&0)2+)oWWhC z*wlHC@(}NN58m^(&lYxP)7QDozXJaO*ynpN!q1%cx2YanqW=o&v(VB;Zf9LZpTnC+ z-$i8mS_Y@UTnVW08Q^1%cNIEdo<(HOxd%?*+vuiej{FLk=NerH-n`#H?JGeyl#*9_ z3YM^qpgDSb^D}^-N8g18kYw@QYW;ExKd^?b06q9!^!^?5FK`Haj(&&gpNF_3_qdE~ zKQ~bOD6y5$8{m!#5VqnoBHu&b!dD-AaYpB^L;U_O@yU_?zvx&ThWSr+@VhaCGlvO7Bw z=pOKQCU@%`#&_|XL#JPYPP9RaUjI|{&2L2Sx9Ijp&M|?mqPvG)Lc8d!J>3V?^L2(J^xF3E@nawE%la*J zHT-MnIzG<#5iEfXJ50AG-&K(@B^O7tt}58>^_=Y0qC zkXzV1Q|Gt>efU0f9h*Da0X2E-y^Adc?)VN_opY_>p8#!lAk6e5sC84QdAtK(kXyjd zJk%cDll|XA{g~7H;|`?HjK!wk^j-yIa@9~}_u)Af;03-HU~(IG>i!D+%;`=}kWbYS z6Of%vKFOkNAIAI!tZ{_i^LAIOAl-_&r3t=m;QpOEN4Gfh*B*d%tuer77r2vU=mYvW ze8UXZH8$Yu9z0k3ZA0y&!E{IXdDh=m=YnVm!oFQaKW**X=xaXlGkEHsXg3E~!RGU2%m}&!%JTpAzr*i6e%}J`hUe~` zI6^)^ehqbwBmh2L_;YkU=oR)|Fx@9||CZUHWr#C+&INLUEPRfAw|)k`&%Os=z;`p= zd9JJY@6o=6+OxUeLv!Nm5-NDsJcpEe*HF*c9Y5W}3Na~|_6{{JK%e@!e{(Kn@lJYn z39`?o`a`Jm%0J%5=igo0=U89RMJ|w=^B5sp!ya2RUU`jNZ$L?|r#-lniFu$!AO_xZ za<&1yan2Z!-E|I4v-qwo;FW!kc!vsdFT=Ot-{GU(=XnFZ3)R;%_I@jmw`ONWTsf2d zUYo<`9k_%425P@6P_YUAmBl(&@XlR9?O_Oi1ngVCnw;)<2%SS8pn+Pa@cI`}pWP06 z>w8|#;3wVUf1s=J@mwCjJ(C;^ z@1lThdM1!*P4u4aK0fx{hOg0i4nA+*>tPl30e%NKul`?w{5HPR0{lZ3`?KaJc;{+E zm#_tNN6Q@HE|M(1qyGW#jMmlv0DX@f=60tm$m_s5 z)|>~<;(44_@wd6>(?Z`uzJ#uV1p6j5r~V=IOdWo6P|u(PJoh)i`>_En@a@32i4A3( zL7(U7{tnLJyCA_Y0MFCAslT;K6*D3H%)N zCtyzVufn@m_mYSBeodhI6zJMe`8Gb**@c&Wz$SGz&uAWb1U~?q_zaNEdjwxY3*i3k z&zU~J>*Fpy!Jk5l+jv(e_>Mu1+ycwkzTji6d8qk|5bwnVyL;?nvk&Xr?#$-4XLGdC@1nZ^AF8;kMR@you8hwTRCyKZope`z zcRRzszr*3(R|+nHyY#MF!+Pd_00n%CzK`9UHS!~J2;_oyH>+TJkFYz}m>e~57b&v! zufQ=d#{#rH72;B=>3>?4qn;4dAfV*+`YhVPg-Jbfu8k<1f z1@v#AEMvXJEb7hOoAE=ad7FCDZ-DQb0POoC@VQ9PzlPescc?|}b#(9GPrv~G-=XGs z5A5MHvXJlKZ*F-%^YGJWwq&NjKCmz1>+cgicMN@jZ;t#M`j^02eSSOWuR($? zL3ZvvXcz2&4)SzX>MqI$=12bs-v!U$CtL9hJ%`W8hfw=6&U`+D>K=e+u>`H50c?ZU z$ohIG?B@pEC79Ml_8i@XauMP?ppo5I3w#1Tx0}!n;4}8Ex@^`1aTD?^BdfoG&Vfx( zA={ImSI`_e0VnW9i0|orWZwz)5a8E=e2Lwh3!nh%>XIDjo!v7SK%LF=^1bjivVMm% z+DDh*KSXZh=b823&GRSp8Q23|kl^R@YP{%FP(RYZ9`|u`ekJ6PXT`oe~G>6BUguC0z!!Q!aHf~8aCtAy9aZeVLL-t z&wN|xt=GmDz$U!=a=+?#fjja{-LdtZtq0$Q??Q_T89$$i6rH;)pyn_xL0$sNWAx_i zLzki6b$jr;$vFndhv+BJO>`+bq2Agn=*;tou7vi`<$-+Y6Li*Zfld~A`$+JU+H(OO z;Jd){Tp(^8Ea2~a_F`|ITLQJVG43KKk9*YiH{c%6K@V6{fAiKMJ}+}w+^gr`!si(I zH{iZnQ19Up7!&gm-6@#v1Kk=3#64n5(f5$`cjgtSeWuVw(8i}$2KF>za0Z;o965YV|NBsLxr-b+fcoje7x-D%S?$5t?}0jZ?LP0|PoO=p3qQiQ zz~*yzjP8aU?%y7*G5v(t!w~s5eBJ`YfPJu@#d8iJ_G`TTouPO4_HYATMrZySzjJH>YF`6=I5*?$MH-0dpjTc8J!J38 zQWoET_EBIT%>3+KzcX-w&nED0x~o6Xc~4THeGgn?ONp^|NiOg01mpo3-EkJ5*$aGq z7QNqo=HRFO;ZwmcXY3PLNBuRrDaHRDpB?n>x(;#o+T6ka04tyiPJq4H{}l*X{Jrk1 z_Sz-aG1PgzPx@IeN0vIf`$@pG4|E;)F8YMJ>Yt%AUkQ$?xLN3Ij8aSpf2Oz2Jf-G10O(&-abqC0{bO&LcD#nfx05l zH+&AS@89u}*7zk^ywmz?uR)3a9ytYf_>Mpk;{LtYo>{XS>-y2RFeZzB`k!Jmehv6( z-$uVAAY_aVweD1gn4_m$f@c2}^#ZX_`>NrmKF}O_0E!ZxIB9}xPJ2lLDC2$bzIV_W z+k+3-T2S8s(`OpLJ7mu_L3a(zJ%-L@f1RUev0BCFZU7x&w{{8wIm};^dkLHOYIGa> zw|8e4V)racd|Oa=X|Lnkcpr20d)Z(0+_(O1U~kj+3i;>BH9P)7fSy2)ve?5nP-~wHt-qpk;pah;z4t?qqh}ULs z1vF&T0OlN5@$TAtj$e+w>BZj}-YxH8z~1QHL&I-B;TPd=@V^E%d7k#*PNsV$)EzhR zqbKo~88ailg5Klfp7c9Lo&!VVyC<^s6*KOo!d2!3pi*$$jPpi_2!;|qKj zfVu+R2X*)zpesVmq2CqI$9^8L%dg?nA?90b`sCOeA7?YqI@HYJ7u>+R7x6Rjo)+Xbw$Xn#v%iCOu-ohO>_W^D@K<(U?$mdR ze$GAG!iTtH;GWdmzkS-n09rytZe4O-qg%qKgx&*toB(${kKS3G(fs}~9vbc9;^G@e1XY)4T0enfBVJhHPa zLep&4GkW{1@pGpW_ynCZSZ5Kx9T1Rjko6hCuc7NfeeMci{YR+!sUNcS?ZrH0h;hw5 z@wsr;4t@zX?I%?{7iY7+dkb6fJQC=8>~~-Xr21la7SHPv_#OLzt`BnZ+1p%3?lQhB z4&mL!GPDnjvDbUxj+@>_+R+X1HMRqvlWzt80XF+~W^Fh4Y@lx;w}C$PYTwNs#<%zb zcW0b2O)aVQjZd(5h~EYieA=Le-Fi8E8R9vn$j+T&3&eOn&D}6pg0H=_pxy)HtW$s? zdi%K7hWMPgU$X%Tk%=!;@{{`vY5|z-aUHf zFYxW5x2NCH<-pi3`UJGm&!anojz9(PJoE783Gl|=Ks`t290PUEqL1h4-3{oi>3%}R zeA=Pzxd`ys^>^p;e?aS!Jj{6wznjJV_Tjtm-V1ZD0PpAsxWi-QS5@q3f}8;F*&}xI zyFX(O@cBD>-;>(SlcL|L;yZN{+54-neH%N#=1$~2r=Mm%)3YV-*T@}WoNENkdk5Cw zZ=f$g8RD6{vkAWT?LAn>$GNVN?Zf;%c>6R@KoqZX z_m0`KyMB$$dE8l`uDLgWdG#?@K)!(1=sZj3-i7+Sq~Krh?r{^E3jE57Ik362Idm&v z0Dpwu-^Au|Mt#lg3<)_}$Tig5p1XG}2hXzj9&%5w;XMa`n~(J4^8>s&?RmRmHtkS- z9`JFGp1-xB54eOo(&&S=S_yl~+qkoC6$Raf^Hy+u(|3Fq}J?B*? zD)>9%d>*_r9qh_`@b=vS&h-j;0DlfO?nmIx-F+K#^zGw&dffB zc96XrOCSf{NA>p9)G)^cB*b54v7dSTZqU`h-Ron$dF&lv9rN^nbRYjiun0mKb9vWV z*xFmjpg9wF>uyHqd?r%l9Ns%=U+R0nyYhg}edgdE*vmHhIp7Yh?R{}R{k$vF_w!b) zcbdiTjuzB&pWu@dx1t~NE_!EEE<(I-CH4-wfP9VrDn3KxrvE>X53Be)Zh(CcJ`m%% zzQm^A55P|z;=3deAF5d2U6kl2P|v)8FX7uW+4_D$h~FR9UdUn(`nZ4d9YLLIv7*i` z?=bi=`liq4$nL_q>k?o5YI4fwg8Q|D?g9IIe7odOAMh!_2zi7pplf4KfxWzgj=&zs z@oVAdCxj`P)-22T9dlIS0}5pAo?%Lk4$!s?+~FAX(B=5dqf4N52?YiGNfvXgV7mt1 zi$2h&P26-wWY564YGSlo%NYF&bRDRAYnt1BdSDlyW^YpOgwIC-_0HwQ-oR@+1CQ`Y z72mhkH-{L4EA-CL^q*75-C9?A2MTn`qpeu0)Q1>-!z}N77SDGM>U=F=ji%4v(YKMe z@h_m$y!ecPb8B&1X$r39ivE zAlLBrpnQnDmc{$iLG}*bV|NB?=H$`0F5?dDH-MMK)z(07?OlAmXRDw)^RdSSq_=Tz zp0znsbTzT&w5C4FetO8-M^Mk_vWjz_6XzV(vcFKq-w8QB&gAz(gLQn3wU>R+#^3Yv zBP~LFj+(Q1O3mHI$Gv*5?5zh@$YcG-TXauq=~IF(`W7)0sB^iOW8`__ZqOAW-q8|$ zYv$A1n!T@(pXPM;1F%5s1Jv4&(5F5>qq_w5?lUt7_CQUYrdRh^fE=AYYI9eeTl6J* zXk#;v&&n9TG5Q|VJ`;3{;2yq>uIb@*#g4%{Y!ASF-Qm;0rw8@C%oWhLkqh9Ncpulm zHFoo-(3-yHp)1v2`K^;9AA)UA;$!~7R_w=K?Ae{q0q04OrxbP>ooBxZQsV5vbFe4- zv$ngDPe2d97V$6}mD0htTPnV4G5O`URl90|xNh#I1m4Z&J^~ zx`BLCN-lTv)W^5_kviT!Pgb%1nkvw3glf!Z(j4y!P~<-eL~!ear4Ln{KaeJ6J*~% ztMK}dkq_ZZbUAbrq+79%nmk?jYtRP&o{_kxyfG!d&9mWt%`pM?>t2g0_ULX~xA7fT zpkF|r6RZFC@Ckf?@8MGbb2YWP*!RIUIL_k#zIBe>oX+GP2H+7mUk!EtU*PpgZsYG4 z?^2Uf`hdQ~_W|r+v%iu!_k4%W-7Hmcz7cX8%pt$V*ZwZxyHM+SkD5Kq!S~Qv<5zr| z-&f%_o^=~|vx(k%O>bAo4KID8otW?NPvAd8 z<)1!Z+vq3wJb*wS_EZAvCEx($)M#St+x~suxbtRq>-|EmJ^TmwI8V?IY@$n{!rrI- z!@GC;KPJw*VN3~v2-GKMj$S`~d=`z}g>UlQ!0UGjy$1!c+SE;upMwGNCGc*#Tl?ri zy}NxdM9!&aFW!|JetKSI%wsO`&+3Vc%`;!ezv)HlPQ1tFPw=@%K0$9^uc7YIGg*cA z{On~JK0xab@9pDltnWFv`=|Ns;|+0dK@SYD>%Ry;pT!+^koS<^fj~}c+iMr>;^TR* zYZJGB&Dwm2bn!{BwShY`_8K~myaGOeW+zXnbJ=g8@0L98#07l+2F`)?p88)QJA-@J z%i=Tm?=^i4kL4SC#~ec8hpB=B8e?$=;}cx89w{O&ss z(XQS;pXPLbjqEuVpaW9;JnwheUv~ZMvyCnX`^2{Ksi4;FVQWF1?KfcT0aya=J_YX0 zbM$-B+9_zEbJlQ6Eo?c|+B;c%=b2*_`!TTpHROCN=Cp=+bK<(#`@ozFV1mAz#dn>( z@1b)qK1b$!dKQoPjzEfjUO)IgXm*o?_}$+4xXUi~5^NK52Q|mP{i)|2xB%uU>2DL> zyeSw1&%On9pC$PRxAaxT--i3>?8RO^TlcjOPSFjB`6u*G(9Dn!AAozPp`MGe`nt1K ze0|S#kj>YJZ=vtPo6Ficyfp)~gIwTy1KQvW{UVsdX3d&9hwx);o?`**Ij8m|yna1w z>dm3fb3Mu8bFY1ZY#l#6=x8f`4`0JS$KC=>{u13!#5&&+`VIJdbbVmm1pOjX0R1;mdC%O~7QB5ugC+s^+@W`Wp7{dw1bB}Yp*vN4ceoS3 zTg~G+Z$e*zF8mtk!0$r$px=R#oCW#R-@{vX2>%Q^KtBQ{^+Or+dtO8Q+>7&?)7^HU zU1$qj!WYEY|7-XSbnC$01G;7SEBG_;0zN@klRt+apmSF@@XoRTPT}3p7{3yD7E8z{ z=;xpb)cxDPcg=H_$zB2tPej>;-WF-8%4G22k(hb`}47R2Lt4eeTc)0oj#AizJ+Hu zyp6xxJ!|_g?l6mI;Jtnaze23ffIINFx3TuC-k#j|A^a)OZ+bTPbdk+*1Khp!oW=fA zu&o`vGn&JD;XDO?6X*eU?@|Fe$oA3&&S4Ms?jG{M%;@jXivd_fehy0bhue7fULm{J z68|>5wcS}j-bZiFYh>fEu)7OqP}hdf;qx-~@PO=`18mM=&q>DLHt^DUVFN!8!b474r{}d-ycFGLQElg>K{X4cN!OE@O?C@E_6X?>jRmwgheb-BlO-n#DW3 z32#1UQ16{Ix3Q1J^svbb`?&`5_}rjBhT5xsd+-~|@ZPfkb+-@D9rP)D3;Yc44umu6 zZ{MEzkLXswD*6NHPeA(}wm!P=firZ`nbSK{fL-`J#CK3q(JOol{dfyKd+kCO@u|_3 z)EdG|-$I?$Gv9{y8K{Z%O#T4s*Q@w0at`<0pXt0y%lMc-AyV%j%5wpVvyjzfGdzzh9|b8h#oPXPixuVy}9<5PsVH+w1JC-`_5 z5~zE#AA4{Q1+nh0LtH@TKEH=|_8i*={1VX5nbfzT_TgO^5YqwP8~g1+hhT!;{pzoe zdA^2Ep&`V1cHobIb?nDHXIVV6ub}eQUBJgVoqHeNI`&dyb1(PWvHbv5Z(sIPhxk3W z3ZD{_K)pjH@Z8*$zM+iw&GWedVdmGG*}M2}XK|KkUi{nW)Z4#z)w}ObCg|4ib)U}u zPiPzcJ^DPvKAxd7k9W?#r{@5*m#Yl@jDLlXv)g+Qw7@6q>fQiz+Vc{;`}+*FegSf7 z?IFL);@x%jP54#deY3ZcIoyNwog;zf#JPhm7;MGgK0DYB!D$xXf%^RjKmE{i3)yG= zC(t3rb9f8@_cAUpHH^ z=5u}EozuN9kP2*;5hzdj5><;yVYmjy3(g@H2D|tbsmp+faArj3u(? z^FkTE4W{pDbQ|iZwF~vEr)M1E-!&T*e)tU0i=4P2vUP=b;S5@!vrqH54|6U9`?$yE z{PyAwuaP}3;r1PhgsYvI^PWex%u>w-GTHP{SZ1PM;T)5ddAL??|`uv z$orti_rI0F256(3hvwiM*r)Z{P_!2mdy`*Ytv!rS{Y)V*wj zeK0}yTNcmj9n@UT=KkLz8)Lt__~u)224|Fifn7TN{?i^}Ezj8f88-)A&f>0Dpw2r+ zXHNmyzU`r;?j!USwsZ6Wnv>63J@n?-2Ko=cPw>+_QpJ8d_`HBW1!L^iy+P;R^y@&a z*};}0uOoYY<`mXRsFy+m*u?gqz&y@*k1mDxJdcpuw{dR?d{V~#yVy%|{0(Z)LwvX4 zZ$KZ|W59k1U&80`i|`5XJrtnEUCeZIP-mF#s*L~bAwk~6cMX_t37-|vMYjI*-RN%d ze}&GR&T|A!kavLe4qbtd`}h@``TFGBhj&*a?EzoVMsNS_XdsW>KHP=970`Wj-urjJ z8Qwr!RrGrUeLz=(F|en9X)`wbb!HdnUS{#Tz&?lgdbiHu?WYG?Gkz7?1NLIRU!>$p zq1MmKxZfeN=Rd0AyS7V?of%(Z%dss$*MK>#`+!e@-g>K5JinTJ&TM>+&jxU>&M^Ib zhU{k>|7H01U<3XN6!3n`HT5TtJ4=vD_zJt{ugzXM=)Irr`w;pOS=|BH0dpC(iECqT zL+!~Ow17FB!@H-=eW&p1yV%PRdC#hYehIpQ-hKJ}xhw1H?|fbC2{gsWJl^dh#PiU{ zn)c%FR`aJ>%yAA~Lv{zJpeANohx4@dgqZ(<-dggD={+R{Wjk zU7F5=O}_v(;r(qetyRX@V{}Jg8cz;=3Vixd_w4?b)se3zM<05Ltp63%b6ry|V{huc zXFeY#x`KFjJy&st;0%5pIFmbBMdw`pP8cD(E9be#ZV&dVd;qMqfb2OaTPr8#4&L7w zhd|#x^p%tx_t@Q)eQ4_;x8Uc|eGLZ4=J6e3zZ0l;NPo{=xkg_E0QK3@zlk;Hc{Mw8 zeA~chBcS)365=}83#fgk=)Qm&EQ2n-`&E2jPj`a6jXWk#7kfZ<=hGciH_UYQu!7$X zvb$P>cLw#l@B?)F*zD7Lr+$p?1v=;2LwAS&NTAn$0_L$LA^y9vV|aIwBb)OTFps-& zMtR|R`YcQB+j9~fc+WzgGVr{?U*O};_P`MPDY_=lB06VLehY7X&(8iDpMpG|(|70_ z-5k7osI3PQ_#U){{t@bK9@N`!h&_2XhWNQxV>Y1frn%2O?AmhRd*TXe-(6rHd)o(p zf=|^fxB21i&-%WT`k+Pp3G@LB&>QEjH<8`ZA@Usf2Kcc~DC6Jgck+Qgfct7@mX68c zte)dGv;{SPr;6_vW%Ik6HgHaR(x)iny>@49blwN`PxC+J?RgaxA@0=O9wFO+5!h4RPn!id5==$Hq<>BkMS)h}nUrwwH5 z+mF2*!%Lf-(rtWFaDYwUTsw3X6e0FI*rFGBXMO>^`->Izpf&oBP-`Z{n$NucoplJm zkNyX3WjsT7?)>^6Rj5OJ&vdEL!{(h&p~kHM{eA}fReW!M4Zi|!?Gd;JfgH^&Cl$5H zrR;aR@ozu}$o~o+$nQD1J7?Yi&glIxPK@wbM1KiRtN32Ng5IDn(HX1VIOF%K7nLA_g^S97)nvboIfIlHF-yjMH3Sa;;TsRMd!bHux=3AzP*`k;-h z?HZk^1^(aS>s$lqDSkdv>+nnP?tKJ*0B?;1fBP<>18mkg1kb@D7~}Ix7W0|Y**^p4 z^30sO>H7r!8OX`we(m)NpPJmEj5Dudf5hH_`rb3%n(kqOyeUsVe!NRV7^gntdJafnWQ0asuEthblN*iC2~ zoxALyAAk|??yO-e$#47)d;;&R_LidaMejN8j}MD=|07&f{Lq?LAw7cbC3j z$H;kzb6#X|ug&@DbB|Ar-LtZWyYn5=)X?W2*w%padI#+N4B2_#f-XAG#XT0}ScL8Z z@602<=UL3{?v1|&U3443y`17}?PDqR8u(k~2$a~I?HBkV{_|OUR-Qux`V^EQ&N_i# zz;A@#*U)J$b@(g+z#{rdxCBQ{vO`A1rS2) z&A3Hmcf1UB4tF!nhmY@5d264dcg7Yv-|J)ed1#mX`n2KO@bA!n2XCz&`~lQ+)n^la zANYQlpj&}Yq0U;D@tu{RGyhQ){eAaM`?-z1-Ju`DTi3hOoACwoint=gK1*nh&e(JG zCBE`~_&rcyPp})O-ds2EXBGDk`}g1t{NJI@qK~!KXZ$g?GkEv01bX0e6?@o#UVP2&Cv&p@-lT{8!Ls4@2T}WOufW?FBHlOO9>ykI(^l z4Zn&_-#ehKgRCuu{|Il~8PvMXjOvmQe?tcJCH^Ho6}-E0e)}!Z1>haAm*?=Fv-!=@ zLhd2k-yS~BAIQ54uijlh2>45UJkKt+Yxo@f0R8}e8QRBY&OWle{|H@y9)aKB-KqQ8 zgE#N=qrW^pU!iYibWh&H0R0tx2(j-9J_qIrz<$ktjBXh?*BsRE6VKh86MVYx_Vx}c zU!%Lj*FOHhr;VF*ov|8);7-?KGU2VV;_M4F432PzCezzd7LZ7FCbfE0iQtZ98~`Sx&e0QZo?;nS&VrP`slBq_Uv5K zPZ`g3qAxo4l!vKhdQZF$1vY!^!TX%#AR$KwxrTR^8oPbA(Yrg(vq0B}+Cz6M{$1j} zQ+WHc|0%BzIL2m;J@xSGU=_ZPZw}1!GqQKseyp3rhY)vTt(VxPE%ff${+-wQ`fE$U z7`cnShPS3V@9P>m_k2yPJ$WwHwcjRBO}$XEQ+@b+txw56_*mlzy?OOF_Ik#9_X4^i zkPqIP%gF9)uoZXzfNlw0iBFABg5Erv@GW>@ZteQMK);XOJarj=1C8*RL;nEv{rd^t zPYQ1>=hR+cFT>O{edh(_C49EAIo|}i2RzF*$bow6&*PJ*$6q-|?{{N&EB;0rLu+6T zZPqlGJ)IHj^S4UukMKQc2e_+K=pq>7e*zuh=O=}F_YSHkC-67$E5QBer@ebikHDQL zWu*3027v5cx|`VwVQ0d*GV zw6_A9f-by!Sb!S$39JM2yYmgGchOj9n8uPXC!alj#-~PJ!*7Cp5!qcW&GbWj2Iv!L z06D(Su#4Ti?%e%%kgf4Mu@(F^x*>dt|2t%Jegk!m9()&S-aDwfGv_t>AMhE&=kV@r z4YcdJR`aO@Y)0V06zlP#06}{%r??aY>tm-(FM+G z4)2P633oWgWN&5q){zI$!wNloz-F#p_`0Mo zWaC=sa^N{lpd;X2^9NqYkR27PA>ALtTz zM4zBHXBYYmyn}lpZ0C^(bj`ke+PA+En`Cb z4m*QiL>G`x;5~0^4d6eZuhIVszX0#~=u@M!Mv3n`^wWEXK1IGnZo`kMwGMCH0!;G~ z_XXV?cKcg~-eGG2^G={Sa87d+)NptDdd|L!a_}wi9-Lc}#>??(SMCTsOQDXDGvxYnZ`@p;}stD&=0G^5QJ!mM2gO>QpkKxVX?gFq^cP?Fk zt`hHgUcr~tONd!R7qA;Yz-I1--eFUAH$Q_pP~+#bunMn!4YZK;`>!fe&#zPmy70aq zQYmr)jCDUDaIO{e-f{<^tKd~OvqA6i>0&d7J>4LCo+}lx&^8#M^LhIY-52B~umH+3 z_PY&lKlb<>x?IH??*0lK0rS{@gL&kDzquQobO*gN+l%MX#-7T9JKzp$bRk5)8r>Rx zE5!T_IJH{J-rvL_VnEz7k@Vk==#+bykrgC&(!__cDSvkM&-I z4fqy*=CWpvk9zl)P^$}XY!`ir-v@Z}=IEX8SExIihxUN)%@wc!a_ru-@8PF?kS|4U z&cm~CAI^BGe;Ik@J!Gjh?Zv*Vd-f#e2z>{=I(-`y=sini=lV__82_bWEMkW6@6_Y- z37zkV9Q+;LzUvVG&bmNXz}w%1xDGa-jca)I#H}G*@e=nz5JX<`bch!5ggMLJu zcf(o*zC)<`bErFS(|yC)tWn5e4gGw ze+FZ9*jvDUegOBx9YYIj{ytfNcRqX4XAzw}<>cOlnlmZ+2mc;EOI6&%e?tR4#yYq3 z3%=%%w@1%aejZ&;{A;Mbzg2NoZGXToqL=sH{RDD+>|+O5b6Xz0yY#FMk*9YG`389; z;PrJLXLL{ARr}Gl2mc)S{uvbSvsQ@FU>+`9Ye+nk9S})B!)R32#50 zS)6@a$5?Yae-7_C+Q%Q@3i$lmUl;ocazQTdo4q)FFZ#6zG z_*3;I|G;Z+@K5y4*TL8CGW9q5-bQ{6-rm()%N@7$3vre?S( zd{^Hxo^^9yy*DNP_fUQB;LYRvD}^tqV{8d>u#J!Q1{=iPqd$kLTZDeb=W-_7gZn&1 z-UkC@pC8Z8bMmaMdk)-#ITmJgSVi9)*?8|v8RA|C$ev*vn|rl4zl*QQog?4FTf=-6 zz9VFN^F6Z%ZS>CNu07vZ=-bHVI79Y6$^Q-g_n;2+3N=<=srUr1-aG6Z)4NfUi`aeS zHnw$W68Hx!!?&;*JAisuyxVV}W2o;Ab=FLv-dpEYKZKv)vjX*8tm_W!{}XTzYf$a; z;O;iQ8v?Ya9V}Pm<>PFxkeeF%kCkV0Neg^_#XpqHzZG-3=chhPGrP0?6W)G3<2&ft zO#TC!0DXLpPM|r|yZwJic^l|B=RA&g+GYu&rWzH5PP;)6Of{M3|IZdrMpO_4K~QU~ zAc!?=5xc>t=tf0FRU1_`s#{T2QB_fGJ(*So-EOumqDu{HiT8M~m*sxG^CaEgd+xdC z+Ili(F6zbX*w^55Ost8&#sL+GbO{W~q4L05+?eky!`80|X-odjKrZXr-U z7X&>2T{sV$>msAm8N0~ugUXcuPK~2$2AMsePovi}w-sJ9*8-1toogFV-?=IOnbEFv z#+^`|i+a;JD_*>yzipd9MutuRVSJ`U{dwZ}W~1{!75b^@2k&$kot~{iQTEt~Er@>- z=*%>$=lb;;@KJoi^WQf)`1WGYf~vPF?1~A3GzftXumHXG*aXVqHJgXNzs#RS{)$gi zNAv3J64x+KsUs<#As?X3Y`y%j)pU%fJXk%K_< zhmpl=r^Qi!o?c|_GqZE_fiC}!rs+|!eaNDMbn^3{9mrEa5zx#oelB{Q>0)$w=yIUy zS9ddntiF5DX&-Sg0o?+i*^AL1MMk8yVw_Y`D?Z?P@OnB%@!ag zfUXtjPEALro~xlvK)vTf`>^N2_v0f-pXn7_5MDj>;?oLMzSZa#z{@X&ya71|0@TwT zbP0p6(F~11bJq>fSQ$G=VFX#|^d3pN{ynZDR=Y_cw_ul#`c;3*tGFb*^6Ol* z^IA{@G;cX{WfWg*9{vT$L3o#Z9mpXt7kd-v#U^_Ko$iBd3FIk2J5bF@=sjY!|M@`A z{KwE9?8@OGf0gpjAdO!Fxf8_D1(3r7a@zCHCD4Trd|}k@OMPmtIPx4~Dxpcx39mEz z0L%yKuOHtIVBl4wLwaHhkTtt{+6eDr)7=Qj4z%}#0RJ9HfePfA&<44XF?L7R{UlQ4M{X3#_uacj~ z-r!X$j*oIFR&hFK)zeSaU2R7{_ zvkzzj$X*>qFF)*>p%nyNKfW9vy}ufO_FN2FfP9+KX`ZnBK`#1SbjsNWdb-f_X@;s7 zLAg4SYf^+lRnvn{f^1L(LeYUcg-vIneA-zM+h88iBG8O2(=U5y#IAkp=%e0(#K!QQ zGGK3oDn`#?2wO2W<&UEqjMW^7WkIzA*;+xO zjCs(-;WP6fs|WecjQYQ8wKwfYd+vn}))(!@rcWGS?OJD}eW|w`PzVP7tI-egqAs}# zuq&<+ns)s@R4*G@JK6?c1>`$;9{tEMQer9gZj}&GiaZK zJ#0qr%GX2Z`t|hQkX<|1ZesYTP7rx27>r$negSems0FG$4PBZ%0eJ0GGiinp)FnqI zAAGtDp6?euQy%^XAfKN02y&w!+WEcOz-b81neE9>Y zzL%6wHKro#j+O)Yy@Kp{{_hsadY)=Q7=PvLAtnyh-BW$-PkE}~HG}p(ct0~c!#Cmj zcT016*j85*Fj&|2p2yB~dJ6dm;jp>xna!}xbX`v>?P>jpgL;%neIqClCob;P+9#Bjc&>mbc7rqH3 z@eP1hbP41oKy{i@ehtwevib|-ryi?-YJ7)HeQ4*Zqk7x$lkR79xj_7s0jl|v*mYi_ zGtsXE*Gmj?}kI-vU@U-iCN`lw&eLmz;zM3*^pXbtEj zz8I>RGdmujX{hV^``5X3;HO?gP}S^#SI!tvzskE7XrJBCBsz~i^xTEoJZieO{8)yPw?MJ=VA%}sUp&3y15Cm1gP5HUo;B{UVpcX7Z zpGKDrmA~wPGIH`PsTb8&y@kLZJ!=3TU#qX4ko-siXH?3|akTa>4h*OI{xJeU?G>Tr{H#5F?vv5B5Bln50Kf z_)P+;son+U@5NtpwjpPFW2f4a@+5&~EJlt)4Rmmi_3*9ea-)793DrQ(L$BDh>pzD9 zsAdYH3wHVUr~p|zsE+zE9oV$PEM)PzbF!(Y7N9ugo(8W^A96czT?BOHKzDR-u1P8M z1J8)gM9)qII`tMS^Y2m^s#>Zc-=2Y(xzH&^kda_$&~@rww@MWgpyC5Zwf17rpLDHN5-+UA|u!UiUFJfHeAr zDgPO)g4emE!Qd?Y=q8gxy{SeAzHxNw?{gr(CTMZg-=TaqBG-W}AORMGSwJ&4D>t+Q zs@XyyH-#VP;<@Y@SMnJH zB)&;x<*&p~ANhGMz^)z7K$eeY(@qoc%Bgu&Q~Oaao%s^%>QVVr!v*S3-@SF92>-US}AbT!0@m%sX|BsK(+_=rx9VjF-#RzBrdY%frc zJy7kd7d{56@h=7ea?g$W=Ti$U z03P~n=$rA0Lsd&yikyR8wQ|9rz6HI`FN`jRe+XWE)*~lf|5+@8PorBlKwWs{%Elsz; zOf7Pxk(FaD(9S1-9_-yv)p|YZ&+0;x_{X3L!Sl~Q3qF?e=hx1ZL+7H|x{;fqs#%53 zBVOmb5Lx$V5-3mkm(RBA`QLXf@QM}YMg9Ax zeJw=ph3ft_z<&zT$j!v%qKgCVJR7Q=7eed65)edA0{N^@;n(HwHbCuqbd?|tA5Qsq zMbC?J<)PmUwByN8&8Qw^3!u~dI>R7Xj;&cX^i$D$&{kv*z0P0?h$E~2!84Rk4*K~( z{T4xMfqqlz4BCOtJq@z3d-NEG&)fxM?Xh2W^!iRxjTWGpnm`a}H>wv;`R9@B^3S;! zz2*ym4s=QKWcu~|_ec(U<us6BVWD=rpg7u>rBsNxbX zKG@NCZ_(>^JdH$LA!mFna za`a%I2EP)>Hwl{IgV?g5l_{TZgAZWKLDsH1;n%^-o+hu(z89T%-Gwl`_9CCO>z}3a zX)o32f4MJ5%y~$7arXOCPc$a^sS?F}vlP)dkoE7zVBOldIl2d1+XGyb$K>#0puWrUC7g_<7 zzYVCb?lRw=P%Nm!9tQO({|vQ5?W+x{p7e>ss}IQw@Kvr(p!>ZIz4FVRMUF6N2D+z; zQ$MZHZgdG~!u6j|?JW!%q{m11Ef-$<$V1)$Rd1TD2wt^&@actS-e1@}^o`JF(2q^} zIJ6RdHGB}L7oBm8xOwpH*lUor>omPoK(~OY(Sd&pgI6yhWY6>GQ{5W$+FcrC!)xD~ zSNoS;zM55a3ee{P^`!oT=wzFPPP;9F*WP-NHJf?}l~u~%RZ!|N;tcUFy0I|!gt z&06>bz7wG7F8{l!KI)%^i(YeTpXKOOL;d&Tr~b9yEaV`3TohS)?KlKg?aaM${od>F zX~swO^gO6{?Q16bDrhlO&@-#p5PTJMEwaALlEkPt?L{BWs5_){aG@>eW!L>yUkPH% zq3X?pS044#4%ObpD|RaOPN07Jf%rsJ89s&4|NU-Ugggnj19r`%{#08tDpv#c9;gcqNBunNbuO~z(2Q|p zgH3aIfi8?aY{auB9=!1HHF z!fRgXRId>%1ui=6KsMb4?Y;@RFhwlAEQ4xB&Cv%Iz{j9EhlK0D@6@l0ZV9?J@HtSN z;xoNt>jBzHFp4i!{V3lgWW{IWub%qR1zi7~lZG!s7l``*o8~#tCS>K%Tws!&pL+iDkafP=PXTDf*Ffbn2v2s9vx8p!=%d~=hw4@lt2;UYUU#AvWC7Jz zPdaOzPd8YKUU4DWKmmLws2QO12lzNN>H2r45MI98jf+k(I`@2_^9f>$Vbe~uH`NIv z=coMVUeCiMc=eBwJzK7n32)bA9i@_WQ?f$I6z`A>rCo`|-A z!I(~X&EUHJu9Em@zJiqh4$gtDz%HN3pcA`#oC6GUF>ui(i4Vf7#^%)j{v9kAT?jc% zPUR350r{y`Hhf_U8E78WY69v(d(=5Ae*$z}{+>%rCsl0f<}w4Auiy9Hn6Z3Zr}0qhoj9qlen ze`#nf_9VP!l%IO_kQbsCuiSbM=sw1fv!cGg_SS=5Ig%iZez0yiyk-ak-FeN>IzV;C z{rIbw067Yg^MDJl-e-VT>;ddakVdaNrgKp~)zS>|)xIUmFAmjnup0e(&<+1Ov>asS zz}^9ZQSLlc_3Ghw+;1DcO+fVuf%dEYslO1I3_m#IJY?CXf+~=YtUc+Rvw*%gM61Z5 z9qG4P(xo0~j{5J$MyPhBbJw$^90BZ}$Bgj07vhy?$3D02AkT+aoNDxg1hRS<>{HYQ zF>D^WMYE9upeE(tr6g2-Ye5T8&;4Eg8R?vaAlR`7?J$8Z49z0OlRf30_Ze% zK6;&5Ep#zRy8apTD2BVS4&5BEW6wMC%3p+Cds+{l2Q;H%w?VgnIJy|f1&K1hhW1jQ zLgw934E3n1*>soGzw&6mo_w(Nclpn|X3~x{Qz2A+=HOEaAA{GNX{h$6eI|kY!sw*e zoFR0}pz3)kwkbgG-)eN_papo;?1t)`1Hc2yp^u9`4qp@X|0eg!{2mos2fDFo2iYL$ z`nHY%f5pe~_s|ERE_(G*6lE45x(u{{5PCf~%G)Epj9ftT1U&yY=4NcUK%WZqF)$BI z27Sb2K@}SY>Sr7ARp`>l+Jo!)?=_u69G&WE)(QBiKfQbOsUtS+;t%Fx)4M>=Q4C+5 ziE0F)%0CIadenEsD_}K9f?x{j`Mt&Pn<*KZCa3xicKPvP^yNVL4bUeJ?Z93e_47}K z=A%i2B6~430aQzMIg^Y>5Y&Sr z;GvhjBITcv`q7;!5Cy6ihz^_;zDaUUMmG&zA+#5ra`yqvEMD(Vow@qzfS-qNKQs&f z2B7_AL*+jKn_{bgVgl$r;##1cK>N_#+F1Y}TIE`hJ=Xn)#~dXEpt^3i>m2?E6DLaU(qU8mY} zfacAC_MlGy&E!E{@>ZgcVOKAT*NzJ1D;-GUqkNUA|NZ`Qkwah`_R08`g9LI6+6h&z zJp6w_u7>LOiOxcM2txCbTk#9QC!t%>E2l@TYVn?b2Q{0AuZOJ#eH<)9&h$zR?JPkK zUD$1XW=0Jk*2ctA)xpLEdiYTzuM~>PPds*u|@M4mO=dGx{R*VelTh z3TVLf&s%4tPZFOHwtj5Q#EQ@C5q&i}oxgUezWagd)`C{}%r2lwC6g%d{sXVG*buA{c@vz{sqtgHogBk zp;aJ=y%W7=Ercp>5BfrUG?VVRFa^j*cI_^PPXbIz`RA!zZRl0E7HE%&F8{Yj7QEs- z*Pp36>dzzJV)#~Kbyq!nV#G++j0VKfYvz|^F|-+~ zzFhRJKt1#$XJbo4Ct>%X?Z_ecBzyq60qDNxLIWxPE_910Uguf}T0jz6XO;yY>hg2N zh!d}!>8{GIxDIqKHr4asRWkt}hL3{|_zCDFYqoxc=e%vs-RPmH>1~GQ>^l=L#~Eb|4IDP)cX-${j{QA4L=j~5EH|$ zbC{2QCAtJYF1+-8AP$tX0NXV5n%6_u4X@`)IYaQGI?Dy13cY%+2gUHpIT^kUUUP|` z3?G1x!Pmnl$QeS`-jnbhqUc>>3-Qq&G_!KGz*m5I=$AuPS9c)>UkPmxkFV~n?yKg~ zJ4&_Gb0tvEL1;jLAFQu=(?GKa%dqJjUH`sMz}H2ehyQ%|B4`VckB5(X@UX`~PRifi z3UuFq703bfGvSk-ep?G_pj_?v)<7qNa0)$C{)6)-k=27@R}b)U?0p~~z3Rt+_B$7v z1(L3xs~Ns0<-fnxn{sv_7r-a+cR?7$(WT4$=Y1i354xH7^g`vA5ATt)6`v{ankx*S z2d|u!&{u$VTMnw=b+5!Lukr+-F8X$K>bo7j7^?bd&p*d1cT{k+#^^^^K z|M4zW-+2Q*F?~TA`{{4jr69c~vWoUeNE>_1FjNDUWJxgAVFM>(EWZH%?3r{7j&a z=juQFcSk~$xHMETF?_;6RNoW23rXyGQFek)5MF&wK+Xb1*lR%*df8OR!=7~g^OD>K z)Q$!}Uw!}S-J$1Fb*JN31JrjVRA(1X3++4s zG-D4r60WcJpvo(`hg?M|fA@i?ADhIcximv3Gz_w_^+VrB7l*EZmZNI}t!4i6oTN?y zTRyt2*cGQ|VaNUok+qjR&=U1yRo6wx$0DHE9ep&DcCTkI1l9Z%=muvJ)w@ae(*Vtw zgi2ovbpPu?E3pA+I`w~!4Un&!m@M>_V8_~nd;@yTqPrxzJnHXDGb?8k@W6VI1!Cwm zoBC7^%`d8Zs2=k{14!einFs3>!|V5rdiSucK-T?v3#fJs$UkHEhyyvuRru-*2lI&P z{^`A^ea(&fdnv>>;QH~4&@DyQ+?_ykwc{T{uk+AO)W7bFo)Mk@j`_tmV;2n%*z56e z(ba;LAcVdZRKu&6^-%eDL+1j;SAsY`3!?tBnFDPEs@H+OKE*7&uS3Xf@V(?wjt%fL zK{g0T2i4A0yEf{dgK}m8&D4QUe#+mu{54Y;eL1=&kaYcLy9%AoPPR(Do4`;q0Jg!ixq(C5MLm~%UiZvj48 zn$a(W$|rz57n-?;&_W>pNh$xH>dxiBYbMPR1Cr&R2U@@^pdMYQ&LiUo^~kNBh3b@l z2gUc`n*ez zUGN|ueJ#iabs&!2!@dowGgRHqluyG@-HR6VI?DjiSuPikpP+l#No+Ab3D@^;5DylE zb)X-+dR9H{K=U-F{NJYHHIwc_!lM`X7|23j4l40WfDY{1Q>!Sb#nz2n0Uhi`v<;hP z&eR;}anRmD9`U)vbO7zD1K*@e-vhZ+M|RzXDJlP14xy7>JroozOJ5S^f% z=nS;q956NI|5j7oB)*vsdfh)kKFXurCtd&EDR(QFhb@MW=3a(8xW5p-E;<*gJML?N4wVx#@2K{Zi@p)M5)^_;z3V(0;e+V3w`K6^ zS$A8#xac$a5nIsfIa`fQ&rcG5Fpu`ri(a#Y;FW7Sd>Fn5em#6SP%ZUY1l2r?(P_u3 zm3IBRtr(qMRmwl#ID9L*R^$-8&P%(~xvRcyv|Ge5zRwa4b)SQ^dN>$8XE5Msh;&1d@*{R z%Sz-hv=%uVUiU2F0*~IHF1{Y9!bf)`3BL#*&7``DRo!J^0eaPxz5+fDUx;3P_e<~j z=Q`pRi(fzdG-x5{#HO6OANu`O50xCpX99fE_0P2mUeDR26nlbKY&)_(nkxX` zi>?*E8z1$iy%)n*!&k|NJ`?b&--=#y>G_eq5?=YmXTukP+LXV4^?pxT>?xB8LFJBoq|umZ$ux*UvYh47J8l40?+`j=P(8zhw5C^LleC6iq~(*7?6KH zRPPKAz75}g;G!!+R(|PJGXRxeE&5)l3w2%eAcU`CRDQ)#e;(KK@0I3iASMUA z1$0MrW*&C=bYf4SpAhx$L;^X0z5|+s23-HylTQBK=tb41H=wJ4pANJO?OD*-YTpJZ zR`;$Q1VP$Gm-62o+J68a^_BTegiZE1b`RewsAf)j{!T@+kma8hB?hY5`+)98X763> z0$yjAi_bRr5WHg3@ z-&@@^-HU#5s>bVzQXOh4s|8{ihb*XIVTT=czU;9Yl zqx~0SYXcs9J^BXV4#c}2HGw{=Em=Lq;FB)O9nLO*tTT{bcbD(0{SWRc^WBcU1D*PA zPWksjJ5tZ}?w#%xfL_0i(^l_ zzD{>C0CX2!bmG;!Fz7F;b5OlHki<{tlOCWRzEwbd2L^OH_sqE@2@Y^omvi7w*Fp?f)q1Ucf!s+p91hyBNQ$XP|MnC!&s+;v}yu`8!~O%Lcr zwIiLY3sqhDi%$yJJ*bN=0hPbbR=G8&=8(Uhe12_5U8z-c>J#y$GB7R_z9;dekQ|Ktt#>ziNho zcCGw_yz8>Zl>eJ!3ckhgML;v@-lef?Z!WYPeJgyXUu@cgb~PD(4sh{Ne%U?zYSGOD z^4kbipQ@b$_3+hfno~Jif%K{wO!>PKuYJf~2k%0IK>0&m{yRrG6Xeo7Y0!*439Te< z$NQZ)E3f(ckY9{ET3Yvtbf$rr5 z{E|_Bzxw+pDd3}zcB$O?=vshg%>lYQ&A@g2nbo7tFX53F>{z!4eH%#1H|78PvijG& z+KFl?PX#*lv16=w{f@2z>QlLkpjkjWnH2SN4DLi{Gyz>&If43#M-XImQ1zKh+&ZXw)V-a89D>SMpCnZI5&~x!aH#?M26Uk=pV!>g#Co3p zj5otC1DYX?Zy2bqJAi&<<&<7M4DK}<^`B+k#}N7i{_3X{zq!b9^mWiWbc(4SpyIV3 z^`$))!mGD5@sf28F0wx5@R~D$oR8l$_yXX${+Yz_m#r9`YP#q%I_B58=**TtrBe-k z2K)D-{yj^&oE>)c-47bkt40mb`Q-uS^029n_S69lNBz5@xNaan^`9nRtjo`>Ig{u# zV=KClWT@_#dQxoC8+d+F{xhNTNs?o*Pf>k(6a#cF0a4IIjQUvs-O;}lS-pFp0VF^U zm<(osG_i_P9?d!ntOwtLP7o;b{dA_;QD4@M(gHE+U-w$~tPy)NkgSwPpw;+K2I^rmKI$)yoN)bJ$KX3bFollTfO2^7#Mes?q;CV-i*_$wGwZ(S zcVe1cRZu;z>SraqdYb^LUkCMb`OfKbxW2OAoyEqPdheKkCn2j814kPU91Rst?WT zy70s*ukwEYZ33CIMlb)#DgU{Y&!AqklN#GVCLeMGeqqV*g6>D!V^72--2dq-&fiNfi(HZwFItrH z=^SV?(A{+hbQ7Zfy-PqfxAaN!s;@$L#gKZp%kKW&`o_p;aI=i|0QdLG)YDDZg^h9N_ic7bLHNDpqzO3tlLj_yGxv6G=X%W%~H8@L;)d!_oSW56r$I)A3LdBVV11atSra z`JgT3KbtjB_2FU*cKLhFK`ut$5A@t`1xaj~okjica`nFueKxXs2tZR{DhQ_h-x-DQ zi$FQDelsXW{m57MME9c=x)9n2ZBrgF8My*fb@|^R;go+L)YDv`x?%j%*h5i2uX5}A z#s%Utz^2|bx8^ItrZWf=mj^A!-Ul7@4LBz^QDw@L>;6Wwy4-EIZ_#H-Kpls~t6&4G`h zTMRmpljK!z`m98!np2_u($PZ|vSw4P@+P4(Q+}-Efa~7}^%D|O_<(*;f$ZX+#3v2a z-qP@zBaR%08lXC>qyBey8oOdPAm;+rmtFf24MTN*HPiNbp8p-EJr|(Ycc6Bn8KhG_ z?NPQ?pdLM9JoIznb$ zs@b$-_3i;ZNArQ^(3$Hjlta&x>-xV_m1i=t@@hvN#O28j24fXJ6Ipv6+==>6;+xqy zxnjg->Z1z*?P4MR3&1z{7J)Q=t;k`ZT;-r1Oo{s6H7)S(f%(Y2)^%MuH9ZdQ6RrgVT`krq_w-Ph~^_36R>>W{L0lx-lH~I-c zJ1IwA4b|?o+c3~h3(@BR&9B*Amw0?Ms{!gWAL#xiu*IPZQvQq{axFIXxftDKXn&c1 z|F+3rdC{vHxpSsE(5Be*4DS`uI`XP)yc!MxHs9v{$D=(1)RV|*IrzA z`Sq6^S$omsi;lF#CB@^98ecN5*f=FeO^l8YocI6!qg|+nE3SBWF@*!d|5Fd+|L=P^ zYW(;TQ#>(x-ojsa126U zS?Z|YEm{>mde3R$kC&Wnuf6aDdqcx$yYlna=h8D{mE&z*BF z*~P4R{wPPW_kVbBhKg?+n{A%Fa!f}4Vb#2nNB=&qWYe#27B9c@Gw1S)56tN6FD$k{ ze)Ai9!t-lQ?6(tQ}dm>-)fDx_cv{?E8oR;+;qrEzg{6#mf2qxx_Kmsq_#T@zwAAiV+hpE) z><}9`c!)jz-38nCrrstz6fOPlXZfY!)=%v3{xjU^otkCd{ZqSnb+c3A{PIajcg=#x z6$=lx7an-7J@m=AU4F!kcGwrK_NlQm>=$P|W8Zza+1@k%1Uuu>N9?|1F1N#8J=>ms zbBURC|GDO_kA~V8=Uir<*j#3Y9kRg;ZT`4q&HgQsso9GoSFXC!Jh1f+^Vz(An7RKr z-4^wpWp}Nr^!-to?d*i*f zWz`+_(YNQ?b9XzS^w>v6mb&L;mrmWg&N;rN&v|jwQRdP^3rp`DKD6}web(EzpWEi_ z6WiVFH+zWrqCLxuufD@PdQ8~%Jv1@$`cC5_>7&nyWMBVpJNNi^>|c{Urjxect?UDLExzFN!tsYShl?L8Jj1N|=xXypbdK5U!gtK(yNB8P{yfHxIOu6R z?T|%wk9Y62*L=`w5B`1J*1Wo>4Q~0#9Q4{qbJE3?&MqT+&HOdznd;MuO!?AT#eaCY zxA^dDhh*MyQ`V1j&Z(GajyrLR8J0WTOt@}s=J{CL-xb+(*jR^1nwkm!zF zv-1ni_(eBIW;Y!a`Dpbc=8g4fbAMuQyJFsRC6$MqUb5dyfyl&*-n1{@+GVf1{BG;q z8Z6yXKeDv%tgr3tcki=PD;~BVt!c64H_Wu7Z@=C?^_OCMOzU3Gs4-2E<#)}AJbwF; zrI)WBQ@U49c4>O-!;urJ^PF+tU+f&PB5nS(e2z2x+!>KeY;MU#4d$q+-|kWT(0MAbw_6Pb66ypxSK7AUC8l)WOU%~wXPW4+FN#OK`|r&6 z!A?2%W%hE%bw5U4e5KR*`tlOzyooP5sfKa3{(};`x?qIe$vwd~Jvq)kdhY)AWhY@z zZamiBd(kUq$z$7`w=bXRJYCr3tSHH|MGHroUFZEba@NU9Oy2yJ=Hq)_HYa~PvGmB- zCYHu`8C803<-w(kZyZ~CS4D2=h6TBH)$vE#w?>~}N4@-*nO(P=efs6!+RKmMcGP}f z_vD$j!^amx{No`P}^X zyIt&FLx_f5aG;Y(g|ejV|;^W*w;&YrjZ6xsWMS0g`WS2@=lw4WI= z>r&^rHRneTn=;lEt?ACp|KCJuq;6_^iSF-i;oHr{DaW+^@doFubM~@V&pFCYiH)^I zk00YaKYADE+~57ld9(FK=kps+bm~sdbN2b~m(KDBFUaKSefbmf$v^&OPPwngTz2wv z&aMZ~ce2mC*lAg|lk>}xot)a7m~+KB$C|(iN0}#99%sgVI>P?$(4Fk0r#~~n_codj zieECHu3VmZRz|e#AK5Z$Y00|M7IXOc7Bgv8lS#F|@2uSTqVwLf9nR>b|8}0R`)2B@ z@1OsaQ!=x1-1zgK4clRJkF(iNmfLS%o^Eok?RNT3o13{i!}E4E`_*6MtbAuxWXl!r zMDDG>-f4d(*Y=eDCiCn*drq@+?pbS`?lHTX4TYiY-+hnmI>+>%d7l}%?_H+G<`swT zKB4%uY0pIF@Aq3fYHX<;@w+j0-fnN1ov#{Vzj|kgojun&A5LBndFHTpOKyDnP&@Iz z1-9jaBQiB^$T`M&Yv!@0ZQVQOrGi~-!{0~Q(pOI~BR~C}xqS5B%;O`fP2T8loUMm_ zw0$2i_!JKu{r{+JIp*z^%y#e!bKawTckP-qC;Vn&<_s5Zc{D>6 z_x7z*%>j4))tq_hE#{!kiB3hwt&wZLZHPR6%;J(&Juf()EUt7$f3>^0{^8lqzY7!2 zo;x38-WX9BsT?`WIs8vCGwJD#re@BewyOVJd;ewK=J#cL+hM!jYi{kj$`l^?u6Z{5 zJ~Q>I6_Eu$Z;EU>$cxsrmz+z2_nR%Nf_Cwgea-&YonVfCGGKPzu(zFXN{Ri`)+J_5{w0w_ZT{%8KOP+k z#m+Ysn}^sx&OY1j;taDt95&Sc^_Kh0u_G(YHKBvdPK&CY{cXV54^J|;Uv;`!ami+L z|3jVT#Aog?w*2kLJ`c1zr%erJX8Ux~RHy&gT2pb}t>%t~m&`YNyku@(b9QFVbNcHe zb8a}^^d0i6xpU;ncHM@1jC1QWbK+OcCN=+R)Az(;(?8~Jb69MsUH`#xc52pz_ShR+ zOkl!;rnzLgId12BoXbx8&e@B1{Ns0DYF{|?ul9lYxwc~4;bvm`TPM)^qVsw03bW^* z$JzbXUv6L6Ct#lab*))Eu{d)NyH7pC+4A9oPHXvEr|RG%OnS~V^UDR3B2_i_M7EeG zN-nwVO?!6txAyPLezvv^+s*j&SIz_H{my)P zz?o*-@MDerb*GYNK1df&`NwXN+vmPv-n}VhPP=EZDUBa#AAkQs`|M?R*t-jgBZs_n zYh>sJdpKVh)OPvW<4KWS9uQ*@b)aLAQ&u?u1-~VMcoxi93N9SLif;x7P z^QF^SG}In<uHX5Q~x|7qWReybf9`*&t?7WTR?bWM}v#0-XwEe-GU`zi!*ZFzIX6L%s z&o-OK9T|E3fp;V879AM*&4McX`}gj&`#j!hANlYEd*q7S?3G8}X2*Shhl%fdoVnB; zWlmcAs+l_BJ@duXVmou)Fxxxi2g z%gs@D&N3IAFv7g^(h3uQ;Bxc!m47lDZ{EE$@yh=*g%(hDNbEj=PKdF|3j)0ST&g(W$WMYDd0#P+#4v-jO^ek1aUJ37OUJ!FaV z@Fj)jy0eqc7x%vAZ2IU{=jIm@krRHr!&!Fv@fjbz>*n7-IrF_f{FbvKyC3{+=KcNJ zmi?TUPP*E0X2l}so;){l@P-SWb9#-l%LT7HZ@zhp^YOfs&CVC?W0pMIlkp$2_034i z`JFTDrOlCJ?jGkXd*GiL{a3r(;9U6L-~C!wzwZqD=NXx2eC(IgohR-cmf?@hKh62^ zfjQ10r=AjdU4khh;z+jqs+`biAh7Zg5OfEah8JVbY4rqSD6kmU(`RU2OIrq&ijqLQpbmxrk zhnkvLp?PB2iRR(6{%Xbzo1J;@zt%r76Fc;%ot#mB+%Lm_J|y8B5T42Ju8)cruGv)5 z_0_4Fx$@rM5}Eb++ZkRv{BhW-$gYvz$UisCch=W6XJUWZHR1g2uk)O*y`3WSI%Cec zp~s!w|5#n}LHSFOg|BRl{80FLEFvslCS-cIm6_Wo#yRBuQQ)J^UMt` zP0l+*uQD^od~ckKr$@dxU}9wAtZ{ZTZ?P-nz~Nf622G51eg&J!_$9JavY% zWWmNv{q9evIyJjI==}Qn6m!Oj!^|S5-HaMP-k$Q*TaK+6Zf?zeKhyK`-`yJdvihjV zg%`~**Zla7`SHZzwr&5N&3TtxY`TLbcKo-e#^TJKa~7&h&nH%30#_cxLDtgMn3=3Zr|gOkew*2tz|)9lK0nKirLVY*7^*)JN6{d>{h%-G&i zd+LlU&C%&T`Non6{0%q8FKV_!dTFH?NpP#Zb%E}MJJp{DK6KRH9s_}09%sxh+Y z&|6D>U3h7v@y@uZT6T>2@6ivKet)Rhuv4~~KQG(dan3uDgTDU3 zx#{71%qcf?MRHzy*ZF4lYv%p@PwY#>3+#KBt}@?VS?M@WmN_@Ay5G6ssa@@FPugNO zH2lH-^~6K$g=2@?z)pX#zt-iIcFrrcQzmXU$DVkptzU4sZ5%tzUi;FNNZtwmaHe-R znYj;7u!pTHvwi6$_SJbOm;+jKoSipb@4UQljCmn@PrK^GC+%1FoK-w2wYubxm+p#8 z-xRZdJM1$1%!f_(^(n`g1*iOEdgn~B>!&=GIftiy+9`97E-x5uX1C5WeOKLTPh9BQ zNspXiU#tAPdAe|k*<*RYF5T}yyXd5Y%-_ZjH>Vx-uqnCedh_(LFPp&fcT1j}*5vfQ z{ZHqkv(B>b4vX1W&*k4YX8&Zq*msOQ^ZGHiVbm;BTynX2s%fnK`=brE;*)di@s)qc z+~;}MjIggia-03~)Uf^M0Y}=up7V?O!??*s1h_>R0T3C2LH>nX{bge|>He2k&B@kNlFE@5h(! ziG23%%aMP4+G5{+Xpy~X%qm;F^9LpG9Z{Bfw{CtQtK_XuR*nx%>>B_5C(lGa`hC!O z`)_wRyL@`R^YCGw({}J!^G)I;yXvBIZ0?E2+udH=#lG?R8)od98K!XLDVZ92&%X41 z?sopJ;d_q1$xJ`$Z?_NI{_m9XIXd?!-aoxg>9(HjzS0Xn@4L46vU@r+{>T4pBUhdB zY@~AO6XxlvMdq=ZM@{#ln!c;6F|FgX_!e zrX!}B;LZh^^W5X5o9%{$=i0YMhHZGpGu!((ZP#b)oaU5$w&xkU_lmW}cmFgiV{`8x zx4oy{mxdj+`XGe_*EG9d{gHJ{fm`$=cLthQ9frDID^9 zGq3$LQ*}p{eXQt3lfU{(v)kBX&1ZL&nqlQ-X3nB%nVQ<``#C>Fe%gIi(7So2(#U6p^Qsl58^$nrlFM7+?qUrp{LFV|}y-n@C$?bPz{KcD!y_;@5sw>tv{*(_Naz6QJ&&*DK zx$?8(>31HRnP=F9+sp}{`{r+$o;K3`61rL=bvkS|9yU@mx7Hy7uWCgYw?I%UUL3?#nH~5L;khB zmtTKBvE;lrPc*me6)>Yeo9--MbGR)yb$%vp*wZt&`%2@K=VpKIJo?ZoQ&<~z4%q7< z=eJ9zI8&ORv`18??4*@n*{JidJ@mF`?8d_$vt#ydGMC5s3efWrH%{#vtV>iwE)I1cp#~HDrKk}zNhDQQltug)8_nVg= zs4^dak+kFXo^1~~IcZWYnb(_uJ=K*{BNoo67-+%3SV}_L~|Klfr>fDpvYG%E+ zuX9LuL*$szFBIR>_6PG*^>=36|3Sr<$7A(9kITMGh-^_RiLyNR&digt6d_cSq_iS! zByB1sTVzR+C1uGH(q4#mq-YbBQfXD9q)pNCdt9%#&+q%keck&!_uM(NojK>snVEab zsl;;;77Qyze%@2u{pSt-@R~ra6HMt4Nq(e$^_j#TRxP+0UTc@YT}UQVb@2U}i5UIK2 z(2JWeKCckxze)zr?S^>t(@{J&TZ8>#Fz+H17YNO&m8Z;zwLH5Tw)sER!oln)mjWbjXQCvG&3BrCgIo^n|lyJ^P% z&f-?(g`i*W45)}&fNRW>$sUAZN7{R^iqA#ibw2nbsS&4OJ3dP|NaKFxkZ5W;oqvA- zUaeVzYP=_$y|`m8gjdy{pdf?-N`Lr)q`WyEf2;{(g{0wp)m$ijGZ}_8%R!f80*>d! z;Ix)v4D|nrcArFH?}265^UZHScrk!O9k^Ai;|rxWKkDE2{HSy(P#+O?T@D$-GcR|K&R6- z%Hp)-etY@KiUpywlB`F8IxDrljJ4Zh#OMFe zw%ycXmCcXh?jeG6(kH5TY;+;+Q>etQNoTP^<1)>WJ5LK!E9iQx3mqrHzCd~rsrXA! ztBwdIPd$m3ujiq+T?|@!hH|{>^<2&F(@kc5rE-7?(t)j-sqDvj+7veEBX(FN;;e|J zklm0EfrfXWI{PO0m0W}qSEVt%a4g1oIpOMKt8mHb$pqG)u$^X z+)zdFZ~FY;Z3g_|8oK<0uF=ra=nSXC&O?A?8-nr;Y<{*3|4GTP9hF}ht$SLmck(Ip zDpulXQktNErxoKtIKl?@%OmU!_|3kssD>A+j&x+pd^%$3L78{Il5&F-|E1DU{<&|{ zX}Om%9aUGQ&{_@tH6>~Ouk}*=*sB|G=S54D%p8XuvQyyq=Rgp>BL#L(j-&rvX|As$ znfYjZZy6r-$%32p?YPP6DPEP8qc=@Hw3VjNk^*_M8SX=FzZa3B|5*C)BN{yoRJm{4 zwQQ*(NQ2D0&SK!QUc4Y)inRd&D05zn|HoFIuX$@ITdND91n`fCJ z!MJ`@rEja%Xvl>Rm^OYq72h36gCu`qk>6`}@Pc|!dA1l&&3u7E*=i(XGmd;*cX9LV z5`+GEc>e+&SlBfU%+0o&P1x|B_hNA|Z|aaxoV1k3>G!OvVRXUXoZjXb(5`^X1M8xq zV`ZR>sUS6G8f-i9jjfgBgJZ*RxN5T#QT;ULIJBYl@71uzu9bbO8o@sJWrt^1m1Bdo z7~Qy0hf!CI=(ERKZr+(1pSz1M^fSuaY@9SReU+Q_woF17}rO z!7h;>Y)45i`|a@w4oBnI1rYakm{7Fn1+(Y3rkJH|x1vK{4rCoXVgODb-{f*7!T8=h zHsF2{yZHQYTy$U-CW*RX#qWDq;H^kuw@gW1Uf{=HJ`6n>V~*!Qs7jb{ef&uTHZjE~D=ihBR%#Wjxig7o=Y8g~(TrpxR!r z-!2ZNv*)%^_Ujzxb&$RQiZEAA*CGLXYYrh>wd!tReyFu zO$z&UQZPFK_Cu%W6}WkI5zdZk06F_~Q2Hy6tmbF7A$~G^`?-meTOSRR>-zoteb46y zjlgZy#gMlx8eGQtz|%W#Szetigm2vr2O?!*q?8Ou1&s$1ZGv%&Ucs4JdKg`>5Y4^{ z;U?AHu-MrEzVZffunQ-Iz%A4L@R5BA!n|JAy5S}pms}3n&+B0NdXxzk-vt|^mV?cs zDya1C#)1qNcHZ~9Oet%}F1)G6!N^JZ4Mm>ISu2?;XnFqS|BOLn!!w|teh%J@%jepz zcj|(xo9@D@F%6Kg>?quP7Xs^RBnIfLMR{a8RnXzycwCe-n?2j>%Gwoqu>4b}pzQTI zXjC^vNhcLFPhX4&y)NUVJ6GWhFBf9Q2;;-C7T7w)2~X#HVtln7*zcMTKYK>v{v=Oq zyC68L_$<%iHe)^?^;FcT!uu=iSE&Y>1?B@dmmWeGT$K%OPSIfZ&kna5)xi$ka9C=( z3!XVk;+4R^P$lriF+Y=-lRm1fdAbDKI8vE{%N1y;jt*syI0MZU3D6KX15Ot`V1jGT zaJJy1P638!9>>o)S@_Rs5b5X0Qz%b}T-6Ss-mwBqaH_)Wt8!>xtASacgYn_1b$HG( z6a_Oc-ZM1C_ZOXU<#WRKHtNiY#e2oF#ylI?H+#PEM)$O9sHkuY_gr|5b94sg97Qk#idZG>Lw`|5ma|QdexxUDsybK#go8aBH zYp_%We|pWY=EXSudXVAg$BC1E^klRSX9$r3^yhFn=S$``IxgnV2vS%Vs(gR;s*OPq`xOt& zEq%X_TprHuwF_jAB_M>483s)jZXmP20`%Sr_V$LcxaxW!g-cYhV`dgWJH=v&7N1f4 zY|X(CKiR?EUrZLogBzRJ=pQ%OUmJKK7iCo*!a}(H%a~+T)JQf3QEyaV-@AJqxYr`J>83-QEA_Fw|p0a_R z6B6KkYY*gBwX;eSb0LjR4!~a{y$;gyI@n<{Zg}!#5dAYTU>a7QV4eiZu@VPQGNtww zY=_E6_~>z;_4D9im6HywaT7!JEHmu*<$+p>jyUUVD1O%7jb;hLI8o#w#IEc2D^~AM zVq0qaZNh&vPOZpeV+Ch2A$bSEE4-0ybCQQUp>rW>i3WS<;2jQc%_(i{4Z{ky+x95T z+BOuzf^UKM)1~-k+HuzK_a#=J%?C$m8#sRA6jTn0!j{qhnB$5!n4kGe*>n|geD1s& zHcCijW*Eoy^`9K2+!(IEqzWT8=71*}EFnlM7G{5c*>2WW z8EOU*Ry-fqG1S;BPnIt`P-jfdfSyY~bYWJ@N3(a|*0ZN;>!2wiLok;{0J~>7&g@A< zA?IS8y!IgmKYfmF<@Fe}SBE(rsc$0d4nQcRF zVA1Pyc&B_Xju6tsuY(W4Gs|gQpM^@QcnhV`)qey1u?nD3!!2l?_Ckyewm`WP!_h9} z9bQ}Z21^!~qisqV-oCDgr3V~9?|UscPw+<5i1lcgI-7KpZRoI)HhtfaM?cnNQ}vWY z`fT8Zqbv$g^J5Z5R-eO(In`+Ia{(W=HW~LOb()kn-ZYz&HG`BUT%ht08NQI44_++U zfu+y);_{d~X8T`%=H{7rgHCw6{xIlHQo)!T26#nH3cW}EFaA?P*%}q=GLo6y2P=3fnPTjs!@Y2*K@KlI+6)g4cEBYk zd0aQq55qs{!CEOpNYow&gMRsf+%P#{w$FG{1LwJt?!I(&KuefzhW+qU!Cc>A+*l~w62Pi|EQ8}^tI*itK2AAe zM(@&O$oHNHIh~Fo^PF0gQW{0Gzxa?>k~oTojK(c;mfW|GCuXq$ibilg>m1%2B1UU; z3`mC8h==Yz$GFk=@af=m9ISf|=UN;=_j-Z<{~FLBCl?YcktCP5t~4Xsiw+3lMov@? zqMbvB(ysi8v}>X{73<4U)22L_s+$e7Qi9>qYZbEdnoo_H5p=5d3kfj{zv-A4|M56E zPKX-~fuKL2%|GUL3$ucC5+N~7%@et)kEW{JXcX7OQ ztaD`_y%l1o{>@_c4)ehB{QelK+AE*A_kCk^;$cxN(mRA#1ipra*Qzk`iXkPx9!DD= zm!t1KGdw40%NDFKMvrm@jvw2#zB7E6eCCLX6LaiLzaQ{+*Eg;A4rzilxhH7@rSsEn}fJ{<9{J%nC#Bgz%(~^p;hYpWV(%LvC}GiIq?Ik=}#g} zlizTBx;B=Mdk$`YF0eg9KN;1=GG5fi1guWlgg-Ve#OfPLq+v9Xqs1mA7EkEx#?wP) zVq10xuFGskqgTZ!X}5?xS1hIIZW-Dbk%2?rPQbdCPeDhfh%6QSDdPDkTrXH7tMbD* zJnM9y;lI;UX=e61@~BxrneO}Pu>P3=``jn%z)SA%pmr_9wD%6sd1Ll4&c3Wp8Hu0o z3+!>%K}am`Lg%|$)bXyB*SI*6aa2rWyiU+KX%UjVj?7Hoo2&2M-H@!z&{XZDovI805RR+Q0}#wdw04@ zl)fIDN0UVF^TsdLV!E#l;qWcKu@jH(gmL z-Tj7bZyXCf!`Cu-5~{4Mqb|h?52WwV}8X`+aNE#w-A@nxC*YT^_>n zHL=v`1FA=VVfM_}&YrKG1MezA;p3t-G_Sple>eKDkymZOHvTOKYl8PE9D8smslHW$ z`6KdBGbDuW8XbiDg>T^5-0vJ8Ue3>B$0P?se40G_B&w3BHC163Ei1(PsWxOjxrho? z+;KzBb*}A?J-b=yyYlP~VF{+Sy8*W>If3uyRbl=dS^D@-ot*PW(ddgSP&BOvW8$3Y z;|L2f9J_-~f1OSN!xGWy@pn-8s{uW0i|~k|FK*^*KKN)nf6KlwSDJO1cgA8|yIKDd*t=GK1N!y3crts}RGgOl)2Cm#)2Vir zB0M}do!u`i$2xAFNWE(hQ;^AJvWhRKknfMlu(q8p+gD-1nI9-~s{^0N9HrUkB>3AC z3b42>4d!j{lGcsOtf zWr&@Eo#K-*xoi-pSGZ<1#=P1txk*o*eY;{WNB66RDYT0@Ovf$5DEGQEZ7l1=l)r1x zXOtCItP;nzGJOn}%pyL!l@7fdPO{GxD9#JWZw> z2mN>f82l^%%JK|gnSV3pSMtd3cPLejO{Dwf9W?BiIzKGw4-`+$hVdz{Va#qb%(?rR z zZY(>hN25k-Q>(2ksqHuiw}SmZ_=6=ZFTM$><+-43_yoRd*nv^o4mh~Z0R1Kl&eRR= zLvWo88n^nety!yB$=qX1eo+6Kvt^JhRFu@?`_}6?ul*V(3UYSRd^O;5mKos84N$mp z79_qj0jIGG;JLvZ7-w1t-KQII$KZ=Nc+f5kjLYWa+$XQZzvn@dnRDefsLTVICFH5YsrD&#Pwh$^Bfkrhe1t( z5lDu7W8|kTnmQK?!W`ulOEwCEM_Qt#y9P@51>iQvwV0a$7#`dSHi@T@e>?-hXEk1X702oI>uuv0 zgAN|Y|3rg8+>kI4&abk;{H1qrCVwMyV6QfNPi8Co)Mp`9>YT)yG0k|tZ5jrQ%tDu} z5)Pj}x&1VF#n2O&^oVdPBn_0%z zyB-VIHlX#|YnVIQ3&XyD!Y3wn@Uc=@wYnNf zQ5om9X5lj@6Wa5>1io1(L+aGMkP>`{{gv+w?@G*Yu%QMnP}IVEaUbCFt)n=0x-{vQ zyU?}y25iQIZsxe@5$2NMOzYe&M*@umdba)r9uOT$es9N;^X?<4vnm8n-|M$oef((( zQGv9xZ?MZf7j>>5K#^)Kd0-DV!!Oiw9B=e+GyHU;7ShX0y>K)(9(up5c$W z-Bd7P5az#l1iDe}pnPBN$;p}TpeLn%ku=yBf zVKur9x`5gmzW7hAk{eSmZ!bs93OA(Q1PCAWgIRLOnQbnPWIo+G2!?;1z<2xuP~3D6 z1B{z+%ieOVmuO~X^wtd6lpBAyf`gUtFR7Y>P_8yzD)A)vlc_(kx=d z=f|_cKiufsjO8?Et37!-EuF0U};8+zaYuJwWXFbd%-qh z2HZKE%I?~*gsObp$ziWK@#bBCv9l6jvxN$L8xn?EGeuD7mpK@=4QBuN3}ITz&CG;+ zKjICYE_C_z9v{D*iY4=g;D(Yh*!)NY+s|%;=J4C>n$AVo{L~Yjc1%X`r%jkq-YoD9 zcH*N+wWw%SgNKcaQRJd7>Eyk`t8dTYI%g>qneZKIpML@W=S$fkV@I&VM^9pzgk^YP zO(%h_G%Cl-cxJ{I9K?OmvUJ_P4IyN10-;&GOJG>%+WOT+9glk~|9 zQgZ%`Gkb61#D5VulU;yUOBUdvs+DMa<0NCXWeBrn=>^{F#)tts@@bVR)PElZJ#W^5 zn}!K19NEDPUHXNwFYm^`A^H@h&!gv06zEMh2;%q+X@+1-G#h=v`t$Sgi69hkkUYsuBokD3MR6zSu;N#tJ|#R*lC!^xZU z$gB_!b*8}4i&EhI>o7SqXj0dZ?RZtbl2JII!=4a2!d}m@#eK6U;_|@N=%cT}bhd0| z$}yi87c+qlz4D~EY0IfSb_KrcoQJu523J&Hd0KMnKIs*c|Og8G`t1- zN=BbbqlZw9Ulm?a*ocOqJ8^XFIV@Wffo{KSX?kS>t=oE-To+1^VL})h+#7`D!xeF& zAY95#aQ>a%;71oekD`&*rC9QI9`5xC$EGF4=#{gdQm^eJx%faz@e1Sgp*;5jytmtf zm@iDF59By|m^Um7U2hI1Gx7b5O|KOf1Kwvl7Y_ZyR+^`QMOQLwRG7(AykgD79SYzD z;T5wtdJ;T0Ukh^a*+-{5kJI&rL9}4IHHm!orW74Vjt`0YiBRVL3ePjas3`La>}nwn@cFB8JHMFns%@B>^^4JR$#TvC`+IAFI2h2^v3_le<@ zT?l8hezVW>oj_^!HFl(ZIxBkmItAVom@FRv#*aiPS-_{-I8nSj zVlsg=sjMxO#D@*$pZYDse|NHq>^5qHbbSW1Zg3j{ zc?k@pPUv#K1Eq^HVbM`zbbAuP_G%4fBdJlmxcH69CkY{;u?AxN9N(W*rd3TY_sp;` zLlugKc*1mp!6^N;8TReG2t!64WQPQWunrPE9G`krg|KD~aq_A8TEq^^Zec~XwZpsT zSGo7kUkGgG3?CX7r3gl3fr_HjC5q^?|o zys9#exAiZi_@ixQ`Es)F26)&f+ZzYtN#xOS+OKk)!j+F8|3!a%?d@kdurSq85QuyU zhi5mU@yq8N9FJyE(l%>Fh3HCbys#OgHr>YYQVpo3xsL8t#L|JOo9VvwEM8~)3~ui} zHJRayU03HvI{*1U{hG{-Wae9TFp^(>Gt5vv)@SG7#I6q*rI3O*!iC6ousu~-*N|;N zBSmb#MzU(Vs6cfprHe^Y+(|o-yxhS$uN=vmX#n5!MKⅈUrK{9wf2H(?AvnAwk8|K7zxSIp?hRb{>mUzIO$}GGVV*;|zWQKa{$Ru&TX2?nR_|=;DIaa}FHD)& z;h#mrI*-!9znKF#^tEj@e9b(JyvHotC??1~_<}V7(v-S-3_Y}pA&0xONyFKc)?5ms zuex)|Sz4dU+sE*)_>1wyrdQCugp0WTLNvy9&cG`Z3pLiZd9X_!m$Ad$gQ+8X4b9)@ z%Jrq2D+8`cQIMx2L&pTUZpR*eL)AB%>F0%+Bp1h!;nJmao8bPd39Cwclf$>ovA;NzbPV5lpzfP(n9J-74uA0N&r(wgNbHIo{{#YeC zlxtCuyfyg-3jB}{E3sYUET{i#g$^+f$F5;6G#c^!UmEkLH*4}UU%6uHQAxO(RLaSB zl2jp6AnnfTrfh{Zqg&a#dFx@R_Zz5pcnoeOc6dj(9n+mA&<*7v`ua1BwuF0;lOPtw zH|YxUij8ou%^5bR*${pIo*b}0BSeKwJsNinoS*jbJZL!&HFkW1q$N$D(fO0>yYGMX zta3&@{5U%`-3oFJ>S1Exb{HerKCxfFp{?0z9A}Y-vmlUz<-O^a+0*X(WF>utwtA+K z`zAxWd1M2^QZ&UXk6xL_kle!Qv^hJC9k$~vGiGlIv$5qI?fmqV zXiys^ExpLb3G(BPjqqX52-V@j9x2S|Ze{#8XfpFR95ku?GTv-BPnDAX{6^0|t@z7L zkRLoum{oX`%&Y#i6y{$|V6!HPuvuT!afW*aZrT0~C4L+rQ-u>$`Z%9PhisxRTUJt5 zmMzIU3F7^dCHb3giSd;@XW>)>HHLcxH16+sI(qX78r2@h*hefUb6ouc z3s-5QkFhA+e7YO;%S~{ni3V=jD#LFXI*4!Xaht|SPUnlCn#g~&MU#IZ>o%@fUW68# ztMHDsFV48>ff@f6aC8rnlrdK-BH`G~bzs>%l+%gs`@5N)E;D)SW#$fy*X0&b-1CDs4LteD5pXc1KkjzT z{=MvE+2yQ_bvJLrzG=*uz;s?|+d1C0nrG~)3*C(E+%_vpa!_ zwW#M+&AZKxj#|aMuajm*XG*{ctpjY3VKtLK_5<&q;!fW2SL=EH;xE|kL7Gg&0-yh* zpP!Z=#VcP;(J6f)+`g?1M+K|FA7SNOn_HUkY;k83Z)(mo5Xn5nwvNeV-tGv4`MiY? zk=l>fxOGO+%P`*42lZw-BMR8%t$eQCWUui6-EQo|v>r3Xh9HnByUqQt)_Hbd z49-b#WmB^C8PAXRd2@@S*)y{J_%wJlv1;D~*tlt92HN)Nz{>2W9E`r_zH5Al3cFNt zJ@ZWY0~}d<7<|(1usuN{Oksluhu8b*Y4Gb%f4!7p^Nk=ObQ~MBx|;c*n$3&$%i=XH zGBA@pb`thq7Quvxx@h~@38s9G2ltNtoTd1KWnBO9QX|krCmdDpr3vCvkHWgdSPl7~Jz#lTFLHcx`KQQh@#v2?t$%iar{aDc*B5+3eqb%;-0~fuM_bzr9zCw$ z;H3XE;qEz}LPB*Aj>d7qyRfB5c&jSX_%TlfI*mjs*@gG6eItXIPhAHXyw- zjFp+)$GuC4kB&*Xu<8c_>}X@V;wBBeTUE7`-86MS`p-ClzCP&#W4!0iYBTx! z3iKjlIQ0ydqD7tAC?%=Ey*K^#7@j!phY0t%Fv~%k%HKIrn~4>Dof`>i;}c-cqDc@i z>jXYN@BjngBnrI}qmsvWac2E!lv}w4B9)8a%1ohwzTU}ff-}35Q6zmI-n}^=W#{aI zRJCv5n<4EyCzu#`xg*2au56&HCH?Wnu+cOIGK?@DQ}a@OSBK?)MZyeTEtwyR{5s z%fw;lroT*@Vk+~irQhFaR}=v+4X=UAx=dKScQdG8I)QOr1voEu8!q^`1D4JRgxR}{ zAbVvErjKq#Lu(nz(HEmP;x^_ZlbSp8IC^k3-3RcrKS~{v~toRIMC9c zL-F5u3%NfE6@PAl62;LlNwS?Sw@yXx19S1m{CcRHIhtJ{CdBE*(S66zU%U>xI*+3D zttk9HYzKaiYKECNQ{ZB64pfx|fvxKV4EeVckJulaG#O-vJM+;PomiVTufQ;4Qm{} z<1ORkcp%>px0L>dOKyVNK|4L@lh0~`F}C!3))K6Ie*l*W=i?#ULR7XthMB=(IJirf z9K4)p#&?#68O=bg@sqLX`h5_q{DTLN2>b|b&J<>I3D=&FN1u7F=oU5tPjnb#x41mI zt=)y9A9GMy;In1;^6c}nnH(Ps_$&?9FreBvC8AIvOwU*hUtTSN?+R-f-dYz%_=zEN zLqQThbxuVgl|*!3sgK`$_HsHH|ILEap~rW!*wNSJD0!(8-JT{ygVa=TQr!yfU16IX z8YjMhNy0+d6Vbp3Z8^!Q1n? z8hT+ugDAHAIS11=4MD%?TEOfzAUx?L)?aQE)L43tO{X*83bawq zM+k+zdH7pV3R8wWg3@1aVf>}Xu%t+k8xp@B*Gv+_>_AD5-mjiRs6h8M%HIBq>S13| z(&!GR_#DR5V~errPrv^&b7KHAv(=9$=gLQe-Jw_!+#j3t=y)E;WICh#w}Z*^W~Af;{~7;ix`wKIY%o!4n6=VU9^5 zyj^`5x_6DCcWoo7DM*BhjB}~s^Kq)l$fQCsBlJjj#F{Hkc)2MXhaWD*RnKlC&pd@` zygq?3x-^A1$^(c3YVhh#ee{k{WWT-EhAl4yw&7$LldUMl>6n7ZacJFlR*-}KlFbj6 zK;E!79Ix-}Oh%nkAvnMF6C922!icX2Fs5c5?%a?K(eCRY#5Wv!$325n_7$v7)esC{ zsLI*vQ$OlCxP5v9VpDL>cy-+AYzOMye$aA1inXZ`hx8O}XipI2!VFS@j!jjpm-z_n z_U{4ZS2BY0*<75J>%qz4efAhuYpNt{*6r6<9n<%5s6s58sE;@8n^N7mrq~|AEN`iI`n3iF0N1q2=pASax6$ z$D^2k7EJ4{{)xd2J~Oz~58B6ezz!|ZM)e585|Ki6pD zQ`4K!VuW1#K3<>gnu=c?9pHIoJi1NO#Sw=t;Ns>`G^`0E`6rGv@7N@IKWhX9Y)C|( zqY?NcM3xL1RB5qTG&%(yz|H%8vGKSPdYq49mY(U?4e7{uJhrYEz9x6UxMMji`|%n? z7u`Xva!F>aWf$Xk>G!}KJ+kN#W1VP2-ZN*LIs#p$dQ}&r% zbZI%7EqnpKzV#f>a&-nFzkN4ud>~0f-^yTKtP*H_T!mA!`+Xggcgesd!J0c%Q12~X zvA>pHs@!&1Iqfz$)mFmKQ3+6;FbXPEu0poWZ*)Ez%WOPn$+K4#Wkmhuxjr5=8*+Cu zpCoAg8HM)?^>KvC7!;kafKSg{1*s$ltPrdjdj1Ab%Hw%7*K9J`d{D#Fwd+BFFAgPL zd-2Ab3=YO%tu6RE3h=VdNbLIJ#=d-_0OMEO2FsQm5Ot`(CSJUK|9Ril@2>2N(o>N4 zd=3OFodb1@#2o@VlqKzkhn&M9?_LV1@@AmRrjKy>Z9i>CiX;R+H^CW$`ZqjfUx1nB zzXjiA!BV4c2+B3ZU(+iw|MYY$-CYbyr!T_NyALrpVz?m3DjO1t?XXELmBZ8IUJ5fP zc(cfi&6VtZ(?g`~qE5ff`{$(p)&a`hVN9@;FB3V~fmt$tB>ISkVdyZk)8zZ^rZvLL9fR)Tru1=#JBhW0~a!0Fpzc3aqN_NG}h)Qpc|^(Xvf3$&_u zHq8$hdAIe90^f)^wy2nKX_IAVKgz)kaV4l|SdKam%=G(kkf>9K%}cytf68bM=kFE~ zkfJsPLe{mj7j%YDWsfRFej7rUD!ZB7e1=U3vS&Z%Z=>@SJZgK@jz?rh^504h<@e@& zpx$Ib?X@Cd`je|hL9^YcQIJdiP2Q3^1ohYcqgRUL_>!TQ=;Xm!v`K3zIK*z@`u*>H z-~G_iyS!=hs(F`FgqSG~Ln%_nvWrCI%u>RkTtpkV*_>Db#Elkz}eS6JVf26fv#i? zjQ*pEuBq_@G+i4}hTjBrSu*ne!7`~lvX=9wslO(Y=F4Zat@4zJvlt32)d68sl|-Sk}aDycd5@BjMxHK~j-_01k`Qk{`x zEIHm5G<|=uXD03fqmzSRVwyd4Ro6pj#!L$TzLZ{=OeZgqINb0f5k;@g$GR;Yup#sm zD_`fy#xqBt{+TfJ8V!Ml?;p(ehRo!ROC8JeTYR6x;o?+j&EW0PsC-cgHRspCaOFH^ zgU~Og{ zPI_Q_ku1Y3AY9BGrhiw3`t|*}=zTn?4l3kjwG1*xGWO^HnEo{qoPAx_!0!pNxVNAye*|+F_#(nf`=+TFSluAhac^wYVG{Rr6mtsQ9 zPO>SBA^X-4@_Rmv?8HRqN=ya5KY0%0^+a$?QyuH^%m*VaC*qpqB+y@ak@wE{k6F*d zBWCk6uCR5x`|~>-ti>_ZtP?HjqP#FV8Vh!8SSicP#>-f zJ$IaxybhY6UgBBYlpsx6Ze_eKK)=F7Pc|)?wR&92~tp7B~Jmj_1FwDM$6P!4I-v{d`ee_+^klvepVLr=t?<7jIJatL5`eU<{DBAj-q>MERGrW8+YkSllAvq zbnu!4-*1`<|Lv+=sOXerHw?SYyF2_Psrl_zESPp@g z>m{_r!j5)J$dbg-Cj30?7sn&9u_3s@Sq2+!YU5D;YRbGBN&C+`QSUrG+P-fI>D5o8 z4EZwDS$`Al2FGH=(@*puF_LU7JTND=KR&yU&taCacumiT);uYu*_*R%Iw6I<8#1SVxgoG2O(7WyqdwT6Y7G5;Nnj-+Nt*)@$UkFUc z??lJ(Vl<-Efy$!pq3TIJnlf$~mA3@|j2(eL1pbqNl|ym<*%};HG?6y^jt8F;>p9vk zFIJ`4aY2OhvT5Iz#ZV@>2=Q+y!xapVvNx%THx6Bj5wn$Bn?FC1+9NZYRfA zQvaXNKxercx^#HMgPl)dmFGP$t^Qx#Y5gnHk#Cue>vx~T+@?|-7P1EoiaqdIs0>bf zF%{dhb@7|pa17oRj#3Ia*t78kj%ol>_-aPmCWk}OD}T76@(MOyb`iu}?OY1~Xf&aJA47%)a&> z)vAx;-{?@B|J73vdtgK>Uc5$Y``b{XBhJYJ)_;bN3aepou`B!#>|K6XuK@ciVURS{ zlG&Z5$IR;yWA6MYfZTm<0RJYkE{T5JcYR}F-4Gwx>oNk2jC$D-rxAvyMZphse~>@N z;FWeKTxsiwxwXav?L~8TG1vFCG0&6|*+yR}Jdo?az4JfiM*C(085Bs9ZTWWEw8WX_ zv4e>xGj`zH%r<>^|D>3GK0BZJb18x>mrbCdk8WYzBM;V1btJ5}IR%;iPg#AJY>p?s z%_A7oE$4aDU*?&;*e*H%r*_0d#z1E{Q}XaO6Mdu^Ec_g?%_SaVS|6jZ^%Xpp5s9N0 z^iWlipvLpFJj&k?ff5;m@Ok?}c;eqh1rv|cA3^PJ+4~8Qvo{{nas_da1uNO_zFJ(r zeSG=vdR?c5+VAee@-dcRlOT-;*B*kCHiN+BN-eJUJA>nX#i4JLBv$%`KN z+3e7;yUZK=uQ*KV2`XCDqSE1!l;|>s9t-l}S5MwZJ5TJTjkosGkJZxX*t-tB#@#{v zfMyIoeham2+tKM@EN-}Z8khUkV2VGVJ@#S;J9$|kyCtR$$L_g`iOn^*Gr$VB>g~r) z3n2<+tznIEI=oZJhvZQ`aN*TwL2h#rWDj4AFDu7m<_a;q{q+NzdOj42LL5WiwAFjW%)1B%5fyMPGxD9g)+N$=X^GMraG$` zq=>zn4x{C+aTLEH6_*YdA`xv(l8IKpg2{t0V{J2Jz5RnylRHo=HV;>+=F>_+ZJ7D4 zg1St@HSt8nE>1?jk44bO*+nFL=K^)h-2hh^X*3G*;oeCn^~VXF`nVO$1hJ-hBL-9E z>Ivla)snt!;M0wk3|t=i7=wE{1@V0U@WHg2=@nS9t{Y9|M$xZ1NM^2vG;8c0^jM{cLUxDY<-C4B zfyHkN3@qcp$|tRCh1_WHuu^ADtq(9V^GsNCy9Ou_=)>pEow$lu&7O5T%?cNZK;8{Y zP!Jmm8^iyX4>M!uQ`Y8?3`Xph#&%WY@c(Z-wk=-822D$5TY`A-Kt5Y=P7%PpSJGZY z`Ib{@`3G%MNRP+)Pc_ldV-rR!4IQu>2Rg2>7Ya++P}^jfXBSNdE&C{~V>dZe7D1EQ z2}pf)lxzEV<6U-}tPThFSL#lt`KJt9AtMb!nTF7io(C`1W-_}qg4sel2K_#L1P2!( zT(n^*zP9$m*(U_C1%}gc!>fIC#e6TlIk|`8_6lMG{}gfX?#+y&;v0G7wd!6K-)73xw04>P*=J|$r|`j45>Y5*ZKJekp=_myloBbuW?n@_mSj() zY*|Z)>?GfLO`o^V_xtWO^E^35|LoSJI`DPNSL%?512+LiuC#RoGd7t4CBrF z;x_9nC|I-<{PJ$#hWdNhvbhrPl)Q$w8!y3-@0X!9hkAT?%?%slqZXHB@4Q(2zZW zhj`5F!#krecg$7noSci%2RCBI4L)|m2s^x(^b1zW-Xy#4y#=kB6HrME;J^)E`n1kk zf&U35N>EnR1OANb1#T`HOwu_`=F@FWX8nK!Vy%0M_B;2AJ^>@)t%T2y&AZ zA@7D65Ak~&2PIfCyI(spHmhcm+Oos+&=P)6eaI^`_YpBeCOyTZ`H75QzXQxOW<9ey zQW4efd1KnyLnvEn$Lt$6pXt#wgQ@l#3AeYDkX811RPU`valNk`bIuN#z#3=xV0DT- z*8D-Q{%nPV^LYQkmrCT#+Gc8l>KamDpzJZ{79@36ycH<%UFLs6R%!jua^(5}7|z?I&Nm3-&(?XH zf~(tvv0>KIon)y|lc3+#x0IRpcl$Bt>g1TMDx>h0>MwBMF{cMRHp0eXO4vX51(f`f zXJUpt#U~j@ut|Fbs-z6Vm(M=Jhw-C9S#c1|H7bG7(NFQ=z_I8tdkX|uOn}&m$c z4d+hDGNoDrnN@14f=%6^_ae-Z91`?!9KUa7uzD^ggz^}pjd2j^&)W~qSxHCEdLoPu zyBh;Aa>pldf70zED%Gj8r{!pcdye*-;`Hrin#xceXHE!xF1?rRNdFuEQ$AmU8-*?~ zYpOZSkcBdAw$Wt%|l3Izn(DY)Ci!{ognM` zHb{*(hq?RR$kM52Da(F=^q8~ZZp7pW@xcjg6p^Op$y7NUal}r$+8dCE+de!1mBNa=OoRTJ88raw^=7)Qw7_ozp@| zdE!!m?;#sC$O-Qo^!V=+>FahEg}lHyt+eO5gJ5cy0QMTjP_p|KOuc3#=xy*{BU~J{ zA6(4e5jk%fYb$$U`M34(YMLQTTK%596^$gp+pQ@7mTA;&$6!jdJA-`OjT;xyl|!ymy>c^!dKVurqR;gXq2)PQIx>3z$)9CJ?_GF~*0#M$ zt}gjZA9jx-7e{JCNar0guPhRt>(3Qbp5 zo$-nxN8mj<9^wJoL5JvQ*K0z3mj`9iAeEe$He`Q{A`m`qOhRcC99He)U@93$b(XtOC-Ndm zvU)$zpXUktmDB+$E#Oz?88TU0LhoD_K(18%B8O(|Bl@R0$*Rbn&?$3-;4L>nm&W(k zB&jupWP6?^T&o%`@GX5erw{DL#ml(-h_+lXjQ%#ml|IPBDi4{on4FJP=J8SHK;-lR zWPDK)5qI#u1*5HaoQ$5Z;^b zCy`4q6qa47g+;ILLhNOmf8&6(?z9Ex$fS<3keNFUV(X zqbOGmbqF)MOYSe!BujJk;CG7xEPbs6dnUh!#{PP!G{*%kX6yjt85~Sne-?`RrGwR} z-B7pUGZ}eO0aCeZMCqwB3EHPl?b?)?CuFBCDe&zq0o1{9WP)5Jb6McT}=&>+_$SvMSb;^$w?#sC<(_a!+2-gwk{lU78 z-yc!xhurxQ=oK4=J6~mV&jb7B{*zaH#BDOaYXeNa*G1>dRVBA=7SiR)ThXxb3_iNt zfG&z<4i>zPyi~5*19!p%-Hq@wJd<4X?n#wz%yY0;ylIz~G8UZTy&)+j0dm>@r9HcF z9DVgvU;2p735NsJ5!k);I!s&1?>j~4;DjN@xG=&B-!@+bneAWTm%?Ajojnp458Qyv z&4W0kc^HxGRpl`Y66mU-GUOs}Lpv~MmBV%0NP+h|tushoN)CJr;C*c_y(E%v3*pJg zIPka^2O;&NAx&wF@J_tD`xEf}KqMua*k{dU;@uoYelKL{m9?!fa7j${$F1jMHfbQd1P2IK{Ms<3} z+h+u+GFwxIFrNZOFdH_=;tiE%XfcU_jV<@7j+n!g`KP^9Nb-1EM|4H#qgmxAh|jn~ zB1LZl0lpz`|U4StgcjDVfXO!t0i6*FkPfr9u)5S=5Qu2iixHyl# zvRskQkIek18~^bq{pCF@Y0o~TWJJbkGQRd4dAD~9-4SL;?lPq$Eh++5)b0YA+xuX; zxeuMoT&AbJSA&KXFBGKDzF75znyF%oL$^j@x>qUQF>J)J-7?JmzI~aATN>faq8(6n zEJu*rnRGeue%6nVcd?zvpIJ>t+6I!aGiJ;@hGE|P1?I@a{p4nA7^&~;PQF-bAYE&Z z8~OR%qL~*c4;gnV#Y2zUq|W0oADj$rcFpj4dI5Pds~fwh@A-3Nv9T(wwp51IIxR3y z94@Z^p$s4LorbN6U5Up81X$3NOLMTIknZTogoclR^# zDDTV1!5Bp7x%vObr`&C$$;w-e|Nfso_7kWKbHdQ^Avkkt3AsN0C+YWR4HP7YU_k0h ztl)jgM@&eeBYGUBdu-Am*}peHSB^D|oA8iyO&8O7OS{kV+w<%hX6QU-^us_VwCOgs zkK=I~9R@QRSBJv)83EuMQw*bc`>X6rWBGI7JrGvhK~1cFD)ixkiN}cdvQOmev?8}~ z`oGKpQ%sQ6bvJg zd6wkLtr~iA=^$br=1uG`+@b?`oU(>I5fsRHL!WUAVEL3}Sa36l3_n~%dOb0NL`@%B z=e<9zo4SvjeaK_2fj=>L$J@<~2SQK2M1BoQ5#(^p;2p7EWGB4$$AdzkGEWZ6c$;TG z%V26x*?a1%`2+gX=w@PVeikBQ_7mfEhhVSa2QbNUf-9w^cyOAK0g-=b z4UrDsnEMB`#$5u(r~Sx*sbk2!J>Q76clWorucVS5;>+VLR1}jwyI&ICUP{oDt_NAP ztBwsl`jR*K{x}J&dw!zZmde7s({1qEHxX zog~j(eL?1ME>-+^H>G)U30+;7PWH`hBUX};@SxXOa@ErSjupGXTAw{IqoI4BX82-7 zoPS3X-x96olY;*x5kd2=6EIRHJtQ%k1|kzyZh_oHkpA?Rf+sqeJD)($B_OWXp+=oN;qdvolCbs| zpV|j40|QavPbzVik%y&2JmK}DD$+631Z>W#0TY)Dx%__dM&)^==4}A&-rC-yXCoqhM^;PijNqd57M|jylx(M1X2R3~c+c z57dtr5FOoHbaCKK`r@wz;Pm?;V_a5{o9fDW%Cu0Aj@{UJe6GUp*Z1fulltbqF$O9%qId42PhBTDR}jxgMdSuf=Bc(fv=USSFt`^Gcv} z&sbqET}_80H3h;gvKSBB~fcn#Qdl z?mTYQ&|~#vg~oOga6$*%+Rl?riie4bT@tq5JAm>B)9~8(wPeiD*L00zH9cbXF|=Np zgL^OUMg5&hjNdOMMrngGvsL~qE*NT~k&=zLN$@*%z4 zpaHF2o@3u{?{J)QCQ%Q4O7w>OBu_5oLgVJ%u%d4l-G`5#up{`j!=Ul9RB`b^jL12J zb9(N;E73_%#DDXo@Az`RNL()#L=l<7e}+5G*3GMM$71CqRFus(zLA8$JcTIL)+2`<4&by>J&qd5+qr-Ru(uR;IVL~LGbgUMd1 z*fMnkK3y>p4<5P?XI3erwcZ`@`C9?XMK!3u^gQPGE5V*6pYS2`4O_1MK>v#~QD5H? zm!#X^w93cuYW*wNzx4-PKXRN}+o#E){`hYy^I#(yn{ar?HtFAd>DA_|!ne&uKF!|O z>YRgn)?8xnsF1wl?Wn`!ox%FYI`I6k8;pY8@zfhDeD2i+%_|LY+Uvg9t5OBawYqJu zJmxK;`=JtD z%qS8tw4Oe?=rt8Lo8Lc8t$+(X713W}|Bq+eWAiE6O^hRS3$XI;9$fbQA&zOT!_Apu zydmN7h57RpY5#6pE5X$3mH1%Ad30KPjdJbWF7Wd8W+SqP&!STA1bnedn;h2bKFiBE z?m*)AJfLhDFZxR3G5am%I~<-&RV(<1r|K=n!^%l_!?@Xj)j+D@E z3QvQK6!V*p&l0W0`+SJ(&?WI3$aY!awh$xCyeWrI{#M|^_$oB<%Es|sPYH8K4zv|` z8*`plA00DFH?x>F8Ly0O9!l6DRz}wkt4RBz0YtMUiLk#9!P9r!!PRUtY?&s5wcb5( z>RcV1>Uy50~KV`7qv|bpmYOy$BvfC-C#yB=8s!Nz|WZ2)ZtL-8yc|OMP_%BbENXAN2aS9@;2c(o zSPg0xz6FL+0KGLi@Nb&LvCdd zjaf0|<>dhKcug^>H#Q>6+pS2`C_lKgRuB3Qx<(H8E~H1ibD@>kRw^U54{_k**j(^z zr*ED^czsMCjt!Fqr57?p!SFGCaJ>eZqL)ZokHwRDy$+B*aEvVMx02Z3IZr|syF<_K zCXk=p9iL0uzBN3y{ot{(@F^jWJbAcEx_L^ESVhiB~fYq#LTc5zFK91L%>$pQr2x#+w+(L ziJI9#W?tsVQH?{yYN|Y}Q#eZQO{gVBf70N|or$o*vz?^rb=&ky+xX%s;&clmES#u; z_w&!f_Km^te4;!w&95YF%~j;cjBZT&0`XGnM|28Rku;mh({f~Ra0{-P8vw1D+QfBT z9ChXK9P0L?k5sQSOQ}y6LPyb5pMak7nM8fVdaV>PHzs+A6QUqudTQB`!f?mO!5s~f_1 z)U?D9ye=*V-|$5wZ}c8IbI&2V@r?rAC!vUrnIJ>f&FTSle=n2Zm&ZxN-zcGf97qQV zx#{Cz5_sB9+V)$mz8JH6D%SlPhzVhL~j8_=210aHcrrI{<D`}7l69uy<%{6vVt4ZX-Earxfh!$k zzkIQGdjH05OD`Id<2kF?Mc(J?y+pwA2U*LzMOOvnZeW*94tR2k6Xun z!BuI`@I^uQIb%YIJKBwV3TqzJV~cz;VdHT3~z4 z65eJmKx+wlkZLu7lfdJ zO}GDoMa(D2f6xSdiV5|##fUc3`b?cSuY#pA9>A?!4zC=$h?{W&nOOLl=0URQpZjts zf3J^}Bufu!T4JDK+B?wdV+^6Iv!HTcC&<@7gAw78@U)ibU7%Dw6{0L|gF63xAO5@B z-&Fb@$(%k<+P5DhqpwGi^-47)IPi+lPi<}y@SDeQ(Hh{5_ZqB7#wlJ_`nE*P$O+$- zjYn4vOAHxWiUZv8(aa|tANa3^pH&+n`|>r|ATt+xsHGw6QH4252_$`YAF{7<1#Nsb z5R+X7V$+Ksu+EvsbELZ0F4YP=-sr+~@cQip{;?61ed8I2K7$w7*YWtk<Q28zV8tWH{p1{X**)<*x zTN0dWAcn;@+VE|=8H81MgM3sfoIW+1j@i`i@Jm~bdLB~^F9-F;m;DUUDQ+!jSaL9~ zNF5i-yoM$6KR|t31VrC`2!UC@RM8IW4}5~z zx(e8Xk7eKjKgqts!yqGIEQIGKp-y!uR`Hm~Gc@>kz{6K#zdtd!);%43IvOFv;}T48 zQ2^^(N5SxCBm64)U%bkmD-=;{7dX&1M6@oDzN==XGaP6@-c;N$94O=xz-*dCeyopQOq}X$8YfSIY)+tC6k4h%!x&$ zEtRus1l8BF++kv_DM@-TK+vV{4r(}TY$*i#EQhcWJnjvD=4GT!NTHD@_33B#H{tZ5 z|6TL1ozw=WGm0|qLP^b+VU}rf|GnP z+}9Qdr5*&?Pn+)kbT9-;0FZtE5%NcH?YG=}J?5 zt1b0swzos#Vqe?Jv)b5Nl?y%YEQRbV^935xGQBf)(-8+!1b+68Jwfafte~gsWVld% z7qSXEz<>4~C^MG@hpoB%_sxs!=oG>GUJYdNDjmQeABgi-g}6}}WWtuG)Sdy|I2K!D zSMk_sS!Dmqeq>Nr4GfJ`2ZiU|Ix4hzO9|ceVjMnws*djaE<^s2LGWmXIx=^*NYql93H_KM5V*30j2)&< z-sT0W$)Vm(q~ao$V~?~vxc zQLr&vmZTRP<$dte=_wy7=@!%Xw2Is^GT9r*`xm)%K)4RQ)a?)XX>AIc$E!)%p%wIt z73xH3Pz))4m;iaDpM`p+?Y)TeWDcOF%u1ZEr+~wqFM&?mLI`U#6Yi%D^(U>JCSkv&5z_VZsq2=XsB4igshkV_iNgb9Vq_OVHq4ntRX(hD82@*f z!yx@|erz2N-y~(At-g<(+{r-by?P#p>L%56Wf0w?WG7Wy-w46H&(n7<8%|$#$M1T7 zz>k;VHM>M&a|MZcoi16rARbq=(U=%=2hLooq33MzB>O&!Aad$lK_|84>!~$olIYpT zc?`y+I4}$t4>4`eVQecOFV1&1m{@-j+9Q>%?BUJ?pm5{G{@e>}aqi%ZimD zV_Yt1%iEKmUfCpiQzcZ48wJ(7Z9w0yopdsh_~XD(%(iTYK{HoC;tW@aUr9s6i62nq z_zn67+e59-a4>LO0J{`sz}w|+l$pkG2bq;})Oj;YX62HB%%)x1jC}euh_7`57#Izz z#jC01^1CUW8T07ln@52{fGKZ-Si##k?|`f^)-ac;B|-KZv2C#n{&jG``%E90`qB}U zrXGL_)2Fa&Zw1(OHG#HhIXvXtp~^`Q4*uCsE5CN6`>VaC)Qb7oM{PBv>Ig+*G%tff z@>zKQAp~xFgp;z@f#h~T2+8)FBFJV`(=&R#c_5|yAoAH}}u4B$n4t2j*lHDD1i0;Y)`tb8|@Jzh~KdRfH_iqucSRhY# zeXXRzjqLD!DheV!%={BLZWLPEF?nCDW1 zdp_jiwNFb>EoMG8{s}>m{X-mavj|K1eIdo2W{~%zlGJ(x3eUM0(t-}nlkJN|m3nwT zLmgiaPJ`*aR{#$PPj+}eg{A5jpss-54_f*Z#>JJ7{1}x3pQ~)zO3s2Q&Zii z4wXEDsl_-c46}G_N~FoOtZU*(1B?pt=UTnBBdFvHRC4Q*@c6 zjaj$e!JYO)#5yF0n0&rL|Ab6>VD2&ca@bOOQB^iQuuGNr^D%N7{&bLGAB^FB-5yZM z>WOb3eTQATPhmmXU6}1v0j4n*;PFdBCNy`S6-nFwB2q(Go38rZ7YASZ0zH06ATz&{ zT%WA~W$R2~XKWkkI2c7twtKCxx37WsfgpQJi^UE>V`cv7O3@or-1WyA&+tBVmxhfXHT!l@ zRn`ap@o98x3H>myosKCOMNEGmrB`p&#Et2yJkIk7d^Ye0J^al|VZLE~Boa-`k79ao zHfnw^#^JJcxa8+!oEf(S-(}jM*JKBr6TBbY*VpiPXFrAYGROC|R2G9T$T4dh6_~~M zJ8_g^xBZATf8y&KpvA}J`=lu%n!V!eTlp=h@k!?Pt37P-NU{TZ=vZP_>v`yF+l@>2 zM)f28BWRJp&w??tVCTDVsOi`QErGMJ_85bQ-Wy{2*PdYJ*BwLZyEp`c#|2t+-$y&d$GzrbZb*4>pJ#mo?o!G;53A*P6tKJS7dx^Fg_}cK+P-9;mwgMup2TT zQp<(_`z#XkR_{bkVJ~*}EXC{@hcQ!cFNRI~15`~hgbb>K2&ZIW98h|Ggt)l%pk~_3 z3vK+Y>JOv-+!F2?Y(7g5h~r~EcozzM&(ij{m!Y>gq!#-SGR#>hOsE?fJCh<^jaOW;R%_DH8ap+egEznRkS1uOScMO_PtH zuOC<1gUiqEb6}{7Ax4gA#NbONIa4){k6m&}D8FEhEWGox=VOLhgUoCd63E4q$QQB1 z&61|q7bel8be>Sh0;ZDXY2AM1%WwJ8x2*?}`UQdH(U9pdC_4zshWJB%GY}=4Zhy8T z{G8*>#SL&F=N!2|?IW?zT})JSo>Hqr_JezBcT9kC*9NlcrI`Fy?cN7yYPv)v8Yei| z+fO8e*7gv1yY8e%j`4oLmxoRxN{3}&yXy~P+DA+pPbq^xe?}_p3yIi*Fqa*OUB&{c z_o;e{Ur@y408q+$n;EF5UCfq;aF+@7tF`c{Qao zaOb-fOlgiM&J2QFNe$`odbMybHeZfT@ThW_K3SocMfvv)q9l<>nx;>dl@p12_2U2g z|NQBRh-lFo-(~*m77z7X6uf9CI#3SwcJ`DVAP_reCwr0C|NlR}GSSK61Dv_VLdmZE zx!h}CLr$6}O?AC=CB9d8NF2Upa7Fw3i=}y|x}rIsmhqg*fM}#Njv|xY1=4)qJZJHk zV=l;Fhi%K;jbU!_aOKcQ?&wh`$^0TSjtQ$}#Vzl-ao;pK1B)rra-yRPk^Nv=DmnRC zA~7sqwSftg;%*@6*K67y^quJ3sHar#7gTaE>32z{LI3J&LDg-^MZJ1 z@LXJ;W+##BpTZ4yUnP`vCkNOQS#tz>T0LXoi75vSA2M032n%?W_!lcj596NEGbLe8 z!!g*(k@>Cd3k!WebIkG^T#MO0E^vAei3T@LBCYSkr6|dVJFG-EYJ`N|UnbdGAm(U; ztC*TUos-^oPHcdAI@8fJbObb-m_Ub20JkA#j^x|JYWS9`&eeWi!Zp^0v5O|`;8ufn z*|l9>=pSamqyWr8k9^Qxu= zupYn1aHjLla2s=S#EI_{xiQOKxWQL1aKHPH<-TMjaaYc6ht$$^3=EvX)eOt!mYts= znc*SBZQI!&`s~;tm7|CqK(Oubb%CyYNjlDrEJc~VBI*BO($tz}fcJ}c;{NZpT5Vp!e7~CrB8~5CzFvsHXZ3XzN!OdZ5|=MTVxQ$Gm9t2ij9=wSww;ZX zEcbuGy>>q%l=ZUR0?FoasFx^*`&>t3nx+Y}QTYJCR|RgNe=EE$F*I!?LYR z%5Ru#moImzD3rUAHV93x4#W9V?t}H0*;1Ne|E1gvTlR+5bM|ku8XCWS$+rGk%x0?J z;qI%ba@)5L;mo^sa3>uTCEU<3&PV=G&wJ!aGwbw}HeD zKgX3VKgfNPTPf(bls9D^U6zU-CAqZz*uT8bQAb(TJUfyiH<(-TF`7Fp7XkTJih}%Q z-HN!BbSyivl{pH(ZuJ@s#3X#&e#P=zP@=aDfi*AM%CwH zD!)#vIWry5MZ(5~RZC(A*095(^SO_D>rnAtxzGo)`2)F%4;Q(OYaRf~32IzQVn^*vTEN@eubhS7e(jBP8+@^d){; zoAC7C-pm-M{D03$pZ&j6l*tpJ?yu?Tk{iV%xV5e`IrG)QTxRGgA%8{V6YkvwBWO=c z1-WNa$+kKD%gWB!l-ZOPN*=Ld$&>cel7S9;;petAj4Hmz{TietNvs(BufEpMNOoGo zZ1Ja=2T;$>n6nycDmk2~%8gzZ$pvi)#L&D~KylHM7Yj1jOJzB%q3#Qe)sz?d&7)Tx z+0}mxbLY)oDZQ{lGZcP*%?xG}4I{V#6Prj_RTK`_&BX!sx~OgW7)Q0Vu)ZOe|MjQz zxjA1R3C~HddpMlK+>`y79|}b{E#o%l*XR} zk;6*P-;wsSX!M!{PO-j@dw&HbvjuxOktelwC!HjNi&w6 z$Y=`fRlM^{eEaNdjvSNc-b^>aEk|vI{6%lKa-S`LJ9FNBMdb+9Ao>5Fpv*)W9u>e$7ewOVU_?vIkZ(iWg6@5UwvdUS!0~ z*pe!hjYx686RtVjm&+L21WIp%ggj}!G(~zSGO>U6LPV!7zB}*q@7~8l(c<%Z&FsAc zYUsW*7T=EPrse3WB5_qdATbX8%-!<%0rqo-NF*W)NteW19K8J%m($AvbHxv(bi)2i z_ojCeU!RFWJ6|=Ga{<)@CFqeV8FN2V@;zBBdFdH1$o$fXcVe%sEG~F{C%ZX&GV5}q zf&Gt9>2J!_tEdu|ha1nuvRCU3aPZMl64vWGw{qVhF2D3SPCVNSZ{C|Iedb@?XUFWs zqg;!0&LQMgI!{A44L5GTID(6LxQTUpe?n-p#fN#^|6P|-=P_*l%JJXYV%(Py1}Uv= z=#~@3SyOyus7l4MjoW)N)x$2M#__Mxc7;{;$YtARjFPPXmVoQRPYLaxvR?+L507CB z#;I|)NEzq9UXNRJdp>*KdI;+>O%@i9H{lW$v?R;gQ@LU}XV!K5N{Ij2le^`1RXptd zEk_Z5y#&gCp0cCrJ;jFPCOejMrFn(&I{<8%Rol4#xa)#UXa@xt@($oc)T0oZY_OtaknvR(WbL z4vQbn+p2Am(h?235Y3G_e;vIe8=!L72Ts|ypCoA7+5hAert&YnZGz$ zDtD1I&6$@E^hO%L%-6?&F%g{3#jBigtp#3RxEquH(zifsS0-teavBqrGg;@Jp=t>r>GBc?NSIofr7n z{g)E^j6B6LPJ=n)t_aRKueYE}>Goh~P&{*asef{%?j-w>;! z3ZXvl4qROyz`ed9!Yd-gd*@XoOJ?uHynq#8u}htq$rZD{J@;})Cm!XVICuDnqbfEt8wzK<<9&y)PFG<@TCQXa2Z7?%y^FRH*k{v7P!4UOQ==b53bnFuP zM0c$mTf*(+e&2k`l}h?a9;y~_7cY)xi@q3e)4mzN*CA8TYwF&!%{?st>&LKBmpQIX zNy43Tj)PU-EVCOTGt!W>5|+p&bWD8>Ud_aL{B zEfRDn$hSXPqO6EXnWxE_&(%UZq&#&L&%wmG8n~0=$MlyL+{+gdVO=1-H-)*$CO1+N z>ug`iVE-uY<(IAO`r2sLUoM9=zS>Wsclw94{i1Igiy$l{jz2?ID;ZKcRf&R7t^^1KcWMWO z?XKjWO;-|hNSdGbWehiCBX3Kj6v3T3lED@GTC!i-`?9Z>6NQpaNnRL6)dZn3QSCIhq>qJ)1m zzOl7`4hZ~9Y19U(LPDbOoeAp|cZ%im?z7Iv(uIEbv%aS!(IHi!%a;{-z5gO^etU*3 zUsB0ktD4O{-`*dk`Kvb%M?d2}O!Y@wPC0!7o^-1xqyMOg|RFv z9PZFRn#tBLxJ!E}sr&`G{JGhi_aCZ&L-X6%=iVnctzJ=2^8&bxQ$^w=dLu5IZXq%2^1(~?dfd^pIjnxs5w@Xt zBd1__LHzmtSd?sz!ce8f*yP+xT90UNUK$%6a2n6lp5f-&=>MmKBI&&YRv{!&MHA)M z#URpgh;|q7a^FjwvHx_uxb7OX2c1X1?@?GfBUaomLWdojycViQr*LwrC%I1{J-EHO ziJYBc0H@&a0ZL2*u&=m6s57Fsj4(z>{FB-AA!p&o$VjLRUxnMA)RAwGzOwa`Z*dzu zqBu(bm-ubWN_=^t2psE1N(MMCU>|iguuQxqmv|`!j!hYl*OgT`(?Cm98el+{p9lxT zZz^opzFSg#2(zh}j5z8P)TdXAAD(^9_20i#61nFbC%qS+nG0i;J-O=q`K-V9cCKhz zCTDzU7Hg=soBMUzgey?gVz++W%#CY|Q#xYyEz3%by~ z_Pcwjnwz`GTw0INSO4=Hw&tZ54)zMfT|L~z>aYL8``3rXmm=2U&jnZ5Io|7}Z5NeB zALh1uws9VEk3gY(49K$KT*+0M9aevxwJX@cT@2E~ZMQ|zz6$%l(=)9?Hpj+^i;lIG z)b>bX-xo~(SO2b-uVAkK7z*bu!^3+b;Xi&>-*00#6^!QY&w34uUDYwXcBHt^wiQ^= zFrED&(++k?Z4mzdA6^7G^V*; zeWWrG$t?W=YaXt_b@J=DLG3j}t7FA~^@MeXIkM%7BiM>Z+1NCvfgS&=FZ8HybP~4-tr<3(AvTwh<|4%j| zQhSQkHrR!+M{XmtG7zP7j(wgfxj5|%_r2{Ww|kNdvwp1;Q~di7TB*#3ccB}_I|uMs z!|!>Um8d=NIBz>!T#^J$tL9_)&3gDaVj|Wz-)4q;g)w>;P7x_R)stfI_{Q54zdMAh zE^bE6VPm0?6G?0oR3s;xrZaWc^26j}kJtKz9u$r$;t^DoPg0dxzMw%Ha>N zN1Owb(i|o5$S(@*Jrku)VP>dhJV+YNX1#6nx8&-BRA z<|0)b*s|!M%)_~Q%%;h#_#*G0CmQF>(8~!EeB=?isp81Bteg4YGa|8s#iL7U5StW57OrEfMan@2E!lcQ3`(U!s zP?)ESbOxuwn4$BkCm7MM6{7}sVzQMF9?9+!cYIJL zbrXZ}b7wy=Uicge!@o+$CV?hYWb^TA4l%L`xdQ*v{Dh3waYN)Sxyt!J z=558aGC2FFYVLZcBWITLL2TnH%S3F91yMp2dat_yAA9IXCe4(viO<(FKfD;G-7Sb| zU8T-Vn7om5(bAPLWkEQux)<}z1n6sN2Vjqetk~srAziYwfRA6=gyto}zNhGa+9h&& z^jW+n!G(P__$n$mWW%Tdo>(+>9LOC@V&4IeW&f207sU+T?{pVveBFm5UNP`!a$m`` z%v#Q6&3ZxKq_i%rt1f*OFRV?&4lE$7s_X-9xycpogR&Xkj0Iq+DP%^(Jr=t50JOg=G+O0H}vk-pm;O5Qw z{@N|HyZ6>tP%?SEa9uV-j=R|!$E~b;$qq|=iVIHN!XF;GT*ldf@cKh3yl45b$LS{q z-t&P0Y6+}j+#a~HM_v5j;Q)5=c{gr}p#>YhGeOYLTK{33$(?7Qu~eQ}T(Oc_Rv^!` z7>9B)2cL8Otaky7i$j*ZNlFY?;}%8UXKZOS9hx8~QBfbv{dgD2j7_=9JT1}y9jlwf zZMQzN>8w1{)FJE(il*HXah5mDxH~1^@lQbEzjuu9N2v~p3XA<%^IBKz_Zykp5@q4N zEGpU08lL*f?rf-nr#_Eh+uJGZfmxe`J{7^>)Ab87EFW9 z6y}YuGegxCV%^IXte>Jkd$no+e0<1a#I{;I;FT`$%kvSoZw`l@N>l$EKSh^K*TV#P z9;31-3M_0p*rKiX*eJIPV&A+xZmh`w#&GyW=2XFBbm=ot;`o^6Bo?JyeP9wdaZx4S zpQeeCwx`&f@)|O{Q=6j}?q+qhbQvq(SR4?)78iWDM@-HglFBH|^5Q`}kzv8ycex0! zo=p<^{eAydOk>@d1vT67hEA0rOQ&7doYQ!DuJrOQseD9(^O(GR^6TLqeF2Rn(`w=-hnr0#iKtR#kWTs~ar7ioE=Pq)O&pIqYBtQ{o@ zad{$@r^se)8s}=L$~FEy$vKoyW|V3+2>K(vcgZ|U+#u6^7SWie$80R%&tv)gFX4vV zs@M?%pNdx;1p4(SM5L(52iC31fP-BMn6>mW#yzuP=Qt+8vB&P{?WV>IJJbvvw`~O8 zrF5LvJE7GQd*<}DF-&p%M69?{OLk55>tCnekUdJB0EI<3X}e7s0YZ7p=4bDWC{ zOBeE`^-pM>&E)1RWxgg0bHT9ANjjXoxUn>Tv5_Fpe)h(kkGZT^t-T#eFRwu9Guszv za6e~^WNcmpaT(zs*!3Qhg|b0c4h#Ai^?#W9@_4SE@BbGPlC@N}L@PoGuQ``Ww2C$r zDUnJ^8`?=PyNV*aHWVU7DX%$K6sc&}qNr%!*LSPd-|LP~eSd$<Z z-qfqoP^j;We)6`?5Os7jQ80cB5wqV2qPKX(fO2)Sm4x(2^{( zXXRg{^!qNgjme@PbW;DyD5;o}%xDRDuIq&{dQ*lSVb{FY`BK^~o4&SLfY}yGbVAxV zCeI96zi)y5P29dxg4>K{;htf_HCc}rNBVfznji zWB|_nMncZq8^rOxF&{tSd4=nScQL85KG(Wn4#FILqP|6l^r&m-$Y@;$$BHpSH*6}i-zd@t*Lq6a36=6yhXGZ{N~KrRTVC&OeE)Q$A~k1XTha!CYW|VPCWbZ zLTVhl5Z-Q6#uxj>XsL1FVm;T@DdvcuVr)l;xH}P=>A4t~`g)soymkJaam6{C~_>LNvU~2^uVw$CyLL zB=<==RyN%Ii$ydk?khC^R|GS&ehM;51le8>9*b6OZt!Q`ZWi}u;Xa_2I&i3ZKAqig z9q07+r+VKu;^jqK(9&|JKue<0?0~yxs$riY#jrEzFnKF|C(SdGd7Je=R#9Em@x2si ztF)|RyuA$cvHW==-4ghRr1rST_BRK%$@rdBr!%jXbm9-{LtUaYK5TrAt8A*_HLTGT>n zOApez4?pRe-o|9b;V)#`;{eXHDo*+;r@N3=vZG-#vq;jryG^Y3&4Zu`*5J>Hpj1VR|CF+VDtd~zzBfm4`>iYCl6QAj)^1I!W>?t)Ff1tn zSllU-sB~E`=!D7eD7BZ}-D2;5*jsleORe>F*!xEV_K4e`)rqyL`r5rM>x@~Qe-Mj` z?oikMgKi4zM4dZxbhG7eyS(SBWLV5aDc^qq=}|icz4s2sp7~vETF0d@I!Oz63fM(U zhgw&}=qk{>#$$Bg06m;PV;eq=Hsy+@cGR;bc>_F0T+rKjHdVE*f`sqtaNx8O#LmvA z)o;hc*Ynzfy_wuE%oB0hvtM+d>MrmwJi+Aean?-Q_aRSh>&2`+=Z82+Rk!l=L2Vux zvwk;`G!Nz6`de^2Jyf|Q!wtCVj6P=18wiaPN5aGQ&Wwigd}o1&8+E2(0?n}bGIEF^HSv0^~aMjpC>VUWRD4>GY_ZJPkA;_d!Y!EUi*kU7g%wH z*JEM&gCopV_aDyT_H1(J^l#~MdYd1?oN0SNMO_Y`KT+efX3pmnEEm(94NK5GXBQRZ z#u-Zl+LEi|pV8fC)G;OTfX(=pMWptF4aCi3K2Z|ujXk&?KJCymWE0~dr5nz78V09w6iLX52=vYGBhJr=hVJj= z>GJgxsKWAcfj6Ts@YdhByW)rOG)V3And!&TeGV+JV*M|X)Ba>KRkactmyBb1YS&Bc z4mi8x+A=xTrtaSy0_LrGu;YvqCoZlA(>H|-SFTDKqc!W8HBFrxO-r6!f8oI0l8SO4f6$Pk0r==iJ>{DB72fm+&5&`2&$B|5)#E`zU)D=kY zi4qc197JMQ{}KDl4TspKOl&c<#0yK!7_Ch?uGD`-8ChCq$9c@!#s!^lgM2aX*#O8mpE?&)qylFRl-PyS=Z#r?x(L``}mckj))qdQ-=Lq;rn; z#Q6htFyAeM<*8YiCAE3d4Srp%Ct(3A4B; zTO~|BkNJ_%!!1)9Ae(oaWAGHuijIQvTu*Mu>|WfOgju-z!AksJnGgv$9O93V`JQUh zV;8o7@|i}dTZ##|K^)XHtYP$;4tsIs8+Hn1K(v0QIX-r2B%kM>16}8#)FmvQF3<{< zdfVNg=hfFq<0DI8q1|GppF!tM&}~-=e&DANA@4}_8hSFuhz=XnM*6g*Li5T&FmCBS z7?8)mzQ;B^enk=2+?_>lox4o@TbF=J&mx#NUIRj(%f^B7 zl1V`9UOMn~8{PI~Cb5+G;-2=)OlEL=d?khZXRX2T=jdUW)!IzM>OdS$gK5i( z6mi(nKeX^eHmx42fu|mI$18dH^oo0LxaoQvcGdJlY4Zj&2}`BrC#4M6%qu71^rr4K zE?^v(y(oq7nzl?rccP`e*>R&KM?(ri0k84%`msF&pI3asj(!KuH5#-=(Af zWV2*H^2;949Ag3thDgx)=Ll@v`$4SH_>yeeVoSQ*H-%4YJ#cy_HxhlIfXRP6aWm6l z#w#K|_2jmAMgWpiv?mQ-7z4xdmP@ax8DZS+NIbvPkQCkskWO&VCbl~}`~v29|6no* za&+py4ZW_f#EJJ0(!(wNaZdSM`e~s8h_}SiE5@#HP_bIDk3<$kck7G9IpM#>Dk~S` zi0lxt+lg1?S1a=kkZioKPU8E=u(FC?55yBqZD8kP*3OEGclW}eHSdTh)Pj|Zdv_8@ z_TD2*2J5ml)N__RENB`l9ym)KM_zkQuD@>}qbwfSeH?imjwjXw*GlQh4W6VQc4jsZ zWI59OlJ0ywny#BQ1a-#7TZXJL;UHsISq8=3}vWS z7emjc7-Cc(OYVh8=^?q-^nmAeTK&1he?}DL38io3n4KO~2B6OnI8FO7et%8KmN$_%;4sSiximzSu z2D@#)NF1|niv_t)q?vI0?Kfi1+i0e%ShQr3-N^I*`fL^HRq@Q`N^S7%6Yg4Y~1?oKCz;asFl;>hlL{FZJRJ(&j*VsXMs? zRnk1qE%DDV-&#KmHq{p~ufIoE)nGNqVnmq+<)k&rCCL^k`6 zag=~_Y+uK&u6AwPCrUYzRqB+wW=0ZMKjBk4=Uv~Kot zJe187svX4Y;F-=@;P7=A)b)=5^I1mliN8fJ^z06r`_7Y}CQ;b<;TUON!~V-8g}FPV z`btRkb{?fV+9SC2`y1fZTJ}9Mc)FxP&fV@xq)`XKK5)n>c!tVB+0dK`K_83h(fez17@ zcrbeNTYNxOHtv_un-BEA&f0|gTRZ-D7uCj6`i+f?F)V`ok$=u(?D;I5Icq;LxL;1s zckKcjQ=?$cmbJLG;wbVvx6pUD;uxNq@e1hVWJW#j_rg1Czu?7frS$w#39ByxtsCDJ z@QY+7r*fr*{_qcF^*|OcGWrk$Z}XqfnTrmAl}-S6 z6U3wIWc@UYLU`=G9{ht?A4F1i`6qqVR!f#{awZ3EH&dg?7&3ieOGVkER7fjX0J}z3 zkt<6ElaAO@3b(*q(wO=cCrqk+00bEA6100<8;#9&axjQOdxS=O2*;u%i z<4f+>%ad4N^o@QqnuKusdoKcpGW4=cwaTNk93inDRpjJ?@@_v7Z-kCoajsL9=RxSC&U| zL`{Rhq*x)n)B4i!k0#m(!l^56YizlQRb1 zoYF}Z*m>?EfEGr2U0v#Cj4t-|JMnUY}oi$pdc&q+_&HHeMm~Uwve5z>h3YB7^4U;Y!DA zH0+ZB>Hhp8Ze1mowRn~|<*Llnr;9n7J}_pugM3PH#UfwuY;+?vtvRT7YP;+ocr=sg z33$_A(&wz41hwhlhfivBruGY*SZ)HR1K9kQ$mp3jTv>jN(UR#xbh$j5)b{!S!o9;r zHQe@6LMl2<;Z)RaV5rGnAzrfa=SfBr2JEJW4|AoT)sM0KJIiBN+vM6w2Ev$b&YhoJqpZsvOcwB*h6D{e&iGFW}JxyI=0;Fzkh+Lpemqt8f^={|yXS6_00sCw;0)u%8_(IPQ z*Pj@|Fs#$rBTgEy7t;0SFsuUJ&apOB&_~JGCb71;9>XI%OTT1Nx@Q87Y1@sPJl&}3 zihO8V{E)cLP=I^C{K#qr3UfdDb9<9=>G|6Sq$4uB@-5F&=yd5Uni74TPPH7!nNu_S z<~T%fC+9?RHOHoNqaSaFZ;|E1Z+>6&?z<7!^y`7@we8SV;TR;pbD&{TZEAi!7t^u} zF|=X>Q7YU_Yu)0}td&5?9uduSSK+Ft7dNG98gwguLoYnrg)Jt(k^ApC7}wqxKVANs zteqFY^qiY868i@^ptaFzS-qHsy>^SyeSjW+yUzxqDoNpP5A4Z39^%C8aBI+fdgg4_ zKbw5rc1gM@To-zUpXD4UW^*MsE^_*r?P6CyHjl}$y8DdA?f7qulf{h)Tuw)>SLbfM zQQ!)Ir&Uspd|Q)A+PW-*vf3EB|JZKyQO#mDxS{2R3;af4_24CNB+`Jh_!PjtZFwQa zxHzpA3L3I-Q)bGSDKGFFh-nCxL3-NW0HPAb*6aT7zx5~jT6ZrQV_d{Fb zZSmPPtzx&F_r!ai6R)Jw&KFzW!Nwy?a74Hq7j(2UI_D+w1xph7`~i;zI+C8%B5uGS zU)wW3TJVmtEyinD+SQISVC8IKsTMcidkco@mZEy+3w*990=%ZN_DYg8TAnWtn~mm* zD_Gg{^9-T;mp!1TY(x3LASXU<%m9AHyAe>IKbh5qw!j~FGP)fPh??>J)FP&TGEJU; zz5X44M`cpwrHk5?%gf*MqbI6XD!y6H*DT!2>#h-1Ru!J-i${nmvjgYxLmsy9ITF{( zXM-11S`5{y#Mtj9Ayk>?!h_`1xW=n%P=YzfK(|IfaB zmAxD0)SSlt2jsEK@2{lW!B|w;@PdrySMewMY4AhcTJfp;WPXo3@YlS4@y&_TXs$wM zD5^a!;FZ+08)Ebhd;Y5KAGlfLgW72a1RNqE+}_`Xm%ikg&V=VdihKCzDXF|{{W{cS zbmvA_)?&>2#c*_lKe*lN$#mr2#s9A!1R4?{DBh_>oiI;4SFm2T=JW)f58X+Y$Id3v z6J+PJUh~244}aRdUmn5ul(d&&{qn8+RlTeK_?mrC#HqQJcwn9qzhGNEzDkzo7b%8f zA2l@)=uiC;$#>&7i?b{*p~QU~Z&#a7YMe9BL-iXvq&%Y`ONQcMvrM$9P3E0F1NmK% z9AD+~1Rj5A0$n0yG)TTCtmqX8ZL z#23ku3u*jL{J%Mm81E)}U)CSUH@hUBUFR(uQ_hy!<{RPn^Qrux((R;V{VyE!A&3vV z(Hs25TX0*^PyUo&r%Hty%8$}{#cyzYiVfrB1^Fe%Lp}KVe$#o)OfB@7s@Hm^%!I@6l2^GiC?7--m;VcE1(UA2%-Iy`)JXCTUm~l*Iq~y$BDw z@Hkh|02T~?iwiT$ahUxy-eOiDv!8|1L=2Rz!;4bvLhnW?a|-jX^SjS_@ZRnw?0pT_ zmoq)D#Qj&U$c4{ng>v)RaJKhkewxX4dTK@-@>{RrF`Mqt|MhHGT($sv2jLXV4H7~agvXM^HzYRey$UMb$l z=BxzVKkxg~&4Hr7wvrrn%Hsnsx$_U6ne(3y59h0-BF1Bs>^y?BJ^lEm;-kFjoFRNh z%y7PRg89F8@yYKJKI?W@zNDG050dQ3slgldT`M2iO`=1F_2C|9$@aalk7Rso)U2@Y zr7{xOuY#0?M3Gc;v7lp-5a#`f`*;60?yGSg$}R0dEN$P?l0$E$hRN!XQT!D*+KKqL z`fhx2+&A2$_PiyN8_c!Dm0U0V`i*u33?WS-kNEAis%B} zJ2pXfR>6FNy*}glgGuD*D+xCH>!R&IISeZ<;>F{?VscFhAN^(^KW$t;kPlb`a@PC# zMXED+xwsLaI4cT*(^m0;^*JDEu;tUUFX4LI2bfV>E$EK%60RrmFA4eTlFSf$qm7hRiq9uOpv_;mgO{anP^XM%+j*DVF(EIBu zTC=4Pb6aj=k>_;&%-eCYHSc4|BNaIuJ^wnc(>h6)#u9v&GKTh?J)D&@fpKRH|lW(}*5z9^c65#@bx#%5Cdicq2av#9NzW=Zfoai{t{~Qx`M%eq9kgubGp1&#%ydJJRuX zzz{B7KNu(H5r*G6_z@l56ou&@Mnc+>Ei7)}Q86|?`%L{lE}^}WmJsCw8st!j9ef!s zhVXaN`fIisILXweW>Zl8QihXX1*pl@^``IK8IJ43cVU%0Lb z@0qLc(svQE#d3O(-sd@;)+-%b-TQ-gMmtyj_7-Qe%Yh@x+n}fNR@u3Jr@*fA3@7M1 z_#N{ZlW1)`3}uf`GX4c)rMNKbjd5$eF#VPx#@V_JJ8GN0LKIVwYFN-1hItLH$nv=YbO1$Rr3i=^(9JU@f z0M{nY2hTkgG=7aA9)9!~YxZ?Q8U(;$^qXemtv@MIsU%e4wp_0;8Laz=k~1c#t)yChoN$p zamy@o_|nje+b?Q|B10dp(}{Up#E1x-m)wJ&rP+o7rnZUkY6UhM+rky+-E`IV9NKDk)V6NxEF7>PUEDmO54=)| zz*vp1IQ?XfAe-nx-90AL#qIm4&ZE9?@wqzeijW8QcYEk*(>U5{aU5;$UdOM|$1pI> zjFo{x6CE+9m?`9w;u=E|`hW4)fyOZPjF(;E*c@St@^;-x9JN7a@4x27X=jevw zW3NbB_PRxUYS<&rJ=4sRSFPo!@-;_${Ud-jD=Cp9xks}!$r%xR17a^uG zm)41W^XsX0%fv@D!k3(XI z{)Oybcs}>|DeZdG5nsG1!s7Q93`5syW4dwaA7WKu2NTyMVT>VL+b0q*guR$zZ#sC2 zy^we9tp|`E5W(JWsPm$}Nvn=+(2b^9V?3GsMI|*L#Lpgl7NXlXz}DXqsfNuFyb$&XTOZ9Q zLuESqwKs;e)et!F6iLjXFSH=flzz-EM|;}^xOT!+-gQWC)}EBw7)ujQEP=72S!A^2 zrht{n&~T=K3rsx9Y1aoaUIITM?0caKH(R@L**b*lDLYHEZtN5I*}YAAsca`rS1(|g z2ekBtukJNCDKAvqAX}ezu;?&WmVBW6PB)xIj6S-ew4`e zT->b&J`ML7jjrUA^yaC8e|l*ttia{*6|~2QU0AK$u?8~9r8_#E)Buy8-^6yJdK|ug zCO>1#aDMzXQ@*X`9#@g_l4~6uCde;o*`r2!O&)+&vnS(*7B#@uSr8SaA@jTT2d4$m ziantZYiK{_H%u?LapE>badhJ!Our{0*_?x9w6v7k`32 zNmnOMqN^Ux#ekEUtZkUH@-Wdm6h*H_uf{KNi_m{h9htn+4=*&k^Vi!e`3gRj*;M1B z8?SOS3-4TC484mtkniqu>Bm9t{N1jTNz+&he%z1$)wkEH2=|utpP2pyEI01u!j*>| zK3Zpf=0Lwg$*lbm-u(@(+n$p8*f{!UdqKtO=Sk3@94~fzv0u=qDCoz0Y*{=4g9}yI zb9(A9EUj~=rj{LJkxz#-;H0Bd|LI=H6WZR26}jxcO|pGMIe42);;yD`;I#i+j_KjC zwA%U+E1Ru_%UC`Sl>zjNTORK0^c1HX>%#3Nk!briAO6Z0a9oa0*_)5q4%0C9$wC%i zUN)aUePT1Z$PeSo69=&JDWt!+SWHbE#MIxUNE)#_7^g{l;E-iXF$Bx5ayF1T6}7l zY)=kaabXVuo2?J%$1eM6N_zu&)VB=#S?q?TBejT`{bbU2{dk7W#7Yhtnxv%Z>l=!GQnMd`n7X6I*VG+ zUZ}(wf1V|4J3VQ-Yd4tW+Jl>Xum^YUL5EMA*|0ikO*k(;cfy1?4GEBLeL4d(|MILoz)I^=Cp#`)TPkN`vDGKaFUul zFyJ<>odjV&JF&7na=biOwn!B&OsIj;mhYrnT*uSwB0bLZmlMO>d$1Dc@ks}*+@0~^ zohkS_Nr~ZU`%(%HUHY-QEMSY691Sba=~Lz0QgUueN1w~s#pj<)zdp%9vH2vfrJn~k zaJ&_$Pd+AI?RuPSq7TLDMu-L9Ojz1gO80QQ+aAcgp8@Ol_QD7q_OBq6Z_!b&F>v0a zAJ^92hnj`Q<4RjL$0I3OHJ8i1e2!b!QpFwkZ;T+5$RRfiEK{R6zsV(>&ZrffTlEjv zV4@Ctf3?De!t>mS)GF>;&;}^^B9=HS${tN$;O(``P}pmuRw3V%1C}km%fbl+27H0 z(l<&EzSXOcWDF!TQrFuKlC8nG7SM;~(cMu8KkYWb^ZY2fDflv3o#@1!nR)}3K9A;t zHw@q;nsc!3YbRFMY-_iJZ_h+d_1!S;W|j$gf5(qroZJp#NuL+WV;Z-ozb))H-%0o?ua{jeftCZn%?q=k&SW6RYq zeFN1O32U%&&KMQ9l4+Dr9M-~fJWk1Fi;-A@cX{iXM6-+;kgf~x&N%c zZJTJkZeMb;(nYB(a^ZgzPn1PD9)L#9itCjg#bdQDpQQ3?2T_6KD#U zrxu^2&I)}|^?)7yv`rfyK5oPM`(O)gCgr zrz6vuKtoMyyLd#37fKBr@$)_l9JDkKRvI+Z>T{M9%#@+_@(Z{Un+``~*M^`Zgix%R>G-`~i81s(Pc4j&3p{g;yWhsJWs)v5f?x$_zRK6`Sg zB=`ic9omWZojOCXt0>vMj;KHJrB$6}`}-)_qinS|I1Y@f7?fK>ublW^K~C#%uM1PT z?Is852hWc0+Ww#@Mlavq6XxBBg3PJ8a4)_GXOj?35~rAgYU>}!2{nQZI??#DrIVm@ z(OBM%4Ek|~DDSvVSH)#XGk4gc%OMk3w=ap$x;lWDyQl`vy651-jzrkvn=CDxIhnhi z+L_z(xB*H?f7YgqJ-L-T=IdyG)u)kj*SiU#E*3Z~ER1hj@f@eTjz|5d88FN19NFzO z2s<0zreAzZXy+k^0j_1h%8++7s=%KNtvbeV3o_Y{-O3$t*5ZOMzk-u*u0hViDMUF( zm#FWXPN$zsmm0r&E&HzwVBhk>qI>kNofA>>al`MoBjGZ2V`b?U(Iwx}0fH^-NVa|f z$sS|^=?aT6CwQb-o<0EEf$pqL75M!P6cgc{u{vCx-ksZ>F6P{2d#tFmaJNj1M*gE-lF-(#NY%l0GEljYjIb#iOeUW)q2?y zAJoh2gv&qQqxvE(xVrI#9R%+o;kSo~oxO6QM>jLLAsNUQZ`%Na-}gpUy?7Y={28<3 zf3<^{erH%3vAUi)+L^mDXexBRJ%wvqcOG87doPrAR$mS-(MKQiBD&}3UO4&a4H?}v znS5FJmQ0?V$E)4DNgut?VX_HW4a}AQ-G6u6jd@k2S0}2NdH*g=l=5Yh; zE|Lprktj-kMS84Fg#liDh<;=Lf9e0nUvgr)njj0) z=XT8bInMq9`%{clg#z-ae(fy2X>^Ql4gIEeC(K zWZ?R4CqrMJqAsV?sLh{^va>f{!SGw&VE(7>uiYi_o+kAAo(ut_}>>Tf4x6id&x(tV{9v1XSBdGRI~7CuSFqL{J#sq-tu zj;r;crR+Py*W2HKw+Yil-_=%l(t&-OB~P->lV2z2OILf(1CAZ=T^Tn0i?WX}5KGF2X=|K8YcQB{DYd+0ziN~OWVN8E9k&iG* z>m~_pp8`8=)tSDw$6Y5MJ&l+g^ZO5B=@dp7L9%BE#Mo=&@7x0#iS3fpwK~ z0!Gnt?^GPF5kdV<<&#N{tDrz4Vb}+DZzrPLXW?i2NZGzLufcTaH->xW&PY;IZN%sb zv{B)j^oLq1vC}d}g^UO^l*(+rv62>C4`6hPZI9r+{YmH~vOve@jzT_Bk+dtc`&c7X zEAx_!c4A#`1%nm#qxLQzym~^FUvj=De>hYfhvl<*6Oo+ESM|1EC)B$s;U^95g7>ef zf}?zgZ{UAD6QsXN>S@~LX?#xaT^N^}BaV8gNp@GMql;w=hM9CDW<#zDb`d2tSi`w9 z9wbe=l5DNZ$C^>uEWRbPQfw6(2$?(9K>CV2td>rPrV&R;*i${gv|^~9B?o~`Ls716 zi-1Ycv1ms0KDuo0793OMax~UG;ByX2lLzwDR5eqEr~ ziKVpsMg70;p7AwB*?xc5rEP|6yi|05>=kj$(KMjTV@Qjah}YQk7|q65z#=c506bDMJR`jRoF!E@amq|fSO@TAu^KIrWj-rw{PEtQQ4>uE%i(5!Y^dAUE# z8C3>TA5>z9oopSka@erEZOP<&=@Z!+-iuN9-CawF+CJ`B+xAdU|w6&2WLj^hflKe^91?kSC591CD!mLCYdU3U5GNrbKZOSZuai_ z4|Tipnkwm`eV5RpQ#y?bpUc_`<$V?~=4=;U-gY^*G#@~ts1CmufnNQeOQa=t8616` zLH;W}ib=Bb5XP;|!0-}%eD~%N^a;MmXbUgh*P`(+ zlF)wmDe1^tJ4k(ao`6@drO0^x5i)kt6X~vHgP?DhLl}5giJ#7eli<1{xVY+rv^Af7 zrzAIG4VV&l1hrq7@S=W|U@`b1qir4*PkLT`h~v8W;)Iiv;C6!+85e6wZ5}HS ziOvq$p0PKGV){}nsW+gbW{#jb{p9hFXC9p!c^MP+?vhpcJ-CmgipH!cq_&F&!JwBN zzCwbG4%sQxXG;p|4ZX+gR`;P9EfzFkO+;^M5aun-7=IX=_x1zF;j(cgGLCO-8-%^e zGU1|K6u96Pu<4?OZ3{QRhG~6~+9`*1eaBNE;T>77L@>6fv$ zNcAuUKCG7}FT_n8UJqaHrl9DQJnz0p_WwCS0bPCvGCUV{8AITj?HKCx167O9@-GUy z^9OB|NPtQzp6oM#$;lVbWr{PuJ{T!^zJ2d7$~^3ybC14B9yl*+xN zi3+UGC=vL+^{cVFo3NLJoSBc}70Qq>TDE_7h=?!nuaU;wxQ5}fJ@6WD8qgCrqENcL z2YlAdfmJ6FTi->u=G_pli3p?5L<`8If>3I{!jW!jJxmvn&Qw3?Gi^N4 z9b+4H`G*H9@R(gK`7f~5?)3X!cG(wFv2<}K%$b+T^nQP+3RoMAr75qz;pAO9ywuG1?JlTpCI@Ujp(u$qY2{Assu3iOA@Q1e%hxv#38Ri2SH%CDRSN zQTY#(sHXE-;-LE+{El5D;_%1Vtt1Nchwg)s?j8U2AMPF%+jKsKG0`1!!UCOrj(_l` zRX;qrWH7j-4W%kk-hxggZAw!3!dZY#*KK&cpb2ku`G`sW_OO29U_N%Zh}%7753Sv| z4dr`{!;&W1UUb8skVAKaVc8uvryvn%7%em4=KD>hi&rnC+utN`CDVHFDnIU$g%0Ck z$)aUwmR(BRFBS-xB&}Q0$b{aHX!O}S*?9DI8lt1Y$CX*|9{m!j;rTGG$+x$CfL{pb z>#4Nk!!M!zO9m#_^19SEFsxg`HZ&vh*U`nz!)t@;Ue+x`6xC z&fw(Pa%lha4Yro`wX;9^fBy$NJX~Pwr~)h{IIb~*d*NMb&9_#4U%1~9kS zNdD7*Mqg@f756@On0#GW2X#(sF{x;wIPU=wbkAs?Et@Lyc{u~aHuc~x4zFW2Nwt$e zW9M!HPs!s+-?=gr`JPu5q=3w77-Emppj}X?jlBInxIYitR zS|HX=EUt}@Beo=xhO791NiVnjoDruWZ9;#|U^wc`m zZVm{p!oT_vt~G%woR0Qwa1H2#!=LHX+5|cA{4bTXCet2zO2Vn_v@_!22Q}gT<|OI8 zD}8v)92G8d<#WM5M$~XEjRa~Ogs?ejki?eA^<&ZWx6 zdaTA>Bc)YU%m%W&68|09Fml2+SnctO=_YAux%7NM4fcJwl7BJF5hVVN#BNQ;+>K`P z2$I)Ff_nTDkl(GqJ(}SHZ@XlZH#ctKy44P-S!KoOh8hh&rk10~4SlJ@Ju8mIUxpC` zZdswmDH90Z!S*R)Z6`eG#3)kH{j$4%FDt>Nu0J~3YWcs<$|2b7i zS&EOxo}!OkZ7_S;X2v7YGl6Wg8ALT6eI~Vy(KO#QT1YQB^m!*4q5Paa8Dr0$1J)Fg z(CI0}(j$%OBB6Xa{o#I0pds4v+X~xo1n1*$2doksxVB4+Lrc#HPP)>{kD z^Pr04Zz15J!qL`4#nAqF0^{#;bvl35g5#Igbmd3=PG$1FYT7|PxIl4fdjsCSTZR*s zUn0GNJ42JfE2?ewlgSn@T7pYXMWCf`hd~q?7g3ApEnIAgAPSn7?-pdEQf%Hnk_y{gtO_TW}N+ zaCn&*!epbd_+P3zSGP@xn{{&(HSN_gAN*onH@;zFA$~9JN30H3lMB|eHI15O@Lgdc zuc&9ock|l^R~lzAc?4Sje|UVWH{3N=!Z)o6aI0cEuDR5T?_Ri|LP9cWUF%8Lf7F!u z8)b?^uKA<2R>vH2M8FaFd@vGDEhwX^%SQ0a?CxS*f*0b9Y}CJ_%nMjZS_y{!C=}0n zR3z{f@Cke+U7I@LjKt$IKiv?B%u>VJwoB;m8w|r{YjTdV9S?*w9|Kfy5?v#mbDGl1 zM}NdN)mIq}Gv`{WzfTi)$DN|Pzw6tLmdpY7UBOI$;{M;z*I_O0ucdhQmKyGHk-)am z(`0+J{-y_xCr}&fRpgVF^FLYEMvo+?eP!pY*U09EFa5@+h1IM)2y(9I%g&?_pQ10(V@?2W$Ja+0B8BqA-0$^SOpigFpyVCEycM7~E zTPBX+W;`9jbkjq997McJ6!?f9_Q@vdrxp{jYdRV@G?UkhqgcG~?&!9Tf7#XB<}mtd z>K*g8e|ZUXCzhr%Y|dI9I5I&M-Ba>~{E~48OUR60*`R$!18x1*p|@Q)_#C_m1C-tI z&6vB$g|7q$w|?TA#@Z-9ss&tZ9@&jLFb3zeXOoGe#CD^k7jd@gaIn;Er{iC4f|3O< z!R>+qXYDzWU-``fOIFCC`n)*k5`>VrY&?9K@JD)8=F21S=9XNc-CPs#XrLwcCf$iE zwXvfSmn(7kcBIa38gTDP)A1Dt$t9z_(slleJBb zo9WOd-SnRMe~X0pr^gLhyT32#4Fm>>sP5kXV7l#jy4c?syH3m}HZRM?XL}FF7eVDP zJ3yH~K3x|Z-`*pSZgnR;-z?{n?wf(Q5UGiVA#U3!!M&?m=`m7_?__&O)PH?Tx}D;% zbdx;4bx8_0SHq7Rr2P>uKfZ?gZ#veI=+{cf3ZtH!DB?1FoXz&Rk!*3gO}7p0%~||c z4ZD;J;pykT;-DFYs8jV6mo4|^_AMC2&6sS=&oS#u1pR2_nnO*@XnZo#00&pB=H7Sw z%Q1Y5^yXt;z04Dp;pFz(C< z92$F8c+Y5jDV5OkQRBhMrx{)yN~Yf@?m~@p2fVZS9O>yI55L^o$xKlfPRa5ID?3(i zFH8INYK7;H_n~2W$G?8})X&GSG$u(?Zj_-;n<1y8IuPznjmN`tRahU-xljV8{WBnb zv=Z++LW%!2rd0eill4mYCCCI~BR9 zc{?E6>Jq6q#M)JffII2@8xp>|4^~454MESE70p=N?=kw8UXczcoPf=;y>}-(w?=ET z>vW844%NNs3eHkUV7e4&+_P+;!aKdbQtGZ(2YH9q(^D}a zXjxxO?xpP`;TzJW{RZtt<5*kT{P7r(W;|hIB0`>Z`@UjDiIPy~MPj{JI%DW5y9M$y zq(fc-q#A_D{7W{`pgz9L1_Q!drH(4+=vOOc?$sv|7ZR?>%6a8mHKG=0fhuZ;nEhq6 zB+DOGq1KEZoO0M1Olp>`1J4{T=#A+?;B|7ov3S|_yRx$b=h9tm%V@xe6nJQt$xT`* z8@K#pf~V5`xfcVc(<58wz-^ZaoHWT#>Uzl(M;i2H`j>9bLYp?Q~4;}FlXDP+f`cM^^qgsx4WWJYfd4b2TIdu3? zZJQwTfe5OCWcww=HDo-?vp{|c+G|WwZ>5EjBD6}SMJPmf zUUO}gC=uGMiI7AR*_W=$k|^3#5{gn$NZpx}l*m#@T2!_Q6)B1oMc;c)AMfAy`}qAa zk9*IYnb-1sz0NlCn%8WZN)y-hOxiL19C_;>$N5-!bO2|YT@w~C1+msJ^9f=x;Pu9z zMW2#yQ_W##!L;ugcFt zVbpfyf6oyPw@%+gI!(ujljZ*X1<;ivPpS_o!|gAYMF0B;2=O+?x5w__&R-wZf%IRb zL+Ev-TiZVTIK(zC1C@U{pR5N@*Nka&_ra}C|n_}Fi zN)zdm|BaIg=Zvl+AFQ@e|FkClqtA<3n!@yUdFB%CBjurrP_@1Yg**`;>86{o(fAqU zM0k>kf_`6L-1mcY;|PA@yA>QhEWfmxhXnO0nyVIC5!=y(+O{1b@5JB!pU2qKPUrqP zoim0>B)+2(&0e2Imue40+FzE#*;P_-V1zSY`P6k7U#;6R-SN2HL;rFn;l_5M*3G!T!e|IR9ZW96$LJ`c~~s z&RGwEYWpDZ+9EypJWlPQ{0Ee!#l1)KEDq1)k9hn8#(yz^Am>PMU!%=e zz2{Dmoh^<3b3&wG_=#$|dx&2eB2etw5ioUI5S(9n4N?Y-1WCU}kzKD1Qd=p;y`v0H zULFCPGp(pjOB4Dayhg%CM5CHp{qtUx!w}lONM4+`B#gZ`3C+4+(P=JSbfL@yaV+&K z&TnedJGgs?ZvHfSHq0EZcjnP=qvw+Ma@SDCJ!AM<6hg0_XJC$!7ix8s;c%9S9D=P! zE(80O;=DRamqglDAOh)Lv@2i;TAe1|pXs&^oj;(R#hRBkx-sVf8V(n8$3dR+c+#El zoIiA+D_yJ3C$`1$WcuM!(OTI~+WWMMxO!)!43!3m40gkF@IB7fnfHRY-_@J$(I_b= zJhS&ORMo|$ljT+=y?o`;xXjk<4WY@-rvO1MHkMl}+-O%Y`8!75T%oy(t+ zjL5f}9-OUq8D1t|yi55GW47`|o`2B$N2idgb`1G_?RHL!YBbrX6boMK9KCR9~J(@l>$ys)F;W8?x5+V+r+;wz1x{w~|8%-_6YhKp9uz(RxE}T!X zSZwgvMEyinl$Tt>Ke*-z*=m%Kb0ISljG6mjs}M*%r50kf{orcNGrFq9BnZz$b(bW~I?Dq!GP}!Y6Nt;HWcJ=3WW9!(`?U#!x zJtjhc$$T>7Rw0q}?|;v{Ij4~+>QkoI8~o@L!4B&4(1@dxJs^fD7tc{ts*I%tdTAo* z+5It!Jr4)Loq=chGfsy6CzqJMaA50Y;{8PZpPn3^z2;}g@im%sZs^Gzt??9%8C`+4 zjQGqi{L_R^*nbvnUjLb#zWo<Tk5&IX?&Sdh(D_$kt*AL~4jov_6`B{vtpABFfPpqVa-__82ueE6J;HC7##|eyw zE$cBrtq^B{I6nxaB_oe#h9a#EvG_I0uk9JVcv-MQA z;U&k5Z#n`EyJ+!GW>#*lNZmO+)vKip6*qZ7&oBg|HuXP;5BXwrFaEaZ`F=T4_FJ06 zneEf4CNIw0wG(xG8}c8{|7c}?*h(dDyvhFksg2Wh!}U|7<>M2wJ<=S_2%LjT_f?^u zx+DK+@z$=ZB5Ys(D@(*`_u9}Y9){HKaVX;y@RZalE}}YHw^41CAXX;c zTY(p;mk-I_{)h5|kEfxpBfZ3V`!>PGs2}8Wv^hNa@D9xg&H0ay`MIB>k3mg2PFJO2 zieI8gFXuNpqcDoZyn7(BwjB#HD|p~6>i~;;quBXeAzO52MM9mNA*@^E$l)OUWh8SV ztA;W8ZNxO+)1p(pjHIUo$(%e2ch!mI2PHC1d@YA(hV!XTFTdS!DNViXMCA{EB%S96 zFuVovG&C!P)_3cZL$?+}>{ zA-1I=XVQ&F+4kxypvm_eV&#kSoF{&F&_`}tqKRq8Dsodx95>U(!%H$(qDPJc>{Jg#b=tL;7^muHk8HwoLNre78?sntuTjDX?-6YgHhpU%ebBxOMoFrf12t-&G zpgsFX!&J`*c&IR!la=K!t&m7Q^rX{wk!B+Hm@R6}n(Y09`~QFa@=~vTLg(re$p#rc zj=o!yB3kdXj7-QLh04V-#n$SsXxg~toL>2VA0xTAH`ISsJX73f3EEM5(5ik55>pTH z-yQjZOgo~Gdt*Y*$b+-#=X4Y4frtIV;!c zLT~D9XpG;Ej=x-l_>T$A)LhTWdGzxs8nyH`nIAQQ(`(L%iLmWL94TCGK%K34lvySP zrwU|{!mB!}CO4O{E-NOg)fcGi^$c2n@~(*g*o-PV{X|(sugLVhgu}sp@@En_my*EH zRMIGM0x7xMp+8gJp%YVjMRm#J(S{xN@NQzeC|WfhEl(>zlk6YKMv-Sq=D1%;BpQyWEpl+Lnv@nX~T=+IRM>FN=- z)Q9(!Bpf=4#?%JT(C|(Y<&*>k7fsN>+cO|hcQ{Jf^fpVRNd*U;%z5@27I%ob@!mXUR=8F>w}(Z2pK2Z+V6;iT67A*OukU zmTJYPNNa(Jt>=!PcC;YKv80{~b5V`zaWbkRjFp=`CuB>QypYbf`urc?CWSUqY_ptL z%vUB+1N-gzuf6xKZFF4gLn8Z2pKZ_5;QgI=4z1U*Wi~qko>1q84mHR9+K0oO-u}J{ zKi%I;96S=}7Q=5eYK8@Kw=b3Kkurj!Gv{#Iw^*Ed<~j`2=z_TiZ=+pNN72QNgTQIC zq@W}>3S08i@V&t!eZm)jgv zmi`yjIvC&{BoFbUd%$k(cl74+C(%*wMsl#$l@3|F9I4iZ!)Kd)@aV&J4o9omZ*j5w zYC-cV7r}x*hXlR;5`v%y&E$p8dYmV%CV2g2wSYumcI^^RGSw6N8DK%r${7L!6Fb3R zM+JdeKnNC}6Z!=+#35r{wyn1*Z5FSW&ul(RZ_7`ElgJ3Ks5}SWr30|b{LirXm<#yM z3&jJj9u_=&9EvkL7vV>~y7+!sf1VqbPt2Q6PG`j&Kb+p8&e^?Fbe#Im3_>=0rhuZp zhPm4@N3^#~MABXiM)E(0;1zsXoST;ptuH3P0r8mG^U8E;cg!Dc$}IWkhwL@MTM?=( zIFz|c5ayvLuw5<3>7`5$@WZJjc@G9~6+d^F$yrUQ|5UJv(DdVyFdpW#HKZ9s4cEUCiqj7)?AD`ai1g$SGayYLb zvVy(Zx3T|_AdW7Jng5b%5)q+HcV>yu$WI>la$z-&&g;R}e?LPdF9~pW`9_u)%z~ zIM!3oxPZP_(5LxmF?J}-16heaQYhbh`*1#$yr)7<|Ce98HiKZZeHh$|_zhPFe9hs5DYi7Z z1&z9e@Y_lb+r;(HwXpQONG?0>CHz+$AGe!t{FF-8P7Dy~wuGR@Vk@w%ZA7V&&g6V| z3{^G^p}eKs+^O)^sIi>=m3kUTl(G!{M}z$xc4&#>p5{p8hx#PD)FOu-e`r&-eG zv*WN7RuX6i*8ysC$7WeMtggJIz2DIMVJZ}@^T8{7IyfI?>;KUly}J^(Y|h|t*|PR3 zs4-te9``!JOUZp@9F?65MvqM$kOB412_l6k&koSTTNHC~ZTDF?P^*Qq#IZLlkM`AB_*Un7 ztn?ughYt_LD~gYBHc)puPdblhqx>g%XyZ`^)o-srbv>pzKbD80p6Ye9*ygCH`sFzK z(cXy4dukC$l}Z?WXF4wSe}}!x-MHua?4jlykHm~)VE=hE-kuo`T?W!P?8_C{6_o|$ zGZ0vX$e>jkAHi>;KeiO;!aSx)~5)q`=g3(t3^PHI!|z6F}I&L@3rnd zocwqkJ{~ZZ!+3W6PLh`O9CuWP;rW-(;#&u-@a=5_=H(rS z%Le2|_m2Z-u6e}iy`|ZO4p{30f1C~@gXs~-Y-B!K9x5T;AGIAmj1B((d?o1;o%!`K znRE3Co%~xHc{F@PcJ2bq+q?sJf0%-=drrXDBhKPe8^f`Z^DcUNgC|xWPvIAi#M={h z;huU8POq}3a&UH=tl(P7IoxOb2-mh8TEQr{dkK!Xk^ycjrD&IA01S(55;?>q zVmZBY*lPMV&OSX&W2xkF7iPQJIL33<91c&deP!ekP9RP#iQ@H&&u@wSLnt*RTATys zi1|&mdmw8kg~xw7Mzp3D&{;8`S$`A?Kc@2e-iJ6pV~=$!>**RrZ_ZbH#ryv@rL`c7 z_&+Rto&jA%e_I5AwX!uuvAaQ8^#o_@8^_YAL~k*9Z1I4@ENopj;(r&w+935>CrXg_`cfZ$Qj%#S-3_In>>cji-8sXyEyXi?c4*yonhq4_Evuh#j4nHzLJcydb9ajeVfPo9V!b`e_+jS?ykyY~CcRUV zqnmqrIJLSthf?wWWGo-Hm^7vE@5^+eWxXfJVmgA1F4IA5y;^$>KYOngQS1B5@n!4! z3g(g{*Ur;jb3SqX1J6ul@f5ag-OT9r-R9~(2Ck-aUY(+K>%IAwZAEm@-tXwnX)`FQ zf0rG;z+UuqlO`!Pt)gGAYJ=;5ICS;ePB>M`t)b!M`5?}p3K|dL)-72Oq;bzF-iX#B zrrcFq~OrV=@f z8bEcO1YnYS0TR0y+PiWjtu$UJGWvXzxJccg<@@^Y_pVwFq|!wEz1Y_P5qn1qHO(1qefS4eyNGk&z+Nb2ML zn_SiYi@NGO={S>tNM7E6zPT!$%gCiO#%mb*?%;iR*{cgf{|XtY4~n_{VVDBe5b6+T6`?I}Z2amsYmrdG11;t)-J^N_0V*iD>gNYVH=N{rn6 z$#7uWC}OayiS`h2PB!6y+`HpVn1x4LbKqhFhgsx^O2^_BI086)B6zRGPMI#FippC-*F{>^UO|mWS z1{%EkJm_?4L&~wUB%(M0`FRAgdJ2`kC{kIa6ma~T!g$Ko(4vw?gfrAYWxfSSKN7OC z@>#ONXL^XwUp7t4<=SwGIhu1!O6 z%Sscp_)9cwOnt@S#?q z@l&FDeJCmV;Y>7*z9FkEs*n(x0&$wp&>Q$lH3OrW$XmZ@de2Qd>KGry2>@YC%5aGP zp#wt_9-!`wp|sBL0C_xl-+#PW{>PqnaJJDF`&iZ;eJXW$JawKshprqP#kG62dK1iF zE8wfW47e%s^K3}3cNVNsbhb-U5W+LP0N zMzlAl-{~oq^lo(`?QV1er^ugZ|M@u1=1sdLs8{%FvVZ<|npd!!7@xRHUr2M~h{9n$ zj#O@NHyL0iOPKn*V5xipemrw!=?MjlDOx-_ow}>paWX0U9OL+~d_T%)!@3i8z2aOm*~yE= z{^`u=KRVEl-oL(pdP@(1b60+%UQaJh_RdoA`=Xe4=JQz8Z7>zgYhyVqHmYh+_019R z$#4`59_0=Ve+}VO%{3d8Kav4wEc(?7bdoM@W%@-akM6ox#nsC_83t&a7u~Qem}y!Y$rz3_W42V) zQHw$KbXa&kHS%dEo%z7;H@MeFs+hl`Q?XmNf&LF!cIp(8I>;3StTaWIm}oidB*4wPrE^yE;J zaT3hzrY0JZxtVNGSj~)gDQBkiZe-FA+feC-T}-Q;4x=SnOy!l&F}2CTOj)lRZE(E^ z!Sa(}am9M%XRwy}9xstQQR)oyH)}f6M5IJ%IXU$8s<(Wm_Z~IhVo!UDj6kxFCd#-I*VtH^j5kpe!wdWilaxU3QYVMc&O`~aZTe9rNyd~J zCauft%ZlLYSb5Akn?TeTS%NNX^kK%X&ZR6ZMU#xPSL)3Cl1xV($NH1dcZs6{3LYBwY zKU?8xJ2!vB9d}$3?=iF=nA1D8nL=^CwSu})w9@uxq9u6Lod*R*;CfqFn&I|)) zne0f;CaZUDWR|ZQ4ChTvAo*kv9i-#JxH0!>$txZ+S0{xQ*-aLuyA=~BsXHX#MmTgR zltbCsQW&`FF1i%>k;r9a(LX#Ny5R0~mL6}>Wj8uMWf#qIGoue>;_12&8>YkR0kd-P ze5S5qJ+N}y*iEKJe*&qQ$4%zfr&^|YQXq5I_!oDMJ^wAniK)VKIQiIPmag$UOYS;j z|L?0j#Nn)E^MOCY!!cF#+14mQhT(791eO!YQI?0 z4_#4oqiYO0aQYg2iy1`+)B>~Fz@9mCa20!v@s96&#l<>({5FJH_Ck(jcn{qi)uFKF zusY*E*?ZvCgbu*G`@z4D=!<;mj0l}%QDq$MmJ{kd=yM&_oV6w8_POZ9J}S)|C+ z4pn5VuLaS4rMA@2Z8WpY5OMOf4Q;0mzl@kY`sZl;Aq$$|I*@|WX?kt?OIlX-o!%`F z$LsC#VLoqlW=6jNv^7SGe_YOuMvqXVC+|$B3zEOk%k7F})rL*ZzRY- zPi-eL^;^y}u8AucuVve*sOJ@3_I4<9aN;7auVHh+fByOT5jUR9*-b3d|MTl0^%?ktpK?pEtFRx{gZM`Z%*Q@k&VyXmR>9(1IAFmpw` zMrPUTdn`ZR8x;i@wr(9&*doPT*{a1XKZ%$tH#adu63gg?p8>RIT^n3m-o?+aucV3t zD>)f|_0&?yNN>t(n*^*}N;fR%wNncqEH#Sc?-4Lj5j&WBaskZQUM1#rj=uRB?_4_R zWIRC!?HK*yQ*=d10e`#nQY6`>iXJ)5q!aHPq%Fok-J4@L|4;F>riXmS{f7g~Bmc>9 z$T43_$p~@m$yIYk@ra_07yKPb)ZG5nwk$>UJ zd^&!V>P?=^%F1JDii+xjVCliTa*;RmCbf{YQ&cK%G}UZ;?me4~H9+2>jz zRr}aP!RQRFy?lwY6{{~>j*hMoU8o$6PA>bv;rcs|Pqvj?i)n7aDNAPyx3SA7yGeTyzksO+PZTj-5nT0e{UYh@!x;-2o1WpmASET57V_` zFRC(QsEl6!JHRzP`<#PisT>WKE?a69r$FA<$sqHkuG)uE-aqDiaB13c z{x^#W$p3vgIW#jDg|6e`X@om+xqat28a7%VDf*Ox_7AE>Nv74DKkT+12}f)mf%f$~ zsO+xzetSfNu~+IH0a^-E&b!1 zCm*cFqibHkXI~K|f8u@FkH+^FXGn?rJ~%vkB;N)j+sK6ZtLQu0WtTm**VYdvLuXh4iTOzd!%O zXuDA(heh$|AU?0NhU1emYyrM_NeyoP`pa*ebpyE_yA1Bilxt@stwcY#FGk)Q6+uoz z0G89V@!{Jxc(?5pu>P5bT)&!fSS#r~LwxU3Fbq|~xO_`-j<77ktNtO5TWtehi5qyI zdkkFk;N$UZqtSoI69>PeXqoc~oaFHk_$tz%5pxDfInPCv8;9ZB zGUcMk#1Q<$^a||t(8Oss6Z!TkH_`Lr&!nqs60>3TWE^zE89$iVjB+-$a5720oB{K8 zWN>^+OV}C z3c7oLv(>#w$|A)Aso!R z27abPAWUHyKG3&?^Pfc#dvmV4MxqB_PeI4(iT}yZ%AsX86e`lrh|=Hg;Am72bmkxO zaz?9cGYNa1E!p=Avj;w%%%8CF|LwaxM3D(9ED6TUdJhu8Ve~mq9lPd^!$kv}U{dK5&M#`#suTG!E1@jRjIUQO0Vt%Gl}C8M z(nF;G>A7fT-vqe(@(S1X>Xfv9vI*Jz8!O{h^A#*BAtXK*Oy)}fguBD#qsL+2z29(L z?>U;29s{#hU*h^+d8NTG&RZ@jci@3z?OpV0@g~k*3ysZyS=(j)1*-XzDu;1x8!YtM zUtaz6uQ^-FYPkRGah`Q7PWQhJQjJ@1)1NW8W~d{+$UKK*Me=mc+5UMFt%OQ`?Rzek zktZYmmiO_WLPzZACB?f^!ScXD=rwtZtQ}`!`_&z=;y3r+FJ$#yH2oyBXkA2Iy>j@} z%NMX@!B$B1Jqd!6hiF-XInIJ`C>Q4-JS0^Ig>N!p$Q4bI*IY#w4`EcycNBhW1zai0 z5V6OLn`}_!00T~!*#XOl{SNUSx8G~PrB#V+KRpYcc1D5L=zKVtehtQY<+6D5MtP4Q z?f06{3kL*M6)GqpX$)m)v1QhUJn{ZD)yR67sc7bq=_E_VE5}ZE0VHj8hDE1rI6m9Y zFBN^=uL+w>i^#R5k{o6zI$1Ql^NA>R^)P&{LIG&j+qhOoh0~xb4pGew7qt|E8z_$lKkxSqjy8n;>5-ysfb*0I0K+Tl& z#-zc3#pbyCbOv0wmjJhgK`77a4JbJ~z(t+gx1>yxIhp3(Wz1um7qa_1@&4~J$AZFS zIsj{lj`0tSw@2ez0=V|`JDN~x^Jef5UxKErQ^}bcFCL%q73W?XtRmiXW)YsSLJL3N z7=nwW0>x{aw^DcUzRBT>59cI4FM_qR9r>d+3rSwAJV&dfx`@-uPLc-?6@8Gx%?N7! zV+q-0GXyu@7=!il3t_IxL8u(oOQNSK;u(@z{QEx!aDD#C`o3}vIPFrwm*`loUROdq zpJ1m$cKY$b!F4sh{&y_LOG#4`*F7|XU0*Gb0pmd<6lUFWem{7sgn~J#e|ML!ZI}Ym z*LYxCED763W)|`aN~~&hH#P@eLzT zyvrK!NVUQIUPGL|zMZ>9B`rkcR^wfGxxE*hr$)l?1Z$S2FyB3n=*DEyg?}H?nz9uz zDK`zSDsbNhya%n$Fuk<@-Me&NI%muO;UkpU@feOgEd}qtzSw+w37E}#kA4RS665h# z;l+C!D6&`yKYRutt@sjT6m$+fZ5_xg8rF#1#uh)?*KCGd5seRQ8ddHI|^e zzmRWXGmfsvq44PA1sHR8IN15i`{YTuR;I4D}@#q-~|u0 z@$wOA@H+S=Rkm8k`iHP#Vl6+@Nrw6G^k?=St?T%5b`*;8ZsYX(ShEPdRLG=SRFZV@ zRPn`$h#&Ja7^48n`NltdCmIZ(3i*yq51K)ztvm~HdfYQmSfIqsi}9Y#J4b#V9?$Ie zw`I~sY2y`A{qMvFE*RoLKHJF1<+nx3%tCZ)fq-mP*~!^X*Qg!w4hKWyA8xFIr`xKG zAB5kAnx=5jF9;HCh!mQC+5eE_9=m{kC66FM;@nsZugj4C$;7itIm+SKAYN}aQ}!2E zKfzX$1|PPgZi7o?num-fNW|E>G7?$RQOXOdy_O#yKTAHWQGoyf`hVQC#y(>T@+ z!hrQ@{BO;MNO`RTO{#oPI%Cx7s>wTG(Y-^`v_n8&Eoj(9WKZHS}JWMb2sx~_WEDR7DFHpw;`cefBnVX82J@e&faISeBxuA1LPtm5Qmr|Fj(zb2ue~yy z?VjvbKM z5Z`ekdwm()_T~s(I5-KmG{(WE=BIQBy+YOGuThnUXT&k*5nz5*nxi)$aU|7hlB7wS z#zWg7UvWI_RBG@?2gg-@WcdguL_C4*&Oz9%bS`{aG7P7D;r0dK&3<|od4^_DtJ!Ao za8DTwUp@&+ClXY(>kyN%!r^PImc0fB9F5_hTAP$CUjlM^pzwRIFT>32zt&V)3 zla(RC;4W>r(r%0sHXlQEmfTnX&+lvmQa+N13P)4ey`c%z8rH+4cZu|9Ng8TUEe02f z7LinoKKBf0|D=a!Mkt~m6GqV#->D!gyao~xy#%fEhOsY8pttTeRiE3<-BZK80{M0o z;<&LyXA1Yu(VnKY^dKrgW!q1pgqPx698*J)t4jZzHH$G@LI#h-<|Omz)?Xx3$ zmj)Am$;DvNbB6@!w89aE!FakjFAA&E*PwMG`xOU?)1!lQ+!;6eLLwO{PcA}??rJ1_dUScgg4^oVS2lyg1c_rJ515h<~-CFp#Z*}Eil?Y0j0j@_AB94U1%i1 z6OE~O@glxA;2yM9^O+wzLq+Wa9+6LLokgRXO|aIX{$}{Z$ z)amEpZRiMM-5f!ys)t}_$-{82btcxVmcb`a37A0}Pjh|O$?l|UMQ4b)r7_oUhM_OI zxDT+L*E4+JKqxKNbwba^N^rqIs3OZtb6^p=*tKsO)=C&iT;05K^s! z2Y2+(*M5E0&sQez41d{oKl=VTH@?MTlQ)w<_SlQav!W17&5{yK){4Y?@?D7R>1g&D z$@4Sa%Pjq*&g@gYgo5odP-u8881NkMBh^p{u&Kcdi~Vrk;76!onhD+^zDCM@zJuoE zVf5acDd?)l3{JO;&6Lc@y9tY&&G1wuZodRx>B0--!lOw78HGiHZ7siWJtH{G&8Y{nkimj+Q40>%QB{dRbjrlz}XVLF5uP+a(YzVSwJmr z_0LUgo7q1%wtGf9ebX1l1m4<2%iI3DS8~42>Sho)k@*yuN3Be_?>OOrOIt-NfBk~> zwdoM^@CPfqaK?&UaQziY3+z1w@&7KBw_ggI%l#PUeaa(!Jd<7r1t{W zzCxK9yTB{;9km}h1UFCSbN7hrg=NU)bsB%r=X`ivvzrF>u0i7u_J8Y@UJ8RxXY?`d z+e>q$`}dXB=?cJqGHP%`;5-~SbSXv_7ttQ)6oegQ(2m$kBxBAUQa!nrthM3hQ+byI zGD&;)1Y8+57?(ImFrx!nxo1vE!e_GC;t)B%KAI$%$_grt)&INCPyX4C%mv1b=b;r; zWEMpCOjMwSKDk7;uLX1$%mH2VXLL4}MhfG_Ya>M!oGmqf`f&Ojemf9owg*c~IKws= zyu&6*jfymJszNEBQE1#X;;nX zbjji=A&<#dx1n5LiN{?mzHD2rJUz1y+{9}|s>f`^_`Zc`(X>0LP4+O#axf<2T*Tw% zQ3GLJtqP6$`av}E`B457X91R}b>d{3_qm+2m8Z>fmbP&6yuZ*PeFlEot)-Ulp0W1g zt=c+AaR2@kfxkx@c6_!Erwo-tcWuto0<0&HQXVM~zU^PD&$jjbz{R)l*#Gi<$2fV} z{|eO)(1ev^sQ#N(^oKSN9d9lF?_8BvAMzYJMxVk1{iBwuubUg`agWcG@mQ_j=yOG8X%2FGJCU&`Vte7#7I zqy5XF3|39+AFF2h?)~x&MU1flHJ2pFdbb8Dv&3V`FVCV=A;Ylbs5e-z{V*<%Tu-VJ z-lJlVZB*^BI$bUOnI?iMbEfADM_VHuqoWnsXqeh>@<15K`W)}YDmPf+-X)s2GaAND zPZG`D7>O?%ALq`o=QXA&3qC4+!I#H>gPKbzkooc(YKBy+EtU>H&@Ai z@mf<>ezwowX<0CO_%i0y7zGHd+(EJju0@*j_QD3a8907>IrWY@1NR0QLhOz8Y#-iJ zeeT)B_Wl2rcTW&JFJD6H8{{e5)_jsDuHL=~|9%(7^=0dRZ_5C?3%2-?*BWv+X%W6@ zBOw?%ERgJ&xr4(pK>rrk_ese$wvSMyXaU%*h!@$m>u`G3N{+&2<~^u&tqfr*2gCOa zJ5()EhD3L`_i7%?<9YOIxY+%Ky1vw4!i*l`;d@VU{6F1j;%KsT*>-Gc+b=iu!{-xoQINy0I*GVek3{hLqMDTxk1bikmLeku1oN}q)KfGsn=tIEFi!@yK11*-m z%QundQ!~F-GHDiPLykVnW5*OPSUY|>9#;AVzf_#f(P!)Tj%or~!wQJfLExkzOFzDr zrZYno;Qr_?q-Qf2jwPo;O?)-FdNzuaJ9l?8@?3b2#QY4UH&!XKcyT;<%g(`rs7i=< zz7*78K7P7y7AQY2p$(CWf?v&=f;W}p>BoVa!L0UG&dXCzke-hZ-(Zm%6Em=X{Ocbb zrf@i}mfM6+Ov)35)n`$e<4f?5qAhSo?=>n|_8T32$gMf}r0vNU^mBK4gjqQ?e}6kXG+;GVN-34yd_;cDi(_HM?~=K-~(IY_4`h01@* zq8{&;QE`8vTS+O0QElWpSWssR@{wEN$<0~7KQ4~HTOkeQNA}QETXSZ~E>GI5W{xhd zTEh0@)f%;f;ngr`Li%OV^z3xm8BFqAZv00WI7Nxsu!+#4 zt2OcYpGWBtT`SS2#oE|PDixL=Oa#*`S*!-k0D5*%vzo ztcuK;=7LN*WP$nMfPT6(Jq|nr#Bu@Me%~_IkN#IX%>DM5-yf&MdPDv)y2iWoT|J+13425u5;sER4 zywDM`FwQ>>9p1=i21Vawn)h(e9M0c9pPRzbSJsfhYDMC;WPO*&o!W3*G&vD#Y9)d6 zQD=~d7J}gF{T$^jI`F{h8jkUoXXan9VEi+bm^`!ua#t=wZL(LzIdD-7+ zzSRnUI2C4$+x-!~P-KA1AE!cATr}(mOow{2nRwQ*G#vfbgmzU~kz@L|kjD@ch`YQJ z*1R@k#@(%;_luL@V$ujoujNw1%y6`8S{`!YPp3~~r6~VOIptUO@!ti8k*$h$l$B|p zrYVUM=es&KB^V#fw8r}P)GJz1P8%`dni~W%g7yrxhA*T;#6P<7X0RMWO+7CB=X_T-izzhiGv0 z7St#~akDRd({mN&_^ASKfdWZY8HBSQi~*m0w}4lePu`dfW&X|+@2@#x40Cdd_}l$_ zD>bsN7WM9bjYgE5CySOVu{sOmZ6CqbL7Sng${8ZR^v5Pko!f$Hx4so=NsR}e;}4+J z)Prl=`bi5b23TP8_d{{%6gj+g*=er6?yWfY;O+>hRuJ!DF;Yk`>!h-B^6DR|qNtC0 zNa=MK_Z@gb`7_dc?!oA7%c3pUMlsVfEm3)UH9dVImTfPLY)qudk9vqY|0}5z?>))Z zv2@sS;O@x(j=%XPq1JIG%y5H5@qW|dy~@QgZ)}^JH4QnTk9N}edrTNx*&JD|rn=Wn z8OLdRsIlH5Tw__pUo@?bXv+^~Tyz*(mTb!U32(El9DnA04Jvo$DD`)W=X|HD<1+u{ zVDXxcxownZ8q2?LnFt0aUy%pdy5#>*_2h9i1>e84tF(~zvTs?^_GTtoDw0Cdu8=+Z zk}c9}->7JlP?WSGQg6;=X_JKPyUMOm*_VXhySFdj-}jICym#-NIs43=J9Flo&=Y?6 zals*oJ{JquzNX;0lKJ{#n``*B?Krk%)jF0l?HIHDF#}9RLRM1M5BIq75u%qGv#wK) z@cC*fDwwfMvS9p-P7jEMpv+!)?Zz%~wr&o&;`SJG^qljb3B}6Juu>`>Oz0mpe4r$T z*#;xn81j@3-m_KQ&tfPGKRAN@zW#t}FWCt{jw|s z4e=nIA^2Hl2R^l9IR{&GU}h<`dOL=UK|RUP+5u$t?~CHxW2We3)B&s*JqlhO8Gtv; zm%(oS7{5CsiD~#}J94?7k35_R4!LJ6`L3M_d!!-*&g(q9(olCZRdX?r>hOL|K28~mZ1T0=|S}3%^GA~ z^#;~U)~RP!pW$iyi@U`3DQ#YqC^=`m9s~i!=u+o&*qBj9m(CX97P~!ohqF0(R$j|jxs%_3 z+em)A7fp!UO7#N0VQSlcl;Kol5lfOKKHvb=1D()@xfW>cr{Qeh%eC+*>pg6f52AhL z-l6v@GGzCYspQPSb?|$3G(7suulw_Kj%3QPg1)3vK8>Jpak=nU9$vGC!ttbJagSY} z@zO>kygIoUjy1fdb`}H_SFM9v3K39lHW8k}D&7Xnuf6Bmik7|J!T)mq|5QYZr@!fs z&s=tc`$g%fPVqd5a`l;y+yL_D;7MHZB#Zu-C}3Cnrb_bumD7%ZO6+%kJ2~U2L#z@8 z6JP5*Z0D;_UMYH%PAo8h=Q#<`bjX(}zmDQ@NCgCAt5$-Z{ED`)Dc6Q)=1*O_G5=c&YDXFNLbv3vb|R$2$FAJ&c%yfjEc6~cR_ zpA*NFtD&mdIYRk^Zr~$)iv8Ywv)ns33~nUF{KMb$M2k$@uO*yuwGy1~48`AKJcaV# z@6ZjCR8VQiMzmDj9&MEr+lQv%4_*q!jMh)g4#&_cVUpMX)jb zEMBp`k-oAzgv^3tu!%!G^~o&(hXt=euYWNwi~U`TNzPhB@HcBZZMV&bK5df1o!jr= z>hwCm=DWD}0&h>NV!W{X(9f`ZTn~_0r39-!jNcMYw>>vH4BWeYpW{!#^%e^FXCqE9S(y(a|kBRz@H z-4Gs*qQU^QJ8d!c_=5TCh2Q(p3saWEhL?}1Tl_HCjv67__8t1*DTSvNS)l(^z9hzUE;^f^Cduh;&(mj7@AKa_C1rbg=3)GH`Z=f&tmE%S285 zXp0{%S#%eFE5Pu$a3?F8(ZHhK`Z2{e4OThO2rl+#_^0U_{8?!?Zri57zt1@tab?ek zBj6Q~1aH^})yOUh@_y_n$x%3cJ@iB+o=O zeIC;CKzVSn^I+N8{n_AIyLf)P4m=hv^ZUi&5^*K%PEYAZpBqvMVcxJXEo$0zIu@J>Qto(s5tDQUvdmNPG z#(&Y1W!zH1oS3yojy?lc4_i@ttHmvhD&@_(;aWv6pbZJEi}@A;>ss0=a>v|8&CL zKcw!!>vPA0JDj};boT^_r#Nczb*}DC;o6E`_4!Qqg(@&I_-onl<=@0#r)l9t7!GM! z&(H>&*-+{iPQ!XTGo^zYVZ;RuShqh%ST8w;!nJjl*hAyDFVr`?1~*7#_kqq}ufP8Q z`MHNBdLM{A@0zmt-zB-u`q|MCx!qJPdNXvU?S$9`t3jvgJ~Z3R2g^67;q8uAuwHl@ z$qOq%vG6|qeES~#Ib;fi*^a=TyB5>_3XlKsDR(y)ooP+|cg^wj|BJE7S$sHT8$QzG z67LW6C21Zhx9vrG-m1Wh4t~QK22+T#i;%p3Ukv5e(QLuD7U2=aM!M9z6wfc;!1f;6 zDJ=e0jlKIIXq(^dv+T9=8tpmc81`4QhQ+HQdEQ&J-T7bX4a@l7sI2XH^>tO6wdNN1 zwXK1aCkE8l;8R)E4pnw!%s`k^*+%!xjDdRhVtOgW6~DdvjF*4T;876%WEAG)DgPo{ zSh9xkJ~c;cCtxex^{6zdjA|)e!b^slkS9y6NUK^JZ`&8Po#+LS?ZMK0t?^3tfj>f~wQaO>yS zVEbMI>T7Bs6nuOy{(Qv-&O4kIKeZgr*AWS=@f!5sarEPwbtrM|8pzwL2WR~xW9`HM zWN+gH8F%x*eWbZC;lp&XMq)3q;i^$s&A$)%a!DvW;g8wdP)+ti`VJ>EUT-UJ3Gnn1 zZ}|AggNE&LM-N`j!%B~{P^+wD9pHi!`_K}OZ}%U==YP<|ueCd6`b4(_u` zk{i$BDPB9N4XfKFVwH(MdAJjoOaJ3*bl=ZK=E=k~z}C3h%Ra^v+k zc-|iRs~`jt*2xPeb=;=jp8j;u+YnkqcZm-LPGJ!;gCN$B@$iS{chD)9uFx|ypEmj* zgn$oByrQ|=Pln@bux5X1m%bB5os?x3p7ng2EAMsDs;fO%$~ifvrme>7gR6V5_##pn zI*gSb&tcj2duWJaGtJA>X1y~0(5uyZapjF%Fgj73Hh`6Zx6ww0u6L1eG^zsFNCzkzC5qTr|OX6-ZdyAswe2_ zmw2y*##pY7~E_frr!Xe;B{pa2}!GwJ2H= zzqNAcR1%%oy*G5+`WZE1#n9yL4OtNd%5SL ziRwwn_1053EZLKFJ@5_^egHVw)pIfu8ML<1ZJz^izp$y~&O|wM?65SC=kDnH^w!Ad z!l-X~@S0sl#W~eNNvdsd&e;V^^d9qYZ_IiotT_=&)lUK}T>DEr$zT%|U-vFu^wN}r z!OKGS+&MfTmm&RK{5{dqz=^PA*G%+z^^~&dRQv9INK)hEK#!&$;HT4O-&^lagX%jI--r9aMoa}kzAErJ9w0n4#1 z98ZFlp%JXS`eoP063@ z3b-UHm*ROpEC=Tkm{sbJra#pX`WJse_1m`KD?KBzY#%?gTuu{yCN|Ms5eFzUC_`3l z(Wqb!#z#l8$w$;Kd@7!49naaR z;BwAw>iR}N)7E&RB=;`a_z`sydz? zoOzBK1rNaOGTViT7N5)F9(aSSlRcETXhT7h4~+Cmp$cb{h4T+JQr`=C=>39&;!m|D z=;GRJ_;GbH7`SbMK5E^ud{j5xN9`>g_-5f<)cEuf9CR4SZcgdTj5_RbZMr#@zo*O7 zAp5$Cr-|cr#gI@$X3Eh*u^Ak0FoI%TJF0)Gng(w37rvh{7w-K`pcNkup=t7I^kujc zzN)_lJwEoE9$tJLf4JHM``_W$4h4N@*no~O2ENH9GUq??Y<*KN*mqirJ?q`Q_b4ql z7Ca>HFS})0WKmb&zp@rRI@Ln=D|de{a&oTvZds}{T#`N@b{TXeMu^LDKUua;_<+v3 zena<6u1n5P3Sm^17HB?Q04lNGyj}fp=w+EJ=Y=~Rm(Yi~Q{hd)K%~B*KVFop4=lov ziPxor-*${Re)BvfaU&P?hEpX+al~~?+^0PjyLnW>i19k~-Rc9vV_R!1A0FZ7%p#5+ zuFd-D%}Cq%5ImC(rNa%Y@EG$DGXJ_N6nXxEZ!?#{)IYJ{J1(53g=-T#Wu+(orNb(o>gJg#qwo-JGoxz~;O`r}`}p$pS` zGM7y!N=vhLQBkBTuWPQ}y9P&k{gn%Pb8Q0jFEXQVizMszj^6Zg8HwOL6JAeX@R|`E~I`a13nsf}Q_dAQ7tVfY| z%_(@|d0qPXsWMW1a!Fi}?GAtLwe$A2?&lCttlmb;oYui({{*~Z+Xg%g^}-d^{JdMF ztMrR5$yg;G{zQ|>Z0q*bAUi+e=YxJ@kA7*yXLc$HkTN7!SMH|C^b|JVIEMT%<4J+~YW&x<1|{pHD9wOhFO-UO_DBoXxY zenZzw?8&2Nu4M5TLWtcyI1psaj3oJ`+|0Fb$qGwIl=~pgl*C=WX!l z%sddU*GJR;xMA-yH?Z2=Nfo>&@p?=f9tn!U@3C9)VG=T{fGj#IiN&@-5$hgk!38=S ziGS`Jt{$fgzKp0-ApwtK$oK2#$+Q|PvOe()a$f=2@?Dtlr=2-rVE=>f0}i6 zoo1`Pra{MIEp#`GkP{)h$hNp`ygd3UXp{QbCwO#C_uk7S{bxw}V-?zQMT(?sGD8FH zfATyRS)b+Vi;#yVJo-MH;vT;(|3uCL<)ba)nY!bN)cxaRXGsD!Iuge^$_!ckt^UH@ zo92Mip8eP`pqLC!tLANJlEWBW-?bL4Y}$gg=9v+dwZ(XVYa6UM?2n#xo}{Bb>k|K` zGUP|5A{?7@0;iL~GpgZSZ(wKptzd8G z&mph9rHOmRVX(gRk)JO*%BeD=ripx?+vn7in5Gz!eqH@=YDg6*?zh9sWyGkrN+Lch zl>#3xO0$5T+n9dc7%=lOCDwQ{Iqf@~#l~Q!9f$CrD%tmW zSw7A%#7AsO@Y*a%-nIsbui(yDR6FVtvHh4t>Lj^S);gQ=Hqv4Cl*;YjiXFq{iC)!J z)K}Cj6@%6=7Q6exuD|Z0GkTfwv|F}yCg>udLe@$h7Q zVsY&;KGI7URGzPavFCKK`-Bz3yXhyewY@dH`E?*}zv2z2mbpOTi*>knc=vkCUzm@6 zYZLF~X2R(X-aH>t$cWRqNcx!v4UnqfamA0oL{Gg1&BGS-*qD-S*perxLvGQep~#MrLXeDM|3K&O*3JSBp+C`*PEKp z>q}C5)R9!P5v0BUE!=yd2*pSzLA!b**Iw|SB2q?^c^*nf8iJb182;I^v9aZiUUB7@ zW2DP>>-S@qGz@v3Ildh`)v#f*qe4@0iXU| z>r61k(txsBS_E zcC&C&P;b0J5)VjSA^#sA-1A1I@p#+B6_oqmSK%=ex+#~JO?|_XRF8n}gFqZ`PjYUv z3uC<%(LAp8iVx{`t--jzS3SrZCF4rZw|G$BZX4~D?~b33V+r*$_Id4;KG>mwQhmpodVr+>0(2Ar%VE=t9#7N>) zPSh&G{zZrI*4Kx?bl$YGBd;I8g3r}R>E?ATS^vZK`md37h9*qeWhHz+$_J)pdWq#4 z#^X1iPoX8W5H%O(PZ&qykIIpxANlxRZ4f=M_d1Hz$U}3lDdY5x zc6`OU0ZVT;!OvIr#kKBB$(dm(xK5)DrdG#u@(~O;bha#Cxfgs)Isk^Psr{aF-r4Nv`Ac|)+yG{;Z^_g=qWR|>ExF)__f3+FH99urO{y)LvhzM|ylan=qOa4q z)KIjt_!p{J^#I#cC}Zggh8Lz>MztHmNZ+Fq$XT-?*uecimpz?vqjmK=IO@bbqhKs5qnp+UEAf&W}rB=qF?0Bgwfgjux(@I|tGr zuOIZ4(G=_%UM+SzwGqZ_=$;F4d`4dq@$h~JnUXW26}az_)40gsAlB48DBk;8Tby+? z0)o~VGoA16aLy71$r*lu@Z2L+ym@6gjY|kaGq%0J>-BkG9}b_uQQ{YSDEKmIY)Zn< zYo+0s+iZ4r^%nSf&;`x5?n@#9XMxn56TE)=os!2-H(tl7kt>LB(|MAxDhRt+D{^)s z0>ihoZ{|RF@q8eeiz{*1+i=J)sHam^o?&;(+o)^_#zl4mvHgSyv^8-fEVLX&+HUS5 zhrW7|wQc8dQgbS2`+`dYSHOe7;G^ODVsF3(fjI4v#Cd;qD7uc1lB_YzTF_Gm@CDh@vWfujtr2ad_?9;lk#jr;(kuC!TgI z6Whhv^7YN%SHK^cg=EJ^O|t6Q8%#_RNb{RuvaZzt&vi}2^L^@x|1<@wZ3zOad+0j( zVG&0<4U5U?VM^ranoJ~@I}9%{4@3pZ_E1*{&{&^|9Rj^9&A&@SL;MG7=g7vq zqL;39_v!0B(opWb2&h(mkFU=g!PEAauT%3R^9>(;e6tecB~N>D_yk;WdbZbU`?%*^ z41dwY`zwi6UvrLr!KwO5Ru-3?t(sFhF*8|-m&xrH%jfLkpV_&+f;At{V_U-u=*!EA zTptttE$i*SVc+OU=u=zwJa4$hQ8YX|6MmfdO#RkG@iZOU-%Jg^*ubBQPw4Z!0XWZ< zuglZZV}Xz5(>G%J{_A|Onqwx`Z_7sm29JY&FLSKan$PmN26R;b7ae*FFNRi7ftvy2 za4**QZ7FXU0a4Rx@sZ-ksBvr+3^?Q?QwrL$7sOjFy0E#EGJ1T@jW%9f#*V)m&(<_-N9n^4k?#Y| zc=;{4FcP*__Q9Q30_hbm4Q%@JGY@;y_(%ASeGm9E)fZWL^LZFJ`3Q;<6v^^O7H~=O z72v6DN4AXI!P{Nr_}{p+BK&{!#npdj>WKTdx6_=>=1`Y9aZQ=OZXwDRrqDB; zvmxoqNS@~16&=W+XOsB;h+bBTB1NmOkB3>!9OwtjF8ScH9$)!)2G{;-ml0mx9jjPS z)As{Cqc<7F-2NyWlXpV+u&#;YhsPuN6`6;qFrz;*C?YkaW98hDLM}3 z^(dDV4|X;3m}?VXcZg~Zyi4zf9H;vf2^5(p@^t^~{{dax^oX9H@dZyZv=#p;P$N6{ zg=1OU1*qJ(4(I7hfxqOe)!}h_aSv@do|pasW9X@4m(eGeaw;;Ph0fMZp^e|F=>GCO zFlgx|UgokP(Xe;hM*K|r3r=y|NM=~iB*mRac)4-(xX72{AwgZxF-ex4Xfnii7W~4W z`#pd+x)@R|M5J|&8LCd&h5Zw!;Qnj8=s;OpGPPqHwHft+m%)$b1e&w#0w*_7fqXuk z3vZ(4EjQ4^8JA(^n#FWt-83ru%7-cRJ3}MUJz>Vo40`%1`X|4~!BcUL<^gV=$jkGK zPYLauZ9wWrMv$ZDdXuNUPxEzO?x?`UDp9!js~_1Ln@cuamA4wH-b(btck;OBA2>~i zPYCB^EV8ZaLWi#&fGf|6pd5CQ-c(F?>Y0-V#^d?tTpg|)d7J_p!sYOR&apgB<(oM3Qgug`pLS?E4C@bDhXPH(Dbi`gSMr^P8&tJjo#21gCwi z!9@v|P>cC`uzy}B@x{9bd)5?!PL!cl#A=b%xue~4*NK0+*X1TwG=t!{8wZca)yKVy z=T}PaEq2hKMGQ-t@rH;8bgi})*_)-#Kg$zEp(4LLLEa{wm4ZNMj0ca@wltkNTQzgQ$u^8oNgx zHrU$GpxP$%K6DqaN1fQ!FvRl$ERFvIyGu>sV5|azdrE9$N-P`PY{D+?zr6{X?`iV7>pC09f;z6VU<=i9v?Gd5-zub2PHYyBH(iT8t5{WCBFRoO7wPbr*N`I8})x5!mJIQ*o?JOpw*ZM<-@Yr z=;L4E#JtA{8875%&F|~RED0`G4;)^8w`B)wx)I4t6API3OpAXub-jNs+g#Kt@r%^L z*=tY3??2gesboL9BF=za{+15HEEP_!qT0*jz)3U`TOK-%+w!~Dw~d#a6z1;`(JKq< z;r-NbWVhrX+L4lh_C=P!wgtrxH}ci+&Pl6mhS-|c`z^O4%pbb2FpAv1`JVN-KPA=&dye1ED@mwa#a z{sDHM_n;@HFZplD7~`4HFwp=PHbuhri+-ST@hRRKQH2wxE`(K=R$}_BKYKM$i_v4D z@bO8e*uvJFb^1(aH^25}c3%sjhkh3gK4=D8Zb*y8mNM+z^mrEA9!IZ5edOpD{Vmdi zM0nd+ZH{)q;OB%0BN{q{`|PxFZf)lO};+R%n=I(iLWm!?4H z1UE1~Kao9KV9D?z!X7rg0gc3eJ}aVk&;W-V*1?vZ7|*#P5^jC2hHAGI2+!t)LuT}N z_VCEZf4b`&V#SnS9$+=GR?JMY)^&UWA48P0Au()^i`R_fd1%WYf~Ll-#@3R!?EBkf(NE)MsGQTi#+|auoA%B=Z?UHJ9rD_t zfH&8M3ymil^YA8*J4i3}XykRs$s+ZtnCdRpWSvfm?A8uG*1JHn(ur=lWXkNbmzV1( zF?f=*gKZs`$ciOvi$(QWC}ts&oWD?Fby8blw^utXHS~r9TX~-_!O4(b%-#AI(m1BY z$zLF|IvdN-Hc3p=R=Dbw$=LzNs~`wcsF_z4R3{j*`r-Frz3xfQW(DKw%3*5VMkf6s z1^lj$A(Q+Cq*t&CPp|5wAviv&3C$iW4Yw1z*9awTL{-h3d3z6j?g346L@Z4;jjh}_ z6O62D*a@!$HhGB7Kl$4j$g!J^*I4$$pHOxyjFX8VK1mYOJw}OLYYJygwwu_7oB^yo zH4~YaZ2K!G(bHMhtair&cJ{Y`tr$@(95O)~8YPM6pBJ{F6HYI{k>$bc%5^-?gI6iA z_p+Zse%(Q~!%&^oJbx)nxO$vTXzo*9_iHb^IwV18rMDeiE3VKU-@L*4wG`X0F^OGK zk6~9A=uo#iQP`um7t*r<+|RwJtg!F_{LD<}VavJJiEoZt&Q1>SVheO%vEN#8?D)cK zyzGL{Yz0pb5zZL9kG^Ychogm#pr#rElRGubx4lzh_$}{$$jM*S5!5MmouLCs8KbF+ zayC+f{o+FH=fd%&Uf?I3!TRQ{XW0MBjpv=4yTz1z@ zFt$9uPN#g#3;z6|AbglV3f~@zpIn#@6BG-C3o09UykqMovvNrc=uPcCCc>NS_{-sPZrElSbG5hH}{{%?$|G~E@ zonMbu&mPRwLodT8&p4Q56i~*seN*+D+RYxw!};>*0DhDg4YDVm(k;6sXAqD#`*O_` zGJOh>(&vHHYu0Tz|F$oKDMHpiYB{_)!>@%4OjbXmlN^eKhFK{vdvh4iD+fEzCluB{ zZxQEOMqv%nblfp^JC@uRT#&?k*{5yJj>oF7{)ZLWNSMrKoLs@!@HOmWAN}%aREF8- z&Hy|ll<%t!iyW~|!%1co6V4i|3P6om(74g-AnV6&INfrWd2H>&7QU$ei&wzm$jWP{ ziLH_FVtgmG>`vrlAPDfA#Y9Hm*_Zn%EYDRRY>_;!nXQBEUpnJ66_vu5r8Tre;yc9A z&Xr}+^WpxbRn+RcI!fKI0kab4bI$}R>4&l0`~~P{NO%9_?r~-R02e4oxe1fH3h3Kj zCy`--2hQ&7UI*mhjU3;M9!{<2W$1+3j6Q&WKRb4VJ}U)$I~vK1~AE2 zXH&mMLWtxH7zbN-=V7Gy+=tz*|3;N+6S0?09CC949#+bP!PNNsA6Svx7h=-#aa!yI z+)L?^FkycKopo7?m_!F)?fD9rts0E%mqwuH;eo<;y+dJCaUM)QBu8ib=w2fmq8R|z z1@$P9U8Xq27G>DvbGSva#kv0@+ifGRqI1$cag@qjc6p-*^SBcT>765B;9)A9l&6b- zEJ#58e}v+e&>{5NMK#c$70JtO&LHWs#?E)x$?-fr=Nf^1Vk0GcMYn~&d`IC~&ila5 zvilsS&;9~DX|yH|k?TY5{CPsJwr*fAAF8qW=g08&@LW8C*}W3LR~$IFDp zBlm~{m88K-9bo3-M%&mBaIVpj7qJn}OVJN_I)5|a~gZ0H>!nowowF;sxYSw*A0`;Pq&J(IJ~ z(VP$8E!W&!haLKNujMD}8>02g`!i9446Xh82IqXxVlF>!K<@foFt_d?ntxLRrwS{O zMwlB;I(~r1x!UYDP26veVm>R7u_ku7M^UQqqmDB#+xen(Xn0sGuHP-@<;Ka*A?_aS zGdd3zx@Z2SN0i*!7su6@^VcT^Vpz2Lk8tHo5pv(TMx3~^0>7LofRx0~;hDiJ#cR%jI*)p%gSayIl8?Qho~ADr6*Us1V_$6H*+E+6c`;YtyG`!)dY zoNd9+$@VN)r=jioc>6I`Y}+0TFIB2E^9^n)3A*#b|@R1{eHGCA`7iw-H1l_3{~J>=z({JJOit`Tr$ zW1^vWb$mZ|{F5)*C5hJ<7Pt>5U5S8B?e4x+u`M1Yf8U1ves;pJ6Ei^btO|t7@;;G* zg?Be#)2wj%dykq>(cg=PNX-X}9g@90`~8-4S9ZXmt5R%j;45BVC*E8|?NjS$-ORbz zzhtzyDb9*M>e6Eky|kH>LoXITMwWF9)`7QM1b9xtelRO2#!_44(a+pebRA2+L2`~` zs_u(niZ;IMyb3S4B}aA+??on_`;N4;F~_$EdsIMhY$VSk$D3(lw=Wa-?4oD@mex!} z$tNb26-#3DJbg0;G+z!wA0v`+VpupnGI%!RT zeBn-HB%Qf@J>0+5?aP_v!}nK#)q-4Mn`CW{!}qteozjOAT8~bi>WyuE-NC9!o|*f( zf@6dUI-a$M-aPse=S&R27P(e<;Ik)4+i@-~XwC)quNa1`n8M48gH^eCF1)_)(?Rzq^0$DY?YS3l839(*p(g@qNd)J?~NlY@vWUL*Iy znoKri82*!&{TnkLmQ=(Z_C&)MUi?{#d-koRKVRIytA;(ICy=?= z{^|vENbMxP6%vlk+W9m6JdRTp82g%Ljb7M*pTs0uB{kX16|N0`w@66yh z`ZlZsjX1@xoeIu7RMCZ=ny9L+uOu&A0I1Lru;kZZy4pscvgFb`L z@2UKK?Q!#QcBu|Ml^RR;4OK2=Dz%@YlgoL%Jb{COQIcIc27f4e=T>~M{z<;^wm zsP0xB+GD>Ctr#auR(-J|i;wjoOP%lW^!1i|h19AIBykRU;M$8n(NJ;{{j@m%S9`nv zr9-f_EI|BV-8X7E^#+|`(2jS7$6{NX-Z;F^M)8z$(PGLhwK%%CG9y2Vj|1n>kAH`Vf(L|%yIf@HmY9ol+}=147E>7bCj+HToq|3% zf8pn(4H*eY<;Q&Swv_`Q%3v*4n!v9gih6!&M?c?s34@;Xht$_cA#ATK^!qRa?xigj zA9hKGYjt_(VL=%kSEYkLxQyY(IMIVwyM`3EzY3B9 zjnbp^Ayvm;i*L~kp&F+Tk%ht#{&(JoX8NhK2lO7q`{nZby|~Gf>6SKudu1svYgxtV zyiUeW)@K3gBcc4n6Pm5J07G&CuHO^}b2HDv)3H@Gq&qj>pOZwaF|L;S zq;{_lE+1sZ!zomngg6==_TC7;?)4$lR$nVqbzKBP!xrk1y^5Y(1h}`cBTlRC-usE4 zxrOaK+>3q5szE`$1rlHN5|o^>fch(F!7RzykCw0ULQNk(+B_t3n zoL%HR?uid7DC79Zb?D@jMfmxjA2jJ#AG~An9ISQY=ihpwfQ8+AdmK;xe(Lb3@d8zh zwSjT{%J_F!U#~*?PS=QboDYMtaa8QRvH|VO@-=Htr|E-35a2VSO!@AyEa?@?6J=OZ2X zD3V@PyMsA+dUq?aaQR1Mb5t20QMp5hgmn9}fAP`7ixY$KJa=_&oE9w(`9g=Rk%bHL zTj`pzL6S53x?*F=yyMi0U_ke}=)GyVz|yk6VDnlU#ipF*_z-Yq{h<(KyrxDhc^bq|R9RFYj*7KDisAn?UVfFz0uPQLTb9yW=HWNPXBRnn3&yGW1jB@D( zNnVfXen%}vH80QxbkTZ zJel$aU9XOU=Ae44uhK^nOZyaF`nd(S`T(`OlS1cc$Y6Wp?wDO1-bP~=jt7zK;{y8e zKfd1op-FUYzyzV*v~_4v-!9Z^OcF0&$s|+w+H?ynHTI=4k&ZJr0*c?O|TdqN!^Z^87X#|Dwisc;9~!M~mC(Zhkj@ngL%8W}xR*~9qyNbxcK-}|d;K}+lNSI+sYAeJUXfUN@-?b4T0mV>W%znS>jsMV`Wo^5 zgoA(IS`U^eMZ)>(>Hpwp=a$mQH@QNciLJs{5nIvGBkAIK>1x>Y^=0vbKtfN4UV|3t zCA6-(3?d|DhJ3eAFb8iGD}+JTk02-)=%*Nd?@J&vTMl|`FGM~{TDV9)3+heyeGFcQ z70q_M-E(bKk_%B=Xf&*ttq0@R0ev&B_#aIrWe&trCy8XfC?LBB?IX;1JyCtKn)t1B zVJDkbu+re=%+zcrJJsKib$&UIZ0e9EgW@t)((w6??YNP4R~1YYljx&MRXZ_w4BBf#-v5{I9EzRi6Re2kEsL+|)R+f~zPS)XOF>ES`J z@pHfpBhu)Hkd4&B+Kp>3;7U#OQaYnwPaR+c)LsLK{o%wv+oF1p9-aJ+78ZBW zM{q{`bee(X(RXRWr=e+dwD(D|i8d+^F88JsPNTGyO~*60+xXuShcdeIkO9`Omi`Sl7nlM+9=XJ%w&%oeO{q>CbJ*Wg6;v-tYT7W^w{GK{tULWS*Fc*apjGWgRBtT@FV zUw3MCisNs-7=gL(MeJ+7#=&ccryzF1zC=+5PI4t{v488@`cl0(@l zBxuA*To5gTf5^^)g+~Ulq+81D;v{RV=XMbX_{B>?giQmJ7xy4w<#l}Hod(%`Y%^XQ zF2c(Po59P0Irwhec>HK~AEMO%JC^D7U3_lDFywpC0S}dnB`Gnj*ziXIny{|hhox+B zsO789+o;mEP?32RbEy~2S zmo|3!)(8WaSM%f0>zA6uQ21>O%|RLc7WP8^#Bw;fq0H7LW%Q0oDsGM?|FBCCKE|cQ+nUe=N$j+jrm9p%+LoLcV@)(?it%8*WbASNDfaC8 z#M2pABZ*xhMe&xvGAvYmj$5{Le-n5q6bsKRcY@MiqtX67-S4~Ja$4B;+yi{8;wFB1 zVJ2_0650j-uZS8?b+5VgTcJa=V)$4Zg6gotIILhUo?~W$rJw5XcsZVT7D;?&tPP0& z*9ttbsC&HrwmuGh0V`qaFab7`_{F$t#NktO&B>m!okVY>6OYq#q8C=HNa6d1z2^%kZ}6&+(a& zW5~E%1)^yahU@37#u087;{7*|VWn0*Y*4q4hksVG9@pBH%m4q&XS?P}VVVGhe{$1M z(`jWgY-{(Ji}D`+`IC)u=ma#{#77$V96*7TD zk;KV)yj__lj^cg=hiNdG;Wdox-jN{IJyS;yr88lrrUp3}IFBTswc_d6l5p`KFFoF5 zL;SKS@RzT@zemPzMUZnDSFKXQrK@jZpFiPX7!-%M*k}>YNDY!P%$}G%8b_udm4cLq ziL~M9FFby`1{u1^1b4diB>@w!V|H1Sr%fSxGd^40%FBJvkdfe^^^qUfk1xCYZ?0gJ zZ-J9{OlQTk4^1cqe4(Wss|Km??Tz(1u+Ev$?BLipXlaEke~*(zmU%o(5oO>6?LjHA0bVYD8~69#iJ$o#Mkl{661OWvvLfA9@!`NaRwSQ`xVrP?^6{r1oAAW2 zg>-j@pl_9gbgy z@M|d|j)(tWMjrRUcSbEo`hv%3#^i0pW=0k1NRqR1>3f>&FgroUHi(E`a3FS2J#9Hq zMeRkKMZpwXO|Y@nxBTe z-Jmlgkh<+|o`;7I<8U8oes4+iesTnMUZBU@(!1UzmUAO?$cEj!h-Tvt9Q@k^of3Kx zyNZ^7y5eYmcF>E+%-%^B+sl!J2ev%?&dV}nf$Tq#S*ruQDmV{Lj6-F`Gra>j80&!3uwEp#|KL)V;a;`D16?yM>nF3p|(kN1WT zgNd+Dby?k}dTf-GMi<_c_@$k2BaJf(;kL~axN8}TfB0U->hhEEzLryX$^1;bujVUm zmE=Ma2TRd3g?7B=ohuomXGEst9KgwQoM_1*%=@ft^Eu7qSr*s>1%%!}r!rn5>)x{D z@to$r_<7n(H-%x|iq*7sayyF7;VDo?;YU-yKXEGs(la2;yNK7(a%3?n;T*5LRq zNj}5N50LKVx%f9Uqr%t&ti*E~o3k;U#_M#^caoSutuD*|kFGb5r|JnG$FG#EWsQoI zrA4&o&T}eR(xwF!C23!^QMBih2-!j*Dk)0|33r~6q&=Y$T4<9tm9(g6``vSV-hIEH z-|zMNV_x^3IWx~XGiT(Q}1vj{VC!Q}+ z=x;5!)UT5uBUOuRfeuyKIr20&B`&;s3H^4gksYs0WFKzYvuk00h}-g2uzzQ)_{2+f z_F!oO3p;e1y>ZuKxe3!@O!hkVVdFCtO&iV8f%1?Yc98zs90ZBMt?<;c9*S(1k@2;< zoY%5IxMW~QZ#4ze#j*FLL(_9%=c~_jXQm-U*nAeRmE8fTG|PcI;nQ!7qn0dAW2K00 zO?1PLdb=PqZLUD$Ig39~m^}dOkKPCWw07}~Yul(^lqrt2eyGG$#a(lyd zsMnQe=Uq;c(~ob0D6(T;%dpc}&Nt*FOxx@V?;;*kKc-eFj$Om+Qj!$thz5q~pqDGy zd_+^U4zn4uZ}x+2S>&ftBqW~x1P^8VvetI#_C=9A-IBv`i%E-<7u6X(2oAS>rGIa&AwiedLP0bK z@%D{$;Bc5}8Az)#+Y_G%=vVMJsGW{~}_j!je>c`^N zLqo}xA>E{2IV<6wn(9eXflVLH%77OT@mS|aG{RC*#QTiPVZ*y`9w((TwH zI_Ih>32GQZj^*x!#xq~@_d4x@;ieg|eord5`Y+*Lj&+n4I%q)GsQuFTZ-en+%6sZC z+l#nZ?6&A z+X0q6y;At*RU|#9+YM^mId*2R1&RL(aI{ek`*l|nFK|y|jbGygdRMC`lFkd8$k-55 zcwK&+)|wxJt}j+FpYwWLk5L|IedILN?f*@XF)wFce#;+aL78atoQt)o`KMR$AWVLVjQD7;AWZ-Oq=^j)i^LB6BT4K88CYu#fw8oH21b%m~}a z?R8tk_4C)@R-|P}3kOAz2WAJM;nR8g^shNLu=+hU@k^I(Y&HQ=?-cm6=`a{-ji!qa zcGz=956ma&4^-LeKl^Y)T0YEd>w(SlMo|~*7E63vD|(b0l2wm=2OP194IhHyBo+ zLAOtv2FL5X@Jdac)N?ezDf{81HvLB7jyG?_JpAA8CR{;bFkIZ;6_(#W{GX0R5AqVB z?C?Yw|9ds0b#|bG``W_Lmo_*+`6{^-Vna-`31{6`Q~c@PT}qlng=|@y6!gB`|$_37pj)2Gg4~xVU5ToYjiEEMb@-EPZ+s<|xkPhMZdsev77K|BvO+ zby^^qGy9+bhv)IXbbU&P9dB9Bv+T{fI;ikpNY;Mw#A{6f;FhM&<;8}>-wbQan0f-Q zkMqSJpL8)l`Z+1Pa|)Iyb;JLmRT6ql7JF*tOmXkP8)U}gQ_yLpB^+%@!Gl4@0uOw- zrN@_XZqqu}30}^|P;pYkMpVp)cUye1m&`{h`df`hWO(_ z;N0~B9I1KC(=X{gRhvyzyULDb+r!<#XQ6SnE##&hl@3q7Na~#L!>^`#h`nn9gLaID z?GDFzor-RD&EodI9>u-Un_pn2FmBY+1)qqn>|WiY6Rrq4KXqZafRiu#+P#qRc=)8W zK@n#?l6kafl;gy0QSfd;C$7}Q5}prkLb-XWm^M(E^WUnkeRelpXC5Zm6>`&I(0MtFewQZ??EdKgT&OvQ)n55Xp#Tzc>MBhp|n9Ij8fMkUEpu)ei~ z6xi;B08>jCpZ-TGU)Pg|FIkcLj!u$4MgQZ8|D7jqEVlh2#>qPUh5U)q9YDJ)&{47X z(dk%6+#H*G$B6sfR4n`%51nf+(^K<};c#XwD&0!LOGo`--VI;K7?=*O*JLpf{Sfw+ zL_ztxGZ6T#h1kxyiPau0*s#KxT`hb7Lu#wZ1NAk)-9JM@YCDrh9!JT+H3-)ImT+}p z6XDxFAAGsM5Q`=KS^0Yf)YlfyEKAIsgK@KO7=2~;7QK@h%4iqOnxhG!*N}0V75k68kG2tfkPdNLpOQ0`~-C{{0H>-t+|c zskxcybW$$jY5p&jTK9*n7bya~v>o+gg`CS|(5a5KkioODmUraW;_5>+xWcdm=Uj=w ziQ5tH_3THB#~p@;O6Rd`&r7lV=7Rsvm{6vl2z1FFu1emG8yRu}Gi7n;N<#g`%K7D> z@WX*`Q=(-9Z$n0MIYx$d#k-%r5$9(LTzttn=-t04f6?X@;8rsg?>_8ka|N~og!PZR z?H@W+v`1R16OP}$*}-gyJkXPq@zRw(WKD2a@QmrWKh3}P0C>wkK<#}y$m@iCY^RdH z%%^iMvmKg?=N65``V9wBS-A{<230}Eg{Pq3qaRO`#MZlsuAg=T^(U0`GLb}Ti8yB? zJG9WTVAGB`339so$cJ;cSxugLmy3JZj)3=JbEvXW0x5ajv7WqTGaiO*KL%HARXEc> z##C$82Z(e$3@1AIfQHt3valfxzFpqMIsLYPLHbvzkvI>3t!~8pPUBf*O9%btF(+}@ zruCTQDYKzfy~YF^Th_dBNy&YG(ym@_9FOq)@*~i#nt}Cd-;# z1^Rg&Cd9@5>$jDolK&f{B>cPIMNW`(eGqO_{zG2e?1#Sox2aA}3caz}480DuOPf_< zNv5<)s9)jz76Lmxq37(3S>}}s@G)ct{O-98ugm%`k3U_hUAivngFqku{IORiOBu1T zNasx|gNB9NypsvsyRTibWw#CUkjNE9q}1cbObu#je+KrMSb$4vH%K_J7Yqg+hb+%< zVy%)4{kco(hvw`Uk}@L<7KI1XhYO2&ngto~zcW%s6+JVlVl%hsvp22s5ZjnU z3~t^B=g3Y{gZ78S@$5E~?ZDt*%_$gy5_VSo6LUYE&Kf2Zvzl*h*jMca)THc2xeqOL z#PRX)ZF)KkEU?2NR|LO7k=z?gRNu1;-VZ$hc_WMk|14FnXzZFfhE)yKg_JQTaj(-u z{G=ENn@BME(S51dUVjv5PwfH2W^1v^8W(u^%N32D-@}hj7z;iqvyI&?aCG%S_HFok2>N~!3w#3z zFSnEZTcz(lh6%Ea|FR$al|D<`K1b(|Kd1^>5orRh|MXI~!jWA%wT?X-^qj5S^PRf* z^=8bYW53IB&{}L+c>t}~tcQSKJCU!`JSL6Qb8}E${u-_b=m9jUR?IbQ5r6$V4jWcp zf(w1K$Gm53 zXm1%Oo6E=3yIp%@b3aAP0moi~ep?-N7*tGVU0Ohk&n_foNsH*DDTd6|{Ug2zPr`-k z{e*tR!>IOBU~fL@vjJ;nGPkcIEwAlnU^>$tUv4>x!2^<^^1CDW^aye?B9-n0XXdf!k`g8>bE3YFT&s39D^N!HeJ~kaZc8{WG z*WDqnJ8ebTQE_@Bejp{Mw~7b+-OBSMYV)~BraHf%f1WpB>xf9{t@#~3mytOdET(EN z>*BbUDNjDa?k&5-GM7d%zE0|&3^HKRDXG3v#~L+rr3%Kk%HwyBWK2F&L_a89g^VAk z@$0q(VcZt)(?Ep-G5N*kx?tRpA>j6S0+_&F9PXI}Gi1L1MGfom$$@|}&nKt9SHzLBI|i~}d4lf`74Dt+YIxLI z0~#?q*!A)OMQW3c;pq!I4y!i{v{>&@<+{7P$5k8Vv*P#8_)alXZ@k*90!lu`7_ zz?J3zmOd39nZy68iQ2( zFGuLswAHw~N`c+8(`L<9!kG<`=Im~ytVbW1Fie-JWIlkNQ{`agr36_#$`rb{AP&BC zGv&^dE`-4RaOi*PEfz;Sr6au)#T&NCbDuP0$%unm95>??SiP7mRxCe&`v=SB!Z{Nm z&UPQn+bxd{>pb9Syb0H1%XfHx$reUMKP6vV^|-9>s_@~BY%fuqg2g?|Fm&l0f#02> zYMk~kFS_iZ5@cPPP1R+7n`i9yqwniRdc@oUo7Qfl`x61Coiq@}o-zHcV1~UZ_q<^e z=dvRj4xelV@ey6&+we6<7w>FZ2od)yq())6@N!)R?6zFTS$Rj(u=xkcviQzycVAZ) zaK4$>gQQ?oH#)B>i4IhW5ajvoY**~P#!b3@bqnor{3g_ew2+rx_4rCN4Uek?agR(+ zay_;v6IY`7RN@Ta|@M4W4{hH0~8G2w5#Cn^_{ zG5&9@K;Ok}%H(F?BXU+#AE%7(k5{*t5c3}0IBCC7upLo|-6u~VF^!6NqEWW@a+#95 z(;@^JFD>aV(8BXPdsZB73CR<1@z4K%X~vW#*f1lTuP@n{XpB~oX9OB#&!W0GiFj$W zK6L(S%pG~fapMwCfNINeGHnlowwU|U;6ig;meLVx@IUy^F0aP%BX5g;H2n}@R@n9x z;GOTAO16p=z(pYe^qLOSDIf0;hlzVh;DZEm@9juwul=U{I!oleCx(cBYmRm`w~@uu ztihB8%kZ2mjtCF`aajXNdmB#ACJ1}cBEI}%4m(YpoNrbh#BznPq{MNh-r!m22QDnE$fOPe^LT0QGhR%-7orZ@)OX%Facab`;u`ElW3D`8;&w4An}C_5Nm%@JfPth8C+QnOf!q{ zJk-^+(i6Ey;G>%y*adI0?D9i{YjR5fe?v3w#9VtWFxC;WGS1T%NvDx7Kk|MUj9NPm z)=7^LSARF@jwHcGDcKyFh0Q^6Xx8rz4z?YH7NCXREi!+J$(0Z=L5@?A#Y_(hxI~;? z_QLw=&ZPHbe<=HV5(ieSC;kdr*iAN0zO$IE@0k`dX zUyxj$PP)p*iEo7wLf)a_duW$#VzROCdaNt6rk7WBd>{L-+Dly2JHB%|wZ|d7JXhd% zDBVMve$1d&Ym$X$o+mzyFFp(XG@ik(k(Q^b}3@bYwKm-Q0p z_DdFvtwh5G8SrHde<-u(E25#h!!;P$V2{n#%jlg!UASJG4LFI~N0^uCz-25n;L8o!^<0 zOt}T_Z&U<0JYVTmW4NNrAK`9L4oMDL1vggJ@%#!hRnbnR2lqAOhfmKiWoB12@;xO^ z8=VK*FU~=O?ExN!=yv;VYFBH6er=}c{bqtd_tgVasQF22e9#<;JJj6RE!jPaJT9L( z{h4;cQ7RrJ_=(6I)D6({wS?rsRbDQV_*Jn`)*~C`$4ltI+7r<8lp5|>RfyH^0@-lA z{#cp5ot1X8=IN7|d=J5mCjGD>;V`zIyN|1_y0Tt6Jy^o27Ga-v?Zp8sG%-#5@yivM?Vkh_?}dqJ(E*-k5ueI8sY1?aS-iTLS&(=@k!%|7 zPO*7AS|n~o*9mcy8{iKcCgrd;_Xa3_e25LwTf)MI+Oe&=&xm^S0sN`7nbt-*;tV^H zAb)}PZTs+x{TF87^8qHiw>jLhZTJ|K9MQ*FUIm62bmJd zrMg3b{$LUG%iPq$g*{qCB=)UKz%`7z=9xhU5}p8ILz%Dw;dj=GjS#RsciQ2V_}s5HQz zM*YczVuu(ESoj(}NiSBcCD>R5+44B{9j&7(+aG}h4F#RYzf$Ht51mEoEA4Pz;xXAC znUw!7i+a2@$Iw|X$-BqtKwm|JZcR@neyPIFwhQZ1*|*LI-?zx(Fct!9nfH@SZa1eaBjBgis6c{rCDJP)>fT!ksF z-BB%98?&d1xp2==I2L{#9bPD5{fPf_o|~tA?a;k=|A#WfdiO?G&v-P-2J~#4fxm3h zV3JCNrR`A$gZ|24Vt5+uY;v6+=S3xko9T}XS>I~e#vZvZ5aeuH8HmRQn&ACE?+G3$ zq3?|T3T2l~thW4Iw3U8+B9BfZ;syEC)h`y_o$+nOPvZtKN7>mAxtg=&_q+{){u@6> zz^JgV^kA_mGjUTUZnAr4dA!$qme8lV9Wj8H59+Wb@cI8G+Z_D zqw0yiQ+!u0#7ZkZ=PT{0AjfKzNzoU5~OD!Q@Nk*U;-RaDvhd+;?COrhqH3F|MUHry zHF#vB--3ADope$9@<|zNlp3)KsnLQQFWLm*(yFHz+_|3pT9C?0;7#wc|{(aUQ|N6~|zk&m7?EKh6llbAtpsv*h)@i7a2*m&IzlL7hV*+4J`I0-nTG zE_75|pQ4X)Av!aXgIWqwf=)TD3MB8!M9oaNYuRX>a4>qu1l{}iuF z+Qs9LTph7hJR&znXpgQYl)62;B=ubPTzKYbm=&sC7E z<#YStvj{b?nAwfI9^Z#^I{bz3JuKb?oIV!bpLBzYDr7 zN;{EV>av>cyEcc>zYhgi5ZRZ#9Z$)sj^v) zBe3Y`VY*;Q5;Xn{CktlOV%W7NTz5Ykq*V_DybALi1RT5!RMSG~k+NuVrMw4hHn{+W zo0K`PueYV$Wq#71b}!~OuOEWCr}bIJb|p5Zb2oP7<~bbOYZ+a3{R}p4TUP{5tJyN& zY&^bTH1{smjr87gjGQ`ch1Km&g1kz06yd=ohsZunVH^}aTfUMhh0J25&Se5EapgOy z#w-g8n1yky!w28>n=j9|=-EPs{7@L=sD)>`+`wLw9?{7gUXtQ4 zdnjl=_^&LFldto%^(wLyYnrDrj^;lYA;@a4r-rywmS9_x6z;kS+szAwdiANXxK(vG z)&$=NbCWir6yQ&u&&j2yszae$x<7ZJWdUb5?K6yN^+X>2(@hTx6~?LK3zMhVVjN4S zYODwQJsta)H7~`54$n9iU>1xT-Hyn7tj^#~zZoF85)2a$`hm^Ig@R0ZI3i+;={Z(J z=VoWls$U+IRSiYs{;#CF{(iB%d-bh&=$1g3DSwtYP7(`v7km$ZFKbk|;sxV<(e>*XF0oOC6iO(3GHZm{BE+fNHgT!5 zksGI}#iiefgqh+S_;7147JOzRdvpxRC3PFltXCLG@ZDTkbht*)F%OeZT9JvcwkU~) zc&(>$)8)ABtpmB~`_9wXzOMAV@**52eN6_Pn9E+gTE_~?Q^bzNWn`c7d(3f@3Ow*I zZrs_&!g~{vv~3Jb+!97y4g_0XGBPJee{}f2+C7~~Y@-Fdl7~~OAuG!Q+!v)w&C=&% z_|$>;<+ncf{LpD`P~&XwSduEYIbs6rpBpH^*H5b@^W8ttQK*CWtalTe>31;vOh2$3 z8Y_-z+>2*NmSU_?FRu0Ma#@^8GwxUGXMluE%l;v97P@9msRk#1LAPKz1VQOR?x#QD-Ecbx($_nEyf;d8NytGFIVTM1>2qm(rd=~ z|JDkt@(sA<5#cbtRE@}0x>BF=Nx0&nANkFe2z^dgPx3!W6!fhG2W(7dOZLqX#y>v4 zso)yzbMbcJ>Au~8J)SLAxNZ--O8VmIg%`m7Q7C*`sRc_Nb+B6(!L}t@G_j71zwv;~ zj{HkgKIeg9CvCV}7Aok)-9i|9MY(^n(Y9h6&l`_h^rs+F;8VO|KYe~kb{GEhBbIvJ zXT%rsi_0Vw z-6#+*{xt;c&yGR8zMo0(Ayu06yc?I-tK)3y9*ct{=J|Cz6syhG=Vc(dY^+oCr{7^l zMQUt^uT18BqiJm)UA$b)pI9%fZ!^LjGstjkNf4Al7A1ge~9Y;ewmhKRNtYM=>Wspri4@ z7$PTL%4tut!+8BqzFf9g_+XD3MR z_KiZXH}q(k5k2TxPG4}Sx(~9CJFl35b#6%({L@Wnj0lf+4vVjRz;%N z{T9!w03+J57x%??E%)G;KNtU4pS$UGU*N&mc`oSh+b1sg(u+>h`OCwS^iTt;I9pqI zukq$Qxnmsq501u|hh%xq5|Z~Y4^9|wgEfy9g4+rYsrGhj&d<6hcWIFc@v&WseHV@f zmG_3wc+3~o_NyXHZk5iywgXc4Tp?XYdZDsQ29(XcOYCEg&>_0{miji;QVmXxi~6aI zopg7wkir^v+2#q`(|(GDO#Ughm6PqwtSUceBZE6f&0Wypw{2ZWAI|{%Fzg1dx#o-| zon-O1)^v98P5$MJJcE|pLtIfA zM^5Q=*aW{H+$NPH?b7@g9d{TEnsi=}_f*KHJ|!A(wb2tR{ntu)IUjm7oeMG=K~2U- z!_9}TU_8~0Iadn)ZIUtBkKjpr3B-Ok2B&^XynaPmsvq&Ol@u07Dx!vl0$e}Y7Y9^# z*f?Ca8qy2hDtWjPm!A%xZG4cFY|_FD^7_oACX;x?45D8jWEbB5{_{WGNRAmQVwrfV z_-*)8zP(EROW~83liSXR>33Ny>9Fi^P&H{X8ULi?UWDa8%>;g?mpl<<%GWuVVhmr- z4hL6Jnb00e8uP`E9M4f7({PBIVF!n!zGK3jL#S^)n+7|?h;`XO)H_@RZSVB){ry;c zJ1h^+pKZX3z(8!ukHje%X2Lk^GdLGIj|l+X-q!TVSCpn4t3bOm-C4$n+c@E=Dzp0J zh0irML*!yr)Z6IJ_ZdmWuxRqC%^S_fRAXA`NWM*qa(ue6-vvRKGNDY`mSjanwR?j0 zMNLRQ@EG6v-y;TF*9-O6;ozARf&CBZiL6# z5HT)|rIS8X!K*)`(eT}5GQl7UAFNcy$@doEx42o%b;>vC#`?+3t5LXrPh`Jmv!Jt4 z=IS^_vZ~N--L=BAQ3*T_5uaj02Nyj5riqi~;)OiJygGQ@d`A{5+mM-D*@xRwJL9i` zZ-npmAv%YJ=Rc)gFOFh3X)d*o%@OMS)M~KI8#x`nOtr_aZ8_2sE>_$(EEwm8ey|Lc z-P?belcNzs9+IBFGl7?L+TLvFKPwzk%^FG4@@w$XkkOen$KXb39ysObvc7|=u;NhC zKN@*DclW;xe#bkZ?*lnud^Fo2f=27}^Q(}|`NrWHf<2sh|8^^W-F-QC3a<_ods3Oj+;DFaoxMYq2Z z#rMZCPD1%S9*5}3+p%~pMW5xo$|;=mt5_&IPx(1^`KFCK6=q<{BTY)XtRq(*3HuwO z`Mu-Gfs+qN(aLV{DL@{_c2!~h-i?Iw`&W=GXOy~#A}K3WlH*Rc@V z8P9X978S~zd#xqV#LIwBmxj;9-59y@XC#z9PD<}}!A z){slshz>GU<9<@z9t^MJsR^i)R7a5_2J^iy>#f^RkTo3o7QF>!;$$} z@W%93TrzDhI_Aa#kL&K7v20LH6}jAePMZH#LK}++lv+he_y5Vo z6tkm{8vN7pLYHDpw>V7JALuUB2|6G>ScdeA^;u6TwChmd$lRiUm|7K=|j%a)A1 zhg;vC#^*kH%%c4yGg{IeKkkk|@@&3%j7BVtQ9VHK6 z=#}n0;L5}+0^FV#!Ub5hd)^89lED?)tGHml%*SP@R5kU7Q0};8GHUl4!lbvA*^!Z> z;AM|r^j35kN!HPWQ&HS@Xn+sk_8;X6W zpCG$V_(>;650pMiZWZ@R*e{gpW~5H+_J)uhZ>{Ocx94cyEE627bW!*=?Jh|pp>r1C zR<%g#8D@rkFQ>wFHQ_vrNP8H$m8xbrFszh ztDGUv56>VW4hu-?Yzu+*|7hdO=SiZ`Eb%yPRM;cb_jZ|#DcvjSh&2>Fi^8a@;|~mS zc4R9?+u+o>Be;`IIl!K@!n^ER2tN8=s{OkM4jlIj#|1Xy{GIw_uzF5GU|4}5UzvB2 z^o(j3+S;^H7*EPt4@wskV^F>**ds)#wGEi}J&VUHayFgBdS2|wE{sf&tpNJ7cji6V z(3n&B;qWDKcNazO{K);_u5ycdD8|qqQSNlODN+AD&VgU`vAs1QN->bf$YtFiUufVl}etyN}!Z!Hyg z5}#^7!}i^HaH2ekR@yJ~@9#rBuF1ZGKZ`8iJPV*T8?49*6GJAKlYwC|rb1f@Ig?HY zofGzhB}aBB{-gc+Q)L!uvl(&5HahEihhJ3A(XOB;vwznZGU+JQ1z@DyBE1lI2)=6e zV#oSEr*m2dLcHE18n~uO{69SNe2c#M%3(jpd}(u8Ij>)l?VejCWz%d>S3ilBJBH&^ zqw4}+ZR1s#)g~RD7txB{C!}Gcq{O;pBHG;jh)4alp<&o-(CD zht#}caC-ELr%$rso(3MMSV`vA*?^!XspX%7RJN~^~@Zb16BAhK9n8?PttP)`JWlbD!5Vh|Hg>v&nptBx9IX;G$ zvm|cCGHTZ|83x7eV}~m`Y%rtqmcXi|*T`(W+hE`40K700@+BdwQ0P;MAw5~IjpL-c zJ1!9iX36s{;$fX>*$<~&|3bwhWnp}{JoU2dJeD~bXj#RMLKosLXl3nmKdbhuqg|cbOw{s77Ea!$QOcB0o2PWPp zho|5zL6aWGoYWx_ID{N&M@ovw!IVsnmfElH|_&mq?JmVzXzgA zuVL(MQ7?3fN)XS{I3oVgvWe7x+d-nQh@jQVis{@BV|_lwk!!YEq@sQyTTUkPx)e>T z?JUT(Y3yknvS1^P^r<4R)#`;c!;jd7*#E31TyGl<)vHcJoy}8TN1`J}2rq{G7TUu3 zogL>n^KPZm6Z>Rq$KxIL$Jz@D6=?`b@|sL9&0QHfOe(Km? zZG4k0U2$~R=)u=s&}R(;SsBR@=-^@U>B+8d%=kz*c4U$px+#m;!xfD0k3v5fHQ*2i zC&seaGs6CdXrTJh!ns#p!K4wZVO55=z`xrh1HQbdx79S%DN>XK(b5I1kMS4M=@c3qBi{^-u16 zc|NHHy{03dY!v!C|J=`W8_}+u2#W^>NJDBUE$Y)w8iIv$_CopeCoW`Hq8pD-Vmx^X ztW*d9!zf3fCzZ%u-eHfc-O7OxxUm~uP#XP^2L?-?;CL0^Bk@L&-cpZp-*frwsryhv6 z+p#wNQsJ3}H-votok8zSU~%3UzJ3j+5iznDN6!+3_TyQ4N38ufjQ9ri#LV3tej39X zhLKfED4rg(mkd7l0n%oxbGFI%X@GYUZItaN_Wj+8i|^;i%+N;)-5-4v2Yh}_S|Tc` zwWJ;%YW3lK>I(7f2_?|nDy*XfIl0*d<1&}B!X!OiS~)P4oqs2bVakOHdRpZ@;UAsz zUfx6#_pbc2ginqsNxU8;)pjE>*G~?;8vVu7Pp;$ZiGmMYu{3RI5D(b?oE}y`BO7m} zye>pN*K~uDd1>%7q6BOXcM@>Sj2+1;)M{|~C4yUV4?a|#g_E~m7W$m#V%5Tfy%g9( zA6Y!;Ew-%vVu$^OhZS9C0nY-NGl|Om=1Ox+3rLni71X@| zdVK93(!U@V)V-46vxe-B>&xBYFSW(g#pyWwW*S{DKLGDH`VpVW7sNblS(1F)eU=%^ z_IRg`-A7E`W@5jJALL+6vmlSr+cL;+~#Ipm(nnBJx=TnX{E~%!LTcDcQ9`o7GtM zMVfBp&n|P3;&Rn%)G*71t}`FaC7d0~Z7!$-?QdZ)&a#r1zv#Hub9#4*1K>0iWIWD3 z8~1dJ!NQht#H!uMvO?nlwbDqy3y)i+azAcK_n-EK-mU#*_x!Hs=@FeVd}yg0J_w!F zbeOL9CF(iM6;0Q!z$G4gaD{Pi_QC5eb#-}3PS_sA{IBuSl1{0Znj}ZwqoHZ zJ7u=<6@O^PmIHWp%Cy3)FDkg#zYBgO0pefc1yZLj`$we>3#Gl|`ytrWg57~PQf>8m z+&ri^dYAv92_F)%V5o@MHKpLJb)9h162j|3A~C*B!d@qkC005pm!l@_-bo})*)KwC z#|)w^KLaNXUypK+B5~B=1ljp;D{A)lBSzi~#<ItMAe-$Fg@(H&`ws&+KY2@t+8;xD>Pr*n~rjvfFAiVZ2xIHQWvU%H%_#nX4OU1 zhEgnTI)UmN3hCW@a&XuDsdUBX8oE7TEM2Ix1WL{4;)DyfXtAazEp2c`&elb2S9|c&=venE?`sjpj`>})$w<=< zTkJ7z2JyUmTgYFj^h&yY&1y{e-bx$SoyLBbdO%Q39Qm#=hxUx>1p98Z)1jd=&?Qv~ z55CC6^fybzc5FMA9&{9Bnt4G4AuqZzQIW8p`cK#9BSwo|ueS5;@n0VQ|3N1mK2J2` zdk)>eY49{iK3LnZA9|uNH6eC+_89oDfgc5IIk4(+lYYd)^0^BnCkY$Z_Bm6u4b z_fha5tAT7RRRN>kB?29XF7Lg=l$i z0J=|XmJU-EY)T?tE1=dnCxaaHoFj#vNzP-~fRmk5Z;Ob_5oEF)>)T^;HI_Cs!o~Ve!hD^s@fmxWO9V)(2FWl)a;dxM$j+VtQ z48Sn=33x<%C9e9T4|2b*5tCVpva`v7G(zkxes(H|pDT%8c)8Ow-D>H~vY~kH@dILA z(1cG$6=S{4NO4W5GrStRv#`5L3F4=}c>L>0%T;|()A@O}*t~K->%kea+}jaoU!R1R z=UQQWZWL}(ErnaRM#Bb`-32c#?4VS71b94rVmo>=6VV3ruv5gbgGzYW^ZAnhlJ0dU z8n-?cH*D=A$m4VJ5pfs$jo7KQO3)KuZbh3p)fsY-UT8SQ!X_X*W3yb7^+m!N8WqIeZF+;KgQ;)#wTID+23IXEIsxt zX`g5;)?1xQQt_eI+5{_YESz~d;LwrvENGfYmPlc!_L zMPY9Jj~6~~NR1lVJ|MZ!DOyW>RwY+#|5QyZryDKQ`OY%v*7jmFxfF-X^1I>8pB-!F z&#y+}v#kNtwD^-CA2AxTS@%-#$JM8J$>=nloF2f}7ipJxB3^5!WhQs9^W#0dEJXIb ze+%D`KDkc##w7~loqGv#`|i71sPi@?1lu@sftT^4t@$xn(pTP$jrpd+vLz~PSucA$ zsi%cOgM+a>DTL&h$+M-I*0`%N1oO8i2=plQPoqv;pfpX%SlT()o&5SJ=t?w>6bW+P zeSSDrADkla$ip9&C2OP2K4jtmUEII^8df>?_pU z&Af2n z+BX?trBr~mr}CHh{yq~7iqoVHC%OpV1inrumkQplA~}CM3ym}%VAs@%IKe-FzT7ZF z7JngEDhWP_!-rnQFP(!>zwr$D+$9T6e@=$$)xw!$NtpJ^fAfz#JJ}rqzE-Gd(oCC@ zWVVWhx9FcJ4cR@tw`h;DZTPIr?|;$k-N_D<=69@@t|SN019R)eepX$`x@n8ZQO7HE z;LHa)pc!el=}FRRy1|b|W6ZcOn*&B<(cz2oNGFSNFycwaoQcQ#r-hQ`Pmhxq znWyp8<3Fg=*@)RZxk~^3w4m|V%n15!k=>EG12sOsf|{cB>`w%N31k?T+13hM-(_9XkAu?f)*o zU*afnw);%D^JhQsvfQGROkCl1VaOkK%d1h>1zjjyw!ydx5lK?vxJ-p)E~74%+E3oV zVr|rz_2(~Aw>}j@Ih*t#Se%+CzI-5_%)VTXu@&>BeV19%wrxIif=&qSZ*rE$A;{Y3 ztqvXZ>MDI{E1c7j^r(DFwf%cQ+@izaYWhT=g)eL0I~pr9jp&?`uW*5#0K2ux=+u82 zuN%qA(m&D)96zW+V|mwH9$S zIFQ-7R$t6Vk0^Hw`tnQl4a4@L{R7h2B{ z#$gGM_gnDmWBoN4`gAZEI8C4HRx*VYrhgz~mxxd1A3Vv#p$*h4^rApVuA@2(%Q+*}yj3GS^R(H0bs{#q z;tJQQm6D&k`;t284J_LGMu0uH*Lki>qb0W-)MS2z9xP*+A3GgpjuO>re0@>RBPSsI z|KaMrT>;*7)s|Id=8OC_ z4q#`$UZCB6jU<}o@O`JCPtsxhsNn(pQ2jstb3FcDGkcPEt4}bRU9M!~=gAX^+ayO6 zK4Tn?Gkt_ydlUfQ-roHu>GC@$DadF>2^}Il@LCC8=I9CMOLLLJJU=|4*A_+-UpHFh zN!%-(8BD#DrMO{p2zV}bB?Ff%B)?x;VD+J!xsDqh_(SF?Y#0>6)u_H=I0k_!>nj(B z-$KW)m%(1`(KM^@w&+5K25g(e*1p9u6M)Ia9!jKlkFqs?LC?i~i1x|@ME<%z_xjxn z$w^Opn?x-IR3*Ecr%9aVC_`n#M^MLDL|&H9FTUa9K@_!B3;Ke;J*Id1NRp#1GSNeEU*Q%cBhr7rQsU; zo_V?XPr6K=G%aH^j=Q%*VmJ7uX!-E7e7U&)bu-B+y)sbuwZ&)eO~6miR`T){k8ahY z8ncp++>wvu-idSA|J!ibr8xq&z(i!wf{@k6LQ$Xh`KZac23{_%Wc@&p|7!+wqx~?Z zv-z2sU0JR z_t=J&dI+)3=zi4XqeXY#T5mjdTn@gnz?mDLnk(#F*N*KeXFsc*Vy38iKBesFtt;ghW!(0BF6AOp;0Iy&Es5&@PdKAwF<}8teTPB@OGD~EA~Cy_!J_Xe6dibr#7KRz6rMFnPHm|q zUO)ZNoDF-CaNH`!GmoE7FRz|PXB#WX)MwX4p_;eRg4xpZss{ZTU#bzgWdErta8ADu z_O`0TMb*G)pRm~oB_G(2u4sF38FUU8&Rt;n+B@QeXHNO!0ppkB3L_;h_+T#636H1L zTmg?#o(^6CT_`S9I!3poc+lpf*kt@i?0P&K9sFsDHhVSVBylKyb@u_2Rfpg%Dl%2$ zwh8-jH|rL2g%N=$;$#i}<&uO4y_DlB(`}&h`chWcz;_CE>lcg%J^6~$!ZfMso@*q> zT$3{iS%n&_hGP}(7akz1hRywKS^ZNnO8Blc{^9ffZU2uv%7;(GlcsHg0T*B5VZ-L* zAzNZ4V_$XeH{F=>7i-*9N70iaPw1y?T8ef9Se}8ye#=a{7Xh>1}j*!!T@k ztru6}dLD1yERE^ldySV5FK_YX;80M{vOsQ@oAAOYif$_ZL6LLpa70-SBX4f(PLX=iKqt%5c0iX)rYQ*5YL&f(lz~^n}B~+nvy=u$|n91uOA# z;Ulg&ECnx^fH<`~1V@r@z~kB2_W%mIti&~(lEaJq4YAm03>oojB558XMuX94oIdL% zYQ036OlQCJ7a4j^$LZq)__7)EOA%|YHGn;y$ML)AeYtP8=S2_tEJ0htn#xkco!Pp; zx;QK7>)dAjKiUPQs$WR>i%Fd0$wk7+OWAl;d@UoDZ1TG=ncL~YwNJ6-Zslq)e0-ZF zWs}IK{s*Da#~xyGXX6_WBFn}^mP-1cS0vi?v!U);G<3dCM%R^h;$2fGqj|!7(ziv1 zHsv(HDrt=DA(yXW|J(0HFD?tX6fOoO8F%mDp3``NZ_d!h({Fvpd6x6=#B)n_?^M|*t-L5@HWELomcUkN~u4hRDVwEQgJ+I z$OR7Wfjt{y_v9HnChf&8e%rBSp87u<67akY4l2SmW)`^UX&w7U z3TpPDh|nJ>6Wd{EoR5_|PxJZW`}!7)wy^R6WS3nsqh;f9>Ai`+gGIAkWbu}?xp;~6 zo=G0&W81$H@8|>McEeAUC%s?3AZjyfyP@w0Zp_J4w6IqKw^Q*qUSnBiF;uc8_J>gt(8fB^tKG z92PsCKna)Rpe1G!RygjAEz6GMDX$}NOZ+lUF|r*8OfUThTM*rUiX`9oUD*ho@i=&i zG0@&wWuMlDG9Gw5eA=fkAqASt_%>qMsaJ%hUzS32fjO}~dI;Y$X-2}O-l+18FIREs z7~XMXC)ZQfoyqsb?pPF+@r>96%5iQ}yg0o`9U4An07;2`OB#cu^W;Wd?y>zOC{1~b zryG>xVYde}9N#X~;ln3Xsk4hZH!wDaOUinPw@q9`5@t&8o}9cK*H<0HM_N+AV?zQ? znl*;5F_FeFof(E)Q*Pq)Zz>@4oq~?`83eIvP3ZE`%Yk zT#QeBA4-yyCNf!!`)dwf&69!WakSDUEO~Ps`T3P#No*Fhlx~2M1qH-RX$dT7EW-yn zXJYwFnrOQ21gvtWlJGJrliSbq;=Ip_*-*gKllW)EEtDApcc6U7hcH0mib|#TEY^lLqapf7@NU~tT=XzUOsm*`n4oh~KREc_8al#? z(I(q7_~N-sc*EEY43GcoGH@tW!*_g!kj|&|_^4SLUedJ%tsSI`e_7~pci+Wxr=L&c zp6?lhbEGkYqKs8JG_Vktc(Z*WLEuhrWI9RuAC`9&X=pZJo!@=%<>?fg9c6JI1amZx ziJsXF1Bdk9qC>aESZ8-6FAMRUTV7Z)>krA&mk5HnLdK?U(b_R)gDc?<~)&cy{@DWJU^*aNAX^FkuN7Yjd0FMQz(CE-V#s3gt)e zphLZI*Zd_c-{SE((d0C5MsI+~p26VtJoe2^q#C#li8grRJ`kjsrd!T7+z02Z`a|G6P+ccWfRcz8X0W1Mw1(N-IcT1t-Y@HxFV4*SX0Aq8=*`Z#EKK96;pbp8-!_wBdZw=8*SzqT)v!xo;JIq1Vb}H?Qe13YMPJ zNG-ffcscPYU0ECJrF=*0a_2zJhOe-ps18lwGRQgaR*WuO!GR&>sC`=#PP$_bQ>?c@ zMeaAgt@vMRRC)n6L6&4mpBNsu;H1e7G;2$v^bQ5Tvbk^48D75bf$>|(pWt_*{RskY zmVFQI(y0pkxnp11E)q#ZX<2aEp&yv4j3C=xy@_A>2Xz0CHs?4n0k=(RlEkL2!I@HD z#iBF1sLOsNzFU-y|F|C`3hQj)-Ox}tU}OuKsa?2T&Vn-+2x|H@Mh?<`Zt72 z{ccU~O7EvR94+!~Rf73bZIJuSi~rdGT%SxL+vItA1V=!X_2)d#Lr&%48&+TcAAJj= zEkqBt3>At_J(B2pJCd@1y+F+Gz$e8+C|7!K&+>Qq5Mn?anSWlh^8WaJL= zBTl{C%li~{*|ts6^`s75Pg@gL@o4(|{TJF083=iQE|B(flNgQv;Ei5$QRHzr;XfT8 z8P$kA3XQSXoKiUCYYlOenXd)Iojl!sK8(8guQAZqF})aI9gLi|6I=wJ<8 znh=IBk3!rw5+$5{T>(1W7ji9+oVht&LD=0ineI)_pt6O9ygUSaB7;;JzJKqJK<+vKMNm5M{~{^bh1f_;cgv$ zp8W=YO^0z-?EhH6)4HyFCD>?wCb#wCuvW?2A>O%?SLWJBdSPv;aQbr zXP`9Z&5$Hywdw=2?SuBO$dWUaoH$IF;7FvwNf4?d**-Z9PN<$kK?#9rkDdYZ2d>0R0!E8; zV{>iFW5Qs(CfhF-+l<%bsxp$%>w?~RMohA3cYFZm+f9_7+xpi&Zjm=d`)9ylHCgPu zyf0qi6M(HRm7#ufMx(4ZJ^$4cYt|<6&w`-816kgWMh)QK-9P$+$BVU+6?Y$#WM{Dm z&Ce-w>e!0@!d@m9{n|LP;K>2e(iffNuX!S9OdKV>m)jUFq=w^y-;43AO^V#D$v7DFLyLD!gas=hvJqRZ{R3tCUdkGI_5b52>Y+NXgh@UKRUgyG1 z4IV*$X;_fIo35fin?4g>o_rb*4@|awUb;z!R!Pr)oj6ANAIyZx@8j?T1ua_Z<^h(& z_rlLmJFGdu1Xf3~xw81Uat1Md_(Jlwr4`lKS2Gy@V6ITL1MQ}naCJyJS?OuX%BFnI zfaJNcXkPE>aOU<2l2CdOFFiI3qTeUr!5?RX>7T6-TyaI>vtuy&lxz&~alTM+r4?E> z9fE5QbI5KV8!(EUFWujLOHPhVX1e;Dwil_c&qXs9ejpPU$TC>{m`lpEMWL~8b(Uv@>ZEVz?08!!u4BMqNm3qS(*RPAT0cn$ZR)XZiXoI zCp$wc9wZw_8n4>@rz0s0ab2Vb^4Ot>-)&6*ORo-cIl9}QXNKSjPM^IBU0#|6&tB-T zHnUoG;syJiu%3M_d3DALcGgYdaWFgVpR$~5{n{#deCZQ3Kb!Ushd7YxV#Nd>tn|~Yq?`WBuhFcTmvHO6M?%?LD-ycxi%{&k4lNBk zE$Z>3QS#pD9v*&hJFCmnGl~wt2jp$IY*~scUV6gKD{+L%n3Mco(WeUeLVzH<@L4 z0%^4#BMVDSadF8Dv|>j0+>Xb4D4-Y`#t@zkG5`Kn{x9)E|QSTK`pICouKip#V44KN;;L?dtiA!`mwkb};O5Z(s zdIbY}%2I8GgILDPg1g(_4>tI^K}50{UTkagi`mLEIcCEGp0@o_B}A=wp(LO|7iBE{NRlmw zL7}%U7~C=BmYvbYk=k{_J>-DY|3JWGsC=dvcJBE`_9c3eHi z!Q|l{spx|Y>JDNho8^)To17Ufbv2P--*+ue{5yghKOhbzwhAPZN9B+hlUWRQV@NS6 z%ItxHUDq(V`lxsN4;Jh^P7J=(;_^q#7eR1fxHhsE48b|YdT2twC{g%l2PU6B2a<{M z{l&0ult1_QwH`FDhil3f-Kor!emjB(u_`mN~61^E;*y^M>P4tWN-PmXzo>z<7;ApsZsB6%jkh^Y=qx zUSB)7pjlPsHpGEUn7WNPuTJK5^-u3i7EWVCOn+RT7|HHw4@DjWb@1o5-FR8@BVxX+ z2bj#8jNJsgMarS`z~OiuYB%pgFBT}!SAF9ctcZ!RW%}v6@c?6GsXxXgtU2|vgvWvU zgn{YEUU)!-8u2+a_rLx|OX~mjIxvd7diGA_E{&JY*9|loN9r9dk=tf9*kxz}HO4l0 zRa&fvUrGLw97eosz5 ze}gyVAH{7y^>NxvORV3c3N4wt9X+kjBBD)Ic*M#NXqqjXF9{UzR~&Y00oQX|6Z+Y_ zOL{4-9LbSq{>n@~Gh+^7^;^mepY^I=lJ%2gB@$9_)56adIPkF z7sBmIX^eEgDA7ad9UMGwx+^STT+cgXy>u^C#YY8Cxb&mU^Rxjr)K7;^`7fdIP&0gK zJImp#2gBlZ81p_QGzK(C~bu1P#gdp{H55tl8|vGRQ+_y9-OxpdbrGLIF#C5!33 zIr8P}?$5-lXY}EqYcVc!S3`wS?b0&~IY48NQoEIAw4lvOdZ)z)@^OeFy|BoU=FEIX zw^pl`t1tcx3tQ-a_=69olI)*9CH?F?ac86xlmDF2u5{|}2IPH4O!a2`1b^jk@Fnjk zJXjG+ZQdHuiX+z4@oX*QmzPJ4?gV;2*_R^wQLyR4F+y{{u)eIzxB%5Pi~^au4$?Tz z5N^&s3_hvR^v~1vG$rT{ZR>nNGy691G8OP}{71V`IkOu2EBr2#^V|)G>8yKtRN>GC zS~+MsMIZD`SJ zssHNXFvhD`QL%h28B~7bt6%x?*Bj}_1@YS4b!K8%m|c|FS? z|H-4joHJ>gpb?DDHKNpJM`}Nwol6qTsxqUmX6=M6aY3~H;U_xtiz&Ui?gEjw528Pg ze1;vp<;%y9%Yi?!*^FOaW_-GQ*OVLWR)lSyZigo?r!!o)3LoGf+w5?^*Y+IGOZHc( z@6)*5bb03+IyANvChRGKpR1B+*M-CU_a|7z{ek&yx>9{^rXP2R84Nx@#G)^>x@f}^ zeGOXhZV73+&-Q2qil=3$?bGRadzS-JCHt`cl*8oB`)Il@bRqqeKb=0AeUs*R$d@nP zZAWdkCi3($nO=V*jV-=cdNxNG0b2%WknLpvAMUt<-VrT^b9E0@s_T})+q+owHi}G` z=g4^NH!ThJdn|;M0U~$)T#g$(OO)B3O zwXl5Y!_iQ>_%+td*~B@OPbhzNU8dY~R5tzbU@P4!E2JwHKZX-p=IC5m46Y3bK=)nz zX-l0EgL7O<407M6z^an{-xTw<~j z&TMK&`D@!?bKqwPJkkrQkC&s4-LL78g-Yc;`x%sv+2&cE{X~!5ITblx0-JtI*XXi` zQMbSGw61d`gU{3Ab9N_Q8}k6GT~?tzdo2ZtZukEqU-mf2oR0dug7&taOSJyA_H#sExb~$B_XSaFjnj1AjDP*8i6D-eE@b9_W{IiZS%Tg+oLz>NPZ95z_KO7f9xzsUYxtiq=%C zuzt`cTA%$#O7ROqY6pPjk6?0rWG1um2SKhhZrw*Z-{u5$S#gBS(b@&Y-v)xQyAe!3 zD4^o%#pLbgGjOi-3#&?gY(`-SI!3pN6Lj`UglByoh36&cv36=$iyT>XnOt^Xf-_{tOrL);0&f3Q8le^nh8KH)18Bmw`U%ixYzcHX>*vMrcwgD-Z|g}h>6v zaXhJXqUQs;eWF%5>{UnMFCU@gjx4HcTP)JZjVH2e&Co|D9X$HGpgeo+aUP%G^Wqqs zFsOU|XP=i2{T6c%5_Du}l2RoYUik{QN^6-ttP4tm=pD+q`jR5m{2K|5!_t6A@0R^f zmfdzDxY_!b)H>xsn?XCons<*kcsz>tZqoXB1$0wo0(D3y)Hq75d|~ou+E(cRk<~X~ z_jeoWC5<`P)Zc;bE$%*#l;31VUmQOOb9{P~@85BZ$>ATqTQ}nH6!{D&IBrJ1OvsXG z9=IW@i(d|wCG`+;^%u!|Uhmj*n+(J}hvH1H#x{aC#`&?p|C)c%F% zMtzsYFx}0|Kv0ylgUT%|B0ZE(!|ex#c%k_%$laeyFBSOH%?lPm%#8|CxNpEeJmN2N zwnJ!p6}b7O@p*#c-uq#tmJRiBN}#{PRA_9jBhwFG#`Ez1PyM1&dAnx$uN2Ji(D8!* z-bd!if~)i%MJq=~X0v>~^iRVljF;x9Epz9=WNuFP|%zfDx zpGV@@otPL8ibc1RR^y*CJ9rueJWR`(g;tZ>U8wHOcJ%(-3G#b&Fh7qL^bKTZN5y=d znyC}v$EHG({5THA3`~Te3v021!y*jQJ1(y&$aBd-HH5GCFX_%0iw};_fm^qN;ONZf zFnp#5z1%5{nX+>R7p5q~c;RtaY&=s&mAZZCn(Z_semh1&XK@o(Z_$r~UDKY&gPx0owXwG-wHnMKaNfenB zS@wFz0b$n2aCA5<4bOcQU7Atfheoxv^4n6E4TmvXAQeFGHIy>I|p&%+n56@;V7J7<@nkAsjAF5kh15b_mo^7Cu)Tj5nL_ zm5#%O*mjH$4pHhUDp2VMblDunbJe1gNa*s0;rmC+>8FkuwD#Zy$Jg?ylkt#v@iLz$;M0lJ z(=?-WKR2WBZePkD&O_M~AMfK1VLb*B+Z@aP@R_EhlD?0o zv9Y{O^L5fSAeLN|?vkMo8L$O+NYr~h;p+{!UsZ;QZk6`ek{rX7! z0%L@4J>z(Mf|T|v$k!kk=bUYlhSQS9S|Gl5m{DZn^D^h>98IP!cjk3fK$=>k|Xj$K4dZRRv^!WOS**4#H^Wkq~ zfT9O!kWC@0KVzn&|G^amJd6?9$?qpx8EK4$8fQbEXK~jk9mM}Wh`S&%4Lr{5X7wB! z99148jm3VA`&8}@^(K!|r0*Ff`^-x{p?|aVjyLJFc-fZ^DDi<0gTupE9IOIov^sI_ zIvqUC`3c%-dtVe_(i0{t+Jfx*&!XMOu9E=^CK6Pu3ThKy+9aJmA{nvXM)Kf(w+(h| zxg=Wr_!Sy|d^mo(yhYNm>m3hUz$Y)4B{=TeL9|Xw56u`Uh)%{3G%qdn0C;}lny?~~&Ur1Nt(mss?TsK=Cn@|9$WLEhNL#TE;72EmO% z-Y_|V`NRl3i)>gMm3O`1$Hfn1WsgXZ!V|PUIE8v$u_|4w6-xHjwUZ>veIz1xGo#A_ z$KX>-LcyoGCoHlqD)S49!U2CX@x2|Jp+0ss+@Grmm1lOKZv(qX!G+PVaAqXQS@#eI z2j4_}ir*p`_sTNU%5{wHdn>C&xuT__sj1z54m>YeON)3K#JgO+6XQ?a=VQzt4Thq} zJ6RjP?Eg+@2OpR0d|wEfcZSiHE&kATY9)CRdV&0r-e+)AC71En8K+8BNjg-dUt#YP zqFzJMa}n6Bk%#ucc82d~_$C-xG6DLk=HWhTuE3`Ga`eis7}4pWHpH=Kf4+{u+sND| z+@%@^xD_%uVVPD?FzFo$epki5Ieo=>4AwvTb~zb}s-C?n{kHKBN~%>vW+fldtHedD z&LNlrCVOP?z-Sq2f9NPYGGu348E+-_S4nf`BvI^=N;1SThQ}oqYbTQcgZCuIX+Ev} z?GEZbzmUIZA`aRZLLT*2LW#lSD817FhNDM;$@-q)a`8U;lr}?Dm0(8=O>RJn-aEYG z@N7K#YX-E-sz8{LGnVV$ZIfv=I*g_P8s%k2ru3jXH{{`qei{y*)NNDy9=PHC=PBQR ziAQ?a0QdfnWQ3d|oi575PRgfP{#Nl2&fYW>>l?-pZ1)ISwn*oz?~9;(^b=G#s|Q}X zBZ0RcK}6y}kpHqsv}xT$%)hTamx1mL2}aIJ2SF`93T&5rXZ-TG(&ug$9kx_}-Thst z`=L-Ao05bl+FnJ2>h1A~@9%_sor5>K$}Z@5;E-|4utC#SEDkyb!wp5KK*L)S@x>Kr z*#mTJv<&zDMF8{gwC)$z=Rn);Ru$JK^qgu&nL`rJX&H z-piF(siOiP8?yhQ)lkrRFE}GfG66tJzurpC_nZoCwx3 zYtY}r94{krA1y0_SY_w|C~E8r z$+r}leRQ>V|4;p1bG1yNa2K49vm~pl`jb&N1biDokX{?JH=duvWuHWYvt3})P>i4b z`O5OM`(CpisJ;N_9Z3V_f(>ZOOJ8!vJq~tnKOqUooR2>23jl|^(hCi4mqT0VQ{rRp zj29Gc6*U@PM_KJFPy$-OX#LxfNVZ-oL|&76qp^1kHdE@y;^30ZaT6r24K5pSR4b(T^E$?x2Y2Du8wE7T$sS7gw1b(F9OYp(E1I+RrJL6w|JUZw zY2}MdgXFMms6JWp=ssFH?;N@_XAOL=7E{Zy88E8#J^U2Tf%3Uogl{jZm8A`{q+_zP zV^MT#A!=V^NQxbm@X7i~a3Ev?ywH^ggB4=Pd+7a-2eDo9aHgxpJ=|Fzw>hbs79QT= zejSp~Jy#@y{EssDi|f+Kg~*<8<*^2g>F~e-6Dtso%;x~zMc>SO;w#pk_~yYAqP790 zrP_m&q0agf<7elENcI~%g~dn`++S+W;}^Wl_a}eG+~Hvh-dsT-e?ShqY3xEL^E`Q5 z5f}cwf;xuwV{O;2EMa*YHnOp*fN#U6xSG>a7yAuNr`65AsD9Wji9=)xcnurP;}F{{ z)P@joBM(DxO~n)hGY>$c;3>++pHQzUHRQMPV@^Xl-&`QlfCDqemQ5Z7WYi*S_#1Q- zcBIXL7_uJ{JzW^AJpr?YA@5zl`-l%n9ir(0xm4Ku{f=bS_C%yI|BmG3{v4rQYa&V? z>c(Wsg5+c-h>5A1@!y;9R?}f({L^VKg@WF@SKZ zc;YhK5{6fG@iG%+CH$YUVS~eYI8)B_&*aCKpFXM!4W|QPNY60Z7Kdn!W-z3E*^2@X z8G>PmE%*g~!r{_$it$@cFZ+I

G#Q3wY)kcLc=RvDS2nmR4V>Zjna%0XI$*b{{Xjnv|%#G4* zt57RQcrDB1!YfI$!67oTag;cEtjob5appw@twhD>_kY zP&AVdkLygz4^%YVkI9XHe||9@oWeBWOSTXXSm%QFtE&`u{_orzPs{n@i-3;_iP_Hw zaBhSH{;P6S8c)WQuC@!MXWQ-Rgpm1kYWUuNd zr%uN6`p&0IycDSB{ju~;Uq!xu65QRj4~nh|=(U??V3pGd+Gm$PK8)Y+?ZxvRt(Km{ zF{i~n_mh=7mc~r`JY}czSv4%d#BU)mDlOU z^zImW`Sy3=K}x;UN23h9YNp`{{pN$&;nVc=>@pfWQj6w&b)=mGRSA827Y3It!A8>f zOO$sqV3NRvdADEP>&=)RiKsBWj;%9g7(T|*t@J$CXbbZa?;W8ZY zeqZ5Tt{koKl%aFx$AjUnT~L3)2|5y{(3c^~box>sy1%F&vA?|w+r_Ar@N}h%?cvcL z>6!e=dthCEwr3@{yX7h(f7h3J>U%*}cE0q^xe#eQ)0O=AO~9u!s=a7$H!Zlo(-U8; zHKB9u_Cc+R2qt;lfl`w&iNGouhaB!c|6zS@BYZE>0{^=qWP@rS-p<9Ur<6ov94pc3 zt8x&r+Y27~PRD#cpU9e%c(QmXS@kwVc$UME?rzVQ2?}PXQA=q&w~hs3`uW)v2pPiu z*93f-xU~vt&NYHZMfQ+f>qnkiN#~}CvfP*fCqklre>-pnx#V?CAvz;Fk_rNRpe^2ue%0Sk zQlx!NoV*qkmTFNiOA2vo&O)g0ATJ+*cKLYZ@^m^mdA&EcTqhRVznw-@HP(!Yzu-CdLI}iBnZLnertYk2^9C3aJC* z*|%W=Dm*UrmFsPeQ{IonPpe)r8Fw|wLjF(_ESt6j%`0;TZ&HbBOzttA>05iFlXfRz zv->pA?K=j(&sU)ZCnfYrg)}CLVlLepF_AlDn28tYq>|v+{-wga#WW*509v9Ebv1c| z>O2Luv)}Av{RgjS7w_5RX-X7SUK-DEo4Fr>X4@|5y)Ax>#=>Z}?;*IVae?kTl}UTO z>yYe-zryP7wf+msl@1veBiz+W}!XfXQNtD#LF=F~X zxaINz{4I{a@NDUso7dI6p2e!!hsXzcS=hW+dbYs64^=XIWG$W5lJLvNptn;oISU6M zPT!9VN;l%xznnpu^{kk!&q?n07yxn`lB9Q0Cku}{zlYP4 z2GCutp0Lc?j633gks~Q`Oo?}uNOYdDvqo}%?-l)lG z$y6JWOJ29lKv5CnM4@ z&kvslsHLOz(*3dLGcUt|p^He0@CI>tWQMHh9|zX_ToO0 ztrCk~oOTwKpS~_#Gk;HltP??R_yf}9$;L6_(?h0Qkjl2YX%`|As z^#rE>u2x4H_NWX>C-N@9QZfL*_*;qbe;B< zxBI6zzPYv?Z@#`A6urk0qYZoE`QdudPM(E6-OtCr)Rv>aH(n!=y8&wtmsL2)uUGB@`3M`Ylo&SjLBpjD4fb97#R+th1Uv7XgC ziOGRj8u{x0XuVBBiu;vt@FfLKr+o(0bTB_aah`sEj+e((Uso7rBZ84K);yi!;^3vY zG~)zkWmW`Ei_2he!(LjPV@!=?9^*c`fb&OJayLIkW1YKl%nlFNJj8vrtKl!wef>`_ z*my$lIq#e3^3~B;uh(82QNIFvY`KVwXPMw1R@2bX3`@}=bw6?(&4;Y~snGvQ4qP9X z4MJ5{ZjkaiE_wPnJT&MRntQv8&Hc=6)wz3z&tv&T=Mmojmh?n#*<6gLl~0p9KtGa+%PtF5Z9mZfv1z5xuYYVWABeW@TV$wsGT+vh9!O^&%=Ae zBH1=5UwI3d~}W_`&L8;3Pt@3`|-9RFdalCdp_S_u+w}Fpom>5_|=rBlCg4r z=&o@uGQrS7lKCk}RF+>G9wLgQilL@z>gC}|c7*|rUOoxdohF#`*1pKx-Szp!}aEp)S~nq<~Zg6EIo z@CoU@+Oic%Skqz%^04=Wk~!H}E^HL{WG2U1w>x6rpj*gSFCTqbzmw4%rWV5F_*c0W z{ZC#3pI7c&@uEtyH2NkKRv+T&68xELN~R3gCBFBAaFXf`zKyss|Fgu1OU3Ouv+=_F z>ZI;ZJgo0+iLE#NB(K^FaEO=mKCFe#*z~U|uDZD%zGiqrLRKzo`#)UbO4Hqpj`Z{v z_IqZ|G1RlBdmn^voh6b=xc7J+-Z3|_MK0um0A0e7c(2E#AThznW}#~EyViZ$-l;XPj- zV;9Q~SktD@X)G(nlPB&LL2u?S!EA3#j0}Ut!~4He!2Vrmt78s2Cb?%#aj0CE`^>4m9`V_8K!_h z`6d4yGMxNx-Fgc&&)Wz-^wPqQOaDK%-aIa+Cj1}2lXj(OwIm)Sk)`FHYf5Fy5)s)V zvV?3QM7Fz9(MF{tr6Q6DZTDO!A&*dstR>1?wh%3Le&?J%p3mp|dj0;G*L|NebFSq* z*UXtSGuJgqO#K-$|6?nOz-NHfCfLBAo0xmDQRLh8di(cF%QScDF~=O@^mc$-^?msm z2WPhW>?~MXsSU$UwK4BTThutKi$9|;h;&RkyThaoQ`qSaaUwPX?bAOap?QEJ9j|6h zZOgXOrPEhYW25OX{eB1NVHae*u=4{@bzBcS7Mzt{bZe7eXet+EkQKYv!+nJ#M8|hN zY~R}OAH63m8pqsU_1|@lMjYtOdkFBfh&X?0WLIBx!txVJ_{{$>Jg6!FsYCnzp-Zof znBB^Otbdy_b-HUz&nno$o$5nO>1Z^bP)}jsper8UItq&2UBLH9A#Cx9E_#@Q0EFDFWTX~=+rx*p%i zg8Geut`Z@cB~~$76wdx#w*$X`GIy!?AN*D}h@G{V&c^#5!6c6xEckY!bgYd&(FqBV z4&wJT7&-4@f!#tOI(RUYw515`O}2Taj4iTRiaku0v)Y{1XxG18)@w?;$^HUo?uy|T zj!ll|%)G-n6_aOpxAwTS@C(e#4oKc}vxNYQcE~UQ) z&J8;z+P}bOdX59e@Nrh7#*N1*#@nqM^rxYj^AIkr&t_oT1NKrUwT9Y#zZ!oX`Gw8sKnytCXB6;4N!{hMN$ ztHEbMFUdq3ey3lK;M^WJA+mqvfzzw#-q5o4ARAuj6R3UbkIp#_uzOxxnWd5TvfnDO8q;vJP3P z%=L>7;cUIr*xAFn7;C34;^yFQh!DJ(u22uA@_l)9(yx)EIKG=}0(X@sW+UgPQtV-12Zzo#>81T4G{}Lj-BoTVc_%+S->|37` zB<%h(lt1)q?vUZM1jP6UGA=7$ zPC~QrC9?CS?ggWK0u7k_c3!<=?v=Fx}v@@H>u-U}~ObNEj5W|^^#$M%qKr>?Sz zbsURWeG_{-w)>Q7E<7lWTV-qN^Uy6 zj5~TWiAz2l#Vvo9$z5=<=M>a#Ktt9lNawa;_HhUPOpP|=BDX^8mf0}*RcI+42|r zIaA)3`^MITtYCu-O)6J$F~1d1{mxFby0epY3R7WQR&5aJB~1v%eYfr-_FRF>4ta1z zSL``AqQIpcIfu~eFiR{hVYk*RqXoUfY^rF%YlVBLUT4Hv%`C*=Sq7Zns7N#n-o)ul zyT`n#pFmS$uH^|?+cdZl+Wc5&L=Kt_m0`NlRpMxyK%OM=r)o%AakKO|zI?SAH(L=> znsosmo!cZ09F!u>Eba()YR}~X4KIXmEy>@cq4aJ9Uk@lUq zA;k??QkBS^2pGonQhXRGmAhkY5t{jz&RTD8>JnBZf$omG(~_1bv-dMcQASjv5o3__FX1Gtf{8C-OY zocreh;X*2w$Spr7k!FKB_^RKFcF;Bw#$1wL3g2;VYZCY4?gi9YM8tcycfVwu-X-+r z!=Ti@ZY4pRzT&jmIoQxQ4!yTb66GnnG>9%QQm2;|T!S`p1S*CfgE@QmqTiJatUhkQ zW_qm`aV$NQgym!?XT`6Z_3}!#d3>XNd@S(ZWAq%%s&13qe^ro#zZC7E<2xmMCFzY} zn>OP-pAp=#o+m|IjTZlq8m&>{-q=NPmCYtxot>#DtGew`>E@Fe*!DFE3oEZ8ecRAZ zho}Eq;T!t)N|Ac~dj@g+6zlmqf$T%K1Zl6kt!Vv5n=AgR&2?MnAkdLDnaW79PJeDl zuqwCUXb(C}xj;I2Sd7T~uU-B9&C3Z!$EU;C-d_P^(EO7)$4`w4d0M^v@$9YjIKW&^ z^6SpP4048C89a>oke#snOA1zB{Dp%r)Zi(PVeoEj5n2E0H~Xo3j9i&tLb?yv? zvI?lZiit{RA^&zKfu`(|*(~XLFazQLlbbI{)7&xS{`W+pvC5dZuCb?i{>Eb7qYn)1 z_&8d|wR7n#1$R1v>xu>6bhx&`EzGSvO8 zDqCwsQ;Ws<6PZBU)ggk;{*H95yn@;FUc${#b>PpT(SoYgX7Z0A!RU0!jIP{bOtU?1 zF^eDHSWmA{HosD|sk#1Cx_}K8*X8+kLxUGE#EgfjTZgcM0xISROD<_ufnzp7uvh!9t4zHk)jO=0w2l0 z^Nr#9I(x$LJ_g!JZ^;W*2s4*n#;KO~;q#9=IORNuK3+c<&rWDx7pRCH2$J}hqKsIX zic4Nuk}p%nbJr{f&@nMRZTGD_PbWAw!A9$|qKpEM#P=rn3Y_8dXbBn2g)pOG8`*J% z^LThw3f<(H0cvHF@pat>+%mYHUf;aIw(p^l;ipkEwyNUeIV&ZV`3LAHu_xJj6odtf_j+rmhS*7*_n@jOk zatjlC&>44)@6QdGVr4vwxRTGVU2ETaB*-YF8h#A;@QZ_pi*T*lYDif30`RzeL(oMQ zQz4#VB{}LFE8Xqo3L_RZ$AQJ_g19qH9Pe&6=$yVALtd-G3NwaUM(n=ZEhl%IRaPs2Ij8TB2wUx=Kb%wTp`(Y7Re&YBa0b+%+gl@a{$nMfxrU?gy6i73~+v=huIa-eAY%gSbo@^9a@ zjl91QfF|Y{IJ9Srw4%cb7_#Ms&{vE0rJN^|su>Mqv;P?-w%ZeP62!b2JrBVCx=e`t zRKXM`IzyJxMd=2AIcOX#CRW|W{oNw&0ygTeZ@}h?&0zdv2A$CqPy5*K{|~zbbyr2b zlBaYPx>2v?>$SvW$kmhLt$ThH9?Uqi+jo5Ht}&xezn2q;8?1a$TxSE|KE>#Is+~k zS%YY9hzSjR+E_4E`zNbk^IUwVIu}L3mZMWKac~HRMvkXj4&RhFDE_3f?Z4qVo}h0 z4=&u%BF5~0m1cz&N4n$GPglr+Y;}AvX#fs5A(wx z)8##T(4p_o!lZrIz_hsoPI=xPdYwN?`o319G6OlxTe%D44`s36dgieEO+3-;eFet) z$VEMO-tiIZi_Dz?x7j1s;mt1T{ohxm6F!Xvk45KMb#)Bwqq>!Tb7-H}GOVv>35i!h z?spTeCMj@6Pk}n>deOHp^WaC7iBzK@SFF1S3ivCs&s%qTV}dexqN2#lq+SKE}kSdX;#;QEhs{E8a2>}wn zunVe0++`m9lo7jhB6ZR=0&PjaKLotO{NPvqadvR)Q1JC#Av`Pl;?m#NY4J4Mn*2CO zg6+`X1wqB!0+b{P?APr=W*o@-9CkWK!hc7KGODiJgP!rXz#uRiO4JI$#A6hE_I^QZ z)0P!{Jn?`;?mhr}V?K*K{h!pzecs<@o%yq5G_`NZEe+1Vs_x=EiR{+fM){`UP%!Uw zs9@K!K|}VTkE1w`Dl7DM=41Dd7WswM2BMIkUtT_ zZA+4$Qtb;n*5t~iSN;LV@QVN+Q`qDVr5N?=3NG*#_bkW+I<{M;Fqg_-%wV%zp8aPY zH#cY-cVS!z=nhteFAw5Dht9%<->1UET{8u~l9?OTxlD@@cvEghx72o|CrK6d;`iJL zc~i>r*t;d2xE4EO%=oql#}2T@VJ};xoMAQF=v9epwcF&5p1o+6Y&hMru_vrpkwz}h zTg$Ew+lAk6bmvsF_hamx!FYE?9FDy*8@pr!_t@zGbdNTn2ht^+X=N=|H22}c%Un5u zpOCKUhO+q@yuZx^Ev#3aXk*!@FK4|?lhXcg<&F++@F)5V3p@WD`#81W^6TXoW1_}6 z7He=zFPuip)LzuGVjnznC}1~Yav*j^DSuW#1Ws|uz(w_k(ATgG-s#>W9S=L=#5Rt( z{;?!yYzN~gmv3OTgOb{o>gOk z{q>coS)R%rs7vCkk00Qgf2`x){R!oKe%j;Q?p--$Lk_IZpxiI&6A**G+*_wByg&bakL#cJ!lG^;cl-k#aItz8E`n?ahTo z?dDdkFtcjCp2}`YN=U!_9N74DA$S~bpBt0hdB9dI&t{dmi={TR_Q4x{bs9b+h*tTH z$IX5BqDImNv|k+so46r#^4<#Rlsi3as;+e5V~4vk|7rnPGsAXw!m?`8^oUoL4$y zSyvU}u_X$-&&xx_yhJu-e2IvefR|8CNMnroc+ukz;R@fUQvZ!id`GPqph%j3-Dj7N zzhly#6GggN`^RCMbpnRo@5Y^-mI4l)g5mc3Gt#V4Qjw<7fpU00myvAcN*q%0p5^?K zke*H2V*3{OOwSU>H?q$(xgcZY2`PaO;Az7Ic;78qGw-u|G)VWE;BX z%zOC~bvH0~{>*JJO5pDAJ&12!MWb3zCrsx3fCnqDnBgif3@~e*J@2n*qj`o0i z{T8#_mNB@}Sgpho`9bu%5f35J z8&KP~8~5#g6PJ5^4A;buuY~e{69$NTK*a0Os}^ISe=&o^?#$qco_z8hRqk!n15sbK z^n(!kIs!g^2*s`2zvH#b^EtI1DQu544t1BW#?rcpc-S%mU`83myso)umz6$F`$El4V8%^8itA^dFmUkPR zE5GU$f?uXdut%F4>Wz&c#hYHU2a&q4z0esNH>hzsi*s?zlG}K$OCVkwz8A}-0dVHs z23-BeU*IXxE-@13|HV2z*|SdUABF1*gK45na$*I4CaKZF$9LtG`M*i;*RF86&o&4R z7$!dN`&*PpreD;LGx_uiHRngjHBuI0>F2Xpv8+bK*V<@5nZd{0R<~#${|S7BG+@I} zZu5|af_+gh;H|TUw0Q0kOtD$YR)HQK7_^TppS4NaY!M^k@N$K_eE-NrFmqm4c33SQ zpIo`lM*e(_tL_Kk!Bc!-#lUN*T-2GXOnAUb&R7uRS{K%BD%OMIUQhhoRE zyNz>jMU=l-?mlUhDySOb`612BdQ37pE;QhFh1kIFeNGU6(hU}FOabrfx8cc)NuvCJ zb=D2MOlIfbBS((hlgI0pu;-`3$&#Jv#B`$;UaZrnZycOx93M-ru*(1}=I5$kU%EsN zABhCng8)+7ArT!Cow@Y4CrJQ3hG!FMFq~cnHU9jim)={1zD;&F-W`|No)dZYR~ioc zoOR`64Qr89XGx8J8oSB@U3CBC_8EOpA6(C4Ls@ZT1;8uO9y^t4yHP18(8m zNnxUW5Qi<&gA=d7h|D_pI5`yaGWBtLhzcsrs|4)AUcW=)d^tI${OHngI2JPN!7eE!%_$T<{Gu1!0LLyrXG zzu8CF?Z=92SW7&3O)P>g?~8^+43Vk3&VNled)u9 zWZ1N~&F0xlJxI5>0JXa{MI1-&Q2wi%h^J=wG*&aE?|=6u`+pPLTO#|+``UX5^(Ps1 za1xmDyXE~lD&tun!rA%%gE}6oPJdfW5%4MV9^8VjB(!LpI$(Pj{sMBSKXqy6$0UtLb+ccJt&_w9r`w&9v zdSg##D{@^emK?vLhds`81%r&oFmFdT@$T3K-n`r*-`y?hFJ=;b%X(0^)Pf<`ZsVG1 zyYNHyufKV+UFUlc#eV|HlSU6gZb2u>HZM!a4R7M(|C->Ud_rD^M-!PLh*%%8m%tzQ z9OgRnGtU$07iE(WD8!1^ENW!o{-Kv^rw;6e@lD4 z8i?h7eg5Jmu`AAl;mrp0hWdUOKZ{?(O^pGq{{H`AD9{gnKS-c05zOwEH-AI!uspSz_8&Yaoh6@x_rJ#c7Ew8@RDW3 z##n!74k6R8^+c!lskHfcwRGE=Ofk=|%#56<%z$xYm$Grk{P0bHCvz(Aj#ZbeMcl76 zOu+O-VVJ+bmeS9Mh50|(Zks;^_K&-w|N8l&o~G|Uv+bus@td*&_0d~OpF0@Q<`WBv z-to^QFZe2n9I{QciB}1S;R5e7T)`te=4E+^xiKu_D(n)^v3pAKB5C z1}rDEn`I(48b!c#ujQDR=)fKao`mLBeYo-E3p$L*7Wm2@^le|~%=@NGI~{&1@*S0Q z71d^aB4$t4Kv(SmI9?tJj~^+}8!s#&k}<=2Jzn_MpaBo^Mc@7;W5&A&s4B?cZ7b;?SAWru7+b~MHXA|{0Uz3 zJ%i8qxMRsfkHDwZderuX8C~^0TYh|YPrCokby2^+x*r%3^WXI|+v#9AHjVT)??pcq z1loMckTap&T?k?)|9j(h`V}Yob|P4cl7WS310W zHI^^(VhV5Hu>|)i!ZQ-#nGx@PG9`aE+Tl+++)~XZHEj>r-ctSyj?N=+@742Ib-WF& zO4|1@rrwxCwvK)eTQXJoy-*qQngk~p5wj4C{MO2yo=k=&nA=B+5} zy?qr>*<&Z(8K}X&jWxi0xJ$SkYx>zBp8C!|0~V3Wbmx~15W2Uwv}-RlZ0M7~J~U?I zl64ud)8aPlkhF;PaPRkU(LO5fO(MoW%SGE4ba1<&E8Lfi9oLJz<6YQNdfsJshnvye z*VCaT{UDh;;Q5B8u%%4C6$?AnDPawYp4ND>XWkf>`S?<7ekF4EhHIme!mOMm)7 zR?q>ltlInnoRhA~-82}?N%x%)>#M0s34M&$0(Zgx;yK~J)ezAM}xPGf+cW&HB#$7x+ zp0g-#-&b#lNi0~Y3wQ1IdnOGH;~pf|78tD9Paf`4G|`*Gr&I9#*VT>REq)};ogF5JQesmG$@QaiZ$RzcLiq(fJ3 zyB5ct^%zmOZ})&g?Imtpi2GU?dozkfZ*}6{&n@TURcgsEb~}MRqCSf9EX~Q3bFr(K z$AEa@S|%hd-vR>KqZrj+R4KGMpX*M(Xp)4&) zokfNiORN`fJV>gWCvX=X%(;5QjimK}8vHvp2KYsPTd<)gUXYj29SDbROz$tDke1#=-cXL9#rIaWAur^q)y`6`U$*FXBc2;wRy zg>nyQlk|E+j`&=<&I`8t!!=PK0f*azEyVwYYj<)-zwX0ZHtqY%g*-ElOlj$XDa`NY zL$*G^n9J-XW1}?k@mab*8V9aLMW@ra_u6^X_2a*Vs^SFtl6;fR+{(H6WNEPl88>YK zKtDO*>|MZB?GwAf`*Tf}cMVRjv(uilut)Pap$=sy^FOlM zz!&o88Q$bUsy#^04noHoeeQ_gGBIB$^Ecg=eHZmVG^-Muzou|4w_b2(S1b}`ROz?6 zVA%FsWS@B~yID9I?>wCZcADb2QFi*Q6x|0_a}T#l3cqwQ<=V8{*Sf7c48-nP&UoUH z9d|}ok*jL_{2$Nl4;$sj`tE^|lY-e68#!BTS88?WjxDpB3g+Hc$WF{~A z;yS$$`TV%?pni*A!!mAQbzOcj$2$^fFlTRByi&RlF|H;|3m2=*3z*y%4N82vttD>YZ;W^9cmii5&M z{Cb?o!a1)CaA%4!=fTGZYXNIfuGrWh z5?!}ml$G0bL%!&`H|_~vj|ZFiH9Fs+IPI-9x542(dYS5?<~3CuUha#7T#NAQn`J1_ zcxAuq9V9e9(ekR;UUtvc zz=@V`S^fkm40N<4^?RR@>g{eK-iOoWOfM#bsXxk>M=t)3$G>EV&;G??kdD~zi@5aF zI*3!Yb>$3K?BP_krgD?LEKqB}KP2hv0oFWdFcw+G;`j*z<(88|aC^%x5*2t*YCdZc z_B;6&BMR)$dR>2!XUsBAT!;U2@Dz4_dKxGDYC!bFId3hKG7`?2ei!=jTadh~BvXygukXyROv698b*sPyU+F z0_-7wF6UO1iStE+Oh&V&re=BBv^~6k=vVe|QyxAp^d)c39mHt9Sj?K`+@$7E7OkcP z2M5{;*h$nsIdlKIZ5C}>X~H7$-W%Ns@NzN zf3d13d>n&Ap3$x6uM%RjUW8|WOSBGlCt)zX#zGj;E z{42XXY-r$3qUrU93~_G~%L(RSyI^vw?8_&>dJcIs}v!$|1j~zo4^3b4Qr8 zC8H}E4#*I39>{gaXS^R{NtTRsnayEv*-c_SwhPp8pKx32KK$1;l*~T1OWGx?g_M@O zl1HrV2tQIMfJ?XwSnaPT2Uc)kio?LhBNZl%y?`fPrQyUkKHTK*;pnn-#NT(1tQ-2{ z;%&FcvnM@8>_=}tCCXgBI!8W#bc|Tng8W2(3AxGl3rBhc!IeWxNr8%!DEDjg4{YJu z1>EM}Ay~FL60Sv`h8KZVP_tzo)NX&x)K=*IpNzLl=`D_56B@ zaX5UKCX=u7(V|OS`r!hzB~qm&c6j1-cdQ>2NSsPd*j5h{&bDza%Pj1Ic@6wJpL!Ie z{j>o6k}3%5vInmJZh{;3kKx4O1~etdSjzq+G^%gHP5}wf)omW9GvF|+ou)^ZTjb)X zPn7qCyusZ417YWycw#WZhJ5H+4#yNzAS>l5SvS%Xe-=%ZdOU3RaVdd*g|S0jxZ#QC z$Q1vhsNN$A<}3PhlU*9P#mNKtxwJHJcbSUM8g__S__S2vhv5U*_08+>oPLyek0@o5 z+?dIvNKygIQu@NZ2shDA8aj?=)crKuk^YCR+NH>JMwR0F-J?LGXf1vpx|EFAnF$l@ zL!@tx%0MaE8m?-OhoqRV(x1y3p?j+%$t~+d{qNhs-Ek^dU~7TjZ&{Pyh0XFYfiA=~ zq8`Tfx0Rpwp3i!1+6#WAnz%Ax2Sk@?gJ5ftZO(WqR*h7y(}q6>Ohh?VZ{_m-6!Fkd z{875*ffd|Vz76xsZ%e=AF9IShg<;j=_k=_yi=T9D&ToZw{WC^-el0TeL%&2OrB`$QRO{ zZ$n|0!f(-zbG2vVxDCbFQ#Osvz586cFDF>IFA>U(Fztk%uX@7DIeH{Tf8bv^WdREO zx%6=fY4*bN=Ig-!{gT-3unGr2y&N|c(FMRZp)_#I*5E{{MN{S*k<<^ zH_0QDFZjMH1*Q%=2fA8?pyO_WNz=x_GZj-2>-&wV?4@$df6rSMsp6OD#n3isI2O&< zrji0TD(J1Tuber!>GSg)9pw4XwEinEt(%KPrBbw*7zDCtGK3Zr2p9qQMq!KRUqWt?SOIM2126_-AZt-b)hn{xNYl ztO=u|o-)_na`1LKj(c+XIpK$WZI-MY2A%pyac%EDY-vj>ce5%NM<%&&YAq7bIDJvj zS@J(PNY?(kEB8G)0n>V_gF!?yyOK1SP0yOiZ9aOE#SQ(*7CRy{?A^YX`E^Ycd)x9+ zIyc_uKfPuJRDg9;e>fUy31bgVXYPxK;Td;h7Q6bOJYql?dg24Lyt9u)ygA8MEfRg7 z1m0rmrlZGcPL%$&H;Lbp7A!bz&#h~Ik++u(7}BB?CR<_Y{JFf zxXG}Hd464rg*{7{e~cv_-Flqq#EWa!vPz3qY1^T2`17Ydj(O*Ij-=jX4FrW0IS;bmsFT)h_gA9vBbRVz^nzr?o&)szxcl$?TB=6I`;H)Ly ztlx(}_8n%!2kv8G-*1D)6B8VK%~izUx0g50jL?z?9+HaV*8iy^neeMBpNMh&!?7%( zKa@ONE%m;WM3g?q3iTj~sZC{@p2vgj@pLHs_7HBrX(0|D)^Q8FAL2B-nsVK8FG2ms zk78N72?e75Sud@~MHUAKpFR=q7p!g|o_Qsr-tX+Lpl9|TeDuLUxRy-WQ;Q>=DCQih z=jWYD$PlL%qN#L-RA1SAPpR*h=G2ZCb%>8s5NRyz>BW*fbjAOI{7IL1e=GiX zL#KS^X*rp>8@KljjrE=A-SqWfd|ZJ)x0K*V|cVk z%f;s%gbTawjMOX+k(0+<ePO>$OntUIb;`kBa;> z_kM-*MeY6DBU44OFH3)E2ht8!@H?jynek*V)FlLpIs{yKjI+%7b*Y)@sB+F5H(a}n z>ksl{`T@S|^vrLVsdk*be|gNtR;PtHzLt<5xl`cXMPquQFr4l`{8Pm6nQ{;Y)+ONf z!qE^C^_pz|r9=IlFGqbNM;by=>(oO$bX^8;6~KiDSLOOH+zbn`mH{JSHGv+8NKtFcDP-R*?B&E9o$Ysn9WR2PCM*~#qiwc1mkKgAG!WF2_7YzFC! zTD<9NLN<7I;p{u+;E1Mun0hf67Z1}H+u2oq{`=U*gL2EA;j6iE;%xw4vLZk&n2cV&^msr1FQf7l{RWO*$)t5J1yiM%YV&(5vFR_X0bcAJf zHWV@EpOu~3q6Q;EZ?Nd1TGrDvR@(1*2cTVT#C&yr{E;r##br}F;D%jpbbsAlk@l5$ z*D*4n6vpH_Q2B8;`1fQv9Bpg@i-);l{+*CjaCoO9l|IzPrlWVnc4=DiU96M!pIkxb zh6S4U7uRP*>^@Wot3n1eA<6SE{xXS2B^&%_4S4sBV$@+a%*>1>!NV8jl9J1KVw`@qWH{0lY0~9Yxz&|fnlp)BR zpTppBlk|uI(;fdngH|4B+Q-U?$z`1EeGBK?`qK)o1|F?FMo0PIq&K@wE|_ue3^7@{ z0ySr8;NhX2gy&@fjnQ-*3-Z>$tKp$&>KqLpUW?dDRNh@-AqUq&lYap-g3T;6>4Au; z*#vQKyku`?F2*fkSQ+(QzM(J;v=$EMZtk?>Mh6b$LhqcDvjKr{p*6gqgi|0}OMgfM zr|97ORyDl8>7+;}l#d;H=3)_3xzPc;r1i&Fv0L!##G5QOQv9ZtaGj!X$b}1dzOx$T z&zmTaeP6+!CvHhc_m2R{LyEU@w!<7Sg*5LkXcE!^9JP8=iAzuFsUh|oGWDl>K`Bj> z_d8>3uWB+3R}X}jTYJLzs(9eG`rzc7M^JxtFFqdD9`Z6`nAl(Xor#9a8#3iO%{Opy zb4T#8^MgU3sPuvUG@8rDNL+X`fow1i!^|Jsn8D&lC}5cPGeaJFbQi{Eit}cYPTXoN zTsaP(mu?hk8pLbD@6P(@_qGB?FA1d0!!2o0@13~gwFDO^6|TPWb2xiUYD3+(G5KUx-2VkpJww*y=bNez*d@7Vf5ZvaX4B>#O)d zw9kkqPW)WpE3?^hSPR;0-qa^)T_7A8rZQae*aX z`mpW);l0)PGQ%QeJY4-lI_Qu(A9LUodF1wz{oC5na$&S4U1sqD`hC>^S7rhIQO=){ zJpmQYAC4_GQgqe2jQb?JS&ObFs;ttWxqjgS&aw>GNu0B3 zCr-J{iaV3&%lbz*3Ni>-$UYcY|6d!C)#W|FQ_mCgpUt+06)~&+VkYZ6X*pSTYaNCT zjl_ZD^hH_Hd(T1jf+!d+)uH+PzNH1rLaDE7RDq7}VN#uZj*JVA!QAd^vDh(JEU%w7 z1=OEL;5*f|xS02pc|Pi+^sgKP-o`-UbyI5OxROy~&OsxNO4D4GE zP>a`HgmZv_)y<$k;zFU>bsOw2w4@8vN7RgZ4C!x-*vWI&?CnQSF@Nn?D=zt1EH^7!3!fYwN>8osMW=63q&Cl2z{C(0 zIwVPnx^2;>9lxy=%m3j0H8ZM0MO-ptCxTP%K4x%gCrrB88D4uXfOY1**{3fGY=lA; z+=z50Tj(~f$B^AHq1Tv|N&3_xBDf_2zQZg9>d`Ue8Vo!GD5#c z0dVdle~xA3P*Q*9Dhkgf1>c8{L_!t&xuC)PN;2-@bNS<6_H5iEIq0P8)4t)q!0~xw zzIOIVmU?#mW(+o%PNNf|?HZhaB^n6)b$@UEUe$r$J(*K=Qu!w*koHf2u*8D+i3 z6^Or%X{YG)YvMeW&?Y25mw&{zKAxQKT;A7v$wr}lNFF$>WAv96O+A-P=lw9J1^sS| z^tVkh!P3k!ko=iNy@-)m2N`oc=+A~f(wnp2LTS)Mn$&x(cwgY@-d{pB^!Cv|&hGTf zrygL~qD`W8+xNZ-e2&k0F3%XU7LbqOB3uj42x*IHDb_rhj0OeT;J@euC~fk`F}WXb z-cc=>_Q4iL|Jo0me%Xu9jNL&n#%CgE`@h7G{%^#4>gV&N$4j4(zPBfk%QL&ezc&5Y zg2oD_Q8a;cd)h$0=xc#}^sK);|Ks~PX(`!Pz7lF4UxbbPKBDI;AymP#md>BvOzX;j zz@D<-#A?++w)egV%JODotIb?gnlm02uit}${M~;ga5lP^xS=~cLSaJ*+z#6S&)z-R=ebTlknKS1&e$%PBLe+_Xrv zkDvdFHX(bzdk1?SMBvG_X3}LxBorR`4SG)ZLDycB2HUH^aG%!ytCz$lqL~ch=7Q#DUNa$HD3twS3$PLIL#Bm?{s(TY)D z^XSwXSK4YY85B}>b7TJ)(S)^cF~j;_`Nq!g`SW?~=~0CIx@l$dIj{Mb0#mR zYrkj0{O7~zuhvu0Kl~Z?P$g8`;as~4vv@(;FQ0aa5XJqQqL9# z_Cqa!SZvdU{j3J{gJ= z3~XU1$uYqOWDB=A;Zv_m#sg#-)wlV_OMcj(t1RiwQ`AZ zf);%HVI*SkuyG%F%8I~b!B>dx6G3zId(hQb3B@DwVPy0gOd4KJoV;e!b1RS2K4X)W>&ZAAaw-wuNH?R=Xeq?a>`Mo`Or)d6Y0z1vpWxUuKE8)tAd?=Of~k#( zRI^)m+F@e=?N!cbN(V*!l$eZN){LYBTu;NqHBAuLv!mF5)IF9`gO35!r<;?=H`3n? z=h&KXUs~Sa2~~>UpQqu;KzCHIoi1(qZb9DU&4$+tufde1We`oZsZ72Jlp=P(%5_fg zYt;>K^D<#ZcTSMb{5j+`N&NT6olVpyr-&{N+e81j-;fXW>kVb4d!(aEIDB))ocpZ0 zmWx{y%hhd&;QCwpiMYleih;$)zrlmcKfu;01G23YFxa9W%i=B)?eYH9{&f(oNzkO{ zDhOt*H>Nkb>(NV{_%UGOYEcKp>tlFdtwnU;steTec8S;)qqSC(`KOnV5#A}Jn@amy z=f}SEZ#$J(nMlb>6HcrTyIuwC6)YeQyJ`y#`E`fVK~YfHzYIF9LGZZ^gaiR+Qrm}{ z=`xRQb{<2&)Tq*1S4`n3zmD~Iof=)a=?IK7>Pz=eKLRT<;$fjmGD&~(O?o+f9IRC? zgE<-e_m$&C@^9Kzajdem=8)Xr=p6YNx=Jpd zohI9;ei=@=Cz5s3caf{{bIJA4BB{!#Gtm8775tN53+v-D;kwyn;yS28^u38`*(PFQ zdE<%Ljs&~%oYsRHq_=`Z;se_z{e+3G1>os2k93@n#*UdWJ5dcyy2MH?!Sq)MIzOm=VUM0vTupVT9L@U`<-+8e*C__-yicjuRC|< zS>Df_IWy19JktgK_`M~OKRrZQ^*lVpHY(KFaAvbu-(M$3_)ns2-=D71lF)6=yua&z z>c8pj1=rr@!@dsPY45{8SKK%S+^y|IZH*DFgyl5y##)f6yac}`;jGt_EC0rclb)9l zy`a@Z-))V-b=5 zm(XDNac&mzE9z6$Zueo@g6$m!2A0sYMc-&hU#6O)* z9agt}f3pv7CW}g5!+rk?&~SDly#LflS}x>De{TvEWmGwFNR+e9g}J9hcq z8C{4gdZ=GJ0f>_IP>QRq!(?OQ}B~$$&T(&O@xCs#512(v)!aa?pi6$)gp-+1oMEAC zJiOI;B%eK8k(xXoMnCcAJT`6WMHaj;5bQ3)?bqO)>n&onAO?=rR>O(IwIVh(-XEE} zNiR6sut-jOKP0c$=+gX3PSkhB-L6o*<2FmniWX@Ibea-|;{C3N;o0>D=+sY#W*=#R zF)vl*`*r%$heloK43`xglAUTP@%(V%9UIH=V9}gsu+1PWc##w6^RC zq>Q1kq4EMhcZi0T6^e3?9m}x2$$Hp?e4O8R0rC&J9od}RRQc&?Q{YxeHXNAN=9jau zc7bgkL9l+-Q9|`@2)cJ3Znr{16 zEXXB0_R<~p&z%6bCf|iml^M_};}`j=k|0$v)Sy*{i|Fmy?P=xIBC)Jc{(fUudNZtx zh^_G3ed``}&te-IK!1#x7tu|3!e54ai zb2tpk((1rrK#GV@O;U+`wyGAj4;#Vnzvw7ptf%24^8H<3ArDRILCv&!!E$pM#F!q3 zmpyX9UuzR2ZaxZYwRgZ1Nt?gv&F2qnTP`1~_2?4#x}qz!E3l-SY+n(5vk9Wi?F;{a zN4rFNzRHQNUhqlG`!+>|?u54JXa}ZSk2p2VZU5#)m?EODpUD%E^S^gW?>p zzIzV&JjEFZiwBxcL4}XmS=^hk>S0mrs>UPsSUOR}pikg=mZAgzFR0PKnbi<;(Gl3Ay1`YVTw6?Bp$mUcfSOV{_-rU8 z2Z~pMEHoaj6&HZZirZkiUJ-c=B&PS1;jFSUTy*ls{u%2*EhSXQ7x4*L+Kru40IZ)e z9Nn+Rl4<2FWM-)Nu1R+I-p9*Lagf*hI2d2v4L(!nl8>6X;I?8QUEUB1fA+qBuZ!-$ zwm;qJtQpGGCC--qF;}rGc7KhU9mVxu$)xyf>=4|O^E&ItH6`3blP|+bo@))nT}y&# z3p>#m3l(~6)7*dZ{3n<2EUdVUfkTgqx(a#-$>!81xxv8!cyTK$4Zr+XdL(e5@XknX zjPHRrROg_@Cvks^Y)LDBAF6H|zKOKp3Zgo48lj7^Z~rc&<*7D3T(5xFJvw4Wj5Qbb zD-ktbH~(xb413S-Vz#582%ac-LS8Z=%fEo(g4}u>Cupx#=p+zM_GT zYqKz+$A0$vN-b-$w84&#@=#&OK$rr@PD#P=F6omdCur+xPg5z zFJXbONaQEooCJ{<6wpacfoloM!6Q$sSw_0tK0S6j-m?z0&&G84p!60;(N0{};GuZ= z!#I)VkGYn74CyM2sL7ST8}w0ny=bl6v3&g1Zsh0|4RWQ@fc{Jx4}FV1k+q}yLIdwycv9Bt@Aqurig|UmoncsU+|G18AQ?o#m`ak4!;&&2tW&mN?4Kmi!D)fJ+Okxtl+lP}lE(e5S4^`+GcwJslvf z@ypJ5wv+bXqXv^FHbLRwO|W~#K&GLvn7QylXL?6i;H4WwSk~Y1^1b~1QA#Dv(w5?( zq}OUb`cMDc#JoqHU@w^8 z+9x|GeFoI_-j+U7dLy|W>4YiH2tw?|4E z^Ktcsix_dh8b>wUB!(Z{>9aLoVVvRs=KndI>_4GPXQnj4^l>R{^oSy|$ZZ&?)us~b zEwdnaY7UMluENc3UH{qYbkS(Y*S#a~mFPeNSvL6*xs#Vcoa7tO@v}Wfe}4j>Yj1*L z!9>)WYXx_o=;JqeGgLh5!qx3o;b=vz^n6H!_>C5@^Xs*TeNE2wk}zX;9L{6yYol$yAjmCu-!(gn88;t<5fb%MU8U1-ns?()v(c0do`UFfV6%HA5x z1m~aQr1wwo@s>U%z+X#CRLa)i{q4s^`hwhVhUn5Y!Q$C^$>HfI*z=Rh@YtyU!-J3E zG&dF6E2SA$TZr>K$>3^Nbn1Q`?VhzV+cZAz%0)da=`;bP20vJ1yE>@)nT)AN^VClCzPU@<220wF%Ms@$P#G| z9?i$~dF@5&I>(E;z6NEcJl2?9T=j>|J*0)3>y05xYbq@FPKB_5b7cWh<6+s8IC81# zFuUa6i}iYY59Qx`z`>t`Y4o`cQ07=H-RN_fbhViv*g!VvxTE;=jvX)X6J?a7_5_D+ z<*bGk;~G5j!l}JYl`icLuq?@eh$1CC(%gC%1 zDzMfqk*o>o2XRBpNtb?wpe2h(y_7f-b@evXcIhRJcKyk|+a|DxHwmD1U@DmYu7dtM z?vgoPOQ5u@n04>)TF_Zi@Gzg~Sl)-{F*P6=m`pxfb|fQyF%f5kM7;ZbEjC8fGB=L| z>Dg;>Z1jTP(&rt`MSZ#qjE5!dw!`PguY}){6SIzUMVEVX-=8~k(^M?5)v5&Jwj_e% z*xnej>>}%sX@JM}q_H1{SL|#%sLnh zP0wLKXYqb-5&x=&OxTydfO>s<3QOv*$QPby+s`(|cRDUr3&!Cee6SmTyTZNrKH%|c zGOYVjBb9XbmFlD`)3@&k^7KVI|NEJ6!j7w~J1Dlf6}o-d#<;~~&{!v$x{+ zhjL`=aq0cj-O+vGdXTU7XO8|#T*QQXu^-y{6FqbGFk39kxN@+vB)u)?$1f_5`1Op9EpEUc>N&&v2(ghkJkA z2Hw>aiETo;DUWT_{{zQ=hhpR0Hh+}SVnct2co5o6?gxu z5h54)Gzoe6KKzP^w{r>2pjT65saAt~Mq^5;3`g!6Wyy{XYZ+RL_P6u4Y~v<5Ttc`{~7xCLQIotC3tbmEk(7l&*>XY{LqUTAd&Z&j6 z#9J_arWU(V7b90Y*AI?lyeB$wKg)(K96~ld9)!y`>!DDu|J5UGpQI<6{TwLv>&iZ9 z5R{T4$|}evB`Gy8zN-V;)OgAjWBub#7ZDl?gnWmqZa5{2Y6(jV5an zqEU5GJMIQ|S?KFzC-x*j*USTSZ-JGF-A->^KAv?M`$NaUJFOxNKbenTvx^}h3P8P! z0+jw*z|E`|W2DN?O74-18a^!3XD6u{`dq$nY7E@2z6_HZ6{zcnRTv?;KqTI}VBxGu zCr&#D1AKkxlG)ueY) z&#?no`|M~~ypqDB<-4InViXhwUWQMNQ(6156L9v5Rgh3oEZuvN_c=Unl@6LN#&nb6 zvM7?FZp1BpH<}yVx~c4l?G7mGIZd#ODxaCA9+ z8&e@~JhPIFR|;b9W@y3Bqj3PMO=)4#BXGSPfj<)4_B|QJMR86e#h8FH2frIc^85wd z%MU8cispTW#m;CKaEER9DkYlsy725!qx{!-3c>d!0)N?jl|88bkzYxVuVd?!&#{)f zfctLc!pvEE#N%B$Ykeb!V3KGj&9kw7) z+=njvrV3E{<(~9SVrP8udopw^KL(v1KZnm1lSF%Qx!RO_noDQIzJYHaZkLTycy2#2 zUy2=!N8q%`OIWc?kxsv{nf9#iKz_78F6!|%n7`LHM`;a|(QFr`N+xO`R@_{j!*Uq`lX zQKj=5R3Ng?XZu|XG)3B-pJuR2WBK!gA#2IXrv}{WjlYqNsTThm_U;6~@4%LmT6cty zx9e!%bs4Cb*&T-VSD@M}Y-r~?8=;(Rn#N)jo&s+nWe1y!`fqX4h0vXZsbA@d!=~_O zkx~sH-c;<`w=~Z4SipZ?++%Et9*>W z>x#?C>Bn|rc|pE`>%Yr`6AH<~NA)=LybpW7;f;OPs&%AzivjI6#|et^pOBr0DoE-3 zjDMJh&WUCf&MC~h97s#N=pUBk1@Fh}d-jNZbuC_HuYtos-T0XL2cp4mhK$Tn*+_Pm zKN4wNULWyKzsC(j(DGHVIOgki!kk!cXh)2_$6|-n_Goi%x0qixtS=~TQ^cG#y|8z$ zNldL%JjUEU2L=hrM9(Dy*4OHBO@G?fjRc*vy!vBD(|Pz}WDq&o@=D+(iHp(3v8w&? zNw^Q2w{jJyZT*RD&&VZey4pBtkuj8Zde8nIy)4%lHi>bcG&z<&0JAo)CReRbW5TIQ zvi@YI{F&Z-w(d_aQHSR~11AdA3f zoBFZ--CVifr?+#8lO4Gg!w=z$pB#pA+wtT0|2zI%7p}xTueIeW8#6dR6L)UU@GvgH zWRytn!ta+X=|M9_W*XQjsQWORr0e$Czaw!%>3$5l)SX*m7ACfZ(5wg2zxUKQfE;NdPS)EFSazeWol0-;Kdurt) z=PF}$Tm-hL`r|{x`_e`=YiR8=1I?b>mRVi;L)IFLb5z+y$Mx)q%p9#UEKu_ditWSY zN)jaZ*+^c6N^r8mGjgK68>+SPcP@W9O~MD?fy7)hj9oPd>dX2fT-_(u5%?ecz8r1_ zN3m4n7uaWy8m9E?jf+p5!AgZkA|`L7!|>|cF!p=ie7JFc4hVJJ)ObH2{fhf1oRzOI zp9}#;&)J<|Q!IlX_$kX%tkgSv_p-oOzpy9Ml z#6+OE`0h37+Y!|`M+0!s5OJ&`sc5NY&kO8v z=Ytc(d5JnC4!uuMXJ}dEX;rZ78_JF(g|XEqVz85ODfzs4C@j8PD%~=)h%`yfATMo= zKv$-cB>MSea|hjK5sMY!*%c>vUFUDVdT29Kn{!xNdVRZ~uLO-Q$}c!L{EX=gr=SNb)+F#dp8l3r>}yu_PbE^P&K}6F)kBmrMK!r-P;SKEG=Au%b&4-5e3O8FbF^*BV0iv$I3EO_ULb0a}7$v)-5ShOJzP3|7T4bn$=J^iRW^KIL?t%b8j zmW^>PW)_zHaHon5l>D$2eD|@2l~P>kDZ_4aMhG-S>?AuANlo7@{@(A$Bvtw8|&Mks|B2BAp ztL0jfA^79mS)3m=kVRh(7U|3lZksPSFH`5-o;q_&RYbpvOg2VU?B^7!)o}se9)iYc zv;B8HQ5jC-yIyO&~)p713Gfa#|Y-|+IO-q()W{7)b4qLV-dO4E> z`~_VlD+l=U`@(iZndfA1bbKbizJHRSpUi$mA(`~Gk?r69ocZ7B{_k5&zBr=pc01BM z5ApHMDss_2U%t}n{QraB!lPPb)Qc)?*Q6-qON11&@gK(GZ-Gbe7!L^5&xYtE37}jR z)Ms>p5ym5l<=C;T;dv%obJ3plh@QirF)<%(pTCv#@_)_r{EhH||6vg;f$mwm6R1HS zqhxhooQPHIdU1Doj+QIG#GsAZ0bPmFNMBO^&IMonOknX!55#s?TNB3C5LIkFe*~>7 zmSH2)E<5U7lh&JevG6k{bEc7auv&8yMXyGEpq>-Fx zWF#x95I?2%lD3Od<#a4WTC&}P2FtUTJ!Hq#va$d926hocrFCf)Qj^Z+RDYcob&bC* z^_d)k$-N>)8;;M@6#ci)YS-|0Y~+h_`TBJZM)ZF=jA9aP}h}H?HbLhE@WV*CH=6&m!+(AYjZ=fXQ7X;tU&qhu`F6A@P-Pc}uvLZl)5|G!gb#UZryLuu(>e!s-J zPui^Cu%~jN{Lz*$h?t+r`;7*Y(BIqO`m?Rl-F&>fgcW0P(yRSALu~+D8#%YE(DS8T z7geRQ@YCK&5({<-(_RCT%AZ9aXpBS zWl=93e02%mM#2UAN!+?bu#@+{!mxLawB^t-^4WVoIE?8n$SoVOievJv&fqbuiu7A3 z_Wz^=#EWn ztVuMy{cMN#D|cdu^Kg8d+$x<<^+`nC38~L=J-p*tNroREg?|I*Ca|%=1or@xKwCg|PTt5QObL2X97L!Uu^iw5}P1 z@9koxUzm~fWvVjN8ZYASVy|M`BBqP&_LsFj`*B7Q@Nxv}IJle~zU9H36|zYCneU-% zp&RT!qK^Uw0!BktD{^7()v$7OS5~SUgaHGFqm8z%-KkxjA?MN*_+6C%YdnsMxC`a? z8>B)1CVoxBbeKRxlGfoQ<|@5_(vp?G;vKn4G z$XdsBr+N;9Q8!*fR|oEryB8(NXCy?z)+{ql|4d)(F;5Ln#Yi~g{5P1@xJ{(5ynhLF zxxE-C&9){_UlR0;3I*m}g_EMb@+_=`TJ z!Cn(YyDXV|9?hyJ;Iq~(ps913X#7%;t0rw`nQJwhC1&s8vVo%R!i#EBwMj!nS9hGjpI|Du(wcsjVdUCV7 z?P0g4{guD!^Fh9wpA+=mmcxoAC!j&W1)mhXC3***liKo8P<(4IV$Mh$Rp~))-)bhe zGM0exJzWtiL8t#O$>uC~hKoZEh?sPr-(OxdV+dR?yGjP_c7h*2b>*w}PXE_VPT^6~ z;O#~7Ms>#?^;bpxPxm&LkKcYCzrLEo1w?j0{k4yUe38$XubMDzLKA;3IuAXpd%|`- zam`EO^FA4OIUt(Nd`CY0>B}i>NEP@??%#SU&pQ@G7H4VvllO1gJL!#X2jOG03F|t% zi3Hm4_vtC`;?Hc|!lO5DAa`T8P*&!_4HR|wAKtQ7B@=YYPLt2Nn<&cTa&B*_{OTGA zQdXur|DJ*HxrVg!EjO|5=v}vo=kP-Lpc!At?{gQ}(u3ldM$+&q2UhjTz-3O&Z0CR& zu)I153Q`x7_UFYtVUkH&)-WY{HRdcblt(;OC5!s*BN{s-#Jj35_aI{${@R!*S8Ym@ z9xK$N<4=wTzc-&@%%EfBw&QF#cu^aBGzEf#Q!+NGci>KcZ)Qo_>UciB1eazVL65Pe z@LSIo#~^MuXv=FBdF3cfp3@#AvwD-Tz5KZ}wuOa_EVeIm`NsBF+p|&0%GhmI z5}RLNC>>F^jIDIpEM4JkFW5u!|EIel$FZs45AI)n2De&y!P>WpuxANB>1|8}|6{|X zgS>3HCEIF=aY@@eyvx2*a&N~0qTF{EX`#)Di!gltZ8`1gt|Aw59KhMon@o0^ z4})%sV}Hq)E+5D!;~NmZqkG<6h z>0w!~xp5cC(qg8+syFHP>@}HF(+P^N+HiqWZj&!p#j&?U<9;T1dCBPh zpzE~LsT7(&X){!`s^aJS)3Ex*Zw!g57HufV`=S3}dbmi3DxHXh$Q`!$Y5dtiYYW9S zTCrWM;|i#3RT|KkT)5?zi8GZ4fJJbB`bG01wVfPJH6ka#?u~8dRlloSupoyiEM1xR zS9V$@+NjL3m7NA@S#eMdj9a!IEZvsE)l$L?`TGYgKP(1=g%enST@1?C9TLkgdRIs? zAAKZVk0W7XlP#QHu?x2?I*1t>ZlwIzZ)SM83+rXOh1D5$;ChUmh#CAGamne?v}EXO zI=eK322J}*w|)1af^Ks%+fhqd0X?$)GBw|rN~;#7!|LPw9xYZ!-==hS*xha`{WUj@ zKPP6xJ-S>)qRI!6J=X%*Og&|Ew2y>^{5hcc9=);M!%Y4h^=bLECy%7_X4}D$uE)r@ z;aiD*bOw#NIgZ|0vKC)XJ<1t+4f{8LDpJ`Z>NDW15o8ZpFUq6Ub1#`3I)xN>E@HAX zawxRhDcZsJQKo$Bq!02BHo@e`wbgVHgGuNlBZWAD=RRJs;&B(gQ>*UeJcQEvBCJX<%lzsEf$FOVm@MG-( zx<2j_Z76o5c{3%{JL~_>&$OC;g5Lf>`q(R1mWYTl!ik-3|g=o#&n$s z-901Vp}hmt_0yvdvRrB5axH4JwFZK3ckoLcjNgd|7LbiDm>*V1^-i2hB&F0T~ zkLcpS-vKFQIcFvGMBOy{Q8S!q-Pwg4mQgdoZnB2N z8h-qzfTw~6!k&fdX!YAlq#@Yk=+ICy?N~l^mpmZm`;LIxhraauy1J3-NlXvs9Skw3(6sZ^wL?-a*G&fE3H5Hq+`9A93YhuJC{$3)z zH#qw14dQ%TTt|?YJ(&eDr>$Z7%Hw2heGa^U-}I_38+8uhd!S-vk*Y*RUPye6>(|YOx7*Q{~Tbs9VwTZ})-93Jua>*(oxc zJSE|dbIE!$Ede`;MwkAih7ll-E5)+n3uN-bG&1|dW2gy6+9TpR?QOV*vE#?!(w!mF)1Rlo zn!t&qrQJdHwtaipGez`?$u4ZQ#YGx%L;*};&mCQKexr)xudI@Lq;8d;d?AHaSvTmt zb}(Jv6hmEqmZ7V8FwssO&a4+ik)P26*2 zn0)Y%QDozy1<-R>d%9}+2@>>r9V-c&fNzaP*p<-b)`5``0smvv21Q;+gNc z_I>%7|4*a4;pCz+Ox61$aj$e8%wykEAk$C0vOT{*d{eKfeo?u8%OwlG1t1cv=^hA~z%L>c#o*UHWM zU1KJHw2+R<$8URwbDQ&|xFT;K7zd}r;O`r$k*7Bexou4~U-zORKV!f!DjT3Ci}kiF zC!4(IljP`m;x~21tEAF8vo+-4cTZxd(F>}I9RJ}UU{UA$MV>sX;s41qF3Dm$KG@OU z;X}kW{ZeIz{MeY;xcAl|lC=5^?jM@X?@O6aW9|2otS>|8xg|dI#%%*QF{?yu=WdI3 z(!u%J)a%z`mNhvQbqv$d>+(+l3(2ed8S>;YR65W-8nox>P~&22I1tuH#QW_&L$WJk zH6*%D#cB1O|H-t8U$31Q;)(-h;<}RT)Vn#jy)KyfbWi2u)_8#NAvs3P=qk5Wvm++I zeqva49lJ8P2f6($SbpcQ1bS_6674m=P%^wdV=(F>stO^!BXa?YY#fQN148r z0~#pm@bOl@lAUjC;ND1Axaxi!O5c2!n|FXy(oX?Y-gWeKN->jjDhemw?Fi}0pCt_rl04$$JrI8+`azEg6y(t2Y&9ZaTskh zw?bC~O`MxzDC#s*wMbgBcL3hB3l#DCl9UV59awBL15$RIyRc z7AWaWU`>cVmY=)E_AY)dcbuKVQj;>-Xhj`-ar2*VQldZcCQ;eG5Bp{XbDeF~xGVGZ zu}`uR`*QL=yZAN}w#+&Jy30ASov+}>9Y!^N?7%P$viE(mz*`bt_=U9k&i(UAM(T z?eJ4LATv|sf6HwS9{9V2>P~D&cTBWl{$o<@D@=Fudy+5U_CXx(y{k)y54E5VWj_U9 zplm?ND0$&8C$cB+I>b4iqSlA>>HYc3;L_(2d>jd7nx?8volk7%<3*;SVVyg^gd(}s zm8Hb4QIB>~Eo0B#Jd(HMKEU1VDJCq7W5@KCQYDwa>`2zef7oR`48dP~Y>sKGmJjqC<2{Gw|CWz__(7=8e%! z1&nV0m<2Sd&~@j!V_{9bede5#r2mn9e4M=_==y0V>JA%-^Ul<>@cTw^Pj3gz4p5?L zz1sE=Ejsys*AVgSJu>IWF(%knEqWX(7wjUQmP#OCLj1pEPQWiDbr6R>@!h;@x-h4^e8CHhx|b|Q2CR)B89D#uB?tUtzv-Cv6RAEwF=9dyL| zr#1ep#1QX}za-RbOQL0dmMJX3ROZl;!C=O;O zE~jC4$9D95=dR)y?fjp%vk=*qF+>ED?N*3$4Bx)sRBCi zZI`bjIr*(KYtHLLLJq%RRnKM!bR|M63~ppfS5i=;e`ju4Q6Tq4HjjIvf_QKKPgXvz zoZUJ%7;o5!-vdeikK){3w#@ws`8CjAD#-Gq{0h47C3yMU0kZe2lc_Ptyx9kAjmK=c|B%L0uPqFy;CJK+6<-Y8q(1^qL0;bhkqAz!Bb zaX_=UV9eAH;at3bF>iLBa;zvI!@rVVBuA%WOB)W1;Z{ zlzi$)OXM#|P|$|5z zyQ4G)uXmY-O{VYJ%jq-7+QFAu*`s(8;^xkJp1%U6sh8=N4%!aySDvIr8&{$r>&w_J zIH>U~dmi|O{ppc`TU+DdjD_MqES0DB<0j9i;QeL`#8^KT`)G}BcktFieeV4Z!U;5l z)U*9?=^<-|JGK$|@9&z_*wYxkx0kRPC-{8=0}l{-Y8y62Pa(-y7KpO)btFoBbhFmy z*Whq3P1wHI7OeB~Nta*N_Qwq(NzBFzuqEmse-7oW(B5R%pEsA9_2czP=nX?3l`?H= z#s!}ngJ1ne;foXoAzQcNnzz@)eqNv*;IR{XuF-*EYJLJgQAXjJKRQD!^P_ztm`d7w za+NQii#+~9Fm4!`iZSJ-!n&G7NT=;`o3fV>%ZSHwt%EFOuhK zE7Cm*MPPk>FL*_?`Aw%nH%Rf^4)a?U(ZJ2c|LA6x50LioiNjMzno#Y}M^q?`g_Gqa zxT@zlwlXk`rKo(wDQBm!fn+}JjOdAi{EKUT2jOX# zhgj5akVt22rZ^sy1#fB=Z6Vv&j1;$$uAjbO(l8rQ7lCfyQA440p&mXO+>x8I$PSV# zVp+t1h0>Y0;iFg{+s^fa=Q7j?94607G^`C<7-*?>>bd(7x{?(#?x(NCt z#myJ%>gA*oi`*3Ec>6JY+AW8r-cHp1=Vh_%*+xCgbnnFJK2Bzu<~c}qeHQb-+}O)B zGgbr@5Y&k-@3&{`zuB(qhr zX9IjYqf3uUylJ&h)Khb)8`me}4r7b416|!2yZ(QD3ZboG#cWs9pdyP!MLZANs;VRoYY%h*e)#I+IZNakf@3D5KKlj_> zJooLcKUcuKVe8u>ut;ded$;3pcCQVn7d02=cRqsVv*u#%SS`};@I-Ez?QX992qo_2 z$x+mPkw1n^kzusm5S+C*548t$@5jYX_ABJALMc7Z%lGK(9fSB(ZPFL230KWrLt91pfH$1{Gt zrRUvDNMLaUnk(sH<3$ZTkpBj*ne)M-Pdc-P!Wy=TzxTELP$j-9+z#63$3mx|$pU|w zyB(&^-pKvQR_Dfj87cpoRf%77+SX?k4%egN?l-8DH=m4|y$3_nbwzoW z&+W*$ELP@Hj?Lh{SsID-1({Y^ye9KYba1KF1pA+-Qb_`6u$dmJ==H1*CF3uH@o`Pu zxb!zm85aUck1w$S_om^ZGu5mtCkz1y7<>54YzN2D1W^C1sT8Lujo%& zTw8|*8_wg@KyQ?48**J2i|0fn8OGmn+-Vi=hC>*a&BrZ1-q0#H2)isb*}NW2vv%Q{ zE9Lw?6SaTuwa{L?nqINa6@9s=``Vl#`zW6Ze_6zBD^|x@{nO>w(mE+T?#b-G>2nuH z8*^4$lEwE&s4wXHwqJVLpJB@Qcat}`HQXS}{A076B^BdEz^~w7#f){|8R@X@^cHRKULTdrVR|#1-}P=gOC~&DYw$ixj`f zg3LOhJ#l2$k=(a4j`G+0+`)duSNT!(arm)M19!A1D4#3?l(MAt^X&@%P)0&Wi~@|IbE@`Rz#(4_{O!Uu{n^q6VGh1MNK>&G{+X;eD4F5ZPEe;v4;Bip{G->xRZ zrublFCKFl8+92^exV5Fb*k(Hh7;&nmPw=~w2j^ZB&EfYl{$83YSi9gN2=Yi`!ocpp z2b353FaxjcpfxUt`;+qj3z?< z<%nbWv}%fI*A0umusu_rxA9gh!t%8iI4W`)*E0VXd;e3F+tO_mdY)X4_VZt``ri(? zI-xx`)oPx-Ho1s3??!Q3j`rnlP!#Jw;A5reO*@Bk_lMw%@E)8+$64HQ_n&yUK8!`! ztt6+X-h^QX8i~K}7yK~Ynhma2W*#s|tebtw3x|KNBrbJS`bjB8%-c{H!FhNYa4zor znlgESQQ0MB`#Ro0vbi%?mKwl?uHB4bi$aC6l0%1P$rs$VM)}1AoEu*))^|KBm764X z;08t+a|uiKGj7gB@bOItzPw;tiDK&^+>#>4JL;C;*=2d8|)#fj`G{vv1-RV7Sz#~Let z@27-pT(w2C&;Q6O>-&BYo3yRJ=r0iR^ru&#-KX86?u)*xWX{Pk5M-K$SB|Yhmx(5r zwZ({i9%{_3;?K>sMrM)v=<66VUx719T8|4NGnlImgQf4=3GG!DcQF^M=1t{xjMU}2 zOclqfvh}!^je0N-2i!H|?tF;B_K|Jhk$d|x*b>u|n5?6O&wozEG$UJfU+hsK-#%I$56u=)NkeA}UI&sq_S;Ieuh?NyeC-y}WySWvl)mK8f(}sdhk``$R+0U;Hc6>9D=D)ip>gICaR(@1x z9XXuik2k%yV}^b-UhWZxEB2^j`$KbyrPKz@G^T-;Y4g8+EvI%cYq3jU_l>Ki0^MOv zhL{nXi2Ht6ak@j5IN!M@s1xiVPhL|@B0Fbe`SiWCE-RWws5C-dc{#W@58#xazU23^ zX2bEf8*#K!9(?lsOE&)9CAK#-;Kz3TGaznq4=NdHj{9z$Kus@kAG@qds~Rd59EjcQ z0%F>)fPLthjnf8cP(hA^>&t}z_wHQWf5IJi0w`He7uO; zhL?UYE-(z*oyxZ>F1jybA#3Q>{sQw0U-&i1SF_!*GAf7HA(IM~^)1#03ul?lRvm?KpIe7?+>J zoAp_E-){stpwQ+I)P8vz{Ic|DpQw6h&6o-UhZc(Oi%UT%ZJMl1H71zTqu(k~t)d7^ z%Z*UJCWx*LH^(&#zu}tbMKJGSFeL6uCs#MLosYl%2g#+Ntzdep9ppQx0t<;o*VUUa zDQ~}hGwp+Bv8zP-3XObRyz(#_7jTju4L?rLB|jl~F;VP-juLI@{TTe$JOihu8)BQP zjQuRskqOD{`bqiA*tyuyVK}}rd4P#a`E^d~v+z!13*9h8#8<>dz$f_9b@q9{A!7fe zn)P`24kypK!d48KM(z3zrdGdfX+rcPP}Iv}??#uG2(;@o`7`8ByXDw;7uOmZ5~HUx zM7~%2i*ZI~dl>p~K4eDikbWKGE7(!;Xh#{Q=bdMxk1#<_iPh>5v3$TcGts6=e^&xA zi;yq+Ka{-*G?q{FIBpAND@$1tTI?#`=b1b6h{}={t=d!yl@gMQibP1r8bT-{Yf0Ak zQOJ^|k`z)%yB1Pfr1kfD`+h&?{Qu|t|L6RiGY2#C%$p8|LE#AANnm*W!Hbd4WuqCp`Wu_%!ZpPIyW zMs=63uMKCPYz}4XOD9)kq&1aC883$YZDOqJ2u<2Ih2!!02JGBJpSiu6kIBaVH8}T| zPr2UFB&;oTq`b!DBaZs3!*e;i~Dfct}vc+(d9Kel+(9oCtr2uPMFfSc{L{(aAlwH>RJ$EWUCan*TtJ$t9; zv0I`SvtRd?aFKH-(^?tLIO)bR)ayAF?+SIvk6ZtAn`;JDNVvEV+o15DeV6}&!QI@V zoqMrgMh~}c-Do3BTj#ks+e66&e~1UWXZNn(!NZPN!2(&!WIB)EiKHak3{JH%}1 z^kh~ZoWk79HQ>Yerr=NN#8kQW)U$YT2~?SLX{-KQi^-0J^^f?shb)%6z~-d<#iM;S z!h21kdl{>cq|5k<7MI!*+YNQZNC6AMx7-#^4vFMG=+ui^Ot8 z9$}jglg5QMOfWbT#Q;sNP~9i)k)vwth00r4xbxE(Z2!$Nvl!)f`PW&F7u2p~-2 zo%cjoR+kZw{pY7S*3S-hvaX$3-156|+`OY!++5e0eEf})21&=ZMrPWfUJ|p3@bT>G z6RFTS_Kyx@&n}vzz^BQ|)J>r9Lz$V47>204C&5d9lfoS#+>4Xs_%vKspipss=Qj2> z&!;upf+1I#05VbRJR@iG_`Q0z7_|;@@lBq1l-jAx} zj={QaUQZGn92unilGM41GUVB8BA)C|>=N|BbK(J9P@=%c`>e+~KEAwX1m4_X*s_X~ z+?3&evUaR4#p?zh3K$36Kiznq@mA71qKStd>W}f7K-iHV+?<60 z*La*|Eb2;_zYf0`TRRbBV6X?~|kL$+z@!>Y^h{4{SmCz)c z!|H8SWxKNYV~@4@KMdT(QVPXoPMiAKbB&1g{9DC!NsGlxnyrXp>?$@eMi@_+X2&Yk zoMG*Mw=)|x&ok$)NiuhCNK|ZjEI<~TT9>DsIY!=vzanx$2bunEDOPs90Flg#VrMn; zd>rC-GA+MmGqRisOz5S#%v%x9@|y~y+}8v0Z06nLJYU;b_AM{})7V@sJ|E}SZn;WS zr#KKXPeD@U{f|$UPqid_EP`-@ua{qU4TkNgEN^I%j{`NB_or(i| z8nzzUiz7Y$-Cg?D@%*OK!ujD$2>!H>b^Kk-I!S4fF0nJD(N2Q=E$J*@()^aspN=l) znM2hdNs{&yvb&HcN4v(84LL2yXkrU%CApi(rEo~7`Z!*$*RNpx@hXvVI7g}j#xq~E zCo#>k+I+ZU>+(u9kBoP=hjFXUm-jdQ-^*J)rX)yKooH_1^MrrBOa;C!Btv%Wiy^Ol z_HZ{esBq11=&>o^e3?1R#mL>j&t&9XJmV=HLt-yZAfcMgL^$~YSzf44?(lMQERg=p zhubYD#RTqI%;>hPU`#TG$n#>q3YYm=d^#V*Z_6Z_3P?xgzx@e$=#gboi6rNGJ^SF} z9}>M(ooo=k&-K5oSJvlvjq%)KZ2Zw{fH`Y#!YswuZPC3Ufz~*ATM51?F|^B9ZSx*eL-wSm%ZsRxrIEC#_E;MYtS~ zDlaKd-ZG6pp3&@IWQ%Mqe?7+V9^;iu0@AEEmCPQlcHz!!=o@Pj5Z3#_O|73=KDgS3 zz1^~a-LH3wD@{qVF3zoZ$?vT=XQ2>ojP+(kd%MY@(IS$$&ycKYGbfsr?@3StFmlB! zNP+J|qO$xa=7mKfr)Ij6ogWVpiJFgOg9P8Ecnr3G4do6K>L=p{k%!o4>*w+^W$lWZ zM6n8iC%akIcjvjrtGn>rlJ)Eg0|CNRP&g>340p?O;qS#Re6YZdkH@X5g?QpJQBrCV9fLQgh-Z)gT!lV0!p@D%QHRZ!%iRO_*s2!qC_r}L2YN_ z{JwL$baMz-A%Nh`hj6cJCi>#5!3}z_8=qD8;Q7k)^h)L9nahGL+}+)S_-B3!e=V*U z&t(7ZsptM!e+R2(@fTY{K9ubP-cBHtJ(o{gFKs~G-Sq1osF?s7x0!|w`r@}T(`9R`cnSvP!xD}0=n%lPC}*693ha-SzV!kfmEn<4t-c3;WZ`N2Q@nlxU$ zJBQo<<_$bP>_db{PLM8f7OI@YlQn4HwRhfLvjsMtZY%{crYHT#azdY{wukA z=Ju0^oCRbCD+<@A%p;e;m%BfA5P!4dFn^Usu*F-Ns2{yXLOHYeV;Zh^hBs<`IlZ*_TR9{Z?gdE9HoxB(u4CmwSsdOq^s2mEw-RsShLYu@r^tAw8mIq#f+Ljc*zCA+^5AqKxfuuL zHV@>;QLPnZ7YbwYJhw2umliU)od=n)KjO^3+Qo#4=H-M=&m|qpN1*kXE1!pWF#MlA z-9eyF2|N;Cve$Xvy`807$*t`-anRLR%nSpmp0}NhV=Tyy{7Yl|roU6BHreIy0UvlX zhds8k?pvj5?yhZ>FjAb$rZT6?2F#kUY<~zhJ<^;k*%{1zAvTVL zoY`KU8sExF7tLVD#_PJgh@CuBk>ngWrWEOW-+CN=9 zwRZx0vhWU0+lD(duqz=02OnbWgfL#4ibs;pcyuTSSE8`u}@=tZp0*=iZXO!LMsv8_0;( z595f4e(ZM1V+^jq^vS9CwBj%Bo|{={xnLsqO5jdb>Ek~;8R$vF$1eAlFYKMpe}3b; zO@#CD0xOcrV8ek}QahE8vwxV5Ah&a4J<)1E$?dGtB07&Ilct_{wr9^3_~;c$lK(Em zt{2;Jw!iq;I09qy+NER)zb{4*pBNufSx|@zlX*V>)9T2XZ{=iAOq{uKAerAc>-RFk zT`Em{<>T4I!*oT=fJKG%sz3a>kBu`{6nMIv*~hmsYV+*5KP*l0!DpH9S>O>73_rp4 z)y$~Sw@6_}zew=U^>e>Wt&qCYN6xOSCQ+VhjMd{)d^p^JTDDsmRg@NU$P$M*vS5h| zfkK^%sjcVOj$NuG!%~g>d0|RS@=yJ@UWvSzd{3n$*^y%9a+^X)ytnid3HR(Hwo>}c zo}*t$wdfxH*xrq!`0<;Y`14NzqImc?QC=oX?xeYrC2ML)*hL*u+Pthh`EDC&5Cp!w z4K8wK9<8%3*Wkr$^-vci?xQ=(-@HmA`-kL7bwv%k)Wp7G`2o)g$F-yQc0d46pU+fu zW(rlLslCFP>e0;e8%0FbiWk@J&@0eP+QC*$oq(4Gn~|ESB3L?i5?Z%hfDzAV#WS7M z81)T9M2Qz4c5Gdi?aO5h@?)9M(+`=%{o{CPY-8P|1u=|dj+7wE@a?)6Ru-K*o|L8 z@k!Zhyf|edte#XZ5mB<>ujR3&M=EO4t;g2Lf5XAX|K!k_?;o-EHpen!5*wOKUR@rAcajopQkrXppP)CpL?xaKB<&+s1!+X$V-;&$tWB0JkzzX< zK0)#5XUv%_%si}8;jd3##069`x|W%ISC*+g@d9UCl!C5?F)w~_4BH}#iQF$I5;)1y|h*~ z((wYDjsM~r5e>HExlYBJW2NN}EjN+U=lj@srig4+JWgU9-r+@Sqp@nFD0$N*!W=he z`7p-Ta;&7S_hr3QTF7<_5oXHUYox==pRA4C#qXmW^$8Tyx3K@y*FSdIGBPP<1|zij z59yA7MpA|)N$7Q+U&gCHkmUK9Ud3-8J;aS0FK`ErFDBoE zjVhEr?5Gf}FeJws7$SA#3hTS=6MuiqdMUxkJ(uIwC&@7Hq;3=ZKQXdJ&-3^#dj})h zV!LT;m#y@g-pdplgT7dse!#WPnd{n>k?sqQO0|97D-H% z!)P>;2wVS)zdveaPe!Bv$>*OJ;Ff5-Bu%QtqO;nve6*X}-KT&@#-E z-gN%||AVW**zacsaU`UW&yW65t#$l%Ln+>O&G0O8{Ld?LYk3P15zobEwDvNhzmN0j z@Sih_H)KCwT{*_j{+I5~VKnz1AqnGweD64bUDW=?+x2cu z;b%#k_D^u1B!ll>IL(|Mbz~fS1et`6vt-_vqx^A#Hdb*PrQ2X#e|s5O z=tSD1Clc|U6|9cKebQ~tFi$&+$l0D7xX0!qxALbxv6sxn1&*S40XK#_P%#cV&i>2G zU^;&J*c=6Fof7$c`L|n`>+j#p9jjX#HWi#O&0g7mi~D`bZ{qW+l;7vT4cG7AqsAtkro(au#NV{{Xkb5f7HZ2KN=+E;h(g^;)9Tu)P66rGPPe7}+LiK%2w zViHjv*g`zM@Zvi1azZABjw8#v0?1*b>Eu(-zgS{p<0wo>#z%DOd3zpzKj5$Hw3Msl zuw@%e4R9xYYdXo5h-}ia#F{xoEym`d>)F#m_tEPJiHfMdYteL&$3GScaQl=G;K{{0 z6&iv!h{j(z=Hi+kWZBBc<;&(RAYUwp$nx!Sq=e_QG6r+3Tt-d{jIHm=V z@@f7=J8T}wDV1V!#!n!T*|J1X;0<=k{zv;2Rhkuk?Oax8))2y4at}Z#6)6$LgrfM@s4Q#~mmmi*Ra9?PE@c z@*|GO4k_wKA8n+4J`}xMH-qlfjixsYeWJH@m{Wi1ep0(XOri2#7}Fo;wbQ*%2Wdg| zJ@i$<@wDvSOpcNlmzFL$Mn@D5bADYuOvUFL(%HVpI7jB}rzbWT(5=Ql>614i=(Dnt5f{9}&Z zhAGt3A}Lzzr3m$F-eSW`r=}sB<8>Sd<4&U~4_&E)eFjE9&!uv>-%E_FT3yjoCtI|0 zy8s#N>7Zd3wAXAavUNL3Kec|z z;fOLE{Z9fM|EMd}WHybXH~Snl;vGQ!nR1DG;4VUch_|4Z?-HW36P4)s;fvAh06!$@ zu7)OG4Mb;F7@|YZUQ#yYw`jGdL-h42{xp5I*vP+0o+=1#Hrg#3M*nboPeod7KyM<_ z==L>lsRseqsKSI3)ZGgbRO}Qk=S?S}oE=m-sh3|;5z>j&9UXi6zK;Pl`PEY7ckBuM zI$%7SHlo1Mn@n*~B5U}4(l94)W)MffTpL0@`ElY;$k3jKsT`e=WyoB39@6|SgaXQn zIY;I2@j848%rS5K=#aSKrj#9pVmosob1jhMsIEAMPr;!_PG_7_C@=~`# z5q7ib4Hk#dc-Qmvk|q_jkW)ZEajN6w7hi_aIAOH>S_kK+O2v3hi-b^v9O`IGYp&Z7Me2K1HMYgA{8CM7*> z2Ninm9Gw~FLP=@>s%r8;nR}|~O7q=Dd!|THUldA>S~8j`7}lW=JS(N?3p+VSTWUDp zmZ(v3VTV!dx)7>nn=1CXDQmRHI(hD!>tBtG)kEmJb|Lh`a$RcSyH0vS-Bns{yEy&B z(uCIf=!4Q#oaoF%Svsu8gtltIw6yOk6tQj(t@FnVEnQ}bdK1?ppS*s0n!P5PWj3Fd zeshg7oo7t%_%4GIFZ$6>Hm=~@Iu=8j+PYF^TI%TO*{Nvv1WVMt#1dZAY=^KpfzbHA z9pWQ8z<#|Hw$I**mGAAw4xx~$#y1Gm z?c(Lk8Gx+HF#K_sE8ZCvich^g1+6a};lzpp81>u<@s)ev*!kLM)wQ_0UjvuNE8v6r+W6J< z5SZ<{5B!D?0{6HA{<2O0yK0WdCtr6!Ty8mB@NI;?mJ*%|PXWF-p%K3nh$c#RJjl9| z!{o5iLa=pP3|c!^!s*XF(5asdzW&uvRdx^k)^kVpQGZZ<_!|6s`&O)R!5_cgunoo) z8ba|t6Yz3W!ls4lxc1si>~0YUU%n(j{P#RqsNMoh<0yP`o`RiEZN{Cv+@ZS;IO6*~ ziO9~g7QKz=N2W1XaEsM7?CtOtZ|^OGy4j`RJJJk`Ubo}Keb@2UvWIwV9mh)iEKRh` zCyVUTEh2RnTgbCwMLhkdJ`N@uvF%3@;(u=(N$8O!Uzg=VwnjSq5w3vyBPZd#Z6uVw zy#u#x7vb&)lW={QKHkYpz{L;caB>I1gWI~{ZR`^`Y&Zwse;5ar2{$4Cz&-f;awQae zUJj{3F5oilCAzTxJ@UMD9#;PhgqcAmV3kLMe32La{KN&%vp$3Oe$9v6_&1=pZzDeN z>Iv{-S>mvkFueZRMOb>j82m(5;iAP8;hQNJRbLVS!K+7(T;|2LA8n#}KCO>o)51@1AW96|N=jkLy^}Dz5b$CB2FxtY#;fOL;@M~V zK#dnCDB^?=zV-e!?rv(qsUpwt4mbjKhZ11&mt@dfbsW$4FUIRtl*rAU`skjy8QsZ? zaigw=;k&ssoDi4+yLQQACqH5QMP?j6Dl!Ud<3(|ey&ew!{1$99N+EecFBsGg;P~Y& z*f2zhkp4H2XH*OWH(o-%?+A)bdW`Nl%febMT^w9G5l>%XfS>HV3*ArZVVvw2*fZlM zZL@U((rxuXZU&)HtaTmCjy?hV4>Mt*gBiSCkO32Ty4e2tH?ZA{Zakwy4zK;7gC%F4 z!xHa^46Ewv2hS?21(*|BQtT~&pI4w+RyVlo$5ht;+ngQHPqFeUpk zxEOqa`zGRe**78M%$NjAZazcJd#`|#%yU>CKM9|js|2+ahTVQA;3!1EpC^}q zQ#J#yOUgprno!iQD-8~fZ;{|g6>9fUfG8&+>~`omIBlPR6?cDw_&>j3cGgd)!1^Ts9Bye>)DlO6lPl zUsvOJUOd`av1vH}a}#z&!d@Vs?0&n8s?%#wrfocu(5f5$aE%iaZkd*(nA@4oG| zW*_t`i{keOf5Pk78E~v15420B;*gi-NUQAv+IIOFDw8Y5nSW~V_-;`m!%@c9ZF%q2 zCp8>WGKkAnCz3%KLLBoVI5sC5IZda8=%vx6Wa_7KvTz&E=5YOfEVelo&#QilEAQUL z?_z)Bn{$+i%==k*GdS$%xDEQWUx46yJ3KNm9O@?JgPOokK-)i{jx*m- zpqda!jNE|{ad9kuVFnI*7zEz?&Vt#7AMhr}gk(GKB$A)bkQ)o6aMid;c=SClE`rP| z92ek)Kg@T-s{)!~&BQLK_fW?uCmw9B)PmzPDO~QWf|Jol(N{Q`bXj{O9VqaZj)*pc^CI=MWo!VaH8GtN zJMKRH#CsN!@SYAjd1c7?$T6tTJ%x*ku24{!PA8|22em6rw2OTfstt;PiF1$B?#Wd~ z=^c+W+;z(++xH{%xw}bdqgx>L^}=;{(OZ#MY=<%_Gj=e6{M1P8B)943~8Iw ztLdFGqqLrW4|4Z&rm1-+>FKMpP(Pzge-$`K-`hL~{cPSvg%2$?I#YAjurNoMaw*iK z+7kLG-RK6|&3^-ZWa$Iy^k+kA-6mD4^S%J`a~YvVf*NSDVgh6-g;0|wgiw#DJG9vC zX8OpTH#ARihXxF$)53vUP;Kx=aNqx$677{kTYkxclTa7^#myB}-LeJqjkT2bhMz_! z`xjAP9}IAA2d|*Kel4XvM&;`?f0U+aaa!=>Vst8$<)nvZP+rq*X=QgV zS~fKu{T=D&^XUKX{X0ar(&Ck&NK926zS>EnA4a0^@rEMKy_QD5`8q@|s@{iMzna1`9*@{XV!?x{D5Q-$sp%bwBPtJx`xSRCyL04l6@~*2<9jLj-TA zp2u&WOTN>8kq9;SdOUr_?J~;nTSbYwq*C+sFVps?meVUHe4*{8nt?`D7~Qj02zAff zg0?5$raiiv(8y3We8bD=e2clr^aKNn79Xk3)SJ}m+Zy!rjefLYFCpb%H zETT3zSX0e&9Qkl1Zrws}OkMeZXC1pj4M~d7&W$SQsjVdl509tp)&Ce}WTkS%ukE3; z?u#HP=|GU28iw{yvVkwM>+pNC2~fq(gglZ9Z#_h5nUjJjHh3HaYyG4r?J7i_HccQd zUq_X^F+#5^Fr1$Co!-|#(<+G{sKX9P==WNCm=DjyeW;?f6g_&-0Bw_NQCryo$nM*ST{;rz z8OyTJ`EmiC9hx($t(*%-dGS{l53Hg3)(uhzS3IKwe(G>!GOtL*PP<`bEp0*XACN@% z8`XL8@*LIBB1xP1d7uWzGp5=5&yHaW=qS2u)py$3+q$J0!q%5~bp}(bt=$WT5 z*vq$5Hw)6~@UmX??m#g8UO@!Hs8KLsjHsFB^&EGf5b8toDo7S!aPw7$NC-bR{9J59 zy^cCcA6Cwwa9Sb#C+Rr)kkN{6E&hb=Y%+y85)09xd};dBc|&>uF{I~?N+H2iV<>Q4 zhoHa^x;(ez4KtPLwwe?4Q~h!T<#u$>s# z(!ptmta)~{OL50!FD%rXfTa>65J$Ql?U4C`&hr9^R>_1y;nh%B5p^9}%-TU!Vh9RN zFsOLjfY!8`U~p+BzMVp2?}F9%iqI~kBIJ%7LcGyQycgO{GN9kK59ZCBgpZgQVEtD% zxY4hGbJu+i73Y;mGxmyfd8#-X$(KUwsuSVZsT@c@SOHof_t!%+m6wiOVPaS~pt$3U9J zRS;j1z&rmsQ;lmzsOoc3lz!-2`lM$!Eq2}jDF%dsYR?|9eYz0dJXD9OWwEf-rw?9E z%mlyi0-pU?8tm_9K*nYhe7AIl_tC#-*QmF2`=W{Ho4*Y)H{MUqFo%f!-VRW{`3C+3 zi(}D?v0xgcgcW4-v53zxRBs@Ta>SRTFxkm?r~GUjzSa~EdA)=i_MhO9ULPF(eGVo~ zzX*{EqS)Mg4aOeF@zj&2@!jkSEP8DSAF-WH{3RbyTAy^OgrNCUWiLV3zZM`V+lR2e%)=RIt^)(;C03*+mHr-9ebZLnG(3OI^3$a-cl`rCF8rP=x7E)tD3%02LN`===P z*GIJDm^PRwO5?3!AMrCJ1v~1xFl{ z;o#FJaB7Vu-k4&FtK;%uokkt}YKVp}XQGhl>r<%N^&;BtB@br}_JGE^QwYYTpiA?2 zp-01NDEZxKcy=Nj{4d6V_og1?7w80AUN3`Zmv2FJz6kEq^#yKp9B^&!f$!fE^z`dg zdi?cB>UEb1eb+V+fvW?mh&MwrmOGGS=>zm)&r394)c_8Cnhiy3Pb16FpLAZzcy!%g zgkG&>NednJr_VWEq}2p0K~FXubgtxrtdu0mS-ptAU+HR0Cz@d^NyH*YlBX^OCgao~ zxM2`I6t9Q5gY_V)Is$9F!;pJl6dL|J1KzEWfz6pxke?_6L%pjpHzNi|f4+fdKIee% zuMre1Edz6nBXC~eb3Cy!8plabp{eK;+Pam{8`uf-vV(He?r=j2m2II1$C=$rjl1$^+;FiQraw1~hCF zps@J?G^hIH{~S1Wt~3t+Ra)3E9ngqy9B;L_B~aLe{Gtd}xG z9BozPWF-PqWBfti?+k3=(aG?JAY>IQ!Vm97F#hcb*7+zvjLylBduBV3&XE%2rCeL$9qu^{)E11+kwr^DERg`0uqj$2l=zc(4EGCubI-BN+_DOEV4pseILO1B7CG?H(Ev@G z10Xl?0&IOlK=G{|1b;sOL1OL@w>ccbZA)Od;s#QkIgWRxFb9vB?+|-_7WCzm!Qp+Y z$%7YrNr+bv8RwS(pIocKs_X-}#Khsy{!IL4auNUiI@QsLZ0kXWxQ# zv?7M^I7rY4g*{#$K~HiVo?xPfx5^pglyi&WnN85*EexW*XpaCj)TOktZOzOAbrf>0^mIuc7gcGOo{?f~TYv zfw9N{n0l~Ru<+(lA(q|SJATzbLr8;(^3D7 zXJ|xnDju&S$BSX2fh*jO!ZqprP#p6bHmu$a6Sa?n-`>lhwq_D}v_h3s@a%d#oi3qs zpP!(K>ows+@m=&pyb_5CT}RYxMM$ZXfp-F;pfhJ8+!`+qB8^FK&v6bi?vh4c(c-9k z=1!3A-v)1c5&Q?~z9?vttzBwCW%=UxqX?2LU{EJ!! z22k841lyXXz^ft^nEBBGOpcjCs>uoT@E-46pk)EN6Pu8gt1+!(Q$d}PPNe-GE+7Mi zn3OCB(mz!UTpHq$ul*8qRv{8(W#++F-|=v=(-jiy_rQ{)HsCy62?Fajz!v|NFj*`L zWel~U?Y@Kbd}$|e%$pBaEB(N=Rsnmw`~~_Yt?=^s22m6x_!1r}qU~!c(G=;QKF?WlE2Ud=hphexJ?^CMsbLKy6X`&XJsk1z0?b`zu!T3M3RwAUoJK!^W#Sus42*4VPcXJ$8Grch&*?QPK*P?X1A{VC@=;ZZCAxh#1L?K;Bhg*c@cN(q zFxz7k=+*f!6qW-URCD0)_1n;vkO}QvXAoUufY$AQO`p+jqDQNg@f9&4{HP}s=5CS% z`HaV??9l@x`^ppTb+kj@bnnt_Q}SR)Vm562*bdik^K{-9qoKcb7A%T+g!TtSppJdv zsA|0?TJU5OYFxUFu6$ucEdIX5SLPOC_f?mnC7Y)YK6C&|{pE@HNqI82MUA|_-;Lr? z4(i%=1PQ-93ccU=gGT!daL|u|NEZecFK$KNJ-e{P7i+BQB!}~yO8IAp1*>CGbSi-t zzX#FI+?8ly2=AM&>=W80?SwX_o1*iwy>yqwRPgv}4%!OekV}0UihJ%4+SnXqo;bqt z)kk2S$`LrzVFd3LKO>1po6*~_7f`kCGb&E;M*9@YU|RuF>+#E>S=hMcE%>?YhQ!!nZoyr!po5`^ULC&Z6Q z`tjy7wEWN+x~0VrW{6}%-o$fYS6`2=-_J!_j^$|mqimS!e;4khJcP=FH|UD^BwA21 z3Vo>ljY#nXP;9aWsazT4W4a0j?EZ$rXSIXUsybMlatF?4|D^|PzSFBMXQ9_SOyPYR zf)Oz**nXxM^$>(cc(`6zA&&fR%|thI_MioaY|%)c4RTTQK}BLOaPqyUc<$5Zc>9er zw8IB6^wlm2eOq)6tDY&u;?=yILo>8#ruMy{WBefHeEk3~pTr3yeY+Id?)5@eiB)K& z$w3$h6vGqkCGbRjb*yqK4xiA;#DWvc@Ei6VnO~Ph9&}$K8dA^s=Zmp((24CTcCe`EpFdE3B7ingnAbTqPn40dVjYRvfE3Lf_McQIZ=-` zn^pg(ms|Tema66X)jdtxipvet@f(qH2sC*OlZKY!B!6R=(76`&$Ob_gmV;e``jBxg z2%;-WIgZGJ5{lr^dynj;%^{5zN?1-i4>!}Im(@|ro=s@-`508q2B7g`7m>X0Nff8t zi0TC#pjcIsqqwJ%vwBfC?X2a7H=eS?sm-f!zC|F)KH84%9AqF${t2!R8o?{V)yc92 z`AF)!D7>&%ho~-f;0$4q&pQnoPn+@d#%uWDq-Os4Ibxmxn$|KO9+j(+^EY`uBdLbu zc;XUTi1UeJA7)e2kFUkbsy=wtr(AyjF&?@8)q!&{Y%?`_KAyALzJ|JWcp5(Es)7SX zO>uKQ@0;^x67=qU3eOgt!;iM)V?o5l%ZUjw*)gA-Oj%EU*L&cc%^~<+MH0@7=|o=? z{-S~7s=%D;L3KI9=&P(ED5Z#_whTEWv3LP`eK`facySp;Sa~2n1Hc_S=VMqo5gYf{ z!9#X3_Pncsk1srjHL8!{nkt?yz2*+CQm?=(3SIGD=@WSE?^rAs%e%+h_z{Yx@a}^8 zCgNAv8`gQ)f;KN#$@XkD9BX2PpC8o0JI23$7U zj!qcSZ-6yTV#H2Njns7^QfjS=w>|g6vzOn-8}_)+RNp(a(D*n!(3L`+aud)^*$6aU zXCGwQO#mC-H<98Z1$<7(2k*Bk!#ks*z+p`-x*;lt0-H|Yjp?EI^>8HqyD0{ls|z71 z!_TNT<0-h_)5kJ1!!h?zJ+ST-P;cfB&Xv-*#atL~F6acY%*}{VJBe}>o$2B3G_?4K zJRJCb4^*cr2Vk-fcK%i!D}?pNe&Ms*_~j4P+VPO5B+VaLV~P-L`KE z`mj*}`^}KY(P^@{RZI`hjGBlGK0kufvAeL&sa$-aq7|=wN0IAM-o&@kpSWrsCSM)9 z$g@s)=H#y!%DF{EWsJ8rIcLL$4MuooaWdZFCkBa6Y-}mWFq`Ud8VXPh)RQOPoc|!O?f8;9F%sK=-vBo-!jD zUwhkyf<9hCQzO&S-l^B|Y(01Ug_g&`D=$HEPBLWg$bjWe+C+1T6N%axLIN`v&}1l> zp8feZZS0dtM#_T8kMBU{$f=Opq?shSgl7+pgvbr?*`z?zjz~mD;pW;R9I91=Um87x zd6lQ&ql7hbD;S3>9~tAk*5!EPtsF2Z9gmL%O~iX%O(r`Ehq0n?EbdwRezPn&fBgbx{rs^~ z;e9yZlY=FiB=Cf=bX0pj09Sp?ge`53)Qa9E{4rAcqNyuy-=pgZW@yu%Tv|Z?IE*)z zhI~~Pk&?S`q<=H=)09N?`zVMw_W_9%1*1NdvuN`{2ej4Pm7eVP5(>8R_S=fP`1hl4 zJd^AI@Rkrl3)4Et^12ViF!CW;YNCYaTuX;79bB|ywkP&FAB}tUL-FOD)gXUZ54#N9 z#|9(^RGRL>hv>JkGf0(84)iAbI#S83VQJWFBY=8zf@sm5dr43JFZ?=m6m|uFqBk$) zP#47-IhOb2__&C%fz)XY0?V%y!>gb3@c28=!0_p8oVa5K{Nb-#f|1zS=O%V4=V?Due9;@X>Cn0P0Ww;%i~p`H%Cbd*?_!~4O)BzxkV31@ zosI-{8zZMPN06CH0-7GD4i(e<;MbQy+VAU4T2W#G_@sKFqddL7^=J%oc8>&~w%OR{ z$3^_PXg501C% zNg23#s}VvhH{dicf2^oggrED*MLAN|XkES}1gI|qpX~RrD>DLb=iNzqS_MME;AiN3 zs0nB=9~FL8h6}l#lq_@CaO|Dx+b@qk%vFWr`wb91*8s|#!cdQWE+5w>+djHa_6gF= zjY7vd^uYg1Hu(h4m!XhKdT050tO%U{=TS&!gZa{*SF>QCf+(ve*!i;T;>K`^8iUJlye zwE0bVbd?rJ`KEwTssd&o8l!>NJJCd*9i7h6@i2OF0z?U>gVM$dI%$t5I`nrwtomII z`z0K(QC1y}@e08vynAbI9xo5UGf8^)q`CBm!35f3t2qQ-$^zesQrK+ZB|Tg>Oz+s2 zj*eVkfv%pnh4Q|sI7#|7+I3$PP}@47!#BgOh-!F|VTm(-2chy!3qV?{5`ORaMw2yO zw5QKjG&t8DF1d@~!?*Y2QzlWA`1~VO)E@y@w&ALmI^!XuuTb z^RTjw!ApZW@SnfqsPujXj%>gMinVJI3j1*YJ}H!APpu-5xxN8s4&1_v-&x@AdynJn z#TEE|)p@isz!G|?UKlmiMo)GNf#vXJXdPINhBjY9(f1<3Wr`?R>sq7uKT6Qx83NJP z^YLxTYV4ybMPJGeMNeI%LD6I>Jvl9m&i+`43=WFJ(CY%AtPQcqQbp7or-UTMo}&jg zEEq}%W4Z7nxFC8Tuw_bkR5BYMePIiK%eP@B;Rj|8n?U3(Ei8DT4A*b(p(~v9k%5aK zxc|LNF{~h})4q)qW$S52o%blW#fCmf250O(@@$PF>4x+F zho>tKr|SFOG7nLil8l*&xcBU})-k0LAw#oDlTw;XR8kU|NeLlDsU$Q?gQ8iJq@qFd zK!c=tQor-L-{<-L(bIXZbN60*-?P_QYp?fxw>|r0?gt@@9r3yOVGNuwgx0@JV_FaQ zLIoFrM~4df(?{;musM_XK4qV|SLq|+)RYWp9k(5-ee7YX=Nou!Xii$PHdGxxnR561 zVS|=tGG1dJTuxLZ?=kBrYt~olZ?K2Cxd^!8MVg>F(S`IC6nLe{Nqohl1KeVZZA?uj z2|zCq?zAFsW~adKygFSFeZ-C4eP~`x3b@^R$Fy#=L-m~yjOzUd+hQ`QpwgBU&ILYqyK z?!!FeL~QQ#4D=n4i?_by;MJQERC~aOCaXEq_Ba{*TdjxfF@CsXpB6=^8&iP(3)I&& zC8cKr=|{XAE&rZ__2s+pwNQ%+G4DrBa=pm!$~V-AJO&;{&!PJ30|>HMfR58<;+Hr5 z(Mf**ANi~wKP6U`pJKHeWqy_6av>j@*dotwkyqibIxF+rRdQ(2%y{a2lRzsw-7uzh z3GRIyf}giUfa|R7;8Ca#Y8x%+eZCq^&}hfj;~U7YegUPqrqILPrg$re!>UIQU}f|J zjQaCTIG5}U;=PZU-~JeL&2FK-xCVpm&tco)$9OC$5uN63#pxQ0Fnpvm>CAeG!Eg-& z_J+Z&?^Q6Y-&_b_Gl7kv9ynZyA)l+>(BhTr4K_hbaUTNO+yx*aqp z_CAR&+@kiBxBND)k^lRqfmeAe&$46XnUhyJr>3n(Gn!PXaI88xwzb26L;K<4#d`P} zc^z*X9>%$~%do`N8ArVgLYdVoaCPGb`mudA^;sN7mR%O~!)Oro`_+#M^qa8!-(wW3 zH{*>qLz?-9r_vDUT<HWNs`&nSx044^4tQhfe! zIezkJ{?fFe32*-VQ0jrA4Ns)i9WzBs&Sb*@;xAJ%QTB`wF>;Ar0Ti(X7)- zG%Udz7oJ;(U8BtK#JA6AbL|6`CbeOlo+U0beFg_-L_x*q{aCQ72-8kQV_eA-3{H_J z@z5cpJNPx8^3@{enm(j%WJy~E%B za<|gch^3?+xsjfl+hFl{fBd@g4J@59h!0w-$6Jpa$an0X0Z~*3QzK*Gu?|P6e;tV* z<3v{t4^m#)ZpwB(O)>m4yma^udd`)nry2qLlat6_(-!l!hZS(z6)`4l*TW4e@zn4s ziKJRBX`b^uoZo7XbJyzQBn30<7drqazkUHD%hV|}U7z-~+tTd;&h-6)9u*o`(!b;F zLax+8Ya;HE>6HwA?Yu<3&@YBh^%_Ga4HM~uS+t-r{}j5m_C|%duc43IGP?F}4%IJo zAuWw1a8BzZBv~E<(|}4z9qIVnZQwU{3p2a*8uFL_K=8VY7aej?$*~lrY&y}&WHNnJ zahKFnqTEaA-fvg_%4R9bs=0~Nm)enE!ZR?*pNymK52C7X$z(afl?L9wjJ{Vsp}eyW zF{fKp;G{?EpRN}2#Y^NZU^G`oCJ1q{690F95N%uMNRRnlG@<@14a$8_nwFt_jM7G^ zOIiopO%m zA#Izwom$RRlby#=sC#dUIw6y2t)5WB+5HGl2F0PSM=1O2xrHrVdK2e_+B1f)^zzl*e1O_5)yZBFy{|H7vs1~jUC3@Um0;)C54*m7;P<7mxz$Nkk&j-?Yy z(ZKo!8kR1g#qP)0?qn1E*tCpR{~pPg_;(Yl*C1QBW~>tS%DoK>whHbUmu%Vcs*7q;V z*ts4@EigunEA!E{r5JbZ>BTQ^@#4h=ONDdqWAO5ZIXJZ^96R4E1NU)7IHm0*EsLLx zO^>c%u}vS+DNv(hkH=DGcM%=$We6@hUr^&*Em?$cV6gcDSoZmhrW*VCZ_Bpvt6Z~% zGYub7w*DGYZPuZG-@_=$sV~2Fay4&cBgI)?9mu?k4Z+~6z2k+Gp^j}8n;jn;oPwUF zIke`913yekhPM6QN2^QM@e>~W!*ur!{8}SVp*HVvL)>{1+j{dFqxayTjruh9%T>B( za17T8n4RIymr>E@5?rNl2+EK_B$YaZs5*#Rs zru}c<5WMfAI^ShMtg@0v*U+rxR6pDH-Grx!mzXBYo_Um)rIXrr+w#`BKqvNSDY zAkQ6N%r6=0PVRH0=$N`HE)eQ)S$ld@>5$Fn(W?OdwS9xo`RQ2N>lXPa%JHrD75JzH zQ*i0;GnhU%pOQ?1vG#pA4h%6T!&+I2)O>~8ZpsMnT@6xuc9TxuHs`zIZ^O4k*QsuC zEC1ufLYiq%MSqQ~_+dwj`!lD}>?xPRsG2$Lz>s_dyvnU*@;6UrqElT zLA;+z(b21`O&kr`ON(F^n6b)RR^4*!cF%`H%!w(eY%N5=DmRb-=~Zc z$xHGCx!b$B)kkC`*V9w(F~8x9C4ON4`$-_J_!l?EP7!I2yUa=(24b~^A@`zW2{=yo z7HVC=;@6PJ?JN{Suvw;q-Rh z=>&L%Y~$XhWI-^zmh_cS*s*X0+$-6G zD%W>P#;~v0fQ4-^5StoIf+-&aeOcjN8U8!BF)x+vi0sd0Dn4blPb($uWSKjPnrdcn zApwg-fmX`QYua)6mA#RhKUR!+>y0ILW_8Fi-H*)bFt~?Z5Gw zgS-HVTqKU|_FzFTcS`J0kdtO(!eu2kCmkIDq7548xJxYRikQJUh0KuL;}dX;nOQ7G zi}_L_@7@Q+I_>rBe##5+f|FZV^L{JIcmL~a&d8kqpR4xf1CY&Z;`aK%6Btow{ExOI{DdTZr5mC zn7mO{e5hIhl%gh~aq?ArrSe+w29FZvRpHELJ-R5wba`|&C|jI1v74jU+u_~HpW<~M zBIpS_igyZX>=hcGi=={#xUU&S6pEq_^oy z*vUL8@@n}>LGjzTr^_3#`FXpf{q1Ayndb^GsC+9Weq0yBi)`5t)?yY{VU z(_>FXx^|MQR{a26HII>dio&g8B<$(7u*F2e+hNSzAKoe?iSBx3!~ z^^(~9gVk~#%Q8X!?knzROd3Q_9ERbaazzU!%Ydx664w3rEzVPm;MQHrl(et6#RU5H zb;3*6jaYNC6^yIT2Q}{u=G!p6+6F`BBCZ_Q zD4sZ|QoP^Oi)|=5%|$3?0L=P<<&*@r6ng7Gv_Y z*MIA*{ofK++*1kfbg#3i7lPonzmmkxmth;(xBH*jhTbQ{na4BP{Mmjux9BC~Mnr*6 z&2<>nI*7%roX>U*yaJPLj)?~@?O|!d`{F~Za#rzSKb(5aN!l5;y$Dh*`{R(*(c<~z zuCpAeCvbkRJh$X$DQh%Qf)m@0iNjULW59lAdgH!M{I;+UOgLtO*U$BcFAVL)YR+`B zPiN~{!^mlDL7oQdw;+ox3fDu;7hZHUUxQmPIEy(`GpHOd;Zn{l67rfzD!#AEH1&5% ze1_7|H{2naH{8b!^WclsP&U1wo@E}K2Tjw~bNkP}XS;@};d(P$7QSUaM7(asn%Lp+ z?0X3A*F7#-7bDdBN&Ir-u1xX4202i(bi=_qSA@;RI&Ng(Q_1f>|4IkkiaLR=PW@PV z_dT|#I#bx&>&*t{n}FqcAwRnvBkq=41dVr6amK~d-1~EJ?38V~4LU4+I5Mp5pU zcCpU@5i4lxlC-I!BJfl${=nG=)41p(zz(>yu$0V!?2x=I7LJ@v4-!Xf$nP5}PYMKVrw!dV;0N;H+kj--=bqV(N2ou;>Z9*ftuK4OBVJwm3MqWWD4! z@_3|ZFfP3mLJvs6z`R{BcIF`nl1t=v=@?Mh&jX}&*w|41#4z9=+avq3MKT`&7wed) zsEXTPbP)?%_36`he=g0-lv^8;$309L$BOeSAiAp&-Jf6L)Qf&#b@MrjEKlZ+FMWnt zOVb42;(M<3r4x0zZ6qU|`{Lp2v&6GvJ&fkKNQ+~$r*V!09GP#Xz^VR}1fSg1C3c26 zPZKl(s`33EOIB{S433s5qw>H;u^*o&D(N4o^<&&vcY(IvDG{TQ*!w+WLF3E+{J_+;AE3nN8z#ugN}in^Z{~54 z;hmiGguZw@aU$JIKP5g;tj2Yp0Lh&9scWToCA0a?tM}}S5|7)rTu5-(XqUycjq&8( zO#LTX7U~bOffca-^D0bC_`xR5E`{7jpJ3}VPwGfZ_r`4 zR}{kQ?3XxV^ipn$LpAs`-$F~{9u~3WI`jQH36AdUEgq2jOwxaKjW5R*{oxwkXfV5R zPAsJ5F{{^%f#wZD9nmd=dpKT+9lv&#dA5v#P_1$(={*S!UYEzAQO5W!4#h{LLPU2P z`ipn05jYkH*FxJn6>NR-S&qYk(0@#WjnH&^ZGISokr}R>nZG8@)-TDDADD<)tt|^63)A5C|CGMPU`vK z99D5Y5>7N7feWX;uwC)Tn9HjYZr_af%@y$@*2`UOs-=8AJZ`iXvx-zn1W z6k@ubW=^h;vhbV#g!E&%m_KhI$(}qew&d4}%@Z6s(CvZ5X_p}{vJRp?u7&TGLTP6Qy-U(-TrfQPs0Vt7E4Ne8d#oCY-fv^vDW2z9KH@7I=N_+weoQ zKE2M(qfg_l*yzj0xqYJ!ac*Ok&~LjYK8zHhT%s13wk?D6Yj?ugYI|nqe}&s$6~Pj3 zUxJ$UU`$=O3*E=BVjUU5tc5?x6jS6uc84Q#jdEZ{o+ZL(BZWQOlQ7}8JZz8~$IO59 zVn-g&$EeMH=t21^8g|kL|EbB-tX5CjIN=}^kJQA>{`0Xs>Lbkjz_4mS0AA~(0kzL% z;O<8Q@Ja5+dPk};wG=gWuvQZ^3fi~?mkiceUJMaiuR?>-Kzw{P20l6p`E&Rr938os zg^ilTMt^5dAW#rh!w%~>NsXML0H3f=a}czX4olz5OkwX;@m46uwuJ_ zkvuNM0ex5FJL6K^zP*f0&Ldhq9xmXKMZDwC_w08^Up%|yHLjDs184u`WA&qjWOw8O zY!`S~TbjN?uWhC<$a)IYt$&3-+*Yu&%3H9kDu$NEH;mV?hR-@K)I^rx zsN4WLc}j~~0_w2Vpf8;&dc<;aHsi-WJ~ZUqR(5a8Vi^9U6r0o|uro~-J9QUezvhi7 z)SH)(!)bSRktTlnK@v0cR+ne~Vix@Xi z1$Bn1QMBx5%02apw~4q9=cDg|!onkXbaFdOJnoA-{uI#krF}@Nr~->VSK{pLlR@Ly zRi@OTfU1S+WVlJ0?(W=6S^27%e7Rx}IUKcN1?GXUJz9mTswd;khI{ZNXcE3@nT#qo22gFRJU>P~lg+uF z#x5qElf*^SwW`TY&*u_VmQ^oAL29!OrW|J zqZ03V2JxzsbT2j6OGK(!>rd$;sjcvl56)My^q&I&-hC|yc8;B0YOZGqGR?6^+ zmB)DB0gCvS`$ay>v-tD^X}m7HCihVU44$08!k!L?*&b;f4^#Mkrl9fQl2P7K*8av)a=Vq(&5%qes$?d<}}vzKc7hyU?M$91C{aV&igUxMp4h)5ciP;k$O|Fx!EGT37KI*V^D` z@?^|wT}9bWi(#mG4g~vTp;<{h9#T1g{ZB8Z6(PRREHfTY_)nqkfz_a6`jU-lejpk! zVK^#w%!ZX3wVeMwfBx>$ZIZR6*zJd0Z)r$S9?cq^w9rA@mL+@}&w3B*jV6oT*be*G zqPBP2VDS1l&fm_DtuQd)8?}c^zUw{iISdXck;Gd59a3;zV}vMTLa=zm-5825NELXf zP2hgr1>F=yl4o{Y-CZos9wxb7QszTt+FNk6YYI1h%zY?Uc+DPtjb}wON6??nUAW@& z7m>x&6?jW)4g^T;NXh5y z<3PN9=m;y?QOrK1d?qKGf0FifgFI+j@gm9f4b>Q4oFHIt1bd_nR51Qi4wN4XWG(vn z_+#2tNjslsUWFu2d&za&jhFxNOV~fXcvh0~9lR!pz+dA$)6SNric^PR@%SZNS)ej@ zXw`AU&g4jke%VY@e>{`)wS~QfXcYxc&P2Dmdg)j%04%c&~>eVX5MmORc%4b&0rqZXd~5&K!&3m1AlpW1q8OGDb2* zaaY$f@xSfqaP~ttm-1VhOJf@Js3S_!SGUVPrmC_}T$oVC=}o)`m)BWJ+L30FxMaO2 z^iSTx6n)y^c-{jpY2k1d^&uTX7S7@#%||g4-K~&#zFw@e!kyK3Z-(mp^DN20n!8n+ zBo#MLJxeLXikgYJ`E zCH5_nvZnX0zVNs(oB2ka$H;BR;ZI&)uF$1}l|-H3Tnc}(>0uYK)AKV-p$Jy;BM839 z-eB&dwu^t)c!KRMCD?A$!5+s7HSo{^bgqoSy2ceWW|I{y%`CvO3ny^p$PwsMtc2a3&@!add3ZuT2V^pRd-Tj#$60eG()*T-q7H%jeNFnG*4xMfIIM|)A<|? zA7e~Rz^#}jXW-40gLv_gfP)Eq3-25B;Fzp7);WKN%9a|Km35bXrbpB3JbN1R^EGo1 zyv>TUMzLk5PGZ>dM6{9bjr)Gy#I4^Cq1WaoxaS2>#>}_4ZpdRSv%drjP8@~_N4ub^ zW*)kAc;odW8*p~YUeKC^?{{mm$9t7E8 z{m@%=CmcH=FF=*Ny6z&v5UW_@r09rDSiy0aUxdFw>{7iNK;gT;9IydH)8~nep4yMb zwCX@sa})F4Sitsfe*>5_25irqhU0tv@W{zw`2O!g3~C=wCae08-um0F~c$S`+?G&}pwl zenAHA?`uGsvV*~FiVe6L*?_X5pc%4Ecz1r^2tTjpLH_6Auzo=>9_pxJt)p+T+SVL4 zZ*2f|cX!~_8OoG7J{=X;%)*?N?$jperyT#G2D7UV;r`eZ_;*zb{{4K154R11b+y{y z+Uy8}+CMNI|3~cU<+&`lG6Q>)yY7kbUO45%6gWiD*W-*sf znk)#7!NFz}{aiMfss~%s8#NCIpL&#a#6N{cS`FmBB8of%+@hZ(nc=FR8GTF;n5vO#xSCC)A(crjdGbB$<21(ty|oK|5s{ z!JQj8$znStj0>e`gFu?Kz=%&AAjO|Gk>Rb*Z=~z(c4WoODZj@NMT2xvUM>$`N*}|Z zoKjTzoQEOjyV!B*!K{8oEK?t=g?Zx;olgnt-l6w9&;nqJ^)NYeQ8MW4QbVNIvwEm_GMPEtUxtp>w`}1K1Eo3k7 zT(>CAA@`O6NuVieH|S182pJ{@(v0sKyph#M(poWtc0n;p-Dtszf|qFIJJ3nK z-)~3dTh|>sU-aV(s+-9gw$Xqpb;s91P5iFhO8&{8Tt498CVuMGS-iT9Jg=62iJspW zMa84c`G=RkQ1-u%bacla{>9TRe8?IN{!#U9zMbFAJIE#Q)joF}%jRY~o==(LsAjhx zqDJk9K7B{Rc{1T^GVfEMLIr*I)}XzO9eDAIz<-^u$o~r7N++|=k-gJsGCA!}lb>pk zae4{wd6oFe<^Sl=Ei0a%EYH_3ZKv)DF?{DlL6^Pm3{AXIOX;%H$-VO-Z8~@q=jkS5 z$eQ=C%Hae512^+K!^`<2oA>b1PuBA$(=>R`Ss(eEf)22^UnZZN&A>PN=`K$9={ru_N^$)0bubi%#ZKH`L%JgUFNXq`G z$IGUu^UX2kv@7=s_`i9_Hc3|sm;noZ_9R8#TXOi?62g-Z>3(v?Y;4=jU zJZCD6Hd8hc=VwH2Rj;VqTAfl)F2+j_J?L)u9-22JipCV#lFrjm{>#}xeD72hzAH=c zD^?8{^(vCCx0yMr23_IPWsmTULSABW&6f&3d63VigZ$7{$M`juHF)>jMqYZ|LB8<& zX5KK7=UdeNlBrG!?|8^nosk)gLDL2x)^3S}p#dd!EtqA_H z)f~sO!dVw{>jygeN5=7HQyh8yQG{P2Bw|i|l*C+ZX z7Y!4jmlcQZ!SQGpS;1O+&jimmJs|dez!^GKvg8}m@cP6e2bI@TM1!OUG5vIF(Xr{l zqQ8+O?z2r=y!+BNE;L4&eSiB_uKtY_lWkqi#>X><`l?3`e@!+>YV7~z4f7f+xyk2r z*uCJB%&xzn%ky4YT;KIYd{tVH4VeB`RCwx^So!w|_Ap@|JM?@sq~Cfh@ky(1juR(t zYGU*AR*1x{2He<;tKtPJDdK|N(%kvUP29Z70&eB4(QNvqYNmM5Mp6e67{(mtSaU5b znJLC@Va~>9#3vscvhUv&m_qXr7I&ozPU+Rdw?_q>*k6aGnqOneiR;9dFMJgrHm>0~ zjC9!a{i1lbO)cj;xLthZ_HFLZ;1!bi`&L+n`1vVUE+{AyertK*AVEv*(w+X?c^6OC zBtFi9o~~rWkLy8mKTo*#JCcU>~IWX9VN5|QBaD$>3r#V}TY3a{n<$K4n zmyd?PdjE|uMdz$TtyF`^uC#{xcvVOID!E16K5@4s=lpLDwMXCJdR)%I{*o<1{HH+G zwmT%4@t}zcV#5t_lDQ5&d!GCF?Gw8cv>yr!`eW3KN-oPVoK0%^D6zj^*Lvff<$S0nH^?aI7+_l1gN4sy%>IOr;Qi1Lp8m9%ZY z#f`OC9D%o8@m#w3N4E6S7q+BzllV?+5?7$mDCx^%wYS5w%43|0!8>vF@*ws(TF?Q? z-R1DbceUtu**$UU@G?=D<+(VtuYIoZVJ-<)_`c7F&oR-!K@>$V4jxC`dXcX zwflC1+FL;bD*qt!+^Pt@oOfao^JQ62l$ly#ow&~np)K!!+_I+) zVvnb$EME5(H#_dXob4r-n`t4LtAfAAoJ+tNPE*!_^G=dsX&K<5E> z^Qn=zZ=gK)^{}O6ECu01xFMt4n1V?SXFFVlJC_CLGkuyg!F(e5lo@ip5! z;#~JquE>0j`1hk^uB7NIH_f_%ozMOXaZw6*XS}Ys+k1%EvOtwP`Q#DXw{|Av%71{f z8QN^a_H?GD^;Ke9?wf7gjg&R8xGhPm@;M#Mjvq)S}sw=bvIwYUO96?i&Tm2eOSev zu=>RnJDcO)49Y2)_G{m^5bAy&4D@b1XRFk#((kmKHqO{3B{G)v+_XZ1ku zvq=HtzW@iRtPp-kcsJQU`RUV+Cr)fNLAPQkDvCeYdI0Nk=wrr)QJ zVK^k=-qJ)i{`NyIhcjc^ldUDcefWqZu}nBZQKCB(|MZ7F0gK6F{tW6g_oZRGYMFtk zfCYNwGb^nw{5;_qPN=z$r^duX-xta7O!)tTp*Kykx1q7G^=Rhi2q-$WlQnDjv2P#q zAZYkb_$SJTna7XNl7ZW3fjEYKm0MDg$3RLEFx(@A9f0pf9Cbbt)9zevXg<=!-nyEw zfy(>&yK_r;mz*vC!?-S6#}T5TkT(1VI`dk9J)x{X1u^BD- z6m%W;9p`ADMYjX_ILyBj$LejyuWpFm{W{_B@T;&a+8s@-&tSm9YJ91^6LW+d&U4vp zobU=T-{!ewPX1dnhcC*}H2rrtYwk(BUam-`M;h?Z>t}c(M+Mz3|AXmLEimjy3TlPF zf?ku#Vbfd}WS>%DRew|1u)H5W>G=uQQ;&mNm?auN?~Og}mtm~TMy3|$!`8koV~eu( z;B+^447E56PyBt*cGW}BIk+2E-?@aN?h3qSEW#st7GQQn8bm$KtgOoeJJxvOz8@Su zyR??}E{&j6+kDz8ry^*t=um@|F=dzxq4tL=G-6geS|2>XIwlDTkz2Bc+ zt?oqiEP(An$QFC&u~iRi&{J*)T6|fG!NPYmX5Yp~z2r!4uRG>G8H%e`3__cMx)>7O z4JL^C@kXy~s2U%5w6v;R7tcj*KVLC>kc^&?tspGkdf zGwF-bUfNW@6~Fv^hFAC6ke#4|HMb#+Zy9j3P zDC49WRpDj0CfHbMU`$;S4!4Lyiye9R@3J!0IjhmK{YI!Tv=X13c#qeMbJ_mezOcY; z8UBqs0Cf|$GEHZH)<@ew;FQNesL>ESr|=H04Uob=s>?C#QW{FDp24giZ-gSSGb+9~ z3F?C!F!X9VysDT8VUrN-Pu#@71Yde?l|${`Ch*B30Pidi(U-L@XdLwe?*|Scvug`r z_gHz<3!jOfvmJ2DoIbcKRM1B9IYGvr+Wc1Gtcle#N};btGYqYa!13n#@M*s@>kn0! z6u%ck_vEAfD;?ZeSi!#ilZB9!ItViAPrciCiilfHinBCW@Q%LF6Zs8Z+&nFb|MJA? zxVtqE%TAp|v#tH9bHGkmT52KrTXT*F3~SrM`1cQ3$3YM7F0N&sGeRNw@IE{|eJbg_ zIz!=j0Xl{fM7qbqpYL0s_jyh9bI_%lcwoL;6PUY!0ZzPY278BDW5nZ=I8$RUeo@Lp z)3kr6GWIlkbtwmi8%d%1A4i(%w4V-ceMrhapP;^SI{p@)!+Yfqar``vF1nni(}SjP zy&?v&in6n8kf8HglE(O6b|w5_-(}dpcr4oATShB)dQ;fR~xd=sh+JB^EeKgUIH%V6eNPpCF> zfXwMR__kLU&g(sts$`~M#f^TnPe+MgSE7ST6Yk=b&NVclPb++kn~aIL}ptQs}T@}!(NfSP=QaO1U~IQGye zx)%^G@o$EntC)X$GD_bH!KJ}@sOscIv-fr5qzm`(^I|WmnK+u%{zg;m`&>-ee-YW; zcX&g43fgUQ!Js=c@Mw2123pNTZHrWFJ(+>~qMeX0aKJ&PS$I}wBObCZ#&)~jc=?4I zerj^WTw~x%58Cho#e`q+?HKOXJBiXyZe!8Tda{+TrUQ1DX}y*T<;Z=;wGpy{?=#0o z19_}fcgGz*XIQ?=6V@5|iv7j;n10?B`<_Weid#pS`=(Mqzj0)6)UeiM!pvwZ1Tj8j!Yca9fMmV4&hS8aX9w&P;_aGKqfaAe~L^6 zzF9av&k){mq=_*bdSmUgK6L;1TiiJHHNJ3q&wqJvo&VmiiPyN%gl{aaU~AbG9C&X9 z?b4V*-}W!24wE>#1IsDxS`f9*T7uRVqi|KSA4VD)k?v?M(ic0>!USC+n{ISgc!2xv zl+k;E_cGTqi}VMk@GW0L`RJ2DyyBlsc#^Cm@jqvNB?~$^E78|H5%Ya7z`X5Sq4xSF zaH_k6@nx4#dXf^QS7qUD|5UWt6o>i-4)lJ6F72t*qyKb`REJN)m0OdMU-B3|9-P3; zlD(KVU*OHNUKFBXO!MkJpz6LqROXI>{7r5Ym+wu>F2vIOfB(>4p$_w=AHlIMH8l2L zH68PLM61WxvI~!GSQgD-BQ#=o$GLNO8J}4GXstc9tU@{*K9N>!)#5Mo9mwCkK3F(I zxD~rK>ruV29%FvSV}#-kd|Pn`!}~nJRpTF`+wD*I)F6}GdPdT)yY}SXzXF409^sfP zy=h{!8IGzq#Embb@bkI^tWb8tSG|m|WK|AWs@#{%kLGGKj4gW(3M+m>xRV-S-^{ad`JZ%^lDL z;^AV~H8^@V3kL8x;4-Wdlx}5WOk)wQmV1qh_l-xR?_oG1Xcx}gITVY!_3@aV6YAPp zlD=1e>RVwzN^5Pif^H>7Gg3D+nsLRrHd_;JY`N-Gk$PHlrI zYkLR%@q0w`7Isk9{1*HXcOUO;`G#LN=25s=0!{14AUSs>>>vLI7A+lw76E!>yEU3D z?_Z|BmiJ)NzHDgOodh+E69_HPdEA}2hq19 zE~N8(w4jU8hreek$J>3>;y)NprGnFbw5}zPqM!J|=RflxZ*LMrFJkoZ?jRalXhMZ^ z2V>nnO^g_@2&d?M<;;HQGN0B)mSy<>%y0gMT6;50&oGqurNzDjAcvR7<9QCKHZ~WN zRd-yquJj`Nl)+>gxnfG4_3tE0UN;FY?A(Cjc@@mr zuwmi0Eav2OCf}=-9!-+wuN~6lGk5+4ABDr1@b4y> z{Ja8hZ+-%eJBV(!(NMZ69v#jNravXw)QLyw8M{f}A6{TLL*$|UNFg+zslo5bU(m_l zjT-!ail0p}W(Awd!EEqoh>WojY(9)ut2vzJScfB=!^y|;8fxBsgzj2hy%R4OLC>J2 zFmav(Mt|{Q6_@N-zbVD6sVfQh7(T}}x@$=5@gg*nO~rTD4xx6}3Ye?Z4aw)SQO<8U zyYY1>v~>q!lj|wCl6D7T)D_WZZW(-aO-1(&!nuFF_i`>#qgeEUO4i(08b&Fkp@)EF z>QZOydr1fzW19xkC$EQ+8%uDStP9=Da^{Xt9LpuPUgFj?T!Ew@p+<%ol;E)pZR-ul zZCWlR-8w|R_s-LznjiFglNPYfZLrsJ5e}IC8}dg);njl%q;33z3%w!_qxD+_K2HEz zE{?^sGfv=FuNSa9_Zbd$ok7Y;8BDW60k$wb?AunuZQD`8#iw?#o&m9tXH<%vL%is7 zb_@&b=c%N>qXhE*)PVV89c=kkirS}nOq`=h>F>6YZuB0|5&XV~&jW%y8C+-ApX|og z(hbAUFk|^ZOjjF&3wBK7GB>HQQKfsJ@>mx8BFctMzrApgyC2x@ip6@rrSv`a7+e)% znrIh2`jr>~RykF`eiPdIM8eMOzn~@PVy=yx2L07r;dR^|tUJ>K9zWXz9N!A^Ii|-B z-|G+U$1QMBrZsdw_kz6+7a+2K49y;#K_P?B5!YG3z)J-lyZnMO;n|oZ6^d_97*fUR zE6k_K2*TH0hB3aIaPG0~6st3c?_ZM1iSB&ol5MAf%F9~ZeEt%?G<}cDM#hrQsSGlk zR!mb(QlYi#Fr1Gw5Mr_~?9v8->(8!ZoJZ+!V{A6EU;#9?vc zaDsIuy4~}{>i418{qF~wSz0sEsC#Vj3>BPcm%+{i>$1zvW7yrxufRs_A>7VA3Xctk zkl(w9*xx%4dyUG&P@`FRXxc0+Yj?w>hmP1<;*HA>rU-j-Lt$u9C|rH^1Uf?FAw(($ zMt}Z_3tn!)uO~FHRA&W@Fsx$987H_mmaSl7mjp$zo53Pq&{+MINyY>6=tXV_-oM!c z0a;t&;O#D)8nOj-JIA5wd`~=^Cy#%kKEZ15xv+fdK=7%vf=S$Vcr@W3TTqq9<_o;N z)M`Pi`rHQWRII}_GtRfCO+X(iB>%n1fEEMkD1Ic!F(C7IZws3GCnzc!V@}?*d z2-*g6{_&u*c`7(8{0dip-+_rAF2StT^KgH1e>~+g0Y3>khKmQ}g7ot=SocF6S9jM! z_4P2gJ}C>OBgbQ|qB&+iy##xM<^Xrs3Xa~iz<^6KcyIT1c>Zz~c14I$>tj0@*Co(D z-QDDuTT07Z7x2$z?0KE<%Dn%#qnLar6UBpKkgXqxCjPN7y*3c$&bmZx8=ugg&?~g+ z-egptWrm77zkrPR6RvK3ixwXrCIyLtmh9$qj#a9#w0t4yIi0nsS*oI zHlUS<9`4XO2zRX}!Nce`Ah)~-zYF^dXN~>PzNr!(-BE|VGX?#Ogf9@-=nTfz7QMzy z9fYsn1wd2FBxIbRQ}|&m*mn)Z^`{Tx>+8F4!OuL5xT1#Hv#OxAU;&i&6~n^PE|zJ! z9&RWu!55pAvB`N34(yadhp&_2jb8!m-QKZ#H`!vv2{j6nTChQi!_2g6e@LB=Z+ ze#EE1(MBVT-z2Ys z3rJC$Cs1IjI&D<`08b_NfPdOlxRVesluXPPzL)V3&fSy(kMu9WDWz&C{waxOU4wA_ z?U4Flp!{#|A1A2Sx^+T z8Scb+gHehKoKMr`e!n8Dm!1tJSA1aSD+#cvo=Z9EE9ri*J*l}~f$XXpXqr<4U-ot2 z`FSFwpDjY&OYTGKx+*9f9R@ODIk0}k5Qx@|#6OEonH$>znfKMlnK3_~2(QLOF-Iq+ zFyBOL@Z_or9533Ar(uX0z5Y2fSk%I7j~fFL+kP-%T8|l>z*a8z*^Vwtvd~ah220$< z(4d~{q)=WA<=-Pf<>P$VRVlz*zOLBDd462ww}E185F}oh22$QmgmWGT5tu0To)K@Zf+fn0;3Us^ zq0)ciUrTxF7}TRT*E?a=G9msvp^PC4mzV{WN0=iqYnesO%`nGS4eJg}$Ax$NV1i~8 zXw-d%UH2k5KV~paRXv6dD$xw{*@xM^riMw;e2xY8KA=wC0~~t==x41y4K0~LYn9Il zw+JMdJ(pCO$tz-U-Gge>QO@V~ng)za>@H>?*OO75e}QqEU=AN%yl40e13|p{3AFb$ z!2ORLvpA;=G$-7I@gOTSeP=8DJ+49MZ+C;SoAHZr@-c-mxEW4aj;SnGcWj{1i~`?*cL+qjR#>y{!ixejcCUD=W{0XDGS5@-N$cVo}<0TLu|6yh@DXZ*p$8qmtA!Q zy`cvXQK=6l94l9p^AevcF^7m5L2xVQ6MXUd4d=h^WDd06X70|+VW#brhs+jRkg*vF zTjzX-a;t^teXA8>uS(#5l~%Y(FcrHe^}&IO3YfTW6t)+?2K5=TxZ76+`<%9c-@Q~Y zdYc5tt)o%JE&_{-_n^C15;Je$2*)zs!?XzcJ&L>)a5`rwBc;^Nd{~yv2&FXehd>c8G#!T1NvhyHWi;f>QHGrZ z=^%6eD}27y2lw|rz&bU$F3N-1PKo)QdqtinU7gT%i33x%qbW8urwIMzZGzEs#e_d^fZ2a znTRhI=b=s89K1Nq3J=TgMVGHyG{sDwcK+(epK|qRwzCpFwI5*N8&e$T=75{`FT+P+ z-6R;%MH=0mbWLR(6@*Jui@y||RC`1&v5j(88USerOEa~5T0BDpUR#> z|AX;xaGe88bW#9!kNfy)Ukh4oD#Ed9VPrSFll;v$(J4HF%ZozLmLHBY&j+ER_zoOt zb_Cl!uAt4`1g_gE51s#uM~BQ)_%Zw>Zf~7|uCXi@ar)9UK?j5DdgPYVm~XSl@|6KORdZSimmj$98H-%o1>Su40)@XiG2ST? z8&gxzMJxsz5A$)e71vvEpby>~4Y0Rnb+C^-@3EgFn^0W79A|8+!6U0=Y3_d`NSEdK zTweO*%<)ytLKHYJCCC4qlZY0NPvhY8iCpj4G%^`uNOME`P~WKo8-KT>bo6D~z3T+A zHnFs)&6)N2KA(NNV;-yOws{0|eOE_dzvN{YGh9yImgSVgF*@tC&B(<;pR{+W(!i@ua(h2GUYsH^*sW40EcXO|5v*RmozZ5L81 z?Bix9Vszj1EmqF2!`oKRQOxH!S``Ok@$N)~@*04TZt!O!((SY=_|Uk*9EfL^O&f4G}ZkMTSR2*r4%ESjx8JiPRQ6v}ifSshH5bQ&Xt_)jY~( zv`9R`fMfnBQgx#O%?XyH+Jon)V`nTmGAHQjp7Urt<|JOe7sNe7meALy;nY?WK%Wdv zvH7A2{**TT6qc{pbta}d@uH0;Heg#%WR%6c1$y7U0l6*KH?*1ubDexX_Hx&p4j`Emvv}~iX!U&g@ z`~jzfeXuyN0}K|u+&NV)aYgCS zMffer9C!Tq1l2o_fiU0)>dyOxBRWUYmCb?p;{)KyXOj4I{W-}09010}3Gi#26+W3| zf^)u2#bK0yZoALSZ!s6YGWZ5_wlMD63tey5~JpX$n!uV0B z$Z30mrb6!84#ujA%X9#XsAfSi1^x-ZDXuFq=%N+Io$tbt56_|Wn$tMYrcXlsUL0qi zfF72nWU@_{o?M?o2M#V{e`_nT(i6I9wxkQj51L?8)=10?^vA(gcbw8`iT`Cp!wju3 zxWLZ@n^|_GS&O9fRD*K5oUuh~DN1&^p%RxFdXc{p&4X*e`Ryv)Fl9WpKN^IaOXNwt zstei2b@*Y%XY_U~LN|j%40?MIv^-~Ud73Qd0S6_1UD-u-cK0|gsv|WQ+tABJ3^9Yv zl+k{X0<ALdgBWC62fP?;G3 zA1hVC%#x2I^rqtYy{+(4?gfhOXhj*W^9wkgO49ug@U}A%M(uI~>$?W9_ed$Oep!gKx0d1B zl({&6jwO1noQa}llEBX@W5z6%g09qWn4a_o$9b3Ffch&)UV91d$!~yZ%nMjx$^EUb z*MQ$XWh&DjL6V#|)Od_KddN+{9krrpQ!hu0zpg;>=c>5#HFq9u=3u0RI$C#`Qwi6@ ztFmth#qU+o9HVsdS$BebQ~txb8#$h!BR8AjS(3NLZ_K~aiYb0o_(P=tC5um@MVJg0 zRh|QXk$6~;?1UZT>~KZlO7zn>j?fj3=1ND=I&msxX)4gBL-M3;wUDB+WGK$!Jf8c_ zalv!PP~ku~@_IhdpBc4epmu_$3Rcthv}x3BAxnq3ERslm87w-M28vuxa>kV!GCNdE zN0wg4+M08y!uisEUG)dS*l1W&76A*64&lZtl2qd$O+BqC}(_(Q~y0a>P;!S#StYRm=toniCuI6;o`xFKn1Nux0MxBIU zbSO^4-`CZ!pJScx_b1r1PnDHgqQbVVRb#jNNdKEZOv)Vte~U{gG)ILZThOk*dSPmYZsh>pF&Mg{&WhIJ+Okc*<_I48xs=3#(s)n?KXYH;R7P{ zAwhx878tUduj{ghxjFRw;<|r7|9`%ekJN+rH3tz?<*9Jr8JxK>98*4>#G8(dbb7`~ zdR!4j_vOd3#V<#*-ofM8i8d-GcN~7P`iZ~U4Msk~^#*?hmm_6_-COJ^*iDnPH|ukK z%Ew{r#N%+t^At2bD8RU)lQ^~X5GolRMm-%fTr}W}%lqC?@$N2?O^{-3GM7@>sWr6E z-h)2;9Zh4_+{TqsC77^=TQjAPr3G>t1bSy6;kg-1S967_Cd;Wubp_2lY)9t??qGv) z1nyswh?yHfnM@~hm30~z+{-)M`=uOhOS6w$131lHy{qL9nc|2N|Q7)Sp*-Yt$3 zgh{H?XtSF!sh;$t_{v5c^}7m1)<3}96JmrKZP?kEg$c3`aDV3p9KIormM-Ijjqk4i z5S&VJSOtp6ZrPyEoA+Y$EaK4|KFeY+bJ7fwoivMn^fTMEOTc4H!bGs zMld`+O7oXKU= zI)y*vjsbtwW(?%mj4m;$!UT3SbXsXKiw!P70Q(#E{0il#>YU;67f3NT%dCYXOD%;v zg1Bdzj|gKd93k>bjMGi?#|dU^7|)nL4`J@2x1gcmjIj8d7Vy;O3Qu%RXO5H(3wx8F z3FRm3Wu|uM!j7r0gn40;fj_sMDIMP?)L$_kw(NNcwXdHuzlFMZBZHy&_ZZKQ@0j&T zg&?)8gkNA(BQQO(PIyJ1>j+;p21Az`?_SA#c$dy}swP21avGD~ z9tD~lM|kSGwZe7%=b5g|V=%tvqF_m_E3;--KEvWzLDu(1!O6a#LMg#Z;fQ__P$*(h z@xf>&Wn3qdWGfGcV@?S@x!H=9_(QO{IZBYBheAJN8jKJLE?RGol=X#U<~BPx4h^v0FHPZV2YM3hrD!>Jn~w90Y%B+|a2@j@Aqn346Xi z{der7PO3o1bwbgF(RizQGSh7=1^rcPAX3)p-@PTu&C!i=qL}P^<~V;qlDsGV7Bokz zKw*&=Gg&o2u)fDxu%tgji1CI(*&r2&JLZc|cgz-C%#jwVCI$(!H|aCtdWA6FG8_}C zeluUUwBX_uEIodFSt$1LJJ@o2&P(M0URita#56lvTJTfQGFMxk z3Usn2GFgq^nDxz58KKP)Fx#St<<5$*CBzz^-4vs6`B=g5Q+wuAejxOItq`0|PZ2Ch z^Mmpg{t$JeANCbmqV9@H;aHth%y`Eg;8e3w_{33-*?UzJeyv?5v~w%q%4aSzJ2&S+ z>v|oU{IZdBhCKOG&#d6C)f3e@z@aq^QG2xo^vt@S%b+OvYG+r#N3ldaG_RTV9&XVaH;$}JL66i}9LLCFvFHS@V`D0gTYej4oGO60V`Xd<9@Gd>`pGlbgr+c zpl}>MHV{F-J>{6%7L9lE!_c*HGU&W#;Md=|Al(^x1$4C)!z0^H;n&I-JmWqGW!x{}&dFRaYY^ANBq1dAm!@R-mQR1{8rjk#Ic)NU zIQCFw3C`VR(NNNk%K^N|xWbcny!P@(4u+N&G!2bVop|<)o+;O~$ zVf_d3#!olYwC{#0@lq&w!D*Y*6)B{31m(|sh7ZRFVDDCMG<~uir`H@MiyKQR?4v40 z-Yo}7+jH<;KNpTWOVX>{A$;i8j-ppO@bZj{II|-McPb}?gUvCBRT9E))gn@2 zcC>xxg_qt>WUU(&*pk~atl`Wa>{F=5*T>Ey<9z^?_^z0Gb0s#^|HkcG9^tl%EWG?Y zoEH4`p#{D`v8Nd1EqMs?8@ED^-)0OBpN72-`gq7F3+><(p5%P1D{skyYQkB@^!9aT z+NQHuzCH&fFk`XWsZXdL8ve+R`iwODf!O zgwhW$rH%LrC-JvX+j5Q{p)N_CZy!*aZ4##uhf)5mwWP@MD38-`*57ymK9bwv*ensu ziQ7o3vCBzwg&FnnJWys$Fy8Q6h1Y&+!n3tKOoGBUX0L@QTI_iPzLngZ$zu))H90<( z|05hxewHHAQb_glahm@l1>0{jxP#*{$gCYnH%t5R+r$^BnR5|NzF!YBS4{z{@?~_P zNkGr~HR-OUI9cX#*(07Dxg7AsmWyjRU6(NUkv1EZt znWXkMh(vuhlly^8Ja!}%eM71+^G^=vBRq~Ot70+V<|8O{zl5rFO4zYC3cSfR4CpTk`FM*y^PgwJ!iV@!W3EB#LJBI)F@)E@r`{f#rQ(R~S-9siB@?E7%pmEnCEot`FVO06ND>r*J015 z@wi6V08N*a;AN02xQtB(*~}fF#`PUsDLzAo=O3qH&uFT-DTB0823wwr<9Qt$T6;-{ z3S7kLNcDWoY*xb0sZZe8zPWU*m7&RzGwAf+aGaiQj6xSjj$NMu?wWO=S5yI>?Y$Jd z{R34_enb1}t?2BEDa1cLiFyNXQqJoO6xW(gwQ^c;w5NiZ5+2Q@)Q>PZ7A$4r^;N{A zjbr(bw~E59*#eB?IQ2R0eY7>{9WCT~+;kROq2Xe4^v$orHS2GqJKjWv5Uy)5{1x7a zdxE#+_EF~oa|&;tM1#xfapbR1^qA_164&cdJ?GqF7O6?Ua%qIqsAo}V3q zLm4V)5@d$c7IHdY%UL{-l8czN2raO4J@;1sVnuyAG)J2Ll*&?dx)>aP z631L!bBB3Q9zpM>L{rS9c=FLvpn=p;GOB#swHM zSDMZ?a`uT+p}@Qky;#D^b-X;>Op*aKAf?=0{t3$Bl1Vy$94Ya(EUa%Iw;-6*42jzO6zj;byvmJv0`{lau%ge)}`lTO=+UtRJOK5mTmezmVM~G zAG$b=<;ZhSXmDPRrdNz{(`pu<*qw#G*Zy#DWinif8B5czno-J=t>mE09k1`+*l&}H zlTta&EZUdJeQC_$ao7aqhB`m69~VE{&0=Hc%3d6-p@i07^Z zV|#5Fn%+vpaobmLJTFaJT{xO97g*tF+Y`9gq6uwI%)%+_x1*r@C`R;#;P6gATvrl| z9_wzwJeA3K`_2S>v!xZ?WXo~0_cMHdy&d0v>cac!Bgo&i4Zk>yppwKPR5dNb$62Gv z_kt;1<$QNR#hC_sjmM+`ShsnMC1u^-1&)4wCQOP&%|h36rwb&^_54oSRNXS9%4_?Mmvd)cO_7IO%)yA@DeA_ z>c#LEsx)chSd9GA2dkcQv!{6hv~Xnrb;`$3_l|Hh7Rkc7zOOJVNJ6M386!Nb$?f@y zUct4zdRQ}X27;z?{fgHoVpty0e7p?}I}&iK$8L0=<%2!u9r!5c1{}Np9u(r{;+8A& zsB=IB2Y#KyN19t9wAvD+BObzn=4Y6`_RdJ#_!OqVaTk2_J<2FIgoDSl^>FTiC@ng& zoqM<1QR(LKOtO|0<8V_PPMmDQwUUyA4n2758`s})n)5+4>_^zQlX+ci2BZDw|J&cc zD79jKJ}L#3YwnO)$HSXn%rSOa3#1y^q2`JiB>MIwW6R~2Ja1fua1M4m>*qCC-e1V^ zS?_>&-XZi_mVy;hQ=n++z14B`QO%?B9SI~N}eb)|2ynL)3ErECam>UTVZSx>~kYpy5r_w6U>E$svScPg+& zTnb`aRPe<)jxAmRn4Ff5T|#eCd3BjIM4wUAl6~;sZ3e=@UT;&x50~5c& zI(`BcZ2E!AD(&I$sZ0p_0{AyQ9q!z+!VtAp_%~oIRxY+i8SKE02irI;%?RDLMB>um z<#fPvHB|&|qAz6(+{HLOw2z(G8-b`RFh%EXqJxj5jPDd0C> z6G(V}VD|s+MghB>*blWNnV*C*61v1Uv&A#Ag}C$fXKd!Ki)tZz$S^F_)n? zVjL`c!tJ|q=HN<^FVO0ffth`~q4@7u=Fnjsre=~E{Pnbi;FU43TtWb@;`aDut}F@v z`vz_t*E>nUf-I-6!KBr*aCVXiB{DX!x#Stsw8joL1oePCjb{$-En({F4Y2jjQOK3~ z1{XvXAy@1t+z+{drd%(;s-P|4V3dRXo4&%{I2Go2WE7m~%!Zc5w_sy-B2-QZgxG-b z@FpS^3>S*Q!}zCIw_ThBRfg!AeN0&JJ)ilS;LJ4DPsQZ13vq|vCUUBGz*p1P;RT=B zIH>(v$R+Z?_0oAbFV5wO4zEPr_(I5%ox+Sg_J!(2B|2(~n6W8tKWoE9>h zjD=3zy!|B_OlpKbBfr6H(GwtuUkclHYhu*!CRo#-1+QKjqe?+8X8bf|CQW_Bl+LPv z?NvK??G3Z|=^cOhb~hvWe*_8qV^2RY+uO$R_fOGg8vKm~pIXBC4z42wV@@_R8dpB? z2X@}(r!EgcDOcxzJcr%KBbYe_ul{}g@L&4dbS7w?Z%+rmR>adb;F}7Ry zx$u#Zc25M)Gxrp9ZYUDGJfe0zgwu~tOb$>N_j8MdikzZcbP+eWqc7u1AeH95~QqD5D2e=;4HT`_$IGl zoUS1&Skvk(to`W0f4b%f-|B%rwACKvY3~}{IwNfOhr`;KSFPvAj4%NleT{n?{uUcKfQGZV;r5wpQ-;_Xrh|H_mq(0 zZ?Ti%_dltFwenYtw<(JVmbr>Cx~p~gCv8mxUlwG+-LigevLPZoyxoRx_9>qKC*4Ef z6Ltce<^=LiUM~_Z;2#l;&NwSj_?)@ z>(O?#fH(7WF;Df+3Fc789-i*YPQj0m7TzH>HQqAS7XG(5J{IQ^&tq#n@6nU@f@00% z{1I!$2!4pohfl7B{1Mv9g0oM9@o1eD|3ZJ{_~1{GxIjdm-@4<9U}(o1ysBit-}#}L zH|nXFurSlW`1^3FadFZqL8Dus@&9H?nl`iGz(HNUZ@!%1zifHjFix3oU|Ygp?YbK5 zzfIvu1y}NtIL`MNxx2j2?;i7blO&nz>&EjYmk05lTPATX5);ay+Z~tTD)ns$?>GSkqRxXXT;do^s*D&it2X;O##KE3={NyV_ zv1gN^C-xI;diD@9D@RdWkrvH};X0QZBxp33RTd8$OWTF}Q2M?X8jT3VEl%B(sqmdN zLZsLi|M6*mxsV=-E~N60Z8TA8YC{)I(mAMl>>0MflVG{%nWsg77r+nb~D>+x`W%4v%EZ#b6e*Z|ztk&9X$ zUg$WagYm5&VfdFS$2XsYJv-0hLLECS8`_Rqn{!b!iOVaDufr8rUt_Y(IMDuC4ALCK zG=k$H*os=B_=$yR=;25~YFsvS?*=-kRDfwSi}B~VI(!+sX~#!fo=n|8$5_xO ze_iU))TU(13#e*YhO1}aK!+(dv@3lJ4LuH_SF%}nx2zbQ=fA)R&hrwXK8>29?CHz< zcpB`Bqv${Xkp>-Sz7#k8n;DrW)WDM=Be1@h0H022p~%4}+`6h8rkQW1t$8b{$Zrox zE{%osv6j%zC+kt7J&PA4j$*e( zH(E&*g4cvlh)7I?+KlfMX!DE2{Dx^q&2swjZZ&0PETYyqk16r>T}s&ZlDzN9G4hcy zjCpqzqxbGDyY26Hw#`(`B(;ApmS1ndbcG3Ie7KlO6}XP^hfVauzZmar%f}6Jck$k? zRyWd2c(J`{~1jXgYiR~LmQ+7W2K@+DsM&%{}LE(d;=^YXOTat!*%*uAtD zbw;G2&94l+w!9ucjwykF?P9ohNiB?idk=Towc^cFomiRVNA>sjlK+ZC+Pbt9&z(>p z!);4wTmBb3G-WjX4wa*hJ27aj{T4U&l;bpUUwkv(2emIA#e?FLF!B=*U)^xPCp)>m zd7aDfaB>e={d$6LqTXY#_B*t;3`FU+RIE>_!}`J_=-?NL8&=;!536yQzM~J03v|%w za2VdM3qh@dB-DTUig4x#_S6S;_VBvVbdMxyMYATg?G#YuHv#>Un9q5~KEU0hVURiT zCG5?fLm&Dq$z-e>NdzAQfodA~6_#R&S~f{0al9t_wO{QBh z#Ckazex8M^IUk8IZ8<)w)WVLSZkV9pfGv~Du#uZdwk>bK{|<6~guw=!xrXxxTL#m# zHC%V!+lN#+;6^LagPy7Tl7g!WnC!m6Y`GT6xW0C#6(JibS#=Y=N`Hda*7sp(xdH`p z>$zBtiz6BM4)%>7!s9j4r1Llub5(fk^nkC#Us8zsxGw6Zy(T2Ek)REtdr|pF7hHaL z9>12U)5i;8=wfV&%VsFyw%|ZG!!fh+8=ukjh;SMx9fs`UH^{2^pl`uj$d{TA8VHVhbuR&0v_Z?yhpMFR;pn6Dux3XLyTWNY8-2=&Y#Qva!rmQH{fi*j zM3;7dI)i2NzQdNYB`9AViP>Bpe!?{_GhPu!a@tB1p_YpG8~QJW!$lm*05V+JbG`@|JAW+fZSy zDC-xIK^^9^=<=6YCY?p^*h^cEv2Oac*cljsDuJ?9BX9+4K0|5eZvt@ZZZ;*?&s5+xO}|)@GjN_i?C-O$J1)R2rSF9 zX1jhCkVcpTEnz4Ao4p7hh%kwlN17y72eLCVUXhzCpIr9jU}do+-RTTt1L9fMXh4rG zjyg_DI=OyJ?LxfJm5I&CswkVZ6MN)$(;?@(=)23E%hq_YK2gN}_i8>_x<=zf<5cLs zy%VK7@X&x`(o7fjd<@Fn3%OiQtSIK!v&*z~2*l!S1dyBFq zcaeguB?ZR;W_*c;^6pS{v2XkLH@~ylf*-RAx!k@1@( z+36U3$_h@+-vf{JH!;bJGa1GA#<)RY36%B;Fn)^#?kHFUQ-&qbye^G^7Mtp*wlba#rjtx7JKY|^4 z^brT8o!JAI<=DG9AIVN7fqHMs(cLE*xa?CnyLf{kYj@{8iIz8V+@FPLdNK$0xNbxh zt%=b3(SrD#pL-3LN1}RN7?Logg2q<7nB{>VCY`0T(kb*>Ig$ptkFy?*Mr_|-9x1&u zVfQ@5UA@exTspL*dUNxSKIc?*ZK>-}nkVjFAJQ5$COXVwC zup;ve-j&tBv3X@6^)`ViW#XAcezI^=ts(xmvKM^+M8g%mTX?v$6xUAA#cf^AbbmCD z%x>PtBVs;iu*?Y;EtJRaDJ(S%E+B(H7Bu+gBJ)&#G8lenVdgTI@S62SJiY$}E_zXq zdz{aq+vJnDAxUv`ji)7Z&?JhfA3Q>$VIee8#Dk75=z&LP@?dR75QG$~(T$~Ibh)+$ zO(kC7bl0)8*T9%&-Sh(YCkf2NkE8zSKZ6U^;N0pS=F8(vRG=Zgma?)jWSIf2jXJOOWgFZ|w61nYFKgNsN%y1xH_%iFGU^N(co zNwGs?yB=r>YC#e0COoWv8SlN>kIDKgaD>woyjN7j@s?-19bJ z6p~_>5_xm%@1+5%RB}<7UM^FIz?=ZchyvJI&D(3jV>+H#|Nz6&V&&>KyL&A!qTZDyEGGOPU9Wb$593)d}m^_Co!nn1jv3k*wear;S}_-rpOU9lct=xk(SAcv{H`G@&@XC$XLe#ad~ zUHFFc=y!W-aQ>jLzy&67qklED@-$%d%yj0J^dRFhV=NqPKERkglz{6EHMF!!%mE7bC7%V4LrDFwu|{ z3VH$=x2#Y`f>(wu7c#K*=~b*VPQ%cxw)pkIa%|cc#+W9$Gf^Xkg(K90Pk;A$&(H zZc3}bX@BnFFGn}}qbWu5hj-wuuszUN{1?iDGO<#k2cAB#fZ4C+fJ2EZHRh}&{p)^o zet9a&d=}$6Aqv3YR|0%kbPQTE`k_tq7VcI2feqCL6h>LVG&zBM?iEnZjm3TmYuszC zipNS0!MF#enARpmZ;h02VH$(B8JE%haSpuHC}SmQ>Dv#4UIe$I-cJAwHoh*m`Ft1zd@wdGeMnjCco*JxT0V-BYBqZU{%H zO5>Uzub{}^1Ku3Fj=R=7;RwsQ82Ua9o&}tP^O-5o*SiE`n;xS1l0!6*wUo;!+F~+y zjlOH-^7YH)(4kEZ-6O8Rq1EFc<9jYF-gOiW%PjDQyeXPZor?0oWl+0085jDw!=;k~ zh|8P~d%_LqyHOOKx+=l09CaH*nksNg*i-y*L4u6V6r+hZp9a*=g6f%Z@ZsSHM*gNb z*@UFi?N>|Kn%T|x`AisY&h5c>H=@XFv=(bAkx zu~}PT(VC~sqr=~z5fZG!ZoyWkZ3aA1p6S$Cy=y*m!C&TJYUzsO%yJ z_|Boio-efL@g|(CEJf3NZqoeRMEtn?0{ZX^@kihYG_Mh0;x5EAi9~p|sRf>Ejl_o9 zyV!YQ78R7Ap$9TX9JebJe;dftR`1CmUA_Sp-j&1LGpaP%_dee3F2{F{*MP4lz;&!N zjdM-Jzpnqm5KoHw5+5+Po92M(>hpN%=oYdMnL&%D|HYqK0rcoH(ZJ=8n3qxmqC4I~ z^O-2j?Q$Y%ZY>aW|0x(RjfdkR93O76KVDEw!FyW9G|m>--3LbC16w(4vev}fpe^Wk z<1lLePR8`i_4MKeLu)kk==OZhvo*_v`@9whH>=a6=6(2m$t0Y^9wo^|zfmHUrE51* z@M-|(p)aV$FEGiJN+vsdbiR;M{+lch{MEymNZ zh8f*auEa;2*TvWMF7kJ|;Y5*rxZ5ip<=1tf(m@Gwlp0HqU%mk;E{kJ*P9Kw{D#^aC zige#MbLYbupt2GIx%^3TQY{&p-ldHeqO8-r)wJ1hHQl-wO_xGzFg3UZr*ob1vqH_l zF;fh>gUsPJ=L7t0`4vkIM$+jQI@GRfO%Ls(>F|b)==d`jp9i17`c?N)df8{xTl*99 ztsC&%vD+LQwjKX?$WvRIC|z_MO)nQ4pyp4GC&78XHe8yAr2~$*U!)XO4)H--X)3&a zGXoaiKTJQ*?xyfoFT$_tbR}{e`R$oSk%NDbCnHK*pDEGkx;JQd;woO|uKy+9>``&Z z3Qztq!nikjU~Ke)i9B$f@i_aC4e7nd&KdEDHJSPuZ%uoGGjDWrj45@t@s}JsLs5(k z?eW4l9`?9w$WN(bWz?M{1{9~E^PyhVi2iuj8E5usy$Zm!rT&rQIvjD#9)xdo1?Q}_hBaO=Q z}~2uyGG9=+NrXfbD%GIjq|5;;e>7}wyRU0eKX#ZohviU z)SkCyrul?2?;EbLdNZ!DMg~Q!sYe}|{VJn$g!yYJ`K}JD3G-$@B zWTo>=oXSCwu>KBcysVFVt}<9u=8wG_ui%m1TR2Ji4PLtLO79PH8R47*v~710I&W*k zLAr|*Ifm3&%fndy!X24r8MLje28+~-=rL{y_62ZVUsDd@Yvl&0jm$>ffM_~%@fK?R z?ZCQTq|@PjD0R6WZ^|wu&FbS^_Y=348W>Nay8*t9sb%ipPl6A&5*Q^Xf?blzSj9WR zu`;tkFk1qTaN(T(d!=Z#Mv+AH4}p#QGq~|K9!rPvg-ZMUg(+>OO!Pq?S{SmK%h86= zqE(4d(tZuba_pN%MJ?7PR=_%zEN5ryUq{`#Zfsa6M{e9XIn}8^s<(9shn~|w$#c?h z>!6ISyZ#?d=N;DL|GxkBQqfdNJCz32>waF>UC2mULL?+BTLV$bXfF-zUD`uaLnxxM zLkU@_^rnc+5RKpS<9q!6aySkSw_f+_Ij-|OA18C}-ZDmPxhgq2SBbqTqQ@FpnX*s% zRzkbO0r;G33?5up(KbI9!>oHyt@00J{Wq8KDt^ltt?FiUA3kArH0hDghitIj^bjuH z>Vc_dG)r#T23ysC!)6iM0l+WKK9dq-A1tIw8doXH-a5bHqKbRJq7S|G|NhDn#~U`*tb7|Ym8Ao%q%EQ)D|(n9Jf z_#=-IJ7uu5?j%m148|3q;TRLn!CDDN)LeT2Lyd*dSe_4sRCMt3DDC^*nE*QO74T(w zD~!B<4_99+W6L@E^Hr$=DUn)GcBI|!^A~~cuOP6`JjIT!31QWr1+i1Sri9e3B!BOI zAwD`c(OA3&lynNB@26A%7et>g@oJIozS!)*|1q1)rO=o z#GJuugH@=(tVge78}V(S8-!lcVs0=}$4{c@8;li;(A(YpSfORcN z;I^*>Zq3{PUrN`2@|XQk((w*PcqSqJnlkqE=ab-HJ;ZwPFT(v*9qI(^pu*4_2Hifv zT^$ir9pd0))tm5=yapp{Ikf5Rg>5^H3-zSURAoj%GmF{9inrS;@}TLkfYwGZn)1fw2LHEx=D5q8N{!;G9}n7J|( zcOMQ$2b~yHJ39l{s&iqAkQ7Sj3F7TcWi&acgadrk|3hc|eDM)bsFF@feap#|$4zpv z?>~Ipz{8%-Q(*(=|Add0f|Tc|j7s|1Se0Cik8~SQ_NN`wy|{)s_$Hp&&9<=_m#(ri z;W7K_b|#iYX5lV6CpzM>fE^QC!M?6{Vq4Ff;;8~dY@D_ZfAT#f{|#uth=wC%3;Mtj z%7)C+wuPe$58+>9SNz?+8v`_Tk>~<8+pNPSuP%ai+I8UYbwMmS2l?B&;c|NwSlOCT zU(8BKNiT%*wpmcoWeg!*UNA&CLt4jK*qq}q)KwM?=WBuc#l`S6r-Q@=zb368`CviXbMO}tK*jSL z(P-~-P>=S6TNV#NYNrZ;b4JAe&r!1Ci~we>roJ818OUj>B8R_r5j9~&(7YwV4kjwH z6LT2W?6w!Q&|J$lKdS$L39fXsL}9fpXnIJPE%p*&lhQ<4OFOE6uc;@(Gk=hElLa`W zn1QyMCHUb_04!=g4S%MJK;&Kw_MEX{Jr`Nc#~GaF3#vYsY>cg3#Ri#JvR>!*u|8TeaPDa_EEFf0 z+wqqKhW#WSLcHMlWjmT0(;l3;3HWv}4)&QAK~dF15W2d7ypJ&;ZL0-I#O+|{FEEDK zFb?$ojX{gS4$xog0$`;BR-5w4u?cTt)>=ou*VW(}Oj!h}uOMM62P|cyVU>joSeq?^ zBO~sxJyifFy9_}6P9!mzD+A7oT_nskgXreRF!oamm<3#t%!1teSpFmq1!qR$PuC!# zE8E72UM$P`v|5yQmz2WXky8MDxp*b@B)a`m!&=gk3d|9`R;d7K=Zl2f-PZ}o3?|m)gq4oeh$0U*; zpMuHSByN;*E`>X{6JYNT0dU!TmxydROSpBVKnr=H?!g2(@h6M1&naf6Ms|{b=(q4g zJ_pik)4}!YLn3pgi+N>l!Z;mEAcuRmk^5Zk1t-X;N%aEfq+ zcEXSLGWb*e6h10Rupf`d;=(I(cqwQh_%#fYrJTEDEPoge%HGA4Wv$pWrvL=)5afSr z!Qu7x@bs-btm_;nRwK!9|AI2ON{GQ|bS5U1t7B~A6r7{oa}HJW@jRCZ+CV?s-ImE} z?F?kMX*z-513rjbuL?tM?EDqus);+PAt+ubISd_oht|h%cN1XqX*uE_Ja-W zwk{40VZ6c!qq@Nc9?WedzwaiIlXZD;KezVZ-1cmNA7&Yu;BdD*zIWLAf8PIAL8e6Z z5uQI`fE9M7pj)yQg@@Gf)ubur{LVw2O{XxA`W5pX1<{k{e>=P{LRh#w`E+v&^S&&G znWtL}jPpM5PhAGpYilsBaW?u(X293V%j8&m3klDX0hI+67uY{X9&zW_S}_@59Wi#z!2l(Gh0Z_$SV{OKF@{4ui}={&Yrz ztAGhTt!=rh|ctA}W&z`8| zGz3a=BrI}_`olW*<65-$J?0R zMRNai(k~Z=iMp8+xn<2yvP@Ut+jWL$dr$@Maoxe2+hwu#P#&y%O7}&pFr3ciieEEU zu-$Kz*(Ih@tbye{P@5e^b1Y71}!;ju#n`#g`w7m?6GzJ%<)9NAOA2QfkoJ%e=>3)keM?3_ePkE{hsK~+bvx=?#*x6v zzl>C;9LUX*#LXo$bBke5#5qt>rE^}-)o@?*IMG_$ z07qx=uofL+?C%Cy)?>CZrbHXy!(?;h(@n$cW7#--x&uT1?nWQ=D7bf`3!)d-LEe&MX`hWTQ9@@=0ccdbq>mZ6q9{9n;4<^Gt3T8E)aOi3%sh9z`IzK z4bqillSh`YW%BEQKaqu%6T9G`HvQ}ss^O%-JXAl&gIkxh!>iSb_?s^todj3mk(3}z z$<79gHJ3nho<8!>yhK~Mby`#HFZopb3u<{Y|#ruMD){)#B8+9brL z2olz4`!3?FSkLKxx`Q#Ma$ww3Ym+piqb3Qd-OgwdTTQUW(XgJs*?zBx3{Z0}-E;LofF&xNgB~6xb?>9iDS>O$6<%RO7`A+UIrr zi3W-uu|m=DB&?lpgN;=S@l4)Q?6>3t`55Y5@6yD0=Pdm2A`>$UY4^baReb5pkNP%L z|6BPVVA3>vd-yqSUa<>(c7@}>Cq;Is@Hbr8WPpmAyzF<&FJQld7Yn}0WB&0TXj{Wi zXZl}({M<%qF-dTJYjot;I zd2t@T%oQbrH=Z#qeF7v)IU95jW`fh&5-9ezMYp|TxN2(;%&l?9>w#Xle_A-&whh6b zWDyj2xe9mhcLv!#3!&R=F-Qy-g4v6!z&ZLJM7`SKaY!}T4;4UxN)0q#D5tr=i*R}8 zT=?WQ4~iANLDq6Hn|fpgDPm@Y7YhYZXB#8fP5uA2{2Xj3|XkX7R)cUQDe_iBJ=At*W6o$id z;{s@PJpzX?1azh!g)dtauxd;aKg3F+c}@xVKFWvhTn(@>c{Nzp>;~7?U@%L)1cz6= zgiLM`ylYNrd@p-RcoPBHS(BjX*as8c4`8q8V%&0Q9rn)Jf$!Fyho-w%;3)YDIuF{x zMPwX4SAPTP=z4fY`xH8MmVjI;;^c$}el1BtvPs0Ged$-W|N9sA*uzR#W|jgDJ!he} zu^L6`=eEVD0CTqa;H>At_~TkSN^%?^de>YS5?cuSc$cE&^Oe}Bv<>I|=ZsRbDCh91 z5QfQ>;(@9#RNfGS7hc?ABR>?dmMy95YySnL>1{DlUBm@#y{%xZn0 z`RJn?gX_C%Fr(Lq)1FwzaZ~eTI`-dYSJYl*i>sPhMR^z8@pvyLT+GL5V?u1qY7y4Y zNrv@G>WB1M8Lx6Q9%)m2#>`xA_HcQWAMzBbscx(5A!L_^|M$|m2L4Hu5@ zf|Ml6DK=OS<>i;j4v%CaKVJwe6NT`_V+}0(u@3|5cR^}G0(3t654hA%gP|$S>U^L* z@Z8%lC7gCYEVn?}d-K3nYZc{_Rl-2%7ra0E9z6_&*t&POA$;fwJPV?|3^#Wmt%}2` z`RZt!uE4f7O0c)qPG?UgUV+KgHL!fuB?vGb!DYKTvF`P2l$E;yzUtX9D3c5A@BQ#n zMHYrM2(a3}sNN8%08+2!!-jgw?Tey6uck8UJBUXCnIv3onSt$PmiW540TynQh2HfxAUY$Bb|}W;s74Xy zWp{zC=|j*6=fd#qW1M#_IgFIXSH>uE8VX;Qz(4w`cxV^kjmx*7DP0=x$M?gc4TrJ2 zLX6#$P=PyEbffhjWj0h=7p?Q$aQpijj2Ey(5qoDmb;kiK4gDdUa*!Kq^WnpLCAd5k z3p4tbQbubRUHS@y?5XQ1Y?xq1U|o}#7fNcVzbg8dpH~1W{g7Bs&se}a})mD z)WF0B=wl-ObFtV1>>+ZO#&Mlevn(+=i=>-Q7qsS!i{Hl zqSA^^bjz}XFX8GqnpJ^b$*+jZ6MEil^#;e+;`r^j2wEB$q4TqL_L`fR$(ff2O!}6K zQj^uf;%wCuJvR0I zX7-JQJgZY^$!6L_vQxwMD4*?*soo_R*z*Ez$%f;j7)kc-!DxD?GSF{ZB8<;nVSI2rm5tHWzg<%P#~r z9y|lBqtbZciy^wrh{LDl`fQZ2D|)>j!)tcC*go6+m>g)#mg|SJGD|48%}<_fJmtuq zwY?96B4r?YycT3$@j*p~61c{c8_BGAw#y!BnRddt;V&B zs_?A%f8>R@5lqqfn&Htp?8&)`#(8(~^0_Q1`Q(72K_5|js~IcC8^x;3?_;f9XTj*EJAOrf^60&rIr08HOej=S%$1d3Ce2AvDPD1;CeX^ zzL>9plhfAYe7g|z<+_JrVT7#d?14|s(a1ern8=!bVti8MNRGh_ph;kO^wb!Xcygf5 zigK=$y-jU?Q80z^;!oH4lOxt@5Tt<#SE}g?{ zS>lDs65e?Hp+5#I?uLUwULdkN9-KZNAv`3ONH6F+O?2la|kLt zO<~4`CGg{9197W(O%A*LA`@>{!$a>q@X62)24|Hq7S+3%c&FHGdQ}68+-;y^l7$&W8u+<5FqTC#y|=aD;As1(t!z`Q@Nsu@N{M-@?8CUK}I#O+ni%NU zDFT^8Pe5pWHN@V{1fTcEU{tP(`k|X3=<6=heXEwV@G5}%6GxEeVIa$z`U;h@$j+<_ zMB&|arg7hLaJ;(_x~)%u!r8fKk)(l(DDzCWCKmGj8=>eTcC=3j1%A&^w1cxYz4Y z9V#9|_q->wHz|VlBSTm}ZU%U-f%r^bAT24mWYL3pBr|#~5t0cc(p#h8+NE;Xt`iHs zgmTqgr9tqg61b0EA=3s&$ooa#Na@~4n67skmOr4`-~Wb4QouA&o23f8>fyL_eHv=* zZ^qu*IV7_*o0KMK!l95Ju;8kID?IsNJGT!`3%>-POXBGGAO%cSD?xZ^6Ua78(wU70 zz6)K8Cc_uWTdi)=xljzoe1*|y)lL-SUW-{T?h*6r; zqC(!T{L3WFiYGiG!YD2+jdxB^ozBFGG#8bT6+*kK-vTV!g<=XC;bV~0PO-`)k8wfo?XKp8kv&$mi6 zAN)Py1c}m1aD|Tro~GHhm6WHqQ-z*yj%31ta!a`UKTnV~59&^EvwPlgvtw@s+5eh# zk^j_eTw6jtjx^(IT^fupN^;TW{xZ^-u1^-WiIJ1M4CBND3~$as#?RsiNfN!poZobp znVDsZI=L(H-s(WCUUD4bOIFbNlRunaa2J|ix50NhKW*V&O;*Qv5aXyEV)RJ{4O%%^ zqqH2S4Ym-Cp)`_qD1*G>@`MQWKoEYI4CgwFAYbDlxN@<$>G2kF_;o$W?-PQXiff=s z_%N6kg+hlJUB{<&P=>%b-abiv6Nl!(@WX8Qc}x=bhG}5du~qnmYHWMg^J2aL51ygk zNP%t1a670PZak1P89k$Hl9&P}!d$OmMbs;Jb5aEFn0#Z_4AU;dPDS#zh?kw~A;OAR zNwHQ21<>)V9(G1H!`#eR44xT>&JXB3?okBM9#AJ>y_tM27e|3_GtgnZ4$8@y!3_2* zN!?sS7M#+B-q>m2SZoMEY5HXTe0MTABZiE>enE16_mgu2N>GzuiWNGQ$mjPNpYn>s z=SmfD7heXJSE8w(_yo9ehk@he$eb_O?1C;Qf3SF4A^6D4zi7$28`l8hw_e<|NM~027}&73(Ue>?e}m)p&sPUeFZ~@v+#m0?cN=+#kUjdX*bVC5DsyNt4=N8vFrqqUsFPy zq6f$yjd^HDXM{V&?&25L1;*28H=YhJzSt~C`2Z6j^jZ@Cruu@fV=}xCX@JDDXJB~I z6)0Ii`J6v=(dDQD4ji$;&M*mFr9KUV^p>E9-e(|1V{reZBKGg|g1?<-C?7ZpVlUa? z^+^tf=GbCwjw<;tERN}ZAIjVh=SH^=(zqgpgW8*t;g*s%M(&7I3uBVQT>SYe9C&o1K*x~IC2DB*Tj5&B8*zo5W0GV?l`J7!<^Row7In&BtUv3WfzH5O*`anMPasJydJaf1QY!msh{vhq=VrO7I z&8oj}pM-OBc7vyK1dM8wfyYHlViy0EDI(l(w!IA^R$d30##&G;i6bYASmOTk05g1b zAuGOQA^U2t37e#n2>&@82SJ5UP(7cITF#{y{J8-Q{H(F?`xbQbJ&J)vd%Z#KObdGgk=!>x z;L;5^tDg!oAs67}mRI0<@HM>r(?k6=KOpU)DR4`2fE8H-Uqeg4?N>SMSI&cxLJv}x za+k!M6`|ZH2Usz-7i?ChLAjV5WgdSfJiqzCB_#wD>>?q`W(8!?%w*(dEl@~TgmCR~ zGAek9m}Olj8jH12Gk`;x)4Nb!+>T9dTgrCcF=Gc+p2NtQ>tJAB4+^pM_;Tnnx~*!! z4X)Ynr#Ba@K8!$R{S{#Q%3<2iD!8Gmi9G?cQINi~pWBUa+dCap-DQcEYN_a$oQTu@ z#GqeW0y|4HgVj2c$L0$ekv8h9tg6$1Oy(*?%()I>=et3pZxQL!QX?~!6^WtkdDuCT zPxI=9RIl3%pVU0T#WWcBPluyRQ3RIk4@Ct&NzB?$8BQCi{vOs1&CiP9b>=PD6Bw zMRS*4GB!!5<-AvXD7G4WA41VUG1EB)XKs27{CEL_iqj6(rHVK>C zKHNch7lE6)IjdJHkyG3(IEOqWw|P^@Qs0jtBU}e&I%nZQ$rG$u-;CyGs!&&G5DJ?k zU_*;LbidF+SSEu>5qDvsyEzFm+ea3qkCP{HPoerfH*PALgTAkeAYxApgsL5ZRlRgZ z`|=E!7iPn?7_NVsWct-G#&hd5a-u7cT-!WK&@Te7QC^yjIyYWA$f9O3H@;E$2sbT` zpwHxd+|f4s$Mb{F2y1Y`4pijnB+YMm(fVezVWn|5Bn*V(0|gT4msmaap#2 zvaP3f&SZ68xzRppFRCw|MCH5p;Y;~vazs#@1n%05{M2h$6}b|R|Mf*)>d`GYB!cc< zhw#VudAR3=0-lQ8hc0? zIQL!jm@5nBfO#<&RG)u9?43`O_Mdfw(VCArgIW;-*bNcz$3#jz=mY&!aJL^wOc< zCo$}-e@{88RDa$mgwdLFadk`%xNnO9x<`WUsQ{?x@P6>PvLcS&@yYpk`iUEk;*##jv3t?;X z0w^lj0{O``aOK`L5H&bYbEX62tmqV3`Be@g7rDbVek&O2Ukp5hIfUy_Fj={L6Ol;_ zg*VNcVSK}S&>wn7d{ZjP{HuXP>bD`Bog+itID_Q8L*CpVUH` zVFPqHG#B6L^5ZR!$6$Z#8%{Vs!B1r!n4_G}T3-)lIgfp4?z0j&2TLHlya?=k@4y?a z9B|?;hnLa=q%`av**8G@OEVL|d-M!cxM zShlAYPS2&YkID|PJ(UM~+($q@P85HvpuX&n(@`LLJE<4fC2kpnq|V{OwkJ)nuCxY1 z_V8e8L?uMmodLC&P815ehJ`B4xa^KUb8go|CP#55dH<`8?wNk5B)kT9yLq7If1A;( zJOuscA^RBSv+ZAu*n3NjG5?eTYEfSZ@6ib0nn;A!<_37Y={CGQSO^#PHGy+UJ*Z_bdgT9jf3~n@amC zqu@+ZG`Q0{D;M=(KK`?p6f&DgZQ>_Vq9|K|V?S5$Zv=27?2!cl&J)kJ) zGzg^6hP4Yc;l8{XD9_Wy{CFX}EF+CB=3KCO_3UjFYo%gIO z`FVFGS+pbo&%UAAhI$TqUw(?GJKOM=VG~}VchTm}G{cd70}4qOdcqp%GQ?i*16lq_e0=;1J`?|jrcZaFvzR^}|Eh$~)JL(a zw*yc2-N)PO{LygEF6^Co2=hem5$%x*qQ#R(UjA@`6!&KY9iA}x#b>a;bUm6n`r?6% zQq;G$Mv42Hc)W}1@xIsL(hFgnp}QIiTx_78cDBB!RX<7RXW~JNKcJk_3ZnL@uxM%z z)PouP7~2e`;me^(^bJf4d;`mV>B zVX|+5smTf<7HcO|ad(Fd8u19=>xyTToA{Mxe}1B`@dM6Ri&=?3;kJ|r{<-Le@zl3-bk$w+7z*|aPCymqFZ8ZzAlYB0!;LMnxP4Lv4R;6PJ>fs} z-q%VjW&I)MiXn`sNWy{Qak5W55FK)@Fj{E?z8BAfr)MHSWKJ<8+cB(btPwksro&d7 zTVwy_#gsp;hq51r&|Zsb<5>=$LYodcPJ@vuE|3TvI7VPgCv zEF03tw;T9Sk$M(vXN96?Qzk}*x8vOM&#P6`<=?Gf%9DEgjV?js z@?w}ws78V6g_tyZ7~EptLD-8d7~510*>q-5|2YqrZ|lO*J|*lwG7f9s&A@`t8e+RE zjwos;5}*B(xaIbBOdY7j;l^c{FDr{XD6{j7ffotlIzloe4l&l>lEG$N1r~M`5`~yZ z^pCg1weM$wKz0B$&ZfCs(VNVqE%h{1r83R$Bgpo%DNrJz0V%UXN&F8(n5~$`n2cUy z(w@p7&!!uY@kO4T(Ar3-yh3W1ZfA0Ib(p-gc}BNi1RJMcbHYwg!5iJ@Vg70BU`S?-eETa8nW!Cnv@_Eegad#p3_Yt{oXKIR{m$ zIAijc$(8xhgm18p`5Em@US||D6|Yw?GONRJqH_t3-%z7I|8NLd(m;xfwvc1rBUt}^ z>a36f?Vg!t!nqgA#|ZG8;E2*SQZD-h$rpUYT=6(U^t;-by8n_m^5ZlkiFX+r>jX}t z{U%ba6~x@VYsKjG>tn;yE*M##LgA1yol^RE2gxjudQD>CJj zL-L9Zm?Y(DXo?MiO6OCgTT&nIdTfRr-)tdFrxsp^&4!xaiKLT93u~5DK#;E()T^d) z)HXXAFNj;l;SSV+IPX}p{%sqRYn6jW)C02ZumS9vl!LXY%Nb{0S(5n9i+XvdnA|1X zIcI09!ML~}NMzq(uFG05$}e6R3I7r0939T&q+V6$H2u~welHL}?pandtuZl7sQmzQ z+jSA=gNY!s{evA@lX-_TFc!gdm)~J-?-(|WjJx{p^L}IW#VDV*)Tr|}G7CHZW8NF{ z8XqW>AbB5+nZYJx&Lm!8yw_G4J9RDR3>1Flwdr-Vr}7d<6B)XPFt**Zw{4<^)xc-(JM= zGCob#C!ICEIIduvwqf`m*G(z);lEy$&-#*R6n`EJLV{r4nl|KKkcx}UXuo33dED$M zjc;}*!fhjMR^>wuF6riAwZ|rKH47x8lOc@tC@-lk8!#>zDCW#~X+VxgU1tW@mT`3Z zWf(1qrT?Cx>ZKH;FS+F3`;haOviURj!GKc?WZsO1qm~YoS6c;hTGer5f-^?^I*j~{ zf+$rp6I~x`;plSeXIA_PPI}T96>0}EtJ9#ux&VAvG{g4EI#{*m9i)Dbh6|@I!P*U9 zz=v|9Uuh@9i`zk<`05YN;gVoC%++QsDyc_T`z`p2@nhFu4SGquMxJS0tkJt`P?SCh zWp`Ata<>#Lm(qip1?1oS`RaRDaL>qUUVH*B>m1Z%f2x38rAlc2UnzJV zXoj9uH{pUN)eGmGpggBEC_VELicY-&Zi`l!N~hg~Z-ZfY^gi^sY=j@j4nyp=L%QoG?RnY%_GrcB~ten}%d*?1qmicw})a1d5D(7pwsVNl)n9SpzI zUqjx)*B3MK%J6JF<-P?@#`^$oV?7WJ4U~|R!H1`pqvFs{)>nbY-GOvsn4Toao)hygHodg(NpzaJvB3&*^~W-}v$9 z;S|{Eo(TG*)3IqA?bDq200tkZfA{GsvOO}8++N;FM5RS3ds>2h8zRTvN8>O36XVugpBFGINIZBQ5=hr1ujVA{%F*xAX0$(yU;sJ9u&xRt>7 z00DgEB#PfEXJFsUa!8-n2{Cgi4@NN;)Q?qx5?x2@bN`U!FKMK8zZ58b`2(x?rsE?& zRU97q4Erz7z}7t$_$}!*bi44NG-Z%QzY#?W+J(kYIW)Xg07dOPV9D}(AfQEi+2XUn z=1e)5?>!E8iub|;%?N17s30-3&XD%IFG-2G8g_lq!58Z}xMy|%Xm`^bmTVDx>|G2Q zUasIdRsavZXm)U!CY~Q&fa#*KP)sx2uky0stVk=KU;Y;z=sA}WegwR;Xy2fUFeW;c zz=5Fv@N54D-rPlK@wNaXkLO{^-D1$>?t#ap5_tFKL8L|TnBrH1(kgAxQ`rRUyGQUm zg6>aIyJ!y22OSUiz;&SrFu9fjN|c55g=%lA#|q%vJPXX7)Wgq_T6m?5=Gr{#h@xB+ zWm21==IvmVH%ddlo*8W82HIt_)}G}Gn1fug_ISWN2=k7=BN5)bAmsgnbSorNPhAB@ zU8qHExmMUahaY7xDWLDrT(WuM0&{e?2}hqZ8&q$~k@|TO%&}joC{)3Vek}`$NWek} ze-lGgLx!1z^CzJzDv2oEzE3=kZUBiAYw~^0Rc7tNdg3P@!n|~Q#(WsCMzz>v*mM0G zvs5vjjBsxJTi=9d(KRvW7^kD-uJNk~dPkX614nLb1I?%o^2J<^*x#Sc*}g6r?2im zUQ#(zo?%)R5sMS;oFleNI14_Xp`OD#jLh16&iBw&kgj!=l*%YDPj@fHvJ;}%p>YhB zXk_A2i83gm?Dx)zzr=xyA(#ATk}2vdv7SjYWVoGq4$8PIwikRZZ-N(kTyTELbfWEb zj3dZ95B$waNc8k_qO7Egp{=L>t;fcC)RSZX=HIOHo!MVJOz)0Hv9E~l^VzuIU^(P| zeL)HXD~Rh+Yoe;ZjB~+Ekyu{IW8^owa5~p{!T!4Y#Aj79nLPB7oS)-IHVRE6$1l@6 zhh!e%v)js`hV;LG=fk>u&V%{N7``SKBL13^O3xYK;GV~1h^%C6C_{N=)+FgnUmBLnjeDAQVpXWwR)s?JZeSuTLhE8Rnqq7&+GeiJ6hMn8c*=fzHVP z^)-5G_M*|`aeTOLF4TV!fwx;8aH@4%n7m;+=lUMcH266;%J>)z%j79poOoODYjv9Y8-Q!Zxo~-fA?eQ)N1bP-5Nfv-?0ghK=fiHY%z%rj z*xm>J#Y)gGFq=6m)JxJ&=@Cl{UM4s=3bt_1A>*E}h>2Sxqs;W8_ioqKtIwGHPv6QO39w}!b(yk928VH9TMshYS8=bY5cR&iB@VRb8Q)WV zqG})U)L%dvro`}=k~D6W7Q_`O`D%)u zcD{_%re93=0zRHd` zI1_Ec12KhrAI4Saz{1rZNZj-$Vsm~9D{MYvYU>wtH3+~r>a>6A&>p;hIGhz8-plsP z+svkV*E3V+o0;r>A4Vlt!Q|^h0h7v`f7u@4*Z65iCDjMcq7)oudB1FBH`P0`?z4+< zL23r}j6`Aoliy@s(FJmBNCMm@6L99|Ks>(N15dP_!rq$)(EG&}JaH@>i)*aVc#{Ky zvLrUgy@vTp6JWnR3oKv2wq&l+bIR80CX#i6z>;F>hwOifKJs^Q zx$b!gJpYq)+gn3n!9wsnq6f`ODFuQexck+=Ka;qDd)aIW;;(3kCRIvzkzbW49q~fFb~I65^+t~ zG1TIJL8O-aU?jO}iB;WYw$rnQjgKv0_fMaO+TRPIf!-^29cjY|I-gZ~pNk4>hS5Q@ z14TS4Q1CX@kHje_-(oqQYzSd7b3c1LdJTJ5YZ0vZwHPmd-k)>Iix5Or9m1fG*B8eXdVnnW*I{=RZ>Xid(NRkLP#o9 z(mZRRIZeOge!lO|`mOc->wDMQ>z;e=p7*oo=Xt$8V%PW0WN>>W`I$XKdTx>-S7g4x z6L&ntf9l!=al0)AwH`@CphOW zJPuFGOX5Y>6C!uLDpBmpqM52c1+jWzJg)Ovl&V;bZoN%M{s%pX$;V@%KIj6!2iFli zfAN|4v=gD)2r;zupA8s!5LT<6tw8xv4OXhkv4Zo?rJ}vwz*KFkE3*eJFFi1X`c_5! zp1;SlyJd-|omCRZ@3<)1?@-#k~-) zt1yxG+J{&@X91cT5RUBmb@BVUI@HHwiT&!92dOzn2^rmRfpT3kVGanTJRc&ITT@`U z%2X8UdYnM-)}S+0*9BMHT!}`jy+o#+7m?chK@b=64=t8{ZMihV!ZK0y0-E7!LQrWr zf~`ldqS{0YqWX3b(JK)ySn^5$n`22Rvx`L!PF4_6)Bh6M+O{}vbO~`iU6+vUe@*n! z=a5S9X(G7Q9*NTr(Cf<~==>A~&?q+-Xy?ZYj<6wu)g}FC+s30P_q-ga)6)gJku1u3 z6KPp;;|_7(tANNUPbXyBGvQZ$3kZq>VRlOm(MN0{E|z!@iYJnZiu@_)z(o(@LP?^a zH9#mh@LNgn?XxAyomxP=YrPCtN136GA=42|4kcDBOh#GbgHhmzX9TUQAqbH6v&xe5 zCYlq@5TRvrP*Al2A)7D;g^&DB%&!n4dBJLd+ao=qe_PR=!pTr`7e@aylo#5>YI+RL6qfMV|yZ2*2*e#;bHW^)ev3s`3h1_$;d8w96Cv* zq8?t~Z^-W{Y*T80d&!~1_2lz{Q@0|B=6GlHNq0H1{{1rOmE?$;u`Pt)(N^NM!WPu< zY8GVcBqPa__X)Yly@JH7iy%H{m%upoi~vvRL`Uo-1!ni`iRR`t0+j_H1e<5hVn)h*V5 zvd4!A-oTi+*u4@(e~luTx-&%1HBUlrK_HsFHx9)m=);0K#NJcAyCa;#>w z3B8euLbTOxsA^Dz&V4(ecf?2V6`PHoG>t%vw=x(U{zYU~EJAI&ra;ZCaKdQpdtRr! zA3m#?V$V|(u;8OT?keKWa;*e2`WVQxs^NF5WAR{>G{ifL!yf%m;4S50%}@bv zqt5dB3!{)kd@QlF)dE!=Ek$wnY#>bhDwOX?2Gi^wI9;iL-2?UzyL=8K_D4QU`CCYw zd3cAgs=JQ5uj~ci?a|O3*apA+c`n0(Gsr#ZD9;%~A>`;73;3p}qVSqRsSo2YdLuH=ac?yiRJD zO)2UzI*knJ!{BpI5g+>Jh^>;Y;?~-^JRZ#?uxM0;>N!u5ncg=vN!0;v`t2cF&q@+z zS!W3Ex#g(sc|7=^OT?R^tRclf5eyRFp-E{W=4Z`*GwLx zvwrzt!DHzWR_C$r<7UFyBUUiB(Ug$ep$SJ474T#4N#vK5Js`vDq1{r>!M;LkY><5v zx1`qK)mjAFFzpE1E@ucHReo^v`))9hRmF=pHKUKM$B8#|+@q-uy>zX4gDuOnE`T9h|wCvtz54H>-My&HYF+v%)be)GU0j4;opat7Kd5HGjaKu@9nj~#125!dz%klPVrw>X%$m`ok z%vcRp-5N`1PoIvQRixl>m>nw7i9$B}`F-k{Uns?X6du3!0^a`K5cVBzhN#R5_&5C! z?PNwGu_X&p*y?vk_26=d?Cpi_j>kw^E)YK6kjG0#Wul$sKhOrNgLuhKO|07h6a+7v1M@u3&`q5O zZYm})abpsgIt`%OoCOeF^BRhx-hm;z7*CB?AS+6op|9vRc$qn3$@e_h+!227#$#&t zha`ZW$^&Q;2=E-qTEcYRIW(@h1srM~6XKt*6I@>$VauN(Kfx++kkSN7GXk>?0z4h* zS^rxb*Z5;ZvisyUWVF8-4Yk)O4hHECz z!!0xHaCD+M?!6(4ufCVZciMUWyMvjq%;66d%Sn@+V}Ia;!S{HT>S+8#+!)(+SYiut zS@Mac80o^FKY#Fh0NMMJU{jtMSbJMS<;6PWu;>taXnhMgg(e`UlKE)f=97?4--h;0 z@_2O6Lp)or8h5Ub#HS65p#fKbuQ-2aJGl!A?-sx=>3mo{bpq(C@px>iC*kwEkD#`q z0fwGe!Q536pgn#Fxz5*y8SipHtnwB(q)TBd9>VJOl6%C96;dcLARD$`NdpEK!`NF9 z_{Y8kthllmUzDE-F1ohxlQ{+B2k$~xbXkr=w2Yu$n+US9uCf%8#LZT|d+w&P1&*15mn?7|aSb$D!XA;JB3$_{+h~kQ6o_ zzNYe+DwkYw#WoL|cVQncO<#`ZY*E8`isE?N?}zxA+eHj3GVup@Z_=XLl%z68km8m5 zL~GBvZ@I7yAOAa%oL4!T9G{{}{`+i--Uvn^Irn?SMOv0} zKm3_=xh_j(-8>BQ=h(yNyYW1;nF-k=J%-e^l_m>(CV?K0O}WfE0p|U^4Ue2&K&T1E z_~%0~Ep3Hqr4Qlj=#x0tZVskAP4M_HcOlcMmba_W!DD%xlfj-QaJis~eVw}?O05yf z1<#>T<_4PbX)=UaWx=^yV{y=cB7UuhaKMX1969$0R$qM_&)?VuJ_#Qne`qSsUBp3= z{zhyd&Enn02{5uz8xJWT!;{6c;JkY-yfJQs85(W?)3!lDV=6=zO+;Q+o=7k71}ZW9 z3wf^=;E*4&cy#D-P~&-T%iRp|Xsse3VpHMXi+nh4G7*1ER>mhUj>fZH62W*!J*+D- z!I@_j$ggh)am(sIc*$ZOXC*`d$A-?rw~`tBV{Is&r@`Mz?B9TeF$?j)JOjLu*Xcc6 zrH@z6w!y!$_F=yn+j>jRlz~ z`)oAY{qzj_xBMs!%-IU6^ACW8n-l(gX#y_d^BL7}2K(~+|H?)dkF08kj=>s8{`nG2 zWMg6W^<=p6>?&{(li@_W6ioE^fK1kG#^DETu(UsKbD$)IA^AqwKaf60mWk8vbJ_g&ng@FyvpupJc9Jf#f~>D!BojQN4ox*0rNA3uYsw zX-3F(+M__?k(0!ki8` z==c}(yfv_HeGNqM*dh^%IgqpSIJ(rb2@>aSuWx?y5Z{L50pU+!) z{ACX3tSQ6?qigXqIWru{V=MBrIewI`0SBWsp;g-w%qTT@>}CYCTofD(495~L95FfF z3|}gV#g_{L@TAvAu}QQ$DH&!#{!5!cO2|AFWNqCeIC0^F!0QtN+m{CL4EXhJeFM)^ z@f0bJv4NBr4<1h_0{f0Si^pXDK?@u2pg6~RwBUIRUVmpTX7DyV>gju!KTCq=qtV6F zz5w1fRtvj5m&HeX%8`U+JGv7t1}uMeB9kUWneknycEu-r_p}rl)S^k&&pC{H?(V>W z6IS7qgY|f#0gIn>v|;J(`e<|PTVn2VP1JC|g}m_ZIvIWP5xHp53XFzU;%Bq>;Tt*e z_|3i@*mp%VUhF#^U*owNvX(BwGX_<`Qg1Q*TX`7*vL>OPsamLP=s7YuzW`_3m}2Ic zIR4wJj?;cl!0I|qIMAmZULS0L76oZ+e5Vgque%0EdHtXFXD`q^^dah}qzSJnwosJ_8XAUwN!yU2J}R zJ7k-0gP6NK-lfhYNb#5jP4#vlIrbF#c0L2W8jyx4-VS7SfjR!!?2SE#mP4&iFetsr zgRMJE5dF~_-8o)`?#b)p&eNl?xS0f2>Q3Wz97^GEy&RSd9YLlS|HXK`0V!cRhF8k8 zz^feHvE1x9+}peg%kXEI%Ev}x)#?Grob&}uPYU3w#wch$xChG3k7B30^KtHX2kbhV zfyZ4}AlzjZ6xVUEyZk~6L$90(i zQ`txAV?E>P!M?}Xz#^Ej6@4FkX08G=G#@uy+Yb>^D`4l9P;d>@AR0U}d0h8Yw9~K_ z3vTXc=`K+ zt8pIIYl^@ZO2*;lupDgHlZ$`sr*OW-N=$Uv;Z;+w0^)TJE69HE{P>OEpB_bj?(jCx zCBERe>;V~n`a0RGe4ecNFbbP(cgJZ5T=Cd@*KnbR8hOusF6psK8+>y|iOvG|RAfPe zA+JloJ%LI!Tl^>JFh(*7xZ>n#e8#&8`@i|j?}hnu@H2IIbMqupZqF`KVPLLkcGI_Tl&E#p2y7{~$Ww5#~=`49B*A!|iVi@$JWm$CAlK`#L3H zXT(BKTGNQ*tu08G*Ip!($m0@|b;PiMMWAfdg&B$moyRBtMvw{r1YlkLRP%Ja-+W zp*xAXc+QB@`YlUYFI6HRMNcJPu2Lfxx33^~Og~H(4xA)E%TB~N=O}iq=dtEERqS7= zj=g`if)o1#7A>>JVyD9K$zpA23qA%4yZQToQW>(rSd*;Y^$WwTl^Clp#HNW_*#7f$ zDF3O?<8MfSG1UORDHEW}?+B_h>jbG=e&E?zg(`xT$rn{lr0p_a^0XX-2Mqgp4ytVY zMeuu-`#vp7Q`vz7?za*Z2&qXL$4Cx%nCel+K;a$l~B(Y&JnrJbAaz5$8 zz2;?bVTTwVP}K%E*LtG=_)FNl`vbtnC-9@g6Lz#kgYC*Qyq%65?3N3Fbpl=7y+<9t z+VTP}t=|eZAIeF$eZNV02{r2d;tO~qn~q1%+<~oJ*5S7i**NU?9qe)U5p?EB;y-F( zxOKM~_GsFV-*sf;@eMuv*^dxQH~5k50SvDnya(>54WJDx_LHsKH<1=AzT@@Z($E1b zDdasvm3Ziqi*jasAP)6P63>GRA?evy2yF|&FArRSl3Z0BYO@#*ig#lLb|M+^*PA@} z@I1L}sDoTI?g5!~MvqkbXG8i1xsvfx*LXV)1<34Q2oDQK;44AAU9JBpkmb3`7(WAC zaY7uM9*jfDkL!@ui8gfErWhv8xd;^(!(qXRF>v5qI*56RVZmD!QYT{yS+2Q&Eb@K? zZB0t}pWveiLe$ptF;JDXVrN#H1To5%Y%!IeDr{wdy`C-`>; zuF=SXUtZOq_(}>d_i!Y0C^1EBubixz3 zP#}(rZkgdXN1uWELm4a^ppEMsLr^X|NIV^Bh)ieq!zn3AyiDmI^xs(ro=bD#mk$d= z_uhhkiauWBw+LIzj6qX!zwqa=laT$JK+)Ls?WPTyYMqInHFaPuT7fh-YJ|=H&md?C z&t*UIgsme zfRQ0}=*!km==|01sAb(1tmFEV*YD6Fzs}At|Sk;BU@67;X z2cD0q^9@K7G=9%xpp8ur##7cGgXcVE_~vU}Ai04*i^wsAFTHv^e(*coBCbU$2P`CQ zqX@i*=Y`w@>#*LpiDXG2MHaO=k!#0D3&J0qC9>=v6VdAL$s0qO)E>*3)cX4qablo7 z_AOe5zkHZNW|+?*mrA&k3wK@yC)p0bpPs|t0z2F|HWUkaJELX%oQJ%E@%P#jctnOj z__$=iDW1QF^LN9_wLv)e*Ae_{FNOE2`{Jb|g0Ra@4R9a{a53kx{p=S)$GS+6(k_5I z?#aB|*F)%R4P59XLl*8HM;aZQOl}X^j6Hi&aoOJ}?9{dwyJVN+tw)|>Cw}Yq!y^C} zr1|4rW5vYO9&bb%*~Sp93$Vz1HuSY`g16T$qwBx_qKOm5;5U4Qx#MiHbw>{7vc<6G z51J3}ixXnky9H97T%#$soMxDVPe8A&)JXMn;C|k@jb&Afw6WktdH!9c_97wOxIMrq8v2pN}G8 zr_>I9?dSntHTZj8hPR7dxe{M^Lg5rWGd#}Q3_r~h;=-O5ymR0U)|V~9GdEns&+Xc< zl;br*K|%uUl|M_U{bV8Wb{#k*?ZA&EXJM~bd-0kRXGHsYDeoFlf4;G80a{UDfL<7T zpcT5^JmtCuniC;`UZ$2IAd1i|M-9j@UXKzAccY|JYhaOY0x@2~ga`;xMr#%uBVxTi zipG6Ny4M9Qc3efc+#E|R^{E$3=-V$yiJM7m*lr^DV>(Zeurn8#KG(wY?&{#@Hj{b0 zuuy`(av`Rb^bmRx*N|sbI)ZuX5M33CgcH{zvy8J>==wTOBIPL)= zhxA2d8n&;@aMl%-XH9cAd|I21{#;5y(+}D}%&(Ey`H&RO>Ku&)^<_kY%>m+o%6;O4 z!DC|bqA$eEhsN-CYY6f9@@Qh|%?HG4u`S>gKL)NVNQvUBii#s@KP`pbiyetRy>-Mq zA8(}g`2%soVHu%SYe(dkrb;K=klCWIE&M^7hcq7%E%qr6QE z(CRJok?*$y&~bqw&i$H*^f!+c^__<;=IHV}8GQA{SoG-MC=_~o0+RV6L7bY`E{M;{ zBfdC}BEssX5v9ujHI5sDE?%05vgTJpxVq_f{^x? z=_p|G7WCC<7?KSF?+aB0zJ!4dfawC!6dTC(68y0YN}95|H&J*$#H%g+)tA7-NhE(S%9oP%P_ zWl-tdIViSA1A_Di(DhDFFuswDTn~C6>FiS|vQHHa?H!Bq)T_}QUvbzm{XLqhG#)No zup~6^n-SXbv8Ysgm*9{ZD>!X1QLxax7X4P*kA|2`q+M)>x7IJj7VUuj-rK=U|869| z`VX4+%mMj+8IQVG=^=??ckmxI8-BYhLEu^jt#Uh!a;B|9WU@JKjF^TEZzy8LyAjAZ ztqDylmjlxb9@C=W7-;%0hn3xF_}Z_7xMDkhHkMxj)#2CRpTRTOl|6uOj=hD8HeJOx zdMCky$SFW&Vc4TuC@{0}6v?acKoqYpcp8ZXSRqw@uY9dt0!u$F1#!s^U^>+B(jPL| zPWA{yr`f=x_Ql}1*cM4>@)-99H;}g4Ox*Wr6h7zt861j_3B(?4N3yTyqAm8b0Iyeq zJi%2I8ZrQ~S9rc=@kJn8J5unxBh#uk?WtAv-Ba*pN*G*kUIDji1bE5HNmxaZ$A=6* z4kw5R;1#|hR^mHgp5wPPgawy|BxvI?SVN1G;lTJpf#kIk!ejqv^lPGk>b^LZn%w`E zJZdYAlRVyn=g4|E_P`UR1uLN|7Y?H-qp#ubvs3VDqeFP_mci|_;Vpk@t^x)3Lix9@gGph~FkCSXF7bA|2iy~h~gzP`T% zy6I8y?x`|d;;}{)DTLWm{v&R4v&F9 z=jr1g&eB-!)dTo>!3R!v9)|#}8!)Rw2ejAN!>^gE!JL)Ad-7dimrXf3uCWXDT|G(m=eBEN)7Ise6PI4k`jT2tSVY?CXVPBuL+B@@4?#TMgrZPLqywOSx~v~y!6oTjtC$Oe0?FJw!L`RS=%cnG)YKhBU+m&RAxd8~#}x~f5Jld*QT?yQf+gQd ziK&rq(ZoaH#M1x|u#TFK<#%=nZWkKEV5AeSs@5f*;_hBy*?up02$6O71cw#av<8r>u|>`ib4`%G-Pe!^$*WT zTD>TYB+je#i0UCpFBBz7?11_749lCD69t!^#ERbYSe@wGs#++FdQM23zDj5*z7@r} z+ag;a&+8@4QW-6n^t?n6>LrF^T*p~0>-r;_^SW}5AYIijyxl2>SC2SGd|FKsFM>y* zAAh3-Xxt^j_65K8mX5H>{MF>AVg~s&fAmqTc`{Pu z|I9R1*i;dQ$?M(dqm=?Q&$kKdLu;RPIDkx z>y9s!#|ZWYA3)+-QKC2%8;=pca^ghKMPHL)LQFi4DcJ$b59Q*FO$MY3e-`-H{UFxK z5hsm2fd~uKLl=1-d7)rD@mVYyjoY*yI=^oroGdWf@KF`E{L+EiK0DqU2qDaO%np*^?Bz(~7gmOe#yPzempU|s^(*;S6r9^zzQkJ}*G(sk*XW{omM76z_hq6s z?mR$?=6;E=NrLtud7^$>hTzqMULr&Hl&F4MF7Y;^S#NQGyao~UU(f#PsoZOa#C*1|S-)Y|I*YYFfj$er&`;Rd@{JZ4z6Wmqs zS$HvZCpX4zANx)D0qZ5aO&0h5WiHN{#-@GiI9E0((rkRx ze!RKi1SJ!=%sx2|oJ!w0wm>(7)s{}Ay*JBKKgW-tZaHUD*>Rq9>g1cuwdo=B z(bp%a&!$V*#u1B|sxEa{Bk`)}PV3G}<8D(p9&_e3H0Qq^0Y`T1PjMb0?;2zS?kqyCkP(WTlw?A`a~ zl-0{%>b>C{CRM+jJGHor``Tv9ZGXC(PG5b1UKJ2PFVnAM_O2Xc>9M8cGOH@4Y(W+I z`P62ve47bJ`1x@u$#I-{%XIFfx+S-(b`p2C_AO^S>jt+gZ36f9cp)|Y_;v2j4l@qP z?c@$yeq(box3OQwA7>6ZD^Tan6xjDolH8$(5$rwZ@!WaC=T$z>$I(Y(qq+2z>1_4| ze|Bz%c#Vg185562^2E93%%jEnHSsPvoV>>s&S|1N9jq%$r%Fy>FJ2F#4`}Mq_>(35 zXP+?_DPzVsMtbA>yTUn{C7PUD_yBX;X$7+{?duTm7)IL)5785wylM9aF)r}7 z1n2+qJDaw?h>a?aW+y$Cr<2J=H5y+p)YyLBR5N<-eO6f5!*VNTvXAbovD*!na?ksx zav#mqsuSq1)i3iNS8sSZpZs*jpKMyLNgE8dkS@{zlmoB**z>KM92`F$$E;c~EQi5* z+A`FgfGfh8m3Ns#MhkeEox=shK4()kwlcGlYp5nA9pQvyp``4c!_=_c4u=_bg)8i+ zoBZ9frExZy8061f`f>&znXky+4Vgj3q%||-rX*4P!(r;CZovPABM06vUj81;$LIoF zPcfwW5{?wC^dl?F)VMsJzjy6Ly6Pt<&HSD$PwI@A#(MvC6s4=^7sLFYIEM2ea*?U2 zutWP!m3BfIIZv;f-234>^Q~hU?2eOW7j3tvq&(*dhy7l=(#GsMaFcRi%L>xbgE~6 zJqzVke=%pE&W$c>W-q!fWk+mI5*nHIF_H@dhU-PF?aM_bO*NitE;Z(MT|LTtX`&%_Lgc=cBN4YGLtr`A3?3^FSGQT3V zM0p?ny@wxbeC|$RhA#e3osOI~l09sk#kPzd$1V$H*%HT0c3f(E_0t)*ad5RW;~#t% z*I5JQI8vXIj+0^!&YzD9RqF6;atn99+KjXOJiz@+G8g6Zzch!xeWv(|+N^W^OnBLN z6lMLkVRna@usxXKESo~BhW#vEufsI`Hm*{a53dUOV#ho3NZn99sek0%a_uN5X*TQRGPm$Rh~FDu6d8H1i6A5XAx5Pola!g@Vu zVx1m@!J-`pMRnp+@vds_-fz_pM{nk~tgWmvu3g8veKX-&*4kCQJvX*Gr$>pckq%;n z$ybNlBUobB7!cn)iyg#fT+>6%3I}93_4E4LqVQM!!d=Cyg=XW#xpT|T3l+@YRV^J* zV}<8NR3si!LXNL=D>r^3D>q1NFN-)7iI(ir7Paqzf0o>ESsPs>gi8d9T(5n*Q02*U zc59vi*%NZPdOtT=7_zsg?5np0%UH^=sqs^}_|Kz+N?E6b$F+p)ETdy08Et=~C;EQ; zY!lIIyrKMa-f6-%SQ*(XDsgYtKVh~my32UklrT4oQI^-T?_Wey{bPUwEClQybpCsL{vi|9d(c9#gORd5f^ne}EmO z{ETfc&SLa!pE9Rh1K8#JuT>r^84ps8nU!~ryR#Bq=a^ZM=4@M9P*p^dD=N`AiUYih z1dqgDlsW6nD@(dIg;kEw;9{SYit7B{Mxy$Ol2PgH&`hGh>s&cp{A^ib{0!!z_pFom zI#E5`Y5&e{+tS6BN|Wrl!E{#Zk|USW7LSxq?-R-_EM?ZXe-qm7N)tXjS|sdLQ4yBw z-V=p$FfL7i-&Tv#u&?JRkw8V~sl;@j_pLaO#5 zH781y)?ZsfMdi$4N=pQEPdKQFPS;VPLtm(+ z$FyldGRx+UD&*2ewlj_+rm(#yA2GK+j$t?b`$GO*lr-G;g8y#lpkiSnHLiF8=@y)EGPJzxOJK zy{Kv9bIJZ?3RFnt4N>>@yY4+hn5BkQ3X|$zn20u$q6{Xud zZ3^pAxrlbx4x!UK2IyTEF3>*fHql-N9rP{6=+vt7#cG9*Jxt)_yyoiK;(O4sLh z?tW1n8Yat{ST8194J%lS$RDEo$Ejm#&Za`zJ20D8u3pCYkK9VTjbq7bRaMT^OM)Iw z`_=ens{7JP>(84!tu4Hatd|%ww6ghT;f|R{*zm>@sy{EA`u1l6cj2cOeN-lodg+=) zzOM+OzvNw`FJ?`lcmKAszODL#e$`S*d-={5l_yYkYITRH8g1JXLXRK)hWZ{=LVZ+= zqRy!5(v@%5v6c(Ii|Woi(+Im7NP}X44Y@DRivCnlL-xj;XR>~T(Am3Xx#AN`Xot`@ zbjgQqdeM(4I)37P>gt`T^yQOLH2tnjBm*s-MwD&RWV86|KkKU@O|-js23_0R zM%&EzPPv$*P?4Eoeh-Qp)&;Ry;yyH8DpT(-j-51Rd!_P`vwv)adjX2b_ zA4SeC$Ao>QC~qTYyb#679om53YE|Ii$)ngWfwr7Y{5N{d1&YpJpCQWQ@)=#_!)f$Pg5`UX8K8%^7P=opIgi+Z7HBht;L+#2?u)97lYw42M_0Y)aU5o zG{uJ9i#5N=11%xkL+g5eEz!#zlR3j3i=4%JydL7z)0}GjFPPPY>6Qqy&*)R{mt>QJ z0$b7i>ou>xT5^9ax2F0#Hx4OtTMbv!^X@LA=2e!lBYg^~$#+vIkBUiD`i@G{U@yn6 z=o-&%kxSuBZBKL0ruwqR=CU=v2i0nxu9{S{woIQs@QK%17?Hrlo^GS!L&R&uidS+w zttD!<*WTk=l>)iC3o6u$^=BBY*G^jH-ex0)`nVPHcetLPgKSX-#feuf;f}f3b9&~^ zRDR<*?)#C6^vD&_ltgzqxAwh&j`!M6e}1UWTAi(?_Pn0X*3`Tejq}5GAiI}lVsCC| z$5`2NXX{->VfW8{Ub*7ORVJ+`o($=qB5JGV@=4U6YAYt?bz9Zq6a{v~*wJMC)XU7z z`RQDGQzX|b{*b%7Dv1t!Z%eN}r_HF@#*uIGtC0(C8-&M-UtJHweS&}TztB_lV05&PDg6C(D|*E)RPU< zxwL29TwdJ-G8J@&WidFbloE}l+AjIzT(dYfYUT;bI^q|#EyD-P6%300%a`+{i!;yB zYOD3B)P*0YS++jH>I2%WW9?2h^=?1YH26fgv|$xJtI~#c&$>Yw_J^{~-|IwSZ)%Mr z!R1kv!p1W=tfYe}bO;osJ3XEzA9%kO#%O;KzJ6K8N-Px38UI_>;qs0yI7+W0DSDYf zv#@wx4y)3iMSpB~OrM$6A$+0oN7R?@w60=x)Hl(Yqg?1I=aw^Hl^&6KhwB;B8Hc&& zq1o(%tEJq)&0KQq6+R zb3OK_IoVa$>BGVkbmW?QjM&^GoOn&laJvT&=ga?;zm->V8*P1B+4}XBYpl#VA5y>R z5mh$cjl0=7+Iq}DG_8DjA-$2DM-^)7kbz61*>Ou(!LbCH>bW^x5dM1%edBgNn-#N$ z3!9@!>3*M0FZdo!Z&?BChrLfon}$?!g2G+u!S7>~Ly8WnSayf4zi!B#aWCdlB05;h z*#(@Rdl7f??5ErvRBZ1AE<3=1wH%wzO8ykk2~PU-5^g0sG$o9_ zyPTuBI$zp2YzM9X(S|9+e@V^sD537cd{RAe1l#{Rm{Y#avM9TO+*}+TLZr$ISYLlZ>^Qn;R$m+r}Q&lNs(~ z!L39%!;_H>k1@e(@AOrx-ZvA*IGJ(&`ReTc1ww9vUOStoIFehYro;W3^A)_FOlDgi z6tgDHS?p5j*KB{o9Cqi~Hg0a|8BSsLA@0(jja*Or5$<(l4r`yXhuLi6z-^S<%~ehA zWW#zUQp-n<=KFC!`6*6`t)6bgp1Ie?`M0KWzXsg6(eL>BOgKT+Up`0HyzU+z`-kNw zxO>h|lG8pTERz|*F8e--Ew#5`j0di>jXpU{#$_9(@!@LG*mBGz87+8VNZMY~=c-0% za0lM@vo$Hb_*>FNR>Pg2$Nqm~f+9Sp>@}xQBjhJA%bVobX4QM7{nIGYahe>r?pr?l z{{9iJTTzPr^(UB>*QgeSz4YZa`*T|+srx60;T9gErma24+Af!=$yryyWxXq5v#U+G zFK<1$SC)Kp%PJ`$~3y7l|X~A@8=2AK^=q~d- zMvHx#C@#v|fedlRW&e88>xdG2D`h@8XN-W=xbTdfvGyNhw)Q0Zw)h5@l-9ttzrN3o z@(bsdj=#k^%4e}rv<7wP{!wbunr?1u%(9x`gfTTXIJhR%D;J-7ZAl%`MO*YWjtF?4i*4T=uOZZj`+bw|liTr?pj)b)NWy>epCWv-!a6norwQ zYp!0u$l2uG;8yD8aelKc@Z5La;QDR_vxePH1#UbqRP7Y2LGvvb;m}cLZIBMLy`_os z{yo5*Q%>TX@~T;{Uw5(e=ntYf{=fEvFco?=l|dbA4WaX&r%)wz6132^zcNwRh@QD& zGR4lJs{49@^BWaPSr(tC!d4g6#Qv(Rk-Ae`BWTHCA1CXv)tmd6ZWDddEdMQD=l7mD z^wS4_K9(t3r}$_1)0-w*)6sVt=(w%XR9)y>YIo_0;l3XnSlmR;JXuS6L(OJYhO^8qaRWerpQn`??ec9Oz$T>Cd-j4y@SaUxlN3&t^;cllE8}DzM(4T zq_LCDjJT_wy4-J_ecXf*${cxW8aEi4!z?SUV-zmWpbJI^F}vREChfPI4!3=9{U{w~ zQ`jiVyVi^`JN`j9H93l&rKU;C+2v8kDsx47`>%dV3zWH0ilaE?PrBTu{bxk|d^nC_ z_a${1DV6aCn3<1ZZf~#9_0dM5{Ii^#%Ug9L%xpD%1)54p+Dji09ovEM;Y>%*aP9C5(hnR3T2qjKW z>KK=Ou$&bWk0fc|3uJeb^KhRQtC-iV2*JA0%3F2y7UvJv$O+poEbOjOw2Zn|`M`BG-NcpS}fMf3)bO|12pc>H*) z5vdjjWPgq|)4sWava|@Nve$j2HePn7XBfM%nJY5bB#Rx~Z6v?;bnz68QUA4r z{g@QWp{%kczT^8vd=Pof^?>NPY7L4Eul4b1k%&GJn0Q%?$fS`Wa-q`8~`O9A*F9=@Tl}`wCxNu&hqB%BBbI{iE~-6{%-i zs;KhkPR!ZvUe?O#gJ>Lj7cjs2o8Dvc+Sgg6a=97RG(U>7{SwJ85UgZM7VOhpE!IsrjnbNQpWL-}40VscpREY`%oNS+U|+lmrgrKsWFB{W z2-Es|NSoqy^xC7_=pd<6BuZY!&dD33R@6M9Un;3n`|UpBC}G}Le0_9+z}o|-osjhDJ~0VKE3=X%-TQ1N_gJo zB+g#qhFk{O8drI2dGi-z;j)}v{%!P?m8!~; z8=V8`OIA*FTcs8q^F@Ir`@Pt9zp-4LmOgi|R+E&*y{yv~L)OnVnq4wDpWb=kDSKq# z2J02$%k6cGqM>^~{c~?6%|og(jrYvRcVllcqfQMnpCsGaJDcQaxj(kTdDU-8|IX=j z)+r@=o9+intj%61ng5fyUr|vRI`u&>-4fbB zn^&G19s|XWAMIuj?0Si{s?ul~pD58gGqg@YBsXGeQG#iwe~H%68yUA{-o@(O2ig}&|&D?OteCQa{*1Aeq-ng7?+j&zYKWX~e%>AI*tk?Tx zjF~}7MPF>KXdjkiZ$DeOQjgJV65&tpGoc7_;AMS}DRkZ< z7A)CG&fM}Po9-3^SUMB`lFTb8+8&38NIxFC#rwP18BRj>O`S5YUxS>-5?#fy?1!uz zbl1HSo!xGWY!7y_2~Kvnz2Oi$z;!Eo;c1YAXip)Jg_Pd8+&cKTGZISv%Ou|1jmrcl zNjGi4i;rFvM85EmVl0viJixl=ER^ykns#JAFmxYXJ?_+NhwTr|u;CEIYW)?FVe2!Y z+3f(>XQCvfMI)+PFE*2OwEIq-CRwKTLF=+guO zzZ?ce`by+U%|NX6d;!~DCh1qZSV8O29MQv_*vm|gWNfw|agk~Gfnz;Zsno^ad;g8m zwLi{;9|!{{=BiOot()wEbIPEl+XTFjkHX5$A9`2f32 zzrgobx55db+i=y0JaodQ4AmJwg!6v#V_-5LqExW+=bjyL)VT;J?u>yZ&t}3KJ{`cd)DN94U4>FD z=OER??c$)nfX}=A+E227;AA@Xf;YOkDI85SJPsT0<)f81-N}%7t+3+k0N#gQFLPiu z;_^s+dl{Bp;6_{;Zjr|(^5|WG39nN=Yq|v`rxr?OAfq;81S;M>i_bS}`DZ+*Z5iYI zZ4Tpj_22x6^c{C1Pc@gO5NF$6NNKGKqiFV%&ntZFxNqnbxM{3ByE(j^QH(jt+%~Y{jEC8CGmgYcHq`k{ zz86;0Rf7-FqS+jCUlaoLCTqbtv%{g>ax1VAm4eR9N$B#OZRk$cADHMk6E%Ckggrrq z$Y$|4lwXpM2K@3sIve%j#Tg&qS(7^6-(_Bl1Sj?mgAZ4lq0iUT;nA6Y;IS@_)XFnN z|6(x|ZJ7){JeHSPH%QR!o5c>L&wu)aJj@jejdpM z>M2X1^US00vs?+NTwsZYxzC_SavMp`*YR-J4u8qn6&EX}gbx%9l3zxDgeXwUrMscC znHi0Hokx8y#XfWoQI{Zq2JiUx zs?aHn`~G+}*D!1`rz?IyuGgE;pPP|j`{z-_)iefwbg6@zHSM6@J9YH*kv5u=6A!c2 zP9s?#Y-r4@}i3 z!`sQzjJ42jjC4QTecU%1&&6(>#m&;sBcXlo$--?W^z!{wsJ*fTt<`)*cRLp1eMv&( z=XsvGj!;CN5r`ZlncP-L%3*gS*l+jaYYdU{ljJFr}svOjvb-7R(-uoIeeN zm3CtMC@351T>3f(BmC#%=b6r}r!Xv3Fm>cjPjYK^_V@-LE>Xn0$0Pf5&d{OliK34(}9e)>mS|+7rp8) zM(2k!fBXi4NoBdPGF=``Z@ou~uJz$!{~W#?_lSsqNIdP^_pSU+zX&Wbs#-eBnG954t(Ch z8;^_4OJjJu%F?-HlvICKLUWX)y4c`hPr}NWToq3sQ*nXAY|_ z{mIOhy~keKRwRfrYlAuIn%9eumFLQp@~PTF>eq`Z2at75-T5kT4C0X&Y< zdw*X2B15+{fjqA!(4VAW4SHIzY~*XfGt>8EvSTV<8qfz@wZF6XXJ?3?AOK{u*cy5 z%y;i5QhJqqufjsTTRcvv)8n-djWf{_y4Pe&?@M_Vtr3Uw??|6F*WBc>c^mZ# z{(b##emp&UHfmbx3mrU8K#y4tuq<~DYuRuO7%mCnX)p`h48BY@gkjfKu)og8iLI1& zqi#!C^nQ{aO6dO+O&k!(*k4OvcV8IH{?f@-WirlmS{a55r4+Q&GRs#lWzB036}98(xOWVtQ?@!z&{-*u7m9PmB7)R18u_ zHq#%G&s`%g zfvssea7|PX-2Ni)KlxksIEi;24-|jeo(5&#ZUFly+ThVykN9%Kvv)&rUmA*1Y!ctd z^sAJXn;BY+_kMZ7Udr}`2K)bg&j#;^!~x&u;o;x2{=+xwNgaE}U;*RgydGFtCqTD+ zPc$>Y2@YL+2gc1wh3lCx_)(mIKZi7vK@t7P-cP4+Tck2M-**pW4Vp_v&JH9N=_6p# zV}x(zWWg=*~{v62vx(A+Wo#X4SebY*8&>dWB6$Lwt zGeCsfK-Ay%-&}Uk#V(v?kHM6cQ{V`TFkYs>Gzlvx=t1tb?0AA0?hl>V?h{kR{ zZbP&><1!w)VghKGAKuQnqXWz7((&=efFpNs0{ed^R0wm-c|Yo*`uQH+j0Y?ln6k zb~ z?u=axZWqZ)*7kCk_k6wac;v8)nOrxsB%Jp7$xKo`@t#M%mqoO>6WZ z7oP-^<3<*cd9ev>+U5uq)*Qe~W}RWo?q!MlUvtBM#sr8j@7&MNdr0&BFLiEsiY`BX zipsnWBWmqL)_oqym)-x@Yw)164~*P=gBhrqihaZW>8mKLAM$aGLj8yRL_vBC$aOR! zpKltFg82w%2CI;cA9+%oMW$94CHoZ!$lUcBJnm5kX5dLu;csBOyrwPY;isx z-iSW{_v&XOm+^a%Sk4cXYwM6L>)(LI@%!P*k4sRN@>SUCb{aK~Qzpmv%aI3l{ZUc+ zLK5Wd0|xbI(q=ARVAq_*`=o|G7jh>1D5L!C9ueHl!RF^J@Uqw|z;MJeLB{cIjNHRU z!I)4pq9}-FwcBOjhg3O1ZSX$uUN?>nJsQLF;T8SQj;mx{aCfm9y!^L<)V}1%-{qxX zsZ$U>KjJz6p6jN6KE*xaERUbP)ALU{npXO#Su;L2xWVv1TH8gQU0T#56vYR?^ zA|4auL=@de(_#7Z$&1<1^z_YOda7^{4fw1?RlF6r3Q<0Nb1<4Xdc+YoWkvWf?i$(l zs}X!%9Rmc5hKYTm7UK8Y4Vc&;Kj6WE+0ba6GwzGC#y@T^Ww?P4d7Y8cBu&Z{!|a}3 zUhkm2MwKl8EWqv>rl-%Ajzu#;(esN{Ccep{%H22=k6IW&FzXLyK@o6 zJ00oRM_YM(%)bWMw^(~u=?rfrkDr^s=K%%ag^53?{d5@UcFR#y*BKn%)k42HP2kM; z{^i^0#rv-V!#oY>+I<~APL2YtVTEMe-Dkwe;}2QUX2`8~$)u*4>2%)C9e{IfK(8f1 z^11WUaoa~naP{;9;_%)HtlDEOHW0@W(+?ip1fNCRjJV}oM))-Dg>@8{zSfR&7l&}# za+A5%@S~jF%wTZb`Tz6VieCOz>RVQlJLaxje!Cub8(MMm`W>OS8}5@6gQk$R1=(a% zsw~g`}? zW*}@y8_Yc}PNZ&K+vwW7lXQ`rJ)LLTK-BxviShwEbZfv|Qk1con;jzLY9m*0E_%NB z&ec6IZv8aim~%wZo>9^@>?OVN_A#znq0P&u+R}!5CnuTtIB=QvwVcI)A>sI*+BCAZ zX$laey5s8&2g$Q<1GwL*v$@sBM{qjr+MJ2=2fB2F6K9m>OP`NEOSf&_KyLcEazBGd zazVFrI3wj!E*4~Q6U$?`S4E4^qMIqmUbh(8r=7u}-!@c7$EcAC63egHRG zWRuWkhp>|Ic>LWzipNi?lhV}Kc!Tf5?JnV}10p}N4=cBj#lklv*>VK4!_l3wo;ri~ z9nyEpPq>h6;vV+ZHPh;^)}Pt;lm30LOP`eto>uMq+q}B@_5krQXCD$4WJGQcR;N>p zIvpI}W%BLogqerg+)IgU+o5#nwak9Ob6$@hrN}Z?lLj%<?|r5m?l37&v?2K5#17jAQQvkWYq6gkF{X+_miqkZBaJ<8Wz@rMC1 z4*GbK#u2Et{Q@i)A5309+y|FdWkcb|0Qku+gM41S3fJ}1BJ%?GF?ZM%%t*UZR@KCb zo5LA!>!%Imwmv_80#`m%2(YW#<%j;+|Dl#t;$n8cSLy6GbM*^unlnmx$~c zA&Ptdnj|#Vl8bu}&^Rkq&PWWnUz^_1YqQI$%N#9GaNK(|H^&Kea8>Z`V>2M@Z$?_j zdehsx2hi8C+fkx{16nkOWlGy}@!$LB#8T{2Li7+UaIC{Kj;U&zlxn0ET7(?JQl=xVtQbbnAF^m!#?f2QkTrMAZ=CUuK( z#VuBczV|Xk_dP4oEAKj_dE5+rT|5PKb<~6J+x`&W6BQI%)? zy0~uR2WH`1$=S#v4b=bM5LEhVu&}0c0N1*DEo>muQRdN^+_njQ^zpnrym7%IvMX#5 zjY>R&m)UP;!p&|AJdA0z)+{w)<&oJ!wmMGukgC#8iPxc_W-NVrCDzB^}Q`0J}pTvYl(DqP-bMv9;eJI>a!DHC@zAj6>qkjGW1m zST7MdA67`m2EL}|5rH(ZZW_9O-W`Qa>?N-}r*f~UJ=Ncv32(f~C%ru}z+{DFZOEM~ z^yBYMv@&iEaf~gX=FvaM_oTfj-(m)L>heS`|DiKiUarOE8R~KYYizkVzba@Hu^ke< zySV=Wk?cCW@Gjd<2Dn9${>sZy$~7@sc3nwmr`iuSdL4)I{VIvp z8F%h)_i4JOHJE-W$$`^a6zId+739F6&+zX;hAb;mA{%~%fTvfcV&iX-;O1LZ;f40Q z=-%-(xb@;elww?hx|DPHJ^~1P0(@Q=GnR7=2k8vIN}Hs>L@(g~B(6+fkEvCEV=F zk;jE9@bYv|Jof2%R=czl;>QHJ&RK@?u)Hv4@jNuz$qcpAfBkN$O^s`@M6p+PLp87& z>aX9!`@Y7PqiA5o8T4|Go^Z$&d7OKZabVZdm%mlWf9&74`Aya0rOg0 zOSpNdny@Ca7o8KP!a3K+!usf20JIGe{*@fz9>)P zm9F50iecbU`hjzn-{Y8*b@eG4RN3_!hU0NwMy3ia!U3!@?wg@5wypqy@9)Lxc| zj$4VzIRgu*G5rKd-!PE+g@lnYEzzi~@G3d#&k5TW>=rt#s1l~#D1uczX2h{zCLWu* z4r;zX1pDUgfOo718>8;c4lqlCx9uoP6#hbh`n6n!mzy;J-1-9ccF#d_w~wJyFPqS~n>&!8 zOF?*HVJli!r$)-IToo&4)v~VP?cf9Z5Eps&rykLD2=82s>m@OzB_xH-FkZ+m8q#Vf zb9)ww)Tk6__2dZxI2oK3)PmRL-G_43-%$06mB3*^omk$*gttYhZm($egTX4d!S-x^ z?nabfkx9Sk-{ISFMgPTY(MMfuc_$8=Y|@}-JWs$+r%t1up;OUI)hJMsY{|hhWCXA#_>Nd}{M#7St900IIg$=-JpVxM;G8aLS=y=z5 zg=-sixpapjy4#PNY15`OQG{GBN_2j40n(VXgS&60 z&80aE<~jn}i1CrJJa5`nF2cPDYQj@BKk+fixG|A!6DP0AB&}h}H16{jI@PuxnGO#! z>nq-}<>A8}ye_taofR5vn%g)W{A3un?WdUbMl6v`Ya4LiGKX<eAGL3(O1UECY#IsUmK&reN-PjOC#@l-k zuYOhJdH>l^-B}VFuyqM_``%96&jWVDX?<>LVFS%wx0A{oI||XVmq33^3b@$h%YV=A zvk%1&Dii4yNgQ~;=BK<5mdaV0FgbuVj|B@l17_J@4VOddxmJu+C$x^7B&L38c*jO9 z@K#m_iTwr=Gv^?7?rC2t4sR!SM;`_8SF~twuQQ$YX$I|5`AfpiPN473l@Z5-{lUiS z$ATXP^|ps545FSIR%G|mG~5?k1lv0$V~mn|b#pbip~lQNBV@KvWVM)1rSiys{1sXjeT%#+)kkNJ=5m`ePjfH#PU8FyrIEuf7r_Z# z7gAcK%I8ZTJ<-lqArc)aa73e*96?4#!>D!oG-{zbfcCVxqrd5&psBD7eoyp-y<@!b z_D*-n`Ag++=zSBE43}Y}>}@13U=MKgSEcU3O(d!=oYdp%c*m1;c&0FijM-*IhHgya zX7#t>66Y&&c^f@>dHhBDK+{+^PFd#<)!lHMYF9NvOCQNx&4o0$&g2TUEOj9}p2d<8 zd!~?^I^V&(9}CedaTvGyq$}rqK%2Y!?JXr|#&M2Mr*Pw**~2Zl%bCEeQ83}RBC@C) zi7u0|w9-NnYwIH>BOG6{>yN#r-a}KU#iGYjeH-@FKFxnNPGXdP)!04yxy@hv{jFZ2hgdOmWX;Fn_2% zep{3Vr0*MrmGO%|^C#a8|f}x;?jR-TQ9y3-k6?BP49(^5nlzu4rD{fpd4?p-=h?ieAwVhObRj`DpfW3hV z^!jZBI#BHh6z=9Opiws05gHmrIyRgp znbvyXz}af7TG@07K( ziiJ9zw&V+`T;@jO{O8hhhaYP81q4h*QZ2 z?3&_9#&?V$7FoB*-9?mGR*6VT=|ONHB?&l3$UwW8#dL7MFS6#kEbl9%7>2ahL!-AN zX^B}cquw@#iV}x_w3q#ndSESba;X-?|GF>L8JVw-4WUuv3b4fB4 zdip4CR_|9@lKfl}>yZQRTaH8CXe)ZB>O|wePokF%9jT_3Cn}13iN0NlLb{$`=_Jd~ zGn=Yektafhir-Hl^J#%pb@fTCUVa&#v51Ekg0E6hv<=5@v*$*c$kNuwpFvbs zBpUH)6zXf9PNOztqdm{G$S?U1tjSVmSTJyHHTkfIakD8P741)u#SV8Ab>=sEILBBx z;MXwW#W%TVi(wBEULS{6nT>{zCGiTw22V#LSXXr0FNqreh#`H0N~l&#lthmeQK-{+ zB%U#v9vgp~m^}3&F?*WC0`?s+!Z9~|JcWuQpy<++KDZLep zUcb4!YP4`tdu=V-cFz}=YPp|Zz`~637$#{;eESm`%QvbzG zw8V`9!(SExRizr%M_~$BKF=Tew2vddtcJjKn$EDndJj3eeG2sNYJpZ4j)0@?39O!> zY*ih*K(beQ0}xC4xZHr#%&R9c5Ntg{K9%<)<95d27t54EdRnPCAz%>dfIGmbe$F6g zi)0;RktX#HJDNXC2*7M`X8?NQ^bdJ9~3QpCsgaewodEVbRMYH6O4!E>T2n(Aw zVofg@zP}^Q8=wrttu;1ipv*1c8GH!tS5$|{5ko?<;6i8O4`*_nRUdHsd{9 zaUV+hu}N@1gD%Oq`;GU<%Og7=EbN54FYW=G)E|MJRjcr|3P->gAK~)`l^hjJ?ij(A zFQ^5M>WZXAMlyD?*%2A{zs;Bi2Qz0Cp0cYPlfd$ViA?g0o6Ow6K+=2f7?IK^NutEs z`7qyi5`CBbhLuSUWN#GlYY0U(ORkBVYlbo-JByi9$yKH@!RCWH zJhrFr9$-En72(=PHgx6wOp@80LN;%bp|5wu(}CET_Sf10_UV|DP1uf2T2_Ofj_SiP z9ZqGde>9mdWB)Mlh@vm)n30bW}%u7vKFTDb0X#Ho0tp}3u7Q;fk>RG*%W|8)= zc5-0O8-fQUlL+~ZWQl1iUb)~bdG<1m2oB!o`C7Z>8}sm}WM`jn%$)Mu0Z!XL#S_tT zaeUNhto!sNzFfDT%yibLPxj8>Wxf68(rR?tm7UUI~+_cjQ}N-hk)(R29XT8On^at~r?zkA)2Ki5>D_)+w0YMDzAf7k+(>>L zh~#O%f7X+&?CB3ryjP|VUx(A|h4r-iO$U(5`?cH?{vGN6&}m+HcvlLUA2f*0PS&F8 zE3U91zYWRGzXyrnu@-&yF@jFDQ=rSV5A%7i{z_&3ZtB3HkE7{z_XG6z+__ZFkR$$I zr_-lTQ14oKUM>?SR+8)e|Bwy0WU8Cl0l4;|3U)hFOy4Ixq3_nLBXNV* zfvu{KaErm8|Jj4g!owZ7l^!O`uQHN8-M_gfC7%&A+f{}3AKi@?6ucs{n0PYQZ8Z7& z_ZCr;9ZWyG_rW>weq_fwFS2QB9QEl=r(H6qY55v0PSb$l{Fl3Oea%-LCVMXw-^dOD z7eD9&S1WUJTc9F|TNDYtBy?8zg;3V&MhvbTtwacpqm5wY~Z zB8J9PJ$mRMZu_gj${AU)l_{o-tn4Y)SVx;R{2mN;mEK}62kDW!I!+*~I+A(1`Ha1p zMy!2my+K7o<89uymy9>4tQ>d5!Js`=Y;G`*J+gWcu3RR^Kg;McI_sEZT{;9CGEM^; z@U$Ff+bg#nq`D!qX5`iJId-q@W7YSw{d9hceZLg2%ZIbP%^znq9a=x;Ih-wC&S-pg zB7R4Tt9V_qIl+c)Q6mnE7LwtUyvX;|Jcm_qDE7bTOPalP#O2#X_?_QcX7GiDoI7 zujAvDhGxCy=}a&y#x));_^DpIefk@J@})E#$A23JHunv)D(e4X%-WNKq&6%Q@_Yhl z&6vlYn;2Z#6D03&)Kl(k$?HxlbMGbP-NUt*S&bX;$T$}3x}IQ%rY>iTUwX3Lmv`av zq%8bG?O)7KZId=XrZm!5mB-IBJP_=NMr3QH4N)sHA)H1se)vs@e?2~gL;d?4tk&#t zXbf{=%zFmogso?o7IzN2Ew>}*VxHp94_~ke1Cwnm@9nFYI(Q0u;p0eNk56XHl7S

PCNgRal?YBDm(ah zBIEnYiXGadgrBV5%|_UKW@lzkV51~^ivHMgjF}Uah~HGEv06Hr2;P+U1Dm!45dvL&xr2OKFoB8n7zm2rSFPD+oC|uLIXH$gge=~y^ub5_?eD~ z?4$F4?05LGNzPuBR9Q7GPl-Or6Vc>DcWC5{2E2j!fcvk$K(>V^aB+XDxh(RRD-&%Z zuOe;9n56OSv^mB2bh0X1p5F@V!tMex=@;)qpH&BP8WSKFw6#BXqD7Aj-J-;WUl_^F zOFBl({tTp}(u`5X;QqMGIS!~Xsm#&!3Ea@sMy|o%nOnR(jSCCN;+)@);>r(y=SEe? z)nxm8=IDfr_{D5r$$2P4s9KyY&QJ^I>l%^a$B~#*+=SvNu6g<+UdJ}bt@X^`PIE_uVrZ@c3-_cP2D68B!8AppHPA4>y6x+H_Dim?QTM3~9ap zS70{p7|b4%4W;i5N$%loZ%&dMe?1whBKPy#_;TK#`;716QzS@rE0|&b2TT}R2~*|_ zLC;Ugkq!M#{;R9_#1rVY)tbNVTr&;YT)!_ceKnm`-|YvYE?t0Ws_t-zp(Q%zqXr(V z@#XboXJRm16!efCCwB?HSP}>o2igF=KVrbhePCjmvZ}XqS_#D0UxQs)ZG!wcGvGpX zPjtGXh|Kdj2W!?PzzmxwaJOzO?7cP_W$g$h#`cxOLdOSff&b#KO(u&$R%SFP&qxyZ zO&W%-0v}Yee;@idbtSYaJqrzWMU1}XOm^_v$KuZwE^MZi3h(>wTDQRs;}Lok@(PC@ zttAb?+mY9ltuTLa81h;_1J+D+Ku1+m(5@>N*r+kK@a=*{c<0Zc zg|^K`SDj;!esnPmYo5X5>d4Lm2Se1WyKLkIO(T~;HJ^M^ddeJdRbz3gRx&L_ z<1jWg0cBLTK&3g#e7%mASyfvft0Jc_Zpe}AMJ;KS(EnW^UxxM&UtVY8jez)ZWfGR_ zS3_PMK8<6}{e`{tN#yCQ6~ujN9rQg~4f5H%F7 zM^Wd@k?HkD_@%23`CsTnQHGyI-OYu6e@AE)iB&FQ%NJf?Z+&!S1_v!* z#u~;#<5Wqk+6RB6lWjv}X4NnT7KdQBWY1|SkKdc!Bz(Jq*>hI${7m{9gq>a@q$61u zKcn;v4m014KXZ4<|Hb2<%}Dau8|>%v=Y%f;cdsfEy|%eP#%d5bGWQL+y7MOqG54mz z6?U+z@n6j4|3ANy$#w@tWi9@@JtC=7!kwmX@u!Yz31WH}t;Y6m|){Wov_ zzw#3|t^otq1VP2{UHE2!ka1Fei=!(5Da!5v#IS&jzwwh84|n8kydWo+HLQFoJ|QZg z&}j$o+Ix!;5yJf_Dh(5{l@?2s=FE)*Uli!p{@f1nxF`N5I(3G)QFgWc8@XH>F zKKBp*fAB9Vc?Qe0%V4=ZhIxS(v3cexFw!jpBtL~P=X^Fq^~I=Xqal0p>`1idT>`EO zf5oow;{?Jpxv2bsoHT9)-aq>iZhK4Ey%SbJGp$ibY$<~4ulK@0831)(PXTUnkYJQL{M#@NH-;3YxZ3%9oT6 zk=GnAFlR+Iqp~K1S=<#+03VHc-LY2=$Gq=W%%L@oy|Zl269LD-r}v z$hSXxQ5Nnlxgi*kzMbvST}X{oUcw2tlHk_|y^QRFa&mp6F<;i|7DKU0JQ2)Vp$Amd zKEb*Tud8DgZuYTZO$d1|9%`Ep@AXy7@gI(j0!wkL&fo1zy z=HrM|P_!kU-Sl3Ot_Yt=$68J%BW3P051vF6;+`V^~HWuQyorOHnXkg5@5EcWW9aYTPn&j zqRAheaGbsx>`ux-6E-d3H83)b7=W9>pL#ZQ__9?gd-OgtU_ESMDBcmu;fd%#62 zSA&;T^LU=5?|f9kY~FhnR>sHw*eBFMmkTxpf}( zNNL4$OKNdk-W4Xf!w2j#djU>csnMl9`lMaMf;i2#Ajif&CT$5Ka!BPT-Lq~Cy*zI* zt^K1d)oYpE?W~kG(d#-QDBGeaF1r2}Tc`fpv#>WHj|N-5Bju*r?nr62PR3@0?x)`9p;d?Gt$Y{RcGk;lKte`|VPXk2HrCK%i*oR-0oKUIypHky-k-xK~x;RbpwgkHYe3T<7xSVH2Om_hF0!UM+LJ+!=BLo(Am+91}qx~qCYCK zAFi*1v!-Rh8(itjGBehjW9OaFz-`T|9aID}(7b!$C}LR^GO$iS7DNwCYm~R`4cCk1o^fU>*(Tv{ZbfU!ecqtAd;qE5l+)qqwR^DdES~&=g zDLOJEk3^9ka)S6?NhXV4XR{6WMC`iQY+Pj)OD3c>kkw)tdXGrf)LpT_RV0r*ZIeUo zEsE%S_7Swi^azOz(MM6+Re708`3;z#TV0Uc$;;5~)n|NVTnibOWJ()9%aBVA58(9C zkKjJ*6-eG&09CwK;;-IjWMjQEJ#2J3-ANZsOrpt0 z{E=+lSD5BLA7&03Bdk{)B8)iSijFUshg-g%;^jM}{64-RvO_N=>*-yW42Fie_sMz5 zUME(2qJ)dw9-^yXgVDlWE_CkT86@=fagg$9BW`lD#qKi<*tX!?bmE6Y^l-L5b$qr2 zySev6eqFA_d_WeQW~&DjM_6LLtax(w$O%wYX-^kKOhOy70NQ%f2OV0FNZgepsKZJY zj=i@Z+F3gyPq`NSXxmeYJD@;%F{!^)2pLDN2iTz#Npag(@~-z4blrLpkKW!+=9}$BW1I6)@I@2kIUtAT4H8p> z{P}ctwhp;@*$D-nltV!%4av@ULq>J}6oj6#=k?!wTMTGW?_}QPZzY%Bco5x`;;QIZ zx0s;ihnae{8f;kuXzU3qYJFFW{;lEHpv!o4jwIG0N%-8cb?j{2wcr?f24t7@LWX+= zeFHAQgYh-Uu4Ny&6ts?Z_krX@@_M>2NuQqSb7BO`hSBz-12D}>4t;Kv zM@x5QqhVtUp_&r_*dz#UJAQ;wcicz<6zcG`sl#|3y*8mA2}Wjk{MFg~b#R$+bw$Q0 zLHSuH0&VvLoh?D&v`GwnMW0Kbi_S*fVMYbtf>rD>==ny53Y)D_K5&5tw%7`A;5=T| z_7>O45*1n6U>8f>-}d8eV-l5|XRUvlEtNfpDsq#MV0j?$oOuWaGnHUrN?+MjB4=8G&Ui!-^ye13wkQYrZa<9fZc3{j zb>5P%7o7ZW?-H4~!hSogpBQ+<>NvE6!(w>zBjDARVlQK*ovhR+A{D=^`dybuC|K&uoGrR$= z*Zm0-Yu~}AN7{k@!og_Kp2B9=WO(m(+E)9E z5j9Q~%=Wnhp3fd6ERU-|rCkYVuz?p2c0CAkAMXLx)pf)w?>OAAl+EM#)~OG;>bV2+ zk;ACS#1FRD+=mS}3J96;1$P8-#5+zG#IJr2oD_Beucou8r);!v`7<}+7-1Lq<=u~t zcxy`jHV>jZo+Oj~n(bnjiX#HY_9*7rI5V7sF2R?IUZ`wU42?H*f>!7B>FL=qRK2SX z8&nlxx2OuIr6Sj(6jg7{)&rjDd?yd$3vgeUZn+u!S?V%alnHie;A0Lvc8^r@2dZx+up$wY=Gr%>!35A^CqCpZ)m z0T;_9!Gx_@=*7ka)R7#EUW`!24Wiq4yru&6EX;F|bJVr7UvDSQG-!m2>ulhckohq7 z(G{YgT@5ZK7{gfZI=LUSnxwQOfE*KJTKU?WKASw8x`INYut1sKD?3i#bla2rwmwvT zc@mAUokL%q?@w%}IejGr=_mUvK`R%7bZWlIefZh>b4H;G-v^?0?XHBpQe5^Y&a z6jYVLCRvQFj(_-_8D4r3508J&_x&a+8lz6hm`J;OI{Lggni#h)qq3@FXk6=IRyFh$ z@8267%}J5-b0YVtmYzI18I^R*Md8;}QGNX&I8tj982RxEywR@`OwJiny=|jenG^%| z%~SE)^7#LKbIWvWn1gOiAOyWkA+J6zqX#AX?yu)of^XYGh}p$e^i8`Tb$h8v{n$hr zrnZ3|+ZajD8C|0*{+^Iommt&v@J2^EoS zaql@Zb1qqm6k3$Blol#&q|!!&NS2VSSu06Y)_c#KLpxeTw5YU+w5zmg`(C%tyVvLQ z`|o#Ouk*V1%$@Vh^E~s+GjnF1WzRCf=;>}E89#OP98lA(h9!=B1)Dlju(vNC-?(-W zUY{~maC+QS!GWv~^&T(#v5nV!+SjNGs|R{B$-DQW&;`K)xkaU5t@$DTd(Wo&1sp5p z>xxLh(V4b%-eYZQ?zsJEwbxV3=m~x2a1T7R%z@fj^&UN##X+kI<(NgA8HD(dIk z^=RepAjBzyB%Vg>cC!ZDBYkQ%JgS(nAaa8nKxs?@GB$0D45DJcCxZm(be@-*@7+97!@yOg7_9J zjNU_aj8DZqGgHuaPhqC}?HibrN%>4_@p0z*8ZPy+Vk|Xgz=pcT)nwjuZDszb-NOY< zQp}Lwdt8#U7pc5-WcK{Ij5tsJ+Bp4(|CGXffvx^y5P}J?sb86P3{nC&zs^NZhYhK% z>>U`Tlm-WH7vt~3vp+hC##h4Ak=LIfv{i~j?^j7eu;Bz+@pK>j5bhy3zat$Dox28e zTjS}aEr}@MsR5LzQeak8?7<&&*Uqz*L3K# zkgKqJ-=+Fh)#c=~)|n&n9Zj9PE*Pxc4{xR%t(V^|jYDI+@Ot{7Aa3^?!JWW3flS>H zH}%I6d~}mNu!6B$`Pkp0!QYE=f3_{1`1wqf%4+P1gd3GzZzF_jK}qzndpezJ09WX+a?ontlc@+kcAQ zmFC8u|J#}jBRdx#X z-Yn~lElrz5<%+F&2M@+*(=R^0AlBdT z0k#3hi3OR8^sSfze*QTG1Ai-no7GnY%2j-UTO*)l!pu?M8ZkK5E0Etr_5GuHgl@6< zeS&jacaZ7_2MJzyO#`#0f%T|4#jf+rV`HYwWKsS;_RP?z5qn@U^xcMk`{Q-Oht@&K zM_cHp(Glofi8}t!X9iyne};Vp9dKdiB0RmR1BoZrFniNxFvs4FMdk_Gha+b>xn8Ct5OCZb<)@+|0Q*Khl0j-lvY zxU&o2hJ1wIJ`4B8XRYJgy%72p*1UqNTI}c$6EC>+tqUBp^(edvAE9Y0Co*d3s-pSR zZ#*J)&AN!+TEsCWkIR|b#0~U|ylg06W=(}Hzs))*?Pt3SUa>hDo=9a`7DzWe3MVcD zlzpWMB|M^_t_2#Rhuc33b~!51`4^6{uCprGI}S(LNo)O>wLvLNnbUaIC+agjLQ6{1 zCiwWgIlZ3Z6GT`d>2*7vo$)D-{T;E51xjx0oE?W)^=)zNnnN$y;h1~uo8OPv#UA1) zZyZO!R*3yupUhPnO}a-PW=CI$XUmONvNM(w^v8rJVA|$Q&{V4n-&p65@0QQQPvqyL zim?RsBv6HlDYe96VRf|5gGkiud=180Z9q@bWEkx!nb=J^9^4z43j6EJnc|LB*z~>@ z-Cxkj*4Z_%*?SAwDMMf2Z2M>E`O~jxOjIjtcDIThJ*R=4%Qa!hHA8|(g$Ve#tAJT{ zL>NzbFat*Q7h*Ry8%pig5!y00i2A2P_7!wLoXnm+T*P*PTvnGgq2>Di@Yf#PCwOPC zOeH+Lg5JD6gi2|GniIDWJv*vSy$eWZW51E?re20IDTzemHP@opFJBu*t8}8Yn_JPP zz1LAmAfG+SZ)8I^Ww1+A)7Xmc1oozT9;<*Y@STDL#$|wHl%7?x3*wKkn|>y;Clo8$ zv}<{+*s^?9O#d@F^hBB(d*2iLEi;2VP2s;5cP%4zEZ)IJ#*U5ZqL;dIm#t4MLyJ%($;e-bl-+Mt)kDAaQ!hqi9qjk<=VsaMlf za7bu5ycDY`_}{Qb@O$RVhj*i9|C5dX_5aBD8M(wpuI)pAeX6~*YUp}DapwByFnWq0 zjOMkv(S?oU;JB00%xb^OcTLv(&e22FlB zPd^Ti6Rb9Uij6N`fc&*n;MA|pDB3xPY1QI0KMEt5lyY~fJ$xqqap~iT-WEGVc0fD! zhlnRr2h{|$2drZTq0W>MJ?v zRs{;OIEZ3gi*dr8E3{rK2@bgRzx97;qh5)c$W^l zw$PcC*kHq?TqtIgwo9{5k0$VE-*3SuU>u$IVFvZQG6>#M(P8dZ)G~#-og#V_tkC}+ zOVN|xABSbuQfQ_2IeNWi4*2v%8g0?cX4G9X(e~B_%;Gs~h!Ni14%S2wUJA^Iq;zJ= zl}4=AZAzV;HjmPkwV;}ERLPrr4uZR`_rNT3f8^s{3fbc`7@KE7OoT9vMaYjn9Q`qn z{Um;f{W(y@e%z3aBd07uDoZ@^2y7Kv46z$Gz*Fvqz$vtuUvo}p9|x?%^1yoJ-?J)c0r4+ z4pL`>`tn#qwHK_*3$e!B=c5}-ZeBnk?HBN)4WZ!Yw$FkSI?n*VgG;;)Jiy-K8nZ9V zQrXrZN1*ZBj*6c-i9)N|g}HP-B9n7RM7$fp)$fcq)8e?0jhsA<{qot5Wen!9SIiGF z8#6T6d9g21$(qTu=Hk?U^%sUzGM8PX8eO)kG(N~vXq>%wBir=r1B1E?nEgYK*y+*I zjYGU1Hgb%m;9<>Sv|w~KGPv5vzWRES%{rOI;=2mW`R*$?d!s2cb~ud98NAJ8-Wz6m zFaBoBC(1N>^lLUw-=vIqo7>P+O>fluXAYYlI?SZaZDu+Kd}!Y9Ks21Y1!3DRR<4m@ zAGzNVtv5z!nJpFprvwDxXPrV8yEW>>T7&w0$B z)e@}A@p^^`e8&0-X^C#rZj|49o6)JMV;&a{i{7n9V1CzHg$i=#|GOV4Q(IXkC}{Db z`*ujeaA6F+6Z*%fkZ?=<$zcfJDpaA!dxo^>LWZsyHWmDMb{l;?{TlM;SfGx|XSij( zIHT*`foVXCg&5PLm; zh?oqrlBbJw<_MiG z$CLP{)Pz0ZgiA-hCa$3L}96V zl-HL#C2SY{-pF$l42BZ@y6rS(_c~4_Wg!b$5m@;!Zc-ScycP0 z%vMu+32{`;n>mz9{WYpO;TN+Y!x|6r;;7o^%c)i`Icmx8FKD5*Ke*Saj>NBBh8L96 zss5e9_?U_kWbM4*_K7jo0rq6Sy!Z~#afr)wt_zb!(H#_Khb*BUh!Xe zs4N<4+AHxq-eyoS*-F$W(@E5si|Uj=w-fn&JV*&_6RDg=hRWNsgyPF(QbWs2sYrC1 zVqYxaeRi#;!0rU<_nG-rP~8FQqQev%*k^(l+dClnh<*Is6SMgTimuT?zRrw3(Zu|o zGL!PXX2iTYw}Ibzt{mmp`=+cPHMbClfY10h^{&;pN?elpHSoPx$dGf3Smisf7nyuO_*yB})DsJ)7)O)Lt zaLbE2F#h7m8X*8ki0&Qlf}q9~p?)+s~fp>=4NpaYz>?xEY%ggLNC3o7t!JR={xgdtazBWt@v z=Bn>8=1=fM@JY)V-@hTik=1+g-G@0S*VmbQ!Z^7}&j<>YokT9V z!PIY_Kl>p@lU+3W4`U>`g0hQjN2hxt>9v1mFh?(3h9CDYMZSSv%#0-w%pY4_toTrc zSDfcV9p_x8YQMf-K-t7TX-*ZZ+a%R^3&fK!8e8} zZ;@foue;7fJ;`AXi2wDo*+hC`z0f+*Ty(LYK zehBcDt=P*U2u`0YCd{389zQv0zzleeVm#Z_sni=Ov_wE3J`17+vmNcZ>sN8<^6#Ti zL$VB0Dsz!O#5n?Y23NpTZ5e7AI*i|(6WSP0%AoF8uBVrs$;B}@uaD3geOK%&NUi-N z2p#iY;L}o1TPf@q8CTKHVSZ$Hlm~ssb)`s7DoicWo+<p|KL$H6oFa~YDE~EF&oms!i zhFW>U1xh-^;h%SBVcQ46IR5tmdZcgvyKc$7r{xB&;F5DBr8T;rnF_SnxRy9ZagR1z zm%5+19de6tOpe3t`Nsfd18J>^4+yQ?%YxtlW7;-vE>rrsmC11(iITjrMe2 zz_wp&F+broU3YOS_AD8LT?FNDy{Z9fTd<6o9YSK=<&UwV<~n9F;$tzJLG0M(hGUQ` z($kxWdP~+bYV$@hpB{WePlS0~=relE{^A~7l`GBkwrAjkV4l#n^m*6N_#r4bQjv6mVK#b zKOC$7MJMyUR>tD{T^{&X%0bli#f*BC(16YjnNY2)3JRX4E_8__;1x6)&GQ@2%)SxK zyxz!X4o;uN$mQS1F`p-}{=?rG?dme-gVG9W{MpNN{;AuDe{MYUX_*~Uf0u`9zK$Px z&x}4dEK85APp1}5%BT38bCi0o&$m* zIU0=`b`GEq{Z~--bt}rnISa)+?MJ3;0h%o}4vjH9h}P}S0@>pi;PYDhQC$uY@y*Tc zBRclno{F%Kp`3XusYTCD37-$7#0K)vx+XU|=U^O?aZVJ?B??1vsN)46$=M&FUrt&< z&+Ev-GiTU<#qZ3i6_fm^zE``*ZAwmr%ynlxYTAC(YbQ$u_o-m?K@z2rSK#s50(#g; z4JhVr1SXLi1b;l-P))--fkEd>5VmXp8+{l{Jyv(1$R|b=uKz*lXAMy$+DV8Uen2<7 zHX`$375qdj5le~tVlQtcCiTl97=5A^9e+QQ;auy&{HQJzoqLN;(xvbk6$vD{LzddN zydKScu@~rG&4C(gjw0i_BKn|j2by{#6mQxS$`m}_fVd2ccB!p_kH2NA3V`9Z;;mfM5A(FK}l;S zT~<3tUK8r9V`7HX_~Wy%>yvOQ{%{)X6~Bu5tt_ajizbxyPzM!ra}SbH-bY^>Jr)it zyu;n%-_Y+3dr)dQO`Sirn)h(#Cf=e{8KkylAGPI1JcafO{RM;5(S~6O-tT}u!I7vv z=rd6Z-HCeY`B@p>SFN{H{O8XASlq!@KOytwNCQ$m7(h9!!-(VDjIO=Rp_`9?7j&*G zB7t!M>3)J1nEdRaYRk=dZW0DOR}C&Wp(BnmCiaT@+9!W2;#_Oy$Hbq)RjO}LwnZRr zx$7&;?X?YU-S!F9O$lW-1gvH%{4{XKaRPmMPz1RzKcWc=2WYe9({ajvManS!A&x)I z!{e=$spn(?uYSx=-fW~~mvR?@VaqnGt187jUr7rdR0p8N73Ju1t2=q*QEtNq@}l5N z;9I|Cr)H@n3Oddiv98{nRE9=v#N(Ic|`5QQ6;z_(NY%_nh z#|pSLaUx2o^Q5GoJ0Z8-Jb||H8#L=n61G@;11&uCj*8K=7RKBC>nHLm-IR#Z$v_Ko zmy!#fnj_b1hmrZW-!xhK8Sk=+0!x&+l*yuu5j_&kx8gCxhU?gFQzy>uh^KV(+o)wV zmDG9uN~U1LFb-ePhGldYW8c+}VD7%B@McXZ>c6;v3DR}N=3j;H3_0u2&XeAh)($0# z-4jD~zpJF)UzDL+?5(JWSp&kjH+5+F!Z&2+(#wL4FYD0P>Gi0)Se4r1vh7%Dct?vi6Rx-kUAHR12bT`wg%^|rx6KNm z{kv*PX2N*he|;Afc?NX|&w?|T{f$-Wk#i5&ic3*Xk~X9D0q@DG1Im=gWnsW~XC6KN z{xrPCW;OhnIhiq$l3>Ia?0`h?aTM~x11LmdtUe9X$wN*kab6)Rz2n1Mp}Uo*awCz~ z@-XXz1fRfsK(NwNz;T^YW86HNh>1`gUd%bQ~3do;{*G324 z?LzuzM4>##}K< zQfS8#si6!HdA7s6jnIW#@J;+gDm3z;;J7mhb;%uq zeRM7*xch;`YBSS>f(Qi7m+1=%6!@~j5+`2yHF|IHvMcz3daKyw5 z_&vXk*-=!^40xL|rm<(CR>Lynec(B|@oE?z_Rzy))&r(VQ-;SG!>2mdy`ga3XvT2t z6NKW*(WD|J-l=ecC)mj2-Pz!R(acn|Q{_2&e>D|aUHwP}+VoKiXI!M4{hm-|)hS4O zlq!3J*~5?~JnH7>duXx;hr0S%oyvH;fq!UE0Q8-G2a5UjP~rR3n4f$n5x(1eROyvj zL3o;kFBKr|MAfo-y#2%H>E11u>BPj_bo&8o=%Zo|zpP%3)omfq>y{6%@mUCO%pNbw zXSFF+_BM_trGHXYB`>L<@wwD8BW>OUp$*<`YjIzKc+M#cxHnU=!rf6g%JL{QTl<>6 zbHf1II^=+DYxL1o&2@r5idO)hwGf=NI6?0sQc>qG8rgD;(Z=UR^y$!dbltlaDq(vD z<(ju0?NrRBCM<)`|F+B-jJ-2JC*_ z2=+sqE*s+b8$Uibh0!(67uq-Wz)I@}V1D{k>YnvxsKum#Bz-AVBDtNNc_@mV%hO^d zI!uwmi+WmG=<6#!r-WJf!;yLXryXr!qrqd9HS{an3{-6)$7ZXqW-@-+Q@HXjv%>Bt za?gH)@_KBkQN~f!-K=rcDV`pq;`x+$gTEo=%Ujsq4espTqmt~iMT3HHpUG(D^JGfG zQG@lnyO9liS-?tj+TdW<1Z0_Q#%s>1-&(v_`8Na_S14jCS%Om(HiNjd0RAFc}tuXMl5Ew zHJoK0AHPj+nUcYT2wL#{nSQK_r5Rg#Er^i{oyYpEPGT#I!`NLJE{r3(gX@}1@P`xo zX)(p8Nbbyhv}fZ*!QtF8=qU85?d=a|S4HsH#|NF*RZ|1~p1>0+pkiSK`oxWNE5*t|C*vZYA@8^Ar25@SxfXH&O*T8z~LN2f%%) z9`;@JQM7&;fzevpgP3nRjM%sPOxAHtVLX^9aQCeky>M&|HG8V3=$w3f=0ssuhQEG| zm+xhvbq1PvTA2ntfiV%NSg9iwLnr)c<0CWz@4xHW0GA1MK8w-$HdJ;(3hif$ng0Dk zADhRU=^d_f;2&j{KH*x&eDC*R?rv4UeF_`lho{d4{o13j#I|v4@qBI8T4@eju;3AX z>zs}2oZrK?n4Q%4m3*W>N(t$<=0F|&S@>LtF!ub#ub8*jmf04h%A9d-!CmTR%pb>M z#{04s7IzZh2XADVl4lo0x+OADTi|)`4pS`n#)ND&V7E-(&D{GqjUD)Pjd4xqvG>{o z*dO1Qu?F4Wkjt8F;Nk%@dd(aUpyHobzjJRnbo6y*HK;{wW#(EoBKsto?EVwq^EASN z>9d%_@rul)t3J&1qdjQ+jMKhQRL5-IA<1li62s^Q*JC)CgRi+z%zPzV*w;CV zv3S%AE4<6`$je`p*uIKQ#TC}Wh2as9c^yUoCtMgqMf$$&aX1(btc*K>{7DXi??#w z!_QLK{Lt+9i_$9OfrdZ*B7#rzn8I^n+R0$dnz>lf}x(bD7Sd@sVh|Uq1z8X6CamP}+{p2}7y!uLOZP$U#Y4wbQSUj`NL!2qxA%_pY{=_e` zQJ_ua#tO#I<_ZQhPBw_qjSahpMDgvT=cmij3SD6h8FGDuU&C2isdf=%}SU@C+@6|olLjO}qF+PH8ET~U1qPTpFBRi)o!&4n5`r>zL;-j~5q zYh+Q*{VmK9i)iL)m_0K=)(Y+NeTvVi$fLTbvG8k!3{F_N8{TYF!f93txa)H)(X1=R zyj2(gM_0@i?n5tRw59s+ko5paF+R<-B;+uk{Sb4=*c>lgdQGq|$^|`UGaGWx4fBuo zs}o};$S{d!?zry!YoxR1AigZj1!kq*Ca94sqASdw!%6wk$RsD4nRs+BzQor=k(Uh7 zsWn4jX>k}j$gyFx4vuGrB*!pj>toS47Xm-mQlQ&a&tksJXYg20g4*Aa#4P%z$OPyg z!JM!On1ybF5q?ch&0wY`Brz7G7~_~YNKbk>j>0nxP_Hxxy!LE^hf>r~XQw~DBhiAb z4me4ZqcH3KAyeH^lu(7y4+c(1*Spq6fYk(PPTw8x{<^fDzus-5U#UMC~WP z(ib@2Cbx4lUH;n6={oDvX1D&v94z>p-{605u}C&|Ys-OGYc7a>k364j%!=9`;X|wS zBw9Cq1Ls6?ZJlhmzGz&PzMV}!^r{nJMY{%pLJ`r#xyp_8$l}v4tk8VWF1?~PRl&AVqIe1e7JqZnZT89sMq~MeJh57I zF1=vpK6(@R_dG4|kcYrSI-Fk_uM3WmJHRkf5VfnSVhcWfR2=yoJwp4)cju8Nep=q~ z5gv%jE{+}}=s){XkSWCyvese$%EPO+)g-j)RF_}Ki(C-I=5Jz4)hK8t;8P!j3nuM=!Uc+Cl# zNsi!i;bos7rcsA}bmI)(>*tGqJ`5EQbHW8$ORQ+i7m?;N-rAGHZ=B)n+N-})k)bdalv@5P;&4K3`o2H&(#ocGn4wf7rq zfZBv5wDj=;e$?WxWIbC)YrOhvKPD409gfhUWu#bp>cnJg@A?h(ts&F+K)HlJceadd z|9Zn*u{2=?;Wc-*q_C>2tnlvt^BF5fiY*UYxh6O~$SQF8mgQEE!{yl8*mAkT%f{Np z$yUtw|Np10qMD=Sf0tA6W+OL)8-Yf}B%vdJN|2LhEIb#S0Aq=L@YzEZ@;1>;Txl6g zdhTlgo_hDeaKkm=y=);IWwizFBctKE70S@>m%p(wBn4BU3>PTtdd~S0+D+i-`@p z+$GmNy+h_rX%yXmUhtLs(yo$-;Y0&Nb`2=7@&p|nV%#NP#*pX3B!N=cW_amFGCY1V z7v9cNg#%U6Fg#fs`nNR^1)Kbcp)04kk8auk4_RAq%#{P;!@rZV(=U-zoi38Kj`?Wm z%WR}{=O7AC)dyE+Xn+};rU1Kd6<}80F;Md55NNXu1rwF|PlI#$LO;~cbD-t|eHd$Q z43%0ZLY=TaG<@L+@;&n){(DD~B$O9fyu>xdKwduL7ETAE52&sMj|iB`(ZI zcIBthCF=$haP&O#4~`|y?lmMgx$?;I;-)A)aT@wTyP{9x3{jb>3FcqQ29H(i;eZK- z223M-TdPdkEr6tLwH$fuNCh}$+y>56J^=|o+sTZc`DCEG9I)+;fc^n{pr>Xb%(x=} zo9ugl`Nl!;z;rAuj2r{str#Qpu|7gQ*jP+Gsw$**a>e1`jy~|)F$)B#@W4z~4Jg<@ zCqp$Ka86nBIosBAxlh!6fy>W6u*5zPemD9=3~%Is2UTUj(^HBV^(BP}jXX#x*@K&e+$C+YBTpAZ6nK#ea>~H_kx&o+PyMuCV?{DQa(}ErpR`SzNOT)Il44sF zf%D7&(%8Wfyyk@iFPv?o3UY0FgYR=59obHbD{Uof*4z+1+c)Vn_dj_5eKYOIR`OAZ zt!4cT?w5`x0zh9G<&$k-a(oU|Pan;-KOwuDiGcSvjYi zyjApfY>vQ7zg9&0ha3TQT97o7w-E7W&H-~mNqiJJ_17g44xH}g1Ug?!Sh-h`%{yJl z`G->p4gMA~Z;kKV_j)py#yikoNS6m`#b1fPh*&7w9FcV6cv1YIQ>pN%8$ zXV;RPEf!#fZw6R?$)2$6)*xJLripNUys3+f{^kSTgpGnny2rq!+G`;bkVC$E8V9Df z^n%cVXc4^4&dK1}GI?^|C^gZ&yzUf`?-LJ}>B+<8qsEEwt10W_?st6-I@l}V^znWX z4?~RpkkK(6Ktb!L?b)Uz8&Mx990du+`=R01FL30(RaF?T#5sr?c0KD~Ul}IfA>sz7 zk~TJ{MD6u%wFbtV0nha?6%!l3L|$r0ePn+)?W+$Q4tblebQ; zBeV^9D6h1~6amt9P zt5i8o#TRYP(}Yc*-X~)AXOax`SVA@~S;Q^8Aq>d3-iM2d3q^2hEKNA}`mA{WIcb{{ zucSCX2l=*dXW4Pgmbwtroy`e;k23*gA0VEkOed0yRuT040q$J+b}rD^ZBilc>Q(jFROcO7hU~8)G6L2;;Rt(_~%wz);)}KK0b#V zCm(3FQ#I0ZsY935#GIK#Ktv~b@aHDc7>PV4Hq~*#Jj=X!rnYj~6Q}-YA7y=g@iOk( z1D(Wu;{@)YrHt5UYezCjr;5DokV{k~^%JL$yOL+eez*C#QkUCZJc^Y1Wz4N{YqxdF zN#X2SSjm|dH7puSF)ORcYMvtCrNn_?qe-^y?L#)o76G>BN6)h@4ctxWwndZ9*+)5( zV{&cd9!};&eOv^$E_Q>)UHhPZb2%w)q6UghZi6DOpN&#aqV1CRrd+*-ab(&FOS0fy z840F1k%GNz$^HX!;PB*0@K3W0w3C#E->YVkmKVm68|TY_ti^_;VU(muzS4Uyk>)ZL zWOl?%(r9%Zh`)UXSWUSK3>JP7$@PEZa)!DT;W7O;R~XVuq}N8ulR|8WEAv=_n170l z{F3{xeo|{QdErDek@U5LSpH|PsGaXKlF2d8w!)|L_29Yd-5`Bw;J^A_nJUmZWi(v3 zwgY_MCyY;D@q|2Z@D<6OR*Z5cCL@b48_}K%QsC5`Gi1fI4P+T*0Y41L!kXWMz~gnB z2+magV$ym|qR@tX8uT5z2_8f>aCJ{SR3~OZ-6i)#b^dhd6B<;m?dwoi+Z|(Oa?@H< zpy19DD0jgEI*q>vA1d#Gy0462;#NGdMII^07$#=UnFtjMwV;pk7}z$`8|d33&^Xr% z*tD&JD1?CN$F*RYy_!hobntgVVo44hBbN-L%Y7ly_5|)~xdmN2zQUdF?uvMrz3zi> z9@vUxm1ZMzB^#91A`1l-A3$aKRj~Z@5mM^Rc@RU^%T0ET%WX;gCvr zyiowVk44|&R#h9Aq~8{*-eg>ZYFOPE5pVa zE+YJlR^}7?&b{DFcVoD4t`rC;TSHPz9vOEy8~Q$204wk7K+N}sMdns;NKpm`=>8zT z)rEtlI=b+K!*&sjrE?ha)3-S4_2KoD+mVHo{wqhwA8>&FmJYD(%n#D6a1UuKv;jP^ zsg;Z3#5o)E(rl-^NP+ROt6_Kmgvqfjq<`>15ic@=t8HHHHW$qsMl1YDyA_e-ikc`= zx>8K^|A>70{qly>j!%VETl684(jm>-%L${VqJMIC;%E?8q3D0s1RcR7sl5@dpR^oq zm#~Kp{T4*QUK3H>Z<8(>;LCsY@+0{P|Ky;l@)*%*^N!mYG|0VN5=YQ>JW>5XXKfL^ zb_#Z+Zyyc%?rMO|uf>4TwMEd@Mh||TeoplN+3Q)N`8QO;9QdglgE==Rf*#c(60tXl z()nDns;UslY|22Le|Dh9C%s7LrF)1^uk^`_QhA{1eIY2V2m;p}E&;i+LXg*O2lf^o zfJS3i!n2b0aL(?p@RBep_sdyXFvKJOA5Ru5dHfr{6BF{Fn_Unz^{3!^OCMPK^p!9_ zSPqca+k-2V54w(~n}^s9=cIJZapLFpI|_&RSE+*4Hn ziXvBo+?WsK{nnpkwb?1sE;^62FJ21|fVt3VvJ)iT2*9uA0F4RCU~6(BRN}0FZks(| z*G~i@kJpj<=M>1$M}F|*Qa5=1$9NcBq>X;Pdj~tOT!VLB7LbW`<4L&#vE<^rQ&3RT zSXBB!4s~CS0AGigeY_BEHxn3ZDZ2bX`rF?=e!raJn zR(3<{)s=AZ(=^dKXGBI^+_kvJWbOj;Y@zVoMFmb^+QIzfFt9Pt43s*{LHFl`sIQLS zno2mLH{7JPo`k`~+vJ{=?qv6;AhK9J1{#K#!Z)FBL3OVoRF)8jSL&{U?*cLMYTzT{ z>dy)Pct6tKOZ!9A_xJHs!n&PQXM{JrdVVpiQul`Y0#}g-cP}R*$Id0F@mDyL!i4GO ztDkZ5mlq*r-FRfOXeqkAIUg2y?15T?(_v!&18)10c9AT8c^)WgW7I2xWUdy0J~v}xgRnjC z4IHKU4A!?xg4HUhvxX_r)u`QfqJF(Mw>br^S z;g?53{r$Wjl(M0qxYP^Cq|XAQwoV56{yt#My@dAorh_f%=C+ zV7Xu?SYWaStygnF7u1mOT~)ZJc;A%N^*us5ri!DhC9k1m6d7XeZASuoi?>@>IhvaN{?-49m*3bjA3Z2`wrgre-Dq`nFULtg!9W8c{oMs8)0Yv zhnPH1iRAwXqYo0^r9-5T&_`M)xRac_bS{Y7E&(=oNQ!95>`W%i)-ERFUv`4zBrT}>bPf#r zGmSJK+eNN9=_#u7-xxW|CW1X%Qh}?}ArNA!1sWb3fb2^%!Q^6f6mdWoIUTh?s}3iT zZ%*$a_h!_I#&bqj1EICTlDu-U5~zP`1<{NO+`VKTx*q9=j%azJ*!!j6`dH!NhO#Ww z(RvHs77O$H&Q*nrN2d}`3QrL&^`%626k zn4dk4wEB60teJO_OcI90Y4_0udv9rhxdmB3yP69PwXKCQ1>K>YaPQ7(+e zRf_N9?i;>M)U4}=kD^*27QzVhD(0wJnv!z*FUi=PEl@`@4K5ojfw{JxMAjt>;$jH> zPZmb{&%`AJ-ZD;ycfnDp5^$BYcZw&&^AD0qn?8W->|Su-VljBQI{_|H+Yfo3$xuSt zpIUP-fRdKpKv{2YBz4&|($e}Ix%$8YGT&?&aYXqj5m_wGEi9A}>Gru3x5JKw4e(*f zJ9y{vD0nH~7>?Q-DAHqOPbIN=#Q>qstt0mvJtogRmj@1ha-jEAG$@l{LHjmMU@LZu z^h>`+_WKW#k8AdUw(w@a$$mi&%ufcAVGZDO-#V}^KNH*w$OEtV%^=oX5jIF^!o9X> zz-K51m{dCfT@6{%X5nX|c?wt5zd>CcMAK9?Qg~cVh&$gRKX(~}ao7iEOH3-NH#wt5l0%NEsxQu8w)H`p2sH@kz-OUpq+(L-YNG*xoO>UwgFc`sNOZVRW+ zh=VnDn@PP3(d7MShG0jX8!(721$%0q0H2_}M3Sd8*%SS@AN~7p5~ixjBD~g9>fpD> zA;4KZ4c_iafU}IRi`sQsr%rZ1_a`gXOyKSbGU9I43+4LN-UNH7m!R~U4QwtQLwH}? z!`06^N(T9e5#F7RL__dJBCY8laJ9YyVghik_6UeX%^c1^RnPqSJBT;Fx<;Me^+=VIbnk;n8nMnI(0g(PkmJa?4Tj#3!0` z>ka}Lf7L@-4VItOQgHk^q0j8Q^ik0Tjm^2XTo~ z(5X`a&0g<{tbawIhxRq1HYO1l2ot9!(wdb6QpYF|y0MS5#6^s={Kp{<6PE;gHsnJ4 z8?i8XLldw*)ef>N^MPxSsR++Vze-yt`z|uMVifq%a+Un{=r8~_4dB%GRFH3XjXY&0 z2S#-l0BW)(2%ArW^hNW*ic{B#dI<@jx@tX;I+qIOSh;}Od1Fa$pZCCO&<#efm;rCR zHUd^Snf#KJM(pdBB^BNU5)X3ZZ7!;fCz(6tgqFlA+pQ;)ML673wjpmO>;!V9FGTd@ z6(9b)uO{Sej9C=<{_I)uMX(q66}o}E%={tC=1RcbPx81gZWW2fc%qR!G4Lyg>@;-& znNC_j|D*;eyc$g|H@ph7ywuS59pb3|z6!{kG6Bqc+CV3oNzZQ*R1v=Xe3G2?CWhONmD9$s)Xeh7FPD z=>#BgdI-F6TnH|v`~psQY~Y>aXJGL8LFhhTUbM!LHa|dC&A%vmrob#&aztkSq$xp} z?E&!Oz~uiUOQ%1*BfyaevNpMpbZ*yx&lc2za^LA-p6qY({b^f}`PvFN7h6Mz2TqWa zG#cJHSxBViD3UXJH-dasW4O0SS_Dt0vH*Tp6u|zdY8dUR4>w*e1Kw_BWYb0tbl-Ll z{y(bTJg&y)3mk8cMA3paDyc-OB?_ap zhehMVVBkqH%#;g;E<-{6^EZN=Tmg+9Y?Ou2gS(*O*hzTUAIAMA`}YF7IjfyzH$UKL zTgvoNU$+e~Y3&$zJNqG3{d$s|8SlY=WOa}kl%`AtDm$v!NaEdcIJl(B{*r%~rtBYN3)Jo&cfC6i73N$2k+^wGx%QXd{j({p{Pk+Kr8ndU-n zC{~lvw}$rN#}flc+Xf!pX}Lnv+$)m*=~E6rOwyl3{}DA>lby-!J$^5LP9AKyMWh;p z=veWi-24r6Jww(#JVXsLOX$~AqiCCB8GrBSyG9F7N$BS4xNBC4bjbeg#uSIT!Z z7{!DtTN`Jn-edN5m6Dlqj>bp-vi!Gm{P<$L5R&tukC4$)G;;bRrmZ@G|Iu|gzvR|g z5*Iy%D2;#0@lo`VCG+R29Qlr;BpyziMf5G}h;!%%dT?9_&F~Zv$}pWEBdxffv|Sw1 zc`a?xWXge$Ou&sH{X=J@liTkW(FWg%jJAe3U#cpG!%cm_hu@H6L_fATP-Wk3w0eCP zQxhe)Ga*`&x$E)TIQO0)fLT6?fAd@`$sDziug%sHA(F-M$==a~7RXBSTX~ZFZnz$2B>mDCv>t8^EgE}`KzP~VFY|_j4N|LU|G5gC52UTzLaz>`}*S$T@ zNO~=2JYx}_oM)kB-7$EyqLawHoK5$SP>0@YLvqt-r%7&aywA^CUe1jf6zE9D>u#X; zd>>Oi>&GyB`bNmqY^09sMakIM4GeGVR#I?kB`^2XPEPLr%~Qu&QsDMw9|zC9hHBdT zMG;EoRFE+-QN-!nHBKgc;){vxOfT+xgnSg^wyJ=C`qd9cF;JQ`I=GVMUEx&epe#&Y z_6HX1naIKNuWv73VcO!mm<*qjK)&Y+bF_UQx{Z7Ksg^T}9U|QK{DSYi?ombLYS{%6 zlT}JZR-XnPw?m+RW-@A=iNlv`!@#XB1PuD0!mQF6V6(#t(wFcddirysB6X4LNY3Q31{-@JD)G~Z^A?7?$l%srVica|6rS!JC}NG6{ow4kJHM625Nre z9Zgf21#-1<7RJZ5z=hX0N&V69 zT)&BfWB5@LIebs+*^G@yE;L!`;Oz$q_|ZIyxo!QN$vY5C)=oJKztrmC?n+f0yc11) z;uN?s-gz#EJ(q;g-z5fjly^&KmIiihzXsuUd|*a1XAZMpQ+!qGH@w20gL$;VfG?2LpPh`-8Tlm~~|Jwiid>m;;?;h`=Zx3vN-Xb~HFlh(7*x)*+ z|0M^y2$R}E!{evZo|aeir|%7zew*cshC48AZGk+HThG2-u#WXf@L;Xo@4=Qe+PKfn z8jp0DFy&h`m}VtPdY*OAgI0C)MuIey^*@2agP+0DQ45)nnIFYe5dLoul z-g5knziH0#N?Gn2quVhF<`vI_FTVcZAh{K8+*l4r#=N5T_xDgKnJ2U}@B(c&xKAZ> zbEu^jfSjWSCY4UYSlyAh_DO*;Pi3Ofs%a~XUvC?Q-H+w4{GAd`82b>)BW^-mSQs?J zRaT~|nw74;#J1Zkpq{%E31e8sJgMk}QMcP*O@0lSXr_Y7fZ(o9_76M@_=!%oBUr-% z9d>iceROiVi9tf8X!_IyzUxRslm0lcQdJ}cR&mS`R)K{6XvLbV&ro#WE#@+}agkpc zzSOG3_@5`)96^1Bt-I6N8yl{Z#H)feV9i-_|o|H&09?2#*mY{gcF&Fozc zQypT!)oeQ~=v@git7m{`lm)mSiN>2<4%lA#2|TaQ#a6}?b=Rq5rpsB}bo!*=96k&S zjqlNqC6W+QKL!%tmt%$29lX+W63b2nv%8PSum*RIuvL!A7!vRp+>QA#-LHrZds)qv zynVnfxT}Ji>xFUJ>G!b0jpEd2uBhS~j~92pg_(t77^|;@TTf3w)#tl$+p06@IUQL& z1tT_Bp0fIpOQ`bjAd)aTk(g@p*sMERZ0WTwJdo%J;%ZAFtzuiDNy_aE|U~gXK zpFwTQOGz#6;ag3420Oll!|c;%Xo=u1&eYs@cqRNB-d~e~2H~erJ|Y<>dWc|2#~hUI z_QNK@_Z1^cWE@=+uxb*qw$54%>{cO z5RKE~aP?~|B#)f^-&}MML=G7_4*3syXWcU9S?H05@^(JIT#GvR=0NA~G^(m7L@QZS z4u)k%)9GMo0@uFibUO2zm&pAtHr|3csJrw(9r16D$B4}#Q#9w&5pNpdLBbT6Vj4+B zTqVgf=anEx8-n%T3J@r`%W0IxScsS_$+exZc0GUOhaufJMzE(i`N5T&V|KsYNoqk2 z-4b35w-(KW!dr{}8_Vs&efT0`2Mz~wXyZ<6TCsctjQ{nNy6rwi7o{DeE#q@J8a$1v zg%>*3pdUMmYa9LU4sooCEEP3B>U6=&HOfu`)|c*Np!I|0B@cN zL09!UCbi=7e`{sc93=(6KNRz*lP=ltbzteaJFxdv z3(EKO{Ue?=JK6AEoeDE$OBEKZoPuRdRTrK1c7!T9WMcnIUW`KhVO%_)Lok z*}7OABqNa*l`xfW)+0~Emij}Da5)GfEYM$mB8~Ufr?HYUaNVVlQPxr;V?RA6BV#}F zKQyWHwXO2`5898@lYg6OeS!|m8{-IRP9bpd(=WKgOZH_b6nnKL& ztXCXfhj*N&Z){8G(gHC~1{wysl|^v&KoVRgPLTN_0oI2fgNA6q9h;h?@z(Eg;47#H z^wjt?ye~F{AO@FAfs>12kI zD3x5hi+jIBRsL_ty{|)#4A5EqCWI7WEmt+Rd)q7^!d+Ub}S>MYR2I{*@|; zz==)^n7ZR8jT$&bCZ#Ro-Z?2c)57bs;oI+M6fL-gb=n$T*!je{m&aOE;p9lN0VZz6?p0 zBN%@up~joC=#GFP8?^ovCT(AMn!Z?ng-UBr z1N*WFR%=8Gd&w@F6@N08ii{AW>$-;Y;XgSz=Jkj4ja#HHR&B-0b(3IvC)9%LFh#tj zW{D&uo*Ct0hg~*Ma;hY`}ZmZ?mrybl!Dz{u9qe2GA8t(&CJ7gWb60OSd{2wlc z!@Kwm(phwa;GTI0=}vl8C4fk4o?ud6%Q6Xb`8`;7*&4*N|CV+SDNX1+|DN zgze)+Fhyz(8t;=NA&aBQsi<6{^+K76zJrWTHj&P>!D9A9ibjFX_bUFV9^|QK5x?WDDaZvPd9Zi`Mamu|)rCB-b#WjM4pZz~y!l^IZ3&uBHDBN1g$RPA;>UE%+Z zhRIhmc6%~8{pd0HntLW(QQ>4DDLI84oLEdo%XTm~;uA>8oso2%b2PiP^ej8h?tYL#_0Ty@Bf}1t5f-o$4&UPJ4Ue2wqHXVok(oTnh$?#i|IgXHeC~y$5Xs1 z$_tEG!mlYU0SVJ|NSvMs&lQH_n(hM@?{i+HCyOtAk%38NiV@#7!5YNOfc$f8vH7e)cV0k~koXEsr09c|`_HdYZ^ATT#s@{^%wKmH{&VgRM1r z3IC(X9pmUt!^r~i3Uc|%ZKkfihnZ~GP3j%xsF3rj|Kz_;wu;!-+yfQ+slcE7l`imh zKG&&V%xeF)M&Cy5WJ>d2qTjlESe~DP#jYakx^-XBPU0qhK6#Pt z5ah2%xvph-V#mn4(qu-XVrWk8*f_zsSoIpCR$qy040Ca_cm~#+H9?bR7zEjCK-ma8 zkZja|aYydchd)#3)Hxp1`gAruv?39tX6bC$ zG$LQ1OhTp{h7sFeQ!P0=kee9>2G=fg<2}s1jIpc2a1n1NYNqQl*Q>1gJKGG1f#(l$ zds_pEN}NmERz1S56_wbQ9fy7sZE*iM!5-2^5JQ#j1<@OqLF;1#RX0AqO_0a@pcjxX zzRtvFr=>|$`&5%)lMdLK(E;PEn_KN#3<_+(XP4? z^iH!q^>i)a?>V)e7~Gi8!Pg6$8Bc?^{IzBg%+|&drub_S$=*MgnA{&n)yCG+Uh`h+ zQ`klq1`gTn|C8bJW76E&zzw|%Oyslq^dS>N_pULaGZsiOw*sNlv%wEP4Wlu8b{RKM7)hrr%%Wd{r^49iYFB;C{t}dyacZh-CV1WqaOAG!!ode&J zc5`b?*9co`2C3A>d@l#@d!1+wpR7~Gyuo=QLYC^I!C=AxcyY6n`+t$u0AoF70?`mV z%-BWbG20Uh8TBo$psaKfDyQEA%eFbpPnVZWDM%8R(pKsvHU|0zhk;1YRp#lD1a1rq z$0V4u_cf?p7|!X3W$rmtF`fkpy(XGeH3#P1iR8w5vf&pUSHBak8-)T4Rzc;A2QbBb z0^xO7GmOwi;lfXknEcmGjObHoGFoc^b$LCGqe0G>B~-ZBm`>hmLeJYO8g(YEGECZZ z-mw1AA?W(u3-qKZF8a2gY_EwX!`Y$OuLx^H&409WJTpiR?F^tfQE4>)^HaL5LKj5W zu7E-*O&TFOpT^ydqd&#ZkU1Z$sDec|)u|uJ`@YAPAC>cnlO?q_XL`m{6?U!r4iDaq z$3KQv$n4%jt(R5Mi&PN*a()8^*;~}N_8m3PnM}Su`NVfh9LaQvH?mm=@339RudxHU z*HHE16TH(|hOZm$5s&I842=sS**|o#cGIMPcfDbgY7suM%|)}lCvf*k9o*Waf!YF} zsDt-c%+~0_bLp>eaP|nCoUMUt24~`t)G%mSJQFIKK2VbxzL*eVgZ&A%SQu#sp}lu$ zfAYXl68oyK#|* zC&+`%=QIr4u^%V8B;oxt{y6WU8D6|R2mSA!;m3=G@y6V)Ggf+}h_5xrp;wy}&e`w` z;N)g#3|&Is))cW%RkB&VdpWG<^xbqxttD;t+07_71VYZL(Qsds4-d|b!KSWJ7?iAv zWhYzb(R^tQBTs>`cV*vhSx8S@Fi%}`(2eh;#z?z)LtW9Vu>!;bowpw`N zrkoMD`g$xhc=WQe%D>o_#1HI;OFLnCA|JX>jDc2rQC7cy4EtS4o!v2ZA>K%L#>$ zG-p&Y`DT;puaQING26sfoZa9%Q>{|zJL?0Gr-g( z4nFKz1L7uqa8gjaIC@SIq-aU8zKjU#An^mW`{N+g!W>rnH_<%Nhp;Cl4=TfzBog%(9l`+Uy0GM#rqn|>n>&uAC|Bi zd7D6TYzwt7y+zvtt00*gf+!8V{btbqSIt zYum(4Zbk{4tl2V;ewFz^@+>`RpXvnW*1A^y)5W>Q2j)J-EsJZ>y8Q|sDzSsl1{*-J zJQC*ajiKY7hEj2XPhlrb$FI-RF+S@qwgs-jH?2xowMh-{T`9z#UpG+Zc^%Sk^^o`b zI_#Gzf;TO4xLrgPXNVYMOu;Mmenbar+4zukG@pvIHI30hVKWXI=drrK@&z@j&i-dV zllM31=_O-WH2p}`ZLA3O?=KV`9ZtkT%{?xXDZbthP{TSwW;hf@A) zYaQ^O9N=lQJEx9i^;YSzS$hg;*dk#_JsSaWX^oh1?+%veU&Efhm2`4YGqGH!1>>$< z1gG>om=$~%V&mWA{JmdryY_F4-PVA@c8}1b{waQZDvgt;4#$PRMxlP}ZWMm%iATD< z5Ki1PsTJIl6>L>u^83XkEP1Gni!SNl`A|Nb-6jNH`L$HI#D!@P4doB)ui#aT_F&n! zE^N?#!QGrWSMg~7DfGI&2#>|Lky;;NGSFVZ=+zvCDA_g8u>CRRbqk}D_dW=0I1ba2 zKGR3GW8l7u8YGPfz!~rNB2Uo`vyaQ*?*bnF-rFFkP2C9VXLeB2p7nHdXajy(dmE>T zUBhou@Y&*9imW{$nDMhd-qn#q5z^_-u+Zaa+b z*$Sj06qY^@gqJg`=!3r@q)j@WeP!p*4qIc(wk}VndBRpy+Q6E+=_bMM`b>E6rxJ4V z=0lUWFPt8C9P)JS*gG8#to?LtHgDG!{CL?7qmu+N3d2g#$S4n;i}G-0*=f$!7}!3* zGxLZ-Rhx8tJuMser7y=PSA9{xYbjPtsbiNI32Mk%UuPwRM{s&iYnv*moj1UX2_9qI z2x3Bc5AUN+S2$+aqu?H>DezQXAKVsuKzIL1T9wmBZ~p#8FP5vL+?`g?iZ}zsHGZg~ zcnBvc?7|&p$IvX*78klGV*)b(s*SRErr8{m!%a{xx(^Vrcdrn90I`s}Ye zs%X7n33w^@(yLFySiM={?35yVwr4^lI4ztD$$kv%d60r%GjcILFd1_fT!L5Uo8d`H zD>TZPvF^w0*xQe`vJEdH*}n?m>_p!~>^ZG*bcOg74!34Oj)Cn-1C*VT4BG^GR=-gR zf^=oUH6jhS*rws1(mXUj-^GO0edT8!v*f=oP=;mMRwQFOkcip@)Rc|GH%m_76R}g+ zY9EUVtueT1v=Zhn9*=pp(IDa$ip0$eo0@d-lbx@L#Nl-&bNiQ=+=v;DgC0}RFHH?A z@(sa3r=8T%U&h4!0Y7c{9qtTaQ&U^Z@xu=QEV8rjy}ekTST)D`~2DGs_U_TmW3ea*B+ZU+v3#) zMtDbG9b3%K^WVv@=WQA(CA4GCFSM1f!$o^5@#mX9tj>6Xqi&X?MnONDd+jT`_;DBe zS)0)0bypafc6qX9hbf1**|p1t6~?OU?iqdf1IqE$tS2Bg7zeYDE`W4h2MD*fhMSqb zpz~W4o-MshM_N6hW(VfrY+Q{!??l{D zTihSL;sY_@u^z5ltcPK&0_GLG!RPLkI4-Rkr;ki#w=PX$cX>px#mZ^)$T?g3BUgx8 z-Z+jEEiBNxObvB%gm9p`SP)xf3s$xtapKZ8+%sY*4sh}BSibpnW5#!XCdr#L6uVWi zw;$cyKj8Q9hv=Aj3+G+Dh&nB2P$*CXC)JI@S;5utcIq+wdi4-$$~fcLHuLAeH8F-?~2AAKA zq(1a7)e1Hg*wy36-FucKcN9zhDEn}7QPzb4hr<0FATg>&}-cU%p8_q(+wP^4z3xqMF394I2L8j6= zh%(s^xv{Uv+8!_JSZe~$sVV(oTS-4XRe{ds7Bq8lKHYM55|oP##Y8j{K@5*z4mXWh zK>atQ(5>2G$2XjI=-$Z-oGMkk6m9R7#Xtl|EQ^v0ib=A&ZB_vCT< zamEu(|lDuQ@M*VnZEaxlpC?x7Q^C1FK+0AwFJ4qNFnz~Tci;}I~Yb$Xfp z^+CkAssZE#XGBqNFF@fNc^G!MT;Pjnhri*cc(?PE$i)6-#No3ko_});|E#UTo$aw4 z9=zW3^qGY=_uWh{m7~|N$Uy33{fwTu_mX}|+5=?*|L{M2UBcq1^PNB#cJT_?8&FA? z1+C|P|M#vpQ=e?#J)BMs5cs*j27%nGaqRW|8`%PdVD|7`Bl^O_g_tKvktu2usLo3{ zjvqIAN*Rr?HH>`sND?HI%g&!t%{qT-V&@N@#s%9h;i6j=n0Bj?Y_oVy+V3KL9-9KE z#PncFgeCNa9>HOmyK$@iLA3vI6>^VTfTg7sX#OljJ@>o#F0u!c4Mp&yn>Yp-D&V$$ zNer4LhCP>cG174io3l@n{WDI2tv(hG^73IYb+`w}yUwAu3uUO4%nDj?A)Z=jMbKda zKikac_cSi0p2~f$rORI*f(FmSP!KbvsM_!nvI1jTf`eVsDV;otc zhF8WX@O@-6IDP6K-wm(QD#2DP2Fm8y!oqw@P}i6UnQk}P{?1A^kEvx#wu!;L{x5XL zpg8RAh=dc41U|>gz^6hRn&ufz7A~-%^vzaQb=y`ps%Jktes>?1dELQ+j(W7txqB8FY>FBX201rP%#EuQhCNomUnn*kxZ_-;c zhGq}%B_dbm(m>xWOzIy)M)>%5X2AR%hLm@r*@=gUw?+x1bMO$AB z>JyAv%HGLd&mONkD9E(}l-ze0ZY@oOYpX}Iz0;-Gxf$YYeCRA{HOiLy;WetG)(lk_ z-+?J_G_IR?8&}z!M!T_v82A1>u3B^zEgY*cyIUWHB+c-|^u?I{G72Ad2-|3zuKxFRL8E$h=U1!M_?X z@A+*Sb*F{6&OO56t1dIhSXSTT8Qsd~jj+vNRvef>*4{E_&U`&fmGzyudGu+twjkbO z7VaB`oc^&hjAo7-t24`_y!opZQzr7nKJsn1BsrIIk*}8Lz*tU*ByobX&IKEX{FJkf z-eM%DPGlPN=Q4*?=W}PKDar;+;rdB@+mRZ4OUYz@fXtBJbN(w)>N4*b$*>!WAJh}b z*QoD?x~%zhOLQtN=wD3>A&-*ki}c>gXEbyA zcY4}8fugaQC|z$ay%yo?IUoVV?>_ zaTvsS=)qRk0MOo^L*mVb(a;SW|ATY(*)Hx3_=ED0k6A=#E-hRmPv4v~;po2CdIQg1 z;sihJ>@vE@Zy{}ZaDe;$pHAQfp5>o$i2!H2ScsK&5ab4E($#}MiD=$iV*MwWI$av_ zpV%DB2I6-D&d*PQ3xT0@-r6zL;=4Xax4O?YG_uot~pKOT2@&{w-t1>K$S{uLN+ow^XS!*JX{#ME~DJI_WW)YB!v2H9Sm6XePNi zZ{&aN-Cqu^=hvjZFsf3a_tmh@3kW$0s?!C2|q+r~QE3`ZThxB$FK9DMvq^i>H^9 z&r$aN0QGlU2BAg!p(-^Qx)LcYZK$A^LSpE%pa|+cE0<>Ji$c1@5>h-Vgf{C+aPnl_ zI~&ZRb^z~tG}W8fO;2d>!AnvI*48V+*)lEeKUrm$3GW6_k)ArT@#iU`;4e(m#|jh8 z*I&qiMu8o`QCJsoR?jNzGh;KiiPSZOP~m6(R73y=@25UT1D$z8`>uLL z^&wsCaG1k?Y2Rvkc0&f$x6y~ZdOuK@a1NG4Ohj*~7+kuolZupQ(swtW({DD{>BW`f zspM#LYOMX7NwM+odzfjsrk=h5dUhGJTd znQJq)M+2E*?ErR0U7&kqE(+-fa_|Rst5TuK`AYxVSPJ6k-!DlAnE*j-$2Wxp}6pV{UtQ@wK+)V7lBIlNR*yf2P?Wp;s`$%Ox!c}cLeD4*lwrFR)1>8~iLS+tu!=ISTriuQTJU%CaCu=nx(G$B@B?;a6$@}&hU ztNxSc;%Rs3=WTxoy%9ik|N!t>YVF~eqpBt-SevcS-gsZc+h#PBKb(*#HZ6__MV~NyX8>%6Q8*#Tq4j zJjB1{qa^m*R@~^J?j+-r`fbK@MKXDZ+cZdp-c(Kwo_XsSDC8JS+puyjQ`D13K363g zMelFsY#s55V)~yS85z07|umBP4tw(z2zj5vKdGQ-vQO0XuQyfK$~R5^t@1>NQe zr=6a_Y84V^w-|EKYa(?XnLsA&P-DXPCNavPZ;0q&XRk1A5UfW5B`4L2R(wU)PY zFciKR&hhuhk7Z2Dw%6pHxU-Q5+e8L7O=g1D_3*F8e3_b(Tr*XDU?|6jzi&HX&0g_V z`fk&=@5?jVd;Zwe&?{}z`szXq>i>E&`iZ(kcX%xMQZH_-^RSs0v(1D#wn>^<;eU`& z?KvF1R{dbPoL3hmEk^$B_F->FXA|=iM@i4jHsbw?rMI`+633cn}8 z;>*5JBGdnWaYzreoA`!IE-mx!pcAi!&{O`}d@5mQ6!AouzW>moFaLe5{^}L6#)&R@ z{AESL`~nqe=A@h?>2-=E-Qp*h?wvh6sb!hGYo`gbS9Tarw|oclX0;rVn0(w&I9L!D z@q8*t&r0Ntue?j#axw_tZkdRh#x{t15)C)R&*`NH-O_tLD#0kUwS=Ey`kX%r)%nwZ z+cOum3wZIDvkm$MdCso4jo8H>r?ETYf>BjB0AFVw!rR+LNq(CpX?!b9UJ5CqeWx;J z?@-14c9(J1t6W?cosKWAs({3;Vc>Ve6!tAY4_y=YzzZLLfq&&NkR~6P^+gaj{PQ~< zc|Dkh%?!7zhx9{wq%zu`SI4s_!(g~)3h1jXfdSV#cEXH@Z0E~PR!Gu=Cdc@-kBNbvQi7> zf7;>C#cQ$Y;u6%{A%rrU#^P_CIru<13`;`^-dwAWc?sccwLvV~wl0=cFWF9XdY>}; zE~JpXu~O_E&*7|x(0g2U-ik<$pycfBWcu!mEZbiy$Exa&V!NHA=q}Z@ z{%f5wx1Myc^6YQbVXTzXcl33AfUPR^*zQ<~V@@PP7PLxQY?AO_X z{D~ACFjoSNuJbTHVak8!lXlO1NYasdve74-%84W3gFO1saT&aBD~Cz1cfqXQZjLu$ zrbGKAwWUfN-0CfVNKMBI?)wL=ksx~R5@@{rKs_de&>wfd(e~3*fcdKev5NmUe!6q_ zUt<4wGpA3czxhnkuO25Bt2OAB2cob*eJpI>zY1(-UI0cvgt%&rV-`$G6`86eLTj!% zlEf8!VpBChizaJP#RxUB^yDCa!etSDulrtJzri;Wf6tlDS|JaO(i7mi%yC#CxL;uI zwbeL5kgGT4$X_g59YVk4&8PQ<3*scARKdGuHiy?3??QV0m_FU+5k?%ph(K7=b$Ft@ z882)dDX5col77lJpeubFNM%tLX@3<+FW+6x!LPP^J{kVb7CweurCKuAX{UY`_^j9r zN1AVdQ2KM2E2@Zdi;v*U`Olc{^r__e=+k7j=w!Zx-yG(EcLHN9_ZYJ3eXzOs9Bybi zN|%0r&GlP3SkHv*dBJQAQl$l7AHa-|KcLd6f**^P(=!8t+HRhP(C~`F^-)(~LHBqx zJvC$t{+qwkv)aIA!$QO{2l4)I7kua+ihrcDFfq3PbOrZS*B{bE|I3N6@aat4S=Wfu zFFHc7{&EQX9R<#_hC|4TUiwLUD39r1fB!zaY|CKQ{?V|$em`~YJq>HtzJsz-SFBi| z$iZv-Yco%_`v?Z^7|B-4%w{c?bkOUxKf$fU*>KFg2%N?XbMiSVv5*>mzXqG`w$on<%VGb`M#y-! zkbPk3$!^w4V~_g_!y^|jD(ZHL8*hAMFFl;)20veF^X<UWx;r3Z%xbReZvpVJG{?jn%oXG9jp45y;9X{2G+0>;jM zF8|F8AI2m#kAqu&`4ku{Y(R^(JO0BXXh}4G{ess#LluN`Q752JQ2XM4`2BlE{H=z_ z)8F8aiU>w)*HVo!K{V*Z92z_+h)TczOV77_g(ZS~qE9oo!`kISsBLLRMQ%t#-Ze2C z>v9VISerwn_As#8R!IF{O@(g*w_(Bg17M>O0$+*{Cin~^-*6{0G0LB@Hrj~ewk*Xq zgP}8!6>q#i?(ABSkv+`R$W9ufb;L5*__!asm2P0VMdp8aEUB1BN5!2a#<6Ooe`+6J z@%dFERA&v!B3s8a`2|pk=@-DE?=<)Sd9#LcM*iVfRV1j1V;RI;(adKipU-20e~8iB zO~>i`oBD9oIh#H6F@_yq9?fpqx|rk5dc%#Ryl)z_y0VC=8f{OjUhm-*KXKq2d`e>Y zT{$?BDZ_z*>uBGrK|h(Mkk#Lf==Gg%`Plz~-(FJ5SX5+@_nTib-u)T;-JK5LWw9N~ z_l1Chxi!bfS%SE!mv7_XTkr{Zu=@-=dOQ{y3+L1O@kfZw;);BmPaj%R%5-D zJif?y3`A!F-4$R(cOJ8+cd>$T?@r-p^$))n51JSqy*|c-H-VP#)S$!UH*$Dv8XryO zoXerFil#E{Z*utgN<4nOykIZ)P!RQ{d<1fObg*6|0))%ssdul>83;N^j`LmFX-QYw*3sH3F@YB#cM5; z-r5W)>bK#@y$G09vmZVl4#3m5CZXzXQM|cfK7@|!p&OPJ(j*&}>`z`pj@_yx{v96h zKGPqvHY7vI>a7qqK?=I!WdX!BShvoX_$}%lS}v{R@ZM}oIX;Jc&Elu4%JJQ{U8S*I zY8>7_t)f7z_BD-Gs3tFGmQc@I6UjZXY6dh#fK)cou9O+z(Xt$D=ks9wr?D{1%nkmy zs=$6dJFq!E5?d>B!FJ(TNDX;JG#zZn-a!|7@vJ1<>7l_sK0lE=Cpl&pM{_QDLvU{a zUJBWOVJ~&@>VyDt@{=V8SF7bR^jo|OOYV7L+7JLky3mV`tqGr%uQ!}oeCIv3JTAwPvn!cd=@qLSEc8UjvBbOq(`XYwBv zm@(d7JZdX4j9R^O7x-v4fm`r8@cqhzyzpOrPciW~$Fb=F3bqHm*~XjStjd zgJLBx+f)lL&ez1OuggjJ+iY_9*hng4WhSVJJRF2~bkeflZ=p`?DcrsF3l>T~Lpiqx znDyi_#*bFP6aCZBVX7%!kU0h0GH*fJ-wc>`7s2m>0bH)Jg7;Bp&_s~C_HKU?r>|^` zgg`>*3k_~QNO)per0ntnQmHbNUhh?>wKH7kON|)v!{IftnmmuBmiE#$EjOs;^mmkZ zxq|)`)D9{QK1T7|c4X(R!2@aQvFD=y?K2tQuSX+vI$w7FxH5|J{gE7d!mUvd3=VaP{yboGytv~~HYjl|` zs0rJf3-iVDVfxlow(eCd`?xrUU2A1QhZmdDP8orppm;~XV0)?qPJxD3@0QmA)gm&QA=4N`(j5K;=^+WT_o_*x3~gQHmC z4I-?0%^O@{qRGk6I*|_~TVT6S4*r3MhyBEQ!_%Pe8wriI3*gG^WL7~qPOyK9V*OLi z*^l!W_HWrFcDlC)T`b5wsWj1`(>EpKn1%khb&nlB5l;Z;>=;gOF8gecwyPGR{hEcS zFl9UGmP#bOzPaS8&KufMai4lc2*XZ6jh9LJGoX0#QdncCFYvP~W1F`!uC%=jH@b@; zNS;8SbU0+BN5U}UbC7;F-ekkQ5R)?}4w*pg5U#m`HRinZdOCH|@uz(~5*tob1 z-`>1~$}>A~jaN5am_CIH>z0!O<0B+1NrjV{TDdejuj&=~Y!=37+6W0f`q*J?ovFkx zSMr8wKQGd^>V9w-CgdK{P8`!KEHIZq=kkvr?% z$1Z`Os*Qpg8SxNie2U}gkKxfowrC4$R_Mo$y5q-w7k8(vyA0^fo=PH_stcx9Mc~k3 z36NSnK*Z8S=^61bdS=2PUYI1fm+I^Q+U(UuGl%Kum|=qXP=+sVlwp=}6}s)HW*#3K zvi**~mEvgmPwy<%U5S=QcVqUYb=b8=aG&y*a&~uNA=|uJp4NAX3Sz%!pxSiXKLdWcG)kf1i*g}OPA69$lEc@Y+F41`y9g8h@tNQkkO1H+x`;j|!E zU{dsPbQzh6x!a#$-JciuV)J`kcGnJ*2i6IaYL?<1`(Zdc;tzO#9*zZh+t6c691dL1 zM&-wrXnRl*Z`yZ5jG+r#wSE~pbBPnTF6cT4((u zUlI{aIHZhc!=pi=(h=4djRW~#I_M)c0XyF)p}b9+$(iC5lQz8=6WRA`;Gz8-*#8?~ z!&gga6iTES5BrFM!UEoyWnXw68GXE!Y9VYK-;?dzW5e2Zm}7^!DrPScN6VSlNV7{m z6Z+&Of9vi)@WAXXxDD%sg0C4Ez9|hyUO0`YW(rpK2%Ibq0{`UWXzX+h*J>QZ;JQW# zbg5|YPhrAd$r%4Bo5b1s z1CM37^=wglDw#98mft$liI2g({xOuX&)*ft1 zv^TYbb6nret~cq9n0NH_j}!ElkTiTM9|fHBVE67>z@Bb1Wp8*_W5{3*I>#Nw z*Db{u)qDv<6_e0p=RUr@mjTb^u?VAZEFCvDh2h)qLuh(Ikh^hxDQ-FMi2mksCJ!Eq znusdCV@n)l=o;sPBw|q|nQ!aJOJ0b^0g}>&N7fF<-&u`-Dub|7#up_hEKg7KWR8G})?LqY?{m*Nd+)Q)x##S=_h(aSU=Wf3arN;~|8@%WsT#v*r!oKb2v1X=PRBO9^L-}%x-%4SyLpI|Yp*BT*$%{%oJP&x&!Qu3m1yeH*RXa= zE1Y`ALDe-G*z5cfYvDV%Ry~?{_J@!YoqnYBTQ-TzK1rIq!$|dVMf7UE9@^_}g{t|u zl?PAw7z@oyc+6S~>HVlcr{-4kvF(FM)dX2GzxO*-CP-1T=QdL9@rtyWMhvpMwe_zK zxG#npRR{2Iv;%F`Wl(a~5iXQ6;AOZNcl-af%bk$l06Esq5Vy<^!s!c4&BzpH`O9$T z_sDJ(_@)~j(jS7wL$A<>7(0Pm_i&Se`DnMah2Sa~A(|ft&Y#?@!Yg=NA~ne!|FXJ` z4|x~S@!y*0nuQYRkZTfhd~_X+lk`Dlqs#!Z_}Bs`U$`G~09GE@1FesOM6w-r(iT0P zB8`>p&!QgP7GzRrg@g9gAxB??ofP+Ct$mVEY497z%W6PJx;S2`qK1oaX=B^?PjpuE zK00R8N_s=a18R>}kLX!&f5AQKjztaqeM1GZm@ya6ydI3_Zb-&^PS8j&btM|vZHZ2t zn}VO#55lX&IuY@fhMGI?u*b-HY`jODmJrxdld4}(ZDo4!YQZS@c5F0E5QJmB!8LfV zUn}+*JAq6ePm>|DEr?`m7`}8U4Xc}Sxa`45*uQi*tkIAK6`Ou+uKN~;D5!w`gkGpV z-UEJoe!O!xdF+EaRhalv4t!#Fpj_LPXmtKar0(ZVFYha+bKZ)ht#appsmy>Ep2s1z z^(&m7{Teprcfk-}ZL%s;mu$N{PBh0JDdBxU4PVl)U*wDS^=}lTE+k=izef+E%~%{1kfMchtTgV=E2=KYe-nB2aAtGcAc)x7Myjp4y~M<~r*j}L5Eh9j45#exUvuo-7Tl6}^{ z@im}N`y9*Z%_}$15u?Y`bLt;JT(La)^iYumJs*dcxvSu+ZQZD>pc(rO$;XvWQMmrk zIrR9~TQt{p7SdVnj?d)<;!FJcFKZSDOVYAHO(zFRJb6seN3Ljs`B60P)pook%nyGn z%fwHF8nE0-1)?etWTx3M*Q{nTvu;>3=~GO>{^SJs=Aa9CCp=M2$qQ6kcpFKZDx23n z`(r+3X|+XKvp(vulg5dt8uxql{IjbJ@DS15L8Z&(|C?`%=zUbw&Y_}nFIO=%{x^?> z82A_4@Akn6!TzfabpFCnRl|9Hi^~g6RbgGevWMQDGo*i5`cq zM{lv1-DyR)>K>=d<91q%>n)_X8iKOtH5M7Q%c*hQwH7Vc&xywRk=@U!OTs&J z^p5+qV^9#P$aF@zVdKz4nK|ey|;F3oTc&Swz9<;a(|512@N|LqkE8fmA z<#{7|->?;pxqAZnYB{3$wsA;uh4#PoZvuN!(8^=A{)dkLd@uRA0QL25qm}zZM07e@ zyNUF=aQ9+CuCSC=wfT#+roH+aGCP@qB*R+i`FWct4|N4}<4zKqgtF*#e`|Vo@g2J1 zq91Nnip7c1J=i_9T%_;&zL-)KdsX3OsRqK*TgK0V0 zaj1>O^u z^e44hKBS>2lJ2QpLl-XD|4%p7#5bEwa2!LHC<{nBA3iz6SPN}idJ4@NF&c+Q+u)lX zc6hwPWIXEmSNtcs1gpJSj9vD~LUB?rE>Qc8y+`hV&WeLDQqc>Rk{t+SCnNIdEb6jM zf#Q?7khLfQjwMVXIjbzli>~qH$eS?SadH~2u)mB7TP_f(9hGEbLp_;1TN`{7bs&(( zzgGKdhZQDSVb!Jnc>md0-0-RfD<%KLzM(myJ{6l*gw!2I62D9>l7RHdgb4xEl$Y6* z_?HFrBN8=TOoQ^r_M&kpWx)P-M-KkyghJUXsGNExkg7py|d zu)+axJV9|kl3w;07N&H-D6uEtu`vbCxSa;$_Dtw`!{>bKYQn!RXhNt{I`k|)4A0i@ z1ux+`X3ddJOmdhj6LYW;R~N{@*k&CldsmKf4fdnO@(yU$*Czto+7}}IxVUmLJnBw{ zOzIivDCXi#exLDH7i-b8?BJiFl%TJGdbu0p=~)MG@UH!M?x0}&Fe(Jc)Uf|_`#-uo z@U5^_8Nd6rm)`ze2KT!M;+;zG(DbFl@tFC%pWPWnI9q89`J<-5OzH*lb+E*0PBU=C zZ{IBi(=l7nH5u$snbbq4wAN%l> z=?twQdynq@Q%)=1Z@}JjRiXci4s1EU7B@RD#t&O!`1h!(L~GYs5;L45%@$T7e!P9c z@az2)F+Izx8QCa*GShdL#=vEBlzZ_k2NbOf~GBVZhM8ukt!L>$kKB+=cNSmxfK^YeS? zb;Z+Z%jYiz?H`b69+x- zz&(2m$Sj_R6?4yHFSZHG2QPw|FCF>y-U1PA25!eF`59Jd4Eq?KFS-w~m4y{2VxaM2 z0sIjDfZW(%Y|P_luAdcvbNdL9R$Wc*>3EP!&!>XI&si{{XF9m{97hi_=3%8bKky*i zFLVr%!56MXi)5x5b`0NIfZ(t%3xj6Ah1Uz!N$p2Ho#fuLlzZP{>Y%DP=`f{;@&`+D z&d8cN;S&0ANX?cRpFmW^XRCd^+! z>^=CF?IL_WnO|F0?ZZjyl<rj zf28)zkiiBkAEOUfqR~tqQ|^7X4Boj$8Gp5I5sh6H*%RpeN;}x3<^W^)_+-Ns!||ZQ z_wks-$#6ZljQZufnXWppmHxW2k4{^Vj7~XVd_(OfYUsL#?9a>K^9f3%sA&|bNSIH4 zx$^JEZ%n}V^Z(*TUVKr4R(#lor{x{NHU+6TvfCFg+-6Q*UYtpa<);xPv2)nOTMpiZ znS*Jp7O@&JoUFbpO^TWa>^O<^+Vd;v^&TtWsQpoR@cJ|uEPIQsuI>V3VG3l;uO~TE z9uRLI{%Bi>PQTlRF6ru^Js;gD9kHE))n5%Q)W%AK#@1nQNU$C73@LQr&pU|j zEQ0izsj$658o%_~fG>~d^MagPP9D@D(s5OfOzA&C-Yz&n=AtAL6>1Tmu2I8|XrSH>S^Q-2PqZ;)4z7yu#047ZSjc12M`bAC?VgQjnRGE4x>g*@ zCrZF;he7b+$Yl6qW&>*%n*Zx-3HF{veW3~@^(w(vZzcHQ$zu_m_QUJ=wOm(!25IU{ zAmuNPlA(PuBKdT^s6+2sBtbdo4NjO_gHsxFLH*rH*qKoT+K;_S(cS|jd&UN`J!v;= zIT!$qQ`~sl+e*CdrWh`W{EfDet+ZB70&Qk3jq2BQfVe{|xb|Izst<}J%S)Qn$bW== z*9%DZmpqcZG>goZ{6#OUHbrV07`@Q{i(#=%Op6XXDMr$d_d)V%DH2o>kH@+m$3JU> zL^^+A{dd}TaweVDR07JQ3SiTV6Oh;N*A_C+w_2CdU^5*JN;bh@eE%d?e^Q2jg%#u1 zqx|7&NjP-Gd&7f7e{!4NPqb@7$zqj$?9|hRUtE>|WrZ|yHuflaN=1`?|3Of+X9VQj z90~qo;$Ty53gkrHf~2;uxI6nBUi7UID@-3vl!luV^}RrT?|cIe4Q)_t`3x2{A4lgd zYop(vFVg!{zJPCp0tp?aMP{^SU>7}qytU#0w&L?5xNbTGvGyC`vtuNWOQ(f0Iy6zr z`yDvo!(!aJWF?-bBSk!~^uk-=8~8SnfB!t!4-Zcd!4D7h;O>%6tQgvl-Gmo`oKJ;K ztpRZGW-#+Z>o6mh>ctp*Uj%8*^I*>f010It-y%#GH$_Wg=R$Xj6^q_kd|aeRNq8+G zZrl7xSZ^ZnI(7|87T{tUfFQXR=oQ3uV-`Y@uL_YZlr8=`9YjHM>= zAQQspT;Tq${~XU+LfhTtFl~D<^mM$zmlplTlQv1g*nWh!+py?%tUXS8;twZ^f}!rg zQ5c({i3@Md5%uS-o$u)|D^(ineWp`)NTCgaUzGRTLCEdLIJDba4~>q_M@Zux@;#J} z+G@P;$gmvq$!jG<`x)!@F#JXP7G4!+0{ghDNUkap`R5J)H#ZLG!>&DTcvQimHMMBR z7CRh$)tZdmy@?o32_!3J#^WZ(E9l;bdAPL754unyIH{z9-x?*+vuv))S=v@Z3J#oB zgR1#+;6v<0@E%0LuB`#k-EfA8`IM2|x9!BMG>DG2KPcv0SVhliDHrX1WD~^c;>gVrcgdmPD6EsSUWud>mw56D{q@kU#|*()>uPlRs#e- z{|r^@hZ3{o<78TPHp${x5;<}l^}slbx@#wcwjnbzyvT&qEzlRu^8o31Pt%rx zi17v>t0Do~QcCcAyFSQO9LD3?kKl~=0eHB7Dm~BjCVee48|{~wjrx}e(T+R2@e%ho zXiEJ^{3z!Vex0EW>T}fL6fp*wQVaBf62~eDEB}rG;Yzy0T?yGIKc<$;IMP;+CX4hj z;VXH*v^*oO9vllJ&ph>ZGBftBtIy_QzT}}BuiW&V8`(AFv;|$*5i9S!M z54U2e8<`X6J;$`+vp@^jbLxDKXbNY)lEy1GbfVEEN#xnF7&84(2+{l_z@HwD!Ro&A zaD=YR9a~#Sh&mo5&Sd*@XnPf!LPdZd4oih1QLyf6Z z70iFvESTr|h`OlP2k)cA$f94rU`eV5jyckcv?3L73mb|jD7j&?rwopKZU9pXm0;DM z=lJ-jP8>G88HcI9!|&1#!`JnjAanOB*zla-mW`vZChvG)5(z?-2Fu~v}oNFFR zT2{xPQ@Q@=;)@29pLQJ|XQFY=lSDqsIf_>J6F_?u*$ZxKOvE19Eyz&X10C;d$1(GR zuw|GR4qG}3oLlRZ^#d3E(uEQAO0aq;pTDEZgK>Shg*jHUnMn-*|`xO1RjE% zqp|e9(q;7Q%1F9!O%yJa&ctghFXM!CWyo-z0{Y)pf{Ns5C@n0-*Hh)7rMU#x-0Q=q zDn>%}YEA5wx(#oPcgKNYe|_(cbWg+u;g+yEkx%beWCXP)KT%!sNFI%BGF}nWL|?y^ zN#FSMLo|2B#!@uQdP=`t(26d|-bWd7SJ4km3EXjn_W?a}06V2*34YZlo6CMnqqLs< zM*8Qv(V|Ivah!x3-p0S%X|mHp$KtyLi>x0~{fFY|!M?X~aYZw}pVWtaS|{*+Z8NcW zZX6#Y;zM4|_9awbAUV_7iiYf2f!973hn73ja8b)0G(3^Q*J|Cdxzjo7fO zBTx3upoqPE71q5!1Uug}!1-CyI5f`~Z{5P@x#^6cE4Q@LcRr~jqZxm2d9)IYxjq+u zO;v`$1x{dHd;;tlXX35Am+WmjLS{I2&>c+^(Tr42wCG4Mb$VwJ)i$<(O4Z|iOGflU zfW9QDU*(I~R|{$Tf__Si`a$n=eoc3j??-R8Yw^Af3ee;%2?uMx;QizO;9wgqn4O;t zvwocbRgcTCf;xkgS}f7L_4iQhaerv9a)nQ+&QP&qGKpE_NrakFM9-xf8!zH9^P@DO z%hrjk)!Wf1j zrU)!N+yu7g+XRJh2iw`7z)zB#aloVdw4{SFnlL*EAMG|GDxbf@PrY6kZ#xp@&Yz6V z2Na@L7PqmkPbk)#*NyZ>n!@yZ60q{@aQJ*T2J{`Lfx-|6Fi#4Hh5ix{*vMeJC7GDs zO5+U~y4Z<&PtDYNNNyIHhP z>30dHj~A>*^LG!Y-oFVIh{;D%Pv?Ha#UKW64=n<_^mrcU|F56;&bk&fbom@Kz4txs zu*?@mUg7Zr15ZHw@gRIbd=_rrxf=U_mxo8CFL2P0Nszn$10EXXAL|xm;1N==?;5CzkIy`KIB)6`fasegP`lUc~gY`*`bx>sar5Iy|~`7ChSXpvm10?<>{7{bje2 zj&VGU^x^vNAUFPl_JT^ScVyd+= zx|5xbw%pnWF_nH`Du{>d*?w5fuo@p~`GwOvw}@;+17pOD&$DDw$)ahWKBH8f5B6A&xK*o!Otvr#k;?V`0LH&J8HTotIQ@}MZ`*B~Oa5ae8n zprLXzat{{6hPpFFx^n5|N4i33Ju+byiOvtG$Kpk6Rzk|dMesS(11xCWBF{9xzRXxPa~K}>iiuB~2& z(X>S9xRU}0kDUSfZ5Xp8JCd1tB8sW)Nx{iGZ{r!m@8e?$P3Qsdhi>}t8yfyITr{r^ z=thE#7fIRcNnY6OA$w08hjCl;VWL(en6aI-(6^JG(EWifyDPve+NR?}&z57e=R3hw zWeddcH1zafc=>iHOlka$k1y4Q!OLafWHtXxKAMDe&zIuyPhQ|3jyZS%?!capRH5#& zII&nGMUpf6KucvI9$~%-PpIC29SlbylNDQ0bWl9{-MbG}4_k^VC4-Uvr+&e>=E;$ORc#nQ06;0!R4#Y!-at^AQd3 zw|bU8)o^GCMBf&N;EOHT;MX{;>b??xpl9P>+#@nQ>M7aV`j`y+{Tyu?r+{z!YU8}3cWH&B9hGKIbzIrt3PCyz5_CG zIfPo)twUyM+bIUEpbl65pw52PB#jdWlR3qT#4z6q4{;2|hjXKGii;tZs@BF9b*)JK z)?(;dycKlA4#O2c4@k(^2W8v=c-Ptp9Sh#W*vg+U|EE5zGSY>WFV(?VIux9LB|}8e zc@Xb+A$R51l0Cu=Bw~~`9J{<38Xg>i{*o$Uu=N(1Z(2iAtx7QKbOHY)A253~9Tr|W z4f#GCY$^VOw+Z-s)FV~l!P=?Bq{M_+yXuq4YUjytDuvuTd5W}1Y9Ws|3Fw^FI#k+Y zNLo{-kmWNslT-9T+`-%G6C&2&3s2f0YGW0g_pSlGCC||L3L$qWwT#3C0uAU80w?zV$KIp@C4?D2s_yh>_i-w4y@xat%Fbiul znfAAN%m|a!;JlCbNB)Umaw@`~BK)z@Z=WN@cM{0qz}@8I zj&YC|IRqj{>%fLH%)c@6cBBw5NIHb&79``dwuhkBXFLQfv4ETF^Fig-WAHv)3a$Jz zuwjrp_*j|1jfZx4oX1O4Z=5b#;|=Ix9iIno*8EB2_4n`K@$)v=w{D}qya}hBHH~rb zguk)7wyR{2nd=%+yUn@R>DC*3o~%h7l(Ll=-N5Io{<8fuYO|L4XWJa8_xb4}u&K<3 zAx>%VNqHNz$@s%Ag+R#nOQ%CiYw4ZFl4!~JVeo9#Xz=G$VbY>6^$zFo( zX1Js8W2fR6cUwqQ7(pTwq)4M@C8W=9fjR0eu;t!e2;kpy&NPs~!{`4&&s|RPhI&VM zrDDR*ff>wBzi6ho)Q?dlm3$u09IUDsiMw;&i||Dz&wI zMG|w+niU>sOJe~&YlbpCEn~8%FQpC+$7G2DzO<(c$}AmR+=sO1xzAM8h0&n7a2UMeeOMiD1k;v1 zqp1VQg_L7$9re|1x%qY91heo@S=5*=Wl_6#RXnKV57#WBcSu|GI8H@7eGO4xh#DGQ zXn;NY<*>`m4rFxeKK4=w#C>Qn4yqarodOk@5-tIIj4DNSSY*x>P_s0l{M#^?D3pb* zYi04So2Ss0`U@zbWF1*0JByfQVB&98ipthLMTwzRh@8HTLlUJRT~!{g%=?blb+LH= zjV9c#BS8*4ehke#25|qbc$j~N_jNU24+oy9;geH7qXw6JBpmgadd{>9w7U!}9?U-n zRP$*Fdl<>b6E?zApK|!{=nPnzt|AL~48;TU29aA`!;rp?7i~X^_is@XS}fo4)3jeD z-^_O3eHeBqALg4U!RYm;@R9^yd|~BM9IiDAC4Wt$6_QWViTaXYVeOD3*brF*dEts1VhfO5|@Z;SZ_&kqh&?j#QXN_LK;{J39*y;nS zr$Uk1s8W&bLHF8Fl2xuu8cw{0b)md-)rqmN^xk+__c{^(Y1)k+tk=ceVfuKat~6F3 zb`M40Ed{5e@$h1YH!NAD1oL#?;91^6{J4^LdR)93JuG36nf5CDZiOL6i}O)lRs}ZR ze;F6GCu1eX0CHyk#*KZK@yL6B^QV1WaZS{3We=~?Mj?sx`VHscl)D$0P&44f)(@!H zUJ5sD`i*XxN24Q`1SmB9E$zdW;Q3w0@vF8u_{x~^%#t2;X0eql^Kph6D6=$pIhu;b z-R0C$^D$e?%u^>to5vf}nmKv)n!BHQLcLzLk#?=kMWaf*EKc3;5Um>wb6B&;Z9+bV z{0?=wI>}_}nnv?ub}ubdrSb*tCVX=JKX+Q6HS1Fj(w* z_-Wd?TWQk4h`|YHmdf$>4q>rftK+XI#uoO`>Rjl zfdcg&AL`D?20?(<5(*uhFRF9l{vMM_`I;i%*BhHd1lwe8isYQ=tS30Wq{Cu1n~X9J zNaB&d7h}hOxv0`%0g~AejK*rr7L2N8{>mh(MKp(~?$8$PE#|bA3%Ukn$tBjBn_aq? zAvitfv*`NtC0dkq_FpU_nO8RiLEt5Db-PW)H>Xpdc;CJ_*C=V7=YIs>-f2)@!VN_4 zF$Vb8b-G}7&iw>+f7)MPt+O##OqJShSrp7y6-2ZLi)b)!Un5aBvBTtRQNBqR>n?CB zO%fbGQ%T8RQZ@VBx6V9wQL*4cd@$8zvyDp4PNIT8j}~O=i&JG%JE*vpu~e+_e5&=6 zyomRKHeF@kB87nd5tCmI7s*M}DnrDBjm0Iwb>836H}o1kSpN&1m+MMREf1q_EIfxK z^cBp!*E^fr>^^UC;{HW?c;5o_V^lkGHLDkl98)83nRbn`{XX6_V3~qQUIXn+n-z=Z zmM+F7a?;dxms^7Jr&5A?KV9>JTzS($1(_48rq2=izV|x!2=>3E1XEvJqWZM|`XF4C zIAi{(M{7{x;e4~W^Dj(IW8a%6+?;8alF@6iD>+YK_$AUJNM?cg!6!%Q@o+)J#|SkW z(KDy`$frqaJB>_Q&)+faaWoP98SzrEV2-RDwh2=cixJ}!Yl(?1m?J4ACMGAxpZ#wi zEiokKvU}a8&7K=3t#{euG6@8fz{1>uLIQJhfvK&vg_!gI`G;?0+5inB&Zp|}4T}vN z|8F!{{6Eluk*O)RwVw2UO2hw8?PGHR{eQE4FgG)$=>Kc{(`M!fEdIZ=?M^T94@+V;QxWc!ON0$ zoBK{w*Fd?N^!1SM>&12`?}OtGW?bkQSGLkkm2GXPDH*tbmRcE3c390maZ2XSNVu>n zPCwWWYoD;4cj7^Jl)H$QN~>=0{IZG++~2*=oE!KaIG|brqFQ#c)YAK`hE_H#4js(O z$b1x)(V5!Ls@KR5-0M4V3^w~v;%zp8tX&#N?8X{$DnC!L=UhCvhdB!e=n&o0JM$TP zT-=ZqRLgOpO)2cl@h92AV&_?dqyF4{i(E)fn#d^~s2B0O@j#!5{{NKSyCey;N^?ri zP>0z~Ohoa9`U1S-<7qbQL@L{yu$rZBH?i}O2Yg;RhgC2tgP3VWC5wZySTom9NP2Sx zde{B|h5RAhZ=(g=E6$lKN!Dh6Exry{Rv%-vCWvwJZ&r%VP^K3%gR@WR+`&rGg zIO)R(wtvG|zAl=Ze)WxrA6S}LvN=hEyw4AV$q%Z*cSJHCpmkmRYc{|mmn~ScPsGRS zj6{)K`TxX>@1%+D8Th@d^7@p~6qAkQiz3aPJa`lChq{nNx(SX|USkK}*Amq^Q2#;0YhZLFl8w*G zM@?Q%tgDGC+rH0CM0?}-JWf7FgY6~Lxai)6?3djMY|i~ttidx0QTxH;7ed9#Kv7*f zE<4yIp9hP6UlsS__76WRy63Wn5gBO5?V3NkZrhN7-@bo}C$r9(+lfqrzlirMTNSyf z7d1Jtq=Ve@a%(Q{fVik^y6bCza}*bm;*aG@EV!U=MXba{9-{A+CD^2;t5^?zG4_D*xf1VjI&3%5WG^qWLRylgCEq-n*?XP`SoJ62tdzox z0UpI9Wmk%5u~^^?>Z9$u4~l4=>s(th-%P;XX}T^tAE?{!l{~Qx8Vh|%mj~_@`_GY*pvN6ZiDY|+ z8FAW$6#Pj2L|fngR`T=08cy%}Jv>Tkh=`wWy|>uf8M>nSq#x##K)o?1@3e^$%G$Gj zn)}!opO5UNimS!LmRYk}^5Zy8Pmi-~vEUkYAF{1u>q`EV&lb_Bs2L4z2Xacji=D>{ z#*XGTo~>jj_1nYL4waJB<_PY}s5fk9qz0G6>)5ON1zhYSO-|zDd{MhOYm7v+T(Qap z&o^ALl<+<4g$BdrYr4dF`)#(w=m=wzSWB$lcaV)2tQn6jVT{32MW$rCCSzIthO84S zXIFgaBu765lD9q)WQ5}ga%sgr?%bO!P|&Gi@|TWalp7_P+-=cJ`aEeStW=!&(=E+_ zVkBd8q-H>0#3DYtBgQXBGDCii;?8|eC3oxRGIyU}=VUexV`4uka_ySGIlpfer0l68 zlkg0fn0t;)waZ+ltY{N)7W)R$XD+d`vdox)x*Rs2VM48^F>U9^FwPq;lYm9L8M_k= zO!*x?LD`OrBy3C$DfyTV`>jqw_u~Z4-o}aO%^Jh#Io&7LC9{|SpC*!XF@rpHS=V&v<&jI$m+mE7ZYnaT!=5lF=9n<{TZH6#xj5PNkl&Xs>LQyKjbn~gFCw~U7BQ0F z%3(`_3Nxo-GGp`gFe4d9m@Vi07}fU=iCf!mGBdxQOnE3^@<-&8)l+?$4+q1T$8<2s z{j`Q?s$C;d%VLOA{6(hp=q#pLf879Iz5~a^%I(aD9V5VeSqbU0U0hGpZkvmP+>HRJL^!-wid*a&X5{CJ~rBSC-FJ-m^0Lv!sMu~WgZ;NFY)}oh!gYZ zBT@FjOqZ8AQ_IV1pseYPmrV21G-l}0HW;9>#mj)X^*)4o`b?4$-LwXk?}*lkS`s!r z2%nDL!6aPKZg6#SOR9eBrC&u(N6Yz$_Y z)5n;yU5Cl|^*c$M%{8{#xPg&cZq6j=D_9QjW)UR8bex{YM10F2+TSC|oMCgAS%-d> zZ0T7_s)HPvwYQEj%UEOX!*_3Hr)CRe8x)Z(yY(3VArF{8J6f4R3BJrbH9L{4HXKQ1 zRNL}M!-FBrd>dQBo7`C1{*s69&~2K+nTZFCSb zwq*s=d%TQE7@AK;23jzKI!}|Tyey$#<~rfM?OvSJY)#>9jT2n@Js0j&uN3#@i8yBz za+BmL?ok+gUh3~UwiT_Ei=SKq9I!&9+54Z~( zPBh}0bX{TSl)c1FIaS!FJyWeSHx%@B$iEE3h@Z>A}HHz|qT zZm~pYG4mI*RB&-U)4)dUXUQHHiVLWB6-EYLV)upE3BUWi=6;S1 z;ZoF!MD@J0o-16z*DI}pN;HD@3tf5xxWL*3uKjEkH=?GGvn{a1i+jq+l}IC!`D3pqq7~vSW4+=H@udu}$_TI1`^utnr~?qU)|7in$eYRD_%EU**Uq18#ItBbQZw znhVQ5&$@=>!>QN?&MHBgsKWV2D?7uRfkl{*s> zDOCO4%{t~G&h71SF2ur_+co41NBSuC#KbXNUfD$9c1t}Wt9*>JoPo(3MHeoHUI<)_ zhOo0tK^Se@3d2tZbA3Ce3za9$;Y^;%3FmZL2%CS^aGqtREM;<&`%`v_z1nq_6Thp= z8QyjkhW}d3IVHT|JbGKWEcP@P_Y;H*qf~|2jbZ=RCHqksKGRahk$@*9b7S*3+n2{# z^B+f55_MtZ-m#c;B`b%D;~0+GgHhhn6fAF6&w)ys3Jad)ns4O@6wX>v)^V z(T{Dodr9*-`KLbIN5?7LGkXp0bWXh3R zSFW$GiK_|7V#ghs%b89O6S4(y+yjNtLS=skVPj?nH%m{7y>xykm+MIj@6Iz5P91lO ztDJFC7%2QC{ATw|DF5oBh!3&JpV*>UH%{>G015m(isYZdLYdB&oY%FpT+XT>*1u*K z`=&RXlib_DDV`iGG#B?58s-#oH=D1rZaVw9GwZi9FE;rI7c8<9X3m+;l@4m-ib76u zuf~ogzfZmqweenYkg&E>iu<(Aen2k1Z+?KVNq!bQZ+gZJzdT-eVx5>!)s5qHWi7Zn z`DxsZF}FzkeR)Pp?+%wYsFzrmN-+-U+d*iZMjZDjG0QTylJz=Xq@lZtgydS04Mx2E z`I#X(a#(>ow00$<|GuI4>+5jR)47_2$`_NgwUW&0k9lO>Ts5Ig<9)I=wh{K0y(xO# zaf{ekN-(B%jifUMnEE5R>1ps`LPLRxf!y(cV?1QrQ`o}XP|$> z)BL$l{zEvmVo4@{f&p{NQ4{FaC8){f`A9ykzsb{~6fng!!Ot{Ck04P_i`$8r~zjbgUN z@;J5%8oYnv5awXs6=It|m=Sy2KqO;2L2HM&aQo$xgpVsDUyjFp4xTJz? zGSE-`yY(TXJd!ypC(SIX?IPMEF5_n8#C;hyhSPdv#3h7+1W4HUVRtuKoQH&ubn z7VHq(B(1D$&oyey><>Y&TB{A8$gIl2#K~9IKfOUZlnJOtuK6=K(w43?lqyAzh z`PXA6XIB^JpP4{pGw z-bp-esxx7S)5wl?Ba-oF2y@^|7ZKFSFn1oxGLO5G|M54#V~OD(@=fd^`Fc1~q^nmS zhLebeio|J6Etz6VlpVNS?-;GdI?aVwV3{SK?(ioaqm3CEJeaGd*f)_@&Nc@{-bL z-4rF6a2GLhdZr^uI(n3x-d0W)31%_YiLTsN18Yw2pbBHsIFuZ*OeWXQ$C2&3OURuO zNn|jYLiP&$zwX-I5jb!_G^k;TWEvAw&Lwn-#0Dj~#C>u&J{6CK98>Ixa2?+g*bTu~V}MOEfa(s;(b zaR!sFd5E-`&Syr?$s%d$QQY8%O)Q-~lMAR2$L|#Cz$!_cQ?9cs=^V+TRqdO}dZ*Fk zoofTzJMWORb}u8zI;oJnp3k`;Ihx3H%P_rD^T^E&r-|%54_I^f9FcafC#jd0 z!u-W&xyTEhM3c{)l4V*Y50WDLSdXkBMs?J+oC_)BEbKEl7>nZSrm z-zM5;96c4m?TVpD(2IC-?wJHB4GG{-N+yRK77^{SZp>oUZ6qKqhdgc?!C0pyfN5|T zI~h*B`OF?%9fD+Tn2q zr|>z)Ha{84XkU?H6_+It1GQ7!fF7&qjAvHgvu4?D)VUWG^o86NiI9T zU^OppCFjy~nfmynr1;c#GWDL2{BGDrW|byG&HLBHpz#oHx^|U(zn==G?mZ-4HWHtq zXECc1&l0B>5=@2JI&xM#hp6#NQ@!UIc@|c{s`^dgUUbI6(wI}3U#Zuy8P7P&F^^Jg#@3eB16*CjZ!yJ?_cMM&>-Inw&`B%7lmV6GT`CR(AhiG6q{ zu_+JcI#+U>+Or_8{A@mHT381QGvnZRw?CP7RGsU}Q4`THSn@G>OuZpD1MZWJT}>qM zQaF)q+rWJnAIiS1cuahr{2@ah^|AxHHy~$ln@NlpS~6Mhy2wZOY25PXwcMcKEYcYi z%6Kl3V=kPJWG9b0RC3a?MO6Pl8Ot+-aaMCA+&DF1m%k608JJA`oXnZ2YqK~@0}Udk zBq`!uCHNQ@cQulvC`^OBfv?Eu05|rElpew6?&NUeU?w;+pFD4mCr91|k;EuRhBG+N zUYS0Ftqrndbe)};pbL*l(bg5r^_Yo_>*v8Do%w84O(dAT%=l@K$vW{+a&v(k(|Z)i z_}nTYoS#kHo?-Is>M}0i=}P?em>L;3JetVN$|Q271&qJZd+1#!Ag%E>jLfdrQ2p}{ zk(|~m;_J$Mck*cYS>hwLniReHL_RC&Gk0IyWhP3MGsZ{Wiga>-ch9WeF7WSl&`WO0Sk+KHwk#d*Se4GaH zgS>&wOy>^V-pK74Vammq4q?RZ3}O!4K;&gg3U}xF1mdzlgZo@|hV5Dq#l~u$VmEqS z1F1|8MXuC#MY;YPo zGrEk^;%*8Rmhv$U@^3+5#!+rI&KGvh2oTz}a@>n!!8ls|EqB|eN>u+#p*7oa)D7l# zI+mzB)-Hihx?JB*6|O}694JcmlY@iy5($?n!tB(I+$y=pMN>=E$+j_x-237BjMPp8 zQthWq{QY!=<63n|M?S^ed0fC6>mT9jNdoZ*WXXkB6aEKPZywjv^F5BYNh{K#l&wNi zincrFmTbuyDauaPP|>2YK3k&+F&)`@H{{ z$9+EUJ#+S%J9p;H8P+2%6ij!A;%;h)FZ)m62JEgS8;wS@BlA`ybt|%!?v_ENbW$HVJXBI{zYG zm1oW)C$$^J;Y^|I{ZJr zCD*i?{rELq3wKQ7(0-Q>Y_Bt>cgGEqzzBRwuEUMCSy%bUF{g9Zl8@U9Ad#VveLj)0 zwBPm2akJ#>wBpa$$Gkf+H zHvbeqcG|bu4>Wr9oVgRxC2g(77!U$cX% zFBLK2Ezh`R50c@Gl_LBf{zcxm&e6abQRIH`1Bq`*{VSt>zaUi}HQ@iX0#oI0qKCn8 z{=zX^i7bMCtjGEn|AUoWMcfXY?SBTY23&@ZO>Nw+EE(J&_uxNxNxi&RkMM=j0lfZL zpMN&s8wP!g0RFHSxmC4+z8w0sWVbdUftN4PM+?u9n4&`^YARA>g7pO9y=?j{x}-0k zhZn7K1i!MLV@v~|rwM+(UvULVv9Khc)E7zCZlomTxZQm`G0uogpGAfG68e(jTcOEx z8vlmSjB6xvMGzKTGUgh>mZIV5tdd?My#?44MZ!9vXi)lIIztwzmg^a!<(@`5($dL* zacWFWd?fhvRRtThaVYvCgC=Flbk78LLDtyGkrMq%S@k?mdh6^V4=Z~RozPK4>sOV) zYs$lM^leEQ)$3S=QK8bP8L0wiyo~X|?`ol*q;A`~RnS^JS^_WXr6tcxEejQF({yZr zgoYIA#G=yh>twNqG=KiE33-0k2nv3!C)31pZ&6-mXy28F1FvV3B?H^3Y`eA1w221R z3a-rmOR+tHQt`>mLh8VD77LX+wp-7{LqUEJ$jOj8(ruJk{xGLU9E}=bpF-N<; zRN-ACS!y`dP2&4fQL)t|``CUi&s&E7Twl-3o%xym?taDTzn(_k=qkh51u{7F`QJ6C z@sBCp0pc^K4|!4_wt+^U-b+#fW#ODpDYe=s3s=UUAzvQ#r_u3M&{}7W#nO+dS)&)u zPTk3@?bTco_|}(3)Yo%&tg>nCT{l5L&yL43HRsoonr%01>Y~G8;@Ed3$8IP9NgdM$C)c+!?!hzQ zQcne_cxuK?${kGXatpEU#s|ExNJN_WF#@kebEc5?tvu@K3@1s~x6;meJlz^1K6z|{ z#eZ$hl+zGw8lifs zoqJ8@1})`Y%m{+B(Y@#u)e{0PdpN+{TVjOgtD@1XKTCWWCvSD|J4j2W#8+K)5)!cJ}5$Ov*!h zJYO~(oowzhqHE_-!+#|%5t%~I)tY4P+X$3*Uyibon~8baA+pcmGyRezqTesqk)s=> z@tDg+#6pkBZqyU({ZD?$)yJiRvs!nH zI&Nv9F{B4Rb8&{XN#|&lT?o9r7fh~c>yhMJFPYq33`&o>k8o-Z{xDf%6BCw&s}ho- z=JGl+^YJVCWARY3c;6bD>N%CWF=-{*-mT@bt&Wlht-DV`{FJ9$2H$!(c zqiI*ntzOU%LbbG6&!V$n@>>zd#;jsK=G~XrNhB!)urK*CeW`f=uOgYa$QXvd9fy}I z9GR2}OX;ZpkJD-3mLnorB&lVi7$`hO4k8RRQqNrDAD4zdu zoP4y3KqJQ~|LL$eA{+fP;Ub~<(WPockGwJ%uk-DPu}=+}~`*!b!gUK`~I zC*^zM<}gLxN14SLpA*qS+aE8T$i#P>YQSWz9D3(u3Hp)fOH%Gk>yM}XW@7y7RJp^?}R;EWX^E2@Q(I_f;-HaA2l9oxxcgm~$nZ5={>d+zJT_k7_sxQ@#4`6dSz6mfGqQosBm+nNK2&dw7dn73VJ0yJG}OMqwBi zZp|B{Df11Qn*8zy?=WYaDxc?>_Fw;Hb~6HdBqk8o))eYfmnA+S!~}ny490OWJ}AF< z28K$BK*Fc3g$L$}=fu_R=F>x;2&aEaj@19o0Jq`0Fx<`+5>9Rs^y)VNOPty&*E<^LC0exVL3D$&mSHs2EOa#rO$SUDtxSoTDK8|1{$}`O|Bv zfU8<4`nXmfErch3vBzqsklR0-#-|h+1Quv$(9DhhwtXfZYmHf0R zJE$h%S1MtC48|W!MMK@H;-RhSRB5cVps(Takxa{JEvgn|g;Dn}m9&-n6x$AoVRp{Q zz{BD^mXf+xf{&6v1I%In&Qju6ki$Lh&ZbKOW4RZvhcV+u{EaQ@?$HO91QHt4rivAR z#>2M_uF%8Lh##2jPWF92Nc-3xpfmjjeci zQWAQi6u%YNd(3$>J^zmWxi^ewE7romv{`)n$&;wAJxqYH4Ja3Mc(-;6_S~LEy80sO zUvd`smEbo~5YP3l*$ww-D4n@8Q-JOD=_UR1Ba&OS?s+jOpGGhGrBYErB_n}Pui1rO zk!f^}Up$rYup?&}T{w6Hwhcdr-(6D$UAeZt!HjdCVQKp&jD91{$9$CIZI@~Z&m=e` zn=X>~%1hbE1yWj{Q33aLdf^l|MV7gDi+h^14?jFH z;Xh1_!QBgQ(=pL9^p1luD=k}!qP#Eg-NcqY3M|C2`io)zz)M&pJB1H7zlCEq_+hzc ze^`6j77x7^VQE(=-TO|Sw;gqgdz*ioge#>HytoYHritf(b{zx1+C%hk=Sm#WW(5x_ zqS5{6B3$;Wo6FHJff-*@Vcm5lm~?IoiMsZX>+f4f>NSio((eM9e{&#KDIb8>hYXpu z4U6E$<;&dR)l?ceyOU>Q}?Jt6@rhdYohRylZV-+2Dfhhy9o@4k@y%$ZxcoTE`k zwfPsy&x#*)da%F5xqA8A-st-6Ixg0|26A`8$?H{Li0$Pb)NyzpjEfZK2iUKSr%ws< z#8TBxPbGdN)iBz=n7xrD_|!oE-XOUbxceqK8ukk~OB+V!Vied`PJ|6-bzqEA+<&%} zz|FZ5g6bzm(9Y+lh>M&%e@D)fuX_-O(^y&(z3>Kod@hkOkPbu3UO$+$DMv8Ntd#lU zDo=wymC%B;&nWeMJS@K8hc#7dutWM4jZts0nSQ^6Tw1S3tNa}qE})eExuXXvY_8x+ zeq5)Eum2=-WA$Ki>u{Q+VZ-}OUW0Q+De*6Q6wk7^1P95?yO`kx2r=*+f&-OQU?kPzvKHO zOCkE*M>0X-AS{8uacdH}CFRoQ7~)&dg0F^bpq~=2V5!I*ABKv@)G?z_^<+GZ-k6Jt z&qlM}nfI8M6%pWUmd*Bzx=C&=-VE$*Ut;4kit4w&<{a@I^_`r;-S=K6^Z^pwg-jgM zT@CcLrLE&fidE}A-#Exlq*7Cq&~`?3>$I>Jly02egD&m z@2#F#elZHyc%5KE4yT|?h8kOZbR}fR%;$HkGUIoID`MKIi)3OyTe3B_4;veti{$eZ zwASvy9_q0aV%rbFA|oeuSMYCU;)gIY$W0%LqCGgN6M7i*#DEyZreWr4J8&N8!>ZVY zVXy3Q#5%eGw0?z5h;J;+RB=z;%;%Amu1>x z!H?M`OTjU)gA^r}kjSlRWX*;>cu3Qh4SjSRopWP|syJulfK++N)=L0eQ8$zNgV0AY z(vYfu74956&UTzK7x1{RRsL_DaHaE7aZa#C(sVN$-rYZpJ^PwtoKmQG?Qk619C?Xs z9)FQbjrC?OCKnPjuN>mlcndTPTuIZChp*hiTwowc4J{TX@CBaGYkwT zCjFi<#LW=GABPbEy(@)n4Bp$N^{@pPwR51KR12VV9` zrA;2oX=Lj|B6|3sq_~r_cXqL&^1z{XO#XkO?e!FbjNT&EUAh zG{{Tr4GuTn68CRb;4jYi`Ecl2s0>J@QDfEUhD*EYfDfkR+m}*eHo=aY7b{J+*qxT( z7ro1xPqlVjDp}k<(*e=Vp-$;=R}Z_^(&0 zqf@PAaP3P)e7({T{&-Fl=l41dhl6C<(_b85_05%_<)Oq{+SY-_ZEto&YB-xy{d1L3zinaRwf&*IHv zoR{2iNIJ8cu9&w5AMG^4mCEDUx16>>_j1o}QX44E&Wm_TBVD3c>rDgLh~B5kh4+_< zTCFcDxA_LWzeSfE-XAac@;^QP-)ln9eW7oX1_k!RtIuRe;}Z-McT5E$!BUVXNQ3rH(ps-+2+YAxDw@qiqGgv%3D1?LplkdS!t= zR24oZ*B3aD9jE8mY?;1$lgWz1oCH&`5-a77P`jn=?i)oSojAA_G7fX{us z@IOA?lro{?W&z$F_5x?0e@VW)G=_njx-flSGIaj7LEHR17&OX^w{D5Ua#wd$wU~tV zDqVzpoX*DYpT>E4%d)p$&&O9c6Ik8yL}FOl#KeaDLGPu0bmEX{(D>1ZUA1E~>c5lY zFKG4SQ(vEBL)dWGT{;Ymev4xs^-h83kA3VPSt)*K&*vDJG@ZVEIvJ6(s zZxJgGBar%dg(l|7@awlUU|h;*>iy^_9_+2i-+LcHC++`?%5%KHZ|4Bs;KN+3T~&mB zA1~qo98N=2tMCGD#)8v(QD)vTOiZ09_RS157+C-loM%efq3EP_34U9ah^{*<_+IVe z`J|-DFmzQW1XWIE=f#cX+dpr@kxx=E(;|W&dNP(z%a8)wn-wLpH}h~eI}a_s9)wl9 zZ;`T^->8-K6f@g(A=C3T&iZJKUNc9c;V50c>3}p`-XD(N`TkU4q!O-(EXHA_pJ-dR z2qUa_VS^QqSNFc4Ek2KmD!TilVW}&BU`7bPZu1G=@LM*v&k2KEl^fiX6X%6KP2%^x z6(7)7`F&8zvVllx9m9Z@V6xywJIy(IlRCLPrw4X@1NC|Pu%?s8ySp!=T1O^x`?NVO za2|wXd&E&Wb3f?FeoBk{Z_p@{131k%hvaE1hWlANzF;nePoFwz;MGuKGtHUwNigDHc6wv2)d2oQrZ4X?=m@Q?t)K^E=YyB0 z3fsRi6;fh)vMWEHMfr1$_`aeOD^=D}rM>2$eCUBhCXu9^VEaOrxfXwET^oM-y$n6} z4`P@7oXig1uvI)hT2dUdCK5%ndy*mj4&soZH8}GoM|Th7=mVw~(cTn>F6-oZ=aj8{ zz|#5rH&1h3XX;~#jirX-6kgAaS_aD|@S5{pAS)v?EsFF4j#{QcXYgIxU;9IKWEzg>#|tD9nS{lDKswwB_E&x6_3 zefC4u_}|=vnUB!)>pFf&ls9h{%L7t9G|B_QPFp(2so>AdYcV>0ZGnF$4MELp9jFGHUF{YkIs^pF?PU z>Y+f@{4oHquMbjpYAr zzZrbxrpaiZZAX>!BJogXPl1oE^N+)4r#jM|Va#uy{S*hRe~R^ns{FUeq5PrWO1wsW zj6`+;r`|wme(lIeh^^SiJ83QAJ63A*fg|tYZP=51W ze#g;id{sgjzf89;e|(k&f4gm*gjbQIJfAJh-HIABkD+R)jQG??9(|HBnQhwAa8Ajd zcsv=*4?7dc$7IC9KfcXV(~!+`0f*b&_~_eV(r3{$2tTtB@_Y`W>;Ao%ax{rvSggYP zR_EhF*@*(J?!ZQRu#n|XC_3}oZSwKdL@fm6x$GlTPaKkM%pX3G!G&F$!kq11gdfQt z@~U(;-=1W~ueiGZKYfkw)0=sx`-M9(=DpxQ>7M8DR^@%P`8pD3*`KCf)w?kJcQK%-u(k7)6=}RoWg}qmFs!U+QYo9Q6v4sL}Bo%33Sc~bN+YuINmAhB~G{A$erlF z0Ic`_DAK!Aj2jQk;a&RXqPLb8fBD%#zF@YA2&LCrr~1$c=a z(KDBq=~s^hUB}VE#+y#(^N6m+Frj{(el_U~-;1sNHjDGYo`k(?RIzk|It|$P0B3n7 zP}i`nc>R19#$SrURW_3$B;+7(&@cw-rc3csbAovqJd|&_X^(M-o?*BAFmwzs{jbga z^ND}gM6eN_nQU~&vldqTKK>?}Zry|OJ8#g>dq$GO1%`b5!2-T?BEp_U1j`_ z&mw1kB=fNvn;a8^`d)K8pOtAv^i)aZ~va<4C$PB1e#2XX7QZ zFL4+@6Q;wm!fi1ODQpCzlp9J~Kt1WTzu>+`}F_CwivX*@*mQzBfb8sz@ zuN;RmY zcN(DIk36DUvV)8muEw67nGf%FwS%h0Dt7#zY?@zjhU^iKS9e}KME8H+3v!8XnbT*l z!15E#aLwm1b)F_-FVEk`PWq_JPUzNR7t2&aOl1~p9zBbkyRS$NpP7qqA2zUK?88}w z-{ow#YCqfg!`s>18NTe}?q76>>>}9j6C}@@PAj=w{a`%Y%;BcyT3 z!b9+6RxkGUQ5CjWjRS>M!Z(=Uf3ay}*dy!*sJAy`yQfcLt8N}(4c>>bq8p3JpxGK^ zZK^sf2x%b3(-%^oG9B!idI~lhOlEIg`wlI4K9XO-pP|g61N>6@(RX`Sfy+w|VlN*D z6Fd4Ly*Z2xG*{ui9+Bs_|4`=*lkZ4m6HPgD5G3+R%GVkmj|8=lfTySNTmEn0b~L~_ zmWOm7S626-5quk638(y2*x?UE>`;RN?8n~9?20Lo>^jwrnDE(%mEbtc4}x%~c6eBm z$+|9cWh38z1mhoraaWE$1bQ8S#JHiX?*&bE+k?AM;baY;A~Rr`>YV?0kl?VartGuw z18nQHeXL@+GZ?6h!8tc(k&m&~tkx+J8Mtp4ob28Q^>-a@xGjO)#zq71?eS8M zihd6(5~i{p3YP4R8y3__r5FMgFY* zLSAY1f+yutZ1LhtP|v0T8YPjKsl_Gz#&G0HU^MCli6~_Os3q6|X0ZTF59I9V(9nQ&c%)lwkwdGLSay&&u6d$i55MR^lYi4edm0XtkRg zdGc~2xW0;@^3SJ%MT82D8aj|DHA%stmY>kOuO>TID}|lE!wEn7o(84o7GD~DiZayc`nOa@(M zmg9bhwK(zKWehav1z4&NQ+$m1fp$Sy`YROcqTWb&6!ndD#25EdX~e;A=rcovb_agY zgsA!0^JzNQ`Rf>c5n0AN`#3WD-3qW*$SAry@C03OWieXq6vxzj(!@YN@wtPkN%;P; z3opu7$5j3^#vL7r>)0CR?$82y=ga}Lov@n^GmYeH6C-fP?s}4Xs|a@_FGFr~0)J#h zD}H@9g;&wdu}-PkOB|9-_@Txu&a?3^Zk+pp_*LHK125U3O~X-Mf1V;AGa8T`CZc_p zWMb>CM1F3>41OxBMW5!6=HE8O(VC-&uxGcGq<l(345PHo z>07^A(t~%z%1pEW#wg`|)c6hNUVN&;9{gHpgdcfnzP&+{_fuYjZ+!Ri_tx1`O}kaN zBKsoNdie2MKd52tDHS?+-)SyVPL*Hh@SPmnoR0G>eL?2YGpaH?m><>j8Sk6CroX*( z`SN)opts;KeD#=GTxd~)3Q~P}bGJx*Sg{P9s>NqFmcPd3>XoEh<{C~y3v}%8!1zNq z@XG}k)Vm*!nip4b;cLub>8D*09YOX2MR}LU=$kQ+PdYG+6kA@#xm_F4 z{ZBbH`reMVYCSL`zZ{*nJjC@|ZFt%H2e_K%eR%Qo8Ym1shOaEr(Pr#h40Mynnj?e+ zv;m)&o5kr3Z{e?PdQ1n6A0g4JNOzz%zMkSn`xbn}Y{wtmn6CF_NyQINKe2; zDG#a9CXIItIFTkXrZge-@*Cu?X+doHyY?$LqqB!vB~E;x=wz~-@Khkx8FA6 zYr+fgt8bR8*ug1``noUQE1ZEm|m2~u4xs6ZW{toNj+R%{)k?*tpCSSDm0aw zX_X?ZG|I>AF-~;7H;;F=Oryh6?+M>Ue&=^b!XDu{U;QYFOJ2)-eKk(83v zXXt=van1(qY51h0ncKSD2&ZVP@v+xc`K61U==(uM#A9L?ojN=jo0QD>aTlublzUIS z89WrG6h%{ob$?@u1{m$Db&&xZ6jfRNs_>; zgm%mFeOU2J2_(N8GadNgSFKq3r9yZfo_!PpT>r)jYI&U!3PE&ocI7w+NFOC$94LmWyjzB?BW0pgxD}@yI=?gj|+pvBZe_|XPF7SS<5Ao%kQc& zDd0rOwBYBoO4|_a2MvR&Hch%n`~Ux*Z&mt<=+S`d6#w!g;jw7KH0s~y6s-@@#?$AI z;P*)Zm~1ACR*F8@Oz^&oXv)=6JgoFdP2CPCU69a`_bM`*W4gF}ha zk|^Bay&GMn2QVuKN8=P9H>#+liD3@=xg#B#n5HT(=;gd)3uaVE6Q=Dcu?Wv3b-w*1 z_a$_sB<1r#L7xKMj~geT|H3ldrPP2s*Pp}%({}NGnuGauQfCGInN=>rZ>I(b>*tcX zJ$jVl@sV4}2|M-w?y0Q3M7;8JIlaRjqDRDct`(J_3Tf6CFJ4yl@#plT!`1@USQ7W9K zz3*${q`%yztPvUn{Cd-G|n6G)xd3JrF znVqg^H&h(Q^(`FkS^SO3_uYJ5;Pp7^4FT7bVE-f&iJnB>PxcD>Qa-*==wBrFB&9}X z8A`QZ!}+G&;v9YT%%fiUR7a^dt~EErXRz)Rw zO(Lp;4icBM(wN@+DE5g+W%g{F1Fi8x1bQa#HgS$cHQaEm>EP^7Z9dW2PF-2q2+gfWrS)mt+uucU+WEAwGpnJ+5j z$76@sH*@av5pap?NPQjXfXahgCH5BOZo5F=Uin0uU%S%5k00RCszQMu$+KIP(!6x< zjo^LqCAsvRNc=z~m33HxMj8j zSmeOz_Op`kE0tdRkbdcJB*1J9aU>S6|HhZDaMi;^Id$QFXP+w4=aI7T`&6B%#4$&O z${u}cJ>O*nzg@pbz>~aB2a`)tpj+_Si45-{C%|_ZA}rTa7yMG?Lko=`9WJ>qT7R}0 zEm~&_btV7L-O;3|mdjbXU$#ju7*afaehH13TUK&iYdBVRcQBvFyU};fe`9HfHVwe; zdkq4u$B#oeBV7-Pj-+Nd$HULd1p-dCYbNI2YY~1=zBe6x(zD3O-hwYl1vj<}ZR@h! zF7#9U!b$4>6z!&CUHwS$B#_|$^D1st;~!C$+R?Xm4t?9{>|c&RU^mHBC}g1 z-}EX3{F?)LeAdw^;8}ZmBVHc8iAp^lfw}kBk*gtuK0KDhh53ikn4FOUJkxtDeK_cV z!2jD`H-&l6y^iWVwS~Ib4QJz zg}NS%F=W>13>vN<%$TVE=86jcaMic;=pcO&$rmrLN#M(?G=ce{BfzhFm&09|G+w~9 zW&VEQo&-j1un?~%CHXBbqEvAXQTwz!T3RKIh4cCfcs5lGz}wv4^}VoGeP-tKV*<~z z$s@_oofjDI(NDRxSN71-yIpj4)j@7tFrxG9f$(GdMb5R9L{a80!QTql8^reI_J7YrRog3>nm+Gnqk<5( zB^oAj;fB1u!X;dg#a{B?sA_J9&DU%h0sqZT1Ag|>rvK_na8GLRrlaX$#+=FvFn0oX zVgBU-!f*E$M-rLzMv%*CS})=z_l5?iu$c8*l^h(un3?rLr)9zZC#Ll@4N5 z{y6gEFU)CL45xMX!f~fT@UTOU?OCnKc84#4-8=Khtl9N=z1IeC65WGuwGm|AL=84L z5#fdEK=hxmA1d>I{nf|rU`*3iWR5xfp$99Kuz80H=`>sqa=Z3mn%85jJ|)GPZmWiy z6qr~o7Xi+)}_n+-DR0)^b?*ni$ne1CVjglAFD&U@fFZwTHFl;KPC zigAGSOZX5b&Fy=2AD(VK#bn*mfhcjVpHE%4m|w-ERPVY!nZ7d-au$XOGD~oLmofmm zF?A?A%nr;f0vXlmgGnrvhPe7Puzab?pK5I{>ED1bOY1mB%z1FRVbGLLMcK1c`8qGG$*Y`k=v|+^9;XW#J=kXA|N2JOp;)X_P z*g^&g{Lfrg4K4F!aN+&I?D$JN=*~CJ?2mN}oWIml;A`a8cLMI_tayyL=fTa%?gu@e z<&!;;F}Uxt9RDDA2%CFu8ESm0HXGuV5Y4K!{s`tfoljE63c=I4dKRK z&0s8k9O0yz;r_W1Qpm=H7?()I__pip;4&MO+sz04sjb1nwC z#N&s97jeJRPf%>NBAEf9B%tXYR(wdsw9$KE<5MHDTHyyg*0+Gr*&{Kl#*kInqz<|F z!@zn`tT@K#G?D0aNOUadRD2;Tr_Tkmah~L7cNOfoU5M{D))BL%17K{B6MPzf3;Z%_ z$TH7`aCAr=(Hqmw475{1izTsCd0G z9=$i-K?yC(%)NM*-$k#dJ%aIfdSIIdhYrW0VRg*{!Co_zLurt~bTnm(!FuIkqP*fQ z>~lT?8{eKFd!|H!rD`yaIuZzd2T5V@;hrE8=e~Dq7N2b!&ES|5Eg13PCnw&{Mq3ux zG2heX(;|CkD9W)CXnR=LqVJzjVi-0CM=#Q4qoWSOz&BsOH?0`rOfz8B>g{m5!h<%p z&mhjtpGbwDB~1Fc2X<;klVRsyGS{w!Lc3EhKGG_kTljDW&R!NKye~A8D<$})BxQK2 z4xLi969Swppv8PQK0Wz&uH91|OXl}I15KlP!ib$&e2u3fE=xaFV&*zboKqzMv(FmC zVTY^K;nClIShPb8^}iY8+lY<0=duqBlTij2Pep#Py$COOiszm_Uc=F;20|ZBM6vabW-6=zP1W zXuhTtDcCrg-89;fSGqTnZChF^p2K;E&9}6O(ywQjVV4F|tVi&x;wNC-mZ9`_USDii zxsUqw92YZW8K2tcJ^t}XlkF9ln+@VQ3}k31tj%VrSvjg1GT zAZ_0DZX9^juI4r@I*9!S4wJx%B;{W4VaA8?<_yx~;qCMgI>{>l4UbGA`e{#aNS-?W zUN{8reD24uOIQb+Di_i!`z(BU{S7Y7EI}WaF?ep6GOT}e6}&Tr7+}%O3GZ-do;U3H z@q-k~nnHA01>}C*iJl&@z^>&4|CP}FN%rvz?&V?V2R(u3;;?jd@Z82{hs7gPD}@V= zu49L}E#XtmCqQm#5byj+huF)OS|2gog<<17=-i$?=_8uL%N=fnNn98sr8b+-j68ws zV|1YLngYB3=TsO*KVXNE9#6-O#pBZ~Sgno>xIp~jgzGrCY&L=~ObJHkfCBoa^uFNt z4*DzMTEuADJlcg!cU{AYMl@jY-UQm(?>X*H&LHNr1;d8#BHQ;3!hugt;&q=_sFb!z z;J?q54_H&2jJuTuT!LGk$9{i? z`U8yifU(ktV8$mWh%t|aZ~AFdtj@+h)Y&6nbi=Hoh72k)GXg9L*iu(aTQgv!uj&m%2Z6$@xW7 ze&WE9p5pb}E2J^5H!WE8ohCVELcp<5@u@UC>2I_XGao*I&l(r_r6GH0W4uU^@t@D! z`Voe^69_+X#zJ2G#(oeN$F+<p>@`w84rihx410Qkd-E-yzEy~Y9b+pc<2=!Wd+T}mQ*Y4M z+m=M1t)yS2>Rwo!{DW94EQiJJyD_mu zjUPDr1w<75B#o2vNo$V?q8RK;7@u>YL4ka#QGCTEpqC67M09|4s4^ zf2MM#?zmy1Ja3bIMc{4ZkKZh=8~#%KTpTiZv+EP4{EIai|}>vWbn&%MEivm+Wxra}<g23O3fXxyn)o!oqJ<{{=$4dSWbvbx63!`<%=u|joSlDz zyyzSZkpXkb_F*NYzGFPK9VLq&zQs`66eX1SrldT!+zITmk<3)>0pvu?0jfDAi&@wd zPk!9FMykJjB{Fs2gmzW0Ih?p|U4~EBo+pRP_R_=gTRDqZaXw*3Y39b&W_oy!p`crl zOELV46z{za7)- ze;h%2ajU`f@g`0j*Z?|1K9EL@VdU(LD4G>m&HV^`MwA-9i{oGS!3$9o>>8&g*m1=#m+5HwO1XqXm8)RlxjLu5Y7hMo4c0<1P$ zg2BSR=%1nhcL!y_$mY?k{A>p{>p>Pb{X;s*^O=U-a;qR=h9{jiVgYm|8xsBYHDsgH zVPd_{QqYl8dJd-hULgY;?lH{H9&G-?BVf3q98S3PqCL;gh67ROX#czSaK*(k%vLNw zy;k9STa=-64bI6&2=b@I{C3!S)3!-MBqJ9a~&RCqyXD>FG|KxqGknQK29ocQ%`uXco(OYm1MJQ;*1iF z@e^Q7WFn^=xCabY*}`($k?^*67S3sVKzlz7p(6%uFSZ*a%e!_lbk9RI{&;;A#%$UL zGA?!WE0Kn%$~I0!3vozqVckU}p_M1wOzc`6ZGP$K;CP$GWPnsTI!(-kZ62wh+A@$= zb)5xcU6MdmIz{mBGq=Z(0~RqPdNPLzH{_807z0VtTk%+7HT|G+1;)(R=5^Q2;keHdknZn+O>+52XaGbsf}2j)Yd%u_+pMg~UeKu^D!61!-K=Z%iYE2DQ$ZE=i=Fn1y) zxu5M23x}A=H0aA%GEjaQai9B#zT0w#$qGR6JXtlJDBh>hU-295e?toIUr7Sx0!{Af z?|ND!`-WcfvB1ObL%2|jMC?CR)y6+R608+35TBVQ7%M*aWci~~B&wDS-?fRT#OKi) z2FIzPK^v*pn@*3^ZzPkm2C<5_%IWbBfhDnzH$jE{~2O-jCA3Cw&%Fmu)BZLu7<{{}=w>Z`acmT!wWC^-qpNahn!wey7S0=yh$P z(>|Bzt}q7IS_QUir7|ghpaC&kdXX99Y?;!QNFsG*Y;ljdcWIuLKbfVLC-h^Mx6}kX z#Fm^BaD8wY1+t$eLuWr#=1uhm5*WOW?5+ArO=F5kzTbX$l`$F?j5rTfcOU&{zkj^x z+|veWjpcNve>Nxk;wViG6JndCeDeah!Gl*}@r+pF8?>8Q$BScMMwvrRZW!1pFW`21 zJtEpk^KjrrRd5lHyFa?skb8A~S)6?k`b2ia5$++mp%9HR^AC~4pvRp4C2R41ZSfuh z=Wt?CxtG~oB+k3zbd(s?>XOQN12IJLRB=*O4}rg+uNRBXx~?W=Tl#~_Se_m~YeItD z~tm_}x)z817xNLY6eO&|5v=6cF%!H$;eLASJA8F?fXaYnA?djU?slijme7&eXxJ@l8B=#V9%jq z<`wV6x$fzQclJzx$#!EQB`lttelr(Tdw*wAUpG_5S&@WP8Q_wxco^5Z8>H5cw0YZf zm-HNap7_shA`jP&qS5KwxRtf(T($p1+$%pGEJm0^-z(>dp?)&C`>rQ5^PLNPrab;? zy+iNb55-*rH6%I{-Hr++rdl`1uT?rQXm>ayIBCMAeG_4k!hAX*^$wlAcO{d$?Eus_ zZY2uGFL7tbB!jfRui(QH*)@7v3%WOw$|ob-rHiMD_q499+C`2}h~oAi{>(I{-X`i- z7E%=FTPQhI#r!|6zB{hxFZw?XE$u-|NJ>$P>UGacWmTv|B{QOs5hdC3Zl|S;Hquau zmel*6mqZ9zp8 zlSQQw)ObS*dW`FhnKiatEpHbtDs$nx7F=d91Fz+<5UQsKGp{)QY`Ee#tMN{qGWW$4aLCI_NSYlAJweUzbM13{diDQZYR<3SWB-pTn4H82gqxUSzNQj0PeTa36!jNVImc3g^6c3p=G+Oc%C|0?!r73 z24KSK(=;IaELOfd%E&F3;$r8PkjQ&coJ{p9Zj9SL9JfLTTlh2Ef{TM`wqY(!JavWF zU+@s_e|5Z0joQ=upvQJu^u;XQcr=T2K#Pmwb(laoK$SbHat>~tcEOST)$sX;Rysa5 zg`V#v0pH37isiAW*I%mFj% zux*G~rV;Xc;B5FJ9NN8_t{6EE?>4nlbE(bn;ch7GHdTQ}KZetGbQ8-#M32v*N_^s& zhO2G1Kn`;Z@4g?7@6Kl8S&N-m-nkUjGzCOHFaX|46^ZAb4qGp#|JW0_UiJcN7srWt zI^tG9b1vV6h>^$W>q|0R*WNXp`vey}dG|O8{hbXYviINbNt{2IAueY@*CBc{{4sqH z=Pxc}Nrwz{r&OSC>Im-BH?f|UV2fuV7Kd-eZwDgjgQCmg`^s;IL(+xOl={>yZ6^WmsOK8*N2HuuS6L%#89XIGY=J{+A*XyI!G|=;S4Ke|4=s)Q> zeti*yH+Q>0kuHaYI$3!1mOSfep~Trva{CQt>BeBF)gVWvIn>8>~fDZuN|OTGZEhf zcTklmNwDapfWH_kcJ?X*cG%`Xib%IJ7$z$6t?sJf6_N~NII+h@+8&Pn}{)Pjs z269(!d}dmX9X~qboFi#f^u=vfbHFRENSrP?%MpAwSl~Ok0-;2zfLfN?Lf8{~^x~$o zM+$uQRq!~09lhnUNswXYZqQ8I^GBH znCpGvyJM%s?fEbL=kJ;WkC2RsXXawN%?8f;&QN4+#v9nONz8`dK-as}e$DqmMG)#+ojaBmvA=IjvTJt&i5e9$-Y3I z;|nnjM;^tK(|0;Z%eKkHNbPt3ug&rXTvZh>I2gIKnI`TAC(#6*qf zar0Cf$-3ap;1MKEEbWgn_hffa{Z-2`GG~~$?7!(x?7T{g4P!BVWD1n5mVow@VYu<$ z6QYRcXzT(rID4u$=8Wr)t})ZmXZ3ZG*(o5&!>7_2Z5$TG3=)+k5QX@2h2WmBkDLo@ z!&_t8nBVue(F>bh827U2;<{9kx=*$qNFWyjRD?HY*wd{hLgrze4OqJPVw~k(diA;u zGj5b3bQ+|Rse@yv-U?kJ@pLb4_Q)n?Q*3dM=~MEx%8tpY+JWc$@cYRCMK-{N$GmcC zC&P5>sr-QJB6&#s4W+){$#_oz9_r~vShE_weUf93huZu{R?c)7CwJ{Kt=NzQXZp4? zisk_%$@8A5orrl7#WIgPSdEjOLPM)_C_hF82fW+~_dB+c0S+7}dMKfrcOFFEZl!zc zi-kt-CCJ8lS|fQ9R{l^fJ6VMU|a1BCIwN%dvGJfwMMd_kq(tgN6Dk1 zKVa=*Z}xX#Ipn+>2om%dY#AUiLt1=Wx#PY&-ZE1b{IUT34Ap!lIQD&liQ}H zB>3YgSWnBuI?q!yLnNEjDwh4vabJzp^lr7nyPnAwTgeOmyLGOF{F!@=S zh!?@*OuM>OtN8N|uO{)cchy{Y-`^PS9Ggtc7yE$mV<#Hl5>BSH%5gf4{_Kbmf6lI7 ztX;t#3CW~mlDa@Ic((A?J$d{~U)`l)uwL>B`PMgzjF?*fZ(b0+uROaH!Sx`XzV_U9 zs_F`S^wk)44A94h_cri*c`&)NcQf-L+CWT;DENlA6hz-m=CUA=w6)%7cZ4C zqIa>MPGRV$I2gQ00`E;ZM?yDf)J)LaFkoW#YgqF0^rJG-i-U-e)%cK6i5!6am z4ughH#DF=;^y#oPJLffj^b}fu9VTa8mJz$Ba&+_B**N5~9M`*g0g??i_;Yn>o&Fsa z(KiLZ))nGtpXVet_!2&xV+)*Tr?6F$$C=8CpiWeUzTJVy>3^d|ef@;y^Uo3I<2Cq1 ztBc;AeSt2Y!(;LBbbpBWgMZryJxr)Zi1uqOu~TOvzrYja0_>8~kcs0oKzhG1mMBbGrK_1b=)) z^U{2YPvskNdgDooT<&#Ah)9#gH1A5B7j=kM3uZ(4hW?!M+j9Eja~+Y0-3nTb{Q5Rh z9d4%d7Rx)jbQI$nLxjxJ4B-XoNp#M~bK>$Qygx@S`TgNZM9ZbYDdKjadP-K5PayhF z6e9ZCW64%idVj@W9KCWkOw|k|ucB{~&_@n%PC5$=I^<~Fi*v%piZc4dDwitN$3aS& z6a>T**nRyX)4|*Pj~uTr=DkVaCXV}D`*YTyR9KEj$sc{<9~a(}{ z=UxmRStEQ>Wq@RA1!FtWfv8KL5S90jul|xDP`uV3$J-f;X*xQ46b##}1sinZVVuD- zni&6;w3s^5DfeTjM0hs&dWi1>8#hBxrMftMFW*?&YV?ghs=i7O-%EmRE9Y|RvVYom zL_-ZNek8}v_~9s)X_$R9IV81>yz@E!j|RyAU3ltQNVCeqz%!wNJmP9`{?kungjYO# z@mD7G>MG>Uj$ArJGFmJLkt{@^5ys(+1~*dGxr#KM7$Mptig*$na`K{0w?- z06%2S7R%(YouW9u(*|k$_WL3+j0{9yum-E>EJo9@mpIMyU43|+Ja=la>bu7Ej_zZlISPnxJ{^){> zXv|{X*(~7ir`HX)hUJ$PL2i@+7z##1$MF=98$3oNKZ%xQx1lm=BTXKrSQlj-OTXIp zr>1NCMdO{A&%bf~fpqqmyZ~Ca-@?k<%cz_FL*my_BQB#?af~>gk(&&sj3%+~c)N@= z`%&C2y_M)TU_ajc-iteuy91oox8e@_czSKlK2*y&DBf59E#FE(8uH2;z(6AbUZ0pv zUImRN&c9v~#XiGW-jEb7%P6t^RphC21n+AxdlJ~{pM{+Dw=q#8PiWt|kHE&afjXYz$_dAIoYt=D;2X97F%ElQbcFrJr{HdQ zvA8|-d&e>9ogLJxe1P3lJ4+s`z5xn06vCKqKd9=RWyt24;sKjwaKohshfcZx1N?co|wR9uUtSc?v%%bLJ2~6eLECokry@T0yGCNb z=h7hA^%f*8-WB@Bybz}o8{!V=vW;&q)@lm-;D)d(Ocb{Y zUHTMJWB)nqkj`nmuaU1JdIXOW#=ymougRm16uR2}7kM!xk4QasXU}CGq;BWd8(BTx4HGK&t=Q`3v-bU|$d#hddgHnRC zrqL;*Smu7#5eVqHTX!n(F3~knhN#Pt@QGA_g1aZUWE*1j`wd_m^p@Ho*$>-z_$NA`#m*U6DN-wh_UDI!iU?r*!Or9 zZku01hU{935tmdrsgQdZT`2?m2I#{Ru9Nttx?_O%8?x-Hcr2C>eM1y196C@G|BP4! zJE!%)B_$bd(g)rrYNa%>O>AcNw`M{0wm~9UNNj2uTf=q>4GxO!%>I4%%Ev)mdXgrq;G;*S zh4JM0r)H8UJ%HMt%wV^gcETc^6zp=A0o7OO(D?Gt+VL;Vt%{dL>kr}{%8tZ`J<0zWgYMlJ!#SR5Ktt(UFz{10o>=n<`$YZkxys#3zPQxs z5#&|A=kc(t(KbYrdtS!l`O25FuS*Wn`H^SH`Q=4qBt+nyiGf&~(+_8SHRUA6g>z|f z!?`5KGI5(9)pUfo$f|Sluad#B_W|zF>$ea)Z6G^FxgQQqFGTmzlem=d3EbJ)aWMVH zdiKZkHE35f5!?EIC;1W`usBPE9A5ACl*=yB zI!#;$Pd}f+mwS7Wn1uyg?7bM)G?X{}nRFIo6Md;)n+zMWV*|TsvVhCMWG=EilG`** zhDg_>GTv!zH0z@ex8qSJ=E^DKm&9(g8*9Uv?)*s41PW;U)eK=Fk2`g|s}e8$?#C5I zm*5;LU2gO|6|Qt%I2O*+!^DWu*!#n9mfmO*o{tW}GbUG<*B9SHs;ew0E_z5c@uu)d zvH{oA*O&|X@C>s{2p71e5D(-`AWK$$K@XE-bisgcn7805o|^R#UFSt}leVPe(IO>W z#h(jBr>j8vqa4V#7W?Xv*fTpA6ujnOZ^18|`{O3OxU`RbF;g2SQXiu9{2lC6z5*^m zo4`7}n@(@t56VmLpbF9Auy`L9YW_xR?yj(>do<^G^gMQy&Sif_-lN}+jbr7*al6MlAN|)FHWj(IBhR^OXrtKu~pu-+{F04m^St(F4XdX1&i{yL!&0a z!jZ2rydxhq6n$9V)-%|kpa*(8p3#eUyl||!6gP7n?~eCv4TkgnZ__^SrpJ%**y_4{ zP)bvVTdJGK${bjMLo{VrUEY5~|3&BR4#fAv?Bmzaf2bU{ue=bsRSy5^I%KjE*V%m% z_3k)wsS9sG=xh_VMt_}Msfsfvf6jnBy^sfHPx7I5RSNEJ(j(Iwj-uDWI2<`S37h8X zU!5mxaehF1j`(vr;|;lmY!PzuxoniS(y>b@d8}Gf_E_x#cP_(7hdW~N znof+lPA|l0PTD~n#YazZKjo_ z480WBh1(o|!qA_;Y2KcIR1u4KGgGTXA>mTMU{g?r>2&JBCmPF9v~r33P(!t|+q*>Tp2?66fWV9;DeJeIYS zzWhG+Z{HQg?)w92aY8Iix;vZvREQ!UEzX0&PzmC;7a9a{}Y@iZyhK7^eT!Jp%;)&pnuF>{x{^B0x-Hzd~IAP1*M zh~^K$R$;iUaiccr+#!dGLEb!;j10tTXTn3XM5^Z%3H9du-5H;EscbXdYcZHDI3rJu zOp>9icRkJ)ysB#|xk*lseN1{zY^0qZuai&849s&2LuOD0CK}aXW7u9=*K0UC%~VP} zriSV8wmGY!#baPdY7_{zW|FTV$<*+f0F7kD%EA zFX2P=r^M*15~)k_CJ7(){?Q!y>o*f)s?AmvRxn;u=E7j#FtD_n#tyo!fMe&FadVU3 zF)ceU5cSGs5;&=Xn!T1LvNFWVl9YuY9n?> z9bjda3zS>x@N3N)5_@g9c#U9PKY)&0ZVDq@$BCcUN7OQtryIf5R!Y*HHqfYIKjHHF z-f-=N2@@U=4DX^fgl3l+FsLg6^N5>7Ni!I7HYbX7C?pzBOcA$-v3^nX^`Hu2(ac)r z^mA*_+P+0-xT+MA)<{#YB{N05N*Fb4ht%PIe{>D0nBinqKSV_f%QhWU21X zfAU=RJ_B6h8sOH~m*R3JcURS=Z6Czjwb{`J3wM($Q%0;u{-V?xhC}i{0#UEZz~=NJ?+*Jsdx{tPKd>wUyk8`Vtb7Gc%}B<@?`iL{DN*u*&$5x zEF}In>@Y4ailzkjMN!!TlSaCA_ITL8B%V`LpUg!~K8R)#x2e$S1YK+CO|O7FH)=p9 z*?wOVvSl1_b95n5P?(F0w_L@O^Sm+ibD+@FaXsvjiKR22r$I=P6BE9k*I9N~5tq+@ zLtx*h3s(phkvjqghZpepdt+{5g=3U>j*v|Kjgj}x9pFPHuoY@0Tyl?2!-?K6L%>{gY{gX&80{OABTw0qny(_1S(b`&2I%_7K#oPUk znrVm%Bv{;rH2T-jpoQ5ue6t1@(x$;_obC@iE{rfLLLP(j)A4%XQB?B&2wcxYVX$i{ zZy#`-rH^S0kAv%&u$pw2!Zq; zLZO+u9DTr#QyGs80p{3nAtsWfX<8szb3q&0ttO&WurzPY=*?q{R?w;XVRVjk08Y^u zM8Y{pP^&%|nU)NdP!p4I74r%l4x}_E1{jyNxceQRmk+Di9!8L2?{Y z#N(ZJxg6`@;tdJ*2MD8*ji-2gsw*R`>A1`+@~)@|zSBcu`my-bC>MJQyzt%92!4IL8m{fzZ+HEZ8@g83(*2hi8uREBmh!m$CY>tS z`r!$x4zMMQ5*cxSJ^5@94*qzRSZdCMhcS`DA?DMl&gxF0o)U;sArW;2Z!)Qy*-xQy zB5#*^&jnwFjv|h2j49;7cX4F>Q_$O zrg6HU#;>tvtsjPm5BcJ8-|-}WC!x-}2a>}ZchiW_Xz`e^B76hMa<(8lhc%NagOLky}f&?EVK>9Lm=|{rHelp~EK%}^iKY#v%7chSALTny3n<*c4oSrZZ zCKGpe5dSZ_ka#Imd~R6h(Ms$Fm~EeOabi># zw#E3M-n5m>S;JiFIL_WqrFjVnI9bEIeH(?&KaxdtD>3(?lUTO3THmN``FP>W^S7zv zjYlMOx+hw85FD^NQdpjEk56OA;p(yb=n$80Bqe4$qkXq8I0Wrt1}16JjHXz6*ZsF$ zcgaGgZSHZJv070iUkTxZ7>aJ)!f{)YsmD26oZxvJ&UoD+$N2LIWus2Ic>6GHIipWs zeoLSo_j!Ea*BAetdzkudpqG4PQQ=w%-K;SP4nDmAXNLZ{H?Z{a7n&FjAyfLxh}-AV zy=R1j#^i{9KM!!iO-i|B=^k6MIam(2kDQ7JT{F5z~pQrnP3cx53x8nU-; zZt-9`c6~K@QII2S821yu_Hv{dBa~6O)dIg7jva|3A;-fBR4j%#TlNE%ii%*f?r~H}3d91ijWcmGc+2d)#cNJw!e}!T!ohi+o z*0IF0Q<>s+8}UpBZGT1JWh)1~uKx<#TbDq)yS*=9hfu^p2-^)cMzuQ*@HDBzB{v)CoSi@~OIH4H0#35RmS zKy6cICri^*)+D$Gg z#tOSu*pieI4c2-}Ief3qLzVCR*>lJ{YUZE-Q~cH0;vtFbUiCt9?EY_f`JTrr@_7bP z^RtP9k|BP-Js9S9wm|BMKlc}ZWo5o33D#=I(jV{N*g13RG%j#EjW&8l&3h)YHGBB; zn)%ftxd=qTJRpr(KGFSXBkv!5#rC<-t?+^>y>9}y&ivY9Hw!o*!Q)1pKLlDWBOzm5 z43yaZU^d+HMjPIgZ|2nwQd3$7=hfuc^_!9C&YceTl)51|awF`1*_W-FxDxVi5_ZAD zb~4Vm2OK{&z^VtYKw@Hbh627abN$-@BfW|D2>5iia}6(q?X5oI1DS6w~?tO`Eaw- z2cNAw3zr}IK%8S08C<=4} ztbR*}z?F<6ATeP!-d{Tf-dT-?f$DA;H}VPX?hdC--_qd3`U5ah)7&oOgc}@-9K#;- zj`}BSkANkZ|Ev~_%C_O=x7OnS*WWjW;j+!3)~d-w$l?Xh_ z=mK2jpx{;p$eeGa!J|81#-Sp(Kl>J}d7;j_zq1h2`8ThdQY1BVT?I07N6`MD49%Ps z3tcM$;drlXwr0|9c6>n=8-Em0Kl2T}dwU1&k(t8$&-W!}7*58TGtowU|ax98~aIJgueK}q_LBQE04ReFFo&r{)~aFAgn~V z$qbmA1vQTn66&J2}&7>*LLBk8_aQ0_5JsUg&`o3r*&&JHcdHNxcH8~p&U)jc8 zyEdCMEck#^7ARrq@RJ<}TuN zO}BIV)Yoweo6hsL`OVa2=|1|}N}g*A?#7p%rQH2RmJo1IpG*neg$?5c96PoM+cmd> zhoBbUNSETfjN>@#Xe}D8wZ>jMlHg~Z3|rkk8+|f=(v;MXLQN?hP~E|wiwu2AT^p3S z5sy{Kq$`ATH>xLJSNF$}T0^*#Ew}K?06*OKScW8D7b&%N625_NR-cNXUr(Q$UB9LSa; z{Q7hf&*e>rFWK|B$-K{}b0h8HDyz$FN^c`k1?ogn zUkj@|qVU7?R+Jc7hemB9ILUDl+{m@vxbkx{%$SpgNe|Z?k+W^aCjZe~($7{r?bRFe zRp+3o?znX>R40xoK8)(Qk;J+B6*<{wf-r1ogx#-0gSmoX+MLh2Ng#jPU3et@ z5Z87kiz^RX$*q%~Nta!6<_z9UK-Cd7SR1O&?cSPb_Rh& z-AD|v@TG!|Dl}ZEjz?Y|K(+j0?(p;|?qbDGZn)M`=+PPsgH|m8hkzQ<_#_DOc}$tx zhj4$44#!zc1D_fV&SJY2y=tYxRi?ZqLD9$Hn?^a-D&550IsL$>WEnG2<$`#u)Ju7c z>|Hr7c3T!MtUF7-hDix;&eq1xX$?4gKncFmp3dpb>c#D^eJU)Pp-Uc}kYhVC&SBvB zSp4{HGAfJ@q$ZYEF@D_y3=dQmj!@Jft35{`9l8o;z!j?ay^J`{RE5H28Zi9IQexz) zjHC54Nq(U-eBj4W`SM7-(6@kcOL}1NP${-(MI{9Pd_kIRe5v`)AX<|bO(mugc1wm7 zy5E+86)LU5KGq&EZpjeVD&K`2?&*Pp*2{sBOACxoPGsz{5iq|ISYm zynH2=`o1H&p3U&Jy8!3@JjHyRUWf|CPT04wN4Q$@E|bOkT9(x-B%7Lnylkw1z;m|j zG; z{eDHqukpma!|Exs+(Ja_Kl)3|qro9tp1XB%BMtCd&Fgii14x=Tl7(=- z&jQ;vR~3AW^(x_kFG%S7-VO;ESQS@-! zB>9ULK3)WF+i%tt%zj0->>dOuZxfhd$0|rdZ-#`g>cUaCo2goB5p&3szju#`rAwWD ziTixWls8QDwIh%beg*~~`${xb0zpumL9R@_%KH{efJ=}2i1}LREkR9H7r`W98(^y* zdpqnbtjUkS>`Z%DH_3%|>D16=y4uK%+6=24htLOF%CIi}2#}dOAofZE^&7(bESO}# zEDCxJ=VKFz@xVX&UrVk{U}E3&g_Bw=Hd8lTwz`hKsaLDrP1C7t$1Vsi>$HtoHR3<* zFi$Ot&ht~oviLLTSf2*BFIW@Za}xM)s0pO))`h&N4Y;h@p1JJZ4I^(Pf?e=9=GL<{ zL{yI!!fPaDi5I>QoWtgMm9**07C2iX&uD(~f|~D?8F2oOZld63HY7>zCZVwn^d-|KL)(0 zLX{^uu*B;znIsd29eqt%GlQknZc`Zq?C%3s?@iG5%X%#OF2xNTJ(r28J4SMLOJe_Q zJ+`0Y6{2Y}9Q$TUgZe0=xCBDvzPSa^Vww(Z0clt=*W3C#{rmNnQ=zGH*TkKGl>89*l(K z&{W~d)ur(9b}Sw~Um>b(Mq{65Hf9}DWKtd8ygoc(epyW?6r9MOG7EIM+Nj#?Lz%*+m!<4-fS+m9*dS~r; z{C@1B@Yp>nuFG8}o5XQIwi!Eb!f=?jQ;rQzd_%OH%OK5b4Zat=r^UJY@XN&(f;L7# z`OKO4`D9leTW}2m1BcV9Clf^Sl92YzC#qh9$d&ZbqIHynDCn7W6VCuUQrfzM8dY4y zOXja|)+tv}TNJ++#hDwPfO79ZaeuTjHaP5|lh)VZq*w+!G!6-A^%ZDa-vGY7=Ck%U(pb&n4{&2k zHIbE*0LjPmm}*OD_EEbzYhu5h4OPt~FQ%w*z3%cL4uhjeDr0~)ffaUPGj4&B(HAh; z*q8M`eHv9tWMEIxV2B~Z*!jktyd8Tyxt2eGecQ8MSY2rY^<(eCnX(M}Q1>)ERNVn5 zGG5_b`tm=0R;_a;qZw|%re3##8aN79Yphua%m&K?nc{WtO-~y%OEY3`dL+Z{HBF%6 zv=u+aW-#HsP76OQ(1z1yZD8ee1f(Q;al?9Iu}WbJboacbTbLchM0O4;?cMsG=b*J}76^^F-X%>GZ^ZM^@AYz?0VMdSN(OQY7X z-i5qBhi<)pb8d^}FS^9zJZTwlmCF8hV^`?rgO17uBzmT-k)ALe*q#wL?as)?v)gR<(`a9J zAe=y0Ke`ixUdfZDN7Hcamk^9s{YmoWlfZobD{|pcAiS`QgS;iY4})oH9F!x?m$e0t z$YwCH|D(frIMz=ro4gP?kh^X~U8Q4)}&ln5lBhA^$?JeMY zUk0-yVsQ_f30kS)>|CF<;`C#55d0KQf^twfTnXZJ`fu+AW=jne_8kVfsn5Z+>b!7s z;XxSp`V2fbY!TL5mqJZUFtg_PZ{lsf304kOfvVbaF+C=VV`$R8I_A4}I!w~GL~wjDuHA6$sm{bTtvIzbn8*kI{_!*3NPN4L3CgSBn7sR+9istLwyZx z){_M7pp!7)NJw9JX2YVL8_DS5{CpWzL$1F~WD2ZyK>Dq|tl{W9(&p4aBvu80Y2qwm zGRO&9dbz=mMPs4FULTzvJYh;5CekLUI(Rp>3!3}(z|2@9F%4JF4i(3$KUKhVfin!` z?J-J%tH_++7Vz8ehH&Vc0%7QNP1ezQn$Yk<4h+^a#HgX!@F;vbjH!qt0^fn`q7@q; zIbg1_IOg=ahIWV+jKE+qP$1m7lC2mK-Ngpjh#J4 z+zth2Ll2A2?F3yO9-tIz0h(pCq(f~ob!dvg^0;~b^8Wq)kYkFd|5cssaegI~KGurH zCZA#K58lSeW*Q{)3?&_5t{}h89KOyF`??owGuut?3>$)V8_bZ0n}MQT(XgdnlPxfm z!tkqFtarl!@;h-j?9-HFrS+vS)!cyWaykQo=f}v|X)i&(Q2<-KbYQ{UqqJ#72>2yT z!~502@Ox`1Y??($fAa`BEz{*6okvh)SE348`rY)b&371o{UGV9yA;w328q&0h{Bc) zQ+OY6OW1`M^jRYJfE0Oz zqI44Hw{E3AqpD$b;X-sAmWLVBtihQ-V{;&#v|inunkJ?}B!3t3yhw>n@sx#8{n*e)&JFBS~SLhZCuwxlnd{JTERxLz9s>mtbGVI6B{73`W<*5TFp3*6GE6R?|0Mc zFq}BsPIOG}GZW@jKuhCvP_XRFuG^!+hTa`dRHa_QvwL^RIbuiqZ&&2Xr(VFNnnQ8X zT{*_IY!QCveWBeBy#^E1Za}LU2M>mQAXmOGg_S%e(}|qh@a20rY;2FfT?1sa@hW@}egIP(^H;a*%zK{1+5 z{f>_p&LPL8qnS;%EOz(wff=NO6e;=rt1pqvW5e?3D7{#+F=7w6_ufNtj)!5&zFlZN zGDkSI$&DNsy^uD4)nauFPr>f>T6lZL5b818o61JcWV5=j!rhM}VbFpJ@az3sh%~JP z?<^n2qJJuQx8*{;aUHW{c^&lL(hehzXES9>q)={g8~wcF(_b0{qV|!kagTnANu_62 zB@o%`iS+V12YC84j^4BD2327K%!u+Ak8>h=|Nro@Wg#i?u);e_8|lo%1eo8wsLm;C zEn3;t<6IMeyg48nizYXbHE%AGr4lv5jFSG;OGlN2npD!Ej)vSVM=#E#UmoghQX~@x zjDk!WQ9|x|Q*0ob>0?B_ML(2_OfhOuWYhBs0G8iSTKz|0OyTXk%qyt+~!OklSC^N z+uHBoun9Y`(fJI@XgZ49#QDj;g^Bw9+-tUxe$Vja)`pgXvqb{S2z*8JUf_)V+~+0g%z%g0`RZp>O6bdNs5erTKe_H!stOPSt<+#oGsafz0-FPzqN@*VG&!3KFdE z2l1Jchz5Zu>~(kJ9xB|yqd8xAo4T9$YNKC>P9zN1%+<8km&-lUS)`P^0NRXz8ZKCRnqmBq58Z#s{(AXE(xw{^8Uq#2jai zGei%kBuEQc#2&tOfb_oTBrY#(eLOofL5kh~n)$cpRQ`4tk{kcrrM-2#%;%^ zVvA;2j5%SZ~5gAyFP@i4y^-KwTooA!RUYU^Z&?M zqHVG|8~i2{cWqzD`Ynk>iE)eAHIH9IZT>gLqof4id{AJY>uI4+s|PE1{uS*??O2Z| zNg!2{1D@@g?Af4ncC^zrc>2bjjC5B;|C1D!<>XVr-SZ&f^%Qg_7lPT{PNw+m1=gI$ zcyC$!nLdr5^siq0NN+Z!!wy3FE3(`37(0JYfBMyq#~2II!h3V3f|H#Vo426^M0NO= z=KuLEdX_k-cNW}02&2BdW>hNT!|`96Q# z*X4fh%$zxM?ks0{AL7WbT>~E^v90ffY-geXr<&&A+ojj>C=esIr_LjjdG7d7vH;jc z7jTr_aCXfN<13%tae2f+h>#kIP3C>T_pFNWrA2Gl_7~2$zN!*yUA@j8 z;!~44mW99>t1;+$`8iH#pM^WZFO#n$q}bO%*YLXS>9{_1E+oZn#iQ?jCplTZIKa{! z8%-R{q6I?y`}Aebp7X$3GWF0}n#j(P719UxvG_=;@yUFt;lm0#l z-y6Lrn$zWABuT}_gC25slh3e&d(7DTl@e^8%W_uZY#MvEr<;56JAs58{)o$GFTj$C zJ5a%jCKMI34qH?oz`MA1+_Uy85p!+E%AuP7q2b5H=kV;*9{l0vRnRpuU{yUXV~8-=g`qnz@gi9{9qhtH|qJCcF58BHg~ch^Wu^ z7xCls4*l$#w%MY5iU0DVbk8|_KqeFa z&{ZJMe)?mts%_ZN>96iuy2HM|PO+lpcesNGgK$e;VkF3(-2RrRzjoa#vxl2!A+~h z_)ZN++?8LK3@bCotc?Jx9sPzLO)Dl3JWrFKzmstLMi0>E@zEA4-on!|WFRiQhjYv3 zbzR**(JreM_;|}EGS>V79^8mLWlY$p0iR z9^!cXdR4rsqn8^SAI_L4Iub{{ljP2x4196iZrorzn4Oon52udV&*V(0!|j{Qd7RRE zvSYp_8l~S%ejU7pi?rLgE`J@|Z1e;dUDJdV!zlcoRYAwc-b1?w%Hf||UyzFjM-p?T z%g8t46*GQj7`Hp71@F8)8#j${#hde!(2d7S*jhC=cF3$8?6P1Oj@|c%IVoO?Zg0|I z2Q99I$WKSeQ?cPldDL(+aKaqsk7+Ot`;~{!b-cvlaR_fJxs8ga7LfZFuVIGAez<%> z97lXF!P7glNo>J9uyK1PIJ)O3vS{z)w&;no#eE*^w#yFe-jAKk*8R zynnZQ?GX2tz!@LTP$71DqjkNwv?=9YI4#od#hqM^|O(VimXSuk0CwFcRq zi-F2#OYyJiddPo>8S|Bo%WU|q%w+kGMvruEp>n82XK0?N{(ru&qjihfo8r!_wud;| zWj7kVI+BG_>cnAZSRAL=(ZeJRX=P+0=Ze}fyW&MwN}s$`Hg6J4m0gLCx_&@r%o0(3 zGaueS+0q(h&blbP{Af1bUfD%Tzu&;B7gO*9hjp+))A>K&wb%ExL@KSFjPCqII%%`u zq>3G}9z7Yc9}PKSfDr3X^CdWNFs{Ep71w{-_@BM;KeYGXX9=gl;z1smv)h5byiGL<2=PuwLliEt3NpW?!snR;DZZ}cKGdCO?4SZKwAJDsr(|+g4w8H> z8FQ{Mc?#4O{(;3!$~4tJlIje!<#TaPr{3aK;8-z|+6@xI_m>(lGw&fee%Fz%ubB(R zAD@Eqjj`mYxyy#i~^K)|tg4F%hBHk+8Xy=rp4A>%FZ8lre6M}u* z==fqj?~Gb4N!ikdPGm%*^N|9wccU!|EO`Xsb5Aja?&8Q{jS@7ju%JONcwLvZ7Jc_z z3Wk>OIPS;FN$ld;Tt{*V+G+TX`{Ni7X?qQ6<*-TQ{vin(Su%t=d{v+;w0c2H)rMAI zzD!K}oaiI}JQ#O<2A zO;CH83={3vQsKLSv^!e^OP-%cY&NGu&XRH>_x=)GHJJt#4>ZZDRfk~ar4~4EIR=G~ z$H=YE7_Tpx18vC&V#@v(gZV}w9Q65|Lrt}(LG_hE$;`lExaY$%ELd$reM6*a%iDO^ z95;*}zd8tXrr2QL4Q|L&M-T6{+)Bp@__Ou0y<~CrP*{D&16-V<;7Vc}KKMfu*1w4b zUCsrZ`^LfkjxN+c{(_njOu*F-0-GfjP%LXf1MKa|`xS=JN_e^8fB)2@%XUs|#wv03 zB3|a@<&m?3G*mF=d`Zpi9QftB04%ikz~^OIkaT+-5w8h?K+^}Hm@^B6QJcXgO_4O! z48~I~Hh@^j6%f}jr88}I6N?8vJnEJ@?n|);8$AL+AqAxD(eX>Oms-J!RUWW_&!t+T zyHG?!S^5Z=IP@YhSd&dW8m5~40bw|0O|w#$Y7TZ_|UWqC3?#| z;hdbLsBDrL5#81Dc>ApXZL;k|1Mk;nfKvqwxTQwI<5llTxMKorc&Gx-;i)Lpn8%OO zQv|QWdhA8}wcPa==4|A$0dT!J9*vp)ch4{I`NXNt2nI&13@Pj!4S(+DfUovYJj9fc zsbki|>grVT_Rvid6lu%cMYPM>sCp z*wH^W!v1ojo?U{vOX09TBndShQo&M2e{OO*XGYB^FQe>sxa_kbr zL2$J80(5l8^H}@i*gxYZ<0MH#e9GqpxG!f!JRreVWWBQn(fU-0+=r{Pie8QQVLZh} z^U5Le>jPZ9qYk#u)`UUj&3J$OEb^r(7EW>rs$3CS|Q6GZGJ$q;~g_N~OR1|4{Wqu%^Nk z@XJjk!vvmkEyU2t|)g|;m^Z0fpR5Bp@|uWr}S&%`!Vu=*Je+N{W*OJi`WkvyMwtA{jfOvdd^ zChX4wYc!wN>-Y?qkF8TeE-Cej!GeL}SmKThaC|-wyHx=&sL=^a+(=*>YN*f-KOJqH&DRkf}U&h^**ad|Ez$-8wTCTl9$u zGF|I%@3tBg{#AfOax!qd(hl5+^pSL(HQT1Q4=nUvCcg#H*ahk(!#}91-z<0GGyYOuq zo^b3iE{VK~US%#}O^)l}!+$K;%;8Vrc)~NhI9Qu1pTB{|4!7j@2!XJ(&=h|_oiU+%k;E@r>cXxrvVj|0aQ+*=D1Wm#=ZP5i{8B6aBDX6pP_m1vWHS z1)~MVcxR+DEG#y`8GA;Nn@4Zrv$8?h-0>RsO(F+e-drFugC;@1@_5|Qa|HSVys^6j zzejEeAqyoE7?s5nV8~PiT!9L{-Zv4GpZOS~ZBT=@z+OgbAKsL%g=~gr!RP`%7~bLv z=Bp@NIqQSn8ZGgK>$@?^^295CD6xU@8}Wd<9A=j670r_&+h^kA(ngH#sC{_SjaBfb z_Xggob^&_Sg?Ml88GQ7CJ2zOa4&EkiM*nxN6b=he$G6HG7~fGG=`KydeJ20&m07)N z7M^d?iNS0EcKvG2x@&ebzdsCQ3uXRd%8b!I!*#wG#3ma3Mq@jcqXV-FvH9&leCp;T z=5^Lo7={FB<-#;vqEf>pUOU2!m~lr$Xa5}CeL=OvGW z_F12>LfIPZ^LrP5Uh#lgHaneBIi-z{Zm2`6JqsD-M;f43n#lNhC*VH{M@iLHAwKe0 z8FPy+ArGU!_Q8MnQny^3Rr&cDZDm5458u-9wdye3T@io=PO>9X)>_zFphrG@IfCAF zY7@ti%aPA9-lzTTX`HG$x5UrP1V$<<5mx>sPOC4)Ws58DbGKbsYHv20Cwm6J(D{v3 zUCuJE+IinG$9*^`QH{ud%_kWum-_kBzlMn}>j^GUvi1pne_}T_KU9fVJul$#2tJ^} zeGOQ>S%OWTavhsm`{6UIpW;R1ZD2vMG?qvkgucj)#cMtk;Z|)cyfJ+ue`cJ^VlncjfbO&4pHoPD}+~-Y4fD-u(af zM~ho8g@2Ag1fx+aO@P)*58I$|8+}yF;0CQkA{|nlB=%q$nRAa*16GyoF>;GpL@Z4%<*}AdWtsV z;3o;M-cG>ITLU5Js2=EcSM;BuL}U1W@1puf*QB9|CgXdv%g=dFT-(yStl~KQtv-c{M(jNfD&6Q za}8J76f#|*BiInNuQ>R|f_|A1(LeU79sKk627Jfx6X9LWVC}q7#Q2pIvOM*HczsG?k`Kjk4G)sIhatm= zbKwqTl%R@7PpyIUC)d%qBSz5O%ln0FUeDgTa}qBw4ien`eu`k_=P2w-B-1P*iKkXg zWHu!*5IRQ-$9tF~CSMX49}>keh&jxAk4{W=gRM@E*lME;Y?pgUinjmg-)o5a(tnqa zSWD7gcVMx%xBv6}pLaHm=^=L~NB!l|>kKlaXg~38d&2FW9|pqCViLos!TPPLP}VvF z+*+K;{81*j_|0HEikGETPXIN$DPghEhlm%ymhU*$jeMhvK{JmMb>&p_M;IaaaeEjO zF5^@ZT2_?lexoJ2O|#lS`DCa@cE0$MIl!%gDuFnhBLQC;3$((GK0 zl%4)J7D3s+(7uik{BW-paKUrY?ZqLY@s9oEh)>9Qk&i#$llV(dID74IkSYly_k3DW zP`w%+@3R)i{Je}djJrO;iBJ&24`dx^63&%kZ_V|IJ(uW?;;3seSE z!tE52yT_Np#%)K*gip%k%tQn9TUbCGO1?3ZWc(m`{5K*uITPtEv4o%>`8am>QzF^K z$4GnMLAq8Ok<4cTljDMs&qIH5Br67e-M5BpSt!E!6SmBZhp$U@@P5r#qIv3N=}B1h z`3sjedJ!{P-x1{Zp5#6r{K0iDRE8;AJ4I*V8tqxQ>PaYjZf`#RYBvs#tm@_xud1P+ z=~28ctq(YvTqy8TrvHp-MZr~O&tfzCz2k=A%F30{Nfyj_f!p_vZ)@P215y5O+NK0RqP2nw~~ zS$z$fDzlaw+vI|ye|=+a4mgjx4Q07$-_Wf2#v(ZlE=VI{cP|jRthMC#xRoNB zlBfMWtNzo^oqQhW_le_S)(1IwbN>_<9r8fn61yL%UGRj2>*g?WOacilD8|*Fmf`^Y zbVRIIpvY?{M0J_37DtWKKXBYj4fI-B9%5X*VZr^AMD16NgKV$nyj!T5K zBxT4rxXxo&?2tXrxIUUpeyE1Q8Tk+{eNPAsTd|k-bqhyM8v3Gf=4lfys%h${S2*5# z7DVdS^Ee50CDV1J(82u<Vvg9O#I7Lk87dA#XN$Lwo8;gzK`o@zyW7WxK_=Z|l<0bcI&(ahD`?FRFz6 zQQgF$VFX>x+w|L=>%h=)BUvvaN#gbOL^A(ud@9VBwIV+^s8X3VnDbU7%;{;)(7b6g z3Rz(Xc_uRvY!ssbdAx0d-lznJf#viXX^Q z?0`R)Zv}GeBquSzi_9Xb+(%i0G!9HzKn}!`opG)bv1OXdW1D?-ZQObOQIQ z;%M>y*)UedhJ=irf-*+}(YED%C3FRF=eP%w%911_SD2zS_e{>G@GhLZ=|+ND710XC zVWjzzHT)RjfeR{LkgwPgU%mKQBx})|(oy@zCZf623FX@-@;IGypwCH^V?fw?tqhtY zhocYKe|?qQOnP92PZFo$cTr%w?w+WR&U_Bm`ltL{%gd(SkOWdaDhNClD&RHzdL#RO zCsc^5(yKK^qTf&CMsS8vq3FQP-{j{Bc~E&~g;xxp-v67hzx^+laZmdyhzNlqU}%pGR&W%{C4M z-?|GU)_Ia$RkrBoqYM#$evoW3V}=-TRTrSqI{^|5q~TF^R{w9p33~gP<%$8^{@rT$ z&$PdOrgc|?k>9mPXk(Ns5nffq4z3xTtH*Of2PhClv+~NJ{O#idW5GNeg9!X}r8bGpf0Djjl zf$#p#A~NBa$1m4lbdT6&kQ3V`qx&o@%=Z zHJGrDc6gn^lcq(I4@-8FkB+9S`he_`%U&DMsOv9?S8)v9DR}^omN-h^u2G~z3@vC* z_#kpp-ibWlH-JrkH4zU$DNySt#bX(dpy`_@({W2Ek_~-`BxNVFlYgm0Ld-}|dL~3m zlzN#ydKUbIPj zA2zH{$Erud!RK->5**Q>s(fDg?K%(9*Ku7WV%h<`MqkMGEEa-%(HC-d%W!V?YI$m= zz7!0CN5S!D=5S6U4ZoP42s*zL;l4kDbs@Wo;Ma`rn z)*bJAG#r0(7{dm{WfJknb5Uf99%M;+Q+qKT8ofFfziV-(JMRlwv61^(JAo7}%QL}YZ-0SrR5{$y7);|F2=lW-7WJ&r z#+%11M&U0n(dIFAB3dnrG^nmsDI~jIBZgA` zxARV+FMsA^`!YS8sA^1S7Y;_jW}jgAVG}%Z)JfQu6%1%$9+9|Piw~}uij7G&yqohI zd4D(rOCKJ_)0H#e`AZ4hem9Ic>GK7mX$lUi8^oRp7(mpXuEo2}E#QNM9KAil3tp!i zQTN|xxdDqh$%!^4vgw07WOf`NkFgjSw6%fJY9o|-{VPQD__L`Sf*E-q4^W@(z%I$H zV6?ohlB2G=@Ml*B{`5VI&tV^l2Wm|u89Xjv-i9lf^P2#{g$O?U4uo}IuL&-u6cWXX zIm}FzJmixd1peK+aPX8Y8~bqy8~E5@yKk`8d@j{X{F!ss*i@{%X9-Qz8-y>#T!Lag zuJcL65M+?y1V-W}yhC;bVLS=_Sb>a|~K)GLgjmYao2>3wu&z@caS= zXld*w-gC7eFCW_`DHv7&*#AJntjguyqq`=*t1`CpW>MyosFeH6pusA7v$h7;o#U z?4Or~slUreuVs&Dj{Z4X1#J2hu)83J2ztfYO~*r^*&q-b4l)qU_glJ7Lhq@i_^47e zksi5+ix?RJ6B}*ZA{$sgub!wRo`c3ykH|ZjI1E3pLG_Fy zq*@c+2R!@$BCw1C&`&RA6#@6 zk=W->SlK@YX3a1Z^&@h+xZvIVKg=?$2>!(*koI{K0vFB643$}=`fVrb!u z@%UkikftX^bF(+xMm_JexiO<# z(e2djq-0M%a_F4|ORSEF=Ab(Uzu>sDJ+@nKfRFu_7xklm{1hwpgY(iZlwEj;Y!)mM zJs;o#$YgV;#JF=5WK&BLVJ|i5=Le|2|Nm3B6Jp&BvUgKLI zR)aCWKOY&b!THuaB45rQA`c$clly);n8<~I$^s6gs}6~1G21+yoH(dUw0C3@-TIH@ zHMvD~*eC9~k{o>}z8V+ZtA`e!Y|#000(E9$W~k;n)c19Yh`uKUTS3oZGD)&O zOIjsWIfKjtqUWgrL%HmBH8`?10v7dr7xAa+vpyWsUQd=!HbYtLV`37xj(pMG29t`? ziMyQ1`U^kqWat`DRGgo z5w*9+=p&JRB!N6vI}($XYnewDuTk?pf8;+j8NJu8#{DuDvC0$P_7+0T*^{C&?RNxF zG-3+A{Ma8;t9Xc=sev~cSYfG8JMi?lHKH<$1AO4mQxW!tuqFGi{&`5@ENDB`4Nj|K zNVL8ulH|f+oU8%)pKMtb< zc0)@BJ6#98zSF4jRTJ@-Ql)42SQEb+_ORRRA}**p2eX5GX{5a!6s|5LBTi{jN%vH^ z9;*k6mptI5&U7YX+J4-p_yy^I|3Y5%oq>xtx8S(P(lqU+3%&1p29ifw(uK(~%(mP% z{#^4JMJrD4=bKPQ`WPDFZ$SqCUWN>7N->z|vg0Hc(fp*7a4~*A-eFV-E&d+Nt%ZE9 z$u~XZl)e>OnvC$Y-x)-w&V~K5vI)Q2>xO3;i?N@--GrDNP4?2k)8Ktt0Z)AY5>yKl z@uclvn9Kgj(8yH?JcR`OW-5~2WH;)0E*TC zE{|I`<-j~`K>6!ml2=J(*wjJ|-WOkiE-zcg<%=A9dE8kr72JaEbDLn8;%3fRCKQ&~ zeIvzsPUv{R9FqM>l1|*3if0w5vW~(Dcw~DR+7cl_!mov)@crM(QXU~L+Pn&RE%k;; zX~yhA-jCq<>O_HUxHwjy9$WI!#Rg_^pYhTg^@2go98nz*UecOT&SRyX!pDV5*f4G} zJ~DGAEJ~1MhwQYVTLw9Tf4m&MnH$OH3uuE{*@38GKn9Wg^p49iY6rt(aa>7eBD2k( z&*l7t$B>bk3aU&maUL0nhaWhCk{^#^y&syfW}^bAZ>ucL-S5!0EOr9az61W1!=9ZIN zuja!2w$X5Z{@<9*_yZbf-(7vOew7WYYUfTz6npSKGH1ESp-bT2zGKXbb{p0!wHSKx zRzMdY^SEM9F-}>z7klt=GvoS>F}{9MWZ;#-U^Ra)Xpbe_=t-;KZ2UylP3kH{V+R<| z$9doLGADKM4rB>=j(@CGAQ}2eNVm9%xGA1QTXv0vV{9>4sf>Z)PWy2EA4T-^^cg&$ ztq#qdSB|VBf1=yo=V5L2X>7H`sDJGh{!CqtG*zDB3SN^OtsRP=vEev;ZU?l~7Ljc& z029l9qO+M1pxaymNw)e})i#`Leb>P?Bz_@FZ?~d#IYDR|BgU-j8oUA(xmKn@T}2k1_n#&ITs&wjHZ6hw%8$JY*lZfjk+ag=KS1Ag_&fs%9(y?blEL!qe4&SPE#gDJ9W1TO@;>-)bP}+!}IQ;4v>{pwN zD_@s@YE3GhdG;O1^}Rx&p)urI<5XPWZ-Oq=If~Ym4SXKHqMUp9%DiUWo%xN}UUo$H z)zs00VaLIh-}6nCQN+&m5-9P(R{TS{oH_p}5c~5Om6962;Nn9!)Hc!vTf}8DXB%&E zZ;NhnNxaSN{e+Kj@R{cK2+vjQ7w;X4*Tmj7$55Q*G>R7AMgqy%DMy)m$==$~b`2FK#GF|H?if>ZI zw~dOJh-1-UVY?qirHtWXw#>$60f0xF@Ua}ntOP;k21q+Wk83gI{k|N}ksuVy#hv;A zv!uu2`0)o|vVg@0dB4p1?;}O?#6R@Qm&Rc^<}sS+Jqicu6V!_($#dBl(fazQF+$YW zCrLwbttQ1??FVp|L@Dw;ZOJ}1pU(Q{YBA4w-=fiBTDUF$E}HdI2N!eu1%X#AvD(nh z_h!kmUC4Ri_F?0PNaBCO^UN_`%ijFOF%cUH* zl?Rp>)&9k~{fEv4gOZT+k!a-R(TDE8HDfI`*Ws%XP5<;om|rc5DgSp|elhiUGkQTR zY>#8_$6{D~o;(_x&fEOqE_xIxg&VH?&!*kqhM~L!o?}&yXY9=roKG6Wt(osk60e>i zmSHn+jvZ0W3xfdRft0`4#1^}*68G?8t%s6zxwaFT1lw>N)z_8 zHpFoWEIwd*id2?ncRM*rnurLb;)BC(0DhGchJ@~cfQT&I|Vtcn9;l9?b z=;yi6*{nryv1mA67C8)^^ZXmbymv=Dx?Rr4FMVslt=<7Rr*tbmeOe!t-+qSI+<(cq zH}mnQzekJoU;pq;ZJ{K@wEsbxI>S-v2xn5YC605vr1p;t3;W;8|Go|}@79vrRXOB^ zt|2;XdL4IflC7$(?l}lo|1^3yvX>MrYpU#k9BIP_Z=t-;Ra6l)KlE5%JA-oT^-O7YflO4!B-v-c))1upSia^Hk5YDhT02iXg zsmlvn(EdD|j^z`ywnZ$a<}Pc&GITvKkJb`{y?%6Mvo~FryPg_U8~}2lf+_9U1`5l> zOErz8N~gu#r+Tw0VeUc`2>2}nZ&Ea9_2S#-A{)$WX&;<_^=;RHj$s5UG^!H*uHmmhLOq3c-H4f#| zqk5aD-Rl%ON#P!kdCx)i>ukuAjf3!JOT7;(X`DxidD1G1@5MNa;dM`I-6h&Z1yWhL*T@@G@ouZE4*PbG!!9 z$>*xM#M}xr^+qJAU#v!#%nzn09vwrdNAVxIr|`l z8!|Bz+S8}e-jfQDGt-jXOs$9OcC8}X*g!;=g)gC3`!11+`-5mwt}$gi{AlPn2^u}8 z3mOO1!-XeRFv&0vW)BVlJGE;tF1vzAXHOt8gKp3VNBro}+-z!*kx17y&7fg>6{)Rt z16&E1OkW><1l0o)Xw@z8(rLmx%7$Ksm2;Ay_UA?zI!Z|QTNZ)o(|G6|oC#?PQgq+m z?0%YrL8%LAqNO;TNS{lG90{QciCgKpQ`%7OY7O^ORr}>exHzYYM(O9fHP_Gt|O1Dq?n1#i{f`R3VI;^9x?(4};J9&q6?vfRDM#f=yC0MeQCxB1;|D4ud^* zDKzlZBgomYhhEwcMI)-k=_&E?RBpBk_1xP9+hg)Ub8}_0wHC&Qrqkm* z9@#Lt@zi9?6Igk~hxWWlq=E)x>TH@%W#0~`ao1*0?>7t;;}qz}6^8UXuTK^F-xKwz z+d2k@)tS*1@zPXy$Ctins3A?096)Ti3}cJWp>`jcdXV!n{Eq9uYAHOclVNa?a%O&uj$;LNO{)R z?H)P4V;d;HiQ?SXqP1*T$1}l{e4&WOEhb8|dch6yH2ndQy)Oyx zf363unha+7COu+*Q;5HJJtInMo|4GHD$qMqpBCSDr-d04NyGdou(C)&C114Q)T~%i zQeMy9E4AeP#^u4^>n0hPlgDJ16I{C!*h+mL7rjg==7T_d^~} zdkd)Sn-4383}Jlu_~`8ARyaY@0K|Ili{4f93}tlxWE1f-lBkQf-zU^}aub)26!m?l z%n1}wl}Y;f)_>_MnDF^T@1R?mkC_pYq4<&SCGb}~#>`18<4$)Bz_@is|GUDb*ODY< zXAjwK=m}31rKrc_Gmxq*K`R#RLo;v)d0QRLCG0;z)J>jq+loc*`^Jlos#sl z{$oJN-ta{?3mNaoK@GpW$+gshaQ|I{XpL8M#E@}Am&g_Ok>SST^m5z*()*i3FYhTp zN7q^6VmuJGyPSlf14Hl{F-g>xJ^**b$S_rQ>M-|=H1w%nM5nxih&b=ZV_dC)G!cVK8r1B(oEgX~&5fbm*xT$hogfk1kNe zBMOF!c(dsJLsA0{xN$=_TG~rUY?39pI#r1KdFEem27cH!!LaMSM0P8KY~n0M&l42x zmaLcA*)Oxg{>w_M2;sPq`1tX75*~aKCSM*5f%;C6`t1ktjhYTSc^t=zkBQ)+oywTr z8;vjK?T5{Ew~*R?OI+k@f$In4mOTB^Ml?Of!lb+tjLiGZScjA%>y!rQ*{lKjYNJS{ zr2*8t^B8WjDR}AL(@?+pJa=!*Q%Yf@%)qo7QuUosmtGZ_!R4>|H|I< zJ05siy$-WmtCPq@pW|~Q4S?Tn)+Dc6q|*?~|J8`3uLYxwQ%Tq>YrUwPy<-4$XkSCG z`FIVlb!k}TlZJ@*{rsM|{iSHW$Xs+5T#da@|1Er0IeG;@Wn&9_# zJnC?hAvsF!sIvj{IDQS}r9T4^qdT!u(_QrO&vxtZEnm>_j<*5>n1DK11I zelLji-OX@@nexxL3zNkZn5U^~@aRk^R1UjDQuXUlcc3;}#Z2?jQgZ3~QS@fCn5Zx0 zM~divwp9_omVHU+tps4D$CDv@PIJtO3xY@6g%Fkg9L;fF0GFd=_`DtYT+WQ&sFCa! zmFwrv=ju+hKz|&G;rZFSCjm{6c!#_$zC>?UWpRG_f7ikg%MVB{+7O4P-Xz=J$bf9J z8Ru{D7uUYO4twuwNdJ-mv`eQ^&Bj|0G>^{}_i#TfV*dIF`if`u^H8j~I+M3Ij>bJh zBt-3hDVJdXu#*s{cMbc(I`sW9I-Iv%ovE;*22(AGJp z-9SGsW=o^A?CFx3Ml|UAE@HNpCN)rqiU3P{(^K`sGB}Zt)F# z!is6}Bw6}sO+Jz5W5Lcn%O_9UedwYK?bJAzkG=BOE#15~lsdjKE`2rCmyBKD0q1=e z7au>rh~^yFL2qT6!4H~13vz5w_J;NJc)=XHp}G;8RiA*lYbWU&RR~(Q-huyIE1D26 zjqcx+M-BP)&8anojvRlU4v{gZh6NcA@UvTB_iHtzmbJlrn_r;9=fz9AyN(Wc5l)@U zSh`}yOS;M-mxhj5K`$?`paod)~HK=sFouKq%u>g!#oQH$Si->{S66$W4M1NUbp&H8@z{4<#PO{riy%oC| zRNp7#MrljqW)z z8d;8^@WuTg%7`+D6z6*&PGJhxEDzj^~BQi<_WMn-iE$?-ek}Er^5*4P*~Sh0q)v~ z#92~_8XAQ%9cEO1Nh$Hqv7;Y#j)0lpL3+Py5gjPpN0)w%qI;g} zQ1eGxG=54u8RcXOGs`#7mi6DEaY-C3s9h>LbMHUvL!Ug)gg@ffAno!L@+v4F=Bl_d zo??TE^txAM@$ivg(lednQy#cqZi~Z$Vf^bUbkV!(Fu3awyq52P3%R4{1m_#DuXGBW zXlRDjfy(sleIwXWA3>9=6X~6c!SwK`6znu+AIKVNP^XP!shC0OiCHbVnvn-Q z7y?aaV(F`2c2J~|Nvy}Hpws3fQSz8IP;FIDj?7s|=igMKT$B}4d4Ca{>|G8YO(iKO zv7D@XZ;g9jJO-v*mkx8arXd?mXwB6(&=ILmr#bnO6@%mH^gCDRvY!vBz9;Vk5O@^M zEnbDQ^^QZ;sRp>$WeYQB?}Lz(L)7g#9|x^tL8Vg8fYR|cu&lqr=ejDVLk8?6UfF$6 zWf=@#ax>vhS0kt-sZykBkD7jMrp28VFw>jB@kQoz-Chgok`_;^yARW(LFqL6tu1sf z7)&?IeTC|umNcR33Dg^@(AB<@^s3Kb+S(IMHwB!d$#PTZsTq>==fjgw;U`bWn;m2v z5B`AcGoOj{UM+fm*AlAHx1CzuTR^RApFrf!Z!pk(4aqy_LFbiTBuv6dFmfyr%`d)f zypD}uFH{1mVBw=Ha7)-nE)*)ztqVpl_0Au_{*?nH?R^MocGF4m_89o7HWN7qyVG@1 zot(B~4G$=jjC55a=+LF+G;)+X4bT`2ww^Po@!jE~dg<|FqWjrniWJ&x>6g5N@OZ~&2_&PAy7R_UAo5P^} z!>ChqEYw=d($+)|x^&S2I;kuL&PV#tfu|4B)T>%_PH_@#JQYD7D;m)mK?*b#a_H72 z2BL3(OUi+ual@#X%>#g)6KHahkWQ44poMxIe1uMz`_c|x1SrrOo0rm?!-MJ3d1kas z&xcwJ_5{-nyP-&6$@@R9q2Kr6|XYV^AAYvSdX0l&{wks7;ZIJ{_Y=UU7ks|^9^$DTDk8booK(pUy zis(>XcZ0mEsQ|R8mTtGp14|zzIy!nH7&IIOlT}yHO&+({=+5&B4ml*xsNZO$NTrkJkEWeGiT6lXU)6UP^u4V@OtT2S)98!t9Ul1+;e-}5Fz<}|Kq7rk{Rk=8BoBK0!&34x}vJ{6Ac;PUE@wRq!`Zj42y<=4;mQrRiix&CC(2Xe*=*O%vl(l_@RT0l1Vq>RhxN{*{sZ%$8L@}dPlZ0N>Z4_ZIJn%Vx$5Z;TW{U}X6PlOQ7ssvV% zF9%zXtI(6SBD}D2gru;mHJn&|0&I29z|#7?+ykc}2*x z`cov+;4odb;Wg3beGz6QW=I}DJUKnDH+}x$3v9blg>gw^$*@o_T5;*6TpV#{UNbnplX z)b-(__ZQ&vnU-|H&n3_pEoaS{o+Xnue20B&EZJ9!1yp~K0^RFb3)!>Tv33&tcgy8_wSI zDf6|S26H#mvbQ&!m}|~jNxxsUaPB`B&>gntfBLcWqwM{h!pt;6Xvx2fEe{#%$; zm@e&E=pnjq@sN34R>%5hMex*Gj*+&H5SzxK_mblYOf9<8;-LBKKM=#Ow`fBviVpm}duHqae|8*br z4>M{18Xf6Qr@D`z=a!`5iXjs0n3X|tHSMG_i~lBjjY-sZL={}neLx>NiE#Lj8nWqq zmQb%{kD!9vsd}Jsv_AwKIY150?fp02>oiv4<z5wulJa3mMI{_ay2X}_eJ{o8j+lt$wEEjXPhRc%T$HmQhAm| zG)dbalq++ONbc>{WgiXl*nZD%%yaV^Qe6-sy3fBR_Jc+es%8P}n_md|uX#@=6aI52 z*Z0q72k$zQcOMoom(UBA(z0R~$wEe6HvzM=P4WNg+PPEU{S>cjD7Tz!_UHP7o71iG zGX@AUk0_6@%D3%BicF44v`->Jwu`4r98& zcNsmQt&usyV0q^M`FHWWeQgLWchj6TDG>aQxg5(bV0UX2N#31eChQPn>*~&leL0emG<{V zPmd}K7?q#Tk)9Lpoi1=_i@%uP7Ro&)-cKU8UL>kY-`UJ&BW361sUL(O+4Q(vs4p( zB(8ayQcT4)+E;`;abuJSCu1()Jm}^$GJTZ`nYz+XDC@BNY9g)c`q-nyaiJHcSBcqX zgKKP7YKDMS?PPDF(Of5B((Q#T%#|pS2SF3r$d-KRI1^9Yc1kE`mfA4TI~Xf^xacR* zE2|^> zV6yPe2%+A>q4zCal#dGer1AxB__L3e{&*>=NW3G+Bdvexlv{%Q(mZ-TKO`Y}?^xEi z+JbkfX~cC-l5ow^=8|Yk3P+CDE+Q}b2W`1;C)%(Td2#;Y_yyVH=>Uo(Z77m!SRWhQFgG>U9gjDtOq ziWJLMKygEVn4Fc#+n-#9Tl~7JN13w7D(Vs39dL%-{9Ht8lCsFP8~n3=Z#}fDUV)CC z-Rbbk5Abbu7z}k;hx3lzh8{15@^QoUXiaOg=)Qg%^i+976z`?6%k72Ws1-ri4jl@z zxK6ZWnj&p-E`bj-UFeAmtJr1dYLQ8wCIM3^HhvEufwqbVNVK+qQ?AWlofo?;iICzd z-WZz-cGt?Gb$B%6O~OI`>L8deHm1Gg;!s<4CXLwS180>DIWNu^?et%QzTFhst4My$ zRCAR)b=6ISnL{4KTsH-7j)fb3>3I=0yBUCAMhaNg4WuSVy3x<-JE>>uMVMePfQ~=n zkL&LGfQ}6B|2$a*#=7%7U9au1#(4tvPKtsn8c)cG-D*^0!YY=?O`*9)`^eJAC%|7v zhQ13f#~6zfcyjp!__TU3Ozy|Wa@gA!=46ME)8{AAcOen*+_fGr4I52QuX$l<^y(8V zP?K)$<&i`yJ*vI!Bzzn?iLN?!m(2`+NgO*y!i6YpdbYL%zCPyT4G(mOlTjPV^Gi+a zWM>Dg{-%ZdQ%8a8xv5~~{u|#LBb#5|2g*jhhOQ<%{^^ONYzLiEG#0#lZ<3shEwD9Z z6n)*IONT0Tp=uUgiGRC4s04k64deRK`^J{Aa@R;W(4`Y<910I%3VRl%W)4}*2l zH*(!A5O((Sha2EW9c8A_ZS|u72X?}<6G@=4%o`T@e1uIoF+_E=1MFy80XsLu1C>pY zgx1Sb$zny&4Ys1mNB;05Np*LGw=v!E_!WLEX@xjjC+ht3KPdBThx47j?DK6qs@H&c z#mg3}lzZXt2k~TWN;Zz=ZNGEhYQVLw?ifDFhMs%A9x}El;dGf#aDSydRSQtzer2vl zg(g?}nH2xed$P=@3Dn3a3uGrc!utzlV05T2&c2Hx;|-d0*o<6Q>^cmanH={aNMfZW zqk=Jx+gN>E8R%XNg)&h(>BtKJMfnJt^HT*wZWlqhA0HEh=MfM!&V$iE=di(jx(jo5 z2y~+sM}HGd$#U>KY6Ibea!Bzho`wwvLZ6IeaKE-m;-wb^;iHkq`wrN&USp99(qYWz zU=YX1&@J2SsI2aKS{uKEep8EvX$fBRN|$odd)ZPNa4sI&wYJf;=yfowUnMv@w-*fR z-UKpF?tsgSKYL)${<~igW1dN#w55QX*B`tq&3Ayw)i~bvrw^$Lai$Yv`qIhyynUS4 zNXdt-r=c#hMRJ^vZT-04NUGXJL#jVA_Hok$Jz7>ao6e|kqTcuI!ONprIQKIxM99;= zz5y;w58-{L9uiU47145!0{y5k z7ls>#lOZ=EKxV22L{3e|?YmX!k)(AXo+^g1P4h_Gl`@#~Fqh06bRHILUIghE{)78Y zZDjczS>9ewpU$71N*l-SrEgCK(V1;gf;@|-c#0HwUOdM^%jn&~+hFSZy|7+66!sjy z0IFp#;n4eJnAal$mOFPq@v=VjdH?m)c*RKJ-gOyoA!S>R@cUq&N}{`8iRnibLGC(b zn7aDBXo~U@YzRz+1fK6Btjh-2uk)KEk*?Te83@iJ{bAf_M?4x9NsQn5zzq6B02X`*B3D;z1^>Y^g3QMU+=H_Q z8PIE965Q``M08}89!get;?8}!Bw(TNE+pgcJsu)^nuGmYU2-A$PpnbnmyVRKMcAHW z39@ck(XNev@9{3m(;WR9^wHJk%?7+HJUFpE(`ykO~IZZB&1NAxj z!Z@pnd(-Q%og_a#NV}@U|5y~n>ha-%U@j#%Pe=m zH*wkZjCOiwqb&GSEtGKXK4$|c^p z+6&ZN6=>8W*1=)Qh3pR0#f!4B*Q~DVf2v8 zB>uM|iFBxAxy38U=PWCtJhKg!?M^2aOT3|CbQ&DUUj?R3EkxJT0c5pC2y#~caEIO6 zd!Ua=227nSd>f0U@~eG`g(1L1Z?+pBphg3)bF>h7nupS;Vq~k|sVrefi>Ou<1VsdL<5~o4(e< z&R%8UQn{Lsy|J3v>HXp1kYelcG6P229U}W9lkw_y^0q5+xQhDghOCt%lxCC!w={1+;lq!}13u zzpz~8Yg)_=MIf&zAv50efPnsgQ{ zOfIsfEtRCs*%Pc)cMJOXY4%fGx+V)kd0xh6sfvY$`Gx!Vqs8*>NZ^ml5M zb6bdd`#!PlyJE>XwOhpfUL=e!eGReOZi$wDe+%=5JR;{~9H8~Z6BtupPwsw*=6yDt zFkrVAJa?D?I$94gD=`I?rfKkTL&tJ^)~j=VZr7nKx03Bz`hq#d_kyLqgQ%jyMZ8sd zT15H;lEK-$ZA|PSs#gCLR2ua0%XVc<%J7H&i9ea2!F`d+n<(bj5r&NyE7ua&y zC!FFr=O^Qi^KRT>CKI-y^d3V$ zRn)BBj{j*N#vP&YxI-@%Ki@iqxrRD4gFA&sEmF7*=VsvVf&q9og^%OxX3uTW>B>Ft z+KQ{cm$E(PBJ}QZNOIIimusA)d}&_(MXq&hBhxao;nq4G=QOVQ;Nlhku{T+>+30ES z;p3{0B5lPUT>0#yT<4WhT%*!TZf|D>H|x}D?sOZ^b71_Afr=)l`b~pdr<8z}C6o9l z$aOe>z#A-S1n!E-WbQ=2d>m-MnY+}Ph@Mu~_}QTl-JkB|J}0Yl`LiqWr*jlDu&u%1 ztCrk^1(n=;PK`5s_<%dQ#Ed)n{Q&k#dySTIRqXD*UR*_KBc6HS#SM1o4@EEga7GRm zcqmBBr8+O*0@9z6nNQp~(e*yu_p|&w%|FNgrsswG@VTTr?)&wV-8cS;Q-A-$i}l^P zHHy8tg&q{A=2+m^%5(qb2r1ToQ?70=CRevh@_J-Bp8e^8Z8aOYAMW8eZ_`P199V^7 zEeGyzJ<{V5pCbHaU5;;g+t&hlSMGc8VD9^ak2t+~1#W1Yj$Yrpv2JNC*!zqfcVfLh z_xbe%w(X%JH&kOGK6mX3I|Et@jCzy*DRaMC0|s3Ff;Zgl;8q^T>Bw~Of!jv^Po9(g zBG6%px}Yb~erBxvmOba3V92d9AIb$6DN2_A4(F2l@%Ah^qqsDMc(BnAMy4+2mf2py zh3ndwhiL$pJof}?>2+JQs$r~9&Vnh*+)3px7&gv=JFq|(KP+s*_L6JZFg{4~zGMVW zyC&=>3OG;SF#+{#<0MNKZ@}O*C(QlmLXOFI;SO*2=i+@7Io!xJXWGO}Qj>)k3sgVG2nrx^dg1W^*TIH2zONdRiEB(|2s=lnrvR zQ{RkyHr;{hpJH)Kc?`zvIV;Hg;g%k!_EwFHJ=?R^YRk3N%Kc~+Jnm&S&a3CGax6c7uPc;oZA>1 zhO_@GMe3k}zxX(zzN03ys8z=>C*cEA*HFgBwS(Ez>ZZJ_p1tu?Rs@&dJd;cL5yS1= z^%?uk>dKwy=*oo}^D#l&>e$PiBB`$Z)6ZpEhj`Dx7*1>KYJ9FYnzP~ta`km)|I@z; z_MRME55XsHluI|^?RN&h#$No~wedot#HB1ASHulR*Zt01PoL$S!hppnSH)rW<5BE! zh$4DjIm>#-7xMfMnfOsZ3MX19O3N0{T{WJKP#7cNX8vuKfb|6!jka)Uog!PlG-2 zfZ))0$h1(QWPC9FY&wnlhW>zEv1)X>dmf~j$-sRpcX(qVhBp_6(a>&NXgO>Fsf^Mj z*Tb4F*ysSm-vqeMF>XJt&iYO_cUs0(#;anhC^Wq$4FAbEN$WDh8XQd3#FwDl7ZYY}ae zMAK6sOTEUdgI?@AKRi- zovz6;p}meTftPohVZn`D_))nDQvYB4y$4fe!69V|%pb$^u61#wrNogAIsc4q&i99a zxd*`Y{aHw9T}{Va8AMZN{9hcF&Ck2h#(rva{fz>66ca%Aubcp*cKzWcoinfi+xu*WmGx0w--#mte=uHOtj*suMaVRZq?oID| zouwZ;_QNSY?$pXFXQa9%=u+~$n16D59NtN$hd(36jjCj+E6?H5k^(EX`+&3M74W_4 z4yPOY!?{~wR$1>3z?++^;B&<%cvKMrK|J=-@-mkNf>Xf;s599B7rV$&n^9Z9Lo|_I zJJyFDeZ^DwjPRvXyBWY}jX&=!GfU^wg#Pi=OEaBT&)NlJ6-opQcbFB!5&H%xvsL@2 z4_^!LpPbUV!rh*c3u{kH!rJ)wAaZUH|G1r)H>gpMm;j*+K95+I45p9oWl$6U-SnV% zD?O&Vkh)vmp%YclKo5l6=PlrO*-r7-&JhYqOZ9;0_(YrsA2Y`VKRfG+fcfAyLl7Sqcw@@TV3DE0+ zWOC{HuD{o0ZYx}X1K#gp*c?yn8`+gB?5HHYZMt$_Qzd8`n8p*Eg|UTh5n!5kiG7=( zj$Jcym}AO2$y##>?o8{$J+M*Wq&WUf>CvO`?{Al%E3og9lVr$Je$M4(`kS_WQ-!Tx zYpwdf-NfB$U4U~R^`Z+ir<3Nuf%?}&! zCxr|&uzG~N~F_w3+4nzK)XjTwA#k=Z=?((t5(|M*{AydgXc!0!PviIxMXBT9KJ~~ zqx(k{qrnIxmM&k#yyuO^U-rHdYa980_%1V!$0m~>IIOQX8{5KQo5dA);v2%fXVI)8 zU@(3u_K@xciLWXvnm@-;k%CXHj8x{*`Hbf# zJq9s%24ZNpaZ;>hb`AL`eMbMqTYPQMW)}7#m)V~05WU-y29D`#nQg>blA7F?8_V-5 zPnr0~4oZ5jG}T5RIchEn*TK7Z-?S~uU$xVZ}E(a$v7G^F97<2#+szs21sab4S zP8YmCP4(a0F2y$U_c7={#+~ZVW^B#7KG@qQiM(R9V9;2>0_FKQm>c)tr!pOGfu}q7 z%&v&#>!`CKjZ?5;!(PVThJ%4=E?5kFjZc~)Ed4)@fcqwA*tBnPII`d_YJ3Vsl>cL! zCT+u=j^#K;Mu&6P2*udNySP`a=GYw&xns&^Tuw$c&v|nSjaLulE?r3CzAyen z^12Qd?b_$Tb<>YS!=#Hi)RIZ`uiLX;X{{I?`GYuYR>Kcz``9Kc6EbMfYOAd7dF~|6r#vf z#`1nv1w*+^lKFg1?VnbuDqFfh?_W~}8lIiJD(U8*I&mbbB4*W#%j@R&HfC z8w}<)U_Y!v}q@HJU@sw5o3r?;tlpVbsbl^cPWln z0aURIk_1FWL)W$1Xu9IhKG>_|Y}6krD{)P8mL%=lL`q-S;33=DnAkTAW1lX@j`WAD zC2112KH(#9^r$2=j;2}l54Yy-*s5?lrfPEbi89>#xue*^*(#i~z6gr9zQn4hR^*gw z39eHMV-~*~XxwyDYqk7KG|X-Togu4f?IgNF0~bcYgAYrn+099G(f#32Hs6B!UbH6{ zkEzna)9PHm7=6hO+Y~xeM!~vyg%f2H_`M`8G3{eHMXK8}(!^Dy(e__G;Md1=a_42YFh;WA0R(5kjSTT+o4T^%zHSX$;L)E1@gr)Iy=62mKTrPq)^}S?g_< zr@L%lf#SJK!ral)M$rY=F4H-~rb>cd4x`=UDl+|imn?Ljs@V!_%lPpRAUE{J4#g|VB_$=M>ke(Z#1m}{(w7a1bblVn7vwL=@C7v5mKTMNu zIHHMVm*VLOrJ3}N=S1S+>;h|k1X0yb^Xb$$2HhTYp)uEhK0TF4`zcCo%;Ce(PO6Mfq`6y+0mrdv5}ddhqV=na}k7rarmp0)5O4Sugh zyJlRZ)*mYc*Lkhz@}JT`gK#G1OYB8}-lCtc}!|P}f1v zQ}CkUGDIA*r7fv9__zj!^czm+<1YpQ9pB1AW;+s1kqeztv0c!;a$lg)9?{e<0BQ5V zYcS)c4Q1h0w2RhU;Jvy*C*ho=qs0&|KAS)lW*xCox*|)hXLt*Bb@&8fxUaKNj#Q4r zsy^{iLNnZj4sF|_XccY2^Dn;y_SO)pi|`V>~;^fynGJ*#X;2kW-sg6 zR|eD44FPmUP!XNzR!GDA2GNbq$<)|d2W~iYqf5%(fbE=~VC9@i$6GbfuC~)jzT0@J zoAeQOyUSCz)HI=Aer^0kceWgX!Mn|9{)yGZ{dghB{B*)+6WS3$_V$S4d2X-^m<-U&x5}rRpbhqRoy9uLl;NB+G-r^%X4;D|jimP(fv5$Q&i5!+i zDy-tAIw0OAAI-<8`m;{8ph3bmjLt%tc}uaTr#!aaFvJ8tuGjPp15hPRRoWJD(U4T^ z*r6+OZ|y6hKCj78^K>?-G@SRBS&qdU_3%wnB)Z3Iz#~mHEOYVUX1iR&%@udZs%9 zP3FIPidL1aqWt+$!u6>Ro7iN_JR)t&II`t{^;%@ABT&X4Cn zZ6JqvzsPgX`m!I3T?9L}z(+4oeS;h^Qy4D#ek2jAMn7lfSKP2Mc&}Ae$ad7+?T9hK z7bLA8n}oU==Due&RUHrgp_dz`-U##$Kgx`5##dnusODfSWi&rD!Rv@=Di-0ebK{MHxq3* zI^P!ERvU1e4{LCrCrWVtkLApC#4Qr!yc*B8nuVqm+ zp}5gamm3lGlg&J#&AD|F%v>pgTRH8l`xT+y6?g@+kId@ zM?!GgPi?F2*+%GCmjtVGZj!B0x>BrVJO{Y5V|D{joal>_G&W-94jpdmiVtXC_{RRui z?js+r+MQka zcFGIx{zP24A&hDD>dm>-UMGQhy}`6}vM`qQ+?TMzPe*uNQNxVg1K~taAdI?sNaE_Y z6&)NBab?nKCepb^zPs^uF4;OT?NX){_pJdJcwQySdmQ2E+BDcWaRA=T4#(d|Pf9L* zK4axIPz~n4N)_GywM6o}|21}TMpMB`o})3$=^F{2C}9>)YKUwWWA!6L;9~qE^2X{g z`>Y)*)mxdughynz_z^7ko+=9L^0T1yb%W%{fGFlPJXsVxOBGi0ao2;xjoAIDu~r*b z%Hbv5cE-Fi(cCqG9q88s#y7N)!xk&SLpucO)@rinZ53qAJb&!BP$Dv*rJ_v^JV(mH z8$w-ovwMM7LbB-E?o!rqQyFYl>A>8ICoH*Ejs0Bg2E(oX*emD#oQnp5Cy2gMlSr(s z4S7ugFf};9%6wiG8m0ANKdbfNw&EQ2uwkE7-=4u>yX81k;5O(I1+=K<`!9=@XGVU!{kK?r0VyV_hus6?OPm}xQ2h`6X!<$XvQZm2qwS=(P@G01H z18)#2b6L`m7D6(!V_2qd4=6OQU^dp)Byj0g_G-%jxZ3%MF>j7V@o};|*A?T?y9u~D z!w2UMNoMaBS+f(rQ`zow%i;XTAI!e5DKX8}2R+qbiOyRcFg`njjC_2n;6eA-Yhp>?|>6dAoE_x<;a_Vc{cLted?1l|pl6s`YC zMog$7al9{+cMAWlW70yJRXfSOI(h6?@{*igkcS_~G!moLw@B`ti4b>Xk}&^_R5ua% z{M48J7Jriii0yw8O65Yqgr8gVpUM*F-X>@?&jcTD8bo~0B$IgA0VKTJA5M#TXZYBB z_0u4wbdY36Oq%Gx*K|>EmJ6{QWlYM3OIUi>O9d{j%Sh$LgCumy1uM;mqJq&DE7{;UsatYG%8?45%&T$)29@+}RCPj);&#i>mvvS}%HIHon z^_pxRAB-vCgV@4lb&@>2!Qg#Y2Gm!Z^Rti#C?&rkKbHqe=B*opV|IGMy#ayXy}v8j zo^^tpTOR;kyl-ZmSrF*tEP&vbj?#HhrZ!L&hM7GjE1m;eI42CIcT>ZO<8;{KX%8ht zeit7v5YTnrCfxQ^O&A|D?idh@ovpaYu7>z3Yr&`ByO4L{k8P-J;w18>mkQR;>IXGE zhr*!5myE>UB~u&^S`AhTMneNcr>@;WWx^F^v~o83s>I=>=iQj@u%DtqFWw4$UvjJh z>sGX4RiX;4?W-ecQ267|Si~=;Em~?Lymw3cYM`xHRPs3y%sdV;gM^-}vpf( z8DW22rY?I9EE<}Dg`>`~`nP+rn2!;vtk4(b#Mw-9&3d8l9|xx3bL|##sKpQiSr%+v z`viU{tRu_Bm!b3d56NSVw`@e@HAz;|D99|nX=S7hQW?cnwZmY`Qp$5W_AIDXb%NRJ z-%4g<5~Rfy7sMn*;KN06BtOty`m7L3F`gK0g4LRru()&paUD5Is5{X&lk}c(1#GXL z1G|BJ!>Rkh z&p<3-G3u;7#V>~?AM#q5g=GzSVcSodPiBX5Di*95j~(L)Ez3G2H4EXnT2Ae@5^!4QbrtgM4wK0aeMv8cHg+yC zjs19-iQy~f!D|)^dpC?1ogCyLx!zZSN$cHmqz9BU6+zZx57fxopk=Hw>pzydeFSc^ zx525FPsWU~SAmLMI%sH-E9rEqW!I;bT6$&85N$E5WJPW%WaZNt7&c3m>Gvrmw_^$I>PJ^pof0K?XbJ7zc*P#Ila>iGgXHEV?P#c!uree_NdX-W);$;%A#BPvuFgYMi@(F0c>(Cczj@x8^U-&xMxMjJWf!ZPR|bC1%Gd$DPUbnYQX=ao2d8t#vk%v*h4KxH^3XBl&sdaV z;Zn5%s=yBm8@J$$rQao0OZKtnANl8^>~At@n1gUn14^L%?qGW0>{cn>;=&XQx-+wa zP2N;Qq-6-0x_03P8~u4-Gss!Xo|Jfr9BuYm{jfI_`be6myNf&x?d30;f9SqYuJl~} z-CyKTLthM7^-DDQk{Yf%xvrqjD3X<2^n}=vnwY*jk&m(GFXVU7TZjub@;2goL(s)( z47~k)moK}4>3q5?n)|6nlCtA4?iu5a)jo&Gyj>5Nb!|P8wSzEldN=9)V)5t)g1pH) z_~#REk5xL9;*RNpzm(91Rf(xAV9gxd+v$x5_o<7VN1KxbNX`Fn;~0L`J%c^0&ZE^H z9US~Rj67Q?VplS2F}ywzC+CE4#*Nj?tI}KASK=VJC3#=&k=*h z9=)B*!WEHSZ#vC>btGZ{Hshf5-due1l8zCf?En99*}6llpaj}x!t1=7r_np}NT(&wanI!3znh5ib z&BkjU=2ly>%_X{aBiNA$HBz?ATM}T{9rIfGc;@*xu)2LBk{eO%-mEN%MO-^8agQO} zf+msPpHt9xW*tUtm*Zx*8gWKLB5>cqEF?iTxFIYQ?^``&<06)bzS-WB?47lrZMMr6 z`h3%hNz88KI_w#$&NX(O%$+{@-Ds}4m<3`b~_IcolOtH`rB1FMJAYC8FUff z8GgcXH$u6*yHS$W@8%Gtk{jghi_2t9?p9Ig)Ktl(Grib^v_2xEO@HS3JR@02`EXfy za&@H8Hj{gkS-Zhp?7nUacQZST`x(EF>`GF_%!R!ruw|bVM{!o=EAry=CnmBQ$u&CW z3inB|jjR_-u00Lr7W8!FPCmVZ9^>rUBmL8CmSqteG9pjFz4nO@851UAYis$~C%?3W zcJDMkE{sc!#ht8SN4P|OTNY8CpvP9uJt0}=SRvVbQIQ&Wm&2NwrjYk|6mM5Co~mjs zlB7>M2~>%ZJKK2M-y65lT5A+2b}UgSf~y;YNwdNNc-pd4bgJV!ZOPKNPOrE@!)S@nwzkwZ%&_C| z-pZq-rzA(D9vDxSYkz?shYhL!F)wfypT+j7UR>`vQ)s*E0P4BZl2&$~1hu#4LZ-(& zkz>ngkdK@Sdmed8@e}5sA(I?n{N*_K8n+lcqJ7}0{Y5ZKZsosi%xTMPUHF>kLax?r z2c1pHDC^<|qJf(9RiCpYx|bi_w_S;nwKDX|c3FD;z)teY#tiER{FFHQJF! zOL|4@tnZO1-tWmUECt<%HnQHwH}PZE878X@!?;<_q#&UcOrM?a3m+v_G*n!`(`7PMh%1GNZtm}NM|E>?@em&@m*XM{? zuy-o)(}^oslPcm)4BX2ejHvl1=bf@M?A`KBD7{abEWFj}GB%M;P_Tin7N>}k>`;>a zaU#E`r2?Y)7o~H6xQB}>G!NSkfnERXdo&tHgSDI?O}+3M^vkA$Th=h49Z#KZ5zCwb zf)4yRYD$})imB|o32;cupENDp4CAxqu&P3xism*7`D;>rAt^iuypl4Z2frRIt;-~~ z6qtX>f*L`85ceNtJmn<{LHJVZi40zTw_mxn`|vieIP@x1@C|}Sqy&Lh{0qe@B5e38;1JzrTgqmU`KcW z_|)=#Fgc^B(%?YhUccfUkag;oR3@3?97?D4zrb=%zXFZkH;8u49^w3a-N`iJRyci9 zCr=e^{Hgx5-t^1yldvXuw9ppmy+giV6E!9V3;C7Iv*FD0v7mcy1Zcf;fzG@auzlPc zs6M!cHJ-F!-SjSq28gX-(zqjFY4Dr09MPtBPSfGPp|W6}*FX}CkI?)5PQZ!kckqgN zNc%?GhvJcE8{wE-1c=J*+5I_j?CJAsWXz@Ibj-aTLR*qw_Qr2I`#|nl4_YxY0y6LF z3Fp5j>?7;NE8xb!R+8Qv%iEa0g$MIWAbI=~$Zs)**)r)Q`J+57*svQ`-416>a|Ay@ z8KWEB|F%#3|74H}t7PQ8&Mac6ewr98IR=|PddoV+eBS@@V<%fATPPx1 zB`MBaqdf}IqO?kj7VT6DEyo^}P$DWK5+X~SnH!-}Y0uiCC|hZlq|NU>cb`{}@B8=1 zJkB}yndO>mu6veiuIqVi*JnJ%<5B2b)Q;jFi@`jD07x143Vu{?f*L;=oIO8_ zqi3ymGFl$s43i|j3o~`D!YAL!_`qQ?xV(M{4EGO)BimKT`|M0MhF3e>UKtEU$JT)F z+F`ioRw)WjP=t@=^5EKy$b~`oNKn~Ju1*iRN;q9TgRZNHVR64_5r306Y)s5{w4g8) zv|=7Y1B+Xc7o!a*WCk}z5fLGfyz&VAzWeB3AN7^*mrN=$qZf}1K(?DRtVC;Ic#j^) zo8Bmi{F5kbdZ7+NCJfR`Jh1U+J{t8&g`XcA52K5uajlm%>klo3ue_?~r@i5m%8I28YNhPM>??BL2J9=$o?(qzxFg#a%Vz(@uGvz54!s(y4nDW@LB1pc|6F?% zi;tSbFDg0!b9B8$Wq1$Xg@DuuC+>U1bnboPWGOA74c<%rf|yI;U>fKoqWwP^6DT|V z#Vl#Q@PLY*Gjdx=!f)4-yraL#^_)JqRCyXk^&CMS-5cpe@g&%q@PVvq835JqG1RL_ zo3?F@rag&6u!*)d_8g}STkDh{ceEUk^D%&LCs_3mG@>USc zG-J>u9~;^$ew?l!^q03uXQiQjRt~(D*a7rI2sE*N%QIhT(!f`tkcLUj2OU=Trz`eqr5!boCh#Tf2+VqBOE*gmw?(^u>$LVz6%p0Jy zwVTk>IP*p117~sk=y8FAn$M;mTK=r z;UAOffqYvU(Vxbj9B7D0q7m+QTM9SiykL&n-}7Qyey6B?!Q94h=u&+Fn=Yv^sxv+a zTlcGgw^0akOa58%!lel%8eaf66KnpH8Y9vABVZG+Yk9c{heu#8=tqXXr_os#&%w6U zHT-vDY+NqO#ge?j?%W;h=g! zML11jFDkh;0d@<-v6PK19RKqkX_`sEzA@c2SfEXdibgZL-M zK9!(vtC@hNRpd!Idq4j$hp1=&BzwLu0*}m6q@8}4w#jbNq{h%2% z@4bQFQKO-2x+j$iGNAKxx1!VaGoZZb44oBnm4E7y37Oxx4jzm%AxD1tBRK~NZcdo@ z%9SMPnxhFa7tumB7r5Xz3FYo(ZAjF-;abZwe9XZg!p9e*%j$Re{l@N8r*9HBF7}=c zAa8#a7Q9u2Z{nVinw16FgB1}k^(*L9rqi=t9{fdWE_mqm2v~J2j`nX55VP0g(URjU!BTz~G zTT59#2Jtkyx#bb{=v+X^`)W8eV-<{Zj^<>QQ5oQ9o7byC%i7kV%A)r)>(v|n*Tfuf zkIf}-&fZ4D_YNS3r^)#EdL{bRS&1Ir_t&Q&PyPufKeaB_OsnK=;kdH(q+NMBa@-Ze z_hq#40dsNKJtQ8I@@!BDq>xqKdtku?JNVeg;%2mbK>@4vA@<>WbZY|NX3f|u!cRTQ zv~zJ4M3hfwabo2;`s%wg$hGb9!WZ4Xpj-BcAHRP;`Bwf4)ql7P+r5Xtch-0RM42Qk zF&RfUP1y~K3S!u_CKamQYY@TB?KJlHV+cO*6Vg~m3`jlAm+xvu3A-;sLyV9+`@dP1 ziRSNihB2)D>C0VVKnG=(G?Hqt9JLs{^DVLVcWpRy$qVN-edg$~(nti2`;X8`SssSHAwTVN+8~RA>s17u4-fgIPXiq*TPNa43RpIyT`Dk&6 zC%R(s3aL*DhxOAGu~qI0ymM45Y+ZMeSdWheg-8>~j?RJV=SGkxUM>tzdk;$#M}Xdm zZ=!jDCrY1osN?*Ha!~MgEQav+!mFuW5IMyd@2F75Y8IpL)LB<;tjgqZ*#~Jn{g}|^ z#?A=%I>HV!XQSX?kPY~{ErO@J9tst{Uk5#_P`W+o1FfqX$7IDRlM=f_pzOAdsy@qQ z$89KH!D1K|PUrI{7!^Y1xC_YfXen}0(B#&~t7~WD&VaL!HHgQp^*t-1Q5~c}|0lwY z?}r07;N^xQSnGNfR`@?hvzFfFo}W1x0@ZzL5V`z27^Y{C)rFtY5apMUR#gqxhDMf* zZ&QVU@0W?+&Gr27p;Bo2vx~xklhRnTY7?&i!UF|AIeN;9hijFzY4w|MNV0v)r#KsA zT~(>>wLD1uK7+mvn~C>oEP#L)@m!n6BUd4>$(6|4WeM258i#`y=7EH=Hasz&NRxN3 zAzvRp6JGLviYD?GfbFEsuw{@HU4L~pTK{t;Upaj*>K|Or)njq3Q6zJMlT#N%iy9jb z4oO4Dw8LPMu>y8YIfC1pmQv9esp%WgxGxrT_jOt9rm9HiXtjg?cP04bpn#2dV`2W6 zB_LRu4uRvX;p(YuxU8BWv@FsT+7+Ay$4p18bmz42cCrIn{u^^LULeoLdP|MPq~&ja zHY+DX)fx{Nd#_wrqNI-7SDb>3erGhu*cn^R{Rw(L>tOr0CN{obG`QW$24Z;9^sIYh@%VDe0H<&wbHrlr%UiffcEGKh6)<@vl zHPL*(b7?4dVKpS~iNoL9i(u%KRGb<07Co+OK#BABz@JP_46hbK$pdAqv1JL~zpWSM zu2>8Gc}Mv2hE?d(p?LCfT!XO7LIv-<*#T>FKSNi%GCjY_9?Dn9Vx#M6(4Fpx9}lwy zG-4uL;GIV7p#}VoWk7LyHSBddU9z;-1qSL3;l%z_=$E#JYh?$)y!ssWxtf9T_DpPW z(gZBdb9)fMfA7oxeX>F1T~G<$$ju?5XFqhW1Ic>JjqA5qQ#dF&633ic0<*@)K>9{4 z{6@zPzjZa?Wc$CmqJ|!WJ2vNG?{Iaj&SKBJID8$}#f`#CcDSRO82gRB>7~&-09q-Q>3irxWJ#GOqpq>i>T~6?Zzolx_21 z|LPZTV(uPrHdZ0MR^2c&)(JZu9Ernc{rR7=wYFbj8r3w@FI}SJ`Itwy6nX625eCoqRr2n1WFBr_mlgnK}S!XopYGlBKyrEF?VFT=X769wUso}ICEKc5&Sg2IG z3(MOnjLc z$==vORc`y!T@Mn;{$L-v=%+dT?U@S0Do;{eaGxa3ScWs>F@C*$25zRF=y;9_D7AW_ zsKx@}$n{@gmiWdJzsIjQISpq0{7t8?=e~=0*t1QF`<@Z>4YK#$q_cl$lMWSq>=Kef zgcCJ5IdTiAD5-YV0nFYlh|2tzlA;_T@bMikkKvpaP}Nk#de3>((SH}&&R9~}QdLsY z7Dcl^=5ytKl0IUt_KVEAqD`Gm*t@(JlZ9Etk7ft%#Z}ebPDPZEJj(o#88~1Dpf42B{^nFu;fY|>D^-qH{&JH(ONB#3_Vx!bQX&-wB#6V zeQFEOq9)LQnjws+pAXL_@jYf9qQh6WlXu61;H`8yd>LU2mpWr0=!OcMwjIpKpuH&f z0H)QWEzwSL9(e1j;KmIHTw0z;?=+VqFP^z@wsbg0kLH8TXtQM)j8b83+$HkJ;M6B1 zHdmc{Ppg$BxjmhwRS{HF>%gy`8MO3a2Y+dL5h&No(e8a`XjN7RxoKl6yqTy;&s}XJ zZ$}T|c)QbDpW{*Gy>m#^ZxQX^`xwX_#NjXTR*pN0ly6T#(-N$a`OrZS7@LgNJ~_#i z(FweboXpHc^zq(}6Qf~#JvuJyggG*hh+ z^Qq;KSn5^RBAlS~MHs#87P6ai6m5@acepT&RW zSBl`IB7I(b5^OiK_Xh5x>C|!^ns79VE2kUb3r}o}@$rCrT;GPAAC080`_NVwX^t-= zjZ)}m--~caIs>*CM)0m1W=ZS;DPxLhNj3_Tpa?f9^i%(k5&=CJk z{DQ&5sGau!9dz*o>f0bq)UThXgvBVoQ=moVWc#70LIoJTBPd<@7OIffq#Bkyru*9@ zWc2kV>UesBtFzxyoZJ?lV&fBl=ynZNi0q0N(J4?Z(Gl@qaNZMu6#w zLVDt`DkL6=hbelBbl8Bra)jxBBUL(W;GdN;ph`^VEH zX6iGt_TEYmHs3~@9xE7DU=z6yo z)IaVc>b&n64(^j2QK5S&8nJ#j-M?oO z(!BH;Rj*Wl#qK}3euWi`B_qG9QU6`Tsa?uU6p1c#&keu$LB3d=aMTEI^fFZ!LO-^0 zZGD+O3ng{N@jI%F_}$r;_#dZ_;P^I*hEhM9rIdFe1Tq8)jPLVAB2c+SET*M#_5DY` ze*XhxtM!Y2d_gStym_MudSs@AG*15YEy{P!AnB)g@S`BJ*u!%Lxngqz>3F)4)wczt zIUtfUmy|(AW+!>_?L3W=D&y+7bK)-^;@rS{$k9p%g|xe($@@=$*vDvcX=oIG(y0R= zb9WLs?_^Bht=5C0N*}b}S%c8eU1YtD4w1;|BC_UIP`~jnzK7s+F}k7HhMtWv5q>^# zLnzbY!14BU>1Ni>*B*t(ZYPbMw~36!2-03On}|m;Xhrl>sC7*%cFdRvk9|I)(*79q zQsNcL{Pl<9&&jWsxjICAD5B@ky~o4ox}6W9;2fc1#g|~wrVj*`54zLdt9y>9Q8oVn)rkh4koomlv!M)QqBjQVut(_tqv@;U*MIcl!B!+G6 zT>t(X--8J)NVe6N6vfS!r$Y>11F=(4^7`OKPN9Wzev&o$e@lg0v9;60d z3+0wn@N3_Du{N;EE^G#gHsM@t~_WBL*LUt$&+vvlO-kk@RD#Af-gFil~D}(vN)`HI{9yYf6O62VQ zao@dWF#dXzI96*g1$AkVVephVejf|+`g`HVo z4ploRAw#VYZY;#bYgkO2QuH4`T#Hg**s*qwM@H{Hq9fCkh37rJ;m_3}ByLY5o6Epo@~=%E{d5%;$z-6iz1r*^`aCSSaf`S%9e|&ei7ftDpm0m2 zD%_Vk3yHORAbsd$_{G|LzRe8*?R6z+gKiv=-Xmv3QItUO|%C z6QtwfW&NeTou6qi7A0I!rH*e^p`6vfEPgm0oHk9ziTx8n^Yt12%Ge!5hSejvA2j7h zdRzmw4L+n{{UEF_bpf{R`6zUl9R*b*e#4c{cS!%>b77^b99%H-LtWQ{!1%QVE^7*e zes=$y$?DnD9}dSiHjRW7RcUDVkcQ;f5im1A7F#4*VVi(koO~v+`^g%wO)#lt4lE7O zg_X)nVU&~>LQ znZt_o`-rzo8=G!<3EzL5gvcvdXz52+Q2o=xUok@ho{tPir8lI&t*Z(}Hq8Oghdk;Q zF&>PXgJIC@)96P?GkKkWAUHq(!#Ae@^uzN4O$PFO1mBX>AK`Z=uJQDOzMxwN= z`*5^q3cRZ+g^P*OApiZHu(072_zpda3rxyTbW}W)6llU)R`)bgT8%242%@X(9l^+> zlN@-5fL5J@-Afmt2bGSbQ`Zk3d`p4JnJm7Y?sc>(Y6Vi>S`9Z_lA$|E3OT4gM=_(m zq0A3#%+(fz)r)eG9;?rVTq*EP4TTHc^HF7D5h~nnh~={Kt>1JBjt3opXDNG8(v$;W zx6lTT$u^Sg5LsMh8h{@M2!(a8U5UF(IgHingnP~`zNIVR2CG;6BwHi!JjKT6vb4wV z-CrW+Z6+GC-4{OB#PW+I3gFe*3N&4%0d)!)n+w3Fo zmR-Ax(lq#$ks+W`V~opd-N4z@o$y^Jism4}F*FE&TYaA#RX&R59A-86A#2FFgHbU0 z*ceW~v?)&q71;&iKR@@Pw+GI^3;zoE6SWCnRN9HRR`6K=-8AsN$Lf@W&)B>*3CE)z zuf`9Jmtq-(4C|`*k0IOhCHf+7jT04qL$YH7I4F7JWns$L^S}x0b5;dEI422NQa9ni z<8=5F_a2H{PGY9@2qZ3M@iNrfVBLXYG)4716rGq4V-ky?{80!*`$~{=rB}ezqXjx% zDPoVwPoOr{6>oc73RM-4kl0jR@S7b6Bi?J%Z!$+{`=aqwd3*z8EPDaDVoUJCM|_so z1ct@kfn+BGsD11T0mfIk`K{ta8(dYMiAQ-x!HruN;K7CGINIimP__RV8Fa${pWohs z275l@hoLwa?wtWb%h51rOfi^h8o={Jgp8Ipz>^!_ki|L^ER;6GUTZhQkD^lkqdB*( zB={M?CKk8!@lOdnB{+G?RW@i?pVz6e#UoKD&~6ZYJ1fs6Z(q25p{yfSbYu&AEc zq-R$Nmb6EO2M3c|6aNUa{H<_vq!C^-a1MVhUkl5Nw!u5rR_5^XYiRIO_I~`BKQ^5# zi=W1rV9$B?5#O*Dnbr=$t6EIK;v)OrJ5?dUz2E#$g_RKR=LLDK-2D(w{k8<2EPov| zK9q>&c7gmR6?`)25*VkoL&M-yAo~u2nA3W2349J6Z{}jt#$epx_#5umD?mE45U<;y zk7bW>I(C8V-obcE(FlBv{}FxB{Kd)ANbf`#DmRR58@(Ou6@#HGJDB6SC+ib;c`J(v zy-f+*9y$*lml7dst^syp?VH90t%3WE<3Ykn4}RKGm}@47FqCUNTAx+z+d^jSTm z-t~n{?C^JQjWLp`<8b7j)jzqGcI+;B#p# zzBQ;H#C&du_VBzb`r>qT(`Jq?Tkl*@yV8OdoqvTEDl7mtkR$N2%(4C}A6$h;*$74aRZND})e9n#t(KuYr+l2BCT@@U@767JcP z34+C(Zdq_r?eD!uhh{J&WKPHTduywwpL9#+ClgD!Y})B-=v8H;ze&F4>9DgZor1Co+o4?FdRZlm<8@WmN{DYI=ztjK<2Bn}Ys|Ue_mm_HSmiI!R zWi!}#fchLQBAJwqR)E_5Q%Fo+9jZzlK(@a&pegsG=_dBx>d<)sf2RQpmu58~MZm-G{$&@HnzDYtd zx33FrjHe*|q7TTzVlCQopb+)woCNimUerDO1B(^yO**S)(l>22wD-3vX_8)o+W5D~ zqtfkUadISj+T;VT_E9;tu^Dp;@2SllyPxrN|+3+Jz|0UOp3@?EnCD3`9pF-IeiRAZ!Y>PlmC^y zlk^4ss&j>y`X$Ke*>K=jR)YMyUesHC7iE`@CUe^-l5I2ZqTuWO!W5^;U>H+Cs+KxY z<4?TDl)bMNI05t*A4M#W` z618_#+8FssbfPL*)(*uO*`PuO*5!(ybLq4yG8Sw;J>uxS~6&rxs+J{3?w`KWuCVDkZVAK2r1Li_rzCndH$9 z8yeMosU*N@HSJ-6x{}3W`0KnAK&9Qg1htCu`#zPDmMjYDdjc89!Yt`mO=hL!A6PSOmv3R!gAG(AmgW_AugxP#s+9CNC3`1JMQKpby^YlV;4ZBIOc@pVJ z3*q`uzrcu%sjWf>26!|$HAEy+f^p+=QD>MK)j$4{f67^vIZj6a&nKKlAOjXB!IEb9 zZ=Nf}PO0P=Nh3>!j-b6MeCEp^H{sQjR%Foj$EdzV-$ubKiC)_ij23LN zX97a)7`bh~=;oU1BvLA!S}kBPo?A8P8tY9oA?`O-+<64vI7bOT*jv!BaTaue`cqU} z8%_m56AAy&D!Om`JDdFGL{Z;(4+S>VJLoy-D;`9LKW&EB_gU;>OF5{haAxeW3O7Fa z%_V%pu!Zzz6+<&Mj?q-aFwU9+}YXXiw7EfFp7-cSx#KHtVA1souS`Z zUzT|4SB2yIbXw=+o@|yrZ2&a=$YLQN0d0jm*c_iRaV@mV66!n*QO5+ zr*mz54)((SUJ7+s5~WPEqSoC7^r`a#S~W5OT$Ku8uGdX|=#r75ast~Ci(%9QCG^th z5LpqOhAupfLDM6$;q~X2^l_;fM??AIb3~N4xzmR$b7^EF>`5W8dv_M!MKMU2+_P2a z?SG3LScYNTcuCqn>^FHgzm>cXzslFu+eIW?E<*CBOeE>9L3`^2bb+oKxDHAIU$Jif zW?yZ(cfTeoJ+g^Rwc0>-_yy8wdo{`N;09u^V#>8AYwilu$|N8+RE9}W)nu{`OE4Xm z428Z;-^mo+E5(Ory7RmEn`oMGg7CYkh47mCHrituN-tcur>Fhjpso#X_|fB6*?cQ7 z1e5h+QC?p+320qN7QZnU#x~y|HHk`eR@~osaYltf^h^qikrpFE!%vSPfrsOXwrds< zKRK1ge=jdtVel0tyVGLPID|GYBd2)l=r*%j5uLndx(yaqFA+|zw}e{fG34<(E2yc{ zBhdkG$OPm{%D3*MuKEk;oa2Y7Zlyi3kLV-jFqdiz9BIRkVcZzy$6lg+gN9LPwk1hl zH_)A$Ur5|}8M>Z*NAfHWQ_t{aR3>mC>n}J8J4LJEihL*L#oBX3^T{7Nq2@T9pWA>g z&anfd!uK%Sq7cNZZlIo)7s7p$gW-o;3H+QC%xJow!f|*F-t3-=@7@W7tfz=hc%lfK zSlpyn_fwhHC|9gCCW!9Q8BT9z8N=-#CmFYq2bqf@b2zyb^-GkNU%mie2N$B*YZXDl z(*VyLyaLykh;w^X(f|AADB-w*F534KAVnh&%#Vs;UI}Z%R5u>Chd+duP4^*E+X`QE zbpZQ8Jm%2VxpYq16sRAU!+Z=*Viv0(XWoBKhp$s&arvo8xVQN*YmYOS);q9Dku1_AJX%6 zaLcY?II82dsElCS%B8Sl^>38)*MH<>b?d}9+7ekudn8_fYhp1jF3y3V+_QMJVm#(+f2Kfe;q>5C zHpaahUS}l4SZf_){yK-0YUv6O-#ZOD9p{;{Hz+YV1o%HWs#t6?5>faZonG}&($ zlf7mr(?a@a&j%Ui$rN?^H6;W*pJ;%E|3hdCoXq^r_(IGj^q9rH>4?9U4}7y$68&iu zBkGeV<(~@X_)N}=1V%EF8L6uB|L``Elfyqi!W4eS;6$OHq$wEm>HWchYg;N!Nmvf!M&qDNaamC z_$~GwV(T$B9}|amQ6uU0s4Qw~4J&tUEtA>2B+gvxJj2TRQajz*D8 zjL;Vdg+=e6fwisLRXr46i`4-8h5-2OegtOx5~rF!k5fPCrPh*Wr{H+(Rs6QP0hgy2 zGD{_uAgOi^-pN?P9D@uHtdfThFZNLDBq^N0V`F4Qs)Dc7GG;;N3P#<380PJ>#p|z3 z#+8d+irVIQb>Z`1X3MECNPc+%w%j8yew`AQp1zd{%fAcm56$4$&vL4#t%qZ5XJEPa zHdrBO3`FP;fdfW^Vf7$OxZx#1vW;5)^<`&RFzBDvrEj&usHhE>2Pvrbk^?;2VT^yf zz>+s!)mSMkR9ZS4FjvkcKIOt5_gCJFkMar=-wL7kj8tmLZe4C6tbRt<02_ z#-j{Y&*&Gb4{w*Z(6RS`XgoMhJNXG@*1F})DVwRxh=YgN`)4zB`FJGRX=Ot%Dt6L9 z@(zt`+|CS6K1PCnn=xgF_t3JyTE5btZ*Y;6(}%1-SNiKKOlV6UrNSw~*{7A5Y3+_o zf9w$&Q=-O<6g$i`&wWiN^$&t&(qT-Nktg$LULYM)I*l(iau8KIYr!}yI?jami_>S_ zy)>*Sj(%qSZYO>ENmf@fbiK!7M){dD>N0;s^SzfbTCq;dkynQyRX7~S=X|DWtHO!e zqb&aVdy~<$gF(zn7JJn6a5s&WX{QGTBN$!T3g$x2TAJ^%ndsV_V}i>}n9jOPm}^)= zEX-J6M_Fyg{G>H=(pX;fZVJrPV^M7J8YXYTR5~vDE`41n$(+)zC8^UM(cSkDW!`6_ z4I$I0!RbV@wYgL{JEM}0J$ZzlItrJoLwuBTF~MzAqEx`b2gFVm#f zFLaX5O0sTA62Ht)k!d=XOqcZ~)3H`s%$H@2#PW$5<7)Pn)@e*;lw)J)#pN&QtGJ_7 zkUAA`@I-hA|6u}WE&}(D$@uL0V;1P6;nf`tgbEozkjf(wF2aYK- z;Q@h6p~-NjlHHqsQk=|e4r!tvOYRe=@FPSwd?+(v#TvRjM3R|qL#bumc{=^pP^RN& zFgZIbl0f-P+On5Y{^cn8*xP{lBQpV++|6U&CA`?bK)eD7wY> z2stq8VbNlbiSWMnCSAL#5xM;+vAMd`4yK;=XE0THZqdm zueyf$b*2`AEqdtJLy63I!C|W8{S*1#x<+j$sWalEl8EWBQ|wvbCUu{s%#42|pbjer zGwoLL%s#)7jKZ3k&@|bQQF&5LC1$EI)&&xb+MZ`LbE*N>Q~yLwczR3cqEWgaICDG~djVneBUr&b@2TY>;qeP2A@(&GUxy z+kRUzwIRi5N8K2b8Tx_7sai78`@-nS;H_YsVZ+*l{FR&bR}uWtY#e6yUP28|R#La+ z>zVE*J|p|xf%)jPhS_WI1~e8wg4Ok(;GydQX77UlFdV&$3GN;unpb#ViZ0WU77^4p zGncND2x8{Cm(%?lHqjfqk21&7vk`wFp7NS@fsSKRaO&MV zfmzaboPNf9CjWC9Hy*4ox>Z}#b{k{rd*c}mlXAvQIl6e7*arAiFNOLDcRu9}Nsgjv%7gK9$15n};~#=t z)^hWtq3kE(sb_}AdVO+F38(Kt34Oc-b&8Tn@w8tUK0&W+A#< zz}jb)&1b%~nlX_Q1C-Y*P4ni>fM1G=jJWjQJLgk}6KP*@6s++-1S`|eKtNb8eO5Au ziEs3R4P~KVFvXIN*YY=+MoP#&-cGBpSqx?XBL*`*ZAS^#QgIj+^bk0pqCEpsF zP~eHBxbwXe#4nmF%*lwOy$jorseA=FnG(id-~~9!bAQ~iBWjh_?i4g%VUtD(|h`Q;tl#$je(rA4~gI6CwyDC zsnpK$E4K%G;FCm>#=b!pGdcTEf&96B&_AF>kDDlgnO_(5FB!t@KCl#e%yYr?P+ zzzi!&9AK@|$F{h1S8Jg{EdTRKq5l!uVS1vfr3L zQTrksymF1-;CGtdbLfYItWR=2a~WnY_oaT1MnUz_FeubFrc7_7#l7kvhzg^cTeN7*8b7>bTN<4gz6gKTSkA3K-|O08LElKG z>=_RgG>wJjE63pIP-BoXbe(jn*wG{-)p zQ+k|)#ZxS?lWHA_dT^guN>`JpMKfu`hg1@#I*9ehEy1rN)v(sB5ZFIhmib=4i?;S1 zL}ilforR?LKbgONXfR3tnt-$yPNgrz(r|@`kjRdm1+wpTg>E^AsNq9rI;`*%br7V` zjmLfA`UpoNwnvSdM@0N)YGq;bOKJM9`wQLE+6HPFv2;?FJ@#Vn6NbioCSUWuLz4L& z(AqkY?p2UrRxK}N4!IZ5xpT*f_$QdqUGmFIT z(mR8ZpAc`1FMxl8!?PRMGg~{(@odO&89oYH`BIBXDE%%qE}>Z)4G*9 z%nhZHn6ER8S==*)S+`x5vClK2xx#tE&FdC3Up@w-lQRc1vp4gYqHY0gcp$@Eu02fm z)IoRL&iWtKh*hWQ5h zv?_fyZLwBjOuAo@sx#3v-Ho-UEZRzspIJloud6aqTT8ilw@G<4xKAx0#x#HyxZk1m zI~dw|QwDNOwea^XI%v&hWkx({IrFXl6+Mt}mcD+Gh1ypgBTJN*(JOgFn9k1+X@|*2 z`Z?Z%ZZ(LZf{ZcLZNoZp48rVuWT)@~bd7FoDn|6V_7KX`S@7 z^Jg0)Elq}6=g}YDMYLQ`46jmmNA@qh>9`LC^y)blL!?5HS>UF|#1(s?=c`zsw{OOb zrd6b<&jLy3J2WD6I#y87k z=|(vOe(`sy%FILL$8A&E(C7nIBiBGx;|W^L{Gre0|JBotSRKJH_##7wF7ZXR4_46q zuUqM1m2^f=S)ExXH9+SKbE9w0_>j9|)0k%}LzpKois*tLjIeN71<_krOvSH@F=NDY z(C?5lG+1&E+;b^JqZdgtr+CG5R%#&k&U^bLP%G0laDCoqSMSTz&QzP=9n;nO4j$#D|41JJ|?nEXp+ zZQ=Ob;Bbl7tghtFb^G=;BZXzUOwHltH1tw0EM{Y}c~mIU!kLjpS}WF((2bAi+~nCz z{kF|O0`+KCMJ7xr(||xcfvM`6z>FU>K(k)5=f0LOa&=TAJuyZ}==b9a>yN${oT{^s z;m*_W(8n4MumBKme*~g{w%aI3CK)XjtcHp&TcE*S6GqD9fX2fmsK1B3XNVj?YO`lU z%bsl@^L{sJ40nR-Thmcd&SjXqx(DK>3uqrx3hnoU(6s<3Hn#F7j>pSX8zDyFCahXl zk4ztBl0iKRSW)FZOw-K;XU{KSe9MS_zwbon7QFzIl(G2Pqq9hQN+0M3zlRqmi(pRY zb2es58BDmG4Z7oP;o|5Uus?1nsD0kbAK~%|MwWQc^X@LNUGowDlIu1&^6V}d&Yr(S z^Z7c~*YVN#dT_3@gW|f`9FK9sb5_?p3{PNnsG>Z3v+@C>!;tYn1#BXk;liMouz+zF z{U<2ClLrx(wQ=JPdn`5n?|J)0<|DY#`3F>H^1*b;GB~g93!%DIU_5ad?49+7zqr8` z(&7{#<47JD30YsRk162A`ZX+eafBi73ejW@1B5j;(@lTM;cdZQm^fV>4Cbn!{cN z*&pde5|PKDs9XhjAr^R9ChMywc@8uedBE>@4bZ4R27}i%!li&nDDFadYD7OW&m2w$ zHs_Inb6sfG<3WUB{g$qU-2zQb1=Kb?9|El{MSBUs_gP(#dLR%&{r$m2piF(ckMd0( zbqH^qR-k3LpTEB~5gJyVL4qDJ_P%8wiw*FAY;rnFx{EoP5^OHA;rcZmFGNA}wxg8m z7ogp-U+AjPgX))@0IU@Wv%;2xVz(+8lRgZp6@x{08iME3_rl?Wwdi%wS^l2Gju3Dj zgL2qTa)Aki+`G-luwxQ@{bT@=WiGJBWD;>4Rmio&i;scRvJ-JYs02rUKWqCD(HM_z zT0Mo3;mu_JoM>9$!Q=9J%?*PcN}f3MwktkAstbCSj)9;PA4uIudHS)t7;W1+6uxXf z1Vx80!rq4Vk}n>4D22rZjyW3%_X|8>%xW=+J^2)zs)9(=2OfITzR3ExpE(e*nx&!&FXM1+><^`&a+}wZmgSxn%pp)8^_o7x*zm z1N;^bXO5+?XQ)?J%rf~Iw6@5RdVSr`{~4XbtZ~p_#C(S_`TJV=Q6ILTgBiiJe9jRl zzQO8Uwv?dXFCNhRqoZkc&;$Nk1pz}Od>NlrpQ&(&Hp#QQK}XFQMb|evGKFJiF}(-a z`@PG*NK=~!KV`{SG|3^nWJez1w><11_fjOG;Jz_UJ3p45mADB;JG0P|X}Z z9p-m}WShN%n|s#L2cCz~xrEo~tXT(=+4YWZs68HaUyx)rlszH|VIOIbWjz0IZ7>ZE z9L^ke`Hhwb@26qf{%m~LH}uZs#f;}UJEoyn3O_5IiEG-Pak3+#Iq#Al@fsCM7xhN* zOTJVR5#2RuS4DL30_G($nGv%Xlb_dxv+~)!wsIcx_|ZJ3_WVVt9e8bX#!ZI4zg0p< zYdoe~3N4u}>j|O{E~1TYcC^l?K`15H#b4Ssm}&A)qCGqGscADCJ4ww8#eNH;t{avx zIyYYsCl-@jDl7`d1^1yDqoXUW!LJ%-Ye-v%TIK5bvhkA#F(}|YNo{d zE4lvpC%w3M53^*^diufjH?(~%CXHw*O>LVEYwv7>`>(Y)-u*}C*1IeI{f;%Xp#v?( zbk*iXblB`a^p8yveI-2kk50CXpuqeYoxLNUo*yei&l?yJSr;I+Y&^W~NvcBOU_G!3 z7{ldXbVdi1&3&B)F$TG zkL48F6u7h4|3})JfYtbY{llk3rKl*%m?}N@$*tB!pCk zsHD!`C#g(jNMtNSrj#-B%zy23zI~tP_q@;bUf2J9&vkvSwf0(Tuf6u#>%Px)-}^Sy z3x}e*cy6=Ho-H3=!&k1%gM1fzK5X_}QtsQ3xCidxom!7jPCUI$HII_i!94i0f$%T? z-_Yr@JFkkn&3o?nF_vWOltaV=To^HKLhW0blaYCYb5q$pQgo^$ zH_>ncUtSjgo4j;k)U__48FYo||1^dHU;44n_tVL>lvJj%!wU=|Lt&E9Hpp4O3)=j) zXPS*2!MKyeA;dj{rBHha@^j~_IypKt#(fYhO}>*OVQuOAbz4B!a2adFOXu{J!zls7Pue%K)7JfV10lWG<-FIjd-3#;yzzvu^w6M`NNGQI$;9@x4Q}U ztz%$Y;z6?F$|JJ!>P82R>;?4em~oUQc2mq)*o~}>?5WhTG$xCuAB8w}S~_ZJM>y&4 z%15v6Lq=ZNsd(yv`>PdygSqSV=?H5po;fR+*DV>tid+)7dvGhIYnyJ+{h}62nrg^w zto2wx(_E?X?oQAGvv7P8_Gm%<{wV?JqnEZAM#!p8=!x{My)4VOsFq=sDS50`) zqa*vgXE0Phd`~{Nzs!`gy0I879f;GbAw%C9(Z1tk5Fc$syKC1#@4FgcYIqFaqobg3 z`EvN{xmoJ}wl%3GGvO5e#`Z0yFEg9lO_AU{9nPCwlP*c$N&Hy}Y__OiYrcOcKlW)! z%Pyyohm8Y>sNI|QUXc}>f02bX!47q=_vWyJ66O)}!Q>bazn0mk95cUup3++j9! zwKziPm@_0TY6tAka*~#{?#!y&Y0|KQ2cS_Lsi^)jj*M&Zf=t=cjcuCxLs9j41M}!? z!7qN;#6FlLl9ntd*SpO!7{4V6F21l+edhT^BVoq?4O;W47bq6A=k`6X<#z5E1&`$k ztaslfY{{}z@?oYCzcels)-PKOJbwmxt@)QMF1^AM502!%;a^A>XCE-Vl%#myb2mE| zdW5;pPa~19caV7_$3fQDL}+&An{?WoK$t%30Hl2#$p_6fh1xNjNy>vQta?zKG|;L7TA?KLxF_xrUc?PXj3bqS`Xs3RNz&hS1V6X`0MCx>kbB)TN*a9oF1WPY z#(uU{UxRym(kf;&&z%$;y2zdwk(?f%XTs-2o5;t_x}@K>a7A33Ib_VS?o83I7&@M6 z0;yV$q4@9^w(#v)S0zVDO+1ns>C7xRorg5nTKYRuw!-buJ;is=a`Lrn4CHGcB~bGmX8+v{X6NJiVy)-IpC?1`nK#w`xSy6RpUFHwvu1{oupkfb6WscnnflBkmIhndL(g2e zJ2ir@+0cSN-PVu4O1j2Q+XYEa6wCyxriSF@flKV;yOpADRPU9$=Q=^@cMZs?h>v(qaQxr6NoCSHCw{i_@U)8mxzi}@=r=Sc~T{UvC zgx2rP;r8%Ju&Ty~-MM>~Kk{+nBd*nu&-Me@xeuDThd(SO$>#^cpSc61yE=xEYKy^q z@9mDn`^`#eesp8#wf846%rqx6F6WT$GmlAKJH_M9IjY)7{|W zoJ{!90YKlSH+Z}3U`F^Hl=b&0`#EAiq*!%^=mueI^DC($HN6L~`nZdGYlf2SW`;1z zeg%8{$Qkc(x=N0HC17`lK)+YpRP)!HbdzlEwF8`MTkzJsV%V;I84k4{vdKEjr9@}Z z7yKWi0pC++$A3N_4ljn^P{{6|BO0UpNcNJpB>5~p18bQ9m3>=(zOF5pTug*SlHE#d)mr&l>4&+a1Ko ztu3@!lc*Rn$sH;ib%GhE?MQsjuF|82oX8SG37mOR2KV|JD2&FB;~zDDkwH(MD{%c) zpBd(@?!bf6x54bfFfz4OZ!+`AEKqtI!fh8<2+3?h4^A&59Zem1#u!)fSvQ{DovZ~t zQu@Q(R-eg_X;CmbKLm`Y$H9q{GogRao$zhBKk2EME8Q_cP8yQEq+*sKU;bUe!du)R zR&V=~nNiQ!k{YbdqaDd{yk1?@Vf2K7U7q^r2Sa=SFs zoRgLF7m+@=56Qhv&2hiLO6Yg@2m}?~2T#)rWSDK9!};emY~ZXw)+yvN8(UV(%629& z$0w(x(;9b&%@1)ek$rWFF~KXzde7J3CLc>yRAjTU0hu`+rZt4mH$1@M0)_7bwRqU& z01|KUMtWoUepnQbKQwTFz=?#Cv1rNQdTSg3C73FiIXAoAC3 zVmr7K7(ChuI`b{L-Dyp-d{!*!?Y$Mc80}Wo>(qJW z>Dp%k8=XB7o_oJl9WyKCGDwXdMQxv_K;KH-FLq;lcy-_c(;%+UJGPRv{2onz2F@hA z%EQF>Zpm2ZT~IUoI%v$80@h2j=|yWt2-Ef=`lf}Xb>lX4eMlib3%!Djv`&Vih>rBg z%MjQT^aYH2EM;chpOXVu8>x=Hb;rAdRY(o|9J7YppR^n3Nqe~aJd9Twjigg&{Ke-o zD#;!1u29wVI}Dd^RkYrE0t}tIQLR_5(6i|em=N=fc)?=Y(WwVbIIH?UR#Jug*R0U{ z0)tJnNS=Fl+Cv%#Z^L82bZ7u5mL$QJyav=uZy${LvlE)_j>#F6TS6pJ*-)8W0zX~1 zgX5T+up}-AuF49)D187NHqj?$!<*AXdZ*yg?}p%&@C$fxb82*H3v{Y$3H|}5w8@`c z(D4iIlk@T-1Vn_v-Zn|_VM;CJy?+T2MoRMFW>Xk`Q^se^dLn99zI@Fla?9YcdgLKE*vlybFo`uqjYB{w0|@?k+^l3jzl=k@woR z9nw5j!kRbPOc zRM*|e|LSAj9$td;UA3ssn8|#%XA~*4dWd_?<-w_=>*1WYF5TnWo&IzWA#JI8PpseX zx)Ym&AK^;eCB^2_Ww4-4GdQ_#Dm3it#^>o?fzz+U;f;3^p%o_dc;+wmo}GoXbsDs} z#VWWNmaLllm6||kbju!|cAF2$XFJou%M;*dWe8M_?m+D~YNou*O#kx@Ru)Wd|<6B`r-_bihe9m+-b-UX zy8mWHhi7}(1BYs89o+zC4XA{=d*=L6@T!~%xh`<3gE4I~=?0kN^Yzx4sje#&=QatC zeFIuayWr>kSSW2!1F{zyv_sSO@cvJ4I^bG9*cT_0uC2bpjc^(5kJp9#p4Al=Z2v^^ z`)I<&=!38#!;&WTdJ5H}p210?qgofM-!tg-=P%)$#zRGg(I>n|pc~C8d<1XjU4uqJ zQo8cDGaOC)4c+fhnzE(^y)r72+|SoNLzrq*NdELFKOX&}2tho;VkIaGrdW|VR ze2E#R=+XC+KAz#(Lk;g;$bwFtvX?7``Xqr=v?> z?7I!+cz<99-upYy%?mutLm+L+7jCeW!}O@Fe3iDF;@u;BrtPR6eE+)-Y*JTmE^56V z_fCFi5AivB9oyTpjESRgpOA27eOkt0mN__V`J1z-a*4yO=njfYS01o8Wg%=(UO4N~ z^q%k`c{t?}u?!j@y|5^XxP|8O`%AsJ|Dw^#rkd~B2(!m-DkgF?@C+Z<(vUzrQ|)6JDl@=f#A7 zsoT$*+|nU;uDkOSsZ!}+&F9j`xeNFP&wbo|;uH4KYL3*&D^$Z-w3H5sLo5Covu!F4gaCONwtaQT--RALonTE~NeCW@K^h zf`8{Ebo;~1lutTOR(2WEiyvJc)F;ikmW(UQVDF+M$^AWr(qXRY(yrSdi}yKs?(N5_ z`0IVPbg1Ej`xNqT+aD6+mTy_!p-MLB?;$4XUCJ^N<$Q694R82-EblhC15a#wly4Z0 z_l3POV~+YKS=%lH_?f{tmnTrUPmSiY@0hR9jK#h2uC48C6m-2Z>f8*L>0F2+fVTP-D)CgswvHhZD&f-jx-?iBg4#{-70 z)S_p7`|yeZ+EnlF1{m^S1pljXQ0m$V?}w;R-&>^HPqqKEL=vh`?X0@cuXqi-<=1n> z__qZOsH*;VzBXow@DRV_iMY}i;58WuPvKeXQS`)u5!|_2!k?GjR16*-EzSw@LFwzr zyIe<>aG*c@nNm)Q3qJgNt`ITbK4=Vy{cpg%qaEqfMBETG=MY5ASP2VTtN;I4-*p|m zG`pD8Dh|N>zPG?kwn2Kmg&};;O;yCS4umtY%^wXT;I40G{0*pboZq|@QSnKQps&t9K4h0Hk+mLG3$$=VxY%l zcJ=-T=GSsKUn?tt>dk~U!)L2DGaCb|eJ!BMI*s^U@5v7>cZSJ5_QBBhMNpKT30Do& z_thLYrlHW=5CI=-Q(=kIM7Fo9F31meFcb^`n zXi&Nv>LzN?uCw|=!GaW66Pd|g-agN(?q3GI?*TNkdWVQd{yeo6&HNOHkNO=*jP*^y zr9~_L+0+E`IwmSKCJa#3UY_Sz7C3JT7xvno=h>_1ov>%o40g`lgf%J-RHW5J!mZw_ z&uo$jk|!|u#sRoKOm)Aie1LJ7bf2*|F}{0}=-cPQlx}NCe%mw{RkWUHI6VLf%OdGt z_K>_6pV@}yuSn+b`Q-M>{m?acA`DqSgswC=1D%uP&|yLc=!tvWZO}TxZ{r?0bA3Df z&-#?>`yU_&zv;mKidg3D^_X8yx=T!^4FgXX3%2^jgn#YlY{+4x%80%#Ye%OI_yc}- zo2&ABzQPcW>gaH%{d#1A<5=i@aTMFxe=T39vqhoV?hHS=Z6~=K5hU^;xwfTB@z|k6 zv8d|Huf8G>&NWaOO!;ZnZ?9G7+c0XkY%mo+zc98l$60;qX zRdv+!#us@1R!ZML(Z~C!iy@*Sml!V?27kY)KSzofuRonYyWnwsupk8NSI&lpv#-(X zcXq*E+}rcY&xLeFpC6&(wfk_vgV(wodh# zR>UIkKzr>oxD@aFwU^&qVbn%LF(&b&9>9(CSXy*&* zaC2`jy5rk+lKp%*EkAq|`qy-#&kDn7yI%9CZAM4ht2O86N{mDL=0V? zcZK#*hSFF04e2YZdx~XC6yUvG0WCiSP|Mo~*@_AMRpS=u9fWYaPhtFx6RaewIi2%k z8-u&u}_~q!|5x&G%~rbDmSLj%xLdTKVim4J37Dj zV9FJD;GJzRsJq{Q{`AzO6S6rrVMy0`Ii}WMLo9rX`{wY4x&-EBAtXm`EmTNS6ri%aA z;c+B(?l>^aDu?vXM(}RU68gFQHtB&tL)yT+gsvT(1R+;@(Jy)B(we7BXlKa}@T4WM zCnp;k8aLnteVl1|b#H3KTZ2o8??)vzH1yw#}8})k2GTGma~l%aQiuX@6nr2Z%b(N;v~4P z_Xq;c$>@g7)#TIT)%3-8UATBSiDg%`R^{M9lt0KG_N2EKc2pJyMRETJ8T5~9GQITO zle+l@(PNiS($U%9;2fDwzu0W1^2PD=QSVd;8Wc#|j8{NN%qQYhXv@DmJ4^>}4yF&Q zz2J?HKDByKN7;&I)Rj8neNMwvV~JcixgI22jYdMziqjC;CY3eqd;&h-HlcfK)-Vb46$RY4mEg-7QUgPWtz_)AQ;K4a;%uJ~jv6VKkl5um>$3=13drl)^y&cZ!1Zgfo0LQ_rby;l#R?tk@J z7V_@1v|;dy3Jo4xlhavY*qHMZLQ3?JxV#Z)Jl;!s+7H(*jzbm z!%jt2ybXSDXsERDr3&NpairN%TS)n|j@d7Gz}>tPctK$@HyC9=vojxy^MU+QSx4Z1 zoLGbH<}lcGQm*f~Q+zA zJ6PinulY&4D4wv=hdagxvZV_j!|1U2;4baQw`q=%8l1_H=9Wvqs5vWkv?GIk-{T>D3Zt%&#o4b(=Skw4R^-{?$DlCD=BY0?^Y?ZiNY8{vJp0jX{!rPLt#*FQ z=E-xp*E}yCbY&G!=pO(v5#|K9e_#bmGg#Y3p&5yYM@Zu3&X`hTN34dvLhQBPo z%U90C-%_q}-e-tJ*){bjcOK=;o-EO0ub#V;q&JgzX5!*gj>bLdnY-)3&M%!TcSz>; zkzxGHvXOj$_n9zli#{LaQ;GL&ujCWvIjQ2Gn30M9gZxmFJOvBn?x#@Z_Fz7OQZd#aT80%ntt zwz({CX9IeoE<|a*@)ENw=hB#nR9M?WkGnj($y=CDCd>XT=$q_p#4VOjr>_imu+O$Z zbmLGDWwu*48oy3MIls9r2{zAHBnO0p!D`$q_f!__zN$?pKWj^+UFSgl%ulL3i}kCM z<2}hwIw|wFd!r*Kbxx>psSIr$l%(NPN8!^Y6Po!{VI*f03<&0Xo<<`;=o zcr{G@=BB*ssjak+Hd5|udXMS1bx>#&t%s?7y3;8IzEXwmQE2|4fkTRZ3aM_h84fgV z$&WW|4r2!zsA4X)|4tn4sjichi}(d9G=K9UrZQ#4qg=jB+m#M|*^#X17sTFH^?{*y?nFCkv~^SiIox#-@eWwi;Zab zv|eEPdmr@wrK8l*3#N1JG9bE(3B9m)8hnZT1Isp!Qud71$!+*53DUY}@z&wj$%zhb z%0@Hx@*Vf6s-Eh{9X0b21TOAMpU-Z_znaM5UUPk=splbf+IFP!+Up#481Dgmn)Qhk zZM?=c^m{68-U3r}uO!z4Mk(|vchX|}Eimp?A8Ptj6RwST#GcuVfvnJ(e05G0pY4^$ ze`NNDlGAOV`dug=Ibo&4!?RlSb&VChc=a%R{_+?Ob@pdYnPD&_YzOeKJ753{hQvGW zbbn$f9lP`j8DS7jr0ci*TN@&`M;&V5>DI~ga_Uh2UY|pwPO-53v6SQ;yT#Y_^WsSw zU;pE?!Qg4vRsZF_$?nH;56_2_TFKC}t{Z(h{Sl1T!@UOjP5^vpCiA)@Hw9?`QyXJEuYVpD`q` z)e@K!I2INcPp8fqM#|9b&al)*#ygZ(z+%|VhuW7yhBH2=&Fs#9)QsZ^9ct327au$1 z3E$hVjHWHBpdP)?iqA`O0ppS-WXRk)XuPWmjIN)c-Y!j{HTwn2ujBLlt{bsc_P(6! zyyX@V&8UCGQaluf+T>Qa!mG^^+R0-VY?h|NgPdIQZj~Qs z84rLF%38MRx%%2-98a>lofllbWvlE`s`@T35wf61XPD=K7RqM}b(HbvonheT3p}dH zaGHDcBHSL;mQFf0fi_v1_wQ#((FcEYfm1SLIyI&m3KqxF6IMfL>+QF}@yS5i@;ifAW zFJcUVgDs+Smpi?LPu552j29E>#F!%XG*<$5A8A9@6Ckr5Ix0tXy{Me|-Ga|HZ34$@ zXa7fC+0IryZ%^fOWy_=FYLF)VGS&db>z<(Lzkl<*8M^f0pGcVKatU(XCU6&0Kx?Ig z>AosTe>(@m`Vn%vqHQK^JHD7(eW>8wOZqGI_m@DJMFVd0a1Qi;y&Hx$*+_qHNuk~> zkakKuPv>17PiHR;;fb5VKtDbU+RgAIt-h9#6uhRab7pSt)M)kZB8|q6Vj;UP@|G7) z^_^JCxc|=4%IT%=`1jAMeNI&|-{@=l?{O6p8Y4Z~<2S4G>#Q`^egSv(;dec=cGTX% z0_G$rVD)W$UUxHVC|P}l?C^g9hok<2Yho0*^uG%yj<=*6Tlq7;s7mf-Vv6_G3?xH# zUV!K2+T_~LQXbkDFTkzAy(Ru!gd=|)sG?{${IuQ2o{wz+FPo(1wxL6yI{Y-*Fm^xO z@*hlJwpvLyTUaaCx?fceU2#RpZmIV1m7F;?i~q{iqm@0wRB?9;(1E?ZBoN*F1aT^T zK_8`0rVqq-!bxUI&+!3djnyWY^VyWX?S8^x3O*0fskhyVD#=BkFRtG41-&6ei$r5Oir2Jo_>Xl=s%b{wW^xgZC)9p{_AM-^LjHUK_!~ zrf*1Pq$}*PT}B;BT*L-O)U}UMpZp5b=9WTTQbCNMK)lJ!?u~d2M?P>UY(uGyy9)aE0 zEQH3Jrm@gosU4a`TjK~)daB!RH!q5SUNEw7H+&rT2z+aUsm|`buzKJe*w8%@+}@@`w{zY8JuarK z)2HjQwCIo8WHrmZxP{b<3laSuvK0BRr6{$${3o?Go-!#EP&)b zy`@K;n<|$-zs9MXf^R6;2Y-k5SH$2mU0Qct3?15yBo?Rp@l|WWX#WPGF!yjZF>TqB zujxI2PmG_z9NuwhW+vV{g>yLPmCQFqt%HR3efh^dZcuK~oI2=f5+m*JWOJhkZhh*q zuv68VkChvd{UEF?y#@cEyB+W2kK==XZG%3(Qf9hmK78I3ER}TZ&Kvuy?z@tIpEmH{ zHJOUn3MiVJ^P@hYT(@))J6ILYhmY(^t*NJ!Ke1QKpea(m{Z9|##pst(EA=RPwyhRJ+rNHXK-Uh)LCtZ`N6)t zxN9$HIK_!iJZAwZxT}4A{5F3+bA!q=aN4N_T#4``4Q_9P<@pgLV0JdypL>${KJ`{~ zK2e~EcIZpy+2Zv!dm2HlX*X3ZzO&uI6$w;XJh+=u+Z>GV% z_pfk&JiKn}(8PZ~1BiIz{SENmVBOpeTepz&%ZF1l%Lq7iGLfFtju-Qhe0kXV-+2l9 zKE}iOhnxbq*R&;^9__^oo#RPQ^H!`;vp$eMyqQvKQaBg=N~FEOS3h6z$@K|eIAS>0 zicG}hP9#u{Qg_yrl{D1#vuLA+VSEcmMI$a^iG0Pi>zT^RzUqu22^3#im&Y#3WQyPSvU9KikITd@uy#!#fxA(_8_6W7@W{2D&| zQpC{k?H0z8bYX0#6c%%D81Px8(p>#xyoc!>W{_~1?Y#Gm1)c9q9X%YF<&83Eec(KM z;T}kq>TghvYpTObLlrz?TL-o+))B^xxWS(G9nGq;TQcwCuF@~d8q(>P@c8NL%?i8N zL6rO?goLVopH75gY5kz()B$kkwKLc(KE)e8`o@Nv zb3bhq00V>fhr7 z^s1x-DwdH6Ys<*!nkisyJJjg=NgZ{VRdc5v-xF3`-E#38deRJUo1&)LayIP7nt z`fW-0o$}F&yx+f_tQ%acsu7WEgA>P~g>xyHzf<*FfyjZR9$V*Tk%eGE&K(R<)p&>B zJlcxwY@VJ|CW*zJO@xs-Ep82@RN}>KkrlB*AF!1t&X{CK<#Qo}a`3{t}R{EpJNa~aIBa+ru!uDRppsbOnlVZKf( z^f~v4_+46{I`(Rc(j69!Zl%blZ6LFJCJk-XT;xM;@7ax99Un(te^fu8x1qrc zFg%__YHg>(wy&lTmYK#rd77|E4W~o_>GX>%iJm3;A?%@mes6rf*r2sRL{w zl|**y32~3vLPp-!1K09_;GB1gM9^q>oHB+zgG10a&xnt_82fJxi1i#fVyJ44nkPyj zKd3Eyl>T!zN9h zc2*OD&%9E!zxM=o;d4wyei-RV3QPi_+q2(b)9kCN|A?x;jBHQOecfu2Xv_7+S%JpH z-nci%6cR`#u%a0)Xy2h2vEg$zK0jaVUnf74b_9$%dD4w*qRI4;Ni1E_kCz+v$(7he zNMocVu5096(%VgL{gKdSXR^S&Xe8O^x{caeC91}lU}gYjB^OkAtDN13ba;9L zdj4vZdpP5aL!IGz@=W7Adz)XY8UvzYq3XU;^*v|A%W?49@D};d=$n|MDi0!tW@8@0 zlaM&+{T98MYiTj;ex=P}0$$@)Y{xsK+ps2928ypK1Z?f2SV7+y=&>&gmcMAje;4b} zeGylPPLrs*|vaR9mkH#?6<}YbAIhWaVp#ugmr{64SdgnCk zj!q%E>n*^w!x;!td>~Gnzi<)1Wn&F)dg?}Qm*6L`s`WDXy=o@})|Hb#;qjoeG){5e z@-=UwUqT~o8!OLWQU4!zEiVz4ooYh8AFU_GecQvg!rP#Kcp~oEdzA{vz+5(c=Qw)x7!o)*WQJvS-3H|Lp)Rnb$!Y2E z7QJEZk}*&$nGB8QF68Aam5QroK`?V`G@Ltm0-hXcpbU$NGQdyIXbUxg7H*fc3GME@u@GGq<;L;UqSTyk^-!&{+)Vjnlp#on0_EjV- zKLexuzA~31Iq>0aE?oRLm1H(=0P)!a6l?wW^NS5V>5iW{dhZN>yzdFA?u~n&6s>~{!>yoaFp^Y`ER)`dQ^0GyKKlER7~akC3b~xM zj${m$P{ZEm;9Ac?5H&3ZOpd7TiIb0La|&+Gu4FqW`Y28|2!OFObnsajE16!`$9#wR zU$*~7E^EzK^X<~rJeeFNy?Ru#MH}`?JK=LZGW#2X`H9cqT-pM5FIWLBmIo?U@6qO; zAFKo2?8U6^m^OgzAQ-qL8zx8IA*0SrfICCY>4go}P%Za?6N{3;vVj5It689&qc5i) z-ab;CbsfiVXXwG*Izw*y=PuvT=o%bNHlZ%p0ks4joy+z}St|D1dAq{$5(%B$+QwAI@I4H)b`nI*9X?BtmDc zVnPt^srw<6Pvo1)?LV=+VyzU0-D$#q@2PbtTsnd4y|U+NQ<@O(U&p13J+k2U^M(-C zem(Tq)_{Gw(Ujkt=gWUvb_BWB1{hUm1rfdXC~6vIfvsNv3=C3Vr*6J zrA?;qWFaN*+4yUjWL(O6)%m94O&b;UZN8P=etr?Joxab%wrJ0tjtpmO8fr@~xEOG0 zauH8oJVF^Uxex68>q9nQyP!BT*+{81E0LtlUN3!}IS%${Tl459>Fm-QGj@DW4}OPq z;l_DCxJ^gZdyqW;MGUz=Bav(W%H(NoSVsRg7VY6v7jM>Wm4&iXXb5Aws37*)MAeRGnX3)uzZ8figvzb$0c;sQ2g+h(cmP{OY6>I;9} zZ|9zz8%AzvP38X0BlyaWHRNM7J}>xW4D4^an~yK-#|K_rK+?6YlInreSig2%6n&FC zx#l5zXgsC~YqX{XzjVc3`RT$hesG@~cj!He+{ir14n=n0F1LKirO9(4<)a~Y-P%yO zclQahaqSDmViSBO^wM%>)>{2Jb&AJ8Zu#v4KeBnIG~#U+K0QVk3>!^Soogpn#d1G(yI*ru{u-^G@$dV<>xC~_w;_3|yfw(50Am`h=9@nqhNe9?@N47Wu!^~} z0H!VFfeTwHeH{DcCNwpGJ?G5%rfQp<4qxUo_q_ew&wo4==ndiD54He5t8QH5ZBym^ zKE259#4D`Ljz%P;O9)@E)R*DfCJyiD1|WV!67lM z53N@CDH&@Wbwgoz>Mm1iY-Gn~kKsFbGZ;Ln4CXcOq?)_`%<249M`!$um9aaesnZa<6G8{?}ihKV{j>pt~WTd-en? z3RuF|QA2)fUpTk1*HgThIGCl9 zWp`RD-A-sKmHmzrY04{}7MH}I+B{;D_q0};&0Im!&IhuBhFW4hB_~=1l8=L1!hnV` zd{cg@;?EINV(HkFY;5_18Flg{(tV4VN7qp1;onKM{u%lBUw!B@@}PGya}K}EW23H% zIZL`NHRU>yo!Hyu2HZK^kH79To{w@Gz~?u3$7s+qa?N?csq0zkDqCUi)F_aZci+c` z+D%a%KXV7|fLn|F_`0#hJjBgV`6o1*KX0@N_s7DyysMVJDV2^gX*#S!H){MX|jxjD+>1&=Qo^7ZyIl$AM3B!3Y8$j2eo zLL?O;3M7&Yq*f?*K^#B`|M94oq1+tdfwBUm87M1|p%0|-gCvqkNQKWNlnEbVOkpSe zN3LHakrbm$fwCavvrsO4-$XtTd104;ut6Ik7ykACkxP))p5w&3L#?7LprVq?F!VkKFDXFov`;o7$ag4+YvIEdX73pI41TjMo5uQ zM6^MaVLvy-I7fAVQLO7l)*>2t5lb3ULzLAa6*&^M zBkDrTL$n2ijSTy{piJaQ=v1*HEk?)@V(cUVM_ePM$B>HgIwKY9Dg0z3ABQ@@SE0Wn z$cyzBIS`PeOkSXlp)=~lItked-$y$=cML_C&=d_+NkdhJC_9%w6c_29_IV_)G@tVuBH zQ&2VuA@b^j5cv~pcnNt!6OGV!R3D2X zc9CBZgE%(L5l0YJi2As`6{>S9^!2iQ55c&-Ctw3JnMC4o4h8Ob1$k+G#pH$cjnOL(NHFQ6`SDekc>` zC}K57hmX#b3v0Z24Po^qW{K=^HXPpDt@#} zDpLEEAQd$@3F%#=;+2MW1%k&s)S+C|s>r8IrtVYVsILF|ac*Gy078ZcMO0y5k;^RP zhZOx+JS&6_4hRSFY>96Z8wC_ zi({k;;fZ`8A{zN|Me1Wnj)gzvn6X#IHpj_P@cbrHF7LcN%Wh)ph2=ehu6 z^+cH%N90_Bas|fL`5##xQhS70S3&&{60{NVhzwi@{5Pe>vz6^Qsz9Og?FC1m^LiI6ak9-pP z7xHIlBi30<=A5c zju5sYS3))o?Zo{1AtDhXo+R|^hI%0rbt7VJz1z4n*w2msnFT)K#HO^b_+EzC}JntqDKT$Oj@k z5hB0!@rZH7T*9m_z8QS72Nv5traeRqfNzm2>A#4q?eH$TaRE#0cryj@) zo$%rBs6G!WP%dOiD2qo~5<)IhueBIk%ry;>QKUXrGyd@*(VZuMW0#(5?#m zi=3EaTdcDLA?7c1PNI!EKS#BVh%XxXB-B@7KT+2;h`WgTbz6*jNrAeytI$rBiz2i~ zhU~%uFUAsWL34`K>m+=MHM@jum0xU&@gJgH5Xyv|6nUX@ zK`L@xhk60APTNuMgK{xbDMgu{P!i zE9Au-exR+WVbMp-U5qQ*SxANKKB5Z!3O)rO@yMGZgikS64bmiRi#bIj zUw}3o+ak|ysJB91)Pz_U(LR9f97HT44!Vn>BWa#4x@>>uhUJ;9k%?){x12MK(4_P7lLcd8eHMPO^IFyTXk(gH%+6sG- z!vbt4qfYR_2q$cd^^+pRIZxP6LkOQjCvqVAdmu07Cvqs(!4Rng#|uU(FG3z=LYId9 zl8`S#nGd!FFX~$KF-KHk9|4;pbxc}Fs|wK$A?ONhi+OEvROd!sp!O@nKEj_E!w~(5 zT!|R#$IC!o=*^K@A}vN6u|6J%BfGUk@>+ z@GbHp>Ppn7@MVuUju5=CwRglh6sX^03~^j>JhVaCIMll#jwA1e5dQuih%Hj*xAQ+#tcMoL3y=!mixJ{gh3$AmooJ&k zkzcn$bv+23kO^DS_kgfLoye__%h1-U2>T;M`@h2nZA86^zLEbx3+)OJP6&~2QD5dr z>({Oh`}jDjzZ(l*qDDo|k`Q9vE@&_0s{Es_4k6aG08xfI9qj9Z&_{^AGDN+-6lHRm zy6)=NR@ez&RtV!Fb!?)}#F``_giajCBCo=av`B3ijpNxMKMv)h-XvlyMK@G};Pkq!UAw;`= z-pHvwj|#bXZU2X+6rmh__+UE;?Zr7&6%$e&^zOT6TcI97rB+*yUPQq%<^+6am| zktadLn3gD$An$_ELWn&48w={iSR!Yl&VC?-E*WVOLL6&iJw)3I(HVIO+8X|Y<0upM zB8TEwaYB6+>XHzm)i_8ABdIlF)CslQuJTPwOFrYgqV*wkL*Q$F;XEHFh|}V zsRAMLEo5?p$V0u2prRg}iq!AeO9T^-^ALdb;P2m1sgM9zzma+KBAU_8nc zD04$uaiO}8=qGX`aukg^5sxYIRWfx=iur%Qwy@K3RLey^JWwv?6oe>1+b2ltp@=pJF}9EkUaadhL9vg>kLX)~5c!f7s$&!O@u<69q@H6k((MQ# z7da8M0I5034j{x>qGlvAoF7sVw}2J$!l%gX5afl;ID}`BdJf{a6~~kqU(8v=B0*e0 zy$3?@acEy3zo1&8jYvbCnBTM_H64dktX+Xjou^3T^$|XWSQCgee;c%~L+D^z=;g>)382mx+nxv!i-4HJkRo->xuIN~>&y|N-xY)h>I7ea zef^O)M4fnv9Ekq$NWBnZyd;_WH`JSs!a{wn6>H=(NIhRc{m@qQ6YJrE@Ii=khNxpx zM|GZ)kQd_yVq1)-gScpBDBK#2ISQF!kwOfI0V}uw>#3b@2WE}Y*ggNrUm#~RP^h2m#*jKa{7pc!*VjZ_& z`!?b(`sj>QA7xgEB-#IA+xczWs&7(k%xyHdARxV zeZA&7=T4rRuYBLX&;P&Ibsf##&yQW^@%~!(6q|1*fx4F*fB9MEBG$8}HnE2;rCt}z z!K-T#a{wQJ^L3&3P~%qNH=*(c(BJ+(1NV9ba*&0%|1a?Rc&6TMXEnBkEVb?js6O(Z zZwGn{9)HN@RMK@*n37jb$%n(ikR2C470SXWIj{e*>N8sq9`};x~>#@0nP0r}Fm<&#;?& z3*9Gp_izLnE3xiV^dnH`|0IvUcjO6tryw3*=j}lkk=?ES&S6daKSr<4n(vWs@Krv6 zFVShYraE)@rk-P)fL{mdg?H~B_$_soA(#Z8fwH^YE{GQ)#yFGxF15)CzX;V=z0~iw z{hR-XN>`$DPJ7Zv)O-DRFZep&po*>a-sAC3h&jZ^dft~ic=Lo1zZ+-p)4*AborTx0 z&U017ca@;CNA>Pon&!|Dd2{|d23~ol(#=Ed`x5A*?V$MTzftkl);`3~`_-VQ3Ft#* zcb^gpHlRWe~vvR{tSME&U@i`U7`2~Bpww4N;^W(z-M_u^eBwFi1dw!Z^V;%82Mm!R%Kd&s~3yXFP@ z5!eGI@j0+pbNk(MF3+VtOV3JMj(!zCXLLU2vX%(sa8~{FNx&6&Qov(V--WjydEq|4 zw9$jd=*{Qs<~u^xCk>T9vUgJ3GV}sy_q}l{?|dG*42<6Zb?$HJE_(gT%05@bca$Rk0>1~+00=Lzhm;gbLq<>-lY-p&&0H@`S%3$9KH3;GXU0f*UtCvc-RrL zw%bat&2vaAc?Vk_;$6`168&rJ?zj(j;fK&ua01+ahVBLMz3icT1mv&41iUrWNu4Rp zsi6$7PJ|-**K!Zrb`|5D^Y?E9yh1m=e<8l(1X~F|4_%PYv_+*FtB%<`3w;2YZ!m z4*D8ozn>{*X4|<7w zgx%e0Gqwx1K|O0i4s9FY!MGys&~sP>89Mh+h8Qy_Vhv;L`vYjgKZRQREwqJP=bu5| z1-}D(YGBLZ|A3ajJFf2(ytKr}`{JGI0_zCpe+{xgz2fV)>Q8v|{3g8L=y_1*x3+KWS749I_r#wOBY#xH{_8!j>4Q#cjssxse}N3%8H_ta zmX>*(X9imVme4OiJNm%~Xcpou8_32iL%mnC&?_(l+~ERLpA3J0C-1}SZ~bH7oc7d$ zCQ$h+pwEjU_Hh7pM@!%wB-GJgy?&cezuE3+5`F@BuiR}4uaCQahx`&u0(JkWY&~pw zPA`GC1h4HSwl2IooyOnu36;O|o9_s{b^GY;b+)oUVE+Nz0Q#oVy@Ps3y{m`t%Rt*1 zd91$&AE3{m=Cg+!`3d|bba)+S%b?!<5~z1y`iePxQ*`!fU;0TKQ0*D?r1Cf475-m9 zQy%oOySpuX%4Ge+5--BTaksN!2t&-n`K{Prat0ncuL-rkKVLv`mW{&P`% z-|+EX+hYUYaqVh0^VIUM_^%>+hwL)}d&o236rCuk;n}JC5xARWs52($&E;;?+uJ8- zoxg)$6P4#06d=*-vpGve!Tln zuq^|>HS6FEWay2bLT{~e;9Tks;Qv6M;Ny%T#C^@9Yk(ZN-;Uoe_h`&$B~rha3AS74 zHoj7I`ZreMJ70tMyjHOr`ySpNgtj~U?BS9)XR>Yw>bHFoy?(~Yd(PH-!0xW}AKQz# zTlq6|&K`hq_F=5EC}+@6#W`;i*?aL8Sof+O-;Vp-tZeRT0lo2J7k!Seb6L-xyg&YJ zJ`XjZos>V0ZJ^%(-^ z(dPL~K}+&_7v95fx4-_L({Bs;18`UNAazGgAT6)C2k6HC&VZjTV&3uYko8|g*8wTA z^IU-d{%prQ7x1g_gB5=N!@JXf?<9N=+5+Y=xBa@?lAd2ewd?C#+Lq9{<865RUxOZ@ z`y2G(2l$oDx{2(}SI7_eHI<>(5hHl-xVwG8X9wDnN8g1e;ItjTGd*)rSU*V|v^Rczj!*XVBa zC(r*N*Z$78hHU^<=bl%gH(Jcp1V8QEMVbN_<4B!C!p@^6|hff3hYU{ zyICx%y+hTD@qUn7z_)o1rr{^sakokM>U)MR0iQt=_hFk(;&!E?4E!A8g+P z>$_Wj51Q*QWbc{0c4-a2MK3GeuNCb>x4}Z2y8zt-`uE8BU+A2}-v0nw@Y)(sztwf# z57_D%AH%PLde+bQ<~hFTUI2UV5^FD>i+vn|FCY!^zxz+NWAB^juHg0e-W(v?V*q{R z=h)OGpaAp8Pk}Y;&AGH?#Ct!Mu%$r$1~|ueK;K>T=B{U&gYOV`fZqQ5&`ESDx_Vac z@g}HcX@aji^M3igx&j55Mwbw){ZgKqzN>q6{Md5}%yEwW2;7ll8hL<^=W+r61?ujb z@Eh>{t?4}DJAnQfdV4jGXF9}YFW#eiHf?E$-zKzUUfm7x4dA|Zim(1l@b;>$1m>y# zEz`r+Q@;{t(8qV?jJv>@7m?k&^R$rj5Z|0Nl{*DGY$bZ%%mAG=rqDkIuiG)FwoPQ? zrh&UZ2FlJH@_1hJ@HxD)dCm13x+V3<|K0_17O>;@ePwq}?}q2~6s&>~vU7FewcjG! zcV?`+;O)tG+wu28z@Fm!9CY#XZ2BMt_G=Hf@Scx9&gpJWp$SL>v>m@E?%w&H0{58V zr~Cu3_61bg*(Qsq>%p&r9iU%IKJT&q-i4ov_S;@J?lfLyHnS}HUXcJ;|jPRXLW9S)b36_Gvnsar6KOEga2c6pTQ;i zEpP<_a)X$E!neTBAScFo_KNs-i2ZuD-jy!;b7iPK3D3~|{#%Cke0{e!&sm`XBzC^d9Kif@=*2yH zXYRlcfPU_(15Kf8==`2}9_sqQy5^Vx^Y95e=TjH*c&^9rgX`GS_cVGQ7a0Y!}!h0SgppR#yk8uH={cM3lWc9v@HRK-1fH}l2ytdGe zZ$4e&o}&A2a1NaD23@UxfS+i`Z@T+3pJ!wZd$_+QFUUgFjpxE=8#M77|NEV~?$^Ee zzITZmf)=_iGy~>a0_it-?BAFJXa=3ZUJ|44GP+4%Khg#;*K^PVo}F{)Yu|yq)(+^J z;1T!&+-C!H;9moOulDdyp!y}mO~HFsd&t&tCi4pK>IJ$Sn&G2v6D$CE-?DRdfP2la z%KsdMLtK;1O*JwDbPf-^8%*_Upe*k}g#dh?agf~8g9-{N#n46#n z`sC4Od_Aw<6lXkws@s6?LYtNR23dWJ*ay%B&X$nJIQt$Udq)!d^6U6-f~W8w!4-OE z4k7+K{6i7X(YdtUg8J|OC2a0CCHE=3xveWM&@X@+^xm%=-rx5HxI_LO`VOoY@s0)f zDUe~a|B@UzJs6`bjI%}t{|lG}CADk&Psk18-2EKzjZGHJ59}!+zV`J@z3TzJci|Dx z=3Uq=;?9Tg#(4&p&>sF1*qosY%+Ul7*g{|reD}Q5L$HJ0Uh^X6T|nNi_(iDpx)wOth=su&3y}KOJa5UWeX8o6rSdpL1Y<&vU4HeJF@wF{kLd_%xBv z)$u!TTnB%9@_jY2PeJ`V!-?zGQ6@OL}@oog@0$kN}@xg&Scflot> zHD3>%eLMrsbBW(U#d{}}-JNI9L9S=K!RIkL!x-`%ukh%wFug~>4 ze1d)eJtM|EXZU2&Ha2okKm+zZ_y_Q8_t4p^_sckWaRuC06CdYP=eg_`@qIh%J@$GA z>vgf~H;*mLqs?7ypi4jl-Z%UT+=39IzRY>g$a6)!SLWPA_MFU<5~tmrIiL1PXb4~X zwQdjlxOe>ZnMJn*c7gikBI^17j=$lz_&J~7W@ij}{H|>wy9;eg@Yb5bz6gB{%yj^5 zfD)e^Uv+hjRd~;)-i`DIKj)dRZ1!q>^GsL1Bd9%E*Rz{~FWYg>Tpm0SQ`h%w-Axal z7r;99vkhPOW6l9Ozp+o?^|^q)0nTFYJ5Y80*ykF23%xnIAb>PP-x0k2GoTMYJ|Apb z@Wy#3JS&lbOLX>h3EzYt?<9}+NSkM>ZiwBT^w1pv-|t&=efR|ZjugL~*cSRLblsJB zC%;#Qw}%wFeVgkTI<7~Yy7eNyEqT9H%JSdhGx80Xz~)^Y;#ZQ>JooVP=r(|I3;6?l zScy3f;Enq>-xRtWs5=D-xfY;*0q2_p+By}rUIKpx>@fp=H|m@tH)I8KLBTS&w+a0>GLA)dL23q*04#mAd%v?qEcQ4L0 zz?Pu<4tQQos6Cv59zFp&4c-9tzSq5qSJpOP#P3^aF6Ojm-P03n9pG+8_>Qny+uX?t zJQ({z*RkjL`F%NsH@`N|F@*TP&2+EhzO_9Ep4Tk!&g*MGX+fXBGg(Cr#GFHO=nb$3 z&u$9Tea^w>)aXJRD{yvi; z?|nn^dfwyT!`RJ#cOAb^3-ErY)Oi*i;2!e8Il}Ar3Hc-T8E63J%aE5p z@CCT1-FDphkhlRc#_pp(Lhrd-;{>_RKM8*Wz6X!m@mtY_cehE1d-q$Op>yU^JvmR2 z2jp^p+n@z+@8-1n z`V{*-zIh&dUx#;(zGv@96F>7h|0}5H^&ILRXQB5{??>l4elKRwWz;){E`uIuw&PBG z7rUjALLs_Wj~fIIv=w=H+a0moj~1}XPbujmUH|9ygU4_BIdloCxQ2T3g~Ar z1^nAVy=UN)r$yYfz9GH| za~fOEyI*|e_t3k)+IEUO{@tB(2Le2@XEOu!PV|efzmtuzFJZl!&*&{7<_~;5-zjWA zf`Hx~IjcQ2@cA2h2AxE24)x!`ySqA{G^ZEufj>j!10d~yM16IJ(qR$rsQWXQx+Um1 z5B}~v1OWA{nqaqx?@iwe^!}_%1Hk^9JT`mCa(XW!Rp-7l{ASUef{Au~6W_xxz@HK0 z+0TL#ppWOV0rueSt0dms?nJ%)-oo3LIh5x?K$ikz#7lIwoRGtQ+}&$z=Dq;x#4@_4 zz_aNufq24FY;W7ML5zkucTd3phYXY6J-x}YDGdW+5pSl_35quA-P2V1R^Ec7E z?=+8hs*i52h-cq`Pk}zp>ui4*kKMZ4P;~?7bLc8~0q&5U*Rw9k`v84}USDhHA-;KM zwKse10B7?|a_j-Vo>BS-vUA^HH_sdhIX++poA-AS+64NKcYRg6*Ch#@nB| z9r$_lyTE?5HQ>$NCC=XURj2m>Un0M(bO%uT5AAs8+}$tm2|C}lJx(K=&zk13-+6d6B{>;XhRJ3HB_+H}V4Bb6SR`&>^)hq1rFu zXVAO%2Gm(6+wpIx6h5HWC&9N1^>BF6fDFY9+|{$G)&iYdD}B|M_@BUUf>Yq{p9I}T5%=&5RKF&)M~>g1o~yGvo3;db2tU(~zau80 z+O7E!TM1oAE-i1E+4t@aMgGLea_lJB4AJ7H-GWa3XcX5Z!`Wxsx z$1Zwf#8iO|r09CUo+g0%zXzH2oV~Z>e(v!pm7xW?acy^l&hzv8XpTFeuf05l_nb$- zdDVMnZ_pbJDha#Fax~r*?u$NK6VT8F^4<$uIQ)i`&tCf zItBbDs5fT+NB^CtK#$fGxJ3O_(+zdxXJ zHuLVo&ja}qop&}1@q2KFJcDdq>r8+lI{9CVDkhIH_E+~Rt?$#j%Fpw1o;&C<@%CY! zJL05M;2wS*>RcB^oM#f6hOd7|jccGY*K#|4LyY(QJ5YJ4^UI5j`1K;*lWAnXLH`ZR z)x)oe%|4`_g)!c}6JRe1@%HfozDpf?LU{}s9u zkbyI>08)H&U{4{B@mKJ(?HIcVH8u@Venrena0o8(O9Xb~J@W_XLp$mwk-vk#N2k9z z8`tqJxEFhShu+=0_qs>>pTp*vwSf9{{LB~dwXVG{VcP=I9!x3gB@TW&b9_` z{6}a?KK~Bez~+6Ng_faXh242GN0PQ8RwpZW`_$}9G8f4@)*LUz`9{beSUaZ$bZ$IPxz}LGu`iB4SivJnf zB3Fl;tMH!vJ-oejp%c(ssQEvGA-sOR3w3qfS#%d*9!vr6dtFyw-%gUpGkDsLalXe9 z`OHxUb_X?PtB7yf^9jKCI@cLK`g`xx)wH&2tMiOw$!|aEXP};M1_tmy0`H!0FKb+C!f-vtA79dLwvc^zjp-t$n`hhKoY!v>gvPl{OkIlQq$kjn$_QV%(U z*5^4O&R*s~02#dme4K9=IW6M3)H?O%kZxhq-@Q%%`%jV6JpPWpL~fFQ2|A|@h!E~v|iZK892f4oEbNbTdU(m!0s@7xUf9x&E^ti6e!HP@g^ z*sN>5d1QC|9n`zF0qvnbMK`V;*d4MvaBtJt)mtNh--J4Q3AGn(o?j2%S(Vk#0`L1H zbUD;J(Ly%`J%o0kFMJr?DBc4NWSf zA->6bDGU5;#vz%~hy*eFqu- zo|kbyqIVAG%HXGg=k%r>e@~xOI_nz0g)BAJ*)#IYq1%FY_6OjbYZUQs(s%Igp>GX* zvk>30`!dfyc4Mc3eO;pWd$flf@b|lV5A^M#^Bd*9>~|LZ`0ikP3vd7W{Tui`M`xbP zc6@*4bq6`N9KCr4*YQ2}pq{tyY|KNqE8d*LBL2Ih&OeXr?gpS^Eb=LE{|)4Yc6^Js z*b`u{9q1S622d7h5%=l+$OHM1t??Y&*p7Y&oqY$Wb@xF@uG|>r82e#&-$P(Ob-cMA z-qge<=amO)0n8E60{y!mH}*pK(8-j8&AuR`3( z88-Vhmv?4P3P$Lzfc``Dv*h% zT)wRh_;Yw`UgBeXpw9-h&b1HkJxS4vCFG}I+?)C$&cB1sm>J{*n&7(#?~bkY8hY3! z4rJJ_h&@GTjP=bofS(55J@=aAfY>g+Nr>lC_w^2Y%}c)&@eK8KPWSdN^yY5V@#tTo z_im=-JH>tu@9a5pz+UHkgw36~EAy#WehaOAYHI#_eC*f$=fSsiy{iqN&KU1sz}K^^ zbJVm&F6;b`pSmV;31*2m?>)RTm`gO__46CyENdVGL;QSy|Mo*ZYsdRt*Yf*+0v3Vi z;eEHx2tK2S3FHN^2e#46o6mQYgn0kg;oZ*~Hh1Gpo_7zvuCs-Ya|B|{JBjShukf3K z55S$s*D)D(cX8N`y1st!&g*{GpysikBj`q(J(0t=VV)i231~`9-@qF3G587G0`tEt z;=fO<>7DalTmXO6-H^}O3t+z=q4oK1!n>zFKGyM!dhlsr@7T;Y{#GG(;p=noeb|33 z=fufp_`JX-{|mft-Z@>-{v*cfor5tM0=$eC=nxfikPgNZ%*rVpuUGy;NA3GTJSH>J6i%WYW@N} zBlZTYwuu4rz+T@2^EHT_g!(?bKjshc8PxeZ(8uVtt8*^{dHT)ar?G88tsCIqLC1T$ zj`^*Z!0$k<=X}~XpbOw5_`A}Lp!@RtJb`Y4H{c35LkKa4d4%`=2wRisR~NZ@huN&hI%*0(ReS_1`*I*C4 z0G)Q+R~F*AdWV+KZ9tu;X$@=x@>;n7uuKG5c@Huwp(k|Vre9nt_hUPp% zcZ~cya9;aQ(5rKAXT->&VtU0Bhi9V1HtqA7Ah7Y7x&_T_CSL9^<1e zg>GXvrmS)kJ$^fWhy0uI59kT7e`j;w-rFX$1%~MKoht~WhG(STCc3(xl3H1a-?Ri8 zuuT_H-i03#HXhfM?i&svE!`Lant8JP-S}#$FMBmwSGT zV2ExT+5yhm1vBk9pE2h1OrbrtCplr@E#50?O*Z%}a&1oNX4r!BG=I|$l zPp{*BDdCr}d-i@S&229sfFhpJ6?%1bkLFv)=9$WSmz>|dTjvOyGd=;vIP*O@+(|=< zkNQXGU$}- z(7y*g{H^m5pAL5S=G`>6cVSG?&!S61d`ITAk7vNVH=u#9=WZ|N-3P`rfoFf%j^8=` z5-@>$1I#achkYRL%=YCSoPzgk{kcQeLN;d#KLBsrao#Sv8{`RS0Cire`|vD0-*s%B zt23{{yN>`);3t9p1E_f#ZR+9IfvO+D@1Qr{oGJV#U><9GM<&p_2mP1Ohd@5$MX-mg zzDc}v9(e$t0dr2`vkTncEYuoe4`jf-wVkLT;VdW$U0uv>2q zp9L_7{xxuR^=XJb-Qye3?}7<1(~kRc?UI2DFK8*ha?goDb-{ zAHv$7!6qp4cxT$4w(b=E6Cj1OAXQ1yE`XzXGs@R_KnwnvRi@EQI& zKK648ya&GVBw&Zy%k4M)xQhh+BM>UR_rw}QuhG(;ZZlcnc)}LXa=P2Hz(gcJ{i>ggf>8) z!EXNFkz2s`{f|oLIc9mhQ|_P#{zA75_3r!*zY9Nv*KX}S&;wW0@VwlI_5B7r|112? zK?8nt9lv=AvOWd0q~03z6Y!opR|a+Go~N~}t^YJW)_o7^vvsE5iumo;CUSE6R?U4+ z+$Oxebnx{jh4w4^uLTO^AD~^}zQqCdIUw)vFn8u#`w^Y}HKBc|aUqYkQ*H3`(5L7V z@C@GB2Tt_ya>O70ab=;w{^Jd7-+5%@lz1RTe@XeR_7`Ii#yQqBzxL0TP zEQatyc;^Tq{vMtw;$}&E z=X8wRz}Gm>rHlSGI`6`F@OSVzcKcOdl22dhOXNvlt~1~r)|TMggFgm!ZE1>4dI_53 z>;ccVo^=Vmx?SijSS+G{POTkyzYVMCog;zv(B0r~&-MJ@qm%yx^<1W*E#U7OYkSTm z`MgKV1@8^ib$mbG<5_gp$?$VN`5Yg6JqOC06@LMBK6hZgMiJkf zecRI&(05uMzZSFyw!r}Z|5%Mea-lcu$39#-l)cvS?403i*0{c?$ ztooUI9)1?s`vYESBDI$NmBdfM|Im){ zSjYGKCv@)TJLnp=Bz*nGyA#jWe9r6Ho2LuyK%L+C1UVtj9^IjLaut0Fy7;(Lb9k?$ z*6O1lfPn8?WM^=`6L{}chHS5|)PqgX0{V$7eA2+3h2H|r>zG^HAqd!OU#Wg6wHnZ7 z5&!LDpCvi0m!O*l+u*1jV;|6uf1e|d$Uj2%?f-yJ2N>^eo7e++5^N>3hi(;n3ut%F z1^6BCfX_#$J>8@40Ke&T$m;fi`*{gQ*qmbn-kNXQ@w-w|!`@PKLv+S_l zE+Ts`+=JiLS$KUju#7AX^wx*>Y_FisnxLPA-vRD-ybJU#&_M1&>)KLx@eb-4*S-(< zouKn~%Ty8PY(N)4z~()%?mfJ_>cP*n>5m+qgMI^SUG40si)`)c2<6L~atJE)7vXi+%=7W19rOfO+7z_6n4h-+TB8V6GN9I`I484A{>VI(vI4 zmpc6&RVZ8+YmN;|1gLfX!FW~F>X7Kj11M1nN{UYXekG>b{c*YrY z5}j`<$>Z5-v#$<1??eOYHz3E~`|p`fA)EKG&3giemEAeEk&l7-X3;;D$Je~8Kq{Q; z4b=YC?*YFZ=Ey;T-M+1Lh2Hn7-urUVCMWP5yc3>vxQ0?^4P6;xjZ_;t;XEg8;_>?x zUfoS)Pw~I2;@zqJ46toMb8Px^~MUW)fY< zzy9}P^Q7qf$?-eF&pMvfJoGto5A1_3I^SiQvmbozcMJU;dUNK19RhdNK(~Xffse7i zvAK2IfdZGZjWfwkLg;J7-8)eG zc!%w0eAmB;b(dowi05q$-+mqE-dn)EtwZhe4E~;64XASk;Jlve7my))9zA&Vr$zK% zfDV8&2|)2AjiG|wT|!Y5`Grl zQ|KmWT(fKBWiY^gkDQ?2MfU9HfVqMch?Jt|jcw=kDqqoFSLk%rjMd zJ%a)K4!RaLcVV2oaSQOyZ%+I83E5h{-Fg;#%|fg{)EE0MzS{4QrH9BJ`0wzY0q#cq z1pFp&N3-Z&g1WXlUdGm~a(G|e=QKKJw}Mzo=1BEHKu{ioy>wzFbCAUZ$W-t)keP$^uH%p7UI1AZFyAbQ*E5$)csMc~^Q0b`6`F5-XJG3P4qejTIh0eA8pRNV=5T(73+mfA7K zT<#_j@NX(A@9(G!uvO`@oZlz#-kEMYzB9jb3qU_>dCvyek`VLVqMN|>BRc17A$!+v zp!T#;#NT)8Pji#Lhy^+kpRA&<5~*_L24R zT=f47y}r)00QKh?)V$ue?IO--u2ZlJhUA!o_CSh{u^Iepa98EkZk=C%{*R$sz`X7; z1C7epKEx)rKDv3RwGW`j_^m>pf_31`buU>U4%s-*xz2N1#P5cC?onI)7kKBifA>}U zN<$vM!Cida(G%>i3VP*D*n@NK0PEJ+gCA6~vzw<|#Ix%`eKY6ac{}Dv$lF40QNy0( zpJ7`9&fvLvHYu=%dvy<6Ma-@42pj_U@Ts70a0#sa3|a^@40OvF9F|~Ka22U0r?JkMsM!<0h$JM&<3^+w28gmjd$PQEAG{r``Eny z_F6*46})GC09{Z|zE|z|-glAh>j&g*_*rlZ@0;-ZlEdGDbx;zguQ){>VV9pS;ygX5 z=eUi{UbVTm8DI~dLzftP?0^mM5?Dvy0zQTJ?=bx(wfOPuS#} zmHZv@26lHYmf>g6>1%)1o)GE9w1wXqv;ln&eNo^SzTQ>$V-D}hytbVChHUNx)v1L{_xe%HNsOOpvLs9%Qmzyf>% z)J-6J2bSTtK_KP_bmO{^Jg4UneJ^Uuvfo}rsC&>1^h1p6!a=SA#y6@DAcf=P7x`nSQaU>9_WQFjQudrf4|vw`f~FQIeD?${hT zy8l)OTKGEOG-wp@Ox=6kn>*UzR;Zyfr`?@oGJgwq4iPd(1Z0#hFA3BTP8Y6YsMIYTIx)LAf&XDEZ zVeKO|{~2-S-U794h|Zmz;qUw@wpSpKd){=^B8L{=I>xzCr-Z>x_R&f4CULghV?u%W1Qc;Oh9w=7tkTJfo=q>odWrF{F4yB zi#>Q}nt*z?+wk}J*qis(zlWyK>DNJL?Ev;ccZGbwxEtsKx)M5sUR}q&?aes%@+-W( zS+4=S+cW6gM-zCDzC&;QfPNa@8Qit=<>=k1eQh8+>l$#j9@M^a;4Jo;GowF?$bne% ze1LDkD=)z}fVKaI+J}8x(|*0PCGr`(v9s;?_op*I0MAUl@P0^lpgm%|pIxZ$;RL8_ zLbE`9=p{(u^%(&B8LSY8-X1(l^W{bS?zyuJ|L4dF)Oh>X=K}fy+4s>Uhxf;sw7t~_ z)aPM6b15glz0QF H#ZMch$d`5He9%qIr;n8!Lp_yn4S_#M1JUIrQRMiKXDFYnNq z_m(*2w8C@ro4|9o{|@#U_{Ac=l?Hskw~0Q3x)v&E%vS(tA!y^70Zv#35cF}qF#`_~?5`G6Xi|Y3Zsg#$1 z=k8rCsk2|i-}`~Q51@n1+}8GP*y9lSHMmFTopvAgV(k`Gy*-xX>$hXiPq78$=g=Fl z3%r+$?f5;tA;x{)5EtNI!f)cU56$tju75+>hj-rI+{HG!8?au)H&RkVz3?n1v_sv4 z`RDM{W{xIujy!~SXV$Qv2K<7Qn9_XkLwIZF$lfRarq(xwcitAVJA45>LT@f>d*A!l zF`sqyX(8+HZnvQNHppSk4m8KU4X@t)*}rw%*Hl42)ojjkQp?yc@ZABs;0k;KuZo!G zCwTXogH3pIn@4o6zw$l!d16k1`i;uZS`B#no`bFe&wB`-fh5GYHr3{Az_>@~2_u;Mk3b-$GC-@8N_Mqzp`6|6TF)qWlhu*urg-xB`*K>3YsC7H&t=R<2z!^&X zoP8B}1YhFkw_4d9OhE13d;$3ml*rOc;J4X(a|SJdvo_H?R|@8ecyFvd4Tjh8-fp5# z;O%7-ECTP+4YB55!+wT5L>}L@%H~e3kpp+-H&1MXQRUk~?=D*Ctoa0f2yH=~`E5n( zGqhIC|0gX2Hu84ynF75wwdvssoqZmm6Yl2&^cV0wxCN&m1KOPXsvURPgSRhd@qTtz z;`{kuwb8jFdpA!?t&(0ow8;UQ=saKlZq7l5ZHUgi=J%WCF5R7Y1Gd0#;2G!x_5S_k zKFx2;9DI+sJ92p5jS4TZ+232Je$Hw99Mt)&>wTC&PQaszI`?O6i^zc3k*2=inT9861L?`dwiC zdvXfT_ARz!bO&IBu7)+ApmUGY@C|%RV4V54(QQCafxhPMV9SB}2jn%V@9PSF9li&h zgbvY}J7ErYbp@QM4Dn3d!9}G{(A9UvoQv%^SHOP?egVHjXaj0r-Vxu{5Ne#cCqN6G zdUe~_oY!+Zhh`vvACZ3ov%uW$Prj~i-=YQHpq~VdO6PokR=y|j*03MXCZ*S`h%r~OJwh_-+~@|Nl)gmXQ54bqTn8YAK`Q2L_qE$ zd#AOf#E$O=ynWt+6JY!dbQU@VeGd-6Jn{(`_nXJx3D!)oW#lsMIlSj0?~c0gJ!=1e zz61XhYH#{%Ll4^Vf7AW|^?qC5yE_8b-bQb|8u#e+@o%iZvAMG=V7?plwclrW`(6Xy z5%u=e124e02%Q9j>v;CgG=c0lZx5OUb`13%dDkY3I8(qr zhu*yIqfzloP;0oKrGj{3{MP7q1cvC%Rg$;Frmr8KZ`EKW)A&d zA#3+M>v^;Z@6Z<1SsU6^FD=o1{c`fRf5erA;cPM=x(p8zk9Id+Cg3C1iSmG zWoep6xk;SzDR%piZ^4iIhR%R~5znW@S088g+p&dy1bqzM1fE*}#@a&}xQFoW)%$&e zO}YV1v8C92PamKO`g?eH-9t9NeU$hu7TB>_dmVlNerv~i2ioXa{0098eJ%ed_1(Lt zwJN4dz6^O5`4ZlmQ~3OeT!J(B*C1faq0VZb_Rs<5_s$()GkzU9KF^9epYk-gDq>yV z-DhB~Ci?H-H=x$AmcA`?#_a)jCjT3lh9B32_r4^^-$R|@jQCriueM3B47^XClX;!T zKKsylbZcN47&Ewz|MzR|`2^W>P%lb!Nr>;r{rYV)U#NWk3BOu=_1}c=V)LyWW4lB? z1j;FSd`rI~H`?*v0?xh;-v#>D=hi`A;_G{QivFgEe;a#G0(s_<@8I>%k$d>Ap>t36 zKf?9`egQ0Mz_Y?AMf8*;O_4gZVFHJB6YxwMfG+>8X!~9!CyfZoaGji2) zS>p=XJFx_|@l)3#*A23L*suPF;7t+tsNWA2oV5i%Am*SQ`%0?%kI`HEC9pqhPryGw zlXm=uEFmY<8DY2fXJqFNAlDDwo<1SYGuy7-8~RD~-rF2x@T=I|zh^(bFVM^D_}y5> zrvDAP9>~eJfUbwm8aZ?i+1^r+AlG~9pzoqvggQg556H&OL7j6AxC?9RldeSjK05u* zE4qbkNDQSO>g#7e z>JQ+J{ZD+to&oQieVhaRo6rvVy+i7zppT$+{73j(>Ntmj^7lz6j<9cmEL3XerS6O4#6{U-j3(w4DK#N*Max zi#=F#Uwb?Dasst(5BLr7?)fd$x9*{V{Go_F^eSpU)=~C88+fR`*as+(Vfz)2??GD)A06)(~op=pWbhTU*G4~T_AfI);4<}H2 zI)@IgJaX^w}ck)auiVHji>hQlxnG7J~F3>Ue4!TbGcpYwIU z#}oc&e1c;h_ip|+=MfBo47tc>+^ng z;cq|_&Ux4MxaJYO@qRm&k^jP8205|L%=iRZuNh!p0rzr-C2=F<6YPZwA&`$X0rc=& z$7^EhJM@V-XI&CE4SxY#dqu4CaIXM7KXv5LAuoZo>sy2~hxgRItoc293NElS zY-?GlY}bf)z&dJM={e68@f=&|a@Lul`yHAC=C`(=I8`=QI+y7C8ZDf-*=JxFD)jWsT{+_x6Nr*L% z&>LU@c(#+kTAk%Iuuk)5>}T!XA>+$}{y|0_>)HYV)U_Mu@5cmxQt_{l)?u#?=mX?Y zJJx9(?%!ZtAf^QFIUsHkSl2kWvjM)zb@(PRtH>PNZ@@lr-T~i}_j?ol58w=($2O>I zn=GgYtipMQ?kNt@ow@hfm|LW09*`;0{d4qr;->Ix*WaIA><%&K6`zK9cztsUoNert5T4fHwmI-jx$F2Eaj`e(VZGrLr*xnDtQt|Ycm#dlb5uhwPF`syB( z6Zn?l)tmU7)fU*7W9`pHyubEpO*t`L_&NOl0O!38rokbwCx2FvAHiA?=W8GC;aNFH z=Q;oZzkH2=zy7;n0>AIm8eW0Gx;gB9^c`fq*F1LDbyZ&Hpzn8Jz3SotJSk#r?)M$K zciTF?2OrTlkn;KmNcX&iJ3}@>8says-m8JHPn>sc0KDVgk3%^3`~%ec*YXN(3%_@- ziL_ri{qFI;@?GH1;jZ8|k@KJjW{Y_D>U>J?P5s_dYV@pg)-4tB9(g_q`2xOm^bDEi zF}HD-1@+_4@U=h(-8b9E=a04A661Yw9xsu`hrmAQ&g&RH$zvU^3+Nd!NBGx3h8?ht zEs^)cEU?BmXD<8m4D7>P_SFT6V9g%Vc~2B<6Npo?ZWHOgO?2~iuy?>Q(Egi7<=>FJeO7Atej&P>8yRzp4Bes!Vf{e zh`DCrn%Ebh371^Q+6U^?_X~2Hm^Cl}KY|W$AGx}o&*(X5v3{zE-_IesIYfPrvk>1) z4_~WtB^G(BT1AkvOh?_%CKvxdChdh!ac7W}j zJO>Hb0T0(Pm$Ui3V1{rV;_Pt^+uD4GUHl)Ao`X663(|le6rm z?MLJuyuG&2-Qyi_4ric*Kg71bd9VjUhAqxIQdWD9mwDx7x9eDH3_FbCEpL& z&c-_;_Xu?19s$o<&Upm#ItSNyPTC7d{Wow)h=04S!8wF~%L2t1Bg*u$@>|A4<_qtj6%-O^@1fJ~!_BL?MU&L7Z?RC7j z^*-aha_SCW|9jguaS3(~l05F?{H;~~ydCFV*OOx#XRSH@dt%4?VAp*%(f12#L$-kX z9ss#H@Dm8s7LfLkqu-)?PtMvgK4p*J(KGgN@9*$C?-B4^JJ|Mp$C@6}Iq$(|#GA)8 ztJp2{ZDa$?6fw8DXO&iK#&aRQyHeGxkKYqLq`nuxde!S#^Uq=1v-jq>h;MTZ*(J7x zG{*W;^hr=soA1Dy^l1;^)oXY+T;m4L(c?uMy?6sa$**duR#;u zy?vjLf%^vH2JpTa_ejybLkaxbD#kgO({HoyE(5R7|F=){JjDL&VH(?9#<`#OGXw75 zLN?$J;I+*$2WJh|a|i;uHLru$Mf^SDy>d3=T=+wXYh6D9Z=ZEN_O=1~-~^aY-&6D> z(8W$!KZ*XPi1T#sEBGe1`k!DGJ%J0zWnewcmFPD{tO0wvsQ40TjjhVAYqG|kxMpwe z#XK>tUx%APx|enVmO)C~1k(3E0@g8w?KuvyjXf^l>1z>b&-GsQo$@8B^q$M62u)`goxcRs7gGLP}&TJSkjV?W_v z0-om?ww!3;w?5~WV!s0&VDCeGu1U~mf#-LC-5|Dwo}$~MYn=NbdXqKQ`4oKyxc(6s zfF>CK(2wV6t*vOc>JduaO>t;zL^MZ8O%O9s~_zK`r; z@1i@mb>w(o;;nB7+n(*~1*r3!5nF-=oH@)bzAGZukoIA`^GneKaCSM;y^J3qJ=Z+M zeVW+Mu=Ux;O%-RJA@)KM|K?1Im-8-Kf0y{DAVYru&fzEQHQ)?9uRV0nbRW3ZcmD}E ze{J_NMmrEMC){@crr=Z11@a@fS#bG z`}%v^Z>8tpnaMvZVx9KXfWH9#*lWGV3-kaU@TJ7PW{u~bV|$j~!5`tww~9Om<1<0N z2Wg1)c)w<_OLCNO$LKBA=jh%=`)%S|fKTC9v0tLkA6+VLB1{*t^+IP==uHu3=5 znb&8Q!l^$1*SLP8;^gICBPYQXdo35t4(tMb&S60Aw6gb<{5Z2#JD$M}`Wb7Dcdfas z#aZ<$-gEVL;0AaGjP1`?aq* z{n#CBYm^fK{xf(Bw()r<@r{!L`=S>z=+0OMT9d-86!seyH_c@K95 zXN`UIN1)zUxq$86ZxQ1;%RQhw>#cTtZ`#h>vkb&f3v6d0+;6Ig@96-2K)iXpd-hr4 z`*zPq#B2j++QC1%j^|P5+kl@0t~tXmR*U%7?8$od%@E@rzV|dlZkMGcU>^xct(P}to@03H z_8j~A;eIXR%#%FYa(8gsAR+FCxW;wdV~GB5_%l$#>AL{VD`&mEFR`W%Cehtb-&>^f zn1bsP?;gT?@c}*wUpe_5xO;ron&LYyI40nJ<8|=XR$~8%bf2rrSLgQ3tgUq&@408~ zJ$Z|?7U$Cf>cVx_zYJEfCqWHuY1<@DF5?~ z{T#iq62B3iSrZ(BE%=Pus49;j+d3s8mz_H+jA0PUDn`wxh3Quz`2;u{T|>&ebn`f)nIhJ9gnfrS?*BT@W`s0{zk@yjXOe*?`YP}o&#;|g z58HTsi{J%tkCMHH@O5wYDKDnM0=QoxFL4RdT-!yQv%e$!o-JYfW6TTm4!rB!%Y5!T z&W~T2Vo%_!_f$6F@|8H#lDHnWdzssN=RVHaoSVS@&hQ7gNn)(mJa3RCd<#sXXGq_r z`diQ@>UoU!AYTW*Uf-zn7wy>hz7o!y0sIX%+VKqs za-E{r`EJnN^P>2g&v^TCW+&)V;8~m5 ztJqWMIdF!~p$jI_U3ZJFykbq=n{uy+oWDo4XOWM9cWxW$_v$UWXIJ-?pwHt=sVm@@ zp9NdsSKw^xGc?Y;i^v0bzgdUKXV`t#E+8A&Ly+WguEuY{-+`y_`V(~TNxh#j3-IQ! zry=lr)j*Drm&9+lMjlL}H<1ZQ1CsSMj`68GGkfuG0e^qILf;0i^*lW1WprcQ<2lmv za>jl)<@XDE$m8EQ9k@qu=KCF3+Xe7kUcj#b*WBX^tWg&UaQ^BeV9!4ze*?aG_jFdC zi#=~4t-o176FYuS#5M6<>Vum>9^&_0%Qq+Zra2#-D13uYc>fr}hVU@4k2D7&+aJ_iF;_cgmdB zwpkEQj5C#U-rD9a$zy*%5uM#Jg7) zegU}$df<#Tf590)1p>GQ=1N%OTe2T#@d?~$vuCG>YmShVVKKGsbk$9Ic0<2td=!9)@L#yHanxEFA{NcZ0Y_T7OqW*@m% z>Dn3l?qZ*{zvfBdx~%a#bAZo$1Gra6drsk}+wr?>>sLV?|MVMu9^F~*x7nZg@4xYFz?E5&u2xeC*q^^8V-K3dp)Ie=iv`S-_!N_X2P> zDX}ehdocGM{tP}Koz)e5ou>)+4o+F0o9AVo*Hzq`Z{l4y!nUqDhcoajj^RIo4!QK5 z!3+E9DdErcgH!AlaK2l>I*nNbDYz$g#5&)3pdZii4BI$!`yQGgg?9$}a$@V6{HCpf z34C2p;(H(_BiAkR9@M#9GbF|wedHYYp&e)7SubE4GmmtQ=Wz^j;x~|2Nb7qHde>FW zNaHj1n`+0NTf~_sMSg}m2RU#q+eO@afG@)itXqfcqC2N1m;l}}_w&rA!BG)wH%G>r z2GUym#=OLzyFc~=Yo?L*=qx(O21w8YG1>`m4|}#od-d)#(ak%aE!?ou2Svrv)7Xb} zO&;RRx9~p@+d|4a55N7BNNZStbI+9cA#xf19$A7ld`F;#o`L3dtkXOPU|bLUW)b&x z4(1jE{KB`ouRp{+ePiC~Cdj}BFrRm90lg$nyT;Gx?rW|FaK?{{_(n&_EASiWz_;4* zJ$A5XKo`zgZ^HWq&ahpdfS33KatVF{yaGA)1~Q?Bd-Pp&XJRcKr1v?+=lQRrE1h|k z!}Hw>1$Pfz;F|%?%)f8l_YJmr%vX}*Z*=dl+$@+b;{EVkZ;9zvKK)(nJ>+}PL-#yA zxAP+QU?1MCLpb-=w-3&WnBRQP@kg)?HsE{6U68Rh#a>3w>0<}!p5}C&b9E*uQr_4n z$PV%yuzzDSdOCr(2IK0xs2tnm8pk3Bm2M8}o+JLIT4(G${72*>@V>}hUB@%q#n(f3 z2G*p00B1~CiTjWDVck9a5&U!HfH*OQT@qK~cSfGaJks?ebn|tPH^_vT7J7znyMRMa zS5H?W4i3*0j!pYO#t?j3NJzr&j&3o*u6YqOUI zI74p`=X#M7ccO%wE67o;n}BnF_n3!oVcYu^(mekHX_ej){NSmy>-qb3t~1wNqujyjh8cS66uTsdWzq;SHOEW zjXZ3}Z%1b(uKQ4N_WJ>!G1lBg-$p+Jbzd1V>)12EJv(3kr|x{L*SQYi=CE%;Ngdw5 z7X{~xJ;FEc0o|Wh$hzNerF%>O>uw;;lZ5#1INzRi-y=8L@tgG=*&%lP9Vc#t&sy!v z95?84FNq8IFR;gbAqMd7JpeDjSZB>~74reEhweVnT3D;na z^=0r~>@qMTY}Z~8?|n!?2cNapIel9ja0&91*#1iV+qL#T$Msowl%JLU|;)tu74`uG~gBK`Kdm*@Hl&YasyxF?{l!*kleK0`hMo`E&^@@aiY%%KE=kdoP~Dy+fA7pI7n*ZVs5&n8&~zv%p$YYMLm%o}YFT z+nT&5uDwBDLgs<9hwFg<0psr>ImUMbxdfLM@$KE9Tdy@vqTBa7@H=_}tRn|gMf|S# zZ+qv`0Iu0a26Lmkzjty5ZnC> z1}QMsejDfw_}z9YCj;Ou-0`$gR23j2N~ zelHTFd-d?;#Ms{+x@WnA%;21%^|rL(OVEY)Jk0X~{T28Vyavu#xMrrHF7O%rtlo}! zSCQ`b2uv4oE-Cu&UJetQz!P3-gbGZ4c`A*GTM3>?fduZk=b~CAvBDJpTUZ5qnCEIpr6?9vEV$ z$P4`D63+M#4Dgw+3*65)vIaj(jCb(>EP;Js&S52uwO2V|PQM`uut)E$KdZ!eZbRe% zUy{cj?0?)Vd7NhwVw`ujgT0CWrX9~N)lZzgc{Wq<>*$_A1L=%9zjd%* zFeZQ=IAYx~a5kq%b8R3!gEge`Y*u^*esB&cOG5jh&AWvzO(P}<=A`3ZD6ileDA?5_A=6Z_F#$MX&ckm~M(#V-IrlP<8~$Ap&(l3USNAucYihX08N375t8LE-vX5=N zwRtzj_b89Qd(QApgO~02onJtj^GhW|h;L=NV7BP*k@n^tF3IJO@lEuNO4`FAan8Mm zdM0T*P-`>>fx%tjRuH(;-)n*vW$UJnmut zIl8|yJ;Qb6Eph_62CncYZT2R{9DZMt5Pu(D7BPqC;ViAeJ|^2=@3Wly4dGhkT|k;g zcs{jW|E*cu&h2Fp@3rzL;;nN6JwqCwQHSv<0ayVwn4(?D(Q z_$D^tGGZFo_OO6m&p^2WtnCyTKnr{+$jQ$l`aQmNY-e(I9p7jQw~EiYpCawud7Yy> zt2A(~aMm@R3*5GL#reLS-wf!2m)Ms;-G0rRkwgC<;1JF`aEAUE1oD~N@8AQm9WY6Z zKTqJ+(aR9;;{`rzbJnhz!{>Yxu!`>8evSOQ;B1j6$Q|T!@Da=c-_eM;-wIj)&P{t8 zsqQ+@;i^Cfo^yRB$`bn+-Z#)E&O81SaP2+F@rOWM9@otm@td`YuR*-`%QI4M!aV`A z#4Lga7~y|`oB^JZ`{m@jLVn+l>$~Xp!2998w<}+3*A^|f?+fC=YcLI0$Ln|R0Wp5t zoR9awnmmIOeV_o-1^rw{e@-1KT)jtv?aw4JewUoTYt7@`S%bHZx;|&=TF=@U1kk{? zpP!NEKrUx(86t1pu5m`*wM*nKa!A}dG6VWOf8VNi!!@3}dvF=-%HBzD4W=+p{yr0NDV40`JBua>ROLtZ5Hf-=#mk z(dDMF55OJqO{DYlzW>sW-~)WJ^B{Xv+pDAV-9!@ z>h}F#VBP_-S&09CaC_)2f$t6Q9QJ^FuOXfFF>;}Z_wx_*2Jws7d%&2n56-)oAai6! ze)FE=ciz@KgFaityXElwjcfdG0HpAq+=9lvYO z;5;j5V$Uh|J+WdE-C29?=DI=un;84?o6rR#xCLxu24EMxi8R+KcnmtkXj@~7Zp;Jv zDl$R#fcISA*j5kmP2E=dr*^F80BM}xIdj^_9qa8^+r2{(=X`)X19EO0T*9pER#~=*!Tzsx{`j z#kQ6de*@imwu^Xw#xtWdYcRIXeMgM*UnEX0sieJS@Hz1%$l)i!7WfQOIP)aLG=V+4 z<`x)ZE@#_CzX9V|;^)u<@C{^v8j1M@-FWMmL%Of_1ZV>HvW_eC7r-BTHKvR7Z2kTZ zDo$Rk7hk`5X^8LVJGe(+#2#5rB5LYmZ?xlm8=`*!_1Rpa&jZhI+yi;8uU=`H0EBC=Bcnj7F<_j9&iu_$*jBBiK{B2`T zds@QofHcJat=u;HGH}0=8ZX-Mex4(9Vn%QYNU>i6<2uOKz`pF~4Cx;Bkg=w&-Team zvm$bR)+~Yq-r77Hb69g8;@NM(&%%2K_TZg9LAox7J1AmJemBN<0O|d7PR6eR*ILs( zGQpO=!yaP$##8h?WUq*^y(-T7Cg8Tg9q>NP0{h=l=B(xSqVjz^=KTp7SnnQwpI1c@ z@1ooe=&(L#jc@ZKw(|+$>s=nepTQfSz!ycVPrF72Z;pU`3xB~H^E@l!jPBW^p1U=8 z&d$eKdNvvM39yz4q_w^P8Mb>&f@LsTiT5Ba;u>e>8T^7h13asGALZ|GwQZhTpnMA4 zLtVe;m4>+fbpeNM?k2kTMtz&u0dS4TfOkz^+cUKGY0$*>e7bPfSF+x}o44T4(CdBO z`-r%@#(nf%@D4QEG0ye+SK-z{Nu8dlJ@_^~>l^GAaSaggn`Z{zecZFmpJj~2lQzwtk!JCE_&5dQ}3V4K@mf0vH$AF<{%Zh(9R|3KVQLGS3s zJFhNw6W*CS3-68RwvI1@y9MUh#Qq!DxB6o+3(O;2(zqR=d`vB+JJw@&S-?#l5*TSEHA+~q+fgB_B9$Y|LuQk~J zDxA`@^c&Lz{#g5X&k$=bS7J?`$ypJ9-zYn*O^_2H1$CX5?Kmsv;rxRjSIT<199Ull zH`k86?!tXUKSApA$C=eRkI*k%gWvOQu6*^mZUASUenXY^U_ZXE31oc--QV>KaNZT; zI>;^fGQ__}x8V-Jh&XFE&hNsbg88Bk(Vc6Slm9x-=oHRe*87gQE`Dov=Jw=FGEe|x ze?qKhx`%y?Tm@f%^)G_VnCcF=hi}{-+*{tWm!K+9%lN4ERrx4Y+&!uJ;>b4rlQ%e0}s3-S=%?>*}FfbB%wa z`)1dG_izVkk57Pk+{+rftjpNbeJiHZM|U>U;0w3_T{!2xgYJA6(Vv1G+~T_e?q?rw;5NV% z_A+uAc?RauJzr}!pFIWS0AI&fxp1G_WbMH_0II*55UDW`H<%FT>R#_U&$UR zYu=$-*IYq8?7fEm6d1Pv=Q?Xl(A~$om;>wWB4>g9uObhN7=MB^=Ly&Y6WGR`vEIDq z%vj@H_!(ax)V;2wmm!{IhLnF-`CF{_jOVeB+VM{5-@?v`wayH^&SkyMxvu{a{ue;M zTnAXA`Lq|o9r3PlhMvW+^0~J$2f+0?`(<#8__fWGz@Cc12i(IuH4XH)umg}6o{f2&!2;5Gct2a@O_AH!b*y`< zckr#jm#mpX$}hn?fBnng2&miFka+VL>pg#q{M3$bcMo|0>fV0A*C56jU$HI^@ozA5 zoxy)bc0dMR0MGDBuApB0##xU&kM9HP19c_Z=f9!!8Uu+)w|>o8^wE}*-I_K1Cz_Lr*T`wQRS*ptA$&EdYT z@jN~AIc$ITy0#={oDb7EsJD;-zC`|l9gsIj}4Jc2im zbJNa&^Yq>Yq;WgMl)$^2Ae-{kGa%*>dM1x8K4UL{lXkoZHShk;JO^^kO4femCF|ya z`l*sNa#P5`H+$4;j*KC;49prmkaG%oQ1NwKhHam7NNcO-q3i&q^ZcubZ)zL;Z*=F7 zlH1w%o$!vVoS4a&p;r)=9JDp6!G5~-beL0;6C+SX0You=CsCUJ8~!3nH*Sz>w;w*;9R?eG}k|10a&N@TX4o&>-Szd=My-6tJrtgFV!oqo^4Lf z9==&(9zYj;54i;LKz{6$81s7G@_jJ=q5kzY8Zr&l_Z2fU9!`8u>@hLs)Sg9m&kJm4 zIgNd@66+kmXXq)I!{;3s=gwIJ=l8D(Z$CTuvJn4Us(m%^>$Be}&?B~iyoG-S?^(P+ z-oUpie_dl;XGT8z)4m|ab9Zj1=!+nuj(McLSoaWLpBU$24aT{@>y)Bb#C0#wjVbXp zfah%w=G{Uc*TVkImDs=aF2g?s2bI5pv`1%D5^H|fdETxYRlM9PKJR7+E+zIeF;~|y zUM|7jCFT~+^R@Oz;2d~Ho_9*zZ|LsxgB*FSU*8k#Ww24i85=Vo-ui~bG_mb#0j!~W zH=FS0c7`SP7SeU1g{^eG(tX~eo8Nrnn%Q4pV1MNi7|#t}X}=9P`v|NVlG7RskYRiF z)=*+wm;ISbxXw3G=lemwv0yxZY}YOj@t-(FYP>9XD{^-G7RB-y7)O{~ppcv+$3>I+y`dpa%k22Hu--Y&*{Hj&%Wa z;r{~Xz`DCw_(om#z&dL_z_!nEUse7-e$PwD zTX)W$SJhtn?c49f2yOxU2)%{wJ#)<$r0e&juzv;a>pj|3XPvg^ zorb92tAiu_&Z3E);cK+xIq$$*(=_(@obb6v3*H}V`Wsu$T-K7o1@IB>8MZNffnA1p zKaBqo?0|zJ#(SQ_Ym!odYriG#5-zZhbF#NOkFud3&a=_>oi*AqSH_w$kGyxJQ^ov%-#a)(Y#_$G_L5-t zh?zwCrUvjc$oJUJ&iej$Z@3KZ5B&DBh228_ub)~Th`0A;;2!FIuwKO2KK@zkB*Y$e z;fHYkHxctrfDSO{d!*}g{MO=~TPWh+_#JO8e^?P|0DDq{|t6LGyNI+_NsN-o|!XFfctlV z^IyfMKhNX&JJ%6>eU_{k-^VntH{TWa6dhvTu+I6pUanijzmwc&89fklh1>?#e*%_) zYmBeYlIj!@6_#S>|sbA>3z;<5F zVFSGY-q*c?b-V?2>+l?Va^(60`5ycW49Vdgbl)UYZwTqUpW&Zv$2B33cPGbZZ-?-o z!E4ZPJwHoD)U&{z@J--7_RP&+q7RAfA>Gqj<#+H~W5PQ7a6Zq#b6`)tqYk|1Y&}D8 zgKaLc1~2CsIhPh(&lqg;f39R8e*^zJeD@WXAf4kPoNr6rHMhtBXD#*lp5RYc;=Ps+ z=s&=zdmlY_@5u9tn}@fy2K*NK9lmp<@&ABjunU}(dz#xH*BK+O;06UGez8#L_we;y zZ^F68Gs}?|$N_%)oCM}^=9h)pi9;R$`8uy>ROZYVxdR@9+CPK-TRZ-@GUGiP&(FS2 z(Z~6T(f1Vj2regA3;zf70B60{St5T&x<=UF9jI-4_0DeMOOXxiSHKwaTc`KnXD|ok z_S*6i7PS6ACH&e$4f#Otr^jQAPi z_P{N=c75-RNwMefS3D=HM-G%?>i)%RtEE-u>&yEfnNKPeRP+ z_w8N9`FnX5-F1HBmasj;DP#lqPJ6Cp5Bs^o9w42^h&4@o``G&W*v1L(k9p+96t*&@ z2KVy!ptCX7yEKV@4$S=sTRvmGvq)LfN4nmcyhkSm{Uf)K%Rt{GauvA+ybE{u>w5ea z9Rg?b9x3lzSOo{zUC;#XGhf8|%{>GyxH&Msr{oO4`<7CddEDEdqaxnV7QPE&Zjj?X z;LDtQK-?tAvE}_1*rR9hEBrVQ_H*nBpq;WVfm_4&PP{}uRR>q?7_);k-dYQ=2gdcb z<6C)!t=~Q8kR$S)SN0MzV~-Hxf0JuszXZZWAGeIfK6cSIq}9#q1&(T+__lC z0F;4Q<`JvJt$~y^##o1UyCmiWdDcc}%__RH-oy4R65^-fd>2i4?|TD0!!}1Bz1fbn zTc2mxTZz1}=J73N>@z-dZ0o!OljxqQJzJmtjJO+gIs3ML`4rvW>-%JXi`XrsIeO^Y z{dSahEAf6j#x~aoJ0rGFjCPF;^x6N1)8~HfyNfiJob{OBzRh(HjM;+g!udP3iJU;b z1usFD7<;uob9gr<(Y>Py_A{jO_B){e|A#|jJnIJ1yS9O}7IW^mpN@->vw7{t0;g9b}SM zAbtURu86gMM4v);FJr8$uAvOf9yBWMmm>N%u$$EJ0QTUfuoLv>$RY9s&bkhe3t$zz zX|o^Fx;!J#?gJQrk~O}C`J$==-Pt_BZlOEBWo+y6T|R^N9Q2#deHz59BfaOw3H!Mr z-kN5y2k5)R{DA%&a$HY{wVn}k2KNsb5ogWbg$%!E?K}C>j@$rSE+>Bq2Kbz_bJ1Q! zegr!}pL=(a^?6xqR>a>?J*0eHvvLBYA)cGI-}OHEQ}jA+gzg=0Rh;v+-|vg~?w?{C z?~nZ_A+G%lcPbD5qhxPqZ?E@A-(Q)-fjhVk+zy;MhUms;NbhJK;u~^ZU6*-ly_T*U z_r%{3`Y*`mebMiH=8y@w2E_t!6y~pfiqu*?LO+_ zLXh_pym`|Q=Q$1Ue9n;mc+X~u*#qvgi(JK5=NaLb?;+C==cn%oct<+$+RjDkJ+Je4 zrYSKSa8tnAy??vy7-QTbwmsYd=TYxFk6yAi5$w^!zCqV-KN-?G?K=sNz~kf6{{gr{avWZsjXrPS3`5d0@^({65UU zuc0p?#XGPFT&GVd^lgHtpi@u_IeOG_g6&WJH$olX#OGYKzkt8NbL@LT4NFMRHzRHx zU4Iw-3^dWlvt<^p&q91NMLgpsd7Rs?aCcQcXH?g+OjZA`t^ns%@!z8Tdk?zgFn$Aj4*L*kU;e&-f$kZ(*B;W` z?)Mhm`)cnu=+47=*83h{`{Q@ny`9e%(00F{(EZlgzkPYezSCupt?*q$z69>s0Fz+3 z9qa88YpqMbZ(JU}ej_??2kkiHFX+xQ#qZnygl%8-`8>v61r2k6#ddt-o`-pj+Xp>j zddQ4AbMzFybAJM-UiZ9>&v^{tU7NG7@T}*H$QgTtZyn5oDfnaX5xC|8`0uqzq&*w& z-_2cgYbwe2D|)8xUP#Y;r_GxHNBCUhxizsb!A}+EnWWh6GX>luDdL-Q%@X!Hc1GTN zaDi=q-h)-J1>D2?_`dksqqQ`#ts}#KgMSV=4`<$abocDSoAXRQL{7io7-#F-biW_* zkAP>ABCqi0#M@7a?S89B{jQ%u53F+!^IX6`NB6Gn<9iNXgDp7Yb8;p4ayVlKz&oI? zT>1L@{uA*Tu`Q&&e-m#l9b%oM`^bxwxFKsh@Ya!lEBy6ckn?@ot21~3oX;83^?h@p z&wvN?1>{F?TSWgWIf-@z$a*M=x_|?D=-&^m&dCNO}AA8@G!75#DtJeXMI?XZYP~1N{T&qxXn+9!+fH zJoat}q_TsEd*6brM;G5U%i*CH<{NHx$wF{r9!znw+z&dN_;WOtFYbNlG z_qJ~V?>#-l=XnmndKHr+PvQSS+MlpK->h+Ii1}LR@)>-Je5;PHysq|OpL=3*VE@*? z4I0>&$k2}W%bCBFuh$X2hR@m`fV1x)%Rmf#KzG*mv5W0@rjMKh%fNf-y~x;8?lo|o z@p8_?p1z$&54*#f5z_nfS-(H5N%L5%Ydn{i$N-lhQ!t{Q6QsXaJb&x6=N72<82>jm zV(euOcs4&E?K!XPPCNeJ9QtrwIKMrfeLV+f?d(br(5=^Z?;hjxfwQJZze<&19eZ=;*TejmYk2PVJ`+z`0;9CjY!J7|)}_4bgk zW(MCpIE33l=5VgrKtG48=i~P!#b=!ZV2&Ly0|w-3Ag4ebpTkXHkB~j$hee#r0-QbB zPeAuREaN)`E%YqJyo=c8y$7Dj0T}nvj(6Y;+dZB29MT^2n@2f519IK6z6lqQ?(2TX z*q(tsj9BlzG0u7ZgKrj?!#m+joA?uQ*~b9=3^@dT-|fNJaXg&;HIU~0ZxMNG57=4G z9mMaxOCX2y+ihMs@3hi-v+MXyybC>0@1gt-zcrgf`;PSqD9LZ%hw$S)h|}*k>CZ~{ zzW$8gnfZ3T8^&y5d!7mCz@JpSxxK^I*n)Gu-UHY4;JZNmriyo8Yni|vqEA))FUS_W z`5Q>%?C}&iM0(d#plxqA=-xSdyh3*mF+^_^Rjp8y=!?YEe&tiRE#MyZ=$m*C{s0ZQ zdpPU%Jbc5S;az8q&UOjtn`y^8eSvh}FG&0HjGf6W(mlpM@UK8l+$T8e4(Mxeu2EXY z^LG3Kl+_ z3f|h~yfX>0uJs<8Q~v}oeipe5obQOWopxNSZvkCfehc}kh~LH%Y3?`3CUPEMiu4S| z?;73p*5?}wNc-Eb{N7jnBe*lTY5YHd4sgCr{6o;eZ?DsEACP`S(>$*8tgX3=^j^FM zYsC2dI0Pry_xR1VfPITSgEW2s?>;|)4?w*OT)PXbwMCpYDtGWXw>$U@zr6YkoNKK` zDQ~X_;CpZcdO)8w`o^4-an>sDz0cr81J3n}z`AaUcb`>kdvTqxPG{!tdw$TrqEC1svOF#cKkO`gO~|)*Y(kT|K{zXdk)Ufc;{+u)^dTA8=}{sfE2o8NwvC3&XcjNJ#`6KlzlzQ145@6g}jGsc-2qdvfwvEKVL_MtnY zMUdnG+xlH|)Q)?e;IsY@*v_p3u7Ky>2j1hK@ppiA__v7ny$Q@y?|)I*Pb%sD)4*EQ zYc$E10?)*H-viIay|2(4D{-Csq}bjWb1WgPwLXUi@jdu$xM#q9^YHZ>p>G3jrXBY# z@qNJF1B-BfAm>0px6TD{i|u@@)xMm$ygx1YfIrV8w$Rr>3uO3=cg~*k<97Tm`W-UY zYhVv0@;$J2*E+)xn6nbUbr^pMtlPZ@U=N&v=_1zexwxO}ytkby-)CaZ!B6l5c+YTv z?l;=Jbzf_6N#%D=_t){>mDKW_xE^>5yb~F)_CJtoU?0@El;-NOzUET6E8=&N-oq(m z?VCn_Tf}dO+>mu$U@whI?%~(oM6MEJoHeXs{{#By)_a9MA+YbTXVAN#g?)}pfZxGc z;2c~t0M7p&xQFmf)w;64SJykfr>x(rIQL!#-c9veP_Ok4cz2w)^~m3>#BZ=~?*hFE zZ_X3eb%@&m?q$vps#y@5p_{L+L%9H!KnHvP`rT)~h`qVrJ9P7_`=*pL#4f|Diwktm zd=j}2n(cTGkI^?Ojzky3ato0GRy*Qh?_Eq##&>?=k zhtl?lE9NJIQC*rPr8zI<;>==LB=Y}Y(1;=c*)dlTLGACTjG<^&Dm<*h;8 z^RrgJch2A^aDhGmyXaGG>R$Q3|2Kus#T+$8#sK@Rxm9$}Ek~C0&?JZNQ2(!T=+5GM zZ090x4a&OCoR}l*7Sf)rcL#d{)ccI!jITLkwa<|5=bY_fNSrlkyH-0#-$Wk~XU`o$ z%wOOJJpl8%hxiqJ3tt{$4t;(2D{u@in!vp3L+~4T+WuN^-Jjp&!AiVC_UX65y7gPr z74`zsT-KK%8|?E6z6E~+XJ5vY@I7E{=5UR3YGE7a>|bD;+wV;u{0x3jXWbC!lQ-u& z@E-RI-ZC(!=#WR*LU!?6e*^!7apVSN_@}I8|c=wL7aQp>jJRPK+a`! z&p_WaG9b;hjh^DSh6JvUZSPNzTfn*0``YL4pte7u4~ab`#u?6Fd-i+S^I!^C$3%e+ zR?&0(bLbbWOVO9$j5Vk87XPkfJ^$q*{{6a$+y{q1-z(sp{I0pbb25+LLif(GouxfL zLF(%uosn~?_j7G|9ls^5c09X#Z1>BE`w7n2MUaNbc~1g%w}^Mr-1auY&dJw9-r(;6 zeM9UXYkZptan^H(J;bkm1y0f3=PmFo?7NSik?&XZ33T6jf}Nmy-!FjqQ@9y8=c<(J z7g4IuqX&FRi0{n%vP4{&vo~@CdT=eI@modwd&$|g(Cc%(G6((|P1X!S7yhzZ>;18| zK5~TIhuv+r*Wub&m$ReJ>#|kQs1K_nv@z28QVWz-iAQQ*yem@OOZm`Z2g6HXzT6I6G%y zKli|U>3w|%?Dt>o_`5M!3qK8P>urFP81r5t?ag_YpaY!eHa`0?uiyMdVyxj$(12e@ z);0FgjkoUEBF<@iUc~)~Jwj@KAm%aHgtOP2_*?J_{VlM!A@GhJVvlpMuHI*|9rYK; zE-@LgH&yHc+~2?&mWp@}didpYq`B;~R1dMQ9c=kIq;+q$zs_2|59e8VzQ^ctzGM6F zZ(4J^pJ#c1Zk>17>-b8fz18pDeUIP+a-18VKjYfa>zO9##%931cR^ie!g}j4?>y3T z@g3IlSta&0(Eb9>!8AApEn?+U@-88rWeD*t1#(!^&qd7HNABP+K|;J|amD&BSVlKi zWC0K}AjZ2n1Rd~85%b%-b22utb{G8t`52kA)}DPIX*+Ed8W0A?IF#(RK)&_u}}TVJg^6xx!g;ffEF-!06XaO$UUU_A6Pd8e#fc5>nUfpwn2 zMmut@5sUcsx$Y8pe!Iw1pntm^e+Sr`b4-A`bqzqq+DAoEJ5^y<$6`9KZQ*;NO58jNsi% z-kkQofL(@|=ihMhVgM$J_+C6y??wsCGY!1=&SI{J`};l2u$@s8{Sd!-7SVI^*Lm-- zSHXz5L{N*^ZO8xa;!IokvplZ3K(~&o750bI=ih$gdP2RFd%pgjI0JQ0PtZH8vsY&`2~y*+eb3L4 z*7YO&1aO_Rbf53h%_o-v&(GPIzlUFa{Pse8ulH~lAj57Jah@s{teENU5cw^8wvSt6j0|g2cOrSu4 zf(aBTP%wc41qzM^1qu{Qpg@6w2^1($Fo6OE3K}8AV1yYKu?$9-#t0*fF~edOVT}K+ z#u#CY5sSHqWiidNT*R_0;&Qo&$B)PV^O@8A?DvNEeO2|XI(4e*obDO<-uK`d*j;1- zme6z7d9TJf*|P~RoQXcA_vE;UcSYX_41vEd3t+7sq?|MNESKByOos5DO-e57s(biN zKlW3k>#g6}G~vn+XChz6|6l3*q2|1|N8j0p=o7$w>zEOFZ-I4o@O=QmJ&Cb4ea1Y7 zcg-u{XA9}pd4qHoXP`%ouaPrFd@jZ@?C}Eb25wvnavoTx&rbvN(d&8MU{B!p zUO5AET?+I~sE8RLg2k%K2X^#5dJ;3=6I7FTl(f1Sj_&rDt>z_9U{TXr^{7}U2 z_CJx%%l*77wXQ81AOz|zkm&B~9n<#-)NxAvm+a9g;->oY0Ip}m05-h5a5gzmYW%aJ!H<`mw(26DtT;XLOVcx4HGKtD#> zgU>=>ZJ#_Z;aroVUju75rh!a}(P!))aRazsJI+gf4$gPYQ@97Dz1WvH2lias*6=GZ z?;3IjJZ;Bk&b8(@r}57E6uSpDuEXP|3~bauWNb^XC0nTkC+bjEHKv&wsUeem&gRUjLbnmUoYZ+2BJT}J;r_u z_Rw2!?$Nr6HC)1{`TyR(to$LwSmSS5=bY9511`Ylf_2E0xFhs6ZF(xPou9G36Lu@! zxvYW)+#1rem;?#6Okt<2^BFdWcT}n0Gqfk|zmb0c=dg!d1T*cpmv)!j18n#Ggk6%e ziFBT09=*rfL!|p2AkRP(Jp~!Mbq~P{@H5aBJCz&}=ZxfLi+Fx@%=glBvCau%a-{iA z(OcSi?B_lD0N*{*{``IoA)dXnFy9S%8pIsoYhw4288Oe1J}0g-r+5jpZ;)9Y-$&+n zhy8tRCvYRyZXoUb61^m+F&_(Nf&2hc^dFJ++Aj7QkbhS3^T;V+9_!4><9#WS&e7T1 zLkDEw0odCB={s@(q{QB1uc9A-4BI}u&+hHo0B3&RlW&R9|A3z4G3E}vwL%Q>=C04n z=jJ`}``{4$8FBzTlLKsh<|?sw(QlC2Nr>})3}@fh?Qe5%2G&uo#P>iGpZ)jYC#$#~ z_5j$g{snyVNN1D5=Stvw&VjY3$c%mM*|V{B^)mmqxBZ!Cx*(qKNcZ!ext}@x43Wq1 z$HYt{(-3?872SJy4FcE$&hP^GeNxZTJ$J!AFkai4WaJqUH&H}x5q=do+i~sWIYYlk z`d++5UgA$#w+-I|8G5H3|2_R9(%jmUMf^Q>!8&`APg!qm32>wAw5$M7@ z56{Uts+&{Y^RqtlIg7>>KClF5jQpSA77XCE{mw4HCU^$i@34qI_gsMMAaBTDvPQpW z;9B8YXJAizK+gF%FZWa)0(-STXSi69C&anzV{c+VLH-3iFZ(yoC33FfTCDY~|5>mW z`2NWAc<$Egni=FBakK5X-dfFL?aF|!uF-m?h;Je7$NQkZ2G=0AgYL)QG0)%*;LOn_ zR+;784P=A3dE|N#?}2A;-|m$L(jqso&Fwn#lILXU!2=7MU<;>N0r_~ zFNx2H@g5pK1%}w_mqnCAbaT}Bgl!&UoRc5t@d910kA4k|e-AeUCP5P7-viz+pH1iD ztj^Kh{}j0k3h*A7D`iblaL0+e1~1*?8CeFcu(Xh_K3J~KF({K zm^Cm1_P|yVf0r8nK%RuyV|3%aSLWWq&XFaMH*T|v>0QNluQRd6@tlyx)v@+j&rfN7 z@6s&N+M1vY@tk@q%$2qC_|%Qh@F{))bXTW!DU4M!_ zYR6}-g|x;5_}=vX4nUd5_mr~`IsReAKVwY~+x>3H^>45NzQ6lvhbqnbX-ra7J>#!?Dbas|U%+pUIi&OOcVh<}0^iZz8|Tr% zm!oe3?@paZnO62SQonm2pw9#E{xw_^*!xHueggRHm}>!a+u!~!>SHg04{-MC9_}AP zJm(g=b$tR2^d7ihAuf;iuwShSa8HWp_Z-Hx5+m=drtvKkdsF%JdscJsejj;v8cO{3 zVqay5e`^_=U=N59J^a?%gnNoS0)Ce_fxHO7eEO3j?rok`^c!-VA-k-1-8etKTlg9H z2GZ|(&(n9AXY7opkag@OG5c^CQqH})=&hBwpYJn!`42H^PTlxtfU`(|`(t?e{G)-mx>_JnlfR(%ml))$b>G;{&!bQ+IoG-6OJbVvP58gTfOX!D z3D9f^u=6_x zP4o@84zYf2u)UAY(zRbe3VPTF@MTWF$RFSy;NBt=)-1#2A)c4Hg>_m(z&C^KJ$wTG zAIOPaY{$Etv)-AnVqYNjtrl^;b9j#a8T9da$H!-bJwkVH?^~a^Cft7k_f6sE(BFXH zLDr@gVto&%Ncj?3|K8}T!#SHBV*eZb4wgYlJ*P#?Ywlxp;;iLvC7%BhantawPlz92 z+q3=yq0Fr?(h5M zBQTdU^6Z@bIkxL3K?b*uYy#IC;~aC~+4isEv*dfS2QJ~WBL2N;UOz*)CAc4f{pff7 z*pD=a`&-vqn|O44a@K>DIE$C?IeusL0eOQAz_?57Ghlt@wAK#mvJmfp_3q-=_ZxBp z+{3x1M0SDS`)|QgJNEDsvLv^4CRNPkRs5Z%{R;m8dyO?MZ1au#Vb3Od2Y-%k@A3oW z55)FBxI%(Uczx=A`oze2F77*BP#dr>*I1+aJ^BT(o-)Mms2p9NI3>;)^=EA)xLApK zt;3qq5TEsN9R56y@eOH8D_;J1H{{Ta<*^X!LJI}tH`#iWN_AjJdA2|ceYy1d# zvl4q7V#_UoJMbrH;#)@A`vdv_-S2tp`=cNyd=LH^`hT?9L!Y<>|5fSQ8=wVdz#~w{ zd1kZNC&=_F?rEM$> zV5x}xO~Bcsy?cJ1{|WjQ@R^)P`rIe1FF}g!o=30ZTibN?RPCENf#%c%j| zHSX&ipMw;=o`rLAs(Uvva*?IRYKf2j;g9XVV1s<}B)5 zwVWo_XIi@tZi#bl=3T}1UE=&b7x@yIhWPKou5rJvfY)viGybrD$iMv!H;r$KxH81& zaK~8i1TJN*vvRh55a7&Xz8-wD(l@a)>;wGfS^$s1Tsy8G5ogZ{GG*@spKJC&iQR(F zS>J%~0DINvnXQ6ZxTnZ1N-J^URo z0tf9F>od8AUia6;H=Zf`e=jeI=>ThZUB!6L_UKycvCa#mcjn*do~!5K8I;tp3TNyc zFuvYXIa$QqU(lV$BkU4s4W5rP90BbMY}cgN-c8@}&ds%d0CO!NSCKc!pMmR~Ne^xc z)V(dDJA;H6rRU>%>t075fIN@?`Hnu%$i0`rufVuTWSwujV6Mc}w%@s3xH)1vm>6B_C@zLBB)JA=iL&T`2GY^BCj3 zNkg2od(D6re5;5()c<}^=JEZ~!M8}v7XBk(tb16K&&ocqCVBT9Am`zR_?<;vZx`PM zw)-{l-JpB^^)qB`UDizipN9dy4zbU%KY}x0t`WXX_&(SNN#K9-*mH`Ul_PGepl*R} zZ|lGsTjVW~-nA#d83^syAjN+H&e5H>XOgkUxM%V^_eY>Z4$sUpO_AG0+}B)tU=Di# z-^2e8Y&N_zZ=5+rA zpZmI({aCB>7{}`;e~mK4->=%+_`ILa$hwZvorUMQfc*n{4>&J#xn>?|-S+4FXNzyS z-Lnb$taoPi<2vDt%xezqEnuB9$md`c6a}s1@fm6mTd(zwIFoBQ&&(VziPirUOoKkQ zy`Q1`KJ+}(m3U8$^^SCb@3#az;J1gL!1zO~dt~tD7xv*DaQ{PO2HaDhv&uvC$+>O< z8NdVa*6;~k-v~PaKBIT=)-1nW#M-T;f$l6f;q@P*J7@dtf*x?+5BLH;_tIz27b`I) zfithZ-GX-(XrX8L=aA2_&%h2?0Is)A&uYFMXYU?Q@cjTLKz$Ai`1ipbu@^|=&w+k( z4Uikyli(Hh4zdGx3+k9@cxxEs z%vhhWW*6JKcj2rfz+b{mf)ABU;nV~E8*+IM(wsAH$M3lzwljK*G*^v1IG-);8GK)W zYebV=&hJO`OVCBn$-Rs|UY}PIy8!m}x!{a}>+Ep}z5~2ZCtx2q2iJcpqTEBi2HRk| z9iRCHbibS2&yV@)nw8eq(ueN4B*Yxf-WhBo?WaWFK#q_ZF>~n7aS}M6dF%w~J#$?G za^O8)CB|C4EAG=n_EzWz`2a81#rFNR2$zw^XXzH?==FOg#SYkg>-rs@}ECJs~#)~|} z`kHO_F(1Bhoq5c=0KWmOr%6uFs;SUXEoi9xt&d@;y%H7*PC$8%lX_O zXW_PqnFb{_=?_b3~E+tl9i$w+`V0OB+uFzV=kagh65oKe-iJx>h&Xff;TN#)ktwKaR8HYnn(vSp$9$uoW7F|uxEiex<%|gLtlc+@w;vsy~!HumTO|a2jdUiLh)@zZ_t-O z3jYjQ*Q`wQXb)E6_dxBtCU(9Z-zTT6`3%m$6x=G90p|2RS$_h2AE@U*`4_NG=XQc@ zA%{qN)ZYcZ2i_rff$O{A?kb*n7UJLWQ^mJ;j{SbXUL$^-gZKcv|Hk#uop;@bGtBWd z@f`vAjI{~5NBElH1m3(2^atb|{wJV??jAme&yb$c4N|To_8vIDFW3R!2Gab?MeMklL%jJ)DofQ>-timM!!> zd`TYn$ly9~x5TCBu2p*e)|#PzC}N*~;X4I~;2p5fI*;{i6xggW-x_wwT4!YqeSP%w z8s6tAgR_3`K@+6l1$^%+<~>IL2;{AKxgF&=lP;?LmB<6Jwy`R#+rB0e|M$R4p>(1LGa+e3~VsKb}YlAQ9wy0?KfYF}K% zJ#O1^o}Rhy$(zdWT)JQ$)H#~0nXT+Oq&=87z`sOhIWud=d*zxD{t#mC2{H%vWiR*m zZo#}dI0w7H&)7%K!y?Adz~2#@!Z{1qx$i%LeQd(rAjfqgv_$|*Mf~07zNg>-{0*ml z2Yv$TZ@~LX|5{0BvIctW=l;IO?Av|3yNCETk=b&HTSfbjmGn)nn_>H@Ub9U=+-lu@xg$w9Dzupz+ z(t=+g-W(;m_wW=u!#+n&1Mls4o}MH8h?tx_lkg|N*_MHGK-#~$^_a_=rr;8gW4nJB zZcm$B<~CPg%__Rj_!`m}^I3BX?7(@)p2Mx9&mqr1f-hxl1J1ny`WN(7^aR-J5i)}} zb`ssad*u3nZk`@|6FC9j1v#P2;*vF*WgzK5Sjme}5dpTIJ< zef=y?Zu>}pxnBUKw)5&&`~uu$5$&8ces@e^dkzaqJ};5ZMsA>N$7gzkzKU;%{ebR_ zUjt`zq#%+(}rAA17ZeKXeje$H5HEkpd~?GWR7?@>aGTo0dX zyeEhFTsMjIIXOkn0%NZ6*ZF*Q>pNzx8TNp<9rPS|hg|||9M6lMjZfLr*;$YIondwdjan7duMy-p4S)j4Pd@^a6|l_(>l8Qde?T4TOb5##5UG>&w?kw znRQs>90KTI_mMr&ZO1$f`0qjg4|@%tvAOZ;+(YX*!FLD} zau~1gTASPn`ab$ount}UXY2VG=X&dXQAvC7JH3naE^MLy6Z{PR0_L1-$9XPbKM*&b z5&Q$TwRtz1#Mra@AA>nC!gqi&*R2gtvzo;5{*q&xQL4 zXI<;&S9e~iV7>c!mt5m{yM762+!V+{JnIIr?ww;Bli+i`^^a{i@E+TDiSx)p{66;_ zo5Y`iSB&#^sxCB{!i+yjs|pL?2XnYGv8A7CBqf{pgK zK0c$n<`}+*oWY(#_Y7M|a};2-Lj9a)4=&(yCTr*^{2a0a_cxsJb>A=G&9w*D0y#0} zbf(U>gKmBHYLEBGJdb-jmq|F+I1h7dfRA7R+~)!`S)=|NXrX&Q>O18g1J>Wb-6O5l zKK6-KcU=wVe_GiSNNY+!7u|h5x8YTs!P_Fv>jJKiFQw+kNavWtnbSL{+y?VSyo2r` z=X}nQJuuG8nhWAR|0dGBm+1EAp4Y&+DCgk57BRpx)FM5ELer>0y%5!q5E!}0q*?+GGo0pW_he_7S5igfwIOw(TCsJo~6D= z!1~Pp3T_T;!Z{CnYk-L&B2ZH*xOd|{;I)luP=k4l@tpq+mt#9?&$xx{oUD5X>1@{E z+}B(Ibl^gWF$uPOr-<`0_6(c=*Brqoyv({Mglo;3a!7uR6H6u=+ z-vI;o-b(y!;GW&e--5q}-+=S1hn3HsvOL}`_w{Zx;d0g^d)NcUn7c+wY#)0U z*qgj-7m=RJ5=aYX0=}T@tFaAN>*IN_-glXEHBSpW12@=d`1T&WfLkJN9k_mc?&^8` zH?tJ~%XWNM*^hTVki#=c$um{N+0}d7SLs@Kxo6mSAZ3m7Y9fsp_ePE!ID7NTS^EQ= zXQf@kvpK|G2F~VJWCrIR_UT#r`3YU_KibIeu>m&+yl=C_&4B~#96JPZkk_;KoKtMi zLD_(7z-?jYMSRy=;}6&;*iB%a_K*Y@C-(9PvGXTOKLT40lNVn18vX4SnK>G*QQU_=^Mhk=LviW zk(1A{e<4o&4|MZQ0q<-9>_MNr(s~;}y|#C-^AOKTzp$TU&}Gj-5xG5WbYZQ&JKf8> za05ETCgiz99$=fN{yuq$UC-w?^c3F&(pf%1e*toGE+Cug)a{v@;{rS{;yjO#=9-6F zLrx)ueGajA(BA-ib9VP&0e%bo0fyKmHC~`QPw&vE9rvkw|MNRLTZv~k2WP*58prct z{Uq3gAA)+VvI`cA_??!rzJ+ex*0_sr9V}zl^FPCOHpW=*LZg_^b)y^^c-g@+{1O?4uEUjYlQ4!PZ48H*5lsR zsXPHsE8QOKPx%gc0<6v2GjiFpGq5JtrpQxb+@l4YscWt44QPO$+wpvtkk;@F&d)1! z>vBzszJ+cd_S#s9{bxw`@tl?70BltHA<}#8bLJg#?=MB1`)4J&oP+D`h`B=_Adjlp zNhLAPWg6Z1G{BK(6>b~ufi=DZUZOX_bNC(P8q(hX02ynn(X%w?9-KL>e-}H!=Nw#T zEuBhQ(*ww`&w=Nfk!ue95V)uNjdPLX6x%a!*4AzQA1f{;&bwiL&se?z+^b8xe^Wdu zqJCd-zQgCi`0VgES>GpiyoMO}tzo|+1m;1U-v^%Q0$2tADZah?5`54C09v$@m2JTz3-WeZbXXqp3CDIpiwUdvfgz(tA?RvYzz>aXny-C&XM6(_`Hf z`ZC-M7|#;<16&tOp*M+1L)9moXWIpJk00PtYV=)UZPx89>fUOZRcjJ*O%rDglSuuq zz<7ShTh`kDW)Xe%K7w=K66x%$vxVR1$TKP7-l5lfEuec|?$HO%rwOdlzWxCYi}=24 zT*ddxE}Zjmc3bG4PZKP{%C)L09~+v z&v%7>_Y);+SFx=pW4~Qu?5%}8K(FslSc%VL2kxaY)m``x`a}`ec?PasLwX-ttoaRJ z7Z@*|eWz>Ni#_}X=I}Xd^T|7dUqF{UJJ?x>=j5E6#Sq-!b6=l@Rb&%suhZyCxeV!E zd-%3Nea6T50zRKz_xl-H*LlD4ZHD$<0?(<5bk6}e!+yA84Ru*_i4^AW4yRxPILixQ ze;bRBNve=aRt&E zt#<(=A?oIqJ1ye>J>YZnjNCn9m(iV_dp$#cMa%;D09;e!?RWZ9^c!+|mUl?^TrT45 z-y-eTy4TTjFlbk226u@465BoP!~Czo0hj>q!9@{s`8;_~Zoy$Y-pzaVat_YE0XIN5 z$2qtwSVte8mA%%rTgM)dm$TMuW95-5!3xZV)dPh_RNOJmzqQlh~8+_GE{5xQ^Y&c!k^`koIxMnn)49f+vaip>tMVO>u-o30OP%HGx*FAka7d$ zX%T0655I)&JpT#o^PV*+{5<*qy~YH7&!!LW{BE)RerzBUY}bXAh!WjdJ;pZ$+&=-z zRuN}p{};qqle2WLp6?&%a_-?g?70loiL~Ag~|O}MAXnqNbI0m_wlo&jmRm?~o2GP3Sj{Q#de8C&P6*L*MSF9FVfmUZ5{ z0t~O>eF%YbXvaJF9GPIJ#5;$2Pv?|_At>SJkbQ82Uhnk|+rDbsv-F)Yx{CAkS-nGl zT*dr|uM6sUd-1Gp@E?Qcz`mZrYj1%KP{-?=#Qq5+dHg;&MV|-u_e#DU`cU&Kw_b|YfA)e7Xy0tWjoxmU9?f2inH7PhKVy=Dkdj85D>lWc!NbiF6x5$$N zW9{D>>_dAM+uU_u4eTsuf8-@`|H9|_x#y93;2Em8K%9Hl@ZS1PsO_yH-amh5cn^13 z*Cf|6(mhIn-NLp$``shX_gMET)--+xz|9e3-Y>|H*hiob0XRpvAz5TeuZlyr;zq)EuXQy(?09& z_dQ+`GX>W|nx_d8>T_K^D|7XLy-y=^V$zlPPG5vGHicJSVms$wDqGz?+|xW4tUEbP2NSDg3n5zl8hVHqN?^f%mtrv6jxdTcM=SPsDjQ z*N~oT9aqaKxE}G1?|ip7MogeR>`iNL-Y9h-Je~Z|keaI^_;?g{x z+W=n|+uupM!28n#3HrkdapbTs*DcDE!}F6{!afEmG3!YCbbdF)CU6=00N+km@8%5TH@Nkg2O&%hR3 z_X-lfxsxKkE1a8LK$lY8_kA#poKXi;z`5qg z1lg)`n_~^wM}YUdFYqZx$Of`LivhOr@4-$HXE24{gFnZ91WvI1n7_;;{O;AB!2f{U z!OoG^(S)<+96g|0vvyCJ$M0p&PwoWn5TxLvemG%nd&on)&&T-8GY=PX`a!Q_l?(7w za3%5DtCfF19%pc^WSzb0_gpW5HFxmaqc!QbC*K7_q_OkJJ7Rhufp?Z2;`E6OyNhoZ zZm;6sA$|9}1ef^BQ2jP7;@<8fJc|Wng5O%~d$)+bBlw^7@p+51PWN*!&mj%*z2Cyl zvFm$!PmUf)iny=)Oaj;L!FjL0pSAaD$5?V0WBd=;*01z@y5wtu&*(mDokE!YAlU$Uf_)+i}f?Jbw2+2VX(6UF}Q!J+T3979{BJke-P>yI0z- zdV?Fll}P9L75)kESulT+$GWWFI-H5MkAU&x^D4*(oV_!7(|(pmaV1CS+M>?YC)Qs6 zBt|_8%mryrlVA>H`0IFO7yli&069MOCU9oI6fxF4F40@ab?_O;mBg51wutw(yy70h z%gw;m^}U5_5POF78K1;wkK+%qbzI25>2h_Q+pc5%8q)WBAL(pDh|gzA&N@DY%ZLqd zQ^aflXK5~<&!4dEIS}WXd*FKSQC)|zO}GTOxAQ1-;#lK(`fL}#bKFK+*F5qLyRKO| zWL*O6-PlRu9uv2T+(GUlJ^P%z?rmN6mK4=HG-q%09KYw|UAVu(4^rY>SNS_g z-$nZ8+p)$Aq<#6Dtweu@-xy~Wkk(*)8ooX6*5Uqp;0btJur`l>i|%9lUbGf#4Xj&6 zx8~ej=somF;F)@M*0_po4=wZzcn&GjIXTz0BJMSTF6Y^Lp1)M~9I^$5AOTCD0X*j( zaNhyaJ>9#172g;3*2U*J`&?>^Wnx@oEzi|MHK#m2=kA^3a|Q|Wut2iTb6!NB0N(3M ze6#w1_u(^G1qp3x?{zEk7^2*kdxIOk=r z&ej>HtT~5sALp!`22G%Do;O8Y@7gIKH;t@wwaC9w@r`!u<1MyznB#BIWG~;_yTBNE z;mo~f-mMGrbb+&}drRR*U>C^M>oVdF+VR*5%f@5xC|M`giL2{?8GkPf;X)q$o~Qm})_8C2tq<;i^B?QTB*geS zug~~CF(cMw`q|4o&Q!SpKH*pIA$O5or2D#m%DMpT#e3yv7Qg42U=NV<#O=T>7@x;8 z^?tZ#N=z4wu+8xrnGp8~%mVlIjyt0!avB@~XJm~#?f72tEHZL?)<0DKYu0qJ&9M&( z_yF$V&0{~FhkidZ*xsL8?EeC{=t+q4vnSVdi2ohg#AhCRPk{CO-i~+h9KD|33%KS= z%;i2UkdU*-niPGth`pxRi`Xq}=dz1*5B;vQCTBiaiQh3fy0hDdcMogv&i3FQfcGkO zAEf7EFFEjhJ&P&O!Cpd|$1}B$XYKg!FW#pvc1n%}UmtxN8Q`5meNU`S1pKvbjh>Zf zHi0aWlSu12B6bg4!i}GM^cm0)aBIl4h|kpnwtJ7{SnGSTh3-BDNJ6Yf`)Lu+W&`P+ z9_K>0*MGF*Z^b!ei=17g{tcuvcV^b}8vPPjL$ipSJ$%KsKG!AenG^q|i1WiJ-OF4(jLq|LcbwqzKA)URS$S~GkoszcjZ&J&g;r9 zk)OaOu#ORVd={+n7&!|%?U<*9oCMyP7QFj7@dmZ~Rum^iFb_7b+TEn;09dz?rGLUJS2d&KuUs z4?(@AB!~JsFxFWI;un$ide6#zQ>1s^vzTeed1P?TH-W#0--NGo`|Lho+m}9Tt9?i4 zvtSjt{sTye$$;_>%;B?_68-ur-Um5r=z=5G%z_4Z1b+_d{I0j}4SXTQ_g{*fCnm*b zf9^TJ{s5F&9?#|{bn`Spx6-xic{Er*RmAu4M`X>TjqBc$BfhT1I%@rUSynM$k@jhx zUDkTud5HIA8G8!9>zwBX()W}xa?Zi|9iY!3&EXx=w+EEp!=9GGLJ{w`y|`w8-2%7R z-YNIF!9D|-V4oZq8}K`$8)EYi-|=6FDbd~Q0^T#4KsVm!D1h+?Th8ZC=^UR`dKdW? zn4@I9*sFXaq;q_ZTm;trx&7_8qx0Cu9uU(6cVG$o1iS;DhjU5bJ4IZVhj_2tZxv2y zU476aR$EwSE!CU&(uyD3#Qy`R*Eq)S44;EEkKbD()>)Hw2mKCA6fw84Tj*(EZ=|t) z|LAXl%XXAANcYOFn5i5%!M48fGk|`Io*`3WtjBxPLHFJm+sEGouF2W&*CM{Nw1ssE zKkjkMTBWrwfEK<1cwZ2JwATxyKJSJ(c7f;l8T^L+GnfMn;*OB^G@ey0sneeB@%w%> zu7Pg=Hrw<9?$I~EGROn#*{eltg6`da3&u9N?%|ujyRrl7b5RbfxUF{N?azDIT8ZC% z_TqcnKI}U|?|>e9na6wX^V)+ug)2ZkyG!)>?`-Grs)+TvPhCeq{{!CpW{o54oZJIt zAdb&Z@V^7sc($wP{|%m^yI-b_oC3Zdf5n~z@@a_oFKaVLeC9XS+H!oJQ?rQg@#~dX zqyAa&BX|vtK%Kv?yJW3gS>;o|2fqV#;a-g*=6Ckil(Jt7>AB7!eLfSU@qRM+@$(tt zyD7)s$A1Lo!8Z6ASjTibo=cW;rS^6FGa- zy7Jc|u2c9swzkLM{;E13A=^Zfh>gE!sMXYraS>iiJ z$}7Di-FA%k9MXz&pOiiB@O!UzKo{LP*>?*xikQb9W`TQpKMSOJt;L?5!&CJ7?mR)? z1D>Uy0sbsR+w(Dx=oj%Gw$PoS`+tE8!K01Z3>9f$ej&)8_c|IA`-c2QTp@$StI`Slh zKrRLDZ5{PlrPy6~ZDZ;^tmQpEd&>g#<9h-;_j%+z@&pXfe@0rT7?RuBTqCue{dg~A ztH4JtZRh4*?&;a&IXw0jb_tHbY#R3G@DAD2GBM+K5Zp4@z;5EZ1?HW=uJ5uv4Y1#W zdM?VZ;1YNSHJ`BF?+SD6z@LI`Q0F{A4@Ha*D{(*P_5uD9xsJ~nKSp|o-UH{j){gfm z;Ohd<)4X|zf1A1g3!s$ypL$s9*&M=qhSnjhVHR!=8A{yYCPT*%}5|3B#KZPp3i=6zMgAIPG{-9_MZaxw%#Xjo50>u>;}B&vJ3P( zyHyYhYM}?~pF)2I@+<1WcCSlt$Jzw@0o{3Zh_NTJ2VCzxNbo&t$DS_G0~qu8GvZ3@ zJ^cR0sb}XL?Z@XrZUUe8${IbxEdMtDDY#`Y0lZK4Z=83pS5O;BLd;<;oA`#H3!J&J z!hLG}2-`cZ{Tbxsx?`<%JfOcp{sua5BXB~@4)zXG-99}RYuYW?4`gs_NM~wo-Z^LI zS*Xjs!`A_>b&h?I!KdIc`W$e*d$y2yh<9!i{yDaJo}v$`HBGp8*eP+ITLU}bvu@{o z3ar&LPCyPy{My#;jLb0))(iH^^$ZWOJ=YJQ54>v`NYKsWeeu({ z;(z7w-(bvRZ_c6v_K7uL8RAR^_!h9e3r9t~gVr_$>V2=_UEd@oC+{(~c{6hK;7^H7 z(6!Ied>3c917X^Be(t_YQh))-K;gdZ$_-hra=H@bwIoO?=A4e~c^`*vLhcLC>jtG4kz4|Q$IdbY+mm)e)1?}89wjVUt0 z-UjA%b|px_7Tg{F0LgAD9vy!H9J^A)h>1l}4a+wnd78@#ih#}4S8OOABz z88Xjfo(1$F__H1RK0$Zg4zdBR(XGuhePC@5d_s30*Ex?%aOcP{r}lbDo!a`gHf zly9n7^}o=q#lE$(z#887`3g4rY@2h*qu;oIy;j8iGuEWUEMkwqCNM|8{q0@g-I)ff z`0U-YagMLyn&2_8pHpC;8z3QP01N0XV%_J5g1lfv4)ge3Jp%40>^HEc0~hei9iY#( z6O<=)%22y2ibYJ;OF<4{oQ3J)a;?;jGiU=owCX zd)U9O-Bq3rCuAb*H z{0oq=#~1j0^f_RjE}VUPuWra^9nQ@8T7QCW?t6Sn*9f0i?J~r9^olqG??=Gb!q#u? zv%uU9q@1___O@2UGuuYD&V;p1&;g#Mdv&g2EhA)$m`mgm>n`!9=+^ALI0too*6Ue4 zhHHW@b^z9rg!tTDz)x#eJx!ud!`*=sAm2xt- z{aHKyeS3*)!QCM@;hl$PWFBYr5joxmaZ6kWIB$Dhz?bPSBG<(?#CHAz{5QxJ(sOCT zuLJE4@l8@~b%@`Y)-$|7U&CkJje_S(8)-hzCkyf4BG%DggC_nP zu+xrzUmEuVdIo3y8Di@7hv@%Pm$T*Bv$g?njWsk`V-0mZb$rgg2{6Y9zdabcf$jba z*nM=@1>&sVGjY~(DVPMVadtj?pRl*UQW5*`S+E}GS?}+h7K-@%2G)2VZm{11YjZ}^ zz`T3yxUPqs2A*jN+&in*&K7aqTet-L0$!b^49Tiw( z4WF@>!SxltA(3(uZ6qj(IfvUoPtm=*fw(?(ouSVG?@fyRp&jF`+kK5Qm(R3wx`#g^ z-Zh?GS@~aKe*wk{=h`6l4*Y>`?e5V)&Vep4|1B^+5VwtfqaEUZ59fJ79xwy9hOF<$ z8n(~AIi}$2Jr}TBK)+|Pi!{gI$R4)-40!?P{TcU%J&kRD^_**I{1LGJ03HFq%bP_! zr&)Ak{a$wd_UcSa>Q{EI;`$q8j%{6aty$qz;%{mF?)yNz=c<2%Zk@hgoR@pcTbuXl8tw&p zKsJyKWVeWUH!69D%tJ&M{~>k@n5P3ue044B@ZK-yAl$zR2CPfk@vP>MyZ8q1^-P9v zzv_qE!PltJWS#w(|9g0j?|^e^fIjfvInz4l6ZC8)z9Sd$En=HvsgmaIV?PJ-tMKD_ zqFakS9?9{!C)St?q%qd1o+58S5_k_8i%%59mU%oU^S;KmCv#e3Al`Y8^AlgfnLi+X zHq_1Y1>3rR0uSy7^kvw7JXhyc?|ZHe% z{}egIeg%}iuab8B|4Y~;pXaUoMZU_rfxpXs-nChD&wLk|g?R6MR;G(+ZzEl2pT_iv zokMy){{joZ*?Au}k<-}r`xdML_X%)2=;nEf?lbGXug|!i!3b^&zxFM9&3lg$@VLrh z9_P>?M$~;J=pW#hfHQWc`dZ4IdBNRW#rL251h#eg+=mcj+%pH}^{yoFBkZFhzK`aS z9ehn>eLfv@^L5+t-=oZN38&l!UG`kZE?F1APjD&n9{4W$pZeKrnYaS#*#a55`#um?ApNd$M$_<*fa|@>3DO!y zaLyzP4_U@YnbsW7~5M_cL6|x`6JU4de-&>rSwL1R0#?sjq=;jh?;n z<2ZZ?n8213x7hmrFU;+^IAeQG^BDgL?w*|95BC}3yIcA8J9Y}Y4Dp?5+zs)QaOy2! zPtL>ryjM&3{st2QUy~SN{dL@*_;R>QunqqbD6PL&#CJ|$&r|fej}m>k9iJ1=x`q83 zso!(Xz=(X#Rh$ERy?{GJx+YtR|2+hs%^B&R{jaU z2R!2d%;DL4SG=ne-~{{zUKP=AjIg)SRs4PSAAD!vNAMDu+ge<^2Ar#(F4AXBc`t{a zA-$*G9rLaN>$SHBqn8hB3|NWY)| z2KMlOVg2j+km^I=bM*}AUY>m)yI$9Yo4_}Xv^Up$MSp_)6{N)fj$Ow)vlpOA4*Q<0 zeD?kb*wa=!=5x&nT!6E;I*0W*AKzV`r)$r^V~~b#|NS2Ck@dJv>3w}xLC!OSeI+8Q8;^Zm+n(>lgQN~&SMolfFuCw^L}|3zJiwP z;LVd1oC$Cif&8AG{WpQRPC*a71Mfa>urt` zr}{4M!2gV`uY-O`ymPxCpEb!li&;45^BHaj-StgS0&A7`U1UrPc@O9Q-o+E_b>N-q zfF|5C&;#Z#19KwY{x?9L$M2pa^aQ@H-Pi`}pSNQ@b5*?goQc>2-W~f^e+ydZ&kAy} z&n!~jT-K;<4?Q5aj?Y@Q;O*0SIA42k4(1(UUt$~YT<)*~w)uWXdas9-%;8)=Sc&(> zJCRdI4=KiWJLcMEjpy&&t*HZ?(IRkPk+F6S`v8ovgUI8o6S!?)EvraBpON;_g1fzn z+!T7wI_tCtF;C36AAEa2-5xjKt;y%}5GlWjd|AZry)?wR%ZnL&_H~BeHJ*w42ljr8 zJO*;!nG51xq91{ui@4sNhuB#jYn+8|5jTL_MfVP!=to-fEO5;tzS1@5?(h8EKY{Du z%aNzV)^%q3;YUb0`*H5ZxQBOYh`olqLH6Oi8`fZqbM+poKLZC~0$7jsJ$YNecSfU# z7$CF2emV0%ZeY)V0l9}je!v=GP3QQg!9K`A+K#jI+?LVpu_TZ4Y@iQ;J@<=meIQ3Iye5zR{6wtv{Rlh{d&}^- zcY@D8Ji`ZUZ6%w@diYHy<=_S#~?;qGHK68m_;69%!yCmO4K@9UshTA_WcOF z1M;5hi1l|!eeUU88rUb;6|2o(P>c4>=@D5BAG0xbpmHrXmK6;bfzoHL_(?7tT z#ooi01Z?CKFt@cVfF3c{whiy>H;^f|>yPECeLRC1VEhccb$JiWZ|yDg2EOIW=ex?e zxA5!xg!Ejm(LM7HXb|h?4Zbb(DWpAYBCXxMvnp18gf6#^-+S1_&O*F@%W$EHdC$3 zd#iGHumdr!yT6L}{T|&OeYVb7}SX<#5H^9##-aQ(XHLN)-@`9 zrj+(?&&Jk$q~ys#51+hf5$j#_z5ED$AIxA|dlT5B`2widjMz5`@xAf{{-lULXVb%H zPW=hG&uq9Nk2caA1N=kaU9`tKZVeumGu34s1Bn&d_Byo~SWof?|+?BO!kdX|yD4!aIezHilXfFDqOi& zmtaLv>R!Uzem*{P-b`%I)xIxl`0zaEInQ~{bDr~K#(9x1(ld}pZ!*@i;o6OV75gE$ z0?uDVI{!uFJS>6dAV=H={`TE$*(Q)n<&SfDkK??#mTz3o2C>!0kk(}YU;o|5{8F&5 zoAz}m&~{C(Q~Qf7)Y18d>DN}EU-=B1b2bv%X|TDE#-4;ZuwP%-qjYVTps&^TGhsI- z^VUIKtVKef&HCatL=y3ETvhn+o$gU?KBNFbB)%b^DdlHPZfJ_eH=u(Hs`heuG#f!H>P{znw;Oe z^BnYzRB2C@!|yT6_^m@9wjOM8EqU~266S224&OiHc=PFnMeNqa+&mi-*oqHBed@?s z%Qk@BJf9<%@P7$O`Z1q1`i#AeJWpQ%e8Ze)6`Om!yI=S?uY0v|I_z_-KIvrfp71 zf^q%PC!sw@Tbpw$n`wB@7SJ0BZS1e@$KQVQb^ZqJAG*L= zX!lH>0rhq4&S~sAc!%7pMnY`bS78*MK@MAuzVlh=^IYZ9;OqEFctqbOxW~?GzUN_! zai3*jFO4yXZ3DK^<5=&9=i)k2n>o9O?zgcfvvAf2ke(~;*41^~!A&qTK>|Y1RbYbt$!oU3`X|OquxDSt8x^2%xBMUy}m?R<3 zB7N2-p}$0b6Z-@*Mb^?#n+AQiu~|=JIcExUq6JdyckyTO)>bI7GXf_9D8}v3FssLlw4Y-$1$s z=kHbr--re|DeutsjEsP}-Un-)CSeWkg?sG2q)6BG3d}2qG`9+N+sm}KX&*wGi@xUR zeSLu5L^f$ZLm!9bEchqPwTyi?CUG9o_ZbY)ug$YF2EO~o%OP*VHu27tLk$LyUFz7C zo(ac0_aCIp9$kBqgmW`W`z*Kz8>d5F(-}3#)(wumW}80q%pva}?JsCY z*02qu7t!Nh-bJrI3~RL(4ce3VSn~#RyRHH@W16#4j5i1Ca0Pm^@EO}+T!Npu{{Z@W z@yQ|0&)=20X!|~nAhkK~4r7gNez#y1yLVUh3{cx9-D7`jWKoPCiY@E zysMK)$4-Iq&u1U+scqksBJ+EOjXw;tTWzm#_C5l zT5WCDvesTBpV0QaeTM!L-Tpps&sLDu%^JMS!hNt_=Ab=)y7d`^O`idz`{i1V)2oh8 zE)6wu4dd9q#IC-(-XzEj+ZZvddzuDYm3EWy?oXa^o-b=`{T`Ot~t^YhX<4krLB7?%*?m zoX^6WkqcQE zqkj$EpAvoX-km}BEr?^MTl%w%n6M0uG`!c%B-~qLRm$Nk7%zv9Yuvz=2j?-TyYQK9 z#+^f2w;m|8a;+jSV)qR$J`DG06Wz5;;WrBA<-K#x-8FFcA7gJow7oz-2i~VVSReDT z-~09jxKD@5VNJ~>`rOa_!|$`2hq2|75VMNie9mK& znKJ&w=tI|T-L|ofLduwB`m`^%`oF|}AKkYl)@KZxadzv3|MQmaUb!~?jAOqfu#U}l z*&1nEqkjZhgoM7)jCq;A2YD0LU>3|Tu3?^b1FYpDpuF;-?9QrDxFpb{@sGxVz zSAdw$SLnVWo=5j|0oj#>^D|G|&t~IWhXlL1t{~l`6uCpYAD`W|5oa0wW~=R4w;tW# zx)0HI4ZX+-Fc0gOOTs@JZ8m)W+LPD`0cF6bH3 zbLm)fa?h+$1!$9IGCUVvp{tiukh(zd3~*+jQCNdm|! zHsf|@p@u!k8W?wX-x21!g6%wL`vQ3b>iBF!p&ZV00bTnTOhAn}dXd)A{FdM~G_bo@ z)hwKgSI7)<*k{WjALpqOuSUBB16lYTejfQ0n)r^C8H1nPM)&hkyFX_$&d+K7{~{pk z;QF+AX05Hh{%oL6g7u2AY|ocN?!I5z?`s3s=vr3MH(_@j*sO!+W|F=)*xl=H^h@Xo zxX;FPUe~4FHy{tjHs%zJqg&@gV106+t%BcBIeeykPWw9OQ%gfm#`mu6&V_H0wmIY& zWBvp98m|J*=f0;&cwe-8PU8BN)p9sDewVueo@r}j{>HVo)^eP_9w=t5`LIpW-a^h| z^Jj&29bMaQ-HGpd^H78_aQ|Fa6~>_tyay9V*QI|K^wX|me}b-U8{MA`>>KLD7HgG0 z4CA9s`8oC~T!dK|%)&XUGWHyLnugkrp*z3vr_sH`uBTWI>rAj4&%JCgMtvRIGW}n| z81@LAKGt>(>70(4!afbw`aIIuyY)ckX?sre?MHWAmth2~MQ`>oj3X#}&pa|G_!}#Q9`HPPwv21tH^9B}Gh__ga28^_2UTp= z`4F;+yh~q#97k5@ckR}@f~=Lpeie`f+Rkg-ToUfHYpBys=(AQOu>MW#7tnLaSO>?s z#_ROo26I`a?OH34LwEi){G5M^_BQP$`tnHU-ay`=ZA`~&+a>w<7Pt>Z`m9L}`e?6# z^Lz`&>dL~n26_|5@ViT20o^s#=*Z&H zYs@j;I_DDpFZ#uDh&73=fv*1$`W;Z8MjB_aj1K3)-z5^pyY@$5oE-TKp_}8)Ea;2a zd^?S2z1;`T+(=7Skh?W&)ozP8+w@<@?s{t}d)o5(9Qn4TJ8uKu3bIBX_mt#ht&F(> zPE|N7`dw>r?Rl;8~A#@eJ6}HKwlo|UKQzceWU1k`U_y5-l;2KpTzxaCas#`V@Eo>E-g<2A% z@7eBL$Od?B^)@Sq z|IDQh{n)R-797gL??031OAvE6&JcZthv9poZTSixLJr#)@&WXLb6Bf+#^&K# zIn>`=b7`opXSxd?>(YzeoZ}qx=r>^ydlA_H*Ls%rdB_pdIPRHa+*9XyLEkWbW5{l} zh~5ut`X`~5`W>Qg5ps-MK~|C4thHE&s4K0lwhA=T(01M;qs{Mnf2zCh_lE!D|A6^V4GtEM9sQBk{|7OD zsZ=^rDwckv`Trp5@BF8VgFn*ze?aR$I(UjZ@FSeRKdAYSpZXE^pSR&1?}6{xb=U#l z3vJ51_=cQ;IY^T5j^(rPJ;Tox-{lI_=vyv_H|HhN&kx5uf)e&sr8fH=RS) zU;>&c&x8=O220rd4x&$kc7hxMznAWSZ)^>_-#zs4r#IvOflcQ0v(dTx>}##@1ARSh z?Uc5Cj^765LQATP{x#;fiF^RQ$FARZKS$g5+s}gyY}UkebZ6nSaS-X}o-zGCV15ny z&m&W8V_?h~cnkNy&yfLm1g^IPzV&%|ywPHck^gu2<>zd28No z>!F=Mo$=J>e`k_AoxrpgoTI`vYv+pEK?mSpRb{l7-Ky74)0v*5U>D zIX#OUEQh(QtNVTz+YG3$($^2x;x<@|8hKwu&oO=u=~~Gr9?R+T*a)vY$sfx990T zvWE108D|85 z;Ze)}xXij+`gx>xc>w8IJwu;yY6)v1moI2Pr_VUnu0*>*`vrW{lJ?DD&oRdJj6eg5 z#JP!V&*kfkdGLFR{XX=7?Xiq8#2>cJxIv`n$^EHfe?{9I*J(S)2T;BQYc&AgC;i&z zvliXhn#?g|AJTZ^8TrwkLPmezUhhj2yXVmPYxwt~Pl7d$^S(oO&F;}+7QWY?MXy4P zr+iPJ^_zisF0Y}d#P#mER{gA{@_I{NZAopeQU}&x3Qwka}DF z)VHu1w+DF%92;x3gTB{4_C;wu4&&$A&1V(43CnN;9REcY_HGNg0x{+c{gt-%t}nJx zXcFr_(%jtp3iIr~=l?9cDUKV&J_)X&2-dAH3*Rkr=$`Ke?Ks~yy1B%*GQn>Qo`Y-h zeQiJ&dKHGUaGv#b|Fu_6^VxyE2<@@djo}?tx(?5<_Rq3#4}81h7;||-zrJtbDm3Z$ zcgkjh%!2P_s}=X`Nolxz5HgE?f7hZ_F9 z$Rg-xjeIYy#RTllCB@&lJfr83F?JRG8jM0pU#UziboWso=dqRvd`6$^@2B72bB)!F zuHO`L3l_1N=PLLbueAGe+N+jB_6M zv6e6v+ejAvEzw+-nah3r7JY-h26%V;J>Fb4!CdaO^cr#u%<&HLK1_mhxDMq6_*OTO zub~^xK+OC1E0-8E1zlinud`2Mc(!(HOTTa1EV7*(?Iz=_b%C*iv|YQO>#lKc9*42b zKriIs26*OO(-3%etyvtmMVx!I@56&~_?(M<%0=wk%-x0boOm|+@Ed_~Fut*u-~)Wy zYHJhsLU|X?f-%3zKF;4hmpOcYPhZBES1=5&s|2gWtb;Mz#~-9GWYi7c+(YgY`d%Mv zl+tIN)t8mD2eV+`oda7f3Hz`^taJ2vc8bUt&zf$-1&DWGw@$SCz}o5KS@7;pBfDV{ zUO=puXKsXc1F625eVk)=E^I6GubmD))@2S{qkEyxNEUKjMH*)qIZodhWXw0pe(g|X ze9Yl%+D+Jm8uk>np+@^9Tn2MUVFv1;UqxFIYT&+uHBZ zFSpuv%3)s~A+Q%Dj z5WDqo{WZ9b?J<~R5&w8*?$c+SKJ4BB>ukUAoIl2Wi$2$CpG7W}L#=y|UC?BVdpU)^ z32~hAYR0~&;dhr&^e)iu8ktS``vmYmdbP+z3qGn*h?$0E9WF;))wdWNoH9|P~6 zV`BYc4%&uVed^{MeSB|L=u08)R{Dt!C^kIan`k?y)iR*pd_-A8Wh@ z+QqdPD@{Uut;q~b!V?&V58ynP;9M5Y@fq|AEJCcKc~>C;^?C6A<%nHGH^%*p_lP;I z)fx00edg9fUoXVG+T}C-TDj)%bv+HFG49c~-RjrJ?{*J}vvE4y#|F|mDdXFxU&fdU zzUC<_;C{8wyH|G~yc@=EGQW4<^LYgpp+`Pg=6AkP`PzUoUVb4PsH1X?4t|AlU0GQKbq_y`P-2msBhrRhG%vA&L zf^~7;LF6Lz!(M-V{IPz8c7(J zpnD#iX9(T;t)M?Ze+=!pzohM&4uN}FGY9k>v2w&+#Ws+I^I=X^uzs<=G1e$Pab26} zeg}GwjOXAW*-d|e_^u;iK4a|qrAXQ__UGlWZ&&C)3kiKc@THD1wrAQgo)u-ZD_65{ z=e|U)gYV%cJ{LgWb=Zx`F#Y2lSj@tAAn#*)JnQavpqQi5cJvvf{{h&(i+n|Y<#h1B zj&9u@a|LEw#IZ`hZ#-}HUqX6LjNd@s%#h%o&4F>;8)ZK{0@qc78PH~5dwg~8`vkbw zBJAzAj*SF+8pc1N{R*yu=RmzlpZ6j`I=-F-8Dm|dy#VttRSw@jX3*XD`?Ot`XWaZ= zW}l9YF{0E*9<-$E_x)IHwO#8i7=qy}oaJ@&HGER&#^xTBkj9B|qkM<`DlCHejUh+N zVax!!_n|=BJK@|3?J-Db&x3JX!z^5fad-nU{@3X9P)j&(jH}qkkK^MSo|M@a_#EQ6 zD67~m!xotTP=*fn*UX`%yBGFPLLFvd1iqiKUl8M6Kwl$|i^v8{mFYwJURXn8v{S@B zj;#a4W&b79F``$($^f;~$pBZpn-q#c<_J7b; z4)3=)K7(OsFh>&_%rQ#4mNJHRo%S*@7Lm5c=yN^pr{nuGa)*oPDfTgNtaEwqVovR{ zg1>KQu`lE2+4%~-1>;q*6`+>zoiy5zp5M5ha~bPSgU>Z|<0kMGe1H_3dlAfI z4hoDvgdEJmyv8zz7}K09_^i=C3>WD$_gU~w+n_I>g|lD{Qf%&p=g2y$uYz{R#kdPu z_{@8aZ5{Oc0_=a5h0mQ;WS#itZ|>IXLzy*!KL_2fd-Sd2XRg!w5$_RtJexDvR>{-4 z)sZnDrF(pX{w-v*FXQ*Jt?gdkf;VsnqF*POB=|6o^>F>AEd1|t`fsAU?w2qRd1#OG zB>T8l&-^yHwjS(#P!Za5NI$FHE7#Ul4&NPe*p25Lds)Ty6qcX>#VmaHYM)>IOYF~~ zAI8ezJ(@zg2hQcc$zG*Ty^76wG8hH(wMMpIz!tQ}cW-^?Ca}K+#~#kY9=H#l&9r5= zFHhV4VRXmVPKS7}uze2gvDMeG_2RpM?-leaRNx}`W9>%D_|Wb{Zi4Ij47}sB$T_gy z4W#x*=*F0Wz4O_2j#@$=vE8#A`WtN4*E)S$4)@Ib{qg&O_huc&=%0rXFjgID{^qg? zUFgju)Fn!DvNoQ3_`H>weJp9Qgjax7NY4ZJsY-6^yfi{S3M_d5pAHjWTU;-aFWmB-{blcLiGss`$B1 z=Wf#Pd+kL^D6 z!Z=vV>&P{vIj2cD7aO#DvQV?{BQMj}g`9*LaIOqldl=5=4!UdEr2S2+U!UP}sIhZR zfid4fT%)xq(yk*5^nU}-U<+=8HFX_lU6$C&0`%fdY>VDn9nW6xkO zz$SKUQiR9g+^%a3Ig7jkdEz^^A3ctbvSLi;DdOWhbhqVGZAtU`5|pwC*61cUUybn% zn9M>Qr?H)dCTzfFMtn$!?Oh*6zXhK2uc48I_n?GrcdzktAH6sFucBLz>)Kj&=P;KB zdB=5ZW#RKP=IOXW@cw$9ns6AO=&yXgr>(6<%wFu{P-lMYcOTxu3vDe5^qKD(aaYRW zjr;;x0&C%(-bFfRny`NSJ#(|=kF|7P*08_B=PUhaU!&h#o!dELzP>H$hw1lD8Q+)- zEqnBAa|-#XQ3_qF-Dj)HyIS09FTHSxQHeH$J_j2YzUb^U7zH6WK)=yC2St=XzRt#}n=AMGM` z*O#J?z-6$O*0Gd@@76=K^_he@bo)}qO`}_f1>_KYC8T4|g6o-p2e1fZ_?hDw@NAfe zcdL)~FxV$M;2cfXq8tR@&?(p;#vSA`-Dq1vtlk&um)Y=+6wq=k&kPB55~9yDYVB|uVLH5|6E(2d;SG&@5KFbIL8&@ zS=09MpV8mmR=URmL_L>szvjv~G3V*uQ{B;GVjMToU$s5nG%~ zX}fy!3a2it{ldi&kCg27LdcxNJ5?^baOLC z4$L>fFUf+AHJ2vg9K67<2=4hD(i&S=Yha%4<?MNQjlkK2Ce%A^AZOJac2%YhcbP zeuuHWMxR8^K!Z8kYolJHe+Y)5#u#H7<7_$Hml|WNbso%T9=r9lUwv;p>)B*XgT9L} z2(GP$G>0Yd&3gduhyE2X<|1rnpT-=)HUZtB|2vol>rh2Hzi*27rO7;d^C`6ACrPN2 zcfxfz_8dIUK7RInjlKmL7{@*6M|UpILN{3Z)ra8?(`Tg|@_O2mOZcR43vR*!_|CNF zQYz`e2l+M29; zBEtvY8^|quH?gO*2Wj6ycRuTHUgo_D>u?*|^N#v9_It3&xM(ZT-T>#SGRG*=dai(L zyNGm5g6`+qVPt}I&I{UzwS&9}?vLM-+|LKJt-tZ7VUqR&IBp%{{82v2s5iWTbFfH^ zGiCh1xVD#(muaWy&+&2WGV&!%!u^)bIy^6j?-I^uZr1$@@>`e%?bqQfq)9lZchI%F zcNx5fZAh5cHCUetIBz2j`~H}Ib8~F}AgjkG=~wt3h@|A4g4 zuH7@d0u`vGt#izn27XVlIrle6&)Ev(>AQ+F?`L4XufVu#;2xNF9b1xwcX=4S2%kX; zH(>OE9Lq2Xl;jwSZn@Ue}ZZYZ&{Ye234o)|h)pa~;F?KC&NT>|r zj~e)QXW?C)M>^g;O4IOfL&jK!EpV-eke9%F^%mL0e+)fAZY#k)&&eWk4yo-B@(uov zz?kMU3fAAYzDW}Hb%C}ynokb4;DLHM)Ut{E68zmYkG;UyG1~XQy^3cqt~1IR`t|F9 zu8cdyxMkQu--Isg@iTshI91xW!M*by=F)Ix&D}VTbuTOU_Sz5KP=ExcVF~nC>OYc& zGq{c(^J=2kAxXl1=zku+>&QB+!a5Yeb7ri&E$lbf8e?O;D4%BGGv9qNzCZ4FBjG&J z7yXo9(H<>__rTAc={^1JzMb?3jGu$=w{SmpU=6ItV)p47KSz6aujw-{@5()F{xp*C z`Pf7@Qa*?2OXzd2eXoaUe+ka@2#VNO!Sh?8T>{U?BJDi(Eu{H*pQ^NX_mg$Euf@GI z-hG&W6x`42pxwE1@L9$@$Zn+j=}(RJ2tM9peZPPh+dH6c9p>N_*l&K`8~4@x29b_+ zzG--ng>e(eN07lbd<$DJRi>XG*J2FMehRL+_>jAqhPAGuSKtX_B_(|A*}3xDjmT>e{ZrdHNbjI6t1XI(^;Pe1n`L&i74AzlhH)xMt^w{mNVT7=M+% zFOg563J>6L7JOeJo6NC^yab!&kNH~nP5PFw8}9|uIm~McJ_qgZvXApxS98)&a#?tP zHd^h3c>U;GVBXIE6r87sbWDd?|IL@Vh6QA}quG&t*Te+JfKKk8iN%w6&3-_*q ze1P4v)D1P*hWqdm8sOaTka1pP*0bT~2d(6)v2ZH|6>rDs7~%yka`_VzYy`z|B5 zh~v7hfP3Y8uJ28V-T-7H%tiUA z)i&Q1m;~!LTn^tUs>I)*?Pp^JTOMg$pQF1LV{FilF=w!uM*~vq-Vg7OIm?rb7@)MK zU7(MBp8rM0S<4#oF*1iVha&xt!MNr(h^{@(yM}F3fATp9^L4)-A;-b_E<(HSYDT?a zcl}7~S|P6U%p;p=_tL%v78 z0j^1UvT)}-w+Yn1efbO?!7}Zha@dzV`q$W;t9lw=-U@Z^yz5uLi?3(HJFyPt@(p+f zO31C2P5TR|!JRhSVoQ&E?Ydt<5pF^y3va2lo`orF)>OGkez6Ww_Tf{2b#RZ(&l>fE z@uFXpFS8J*8<~Oj8%XC;H}4H_P3pstCgJn>3Hn=j16$-|jjWB*9E!-hNaJmT@dnDF zb`{z?4_ogYw(Bqt`Yh1*4XC&C9Gf{@0oR!z( zW$bi#6PxH6xQClyeCuSrs`#%X-(}&ufaktFpC&Q%p8@y8`{W*LTY^{6?*BMzeJ8=M zmWI!HYqSPa_&ToL{~+BpTSMph04v~J+r(IgbLbU_e(wD|xK7{13FJF)u2nc_4s9>9 z&}STd@94`SU5o1*MNY;(B{u!4=+|0y>t_6Wa33yXe*zz1 zl|KD%pz2W<$~@NYld;5h&vwA|dXC4y{i))AHVb>`nJtpb zE9_6v*FoDb?He!&o@dXe@60HD-(Xv(?Ko??fi2E`4}GWAzKu*JVU74UTXuclqR*De zhw)A19azDpJ;rL6+t|&yuN?ZlpI@L~1<$AJH2xuQ&l_&cQ>of03Z28mFEoCgy9DR4kph)`_ZO`IG@ZC${0XW99aswQ@ zjBNrwsRi+^t+w|_b1>#1{Ax+acLUvVGur6iyZ&YTuhZvxy)z40@EJkYXlplT&yICJ z2Pv@&$OO!-n1#D=n6|mEg7qGSCcb^hX}Ao}!I)xgJ(oRMShwpRL!WIih-_c)IQ`x? z&s>7EM&|F@&Vw=EW#R2KFa6rb`-a_tXnTv^4ZYwR9N#3~U8MRqU=573oP|A$W80#;rA%c_+>N5_}6^z&beI&yX$n0G^vFatC7kDAP3DnG5*1ztL{IDs#m4LGmp=&e`7A zAdXnW(Uv}rtb%V^clN3O6>OfRi{Lo-MP9)ZSOL%09b^t^o{QwUi5$_!T1#1w>NR{8 zX>YW0e~*uGo`Er9eC2KM4y(^XFPQ5VXv@==5HHEX8<%2RfNS7becIQqu1&6kXX8n> z-=`OweflZ4U;tYaj2&aFq4#5(f5>`a1$`P@ynFh2PCS3{>`Y=?BcAs>1!V&^PKTQA z&9{huoUfLA)U~~A>wAG-g1zJ2*H;kt&U@y)@$Fm7!u!>Qya8SKc`n;y9Kv=6dY})U z!5!#_9GJrroCV*Cku20?jQADwn0u72c@>%|Ip9CVIQPQ4V$3Qw;~GzyhcpQ_Grr%I zN*R6FZ2KP6!9105o+uM+o=ewl>|y){;4wUhue0!$Eg~<%Ic($LSu=L5#Tb2~;9i<{ z^J?-@HtTrOdI1;*cmp)BNQZ4>gl zLHi7N-mPmL{VGhr5~SD~_!a5*3?$(DoM5wN?e$WxF=iNwU=GgfJ2VfDYhW9Ni=eOj z*d&(o06=QrA;0GbLjf?z}7?j%H#rHfoIfOdnc=) z@3-afUG_F|5%iC1RX&63a1Pu{&&>u{muavL1?U5FHisIxH@-zzvoPOvF{uhb=O!dE*-aGmiZvXv4{JXlp_{qp~ z#A_cN|DJL20adjtLEZ~lXogELy_6zEj(6n|cLZ=m)+ zS^VkJdj8bG8M&^~{-yo)Pygw^m&<;PYw}~cPXjG{N1#(jOM5e#d-c2jS^PuNp9aeR zn;$!{*hfmo2L}iD1?s$O_5ZZ%z(7Zj96iEc0p6R@fB7em{JD6vKh3D;S33n7Ji(vx z-5cmTGP>F!(9u(+6NCG9t@HNx{7;<&@dvi|X4HB67dizxRy=iLf3e@bx%|_)0^1LJ zYILnbMkh`jJ+l8m{?1SS5B&!hyMF&C58kyCrQ`hF?!8NE|6|ynW|aTK-#a+a$)hKU zv^UUyGWpqyIhm*IfrMt>;6BjE zGPwWJcFw5&sQbWTA3HW!I==rsn)um2+Bo=O>-py$0v$hk9=j!Yl ze7bA(-~K}fXLMrlNOAw|@4U2p?@u2bsC4pVX@8*32XdiPpi`%g9NWLN|L*qte{%5B zdVc=kj0TH?$4~CRBApLp{U3D-blkt%zxQP4e|PEcbvQ@2ItMzj|IPl*zx)q>>EPR6 z|E)s@7JIOGqBwYD-xcY+wEX}0A9W0L{K&p@^c}m_DbSIlM-TWCasK+BIMU%{FLcQ0 z7$6%*cYht(t1`p2Rgq0v%hmj3!MU;{I=a8uzw*6* z+3B;tQy`Ax@%`K1Iivj3pFQ{-@zNgI|81mmpoI>Bju(#%9@+nl{LVlAKmVJ9AGV%< zb#O+<`QR`8Fkd3-|Mrg@xU}O(_z>BDAeVo7>MtC8X?wOi1{yqi^oKZ*_0N8;Lq>xo zK1P23x4-9~bO?0x*s%ld+Ml}r|I8eGX%{*NI(5Jc`EUQ%fBwI9`0Z%;j~=|Vqoq@& z{cm>Xhpm3Cw?m*)C;0Ef?5)wsm-D~Z;T-*L=RgO1_50V;h2J^2M)iOC^Bprfwf{>* z>CayHBL@e{|M_1yIMDH<#}4=s(fO{`Uv&t?dvtXF7p=~LdVa4{pwh`=ao?v%=Rgac z0+o&qp7>z`^)x$Vbn56q-~0#u-}C?S;7hyEA<&5K}gT;H8EC9@gHw z*11MK|Gs0OgT5YjzWocG0)_uN)!vLc2kQBLhm691W@>Ms&P!Y95UBOvi`pCLJ2LwJ z4$i3cAAb7%2U_SH$p3=V-azgDH2UfG_vHTgf!j|;NBK`N?G4oaFPuN~$-hI^$3F&Y z{k!VeuYd1*11+at`iVdN1NUERX;J>f&jrKJ{>4AOt1#}TRDSL^Lz{oo;h$4~@_X8T LjsK&6@csV*H8d3= diff --git a/trained/ai85-kws20_v3-qat8.log b/trained/ai85-kws20_v3-qat8.log index 64b373ee..5cdbaf71 100644 --- a/trained/ai85-kws20_v3-qat8.log +++ b/trained/ai85-kws20_v3-qat8.log @@ -1,33252 +1,33653 @@ -2022-12-06 09:30:55,294 - Log file for this run: /data/ml/afshin/ai/kws20-enhancement/ai8x-training/logs/2022.12.06-093055/2022.12.06-093055.log -2022-12-06 09:30:57,423 - Optimizer Type: -2022-12-06 09:30:57,423 - Optimizer Args: {'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0.0, 'amsgrad': False} -2022-12-06 10:23:10,031 - Dataset sizes: - training=307166 - validation=34129 - test=38058 -2022-12-06 10:23:10,039 - Reading compression schedule from: policies/schedule_kws20.yaml -2022-12-06 10:23:10,082 - - -2022-12-06 10:23:10,082 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:23:13,247 - Epoch: [0][ 10/ 1200] Overall Loss 3.044527 Objective Loss 3.044527 LR 0.001000 Time 0.316427 -2022-12-06 10:23:13,405 - Epoch: [0][ 20/ 1200] Overall Loss 3.043691 Objective Loss 3.043691 LR 0.001000 Time 0.166083 -2022-12-06 10:23:13,539 - Epoch: [0][ 30/ 1200] Overall Loss 3.042493 Objective Loss 3.042493 LR 0.001000 Time 0.115184 -2022-12-06 10:23:13,673 - Epoch: [0][ 40/ 1200] Overall Loss 3.040544 Objective Loss 3.040544 LR 0.001000 Time 0.089724 -2022-12-06 10:23:13,806 - Epoch: [0][ 50/ 1200] Overall Loss 3.036430 Objective Loss 3.036430 LR 0.001000 Time 0.074420 -2022-12-06 10:23:13,940 - Epoch: [0][ 60/ 1200] Overall Loss 3.026586 Objective Loss 3.026586 LR 0.001000 Time 0.064236 -2022-12-06 10:23:14,074 - Epoch: [0][ 70/ 1200] Overall Loss 3.015887 Objective Loss 3.015887 LR 0.001000 Time 0.056966 -2022-12-06 10:23:14,208 - Epoch: [0][ 80/ 1200] Overall Loss 3.001765 Objective Loss 3.001765 LR 0.001000 Time 0.051511 -2022-12-06 10:23:14,342 - Epoch: [0][ 90/ 1200] Overall Loss 2.979484 Objective Loss 2.979484 LR 0.001000 Time 0.047269 -2022-12-06 10:23:14,476 - Epoch: [0][ 100/ 1200] Overall Loss 2.952368 Objective Loss 2.952368 LR 0.001000 Time 0.043876 -2022-12-06 10:23:14,610 - Epoch: [0][ 110/ 1200] Overall Loss 2.924516 Objective Loss 2.924516 LR 0.001000 Time 0.041100 -2022-12-06 10:23:14,743 - Epoch: [0][ 120/ 1200] Overall Loss 2.898162 Objective Loss 2.898162 LR 0.001000 Time 0.038784 -2022-12-06 10:23:14,876 - Epoch: [0][ 130/ 1200] Overall Loss 2.873456 Objective Loss 2.873456 LR 0.001000 Time 0.036817 -2022-12-06 10:23:15,011 - Epoch: [0][ 140/ 1200] Overall Loss 2.846835 Objective Loss 2.846835 LR 0.001000 Time 0.035138 -2022-12-06 10:23:15,143 - Epoch: [0][ 150/ 1200] Overall Loss 2.821690 Objective Loss 2.821690 LR 0.001000 Time 0.033675 -2022-12-06 10:23:15,278 - Epoch: [0][ 160/ 1200] Overall Loss 2.797836 Objective Loss 2.797836 LR 0.001000 Time 0.032409 -2022-12-06 10:23:15,412 - Epoch: [0][ 170/ 1200] Overall Loss 2.777086 Objective Loss 2.777086 LR 0.001000 Time 0.031289 -2022-12-06 10:23:15,546 - Epoch: [0][ 180/ 1200] Overall Loss 2.757341 Objective Loss 2.757341 LR 0.001000 Time 0.030293 -2022-12-06 10:23:15,680 - Epoch: [0][ 190/ 1200] Overall Loss 2.735245 Objective Loss 2.735245 LR 0.001000 Time 0.029402 -2022-12-06 10:23:15,814 - Epoch: [0][ 200/ 1200] Overall Loss 2.712491 Objective Loss 2.712491 LR 0.001000 Time 0.028598 -2022-12-06 10:23:15,948 - Epoch: [0][ 210/ 1200] Overall Loss 2.690974 Objective Loss 2.690974 LR 0.001000 Time 0.027871 -2022-12-06 10:23:16,083 - Epoch: [0][ 220/ 1200] Overall Loss 2.670238 Objective Loss 2.670238 LR 0.001000 Time 0.027215 -2022-12-06 10:23:16,216 - Epoch: [0][ 230/ 1200] Overall Loss 2.648976 Objective Loss 2.648976 LR 0.001000 Time 0.026609 -2022-12-06 10:23:16,353 - Epoch: [0][ 240/ 1200] Overall Loss 2.629306 Objective Loss 2.629306 LR 0.001000 Time 0.026060 -2022-12-06 10:23:16,488 - Epoch: [0][ 250/ 1200] Overall Loss 2.609136 Objective Loss 2.609136 LR 0.001000 Time 0.025557 -2022-12-06 10:23:16,623 - Epoch: [0][ 260/ 1200] Overall Loss 2.589631 Objective Loss 2.589631 LR 0.001000 Time 0.025090 -2022-12-06 10:23:16,758 - Epoch: [0][ 270/ 1200] Overall Loss 2.569626 Objective Loss 2.569626 LR 0.001000 Time 0.024660 -2022-12-06 10:23:16,893 - Epoch: [0][ 280/ 1200] Overall Loss 2.550687 Objective Loss 2.550687 LR 0.001000 Time 0.024259 -2022-12-06 10:23:17,028 - Epoch: [0][ 290/ 1200] Overall Loss 2.532224 Objective Loss 2.532224 LR 0.001000 Time 0.023888 -2022-12-06 10:23:17,163 - Epoch: [0][ 300/ 1200] Overall Loss 2.515079 Objective Loss 2.515079 LR 0.001000 Time 0.023538 -2022-12-06 10:23:17,297 - Epoch: [0][ 310/ 1200] Overall Loss 2.497188 Objective Loss 2.497188 LR 0.001000 Time 0.023209 -2022-12-06 10:23:17,433 - Epoch: [0][ 320/ 1200] Overall Loss 2.481312 Objective Loss 2.481312 LR 0.001000 Time 0.022903 -2022-12-06 10:23:17,567 - Epoch: [0][ 330/ 1200] Overall Loss 2.464567 Objective Loss 2.464567 LR 0.001000 Time 0.022612 -2022-12-06 10:23:17,703 - Epoch: [0][ 340/ 1200] Overall Loss 2.448389 Objective Loss 2.448389 LR 0.001000 Time 0.022342 -2022-12-06 10:23:17,837 - Epoch: [0][ 350/ 1200] Overall Loss 2.432008 Objective Loss 2.432008 LR 0.001000 Time 0.022084 -2022-12-06 10:23:17,974 - Epoch: [0][ 360/ 1200] Overall Loss 2.416404 Objective Loss 2.416404 LR 0.001000 Time 0.021845 -2022-12-06 10:23:18,107 - Epoch: [0][ 370/ 1200] Overall Loss 2.400350 Objective Loss 2.400350 LR 0.001000 Time 0.021613 -2022-12-06 10:23:18,244 - Epoch: [0][ 380/ 1200] Overall Loss 2.384526 Objective Loss 2.384526 LR 0.001000 Time 0.021398 -2022-12-06 10:23:18,377 - Epoch: [0][ 390/ 1200] Overall Loss 2.369071 Objective Loss 2.369071 LR 0.001000 Time 0.021190 -2022-12-06 10:23:18,514 - Epoch: [0][ 400/ 1200] Overall Loss 2.355853 Objective Loss 2.355853 LR 0.001000 Time 0.020997 -2022-12-06 10:23:18,647 - Epoch: [0][ 410/ 1200] Overall Loss 2.341331 Objective Loss 2.341331 LR 0.001000 Time 0.020808 -2022-12-06 10:23:18,784 - Epoch: [0][ 420/ 1200] Overall Loss 2.328190 Objective Loss 2.328190 LR 0.001000 Time 0.020634 -2022-12-06 10:23:18,918 - Epoch: [0][ 430/ 1200] Overall Loss 2.314809 Objective Loss 2.314809 LR 0.001000 Time 0.020463 -2022-12-06 10:23:19,054 - Epoch: [0][ 440/ 1200] Overall Loss 2.302555 Objective Loss 2.302555 LR 0.001000 Time 0.020303 -2022-12-06 10:23:19,188 - Epoch: [0][ 450/ 1200] Overall Loss 2.291263 Objective Loss 2.291263 LR 0.001000 Time 0.020147 -2022-12-06 10:23:19,324 - Epoch: [0][ 460/ 1200] Overall Loss 2.277928 Objective Loss 2.277928 LR 0.001000 Time 0.020001 -2022-12-06 10:23:19,458 - Epoch: [0][ 470/ 1200] Overall Loss 2.265401 Objective Loss 2.265401 LR 0.001000 Time 0.019859 -2022-12-06 10:23:19,595 - Epoch: [0][ 480/ 1200] Overall Loss 2.253719 Objective Loss 2.253719 LR 0.001000 Time 0.019725 -2022-12-06 10:23:19,728 - Epoch: [0][ 490/ 1200] Overall Loss 2.242318 Objective Loss 2.242318 LR 0.001000 Time 0.019594 -2022-12-06 10:23:19,865 - Epoch: [0][ 500/ 1200] Overall Loss 2.230448 Objective Loss 2.230448 LR 0.001000 Time 0.019475 -2022-12-06 10:23:19,998 - Epoch: [0][ 510/ 1200] Overall Loss 2.219640 Objective Loss 2.219640 LR 0.001000 Time 0.019353 -2022-12-06 10:23:20,135 - Epoch: [0][ 520/ 1200] Overall Loss 2.209252 Objective Loss 2.209252 LR 0.001000 Time 0.019239 -2022-12-06 10:23:20,269 - Epoch: [0][ 530/ 1200] Overall Loss 2.197978 Objective Loss 2.197978 LR 0.001000 Time 0.019128 -2022-12-06 10:23:20,406 - Epoch: [0][ 540/ 1200] Overall Loss 2.186446 Objective Loss 2.186446 LR 0.001000 Time 0.019023 -2022-12-06 10:23:20,539 - Epoch: [0][ 550/ 1200] Overall Loss 2.175276 Objective Loss 2.175276 LR 0.001000 Time 0.018919 -2022-12-06 10:23:20,676 - Epoch: [0][ 560/ 1200] Overall Loss 2.164629 Objective Loss 2.164629 LR 0.001000 Time 0.018821 -2022-12-06 10:23:20,811 - Epoch: [0][ 570/ 1200] Overall Loss 2.154617 Objective Loss 2.154617 LR 0.001000 Time 0.018728 -2022-12-06 10:23:20,946 - Epoch: [0][ 580/ 1200] Overall Loss 2.144240 Objective Loss 2.144240 LR 0.001000 Time 0.018637 -2022-12-06 10:23:21,080 - Epoch: [0][ 590/ 1200] Overall Loss 2.133819 Objective Loss 2.133819 LR 0.001000 Time 0.018546 -2022-12-06 10:23:21,217 - Epoch: [0][ 600/ 1200] Overall Loss 2.123402 Objective Loss 2.123402 LR 0.001000 Time 0.018464 -2022-12-06 10:23:21,352 - Epoch: [0][ 610/ 1200] Overall Loss 2.113831 Objective Loss 2.113831 LR 0.001000 Time 0.018382 -2022-12-06 10:23:21,487 - Epoch: [0][ 620/ 1200] Overall Loss 2.104111 Objective Loss 2.104111 LR 0.001000 Time 0.018303 -2022-12-06 10:23:21,621 - Epoch: [0][ 630/ 1200] Overall Loss 2.094943 Objective Loss 2.094943 LR 0.001000 Time 0.018224 -2022-12-06 10:23:21,758 - Epoch: [0][ 640/ 1200] Overall Loss 2.085203 Objective Loss 2.085203 LR 0.001000 Time 0.018149 -2022-12-06 10:23:21,891 - Epoch: [0][ 650/ 1200] Overall Loss 2.075811 Objective Loss 2.075811 LR 0.001000 Time 0.018074 -2022-12-06 10:23:22,028 - Epoch: [0][ 660/ 1200] Overall Loss 2.066695 Objective Loss 2.066695 LR 0.001000 Time 0.018004 -2022-12-06 10:23:22,161 - Epoch: [0][ 670/ 1200] Overall Loss 2.057376 Objective Loss 2.057376 LR 0.001000 Time 0.017934 -2022-12-06 10:23:22,298 - Epoch: [0][ 680/ 1200] Overall Loss 2.048061 Objective Loss 2.048061 LR 0.001000 Time 0.017868 -2022-12-06 10:23:22,431 - Epoch: [0][ 690/ 1200] Overall Loss 2.039763 Objective Loss 2.039763 LR 0.001000 Time 0.017802 -2022-12-06 10:23:22,568 - Epoch: [0][ 700/ 1200] Overall Loss 2.030945 Objective Loss 2.030945 LR 0.001000 Time 0.017739 -2022-12-06 10:23:22,702 - Epoch: [0][ 710/ 1200] Overall Loss 2.021659 Objective Loss 2.021659 LR 0.001000 Time 0.017677 -2022-12-06 10:23:22,839 - Epoch: [0][ 720/ 1200] Overall Loss 2.013160 Objective Loss 2.013160 LR 0.001000 Time 0.017618 -2022-12-06 10:23:22,972 - Epoch: [0][ 730/ 1200] Overall Loss 2.004788 Objective Loss 2.004788 LR 0.001000 Time 0.017560 -2022-12-06 10:23:23,109 - Epoch: [0][ 740/ 1200] Overall Loss 1.996688 Objective Loss 1.996688 LR 0.001000 Time 0.017504 -2022-12-06 10:23:23,242 - Epoch: [0][ 750/ 1200] Overall Loss 1.988332 Objective Loss 1.988332 LR 0.001000 Time 0.017448 -2022-12-06 10:23:23,379 - Epoch: [0][ 760/ 1200] Overall Loss 1.981089 Objective Loss 1.981089 LR 0.001000 Time 0.017395 -2022-12-06 10:23:23,511 - Epoch: [0][ 770/ 1200] Overall Loss 1.973393 Objective Loss 1.973393 LR 0.001000 Time 0.017340 -2022-12-06 10:23:23,646 - Epoch: [0][ 780/ 1200] Overall Loss 1.965322 Objective Loss 1.965322 LR 0.001000 Time 0.017288 -2022-12-06 10:23:23,780 - Epoch: [0][ 790/ 1200] Overall Loss 1.957256 Objective Loss 1.957256 LR 0.001000 Time 0.017238 -2022-12-06 10:23:23,914 - Epoch: [0][ 800/ 1200] Overall Loss 1.950230 Objective Loss 1.950230 LR 0.001000 Time 0.017190 -2022-12-06 10:23:24,047 - Epoch: [0][ 810/ 1200] Overall Loss 1.942565 Objective Loss 1.942565 LR 0.001000 Time 0.017142 -2022-12-06 10:23:24,183 - Epoch: [0][ 820/ 1200] Overall Loss 1.935152 Objective Loss 1.935152 LR 0.001000 Time 0.017096 -2022-12-06 10:23:24,316 - Epoch: [0][ 830/ 1200] Overall Loss 1.928047 Objective Loss 1.928047 LR 0.001000 Time 0.017050 -2022-12-06 10:23:24,452 - Epoch: [0][ 840/ 1200] Overall Loss 1.921582 Objective Loss 1.921582 LR 0.001000 Time 0.017006 -2022-12-06 10:23:24,584 - Epoch: [0][ 850/ 1200] Overall Loss 1.915197 Objective Loss 1.915197 LR 0.001000 Time 0.016962 -2022-12-06 10:23:24,720 - Epoch: [0][ 860/ 1200] Overall Loss 1.908192 Objective Loss 1.908192 LR 0.001000 Time 0.016920 -2022-12-06 10:23:24,853 - Epoch: [0][ 870/ 1200] Overall Loss 1.901033 Objective Loss 1.901033 LR 0.001000 Time 0.016878 -2022-12-06 10:23:24,989 - Epoch: [0][ 880/ 1200] Overall Loss 1.894491 Objective Loss 1.894491 LR 0.001000 Time 0.016839 -2022-12-06 10:23:25,122 - Epoch: [0][ 890/ 1200] Overall Loss 1.888162 Objective Loss 1.888162 LR 0.001000 Time 0.016798 -2022-12-06 10:23:25,257 - Epoch: [0][ 900/ 1200] Overall Loss 1.881402 Objective Loss 1.881402 LR 0.001000 Time 0.016760 -2022-12-06 10:23:25,392 - Epoch: [0][ 910/ 1200] Overall Loss 1.874901 Objective Loss 1.874901 LR 0.001000 Time 0.016723 -2022-12-06 10:23:25,526 - Epoch: [0][ 920/ 1200] Overall Loss 1.868045 Objective Loss 1.868045 LR 0.001000 Time 0.016687 -2022-12-06 10:23:25,659 - Epoch: [0][ 930/ 1200] Overall Loss 1.861223 Objective Loss 1.861223 LR 0.001000 Time 0.016650 -2022-12-06 10:23:25,795 - Epoch: [0][ 940/ 1200] Overall Loss 1.854793 Objective Loss 1.854793 LR 0.001000 Time 0.016616 -2022-12-06 10:23:25,929 - Epoch: [0][ 950/ 1200] Overall Loss 1.848308 Objective Loss 1.848308 LR 0.001000 Time 0.016581 -2022-12-06 10:23:26,064 - Epoch: [0][ 960/ 1200] Overall Loss 1.841633 Objective Loss 1.841633 LR 0.001000 Time 0.016547 -2022-12-06 10:23:26,198 - Epoch: [0][ 970/ 1200] Overall Loss 1.835281 Objective Loss 1.835281 LR 0.001000 Time 0.016515 -2022-12-06 10:23:26,333 - Epoch: [0][ 980/ 1200] Overall Loss 1.829096 Objective Loss 1.829096 LR 0.001000 Time 0.016483 -2022-12-06 10:23:26,466 - Epoch: [0][ 990/ 1200] Overall Loss 1.822895 Objective Loss 1.822895 LR 0.001000 Time 0.016450 -2022-12-06 10:23:26,601 - Epoch: [0][ 1000/ 1200] Overall Loss 1.816676 Objective Loss 1.816676 LR 0.001000 Time 0.016420 -2022-12-06 10:23:26,736 - Epoch: [0][ 1010/ 1200] Overall Loss 1.810998 Objective Loss 1.810998 LR 0.001000 Time 0.016390 -2022-12-06 10:23:26,870 - Epoch: [0][ 1020/ 1200] Overall Loss 1.804475 Objective Loss 1.804475 LR 0.001000 Time 0.016360 -2022-12-06 10:23:27,004 - Epoch: [0][ 1030/ 1200] Overall Loss 1.798545 Objective Loss 1.798545 LR 0.001000 Time 0.016331 -2022-12-06 10:23:27,139 - Epoch: [0][ 1040/ 1200] Overall Loss 1.792102 Objective Loss 1.792102 LR 0.001000 Time 0.016303 -2022-12-06 10:23:27,272 - Epoch: [0][ 1050/ 1200] Overall Loss 1.785779 Objective Loss 1.785779 LR 0.001000 Time 0.016274 -2022-12-06 10:23:27,407 - Epoch: [0][ 1060/ 1200] Overall Loss 1.779384 Objective Loss 1.779384 LR 0.001000 Time 0.016246 -2022-12-06 10:23:27,540 - Epoch: [0][ 1070/ 1200] Overall Loss 1.773902 Objective Loss 1.773902 LR 0.001000 Time 0.016219 -2022-12-06 10:23:27,676 - Epoch: [0][ 1080/ 1200] Overall Loss 1.768708 Objective Loss 1.768708 LR 0.001000 Time 0.016192 -2022-12-06 10:23:27,809 - Epoch: [0][ 1090/ 1200] Overall Loss 1.763015 Objective Loss 1.763015 LR 0.001000 Time 0.016165 -2022-12-06 10:23:27,945 - Epoch: [0][ 1100/ 1200] Overall Loss 1.757327 Objective Loss 1.757327 LR 0.001000 Time 0.016140 -2022-12-06 10:23:28,078 - Epoch: [0][ 1110/ 1200] Overall Loss 1.751611 Objective Loss 1.751611 LR 0.001000 Time 0.016114 -2022-12-06 10:23:28,213 - Epoch: [0][ 1120/ 1200] Overall Loss 1.746622 Objective Loss 1.746622 LR 0.001000 Time 0.016090 -2022-12-06 10:23:28,348 - Epoch: [0][ 1130/ 1200] Overall Loss 1.740810 Objective Loss 1.740810 LR 0.001000 Time 0.016066 -2022-12-06 10:23:28,619 - Epoch: [0][ 1140/ 1200] Overall Loss 1.735263 Objective Loss 1.735263 LR 0.001000 Time 0.016084 -2022-12-06 10:23:28,758 - Epoch: [0][ 1150/ 1200] Overall Loss 1.729960 Objective Loss 1.729960 LR 0.001000 Time 0.016065 -2022-12-06 10:23:28,899 - Epoch: [0][ 1160/ 1200] Overall Loss 1.724784 Objective Loss 1.724784 LR 0.001000 Time 0.016046 -2022-12-06 10:23:29,039 - Epoch: [0][ 1170/ 1200] Overall Loss 1.719753 Objective Loss 1.719753 LR 0.001000 Time 0.016028 -2022-12-06 10:23:29,176 - Epoch: [0][ 1180/ 1200] Overall Loss 1.714554 Objective Loss 1.714554 LR 0.001000 Time 0.016008 -2022-12-06 10:23:29,311 - Epoch: [0][ 1190/ 1200] Overall Loss 1.710000 Objective Loss 1.710000 LR 0.001000 Time 0.015986 -2022-12-06 10:23:29,848 - Epoch: [0][ 1200/ 1200] Overall Loss 1.705069 Objective Loss 1.705069 Top1 52.719665 Top5 88.702929 LR 0.001000 Time 0.016300 -2022-12-06 10:23:29,946 - --- validate (epoch=0)----------- -2022-12-06 10:23:29,946 - 34129 samples (256 per mini-batch) -2022-12-06 10:23:30,337 - Epoch: [0][ 10/ 134] Loss 1.090361 Top1 57.656250 Top5 89.960938 -2022-12-06 10:23:30,430 - Epoch: [0][ 20/ 134] Loss 1.076431 Top1 56.835938 Top5 89.667969 -2022-12-06 10:23:30,519 - Epoch: [0][ 30/ 134] Loss 1.087722 Top1 56.614583 Top5 89.739583 -2022-12-06 10:23:30,616 - Epoch: [0][ 40/ 134] Loss 1.083257 Top1 56.708984 Top5 89.902344 -2022-12-06 10:23:30,703 - Epoch: [0][ 50/ 134] Loss 1.088337 Top1 56.750000 Top5 89.601562 -2022-12-06 10:23:30,791 - Epoch: [0][ 60/ 134] Loss 1.077873 Top1 56.907552 Top5 89.648438 -2022-12-06 10:23:30,875 - Epoch: [0][ 70/ 134] Loss 1.084868 Top1 56.623884 Top5 89.620536 -2022-12-06 10:23:30,966 - Epoch: [0][ 80/ 134] Loss 1.081076 Top1 56.708984 Top5 89.653320 -2022-12-06 10:23:31,051 - Epoch: [0][ 90/ 134] Loss 1.084737 Top1 56.601562 Top5 89.570312 -2022-12-06 10:23:31,142 - Epoch: [0][ 100/ 134] Loss 1.083871 Top1 56.507812 Top5 89.480469 -2022-12-06 10:23:31,234 - Epoch: [0][ 110/ 134] Loss 1.086376 Top1 56.498580 Top5 89.460227 -2022-12-06 10:23:31,323 - Epoch: [0][ 120/ 134] Loss 1.080633 Top1 56.634115 Top5 89.472656 -2022-12-06 10:23:31,409 - Epoch: [0][ 130/ 134] Loss 1.080394 Top1 56.556490 Top5 89.516226 -2022-12-06 10:23:31,773 - Epoch: [0][ 134/ 134] Loss 1.077132 Top1 56.620469 Top5 89.513317 -2022-12-06 10:23:31,992 - ==> Top1: 56.620 Top5: 89.513 Loss: 1.077 - -2022-12-06 10:23:32,170 - ==> Confusion: -[[ 600 8 8 0 32 1 0 2 18 265 0 0 2 18 17 1 5 5 4 3 7] - [ 0 733 20 2 34 36 1 23 8 5 36 2 17 9 25 2 12 0 38 7 17] - [ 18 11 819 2 26 8 57 49 1 9 9 2 6 11 0 11 3 3 5 39 14] - [ 8 14 78 559 7 24 1 16 11 3 115 1 9 10 48 2 8 16 66 3 21] - [ 16 21 16 0 855 12 0 0 1 19 0 2 1 17 10 14 15 1 3 3 14] - [ 6 163 63 10 22 492 2 44 10 5 22 12 19 121 10 1 7 1 14 40 5] - [ 2 5 249 3 5 6 737 26 1 0 4 9 2 3 0 9 1 5 4 39 8] - [ 0 25 73 7 8 48 1 666 2 6 16 8 8 9 5 0 2 0 98 64 8] - [ 24 30 3 0 5 0 0 1 752 84 8 1 13 18 106 0 3 2 5 0 9] - [ 201 3 14 0 35 0 0 1 34 638 0 0 2 37 16 1 1 0 0 0 18] - [ 3 65 85 18 9 20 2 26 20 5 612 0 3 9 21 0 6 0 104 7 4] - [ 0 2 1 2 1 37 5 4 0 0 0 631 119 60 2 8 14 35 5 116 9] - [ 0 1 1 5 1 10 1 2 8 0 3 104 650 11 14 4 6 118 6 15 9] - [ 22 18 17 0 17 54 0 1 19 52 1 14 10 743 15 1 6 3 2 19 9] - [ 29 79 3 18 53 7 0 1 37 22 5 0 6 9 834 0 5 3 6 0 13] - [ 10 3 24 3 42 15 7 0 1 0 0 45 10 14 3 759 14 73 0 7 13] - [ 5 75 6 3 110 31 0 0 12 2 3 26 22 21 34 17 671 3 3 8 20] - [ 8 3 1 8 1 4 3 1 4 1 0 30 167 11 18 7 3 750 0 7 9] - [ 1 57 12 13 3 6 1 76 8 3 79 5 10 1 20 0 4 1 692 10 6] - [ 0 11 19 0 4 9 7 18 1 0 0 54 13 10 1 2 2 3 1 913 12] - [ 290 837 533 148 591 424 64 281 179 246 278 253 521 609 698 123 237 239 541 916 5218]] - -2022-12-06 10:23:35,281 - ==> Best [Top1: 56.620 Top5: 89.513 Sparsity:0.00 Params: 5376 on epoch: 0] -2022-12-06 10:23:35,282 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:23:35,330 - - -2022-12-06 10:23:35,331 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:23:36,205 - Epoch: [1][ 10/ 1200] Overall Loss 1.017178 Objective Loss 1.017178 LR 0.001000 Time 0.087316 -2022-12-06 10:23:36,351 - Epoch: [1][ 20/ 1200] Overall Loss 1.039702 Objective Loss 1.039702 LR 0.001000 Time 0.050932 -2022-12-06 10:23:36,489 - Epoch: [1][ 30/ 1200] Overall Loss 1.031120 Objective Loss 1.031120 LR 0.001000 Time 0.038508 -2022-12-06 10:23:36,625 - Epoch: [1][ 40/ 1200] Overall Loss 1.025267 Objective Loss 1.025267 LR 0.001000 Time 0.032274 -2022-12-06 10:23:36,761 - Epoch: [1][ 50/ 1200] Overall Loss 1.020518 Objective Loss 1.020518 LR 0.001000 Time 0.028538 -2022-12-06 10:23:36,897 - Epoch: [1][ 60/ 1200] Overall Loss 1.025644 Objective Loss 1.025644 LR 0.001000 Time 0.026037 -2022-12-06 10:23:37,031 - Epoch: [1][ 70/ 1200] Overall Loss 1.026288 Objective Loss 1.026288 LR 0.001000 Time 0.024228 -2022-12-06 10:23:37,165 - Epoch: [1][ 80/ 1200] Overall Loss 1.028399 Objective Loss 1.028399 LR 0.001000 Time 0.022871 -2022-12-06 10:23:37,301 - Epoch: [1][ 90/ 1200] Overall Loss 1.024401 Objective Loss 1.024401 LR 0.001000 Time 0.021836 -2022-12-06 10:23:37,439 - Epoch: [1][ 100/ 1200] Overall Loss 1.025369 Objective Loss 1.025369 LR 0.001000 Time 0.021021 -2022-12-06 10:23:37,576 - Epoch: [1][ 110/ 1200] Overall Loss 1.028851 Objective Loss 1.028851 LR 0.001000 Time 0.020357 -2022-12-06 10:23:37,713 - Epoch: [1][ 120/ 1200] Overall Loss 1.030061 Objective Loss 1.030061 LR 0.001000 Time 0.019797 -2022-12-06 10:23:37,850 - Epoch: [1][ 130/ 1200] Overall Loss 1.029831 Objective Loss 1.029831 LR 0.001000 Time 0.019327 -2022-12-06 10:23:37,987 - Epoch: [1][ 140/ 1200] Overall Loss 1.030600 Objective Loss 1.030600 LR 0.001000 Time 0.018922 -2022-12-06 10:23:38,125 - Epoch: [1][ 150/ 1200] Overall Loss 1.031781 Objective Loss 1.031781 LR 0.001000 Time 0.018572 -2022-12-06 10:23:38,262 - Epoch: [1][ 160/ 1200] Overall Loss 1.027710 Objective Loss 1.027710 LR 0.001000 Time 0.018268 -2022-12-06 10:23:38,400 - Epoch: [1][ 170/ 1200] Overall Loss 1.027969 Objective Loss 1.027969 LR 0.001000 Time 0.017999 -2022-12-06 10:23:38,537 - Epoch: [1][ 180/ 1200] Overall Loss 1.026042 Objective Loss 1.026042 LR 0.001000 Time 0.017761 -2022-12-06 10:23:38,675 - Epoch: [1][ 190/ 1200] Overall Loss 1.026111 Objective Loss 1.026111 LR 0.001000 Time 0.017551 -2022-12-06 10:23:38,812 - Epoch: [1][ 200/ 1200] Overall Loss 1.024130 Objective Loss 1.024130 LR 0.001000 Time 0.017355 -2022-12-06 10:23:38,950 - Epoch: [1][ 210/ 1200] Overall Loss 1.022185 Objective Loss 1.022185 LR 0.001000 Time 0.017181 -2022-12-06 10:23:39,087 - Epoch: [1][ 220/ 1200] Overall Loss 1.020500 Objective Loss 1.020500 LR 0.001000 Time 0.017022 -2022-12-06 10:23:39,224 - Epoch: [1][ 230/ 1200] Overall Loss 1.018916 Objective Loss 1.018916 LR 0.001000 Time 0.016875 -2022-12-06 10:23:39,360 - Epoch: [1][ 240/ 1200] Overall Loss 1.018961 Objective Loss 1.018961 LR 0.001000 Time 0.016737 -2022-12-06 10:23:39,498 - Epoch: [1][ 250/ 1200] Overall Loss 1.020782 Objective Loss 1.020782 LR 0.001000 Time 0.016616 -2022-12-06 10:23:39,635 - Epoch: [1][ 260/ 1200] Overall Loss 1.019195 Objective Loss 1.019195 LR 0.001000 Time 0.016503 -2022-12-06 10:23:39,772 - Epoch: [1][ 270/ 1200] Overall Loss 1.018989 Objective Loss 1.018989 LR 0.001000 Time 0.016399 -2022-12-06 10:23:39,909 - Epoch: [1][ 280/ 1200] Overall Loss 1.016756 Objective Loss 1.016756 LR 0.001000 Time 0.016301 -2022-12-06 10:23:40,046 - Epoch: [1][ 290/ 1200] Overall Loss 1.015611 Objective Loss 1.015611 LR 0.001000 Time 0.016209 -2022-12-06 10:23:40,183 - Epoch: [1][ 300/ 1200] Overall Loss 1.012645 Objective Loss 1.012645 LR 0.001000 Time 0.016124 -2022-12-06 10:23:40,321 - Epoch: [1][ 310/ 1200] Overall Loss 1.011103 Objective Loss 1.011103 LR 0.001000 Time 0.016047 -2022-12-06 10:23:40,458 - Epoch: [1][ 320/ 1200] Overall Loss 1.008867 Objective Loss 1.008867 LR 0.001000 Time 0.015973 -2022-12-06 10:23:40,595 - Epoch: [1][ 330/ 1200] Overall Loss 1.007577 Objective Loss 1.007577 LR 0.001000 Time 0.015903 -2022-12-06 10:23:40,733 - Epoch: [1][ 340/ 1200] Overall Loss 1.006834 Objective Loss 1.006834 LR 0.001000 Time 0.015838 -2022-12-06 10:23:40,870 - Epoch: [1][ 350/ 1200] Overall Loss 1.004084 Objective Loss 1.004084 LR 0.001000 Time 0.015776 -2022-12-06 10:23:41,007 - Epoch: [1][ 360/ 1200] Overall Loss 1.002173 Objective Loss 1.002173 LR 0.001000 Time 0.015717 -2022-12-06 10:23:41,144 - Epoch: [1][ 370/ 1200] Overall Loss 1.001414 Objective Loss 1.001414 LR 0.001000 Time 0.015662 -2022-12-06 10:23:41,281 - Epoch: [1][ 380/ 1200] Overall Loss 0.998984 Objective Loss 0.998984 LR 0.001000 Time 0.015610 -2022-12-06 10:23:41,419 - Epoch: [1][ 390/ 1200] Overall Loss 0.998538 Objective Loss 0.998538 LR 0.001000 Time 0.015562 -2022-12-06 10:23:41,556 - Epoch: [1][ 400/ 1200] Overall Loss 0.997530 Objective Loss 0.997530 LR 0.001000 Time 0.015515 -2022-12-06 10:23:41,694 - Epoch: [1][ 410/ 1200] Overall Loss 0.995399 Objective Loss 0.995399 LR 0.001000 Time 0.015471 -2022-12-06 10:23:41,831 - Epoch: [1][ 420/ 1200] Overall Loss 0.994832 Objective Loss 0.994832 LR 0.001000 Time 0.015429 -2022-12-06 10:23:41,970 - Epoch: [1][ 430/ 1200] Overall Loss 0.992581 Objective Loss 0.992581 LR 0.001000 Time 0.015391 -2022-12-06 10:23:42,107 - Epoch: [1][ 440/ 1200] Overall Loss 0.990699 Objective Loss 0.990699 LR 0.001000 Time 0.015352 -2022-12-06 10:23:42,245 - Epoch: [1][ 450/ 1200] Overall Loss 0.989725 Objective Loss 0.989725 LR 0.001000 Time 0.015316 -2022-12-06 10:23:42,383 - Epoch: [1][ 460/ 1200] Overall Loss 0.989078 Objective Loss 0.989078 LR 0.001000 Time 0.015281 -2022-12-06 10:23:42,519 - Epoch: [1][ 470/ 1200] Overall Loss 0.988577 Objective Loss 0.988577 LR 0.001000 Time 0.015246 -2022-12-06 10:23:42,654 - Epoch: [1][ 480/ 1200] Overall Loss 0.987746 Objective Loss 0.987746 LR 0.001000 Time 0.015207 -2022-12-06 10:23:42,789 - Epoch: [1][ 490/ 1200] Overall Loss 0.986643 Objective Loss 0.986643 LR 0.001000 Time 0.015172 -2022-12-06 10:23:42,923 - Epoch: [1][ 500/ 1200] Overall Loss 0.984439 Objective Loss 0.984439 LR 0.001000 Time 0.015135 -2022-12-06 10:23:43,059 - Epoch: [1][ 510/ 1200] Overall Loss 0.983601 Objective Loss 0.983601 LR 0.001000 Time 0.015104 -2022-12-06 10:23:43,194 - Epoch: [1][ 520/ 1200] Overall Loss 0.981873 Objective Loss 0.981873 LR 0.001000 Time 0.015072 -2022-12-06 10:23:43,329 - Epoch: [1][ 530/ 1200] Overall Loss 0.979255 Objective Loss 0.979255 LR 0.001000 Time 0.015042 -2022-12-06 10:23:43,464 - Epoch: [1][ 540/ 1200] Overall Loss 0.977785 Objective Loss 0.977785 LR 0.001000 Time 0.015013 -2022-12-06 10:23:43,707 - Epoch: [1][ 550/ 1200] Overall Loss 0.976411 Objective Loss 0.976411 LR 0.001000 Time 0.014999 -2022-12-06 10:23:43,850 - Epoch: [1][ 560/ 1200] Overall Loss 0.975401 Objective Loss 0.975401 LR 0.001000 Time 0.014985 -2022-12-06 10:23:43,992 - Epoch: [1][ 570/ 1200] Overall Loss 0.974128 Objective Loss 0.974128 LR 0.001000 Time 0.014971 -2022-12-06 10:23:44,134 - Epoch: [1][ 580/ 1200] Overall Loss 0.974230 Objective Loss 0.974230 LR 0.001000 Time 0.014957 -2022-12-06 10:23:44,277 - Epoch: [1][ 590/ 1200] Overall Loss 0.972603 Objective Loss 0.972603 LR 0.001000 Time 0.014945 -2022-12-06 10:23:44,420 - Epoch: [1][ 600/ 1200] Overall Loss 0.971695 Objective Loss 0.971695 LR 0.001000 Time 0.014934 -2022-12-06 10:23:44,563 - Epoch: [1][ 610/ 1200] Overall Loss 0.971112 Objective Loss 0.971112 LR 0.001000 Time 0.014922 -2022-12-06 10:23:44,706 - Epoch: [1][ 620/ 1200] Overall Loss 0.969745 Objective Loss 0.969745 LR 0.001000 Time 0.014910 -2022-12-06 10:23:44,849 - Epoch: [1][ 630/ 1200] Overall Loss 0.968408 Objective Loss 0.968408 LR 0.001000 Time 0.014900 -2022-12-06 10:23:44,991 - Epoch: [1][ 640/ 1200] Overall Loss 0.966876 Objective Loss 0.966876 LR 0.001000 Time 0.014889 -2022-12-06 10:23:45,134 - Epoch: [1][ 650/ 1200] Overall Loss 0.965588 Objective Loss 0.965588 LR 0.001000 Time 0.014879 -2022-12-06 10:23:45,277 - Epoch: [1][ 660/ 1200] Overall Loss 0.964347 Objective Loss 0.964347 LR 0.001000 Time 0.014869 -2022-12-06 10:23:45,420 - Epoch: [1][ 670/ 1200] Overall Loss 0.963506 Objective Loss 0.963506 LR 0.001000 Time 0.014860 -2022-12-06 10:23:45,562 - Epoch: [1][ 680/ 1200] Overall Loss 0.962616 Objective Loss 0.962616 LR 0.001000 Time 0.014850 -2022-12-06 10:23:45,705 - Epoch: [1][ 690/ 1200] Overall Loss 0.961175 Objective Loss 0.961175 LR 0.001000 Time 0.014841 -2022-12-06 10:23:45,848 - Epoch: [1][ 700/ 1200] Overall Loss 0.959809 Objective Loss 0.959809 LR 0.001000 Time 0.014832 -2022-12-06 10:23:45,990 - Epoch: [1][ 710/ 1200] Overall Loss 0.957868 Objective Loss 0.957868 LR 0.001000 Time 0.014823 -2022-12-06 10:23:46,134 - Epoch: [1][ 720/ 1200] Overall Loss 0.956720 Objective Loss 0.956720 LR 0.001000 Time 0.014815 -2022-12-06 10:23:46,276 - Epoch: [1][ 730/ 1200] Overall Loss 0.955172 Objective Loss 0.955172 LR 0.001000 Time 0.014807 -2022-12-06 10:23:46,421 - Epoch: [1][ 740/ 1200] Overall Loss 0.954434 Objective Loss 0.954434 LR 0.001000 Time 0.014801 -2022-12-06 10:23:46,563 - Epoch: [1][ 750/ 1200] Overall Loss 0.953573 Objective Loss 0.953573 LR 0.001000 Time 0.014793 -2022-12-06 10:23:46,705 - Epoch: [1][ 760/ 1200] Overall Loss 0.952650 Objective Loss 0.952650 LR 0.001000 Time 0.014785 -2022-12-06 10:23:46,847 - Epoch: [1][ 770/ 1200] Overall Loss 0.951153 Objective Loss 0.951153 LR 0.001000 Time 0.014777 -2022-12-06 10:23:46,990 - Epoch: [1][ 780/ 1200] Overall Loss 0.949971 Objective Loss 0.949971 LR 0.001000 Time 0.014769 -2022-12-06 10:23:47,131 - Epoch: [1][ 790/ 1200] Overall Loss 0.949100 Objective Loss 0.949100 LR 0.001000 Time 0.014761 -2022-12-06 10:23:47,274 - Epoch: [1][ 800/ 1200] Overall Loss 0.947712 Objective Loss 0.947712 LR 0.001000 Time 0.014754 -2022-12-06 10:23:47,415 - Epoch: [1][ 810/ 1200] Overall Loss 0.946473 Objective Loss 0.946473 LR 0.001000 Time 0.014746 -2022-12-06 10:23:47,557 - Epoch: [1][ 820/ 1200] Overall Loss 0.945405 Objective Loss 0.945405 LR 0.001000 Time 0.014738 -2022-12-06 10:23:47,698 - Epoch: [1][ 830/ 1200] Overall Loss 0.943859 Objective Loss 0.943859 LR 0.001000 Time 0.014730 -2022-12-06 10:23:47,840 - Epoch: [1][ 840/ 1200] Overall Loss 0.942327 Objective Loss 0.942327 LR 0.001000 Time 0.014722 -2022-12-06 10:23:47,980 - Epoch: [1][ 850/ 1200] Overall Loss 0.940892 Objective Loss 0.940892 LR 0.001000 Time 0.014714 -2022-12-06 10:23:48,121 - Epoch: [1][ 860/ 1200] Overall Loss 0.939702 Objective Loss 0.939702 LR 0.001000 Time 0.014706 -2022-12-06 10:23:48,262 - Epoch: [1][ 870/ 1200] Overall Loss 0.938373 Objective Loss 0.938373 LR 0.001000 Time 0.014699 -2022-12-06 10:23:48,404 - Epoch: [1][ 880/ 1200] Overall Loss 0.937009 Objective Loss 0.937009 LR 0.001000 Time 0.014691 -2022-12-06 10:23:48,545 - Epoch: [1][ 890/ 1200] Overall Loss 0.935680 Objective Loss 0.935680 LR 0.001000 Time 0.014685 -2022-12-06 10:23:48,699 - Epoch: [1][ 900/ 1200] Overall Loss 0.933973 Objective Loss 0.933973 LR 0.001000 Time 0.014678 -2022-12-06 10:23:48,840 - Epoch: [1][ 910/ 1200] Overall Loss 0.933016 Objective Loss 0.933016 LR 0.001000 Time 0.014671 -2022-12-06 10:23:48,981 - Epoch: [1][ 920/ 1200] Overall Loss 0.931403 Objective Loss 0.931403 LR 0.001000 Time 0.014664 -2022-12-06 10:23:49,123 - Epoch: [1][ 930/ 1200] Overall Loss 0.930185 Objective Loss 0.930185 LR 0.001000 Time 0.014659 -2022-12-06 10:23:49,264 - Epoch: [1][ 940/ 1200] Overall Loss 0.928448 Objective Loss 0.928448 LR 0.001000 Time 0.014652 -2022-12-06 10:23:49,406 - Epoch: [1][ 950/ 1200] Overall Loss 0.927243 Objective Loss 0.927243 LR 0.001000 Time 0.014647 -2022-12-06 10:23:49,548 - Epoch: [1][ 960/ 1200] Overall Loss 0.926071 Objective Loss 0.926071 LR 0.001000 Time 0.014641 -2022-12-06 10:23:49,688 - Epoch: [1][ 970/ 1200] Overall Loss 0.924978 Objective Loss 0.924978 LR 0.001000 Time 0.014635 -2022-12-06 10:23:49,830 - Epoch: [1][ 980/ 1200] Overall Loss 0.924152 Objective Loss 0.924152 LR 0.001000 Time 0.014629 -2022-12-06 10:23:49,971 - Epoch: [1][ 990/ 1200] Overall Loss 0.922882 Objective Loss 0.922882 LR 0.001000 Time 0.014623 -2022-12-06 10:23:50,113 - Epoch: [1][ 1000/ 1200] Overall Loss 0.921974 Objective Loss 0.921974 LR 0.001000 Time 0.014619 -2022-12-06 10:23:50,252 - Epoch: [1][ 1010/ 1200] Overall Loss 0.921249 Objective Loss 0.921249 LR 0.001000 Time 0.014611 -2022-12-06 10:23:50,387 - Epoch: [1][ 1020/ 1200] Overall Loss 0.920243 Objective Loss 0.920243 LR 0.001000 Time 0.014600 -2022-12-06 10:23:50,525 - Epoch: [1][ 1030/ 1200] Overall Loss 0.919140 Objective Loss 0.919140 LR 0.001000 Time 0.014591 -2022-12-06 10:23:50,660 - Epoch: [1][ 1040/ 1200] Overall Loss 0.918039 Objective Loss 0.918039 LR 0.001000 Time 0.014581 -2022-12-06 10:23:50,797 - Epoch: [1][ 1050/ 1200] Overall Loss 0.916677 Objective Loss 0.916677 LR 0.001000 Time 0.014572 -2022-12-06 10:23:50,936 - Epoch: [1][ 1060/ 1200] Overall Loss 0.915409 Objective Loss 0.915409 LR 0.001000 Time 0.014565 -2022-12-06 10:23:51,074 - Epoch: [1][ 1070/ 1200] Overall Loss 0.914316 Objective Loss 0.914316 LR 0.001000 Time 0.014558 -2022-12-06 10:23:51,211 - Epoch: [1][ 1080/ 1200] Overall Loss 0.912943 Objective Loss 0.912943 LR 0.001000 Time 0.014549 -2022-12-06 10:23:51,348 - Epoch: [1][ 1090/ 1200] Overall Loss 0.912168 Objective Loss 0.912168 LR 0.001000 Time 0.014541 -2022-12-06 10:23:51,486 - Epoch: [1][ 1100/ 1200] Overall Loss 0.911208 Objective Loss 0.911208 LR 0.001000 Time 0.014534 -2022-12-06 10:23:51,624 - Epoch: [1][ 1110/ 1200] Overall Loss 0.910468 Objective Loss 0.910468 LR 0.001000 Time 0.014526 -2022-12-06 10:23:51,760 - Epoch: [1][ 1120/ 1200] Overall Loss 0.909720 Objective Loss 0.909720 LR 0.001000 Time 0.014518 -2022-12-06 10:23:51,899 - Epoch: [1][ 1130/ 1200] Overall Loss 0.908955 Objective Loss 0.908955 LR 0.001000 Time 0.014512 -2022-12-06 10:23:52,035 - Epoch: [1][ 1140/ 1200] Overall Loss 0.907981 Objective Loss 0.907981 LR 0.001000 Time 0.014504 -2022-12-06 10:23:52,173 - Epoch: [1][ 1150/ 1200] Overall Loss 0.906924 Objective Loss 0.906924 LR 0.001000 Time 0.014497 -2022-12-06 10:23:52,309 - Epoch: [1][ 1160/ 1200] Overall Loss 0.905732 Objective Loss 0.905732 LR 0.001000 Time 0.014489 -2022-12-06 10:23:52,447 - Epoch: [1][ 1170/ 1200] Overall Loss 0.904577 Objective Loss 0.904577 LR 0.001000 Time 0.014482 -2022-12-06 10:23:52,583 - Epoch: [1][ 1180/ 1200] Overall Loss 0.903251 Objective Loss 0.903251 LR 0.001000 Time 0.014474 -2022-12-06 10:23:52,720 - Epoch: [1][ 1190/ 1200] Overall Loss 0.901996 Objective Loss 0.901996 LR 0.001000 Time 0.014468 -2022-12-06 10:23:52,906 - Epoch: [1][ 1200/ 1200] Overall Loss 0.901044 Objective Loss 0.901044 Top1 61.715481 Top5 92.259414 LR 0.001000 Time 0.014502 -2022-12-06 10:23:52,993 - --- validate (epoch=1)----------- -2022-12-06 10:23:52,993 - 34129 samples (256 per mini-batch) -2022-12-06 10:23:53,395 - Epoch: [1][ 10/ 134] Loss 0.711485 Top1 65.078125 Top5 93.085938 -2022-12-06 10:23:53,500 - Epoch: [1][ 20/ 134] Loss 0.722656 Top1 65.117188 Top5 93.261719 -2022-12-06 10:23:53,608 - Epoch: [1][ 30/ 134] Loss 0.712870 Top1 65.039062 Top5 93.072917 -2022-12-06 10:23:53,714 - Epoch: [1][ 40/ 134] Loss 0.721404 Top1 64.824219 Top5 92.910156 -2022-12-06 10:23:53,814 - Epoch: [1][ 50/ 134] Loss 0.733524 Top1 64.562500 Top5 92.828125 -2022-12-06 10:23:53,919 - Epoch: [1][ 60/ 134] Loss 0.730856 Top1 64.837240 Top5 92.779948 -2022-12-06 10:23:54,019 - Epoch: [1][ 70/ 134] Loss 0.735242 Top1 64.637277 Top5 92.767857 -2022-12-06 10:23:54,124 - Epoch: [1][ 80/ 134] Loss 0.735520 Top1 64.804688 Top5 92.900391 -2022-12-06 10:23:54,225 - Epoch: [1][ 90/ 134] Loss 0.734901 Top1 64.904514 Top5 92.968750 -2022-12-06 10:23:54,329 - Epoch: [1][ 100/ 134] Loss 0.729056 Top1 65.160156 Top5 93.046875 -2022-12-06 10:23:54,429 - Epoch: [1][ 110/ 134] Loss 0.727863 Top1 65.042614 Top5 93.053977 -2022-12-06 10:23:54,532 - Epoch: [1][ 120/ 134] Loss 0.726684 Top1 65.166016 Top5 93.043620 -2022-12-06 10:23:54,625 - Epoch: [1][ 130/ 134] Loss 0.725995 Top1 65.228365 Top5 93.043870 -2022-12-06 10:23:54,646 - Epoch: [1][ 134/ 134] Loss 0.726494 Top1 65.228984 Top5 93.049899 -2022-12-06 10:23:54,732 - ==> Top1: 65.229 Top5: 93.050 Loss: 0.726 - -2022-12-06 10:23:54,732 - ==> Confusion: -[[ 755 1 5 0 10 5 0 5 15 149 0 2 1 18 11 1 3 3 3 2 7] - [ 1 773 5 4 17 57 5 28 4 1 31 6 6 6 23 1 20 0 26 2 11] - [ 19 4 846 27 11 11 53 41 1 3 29 6 5 19 2 5 2 1 8 2 8] - [ 6 5 34 763 2 7 1 5 10 0 89 2 9 9 34 3 6 6 25 1 3] - [ 18 13 8 2 856 23 0 1 0 14 2 1 2 23 11 9 22 2 2 2 9] - [ 2 74 9 5 10 722 13 49 11 2 18 22 5 84 8 0 7 2 12 7 7] - [ 1 9 79 11 2 8 945 23 0 0 7 5 4 1 0 4 1 1 5 10 2] - [ 1 19 21 3 2 59 10 804 0 1 18 11 2 7 1 0 0 0 68 18 9] - [ 18 7 0 1 0 8 0 2 882 53 12 2 7 34 28 0 1 0 8 0 1] - [ 141 4 2 0 3 8 0 4 55 711 1 0 0 56 7 0 1 0 1 0 7] - [ 2 11 13 31 4 22 1 8 23 2 846 2 2 15 7 0 1 1 24 3 1] - [ 3 1 2 0 1 24 3 7 2 0 1 815 87 33 0 11 16 7 5 27 6] - [ 1 2 1 14 4 5 0 4 7 0 1 80 741 8 11 15 10 45 6 8 6] - [ 9 5 5 1 8 63 0 2 13 12 3 13 10 853 3 3 8 1 1 4 6] - [ 21 5 0 19 11 9 0 1 60 13 8 1 6 13 937 1 6 2 12 0 5] - [ 5 5 8 3 12 5 18 2 1 0 0 15 8 5 2 892 22 27 0 3 10] - [ 6 12 4 1 28 10 1 3 4 0 3 6 2 16 5 16 934 2 4 7 8] - [ 5 0 1 9 0 2 6 1 5 0 2 22 138 6 6 25 6 790 3 2 7] - [ 0 11 9 21 0 9 2 50 9 0 65 4 9 1 17 0 1 0 794 4 2] - [ 0 8 4 1 1 17 18 34 0 0 5 39 12 17 0 4 9 5 5 893 8] - [ 349 374 360 293 271 510 171 295 204 206 561 212 524 724 500 269 499 112 432 650 5710]] - -2022-12-06 10:23:55,417 - ==> Best [Top1: 65.229 Top5: 93.050 Sparsity:0.00 Params: 5376 on epoch: 1] -2022-12-06 10:23:55,417 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:23:55,424 - - -2022-12-06 10:23:55,424 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:23:56,271 - Epoch: [2][ 10/ 1200] Overall Loss 0.709675 Objective Loss 0.709675 LR 0.001000 Time 0.084603 -2022-12-06 10:23:56,417 - Epoch: [2][ 20/ 1200] Overall Loss 0.719149 Objective Loss 0.719149 LR 0.001000 Time 0.049599 -2022-12-06 10:23:56,552 - Epoch: [2][ 30/ 1200] Overall Loss 0.731214 Objective Loss 0.731214 LR 0.001000 Time 0.037531 -2022-12-06 10:23:56,686 - Epoch: [2][ 40/ 1200] Overall Loss 0.721774 Objective Loss 0.721774 LR 0.001000 Time 0.031495 -2022-12-06 10:23:56,820 - Epoch: [2][ 50/ 1200] Overall Loss 0.708380 Objective Loss 0.708380 LR 0.001000 Time 0.027860 -2022-12-06 10:23:56,952 - Epoch: [2][ 60/ 1200] Overall Loss 0.715947 Objective Loss 0.715947 LR 0.001000 Time 0.025418 -2022-12-06 10:23:57,087 - Epoch: [2][ 70/ 1200] Overall Loss 0.725251 Objective Loss 0.725251 LR 0.001000 Time 0.023701 -2022-12-06 10:23:57,221 - Epoch: [2][ 80/ 1200] Overall Loss 0.729301 Objective Loss 0.729301 LR 0.001000 Time 0.022405 -2022-12-06 10:23:57,354 - Epoch: [2][ 90/ 1200] Overall Loss 0.729321 Objective Loss 0.729321 LR 0.001000 Time 0.021388 -2022-12-06 10:23:57,488 - Epoch: [2][ 100/ 1200] Overall Loss 0.731972 Objective Loss 0.731972 LR 0.001000 Time 0.020585 -2022-12-06 10:23:57,621 - Epoch: [2][ 110/ 1200] Overall Loss 0.734960 Objective Loss 0.734960 LR 0.001000 Time 0.019925 -2022-12-06 10:23:57,755 - Epoch: [2][ 120/ 1200] Overall Loss 0.736894 Objective Loss 0.736894 LR 0.001000 Time 0.019375 -2022-12-06 10:23:57,888 - Epoch: [2][ 130/ 1200] Overall Loss 0.737987 Objective Loss 0.737987 LR 0.001000 Time 0.018902 -2022-12-06 10:23:58,022 - Epoch: [2][ 140/ 1200] Overall Loss 0.737893 Objective Loss 0.737893 LR 0.001000 Time 0.018504 -2022-12-06 10:23:58,155 - Epoch: [2][ 150/ 1200] Overall Loss 0.742084 Objective Loss 0.742084 LR 0.001000 Time 0.018156 -2022-12-06 10:23:58,288 - Epoch: [2][ 160/ 1200] Overall Loss 0.738949 Objective Loss 0.738949 LR 0.001000 Time 0.017852 -2022-12-06 10:23:58,421 - Epoch: [2][ 170/ 1200] Overall Loss 0.738005 Objective Loss 0.738005 LR 0.001000 Time 0.017582 -2022-12-06 10:23:58,557 - Epoch: [2][ 180/ 1200] Overall Loss 0.737381 Objective Loss 0.737381 LR 0.001000 Time 0.017355 -2022-12-06 10:23:58,691 - Epoch: [2][ 190/ 1200] Overall Loss 0.736766 Objective Loss 0.736766 LR 0.001000 Time 0.017144 -2022-12-06 10:23:58,825 - Epoch: [2][ 200/ 1200] Overall Loss 0.738674 Objective Loss 0.738674 LR 0.001000 Time 0.016957 -2022-12-06 10:23:58,959 - Epoch: [2][ 210/ 1200] Overall Loss 0.737027 Objective Loss 0.737027 LR 0.001000 Time 0.016784 -2022-12-06 10:23:59,092 - Epoch: [2][ 220/ 1200] Overall Loss 0.737247 Objective Loss 0.737247 LR 0.001000 Time 0.016624 -2022-12-06 10:23:59,225 - Epoch: [2][ 230/ 1200] Overall Loss 0.738562 Objective Loss 0.738562 LR 0.001000 Time 0.016478 -2022-12-06 10:23:59,359 - Epoch: [2][ 240/ 1200] Overall Loss 0.738484 Objective Loss 0.738484 LR 0.001000 Time 0.016347 -2022-12-06 10:23:59,492 - Epoch: [2][ 250/ 1200] Overall Loss 0.739408 Objective Loss 0.739408 LR 0.001000 Time 0.016222 -2022-12-06 10:23:59,626 - Epoch: [2][ 260/ 1200] Overall Loss 0.739900 Objective Loss 0.739900 LR 0.001000 Time 0.016112 -2022-12-06 10:23:59,759 - Epoch: [2][ 270/ 1200] Overall Loss 0.741725 Objective Loss 0.741725 LR 0.001000 Time 0.016006 -2022-12-06 10:23:59,892 - Epoch: [2][ 280/ 1200] Overall Loss 0.742410 Objective Loss 0.742410 LR 0.001000 Time 0.015910 -2022-12-06 10:24:00,026 - Epoch: [2][ 290/ 1200] Overall Loss 0.743062 Objective Loss 0.743062 LR 0.001000 Time 0.015821 -2022-12-06 10:24:00,160 - Epoch: [2][ 300/ 1200] Overall Loss 0.742994 Objective Loss 0.742994 LR 0.001000 Time 0.015738 -2022-12-06 10:24:00,295 - Epoch: [2][ 310/ 1200] Overall Loss 0.743787 Objective Loss 0.743787 LR 0.001000 Time 0.015663 -2022-12-06 10:24:00,429 - Epoch: [2][ 320/ 1200] Overall Loss 0.742527 Objective Loss 0.742527 LR 0.001000 Time 0.015592 -2022-12-06 10:24:00,563 - Epoch: [2][ 330/ 1200] Overall Loss 0.741873 Objective Loss 0.741873 LR 0.001000 Time 0.015524 -2022-12-06 10:24:00,696 - Epoch: [2][ 340/ 1200] Overall Loss 0.741031 Objective Loss 0.741031 LR 0.001000 Time 0.015459 -2022-12-06 10:24:00,831 - Epoch: [2][ 350/ 1200] Overall Loss 0.740058 Objective Loss 0.740058 LR 0.001000 Time 0.015399 -2022-12-06 10:24:00,965 - Epoch: [2][ 360/ 1200] Overall Loss 0.739156 Objective Loss 0.739156 LR 0.001000 Time 0.015343 -2022-12-06 10:24:01,099 - Epoch: [2][ 370/ 1200] Overall Loss 0.739656 Objective Loss 0.739656 LR 0.001000 Time 0.015289 -2022-12-06 10:24:01,234 - Epoch: [2][ 380/ 1200] Overall Loss 0.739199 Objective Loss 0.739199 LR 0.001000 Time 0.015241 -2022-12-06 10:24:01,368 - Epoch: [2][ 390/ 1200] Overall Loss 0.739611 Objective Loss 0.739611 LR 0.001000 Time 0.015192 -2022-12-06 10:24:01,501 - Epoch: [2][ 400/ 1200] Overall Loss 0.737561 Objective Loss 0.737561 LR 0.001000 Time 0.015144 -2022-12-06 10:24:01,635 - Epoch: [2][ 410/ 1200] Overall Loss 0.737221 Objective Loss 0.737221 LR 0.001000 Time 0.015100 -2022-12-06 10:24:01,769 - Epoch: [2][ 420/ 1200] Overall Loss 0.738094 Objective Loss 0.738094 LR 0.001000 Time 0.015058 -2022-12-06 10:24:01,903 - Epoch: [2][ 430/ 1200] Overall Loss 0.737316 Objective Loss 0.737316 LR 0.001000 Time 0.015019 -2022-12-06 10:24:02,037 - Epoch: [2][ 440/ 1200] Overall Loss 0.737244 Objective Loss 0.737244 LR 0.001000 Time 0.014981 -2022-12-06 10:24:02,172 - Epoch: [2][ 450/ 1200] Overall Loss 0.736510 Objective Loss 0.736510 LR 0.001000 Time 0.014945 -2022-12-06 10:24:02,307 - Epoch: [2][ 460/ 1200] Overall Loss 0.736366 Objective Loss 0.736366 LR 0.001000 Time 0.014913 -2022-12-06 10:24:02,441 - Epoch: [2][ 470/ 1200] Overall Loss 0.736824 Objective Loss 0.736824 LR 0.001000 Time 0.014879 -2022-12-06 10:24:02,575 - Epoch: [2][ 480/ 1200] Overall Loss 0.736093 Objective Loss 0.736093 LR 0.001000 Time 0.014847 -2022-12-06 10:24:02,709 - Epoch: [2][ 490/ 1200] Overall Loss 0.735496 Objective Loss 0.735496 LR 0.001000 Time 0.014817 -2022-12-06 10:24:02,843 - Epoch: [2][ 500/ 1200] Overall Loss 0.735879 Objective Loss 0.735879 LR 0.001000 Time 0.014788 -2022-12-06 10:24:02,978 - Epoch: [2][ 510/ 1200] Overall Loss 0.734869 Objective Loss 0.734869 LR 0.001000 Time 0.014760 -2022-12-06 10:24:03,111 - Epoch: [2][ 520/ 1200] Overall Loss 0.734459 Objective Loss 0.734459 LR 0.001000 Time 0.014732 -2022-12-06 10:24:03,245 - Epoch: [2][ 530/ 1200] Overall Loss 0.734358 Objective Loss 0.734358 LR 0.001000 Time 0.014705 -2022-12-06 10:24:03,379 - Epoch: [2][ 540/ 1200] Overall Loss 0.734239 Objective Loss 0.734239 LR 0.001000 Time 0.014680 -2022-12-06 10:24:03,514 - Epoch: [2][ 550/ 1200] Overall Loss 0.732672 Objective Loss 0.732672 LR 0.001000 Time 0.014657 -2022-12-06 10:24:03,647 - Epoch: [2][ 560/ 1200] Overall Loss 0.732720 Objective Loss 0.732720 LR 0.001000 Time 0.014633 -2022-12-06 10:24:03,780 - Epoch: [2][ 570/ 1200] Overall Loss 0.731889 Objective Loss 0.731889 LR 0.001000 Time 0.014609 -2022-12-06 10:24:03,915 - Epoch: [2][ 580/ 1200] Overall Loss 0.730702 Objective Loss 0.730702 LR 0.001000 Time 0.014589 -2022-12-06 10:24:04,048 - Epoch: [2][ 590/ 1200] Overall Loss 0.730113 Objective Loss 0.730113 LR 0.001000 Time 0.014566 -2022-12-06 10:24:04,182 - Epoch: [2][ 600/ 1200] Overall Loss 0.729431 Objective Loss 0.729431 LR 0.001000 Time 0.014546 -2022-12-06 10:24:04,315 - Epoch: [2][ 610/ 1200] Overall Loss 0.729414 Objective Loss 0.729414 LR 0.001000 Time 0.014525 -2022-12-06 10:24:04,450 - Epoch: [2][ 620/ 1200] Overall Loss 0.728783 Objective Loss 0.728783 LR 0.001000 Time 0.014507 -2022-12-06 10:24:04,583 - Epoch: [2][ 630/ 1200] Overall Loss 0.728625 Objective Loss 0.728625 LR 0.001000 Time 0.014487 -2022-12-06 10:24:04,717 - Epoch: [2][ 640/ 1200] Overall Loss 0.727550 Objective Loss 0.727550 LR 0.001000 Time 0.014470 -2022-12-06 10:24:04,851 - Epoch: [2][ 650/ 1200] Overall Loss 0.726435 Objective Loss 0.726435 LR 0.001000 Time 0.014452 -2022-12-06 10:24:04,985 - Epoch: [2][ 660/ 1200] Overall Loss 0.726096 Objective Loss 0.726096 LR 0.001000 Time 0.014436 -2022-12-06 10:24:05,118 - Epoch: [2][ 670/ 1200] Overall Loss 0.725573 Objective Loss 0.725573 LR 0.001000 Time 0.014419 -2022-12-06 10:24:05,252 - Epoch: [2][ 680/ 1200] Overall Loss 0.725355 Objective Loss 0.725355 LR 0.001000 Time 0.014403 -2022-12-06 10:24:05,385 - Epoch: [2][ 690/ 1200] Overall Loss 0.725087 Objective Loss 0.725087 LR 0.001000 Time 0.014386 -2022-12-06 10:24:05,519 - Epoch: [2][ 700/ 1200] Overall Loss 0.724929 Objective Loss 0.724929 LR 0.001000 Time 0.014372 -2022-12-06 10:24:05,653 - Epoch: [2][ 710/ 1200] Overall Loss 0.724343 Objective Loss 0.724343 LR 0.001000 Time 0.014357 -2022-12-06 10:24:05,787 - Epoch: [2][ 720/ 1200] Overall Loss 0.723641 Objective Loss 0.723641 LR 0.001000 Time 0.014343 -2022-12-06 10:24:05,920 - Epoch: [2][ 730/ 1200] Overall Loss 0.722725 Objective Loss 0.722725 LR 0.001000 Time 0.014328 -2022-12-06 10:24:06,054 - Epoch: [2][ 740/ 1200] Overall Loss 0.722099 Objective Loss 0.722099 LR 0.001000 Time 0.014315 -2022-12-06 10:24:06,187 - Epoch: [2][ 750/ 1200] Overall Loss 0.721411 Objective Loss 0.721411 LR 0.001000 Time 0.014301 -2022-12-06 10:24:06,322 - Epoch: [2][ 760/ 1200] Overall Loss 0.720992 Objective Loss 0.720992 LR 0.001000 Time 0.014289 -2022-12-06 10:24:06,455 - Epoch: [2][ 770/ 1200] Overall Loss 0.721251 Objective Loss 0.721251 LR 0.001000 Time 0.014276 -2022-12-06 10:24:06,589 - Epoch: [2][ 780/ 1200] Overall Loss 0.720656 Objective Loss 0.720656 LR 0.001000 Time 0.014265 -2022-12-06 10:24:06,723 - Epoch: [2][ 790/ 1200] Overall Loss 0.720359 Objective Loss 0.720359 LR 0.001000 Time 0.014252 -2022-12-06 10:24:06,857 - Epoch: [2][ 800/ 1200] Overall Loss 0.720059 Objective Loss 0.720059 LR 0.001000 Time 0.014242 -2022-12-06 10:24:06,991 - Epoch: [2][ 810/ 1200] Overall Loss 0.719216 Objective Loss 0.719216 LR 0.001000 Time 0.014230 -2022-12-06 10:24:07,125 - Epoch: [2][ 820/ 1200] Overall Loss 0.718831 Objective Loss 0.718831 LR 0.001000 Time 0.014220 -2022-12-06 10:24:07,259 - Epoch: [2][ 830/ 1200] Overall Loss 0.718156 Objective Loss 0.718156 LR 0.001000 Time 0.014210 -2022-12-06 10:24:07,395 - Epoch: [2][ 840/ 1200] Overall Loss 0.718369 Objective Loss 0.718369 LR 0.001000 Time 0.014201 -2022-12-06 10:24:07,529 - Epoch: [2][ 850/ 1200] Overall Loss 0.717779 Objective Loss 0.717779 LR 0.001000 Time 0.014191 -2022-12-06 10:24:07,662 - Epoch: [2][ 860/ 1200] Overall Loss 0.717421 Objective Loss 0.717421 LR 0.001000 Time 0.014181 -2022-12-06 10:24:07,798 - Epoch: [2][ 870/ 1200] Overall Loss 0.717028 Objective Loss 0.717028 LR 0.001000 Time 0.014172 -2022-12-06 10:24:07,933 - Epoch: [2][ 880/ 1200] Overall Loss 0.716437 Objective Loss 0.716437 LR 0.001000 Time 0.014163 -2022-12-06 10:24:08,067 - Epoch: [2][ 890/ 1200] Overall Loss 0.716382 Objective Loss 0.716382 LR 0.001000 Time 0.014154 -2022-12-06 10:24:08,201 - Epoch: [2][ 900/ 1200] Overall Loss 0.716233 Objective Loss 0.716233 LR 0.001000 Time 0.014146 -2022-12-06 10:24:08,336 - Epoch: [2][ 910/ 1200] Overall Loss 0.715984 Objective Loss 0.715984 LR 0.001000 Time 0.014137 -2022-12-06 10:24:08,469 - Epoch: [2][ 920/ 1200] Overall Loss 0.715501 Objective Loss 0.715501 LR 0.001000 Time 0.014128 -2022-12-06 10:24:08,604 - Epoch: [2][ 930/ 1200] Overall Loss 0.715516 Objective Loss 0.715516 LR 0.001000 Time 0.014120 -2022-12-06 10:24:08,738 - Epoch: [2][ 940/ 1200] Overall Loss 0.714638 Objective Loss 0.714638 LR 0.001000 Time 0.014111 -2022-12-06 10:24:08,874 - Epoch: [2][ 950/ 1200] Overall Loss 0.714186 Objective Loss 0.714186 LR 0.001000 Time 0.014104 -2022-12-06 10:24:09,009 - Epoch: [2][ 960/ 1200] Overall Loss 0.713467 Objective Loss 0.713467 LR 0.001000 Time 0.014097 -2022-12-06 10:24:09,142 - Epoch: [2][ 970/ 1200] Overall Loss 0.713123 Objective Loss 0.713123 LR 0.001000 Time 0.014089 -2022-12-06 10:24:09,277 - Epoch: [2][ 980/ 1200] Overall Loss 0.712383 Objective Loss 0.712383 LR 0.001000 Time 0.014082 -2022-12-06 10:24:09,410 - Epoch: [2][ 990/ 1200] Overall Loss 0.711981 Objective Loss 0.711981 LR 0.001000 Time 0.014074 -2022-12-06 10:24:09,545 - Epoch: [2][ 1000/ 1200] Overall Loss 0.711269 Objective Loss 0.711269 LR 0.001000 Time 0.014068 -2022-12-06 10:24:09,679 - Epoch: [2][ 1010/ 1200] Overall Loss 0.710959 Objective Loss 0.710959 LR 0.001000 Time 0.014060 -2022-12-06 10:24:09,814 - Epoch: [2][ 1020/ 1200] Overall Loss 0.710629 Objective Loss 0.710629 LR 0.001000 Time 0.014055 -2022-12-06 10:24:09,948 - Epoch: [2][ 1030/ 1200] Overall Loss 0.710448 Objective Loss 0.710448 LR 0.001000 Time 0.014048 -2022-12-06 10:24:10,082 - Epoch: [2][ 1040/ 1200] Overall Loss 0.710036 Objective Loss 0.710036 LR 0.001000 Time 0.014041 -2022-12-06 10:24:10,216 - Epoch: [2][ 1050/ 1200] Overall Loss 0.709765 Objective Loss 0.709765 LR 0.001000 Time 0.014034 -2022-12-06 10:24:10,350 - Epoch: [2][ 1060/ 1200] Overall Loss 0.709271 Objective Loss 0.709271 LR 0.001000 Time 0.014029 -2022-12-06 10:24:10,484 - Epoch: [2][ 1070/ 1200] Overall Loss 0.708688 Objective Loss 0.708688 LR 0.001000 Time 0.014022 -2022-12-06 10:24:10,618 - Epoch: [2][ 1080/ 1200] Overall Loss 0.708432 Objective Loss 0.708432 LR 0.001000 Time 0.014016 -2022-12-06 10:24:10,751 - Epoch: [2][ 1090/ 1200] Overall Loss 0.707449 Objective Loss 0.707449 LR 0.001000 Time 0.014009 -2022-12-06 10:24:10,886 - Epoch: [2][ 1100/ 1200] Overall Loss 0.706852 Objective Loss 0.706852 LR 0.001000 Time 0.014004 -2022-12-06 10:24:11,020 - Epoch: [2][ 1110/ 1200] Overall Loss 0.706314 Objective Loss 0.706314 LR 0.001000 Time 0.013998 -2022-12-06 10:24:11,155 - Epoch: [2][ 1120/ 1200] Overall Loss 0.705905 Objective Loss 0.705905 LR 0.001000 Time 0.013993 -2022-12-06 10:24:11,288 - Epoch: [2][ 1130/ 1200] Overall Loss 0.705373 Objective Loss 0.705373 LR 0.001000 Time 0.013987 -2022-12-06 10:24:11,423 - Epoch: [2][ 1140/ 1200] Overall Loss 0.705004 Objective Loss 0.705004 LR 0.001000 Time 0.013982 -2022-12-06 10:24:11,556 - Epoch: [2][ 1150/ 1200] Overall Loss 0.704212 Objective Loss 0.704212 LR 0.001000 Time 0.013976 -2022-12-06 10:24:11,691 - Epoch: [2][ 1160/ 1200] Overall Loss 0.703526 Objective Loss 0.703526 LR 0.001000 Time 0.013971 -2022-12-06 10:24:11,825 - Epoch: [2][ 1170/ 1200] Overall Loss 0.702851 Objective Loss 0.702851 LR 0.001000 Time 0.013966 -2022-12-06 10:24:11,959 - Epoch: [2][ 1180/ 1200] Overall Loss 0.703189 Objective Loss 0.703189 LR 0.001000 Time 0.013961 -2022-12-06 10:24:12,092 - Epoch: [2][ 1190/ 1200] Overall Loss 0.702657 Objective Loss 0.702657 LR 0.001000 Time 0.013955 -2022-12-06 10:24:12,271 - Epoch: [2][ 1200/ 1200] Overall Loss 0.702163 Objective Loss 0.702163 Top1 73.849372 Top5 96.025105 LR 0.001000 Time 0.013987 -2022-12-06 10:24:12,363 - --- validate (epoch=2)----------- -2022-12-06 10:24:12,363 - 34129 samples (256 per mini-batch) -2022-12-06 10:24:12,769 - Epoch: [2][ 10/ 134] Loss 0.631416 Top1 70.546875 Top5 95.781250 -2022-12-06 10:24:12,862 - Epoch: [2][ 20/ 134] Loss 0.603505 Top1 71.171875 Top5 95.488281 -2022-12-06 10:24:12,957 - Epoch: [2][ 30/ 134] Loss 0.606596 Top1 71.145833 Top5 95.638021 -2022-12-06 10:24:13,051 - Epoch: [2][ 40/ 134] Loss 0.608564 Top1 70.810547 Top5 95.615234 -2022-12-06 10:24:13,136 - Epoch: [2][ 50/ 134] Loss 0.600759 Top1 70.914063 Top5 95.750000 -2022-12-06 10:24:13,233 - Epoch: [2][ 60/ 134] Loss 0.594019 Top1 71.197917 Top5 95.878906 -2022-12-06 10:24:13,325 - Epoch: [2][ 70/ 134] Loss 0.597730 Top1 71.238839 Top5 95.904018 -2022-12-06 10:24:13,409 - Epoch: [2][ 80/ 134] Loss 0.601800 Top1 71.264648 Top5 95.898438 -2022-12-06 10:24:13,493 - Epoch: [2][ 90/ 134] Loss 0.602240 Top1 71.276042 Top5 95.954861 -2022-12-06 10:24:13,577 - Epoch: [2][ 100/ 134] Loss 0.604735 Top1 71.183594 Top5 95.878906 -2022-12-06 10:24:13,661 - Epoch: [2][ 110/ 134] Loss 0.601193 Top1 71.306818 Top5 95.884233 -2022-12-06 10:24:13,746 - Epoch: [2][ 120/ 134] Loss 0.601902 Top1 71.396484 Top5 95.882161 -2022-12-06 10:24:13,833 - Epoch: [2][ 130/ 134] Loss 0.604689 Top1 71.340144 Top5 95.883413 -2022-12-06 10:24:13,853 - Epoch: [2][ 134/ 134] Loss 0.604769 Top1 71.352808 Top5 95.883266 -2022-12-06 10:24:13,942 - ==> Top1: 71.353 Top5: 95.883 Loss: 0.605 - -2022-12-06 10:24:13,943 - ==> Confusion: -[[ 823 5 4 1 5 4 0 0 14 109 0 0 0 14 1 1 1 4 4 1 5] - [ 1 848 7 0 7 80 1 8 5 0 7 7 2 6 4 2 10 0 16 1 15] - [ 21 8 923 12 6 11 33 20 0 4 5 6 3 16 1 9 0 2 7 3 13] - [ 7 4 59 770 1 16 1 2 12 0 42 1 7 8 34 1 4 9 29 1 12] - [ 24 17 5 0 852 23 0 1 1 13 0 3 1 26 9 12 16 2 4 1 10] - [ 5 70 4 1 5 836 6 22 9 6 3 10 2 66 4 1 4 0 4 5 6] - [ 1 9 55 3 1 16 964 11 1 0 5 4 2 1 0 9 3 3 4 24 2] - [ 2 22 37 2 1 113 6 757 1 1 4 5 0 6 0 1 0 1 69 17 9] - [ 11 7 1 1 0 6 0 1 921 53 7 1 1 35 14 0 1 0 2 0 2] - [ 108 2 3 0 1 4 0 1 63 762 0 2 0 36 4 0 0 0 3 0 12] - [ 1 12 23 29 2 27 3 5 28 2 830 1 1 27 5 0 1 0 16 2 4] - [ 5 12 0 0 0 43 3 3 7 0 0 854 24 36 1 15 4 12 4 21 7] - [ 4 5 0 9 2 11 1 1 20 1 1 103 697 15 6 12 5 60 2 2 12] - [ 5 5 2 1 3 34 0 1 10 18 1 8 3 909 1 4 3 1 1 4 9] - [ 21 9 0 10 10 5 0 0 64 10 5 1 6 14 936 0 6 2 10 0 21] - [ 6 2 8 3 4 6 6 0 1 0 0 10 9 4 0 931 12 22 3 4 12] - [ 8 11 2 0 18 9 2 0 6 2 0 5 2 8 1 15 951 1 5 6 20] - [ 5 1 3 2 0 8 4 0 5 0 1 24 50 7 4 12 0 899 3 2 6] - [ 1 12 13 14 0 19 1 32 11 1 15 2 1 1 8 0 3 0 864 6 4] - [ 1 8 4 1 1 30 4 14 0 0 0 28 9 20 0 7 8 6 4 923 12] - [ 389 447 326 97 184 643 66 151 178 211 245 242 358 759 246 237 266 129 364 586 7102]] - -2022-12-06 10:24:14,517 - ==> Best [Top1: 71.353 Top5: 95.883 Sparsity:0.00 Params: 5376 on epoch: 2] -2022-12-06 10:24:14,517 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:24:14,524 - - -2022-12-06 10:24:14,524 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:24:15,463 - Epoch: [3][ 10/ 1200] Overall Loss 0.643151 Objective Loss 0.643151 LR 0.001000 Time 0.093909 -2022-12-06 10:24:15,620 - Epoch: [3][ 20/ 1200] Overall Loss 0.620181 Objective Loss 0.620181 LR 0.001000 Time 0.054731 -2022-12-06 10:24:15,760 - Epoch: [3][ 30/ 1200] Overall Loss 0.620896 Objective Loss 0.620896 LR 0.001000 Time 0.041133 -2022-12-06 10:24:15,893 - Epoch: [3][ 40/ 1200] Overall Loss 0.617169 Objective Loss 0.617169 LR 0.001000 Time 0.034167 -2022-12-06 10:24:16,027 - Epoch: [3][ 50/ 1200] Overall Loss 0.619608 Objective Loss 0.619608 LR 0.001000 Time 0.030012 -2022-12-06 10:24:16,162 - Epoch: [3][ 60/ 1200] Overall Loss 0.615352 Objective Loss 0.615352 LR 0.001000 Time 0.027244 -2022-12-06 10:24:16,300 - Epoch: [3][ 70/ 1200] Overall Loss 0.614280 Objective Loss 0.614280 LR 0.001000 Time 0.025322 -2022-12-06 10:24:16,435 - Epoch: [3][ 80/ 1200] Overall Loss 0.615231 Objective Loss 0.615231 LR 0.001000 Time 0.023833 -2022-12-06 10:24:16,567 - Epoch: [3][ 90/ 1200] Overall Loss 0.617019 Objective Loss 0.617019 LR 0.001000 Time 0.022649 -2022-12-06 10:24:16,700 - Epoch: [3][ 100/ 1200] Overall Loss 0.620464 Objective Loss 0.620464 LR 0.001000 Time 0.021709 -2022-12-06 10:24:16,834 - Epoch: [3][ 110/ 1200] Overall Loss 0.618838 Objective Loss 0.618838 LR 0.001000 Time 0.020951 -2022-12-06 10:24:16,967 - Epoch: [3][ 120/ 1200] Overall Loss 0.621793 Objective Loss 0.621793 LR 0.001000 Time 0.020309 -2022-12-06 10:24:17,101 - Epoch: [3][ 130/ 1200] Overall Loss 0.622304 Objective Loss 0.622304 LR 0.001000 Time 0.019766 -2022-12-06 10:24:17,234 - Epoch: [3][ 140/ 1200] Overall Loss 0.623876 Objective Loss 0.623876 LR 0.001000 Time 0.019307 -2022-12-06 10:24:17,369 - Epoch: [3][ 150/ 1200] Overall Loss 0.625075 Objective Loss 0.625075 LR 0.001000 Time 0.018914 -2022-12-06 10:24:17,503 - Epoch: [3][ 160/ 1200] Overall Loss 0.624223 Objective Loss 0.624223 LR 0.001000 Time 0.018567 -2022-12-06 10:24:17,638 - Epoch: [3][ 170/ 1200] Overall Loss 0.624775 Objective Loss 0.624775 LR 0.001000 Time 0.018264 -2022-12-06 10:24:17,772 - Epoch: [3][ 180/ 1200] Overall Loss 0.625653 Objective Loss 0.625653 LR 0.001000 Time 0.017990 -2022-12-06 10:24:17,904 - Epoch: [3][ 190/ 1200] Overall Loss 0.625585 Objective Loss 0.625585 LR 0.001000 Time 0.017737 -2022-12-06 10:24:18,037 - Epoch: [3][ 200/ 1200] Overall Loss 0.623824 Objective Loss 0.623824 LR 0.001000 Time 0.017512 -2022-12-06 10:24:18,172 - Epoch: [3][ 210/ 1200] Overall Loss 0.622363 Objective Loss 0.622363 LR 0.001000 Time 0.017315 -2022-12-06 10:24:18,305 - Epoch: [3][ 220/ 1200] Overall Loss 0.622576 Objective Loss 0.622576 LR 0.001000 Time 0.017132 -2022-12-06 10:24:18,437 - Epoch: [3][ 230/ 1200] Overall Loss 0.623865 Objective Loss 0.623865 LR 0.001000 Time 0.016959 -2022-12-06 10:24:18,571 - Epoch: [3][ 240/ 1200] Overall Loss 0.624126 Objective Loss 0.624126 LR 0.001000 Time 0.016806 -2022-12-06 10:24:18,702 - Epoch: [3][ 250/ 1200] Overall Loss 0.624248 Objective Loss 0.624248 LR 0.001000 Time 0.016659 -2022-12-06 10:24:18,834 - Epoch: [3][ 260/ 1200] Overall Loss 0.623770 Objective Loss 0.623770 LR 0.001000 Time 0.016523 -2022-12-06 10:24:18,968 - Epoch: [3][ 270/ 1200] Overall Loss 0.622427 Objective Loss 0.622427 LR 0.001000 Time 0.016403 -2022-12-06 10:24:19,102 - Epoch: [3][ 280/ 1200] Overall Loss 0.621536 Objective Loss 0.621536 LR 0.001000 Time 0.016294 -2022-12-06 10:24:19,235 - Epoch: [3][ 290/ 1200] Overall Loss 0.620676 Objective Loss 0.620676 LR 0.001000 Time 0.016190 -2022-12-06 10:24:19,369 - Epoch: [3][ 300/ 1200] Overall Loss 0.621519 Objective Loss 0.621519 LR 0.001000 Time 0.016096 -2022-12-06 10:24:19,502 - Epoch: [3][ 310/ 1200] Overall Loss 0.620028 Objective Loss 0.620028 LR 0.001000 Time 0.016003 -2022-12-06 10:24:19,634 - Epoch: [3][ 320/ 1200] Overall Loss 0.619545 Objective Loss 0.619545 LR 0.001000 Time 0.015914 -2022-12-06 10:24:19,766 - Epoch: [3][ 330/ 1200] Overall Loss 0.620728 Objective Loss 0.620728 LR 0.001000 Time 0.015830 -2022-12-06 10:24:19,898 - Epoch: [3][ 340/ 1200] Overall Loss 0.619437 Objective Loss 0.619437 LR 0.001000 Time 0.015751 -2022-12-06 10:24:20,031 - Epoch: [3][ 350/ 1200] Overall Loss 0.619587 Objective Loss 0.619587 LR 0.001000 Time 0.015679 -2022-12-06 10:24:20,163 - Epoch: [3][ 360/ 1200] Overall Loss 0.619060 Objective Loss 0.619060 LR 0.001000 Time 0.015608 -2022-12-06 10:24:20,296 - Epoch: [3][ 370/ 1200] Overall Loss 0.618351 Objective Loss 0.618351 LR 0.001000 Time 0.015546 -2022-12-06 10:24:20,429 - Epoch: [3][ 380/ 1200] Overall Loss 0.617651 Objective Loss 0.617651 LR 0.001000 Time 0.015484 -2022-12-06 10:24:20,562 - Epoch: [3][ 390/ 1200] Overall Loss 0.617097 Objective Loss 0.617097 LR 0.001000 Time 0.015427 -2022-12-06 10:24:20,696 - Epoch: [3][ 400/ 1200] Overall Loss 0.616624 Objective Loss 0.616624 LR 0.001000 Time 0.015375 -2022-12-06 10:24:20,828 - Epoch: [3][ 410/ 1200] Overall Loss 0.615935 Objective Loss 0.615935 LR 0.001000 Time 0.015322 -2022-12-06 10:24:20,961 - Epoch: [3][ 420/ 1200] Overall Loss 0.614805 Objective Loss 0.614805 LR 0.001000 Time 0.015271 -2022-12-06 10:24:21,092 - Epoch: [3][ 430/ 1200] Overall Loss 0.614792 Objective Loss 0.614792 LR 0.001000 Time 0.015220 -2022-12-06 10:24:21,225 - Epoch: [3][ 440/ 1200] Overall Loss 0.614507 Objective Loss 0.614507 LR 0.001000 Time 0.015176 -2022-12-06 10:24:21,359 - Epoch: [3][ 450/ 1200] Overall Loss 0.614185 Objective Loss 0.614185 LR 0.001000 Time 0.015134 -2022-12-06 10:24:21,493 - Epoch: [3][ 460/ 1200] Overall Loss 0.612787 Objective Loss 0.612787 LR 0.001000 Time 0.015097 -2022-12-06 10:24:21,626 - Epoch: [3][ 470/ 1200] Overall Loss 0.611728 Objective Loss 0.611728 LR 0.001000 Time 0.015057 -2022-12-06 10:24:21,758 - Epoch: [3][ 480/ 1200] Overall Loss 0.611587 Objective Loss 0.611587 LR 0.001000 Time 0.015017 -2022-12-06 10:24:21,890 - Epoch: [3][ 490/ 1200] Overall Loss 0.610931 Objective Loss 0.610931 LR 0.001000 Time 0.014978 -2022-12-06 10:24:22,024 - Epoch: [3][ 500/ 1200] Overall Loss 0.611152 Objective Loss 0.611152 LR 0.001000 Time 0.014945 -2022-12-06 10:24:22,157 - Epoch: [3][ 510/ 1200] Overall Loss 0.610896 Objective Loss 0.610896 LR 0.001000 Time 0.014913 -2022-12-06 10:24:22,292 - Epoch: [3][ 520/ 1200] Overall Loss 0.610524 Objective Loss 0.610524 LR 0.001000 Time 0.014884 -2022-12-06 10:24:22,425 - Epoch: [3][ 530/ 1200] Overall Loss 0.610833 Objective Loss 0.610833 LR 0.001000 Time 0.014853 -2022-12-06 10:24:22,558 - Epoch: [3][ 540/ 1200] Overall Loss 0.610216 Objective Loss 0.610216 LR 0.001000 Time 0.014824 -2022-12-06 10:24:22,693 - Epoch: [3][ 550/ 1200] Overall Loss 0.610610 Objective Loss 0.610610 LR 0.001000 Time 0.014798 -2022-12-06 10:24:22,827 - Epoch: [3][ 560/ 1200] Overall Loss 0.609781 Objective Loss 0.609781 LR 0.001000 Time 0.014772 -2022-12-06 10:24:22,960 - Epoch: [3][ 570/ 1200] Overall Loss 0.609702 Objective Loss 0.609702 LR 0.001000 Time 0.014746 -2022-12-06 10:24:23,092 - Epoch: [3][ 580/ 1200] Overall Loss 0.610244 Objective Loss 0.610244 LR 0.001000 Time 0.014718 -2022-12-06 10:24:23,224 - Epoch: [3][ 590/ 1200] Overall Loss 0.609858 Objective Loss 0.609858 LR 0.001000 Time 0.014691 -2022-12-06 10:24:23,358 - Epoch: [3][ 600/ 1200] Overall Loss 0.609278 Objective Loss 0.609278 LR 0.001000 Time 0.014669 -2022-12-06 10:24:23,492 - Epoch: [3][ 610/ 1200] Overall Loss 0.608850 Objective Loss 0.608850 LR 0.001000 Time 0.014648 -2022-12-06 10:24:23,626 - Epoch: [3][ 620/ 1200] Overall Loss 0.608331 Objective Loss 0.608331 LR 0.001000 Time 0.014626 -2022-12-06 10:24:23,759 - Epoch: [3][ 630/ 1200] Overall Loss 0.607558 Objective Loss 0.607558 LR 0.001000 Time 0.014605 -2022-12-06 10:24:23,891 - Epoch: [3][ 640/ 1200] Overall Loss 0.606979 Objective Loss 0.606979 LR 0.001000 Time 0.014582 -2022-12-06 10:24:24,024 - Epoch: [3][ 650/ 1200] Overall Loss 0.606292 Objective Loss 0.606292 LR 0.001000 Time 0.014562 -2022-12-06 10:24:24,156 - Epoch: [3][ 660/ 1200] Overall Loss 0.605943 Objective Loss 0.605943 LR 0.001000 Time 0.014540 -2022-12-06 10:24:24,290 - Epoch: [3][ 670/ 1200] Overall Loss 0.605205 Objective Loss 0.605205 LR 0.001000 Time 0.014522 -2022-12-06 10:24:24,424 - Epoch: [3][ 680/ 1200] Overall Loss 0.605137 Objective Loss 0.605137 LR 0.001000 Time 0.014505 -2022-12-06 10:24:24,558 - Epoch: [3][ 690/ 1200] Overall Loss 0.605500 Objective Loss 0.605500 LR 0.001000 Time 0.014488 -2022-12-06 10:24:24,690 - Epoch: [3][ 700/ 1200] Overall Loss 0.605779 Objective Loss 0.605779 LR 0.001000 Time 0.014469 -2022-12-06 10:24:24,824 - Epoch: [3][ 710/ 1200] Overall Loss 0.606348 Objective Loss 0.606348 LR 0.001000 Time 0.014453 -2022-12-06 10:24:24,958 - Epoch: [3][ 720/ 1200] Overall Loss 0.605778 Objective Loss 0.605778 LR 0.001000 Time 0.014438 -2022-12-06 10:24:25,103 - Epoch: [3][ 730/ 1200] Overall Loss 0.605192 Objective Loss 0.605192 LR 0.001000 Time 0.014438 -2022-12-06 10:24:25,252 - Epoch: [3][ 740/ 1200] Overall Loss 0.605124 Objective Loss 0.605124 LR 0.001000 Time 0.014444 -2022-12-06 10:24:25,396 - Epoch: [3][ 750/ 1200] Overall Loss 0.605844 Objective Loss 0.605844 LR 0.001000 Time 0.014441 -2022-12-06 10:24:25,545 - Epoch: [3][ 760/ 1200] Overall Loss 0.605680 Objective Loss 0.605680 LR 0.001000 Time 0.014447 -2022-12-06 10:24:25,689 - Epoch: [3][ 770/ 1200] Overall Loss 0.604919 Objective Loss 0.604919 LR 0.001000 Time 0.014446 -2022-12-06 10:24:25,838 - Epoch: [3][ 780/ 1200] Overall Loss 0.604447 Objective Loss 0.604447 LR 0.001000 Time 0.014451 -2022-12-06 10:24:25,982 - Epoch: [3][ 790/ 1200] Overall Loss 0.604039 Objective Loss 0.604039 LR 0.001000 Time 0.014450 -2022-12-06 10:24:26,131 - Epoch: [3][ 800/ 1200] Overall Loss 0.603863 Objective Loss 0.603863 LR 0.001000 Time 0.014455 -2022-12-06 10:24:26,274 - Epoch: [3][ 810/ 1200] Overall Loss 0.603672 Objective Loss 0.603672 LR 0.001000 Time 0.014453 -2022-12-06 10:24:26,423 - Epoch: [3][ 820/ 1200] Overall Loss 0.603790 Objective Loss 0.603790 LR 0.001000 Time 0.014458 -2022-12-06 10:24:26,567 - Epoch: [3][ 830/ 1200] Overall Loss 0.603207 Objective Loss 0.603207 LR 0.001000 Time 0.014456 -2022-12-06 10:24:26,716 - Epoch: [3][ 840/ 1200] Overall Loss 0.603094 Objective Loss 0.603094 LR 0.001000 Time 0.014461 -2022-12-06 10:24:26,859 - Epoch: [3][ 850/ 1200] Overall Loss 0.602903 Objective Loss 0.602903 LR 0.001000 Time 0.014459 -2022-12-06 10:24:27,008 - Epoch: [3][ 860/ 1200] Overall Loss 0.602890 Objective Loss 0.602890 LR 0.001000 Time 0.014463 -2022-12-06 10:24:27,155 - Epoch: [3][ 870/ 1200] Overall Loss 0.602830 Objective Loss 0.602830 LR 0.001000 Time 0.014465 -2022-12-06 10:24:27,307 - Epoch: [3][ 880/ 1200] Overall Loss 0.603070 Objective Loss 0.603070 LR 0.001000 Time 0.014473 -2022-12-06 10:24:27,452 - Epoch: [3][ 890/ 1200] Overall Loss 0.602863 Objective Loss 0.602863 LR 0.001000 Time 0.014473 -2022-12-06 10:24:27,602 - Epoch: [3][ 900/ 1200] Overall Loss 0.602577 Objective Loss 0.602577 LR 0.001000 Time 0.014478 -2022-12-06 10:24:27,745 - Epoch: [3][ 910/ 1200] Overall Loss 0.602643 Objective Loss 0.602643 LR 0.001000 Time 0.014476 -2022-12-06 10:24:27,894 - Epoch: [3][ 920/ 1200] Overall Loss 0.602727 Objective Loss 0.602727 LR 0.001000 Time 0.014480 -2022-12-06 10:24:28,038 - Epoch: [3][ 930/ 1200] Overall Loss 0.602266 Objective Loss 0.602266 LR 0.001000 Time 0.014478 -2022-12-06 10:24:28,187 - Epoch: [3][ 940/ 1200] Overall Loss 0.601514 Objective Loss 0.601514 LR 0.001000 Time 0.014482 -2022-12-06 10:24:28,330 - Epoch: [3][ 950/ 1200] Overall Loss 0.601161 Objective Loss 0.601161 LR 0.001000 Time 0.014480 -2022-12-06 10:24:28,479 - Epoch: [3][ 960/ 1200] Overall Loss 0.600795 Objective Loss 0.600795 LR 0.001000 Time 0.014483 -2022-12-06 10:24:28,622 - Epoch: [3][ 970/ 1200] Overall Loss 0.600714 Objective Loss 0.600714 LR 0.001000 Time 0.014481 -2022-12-06 10:24:28,770 - Epoch: [3][ 980/ 1200] Overall Loss 0.600303 Objective Loss 0.600303 LR 0.001000 Time 0.014484 -2022-12-06 10:24:28,913 - Epoch: [3][ 990/ 1200] Overall Loss 0.599970 Objective Loss 0.599970 LR 0.001000 Time 0.014482 -2022-12-06 10:24:29,062 - Epoch: [3][ 1000/ 1200] Overall Loss 0.599353 Objective Loss 0.599353 LR 0.001000 Time 0.014486 -2022-12-06 10:24:29,207 - Epoch: [3][ 1010/ 1200] Overall Loss 0.599259 Objective Loss 0.599259 LR 0.001000 Time 0.014485 -2022-12-06 10:24:29,358 - Epoch: [3][ 1020/ 1200] Overall Loss 0.598783 Objective Loss 0.598783 LR 0.001000 Time 0.014491 -2022-12-06 10:24:29,495 - Epoch: [3][ 1030/ 1200] Overall Loss 0.598498 Objective Loss 0.598498 LR 0.001000 Time 0.014482 -2022-12-06 10:24:29,632 - Epoch: [3][ 1040/ 1200] Overall Loss 0.598156 Objective Loss 0.598156 LR 0.001000 Time 0.014475 -2022-12-06 10:24:29,767 - Epoch: [3][ 1050/ 1200] Overall Loss 0.597892 Objective Loss 0.597892 LR 0.001000 Time 0.014465 -2022-12-06 10:24:29,904 - Epoch: [3][ 1060/ 1200] Overall Loss 0.597763 Objective Loss 0.597763 LR 0.001000 Time 0.014457 -2022-12-06 10:24:30,040 - Epoch: [3][ 1070/ 1200] Overall Loss 0.597299 Objective Loss 0.597299 LR 0.001000 Time 0.014448 -2022-12-06 10:24:30,177 - Epoch: [3][ 1080/ 1200] Overall Loss 0.597158 Objective Loss 0.597158 LR 0.001000 Time 0.014441 -2022-12-06 10:24:30,312 - Epoch: [3][ 1090/ 1200] Overall Loss 0.596763 Objective Loss 0.596763 LR 0.001000 Time 0.014432 -2022-12-06 10:24:30,448 - Epoch: [3][ 1100/ 1200] Overall Loss 0.596446 Objective Loss 0.596446 LR 0.001000 Time 0.014424 -2022-12-06 10:24:30,584 - Epoch: [3][ 1110/ 1200] Overall Loss 0.596009 Objective Loss 0.596009 LR 0.001000 Time 0.014415 -2022-12-06 10:24:30,721 - Epoch: [3][ 1120/ 1200] Overall Loss 0.595940 Objective Loss 0.595940 LR 0.001000 Time 0.014408 -2022-12-06 10:24:30,858 - Epoch: [3][ 1130/ 1200] Overall Loss 0.595980 Objective Loss 0.595980 LR 0.001000 Time 0.014400 -2022-12-06 10:24:30,996 - Epoch: [3][ 1140/ 1200] Overall Loss 0.595737 Objective Loss 0.595737 LR 0.001000 Time 0.014394 -2022-12-06 10:24:31,131 - Epoch: [3][ 1150/ 1200] Overall Loss 0.595302 Objective Loss 0.595302 LR 0.001000 Time 0.014387 -2022-12-06 10:24:31,269 - Epoch: [3][ 1160/ 1200] Overall Loss 0.595023 Objective Loss 0.595023 LR 0.001000 Time 0.014381 -2022-12-06 10:24:31,404 - Epoch: [3][ 1170/ 1200] Overall Loss 0.595026 Objective Loss 0.595026 LR 0.001000 Time 0.014373 -2022-12-06 10:24:31,542 - Epoch: [3][ 1180/ 1200] Overall Loss 0.594930 Objective Loss 0.594930 LR 0.001000 Time 0.014368 -2022-12-06 10:24:31,677 - Epoch: [3][ 1190/ 1200] Overall Loss 0.594289 Objective Loss 0.594289 LR 0.001000 Time 0.014360 -2022-12-06 10:24:31,859 - Epoch: [3][ 1200/ 1200] Overall Loss 0.593909 Objective Loss 0.593909 Top1 74.476987 Top5 96.025105 LR 0.001000 Time 0.014392 -2022-12-06 10:24:31,947 - --- validate (epoch=3)----------- -2022-12-06 10:24:31,947 - 34129 samples (256 per mini-batch) -2022-12-06 10:24:32,349 - Epoch: [3][ 10/ 134] Loss 0.531810 Top1 73.242188 Top5 95.664062 -2022-12-06 10:24:32,438 - Epoch: [3][ 20/ 134] Loss 0.534855 Top1 73.242188 Top5 95.800781 -2022-12-06 10:24:32,528 - Epoch: [3][ 30/ 134] Loss 0.538278 Top1 72.942708 Top5 96.080729 -2022-12-06 10:24:32,620 - Epoch: [3][ 40/ 134] Loss 0.540035 Top1 72.998047 Top5 96.142578 -2022-12-06 10:24:32,709 - Epoch: [3][ 50/ 134] Loss 0.541886 Top1 72.664062 Top5 96.101562 -2022-12-06 10:24:32,801 - Epoch: [3][ 60/ 134] Loss 0.541521 Top1 72.578125 Top5 96.074219 -2022-12-06 10:24:32,894 - Epoch: [3][ 70/ 134] Loss 0.541844 Top1 72.410714 Top5 96.054688 -2022-12-06 10:24:32,988 - Epoch: [3][ 80/ 134] Loss 0.542204 Top1 72.446289 Top5 96.044922 -2022-12-06 10:24:33,080 - Epoch: [3][ 90/ 134] Loss 0.539568 Top1 72.534722 Top5 96.154514 -2022-12-06 10:24:33,171 - Epoch: [3][ 100/ 134] Loss 0.539513 Top1 72.562500 Top5 96.199219 -2022-12-06 10:24:33,264 - Epoch: [3][ 110/ 134] Loss 0.539080 Top1 72.553267 Top5 96.235795 -2022-12-06 10:24:33,358 - Epoch: [3][ 120/ 134] Loss 0.540234 Top1 72.490234 Top5 96.236979 -2022-12-06 10:24:33,453 - Epoch: [3][ 130/ 134] Loss 0.538760 Top1 72.569111 Top5 96.207933 -2022-12-06 10:24:33,475 - Epoch: [3][ 134/ 134] Loss 0.540661 Top1 72.601014 Top5 96.211433 -2022-12-06 10:24:33,564 - ==> Top1: 72.601 Top5: 96.211 Loss: 0.541 - -2022-12-06 10:24:33,564 - ==> Confusion: -[[ 721 1 5 0 20 9 1 1 12 185 1 6 1 13 7 2 1 3 2 0 5] - [ 0 853 3 1 11 62 2 32 5 0 5 13 4 4 7 0 5 2 7 3 8] - [ 8 4 915 8 13 11 39 43 4 3 10 11 1 12 1 1 0 0 6 6 7] - [ 3 4 57 759 6 9 2 7 8 0 44 4 24 3 31 1 0 8 36 2 12] - [ 8 13 1 0 921 14 0 2 2 8 1 10 1 12 5 6 4 3 0 2 7] - [ 1 61 1 1 7 819 7 34 7 4 5 29 1 62 3 0 2 0 2 18 5] - [ 0 8 29 1 3 3 997 22 0 1 5 8 4 0 0 9 1 1 2 21 3] - [ 3 11 9 2 2 61 5 885 0 2 4 10 1 2 0 1 0 0 26 20 10] - [ 7 3 0 0 0 5 0 2 925 40 16 5 2 36 12 0 2 0 8 1 0] - [ 50 2 4 0 8 4 1 1 63 773 0 5 0 76 2 0 0 0 1 0 11] - [ 2 5 8 9 5 5 1 9 18 2 906 6 0 18 7 0 1 0 14 1 2] - [ 0 3 2 0 0 11 0 5 2 0 1 961 16 22 0 4 2 5 4 11 2] - [ 3 1 1 1 1 2 3 0 2 0 0 142 760 5 4 6 1 24 3 4 6] - [ 3 3 1 0 6 15 0 3 6 5 5 19 5 937 0 1 4 1 0 4 5] - [ 13 6 0 12 21 1 0 1 62 5 10 2 10 11 942 0 4 5 8 1 16] - [ 1 3 2 2 12 6 10 0 2 0 0 29 7 4 1 934 3 14 0 6 7] - [ 4 11 2 0 29 6 2 1 1 2 0 24 3 7 1 14 938 0 2 11 14] - [ 4 2 3 2 0 3 2 1 1 1 0 52 79 6 4 8 0 860 0 1 7] - [ 2 3 4 9 0 7 1 88 5 1 14 7 7 0 10 0 0 0 842 3 5] - [ 0 4 0 0 1 11 7 9 0 0 0 57 3 2 2 6 1 3 0 967 7] - [ 233 430 279 68 382 471 92 317 162 154 240 456 410 688 216 209 217 88 338 613 7163]] - -2022-12-06 10:24:34,144 - ==> Best [Top1: 72.601 Top5: 96.211 Sparsity:0.00 Params: 5376 on epoch: 3] -2022-12-06 10:24:34,144 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:24:34,209 - - -2022-12-06 10:24:34,209 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:24:35,196 - Epoch: [4][ 10/ 1200] Overall Loss 0.538873 Objective Loss 0.538873 LR 0.001000 Time 0.098617 -2022-12-06 10:24:35,342 - Epoch: [4][ 20/ 1200] Overall Loss 0.535076 Objective Loss 0.535076 LR 0.001000 Time 0.056591 -2022-12-06 10:24:35,483 - Epoch: [4][ 30/ 1200] Overall Loss 0.530378 Objective Loss 0.530378 LR 0.001000 Time 0.042395 -2022-12-06 10:24:35,625 - Epoch: [4][ 40/ 1200] Overall Loss 0.528950 Objective Loss 0.528950 LR 0.001000 Time 0.035338 -2022-12-06 10:24:35,764 - Epoch: [4][ 50/ 1200] Overall Loss 0.525031 Objective Loss 0.525031 LR 0.001000 Time 0.031035 -2022-12-06 10:24:35,905 - Epoch: [4][ 60/ 1200] Overall Loss 0.527927 Objective Loss 0.527927 LR 0.001000 Time 0.028212 -2022-12-06 10:24:36,044 - Epoch: [4][ 70/ 1200] Overall Loss 0.529074 Objective Loss 0.529074 LR 0.001000 Time 0.026160 -2022-12-06 10:24:36,186 - Epoch: [4][ 80/ 1200] Overall Loss 0.531465 Objective Loss 0.531465 LR 0.001000 Time 0.024656 -2022-12-06 10:24:36,324 - Epoch: [4][ 90/ 1200] Overall Loss 0.535231 Objective Loss 0.535231 LR 0.001000 Time 0.023450 -2022-12-06 10:24:36,465 - Epoch: [4][ 100/ 1200] Overall Loss 0.540645 Objective Loss 0.540645 LR 0.001000 Time 0.022510 -2022-12-06 10:24:36,604 - Epoch: [4][ 110/ 1200] Overall Loss 0.540722 Objective Loss 0.540722 LR 0.001000 Time 0.021716 -2022-12-06 10:24:36,745 - Epoch: [4][ 120/ 1200] Overall Loss 0.540938 Objective Loss 0.540938 LR 0.001000 Time 0.021080 -2022-12-06 10:24:36,883 - Epoch: [4][ 130/ 1200] Overall Loss 0.540617 Objective Loss 0.540617 LR 0.001000 Time 0.020519 -2022-12-06 10:24:37,025 - Epoch: [4][ 140/ 1200] Overall Loss 0.539418 Objective Loss 0.539418 LR 0.001000 Time 0.020061 -2022-12-06 10:24:37,163 - Epoch: [4][ 150/ 1200] Overall Loss 0.541328 Objective Loss 0.541328 LR 0.001000 Time 0.019643 -2022-12-06 10:24:37,303 - Epoch: [4][ 160/ 1200] Overall Loss 0.540402 Objective Loss 0.540402 LR 0.001000 Time 0.019287 -2022-12-06 10:24:37,443 - Epoch: [4][ 170/ 1200] Overall Loss 0.539171 Objective Loss 0.539171 LR 0.001000 Time 0.018962 -2022-12-06 10:24:37,583 - Epoch: [4][ 180/ 1200] Overall Loss 0.537203 Objective Loss 0.537203 LR 0.001000 Time 0.018688 -2022-12-06 10:24:37,722 - Epoch: [4][ 190/ 1200] Overall Loss 0.535312 Objective Loss 0.535312 LR 0.001000 Time 0.018429 -2022-12-06 10:24:37,861 - Epoch: [4][ 200/ 1200] Overall Loss 0.537096 Objective Loss 0.537096 LR 0.001000 Time 0.018202 -2022-12-06 10:24:37,996 - Epoch: [4][ 210/ 1200] Overall Loss 0.537552 Objective Loss 0.537552 LR 0.001000 Time 0.017972 -2022-12-06 10:24:38,130 - Epoch: [4][ 220/ 1200] Overall Loss 0.538582 Objective Loss 0.538582 LR 0.001000 Time 0.017762 -2022-12-06 10:24:38,265 - Epoch: [4][ 230/ 1200] Overall Loss 0.539457 Objective Loss 0.539457 LR 0.001000 Time 0.017571 -2022-12-06 10:24:38,399 - Epoch: [4][ 240/ 1200] Overall Loss 0.538633 Objective Loss 0.538633 LR 0.001000 Time 0.017393 -2022-12-06 10:24:38,532 - Epoch: [4][ 250/ 1200] Overall Loss 0.538545 Objective Loss 0.538545 LR 0.001000 Time 0.017228 -2022-12-06 10:24:38,667 - Epoch: [4][ 260/ 1200] Overall Loss 0.538749 Objective Loss 0.538749 LR 0.001000 Time 0.017081 -2022-12-06 10:24:38,802 - Epoch: [4][ 270/ 1200] Overall Loss 0.538077 Objective Loss 0.538077 LR 0.001000 Time 0.016946 -2022-12-06 10:24:38,936 - Epoch: [4][ 280/ 1200] Overall Loss 0.537831 Objective Loss 0.537831 LR 0.001000 Time 0.016815 -2022-12-06 10:24:39,069 - Epoch: [4][ 290/ 1200] Overall Loss 0.538808 Objective Loss 0.538808 LR 0.001000 Time 0.016691 -2022-12-06 10:24:39,204 - Epoch: [4][ 300/ 1200] Overall Loss 0.539464 Objective Loss 0.539464 LR 0.001000 Time 0.016583 -2022-12-06 10:24:39,337 - Epoch: [4][ 310/ 1200] Overall Loss 0.538054 Objective Loss 0.538054 LR 0.001000 Time 0.016477 -2022-12-06 10:24:39,472 - Epoch: [4][ 320/ 1200] Overall Loss 0.537206 Objective Loss 0.537206 LR 0.001000 Time 0.016382 -2022-12-06 10:24:39,605 - Epoch: [4][ 330/ 1200] Overall Loss 0.534740 Objective Loss 0.534740 LR 0.001000 Time 0.016288 -2022-12-06 10:24:39,739 - Epoch: [4][ 340/ 1200] Overall Loss 0.533593 Objective Loss 0.533593 LR 0.001000 Time 0.016200 -2022-12-06 10:24:39,872 - Epoch: [4][ 350/ 1200] Overall Loss 0.533675 Objective Loss 0.533675 LR 0.001000 Time 0.016117 -2022-12-06 10:24:40,006 - Epoch: [4][ 360/ 1200] Overall Loss 0.533868 Objective Loss 0.533868 LR 0.001000 Time 0.016041 -2022-12-06 10:24:40,141 - Epoch: [4][ 370/ 1200] Overall Loss 0.532846 Objective Loss 0.532846 LR 0.001000 Time 0.015969 -2022-12-06 10:24:40,275 - Epoch: [4][ 380/ 1200] Overall Loss 0.533230 Objective Loss 0.533230 LR 0.001000 Time 0.015901 -2022-12-06 10:24:40,409 - Epoch: [4][ 390/ 1200] Overall Loss 0.532753 Objective Loss 0.532753 LR 0.001000 Time 0.015835 -2022-12-06 10:24:40,543 - Epoch: [4][ 400/ 1200] Overall Loss 0.532483 Objective Loss 0.532483 LR 0.001000 Time 0.015773 -2022-12-06 10:24:40,677 - Epoch: [4][ 410/ 1200] Overall Loss 0.532690 Objective Loss 0.532690 LR 0.001000 Time 0.015716 -2022-12-06 10:24:40,812 - Epoch: [4][ 420/ 1200] Overall Loss 0.533091 Objective Loss 0.533091 LR 0.001000 Time 0.015661 -2022-12-06 10:24:40,947 - Epoch: [4][ 430/ 1200] Overall Loss 0.531358 Objective Loss 0.531358 LR 0.001000 Time 0.015610 -2022-12-06 10:24:41,081 - Epoch: [4][ 440/ 1200] Overall Loss 0.531498 Objective Loss 0.531498 LR 0.001000 Time 0.015558 -2022-12-06 10:24:41,215 - Epoch: [4][ 450/ 1200] Overall Loss 0.531284 Objective Loss 0.531284 LR 0.001000 Time 0.015510 -2022-12-06 10:24:41,349 - Epoch: [4][ 460/ 1200] Overall Loss 0.532112 Objective Loss 0.532112 LR 0.001000 Time 0.015464 -2022-12-06 10:24:41,483 - Epoch: [4][ 470/ 1200] Overall Loss 0.533136 Objective Loss 0.533136 LR 0.001000 Time 0.015419 -2022-12-06 10:24:41,617 - Epoch: [4][ 480/ 1200] Overall Loss 0.532382 Objective Loss 0.532382 LR 0.001000 Time 0.015376 -2022-12-06 10:24:41,752 - Epoch: [4][ 490/ 1200] Overall Loss 0.532529 Objective Loss 0.532529 LR 0.001000 Time 0.015335 -2022-12-06 10:24:41,886 - Epoch: [4][ 500/ 1200] Overall Loss 0.532043 Objective Loss 0.532043 LR 0.001000 Time 0.015296 -2022-12-06 10:24:42,020 - Epoch: [4][ 510/ 1200] Overall Loss 0.532327 Objective Loss 0.532327 LR 0.001000 Time 0.015259 -2022-12-06 10:24:42,155 - Epoch: [4][ 520/ 1200] Overall Loss 0.532419 Objective Loss 0.532419 LR 0.001000 Time 0.015224 -2022-12-06 10:24:42,290 - Epoch: [4][ 530/ 1200] Overall Loss 0.532719 Objective Loss 0.532719 LR 0.001000 Time 0.015189 -2022-12-06 10:24:42,424 - Epoch: [4][ 540/ 1200] Overall Loss 0.532664 Objective Loss 0.532664 LR 0.001000 Time 0.015155 -2022-12-06 10:24:42,558 - Epoch: [4][ 550/ 1200] Overall Loss 0.531891 Objective Loss 0.531891 LR 0.001000 Time 0.015123 -2022-12-06 10:24:42,693 - Epoch: [4][ 560/ 1200] Overall Loss 0.532181 Objective Loss 0.532181 LR 0.001000 Time 0.015093 -2022-12-06 10:24:42,827 - Epoch: [4][ 570/ 1200] Overall Loss 0.531823 Objective Loss 0.531823 LR 0.001000 Time 0.015064 -2022-12-06 10:24:42,961 - Epoch: [4][ 580/ 1200] Overall Loss 0.531397 Objective Loss 0.531397 LR 0.001000 Time 0.015034 -2022-12-06 10:24:43,096 - Epoch: [4][ 590/ 1200] Overall Loss 0.531974 Objective Loss 0.531974 LR 0.001000 Time 0.015007 -2022-12-06 10:24:43,231 - Epoch: [4][ 600/ 1200] Overall Loss 0.532042 Objective Loss 0.532042 LR 0.001000 Time 0.014980 -2022-12-06 10:24:43,365 - Epoch: [4][ 610/ 1200] Overall Loss 0.531869 Objective Loss 0.531869 LR 0.001000 Time 0.014955 -2022-12-06 10:24:43,500 - Epoch: [4][ 620/ 1200] Overall Loss 0.531862 Objective Loss 0.531862 LR 0.001000 Time 0.014930 -2022-12-06 10:24:43,636 - Epoch: [4][ 630/ 1200] Overall Loss 0.531544 Objective Loss 0.531544 LR 0.001000 Time 0.014907 -2022-12-06 10:24:43,770 - Epoch: [4][ 640/ 1200] Overall Loss 0.531001 Objective Loss 0.531001 LR 0.001000 Time 0.014884 -2022-12-06 10:24:43,904 - Epoch: [4][ 650/ 1200] Overall Loss 0.531015 Objective Loss 0.531015 LR 0.001000 Time 0.014860 -2022-12-06 10:24:44,039 - Epoch: [4][ 660/ 1200] Overall Loss 0.530800 Objective Loss 0.530800 LR 0.001000 Time 0.014838 -2022-12-06 10:24:44,174 - Epoch: [4][ 670/ 1200] Overall Loss 0.530201 Objective Loss 0.530201 LR 0.001000 Time 0.014818 -2022-12-06 10:24:44,308 - Epoch: [4][ 680/ 1200] Overall Loss 0.530248 Objective Loss 0.530248 LR 0.001000 Time 0.014797 -2022-12-06 10:24:44,442 - Epoch: [4][ 690/ 1200] Overall Loss 0.530141 Objective Loss 0.530141 LR 0.001000 Time 0.014776 -2022-12-06 10:24:44,577 - Epoch: [4][ 700/ 1200] Overall Loss 0.530581 Objective Loss 0.530581 LR 0.001000 Time 0.014757 -2022-12-06 10:24:44,711 - Epoch: [4][ 710/ 1200] Overall Loss 0.529802 Objective Loss 0.529802 LR 0.001000 Time 0.014738 -2022-12-06 10:24:44,847 - Epoch: [4][ 720/ 1200] Overall Loss 0.529378 Objective Loss 0.529378 LR 0.001000 Time 0.014722 -2022-12-06 10:24:44,982 - Epoch: [4][ 730/ 1200] Overall Loss 0.529404 Objective Loss 0.529404 LR 0.001000 Time 0.014704 -2022-12-06 10:24:45,117 - Epoch: [4][ 740/ 1200] Overall Loss 0.529718 Objective Loss 0.529718 LR 0.001000 Time 0.014687 -2022-12-06 10:24:45,252 - Epoch: [4][ 750/ 1200] Overall Loss 0.529377 Objective Loss 0.529377 LR 0.001000 Time 0.014670 -2022-12-06 10:24:45,388 - Epoch: [4][ 760/ 1200] Overall Loss 0.528935 Objective Loss 0.528935 LR 0.001000 Time 0.014656 -2022-12-06 10:24:45,523 - Epoch: [4][ 770/ 1200] Overall Loss 0.528762 Objective Loss 0.528762 LR 0.001000 Time 0.014640 -2022-12-06 10:24:45,657 - Epoch: [4][ 780/ 1200] Overall Loss 0.529110 Objective Loss 0.529110 LR 0.001000 Time 0.014624 -2022-12-06 10:24:45,792 - Epoch: [4][ 790/ 1200] Overall Loss 0.528898 Objective Loss 0.528898 LR 0.001000 Time 0.014608 -2022-12-06 10:24:45,926 - Epoch: [4][ 800/ 1200] Overall Loss 0.529021 Objective Loss 0.529021 LR 0.001000 Time 0.014593 -2022-12-06 10:24:46,060 - Epoch: [4][ 810/ 1200] Overall Loss 0.528845 Objective Loss 0.528845 LR 0.001000 Time 0.014578 -2022-12-06 10:24:46,195 - Epoch: [4][ 820/ 1200] Overall Loss 0.528861 Objective Loss 0.528861 LR 0.001000 Time 0.014564 -2022-12-06 10:24:46,330 - Epoch: [4][ 830/ 1200] Overall Loss 0.528401 Objective Loss 0.528401 LR 0.001000 Time 0.014551 -2022-12-06 10:24:46,464 - Epoch: [4][ 840/ 1200] Overall Loss 0.528620 Objective Loss 0.528620 LR 0.001000 Time 0.014537 -2022-12-06 10:24:46,599 - Epoch: [4][ 850/ 1200] Overall Loss 0.528646 Objective Loss 0.528646 LR 0.001000 Time 0.014524 -2022-12-06 10:24:46,733 - Epoch: [4][ 860/ 1200] Overall Loss 0.529421 Objective Loss 0.529421 LR 0.001000 Time 0.014511 -2022-12-06 10:24:46,867 - Epoch: [4][ 870/ 1200] Overall Loss 0.529559 Objective Loss 0.529559 LR 0.001000 Time 0.014497 -2022-12-06 10:24:47,002 - Epoch: [4][ 880/ 1200] Overall Loss 0.529164 Objective Loss 0.529164 LR 0.001000 Time 0.014485 -2022-12-06 10:24:47,138 - Epoch: [4][ 890/ 1200] Overall Loss 0.529569 Objective Loss 0.529569 LR 0.001000 Time 0.014474 -2022-12-06 10:24:47,273 - Epoch: [4][ 900/ 1200] Overall Loss 0.529602 Objective Loss 0.529602 LR 0.001000 Time 0.014463 -2022-12-06 10:24:47,408 - Epoch: [4][ 910/ 1200] Overall Loss 0.529403 Objective Loss 0.529403 LR 0.001000 Time 0.014452 -2022-12-06 10:24:47,542 - Epoch: [4][ 920/ 1200] Overall Loss 0.528944 Objective Loss 0.528944 LR 0.001000 Time 0.014440 -2022-12-06 10:24:47,676 - Epoch: [4][ 930/ 1200] Overall Loss 0.528849 Objective Loss 0.528849 LR 0.001000 Time 0.014429 -2022-12-06 10:24:47,811 - Epoch: [4][ 940/ 1200] Overall Loss 0.528747 Objective Loss 0.528747 LR 0.001000 Time 0.014418 -2022-12-06 10:24:47,945 - Epoch: [4][ 950/ 1200] Overall Loss 0.528560 Objective Loss 0.528560 LR 0.001000 Time 0.014407 -2022-12-06 10:24:48,079 - Epoch: [4][ 960/ 1200] Overall Loss 0.528317 Objective Loss 0.528317 LR 0.001000 Time 0.014397 -2022-12-06 10:24:48,214 - Epoch: [4][ 970/ 1200] Overall Loss 0.528426 Objective Loss 0.528426 LR 0.001000 Time 0.014386 -2022-12-06 10:24:48,348 - Epoch: [4][ 980/ 1200] Overall Loss 0.528630 Objective Loss 0.528630 LR 0.001000 Time 0.014376 -2022-12-06 10:24:48,483 - Epoch: [4][ 990/ 1200] Overall Loss 0.528782 Objective Loss 0.528782 LR 0.001000 Time 0.014366 -2022-12-06 10:24:48,618 - Epoch: [4][ 1000/ 1200] Overall Loss 0.528299 Objective Loss 0.528299 LR 0.001000 Time 0.014358 -2022-12-06 10:24:48,754 - Epoch: [4][ 1010/ 1200] Overall Loss 0.527912 Objective Loss 0.527912 LR 0.001000 Time 0.014349 -2022-12-06 10:24:48,889 - Epoch: [4][ 1020/ 1200] Overall Loss 0.527846 Objective Loss 0.527846 LR 0.001000 Time 0.014341 -2022-12-06 10:24:49,023 - Epoch: [4][ 1030/ 1200] Overall Loss 0.527688 Objective Loss 0.527688 LR 0.001000 Time 0.014331 -2022-12-06 10:24:49,159 - Epoch: [4][ 1040/ 1200] Overall Loss 0.527327 Objective Loss 0.527327 LR 0.001000 Time 0.014324 -2022-12-06 10:24:49,294 - Epoch: [4][ 1050/ 1200] Overall Loss 0.527342 Objective Loss 0.527342 LR 0.001000 Time 0.014315 -2022-12-06 10:24:49,428 - Epoch: [4][ 1060/ 1200] Overall Loss 0.527482 Objective Loss 0.527482 LR 0.001000 Time 0.014307 -2022-12-06 10:24:49,564 - Epoch: [4][ 1070/ 1200] Overall Loss 0.527248 Objective Loss 0.527248 LR 0.001000 Time 0.014299 -2022-12-06 10:24:49,700 - Epoch: [4][ 1080/ 1200] Overall Loss 0.527235 Objective Loss 0.527235 LR 0.001000 Time 0.014293 -2022-12-06 10:24:49,835 - Epoch: [4][ 1090/ 1200] Overall Loss 0.526743 Objective Loss 0.526743 LR 0.001000 Time 0.014285 -2022-12-06 10:24:49,970 - Epoch: [4][ 1100/ 1200] Overall Loss 0.526366 Objective Loss 0.526366 LR 0.001000 Time 0.014277 -2022-12-06 10:24:50,105 - Epoch: [4][ 1110/ 1200] Overall Loss 0.526095 Objective Loss 0.526095 LR 0.001000 Time 0.014270 -2022-12-06 10:24:50,240 - Epoch: [4][ 1120/ 1200] Overall Loss 0.525785 Objective Loss 0.525785 LR 0.001000 Time 0.014263 -2022-12-06 10:24:50,375 - Epoch: [4][ 1130/ 1200] Overall Loss 0.525450 Objective Loss 0.525450 LR 0.001000 Time 0.014255 -2022-12-06 10:24:50,510 - Epoch: [4][ 1140/ 1200] Overall Loss 0.525584 Objective Loss 0.525584 LR 0.001000 Time 0.014248 -2022-12-06 10:24:50,645 - Epoch: [4][ 1150/ 1200] Overall Loss 0.525701 Objective Loss 0.525701 LR 0.001000 Time 0.014242 -2022-12-06 10:24:50,781 - Epoch: [4][ 1160/ 1200] Overall Loss 0.525867 Objective Loss 0.525867 LR 0.001000 Time 0.014235 -2022-12-06 10:24:50,916 - Epoch: [4][ 1170/ 1200] Overall Loss 0.525664 Objective Loss 0.525664 LR 0.001000 Time 0.014229 -2022-12-06 10:24:51,051 - Epoch: [4][ 1180/ 1200] Overall Loss 0.525711 Objective Loss 0.525711 LR 0.001000 Time 0.014222 -2022-12-06 10:24:51,186 - Epoch: [4][ 1190/ 1200] Overall Loss 0.525957 Objective Loss 0.525957 LR 0.001000 Time 0.014216 -2022-12-06 10:24:51,372 - Epoch: [4][ 1200/ 1200] Overall Loss 0.526199 Objective Loss 0.526199 Top1 74.267782 Top5 96.025105 LR 0.001000 Time 0.014252 -2022-12-06 10:24:51,459 - --- validate (epoch=4)----------- -2022-12-06 10:24:51,459 - 34129 samples (256 per mini-batch) -2022-12-06 10:24:51,849 - Epoch: [4][ 10/ 134] Loss 0.479149 Top1 74.687500 Top5 96.640625 -2022-12-06 10:24:51,946 - Epoch: [4][ 20/ 134] Loss 0.496355 Top1 73.808594 Top5 96.191406 -2022-12-06 10:24:52,044 - Epoch: [4][ 30/ 134] Loss 0.507206 Top1 73.372396 Top5 96.145833 -2022-12-06 10:24:52,138 - Epoch: [4][ 40/ 134] Loss 0.505377 Top1 73.408203 Top5 96.269531 -2022-12-06 10:24:52,231 - Epoch: [4][ 50/ 134] Loss 0.501442 Top1 73.382812 Top5 96.085938 -2022-12-06 10:24:52,325 - Epoch: [4][ 60/ 134] Loss 0.498974 Top1 73.444010 Top5 96.132812 -2022-12-06 10:24:52,417 - Epoch: [4][ 70/ 134] Loss 0.502212 Top1 73.381696 Top5 96.104911 -2022-12-06 10:24:52,513 - Epoch: [4][ 80/ 134] Loss 0.499400 Top1 73.408203 Top5 96.113281 -2022-12-06 10:24:52,604 - Epoch: [4][ 90/ 134] Loss 0.496909 Top1 73.537326 Top5 96.189236 -2022-12-06 10:24:52,693 - Epoch: [4][ 100/ 134] Loss 0.498831 Top1 73.484375 Top5 96.148438 -2022-12-06 10:24:52,785 - Epoch: [4][ 110/ 134] Loss 0.498777 Top1 73.313210 Top5 96.147017 -2022-12-06 10:24:52,879 - Epoch: [4][ 120/ 134] Loss 0.499587 Top1 73.313802 Top5 96.126302 -2022-12-06 10:24:52,970 - Epoch: [4][ 130/ 134] Loss 0.497389 Top1 73.392428 Top5 96.156851 -2022-12-06 10:24:52,992 - Epoch: [4][ 134/ 134] Loss 0.495995 Top1 73.459521 Top5 96.155762 -2022-12-06 10:24:53,079 - ==> Top1: 73.460 Top5: 96.156 Loss: 0.496 - -2022-12-06 10:24:53,080 - ==> Confusion: -[[ 875 1 5 2 14 5 1 0 5 64 0 3 2 2 9 1 0 0 5 0 2] - [ 3 911 1 0 24 12 1 17 6 0 2 6 4 0 13 1 11 2 10 2 1] - [ 12 10 959 22 11 3 25 19 2 3 9 3 3 4 4 1 1 2 6 0 4] - [ 4 8 29 874 1 4 0 2 5 0 16 2 5 0 38 1 4 2 22 1 2] - [ 11 6 2 0 956 5 0 0 3 7 0 6 3 0 7 3 4 2 1 0 4] - [ 6 148 3 3 20 765 0 34 4 4 2 29 7 17 7 0 6 1 4 5 4] - [ 1 15 32 7 7 1 1000 9 3 1 2 6 2 0 0 10 0 2 2 17 1] - [ 3 37 20 3 3 33 5 849 3 4 2 7 3 1 0 0 4 0 60 13 4] - [ 11 5 0 0 0 1 0 3 920 66 10 3 2 9 27 1 2 0 3 0 1] - [ 120 0 6 0 13 4 0 2 29 799 1 0 1 10 9 0 0 0 0 0 7] - [ 1 6 12 14 5 6 3 5 19 2 897 5 1 8 10 0 1 0 23 0 1] - [ 5 6 2 0 3 16 2 6 5 0 0 938 28 5 2 6 7 7 5 7 1] - [ 1 2 1 18 3 2 1 4 2 0 0 102 777 2 14 6 4 17 2 3 8] - [ 8 8 2 3 27 23 0 2 18 26 18 18 4 822 16 0 8 1 0 7 12] - [ 11 5 1 13 16 1 0 1 22 9 4 2 2 0 1028 0 4 0 7 0 4] - [ 4 3 11 0 12 1 8 0 1 0 0 24 13 1 4 917 15 15 2 4 8] - [ 6 11 4 1 33 3 1 0 2 1 0 9 1 0 4 5 979 1 3 2 6] - [ 6 3 5 13 0 1 4 1 5 1 0 29 75 1 9 8 4 862 3 0 6] - [ 3 11 10 16 4 3 0 33 5 0 7 4 6 0 18 0 1 0 884 2 1] - [ 1 10 6 1 1 12 6 14 0 0 1 47 6 2 1 5 9 6 2 938 12] - [ 387 530 379 227 551 230 53 185 165 160 226 294 450 267 508 143 461 61 384 444 7121]] - -2022-12-06 10:24:53,641 - ==> Best [Top1: 73.460 Top5: 96.156 Sparsity:0.00 Params: 5376 on epoch: 4] -2022-12-06 10:24:53,641 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:24:53,648 - - -2022-12-06 10:24:53,648 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:24:54,512 - Epoch: [5][ 10/ 1200] Overall Loss 0.464674 Objective Loss 0.464674 LR 0.001000 Time 0.086384 -2022-12-06 10:24:54,660 - Epoch: [5][ 20/ 1200] Overall Loss 0.469439 Objective Loss 0.469439 LR 0.001000 Time 0.050555 -2022-12-06 10:24:54,798 - Epoch: [5][ 30/ 1200] Overall Loss 0.476316 Objective Loss 0.476316 LR 0.001000 Time 0.038295 -2022-12-06 10:24:54,932 - Epoch: [5][ 40/ 1200] Overall Loss 0.486924 Objective Loss 0.486924 LR 0.001000 Time 0.032054 -2022-12-06 10:24:55,066 - Epoch: [5][ 50/ 1200] Overall Loss 0.489994 Objective Loss 0.489994 LR 0.001000 Time 0.028313 -2022-12-06 10:24:55,200 - Epoch: [5][ 60/ 1200] Overall Loss 0.494996 Objective Loss 0.494996 LR 0.001000 Time 0.025814 -2022-12-06 10:24:55,334 - Epoch: [5][ 70/ 1200] Overall Loss 0.493443 Objective Loss 0.493443 LR 0.001000 Time 0.024039 -2022-12-06 10:24:55,468 - Epoch: [5][ 80/ 1200] Overall Loss 0.490623 Objective Loss 0.490623 LR 0.001000 Time 0.022702 -2022-12-06 10:24:55,602 - Epoch: [5][ 90/ 1200] Overall Loss 0.495937 Objective Loss 0.495937 LR 0.001000 Time 0.021663 -2022-12-06 10:24:55,736 - Epoch: [5][ 100/ 1200] Overall Loss 0.495524 Objective Loss 0.495524 LR 0.001000 Time 0.020827 -2022-12-06 10:24:55,870 - Epoch: [5][ 110/ 1200] Overall Loss 0.498995 Objective Loss 0.498995 LR 0.001000 Time 0.020147 -2022-12-06 10:24:56,004 - Epoch: [5][ 120/ 1200] Overall Loss 0.499811 Objective Loss 0.499811 LR 0.001000 Time 0.019581 -2022-12-06 10:24:56,137 - Epoch: [5][ 130/ 1200] Overall Loss 0.499039 Objective Loss 0.499039 LR 0.001000 Time 0.019098 -2022-12-06 10:24:56,271 - Epoch: [5][ 140/ 1200] Overall Loss 0.498756 Objective Loss 0.498756 LR 0.001000 Time 0.018685 -2022-12-06 10:24:56,404 - Epoch: [5][ 150/ 1200] Overall Loss 0.497971 Objective Loss 0.497971 LR 0.001000 Time 0.018327 -2022-12-06 10:24:56,537 - Epoch: [5][ 160/ 1200] Overall Loss 0.497924 Objective Loss 0.497924 LR 0.001000 Time 0.018010 -2022-12-06 10:24:56,671 - Epoch: [5][ 170/ 1200] Overall Loss 0.495743 Objective Loss 0.495743 LR 0.001000 Time 0.017737 -2022-12-06 10:24:56,805 - Epoch: [5][ 180/ 1200] Overall Loss 0.495714 Objective Loss 0.495714 LR 0.001000 Time 0.017490 -2022-12-06 10:24:56,939 - Epoch: [5][ 190/ 1200] Overall Loss 0.496774 Objective Loss 0.496774 LR 0.001000 Time 0.017271 -2022-12-06 10:24:57,072 - Epoch: [5][ 200/ 1200] Overall Loss 0.495270 Objective Loss 0.495270 LR 0.001000 Time 0.017072 -2022-12-06 10:24:57,206 - Epoch: [5][ 210/ 1200] Overall Loss 0.493047 Objective Loss 0.493047 LR 0.001000 Time 0.016895 -2022-12-06 10:24:57,339 - Epoch: [5][ 220/ 1200] Overall Loss 0.492509 Objective Loss 0.492509 LR 0.001000 Time 0.016731 -2022-12-06 10:24:57,473 - Epoch: [5][ 230/ 1200] Overall Loss 0.492224 Objective Loss 0.492224 LR 0.001000 Time 0.016584 -2022-12-06 10:24:57,607 - Epoch: [5][ 240/ 1200] Overall Loss 0.492867 Objective Loss 0.492867 LR 0.001000 Time 0.016449 -2022-12-06 10:24:57,741 - Epoch: [5][ 250/ 1200] Overall Loss 0.493569 Objective Loss 0.493569 LR 0.001000 Time 0.016324 -2022-12-06 10:24:57,875 - Epoch: [5][ 260/ 1200] Overall Loss 0.492700 Objective Loss 0.492700 LR 0.001000 Time 0.016209 -2022-12-06 10:24:58,009 - Epoch: [5][ 270/ 1200] Overall Loss 0.491796 Objective Loss 0.491796 LR 0.001000 Time 0.016104 -2022-12-06 10:24:58,142 - Epoch: [5][ 280/ 1200] Overall Loss 0.491358 Objective Loss 0.491358 LR 0.001000 Time 0.016003 -2022-12-06 10:24:58,277 - Epoch: [5][ 290/ 1200] Overall Loss 0.490177 Objective Loss 0.490177 LR 0.001000 Time 0.015915 -2022-12-06 10:24:58,411 - Epoch: [5][ 300/ 1200] Overall Loss 0.489676 Objective Loss 0.489676 LR 0.001000 Time 0.015828 -2022-12-06 10:24:58,545 - Epoch: [5][ 310/ 1200] Overall Loss 0.489820 Objective Loss 0.489820 LR 0.001000 Time 0.015748 -2022-12-06 10:24:58,678 - Epoch: [5][ 320/ 1200] Overall Loss 0.490073 Objective Loss 0.490073 LR 0.001000 Time 0.015672 -2022-12-06 10:24:58,812 - Epoch: [5][ 330/ 1200] Overall Loss 0.489669 Objective Loss 0.489669 LR 0.001000 Time 0.015601 -2022-12-06 10:24:58,946 - Epoch: [5][ 340/ 1200] Overall Loss 0.490148 Objective Loss 0.490148 LR 0.001000 Time 0.015533 -2022-12-06 10:24:59,083 - Epoch: [5][ 350/ 1200] Overall Loss 0.489328 Objective Loss 0.489328 LR 0.001000 Time 0.015481 -2022-12-06 10:24:59,224 - Epoch: [5][ 360/ 1200] Overall Loss 0.490959 Objective Loss 0.490959 LR 0.001000 Time 0.015441 -2022-12-06 10:24:59,363 - Epoch: [5][ 370/ 1200] Overall Loss 0.489630 Objective Loss 0.489630 LR 0.001000 Time 0.015398 -2022-12-06 10:24:59,506 - Epoch: [5][ 380/ 1200] Overall Loss 0.489827 Objective Loss 0.489827 LR 0.001000 Time 0.015367 -2022-12-06 10:24:59,645 - Epoch: [5][ 390/ 1200] Overall Loss 0.489097 Objective Loss 0.489097 LR 0.001000 Time 0.015330 -2022-12-06 10:24:59,788 - Epoch: [5][ 400/ 1200] Overall Loss 0.488814 Objective Loss 0.488814 LR 0.001000 Time 0.015302 -2022-12-06 10:24:59,927 - Epoch: [5][ 410/ 1200] Overall Loss 0.489159 Objective Loss 0.489159 LR 0.001000 Time 0.015268 -2022-12-06 10:25:00,071 - Epoch: [5][ 420/ 1200] Overall Loss 0.489044 Objective Loss 0.489044 LR 0.001000 Time 0.015244 -2022-12-06 10:25:00,211 - Epoch: [5][ 430/ 1200] Overall Loss 0.487800 Objective Loss 0.487800 LR 0.001000 Time 0.015215 -2022-12-06 10:25:00,352 - Epoch: [5][ 440/ 1200] Overall Loss 0.487186 Objective Loss 0.487186 LR 0.001000 Time 0.015189 -2022-12-06 10:25:00,492 - Epoch: [5][ 450/ 1200] Overall Loss 0.486257 Objective Loss 0.486257 LR 0.001000 Time 0.015160 -2022-12-06 10:25:00,635 - Epoch: [5][ 460/ 1200] Overall Loss 0.486375 Objective Loss 0.486375 LR 0.001000 Time 0.015141 -2022-12-06 10:25:00,775 - Epoch: [5][ 470/ 1200] Overall Loss 0.487489 Objective Loss 0.487489 LR 0.001000 Time 0.015115 -2022-12-06 10:25:00,917 - Epoch: [5][ 480/ 1200] Overall Loss 0.488044 Objective Loss 0.488044 LR 0.001000 Time 0.015096 -2022-12-06 10:25:01,058 - Epoch: [5][ 490/ 1200] Overall Loss 0.488476 Objective Loss 0.488476 LR 0.001000 Time 0.015073 -2022-12-06 10:25:01,193 - Epoch: [5][ 500/ 1200] Overall Loss 0.488379 Objective Loss 0.488379 LR 0.001000 Time 0.015042 -2022-12-06 10:25:01,328 - Epoch: [5][ 510/ 1200] Overall Loss 0.488744 Objective Loss 0.488744 LR 0.001000 Time 0.015010 -2022-12-06 10:25:01,461 - Epoch: [5][ 520/ 1200] Overall Loss 0.488443 Objective Loss 0.488443 LR 0.001000 Time 0.014976 -2022-12-06 10:25:01,595 - Epoch: [5][ 530/ 1200] Overall Loss 0.488207 Objective Loss 0.488207 LR 0.001000 Time 0.014946 -2022-12-06 10:25:01,729 - Epoch: [5][ 540/ 1200] Overall Loss 0.488183 Objective Loss 0.488183 LR 0.001000 Time 0.014916 -2022-12-06 10:25:01,864 - Epoch: [5][ 550/ 1200] Overall Loss 0.488048 Objective Loss 0.488048 LR 0.001000 Time 0.014890 -2022-12-06 10:25:01,998 - Epoch: [5][ 560/ 1200] Overall Loss 0.487731 Objective Loss 0.487731 LR 0.001000 Time 0.014862 -2022-12-06 10:25:02,132 - Epoch: [5][ 570/ 1200] Overall Loss 0.487263 Objective Loss 0.487263 LR 0.001000 Time 0.014835 -2022-12-06 10:25:02,268 - Epoch: [5][ 580/ 1200] Overall Loss 0.487423 Objective Loss 0.487423 LR 0.001000 Time 0.014813 -2022-12-06 10:25:02,402 - Epoch: [5][ 590/ 1200] Overall Loss 0.486346 Objective Loss 0.486346 LR 0.001000 Time 0.014788 -2022-12-06 10:25:02,536 - Epoch: [5][ 600/ 1200] Overall Loss 0.485927 Objective Loss 0.485927 LR 0.001000 Time 0.014763 -2022-12-06 10:25:02,670 - Epoch: [5][ 610/ 1200] Overall Loss 0.485487 Objective Loss 0.485487 LR 0.001000 Time 0.014741 -2022-12-06 10:25:02,806 - Epoch: [5][ 620/ 1200] Overall Loss 0.484851 Objective Loss 0.484851 LR 0.001000 Time 0.014720 -2022-12-06 10:25:02,941 - Epoch: [5][ 630/ 1200] Overall Loss 0.484954 Objective Loss 0.484954 LR 0.001000 Time 0.014701 -2022-12-06 10:25:03,077 - Epoch: [5][ 640/ 1200] Overall Loss 0.484933 Objective Loss 0.484933 LR 0.001000 Time 0.014682 -2022-12-06 10:25:03,211 - Epoch: [5][ 650/ 1200] Overall Loss 0.484711 Objective Loss 0.484711 LR 0.001000 Time 0.014662 -2022-12-06 10:25:03,345 - Epoch: [5][ 660/ 1200] Overall Loss 0.484639 Objective Loss 0.484639 LR 0.001000 Time 0.014642 -2022-12-06 10:25:03,481 - Epoch: [5][ 670/ 1200] Overall Loss 0.484956 Objective Loss 0.484956 LR 0.001000 Time 0.014625 -2022-12-06 10:25:03,618 - Epoch: [5][ 680/ 1200] Overall Loss 0.485014 Objective Loss 0.485014 LR 0.001000 Time 0.014611 -2022-12-06 10:25:03,752 - Epoch: [5][ 690/ 1200] Overall Loss 0.485273 Objective Loss 0.485273 LR 0.001000 Time 0.014593 -2022-12-06 10:25:03,887 - Epoch: [5][ 700/ 1200] Overall Loss 0.485368 Objective Loss 0.485368 LR 0.001000 Time 0.014577 -2022-12-06 10:25:04,022 - Epoch: [5][ 710/ 1200] Overall Loss 0.484927 Objective Loss 0.484927 LR 0.001000 Time 0.014560 -2022-12-06 10:25:04,157 - Epoch: [5][ 720/ 1200] Overall Loss 0.484474 Objective Loss 0.484474 LR 0.001000 Time 0.014544 -2022-12-06 10:25:04,293 - Epoch: [5][ 730/ 1200] Overall Loss 0.484799 Objective Loss 0.484799 LR 0.001000 Time 0.014530 -2022-12-06 10:25:04,429 - Epoch: [5][ 740/ 1200] Overall Loss 0.485154 Objective Loss 0.485154 LR 0.001000 Time 0.014517 -2022-12-06 10:25:04,564 - Epoch: [5][ 750/ 1200] Overall Loss 0.485023 Objective Loss 0.485023 LR 0.001000 Time 0.014503 -2022-12-06 10:25:04,702 - Epoch: [5][ 760/ 1200] Overall Loss 0.485232 Objective Loss 0.485232 LR 0.001000 Time 0.014493 -2022-12-06 10:25:04,840 - Epoch: [5][ 770/ 1200] Overall Loss 0.484842 Objective Loss 0.484842 LR 0.001000 Time 0.014484 -2022-12-06 10:25:04,977 - Epoch: [5][ 780/ 1200] Overall Loss 0.484899 Objective Loss 0.484899 LR 0.001000 Time 0.014474 -2022-12-06 10:25:05,115 - Epoch: [5][ 790/ 1200] Overall Loss 0.484748 Objective Loss 0.484748 LR 0.001000 Time 0.014464 -2022-12-06 10:25:05,253 - Epoch: [5][ 800/ 1200] Overall Loss 0.485072 Objective Loss 0.485072 LR 0.001000 Time 0.014455 -2022-12-06 10:25:05,389 - Epoch: [5][ 810/ 1200] Overall Loss 0.485306 Objective Loss 0.485306 LR 0.001000 Time 0.014444 -2022-12-06 10:25:05,527 - Epoch: [5][ 820/ 1200] Overall Loss 0.485686 Objective Loss 0.485686 LR 0.001000 Time 0.014436 -2022-12-06 10:25:05,665 - Epoch: [5][ 830/ 1200] Overall Loss 0.484928 Objective Loss 0.484928 LR 0.001000 Time 0.014428 -2022-12-06 10:25:05,804 - Epoch: [5][ 840/ 1200] Overall Loss 0.484704 Objective Loss 0.484704 LR 0.001000 Time 0.014420 -2022-12-06 10:25:05,942 - Epoch: [5][ 850/ 1200] Overall Loss 0.484186 Objective Loss 0.484186 LR 0.001000 Time 0.014412 -2022-12-06 10:25:06,079 - Epoch: [5][ 860/ 1200] Overall Loss 0.483935 Objective Loss 0.483935 LR 0.001000 Time 0.014403 -2022-12-06 10:25:06,217 - Epoch: [5][ 870/ 1200] Overall Loss 0.483787 Objective Loss 0.483787 LR 0.001000 Time 0.014396 -2022-12-06 10:25:06,353 - Epoch: [5][ 880/ 1200] Overall Loss 0.483410 Objective Loss 0.483410 LR 0.001000 Time 0.014387 -2022-12-06 10:25:06,489 - Epoch: [5][ 890/ 1200] Overall Loss 0.483751 Objective Loss 0.483751 LR 0.001000 Time 0.014378 -2022-12-06 10:25:06,627 - Epoch: [5][ 900/ 1200] Overall Loss 0.483377 Objective Loss 0.483377 LR 0.001000 Time 0.014370 -2022-12-06 10:25:06,764 - Epoch: [5][ 910/ 1200] Overall Loss 0.482924 Objective Loss 0.482924 LR 0.001000 Time 0.014363 -2022-12-06 10:25:06,901 - Epoch: [5][ 920/ 1200] Overall Loss 0.482268 Objective Loss 0.482268 LR 0.001000 Time 0.014355 -2022-12-06 10:25:07,039 - Epoch: [5][ 930/ 1200] Overall Loss 0.482162 Objective Loss 0.482162 LR 0.001000 Time 0.014348 -2022-12-06 10:25:07,176 - Epoch: [5][ 940/ 1200] Overall Loss 0.481974 Objective Loss 0.481974 LR 0.001000 Time 0.014341 -2022-12-06 10:25:07,313 - Epoch: [5][ 950/ 1200] Overall Loss 0.481899 Objective Loss 0.481899 LR 0.001000 Time 0.014334 -2022-12-06 10:25:07,451 - Epoch: [5][ 960/ 1200] Overall Loss 0.481600 Objective Loss 0.481600 LR 0.001000 Time 0.014328 -2022-12-06 10:25:07,588 - Epoch: [5][ 970/ 1200] Overall Loss 0.481809 Objective Loss 0.481809 LR 0.001000 Time 0.014321 -2022-12-06 10:25:07,726 - Epoch: [5][ 980/ 1200] Overall Loss 0.481689 Objective Loss 0.481689 LR 0.001000 Time 0.014315 -2022-12-06 10:25:07,864 - Epoch: [5][ 990/ 1200] Overall Loss 0.481740 Objective Loss 0.481740 LR 0.001000 Time 0.014309 -2022-12-06 10:25:08,002 - Epoch: [5][ 1000/ 1200] Overall Loss 0.481571 Objective Loss 0.481571 LR 0.001000 Time 0.014303 -2022-12-06 10:25:08,139 - Epoch: [5][ 1010/ 1200] Overall Loss 0.481159 Objective Loss 0.481159 LR 0.001000 Time 0.014297 -2022-12-06 10:25:08,277 - Epoch: [5][ 1020/ 1200] Overall Loss 0.481414 Objective Loss 0.481414 LR 0.001000 Time 0.014292 -2022-12-06 10:25:08,414 - Epoch: [5][ 1030/ 1200] Overall Loss 0.481397 Objective Loss 0.481397 LR 0.001000 Time 0.014285 -2022-12-06 10:25:08,551 - Epoch: [5][ 1040/ 1200] Overall Loss 0.481073 Objective Loss 0.481073 LR 0.001000 Time 0.014280 -2022-12-06 10:25:08,689 - Epoch: [5][ 1050/ 1200] Overall Loss 0.480872 Objective Loss 0.480872 LR 0.001000 Time 0.014275 -2022-12-06 10:25:08,827 - Epoch: [5][ 1060/ 1200] Overall Loss 0.480496 Objective Loss 0.480496 LR 0.001000 Time 0.014269 -2022-12-06 10:25:08,964 - Epoch: [5][ 1070/ 1200] Overall Loss 0.480092 Objective Loss 0.480092 LR 0.001000 Time 0.014264 -2022-12-06 10:25:09,100 - Epoch: [5][ 1080/ 1200] Overall Loss 0.480307 Objective Loss 0.480307 LR 0.001000 Time 0.014258 -2022-12-06 10:25:09,238 - Epoch: [5][ 1090/ 1200] Overall Loss 0.480333 Objective Loss 0.480333 LR 0.001000 Time 0.014253 -2022-12-06 10:25:09,376 - Epoch: [5][ 1100/ 1200] Overall Loss 0.480360 Objective Loss 0.480360 LR 0.001000 Time 0.014248 -2022-12-06 10:25:09,513 - Epoch: [5][ 1110/ 1200] Overall Loss 0.480342 Objective Loss 0.480342 LR 0.001000 Time 0.014243 -2022-12-06 10:25:09,650 - Epoch: [5][ 1120/ 1200] Overall Loss 0.480107 Objective Loss 0.480107 LR 0.001000 Time 0.014238 -2022-12-06 10:25:09,786 - Epoch: [5][ 1130/ 1200] Overall Loss 0.480261 Objective Loss 0.480261 LR 0.001000 Time 0.014232 -2022-12-06 10:25:09,924 - Epoch: [5][ 1140/ 1200] Overall Loss 0.480233 Objective Loss 0.480233 LR 0.001000 Time 0.014227 -2022-12-06 10:25:10,062 - Epoch: [5][ 1150/ 1200] Overall Loss 0.479977 Objective Loss 0.479977 LR 0.001000 Time 0.014223 -2022-12-06 10:25:10,200 - Epoch: [5][ 1160/ 1200] Overall Loss 0.479962 Objective Loss 0.479962 LR 0.001000 Time 0.014219 -2022-12-06 10:25:10,337 - Epoch: [5][ 1170/ 1200] Overall Loss 0.480452 Objective Loss 0.480452 LR 0.001000 Time 0.014215 -2022-12-06 10:25:10,475 - Epoch: [5][ 1180/ 1200] Overall Loss 0.480462 Objective Loss 0.480462 LR 0.001000 Time 0.014210 -2022-12-06 10:25:10,611 - Epoch: [5][ 1190/ 1200] Overall Loss 0.480546 Objective Loss 0.480546 LR 0.001000 Time 0.014205 -2022-12-06 10:25:10,799 - Epoch: [5][ 1200/ 1200] Overall Loss 0.480651 Objective Loss 0.480651 Top1 73.012552 Top5 96.025105 LR 0.001000 Time 0.014243 -2022-12-06 10:25:10,887 - --- validate (epoch=5)----------- -2022-12-06 10:25:10,887 - 34129 samples (256 per mini-batch) -2022-12-06 10:25:11,296 - Epoch: [5][ 10/ 134] Loss 0.456080 Top1 76.601562 Top5 96.484375 -2022-12-06 10:25:11,389 - Epoch: [5][ 20/ 134] Loss 0.454280 Top1 76.816406 Top5 96.757812 -2022-12-06 10:25:11,490 - Epoch: [5][ 30/ 134] Loss 0.450962 Top1 77.109375 Top5 96.861979 -2022-12-06 10:25:11,595 - Epoch: [5][ 40/ 134] Loss 0.442858 Top1 77.373047 Top5 96.767578 -2022-12-06 10:25:11,695 - Epoch: [5][ 50/ 134] Loss 0.448536 Top1 77.140625 Top5 96.789062 -2022-12-06 10:25:11,800 - Epoch: [5][ 60/ 134] Loss 0.453538 Top1 76.979167 Top5 96.744792 -2022-12-06 10:25:11,899 - Epoch: [5][ 70/ 134] Loss 0.458082 Top1 76.819196 Top5 96.746652 -2022-12-06 10:25:12,004 - Epoch: [5][ 80/ 134] Loss 0.455315 Top1 76.816406 Top5 96.816406 -2022-12-06 10:25:12,097 - Epoch: [5][ 90/ 134] Loss 0.458823 Top1 76.783854 Top5 96.744792 -2022-12-06 10:25:12,197 - Epoch: [5][ 100/ 134] Loss 0.459809 Top1 76.691406 Top5 96.746094 -2022-12-06 10:25:12,301 - Epoch: [5][ 110/ 134] Loss 0.459156 Top1 76.704545 Top5 96.732955 -2022-12-06 10:25:12,403 - Epoch: [5][ 120/ 134] Loss 0.457865 Top1 76.764323 Top5 96.708984 -2022-12-06 10:25:12,503 - Epoch: [5][ 130/ 134] Loss 0.460241 Top1 76.703726 Top5 96.712740 -2022-12-06 10:25:12,523 - Epoch: [5][ 134/ 134] Loss 0.459459 Top1 76.720677 Top5 96.706613 -2022-12-06 10:25:12,612 - ==> Top1: 76.721 Top5: 96.707 Loss: 0.459 - -2022-12-06 10:25:12,613 - ==> Confusion: -[[ 854 0 2 0 11 11 1 0 6 79 0 6 5 0 4 2 2 2 2 0 9] - [ 0 848 2 1 22 57 3 11 5 1 2 9 9 0 7 2 17 1 16 2 12] - [ 7 5 932 19 7 8 39 10 3 4 10 6 4 9 2 4 4 4 6 6 14] - [ 2 3 24 868 2 10 1 3 5 0 20 1 10 1 26 2 3 6 20 1 12] - [ 11 4 0 0 937 10 2 0 3 5 1 3 2 3 9 4 13 3 1 2 7] - [ 1 39 0 2 7 892 6 16 7 3 4 29 9 26 2 1 4 1 1 8 11] - [ 0 8 18 2 2 5 1010 7 1 0 10 7 4 0 0 11 3 2 2 22 4] - [ 0 18 16 1 3 67 9 812 3 3 12 6 5 1 0 2 1 1 66 14 14] - [ 8 5 1 0 0 3 0 0 948 56 3 6 5 3 22 0 1 0 1 2 0] - [ 79 2 0 0 6 5 0 2 35 840 2 2 2 7 8 0 2 0 1 0 8] - [ 1 3 3 10 1 3 1 3 24 2 916 4 3 8 6 0 4 0 19 0 8] - [ 2 3 1 0 2 19 2 2 1 1 1 884 80 15 0 11 6 6 3 9 3] - [ 2 1 0 2 1 6 0 0 3 1 0 46 846 5 4 12 6 19 2 6 7] - [ 1 7 0 2 8 19 0 5 20 40 13 8 6 862 5 2 6 0 0 2 17] - [ 12 5 3 13 13 4 0 0 25 7 2 2 9 2 1004 0 7 4 7 0 11] - [ 2 3 2 0 5 0 8 0 1 1 1 8 9 5 1 954 19 13 1 4 6] - [ 3 7 1 1 8 4 0 0 5 1 0 4 3 2 1 9 1009 1 1 5 7] - [ 6 2 0 3 0 4 4 1 5 3 2 16 64 3 3 20 3 892 1 0 4] - [ 4 6 6 17 0 5 0 24 8 0 12 6 8 1 17 0 2 0 880 3 9] - [ 0 5 1 0 1 12 5 6 0 1 0 18 14 6 0 5 10 4 0 978 14] - [ 253 337 179 92 319 375 88 98 157 151 228 208 521 347 285 236 550 83 296 405 8018]] - -2022-12-06 10:25:13,277 - ==> Best [Top1: 76.721 Top5: 96.707 Sparsity:0.00 Params: 5376 on epoch: 5] -2022-12-06 10:25:13,277 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:25:13,284 - - -2022-12-06 10:25:13,284 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:25:14,144 - Epoch: [6][ 10/ 1200] Overall Loss 0.455745 Objective Loss 0.455745 LR 0.001000 Time 0.085940 -2022-12-06 10:25:14,281 - Epoch: [6][ 20/ 1200] Overall Loss 0.429553 Objective Loss 0.429553 LR 0.001000 Time 0.049810 -2022-12-06 10:25:14,414 - Epoch: [6][ 30/ 1200] Overall Loss 0.432428 Objective Loss 0.432428 LR 0.001000 Time 0.037625 -2022-12-06 10:25:14,547 - Epoch: [6][ 40/ 1200] Overall Loss 0.442299 Objective Loss 0.442299 LR 0.001000 Time 0.031535 -2022-12-06 10:25:14,680 - Epoch: [6][ 50/ 1200] Overall Loss 0.444089 Objective Loss 0.444089 LR 0.001000 Time 0.027869 -2022-12-06 10:25:14,813 - Epoch: [6][ 60/ 1200] Overall Loss 0.446938 Objective Loss 0.446938 LR 0.001000 Time 0.025429 -2022-12-06 10:25:14,946 - Epoch: [6][ 70/ 1200] Overall Loss 0.447547 Objective Loss 0.447547 LR 0.001000 Time 0.023688 -2022-12-06 10:25:15,078 - Epoch: [6][ 80/ 1200] Overall Loss 0.447731 Objective Loss 0.447731 LR 0.001000 Time 0.022376 -2022-12-06 10:25:15,211 - Epoch: [6][ 90/ 1200] Overall Loss 0.446939 Objective Loss 0.446939 LR 0.001000 Time 0.021356 -2022-12-06 10:25:15,344 - Epoch: [6][ 100/ 1200] Overall Loss 0.449881 Objective Loss 0.449881 LR 0.001000 Time 0.020546 -2022-12-06 10:25:15,477 - Epoch: [6][ 110/ 1200] Overall Loss 0.453242 Objective Loss 0.453242 LR 0.001000 Time 0.019887 -2022-12-06 10:25:15,610 - Epoch: [6][ 120/ 1200] Overall Loss 0.451330 Objective Loss 0.451330 LR 0.001000 Time 0.019336 -2022-12-06 10:25:15,744 - Epoch: [6][ 130/ 1200] Overall Loss 0.450259 Objective Loss 0.450259 LR 0.001000 Time 0.018872 -2022-12-06 10:25:15,877 - Epoch: [6][ 140/ 1200] Overall Loss 0.449723 Objective Loss 0.449723 LR 0.001000 Time 0.018472 -2022-12-06 10:25:16,011 - Epoch: [6][ 150/ 1200] Overall Loss 0.451819 Objective Loss 0.451819 LR 0.001000 Time 0.018130 -2022-12-06 10:25:16,144 - Epoch: [6][ 160/ 1200] Overall Loss 0.453614 Objective Loss 0.453614 LR 0.001000 Time 0.017826 -2022-12-06 10:25:16,277 - Epoch: [6][ 170/ 1200] Overall Loss 0.455875 Objective Loss 0.455875 LR 0.001000 Time 0.017556 -2022-12-06 10:25:16,410 - Epoch: [6][ 180/ 1200] Overall Loss 0.454002 Objective Loss 0.454002 LR 0.001000 Time 0.017318 -2022-12-06 10:25:16,544 - Epoch: [6][ 190/ 1200] Overall Loss 0.453353 Objective Loss 0.453353 LR 0.001000 Time 0.017108 -2022-12-06 10:25:16,678 - Epoch: [6][ 200/ 1200] Overall Loss 0.452571 Objective Loss 0.452571 LR 0.001000 Time 0.016918 -2022-12-06 10:25:16,811 - Epoch: [6][ 210/ 1200] Overall Loss 0.452983 Objective Loss 0.452983 LR 0.001000 Time 0.016747 -2022-12-06 10:25:16,945 - Epoch: [6][ 220/ 1200] Overall Loss 0.452641 Objective Loss 0.452641 LR 0.001000 Time 0.016591 -2022-12-06 10:25:17,079 - Epoch: [6][ 230/ 1200] Overall Loss 0.452364 Objective Loss 0.452364 LR 0.001000 Time 0.016449 -2022-12-06 10:25:17,212 - Epoch: [6][ 240/ 1200] Overall Loss 0.451153 Objective Loss 0.451153 LR 0.001000 Time 0.016319 -2022-12-06 10:25:17,346 - Epoch: [6][ 250/ 1200] Overall Loss 0.451507 Objective Loss 0.451507 LR 0.001000 Time 0.016200 -2022-12-06 10:25:17,480 - Epoch: [6][ 260/ 1200] Overall Loss 0.450358 Objective Loss 0.450358 LR 0.001000 Time 0.016089 -2022-12-06 10:25:17,613 - Epoch: [6][ 270/ 1200] Overall Loss 0.450414 Objective Loss 0.450414 LR 0.001000 Time 0.015985 -2022-12-06 10:25:17,748 - Epoch: [6][ 280/ 1200] Overall Loss 0.450956 Objective Loss 0.450956 LR 0.001000 Time 0.015893 -2022-12-06 10:25:17,881 - Epoch: [6][ 290/ 1200] Overall Loss 0.451237 Objective Loss 0.451237 LR 0.001000 Time 0.015804 -2022-12-06 10:25:18,015 - Epoch: [6][ 300/ 1200] Overall Loss 0.449724 Objective Loss 0.449724 LR 0.001000 Time 0.015721 -2022-12-06 10:25:18,149 - Epoch: [6][ 310/ 1200] Overall Loss 0.450114 Objective Loss 0.450114 LR 0.001000 Time 0.015645 -2022-12-06 10:25:18,282 - Epoch: [6][ 320/ 1200] Overall Loss 0.450099 Objective Loss 0.450099 LR 0.001000 Time 0.015571 -2022-12-06 10:25:18,415 - Epoch: [6][ 330/ 1200] Overall Loss 0.449049 Objective Loss 0.449049 LR 0.001000 Time 0.015501 -2022-12-06 10:25:18,550 - Epoch: [6][ 340/ 1200] Overall Loss 0.448511 Objective Loss 0.448511 LR 0.001000 Time 0.015441 -2022-12-06 10:25:18,684 - Epoch: [6][ 350/ 1200] Overall Loss 0.447725 Objective Loss 0.447725 LR 0.001000 Time 0.015380 -2022-12-06 10:25:18,818 - Epoch: [6][ 360/ 1200] Overall Loss 0.447608 Objective Loss 0.447608 LR 0.001000 Time 0.015323 -2022-12-06 10:25:18,951 - Epoch: [6][ 370/ 1200] Overall Loss 0.447286 Objective Loss 0.447286 LR 0.001000 Time 0.015269 -2022-12-06 10:25:19,085 - Epoch: [6][ 380/ 1200] Overall Loss 0.445946 Objective Loss 0.445946 LR 0.001000 Time 0.015218 -2022-12-06 10:25:19,219 - Epoch: [6][ 390/ 1200] Overall Loss 0.446607 Objective Loss 0.446607 LR 0.001000 Time 0.015169 -2022-12-06 10:25:19,352 - Epoch: [6][ 400/ 1200] Overall Loss 0.445997 Objective Loss 0.445997 LR 0.001000 Time 0.015121 -2022-12-06 10:25:19,485 - Epoch: [6][ 410/ 1200] Overall Loss 0.446248 Objective Loss 0.446248 LR 0.001000 Time 0.015076 -2022-12-06 10:25:19,618 - Epoch: [6][ 420/ 1200] Overall Loss 0.446476 Objective Loss 0.446476 LR 0.001000 Time 0.015033 -2022-12-06 10:25:19,752 - Epoch: [6][ 430/ 1200] Overall Loss 0.446380 Objective Loss 0.446380 LR 0.001000 Time 0.014993 -2022-12-06 10:25:19,886 - Epoch: [6][ 440/ 1200] Overall Loss 0.445370 Objective Loss 0.445370 LR 0.001000 Time 0.014956 -2022-12-06 10:25:20,019 - Epoch: [6][ 450/ 1200] Overall Loss 0.445397 Objective Loss 0.445397 LR 0.001000 Time 0.014919 -2022-12-06 10:25:20,153 - Epoch: [6][ 460/ 1200] Overall Loss 0.445437 Objective Loss 0.445437 LR 0.001000 Time 0.014885 -2022-12-06 10:25:20,286 - Epoch: [6][ 470/ 1200] Overall Loss 0.445184 Objective Loss 0.445184 LR 0.001000 Time 0.014850 -2022-12-06 10:25:20,420 - Epoch: [6][ 480/ 1200] Overall Loss 0.445205 Objective Loss 0.445205 LR 0.001000 Time 0.014818 -2022-12-06 10:25:20,553 - Epoch: [6][ 490/ 1200] Overall Loss 0.445199 Objective Loss 0.445199 LR 0.001000 Time 0.014786 -2022-12-06 10:25:20,687 - Epoch: [6][ 500/ 1200] Overall Loss 0.444595 Objective Loss 0.444595 LR 0.001000 Time 0.014757 -2022-12-06 10:25:20,821 - Epoch: [6][ 510/ 1200] Overall Loss 0.445239 Objective Loss 0.445239 LR 0.001000 Time 0.014730 -2022-12-06 10:25:20,955 - Epoch: [6][ 520/ 1200] Overall Loss 0.444560 Objective Loss 0.444560 LR 0.001000 Time 0.014703 -2022-12-06 10:25:21,088 - Epoch: [6][ 530/ 1200] Overall Loss 0.445164 Objective Loss 0.445164 LR 0.001000 Time 0.014676 -2022-12-06 10:25:21,221 - Epoch: [6][ 540/ 1200] Overall Loss 0.444617 Objective Loss 0.444617 LR 0.001000 Time 0.014650 -2022-12-06 10:25:21,354 - Epoch: [6][ 550/ 1200] Overall Loss 0.444782 Objective Loss 0.444782 LR 0.001000 Time 0.014625 -2022-12-06 10:25:21,488 - Epoch: [6][ 560/ 1200] Overall Loss 0.445351 Objective Loss 0.445351 LR 0.001000 Time 0.014601 -2022-12-06 10:25:21,621 - Epoch: [6][ 570/ 1200] Overall Loss 0.445357 Objective Loss 0.445357 LR 0.001000 Time 0.014578 -2022-12-06 10:25:21,754 - Epoch: [6][ 580/ 1200] Overall Loss 0.445807 Objective Loss 0.445807 LR 0.001000 Time 0.014556 -2022-12-06 10:25:21,888 - Epoch: [6][ 590/ 1200] Overall Loss 0.445687 Objective Loss 0.445687 LR 0.001000 Time 0.014535 -2022-12-06 10:25:22,022 - Epoch: [6][ 600/ 1200] Overall Loss 0.445931 Objective Loss 0.445931 LR 0.001000 Time 0.014516 -2022-12-06 10:25:22,156 - Epoch: [6][ 610/ 1200] Overall Loss 0.446029 Objective Loss 0.446029 LR 0.001000 Time 0.014496 -2022-12-06 10:25:22,289 - Epoch: [6][ 620/ 1200] Overall Loss 0.446038 Objective Loss 0.446038 LR 0.001000 Time 0.014476 -2022-12-06 10:25:22,422 - Epoch: [6][ 630/ 1200] Overall Loss 0.445496 Objective Loss 0.445496 LR 0.001000 Time 0.014457 -2022-12-06 10:25:22,556 - Epoch: [6][ 640/ 1200] Overall Loss 0.445203 Objective Loss 0.445203 LR 0.001000 Time 0.014439 -2022-12-06 10:25:22,689 - Epoch: [6][ 650/ 1200] Overall Loss 0.445221 Objective Loss 0.445221 LR 0.001000 Time 0.014422 -2022-12-06 10:25:22,822 - Epoch: [6][ 660/ 1200] Overall Loss 0.445472 Objective Loss 0.445472 LR 0.001000 Time 0.014404 -2022-12-06 10:25:22,955 - Epoch: [6][ 670/ 1200] Overall Loss 0.445556 Objective Loss 0.445556 LR 0.001000 Time 0.014387 -2022-12-06 10:25:23,089 - Epoch: [6][ 680/ 1200] Overall Loss 0.445455 Objective Loss 0.445455 LR 0.001000 Time 0.014372 -2022-12-06 10:25:23,222 - Epoch: [6][ 690/ 1200] Overall Loss 0.444980 Objective Loss 0.444980 LR 0.001000 Time 0.014356 -2022-12-06 10:25:23,356 - Epoch: [6][ 700/ 1200] Overall Loss 0.444327 Objective Loss 0.444327 LR 0.001000 Time 0.014340 -2022-12-06 10:25:23,489 - Epoch: [6][ 710/ 1200] Overall Loss 0.443759 Objective Loss 0.443759 LR 0.001000 Time 0.014325 -2022-12-06 10:25:23,622 - Epoch: [6][ 720/ 1200] Overall Loss 0.443979 Objective Loss 0.443979 LR 0.001000 Time 0.014311 -2022-12-06 10:25:23,756 - Epoch: [6][ 730/ 1200] Overall Loss 0.444597 Objective Loss 0.444597 LR 0.001000 Time 0.014298 -2022-12-06 10:25:23,890 - Epoch: [6][ 740/ 1200] Overall Loss 0.444456 Objective Loss 0.444456 LR 0.001000 Time 0.014285 -2022-12-06 10:25:24,024 - Epoch: [6][ 750/ 1200] Overall Loss 0.444564 Objective Loss 0.444564 LR 0.001000 Time 0.014272 -2022-12-06 10:25:24,158 - Epoch: [6][ 760/ 1200] Overall Loss 0.444919 Objective Loss 0.444919 LR 0.001000 Time 0.014260 -2022-12-06 10:25:24,296 - Epoch: [6][ 770/ 1200] Overall Loss 0.444386 Objective Loss 0.444386 LR 0.001000 Time 0.014253 -2022-12-06 10:25:24,435 - Epoch: [6][ 780/ 1200] Overall Loss 0.444400 Objective Loss 0.444400 LR 0.001000 Time 0.014247 -2022-12-06 10:25:24,576 - Epoch: [6][ 790/ 1200] Overall Loss 0.444226 Objective Loss 0.444226 LR 0.001000 Time 0.014244 -2022-12-06 10:25:24,714 - Epoch: [6][ 800/ 1200] Overall Loss 0.444271 Objective Loss 0.444271 LR 0.001000 Time 0.014238 -2022-12-06 10:25:24,854 - Epoch: [6][ 810/ 1200] Overall Loss 0.444892 Objective Loss 0.444892 LR 0.001000 Time 0.014234 -2022-12-06 10:25:24,993 - Epoch: [6][ 820/ 1200] Overall Loss 0.444848 Objective Loss 0.444848 LR 0.001000 Time 0.014229 -2022-12-06 10:25:25,133 - Epoch: [6][ 830/ 1200] Overall Loss 0.444984 Objective Loss 0.444984 LR 0.001000 Time 0.014225 -2022-12-06 10:25:25,272 - Epoch: [6][ 840/ 1200] Overall Loss 0.444784 Objective Loss 0.444784 LR 0.001000 Time 0.014219 -2022-12-06 10:25:25,411 - Epoch: [6][ 850/ 1200] Overall Loss 0.444725 Objective Loss 0.444725 LR 0.001000 Time 0.014216 -2022-12-06 10:25:25,551 - Epoch: [6][ 860/ 1200] Overall Loss 0.444992 Objective Loss 0.444992 LR 0.001000 Time 0.014210 -2022-12-06 10:25:25,692 - Epoch: [6][ 870/ 1200] Overall Loss 0.445176 Objective Loss 0.445176 LR 0.001000 Time 0.014208 -2022-12-06 10:25:25,830 - Epoch: [6][ 880/ 1200] Overall Loss 0.445060 Objective Loss 0.445060 LR 0.001000 Time 0.014203 -2022-12-06 10:25:25,969 - Epoch: [6][ 890/ 1200] Overall Loss 0.445129 Objective Loss 0.445129 LR 0.001000 Time 0.014200 -2022-12-06 10:25:26,109 - Epoch: [6][ 900/ 1200] Overall Loss 0.445090 Objective Loss 0.445090 LR 0.001000 Time 0.014195 -2022-12-06 10:25:26,248 - Epoch: [6][ 910/ 1200] Overall Loss 0.445537 Objective Loss 0.445537 LR 0.001000 Time 0.014192 -2022-12-06 10:25:26,387 - Epoch: [6][ 920/ 1200] Overall Loss 0.445256 Objective Loss 0.445256 LR 0.001000 Time 0.014187 -2022-12-06 10:25:26,528 - Epoch: [6][ 930/ 1200] Overall Loss 0.445536 Objective Loss 0.445536 LR 0.001000 Time 0.014186 -2022-12-06 10:25:26,667 - Epoch: [6][ 940/ 1200] Overall Loss 0.445616 Objective Loss 0.445616 LR 0.001000 Time 0.014181 -2022-12-06 10:25:26,807 - Epoch: [6][ 950/ 1200] Overall Loss 0.445590 Objective Loss 0.445590 LR 0.001000 Time 0.014180 -2022-12-06 10:25:26,946 - Epoch: [6][ 960/ 1200] Overall Loss 0.445507 Objective Loss 0.445507 LR 0.001000 Time 0.014176 -2022-12-06 10:25:27,085 - Epoch: [6][ 970/ 1200] Overall Loss 0.445322 Objective Loss 0.445322 LR 0.001000 Time 0.014173 -2022-12-06 10:25:27,225 - Epoch: [6][ 980/ 1200] Overall Loss 0.445176 Objective Loss 0.445176 LR 0.001000 Time 0.014169 -2022-12-06 10:25:27,364 - Epoch: [6][ 990/ 1200] Overall Loss 0.445584 Objective Loss 0.445584 LR 0.001000 Time 0.014166 -2022-12-06 10:25:27,503 - Epoch: [6][ 1000/ 1200] Overall Loss 0.445943 Objective Loss 0.445943 LR 0.001000 Time 0.014162 -2022-12-06 10:25:27,644 - Epoch: [6][ 1010/ 1200] Overall Loss 0.445951 Objective Loss 0.445951 LR 0.001000 Time 0.014160 -2022-12-06 10:25:27,782 - Epoch: [6][ 1020/ 1200] Overall Loss 0.446126 Objective Loss 0.446126 LR 0.001000 Time 0.014156 -2022-12-06 10:25:27,922 - Epoch: [6][ 1030/ 1200] Overall Loss 0.446098 Objective Loss 0.446098 LR 0.001000 Time 0.014154 -2022-12-06 10:25:28,061 - Epoch: [6][ 1040/ 1200] Overall Loss 0.445959 Objective Loss 0.445959 LR 0.001000 Time 0.014150 -2022-12-06 10:25:28,201 - Epoch: [6][ 1050/ 1200] Overall Loss 0.446112 Objective Loss 0.446112 LR 0.001000 Time 0.014148 -2022-12-06 10:25:28,340 - Epoch: [6][ 1060/ 1200] Overall Loss 0.445946 Objective Loss 0.445946 LR 0.001000 Time 0.014144 -2022-12-06 10:25:28,479 - Epoch: [6][ 1070/ 1200] Overall Loss 0.446010 Objective Loss 0.446010 LR 0.001000 Time 0.014142 -2022-12-06 10:25:28,619 - Epoch: [6][ 1080/ 1200] Overall Loss 0.446320 Objective Loss 0.446320 LR 0.001000 Time 0.014139 -2022-12-06 10:25:28,759 - Epoch: [6][ 1090/ 1200] Overall Loss 0.445812 Objective Loss 0.445812 LR 0.001000 Time 0.014137 -2022-12-06 10:25:28,898 - Epoch: [6][ 1100/ 1200] Overall Loss 0.445811 Objective Loss 0.445811 LR 0.001000 Time 0.014134 -2022-12-06 10:25:29,037 - Epoch: [6][ 1110/ 1200] Overall Loss 0.445834 Objective Loss 0.445834 LR 0.001000 Time 0.014132 -2022-12-06 10:25:29,176 - Epoch: [6][ 1120/ 1200] Overall Loss 0.445764 Objective Loss 0.445764 LR 0.001000 Time 0.014129 -2022-12-06 10:25:29,316 - Epoch: [6][ 1130/ 1200] Overall Loss 0.445848 Objective Loss 0.445848 LR 0.001000 Time 0.014127 -2022-12-06 10:25:29,455 - Epoch: [6][ 1140/ 1200] Overall Loss 0.445737 Objective Loss 0.445737 LR 0.001000 Time 0.014123 -2022-12-06 10:25:29,596 - Epoch: [6][ 1150/ 1200] Overall Loss 0.445663 Objective Loss 0.445663 LR 0.001000 Time 0.014123 -2022-12-06 10:25:29,734 - Epoch: [6][ 1160/ 1200] Overall Loss 0.445999 Objective Loss 0.445999 LR 0.001000 Time 0.014120 -2022-12-06 10:25:29,873 - Epoch: [6][ 1170/ 1200] Overall Loss 0.446189 Objective Loss 0.446189 LR 0.001000 Time 0.014117 -2022-12-06 10:25:30,013 - Epoch: [6][ 1180/ 1200] Overall Loss 0.445752 Objective Loss 0.445752 LR 0.001000 Time 0.014115 -2022-12-06 10:25:30,152 - Epoch: [6][ 1190/ 1200] Overall Loss 0.445320 Objective Loss 0.445320 LR 0.001000 Time 0.014113 -2022-12-06 10:25:30,333 - Epoch: [6][ 1200/ 1200] Overall Loss 0.445241 Objective Loss 0.445241 Top1 80.753138 Top5 98.326360 LR 0.001000 Time 0.014145 -2022-12-06 10:25:30,430 - --- validate (epoch=6)----------- -2022-12-06 10:25:30,430 - 34129 samples (256 per mini-batch) -2022-12-06 10:25:30,834 - Epoch: [6][ 10/ 134] Loss 0.425556 Top1 79.414062 Top5 96.679688 -2022-12-06 10:25:30,933 - Epoch: [6][ 20/ 134] Loss 0.424086 Top1 78.750000 Top5 96.816406 -2022-12-06 10:25:31,031 - Epoch: [6][ 30/ 134] Loss 0.422523 Top1 78.958333 Top5 96.953125 -2022-12-06 10:25:31,123 - Epoch: [6][ 40/ 134] Loss 0.418009 Top1 79.199219 Top5 96.992188 -2022-12-06 10:25:31,221 - Epoch: [6][ 50/ 134] Loss 0.417443 Top1 79.179688 Top5 97.015625 -2022-12-06 10:25:31,311 - Epoch: [6][ 60/ 134] Loss 0.410444 Top1 79.563802 Top5 97.174479 -2022-12-06 10:25:31,408 - Epoch: [6][ 70/ 134] Loss 0.408637 Top1 79.637277 Top5 97.142857 -2022-12-06 10:25:31,499 - Epoch: [6][ 80/ 134] Loss 0.406926 Top1 79.760742 Top5 97.172852 -2022-12-06 10:25:31,596 - Epoch: [6][ 90/ 134] Loss 0.409977 Top1 79.691840 Top5 97.196181 -2022-12-06 10:25:31,686 - Epoch: [6][ 100/ 134] Loss 0.413227 Top1 79.464844 Top5 97.214844 -2022-12-06 10:25:31,783 - Epoch: [6][ 110/ 134] Loss 0.412591 Top1 79.534801 Top5 97.237216 -2022-12-06 10:25:31,868 - Epoch: [6][ 120/ 134] Loss 0.413428 Top1 79.534505 Top5 97.255859 -2022-12-06 10:25:31,963 - Epoch: [6][ 130/ 134] Loss 0.414392 Top1 79.537260 Top5 97.292668 -2022-12-06 10:25:31,985 - Epoch: [6][ 134/ 134] Loss 0.411725 Top1 79.583346 Top5 97.310205 -2022-12-06 10:25:32,080 - ==> Top1: 79.583 Top5: 97.310 Loss: 0.412 - -2022-12-06 10:25:32,081 - ==> Confusion: -[[ 890 0 10 1 8 8 0 0 4 43 0 7 4 3 3 2 1 2 1 0 9] - [ 0 860 5 2 10 52 2 26 2 1 3 9 6 1 3 1 9 1 14 6 14] - [ 6 1 1006 8 3 8 15 9 0 0 5 11 2 3 1 5 0 2 2 4 12] - [ 1 4 42 881 0 7 0 2 3 0 15 6 5 3 17 0 3 3 17 1 10] - [ 13 6 5 0 926 9 0 4 1 5 1 5 1 2 9 7 13 2 1 2 8] - [ 2 36 4 3 6 903 3 29 3 3 2 26 7 12 1 1 3 0 4 14 7] - [ 0 7 48 1 1 3 1001 8 0 0 4 8 0 0 0 9 0 2 1 24 1] - [ 2 8 29 0 1 41 2 887 0 2 2 12 0 0 0 2 1 1 38 15 11] - [ 5 3 3 2 1 7 0 1 907 59 11 6 6 27 11 1 3 0 6 1 4] - [ 110 1 4 0 13 10 0 3 26 784 0 1 1 29 5 0 3 1 1 0 9] - [ 0 4 14 13 0 0 5 6 6 2 912 8 1 14 3 0 4 0 17 3 7] - [ 3 3 3 1 0 11 3 2 1 0 0 964 22 4 1 7 5 4 3 11 3] - [ 1 3 3 5 2 1 0 2 0 0 1 87 821 1 1 9 1 17 3 3 8] - [ 1 4 0 1 4 24 1 2 3 11 7 17 6 912 1 2 5 0 0 7 15] - [ 12 4 8 25 9 3 0 1 15 2 3 5 6 5 1003 0 2 2 8 1 16] - [ 3 0 12 2 2 1 9 0 1 0 0 18 7 2 0 956 9 7 3 6 5] - [ 4 4 2 2 6 2 0 0 1 0 1 8 2 0 2 10 1014 0 2 6 6] - [ 2 2 1 7 0 3 2 0 0 1 1 36 46 3 2 19 5 898 1 1 6] - [ 1 3 14 11 0 3 1 35 4 0 8 7 5 0 9 0 0 0 899 2 6] - [ 0 3 5 0 0 3 5 10 0 0 0 32 6 2 0 5 5 7 0 985 12] - [ 200 244 412 99 177 310 41 161 76 78 153 322 433 379 167 165 342 48 241 426 8752]] - -2022-12-06 10:25:32,745 - ==> Best [Top1: 79.583 Top5: 97.310 Sparsity:0.00 Params: 5376 on epoch: 6] -2022-12-06 10:25:32,745 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:25:32,752 - - -2022-12-06 10:25:32,752 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:25:33,615 - Epoch: [7][ 10/ 1200] Overall Loss 0.415669 Objective Loss 0.415669 LR 0.001000 Time 0.086267 -2022-12-06 10:25:33,754 - Epoch: [7][ 20/ 1200] Overall Loss 0.391668 Objective Loss 0.391668 LR 0.001000 Time 0.050050 -2022-12-06 10:25:33,892 - Epoch: [7][ 30/ 1200] Overall Loss 0.387103 Objective Loss 0.387103 LR 0.001000 Time 0.037938 -2022-12-06 10:25:34,028 - Epoch: [7][ 40/ 1200] Overall Loss 0.392236 Objective Loss 0.392236 LR 0.001000 Time 0.031837 -2022-12-06 10:25:34,164 - Epoch: [7][ 50/ 1200] Overall Loss 0.400623 Objective Loss 0.400623 LR 0.001000 Time 0.028176 -2022-12-06 10:25:34,300 - Epoch: [7][ 60/ 1200] Overall Loss 0.400609 Objective Loss 0.400609 LR 0.001000 Time 0.025737 -2022-12-06 10:25:34,435 - Epoch: [7][ 70/ 1200] Overall Loss 0.404330 Objective Loss 0.404330 LR 0.001000 Time 0.023992 -2022-12-06 10:25:34,571 - Epoch: [7][ 80/ 1200] Overall Loss 0.405774 Objective Loss 0.405774 LR 0.001000 Time 0.022682 -2022-12-06 10:25:34,706 - Epoch: [7][ 90/ 1200] Overall Loss 0.406900 Objective Loss 0.406900 LR 0.001000 Time 0.021661 -2022-12-06 10:25:34,842 - Epoch: [7][ 100/ 1200] Overall Loss 0.407368 Objective Loss 0.407368 LR 0.001000 Time 0.020847 -2022-12-06 10:25:34,978 - Epoch: [7][ 110/ 1200] Overall Loss 0.407452 Objective Loss 0.407452 LR 0.001000 Time 0.020182 -2022-12-06 10:25:35,114 - Epoch: [7][ 120/ 1200] Overall Loss 0.413847 Objective Loss 0.413847 LR 0.001000 Time 0.019632 -2022-12-06 10:25:35,250 - Epoch: [7][ 130/ 1200] Overall Loss 0.411698 Objective Loss 0.411698 LR 0.001000 Time 0.019165 -2022-12-06 10:25:35,387 - Epoch: [7][ 140/ 1200] Overall Loss 0.412570 Objective Loss 0.412570 LR 0.001000 Time 0.018765 -2022-12-06 10:25:35,523 - Epoch: [7][ 150/ 1200] Overall Loss 0.414594 Objective Loss 0.414594 LR 0.001000 Time 0.018419 -2022-12-06 10:25:35,659 - Epoch: [7][ 160/ 1200] Overall Loss 0.413628 Objective Loss 0.413628 LR 0.001000 Time 0.018115 -2022-12-06 10:25:35,795 - Epoch: [7][ 170/ 1200] Overall Loss 0.414281 Objective Loss 0.414281 LR 0.001000 Time 0.017848 -2022-12-06 10:25:35,931 - Epoch: [7][ 180/ 1200] Overall Loss 0.415555 Objective Loss 0.415555 LR 0.001000 Time 0.017610 -2022-12-06 10:25:36,068 - Epoch: [7][ 190/ 1200] Overall Loss 0.416070 Objective Loss 0.416070 LR 0.001000 Time 0.017397 -2022-12-06 10:25:36,204 - Epoch: [7][ 200/ 1200] Overall Loss 0.417564 Objective Loss 0.417564 LR 0.001000 Time 0.017205 -2022-12-06 10:25:36,339 - Epoch: [7][ 210/ 1200] Overall Loss 0.417606 Objective Loss 0.417606 LR 0.001000 Time 0.017030 -2022-12-06 10:25:36,475 - Epoch: [7][ 220/ 1200] Overall Loss 0.416889 Objective Loss 0.416889 LR 0.001000 Time 0.016871 -2022-12-06 10:25:36,611 - Epoch: [7][ 230/ 1200] Overall Loss 0.417846 Objective Loss 0.417846 LR 0.001000 Time 0.016727 -2022-12-06 10:25:36,747 - Epoch: [7][ 240/ 1200] Overall Loss 0.416819 Objective Loss 0.416819 LR 0.001000 Time 0.016593 -2022-12-06 10:25:36,883 - Epoch: [7][ 250/ 1200] Overall Loss 0.416304 Objective Loss 0.416304 LR 0.001000 Time 0.016471 -2022-12-06 10:25:37,019 - Epoch: [7][ 260/ 1200] Overall Loss 0.416198 Objective Loss 0.416198 LR 0.001000 Time 0.016359 -2022-12-06 10:25:37,155 - Epoch: [7][ 270/ 1200] Overall Loss 0.417765 Objective Loss 0.417765 LR 0.001000 Time 0.016255 -2022-12-06 10:25:37,291 - Epoch: [7][ 280/ 1200] Overall Loss 0.417417 Objective Loss 0.417417 LR 0.001000 Time 0.016158 -2022-12-06 10:25:37,427 - Epoch: [7][ 290/ 1200] Overall Loss 0.417327 Objective Loss 0.417327 LR 0.001000 Time 0.016068 -2022-12-06 10:25:37,563 - Epoch: [7][ 300/ 1200] Overall Loss 0.418076 Objective Loss 0.418076 LR 0.001000 Time 0.015984 -2022-12-06 10:25:37,699 - Epoch: [7][ 310/ 1200] Overall Loss 0.418651 Objective Loss 0.418651 LR 0.001000 Time 0.015905 -2022-12-06 10:25:37,835 - Epoch: [7][ 320/ 1200] Overall Loss 0.418218 Objective Loss 0.418218 LR 0.001000 Time 0.015831 -2022-12-06 10:25:37,971 - Epoch: [7][ 330/ 1200] Overall Loss 0.418693 Objective Loss 0.418693 LR 0.001000 Time 0.015762 -2022-12-06 10:25:38,107 - Epoch: [7][ 340/ 1200] Overall Loss 0.418258 Objective Loss 0.418258 LR 0.001000 Time 0.015699 -2022-12-06 10:25:38,242 - Epoch: [7][ 350/ 1200] Overall Loss 0.418579 Objective Loss 0.418579 LR 0.001000 Time 0.015633 -2022-12-06 10:25:38,380 - Epoch: [7][ 360/ 1200] Overall Loss 0.418396 Objective Loss 0.418396 LR 0.001000 Time 0.015576 -2022-12-06 10:25:38,515 - Epoch: [7][ 370/ 1200] Overall Loss 0.418397 Objective Loss 0.418397 LR 0.001000 Time 0.015518 -2022-12-06 10:25:38,653 - Epoch: [7][ 380/ 1200] Overall Loss 0.417837 Objective Loss 0.417837 LR 0.001000 Time 0.015466 -2022-12-06 10:25:38,787 - Epoch: [7][ 390/ 1200] Overall Loss 0.417241 Objective Loss 0.417241 LR 0.001000 Time 0.015413 -2022-12-06 10:25:38,924 - Epoch: [7][ 400/ 1200] Overall Loss 0.416803 Objective Loss 0.416803 LR 0.001000 Time 0.015365 -2022-12-06 10:25:39,059 - Epoch: [7][ 410/ 1200] Overall Loss 0.416211 Objective Loss 0.416211 LR 0.001000 Time 0.015318 -2022-12-06 10:25:39,197 - Epoch: [7][ 420/ 1200] Overall Loss 0.415589 Objective Loss 0.415589 LR 0.001000 Time 0.015276 -2022-12-06 10:25:39,332 - Epoch: [7][ 430/ 1200] Overall Loss 0.415672 Objective Loss 0.415672 LR 0.001000 Time 0.015233 -2022-12-06 10:25:39,469 - Epoch: [7][ 440/ 1200] Overall Loss 0.415103 Objective Loss 0.415103 LR 0.001000 Time 0.015194 -2022-12-06 10:25:39,606 - Epoch: [7][ 450/ 1200] Overall Loss 0.414911 Objective Loss 0.414911 LR 0.001000 Time 0.015159 -2022-12-06 10:25:39,740 - Epoch: [7][ 460/ 1200] Overall Loss 0.414791 Objective Loss 0.414791 LR 0.001000 Time 0.015120 -2022-12-06 10:25:39,883 - Epoch: [7][ 470/ 1200] Overall Loss 0.415173 Objective Loss 0.415173 LR 0.001000 Time 0.015099 -2022-12-06 10:25:40,025 - Epoch: [7][ 480/ 1200] Overall Loss 0.414758 Objective Loss 0.414758 LR 0.001000 Time 0.015075 -2022-12-06 10:25:40,166 - Epoch: [7][ 490/ 1200] Overall Loss 0.414922 Objective Loss 0.414922 LR 0.001000 Time 0.015055 -2022-12-06 10:25:40,307 - Epoch: [7][ 500/ 1200] Overall Loss 0.415681 Objective Loss 0.415681 LR 0.001000 Time 0.015032 -2022-12-06 10:25:40,448 - Epoch: [7][ 510/ 1200] Overall Loss 0.415564 Objective Loss 0.415564 LR 0.001000 Time 0.015014 -2022-12-06 10:25:40,589 - Epoch: [7][ 520/ 1200] Overall Loss 0.415728 Objective Loss 0.415728 LR 0.001000 Time 0.014993 -2022-12-06 10:25:40,731 - Epoch: [7][ 530/ 1200] Overall Loss 0.415874 Objective Loss 0.415874 LR 0.001000 Time 0.014976 -2022-12-06 10:25:40,872 - Epoch: [7][ 540/ 1200] Overall Loss 0.416210 Objective Loss 0.416210 LR 0.001000 Time 0.014957 -2022-12-06 10:25:41,014 - Epoch: [7][ 550/ 1200] Overall Loss 0.416251 Objective Loss 0.416251 LR 0.001000 Time 0.014942 -2022-12-06 10:25:41,155 - Epoch: [7][ 560/ 1200] Overall Loss 0.417134 Objective Loss 0.417134 LR 0.001000 Time 0.014924 -2022-12-06 10:25:41,296 - Epoch: [7][ 570/ 1200] Overall Loss 0.417208 Objective Loss 0.417208 LR 0.001000 Time 0.014909 -2022-12-06 10:25:41,437 - Epoch: [7][ 580/ 1200] Overall Loss 0.417207 Objective Loss 0.417207 LR 0.001000 Time 0.014893 -2022-12-06 10:25:41,579 - Epoch: [7][ 590/ 1200] Overall Loss 0.417138 Objective Loss 0.417138 LR 0.001000 Time 0.014879 -2022-12-06 10:25:41,720 - Epoch: [7][ 600/ 1200] Overall Loss 0.417813 Objective Loss 0.417813 LR 0.001000 Time 0.014863 -2022-12-06 10:25:41,862 - Epoch: [7][ 610/ 1200] Overall Loss 0.418187 Objective Loss 0.418187 LR 0.001000 Time 0.014851 -2022-12-06 10:25:42,003 - Epoch: [7][ 620/ 1200] Overall Loss 0.418084 Objective Loss 0.418084 LR 0.001000 Time 0.014836 -2022-12-06 10:25:42,145 - Epoch: [7][ 630/ 1200] Overall Loss 0.418454 Objective Loss 0.418454 LR 0.001000 Time 0.014825 -2022-12-06 10:25:42,286 - Epoch: [7][ 640/ 1200] Overall Loss 0.418629 Objective Loss 0.418629 LR 0.001000 Time 0.014812 -2022-12-06 10:25:42,428 - Epoch: [7][ 650/ 1200] Overall Loss 0.418208 Objective Loss 0.418208 LR 0.001000 Time 0.014801 -2022-12-06 10:25:42,569 - Epoch: [7][ 660/ 1200] Overall Loss 0.418308 Objective Loss 0.418308 LR 0.001000 Time 0.014788 -2022-12-06 10:25:42,711 - Epoch: [7][ 670/ 1200] Overall Loss 0.418136 Objective Loss 0.418136 LR 0.001000 Time 0.014778 -2022-12-06 10:25:42,852 - Epoch: [7][ 680/ 1200] Overall Loss 0.418009 Objective Loss 0.418009 LR 0.001000 Time 0.014766 -2022-12-06 10:25:42,994 - Epoch: [7][ 690/ 1200] Overall Loss 0.417915 Objective Loss 0.417915 LR 0.001000 Time 0.014757 -2022-12-06 10:25:43,135 - Epoch: [7][ 700/ 1200] Overall Loss 0.418015 Objective Loss 0.418015 LR 0.001000 Time 0.014746 -2022-12-06 10:25:43,277 - Epoch: [7][ 710/ 1200] Overall Loss 0.418060 Objective Loss 0.418060 LR 0.001000 Time 0.014737 -2022-12-06 10:25:43,419 - Epoch: [7][ 720/ 1200] Overall Loss 0.417698 Objective Loss 0.417698 LR 0.001000 Time 0.014726 -2022-12-06 10:25:43,560 - Epoch: [7][ 730/ 1200] Overall Loss 0.418024 Objective Loss 0.418024 LR 0.001000 Time 0.014718 -2022-12-06 10:25:43,702 - Epoch: [7][ 740/ 1200] Overall Loss 0.418319 Objective Loss 0.418319 LR 0.001000 Time 0.014708 -2022-12-06 10:25:43,844 - Epoch: [7][ 750/ 1200] Overall Loss 0.418223 Objective Loss 0.418223 LR 0.001000 Time 0.014701 -2022-12-06 10:25:43,986 - Epoch: [7][ 760/ 1200] Overall Loss 0.417740 Objective Loss 0.417740 LR 0.001000 Time 0.014692 -2022-12-06 10:25:44,128 - Epoch: [7][ 770/ 1200] Overall Loss 0.417951 Objective Loss 0.417951 LR 0.001000 Time 0.014684 -2022-12-06 10:25:44,269 - Epoch: [7][ 780/ 1200] Overall Loss 0.418464 Objective Loss 0.418464 LR 0.001000 Time 0.014675 -2022-12-06 10:25:44,411 - Epoch: [7][ 790/ 1200] Overall Loss 0.418199 Objective Loss 0.418199 LR 0.001000 Time 0.014668 -2022-12-06 10:25:44,552 - Epoch: [7][ 800/ 1200] Overall Loss 0.418107 Objective Loss 0.418107 LR 0.001000 Time 0.014659 -2022-12-06 10:25:44,694 - Epoch: [7][ 810/ 1200] Overall Loss 0.418517 Objective Loss 0.418517 LR 0.001000 Time 0.014653 -2022-12-06 10:25:44,835 - Epoch: [7][ 820/ 1200] Overall Loss 0.418404 Objective Loss 0.418404 LR 0.001000 Time 0.014645 -2022-12-06 10:25:44,977 - Epoch: [7][ 830/ 1200] Overall Loss 0.419103 Objective Loss 0.419103 LR 0.001000 Time 0.014638 -2022-12-06 10:25:45,119 - Epoch: [7][ 840/ 1200] Overall Loss 0.418972 Objective Loss 0.418972 LR 0.001000 Time 0.014631 -2022-12-06 10:25:45,261 - Epoch: [7][ 850/ 1200] Overall Loss 0.419284 Objective Loss 0.419284 LR 0.001000 Time 0.014625 -2022-12-06 10:25:45,402 - Epoch: [7][ 860/ 1200] Overall Loss 0.419144 Objective Loss 0.419144 LR 0.001000 Time 0.014617 -2022-12-06 10:25:45,544 - Epoch: [7][ 870/ 1200] Overall Loss 0.419131 Objective Loss 0.419131 LR 0.001000 Time 0.014611 -2022-12-06 10:25:45,685 - Epoch: [7][ 880/ 1200] Overall Loss 0.419074 Objective Loss 0.419074 LR 0.001000 Time 0.014604 -2022-12-06 10:25:45,827 - Epoch: [7][ 890/ 1200] Overall Loss 0.419006 Objective Loss 0.419006 LR 0.001000 Time 0.014599 -2022-12-06 10:25:45,968 - Epoch: [7][ 900/ 1200] Overall Loss 0.418963 Objective Loss 0.418963 LR 0.001000 Time 0.014592 -2022-12-06 10:25:46,110 - Epoch: [7][ 910/ 1200] Overall Loss 0.419083 Objective Loss 0.419083 LR 0.001000 Time 0.014586 -2022-12-06 10:25:46,251 - Epoch: [7][ 920/ 1200] Overall Loss 0.419240 Objective Loss 0.419240 LR 0.001000 Time 0.014580 -2022-12-06 10:25:46,393 - Epoch: [7][ 930/ 1200] Overall Loss 0.419468 Objective Loss 0.419468 LR 0.001000 Time 0.014575 -2022-12-06 10:25:46,535 - Epoch: [7][ 940/ 1200] Overall Loss 0.419577 Objective Loss 0.419577 LR 0.001000 Time 0.014568 -2022-12-06 10:25:46,676 - Epoch: [7][ 950/ 1200] Overall Loss 0.419327 Objective Loss 0.419327 LR 0.001000 Time 0.014564 -2022-12-06 10:25:46,818 - Epoch: [7][ 960/ 1200] Overall Loss 0.419499 Objective Loss 0.419499 LR 0.001000 Time 0.014557 -2022-12-06 10:25:46,959 - Epoch: [7][ 970/ 1200] Overall Loss 0.419659 Objective Loss 0.419659 LR 0.001000 Time 0.014553 -2022-12-06 10:25:47,101 - Epoch: [7][ 980/ 1200] Overall Loss 0.419811 Objective Loss 0.419811 LR 0.001000 Time 0.014547 -2022-12-06 10:25:47,243 - Epoch: [7][ 990/ 1200] Overall Loss 0.419538 Objective Loss 0.419538 LR 0.001000 Time 0.014543 -2022-12-06 10:25:47,385 - Epoch: [7][ 1000/ 1200] Overall Loss 0.419315 Objective Loss 0.419315 LR 0.001000 Time 0.014537 -2022-12-06 10:25:47,526 - Epoch: [7][ 1010/ 1200] Overall Loss 0.419408 Objective Loss 0.419408 LR 0.001000 Time 0.014533 -2022-12-06 10:25:47,668 - Epoch: [7][ 1020/ 1200] Overall Loss 0.419323 Objective Loss 0.419323 LR 0.001000 Time 0.014528 -2022-12-06 10:25:47,812 - Epoch: [7][ 1030/ 1200] Overall Loss 0.419164 Objective Loss 0.419164 LR 0.001000 Time 0.014526 -2022-12-06 10:25:47,952 - Epoch: [7][ 1040/ 1200] Overall Loss 0.419001 Objective Loss 0.419001 LR 0.001000 Time 0.014521 -2022-12-06 10:25:48,096 - Epoch: [7][ 1050/ 1200] Overall Loss 0.419164 Objective Loss 0.419164 LR 0.001000 Time 0.014519 -2022-12-06 10:25:48,236 - Epoch: [7][ 1060/ 1200] Overall Loss 0.419286 Objective Loss 0.419286 LR 0.001000 Time 0.014514 -2022-12-06 10:25:48,380 - Epoch: [7][ 1070/ 1200] Overall Loss 0.419168 Objective Loss 0.419168 LR 0.001000 Time 0.014512 -2022-12-06 10:25:48,520 - Epoch: [7][ 1080/ 1200] Overall Loss 0.419356 Objective Loss 0.419356 LR 0.001000 Time 0.014507 -2022-12-06 10:25:48,664 - Epoch: [7][ 1090/ 1200] Overall Loss 0.418926 Objective Loss 0.418926 LR 0.001000 Time 0.014505 -2022-12-06 10:25:48,805 - Epoch: [7][ 1100/ 1200] Overall Loss 0.418651 Objective Loss 0.418651 LR 0.001000 Time 0.014501 -2022-12-06 10:25:48,949 - Epoch: [7][ 1110/ 1200] Overall Loss 0.418753 Objective Loss 0.418753 LR 0.001000 Time 0.014500 -2022-12-06 10:25:49,090 - Epoch: [7][ 1120/ 1200] Overall Loss 0.418391 Objective Loss 0.418391 LR 0.001000 Time 0.014496 -2022-12-06 10:25:49,233 - Epoch: [7][ 1130/ 1200] Overall Loss 0.417996 Objective Loss 0.417996 LR 0.001000 Time 0.014494 -2022-12-06 10:25:49,373 - Epoch: [7][ 1140/ 1200] Overall Loss 0.417606 Objective Loss 0.417606 LR 0.001000 Time 0.014490 -2022-12-06 10:25:49,517 - Epoch: [7][ 1150/ 1200] Overall Loss 0.417385 Objective Loss 0.417385 LR 0.001000 Time 0.014488 -2022-12-06 10:25:49,659 - Epoch: [7][ 1160/ 1200] Overall Loss 0.417458 Objective Loss 0.417458 LR 0.001000 Time 0.014485 -2022-12-06 10:25:49,803 - Epoch: [7][ 1170/ 1200] Overall Loss 0.417501 Objective Loss 0.417501 LR 0.001000 Time 0.014484 -2022-12-06 10:25:49,943 - Epoch: [7][ 1180/ 1200] Overall Loss 0.417825 Objective Loss 0.417825 LR 0.001000 Time 0.014480 -2022-12-06 10:25:50,087 - Epoch: [7][ 1190/ 1200] Overall Loss 0.417388 Objective Loss 0.417388 LR 0.001000 Time 0.014479 -2022-12-06 10:25:50,268 - Epoch: [7][ 1200/ 1200] Overall Loss 0.417079 Objective Loss 0.417079 Top1 80.962343 Top5 97.489540 LR 0.001000 Time 0.014509 -2022-12-06 10:25:50,362 - --- validate (epoch=7)----------- -2022-12-06 10:25:50,362 - 34129 samples (256 per mini-batch) -2022-12-06 10:25:50,770 - Epoch: [7][ 10/ 134] Loss 0.390837 Top1 78.867188 Top5 97.070312 -2022-12-06 10:25:50,862 - Epoch: [7][ 20/ 134] Loss 0.385288 Top1 79.687500 Top5 97.050781 -2022-12-06 10:25:50,952 - Epoch: [7][ 30/ 134] Loss 0.384682 Top1 80.117188 Top5 97.135417 -2022-12-06 10:25:51,041 - Epoch: [7][ 40/ 134] Loss 0.394770 Top1 79.755859 Top5 97.197266 -2022-12-06 10:25:51,131 - Epoch: [7][ 50/ 134] Loss 0.397878 Top1 79.664062 Top5 97.179688 -2022-12-06 10:25:51,229 - Epoch: [7][ 60/ 134] Loss 0.399916 Top1 79.576823 Top5 97.174479 -2022-12-06 10:25:51,325 - Epoch: [7][ 70/ 134] Loss 0.398465 Top1 79.720982 Top5 97.198661 -2022-12-06 10:25:51,422 - Epoch: [7][ 80/ 134] Loss 0.391737 Top1 80.009766 Top5 97.270508 -2022-12-06 10:25:51,517 - Epoch: [7][ 90/ 134] Loss 0.389738 Top1 80.034722 Top5 97.309028 -2022-12-06 10:25:51,610 - Epoch: [7][ 100/ 134] Loss 0.388554 Top1 80.074219 Top5 97.359375 -2022-12-06 10:25:51,706 - Epoch: [7][ 110/ 134] Loss 0.389315 Top1 80.000000 Top5 97.343750 -2022-12-06 10:25:51,795 - Epoch: [7][ 120/ 134] Loss 0.390331 Top1 79.954427 Top5 97.402344 -2022-12-06 10:25:51,889 - Epoch: [7][ 130/ 134] Loss 0.392975 Top1 79.879808 Top5 97.370793 -2022-12-06 10:25:51,911 - Epoch: [7][ 134/ 134] Loss 0.392779 Top1 79.917372 Top5 97.365877 -2022-12-06 10:25:51,997 - ==> Top1: 79.917 Top5: 97.366 Loss: 0.393 - -2022-12-06 10:25:51,998 - ==> Confusion: -[[ 864 4 0 1 3 5 0 1 9 73 0 3 1 6 7 1 3 7 4 0 4] - [ 2 904 1 2 5 30 3 22 5 1 6 3 2 3 5 0 7 2 15 0 9] - [ 11 3 929 32 3 6 24 26 0 2 12 4 1 8 3 3 1 4 5 1 25] - [ 2 11 11 896 1 4 2 2 2 0 20 0 1 3 31 0 0 7 23 0 4] - [ 19 18 1 1 899 10 0 5 1 12 2 1 1 10 11 6 13 3 1 0 6] - [ 3 54 0 3 3 886 0 36 4 2 2 21 1 31 1 1 1 1 4 6 9] - [ 0 10 16 4 2 2 1028 12 0 0 8 4 5 1 0 6 2 4 1 7 6] - [ 2 13 3 2 1 27 3 923 3 0 8 3 1 2 0 2 0 1 46 4 10] - [ 6 2 0 0 0 0 0 5 958 41 5 2 5 12 19 0 0 2 6 0 1] - [ 74 1 1 0 0 1 0 1 39 848 0 1 0 18 2 0 1 3 1 1 9] - [ 1 5 2 7 0 0 1 5 20 2 934 1 2 18 9 0 0 0 11 0 1] - [ 2 5 2 0 1 17 1 5 2 0 3 928 34 10 2 3 8 13 4 8 3] - [ 0 5 1 9 3 3 1 1 0 0 1 43 803 6 4 9 2 62 3 3 10] - [ 4 3 0 1 0 10 0 4 21 16 11 10 4 925 1 1 4 1 0 0 7] - [ 9 5 0 13 3 1 0 1 38 7 3 3 4 6 1018 0 2 5 6 0 6] - [ 4 3 2 4 3 3 4 1 1 0 0 7 10 6 0 946 13 22 3 3 8] - [ 4 9 2 3 5 2 2 1 1 0 0 2 2 1 4 4 1009 4 3 4 10] - [ 2 1 0 2 0 0 1 1 3 1 1 7 13 2 2 5 2 983 4 0 6] - [ 4 2 3 14 0 1 1 26 3 0 8 4 5 1 15 0 1 1 911 2 6] - [ 1 9 1 0 1 10 9 23 0 0 0 32 7 6 0 3 9 9 1 943 16] - [ 204 376 147 133 91 317 40 205 147 135 264 201 350 487 361 123 262 117 360 166 8740]] - -2022-12-06 10:25:52,667 - ==> Best [Top1: 79.917 Top5: 97.366 Sparsity:0.00 Params: 5376 on epoch: 7] -2022-12-06 10:25:52,667 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:25:52,674 - - -2022-12-06 10:25:52,674 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:25:53,563 - Epoch: [8][ 10/ 1200] Overall Loss 0.376168 Objective Loss 0.376168 LR 0.001000 Time 0.088827 -2022-12-06 10:25:53,728 - Epoch: [8][ 20/ 1200] Overall Loss 0.377803 Objective Loss 0.377803 LR 0.001000 Time 0.052638 -2022-12-06 10:25:53,880 - Epoch: [8][ 30/ 1200] Overall Loss 0.374163 Objective Loss 0.374163 LR 0.001000 Time 0.040151 -2022-12-06 10:25:54,032 - Epoch: [8][ 40/ 1200] Overall Loss 0.379952 Objective Loss 0.379952 LR 0.001000 Time 0.033890 -2022-12-06 10:25:54,174 - Epoch: [8][ 50/ 1200] Overall Loss 0.377196 Objective Loss 0.377196 LR 0.001000 Time 0.029935 -2022-12-06 10:25:54,312 - Epoch: [8][ 60/ 1200] Overall Loss 0.383835 Objective Loss 0.383835 LR 0.001000 Time 0.027241 -2022-12-06 10:25:54,449 - Epoch: [8][ 70/ 1200] Overall Loss 0.384392 Objective Loss 0.384392 LR 0.001000 Time 0.025295 -2022-12-06 10:25:54,586 - Epoch: [8][ 80/ 1200] Overall Loss 0.382369 Objective Loss 0.382369 LR 0.001000 Time 0.023837 -2022-12-06 10:25:54,723 - Epoch: [8][ 90/ 1200] Overall Loss 0.387430 Objective Loss 0.387430 LR 0.001000 Time 0.022710 -2022-12-06 10:25:54,862 - Epoch: [8][ 100/ 1200] Overall Loss 0.387651 Objective Loss 0.387651 LR 0.001000 Time 0.021817 -2022-12-06 10:25:54,999 - Epoch: [8][ 110/ 1200] Overall Loss 0.383651 Objective Loss 0.383651 LR 0.001000 Time 0.021078 -2022-12-06 10:25:55,136 - Epoch: [8][ 120/ 1200] Overall Loss 0.383262 Objective Loss 0.383262 LR 0.001000 Time 0.020455 -2022-12-06 10:25:55,274 - Epoch: [8][ 130/ 1200] Overall Loss 0.386091 Objective Loss 0.386091 LR 0.001000 Time 0.019945 -2022-12-06 10:25:55,412 - Epoch: [8][ 140/ 1200] Overall Loss 0.385883 Objective Loss 0.385883 LR 0.001000 Time 0.019498 -2022-12-06 10:25:55,550 - Epoch: [8][ 150/ 1200] Overall Loss 0.387052 Objective Loss 0.387052 LR 0.001000 Time 0.019114 -2022-12-06 10:25:55,688 - Epoch: [8][ 160/ 1200] Overall Loss 0.387014 Objective Loss 0.387014 LR 0.001000 Time 0.018779 -2022-12-06 10:25:55,826 - Epoch: [8][ 170/ 1200] Overall Loss 0.389722 Objective Loss 0.389722 LR 0.001000 Time 0.018485 -2022-12-06 10:25:55,964 - Epoch: [8][ 180/ 1200] Overall Loss 0.390300 Objective Loss 0.390300 LR 0.001000 Time 0.018218 -2022-12-06 10:25:56,101 - Epoch: [8][ 190/ 1200] Overall Loss 0.391449 Objective Loss 0.391449 LR 0.001000 Time 0.017981 -2022-12-06 10:25:56,238 - Epoch: [8][ 200/ 1200] Overall Loss 0.393634 Objective Loss 0.393634 LR 0.001000 Time 0.017765 -2022-12-06 10:25:56,376 - Epoch: [8][ 210/ 1200] Overall Loss 0.393786 Objective Loss 0.393786 LR 0.001000 Time 0.017569 -2022-12-06 10:25:56,513 - Epoch: [8][ 220/ 1200] Overall Loss 0.393160 Objective Loss 0.393160 LR 0.001000 Time 0.017395 -2022-12-06 10:25:56,651 - Epoch: [8][ 230/ 1200] Overall Loss 0.393694 Objective Loss 0.393694 LR 0.001000 Time 0.017235 -2022-12-06 10:25:56,788 - Epoch: [8][ 240/ 1200] Overall Loss 0.393295 Objective Loss 0.393295 LR 0.001000 Time 0.017085 -2022-12-06 10:25:56,926 - Epoch: [8][ 250/ 1200] Overall Loss 0.392059 Objective Loss 0.392059 LR 0.001000 Time 0.016950 -2022-12-06 10:25:57,062 - Epoch: [8][ 260/ 1200] Overall Loss 0.391789 Objective Loss 0.391789 LR 0.001000 Time 0.016821 -2022-12-06 10:25:57,200 - Epoch: [8][ 270/ 1200] Overall Loss 0.391395 Objective Loss 0.391395 LR 0.001000 Time 0.016706 -2022-12-06 10:25:57,337 - Epoch: [8][ 280/ 1200] Overall Loss 0.392106 Objective Loss 0.392106 LR 0.001000 Time 0.016598 -2022-12-06 10:25:57,475 - Epoch: [8][ 290/ 1200] Overall Loss 0.392178 Objective Loss 0.392178 LR 0.001000 Time 0.016500 -2022-12-06 10:25:57,614 - Epoch: [8][ 300/ 1200] Overall Loss 0.393083 Objective Loss 0.393083 LR 0.001000 Time 0.016410 -2022-12-06 10:25:57,752 - Epoch: [8][ 310/ 1200] Overall Loss 0.392217 Objective Loss 0.392217 LR 0.001000 Time 0.016323 -2022-12-06 10:25:57,889 - Epoch: [8][ 320/ 1200] Overall Loss 0.391812 Objective Loss 0.391812 LR 0.001000 Time 0.016241 -2022-12-06 10:25:58,027 - Epoch: [8][ 330/ 1200] Overall Loss 0.390827 Objective Loss 0.390827 LR 0.001000 Time 0.016164 -2022-12-06 10:25:58,164 - Epoch: [8][ 340/ 1200] Overall Loss 0.390686 Objective Loss 0.390686 LR 0.001000 Time 0.016089 -2022-12-06 10:25:58,303 - Epoch: [8][ 350/ 1200] Overall Loss 0.391387 Objective Loss 0.391387 LR 0.001000 Time 0.016027 -2022-12-06 10:25:58,442 - Epoch: [8][ 360/ 1200] Overall Loss 0.391598 Objective Loss 0.391598 LR 0.001000 Time 0.015965 -2022-12-06 10:25:58,579 - Epoch: [8][ 370/ 1200] Overall Loss 0.391084 Objective Loss 0.391084 LR 0.001000 Time 0.015904 -2022-12-06 10:25:58,717 - Epoch: [8][ 380/ 1200] Overall Loss 0.390387 Objective Loss 0.390387 LR 0.001000 Time 0.015845 -2022-12-06 10:25:58,855 - Epoch: [8][ 390/ 1200] Overall Loss 0.389857 Objective Loss 0.389857 LR 0.001000 Time 0.015792 -2022-12-06 10:25:58,992 - Epoch: [8][ 400/ 1200] Overall Loss 0.388778 Objective Loss 0.388778 LR 0.001000 Time 0.015739 -2022-12-06 10:25:59,131 - Epoch: [8][ 410/ 1200] Overall Loss 0.388559 Objective Loss 0.388559 LR 0.001000 Time 0.015692 -2022-12-06 10:25:59,268 - Epoch: [8][ 420/ 1200] Overall Loss 0.387942 Objective Loss 0.387942 LR 0.001000 Time 0.015644 -2022-12-06 10:25:59,407 - Epoch: [8][ 430/ 1200] Overall Loss 0.388118 Objective Loss 0.388118 LR 0.001000 Time 0.015602 -2022-12-06 10:25:59,545 - Epoch: [8][ 440/ 1200] Overall Loss 0.388474 Objective Loss 0.388474 LR 0.001000 Time 0.015560 -2022-12-06 10:25:59,683 - Epoch: [8][ 450/ 1200] Overall Loss 0.388517 Objective Loss 0.388517 LR 0.001000 Time 0.015520 -2022-12-06 10:25:59,820 - Epoch: [8][ 460/ 1200] Overall Loss 0.388938 Objective Loss 0.388938 LR 0.001000 Time 0.015479 -2022-12-06 10:25:59,959 - Epoch: [8][ 470/ 1200] Overall Loss 0.389105 Objective Loss 0.389105 LR 0.001000 Time 0.015444 -2022-12-06 10:26:00,097 - Epoch: [8][ 480/ 1200] Overall Loss 0.389051 Objective Loss 0.389051 LR 0.001000 Time 0.015408 -2022-12-06 10:26:00,235 - Epoch: [8][ 490/ 1200] Overall Loss 0.388787 Objective Loss 0.388787 LR 0.001000 Time 0.015374 -2022-12-06 10:26:00,372 - Epoch: [8][ 500/ 1200] Overall Loss 0.388076 Objective Loss 0.388076 LR 0.001000 Time 0.015339 -2022-12-06 10:26:00,510 - Epoch: [8][ 510/ 1200] Overall Loss 0.388422 Objective Loss 0.388422 LR 0.001000 Time 0.015309 -2022-12-06 10:26:00,648 - Epoch: [8][ 520/ 1200] Overall Loss 0.388913 Objective Loss 0.388913 LR 0.001000 Time 0.015278 -2022-12-06 10:26:00,786 - Epoch: [8][ 530/ 1200] Overall Loss 0.389265 Objective Loss 0.389265 LR 0.001000 Time 0.015249 -2022-12-06 10:26:00,924 - Epoch: [8][ 540/ 1200] Overall Loss 0.389010 Objective Loss 0.389010 LR 0.001000 Time 0.015221 -2022-12-06 10:26:01,062 - Epoch: [8][ 550/ 1200] Overall Loss 0.389450 Objective Loss 0.389450 LR 0.001000 Time 0.015195 -2022-12-06 10:26:01,200 - Epoch: [8][ 560/ 1200] Overall Loss 0.389434 Objective Loss 0.389434 LR 0.001000 Time 0.015168 -2022-12-06 10:26:01,338 - Epoch: [8][ 570/ 1200] Overall Loss 0.389731 Objective Loss 0.389731 LR 0.001000 Time 0.015144 -2022-12-06 10:26:01,476 - Epoch: [8][ 580/ 1200] Overall Loss 0.390356 Objective Loss 0.390356 LR 0.001000 Time 0.015119 -2022-12-06 10:26:01,615 - Epoch: [8][ 590/ 1200] Overall Loss 0.390532 Objective Loss 0.390532 LR 0.001000 Time 0.015099 -2022-12-06 10:26:01,752 - Epoch: [8][ 600/ 1200] Overall Loss 0.390172 Objective Loss 0.390172 LR 0.001000 Time 0.015074 -2022-12-06 10:26:01,891 - Epoch: [8][ 610/ 1200] Overall Loss 0.390195 Objective Loss 0.390195 LR 0.001000 Time 0.015054 -2022-12-06 10:26:02,029 - Epoch: [8][ 620/ 1200] Overall Loss 0.389960 Objective Loss 0.389960 LR 0.001000 Time 0.015033 -2022-12-06 10:26:02,167 - Epoch: [8][ 630/ 1200] Overall Loss 0.390190 Objective Loss 0.390190 LR 0.001000 Time 0.015012 -2022-12-06 10:26:02,304 - Epoch: [8][ 640/ 1200] Overall Loss 0.390634 Objective Loss 0.390634 LR 0.001000 Time 0.014991 -2022-12-06 10:26:02,443 - Epoch: [8][ 650/ 1200] Overall Loss 0.390908 Objective Loss 0.390908 LR 0.001000 Time 0.014973 -2022-12-06 10:26:02,582 - Epoch: [8][ 660/ 1200] Overall Loss 0.390686 Objective Loss 0.390686 LR 0.001000 Time 0.014956 -2022-12-06 10:26:02,720 - Epoch: [8][ 670/ 1200] Overall Loss 0.390951 Objective Loss 0.390951 LR 0.001000 Time 0.014938 -2022-12-06 10:26:02,857 - Epoch: [8][ 680/ 1200] Overall Loss 0.391088 Objective Loss 0.391088 LR 0.001000 Time 0.014918 -2022-12-06 10:26:02,995 - Epoch: [8][ 690/ 1200] Overall Loss 0.391434 Objective Loss 0.391434 LR 0.001000 Time 0.014902 -2022-12-06 10:26:03,132 - Epoch: [8][ 700/ 1200] Overall Loss 0.391668 Objective Loss 0.391668 LR 0.001000 Time 0.014884 -2022-12-06 10:26:03,271 - Epoch: [8][ 710/ 1200] Overall Loss 0.392094 Objective Loss 0.392094 LR 0.001000 Time 0.014869 -2022-12-06 10:26:03,407 - Epoch: [8][ 720/ 1200] Overall Loss 0.392188 Objective Loss 0.392188 LR 0.001000 Time 0.014851 -2022-12-06 10:26:03,545 - Epoch: [8][ 730/ 1200] Overall Loss 0.392290 Objective Loss 0.392290 LR 0.001000 Time 0.014836 -2022-12-06 10:26:03,684 - Epoch: [8][ 740/ 1200] Overall Loss 0.392430 Objective Loss 0.392430 LR 0.001000 Time 0.014822 -2022-12-06 10:26:03,825 - Epoch: [8][ 750/ 1200] Overall Loss 0.392883 Objective Loss 0.392883 LR 0.001000 Time 0.014812 -2022-12-06 10:26:03,967 - Epoch: [8][ 760/ 1200] Overall Loss 0.392912 Objective Loss 0.392912 LR 0.001000 Time 0.014802 -2022-12-06 10:26:04,106 - Epoch: [8][ 770/ 1200] Overall Loss 0.392756 Objective Loss 0.392756 LR 0.001000 Time 0.014790 -2022-12-06 10:26:04,246 - Epoch: [8][ 780/ 1200] Overall Loss 0.392359 Objective Loss 0.392359 LR 0.001000 Time 0.014780 -2022-12-06 10:26:04,387 - Epoch: [8][ 790/ 1200] Overall Loss 0.392647 Objective Loss 0.392647 LR 0.001000 Time 0.014770 -2022-12-06 10:26:04,528 - Epoch: [8][ 800/ 1200] Overall Loss 0.392733 Objective Loss 0.392733 LR 0.001000 Time 0.014761 -2022-12-06 10:26:04,668 - Epoch: [8][ 810/ 1200] Overall Loss 0.392797 Objective Loss 0.392797 LR 0.001000 Time 0.014750 -2022-12-06 10:26:04,808 - Epoch: [8][ 820/ 1200] Overall Loss 0.393140 Objective Loss 0.393140 LR 0.001000 Time 0.014741 -2022-12-06 10:26:04,947 - Epoch: [8][ 830/ 1200] Overall Loss 0.393145 Objective Loss 0.393145 LR 0.001000 Time 0.014729 -2022-12-06 10:26:05,086 - Epoch: [8][ 840/ 1200] Overall Loss 0.393009 Objective Loss 0.393009 LR 0.001000 Time 0.014719 -2022-12-06 10:26:05,227 - Epoch: [8][ 850/ 1200] Overall Loss 0.393364 Objective Loss 0.393364 LR 0.001000 Time 0.014709 -2022-12-06 10:26:05,368 - Epoch: [8][ 860/ 1200] Overall Loss 0.393285 Objective Loss 0.393285 LR 0.001000 Time 0.014700 -2022-12-06 10:26:05,508 - Epoch: [8][ 870/ 1200] Overall Loss 0.393324 Objective Loss 0.393324 LR 0.001000 Time 0.014692 -2022-12-06 10:26:05,649 - Epoch: [8][ 880/ 1200] Overall Loss 0.393847 Objective Loss 0.393847 LR 0.001000 Time 0.014684 -2022-12-06 10:26:05,789 - Epoch: [8][ 890/ 1200] Overall Loss 0.393797 Objective Loss 0.393797 LR 0.001000 Time 0.014676 -2022-12-06 10:26:05,930 - Epoch: [8][ 900/ 1200] Overall Loss 0.394502 Objective Loss 0.394502 LR 0.001000 Time 0.014669 -2022-12-06 10:26:06,070 - Epoch: [8][ 910/ 1200] Overall Loss 0.394590 Objective Loss 0.394590 LR 0.001000 Time 0.014661 -2022-12-06 10:26:06,208 - Epoch: [8][ 920/ 1200] Overall Loss 0.394717 Objective Loss 0.394717 LR 0.001000 Time 0.014650 -2022-12-06 10:26:06,349 - Epoch: [8][ 930/ 1200] Overall Loss 0.394854 Objective Loss 0.394854 LR 0.001000 Time 0.014643 -2022-12-06 10:26:06,488 - Epoch: [8][ 940/ 1200] Overall Loss 0.395310 Objective Loss 0.395310 LR 0.001000 Time 0.014635 -2022-12-06 10:26:06,630 - Epoch: [8][ 950/ 1200] Overall Loss 0.395367 Objective Loss 0.395367 LR 0.001000 Time 0.014629 -2022-12-06 10:26:06,771 - Epoch: [8][ 960/ 1200] Overall Loss 0.395427 Objective Loss 0.395427 LR 0.001000 Time 0.014623 -2022-12-06 10:26:06,910 - Epoch: [8][ 970/ 1200] Overall Loss 0.395098 Objective Loss 0.395098 LR 0.001000 Time 0.014615 -2022-12-06 10:26:07,051 - Epoch: [8][ 980/ 1200] Overall Loss 0.394789 Objective Loss 0.394789 LR 0.001000 Time 0.014609 -2022-12-06 10:26:07,192 - Epoch: [8][ 990/ 1200] Overall Loss 0.394559 Objective Loss 0.394559 LR 0.001000 Time 0.014603 -2022-12-06 10:26:07,332 - Epoch: [8][ 1000/ 1200] Overall Loss 0.394522 Objective Loss 0.394522 LR 0.001000 Time 0.014595 -2022-12-06 10:26:07,471 - Epoch: [8][ 1010/ 1200] Overall Loss 0.394802 Objective Loss 0.394802 LR 0.001000 Time 0.014588 -2022-12-06 10:26:07,612 - Epoch: [8][ 1020/ 1200] Overall Loss 0.395055 Objective Loss 0.395055 LR 0.001000 Time 0.014582 -2022-12-06 10:26:07,753 - Epoch: [8][ 1030/ 1200] Overall Loss 0.394639 Objective Loss 0.394639 LR 0.001000 Time 0.014577 -2022-12-06 10:26:07,895 - Epoch: [8][ 1040/ 1200] Overall Loss 0.394477 Objective Loss 0.394477 LR 0.001000 Time 0.014571 -2022-12-06 10:26:08,033 - Epoch: [8][ 1050/ 1200] Overall Loss 0.394297 Objective Loss 0.394297 LR 0.001000 Time 0.014564 -2022-12-06 10:26:08,175 - Epoch: [8][ 1060/ 1200] Overall Loss 0.394498 Objective Loss 0.394498 LR 0.001000 Time 0.014560 -2022-12-06 10:26:08,315 - Epoch: [8][ 1070/ 1200] Overall Loss 0.394257 Objective Loss 0.394257 LR 0.001000 Time 0.014554 -2022-12-06 10:26:08,454 - Epoch: [8][ 1080/ 1200] Overall Loss 0.394183 Objective Loss 0.394183 LR 0.001000 Time 0.014547 -2022-12-06 10:26:08,594 - Epoch: [8][ 1090/ 1200] Overall Loss 0.394345 Objective Loss 0.394345 LR 0.001000 Time 0.014540 -2022-12-06 10:26:08,736 - Epoch: [8][ 1100/ 1200] Overall Loss 0.394234 Objective Loss 0.394234 LR 0.001000 Time 0.014535 -2022-12-06 10:26:08,876 - Epoch: [8][ 1110/ 1200] Overall Loss 0.394463 Objective Loss 0.394463 LR 0.001000 Time 0.014530 -2022-12-06 10:26:09,018 - Epoch: [8][ 1120/ 1200] Overall Loss 0.394400 Objective Loss 0.394400 LR 0.001000 Time 0.014525 -2022-12-06 10:26:09,157 - Epoch: [8][ 1130/ 1200] Overall Loss 0.394558 Objective Loss 0.394558 LR 0.001000 Time 0.014520 -2022-12-06 10:26:09,301 - Epoch: [8][ 1140/ 1200] Overall Loss 0.394463 Objective Loss 0.394463 LR 0.001000 Time 0.014518 -2022-12-06 10:26:09,443 - Epoch: [8][ 1150/ 1200] Overall Loss 0.394598 Objective Loss 0.394598 LR 0.001000 Time 0.014514 -2022-12-06 10:26:09,583 - Epoch: [8][ 1160/ 1200] Overall Loss 0.394388 Objective Loss 0.394388 LR 0.001000 Time 0.014508 -2022-12-06 10:26:09,722 - Epoch: [8][ 1170/ 1200] Overall Loss 0.394499 Objective Loss 0.394499 LR 0.001000 Time 0.014503 -2022-12-06 10:26:09,862 - Epoch: [8][ 1180/ 1200] Overall Loss 0.394463 Objective Loss 0.394463 LR 0.001000 Time 0.014497 -2022-12-06 10:26:10,002 - Epoch: [8][ 1190/ 1200] Overall Loss 0.394580 Objective Loss 0.394580 LR 0.001000 Time 0.014492 -2022-12-06 10:26:10,186 - Epoch: [8][ 1200/ 1200] Overall Loss 0.394589 Objective Loss 0.394589 Top1 76.569038 Top5 98.326360 LR 0.001000 Time 0.014523 -2022-12-06 10:26:10,285 - --- validate (epoch=8)----------- -2022-12-06 10:26:10,285 - 34129 samples (256 per mini-batch) -2022-12-06 10:26:10,704 - Epoch: [8][ 10/ 134] Loss 0.386521 Top1 78.476562 Top5 97.382812 -2022-12-06 10:26:10,800 - Epoch: [8][ 20/ 134] Loss 0.390789 Top1 78.496094 Top5 97.402344 -2022-12-06 10:26:10,898 - Epoch: [8][ 30/ 134] Loss 0.387783 Top1 78.684896 Top5 97.447917 -2022-12-06 10:26:10,995 - Epoch: [8][ 40/ 134] Loss 0.379797 Top1 79.443359 Top5 97.441406 -2022-12-06 10:26:11,094 - Epoch: [8][ 50/ 134] Loss 0.386155 Top1 79.460938 Top5 97.320312 -2022-12-06 10:26:11,182 - Epoch: [8][ 60/ 134] Loss 0.389353 Top1 79.641927 Top5 97.330729 -2022-12-06 10:26:11,272 - Epoch: [8][ 70/ 134] Loss 0.389516 Top1 79.732143 Top5 97.405134 -2022-12-06 10:26:11,368 - Epoch: [8][ 80/ 134] Loss 0.389021 Top1 79.580078 Top5 97.353516 -2022-12-06 10:26:11,462 - Epoch: [8][ 90/ 134] Loss 0.385721 Top1 79.765625 Top5 97.356771 -2022-12-06 10:26:11,554 - Epoch: [8][ 100/ 134] Loss 0.387039 Top1 79.613281 Top5 97.339844 -2022-12-06 10:26:11,645 - Epoch: [8][ 110/ 134] Loss 0.383308 Top1 79.659091 Top5 97.340199 -2022-12-06 10:26:11,734 - Epoch: [8][ 120/ 134] Loss 0.384433 Top1 79.599609 Top5 97.314453 -2022-12-06 10:26:11,836 - Epoch: [8][ 130/ 134] Loss 0.384918 Top1 79.648438 Top5 97.298678 -2022-12-06 10:26:11,857 - Epoch: [8][ 134/ 134] Loss 0.386257 Top1 79.633157 Top5 97.304345 -2022-12-06 10:26:11,947 - ==> Top1: 79.633 Top5: 97.304 Loss: 0.386 - -2022-12-06 10:26:11,948 - ==> Confusion: -[[ 846 5 7 0 13 3 0 3 7 80 0 1 1 5 13 1 3 0 1 0 7] - [ 0 890 3 0 16 40 3 28 3 2 3 6 1 1 7 1 6 1 5 1 10] - [ 5 3 957 12 3 7 38 29 1 2 7 5 1 9 4 4 0 0 4 4 8] - [ 1 7 41 874 0 5 4 5 2 0 21 2 6 3 30 0 0 5 8 1 5] - [ 13 7 2 0 945 6 0 3 1 8 1 4 0 4 4 4 8 1 2 3 4] - [ 1 42 1 0 10 905 4 35 3 2 1 13 0 22 3 3 1 0 4 12 7] - [ 1 9 21 0 0 2 1046 11 0 0 2 3 0 0 1 4 0 0 3 14 1] - [ 0 13 4 3 0 32 5 953 0 1 3 5 0 1 2 1 0 0 14 12 5] - [ 6 4 0 1 1 3 0 2 949 47 4 3 2 15 17 0 0 1 3 2 4] - [ 69 1 4 0 4 3 1 3 26 858 0 0 0 15 9 1 0 0 0 2 5] - [ 1 5 6 3 2 2 5 9 12 1 934 2 1 16 6 0 0 0 9 1 4] - [ 2 4 6 0 1 17 5 11 3 0 2 941 16 9 0 9 5 4 1 12 3] - [ 3 5 4 4 2 3 3 4 0 0 0 73 807 4 4 14 5 14 2 6 12] - [ 2 4 0 0 1 13 0 4 7 17 4 5 2 936 3 4 3 0 0 3 15] - [ 9 3 2 9 9 4 0 1 14 1 5 1 7 5 1043 1 3 1 6 0 6] - [ 1 4 2 2 5 2 6 0 0 0 0 12 4 2 2 972 9 3 3 8 6] - [ 3 2 7 0 9 2 1 1 2 0 0 4 1 1 5 9 1009 1 0 8 7] - [ 2 4 4 5 0 3 5 1 0 1 2 24 46 2 2 40 3 882 1 2 7] - [ 4 2 7 17 0 2 1 61 5 0 12 1 2 2 13 0 1 0 861 7 10] - [ 0 9 3 0 2 9 14 9 0 1 1 16 2 4 1 6 10 0 0 988 5] - [ 204 362 284 71 259 305 133 265 94 113 227 183 354 444 251 195 266 41 164 429 8582]] - -2022-12-06 10:26:12,528 - ==> Best [Top1: 79.917 Top5: 97.366 Sparsity:0.00 Params: 5376 on epoch: 7] -2022-12-06 10:26:12,529 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:26:12,535 - - -2022-12-06 10:26:12,535 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:26:13,668 - Epoch: [9][ 10/ 1200] Overall Loss 0.385909 Objective Loss 0.385909 LR 0.001000 Time 0.113273 -2022-12-06 10:26:13,828 - Epoch: [9][ 20/ 1200] Overall Loss 0.402686 Objective Loss 0.402686 LR 0.001000 Time 0.064583 -2022-12-06 10:26:13,972 - Epoch: [9][ 30/ 1200] Overall Loss 0.395198 Objective Loss 0.395198 LR 0.001000 Time 0.047851 -2022-12-06 10:26:14,123 - Epoch: [9][ 40/ 1200] Overall Loss 0.386289 Objective Loss 0.386289 LR 0.001000 Time 0.039659 -2022-12-06 10:26:14,267 - Epoch: [9][ 50/ 1200] Overall Loss 0.380852 Objective Loss 0.380852 LR 0.001000 Time 0.034596 -2022-12-06 10:26:14,419 - Epoch: [9][ 60/ 1200] Overall Loss 0.380054 Objective Loss 0.380054 LR 0.001000 Time 0.031341 -2022-12-06 10:26:14,568 - Epoch: [9][ 70/ 1200] Overall Loss 0.379258 Objective Loss 0.379258 LR 0.001000 Time 0.028991 -2022-12-06 10:26:14,720 - Epoch: [9][ 80/ 1200] Overall Loss 0.380401 Objective Loss 0.380401 LR 0.001000 Time 0.027254 -2022-12-06 10:26:14,864 - Epoch: [9][ 90/ 1200] Overall Loss 0.378457 Objective Loss 0.378457 LR 0.001000 Time 0.025828 -2022-12-06 10:26:15,010 - Epoch: [9][ 100/ 1200] Overall Loss 0.376138 Objective Loss 0.376138 LR 0.001000 Time 0.024696 -2022-12-06 10:26:15,154 - Epoch: [9][ 110/ 1200] Overall Loss 0.375139 Objective Loss 0.375139 LR 0.001000 Time 0.023753 -2022-12-06 10:26:15,300 - Epoch: [9][ 120/ 1200] Overall Loss 0.375217 Objective Loss 0.375217 LR 0.001000 Time 0.022989 -2022-12-06 10:26:15,443 - Epoch: [9][ 130/ 1200] Overall Loss 0.374734 Objective Loss 0.374734 LR 0.001000 Time 0.022314 -2022-12-06 10:26:15,589 - Epoch: [9][ 140/ 1200] Overall Loss 0.373754 Objective Loss 0.373754 LR 0.001000 Time 0.021757 -2022-12-06 10:26:15,731 - Epoch: [9][ 150/ 1200] Overall Loss 0.372668 Objective Loss 0.372668 LR 0.001000 Time 0.021253 -2022-12-06 10:26:15,877 - Epoch: [9][ 160/ 1200] Overall Loss 0.373574 Objective Loss 0.373574 LR 0.001000 Time 0.020833 -2022-12-06 10:26:16,020 - Epoch: [9][ 170/ 1200] Overall Loss 0.373918 Objective Loss 0.373918 LR 0.001000 Time 0.020448 -2022-12-06 10:26:16,166 - Epoch: [9][ 180/ 1200] Overall Loss 0.375466 Objective Loss 0.375466 LR 0.001000 Time 0.020118 -2022-12-06 10:26:16,309 - Epoch: [9][ 190/ 1200] Overall Loss 0.376169 Objective Loss 0.376169 LR 0.001000 Time 0.019810 -2022-12-06 10:26:16,455 - Epoch: [9][ 200/ 1200] Overall Loss 0.376848 Objective Loss 0.376848 LR 0.001000 Time 0.019545 -2022-12-06 10:26:16,599 - Epoch: [9][ 210/ 1200] Overall Loss 0.376053 Objective Loss 0.376053 LR 0.001000 Time 0.019296 -2022-12-06 10:26:16,745 - Epoch: [9][ 220/ 1200] Overall Loss 0.376377 Objective Loss 0.376377 LR 0.001000 Time 0.019081 -2022-12-06 10:26:16,887 - Epoch: [9][ 230/ 1200] Overall Loss 0.375681 Objective Loss 0.375681 LR 0.001000 Time 0.018869 -2022-12-06 10:26:17,033 - Epoch: [9][ 240/ 1200] Overall Loss 0.376124 Objective Loss 0.376124 LR 0.001000 Time 0.018688 -2022-12-06 10:26:17,177 - Epoch: [9][ 250/ 1200] Overall Loss 0.376249 Objective Loss 0.376249 LR 0.001000 Time 0.018511 -2022-12-06 10:26:17,322 - Epoch: [9][ 260/ 1200] Overall Loss 0.377317 Objective Loss 0.377317 LR 0.001000 Time 0.018358 -2022-12-06 10:26:17,466 - Epoch: [9][ 270/ 1200] Overall Loss 0.377279 Objective Loss 0.377279 LR 0.001000 Time 0.018208 -2022-12-06 10:26:17,612 - Epoch: [9][ 280/ 1200] Overall Loss 0.377842 Objective Loss 0.377842 LR 0.001000 Time 0.018076 -2022-12-06 10:26:17,755 - Epoch: [9][ 290/ 1200] Overall Loss 0.378597 Objective Loss 0.378597 LR 0.001000 Time 0.017946 -2022-12-06 10:26:17,902 - Epoch: [9][ 300/ 1200] Overall Loss 0.378372 Objective Loss 0.378372 LR 0.001000 Time 0.017834 -2022-12-06 10:26:18,047 - Epoch: [9][ 310/ 1200] Overall Loss 0.379383 Objective Loss 0.379383 LR 0.001000 Time 0.017724 -2022-12-06 10:26:18,192 - Epoch: [9][ 320/ 1200] Overall Loss 0.380430 Objective Loss 0.380430 LR 0.001000 Time 0.017624 -2022-12-06 10:26:18,336 - Epoch: [9][ 330/ 1200] Overall Loss 0.380942 Objective Loss 0.380942 LR 0.001000 Time 0.017523 -2022-12-06 10:26:18,482 - Epoch: [9][ 340/ 1200] Overall Loss 0.381099 Objective Loss 0.381099 LR 0.001000 Time 0.017437 -2022-12-06 10:26:18,626 - Epoch: [9][ 350/ 1200] Overall Loss 0.382182 Objective Loss 0.382182 LR 0.001000 Time 0.017344 -2022-12-06 10:26:18,772 - Epoch: [9][ 360/ 1200] Overall Loss 0.382732 Objective Loss 0.382732 LR 0.001000 Time 0.017266 -2022-12-06 10:26:18,916 - Epoch: [9][ 370/ 1200] Overall Loss 0.381879 Objective Loss 0.381879 LR 0.001000 Time 0.017187 -2022-12-06 10:26:19,062 - Epoch: [9][ 380/ 1200] Overall Loss 0.382005 Objective Loss 0.382005 LR 0.001000 Time 0.017117 -2022-12-06 10:26:19,206 - Epoch: [9][ 390/ 1200] Overall Loss 0.381282 Objective Loss 0.381282 LR 0.001000 Time 0.017048 -2022-12-06 10:26:19,353 - Epoch: [9][ 400/ 1200] Overall Loss 0.381164 Objective Loss 0.381164 LR 0.001000 Time 0.016987 -2022-12-06 10:26:19,499 - Epoch: [9][ 410/ 1200] Overall Loss 0.380565 Objective Loss 0.380565 LR 0.001000 Time 0.016928 -2022-12-06 10:26:19,647 - Epoch: [9][ 420/ 1200] Overall Loss 0.380365 Objective Loss 0.380365 LR 0.001000 Time 0.016877 -2022-12-06 10:26:19,796 - Epoch: [9][ 430/ 1200] Overall Loss 0.380382 Objective Loss 0.380382 LR 0.001000 Time 0.016829 -2022-12-06 10:26:19,946 - Epoch: [9][ 440/ 1200] Overall Loss 0.381094 Objective Loss 0.381094 LR 0.001000 Time 0.016786 -2022-12-06 10:26:20,096 - Epoch: [9][ 450/ 1200] Overall Loss 0.380236 Objective Loss 0.380236 LR 0.001000 Time 0.016744 -2022-12-06 10:26:20,246 - Epoch: [9][ 460/ 1200] Overall Loss 0.381017 Objective Loss 0.381017 LR 0.001000 Time 0.016705 -2022-12-06 10:26:20,396 - Epoch: [9][ 470/ 1200] Overall Loss 0.380109 Objective Loss 0.380109 LR 0.001000 Time 0.016667 -2022-12-06 10:26:20,543 - Epoch: [9][ 480/ 1200] Overall Loss 0.380352 Objective Loss 0.380352 LR 0.001000 Time 0.016627 -2022-12-06 10:26:20,686 - Epoch: [9][ 490/ 1200] Overall Loss 0.380198 Objective Loss 0.380198 LR 0.001000 Time 0.016577 -2022-12-06 10:26:20,831 - Epoch: [9][ 500/ 1200] Overall Loss 0.380172 Objective Loss 0.380172 LR 0.001000 Time 0.016535 -2022-12-06 10:26:20,972 - Epoch: [9][ 510/ 1200] Overall Loss 0.380167 Objective Loss 0.380167 LR 0.001000 Time 0.016487 -2022-12-06 10:26:21,106 - Epoch: [9][ 520/ 1200] Overall Loss 0.380547 Objective Loss 0.380547 LR 0.001000 Time 0.016423 -2022-12-06 10:26:21,240 - Epoch: [9][ 530/ 1200] Overall Loss 0.380280 Objective Loss 0.380280 LR 0.001000 Time 0.016362 -2022-12-06 10:26:21,373 - Epoch: [9][ 540/ 1200] Overall Loss 0.380033 Objective Loss 0.380033 LR 0.001000 Time 0.016303 -2022-12-06 10:26:21,507 - Epoch: [9][ 550/ 1200] Overall Loss 0.380425 Objective Loss 0.380425 LR 0.001000 Time 0.016246 -2022-12-06 10:26:21,640 - Epoch: [9][ 560/ 1200] Overall Loss 0.380154 Objective Loss 0.380154 LR 0.001000 Time 0.016192 -2022-12-06 10:26:21,774 - Epoch: [9][ 570/ 1200] Overall Loss 0.379314 Objective Loss 0.379314 LR 0.001000 Time 0.016142 -2022-12-06 10:26:21,908 - Epoch: [9][ 580/ 1200] Overall Loss 0.379339 Objective Loss 0.379339 LR 0.001000 Time 0.016091 -2022-12-06 10:26:22,042 - Epoch: [9][ 590/ 1200] Overall Loss 0.379265 Objective Loss 0.379265 LR 0.001000 Time 0.016042 -2022-12-06 10:26:22,176 - Epoch: [9][ 600/ 1200] Overall Loss 0.379867 Objective Loss 0.379867 LR 0.001000 Time 0.015995 -2022-12-06 10:26:22,309 - Epoch: [9][ 610/ 1200] Overall Loss 0.380168 Objective Loss 0.380168 LR 0.001000 Time 0.015949 -2022-12-06 10:26:22,444 - Epoch: [9][ 620/ 1200] Overall Loss 0.378818 Objective Loss 0.378818 LR 0.001000 Time 0.015906 -2022-12-06 10:26:22,577 - Epoch: [9][ 630/ 1200] Overall Loss 0.379085 Objective Loss 0.379085 LR 0.001000 Time 0.015863 -2022-12-06 10:26:22,711 - Epoch: [9][ 640/ 1200] Overall Loss 0.378823 Objective Loss 0.378823 LR 0.001000 Time 0.015822 -2022-12-06 10:26:22,845 - Epoch: [9][ 650/ 1200] Overall Loss 0.378945 Objective Loss 0.378945 LR 0.001000 Time 0.015781 -2022-12-06 10:26:22,980 - Epoch: [9][ 660/ 1200] Overall Loss 0.378226 Objective Loss 0.378226 LR 0.001000 Time 0.015745 -2022-12-06 10:26:23,113 - Epoch: [9][ 670/ 1200] Overall Loss 0.378223 Objective Loss 0.378223 LR 0.001000 Time 0.015707 -2022-12-06 10:26:23,249 - Epoch: [9][ 680/ 1200] Overall Loss 0.378641 Objective Loss 0.378641 LR 0.001000 Time 0.015673 -2022-12-06 10:26:23,385 - Epoch: [9][ 690/ 1200] Overall Loss 0.377917 Objective Loss 0.377917 LR 0.001000 Time 0.015642 -2022-12-06 10:26:23,520 - Epoch: [9][ 700/ 1200] Overall Loss 0.378287 Objective Loss 0.378287 LR 0.001000 Time 0.015612 -2022-12-06 10:26:23,656 - Epoch: [9][ 710/ 1200] Overall Loss 0.378390 Objective Loss 0.378390 LR 0.001000 Time 0.015582 -2022-12-06 10:26:23,792 - Epoch: [9][ 720/ 1200] Overall Loss 0.377937 Objective Loss 0.377937 LR 0.001000 Time 0.015554 -2022-12-06 10:26:23,928 - Epoch: [9][ 730/ 1200] Overall Loss 0.378002 Objective Loss 0.378002 LR 0.001000 Time 0.015526 -2022-12-06 10:26:24,064 - Epoch: [9][ 740/ 1200] Overall Loss 0.377813 Objective Loss 0.377813 LR 0.001000 Time 0.015499 -2022-12-06 10:26:24,199 - Epoch: [9][ 750/ 1200] Overall Loss 0.378102 Objective Loss 0.378102 LR 0.001000 Time 0.015472 -2022-12-06 10:26:24,335 - Epoch: [9][ 760/ 1200] Overall Loss 0.378018 Objective Loss 0.378018 LR 0.001000 Time 0.015447 -2022-12-06 10:26:24,471 - Epoch: [9][ 770/ 1200] Overall Loss 0.378377 Objective Loss 0.378377 LR 0.001000 Time 0.015422 -2022-12-06 10:26:24,606 - Epoch: [9][ 780/ 1200] Overall Loss 0.378063 Objective Loss 0.378063 LR 0.001000 Time 0.015397 -2022-12-06 10:26:24,742 - Epoch: [9][ 790/ 1200] Overall Loss 0.378147 Objective Loss 0.378147 LR 0.001000 Time 0.015374 -2022-12-06 10:26:24,877 - Epoch: [9][ 800/ 1200] Overall Loss 0.378188 Objective Loss 0.378188 LR 0.001000 Time 0.015350 -2022-12-06 10:26:25,013 - Epoch: [9][ 810/ 1200] Overall Loss 0.378339 Objective Loss 0.378339 LR 0.001000 Time 0.015328 -2022-12-06 10:26:25,149 - Epoch: [9][ 820/ 1200] Overall Loss 0.378125 Objective Loss 0.378125 LR 0.001000 Time 0.015306 -2022-12-06 10:26:25,285 - Epoch: [9][ 830/ 1200] Overall Loss 0.377919 Objective Loss 0.377919 LR 0.001000 Time 0.015285 -2022-12-06 10:26:25,421 - Epoch: [9][ 840/ 1200] Overall Loss 0.377692 Objective Loss 0.377692 LR 0.001000 Time 0.015265 -2022-12-06 10:26:25,557 - Epoch: [9][ 850/ 1200] Overall Loss 0.377872 Objective Loss 0.377872 LR 0.001000 Time 0.015244 -2022-12-06 10:26:25,694 - Epoch: [9][ 860/ 1200] Overall Loss 0.377867 Objective Loss 0.377867 LR 0.001000 Time 0.015225 -2022-12-06 10:26:25,830 - Epoch: [9][ 870/ 1200] Overall Loss 0.377840 Objective Loss 0.377840 LR 0.001000 Time 0.015206 -2022-12-06 10:26:25,966 - Epoch: [9][ 880/ 1200] Overall Loss 0.377737 Objective Loss 0.377737 LR 0.001000 Time 0.015187 -2022-12-06 10:26:26,102 - Epoch: [9][ 890/ 1200] Overall Loss 0.377371 Objective Loss 0.377371 LR 0.001000 Time 0.015168 -2022-12-06 10:26:26,238 - Epoch: [9][ 900/ 1200] Overall Loss 0.376775 Objective Loss 0.376775 LR 0.001000 Time 0.015151 -2022-12-06 10:26:26,374 - Epoch: [9][ 910/ 1200] Overall Loss 0.376975 Objective Loss 0.376975 LR 0.001000 Time 0.015133 -2022-12-06 10:26:26,510 - Epoch: [9][ 920/ 1200] Overall Loss 0.377362 Objective Loss 0.377362 LR 0.001000 Time 0.015116 -2022-12-06 10:26:26,646 - Epoch: [9][ 930/ 1200] Overall Loss 0.377423 Objective Loss 0.377423 LR 0.001000 Time 0.015099 -2022-12-06 10:26:26,782 - Epoch: [9][ 940/ 1200] Overall Loss 0.376989 Objective Loss 0.376989 LR 0.001000 Time 0.015083 -2022-12-06 10:26:26,918 - Epoch: [9][ 950/ 1200] Overall Loss 0.377061 Objective Loss 0.377061 LR 0.001000 Time 0.015067 -2022-12-06 10:26:27,054 - Epoch: [9][ 960/ 1200] Overall Loss 0.377314 Objective Loss 0.377314 LR 0.001000 Time 0.015051 -2022-12-06 10:26:27,191 - Epoch: [9][ 970/ 1200] Overall Loss 0.377420 Objective Loss 0.377420 LR 0.001000 Time 0.015036 -2022-12-06 10:26:27,327 - Epoch: [9][ 980/ 1200] Overall Loss 0.377225 Objective Loss 0.377225 LR 0.001000 Time 0.015021 -2022-12-06 10:26:27,463 - Epoch: [9][ 990/ 1200] Overall Loss 0.377156 Objective Loss 0.377156 LR 0.001000 Time 0.015006 -2022-12-06 10:26:27,599 - Epoch: [9][ 1000/ 1200] Overall Loss 0.377468 Objective Loss 0.377468 LR 0.001000 Time 0.014992 -2022-12-06 10:26:27,735 - Epoch: [9][ 1010/ 1200] Overall Loss 0.378081 Objective Loss 0.378081 LR 0.001000 Time 0.014978 -2022-12-06 10:26:27,872 - Epoch: [9][ 1020/ 1200] Overall Loss 0.378221 Objective Loss 0.378221 LR 0.001000 Time 0.014964 -2022-12-06 10:26:28,008 - Epoch: [9][ 1030/ 1200] Overall Loss 0.378637 Objective Loss 0.378637 LR 0.001000 Time 0.014951 -2022-12-06 10:26:28,144 - Epoch: [9][ 1040/ 1200] Overall Loss 0.378963 Objective Loss 0.378963 LR 0.001000 Time 0.014938 -2022-12-06 10:26:28,280 - Epoch: [9][ 1050/ 1200] Overall Loss 0.378880 Objective Loss 0.378880 LR 0.001000 Time 0.014924 -2022-12-06 10:26:28,416 - Epoch: [9][ 1060/ 1200] Overall Loss 0.379312 Objective Loss 0.379312 LR 0.001000 Time 0.014912 -2022-12-06 10:26:28,552 - Epoch: [9][ 1070/ 1200] Overall Loss 0.379453 Objective Loss 0.379453 LR 0.001000 Time 0.014899 -2022-12-06 10:26:28,689 - Epoch: [9][ 1080/ 1200] Overall Loss 0.379885 Objective Loss 0.379885 LR 0.001000 Time 0.014887 -2022-12-06 10:26:28,825 - Epoch: [9][ 1090/ 1200] Overall Loss 0.379869 Objective Loss 0.379869 LR 0.001000 Time 0.014875 -2022-12-06 10:26:28,962 - Epoch: [9][ 1100/ 1200] Overall Loss 0.379819 Objective Loss 0.379819 LR 0.001000 Time 0.014864 -2022-12-06 10:26:29,098 - Epoch: [9][ 1110/ 1200] Overall Loss 0.379692 Objective Loss 0.379692 LR 0.001000 Time 0.014852 -2022-12-06 10:26:29,234 - Epoch: [9][ 1120/ 1200] Overall Loss 0.379719 Objective Loss 0.379719 LR 0.001000 Time 0.014841 -2022-12-06 10:26:29,370 - Epoch: [9][ 1130/ 1200] Overall Loss 0.379889 Objective Loss 0.379889 LR 0.001000 Time 0.014829 -2022-12-06 10:26:29,506 - Epoch: [9][ 1140/ 1200] Overall Loss 0.379720 Objective Loss 0.379720 LR 0.001000 Time 0.014818 -2022-12-06 10:26:29,642 - Epoch: [9][ 1150/ 1200] Overall Loss 0.379943 Objective Loss 0.379943 LR 0.001000 Time 0.014807 -2022-12-06 10:26:29,778 - Epoch: [9][ 1160/ 1200] Overall Loss 0.379788 Objective Loss 0.379788 LR 0.001000 Time 0.014796 -2022-12-06 10:26:29,914 - Epoch: [9][ 1170/ 1200] Overall Loss 0.380075 Objective Loss 0.380075 LR 0.001000 Time 0.014786 -2022-12-06 10:26:30,050 - Epoch: [9][ 1180/ 1200] Overall Loss 0.379906 Objective Loss 0.379906 LR 0.001000 Time 0.014775 -2022-12-06 10:26:30,186 - Epoch: [9][ 1190/ 1200] Overall Loss 0.380119 Objective Loss 0.380119 LR 0.001000 Time 0.014765 -2022-12-06 10:26:30,369 - Epoch: [9][ 1200/ 1200] Overall Loss 0.379927 Objective Loss 0.379927 Top1 82.217573 Top5 97.698745 LR 0.001000 Time 0.014794 -2022-12-06 10:26:30,461 - --- validate (epoch=9)----------- -2022-12-06 10:26:30,461 - 34129 samples (256 per mini-batch) -2022-12-06 10:26:30,864 - Epoch: [9][ 10/ 134] Loss 0.376466 Top1 81.210938 Top5 97.578125 -2022-12-06 10:26:30,956 - Epoch: [9][ 20/ 134] Loss 0.361710 Top1 81.601562 Top5 97.968750 -2022-12-06 10:26:31,048 - Epoch: [9][ 30/ 134] Loss 0.359514 Top1 81.549479 Top5 98.085938 -2022-12-06 10:26:31,142 - Epoch: [9][ 40/ 134] Loss 0.360605 Top1 81.748047 Top5 97.900391 -2022-12-06 10:26:31,229 - Epoch: [9][ 50/ 134] Loss 0.361306 Top1 82.101562 Top5 97.976562 -2022-12-06 10:26:31,323 - Epoch: [9][ 60/ 134] Loss 0.370572 Top1 81.868490 Top5 97.884115 -2022-12-06 10:26:31,412 - Epoch: [9][ 70/ 134] Loss 0.369336 Top1 81.835938 Top5 97.767857 -2022-12-06 10:26:31,506 - Epoch: [9][ 80/ 134] Loss 0.369026 Top1 81.669922 Top5 97.690430 -2022-12-06 10:26:31,600 - Epoch: [9][ 90/ 134] Loss 0.365289 Top1 81.701389 Top5 97.721354 -2022-12-06 10:26:31,691 - Epoch: [9][ 100/ 134] Loss 0.366288 Top1 81.695312 Top5 97.664062 -2022-12-06 10:26:31,781 - Epoch: [9][ 110/ 134] Loss 0.370115 Top1 81.626420 Top5 97.645597 -2022-12-06 10:26:31,874 - Epoch: [9][ 120/ 134] Loss 0.371313 Top1 81.575521 Top5 97.666016 -2022-12-06 10:26:31,972 - Epoch: [9][ 130/ 134] Loss 0.371235 Top1 81.610577 Top5 97.641226 -2022-12-06 10:26:31,994 - Epoch: [9][ 134/ 134] Loss 0.373115 Top1 81.543555 Top5 97.632512 -2022-12-06 10:26:32,097 - ==> Top1: 81.544 Top5: 97.633 Loss: 0.373 - -2022-12-06 10:26:32,098 - ==> Confusion: -[[ 894 2 2 0 9 5 1 4 4 42 0 3 3 2 5 2 4 5 0 1 8] - [ 1 908 1 1 14 27 1 24 1 1 4 6 5 0 4 0 7 1 11 2 8] - [ 7 4 989 14 2 4 7 17 1 0 6 7 4 2 3 3 1 3 11 2 16] - [ 1 5 27 900 2 5 0 2 2 0 12 2 10 2 18 0 0 7 18 1 6] - [ 9 8 3 0 941 11 0 4 0 3 1 5 2 3 9 3 9 2 0 1 6] - [ 1 58 1 0 9 906 0 31 0 0 3 17 5 16 0 1 2 2 3 8 6] - [ 0 8 48 5 1 3 976 21 0 0 6 2 1 0 0 13 1 2 6 19 6] - [ 0 18 8 4 2 26 0 917 0 1 3 9 1 1 0 1 0 1 45 11 6] - [ 12 7 0 2 0 2 0 3 942 45 13 5 1 10 12 0 1 3 3 0 3] - [ 109 3 2 0 10 6 0 3 30 796 4 1 1 17 3 0 1 3 1 1 10] - [ 0 8 9 7 1 4 0 9 3 1 929 2 3 10 7 0 2 2 19 0 3] - [ 0 5 4 0 0 15 0 5 1 0 0 954 27 4 2 6 3 8 1 14 2] - [ 0 4 0 2 2 3 0 3 0 0 0 56 856 0 2 7 1 20 1 4 8] - [ 1 8 1 2 3 12 0 2 12 16 17 12 6 903 0 4 4 1 1 2 16] - [ 11 3 1 14 7 3 0 1 17 1 7 2 6 2 1023 0 3 7 10 0 12] - [ 1 2 3 3 5 4 1 0 0 0 1 8 10 1 1 964 14 11 2 5 7] - [ 4 8 2 2 9 3 1 1 0 0 0 1 6 1 0 9 1007 2 3 7 6] - [ 0 3 2 5 0 2 1 0 1 1 1 17 44 2 0 14 2 931 2 1 7] - [ 1 8 3 9 0 3 0 33 3 0 5 5 3 1 13 0 1 1 910 5 4] - [ 0 9 1 1 0 9 5 11 0 0 1 13 6 2 1 7 6 3 5 993 7] - [ 158 366 273 101 224 211 27 191 90 57 207 172 425 276 207 152 205 86 294 313 9191]] - -2022-12-06 10:26:32,671 - ==> Best [Top1: 81.544 Top5: 97.633 Sparsity:0.00 Params: 5376 on epoch: 9] -2022-12-06 10:26:32,671 - Saving checkpoint to: logs/2022.12.06-093055/checkpoint.pth.tar -2022-12-06 10:26:32,691 - - -2022-12-06 10:26:32,692 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:26:33,772 - Epoch: [10][ 10/ 1200] Overall Loss 1.141923 Objective Loss 1.141923 LR 0.001000 Time 0.107996 -2022-12-06 10:26:33,966 - Epoch: [10][ 20/ 1200] Overall Loss 1.053313 Objective Loss 1.053313 LR 0.001000 Time 0.063683 -2022-12-06 10:26:34,158 - Epoch: [10][ 30/ 1200] Overall Loss 0.993764 Objective Loss 0.993764 LR 0.001000 Time 0.048820 -2022-12-06 10:26:34,350 - Epoch: [10][ 40/ 1200] Overall Loss 0.948645 Objective Loss 0.948645 LR 0.001000 Time 0.041395 -2022-12-06 10:26:34,542 - Epoch: [10][ 50/ 1200] Overall Loss 0.912657 Objective Loss 0.912657 LR 0.001000 Time 0.036946 -2022-12-06 10:26:34,732 - Epoch: [10][ 60/ 1200] Overall Loss 0.881950 Objective Loss 0.881950 LR 0.001000 Time 0.033960 -2022-12-06 10:26:34,924 - Epoch: [10][ 70/ 1200] Overall Loss 0.857007 Objective Loss 0.857007 LR 0.001000 Time 0.031837 -2022-12-06 10:26:35,115 - Epoch: [10][ 80/ 1200] Overall Loss 0.833704 Objective Loss 0.833704 LR 0.001000 Time 0.030239 -2022-12-06 10:26:35,306 - Epoch: [10][ 90/ 1200] Overall Loss 0.814229 Objective Loss 0.814229 LR 0.001000 Time 0.028998 -2022-12-06 10:26:35,497 - Epoch: [10][ 100/ 1200] Overall Loss 0.795302 Objective Loss 0.795302 LR 0.001000 Time 0.028002 -2022-12-06 10:26:35,689 - Epoch: [10][ 110/ 1200] Overall Loss 0.775028 Objective Loss 0.775028 LR 0.001000 Time 0.027194 -2022-12-06 10:26:35,880 - Epoch: [10][ 120/ 1200] Overall Loss 0.762147 Objective Loss 0.762147 LR 0.001000 Time 0.026518 -2022-12-06 10:26:36,071 - Epoch: [10][ 130/ 1200] Overall Loss 0.750461 Objective Loss 0.750461 LR 0.001000 Time 0.025944 -2022-12-06 10:26:36,261 - Epoch: [10][ 140/ 1200] Overall Loss 0.740606 Objective Loss 0.740606 LR 0.001000 Time 0.025445 -2022-12-06 10:26:36,452 - Epoch: [10][ 150/ 1200] Overall Loss 0.730540 Objective Loss 0.730540 LR 0.001000 Time 0.025016 -2022-12-06 10:26:36,644 - Epoch: [10][ 160/ 1200] Overall Loss 0.719931 Objective Loss 0.719931 LR 0.001000 Time 0.024646 -2022-12-06 10:26:36,834 - Epoch: [10][ 170/ 1200] Overall Loss 0.710939 Objective Loss 0.710939 LR 0.001000 Time 0.024315 -2022-12-06 10:26:37,025 - Epoch: [10][ 180/ 1200] Overall Loss 0.701865 Objective Loss 0.701865 LR 0.001000 Time 0.024023 -2022-12-06 10:26:37,216 - Epoch: [10][ 190/ 1200] Overall Loss 0.693174 Objective Loss 0.693174 LR 0.001000 Time 0.023761 -2022-12-06 10:26:37,408 - Epoch: [10][ 200/ 1200] Overall Loss 0.687282 Objective Loss 0.687282 LR 0.001000 Time 0.023527 -2022-12-06 10:26:37,599 - Epoch: [10][ 210/ 1200] Overall Loss 0.680413 Objective Loss 0.680413 LR 0.001000 Time 0.023315 -2022-12-06 10:26:37,789 - Epoch: [10][ 220/ 1200] Overall Loss 0.674383 Objective Loss 0.674383 LR 0.001000 Time 0.023118 -2022-12-06 10:26:37,980 - Epoch: [10][ 230/ 1200] Overall Loss 0.667741 Objective Loss 0.667741 LR 0.001000 Time 0.022941 -2022-12-06 10:26:38,176 - Epoch: [10][ 240/ 1200] Overall Loss 0.661140 Objective Loss 0.661140 LR 0.001000 Time 0.022799 -2022-12-06 10:26:38,378 - Epoch: [10][ 250/ 1200] Overall Loss 0.655787 Objective Loss 0.655787 LR 0.001000 Time 0.022690 -2022-12-06 10:26:38,582 - Epoch: [10][ 260/ 1200] Overall Loss 0.650062 Objective Loss 0.650062 LR 0.001000 Time 0.022602 -2022-12-06 10:26:38,784 - Epoch: [10][ 270/ 1200] Overall Loss 0.645993 Objective Loss 0.645993 LR 0.001000 Time 0.022509 -2022-12-06 10:26:38,988 - Epoch: [10][ 280/ 1200] Overall Loss 0.642223 Objective Loss 0.642223 LR 0.001000 Time 0.022433 -2022-12-06 10:26:39,189 - Epoch: [10][ 290/ 1200] Overall Loss 0.639140 Objective Loss 0.639140 LR 0.001000 Time 0.022351 -2022-12-06 10:26:39,395 - Epoch: [10][ 300/ 1200] Overall Loss 0.636022 Objective Loss 0.636022 LR 0.001000 Time 0.022291 -2022-12-06 10:26:39,597 - Epoch: [10][ 310/ 1200] Overall Loss 0.631776 Objective Loss 0.631776 LR 0.001000 Time 0.022219 -2022-12-06 10:26:39,802 - Epoch: [10][ 320/ 1200] Overall Loss 0.628054 Objective Loss 0.628054 LR 0.001000 Time 0.022165 -2022-12-06 10:26:40,003 - Epoch: [10][ 330/ 1200] Overall Loss 0.624266 Objective Loss 0.624266 LR 0.001000 Time 0.022101 -2022-12-06 10:26:40,207 - Epoch: [10][ 340/ 1200] Overall Loss 0.620808 Objective Loss 0.620808 LR 0.001000 Time 0.022050 -2022-12-06 10:26:40,409 - Epoch: [10][ 350/ 1200] Overall Loss 0.617151 Objective Loss 0.617151 LR 0.001000 Time 0.021994 -2022-12-06 10:26:40,613 - Epoch: [10][ 360/ 1200] Overall Loss 0.614032 Objective Loss 0.614032 LR 0.001000 Time 0.021949 -2022-12-06 10:26:40,815 - Epoch: [10][ 370/ 1200] Overall Loss 0.611182 Objective Loss 0.611182 LR 0.001000 Time 0.021898 -2022-12-06 10:26:41,020 - Epoch: [10][ 380/ 1200] Overall Loss 0.607674 Objective Loss 0.607674 LR 0.001000 Time 0.021861 -2022-12-06 10:26:41,221 - Epoch: [10][ 390/ 1200] Overall Loss 0.604576 Objective Loss 0.604576 LR 0.001000 Time 0.021813 -2022-12-06 10:26:41,425 - Epoch: [10][ 400/ 1200] Overall Loss 0.602270 Objective Loss 0.602270 LR 0.001000 Time 0.021779 -2022-12-06 10:26:41,626 - Epoch: [10][ 410/ 1200] Overall Loss 0.599630 Objective Loss 0.599630 LR 0.001000 Time 0.021734 -2022-12-06 10:26:41,831 - Epoch: [10][ 420/ 1200] Overall Loss 0.597827 Objective Loss 0.597827 LR 0.001000 Time 0.021704 -2022-12-06 10:26:42,032 - Epoch: [10][ 430/ 1200] Overall Loss 0.595526 Objective Loss 0.595526 LR 0.001000 Time 0.021665 -2022-12-06 10:26:42,236 - Epoch: [10][ 440/ 1200] Overall Loss 0.592890 Objective Loss 0.592890 LR 0.001000 Time 0.021636 -2022-12-06 10:26:42,438 - Epoch: [10][ 450/ 1200] Overall Loss 0.590779 Objective Loss 0.590779 LR 0.001000 Time 0.021602 -2022-12-06 10:26:42,643 - Epoch: [10][ 460/ 1200] Overall Loss 0.587900 Objective Loss 0.587900 LR 0.001000 Time 0.021576 -2022-12-06 10:26:42,844 - Epoch: [10][ 470/ 1200] Overall Loss 0.586463 Objective Loss 0.586463 LR 0.001000 Time 0.021544 -2022-12-06 10:26:43,049 - Epoch: [10][ 480/ 1200] Overall Loss 0.584100 Objective Loss 0.584100 LR 0.001000 Time 0.021521 -2022-12-06 10:26:43,250 - Epoch: [10][ 490/ 1200] Overall Loss 0.581137 Objective Loss 0.581137 LR 0.001000 Time 0.021491 -2022-12-06 10:26:43,455 - Epoch: [10][ 500/ 1200] Overall Loss 0.578950 Objective Loss 0.578950 LR 0.001000 Time 0.021470 -2022-12-06 10:26:43,657 - Epoch: [10][ 510/ 1200] Overall Loss 0.577041 Objective Loss 0.577041 LR 0.001000 Time 0.021443 -2022-12-06 10:26:43,861 - Epoch: [10][ 520/ 1200] Overall Loss 0.575484 Objective Loss 0.575484 LR 0.001000 Time 0.021423 -2022-12-06 10:26:44,062 - Epoch: [10][ 530/ 1200] Overall Loss 0.573448 Objective Loss 0.573448 LR 0.001000 Time 0.021397 -2022-12-06 10:26:44,266 - Epoch: [10][ 540/ 1200] Overall Loss 0.571955 Objective Loss 0.571955 LR 0.001000 Time 0.021378 -2022-12-06 10:26:44,467 - Epoch: [10][ 550/ 1200] Overall Loss 0.569924 Objective Loss 0.569924 LR 0.001000 Time 0.021353 -2022-12-06 10:26:44,672 - Epoch: [10][ 560/ 1200] Overall Loss 0.568341 Objective Loss 0.568341 LR 0.001000 Time 0.021337 -2022-12-06 10:26:44,873 - Epoch: [10][ 570/ 1200] Overall Loss 0.566455 Objective Loss 0.566455 LR 0.001000 Time 0.021313 -2022-12-06 10:26:45,078 - Epoch: [10][ 580/ 1200] Overall Loss 0.565111 Objective Loss 0.565111 LR 0.001000 Time 0.021298 -2022-12-06 10:26:45,279 - Epoch: [10][ 590/ 1200] Overall Loss 0.563716 Objective Loss 0.563716 LR 0.001000 Time 0.021277 -2022-12-06 10:26:45,484 - Epoch: [10][ 600/ 1200] Overall Loss 0.561900 Objective Loss 0.561900 LR 0.001000 Time 0.021263 -2022-12-06 10:26:45,685 - Epoch: [10][ 610/ 1200] Overall Loss 0.560409 Objective Loss 0.560409 LR 0.001000 Time 0.021244 -2022-12-06 10:26:45,892 - Epoch: [10][ 620/ 1200] Overall Loss 0.558516 Objective Loss 0.558516 LR 0.001000 Time 0.021234 -2022-12-06 10:26:46,093 - Epoch: [10][ 630/ 1200] Overall Loss 0.557320 Objective Loss 0.557320 LR 0.001000 Time 0.021214 -2022-12-06 10:26:46,297 - Epoch: [10][ 640/ 1200] Overall Loss 0.556617 Objective Loss 0.556617 LR 0.001000 Time 0.021202 -2022-12-06 10:26:46,498 - Epoch: [10][ 650/ 1200] Overall Loss 0.554842 Objective Loss 0.554842 LR 0.001000 Time 0.021184 -2022-12-06 10:26:46,704 - Epoch: [10][ 660/ 1200] Overall Loss 0.553748 Objective Loss 0.553748 LR 0.001000 Time 0.021173 -2022-12-06 10:26:46,904 - Epoch: [10][ 670/ 1200] Overall Loss 0.552838 Objective Loss 0.552838 LR 0.001000 Time 0.021156 -2022-12-06 10:26:47,109 - Epoch: [10][ 680/ 1200] Overall Loss 0.551563 Objective Loss 0.551563 LR 0.001000 Time 0.021145 -2022-12-06 10:26:47,310 - Epoch: [10][ 690/ 1200] Overall Loss 0.550130 Objective Loss 0.550130 LR 0.001000 Time 0.021129 -2022-12-06 10:26:47,515 - Epoch: [10][ 700/ 1200] Overall Loss 0.548500 Objective Loss 0.548500 LR 0.001000 Time 0.021119 -2022-12-06 10:26:47,717 - Epoch: [10][ 710/ 1200] Overall Loss 0.547166 Objective Loss 0.547166 LR 0.001000 Time 0.021105 -2022-12-06 10:26:47,923 - Epoch: [10][ 720/ 1200] Overall Loss 0.545746 Objective Loss 0.545746 LR 0.001000 Time 0.021097 -2022-12-06 10:26:48,124 - Epoch: [10][ 730/ 1200] Overall Loss 0.544285 Objective Loss 0.544285 LR 0.001000 Time 0.021083 -2022-12-06 10:26:48,329 - Epoch: [10][ 740/ 1200] Overall Loss 0.543116 Objective Loss 0.543116 LR 0.001000 Time 0.021075 -2022-12-06 10:26:48,531 - Epoch: [10][ 750/ 1200] Overall Loss 0.542410 Objective Loss 0.542410 LR 0.001000 Time 0.021061 -2022-12-06 10:26:48,736 - Epoch: [10][ 760/ 1200] Overall Loss 0.541022 Objective Loss 0.541022 LR 0.001000 Time 0.021054 -2022-12-06 10:26:48,938 - Epoch: [10][ 770/ 1200] Overall Loss 0.539824 Objective Loss 0.539824 LR 0.001000 Time 0.021041 -2022-12-06 10:26:49,143 - Epoch: [10][ 780/ 1200] Overall Loss 0.538459 Objective Loss 0.538459 LR 0.001000 Time 0.021033 -2022-12-06 10:26:49,344 - Epoch: [10][ 790/ 1200] Overall Loss 0.537655 Objective Loss 0.537655 LR 0.001000 Time 0.021021 -2022-12-06 10:26:49,549 - Epoch: [10][ 800/ 1200] Overall Loss 0.536103 Objective Loss 0.536103 LR 0.001000 Time 0.021013 -2022-12-06 10:26:49,750 - Epoch: [10][ 810/ 1200] Overall Loss 0.535171 Objective Loss 0.535171 LR 0.001000 Time 0.021002 -2022-12-06 10:26:49,954 - Epoch: [10][ 820/ 1200] Overall Loss 0.534022 Objective Loss 0.534022 LR 0.001000 Time 0.020994 -2022-12-06 10:26:50,156 - Epoch: [10][ 830/ 1200] Overall Loss 0.533396 Objective Loss 0.533396 LR 0.001000 Time 0.020983 -2022-12-06 10:26:50,361 - Epoch: [10][ 840/ 1200] Overall Loss 0.532365 Objective Loss 0.532365 LR 0.001000 Time 0.020977 -2022-12-06 10:26:50,562 - Epoch: [10][ 850/ 1200] Overall Loss 0.531243 Objective Loss 0.531243 LR 0.001000 Time 0.020967 -2022-12-06 10:26:50,768 - Epoch: [10][ 860/ 1200] Overall Loss 0.530601 Objective Loss 0.530601 LR 0.001000 Time 0.020961 -2022-12-06 10:26:50,969 - Epoch: [10][ 870/ 1200] Overall Loss 0.529547 Objective Loss 0.529547 LR 0.001000 Time 0.020950 -2022-12-06 10:26:51,173 - Epoch: [10][ 880/ 1200] Overall Loss 0.528385 Objective Loss 0.528385 LR 0.001000 Time 0.020944 -2022-12-06 10:26:51,374 - Epoch: [10][ 890/ 1200] Overall Loss 0.527067 Objective Loss 0.527067 LR 0.001000 Time 0.020934 -2022-12-06 10:26:51,580 - Epoch: [10][ 900/ 1200] Overall Loss 0.526068 Objective Loss 0.526068 LR 0.001000 Time 0.020929 -2022-12-06 10:26:51,782 - Epoch: [10][ 910/ 1200] Overall Loss 0.525173 Objective Loss 0.525173 LR 0.001000 Time 0.020920 -2022-12-06 10:26:51,986 - Epoch: [10][ 920/ 1200] Overall Loss 0.523736 Objective Loss 0.523736 LR 0.001000 Time 0.020914 -2022-12-06 10:26:52,187 - Epoch: [10][ 930/ 1200] Overall Loss 0.522630 Objective Loss 0.522630 LR 0.001000 Time 0.020905 -2022-12-06 10:26:52,392 - Epoch: [10][ 940/ 1200] Overall Loss 0.521626 Objective Loss 0.521626 LR 0.001000 Time 0.020900 -2022-12-06 10:26:52,593 - Epoch: [10][ 950/ 1200] Overall Loss 0.520525 Objective Loss 0.520525 LR 0.001000 Time 0.020891 -2022-12-06 10:26:52,798 - Epoch: [10][ 960/ 1200] Overall Loss 0.519566 Objective Loss 0.519566 LR 0.001000 Time 0.020887 -2022-12-06 10:26:53,000 - Epoch: [10][ 970/ 1200] Overall Loss 0.518682 Objective Loss 0.518682 LR 0.001000 Time 0.020878 -2022-12-06 10:26:53,205 - Epoch: [10][ 980/ 1200] Overall Loss 0.518106 Objective Loss 0.518106 LR 0.001000 Time 0.020874 -2022-12-06 10:26:53,406 - Epoch: [10][ 990/ 1200] Overall Loss 0.517604 Objective Loss 0.517604 LR 0.001000 Time 0.020865 -2022-12-06 10:26:53,611 - Epoch: [10][ 1000/ 1200] Overall Loss 0.516803 Objective Loss 0.516803 LR 0.001000 Time 0.020861 -2022-12-06 10:26:53,812 - Epoch: [10][ 1010/ 1200] Overall Loss 0.515668 Objective Loss 0.515668 LR 0.001000 Time 0.020853 -2022-12-06 10:26:54,017 - Epoch: [10][ 1020/ 1200] Overall Loss 0.514778 Objective Loss 0.514778 LR 0.001000 Time 0.020850 -2022-12-06 10:26:54,218 - Epoch: [10][ 1030/ 1200] Overall Loss 0.513773 Objective Loss 0.513773 LR 0.001000 Time 0.020841 -2022-12-06 10:26:54,423 - Epoch: [10][ 1040/ 1200] Overall Loss 0.513006 Objective Loss 0.513006 LR 0.001000 Time 0.020838 -2022-12-06 10:26:54,624 - Epoch: [10][ 1050/ 1200] Overall Loss 0.512239 Objective Loss 0.512239 LR 0.001000 Time 0.020830 -2022-12-06 10:26:54,829 - Epoch: [10][ 1060/ 1200] Overall Loss 0.511757 Objective Loss 0.511757 LR 0.001000 Time 0.020826 -2022-12-06 10:26:55,030 - Epoch: [10][ 1070/ 1200] Overall Loss 0.510979 Objective Loss 0.510979 LR 0.001000 Time 0.020819 -2022-12-06 10:26:55,234 - Epoch: [10][ 1080/ 1200] Overall Loss 0.510492 Objective Loss 0.510492 LR 0.001000 Time 0.020815 -2022-12-06 10:26:55,436 - Epoch: [10][ 1090/ 1200] Overall Loss 0.509618 Objective Loss 0.509618 LR 0.001000 Time 0.020808 -2022-12-06 10:26:55,640 - Epoch: [10][ 1100/ 1200] Overall Loss 0.508941 Objective Loss 0.508941 LR 0.001000 Time 0.020804 -2022-12-06 10:26:55,841 - Epoch: [10][ 1110/ 1200] Overall Loss 0.508270 Objective Loss 0.508270 LR 0.001000 Time 0.020797 -2022-12-06 10:26:56,046 - Epoch: [10][ 1120/ 1200] Overall Loss 0.507364 Objective Loss 0.507364 LR 0.001000 Time 0.020794 -2022-12-06 10:26:56,247 - Epoch: [10][ 1130/ 1200] Overall Loss 0.506595 Objective Loss 0.506595 LR 0.001000 Time 0.020787 -2022-12-06 10:26:56,452 - Epoch: [10][ 1140/ 1200] Overall Loss 0.505924 Objective Loss 0.505924 LR 0.001000 Time 0.020784 -2022-12-06 10:26:56,654 - Epoch: [10][ 1150/ 1200] Overall Loss 0.505654 Objective Loss 0.505654 LR 0.001000 Time 0.020778 -2022-12-06 10:26:56,858 - Epoch: [10][ 1160/ 1200] Overall Loss 0.505145 Objective Loss 0.505145 LR 0.001000 Time 0.020775 -2022-12-06 10:26:57,059 - Epoch: [10][ 1170/ 1200] Overall Loss 0.504637 Objective Loss 0.504637 LR 0.001000 Time 0.020769 -2022-12-06 10:26:57,264 - Epoch: [10][ 1180/ 1200] Overall Loss 0.503891 Objective Loss 0.503891 LR 0.001000 Time 0.020766 -2022-12-06 10:26:57,466 - Epoch: [10][ 1190/ 1200] Overall Loss 0.503069 Objective Loss 0.503069 LR 0.001000 Time 0.020761 -2022-12-06 10:26:57,704 - Epoch: [10][ 1200/ 1200] Overall Loss 0.502279 Objective Loss 0.502279 Top1 79.288703 Top5 96.861925 LR 0.001000 Time 0.020786 -2022-12-06 10:26:57,793 - --- validate (epoch=10)----------- -2022-12-06 10:26:57,793 - 34129 samples (256 per mini-batch) -2022-12-06 10:26:58,250 - Epoch: [10][ 10/ 134] Loss 0.422028 Top1 77.734375 Top5 97.070312 -2022-12-06 10:26:58,379 - Epoch: [10][ 20/ 134] Loss 0.429431 Top1 77.988281 Top5 97.031250 -2022-12-06 10:26:58,510 - Epoch: [10][ 30/ 134] Loss 0.418608 Top1 78.294271 Top5 97.031250 -2022-12-06 10:26:58,640 - Epoch: [10][ 40/ 134] Loss 0.417271 Top1 78.300781 Top5 96.904297 -2022-12-06 10:26:58,770 - Epoch: [10][ 50/ 134] Loss 0.415767 Top1 78.398438 Top5 96.843750 -2022-12-06 10:26:58,900 - Epoch: [10][ 60/ 134] Loss 0.411534 Top1 78.561198 Top5 96.946615 -2022-12-06 10:26:59,030 - Epoch: [10][ 70/ 134] Loss 0.410536 Top1 78.593750 Top5 96.947545 -2022-12-06 10:26:59,162 - Epoch: [10][ 80/ 134] Loss 0.412753 Top1 78.535156 Top5 96.938477 -2022-12-06 10:26:59,293 - Epoch: [10][ 90/ 134] Loss 0.410081 Top1 78.489583 Top5 96.944444 -2022-12-06 10:26:59,424 - Epoch: [10][ 100/ 134] Loss 0.411377 Top1 78.542969 Top5 96.972656 -2022-12-06 10:26:59,556 - Epoch: [10][ 110/ 134] Loss 0.413854 Top1 78.494318 Top5 96.938920 -2022-12-06 10:26:59,687 - Epoch: [10][ 120/ 134] Loss 0.408454 Top1 78.610026 Top5 96.949870 -2022-12-06 10:26:59,820 - Epoch: [10][ 130/ 134] Loss 0.408377 Top1 78.587740 Top5 96.947115 -2022-12-06 10:26:59,860 - Epoch: [10][ 134/ 134] Loss 0.408741 Top1 78.584195 Top5 96.949808 -2022-12-06 10:26:59,947 - ==> Top1: 78.584 Top5: 96.950 Loss: 0.409 - -2022-12-06 10:26:59,948 - ==> Confusion: -[[ 891 4 2 1 11 2 0 0 5 54 0 5 2 2 5 3 0 2 0 0 7] - [ 1 897 2 2 18 28 1 20 2 2 3 6 6 2 6 2 7 1 17 0 4] - [ 9 2 955 16 3 5 48 14 0 1 6 5 4 2 4 9 0 4 7 2 7] - [ 2 3 29 902 0 4 0 0 1 0 10 3 4 0 33 1 3 7 14 0 4] - [ 14 5 3 2 944 5 0 1 0 10 0 4 0 1 9 12 6 1 0 0 3] - [ 3 43 0 5 12 915 7 22 4 2 0 13 10 14 1 2 0 1 3 5 7] - [ 1 6 15 3 0 1 1053 6 1 0 1 4 2 0 0 9 2 1 2 9 2] - [ 2 19 14 2 2 36 3 884 0 0 2 8 4 0 0 4 0 3 56 12 3] - [ 7 9 0 1 0 4 0 0 944 53 4 1 4 11 18 1 3 1 1 0 2] - [ 114 2 3 0 10 0 0 1 29 806 1 4 0 10 9 4 0 1 0 0 7] - [ 0 7 9 20 2 1 5 3 16 3 897 2 3 11 13 1 0 2 20 0 4] - [ 3 5 4 0 1 14 4 4 3 0 1 886 76 6 0 9 5 8 3 10 9] - [ 1 4 1 2 2 3 2 2 0 0 0 17 898 0 1 14 3 17 0 2 0] - [ 5 1 1 1 4 15 0 4 16 18 4 8 9 901 7 7 4 3 0 2 13] - [ 20 3 0 8 9 3 0 1 22 0 0 1 5 2 1038 2 2 2 3 0 9] - [ 1 1 2 3 0 4 3 0 2 1 0 6 7 1 0 983 7 11 1 4 6] - [ 2 5 1 1 7 2 1 0 0 0 0 6 5 0 3 15 1007 5 0 5 7] - [ 2 1 1 6 0 1 2 0 0 0 1 5 52 1 2 19 3 935 1 1 3] - [ 4 3 8 13 1 2 1 25 2 0 4 3 6 1 24 1 3 1 901 2 3] - [ 0 5 2 0 1 8 19 8 1 0 0 28 7 3 0 6 10 4 0 973 5] - [ 271 384 260 132 267 214 133 168 96 120 171 143 608 352 320 311 265 96 340 372 8203]] - -2022-12-06 10:27:00,524 - ==> Best [Top1: 78.584 Top5: 96.950 Sparsity:0.00 Params: 5376 on epoch: 10] -2022-12-06 10:27:00,524 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:27:00,530 - - -2022-12-06 10:27:00,530 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:27:01,574 - Epoch: [11][ 10/ 1200] Overall Loss 0.431666 Objective Loss 0.431666 LR 0.001000 Time 0.104245 -2022-12-06 10:27:01,790 - Epoch: [11][ 20/ 1200] Overall Loss 0.428608 Objective Loss 0.428608 LR 0.001000 Time 0.062931 -2022-12-06 10:27:01,994 - Epoch: [11][ 30/ 1200] Overall Loss 0.401321 Objective Loss 0.401321 LR 0.001000 Time 0.048723 -2022-12-06 10:27:02,200 - Epoch: [11][ 40/ 1200] Overall Loss 0.409946 Objective Loss 0.409946 LR 0.001000 Time 0.041680 -2022-12-06 10:27:02,402 - Epoch: [11][ 50/ 1200] Overall Loss 0.413751 Objective Loss 0.413751 LR 0.001000 Time 0.037366 -2022-12-06 10:27:02,608 - Epoch: [11][ 60/ 1200] Overall Loss 0.409291 Objective Loss 0.409291 LR 0.001000 Time 0.034564 -2022-12-06 10:27:02,810 - Epoch: [11][ 70/ 1200] Overall Loss 0.409756 Objective Loss 0.409756 LR 0.001000 Time 0.032500 -2022-12-06 10:27:03,015 - Epoch: [11][ 80/ 1200] Overall Loss 0.409843 Objective Loss 0.409843 LR 0.001000 Time 0.030994 -2022-12-06 10:27:03,216 - Epoch: [11][ 90/ 1200] Overall Loss 0.403624 Objective Loss 0.403624 LR 0.001000 Time 0.029781 -2022-12-06 10:27:03,422 - Epoch: [11][ 100/ 1200] Overall Loss 0.404084 Objective Loss 0.404084 LR 0.001000 Time 0.028853 -2022-12-06 10:27:03,624 - Epoch: [11][ 110/ 1200] Overall Loss 0.404963 Objective Loss 0.404963 LR 0.001000 Time 0.028065 -2022-12-06 10:27:03,830 - Epoch: [11][ 120/ 1200] Overall Loss 0.404515 Objective Loss 0.404515 LR 0.001000 Time 0.027434 -2022-12-06 10:27:04,032 - Epoch: [11][ 130/ 1200] Overall Loss 0.406447 Objective Loss 0.406447 LR 0.001000 Time 0.026873 -2022-12-06 10:27:04,238 - Epoch: [11][ 140/ 1200] Overall Loss 0.406276 Objective Loss 0.406276 LR 0.001000 Time 0.026420 -2022-12-06 10:27:04,440 - Epoch: [11][ 150/ 1200] Overall Loss 0.406438 Objective Loss 0.406438 LR 0.001000 Time 0.026004 -2022-12-06 10:27:04,646 - Epoch: [11][ 160/ 1200] Overall Loss 0.406523 Objective Loss 0.406523 LR 0.001000 Time 0.025659 -2022-12-06 10:27:04,848 - Epoch: [11][ 170/ 1200] Overall Loss 0.405781 Objective Loss 0.405781 LR 0.001000 Time 0.025337 -2022-12-06 10:27:05,055 - Epoch: [11][ 180/ 1200] Overall Loss 0.406362 Objective Loss 0.406362 LR 0.001000 Time 0.025075 -2022-12-06 10:27:05,257 - Epoch: [11][ 190/ 1200] Overall Loss 0.404722 Objective Loss 0.404722 LR 0.001000 Time 0.024818 -2022-12-06 10:27:05,462 - Epoch: [11][ 200/ 1200] Overall Loss 0.407787 Objective Loss 0.407787 LR 0.001000 Time 0.024599 -2022-12-06 10:27:05,665 - Epoch: [11][ 210/ 1200] Overall Loss 0.406887 Objective Loss 0.406887 LR 0.001000 Time 0.024390 -2022-12-06 10:27:05,871 - Epoch: [11][ 220/ 1200] Overall Loss 0.405822 Objective Loss 0.405822 LR 0.001000 Time 0.024214 -2022-12-06 10:27:06,072 - Epoch: [11][ 230/ 1200] Overall Loss 0.406309 Objective Loss 0.406309 LR 0.001000 Time 0.024034 -2022-12-06 10:27:06,279 - Epoch: [11][ 240/ 1200] Overall Loss 0.406221 Objective Loss 0.406221 LR 0.001000 Time 0.023892 -2022-12-06 10:27:06,481 - Epoch: [11][ 250/ 1200] Overall Loss 0.406382 Objective Loss 0.406382 LR 0.001000 Time 0.023740 -2022-12-06 10:27:06,686 - Epoch: [11][ 260/ 1200] Overall Loss 0.405432 Objective Loss 0.405432 LR 0.001000 Time 0.023615 -2022-12-06 10:27:06,887 - Epoch: [11][ 270/ 1200] Overall Loss 0.405282 Objective Loss 0.405282 LR 0.001000 Time 0.023484 -2022-12-06 10:27:07,094 - Epoch: [11][ 280/ 1200] Overall Loss 0.404303 Objective Loss 0.404303 LR 0.001000 Time 0.023380 -2022-12-06 10:27:07,296 - Epoch: [11][ 290/ 1200] Overall Loss 0.403984 Objective Loss 0.403984 LR 0.001000 Time 0.023268 -2022-12-06 10:27:07,501 - Epoch: [11][ 300/ 1200] Overall Loss 0.403542 Objective Loss 0.403542 LR 0.001000 Time 0.023175 -2022-12-06 10:27:07,704 - Epoch: [11][ 310/ 1200] Overall Loss 0.403300 Objective Loss 0.403300 LR 0.001000 Time 0.023078 -2022-12-06 10:27:07,909 - Epoch: [11][ 320/ 1200] Overall Loss 0.403556 Objective Loss 0.403556 LR 0.001000 Time 0.022998 -2022-12-06 10:27:08,110 - Epoch: [11][ 330/ 1200] Overall Loss 0.402953 Objective Loss 0.402953 LR 0.001000 Time 0.022908 -2022-12-06 10:27:08,316 - Epoch: [11][ 340/ 1200] Overall Loss 0.404025 Objective Loss 0.404025 LR 0.001000 Time 0.022838 -2022-12-06 10:27:08,518 - Epoch: [11][ 350/ 1200] Overall Loss 0.404305 Objective Loss 0.404305 LR 0.001000 Time 0.022760 -2022-12-06 10:27:08,723 - Epoch: [11][ 360/ 1200] Overall Loss 0.403964 Objective Loss 0.403964 LR 0.001000 Time 0.022697 -2022-12-06 10:27:08,925 - Epoch: [11][ 370/ 1200] Overall Loss 0.403925 Objective Loss 0.403925 LR 0.001000 Time 0.022628 -2022-12-06 10:27:09,131 - Epoch: [11][ 380/ 1200] Overall Loss 0.404482 Objective Loss 0.404482 LR 0.001000 Time 0.022573 -2022-12-06 10:27:09,334 - Epoch: [11][ 390/ 1200] Overall Loss 0.404622 Objective Loss 0.404622 LR 0.001000 Time 0.022513 -2022-12-06 10:27:09,539 - Epoch: [11][ 400/ 1200] Overall Loss 0.404747 Objective Loss 0.404747 LR 0.001000 Time 0.022460 -2022-12-06 10:27:09,741 - Epoch: [11][ 410/ 1200] Overall Loss 0.404347 Objective Loss 0.404347 LR 0.001000 Time 0.022403 -2022-12-06 10:27:09,946 - Epoch: [11][ 420/ 1200] Overall Loss 0.404211 Objective Loss 0.404211 LR 0.001000 Time 0.022357 -2022-12-06 10:27:10,148 - Epoch: [11][ 430/ 1200] Overall Loss 0.404620 Objective Loss 0.404620 LR 0.001000 Time 0.022306 -2022-12-06 10:27:10,354 - Epoch: [11][ 440/ 1200] Overall Loss 0.403932 Objective Loss 0.403932 LR 0.001000 Time 0.022264 -2022-12-06 10:27:10,555 - Epoch: [11][ 450/ 1200] Overall Loss 0.403907 Objective Loss 0.403907 LR 0.001000 Time 0.022216 -2022-12-06 10:27:10,761 - Epoch: [11][ 460/ 1200] Overall Loss 0.403329 Objective Loss 0.403329 LR 0.001000 Time 0.022180 -2022-12-06 10:27:10,963 - Epoch: [11][ 470/ 1200] Overall Loss 0.404481 Objective Loss 0.404481 LR 0.001000 Time 0.022137 -2022-12-06 10:27:11,170 - Epoch: [11][ 480/ 1200] Overall Loss 0.404197 Objective Loss 0.404197 LR 0.001000 Time 0.022104 -2022-12-06 10:27:11,371 - Epoch: [11][ 490/ 1200] Overall Loss 0.403583 Objective Loss 0.403583 LR 0.001000 Time 0.022063 -2022-12-06 10:27:11,576 - Epoch: [11][ 500/ 1200] Overall Loss 0.403700 Objective Loss 0.403700 LR 0.001000 Time 0.022031 -2022-12-06 10:27:11,778 - Epoch: [11][ 510/ 1200] Overall Loss 0.403730 Objective Loss 0.403730 LR 0.001000 Time 0.021993 -2022-12-06 10:27:11,984 - Epoch: [11][ 520/ 1200] Overall Loss 0.403438 Objective Loss 0.403438 LR 0.001000 Time 0.021964 -2022-12-06 10:27:12,185 - Epoch: [11][ 530/ 1200] Overall Loss 0.402887 Objective Loss 0.402887 LR 0.001000 Time 0.021929 -2022-12-06 10:27:12,391 - Epoch: [11][ 540/ 1200] Overall Loss 0.403523 Objective Loss 0.403523 LR 0.001000 Time 0.021903 -2022-12-06 10:27:12,591 - Epoch: [11][ 550/ 1200] Overall Loss 0.403282 Objective Loss 0.403282 LR 0.001000 Time 0.021868 -2022-12-06 10:27:12,797 - Epoch: [11][ 560/ 1200] Overall Loss 0.403159 Objective Loss 0.403159 LR 0.001000 Time 0.021843 -2022-12-06 10:27:12,998 - Epoch: [11][ 570/ 1200] Overall Loss 0.402629 Objective Loss 0.402629 LR 0.001000 Time 0.021812 -2022-12-06 10:27:13,205 - Epoch: [11][ 580/ 1200] Overall Loss 0.402274 Objective Loss 0.402274 LR 0.001000 Time 0.021791 -2022-12-06 10:27:13,407 - Epoch: [11][ 590/ 1200] Overall Loss 0.402604 Objective Loss 0.402604 LR 0.001000 Time 0.021763 -2022-12-06 10:27:13,612 - Epoch: [11][ 600/ 1200] Overall Loss 0.402293 Objective Loss 0.402293 LR 0.001000 Time 0.021742 -2022-12-06 10:27:13,814 - Epoch: [11][ 610/ 1200] Overall Loss 0.401913 Objective Loss 0.401913 LR 0.001000 Time 0.021715 -2022-12-06 10:27:14,020 - Epoch: [11][ 620/ 1200] Overall Loss 0.402073 Objective Loss 0.402073 LR 0.001000 Time 0.021696 -2022-12-06 10:27:14,220 - Epoch: [11][ 630/ 1200] Overall Loss 0.402176 Objective Loss 0.402176 LR 0.001000 Time 0.021669 -2022-12-06 10:27:14,426 - Epoch: [11][ 640/ 1200] Overall Loss 0.402242 Objective Loss 0.402242 LR 0.001000 Time 0.021651 -2022-12-06 10:27:14,628 - Epoch: [11][ 650/ 1200] Overall Loss 0.401780 Objective Loss 0.401780 LR 0.001000 Time 0.021627 -2022-12-06 10:27:14,833 - Epoch: [11][ 660/ 1200] Overall Loss 0.401703 Objective Loss 0.401703 LR 0.001000 Time 0.021609 -2022-12-06 10:27:15,035 - Epoch: [11][ 670/ 1200] Overall Loss 0.401608 Objective Loss 0.401608 LR 0.001000 Time 0.021587 -2022-12-06 10:27:15,239 - Epoch: [11][ 680/ 1200] Overall Loss 0.401351 Objective Loss 0.401351 LR 0.001000 Time 0.021569 -2022-12-06 10:27:15,441 - Epoch: [11][ 690/ 1200] Overall Loss 0.401186 Objective Loss 0.401186 LR 0.001000 Time 0.021548 -2022-12-06 10:27:15,647 - Epoch: [11][ 700/ 1200] Overall Loss 0.400844 Objective Loss 0.400844 LR 0.001000 Time 0.021534 -2022-12-06 10:27:15,849 - Epoch: [11][ 710/ 1200] Overall Loss 0.400687 Objective Loss 0.400687 LR 0.001000 Time 0.021514 -2022-12-06 10:27:16,054 - Epoch: [11][ 720/ 1200] Overall Loss 0.400474 Objective Loss 0.400474 LR 0.001000 Time 0.021500 -2022-12-06 10:27:16,256 - Epoch: [11][ 730/ 1200] Overall Loss 0.400739 Objective Loss 0.400739 LR 0.001000 Time 0.021481 -2022-12-06 10:27:16,462 - Epoch: [11][ 740/ 1200] Overall Loss 0.400474 Objective Loss 0.400474 LR 0.001000 Time 0.021468 -2022-12-06 10:27:16,664 - Epoch: [11][ 750/ 1200] Overall Loss 0.400380 Objective Loss 0.400380 LR 0.001000 Time 0.021450 -2022-12-06 10:27:16,871 - Epoch: [11][ 760/ 1200] Overall Loss 0.400722 Objective Loss 0.400722 LR 0.001000 Time 0.021439 -2022-12-06 10:27:17,074 - Epoch: [11][ 770/ 1200] Overall Loss 0.400659 Objective Loss 0.400659 LR 0.001000 Time 0.021424 -2022-12-06 10:27:17,280 - Epoch: [11][ 780/ 1200] Overall Loss 0.400722 Objective Loss 0.400722 LR 0.001000 Time 0.021413 -2022-12-06 10:27:17,482 - Epoch: [11][ 790/ 1200] Overall Loss 0.400587 Objective Loss 0.400587 LR 0.001000 Time 0.021397 -2022-12-06 10:27:17,688 - Epoch: [11][ 800/ 1200] Overall Loss 0.400684 Objective Loss 0.400684 LR 0.001000 Time 0.021386 -2022-12-06 10:27:17,890 - Epoch: [11][ 810/ 1200] Overall Loss 0.400987 Objective Loss 0.400987 LR 0.001000 Time 0.021370 -2022-12-06 10:27:18,096 - Epoch: [11][ 820/ 1200] Overall Loss 0.400915 Objective Loss 0.400915 LR 0.001000 Time 0.021360 -2022-12-06 10:27:18,298 - Epoch: [11][ 830/ 1200] Overall Loss 0.400840 Objective Loss 0.400840 LR 0.001000 Time 0.021346 -2022-12-06 10:27:18,505 - Epoch: [11][ 840/ 1200] Overall Loss 0.400768 Objective Loss 0.400768 LR 0.001000 Time 0.021337 -2022-12-06 10:27:18,707 - Epoch: [11][ 850/ 1200] Overall Loss 0.400948 Objective Loss 0.400948 LR 0.001000 Time 0.021323 -2022-12-06 10:27:18,912 - Epoch: [11][ 860/ 1200] Overall Loss 0.401115 Objective Loss 0.401115 LR 0.001000 Time 0.021313 -2022-12-06 10:27:19,114 - Epoch: [11][ 870/ 1200] Overall Loss 0.400831 Objective Loss 0.400831 LR 0.001000 Time 0.021299 -2022-12-06 10:27:19,319 - Epoch: [11][ 880/ 1200] Overall Loss 0.400432 Objective Loss 0.400432 LR 0.001000 Time 0.021290 -2022-12-06 10:27:19,522 - Epoch: [11][ 890/ 1200] Overall Loss 0.400665 Objective Loss 0.400665 LR 0.001000 Time 0.021277 -2022-12-06 10:27:19,727 - Epoch: [11][ 900/ 1200] Overall Loss 0.400670 Objective Loss 0.400670 LR 0.001000 Time 0.021268 -2022-12-06 10:27:19,929 - Epoch: [11][ 910/ 1200] Overall Loss 0.400571 Objective Loss 0.400571 LR 0.001000 Time 0.021256 -2022-12-06 10:27:20,134 - Epoch: [11][ 920/ 1200] Overall Loss 0.400248 Objective Loss 0.400248 LR 0.001000 Time 0.021247 -2022-12-06 10:27:20,336 - Epoch: [11][ 930/ 1200] Overall Loss 0.400009 Objective Loss 0.400009 LR 0.001000 Time 0.021235 -2022-12-06 10:27:20,543 - Epoch: [11][ 940/ 1200] Overall Loss 0.399992 Objective Loss 0.399992 LR 0.001000 Time 0.021229 -2022-12-06 10:27:20,745 - Epoch: [11][ 950/ 1200] Overall Loss 0.399937 Objective Loss 0.399937 LR 0.001000 Time 0.021218 -2022-12-06 10:27:20,952 - Epoch: [11][ 960/ 1200] Overall Loss 0.399888 Objective Loss 0.399888 LR 0.001000 Time 0.021211 -2022-12-06 10:27:21,154 - Epoch: [11][ 970/ 1200] Overall Loss 0.399425 Objective Loss 0.399425 LR 0.001000 Time 0.021200 -2022-12-06 10:27:21,360 - Epoch: [11][ 980/ 1200] Overall Loss 0.399020 Objective Loss 0.399020 LR 0.001000 Time 0.021194 -2022-12-06 10:27:21,563 - Epoch: [11][ 990/ 1200] Overall Loss 0.398988 Objective Loss 0.398988 LR 0.001000 Time 0.021184 -2022-12-06 10:27:21,768 - Epoch: [11][ 1000/ 1200] Overall Loss 0.398694 Objective Loss 0.398694 LR 0.001000 Time 0.021177 -2022-12-06 10:27:21,970 - Epoch: [11][ 1010/ 1200] Overall Loss 0.398509 Objective Loss 0.398509 LR 0.001000 Time 0.021167 -2022-12-06 10:27:22,176 - Epoch: [11][ 1020/ 1200] Overall Loss 0.398291 Objective Loss 0.398291 LR 0.001000 Time 0.021160 -2022-12-06 10:27:22,378 - Epoch: [11][ 1030/ 1200] Overall Loss 0.398350 Objective Loss 0.398350 LR 0.001000 Time 0.021150 -2022-12-06 10:27:22,585 - Epoch: [11][ 1040/ 1200] Overall Loss 0.398115 Objective Loss 0.398115 LR 0.001000 Time 0.021145 -2022-12-06 10:27:22,786 - Epoch: [11][ 1050/ 1200] Overall Loss 0.398197 Objective Loss 0.398197 LR 0.001000 Time 0.021135 -2022-12-06 10:27:22,992 - Epoch: [11][ 1060/ 1200] Overall Loss 0.398203 Objective Loss 0.398203 LR 0.001000 Time 0.021129 -2022-12-06 10:27:23,194 - Epoch: [11][ 1070/ 1200] Overall Loss 0.398475 Objective Loss 0.398475 LR 0.001000 Time 0.021120 -2022-12-06 10:27:23,400 - Epoch: [11][ 1080/ 1200] Overall Loss 0.398333 Objective Loss 0.398333 LR 0.001000 Time 0.021114 -2022-12-06 10:27:23,602 - Epoch: [11][ 1090/ 1200] Overall Loss 0.398538 Objective Loss 0.398538 LR 0.001000 Time 0.021106 -2022-12-06 10:27:23,808 - Epoch: [11][ 1100/ 1200] Overall Loss 0.398481 Objective Loss 0.398481 LR 0.001000 Time 0.021101 -2022-12-06 10:27:24,010 - Epoch: [11][ 1110/ 1200] Overall Loss 0.398202 Objective Loss 0.398202 LR 0.001000 Time 0.021091 -2022-12-06 10:27:24,215 - Epoch: [11][ 1120/ 1200] Overall Loss 0.397878 Objective Loss 0.397878 LR 0.001000 Time 0.021086 -2022-12-06 10:27:24,418 - Epoch: [11][ 1130/ 1200] Overall Loss 0.397910 Objective Loss 0.397910 LR 0.001000 Time 0.021078 -2022-12-06 10:27:24,623 - Epoch: [11][ 1140/ 1200] Overall Loss 0.398229 Objective Loss 0.398229 LR 0.001000 Time 0.021073 -2022-12-06 10:27:24,825 - Epoch: [11][ 1150/ 1200] Overall Loss 0.398408 Objective Loss 0.398408 LR 0.001000 Time 0.021065 -2022-12-06 10:27:25,027 - Epoch: [11][ 1160/ 1200] Overall Loss 0.398355 Objective Loss 0.398355 LR 0.001000 Time 0.021056 -2022-12-06 10:27:25,218 - Epoch: [11][ 1170/ 1200] Overall Loss 0.398313 Objective Loss 0.398313 LR 0.001000 Time 0.021039 -2022-12-06 10:27:25,409 - Epoch: [11][ 1180/ 1200] Overall Loss 0.398299 Objective Loss 0.398299 LR 0.001000 Time 0.021023 -2022-12-06 10:27:25,600 - Epoch: [11][ 1190/ 1200] Overall Loss 0.398296 Objective Loss 0.398296 LR 0.001000 Time 0.021006 -2022-12-06 10:27:25,835 - Epoch: [11][ 1200/ 1200] Overall Loss 0.398043 Objective Loss 0.398043 Top1 79.079498 Top5 97.698745 LR 0.001000 Time 0.021026 -2022-12-06 10:27:25,924 - --- validate (epoch=11)----------- -2022-12-06 10:27:25,925 - 34129 samples (256 per mini-batch) -2022-12-06 10:27:26,368 - Epoch: [11][ 10/ 134] Loss 0.639216 Top1 81.093750 Top5 97.304688 -2022-12-06 10:27:26,502 - Epoch: [11][ 20/ 134] Loss 0.646298 Top1 81.406250 Top5 97.382812 -2022-12-06 10:27:26,627 - Epoch: [11][ 30/ 134] Loss 0.653238 Top1 81.250000 Top5 97.369792 -2022-12-06 10:27:26,757 - Epoch: [11][ 40/ 134] Loss 0.660354 Top1 80.664062 Top5 97.353516 -2022-12-06 10:27:26,887 - Epoch: [11][ 50/ 134] Loss 0.660009 Top1 80.960938 Top5 97.414062 -2022-12-06 10:27:27,018 - Epoch: [11][ 60/ 134] Loss 0.660193 Top1 81.028646 Top5 97.421875 -2022-12-06 10:27:27,150 - Epoch: [11][ 70/ 134] Loss 0.659537 Top1 81.037946 Top5 97.410714 -2022-12-06 10:27:27,277 - Epoch: [11][ 80/ 134] Loss 0.658895 Top1 80.961914 Top5 97.421875 -2022-12-06 10:27:27,410 - Epoch: [11][ 90/ 134] Loss 0.656654 Top1 81.059028 Top5 97.430556 -2022-12-06 10:27:27,532 - Epoch: [11][ 100/ 134] Loss 0.660610 Top1 80.843750 Top5 97.382812 -2022-12-06 10:27:27,660 - Epoch: [11][ 110/ 134] Loss 0.659392 Top1 80.894886 Top5 97.428977 -2022-12-06 10:27:27,787 - Epoch: [11][ 120/ 134] Loss 0.658571 Top1 80.947266 Top5 97.438151 -2022-12-06 10:27:27,915 - Epoch: [11][ 130/ 134] Loss 0.659044 Top1 80.868389 Top5 97.448918 -2022-12-06 10:27:27,952 - Epoch: [11][ 134/ 134] Loss 0.658332 Top1 80.828621 Top5 97.450848 -2022-12-06 10:27:28,039 - ==> Top1: 80.829 Top5: 97.451 Loss: 0.658 - -2022-12-06 10:27:28,039 - ==> Confusion: -[[ 852 7 2 2 8 3 0 0 6 88 1 5 1 2 7 2 1 1 2 0 6] - [ 1 948 1 0 4 14 0 18 2 1 1 2 4 0 3 0 6 1 13 1 7] - [ 6 9 976 9 5 3 32 12 2 2 6 7 2 6 2 4 0 0 7 3 10] - [ 3 5 40 890 0 0 1 1 1 0 14 0 9 1 19 1 2 6 21 0 6] - [ 12 21 4 0 922 6 0 0 2 9 0 3 2 4 10 11 7 1 0 1 5] - [ 2 79 0 2 9 865 2 24 4 3 1 17 4 36 0 2 0 2 3 4 10] - [ 0 8 9 2 0 0 1057 10 3 0 1 4 2 0 0 6 3 2 2 7 2] - [ 3 29 11 1 1 25 6 908 1 1 1 4 2 1 0 1 0 3 43 8 5] - [ 8 6 0 0 1 1 0 2 955 43 7 2 4 6 17 0 1 4 2 1 4] - [ 69 1 1 0 2 0 0 2 31 869 1 1 0 9 3 1 0 2 0 0 9] - [ 1 4 11 6 2 1 0 4 12 3 923 3 2 11 7 0 1 1 22 0 5] - [ 2 4 3 0 0 6 3 5 2 0 0 950 41 13 0 1 0 8 4 5 4] - [ 1 4 2 3 1 2 0 1 0 0 0 47 877 1 0 12 2 8 1 2 5] - [ 1 5 1 1 2 8 0 1 17 25 10 8 5 920 3 3 2 1 0 0 10] - [ 13 5 2 7 6 2 0 1 22 3 1 2 4 4 1038 0 5 1 6 0 8] - [ 2 5 6 0 2 0 0 0 1 1 0 14 7 2 3 977 9 8 0 3 3] - [ 3 9 3 1 4 1 0 0 2 0 1 8 1 1 1 13 1004 2 2 6 10] - [ 3 5 1 3 0 0 1 0 1 0 1 18 37 1 2 13 2 941 2 1 4] - [ 2 5 8 4 0 1 0 26 5 0 5 2 5 1 13 1 1 0 921 1 7] - [ 0 9 6 1 1 7 7 16 1 0 1 30 5 6 0 2 9 5 5 965 4] - [ 172 523 247 84 157 140 81 158 127 122 183 180 475 435 224 184 212 86 315 299 8822]] - -2022-12-06 10:27:28,607 - ==> Best [Top1: 80.829 Top5: 97.451 Sparsity:0.00 Params: 5376 on epoch: 11] -2022-12-06 10:27:28,607 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:27:28,614 - - -2022-12-06 10:27:28,614 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:27:29,555 - Epoch: [12][ 10/ 1200] Overall Loss 0.678931 Objective Loss 0.678931 LR 0.001000 Time 0.094043 -2022-12-06 10:27:29,753 - Epoch: [12][ 20/ 1200] Overall Loss 0.652334 Objective Loss 0.652334 LR 0.001000 Time 0.056905 -2022-12-06 10:27:29,953 - Epoch: [12][ 30/ 1200] Overall Loss 0.631245 Objective Loss 0.631245 LR 0.001000 Time 0.044580 -2022-12-06 10:27:30,149 - Epoch: [12][ 40/ 1200] Overall Loss 0.614156 Objective Loss 0.614156 LR 0.001000 Time 0.038313 -2022-12-06 10:27:30,349 - Epoch: [12][ 50/ 1200] Overall Loss 0.602798 Objective Loss 0.602798 LR 0.001000 Time 0.034632 -2022-12-06 10:27:30,544 - Epoch: [12][ 60/ 1200] Overall Loss 0.594952 Objective Loss 0.594952 LR 0.001000 Time 0.032106 -2022-12-06 10:27:30,742 - Epoch: [12][ 70/ 1200] Overall Loss 0.587817 Objective Loss 0.587817 LR 0.001000 Time 0.030347 -2022-12-06 10:27:30,938 - Epoch: [12][ 80/ 1200] Overall Loss 0.581693 Objective Loss 0.581693 LR 0.001000 Time 0.028988 -2022-12-06 10:27:31,135 - Epoch: [12][ 90/ 1200] Overall Loss 0.576400 Objective Loss 0.576400 LR 0.001000 Time 0.027954 -2022-12-06 10:27:31,331 - Epoch: [12][ 100/ 1200] Overall Loss 0.571405 Objective Loss 0.571405 LR 0.001000 Time 0.027107 -2022-12-06 10:27:31,529 - Epoch: [12][ 110/ 1200] Overall Loss 0.566514 Objective Loss 0.566514 LR 0.001000 Time 0.026446 -2022-12-06 10:27:31,725 - Epoch: [12][ 120/ 1200] Overall Loss 0.562465 Objective Loss 0.562465 LR 0.001000 Time 0.025868 -2022-12-06 10:27:31,923 - Epoch: [12][ 130/ 1200] Overall Loss 0.559220 Objective Loss 0.559220 LR 0.001000 Time 0.025400 -2022-12-06 10:27:32,120 - Epoch: [12][ 140/ 1200] Overall Loss 0.555751 Objective Loss 0.555751 LR 0.001000 Time 0.024987 -2022-12-06 10:27:32,318 - Epoch: [12][ 150/ 1200] Overall Loss 0.552097 Objective Loss 0.552097 LR 0.001000 Time 0.024638 -2022-12-06 10:27:32,513 - Epoch: [12][ 160/ 1200] Overall Loss 0.548772 Objective Loss 0.548772 LR 0.001000 Time 0.024310 -2022-12-06 10:27:32,711 - Epoch: [12][ 170/ 1200] Overall Loss 0.545203 Objective Loss 0.545203 LR 0.001000 Time 0.024042 -2022-12-06 10:27:32,906 - Epoch: [12][ 180/ 1200] Overall Loss 0.541407 Objective Loss 0.541407 LR 0.001000 Time 0.023789 -2022-12-06 10:27:33,104 - Epoch: [12][ 190/ 1200] Overall Loss 0.539404 Objective Loss 0.539404 LR 0.001000 Time 0.023578 -2022-12-06 10:27:33,299 - Epoch: [12][ 200/ 1200] Overall Loss 0.537056 Objective Loss 0.537056 LR 0.001000 Time 0.023370 -2022-12-06 10:27:33,497 - Epoch: [12][ 210/ 1200] Overall Loss 0.533644 Objective Loss 0.533644 LR 0.001000 Time 0.023199 -2022-12-06 10:27:33,692 - Epoch: [12][ 220/ 1200] Overall Loss 0.531879 Objective Loss 0.531879 LR 0.001000 Time 0.023028 -2022-12-06 10:27:33,891 - Epoch: [12][ 230/ 1200] Overall Loss 0.528860 Objective Loss 0.528860 LR 0.001000 Time 0.022885 -2022-12-06 10:27:34,085 - Epoch: [12][ 240/ 1200] Overall Loss 0.527327 Objective Loss 0.527327 LR 0.001000 Time 0.022741 -2022-12-06 10:27:34,284 - Epoch: [12][ 250/ 1200] Overall Loss 0.524863 Objective Loss 0.524863 LR 0.001000 Time 0.022623 -2022-12-06 10:27:34,477 - Epoch: [12][ 260/ 1200] Overall Loss 0.523454 Objective Loss 0.523454 LR 0.001000 Time 0.022493 -2022-12-06 10:27:34,667 - Epoch: [12][ 270/ 1200] Overall Loss 0.521085 Objective Loss 0.521085 LR 0.001000 Time 0.022363 -2022-12-06 10:27:34,857 - Epoch: [12][ 280/ 1200] Overall Loss 0.518575 Objective Loss 0.518575 LR 0.001000 Time 0.022240 -2022-12-06 10:27:35,047 - Epoch: [12][ 290/ 1200] Overall Loss 0.516637 Objective Loss 0.516637 LR 0.001000 Time 0.022126 -2022-12-06 10:27:35,237 - Epoch: [12][ 300/ 1200] Overall Loss 0.515137 Objective Loss 0.515137 LR 0.001000 Time 0.022020 -2022-12-06 10:27:35,426 - Epoch: [12][ 310/ 1200] Overall Loss 0.513835 Objective Loss 0.513835 LR 0.001000 Time 0.021920 -2022-12-06 10:27:35,617 - Epoch: [12][ 320/ 1200] Overall Loss 0.511974 Objective Loss 0.511974 LR 0.001000 Time 0.021828 -2022-12-06 10:27:35,807 - Epoch: [12][ 330/ 1200] Overall Loss 0.509834 Objective Loss 0.509834 LR 0.001000 Time 0.021742 -2022-12-06 10:27:35,997 - Epoch: [12][ 340/ 1200] Overall Loss 0.507229 Objective Loss 0.507229 LR 0.001000 Time 0.021660 -2022-12-06 10:27:36,187 - Epoch: [12][ 350/ 1200] Overall Loss 0.505590 Objective Loss 0.505590 LR 0.001000 Time 0.021583 -2022-12-06 10:27:36,377 - Epoch: [12][ 360/ 1200] Overall Loss 0.504484 Objective Loss 0.504484 LR 0.001000 Time 0.021509 -2022-12-06 10:27:36,568 - Epoch: [12][ 370/ 1200] Overall Loss 0.502592 Objective Loss 0.502592 LR 0.001000 Time 0.021441 -2022-12-06 10:27:36,758 - Epoch: [12][ 380/ 1200] Overall Loss 0.501403 Objective Loss 0.501403 LR 0.001000 Time 0.021376 -2022-12-06 10:27:36,948 - Epoch: [12][ 390/ 1200] Overall Loss 0.500252 Objective Loss 0.500252 LR 0.001000 Time 0.021314 -2022-12-06 10:27:37,139 - Epoch: [12][ 400/ 1200] Overall Loss 0.498790 Objective Loss 0.498790 LR 0.001000 Time 0.021257 -2022-12-06 10:27:37,330 - Epoch: [12][ 410/ 1200] Overall Loss 0.496610 Objective Loss 0.496610 LR 0.001000 Time 0.021202 -2022-12-06 10:27:37,520 - Epoch: [12][ 420/ 1200] Overall Loss 0.495569 Objective Loss 0.495569 LR 0.001000 Time 0.021150 -2022-12-06 10:27:37,711 - Epoch: [12][ 430/ 1200] Overall Loss 0.494776 Objective Loss 0.494776 LR 0.001000 Time 0.021100 -2022-12-06 10:27:37,901 - Epoch: [12][ 440/ 1200] Overall Loss 0.494030 Objective Loss 0.494030 LR 0.001000 Time 0.021052 -2022-12-06 10:27:38,092 - Epoch: [12][ 450/ 1200] Overall Loss 0.492873 Objective Loss 0.492873 LR 0.001000 Time 0.021007 -2022-12-06 10:27:38,282 - Epoch: [12][ 460/ 1200] Overall Loss 0.491452 Objective Loss 0.491452 LR 0.001000 Time 0.020963 -2022-12-06 10:27:38,472 - Epoch: [12][ 470/ 1200] Overall Loss 0.489828 Objective Loss 0.489828 LR 0.001000 Time 0.020920 -2022-12-06 10:27:38,663 - Epoch: [12][ 480/ 1200] Overall Loss 0.488055 Objective Loss 0.488055 LR 0.001000 Time 0.020880 -2022-12-06 10:27:38,853 - Epoch: [12][ 490/ 1200] Overall Loss 0.486978 Objective Loss 0.486978 LR 0.001000 Time 0.020841 -2022-12-06 10:27:39,044 - Epoch: [12][ 500/ 1200] Overall Loss 0.485455 Objective Loss 0.485455 LR 0.001000 Time 0.020805 -2022-12-06 10:27:39,234 - Epoch: [12][ 510/ 1200] Overall Loss 0.484739 Objective Loss 0.484739 LR 0.001000 Time 0.020769 -2022-12-06 10:27:39,425 - Epoch: [12][ 520/ 1200] Overall Loss 0.483291 Objective Loss 0.483291 LR 0.001000 Time 0.020735 -2022-12-06 10:27:39,615 - Epoch: [12][ 530/ 1200] Overall Loss 0.482677 Objective Loss 0.482677 LR 0.001000 Time 0.020702 -2022-12-06 10:27:39,806 - Epoch: [12][ 540/ 1200] Overall Loss 0.481723 Objective Loss 0.481723 LR 0.001000 Time 0.020671 -2022-12-06 10:27:39,996 - Epoch: [12][ 550/ 1200] Overall Loss 0.480532 Objective Loss 0.480532 LR 0.001000 Time 0.020639 -2022-12-06 10:27:40,186 - Epoch: [12][ 560/ 1200] Overall Loss 0.479944 Objective Loss 0.479944 LR 0.001000 Time 0.020609 -2022-12-06 10:27:40,376 - Epoch: [12][ 570/ 1200] Overall Loss 0.479064 Objective Loss 0.479064 LR 0.001000 Time 0.020580 -2022-12-06 10:27:40,566 - Epoch: [12][ 580/ 1200] Overall Loss 0.478035 Objective Loss 0.478035 LR 0.001000 Time 0.020551 -2022-12-06 10:27:40,756 - Epoch: [12][ 590/ 1200] Overall Loss 0.477507 Objective Loss 0.477507 LR 0.001000 Time 0.020524 -2022-12-06 10:27:40,946 - Epoch: [12][ 600/ 1200] Overall Loss 0.476901 Objective Loss 0.476901 LR 0.001000 Time 0.020498 -2022-12-06 10:27:41,136 - Epoch: [12][ 610/ 1200] Overall Loss 0.475888 Objective Loss 0.475888 LR 0.001000 Time 0.020474 -2022-12-06 10:27:41,326 - Epoch: [12][ 620/ 1200] Overall Loss 0.475238 Objective Loss 0.475238 LR 0.001000 Time 0.020448 -2022-12-06 10:27:41,516 - Epoch: [12][ 630/ 1200] Overall Loss 0.474072 Objective Loss 0.474072 LR 0.001000 Time 0.020424 -2022-12-06 10:27:41,706 - Epoch: [12][ 640/ 1200] Overall Loss 0.473648 Objective Loss 0.473648 LR 0.001000 Time 0.020401 -2022-12-06 10:27:41,897 - Epoch: [12][ 650/ 1200] Overall Loss 0.472764 Objective Loss 0.472764 LR 0.001000 Time 0.020380 -2022-12-06 10:27:42,086 - Epoch: [12][ 660/ 1200] Overall Loss 0.471777 Objective Loss 0.471777 LR 0.001000 Time 0.020358 -2022-12-06 10:27:42,276 - Epoch: [12][ 670/ 1200] Overall Loss 0.470925 Objective Loss 0.470925 LR 0.001000 Time 0.020336 -2022-12-06 10:27:42,465 - Epoch: [12][ 680/ 1200] Overall Loss 0.470434 Objective Loss 0.470434 LR 0.001000 Time 0.020315 -2022-12-06 10:27:42,658 - Epoch: [12][ 690/ 1200] Overall Loss 0.469733 Objective Loss 0.469733 LR 0.001000 Time 0.020299 -2022-12-06 10:27:42,848 - Epoch: [12][ 700/ 1200] Overall Loss 0.469075 Objective Loss 0.469075 LR 0.001000 Time 0.020280 -2022-12-06 10:27:43,041 - Epoch: [12][ 710/ 1200] Overall Loss 0.468239 Objective Loss 0.468239 LR 0.001000 Time 0.020264 -2022-12-06 10:27:43,232 - Epoch: [12][ 720/ 1200] Overall Loss 0.467255 Objective Loss 0.467255 LR 0.001000 Time 0.020248 -2022-12-06 10:27:43,425 - Epoch: [12][ 730/ 1200] Overall Loss 0.466561 Objective Loss 0.466561 LR 0.001000 Time 0.020234 -2022-12-06 10:27:43,616 - Epoch: [12][ 740/ 1200] Overall Loss 0.466133 Objective Loss 0.466133 LR 0.001000 Time 0.020218 -2022-12-06 10:27:43,808 - Epoch: [12][ 750/ 1200] Overall Loss 0.465741 Objective Loss 0.465741 LR 0.001000 Time 0.020204 -2022-12-06 10:27:44,000 - Epoch: [12][ 760/ 1200] Overall Loss 0.464909 Objective Loss 0.464909 LR 0.001000 Time 0.020190 -2022-12-06 10:27:44,193 - Epoch: [12][ 770/ 1200] Overall Loss 0.464474 Objective Loss 0.464474 LR 0.001000 Time 0.020177 -2022-12-06 10:27:44,385 - Epoch: [12][ 780/ 1200] Overall Loss 0.464053 Objective Loss 0.464053 LR 0.001000 Time 0.020164 -2022-12-06 10:27:44,578 - Epoch: [12][ 790/ 1200] Overall Loss 0.463037 Objective Loss 0.463037 LR 0.001000 Time 0.020152 -2022-12-06 10:27:44,770 - Epoch: [12][ 800/ 1200] Overall Loss 0.462253 Objective Loss 0.462253 LR 0.001000 Time 0.020140 -2022-12-06 10:27:44,962 - Epoch: [12][ 810/ 1200] Overall Loss 0.461732 Objective Loss 0.461732 LR 0.001000 Time 0.020128 -2022-12-06 10:27:45,155 - Epoch: [12][ 820/ 1200] Overall Loss 0.461018 Objective Loss 0.461018 LR 0.001000 Time 0.020117 -2022-12-06 10:27:45,348 - Epoch: [12][ 830/ 1200] Overall Loss 0.460485 Objective Loss 0.460485 LR 0.001000 Time 0.020106 -2022-12-06 10:27:45,539 - Epoch: [12][ 840/ 1200] Overall Loss 0.459910 Objective Loss 0.459910 LR 0.001000 Time 0.020094 -2022-12-06 10:27:45,731 - Epoch: [12][ 850/ 1200] Overall Loss 0.459408 Objective Loss 0.459408 LR 0.001000 Time 0.020083 -2022-12-06 10:27:45,923 - Epoch: [12][ 860/ 1200] Overall Loss 0.458915 Objective Loss 0.458915 LR 0.001000 Time 0.020072 -2022-12-06 10:27:46,115 - Epoch: [12][ 870/ 1200] Overall Loss 0.458109 Objective Loss 0.458109 LR 0.001000 Time 0.020062 -2022-12-06 10:27:46,307 - Epoch: [12][ 880/ 1200] Overall Loss 0.457307 Objective Loss 0.457307 LR 0.001000 Time 0.020051 -2022-12-06 10:27:46,500 - Epoch: [12][ 890/ 1200] Overall Loss 0.456817 Objective Loss 0.456817 LR 0.001000 Time 0.020042 -2022-12-06 10:27:46,692 - Epoch: [12][ 900/ 1200] Overall Loss 0.455943 Objective Loss 0.455943 LR 0.001000 Time 0.020032 -2022-12-06 10:27:46,884 - Epoch: [12][ 910/ 1200] Overall Loss 0.455853 Objective Loss 0.455853 LR 0.001000 Time 0.020022 -2022-12-06 10:27:47,076 - Epoch: [12][ 920/ 1200] Overall Loss 0.455289 Objective Loss 0.455289 LR 0.001000 Time 0.020012 -2022-12-06 10:27:47,269 - Epoch: [12][ 930/ 1200] Overall Loss 0.454754 Objective Loss 0.454754 LR 0.001000 Time 0.020005 -2022-12-06 10:27:47,461 - Epoch: [12][ 940/ 1200] Overall Loss 0.454351 Objective Loss 0.454351 LR 0.001000 Time 0.019995 -2022-12-06 10:27:47,654 - Epoch: [12][ 950/ 1200] Overall Loss 0.454102 Objective Loss 0.454102 LR 0.001000 Time 0.019987 -2022-12-06 10:27:47,845 - Epoch: [12][ 960/ 1200] Overall Loss 0.453746 Objective Loss 0.453746 LR 0.001000 Time 0.019978 -2022-12-06 10:27:48,038 - Epoch: [12][ 970/ 1200] Overall Loss 0.453057 Objective Loss 0.453057 LR 0.001000 Time 0.019970 -2022-12-06 10:27:48,230 - Epoch: [12][ 980/ 1200] Overall Loss 0.452885 Objective Loss 0.452885 LR 0.001000 Time 0.019962 -2022-12-06 10:27:48,423 - Epoch: [12][ 990/ 1200] Overall Loss 0.452629 Objective Loss 0.452629 LR 0.001000 Time 0.019954 -2022-12-06 10:27:48,614 - Epoch: [12][ 1000/ 1200] Overall Loss 0.452131 Objective Loss 0.452131 LR 0.001000 Time 0.019946 -2022-12-06 10:27:48,806 - Epoch: [12][ 1010/ 1200] Overall Loss 0.451537 Objective Loss 0.451537 LR 0.001000 Time 0.019938 -2022-12-06 10:27:48,999 - Epoch: [12][ 1020/ 1200] Overall Loss 0.450896 Objective Loss 0.450896 LR 0.001000 Time 0.019930 -2022-12-06 10:27:49,191 - Epoch: [12][ 1030/ 1200] Overall Loss 0.450230 Objective Loss 0.450230 LR 0.001000 Time 0.019923 -2022-12-06 10:27:49,384 - Epoch: [12][ 1040/ 1200] Overall Loss 0.449731 Objective Loss 0.449731 LR 0.001000 Time 0.019916 -2022-12-06 10:27:49,576 - Epoch: [12][ 1050/ 1200] Overall Loss 0.449261 Objective Loss 0.449261 LR 0.001000 Time 0.019909 -2022-12-06 10:27:49,767 - Epoch: [12][ 1060/ 1200] Overall Loss 0.448783 Objective Loss 0.448783 LR 0.001000 Time 0.019901 -2022-12-06 10:27:49,959 - Epoch: [12][ 1070/ 1200] Overall Loss 0.448449 Objective Loss 0.448449 LR 0.001000 Time 0.019894 -2022-12-06 10:27:50,151 - Epoch: [12][ 1080/ 1200] Overall Loss 0.448187 Objective Loss 0.448187 LR 0.001000 Time 0.019887 -2022-12-06 10:27:50,343 - Epoch: [12][ 1090/ 1200] Overall Loss 0.447868 Objective Loss 0.447868 LR 0.001000 Time 0.019880 -2022-12-06 10:27:50,534 - Epoch: [12][ 1100/ 1200] Overall Loss 0.447667 Objective Loss 0.447667 LR 0.001000 Time 0.019873 -2022-12-06 10:27:50,727 - Epoch: [12][ 1110/ 1200] Overall Loss 0.447479 Objective Loss 0.447479 LR 0.001000 Time 0.019866 -2022-12-06 10:27:50,919 - Epoch: [12][ 1120/ 1200] Overall Loss 0.447190 Objective Loss 0.447190 LR 0.001000 Time 0.019860 -2022-12-06 10:27:51,111 - Epoch: [12][ 1130/ 1200] Overall Loss 0.446666 Objective Loss 0.446666 LR 0.001000 Time 0.019854 -2022-12-06 10:27:51,303 - Epoch: [12][ 1140/ 1200] Overall Loss 0.446343 Objective Loss 0.446343 LR 0.001000 Time 0.019848 -2022-12-06 10:27:51,495 - Epoch: [12][ 1150/ 1200] Overall Loss 0.445992 Objective Loss 0.445992 LR 0.001000 Time 0.019842 -2022-12-06 10:27:51,687 - Epoch: [12][ 1160/ 1200] Overall Loss 0.445846 Objective Loss 0.445846 LR 0.001000 Time 0.019836 -2022-12-06 10:27:51,879 - Epoch: [12][ 1170/ 1200] Overall Loss 0.445790 Objective Loss 0.445790 LR 0.001000 Time 0.019830 -2022-12-06 10:27:52,071 - Epoch: [12][ 1180/ 1200] Overall Loss 0.445515 Objective Loss 0.445515 LR 0.001000 Time 0.019824 -2022-12-06 10:27:52,263 - Epoch: [12][ 1190/ 1200] Overall Loss 0.445301 Objective Loss 0.445301 LR 0.001000 Time 0.019819 -2022-12-06 10:27:52,487 - Epoch: [12][ 1200/ 1200] Overall Loss 0.445301 Objective Loss 0.445301 Top1 79.288703 Top5 96.443515 LR 0.001000 Time 0.019839 -2022-12-06 10:27:52,575 - --- validate (epoch=12)----------- -2022-12-06 10:27:52,575 - 34129 samples (256 per mini-batch) -2022-12-06 10:27:53,015 - Epoch: [12][ 10/ 134] Loss 0.399628 Top1 79.179688 Top5 96.953125 -2022-12-06 10:27:53,151 - Epoch: [12][ 20/ 134] Loss 0.396413 Top1 78.984375 Top5 97.128906 -2022-12-06 10:27:53,284 - Epoch: [12][ 30/ 134] Loss 0.392413 Top1 79.257812 Top5 97.122396 -2022-12-06 10:27:53,418 - Epoch: [12][ 40/ 134] Loss 0.393617 Top1 79.248047 Top5 97.080078 -2022-12-06 10:27:53,550 - Epoch: [12][ 50/ 134] Loss 0.391457 Top1 79.164062 Top5 97.031250 -2022-12-06 10:27:53,682 - Epoch: [12][ 60/ 134] Loss 0.390834 Top1 79.101562 Top5 97.024740 -2022-12-06 10:27:53,814 - Epoch: [12][ 70/ 134] Loss 0.388925 Top1 79.363839 Top5 97.020089 -2022-12-06 10:27:53,947 - Epoch: [12][ 80/ 134] Loss 0.392303 Top1 79.521484 Top5 96.977539 -2022-12-06 10:27:54,079 - Epoch: [12][ 90/ 134] Loss 0.388781 Top1 79.500868 Top5 97.022569 -2022-12-06 10:27:54,213 - Epoch: [12][ 100/ 134] Loss 0.387788 Top1 79.546875 Top5 96.949219 -2022-12-06 10:27:54,345 - Epoch: [12][ 110/ 134] Loss 0.387996 Top1 79.520597 Top5 96.960227 -2022-12-06 10:27:54,478 - Epoch: [12][ 120/ 134] Loss 0.386978 Top1 79.550781 Top5 96.972656 -2022-12-06 10:27:54,613 - Epoch: [12][ 130/ 134] Loss 0.387882 Top1 79.561298 Top5 96.953125 -2022-12-06 10:27:54,652 - Epoch: [12][ 134/ 134] Loss 0.387954 Top1 79.504234 Top5 96.964458 -2022-12-06 10:27:54,739 - ==> Top1: 79.504 Top5: 96.964 Loss: 0.388 - -2022-12-06 10:27:54,740 - ==> Confusion: -[[ 887 0 4 1 13 3 0 5 4 54 0 9 1 1 4 1 1 1 1 1 5] - [ 0 893 1 1 12 40 2 30 2 4 0 6 3 1 3 4 11 1 10 1 2] - [ 7 3 976 23 2 8 22 20 2 2 3 6 1 5 3 2 1 3 2 5 7] - [ 1 2 29 906 1 5 1 3 0 1 7 3 4 0 24 0 3 6 18 0 6] - [ 12 5 3 1 948 10 0 5 0 3 1 6 3 0 8 5 5 2 1 1 1] - [ 2 24 4 2 6 949 4 35 2 3 0 14 2 5 2 1 0 0 1 8 5] - [ 0 4 19 3 0 2 1045 6 1 1 3 4 4 0 0 6 3 5 0 12 0] - [ 3 4 12 2 1 22 3 939 0 2 1 8 2 1 1 3 1 0 33 13 3] - [ 9 3 0 1 0 2 1 1 943 58 4 3 0 7 16 2 1 2 8 1 2] - [ 84 2 1 0 11 5 0 1 22 856 0 1 2 8 1 0 0 0 0 0 7] - [ 0 4 8 20 1 1 2 7 10 4 904 5 2 14 8 0 3 0 20 2 4] - [ 4 3 3 0 0 19 0 6 1 1 0 975 16 2 1 2 3 8 0 6 1] - [ 2 0 0 1 2 7 0 1 0 0 0 76 855 0 0 8 1 10 0 3 3] - [ 2 1 0 1 4 27 0 1 14 20 6 11 2 909 5 3 2 1 0 3 11] - [ 11 6 0 20 11 3 0 2 15 3 0 1 3 4 1027 1 1 2 4 1 15] - [ 5 2 4 4 5 7 2 0 1 1 0 18 6 1 0 949 13 13 1 5 6] - [ 1 5 2 2 13 6 1 1 0 0 0 6 2 0 0 11 1006 2 0 5 9] - [ 4 1 3 3 0 3 0 1 4 1 1 19 29 1 3 6 2 952 1 0 2] - [ 2 7 5 15 0 3 0 42 2 1 5 7 4 0 8 0 1 0 901 1 4] - [ 1 1 6 1 1 6 6 15 0 0 0 40 5 7 0 4 6 3 1 974 3] - [ 240 373 274 165 232 376 82 298 90 129 152 216 504 361 203 132 292 95 277 403 8332]] - -2022-12-06 10:27:55,321 - ==> Best [Top1: 80.829 Top5: 97.451 Sparsity:0.00 Params: 5376 on epoch: 11] -2022-12-06 10:27:55,321 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:27:55,327 - - -2022-12-06 10:27:55,328 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:27:56,356 - Epoch: [13][ 10/ 1200] Overall Loss 0.376489 Objective Loss 0.376489 LR 0.001000 Time 0.102796 -2022-12-06 10:27:56,561 - Epoch: [13][ 20/ 1200] Overall Loss 0.403784 Objective Loss 0.403784 LR 0.001000 Time 0.061589 -2022-12-06 10:27:56,753 - Epoch: [13][ 30/ 1200] Overall Loss 0.409368 Objective Loss 0.409368 LR 0.001000 Time 0.047443 -2022-12-06 10:27:56,944 - Epoch: [13][ 40/ 1200] Overall Loss 0.405892 Objective Loss 0.405892 LR 0.001000 Time 0.040345 -2022-12-06 10:27:57,135 - Epoch: [13][ 50/ 1200] Overall Loss 0.401838 Objective Loss 0.401838 LR 0.001000 Time 0.036088 -2022-12-06 10:27:57,325 - Epoch: [13][ 60/ 1200] Overall Loss 0.400664 Objective Loss 0.400664 LR 0.001000 Time 0.033242 -2022-12-06 10:27:57,516 - Epoch: [13][ 70/ 1200] Overall Loss 0.394609 Objective Loss 0.394609 LR 0.001000 Time 0.031206 -2022-12-06 10:27:57,707 - Epoch: [13][ 80/ 1200] Overall Loss 0.393892 Objective Loss 0.393892 LR 0.001000 Time 0.029684 -2022-12-06 10:27:57,897 - Epoch: [13][ 90/ 1200] Overall Loss 0.391685 Objective Loss 0.391685 LR 0.001000 Time 0.028495 -2022-12-06 10:27:58,088 - Epoch: [13][ 100/ 1200] Overall Loss 0.391968 Objective Loss 0.391968 LR 0.001000 Time 0.027546 -2022-12-06 10:27:58,279 - Epoch: [13][ 110/ 1200] Overall Loss 0.392279 Objective Loss 0.392279 LR 0.001000 Time 0.026772 -2022-12-06 10:27:58,470 - Epoch: [13][ 120/ 1200] Overall Loss 0.389314 Objective Loss 0.389314 LR 0.001000 Time 0.026132 -2022-12-06 10:27:58,660 - Epoch: [13][ 130/ 1200] Overall Loss 0.388161 Objective Loss 0.388161 LR 0.001000 Time 0.025582 -2022-12-06 10:27:58,851 - Epoch: [13][ 140/ 1200] Overall Loss 0.388378 Objective Loss 0.388378 LR 0.001000 Time 0.025111 -2022-12-06 10:27:59,041 - Epoch: [13][ 150/ 1200] Overall Loss 0.386798 Objective Loss 0.386798 LR 0.001000 Time 0.024702 -2022-12-06 10:27:59,232 - Epoch: [13][ 160/ 1200] Overall Loss 0.385644 Objective Loss 0.385644 LR 0.001000 Time 0.024349 -2022-12-06 10:27:59,423 - Epoch: [13][ 170/ 1200] Overall Loss 0.386509 Objective Loss 0.386509 LR 0.001000 Time 0.024038 -2022-12-06 10:27:59,615 - Epoch: [13][ 180/ 1200] Overall Loss 0.385658 Objective Loss 0.385658 LR 0.001000 Time 0.023762 -2022-12-06 10:27:59,805 - Epoch: [13][ 190/ 1200] Overall Loss 0.385642 Objective Loss 0.385642 LR 0.001000 Time 0.023512 -2022-12-06 10:27:59,995 - Epoch: [13][ 200/ 1200] Overall Loss 0.386762 Objective Loss 0.386762 LR 0.001000 Time 0.023284 -2022-12-06 10:28:00,186 - Epoch: [13][ 210/ 1200] Overall Loss 0.385682 Objective Loss 0.385682 LR 0.001000 Time 0.023080 -2022-12-06 10:28:00,377 - Epoch: [13][ 220/ 1200] Overall Loss 0.384945 Objective Loss 0.384945 LR 0.001000 Time 0.022896 -2022-12-06 10:28:00,568 - Epoch: [13][ 230/ 1200] Overall Loss 0.382894 Objective Loss 0.382894 LR 0.001000 Time 0.022727 -2022-12-06 10:28:00,759 - Epoch: [13][ 240/ 1200] Overall Loss 0.383516 Objective Loss 0.383516 LR 0.001000 Time 0.022576 -2022-12-06 10:28:00,949 - Epoch: [13][ 250/ 1200] Overall Loss 0.384939 Objective Loss 0.384939 LR 0.001000 Time 0.022432 -2022-12-06 10:28:01,140 - Epoch: [13][ 260/ 1200] Overall Loss 0.385812 Objective Loss 0.385812 LR 0.001000 Time 0.022301 -2022-12-06 10:28:01,331 - Epoch: [13][ 270/ 1200] Overall Loss 0.386098 Objective Loss 0.386098 LR 0.001000 Time 0.022181 -2022-12-06 10:28:01,522 - Epoch: [13][ 280/ 1200] Overall Loss 0.385357 Objective Loss 0.385357 LR 0.001000 Time 0.022068 -2022-12-06 10:28:01,714 - Epoch: [13][ 290/ 1200] Overall Loss 0.384210 Objective Loss 0.384210 LR 0.001000 Time 0.021967 -2022-12-06 10:28:01,904 - Epoch: [13][ 300/ 1200] Overall Loss 0.383635 Objective Loss 0.383635 LR 0.001000 Time 0.021867 -2022-12-06 10:28:02,095 - Epoch: [13][ 310/ 1200] Overall Loss 0.382988 Objective Loss 0.382988 LR 0.001000 Time 0.021775 -2022-12-06 10:28:02,285 - Epoch: [13][ 320/ 1200] Overall Loss 0.383605 Objective Loss 0.383605 LR 0.001000 Time 0.021688 -2022-12-06 10:28:02,476 - Epoch: [13][ 330/ 1200] Overall Loss 0.383142 Objective Loss 0.383142 LR 0.001000 Time 0.021607 -2022-12-06 10:28:02,667 - Epoch: [13][ 340/ 1200] Overall Loss 0.382387 Objective Loss 0.382387 LR 0.001000 Time 0.021533 -2022-12-06 10:28:02,857 - Epoch: [13][ 350/ 1200] Overall Loss 0.381970 Objective Loss 0.381970 LR 0.001000 Time 0.021458 -2022-12-06 10:28:03,049 - Epoch: [13][ 360/ 1200] Overall Loss 0.381436 Objective Loss 0.381436 LR 0.001000 Time 0.021393 -2022-12-06 10:28:03,239 - Epoch: [13][ 370/ 1200] Overall Loss 0.381154 Objective Loss 0.381154 LR 0.001000 Time 0.021327 -2022-12-06 10:28:03,431 - Epoch: [13][ 380/ 1200] Overall Loss 0.382040 Objective Loss 0.382040 LR 0.001000 Time 0.021270 -2022-12-06 10:28:03,622 - Epoch: [13][ 390/ 1200] Overall Loss 0.382010 Objective Loss 0.382010 LR 0.001000 Time 0.021213 -2022-12-06 10:28:03,814 - Epoch: [13][ 400/ 1200] Overall Loss 0.382679 Objective Loss 0.382679 LR 0.001000 Time 0.021162 -2022-12-06 10:28:04,005 - Epoch: [13][ 410/ 1200] Overall Loss 0.382652 Objective Loss 0.382652 LR 0.001000 Time 0.021109 -2022-12-06 10:28:04,196 - Epoch: [13][ 420/ 1200] Overall Loss 0.382664 Objective Loss 0.382664 LR 0.001000 Time 0.021060 -2022-12-06 10:28:04,387 - Epoch: [13][ 430/ 1200] Overall Loss 0.382671 Objective Loss 0.382671 LR 0.001000 Time 0.021012 -2022-12-06 10:28:04,578 - Epoch: [13][ 440/ 1200] Overall Loss 0.382651 Objective Loss 0.382651 LR 0.001000 Time 0.020968 -2022-12-06 10:28:04,768 - Epoch: [13][ 450/ 1200] Overall Loss 0.383231 Objective Loss 0.383231 LR 0.001000 Time 0.020924 -2022-12-06 10:28:04,959 - Epoch: [13][ 460/ 1200] Overall Loss 0.383248 Objective Loss 0.383248 LR 0.001000 Time 0.020882 -2022-12-06 10:28:05,149 - Epoch: [13][ 470/ 1200] Overall Loss 0.383345 Objective Loss 0.383345 LR 0.001000 Time 0.020843 -2022-12-06 10:28:05,341 - Epoch: [13][ 480/ 1200] Overall Loss 0.382913 Objective Loss 0.382913 LR 0.001000 Time 0.020807 -2022-12-06 10:28:05,532 - Epoch: [13][ 490/ 1200] Overall Loss 0.383475 Objective Loss 0.383475 LR 0.001000 Time 0.020770 -2022-12-06 10:28:05,722 - Epoch: [13][ 500/ 1200] Overall Loss 0.383671 Objective Loss 0.383671 LR 0.001000 Time 0.020735 -2022-12-06 10:28:05,913 - Epoch: [13][ 510/ 1200] Overall Loss 0.383403 Objective Loss 0.383403 LR 0.001000 Time 0.020701 -2022-12-06 10:28:06,104 - Epoch: [13][ 520/ 1200] Overall Loss 0.383521 Objective Loss 0.383521 LR 0.001000 Time 0.020669 -2022-12-06 10:28:06,296 - Epoch: [13][ 530/ 1200] Overall Loss 0.383467 Objective Loss 0.383467 LR 0.001000 Time 0.020640 -2022-12-06 10:28:06,487 - Epoch: [13][ 540/ 1200] Overall Loss 0.382640 Objective Loss 0.382640 LR 0.001000 Time 0.020610 -2022-12-06 10:28:06,677 - Epoch: [13][ 550/ 1200] Overall Loss 0.382787 Objective Loss 0.382787 LR 0.001000 Time 0.020581 -2022-12-06 10:28:06,868 - Epoch: [13][ 560/ 1200] Overall Loss 0.383009 Objective Loss 0.383009 LR 0.001000 Time 0.020553 -2022-12-06 10:28:07,059 - Epoch: [13][ 570/ 1200] Overall Loss 0.383261 Objective Loss 0.383261 LR 0.001000 Time 0.020527 -2022-12-06 10:28:07,251 - Epoch: [13][ 580/ 1200] Overall Loss 0.383406 Objective Loss 0.383406 LR 0.001000 Time 0.020502 -2022-12-06 10:28:07,441 - Epoch: [13][ 590/ 1200] Overall Loss 0.383893 Objective Loss 0.383893 LR 0.001000 Time 0.020477 -2022-12-06 10:28:07,633 - Epoch: [13][ 600/ 1200] Overall Loss 0.384750 Objective Loss 0.384750 LR 0.001000 Time 0.020454 -2022-12-06 10:28:07,823 - Epoch: [13][ 610/ 1200] Overall Loss 0.384680 Objective Loss 0.384680 LR 0.001000 Time 0.020430 -2022-12-06 10:28:08,014 - Epoch: [13][ 620/ 1200] Overall Loss 0.384430 Objective Loss 0.384430 LR 0.001000 Time 0.020408 -2022-12-06 10:28:08,205 - Epoch: [13][ 630/ 1200] Overall Loss 0.383933 Objective Loss 0.383933 LR 0.001000 Time 0.020386 -2022-12-06 10:28:08,397 - Epoch: [13][ 640/ 1200] Overall Loss 0.383978 Objective Loss 0.383978 LR 0.001000 Time 0.020366 -2022-12-06 10:28:08,588 - Epoch: [13][ 650/ 1200] Overall Loss 0.383997 Objective Loss 0.383997 LR 0.001000 Time 0.020345 -2022-12-06 10:28:08,779 - Epoch: [13][ 660/ 1200] Overall Loss 0.384486 Objective Loss 0.384486 LR 0.001000 Time 0.020326 -2022-12-06 10:28:08,970 - Epoch: [13][ 670/ 1200] Overall Loss 0.384685 Objective Loss 0.384685 LR 0.001000 Time 0.020307 -2022-12-06 10:28:09,161 - Epoch: [13][ 680/ 1200] Overall Loss 0.384569 Objective Loss 0.384569 LR 0.001000 Time 0.020288 -2022-12-06 10:28:09,352 - Epoch: [13][ 690/ 1200] Overall Loss 0.384660 Objective Loss 0.384660 LR 0.001000 Time 0.020270 -2022-12-06 10:28:09,542 - Epoch: [13][ 700/ 1200] Overall Loss 0.385016 Objective Loss 0.385016 LR 0.001000 Time 0.020252 -2022-12-06 10:28:09,733 - Epoch: [13][ 710/ 1200] Overall Loss 0.384809 Objective Loss 0.384809 LR 0.001000 Time 0.020234 -2022-12-06 10:28:09,924 - Epoch: [13][ 720/ 1200] Overall Loss 0.384122 Objective Loss 0.384122 LR 0.001000 Time 0.020218 -2022-12-06 10:28:10,115 - Epoch: [13][ 730/ 1200] Overall Loss 0.384262 Objective Loss 0.384262 LR 0.001000 Time 0.020202 -2022-12-06 10:28:10,306 - Epoch: [13][ 740/ 1200] Overall Loss 0.384041 Objective Loss 0.384041 LR 0.001000 Time 0.020186 -2022-12-06 10:28:10,497 - Epoch: [13][ 750/ 1200] Overall Loss 0.383847 Objective Loss 0.383847 LR 0.001000 Time 0.020172 -2022-12-06 10:28:10,689 - Epoch: [13][ 760/ 1200] Overall Loss 0.383896 Objective Loss 0.383896 LR 0.001000 Time 0.020157 -2022-12-06 10:28:10,879 - Epoch: [13][ 770/ 1200] Overall Loss 0.384363 Objective Loss 0.384363 LR 0.001000 Time 0.020142 -2022-12-06 10:28:11,069 - Epoch: [13][ 780/ 1200] Overall Loss 0.383928 Objective Loss 0.383928 LR 0.001000 Time 0.020127 -2022-12-06 10:28:11,260 - Epoch: [13][ 790/ 1200] Overall Loss 0.383859 Objective Loss 0.383859 LR 0.001000 Time 0.020113 -2022-12-06 10:28:11,451 - Epoch: [13][ 800/ 1200] Overall Loss 0.383883 Objective Loss 0.383883 LR 0.001000 Time 0.020099 -2022-12-06 10:28:11,642 - Epoch: [13][ 810/ 1200] Overall Loss 0.383767 Objective Loss 0.383767 LR 0.001000 Time 0.020087 -2022-12-06 10:28:11,833 - Epoch: [13][ 820/ 1200] Overall Loss 0.383400 Objective Loss 0.383400 LR 0.001000 Time 0.020074 -2022-12-06 10:28:12,023 - Epoch: [13][ 830/ 1200] Overall Loss 0.383203 Objective Loss 0.383203 LR 0.001000 Time 0.020061 -2022-12-06 10:28:12,214 - Epoch: [13][ 840/ 1200] Overall Loss 0.383423 Objective Loss 0.383423 LR 0.001000 Time 0.020048 -2022-12-06 10:28:12,405 - Epoch: [13][ 850/ 1200] Overall Loss 0.383731 Objective Loss 0.383731 LR 0.001000 Time 0.020037 -2022-12-06 10:28:12,596 - Epoch: [13][ 860/ 1200] Overall Loss 0.383893 Objective Loss 0.383893 LR 0.001000 Time 0.020025 -2022-12-06 10:28:12,787 - Epoch: [13][ 870/ 1200] Overall Loss 0.383676 Objective Loss 0.383676 LR 0.001000 Time 0.020014 -2022-12-06 10:28:12,978 - Epoch: [13][ 880/ 1200] Overall Loss 0.383308 Objective Loss 0.383308 LR 0.001000 Time 0.020003 -2022-12-06 10:28:13,169 - Epoch: [13][ 890/ 1200] Overall Loss 0.383216 Objective Loss 0.383216 LR 0.001000 Time 0.019991 -2022-12-06 10:28:13,359 - Epoch: [13][ 900/ 1200] Overall Loss 0.383180 Objective Loss 0.383180 LR 0.001000 Time 0.019980 -2022-12-06 10:28:13,549 - Epoch: [13][ 910/ 1200] Overall Loss 0.383389 Objective Loss 0.383389 LR 0.001000 Time 0.019969 -2022-12-06 10:28:13,741 - Epoch: [13][ 920/ 1200] Overall Loss 0.383730 Objective Loss 0.383730 LR 0.001000 Time 0.019960 -2022-12-06 10:28:13,932 - Epoch: [13][ 930/ 1200] Overall Loss 0.383679 Objective Loss 0.383679 LR 0.001000 Time 0.019950 -2022-12-06 10:28:14,123 - Epoch: [13][ 940/ 1200] Overall Loss 0.384688 Objective Loss 0.384688 LR 0.001000 Time 0.019941 -2022-12-06 10:28:14,315 - Epoch: [13][ 950/ 1200] Overall Loss 0.385280 Objective Loss 0.385280 LR 0.001000 Time 0.019932 -2022-12-06 10:28:14,506 - Epoch: [13][ 960/ 1200] Overall Loss 0.385239 Objective Loss 0.385239 LR 0.001000 Time 0.019923 -2022-12-06 10:28:14,698 - Epoch: [13][ 970/ 1200] Overall Loss 0.385240 Objective Loss 0.385240 LR 0.001000 Time 0.019914 -2022-12-06 10:28:14,888 - Epoch: [13][ 980/ 1200] Overall Loss 0.385520 Objective Loss 0.385520 LR 0.001000 Time 0.019905 -2022-12-06 10:28:15,079 - Epoch: [13][ 990/ 1200] Overall Loss 0.385387 Objective Loss 0.385387 LR 0.001000 Time 0.019896 -2022-12-06 10:28:15,270 - Epoch: [13][ 1000/ 1200] Overall Loss 0.385562 Objective Loss 0.385562 LR 0.001000 Time 0.019888 -2022-12-06 10:28:15,461 - Epoch: [13][ 1010/ 1200] Overall Loss 0.385369 Objective Loss 0.385369 LR 0.001000 Time 0.019880 -2022-12-06 10:28:15,653 - Epoch: [13][ 1020/ 1200] Overall Loss 0.385226 Objective Loss 0.385226 LR 0.001000 Time 0.019872 -2022-12-06 10:28:15,844 - Epoch: [13][ 1030/ 1200] Overall Loss 0.385535 Objective Loss 0.385535 LR 0.001000 Time 0.019864 -2022-12-06 10:28:16,035 - Epoch: [13][ 1040/ 1200] Overall Loss 0.385386 Objective Loss 0.385386 LR 0.001000 Time 0.019856 -2022-12-06 10:28:16,226 - Epoch: [13][ 1050/ 1200] Overall Loss 0.385392 Objective Loss 0.385392 LR 0.001000 Time 0.019848 -2022-12-06 10:28:16,416 - Epoch: [13][ 1060/ 1200] Overall Loss 0.385260 Objective Loss 0.385260 LR 0.001000 Time 0.019840 -2022-12-06 10:28:16,607 - Epoch: [13][ 1070/ 1200] Overall Loss 0.385083 Objective Loss 0.385083 LR 0.001000 Time 0.019833 -2022-12-06 10:28:16,799 - Epoch: [13][ 1080/ 1200] Overall Loss 0.385131 Objective Loss 0.385131 LR 0.001000 Time 0.019826 -2022-12-06 10:28:16,989 - Epoch: [13][ 1090/ 1200] Overall Loss 0.385365 Objective Loss 0.385365 LR 0.001000 Time 0.019818 -2022-12-06 10:28:17,180 - Epoch: [13][ 1100/ 1200] Overall Loss 0.385402 Objective Loss 0.385402 LR 0.001000 Time 0.019812 -2022-12-06 10:28:17,372 - Epoch: [13][ 1110/ 1200] Overall Loss 0.385572 Objective Loss 0.385572 LR 0.001000 Time 0.019805 -2022-12-06 10:28:17,562 - Epoch: [13][ 1120/ 1200] Overall Loss 0.385432 Objective Loss 0.385432 LR 0.001000 Time 0.019798 -2022-12-06 10:28:17,753 - Epoch: [13][ 1130/ 1200] Overall Loss 0.385087 Objective Loss 0.385087 LR 0.001000 Time 0.019791 -2022-12-06 10:28:17,944 - Epoch: [13][ 1140/ 1200] Overall Loss 0.385023 Objective Loss 0.385023 LR 0.001000 Time 0.019784 -2022-12-06 10:28:18,136 - Epoch: [13][ 1150/ 1200] Overall Loss 0.385078 Objective Loss 0.385078 LR 0.001000 Time 0.019778 -2022-12-06 10:28:18,326 - Epoch: [13][ 1160/ 1200] Overall Loss 0.385173 Objective Loss 0.385173 LR 0.001000 Time 0.019772 -2022-12-06 10:28:18,517 - Epoch: [13][ 1170/ 1200] Overall Loss 0.385041 Objective Loss 0.385041 LR 0.001000 Time 0.019765 -2022-12-06 10:28:18,707 - Epoch: [13][ 1180/ 1200] Overall Loss 0.385024 Objective Loss 0.385024 LR 0.001000 Time 0.019759 -2022-12-06 10:28:18,898 - Epoch: [13][ 1190/ 1200] Overall Loss 0.384831 Objective Loss 0.384831 LR 0.001000 Time 0.019752 -2022-12-06 10:28:19,130 - Epoch: [13][ 1200/ 1200] Overall Loss 0.384749 Objective Loss 0.384749 Top1 83.263598 Top5 97.071130 LR 0.001000 Time 0.019781 -2022-12-06 10:28:19,219 - --- validate (epoch=13)----------- -2022-12-06 10:28:19,219 - 34129 samples (256 per mini-batch) -2022-12-06 10:28:19,668 - Epoch: [13][ 10/ 134] Loss 0.368610 Top1 80.898438 Top5 97.148438 -2022-12-06 10:28:19,807 - Epoch: [13][ 20/ 134] Loss 0.367331 Top1 80.742188 Top5 97.324219 -2022-12-06 10:28:19,940 - Epoch: [13][ 30/ 134] Loss 0.364072 Top1 80.429688 Top5 97.174479 -2022-12-06 10:28:20,072 - Epoch: [13][ 40/ 134] Loss 0.360058 Top1 80.800781 Top5 97.294922 -2022-12-06 10:28:20,204 - Epoch: [13][ 50/ 134] Loss 0.362604 Top1 80.796875 Top5 97.265625 -2022-12-06 10:28:20,336 - Epoch: [13][ 60/ 134] Loss 0.366213 Top1 80.820312 Top5 97.265625 -2022-12-06 10:28:20,468 - Epoch: [13][ 70/ 134] Loss 0.370441 Top1 80.714286 Top5 97.226562 -2022-12-06 10:28:20,598 - Epoch: [13][ 80/ 134] Loss 0.370583 Top1 80.708008 Top5 97.246094 -2022-12-06 10:28:20,728 - Epoch: [13][ 90/ 134] Loss 0.368993 Top1 80.759549 Top5 97.261285 -2022-12-06 10:28:20,861 - Epoch: [13][ 100/ 134] Loss 0.370925 Top1 80.605469 Top5 97.246094 -2022-12-06 10:28:20,992 - Epoch: [13][ 110/ 134] Loss 0.371482 Top1 80.511364 Top5 97.215909 -2022-12-06 10:28:21,125 - Epoch: [13][ 120/ 134] Loss 0.370669 Top1 80.530599 Top5 97.229818 -2022-12-06 10:28:21,260 - Epoch: [13][ 130/ 134] Loss 0.371899 Top1 80.456731 Top5 97.211538 -2022-12-06 10:28:21,299 - Epoch: [13][ 134/ 134] Loss 0.371630 Top1 80.424273 Top5 97.198863 -2022-12-06 10:28:21,388 - ==> Top1: 80.424 Top5: 97.199 Loss: 0.372 - -2022-12-06 10:28:21,389 - ==> Confusion: -[[ 879 1 3 2 13 4 0 2 12 49 0 4 4 1 8 2 2 2 1 1 6] - [ 0 861 0 5 12 62 2 24 1 2 4 5 4 2 4 3 13 1 15 1 6] - [ 7 7 973 21 5 5 23 13 1 1 6 3 4 4 3 6 1 1 3 8 8] - [ 3 2 18 924 2 1 1 0 2 0 14 3 9 2 13 0 4 1 14 0 7] - [ 12 9 6 0 931 8 2 3 0 5 0 3 3 3 6 8 14 1 0 2 4] - [ 0 13 2 4 4 957 1 19 3 1 0 19 3 14 2 2 4 0 3 10 8] - [ 0 1 14 2 1 3 1048 4 0 0 3 4 2 1 0 8 0 3 2 20 2] - [ 0 8 9 2 1 45 5 899 3 1 3 9 1 2 1 2 0 1 43 17 2] - [ 5 4 1 2 0 8 0 0 963 35 7 3 4 9 10 2 2 1 4 2 2] - [ 85 0 1 0 5 6 1 0 32 836 0 1 2 14 6 2 0 0 4 0 6] - [ 0 2 4 19 1 3 1 0 15 1 928 2 5 19 4 0 1 1 9 2 2] - [ 3 2 3 0 0 9 0 2 3 0 0 976 32 4 1 2 1 6 1 4 2] - [ 4 1 1 1 1 5 1 1 1 0 0 55 867 1 0 6 1 13 4 1 5] - [ 0 4 1 1 1 12 0 0 11 13 8 10 4 932 3 6 4 1 0 4 8] - [ 4 3 1 16 6 2 0 1 20 1 2 2 11 5 1019 0 8 2 12 1 14] - [ 4 2 4 2 1 2 2 0 0 0 0 13 15 1 0 955 19 10 1 5 7] - [ 0 3 2 3 1 2 1 1 1 0 0 6 6 0 0 12 1014 1 1 10 8] - [ 4 2 3 2 0 2 2 1 1 0 1 21 39 2 3 9 2 936 1 1 4] - [ 1 4 3 7 0 4 0 27 1 1 8 4 7 1 8 0 3 1 924 0 4] - [ 0 4 4 1 0 8 3 4 0 0 1 27 7 5 0 5 4 3 0 1003 1] - [ 163 261 186 154 166 339 77 164 116 77 223 211 577 443 178 118 350 70 319 419 8615]] - -2022-12-06 10:28:21,970 - ==> Best [Top1: 80.829 Top5: 97.451 Sparsity:0.00 Params: 5376 on epoch: 11] -2022-12-06 10:28:21,970 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:28:21,976 - - -2022-12-06 10:28:21,976 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:28:22,997 - Epoch: [14][ 10/ 1200] Overall Loss 0.354915 Objective Loss 0.354915 LR 0.001000 Time 0.102013 -2022-12-06 10:28:23,192 - Epoch: [14][ 20/ 1200] Overall Loss 0.355342 Objective Loss 0.355342 LR 0.001000 Time 0.060733 -2022-12-06 10:28:23,384 - Epoch: [14][ 30/ 1200] Overall Loss 0.344666 Objective Loss 0.344666 LR 0.001000 Time 0.046865 -2022-12-06 10:28:23,576 - Epoch: [14][ 40/ 1200] Overall Loss 0.346018 Objective Loss 0.346018 LR 0.001000 Time 0.039927 -2022-12-06 10:28:23,767 - Epoch: [14][ 50/ 1200] Overall Loss 0.349969 Objective Loss 0.349969 LR 0.001000 Time 0.035765 -2022-12-06 10:28:23,959 - Epoch: [14][ 60/ 1200] Overall Loss 0.346271 Objective Loss 0.346271 LR 0.001000 Time 0.032979 -2022-12-06 10:28:24,150 - Epoch: [14][ 70/ 1200] Overall Loss 0.345847 Objective Loss 0.345847 LR 0.001000 Time 0.030991 -2022-12-06 10:28:24,341 - Epoch: [14][ 80/ 1200] Overall Loss 0.353135 Objective Loss 0.353135 LR 0.001000 Time 0.029504 -2022-12-06 10:28:24,532 - Epoch: [14][ 90/ 1200] Overall Loss 0.355480 Objective Loss 0.355480 LR 0.001000 Time 0.028343 -2022-12-06 10:28:24,725 - Epoch: [14][ 100/ 1200] Overall Loss 0.356328 Objective Loss 0.356328 LR 0.001000 Time 0.027433 -2022-12-06 10:28:24,919 - Epoch: [14][ 110/ 1200] Overall Loss 0.353040 Objective Loss 0.353040 LR 0.001000 Time 0.026696 -2022-12-06 10:28:25,112 - Epoch: [14][ 120/ 1200] Overall Loss 0.351418 Objective Loss 0.351418 LR 0.001000 Time 0.026077 -2022-12-06 10:28:25,307 - Epoch: [14][ 130/ 1200] Overall Loss 0.352874 Objective Loss 0.352874 LR 0.001000 Time 0.025563 -2022-12-06 10:28:25,499 - Epoch: [14][ 140/ 1200] Overall Loss 0.354041 Objective Loss 0.354041 LR 0.001000 Time 0.025108 -2022-12-06 10:28:25,694 - Epoch: [14][ 150/ 1200] Overall Loss 0.353991 Objective Loss 0.353991 LR 0.001000 Time 0.024727 -2022-12-06 10:28:25,888 - Epoch: [14][ 160/ 1200] Overall Loss 0.354311 Objective Loss 0.354311 LR 0.001000 Time 0.024390 -2022-12-06 10:28:26,082 - Epoch: [14][ 170/ 1200] Overall Loss 0.353508 Objective Loss 0.353508 LR 0.001000 Time 0.024096 -2022-12-06 10:28:26,275 - Epoch: [14][ 180/ 1200] Overall Loss 0.353364 Objective Loss 0.353364 LR 0.001000 Time 0.023828 -2022-12-06 10:28:26,469 - Epoch: [14][ 190/ 1200] Overall Loss 0.352589 Objective Loss 0.352589 LR 0.001000 Time 0.023592 -2022-12-06 10:28:26,663 - Epoch: [14][ 200/ 1200] Overall Loss 0.352769 Objective Loss 0.352769 LR 0.001000 Time 0.023379 -2022-12-06 10:28:26,857 - Epoch: [14][ 210/ 1200] Overall Loss 0.351160 Objective Loss 0.351160 LR 0.001000 Time 0.023186 -2022-12-06 10:28:27,050 - Epoch: [14][ 220/ 1200] Overall Loss 0.351325 Objective Loss 0.351325 LR 0.001000 Time 0.023006 -2022-12-06 10:28:27,244 - Epoch: [14][ 230/ 1200] Overall Loss 0.351288 Objective Loss 0.351288 LR 0.001000 Time 0.022845 -2022-12-06 10:28:27,436 - Epoch: [14][ 240/ 1200] Overall Loss 0.351754 Objective Loss 0.351754 LR 0.001000 Time 0.022694 -2022-12-06 10:28:27,630 - Epoch: [14][ 250/ 1200] Overall Loss 0.352706 Objective Loss 0.352706 LR 0.001000 Time 0.022559 -2022-12-06 10:28:27,823 - Epoch: [14][ 260/ 1200] Overall Loss 0.353500 Objective Loss 0.353500 LR 0.001000 Time 0.022432 -2022-12-06 10:28:28,017 - Epoch: [14][ 270/ 1200] Overall Loss 0.354056 Objective Loss 0.354056 LR 0.001000 Time 0.022316 -2022-12-06 10:28:28,210 - Epoch: [14][ 280/ 1200] Overall Loss 0.354758 Objective Loss 0.354758 LR 0.001000 Time 0.022206 -2022-12-06 10:28:28,403 - Epoch: [14][ 290/ 1200] Overall Loss 0.355321 Objective Loss 0.355321 LR 0.001000 Time 0.022107 -2022-12-06 10:28:28,597 - Epoch: [14][ 300/ 1200] Overall Loss 0.355180 Objective Loss 0.355180 LR 0.001000 Time 0.022012 -2022-12-06 10:28:28,791 - Epoch: [14][ 310/ 1200] Overall Loss 0.356081 Objective Loss 0.356081 LR 0.001000 Time 0.021926 -2022-12-06 10:28:28,984 - Epoch: [14][ 320/ 1200] Overall Loss 0.356628 Objective Loss 0.356628 LR 0.001000 Time 0.021842 -2022-12-06 10:28:29,176 - Epoch: [14][ 330/ 1200] Overall Loss 0.355975 Objective Loss 0.355975 LR 0.001000 Time 0.021762 -2022-12-06 10:28:29,368 - Epoch: [14][ 340/ 1200] Overall Loss 0.356015 Objective Loss 0.356015 LR 0.001000 Time 0.021685 -2022-12-06 10:28:29,563 - Epoch: [14][ 350/ 1200] Overall Loss 0.356795 Objective Loss 0.356795 LR 0.001000 Time 0.021621 -2022-12-06 10:28:29,756 - Epoch: [14][ 360/ 1200] Overall Loss 0.356932 Objective Loss 0.356932 LR 0.001000 Time 0.021555 -2022-12-06 10:28:29,950 - Epoch: [14][ 370/ 1200] Overall Loss 0.357110 Objective Loss 0.357110 LR 0.001000 Time 0.021495 -2022-12-06 10:28:30,143 - Epoch: [14][ 380/ 1200] Overall Loss 0.356554 Objective Loss 0.356554 LR 0.001000 Time 0.021436 -2022-12-06 10:28:30,336 - Epoch: [14][ 390/ 1200] Overall Loss 0.357275 Objective Loss 0.357275 LR 0.001000 Time 0.021379 -2022-12-06 10:28:30,529 - Epoch: [14][ 400/ 1200] Overall Loss 0.357186 Objective Loss 0.357186 LR 0.001000 Time 0.021326 -2022-12-06 10:28:30,723 - Epoch: [14][ 410/ 1200] Overall Loss 0.356683 Objective Loss 0.356683 LR 0.001000 Time 0.021278 -2022-12-06 10:28:30,916 - Epoch: [14][ 420/ 1200] Overall Loss 0.356477 Objective Loss 0.356477 LR 0.001000 Time 0.021229 -2022-12-06 10:28:31,109 - Epoch: [14][ 430/ 1200] Overall Loss 0.356250 Objective Loss 0.356250 LR 0.001000 Time 0.021185 -2022-12-06 10:28:31,302 - Epoch: [14][ 440/ 1200] Overall Loss 0.356550 Objective Loss 0.356550 LR 0.001000 Time 0.021140 -2022-12-06 10:28:31,497 - Epoch: [14][ 450/ 1200] Overall Loss 0.356478 Objective Loss 0.356478 LR 0.001000 Time 0.021101 -2022-12-06 10:28:31,689 - Epoch: [14][ 460/ 1200] Overall Loss 0.356979 Objective Loss 0.356979 LR 0.001000 Time 0.021060 -2022-12-06 10:28:31,883 - Epoch: [14][ 470/ 1200] Overall Loss 0.356863 Objective Loss 0.356863 LR 0.001000 Time 0.021023 -2022-12-06 10:28:32,075 - Epoch: [14][ 480/ 1200] Overall Loss 0.357390 Objective Loss 0.357390 LR 0.001000 Time 0.020985 -2022-12-06 10:28:32,269 - Epoch: [14][ 490/ 1200] Overall Loss 0.357770 Objective Loss 0.357770 LR 0.001000 Time 0.020950 -2022-12-06 10:28:32,462 - Epoch: [14][ 500/ 1200] Overall Loss 0.357951 Objective Loss 0.357951 LR 0.001000 Time 0.020916 -2022-12-06 10:28:32,656 - Epoch: [14][ 510/ 1200] Overall Loss 0.358183 Objective Loss 0.358183 LR 0.001000 Time 0.020886 -2022-12-06 10:28:32,849 - Epoch: [14][ 520/ 1200] Overall Loss 0.358609 Objective Loss 0.358609 LR 0.001000 Time 0.020854 -2022-12-06 10:28:33,043 - Epoch: [14][ 530/ 1200] Overall Loss 0.358356 Objective Loss 0.358356 LR 0.001000 Time 0.020826 -2022-12-06 10:28:33,237 - Epoch: [14][ 540/ 1200] Overall Loss 0.358296 Objective Loss 0.358296 LR 0.001000 Time 0.020797 -2022-12-06 10:28:33,431 - Epoch: [14][ 550/ 1200] Overall Loss 0.358711 Objective Loss 0.358711 LR 0.001000 Time 0.020772 -2022-12-06 10:28:33,624 - Epoch: [14][ 560/ 1200] Overall Loss 0.358467 Objective Loss 0.358467 LR 0.001000 Time 0.020744 -2022-12-06 10:28:33,818 - Epoch: [14][ 570/ 1200] Overall Loss 0.358310 Objective Loss 0.358310 LR 0.001000 Time 0.020720 -2022-12-06 10:28:34,011 - Epoch: [14][ 580/ 1200] Overall Loss 0.358222 Objective Loss 0.358222 LR 0.001000 Time 0.020694 -2022-12-06 10:28:34,205 - Epoch: [14][ 590/ 1200] Overall Loss 0.358153 Objective Loss 0.358153 LR 0.001000 Time 0.020671 -2022-12-06 10:28:34,397 - Epoch: [14][ 600/ 1200] Overall Loss 0.358623 Objective Loss 0.358623 LR 0.001000 Time 0.020646 -2022-12-06 10:28:34,591 - Epoch: [14][ 610/ 1200] Overall Loss 0.358767 Objective Loss 0.358767 LR 0.001000 Time 0.020625 -2022-12-06 10:28:34,784 - Epoch: [14][ 620/ 1200] Overall Loss 0.359234 Objective Loss 0.359234 LR 0.001000 Time 0.020602 -2022-12-06 10:28:34,977 - Epoch: [14][ 630/ 1200] Overall Loss 0.358845 Objective Loss 0.358845 LR 0.001000 Time 0.020581 -2022-12-06 10:28:35,171 - Epoch: [14][ 640/ 1200] Overall Loss 0.359036 Objective Loss 0.359036 LR 0.001000 Time 0.020562 -2022-12-06 10:28:35,364 - Epoch: [14][ 650/ 1200] Overall Loss 0.359386 Objective Loss 0.359386 LR 0.001000 Time 0.020542 -2022-12-06 10:28:35,558 - Epoch: [14][ 660/ 1200] Overall Loss 0.359475 Objective Loss 0.359475 LR 0.001000 Time 0.020523 -2022-12-06 10:28:35,752 - Epoch: [14][ 670/ 1200] Overall Loss 0.359621 Objective Loss 0.359621 LR 0.001000 Time 0.020505 -2022-12-06 10:28:35,945 - Epoch: [14][ 680/ 1200] Overall Loss 0.359417 Objective Loss 0.359417 LR 0.001000 Time 0.020487 -2022-12-06 10:28:36,139 - Epoch: [14][ 690/ 1200] Overall Loss 0.359364 Objective Loss 0.359364 LR 0.001000 Time 0.020470 -2022-12-06 10:28:36,331 - Epoch: [14][ 700/ 1200] Overall Loss 0.359142 Objective Loss 0.359142 LR 0.001000 Time 0.020452 -2022-12-06 10:28:36,525 - Epoch: [14][ 710/ 1200] Overall Loss 0.359213 Objective Loss 0.359213 LR 0.001000 Time 0.020436 -2022-12-06 10:28:36,718 - Epoch: [14][ 720/ 1200] Overall Loss 0.359363 Objective Loss 0.359363 LR 0.001000 Time 0.020420 -2022-12-06 10:28:36,911 - Epoch: [14][ 730/ 1200] Overall Loss 0.359580 Objective Loss 0.359580 LR 0.001000 Time 0.020403 -2022-12-06 10:28:37,103 - Epoch: [14][ 740/ 1200] Overall Loss 0.359781 Objective Loss 0.359781 LR 0.001000 Time 0.020387 -2022-12-06 10:28:37,297 - Epoch: [14][ 750/ 1200] Overall Loss 0.359584 Objective Loss 0.359584 LR 0.001000 Time 0.020372 -2022-12-06 10:28:37,490 - Epoch: [14][ 760/ 1200] Overall Loss 0.359713 Objective Loss 0.359713 LR 0.001000 Time 0.020357 -2022-12-06 10:28:37,684 - Epoch: [14][ 770/ 1200] Overall Loss 0.359240 Objective Loss 0.359240 LR 0.001000 Time 0.020345 -2022-12-06 10:28:37,877 - Epoch: [14][ 780/ 1200] Overall Loss 0.359174 Objective Loss 0.359174 LR 0.001000 Time 0.020330 -2022-12-06 10:28:38,071 - Epoch: [14][ 790/ 1200] Overall Loss 0.359322 Objective Loss 0.359322 LR 0.001000 Time 0.020318 -2022-12-06 10:28:38,264 - Epoch: [14][ 800/ 1200] Overall Loss 0.359688 Objective Loss 0.359688 LR 0.001000 Time 0.020304 -2022-12-06 10:28:38,457 - Epoch: [14][ 810/ 1200] Overall Loss 0.359675 Objective Loss 0.359675 LR 0.001000 Time 0.020292 -2022-12-06 10:28:38,650 - Epoch: [14][ 820/ 1200] Overall Loss 0.359644 Objective Loss 0.359644 LR 0.001000 Time 0.020278 -2022-12-06 10:28:38,843 - Epoch: [14][ 830/ 1200] Overall Loss 0.359535 Objective Loss 0.359535 LR 0.001000 Time 0.020266 -2022-12-06 10:28:39,035 - Epoch: [14][ 840/ 1200] Overall Loss 0.359620 Objective Loss 0.359620 LR 0.001000 Time 0.020253 -2022-12-06 10:28:39,228 - Epoch: [14][ 850/ 1200] Overall Loss 0.359778 Objective Loss 0.359778 LR 0.001000 Time 0.020241 -2022-12-06 10:28:39,421 - Epoch: [14][ 860/ 1200] Overall Loss 0.359603 Objective Loss 0.359603 LR 0.001000 Time 0.020230 -2022-12-06 10:28:39,615 - Epoch: [14][ 870/ 1200] Overall Loss 0.359669 Objective Loss 0.359669 LR 0.001000 Time 0.020219 -2022-12-06 10:28:39,807 - Epoch: [14][ 880/ 1200] Overall Loss 0.359459 Objective Loss 0.359459 LR 0.001000 Time 0.020207 -2022-12-06 10:28:40,001 - Epoch: [14][ 890/ 1200] Overall Loss 0.359785 Objective Loss 0.359785 LR 0.001000 Time 0.020197 -2022-12-06 10:28:40,194 - Epoch: [14][ 900/ 1200] Overall Loss 0.359572 Objective Loss 0.359572 LR 0.001000 Time 0.020187 -2022-12-06 10:28:40,388 - Epoch: [14][ 910/ 1200] Overall Loss 0.359343 Objective Loss 0.359343 LR 0.001000 Time 0.020177 -2022-12-06 10:28:40,581 - Epoch: [14][ 920/ 1200] Overall Loss 0.359158 Objective Loss 0.359158 LR 0.001000 Time 0.020167 -2022-12-06 10:28:40,775 - Epoch: [14][ 930/ 1200] Overall Loss 0.359464 Objective Loss 0.359464 LR 0.001000 Time 0.020159 -2022-12-06 10:28:40,968 - Epoch: [14][ 940/ 1200] Overall Loss 0.359349 Objective Loss 0.359349 LR 0.001000 Time 0.020149 -2022-12-06 10:28:41,162 - Epoch: [14][ 950/ 1200] Overall Loss 0.359621 Objective Loss 0.359621 LR 0.001000 Time 0.020140 -2022-12-06 10:28:41,355 - Epoch: [14][ 960/ 1200] Overall Loss 0.359303 Objective Loss 0.359303 LR 0.001000 Time 0.020131 -2022-12-06 10:28:41,548 - Epoch: [14][ 970/ 1200] Overall Loss 0.359652 Objective Loss 0.359652 LR 0.001000 Time 0.020122 -2022-12-06 10:28:41,741 - Epoch: [14][ 980/ 1200] Overall Loss 0.359470 Objective Loss 0.359470 LR 0.001000 Time 0.020112 -2022-12-06 10:28:41,935 - Epoch: [14][ 990/ 1200] Overall Loss 0.359813 Objective Loss 0.359813 LR 0.001000 Time 0.020105 -2022-12-06 10:28:42,128 - Epoch: [14][ 1000/ 1200] Overall Loss 0.360139 Objective Loss 0.360139 LR 0.001000 Time 0.020096 -2022-12-06 10:28:42,322 - Epoch: [14][ 1010/ 1200] Overall Loss 0.359966 Objective Loss 0.359966 LR 0.001000 Time 0.020089 -2022-12-06 10:28:42,515 - Epoch: [14][ 1020/ 1200] Overall Loss 0.360091 Objective Loss 0.360091 LR 0.001000 Time 0.020081 -2022-12-06 10:28:42,707 - Epoch: [14][ 1030/ 1200] Overall Loss 0.360056 Objective Loss 0.360056 LR 0.001000 Time 0.020072 -2022-12-06 10:28:42,899 - Epoch: [14][ 1040/ 1200] Overall Loss 0.359957 Objective Loss 0.359957 LR 0.001000 Time 0.020063 -2022-12-06 10:28:43,090 - Epoch: [14][ 1050/ 1200] Overall Loss 0.360315 Objective Loss 0.360315 LR 0.001000 Time 0.020053 -2022-12-06 10:28:43,281 - Epoch: [14][ 1060/ 1200] Overall Loss 0.360335 Objective Loss 0.360335 LR 0.001000 Time 0.020044 -2022-12-06 10:28:43,473 - Epoch: [14][ 1070/ 1200] Overall Loss 0.360635 Objective Loss 0.360635 LR 0.001000 Time 0.020035 -2022-12-06 10:28:43,663 - Epoch: [14][ 1080/ 1200] Overall Loss 0.360892 Objective Loss 0.360892 LR 0.001000 Time 0.020025 -2022-12-06 10:28:43,854 - Epoch: [14][ 1090/ 1200] Overall Loss 0.361010 Objective Loss 0.361010 LR 0.001000 Time 0.020016 -2022-12-06 10:28:44,044 - Epoch: [14][ 1100/ 1200] Overall Loss 0.361040 Objective Loss 0.361040 LR 0.001000 Time 0.020007 -2022-12-06 10:28:44,235 - Epoch: [14][ 1110/ 1200] Overall Loss 0.361015 Objective Loss 0.361015 LR 0.001000 Time 0.019998 -2022-12-06 10:28:44,426 - Epoch: [14][ 1120/ 1200] Overall Loss 0.361323 Objective Loss 0.361323 LR 0.001000 Time 0.019989 -2022-12-06 10:28:44,616 - Epoch: [14][ 1130/ 1200] Overall Loss 0.361514 Objective Loss 0.361514 LR 0.001000 Time 0.019980 -2022-12-06 10:28:44,807 - Epoch: [14][ 1140/ 1200] Overall Loss 0.361304 Objective Loss 0.361304 LR 0.001000 Time 0.019972 -2022-12-06 10:28:44,997 - Epoch: [14][ 1150/ 1200] Overall Loss 0.361265 Objective Loss 0.361265 LR 0.001000 Time 0.019963 -2022-12-06 10:28:45,188 - Epoch: [14][ 1160/ 1200] Overall Loss 0.361374 Objective Loss 0.361374 LR 0.001000 Time 0.019955 -2022-12-06 10:28:45,379 - Epoch: [14][ 1170/ 1200] Overall Loss 0.361621 Objective Loss 0.361621 LR 0.001000 Time 0.019947 -2022-12-06 10:28:45,571 - Epoch: [14][ 1180/ 1200] Overall Loss 0.361716 Objective Loss 0.361716 LR 0.001000 Time 0.019940 -2022-12-06 10:28:45,761 - Epoch: [14][ 1190/ 1200] Overall Loss 0.361739 Objective Loss 0.361739 LR 0.001000 Time 0.019932 -2022-12-06 10:28:45,989 - Epoch: [14][ 1200/ 1200] Overall Loss 0.361374 Objective Loss 0.361374 Top1 83.263598 Top5 97.280335 LR 0.001000 Time 0.019956 -2022-12-06 10:28:46,078 - --- validate (epoch=14)----------- -2022-12-06 10:28:46,078 - 34129 samples (256 per mini-batch) -2022-12-06 10:28:46,518 - Epoch: [14][ 10/ 134] Loss 0.321692 Top1 84.414062 Top5 97.812500 -2022-12-06 10:28:46,650 - Epoch: [14][ 20/ 134] Loss 0.331137 Top1 83.457031 Top5 97.773438 -2022-12-06 10:28:46,784 - Epoch: [14][ 30/ 134] Loss 0.344311 Top1 82.851562 Top5 97.630208 -2022-12-06 10:28:46,914 - Epoch: [14][ 40/ 134] Loss 0.348852 Top1 82.724609 Top5 97.626953 -2022-12-06 10:28:47,047 - Epoch: [14][ 50/ 134] Loss 0.349662 Top1 82.843750 Top5 97.640625 -2022-12-06 10:28:47,178 - Epoch: [14][ 60/ 134] Loss 0.349880 Top1 82.760417 Top5 97.649740 -2022-12-06 10:28:47,310 - Epoch: [14][ 70/ 134] Loss 0.354235 Top1 82.684152 Top5 97.600446 -2022-12-06 10:28:47,440 - Epoch: [14][ 80/ 134] Loss 0.351520 Top1 82.685547 Top5 97.548828 -2022-12-06 10:28:47,572 - Epoch: [14][ 90/ 134] Loss 0.347465 Top1 82.695312 Top5 97.565104 -2022-12-06 10:28:47,702 - Epoch: [14][ 100/ 134] Loss 0.347479 Top1 82.726562 Top5 97.550781 -2022-12-06 10:28:47,834 - Epoch: [14][ 110/ 134] Loss 0.346325 Top1 82.755682 Top5 97.578125 -2022-12-06 10:28:47,963 - Epoch: [14][ 120/ 134] Loss 0.347021 Top1 82.705078 Top5 97.574870 -2022-12-06 10:28:48,098 - Epoch: [14][ 130/ 134] Loss 0.345786 Top1 82.668269 Top5 97.590144 -2022-12-06 10:28:48,137 - Epoch: [14][ 134/ 134] Loss 0.345781 Top1 82.671628 Top5 97.594421 -2022-12-06 10:28:48,227 - ==> Top1: 82.672 Top5: 97.594 Loss: 0.346 - -2022-12-06 10:28:48,228 - ==> Confusion: -[[ 875 1 1 1 7 1 0 1 5 74 0 7 2 3 6 3 2 1 0 0 6] - [ 2 908 1 1 10 36 4 23 3 2 1 6 2 2 3 2 6 3 4 0 8] - [ 11 5 970 14 4 3 40 10 0 0 4 8 1 5 2 6 3 0 2 3 12] - [ 3 2 36 880 0 1 2 3 4 0 8 1 8 3 24 2 2 7 17 0 17] - [ 12 10 3 0 938 6 1 2 1 7 0 3 1 4 9 9 8 1 1 0 4] - [ 1 19 1 1 6 947 4 21 6 3 0 16 1 26 0 1 1 1 2 6 6] - [ 0 4 7 0 0 3 1070 4 1 0 2 3 1 0 2 5 1 1 0 11 3] - [ 0 9 15 1 2 39 6 923 1 3 0 15 0 0 0 1 0 1 19 14 5] - [ 5 1 0 0 0 1 0 1 976 50 2 2 4 5 6 1 0 1 3 2 4] - [ 63 0 1 0 3 2 0 3 29 875 0 3 0 11 1 1 0 0 0 0 9] - [ 0 3 10 5 1 3 3 2 17 4 920 4 5 17 7 0 0 1 8 2 7] - [ 4 0 4 0 0 15 3 1 2 1 0 969 14 12 0 6 3 9 1 5 2] - [ 0 1 0 1 1 3 1 1 1 0 1 64 853 4 0 8 1 14 1 6 8] - [ 1 4 0 0 2 6 0 1 14 14 6 7 4 942 0 4 2 4 1 1 10] - [ 14 3 0 12 6 3 1 0 39 6 1 1 3 2 1006 0 4 7 7 1 14] - [ 6 1 2 2 1 0 2 0 1 1 0 18 5 4 0 972 10 7 1 5 5] - [ 2 6 4 0 3 2 1 0 0 0 0 6 2 1 1 7 1023 2 0 4 8] - [ 4 1 3 1 0 1 2 0 2 3 1 16 26 4 1 14 3 949 1 1 3] - [ 3 8 5 6 0 2 0 31 6 1 5 4 6 1 11 0 0 1 904 4 10] - [ 1 3 4 1 0 7 5 5 1 0 1 34 5 3 0 6 4 2 0 993 5] - [ 197 252 217 59 138 204 91 179 134 121 137 230 407 396 117 160 303 62 163 344 9315]] - -2022-12-06 10:28:48,811 - ==> Best [Top1: 82.672 Top5: 97.594 Sparsity:0.00 Params: 5376 on epoch: 14] -2022-12-06 10:28:48,812 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:28:48,818 - - -2022-12-06 10:28:48,819 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:28:49,751 - Epoch: [15][ 10/ 1200] Overall Loss 0.311702 Objective Loss 0.311702 LR 0.001000 Time 0.093188 -2022-12-06 10:28:49,954 - Epoch: [15][ 20/ 1200] Overall Loss 0.316684 Objective Loss 0.316684 LR 0.001000 Time 0.056686 -2022-12-06 10:28:50,147 - Epoch: [15][ 30/ 1200] Overall Loss 0.328698 Objective Loss 0.328698 LR 0.001000 Time 0.044208 -2022-12-06 10:28:50,339 - Epoch: [15][ 40/ 1200] Overall Loss 0.321271 Objective Loss 0.321271 LR 0.001000 Time 0.037948 -2022-12-06 10:28:50,531 - Epoch: [15][ 50/ 1200] Overall Loss 0.330999 Objective Loss 0.330999 LR 0.001000 Time 0.034191 -2022-12-06 10:28:50,723 - Epoch: [15][ 60/ 1200] Overall Loss 0.329385 Objective Loss 0.329385 LR 0.001000 Time 0.031688 -2022-12-06 10:28:50,915 - Epoch: [15][ 70/ 1200] Overall Loss 0.327896 Objective Loss 0.327896 LR 0.001000 Time 0.029895 -2022-12-06 10:28:51,107 - Epoch: [15][ 80/ 1200] Overall Loss 0.332061 Objective Loss 0.332061 LR 0.001000 Time 0.028549 -2022-12-06 10:28:51,299 - Epoch: [15][ 90/ 1200] Overall Loss 0.329389 Objective Loss 0.329389 LR 0.001000 Time 0.027505 -2022-12-06 10:28:51,491 - Epoch: [15][ 100/ 1200] Overall Loss 0.333015 Objective Loss 0.333015 LR 0.001000 Time 0.026669 -2022-12-06 10:28:51,683 - Epoch: [15][ 110/ 1200] Overall Loss 0.333481 Objective Loss 0.333481 LR 0.001000 Time 0.025983 -2022-12-06 10:28:51,875 - Epoch: [15][ 120/ 1200] Overall Loss 0.335563 Objective Loss 0.335563 LR 0.001000 Time 0.025412 -2022-12-06 10:28:52,067 - Epoch: [15][ 130/ 1200] Overall Loss 0.334045 Objective Loss 0.334045 LR 0.001000 Time 0.024934 -2022-12-06 10:28:52,259 - Epoch: [15][ 140/ 1200] Overall Loss 0.337054 Objective Loss 0.337054 LR 0.001000 Time 0.024520 -2022-12-06 10:28:52,451 - Epoch: [15][ 150/ 1200] Overall Loss 0.335918 Objective Loss 0.335918 LR 0.001000 Time 0.024165 -2022-12-06 10:28:52,644 - Epoch: [15][ 160/ 1200] Overall Loss 0.334804 Objective Loss 0.334804 LR 0.001000 Time 0.023853 -2022-12-06 10:28:52,836 - Epoch: [15][ 170/ 1200] Overall Loss 0.334031 Objective Loss 0.334031 LR 0.001000 Time 0.023578 -2022-12-06 10:28:53,028 - Epoch: [15][ 180/ 1200] Overall Loss 0.331909 Objective Loss 0.331909 LR 0.001000 Time 0.023329 -2022-12-06 10:28:53,220 - Epoch: [15][ 190/ 1200] Overall Loss 0.329885 Objective Loss 0.329885 LR 0.001000 Time 0.023111 -2022-12-06 10:28:53,411 - Epoch: [15][ 200/ 1200] Overall Loss 0.331191 Objective Loss 0.331191 LR 0.001000 Time 0.022909 -2022-12-06 10:28:53,603 - Epoch: [15][ 210/ 1200] Overall Loss 0.332681 Objective Loss 0.332681 LR 0.001000 Time 0.022731 -2022-12-06 10:28:53,795 - Epoch: [15][ 220/ 1200] Overall Loss 0.334003 Objective Loss 0.334003 LR 0.001000 Time 0.022566 -2022-12-06 10:28:53,987 - Epoch: [15][ 230/ 1200] Overall Loss 0.333495 Objective Loss 0.333495 LR 0.001000 Time 0.022416 -2022-12-06 10:28:54,180 - Epoch: [15][ 240/ 1200] Overall Loss 0.334726 Objective Loss 0.334726 LR 0.001000 Time 0.022284 -2022-12-06 10:28:54,372 - Epoch: [15][ 250/ 1200] Overall Loss 0.334201 Objective Loss 0.334201 LR 0.001000 Time 0.022158 -2022-12-06 10:28:54,563 - Epoch: [15][ 260/ 1200] Overall Loss 0.335836 Objective Loss 0.335836 LR 0.001000 Time 0.022040 -2022-12-06 10:28:54,755 - Epoch: [15][ 270/ 1200] Overall Loss 0.338112 Objective Loss 0.338112 LR 0.001000 Time 0.021933 -2022-12-06 10:28:54,946 - Epoch: [15][ 280/ 1200] Overall Loss 0.337398 Objective Loss 0.337398 LR 0.001000 Time 0.021830 -2022-12-06 10:28:55,138 - Epoch: [15][ 290/ 1200] Overall Loss 0.338491 Objective Loss 0.338491 LR 0.001000 Time 0.021738 -2022-12-06 10:28:55,330 - Epoch: [15][ 300/ 1200] Overall Loss 0.338411 Objective Loss 0.338411 LR 0.001000 Time 0.021650 -2022-12-06 10:28:55,522 - Epoch: [15][ 310/ 1200] Overall Loss 0.338401 Objective Loss 0.338401 LR 0.001000 Time 0.021570 -2022-12-06 10:28:55,715 - Epoch: [15][ 320/ 1200] Overall Loss 0.338762 Objective Loss 0.338762 LR 0.001000 Time 0.021497 -2022-12-06 10:28:55,908 - Epoch: [15][ 330/ 1200] Overall Loss 0.339525 Objective Loss 0.339525 LR 0.001000 Time 0.021429 -2022-12-06 10:28:56,100 - Epoch: [15][ 340/ 1200] Overall Loss 0.339487 Objective Loss 0.339487 LR 0.001000 Time 0.021362 -2022-12-06 10:28:56,293 - Epoch: [15][ 350/ 1200] Overall Loss 0.339759 Objective Loss 0.339759 LR 0.001000 Time 0.021301 -2022-12-06 10:28:56,484 - Epoch: [15][ 360/ 1200] Overall Loss 0.339971 Objective Loss 0.339971 LR 0.001000 Time 0.021240 -2022-12-06 10:28:56,677 - Epoch: [15][ 370/ 1200] Overall Loss 0.340224 Objective Loss 0.340224 LR 0.001000 Time 0.021184 -2022-12-06 10:28:56,868 - Epoch: [15][ 380/ 1200] Overall Loss 0.340228 Objective Loss 0.340228 LR 0.001000 Time 0.021129 -2022-12-06 10:28:57,060 - Epoch: [15][ 390/ 1200] Overall Loss 0.340834 Objective Loss 0.340834 LR 0.001000 Time 0.021079 -2022-12-06 10:28:57,252 - Epoch: [15][ 400/ 1200] Overall Loss 0.341400 Objective Loss 0.341400 LR 0.001000 Time 0.021031 -2022-12-06 10:28:57,444 - Epoch: [15][ 410/ 1200] Overall Loss 0.341598 Objective Loss 0.341598 LR 0.001000 Time 0.020985 -2022-12-06 10:28:57,636 - Epoch: [15][ 420/ 1200] Overall Loss 0.342186 Objective Loss 0.342186 LR 0.001000 Time 0.020940 -2022-12-06 10:28:57,828 - Epoch: [15][ 430/ 1200] Overall Loss 0.341798 Objective Loss 0.341798 LR 0.001000 Time 0.020898 -2022-12-06 10:28:58,019 - Epoch: [15][ 440/ 1200] Overall Loss 0.340962 Objective Loss 0.340962 LR 0.001000 Time 0.020857 -2022-12-06 10:28:58,211 - Epoch: [15][ 450/ 1200] Overall Loss 0.341450 Objective Loss 0.341450 LR 0.001000 Time 0.020819 -2022-12-06 10:28:58,403 - Epoch: [15][ 460/ 1200] Overall Loss 0.341329 Objective Loss 0.341329 LR 0.001000 Time 0.020781 -2022-12-06 10:28:58,595 - Epoch: [15][ 470/ 1200] Overall Loss 0.341803 Objective Loss 0.341803 LR 0.001000 Time 0.020747 -2022-12-06 10:28:58,787 - Epoch: [15][ 480/ 1200] Overall Loss 0.342262 Objective Loss 0.342262 LR 0.001000 Time 0.020713 -2022-12-06 10:28:58,979 - Epoch: [15][ 490/ 1200] Overall Loss 0.342543 Objective Loss 0.342543 LR 0.001000 Time 0.020683 -2022-12-06 10:28:59,171 - Epoch: [15][ 500/ 1200] Overall Loss 0.342992 Objective Loss 0.342992 LR 0.001000 Time 0.020651 -2022-12-06 10:28:59,363 - Epoch: [15][ 510/ 1200] Overall Loss 0.343304 Objective Loss 0.343304 LR 0.001000 Time 0.020622 -2022-12-06 10:28:59,555 - Epoch: [15][ 520/ 1200] Overall Loss 0.343721 Objective Loss 0.343721 LR 0.001000 Time 0.020593 -2022-12-06 10:28:59,747 - Epoch: [15][ 530/ 1200] Overall Loss 0.343973 Objective Loss 0.343973 LR 0.001000 Time 0.020567 -2022-12-06 10:28:59,939 - Epoch: [15][ 540/ 1200] Overall Loss 0.343953 Objective Loss 0.343953 LR 0.001000 Time 0.020540 -2022-12-06 10:29:00,131 - Epoch: [15][ 550/ 1200] Overall Loss 0.343986 Objective Loss 0.343986 LR 0.001000 Time 0.020514 -2022-12-06 10:29:00,322 - Epoch: [15][ 560/ 1200] Overall Loss 0.343534 Objective Loss 0.343534 LR 0.001000 Time 0.020489 -2022-12-06 10:29:00,514 - Epoch: [15][ 570/ 1200] Overall Loss 0.343934 Objective Loss 0.343934 LR 0.001000 Time 0.020465 -2022-12-06 10:29:00,706 - Epoch: [15][ 580/ 1200] Overall Loss 0.343900 Objective Loss 0.343900 LR 0.001000 Time 0.020443 -2022-12-06 10:29:00,899 - Epoch: [15][ 590/ 1200] Overall Loss 0.344273 Objective Loss 0.344273 LR 0.001000 Time 0.020422 -2022-12-06 10:29:01,090 - Epoch: [15][ 600/ 1200] Overall Loss 0.344505 Objective Loss 0.344505 LR 0.001000 Time 0.020399 -2022-12-06 10:29:01,283 - Epoch: [15][ 610/ 1200] Overall Loss 0.345009 Objective Loss 0.345009 LR 0.001000 Time 0.020379 -2022-12-06 10:29:01,474 - Epoch: [15][ 620/ 1200] Overall Loss 0.344959 Objective Loss 0.344959 LR 0.001000 Time 0.020359 -2022-12-06 10:29:01,666 - Epoch: [15][ 630/ 1200] Overall Loss 0.345054 Objective Loss 0.345054 LR 0.001000 Time 0.020340 -2022-12-06 10:29:01,858 - Epoch: [15][ 640/ 1200] Overall Loss 0.345252 Objective Loss 0.345252 LR 0.001000 Time 0.020321 -2022-12-06 10:29:02,051 - Epoch: [15][ 650/ 1200] Overall Loss 0.345708 Objective Loss 0.345708 LR 0.001000 Time 0.020304 -2022-12-06 10:29:02,242 - Epoch: [15][ 660/ 1200] Overall Loss 0.345705 Objective Loss 0.345705 LR 0.001000 Time 0.020286 -2022-12-06 10:29:02,435 - Epoch: [15][ 670/ 1200] Overall Loss 0.345943 Objective Loss 0.345943 LR 0.001000 Time 0.020269 -2022-12-06 10:29:02,626 - Epoch: [15][ 680/ 1200] Overall Loss 0.345697 Objective Loss 0.345697 LR 0.001000 Time 0.020252 -2022-12-06 10:29:02,819 - Epoch: [15][ 690/ 1200] Overall Loss 0.345793 Objective Loss 0.345793 LR 0.001000 Time 0.020236 -2022-12-06 10:29:03,010 - Epoch: [15][ 700/ 1200] Overall Loss 0.345369 Objective Loss 0.345369 LR 0.001000 Time 0.020220 -2022-12-06 10:29:03,202 - Epoch: [15][ 710/ 1200] Overall Loss 0.345659 Objective Loss 0.345659 LR 0.001000 Time 0.020205 -2022-12-06 10:29:03,394 - Epoch: [15][ 720/ 1200] Overall Loss 0.345247 Objective Loss 0.345247 LR 0.001000 Time 0.020190 -2022-12-06 10:29:03,586 - Epoch: [15][ 730/ 1200] Overall Loss 0.344444 Objective Loss 0.344444 LR 0.001000 Time 0.020176 -2022-12-06 10:29:03,778 - Epoch: [15][ 740/ 1200] Overall Loss 0.344567 Objective Loss 0.344567 LR 0.001000 Time 0.020162 -2022-12-06 10:29:03,971 - Epoch: [15][ 750/ 1200] Overall Loss 0.344782 Objective Loss 0.344782 LR 0.001000 Time 0.020149 -2022-12-06 10:29:04,162 - Epoch: [15][ 760/ 1200] Overall Loss 0.344625 Objective Loss 0.344625 LR 0.001000 Time 0.020135 -2022-12-06 10:29:04,355 - Epoch: [15][ 770/ 1200] Overall Loss 0.344880 Objective Loss 0.344880 LR 0.001000 Time 0.020123 -2022-12-06 10:29:04,546 - Epoch: [15][ 780/ 1200] Overall Loss 0.345103 Objective Loss 0.345103 LR 0.001000 Time 0.020110 -2022-12-06 10:29:04,739 - Epoch: [15][ 790/ 1200] Overall Loss 0.344830 Objective Loss 0.344830 LR 0.001000 Time 0.020098 -2022-12-06 10:29:04,931 - Epoch: [15][ 800/ 1200] Overall Loss 0.344994 Objective Loss 0.344994 LR 0.001000 Time 0.020086 -2022-12-06 10:29:05,122 - Epoch: [15][ 810/ 1200] Overall Loss 0.344868 Objective Loss 0.344868 LR 0.001000 Time 0.020074 -2022-12-06 10:29:05,314 - Epoch: [15][ 820/ 1200] Overall Loss 0.344994 Objective Loss 0.344994 LR 0.001000 Time 0.020063 -2022-12-06 10:29:05,506 - Epoch: [15][ 830/ 1200] Overall Loss 0.344953 Objective Loss 0.344953 LR 0.001000 Time 0.020052 -2022-12-06 10:29:05,698 - Epoch: [15][ 840/ 1200] Overall Loss 0.345472 Objective Loss 0.345472 LR 0.001000 Time 0.020041 -2022-12-06 10:29:05,889 - Epoch: [15][ 850/ 1200] Overall Loss 0.345271 Objective Loss 0.345271 LR 0.001000 Time 0.020030 -2022-12-06 10:29:06,081 - Epoch: [15][ 860/ 1200] Overall Loss 0.345815 Objective Loss 0.345815 LR 0.001000 Time 0.020020 -2022-12-06 10:29:06,274 - Epoch: [15][ 870/ 1200] Overall Loss 0.345937 Objective Loss 0.345937 LR 0.001000 Time 0.020010 -2022-12-06 10:29:06,466 - Epoch: [15][ 880/ 1200] Overall Loss 0.345905 Objective Loss 0.345905 LR 0.001000 Time 0.020000 -2022-12-06 10:29:06,658 - Epoch: [15][ 890/ 1200] Overall Loss 0.345628 Objective Loss 0.345628 LR 0.001000 Time 0.019991 -2022-12-06 10:29:06,850 - Epoch: [15][ 900/ 1200] Overall Loss 0.346095 Objective Loss 0.346095 LR 0.001000 Time 0.019981 -2022-12-06 10:29:07,042 - Epoch: [15][ 910/ 1200] Overall Loss 0.345920 Objective Loss 0.345920 LR 0.001000 Time 0.019972 -2022-12-06 10:29:07,234 - Epoch: [15][ 920/ 1200] Overall Loss 0.346325 Objective Loss 0.346325 LR 0.001000 Time 0.019963 -2022-12-06 10:29:07,427 - Epoch: [15][ 930/ 1200] Overall Loss 0.346465 Objective Loss 0.346465 LR 0.001000 Time 0.019955 -2022-12-06 10:29:07,619 - Epoch: [15][ 940/ 1200] Overall Loss 0.346701 Objective Loss 0.346701 LR 0.001000 Time 0.019947 -2022-12-06 10:29:07,811 - Epoch: [15][ 950/ 1200] Overall Loss 0.346610 Objective Loss 0.346610 LR 0.001000 Time 0.019938 -2022-12-06 10:29:08,002 - Epoch: [15][ 960/ 1200] Overall Loss 0.346875 Objective Loss 0.346875 LR 0.001000 Time 0.019930 -2022-12-06 10:29:08,195 - Epoch: [15][ 970/ 1200] Overall Loss 0.347018 Objective Loss 0.347018 LR 0.001000 Time 0.019922 -2022-12-06 10:29:08,386 - Epoch: [15][ 980/ 1200] Overall Loss 0.346806 Objective Loss 0.346806 LR 0.001000 Time 0.019914 -2022-12-06 10:29:08,578 - Epoch: [15][ 990/ 1200] Overall Loss 0.347001 Objective Loss 0.347001 LR 0.001000 Time 0.019906 -2022-12-06 10:29:08,770 - Epoch: [15][ 1000/ 1200] Overall Loss 0.347050 Objective Loss 0.347050 LR 0.001000 Time 0.019898 -2022-12-06 10:29:08,963 - Epoch: [15][ 1010/ 1200] Overall Loss 0.347132 Objective Loss 0.347132 LR 0.001000 Time 0.019891 -2022-12-06 10:29:09,154 - Epoch: [15][ 1020/ 1200] Overall Loss 0.347352 Objective Loss 0.347352 LR 0.001000 Time 0.019883 -2022-12-06 10:29:09,346 - Epoch: [15][ 1030/ 1200] Overall Loss 0.347245 Objective Loss 0.347245 LR 0.001000 Time 0.019877 -2022-12-06 10:29:09,538 - Epoch: [15][ 1040/ 1200] Overall Loss 0.347258 Objective Loss 0.347258 LR 0.001000 Time 0.019869 -2022-12-06 10:29:09,730 - Epoch: [15][ 1050/ 1200] Overall Loss 0.346980 Objective Loss 0.346980 LR 0.001000 Time 0.019862 -2022-12-06 10:29:09,922 - Epoch: [15][ 1060/ 1200] Overall Loss 0.346947 Objective Loss 0.346947 LR 0.001000 Time 0.019855 -2022-12-06 10:29:10,114 - Epoch: [15][ 1070/ 1200] Overall Loss 0.346958 Objective Loss 0.346958 LR 0.001000 Time 0.019848 -2022-12-06 10:29:10,305 - Epoch: [15][ 1080/ 1200] Overall Loss 0.347186 Objective Loss 0.347186 LR 0.001000 Time 0.019842 -2022-12-06 10:29:10,498 - Epoch: [15][ 1090/ 1200] Overall Loss 0.347003 Objective Loss 0.347003 LR 0.001000 Time 0.019836 -2022-12-06 10:29:10,690 - Epoch: [15][ 1100/ 1200] Overall Loss 0.347029 Objective Loss 0.347029 LR 0.001000 Time 0.019829 -2022-12-06 10:29:10,882 - Epoch: [15][ 1110/ 1200] Overall Loss 0.346949 Objective Loss 0.346949 LR 0.001000 Time 0.019823 -2022-12-06 10:29:11,074 - Epoch: [15][ 1120/ 1200] Overall Loss 0.347077 Objective Loss 0.347077 LR 0.001000 Time 0.019817 -2022-12-06 10:29:11,266 - Epoch: [15][ 1130/ 1200] Overall Loss 0.347005 Objective Loss 0.347005 LR 0.001000 Time 0.019812 -2022-12-06 10:29:11,458 - Epoch: [15][ 1140/ 1200] Overall Loss 0.346824 Objective Loss 0.346824 LR 0.001000 Time 0.019805 -2022-12-06 10:29:11,650 - Epoch: [15][ 1150/ 1200] Overall Loss 0.346700 Objective Loss 0.346700 LR 0.001000 Time 0.019800 -2022-12-06 10:29:11,842 - Epoch: [15][ 1160/ 1200] Overall Loss 0.346972 Objective Loss 0.346972 LR 0.001000 Time 0.019794 -2022-12-06 10:29:12,034 - Epoch: [15][ 1170/ 1200] Overall Loss 0.346784 Objective Loss 0.346784 LR 0.001000 Time 0.019789 -2022-12-06 10:29:12,226 - Epoch: [15][ 1180/ 1200] Overall Loss 0.346757 Objective Loss 0.346757 LR 0.001000 Time 0.019783 -2022-12-06 10:29:12,418 - Epoch: [15][ 1190/ 1200] Overall Loss 0.346907 Objective Loss 0.346907 LR 0.001000 Time 0.019778 -2022-12-06 10:29:12,648 - Epoch: [15][ 1200/ 1200] Overall Loss 0.346868 Objective Loss 0.346868 Top1 82.008368 Top5 97.280335 LR 0.001000 Time 0.019804 -2022-12-06 10:29:12,735 - --- validate (epoch=15)----------- -2022-12-06 10:29:12,736 - 34129 samples (256 per mini-batch) -2022-12-06 10:29:13,180 - Epoch: [15][ 10/ 134] Loss 0.329314 Top1 82.929688 Top5 97.460938 -2022-12-06 10:29:13,316 - Epoch: [15][ 20/ 134] Loss 0.339747 Top1 82.890625 Top5 97.265625 -2022-12-06 10:29:13,448 - Epoch: [15][ 30/ 134] Loss 0.332138 Top1 82.734375 Top5 97.460938 -2022-12-06 10:29:13,573 - Epoch: [15][ 40/ 134] Loss 0.337841 Top1 82.480469 Top5 97.412109 -2022-12-06 10:29:13,696 - Epoch: [15][ 50/ 134] Loss 0.345458 Top1 82.257812 Top5 97.398438 -2022-12-06 10:29:13,822 - Epoch: [15][ 60/ 134] Loss 0.340751 Top1 82.200521 Top5 97.389323 -2022-12-06 10:29:13,946 - Epoch: [15][ 70/ 134] Loss 0.341480 Top1 82.120536 Top5 97.410714 -2022-12-06 10:29:14,070 - Epoch: [15][ 80/ 134] Loss 0.344848 Top1 82.109375 Top5 97.363281 -2022-12-06 10:29:14,195 - Epoch: [15][ 90/ 134] Loss 0.342715 Top1 82.052951 Top5 97.408854 -2022-12-06 10:29:14,318 - Epoch: [15][ 100/ 134] Loss 0.340744 Top1 82.105469 Top5 97.464844 -2022-12-06 10:29:14,443 - Epoch: [15][ 110/ 134] Loss 0.337840 Top1 82.088068 Top5 97.471591 -2022-12-06 10:29:14,568 - Epoch: [15][ 120/ 134] Loss 0.337049 Top1 82.106120 Top5 97.513021 -2022-12-06 10:29:14,695 - Epoch: [15][ 130/ 134] Loss 0.335486 Top1 82.142428 Top5 97.527043 -2022-12-06 10:29:14,731 - Epoch: [15][ 134/ 134] Loss 0.336775 Top1 82.123707 Top5 97.503589 -2022-12-06 10:29:14,818 - ==> Top1: 82.124 Top5: 97.504 Loss: 0.337 - -2022-12-06 10:29:14,819 - ==> Confusion: -[[ 836 5 4 3 7 6 0 4 5 104 0 4 2 1 4 3 0 0 2 1 5] - [ 0 937 1 3 12 18 2 16 2 2 6 3 1 0 3 3 5 0 4 2 7] - [ 6 8 987 13 2 6 24 16 0 4 5 5 2 2 0 5 2 1 4 5 6] - [ 1 4 38 907 0 3 0 2 1 1 12 2 2 0 20 1 1 3 18 0 4] - [ 7 7 1 1 950 6 1 2 0 11 1 3 0 4 9 6 3 1 0 1 6] - [ 0 49 1 1 11 911 4 33 3 2 0 10 1 22 3 0 1 0 2 6 9] - [ 0 9 11 2 0 2 1060 12 0 1 3 4 0 2 0 6 1 0 0 5 0] - [ 0 12 5 1 2 29 4 955 0 0 1 4 0 1 1 3 0 1 23 10 2] - [ 3 9 0 0 1 2 0 1 979 42 7 0 1 8 5 1 1 0 2 1 1] - [ 35 0 2 1 3 4 0 3 33 904 0 1 0 9 2 1 0 0 0 0 3] - [ 0 5 3 6 0 0 1 7 15 3 947 2 1 14 2 1 1 0 5 1 5] - [ 3 2 5 0 0 13 4 8 1 1 2 951 16 8 0 6 5 11 0 13 2] - [ 0 3 1 5 1 2 2 2 1 0 1 66 834 3 1 11 1 19 1 6 9] - [ 1 6 1 1 1 14 0 0 10 21 4 6 1 943 1 2 1 1 0 1 8] - [ 4 8 3 16 6 3 0 1 26 5 1 0 3 3 1031 1 1 2 4 1 11] - [ 1 6 3 2 2 2 7 0 0 1 0 9 5 5 0 966 10 13 0 4 7] - [ 2 8 0 4 9 2 2 1 1 0 0 6 1 3 1 5 1007 2 0 6 12] - [ 1 3 1 6 0 1 2 0 1 2 1 12 14 1 3 7 3 969 2 3 4] - [ 1 11 7 15 2 4 0 37 5 1 9 3 2 1 3 1 2 0 895 4 5] - [ 0 7 3 1 1 10 12 14 0 0 0 18 5 5 0 3 5 4 3 984 5] - [ 142 383 253 137 217 180 80 219 106 169 197 154 348 442 185 131 180 86 195 349 9073]] - -2022-12-06 10:29:15,474 - ==> Best [Top1: 82.672 Top5: 97.594 Sparsity:0.00 Params: 5376 on epoch: 14] -2022-12-06 10:29:15,475 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:29:15,480 - - -2022-12-06 10:29:15,481 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:29:16,401 - Epoch: [16][ 10/ 1200] Overall Loss 0.299767 Objective Loss 0.299767 LR 0.001000 Time 0.091974 -2022-12-06 10:29:16,604 - Epoch: [16][ 20/ 1200] Overall Loss 0.324813 Objective Loss 0.324813 LR 0.001000 Time 0.056106 -2022-12-06 10:29:16,796 - Epoch: [16][ 30/ 1200] Overall Loss 0.328281 Objective Loss 0.328281 LR 0.001000 Time 0.043772 -2022-12-06 10:29:16,986 - Epoch: [16][ 40/ 1200] Overall Loss 0.328205 Objective Loss 0.328205 LR 0.001000 Time 0.037571 -2022-12-06 10:29:17,176 - Epoch: [16][ 50/ 1200] Overall Loss 0.328339 Objective Loss 0.328339 LR 0.001000 Time 0.033858 -2022-12-06 10:29:17,367 - Epoch: [16][ 60/ 1200] Overall Loss 0.326309 Objective Loss 0.326309 LR 0.001000 Time 0.031388 -2022-12-06 10:29:17,558 - Epoch: [16][ 70/ 1200] Overall Loss 0.323992 Objective Loss 0.323992 LR 0.001000 Time 0.029620 -2022-12-06 10:29:17,748 - Epoch: [16][ 80/ 1200] Overall Loss 0.324168 Objective Loss 0.324168 LR 0.001000 Time 0.028283 -2022-12-06 10:29:17,938 - Epoch: [16][ 90/ 1200] Overall Loss 0.325880 Objective Loss 0.325880 LR 0.001000 Time 0.027252 -2022-12-06 10:29:18,129 - Epoch: [16][ 100/ 1200] Overall Loss 0.323860 Objective Loss 0.323860 LR 0.001000 Time 0.026424 -2022-12-06 10:29:18,319 - Epoch: [16][ 110/ 1200] Overall Loss 0.327511 Objective Loss 0.327511 LR 0.001000 Time 0.025753 -2022-12-06 10:29:18,510 - Epoch: [16][ 120/ 1200] Overall Loss 0.326849 Objective Loss 0.326849 LR 0.001000 Time 0.025188 -2022-12-06 10:29:18,701 - Epoch: [16][ 130/ 1200] Overall Loss 0.326784 Objective Loss 0.326784 LR 0.001000 Time 0.024715 -2022-12-06 10:29:18,892 - Epoch: [16][ 140/ 1200] Overall Loss 0.325861 Objective Loss 0.325861 LR 0.001000 Time 0.024312 -2022-12-06 10:29:19,082 - Epoch: [16][ 150/ 1200] Overall Loss 0.326183 Objective Loss 0.326183 LR 0.001000 Time 0.023956 -2022-12-06 10:29:19,273 - Epoch: [16][ 160/ 1200] Overall Loss 0.326153 Objective Loss 0.326153 LR 0.001000 Time 0.023647 -2022-12-06 10:29:19,463 - Epoch: [16][ 170/ 1200] Overall Loss 0.326598 Objective Loss 0.326598 LR 0.001000 Time 0.023374 -2022-12-06 10:29:19,654 - Epoch: [16][ 180/ 1200] Overall Loss 0.325148 Objective Loss 0.325148 LR 0.001000 Time 0.023129 -2022-12-06 10:29:19,844 - Epoch: [16][ 190/ 1200] Overall Loss 0.324931 Objective Loss 0.324931 LR 0.001000 Time 0.022911 -2022-12-06 10:29:20,035 - Epoch: [16][ 200/ 1200] Overall Loss 0.325795 Objective Loss 0.325795 LR 0.001000 Time 0.022717 -2022-12-06 10:29:20,226 - Epoch: [16][ 210/ 1200] Overall Loss 0.325372 Objective Loss 0.325372 LR 0.001000 Time 0.022541 -2022-12-06 10:29:20,417 - Epoch: [16][ 220/ 1200] Overall Loss 0.324483 Objective Loss 0.324483 LR 0.001000 Time 0.022382 -2022-12-06 10:29:20,607 - Epoch: [16][ 230/ 1200] Overall Loss 0.323459 Objective Loss 0.323459 LR 0.001000 Time 0.022235 -2022-12-06 10:29:20,798 - Epoch: [16][ 240/ 1200] Overall Loss 0.323121 Objective Loss 0.323121 LR 0.001000 Time 0.022100 -2022-12-06 10:29:20,988 - Epoch: [16][ 250/ 1200] Overall Loss 0.324328 Objective Loss 0.324328 LR 0.001000 Time 0.021973 -2022-12-06 10:29:21,178 - Epoch: [16][ 260/ 1200] Overall Loss 0.324558 Objective Loss 0.324558 LR 0.001000 Time 0.021858 -2022-12-06 10:29:21,369 - Epoch: [16][ 270/ 1200] Overall Loss 0.327502 Objective Loss 0.327502 LR 0.001000 Time 0.021752 -2022-12-06 10:29:21,559 - Epoch: [16][ 280/ 1200] Overall Loss 0.329437 Objective Loss 0.329437 LR 0.001000 Time 0.021653 -2022-12-06 10:29:21,750 - Epoch: [16][ 290/ 1200] Overall Loss 0.329634 Objective Loss 0.329634 LR 0.001000 Time 0.021565 -2022-12-06 10:29:21,942 - Epoch: [16][ 300/ 1200] Overall Loss 0.330466 Objective Loss 0.330466 LR 0.001000 Time 0.021483 -2022-12-06 10:29:22,133 - Epoch: [16][ 310/ 1200] Overall Loss 0.330377 Objective Loss 0.330377 LR 0.001000 Time 0.021405 -2022-12-06 10:29:22,324 - Epoch: [16][ 320/ 1200] Overall Loss 0.329880 Objective Loss 0.329880 LR 0.001000 Time 0.021331 -2022-12-06 10:29:22,514 - Epoch: [16][ 330/ 1200] Overall Loss 0.329934 Objective Loss 0.329934 LR 0.001000 Time 0.021259 -2022-12-06 10:29:22,705 - Epoch: [16][ 340/ 1200] Overall Loss 0.329703 Objective Loss 0.329703 LR 0.001000 Time 0.021193 -2022-12-06 10:29:22,895 - Epoch: [16][ 350/ 1200] Overall Loss 0.329679 Objective Loss 0.329679 LR 0.001000 Time 0.021130 -2022-12-06 10:29:23,086 - Epoch: [16][ 360/ 1200] Overall Loss 0.329667 Objective Loss 0.329667 LR 0.001000 Time 0.021072 -2022-12-06 10:29:23,277 - Epoch: [16][ 370/ 1200] Overall Loss 0.329211 Objective Loss 0.329211 LR 0.001000 Time 0.021018 -2022-12-06 10:29:23,469 - Epoch: [16][ 380/ 1200] Overall Loss 0.329278 Objective Loss 0.329278 LR 0.001000 Time 0.020966 -2022-12-06 10:29:23,659 - Epoch: [16][ 390/ 1200] Overall Loss 0.329267 Objective Loss 0.329267 LR 0.001000 Time 0.020916 -2022-12-06 10:29:23,850 - Epoch: [16][ 400/ 1200] Overall Loss 0.329369 Objective Loss 0.329369 LR 0.001000 Time 0.020868 -2022-12-06 10:29:24,041 - Epoch: [16][ 410/ 1200] Overall Loss 0.329346 Objective Loss 0.329346 LR 0.001000 Time 0.020824 -2022-12-06 10:29:24,232 - Epoch: [16][ 420/ 1200] Overall Loss 0.328905 Objective Loss 0.328905 LR 0.001000 Time 0.020781 -2022-12-06 10:29:24,422 - Epoch: [16][ 430/ 1200] Overall Loss 0.329583 Objective Loss 0.329583 LR 0.001000 Time 0.020740 -2022-12-06 10:29:24,613 - Epoch: [16][ 440/ 1200] Overall Loss 0.330005 Objective Loss 0.330005 LR 0.001000 Time 0.020701 -2022-12-06 10:29:24,804 - Epoch: [16][ 450/ 1200] Overall Loss 0.330599 Objective Loss 0.330599 LR 0.001000 Time 0.020663 -2022-12-06 10:29:24,995 - Epoch: [16][ 460/ 1200] Overall Loss 0.330370 Objective Loss 0.330370 LR 0.001000 Time 0.020628 -2022-12-06 10:29:25,185 - Epoch: [16][ 470/ 1200] Overall Loss 0.331671 Objective Loss 0.331671 LR 0.001000 Time 0.020594 -2022-12-06 10:29:25,376 - Epoch: [16][ 480/ 1200] Overall Loss 0.332415 Objective Loss 0.332415 LR 0.001000 Time 0.020561 -2022-12-06 10:29:25,567 - Epoch: [16][ 490/ 1200] Overall Loss 0.332606 Objective Loss 0.332606 LR 0.001000 Time 0.020529 -2022-12-06 10:29:25,757 - Epoch: [16][ 500/ 1200] Overall Loss 0.332866 Objective Loss 0.332866 LR 0.001000 Time 0.020498 -2022-12-06 10:29:25,948 - Epoch: [16][ 510/ 1200] Overall Loss 0.332714 Objective Loss 0.332714 LR 0.001000 Time 0.020470 -2022-12-06 10:29:26,139 - Epoch: [16][ 520/ 1200] Overall Loss 0.333274 Objective Loss 0.333274 LR 0.001000 Time 0.020442 -2022-12-06 10:29:26,330 - Epoch: [16][ 530/ 1200] Overall Loss 0.333217 Objective Loss 0.333217 LR 0.001000 Time 0.020415 -2022-12-06 10:29:26,520 - Epoch: [16][ 540/ 1200] Overall Loss 0.333739 Objective Loss 0.333739 LR 0.001000 Time 0.020388 -2022-12-06 10:29:26,710 - Epoch: [16][ 550/ 1200] Overall Loss 0.334158 Objective Loss 0.334158 LR 0.001000 Time 0.020362 -2022-12-06 10:29:26,901 - Epoch: [16][ 560/ 1200] Overall Loss 0.333856 Objective Loss 0.333856 LR 0.001000 Time 0.020339 -2022-12-06 10:29:27,091 - Epoch: [16][ 570/ 1200] Overall Loss 0.334278 Objective Loss 0.334278 LR 0.001000 Time 0.020315 -2022-12-06 10:29:27,282 - Epoch: [16][ 580/ 1200] Overall Loss 0.334186 Objective Loss 0.334186 LR 0.001000 Time 0.020292 -2022-12-06 10:29:27,472 - Epoch: [16][ 590/ 1200] Overall Loss 0.334410 Objective Loss 0.334410 LR 0.001000 Time 0.020269 -2022-12-06 10:29:27,662 - Epoch: [16][ 600/ 1200] Overall Loss 0.334288 Objective Loss 0.334288 LR 0.001000 Time 0.020248 -2022-12-06 10:29:27,853 - Epoch: [16][ 610/ 1200] Overall Loss 0.333361 Objective Loss 0.333361 LR 0.001000 Time 0.020228 -2022-12-06 10:29:28,044 - Epoch: [16][ 620/ 1200] Overall Loss 0.333433 Objective Loss 0.333433 LR 0.001000 Time 0.020210 -2022-12-06 10:29:28,236 - Epoch: [16][ 630/ 1200] Overall Loss 0.333398 Objective Loss 0.333398 LR 0.001000 Time 0.020192 -2022-12-06 10:29:28,427 - Epoch: [16][ 640/ 1200] Overall Loss 0.333335 Objective Loss 0.333335 LR 0.001000 Time 0.020174 -2022-12-06 10:29:28,618 - Epoch: [16][ 650/ 1200] Overall Loss 0.333020 Objective Loss 0.333020 LR 0.001000 Time 0.020157 -2022-12-06 10:29:28,809 - Epoch: [16][ 660/ 1200] Overall Loss 0.332952 Objective Loss 0.332952 LR 0.001000 Time 0.020140 -2022-12-06 10:29:28,999 - Epoch: [16][ 670/ 1200] Overall Loss 0.332824 Objective Loss 0.332824 LR 0.001000 Time 0.020123 -2022-12-06 10:29:29,190 - Epoch: [16][ 680/ 1200] Overall Loss 0.332759 Objective Loss 0.332759 LR 0.001000 Time 0.020107 -2022-12-06 10:29:29,381 - Epoch: [16][ 690/ 1200] Overall Loss 0.333078 Objective Loss 0.333078 LR 0.001000 Time 0.020091 -2022-12-06 10:29:29,572 - Epoch: [16][ 700/ 1200] Overall Loss 0.333450 Objective Loss 0.333450 LR 0.001000 Time 0.020076 -2022-12-06 10:29:29,762 - Epoch: [16][ 710/ 1200] Overall Loss 0.333657 Objective Loss 0.333657 LR 0.001000 Time 0.020061 -2022-12-06 10:29:29,953 - Epoch: [16][ 720/ 1200] Overall Loss 0.333198 Objective Loss 0.333198 LR 0.001000 Time 0.020046 -2022-12-06 10:29:30,145 - Epoch: [16][ 730/ 1200] Overall Loss 0.333005 Objective Loss 0.333005 LR 0.001000 Time 0.020033 -2022-12-06 10:29:30,335 - Epoch: [16][ 740/ 1200] Overall Loss 0.333287 Objective Loss 0.333287 LR 0.001000 Time 0.020020 -2022-12-06 10:29:30,527 - Epoch: [16][ 750/ 1200] Overall Loss 0.333667 Objective Loss 0.333667 LR 0.001000 Time 0.020007 -2022-12-06 10:29:30,718 - Epoch: [16][ 760/ 1200] Overall Loss 0.334340 Objective Loss 0.334340 LR 0.001000 Time 0.019994 -2022-12-06 10:29:30,908 - Epoch: [16][ 770/ 1200] Overall Loss 0.334161 Objective Loss 0.334161 LR 0.001000 Time 0.019982 -2022-12-06 10:29:31,099 - Epoch: [16][ 780/ 1200] Overall Loss 0.334613 Objective Loss 0.334613 LR 0.001000 Time 0.019969 -2022-12-06 10:29:31,291 - Epoch: [16][ 790/ 1200] Overall Loss 0.334972 Objective Loss 0.334972 LR 0.001000 Time 0.019958 -2022-12-06 10:29:31,482 - Epoch: [16][ 800/ 1200] Overall Loss 0.335059 Objective Loss 0.335059 LR 0.001000 Time 0.019947 -2022-12-06 10:29:31,673 - Epoch: [16][ 810/ 1200] Overall Loss 0.335228 Objective Loss 0.335228 LR 0.001000 Time 0.019937 -2022-12-06 10:29:31,864 - Epoch: [16][ 820/ 1200] Overall Loss 0.335147 Objective Loss 0.335147 LR 0.001000 Time 0.019926 -2022-12-06 10:29:32,055 - Epoch: [16][ 830/ 1200] Overall Loss 0.335590 Objective Loss 0.335590 LR 0.001000 Time 0.019915 -2022-12-06 10:29:32,248 - Epoch: [16][ 840/ 1200] Overall Loss 0.335684 Objective Loss 0.335684 LR 0.001000 Time 0.019907 -2022-12-06 10:29:32,441 - Epoch: [16][ 850/ 1200] Overall Loss 0.336018 Objective Loss 0.336018 LR 0.001000 Time 0.019899 -2022-12-06 10:29:32,634 - Epoch: [16][ 860/ 1200] Overall Loss 0.336193 Objective Loss 0.336193 LR 0.001000 Time 0.019891 -2022-12-06 10:29:32,826 - Epoch: [16][ 870/ 1200] Overall Loss 0.336660 Objective Loss 0.336660 LR 0.001000 Time 0.019883 -2022-12-06 10:29:33,019 - Epoch: [16][ 880/ 1200] Overall Loss 0.336808 Objective Loss 0.336808 LR 0.001000 Time 0.019876 -2022-12-06 10:29:33,212 - Epoch: [16][ 890/ 1200] Overall Loss 0.336590 Objective Loss 0.336590 LR 0.001000 Time 0.019869 -2022-12-06 10:29:33,405 - Epoch: [16][ 900/ 1200] Overall Loss 0.337117 Objective Loss 0.337117 LR 0.001000 Time 0.019861 -2022-12-06 10:29:33,597 - Epoch: [16][ 910/ 1200] Overall Loss 0.337333 Objective Loss 0.337333 LR 0.001000 Time 0.019854 -2022-12-06 10:29:33,790 - Epoch: [16][ 920/ 1200] Overall Loss 0.337159 Objective Loss 0.337159 LR 0.001000 Time 0.019848 -2022-12-06 10:29:33,983 - Epoch: [16][ 930/ 1200] Overall Loss 0.337252 Objective Loss 0.337252 LR 0.001000 Time 0.019841 -2022-12-06 10:29:34,177 - Epoch: [16][ 940/ 1200] Overall Loss 0.337397 Objective Loss 0.337397 LR 0.001000 Time 0.019835 -2022-12-06 10:29:34,369 - Epoch: [16][ 950/ 1200] Overall Loss 0.337283 Objective Loss 0.337283 LR 0.001000 Time 0.019828 -2022-12-06 10:29:34,562 - Epoch: [16][ 960/ 1200] Overall Loss 0.337655 Objective Loss 0.337655 LR 0.001000 Time 0.019822 -2022-12-06 10:29:34,755 - Epoch: [16][ 970/ 1200] Overall Loss 0.337857 Objective Loss 0.337857 LR 0.001000 Time 0.019817 -2022-12-06 10:29:34,948 - Epoch: [16][ 980/ 1200] Overall Loss 0.337598 Objective Loss 0.337598 LR 0.001000 Time 0.019810 -2022-12-06 10:29:35,141 - Epoch: [16][ 990/ 1200] Overall Loss 0.337440 Objective Loss 0.337440 LR 0.001000 Time 0.019804 -2022-12-06 10:29:35,334 - Epoch: [16][ 1000/ 1200] Overall Loss 0.337189 Objective Loss 0.337189 LR 0.001000 Time 0.019799 -2022-12-06 10:29:35,527 - Epoch: [16][ 1010/ 1200] Overall Loss 0.337773 Objective Loss 0.337773 LR 0.001000 Time 0.019794 -2022-12-06 10:29:35,720 - Epoch: [16][ 1020/ 1200] Overall Loss 0.337769 Objective Loss 0.337769 LR 0.001000 Time 0.019788 -2022-12-06 10:29:35,913 - Epoch: [16][ 1030/ 1200] Overall Loss 0.338171 Objective Loss 0.338171 LR 0.001000 Time 0.019783 -2022-12-06 10:29:36,107 - Epoch: [16][ 1040/ 1200] Overall Loss 0.338173 Objective Loss 0.338173 LR 0.001000 Time 0.019779 -2022-12-06 10:29:36,300 - Epoch: [16][ 1050/ 1200] Overall Loss 0.338155 Objective Loss 0.338155 LR 0.001000 Time 0.019774 -2022-12-06 10:29:36,493 - Epoch: [16][ 1060/ 1200] Overall Loss 0.338278 Objective Loss 0.338278 LR 0.001000 Time 0.019769 -2022-12-06 10:29:36,686 - Epoch: [16][ 1070/ 1200] Overall Loss 0.338408 Objective Loss 0.338408 LR 0.001000 Time 0.019764 -2022-12-06 10:29:36,878 - Epoch: [16][ 1080/ 1200] Overall Loss 0.338561 Objective Loss 0.338561 LR 0.001000 Time 0.019759 -2022-12-06 10:29:37,071 - Epoch: [16][ 1090/ 1200] Overall Loss 0.338663 Objective Loss 0.338663 LR 0.001000 Time 0.019754 -2022-12-06 10:29:37,264 - Epoch: [16][ 1100/ 1200] Overall Loss 0.338630 Objective Loss 0.338630 LR 0.001000 Time 0.019749 -2022-12-06 10:29:37,457 - Epoch: [16][ 1110/ 1200] Overall Loss 0.338178 Objective Loss 0.338178 LR 0.001000 Time 0.019744 -2022-12-06 10:29:37,649 - Epoch: [16][ 1120/ 1200] Overall Loss 0.338499 Objective Loss 0.338499 LR 0.001000 Time 0.019739 -2022-12-06 10:29:37,841 - Epoch: [16][ 1130/ 1200] Overall Loss 0.338668 Objective Loss 0.338668 LR 0.001000 Time 0.019734 -2022-12-06 10:29:38,035 - Epoch: [16][ 1140/ 1200] Overall Loss 0.338454 Objective Loss 0.338454 LR 0.001000 Time 0.019730 -2022-12-06 10:29:38,227 - Epoch: [16][ 1150/ 1200] Overall Loss 0.338446 Objective Loss 0.338446 LR 0.001000 Time 0.019726 -2022-12-06 10:29:38,421 - Epoch: [16][ 1160/ 1200] Overall Loss 0.338555 Objective Loss 0.338555 LR 0.001000 Time 0.019722 -2022-12-06 10:29:38,613 - Epoch: [16][ 1170/ 1200] Overall Loss 0.338908 Objective Loss 0.338908 LR 0.001000 Time 0.019717 -2022-12-06 10:29:38,806 - Epoch: [16][ 1180/ 1200] Overall Loss 0.338784 Objective Loss 0.338784 LR 0.001000 Time 0.019713 -2022-12-06 10:29:38,999 - Epoch: [16][ 1190/ 1200] Overall Loss 0.338650 Objective Loss 0.338650 LR 0.001000 Time 0.019709 -2022-12-06 10:29:39,230 - Epoch: [16][ 1200/ 1200] Overall Loss 0.338379 Objective Loss 0.338379 Top1 82.635983 Top5 98.117155 LR 0.001000 Time 0.019737 -2022-12-06 10:29:39,318 - --- validate (epoch=16)----------- -2022-12-06 10:29:39,319 - 34129 samples (256 per mini-batch) -2022-12-06 10:29:39,760 - Epoch: [16][ 10/ 134] Loss 0.316538 Top1 82.265625 Top5 97.578125 -2022-12-06 10:29:39,888 - Epoch: [16][ 20/ 134] Loss 0.337981 Top1 82.363281 Top5 97.500000 -2022-12-06 10:29:40,014 - Epoch: [16][ 30/ 134] Loss 0.342231 Top1 82.330729 Top5 97.343750 -2022-12-06 10:29:40,142 - Epoch: [16][ 40/ 134] Loss 0.343930 Top1 82.050781 Top5 97.314453 -2022-12-06 10:29:40,283 - Epoch: [16][ 50/ 134] Loss 0.341919 Top1 82.007812 Top5 97.429688 -2022-12-06 10:29:40,425 - Epoch: [16][ 60/ 134] Loss 0.338711 Top1 81.927083 Top5 97.434896 -2022-12-06 10:29:40,557 - Epoch: [16][ 70/ 134] Loss 0.337772 Top1 82.020089 Top5 97.483259 -2022-12-06 10:29:40,686 - Epoch: [16][ 80/ 134] Loss 0.336533 Top1 81.962891 Top5 97.500000 -2022-12-06 10:29:40,833 - Epoch: [16][ 90/ 134] Loss 0.338846 Top1 81.966146 Top5 97.482639 -2022-12-06 10:29:40,974 - Epoch: [16][ 100/ 134] Loss 0.339062 Top1 81.976562 Top5 97.496094 -2022-12-06 10:29:41,122 - Epoch: [16][ 110/ 134] Loss 0.339450 Top1 81.946023 Top5 97.471591 -2022-12-06 10:29:41,263 - Epoch: [16][ 120/ 134] Loss 0.335605 Top1 82.011719 Top5 97.503255 -2022-12-06 10:29:41,404 - Epoch: [16][ 130/ 134] Loss 0.337801 Top1 81.956130 Top5 97.457933 -2022-12-06 10:29:41,441 - Epoch: [16][ 134/ 134] Loss 0.339596 Top1 81.933253 Top5 97.447918 -2022-12-06 10:29:41,529 - ==> Top1: 81.933 Top5: 97.448 Loss: 0.340 - -2022-12-06 10:29:41,529 - ==> Confusion: -[[ 889 6 3 1 7 7 0 0 6 53 0 7 2 1 7 1 1 1 1 0 3] - [ 0 932 3 3 9 35 4 13 0 0 1 1 4 0 4 3 4 0 6 0 5] - [ 9 7 971 19 5 4 27 15 0 0 2 5 3 2 2 11 3 1 4 3 10] - [ 3 2 19 921 2 4 0 1 0 0 10 1 8 1 26 3 2 2 11 0 4] - [ 14 9 0 2 935 15 1 0 0 1 0 4 1 1 11 8 10 2 0 1 5] - [ 2 29 1 1 4 955 4 10 4 0 0 18 5 9 3 1 4 1 1 11 6] - [ 0 8 13 1 1 3 1049 5 0 0 3 5 1 0 0 12 0 3 0 9 5] - [ 1 21 9 1 1 46 5 917 0 0 0 5 2 1 2 3 0 0 26 9 5] - [ 5 6 2 0 1 5 0 1 951 38 11 0 6 15 12 2 2 0 0 2 5] - [ 81 2 3 0 12 5 0 4 26 821 1 1 2 22 7 1 0 2 0 2 9] - [ 1 5 9 11 0 7 3 3 6 2 933 1 4 11 10 0 2 0 6 1 4] - [ 6 3 3 0 1 9 1 5 1 0 1 959 38 3 0 3 3 4 0 8 3] - [ 2 0 0 1 1 4 0 2 0 0 1 43 885 0 0 9 1 9 2 2 7] - [ 0 4 2 1 4 20 0 1 4 7 10 10 5 918 1 7 8 1 0 12 8] - [ 5 6 1 4 5 3 1 1 11 0 1 4 3 2 1061 0 2 3 4 1 12] - [ 0 3 2 2 0 3 1 1 0 0 0 13 9 0 0 979 12 7 1 6 4] - [ 2 6 0 3 3 1 4 1 1 0 0 2 4 0 3 11 1019 2 0 4 6] - [ 0 1 3 2 0 0 4 0 0 0 1 12 33 3 1 18 1 947 0 4 6] - [ 4 9 5 14 0 5 0 31 4 0 3 5 6 1 16 1 0 0 894 2 8] - [ 0 6 4 1 0 8 9 10 0 0 0 26 7 2 1 5 6 3 1 989 2] - [ 171 355 215 120 125 284 77 164 69 80 156 176 515 320 209 237 281 72 189 379 9032]] - -2022-12-06 10:29:42,184 - ==> Best [Top1: 82.672 Top5: 97.594 Sparsity:0.00 Params: 5376 on epoch: 14] -2022-12-06 10:29:42,184 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:29:42,190 - - -2022-12-06 10:29:42,190 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:29:43,123 - Epoch: [17][ 10/ 1200] Overall Loss 0.345534 Objective Loss 0.345534 LR 0.001000 Time 0.093213 -2022-12-06 10:29:43,325 - Epoch: [17][ 20/ 1200] Overall Loss 0.316213 Objective Loss 0.316213 LR 0.001000 Time 0.056691 -2022-12-06 10:29:43,524 - Epoch: [17][ 30/ 1200] Overall Loss 0.306010 Objective Loss 0.306010 LR 0.001000 Time 0.044386 -2022-12-06 10:29:43,720 - Epoch: [17][ 40/ 1200] Overall Loss 0.306784 Objective Loss 0.306784 LR 0.001000 Time 0.038176 -2022-12-06 10:29:43,918 - Epoch: [17][ 50/ 1200] Overall Loss 0.314237 Objective Loss 0.314237 LR 0.001000 Time 0.034487 -2022-12-06 10:29:44,113 - Epoch: [17][ 60/ 1200] Overall Loss 0.315081 Objective Loss 0.315081 LR 0.001000 Time 0.031986 -2022-12-06 10:29:44,310 - Epoch: [17][ 70/ 1200] Overall Loss 0.317803 Objective Loss 0.317803 LR 0.001000 Time 0.030228 -2022-12-06 10:29:44,506 - Epoch: [17][ 80/ 1200] Overall Loss 0.318742 Objective Loss 0.318742 LR 0.001000 Time 0.028890 -2022-12-06 10:29:44,703 - Epoch: [17][ 90/ 1200] Overall Loss 0.318355 Objective Loss 0.318355 LR 0.001000 Time 0.027866 -2022-12-06 10:29:44,898 - Epoch: [17][ 100/ 1200] Overall Loss 0.321073 Objective Loss 0.321073 LR 0.001000 Time 0.027019 -2022-12-06 10:29:45,095 - Epoch: [17][ 110/ 1200] Overall Loss 0.320520 Objective Loss 0.320520 LR 0.001000 Time 0.026350 -2022-12-06 10:29:45,290 - Epoch: [17][ 120/ 1200] Overall Loss 0.321136 Objective Loss 0.321136 LR 0.001000 Time 0.025778 -2022-12-06 10:29:45,488 - Epoch: [17][ 130/ 1200] Overall Loss 0.320503 Objective Loss 0.320503 LR 0.001000 Time 0.025315 -2022-12-06 10:29:45,684 - Epoch: [17][ 140/ 1200] Overall Loss 0.321992 Objective Loss 0.321992 LR 0.001000 Time 0.024900 -2022-12-06 10:29:45,881 - Epoch: [17][ 150/ 1200] Overall Loss 0.321817 Objective Loss 0.321817 LR 0.001000 Time 0.024549 -2022-12-06 10:29:46,076 - Epoch: [17][ 160/ 1200] Overall Loss 0.321389 Objective Loss 0.321389 LR 0.001000 Time 0.024231 -2022-12-06 10:29:46,273 - Epoch: [17][ 170/ 1200] Overall Loss 0.320183 Objective Loss 0.320183 LR 0.001000 Time 0.023963 -2022-12-06 10:29:46,468 - Epoch: [17][ 180/ 1200] Overall Loss 0.320237 Objective Loss 0.320237 LR 0.001000 Time 0.023713 -2022-12-06 10:29:46,666 - Epoch: [17][ 190/ 1200] Overall Loss 0.320249 Objective Loss 0.320249 LR 0.001000 Time 0.023502 -2022-12-06 10:29:46,861 - Epoch: [17][ 200/ 1200] Overall Loss 0.321282 Objective Loss 0.321282 LR 0.001000 Time 0.023301 -2022-12-06 10:29:47,059 - Epoch: [17][ 210/ 1200] Overall Loss 0.322284 Objective Loss 0.322284 LR 0.001000 Time 0.023129 -2022-12-06 10:29:47,254 - Epoch: [17][ 220/ 1200] Overall Loss 0.323094 Objective Loss 0.323094 LR 0.001000 Time 0.022962 -2022-12-06 10:29:47,451 - Epoch: [17][ 230/ 1200] Overall Loss 0.323455 Objective Loss 0.323455 LR 0.001000 Time 0.022820 -2022-12-06 10:29:47,647 - Epoch: [17][ 240/ 1200] Overall Loss 0.323227 Objective Loss 0.323227 LR 0.001000 Time 0.022680 -2022-12-06 10:29:47,844 - Epoch: [17][ 250/ 1200] Overall Loss 0.323224 Objective Loss 0.323224 LR 0.001000 Time 0.022560 -2022-12-06 10:29:48,040 - Epoch: [17][ 260/ 1200] Overall Loss 0.323379 Objective Loss 0.323379 LR 0.001000 Time 0.022444 -2022-12-06 10:29:48,237 - Epoch: [17][ 270/ 1200] Overall Loss 0.324734 Objective Loss 0.324734 LR 0.001000 Time 0.022341 -2022-12-06 10:29:48,432 - Epoch: [17][ 280/ 1200] Overall Loss 0.325575 Objective Loss 0.325575 LR 0.001000 Time 0.022238 -2022-12-06 10:29:48,630 - Epoch: [17][ 290/ 1200] Overall Loss 0.326244 Objective Loss 0.326244 LR 0.001000 Time 0.022152 -2022-12-06 10:29:48,825 - Epoch: [17][ 300/ 1200] Overall Loss 0.325667 Objective Loss 0.325667 LR 0.001000 Time 0.022063 -2022-12-06 10:29:49,023 - Epoch: [17][ 310/ 1200] Overall Loss 0.325787 Objective Loss 0.325787 LR 0.001000 Time 0.021986 -2022-12-06 10:29:49,218 - Epoch: [17][ 320/ 1200] Overall Loss 0.324519 Objective Loss 0.324519 LR 0.001000 Time 0.021907 -2022-12-06 10:29:49,416 - Epoch: [17][ 330/ 1200] Overall Loss 0.323669 Objective Loss 0.323669 LR 0.001000 Time 0.021841 -2022-12-06 10:29:49,611 - Epoch: [17][ 340/ 1200] Overall Loss 0.322852 Objective Loss 0.322852 LR 0.001000 Time 0.021771 -2022-12-06 10:29:49,808 - Epoch: [17][ 350/ 1200] Overall Loss 0.322443 Objective Loss 0.322443 LR 0.001000 Time 0.021711 -2022-12-06 10:29:50,004 - Epoch: [17][ 360/ 1200] Overall Loss 0.322754 Objective Loss 0.322754 LR 0.001000 Time 0.021649 -2022-12-06 10:29:50,201 - Epoch: [17][ 370/ 1200] Overall Loss 0.322776 Objective Loss 0.322776 LR 0.001000 Time 0.021596 -2022-12-06 10:29:50,396 - Epoch: [17][ 380/ 1200] Overall Loss 0.323632 Objective Loss 0.323632 LR 0.001000 Time 0.021539 -2022-12-06 10:29:50,593 - Epoch: [17][ 390/ 1200] Overall Loss 0.323896 Objective Loss 0.323896 LR 0.001000 Time 0.021490 -2022-12-06 10:29:50,789 - Epoch: [17][ 400/ 1200] Overall Loss 0.323664 Objective Loss 0.323664 LR 0.001000 Time 0.021442 -2022-12-06 10:29:50,986 - Epoch: [17][ 410/ 1200] Overall Loss 0.323889 Objective Loss 0.323889 LR 0.001000 Time 0.021399 -2022-12-06 10:29:51,182 - Epoch: [17][ 420/ 1200] Overall Loss 0.325053 Objective Loss 0.325053 LR 0.001000 Time 0.021355 -2022-12-06 10:29:51,379 - Epoch: [17][ 430/ 1200] Overall Loss 0.325122 Objective Loss 0.325122 LR 0.001000 Time 0.021315 -2022-12-06 10:29:51,574 - Epoch: [17][ 440/ 1200] Overall Loss 0.324945 Objective Loss 0.324945 LR 0.001000 Time 0.021273 -2022-12-06 10:29:51,772 - Epoch: [17][ 450/ 1200] Overall Loss 0.324496 Objective Loss 0.324496 LR 0.001000 Time 0.021239 -2022-12-06 10:29:51,967 - Epoch: [17][ 460/ 1200] Overall Loss 0.324766 Objective Loss 0.324766 LR 0.001000 Time 0.021200 -2022-12-06 10:29:52,165 - Epoch: [17][ 470/ 1200] Overall Loss 0.324748 Objective Loss 0.324748 LR 0.001000 Time 0.021170 -2022-12-06 10:29:52,361 - Epoch: [17][ 480/ 1200] Overall Loss 0.324522 Objective Loss 0.324522 LR 0.001000 Time 0.021134 -2022-12-06 10:29:52,558 - Epoch: [17][ 490/ 1200] Overall Loss 0.323906 Objective Loss 0.323906 LR 0.001000 Time 0.021105 -2022-12-06 10:29:52,753 - Epoch: [17][ 500/ 1200] Overall Loss 0.324439 Objective Loss 0.324439 LR 0.001000 Time 0.021072 -2022-12-06 10:29:52,950 - Epoch: [17][ 510/ 1200] Overall Loss 0.323628 Objective Loss 0.323628 LR 0.001000 Time 0.021044 -2022-12-06 10:29:53,145 - Epoch: [17][ 520/ 1200] Overall Loss 0.323541 Objective Loss 0.323541 LR 0.001000 Time 0.021013 -2022-12-06 10:29:53,342 - Epoch: [17][ 530/ 1200] Overall Loss 0.324012 Objective Loss 0.324012 LR 0.001000 Time 0.020988 -2022-12-06 10:29:53,538 - Epoch: [17][ 540/ 1200] Overall Loss 0.323696 Objective Loss 0.323696 LR 0.001000 Time 0.020960 -2022-12-06 10:29:53,736 - Epoch: [17][ 550/ 1200] Overall Loss 0.324194 Objective Loss 0.324194 LR 0.001000 Time 0.020937 -2022-12-06 10:29:53,931 - Epoch: [17][ 560/ 1200] Overall Loss 0.323695 Objective Loss 0.323695 LR 0.001000 Time 0.020911 -2022-12-06 10:29:54,128 - Epoch: [17][ 570/ 1200] Overall Loss 0.323770 Objective Loss 0.323770 LR 0.001000 Time 0.020889 -2022-12-06 10:29:54,323 - Epoch: [17][ 580/ 1200] Overall Loss 0.323408 Objective Loss 0.323408 LR 0.001000 Time 0.020864 -2022-12-06 10:29:54,520 - Epoch: [17][ 590/ 1200] Overall Loss 0.323928 Objective Loss 0.323928 LR 0.001000 Time 0.020844 -2022-12-06 10:29:54,715 - Epoch: [17][ 600/ 1200] Overall Loss 0.323910 Objective Loss 0.323910 LR 0.001000 Time 0.020820 -2022-12-06 10:29:54,913 - Epoch: [17][ 610/ 1200] Overall Loss 0.323809 Objective Loss 0.323809 LR 0.001000 Time 0.020802 -2022-12-06 10:29:55,108 - Epoch: [17][ 620/ 1200] Overall Loss 0.324189 Objective Loss 0.324189 LR 0.001000 Time 0.020781 -2022-12-06 10:29:55,305 - Epoch: [17][ 630/ 1200] Overall Loss 0.325090 Objective Loss 0.325090 LR 0.001000 Time 0.020763 -2022-12-06 10:29:55,500 - Epoch: [17][ 640/ 1200] Overall Loss 0.325738 Objective Loss 0.325738 LR 0.001000 Time 0.020743 -2022-12-06 10:29:55,698 - Epoch: [17][ 650/ 1200] Overall Loss 0.325870 Objective Loss 0.325870 LR 0.001000 Time 0.020727 -2022-12-06 10:29:55,893 - Epoch: [17][ 660/ 1200] Overall Loss 0.326702 Objective Loss 0.326702 LR 0.001000 Time 0.020708 -2022-12-06 10:29:56,091 - Epoch: [17][ 670/ 1200] Overall Loss 0.327260 Objective Loss 0.327260 LR 0.001000 Time 0.020693 -2022-12-06 10:29:56,286 - Epoch: [17][ 680/ 1200] Overall Loss 0.327663 Objective Loss 0.327663 LR 0.001000 Time 0.020675 -2022-12-06 10:29:56,484 - Epoch: [17][ 690/ 1200] Overall Loss 0.327575 Objective Loss 0.327575 LR 0.001000 Time 0.020661 -2022-12-06 10:29:56,680 - Epoch: [17][ 700/ 1200] Overall Loss 0.327736 Objective Loss 0.327736 LR 0.001000 Time 0.020645 -2022-12-06 10:29:56,877 - Epoch: [17][ 710/ 1200] Overall Loss 0.328224 Objective Loss 0.328224 LR 0.001000 Time 0.020631 -2022-12-06 10:29:57,073 - Epoch: [17][ 720/ 1200] Overall Loss 0.327997 Objective Loss 0.327997 LR 0.001000 Time 0.020616 -2022-12-06 10:29:57,270 - Epoch: [17][ 730/ 1200] Overall Loss 0.328317 Objective Loss 0.328317 LR 0.001000 Time 0.020603 -2022-12-06 10:29:57,465 - Epoch: [17][ 740/ 1200] Overall Loss 0.328286 Objective Loss 0.328286 LR 0.001000 Time 0.020588 -2022-12-06 10:29:57,663 - Epoch: [17][ 750/ 1200] Overall Loss 0.328250 Objective Loss 0.328250 LR 0.001000 Time 0.020576 -2022-12-06 10:29:57,859 - Epoch: [17][ 760/ 1200] Overall Loss 0.328488 Objective Loss 0.328488 LR 0.001000 Time 0.020562 -2022-12-06 10:29:58,056 - Epoch: [17][ 770/ 1200] Overall Loss 0.328598 Objective Loss 0.328598 LR 0.001000 Time 0.020551 -2022-12-06 10:29:58,252 - Epoch: [17][ 780/ 1200] Overall Loss 0.328381 Objective Loss 0.328381 LR 0.001000 Time 0.020537 -2022-12-06 10:29:58,449 - Epoch: [17][ 790/ 1200] Overall Loss 0.328344 Objective Loss 0.328344 LR 0.001000 Time 0.020526 -2022-12-06 10:29:58,644 - Epoch: [17][ 800/ 1200] Overall Loss 0.328694 Objective Loss 0.328694 LR 0.001000 Time 0.020513 -2022-12-06 10:29:58,842 - Epoch: [17][ 810/ 1200] Overall Loss 0.328717 Objective Loss 0.328717 LR 0.001000 Time 0.020504 -2022-12-06 10:29:59,037 - Epoch: [17][ 820/ 1200] Overall Loss 0.329048 Objective Loss 0.329048 LR 0.001000 Time 0.020491 -2022-12-06 10:29:59,235 - Epoch: [17][ 830/ 1200] Overall Loss 0.329521 Objective Loss 0.329521 LR 0.001000 Time 0.020481 -2022-12-06 10:29:59,430 - Epoch: [17][ 840/ 1200] Overall Loss 0.329650 Objective Loss 0.329650 LR 0.001000 Time 0.020470 -2022-12-06 10:29:59,628 - Epoch: [17][ 850/ 1200] Overall Loss 0.329681 Objective Loss 0.329681 LR 0.001000 Time 0.020461 -2022-12-06 10:29:59,824 - Epoch: [17][ 860/ 1200] Overall Loss 0.330150 Objective Loss 0.330150 LR 0.001000 Time 0.020450 -2022-12-06 10:30:00,021 - Epoch: [17][ 870/ 1200] Overall Loss 0.329987 Objective Loss 0.329987 LR 0.001000 Time 0.020441 -2022-12-06 10:30:00,217 - Epoch: [17][ 880/ 1200] Overall Loss 0.330199 Objective Loss 0.330199 LR 0.001000 Time 0.020431 -2022-12-06 10:30:00,414 - Epoch: [17][ 890/ 1200] Overall Loss 0.330495 Objective Loss 0.330495 LR 0.001000 Time 0.020422 -2022-12-06 10:30:00,610 - Epoch: [17][ 900/ 1200] Overall Loss 0.330697 Objective Loss 0.330697 LR 0.001000 Time 0.020412 -2022-12-06 10:30:00,807 - Epoch: [17][ 910/ 1200] Overall Loss 0.330069 Objective Loss 0.330069 LR 0.001000 Time 0.020403 -2022-12-06 10:30:01,002 - Epoch: [17][ 920/ 1200] Overall Loss 0.330038 Objective Loss 0.330038 LR 0.001000 Time 0.020394 -2022-12-06 10:30:01,199 - Epoch: [17][ 930/ 1200] Overall Loss 0.329864 Objective Loss 0.329864 LR 0.001000 Time 0.020385 -2022-12-06 10:30:01,395 - Epoch: [17][ 940/ 1200] Overall Loss 0.330279 Objective Loss 0.330279 LR 0.001000 Time 0.020376 -2022-12-06 10:30:01,593 - Epoch: [17][ 950/ 1200] Overall Loss 0.329888 Objective Loss 0.329888 LR 0.001000 Time 0.020369 -2022-12-06 10:30:01,788 - Epoch: [17][ 960/ 1200] Overall Loss 0.329808 Objective Loss 0.329808 LR 0.001000 Time 0.020361 -2022-12-06 10:30:01,985 - Epoch: [17][ 970/ 1200] Overall Loss 0.329817 Objective Loss 0.329817 LR 0.001000 Time 0.020353 -2022-12-06 10:30:02,181 - Epoch: [17][ 980/ 1200] Overall Loss 0.329744 Objective Loss 0.329744 LR 0.001000 Time 0.020345 -2022-12-06 10:30:02,378 - Epoch: [17][ 990/ 1200] Overall Loss 0.329966 Objective Loss 0.329966 LR 0.001000 Time 0.020338 -2022-12-06 10:30:02,574 - Epoch: [17][ 1000/ 1200] Overall Loss 0.329791 Objective Loss 0.329791 LR 0.001000 Time 0.020329 -2022-12-06 10:30:02,771 - Epoch: [17][ 1010/ 1200] Overall Loss 0.329917 Objective Loss 0.329917 LR 0.001000 Time 0.020323 -2022-12-06 10:30:02,966 - Epoch: [17][ 1020/ 1200] Overall Loss 0.329547 Objective Loss 0.329547 LR 0.001000 Time 0.020314 -2022-12-06 10:30:03,163 - Epoch: [17][ 1030/ 1200] Overall Loss 0.329827 Objective Loss 0.329827 LR 0.001000 Time 0.020308 -2022-12-06 10:30:03,359 - Epoch: [17][ 1040/ 1200] Overall Loss 0.329699 Objective Loss 0.329699 LR 0.001000 Time 0.020301 -2022-12-06 10:30:03,557 - Epoch: [17][ 1050/ 1200] Overall Loss 0.329413 Objective Loss 0.329413 LR 0.001000 Time 0.020295 -2022-12-06 10:30:03,752 - Epoch: [17][ 1060/ 1200] Overall Loss 0.329413 Objective Loss 0.329413 LR 0.001000 Time 0.020287 -2022-12-06 10:30:03,950 - Epoch: [17][ 1070/ 1200] Overall Loss 0.329579 Objective Loss 0.329579 LR 0.001000 Time 0.020282 -2022-12-06 10:30:04,145 - Epoch: [17][ 1080/ 1200] Overall Loss 0.329873 Objective Loss 0.329873 LR 0.001000 Time 0.020274 -2022-12-06 10:30:04,342 - Epoch: [17][ 1090/ 1200] Overall Loss 0.329820 Objective Loss 0.329820 LR 0.001000 Time 0.020269 -2022-12-06 10:30:04,537 - Epoch: [17][ 1100/ 1200] Overall Loss 0.329775 Objective Loss 0.329775 LR 0.001000 Time 0.020261 -2022-12-06 10:30:04,735 - Epoch: [17][ 1110/ 1200] Overall Loss 0.329609 Objective Loss 0.329609 LR 0.001000 Time 0.020257 -2022-12-06 10:30:04,931 - Epoch: [17][ 1120/ 1200] Overall Loss 0.329648 Objective Loss 0.329648 LR 0.001000 Time 0.020250 -2022-12-06 10:30:05,129 - Epoch: [17][ 1130/ 1200] Overall Loss 0.329537 Objective Loss 0.329537 LR 0.001000 Time 0.020245 -2022-12-06 10:30:05,324 - Epoch: [17][ 1140/ 1200] Overall Loss 0.329817 Objective Loss 0.329817 LR 0.001000 Time 0.020238 -2022-12-06 10:30:05,521 - Epoch: [17][ 1150/ 1200] Overall Loss 0.329704 Objective Loss 0.329704 LR 0.001000 Time 0.020233 -2022-12-06 10:30:05,716 - Epoch: [17][ 1160/ 1200] Overall Loss 0.329483 Objective Loss 0.329483 LR 0.001000 Time 0.020226 -2022-12-06 10:30:05,913 - Epoch: [17][ 1170/ 1200] Overall Loss 0.329505 Objective Loss 0.329505 LR 0.001000 Time 0.020222 -2022-12-06 10:30:06,109 - Epoch: [17][ 1180/ 1200] Overall Loss 0.329320 Objective Loss 0.329320 LR 0.001000 Time 0.020216 -2022-12-06 10:30:06,307 - Epoch: [17][ 1190/ 1200] Overall Loss 0.329296 Objective Loss 0.329296 LR 0.001000 Time 0.020212 -2022-12-06 10:30:06,532 - Epoch: [17][ 1200/ 1200] Overall Loss 0.329584 Objective Loss 0.329584 Top1 80.753138 Top5 97.489540 LR 0.001000 Time 0.020230 -2022-12-06 10:30:06,628 - --- validate (epoch=17)----------- -2022-12-06 10:30:06,629 - 34129 samples (256 per mini-batch) -2022-12-06 10:30:07,105 - Epoch: [17][ 10/ 134] Loss 0.339878 Top1 82.617188 Top5 97.304688 -2022-12-06 10:30:07,247 - Epoch: [17][ 20/ 134] Loss 0.314008 Top1 82.832031 Top5 97.167969 -2022-12-06 10:30:07,394 - Epoch: [17][ 30/ 134] Loss 0.314661 Top1 82.929688 Top5 97.226562 -2022-12-06 10:30:07,535 - Epoch: [17][ 40/ 134] Loss 0.325465 Top1 82.666016 Top5 97.265625 -2022-12-06 10:30:07,681 - Epoch: [17][ 50/ 134] Loss 0.331950 Top1 82.460938 Top5 97.148438 -2022-12-06 10:30:07,820 - Epoch: [17][ 60/ 134] Loss 0.335213 Top1 82.220052 Top5 97.161458 -2022-12-06 10:30:07,962 - Epoch: [17][ 70/ 134] Loss 0.340470 Top1 82.064732 Top5 97.114955 -2022-12-06 10:30:08,092 - Epoch: [17][ 80/ 134] Loss 0.339155 Top1 82.011719 Top5 97.192383 -2022-12-06 10:30:08,219 - Epoch: [17][ 90/ 134] Loss 0.338077 Top1 82.013889 Top5 97.230903 -2022-12-06 10:30:08,345 - Epoch: [17][ 100/ 134] Loss 0.341631 Top1 81.917969 Top5 97.226562 -2022-12-06 10:30:08,472 - Epoch: [17][ 110/ 134] Loss 0.341586 Top1 81.960227 Top5 97.279830 -2022-12-06 10:30:08,599 - Epoch: [17][ 120/ 134] Loss 0.341315 Top1 82.018229 Top5 97.327474 -2022-12-06 10:30:08,728 - Epoch: [17][ 130/ 134] Loss 0.340727 Top1 82.130409 Top5 97.325721 -2022-12-06 10:30:08,765 - Epoch: [17][ 134/ 134] Loss 0.338678 Top1 82.155938 Top5 97.345366 -2022-12-06 10:30:08,853 - ==> Top1: 82.156 Top5: 97.345 Loss: 0.339 - -2022-12-06 10:30:08,853 - ==> Confusion: -[[ 930 3 1 0 2 2 0 0 3 36 0 0 1 4 7 2 3 0 0 0 2] - [ 5 897 4 3 15 29 2 14 2 3 5 3 1 3 9 2 13 0 8 1 8] - [ 10 3 994 12 2 2 33 11 1 2 1 4 0 4 3 6 1 1 2 4 7] - [ 4 4 43 891 0 7 1 0 1 1 13 1 2 3 30 0 2 3 8 0 6] - [ 20 4 5 0 936 5 0 0 4 9 0 2 1 4 12 9 4 1 0 0 4] - [ 5 24 0 0 9 940 3 18 3 2 1 9 2 24 6 1 5 2 2 6 7] - [ 1 4 15 0 0 2 1063 5 0 0 1 3 0 0 1 5 4 1 0 11 2] - [ 5 9 16 0 1 40 17 886 1 0 3 7 1 3 1 3 2 0 37 19 3] - [ 8 2 0 0 0 1 0 1 974 48 6 1 1 4 7 1 4 1 0 1 4] - [ 128 0 3 0 1 0 0 2 19 826 1 0 0 8 7 1 1 0 0 1 3] - [ 2 1 7 5 1 2 3 2 11 2 932 4 2 23 9 0 2 0 6 1 4] - [ 9 3 2 0 1 12 3 6 0 0 0 945 14 14 0 9 4 11 0 15 3] - [ 5 0 0 3 2 6 3 2 0 0 0 66 812 5 3 12 2 28 1 4 15] - [ 4 1 0 0 1 6 0 1 10 19 4 2 1 958 1 3 2 2 0 1 7] - [ 19 3 1 4 4 3 0 1 22 3 2 2 3 2 1048 0 3 1 4 0 5] - [ 8 1 3 1 1 1 2 0 0 0 0 9 2 5 1 981 13 7 1 4 3] - [ 3 3 2 3 2 2 1 0 0 0 0 3 0 3 3 9 1022 2 1 7 6] - [ 10 3 4 1 0 2 1 0 4 0 1 15 7 2 3 18 0 960 1 1 3] - [ 3 2 10 8 1 2 1 21 4 1 11 5 1 1 16 1 1 0 910 4 5] - [ 1 3 4 0 1 5 9 9 0 0 1 24 3 6 2 4 8 3 0 995 2] - [ 314 251 289 87 155 190 98 136 118 112 203 132 325 391 294 149 283 74 166 330 9129]] - -2022-12-06 10:30:09,503 - ==> Best [Top1: 82.672 Top5: 97.594 Sparsity:0.00 Params: 5376 on epoch: 14] -2022-12-06 10:30:09,503 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:30:09,509 - - -2022-12-06 10:30:09,509 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:30:10,445 - Epoch: [18][ 10/ 1200] Overall Loss 0.321481 Objective Loss 0.321481 LR 0.001000 Time 0.093488 -2022-12-06 10:30:10,641 - Epoch: [18][ 20/ 1200] Overall Loss 0.302294 Objective Loss 0.302294 LR 0.001000 Time 0.056538 -2022-12-06 10:30:10,834 - Epoch: [18][ 30/ 1200] Overall Loss 0.304019 Objective Loss 0.304019 LR 0.001000 Time 0.044081 -2022-12-06 10:30:11,024 - Epoch: [18][ 40/ 1200] Overall Loss 0.303440 Objective Loss 0.303440 LR 0.001000 Time 0.037804 -2022-12-06 10:30:11,215 - Epoch: [18][ 50/ 1200] Overall Loss 0.305630 Objective Loss 0.305630 LR 0.001000 Time 0.034054 -2022-12-06 10:30:11,406 - Epoch: [18][ 60/ 1200] Overall Loss 0.303439 Objective Loss 0.303439 LR 0.001000 Time 0.031553 -2022-12-06 10:30:11,597 - Epoch: [18][ 70/ 1200] Overall Loss 0.303641 Objective Loss 0.303641 LR 0.001000 Time 0.029771 -2022-12-06 10:30:11,788 - Epoch: [18][ 80/ 1200] Overall Loss 0.305856 Objective Loss 0.305856 LR 0.001000 Time 0.028423 -2022-12-06 10:30:11,978 - Epoch: [18][ 90/ 1200] Overall Loss 0.305728 Objective Loss 0.305728 LR 0.001000 Time 0.027369 -2022-12-06 10:30:12,169 - Epoch: [18][ 100/ 1200] Overall Loss 0.304903 Objective Loss 0.304903 LR 0.001000 Time 0.026543 -2022-12-06 10:30:12,360 - Epoch: [18][ 110/ 1200] Overall Loss 0.305811 Objective Loss 0.305811 LR 0.001000 Time 0.025857 -2022-12-06 10:30:12,551 - Epoch: [18][ 120/ 1200] Overall Loss 0.305257 Objective Loss 0.305257 LR 0.001000 Time 0.025291 -2022-12-06 10:30:12,743 - Epoch: [18][ 130/ 1200] Overall Loss 0.306608 Objective Loss 0.306608 LR 0.001000 Time 0.024819 -2022-12-06 10:30:12,933 - Epoch: [18][ 140/ 1200] Overall Loss 0.306066 Objective Loss 0.306066 LR 0.001000 Time 0.024401 -2022-12-06 10:30:13,124 - Epoch: [18][ 150/ 1200] Overall Loss 0.306107 Objective Loss 0.306107 LR 0.001000 Time 0.024042 -2022-12-06 10:30:13,315 - Epoch: [18][ 160/ 1200] Overall Loss 0.308172 Objective Loss 0.308172 LR 0.001000 Time 0.023729 -2022-12-06 10:30:13,507 - Epoch: [18][ 170/ 1200] Overall Loss 0.307464 Objective Loss 0.307464 LR 0.001000 Time 0.023459 -2022-12-06 10:30:13,698 - Epoch: [18][ 180/ 1200] Overall Loss 0.306248 Objective Loss 0.306248 LR 0.001000 Time 0.023213 -2022-12-06 10:30:13,889 - Epoch: [18][ 190/ 1200] Overall Loss 0.305583 Objective Loss 0.305583 LR 0.001000 Time 0.022995 -2022-12-06 10:30:14,081 - Epoch: [18][ 200/ 1200] Overall Loss 0.305563 Objective Loss 0.305563 LR 0.001000 Time 0.022801 -2022-12-06 10:30:14,271 - Epoch: [18][ 210/ 1200] Overall Loss 0.306333 Objective Loss 0.306333 LR 0.001000 Time 0.022619 -2022-12-06 10:30:14,462 - Epoch: [18][ 220/ 1200] Overall Loss 0.306886 Objective Loss 0.306886 LR 0.001000 Time 0.022456 -2022-12-06 10:30:14,654 - Epoch: [18][ 230/ 1200] Overall Loss 0.307663 Objective Loss 0.307663 LR 0.001000 Time 0.022311 -2022-12-06 10:30:14,845 - Epoch: [18][ 240/ 1200] Overall Loss 0.308684 Objective Loss 0.308684 LR 0.001000 Time 0.022175 -2022-12-06 10:30:15,036 - Epoch: [18][ 250/ 1200] Overall Loss 0.307534 Objective Loss 0.307534 LR 0.001000 Time 0.022052 -2022-12-06 10:30:15,228 - Epoch: [18][ 260/ 1200] Overall Loss 0.306834 Objective Loss 0.306834 LR 0.001000 Time 0.021939 -2022-12-06 10:30:15,420 - Epoch: [18][ 270/ 1200] Overall Loss 0.307196 Objective Loss 0.307196 LR 0.001000 Time 0.021835 -2022-12-06 10:30:15,611 - Epoch: [18][ 280/ 1200] Overall Loss 0.307375 Objective Loss 0.307375 LR 0.001000 Time 0.021734 -2022-12-06 10:30:15,802 - Epoch: [18][ 290/ 1200] Overall Loss 0.307098 Objective Loss 0.307098 LR 0.001000 Time 0.021643 -2022-12-06 10:30:15,992 - Epoch: [18][ 300/ 1200] Overall Loss 0.307983 Objective Loss 0.307983 LR 0.001000 Time 0.021553 -2022-12-06 10:30:16,183 - Epoch: [18][ 310/ 1200] Overall Loss 0.308055 Objective Loss 0.308055 LR 0.001000 Time 0.021471 -2022-12-06 10:30:16,373 - Epoch: [18][ 320/ 1200] Overall Loss 0.307349 Objective Loss 0.307349 LR 0.001000 Time 0.021392 -2022-12-06 10:30:16,564 - Epoch: [18][ 330/ 1200] Overall Loss 0.306772 Objective Loss 0.306772 LR 0.001000 Time 0.021321 -2022-12-06 10:30:16,754 - Epoch: [18][ 340/ 1200] Overall Loss 0.306809 Objective Loss 0.306809 LR 0.001000 Time 0.021251 -2022-12-06 10:30:16,946 - Epoch: [18][ 350/ 1200] Overall Loss 0.307131 Objective Loss 0.307131 LR 0.001000 Time 0.021191 -2022-12-06 10:30:17,136 - Epoch: [18][ 360/ 1200] Overall Loss 0.306894 Objective Loss 0.306894 LR 0.001000 Time 0.021130 -2022-12-06 10:30:17,327 - Epoch: [18][ 370/ 1200] Overall Loss 0.307759 Objective Loss 0.307759 LR 0.001000 Time 0.021074 -2022-12-06 10:30:17,518 - Epoch: [18][ 380/ 1200] Overall Loss 0.308723 Objective Loss 0.308723 LR 0.001000 Time 0.021020 -2022-12-06 10:30:17,709 - Epoch: [18][ 390/ 1200] Overall Loss 0.309748 Objective Loss 0.309748 LR 0.001000 Time 0.020970 -2022-12-06 10:30:17,901 - Epoch: [18][ 400/ 1200] Overall Loss 0.310598 Objective Loss 0.310598 LR 0.001000 Time 0.020922 -2022-12-06 10:30:18,091 - Epoch: [18][ 410/ 1200] Overall Loss 0.311877 Objective Loss 0.311877 LR 0.001000 Time 0.020876 -2022-12-06 10:30:18,282 - Epoch: [18][ 420/ 1200] Overall Loss 0.312066 Objective Loss 0.312066 LR 0.001000 Time 0.020831 -2022-12-06 10:30:18,473 - Epoch: [18][ 430/ 1200] Overall Loss 0.312987 Objective Loss 0.312987 LR 0.001000 Time 0.020789 -2022-12-06 10:30:18,664 - Epoch: [18][ 440/ 1200] Overall Loss 0.312824 Objective Loss 0.312824 LR 0.001000 Time 0.020750 -2022-12-06 10:30:18,854 - Epoch: [18][ 450/ 1200] Overall Loss 0.312911 Objective Loss 0.312911 LR 0.001000 Time 0.020710 -2022-12-06 10:30:19,044 - Epoch: [18][ 460/ 1200] Overall Loss 0.312314 Objective Loss 0.312314 LR 0.001000 Time 0.020673 -2022-12-06 10:30:19,236 - Epoch: [18][ 470/ 1200] Overall Loss 0.312962 Objective Loss 0.312962 LR 0.001000 Time 0.020639 -2022-12-06 10:30:19,426 - Epoch: [18][ 480/ 1200] Overall Loss 0.313335 Objective Loss 0.313335 LR 0.001000 Time 0.020604 -2022-12-06 10:30:19,617 - Epoch: [18][ 490/ 1200] Overall Loss 0.313964 Objective Loss 0.313964 LR 0.001000 Time 0.020572 -2022-12-06 10:30:19,808 - Epoch: [18][ 500/ 1200] Overall Loss 0.314366 Objective Loss 0.314366 LR 0.001000 Time 0.020542 -2022-12-06 10:30:19,999 - Epoch: [18][ 510/ 1200] Overall Loss 0.314180 Objective Loss 0.314180 LR 0.001000 Time 0.020512 -2022-12-06 10:30:20,189 - Epoch: [18][ 520/ 1200] Overall Loss 0.314446 Objective Loss 0.314446 LR 0.001000 Time 0.020482 -2022-12-06 10:30:20,380 - Epoch: [18][ 530/ 1200] Overall Loss 0.314181 Objective Loss 0.314181 LR 0.001000 Time 0.020455 -2022-12-06 10:30:20,572 - Epoch: [18][ 540/ 1200] Overall Loss 0.314018 Objective Loss 0.314018 LR 0.001000 Time 0.020431 -2022-12-06 10:30:20,763 - Epoch: [18][ 550/ 1200] Overall Loss 0.314167 Objective Loss 0.314167 LR 0.001000 Time 0.020406 -2022-12-06 10:30:20,954 - Epoch: [18][ 560/ 1200] Overall Loss 0.314305 Objective Loss 0.314305 LR 0.001000 Time 0.020382 -2022-12-06 10:30:21,145 - Epoch: [18][ 570/ 1200] Overall Loss 0.313728 Objective Loss 0.313728 LR 0.001000 Time 0.020358 -2022-12-06 10:30:21,337 - Epoch: [18][ 580/ 1200] Overall Loss 0.313974 Objective Loss 0.313974 LR 0.001000 Time 0.020336 -2022-12-06 10:30:21,528 - Epoch: [18][ 590/ 1200] Overall Loss 0.313989 Objective Loss 0.313989 LR 0.001000 Time 0.020316 -2022-12-06 10:30:21,720 - Epoch: [18][ 600/ 1200] Overall Loss 0.313959 Objective Loss 0.313959 LR 0.001000 Time 0.020296 -2022-12-06 10:30:21,910 - Epoch: [18][ 610/ 1200] Overall Loss 0.313849 Objective Loss 0.313849 LR 0.001000 Time 0.020273 -2022-12-06 10:30:22,101 - Epoch: [18][ 620/ 1200] Overall Loss 0.313820 Objective Loss 0.313820 LR 0.001000 Time 0.020254 -2022-12-06 10:30:22,293 - Epoch: [18][ 630/ 1200] Overall Loss 0.313284 Objective Loss 0.313284 LR 0.001000 Time 0.020235 -2022-12-06 10:30:22,483 - Epoch: [18][ 640/ 1200] Overall Loss 0.313261 Objective Loss 0.313261 LR 0.001000 Time 0.020216 -2022-12-06 10:30:22,674 - Epoch: [18][ 650/ 1200] Overall Loss 0.312816 Objective Loss 0.312816 LR 0.001000 Time 0.020198 -2022-12-06 10:30:22,864 - Epoch: [18][ 660/ 1200] Overall Loss 0.312769 Objective Loss 0.312769 LR 0.001000 Time 0.020178 -2022-12-06 10:30:23,054 - Epoch: [18][ 670/ 1200] Overall Loss 0.312667 Objective Loss 0.312667 LR 0.001000 Time 0.020161 -2022-12-06 10:30:23,245 - Epoch: [18][ 680/ 1200] Overall Loss 0.312667 Objective Loss 0.312667 LR 0.001000 Time 0.020144 -2022-12-06 10:30:23,436 - Epoch: [18][ 690/ 1200] Overall Loss 0.312590 Objective Loss 0.312590 LR 0.001000 Time 0.020129 -2022-12-06 10:30:23,627 - Epoch: [18][ 700/ 1200] Overall Loss 0.312259 Objective Loss 0.312259 LR 0.001000 Time 0.020113 -2022-12-06 10:30:23,818 - Epoch: [18][ 710/ 1200] Overall Loss 0.312556 Objective Loss 0.312556 LR 0.001000 Time 0.020098 -2022-12-06 10:30:24,009 - Epoch: [18][ 720/ 1200] Overall Loss 0.312264 Objective Loss 0.312264 LR 0.001000 Time 0.020083 -2022-12-06 10:30:24,200 - Epoch: [18][ 730/ 1200] Overall Loss 0.312766 Objective Loss 0.312766 LR 0.001000 Time 0.020069 -2022-12-06 10:30:24,391 - Epoch: [18][ 740/ 1200] Overall Loss 0.312530 Objective Loss 0.312530 LR 0.001000 Time 0.020054 -2022-12-06 10:30:24,581 - Epoch: [18][ 750/ 1200] Overall Loss 0.312491 Objective Loss 0.312491 LR 0.001000 Time 0.020040 -2022-12-06 10:30:24,772 - Epoch: [18][ 760/ 1200] Overall Loss 0.312559 Objective Loss 0.312559 LR 0.001000 Time 0.020027 -2022-12-06 10:30:24,963 - Epoch: [18][ 770/ 1200] Overall Loss 0.312991 Objective Loss 0.312991 LR 0.001000 Time 0.020013 -2022-12-06 10:30:25,154 - Epoch: [18][ 780/ 1200] Overall Loss 0.313446 Objective Loss 0.313446 LR 0.001000 Time 0.020001 -2022-12-06 10:30:25,344 - Epoch: [18][ 790/ 1200] Overall Loss 0.313404 Objective Loss 0.313404 LR 0.001000 Time 0.019989 -2022-12-06 10:30:25,535 - Epoch: [18][ 800/ 1200] Overall Loss 0.313683 Objective Loss 0.313683 LR 0.001000 Time 0.019977 -2022-12-06 10:30:25,726 - Epoch: [18][ 810/ 1200] Overall Loss 0.314017 Objective Loss 0.314017 LR 0.001000 Time 0.019965 -2022-12-06 10:30:25,917 - Epoch: [18][ 820/ 1200] Overall Loss 0.314136 Objective Loss 0.314136 LR 0.001000 Time 0.019954 -2022-12-06 10:30:26,108 - Epoch: [18][ 830/ 1200] Overall Loss 0.314355 Objective Loss 0.314355 LR 0.001000 Time 0.019943 -2022-12-06 10:30:26,299 - Epoch: [18][ 840/ 1200] Overall Loss 0.314783 Objective Loss 0.314783 LR 0.001000 Time 0.019933 -2022-12-06 10:30:26,491 - Epoch: [18][ 850/ 1200] Overall Loss 0.314689 Objective Loss 0.314689 LR 0.001000 Time 0.019922 -2022-12-06 10:30:26,682 - Epoch: [18][ 860/ 1200] Overall Loss 0.315095 Objective Loss 0.315095 LR 0.001000 Time 0.019912 -2022-12-06 10:30:26,873 - Epoch: [18][ 870/ 1200] Overall Loss 0.315220 Objective Loss 0.315220 LR 0.001000 Time 0.019902 -2022-12-06 10:30:27,064 - Epoch: [18][ 880/ 1200] Overall Loss 0.315335 Objective Loss 0.315335 LR 0.001000 Time 0.019893 -2022-12-06 10:30:27,255 - Epoch: [18][ 890/ 1200] Overall Loss 0.315435 Objective Loss 0.315435 LR 0.001000 Time 0.019883 -2022-12-06 10:30:27,446 - Epoch: [18][ 900/ 1200] Overall Loss 0.315589 Objective Loss 0.315589 LR 0.001000 Time 0.019873 -2022-12-06 10:30:27,636 - Epoch: [18][ 910/ 1200] Overall Loss 0.315739 Objective Loss 0.315739 LR 0.001000 Time 0.019864 -2022-12-06 10:30:27,826 - Epoch: [18][ 920/ 1200] Overall Loss 0.315747 Objective Loss 0.315747 LR 0.001000 Time 0.019854 -2022-12-06 10:30:28,018 - Epoch: [18][ 930/ 1200] Overall Loss 0.315742 Objective Loss 0.315742 LR 0.001000 Time 0.019846 -2022-12-06 10:30:28,209 - Epoch: [18][ 940/ 1200] Overall Loss 0.315648 Objective Loss 0.315648 LR 0.001000 Time 0.019837 -2022-12-06 10:30:28,400 - Epoch: [18][ 950/ 1200] Overall Loss 0.315996 Objective Loss 0.315996 LR 0.001000 Time 0.019829 -2022-12-06 10:30:28,591 - Epoch: [18][ 960/ 1200] Overall Loss 0.316197 Objective Loss 0.316197 LR 0.001000 Time 0.019821 -2022-12-06 10:30:28,782 - Epoch: [18][ 970/ 1200] Overall Loss 0.316360 Objective Loss 0.316360 LR 0.001000 Time 0.019813 -2022-12-06 10:30:28,973 - Epoch: [18][ 980/ 1200] Overall Loss 0.316756 Objective Loss 0.316756 LR 0.001000 Time 0.019805 -2022-12-06 10:30:29,164 - Epoch: [18][ 990/ 1200] Overall Loss 0.316827 Objective Loss 0.316827 LR 0.001000 Time 0.019797 -2022-12-06 10:30:29,355 - Epoch: [18][ 1000/ 1200] Overall Loss 0.316514 Objective Loss 0.316514 LR 0.001000 Time 0.019790 -2022-12-06 10:30:29,546 - Epoch: [18][ 1010/ 1200] Overall Loss 0.316692 Objective Loss 0.316692 LR 0.001000 Time 0.019782 -2022-12-06 10:30:29,737 - Epoch: [18][ 1020/ 1200] Overall Loss 0.316593 Objective Loss 0.316593 LR 0.001000 Time 0.019775 -2022-12-06 10:30:29,928 - Epoch: [18][ 1030/ 1200] Overall Loss 0.316480 Objective Loss 0.316480 LR 0.001000 Time 0.019768 -2022-12-06 10:30:30,119 - Epoch: [18][ 1040/ 1200] Overall Loss 0.316586 Objective Loss 0.316586 LR 0.001000 Time 0.019761 -2022-12-06 10:30:30,310 - Epoch: [18][ 1050/ 1200] Overall Loss 0.316918 Objective Loss 0.316918 LR 0.001000 Time 0.019755 -2022-12-06 10:30:30,501 - Epoch: [18][ 1060/ 1200] Overall Loss 0.316582 Objective Loss 0.316582 LR 0.001000 Time 0.019748 -2022-12-06 10:30:30,692 - Epoch: [18][ 1070/ 1200] Overall Loss 0.316763 Objective Loss 0.316763 LR 0.001000 Time 0.019742 -2022-12-06 10:30:30,883 - Epoch: [18][ 1080/ 1200] Overall Loss 0.316967 Objective Loss 0.316967 LR 0.001000 Time 0.019735 -2022-12-06 10:30:31,075 - Epoch: [18][ 1090/ 1200] Overall Loss 0.316903 Objective Loss 0.316903 LR 0.001000 Time 0.019729 -2022-12-06 10:30:31,266 - Epoch: [18][ 1100/ 1200] Overall Loss 0.316855 Objective Loss 0.316855 LR 0.001000 Time 0.019723 -2022-12-06 10:30:31,457 - Epoch: [18][ 1110/ 1200] Overall Loss 0.317082 Objective Loss 0.317082 LR 0.001000 Time 0.019717 -2022-12-06 10:30:31,648 - Epoch: [18][ 1120/ 1200] Overall Loss 0.317126 Objective Loss 0.317126 LR 0.001000 Time 0.019711 -2022-12-06 10:30:31,839 - Epoch: [18][ 1130/ 1200] Overall Loss 0.317492 Objective Loss 0.317492 LR 0.001000 Time 0.019705 -2022-12-06 10:30:32,030 - Epoch: [18][ 1140/ 1200] Overall Loss 0.317607 Objective Loss 0.317607 LR 0.001000 Time 0.019699 -2022-12-06 10:30:32,221 - Epoch: [18][ 1150/ 1200] Overall Loss 0.317536 Objective Loss 0.317536 LR 0.001000 Time 0.019693 -2022-12-06 10:30:32,411 - Epoch: [18][ 1160/ 1200] Overall Loss 0.317488 Objective Loss 0.317488 LR 0.001000 Time 0.019687 -2022-12-06 10:30:32,602 - Epoch: [18][ 1170/ 1200] Overall Loss 0.317497 Objective Loss 0.317497 LR 0.001000 Time 0.019682 -2022-12-06 10:30:32,793 - Epoch: [18][ 1180/ 1200] Overall Loss 0.317731 Objective Loss 0.317731 LR 0.001000 Time 0.019676 -2022-12-06 10:30:32,985 - Epoch: [18][ 1190/ 1200] Overall Loss 0.317915 Objective Loss 0.317915 LR 0.001000 Time 0.019671 -2022-12-06 10:30:33,219 - Epoch: [18][ 1200/ 1200] Overall Loss 0.318087 Objective Loss 0.318087 Top1 82.008368 Top5 98.117155 LR 0.001000 Time 0.019703 -2022-12-06 10:30:33,308 - --- validate (epoch=18)----------- -2022-12-06 10:30:33,309 - 34129 samples (256 per mini-batch) -2022-12-06 10:30:33,765 - Epoch: [18][ 10/ 134] Loss 0.331606 Top1 83.007812 Top5 97.148438 -2022-12-06 10:30:33,895 - Epoch: [18][ 20/ 134] Loss 0.338938 Top1 82.949219 Top5 97.500000 -2022-12-06 10:30:34,027 - Epoch: [18][ 30/ 134] Loss 0.336357 Top1 83.111979 Top5 97.617188 -2022-12-06 10:30:34,158 - Epoch: [18][ 40/ 134] Loss 0.339056 Top1 82.851562 Top5 97.490234 -2022-12-06 10:30:34,290 - Epoch: [18][ 50/ 134] Loss 0.342292 Top1 82.710938 Top5 97.414062 -2022-12-06 10:30:34,422 - Epoch: [18][ 60/ 134] Loss 0.334838 Top1 82.812500 Top5 97.473958 -2022-12-06 10:30:34,550 - Epoch: [18][ 70/ 134] Loss 0.334002 Top1 82.818080 Top5 97.589286 -2022-12-06 10:30:34,683 - Epoch: [18][ 80/ 134] Loss 0.335331 Top1 82.719727 Top5 97.524414 -2022-12-06 10:30:34,815 - Epoch: [18][ 90/ 134] Loss 0.334839 Top1 82.786458 Top5 97.569444 -2022-12-06 10:30:34,946 - Epoch: [18][ 100/ 134] Loss 0.329897 Top1 82.929688 Top5 97.621094 -2022-12-06 10:30:35,080 - Epoch: [18][ 110/ 134] Loss 0.330090 Top1 82.816051 Top5 97.631392 -2022-12-06 10:30:35,213 - Epoch: [18][ 120/ 134] Loss 0.328811 Top1 82.832031 Top5 97.620443 -2022-12-06 10:30:35,347 - Epoch: [18][ 130/ 134] Loss 0.329816 Top1 82.872596 Top5 97.617188 -2022-12-06 10:30:35,386 - Epoch: [18][ 134/ 134] Loss 0.330415 Top1 82.847432 Top5 97.603211 -2022-12-06 10:30:35,474 - ==> Top1: 82.847 Top5: 97.603 Loss: 0.330 - -2022-12-06 10:30:35,474 - ==> Confusion: -[[ 821 5 2 1 11 2 0 1 14 115 0 3 1 4 11 1 0 0 1 0 3] - [ 0 905 1 2 14 19 0 28 3 0 6 2 3 2 4 5 8 0 12 1 12] - [ 6 6 996 25 4 1 17 11 1 0 4 2 3 3 1 1 0 0 7 1 14] - [ 2 3 19 937 0 1 0 1 1 0 11 0 1 2 18 0 1 7 11 0 5] - [ 7 4 1 1 955 1 1 2 0 9 1 3 0 4 11 7 8 2 1 0 2] - [ 0 30 2 5 9 909 5 33 5 7 1 14 7 24 5 1 4 2 1 2 3] - [ 0 7 25 3 1 1 1045 7 1 0 0 6 0 0 0 4 2 3 1 8 4] - [ 1 12 13 2 1 15 1 948 1 2 2 2 0 1 3 1 1 1 33 10 4] - [ 2 1 0 1 0 0 0 4 978 47 5 1 3 7 9 1 0 0 3 0 2] - [ 37 0 1 1 3 0 0 2 36 904 0 1 1 8 2 0 1 0 2 0 2] - [ 0 4 9 23 1 3 0 3 20 3 907 3 3 19 3 1 0 0 12 2 3] - [ 5 5 4 0 1 10 2 7 2 0 0 977 12 4 0 2 0 9 3 2 6] - [ 0 4 3 8 2 1 0 2 0 0 0 58 829 2 3 7 3 31 0 2 14] - [ 0 5 2 1 3 3 0 2 18 21 3 3 4 935 1 3 1 1 1 6 10] - [ 5 3 0 17 7 1 0 1 31 3 3 1 4 4 1033 0 2 3 4 1 7] - [ 2 3 2 4 3 1 3 1 0 1 0 13 5 3 1 974 11 12 0 3 1] - [ 0 4 4 1 9 0 1 1 5 0 0 6 0 0 1 5 1020 3 0 5 7] - [ 0 1 2 3 0 0 1 2 2 1 2 13 9 2 2 13 1 980 0 1 1] - [ 3 2 6 16 0 2 0 26 4 0 4 3 2 0 11 0 0 0 917 5 7] - [ 0 5 5 1 1 6 8 9 0 0 1 32 5 5 1 1 4 5 1 983 7] - [ 130 220 226 140 242 124 60 189 127 174 158 174 332 363 257 107 290 118 231 245 9319]] - -2022-12-06 10:30:36,136 - ==> Best [Top1: 82.847 Top5: 97.603 Sparsity:0.00 Params: 5376 on epoch: 18] -2022-12-06 10:30:36,136 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:30:36,143 - - -2022-12-06 10:30:36,143 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:30:37,073 - Epoch: [19][ 10/ 1200] Overall Loss 0.310555 Objective Loss 0.310555 LR 0.001000 Time 0.092890 -2022-12-06 10:30:37,273 - Epoch: [19][ 20/ 1200] Overall Loss 0.311963 Objective Loss 0.311963 LR 0.001000 Time 0.056429 -2022-12-06 10:30:37,464 - Epoch: [19][ 30/ 1200] Overall Loss 0.300396 Objective Loss 0.300396 LR 0.001000 Time 0.043961 -2022-12-06 10:30:37,654 - Epoch: [19][ 40/ 1200] Overall Loss 0.295301 Objective Loss 0.295301 LR 0.001000 Time 0.037718 -2022-12-06 10:30:37,844 - Epoch: [19][ 50/ 1200] Overall Loss 0.292620 Objective Loss 0.292620 LR 0.001000 Time 0.033967 -2022-12-06 10:30:38,035 - Epoch: [19][ 60/ 1200] Overall Loss 0.296643 Objective Loss 0.296643 LR 0.001000 Time 0.031481 -2022-12-06 10:30:38,226 - Epoch: [19][ 70/ 1200] Overall Loss 0.294196 Objective Loss 0.294196 LR 0.001000 Time 0.029694 -2022-12-06 10:30:38,416 - Epoch: [19][ 80/ 1200] Overall Loss 0.297441 Objective Loss 0.297441 LR 0.001000 Time 0.028354 -2022-12-06 10:30:38,606 - Epoch: [19][ 90/ 1200] Overall Loss 0.298009 Objective Loss 0.298009 LR 0.001000 Time 0.027311 -2022-12-06 10:30:38,797 - Epoch: [19][ 100/ 1200] Overall Loss 0.300676 Objective Loss 0.300676 LR 0.001000 Time 0.026483 -2022-12-06 10:30:38,988 - Epoch: [19][ 110/ 1200] Overall Loss 0.301033 Objective Loss 0.301033 LR 0.001000 Time 0.025806 -2022-12-06 10:30:39,178 - Epoch: [19][ 120/ 1200] Overall Loss 0.301185 Objective Loss 0.301185 LR 0.001000 Time 0.025237 -2022-12-06 10:30:39,368 - Epoch: [19][ 130/ 1200] Overall Loss 0.302827 Objective Loss 0.302827 LR 0.001000 Time 0.024755 -2022-12-06 10:30:39,559 - Epoch: [19][ 140/ 1200] Overall Loss 0.304149 Objective Loss 0.304149 LR 0.001000 Time 0.024343 -2022-12-06 10:30:39,749 - Epoch: [19][ 150/ 1200] Overall Loss 0.303943 Objective Loss 0.303943 LR 0.001000 Time 0.023987 -2022-12-06 10:30:39,940 - Epoch: [19][ 160/ 1200] Overall Loss 0.304326 Objective Loss 0.304326 LR 0.001000 Time 0.023675 -2022-12-06 10:30:40,130 - Epoch: [19][ 170/ 1200] Overall Loss 0.303360 Objective Loss 0.303360 LR 0.001000 Time 0.023398 -2022-12-06 10:30:40,321 - Epoch: [19][ 180/ 1200] Overall Loss 0.303846 Objective Loss 0.303846 LR 0.001000 Time 0.023154 -2022-12-06 10:30:40,511 - Epoch: [19][ 190/ 1200] Overall Loss 0.302633 Objective Loss 0.302633 LR 0.001000 Time 0.022935 -2022-12-06 10:30:40,702 - Epoch: [19][ 200/ 1200] Overall Loss 0.303404 Objective Loss 0.303404 LR 0.001000 Time 0.022739 -2022-12-06 10:30:40,892 - Epoch: [19][ 210/ 1200] Overall Loss 0.302673 Objective Loss 0.302673 LR 0.001000 Time 0.022561 -2022-12-06 10:30:41,083 - Epoch: [19][ 220/ 1200] Overall Loss 0.302676 Objective Loss 0.302676 LR 0.001000 Time 0.022399 -2022-12-06 10:30:41,273 - Epoch: [19][ 230/ 1200] Overall Loss 0.303755 Objective Loss 0.303755 LR 0.001000 Time 0.022251 -2022-12-06 10:30:41,464 - Epoch: [19][ 240/ 1200] Overall Loss 0.304068 Objective Loss 0.304068 LR 0.001000 Time 0.022115 -2022-12-06 10:30:41,654 - Epoch: [19][ 250/ 1200] Overall Loss 0.305983 Objective Loss 0.305983 LR 0.001000 Time 0.021988 -2022-12-06 10:30:41,844 - Epoch: [19][ 260/ 1200] Overall Loss 0.305363 Objective Loss 0.305363 LR 0.001000 Time 0.021871 -2022-12-06 10:30:42,034 - Epoch: [19][ 270/ 1200] Overall Loss 0.305690 Objective Loss 0.305690 LR 0.001000 Time 0.021763 -2022-12-06 10:30:42,224 - Epoch: [19][ 280/ 1200] Overall Loss 0.305313 Objective Loss 0.305313 LR 0.001000 Time 0.021663 -2022-12-06 10:30:42,414 - Epoch: [19][ 290/ 1200] Overall Loss 0.305547 Objective Loss 0.305547 LR 0.001000 Time 0.021571 -2022-12-06 10:30:42,604 - Epoch: [19][ 300/ 1200] Overall Loss 0.304753 Objective Loss 0.304753 LR 0.001000 Time 0.021483 -2022-12-06 10:30:42,795 - Epoch: [19][ 310/ 1200] Overall Loss 0.305090 Objective Loss 0.305090 LR 0.001000 Time 0.021404 -2022-12-06 10:30:42,986 - Epoch: [19][ 320/ 1200] Overall Loss 0.305686 Objective Loss 0.305686 LR 0.001000 Time 0.021329 -2022-12-06 10:30:43,176 - Epoch: [19][ 330/ 1200] Overall Loss 0.306246 Objective Loss 0.306246 LR 0.001000 Time 0.021257 -2022-12-06 10:30:43,366 - Epoch: [19][ 340/ 1200] Overall Loss 0.306872 Objective Loss 0.306872 LR 0.001000 Time 0.021190 -2022-12-06 10:30:43,557 - Epoch: [19][ 350/ 1200] Overall Loss 0.306921 Objective Loss 0.306921 LR 0.001000 Time 0.021127 -2022-12-06 10:30:43,747 - Epoch: [19][ 360/ 1200] Overall Loss 0.307590 Objective Loss 0.307590 LR 0.001000 Time 0.021069 -2022-12-06 10:30:43,938 - Epoch: [19][ 370/ 1200] Overall Loss 0.307733 Objective Loss 0.307733 LR 0.001000 Time 0.021013 -2022-12-06 10:30:44,128 - Epoch: [19][ 380/ 1200] Overall Loss 0.307877 Objective Loss 0.307877 LR 0.001000 Time 0.020958 -2022-12-06 10:30:44,318 - Epoch: [19][ 390/ 1200] Overall Loss 0.307699 Objective Loss 0.307699 LR 0.001000 Time 0.020907 -2022-12-06 10:30:44,508 - Epoch: [19][ 400/ 1200] Overall Loss 0.307044 Objective Loss 0.307044 LR 0.001000 Time 0.020858 -2022-12-06 10:30:44,698 - Epoch: [19][ 410/ 1200] Overall Loss 0.307615 Objective Loss 0.307615 LR 0.001000 Time 0.020812 -2022-12-06 10:30:44,888 - Epoch: [19][ 420/ 1200] Overall Loss 0.308170 Objective Loss 0.308170 LR 0.001000 Time 0.020767 -2022-12-06 10:30:45,078 - Epoch: [19][ 430/ 1200] Overall Loss 0.308575 Objective Loss 0.308575 LR 0.001000 Time 0.020725 -2022-12-06 10:30:45,269 - Epoch: [19][ 440/ 1200] Overall Loss 0.308393 Objective Loss 0.308393 LR 0.001000 Time 0.020686 -2022-12-06 10:30:45,459 - Epoch: [19][ 450/ 1200] Overall Loss 0.309008 Objective Loss 0.309008 LR 0.001000 Time 0.020648 -2022-12-06 10:30:45,650 - Epoch: [19][ 460/ 1200] Overall Loss 0.308758 Objective Loss 0.308758 LR 0.001000 Time 0.020612 -2022-12-06 10:30:45,841 - Epoch: [19][ 470/ 1200] Overall Loss 0.309019 Objective Loss 0.309019 LR 0.001000 Time 0.020580 -2022-12-06 10:30:46,031 - Epoch: [19][ 480/ 1200] Overall Loss 0.309061 Objective Loss 0.309061 LR 0.001000 Time 0.020546 -2022-12-06 10:30:46,222 - Epoch: [19][ 490/ 1200] Overall Loss 0.309136 Objective Loss 0.309136 LR 0.001000 Time 0.020515 -2022-12-06 10:30:46,412 - Epoch: [19][ 500/ 1200] Overall Loss 0.309144 Objective Loss 0.309144 LR 0.001000 Time 0.020483 -2022-12-06 10:30:46,602 - Epoch: [19][ 510/ 1200] Overall Loss 0.309411 Objective Loss 0.309411 LR 0.001000 Time 0.020453 -2022-12-06 10:30:46,792 - Epoch: [19][ 520/ 1200] Overall Loss 0.309982 Objective Loss 0.309982 LR 0.001000 Time 0.020424 -2022-12-06 10:30:46,982 - Epoch: [19][ 530/ 1200] Overall Loss 0.309694 Objective Loss 0.309694 LR 0.001000 Time 0.020397 -2022-12-06 10:30:47,173 - Epoch: [19][ 540/ 1200] Overall Loss 0.309617 Objective Loss 0.309617 LR 0.001000 Time 0.020371 -2022-12-06 10:30:47,363 - Epoch: [19][ 550/ 1200] Overall Loss 0.310259 Objective Loss 0.310259 LR 0.001000 Time 0.020346 -2022-12-06 10:30:47,554 - Epoch: [19][ 560/ 1200] Overall Loss 0.310493 Objective Loss 0.310493 LR 0.001000 Time 0.020323 -2022-12-06 10:30:47,745 - Epoch: [19][ 570/ 1200] Overall Loss 0.310502 Objective Loss 0.310502 LR 0.001000 Time 0.020301 -2022-12-06 10:30:47,937 - Epoch: [19][ 580/ 1200] Overall Loss 0.310917 Objective Loss 0.310917 LR 0.001000 Time 0.020281 -2022-12-06 10:30:48,129 - Epoch: [19][ 590/ 1200] Overall Loss 0.311081 Objective Loss 0.311081 LR 0.001000 Time 0.020262 -2022-12-06 10:30:48,322 - Epoch: [19][ 600/ 1200] Overall Loss 0.311635 Objective Loss 0.311635 LR 0.001000 Time 0.020244 -2022-12-06 10:30:48,515 - Epoch: [19][ 610/ 1200] Overall Loss 0.311888 Objective Loss 0.311888 LR 0.001000 Time 0.020227 -2022-12-06 10:30:48,707 - Epoch: [19][ 620/ 1200] Overall Loss 0.311624 Objective Loss 0.311624 LR 0.001000 Time 0.020210 -2022-12-06 10:30:48,899 - Epoch: [19][ 630/ 1200] Overall Loss 0.311911 Objective Loss 0.311911 LR 0.001000 Time 0.020193 -2022-12-06 10:30:49,092 - Epoch: [19][ 640/ 1200] Overall Loss 0.311873 Objective Loss 0.311873 LR 0.001000 Time 0.020178 -2022-12-06 10:30:49,284 - Epoch: [19][ 650/ 1200] Overall Loss 0.311864 Objective Loss 0.311864 LR 0.001000 Time 0.020163 -2022-12-06 10:30:49,476 - Epoch: [19][ 660/ 1200] Overall Loss 0.312291 Objective Loss 0.312291 LR 0.001000 Time 0.020148 -2022-12-06 10:30:49,668 - Epoch: [19][ 670/ 1200] Overall Loss 0.312709 Objective Loss 0.312709 LR 0.001000 Time 0.020133 -2022-12-06 10:30:49,860 - Epoch: [19][ 680/ 1200] Overall Loss 0.313329 Objective Loss 0.313329 LR 0.001000 Time 0.020118 -2022-12-06 10:30:50,052 - Epoch: [19][ 690/ 1200] Overall Loss 0.313643 Objective Loss 0.313643 LR 0.001000 Time 0.020104 -2022-12-06 10:30:50,244 - Epoch: [19][ 700/ 1200] Overall Loss 0.314140 Objective Loss 0.314140 LR 0.001000 Time 0.020090 -2022-12-06 10:30:50,436 - Epoch: [19][ 710/ 1200] Overall Loss 0.314400 Objective Loss 0.314400 LR 0.001000 Time 0.020077 -2022-12-06 10:30:50,628 - Epoch: [19][ 720/ 1200] Overall Loss 0.314690 Objective Loss 0.314690 LR 0.001000 Time 0.020064 -2022-12-06 10:30:50,820 - Epoch: [19][ 730/ 1200] Overall Loss 0.314819 Objective Loss 0.314819 LR 0.001000 Time 0.020052 -2022-12-06 10:30:51,012 - Epoch: [19][ 740/ 1200] Overall Loss 0.314888 Objective Loss 0.314888 LR 0.001000 Time 0.020039 -2022-12-06 10:30:51,204 - Epoch: [19][ 750/ 1200] Overall Loss 0.314850 Objective Loss 0.314850 LR 0.001000 Time 0.020027 -2022-12-06 10:30:51,397 - Epoch: [19][ 760/ 1200] Overall Loss 0.314850 Objective Loss 0.314850 LR 0.001000 Time 0.020017 -2022-12-06 10:30:51,589 - Epoch: [19][ 770/ 1200] Overall Loss 0.314623 Objective Loss 0.314623 LR 0.001000 Time 0.020005 -2022-12-06 10:30:51,782 - Epoch: [19][ 780/ 1200] Overall Loss 0.314779 Objective Loss 0.314779 LR 0.001000 Time 0.019995 -2022-12-06 10:30:51,973 - Epoch: [19][ 790/ 1200] Overall Loss 0.314715 Objective Loss 0.314715 LR 0.001000 Time 0.019984 -2022-12-06 10:30:52,166 - Epoch: [19][ 800/ 1200] Overall Loss 0.314737 Objective Loss 0.314737 LR 0.001000 Time 0.019975 -2022-12-06 10:30:52,358 - Epoch: [19][ 810/ 1200] Overall Loss 0.314414 Objective Loss 0.314414 LR 0.001000 Time 0.019964 -2022-12-06 10:30:52,550 - Epoch: [19][ 820/ 1200] Overall Loss 0.314594 Objective Loss 0.314594 LR 0.001000 Time 0.019955 -2022-12-06 10:30:52,742 - Epoch: [19][ 830/ 1200] Overall Loss 0.314753 Objective Loss 0.314753 LR 0.001000 Time 0.019944 -2022-12-06 10:30:52,935 - Epoch: [19][ 840/ 1200] Overall Loss 0.315113 Objective Loss 0.315113 LR 0.001000 Time 0.019936 -2022-12-06 10:30:53,128 - Epoch: [19][ 850/ 1200] Overall Loss 0.314942 Objective Loss 0.314942 LR 0.001000 Time 0.019928 -2022-12-06 10:30:53,321 - Epoch: [19][ 860/ 1200] Overall Loss 0.314689 Objective Loss 0.314689 LR 0.001000 Time 0.019920 -2022-12-06 10:30:53,513 - Epoch: [19][ 870/ 1200] Overall Loss 0.314364 Objective Loss 0.314364 LR 0.001000 Time 0.019911 -2022-12-06 10:30:53,706 - Epoch: [19][ 880/ 1200] Overall Loss 0.314473 Objective Loss 0.314473 LR 0.001000 Time 0.019904 -2022-12-06 10:30:53,897 - Epoch: [19][ 890/ 1200] Overall Loss 0.314614 Objective Loss 0.314614 LR 0.001000 Time 0.019895 -2022-12-06 10:30:54,091 - Epoch: [19][ 900/ 1200] Overall Loss 0.314546 Objective Loss 0.314546 LR 0.001000 Time 0.019888 -2022-12-06 10:30:54,283 - Epoch: [19][ 910/ 1200] Overall Loss 0.314408 Objective Loss 0.314408 LR 0.001000 Time 0.019880 -2022-12-06 10:30:54,476 - Epoch: [19][ 920/ 1200] Overall Loss 0.314179 Objective Loss 0.314179 LR 0.001000 Time 0.019873 -2022-12-06 10:30:54,668 - Epoch: [19][ 930/ 1200] Overall Loss 0.314102 Objective Loss 0.314102 LR 0.001000 Time 0.019865 -2022-12-06 10:30:54,860 - Epoch: [19][ 940/ 1200] Overall Loss 0.314007 Objective Loss 0.314007 LR 0.001000 Time 0.019858 -2022-12-06 10:30:55,053 - Epoch: [19][ 950/ 1200] Overall Loss 0.313843 Objective Loss 0.313843 LR 0.001000 Time 0.019852 -2022-12-06 10:30:55,246 - Epoch: [19][ 960/ 1200] Overall Loss 0.313676 Objective Loss 0.313676 LR 0.001000 Time 0.019845 -2022-12-06 10:30:55,438 - Epoch: [19][ 970/ 1200] Overall Loss 0.313397 Objective Loss 0.313397 LR 0.001000 Time 0.019838 -2022-12-06 10:30:55,631 - Epoch: [19][ 980/ 1200] Overall Loss 0.313384 Objective Loss 0.313384 LR 0.001000 Time 0.019832 -2022-12-06 10:30:55,824 - Epoch: [19][ 990/ 1200] Overall Loss 0.313486 Objective Loss 0.313486 LR 0.001000 Time 0.019825 -2022-12-06 10:30:56,016 - Epoch: [19][ 1000/ 1200] Overall Loss 0.313609 Objective Loss 0.313609 LR 0.001000 Time 0.019819 -2022-12-06 10:30:56,209 - Epoch: [19][ 1010/ 1200] Overall Loss 0.313946 Objective Loss 0.313946 LR 0.001000 Time 0.019813 -2022-12-06 10:30:56,401 - Epoch: [19][ 1020/ 1200] Overall Loss 0.313987 Objective Loss 0.313987 LR 0.001000 Time 0.019807 -2022-12-06 10:30:56,593 - Epoch: [19][ 1030/ 1200] Overall Loss 0.313857 Objective Loss 0.313857 LR 0.001000 Time 0.019800 -2022-12-06 10:30:56,785 - Epoch: [19][ 1040/ 1200] Overall Loss 0.313822 Objective Loss 0.313822 LR 0.001000 Time 0.019794 -2022-12-06 10:30:56,977 - Epoch: [19][ 1050/ 1200] Overall Loss 0.313763 Objective Loss 0.313763 LR 0.001000 Time 0.019788 -2022-12-06 10:30:57,169 - Epoch: [19][ 1060/ 1200] Overall Loss 0.313857 Objective Loss 0.313857 LR 0.001000 Time 0.019782 -2022-12-06 10:30:57,361 - Epoch: [19][ 1070/ 1200] Overall Loss 0.314223 Objective Loss 0.314223 LR 0.001000 Time 0.019776 -2022-12-06 10:30:57,553 - Epoch: [19][ 1080/ 1200] Overall Loss 0.314191 Objective Loss 0.314191 LR 0.001000 Time 0.019771 -2022-12-06 10:30:57,745 - Epoch: [19][ 1090/ 1200] Overall Loss 0.314474 Objective Loss 0.314474 LR 0.001000 Time 0.019765 -2022-12-06 10:30:57,938 - Epoch: [19][ 1100/ 1200] Overall Loss 0.314777 Objective Loss 0.314777 LR 0.001000 Time 0.019759 -2022-12-06 10:30:58,130 - Epoch: [19][ 1110/ 1200] Overall Loss 0.314994 Objective Loss 0.314994 LR 0.001000 Time 0.019754 -2022-12-06 10:30:58,322 - Epoch: [19][ 1120/ 1200] Overall Loss 0.314807 Objective Loss 0.314807 LR 0.001000 Time 0.019749 -2022-12-06 10:30:58,515 - Epoch: [19][ 1130/ 1200] Overall Loss 0.314864 Objective Loss 0.314864 LR 0.001000 Time 0.019744 -2022-12-06 10:30:58,705 - Epoch: [19][ 1140/ 1200] Overall Loss 0.315066 Objective Loss 0.315066 LR 0.001000 Time 0.019737 -2022-12-06 10:30:58,896 - Epoch: [19][ 1150/ 1200] Overall Loss 0.314917 Objective Loss 0.314917 LR 0.001000 Time 0.019731 -2022-12-06 10:30:59,086 - Epoch: [19][ 1160/ 1200] Overall Loss 0.314814 Objective Loss 0.314814 LR 0.001000 Time 0.019725 -2022-12-06 10:30:59,276 - Epoch: [19][ 1170/ 1200] Overall Loss 0.314700 Objective Loss 0.314700 LR 0.001000 Time 0.019718 -2022-12-06 10:30:59,467 - Epoch: [19][ 1180/ 1200] Overall Loss 0.314699 Objective Loss 0.314699 LR 0.001000 Time 0.019713 -2022-12-06 10:30:59,657 - Epoch: [19][ 1190/ 1200] Overall Loss 0.314757 Objective Loss 0.314757 LR 0.001000 Time 0.019706 -2022-12-06 10:30:59,889 - Epoch: [19][ 1200/ 1200] Overall Loss 0.314732 Objective Loss 0.314732 Top1 79.916318 Top5 97.907950 LR 0.001000 Time 0.019735 -2022-12-06 10:30:59,977 - --- validate (epoch=19)----------- -2022-12-06 10:30:59,977 - 34129 samples (256 per mini-batch) -2022-12-06 10:31:00,421 - Epoch: [19][ 10/ 134] Loss 0.336194 Top1 82.148438 Top5 97.578125 -2022-12-06 10:31:00,551 - Epoch: [19][ 20/ 134] Loss 0.345129 Top1 82.714844 Top5 97.675781 -2022-12-06 10:31:00,678 - Epoch: [19][ 30/ 134] Loss 0.330041 Top1 82.890625 Top5 97.825521 -2022-12-06 10:31:00,806 - Epoch: [19][ 40/ 134] Loss 0.335232 Top1 82.509766 Top5 97.802734 -2022-12-06 10:31:00,935 - Epoch: [19][ 50/ 134] Loss 0.337540 Top1 82.281250 Top5 97.789062 -2022-12-06 10:31:01,069 - Epoch: [19][ 60/ 134] Loss 0.337749 Top1 82.376302 Top5 97.695312 -2022-12-06 10:31:01,198 - Epoch: [19][ 70/ 134] Loss 0.337927 Top1 82.265625 Top5 97.667411 -2022-12-06 10:31:01,332 - Epoch: [19][ 80/ 134] Loss 0.340956 Top1 82.221680 Top5 97.695312 -2022-12-06 10:31:01,461 - Epoch: [19][ 90/ 134] Loss 0.337529 Top1 82.322049 Top5 97.690972 -2022-12-06 10:31:01,595 - Epoch: [19][ 100/ 134] Loss 0.337850 Top1 82.300781 Top5 97.683594 -2022-12-06 10:31:01,724 - Epoch: [19][ 110/ 134] Loss 0.338254 Top1 82.418324 Top5 97.642045 -2022-12-06 10:31:01,858 - Epoch: [19][ 120/ 134] Loss 0.338457 Top1 82.395833 Top5 97.600911 -2022-12-06 10:31:01,988 - Epoch: [19][ 130/ 134] Loss 0.338797 Top1 82.370793 Top5 97.599159 -2022-12-06 10:31:02,026 - Epoch: [19][ 134/ 134] Loss 0.337775 Top1 82.366902 Top5 97.603211 -2022-12-06 10:31:02,113 - ==> Top1: 82.367 Top5: 97.603 Loss: 0.338 - -2022-12-06 10:31:02,114 - ==> Confusion: -[[ 876 3 3 2 7 3 0 2 9 66 0 2 1 1 11 1 3 0 2 0 4] - [ 0 899 3 3 14 36 4 15 0 0 7 2 7 1 1 3 7 0 11 5 9] - [ 5 3 1000 15 1 1 29 7 0 0 5 4 3 2 3 3 4 2 6 3 7] - [ 3 4 34 918 0 2 1 2 1 0 10 4 6 1 9 0 3 3 11 1 7] - [ 11 4 5 3 939 5 1 2 0 3 1 3 1 4 11 5 12 3 2 1 4] - [ 4 18 0 11 6 944 5 11 4 1 0 23 11 7 2 0 5 0 4 4 9] - [ 0 3 18 2 0 1 1056 1 0 0 2 5 0 0 0 5 1 1 1 17 5] - [ 0 18 24 4 2 37 6 880 0 1 0 10 1 1 1 1 0 1 45 15 7] - [ 3 4 0 1 0 2 1 2 951 47 12 2 3 4 16 0 4 2 4 2 4] - [ 64 0 4 1 5 2 0 4 23 874 2 0 0 8 4 1 1 0 2 0 6] - [ 0 4 7 15 2 2 1 4 8 2 936 4 4 17 4 0 2 0 4 1 2] - [ 4 4 3 0 0 10 3 4 1 0 0 969 32 1 1 2 3 2 3 5 4] - [ 0 2 0 2 0 0 1 3 1 0 1 65 857 0 1 5 3 11 1 4 12] - [ 2 4 2 1 3 21 0 2 10 18 12 25 10 880 4 3 7 1 0 5 13] - [ 6 4 1 20 4 3 0 1 11 2 3 2 5 2 1026 0 7 6 13 1 13] - [ 1 3 5 1 1 1 5 0 0 0 2 14 7 0 0 966 15 7 0 4 11] - [ 2 5 3 3 2 0 1 0 0 0 0 8 3 0 2 9 1017 3 1 8 5] - [ 3 3 1 4 0 1 2 1 1 1 1 19 32 1 1 7 3 947 2 2 4] - [ 3 5 10 16 0 2 0 17 2 0 14 5 6 2 4 0 1 0 914 2 5] - [ 1 5 3 0 0 6 11 7 0 0 0 36 7 1 0 2 5 3 4 983 6] - [ 150 261 245 131 163 218 89 114 87 109 220 198 453 253 182 129 366 76 214 295 9273]] - -2022-12-06 10:31:02,773 - ==> Best [Top1: 82.847 Top5: 97.603 Sparsity:0.00 Params: 5376 on epoch: 18] -2022-12-06 10:31:02,773 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:31:02,779 - - -2022-12-06 10:31:02,779 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:31:03,714 - Epoch: [20][ 10/ 1200] Overall Loss 0.333310 Objective Loss 0.333310 LR 0.001000 Time 0.093457 -2022-12-06 10:31:03,915 - Epoch: [20][ 20/ 1200] Overall Loss 0.334331 Objective Loss 0.334331 LR 0.001000 Time 0.056771 -2022-12-06 10:31:04,107 - Epoch: [20][ 30/ 1200] Overall Loss 0.341550 Objective Loss 0.341550 LR 0.001000 Time 0.044226 -2022-12-06 10:31:04,300 - Epoch: [20][ 40/ 1200] Overall Loss 0.339234 Objective Loss 0.339234 LR 0.001000 Time 0.037960 -2022-12-06 10:31:04,490 - Epoch: [20][ 50/ 1200] Overall Loss 0.331024 Objective Loss 0.331024 LR 0.001000 Time 0.034169 -2022-12-06 10:31:04,680 - Epoch: [20][ 60/ 1200] Overall Loss 0.323220 Objective Loss 0.323220 LR 0.001000 Time 0.031632 -2022-12-06 10:31:04,869 - Epoch: [20][ 70/ 1200] Overall Loss 0.325644 Objective Loss 0.325644 LR 0.001000 Time 0.029804 -2022-12-06 10:31:05,059 - Epoch: [20][ 80/ 1200] Overall Loss 0.320523 Objective Loss 0.320523 LR 0.001000 Time 0.028445 -2022-12-06 10:31:05,251 - Epoch: [20][ 90/ 1200] Overall Loss 0.320959 Objective Loss 0.320959 LR 0.001000 Time 0.027407 -2022-12-06 10:31:05,440 - Epoch: [20][ 100/ 1200] Overall Loss 0.315914 Objective Loss 0.315914 LR 0.001000 Time 0.026556 -2022-12-06 10:31:05,631 - Epoch: [20][ 110/ 1200] Overall Loss 0.313463 Objective Loss 0.313463 LR 0.001000 Time 0.025869 -2022-12-06 10:31:05,822 - Epoch: [20][ 120/ 1200] Overall Loss 0.314125 Objective Loss 0.314125 LR 0.001000 Time 0.025301 -2022-12-06 10:31:06,012 - Epoch: [20][ 130/ 1200] Overall Loss 0.314067 Objective Loss 0.314067 LR 0.001000 Time 0.024814 -2022-12-06 10:31:06,202 - Epoch: [20][ 140/ 1200] Overall Loss 0.314678 Objective Loss 0.314678 LR 0.001000 Time 0.024395 -2022-12-06 10:31:06,392 - Epoch: [20][ 150/ 1200] Overall Loss 0.315726 Objective Loss 0.315726 LR 0.001000 Time 0.024035 -2022-12-06 10:31:06,584 - Epoch: [20][ 160/ 1200] Overall Loss 0.315465 Objective Loss 0.315465 LR 0.001000 Time 0.023726 -2022-12-06 10:31:06,775 - Epoch: [20][ 170/ 1200] Overall Loss 0.313835 Objective Loss 0.313835 LR 0.001000 Time 0.023449 -2022-12-06 10:31:06,965 - Epoch: [20][ 180/ 1200] Overall Loss 0.312625 Objective Loss 0.312625 LR 0.001000 Time 0.023203 -2022-12-06 10:31:07,156 - Epoch: [20][ 190/ 1200] Overall Loss 0.313279 Objective Loss 0.313279 LR 0.001000 Time 0.022982 -2022-12-06 10:31:07,347 - Epoch: [20][ 200/ 1200] Overall Loss 0.314021 Objective Loss 0.314021 LR 0.001000 Time 0.022783 -2022-12-06 10:31:07,539 - Epoch: [20][ 210/ 1200] Overall Loss 0.313312 Objective Loss 0.313312 LR 0.001000 Time 0.022611 -2022-12-06 10:31:07,728 - Epoch: [20][ 220/ 1200] Overall Loss 0.311900 Objective Loss 0.311900 LR 0.001000 Time 0.022442 -2022-12-06 10:31:07,919 - Epoch: [20][ 230/ 1200] Overall Loss 0.311843 Objective Loss 0.311843 LR 0.001000 Time 0.022295 -2022-12-06 10:31:08,110 - Epoch: [20][ 240/ 1200] Overall Loss 0.310314 Objective Loss 0.310314 LR 0.001000 Time 0.022158 -2022-12-06 10:31:08,301 - Epoch: [20][ 250/ 1200] Overall Loss 0.308074 Objective Loss 0.308074 LR 0.001000 Time 0.022031 -2022-12-06 10:31:08,491 - Epoch: [20][ 260/ 1200] Overall Loss 0.307536 Objective Loss 0.307536 LR 0.001000 Time 0.021914 -2022-12-06 10:31:08,681 - Epoch: [20][ 270/ 1200] Overall Loss 0.307584 Objective Loss 0.307584 LR 0.001000 Time 0.021805 -2022-12-06 10:31:08,871 - Epoch: [20][ 280/ 1200] Overall Loss 0.306945 Objective Loss 0.306945 LR 0.001000 Time 0.021702 -2022-12-06 10:31:09,062 - Epoch: [20][ 290/ 1200] Overall Loss 0.306940 Objective Loss 0.306940 LR 0.001000 Time 0.021611 -2022-12-06 10:31:09,252 - Epoch: [20][ 300/ 1200] Overall Loss 0.306555 Objective Loss 0.306555 LR 0.001000 Time 0.021521 -2022-12-06 10:31:09,441 - Epoch: [20][ 310/ 1200] Overall Loss 0.305849 Objective Loss 0.305849 LR 0.001000 Time 0.021436 -2022-12-06 10:31:09,632 - Epoch: [20][ 320/ 1200] Overall Loss 0.305712 Objective Loss 0.305712 LR 0.001000 Time 0.021360 -2022-12-06 10:31:09,822 - Epoch: [20][ 330/ 1200] Overall Loss 0.305295 Objective Loss 0.305295 LR 0.001000 Time 0.021287 -2022-12-06 10:31:10,012 - Epoch: [20][ 340/ 1200] Overall Loss 0.305903 Objective Loss 0.305903 LR 0.001000 Time 0.021218 -2022-12-06 10:31:10,203 - Epoch: [20][ 350/ 1200] Overall Loss 0.306691 Objective Loss 0.306691 LR 0.001000 Time 0.021156 -2022-12-06 10:31:10,394 - Epoch: [20][ 360/ 1200] Overall Loss 0.306511 Objective Loss 0.306511 LR 0.001000 Time 0.021098 -2022-12-06 10:31:10,584 - Epoch: [20][ 370/ 1200] Overall Loss 0.305999 Objective Loss 0.305999 LR 0.001000 Time 0.021040 -2022-12-06 10:31:10,775 - Epoch: [20][ 380/ 1200] Overall Loss 0.305793 Objective Loss 0.305793 LR 0.001000 Time 0.020988 -2022-12-06 10:31:10,967 - Epoch: [20][ 390/ 1200] Overall Loss 0.305260 Objective Loss 0.305260 LR 0.001000 Time 0.020939 -2022-12-06 10:31:11,156 - Epoch: [20][ 400/ 1200] Overall Loss 0.305720 Objective Loss 0.305720 LR 0.001000 Time 0.020888 -2022-12-06 10:31:11,347 - Epoch: [20][ 410/ 1200] Overall Loss 0.305057 Objective Loss 0.305057 LR 0.001000 Time 0.020843 -2022-12-06 10:31:11,539 - Epoch: [20][ 420/ 1200] Overall Loss 0.305292 Objective Loss 0.305292 LR 0.001000 Time 0.020801 -2022-12-06 10:31:11,730 - Epoch: [20][ 430/ 1200] Overall Loss 0.305524 Objective Loss 0.305524 LR 0.001000 Time 0.020760 -2022-12-06 10:31:11,920 - Epoch: [20][ 440/ 1200] Overall Loss 0.306233 Objective Loss 0.306233 LR 0.001000 Time 0.020719 -2022-12-06 10:31:12,111 - Epoch: [20][ 450/ 1200] Overall Loss 0.306552 Objective Loss 0.306552 LR 0.001000 Time 0.020682 -2022-12-06 10:31:12,302 - Epoch: [20][ 460/ 1200] Overall Loss 0.307156 Objective Loss 0.307156 LR 0.001000 Time 0.020645 -2022-12-06 10:31:12,492 - Epoch: [20][ 470/ 1200] Overall Loss 0.307477 Objective Loss 0.307477 LR 0.001000 Time 0.020610 -2022-12-06 10:31:12,683 - Epoch: [20][ 480/ 1200] Overall Loss 0.307696 Objective Loss 0.307696 LR 0.001000 Time 0.020577 -2022-12-06 10:31:12,874 - Epoch: [20][ 490/ 1200] Overall Loss 0.307498 Objective Loss 0.307498 LR 0.001000 Time 0.020547 -2022-12-06 10:31:13,065 - Epoch: [20][ 500/ 1200] Overall Loss 0.307597 Objective Loss 0.307597 LR 0.001000 Time 0.020516 -2022-12-06 10:31:13,256 - Epoch: [20][ 510/ 1200] Overall Loss 0.307453 Objective Loss 0.307453 LR 0.001000 Time 0.020487 -2022-12-06 10:31:13,447 - Epoch: [20][ 520/ 1200] Overall Loss 0.307540 Objective Loss 0.307540 LR 0.001000 Time 0.020460 -2022-12-06 10:31:13,639 - Epoch: [20][ 530/ 1200] Overall Loss 0.307761 Objective Loss 0.307761 LR 0.001000 Time 0.020434 -2022-12-06 10:31:13,830 - Epoch: [20][ 540/ 1200] Overall Loss 0.307603 Objective Loss 0.307603 LR 0.001000 Time 0.020408 -2022-12-06 10:31:14,021 - Epoch: [20][ 550/ 1200] Overall Loss 0.307812 Objective Loss 0.307812 LR 0.001000 Time 0.020384 -2022-12-06 10:31:14,212 - Epoch: [20][ 560/ 1200] Overall Loss 0.308074 Objective Loss 0.308074 LR 0.001000 Time 0.020361 -2022-12-06 10:31:14,404 - Epoch: [20][ 570/ 1200] Overall Loss 0.307719 Objective Loss 0.307719 LR 0.001000 Time 0.020338 -2022-12-06 10:31:14,595 - Epoch: [20][ 580/ 1200] Overall Loss 0.307623 Objective Loss 0.307623 LR 0.001000 Time 0.020317 -2022-12-06 10:31:14,786 - Epoch: [20][ 590/ 1200] Overall Loss 0.307905 Objective Loss 0.307905 LR 0.001000 Time 0.020296 -2022-12-06 10:31:14,977 - Epoch: [20][ 600/ 1200] Overall Loss 0.308187 Objective Loss 0.308187 LR 0.001000 Time 0.020273 -2022-12-06 10:31:15,168 - Epoch: [20][ 610/ 1200] Overall Loss 0.308457 Objective Loss 0.308457 LR 0.001000 Time 0.020253 -2022-12-06 10:31:15,358 - Epoch: [20][ 620/ 1200] Overall Loss 0.308342 Objective Loss 0.308342 LR 0.001000 Time 0.020233 -2022-12-06 10:31:15,550 - Epoch: [20][ 630/ 1200] Overall Loss 0.307813 Objective Loss 0.307813 LR 0.001000 Time 0.020215 -2022-12-06 10:31:15,740 - Epoch: [20][ 640/ 1200] Overall Loss 0.307685 Objective Loss 0.307685 LR 0.001000 Time 0.020195 -2022-12-06 10:31:15,931 - Epoch: [20][ 650/ 1200] Overall Loss 0.307567 Objective Loss 0.307567 LR 0.001000 Time 0.020178 -2022-12-06 10:31:16,122 - Epoch: [20][ 660/ 1200] Overall Loss 0.307495 Objective Loss 0.307495 LR 0.001000 Time 0.020161 -2022-12-06 10:31:16,314 - Epoch: [20][ 670/ 1200] Overall Loss 0.308032 Objective Loss 0.308032 LR 0.001000 Time 0.020145 -2022-12-06 10:31:16,504 - Epoch: [20][ 680/ 1200] Overall Loss 0.308202 Objective Loss 0.308202 LR 0.001000 Time 0.020128 -2022-12-06 10:31:16,695 - Epoch: [20][ 690/ 1200] Overall Loss 0.308099 Objective Loss 0.308099 LR 0.001000 Time 0.020112 -2022-12-06 10:31:16,886 - Epoch: [20][ 700/ 1200] Overall Loss 0.307990 Objective Loss 0.307990 LR 0.001000 Time 0.020097 -2022-12-06 10:31:17,078 - Epoch: [20][ 710/ 1200] Overall Loss 0.307948 Objective Loss 0.307948 LR 0.001000 Time 0.020084 -2022-12-06 10:31:17,269 - Epoch: [20][ 720/ 1200] Overall Loss 0.307927 Objective Loss 0.307927 LR 0.001000 Time 0.020068 -2022-12-06 10:31:17,460 - Epoch: [20][ 730/ 1200] Overall Loss 0.308397 Objective Loss 0.308397 LR 0.001000 Time 0.020054 -2022-12-06 10:31:17,651 - Epoch: [20][ 740/ 1200] Overall Loss 0.308783 Objective Loss 0.308783 LR 0.001000 Time 0.020042 -2022-12-06 10:31:17,843 - Epoch: [20][ 750/ 1200] Overall Loss 0.309011 Objective Loss 0.309011 LR 0.001000 Time 0.020030 -2022-12-06 10:31:18,034 - Epoch: [20][ 760/ 1200] Overall Loss 0.309110 Objective Loss 0.309110 LR 0.001000 Time 0.020016 -2022-12-06 10:31:18,226 - Epoch: [20][ 770/ 1200] Overall Loss 0.309131 Objective Loss 0.309131 LR 0.001000 Time 0.020005 -2022-12-06 10:31:18,417 - Epoch: [20][ 780/ 1200] Overall Loss 0.309229 Objective Loss 0.309229 LR 0.001000 Time 0.019992 -2022-12-06 10:31:18,610 - Epoch: [20][ 790/ 1200] Overall Loss 0.309339 Objective Loss 0.309339 LR 0.001000 Time 0.019982 -2022-12-06 10:31:18,803 - Epoch: [20][ 800/ 1200] Overall Loss 0.309491 Objective Loss 0.309491 LR 0.001000 Time 0.019973 -2022-12-06 10:31:18,995 - Epoch: [20][ 810/ 1200] Overall Loss 0.309524 Objective Loss 0.309524 LR 0.001000 Time 0.019963 -2022-12-06 10:31:19,189 - Epoch: [20][ 820/ 1200] Overall Loss 0.309306 Objective Loss 0.309306 LR 0.001000 Time 0.019956 -2022-12-06 10:31:19,381 - Epoch: [20][ 830/ 1200] Overall Loss 0.309202 Objective Loss 0.309202 LR 0.001000 Time 0.019947 -2022-12-06 10:31:19,575 - Epoch: [20][ 840/ 1200] Overall Loss 0.309619 Objective Loss 0.309619 LR 0.001000 Time 0.019939 -2022-12-06 10:31:19,768 - Epoch: [20][ 850/ 1200] Overall Loss 0.309572 Objective Loss 0.309572 LR 0.001000 Time 0.019931 -2022-12-06 10:31:19,961 - Epoch: [20][ 860/ 1200] Overall Loss 0.309842 Objective Loss 0.309842 LR 0.001000 Time 0.019923 -2022-12-06 10:31:20,155 - Epoch: [20][ 870/ 1200] Overall Loss 0.312239 Objective Loss 0.312239 LR 0.001000 Time 0.019916 -2022-12-06 10:31:20,348 - Epoch: [20][ 880/ 1200] Overall Loss 0.314415 Objective Loss 0.314415 LR 0.001000 Time 0.019909 -2022-12-06 10:31:20,541 - Epoch: [20][ 890/ 1200] Overall Loss 0.316034 Objective Loss 0.316034 LR 0.001000 Time 0.019901 -2022-12-06 10:31:20,734 - Epoch: [20][ 900/ 1200] Overall Loss 0.317684 Objective Loss 0.317684 LR 0.001000 Time 0.019894 -2022-12-06 10:31:20,927 - Epoch: [20][ 910/ 1200] Overall Loss 0.319133 Objective Loss 0.319133 LR 0.001000 Time 0.019887 -2022-12-06 10:31:21,121 - Epoch: [20][ 920/ 1200] Overall Loss 0.320310 Objective Loss 0.320310 LR 0.001000 Time 0.019881 -2022-12-06 10:31:21,312 - Epoch: [20][ 930/ 1200] Overall Loss 0.321855 Objective Loss 0.321855 LR 0.001000 Time 0.019872 -2022-12-06 10:31:21,506 - Epoch: [20][ 940/ 1200] Overall Loss 0.322991 Objective Loss 0.322991 LR 0.001000 Time 0.019866 -2022-12-06 10:31:21,699 - Epoch: [20][ 950/ 1200] Overall Loss 0.323939 Objective Loss 0.323939 LR 0.001000 Time 0.019860 -2022-12-06 10:31:21,892 - Epoch: [20][ 960/ 1200] Overall Loss 0.324905 Objective Loss 0.324905 LR 0.001000 Time 0.019854 -2022-12-06 10:31:22,085 - Epoch: [20][ 970/ 1200] Overall Loss 0.326062 Objective Loss 0.326062 LR 0.001000 Time 0.019847 -2022-12-06 10:31:22,278 - Epoch: [20][ 980/ 1200] Overall Loss 0.326844 Objective Loss 0.326844 LR 0.001000 Time 0.019841 -2022-12-06 10:31:22,470 - Epoch: [20][ 990/ 1200] Overall Loss 0.327710 Objective Loss 0.327710 LR 0.001000 Time 0.019834 -2022-12-06 10:31:22,664 - Epoch: [20][ 1000/ 1200] Overall Loss 0.328322 Objective Loss 0.328322 LR 0.001000 Time 0.019829 -2022-12-06 10:31:22,856 - Epoch: [20][ 1010/ 1200] Overall Loss 0.328963 Objective Loss 0.328963 LR 0.001000 Time 0.019822 -2022-12-06 10:31:23,049 - Epoch: [20][ 1020/ 1200] Overall Loss 0.329534 Objective Loss 0.329534 LR 0.001000 Time 0.019816 -2022-12-06 10:31:23,242 - Epoch: [20][ 1030/ 1200] Overall Loss 0.330191 Objective Loss 0.330191 LR 0.001000 Time 0.019811 -2022-12-06 10:31:23,437 - Epoch: [20][ 1040/ 1200] Overall Loss 0.330889 Objective Loss 0.330889 LR 0.001000 Time 0.019808 -2022-12-06 10:31:23,632 - Epoch: [20][ 1050/ 1200] Overall Loss 0.331514 Objective Loss 0.331514 LR 0.001000 Time 0.019804 -2022-12-06 10:31:23,827 - Epoch: [20][ 1060/ 1200] Overall Loss 0.332071 Objective Loss 0.332071 LR 0.001000 Time 0.019800 -2022-12-06 10:31:24,021 - Epoch: [20][ 1070/ 1200] Overall Loss 0.332794 Objective Loss 0.332794 LR 0.001000 Time 0.019796 -2022-12-06 10:31:24,216 - Epoch: [20][ 1080/ 1200] Overall Loss 0.333420 Objective Loss 0.333420 LR 0.001000 Time 0.019793 -2022-12-06 10:31:24,410 - Epoch: [20][ 1090/ 1200] Overall Loss 0.334157 Objective Loss 0.334157 LR 0.001000 Time 0.019789 -2022-12-06 10:31:24,605 - Epoch: [20][ 1100/ 1200] Overall Loss 0.334767 Objective Loss 0.334767 LR 0.001000 Time 0.019786 -2022-12-06 10:31:24,799 - Epoch: [20][ 1110/ 1200] Overall Loss 0.335432 Objective Loss 0.335432 LR 0.001000 Time 0.019782 -2022-12-06 10:31:24,994 - Epoch: [20][ 1120/ 1200] Overall Loss 0.336308 Objective Loss 0.336308 LR 0.001000 Time 0.019779 -2022-12-06 10:31:25,188 - Epoch: [20][ 1130/ 1200] Overall Loss 0.336745 Objective Loss 0.336745 LR 0.001000 Time 0.019775 -2022-12-06 10:31:25,383 - Epoch: [20][ 1140/ 1200] Overall Loss 0.337129 Objective Loss 0.337129 LR 0.001000 Time 0.019772 -2022-12-06 10:31:25,577 - Epoch: [20][ 1150/ 1200] Overall Loss 0.337371 Objective Loss 0.337371 LR 0.001000 Time 0.019769 -2022-12-06 10:31:25,772 - Epoch: [20][ 1160/ 1200] Overall Loss 0.337793 Objective Loss 0.337793 LR 0.001000 Time 0.019766 -2022-12-06 10:31:25,966 - Epoch: [20][ 1170/ 1200] Overall Loss 0.338248 Objective Loss 0.338248 LR 0.001000 Time 0.019762 -2022-12-06 10:31:26,161 - Epoch: [20][ 1180/ 1200] Overall Loss 0.338694 Objective Loss 0.338694 LR 0.001000 Time 0.019760 -2022-12-06 10:31:26,355 - Epoch: [20][ 1190/ 1200] Overall Loss 0.339273 Objective Loss 0.339273 LR 0.001000 Time 0.019756 -2022-12-06 10:31:26,581 - Epoch: [20][ 1200/ 1200] Overall Loss 0.339429 Objective Loss 0.339429 Top1 83.263598 Top5 97.280335 LR 0.001000 Time 0.019779 -2022-12-06 10:31:26,669 - --- validate (epoch=20)----------- -2022-12-06 10:31:26,669 - 34129 samples (256 per mini-batch) -2022-12-06 10:31:27,115 - Epoch: [20][ 10/ 134] Loss 0.396750 Top1 82.265625 Top5 97.304688 -2022-12-06 10:31:27,248 - Epoch: [20][ 20/ 134] Loss 0.397737 Top1 82.285156 Top5 97.207031 -2022-12-06 10:31:27,381 - Epoch: [20][ 30/ 134] Loss 0.386368 Top1 82.630208 Top5 97.252604 -2022-12-06 10:31:27,512 - Epoch: [20][ 40/ 134] Loss 0.384768 Top1 82.744141 Top5 97.451172 -2022-12-06 10:31:27,645 - Epoch: [20][ 50/ 134] Loss 0.386863 Top1 82.710938 Top5 97.500000 -2022-12-06 10:31:27,776 - Epoch: [20][ 60/ 134] Loss 0.385165 Top1 82.714844 Top5 97.545573 -2022-12-06 10:31:27,908 - Epoch: [20][ 70/ 134] Loss 0.382801 Top1 82.606027 Top5 97.583705 -2022-12-06 10:31:28,039 - Epoch: [20][ 80/ 134] Loss 0.383281 Top1 82.617188 Top5 97.573242 -2022-12-06 10:31:28,171 - Epoch: [20][ 90/ 134] Loss 0.383319 Top1 82.634549 Top5 97.604167 -2022-12-06 10:31:28,303 - Epoch: [20][ 100/ 134] Loss 0.380257 Top1 82.640625 Top5 97.621094 -2022-12-06 10:31:28,436 - Epoch: [20][ 110/ 134] Loss 0.383235 Top1 82.539062 Top5 97.613636 -2022-12-06 10:31:28,567 - Epoch: [20][ 120/ 134] Loss 0.381097 Top1 82.659505 Top5 97.607422 -2022-12-06 10:31:28,700 - Epoch: [20][ 130/ 134] Loss 0.381813 Top1 82.617188 Top5 97.563101 -2022-12-06 10:31:28,739 - Epoch: [20][ 134/ 134] Loss 0.380315 Top1 82.645258 Top5 97.570981 -2022-12-06 10:31:28,829 - ==> Top1: 82.645 Top5: 97.571 Loss: 0.380 - -2022-12-06 10:31:28,829 - ==> Confusion: -[[ 873 2 6 2 11 2 0 0 6 67 1 6 2 1 4 1 1 1 1 0 9] - [ 1 928 4 3 11 25 1 18 0 0 2 3 2 0 4 2 4 1 12 3 3] - [ 5 4 1010 11 4 1 22 8 0 1 4 6 1 2 1 2 2 1 5 3 10] - [ 2 2 34 921 0 1 1 2 1 0 7 2 5 1 20 0 2 2 12 0 5] - [ 8 6 2 1 951 7 0 2 0 2 0 5 2 3 7 8 9 0 2 0 5] - [ 1 32 0 1 7 949 3 18 3 1 1 15 6 14 2 2 4 0 3 4 3] - [ 0 2 14 2 0 3 1066 4 2 0 1 2 1 0 0 7 0 1 0 11 2] - [ 1 11 18 2 3 32 5 909 0 1 3 9 1 1 1 3 0 0 37 14 3] - [ 4 3 1 2 0 5 0 0 957 45 10 2 4 8 11 1 4 0 3 2 2] - [ 62 0 5 0 5 2 0 2 20 874 1 2 0 16 4 1 0 2 0 0 5] - [ 0 3 6 10 0 3 0 2 4 3 943 4 2 15 10 0 1 0 11 0 2] - [ 4 3 3 0 0 9 2 5 0 1 0 976 26 7 0 4 2 3 1 4 1] - [ 0 1 1 5 1 3 0 2 1 0 0 38 886 0 0 9 4 8 0 4 6] - [ 2 2 1 1 1 7 0 2 6 14 8 8 4 944 3 3 7 0 0 2 8] - [ 6 3 2 14 8 2 0 1 16 4 1 1 4 5 1050 0 3 0 4 2 4] - [ 2 3 2 0 4 0 0 0 0 0 1 13 8 1 2 976 12 6 1 5 7] - [ 2 6 2 1 5 2 2 1 0 0 0 6 1 1 0 7 1023 3 1 3 6] - [ 2 0 1 5 0 0 1 1 1 1 0 16 19 2 1 12 3 964 2 2 3] - [ 3 4 4 12 0 3 0 23 2 0 6 4 3 2 12 0 0 1 922 1 6] - [ 0 2 3 0 1 11 9 8 0 0 1 37 3 2 3 2 7 2 1 983 5] - [ 122 366 318 138 153 210 98 136 78 104 229 169 475 398 247 119 205 60 244 265 9092]] - -2022-12-06 10:31:29,485 - ==> Best [Top1: 82.847 Top5: 97.603 Sparsity:0.00 Params: 5376 on epoch: 18] -2022-12-06 10:31:29,485 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:31:29,491 - - -2022-12-06 10:31:29,491 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:31:30,424 - Epoch: [21][ 10/ 1200] Overall Loss 0.370830 Objective Loss 0.370830 LR 0.001000 Time 0.093184 -2022-12-06 10:31:30,630 - Epoch: [21][ 20/ 1200] Overall Loss 0.364587 Objective Loss 0.364587 LR 0.001000 Time 0.056850 -2022-12-06 10:31:30,826 - Epoch: [21][ 30/ 1200] Overall Loss 0.364232 Objective Loss 0.364232 LR 0.001000 Time 0.044422 -2022-12-06 10:31:31,024 - Epoch: [21][ 40/ 1200] Overall Loss 0.362351 Objective Loss 0.362351 LR 0.001000 Time 0.038249 -2022-12-06 10:31:31,220 - Epoch: [21][ 50/ 1200] Overall Loss 0.360507 Objective Loss 0.360507 LR 0.001000 Time 0.034505 -2022-12-06 10:31:31,417 - Epoch: [21][ 60/ 1200] Overall Loss 0.362772 Objective Loss 0.362772 LR 0.001000 Time 0.032038 -2022-12-06 10:31:31,613 - Epoch: [21][ 70/ 1200] Overall Loss 0.361257 Objective Loss 0.361257 LR 0.001000 Time 0.030246 -2022-12-06 10:31:31,811 - Epoch: [21][ 80/ 1200] Overall Loss 0.358770 Objective Loss 0.358770 LR 0.001000 Time 0.028937 -2022-12-06 10:31:32,007 - Epoch: [21][ 90/ 1200] Overall Loss 0.360505 Objective Loss 0.360505 LR 0.001000 Time 0.027889 -2022-12-06 10:31:32,204 - Epoch: [21][ 100/ 1200] Overall Loss 0.360483 Objective Loss 0.360483 LR 0.001000 Time 0.027068 -2022-12-06 10:31:32,399 - Epoch: [21][ 110/ 1200] Overall Loss 0.359255 Objective Loss 0.359255 LR 0.001000 Time 0.026380 -2022-12-06 10:31:32,597 - Epoch: [21][ 120/ 1200] Overall Loss 0.359041 Objective Loss 0.359041 LR 0.001000 Time 0.025820 -2022-12-06 10:31:32,792 - Epoch: [21][ 130/ 1200] Overall Loss 0.357547 Objective Loss 0.357547 LR 0.001000 Time 0.025331 -2022-12-06 10:31:32,990 - Epoch: [21][ 140/ 1200] Overall Loss 0.359227 Objective Loss 0.359227 LR 0.001000 Time 0.024932 -2022-12-06 10:31:33,185 - Epoch: [21][ 150/ 1200] Overall Loss 0.359403 Objective Loss 0.359403 LR 0.001000 Time 0.024569 -2022-12-06 10:31:33,383 - Epoch: [21][ 160/ 1200] Overall Loss 0.361553 Objective Loss 0.361553 LR 0.001000 Time 0.024268 -2022-12-06 10:31:33,578 - Epoch: [21][ 170/ 1200] Overall Loss 0.361895 Objective Loss 0.361895 LR 0.001000 Time 0.023985 -2022-12-06 10:31:33,776 - Epoch: [21][ 180/ 1200] Overall Loss 0.362149 Objective Loss 0.362149 LR 0.001000 Time 0.023747 -2022-12-06 10:31:33,971 - Epoch: [21][ 190/ 1200] Overall Loss 0.362350 Objective Loss 0.362350 LR 0.001000 Time 0.023522 -2022-12-06 10:31:34,169 - Epoch: [21][ 200/ 1200] Overall Loss 0.361576 Objective Loss 0.361576 LR 0.001000 Time 0.023331 -2022-12-06 10:31:34,364 - Epoch: [21][ 210/ 1200] Overall Loss 0.361934 Objective Loss 0.361934 LR 0.001000 Time 0.023148 -2022-12-06 10:31:34,562 - Epoch: [21][ 220/ 1200] Overall Loss 0.361950 Objective Loss 0.361950 LR 0.001000 Time 0.022992 -2022-12-06 10:31:34,757 - Epoch: [21][ 230/ 1200] Overall Loss 0.361445 Objective Loss 0.361445 LR 0.001000 Time 0.022839 -2022-12-06 10:31:34,955 - Epoch: [21][ 240/ 1200] Overall Loss 0.361121 Objective Loss 0.361121 LR 0.001000 Time 0.022708 -2022-12-06 10:31:35,150 - Epoch: [21][ 250/ 1200] Overall Loss 0.361415 Objective Loss 0.361415 LR 0.001000 Time 0.022579 -2022-12-06 10:31:35,347 - Epoch: [21][ 260/ 1200] Overall Loss 0.360918 Objective Loss 0.360918 LR 0.001000 Time 0.022468 -2022-12-06 10:31:35,543 - Epoch: [21][ 270/ 1200] Overall Loss 0.360091 Objective Loss 0.360091 LR 0.001000 Time 0.022357 -2022-12-06 10:31:35,741 - Epoch: [21][ 280/ 1200] Overall Loss 0.359210 Objective Loss 0.359210 LR 0.001000 Time 0.022265 -2022-12-06 10:31:35,937 - Epoch: [21][ 290/ 1200] Overall Loss 0.359209 Objective Loss 0.359209 LR 0.001000 Time 0.022172 -2022-12-06 10:31:36,137 - Epoch: [21][ 300/ 1200] Overall Loss 0.358263 Objective Loss 0.358263 LR 0.001000 Time 0.022095 -2022-12-06 10:31:36,333 - Epoch: [21][ 310/ 1200] Overall Loss 0.357627 Objective Loss 0.357627 LR 0.001000 Time 0.022015 -2022-12-06 10:31:36,532 - Epoch: [21][ 320/ 1200] Overall Loss 0.357428 Objective Loss 0.357428 LR 0.001000 Time 0.021946 -2022-12-06 10:31:36,728 - Epoch: [21][ 330/ 1200] Overall Loss 0.357595 Objective Loss 0.357595 LR 0.001000 Time 0.021875 -2022-12-06 10:31:36,927 - Epoch: [21][ 340/ 1200] Overall Loss 0.357687 Objective Loss 0.357687 LR 0.001000 Time 0.021815 -2022-12-06 10:31:37,124 - Epoch: [21][ 350/ 1200] Overall Loss 0.356987 Objective Loss 0.356987 LR 0.001000 Time 0.021751 -2022-12-06 10:31:37,323 - Epoch: [21][ 360/ 1200] Overall Loss 0.357445 Objective Loss 0.357445 LR 0.001000 Time 0.021700 -2022-12-06 10:31:37,520 - Epoch: [21][ 370/ 1200] Overall Loss 0.358180 Objective Loss 0.358180 LR 0.001000 Time 0.021644 -2022-12-06 10:31:37,719 - Epoch: [21][ 380/ 1200] Overall Loss 0.357710 Objective Loss 0.357710 LR 0.001000 Time 0.021596 -2022-12-06 10:31:37,916 - Epoch: [21][ 390/ 1200] Overall Loss 0.356831 Objective Loss 0.356831 LR 0.001000 Time 0.021546 -2022-12-06 10:31:38,115 - Epoch: [21][ 400/ 1200] Overall Loss 0.356544 Objective Loss 0.356544 LR 0.001000 Time 0.021504 -2022-12-06 10:31:38,313 - Epoch: [21][ 410/ 1200] Overall Loss 0.356640 Objective Loss 0.356640 LR 0.001000 Time 0.021460 -2022-12-06 10:31:38,512 - Epoch: [21][ 420/ 1200] Overall Loss 0.356296 Objective Loss 0.356296 LR 0.001000 Time 0.021421 -2022-12-06 10:31:38,708 - Epoch: [21][ 430/ 1200] Overall Loss 0.356688 Objective Loss 0.356688 LR 0.001000 Time 0.021379 -2022-12-06 10:31:38,908 - Epoch: [21][ 440/ 1200] Overall Loss 0.356103 Objective Loss 0.356103 LR 0.001000 Time 0.021345 -2022-12-06 10:31:39,105 - Epoch: [21][ 450/ 1200] Overall Loss 0.356550 Objective Loss 0.356550 LR 0.001000 Time 0.021308 -2022-12-06 10:31:39,304 - Epoch: [21][ 460/ 1200] Overall Loss 0.356315 Objective Loss 0.356315 LR 0.001000 Time 0.021276 -2022-12-06 10:31:39,500 - Epoch: [21][ 470/ 1200] Overall Loss 0.356614 Objective Loss 0.356614 LR 0.001000 Time 0.021240 -2022-12-06 10:31:39,695 - Epoch: [21][ 480/ 1200] Overall Loss 0.356151 Objective Loss 0.356151 LR 0.001000 Time 0.021203 -2022-12-06 10:31:39,888 - Epoch: [21][ 490/ 1200] Overall Loss 0.356332 Objective Loss 0.356332 LR 0.001000 Time 0.021162 -2022-12-06 10:31:40,080 - Epoch: [21][ 500/ 1200] Overall Loss 0.356153 Objective Loss 0.356153 LR 0.001000 Time 0.021122 -2022-12-06 10:31:40,273 - Epoch: [21][ 510/ 1200] Overall Loss 0.355494 Objective Loss 0.355494 LR 0.001000 Time 0.021084 -2022-12-06 10:31:40,465 - Epoch: [21][ 520/ 1200] Overall Loss 0.355059 Objective Loss 0.355059 LR 0.001000 Time 0.021048 -2022-12-06 10:31:40,657 - Epoch: [21][ 530/ 1200] Overall Loss 0.354741 Objective Loss 0.354741 LR 0.001000 Time 0.021012 -2022-12-06 10:31:40,849 - Epoch: [21][ 540/ 1200] Overall Loss 0.354599 Objective Loss 0.354599 LR 0.001000 Time 0.020978 -2022-12-06 10:31:41,042 - Epoch: [21][ 550/ 1200] Overall Loss 0.354122 Objective Loss 0.354122 LR 0.001000 Time 0.020945 -2022-12-06 10:31:41,235 - Epoch: [21][ 560/ 1200] Overall Loss 0.353516 Objective Loss 0.353516 LR 0.001000 Time 0.020915 -2022-12-06 10:31:41,428 - Epoch: [21][ 570/ 1200] Overall Loss 0.353566 Objective Loss 0.353566 LR 0.001000 Time 0.020885 -2022-12-06 10:31:41,620 - Epoch: [21][ 580/ 1200] Overall Loss 0.353849 Objective Loss 0.353849 LR 0.001000 Time 0.020856 -2022-12-06 10:31:41,812 - Epoch: [21][ 590/ 1200] Overall Loss 0.353434 Objective Loss 0.353434 LR 0.001000 Time 0.020827 -2022-12-06 10:31:42,004 - Epoch: [21][ 600/ 1200] Overall Loss 0.353179 Objective Loss 0.353179 LR 0.001000 Time 0.020800 -2022-12-06 10:31:42,196 - Epoch: [21][ 610/ 1200] Overall Loss 0.352587 Objective Loss 0.352587 LR 0.001000 Time 0.020773 -2022-12-06 10:31:42,389 - Epoch: [21][ 620/ 1200] Overall Loss 0.352679 Objective Loss 0.352679 LR 0.001000 Time 0.020748 -2022-12-06 10:31:42,581 - Epoch: [21][ 630/ 1200] Overall Loss 0.352750 Objective Loss 0.352750 LR 0.001000 Time 0.020722 -2022-12-06 10:31:42,773 - Epoch: [21][ 640/ 1200] Overall Loss 0.353084 Objective Loss 0.353084 LR 0.001000 Time 0.020698 -2022-12-06 10:31:42,966 - Epoch: [21][ 650/ 1200] Overall Loss 0.352660 Objective Loss 0.352660 LR 0.001000 Time 0.020674 -2022-12-06 10:31:43,158 - Epoch: [21][ 660/ 1200] Overall Loss 0.352590 Objective Loss 0.352590 LR 0.001000 Time 0.020652 -2022-12-06 10:31:43,351 - Epoch: [21][ 670/ 1200] Overall Loss 0.352766 Objective Loss 0.352766 LR 0.001000 Time 0.020630 -2022-12-06 10:31:43,543 - Epoch: [21][ 680/ 1200] Overall Loss 0.352954 Objective Loss 0.352954 LR 0.001000 Time 0.020609 -2022-12-06 10:31:43,735 - Epoch: [21][ 690/ 1200] Overall Loss 0.352542 Objective Loss 0.352542 LR 0.001000 Time 0.020588 -2022-12-06 10:31:43,927 - Epoch: [21][ 700/ 1200] Overall Loss 0.352525 Objective Loss 0.352525 LR 0.001000 Time 0.020568 -2022-12-06 10:31:44,119 - Epoch: [21][ 710/ 1200] Overall Loss 0.352534 Objective Loss 0.352534 LR 0.001000 Time 0.020547 -2022-12-06 10:31:44,312 - Epoch: [21][ 720/ 1200] Overall Loss 0.352657 Objective Loss 0.352657 LR 0.001000 Time 0.020528 -2022-12-06 10:31:44,504 - Epoch: [21][ 730/ 1200] Overall Loss 0.352655 Objective Loss 0.352655 LR 0.001000 Time 0.020510 -2022-12-06 10:31:44,696 - Epoch: [21][ 740/ 1200] Overall Loss 0.352494 Objective Loss 0.352494 LR 0.001000 Time 0.020492 -2022-12-06 10:31:44,889 - Epoch: [21][ 750/ 1200] Overall Loss 0.352368 Objective Loss 0.352368 LR 0.001000 Time 0.020475 -2022-12-06 10:31:45,081 - Epoch: [21][ 760/ 1200] Overall Loss 0.352175 Objective Loss 0.352175 LR 0.001000 Time 0.020458 -2022-12-06 10:31:45,273 - Epoch: [21][ 770/ 1200] Overall Loss 0.352309 Objective Loss 0.352309 LR 0.001000 Time 0.020440 -2022-12-06 10:31:45,466 - Epoch: [21][ 780/ 1200] Overall Loss 0.352418 Objective Loss 0.352418 LR 0.001000 Time 0.020425 -2022-12-06 10:31:45,659 - Epoch: [21][ 790/ 1200] Overall Loss 0.352432 Objective Loss 0.352432 LR 0.001000 Time 0.020409 -2022-12-06 10:31:45,851 - Epoch: [21][ 800/ 1200] Overall Loss 0.352245 Objective Loss 0.352245 LR 0.001000 Time 0.020394 -2022-12-06 10:31:46,044 - Epoch: [21][ 810/ 1200] Overall Loss 0.352555 Objective Loss 0.352555 LR 0.001000 Time 0.020380 -2022-12-06 10:31:46,236 - Epoch: [21][ 820/ 1200] Overall Loss 0.352012 Objective Loss 0.352012 LR 0.001000 Time 0.020365 -2022-12-06 10:31:46,428 - Epoch: [21][ 830/ 1200] Overall Loss 0.351802 Objective Loss 0.351802 LR 0.001000 Time 0.020350 -2022-12-06 10:31:46,620 - Epoch: [21][ 840/ 1200] Overall Loss 0.351962 Objective Loss 0.351962 LR 0.001000 Time 0.020336 -2022-12-06 10:31:46,812 - Epoch: [21][ 850/ 1200] Overall Loss 0.352134 Objective Loss 0.352134 LR 0.001000 Time 0.020322 -2022-12-06 10:31:47,004 - Epoch: [21][ 860/ 1200] Overall Loss 0.352166 Objective Loss 0.352166 LR 0.001000 Time 0.020309 -2022-12-06 10:31:47,197 - Epoch: [21][ 870/ 1200] Overall Loss 0.352341 Objective Loss 0.352341 LR 0.001000 Time 0.020296 -2022-12-06 10:31:47,389 - Epoch: [21][ 880/ 1200] Overall Loss 0.352364 Objective Loss 0.352364 LR 0.001000 Time 0.020283 -2022-12-06 10:31:47,582 - Epoch: [21][ 890/ 1200] Overall Loss 0.352095 Objective Loss 0.352095 LR 0.001000 Time 0.020271 -2022-12-06 10:31:47,774 - Epoch: [21][ 900/ 1200] Overall Loss 0.351848 Objective Loss 0.351848 LR 0.001000 Time 0.020259 -2022-12-06 10:31:47,967 - Epoch: [21][ 910/ 1200] Overall Loss 0.351643 Objective Loss 0.351643 LR 0.001000 Time 0.020248 -2022-12-06 10:31:48,160 - Epoch: [21][ 920/ 1200] Overall Loss 0.351719 Objective Loss 0.351719 LR 0.001000 Time 0.020236 -2022-12-06 10:31:48,352 - Epoch: [21][ 930/ 1200] Overall Loss 0.351551 Objective Loss 0.351551 LR 0.001000 Time 0.020225 -2022-12-06 10:31:48,545 - Epoch: [21][ 940/ 1200] Overall Loss 0.351254 Objective Loss 0.351254 LR 0.001000 Time 0.020215 -2022-12-06 10:31:48,738 - Epoch: [21][ 950/ 1200] Overall Loss 0.351193 Objective Loss 0.351193 LR 0.001000 Time 0.020204 -2022-12-06 10:31:48,930 - Epoch: [21][ 960/ 1200] Overall Loss 0.350913 Objective Loss 0.350913 LR 0.001000 Time 0.020193 -2022-12-06 10:31:49,122 - Epoch: [21][ 970/ 1200] Overall Loss 0.350791 Objective Loss 0.350791 LR 0.001000 Time 0.020183 -2022-12-06 10:31:49,316 - Epoch: [21][ 980/ 1200] Overall Loss 0.350845 Objective Loss 0.350845 LR 0.001000 Time 0.020174 -2022-12-06 10:31:49,509 - Epoch: [21][ 990/ 1200] Overall Loss 0.350795 Objective Loss 0.350795 LR 0.001000 Time 0.020165 -2022-12-06 10:31:49,701 - Epoch: [21][ 1000/ 1200] Overall Loss 0.351011 Objective Loss 0.351011 LR 0.001000 Time 0.020155 -2022-12-06 10:31:49,894 - Epoch: [21][ 1010/ 1200] Overall Loss 0.351029 Objective Loss 0.351029 LR 0.001000 Time 0.020146 -2022-12-06 10:31:50,087 - Epoch: [21][ 1020/ 1200] Overall Loss 0.350816 Objective Loss 0.350816 LR 0.001000 Time 0.020136 -2022-12-06 10:31:50,279 - Epoch: [21][ 1030/ 1200] Overall Loss 0.350633 Objective Loss 0.350633 LR 0.001000 Time 0.020127 -2022-12-06 10:31:50,472 - Epoch: [21][ 1040/ 1200] Overall Loss 0.350879 Objective Loss 0.350879 LR 0.001000 Time 0.020119 -2022-12-06 10:31:50,664 - Epoch: [21][ 1050/ 1200] Overall Loss 0.350617 Objective Loss 0.350617 LR 0.001000 Time 0.020110 -2022-12-06 10:31:50,857 - Epoch: [21][ 1060/ 1200] Overall Loss 0.350387 Objective Loss 0.350387 LR 0.001000 Time 0.020101 -2022-12-06 10:31:51,051 - Epoch: [21][ 1070/ 1200] Overall Loss 0.350313 Objective Loss 0.350313 LR 0.001000 Time 0.020094 -2022-12-06 10:31:51,243 - Epoch: [21][ 1080/ 1200] Overall Loss 0.350293 Objective Loss 0.350293 LR 0.001000 Time 0.020085 -2022-12-06 10:31:51,436 - Epoch: [21][ 1090/ 1200] Overall Loss 0.350089 Objective Loss 0.350089 LR 0.001000 Time 0.020077 -2022-12-06 10:31:51,628 - Epoch: [21][ 1100/ 1200] Overall Loss 0.350154 Objective Loss 0.350154 LR 0.001000 Time 0.020069 -2022-12-06 10:31:51,821 - Epoch: [21][ 1110/ 1200] Overall Loss 0.350067 Objective Loss 0.350067 LR 0.001000 Time 0.020061 -2022-12-06 10:31:52,013 - Epoch: [21][ 1120/ 1200] Overall Loss 0.349880 Objective Loss 0.349880 LR 0.001000 Time 0.020054 -2022-12-06 10:31:52,206 - Epoch: [21][ 1130/ 1200] Overall Loss 0.349798 Objective Loss 0.349798 LR 0.001000 Time 0.020046 -2022-12-06 10:31:52,398 - Epoch: [21][ 1140/ 1200] Overall Loss 0.349782 Objective Loss 0.349782 LR 0.001000 Time 0.020039 -2022-12-06 10:31:52,591 - Epoch: [21][ 1150/ 1200] Overall Loss 0.349663 Objective Loss 0.349663 LR 0.001000 Time 0.020031 -2022-12-06 10:31:52,783 - Epoch: [21][ 1160/ 1200] Overall Loss 0.349563 Objective Loss 0.349563 LR 0.001000 Time 0.020024 -2022-12-06 10:31:52,975 - Epoch: [21][ 1170/ 1200] Overall Loss 0.349352 Objective Loss 0.349352 LR 0.001000 Time 0.020016 -2022-12-06 10:31:53,168 - Epoch: [21][ 1180/ 1200] Overall Loss 0.349301 Objective Loss 0.349301 LR 0.001000 Time 0.020009 -2022-12-06 10:31:53,360 - Epoch: [21][ 1190/ 1200] Overall Loss 0.349323 Objective Loss 0.349323 LR 0.001000 Time 0.020003 -2022-12-06 10:31:53,585 - Epoch: [21][ 1200/ 1200] Overall Loss 0.349233 Objective Loss 0.349233 Top1 84.728033 Top5 97.280335 LR 0.001000 Time 0.020023 -2022-12-06 10:31:53,674 - --- validate (epoch=21)----------- -2022-12-06 10:31:53,674 - 34129 samples (256 per mini-batch) -2022-12-06 10:31:54,120 - Epoch: [21][ 10/ 134] Loss 0.307225 Top1 83.203125 Top5 98.125000 -2022-12-06 10:31:54,246 - Epoch: [21][ 20/ 134] Loss 0.315406 Top1 83.574219 Top5 97.832031 -2022-12-06 10:31:54,376 - Epoch: [21][ 30/ 134] Loss 0.332085 Top1 83.125000 Top5 97.825521 -2022-12-06 10:31:54,504 - Epoch: [21][ 40/ 134] Loss 0.334520 Top1 83.212891 Top5 97.851562 -2022-12-06 10:31:54,638 - Epoch: [21][ 50/ 134] Loss 0.330372 Top1 83.328125 Top5 97.750000 -2022-12-06 10:31:54,785 - Epoch: [21][ 60/ 134] Loss 0.332259 Top1 83.235677 Top5 97.786458 -2022-12-06 10:31:54,925 - Epoch: [21][ 70/ 134] Loss 0.332833 Top1 83.387277 Top5 97.756696 -2022-12-06 10:31:55,071 - Epoch: [21][ 80/ 134] Loss 0.331813 Top1 83.530273 Top5 97.739258 -2022-12-06 10:31:55,212 - Epoch: [21][ 90/ 134] Loss 0.332852 Top1 83.546007 Top5 97.730035 -2022-12-06 10:31:55,359 - Epoch: [21][ 100/ 134] Loss 0.331404 Top1 83.500000 Top5 97.683594 -2022-12-06 10:31:55,499 - Epoch: [21][ 110/ 134] Loss 0.334315 Top1 83.501420 Top5 97.705966 -2022-12-06 10:31:55,646 - Epoch: [21][ 120/ 134] Loss 0.336452 Top1 83.476562 Top5 97.701823 -2022-12-06 10:31:55,782 - Epoch: [21][ 130/ 134] Loss 0.337062 Top1 83.491587 Top5 97.716346 -2022-12-06 10:31:55,818 - Epoch: [21][ 134/ 134] Loss 0.336761 Top1 83.489115 Top5 97.726274 -2022-12-06 10:31:55,906 - ==> Top1: 83.489 Top5: 97.726 Loss: 0.337 - -2022-12-06 10:31:55,907 - ==> Confusion: -[[ 904 1 2 2 4 4 2 0 5 52 0 5 1 0 4 1 0 0 1 1 7] - [ 0 939 0 2 9 25 4 13 0 0 2 8 0 1 3 0 3 1 8 2 7] - [ 6 7 998 12 1 6 24 7 0 1 5 11 3 1 1 4 1 1 4 3 7] - [ 2 4 37 916 0 3 0 0 0 0 12 1 3 1 19 2 2 4 7 1 6] - [ 11 9 6 0 954 6 2 1 0 4 1 3 1 0 4 4 8 2 0 1 3] - [ 0 46 1 2 7 959 5 11 3 2 0 12 3 5 0 2 2 0 1 4 4] - [ 0 0 21 1 0 1 1058 2 0 0 1 4 0 0 0 10 0 3 2 12 3] - [ 2 20 17 0 3 43 8 894 0 0 1 9 1 1 0 2 0 0 34 15 4] - [ 10 6 1 3 0 4 1 0 944 48 11 4 2 8 15 0 2 1 1 0 3] - [ 73 0 5 1 6 2 0 1 17 873 1 1 0 5 5 1 1 0 0 0 9] - [ 0 4 8 8 1 4 2 2 6 3 945 2 1 11 9 1 0 0 5 3 4] - [ 5 3 2 0 0 17 4 3 0 0 1 987 8 3 0 5 1 2 3 4 3] - [ 1 1 1 3 1 3 0 2 0 0 0 73 855 0 0 8 0 8 0 4 9] - [ 1 3 1 0 5 15 0 2 7 26 11 9 3 918 2 4 2 3 0 3 8] - [ 9 4 2 15 12 3 0 1 10 6 0 4 2 1 1036 0 1 4 6 1 13] - [ 1 0 6 2 3 1 5 0 0 0 1 12 5 1 0 981 7 8 0 5 5] - [ 1 6 5 1 12 2 2 0 1 1 0 9 0 0 0 10 1007 1 0 4 10] - [ 1 1 2 3 0 1 2 1 2 1 1 16 25 2 1 16 2 956 0 1 2] - [ 7 10 6 8 1 5 0 28 2 0 9 5 2 1 11 1 1 1 900 4 6] - [ 1 6 2 0 0 10 8 5 0 0 0 27 4 2 0 5 5 4 0 998 3] - [ 151 368 271 115 198 218 107 130 69 95 193 204 384 284 176 125 132 71 149 314 9472]] - -2022-12-06 10:31:56,487 - ==> Best [Top1: 83.489 Top5: 97.726 Sparsity:0.00 Params: 5376 on epoch: 21] -2022-12-06 10:31:56,487 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:31:56,494 - - -2022-12-06 10:31:56,494 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:31:57,532 - Epoch: [22][ 10/ 1200] Overall Loss 0.311206 Objective Loss 0.311206 LR 0.001000 Time 0.103750 -2022-12-06 10:31:57,737 - Epoch: [22][ 20/ 1200] Overall Loss 0.319388 Objective Loss 0.319388 LR 0.001000 Time 0.062049 -2022-12-06 10:31:57,930 - Epoch: [22][ 30/ 1200] Overall Loss 0.312663 Objective Loss 0.312663 LR 0.001000 Time 0.047804 -2022-12-06 10:31:58,123 - Epoch: [22][ 40/ 1200] Overall Loss 0.316382 Objective Loss 0.316382 LR 0.001000 Time 0.040661 -2022-12-06 10:31:58,316 - Epoch: [22][ 50/ 1200] Overall Loss 0.313239 Objective Loss 0.313239 LR 0.001000 Time 0.036379 -2022-12-06 10:31:58,508 - Epoch: [22][ 60/ 1200] Overall Loss 0.312923 Objective Loss 0.312923 LR 0.001000 Time 0.033504 -2022-12-06 10:31:58,701 - Epoch: [22][ 70/ 1200] Overall Loss 0.317026 Objective Loss 0.317026 LR 0.001000 Time 0.031468 -2022-12-06 10:31:58,893 - Epoch: [22][ 80/ 1200] Overall Loss 0.318670 Objective Loss 0.318670 LR 0.001000 Time 0.029929 -2022-12-06 10:31:59,086 - Epoch: [22][ 90/ 1200] Overall Loss 0.319405 Objective Loss 0.319405 LR 0.001000 Time 0.028734 -2022-12-06 10:31:59,278 - Epoch: [22][ 100/ 1200] Overall Loss 0.320741 Objective Loss 0.320741 LR 0.001000 Time 0.027775 -2022-12-06 10:31:59,470 - Epoch: [22][ 110/ 1200] Overall Loss 0.321989 Objective Loss 0.321989 LR 0.001000 Time 0.026992 -2022-12-06 10:31:59,662 - Epoch: [22][ 120/ 1200] Overall Loss 0.321759 Objective Loss 0.321759 LR 0.001000 Time 0.026340 -2022-12-06 10:31:59,854 - Epoch: [22][ 130/ 1200] Overall Loss 0.321089 Objective Loss 0.321089 LR 0.001000 Time 0.025788 -2022-12-06 10:32:00,046 - Epoch: [22][ 140/ 1200] Overall Loss 0.321814 Objective Loss 0.321814 LR 0.001000 Time 0.025316 -2022-12-06 10:32:00,239 - Epoch: [22][ 150/ 1200] Overall Loss 0.323031 Objective Loss 0.323031 LR 0.001000 Time 0.024909 -2022-12-06 10:32:00,431 - Epoch: [22][ 160/ 1200] Overall Loss 0.323355 Objective Loss 0.323355 LR 0.001000 Time 0.024549 -2022-12-06 10:32:00,623 - Epoch: [22][ 170/ 1200] Overall Loss 0.323722 Objective Loss 0.323722 LR 0.001000 Time 0.024233 -2022-12-06 10:32:00,815 - Epoch: [22][ 180/ 1200] Overall Loss 0.324292 Objective Loss 0.324292 LR 0.001000 Time 0.023950 -2022-12-06 10:32:01,008 - Epoch: [22][ 190/ 1200] Overall Loss 0.324775 Objective Loss 0.324775 LR 0.001000 Time 0.023702 -2022-12-06 10:32:01,201 - Epoch: [22][ 200/ 1200] Overall Loss 0.324594 Objective Loss 0.324594 LR 0.001000 Time 0.023478 -2022-12-06 10:32:01,394 - Epoch: [22][ 210/ 1200] Overall Loss 0.326158 Objective Loss 0.326158 LR 0.001000 Time 0.023275 -2022-12-06 10:32:01,587 - Epoch: [22][ 220/ 1200] Overall Loss 0.324713 Objective Loss 0.324713 LR 0.001000 Time 0.023091 -2022-12-06 10:32:01,779 - Epoch: [22][ 230/ 1200] Overall Loss 0.323889 Objective Loss 0.323889 LR 0.001000 Time 0.022920 -2022-12-06 10:32:01,972 - Epoch: [22][ 240/ 1200] Overall Loss 0.323384 Objective Loss 0.323384 LR 0.001000 Time 0.022766 -2022-12-06 10:32:02,164 - Epoch: [22][ 250/ 1200] Overall Loss 0.323207 Objective Loss 0.323207 LR 0.001000 Time 0.022623 -2022-12-06 10:32:02,356 - Epoch: [22][ 260/ 1200] Overall Loss 0.322978 Objective Loss 0.322978 LR 0.001000 Time 0.022490 -2022-12-06 10:32:02,549 - Epoch: [22][ 270/ 1200] Overall Loss 0.323815 Objective Loss 0.323815 LR 0.001000 Time 0.022369 -2022-12-06 10:32:02,741 - Epoch: [22][ 280/ 1200] Overall Loss 0.323141 Objective Loss 0.323141 LR 0.001000 Time 0.022254 -2022-12-06 10:32:02,934 - Epoch: [22][ 290/ 1200] Overall Loss 0.323342 Objective Loss 0.323342 LR 0.001000 Time 0.022151 -2022-12-06 10:32:03,127 - Epoch: [22][ 300/ 1200] Overall Loss 0.323412 Objective Loss 0.323412 LR 0.001000 Time 0.022053 -2022-12-06 10:32:03,320 - Epoch: [22][ 310/ 1200] Overall Loss 0.323105 Objective Loss 0.323105 LR 0.001000 Time 0.021961 -2022-12-06 10:32:03,512 - Epoch: [22][ 320/ 1200] Overall Loss 0.322468 Objective Loss 0.322468 LR 0.001000 Time 0.021874 -2022-12-06 10:32:03,705 - Epoch: [22][ 330/ 1200] Overall Loss 0.323024 Objective Loss 0.323024 LR 0.001000 Time 0.021793 -2022-12-06 10:32:03,898 - Epoch: [22][ 340/ 1200] Overall Loss 0.321722 Objective Loss 0.321722 LR 0.001000 Time 0.021718 -2022-12-06 10:32:04,090 - Epoch: [22][ 350/ 1200] Overall Loss 0.321302 Objective Loss 0.321302 LR 0.001000 Time 0.021646 -2022-12-06 10:32:04,282 - Epoch: [22][ 360/ 1200] Overall Loss 0.321901 Objective Loss 0.321901 LR 0.001000 Time 0.021578 -2022-12-06 10:32:04,475 - Epoch: [22][ 370/ 1200] Overall Loss 0.321505 Objective Loss 0.321505 LR 0.001000 Time 0.021513 -2022-12-06 10:32:04,667 - Epoch: [22][ 380/ 1200] Overall Loss 0.321792 Objective Loss 0.321792 LR 0.001000 Time 0.021452 -2022-12-06 10:32:04,860 - Epoch: [22][ 390/ 1200] Overall Loss 0.321508 Objective Loss 0.321508 LR 0.001000 Time 0.021394 -2022-12-06 10:32:05,052 - Epoch: [22][ 400/ 1200] Overall Loss 0.322528 Objective Loss 0.322528 LR 0.001000 Time 0.021338 -2022-12-06 10:32:05,244 - Epoch: [22][ 410/ 1200] Overall Loss 0.323303 Objective Loss 0.323303 LR 0.001000 Time 0.021285 -2022-12-06 10:32:05,436 - Epoch: [22][ 420/ 1200] Overall Loss 0.323111 Objective Loss 0.323111 LR 0.001000 Time 0.021234 -2022-12-06 10:32:05,629 - Epoch: [22][ 430/ 1200] Overall Loss 0.323130 Objective Loss 0.323130 LR 0.001000 Time 0.021188 -2022-12-06 10:32:05,822 - Epoch: [22][ 440/ 1200] Overall Loss 0.323571 Objective Loss 0.323571 LR 0.001000 Time 0.021143 -2022-12-06 10:32:06,014 - Epoch: [22][ 450/ 1200] Overall Loss 0.324248 Objective Loss 0.324248 LR 0.001000 Time 0.021100 -2022-12-06 10:32:06,207 - Epoch: [22][ 460/ 1200] Overall Loss 0.324468 Objective Loss 0.324468 LR 0.001000 Time 0.021058 -2022-12-06 10:32:06,399 - Epoch: [22][ 470/ 1200] Overall Loss 0.324768 Objective Loss 0.324768 LR 0.001000 Time 0.021018 -2022-12-06 10:32:06,593 - Epoch: [22][ 480/ 1200] Overall Loss 0.324959 Objective Loss 0.324959 LR 0.001000 Time 0.020983 -2022-12-06 10:32:06,785 - Epoch: [22][ 490/ 1200] Overall Loss 0.325268 Objective Loss 0.325268 LR 0.001000 Time 0.020946 -2022-12-06 10:32:06,979 - Epoch: [22][ 500/ 1200] Overall Loss 0.324719 Objective Loss 0.324719 LR 0.001000 Time 0.020913 -2022-12-06 10:32:07,172 - Epoch: [22][ 510/ 1200] Overall Loss 0.324366 Objective Loss 0.324366 LR 0.001000 Time 0.020880 -2022-12-06 10:32:07,365 - Epoch: [22][ 520/ 1200] Overall Loss 0.323869 Objective Loss 0.323869 LR 0.001000 Time 0.020849 -2022-12-06 10:32:07,558 - Epoch: [22][ 530/ 1200] Overall Loss 0.323736 Objective Loss 0.323736 LR 0.001000 Time 0.020819 -2022-12-06 10:32:07,750 - Epoch: [22][ 540/ 1200] Overall Loss 0.323899 Objective Loss 0.323899 LR 0.001000 Time 0.020789 -2022-12-06 10:32:07,942 - Epoch: [22][ 550/ 1200] Overall Loss 0.324217 Objective Loss 0.324217 LR 0.001000 Time 0.020758 -2022-12-06 10:32:08,134 - Epoch: [22][ 560/ 1200] Overall Loss 0.324475 Objective Loss 0.324475 LR 0.001000 Time 0.020729 -2022-12-06 10:32:08,325 - Epoch: [22][ 570/ 1200] Overall Loss 0.324660 Objective Loss 0.324660 LR 0.001000 Time 0.020700 -2022-12-06 10:32:08,516 - Epoch: [22][ 580/ 1200] Overall Loss 0.324783 Objective Loss 0.324783 LR 0.001000 Time 0.020672 -2022-12-06 10:32:08,709 - Epoch: [22][ 590/ 1200] Overall Loss 0.324266 Objective Loss 0.324266 LR 0.001000 Time 0.020647 -2022-12-06 10:32:08,901 - Epoch: [22][ 600/ 1200] Overall Loss 0.325075 Objective Loss 0.325075 LR 0.001000 Time 0.020623 -2022-12-06 10:32:09,094 - Epoch: [22][ 610/ 1200] Overall Loss 0.324580 Objective Loss 0.324580 LR 0.001000 Time 0.020599 -2022-12-06 10:32:09,285 - Epoch: [22][ 620/ 1200] Overall Loss 0.324521 Objective Loss 0.324521 LR 0.001000 Time 0.020574 -2022-12-06 10:32:09,477 - Epoch: [22][ 630/ 1200] Overall Loss 0.324894 Objective Loss 0.324894 LR 0.001000 Time 0.020551 -2022-12-06 10:32:09,668 - Epoch: [22][ 640/ 1200] Overall Loss 0.324651 Objective Loss 0.324651 LR 0.001000 Time 0.020529 -2022-12-06 10:32:09,860 - Epoch: [22][ 650/ 1200] Overall Loss 0.324670 Objective Loss 0.324670 LR 0.001000 Time 0.020508 -2022-12-06 10:32:10,052 - Epoch: [22][ 660/ 1200] Overall Loss 0.324923 Objective Loss 0.324923 LR 0.001000 Time 0.020486 -2022-12-06 10:32:10,244 - Epoch: [22][ 670/ 1200] Overall Loss 0.325101 Objective Loss 0.325101 LR 0.001000 Time 0.020466 -2022-12-06 10:32:10,436 - Epoch: [22][ 680/ 1200] Overall Loss 0.324869 Objective Loss 0.324869 LR 0.001000 Time 0.020446 -2022-12-06 10:32:10,627 - Epoch: [22][ 690/ 1200] Overall Loss 0.325159 Objective Loss 0.325159 LR 0.001000 Time 0.020427 -2022-12-06 10:32:10,818 - Epoch: [22][ 700/ 1200] Overall Loss 0.325120 Objective Loss 0.325120 LR 0.001000 Time 0.020407 -2022-12-06 10:32:11,010 - Epoch: [22][ 710/ 1200] Overall Loss 0.325165 Objective Loss 0.325165 LR 0.001000 Time 0.020389 -2022-12-06 10:32:11,202 - Epoch: [22][ 720/ 1200] Overall Loss 0.325013 Objective Loss 0.325013 LR 0.001000 Time 0.020372 -2022-12-06 10:32:11,394 - Epoch: [22][ 730/ 1200] Overall Loss 0.325053 Objective Loss 0.325053 LR 0.001000 Time 0.020355 -2022-12-06 10:32:11,586 - Epoch: [22][ 740/ 1200] Overall Loss 0.325353 Objective Loss 0.325353 LR 0.001000 Time 0.020339 -2022-12-06 10:32:11,778 - Epoch: [22][ 750/ 1200] Overall Loss 0.325116 Objective Loss 0.325116 LR 0.001000 Time 0.020323 -2022-12-06 10:32:11,970 - Epoch: [22][ 760/ 1200] Overall Loss 0.325311 Objective Loss 0.325311 LR 0.001000 Time 0.020307 -2022-12-06 10:32:12,162 - Epoch: [22][ 770/ 1200] Overall Loss 0.325593 Objective Loss 0.325593 LR 0.001000 Time 0.020292 -2022-12-06 10:32:12,354 - Epoch: [22][ 780/ 1200] Overall Loss 0.325708 Objective Loss 0.325708 LR 0.001000 Time 0.020277 -2022-12-06 10:32:12,546 - Epoch: [22][ 790/ 1200] Overall Loss 0.326144 Objective Loss 0.326144 LR 0.001000 Time 0.020263 -2022-12-06 10:32:12,737 - Epoch: [22][ 800/ 1200] Overall Loss 0.326081 Objective Loss 0.326081 LR 0.001000 Time 0.020248 -2022-12-06 10:32:12,928 - Epoch: [22][ 810/ 1200] Overall Loss 0.325990 Objective Loss 0.325990 LR 0.001000 Time 0.020233 -2022-12-06 10:32:13,120 - Epoch: [22][ 820/ 1200] Overall Loss 0.326075 Objective Loss 0.326075 LR 0.001000 Time 0.020220 -2022-12-06 10:32:13,312 - Epoch: [22][ 830/ 1200] Overall Loss 0.326012 Objective Loss 0.326012 LR 0.001000 Time 0.020207 -2022-12-06 10:32:13,504 - Epoch: [22][ 840/ 1200] Overall Loss 0.325937 Objective Loss 0.325937 LR 0.001000 Time 0.020195 -2022-12-06 10:32:13,697 - Epoch: [22][ 850/ 1200] Overall Loss 0.326040 Objective Loss 0.326040 LR 0.001000 Time 0.020182 -2022-12-06 10:32:13,888 - Epoch: [22][ 860/ 1200] Overall Loss 0.326022 Objective Loss 0.326022 LR 0.001000 Time 0.020169 -2022-12-06 10:32:14,079 - Epoch: [22][ 870/ 1200] Overall Loss 0.325928 Objective Loss 0.325928 LR 0.001000 Time 0.020157 -2022-12-06 10:32:14,271 - Epoch: [22][ 880/ 1200] Overall Loss 0.326294 Objective Loss 0.326294 LR 0.001000 Time 0.020145 -2022-12-06 10:32:14,463 - Epoch: [22][ 890/ 1200] Overall Loss 0.326734 Objective Loss 0.326734 LR 0.001000 Time 0.020134 -2022-12-06 10:32:14,655 - Epoch: [22][ 900/ 1200] Overall Loss 0.326937 Objective Loss 0.326937 LR 0.001000 Time 0.020123 -2022-12-06 10:32:14,847 - Epoch: [22][ 910/ 1200] Overall Loss 0.327040 Objective Loss 0.327040 LR 0.001000 Time 0.020112 -2022-12-06 10:32:15,039 - Epoch: [22][ 920/ 1200] Overall Loss 0.327023 Objective Loss 0.327023 LR 0.001000 Time 0.020101 -2022-12-06 10:32:15,231 - Epoch: [22][ 930/ 1200] Overall Loss 0.327245 Objective Loss 0.327245 LR 0.001000 Time 0.020091 -2022-12-06 10:32:15,422 - Epoch: [22][ 940/ 1200] Overall Loss 0.327201 Objective Loss 0.327201 LR 0.001000 Time 0.020080 -2022-12-06 10:32:15,613 - Epoch: [22][ 950/ 1200] Overall Loss 0.327493 Objective Loss 0.327493 LR 0.001000 Time 0.020070 -2022-12-06 10:32:15,806 - Epoch: [22][ 960/ 1200] Overall Loss 0.327578 Objective Loss 0.327578 LR 0.001000 Time 0.020061 -2022-12-06 10:32:15,998 - Epoch: [22][ 970/ 1200] Overall Loss 0.327894 Objective Loss 0.327894 LR 0.001000 Time 0.020052 -2022-12-06 10:32:16,190 - Epoch: [22][ 980/ 1200] Overall Loss 0.328379 Objective Loss 0.328379 LR 0.001000 Time 0.020042 -2022-12-06 10:32:16,382 - Epoch: [22][ 990/ 1200] Overall Loss 0.328470 Objective Loss 0.328470 LR 0.001000 Time 0.020033 -2022-12-06 10:32:16,574 - Epoch: [22][ 1000/ 1200] Overall Loss 0.328390 Objective Loss 0.328390 LR 0.001000 Time 0.020024 -2022-12-06 10:32:16,766 - Epoch: [22][ 1010/ 1200] Overall Loss 0.328515 Objective Loss 0.328515 LR 0.001000 Time 0.020016 -2022-12-06 10:32:16,959 - Epoch: [22][ 1020/ 1200] Overall Loss 0.328357 Objective Loss 0.328357 LR 0.001000 Time 0.020008 -2022-12-06 10:32:17,151 - Epoch: [22][ 1030/ 1200] Overall Loss 0.328499 Objective Loss 0.328499 LR 0.001000 Time 0.019999 -2022-12-06 10:32:17,343 - Epoch: [22][ 1040/ 1200] Overall Loss 0.328538 Objective Loss 0.328538 LR 0.001000 Time 0.019991 -2022-12-06 10:32:17,535 - Epoch: [22][ 1050/ 1200] Overall Loss 0.328432 Objective Loss 0.328432 LR 0.001000 Time 0.019984 -2022-12-06 10:32:17,727 - Epoch: [22][ 1060/ 1200] Overall Loss 0.328370 Objective Loss 0.328370 LR 0.001000 Time 0.019975 -2022-12-06 10:32:17,919 - Epoch: [22][ 1070/ 1200] Overall Loss 0.328292 Objective Loss 0.328292 LR 0.001000 Time 0.019968 -2022-12-06 10:32:18,111 - Epoch: [22][ 1080/ 1200] Overall Loss 0.328388 Objective Loss 0.328388 LR 0.001000 Time 0.019960 -2022-12-06 10:32:18,303 - Epoch: [22][ 1090/ 1200] Overall Loss 0.328235 Objective Loss 0.328235 LR 0.001000 Time 0.019953 -2022-12-06 10:32:18,495 - Epoch: [22][ 1100/ 1200] Overall Loss 0.327926 Objective Loss 0.327926 LR 0.001000 Time 0.019945 -2022-12-06 10:32:18,687 - Epoch: [22][ 1110/ 1200] Overall Loss 0.328147 Objective Loss 0.328147 LR 0.001000 Time 0.019938 -2022-12-06 10:32:18,878 - Epoch: [22][ 1120/ 1200] Overall Loss 0.328232 Objective Loss 0.328232 LR 0.001000 Time 0.019930 -2022-12-06 10:32:19,070 - Epoch: [22][ 1130/ 1200] Overall Loss 0.328191 Objective Loss 0.328191 LR 0.001000 Time 0.019923 -2022-12-06 10:32:19,262 - Epoch: [22][ 1140/ 1200] Overall Loss 0.328096 Objective Loss 0.328096 LR 0.001000 Time 0.019916 -2022-12-06 10:32:19,453 - Epoch: [22][ 1150/ 1200] Overall Loss 0.328074 Objective Loss 0.328074 LR 0.001000 Time 0.019909 -2022-12-06 10:32:19,644 - Epoch: [22][ 1160/ 1200] Overall Loss 0.328202 Objective Loss 0.328202 LR 0.001000 Time 0.019901 -2022-12-06 10:32:19,838 - Epoch: [22][ 1170/ 1200] Overall Loss 0.328226 Objective Loss 0.328226 LR 0.001000 Time 0.019897 -2022-12-06 10:32:20,030 - Epoch: [22][ 1180/ 1200] Overall Loss 0.328296 Objective Loss 0.328296 LR 0.001000 Time 0.019890 -2022-12-06 10:32:20,221 - Epoch: [22][ 1190/ 1200] Overall Loss 0.328233 Objective Loss 0.328233 LR 0.001000 Time 0.019883 -2022-12-06 10:32:20,454 - Epoch: [22][ 1200/ 1200] Overall Loss 0.328678 Objective Loss 0.328678 Top1 79.497908 Top5 95.815900 LR 0.001000 Time 0.019911 -2022-12-06 10:32:20,543 - --- validate (epoch=22)----------- -2022-12-06 10:32:20,543 - 34129 samples (256 per mini-batch) -2022-12-06 10:32:20,992 - Epoch: [22][ 10/ 134] Loss 0.328539 Top1 81.484375 Top5 96.835938 -2022-12-06 10:32:21,138 - Epoch: [22][ 20/ 134] Loss 0.325886 Top1 81.601562 Top5 96.914062 -2022-12-06 10:32:21,268 - Epoch: [22][ 30/ 134] Loss 0.319512 Top1 82.213542 Top5 97.148438 -2022-12-06 10:32:21,400 - Epoch: [22][ 40/ 134] Loss 0.326369 Top1 81.796875 Top5 97.060547 -2022-12-06 10:32:21,530 - Epoch: [22][ 50/ 134] Loss 0.330673 Top1 81.531250 Top5 97.109375 -2022-12-06 10:32:21,661 - Epoch: [22][ 60/ 134] Loss 0.328091 Top1 81.731771 Top5 97.180990 -2022-12-06 10:32:21,791 - Epoch: [22][ 70/ 134] Loss 0.329014 Top1 81.735491 Top5 97.131696 -2022-12-06 10:32:21,923 - Epoch: [22][ 80/ 134] Loss 0.329857 Top1 81.655273 Top5 97.075195 -2022-12-06 10:32:22,053 - Epoch: [22][ 90/ 134] Loss 0.331292 Top1 81.592882 Top5 97.087674 -2022-12-06 10:32:22,184 - Epoch: [22][ 100/ 134] Loss 0.332501 Top1 81.660156 Top5 97.117188 -2022-12-06 10:32:22,314 - Epoch: [22][ 110/ 134] Loss 0.327337 Top1 81.821733 Top5 97.183949 -2022-12-06 10:32:22,447 - Epoch: [22][ 120/ 134] Loss 0.328791 Top1 81.861979 Top5 97.190755 -2022-12-06 10:32:22,579 - Epoch: [22][ 130/ 134] Loss 0.330745 Top1 81.859976 Top5 97.187500 -2022-12-06 10:32:22,617 - Epoch: [22][ 134/ 134] Loss 0.330040 Top1 81.915673 Top5 97.225234 -2022-12-06 10:32:22,705 - ==> Top1: 81.916 Top5: 97.225 Loss: 0.330 - -2022-12-06 10:32:22,706 - ==> Confusion: -[[ 858 4 3 0 5 4 0 1 10 92 0 5 2 1 3 2 1 0 1 2 2] - [ 2 904 2 2 15 31 4 19 4 0 7 3 2 4 1 3 5 1 13 3 2] - [ 2 9 996 14 1 5 18 12 0 2 12 5 1 3 0 8 1 0 3 6 5] - [ 1 2 24 923 0 4 0 1 1 0 21 1 3 2 18 2 1 2 8 2 4] - [ 12 5 2 0 947 5 1 1 1 10 0 6 0 2 9 6 9 3 0 1 0] - [ 3 21 0 0 10 962 4 14 3 2 2 14 3 13 1 2 3 0 2 7 3] - [ 0 2 14 0 0 2 1057 7 0 0 3 5 1 1 0 6 1 1 1 14 3] - [ 2 14 15 2 3 42 4 912 3 0 3 4 0 2 0 1 1 0 24 19 3] - [ 6 2 1 0 0 3 0 0 975 42 14 3 1 9 5 0 2 0 1 0 0] - [ 42 0 3 0 4 3 0 2 40 889 2 1 2 5 1 1 0 0 2 1 3] - [ 0 2 2 3 0 2 0 1 4 3 976 2 2 10 4 0 1 0 5 1 1] - [ 3 3 2 0 1 15 1 6 0 2 2 966 11 9 0 3 2 6 1 16 2] - [ 1 2 0 4 3 8 0 2 1 0 1 83 823 2 1 7 3 24 1 2 1] - [ 0 3 0 0 2 9 0 2 19 23 12 10 2 920 0 4 1 1 2 2 11] - [ 7 6 3 10 8 2 0 1 29 5 4 1 2 3 1025 1 4 4 5 3 7] - [ 3 2 2 1 2 3 2 0 1 0 1 12 3 3 1 980 10 11 0 6 0] - [ 2 2 1 1 7 6 2 0 1 0 1 6 1 1 1 11 1018 1 1 5 4] - [ 1 3 1 1 0 3 1 2 1 0 4 15 11 2 2 11 2 974 0 1 1] - [ 0 2 7 10 1 8 0 25 2 1 13 5 3 2 11 0 3 1 908 3 3] - [ 1 4 4 0 0 6 6 4 1 0 1 18 2 5 0 6 5 2 1 1009 5] - [ 156 303 241 120 171 253 89 145 139 137 365 202 417 360 190 153 270 89 189 307 8930]] - -2022-12-06 10:32:23,280 - ==> Best [Top1: 83.489 Top5: 97.726 Sparsity:0.00 Params: 5376 on epoch: 21] -2022-12-06 10:32:23,281 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:32:23,286 - - -2022-12-06 10:32:23,287 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:32:24,316 - Epoch: [23][ 10/ 1200] Overall Loss 0.311389 Objective Loss 0.311389 LR 0.001000 Time 0.102876 -2022-12-06 10:32:24,510 - Epoch: [23][ 20/ 1200] Overall Loss 0.317862 Objective Loss 0.317862 LR 0.001000 Time 0.061106 -2022-12-06 10:32:24,701 - Epoch: [23][ 30/ 1200] Overall Loss 0.314422 Objective Loss 0.314422 LR 0.001000 Time 0.047100 -2022-12-06 10:32:24,891 - Epoch: [23][ 40/ 1200] Overall Loss 0.313358 Objective Loss 0.313358 LR 0.001000 Time 0.040059 -2022-12-06 10:32:25,082 - Epoch: [23][ 50/ 1200] Overall Loss 0.314086 Objective Loss 0.314086 LR 0.001000 Time 0.035839 -2022-12-06 10:32:25,271 - Epoch: [23][ 60/ 1200] Overall Loss 0.313647 Objective Loss 0.313647 LR 0.001000 Time 0.033022 -2022-12-06 10:32:25,462 - Epoch: [23][ 70/ 1200] Overall Loss 0.315366 Objective Loss 0.315366 LR 0.001000 Time 0.031015 -2022-12-06 10:32:25,652 - Epoch: [23][ 80/ 1200] Overall Loss 0.314704 Objective Loss 0.314704 LR 0.001000 Time 0.029508 -2022-12-06 10:32:25,842 - Epoch: [23][ 90/ 1200] Overall Loss 0.314709 Objective Loss 0.314709 LR 0.001000 Time 0.028335 -2022-12-06 10:32:26,031 - Epoch: [23][ 100/ 1200] Overall Loss 0.313959 Objective Loss 0.313959 LR 0.001000 Time 0.027393 -2022-12-06 10:32:26,222 - Epoch: [23][ 110/ 1200] Overall Loss 0.314977 Objective Loss 0.314977 LR 0.001000 Time 0.026629 -2022-12-06 10:32:26,412 - Epoch: [23][ 120/ 1200] Overall Loss 0.316048 Objective Loss 0.316048 LR 0.001000 Time 0.025989 -2022-12-06 10:32:26,603 - Epoch: [23][ 130/ 1200] Overall Loss 0.316659 Objective Loss 0.316659 LR 0.001000 Time 0.025453 -2022-12-06 10:32:26,792 - Epoch: [23][ 140/ 1200] Overall Loss 0.316508 Objective Loss 0.316508 LR 0.001000 Time 0.024984 -2022-12-06 10:32:26,982 - Epoch: [23][ 150/ 1200] Overall Loss 0.316768 Objective Loss 0.316768 LR 0.001000 Time 0.024583 -2022-12-06 10:32:27,172 - Epoch: [23][ 160/ 1200] Overall Loss 0.316375 Objective Loss 0.316375 LR 0.001000 Time 0.024229 -2022-12-06 10:32:27,362 - Epoch: [23][ 170/ 1200] Overall Loss 0.316306 Objective Loss 0.316306 LR 0.001000 Time 0.023917 -2022-12-06 10:32:27,552 - Epoch: [23][ 180/ 1200] Overall Loss 0.315191 Objective Loss 0.315191 LR 0.001000 Time 0.023640 -2022-12-06 10:32:27,742 - Epoch: [23][ 190/ 1200] Overall Loss 0.315308 Objective Loss 0.315308 LR 0.001000 Time 0.023394 -2022-12-06 10:32:27,932 - Epoch: [23][ 200/ 1200] Overall Loss 0.315633 Objective Loss 0.315633 LR 0.001000 Time 0.023173 -2022-12-06 10:32:28,122 - Epoch: [23][ 210/ 1200] Overall Loss 0.315733 Objective Loss 0.315733 LR 0.001000 Time 0.022971 -2022-12-06 10:32:28,311 - Epoch: [23][ 220/ 1200] Overall Loss 0.315524 Objective Loss 0.315524 LR 0.001000 Time 0.022785 -2022-12-06 10:32:28,502 - Epoch: [23][ 230/ 1200] Overall Loss 0.314390 Objective Loss 0.314390 LR 0.001000 Time 0.022621 -2022-12-06 10:32:28,691 - Epoch: [23][ 240/ 1200] Overall Loss 0.315082 Objective Loss 0.315082 LR 0.001000 Time 0.022466 -2022-12-06 10:32:28,881 - Epoch: [23][ 250/ 1200] Overall Loss 0.315304 Objective Loss 0.315304 LR 0.001000 Time 0.022325 -2022-12-06 10:32:29,072 - Epoch: [23][ 260/ 1200] Overall Loss 0.315822 Objective Loss 0.315822 LR 0.001000 Time 0.022196 -2022-12-06 10:32:29,262 - Epoch: [23][ 270/ 1200] Overall Loss 0.315341 Objective Loss 0.315341 LR 0.001000 Time 0.022078 -2022-12-06 10:32:29,452 - Epoch: [23][ 280/ 1200] Overall Loss 0.317280 Objective Loss 0.317280 LR 0.001000 Time 0.021964 -2022-12-06 10:32:29,642 - Epoch: [23][ 290/ 1200] Overall Loss 0.317536 Objective Loss 0.317536 LR 0.001000 Time 0.021860 -2022-12-06 10:32:29,831 - Epoch: [23][ 300/ 1200] Overall Loss 0.317397 Objective Loss 0.317397 LR 0.001000 Time 0.021761 -2022-12-06 10:32:30,022 - Epoch: [23][ 310/ 1200] Overall Loss 0.317470 Objective Loss 0.317470 LR 0.001000 Time 0.021672 -2022-12-06 10:32:30,211 - Epoch: [23][ 320/ 1200] Overall Loss 0.316989 Objective Loss 0.316989 LR 0.001000 Time 0.021586 -2022-12-06 10:32:30,402 - Epoch: [23][ 330/ 1200] Overall Loss 0.316819 Objective Loss 0.316819 LR 0.001000 Time 0.021506 -2022-12-06 10:32:30,592 - Epoch: [23][ 340/ 1200] Overall Loss 0.317698 Objective Loss 0.317698 LR 0.001000 Time 0.021431 -2022-12-06 10:32:30,782 - Epoch: [23][ 350/ 1200] Overall Loss 0.317914 Objective Loss 0.317914 LR 0.001000 Time 0.021361 -2022-12-06 10:32:30,972 - Epoch: [23][ 360/ 1200] Overall Loss 0.319370 Objective Loss 0.319370 LR 0.001000 Time 0.021293 -2022-12-06 10:32:31,162 - Epoch: [23][ 370/ 1200] Overall Loss 0.320358 Objective Loss 0.320358 LR 0.001000 Time 0.021229 -2022-12-06 10:32:31,351 - Epoch: [23][ 380/ 1200] Overall Loss 0.320141 Objective Loss 0.320141 LR 0.001000 Time 0.021169 -2022-12-06 10:32:31,542 - Epoch: [23][ 390/ 1200] Overall Loss 0.320262 Objective Loss 0.320262 LR 0.001000 Time 0.021114 -2022-12-06 10:32:31,732 - Epoch: [23][ 400/ 1200] Overall Loss 0.319647 Objective Loss 0.319647 LR 0.001000 Time 0.021059 -2022-12-06 10:32:31,922 - Epoch: [23][ 410/ 1200] Overall Loss 0.320063 Objective Loss 0.320063 LR 0.001000 Time 0.021007 -2022-12-06 10:32:32,111 - Epoch: [23][ 420/ 1200] Overall Loss 0.320620 Objective Loss 0.320620 LR 0.001000 Time 0.020957 -2022-12-06 10:32:32,301 - Epoch: [23][ 430/ 1200] Overall Loss 0.320127 Objective Loss 0.320127 LR 0.001000 Time 0.020910 -2022-12-06 10:32:32,491 - Epoch: [23][ 440/ 1200] Overall Loss 0.319918 Objective Loss 0.319918 LR 0.001000 Time 0.020864 -2022-12-06 10:32:32,681 - Epoch: [23][ 450/ 1200] Overall Loss 0.320319 Objective Loss 0.320319 LR 0.001000 Time 0.020823 -2022-12-06 10:32:32,872 - Epoch: [23][ 460/ 1200] Overall Loss 0.320244 Objective Loss 0.320244 LR 0.001000 Time 0.020783 -2022-12-06 10:32:33,061 - Epoch: [23][ 470/ 1200] Overall Loss 0.320578 Objective Loss 0.320578 LR 0.001000 Time 0.020743 -2022-12-06 10:32:33,251 - Epoch: [23][ 480/ 1200] Overall Loss 0.321301 Objective Loss 0.321301 LR 0.001000 Time 0.020705 -2022-12-06 10:32:33,441 - Epoch: [23][ 490/ 1200] Overall Loss 0.321747 Objective Loss 0.321747 LR 0.001000 Time 0.020669 -2022-12-06 10:32:33,632 - Epoch: [23][ 500/ 1200] Overall Loss 0.321429 Objective Loss 0.321429 LR 0.001000 Time 0.020636 -2022-12-06 10:32:33,821 - Epoch: [23][ 510/ 1200] Overall Loss 0.321014 Objective Loss 0.321014 LR 0.001000 Time 0.020602 -2022-12-06 10:32:34,011 - Epoch: [23][ 520/ 1200] Overall Loss 0.321627 Objective Loss 0.321627 LR 0.001000 Time 0.020570 -2022-12-06 10:32:34,202 - Epoch: [23][ 530/ 1200] Overall Loss 0.321929 Objective Loss 0.321929 LR 0.001000 Time 0.020540 -2022-12-06 10:32:34,391 - Epoch: [23][ 540/ 1200] Overall Loss 0.321853 Objective Loss 0.321853 LR 0.001000 Time 0.020510 -2022-12-06 10:32:34,581 - Epoch: [23][ 550/ 1200] Overall Loss 0.321738 Objective Loss 0.321738 LR 0.001000 Time 0.020482 -2022-12-06 10:32:34,771 - Epoch: [23][ 560/ 1200] Overall Loss 0.321675 Objective Loss 0.321675 LR 0.001000 Time 0.020455 -2022-12-06 10:32:34,962 - Epoch: [23][ 570/ 1200] Overall Loss 0.322171 Objective Loss 0.322171 LR 0.001000 Time 0.020428 -2022-12-06 10:32:35,152 - Epoch: [23][ 580/ 1200] Overall Loss 0.321792 Objective Loss 0.321792 LR 0.001000 Time 0.020403 -2022-12-06 10:32:35,342 - Epoch: [23][ 590/ 1200] Overall Loss 0.321864 Objective Loss 0.321864 LR 0.001000 Time 0.020378 -2022-12-06 10:32:35,531 - Epoch: [23][ 600/ 1200] Overall Loss 0.322024 Objective Loss 0.322024 LR 0.001000 Time 0.020354 -2022-12-06 10:32:35,722 - Epoch: [23][ 610/ 1200] Overall Loss 0.321731 Objective Loss 0.321731 LR 0.001000 Time 0.020332 -2022-12-06 10:32:35,911 - Epoch: [23][ 620/ 1200] Overall Loss 0.321722 Objective Loss 0.321722 LR 0.001000 Time 0.020309 -2022-12-06 10:32:36,102 - Epoch: [23][ 630/ 1200] Overall Loss 0.322067 Objective Loss 0.322067 LR 0.001000 Time 0.020288 -2022-12-06 10:32:36,292 - Epoch: [23][ 640/ 1200] Overall Loss 0.322100 Objective Loss 0.322100 LR 0.001000 Time 0.020267 -2022-12-06 10:32:36,483 - Epoch: [23][ 650/ 1200] Overall Loss 0.322072 Objective Loss 0.322072 LR 0.001000 Time 0.020247 -2022-12-06 10:32:36,672 - Epoch: [23][ 660/ 1200] Overall Loss 0.322427 Objective Loss 0.322427 LR 0.001000 Time 0.020227 -2022-12-06 10:32:36,863 - Epoch: [23][ 670/ 1200] Overall Loss 0.322403 Objective Loss 0.322403 LR 0.001000 Time 0.020209 -2022-12-06 10:32:37,053 - Epoch: [23][ 680/ 1200] Overall Loss 0.323119 Objective Loss 0.323119 LR 0.001000 Time 0.020191 -2022-12-06 10:32:37,244 - Epoch: [23][ 690/ 1200] Overall Loss 0.323177 Objective Loss 0.323177 LR 0.001000 Time 0.020174 -2022-12-06 10:32:37,434 - Epoch: [23][ 700/ 1200] Overall Loss 0.323009 Objective Loss 0.323009 LR 0.001000 Time 0.020157 -2022-12-06 10:32:37,625 - Epoch: [23][ 710/ 1200] Overall Loss 0.323055 Objective Loss 0.323055 LR 0.001000 Time 0.020140 -2022-12-06 10:32:37,815 - Epoch: [23][ 720/ 1200] Overall Loss 0.322906 Objective Loss 0.322906 LR 0.001000 Time 0.020125 -2022-12-06 10:32:38,005 - Epoch: [23][ 730/ 1200] Overall Loss 0.323048 Objective Loss 0.323048 LR 0.001000 Time 0.020108 -2022-12-06 10:32:38,195 - Epoch: [23][ 740/ 1200] Overall Loss 0.322719 Objective Loss 0.322719 LR 0.001000 Time 0.020092 -2022-12-06 10:32:38,385 - Epoch: [23][ 750/ 1200] Overall Loss 0.322674 Objective Loss 0.322674 LR 0.001000 Time 0.020077 -2022-12-06 10:32:38,575 - Epoch: [23][ 760/ 1200] Overall Loss 0.322321 Objective Loss 0.322321 LR 0.001000 Time 0.020062 -2022-12-06 10:32:38,765 - Epoch: [23][ 770/ 1200] Overall Loss 0.322437 Objective Loss 0.322437 LR 0.001000 Time 0.020048 -2022-12-06 10:32:38,955 - Epoch: [23][ 780/ 1200] Overall Loss 0.322762 Objective Loss 0.322762 LR 0.001000 Time 0.020033 -2022-12-06 10:32:39,145 - Epoch: [23][ 790/ 1200] Overall Loss 0.323249 Objective Loss 0.323249 LR 0.001000 Time 0.020020 -2022-12-06 10:32:39,335 - Epoch: [23][ 800/ 1200] Overall Loss 0.323140 Objective Loss 0.323140 LR 0.001000 Time 0.020007 -2022-12-06 10:32:39,525 - Epoch: [23][ 810/ 1200] Overall Loss 0.322784 Objective Loss 0.322784 LR 0.001000 Time 0.019994 -2022-12-06 10:32:39,715 - Epoch: [23][ 820/ 1200] Overall Loss 0.322738 Objective Loss 0.322738 LR 0.001000 Time 0.019981 -2022-12-06 10:32:39,906 - Epoch: [23][ 830/ 1200] Overall Loss 0.322563 Objective Loss 0.322563 LR 0.001000 Time 0.019969 -2022-12-06 10:32:40,095 - Epoch: [23][ 840/ 1200] Overall Loss 0.322640 Objective Loss 0.322640 LR 0.001000 Time 0.019956 -2022-12-06 10:32:40,285 - Epoch: [23][ 850/ 1200] Overall Loss 0.322526 Objective Loss 0.322526 LR 0.001000 Time 0.019944 -2022-12-06 10:32:40,475 - Epoch: [23][ 860/ 1200] Overall Loss 0.322348 Objective Loss 0.322348 LR 0.001000 Time 0.019932 -2022-12-06 10:32:40,665 - Epoch: [23][ 870/ 1200] Overall Loss 0.322598 Objective Loss 0.322598 LR 0.001000 Time 0.019921 -2022-12-06 10:32:40,855 - Epoch: [23][ 880/ 1200] Overall Loss 0.323174 Objective Loss 0.323174 LR 0.001000 Time 0.019910 -2022-12-06 10:32:41,045 - Epoch: [23][ 890/ 1200] Overall Loss 0.323160 Objective Loss 0.323160 LR 0.001000 Time 0.019900 -2022-12-06 10:32:41,235 - Epoch: [23][ 900/ 1200] Overall Loss 0.323152 Objective Loss 0.323152 LR 0.001000 Time 0.019889 -2022-12-06 10:32:41,425 - Epoch: [23][ 910/ 1200] Overall Loss 0.323154 Objective Loss 0.323154 LR 0.001000 Time 0.019879 -2022-12-06 10:32:41,615 - Epoch: [23][ 920/ 1200] Overall Loss 0.323359 Objective Loss 0.323359 LR 0.001000 Time 0.019868 -2022-12-06 10:32:41,805 - Epoch: [23][ 930/ 1200] Overall Loss 0.323280 Objective Loss 0.323280 LR 0.001000 Time 0.019859 -2022-12-06 10:32:41,995 - Epoch: [23][ 940/ 1200] Overall Loss 0.323564 Objective Loss 0.323564 LR 0.001000 Time 0.019848 -2022-12-06 10:32:42,185 - Epoch: [23][ 950/ 1200] Overall Loss 0.323301 Objective Loss 0.323301 LR 0.001000 Time 0.019839 -2022-12-06 10:32:42,375 - Epoch: [23][ 960/ 1200] Overall Loss 0.323223 Objective Loss 0.323223 LR 0.001000 Time 0.019829 -2022-12-06 10:32:42,565 - Epoch: [23][ 970/ 1200] Overall Loss 0.323049 Objective Loss 0.323049 LR 0.001000 Time 0.019821 -2022-12-06 10:32:42,755 - Epoch: [23][ 980/ 1200] Overall Loss 0.323462 Objective Loss 0.323462 LR 0.001000 Time 0.019812 -2022-12-06 10:32:42,946 - Epoch: [23][ 990/ 1200] Overall Loss 0.323308 Objective Loss 0.323308 LR 0.001000 Time 0.019804 -2022-12-06 10:32:43,136 - Epoch: [23][ 1000/ 1200] Overall Loss 0.323298 Objective Loss 0.323298 LR 0.001000 Time 0.019795 -2022-12-06 10:32:43,326 - Epoch: [23][ 1010/ 1200] Overall Loss 0.323140 Objective Loss 0.323140 LR 0.001000 Time 0.019787 -2022-12-06 10:32:43,516 - Epoch: [23][ 1020/ 1200] Overall Loss 0.322882 Objective Loss 0.322882 LR 0.001000 Time 0.019778 -2022-12-06 10:32:43,706 - Epoch: [23][ 1030/ 1200] Overall Loss 0.323159 Objective Loss 0.323159 LR 0.001000 Time 0.019771 -2022-12-06 10:32:43,895 - Epoch: [23][ 1040/ 1200] Overall Loss 0.323188 Objective Loss 0.323188 LR 0.001000 Time 0.019762 -2022-12-06 10:32:44,086 - Epoch: [23][ 1050/ 1200] Overall Loss 0.323051 Objective Loss 0.323051 LR 0.001000 Time 0.019755 -2022-12-06 10:32:44,275 - Epoch: [23][ 1060/ 1200] Overall Loss 0.323279 Objective Loss 0.323279 LR 0.001000 Time 0.019747 -2022-12-06 10:32:44,466 - Epoch: [23][ 1070/ 1200] Overall Loss 0.323394 Objective Loss 0.323394 LR 0.001000 Time 0.019740 -2022-12-06 10:32:44,656 - Epoch: [23][ 1080/ 1200] Overall Loss 0.323494 Objective Loss 0.323494 LR 0.001000 Time 0.019732 -2022-12-06 10:32:44,846 - Epoch: [23][ 1090/ 1200] Overall Loss 0.323353 Objective Loss 0.323353 LR 0.001000 Time 0.019725 -2022-12-06 10:32:45,035 - Epoch: [23][ 1100/ 1200] Overall Loss 0.323513 Objective Loss 0.323513 LR 0.001000 Time 0.019718 -2022-12-06 10:32:45,225 - Epoch: [23][ 1110/ 1200] Overall Loss 0.323495 Objective Loss 0.323495 LR 0.001000 Time 0.019711 -2022-12-06 10:32:45,416 - Epoch: [23][ 1120/ 1200] Overall Loss 0.323679 Objective Loss 0.323679 LR 0.001000 Time 0.019704 -2022-12-06 10:32:45,606 - Epoch: [23][ 1130/ 1200] Overall Loss 0.323637 Objective Loss 0.323637 LR 0.001000 Time 0.019697 -2022-12-06 10:32:45,795 - Epoch: [23][ 1140/ 1200] Overall Loss 0.323596 Objective Loss 0.323596 LR 0.001000 Time 0.019691 -2022-12-06 10:32:45,986 - Epoch: [23][ 1150/ 1200] Overall Loss 0.323482 Objective Loss 0.323482 LR 0.001000 Time 0.019684 -2022-12-06 10:32:46,175 - Epoch: [23][ 1160/ 1200] Overall Loss 0.323605 Objective Loss 0.323605 LR 0.001000 Time 0.019678 -2022-12-06 10:32:46,365 - Epoch: [23][ 1170/ 1200] Overall Loss 0.323452 Objective Loss 0.323452 LR 0.001000 Time 0.019671 -2022-12-06 10:32:46,555 - Epoch: [23][ 1180/ 1200] Overall Loss 0.323227 Objective Loss 0.323227 LR 0.001000 Time 0.019665 -2022-12-06 10:32:46,745 - Epoch: [23][ 1190/ 1200] Overall Loss 0.323507 Objective Loss 0.323507 LR 0.001000 Time 0.019659 -2022-12-06 10:32:46,967 - Epoch: [23][ 1200/ 1200] Overall Loss 0.323462 Objective Loss 0.323462 Top1 81.799163 Top5 97.698745 LR 0.001000 Time 0.019680 -2022-12-06 10:32:47,056 - --- validate (epoch=23)----------- -2022-12-06 10:32:47,056 - 34129 samples (256 per mini-batch) -2022-12-06 10:32:47,498 - Epoch: [23][ 10/ 134] Loss 0.317886 Top1 83.750000 Top5 97.539062 -2022-12-06 10:32:47,627 - Epoch: [23][ 20/ 134] Loss 0.314322 Top1 84.199219 Top5 97.929688 -2022-12-06 10:32:47,754 - Epoch: [23][ 30/ 134] Loss 0.321553 Top1 84.231771 Top5 97.760417 -2022-12-06 10:32:47,886 - Epoch: [23][ 40/ 134] Loss 0.328126 Top1 84.257812 Top5 97.822266 -2022-12-06 10:32:48,018 - Epoch: [23][ 50/ 134] Loss 0.323498 Top1 84.007812 Top5 97.851562 -2022-12-06 10:32:48,146 - Epoch: [23][ 60/ 134] Loss 0.321609 Top1 83.860677 Top5 97.871094 -2022-12-06 10:32:48,269 - Epoch: [23][ 70/ 134] Loss 0.326365 Top1 83.733259 Top5 97.823661 -2022-12-06 10:32:48,394 - Epoch: [23][ 80/ 134] Loss 0.329448 Top1 83.583984 Top5 97.802734 -2022-12-06 10:32:48,527 - Epoch: [23][ 90/ 134] Loss 0.327789 Top1 83.689236 Top5 97.834201 -2022-12-06 10:32:48,655 - Epoch: [23][ 100/ 134] Loss 0.327666 Top1 83.652344 Top5 97.851562 -2022-12-06 10:32:48,782 - Epoch: [23][ 110/ 134] Loss 0.330071 Top1 83.657670 Top5 97.844460 -2022-12-06 10:32:48,908 - Epoch: [23][ 120/ 134] Loss 0.328780 Top1 83.746745 Top5 97.854818 -2022-12-06 10:32:49,035 - Epoch: [23][ 130/ 134] Loss 0.329018 Top1 83.783053 Top5 97.854567 -2022-12-06 10:32:49,072 - Epoch: [23][ 134/ 134] Loss 0.330696 Top1 83.755750 Top5 97.858127 -2022-12-06 10:32:49,158 - ==> Top1: 83.756 Top5: 97.858 Loss: 0.331 - -2022-12-06 10:32:49,159 - ==> Confusion: -[[ 873 1 2 1 15 3 0 1 4 74 0 4 1 3 2 3 2 1 1 0 5] - [ 1 888 3 1 26 31 6 21 3 0 6 5 4 5 2 1 7 0 8 2 7] - [ 6 2 981 24 5 2 21 15 0 4 4 5 3 6 0 6 3 2 2 3 9] - [ 2 0 15 922 0 6 0 1 3 1 14 0 6 4 21 0 3 3 16 0 3] - [ 6 2 2 0 962 3 0 1 0 10 0 3 0 8 4 6 4 3 0 2 4] - [ 7 16 2 1 12 918 8 30 2 2 1 16 5 22 0 2 6 0 4 8 7] - [ 0 4 17 5 0 3 1050 9 0 0 3 3 2 0 0 7 1 1 1 6 6] - [ 3 7 14 2 1 21 4 927 0 1 1 9 1 2 1 2 0 0 40 14 4] - [ 5 3 1 0 0 2 1 0 960 55 4 1 2 12 7 0 4 2 1 1 3] - [ 61 0 1 0 6 2 0 3 16 888 1 1 0 13 1 0 0 1 1 0 6] - [ 0 3 4 11 1 1 1 2 12 3 924 1 3 26 5 0 1 1 8 2 10] - [ 6 2 3 0 0 10 3 5 0 0 1 941 39 8 0 5 4 7 1 8 8] - [ 1 1 1 1 0 2 0 1 0 1 0 24 899 2 0 13 2 8 1 3 9] - [ 1 3 0 0 1 3 0 2 12 15 5 7 2 959 0 3 2 0 0 0 8] - [ 12 5 1 12 11 2 0 1 24 10 0 1 3 13 1011 0 2 3 5 1 13] - [ 2 0 2 0 2 2 2 0 0 2 0 9 7 5 0 991 3 4 0 5 7] - [ 5 6 2 0 0 1 1 1 2 1 0 2 0 2 1 18 1018 2 1 3 6] - [ 3 1 2 5 0 1 1 2 2 2 0 7 33 2 1 24 2 944 1 1 2] - [ 2 4 3 11 1 1 1 16 3 0 3 6 2 1 8 0 5 0 935 2 4] - [ 1 2 3 0 0 5 8 7 0 0 0 24 5 6 1 5 11 2 1 985 14] - [ 138 193 181 108 233 140 75 141 97 102 147 121 452 463 129 191 189 68 243 206 9609]] - -2022-12-06 10:32:49,732 - ==> Best [Top1: 83.756 Top5: 97.858 Sparsity:0.00 Params: 5376 on epoch: 23] -2022-12-06 10:32:49,732 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:32:49,739 - - -2022-12-06 10:32:49,739 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:32:50,775 - Epoch: [24][ 10/ 1200] Overall Loss 0.323147 Objective Loss 0.323147 LR 0.001000 Time 0.103559 -2022-12-06 10:32:50,966 - Epoch: [24][ 20/ 1200] Overall Loss 0.334270 Objective Loss 0.334270 LR 0.001000 Time 0.061310 -2022-12-06 10:32:51,156 - Epoch: [24][ 30/ 1200] Overall Loss 0.333269 Objective Loss 0.333269 LR 0.001000 Time 0.047184 -2022-12-06 10:32:51,346 - Epoch: [24][ 40/ 1200] Overall Loss 0.326433 Objective Loss 0.326433 LR 0.001000 Time 0.040122 -2022-12-06 10:32:51,536 - Epoch: [24][ 50/ 1200] Overall Loss 0.325734 Objective Loss 0.325734 LR 0.001000 Time 0.035876 -2022-12-06 10:32:51,725 - Epoch: [24][ 60/ 1200] Overall Loss 0.321005 Objective Loss 0.321005 LR 0.001000 Time 0.033047 -2022-12-06 10:32:51,915 - Epoch: [24][ 70/ 1200] Overall Loss 0.321967 Objective Loss 0.321967 LR 0.001000 Time 0.031023 -2022-12-06 10:32:52,104 - Epoch: [24][ 80/ 1200] Overall Loss 0.316092 Objective Loss 0.316092 LR 0.001000 Time 0.029507 -2022-12-06 10:32:52,293 - Epoch: [24][ 90/ 1200] Overall Loss 0.314867 Objective Loss 0.314867 LR 0.001000 Time 0.028323 -2022-12-06 10:32:52,481 - Epoch: [24][ 100/ 1200] Overall Loss 0.313250 Objective Loss 0.313250 LR 0.001000 Time 0.027369 -2022-12-06 10:32:52,670 - Epoch: [24][ 110/ 1200] Overall Loss 0.313476 Objective Loss 0.313476 LR 0.001000 Time 0.026591 -2022-12-06 10:32:52,860 - Epoch: [24][ 120/ 1200] Overall Loss 0.311301 Objective Loss 0.311301 LR 0.001000 Time 0.025952 -2022-12-06 10:32:53,050 - Epoch: [24][ 130/ 1200] Overall Loss 0.309366 Objective Loss 0.309366 LR 0.001000 Time 0.025410 -2022-12-06 10:32:53,239 - Epoch: [24][ 140/ 1200] Overall Loss 0.311188 Objective Loss 0.311188 LR 0.001000 Time 0.024945 -2022-12-06 10:32:53,429 - Epoch: [24][ 150/ 1200] Overall Loss 0.311721 Objective Loss 0.311721 LR 0.001000 Time 0.024545 -2022-12-06 10:32:53,619 - Epoch: [24][ 160/ 1200] Overall Loss 0.309536 Objective Loss 0.309536 LR 0.001000 Time 0.024195 -2022-12-06 10:32:53,809 - Epoch: [24][ 170/ 1200] Overall Loss 0.308509 Objective Loss 0.308509 LR 0.001000 Time 0.023885 -2022-12-06 10:32:53,998 - Epoch: [24][ 180/ 1200] Overall Loss 0.308210 Objective Loss 0.308210 LR 0.001000 Time 0.023606 -2022-12-06 10:32:54,187 - Epoch: [24][ 190/ 1200] Overall Loss 0.307663 Objective Loss 0.307663 LR 0.001000 Time 0.023357 -2022-12-06 10:32:54,376 - Epoch: [24][ 200/ 1200] Overall Loss 0.306444 Objective Loss 0.306444 LR 0.001000 Time 0.023132 -2022-12-06 10:32:54,566 - Epoch: [24][ 210/ 1200] Overall Loss 0.306023 Objective Loss 0.306023 LR 0.001000 Time 0.022932 -2022-12-06 10:32:54,756 - Epoch: [24][ 220/ 1200] Overall Loss 0.306011 Objective Loss 0.306011 LR 0.001000 Time 0.022748 -2022-12-06 10:32:54,945 - Epoch: [24][ 230/ 1200] Overall Loss 0.306644 Objective Loss 0.306644 LR 0.001000 Time 0.022580 -2022-12-06 10:32:55,134 - Epoch: [24][ 240/ 1200] Overall Loss 0.306414 Objective Loss 0.306414 LR 0.001000 Time 0.022426 -2022-12-06 10:32:55,324 - Epoch: [24][ 250/ 1200] Overall Loss 0.307017 Objective Loss 0.307017 LR 0.001000 Time 0.022284 -2022-12-06 10:32:55,513 - Epoch: [24][ 260/ 1200] Overall Loss 0.306418 Objective Loss 0.306418 LR 0.001000 Time 0.022154 -2022-12-06 10:32:55,703 - Epoch: [24][ 270/ 1200] Overall Loss 0.306616 Objective Loss 0.306616 LR 0.001000 Time 0.022033 -2022-12-06 10:32:55,892 - Epoch: [24][ 280/ 1200] Overall Loss 0.307077 Objective Loss 0.307077 LR 0.001000 Time 0.021920 -2022-12-06 10:32:56,081 - Epoch: [24][ 290/ 1200] Overall Loss 0.308434 Objective Loss 0.308434 LR 0.001000 Time 0.021815 -2022-12-06 10:32:56,271 - Epoch: [24][ 300/ 1200] Overall Loss 0.308758 Objective Loss 0.308758 LR 0.001000 Time 0.021717 -2022-12-06 10:32:56,460 - Epoch: [24][ 310/ 1200] Overall Loss 0.309421 Objective Loss 0.309421 LR 0.001000 Time 0.021625 -2022-12-06 10:32:56,650 - Epoch: [24][ 320/ 1200] Overall Loss 0.309651 Objective Loss 0.309651 LR 0.001000 Time 0.021541 -2022-12-06 10:32:56,840 - Epoch: [24][ 330/ 1200] Overall Loss 0.309431 Objective Loss 0.309431 LR 0.001000 Time 0.021462 -2022-12-06 10:32:57,029 - Epoch: [24][ 340/ 1200] Overall Loss 0.310523 Objective Loss 0.310523 LR 0.001000 Time 0.021387 -2022-12-06 10:32:57,219 - Epoch: [24][ 350/ 1200] Overall Loss 0.310675 Objective Loss 0.310675 LR 0.001000 Time 0.021316 -2022-12-06 10:32:57,408 - Epoch: [24][ 360/ 1200] Overall Loss 0.310594 Objective Loss 0.310594 LR 0.001000 Time 0.021249 -2022-12-06 10:32:57,598 - Epoch: [24][ 370/ 1200] Overall Loss 0.310482 Objective Loss 0.310482 LR 0.001000 Time 0.021186 -2022-12-06 10:32:57,788 - Epoch: [24][ 380/ 1200] Overall Loss 0.310122 Objective Loss 0.310122 LR 0.001000 Time 0.021126 -2022-12-06 10:32:57,978 - Epoch: [24][ 390/ 1200] Overall Loss 0.309704 Objective Loss 0.309704 LR 0.001000 Time 0.021070 -2022-12-06 10:32:58,167 - Epoch: [24][ 400/ 1200] Overall Loss 0.309765 Objective Loss 0.309765 LR 0.001000 Time 0.021015 -2022-12-06 10:32:58,357 - Epoch: [24][ 410/ 1200] Overall Loss 0.309588 Objective Loss 0.309588 LR 0.001000 Time 0.020964 -2022-12-06 10:32:58,546 - Epoch: [24][ 420/ 1200] Overall Loss 0.308951 Objective Loss 0.308951 LR 0.001000 Time 0.020913 -2022-12-06 10:32:58,735 - Epoch: [24][ 430/ 1200] Overall Loss 0.308671 Objective Loss 0.308671 LR 0.001000 Time 0.020866 -2022-12-06 10:32:58,924 - Epoch: [24][ 440/ 1200] Overall Loss 0.308078 Objective Loss 0.308078 LR 0.001000 Time 0.020820 -2022-12-06 10:32:59,114 - Epoch: [24][ 450/ 1200] Overall Loss 0.308077 Objective Loss 0.308077 LR 0.001000 Time 0.020778 -2022-12-06 10:32:59,303 - Epoch: [24][ 460/ 1200] Overall Loss 0.307872 Objective Loss 0.307872 LR 0.001000 Time 0.020737 -2022-12-06 10:32:59,493 - Epoch: [24][ 470/ 1200] Overall Loss 0.307940 Objective Loss 0.307940 LR 0.001000 Time 0.020699 -2022-12-06 10:32:59,683 - Epoch: [24][ 480/ 1200] Overall Loss 0.308650 Objective Loss 0.308650 LR 0.001000 Time 0.020663 -2022-12-06 10:32:59,874 - Epoch: [24][ 490/ 1200] Overall Loss 0.309411 Objective Loss 0.309411 LR 0.001000 Time 0.020629 -2022-12-06 10:33:00,064 - Epoch: [24][ 500/ 1200] Overall Loss 0.310004 Objective Loss 0.310004 LR 0.001000 Time 0.020595 -2022-12-06 10:33:00,255 - Epoch: [24][ 510/ 1200] Overall Loss 0.310174 Objective Loss 0.310174 LR 0.001000 Time 0.020564 -2022-12-06 10:33:00,444 - Epoch: [24][ 520/ 1200] Overall Loss 0.311522 Objective Loss 0.311522 LR 0.001000 Time 0.020532 -2022-12-06 10:33:00,635 - Epoch: [24][ 530/ 1200] Overall Loss 0.311993 Objective Loss 0.311993 LR 0.001000 Time 0.020504 -2022-12-06 10:33:00,824 - Epoch: [24][ 540/ 1200] Overall Loss 0.311856 Objective Loss 0.311856 LR 0.001000 Time 0.020473 -2022-12-06 10:33:01,015 - Epoch: [24][ 550/ 1200] Overall Loss 0.312085 Objective Loss 0.312085 LR 0.001000 Time 0.020447 -2022-12-06 10:33:01,204 - Epoch: [24][ 560/ 1200] Overall Loss 0.311677 Objective Loss 0.311677 LR 0.001000 Time 0.020419 -2022-12-06 10:33:01,396 - Epoch: [24][ 570/ 1200] Overall Loss 0.311588 Objective Loss 0.311588 LR 0.001000 Time 0.020397 -2022-12-06 10:33:01,592 - Epoch: [24][ 580/ 1200] Overall Loss 0.311706 Objective Loss 0.311706 LR 0.001000 Time 0.020382 -2022-12-06 10:33:01,789 - Epoch: [24][ 590/ 1200] Overall Loss 0.311563 Objective Loss 0.311563 LR 0.001000 Time 0.020368 -2022-12-06 10:33:01,984 - Epoch: [24][ 600/ 1200] Overall Loss 0.311244 Objective Loss 0.311244 LR 0.001000 Time 0.020354 -2022-12-06 10:33:02,180 - Epoch: [24][ 610/ 1200] Overall Loss 0.311349 Objective Loss 0.311349 LR 0.001000 Time 0.020340 -2022-12-06 10:33:02,376 - Epoch: [24][ 620/ 1200] Overall Loss 0.311461 Objective Loss 0.311461 LR 0.001000 Time 0.020327 -2022-12-06 10:33:02,571 - Epoch: [24][ 630/ 1200] Overall Loss 0.312014 Objective Loss 0.312014 LR 0.001000 Time 0.020314 -2022-12-06 10:33:02,767 - Epoch: [24][ 640/ 1200] Overall Loss 0.311665 Objective Loss 0.311665 LR 0.001000 Time 0.020302 -2022-12-06 10:33:02,963 - Epoch: [24][ 650/ 1200] Overall Loss 0.311447 Objective Loss 0.311447 LR 0.001000 Time 0.020290 -2022-12-06 10:33:03,159 - Epoch: [24][ 660/ 1200] Overall Loss 0.311840 Objective Loss 0.311840 LR 0.001000 Time 0.020278 -2022-12-06 10:33:03,355 - Epoch: [24][ 670/ 1200] Overall Loss 0.311405 Objective Loss 0.311405 LR 0.001000 Time 0.020267 -2022-12-06 10:33:03,550 - Epoch: [24][ 680/ 1200] Overall Loss 0.311583 Objective Loss 0.311583 LR 0.001000 Time 0.020256 -2022-12-06 10:33:03,746 - Epoch: [24][ 690/ 1200] Overall Loss 0.311596 Objective Loss 0.311596 LR 0.001000 Time 0.020245 -2022-12-06 10:33:03,942 - Epoch: [24][ 700/ 1200] Overall Loss 0.311792 Objective Loss 0.311792 LR 0.001000 Time 0.020235 -2022-12-06 10:33:04,138 - Epoch: [24][ 710/ 1200] Overall Loss 0.311846 Objective Loss 0.311846 LR 0.001000 Time 0.020225 -2022-12-06 10:33:04,334 - Epoch: [24][ 720/ 1200] Overall Loss 0.311980 Objective Loss 0.311980 LR 0.001000 Time 0.020215 -2022-12-06 10:33:04,530 - Epoch: [24][ 730/ 1200] Overall Loss 0.312336 Objective Loss 0.312336 LR 0.001000 Time 0.020206 -2022-12-06 10:33:04,726 - Epoch: [24][ 740/ 1200] Overall Loss 0.312196 Objective Loss 0.312196 LR 0.001000 Time 0.020197 -2022-12-06 10:33:04,922 - Epoch: [24][ 750/ 1200] Overall Loss 0.312212 Objective Loss 0.312212 LR 0.001000 Time 0.020188 -2022-12-06 10:33:05,117 - Epoch: [24][ 760/ 1200] Overall Loss 0.312282 Objective Loss 0.312282 LR 0.001000 Time 0.020180 -2022-12-06 10:33:05,313 - Epoch: [24][ 770/ 1200] Overall Loss 0.312040 Objective Loss 0.312040 LR 0.001000 Time 0.020171 -2022-12-06 10:33:05,509 - Epoch: [24][ 780/ 1200] Overall Loss 0.311905 Objective Loss 0.311905 LR 0.001000 Time 0.020163 -2022-12-06 10:33:05,705 - Epoch: [24][ 790/ 1200] Overall Loss 0.311883 Objective Loss 0.311883 LR 0.001000 Time 0.020155 -2022-12-06 10:33:05,901 - Epoch: [24][ 800/ 1200] Overall Loss 0.312267 Objective Loss 0.312267 LR 0.001000 Time 0.020147 -2022-12-06 10:33:06,097 - Epoch: [24][ 810/ 1200] Overall Loss 0.312181 Objective Loss 0.312181 LR 0.001000 Time 0.020140 -2022-12-06 10:33:06,293 - Epoch: [24][ 820/ 1200] Overall Loss 0.311890 Objective Loss 0.311890 LR 0.001000 Time 0.020132 -2022-12-06 10:33:06,489 - Epoch: [24][ 830/ 1200] Overall Loss 0.312034 Objective Loss 0.312034 LR 0.001000 Time 0.020125 -2022-12-06 10:33:06,683 - Epoch: [24][ 840/ 1200] Overall Loss 0.311773 Objective Loss 0.311773 LR 0.001000 Time 0.020117 -2022-12-06 10:33:06,874 - Epoch: [24][ 850/ 1200] Overall Loss 0.311782 Objective Loss 0.311782 LR 0.001000 Time 0.020103 -2022-12-06 10:33:07,063 - Epoch: [24][ 860/ 1200] Overall Loss 0.311630 Objective Loss 0.311630 LR 0.001000 Time 0.020089 -2022-12-06 10:33:07,254 - Epoch: [24][ 870/ 1200] Overall Loss 0.311637 Objective Loss 0.311637 LR 0.001000 Time 0.020077 -2022-12-06 10:33:07,444 - Epoch: [24][ 880/ 1200] Overall Loss 0.311456 Objective Loss 0.311456 LR 0.001000 Time 0.020064 -2022-12-06 10:33:07,635 - Epoch: [24][ 890/ 1200] Overall Loss 0.311001 Objective Loss 0.311001 LR 0.001000 Time 0.020052 -2022-12-06 10:33:07,825 - Epoch: [24][ 900/ 1200] Overall Loss 0.311443 Objective Loss 0.311443 LR 0.001000 Time 0.020040 -2022-12-06 10:33:08,014 - Epoch: [24][ 910/ 1200] Overall Loss 0.311607 Objective Loss 0.311607 LR 0.001000 Time 0.020027 -2022-12-06 10:33:08,204 - Epoch: [24][ 920/ 1200] Overall Loss 0.311597 Objective Loss 0.311597 LR 0.001000 Time 0.020015 -2022-12-06 10:33:08,394 - Epoch: [24][ 930/ 1200] Overall Loss 0.311565 Objective Loss 0.311565 LR 0.001000 Time 0.020003 -2022-12-06 10:33:08,583 - Epoch: [24][ 940/ 1200] Overall Loss 0.312205 Objective Loss 0.312205 LR 0.001000 Time 0.019992 -2022-12-06 10:33:08,773 - Epoch: [24][ 950/ 1200] Overall Loss 0.311901 Objective Loss 0.311901 LR 0.001000 Time 0.019981 -2022-12-06 10:33:08,964 - Epoch: [24][ 960/ 1200] Overall Loss 0.312047 Objective Loss 0.312047 LR 0.001000 Time 0.019971 -2022-12-06 10:33:09,155 - Epoch: [24][ 970/ 1200] Overall Loss 0.312347 Objective Loss 0.312347 LR 0.001000 Time 0.019961 -2022-12-06 10:33:09,344 - Epoch: [24][ 980/ 1200] Overall Loss 0.312015 Objective Loss 0.312015 LR 0.001000 Time 0.019950 -2022-12-06 10:33:09,534 - Epoch: [24][ 990/ 1200] Overall Loss 0.312535 Objective Loss 0.312535 LR 0.001000 Time 0.019940 -2022-12-06 10:33:09,724 - Epoch: [24][ 1000/ 1200] Overall Loss 0.312475 Objective Loss 0.312475 LR 0.001000 Time 0.019930 -2022-12-06 10:33:09,914 - Epoch: [24][ 1010/ 1200] Overall Loss 0.312650 Objective Loss 0.312650 LR 0.001000 Time 0.019920 -2022-12-06 10:33:10,104 - Epoch: [24][ 1020/ 1200] Overall Loss 0.312837 Objective Loss 0.312837 LR 0.001000 Time 0.019910 -2022-12-06 10:33:10,294 - Epoch: [24][ 1030/ 1200] Overall Loss 0.313079 Objective Loss 0.313079 LR 0.001000 Time 0.019900 -2022-12-06 10:33:10,483 - Epoch: [24][ 1040/ 1200] Overall Loss 0.313117 Objective Loss 0.313117 LR 0.001000 Time 0.019891 -2022-12-06 10:33:10,674 - Epoch: [24][ 1050/ 1200] Overall Loss 0.312916 Objective Loss 0.312916 LR 0.001000 Time 0.019883 -2022-12-06 10:33:10,864 - Epoch: [24][ 1060/ 1200] Overall Loss 0.312806 Objective Loss 0.312806 LR 0.001000 Time 0.019874 -2022-12-06 10:33:11,053 - Epoch: [24][ 1070/ 1200] Overall Loss 0.312732 Objective Loss 0.312732 LR 0.001000 Time 0.019864 -2022-12-06 10:33:11,243 - Epoch: [24][ 1080/ 1200] Overall Loss 0.312652 Objective Loss 0.312652 LR 0.001000 Time 0.019856 -2022-12-06 10:33:11,434 - Epoch: [24][ 1090/ 1200] Overall Loss 0.312642 Objective Loss 0.312642 LR 0.001000 Time 0.019848 -2022-12-06 10:33:11,623 - Epoch: [24][ 1100/ 1200] Overall Loss 0.312587 Objective Loss 0.312587 LR 0.001000 Time 0.019839 -2022-12-06 10:33:11,813 - Epoch: [24][ 1110/ 1200] Overall Loss 0.312662 Objective Loss 0.312662 LR 0.001000 Time 0.019831 -2022-12-06 10:33:12,003 - Epoch: [24][ 1120/ 1200] Overall Loss 0.312755 Objective Loss 0.312755 LR 0.001000 Time 0.019823 -2022-12-06 10:33:12,193 - Epoch: [24][ 1130/ 1200] Overall Loss 0.312622 Objective Loss 0.312622 LR 0.001000 Time 0.019815 -2022-12-06 10:33:12,383 - Epoch: [24][ 1140/ 1200] Overall Loss 0.312522 Objective Loss 0.312522 LR 0.001000 Time 0.019808 -2022-12-06 10:33:12,573 - Epoch: [24][ 1150/ 1200] Overall Loss 0.312294 Objective Loss 0.312294 LR 0.001000 Time 0.019800 -2022-12-06 10:33:12,763 - Epoch: [24][ 1160/ 1200] Overall Loss 0.312461 Objective Loss 0.312461 LR 0.001000 Time 0.019793 -2022-12-06 10:33:12,953 - Epoch: [24][ 1170/ 1200] Overall Loss 0.312869 Objective Loss 0.312869 LR 0.001000 Time 0.019786 -2022-12-06 10:33:13,143 - Epoch: [24][ 1180/ 1200] Overall Loss 0.313059 Objective Loss 0.313059 LR 0.001000 Time 0.019779 -2022-12-06 10:33:13,334 - Epoch: [24][ 1190/ 1200] Overall Loss 0.312934 Objective Loss 0.312934 LR 0.001000 Time 0.019772 -2022-12-06 10:33:13,562 - Epoch: [24][ 1200/ 1200] Overall Loss 0.313023 Objective Loss 0.313023 Top1 82.426778 Top5 97.280335 LR 0.001000 Time 0.019797 -2022-12-06 10:33:13,650 - --- validate (epoch=24)----------- -2022-12-06 10:33:13,650 - 34129 samples (256 per mini-batch) -2022-12-06 10:33:14,111 - Epoch: [24][ 10/ 134] Loss 0.330202 Top1 82.578125 Top5 97.460938 -2022-12-06 10:33:14,255 - Epoch: [24][ 20/ 134] Loss 0.316887 Top1 82.343750 Top5 97.734375 -2022-12-06 10:33:14,404 - Epoch: [24][ 30/ 134] Loss 0.308437 Top1 82.486979 Top5 97.656250 -2022-12-06 10:33:14,538 - Epoch: [24][ 40/ 134] Loss 0.312534 Top1 82.236328 Top5 97.685547 -2022-12-06 10:33:14,682 - Epoch: [24][ 50/ 134] Loss 0.312679 Top1 82.312500 Top5 97.679688 -2022-12-06 10:33:14,828 - Epoch: [24][ 60/ 134] Loss 0.310812 Top1 82.246094 Top5 97.701823 -2022-12-06 10:33:14,975 - Epoch: [24][ 70/ 134] Loss 0.308364 Top1 82.271205 Top5 97.639509 -2022-12-06 10:33:15,120 - Epoch: [24][ 80/ 134] Loss 0.311006 Top1 82.192383 Top5 97.636719 -2022-12-06 10:33:15,268 - Epoch: [24][ 90/ 134] Loss 0.309826 Top1 82.083333 Top5 97.604167 -2022-12-06 10:33:15,413 - Epoch: [24][ 100/ 134] Loss 0.312182 Top1 82.101562 Top5 97.582031 -2022-12-06 10:33:15,560 - Epoch: [24][ 110/ 134] Loss 0.313140 Top1 82.123580 Top5 97.585227 -2022-12-06 10:33:15,707 - Epoch: [24][ 120/ 134] Loss 0.314909 Top1 82.070312 Top5 97.535807 -2022-12-06 10:33:15,841 - Epoch: [24][ 130/ 134] Loss 0.313726 Top1 82.085337 Top5 97.506010 -2022-12-06 10:33:15,878 - Epoch: [24][ 134/ 134] Loss 0.314279 Top1 82.076826 Top5 97.521170 -2022-12-06 10:33:15,973 - ==> Top1: 82.077 Top5: 97.521 Loss: 0.314 - -2022-12-06 10:33:15,974 - ==> Confusion: -[[ 909 4 0 0 5 2 0 5 9 50 0 1 1 0 4 2 2 0 0 0 2] - [ 2 939 0 2 6 32 3 18 0 3 4 1 0 1 2 1 3 0 4 1 5] - [ 9 3 987 13 2 1 32 15 0 2 3 6 2 4 1 3 2 1 3 7 7] - [ 3 2 27 910 1 4 0 4 1 0 12 1 2 2 23 0 3 7 12 1 5] - [ 10 13 3 0 948 3 1 3 1 7 0 2 1 1 8 8 5 3 1 2 0] - [ 3 30 1 2 7 942 7 29 1 3 0 10 1 11 2 0 7 1 1 7 4] - [ 0 2 21 1 0 2 1062 6 0 0 0 1 0 0 0 7 2 1 2 9 2] - [ 2 10 6 0 1 27 3 957 0 1 0 3 1 1 0 4 1 1 21 11 4] - [ 7 6 0 1 0 0 0 3 979 47 4 0 3 1 6 0 3 1 0 1 2] - [ 88 1 2 0 1 3 0 1 22 870 1 1 0 3 1 1 0 2 1 1 2] - [ 1 2 3 11 0 2 4 8 11 1 945 2 1 13 6 1 0 0 5 1 2] - [ 7 4 3 0 0 18 4 7 2 0 2 933 15 2 0 12 8 10 1 20 3] - [ 1 2 3 7 1 2 1 3 3 0 1 30 848 1 0 13 2 29 3 8 11] - [ 2 1 1 0 1 12 0 3 12 27 9 10 2 918 1 6 3 3 0 5 7] - [ 13 6 2 13 3 3 0 1 22 2 2 2 0 6 1027 2 5 4 6 3 8] - [ 2 4 4 1 2 2 3 0 0 1 0 6 1 3 3 983 5 15 0 5 3] - [ 3 6 1 1 4 2 2 1 0 1 0 0 0 0 1 16 1019 3 1 7 4] - [ 3 1 2 4 0 1 1 1 0 2 1 5 5 1 0 17 1 985 1 4 1] - [ 3 6 1 5 0 3 1 34 2 0 9 2 2 2 6 0 1 1 923 4 3] - [ 4 1 3 0 0 5 10 8 1 1 0 13 2 2 0 1 7 6 1 1013 2] - [ 233 327 238 135 171 190 115 254 115 119 226 98 353 337 196 189 303 123 182 409 8913]] - -2022-12-06 10:33:16,555 - ==> Best [Top1: 83.756 Top5: 97.858 Sparsity:0.00 Params: 5376 on epoch: 23] -2022-12-06 10:33:16,555 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:33:16,561 - - -2022-12-06 10:33:16,561 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:33:17,498 - Epoch: [25][ 10/ 1200] Overall Loss 0.313986 Objective Loss 0.313986 LR 0.001000 Time 0.093663 -2022-12-06 10:33:17,698 - Epoch: [25][ 20/ 1200] Overall Loss 0.289490 Objective Loss 0.289490 LR 0.001000 Time 0.056774 -2022-12-06 10:33:17,891 - Epoch: [25][ 30/ 1200] Overall Loss 0.297929 Objective Loss 0.297929 LR 0.001000 Time 0.044255 -2022-12-06 10:33:18,082 - Epoch: [25][ 40/ 1200] Overall Loss 0.298628 Objective Loss 0.298628 LR 0.001000 Time 0.037966 -2022-12-06 10:33:18,274 - Epoch: [25][ 50/ 1200] Overall Loss 0.294001 Objective Loss 0.294001 LR 0.001000 Time 0.034197 -2022-12-06 10:33:18,466 - Epoch: [25][ 60/ 1200] Overall Loss 0.296717 Objective Loss 0.296717 LR 0.001000 Time 0.031683 -2022-12-06 10:33:18,657 - Epoch: [25][ 70/ 1200] Overall Loss 0.296324 Objective Loss 0.296324 LR 0.001000 Time 0.029879 -2022-12-06 10:33:18,848 - Epoch: [25][ 80/ 1200] Overall Loss 0.295146 Objective Loss 0.295146 LR 0.001000 Time 0.028524 -2022-12-06 10:33:19,039 - Epoch: [25][ 90/ 1200] Overall Loss 0.293368 Objective Loss 0.293368 LR 0.001000 Time 0.027473 -2022-12-06 10:33:19,230 - Epoch: [25][ 100/ 1200] Overall Loss 0.295209 Objective Loss 0.295209 LR 0.001000 Time 0.026633 -2022-12-06 10:33:19,422 - Epoch: [25][ 110/ 1200] Overall Loss 0.294333 Objective Loss 0.294333 LR 0.001000 Time 0.025948 -2022-12-06 10:33:19,612 - Epoch: [25][ 120/ 1200] Overall Loss 0.295403 Objective Loss 0.295403 LR 0.001000 Time 0.025371 -2022-12-06 10:33:19,804 - Epoch: [25][ 130/ 1200] Overall Loss 0.295077 Objective Loss 0.295077 LR 0.001000 Time 0.024888 -2022-12-06 10:33:19,996 - Epoch: [25][ 140/ 1200] Overall Loss 0.292967 Objective Loss 0.292967 LR 0.001000 Time 0.024475 -2022-12-06 10:33:20,187 - Epoch: [25][ 150/ 1200] Overall Loss 0.296069 Objective Loss 0.296069 LR 0.001000 Time 0.024113 -2022-12-06 10:33:20,378 - Epoch: [25][ 160/ 1200] Overall Loss 0.296847 Objective Loss 0.296847 LR 0.001000 Time 0.023800 -2022-12-06 10:33:20,570 - Epoch: [25][ 170/ 1200] Overall Loss 0.297938 Objective Loss 0.297938 LR 0.001000 Time 0.023523 -2022-12-06 10:33:20,762 - Epoch: [25][ 180/ 1200] Overall Loss 0.299833 Objective Loss 0.299833 LR 0.001000 Time 0.023280 -2022-12-06 10:33:20,953 - Epoch: [25][ 190/ 1200] Overall Loss 0.299383 Objective Loss 0.299383 LR 0.001000 Time 0.023058 -2022-12-06 10:33:21,145 - Epoch: [25][ 200/ 1200] Overall Loss 0.297290 Objective Loss 0.297290 LR 0.001000 Time 0.022862 -2022-12-06 10:33:21,336 - Epoch: [25][ 210/ 1200] Overall Loss 0.298782 Objective Loss 0.298782 LR 0.001000 Time 0.022680 -2022-12-06 10:33:21,528 - Epoch: [25][ 220/ 1200] Overall Loss 0.299926 Objective Loss 0.299926 LR 0.001000 Time 0.022518 -2022-12-06 10:33:21,720 - Epoch: [25][ 230/ 1200] Overall Loss 0.299733 Objective Loss 0.299733 LR 0.001000 Time 0.022371 -2022-12-06 10:33:21,911 - Epoch: [25][ 240/ 1200] Overall Loss 0.300070 Objective Loss 0.300070 LR 0.001000 Time 0.022235 -2022-12-06 10:33:22,103 - Epoch: [25][ 250/ 1200] Overall Loss 0.300142 Objective Loss 0.300142 LR 0.001000 Time 0.022111 -2022-12-06 10:33:22,295 - Epoch: [25][ 260/ 1200] Overall Loss 0.301897 Objective Loss 0.301897 LR 0.001000 Time 0.021995 -2022-12-06 10:33:22,486 - Epoch: [25][ 270/ 1200] Overall Loss 0.301168 Objective Loss 0.301168 LR 0.001000 Time 0.021886 -2022-12-06 10:33:22,677 - Epoch: [25][ 280/ 1200] Overall Loss 0.301834 Objective Loss 0.301834 LR 0.001000 Time 0.021787 -2022-12-06 10:33:22,869 - Epoch: [25][ 290/ 1200] Overall Loss 0.301931 Objective Loss 0.301931 LR 0.001000 Time 0.021695 -2022-12-06 10:33:23,060 - Epoch: [25][ 300/ 1200] Overall Loss 0.301918 Objective Loss 0.301918 LR 0.001000 Time 0.021608 -2022-12-06 10:33:23,252 - Epoch: [25][ 310/ 1200] Overall Loss 0.301293 Objective Loss 0.301293 LR 0.001000 Time 0.021527 -2022-12-06 10:33:23,444 - Epoch: [25][ 320/ 1200] Overall Loss 0.301885 Objective Loss 0.301885 LR 0.001000 Time 0.021452 -2022-12-06 10:33:23,635 - Epoch: [25][ 330/ 1200] Overall Loss 0.303359 Objective Loss 0.303359 LR 0.001000 Time 0.021381 -2022-12-06 10:33:23,827 - Epoch: [25][ 340/ 1200] Overall Loss 0.304353 Objective Loss 0.304353 LR 0.001000 Time 0.021313 -2022-12-06 10:33:24,018 - Epoch: [25][ 350/ 1200] Overall Loss 0.304810 Objective Loss 0.304810 LR 0.001000 Time 0.021250 -2022-12-06 10:33:24,210 - Epoch: [25][ 360/ 1200] Overall Loss 0.304390 Objective Loss 0.304390 LR 0.001000 Time 0.021190 -2022-12-06 10:33:24,401 - Epoch: [25][ 370/ 1200] Overall Loss 0.304092 Objective Loss 0.304092 LR 0.001000 Time 0.021133 -2022-12-06 10:33:24,593 - Epoch: [25][ 380/ 1200] Overall Loss 0.304556 Objective Loss 0.304556 LR 0.001000 Time 0.021081 -2022-12-06 10:33:24,785 - Epoch: [25][ 390/ 1200] Overall Loss 0.304191 Objective Loss 0.304191 LR 0.001000 Time 0.021029 -2022-12-06 10:33:24,976 - Epoch: [25][ 400/ 1200] Overall Loss 0.304164 Objective Loss 0.304164 LR 0.001000 Time 0.020981 -2022-12-06 10:33:25,168 - Epoch: [25][ 410/ 1200] Overall Loss 0.304141 Objective Loss 0.304141 LR 0.001000 Time 0.020935 -2022-12-06 10:33:25,359 - Epoch: [25][ 420/ 1200] Overall Loss 0.303420 Objective Loss 0.303420 LR 0.001000 Time 0.020891 -2022-12-06 10:33:25,551 - Epoch: [25][ 430/ 1200] Overall Loss 0.303072 Objective Loss 0.303072 LR 0.001000 Time 0.020850 -2022-12-06 10:33:25,743 - Epoch: [25][ 440/ 1200] Overall Loss 0.303429 Objective Loss 0.303429 LR 0.001000 Time 0.020811 -2022-12-06 10:33:25,934 - Epoch: [25][ 450/ 1200] Overall Loss 0.303534 Objective Loss 0.303534 LR 0.001000 Time 0.020773 -2022-12-06 10:33:26,126 - Epoch: [25][ 460/ 1200] Overall Loss 0.303532 Objective Loss 0.303532 LR 0.001000 Time 0.020738 -2022-12-06 10:33:26,318 - Epoch: [25][ 470/ 1200] Overall Loss 0.303868 Objective Loss 0.303868 LR 0.001000 Time 0.020704 -2022-12-06 10:33:26,510 - Epoch: [25][ 480/ 1200] Overall Loss 0.303956 Objective Loss 0.303956 LR 0.001000 Time 0.020670 -2022-12-06 10:33:26,702 - Epoch: [25][ 490/ 1200] Overall Loss 0.303848 Objective Loss 0.303848 LR 0.001000 Time 0.020639 -2022-12-06 10:33:26,894 - Epoch: [25][ 500/ 1200] Overall Loss 0.303922 Objective Loss 0.303922 LR 0.001000 Time 0.020609 -2022-12-06 10:33:27,085 - Epoch: [25][ 510/ 1200] Overall Loss 0.304558 Objective Loss 0.304558 LR 0.001000 Time 0.020579 -2022-12-06 10:33:27,277 - Epoch: [25][ 520/ 1200] Overall Loss 0.304756 Objective Loss 0.304756 LR 0.001000 Time 0.020551 -2022-12-06 10:33:27,468 - Epoch: [25][ 530/ 1200] Overall Loss 0.304937 Objective Loss 0.304937 LR 0.001000 Time 0.020522 -2022-12-06 10:33:27,659 - Epoch: [25][ 540/ 1200] Overall Loss 0.304359 Objective Loss 0.304359 LR 0.001000 Time 0.020496 -2022-12-06 10:33:27,851 - Epoch: [25][ 550/ 1200] Overall Loss 0.304248 Objective Loss 0.304248 LR 0.001000 Time 0.020471 -2022-12-06 10:33:28,042 - Epoch: [25][ 560/ 1200] Overall Loss 0.304298 Objective Loss 0.304298 LR 0.001000 Time 0.020446 -2022-12-06 10:33:28,234 - Epoch: [25][ 570/ 1200] Overall Loss 0.304082 Objective Loss 0.304082 LR 0.001000 Time 0.020422 -2022-12-06 10:33:28,426 - Epoch: [25][ 580/ 1200] Overall Loss 0.304366 Objective Loss 0.304366 LR 0.001000 Time 0.020400 -2022-12-06 10:33:28,617 - Epoch: [25][ 590/ 1200] Overall Loss 0.304186 Objective Loss 0.304186 LR 0.001000 Time 0.020377 -2022-12-06 10:33:28,808 - Epoch: [25][ 600/ 1200] Overall Loss 0.304768 Objective Loss 0.304768 LR 0.001000 Time 0.020355 -2022-12-06 10:33:29,000 - Epoch: [25][ 610/ 1200] Overall Loss 0.304552 Objective Loss 0.304552 LR 0.001000 Time 0.020335 -2022-12-06 10:33:29,191 - Epoch: [25][ 620/ 1200] Overall Loss 0.304798 Objective Loss 0.304798 LR 0.001000 Time 0.020315 -2022-12-06 10:33:29,383 - Epoch: [25][ 630/ 1200] Overall Loss 0.304822 Objective Loss 0.304822 LR 0.001000 Time 0.020296 -2022-12-06 10:33:29,575 - Epoch: [25][ 640/ 1200] Overall Loss 0.305076 Objective Loss 0.305076 LR 0.001000 Time 0.020278 -2022-12-06 10:33:29,767 - Epoch: [25][ 650/ 1200] Overall Loss 0.305181 Objective Loss 0.305181 LR 0.001000 Time 0.020261 -2022-12-06 10:33:29,958 - Epoch: [25][ 660/ 1200] Overall Loss 0.305256 Objective Loss 0.305256 LR 0.001000 Time 0.020242 -2022-12-06 10:33:30,149 - Epoch: [25][ 670/ 1200] Overall Loss 0.305203 Objective Loss 0.305203 LR 0.001000 Time 0.020225 -2022-12-06 10:33:30,341 - Epoch: [25][ 680/ 1200] Overall Loss 0.305455 Objective Loss 0.305455 LR 0.001000 Time 0.020209 -2022-12-06 10:33:30,533 - Epoch: [25][ 690/ 1200] Overall Loss 0.305844 Objective Loss 0.305844 LR 0.001000 Time 0.020192 -2022-12-06 10:33:30,724 - Epoch: [25][ 700/ 1200] Overall Loss 0.305832 Objective Loss 0.305832 LR 0.001000 Time 0.020177 -2022-12-06 10:33:30,915 - Epoch: [25][ 710/ 1200] Overall Loss 0.306446 Objective Loss 0.306446 LR 0.001000 Time 0.020161 -2022-12-06 10:33:31,107 - Epoch: [25][ 720/ 1200] Overall Loss 0.306555 Objective Loss 0.306555 LR 0.001000 Time 0.020147 -2022-12-06 10:33:31,299 - Epoch: [25][ 730/ 1200] Overall Loss 0.306373 Objective Loss 0.306373 LR 0.001000 Time 0.020133 -2022-12-06 10:33:31,490 - Epoch: [25][ 740/ 1200] Overall Loss 0.306159 Objective Loss 0.306159 LR 0.001000 Time 0.020118 -2022-12-06 10:33:31,681 - Epoch: [25][ 750/ 1200] Overall Loss 0.305958 Objective Loss 0.305958 LR 0.001000 Time 0.020104 -2022-12-06 10:33:31,874 - Epoch: [25][ 760/ 1200] Overall Loss 0.305628 Objective Loss 0.305628 LR 0.001000 Time 0.020092 -2022-12-06 10:33:32,065 - Epoch: [25][ 770/ 1200] Overall Loss 0.305893 Objective Loss 0.305893 LR 0.001000 Time 0.020078 -2022-12-06 10:33:32,256 - Epoch: [25][ 780/ 1200] Overall Loss 0.306141 Objective Loss 0.306141 LR 0.001000 Time 0.020066 -2022-12-06 10:33:32,448 - Epoch: [25][ 790/ 1200] Overall Loss 0.306070 Objective Loss 0.306070 LR 0.001000 Time 0.020053 -2022-12-06 10:33:32,640 - Epoch: [25][ 800/ 1200] Overall Loss 0.305659 Objective Loss 0.305659 LR 0.001000 Time 0.020042 -2022-12-06 10:33:32,832 - Epoch: [25][ 810/ 1200] Overall Loss 0.305486 Objective Loss 0.305486 LR 0.001000 Time 0.020031 -2022-12-06 10:33:33,023 - Epoch: [25][ 820/ 1200] Overall Loss 0.305734 Objective Loss 0.305734 LR 0.001000 Time 0.020020 -2022-12-06 10:33:33,216 - Epoch: [25][ 830/ 1200] Overall Loss 0.305885 Objective Loss 0.305885 LR 0.001000 Time 0.020010 -2022-12-06 10:33:33,408 - Epoch: [25][ 840/ 1200] Overall Loss 0.305621 Objective Loss 0.305621 LR 0.001000 Time 0.019999 -2022-12-06 10:33:33,599 - Epoch: [25][ 850/ 1200] Overall Loss 0.305705 Objective Loss 0.305705 LR 0.001000 Time 0.019989 -2022-12-06 10:33:33,790 - Epoch: [25][ 860/ 1200] Overall Loss 0.305703 Objective Loss 0.305703 LR 0.001000 Time 0.019978 -2022-12-06 10:33:33,982 - Epoch: [25][ 870/ 1200] Overall Loss 0.305688 Objective Loss 0.305688 LR 0.001000 Time 0.019967 -2022-12-06 10:33:34,173 - Epoch: [25][ 880/ 1200] Overall Loss 0.306011 Objective Loss 0.306011 LR 0.001000 Time 0.019957 -2022-12-06 10:33:34,364 - Epoch: [25][ 890/ 1200] Overall Loss 0.306350 Objective Loss 0.306350 LR 0.001000 Time 0.019947 -2022-12-06 10:33:34,556 - Epoch: [25][ 900/ 1200] Overall Loss 0.306162 Objective Loss 0.306162 LR 0.001000 Time 0.019938 -2022-12-06 10:33:34,747 - Epoch: [25][ 910/ 1200] Overall Loss 0.306506 Objective Loss 0.306506 LR 0.001000 Time 0.019928 -2022-12-06 10:33:34,938 - Epoch: [25][ 920/ 1200] Overall Loss 0.306453 Objective Loss 0.306453 LR 0.001000 Time 0.019919 -2022-12-06 10:33:35,129 - Epoch: [25][ 930/ 1200] Overall Loss 0.306495 Objective Loss 0.306495 LR 0.001000 Time 0.019910 -2022-12-06 10:33:35,320 - Epoch: [25][ 940/ 1200] Overall Loss 0.306734 Objective Loss 0.306734 LR 0.001000 Time 0.019901 -2022-12-06 10:33:35,512 - Epoch: [25][ 950/ 1200] Overall Loss 0.307347 Objective Loss 0.307347 LR 0.001000 Time 0.019892 -2022-12-06 10:33:35,703 - Epoch: [25][ 960/ 1200] Overall Loss 0.307502 Objective Loss 0.307502 LR 0.001000 Time 0.019883 -2022-12-06 10:33:35,894 - Epoch: [25][ 970/ 1200] Overall Loss 0.307770 Objective Loss 0.307770 LR 0.001000 Time 0.019875 -2022-12-06 10:33:36,085 - Epoch: [25][ 980/ 1200] Overall Loss 0.308018 Objective Loss 0.308018 LR 0.001000 Time 0.019867 -2022-12-06 10:33:36,277 - Epoch: [25][ 990/ 1200] Overall Loss 0.308310 Objective Loss 0.308310 LR 0.001000 Time 0.019859 -2022-12-06 10:33:36,469 - Epoch: [25][ 1000/ 1200] Overall Loss 0.308139 Objective Loss 0.308139 LR 0.001000 Time 0.019851 -2022-12-06 10:33:36,660 - Epoch: [25][ 1010/ 1200] Overall Loss 0.307890 Objective Loss 0.307890 LR 0.001000 Time 0.019844 -2022-12-06 10:33:36,852 - Epoch: [25][ 1020/ 1200] Overall Loss 0.307688 Objective Loss 0.307688 LR 0.001000 Time 0.019837 -2022-12-06 10:33:37,044 - Epoch: [25][ 1030/ 1200] Overall Loss 0.307644 Objective Loss 0.307644 LR 0.001000 Time 0.019830 -2022-12-06 10:33:37,235 - Epoch: [25][ 1040/ 1200] Overall Loss 0.307721 Objective Loss 0.307721 LR 0.001000 Time 0.019823 -2022-12-06 10:33:37,428 - Epoch: [25][ 1050/ 1200] Overall Loss 0.307635 Objective Loss 0.307635 LR 0.001000 Time 0.019817 -2022-12-06 10:33:37,619 - Epoch: [25][ 1060/ 1200] Overall Loss 0.307371 Objective Loss 0.307371 LR 0.001000 Time 0.019810 -2022-12-06 10:33:37,811 - Epoch: [25][ 1070/ 1200] Overall Loss 0.307240 Objective Loss 0.307240 LR 0.001000 Time 0.019804 -2022-12-06 10:33:38,002 - Epoch: [25][ 1080/ 1200] Overall Loss 0.307020 Objective Loss 0.307020 LR 0.001000 Time 0.019797 -2022-12-06 10:33:38,194 - Epoch: [25][ 1090/ 1200] Overall Loss 0.307128 Objective Loss 0.307128 LR 0.001000 Time 0.019790 -2022-12-06 10:33:38,386 - Epoch: [25][ 1100/ 1200] Overall Loss 0.307080 Objective Loss 0.307080 LR 0.001000 Time 0.019785 -2022-12-06 10:33:38,578 - Epoch: [25][ 1110/ 1200] Overall Loss 0.306948 Objective Loss 0.306948 LR 0.001000 Time 0.019779 -2022-12-06 10:33:38,769 - Epoch: [25][ 1120/ 1200] Overall Loss 0.306997 Objective Loss 0.306997 LR 0.001000 Time 0.019773 -2022-12-06 10:33:38,961 - Epoch: [25][ 1130/ 1200] Overall Loss 0.306777 Objective Loss 0.306777 LR 0.001000 Time 0.019767 -2022-12-06 10:33:39,153 - Epoch: [25][ 1140/ 1200] Overall Loss 0.306795 Objective Loss 0.306795 LR 0.001000 Time 0.019761 -2022-12-06 10:33:39,344 - Epoch: [25][ 1150/ 1200] Overall Loss 0.306975 Objective Loss 0.306975 LR 0.001000 Time 0.019755 -2022-12-06 10:33:39,536 - Epoch: [25][ 1160/ 1200] Overall Loss 0.306873 Objective Loss 0.306873 LR 0.001000 Time 0.019750 -2022-12-06 10:33:39,728 - Epoch: [25][ 1170/ 1200] Overall Loss 0.307017 Objective Loss 0.307017 LR 0.001000 Time 0.019744 -2022-12-06 10:33:39,919 - Epoch: [25][ 1180/ 1200] Overall Loss 0.307170 Objective Loss 0.307170 LR 0.001000 Time 0.019739 -2022-12-06 10:33:40,111 - Epoch: [25][ 1190/ 1200] Overall Loss 0.307042 Objective Loss 0.307042 LR 0.001000 Time 0.019734 -2022-12-06 10:33:40,347 - Epoch: [25][ 1200/ 1200] Overall Loss 0.307146 Objective Loss 0.307146 Top1 83.891213 Top5 97.907950 LR 0.001000 Time 0.019765 -2022-12-06 10:33:40,435 - --- validate (epoch=25)----------- -2022-12-06 10:33:40,435 - 34129 samples (256 per mini-batch) -2022-12-06 10:33:40,886 - Epoch: [25][ 10/ 134] Loss 0.270211 Top1 82.734375 Top5 97.890625 -2022-12-06 10:33:41,016 - Epoch: [25][ 20/ 134] Loss 0.295524 Top1 82.500000 Top5 97.675781 -2022-12-06 10:33:41,150 - Epoch: [25][ 30/ 134] Loss 0.306627 Top1 82.096354 Top5 97.513021 -2022-12-06 10:33:41,279 - Epoch: [25][ 40/ 134] Loss 0.313368 Top1 82.109375 Top5 97.539062 -2022-12-06 10:33:41,411 - Epoch: [25][ 50/ 134] Loss 0.307326 Top1 82.507812 Top5 97.515625 -2022-12-06 10:33:41,543 - Epoch: [25][ 60/ 134] Loss 0.308423 Top1 82.636719 Top5 97.506510 -2022-12-06 10:33:41,674 - Epoch: [25][ 70/ 134] Loss 0.307502 Top1 82.505580 Top5 97.583705 -2022-12-06 10:33:41,806 - Epoch: [25][ 80/ 134] Loss 0.311372 Top1 82.441406 Top5 97.529297 -2022-12-06 10:33:41,937 - Epoch: [25][ 90/ 134] Loss 0.314028 Top1 82.447917 Top5 97.556424 -2022-12-06 10:33:42,069 - Epoch: [25][ 100/ 134] Loss 0.312742 Top1 82.507812 Top5 97.562500 -2022-12-06 10:33:42,200 - Epoch: [25][ 110/ 134] Loss 0.311148 Top1 82.656250 Top5 97.556818 -2022-12-06 10:33:42,328 - Epoch: [25][ 120/ 134] Loss 0.313599 Top1 82.633464 Top5 97.522786 -2022-12-06 10:33:42,461 - Epoch: [25][ 130/ 134] Loss 0.314320 Top1 82.665264 Top5 97.545072 -2022-12-06 10:33:42,498 - Epoch: [25][ 134/ 134] Loss 0.313731 Top1 82.624747 Top5 97.559260 -2022-12-06 10:33:42,585 - ==> Top1: 82.625 Top5: 97.559 Loss: 0.314 - -2022-12-06 10:33:42,586 - ==> Confusion: -[[ 869 4 3 1 10 5 2 2 10 69 1 1 2 3 6 2 1 1 1 0 3] - [ 0 908 0 2 14 34 6 16 2 1 11 4 2 5 2 1 4 1 7 2 5] - [ 2 5 1003 7 6 2 29 13 0 2 3 5 1 1 2 2 0 3 5 3 9] - [ 1 4 35 922 0 4 0 1 1 0 12 0 5 3 18 0 1 1 8 0 4] - [ 9 5 1 0 958 4 0 3 0 7 2 5 0 2 7 5 4 3 0 2 3] - [ 1 13 1 1 10 972 4 13 2 1 1 17 2 13 6 0 2 1 3 4 2] - [ 0 4 10 1 1 2 1064 6 0 0 3 0 0 3 0 10 2 1 1 8 2] - [ 0 9 8 2 2 41 3 930 0 1 4 8 2 0 0 3 0 1 20 16 4] - [ 5 7 0 3 0 7 0 0 921 58 14 4 4 25 10 0 1 1 3 1 0] - [ 59 0 3 0 9 9 0 1 17 871 2 1 1 19 3 0 0 0 1 0 5] - [ 0 4 8 6 0 2 1 6 3 1 958 3 3 12 3 0 0 1 4 3 1] - [ 2 5 1 0 0 13 1 6 0 0 2 963 25 13 1 3 3 6 0 6 1] - [ 1 2 0 8 2 3 2 0 0 0 0 31 883 4 1 14 1 11 0 3 3] - [ 0 3 0 0 1 12 1 3 2 11 12 4 1 961 1 4 2 0 0 1 4] - [ 5 2 1 17 6 3 0 1 17 4 4 0 4 9 1042 0 0 0 5 2 8] - [ 0 4 4 1 4 4 3 0 0 0 1 9 5 5 0 976 8 8 0 4 7] - [ 1 5 2 1 7 4 3 1 0 0 1 4 1 2 1 11 1013 3 1 5 6] - [ 1 4 2 4 0 1 2 0 1 0 2 13 21 4 3 13 2 958 2 2 1] - [ 1 6 7 13 1 3 0 23 1 0 19 6 4 5 10 0 0 2 900 1 6] - [ 0 5 2 0 0 8 9 7 1 0 0 17 7 8 1 2 6 3 0 995 9] - [ 130 273 266 170 210 287 95 145 57 120 283 136 456 425 161 165 179 77 192 268 9131]] - -2022-12-06 10:33:43,153 - ==> Best [Top1: 83.756 Top5: 97.858 Sparsity:0.00 Params: 5376 on epoch: 23] -2022-12-06 10:33:43,153 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:33:43,159 - - -2022-12-06 10:33:43,159 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:33:44,173 - Epoch: [26][ 10/ 1200] Overall Loss 0.292935 Objective Loss 0.292935 LR 0.001000 Time 0.101307 -2022-12-06 10:33:44,369 - Epoch: [26][ 20/ 1200] Overall Loss 0.302326 Objective Loss 0.302326 LR 0.001000 Time 0.060430 -2022-12-06 10:33:44,560 - Epoch: [26][ 30/ 1200] Overall Loss 0.308679 Objective Loss 0.308679 LR 0.001000 Time 0.046636 -2022-12-06 10:33:44,750 - Epoch: [26][ 40/ 1200] Overall Loss 0.297000 Objective Loss 0.297000 LR 0.001000 Time 0.039715 -2022-12-06 10:33:44,941 - Epoch: [26][ 50/ 1200] Overall Loss 0.293270 Objective Loss 0.293270 LR 0.001000 Time 0.035572 -2022-12-06 10:33:45,131 - Epoch: [26][ 60/ 1200] Overall Loss 0.293157 Objective Loss 0.293157 LR 0.001000 Time 0.032803 -2022-12-06 10:33:45,322 - Epoch: [26][ 70/ 1200] Overall Loss 0.296172 Objective Loss 0.296172 LR 0.001000 Time 0.030836 -2022-12-06 10:33:45,512 - Epoch: [26][ 80/ 1200] Overall Loss 0.293034 Objective Loss 0.293034 LR 0.001000 Time 0.029355 -2022-12-06 10:33:45,702 - Epoch: [26][ 90/ 1200] Overall Loss 0.294442 Objective Loss 0.294442 LR 0.001000 Time 0.028198 -2022-12-06 10:33:45,892 - Epoch: [26][ 100/ 1200] Overall Loss 0.299319 Objective Loss 0.299319 LR 0.001000 Time 0.027272 -2022-12-06 10:33:46,083 - Epoch: [26][ 110/ 1200] Overall Loss 0.294641 Objective Loss 0.294641 LR 0.001000 Time 0.026519 -2022-12-06 10:33:46,272 - Epoch: [26][ 120/ 1200] Overall Loss 0.292946 Objective Loss 0.292946 LR 0.001000 Time 0.025887 -2022-12-06 10:33:46,463 - Epoch: [26][ 130/ 1200] Overall Loss 0.294878 Objective Loss 0.294878 LR 0.001000 Time 0.025354 -2022-12-06 10:33:46,653 - Epoch: [26][ 140/ 1200] Overall Loss 0.294472 Objective Loss 0.294472 LR 0.001000 Time 0.024896 -2022-12-06 10:33:46,842 - Epoch: [26][ 150/ 1200] Overall Loss 0.292485 Objective Loss 0.292485 LR 0.001000 Time 0.024498 -2022-12-06 10:33:47,033 - Epoch: [26][ 160/ 1200] Overall Loss 0.290408 Objective Loss 0.290408 LR 0.001000 Time 0.024152 -2022-12-06 10:33:47,223 - Epoch: [26][ 170/ 1200] Overall Loss 0.288946 Objective Loss 0.288946 LR 0.001000 Time 0.023847 -2022-12-06 10:33:47,413 - Epoch: [26][ 180/ 1200] Overall Loss 0.288765 Objective Loss 0.288765 LR 0.001000 Time 0.023576 -2022-12-06 10:33:47,603 - Epoch: [26][ 190/ 1200] Overall Loss 0.290004 Objective Loss 0.290004 LR 0.001000 Time 0.023330 -2022-12-06 10:33:47,793 - Epoch: [26][ 200/ 1200] Overall Loss 0.289507 Objective Loss 0.289507 LR 0.001000 Time 0.023112 -2022-12-06 10:33:47,982 - Epoch: [26][ 210/ 1200] Overall Loss 0.289811 Objective Loss 0.289811 LR 0.001000 Time 0.022912 -2022-12-06 10:33:48,173 - Epoch: [26][ 220/ 1200] Overall Loss 0.289861 Objective Loss 0.289861 LR 0.001000 Time 0.022733 -2022-12-06 10:33:48,363 - Epoch: [26][ 230/ 1200] Overall Loss 0.289337 Objective Loss 0.289337 LR 0.001000 Time 0.022572 -2022-12-06 10:33:48,553 - Epoch: [26][ 240/ 1200] Overall Loss 0.287570 Objective Loss 0.287570 LR 0.001000 Time 0.022419 -2022-12-06 10:33:48,743 - Epoch: [26][ 250/ 1200] Overall Loss 0.287463 Objective Loss 0.287463 LR 0.001000 Time 0.022279 -2022-12-06 10:33:48,932 - Epoch: [26][ 260/ 1200] Overall Loss 0.288141 Objective Loss 0.288141 LR 0.001000 Time 0.022149 -2022-12-06 10:33:49,122 - Epoch: [26][ 270/ 1200] Overall Loss 0.289334 Objective Loss 0.289334 LR 0.001000 Time 0.022031 -2022-12-06 10:33:49,313 - Epoch: [26][ 280/ 1200] Overall Loss 0.290042 Objective Loss 0.290042 LR 0.001000 Time 0.021923 -2022-12-06 10:33:49,503 - Epoch: [26][ 290/ 1200] Overall Loss 0.290083 Objective Loss 0.290083 LR 0.001000 Time 0.021821 -2022-12-06 10:33:49,693 - Epoch: [26][ 300/ 1200] Overall Loss 0.290612 Objective Loss 0.290612 LR 0.001000 Time 0.021723 -2022-12-06 10:33:49,882 - Epoch: [26][ 310/ 1200] Overall Loss 0.290878 Objective Loss 0.290878 LR 0.001000 Time 0.021633 -2022-12-06 10:33:50,073 - Epoch: [26][ 320/ 1200] Overall Loss 0.289920 Objective Loss 0.289920 LR 0.001000 Time 0.021551 -2022-12-06 10:33:50,263 - Epoch: [26][ 330/ 1200] Overall Loss 0.290075 Objective Loss 0.290075 LR 0.001000 Time 0.021471 -2022-12-06 10:33:50,453 - Epoch: [26][ 340/ 1200] Overall Loss 0.289862 Objective Loss 0.289862 LR 0.001000 Time 0.021396 -2022-12-06 10:33:50,643 - Epoch: [26][ 350/ 1200] Overall Loss 0.289633 Objective Loss 0.289633 LR 0.001000 Time 0.021327 -2022-12-06 10:33:50,833 - Epoch: [26][ 360/ 1200] Overall Loss 0.290242 Objective Loss 0.290242 LR 0.001000 Time 0.021261 -2022-12-06 10:33:51,023 - Epoch: [26][ 370/ 1200] Overall Loss 0.291354 Objective Loss 0.291354 LR 0.001000 Time 0.021197 -2022-12-06 10:33:51,212 - Epoch: [26][ 380/ 1200] Overall Loss 0.291424 Objective Loss 0.291424 LR 0.001000 Time 0.021138 -2022-12-06 10:33:51,403 - Epoch: [26][ 390/ 1200] Overall Loss 0.291860 Objective Loss 0.291860 LR 0.001000 Time 0.021083 -2022-12-06 10:33:51,593 - Epoch: [26][ 400/ 1200] Overall Loss 0.291231 Objective Loss 0.291231 LR 0.001000 Time 0.021029 -2022-12-06 10:33:51,783 - Epoch: [26][ 410/ 1200] Overall Loss 0.290951 Objective Loss 0.290951 LR 0.001000 Time 0.020978 -2022-12-06 10:33:51,973 - Epoch: [26][ 420/ 1200] Overall Loss 0.291102 Objective Loss 0.291102 LR 0.001000 Time 0.020931 -2022-12-06 10:33:52,163 - Epoch: [26][ 430/ 1200] Overall Loss 0.290797 Objective Loss 0.290797 LR 0.001000 Time 0.020884 -2022-12-06 10:33:52,353 - Epoch: [26][ 440/ 1200] Overall Loss 0.291679 Objective Loss 0.291679 LR 0.001000 Time 0.020840 -2022-12-06 10:33:52,543 - Epoch: [26][ 450/ 1200] Overall Loss 0.292012 Objective Loss 0.292012 LR 0.001000 Time 0.020797 -2022-12-06 10:33:52,733 - Epoch: [26][ 460/ 1200] Overall Loss 0.291731 Objective Loss 0.291731 LR 0.001000 Time 0.020758 -2022-12-06 10:33:52,922 - Epoch: [26][ 470/ 1200] Overall Loss 0.292578 Objective Loss 0.292578 LR 0.001000 Time 0.020718 -2022-12-06 10:33:53,112 - Epoch: [26][ 480/ 1200] Overall Loss 0.293722 Objective Loss 0.293722 LR 0.001000 Time 0.020681 -2022-12-06 10:33:53,303 - Epoch: [26][ 490/ 1200] Overall Loss 0.293673 Objective Loss 0.293673 LR 0.001000 Time 0.020646 -2022-12-06 10:33:53,493 - Epoch: [26][ 500/ 1200] Overall Loss 0.293929 Objective Loss 0.293929 LR 0.001000 Time 0.020614 -2022-12-06 10:33:53,683 - Epoch: [26][ 510/ 1200] Overall Loss 0.294077 Objective Loss 0.294077 LR 0.001000 Time 0.020580 -2022-12-06 10:33:53,873 - Epoch: [26][ 520/ 1200] Overall Loss 0.294645 Objective Loss 0.294645 LR 0.001000 Time 0.020549 -2022-12-06 10:33:54,064 - Epoch: [26][ 530/ 1200] Overall Loss 0.295057 Objective Loss 0.295057 LR 0.001000 Time 0.020520 -2022-12-06 10:33:54,254 - Epoch: [26][ 540/ 1200] Overall Loss 0.294798 Objective Loss 0.294798 LR 0.001000 Time 0.020492 -2022-12-06 10:33:54,445 - Epoch: [26][ 550/ 1200] Overall Loss 0.294750 Objective Loss 0.294750 LR 0.001000 Time 0.020465 -2022-12-06 10:33:54,635 - Epoch: [26][ 560/ 1200] Overall Loss 0.294605 Objective Loss 0.294605 LR 0.001000 Time 0.020439 -2022-12-06 10:33:54,826 - Epoch: [26][ 570/ 1200] Overall Loss 0.294459 Objective Loss 0.294459 LR 0.001000 Time 0.020414 -2022-12-06 10:33:55,016 - Epoch: [26][ 580/ 1200] Overall Loss 0.294075 Objective Loss 0.294075 LR 0.001000 Time 0.020389 -2022-12-06 10:33:55,206 - Epoch: [26][ 590/ 1200] Overall Loss 0.294153 Objective Loss 0.294153 LR 0.001000 Time 0.020364 -2022-12-06 10:33:55,397 - Epoch: [26][ 600/ 1200] Overall Loss 0.294831 Objective Loss 0.294831 LR 0.001000 Time 0.020341 -2022-12-06 10:33:55,587 - Epoch: [26][ 610/ 1200] Overall Loss 0.295222 Objective Loss 0.295222 LR 0.001000 Time 0.020319 -2022-12-06 10:33:55,777 - Epoch: [26][ 620/ 1200] Overall Loss 0.294733 Objective Loss 0.294733 LR 0.001000 Time 0.020297 -2022-12-06 10:33:55,967 - Epoch: [26][ 630/ 1200] Overall Loss 0.294420 Objective Loss 0.294420 LR 0.001000 Time 0.020275 -2022-12-06 10:33:56,157 - Epoch: [26][ 640/ 1200] Overall Loss 0.294284 Objective Loss 0.294284 LR 0.001000 Time 0.020255 -2022-12-06 10:33:56,347 - Epoch: [26][ 650/ 1200] Overall Loss 0.294199 Objective Loss 0.294199 LR 0.001000 Time 0.020235 -2022-12-06 10:33:56,538 - Epoch: [26][ 660/ 1200] Overall Loss 0.294518 Objective Loss 0.294518 LR 0.001000 Time 0.020216 -2022-12-06 10:33:56,730 - Epoch: [26][ 670/ 1200] Overall Loss 0.295287 Objective Loss 0.295287 LR 0.001000 Time 0.020201 -2022-12-06 10:33:56,922 - Epoch: [26][ 680/ 1200] Overall Loss 0.295638 Objective Loss 0.295638 LR 0.001000 Time 0.020185 -2022-12-06 10:33:57,114 - Epoch: [26][ 690/ 1200] Overall Loss 0.295601 Objective Loss 0.295601 LR 0.001000 Time 0.020170 -2022-12-06 10:33:57,307 - Epoch: [26][ 700/ 1200] Overall Loss 0.295609 Objective Loss 0.295609 LR 0.001000 Time 0.020157 -2022-12-06 10:33:57,499 - Epoch: [26][ 710/ 1200] Overall Loss 0.295633 Objective Loss 0.295633 LR 0.001000 Time 0.020143 -2022-12-06 10:33:57,692 - Epoch: [26][ 720/ 1200] Overall Loss 0.295983 Objective Loss 0.295983 LR 0.001000 Time 0.020130 -2022-12-06 10:33:57,884 - Epoch: [26][ 730/ 1200] Overall Loss 0.296357 Objective Loss 0.296357 LR 0.001000 Time 0.020117 -2022-12-06 10:33:58,077 - Epoch: [26][ 740/ 1200] Overall Loss 0.296705 Objective Loss 0.296705 LR 0.001000 Time 0.020105 -2022-12-06 10:33:58,270 - Epoch: [26][ 750/ 1200] Overall Loss 0.296354 Objective Loss 0.296354 LR 0.001000 Time 0.020093 -2022-12-06 10:33:58,463 - Epoch: [26][ 760/ 1200] Overall Loss 0.296474 Objective Loss 0.296474 LR 0.001000 Time 0.020082 -2022-12-06 10:33:58,655 - Epoch: [26][ 770/ 1200] Overall Loss 0.296494 Objective Loss 0.296494 LR 0.001000 Time 0.020070 -2022-12-06 10:33:58,848 - Epoch: [26][ 780/ 1200] Overall Loss 0.296267 Objective Loss 0.296267 LR 0.001000 Time 0.020059 -2022-12-06 10:33:59,040 - Epoch: [26][ 790/ 1200] Overall Loss 0.295884 Objective Loss 0.295884 LR 0.001000 Time 0.020048 -2022-12-06 10:33:59,233 - Epoch: [26][ 800/ 1200] Overall Loss 0.295995 Objective Loss 0.295995 LR 0.001000 Time 0.020038 -2022-12-06 10:33:59,426 - Epoch: [26][ 810/ 1200] Overall Loss 0.296163 Objective Loss 0.296163 LR 0.001000 Time 0.020028 -2022-12-06 10:33:59,619 - Epoch: [26][ 820/ 1200] Overall Loss 0.296284 Objective Loss 0.296284 LR 0.001000 Time 0.020018 -2022-12-06 10:33:59,811 - Epoch: [26][ 830/ 1200] Overall Loss 0.296390 Objective Loss 0.296390 LR 0.001000 Time 0.020008 -2022-12-06 10:34:00,003 - Epoch: [26][ 840/ 1200] Overall Loss 0.296849 Objective Loss 0.296849 LR 0.001000 Time 0.019998 -2022-12-06 10:34:00,196 - Epoch: [26][ 850/ 1200] Overall Loss 0.297039 Objective Loss 0.297039 LR 0.001000 Time 0.019988 -2022-12-06 10:34:00,388 - Epoch: [26][ 860/ 1200] Overall Loss 0.296885 Objective Loss 0.296885 LR 0.001000 Time 0.019979 -2022-12-06 10:34:00,580 - Epoch: [26][ 870/ 1200] Overall Loss 0.297075 Objective Loss 0.297075 LR 0.001000 Time 0.019970 -2022-12-06 10:34:00,773 - Epoch: [26][ 880/ 1200] Overall Loss 0.297224 Objective Loss 0.297224 LR 0.001000 Time 0.019961 -2022-12-06 10:34:00,965 - Epoch: [26][ 890/ 1200] Overall Loss 0.297275 Objective Loss 0.297275 LR 0.001000 Time 0.019952 -2022-12-06 10:34:01,157 - Epoch: [26][ 900/ 1200] Overall Loss 0.297443 Objective Loss 0.297443 LR 0.001000 Time 0.019944 -2022-12-06 10:34:01,350 - Epoch: [26][ 910/ 1200] Overall Loss 0.297676 Objective Loss 0.297676 LR 0.001000 Time 0.019935 -2022-12-06 10:34:01,542 - Epoch: [26][ 920/ 1200] Overall Loss 0.297382 Objective Loss 0.297382 LR 0.001000 Time 0.019927 -2022-12-06 10:34:01,735 - Epoch: [26][ 930/ 1200] Overall Loss 0.297805 Objective Loss 0.297805 LR 0.001000 Time 0.019919 -2022-12-06 10:34:01,927 - Epoch: [26][ 940/ 1200] Overall Loss 0.298237 Objective Loss 0.298237 LR 0.001000 Time 0.019912 -2022-12-06 10:34:02,120 - Epoch: [26][ 950/ 1200] Overall Loss 0.298443 Objective Loss 0.298443 LR 0.001000 Time 0.019904 -2022-12-06 10:34:02,312 - Epoch: [26][ 960/ 1200] Overall Loss 0.298628 Objective Loss 0.298628 LR 0.001000 Time 0.019896 -2022-12-06 10:34:02,504 - Epoch: [26][ 970/ 1200] Overall Loss 0.298816 Objective Loss 0.298816 LR 0.001000 Time 0.019889 -2022-12-06 10:34:02,697 - Epoch: [26][ 980/ 1200] Overall Loss 0.298551 Objective Loss 0.298551 LR 0.001000 Time 0.019882 -2022-12-06 10:34:02,889 - Epoch: [26][ 990/ 1200] Overall Loss 0.298609 Objective Loss 0.298609 LR 0.001000 Time 0.019875 -2022-12-06 10:34:03,082 - Epoch: [26][ 1000/ 1200] Overall Loss 0.298843 Objective Loss 0.298843 LR 0.001000 Time 0.019868 -2022-12-06 10:34:03,274 - Epoch: [26][ 1010/ 1200] Overall Loss 0.298680 Objective Loss 0.298680 LR 0.001000 Time 0.019862 -2022-12-06 10:34:03,466 - Epoch: [26][ 1020/ 1200] Overall Loss 0.298815 Objective Loss 0.298815 LR 0.001000 Time 0.019855 -2022-12-06 10:34:03,658 - Epoch: [26][ 1030/ 1200] Overall Loss 0.298614 Objective Loss 0.298614 LR 0.001000 Time 0.019848 -2022-12-06 10:34:03,851 - Epoch: [26][ 1040/ 1200] Overall Loss 0.298867 Objective Loss 0.298867 LR 0.001000 Time 0.019842 -2022-12-06 10:34:04,043 - Epoch: [26][ 1050/ 1200] Overall Loss 0.298790 Objective Loss 0.298790 LR 0.001000 Time 0.019835 -2022-12-06 10:34:04,236 - Epoch: [26][ 1060/ 1200] Overall Loss 0.298735 Objective Loss 0.298735 LR 0.001000 Time 0.019829 -2022-12-06 10:34:04,428 - Epoch: [26][ 1070/ 1200] Overall Loss 0.298664 Objective Loss 0.298664 LR 0.001000 Time 0.019823 -2022-12-06 10:34:04,621 - Epoch: [26][ 1080/ 1200] Overall Loss 0.298673 Objective Loss 0.298673 LR 0.001000 Time 0.019818 -2022-12-06 10:34:04,814 - Epoch: [26][ 1090/ 1200] Overall Loss 0.298709 Objective Loss 0.298709 LR 0.001000 Time 0.019813 -2022-12-06 10:34:05,007 - Epoch: [26][ 1100/ 1200] Overall Loss 0.298840 Objective Loss 0.298840 LR 0.001000 Time 0.019807 -2022-12-06 10:34:05,199 - Epoch: [26][ 1110/ 1200] Overall Loss 0.298663 Objective Loss 0.298663 LR 0.001000 Time 0.019802 -2022-12-06 10:34:05,392 - Epoch: [26][ 1120/ 1200] Overall Loss 0.298770 Objective Loss 0.298770 LR 0.001000 Time 0.019796 -2022-12-06 10:34:05,584 - Epoch: [26][ 1130/ 1200] Overall Loss 0.298974 Objective Loss 0.298974 LR 0.001000 Time 0.019791 -2022-12-06 10:34:05,777 - Epoch: [26][ 1140/ 1200] Overall Loss 0.298900 Objective Loss 0.298900 LR 0.001000 Time 0.019786 -2022-12-06 10:34:05,970 - Epoch: [26][ 1150/ 1200] Overall Loss 0.298888 Objective Loss 0.298888 LR 0.001000 Time 0.019781 -2022-12-06 10:34:06,162 - Epoch: [26][ 1160/ 1200] Overall Loss 0.298939 Objective Loss 0.298939 LR 0.001000 Time 0.019776 -2022-12-06 10:34:06,354 - Epoch: [26][ 1170/ 1200] Overall Loss 0.299087 Objective Loss 0.299087 LR 0.001000 Time 0.019771 -2022-12-06 10:34:06,547 - Epoch: [26][ 1180/ 1200] Overall Loss 0.299087 Objective Loss 0.299087 LR 0.001000 Time 0.019766 -2022-12-06 10:34:06,739 - Epoch: [26][ 1190/ 1200] Overall Loss 0.299130 Objective Loss 0.299130 LR 0.001000 Time 0.019761 -2022-12-06 10:34:06,963 - Epoch: [26][ 1200/ 1200] Overall Loss 0.299061 Objective Loss 0.299061 Top1 84.309623 Top5 97.907950 LR 0.001000 Time 0.019782 -2022-12-06 10:34:07,051 - --- validate (epoch=26)----------- -2022-12-06 10:34:07,051 - 34129 samples (256 per mini-batch) -2022-12-06 10:34:07,492 - Epoch: [26][ 10/ 134] Loss 0.328490 Top1 82.890625 Top5 97.500000 -2022-12-06 10:34:07,623 - Epoch: [26][ 20/ 134] Loss 0.320333 Top1 83.125000 Top5 97.460938 -2022-12-06 10:34:07,751 - Epoch: [26][ 30/ 134] Loss 0.309775 Top1 83.372396 Top5 97.513021 -2022-12-06 10:34:07,881 - Epoch: [26][ 40/ 134] Loss 0.308703 Top1 83.476562 Top5 97.558594 -2022-12-06 10:34:08,009 - Epoch: [26][ 50/ 134] Loss 0.311813 Top1 83.421875 Top5 97.515625 -2022-12-06 10:34:08,139 - Epoch: [26][ 60/ 134] Loss 0.316472 Top1 83.287760 Top5 97.441406 -2022-12-06 10:34:08,270 - Epoch: [26][ 70/ 134] Loss 0.315874 Top1 83.203125 Top5 97.483259 -2022-12-06 10:34:08,401 - Epoch: [26][ 80/ 134] Loss 0.314488 Top1 83.154297 Top5 97.500000 -2022-12-06 10:34:08,531 - Epoch: [26][ 90/ 134] Loss 0.316828 Top1 83.042535 Top5 97.495660 -2022-12-06 10:34:08,663 - Epoch: [26][ 100/ 134] Loss 0.316443 Top1 83.066406 Top5 97.546875 -2022-12-06 10:34:08,791 - Epoch: [26][ 110/ 134] Loss 0.318287 Top1 83.036222 Top5 97.546165 -2022-12-06 10:34:08,923 - Epoch: [26][ 120/ 134] Loss 0.320718 Top1 82.903646 Top5 97.509766 -2022-12-06 10:34:09,057 - Epoch: [26][ 130/ 134] Loss 0.321180 Top1 82.842548 Top5 97.484976 -2022-12-06 10:34:09,096 - Epoch: [26][ 134/ 134] Loss 0.323030 Top1 82.815201 Top5 97.486009 -2022-12-06 10:34:09,191 - ==> Top1: 82.815 Top5: 97.486 Loss: 0.323 - -2022-12-06 10:34:09,192 - ==> Confusion: -[[ 892 3 0 2 4 3 1 1 6 67 0 2 2 1 3 1 1 3 0 2 2] - [ 1 911 1 1 15 31 5 15 5 1 3 3 3 4 5 1 7 2 5 4 4] - [ 12 3 961 20 3 3 39 20 0 4 5 2 2 2 2 7 1 1 2 4 10] - [ 2 2 12 930 0 6 0 1 4 0 13 0 5 3 27 1 1 4 6 0 3] - [ 10 3 2 0 953 7 1 1 0 11 2 1 0 1 11 6 4 4 0 1 2] - [ 2 18 0 2 8 962 6 21 4 3 0 8 6 10 3 2 1 1 1 5 6] - [ 0 3 11 3 1 4 1065 5 0 1 3 0 1 1 0 8 1 1 0 9 1] - [ 0 12 5 1 1 32 5 942 0 2 1 5 2 0 0 2 0 1 23 16 4] - [ 4 0 0 1 1 5 0 0 967 55 9 2 3 4 8 0 1 0 0 1 3] - [ 58 0 1 1 5 4 0 3 24 889 1 1 0 4 2 3 0 1 0 1 3] - [ 0 6 4 11 1 2 7 5 10 4 929 2 3 9 10 0 0 1 5 2 8] - [ 3 6 2 0 0 12 4 4 1 3 0 921 49 5 1 8 3 5 0 22 2] - [ 1 2 1 2 1 2 2 1 0 0 2 16 889 2 2 12 2 13 1 6 12] - [ 2 2 0 0 1 14 0 2 19 24 7 7 8 914 2 4 3 0 0 4 10] - [ 11 4 1 4 6 2 0 2 29 9 1 1 5 2 1040 1 2 2 1 1 6] - [ 1 4 3 0 2 1 5 1 0 0 1 7 8 1 0 985 10 8 0 3 3] - [ 5 7 1 1 3 2 0 1 1 1 1 3 1 0 3 17 1011 0 0 11 3] - [ 4 0 0 8 0 2 4 0 0 2 1 6 35 4 3 22 1 941 1 1 1] - [ 2 6 8 9 1 3 1 23 5 2 7 2 4 2 13 1 2 2 908 4 3] - [ 3 3 1 0 0 6 8 6 0 1 1 10 7 1 0 3 3 1 0 1021 5] - [ 211 304 162 120 192 225 107 179 126 172 177 90 428 289 208 184 213 74 145 389 9231]] - -2022-12-06 10:34:09,772 - ==> Best [Top1: 83.756 Top5: 97.858 Sparsity:0.00 Params: 5376 on epoch: 23] -2022-12-06 10:34:09,773 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:34:09,779 - - -2022-12-06 10:34:09,779 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:34:10,816 - Epoch: [27][ 10/ 1200] Overall Loss 0.280769 Objective Loss 0.280769 LR 0.001000 Time 0.103688 -2022-12-06 10:34:11,010 - Epoch: [27][ 20/ 1200] Overall Loss 0.281269 Objective Loss 0.281269 LR 0.001000 Time 0.061482 -2022-12-06 10:34:11,202 - Epoch: [27][ 30/ 1200] Overall Loss 0.291781 Objective Loss 0.291781 LR 0.001000 Time 0.047385 -2022-12-06 10:34:11,394 - Epoch: [27][ 40/ 1200] Overall Loss 0.293833 Objective Loss 0.293833 LR 0.001000 Time 0.040334 -2022-12-06 10:34:11,586 - Epoch: [27][ 50/ 1200] Overall Loss 0.292838 Objective Loss 0.292838 LR 0.001000 Time 0.036087 -2022-12-06 10:34:11,777 - Epoch: [27][ 60/ 1200] Overall Loss 0.291396 Objective Loss 0.291396 LR 0.001000 Time 0.033255 -2022-12-06 10:34:11,969 - Epoch: [27][ 70/ 1200] Overall Loss 0.291668 Objective Loss 0.291668 LR 0.001000 Time 0.031234 -2022-12-06 10:34:12,161 - Epoch: [27][ 80/ 1200] Overall Loss 0.291094 Objective Loss 0.291094 LR 0.001000 Time 0.029720 -2022-12-06 10:34:12,352 - Epoch: [27][ 90/ 1200] Overall Loss 0.292148 Objective Loss 0.292148 LR 0.001000 Time 0.028532 -2022-12-06 10:34:12,543 - Epoch: [27][ 100/ 1200] Overall Loss 0.290022 Objective Loss 0.290022 LR 0.001000 Time 0.027590 -2022-12-06 10:34:12,735 - Epoch: [27][ 110/ 1200] Overall Loss 0.287275 Objective Loss 0.287275 LR 0.001000 Time 0.026815 -2022-12-06 10:34:12,927 - Epoch: [27][ 120/ 1200] Overall Loss 0.286350 Objective Loss 0.286350 LR 0.001000 Time 0.026176 -2022-12-06 10:34:13,118 - Epoch: [27][ 130/ 1200] Overall Loss 0.288780 Objective Loss 0.288780 LR 0.001000 Time 0.025633 -2022-12-06 10:34:13,310 - Epoch: [27][ 140/ 1200] Overall Loss 0.291098 Objective Loss 0.291098 LR 0.001000 Time 0.025165 -2022-12-06 10:34:13,501 - Epoch: [27][ 150/ 1200] Overall Loss 0.290692 Objective Loss 0.290692 LR 0.001000 Time 0.024762 -2022-12-06 10:34:13,693 - Epoch: [27][ 160/ 1200] Overall Loss 0.289661 Objective Loss 0.289661 LR 0.001000 Time 0.024411 -2022-12-06 10:34:13,884 - Epoch: [27][ 170/ 1200] Overall Loss 0.290199 Objective Loss 0.290199 LR 0.001000 Time 0.024093 -2022-12-06 10:34:14,076 - Epoch: [27][ 180/ 1200] Overall Loss 0.289496 Objective Loss 0.289496 LR 0.001000 Time 0.023817 -2022-12-06 10:34:14,267 - Epoch: [27][ 190/ 1200] Overall Loss 0.290780 Objective Loss 0.290780 LR 0.001000 Time 0.023569 -2022-12-06 10:34:14,460 - Epoch: [27][ 200/ 1200] Overall Loss 0.290389 Objective Loss 0.290389 LR 0.001000 Time 0.023350 -2022-12-06 10:34:14,652 - Epoch: [27][ 210/ 1200] Overall Loss 0.289823 Objective Loss 0.289823 LR 0.001000 Time 0.023150 -2022-12-06 10:34:14,844 - Epoch: [27][ 220/ 1200] Overall Loss 0.290185 Objective Loss 0.290185 LR 0.001000 Time 0.022967 -2022-12-06 10:34:15,036 - Epoch: [27][ 230/ 1200] Overall Loss 0.289873 Objective Loss 0.289873 LR 0.001000 Time 0.022801 -2022-12-06 10:34:15,227 - Epoch: [27][ 240/ 1200] Overall Loss 0.289547 Objective Loss 0.289547 LR 0.001000 Time 0.022646 -2022-12-06 10:34:15,419 - Epoch: [27][ 250/ 1200] Overall Loss 0.289702 Objective Loss 0.289702 LR 0.001000 Time 0.022505 -2022-12-06 10:34:15,610 - Epoch: [27][ 260/ 1200] Overall Loss 0.290545 Objective Loss 0.290545 LR 0.001000 Time 0.022371 -2022-12-06 10:34:15,802 - Epoch: [27][ 270/ 1200] Overall Loss 0.291122 Objective Loss 0.291122 LR 0.001000 Time 0.022252 -2022-12-06 10:34:15,993 - Epoch: [27][ 280/ 1200] Overall Loss 0.291905 Objective Loss 0.291905 LR 0.001000 Time 0.022139 -2022-12-06 10:34:16,185 - Epoch: [27][ 290/ 1200] Overall Loss 0.292021 Objective Loss 0.292021 LR 0.001000 Time 0.022036 -2022-12-06 10:34:16,377 - Epoch: [27][ 300/ 1200] Overall Loss 0.292592 Objective Loss 0.292592 LR 0.001000 Time 0.021938 -2022-12-06 10:34:16,568 - Epoch: [27][ 310/ 1200] Overall Loss 0.292764 Objective Loss 0.292764 LR 0.001000 Time 0.021847 -2022-12-06 10:34:16,762 - Epoch: [27][ 320/ 1200] Overall Loss 0.293310 Objective Loss 0.293310 LR 0.001000 Time 0.021766 -2022-12-06 10:34:16,954 - Epoch: [27][ 330/ 1200] Overall Loss 0.293834 Objective Loss 0.293834 LR 0.001000 Time 0.021689 -2022-12-06 10:34:17,148 - Epoch: [27][ 340/ 1200] Overall Loss 0.293146 Objective Loss 0.293146 LR 0.001000 Time 0.021618 -2022-12-06 10:34:17,342 - Epoch: [27][ 350/ 1200] Overall Loss 0.293222 Objective Loss 0.293222 LR 0.001000 Time 0.021555 -2022-12-06 10:34:17,536 - Epoch: [27][ 360/ 1200] Overall Loss 0.293587 Objective Loss 0.293587 LR 0.001000 Time 0.021492 -2022-12-06 10:34:17,728 - Epoch: [27][ 370/ 1200] Overall Loss 0.294334 Objective Loss 0.294334 LR 0.001000 Time 0.021430 -2022-12-06 10:34:17,922 - Epoch: [27][ 380/ 1200] Overall Loss 0.294711 Objective Loss 0.294711 LR 0.001000 Time 0.021374 -2022-12-06 10:34:18,115 - Epoch: [27][ 390/ 1200] Overall Loss 0.295065 Objective Loss 0.295065 LR 0.001000 Time 0.021319 -2022-12-06 10:34:18,308 - Epoch: [27][ 400/ 1200] Overall Loss 0.295475 Objective Loss 0.295475 LR 0.001000 Time 0.021269 -2022-12-06 10:34:18,502 - Epoch: [27][ 410/ 1200] Overall Loss 0.295325 Objective Loss 0.295325 LR 0.001000 Time 0.021220 -2022-12-06 10:34:18,695 - Epoch: [27][ 420/ 1200] Overall Loss 0.295450 Objective Loss 0.295450 LR 0.001000 Time 0.021175 -2022-12-06 10:34:18,888 - Epoch: [27][ 430/ 1200] Overall Loss 0.295211 Objective Loss 0.295211 LR 0.001000 Time 0.021130 -2022-12-06 10:34:19,082 - Epoch: [27][ 440/ 1200] Overall Loss 0.295721 Objective Loss 0.295721 LR 0.001000 Time 0.021088 -2022-12-06 10:34:19,276 - Epoch: [27][ 450/ 1200] Overall Loss 0.295347 Objective Loss 0.295347 LR 0.001000 Time 0.021050 -2022-12-06 10:34:19,469 - Epoch: [27][ 460/ 1200] Overall Loss 0.295755 Objective Loss 0.295755 LR 0.001000 Time 0.021011 -2022-12-06 10:34:19,662 - Epoch: [27][ 470/ 1200] Overall Loss 0.295331 Objective Loss 0.295331 LR 0.001000 Time 0.020973 -2022-12-06 10:34:19,855 - Epoch: [27][ 480/ 1200] Overall Loss 0.295672 Objective Loss 0.295672 LR 0.001000 Time 0.020937 -2022-12-06 10:34:20,048 - Epoch: [27][ 490/ 1200] Overall Loss 0.295472 Objective Loss 0.295472 LR 0.001000 Time 0.020903 -2022-12-06 10:34:20,241 - Epoch: [27][ 500/ 1200] Overall Loss 0.295791 Objective Loss 0.295791 LR 0.001000 Time 0.020870 -2022-12-06 10:34:20,434 - Epoch: [27][ 510/ 1200] Overall Loss 0.295982 Objective Loss 0.295982 LR 0.001000 Time 0.020837 -2022-12-06 10:34:20,628 - Epoch: [27][ 520/ 1200] Overall Loss 0.295888 Objective Loss 0.295888 LR 0.001000 Time 0.020809 -2022-12-06 10:34:20,820 - Epoch: [27][ 530/ 1200] Overall Loss 0.295515 Objective Loss 0.295515 LR 0.001000 Time 0.020778 -2022-12-06 10:34:21,014 - Epoch: [27][ 540/ 1200] Overall Loss 0.295177 Objective Loss 0.295177 LR 0.001000 Time 0.020750 -2022-12-06 10:34:21,207 - Epoch: [27][ 550/ 1200] Overall Loss 0.295106 Objective Loss 0.295106 LR 0.001000 Time 0.020724 -2022-12-06 10:34:21,401 - Epoch: [27][ 560/ 1200] Overall Loss 0.294826 Objective Loss 0.294826 LR 0.001000 Time 0.020699 -2022-12-06 10:34:21,595 - Epoch: [27][ 570/ 1200] Overall Loss 0.295427 Objective Loss 0.295427 LR 0.001000 Time 0.020675 -2022-12-06 10:34:21,789 - Epoch: [27][ 580/ 1200] Overall Loss 0.295803 Objective Loss 0.295803 LR 0.001000 Time 0.020653 -2022-12-06 10:34:21,983 - Epoch: [27][ 590/ 1200] Overall Loss 0.295845 Objective Loss 0.295845 LR 0.001000 Time 0.020630 -2022-12-06 10:34:22,176 - Epoch: [27][ 600/ 1200] Overall Loss 0.296193 Objective Loss 0.296193 LR 0.001000 Time 0.020607 -2022-12-06 10:34:22,368 - Epoch: [27][ 610/ 1200] Overall Loss 0.296277 Objective Loss 0.296277 LR 0.001000 Time 0.020584 -2022-12-06 10:34:22,561 - Epoch: [27][ 620/ 1200] Overall Loss 0.296218 Objective Loss 0.296218 LR 0.001000 Time 0.020563 -2022-12-06 10:34:22,754 - Epoch: [27][ 630/ 1200] Overall Loss 0.296039 Objective Loss 0.296039 LR 0.001000 Time 0.020542 -2022-12-06 10:34:22,948 - Epoch: [27][ 640/ 1200] Overall Loss 0.296472 Objective Loss 0.296472 LR 0.001000 Time 0.020523 -2022-12-06 10:34:23,141 - Epoch: [27][ 650/ 1200] Overall Loss 0.296408 Objective Loss 0.296408 LR 0.001000 Time 0.020503 -2022-12-06 10:34:23,334 - Epoch: [27][ 660/ 1200] Overall Loss 0.296360 Objective Loss 0.296360 LR 0.001000 Time 0.020484 -2022-12-06 10:34:23,528 - Epoch: [27][ 670/ 1200] Overall Loss 0.296262 Objective Loss 0.296262 LR 0.001000 Time 0.020466 -2022-12-06 10:34:23,721 - Epoch: [27][ 680/ 1200] Overall Loss 0.296308 Objective Loss 0.296308 LR 0.001000 Time 0.020449 -2022-12-06 10:34:23,914 - Epoch: [27][ 690/ 1200] Overall Loss 0.296122 Objective Loss 0.296122 LR 0.001000 Time 0.020430 -2022-12-06 10:34:24,106 - Epoch: [27][ 700/ 1200] Overall Loss 0.296219 Objective Loss 0.296219 LR 0.001000 Time 0.020413 -2022-12-06 10:34:24,300 - Epoch: [27][ 710/ 1200] Overall Loss 0.296393 Objective Loss 0.296393 LR 0.001000 Time 0.020397 -2022-12-06 10:34:24,493 - Epoch: [27][ 720/ 1200] Overall Loss 0.296695 Objective Loss 0.296695 LR 0.001000 Time 0.020382 -2022-12-06 10:34:24,687 - Epoch: [27][ 730/ 1200] Overall Loss 0.296577 Objective Loss 0.296577 LR 0.001000 Time 0.020367 -2022-12-06 10:34:24,881 - Epoch: [27][ 740/ 1200] Overall Loss 0.296424 Objective Loss 0.296424 LR 0.001000 Time 0.020353 -2022-12-06 10:34:25,073 - Epoch: [27][ 750/ 1200] Overall Loss 0.296536 Objective Loss 0.296536 LR 0.001000 Time 0.020338 -2022-12-06 10:34:25,266 - Epoch: [27][ 760/ 1200] Overall Loss 0.296780 Objective Loss 0.296780 LR 0.001000 Time 0.020324 -2022-12-06 10:34:25,460 - Epoch: [27][ 770/ 1200] Overall Loss 0.296409 Objective Loss 0.296409 LR 0.001000 Time 0.020311 -2022-12-06 10:34:25,652 - Epoch: [27][ 780/ 1200] Overall Loss 0.296283 Objective Loss 0.296283 LR 0.001000 Time 0.020295 -2022-12-06 10:34:25,843 - Epoch: [27][ 790/ 1200] Overall Loss 0.296090 Objective Loss 0.296090 LR 0.001000 Time 0.020280 -2022-12-06 10:34:26,035 - Epoch: [27][ 800/ 1200] Overall Loss 0.296285 Objective Loss 0.296285 LR 0.001000 Time 0.020265 -2022-12-06 10:34:26,227 - Epoch: [27][ 810/ 1200] Overall Loss 0.296609 Objective Loss 0.296609 LR 0.001000 Time 0.020251 -2022-12-06 10:34:26,419 - Epoch: [27][ 820/ 1200] Overall Loss 0.296664 Objective Loss 0.296664 LR 0.001000 Time 0.020238 -2022-12-06 10:34:26,610 - Epoch: [27][ 830/ 1200] Overall Loss 0.296798 Objective Loss 0.296798 LR 0.001000 Time 0.020224 -2022-12-06 10:34:26,801 - Epoch: [27][ 840/ 1200] Overall Loss 0.296664 Objective Loss 0.296664 LR 0.001000 Time 0.020210 -2022-12-06 10:34:26,993 - Epoch: [27][ 850/ 1200] Overall Loss 0.297009 Objective Loss 0.297009 LR 0.001000 Time 0.020197 -2022-12-06 10:34:27,185 - Epoch: [27][ 860/ 1200] Overall Loss 0.296848 Objective Loss 0.296848 LR 0.001000 Time 0.020185 -2022-12-06 10:34:27,376 - Epoch: [27][ 870/ 1200] Overall Loss 0.296962 Objective Loss 0.296962 LR 0.001000 Time 0.020173 -2022-12-06 10:34:27,568 - Epoch: [27][ 880/ 1200] Overall Loss 0.296744 Objective Loss 0.296744 LR 0.001000 Time 0.020160 -2022-12-06 10:34:27,760 - Epoch: [27][ 890/ 1200] Overall Loss 0.296841 Objective Loss 0.296841 LR 0.001000 Time 0.020149 -2022-12-06 10:34:27,951 - Epoch: [27][ 900/ 1200] Overall Loss 0.296929 Objective Loss 0.296929 LR 0.001000 Time 0.020136 -2022-12-06 10:34:28,142 - Epoch: [27][ 910/ 1200] Overall Loss 0.297156 Objective Loss 0.297156 LR 0.001000 Time 0.020124 -2022-12-06 10:34:28,334 - Epoch: [27][ 920/ 1200] Overall Loss 0.296960 Objective Loss 0.296960 LR 0.001000 Time 0.020114 -2022-12-06 10:34:28,525 - Epoch: [27][ 930/ 1200] Overall Loss 0.296852 Objective Loss 0.296852 LR 0.001000 Time 0.020102 -2022-12-06 10:34:28,716 - Epoch: [27][ 940/ 1200] Overall Loss 0.296643 Objective Loss 0.296643 LR 0.001000 Time 0.020092 -2022-12-06 10:34:28,908 - Epoch: [27][ 950/ 1200] Overall Loss 0.296894 Objective Loss 0.296894 LR 0.001000 Time 0.020081 -2022-12-06 10:34:29,100 - Epoch: [27][ 960/ 1200] Overall Loss 0.296842 Objective Loss 0.296842 LR 0.001000 Time 0.020071 -2022-12-06 10:34:29,292 - Epoch: [27][ 970/ 1200] Overall Loss 0.296809 Objective Loss 0.296809 LR 0.001000 Time 0.020062 -2022-12-06 10:34:29,483 - Epoch: [27][ 980/ 1200] Overall Loss 0.296695 Objective Loss 0.296695 LR 0.001000 Time 0.020052 -2022-12-06 10:34:29,675 - Epoch: [27][ 990/ 1200] Overall Loss 0.296283 Objective Loss 0.296283 LR 0.001000 Time 0.020043 -2022-12-06 10:34:29,867 - Epoch: [27][ 1000/ 1200] Overall Loss 0.296263 Objective Loss 0.296263 LR 0.001000 Time 0.020034 -2022-12-06 10:34:30,059 - Epoch: [27][ 1010/ 1200] Overall Loss 0.296338 Objective Loss 0.296338 LR 0.001000 Time 0.020024 -2022-12-06 10:34:30,250 - Epoch: [27][ 1020/ 1200] Overall Loss 0.296208 Objective Loss 0.296208 LR 0.001000 Time 0.020015 -2022-12-06 10:34:30,441 - Epoch: [27][ 1030/ 1200] Overall Loss 0.296064 Objective Loss 0.296064 LR 0.001000 Time 0.020006 -2022-12-06 10:34:30,633 - Epoch: [27][ 1040/ 1200] Overall Loss 0.296298 Objective Loss 0.296298 LR 0.001000 Time 0.019998 -2022-12-06 10:34:30,824 - Epoch: [27][ 1050/ 1200] Overall Loss 0.296541 Objective Loss 0.296541 LR 0.001000 Time 0.019988 -2022-12-06 10:34:31,016 - Epoch: [27][ 1060/ 1200] Overall Loss 0.296739 Objective Loss 0.296739 LR 0.001000 Time 0.019980 -2022-12-06 10:34:31,207 - Epoch: [27][ 1070/ 1200] Overall Loss 0.296757 Objective Loss 0.296757 LR 0.001000 Time 0.019972 -2022-12-06 10:34:31,399 - Epoch: [27][ 1080/ 1200] Overall Loss 0.296602 Objective Loss 0.296602 LR 0.001000 Time 0.019964 -2022-12-06 10:34:31,590 - Epoch: [27][ 1090/ 1200] Overall Loss 0.297061 Objective Loss 0.297061 LR 0.001000 Time 0.019956 -2022-12-06 10:34:31,782 - Epoch: [27][ 1100/ 1200] Overall Loss 0.296978 Objective Loss 0.296978 LR 0.001000 Time 0.019948 -2022-12-06 10:34:31,973 - Epoch: [27][ 1110/ 1200] Overall Loss 0.297027 Objective Loss 0.297027 LR 0.001000 Time 0.019940 -2022-12-06 10:34:32,165 - Epoch: [27][ 1120/ 1200] Overall Loss 0.296732 Objective Loss 0.296732 LR 0.001000 Time 0.019933 -2022-12-06 10:34:32,358 - Epoch: [27][ 1130/ 1200] Overall Loss 0.296803 Objective Loss 0.296803 LR 0.001000 Time 0.019927 -2022-12-06 10:34:32,550 - Epoch: [27][ 1140/ 1200] Overall Loss 0.296837 Objective Loss 0.296837 LR 0.001000 Time 0.019920 -2022-12-06 10:34:32,741 - Epoch: [27][ 1150/ 1200] Overall Loss 0.296645 Objective Loss 0.296645 LR 0.001000 Time 0.019913 -2022-12-06 10:34:32,933 - Epoch: [27][ 1160/ 1200] Overall Loss 0.296851 Objective Loss 0.296851 LR 0.001000 Time 0.019906 -2022-12-06 10:34:33,124 - Epoch: [27][ 1170/ 1200] Overall Loss 0.297125 Objective Loss 0.297125 LR 0.001000 Time 0.019898 -2022-12-06 10:34:33,316 - Epoch: [27][ 1180/ 1200] Overall Loss 0.297255 Objective Loss 0.297255 LR 0.001000 Time 0.019892 -2022-12-06 10:34:33,506 - Epoch: [27][ 1190/ 1200] Overall Loss 0.297249 Objective Loss 0.297249 LR 0.001000 Time 0.019884 -2022-12-06 10:34:33,734 - Epoch: [27][ 1200/ 1200] Overall Loss 0.297416 Objective Loss 0.297416 Top1 84.518828 Top5 97.907950 LR 0.001000 Time 0.019908 -2022-12-06 10:34:33,822 - --- validate (epoch=27)----------- -2022-12-06 10:34:33,822 - 34129 samples (256 per mini-batch) -2022-12-06 10:34:34,267 - Epoch: [27][ 10/ 134] Loss 0.318658 Top1 82.929688 Top5 97.734375 -2022-12-06 10:34:34,397 - Epoch: [27][ 20/ 134] Loss 0.326042 Top1 82.812500 Top5 97.656250 -2022-12-06 10:34:34,527 - Epoch: [27][ 30/ 134] Loss 0.313674 Top1 82.838542 Top5 97.513021 -2022-12-06 10:34:34,660 - Epoch: [27][ 40/ 134] Loss 0.312015 Top1 83.027344 Top5 97.451172 -2022-12-06 10:34:34,791 - Epoch: [27][ 50/ 134] Loss 0.314967 Top1 82.828125 Top5 97.429688 -2022-12-06 10:34:34,923 - Epoch: [27][ 60/ 134] Loss 0.314776 Top1 82.864583 Top5 97.480469 -2022-12-06 10:34:35,052 - Epoch: [27][ 70/ 134] Loss 0.316968 Top1 82.890625 Top5 97.500000 -2022-12-06 10:34:35,184 - Epoch: [27][ 80/ 134] Loss 0.310981 Top1 83.115234 Top5 97.573242 -2022-12-06 10:34:35,316 - Epoch: [27][ 90/ 134] Loss 0.316867 Top1 83.007812 Top5 97.547743 -2022-12-06 10:34:35,448 - Epoch: [27][ 100/ 134] Loss 0.316828 Top1 82.945312 Top5 97.578125 -2022-12-06 10:34:35,582 - Epoch: [27][ 110/ 134] Loss 0.318438 Top1 82.982955 Top5 97.553267 -2022-12-06 10:34:35,714 - Epoch: [27][ 120/ 134] Loss 0.314898 Top1 83.059896 Top5 97.542318 -2022-12-06 10:34:35,847 - Epoch: [27][ 130/ 134] Loss 0.312919 Top1 83.088942 Top5 97.578125 -2022-12-06 10:34:35,886 - Epoch: [27][ 134/ 134] Loss 0.314347 Top1 83.052536 Top5 97.570981 -2022-12-06 10:34:35,973 - ==> Top1: 83.053 Top5: 97.571 Loss: 0.314 - -2022-12-06 10:34:35,974 - ==> Confusion: -[[ 895 3 4 0 3 3 1 0 6 59 1 3 2 1 2 3 1 2 2 1 4] - [ 2 917 1 2 15 19 9 19 1 0 9 2 2 1 3 3 2 2 4 5 9] - [ 6 2 1015 12 3 2 27 10 0 0 8 4 2 1 2 1 1 1 1 2 3] - [ 0 1 29 927 0 5 0 1 1 0 8 0 4 1 23 2 1 3 10 0 4] - [ 12 4 0 0 955 5 1 1 0 6 4 1 0 5 10 8 3 1 0 1 3] - [ 4 22 1 3 10 945 3 23 4 0 3 17 4 8 2 2 2 0 3 7 6] - [ 0 1 16 2 0 4 1070 2 0 0 5 1 0 0 0 5 0 1 0 8 3] - [ 0 10 14 2 1 30 6 921 1 1 4 5 0 2 0 4 0 1 27 21 4] - [ 5 4 0 1 0 1 1 0 959 54 8 2 1 6 16 0 1 1 1 2 1] - [ 62 0 5 0 4 3 1 1 17 891 2 1 0 7 2 3 0 0 0 0 2] - [ 0 5 4 8 1 0 0 1 7 3 954 3 1 11 6 0 1 1 6 4 3] - [ 3 3 3 0 0 13 3 8 2 0 1 954 26 3 0 8 2 3 0 17 2] - [ 2 2 4 6 1 3 2 0 0 0 1 44 846 1 3 18 0 16 0 11 9] - [ 0 3 1 1 2 12 0 2 11 19 16 8 5 925 2 6 1 0 0 5 4] - [ 12 2 0 7 3 0 0 1 20 7 4 1 1 4 1055 2 1 0 1 1 8] - [ 2 2 5 1 2 2 4 0 0 0 1 5 1 2 0 991 7 9 0 5 4] - [ 4 2 2 3 4 3 2 1 1 0 3 5 2 2 1 20 1002 0 1 6 8] - [ 3 1 3 1 1 1 5 1 2 2 1 15 19 1 4 22 0 950 1 2 1] - [ 2 6 8 12 0 4 2 24 2 1 15 3 2 2 17 0 0 1 899 4 4] - [ 1 2 4 0 0 8 13 6 0 0 0 13 5 3 0 4 4 2 0 1010 5] - [ 164 243 327 140 178 187 133 128 86 118 257 137 385 348 234 227 131 61 140 343 9259]] - -2022-12-06 10:34:36,542 - ==> Best [Top1: 83.756 Top5: 97.858 Sparsity:0.00 Params: 5376 on epoch: 23] -2022-12-06 10:34:36,542 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:34:36,548 - - -2022-12-06 10:34:36,548 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:34:37,476 - Epoch: [28][ 10/ 1200] Overall Loss 0.268578 Objective Loss 0.268578 LR 0.001000 Time 0.092739 -2022-12-06 10:34:37,675 - Epoch: [28][ 20/ 1200] Overall Loss 0.291606 Objective Loss 0.291606 LR 0.001000 Time 0.056315 -2022-12-06 10:34:37,871 - Epoch: [28][ 30/ 1200] Overall Loss 0.287348 Objective Loss 0.287348 LR 0.001000 Time 0.044032 -2022-12-06 10:34:38,068 - Epoch: [28][ 40/ 1200] Overall Loss 0.287690 Objective Loss 0.287690 LR 0.001000 Time 0.037944 -2022-12-06 10:34:38,263 - Epoch: [28][ 50/ 1200] Overall Loss 0.284650 Objective Loss 0.284650 LR 0.001000 Time 0.034256 -2022-12-06 10:34:38,461 - Epoch: [28][ 60/ 1200] Overall Loss 0.279493 Objective Loss 0.279493 LR 0.001000 Time 0.031833 -2022-12-06 10:34:38,657 - Epoch: [28][ 70/ 1200] Overall Loss 0.276820 Objective Loss 0.276820 LR 0.001000 Time 0.030078 -2022-12-06 10:34:38,855 - Epoch: [28][ 80/ 1200] Overall Loss 0.277197 Objective Loss 0.277197 LR 0.001000 Time 0.028777 -2022-12-06 10:34:39,050 - Epoch: [28][ 90/ 1200] Overall Loss 0.278431 Objective Loss 0.278431 LR 0.001000 Time 0.027745 -2022-12-06 10:34:39,248 - Epoch: [28][ 100/ 1200] Overall Loss 0.277802 Objective Loss 0.277802 LR 0.001000 Time 0.026942 -2022-12-06 10:34:39,443 - Epoch: [28][ 110/ 1200] Overall Loss 0.278404 Objective Loss 0.278404 LR 0.001000 Time 0.026266 -2022-12-06 10:34:39,641 - Epoch: [28][ 120/ 1200] Overall Loss 0.276116 Objective Loss 0.276116 LR 0.001000 Time 0.025723 -2022-12-06 10:34:39,837 - Epoch: [28][ 130/ 1200] Overall Loss 0.277915 Objective Loss 0.277915 LR 0.001000 Time 0.025249 -2022-12-06 10:34:40,035 - Epoch: [28][ 140/ 1200] Overall Loss 0.279329 Objective Loss 0.279329 LR 0.001000 Time 0.024853 -2022-12-06 10:34:40,231 - Epoch: [28][ 150/ 1200] Overall Loss 0.280082 Objective Loss 0.280082 LR 0.001000 Time 0.024497 -2022-12-06 10:34:40,427 - Epoch: [28][ 160/ 1200] Overall Loss 0.278635 Objective Loss 0.278635 LR 0.001000 Time 0.024190 -2022-12-06 10:34:40,623 - Epoch: [28][ 170/ 1200] Overall Loss 0.278647 Objective Loss 0.278647 LR 0.001000 Time 0.023915 -2022-12-06 10:34:40,820 - Epoch: [28][ 180/ 1200] Overall Loss 0.280384 Objective Loss 0.280384 LR 0.001000 Time 0.023680 -2022-12-06 10:34:41,015 - Epoch: [28][ 190/ 1200] Overall Loss 0.280331 Objective Loss 0.280331 LR 0.001000 Time 0.023456 -2022-12-06 10:34:41,213 - Epoch: [28][ 200/ 1200] Overall Loss 0.280802 Objective Loss 0.280802 LR 0.001000 Time 0.023271 -2022-12-06 10:34:41,408 - Epoch: [28][ 210/ 1200] Overall Loss 0.281618 Objective Loss 0.281618 LR 0.001000 Time 0.023091 -2022-12-06 10:34:41,606 - Epoch: [28][ 220/ 1200] Overall Loss 0.282344 Objective Loss 0.282344 LR 0.001000 Time 0.022939 -2022-12-06 10:34:41,802 - Epoch: [28][ 230/ 1200] Overall Loss 0.281431 Objective Loss 0.281431 LR 0.001000 Time 0.022788 -2022-12-06 10:34:41,999 - Epoch: [28][ 240/ 1200] Overall Loss 0.280506 Objective Loss 0.280506 LR 0.001000 Time 0.022657 -2022-12-06 10:34:42,195 - Epoch: [28][ 250/ 1200] Overall Loss 0.280234 Objective Loss 0.280234 LR 0.001000 Time 0.022532 -2022-12-06 10:34:42,392 - Epoch: [28][ 260/ 1200] Overall Loss 0.280334 Objective Loss 0.280334 LR 0.001000 Time 0.022422 -2022-12-06 10:34:42,588 - Epoch: [28][ 270/ 1200] Overall Loss 0.279789 Objective Loss 0.279789 LR 0.001000 Time 0.022314 -2022-12-06 10:34:42,784 - Epoch: [28][ 280/ 1200] Overall Loss 0.279305 Objective Loss 0.279305 LR 0.001000 Time 0.022218 -2022-12-06 10:34:42,980 - Epoch: [28][ 290/ 1200] Overall Loss 0.279065 Objective Loss 0.279065 LR 0.001000 Time 0.022124 -2022-12-06 10:34:43,178 - Epoch: [28][ 300/ 1200] Overall Loss 0.278164 Objective Loss 0.278164 LR 0.001000 Time 0.022044 -2022-12-06 10:34:43,373 - Epoch: [28][ 310/ 1200] Overall Loss 0.278093 Objective Loss 0.278093 LR 0.001000 Time 0.021963 -2022-12-06 10:34:43,571 - Epoch: [28][ 320/ 1200] Overall Loss 0.278429 Objective Loss 0.278429 LR 0.001000 Time 0.021893 -2022-12-06 10:34:43,767 - Epoch: [28][ 330/ 1200] Overall Loss 0.278454 Objective Loss 0.278454 LR 0.001000 Time 0.021822 -2022-12-06 10:34:43,965 - Epoch: [28][ 340/ 1200] Overall Loss 0.279531 Objective Loss 0.279531 LR 0.001000 Time 0.021760 -2022-12-06 10:34:44,161 - Epoch: [28][ 350/ 1200] Overall Loss 0.279969 Objective Loss 0.279969 LR 0.001000 Time 0.021697 -2022-12-06 10:34:44,359 - Epoch: [28][ 360/ 1200] Overall Loss 0.280373 Objective Loss 0.280373 LR 0.001000 Time 0.021642 -2022-12-06 10:34:44,554 - Epoch: [28][ 370/ 1200] Overall Loss 0.281612 Objective Loss 0.281612 LR 0.001000 Time 0.021584 -2022-12-06 10:34:44,751 - Epoch: [28][ 380/ 1200] Overall Loss 0.281595 Objective Loss 0.281595 LR 0.001000 Time 0.021533 -2022-12-06 10:34:44,948 - Epoch: [28][ 390/ 1200] Overall Loss 0.282217 Objective Loss 0.282217 LR 0.001000 Time 0.021484 -2022-12-06 10:34:45,145 - Epoch: [28][ 400/ 1200] Overall Loss 0.283293 Objective Loss 0.283293 LR 0.001000 Time 0.021439 -2022-12-06 10:34:45,342 - Epoch: [28][ 410/ 1200] Overall Loss 0.282873 Objective Loss 0.282873 LR 0.001000 Time 0.021394 -2022-12-06 10:34:45,539 - Epoch: [28][ 420/ 1200] Overall Loss 0.283051 Objective Loss 0.283051 LR 0.001000 Time 0.021353 -2022-12-06 10:34:45,736 - Epoch: [28][ 430/ 1200] Overall Loss 0.283570 Objective Loss 0.283570 LR 0.001000 Time 0.021312 -2022-12-06 10:34:45,936 - Epoch: [28][ 440/ 1200] Overall Loss 0.284166 Objective Loss 0.284166 LR 0.001000 Time 0.021282 -2022-12-06 10:34:46,134 - Epoch: [28][ 450/ 1200] Overall Loss 0.284074 Objective Loss 0.284074 LR 0.001000 Time 0.021248 -2022-12-06 10:34:46,334 - Epoch: [28][ 460/ 1200] Overall Loss 0.284499 Objective Loss 0.284499 LR 0.001000 Time 0.021220 -2022-12-06 10:34:46,532 - Epoch: [28][ 470/ 1200] Overall Loss 0.284701 Objective Loss 0.284701 LR 0.001000 Time 0.021188 -2022-12-06 10:34:46,733 - Epoch: [28][ 480/ 1200] Overall Loss 0.284979 Objective Loss 0.284979 LR 0.001000 Time 0.021163 -2022-12-06 10:34:46,930 - Epoch: [28][ 490/ 1200] Overall Loss 0.285385 Objective Loss 0.285385 LR 0.001000 Time 0.021134 -2022-12-06 10:34:47,131 - Epoch: [28][ 500/ 1200] Overall Loss 0.285827 Objective Loss 0.285827 LR 0.001000 Time 0.021112 -2022-12-06 10:34:47,329 - Epoch: [28][ 510/ 1200] Overall Loss 0.285532 Objective Loss 0.285532 LR 0.001000 Time 0.021084 -2022-12-06 10:34:47,529 - Epoch: [28][ 520/ 1200] Overall Loss 0.285300 Objective Loss 0.285300 LR 0.001000 Time 0.021063 -2022-12-06 10:34:47,726 - Epoch: [28][ 530/ 1200] Overall Loss 0.285199 Objective Loss 0.285199 LR 0.001000 Time 0.021037 -2022-12-06 10:34:47,927 - Epoch: [28][ 540/ 1200] Overall Loss 0.284825 Objective Loss 0.284825 LR 0.001000 Time 0.021017 -2022-12-06 10:34:48,124 - Epoch: [28][ 550/ 1200] Overall Loss 0.284186 Objective Loss 0.284186 LR 0.001000 Time 0.020993 -2022-12-06 10:34:48,325 - Epoch: [28][ 560/ 1200] Overall Loss 0.284549 Objective Loss 0.284549 LR 0.001000 Time 0.020976 -2022-12-06 10:34:48,523 - Epoch: [28][ 570/ 1200] Overall Loss 0.284314 Objective Loss 0.284314 LR 0.001000 Time 0.020954 -2022-12-06 10:34:48,725 - Epoch: [28][ 580/ 1200] Overall Loss 0.284405 Objective Loss 0.284405 LR 0.001000 Time 0.020940 -2022-12-06 10:34:48,923 - Epoch: [28][ 590/ 1200] Overall Loss 0.284627 Objective Loss 0.284627 LR 0.001000 Time 0.020920 -2022-12-06 10:34:49,124 - Epoch: [28][ 600/ 1200] Overall Loss 0.284547 Objective Loss 0.284547 LR 0.001000 Time 0.020906 -2022-12-06 10:34:49,323 - Epoch: [28][ 610/ 1200] Overall Loss 0.284052 Objective Loss 0.284052 LR 0.001000 Time 0.020888 -2022-12-06 10:34:49,524 - Epoch: [28][ 620/ 1200] Overall Loss 0.284164 Objective Loss 0.284164 LR 0.001000 Time 0.020875 -2022-12-06 10:34:49,723 - Epoch: [28][ 630/ 1200] Overall Loss 0.284460 Objective Loss 0.284460 LR 0.001000 Time 0.020858 -2022-12-06 10:34:49,924 - Epoch: [28][ 640/ 1200] Overall Loss 0.284610 Objective Loss 0.284610 LR 0.001000 Time 0.020846 -2022-12-06 10:34:50,122 - Epoch: [28][ 650/ 1200] Overall Loss 0.284634 Objective Loss 0.284634 LR 0.001000 Time 0.020829 -2022-12-06 10:34:50,324 - Epoch: [28][ 660/ 1200] Overall Loss 0.284212 Objective Loss 0.284212 LR 0.001000 Time 0.020819 -2022-12-06 10:34:50,523 - Epoch: [28][ 670/ 1200] Overall Loss 0.284419 Objective Loss 0.284419 LR 0.001000 Time 0.020804 -2022-12-06 10:34:50,723 - Epoch: [28][ 680/ 1200] Overall Loss 0.284657 Objective Loss 0.284657 LR 0.001000 Time 0.020791 -2022-12-06 10:34:50,921 - Epoch: [28][ 690/ 1200] Overall Loss 0.284878 Objective Loss 0.284878 LR 0.001000 Time 0.020776 -2022-12-06 10:34:51,122 - Epoch: [28][ 700/ 1200] Overall Loss 0.285112 Objective Loss 0.285112 LR 0.001000 Time 0.020765 -2022-12-06 10:34:51,319 - Epoch: [28][ 710/ 1200] Overall Loss 0.285078 Objective Loss 0.285078 LR 0.001000 Time 0.020750 -2022-12-06 10:34:51,520 - Epoch: [28][ 720/ 1200] Overall Loss 0.285352 Objective Loss 0.285352 LR 0.001000 Time 0.020740 -2022-12-06 10:34:51,719 - Epoch: [28][ 730/ 1200] Overall Loss 0.285481 Objective Loss 0.285481 LR 0.001000 Time 0.020727 -2022-12-06 10:34:51,920 - Epoch: [28][ 740/ 1200] Overall Loss 0.285580 Objective Loss 0.285580 LR 0.001000 Time 0.020719 -2022-12-06 10:34:52,118 - Epoch: [28][ 750/ 1200] Overall Loss 0.285406 Objective Loss 0.285406 LR 0.001000 Time 0.020706 -2022-12-06 10:34:52,319 - Epoch: [28][ 760/ 1200] Overall Loss 0.285620 Objective Loss 0.285620 LR 0.001000 Time 0.020696 -2022-12-06 10:34:52,517 - Epoch: [28][ 770/ 1200] Overall Loss 0.285681 Objective Loss 0.285681 LR 0.001000 Time 0.020684 -2022-12-06 10:34:52,718 - Epoch: [28][ 780/ 1200] Overall Loss 0.285897 Objective Loss 0.285897 LR 0.001000 Time 0.020676 -2022-12-06 10:34:52,917 - Epoch: [28][ 790/ 1200] Overall Loss 0.285942 Objective Loss 0.285942 LR 0.001000 Time 0.020666 -2022-12-06 10:34:53,118 - Epoch: [28][ 800/ 1200] Overall Loss 0.286344 Objective Loss 0.286344 LR 0.001000 Time 0.020658 -2022-12-06 10:34:53,316 - Epoch: [28][ 810/ 1200] Overall Loss 0.286407 Objective Loss 0.286407 LR 0.001000 Time 0.020647 -2022-12-06 10:34:53,517 - Epoch: [28][ 820/ 1200] Overall Loss 0.286765 Objective Loss 0.286765 LR 0.001000 Time 0.020640 -2022-12-06 10:34:53,716 - Epoch: [28][ 830/ 1200] Overall Loss 0.286687 Objective Loss 0.286687 LR 0.001000 Time 0.020630 -2022-12-06 10:34:53,917 - Epoch: [28][ 840/ 1200] Overall Loss 0.287409 Objective Loss 0.287409 LR 0.001000 Time 0.020623 -2022-12-06 10:34:54,115 - Epoch: [28][ 850/ 1200] Overall Loss 0.287393 Objective Loss 0.287393 LR 0.001000 Time 0.020612 -2022-12-06 10:34:54,316 - Epoch: [28][ 860/ 1200] Overall Loss 0.287531 Objective Loss 0.287531 LR 0.001000 Time 0.020606 -2022-12-06 10:34:54,514 - Epoch: [28][ 870/ 1200] Overall Loss 0.287285 Objective Loss 0.287285 LR 0.001000 Time 0.020596 -2022-12-06 10:34:54,715 - Epoch: [28][ 880/ 1200] Overall Loss 0.287411 Objective Loss 0.287411 LR 0.001000 Time 0.020590 -2022-12-06 10:34:54,913 - Epoch: [28][ 890/ 1200] Overall Loss 0.287514 Objective Loss 0.287514 LR 0.001000 Time 0.020581 -2022-12-06 10:34:55,115 - Epoch: [28][ 900/ 1200] Overall Loss 0.287641 Objective Loss 0.287641 LR 0.001000 Time 0.020576 -2022-12-06 10:34:55,314 - Epoch: [28][ 910/ 1200] Overall Loss 0.287901 Objective Loss 0.287901 LR 0.001000 Time 0.020567 -2022-12-06 10:34:55,515 - Epoch: [28][ 920/ 1200] Overall Loss 0.288014 Objective Loss 0.288014 LR 0.001000 Time 0.020562 -2022-12-06 10:34:55,713 - Epoch: [28][ 930/ 1200] Overall Loss 0.288173 Objective Loss 0.288173 LR 0.001000 Time 0.020554 -2022-12-06 10:34:55,915 - Epoch: [28][ 940/ 1200] Overall Loss 0.288760 Objective Loss 0.288760 LR 0.001000 Time 0.020549 -2022-12-06 10:34:56,113 - Epoch: [28][ 950/ 1200] Overall Loss 0.288591 Objective Loss 0.288591 LR 0.001000 Time 0.020540 -2022-12-06 10:34:56,314 - Epoch: [28][ 960/ 1200] Overall Loss 0.288851 Objective Loss 0.288851 LR 0.001000 Time 0.020536 -2022-12-06 10:34:56,513 - Epoch: [28][ 970/ 1200] Overall Loss 0.289060 Objective Loss 0.289060 LR 0.001000 Time 0.020528 -2022-12-06 10:34:56,713 - Epoch: [28][ 980/ 1200] Overall Loss 0.289102 Objective Loss 0.289102 LR 0.001000 Time 0.020523 -2022-12-06 10:34:56,912 - Epoch: [28][ 990/ 1200] Overall Loss 0.289276 Objective Loss 0.289276 LR 0.001000 Time 0.020516 -2022-12-06 10:34:57,113 - Epoch: [28][ 1000/ 1200] Overall Loss 0.289497 Objective Loss 0.289497 LR 0.001000 Time 0.020511 -2022-12-06 10:34:57,312 - Epoch: [28][ 1010/ 1200] Overall Loss 0.289606 Objective Loss 0.289606 LR 0.001000 Time 0.020504 -2022-12-06 10:34:57,512 - Epoch: [28][ 1020/ 1200] Overall Loss 0.289615 Objective Loss 0.289615 LR 0.001000 Time 0.020499 -2022-12-06 10:34:57,710 - Epoch: [28][ 1030/ 1200] Overall Loss 0.289747 Objective Loss 0.289747 LR 0.001000 Time 0.020492 -2022-12-06 10:34:57,912 - Epoch: [28][ 1040/ 1200] Overall Loss 0.289852 Objective Loss 0.289852 LR 0.001000 Time 0.020488 -2022-12-06 10:34:58,110 - Epoch: [28][ 1050/ 1200] Overall Loss 0.289974 Objective Loss 0.289974 LR 0.001000 Time 0.020481 -2022-12-06 10:34:58,311 - Epoch: [28][ 1060/ 1200] Overall Loss 0.290041 Objective Loss 0.290041 LR 0.001000 Time 0.020477 -2022-12-06 10:34:58,509 - Epoch: [28][ 1070/ 1200] Overall Loss 0.290217 Objective Loss 0.290217 LR 0.001000 Time 0.020470 -2022-12-06 10:34:58,709 - Epoch: [28][ 1080/ 1200] Overall Loss 0.290415 Objective Loss 0.290415 LR 0.001000 Time 0.020465 -2022-12-06 10:34:58,908 - Epoch: [28][ 1090/ 1200] Overall Loss 0.290785 Objective Loss 0.290785 LR 0.001000 Time 0.020460 -2022-12-06 10:34:59,108 - Epoch: [28][ 1100/ 1200] Overall Loss 0.290748 Objective Loss 0.290748 LR 0.001000 Time 0.020455 -2022-12-06 10:34:59,307 - Epoch: [28][ 1110/ 1200] Overall Loss 0.291042 Objective Loss 0.291042 LR 0.001000 Time 0.020450 -2022-12-06 10:34:59,508 - Epoch: [28][ 1120/ 1200] Overall Loss 0.290974 Objective Loss 0.290974 LR 0.001000 Time 0.020446 -2022-12-06 10:34:59,705 - Epoch: [28][ 1130/ 1200] Overall Loss 0.291144 Objective Loss 0.291144 LR 0.001000 Time 0.020439 -2022-12-06 10:34:59,905 - Epoch: [28][ 1140/ 1200] Overall Loss 0.291394 Objective Loss 0.291394 LR 0.001000 Time 0.020435 -2022-12-06 10:35:00,104 - Epoch: [28][ 1150/ 1200] Overall Loss 0.291533 Objective Loss 0.291533 LR 0.001000 Time 0.020429 -2022-12-06 10:35:00,304 - Epoch: [28][ 1160/ 1200] Overall Loss 0.291448 Objective Loss 0.291448 LR 0.001000 Time 0.020426 -2022-12-06 10:35:00,503 - Epoch: [28][ 1170/ 1200] Overall Loss 0.291721 Objective Loss 0.291721 LR 0.001000 Time 0.020420 -2022-12-06 10:35:00,704 - Epoch: [28][ 1180/ 1200] Overall Loss 0.292007 Objective Loss 0.292007 LR 0.001000 Time 0.020417 -2022-12-06 10:35:00,903 - Epoch: [28][ 1190/ 1200] Overall Loss 0.291716 Objective Loss 0.291716 LR 0.001000 Time 0.020412 -2022-12-06 10:35:01,137 - Epoch: [28][ 1200/ 1200] Overall Loss 0.291843 Objective Loss 0.291843 Top1 84.728033 Top5 97.907950 LR 0.001000 Time 0.020436 -2022-12-06 10:35:01,225 - --- validate (epoch=28)----------- -2022-12-06 10:35:01,225 - 34129 samples (256 per mini-batch) -2022-12-06 10:35:01,676 - Epoch: [28][ 10/ 134] Loss 0.285005 Top1 84.257812 Top5 97.968750 -2022-12-06 10:35:01,809 - Epoch: [28][ 20/ 134] Loss 0.294440 Top1 84.375000 Top5 97.910156 -2022-12-06 10:35:01,942 - Epoch: [28][ 30/ 134] Loss 0.302849 Top1 84.244792 Top5 98.111979 -2022-12-06 10:35:02,075 - Epoch: [28][ 40/ 134] Loss 0.298384 Top1 84.492188 Top5 98.154297 -2022-12-06 10:35:02,207 - Epoch: [28][ 50/ 134] Loss 0.298620 Top1 84.351562 Top5 98.164062 -2022-12-06 10:35:02,340 - Epoch: [28][ 60/ 134] Loss 0.302452 Top1 84.348958 Top5 98.020833 -2022-12-06 10:35:02,469 - Epoch: [28][ 70/ 134] Loss 0.302270 Top1 84.458705 Top5 97.979911 -2022-12-06 10:35:02,597 - Epoch: [28][ 80/ 134] Loss 0.303849 Top1 84.448242 Top5 97.973633 -2022-12-06 10:35:02,725 - Epoch: [28][ 90/ 134] Loss 0.307409 Top1 84.231771 Top5 97.999132 -2022-12-06 10:35:02,855 - Epoch: [28][ 100/ 134] Loss 0.309022 Top1 84.171875 Top5 98.000000 -2022-12-06 10:35:02,985 - Epoch: [28][ 110/ 134] Loss 0.310514 Top1 84.151278 Top5 97.950994 -2022-12-06 10:35:03,116 - Epoch: [28][ 120/ 134] Loss 0.311127 Top1 84.163411 Top5 97.929688 -2022-12-06 10:35:03,248 - Epoch: [28][ 130/ 134] Loss 0.309161 Top1 84.179688 Top5 97.932692 -2022-12-06 10:35:03,288 - Epoch: [28][ 134/ 134] Loss 0.311389 Top1 84.139588 Top5 97.922588 -2022-12-06 10:35:03,378 - ==> Top1: 84.140 Top5: 97.923 Loss: 0.311 - -2022-12-06 10:35:03,378 - ==> Confusion: -[[ 895 1 1 1 8 4 1 2 1 47 1 5 2 1 9 1 4 3 2 2 5] - [ 2 888 2 3 9 36 7 26 0 1 3 2 6 1 3 1 3 1 13 6 14] - [ 3 3 1002 13 1 1 34 14 0 0 2 5 0 2 1 4 1 1 4 3 9] - [ 3 2 24 943 0 1 0 1 1 1 10 3 4 2 6 0 2 1 10 1 5] - [ 8 3 3 0 942 8 2 2 0 4 1 2 0 0 15 7 10 4 0 3 6] - [ 4 11 0 2 9 931 7 37 2 1 2 17 2 14 6 0 0 1 1 17 5] - [ 0 2 14 1 1 5 1066 6 0 0 1 2 1 0 1 5 0 0 2 7 4] - [ 0 4 6 1 1 14 2 960 0 1 2 5 1 1 1 3 0 1 28 20 3] - [ 10 1 0 1 0 5 0 1 929 50 19 2 3 8 21 0 1 4 3 1 5] - [ 93 0 4 0 6 1 0 3 19 842 3 4 0 14 4 0 1 1 0 0 6] - [ 0 2 7 11 2 1 2 4 1 1 951 2 2 9 6 0 2 0 7 4 5] - [ 2 0 4 0 0 8 3 7 2 0 0 972 20 3 2 2 2 3 0 19 2] - [ 2 0 1 6 2 2 1 4 0 0 0 53 863 1 0 10 0 6 1 6 11] - [ 0 2 1 0 1 13 0 3 5 10 15 15 3 932 2 4 3 0 0 8 6] - [ 5 2 3 25 3 0 0 1 14 2 0 2 5 6 1042 0 3 1 4 2 10] - [ 4 1 2 0 3 1 5 1 0 0 0 11 4 2 0 985 4 9 0 4 7] - [ 2 3 1 2 2 0 3 2 0 1 0 10 3 0 1 13 1006 1 1 10 11] - [ 2 1 1 2 0 1 1 2 0 1 1 20 15 2 1 21 2 958 1 3 1] - [ 3 1 4 16 0 1 0 28 0 0 6 6 3 1 8 0 2 2 919 3 5] - [ 1 1 5 1 1 4 7 2 0 0 0 17 6 2 1 2 1 0 1 1020 8] - [ 137 171 222 137 126 193 91 200 39 78 185 163 382 323 172 163 154 74 184 368 9664]] - -2022-12-06 10:35:04,050 - ==> Best [Top1: 84.140 Top5: 97.923 Sparsity:0.00 Params: 5376 on epoch: 28] -2022-12-06 10:35:04,050 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:35:04,057 - - -2022-12-06 10:35:04,057 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:35:04,985 - Epoch: [29][ 10/ 1200] Overall Loss 0.298452 Objective Loss 0.298452 LR 0.001000 Time 0.092705 -2022-12-06 10:35:05,175 - Epoch: [29][ 20/ 1200] Overall Loss 0.276364 Objective Loss 0.276364 LR 0.001000 Time 0.055841 -2022-12-06 10:35:05,365 - Epoch: [29][ 30/ 1200] Overall Loss 0.268746 Objective Loss 0.268746 LR 0.001000 Time 0.043538 -2022-12-06 10:35:05,555 - Epoch: [29][ 40/ 1200] Overall Loss 0.270686 Objective Loss 0.270686 LR 0.001000 Time 0.037382 -2022-12-06 10:35:05,745 - Epoch: [29][ 50/ 1200] Overall Loss 0.270450 Objective Loss 0.270450 LR 0.001000 Time 0.033692 -2022-12-06 10:35:05,934 - Epoch: [29][ 60/ 1200] Overall Loss 0.265525 Objective Loss 0.265525 LR 0.001000 Time 0.031228 -2022-12-06 10:35:06,123 - Epoch: [29][ 70/ 1200] Overall Loss 0.267317 Objective Loss 0.267317 LR 0.001000 Time 0.029458 -2022-12-06 10:35:06,313 - Epoch: [29][ 80/ 1200] Overall Loss 0.267286 Objective Loss 0.267286 LR 0.001000 Time 0.028142 -2022-12-06 10:35:06,503 - Epoch: [29][ 90/ 1200] Overall Loss 0.272186 Objective Loss 0.272186 LR 0.001000 Time 0.027121 -2022-12-06 10:35:06,693 - Epoch: [29][ 100/ 1200] Overall Loss 0.273163 Objective Loss 0.273163 LR 0.001000 Time 0.026305 -2022-12-06 10:35:06,883 - Epoch: [29][ 110/ 1200] Overall Loss 0.273848 Objective Loss 0.273848 LR 0.001000 Time 0.025635 -2022-12-06 10:35:07,073 - Epoch: [29][ 120/ 1200] Overall Loss 0.272099 Objective Loss 0.272099 LR 0.001000 Time 0.025077 -2022-12-06 10:35:07,263 - Epoch: [29][ 130/ 1200] Overall Loss 0.271362 Objective Loss 0.271362 LR 0.001000 Time 0.024603 -2022-12-06 10:35:07,452 - Epoch: [29][ 140/ 1200] Overall Loss 0.272887 Objective Loss 0.272887 LR 0.001000 Time 0.024193 -2022-12-06 10:35:07,643 - Epoch: [29][ 150/ 1200] Overall Loss 0.275884 Objective Loss 0.275884 LR 0.001000 Time 0.023847 -2022-12-06 10:35:07,832 - Epoch: [29][ 160/ 1200] Overall Loss 0.275536 Objective Loss 0.275536 LR 0.001000 Time 0.023540 -2022-12-06 10:35:08,022 - Epoch: [29][ 170/ 1200] Overall Loss 0.276518 Objective Loss 0.276518 LR 0.001000 Time 0.023270 -2022-12-06 10:35:08,212 - Epoch: [29][ 180/ 1200] Overall Loss 0.276551 Objective Loss 0.276551 LR 0.001000 Time 0.023029 -2022-12-06 10:35:08,403 - Epoch: [29][ 190/ 1200] Overall Loss 0.275392 Objective Loss 0.275392 LR 0.001000 Time 0.022819 -2022-12-06 10:35:08,593 - Epoch: [29][ 200/ 1200] Overall Loss 0.277170 Objective Loss 0.277170 LR 0.001000 Time 0.022626 -2022-12-06 10:35:08,783 - Epoch: [29][ 210/ 1200] Overall Loss 0.276972 Objective Loss 0.276972 LR 0.001000 Time 0.022450 -2022-12-06 10:35:08,974 - Epoch: [29][ 220/ 1200] Overall Loss 0.277093 Objective Loss 0.277093 LR 0.001000 Time 0.022293 -2022-12-06 10:35:09,163 - Epoch: [29][ 230/ 1200] Overall Loss 0.277147 Objective Loss 0.277147 LR 0.001000 Time 0.022145 -2022-12-06 10:35:09,354 - Epoch: [29][ 240/ 1200] Overall Loss 0.278014 Objective Loss 0.278014 LR 0.001000 Time 0.022014 -2022-12-06 10:35:09,543 - Epoch: [29][ 250/ 1200] Overall Loss 0.278113 Objective Loss 0.278113 LR 0.001000 Time 0.021890 -2022-12-06 10:35:09,733 - Epoch: [29][ 260/ 1200] Overall Loss 0.278610 Objective Loss 0.278610 LR 0.001000 Time 0.021775 -2022-12-06 10:35:09,923 - Epoch: [29][ 270/ 1200] Overall Loss 0.279815 Objective Loss 0.279815 LR 0.001000 Time 0.021670 -2022-12-06 10:35:10,114 - Epoch: [29][ 280/ 1200] Overall Loss 0.279880 Objective Loss 0.279880 LR 0.001000 Time 0.021575 -2022-12-06 10:35:10,304 - Epoch: [29][ 290/ 1200] Overall Loss 0.280429 Objective Loss 0.280429 LR 0.001000 Time 0.021484 -2022-12-06 10:35:10,493 - Epoch: [29][ 300/ 1200] Overall Loss 0.279834 Objective Loss 0.279834 LR 0.001000 Time 0.021398 -2022-12-06 10:35:10,683 - Epoch: [29][ 310/ 1200] Overall Loss 0.279621 Objective Loss 0.279621 LR 0.001000 Time 0.021319 -2022-12-06 10:35:10,873 - Epoch: [29][ 320/ 1200] Overall Loss 0.279198 Objective Loss 0.279198 LR 0.001000 Time 0.021245 -2022-12-06 10:35:11,064 - Epoch: [29][ 330/ 1200] Overall Loss 0.278563 Objective Loss 0.278563 LR 0.001000 Time 0.021176 -2022-12-06 10:35:11,254 - Epoch: [29][ 340/ 1200] Overall Loss 0.278381 Objective Loss 0.278381 LR 0.001000 Time 0.021110 -2022-12-06 10:35:11,444 - Epoch: [29][ 350/ 1200] Overall Loss 0.278344 Objective Loss 0.278344 LR 0.001000 Time 0.021050 -2022-12-06 10:35:11,634 - Epoch: [29][ 360/ 1200] Overall Loss 0.278394 Objective Loss 0.278394 LR 0.001000 Time 0.020991 -2022-12-06 10:35:11,824 - Epoch: [29][ 370/ 1200] Overall Loss 0.278691 Objective Loss 0.278691 LR 0.001000 Time 0.020936 -2022-12-06 10:35:12,014 - Epoch: [29][ 380/ 1200] Overall Loss 0.277759 Objective Loss 0.277759 LR 0.001000 Time 0.020883 -2022-12-06 10:35:12,204 - Epoch: [29][ 390/ 1200] Overall Loss 0.278431 Objective Loss 0.278431 LR 0.001000 Time 0.020833 -2022-12-06 10:35:12,393 - Epoch: [29][ 400/ 1200] Overall Loss 0.278533 Objective Loss 0.278533 LR 0.001000 Time 0.020785 -2022-12-06 10:35:12,584 - Epoch: [29][ 410/ 1200] Overall Loss 0.278873 Objective Loss 0.278873 LR 0.001000 Time 0.020741 -2022-12-06 10:35:12,773 - Epoch: [29][ 420/ 1200] Overall Loss 0.278737 Objective Loss 0.278737 LR 0.001000 Time 0.020697 -2022-12-06 10:35:12,963 - Epoch: [29][ 430/ 1200] Overall Loss 0.278821 Objective Loss 0.278821 LR 0.001000 Time 0.020655 -2022-12-06 10:35:13,153 - Epoch: [29][ 440/ 1200] Overall Loss 0.278684 Objective Loss 0.278684 LR 0.001000 Time 0.020617 -2022-12-06 10:35:13,343 - Epoch: [29][ 450/ 1200] Overall Loss 0.278955 Objective Loss 0.278955 LR 0.001000 Time 0.020580 -2022-12-06 10:35:13,533 - Epoch: [29][ 460/ 1200] Overall Loss 0.278927 Objective Loss 0.278927 LR 0.001000 Time 0.020545 -2022-12-06 10:35:13,723 - Epoch: [29][ 470/ 1200] Overall Loss 0.278658 Objective Loss 0.278658 LR 0.001000 Time 0.020511 -2022-12-06 10:35:13,914 - Epoch: [29][ 480/ 1200] Overall Loss 0.278524 Objective Loss 0.278524 LR 0.001000 Time 0.020478 -2022-12-06 10:35:14,104 - Epoch: [29][ 490/ 1200] Overall Loss 0.278991 Objective Loss 0.278991 LR 0.001000 Time 0.020448 -2022-12-06 10:35:14,294 - Epoch: [29][ 500/ 1200] Overall Loss 0.279513 Objective Loss 0.279513 LR 0.001000 Time 0.020417 -2022-12-06 10:35:14,484 - Epoch: [29][ 510/ 1200] Overall Loss 0.280284 Objective Loss 0.280284 LR 0.001000 Time 0.020390 -2022-12-06 10:35:14,675 - Epoch: [29][ 520/ 1200] Overall Loss 0.281084 Objective Loss 0.281084 LR 0.001000 Time 0.020363 -2022-12-06 10:35:14,864 - Epoch: [29][ 530/ 1200] Overall Loss 0.282806 Objective Loss 0.282806 LR 0.001000 Time 0.020336 -2022-12-06 10:35:15,055 - Epoch: [29][ 540/ 1200] Overall Loss 0.283282 Objective Loss 0.283282 LR 0.001000 Time 0.020310 -2022-12-06 10:35:15,244 - Epoch: [29][ 550/ 1200] Overall Loss 0.283402 Objective Loss 0.283402 LR 0.001000 Time 0.020285 -2022-12-06 10:35:15,435 - Epoch: [29][ 560/ 1200] Overall Loss 0.283539 Objective Loss 0.283539 LR 0.001000 Time 0.020262 -2022-12-06 10:35:15,627 - Epoch: [29][ 570/ 1200] Overall Loss 0.283848 Objective Loss 0.283848 LR 0.001000 Time 0.020242 -2022-12-06 10:35:15,818 - Epoch: [29][ 580/ 1200] Overall Loss 0.284116 Objective Loss 0.284116 LR 0.001000 Time 0.020222 -2022-12-06 10:35:16,009 - Epoch: [29][ 590/ 1200] Overall Loss 0.284158 Objective Loss 0.284158 LR 0.001000 Time 0.020201 -2022-12-06 10:35:16,201 - Epoch: [29][ 600/ 1200] Overall Loss 0.284344 Objective Loss 0.284344 LR 0.001000 Time 0.020184 -2022-12-06 10:35:16,392 - Epoch: [29][ 610/ 1200] Overall Loss 0.284446 Objective Loss 0.284446 LR 0.001000 Time 0.020166 -2022-12-06 10:35:16,584 - Epoch: [29][ 620/ 1200] Overall Loss 0.284862 Objective Loss 0.284862 LR 0.001000 Time 0.020148 -2022-12-06 10:35:16,776 - Epoch: [29][ 630/ 1200] Overall Loss 0.285229 Objective Loss 0.285229 LR 0.001000 Time 0.020132 -2022-12-06 10:35:16,967 - Epoch: [29][ 640/ 1200] Overall Loss 0.285240 Objective Loss 0.285240 LR 0.001000 Time 0.020117 -2022-12-06 10:35:17,159 - Epoch: [29][ 650/ 1200] Overall Loss 0.285560 Objective Loss 0.285560 LR 0.001000 Time 0.020101 -2022-12-06 10:35:17,351 - Epoch: [29][ 660/ 1200] Overall Loss 0.285661 Objective Loss 0.285661 LR 0.001000 Time 0.020087 -2022-12-06 10:35:17,543 - Epoch: [29][ 670/ 1200] Overall Loss 0.285877 Objective Loss 0.285877 LR 0.001000 Time 0.020072 -2022-12-06 10:35:17,734 - Epoch: [29][ 680/ 1200] Overall Loss 0.285631 Objective Loss 0.285631 LR 0.001000 Time 0.020057 -2022-12-06 10:35:17,925 - Epoch: [29][ 690/ 1200] Overall Loss 0.285584 Objective Loss 0.285584 LR 0.001000 Time 0.020043 -2022-12-06 10:35:18,118 - Epoch: [29][ 700/ 1200] Overall Loss 0.285553 Objective Loss 0.285553 LR 0.001000 Time 0.020030 -2022-12-06 10:35:18,310 - Epoch: [29][ 710/ 1200] Overall Loss 0.285664 Objective Loss 0.285664 LR 0.001000 Time 0.020018 -2022-12-06 10:35:18,502 - Epoch: [29][ 720/ 1200] Overall Loss 0.285732 Objective Loss 0.285732 LR 0.001000 Time 0.020007 -2022-12-06 10:35:18,693 - Epoch: [29][ 730/ 1200] Overall Loss 0.286555 Objective Loss 0.286555 LR 0.001000 Time 0.019994 -2022-12-06 10:35:18,887 - Epoch: [29][ 740/ 1200] Overall Loss 0.287246 Objective Loss 0.287246 LR 0.001000 Time 0.019985 -2022-12-06 10:35:19,080 - Epoch: [29][ 750/ 1200] Overall Loss 0.287321 Objective Loss 0.287321 LR 0.001000 Time 0.019974 -2022-12-06 10:35:19,272 - Epoch: [29][ 760/ 1200] Overall Loss 0.287163 Objective Loss 0.287163 LR 0.001000 Time 0.019964 -2022-12-06 10:35:19,466 - Epoch: [29][ 770/ 1200] Overall Loss 0.287360 Objective Loss 0.287360 LR 0.001000 Time 0.019955 -2022-12-06 10:35:19,658 - Epoch: [29][ 780/ 1200] Overall Loss 0.287871 Objective Loss 0.287871 LR 0.001000 Time 0.019945 -2022-12-06 10:35:19,851 - Epoch: [29][ 790/ 1200] Overall Loss 0.288226 Objective Loss 0.288226 LR 0.001000 Time 0.019936 -2022-12-06 10:35:20,043 - Epoch: [29][ 800/ 1200] Overall Loss 0.288708 Objective Loss 0.288708 LR 0.001000 Time 0.019926 -2022-12-06 10:35:20,236 - Epoch: [29][ 810/ 1200] Overall Loss 0.288927 Objective Loss 0.288927 LR 0.001000 Time 0.019918 -2022-12-06 10:35:20,428 - Epoch: [29][ 820/ 1200] Overall Loss 0.289279 Objective Loss 0.289279 LR 0.001000 Time 0.019909 -2022-12-06 10:35:20,620 - Epoch: [29][ 830/ 1200] Overall Loss 0.289210 Objective Loss 0.289210 LR 0.001000 Time 0.019900 -2022-12-06 10:35:20,813 - Epoch: [29][ 840/ 1200] Overall Loss 0.289396 Objective Loss 0.289396 LR 0.001000 Time 0.019891 -2022-12-06 10:35:21,005 - Epoch: [29][ 850/ 1200] Overall Loss 0.289561 Objective Loss 0.289561 LR 0.001000 Time 0.019883 -2022-12-06 10:35:21,198 - Epoch: [29][ 860/ 1200] Overall Loss 0.289353 Objective Loss 0.289353 LR 0.001000 Time 0.019875 -2022-12-06 10:35:21,390 - Epoch: [29][ 870/ 1200] Overall Loss 0.289432 Objective Loss 0.289432 LR 0.001000 Time 0.019867 -2022-12-06 10:35:21,583 - Epoch: [29][ 880/ 1200] Overall Loss 0.289388 Objective Loss 0.289388 LR 0.001000 Time 0.019859 -2022-12-06 10:35:21,774 - Epoch: [29][ 890/ 1200] Overall Loss 0.289554 Objective Loss 0.289554 LR 0.001000 Time 0.019851 -2022-12-06 10:35:21,966 - Epoch: [29][ 900/ 1200] Overall Loss 0.289507 Objective Loss 0.289507 LR 0.001000 Time 0.019843 -2022-12-06 10:35:22,158 - Epoch: [29][ 910/ 1200] Overall Loss 0.289446 Objective Loss 0.289446 LR 0.001000 Time 0.019836 -2022-12-06 10:35:22,351 - Epoch: [29][ 920/ 1200] Overall Loss 0.289338 Objective Loss 0.289338 LR 0.001000 Time 0.019828 -2022-12-06 10:35:22,543 - Epoch: [29][ 930/ 1200] Overall Loss 0.289552 Objective Loss 0.289552 LR 0.001000 Time 0.019821 -2022-12-06 10:35:22,735 - Epoch: [29][ 940/ 1200] Overall Loss 0.289301 Objective Loss 0.289301 LR 0.001000 Time 0.019814 -2022-12-06 10:35:22,928 - Epoch: [29][ 950/ 1200] Overall Loss 0.289368 Objective Loss 0.289368 LR 0.001000 Time 0.019808 -2022-12-06 10:35:23,120 - Epoch: [29][ 960/ 1200] Overall Loss 0.289762 Objective Loss 0.289762 LR 0.001000 Time 0.019801 -2022-12-06 10:35:23,312 - Epoch: [29][ 970/ 1200] Overall Loss 0.289799 Objective Loss 0.289799 LR 0.001000 Time 0.019795 -2022-12-06 10:35:23,505 - Epoch: [29][ 980/ 1200] Overall Loss 0.290047 Objective Loss 0.290047 LR 0.001000 Time 0.019789 -2022-12-06 10:35:23,697 - Epoch: [29][ 990/ 1200] Overall Loss 0.290121 Objective Loss 0.290121 LR 0.001000 Time 0.019783 -2022-12-06 10:35:23,890 - Epoch: [29][ 1000/ 1200] Overall Loss 0.290272 Objective Loss 0.290272 LR 0.001000 Time 0.019777 -2022-12-06 10:35:24,082 - Epoch: [29][ 1010/ 1200] Overall Loss 0.290198 Objective Loss 0.290198 LR 0.001000 Time 0.019771 -2022-12-06 10:35:24,275 - Epoch: [29][ 1020/ 1200] Overall Loss 0.290615 Objective Loss 0.290615 LR 0.001000 Time 0.019765 -2022-12-06 10:35:24,467 - Epoch: [29][ 1030/ 1200] Overall Loss 0.290467 Objective Loss 0.290467 LR 0.001000 Time 0.019760 -2022-12-06 10:35:24,659 - Epoch: [29][ 1040/ 1200] Overall Loss 0.290443 Objective Loss 0.290443 LR 0.001000 Time 0.019754 -2022-12-06 10:35:24,851 - Epoch: [29][ 1050/ 1200] Overall Loss 0.290604 Objective Loss 0.290604 LR 0.001000 Time 0.019748 -2022-12-06 10:35:25,043 - Epoch: [29][ 1060/ 1200] Overall Loss 0.290910 Objective Loss 0.290910 LR 0.001000 Time 0.019742 -2022-12-06 10:35:25,236 - Epoch: [29][ 1070/ 1200] Overall Loss 0.291129 Objective Loss 0.291129 LR 0.001000 Time 0.019737 -2022-12-06 10:35:25,428 - Epoch: [29][ 1080/ 1200] Overall Loss 0.291090 Objective Loss 0.291090 LR 0.001000 Time 0.019732 -2022-12-06 10:35:25,620 - Epoch: [29][ 1090/ 1200] Overall Loss 0.291118 Objective Loss 0.291118 LR 0.001000 Time 0.019726 -2022-12-06 10:35:25,812 - Epoch: [29][ 1100/ 1200] Overall Loss 0.291078 Objective Loss 0.291078 LR 0.001000 Time 0.019721 -2022-12-06 10:35:26,004 - Epoch: [29][ 1110/ 1200] Overall Loss 0.290911 Objective Loss 0.290911 LR 0.001000 Time 0.019716 -2022-12-06 10:35:26,196 - Epoch: [29][ 1120/ 1200] Overall Loss 0.291058 Objective Loss 0.291058 LR 0.001000 Time 0.019711 -2022-12-06 10:35:26,388 - Epoch: [29][ 1130/ 1200] Overall Loss 0.291146 Objective Loss 0.291146 LR 0.001000 Time 0.019706 -2022-12-06 10:35:26,580 - Epoch: [29][ 1140/ 1200] Overall Loss 0.290976 Objective Loss 0.290976 LR 0.001000 Time 0.019701 -2022-12-06 10:35:26,772 - Epoch: [29][ 1150/ 1200] Overall Loss 0.291526 Objective Loss 0.291526 LR 0.001000 Time 0.019696 -2022-12-06 10:35:26,964 - Epoch: [29][ 1160/ 1200] Overall Loss 0.291666 Objective Loss 0.291666 LR 0.001000 Time 0.019691 -2022-12-06 10:35:27,156 - Epoch: [29][ 1170/ 1200] Overall Loss 0.291940 Objective Loss 0.291940 LR 0.001000 Time 0.019686 -2022-12-06 10:35:27,347 - Epoch: [29][ 1180/ 1200] Overall Loss 0.291894 Objective Loss 0.291894 LR 0.001000 Time 0.019681 -2022-12-06 10:35:27,539 - Epoch: [29][ 1190/ 1200] Overall Loss 0.291940 Objective Loss 0.291940 LR 0.001000 Time 0.019677 -2022-12-06 10:35:27,764 - Epoch: [29][ 1200/ 1200] Overall Loss 0.292036 Objective Loss 0.292036 Top1 85.146444 Top5 99.163180 LR 0.001000 Time 0.019700 -2022-12-06 10:35:27,856 - --- validate (epoch=29)----------- -2022-12-06 10:35:27,856 - 34129 samples (256 per mini-batch) -2022-12-06 10:35:28,309 - Epoch: [29][ 10/ 134] Loss 0.332211 Top1 83.671875 Top5 97.382812 -2022-12-06 10:35:28,442 - Epoch: [29][ 20/ 134] Loss 0.314137 Top1 83.886719 Top5 97.734375 -2022-12-06 10:35:28,575 - Epoch: [29][ 30/ 134] Loss 0.312418 Top1 83.632812 Top5 97.864583 -2022-12-06 10:35:28,706 - Epoch: [29][ 40/ 134] Loss 0.311854 Top1 83.847656 Top5 97.763672 -2022-12-06 10:35:28,840 - Epoch: [29][ 50/ 134] Loss 0.309023 Top1 83.695312 Top5 97.882812 -2022-12-06 10:35:28,970 - Epoch: [29][ 60/ 134] Loss 0.314336 Top1 83.574219 Top5 97.838542 -2022-12-06 10:35:29,114 - Epoch: [29][ 70/ 134] Loss 0.314826 Top1 83.521205 Top5 97.812500 -2022-12-06 10:35:29,257 - Epoch: [29][ 80/ 134] Loss 0.315602 Top1 83.496094 Top5 97.841797 -2022-12-06 10:35:29,397 - Epoch: [29][ 90/ 134] Loss 0.311734 Top1 83.563368 Top5 97.808160 -2022-12-06 10:35:29,545 - Epoch: [29][ 100/ 134] Loss 0.312106 Top1 83.542969 Top5 97.785156 -2022-12-06 10:35:29,687 - Epoch: [29][ 110/ 134] Loss 0.311342 Top1 83.551136 Top5 97.787642 -2022-12-06 10:35:29,831 - Epoch: [29][ 120/ 134] Loss 0.309940 Top1 83.619792 Top5 97.770182 -2022-12-06 10:35:29,960 - Epoch: [29][ 130/ 134] Loss 0.310627 Top1 83.620793 Top5 97.740385 -2022-12-06 10:35:29,997 - Epoch: [29][ 134/ 134] Loss 0.310178 Top1 83.656128 Top5 97.749714 -2022-12-06 10:35:30,089 - ==> Top1: 83.656 Top5: 97.750 Loss: 0.310 - -2022-12-06 10:35:30,090 - ==> Confusion: -[[ 885 3 1 0 3 7 1 1 15 55 0 6 2 2 2 2 6 1 0 1 3] - [ 0 929 0 0 3 40 2 13 1 1 5 7 0 2 1 0 5 2 5 2 9] - [ 4 5 979 25 2 3 28 7 0 0 4 8 2 3 3 6 0 2 7 3 12] - [ 3 3 14 934 0 6 0 0 1 0 9 1 5 3 14 0 3 5 13 1 5] - [ 9 15 1 1 909 19 0 1 1 4 2 7 0 2 9 5 24 3 0 2 6] - [ 2 18 2 0 5 971 1 9 3 1 0 25 4 10 2 0 3 1 1 5 6] - [ 0 4 10 4 0 3 1060 3 0 1 4 4 2 2 1 8 1 1 0 7 3] - [ 1 11 10 2 2 54 7 902 0 0 1 15 1 2 0 2 1 0 22 17 4] - [ 4 6 0 1 0 5 0 0 978 31 5 3 2 8 11 1 2 0 3 2 2] - [ 55 0 4 1 3 4 0 2 43 857 1 3 0 16 2 0 0 0 1 0 9] - [ 0 4 6 14 1 1 1 2 8 2 941 4 3 13 3 0 0 0 9 2 5] - [ 2 4 2 0 0 6 2 2 0 1 1 1001 13 2 1 2 2 4 0 6 0] - [ 2 3 0 4 1 2 0 1 0 0 0 65 864 1 2 7 1 8 1 2 5] - [ 0 3 1 0 1 14 0 0 11 8 8 10 5 942 1 3 2 0 1 4 9] - [ 10 3 0 13 1 3 0 0 31 1 0 2 3 6 1025 0 4 2 15 1 10] - [ 2 4 1 2 2 2 3 0 0 0 1 18 8 0 0 972 9 12 0 2 5] - [ 2 4 3 2 1 3 0 1 1 0 0 6 1 1 1 9 1025 1 0 5 6] - [ 1 2 1 0 0 2 1 1 0 2 1 18 29 1 0 10 1 963 0 2 1] - [ 3 6 4 13 0 6 1 14 3 2 8 7 2 2 2 0 1 1 926 4 3] - [ 3 2 1 0 0 7 3 1 0 0 0 41 7 6 0 4 9 2 0 993 1] - [ 108 294 151 117 63 293 84 131 94 77 161 232 423 322 126 137 305 95 201 325 9487]] - -2022-12-06 10:35:30,757 - ==> Best [Top1: 84.140 Top5: 97.923 Sparsity:0.00 Params: 5376 on epoch: 28] -2022-12-06 10:35:30,757 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:35:30,763 - - -2022-12-06 10:35:30,763 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:35:31,692 - Epoch: [30][ 10/ 1200] Overall Loss 0.269829 Objective Loss 0.269829 LR 0.001000 Time 0.092817 -2022-12-06 10:35:31,887 - Epoch: [30][ 20/ 1200] Overall Loss 0.272487 Objective Loss 0.272487 LR 0.001000 Time 0.056139 -2022-12-06 10:35:32,079 - Epoch: [30][ 30/ 1200] Overall Loss 0.275721 Objective Loss 0.275721 LR 0.001000 Time 0.043794 -2022-12-06 10:35:32,271 - Epoch: [30][ 40/ 1200] Overall Loss 0.275633 Objective Loss 0.275633 LR 0.001000 Time 0.037630 -2022-12-06 10:35:32,462 - Epoch: [30][ 50/ 1200] Overall Loss 0.273263 Objective Loss 0.273263 LR 0.001000 Time 0.033924 -2022-12-06 10:35:32,654 - Epoch: [30][ 60/ 1200] Overall Loss 0.275596 Objective Loss 0.275596 LR 0.001000 Time 0.031452 -2022-12-06 10:35:32,845 - Epoch: [30][ 70/ 1200] Overall Loss 0.278737 Objective Loss 0.278737 LR 0.001000 Time 0.029689 -2022-12-06 10:35:33,037 - Epoch: [30][ 80/ 1200] Overall Loss 0.276425 Objective Loss 0.276425 LR 0.001000 Time 0.028368 -2022-12-06 10:35:33,228 - Epoch: [30][ 90/ 1200] Overall Loss 0.273445 Objective Loss 0.273445 LR 0.001000 Time 0.027333 -2022-12-06 10:35:33,420 - Epoch: [30][ 100/ 1200] Overall Loss 0.275708 Objective Loss 0.275708 LR 0.001000 Time 0.026513 -2022-12-06 10:35:33,612 - Epoch: [30][ 110/ 1200] Overall Loss 0.275575 Objective Loss 0.275575 LR 0.001000 Time 0.025842 -2022-12-06 10:35:33,804 - Epoch: [30][ 120/ 1200] Overall Loss 0.275344 Objective Loss 0.275344 LR 0.001000 Time 0.025284 -2022-12-06 10:35:33,996 - Epoch: [30][ 130/ 1200] Overall Loss 0.274160 Objective Loss 0.274160 LR 0.001000 Time 0.024809 -2022-12-06 10:35:34,187 - Epoch: [30][ 140/ 1200] Overall Loss 0.274735 Objective Loss 0.274735 LR 0.001000 Time 0.024402 -2022-12-06 10:35:34,380 - Epoch: [30][ 150/ 1200] Overall Loss 0.274377 Objective Loss 0.274377 LR 0.001000 Time 0.024053 -2022-12-06 10:35:34,572 - Epoch: [30][ 160/ 1200] Overall Loss 0.274413 Objective Loss 0.274413 LR 0.001000 Time 0.023747 -2022-12-06 10:35:34,764 - Epoch: [30][ 170/ 1200] Overall Loss 0.275936 Objective Loss 0.275936 LR 0.001000 Time 0.023478 -2022-12-06 10:35:34,956 - Epoch: [30][ 180/ 1200] Overall Loss 0.274794 Objective Loss 0.274794 LR 0.001000 Time 0.023237 -2022-12-06 10:35:35,148 - Epoch: [30][ 190/ 1200] Overall Loss 0.275272 Objective Loss 0.275272 LR 0.001000 Time 0.023020 -2022-12-06 10:35:35,340 - Epoch: [30][ 200/ 1200] Overall Loss 0.274796 Objective Loss 0.274796 LR 0.001000 Time 0.022826 -2022-12-06 10:35:35,531 - Epoch: [30][ 210/ 1200] Overall Loss 0.275166 Objective Loss 0.275166 LR 0.001000 Time 0.022648 -2022-12-06 10:35:35,723 - Epoch: [30][ 220/ 1200] Overall Loss 0.276253 Objective Loss 0.276253 LR 0.001000 Time 0.022486 -2022-12-06 10:35:35,915 - Epoch: [30][ 230/ 1200] Overall Loss 0.278133 Objective Loss 0.278133 LR 0.001000 Time 0.022341 -2022-12-06 10:35:36,106 - Epoch: [30][ 240/ 1200] Overall Loss 0.278657 Objective Loss 0.278657 LR 0.001000 Time 0.022206 -2022-12-06 10:35:36,298 - Epoch: [30][ 250/ 1200] Overall Loss 0.278876 Objective Loss 0.278876 LR 0.001000 Time 0.022082 -2022-12-06 10:35:36,489 - Epoch: [30][ 260/ 1200] Overall Loss 0.278707 Objective Loss 0.278707 LR 0.001000 Time 0.021967 -2022-12-06 10:35:36,681 - Epoch: [30][ 270/ 1200] Overall Loss 0.278342 Objective Loss 0.278342 LR 0.001000 Time 0.021861 -2022-12-06 10:35:36,872 - Epoch: [30][ 280/ 1200] Overall Loss 0.279702 Objective Loss 0.279702 LR 0.001000 Time 0.021762 -2022-12-06 10:35:37,064 - Epoch: [30][ 290/ 1200] Overall Loss 0.280501 Objective Loss 0.280501 LR 0.001000 Time 0.021672 -2022-12-06 10:35:37,259 - Epoch: [30][ 300/ 1200] Overall Loss 0.280073 Objective Loss 0.280073 LR 0.001000 Time 0.021598 -2022-12-06 10:35:37,456 - Epoch: [30][ 310/ 1200] Overall Loss 0.280124 Objective Loss 0.280124 LR 0.001000 Time 0.021533 -2022-12-06 10:35:37,652 - Epoch: [30][ 320/ 1200] Overall Loss 0.279633 Objective Loss 0.279633 LR 0.001000 Time 0.021472 -2022-12-06 10:35:37,848 - Epoch: [30][ 330/ 1200] Overall Loss 0.279891 Objective Loss 0.279891 LR 0.001000 Time 0.021414 -2022-12-06 10:35:38,044 - Epoch: [30][ 340/ 1200] Overall Loss 0.279517 Objective Loss 0.279517 LR 0.001000 Time 0.021359 -2022-12-06 10:35:38,240 - Epoch: [30][ 350/ 1200] Overall Loss 0.280452 Objective Loss 0.280452 LR 0.001000 Time 0.021307 -2022-12-06 10:35:38,436 - Epoch: [30][ 360/ 1200] Overall Loss 0.280317 Objective Loss 0.280317 LR 0.001000 Time 0.021258 -2022-12-06 10:35:38,632 - Epoch: [30][ 370/ 1200] Overall Loss 0.279568 Objective Loss 0.279568 LR 0.001000 Time 0.021212 -2022-12-06 10:35:38,829 - Epoch: [30][ 380/ 1200] Overall Loss 0.279117 Objective Loss 0.279117 LR 0.001000 Time 0.021169 -2022-12-06 10:35:39,024 - Epoch: [30][ 390/ 1200] Overall Loss 0.278976 Objective Loss 0.278976 LR 0.001000 Time 0.021127 -2022-12-06 10:35:39,221 - Epoch: [30][ 400/ 1200] Overall Loss 0.279060 Objective Loss 0.279060 LR 0.001000 Time 0.021088 -2022-12-06 10:35:39,417 - Epoch: [30][ 410/ 1200] Overall Loss 0.279234 Objective Loss 0.279234 LR 0.001000 Time 0.021050 -2022-12-06 10:35:39,613 - Epoch: [30][ 420/ 1200] Overall Loss 0.279756 Objective Loss 0.279756 LR 0.001000 Time 0.021014 -2022-12-06 10:35:39,809 - Epoch: [30][ 430/ 1200] Overall Loss 0.279501 Objective Loss 0.279501 LR 0.001000 Time 0.020980 -2022-12-06 10:35:40,004 - Epoch: [30][ 440/ 1200] Overall Loss 0.279611 Objective Loss 0.279611 LR 0.001000 Time 0.020947 -2022-12-06 10:35:40,201 - Epoch: [30][ 450/ 1200] Overall Loss 0.280042 Objective Loss 0.280042 LR 0.001000 Time 0.020916 -2022-12-06 10:35:40,397 - Epoch: [30][ 460/ 1200] Overall Loss 0.280031 Objective Loss 0.280031 LR 0.001000 Time 0.020887 -2022-12-06 10:35:40,593 - Epoch: [30][ 470/ 1200] Overall Loss 0.279670 Objective Loss 0.279670 LR 0.001000 Time 0.020860 -2022-12-06 10:35:40,789 - Epoch: [30][ 480/ 1200] Overall Loss 0.279515 Objective Loss 0.279515 LR 0.001000 Time 0.020832 -2022-12-06 10:35:40,985 - Epoch: [30][ 490/ 1200] Overall Loss 0.279211 Objective Loss 0.279211 LR 0.001000 Time 0.020805 -2022-12-06 10:35:41,181 - Epoch: [30][ 500/ 1200] Overall Loss 0.279154 Objective Loss 0.279154 LR 0.001000 Time 0.020780 -2022-12-06 10:35:41,377 - Epoch: [30][ 510/ 1200] Overall Loss 0.279366 Objective Loss 0.279366 LR 0.001000 Time 0.020755 -2022-12-06 10:35:41,573 - Epoch: [30][ 520/ 1200] Overall Loss 0.279128 Objective Loss 0.279128 LR 0.001000 Time 0.020732 -2022-12-06 10:35:41,768 - Epoch: [30][ 530/ 1200] Overall Loss 0.278778 Objective Loss 0.278778 LR 0.001000 Time 0.020709 -2022-12-06 10:35:41,964 - Epoch: [30][ 540/ 1200] Overall Loss 0.278388 Objective Loss 0.278388 LR 0.001000 Time 0.020687 -2022-12-06 10:35:42,160 - Epoch: [30][ 550/ 1200] Overall Loss 0.277874 Objective Loss 0.277874 LR 0.001000 Time 0.020665 -2022-12-06 10:35:42,355 - Epoch: [30][ 560/ 1200] Overall Loss 0.278466 Objective Loss 0.278466 LR 0.001000 Time 0.020644 -2022-12-06 10:35:42,551 - Epoch: [30][ 570/ 1200] Overall Loss 0.278703 Objective Loss 0.278703 LR 0.001000 Time 0.020625 -2022-12-06 10:35:42,747 - Epoch: [30][ 580/ 1200] Overall Loss 0.278844 Objective Loss 0.278844 LR 0.001000 Time 0.020606 -2022-12-06 10:35:42,942 - Epoch: [30][ 590/ 1200] Overall Loss 0.278836 Objective Loss 0.278836 LR 0.001000 Time 0.020587 -2022-12-06 10:35:43,138 - Epoch: [30][ 600/ 1200] Overall Loss 0.279003 Objective Loss 0.279003 LR 0.001000 Time 0.020569 -2022-12-06 10:35:43,334 - Epoch: [30][ 610/ 1200] Overall Loss 0.279427 Objective Loss 0.279427 LR 0.001000 Time 0.020552 -2022-12-06 10:35:43,530 - Epoch: [30][ 620/ 1200] Overall Loss 0.279341 Objective Loss 0.279341 LR 0.001000 Time 0.020536 -2022-12-06 10:35:43,725 - Epoch: [30][ 630/ 1200] Overall Loss 0.279316 Objective Loss 0.279316 LR 0.001000 Time 0.020519 -2022-12-06 10:35:43,920 - Epoch: [30][ 640/ 1200] Overall Loss 0.279150 Objective Loss 0.279150 LR 0.001000 Time 0.020503 -2022-12-06 10:35:44,116 - Epoch: [30][ 650/ 1200] Overall Loss 0.279277 Objective Loss 0.279277 LR 0.001000 Time 0.020487 -2022-12-06 10:35:44,312 - Epoch: [30][ 660/ 1200] Overall Loss 0.280193 Objective Loss 0.280193 LR 0.001000 Time 0.020473 -2022-12-06 10:35:44,508 - Epoch: [30][ 670/ 1200] Overall Loss 0.280174 Objective Loss 0.280174 LR 0.001000 Time 0.020459 -2022-12-06 10:35:44,704 - Epoch: [30][ 680/ 1200] Overall Loss 0.280477 Objective Loss 0.280477 LR 0.001000 Time 0.020446 -2022-12-06 10:35:44,900 - Epoch: [30][ 690/ 1200] Overall Loss 0.279893 Objective Loss 0.279893 LR 0.001000 Time 0.020433 -2022-12-06 10:35:45,096 - Epoch: [30][ 700/ 1200] Overall Loss 0.279552 Objective Loss 0.279552 LR 0.001000 Time 0.020420 -2022-12-06 10:35:45,292 - Epoch: [30][ 710/ 1200] Overall Loss 0.279767 Objective Loss 0.279767 LR 0.001000 Time 0.020408 -2022-12-06 10:35:45,488 - Epoch: [30][ 720/ 1200] Overall Loss 0.280332 Objective Loss 0.280332 LR 0.001000 Time 0.020396 -2022-12-06 10:35:45,684 - Epoch: [30][ 730/ 1200] Overall Loss 0.280742 Objective Loss 0.280742 LR 0.001000 Time 0.020384 -2022-12-06 10:35:45,880 - Epoch: [30][ 740/ 1200] Overall Loss 0.281021 Objective Loss 0.281021 LR 0.001000 Time 0.020373 -2022-12-06 10:35:46,076 - Epoch: [30][ 750/ 1200] Overall Loss 0.280884 Objective Loss 0.280884 LR 0.001000 Time 0.020362 -2022-12-06 10:35:46,272 - Epoch: [30][ 760/ 1200] Overall Loss 0.281148 Objective Loss 0.281148 LR 0.001000 Time 0.020351 -2022-12-06 10:35:46,468 - Epoch: [30][ 770/ 1200] Overall Loss 0.281264 Objective Loss 0.281264 LR 0.001000 Time 0.020340 -2022-12-06 10:35:46,663 - Epoch: [30][ 780/ 1200] Overall Loss 0.281580 Objective Loss 0.281580 LR 0.001000 Time 0.020330 -2022-12-06 10:35:46,859 - Epoch: [30][ 790/ 1200] Overall Loss 0.281773 Objective Loss 0.281773 LR 0.001000 Time 0.020319 -2022-12-06 10:35:47,054 - Epoch: [30][ 800/ 1200] Overall Loss 0.281813 Objective Loss 0.281813 LR 0.001000 Time 0.020308 -2022-12-06 10:35:47,249 - Epoch: [30][ 810/ 1200] Overall Loss 0.281682 Objective Loss 0.281682 LR 0.001000 Time 0.020298 -2022-12-06 10:35:47,446 - Epoch: [30][ 820/ 1200] Overall Loss 0.281949 Objective Loss 0.281949 LR 0.001000 Time 0.020289 -2022-12-06 10:35:47,642 - Epoch: [30][ 830/ 1200] Overall Loss 0.282262 Objective Loss 0.282262 LR 0.001000 Time 0.020280 -2022-12-06 10:35:47,838 - Epoch: [30][ 840/ 1200] Overall Loss 0.282233 Objective Loss 0.282233 LR 0.001000 Time 0.020272 -2022-12-06 10:35:48,034 - Epoch: [30][ 850/ 1200] Overall Loss 0.282285 Objective Loss 0.282285 LR 0.001000 Time 0.020263 -2022-12-06 10:35:48,230 - Epoch: [30][ 860/ 1200] Overall Loss 0.282588 Objective Loss 0.282588 LR 0.001000 Time 0.020255 -2022-12-06 10:35:48,426 - Epoch: [30][ 870/ 1200] Overall Loss 0.283299 Objective Loss 0.283299 LR 0.001000 Time 0.020246 -2022-12-06 10:35:48,622 - Epoch: [30][ 880/ 1200] Overall Loss 0.283433 Objective Loss 0.283433 LR 0.001000 Time 0.020239 -2022-12-06 10:35:48,818 - Epoch: [30][ 890/ 1200] Overall Loss 0.283586 Objective Loss 0.283586 LR 0.001000 Time 0.020231 -2022-12-06 10:35:49,014 - Epoch: [30][ 900/ 1200] Overall Loss 0.283758 Objective Loss 0.283758 LR 0.001000 Time 0.020223 -2022-12-06 10:35:49,210 - Epoch: [30][ 910/ 1200] Overall Loss 0.283941 Objective Loss 0.283941 LR 0.001000 Time 0.020216 -2022-12-06 10:35:49,406 - Epoch: [30][ 920/ 1200] Overall Loss 0.284159 Objective Loss 0.284159 LR 0.001000 Time 0.020208 -2022-12-06 10:35:49,602 - Epoch: [30][ 930/ 1200] Overall Loss 0.284394 Objective Loss 0.284394 LR 0.001000 Time 0.020201 -2022-12-06 10:35:49,798 - Epoch: [30][ 940/ 1200] Overall Loss 0.284402 Objective Loss 0.284402 LR 0.001000 Time 0.020194 -2022-12-06 10:35:49,993 - Epoch: [30][ 950/ 1200] Overall Loss 0.284310 Objective Loss 0.284310 LR 0.001000 Time 0.020187 -2022-12-06 10:35:50,189 - Epoch: [30][ 960/ 1200] Overall Loss 0.284407 Objective Loss 0.284407 LR 0.001000 Time 0.020180 -2022-12-06 10:35:50,385 - Epoch: [30][ 970/ 1200] Overall Loss 0.284880 Objective Loss 0.284880 LR 0.001000 Time 0.020173 -2022-12-06 10:35:50,581 - Epoch: [30][ 980/ 1200] Overall Loss 0.284595 Objective Loss 0.284595 LR 0.001000 Time 0.020166 -2022-12-06 10:35:50,777 - Epoch: [30][ 990/ 1200] Overall Loss 0.284461 Objective Loss 0.284461 LR 0.001000 Time 0.020160 -2022-12-06 10:35:50,972 - Epoch: [30][ 1000/ 1200] Overall Loss 0.284412 Objective Loss 0.284412 LR 0.001000 Time 0.020154 -2022-12-06 10:35:51,168 - Epoch: [30][ 1010/ 1200] Overall Loss 0.284441 Objective Loss 0.284441 LR 0.001000 Time 0.020148 -2022-12-06 10:35:51,365 - Epoch: [30][ 1020/ 1200] Overall Loss 0.284297 Objective Loss 0.284297 LR 0.001000 Time 0.020142 -2022-12-06 10:35:51,561 - Epoch: [30][ 1030/ 1200] Overall Loss 0.284405 Objective Loss 0.284405 LR 0.001000 Time 0.020136 -2022-12-06 10:35:51,757 - Epoch: [30][ 1040/ 1200] Overall Loss 0.284522 Objective Loss 0.284522 LR 0.001000 Time 0.020130 -2022-12-06 10:35:51,952 - Epoch: [30][ 1050/ 1200] Overall Loss 0.284551 Objective Loss 0.284551 LR 0.001000 Time 0.020124 -2022-12-06 10:35:52,148 - Epoch: [30][ 1060/ 1200] Overall Loss 0.284432 Objective Loss 0.284432 LR 0.001000 Time 0.020119 -2022-12-06 10:35:52,344 - Epoch: [30][ 1070/ 1200] Overall Loss 0.284564 Objective Loss 0.284564 LR 0.001000 Time 0.020113 -2022-12-06 10:35:52,536 - Epoch: [30][ 1080/ 1200] Overall Loss 0.284656 Objective Loss 0.284656 LR 0.001000 Time 0.020105 -2022-12-06 10:35:52,728 - Epoch: [30][ 1090/ 1200] Overall Loss 0.284700 Objective Loss 0.284700 LR 0.001000 Time 0.020096 -2022-12-06 10:35:52,920 - Epoch: [30][ 1100/ 1200] Overall Loss 0.284721 Objective Loss 0.284721 LR 0.001000 Time 0.020087 -2022-12-06 10:35:53,112 - Epoch: [30][ 1110/ 1200] Overall Loss 0.284457 Objective Loss 0.284457 LR 0.001000 Time 0.020079 -2022-12-06 10:35:53,305 - Epoch: [30][ 1120/ 1200] Overall Loss 0.284764 Objective Loss 0.284764 LR 0.001000 Time 0.020071 -2022-12-06 10:35:53,497 - Epoch: [30][ 1130/ 1200] Overall Loss 0.285033 Objective Loss 0.285033 LR 0.001000 Time 0.020063 -2022-12-06 10:35:53,689 - Epoch: [30][ 1140/ 1200] Overall Loss 0.285315 Objective Loss 0.285315 LR 0.001000 Time 0.020055 -2022-12-06 10:35:53,881 - Epoch: [30][ 1150/ 1200] Overall Loss 0.285565 Objective Loss 0.285565 LR 0.001000 Time 0.020047 -2022-12-06 10:35:54,073 - Epoch: [30][ 1160/ 1200] Overall Loss 0.285886 Objective Loss 0.285886 LR 0.001000 Time 0.020039 -2022-12-06 10:35:54,266 - Epoch: [30][ 1170/ 1200] Overall Loss 0.286254 Objective Loss 0.286254 LR 0.001000 Time 0.020032 -2022-12-06 10:35:54,458 - Epoch: [30][ 1180/ 1200] Overall Loss 0.286158 Objective Loss 0.286158 LR 0.001000 Time 0.020024 -2022-12-06 10:35:54,650 - Epoch: [30][ 1190/ 1200] Overall Loss 0.286217 Objective Loss 0.286217 LR 0.001000 Time 0.020017 -2022-12-06 10:35:54,876 - Epoch: [30][ 1200/ 1200] Overall Loss 0.286135 Objective Loss 0.286135 Top1 82.845188 Top5 97.907950 LR 0.001000 Time 0.020039 -2022-12-06 10:35:54,964 - --- validate (epoch=30)----------- -2022-12-06 10:35:54,965 - 34129 samples (256 per mini-batch) -2022-12-06 10:35:55,411 - Epoch: [30][ 10/ 134] Loss 0.312101 Top1 84.140625 Top5 97.851562 -2022-12-06 10:35:55,542 - Epoch: [30][ 20/ 134] Loss 0.297810 Top1 84.550781 Top5 98.242188 -2022-12-06 10:35:55,671 - Epoch: [30][ 30/ 134] Loss 0.299079 Top1 84.414062 Top5 98.320312 -2022-12-06 10:35:55,803 - Epoch: [30][ 40/ 134] Loss 0.300242 Top1 84.570312 Top5 98.271484 -2022-12-06 10:35:55,933 - Epoch: [30][ 50/ 134] Loss 0.295568 Top1 84.671875 Top5 98.210938 -2022-12-06 10:35:56,064 - Epoch: [30][ 60/ 134] Loss 0.296906 Top1 84.687500 Top5 98.157552 -2022-12-06 10:35:56,201 - Epoch: [30][ 70/ 134] Loss 0.297940 Top1 84.715402 Top5 98.147321 -2022-12-06 10:35:56,348 - Epoch: [30][ 80/ 134] Loss 0.299694 Top1 84.624023 Top5 98.115234 -2022-12-06 10:35:56,486 - Epoch: [30][ 90/ 134] Loss 0.298661 Top1 84.635417 Top5 98.103299 -2022-12-06 10:35:56,618 - Epoch: [30][ 100/ 134] Loss 0.299804 Top1 84.554688 Top5 98.039062 -2022-12-06 10:35:56,752 - Epoch: [30][ 110/ 134] Loss 0.301976 Top1 84.577415 Top5 98.046875 -2022-12-06 10:35:56,885 - Epoch: [30][ 120/ 134] Loss 0.302087 Top1 84.635417 Top5 97.988281 -2022-12-06 10:35:57,019 - Epoch: [30][ 130/ 134] Loss 0.299864 Top1 84.702524 Top5 98.001803 -2022-12-06 10:35:57,058 - Epoch: [30][ 134/ 134] Loss 0.299809 Top1 84.702159 Top5 98.013420 -2022-12-06 10:35:57,146 - ==> Top1: 84.702 Top5: 98.013 Loss: 0.300 - -2022-12-06 10:35:57,147 - ==> Confusion: -[[ 855 5 1 3 2 3 1 3 7 86 0 3 3 3 8 2 2 0 2 1 6] - [ 0 917 5 2 9 19 3 19 2 1 4 2 1 3 2 1 5 2 16 2 12] - [ 2 5 1004 17 2 3 29 10 0 3 2 3 0 2 2 4 0 0 3 4 8] - [ 2 0 19 946 0 3 0 1 2 0 9 0 3 0 11 0 1 7 10 1 5] - [ 6 11 6 1 940 3 1 3 0 7 2 1 0 1 14 4 5 2 0 3 10] - [ 2 24 1 1 5 959 5 20 5 2 0 5 3 18 1 0 0 0 3 6 9] - [ 0 4 12 1 0 2 1073 2 0 0 2 3 0 0 1 6 2 0 1 6 3] - [ 0 13 7 2 1 28 3 925 1 0 2 5 2 1 0 2 0 1 46 11 4] - [ 4 5 0 1 0 1 0 0 973 37 6 2 2 11 13 0 1 0 5 0 3] - [ 39 2 2 0 2 0 0 4 40 890 1 1 1 8 3 0 0 0 0 0 8] - [ 0 3 5 20 0 1 2 3 6 2 934 1 1 15 6 0 1 0 11 2 6] - [ 2 5 4 0 0 18 2 7 1 1 1 923 43 10 0 2 3 7 1 14 7] - [ 3 3 1 4 1 2 0 3 0 0 0 34 877 1 1 9 1 11 1 6 11] - [ 0 2 2 0 0 5 0 2 11 16 6 1 5 953 2 1 1 0 2 1 13] - [ 9 5 2 23 2 0 0 1 16 4 4 1 4 7 1030 0 1 3 7 0 11] - [ 0 4 6 1 3 1 2 1 0 1 0 8 6 4 2 979 5 9 0 5 6] - [ 2 8 2 2 3 1 1 0 1 0 0 5 0 4 2 12 1002 0 2 8 17] - [ 2 2 1 5 0 2 1 2 0 3 0 11 25 2 1 17 2 956 0 1 3] - [ 3 4 9 15 3 2 0 17 2 1 4 2 1 1 7 0 0 2 926 1 8] - [ 1 4 4 0 1 10 4 5 0 0 1 6 8 5 0 3 2 1 2 1016 7] - [ 85 231 261 176 92 141 101 139 78 100 132 106 382 316 160 170 162 60 241 265 9828]] - -2022-12-06 10:35:57,727 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:35:57,728 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:35:57,735 - - -2022-12-06 10:35:57,735 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:35:58,757 - Epoch: [31][ 10/ 1200] Overall Loss 0.275813 Objective Loss 0.275813 LR 0.001000 Time 0.102159 -2022-12-06 10:35:58,952 - Epoch: [31][ 20/ 1200] Overall Loss 0.272177 Objective Loss 0.272177 LR 0.001000 Time 0.060800 -2022-12-06 10:35:59,146 - Epoch: [31][ 30/ 1200] Overall Loss 0.258992 Objective Loss 0.258992 LR 0.001000 Time 0.046981 -2022-12-06 10:35:59,339 - Epoch: [31][ 40/ 1200] Overall Loss 0.260427 Objective Loss 0.260427 LR 0.001000 Time 0.040053 -2022-12-06 10:35:59,533 - Epoch: [31][ 50/ 1200] Overall Loss 0.260218 Objective Loss 0.260218 LR 0.001000 Time 0.035904 -2022-12-06 10:35:59,727 - Epoch: [31][ 60/ 1200] Overall Loss 0.261111 Objective Loss 0.261111 LR 0.001000 Time 0.033136 -2022-12-06 10:35:59,921 - Epoch: [31][ 70/ 1200] Overall Loss 0.264108 Objective Loss 0.264108 LR 0.001000 Time 0.031171 -2022-12-06 10:36:00,114 - Epoch: [31][ 80/ 1200] Overall Loss 0.265811 Objective Loss 0.265811 LR 0.001000 Time 0.029678 -2022-12-06 10:36:00,308 - Epoch: [31][ 90/ 1200] Overall Loss 0.267597 Objective Loss 0.267597 LR 0.001000 Time 0.028530 -2022-12-06 10:36:00,501 - Epoch: [31][ 100/ 1200] Overall Loss 0.267071 Objective Loss 0.267071 LR 0.001000 Time 0.027601 -2022-12-06 10:36:00,695 - Epoch: [31][ 110/ 1200] Overall Loss 0.269047 Objective Loss 0.269047 LR 0.001000 Time 0.026848 -2022-12-06 10:36:00,888 - Epoch: [31][ 120/ 1200] Overall Loss 0.270066 Objective Loss 0.270066 LR 0.001000 Time 0.026217 -2022-12-06 10:36:01,081 - Epoch: [31][ 130/ 1200] Overall Loss 0.273279 Objective Loss 0.273279 LR 0.001000 Time 0.025684 -2022-12-06 10:36:01,274 - Epoch: [31][ 140/ 1200] Overall Loss 0.274102 Objective Loss 0.274102 LR 0.001000 Time 0.025222 -2022-12-06 10:36:01,468 - Epoch: [31][ 150/ 1200] Overall Loss 0.274464 Objective Loss 0.274464 LR 0.001000 Time 0.024832 -2022-12-06 10:36:01,661 - Epoch: [31][ 160/ 1200] Overall Loss 0.275039 Objective Loss 0.275039 LR 0.001000 Time 0.024484 -2022-12-06 10:36:01,854 - Epoch: [31][ 170/ 1200] Overall Loss 0.276265 Objective Loss 0.276265 LR 0.001000 Time 0.024174 -2022-12-06 10:36:02,048 - Epoch: [31][ 180/ 1200] Overall Loss 0.277338 Objective Loss 0.277338 LR 0.001000 Time 0.023904 -2022-12-06 10:36:02,241 - Epoch: [31][ 190/ 1200] Overall Loss 0.278239 Objective Loss 0.278239 LR 0.001000 Time 0.023661 -2022-12-06 10:36:02,435 - Epoch: [31][ 200/ 1200] Overall Loss 0.278276 Objective Loss 0.278276 LR 0.001000 Time 0.023442 -2022-12-06 10:36:02,628 - Epoch: [31][ 210/ 1200] Overall Loss 0.279213 Objective Loss 0.279213 LR 0.001000 Time 0.023246 -2022-12-06 10:36:02,822 - Epoch: [31][ 220/ 1200] Overall Loss 0.279161 Objective Loss 0.279161 LR 0.001000 Time 0.023064 -2022-12-06 10:36:03,015 - Epoch: [31][ 230/ 1200] Overall Loss 0.280032 Objective Loss 0.280032 LR 0.001000 Time 0.022902 -2022-12-06 10:36:03,209 - Epoch: [31][ 240/ 1200] Overall Loss 0.280080 Objective Loss 0.280080 LR 0.001000 Time 0.022751 -2022-12-06 10:36:03,403 - Epoch: [31][ 250/ 1200] Overall Loss 0.279185 Objective Loss 0.279185 LR 0.001000 Time 0.022614 -2022-12-06 10:36:03,596 - Epoch: [31][ 260/ 1200] Overall Loss 0.279696 Objective Loss 0.279696 LR 0.001000 Time 0.022485 -2022-12-06 10:36:03,790 - Epoch: [31][ 270/ 1200] Overall Loss 0.279905 Objective Loss 0.279905 LR 0.001000 Time 0.022368 -2022-12-06 10:36:03,984 - Epoch: [31][ 280/ 1200] Overall Loss 0.280597 Objective Loss 0.280597 LR 0.001000 Time 0.022260 -2022-12-06 10:36:04,177 - Epoch: [31][ 290/ 1200] Overall Loss 0.280093 Objective Loss 0.280093 LR 0.001000 Time 0.022157 -2022-12-06 10:36:04,370 - Epoch: [31][ 300/ 1200] Overall Loss 0.281328 Objective Loss 0.281328 LR 0.001000 Time 0.022060 -2022-12-06 10:36:04,564 - Epoch: [31][ 310/ 1200] Overall Loss 0.281528 Objective Loss 0.281528 LR 0.001000 Time 0.021973 -2022-12-06 10:36:04,758 - Epoch: [31][ 320/ 1200] Overall Loss 0.282175 Objective Loss 0.282175 LR 0.001000 Time 0.021890 -2022-12-06 10:36:04,952 - Epoch: [31][ 330/ 1200] Overall Loss 0.282307 Objective Loss 0.282307 LR 0.001000 Time 0.021812 -2022-12-06 10:36:05,146 - Epoch: [31][ 340/ 1200] Overall Loss 0.281782 Objective Loss 0.281782 LR 0.001000 Time 0.021739 -2022-12-06 10:36:05,339 - Epoch: [31][ 350/ 1200] Overall Loss 0.282879 Objective Loss 0.282879 LR 0.001000 Time 0.021670 -2022-12-06 10:36:05,533 - Epoch: [31][ 360/ 1200] Overall Loss 0.283148 Objective Loss 0.283148 LR 0.001000 Time 0.021605 -2022-12-06 10:36:05,727 - Epoch: [31][ 370/ 1200] Overall Loss 0.283122 Objective Loss 0.283122 LR 0.001000 Time 0.021544 -2022-12-06 10:36:05,921 - Epoch: [31][ 380/ 1200] Overall Loss 0.282629 Objective Loss 0.282629 LR 0.001000 Time 0.021486 -2022-12-06 10:36:06,115 - Epoch: [31][ 390/ 1200] Overall Loss 0.282924 Objective Loss 0.282924 LR 0.001000 Time 0.021430 -2022-12-06 10:36:06,309 - Epoch: [31][ 400/ 1200] Overall Loss 0.283594 Objective Loss 0.283594 LR 0.001000 Time 0.021377 -2022-12-06 10:36:06,503 - Epoch: [31][ 410/ 1200] Overall Loss 0.283168 Objective Loss 0.283168 LR 0.001000 Time 0.021328 -2022-12-06 10:36:06,696 - Epoch: [31][ 420/ 1200] Overall Loss 0.283256 Objective Loss 0.283256 LR 0.001000 Time 0.021279 -2022-12-06 10:36:06,890 - Epoch: [31][ 430/ 1200] Overall Loss 0.283023 Objective Loss 0.283023 LR 0.001000 Time 0.021233 -2022-12-06 10:36:07,084 - Epoch: [31][ 440/ 1200] Overall Loss 0.282426 Objective Loss 0.282426 LR 0.001000 Time 0.021190 -2022-12-06 10:36:07,277 - Epoch: [31][ 450/ 1200] Overall Loss 0.282857 Objective Loss 0.282857 LR 0.001000 Time 0.021148 -2022-12-06 10:36:07,471 - Epoch: [31][ 460/ 1200] Overall Loss 0.283060 Objective Loss 0.283060 LR 0.001000 Time 0.021108 -2022-12-06 10:36:07,665 - Epoch: [31][ 470/ 1200] Overall Loss 0.283036 Objective Loss 0.283036 LR 0.001000 Time 0.021071 -2022-12-06 10:36:07,859 - Epoch: [31][ 480/ 1200] Overall Loss 0.283017 Objective Loss 0.283017 LR 0.001000 Time 0.021034 -2022-12-06 10:36:08,052 - Epoch: [31][ 490/ 1200] Overall Loss 0.283427 Objective Loss 0.283427 LR 0.001000 Time 0.020999 -2022-12-06 10:36:08,245 - Epoch: [31][ 500/ 1200] Overall Loss 0.283833 Objective Loss 0.283833 LR 0.001000 Time 0.020963 -2022-12-06 10:36:08,438 - Epoch: [31][ 510/ 1200] Overall Loss 0.284397 Objective Loss 0.284397 LR 0.001000 Time 0.020930 -2022-12-06 10:36:08,632 - Epoch: [31][ 520/ 1200] Overall Loss 0.284276 Objective Loss 0.284276 LR 0.001000 Time 0.020899 -2022-12-06 10:36:08,826 - Epoch: [31][ 530/ 1200] Overall Loss 0.284167 Objective Loss 0.284167 LR 0.001000 Time 0.020870 -2022-12-06 10:36:09,019 - Epoch: [31][ 540/ 1200] Overall Loss 0.284162 Objective Loss 0.284162 LR 0.001000 Time 0.020839 -2022-12-06 10:36:09,212 - Epoch: [31][ 550/ 1200] Overall Loss 0.285174 Objective Loss 0.285174 LR 0.001000 Time 0.020811 -2022-12-06 10:36:09,406 - Epoch: [31][ 560/ 1200] Overall Loss 0.284998 Objective Loss 0.284998 LR 0.001000 Time 0.020784 -2022-12-06 10:36:09,600 - Epoch: [31][ 570/ 1200] Overall Loss 0.284860 Objective Loss 0.284860 LR 0.001000 Time 0.020759 -2022-12-06 10:36:09,794 - Epoch: [31][ 580/ 1200] Overall Loss 0.285008 Objective Loss 0.285008 LR 0.001000 Time 0.020734 -2022-12-06 10:36:09,988 - Epoch: [31][ 590/ 1200] Overall Loss 0.285134 Objective Loss 0.285134 LR 0.001000 Time 0.020711 -2022-12-06 10:36:10,181 - Epoch: [31][ 600/ 1200] Overall Loss 0.285029 Objective Loss 0.285029 LR 0.001000 Time 0.020687 -2022-12-06 10:36:10,375 - Epoch: [31][ 610/ 1200] Overall Loss 0.285160 Objective Loss 0.285160 LR 0.001000 Time 0.020665 -2022-12-06 10:36:10,569 - Epoch: [31][ 620/ 1200] Overall Loss 0.285173 Objective Loss 0.285173 LR 0.001000 Time 0.020643 -2022-12-06 10:36:10,762 - Epoch: [31][ 630/ 1200] Overall Loss 0.285470 Objective Loss 0.285470 LR 0.001000 Time 0.020622 -2022-12-06 10:36:10,956 - Epoch: [31][ 640/ 1200] Overall Loss 0.285630 Objective Loss 0.285630 LR 0.001000 Time 0.020601 -2022-12-06 10:36:11,150 - Epoch: [31][ 650/ 1200] Overall Loss 0.285803 Objective Loss 0.285803 LR 0.001000 Time 0.020582 -2022-12-06 10:36:11,344 - Epoch: [31][ 660/ 1200] Overall Loss 0.285436 Objective Loss 0.285436 LR 0.001000 Time 0.020562 -2022-12-06 10:36:11,537 - Epoch: [31][ 670/ 1200] Overall Loss 0.285517 Objective Loss 0.285517 LR 0.001000 Time 0.020543 -2022-12-06 10:36:11,730 - Epoch: [31][ 680/ 1200] Overall Loss 0.285503 Objective Loss 0.285503 LR 0.001000 Time 0.020524 -2022-12-06 10:36:11,924 - Epoch: [31][ 690/ 1200] Overall Loss 0.285692 Objective Loss 0.285692 LR 0.001000 Time 0.020508 -2022-12-06 10:36:12,118 - Epoch: [31][ 700/ 1200] Overall Loss 0.285904 Objective Loss 0.285904 LR 0.001000 Time 0.020490 -2022-12-06 10:36:12,312 - Epoch: [31][ 710/ 1200] Overall Loss 0.285492 Objective Loss 0.285492 LR 0.001000 Time 0.020474 -2022-12-06 10:36:12,505 - Epoch: [31][ 720/ 1200] Overall Loss 0.285360 Objective Loss 0.285360 LR 0.001000 Time 0.020457 -2022-12-06 10:36:12,699 - Epoch: [31][ 730/ 1200] Overall Loss 0.285470 Objective Loss 0.285470 LR 0.001000 Time 0.020442 -2022-12-06 10:36:12,892 - Epoch: [31][ 740/ 1200] Overall Loss 0.285634 Objective Loss 0.285634 LR 0.001000 Time 0.020426 -2022-12-06 10:36:13,086 - Epoch: [31][ 750/ 1200] Overall Loss 0.285587 Objective Loss 0.285587 LR 0.001000 Time 0.020411 -2022-12-06 10:36:13,279 - Epoch: [31][ 760/ 1200] Overall Loss 0.285687 Objective Loss 0.285687 LR 0.001000 Time 0.020396 -2022-12-06 10:36:13,474 - Epoch: [31][ 770/ 1200] Overall Loss 0.285618 Objective Loss 0.285618 LR 0.001000 Time 0.020383 -2022-12-06 10:36:13,667 - Epoch: [31][ 780/ 1200] Overall Loss 0.285120 Objective Loss 0.285120 LR 0.001000 Time 0.020369 -2022-12-06 10:36:13,860 - Epoch: [31][ 790/ 1200] Overall Loss 0.285516 Objective Loss 0.285516 LR 0.001000 Time 0.020354 -2022-12-06 10:36:14,053 - Epoch: [31][ 800/ 1200] Overall Loss 0.285653 Objective Loss 0.285653 LR 0.001000 Time 0.020341 -2022-12-06 10:36:14,247 - Epoch: [31][ 810/ 1200] Overall Loss 0.285317 Objective Loss 0.285317 LR 0.001000 Time 0.020329 -2022-12-06 10:36:14,441 - Epoch: [31][ 820/ 1200] Overall Loss 0.285035 Objective Loss 0.285035 LR 0.001000 Time 0.020316 -2022-12-06 10:36:14,635 - Epoch: [31][ 830/ 1200] Overall Loss 0.285226 Objective Loss 0.285226 LR 0.001000 Time 0.020305 -2022-12-06 10:36:14,828 - Epoch: [31][ 840/ 1200] Overall Loss 0.285595 Objective Loss 0.285595 LR 0.001000 Time 0.020292 -2022-12-06 10:36:15,023 - Epoch: [31][ 850/ 1200] Overall Loss 0.285779 Objective Loss 0.285779 LR 0.001000 Time 0.020282 -2022-12-06 10:36:15,217 - Epoch: [31][ 860/ 1200] Overall Loss 0.285952 Objective Loss 0.285952 LR 0.001000 Time 0.020271 -2022-12-06 10:36:15,411 - Epoch: [31][ 870/ 1200] Overall Loss 0.286593 Objective Loss 0.286593 LR 0.001000 Time 0.020260 -2022-12-06 10:36:15,604 - Epoch: [31][ 880/ 1200] Overall Loss 0.286690 Objective Loss 0.286690 LR 0.001000 Time 0.020249 -2022-12-06 10:36:15,799 - Epoch: [31][ 890/ 1200] Overall Loss 0.286871 Objective Loss 0.286871 LR 0.001000 Time 0.020240 -2022-12-06 10:36:15,992 - Epoch: [31][ 900/ 1200] Overall Loss 0.286850 Objective Loss 0.286850 LR 0.001000 Time 0.020228 -2022-12-06 10:36:16,186 - Epoch: [31][ 910/ 1200] Overall Loss 0.287077 Objective Loss 0.287077 LR 0.001000 Time 0.020219 -2022-12-06 10:36:16,379 - Epoch: [31][ 920/ 1200] Overall Loss 0.287097 Objective Loss 0.287097 LR 0.001000 Time 0.020209 -2022-12-06 10:36:16,573 - Epoch: [31][ 930/ 1200] Overall Loss 0.287107 Objective Loss 0.287107 LR 0.001000 Time 0.020199 -2022-12-06 10:36:16,767 - Epoch: [31][ 940/ 1200] Overall Loss 0.287229 Objective Loss 0.287229 LR 0.001000 Time 0.020190 -2022-12-06 10:36:16,961 - Epoch: [31][ 950/ 1200] Overall Loss 0.287171 Objective Loss 0.287171 LR 0.001000 Time 0.020181 -2022-12-06 10:36:17,154 - Epoch: [31][ 960/ 1200] Overall Loss 0.287616 Objective Loss 0.287616 LR 0.001000 Time 0.020172 -2022-12-06 10:36:17,348 - Epoch: [31][ 970/ 1200] Overall Loss 0.287692 Objective Loss 0.287692 LR 0.001000 Time 0.020163 -2022-12-06 10:36:17,542 - Epoch: [31][ 980/ 1200] Overall Loss 0.288137 Objective Loss 0.288137 LR 0.001000 Time 0.020154 -2022-12-06 10:36:17,736 - Epoch: [31][ 990/ 1200] Overall Loss 0.287999 Objective Loss 0.287999 LR 0.001000 Time 0.020147 -2022-12-06 10:36:17,930 - Epoch: [31][ 1000/ 1200] Overall Loss 0.288075 Objective Loss 0.288075 LR 0.001000 Time 0.020138 -2022-12-06 10:36:18,124 - Epoch: [31][ 1010/ 1200] Overall Loss 0.288209 Objective Loss 0.288209 LR 0.001000 Time 0.020130 -2022-12-06 10:36:18,317 - Epoch: [31][ 1020/ 1200] Overall Loss 0.288213 Objective Loss 0.288213 LR 0.001000 Time 0.020122 -2022-12-06 10:36:18,511 - Epoch: [31][ 1030/ 1200] Overall Loss 0.288276 Objective Loss 0.288276 LR 0.001000 Time 0.020115 -2022-12-06 10:36:18,705 - Epoch: [31][ 1040/ 1200] Overall Loss 0.288121 Objective Loss 0.288121 LR 0.001000 Time 0.020107 -2022-12-06 10:36:18,899 - Epoch: [31][ 1050/ 1200] Overall Loss 0.288368 Objective Loss 0.288368 LR 0.001000 Time 0.020099 -2022-12-06 10:36:19,092 - Epoch: [31][ 1060/ 1200] Overall Loss 0.288494 Objective Loss 0.288494 LR 0.001000 Time 0.020092 -2022-12-06 10:36:19,287 - Epoch: [31][ 1070/ 1200] Overall Loss 0.288579 Objective Loss 0.288579 LR 0.001000 Time 0.020085 -2022-12-06 10:36:19,480 - Epoch: [31][ 1080/ 1200] Overall Loss 0.288710 Objective Loss 0.288710 LR 0.001000 Time 0.020078 -2022-12-06 10:36:19,674 - Epoch: [31][ 1090/ 1200] Overall Loss 0.288853 Objective Loss 0.288853 LR 0.001000 Time 0.020071 -2022-12-06 10:36:19,868 - Epoch: [31][ 1100/ 1200] Overall Loss 0.288971 Objective Loss 0.288971 LR 0.001000 Time 0.020064 -2022-12-06 10:36:20,061 - Epoch: [31][ 1110/ 1200] Overall Loss 0.288905 Objective Loss 0.288905 LR 0.001000 Time 0.020057 -2022-12-06 10:36:20,254 - Epoch: [31][ 1120/ 1200] Overall Loss 0.288806 Objective Loss 0.288806 LR 0.001000 Time 0.020050 -2022-12-06 10:36:20,448 - Epoch: [31][ 1130/ 1200] Overall Loss 0.288647 Objective Loss 0.288647 LR 0.001000 Time 0.020043 -2022-12-06 10:36:20,641 - Epoch: [31][ 1140/ 1200] Overall Loss 0.288472 Objective Loss 0.288472 LR 0.001000 Time 0.020037 -2022-12-06 10:36:20,835 - Epoch: [31][ 1150/ 1200] Overall Loss 0.288526 Objective Loss 0.288526 LR 0.001000 Time 0.020031 -2022-12-06 10:36:21,029 - Epoch: [31][ 1160/ 1200] Overall Loss 0.288733 Objective Loss 0.288733 LR 0.001000 Time 0.020025 -2022-12-06 10:36:21,224 - Epoch: [31][ 1170/ 1200] Overall Loss 0.288824 Objective Loss 0.288824 LR 0.001000 Time 0.020020 -2022-12-06 10:36:21,417 - Epoch: [31][ 1180/ 1200] Overall Loss 0.289264 Objective Loss 0.289264 LR 0.001000 Time 0.020013 -2022-12-06 10:36:21,611 - Epoch: [31][ 1190/ 1200] Overall Loss 0.289370 Objective Loss 0.289370 LR 0.001000 Time 0.020007 -2022-12-06 10:36:21,845 - Epoch: [31][ 1200/ 1200] Overall Loss 0.289359 Objective Loss 0.289359 Top1 86.610879 Top5 98.117155 LR 0.001000 Time 0.020035 -2022-12-06 10:36:21,933 - --- validate (epoch=31)----------- -2022-12-06 10:36:21,934 - 34129 samples (256 per mini-batch) -2022-12-06 10:36:22,385 - Epoch: [31][ 10/ 134] Loss 0.302958 Top1 84.687500 Top5 97.890625 -2022-12-06 10:36:22,517 - Epoch: [31][ 20/ 134] Loss 0.293867 Top1 84.882812 Top5 97.871094 -2022-12-06 10:36:22,648 - Epoch: [31][ 30/ 134] Loss 0.307751 Top1 84.427083 Top5 97.838542 -2022-12-06 10:36:22,777 - Epoch: [31][ 40/ 134] Loss 0.307560 Top1 84.365234 Top5 97.851562 -2022-12-06 10:36:22,908 - Epoch: [31][ 50/ 134] Loss 0.306213 Top1 84.406250 Top5 97.851562 -2022-12-06 10:36:23,037 - Epoch: [31][ 60/ 134] Loss 0.304458 Top1 84.388021 Top5 97.877604 -2022-12-06 10:36:23,167 - Epoch: [31][ 70/ 134] Loss 0.306291 Top1 84.335938 Top5 97.907366 -2022-12-06 10:36:23,297 - Epoch: [31][ 80/ 134] Loss 0.302747 Top1 84.379883 Top5 97.929688 -2022-12-06 10:36:23,427 - Epoch: [31][ 90/ 134] Loss 0.303533 Top1 84.448785 Top5 97.907986 -2022-12-06 10:36:23,559 - Epoch: [31][ 100/ 134] Loss 0.299237 Top1 84.609375 Top5 97.953125 -2022-12-06 10:36:23,687 - Epoch: [31][ 110/ 134] Loss 0.299916 Top1 84.683949 Top5 97.926136 -2022-12-06 10:36:23,818 - Epoch: [31][ 120/ 134] Loss 0.300358 Top1 84.677734 Top5 97.939453 -2022-12-06 10:36:23,950 - Epoch: [31][ 130/ 134] Loss 0.300434 Top1 84.618389 Top5 97.932692 -2022-12-06 10:36:23,988 - Epoch: [31][ 134/ 134] Loss 0.301603 Top1 84.564447 Top5 97.922588 -2022-12-06 10:36:24,076 - ==> Top1: 84.564 Top5: 97.923 Loss: 0.302 - -2022-12-06 10:36:24,077 - ==> Confusion: -[[ 865 2 1 1 6 5 0 1 7 80 0 2 1 2 6 1 3 3 2 1 7] - [ 1 928 1 1 13 17 4 9 1 1 2 5 2 7 1 1 11 2 8 5 7] - [ 6 5 967 25 2 1 28 5 0 0 7 6 3 10 2 5 1 2 4 5 19] - [ 2 2 14 905 0 5 2 0 0 1 16 0 10 4 22 0 5 5 17 0 10] - [ 9 5 1 1 949 5 0 0 1 10 2 3 0 3 10 8 10 1 1 0 1] - [ 2 20 0 1 12 940 2 17 3 4 1 15 2 21 3 2 5 1 2 6 10] - [ 0 3 6 2 0 6 1065 1 0 0 1 1 1 2 0 12 2 3 1 9 3] - [ 1 10 12 2 1 43 9 915 0 0 2 5 0 0 0 1 2 3 26 17 5] - [ 5 1 1 2 1 2 0 1 961 58 3 1 2 8 10 0 3 1 0 2 2] - [ 44 0 2 1 4 2 0 0 17 914 0 2 0 8 0 1 1 0 0 1 4] - [ 0 7 5 3 1 2 2 3 14 4 929 3 2 23 7 0 3 1 5 1 4] - [ 2 4 1 0 0 5 3 1 0 0 0 975 15 11 1 8 0 8 1 11 5] - [ 0 3 0 1 1 6 1 0 1 1 0 45 866 4 0 10 1 11 0 4 14] - [ 0 2 0 0 1 3 0 0 12 18 3 7 4 950 2 1 1 3 1 4 11] - [ 8 4 2 7 4 2 0 0 22 5 1 2 2 3 1049 1 4 0 3 0 11] - [ 2 4 2 0 1 2 3 1 0 0 1 7 5 5 0 977 10 12 0 2 9] - [ 3 4 2 0 0 0 0 0 0 1 0 2 2 2 1 11 1028 6 0 6 4] - [ 1 2 0 3 0 1 1 0 0 2 0 13 19 1 2 13 2 972 1 0 3] - [ 1 4 6 7 1 3 1 20 5 0 8 3 4 3 7 1 1 3 924 3 3] - [ 1 2 4 0 0 6 6 3 0 0 0 18 5 6 1 4 4 6 0 1010 4] - [ 89 291 153 91 145 150 68 124 79 144 164 107 368 356 151 145 278 80 180 297 9766]] - -2022-12-06 10:36:24,645 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:36:24,645 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:36:24,651 - - -2022-12-06 10:36:24,651 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:36:25,684 - Epoch: [32][ 10/ 1200] Overall Loss 0.285665 Objective Loss 0.285665 LR 0.001000 Time 0.103258 -2022-12-06 10:36:25,888 - Epoch: [32][ 20/ 1200] Overall Loss 0.276611 Objective Loss 0.276611 LR 0.001000 Time 0.061781 -2022-12-06 10:36:26,080 - Epoch: [32][ 30/ 1200] Overall Loss 0.272412 Objective Loss 0.272412 LR 0.001000 Time 0.047568 -2022-12-06 10:36:26,271 - Epoch: [32][ 40/ 1200] Overall Loss 0.276239 Objective Loss 0.276239 LR 0.001000 Time 0.040457 -2022-12-06 10:36:26,463 - Epoch: [32][ 50/ 1200] Overall Loss 0.275319 Objective Loss 0.275319 LR 0.001000 Time 0.036185 -2022-12-06 10:36:26,655 - Epoch: [32][ 60/ 1200] Overall Loss 0.275574 Objective Loss 0.275574 LR 0.001000 Time 0.033353 -2022-12-06 10:36:26,847 - Epoch: [32][ 70/ 1200] Overall Loss 0.275093 Objective Loss 0.275093 LR 0.001000 Time 0.031324 -2022-12-06 10:36:27,039 - Epoch: [32][ 80/ 1200] Overall Loss 0.273975 Objective Loss 0.273975 LR 0.001000 Time 0.029799 -2022-12-06 10:36:27,232 - Epoch: [32][ 90/ 1200] Overall Loss 0.273260 Objective Loss 0.273260 LR 0.001000 Time 0.028619 -2022-12-06 10:36:27,423 - Epoch: [32][ 100/ 1200] Overall Loss 0.270139 Objective Loss 0.270139 LR 0.001000 Time 0.027668 -2022-12-06 10:36:27,616 - Epoch: [32][ 110/ 1200] Overall Loss 0.269705 Objective Loss 0.269705 LR 0.001000 Time 0.026898 -2022-12-06 10:36:27,810 - Epoch: [32][ 120/ 1200] Overall Loss 0.270313 Objective Loss 0.270313 LR 0.001000 Time 0.026267 -2022-12-06 10:36:28,002 - Epoch: [32][ 130/ 1200] Overall Loss 0.272301 Objective Loss 0.272301 LR 0.001000 Time 0.025722 -2022-12-06 10:36:28,195 - Epoch: [32][ 140/ 1200] Overall Loss 0.273567 Objective Loss 0.273567 LR 0.001000 Time 0.025256 -2022-12-06 10:36:28,387 - Epoch: [32][ 150/ 1200] Overall Loss 0.274514 Objective Loss 0.274514 LR 0.001000 Time 0.024852 -2022-12-06 10:36:28,579 - Epoch: [32][ 160/ 1200] Overall Loss 0.274989 Objective Loss 0.274989 LR 0.001000 Time 0.024494 -2022-12-06 10:36:28,768 - Epoch: [32][ 170/ 1200] Overall Loss 0.274492 Objective Loss 0.274492 LR 0.001000 Time 0.024163 -2022-12-06 10:36:28,958 - Epoch: [32][ 180/ 1200] Overall Loss 0.273802 Objective Loss 0.273802 LR 0.001000 Time 0.023875 -2022-12-06 10:36:29,148 - Epoch: [32][ 190/ 1200] Overall Loss 0.271835 Objective Loss 0.271835 LR 0.001000 Time 0.023615 -2022-12-06 10:36:29,338 - Epoch: [32][ 200/ 1200] Overall Loss 0.272684 Objective Loss 0.272684 LR 0.001000 Time 0.023382 -2022-12-06 10:36:29,528 - Epoch: [32][ 210/ 1200] Overall Loss 0.273006 Objective Loss 0.273006 LR 0.001000 Time 0.023168 -2022-12-06 10:36:29,718 - Epoch: [32][ 220/ 1200] Overall Loss 0.272905 Objective Loss 0.272905 LR 0.001000 Time 0.022977 -2022-12-06 10:36:29,908 - Epoch: [32][ 230/ 1200] Overall Loss 0.274058 Objective Loss 0.274058 LR 0.001000 Time 0.022799 -2022-12-06 10:36:30,098 - Epoch: [32][ 240/ 1200] Overall Loss 0.274145 Objective Loss 0.274145 LR 0.001000 Time 0.022639 -2022-12-06 10:36:30,287 - Epoch: [32][ 250/ 1200] Overall Loss 0.273183 Objective Loss 0.273183 LR 0.001000 Time 0.022491 -2022-12-06 10:36:30,478 - Epoch: [32][ 260/ 1200] Overall Loss 0.273057 Objective Loss 0.273057 LR 0.001000 Time 0.022354 -2022-12-06 10:36:30,667 - Epoch: [32][ 270/ 1200] Overall Loss 0.273842 Objective Loss 0.273842 LR 0.001000 Time 0.022226 -2022-12-06 10:36:30,857 - Epoch: [32][ 280/ 1200] Overall Loss 0.274381 Objective Loss 0.274381 LR 0.001000 Time 0.022108 -2022-12-06 10:36:31,046 - Epoch: [32][ 290/ 1200] Overall Loss 0.274963 Objective Loss 0.274963 LR 0.001000 Time 0.021998 -2022-12-06 10:36:31,237 - Epoch: [32][ 300/ 1200] Overall Loss 0.274818 Objective Loss 0.274818 LR 0.001000 Time 0.021897 -2022-12-06 10:36:31,427 - Epoch: [32][ 310/ 1200] Overall Loss 0.275007 Objective Loss 0.275007 LR 0.001000 Time 0.021801 -2022-12-06 10:36:31,616 - Epoch: [32][ 320/ 1200] Overall Loss 0.275059 Objective Loss 0.275059 LR 0.001000 Time 0.021711 -2022-12-06 10:36:31,806 - Epoch: [32][ 330/ 1200] Overall Loss 0.276077 Objective Loss 0.276077 LR 0.001000 Time 0.021627 -2022-12-06 10:36:31,996 - Epoch: [32][ 340/ 1200] Overall Loss 0.276341 Objective Loss 0.276341 LR 0.001000 Time 0.021547 -2022-12-06 10:36:32,186 - Epoch: [32][ 350/ 1200] Overall Loss 0.276481 Objective Loss 0.276481 LR 0.001000 Time 0.021472 -2022-12-06 10:36:32,375 - Epoch: [32][ 360/ 1200] Overall Loss 0.276565 Objective Loss 0.276565 LR 0.001000 Time 0.021401 -2022-12-06 10:36:32,565 - Epoch: [32][ 370/ 1200] Overall Loss 0.277416 Objective Loss 0.277416 LR 0.001000 Time 0.021335 -2022-12-06 10:36:32,755 - Epoch: [32][ 380/ 1200] Overall Loss 0.278468 Objective Loss 0.278468 LR 0.001000 Time 0.021272 -2022-12-06 10:36:32,945 - Epoch: [32][ 390/ 1200] Overall Loss 0.278745 Objective Loss 0.278745 LR 0.001000 Time 0.021211 -2022-12-06 10:36:33,135 - Epoch: [32][ 400/ 1200] Overall Loss 0.279318 Objective Loss 0.279318 LR 0.001000 Time 0.021155 -2022-12-06 10:36:33,325 - Epoch: [32][ 410/ 1200] Overall Loss 0.278922 Objective Loss 0.278922 LR 0.001000 Time 0.021102 -2022-12-06 10:36:33,516 - Epoch: [32][ 420/ 1200] Overall Loss 0.278969 Objective Loss 0.278969 LR 0.001000 Time 0.021052 -2022-12-06 10:36:33,706 - Epoch: [32][ 430/ 1200] Overall Loss 0.278810 Objective Loss 0.278810 LR 0.001000 Time 0.021003 -2022-12-06 10:36:33,897 - Epoch: [32][ 440/ 1200] Overall Loss 0.278400 Objective Loss 0.278400 LR 0.001000 Time 0.020957 -2022-12-06 10:36:34,086 - Epoch: [32][ 450/ 1200] Overall Loss 0.278439 Objective Loss 0.278439 LR 0.001000 Time 0.020912 -2022-12-06 10:36:34,276 - Epoch: [32][ 460/ 1200] Overall Loss 0.278813 Objective Loss 0.278813 LR 0.001000 Time 0.020870 -2022-12-06 10:36:34,466 - Epoch: [32][ 470/ 1200] Overall Loss 0.279258 Objective Loss 0.279258 LR 0.001000 Time 0.020828 -2022-12-06 10:36:34,656 - Epoch: [32][ 480/ 1200] Overall Loss 0.279327 Objective Loss 0.279327 LR 0.001000 Time 0.020789 -2022-12-06 10:36:34,846 - Epoch: [32][ 490/ 1200] Overall Loss 0.280484 Objective Loss 0.280484 LR 0.001000 Time 0.020751 -2022-12-06 10:36:35,036 - Epoch: [32][ 500/ 1200] Overall Loss 0.280615 Objective Loss 0.280615 LR 0.001000 Time 0.020715 -2022-12-06 10:36:35,226 - Epoch: [32][ 510/ 1200] Overall Loss 0.280678 Objective Loss 0.280678 LR 0.001000 Time 0.020681 -2022-12-06 10:36:35,416 - Epoch: [32][ 520/ 1200] Overall Loss 0.281023 Objective Loss 0.281023 LR 0.001000 Time 0.020646 -2022-12-06 10:36:35,606 - Epoch: [32][ 530/ 1200] Overall Loss 0.281337 Objective Loss 0.281337 LR 0.001000 Time 0.020615 -2022-12-06 10:36:35,796 - Epoch: [32][ 540/ 1200] Overall Loss 0.281390 Objective Loss 0.281390 LR 0.001000 Time 0.020584 -2022-12-06 10:36:35,986 - Epoch: [32][ 550/ 1200] Overall Loss 0.281329 Objective Loss 0.281329 LR 0.001000 Time 0.020554 -2022-12-06 10:36:36,177 - Epoch: [32][ 560/ 1200] Overall Loss 0.281343 Objective Loss 0.281343 LR 0.001000 Time 0.020526 -2022-12-06 10:36:36,366 - Epoch: [32][ 570/ 1200] Overall Loss 0.281213 Objective Loss 0.281213 LR 0.001000 Time 0.020498 -2022-12-06 10:36:36,556 - Epoch: [32][ 580/ 1200] Overall Loss 0.281325 Objective Loss 0.281325 LR 0.001000 Time 0.020471 -2022-12-06 10:36:36,746 - Epoch: [32][ 590/ 1200] Overall Loss 0.281070 Objective Loss 0.281070 LR 0.001000 Time 0.020445 -2022-12-06 10:36:36,937 - Epoch: [32][ 600/ 1200] Overall Loss 0.280721 Objective Loss 0.280721 LR 0.001000 Time 0.020420 -2022-12-06 10:36:37,126 - Epoch: [32][ 610/ 1200] Overall Loss 0.280336 Objective Loss 0.280336 LR 0.001000 Time 0.020396 -2022-12-06 10:36:37,317 - Epoch: [32][ 620/ 1200] Overall Loss 0.280289 Objective Loss 0.280289 LR 0.001000 Time 0.020373 -2022-12-06 10:36:37,507 - Epoch: [32][ 630/ 1200] Overall Loss 0.279802 Objective Loss 0.279802 LR 0.001000 Time 0.020351 -2022-12-06 10:36:37,697 - Epoch: [32][ 640/ 1200] Overall Loss 0.280162 Objective Loss 0.280162 LR 0.001000 Time 0.020329 -2022-12-06 10:36:37,886 - Epoch: [32][ 650/ 1200] Overall Loss 0.280452 Objective Loss 0.280452 LR 0.001000 Time 0.020306 -2022-12-06 10:36:38,077 - Epoch: [32][ 660/ 1200] Overall Loss 0.280644 Objective Loss 0.280644 LR 0.001000 Time 0.020286 -2022-12-06 10:36:38,266 - Epoch: [32][ 670/ 1200] Overall Loss 0.280670 Objective Loss 0.280670 LR 0.001000 Time 0.020265 -2022-12-06 10:36:38,456 - Epoch: [32][ 680/ 1200] Overall Loss 0.280360 Objective Loss 0.280360 LR 0.001000 Time 0.020246 -2022-12-06 10:36:38,645 - Epoch: [32][ 690/ 1200] Overall Loss 0.280254 Objective Loss 0.280254 LR 0.001000 Time 0.020226 -2022-12-06 10:36:38,836 - Epoch: [32][ 700/ 1200] Overall Loss 0.280462 Objective Loss 0.280462 LR 0.001000 Time 0.020208 -2022-12-06 10:36:39,025 - Epoch: [32][ 710/ 1200] Overall Loss 0.280400 Objective Loss 0.280400 LR 0.001000 Time 0.020190 -2022-12-06 10:36:39,215 - Epoch: [32][ 720/ 1200] Overall Loss 0.280276 Objective Loss 0.280276 LR 0.001000 Time 0.020173 -2022-12-06 10:36:39,405 - Epoch: [32][ 730/ 1200] Overall Loss 0.280415 Objective Loss 0.280415 LR 0.001000 Time 0.020155 -2022-12-06 10:36:39,594 - Epoch: [32][ 740/ 1200] Overall Loss 0.280696 Objective Loss 0.280696 LR 0.001000 Time 0.020138 -2022-12-06 10:36:39,784 - Epoch: [32][ 750/ 1200] Overall Loss 0.280869 Objective Loss 0.280869 LR 0.001000 Time 0.020122 -2022-12-06 10:36:39,974 - Epoch: [32][ 760/ 1200] Overall Loss 0.280992 Objective Loss 0.280992 LR 0.001000 Time 0.020106 -2022-12-06 10:36:40,164 - Epoch: [32][ 770/ 1200] Overall Loss 0.281044 Objective Loss 0.281044 LR 0.001000 Time 0.020091 -2022-12-06 10:36:40,354 - Epoch: [32][ 780/ 1200] Overall Loss 0.280472 Objective Loss 0.280472 LR 0.001000 Time 0.020076 -2022-12-06 10:36:40,544 - Epoch: [32][ 790/ 1200] Overall Loss 0.280645 Objective Loss 0.280645 LR 0.001000 Time 0.020062 -2022-12-06 10:36:40,734 - Epoch: [32][ 800/ 1200] Overall Loss 0.280600 Objective Loss 0.280600 LR 0.001000 Time 0.020048 -2022-12-06 10:36:40,924 - Epoch: [32][ 810/ 1200] Overall Loss 0.280450 Objective Loss 0.280450 LR 0.001000 Time 0.020035 -2022-12-06 10:36:41,114 - Epoch: [32][ 820/ 1200] Overall Loss 0.280468 Objective Loss 0.280468 LR 0.001000 Time 0.020022 -2022-12-06 10:36:41,304 - Epoch: [32][ 830/ 1200] Overall Loss 0.280993 Objective Loss 0.280993 LR 0.001000 Time 0.020009 -2022-12-06 10:36:41,495 - Epoch: [32][ 840/ 1200] Overall Loss 0.280931 Objective Loss 0.280931 LR 0.001000 Time 0.019997 -2022-12-06 10:36:41,684 - Epoch: [32][ 850/ 1200] Overall Loss 0.280749 Objective Loss 0.280749 LR 0.001000 Time 0.019984 -2022-12-06 10:36:41,874 - Epoch: [32][ 860/ 1200] Overall Loss 0.280801 Objective Loss 0.280801 LR 0.001000 Time 0.019972 -2022-12-06 10:36:42,064 - Epoch: [32][ 870/ 1200] Overall Loss 0.280675 Objective Loss 0.280675 LR 0.001000 Time 0.019959 -2022-12-06 10:36:42,254 - Epoch: [32][ 880/ 1200] Overall Loss 0.280623 Objective Loss 0.280623 LR 0.001000 Time 0.019948 -2022-12-06 10:36:42,444 - Epoch: [32][ 890/ 1200] Overall Loss 0.280541 Objective Loss 0.280541 LR 0.001000 Time 0.019937 -2022-12-06 10:36:42,634 - Epoch: [32][ 900/ 1200] Overall Loss 0.280215 Objective Loss 0.280215 LR 0.001000 Time 0.019926 -2022-12-06 10:36:42,824 - Epoch: [32][ 910/ 1200] Overall Loss 0.280604 Objective Loss 0.280604 LR 0.001000 Time 0.019915 -2022-12-06 10:36:43,013 - Epoch: [32][ 920/ 1200] Overall Loss 0.280800 Objective Loss 0.280800 LR 0.001000 Time 0.019904 -2022-12-06 10:36:43,203 - Epoch: [32][ 930/ 1200] Overall Loss 0.280571 Objective Loss 0.280571 LR 0.001000 Time 0.019893 -2022-12-06 10:36:43,393 - Epoch: [32][ 940/ 1200] Overall Loss 0.280815 Objective Loss 0.280815 LR 0.001000 Time 0.019883 -2022-12-06 10:36:43,583 - Epoch: [32][ 950/ 1200] Overall Loss 0.280754 Objective Loss 0.280754 LR 0.001000 Time 0.019873 -2022-12-06 10:36:43,773 - Epoch: [32][ 960/ 1200] Overall Loss 0.280902 Objective Loss 0.280902 LR 0.001000 Time 0.019864 -2022-12-06 10:36:43,963 - Epoch: [32][ 970/ 1200] Overall Loss 0.281067 Objective Loss 0.281067 LR 0.001000 Time 0.019854 -2022-12-06 10:36:44,153 - Epoch: [32][ 980/ 1200] Overall Loss 0.281030 Objective Loss 0.281030 LR 0.001000 Time 0.019845 -2022-12-06 10:36:44,343 - Epoch: [32][ 990/ 1200] Overall Loss 0.281348 Objective Loss 0.281348 LR 0.001000 Time 0.019835 -2022-12-06 10:36:44,533 - Epoch: [32][ 1000/ 1200] Overall Loss 0.281710 Objective Loss 0.281710 LR 0.001000 Time 0.019827 -2022-12-06 10:36:44,723 - Epoch: [32][ 1010/ 1200] Overall Loss 0.282036 Objective Loss 0.282036 LR 0.001000 Time 0.019818 -2022-12-06 10:36:44,913 - Epoch: [32][ 1020/ 1200] Overall Loss 0.282068 Objective Loss 0.282068 LR 0.001000 Time 0.019810 -2022-12-06 10:36:45,103 - Epoch: [32][ 1030/ 1200] Overall Loss 0.282179 Objective Loss 0.282179 LR 0.001000 Time 0.019801 -2022-12-06 10:36:45,293 - Epoch: [32][ 1040/ 1200] Overall Loss 0.282271 Objective Loss 0.282271 LR 0.001000 Time 0.019793 -2022-12-06 10:36:45,483 - Epoch: [32][ 1050/ 1200] Overall Loss 0.282560 Objective Loss 0.282560 LR 0.001000 Time 0.019785 -2022-12-06 10:36:45,674 - Epoch: [32][ 1060/ 1200] Overall Loss 0.282703 Objective Loss 0.282703 LR 0.001000 Time 0.019778 -2022-12-06 10:36:45,864 - Epoch: [32][ 1070/ 1200] Overall Loss 0.282892 Objective Loss 0.282892 LR 0.001000 Time 0.019770 -2022-12-06 10:36:46,054 - Epoch: [32][ 1080/ 1200] Overall Loss 0.282755 Objective Loss 0.282755 LR 0.001000 Time 0.019763 -2022-12-06 10:36:46,244 - Epoch: [32][ 1090/ 1200] Overall Loss 0.282998 Objective Loss 0.282998 LR 0.001000 Time 0.019755 -2022-12-06 10:36:46,434 - Epoch: [32][ 1100/ 1200] Overall Loss 0.282860 Objective Loss 0.282860 LR 0.001000 Time 0.019747 -2022-12-06 10:36:46,623 - Epoch: [32][ 1110/ 1200] Overall Loss 0.283025 Objective Loss 0.283025 LR 0.001000 Time 0.019740 -2022-12-06 10:36:46,814 - Epoch: [32][ 1120/ 1200] Overall Loss 0.283034 Objective Loss 0.283034 LR 0.001000 Time 0.019733 -2022-12-06 10:36:47,004 - Epoch: [32][ 1130/ 1200] Overall Loss 0.283193 Objective Loss 0.283193 LR 0.001000 Time 0.019726 -2022-12-06 10:36:47,194 - Epoch: [32][ 1140/ 1200] Overall Loss 0.283123 Objective Loss 0.283123 LR 0.001000 Time 0.019719 -2022-12-06 10:36:47,384 - Epoch: [32][ 1150/ 1200] Overall Loss 0.283411 Objective Loss 0.283411 LR 0.001000 Time 0.019712 -2022-12-06 10:36:47,574 - Epoch: [32][ 1160/ 1200] Overall Loss 0.283522 Objective Loss 0.283522 LR 0.001000 Time 0.019706 -2022-12-06 10:36:47,764 - Epoch: [32][ 1170/ 1200] Overall Loss 0.283634 Objective Loss 0.283634 LR 0.001000 Time 0.019699 -2022-12-06 10:36:47,954 - Epoch: [32][ 1180/ 1200] Overall Loss 0.283828 Objective Loss 0.283828 LR 0.001000 Time 0.019693 -2022-12-06 10:36:48,144 - Epoch: [32][ 1190/ 1200] Overall Loss 0.283861 Objective Loss 0.283861 LR 0.001000 Time 0.019687 -2022-12-06 10:36:48,377 - Epoch: [32][ 1200/ 1200] Overall Loss 0.284385 Objective Loss 0.284385 Top1 81.380753 Top5 97.489540 LR 0.001000 Time 0.019717 -2022-12-06 10:36:48,466 - --- validate (epoch=32)----------- -2022-12-06 10:36:48,466 - 34129 samples (256 per mini-batch) -2022-12-06 10:36:48,909 - Epoch: [32][ 10/ 134] Loss 0.320737 Top1 81.289062 Top5 97.265625 -2022-12-06 10:36:49,037 - Epoch: [32][ 20/ 134] Loss 0.309582 Top1 81.894531 Top5 97.656250 -2022-12-06 10:36:49,167 - Epoch: [32][ 30/ 134] Loss 0.308348 Top1 82.565104 Top5 97.552083 -2022-12-06 10:36:49,299 - Epoch: [32][ 40/ 134] Loss 0.301000 Top1 82.666016 Top5 97.568359 -2022-12-06 10:36:49,432 - Epoch: [32][ 50/ 134] Loss 0.306345 Top1 82.562500 Top5 97.609375 -2022-12-06 10:36:49,561 - Epoch: [32][ 60/ 134] Loss 0.303433 Top1 82.740885 Top5 97.643229 -2022-12-06 10:36:49,692 - Epoch: [32][ 70/ 134] Loss 0.304766 Top1 82.606027 Top5 97.578125 -2022-12-06 10:36:49,823 - Epoch: [32][ 80/ 134] Loss 0.302160 Top1 82.656250 Top5 97.583008 -2022-12-06 10:36:49,956 - Epoch: [32][ 90/ 134] Loss 0.303050 Top1 82.478299 Top5 97.500000 -2022-12-06 10:36:50,087 - Epoch: [32][ 100/ 134] Loss 0.302321 Top1 82.539062 Top5 97.515625 -2022-12-06 10:36:50,219 - Epoch: [32][ 110/ 134] Loss 0.301856 Top1 82.567472 Top5 97.524858 -2022-12-06 10:36:50,352 - Epoch: [32][ 120/ 134] Loss 0.303248 Top1 82.470703 Top5 97.539062 -2022-12-06 10:36:50,485 - Epoch: [32][ 130/ 134] Loss 0.303744 Top1 82.388822 Top5 97.521034 -2022-12-06 10:36:50,525 - Epoch: [32][ 134/ 134] Loss 0.303156 Top1 82.381552 Top5 97.509449 -2022-12-06 10:36:50,614 - ==> Top1: 82.382 Top5: 97.509 Loss: 0.303 - -2022-12-06 10:36:50,615 - ==> Confusion: -[[ 892 3 0 4 5 2 0 1 11 57 0 0 0 4 6 2 2 3 3 0 1] - [ 2 938 1 4 15 18 4 10 1 0 10 1 0 3 1 0 3 2 6 0 8] - [ 1 9 987 16 2 3 39 10 0 1 2 3 1 4 2 4 3 1 6 0 9] - [ 3 3 18 937 1 1 0 0 3 1 14 1 3 2 14 1 0 1 10 0 7] - [ 11 7 2 2 952 2 0 1 2 10 4 0 0 2 7 4 5 4 0 2 3] - [ 4 19 0 4 10 965 3 14 5 3 2 9 3 11 4 0 0 1 1 2 9] - [ 0 6 10 2 0 3 1071 6 1 1 1 0 0 2 1 3 1 2 0 8 0] - [ 1 29 13 0 2 33 11 918 1 1 2 3 1 0 1 1 0 2 23 10 2] - [ 6 4 1 1 1 0 0 1 969 40 7 3 0 8 16 0 3 2 0 0 2] - [ 71 1 1 1 0 3 1 5 28 872 1 0 0 9 3 1 1 1 0 0 2] - [ 0 3 5 10 0 3 1 3 8 1 954 1 0 16 5 0 2 0 3 1 3] - [ 7 7 3 0 0 21 0 4 4 0 0 937 29 11 1 2 5 6 1 10 3] - [ 2 3 1 7 3 5 1 1 3 1 0 47 850 2 7 7 2 13 1 4 9] - [ 1 1 0 1 2 10 0 2 9 16 8 3 4 957 0 1 1 1 0 1 5] - [ 7 3 1 17 2 1 1 1 23 2 3 0 1 4 1042 0 4 1 6 2 9] - [ 1 3 3 2 4 1 3 1 0 0 2 7 8 4 0 970 10 15 0 6 3] - [ 5 5 1 1 6 2 0 0 1 1 0 4 0 2 4 12 1020 1 0 4 3] - [ 1 4 1 3 0 1 4 2 0 4 2 8 18 2 4 8 3 967 1 1 2] - [ 2 8 5 11 1 5 2 16 3 0 9 1 1 4 11 0 0 3 918 2 6] - [ 2 6 4 3 2 11 10 6 0 0 1 19 7 3 1 2 7 1 1 991 3] - [ 144 352 235 185 183 228 142 127 98 126 261 140 403 368 213 137 392 92 136 263 9001]] - -2022-12-06 10:36:51,187 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:36:51,187 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:36:51,193 - - -2022-12-06 10:36:51,194 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:36:52,108 - Epoch: [33][ 10/ 1200] Overall Loss 0.283619 Objective Loss 0.283619 LR 0.001000 Time 0.091408 -2022-12-06 10:36:52,298 - Epoch: [33][ 20/ 1200] Overall Loss 0.293073 Objective Loss 0.293073 LR 0.001000 Time 0.055137 -2022-12-06 10:36:52,487 - Epoch: [33][ 30/ 1200] Overall Loss 0.288777 Objective Loss 0.288777 LR 0.001000 Time 0.043043 -2022-12-06 10:36:52,676 - Epoch: [33][ 40/ 1200] Overall Loss 0.285393 Objective Loss 0.285393 LR 0.001000 Time 0.037001 -2022-12-06 10:36:52,865 - Epoch: [33][ 50/ 1200] Overall Loss 0.288785 Objective Loss 0.288785 LR 0.001000 Time 0.033363 -2022-12-06 10:36:53,053 - Epoch: [33][ 60/ 1200] Overall Loss 0.284344 Objective Loss 0.284344 LR 0.001000 Time 0.030941 -2022-12-06 10:36:53,242 - Epoch: [33][ 70/ 1200] Overall Loss 0.280845 Objective Loss 0.280845 LR 0.001000 Time 0.029203 -2022-12-06 10:36:53,431 - Epoch: [33][ 80/ 1200] Overall Loss 0.279763 Objective Loss 0.279763 LR 0.001000 Time 0.027907 -2022-12-06 10:36:53,619 - Epoch: [33][ 90/ 1200] Overall Loss 0.281587 Objective Loss 0.281587 LR 0.001000 Time 0.026896 -2022-12-06 10:36:53,808 - Epoch: [33][ 100/ 1200] Overall Loss 0.279852 Objective Loss 0.279852 LR 0.001000 Time 0.026087 -2022-12-06 10:36:53,996 - Epoch: [33][ 110/ 1200] Overall Loss 0.279058 Objective Loss 0.279058 LR 0.001000 Time 0.025422 -2022-12-06 10:36:54,185 - Epoch: [33][ 120/ 1200] Overall Loss 0.277805 Objective Loss 0.277805 LR 0.001000 Time 0.024872 -2022-12-06 10:36:54,373 - Epoch: [33][ 130/ 1200] Overall Loss 0.275727 Objective Loss 0.275727 LR 0.001000 Time 0.024401 -2022-12-06 10:36:54,562 - Epoch: [33][ 140/ 1200] Overall Loss 0.275393 Objective Loss 0.275393 LR 0.001000 Time 0.024004 -2022-12-06 10:36:54,750 - Epoch: [33][ 150/ 1200] Overall Loss 0.275458 Objective Loss 0.275458 LR 0.001000 Time 0.023655 -2022-12-06 10:36:54,941 - Epoch: [33][ 160/ 1200] Overall Loss 0.276004 Objective Loss 0.276004 LR 0.001000 Time 0.023366 -2022-12-06 10:36:55,131 - Epoch: [33][ 170/ 1200] Overall Loss 0.276046 Objective Loss 0.276046 LR 0.001000 Time 0.023107 -2022-12-06 10:36:55,322 - Epoch: [33][ 180/ 1200] Overall Loss 0.278687 Objective Loss 0.278687 LR 0.001000 Time 0.022879 -2022-12-06 10:36:55,512 - Epoch: [33][ 190/ 1200] Overall Loss 0.278642 Objective Loss 0.278642 LR 0.001000 Time 0.022674 -2022-12-06 10:36:55,703 - Epoch: [33][ 200/ 1200] Overall Loss 0.279858 Objective Loss 0.279858 LR 0.001000 Time 0.022492 -2022-12-06 10:36:55,893 - Epoch: [33][ 210/ 1200] Overall Loss 0.279611 Objective Loss 0.279611 LR 0.001000 Time 0.022324 -2022-12-06 10:36:56,085 - Epoch: [33][ 220/ 1200] Overall Loss 0.280157 Objective Loss 0.280157 LR 0.001000 Time 0.022179 -2022-12-06 10:36:56,275 - Epoch: [33][ 230/ 1200] Overall Loss 0.279425 Objective Loss 0.279425 LR 0.001000 Time 0.022037 -2022-12-06 10:36:56,466 - Epoch: [33][ 240/ 1200] Overall Loss 0.280488 Objective Loss 0.280488 LR 0.001000 Time 0.021914 -2022-12-06 10:36:56,656 - Epoch: [33][ 250/ 1200] Overall Loss 0.280101 Objective Loss 0.280101 LR 0.001000 Time 0.021794 -2022-12-06 10:36:56,847 - Epoch: [33][ 260/ 1200] Overall Loss 0.282019 Objective Loss 0.282019 LR 0.001000 Time 0.021688 -2022-12-06 10:36:57,037 - Epoch: [33][ 270/ 1200] Overall Loss 0.281911 Objective Loss 0.281911 LR 0.001000 Time 0.021587 -2022-12-06 10:36:57,228 - Epoch: [33][ 280/ 1200] Overall Loss 0.282395 Objective Loss 0.282395 LR 0.001000 Time 0.021496 -2022-12-06 10:36:57,417 - Epoch: [33][ 290/ 1200] Overall Loss 0.283647 Objective Loss 0.283647 LR 0.001000 Time 0.021407 -2022-12-06 10:36:57,608 - Epoch: [33][ 300/ 1200] Overall Loss 0.284023 Objective Loss 0.284023 LR 0.001000 Time 0.021327 -2022-12-06 10:36:57,798 - Epoch: [33][ 310/ 1200] Overall Loss 0.283754 Objective Loss 0.283754 LR 0.001000 Time 0.021252 -2022-12-06 10:36:57,989 - Epoch: [33][ 320/ 1200] Overall Loss 0.284382 Objective Loss 0.284382 LR 0.001000 Time 0.021182 -2022-12-06 10:36:58,180 - Epoch: [33][ 330/ 1200] Overall Loss 0.284059 Objective Loss 0.284059 LR 0.001000 Time 0.021115 -2022-12-06 10:36:58,371 - Epoch: [33][ 340/ 1200] Overall Loss 0.283952 Objective Loss 0.283952 LR 0.001000 Time 0.021055 -2022-12-06 10:36:58,561 - Epoch: [33][ 350/ 1200] Overall Loss 0.283936 Objective Loss 0.283936 LR 0.001000 Time 0.020995 -2022-12-06 10:36:58,752 - Epoch: [33][ 360/ 1200] Overall Loss 0.284254 Objective Loss 0.284254 LR 0.001000 Time 0.020941 -2022-12-06 10:36:58,942 - Epoch: [33][ 370/ 1200] Overall Loss 0.283909 Objective Loss 0.283909 LR 0.001000 Time 0.020887 -2022-12-06 10:36:59,133 - Epoch: [33][ 380/ 1200] Overall Loss 0.283756 Objective Loss 0.283756 LR 0.001000 Time 0.020839 -2022-12-06 10:36:59,323 - Epoch: [33][ 390/ 1200] Overall Loss 0.283227 Objective Loss 0.283227 LR 0.001000 Time 0.020790 -2022-12-06 10:36:59,513 - Epoch: [33][ 400/ 1200] Overall Loss 0.283471 Objective Loss 0.283471 LR 0.001000 Time 0.020745 -2022-12-06 10:36:59,703 - Epoch: [33][ 410/ 1200] Overall Loss 0.283278 Objective Loss 0.283278 LR 0.001000 Time 0.020700 -2022-12-06 10:36:59,893 - Epoch: [33][ 420/ 1200] Overall Loss 0.283365 Objective Loss 0.283365 LR 0.001000 Time 0.020660 -2022-12-06 10:37:00,083 - Epoch: [33][ 430/ 1200] Overall Loss 0.283113 Objective Loss 0.283113 LR 0.001000 Time 0.020618 -2022-12-06 10:37:00,274 - Epoch: [33][ 440/ 1200] Overall Loss 0.284343 Objective Loss 0.284343 LR 0.001000 Time 0.020582 -2022-12-06 10:37:00,463 - Epoch: [33][ 450/ 1200] Overall Loss 0.285785 Objective Loss 0.285785 LR 0.001000 Time 0.020545 -2022-12-06 10:37:00,654 - Epoch: [33][ 460/ 1200] Overall Loss 0.286040 Objective Loss 0.286040 LR 0.001000 Time 0.020513 -2022-12-06 10:37:00,844 - Epoch: [33][ 470/ 1200] Overall Loss 0.285512 Objective Loss 0.285512 LR 0.001000 Time 0.020480 -2022-12-06 10:37:01,036 - Epoch: [33][ 480/ 1200] Overall Loss 0.284744 Objective Loss 0.284744 LR 0.001000 Time 0.020450 -2022-12-06 10:37:01,226 - Epoch: [33][ 490/ 1200] Overall Loss 0.285112 Objective Loss 0.285112 LR 0.001000 Time 0.020420 -2022-12-06 10:37:01,417 - Epoch: [33][ 500/ 1200] Overall Loss 0.284525 Objective Loss 0.284525 LR 0.001000 Time 0.020392 -2022-12-06 10:37:01,607 - Epoch: [33][ 510/ 1200] Overall Loss 0.284457 Objective Loss 0.284457 LR 0.001000 Time 0.020364 -2022-12-06 10:37:01,798 - Epoch: [33][ 520/ 1200] Overall Loss 0.284210 Objective Loss 0.284210 LR 0.001000 Time 0.020338 -2022-12-06 10:37:01,987 - Epoch: [33][ 530/ 1200] Overall Loss 0.284618 Objective Loss 0.284618 LR 0.001000 Time 0.020311 -2022-12-06 10:37:02,177 - Epoch: [33][ 540/ 1200] Overall Loss 0.284283 Objective Loss 0.284283 LR 0.001000 Time 0.020286 -2022-12-06 10:37:02,367 - Epoch: [33][ 550/ 1200] Overall Loss 0.283739 Objective Loss 0.283739 LR 0.001000 Time 0.020262 -2022-12-06 10:37:02,558 - Epoch: [33][ 560/ 1200] Overall Loss 0.283447 Objective Loss 0.283447 LR 0.001000 Time 0.020240 -2022-12-06 10:37:02,749 - Epoch: [33][ 570/ 1200] Overall Loss 0.283462 Objective Loss 0.283462 LR 0.001000 Time 0.020218 -2022-12-06 10:37:02,939 - Epoch: [33][ 580/ 1200] Overall Loss 0.283517 Objective Loss 0.283517 LR 0.001000 Time 0.020197 -2022-12-06 10:37:03,129 - Epoch: [33][ 590/ 1200] Overall Loss 0.283494 Objective Loss 0.283494 LR 0.001000 Time 0.020176 -2022-12-06 10:37:03,320 - Epoch: [33][ 600/ 1200] Overall Loss 0.283700 Objective Loss 0.283700 LR 0.001000 Time 0.020157 -2022-12-06 10:37:03,510 - Epoch: [33][ 610/ 1200] Overall Loss 0.283650 Objective Loss 0.283650 LR 0.001000 Time 0.020137 -2022-12-06 10:37:03,701 - Epoch: [33][ 620/ 1200] Overall Loss 0.283839 Objective Loss 0.283839 LR 0.001000 Time 0.020120 -2022-12-06 10:37:03,890 - Epoch: [33][ 630/ 1200] Overall Loss 0.283787 Objective Loss 0.283787 LR 0.001000 Time 0.020100 -2022-12-06 10:37:04,082 - Epoch: [33][ 640/ 1200] Overall Loss 0.283910 Objective Loss 0.283910 LR 0.001000 Time 0.020083 -2022-12-06 10:37:04,271 - Epoch: [33][ 650/ 1200] Overall Loss 0.283562 Objective Loss 0.283562 LR 0.001000 Time 0.020065 -2022-12-06 10:37:04,462 - Epoch: [33][ 660/ 1200] Overall Loss 0.283755 Objective Loss 0.283755 LR 0.001000 Time 0.020050 -2022-12-06 10:37:04,652 - Epoch: [33][ 670/ 1200] Overall Loss 0.283682 Objective Loss 0.283682 LR 0.001000 Time 0.020033 -2022-12-06 10:37:04,843 - Epoch: [33][ 680/ 1200] Overall Loss 0.284176 Objective Loss 0.284176 LR 0.001000 Time 0.020019 -2022-12-06 10:37:05,034 - Epoch: [33][ 690/ 1200] Overall Loss 0.284260 Objective Loss 0.284260 LR 0.001000 Time 0.020004 -2022-12-06 10:37:05,225 - Epoch: [33][ 700/ 1200] Overall Loss 0.284383 Objective Loss 0.284383 LR 0.001000 Time 0.019991 -2022-12-06 10:37:05,414 - Epoch: [33][ 710/ 1200] Overall Loss 0.285199 Objective Loss 0.285199 LR 0.001000 Time 0.019975 -2022-12-06 10:37:05,606 - Epoch: [33][ 720/ 1200] Overall Loss 0.285489 Objective Loss 0.285489 LR 0.001000 Time 0.019963 -2022-12-06 10:37:05,796 - Epoch: [33][ 730/ 1200] Overall Loss 0.285460 Objective Loss 0.285460 LR 0.001000 Time 0.019949 -2022-12-06 10:37:05,987 - Epoch: [33][ 740/ 1200] Overall Loss 0.285427 Objective Loss 0.285427 LR 0.001000 Time 0.019937 -2022-12-06 10:37:06,177 - Epoch: [33][ 750/ 1200] Overall Loss 0.285018 Objective Loss 0.285018 LR 0.001000 Time 0.019924 -2022-12-06 10:37:06,368 - Epoch: [33][ 760/ 1200] Overall Loss 0.284858 Objective Loss 0.284858 LR 0.001000 Time 0.019912 -2022-12-06 10:37:06,559 - Epoch: [33][ 770/ 1200] Overall Loss 0.285012 Objective Loss 0.285012 LR 0.001000 Time 0.019901 -2022-12-06 10:37:06,749 - Epoch: [33][ 780/ 1200] Overall Loss 0.285159 Objective Loss 0.285159 LR 0.001000 Time 0.019889 -2022-12-06 10:37:06,940 - Epoch: [33][ 790/ 1200] Overall Loss 0.285174 Objective Loss 0.285174 LR 0.001000 Time 0.019878 -2022-12-06 10:37:07,132 - Epoch: [33][ 800/ 1200] Overall Loss 0.284921 Objective Loss 0.284921 LR 0.001000 Time 0.019868 -2022-12-06 10:37:07,321 - Epoch: [33][ 810/ 1200] Overall Loss 0.284802 Objective Loss 0.284802 LR 0.001000 Time 0.019856 -2022-12-06 10:37:07,513 - Epoch: [33][ 820/ 1200] Overall Loss 0.285287 Objective Loss 0.285287 LR 0.001000 Time 0.019848 -2022-12-06 10:37:07,703 - Epoch: [33][ 830/ 1200] Overall Loss 0.285403 Objective Loss 0.285403 LR 0.001000 Time 0.019837 -2022-12-06 10:37:07,895 - Epoch: [33][ 840/ 1200] Overall Loss 0.285188 Objective Loss 0.285188 LR 0.001000 Time 0.019829 -2022-12-06 10:37:08,085 - Epoch: [33][ 850/ 1200] Overall Loss 0.285287 Objective Loss 0.285287 LR 0.001000 Time 0.019818 -2022-12-06 10:37:08,276 - Epoch: [33][ 860/ 1200] Overall Loss 0.285323 Objective Loss 0.285323 LR 0.001000 Time 0.019809 -2022-12-06 10:37:08,466 - Epoch: [33][ 870/ 1200] Overall Loss 0.285464 Objective Loss 0.285464 LR 0.001000 Time 0.019799 -2022-12-06 10:37:08,657 - Epoch: [33][ 880/ 1200] Overall Loss 0.285672 Objective Loss 0.285672 LR 0.001000 Time 0.019790 -2022-12-06 10:37:08,847 - Epoch: [33][ 890/ 1200] Overall Loss 0.285397 Objective Loss 0.285397 LR 0.001000 Time 0.019781 -2022-12-06 10:37:09,038 - Epoch: [33][ 900/ 1200] Overall Loss 0.285385 Objective Loss 0.285385 LR 0.001000 Time 0.019773 -2022-12-06 10:37:09,228 - Epoch: [33][ 910/ 1200] Overall Loss 0.285505 Objective Loss 0.285505 LR 0.001000 Time 0.019764 -2022-12-06 10:37:09,420 - Epoch: [33][ 920/ 1200] Overall Loss 0.285721 Objective Loss 0.285721 LR 0.001000 Time 0.019757 -2022-12-06 10:37:09,610 - Epoch: [33][ 930/ 1200] Overall Loss 0.285817 Objective Loss 0.285817 LR 0.001000 Time 0.019749 -2022-12-06 10:37:09,801 - Epoch: [33][ 940/ 1200] Overall Loss 0.285892 Objective Loss 0.285892 LR 0.001000 Time 0.019741 -2022-12-06 10:37:09,990 - Epoch: [33][ 950/ 1200] Overall Loss 0.285771 Objective Loss 0.285771 LR 0.001000 Time 0.019732 -2022-12-06 10:37:10,181 - Epoch: [33][ 960/ 1200] Overall Loss 0.285938 Objective Loss 0.285938 LR 0.001000 Time 0.019724 -2022-12-06 10:37:10,371 - Epoch: [33][ 970/ 1200] Overall Loss 0.285894 Objective Loss 0.285894 LR 0.001000 Time 0.019716 -2022-12-06 10:37:10,562 - Epoch: [33][ 980/ 1200] Overall Loss 0.286112 Objective Loss 0.286112 LR 0.001000 Time 0.019709 -2022-12-06 10:37:10,751 - Epoch: [33][ 990/ 1200] Overall Loss 0.285775 Objective Loss 0.285775 LR 0.001000 Time 0.019701 -2022-12-06 10:37:10,942 - Epoch: [33][ 1000/ 1200] Overall Loss 0.285741 Objective Loss 0.285741 LR 0.001000 Time 0.019695 -2022-12-06 10:37:11,132 - Epoch: [33][ 1010/ 1200] Overall Loss 0.285771 Objective Loss 0.285771 LR 0.001000 Time 0.019687 -2022-12-06 10:37:11,324 - Epoch: [33][ 1020/ 1200] Overall Loss 0.285959 Objective Loss 0.285959 LR 0.001000 Time 0.019682 -2022-12-06 10:37:11,514 - Epoch: [33][ 1030/ 1200] Overall Loss 0.285763 Objective Loss 0.285763 LR 0.001000 Time 0.019674 -2022-12-06 10:37:11,705 - Epoch: [33][ 1040/ 1200] Overall Loss 0.285660 Objective Loss 0.285660 LR 0.001000 Time 0.019668 -2022-12-06 10:37:11,894 - Epoch: [33][ 1050/ 1200] Overall Loss 0.285850 Objective Loss 0.285850 LR 0.001000 Time 0.019661 -2022-12-06 10:37:12,085 - Epoch: [33][ 1060/ 1200] Overall Loss 0.285639 Objective Loss 0.285639 LR 0.001000 Time 0.019655 -2022-12-06 10:37:12,275 - Epoch: [33][ 1070/ 1200] Overall Loss 0.285931 Objective Loss 0.285931 LR 0.001000 Time 0.019648 -2022-12-06 10:37:12,466 - Epoch: [33][ 1080/ 1200] Overall Loss 0.286153 Objective Loss 0.286153 LR 0.001000 Time 0.019642 -2022-12-06 10:37:12,656 - Epoch: [33][ 1090/ 1200] Overall Loss 0.285848 Objective Loss 0.285848 LR 0.001000 Time 0.019636 -2022-12-06 10:37:12,847 - Epoch: [33][ 1100/ 1200] Overall Loss 0.285944 Objective Loss 0.285944 LR 0.001000 Time 0.019631 -2022-12-06 10:37:13,037 - Epoch: [33][ 1110/ 1200] Overall Loss 0.285948 Objective Loss 0.285948 LR 0.001000 Time 0.019625 -2022-12-06 10:37:13,228 - Epoch: [33][ 1120/ 1200] Overall Loss 0.285822 Objective Loss 0.285822 LR 0.001000 Time 0.019619 -2022-12-06 10:37:13,418 - Epoch: [33][ 1130/ 1200] Overall Loss 0.285890 Objective Loss 0.285890 LR 0.001000 Time 0.019613 -2022-12-06 10:37:13,609 - Epoch: [33][ 1140/ 1200] Overall Loss 0.286121 Objective Loss 0.286121 LR 0.001000 Time 0.019608 -2022-12-06 10:37:13,798 - Epoch: [33][ 1150/ 1200] Overall Loss 0.286373 Objective Loss 0.286373 LR 0.001000 Time 0.019602 -2022-12-06 10:37:13,989 - Epoch: [33][ 1160/ 1200] Overall Loss 0.286426 Objective Loss 0.286426 LR 0.001000 Time 0.019597 -2022-12-06 10:37:14,178 - Epoch: [33][ 1170/ 1200] Overall Loss 0.286365 Objective Loss 0.286365 LR 0.001000 Time 0.019591 -2022-12-06 10:37:14,369 - Epoch: [33][ 1180/ 1200] Overall Loss 0.286337 Objective Loss 0.286337 LR 0.001000 Time 0.019586 -2022-12-06 10:37:14,558 - Epoch: [33][ 1190/ 1200] Overall Loss 0.286421 Objective Loss 0.286421 LR 0.001000 Time 0.019580 -2022-12-06 10:37:14,781 - Epoch: [33][ 1200/ 1200] Overall Loss 0.286556 Objective Loss 0.286556 Top1 85.774059 Top5 97.280335 LR 0.001000 Time 0.019602 -2022-12-06 10:37:14,870 - --- validate (epoch=33)----------- -2022-12-06 10:37:14,870 - 34129 samples (256 per mini-batch) -2022-12-06 10:37:15,429 - Epoch: [33][ 10/ 134] Loss 0.276827 Top1 85.156250 Top5 97.929688 -2022-12-06 10:37:15,562 - Epoch: [33][ 20/ 134] Loss 0.276727 Top1 85.058594 Top5 98.144531 -2022-12-06 10:37:15,694 - Epoch: [33][ 30/ 134] Loss 0.279763 Top1 84.895833 Top5 98.007812 -2022-12-06 10:37:15,825 - Epoch: [33][ 40/ 134] Loss 0.287995 Top1 84.736328 Top5 97.919922 -2022-12-06 10:37:15,955 - Epoch: [33][ 50/ 134] Loss 0.293520 Top1 84.593750 Top5 97.960938 -2022-12-06 10:37:16,085 - Epoch: [33][ 60/ 134] Loss 0.290467 Top1 84.583333 Top5 98.014323 -2022-12-06 10:37:16,215 - Epoch: [33][ 70/ 134] Loss 0.297927 Top1 84.592634 Top5 97.996652 -2022-12-06 10:37:16,346 - Epoch: [33][ 80/ 134] Loss 0.303734 Top1 84.448242 Top5 97.944336 -2022-12-06 10:37:16,476 - Epoch: [33][ 90/ 134] Loss 0.302363 Top1 84.453125 Top5 97.942708 -2022-12-06 10:37:16,607 - Epoch: [33][ 100/ 134] Loss 0.305973 Top1 84.425781 Top5 97.964844 -2022-12-06 10:37:16,738 - Epoch: [33][ 110/ 134] Loss 0.306216 Top1 84.392756 Top5 97.975852 -2022-12-06 10:37:16,868 - Epoch: [33][ 120/ 134] Loss 0.306326 Top1 84.326172 Top5 97.985026 -2022-12-06 10:37:17,000 - Epoch: [33][ 130/ 134] Loss 0.306459 Top1 84.323918 Top5 98.007812 -2022-12-06 10:37:17,039 - Epoch: [33][ 134/ 134] Loss 0.305761 Top1 84.303671 Top5 98.004629 -2022-12-06 10:37:17,127 - ==> Top1: 84.304 Top5: 98.005 Loss: 0.306 - -2022-12-06 10:37:17,127 - ==> Confusion: -[[ 888 2 2 2 5 9 0 4 8 55 1 3 1 2 1 2 1 1 2 0 7] - [ 0 903 1 2 8 43 2 19 0 1 2 2 5 3 1 1 3 4 11 7 9] - [ 6 6 950 36 0 8 42 12 0 2 3 3 1 7 1 5 2 1 4 3 11] - [ 1 2 14 952 0 3 1 1 3 0 11 0 3 2 7 0 2 6 9 0 3] - [ 9 8 1 0 939 6 1 2 0 9 1 4 1 4 10 6 7 3 0 2 7] - [ 1 8 0 3 4 979 2 13 3 2 0 16 6 16 2 1 2 1 0 4 6] - [ 0 3 7 3 0 4 1069 5 1 0 1 4 1 0 0 5 0 2 2 9 2] - [ 0 10 5 4 1 34 6 937 0 0 1 6 2 2 0 3 0 1 20 18 4] - [ 2 4 0 2 2 6 0 0 967 45 8 3 2 6 10 0 1 1 2 1 2] - [ 65 0 1 1 1 4 1 2 39 861 1 2 1 12 1 1 0 0 1 0 7] - [ 0 2 2 16 0 4 5 5 7 3 926 4 3 17 4 0 0 1 11 1 8] - [ 4 1 0 0 0 6 2 3 2 3 1 967 25 7 0 5 2 7 0 15 1] - [ 0 0 1 7 1 2 0 1 4 0 1 37 883 1 2 8 1 6 1 3 10] - [ 0 3 1 0 2 9 0 3 16 10 3 5 5 951 1 3 0 2 0 2 7] - [ 13 4 0 37 3 1 1 2 20 4 0 2 4 4 998 1 3 3 16 0 14] - [ 1 1 0 4 4 0 4 0 0 1 0 9 6 4 0 988 3 12 0 2 4] - [ 0 3 0 2 2 0 1 0 1 2 0 5 4 6 2 20 1006 2 0 7 9] - [ 1 2 0 3 0 2 1 1 3 2 0 19 22 2 2 7 1 966 0 0 2] - [ 4 4 2 17 0 3 0 34 1 1 4 5 5 1 5 1 1 2 911 3 4] - [ 1 3 0 0 0 12 4 4 0 0 3 19 8 5 0 3 4 3 1 1004 6] - [ 138 243 147 186 94 267 102 177 81 96 140 134 391 339 121 155 145 102 167 279 9722]] - -2022-12-06 10:37:17,708 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:37:17,708 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:37:17,714 - - -2022-12-06 10:37:17,714 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:37:18,650 - Epoch: [34][ 10/ 1200] Overall Loss 0.251312 Objective Loss 0.251312 LR 0.001000 Time 0.093571 -2022-12-06 10:37:18,854 - Epoch: [34][ 20/ 1200] Overall Loss 0.257822 Objective Loss 0.257822 LR 0.001000 Time 0.056950 -2022-12-06 10:37:19,046 - Epoch: [34][ 30/ 1200] Overall Loss 0.256258 Objective Loss 0.256258 LR 0.001000 Time 0.044340 -2022-12-06 10:37:19,237 - Epoch: [34][ 40/ 1200] Overall Loss 0.265850 Objective Loss 0.265850 LR 0.001000 Time 0.038013 -2022-12-06 10:37:19,428 - Epoch: [34][ 50/ 1200] Overall Loss 0.272937 Objective Loss 0.272937 LR 0.001000 Time 0.034226 -2022-12-06 10:37:19,619 - Epoch: [34][ 60/ 1200] Overall Loss 0.276385 Objective Loss 0.276385 LR 0.001000 Time 0.031697 -2022-12-06 10:37:19,811 - Epoch: [34][ 70/ 1200] Overall Loss 0.278580 Objective Loss 0.278580 LR 0.001000 Time 0.029900 -2022-12-06 10:37:20,002 - Epoch: [34][ 80/ 1200] Overall Loss 0.277093 Objective Loss 0.277093 LR 0.001000 Time 0.028540 -2022-12-06 10:37:20,194 - Epoch: [34][ 90/ 1200] Overall Loss 0.276302 Objective Loss 0.276302 LR 0.001000 Time 0.027495 -2022-12-06 10:37:20,385 - Epoch: [34][ 100/ 1200] Overall Loss 0.277805 Objective Loss 0.277805 LR 0.001000 Time 0.026651 -2022-12-06 10:37:20,576 - Epoch: [34][ 110/ 1200] Overall Loss 0.277339 Objective Loss 0.277339 LR 0.001000 Time 0.025963 -2022-12-06 10:37:20,767 - Epoch: [34][ 120/ 1200] Overall Loss 0.278274 Objective Loss 0.278274 LR 0.001000 Time 0.025386 -2022-12-06 10:37:20,959 - Epoch: [34][ 130/ 1200] Overall Loss 0.276767 Objective Loss 0.276767 LR 0.001000 Time 0.024903 -2022-12-06 10:37:21,150 - Epoch: [34][ 140/ 1200] Overall Loss 0.276020 Objective Loss 0.276020 LR 0.001000 Time 0.024484 -2022-12-06 10:37:21,341 - Epoch: [34][ 150/ 1200] Overall Loss 0.274751 Objective Loss 0.274751 LR 0.001000 Time 0.024123 -2022-12-06 10:37:21,532 - Epoch: [34][ 160/ 1200] Overall Loss 0.273375 Objective Loss 0.273375 LR 0.001000 Time 0.023808 -2022-12-06 10:37:21,724 - Epoch: [34][ 170/ 1200] Overall Loss 0.273900 Objective Loss 0.273900 LR 0.001000 Time 0.023530 -2022-12-06 10:37:21,915 - Epoch: [34][ 180/ 1200] Overall Loss 0.273761 Objective Loss 0.273761 LR 0.001000 Time 0.023282 -2022-12-06 10:37:22,106 - Epoch: [34][ 190/ 1200] Overall Loss 0.272316 Objective Loss 0.272316 LR 0.001000 Time 0.023061 -2022-12-06 10:37:22,297 - Epoch: [34][ 200/ 1200] Overall Loss 0.271615 Objective Loss 0.271615 LR 0.001000 Time 0.022861 -2022-12-06 10:37:22,489 - Epoch: [34][ 210/ 1200] Overall Loss 0.270315 Objective Loss 0.270315 LR 0.001000 Time 0.022683 -2022-12-06 10:37:22,680 - Epoch: [34][ 220/ 1200] Overall Loss 0.271879 Objective Loss 0.271879 LR 0.001000 Time 0.022519 -2022-12-06 10:37:22,872 - Epoch: [34][ 230/ 1200] Overall Loss 0.272826 Objective Loss 0.272826 LR 0.001000 Time 0.022370 -2022-12-06 10:37:23,063 - Epoch: [34][ 240/ 1200] Overall Loss 0.271479 Objective Loss 0.271479 LR 0.001000 Time 0.022233 -2022-12-06 10:37:23,255 - Epoch: [34][ 250/ 1200] Overall Loss 0.272053 Objective Loss 0.272053 LR 0.001000 Time 0.022109 -2022-12-06 10:37:23,446 - Epoch: [34][ 260/ 1200] Overall Loss 0.273131 Objective Loss 0.273131 LR 0.001000 Time 0.021992 -2022-12-06 10:37:23,638 - Epoch: [34][ 270/ 1200] Overall Loss 0.273436 Objective Loss 0.273436 LR 0.001000 Time 0.021887 -2022-12-06 10:37:23,830 - Epoch: [34][ 280/ 1200] Overall Loss 0.273013 Objective Loss 0.273013 LR 0.001000 Time 0.021787 -2022-12-06 10:37:24,021 - Epoch: [34][ 290/ 1200] Overall Loss 0.272997 Objective Loss 0.272997 LR 0.001000 Time 0.021694 -2022-12-06 10:37:24,212 - Epoch: [34][ 300/ 1200] Overall Loss 0.273019 Objective Loss 0.273019 LR 0.001000 Time 0.021604 -2022-12-06 10:37:24,404 - Epoch: [34][ 310/ 1200] Overall Loss 0.273456 Objective Loss 0.273456 LR 0.001000 Time 0.021525 -2022-12-06 10:37:24,595 - Epoch: [34][ 320/ 1200] Overall Loss 0.273412 Objective Loss 0.273412 LR 0.001000 Time 0.021449 -2022-12-06 10:37:24,787 - Epoch: [34][ 330/ 1200] Overall Loss 0.273663 Objective Loss 0.273663 LR 0.001000 Time 0.021378 -2022-12-06 10:37:24,979 - Epoch: [34][ 340/ 1200] Overall Loss 0.273019 Objective Loss 0.273019 LR 0.001000 Time 0.021312 -2022-12-06 10:37:25,170 - Epoch: [34][ 350/ 1200] Overall Loss 0.273025 Objective Loss 0.273025 LR 0.001000 Time 0.021247 -2022-12-06 10:37:25,361 - Epoch: [34][ 360/ 1200] Overall Loss 0.273242 Objective Loss 0.273242 LR 0.001000 Time 0.021186 -2022-12-06 10:37:25,553 - Epoch: [34][ 370/ 1200] Overall Loss 0.273678 Objective Loss 0.273678 LR 0.001000 Time 0.021130 -2022-12-06 10:37:25,744 - Epoch: [34][ 380/ 1200] Overall Loss 0.273298 Objective Loss 0.273298 LR 0.001000 Time 0.021076 -2022-12-06 10:37:25,935 - Epoch: [34][ 390/ 1200] Overall Loss 0.273433 Objective Loss 0.273433 LR 0.001000 Time 0.021024 -2022-12-06 10:37:26,128 - Epoch: [34][ 400/ 1200] Overall Loss 0.274283 Objective Loss 0.274283 LR 0.001000 Time 0.020979 -2022-12-06 10:37:26,319 - Epoch: [34][ 410/ 1200] Overall Loss 0.274164 Objective Loss 0.274164 LR 0.001000 Time 0.020933 -2022-12-06 10:37:26,510 - Epoch: [34][ 420/ 1200] Overall Loss 0.274172 Objective Loss 0.274172 LR 0.001000 Time 0.020888 -2022-12-06 10:37:26,702 - Epoch: [34][ 430/ 1200] Overall Loss 0.274629 Objective Loss 0.274629 LR 0.001000 Time 0.020847 -2022-12-06 10:37:26,894 - Epoch: [34][ 440/ 1200] Overall Loss 0.274910 Objective Loss 0.274910 LR 0.001000 Time 0.020807 -2022-12-06 10:37:27,085 - Epoch: [34][ 450/ 1200] Overall Loss 0.275302 Objective Loss 0.275302 LR 0.001000 Time 0.020768 -2022-12-06 10:37:27,276 - Epoch: [34][ 460/ 1200] Overall Loss 0.275326 Objective Loss 0.275326 LR 0.001000 Time 0.020731 -2022-12-06 10:37:27,467 - Epoch: [34][ 470/ 1200] Overall Loss 0.275952 Objective Loss 0.275952 LR 0.001000 Time 0.020695 -2022-12-06 10:37:27,658 - Epoch: [34][ 480/ 1200] Overall Loss 0.276237 Objective Loss 0.276237 LR 0.001000 Time 0.020661 -2022-12-06 10:37:27,849 - Epoch: [34][ 490/ 1200] Overall Loss 0.276513 Objective Loss 0.276513 LR 0.001000 Time 0.020628 -2022-12-06 10:37:28,040 - Epoch: [34][ 500/ 1200] Overall Loss 0.276814 Objective Loss 0.276814 LR 0.001000 Time 0.020596 -2022-12-06 10:37:28,232 - Epoch: [34][ 510/ 1200] Overall Loss 0.277511 Objective Loss 0.277511 LR 0.001000 Time 0.020567 -2022-12-06 10:37:28,423 - Epoch: [34][ 520/ 1200] Overall Loss 0.277507 Objective Loss 0.277507 LR 0.001000 Time 0.020539 -2022-12-06 10:37:28,614 - Epoch: [34][ 530/ 1200] Overall Loss 0.277424 Objective Loss 0.277424 LR 0.001000 Time 0.020511 -2022-12-06 10:37:28,806 - Epoch: [34][ 540/ 1200] Overall Loss 0.277270 Objective Loss 0.277270 LR 0.001000 Time 0.020485 -2022-12-06 10:37:28,997 - Epoch: [34][ 550/ 1200] Overall Loss 0.277537 Objective Loss 0.277537 LR 0.001000 Time 0.020459 -2022-12-06 10:37:29,188 - Epoch: [34][ 560/ 1200] Overall Loss 0.277004 Objective Loss 0.277004 LR 0.001000 Time 0.020434 -2022-12-06 10:37:29,379 - Epoch: [34][ 570/ 1200] Overall Loss 0.277384 Objective Loss 0.277384 LR 0.001000 Time 0.020410 -2022-12-06 10:37:29,571 - Epoch: [34][ 580/ 1200] Overall Loss 0.277310 Objective Loss 0.277310 LR 0.001000 Time 0.020387 -2022-12-06 10:37:29,763 - Epoch: [34][ 590/ 1200] Overall Loss 0.277090 Objective Loss 0.277090 LR 0.001000 Time 0.020367 -2022-12-06 10:37:29,955 - Epoch: [34][ 600/ 1200] Overall Loss 0.277504 Objective Loss 0.277504 LR 0.001000 Time 0.020346 -2022-12-06 10:37:30,146 - Epoch: [34][ 610/ 1200] Overall Loss 0.277790 Objective Loss 0.277790 LR 0.001000 Time 0.020325 -2022-12-06 10:37:30,337 - Epoch: [34][ 620/ 1200] Overall Loss 0.278485 Objective Loss 0.278485 LR 0.001000 Time 0.020305 -2022-12-06 10:37:30,528 - Epoch: [34][ 630/ 1200] Overall Loss 0.278675 Objective Loss 0.278675 LR 0.001000 Time 0.020285 -2022-12-06 10:37:30,720 - Epoch: [34][ 640/ 1200] Overall Loss 0.278601 Objective Loss 0.278601 LR 0.001000 Time 0.020266 -2022-12-06 10:37:30,911 - Epoch: [34][ 650/ 1200] Overall Loss 0.278818 Objective Loss 0.278818 LR 0.001000 Time 0.020248 -2022-12-06 10:37:31,102 - Epoch: [34][ 660/ 1200] Overall Loss 0.278905 Objective Loss 0.278905 LR 0.001000 Time 0.020230 -2022-12-06 10:37:31,294 - Epoch: [34][ 670/ 1200] Overall Loss 0.279312 Objective Loss 0.279312 LR 0.001000 Time 0.020212 -2022-12-06 10:37:31,484 - Epoch: [34][ 680/ 1200] Overall Loss 0.279426 Objective Loss 0.279426 LR 0.001000 Time 0.020195 -2022-12-06 10:37:31,676 - Epoch: [34][ 690/ 1200] Overall Loss 0.279782 Objective Loss 0.279782 LR 0.001000 Time 0.020180 -2022-12-06 10:37:31,868 - Epoch: [34][ 700/ 1200] Overall Loss 0.279806 Objective Loss 0.279806 LR 0.001000 Time 0.020164 -2022-12-06 10:37:32,059 - Epoch: [34][ 710/ 1200] Overall Loss 0.279316 Objective Loss 0.279316 LR 0.001000 Time 0.020149 -2022-12-06 10:37:32,251 - Epoch: [34][ 720/ 1200] Overall Loss 0.279437 Objective Loss 0.279437 LR 0.001000 Time 0.020134 -2022-12-06 10:37:32,442 - Epoch: [34][ 730/ 1200] Overall Loss 0.279510 Objective Loss 0.279510 LR 0.001000 Time 0.020120 -2022-12-06 10:37:32,633 - Epoch: [34][ 740/ 1200] Overall Loss 0.279645 Objective Loss 0.279645 LR 0.001000 Time 0.020106 -2022-12-06 10:37:32,824 - Epoch: [34][ 750/ 1200] Overall Loss 0.279819 Objective Loss 0.279819 LR 0.001000 Time 0.020092 -2022-12-06 10:37:33,016 - Epoch: [34][ 760/ 1200] Overall Loss 0.279929 Objective Loss 0.279929 LR 0.001000 Time 0.020078 -2022-12-06 10:37:33,207 - Epoch: [34][ 770/ 1200] Overall Loss 0.279743 Objective Loss 0.279743 LR 0.001000 Time 0.020065 -2022-12-06 10:37:33,399 - Epoch: [34][ 780/ 1200] Overall Loss 0.279711 Objective Loss 0.279711 LR 0.001000 Time 0.020054 -2022-12-06 10:37:33,590 - Epoch: [34][ 790/ 1200] Overall Loss 0.279844 Objective Loss 0.279844 LR 0.001000 Time 0.020041 -2022-12-06 10:37:33,782 - Epoch: [34][ 800/ 1200] Overall Loss 0.279994 Objective Loss 0.279994 LR 0.001000 Time 0.020029 -2022-12-06 10:37:33,973 - Epoch: [34][ 810/ 1200] Overall Loss 0.280276 Objective Loss 0.280276 LR 0.001000 Time 0.020018 -2022-12-06 10:37:34,165 - Epoch: [34][ 820/ 1200] Overall Loss 0.280910 Objective Loss 0.280910 LR 0.001000 Time 0.020006 -2022-12-06 10:37:34,356 - Epoch: [34][ 830/ 1200] Overall Loss 0.280778 Objective Loss 0.280778 LR 0.001000 Time 0.019995 -2022-12-06 10:37:34,548 - Epoch: [34][ 840/ 1200] Overall Loss 0.280883 Objective Loss 0.280883 LR 0.001000 Time 0.019985 -2022-12-06 10:37:34,739 - Epoch: [34][ 850/ 1200] Overall Loss 0.281001 Objective Loss 0.281001 LR 0.001000 Time 0.019973 -2022-12-06 10:37:34,930 - Epoch: [34][ 860/ 1200] Overall Loss 0.280935 Objective Loss 0.280935 LR 0.001000 Time 0.019963 -2022-12-06 10:37:35,122 - Epoch: [34][ 870/ 1200] Overall Loss 0.280942 Objective Loss 0.280942 LR 0.001000 Time 0.019953 -2022-12-06 10:37:35,313 - Epoch: [34][ 880/ 1200] Overall Loss 0.281164 Objective Loss 0.281164 LR 0.001000 Time 0.019943 -2022-12-06 10:37:35,504 - Epoch: [34][ 890/ 1200] Overall Loss 0.281180 Objective Loss 0.281180 LR 0.001000 Time 0.019934 -2022-12-06 10:37:35,696 - Epoch: [34][ 900/ 1200] Overall Loss 0.281064 Objective Loss 0.281064 LR 0.001000 Time 0.019924 -2022-12-06 10:37:35,887 - Epoch: [34][ 910/ 1200] Overall Loss 0.281260 Objective Loss 0.281260 LR 0.001000 Time 0.019915 -2022-12-06 10:37:36,079 - Epoch: [34][ 920/ 1200] Overall Loss 0.281220 Objective Loss 0.281220 LR 0.001000 Time 0.019906 -2022-12-06 10:37:36,270 - Epoch: [34][ 930/ 1200] Overall Loss 0.281266 Objective Loss 0.281266 LR 0.001000 Time 0.019897 -2022-12-06 10:37:36,461 - Epoch: [34][ 940/ 1200] Overall Loss 0.281518 Objective Loss 0.281518 LR 0.001000 Time 0.019888 -2022-12-06 10:37:36,652 - Epoch: [34][ 950/ 1200] Overall Loss 0.281432 Objective Loss 0.281432 LR 0.001000 Time 0.019879 -2022-12-06 10:37:36,844 - Epoch: [34][ 960/ 1200] Overall Loss 0.280933 Objective Loss 0.280933 LR 0.001000 Time 0.019871 -2022-12-06 10:37:37,034 - Epoch: [34][ 970/ 1200] Overall Loss 0.280758 Objective Loss 0.280758 LR 0.001000 Time 0.019862 -2022-12-06 10:37:37,225 - Epoch: [34][ 980/ 1200] Overall Loss 0.280667 Objective Loss 0.280667 LR 0.001000 Time 0.019854 -2022-12-06 10:37:37,416 - Epoch: [34][ 990/ 1200] Overall Loss 0.280839 Objective Loss 0.280839 LR 0.001000 Time 0.019845 -2022-12-06 10:37:37,607 - Epoch: [34][ 1000/ 1200] Overall Loss 0.281009 Objective Loss 0.281009 LR 0.001000 Time 0.019837 -2022-12-06 10:37:37,797 - Epoch: [34][ 1010/ 1200] Overall Loss 0.281170 Objective Loss 0.281170 LR 0.001000 Time 0.019829 -2022-12-06 10:37:37,988 - Epoch: [34][ 1020/ 1200] Overall Loss 0.281010 Objective Loss 0.281010 LR 0.001000 Time 0.019821 -2022-12-06 10:37:38,179 - Epoch: [34][ 1030/ 1200] Overall Loss 0.280954 Objective Loss 0.280954 LR 0.001000 Time 0.019813 -2022-12-06 10:37:38,371 - Epoch: [34][ 1040/ 1200] Overall Loss 0.280935 Objective Loss 0.280935 LR 0.001000 Time 0.019806 -2022-12-06 10:37:38,561 - Epoch: [34][ 1050/ 1200] Overall Loss 0.281039 Objective Loss 0.281039 LR 0.001000 Time 0.019799 -2022-12-06 10:37:38,752 - Epoch: [34][ 1060/ 1200] Overall Loss 0.281387 Objective Loss 0.281387 LR 0.001000 Time 0.019792 -2022-12-06 10:37:38,943 - Epoch: [34][ 1070/ 1200] Overall Loss 0.281200 Objective Loss 0.281200 LR 0.001000 Time 0.019785 -2022-12-06 10:37:39,135 - Epoch: [34][ 1080/ 1200] Overall Loss 0.281514 Objective Loss 0.281514 LR 0.001000 Time 0.019779 -2022-12-06 10:37:39,326 - Epoch: [34][ 1090/ 1200] Overall Loss 0.281816 Objective Loss 0.281816 LR 0.001000 Time 0.019772 -2022-12-06 10:37:39,517 - Epoch: [34][ 1100/ 1200] Overall Loss 0.282148 Objective Loss 0.282148 LR 0.001000 Time 0.019765 -2022-12-06 10:37:39,708 - Epoch: [34][ 1110/ 1200] Overall Loss 0.282321 Objective Loss 0.282321 LR 0.001000 Time 0.019759 -2022-12-06 10:37:39,900 - Epoch: [34][ 1120/ 1200] Overall Loss 0.282575 Objective Loss 0.282575 LR 0.001000 Time 0.019753 -2022-12-06 10:37:40,091 - Epoch: [34][ 1130/ 1200] Overall Loss 0.282713 Objective Loss 0.282713 LR 0.001000 Time 0.019747 -2022-12-06 10:37:40,282 - Epoch: [34][ 1140/ 1200] Overall Loss 0.282564 Objective Loss 0.282564 LR 0.001000 Time 0.019741 -2022-12-06 10:37:40,474 - Epoch: [34][ 1150/ 1200] Overall Loss 0.282764 Objective Loss 0.282764 LR 0.001000 Time 0.019736 -2022-12-06 10:37:40,666 - Epoch: [34][ 1160/ 1200] Overall Loss 0.282780 Objective Loss 0.282780 LR 0.001000 Time 0.019730 -2022-12-06 10:37:40,857 - Epoch: [34][ 1170/ 1200] Overall Loss 0.282868 Objective Loss 0.282868 LR 0.001000 Time 0.019724 -2022-12-06 10:37:41,048 - Epoch: [34][ 1180/ 1200] Overall Loss 0.282692 Objective Loss 0.282692 LR 0.001000 Time 0.019719 -2022-12-06 10:37:41,239 - Epoch: [34][ 1190/ 1200] Overall Loss 0.282813 Objective Loss 0.282813 LR 0.001000 Time 0.019713 -2022-12-06 10:37:41,461 - Epoch: [34][ 1200/ 1200] Overall Loss 0.282587 Objective Loss 0.282587 Top1 88.912134 Top5 98.535565 LR 0.001000 Time 0.019734 -2022-12-06 10:37:41,550 - --- validate (epoch=34)----------- -2022-12-06 10:37:41,550 - 34129 samples (256 per mini-batch) -2022-12-06 10:37:41,987 - Epoch: [34][ 10/ 134] Loss 0.281747 Top1 85.312500 Top5 98.085938 -2022-12-06 10:37:42,114 - Epoch: [34][ 20/ 134] Loss 0.268699 Top1 85.058594 Top5 97.968750 -2022-12-06 10:37:42,240 - Epoch: [34][ 30/ 134] Loss 0.273456 Top1 84.947917 Top5 97.955729 -2022-12-06 10:37:42,366 - Epoch: [34][ 40/ 134] Loss 0.276233 Top1 84.707031 Top5 97.998047 -2022-12-06 10:37:42,494 - Epoch: [34][ 50/ 134] Loss 0.283778 Top1 84.601562 Top5 97.882812 -2022-12-06 10:37:42,620 - Epoch: [34][ 60/ 134] Loss 0.287425 Top1 84.576823 Top5 97.845052 -2022-12-06 10:37:42,748 - Epoch: [34][ 70/ 134] Loss 0.288245 Top1 84.715402 Top5 97.857143 -2022-12-06 10:37:42,875 - Epoch: [34][ 80/ 134] Loss 0.289904 Top1 84.677734 Top5 97.851562 -2022-12-06 10:37:43,001 - Epoch: [34][ 90/ 134] Loss 0.291153 Top1 84.587674 Top5 97.847222 -2022-12-06 10:37:43,126 - Epoch: [34][ 100/ 134] Loss 0.291453 Top1 84.511719 Top5 97.878906 -2022-12-06 10:37:43,250 - Epoch: [34][ 110/ 134] Loss 0.290896 Top1 84.541903 Top5 97.904830 -2022-12-06 10:37:43,379 - Epoch: [34][ 120/ 134] Loss 0.290850 Top1 84.488932 Top5 97.910156 -2022-12-06 10:37:43,507 - Epoch: [34][ 130/ 134] Loss 0.290984 Top1 84.402043 Top5 97.911659 -2022-12-06 10:37:43,544 - Epoch: [34][ 134/ 134] Loss 0.290522 Top1 84.397433 Top5 97.919658 -2022-12-06 10:37:43,633 - ==> Top1: 84.397 Top5: 97.920 Loss: 0.291 - -2022-12-06 10:37:43,634 - ==> Confusion: -[[ 884 2 6 2 9 5 0 0 9 59 1 4 1 1 5 1 0 1 2 1 3] - [ 0 936 2 2 11 19 1 14 1 2 4 3 2 1 0 0 8 3 7 2 9] - [ 5 2 1018 9 2 2 19 7 2 2 4 3 3 1 2 2 1 1 5 3 10] - [ 2 1 29 928 0 5 0 0 0 0 14 1 2 1 11 0 3 4 14 1 4] - [ 8 5 1 2 945 6 2 1 3 5 2 3 0 3 14 4 9 1 0 2 4] - [ 1 12 1 3 11 956 3 19 9 1 1 18 3 14 2 1 3 0 0 6 5] - [ 1 4 19 0 0 5 1055 3 0 0 4 2 3 0 2 4 2 2 1 8 3] - [ 1 16 15 2 2 26 8 928 1 0 3 7 0 2 0 0 0 0 24 15 4] - [ 5 4 0 1 0 3 1 0 969 28 12 4 0 16 13 0 2 1 2 1 2] - [ 70 0 3 0 4 3 0 2 36 849 0 5 0 14 4 1 0 1 0 0 9] - [ 0 1 3 4 0 2 1 1 9 1 956 3 2 16 6 0 1 1 5 4 3] - [ 3 6 5 0 0 12 1 4 2 0 0 955 28 4 1 4 6 3 0 16 1] - [ 0 1 2 4 2 3 1 0 1 0 0 40 882 1 1 6 1 11 1 5 7] - [ 1 2 1 0 1 6 0 2 8 5 11 8 3 959 0 1 6 1 0 4 4] - [ 5 6 5 14 6 0 0 2 18 2 2 2 3 3 1045 0 1 0 9 0 7] - [ 1 2 2 4 3 1 5 1 0 0 0 7 9 3 1 980 8 5 0 4 7] - [ 1 6 2 2 5 2 0 0 2 0 0 5 2 3 1 8 1024 4 0 4 1] - [ 2 2 1 2 0 0 3 0 1 1 1 8 25 2 3 9 3 967 0 4 2] - [ 3 9 8 6 0 2 1 22 2 0 11 3 3 1 10 0 2 1 918 2 4] - [ 1 8 4 0 1 5 5 6 0 0 0 17 6 4 1 2 8 1 1 1005 5] - [ 131 298 224 117 132 223 68 142 87 81 183 128 359 396 151 106 282 56 174 246 9642]] - -2022-12-06 10:37:44,294 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:37:44,294 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:37:44,300 - - -2022-12-06 10:37:44,300 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:37:45,232 - Epoch: [35][ 10/ 1200] Overall Loss 0.292100 Objective Loss 0.292100 LR 0.001000 Time 0.093057 -2022-12-06 10:37:45,426 - Epoch: [35][ 20/ 1200] Overall Loss 0.275881 Objective Loss 0.275881 LR 0.001000 Time 0.056225 -2022-12-06 10:37:45,615 - Epoch: [35][ 30/ 1200] Overall Loss 0.263143 Objective Loss 0.263143 LR 0.001000 Time 0.043762 -2022-12-06 10:37:45,805 - Epoch: [35][ 40/ 1200] Overall Loss 0.267040 Objective Loss 0.267040 LR 0.001000 Time 0.037548 -2022-12-06 10:37:45,994 - Epoch: [35][ 50/ 1200] Overall Loss 0.261540 Objective Loss 0.261540 LR 0.001000 Time 0.033823 -2022-12-06 10:37:46,184 - Epoch: [35][ 60/ 1200] Overall Loss 0.264666 Objective Loss 0.264666 LR 0.001000 Time 0.031341 -2022-12-06 10:37:46,374 - Epoch: [35][ 70/ 1200] Overall Loss 0.264406 Objective Loss 0.264406 LR 0.001000 Time 0.029565 -2022-12-06 10:37:46,564 - Epoch: [35][ 80/ 1200] Overall Loss 0.262506 Objective Loss 0.262506 LR 0.001000 Time 0.028241 -2022-12-06 10:37:46,754 - Epoch: [35][ 90/ 1200] Overall Loss 0.264482 Objective Loss 0.264482 LR 0.001000 Time 0.027200 -2022-12-06 10:37:46,942 - Epoch: [35][ 100/ 1200] Overall Loss 0.266682 Objective Loss 0.266682 LR 0.001000 Time 0.026365 -2022-12-06 10:37:47,132 - Epoch: [35][ 110/ 1200] Overall Loss 0.267111 Objective Loss 0.267111 LR 0.001000 Time 0.025690 -2022-12-06 10:37:47,323 - Epoch: [35][ 120/ 1200] Overall Loss 0.268647 Objective Loss 0.268647 LR 0.001000 Time 0.025128 -2022-12-06 10:37:47,511 - Epoch: [35][ 130/ 1200] Overall Loss 0.269689 Objective Loss 0.269689 LR 0.001000 Time 0.024644 -2022-12-06 10:37:47,700 - Epoch: [35][ 140/ 1200] Overall Loss 0.269127 Objective Loss 0.269127 LR 0.001000 Time 0.024229 -2022-12-06 10:37:47,889 - Epoch: [35][ 150/ 1200] Overall Loss 0.268784 Objective Loss 0.268784 LR 0.001000 Time 0.023871 -2022-12-06 10:37:48,079 - Epoch: [35][ 160/ 1200] Overall Loss 0.267741 Objective Loss 0.267741 LR 0.001000 Time 0.023563 -2022-12-06 10:37:48,268 - Epoch: [35][ 170/ 1200] Overall Loss 0.267672 Objective Loss 0.267672 LR 0.001000 Time 0.023285 -2022-12-06 10:37:48,458 - Epoch: [35][ 180/ 1200] Overall Loss 0.268593 Objective Loss 0.268593 LR 0.001000 Time 0.023043 -2022-12-06 10:37:48,649 - Epoch: [35][ 190/ 1200] Overall Loss 0.268261 Objective Loss 0.268261 LR 0.001000 Time 0.022829 -2022-12-06 10:37:48,839 - Epoch: [35][ 200/ 1200] Overall Loss 0.269211 Objective Loss 0.269211 LR 0.001000 Time 0.022636 -2022-12-06 10:37:49,028 - Epoch: [35][ 210/ 1200] Overall Loss 0.271613 Objective Loss 0.271613 LR 0.001000 Time 0.022455 -2022-12-06 10:37:49,217 - Epoch: [35][ 220/ 1200] Overall Loss 0.271486 Objective Loss 0.271486 LR 0.001000 Time 0.022293 -2022-12-06 10:37:49,407 - Epoch: [35][ 230/ 1200] Overall Loss 0.272293 Objective Loss 0.272293 LR 0.001000 Time 0.022146 -2022-12-06 10:37:49,597 - Epoch: [35][ 240/ 1200] Overall Loss 0.272860 Objective Loss 0.272860 LR 0.001000 Time 0.022013 -2022-12-06 10:37:49,787 - Epoch: [35][ 250/ 1200] Overall Loss 0.272073 Objective Loss 0.272073 LR 0.001000 Time 0.021889 -2022-12-06 10:37:49,977 - Epoch: [35][ 260/ 1200] Overall Loss 0.272949 Objective Loss 0.272949 LR 0.001000 Time 0.021779 -2022-12-06 10:37:50,167 - Epoch: [35][ 270/ 1200] Overall Loss 0.273202 Objective Loss 0.273202 LR 0.001000 Time 0.021672 -2022-12-06 10:37:50,356 - Epoch: [35][ 280/ 1200] Overall Loss 0.274075 Objective Loss 0.274075 LR 0.001000 Time 0.021572 -2022-12-06 10:37:50,546 - Epoch: [35][ 290/ 1200] Overall Loss 0.273322 Objective Loss 0.273322 LR 0.001000 Time 0.021480 -2022-12-06 10:37:50,736 - Epoch: [35][ 300/ 1200] Overall Loss 0.273185 Objective Loss 0.273185 LR 0.001000 Time 0.021397 -2022-12-06 10:37:50,926 - Epoch: [35][ 310/ 1200] Overall Loss 0.273905 Objective Loss 0.273905 LR 0.001000 Time 0.021316 -2022-12-06 10:37:51,116 - Epoch: [35][ 320/ 1200] Overall Loss 0.273044 Objective Loss 0.273044 LR 0.001000 Time 0.021243 -2022-12-06 10:37:51,305 - Epoch: [35][ 330/ 1200] Overall Loss 0.273851 Objective Loss 0.273851 LR 0.001000 Time 0.021170 -2022-12-06 10:37:51,495 - Epoch: [35][ 340/ 1200] Overall Loss 0.274491 Objective Loss 0.274491 LR 0.001000 Time 0.021104 -2022-12-06 10:37:51,685 - Epoch: [35][ 350/ 1200] Overall Loss 0.274869 Objective Loss 0.274869 LR 0.001000 Time 0.021043 -2022-12-06 10:37:51,875 - Epoch: [35][ 360/ 1200] Overall Loss 0.275068 Objective Loss 0.275068 LR 0.001000 Time 0.020985 -2022-12-06 10:37:52,064 - Epoch: [35][ 370/ 1200] Overall Loss 0.275564 Objective Loss 0.275564 LR 0.001000 Time 0.020928 -2022-12-06 10:37:52,255 - Epoch: [35][ 380/ 1200] Overall Loss 0.275286 Objective Loss 0.275286 LR 0.001000 Time 0.020876 -2022-12-06 10:37:52,444 - Epoch: [35][ 390/ 1200] Overall Loss 0.275058 Objective Loss 0.275058 LR 0.001000 Time 0.020825 -2022-12-06 10:37:52,634 - Epoch: [35][ 400/ 1200] Overall Loss 0.274776 Objective Loss 0.274776 LR 0.001000 Time 0.020778 -2022-12-06 10:37:52,823 - Epoch: [35][ 410/ 1200] Overall Loss 0.274399 Objective Loss 0.274399 LR 0.001000 Time 0.020732 -2022-12-06 10:37:53,013 - Epoch: [35][ 420/ 1200] Overall Loss 0.274536 Objective Loss 0.274536 LR 0.001000 Time 0.020688 -2022-12-06 10:37:53,202 - Epoch: [35][ 430/ 1200] Overall Loss 0.274341 Objective Loss 0.274341 LR 0.001000 Time 0.020646 -2022-12-06 10:37:53,392 - Epoch: [35][ 440/ 1200] Overall Loss 0.274306 Objective Loss 0.274306 LR 0.001000 Time 0.020607 -2022-12-06 10:37:53,581 - Epoch: [35][ 450/ 1200] Overall Loss 0.273822 Objective Loss 0.273822 LR 0.001000 Time 0.020569 -2022-12-06 10:37:53,772 - Epoch: [35][ 460/ 1200] Overall Loss 0.274350 Objective Loss 0.274350 LR 0.001000 Time 0.020534 -2022-12-06 10:37:53,961 - Epoch: [35][ 470/ 1200] Overall Loss 0.274316 Objective Loss 0.274316 LR 0.001000 Time 0.020498 -2022-12-06 10:37:54,151 - Epoch: [35][ 480/ 1200] Overall Loss 0.274128 Objective Loss 0.274128 LR 0.001000 Time 0.020468 -2022-12-06 10:37:54,341 - Epoch: [35][ 490/ 1200] Overall Loss 0.273911 Objective Loss 0.273911 LR 0.001000 Time 0.020436 -2022-12-06 10:37:54,531 - Epoch: [35][ 500/ 1200] Overall Loss 0.274641 Objective Loss 0.274641 LR 0.001000 Time 0.020405 -2022-12-06 10:37:54,720 - Epoch: [35][ 510/ 1200] Overall Loss 0.274375 Objective Loss 0.274375 LR 0.001000 Time 0.020374 -2022-12-06 10:37:54,909 - Epoch: [35][ 520/ 1200] Overall Loss 0.274368 Objective Loss 0.274368 LR 0.001000 Time 0.020346 -2022-12-06 10:37:55,099 - Epoch: [35][ 530/ 1200] Overall Loss 0.274570 Objective Loss 0.274570 LR 0.001000 Time 0.020318 -2022-12-06 10:37:55,288 - Epoch: [35][ 540/ 1200] Overall Loss 0.274085 Objective Loss 0.274085 LR 0.001000 Time 0.020292 -2022-12-06 10:37:55,477 - Epoch: [35][ 550/ 1200] Overall Loss 0.273791 Objective Loss 0.273791 LR 0.001000 Time 0.020266 -2022-12-06 10:37:55,667 - Epoch: [35][ 560/ 1200] Overall Loss 0.273912 Objective Loss 0.273912 LR 0.001000 Time 0.020242 -2022-12-06 10:37:55,856 - Epoch: [35][ 570/ 1200] Overall Loss 0.274590 Objective Loss 0.274590 LR 0.001000 Time 0.020217 -2022-12-06 10:37:56,045 - Epoch: [35][ 580/ 1200] Overall Loss 0.274675 Objective Loss 0.274675 LR 0.001000 Time 0.020195 -2022-12-06 10:37:56,235 - Epoch: [35][ 590/ 1200] Overall Loss 0.275141 Objective Loss 0.275141 LR 0.001000 Time 0.020172 -2022-12-06 10:37:56,425 - Epoch: [35][ 600/ 1200] Overall Loss 0.275787 Objective Loss 0.275787 LR 0.001000 Time 0.020152 -2022-12-06 10:37:56,614 - Epoch: [35][ 610/ 1200] Overall Loss 0.276202 Objective Loss 0.276202 LR 0.001000 Time 0.020130 -2022-12-06 10:37:56,803 - Epoch: [35][ 620/ 1200] Overall Loss 0.276098 Objective Loss 0.276098 LR 0.001000 Time 0.020111 -2022-12-06 10:37:56,992 - Epoch: [35][ 630/ 1200] Overall Loss 0.276548 Objective Loss 0.276548 LR 0.001000 Time 0.020090 -2022-12-06 10:37:57,181 - Epoch: [35][ 640/ 1200] Overall Loss 0.276717 Objective Loss 0.276717 LR 0.001000 Time 0.020071 -2022-12-06 10:37:57,371 - Epoch: [35][ 650/ 1200] Overall Loss 0.276344 Objective Loss 0.276344 LR 0.001000 Time 0.020053 -2022-12-06 10:37:57,561 - Epoch: [35][ 660/ 1200] Overall Loss 0.276472 Objective Loss 0.276472 LR 0.001000 Time 0.020036 -2022-12-06 10:37:57,750 - Epoch: [35][ 670/ 1200] Overall Loss 0.276284 Objective Loss 0.276284 LR 0.001000 Time 0.020018 -2022-12-06 10:37:57,939 - Epoch: [35][ 680/ 1200] Overall Loss 0.276779 Objective Loss 0.276779 LR 0.001000 Time 0.020002 -2022-12-06 10:37:58,128 - Epoch: [35][ 690/ 1200] Overall Loss 0.277081 Objective Loss 0.277081 LR 0.001000 Time 0.019985 -2022-12-06 10:37:58,319 - Epoch: [35][ 700/ 1200] Overall Loss 0.277091 Objective Loss 0.277091 LR 0.001000 Time 0.019971 -2022-12-06 10:37:58,508 - Epoch: [35][ 710/ 1200] Overall Loss 0.277420 Objective Loss 0.277420 LR 0.001000 Time 0.019956 -2022-12-06 10:37:58,698 - Epoch: [35][ 720/ 1200] Overall Loss 0.277482 Objective Loss 0.277482 LR 0.001000 Time 0.019941 -2022-12-06 10:37:58,887 - Epoch: [35][ 730/ 1200] Overall Loss 0.277495 Objective Loss 0.277495 LR 0.001000 Time 0.019927 -2022-12-06 10:37:59,077 - Epoch: [35][ 740/ 1200] Overall Loss 0.277272 Objective Loss 0.277272 LR 0.001000 Time 0.019913 -2022-12-06 10:37:59,266 - Epoch: [35][ 750/ 1200] Overall Loss 0.277300 Objective Loss 0.277300 LR 0.001000 Time 0.019899 -2022-12-06 10:37:59,456 - Epoch: [35][ 760/ 1200] Overall Loss 0.277588 Objective Loss 0.277588 LR 0.001000 Time 0.019886 -2022-12-06 10:37:59,646 - Epoch: [35][ 770/ 1200] Overall Loss 0.277951 Objective Loss 0.277951 LR 0.001000 Time 0.019874 -2022-12-06 10:37:59,836 - Epoch: [35][ 780/ 1200] Overall Loss 0.277954 Objective Loss 0.277954 LR 0.001000 Time 0.019862 -2022-12-06 10:38:00,026 - Epoch: [35][ 790/ 1200] Overall Loss 0.278090 Objective Loss 0.278090 LR 0.001000 Time 0.019850 -2022-12-06 10:38:00,215 - Epoch: [35][ 800/ 1200] Overall Loss 0.278192 Objective Loss 0.278192 LR 0.001000 Time 0.019839 -2022-12-06 10:38:00,405 - Epoch: [35][ 810/ 1200] Overall Loss 0.278242 Objective Loss 0.278242 LR 0.001000 Time 0.019827 -2022-12-06 10:38:00,595 - Epoch: [35][ 820/ 1200] Overall Loss 0.278476 Objective Loss 0.278476 LR 0.001000 Time 0.019816 -2022-12-06 10:38:00,785 - Epoch: [35][ 830/ 1200] Overall Loss 0.278410 Objective Loss 0.278410 LR 0.001000 Time 0.019805 -2022-12-06 10:38:00,975 - Epoch: [35][ 840/ 1200] Overall Loss 0.278472 Objective Loss 0.278472 LR 0.001000 Time 0.019795 -2022-12-06 10:38:01,164 - Epoch: [35][ 850/ 1200] Overall Loss 0.278665 Objective Loss 0.278665 LR 0.001000 Time 0.019784 -2022-12-06 10:38:01,354 - Epoch: [35][ 860/ 1200] Overall Loss 0.278640 Objective Loss 0.278640 LR 0.001000 Time 0.019775 -2022-12-06 10:38:01,544 - Epoch: [35][ 870/ 1200] Overall Loss 0.278713 Objective Loss 0.278713 LR 0.001000 Time 0.019765 -2022-12-06 10:38:01,735 - Epoch: [35][ 880/ 1200] Overall Loss 0.278901 Objective Loss 0.278901 LR 0.001000 Time 0.019756 -2022-12-06 10:38:01,925 - Epoch: [35][ 890/ 1200] Overall Loss 0.279107 Objective Loss 0.279107 LR 0.001000 Time 0.019747 -2022-12-06 10:38:02,117 - Epoch: [35][ 900/ 1200] Overall Loss 0.279577 Objective Loss 0.279577 LR 0.001000 Time 0.019741 -2022-12-06 10:38:02,315 - Epoch: [35][ 910/ 1200] Overall Loss 0.279388 Objective Loss 0.279388 LR 0.001000 Time 0.019741 -2022-12-06 10:38:02,510 - Epoch: [35][ 920/ 1200] Overall Loss 0.279364 Objective Loss 0.279364 LR 0.001000 Time 0.019738 -2022-12-06 10:38:02,707 - Epoch: [35][ 930/ 1200] Overall Loss 0.279345 Objective Loss 0.279345 LR 0.001000 Time 0.019737 -2022-12-06 10:38:02,902 - Epoch: [35][ 940/ 1200] Overall Loss 0.279622 Objective Loss 0.279622 LR 0.001000 Time 0.019734 -2022-12-06 10:38:03,100 - Epoch: [35][ 950/ 1200] Overall Loss 0.279642 Objective Loss 0.279642 LR 0.001000 Time 0.019734 -2022-12-06 10:38:03,295 - Epoch: [35][ 960/ 1200] Overall Loss 0.279664 Objective Loss 0.279664 LR 0.001000 Time 0.019731 -2022-12-06 10:38:03,492 - Epoch: [35][ 970/ 1200] Overall Loss 0.279421 Objective Loss 0.279421 LR 0.001000 Time 0.019730 -2022-12-06 10:38:03,687 - Epoch: [35][ 980/ 1200] Overall Loss 0.279047 Objective Loss 0.279047 LR 0.001000 Time 0.019727 -2022-12-06 10:38:03,884 - Epoch: [35][ 990/ 1200] Overall Loss 0.279421 Objective Loss 0.279421 LR 0.001000 Time 0.019727 -2022-12-06 10:38:04,079 - Epoch: [35][ 1000/ 1200] Overall Loss 0.279291 Objective Loss 0.279291 LR 0.001000 Time 0.019724 -2022-12-06 10:38:04,277 - Epoch: [35][ 1010/ 1200] Overall Loss 0.279482 Objective Loss 0.279482 LR 0.001000 Time 0.019724 -2022-12-06 10:38:04,472 - Epoch: [35][ 1020/ 1200] Overall Loss 0.279804 Objective Loss 0.279804 LR 0.001000 Time 0.019721 -2022-12-06 10:38:04,671 - Epoch: [35][ 1030/ 1200] Overall Loss 0.279909 Objective Loss 0.279909 LR 0.001000 Time 0.019722 -2022-12-06 10:38:04,865 - Epoch: [35][ 1040/ 1200] Overall Loss 0.280028 Objective Loss 0.280028 LR 0.001000 Time 0.019719 -2022-12-06 10:38:05,063 - Epoch: [35][ 1050/ 1200] Overall Loss 0.280005 Objective Loss 0.280005 LR 0.001000 Time 0.019719 -2022-12-06 10:38:05,257 - Epoch: [35][ 1060/ 1200] Overall Loss 0.279878 Objective Loss 0.279878 LR 0.001000 Time 0.019716 -2022-12-06 10:38:05,455 - Epoch: [35][ 1070/ 1200] Overall Loss 0.280102 Objective Loss 0.280102 LR 0.001000 Time 0.019716 -2022-12-06 10:38:05,650 - Epoch: [35][ 1080/ 1200] Overall Loss 0.280417 Objective Loss 0.280417 LR 0.001000 Time 0.019713 -2022-12-06 10:38:05,848 - Epoch: [35][ 1090/ 1200] Overall Loss 0.280461 Objective Loss 0.280461 LR 0.001000 Time 0.019713 -2022-12-06 10:38:06,042 - Epoch: [35][ 1100/ 1200] Overall Loss 0.280442 Objective Loss 0.280442 LR 0.001000 Time 0.019710 -2022-12-06 10:38:06,240 - Epoch: [35][ 1110/ 1200] Overall Loss 0.280367 Objective Loss 0.280367 LR 0.001000 Time 0.019710 -2022-12-06 10:38:06,435 - Epoch: [35][ 1120/ 1200] Overall Loss 0.280542 Objective Loss 0.280542 LR 0.001000 Time 0.019708 -2022-12-06 10:38:06,633 - Epoch: [35][ 1130/ 1200] Overall Loss 0.280896 Objective Loss 0.280896 LR 0.001000 Time 0.019708 -2022-12-06 10:38:06,828 - Epoch: [35][ 1140/ 1200] Overall Loss 0.280959 Objective Loss 0.280959 LR 0.001000 Time 0.019706 -2022-12-06 10:38:07,026 - Epoch: [35][ 1150/ 1200] Overall Loss 0.280735 Objective Loss 0.280735 LR 0.001000 Time 0.019707 -2022-12-06 10:38:07,221 - Epoch: [35][ 1160/ 1200] Overall Loss 0.281049 Objective Loss 0.281049 LR 0.001000 Time 0.019704 -2022-12-06 10:38:07,419 - Epoch: [35][ 1170/ 1200] Overall Loss 0.281257 Objective Loss 0.281257 LR 0.001000 Time 0.019704 -2022-12-06 10:38:07,614 - Epoch: [35][ 1180/ 1200] Overall Loss 0.281224 Objective Loss 0.281224 LR 0.001000 Time 0.019702 -2022-12-06 10:38:07,811 - Epoch: [35][ 1190/ 1200] Overall Loss 0.281288 Objective Loss 0.281288 LR 0.001000 Time 0.019702 -2022-12-06 10:38:08,043 - Epoch: [35][ 1200/ 1200] Overall Loss 0.281086 Objective Loss 0.281086 Top1 80.753138 Top5 97.489540 LR 0.001000 Time 0.019731 -2022-12-06 10:38:08,131 - --- validate (epoch=35)----------- -2022-12-06 10:38:08,131 - 34129 samples (256 per mini-batch) -2022-12-06 10:38:08,592 - Epoch: [35][ 10/ 134] Loss 0.313208 Top1 83.906250 Top5 97.617188 -2022-12-06 10:38:08,727 - Epoch: [35][ 20/ 134] Loss 0.311442 Top1 84.140625 Top5 97.832031 -2022-12-06 10:38:08,860 - Epoch: [35][ 30/ 134] Loss 0.317022 Top1 83.984375 Top5 97.903646 -2022-12-06 10:38:08,992 - Epoch: [35][ 40/ 134] Loss 0.308895 Top1 84.345703 Top5 97.910156 -2022-12-06 10:38:09,125 - Epoch: [35][ 50/ 134] Loss 0.312067 Top1 84.156250 Top5 97.851562 -2022-12-06 10:38:09,258 - Epoch: [35][ 60/ 134] Loss 0.307030 Top1 84.160156 Top5 97.871094 -2022-12-06 10:38:09,390 - Epoch: [35][ 70/ 134] Loss 0.303897 Top1 84.162946 Top5 97.907366 -2022-12-06 10:38:09,525 - Epoch: [35][ 80/ 134] Loss 0.304156 Top1 84.135742 Top5 97.866211 -2022-12-06 10:38:09,658 - Epoch: [35][ 90/ 134] Loss 0.305635 Top1 84.092882 Top5 97.847222 -2022-12-06 10:38:09,791 - Epoch: [35][ 100/ 134] Loss 0.302265 Top1 84.003906 Top5 97.843750 -2022-12-06 10:38:09,925 - Epoch: [35][ 110/ 134] Loss 0.299585 Top1 83.995028 Top5 97.855114 -2022-12-06 10:38:10,058 - Epoch: [35][ 120/ 134] Loss 0.301155 Top1 83.919271 Top5 97.845052 -2022-12-06 10:38:10,192 - Epoch: [35][ 130/ 134] Loss 0.300717 Top1 83.897236 Top5 97.851562 -2022-12-06 10:38:10,233 - Epoch: [35][ 134/ 134] Loss 0.300141 Top1 83.925694 Top5 97.861057 -2022-12-06 10:38:10,320 - ==> Top1: 83.926 Top5: 97.861 Loss: 0.300 - -2022-12-06 10:38:10,321 - ==> Confusion: -[[ 843 4 4 1 5 5 0 1 10 97 0 2 3 2 5 2 2 2 3 1 4] - [ 1 917 1 3 10 17 5 20 1 1 7 4 2 5 1 1 7 1 14 3 6] - [ 4 3 989 23 1 4 27 13 0 2 7 5 1 1 2 3 1 0 8 2 7] - [ 2 1 14 937 0 3 0 0 1 0 14 0 3 3 21 0 3 2 13 0 3] - [ 8 4 1 0 943 5 2 1 1 8 1 5 0 4 11 8 8 3 0 3 4] - [ 0 18 1 1 8 967 1 20 5 0 2 15 2 9 1 0 2 2 3 6 6] - [ 0 0 12 3 0 1 1069 1 0 0 7 4 2 2 0 7 0 0 3 3 4] - [ 1 5 11 1 1 32 9 921 1 1 6 4 0 2 0 2 0 2 36 14 5] - [ 1 4 0 0 1 4 0 1 967 40 11 1 3 5 17 1 2 0 4 1 1] - [ 34 0 2 0 4 2 1 4 25 899 2 3 0 13 2 0 1 1 1 0 7] - [ 0 1 2 5 0 3 2 2 12 4 956 2 1 10 8 0 0 0 10 1 0] - [ 4 1 4 0 0 17 2 4 2 1 2 943 25 8 0 9 2 4 2 19 2] - [ 0 0 2 5 0 4 2 2 2 0 1 37 876 2 1 7 3 11 2 6 6] - [ 0 1 0 1 1 12 0 4 16 14 15 4 2 933 2 1 2 1 1 2 11] - [ 9 3 3 12 2 2 1 0 25 0 0 1 2 3 1044 1 2 0 13 2 5] - [ 1 0 3 0 3 1 6 1 0 0 2 5 6 5 0 983 7 11 1 4 4] - [ 4 2 2 0 2 0 2 0 4 0 0 4 0 3 2 10 1020 1 1 8 7] - [ 3 1 1 6 0 1 3 0 3 1 2 13 21 5 1 15 2 951 3 2 2] - [ 3 3 3 8 0 4 2 25 2 0 7 1 2 0 12 1 1 0 927 2 5] - [ 2 2 1 0 1 7 8 9 2 0 1 11 7 4 1 1 5 3 1 1012 2] - [ 96 206 192 163 92 192 88 173 92 110 215 123 400 296 202 126 266 65 247 342 9540]] - -2022-12-06 10:38:10,996 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:38:10,996 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:38:11,002 - - -2022-12-06 10:38:11,002 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:38:11,943 - Epoch: [36][ 10/ 1200] Overall Loss 0.263591 Objective Loss 0.263591 LR 0.001000 Time 0.094018 -2022-12-06 10:38:12,137 - Epoch: [36][ 20/ 1200] Overall Loss 0.263255 Objective Loss 0.263255 LR 0.001000 Time 0.056691 -2022-12-06 10:38:12,331 - Epoch: [36][ 30/ 1200] Overall Loss 0.262813 Objective Loss 0.262813 LR 0.001000 Time 0.044258 -2022-12-06 10:38:12,525 - Epoch: [36][ 40/ 1200] Overall Loss 0.260725 Objective Loss 0.260725 LR 0.001000 Time 0.038010 -2022-12-06 10:38:12,717 - Epoch: [36][ 50/ 1200] Overall Loss 0.259488 Objective Loss 0.259488 LR 0.001000 Time 0.034248 -2022-12-06 10:38:12,910 - Epoch: [36][ 60/ 1200] Overall Loss 0.257704 Objective Loss 0.257704 LR 0.001000 Time 0.031749 -2022-12-06 10:38:13,103 - Epoch: [36][ 70/ 1200] Overall Loss 0.260694 Objective Loss 0.260694 LR 0.001000 Time 0.029960 -2022-12-06 10:38:13,296 - Epoch: [36][ 80/ 1200] Overall Loss 0.262304 Objective Loss 0.262304 LR 0.001000 Time 0.028619 -2022-12-06 10:38:13,488 - Epoch: [36][ 90/ 1200] Overall Loss 0.262926 Objective Loss 0.262926 LR 0.001000 Time 0.027571 -2022-12-06 10:38:13,681 - Epoch: [36][ 100/ 1200] Overall Loss 0.263717 Objective Loss 0.263717 LR 0.001000 Time 0.026730 -2022-12-06 10:38:13,872 - Epoch: [36][ 110/ 1200] Overall Loss 0.263431 Objective Loss 0.263431 LR 0.001000 Time 0.026037 -2022-12-06 10:38:14,065 - Epoch: [36][ 120/ 1200] Overall Loss 0.263947 Objective Loss 0.263947 LR 0.001000 Time 0.025465 -2022-12-06 10:38:14,257 - Epoch: [36][ 130/ 1200] Overall Loss 0.265498 Objective Loss 0.265498 LR 0.001000 Time 0.024979 -2022-12-06 10:38:14,449 - Epoch: [36][ 140/ 1200] Overall Loss 0.264834 Objective Loss 0.264834 LR 0.001000 Time 0.024565 -2022-12-06 10:38:14,641 - Epoch: [36][ 150/ 1200] Overall Loss 0.264166 Objective Loss 0.264166 LR 0.001000 Time 0.024206 -2022-12-06 10:38:14,833 - Epoch: [36][ 160/ 1200] Overall Loss 0.263989 Objective Loss 0.263989 LR 0.001000 Time 0.023890 -2022-12-06 10:38:15,026 - Epoch: [36][ 170/ 1200] Overall Loss 0.262932 Objective Loss 0.262932 LR 0.001000 Time 0.023614 -2022-12-06 10:38:15,217 - Epoch: [36][ 180/ 1200] Overall Loss 0.263970 Objective Loss 0.263970 LR 0.001000 Time 0.023363 -2022-12-06 10:38:15,411 - Epoch: [36][ 190/ 1200] Overall Loss 0.264444 Objective Loss 0.264444 LR 0.001000 Time 0.023149 -2022-12-06 10:38:15,603 - Epoch: [36][ 200/ 1200] Overall Loss 0.264624 Objective Loss 0.264624 LR 0.001000 Time 0.022951 -2022-12-06 10:38:15,795 - Epoch: [36][ 210/ 1200] Overall Loss 0.264843 Objective Loss 0.264843 LR 0.001000 Time 0.022770 -2022-12-06 10:38:15,988 - Epoch: [36][ 220/ 1200] Overall Loss 0.266144 Objective Loss 0.266144 LR 0.001000 Time 0.022607 -2022-12-06 10:38:16,180 - Epoch: [36][ 230/ 1200] Overall Loss 0.266799 Objective Loss 0.266799 LR 0.001000 Time 0.022459 -2022-12-06 10:38:16,373 - Epoch: [36][ 240/ 1200] Overall Loss 0.267583 Objective Loss 0.267583 LR 0.001000 Time 0.022325 -2022-12-06 10:38:16,566 - Epoch: [36][ 250/ 1200] Overall Loss 0.266815 Objective Loss 0.266815 LR 0.001000 Time 0.022199 -2022-12-06 10:38:16,758 - Epoch: [36][ 260/ 1200] Overall Loss 0.266260 Objective Loss 0.266260 LR 0.001000 Time 0.022084 -2022-12-06 10:38:16,951 - Epoch: [36][ 270/ 1200] Overall Loss 0.266572 Objective Loss 0.266572 LR 0.001000 Time 0.021977 -2022-12-06 10:38:17,143 - Epoch: [36][ 280/ 1200] Overall Loss 0.266867 Objective Loss 0.266867 LR 0.001000 Time 0.021876 -2022-12-06 10:38:17,335 - Epoch: [36][ 290/ 1200] Overall Loss 0.266433 Objective Loss 0.266433 LR 0.001000 Time 0.021782 -2022-12-06 10:38:17,527 - Epoch: [36][ 300/ 1200] Overall Loss 0.266084 Objective Loss 0.266084 LR 0.001000 Time 0.021693 -2022-12-06 10:38:17,719 - Epoch: [36][ 310/ 1200] Overall Loss 0.266738 Objective Loss 0.266738 LR 0.001000 Time 0.021612 -2022-12-06 10:38:17,911 - Epoch: [36][ 320/ 1200] Overall Loss 0.267952 Objective Loss 0.267952 LR 0.001000 Time 0.021535 -2022-12-06 10:38:18,103 - Epoch: [36][ 330/ 1200] Overall Loss 0.268000 Objective Loss 0.268000 LR 0.001000 Time 0.021463 -2022-12-06 10:38:18,295 - Epoch: [36][ 340/ 1200] Overall Loss 0.268303 Objective Loss 0.268303 LR 0.001000 Time 0.021395 -2022-12-06 10:38:18,488 - Epoch: [36][ 350/ 1200] Overall Loss 0.268352 Objective Loss 0.268352 LR 0.001000 Time 0.021332 -2022-12-06 10:38:18,680 - Epoch: [36][ 360/ 1200] Overall Loss 0.269636 Objective Loss 0.269636 LR 0.001000 Time 0.021272 -2022-12-06 10:38:18,871 - Epoch: [36][ 370/ 1200] Overall Loss 0.269504 Objective Loss 0.269504 LR 0.001000 Time 0.021210 -2022-12-06 10:38:19,061 - Epoch: [36][ 380/ 1200] Overall Loss 0.270554 Objective Loss 0.270554 LR 0.001000 Time 0.021153 -2022-12-06 10:38:19,252 - Epoch: [36][ 390/ 1200] Overall Loss 0.270837 Objective Loss 0.270837 LR 0.001000 Time 0.021098 -2022-12-06 10:38:19,443 - Epoch: [36][ 400/ 1200] Overall Loss 0.271413 Objective Loss 0.271413 LR 0.001000 Time 0.021046 -2022-12-06 10:38:19,633 - Epoch: [36][ 410/ 1200] Overall Loss 0.271924 Objective Loss 0.271924 LR 0.001000 Time 0.020996 -2022-12-06 10:38:19,824 - Epoch: [36][ 420/ 1200] Overall Loss 0.271777 Objective Loss 0.271777 LR 0.001000 Time 0.020949 -2022-12-06 10:38:20,015 - Epoch: [36][ 430/ 1200] Overall Loss 0.272103 Objective Loss 0.272103 LR 0.001000 Time 0.020905 -2022-12-06 10:38:20,207 - Epoch: [36][ 440/ 1200] Overall Loss 0.272160 Objective Loss 0.272160 LR 0.001000 Time 0.020864 -2022-12-06 10:38:20,397 - Epoch: [36][ 450/ 1200] Overall Loss 0.272633 Objective Loss 0.272633 LR 0.001000 Time 0.020823 -2022-12-06 10:38:20,588 - Epoch: [36][ 460/ 1200] Overall Loss 0.272639 Objective Loss 0.272639 LR 0.001000 Time 0.020784 -2022-12-06 10:38:20,779 - Epoch: [36][ 470/ 1200] Overall Loss 0.273006 Objective Loss 0.273006 LR 0.001000 Time 0.020748 -2022-12-06 10:38:20,970 - Epoch: [36][ 480/ 1200] Overall Loss 0.273001 Objective Loss 0.273001 LR 0.001000 Time 0.020711 -2022-12-06 10:38:21,161 - Epoch: [36][ 490/ 1200] Overall Loss 0.273620 Objective Loss 0.273620 LR 0.001000 Time 0.020678 -2022-12-06 10:38:21,352 - Epoch: [36][ 500/ 1200] Overall Loss 0.273739 Objective Loss 0.273739 LR 0.001000 Time 0.020644 -2022-12-06 10:38:21,542 - Epoch: [36][ 510/ 1200] Overall Loss 0.273666 Objective Loss 0.273666 LR 0.001000 Time 0.020612 -2022-12-06 10:38:21,733 - Epoch: [36][ 520/ 1200] Overall Loss 0.273557 Objective Loss 0.273557 LR 0.001000 Time 0.020581 -2022-12-06 10:38:21,924 - Epoch: [36][ 530/ 1200] Overall Loss 0.273302 Objective Loss 0.273302 LR 0.001000 Time 0.020553 -2022-12-06 10:38:22,116 - Epoch: [36][ 540/ 1200] Overall Loss 0.273040 Objective Loss 0.273040 LR 0.001000 Time 0.020525 -2022-12-06 10:38:22,306 - Epoch: [36][ 550/ 1200] Overall Loss 0.273107 Objective Loss 0.273107 LR 0.001000 Time 0.020497 -2022-12-06 10:38:22,497 - Epoch: [36][ 560/ 1200] Overall Loss 0.272739 Objective Loss 0.272739 LR 0.001000 Time 0.020470 -2022-12-06 10:38:22,687 - Epoch: [36][ 570/ 1200] Overall Loss 0.272516 Objective Loss 0.272516 LR 0.001000 Time 0.020445 -2022-12-06 10:38:22,879 - Epoch: [36][ 580/ 1200] Overall Loss 0.273061 Objective Loss 0.273061 LR 0.001000 Time 0.020421 -2022-12-06 10:38:23,070 - Epoch: [36][ 590/ 1200] Overall Loss 0.272891 Objective Loss 0.272891 LR 0.001000 Time 0.020398 -2022-12-06 10:38:23,261 - Epoch: [36][ 600/ 1200] Overall Loss 0.273335 Objective Loss 0.273335 LR 0.001000 Time 0.020376 -2022-12-06 10:38:23,451 - Epoch: [36][ 610/ 1200] Overall Loss 0.273529 Objective Loss 0.273529 LR 0.001000 Time 0.020353 -2022-12-06 10:38:23,641 - Epoch: [36][ 620/ 1200] Overall Loss 0.273546 Objective Loss 0.273546 LR 0.001000 Time 0.020330 -2022-12-06 10:38:23,833 - Epoch: [36][ 630/ 1200] Overall Loss 0.273446 Objective Loss 0.273446 LR 0.001000 Time 0.020311 -2022-12-06 10:38:24,023 - Epoch: [36][ 640/ 1200] Overall Loss 0.274284 Objective Loss 0.274284 LR 0.001000 Time 0.020289 -2022-12-06 10:38:24,213 - Epoch: [36][ 650/ 1200] Overall Loss 0.274646 Objective Loss 0.274646 LR 0.001000 Time 0.020269 -2022-12-06 10:38:24,404 - Epoch: [36][ 660/ 1200] Overall Loss 0.274848 Objective Loss 0.274848 LR 0.001000 Time 0.020250 -2022-12-06 10:38:24,595 - Epoch: [36][ 670/ 1200] Overall Loss 0.274814 Objective Loss 0.274814 LR 0.001000 Time 0.020232 -2022-12-06 10:38:24,785 - Epoch: [36][ 680/ 1200] Overall Loss 0.274952 Objective Loss 0.274952 LR 0.001000 Time 0.020214 -2022-12-06 10:38:24,976 - Epoch: [36][ 690/ 1200] Overall Loss 0.275083 Objective Loss 0.275083 LR 0.001000 Time 0.020197 -2022-12-06 10:38:25,168 - Epoch: [36][ 700/ 1200] Overall Loss 0.275024 Objective Loss 0.275024 LR 0.001000 Time 0.020181 -2022-12-06 10:38:25,359 - Epoch: [36][ 710/ 1200] Overall Loss 0.274973 Objective Loss 0.274973 LR 0.001000 Time 0.020165 -2022-12-06 10:38:25,549 - Epoch: [36][ 720/ 1200] Overall Loss 0.275586 Objective Loss 0.275586 LR 0.001000 Time 0.020149 -2022-12-06 10:38:25,740 - Epoch: [36][ 730/ 1200] Overall Loss 0.275635 Objective Loss 0.275635 LR 0.001000 Time 0.020133 -2022-12-06 10:38:25,931 - Epoch: [36][ 740/ 1200] Overall Loss 0.275992 Objective Loss 0.275992 LR 0.001000 Time 0.020118 -2022-12-06 10:38:26,121 - Epoch: [36][ 750/ 1200] Overall Loss 0.276310 Objective Loss 0.276310 LR 0.001000 Time 0.020104 -2022-12-06 10:38:26,312 - Epoch: [36][ 760/ 1200] Overall Loss 0.276478 Objective Loss 0.276478 LR 0.001000 Time 0.020090 -2022-12-06 10:38:26,503 - Epoch: [36][ 770/ 1200] Overall Loss 0.276362 Objective Loss 0.276362 LR 0.001000 Time 0.020076 -2022-12-06 10:38:26,694 - Epoch: [36][ 780/ 1200] Overall Loss 0.276663 Objective Loss 0.276663 LR 0.001000 Time 0.020062 -2022-12-06 10:38:26,884 - Epoch: [36][ 790/ 1200] Overall Loss 0.276434 Objective Loss 0.276434 LR 0.001000 Time 0.020049 -2022-12-06 10:38:27,076 - Epoch: [36][ 800/ 1200] Overall Loss 0.276909 Objective Loss 0.276909 LR 0.001000 Time 0.020037 -2022-12-06 10:38:27,267 - Epoch: [36][ 810/ 1200] Overall Loss 0.277064 Objective Loss 0.277064 LR 0.001000 Time 0.020024 -2022-12-06 10:38:27,457 - Epoch: [36][ 820/ 1200] Overall Loss 0.277136 Objective Loss 0.277136 LR 0.001000 Time 0.020012 -2022-12-06 10:38:27,649 - Epoch: [36][ 830/ 1200] Overall Loss 0.277176 Objective Loss 0.277176 LR 0.001000 Time 0.020001 -2022-12-06 10:38:27,842 - Epoch: [36][ 840/ 1200] Overall Loss 0.277137 Objective Loss 0.277137 LR 0.001000 Time 0.019992 -2022-12-06 10:38:28,034 - Epoch: [36][ 850/ 1200] Overall Loss 0.277453 Objective Loss 0.277453 LR 0.001000 Time 0.019982 -2022-12-06 10:38:28,226 - Epoch: [36][ 860/ 1200] Overall Loss 0.277724 Objective Loss 0.277724 LR 0.001000 Time 0.019973 -2022-12-06 10:38:28,418 - Epoch: [36][ 870/ 1200] Overall Loss 0.278051 Objective Loss 0.278051 LR 0.001000 Time 0.019964 -2022-12-06 10:38:28,611 - Epoch: [36][ 880/ 1200] Overall Loss 0.278442 Objective Loss 0.278442 LR 0.001000 Time 0.019955 -2022-12-06 10:38:28,803 - Epoch: [36][ 890/ 1200] Overall Loss 0.278769 Objective Loss 0.278769 LR 0.001000 Time 0.019946 -2022-12-06 10:38:28,995 - Epoch: [36][ 900/ 1200] Overall Loss 0.279044 Objective Loss 0.279044 LR 0.001000 Time 0.019937 -2022-12-06 10:38:29,187 - Epoch: [36][ 910/ 1200] Overall Loss 0.279270 Objective Loss 0.279270 LR 0.001000 Time 0.019928 -2022-12-06 10:38:29,379 - Epoch: [36][ 920/ 1200] Overall Loss 0.279521 Objective Loss 0.279521 LR 0.001000 Time 0.019920 -2022-12-06 10:38:29,571 - Epoch: [36][ 930/ 1200] Overall Loss 0.279761 Objective Loss 0.279761 LR 0.001000 Time 0.019912 -2022-12-06 10:38:29,764 - Epoch: [36][ 940/ 1200] Overall Loss 0.279807 Objective Loss 0.279807 LR 0.001000 Time 0.019904 -2022-12-06 10:38:29,955 - Epoch: [36][ 950/ 1200] Overall Loss 0.279878 Objective Loss 0.279878 LR 0.001000 Time 0.019895 -2022-12-06 10:38:30,147 - Epoch: [36][ 960/ 1200] Overall Loss 0.279831 Objective Loss 0.279831 LR 0.001000 Time 0.019888 -2022-12-06 10:38:30,339 - Epoch: [36][ 970/ 1200] Overall Loss 0.279982 Objective Loss 0.279982 LR 0.001000 Time 0.019880 -2022-12-06 10:38:30,531 - Epoch: [36][ 980/ 1200] Overall Loss 0.279913 Objective Loss 0.279913 LR 0.001000 Time 0.019872 -2022-12-06 10:38:30,723 - Epoch: [36][ 990/ 1200] Overall Loss 0.280297 Objective Loss 0.280297 LR 0.001000 Time 0.019865 -2022-12-06 10:38:30,915 - Epoch: [36][ 1000/ 1200] Overall Loss 0.280280 Objective Loss 0.280280 LR 0.001000 Time 0.019858 -2022-12-06 10:38:31,108 - Epoch: [36][ 1010/ 1200] Overall Loss 0.280554 Objective Loss 0.280554 LR 0.001000 Time 0.019852 -2022-12-06 10:38:31,300 - Epoch: [36][ 1020/ 1200] Overall Loss 0.280643 Objective Loss 0.280643 LR 0.001000 Time 0.019845 -2022-12-06 10:38:31,492 - Epoch: [36][ 1030/ 1200] Overall Loss 0.280685 Objective Loss 0.280685 LR 0.001000 Time 0.019838 -2022-12-06 10:38:31,684 - Epoch: [36][ 1040/ 1200] Overall Loss 0.280847 Objective Loss 0.280847 LR 0.001000 Time 0.019831 -2022-12-06 10:38:31,876 - Epoch: [36][ 1050/ 1200] Overall Loss 0.280880 Objective Loss 0.280880 LR 0.001000 Time 0.019825 -2022-12-06 10:38:32,068 - Epoch: [36][ 1060/ 1200] Overall Loss 0.281306 Objective Loss 0.281306 LR 0.001000 Time 0.019818 -2022-12-06 10:38:32,260 - Epoch: [36][ 1070/ 1200] Overall Loss 0.281073 Objective Loss 0.281073 LR 0.001000 Time 0.019812 -2022-12-06 10:38:32,453 - Epoch: [36][ 1080/ 1200] Overall Loss 0.280984 Objective Loss 0.280984 LR 0.001000 Time 0.019806 -2022-12-06 10:38:32,645 - Epoch: [36][ 1090/ 1200] Overall Loss 0.280865 Objective Loss 0.280865 LR 0.001000 Time 0.019801 -2022-12-06 10:38:32,837 - Epoch: [36][ 1100/ 1200] Overall Loss 0.280778 Objective Loss 0.280778 LR 0.001000 Time 0.019795 -2022-12-06 10:38:33,030 - Epoch: [36][ 1110/ 1200] Overall Loss 0.280953 Objective Loss 0.280953 LR 0.001000 Time 0.019790 -2022-12-06 10:38:33,222 - Epoch: [36][ 1120/ 1200] Overall Loss 0.281055 Objective Loss 0.281055 LR 0.001000 Time 0.019784 -2022-12-06 10:38:33,415 - Epoch: [36][ 1130/ 1200] Overall Loss 0.281258 Objective Loss 0.281258 LR 0.001000 Time 0.019779 -2022-12-06 10:38:33,608 - Epoch: [36][ 1140/ 1200] Overall Loss 0.281607 Objective Loss 0.281607 LR 0.001000 Time 0.019774 -2022-12-06 10:38:33,800 - Epoch: [36][ 1150/ 1200] Overall Loss 0.281698 Objective Loss 0.281698 LR 0.001000 Time 0.019769 -2022-12-06 10:38:33,992 - Epoch: [36][ 1160/ 1200] Overall Loss 0.281977 Objective Loss 0.281977 LR 0.001000 Time 0.019764 -2022-12-06 10:38:34,185 - Epoch: [36][ 1170/ 1200] Overall Loss 0.282241 Objective Loss 0.282241 LR 0.001000 Time 0.019759 -2022-12-06 10:38:34,377 - Epoch: [36][ 1180/ 1200] Overall Loss 0.282237 Objective Loss 0.282237 LR 0.001000 Time 0.019754 -2022-12-06 10:38:34,570 - Epoch: [36][ 1190/ 1200] Overall Loss 0.282236 Objective Loss 0.282236 LR 0.001000 Time 0.019749 -2022-12-06 10:38:34,805 - Epoch: [36][ 1200/ 1200] Overall Loss 0.282387 Objective Loss 0.282387 Top1 85.146444 Top5 98.535565 LR 0.001000 Time 0.019781 -2022-12-06 10:38:34,894 - --- validate (epoch=36)----------- -2022-12-06 10:38:34,894 - 34129 samples (256 per mini-batch) -2022-12-06 10:38:35,337 - Epoch: [36][ 10/ 134] Loss 0.302831 Top1 83.906250 Top5 97.890625 -2022-12-06 10:38:35,469 - Epoch: [36][ 20/ 134] Loss 0.289137 Top1 83.906250 Top5 97.695312 -2022-12-06 10:38:35,599 - Epoch: [36][ 30/ 134] Loss 0.289787 Top1 83.606771 Top5 97.903646 -2022-12-06 10:38:35,731 - Epoch: [36][ 40/ 134] Loss 0.298638 Top1 83.417969 Top5 97.753906 -2022-12-06 10:38:35,864 - Epoch: [36][ 50/ 134] Loss 0.305127 Top1 83.375000 Top5 97.679688 -2022-12-06 10:38:35,995 - Epoch: [36][ 60/ 134] Loss 0.308990 Top1 83.274740 Top5 97.727865 -2022-12-06 10:38:36,125 - Epoch: [36][ 70/ 134] Loss 0.308091 Top1 83.376116 Top5 97.700893 -2022-12-06 10:38:36,256 - Epoch: [36][ 80/ 134] Loss 0.306080 Top1 83.413086 Top5 97.729492 -2022-12-06 10:38:36,386 - Epoch: [36][ 90/ 134] Loss 0.303913 Top1 83.411458 Top5 97.747396 -2022-12-06 10:38:36,517 - Epoch: [36][ 100/ 134] Loss 0.306843 Top1 83.363281 Top5 97.703125 -2022-12-06 10:38:36,649 - Epoch: [36][ 110/ 134] Loss 0.307317 Top1 83.338068 Top5 97.634943 -2022-12-06 10:38:36,780 - Epoch: [36][ 120/ 134] Loss 0.308207 Top1 83.336589 Top5 97.626953 -2022-12-06 10:38:36,912 - Epoch: [36][ 130/ 134] Loss 0.307500 Top1 83.362380 Top5 97.638221 -2022-12-06 10:38:36,948 - Epoch: [36][ 134/ 134] Loss 0.305975 Top1 83.377773 Top5 97.635442 -2022-12-06 10:38:37,035 - ==> Top1: 83.378 Top5: 97.635 Loss: 0.306 - -2022-12-06 10:38:37,036 - ==> Confusion: -[[ 850 0 6 3 6 7 1 6 6 92 0 4 2 1 2 2 2 1 0 1 4] - [ 1 887 1 2 7 43 5 31 1 1 8 4 4 1 0 1 2 4 14 3 7] - [ 5 2 959 20 2 3 55 14 1 1 5 6 0 2 1 6 0 2 8 4 7] - [ 2 2 16 950 0 3 2 2 0 0 9 2 5 2 5 1 0 4 9 0 6] - [ 11 4 2 0 930 5 2 1 0 9 2 6 2 4 10 7 10 4 0 1 10] - [ 2 11 1 3 5 954 7 36 3 0 1 11 5 11 2 1 1 0 2 5 8] - [ 1 1 3 2 0 0 1082 4 0 0 3 2 0 0 0 7 0 1 0 8 4] - [ 0 4 5 2 1 25 9 958 0 1 2 5 3 1 0 3 0 0 20 10 5] - [ 3 3 0 2 1 3 1 4 949 44 11 1 4 16 9 1 3 0 4 2 3] - [ 49 0 1 0 1 3 2 1 15 907 1 3 1 10 1 0 1 1 0 0 4] - [ 0 2 4 10 0 2 3 3 8 1 951 2 2 12 3 0 0 0 9 1 6] - [ 2 0 3 0 0 15 5 3 0 1 0 951 34 8 2 3 1 6 1 14 2] - [ 0 1 4 5 0 1 0 1 0 0 1 29 888 0 0 15 2 10 0 3 9] - [ 0 2 1 0 0 11 1 3 6 18 10 14 7 937 0 2 3 0 0 2 6] - [ 7 2 0 32 3 4 1 1 24 11 2 2 2 8 1000 0 3 5 8 3 12] - [ 0 2 3 0 1 1 5 0 0 0 1 12 6 1 0 987 8 10 0 2 4] - [ 0 2 1 3 2 1 2 0 1 1 0 6 1 0 0 10 1024 3 1 9 5] - [ 1 0 1 2 0 0 5 1 0 1 0 8 32 1 0 15 0 964 1 2 2] - [ 2 2 6 16 0 4 1 32 1 1 7 1 2 1 4 0 0 1 919 4 4] - [ 2 2 3 0 0 7 12 4 0 0 2 13 8 2 1 1 0 1 2 1015 5] - [ 101 167 164 160 104 236 162 240 78 153 177 127 466 366 107 188 213 92 180 355 9390]] - -2022-12-06 10:38:37,694 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:38:37,694 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:38:37,700 - - -2022-12-06 10:38:37,700 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:38:38,621 - Epoch: [37][ 10/ 1200] Overall Loss 0.300464 Objective Loss 0.300464 LR 0.001000 Time 0.092037 -2022-12-06 10:38:38,814 - Epoch: [37][ 20/ 1200] Overall Loss 0.280088 Objective Loss 0.280088 LR 0.001000 Time 0.055642 -2022-12-06 10:38:39,006 - Epoch: [37][ 30/ 1200] Overall Loss 0.281319 Objective Loss 0.281319 LR 0.001000 Time 0.043476 -2022-12-06 10:38:39,198 - Epoch: [37][ 40/ 1200] Overall Loss 0.278042 Objective Loss 0.278042 LR 0.001000 Time 0.037393 -2022-12-06 10:38:39,390 - Epoch: [37][ 50/ 1200] Overall Loss 0.273737 Objective Loss 0.273737 LR 0.001000 Time 0.033733 -2022-12-06 10:38:39,581 - Epoch: [37][ 60/ 1200] Overall Loss 0.268462 Objective Loss 0.268462 LR 0.001000 Time 0.031295 -2022-12-06 10:38:39,772 - Epoch: [37][ 70/ 1200] Overall Loss 0.265543 Objective Loss 0.265543 LR 0.001000 Time 0.029538 -2022-12-06 10:38:39,963 - Epoch: [37][ 80/ 1200] Overall Loss 0.267166 Objective Loss 0.267166 LR 0.001000 Time 0.028236 -2022-12-06 10:38:40,155 - Epoch: [37][ 90/ 1200] Overall Loss 0.266760 Objective Loss 0.266760 LR 0.001000 Time 0.027219 -2022-12-06 10:38:40,346 - Epoch: [37][ 100/ 1200] Overall Loss 0.269342 Objective Loss 0.269342 LR 0.001000 Time 0.026401 -2022-12-06 10:38:40,537 - Epoch: [37][ 110/ 1200] Overall Loss 0.267706 Objective Loss 0.267706 LR 0.001000 Time 0.025734 -2022-12-06 10:38:40,728 - Epoch: [37][ 120/ 1200] Overall Loss 0.266678 Objective Loss 0.266678 LR 0.001000 Time 0.025179 -2022-12-06 10:38:40,920 - Epoch: [37][ 130/ 1200] Overall Loss 0.266528 Objective Loss 0.266528 LR 0.001000 Time 0.024711 -2022-12-06 10:38:41,112 - Epoch: [37][ 140/ 1200] Overall Loss 0.266458 Objective Loss 0.266458 LR 0.001000 Time 0.024313 -2022-12-06 10:38:41,302 - Epoch: [37][ 150/ 1200] Overall Loss 0.266989 Objective Loss 0.266989 LR 0.001000 Time 0.023959 -2022-12-06 10:38:41,493 - Epoch: [37][ 160/ 1200] Overall Loss 0.268561 Objective Loss 0.268561 LR 0.001000 Time 0.023653 -2022-12-06 10:38:41,685 - Epoch: [37][ 170/ 1200] Overall Loss 0.268883 Objective Loss 0.268883 LR 0.001000 Time 0.023384 -2022-12-06 10:38:41,876 - Epoch: [37][ 180/ 1200] Overall Loss 0.269256 Objective Loss 0.269256 LR 0.001000 Time 0.023144 -2022-12-06 10:38:42,068 - Epoch: [37][ 190/ 1200] Overall Loss 0.269353 Objective Loss 0.269353 LR 0.001000 Time 0.022931 -2022-12-06 10:38:42,259 - Epoch: [37][ 200/ 1200] Overall Loss 0.268787 Objective Loss 0.268787 LR 0.001000 Time 0.022736 -2022-12-06 10:38:42,449 - Epoch: [37][ 210/ 1200] Overall Loss 0.268976 Objective Loss 0.268976 LR 0.001000 Time 0.022558 -2022-12-06 10:38:42,640 - Epoch: [37][ 220/ 1200] Overall Loss 0.270498 Objective Loss 0.270498 LR 0.001000 Time 0.022399 -2022-12-06 10:38:42,831 - Epoch: [37][ 230/ 1200] Overall Loss 0.270305 Objective Loss 0.270305 LR 0.001000 Time 0.022251 -2022-12-06 10:38:43,023 - Epoch: [37][ 240/ 1200] Overall Loss 0.269597 Objective Loss 0.269597 LR 0.001000 Time 0.022123 -2022-12-06 10:38:43,215 - Epoch: [37][ 250/ 1200] Overall Loss 0.271177 Objective Loss 0.271177 LR 0.001000 Time 0.022003 -2022-12-06 10:38:43,406 - Epoch: [37][ 260/ 1200] Overall Loss 0.270282 Objective Loss 0.270282 LR 0.001000 Time 0.021891 -2022-12-06 10:38:43,597 - Epoch: [37][ 270/ 1200] Overall Loss 0.270419 Objective Loss 0.270419 LR 0.001000 Time 0.021786 -2022-12-06 10:38:43,789 - Epoch: [37][ 280/ 1200] Overall Loss 0.270843 Objective Loss 0.270843 LR 0.001000 Time 0.021690 -2022-12-06 10:38:43,981 - Epoch: [37][ 290/ 1200] Overall Loss 0.272256 Objective Loss 0.272256 LR 0.001000 Time 0.021604 -2022-12-06 10:38:44,172 - Epoch: [37][ 300/ 1200] Overall Loss 0.272546 Objective Loss 0.272546 LR 0.001000 Time 0.021518 -2022-12-06 10:38:44,364 - Epoch: [37][ 310/ 1200] Overall Loss 0.273695 Objective Loss 0.273695 LR 0.001000 Time 0.021439 -2022-12-06 10:38:44,554 - Epoch: [37][ 320/ 1200] Overall Loss 0.274099 Objective Loss 0.274099 LR 0.001000 Time 0.021364 -2022-12-06 10:38:44,746 - Epoch: [37][ 330/ 1200] Overall Loss 0.274942 Objective Loss 0.274942 LR 0.001000 Time 0.021296 -2022-12-06 10:38:44,937 - Epoch: [37][ 340/ 1200] Overall Loss 0.274796 Objective Loss 0.274796 LR 0.001000 Time 0.021230 -2022-12-06 10:38:45,129 - Epoch: [37][ 350/ 1200] Overall Loss 0.275430 Objective Loss 0.275430 LR 0.001000 Time 0.021171 -2022-12-06 10:38:45,320 - Epoch: [37][ 360/ 1200] Overall Loss 0.276725 Objective Loss 0.276725 LR 0.001000 Time 0.021112 -2022-12-06 10:38:45,512 - Epoch: [37][ 370/ 1200] Overall Loss 0.276698 Objective Loss 0.276698 LR 0.001000 Time 0.021059 -2022-12-06 10:38:45,704 - Epoch: [37][ 380/ 1200] Overall Loss 0.276690 Objective Loss 0.276690 LR 0.001000 Time 0.021007 -2022-12-06 10:38:45,895 - Epoch: [37][ 390/ 1200] Overall Loss 0.275971 Objective Loss 0.275971 LR 0.001000 Time 0.020958 -2022-12-06 10:38:46,087 - Epoch: [37][ 400/ 1200] Overall Loss 0.276220 Objective Loss 0.276220 LR 0.001000 Time 0.020913 -2022-12-06 10:38:46,280 - Epoch: [37][ 410/ 1200] Overall Loss 0.275513 Objective Loss 0.275513 LR 0.001000 Time 0.020870 -2022-12-06 10:38:46,470 - Epoch: [37][ 420/ 1200] Overall Loss 0.274946 Objective Loss 0.274946 LR 0.001000 Time 0.020826 -2022-12-06 10:38:46,662 - Epoch: [37][ 430/ 1200] Overall Loss 0.275146 Objective Loss 0.275146 LR 0.001000 Time 0.020786 -2022-12-06 10:38:46,854 - Epoch: [37][ 440/ 1200] Overall Loss 0.275439 Objective Loss 0.275439 LR 0.001000 Time 0.020749 -2022-12-06 10:38:47,045 - Epoch: [37][ 450/ 1200] Overall Loss 0.275292 Objective Loss 0.275292 LR 0.001000 Time 0.020711 -2022-12-06 10:38:47,236 - Epoch: [37][ 460/ 1200] Overall Loss 0.275210 Objective Loss 0.275210 LR 0.001000 Time 0.020675 -2022-12-06 10:38:47,428 - Epoch: [37][ 470/ 1200] Overall Loss 0.275029 Objective Loss 0.275029 LR 0.001000 Time 0.020641 -2022-12-06 10:38:47,620 - Epoch: [37][ 480/ 1200] Overall Loss 0.275348 Objective Loss 0.275348 LR 0.001000 Time 0.020610 -2022-12-06 10:38:47,812 - Epoch: [37][ 490/ 1200] Overall Loss 0.274790 Objective Loss 0.274790 LR 0.001000 Time 0.020580 -2022-12-06 10:38:48,003 - Epoch: [37][ 500/ 1200] Overall Loss 0.274336 Objective Loss 0.274336 LR 0.001000 Time 0.020551 -2022-12-06 10:38:48,194 - Epoch: [37][ 510/ 1200] Overall Loss 0.274147 Objective Loss 0.274147 LR 0.001000 Time 0.020521 -2022-12-06 10:38:48,386 - Epoch: [37][ 520/ 1200] Overall Loss 0.274787 Objective Loss 0.274787 LR 0.001000 Time 0.020493 -2022-12-06 10:38:48,578 - Epoch: [37][ 530/ 1200] Overall Loss 0.275207 Objective Loss 0.275207 LR 0.001000 Time 0.020468 -2022-12-06 10:38:48,769 - Epoch: [37][ 540/ 1200] Overall Loss 0.275038 Objective Loss 0.275038 LR 0.001000 Time 0.020443 -2022-12-06 10:38:48,960 - Epoch: [37][ 550/ 1200] Overall Loss 0.274520 Objective Loss 0.274520 LR 0.001000 Time 0.020416 -2022-12-06 10:38:49,151 - Epoch: [37][ 560/ 1200] Overall Loss 0.274022 Objective Loss 0.274022 LR 0.001000 Time 0.020393 -2022-12-06 10:38:49,344 - Epoch: [37][ 570/ 1200] Overall Loss 0.274593 Objective Loss 0.274593 LR 0.001000 Time 0.020373 -2022-12-06 10:38:49,535 - Epoch: [37][ 580/ 1200] Overall Loss 0.274741 Objective Loss 0.274741 LR 0.001000 Time 0.020349 -2022-12-06 10:38:49,725 - Epoch: [37][ 590/ 1200] Overall Loss 0.274379 Objective Loss 0.274379 LR 0.001000 Time 0.020325 -2022-12-06 10:38:49,917 - Epoch: [37][ 600/ 1200] Overall Loss 0.274659 Objective Loss 0.274659 LR 0.001000 Time 0.020306 -2022-12-06 10:38:50,108 - Epoch: [37][ 610/ 1200] Overall Loss 0.274757 Objective Loss 0.274757 LR 0.001000 Time 0.020284 -2022-12-06 10:38:50,298 - Epoch: [37][ 620/ 1200] Overall Loss 0.274848 Objective Loss 0.274848 LR 0.001000 Time 0.020264 -2022-12-06 10:38:50,490 - Epoch: [37][ 630/ 1200] Overall Loss 0.275063 Objective Loss 0.275063 LR 0.001000 Time 0.020245 -2022-12-06 10:38:50,680 - Epoch: [37][ 640/ 1200] Overall Loss 0.274936 Objective Loss 0.274936 LR 0.001000 Time 0.020225 -2022-12-06 10:38:50,871 - Epoch: [37][ 650/ 1200] Overall Loss 0.274945 Objective Loss 0.274945 LR 0.001000 Time 0.020207 -2022-12-06 10:38:51,063 - Epoch: [37][ 660/ 1200] Overall Loss 0.275322 Objective Loss 0.275322 LR 0.001000 Time 0.020191 -2022-12-06 10:38:51,254 - Epoch: [37][ 670/ 1200] Overall Loss 0.275416 Objective Loss 0.275416 LR 0.001000 Time 0.020173 -2022-12-06 10:38:51,444 - Epoch: [37][ 680/ 1200] Overall Loss 0.276028 Objective Loss 0.276028 LR 0.001000 Time 0.020156 -2022-12-06 10:38:51,634 - Epoch: [37][ 690/ 1200] Overall Loss 0.276405 Objective Loss 0.276405 LR 0.001000 Time 0.020138 -2022-12-06 10:38:51,825 - Epoch: [37][ 700/ 1200] Overall Loss 0.276527 Objective Loss 0.276527 LR 0.001000 Time 0.020123 -2022-12-06 10:38:52,016 - Epoch: [37][ 710/ 1200] Overall Loss 0.276475 Objective Loss 0.276475 LR 0.001000 Time 0.020107 -2022-12-06 10:38:52,207 - Epoch: [37][ 720/ 1200] Overall Loss 0.276637 Objective Loss 0.276637 LR 0.001000 Time 0.020093 -2022-12-06 10:38:52,398 - Epoch: [37][ 730/ 1200] Overall Loss 0.276824 Objective Loss 0.276824 LR 0.001000 Time 0.020079 -2022-12-06 10:38:52,590 - Epoch: [37][ 740/ 1200] Overall Loss 0.276983 Objective Loss 0.276983 LR 0.001000 Time 0.020065 -2022-12-06 10:38:52,781 - Epoch: [37][ 750/ 1200] Overall Loss 0.277199 Objective Loss 0.277199 LR 0.001000 Time 0.020052 -2022-12-06 10:38:52,973 - Epoch: [37][ 760/ 1200] Overall Loss 0.276987 Objective Loss 0.276987 LR 0.001000 Time 0.020040 -2022-12-06 10:38:53,164 - Epoch: [37][ 770/ 1200] Overall Loss 0.277072 Objective Loss 0.277072 LR 0.001000 Time 0.020027 -2022-12-06 10:38:53,356 - Epoch: [37][ 780/ 1200] Overall Loss 0.277206 Objective Loss 0.277206 LR 0.001000 Time 0.020016 -2022-12-06 10:38:53,547 - Epoch: [37][ 790/ 1200] Overall Loss 0.277691 Objective Loss 0.277691 LR 0.001000 Time 0.020004 -2022-12-06 10:38:53,738 - Epoch: [37][ 800/ 1200] Overall Loss 0.278131 Objective Loss 0.278131 LR 0.001000 Time 0.019992 -2022-12-06 10:38:53,929 - Epoch: [37][ 810/ 1200] Overall Loss 0.278431 Objective Loss 0.278431 LR 0.001000 Time 0.019981 -2022-12-06 10:38:54,121 - Epoch: [37][ 820/ 1200] Overall Loss 0.278604 Objective Loss 0.278604 LR 0.001000 Time 0.019970 -2022-12-06 10:38:54,311 - Epoch: [37][ 830/ 1200] Overall Loss 0.279357 Objective Loss 0.279357 LR 0.001000 Time 0.019958 -2022-12-06 10:38:54,502 - Epoch: [37][ 840/ 1200] Overall Loss 0.279950 Objective Loss 0.279950 LR 0.001000 Time 0.019947 -2022-12-06 10:38:54,693 - Epoch: [37][ 850/ 1200] Overall Loss 0.279835 Objective Loss 0.279835 LR 0.001000 Time 0.019936 -2022-12-06 10:38:54,886 - Epoch: [37][ 860/ 1200] Overall Loss 0.279764 Objective Loss 0.279764 LR 0.001000 Time 0.019928 -2022-12-06 10:38:55,077 - Epoch: [37][ 870/ 1200] Overall Loss 0.279799 Objective Loss 0.279799 LR 0.001000 Time 0.019918 -2022-12-06 10:38:55,268 - Epoch: [37][ 880/ 1200] Overall Loss 0.279705 Objective Loss 0.279705 LR 0.001000 Time 0.019908 -2022-12-06 10:38:55,459 - Epoch: [37][ 890/ 1200] Overall Loss 0.279863 Objective Loss 0.279863 LR 0.001000 Time 0.019898 -2022-12-06 10:38:55,650 - Epoch: [37][ 900/ 1200] Overall Loss 0.279639 Objective Loss 0.279639 LR 0.001000 Time 0.019889 -2022-12-06 10:38:55,841 - Epoch: [37][ 910/ 1200] Overall Loss 0.279525 Objective Loss 0.279525 LR 0.001000 Time 0.019880 -2022-12-06 10:38:56,032 - Epoch: [37][ 920/ 1200] Overall Loss 0.279573 Objective Loss 0.279573 LR 0.001000 Time 0.019870 -2022-12-06 10:38:56,224 - Epoch: [37][ 930/ 1200] Overall Loss 0.279655 Objective Loss 0.279655 LR 0.001000 Time 0.019863 -2022-12-06 10:38:56,415 - Epoch: [37][ 940/ 1200] Overall Loss 0.279535 Objective Loss 0.279535 LR 0.001000 Time 0.019854 -2022-12-06 10:38:56,607 - Epoch: [37][ 950/ 1200] Overall Loss 0.279595 Objective Loss 0.279595 LR 0.001000 Time 0.019846 -2022-12-06 10:38:56,797 - Epoch: [37][ 960/ 1200] Overall Loss 0.279790 Objective Loss 0.279790 LR 0.001000 Time 0.019837 -2022-12-06 10:38:56,988 - Epoch: [37][ 970/ 1200] Overall Loss 0.280002 Objective Loss 0.280002 LR 0.001000 Time 0.019829 -2022-12-06 10:38:57,179 - Epoch: [37][ 980/ 1200] Overall Loss 0.279832 Objective Loss 0.279832 LR 0.001000 Time 0.019821 -2022-12-06 10:38:57,369 - Epoch: [37][ 990/ 1200] Overall Loss 0.279951 Objective Loss 0.279951 LR 0.001000 Time 0.019812 -2022-12-06 10:38:57,560 - Epoch: [37][ 1000/ 1200] Overall Loss 0.279901 Objective Loss 0.279901 LR 0.001000 Time 0.019805 -2022-12-06 10:38:57,752 - Epoch: [37][ 1010/ 1200] Overall Loss 0.279932 Objective Loss 0.279932 LR 0.001000 Time 0.019798 -2022-12-06 10:38:57,943 - Epoch: [37][ 1020/ 1200] Overall Loss 0.280371 Objective Loss 0.280371 LR 0.001000 Time 0.019790 -2022-12-06 10:38:58,135 - Epoch: [37][ 1030/ 1200] Overall Loss 0.280483 Objective Loss 0.280483 LR 0.001000 Time 0.019784 -2022-12-06 10:38:58,326 - Epoch: [37][ 1040/ 1200] Overall Loss 0.280520 Objective Loss 0.280520 LR 0.001000 Time 0.019777 -2022-12-06 10:38:58,517 - Epoch: [37][ 1050/ 1200] Overall Loss 0.281274 Objective Loss 0.281274 LR 0.001000 Time 0.019771 -2022-12-06 10:38:58,708 - Epoch: [37][ 1060/ 1200] Overall Loss 0.283035 Objective Loss 0.283035 LR 0.001000 Time 0.019764 -2022-12-06 10:38:58,900 - Epoch: [37][ 1070/ 1200] Overall Loss 0.284874 Objective Loss 0.284874 LR 0.001000 Time 0.019757 -2022-12-06 10:38:59,092 - Epoch: [37][ 1080/ 1200] Overall Loss 0.286167 Objective Loss 0.286167 LR 0.001000 Time 0.019752 -2022-12-06 10:38:59,282 - Epoch: [37][ 1090/ 1200] Overall Loss 0.287353 Objective Loss 0.287353 LR 0.001000 Time 0.019745 -2022-12-06 10:38:59,473 - Epoch: [37][ 1100/ 1200] Overall Loss 0.288527 Objective Loss 0.288527 LR 0.001000 Time 0.019738 -2022-12-06 10:38:59,665 - Epoch: [37][ 1110/ 1200] Overall Loss 0.289487 Objective Loss 0.289487 LR 0.001000 Time 0.019733 -2022-12-06 10:38:59,856 - Epoch: [37][ 1120/ 1200] Overall Loss 0.290648 Objective Loss 0.290648 LR 0.001000 Time 0.019727 -2022-12-06 10:39:00,046 - Epoch: [37][ 1130/ 1200] Overall Loss 0.291734 Objective Loss 0.291734 LR 0.001000 Time 0.019720 -2022-12-06 10:39:00,237 - Epoch: [37][ 1140/ 1200] Overall Loss 0.292615 Objective Loss 0.292615 LR 0.001000 Time 0.019714 -2022-12-06 10:39:00,428 - Epoch: [37][ 1150/ 1200] Overall Loss 0.293513 Objective Loss 0.293513 LR 0.001000 Time 0.019708 -2022-12-06 10:39:00,620 - Epoch: [37][ 1160/ 1200] Overall Loss 0.294401 Objective Loss 0.294401 LR 0.001000 Time 0.019703 -2022-12-06 10:39:00,810 - Epoch: [37][ 1170/ 1200] Overall Loss 0.295069 Objective Loss 0.295069 LR 0.001000 Time 0.019697 -2022-12-06 10:39:01,001 - Epoch: [37][ 1180/ 1200] Overall Loss 0.295762 Objective Loss 0.295762 LR 0.001000 Time 0.019691 -2022-12-06 10:39:01,193 - Epoch: [37][ 1190/ 1200] Overall Loss 0.296372 Objective Loss 0.296372 LR 0.001000 Time 0.019686 -2022-12-06 10:39:01,417 - Epoch: [37][ 1200/ 1200] Overall Loss 0.297230 Objective Loss 0.297230 Top1 82.008368 Top5 97.489540 LR 0.001000 Time 0.019709 -2022-12-06 10:39:01,506 - --- validate (epoch=37)----------- -2022-12-06 10:39:01,507 - 34129 samples (256 per mini-batch) -2022-12-06 10:39:01,970 - Epoch: [37][ 10/ 134] Loss 0.381468 Top1 80.742188 Top5 97.656250 -2022-12-06 10:39:02,117 - Epoch: [37][ 20/ 134] Loss 0.393127 Top1 80.625000 Top5 97.500000 -2022-12-06 10:39:02,260 - Epoch: [37][ 30/ 134] Loss 0.392006 Top1 81.119792 Top5 97.539062 -2022-12-06 10:39:02,396 - Epoch: [37][ 40/ 134] Loss 0.387958 Top1 81.328125 Top5 97.451172 -2022-12-06 10:39:02,527 - Epoch: [37][ 50/ 134] Loss 0.389895 Top1 81.507812 Top5 97.539062 -2022-12-06 10:39:02,659 - Epoch: [37][ 60/ 134] Loss 0.384703 Top1 81.751302 Top5 97.591146 -2022-12-06 10:39:02,791 - Epoch: [37][ 70/ 134] Loss 0.389172 Top1 81.679688 Top5 97.589286 -2022-12-06 10:39:02,922 - Epoch: [37][ 80/ 134] Loss 0.389472 Top1 81.694336 Top5 97.656250 -2022-12-06 10:39:03,054 - Epoch: [37][ 90/ 134] Loss 0.387054 Top1 81.788194 Top5 97.703993 -2022-12-06 10:39:03,185 - Epoch: [37][ 100/ 134] Loss 0.387421 Top1 81.871094 Top5 97.695312 -2022-12-06 10:39:03,317 - Epoch: [37][ 110/ 134] Loss 0.385865 Top1 82.002841 Top5 97.691761 -2022-12-06 10:39:03,449 - Epoch: [37][ 120/ 134] Loss 0.389199 Top1 81.959635 Top5 97.662760 -2022-12-06 10:39:03,582 - Epoch: [37][ 130/ 134] Loss 0.387870 Top1 82.016226 Top5 97.632212 -2022-12-06 10:39:03,621 - Epoch: [37][ 134/ 134] Loss 0.388165 Top1 81.994784 Top5 97.612002 -2022-12-06 10:39:03,710 - ==> Top1: 81.995 Top5: 97.612 Loss: 0.388 - -2022-12-06 10:39:03,710 - ==> Confusion: -[[ 896 1 2 0 6 8 0 2 6 61 0 0 3 2 4 1 0 1 2 0 1] - [ 3 917 0 2 9 42 1 20 1 0 8 4 1 1 2 1 4 1 8 0 2] - [ 6 3 992 19 4 6 16 9 0 3 4 5 4 1 1 9 2 0 5 4 10] - [ 2 1 21 936 0 6 0 1 1 1 10 0 5 1 23 0 3 1 6 0 2] - [ 10 1 1 0 948 8 0 2 0 12 2 7 1 4 8 3 7 3 0 1 2] - [ 6 11 1 0 9 983 0 14 3 1 2 13 3 12 1 0 2 0 1 4 3] - [ 0 2 9 5 1 9 1051 5 0 0 2 3 3 0 0 9 1 1 2 12 3] - [ 0 5 7 3 1 39 3 939 0 0 1 8 3 0 0 1 1 0 29 12 2] - [ 3 2 0 0 0 8 1 1 969 39 10 1 2 12 10 0 1 0 2 2 1] - [ 65 0 0 0 2 3 0 1 21 890 0 1 0 11 2 1 0 0 1 0 3] - [ 0 4 4 7 0 3 2 1 9 2 946 4 2 14 7 0 0 0 10 2 2] - [ 4 0 2 0 1 11 1 4 2 0 0 988 15 4 1 5 3 2 0 7 1] - [ 0 1 0 4 0 4 2 2 1 0 0 57 872 0 1 9 2 5 0 4 5] - [ 1 0 1 1 2 13 0 3 11 12 6 9 1 951 1 3 1 0 1 2 4] - [ 9 5 1 8 4 3 0 0 27 6 0 2 3 5 1042 0 4 0 6 1 4] - [ 2 1 2 1 3 4 4 0 0 0 1 9 7 4 0 982 7 8 1 3 4] - [ 3 2 0 2 2 1 0 0 1 0 1 4 0 2 2 14 1025 4 2 4 3] - [ 3 1 0 5 0 3 2 1 3 1 0 14 22 2 3 15 3 956 0 2 0] - [ 4 4 4 8 1 8 0 22 1 0 5 2 1 2 11 1 0 0 931 2 1] - [ 4 3 0 0 1 9 3 8 0 0 1 19 5 5 1 1 4 1 0 1013 2] - [ 173 289 210 140 158 306 69 169 115 126 211 169 487 493 232 168 285 61 243 376 8746]] - -2022-12-06 10:39:04,386 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:39:04,386 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:39:04,392 - - -2022-12-06 10:39:04,392 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:39:05,329 - Epoch: [38][ 10/ 1200] Overall Loss 0.348101 Objective Loss 0.348101 LR 0.001000 Time 0.093608 -2022-12-06 10:39:05,520 - Epoch: [38][ 20/ 1200] Overall Loss 0.359428 Objective Loss 0.359428 LR 0.001000 Time 0.056352 -2022-12-06 10:39:05,712 - Epoch: [38][ 30/ 1200] Overall Loss 0.362771 Objective Loss 0.362771 LR 0.001000 Time 0.043954 -2022-12-06 10:39:05,904 - Epoch: [38][ 40/ 1200] Overall Loss 0.364305 Objective Loss 0.364305 LR 0.001000 Time 0.037751 -2022-12-06 10:39:06,096 - Epoch: [38][ 50/ 1200] Overall Loss 0.363490 Objective Loss 0.363490 LR 0.001000 Time 0.034016 -2022-12-06 10:39:06,287 - Epoch: [38][ 60/ 1200] Overall Loss 0.360661 Objective Loss 0.360661 LR 0.001000 Time 0.031529 -2022-12-06 10:39:06,479 - Epoch: [38][ 70/ 1200] Overall Loss 0.362022 Objective Loss 0.362022 LR 0.001000 Time 0.029761 -2022-12-06 10:39:06,671 - Epoch: [38][ 80/ 1200] Overall Loss 0.364999 Objective Loss 0.364999 LR 0.001000 Time 0.028433 -2022-12-06 10:39:06,863 - Epoch: [38][ 90/ 1200] Overall Loss 0.363615 Objective Loss 0.363615 LR 0.001000 Time 0.027395 -2022-12-06 10:39:07,054 - Epoch: [38][ 100/ 1200] Overall Loss 0.363227 Objective Loss 0.363227 LR 0.001000 Time 0.026561 -2022-12-06 10:39:07,246 - Epoch: [38][ 110/ 1200] Overall Loss 0.362316 Objective Loss 0.362316 LR 0.001000 Time 0.025884 -2022-12-06 10:39:07,437 - Epoch: [38][ 120/ 1200] Overall Loss 0.361758 Objective Loss 0.361758 LR 0.001000 Time 0.025318 -2022-12-06 10:39:07,630 - Epoch: [38][ 130/ 1200] Overall Loss 0.361160 Objective Loss 0.361160 LR 0.001000 Time 0.024849 -2022-12-06 10:39:07,822 - Epoch: [38][ 140/ 1200] Overall Loss 0.360705 Objective Loss 0.360705 LR 0.001000 Time 0.024441 -2022-12-06 10:39:08,014 - Epoch: [38][ 150/ 1200] Overall Loss 0.360192 Objective Loss 0.360192 LR 0.001000 Time 0.024087 -2022-12-06 10:39:08,205 - Epoch: [38][ 160/ 1200] Overall Loss 0.359706 Objective Loss 0.359706 LR 0.001000 Time 0.023775 -2022-12-06 10:39:08,397 - Epoch: [38][ 170/ 1200] Overall Loss 0.358158 Objective Loss 0.358158 LR 0.001000 Time 0.023499 -2022-12-06 10:39:08,589 - Epoch: [38][ 180/ 1200] Overall Loss 0.359875 Objective Loss 0.359875 LR 0.001000 Time 0.023258 -2022-12-06 10:39:08,782 - Epoch: [38][ 190/ 1200] Overall Loss 0.360096 Objective Loss 0.360096 LR 0.001000 Time 0.023046 -2022-12-06 10:39:08,974 - Epoch: [38][ 200/ 1200] Overall Loss 0.360817 Objective Loss 0.360817 LR 0.001000 Time 0.022851 -2022-12-06 10:39:09,166 - Epoch: [38][ 210/ 1200] Overall Loss 0.359604 Objective Loss 0.359604 LR 0.001000 Time 0.022676 -2022-12-06 10:39:09,358 - Epoch: [38][ 220/ 1200] Overall Loss 0.358372 Objective Loss 0.358372 LR 0.001000 Time 0.022516 -2022-12-06 10:39:09,551 - Epoch: [38][ 230/ 1200] Overall Loss 0.358999 Objective Loss 0.358999 LR 0.001000 Time 0.022371 -2022-12-06 10:39:09,743 - Epoch: [38][ 240/ 1200] Overall Loss 0.358757 Objective Loss 0.358757 LR 0.001000 Time 0.022240 -2022-12-06 10:39:09,936 - Epoch: [38][ 250/ 1200] Overall Loss 0.358167 Objective Loss 0.358167 LR 0.001000 Time 0.022117 -2022-12-06 10:39:10,128 - Epoch: [38][ 260/ 1200] Overall Loss 0.358174 Objective Loss 0.358174 LR 0.001000 Time 0.022002 -2022-12-06 10:39:10,320 - Epoch: [38][ 270/ 1200] Overall Loss 0.356402 Objective Loss 0.356402 LR 0.001000 Time 0.021896 -2022-12-06 10:39:10,512 - Epoch: [38][ 280/ 1200] Overall Loss 0.355515 Objective Loss 0.355515 LR 0.001000 Time 0.021799 -2022-12-06 10:39:10,704 - Epoch: [38][ 290/ 1200] Overall Loss 0.354823 Objective Loss 0.354823 LR 0.001000 Time 0.021706 -2022-12-06 10:39:10,895 - Epoch: [38][ 300/ 1200] Overall Loss 0.353523 Objective Loss 0.353523 LR 0.001000 Time 0.021619 -2022-12-06 10:39:11,087 - Epoch: [38][ 310/ 1200] Overall Loss 0.353193 Objective Loss 0.353193 LR 0.001000 Time 0.021540 -2022-12-06 10:39:11,280 - Epoch: [38][ 320/ 1200] Overall Loss 0.353264 Objective Loss 0.353264 LR 0.001000 Time 0.021466 -2022-12-06 10:39:11,471 - Epoch: [38][ 330/ 1200] Overall Loss 0.352871 Objective Loss 0.352871 LR 0.001000 Time 0.021395 -2022-12-06 10:39:11,663 - Epoch: [38][ 340/ 1200] Overall Loss 0.351731 Objective Loss 0.351731 LR 0.001000 Time 0.021328 -2022-12-06 10:39:11,855 - Epoch: [38][ 350/ 1200] Overall Loss 0.351590 Objective Loss 0.351590 LR 0.001000 Time 0.021265 -2022-12-06 10:39:12,046 - Epoch: [38][ 360/ 1200] Overall Loss 0.352069 Objective Loss 0.352069 LR 0.001000 Time 0.021204 -2022-12-06 10:39:12,238 - Epoch: [38][ 370/ 1200] Overall Loss 0.352076 Objective Loss 0.352076 LR 0.001000 Time 0.021149 -2022-12-06 10:39:12,430 - Epoch: [38][ 380/ 1200] Overall Loss 0.351464 Objective Loss 0.351464 LR 0.001000 Time 0.021094 -2022-12-06 10:39:12,622 - Epoch: [38][ 390/ 1200] Overall Loss 0.350915 Objective Loss 0.350915 LR 0.001000 Time 0.021044 -2022-12-06 10:39:12,813 - Epoch: [38][ 400/ 1200] Overall Loss 0.350443 Objective Loss 0.350443 LR 0.001000 Time 0.020994 -2022-12-06 10:39:13,005 - Epoch: [38][ 410/ 1200] Overall Loss 0.349921 Objective Loss 0.349921 LR 0.001000 Time 0.020950 -2022-12-06 10:39:13,197 - Epoch: [38][ 420/ 1200] Overall Loss 0.350254 Objective Loss 0.350254 LR 0.001000 Time 0.020906 -2022-12-06 10:39:13,389 - Epoch: [38][ 430/ 1200] Overall Loss 0.349334 Objective Loss 0.349334 LR 0.001000 Time 0.020865 -2022-12-06 10:39:13,581 - Epoch: [38][ 440/ 1200] Overall Loss 0.349629 Objective Loss 0.349629 LR 0.001000 Time 0.020827 -2022-12-06 10:39:13,773 - Epoch: [38][ 450/ 1200] Overall Loss 0.349570 Objective Loss 0.349570 LR 0.001000 Time 0.020789 -2022-12-06 10:39:13,965 - Epoch: [38][ 460/ 1200] Overall Loss 0.349383 Objective Loss 0.349383 LR 0.001000 Time 0.020753 -2022-12-06 10:39:14,157 - Epoch: [38][ 470/ 1200] Overall Loss 0.349291 Objective Loss 0.349291 LR 0.001000 Time 0.020718 -2022-12-06 10:39:14,348 - Epoch: [38][ 480/ 1200] Overall Loss 0.348677 Objective Loss 0.348677 LR 0.001000 Time 0.020685 -2022-12-06 10:39:14,540 - Epoch: [38][ 490/ 1200] Overall Loss 0.348184 Objective Loss 0.348184 LR 0.001000 Time 0.020652 -2022-12-06 10:39:14,732 - Epoch: [38][ 500/ 1200] Overall Loss 0.347245 Objective Loss 0.347245 LR 0.001000 Time 0.020622 -2022-12-06 10:39:14,924 - Epoch: [38][ 510/ 1200] Overall Loss 0.347044 Objective Loss 0.347044 LR 0.001000 Time 0.020593 -2022-12-06 10:39:15,115 - Epoch: [38][ 520/ 1200] Overall Loss 0.347148 Objective Loss 0.347148 LR 0.001000 Time 0.020564 -2022-12-06 10:39:15,308 - Epoch: [38][ 530/ 1200] Overall Loss 0.346742 Objective Loss 0.346742 LR 0.001000 Time 0.020538 -2022-12-06 10:39:15,499 - Epoch: [38][ 540/ 1200] Overall Loss 0.346137 Objective Loss 0.346137 LR 0.001000 Time 0.020512 -2022-12-06 10:39:15,691 - Epoch: [38][ 550/ 1200] Overall Loss 0.345619 Objective Loss 0.345619 LR 0.001000 Time 0.020486 -2022-12-06 10:39:15,883 - Epoch: [38][ 560/ 1200] Overall Loss 0.345073 Objective Loss 0.345073 LR 0.001000 Time 0.020462 -2022-12-06 10:39:16,076 - Epoch: [38][ 570/ 1200] Overall Loss 0.344286 Objective Loss 0.344286 LR 0.001000 Time 0.020440 -2022-12-06 10:39:16,268 - Epoch: [38][ 580/ 1200] Overall Loss 0.343928 Objective Loss 0.343928 LR 0.001000 Time 0.020418 -2022-12-06 10:39:16,460 - Epoch: [38][ 590/ 1200] Overall Loss 0.343320 Objective Loss 0.343320 LR 0.001000 Time 0.020397 -2022-12-06 10:39:16,652 - Epoch: [38][ 600/ 1200] Overall Loss 0.343392 Objective Loss 0.343392 LR 0.001000 Time 0.020376 -2022-12-06 10:39:16,844 - Epoch: [38][ 610/ 1200] Overall Loss 0.342676 Objective Loss 0.342676 LR 0.001000 Time 0.020355 -2022-12-06 10:39:17,036 - Epoch: [38][ 620/ 1200] Overall Loss 0.342595 Objective Loss 0.342595 LR 0.001000 Time 0.020335 -2022-12-06 10:39:17,227 - Epoch: [38][ 630/ 1200] Overall Loss 0.342241 Objective Loss 0.342241 LR 0.001000 Time 0.020316 -2022-12-06 10:39:17,420 - Epoch: [38][ 640/ 1200] Overall Loss 0.341597 Objective Loss 0.341597 LR 0.001000 Time 0.020298 -2022-12-06 10:39:17,612 - Epoch: [38][ 650/ 1200] Overall Loss 0.341725 Objective Loss 0.341725 LR 0.001000 Time 0.020280 -2022-12-06 10:39:17,804 - Epoch: [38][ 660/ 1200] Overall Loss 0.341535 Objective Loss 0.341535 LR 0.001000 Time 0.020263 -2022-12-06 10:39:17,996 - Epoch: [38][ 670/ 1200] Overall Loss 0.341536 Objective Loss 0.341536 LR 0.001000 Time 0.020246 -2022-12-06 10:39:18,188 - Epoch: [38][ 680/ 1200] Overall Loss 0.341557 Objective Loss 0.341557 LR 0.001000 Time 0.020230 -2022-12-06 10:39:18,380 - Epoch: [38][ 690/ 1200] Overall Loss 0.341843 Objective Loss 0.341843 LR 0.001000 Time 0.020214 -2022-12-06 10:39:18,572 - Epoch: [38][ 700/ 1200] Overall Loss 0.341282 Objective Loss 0.341282 LR 0.001000 Time 0.020199 -2022-12-06 10:39:18,764 - Epoch: [38][ 710/ 1200] Overall Loss 0.341397 Objective Loss 0.341397 LR 0.001000 Time 0.020184 -2022-12-06 10:39:18,956 - Epoch: [38][ 720/ 1200] Overall Loss 0.341130 Objective Loss 0.341130 LR 0.001000 Time 0.020170 -2022-12-06 10:39:19,148 - Epoch: [38][ 730/ 1200] Overall Loss 0.340639 Objective Loss 0.340639 LR 0.001000 Time 0.020156 -2022-12-06 10:39:19,340 - Epoch: [38][ 740/ 1200] Overall Loss 0.340214 Objective Loss 0.340214 LR 0.001000 Time 0.020142 -2022-12-06 10:39:19,532 - Epoch: [38][ 750/ 1200] Overall Loss 0.339762 Objective Loss 0.339762 LR 0.001000 Time 0.020129 -2022-12-06 10:39:19,724 - Epoch: [38][ 760/ 1200] Overall Loss 0.339754 Objective Loss 0.339754 LR 0.001000 Time 0.020116 -2022-12-06 10:39:19,916 - Epoch: [38][ 770/ 1200] Overall Loss 0.339449 Objective Loss 0.339449 LR 0.001000 Time 0.020103 -2022-12-06 10:39:20,108 - Epoch: [38][ 780/ 1200] Overall Loss 0.339241 Objective Loss 0.339241 LR 0.001000 Time 0.020091 -2022-12-06 10:39:20,299 - Epoch: [38][ 790/ 1200] Overall Loss 0.339039 Objective Loss 0.339039 LR 0.001000 Time 0.020078 -2022-12-06 10:39:20,492 - Epoch: [38][ 800/ 1200] Overall Loss 0.338705 Objective Loss 0.338705 LR 0.001000 Time 0.020067 -2022-12-06 10:39:20,684 - Epoch: [38][ 810/ 1200] Overall Loss 0.338460 Objective Loss 0.338460 LR 0.001000 Time 0.020056 -2022-12-06 10:39:20,876 - Epoch: [38][ 820/ 1200] Overall Loss 0.338399 Objective Loss 0.338399 LR 0.001000 Time 0.020045 -2022-12-06 10:39:21,068 - Epoch: [38][ 830/ 1200] Overall Loss 0.338158 Objective Loss 0.338158 LR 0.001000 Time 0.020033 -2022-12-06 10:39:21,260 - Epoch: [38][ 840/ 1200] Overall Loss 0.337626 Objective Loss 0.337626 LR 0.001000 Time 0.020023 -2022-12-06 10:39:21,451 - Epoch: [38][ 850/ 1200] Overall Loss 0.337369 Objective Loss 0.337369 LR 0.001000 Time 0.020012 -2022-12-06 10:39:21,644 - Epoch: [38][ 860/ 1200] Overall Loss 0.337085 Objective Loss 0.337085 LR 0.001000 Time 0.020003 -2022-12-06 10:39:21,836 - Epoch: [38][ 870/ 1200] Overall Loss 0.337105 Objective Loss 0.337105 LR 0.001000 Time 0.019992 -2022-12-06 10:39:22,027 - Epoch: [38][ 880/ 1200] Overall Loss 0.337192 Objective Loss 0.337192 LR 0.001000 Time 0.019983 -2022-12-06 10:39:22,219 - Epoch: [38][ 890/ 1200] Overall Loss 0.336948 Objective Loss 0.336948 LR 0.001000 Time 0.019972 -2022-12-06 10:39:22,410 - Epoch: [38][ 900/ 1200] Overall Loss 0.336991 Objective Loss 0.336991 LR 0.001000 Time 0.019963 -2022-12-06 10:39:22,603 - Epoch: [38][ 910/ 1200] Overall Loss 0.336888 Objective Loss 0.336888 LR 0.001000 Time 0.019954 -2022-12-06 10:39:22,795 - Epoch: [38][ 920/ 1200] Overall Loss 0.336692 Objective Loss 0.336692 LR 0.001000 Time 0.019945 -2022-12-06 10:39:22,987 - Epoch: [38][ 930/ 1200] Overall Loss 0.336378 Objective Loss 0.336378 LR 0.001000 Time 0.019937 -2022-12-06 10:39:23,178 - Epoch: [38][ 940/ 1200] Overall Loss 0.336221 Objective Loss 0.336221 LR 0.001000 Time 0.019927 -2022-12-06 10:39:23,370 - Epoch: [38][ 950/ 1200] Overall Loss 0.335996 Objective Loss 0.335996 LR 0.001000 Time 0.019919 -2022-12-06 10:39:23,562 - Epoch: [38][ 960/ 1200] Overall Loss 0.336119 Objective Loss 0.336119 LR 0.001000 Time 0.019911 -2022-12-06 10:39:23,754 - Epoch: [38][ 970/ 1200] Overall Loss 0.336115 Objective Loss 0.336115 LR 0.001000 Time 0.019903 -2022-12-06 10:39:23,946 - Epoch: [38][ 980/ 1200] Overall Loss 0.335869 Objective Loss 0.335869 LR 0.001000 Time 0.019895 -2022-12-06 10:39:24,138 - Epoch: [38][ 990/ 1200] Overall Loss 0.335876 Objective Loss 0.335876 LR 0.001000 Time 0.019888 -2022-12-06 10:39:24,330 - Epoch: [38][ 1000/ 1200] Overall Loss 0.335689 Objective Loss 0.335689 LR 0.001000 Time 0.019880 -2022-12-06 10:39:24,522 - Epoch: [38][ 1010/ 1200] Overall Loss 0.335742 Objective Loss 0.335742 LR 0.001000 Time 0.019873 -2022-12-06 10:39:24,714 - Epoch: [38][ 1020/ 1200] Overall Loss 0.335500 Objective Loss 0.335500 LR 0.001000 Time 0.019866 -2022-12-06 10:39:24,906 - Epoch: [38][ 1030/ 1200] Overall Loss 0.335341 Objective Loss 0.335341 LR 0.001000 Time 0.019858 -2022-12-06 10:39:25,098 - Epoch: [38][ 1040/ 1200] Overall Loss 0.335686 Objective Loss 0.335686 LR 0.001000 Time 0.019852 -2022-12-06 10:39:25,290 - Epoch: [38][ 1050/ 1200] Overall Loss 0.335511 Objective Loss 0.335511 LR 0.001000 Time 0.019845 -2022-12-06 10:39:25,482 - Epoch: [38][ 1060/ 1200] Overall Loss 0.335421 Objective Loss 0.335421 LR 0.001000 Time 0.019839 -2022-12-06 10:39:25,674 - Epoch: [38][ 1070/ 1200] Overall Loss 0.335171 Objective Loss 0.335171 LR 0.001000 Time 0.019832 -2022-12-06 10:39:25,866 - Epoch: [38][ 1080/ 1200] Overall Loss 0.335230 Objective Loss 0.335230 LR 0.001000 Time 0.019826 -2022-12-06 10:39:26,058 - Epoch: [38][ 1090/ 1200] Overall Loss 0.335065 Objective Loss 0.335065 LR 0.001000 Time 0.019820 -2022-12-06 10:39:26,250 - Epoch: [38][ 1100/ 1200] Overall Loss 0.335059 Objective Loss 0.335059 LR 0.001000 Time 0.019813 -2022-12-06 10:39:26,442 - Epoch: [38][ 1110/ 1200] Overall Loss 0.334979 Objective Loss 0.334979 LR 0.001000 Time 0.019807 -2022-12-06 10:39:26,634 - Epoch: [38][ 1120/ 1200] Overall Loss 0.334957 Objective Loss 0.334957 LR 0.001000 Time 0.019802 -2022-12-06 10:39:26,826 - Epoch: [38][ 1130/ 1200] Overall Loss 0.334844 Objective Loss 0.334844 LR 0.001000 Time 0.019796 -2022-12-06 10:39:27,019 - Epoch: [38][ 1140/ 1200] Overall Loss 0.334803 Objective Loss 0.334803 LR 0.001000 Time 0.019790 -2022-12-06 10:39:27,211 - Epoch: [38][ 1150/ 1200] Overall Loss 0.334741 Objective Loss 0.334741 LR 0.001000 Time 0.019785 -2022-12-06 10:39:27,403 - Epoch: [38][ 1160/ 1200] Overall Loss 0.334716 Objective Loss 0.334716 LR 0.001000 Time 0.019780 -2022-12-06 10:39:27,596 - Epoch: [38][ 1170/ 1200] Overall Loss 0.334676 Objective Loss 0.334676 LR 0.001000 Time 0.019775 -2022-12-06 10:39:27,788 - Epoch: [38][ 1180/ 1200] Overall Loss 0.334478 Objective Loss 0.334478 LR 0.001000 Time 0.019769 -2022-12-06 10:39:27,980 - Epoch: [38][ 1190/ 1200] Overall Loss 0.334445 Objective Loss 0.334445 LR 0.001000 Time 0.019764 -2022-12-06 10:39:28,213 - Epoch: [38][ 1200/ 1200] Overall Loss 0.334014 Objective Loss 0.334014 Top1 83.682008 Top5 97.489540 LR 0.001000 Time 0.019793 -2022-12-06 10:39:28,302 - --- validate (epoch=38)----------- -2022-12-06 10:39:28,302 - 34129 samples (256 per mini-batch) -2022-12-06 10:39:28,754 - Epoch: [38][ 10/ 134] Loss 0.317331 Top1 83.437500 Top5 97.304688 -2022-12-06 10:39:28,884 - Epoch: [38][ 20/ 134] Loss 0.322436 Top1 82.832031 Top5 97.441406 -2022-12-06 10:39:29,012 - Epoch: [38][ 30/ 134] Loss 0.324603 Top1 82.747396 Top5 97.513021 -2022-12-06 10:39:29,158 - Epoch: [38][ 40/ 134] Loss 0.325134 Top1 82.529297 Top5 97.509766 -2022-12-06 10:39:29,285 - Epoch: [38][ 50/ 134] Loss 0.327904 Top1 82.515625 Top5 97.601562 -2022-12-06 10:39:29,412 - Epoch: [38][ 60/ 134] Loss 0.327269 Top1 82.591146 Top5 97.591146 -2022-12-06 10:39:29,536 - Epoch: [38][ 70/ 134] Loss 0.330981 Top1 82.522321 Top5 97.645089 -2022-12-06 10:39:29,661 - Epoch: [38][ 80/ 134] Loss 0.332825 Top1 82.456055 Top5 97.636719 -2022-12-06 10:39:29,788 - Epoch: [38][ 90/ 134] Loss 0.335217 Top1 82.404514 Top5 97.625868 -2022-12-06 10:39:29,926 - Epoch: [38][ 100/ 134] Loss 0.335089 Top1 82.429688 Top5 97.664062 -2022-12-06 10:39:30,067 - Epoch: [38][ 110/ 134] Loss 0.336308 Top1 82.492898 Top5 97.656250 -2022-12-06 10:39:30,204 - Epoch: [38][ 120/ 134] Loss 0.337807 Top1 82.454427 Top5 97.679036 -2022-12-06 10:39:30,331 - Epoch: [38][ 130/ 134] Loss 0.337493 Top1 82.548077 Top5 97.671274 -2022-12-06 10:39:30,367 - Epoch: [38][ 134/ 134] Loss 0.337896 Top1 82.498755 Top5 97.664743 -2022-12-06 10:39:30,457 - ==> Top1: 82.499 Top5: 97.665 Loss: 0.338 - -2022-12-06 10:39:30,458 - ==> Confusion: -[[ 887 2 3 0 5 5 0 3 5 63 0 4 1 2 6 2 2 2 3 1 0] - [ 1 926 1 2 10 24 1 24 0 1 4 2 3 2 2 1 5 2 8 0 8] - [ 6 6 986 19 2 2 16 11 0 2 5 3 2 7 2 3 0 4 6 8 13] - [ 1 4 26 907 0 3 1 2 1 0 9 0 5 3 29 1 1 8 13 0 6] - [ 10 4 1 0 959 6 1 2 2 4 2 3 1 1 8 6 4 3 0 2 1] - [ 5 18 1 2 12 945 3 22 4 2 0 15 1 21 3 1 0 0 2 6 6] - [ 0 2 8 1 0 5 1070 10 0 0 0 1 0 0 0 8 0 1 1 8 3] - [ 1 11 14 1 1 30 6 930 1 0 0 5 1 1 0 1 2 0 29 17 3] - [ 4 4 1 2 0 4 0 1 951 52 4 1 0 15 16 0 2 2 2 2 1] - [ 50 1 1 0 2 3 0 3 13 911 1 1 0 7 3 0 0 0 0 0 5] - [ 0 2 9 9 0 3 1 1 11 6 932 2 1 19 10 0 1 0 8 0 4] - [ 3 5 2 0 0 12 2 2 1 0 0 987 10 8 0 6 1 4 1 7 0] - [ 0 1 0 3 0 4 0 0 1 0 0 65 861 0 0 7 2 16 0 3 6] - [ 1 2 0 0 1 7 0 2 9 15 4 8 0 965 2 0 2 1 0 1 3] - [ 10 3 2 6 8 2 0 0 13 5 0 1 4 8 1051 0 2 2 6 1 6] - [ 0 0 2 0 6 3 4 0 0 1 2 7 3 3 0 981 8 13 1 5 4] - [ 3 4 2 1 5 3 2 1 3 1 0 7 0 1 3 2 1020 1 0 10 3] - [ 2 2 1 2 0 1 1 1 5 2 0 11 13 4 1 13 1 973 1 1 1] - [ 5 6 4 10 0 3 0 26 1 0 3 3 5 3 14 0 1 0 919 2 3] - [ 1 3 1 1 1 7 4 7 0 0 1 25 4 6 1 2 8 5 2 997 4] - [ 140 282 234 92 197 203 89 173 93 133 179 204 428 542 219 158 238 102 215 313 8992]] - -2022-12-06 10:39:31,025 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:39:31,025 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:39:31,031 - - -2022-12-06 10:39:31,031 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:39:32,063 - Epoch: [39][ 10/ 1200] Overall Loss 0.297948 Objective Loss 0.297948 LR 0.001000 Time 0.103107 -2022-12-06 10:39:32,264 - Epoch: [39][ 20/ 1200] Overall Loss 0.310091 Objective Loss 0.310091 LR 0.001000 Time 0.061592 -2022-12-06 10:39:32,457 - Epoch: [39][ 30/ 1200] Overall Loss 0.307037 Objective Loss 0.307037 LR 0.001000 Time 0.047484 -2022-12-06 10:39:32,651 - Epoch: [39][ 40/ 1200] Overall Loss 0.308415 Objective Loss 0.308415 LR 0.001000 Time 0.040436 -2022-12-06 10:39:32,844 - Epoch: [39][ 50/ 1200] Overall Loss 0.308702 Objective Loss 0.308702 LR 0.001000 Time 0.036209 -2022-12-06 10:39:33,037 - Epoch: [39][ 60/ 1200] Overall Loss 0.311296 Objective Loss 0.311296 LR 0.001000 Time 0.033382 -2022-12-06 10:39:33,230 - Epoch: [39][ 70/ 1200] Overall Loss 0.310940 Objective Loss 0.310940 LR 0.001000 Time 0.031358 -2022-12-06 10:39:33,423 - Epoch: [39][ 80/ 1200] Overall Loss 0.308196 Objective Loss 0.308196 LR 0.001000 Time 0.029842 -2022-12-06 10:39:33,615 - Epoch: [39][ 90/ 1200] Overall Loss 0.306612 Objective Loss 0.306612 LR 0.001000 Time 0.028661 -2022-12-06 10:39:33,807 - Epoch: [39][ 100/ 1200] Overall Loss 0.306539 Objective Loss 0.306539 LR 0.001000 Time 0.027711 -2022-12-06 10:39:34,001 - Epoch: [39][ 110/ 1200] Overall Loss 0.307003 Objective Loss 0.307003 LR 0.001000 Time 0.026942 -2022-12-06 10:39:34,193 - Epoch: [39][ 120/ 1200] Overall Loss 0.309041 Objective Loss 0.309041 LR 0.001000 Time 0.026297 -2022-12-06 10:39:34,386 - Epoch: [39][ 130/ 1200] Overall Loss 0.310072 Objective Loss 0.310072 LR 0.001000 Time 0.025754 -2022-12-06 10:39:34,579 - Epoch: [39][ 140/ 1200] Overall Loss 0.310593 Objective Loss 0.310593 LR 0.001000 Time 0.025288 -2022-12-06 10:39:34,772 - Epoch: [39][ 150/ 1200] Overall Loss 0.309497 Objective Loss 0.309497 LR 0.001000 Time 0.024890 -2022-12-06 10:39:34,966 - Epoch: [39][ 160/ 1200] Overall Loss 0.310478 Objective Loss 0.310478 LR 0.001000 Time 0.024543 -2022-12-06 10:39:35,161 - Epoch: [39][ 170/ 1200] Overall Loss 0.311526 Objective Loss 0.311526 LR 0.001000 Time 0.024240 -2022-12-06 10:39:35,354 - Epoch: [39][ 180/ 1200] Overall Loss 0.310329 Objective Loss 0.310329 LR 0.001000 Time 0.023963 -2022-12-06 10:39:35,547 - Epoch: [39][ 190/ 1200] Overall Loss 0.309085 Objective Loss 0.309085 LR 0.001000 Time 0.023716 -2022-12-06 10:39:35,741 - Epoch: [39][ 200/ 1200] Overall Loss 0.309713 Objective Loss 0.309713 LR 0.001000 Time 0.023494 -2022-12-06 10:39:35,935 - Epoch: [39][ 210/ 1200] Overall Loss 0.308700 Objective Loss 0.308700 LR 0.001000 Time 0.023297 -2022-12-06 10:39:36,128 - Epoch: [39][ 220/ 1200] Overall Loss 0.309720 Objective Loss 0.309720 LR 0.001000 Time 0.023113 -2022-12-06 10:39:36,321 - Epoch: [39][ 230/ 1200] Overall Loss 0.310856 Objective Loss 0.310856 LR 0.001000 Time 0.022947 -2022-12-06 10:39:36,514 - Epoch: [39][ 240/ 1200] Overall Loss 0.311419 Objective Loss 0.311419 LR 0.001000 Time 0.022793 -2022-12-06 10:39:36,708 - Epoch: [39][ 250/ 1200] Overall Loss 0.313195 Objective Loss 0.313195 LR 0.001000 Time 0.022655 -2022-12-06 10:39:36,900 - Epoch: [39][ 260/ 1200] Overall Loss 0.314070 Objective Loss 0.314070 LR 0.001000 Time 0.022521 -2022-12-06 10:39:37,094 - Epoch: [39][ 270/ 1200] Overall Loss 0.313826 Objective Loss 0.313826 LR 0.001000 Time 0.022402 -2022-12-06 10:39:37,287 - Epoch: [39][ 280/ 1200] Overall Loss 0.313534 Objective Loss 0.313534 LR 0.001000 Time 0.022290 -2022-12-06 10:39:37,481 - Epoch: [39][ 290/ 1200] Overall Loss 0.313844 Objective Loss 0.313844 LR 0.001000 Time 0.022189 -2022-12-06 10:39:37,675 - Epoch: [39][ 300/ 1200] Overall Loss 0.312058 Objective Loss 0.312058 LR 0.001000 Time 0.022092 -2022-12-06 10:39:37,869 - Epoch: [39][ 310/ 1200] Overall Loss 0.312002 Objective Loss 0.312002 LR 0.001000 Time 0.022004 -2022-12-06 10:39:38,062 - Epoch: [39][ 320/ 1200] Overall Loss 0.311234 Objective Loss 0.311234 LR 0.001000 Time 0.021919 -2022-12-06 10:39:38,256 - Epoch: [39][ 330/ 1200] Overall Loss 0.310079 Objective Loss 0.310079 LR 0.001000 Time 0.021841 -2022-12-06 10:39:38,449 - Epoch: [39][ 340/ 1200] Overall Loss 0.309490 Objective Loss 0.309490 LR 0.001000 Time 0.021763 -2022-12-06 10:39:38,642 - Epoch: [39][ 350/ 1200] Overall Loss 0.309500 Objective Loss 0.309500 LR 0.001000 Time 0.021692 -2022-12-06 10:39:38,835 - Epoch: [39][ 360/ 1200] Overall Loss 0.309539 Objective Loss 0.309539 LR 0.001000 Time 0.021624 -2022-12-06 10:39:39,029 - Epoch: [39][ 370/ 1200] Overall Loss 0.309669 Objective Loss 0.309669 LR 0.001000 Time 0.021562 -2022-12-06 10:39:39,221 - Epoch: [39][ 380/ 1200] Overall Loss 0.310351 Objective Loss 0.310351 LR 0.001000 Time 0.021501 -2022-12-06 10:39:39,415 - Epoch: [39][ 390/ 1200] Overall Loss 0.309969 Objective Loss 0.309969 LR 0.001000 Time 0.021445 -2022-12-06 10:39:39,609 - Epoch: [39][ 400/ 1200] Overall Loss 0.310381 Objective Loss 0.310381 LR 0.001000 Time 0.021392 -2022-12-06 10:39:39,803 - Epoch: [39][ 410/ 1200] Overall Loss 0.309236 Objective Loss 0.309236 LR 0.001000 Time 0.021343 -2022-12-06 10:39:39,996 - Epoch: [39][ 420/ 1200] Overall Loss 0.308912 Objective Loss 0.308912 LR 0.001000 Time 0.021292 -2022-12-06 10:39:40,190 - Epoch: [39][ 430/ 1200] Overall Loss 0.309106 Objective Loss 0.309106 LR 0.001000 Time 0.021246 -2022-12-06 10:39:40,383 - Epoch: [39][ 440/ 1200] Overall Loss 0.309160 Objective Loss 0.309160 LR 0.001000 Time 0.021202 -2022-12-06 10:39:40,577 - Epoch: [39][ 450/ 1200] Overall Loss 0.309533 Objective Loss 0.309533 LR 0.001000 Time 0.021161 -2022-12-06 10:39:40,771 - Epoch: [39][ 460/ 1200] Overall Loss 0.309678 Objective Loss 0.309678 LR 0.001000 Time 0.021120 -2022-12-06 10:39:40,964 - Epoch: [39][ 470/ 1200] Overall Loss 0.309624 Objective Loss 0.309624 LR 0.001000 Time 0.021081 -2022-12-06 10:39:41,157 - Epoch: [39][ 480/ 1200] Overall Loss 0.308937 Objective Loss 0.308937 LR 0.001000 Time 0.021042 -2022-12-06 10:39:41,350 - Epoch: [39][ 490/ 1200] Overall Loss 0.309204 Objective Loss 0.309204 LR 0.001000 Time 0.021006 -2022-12-06 10:39:41,543 - Epoch: [39][ 500/ 1200] Overall Loss 0.309903 Objective Loss 0.309903 LR 0.001000 Time 0.020972 -2022-12-06 10:39:41,738 - Epoch: [39][ 510/ 1200] Overall Loss 0.310052 Objective Loss 0.310052 LR 0.001000 Time 0.020941 -2022-12-06 10:39:41,932 - Epoch: [39][ 520/ 1200] Overall Loss 0.310681 Objective Loss 0.310681 LR 0.001000 Time 0.020910 -2022-12-06 10:39:42,126 - Epoch: [39][ 530/ 1200] Overall Loss 0.310654 Objective Loss 0.310654 LR 0.001000 Time 0.020880 -2022-12-06 10:39:42,320 - Epoch: [39][ 540/ 1200] Overall Loss 0.310511 Objective Loss 0.310511 LR 0.001000 Time 0.020852 -2022-12-06 10:39:42,514 - Epoch: [39][ 550/ 1200] Overall Loss 0.310316 Objective Loss 0.310316 LR 0.001000 Time 0.020825 -2022-12-06 10:39:42,708 - Epoch: [39][ 560/ 1200] Overall Loss 0.310602 Objective Loss 0.310602 LR 0.001000 Time 0.020799 -2022-12-06 10:39:42,903 - Epoch: [39][ 570/ 1200] Overall Loss 0.310516 Objective Loss 0.310516 LR 0.001000 Time 0.020775 -2022-12-06 10:39:43,096 - Epoch: [39][ 580/ 1200] Overall Loss 0.310019 Objective Loss 0.310019 LR 0.001000 Time 0.020749 -2022-12-06 10:39:43,290 - Epoch: [39][ 590/ 1200] Overall Loss 0.309819 Objective Loss 0.309819 LR 0.001000 Time 0.020725 -2022-12-06 10:39:43,484 - Epoch: [39][ 600/ 1200] Overall Loss 0.310007 Objective Loss 0.310007 LR 0.001000 Time 0.020702 -2022-12-06 10:39:43,678 - Epoch: [39][ 610/ 1200] Overall Loss 0.310210 Objective Loss 0.310210 LR 0.001000 Time 0.020680 -2022-12-06 10:39:43,871 - Epoch: [39][ 620/ 1200] Overall Loss 0.310424 Objective Loss 0.310424 LR 0.001000 Time 0.020656 -2022-12-06 10:39:44,065 - Epoch: [39][ 630/ 1200] Overall Loss 0.310673 Objective Loss 0.310673 LR 0.001000 Time 0.020635 -2022-12-06 10:39:44,258 - Epoch: [39][ 640/ 1200] Overall Loss 0.311041 Objective Loss 0.311041 LR 0.001000 Time 0.020613 -2022-12-06 10:39:44,451 - Epoch: [39][ 650/ 1200] Overall Loss 0.311153 Objective Loss 0.311153 LR 0.001000 Time 0.020594 -2022-12-06 10:39:44,645 - Epoch: [39][ 660/ 1200] Overall Loss 0.311263 Objective Loss 0.311263 LR 0.001000 Time 0.020574 -2022-12-06 10:39:44,839 - Epoch: [39][ 670/ 1200] Overall Loss 0.311478 Objective Loss 0.311478 LR 0.001000 Time 0.020555 -2022-12-06 10:39:45,033 - Epoch: [39][ 680/ 1200] Overall Loss 0.311688 Objective Loss 0.311688 LR 0.001000 Time 0.020537 -2022-12-06 10:39:45,226 - Epoch: [39][ 690/ 1200] Overall Loss 0.311671 Objective Loss 0.311671 LR 0.001000 Time 0.020519 -2022-12-06 10:39:45,418 - Epoch: [39][ 700/ 1200] Overall Loss 0.312008 Objective Loss 0.312008 LR 0.001000 Time 0.020500 -2022-12-06 10:39:45,610 - Epoch: [39][ 710/ 1200] Overall Loss 0.311937 Objective Loss 0.311937 LR 0.001000 Time 0.020480 -2022-12-06 10:39:45,802 - Epoch: [39][ 720/ 1200] Overall Loss 0.312208 Objective Loss 0.312208 LR 0.001000 Time 0.020462 -2022-12-06 10:39:45,994 - Epoch: [39][ 730/ 1200] Overall Loss 0.312116 Objective Loss 0.312116 LR 0.001000 Time 0.020444 -2022-12-06 10:39:46,186 - Epoch: [39][ 740/ 1200] Overall Loss 0.312293 Objective Loss 0.312293 LR 0.001000 Time 0.020426 -2022-12-06 10:39:46,378 - Epoch: [39][ 750/ 1200] Overall Loss 0.312221 Objective Loss 0.312221 LR 0.001000 Time 0.020409 -2022-12-06 10:39:46,570 - Epoch: [39][ 760/ 1200] Overall Loss 0.312225 Objective Loss 0.312225 LR 0.001000 Time 0.020393 -2022-12-06 10:39:46,762 - Epoch: [39][ 770/ 1200] Overall Loss 0.312236 Objective Loss 0.312236 LR 0.001000 Time 0.020377 -2022-12-06 10:39:46,955 - Epoch: [39][ 780/ 1200] Overall Loss 0.312235 Objective Loss 0.312235 LR 0.001000 Time 0.020362 -2022-12-06 10:39:47,147 - Epoch: [39][ 790/ 1200] Overall Loss 0.312194 Objective Loss 0.312194 LR 0.001000 Time 0.020347 -2022-12-06 10:39:47,339 - Epoch: [39][ 800/ 1200] Overall Loss 0.312246 Objective Loss 0.312246 LR 0.001000 Time 0.020331 -2022-12-06 10:39:47,531 - Epoch: [39][ 810/ 1200] Overall Loss 0.312416 Objective Loss 0.312416 LR 0.001000 Time 0.020317 -2022-12-06 10:39:47,723 - Epoch: [39][ 820/ 1200] Overall Loss 0.312359 Objective Loss 0.312359 LR 0.001000 Time 0.020302 -2022-12-06 10:39:47,914 - Epoch: [39][ 830/ 1200] Overall Loss 0.312756 Objective Loss 0.312756 LR 0.001000 Time 0.020288 -2022-12-06 10:39:48,106 - Epoch: [39][ 840/ 1200] Overall Loss 0.312815 Objective Loss 0.312815 LR 0.001000 Time 0.020274 -2022-12-06 10:39:48,298 - Epoch: [39][ 850/ 1200] Overall Loss 0.312828 Objective Loss 0.312828 LR 0.001000 Time 0.020261 -2022-12-06 10:39:48,490 - Epoch: [39][ 860/ 1200] Overall Loss 0.312487 Objective Loss 0.312487 LR 0.001000 Time 0.020248 -2022-12-06 10:39:48,682 - Epoch: [39][ 870/ 1200] Overall Loss 0.312472 Objective Loss 0.312472 LR 0.001000 Time 0.020236 -2022-12-06 10:39:48,874 - Epoch: [39][ 880/ 1200] Overall Loss 0.312325 Objective Loss 0.312325 LR 0.001000 Time 0.020223 -2022-12-06 10:39:49,066 - Epoch: [39][ 890/ 1200] Overall Loss 0.312253 Objective Loss 0.312253 LR 0.001000 Time 0.020211 -2022-12-06 10:39:49,258 - Epoch: [39][ 900/ 1200] Overall Loss 0.312534 Objective Loss 0.312534 LR 0.001000 Time 0.020199 -2022-12-06 10:39:49,450 - Epoch: [39][ 910/ 1200] Overall Loss 0.312498 Objective Loss 0.312498 LR 0.001000 Time 0.020187 -2022-12-06 10:39:49,642 - Epoch: [39][ 920/ 1200] Overall Loss 0.312640 Objective Loss 0.312640 LR 0.001000 Time 0.020176 -2022-12-06 10:39:49,834 - Epoch: [39][ 930/ 1200] Overall Loss 0.312574 Objective Loss 0.312574 LR 0.001000 Time 0.020165 -2022-12-06 10:39:50,026 - Epoch: [39][ 940/ 1200] Overall Loss 0.312725 Objective Loss 0.312725 LR 0.001000 Time 0.020154 -2022-12-06 10:39:50,218 - Epoch: [39][ 950/ 1200] Overall Loss 0.312913 Objective Loss 0.312913 LR 0.001000 Time 0.020143 -2022-12-06 10:39:50,410 - Epoch: [39][ 960/ 1200] Overall Loss 0.312895 Objective Loss 0.312895 LR 0.001000 Time 0.020133 -2022-12-06 10:39:50,602 - Epoch: [39][ 970/ 1200] Overall Loss 0.312698 Objective Loss 0.312698 LR 0.001000 Time 0.020123 -2022-12-06 10:39:50,794 - Epoch: [39][ 980/ 1200] Overall Loss 0.312679 Objective Loss 0.312679 LR 0.001000 Time 0.020114 -2022-12-06 10:39:50,986 - Epoch: [39][ 990/ 1200] Overall Loss 0.312620 Objective Loss 0.312620 LR 0.001000 Time 0.020103 -2022-12-06 10:39:51,178 - Epoch: [39][ 1000/ 1200] Overall Loss 0.312677 Objective Loss 0.312677 LR 0.001000 Time 0.020094 -2022-12-06 10:39:51,370 - Epoch: [39][ 1010/ 1200] Overall Loss 0.312744 Objective Loss 0.312744 LR 0.001000 Time 0.020085 -2022-12-06 10:39:51,562 - Epoch: [39][ 1020/ 1200] Overall Loss 0.312801 Objective Loss 0.312801 LR 0.001000 Time 0.020075 -2022-12-06 10:39:51,754 - Epoch: [39][ 1030/ 1200] Overall Loss 0.313143 Objective Loss 0.313143 LR 0.001000 Time 0.020066 -2022-12-06 10:39:51,945 - Epoch: [39][ 1040/ 1200] Overall Loss 0.313171 Objective Loss 0.313171 LR 0.001000 Time 0.020056 -2022-12-06 10:39:52,138 - Epoch: [39][ 1050/ 1200] Overall Loss 0.313546 Objective Loss 0.313546 LR 0.001000 Time 0.020048 -2022-12-06 10:39:52,329 - Epoch: [39][ 1060/ 1200] Overall Loss 0.313433 Objective Loss 0.313433 LR 0.001000 Time 0.020039 -2022-12-06 10:39:52,521 - Epoch: [39][ 1070/ 1200] Overall Loss 0.313328 Objective Loss 0.313328 LR 0.001000 Time 0.020031 -2022-12-06 10:39:52,712 - Epoch: [39][ 1080/ 1200] Overall Loss 0.313334 Objective Loss 0.313334 LR 0.001000 Time 0.020022 -2022-12-06 10:39:52,905 - Epoch: [39][ 1090/ 1200] Overall Loss 0.313506 Objective Loss 0.313506 LR 0.001000 Time 0.020014 -2022-12-06 10:39:53,097 - Epoch: [39][ 1100/ 1200] Overall Loss 0.313364 Objective Loss 0.313364 LR 0.001000 Time 0.020007 -2022-12-06 10:39:53,290 - Epoch: [39][ 1110/ 1200] Overall Loss 0.312998 Objective Loss 0.312998 LR 0.001000 Time 0.020000 -2022-12-06 10:39:53,483 - Epoch: [39][ 1120/ 1200] Overall Loss 0.313196 Objective Loss 0.313196 LR 0.001000 Time 0.019993 -2022-12-06 10:39:53,675 - Epoch: [39][ 1130/ 1200] Overall Loss 0.313163 Objective Loss 0.313163 LR 0.001000 Time 0.019986 -2022-12-06 10:39:53,867 - Epoch: [39][ 1140/ 1200] Overall Loss 0.313153 Objective Loss 0.313153 LR 0.001000 Time 0.019978 -2022-12-06 10:39:54,060 - Epoch: [39][ 1150/ 1200] Overall Loss 0.313218 Objective Loss 0.313218 LR 0.001000 Time 0.019972 -2022-12-06 10:39:54,252 - Epoch: [39][ 1160/ 1200] Overall Loss 0.313071 Objective Loss 0.313071 LR 0.001000 Time 0.019964 -2022-12-06 10:39:54,444 - Epoch: [39][ 1170/ 1200] Overall Loss 0.313060 Objective Loss 0.313060 LR 0.001000 Time 0.019958 -2022-12-06 10:39:54,636 - Epoch: [39][ 1180/ 1200] Overall Loss 0.313021 Objective Loss 0.313021 LR 0.001000 Time 0.019951 -2022-12-06 10:39:54,828 - Epoch: [39][ 1190/ 1200] Overall Loss 0.313112 Objective Loss 0.313112 LR 0.001000 Time 0.019945 -2022-12-06 10:39:55,061 - Epoch: [39][ 1200/ 1200] Overall Loss 0.313084 Objective Loss 0.313084 Top1 83.682008 Top5 97.698745 LR 0.001000 Time 0.019972 -2022-12-06 10:39:55,150 - --- validate (epoch=39)----------- -2022-12-06 10:39:55,150 - 34129 samples (256 per mini-batch) -2022-12-06 10:39:55,592 - Epoch: [39][ 10/ 134] Loss 0.290559 Top1 82.929688 Top5 98.046875 -2022-12-06 10:39:55,728 - Epoch: [39][ 20/ 134] Loss 0.308812 Top1 82.539062 Top5 98.066406 -2022-12-06 10:39:55,858 - Epoch: [39][ 30/ 134] Loss 0.311782 Top1 82.369792 Top5 97.656250 -2022-12-06 10:39:55,993 - Epoch: [39][ 40/ 134] Loss 0.313320 Top1 82.216797 Top5 97.470703 -2022-12-06 10:39:56,122 - Epoch: [39][ 50/ 134] Loss 0.320931 Top1 82.039062 Top5 97.398438 -2022-12-06 10:39:56,257 - Epoch: [39][ 60/ 134] Loss 0.318079 Top1 82.011719 Top5 97.500000 -2022-12-06 10:39:56,387 - Epoch: [39][ 70/ 134] Loss 0.320617 Top1 81.997768 Top5 97.388393 -2022-12-06 10:39:56,516 - Epoch: [39][ 80/ 134] Loss 0.319222 Top1 82.016602 Top5 97.392578 -2022-12-06 10:39:56,641 - Epoch: [39][ 90/ 134] Loss 0.322197 Top1 81.996528 Top5 97.378472 -2022-12-06 10:39:56,770 - Epoch: [39][ 100/ 134] Loss 0.322004 Top1 82.054688 Top5 97.367188 -2022-12-06 10:39:56,899 - Epoch: [39][ 110/ 134] Loss 0.321125 Top1 82.144886 Top5 97.418324 -2022-12-06 10:39:57,029 - Epoch: [39][ 120/ 134] Loss 0.321430 Top1 82.096354 Top5 97.369792 -2022-12-06 10:39:57,160 - Epoch: [39][ 130/ 134] Loss 0.319795 Top1 82.175481 Top5 97.364784 -2022-12-06 10:39:57,197 - Epoch: [39][ 134/ 134] Loss 0.319530 Top1 82.194029 Top5 97.386387 -2022-12-06 10:39:57,287 - ==> Top1: 82.194 Top5: 97.386 Loss: 0.320 - -2022-12-06 10:39:57,288 - ==> Confusion: -[[ 884 4 3 0 11 4 0 4 11 53 0 5 2 2 4 2 3 1 2 1 0] - [ 0 937 0 1 15 21 5 13 3 1 4 6 0 3 1 1 6 3 3 0 4] - [ 4 6 993 11 5 3 31 9 0 3 6 9 1 2 3 5 1 0 0 4 7] - [ 2 4 26 909 0 9 1 0 2 1 16 2 3 1 19 0 1 3 15 0 6] - [ 5 6 3 0 962 6 0 1 0 6 2 4 0 1 7 3 7 3 0 2 2] - [ 2 22 1 0 11 976 1 14 6 2 0 10 2 11 1 1 1 1 1 4 2] - [ 1 5 13 2 0 2 1061 7 0 0 3 6 0 1 0 4 0 1 1 11 0] - [ 1 14 8 1 2 28 6 943 0 0 1 14 0 2 0 0 1 0 18 13 2] - [ 3 5 0 2 0 3 1 0 984 36 7 3 1 8 4 0 2 1 2 2 0] - [ 60 0 2 0 5 1 0 3 31 882 1 4 0 8 1 1 0 0 0 0 2] - [ 0 4 3 3 0 1 0 5 11 2 958 3 1 16 3 0 3 1 1 3 1] - [ 3 2 1 0 0 12 3 2 3 1 1 984 5 9 1 4 4 2 0 12 2] - [ 0 2 2 4 2 2 3 1 0 0 0 71 838 2 0 15 2 17 1 3 4] - [ 0 1 0 0 3 8 0 1 14 14 8 10 1 951 1 3 2 0 0 1 5] - [ 7 6 2 10 7 1 0 0 29 4 3 3 1 4 1035 0 3 0 7 2 6] - [ 0 1 1 1 5 2 1 0 0 0 0 15 3 4 0 986 9 8 1 3 3] - [ 3 6 3 0 2 0 0 0 2 0 0 5 0 1 0 7 1027 3 0 9 4] - [ 3 1 0 2 1 1 2 0 2 1 0 16 12 2 1 12 4 971 1 2 2] - [ 1 11 9 5 1 2 0 30 3 0 15 5 1 2 8 1 1 1 902 6 4] - [ 1 4 1 0 1 6 2 6 0 0 2 18 1 5 1 1 5 4 0 1018 4] - [ 177 361 244 79 220 238 91 183 126 129 296 214 332 432 181 148 352 79 138 364 8842]] - -2022-12-06 10:39:57,853 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:39:57,853 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:39:57,859 - - -2022-12-06 10:39:57,859 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:39:58,897 - Epoch: [40][ 10/ 1200] Overall Loss 0.301325 Objective Loss 0.301325 LR 0.001000 Time 0.103744 -2022-12-06 10:39:59,095 - Epoch: [40][ 20/ 1200] Overall Loss 0.290732 Objective Loss 0.290732 LR 0.001000 Time 0.061751 -2022-12-06 10:39:59,291 - Epoch: [40][ 30/ 1200] Overall Loss 0.288576 Objective Loss 0.288576 LR 0.001000 Time 0.047651 -2022-12-06 10:39:59,488 - Epoch: [40][ 40/ 1200] Overall Loss 0.293271 Objective Loss 0.293271 LR 0.001000 Time 0.040663 -2022-12-06 10:39:59,683 - Epoch: [40][ 50/ 1200] Overall Loss 0.295063 Objective Loss 0.295063 LR 0.001000 Time 0.036413 -2022-12-06 10:39:59,880 - Epoch: [40][ 60/ 1200] Overall Loss 0.293865 Objective Loss 0.293865 LR 0.001000 Time 0.033621 -2022-12-06 10:40:00,075 - Epoch: [40][ 70/ 1200] Overall Loss 0.293916 Objective Loss 0.293916 LR 0.001000 Time 0.031597 -2022-12-06 10:40:00,272 - Epoch: [40][ 80/ 1200] Overall Loss 0.291454 Objective Loss 0.291454 LR 0.001000 Time 0.030105 -2022-12-06 10:40:00,466 - Epoch: [40][ 90/ 1200] Overall Loss 0.293411 Objective Loss 0.293411 LR 0.001000 Time 0.028911 -2022-12-06 10:40:00,665 - Epoch: [40][ 100/ 1200] Overall Loss 0.294537 Objective Loss 0.294537 LR 0.001000 Time 0.028000 -2022-12-06 10:40:00,859 - Epoch: [40][ 110/ 1200] Overall Loss 0.294875 Objective Loss 0.294875 LR 0.001000 Time 0.027214 -2022-12-06 10:40:01,055 - Epoch: [40][ 120/ 1200] Overall Loss 0.293127 Objective Loss 0.293127 LR 0.001000 Time 0.026577 -2022-12-06 10:40:01,250 - Epoch: [40][ 130/ 1200] Overall Loss 0.293504 Objective Loss 0.293504 LR 0.001000 Time 0.026026 -2022-12-06 10:40:01,447 - Epoch: [40][ 140/ 1200] Overall Loss 0.294030 Objective Loss 0.294030 LR 0.001000 Time 0.025570 -2022-12-06 10:40:01,641 - Epoch: [40][ 150/ 1200] Overall Loss 0.293483 Objective Loss 0.293483 LR 0.001000 Time 0.025159 -2022-12-06 10:40:01,839 - Epoch: [40][ 160/ 1200] Overall Loss 0.294377 Objective Loss 0.294377 LR 0.001000 Time 0.024816 -2022-12-06 10:40:02,033 - Epoch: [40][ 170/ 1200] Overall Loss 0.293721 Objective Loss 0.293721 LR 0.001000 Time 0.024494 -2022-12-06 10:40:02,230 - Epoch: [40][ 180/ 1200] Overall Loss 0.293683 Objective Loss 0.293683 LR 0.001000 Time 0.024226 -2022-12-06 10:40:02,425 - Epoch: [40][ 190/ 1200] Overall Loss 0.293016 Objective Loss 0.293016 LR 0.001000 Time 0.023973 -2022-12-06 10:40:02,621 - Epoch: [40][ 200/ 1200] Overall Loss 0.292870 Objective Loss 0.292870 LR 0.001000 Time 0.023754 -2022-12-06 10:40:02,816 - Epoch: [40][ 210/ 1200] Overall Loss 0.292975 Objective Loss 0.292975 LR 0.001000 Time 0.023549 -2022-12-06 10:40:03,013 - Epoch: [40][ 220/ 1200] Overall Loss 0.293529 Objective Loss 0.293529 LR 0.001000 Time 0.023371 -2022-12-06 10:40:03,208 - Epoch: [40][ 230/ 1200] Overall Loss 0.293723 Objective Loss 0.293723 LR 0.001000 Time 0.023198 -2022-12-06 10:40:03,405 - Epoch: [40][ 240/ 1200] Overall Loss 0.294763 Objective Loss 0.294763 LR 0.001000 Time 0.023051 -2022-12-06 10:40:03,599 - Epoch: [40][ 250/ 1200] Overall Loss 0.295727 Objective Loss 0.295727 LR 0.001000 Time 0.022906 -2022-12-06 10:40:03,797 - Epoch: [40][ 260/ 1200] Overall Loss 0.295497 Objective Loss 0.295497 LR 0.001000 Time 0.022781 -2022-12-06 10:40:03,991 - Epoch: [40][ 270/ 1200] Overall Loss 0.295552 Objective Loss 0.295552 LR 0.001000 Time 0.022655 -2022-12-06 10:40:04,188 - Epoch: [40][ 280/ 1200] Overall Loss 0.294665 Objective Loss 0.294665 LR 0.001000 Time 0.022547 -2022-12-06 10:40:04,382 - Epoch: [40][ 290/ 1200] Overall Loss 0.294494 Objective Loss 0.294494 LR 0.001000 Time 0.022437 -2022-12-06 10:40:04,579 - Epoch: [40][ 300/ 1200] Overall Loss 0.294108 Objective Loss 0.294108 LR 0.001000 Time 0.022345 -2022-12-06 10:40:04,774 - Epoch: [40][ 310/ 1200] Overall Loss 0.293936 Objective Loss 0.293936 LR 0.001000 Time 0.022251 -2022-12-06 10:40:04,972 - Epoch: [40][ 320/ 1200] Overall Loss 0.293120 Objective Loss 0.293120 LR 0.001000 Time 0.022171 -2022-12-06 10:40:05,166 - Epoch: [40][ 330/ 1200] Overall Loss 0.292530 Objective Loss 0.292530 LR 0.001000 Time 0.022087 -2022-12-06 10:40:05,363 - Epoch: [40][ 340/ 1200] Overall Loss 0.292460 Objective Loss 0.292460 LR 0.001000 Time 0.022014 -2022-12-06 10:40:05,557 - Epoch: [40][ 350/ 1200] Overall Loss 0.293843 Objective Loss 0.293843 LR 0.001000 Time 0.021938 -2022-12-06 10:40:05,754 - Epoch: [40][ 360/ 1200] Overall Loss 0.294328 Objective Loss 0.294328 LR 0.001000 Time 0.021875 -2022-12-06 10:40:05,950 - Epoch: [40][ 370/ 1200] Overall Loss 0.294827 Objective Loss 0.294827 LR 0.001000 Time 0.021811 -2022-12-06 10:40:06,148 - Epoch: [40][ 380/ 1200] Overall Loss 0.294517 Objective Loss 0.294517 LR 0.001000 Time 0.021757 -2022-12-06 10:40:06,342 - Epoch: [40][ 390/ 1200] Overall Loss 0.294234 Objective Loss 0.294234 LR 0.001000 Time 0.021695 -2022-12-06 10:40:06,539 - Epoch: [40][ 400/ 1200] Overall Loss 0.293876 Objective Loss 0.293876 LR 0.001000 Time 0.021644 -2022-12-06 10:40:06,734 - Epoch: [40][ 410/ 1200] Overall Loss 0.294274 Objective Loss 0.294274 LR 0.001000 Time 0.021591 -2022-12-06 10:40:06,931 - Epoch: [40][ 420/ 1200] Overall Loss 0.295281 Objective Loss 0.295281 LR 0.001000 Time 0.021545 -2022-12-06 10:40:07,126 - Epoch: [40][ 430/ 1200] Overall Loss 0.295550 Objective Loss 0.295550 LR 0.001000 Time 0.021495 -2022-12-06 10:40:07,323 - Epoch: [40][ 440/ 1200] Overall Loss 0.296343 Objective Loss 0.296343 LR 0.001000 Time 0.021453 -2022-12-06 10:40:07,517 - Epoch: [40][ 450/ 1200] Overall Loss 0.296516 Objective Loss 0.296516 LR 0.001000 Time 0.021406 -2022-12-06 10:40:07,714 - Epoch: [40][ 460/ 1200] Overall Loss 0.296536 Objective Loss 0.296536 LR 0.001000 Time 0.021368 -2022-12-06 10:40:07,909 - Epoch: [40][ 470/ 1200] Overall Loss 0.296591 Objective Loss 0.296591 LR 0.001000 Time 0.021327 -2022-12-06 10:40:08,105 - Epoch: [40][ 480/ 1200] Overall Loss 0.296718 Objective Loss 0.296718 LR 0.001000 Time 0.021291 -2022-12-06 10:40:08,301 - Epoch: [40][ 490/ 1200] Overall Loss 0.296775 Objective Loss 0.296775 LR 0.001000 Time 0.021254 -2022-12-06 10:40:08,498 - Epoch: [40][ 500/ 1200] Overall Loss 0.296305 Objective Loss 0.296305 LR 0.001000 Time 0.021222 -2022-12-06 10:40:08,692 - Epoch: [40][ 510/ 1200] Overall Loss 0.296045 Objective Loss 0.296045 LR 0.001000 Time 0.021187 -2022-12-06 10:40:08,890 - Epoch: [40][ 520/ 1200] Overall Loss 0.296490 Objective Loss 0.296490 LR 0.001000 Time 0.021158 -2022-12-06 10:40:09,084 - Epoch: [40][ 530/ 1200] Overall Loss 0.296946 Objective Loss 0.296946 LR 0.001000 Time 0.021124 -2022-12-06 10:40:09,282 - Epoch: [40][ 540/ 1200] Overall Loss 0.296946 Objective Loss 0.296946 LR 0.001000 Time 0.021099 -2022-12-06 10:40:09,476 - Epoch: [40][ 550/ 1200] Overall Loss 0.296909 Objective Loss 0.296909 LR 0.001000 Time 0.021067 -2022-12-06 10:40:09,673 - Epoch: [40][ 560/ 1200] Overall Loss 0.297181 Objective Loss 0.297181 LR 0.001000 Time 0.021042 -2022-12-06 10:40:09,868 - Epoch: [40][ 570/ 1200] Overall Loss 0.297678 Objective Loss 0.297678 LR 0.001000 Time 0.021014 -2022-12-06 10:40:10,065 - Epoch: [40][ 580/ 1200] Overall Loss 0.298312 Objective Loss 0.298312 LR 0.001000 Time 0.020990 -2022-12-06 10:40:10,260 - Epoch: [40][ 590/ 1200] Overall Loss 0.298466 Objective Loss 0.298466 LR 0.001000 Time 0.020964 -2022-12-06 10:40:10,457 - Epoch: [40][ 600/ 1200] Overall Loss 0.298380 Objective Loss 0.298380 LR 0.001000 Time 0.020942 -2022-12-06 10:40:10,652 - Epoch: [40][ 610/ 1200] Overall Loss 0.298494 Objective Loss 0.298494 LR 0.001000 Time 0.020917 -2022-12-06 10:40:10,849 - Epoch: [40][ 620/ 1200] Overall Loss 0.298349 Objective Loss 0.298349 LR 0.001000 Time 0.020897 -2022-12-06 10:40:11,045 - Epoch: [40][ 630/ 1200] Overall Loss 0.298318 Objective Loss 0.298318 LR 0.001000 Time 0.020875 -2022-12-06 10:40:11,242 - Epoch: [40][ 640/ 1200] Overall Loss 0.298200 Objective Loss 0.298200 LR 0.001000 Time 0.020857 -2022-12-06 10:40:11,437 - Epoch: [40][ 650/ 1200] Overall Loss 0.298494 Objective Loss 0.298494 LR 0.001000 Time 0.020834 -2022-12-06 10:40:11,632 - Epoch: [40][ 660/ 1200] Overall Loss 0.298756 Objective Loss 0.298756 LR 0.001000 Time 0.020813 -2022-12-06 10:40:11,822 - Epoch: [40][ 670/ 1200] Overall Loss 0.298778 Objective Loss 0.298778 LR 0.001000 Time 0.020786 -2022-12-06 10:40:12,013 - Epoch: [40][ 680/ 1200] Overall Loss 0.298953 Objective Loss 0.298953 LR 0.001000 Time 0.020759 -2022-12-06 10:40:12,203 - Epoch: [40][ 690/ 1200] Overall Loss 0.298818 Objective Loss 0.298818 LR 0.001000 Time 0.020733 -2022-12-06 10:40:12,392 - Epoch: [40][ 700/ 1200] Overall Loss 0.298857 Objective Loss 0.298857 LR 0.001000 Time 0.020706 -2022-12-06 10:40:12,582 - Epoch: [40][ 710/ 1200] Overall Loss 0.298855 Objective Loss 0.298855 LR 0.001000 Time 0.020681 -2022-12-06 10:40:12,772 - Epoch: [40][ 720/ 1200] Overall Loss 0.298915 Objective Loss 0.298915 LR 0.001000 Time 0.020657 -2022-12-06 10:40:12,963 - Epoch: [40][ 730/ 1200] Overall Loss 0.299009 Objective Loss 0.299009 LR 0.001000 Time 0.020635 -2022-12-06 10:40:13,153 - Epoch: [40][ 740/ 1200] Overall Loss 0.298660 Objective Loss 0.298660 LR 0.001000 Time 0.020612 -2022-12-06 10:40:13,343 - Epoch: [40][ 750/ 1200] Overall Loss 0.298730 Objective Loss 0.298730 LR 0.001000 Time 0.020591 -2022-12-06 10:40:13,534 - Epoch: [40][ 760/ 1200] Overall Loss 0.299230 Objective Loss 0.299230 LR 0.001000 Time 0.020570 -2022-12-06 10:40:13,723 - Epoch: [40][ 770/ 1200] Overall Loss 0.299515 Objective Loss 0.299515 LR 0.001000 Time 0.020548 -2022-12-06 10:40:13,913 - Epoch: [40][ 780/ 1200] Overall Loss 0.299736 Objective Loss 0.299736 LR 0.001000 Time 0.020527 -2022-12-06 10:40:14,103 - Epoch: [40][ 790/ 1200] Overall Loss 0.300269 Objective Loss 0.300269 LR 0.001000 Time 0.020507 -2022-12-06 10:40:14,293 - Epoch: [40][ 800/ 1200] Overall Loss 0.300514 Objective Loss 0.300514 LR 0.001000 Time 0.020487 -2022-12-06 10:40:14,483 - Epoch: [40][ 810/ 1200] Overall Loss 0.300389 Objective Loss 0.300389 LR 0.001000 Time 0.020468 -2022-12-06 10:40:14,674 - Epoch: [40][ 820/ 1200] Overall Loss 0.300856 Objective Loss 0.300856 LR 0.001000 Time 0.020451 -2022-12-06 10:40:14,864 - Epoch: [40][ 830/ 1200] Overall Loss 0.301085 Objective Loss 0.301085 LR 0.001000 Time 0.020433 -2022-12-06 10:40:15,055 - Epoch: [40][ 840/ 1200] Overall Loss 0.300837 Objective Loss 0.300837 LR 0.001000 Time 0.020417 -2022-12-06 10:40:15,244 - Epoch: [40][ 850/ 1200] Overall Loss 0.300953 Objective Loss 0.300953 LR 0.001000 Time 0.020398 -2022-12-06 10:40:15,435 - Epoch: [40][ 860/ 1200] Overall Loss 0.301207 Objective Loss 0.301207 LR 0.001000 Time 0.020383 -2022-12-06 10:40:15,626 - Epoch: [40][ 870/ 1200] Overall Loss 0.301009 Objective Loss 0.301009 LR 0.001000 Time 0.020366 -2022-12-06 10:40:15,816 - Epoch: [40][ 880/ 1200] Overall Loss 0.300632 Objective Loss 0.300632 LR 0.001000 Time 0.020350 -2022-12-06 10:40:16,005 - Epoch: [40][ 890/ 1200] Overall Loss 0.300249 Objective Loss 0.300249 LR 0.001000 Time 0.020334 -2022-12-06 10:40:16,195 - Epoch: [40][ 900/ 1200] Overall Loss 0.300033 Objective Loss 0.300033 LR 0.001000 Time 0.020319 -2022-12-06 10:40:16,385 - Epoch: [40][ 910/ 1200] Overall Loss 0.300162 Objective Loss 0.300162 LR 0.001000 Time 0.020303 -2022-12-06 10:40:16,575 - Epoch: [40][ 920/ 1200] Overall Loss 0.300166 Objective Loss 0.300166 LR 0.001000 Time 0.020289 -2022-12-06 10:40:16,765 - Epoch: [40][ 930/ 1200] Overall Loss 0.300187 Objective Loss 0.300187 LR 0.001000 Time 0.020274 -2022-12-06 10:40:16,955 - Epoch: [40][ 940/ 1200] Overall Loss 0.300135 Objective Loss 0.300135 LR 0.001000 Time 0.020260 -2022-12-06 10:40:17,145 - Epoch: [40][ 950/ 1200] Overall Loss 0.300233 Objective Loss 0.300233 LR 0.001000 Time 0.020246 -2022-12-06 10:40:17,335 - Epoch: [40][ 960/ 1200] Overall Loss 0.300172 Objective Loss 0.300172 LR 0.001000 Time 0.020232 -2022-12-06 10:40:17,524 - Epoch: [40][ 970/ 1200] Overall Loss 0.300058 Objective Loss 0.300058 LR 0.001000 Time 0.020219 -2022-12-06 10:40:17,714 - Epoch: [40][ 980/ 1200] Overall Loss 0.299971 Objective Loss 0.299971 LR 0.001000 Time 0.020206 -2022-12-06 10:40:17,904 - Epoch: [40][ 990/ 1200] Overall Loss 0.300007 Objective Loss 0.300007 LR 0.001000 Time 0.020193 -2022-12-06 10:40:18,094 - Epoch: [40][ 1000/ 1200] Overall Loss 0.300112 Objective Loss 0.300112 LR 0.001000 Time 0.020180 -2022-12-06 10:40:18,284 - Epoch: [40][ 1010/ 1200] Overall Loss 0.300460 Objective Loss 0.300460 LR 0.001000 Time 0.020168 -2022-12-06 10:40:18,474 - Epoch: [40][ 1020/ 1200] Overall Loss 0.300878 Objective Loss 0.300878 LR 0.001000 Time 0.020156 -2022-12-06 10:40:18,664 - Epoch: [40][ 1030/ 1200] Overall Loss 0.301051 Objective Loss 0.301051 LR 0.001000 Time 0.020144 -2022-12-06 10:40:18,854 - Epoch: [40][ 1040/ 1200] Overall Loss 0.301371 Objective Loss 0.301371 LR 0.001000 Time 0.020132 -2022-12-06 10:40:19,043 - Epoch: [40][ 1050/ 1200] Overall Loss 0.301240 Objective Loss 0.301240 LR 0.001000 Time 0.020121 -2022-12-06 10:40:19,233 - Epoch: [40][ 1060/ 1200] Overall Loss 0.301143 Objective Loss 0.301143 LR 0.001000 Time 0.020110 -2022-12-06 10:40:19,423 - Epoch: [40][ 1070/ 1200] Overall Loss 0.301320 Objective Loss 0.301320 LR 0.001000 Time 0.020099 -2022-12-06 10:40:19,614 - Epoch: [40][ 1080/ 1200] Overall Loss 0.301414 Objective Loss 0.301414 LR 0.001000 Time 0.020088 -2022-12-06 10:40:19,804 - Epoch: [40][ 1090/ 1200] Overall Loss 0.301441 Objective Loss 0.301441 LR 0.001000 Time 0.020078 -2022-12-06 10:40:19,994 - Epoch: [40][ 1100/ 1200] Overall Loss 0.301326 Objective Loss 0.301326 LR 0.001000 Time 0.020068 -2022-12-06 10:40:20,184 - Epoch: [40][ 1110/ 1200] Overall Loss 0.301195 Objective Loss 0.301195 LR 0.001000 Time 0.020058 -2022-12-06 10:40:20,374 - Epoch: [40][ 1120/ 1200] Overall Loss 0.301020 Objective Loss 0.301020 LR 0.001000 Time 0.020048 -2022-12-06 10:40:20,564 - Epoch: [40][ 1130/ 1200] Overall Loss 0.301026 Objective Loss 0.301026 LR 0.001000 Time 0.020038 -2022-12-06 10:40:20,753 - Epoch: [40][ 1140/ 1200] Overall Loss 0.301252 Objective Loss 0.301252 LR 0.001000 Time 0.020028 -2022-12-06 10:40:20,943 - Epoch: [40][ 1150/ 1200] Overall Loss 0.301227 Objective Loss 0.301227 LR 0.001000 Time 0.020019 -2022-12-06 10:40:21,134 - Epoch: [40][ 1160/ 1200] Overall Loss 0.301294 Objective Loss 0.301294 LR 0.001000 Time 0.020010 -2022-12-06 10:40:21,323 - Epoch: [40][ 1170/ 1200] Overall Loss 0.301142 Objective Loss 0.301142 LR 0.001000 Time 0.020000 -2022-12-06 10:40:21,513 - Epoch: [40][ 1180/ 1200] Overall Loss 0.301371 Objective Loss 0.301371 LR 0.001000 Time 0.019991 -2022-12-06 10:40:21,703 - Epoch: [40][ 1190/ 1200] Overall Loss 0.301281 Objective Loss 0.301281 LR 0.001000 Time 0.019982 -2022-12-06 10:40:21,935 - Epoch: [40][ 1200/ 1200] Overall Loss 0.301313 Objective Loss 0.301313 Top1 82.426778 Top5 97.907950 LR 0.001000 Time 0.020009 -2022-12-06 10:40:22,024 - --- validate (epoch=40)----------- -2022-12-06 10:40:22,024 - 34129 samples (256 per mini-batch) -2022-12-06 10:40:22,469 - Epoch: [40][ 10/ 134] Loss 0.319633 Top1 85.273438 Top5 97.109375 -2022-12-06 10:40:22,598 - Epoch: [40][ 20/ 134] Loss 0.301605 Top1 85.000000 Top5 97.617188 -2022-12-06 10:40:22,727 - Epoch: [40][ 30/ 134] Loss 0.299373 Top1 84.739583 Top5 97.786458 -2022-12-06 10:40:22,855 - Epoch: [40][ 40/ 134] Loss 0.302186 Top1 84.667969 Top5 97.802734 -2022-12-06 10:40:22,981 - Epoch: [40][ 50/ 134] Loss 0.312085 Top1 84.296875 Top5 97.804688 -2022-12-06 10:40:23,107 - Epoch: [40][ 60/ 134] Loss 0.315997 Top1 84.088542 Top5 97.786458 -2022-12-06 10:40:23,235 - Epoch: [40][ 70/ 134] Loss 0.315563 Top1 83.956473 Top5 97.806920 -2022-12-06 10:40:23,362 - Epoch: [40][ 80/ 134] Loss 0.315521 Top1 83.989258 Top5 97.807617 -2022-12-06 10:40:23,484 - Epoch: [40][ 90/ 134] Loss 0.314794 Top1 83.988715 Top5 97.812500 -2022-12-06 10:40:23,615 - Epoch: [40][ 100/ 134] Loss 0.315442 Top1 83.968750 Top5 97.800781 -2022-12-06 10:40:23,744 - Epoch: [40][ 110/ 134] Loss 0.315997 Top1 83.938210 Top5 97.766335 -2022-12-06 10:40:23,876 - Epoch: [40][ 120/ 134] Loss 0.314313 Top1 84.003906 Top5 97.747396 -2022-12-06 10:40:24,001 - Epoch: [40][ 130/ 134] Loss 0.316605 Top1 83.912260 Top5 97.719351 -2022-12-06 10:40:24,038 - Epoch: [40][ 134/ 134] Loss 0.315233 Top1 83.911043 Top5 97.737994 -2022-12-06 10:40:24,133 - ==> Top1: 83.911 Top5: 97.738 Loss: 0.315 - -2022-12-06 10:40:24,134 - ==> Confusion: -[[ 875 2 6 1 3 10 0 3 7 65 0 2 1 1 7 2 3 1 2 2 3] - [ 1 907 2 2 7 26 3 23 0 0 7 2 3 2 1 2 3 3 19 3 11] - [ 3 6 989 12 1 1 42 12 0 2 6 5 0 1 4 4 1 0 3 1 10] - [ 1 2 28 912 0 4 1 1 0 0 17 3 3 1 19 1 0 5 17 0 5] - [ 8 5 3 2 929 9 2 1 2 7 2 6 0 2 9 8 9 5 1 3 7] - [ 0 11 0 3 4 978 3 23 5 0 1 13 1 7 2 0 1 5 1 4 7] - [ 0 3 4 1 0 1 1081 3 0 0 2 3 1 2 0 6 1 1 1 6 2] - [ 0 4 10 0 2 21 12 940 2 0 3 7 0 0 1 1 0 0 32 16 3] - [ 2 3 0 0 0 4 1 1 959 43 9 2 2 11 10 0 1 2 6 4 4] - [ 47 0 3 0 0 9 1 6 16 892 2 3 0 9 2 0 0 1 2 0 8] - [ 1 1 5 7 2 1 5 1 7 2 953 2 2 13 4 0 1 0 8 1 3] - [ 2 0 2 0 0 10 6 4 1 0 1 963 19 3 0 10 2 13 2 11 2] - [ 1 0 2 6 0 2 4 1 0 0 0 38 864 2 2 8 0 24 1 6 8] - [ 0 1 0 0 0 19 0 4 5 13 14 13 3 934 0 2 0 3 0 5 7] - [ 3 2 6 14 2 2 0 0 21 2 1 4 1 2 1041 1 1 1 11 1 14] - [ 0 1 1 0 0 3 4 0 0 0 1 6 4 4 1 992 3 13 1 5 4] - [ 0 4 2 1 3 3 1 1 1 1 1 7 1 2 2 19 998 2 2 10 11] - [ 2 0 0 1 1 4 3 0 1 1 0 11 13 2 2 8 0 978 4 2 3] - [ 1 5 7 11 0 3 2 27 0 0 6 4 2 1 7 0 1 1 923 4 3] - [ 1 2 1 0 0 6 7 6 0 0 2 22 5 4 1 5 2 4 4 1005 3] - [ 94 240 242 93 76 231 141 203 94 91 221 140 388 324 153 160 131 128 250 309 9517]] - -2022-12-06 10:40:24,716 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:40:24,716 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:40:24,722 - - -2022-12-06 10:40:24,722 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:40:25,645 - Epoch: [41][ 10/ 1200] Overall Loss 0.301122 Objective Loss 0.301122 LR 0.001000 Time 0.092224 -2022-12-06 10:40:25,847 - Epoch: [41][ 20/ 1200] Overall Loss 0.299037 Objective Loss 0.299037 LR 0.001000 Time 0.056145 -2022-12-06 10:40:26,037 - Epoch: [41][ 30/ 1200] Overall Loss 0.296455 Objective Loss 0.296455 LR 0.001000 Time 0.043776 -2022-12-06 10:40:26,227 - Epoch: [41][ 40/ 1200] Overall Loss 0.293846 Objective Loss 0.293846 LR 0.001000 Time 0.037562 -2022-12-06 10:40:26,417 - Epoch: [41][ 50/ 1200] Overall Loss 0.295604 Objective Loss 0.295604 LR 0.001000 Time 0.033841 -2022-12-06 10:40:26,607 - Epoch: [41][ 60/ 1200] Overall Loss 0.295906 Objective Loss 0.295906 LR 0.001000 Time 0.031350 -2022-12-06 10:40:26,797 - Epoch: [41][ 70/ 1200] Overall Loss 0.294767 Objective Loss 0.294767 LR 0.001000 Time 0.029578 -2022-12-06 10:40:26,988 - Epoch: [41][ 80/ 1200] Overall Loss 0.292369 Objective Loss 0.292369 LR 0.001000 Time 0.028258 -2022-12-06 10:40:27,177 - Epoch: [41][ 90/ 1200] Overall Loss 0.291207 Objective Loss 0.291207 LR 0.001000 Time 0.027223 -2022-12-06 10:40:27,368 - Epoch: [41][ 100/ 1200] Overall Loss 0.292105 Objective Loss 0.292105 LR 0.001000 Time 0.026395 -2022-12-06 10:40:27,557 - Epoch: [41][ 110/ 1200] Overall Loss 0.295615 Objective Loss 0.295615 LR 0.001000 Time 0.025718 -2022-12-06 10:40:27,747 - Epoch: [41][ 120/ 1200] Overall Loss 0.295017 Objective Loss 0.295017 LR 0.001000 Time 0.025151 -2022-12-06 10:40:27,938 - Epoch: [41][ 130/ 1200] Overall Loss 0.293084 Objective Loss 0.293084 LR 0.001000 Time 0.024678 -2022-12-06 10:40:28,127 - Epoch: [41][ 140/ 1200] Overall Loss 0.293757 Objective Loss 0.293757 LR 0.001000 Time 0.024266 -2022-12-06 10:40:28,318 - Epoch: [41][ 150/ 1200] Overall Loss 0.293793 Objective Loss 0.293793 LR 0.001000 Time 0.023915 -2022-12-06 10:40:28,507 - Epoch: [41][ 160/ 1200] Overall Loss 0.294055 Objective Loss 0.294055 LR 0.001000 Time 0.023602 -2022-12-06 10:40:28,698 - Epoch: [41][ 170/ 1200] Overall Loss 0.294821 Objective Loss 0.294821 LR 0.001000 Time 0.023329 -2022-12-06 10:40:28,888 - Epoch: [41][ 180/ 1200] Overall Loss 0.293281 Objective Loss 0.293281 LR 0.001000 Time 0.023087 -2022-12-06 10:40:29,079 - Epoch: [41][ 190/ 1200] Overall Loss 0.295514 Objective Loss 0.295514 LR 0.001000 Time 0.022873 -2022-12-06 10:40:29,269 - Epoch: [41][ 200/ 1200] Overall Loss 0.295198 Objective Loss 0.295198 LR 0.001000 Time 0.022679 -2022-12-06 10:40:29,460 - Epoch: [41][ 210/ 1200] Overall Loss 0.295172 Objective Loss 0.295172 LR 0.001000 Time 0.022505 -2022-12-06 10:40:29,650 - Epoch: [41][ 220/ 1200] Overall Loss 0.295442 Objective Loss 0.295442 LR 0.001000 Time 0.022342 -2022-12-06 10:40:29,841 - Epoch: [41][ 230/ 1200] Overall Loss 0.295948 Objective Loss 0.295948 LR 0.001000 Time 0.022199 -2022-12-06 10:40:30,032 - Epoch: [41][ 240/ 1200] Overall Loss 0.296106 Objective Loss 0.296106 LR 0.001000 Time 0.022068 -2022-12-06 10:40:30,222 - Epoch: [41][ 250/ 1200] Overall Loss 0.296133 Objective Loss 0.296133 LR 0.001000 Time 0.021944 -2022-12-06 10:40:30,413 - Epoch: [41][ 260/ 1200] Overall Loss 0.296279 Objective Loss 0.296279 LR 0.001000 Time 0.021832 -2022-12-06 10:40:30,604 - Epoch: [41][ 270/ 1200] Overall Loss 0.295768 Objective Loss 0.295768 LR 0.001000 Time 0.021728 -2022-12-06 10:40:30,794 - Epoch: [41][ 280/ 1200] Overall Loss 0.294566 Objective Loss 0.294566 LR 0.001000 Time 0.021631 -2022-12-06 10:40:30,984 - Epoch: [41][ 290/ 1200] Overall Loss 0.294155 Objective Loss 0.294155 LR 0.001000 Time 0.021539 -2022-12-06 10:40:31,175 - Epoch: [41][ 300/ 1200] Overall Loss 0.293192 Objective Loss 0.293192 LR 0.001000 Time 0.021453 -2022-12-06 10:40:31,366 - Epoch: [41][ 310/ 1200] Overall Loss 0.293439 Objective Loss 0.293439 LR 0.001000 Time 0.021375 -2022-12-06 10:40:31,557 - Epoch: [41][ 320/ 1200] Overall Loss 0.292893 Objective Loss 0.292893 LR 0.001000 Time 0.021302 -2022-12-06 10:40:31,747 - Epoch: [41][ 330/ 1200] Overall Loss 0.292709 Objective Loss 0.292709 LR 0.001000 Time 0.021231 -2022-12-06 10:40:31,936 - Epoch: [41][ 340/ 1200] Overall Loss 0.292623 Objective Loss 0.292623 LR 0.001000 Time 0.021163 -2022-12-06 10:40:32,127 - Epoch: [41][ 350/ 1200] Overall Loss 0.292603 Objective Loss 0.292603 LR 0.001000 Time 0.021102 -2022-12-06 10:40:32,318 - Epoch: [41][ 360/ 1200] Overall Loss 0.292951 Objective Loss 0.292951 LR 0.001000 Time 0.021044 -2022-12-06 10:40:32,508 - Epoch: [41][ 370/ 1200] Overall Loss 0.292740 Objective Loss 0.292740 LR 0.001000 Time 0.020989 -2022-12-06 10:40:32,699 - Epoch: [41][ 380/ 1200] Overall Loss 0.292315 Objective Loss 0.292315 LR 0.001000 Time 0.020937 -2022-12-06 10:40:32,890 - Epoch: [41][ 390/ 1200] Overall Loss 0.291946 Objective Loss 0.291946 LR 0.001000 Time 0.020889 -2022-12-06 10:40:33,082 - Epoch: [41][ 400/ 1200] Overall Loss 0.292071 Objective Loss 0.292071 LR 0.001000 Time 0.020845 -2022-12-06 10:40:33,273 - Epoch: [41][ 410/ 1200] Overall Loss 0.292738 Objective Loss 0.292738 LR 0.001000 Time 0.020801 -2022-12-06 10:40:33,463 - Epoch: [41][ 420/ 1200] Overall Loss 0.292560 Objective Loss 0.292560 LR 0.001000 Time 0.020757 -2022-12-06 10:40:33,655 - Epoch: [41][ 430/ 1200] Overall Loss 0.291771 Objective Loss 0.291771 LR 0.001000 Time 0.020718 -2022-12-06 10:40:33,845 - Epoch: [41][ 440/ 1200] Overall Loss 0.291671 Objective Loss 0.291671 LR 0.001000 Time 0.020679 -2022-12-06 10:40:34,036 - Epoch: [41][ 450/ 1200] Overall Loss 0.291717 Objective Loss 0.291717 LR 0.001000 Time 0.020642 -2022-12-06 10:40:34,227 - Epoch: [41][ 460/ 1200] Overall Loss 0.291676 Objective Loss 0.291676 LR 0.001000 Time 0.020608 -2022-12-06 10:40:34,417 - Epoch: [41][ 470/ 1200] Overall Loss 0.291706 Objective Loss 0.291706 LR 0.001000 Time 0.020573 -2022-12-06 10:40:34,608 - Epoch: [41][ 480/ 1200] Overall Loss 0.291361 Objective Loss 0.291361 LR 0.001000 Time 0.020540 -2022-12-06 10:40:34,798 - Epoch: [41][ 490/ 1200] Overall Loss 0.291140 Objective Loss 0.291140 LR 0.001000 Time 0.020509 -2022-12-06 10:40:34,988 - Epoch: [41][ 500/ 1200] Overall Loss 0.290992 Objective Loss 0.290992 LR 0.001000 Time 0.020478 -2022-12-06 10:40:35,180 - Epoch: [41][ 510/ 1200] Overall Loss 0.290784 Objective Loss 0.290784 LR 0.001000 Time 0.020450 -2022-12-06 10:40:35,370 - Epoch: [41][ 520/ 1200] Overall Loss 0.290644 Objective Loss 0.290644 LR 0.001000 Time 0.020422 -2022-12-06 10:40:35,561 - Epoch: [41][ 530/ 1200] Overall Loss 0.290702 Objective Loss 0.290702 LR 0.001000 Time 0.020395 -2022-12-06 10:40:35,751 - Epoch: [41][ 540/ 1200] Overall Loss 0.290683 Objective Loss 0.290683 LR 0.001000 Time 0.020370 -2022-12-06 10:40:35,943 - Epoch: [41][ 550/ 1200] Overall Loss 0.290788 Objective Loss 0.290788 LR 0.001000 Time 0.020347 -2022-12-06 10:40:36,133 - Epoch: [41][ 560/ 1200] Overall Loss 0.290887 Objective Loss 0.290887 LR 0.001000 Time 0.020322 -2022-12-06 10:40:36,324 - Epoch: [41][ 570/ 1200] Overall Loss 0.290653 Objective Loss 0.290653 LR 0.001000 Time 0.020299 -2022-12-06 10:40:36,514 - Epoch: [41][ 580/ 1200] Overall Loss 0.290593 Objective Loss 0.290593 LR 0.001000 Time 0.020276 -2022-12-06 10:40:36,705 - Epoch: [41][ 590/ 1200] Overall Loss 0.290651 Objective Loss 0.290651 LR 0.001000 Time 0.020255 -2022-12-06 10:40:36,896 - Epoch: [41][ 600/ 1200] Overall Loss 0.291041 Objective Loss 0.291041 LR 0.001000 Time 0.020234 -2022-12-06 10:40:37,086 - Epoch: [41][ 610/ 1200] Overall Loss 0.291062 Objective Loss 0.291062 LR 0.001000 Time 0.020214 -2022-12-06 10:40:37,276 - Epoch: [41][ 620/ 1200] Overall Loss 0.290959 Objective Loss 0.290959 LR 0.001000 Time 0.020193 -2022-12-06 10:40:37,467 - Epoch: [41][ 630/ 1200] Overall Loss 0.291501 Objective Loss 0.291501 LR 0.001000 Time 0.020176 -2022-12-06 10:40:37,658 - Epoch: [41][ 640/ 1200] Overall Loss 0.291493 Objective Loss 0.291493 LR 0.001000 Time 0.020157 -2022-12-06 10:40:37,849 - Epoch: [41][ 650/ 1200] Overall Loss 0.291825 Objective Loss 0.291825 LR 0.001000 Time 0.020140 -2022-12-06 10:40:38,039 - Epoch: [41][ 660/ 1200] Overall Loss 0.292045 Objective Loss 0.292045 LR 0.001000 Time 0.020123 -2022-12-06 10:40:38,230 - Epoch: [41][ 670/ 1200] Overall Loss 0.292160 Objective Loss 0.292160 LR 0.001000 Time 0.020105 -2022-12-06 10:40:38,421 - Epoch: [41][ 680/ 1200] Overall Loss 0.292431 Objective Loss 0.292431 LR 0.001000 Time 0.020091 -2022-12-06 10:40:38,613 - Epoch: [41][ 690/ 1200] Overall Loss 0.292752 Objective Loss 0.292752 LR 0.001000 Time 0.020077 -2022-12-06 10:40:38,804 - Epoch: [41][ 700/ 1200] Overall Loss 0.292899 Objective Loss 0.292899 LR 0.001000 Time 0.020062 -2022-12-06 10:40:38,995 - Epoch: [41][ 710/ 1200] Overall Loss 0.293005 Objective Loss 0.293005 LR 0.001000 Time 0.020047 -2022-12-06 10:40:39,187 - Epoch: [41][ 720/ 1200] Overall Loss 0.293204 Objective Loss 0.293204 LR 0.001000 Time 0.020034 -2022-12-06 10:40:39,378 - Epoch: [41][ 730/ 1200] Overall Loss 0.293169 Objective Loss 0.293169 LR 0.001000 Time 0.020021 -2022-12-06 10:40:39,569 - Epoch: [41][ 740/ 1200] Overall Loss 0.293231 Objective Loss 0.293231 LR 0.001000 Time 0.020008 -2022-12-06 10:40:39,760 - Epoch: [41][ 750/ 1200] Overall Loss 0.293648 Objective Loss 0.293648 LR 0.001000 Time 0.019996 -2022-12-06 10:40:39,951 - Epoch: [41][ 760/ 1200] Overall Loss 0.293778 Objective Loss 0.293778 LR 0.001000 Time 0.019983 -2022-12-06 10:40:40,142 - Epoch: [41][ 770/ 1200] Overall Loss 0.293460 Objective Loss 0.293460 LR 0.001000 Time 0.019970 -2022-12-06 10:40:40,332 - Epoch: [41][ 780/ 1200] Overall Loss 0.293508 Objective Loss 0.293508 LR 0.001000 Time 0.019958 -2022-12-06 10:40:40,523 - Epoch: [41][ 790/ 1200] Overall Loss 0.293532 Objective Loss 0.293532 LR 0.001000 Time 0.019946 -2022-12-06 10:40:40,714 - Epoch: [41][ 800/ 1200] Overall Loss 0.293502 Objective Loss 0.293502 LR 0.001000 Time 0.019934 -2022-12-06 10:40:40,904 - Epoch: [41][ 810/ 1200] Overall Loss 0.293003 Objective Loss 0.293003 LR 0.001000 Time 0.019923 -2022-12-06 10:40:41,095 - Epoch: [41][ 820/ 1200] Overall Loss 0.293127 Objective Loss 0.293127 LR 0.001000 Time 0.019911 -2022-12-06 10:40:41,285 - Epoch: [41][ 830/ 1200] Overall Loss 0.293147 Objective Loss 0.293147 LR 0.001000 Time 0.019900 -2022-12-06 10:40:41,476 - Epoch: [41][ 840/ 1200] Overall Loss 0.293166 Objective Loss 0.293166 LR 0.001000 Time 0.019890 -2022-12-06 10:40:41,668 - Epoch: [41][ 850/ 1200] Overall Loss 0.293043 Objective Loss 0.293043 LR 0.001000 Time 0.019881 -2022-12-06 10:40:41,859 - Epoch: [41][ 860/ 1200] Overall Loss 0.293062 Objective Loss 0.293062 LR 0.001000 Time 0.019871 -2022-12-06 10:40:42,051 - Epoch: [41][ 870/ 1200] Overall Loss 0.292965 Objective Loss 0.292965 LR 0.001000 Time 0.019863 -2022-12-06 10:40:42,241 - Epoch: [41][ 880/ 1200] Overall Loss 0.293122 Objective Loss 0.293122 LR 0.001000 Time 0.019852 -2022-12-06 10:40:42,431 - Epoch: [41][ 890/ 1200] Overall Loss 0.293360 Objective Loss 0.293360 LR 0.001000 Time 0.019842 -2022-12-06 10:40:42,622 - Epoch: [41][ 900/ 1200] Overall Loss 0.293733 Objective Loss 0.293733 LR 0.001000 Time 0.019834 -2022-12-06 10:40:42,812 - Epoch: [41][ 910/ 1200] Overall Loss 0.293893 Objective Loss 0.293893 LR 0.001000 Time 0.019824 -2022-12-06 10:40:43,003 - Epoch: [41][ 920/ 1200] Overall Loss 0.294131 Objective Loss 0.294131 LR 0.001000 Time 0.019816 -2022-12-06 10:40:43,194 - Epoch: [41][ 930/ 1200] Overall Loss 0.294302 Objective Loss 0.294302 LR 0.001000 Time 0.019807 -2022-12-06 10:40:43,384 - Epoch: [41][ 940/ 1200] Overall Loss 0.294331 Objective Loss 0.294331 LR 0.001000 Time 0.019798 -2022-12-06 10:40:43,576 - Epoch: [41][ 950/ 1200] Overall Loss 0.294563 Objective Loss 0.294563 LR 0.001000 Time 0.019791 -2022-12-06 10:40:43,766 - Epoch: [41][ 960/ 1200] Overall Loss 0.294965 Objective Loss 0.294965 LR 0.001000 Time 0.019782 -2022-12-06 10:40:43,957 - Epoch: [41][ 970/ 1200] Overall Loss 0.294886 Objective Loss 0.294886 LR 0.001000 Time 0.019775 -2022-12-06 10:40:44,148 - Epoch: [41][ 980/ 1200] Overall Loss 0.295313 Objective Loss 0.295313 LR 0.001000 Time 0.019767 -2022-12-06 10:40:44,338 - Epoch: [41][ 990/ 1200] Overall Loss 0.295571 Objective Loss 0.295571 LR 0.001000 Time 0.019759 -2022-12-06 10:40:44,529 - Epoch: [41][ 1000/ 1200] Overall Loss 0.295838 Objective Loss 0.295838 LR 0.001000 Time 0.019752 -2022-12-06 10:40:44,720 - Epoch: [41][ 1010/ 1200] Overall Loss 0.296037 Objective Loss 0.296037 LR 0.001000 Time 0.019745 -2022-12-06 10:40:44,910 - Epoch: [41][ 1020/ 1200] Overall Loss 0.295945 Objective Loss 0.295945 LR 0.001000 Time 0.019737 -2022-12-06 10:40:45,102 - Epoch: [41][ 1030/ 1200] Overall Loss 0.296230 Objective Loss 0.296230 LR 0.001000 Time 0.019731 -2022-12-06 10:40:45,293 - Epoch: [41][ 1040/ 1200] Overall Loss 0.296301 Objective Loss 0.296301 LR 0.001000 Time 0.019724 -2022-12-06 10:40:45,483 - Epoch: [41][ 1050/ 1200] Overall Loss 0.296326 Objective Loss 0.296326 LR 0.001000 Time 0.019717 -2022-12-06 10:40:45,672 - Epoch: [41][ 1060/ 1200] Overall Loss 0.296177 Objective Loss 0.296177 LR 0.001000 Time 0.019709 -2022-12-06 10:40:45,864 - Epoch: [41][ 1070/ 1200] Overall Loss 0.296525 Objective Loss 0.296525 LR 0.001000 Time 0.019703 -2022-12-06 10:40:46,054 - Epoch: [41][ 1080/ 1200] Overall Loss 0.296693 Objective Loss 0.296693 LR 0.001000 Time 0.019697 -2022-12-06 10:40:46,245 - Epoch: [41][ 1090/ 1200] Overall Loss 0.296845 Objective Loss 0.296845 LR 0.001000 Time 0.019691 -2022-12-06 10:40:46,435 - Epoch: [41][ 1100/ 1200] Overall Loss 0.296643 Objective Loss 0.296643 LR 0.001000 Time 0.019684 -2022-12-06 10:40:46,626 - Epoch: [41][ 1110/ 1200] Overall Loss 0.296769 Objective Loss 0.296769 LR 0.001000 Time 0.019678 -2022-12-06 10:40:46,818 - Epoch: [41][ 1120/ 1200] Overall Loss 0.296836 Objective Loss 0.296836 LR 0.001000 Time 0.019673 -2022-12-06 10:40:47,009 - Epoch: [41][ 1130/ 1200] Overall Loss 0.296863 Objective Loss 0.296863 LR 0.001000 Time 0.019668 -2022-12-06 10:40:47,200 - Epoch: [41][ 1140/ 1200] Overall Loss 0.296802 Objective Loss 0.296802 LR 0.001000 Time 0.019663 -2022-12-06 10:40:47,392 - Epoch: [41][ 1150/ 1200] Overall Loss 0.297108 Objective Loss 0.297108 LR 0.001000 Time 0.019658 -2022-12-06 10:40:47,582 - Epoch: [41][ 1160/ 1200] Overall Loss 0.297256 Objective Loss 0.297256 LR 0.001000 Time 0.019652 -2022-12-06 10:40:47,773 - Epoch: [41][ 1170/ 1200] Overall Loss 0.297268 Objective Loss 0.297268 LR 0.001000 Time 0.019647 -2022-12-06 10:40:47,965 - Epoch: [41][ 1180/ 1200] Overall Loss 0.297184 Objective Loss 0.297184 LR 0.001000 Time 0.019642 -2022-12-06 10:40:48,156 - Epoch: [41][ 1190/ 1200] Overall Loss 0.296992 Objective Loss 0.296992 LR 0.001000 Time 0.019637 -2022-12-06 10:40:48,379 - Epoch: [41][ 1200/ 1200] Overall Loss 0.297203 Objective Loss 0.297203 Top1 84.518828 Top5 97.698745 LR 0.001000 Time 0.019659 -2022-12-06 10:40:48,468 - --- validate (epoch=41)----------- -2022-12-06 10:40:48,468 - 34129 samples (256 per mini-batch) -2022-12-06 10:40:48,912 - Epoch: [41][ 10/ 134] Loss 0.333548 Top1 83.242188 Top5 97.031250 -2022-12-06 10:40:49,045 - Epoch: [41][ 20/ 134] Loss 0.311897 Top1 83.535156 Top5 97.519531 -2022-12-06 10:40:49,178 - Epoch: [41][ 30/ 134] Loss 0.312481 Top1 83.723958 Top5 97.656250 -2022-12-06 10:40:49,311 - Epoch: [41][ 40/ 134] Loss 0.316465 Top1 83.701172 Top5 97.607422 -2022-12-06 10:40:49,441 - Epoch: [41][ 50/ 134] Loss 0.310072 Top1 83.843750 Top5 97.640625 -2022-12-06 10:40:49,569 - Epoch: [41][ 60/ 134] Loss 0.308050 Top1 83.958333 Top5 97.740885 -2022-12-06 10:40:49,697 - Epoch: [41][ 70/ 134] Loss 0.307449 Top1 83.939732 Top5 97.700893 -2022-12-06 10:40:49,835 - Epoch: [41][ 80/ 134] Loss 0.304541 Top1 84.018555 Top5 97.709961 -2022-12-06 10:40:49,965 - Epoch: [41][ 90/ 134] Loss 0.305732 Top1 84.105903 Top5 97.795139 -2022-12-06 10:40:50,098 - Epoch: [41][ 100/ 134] Loss 0.305089 Top1 84.121094 Top5 97.808594 -2022-12-06 10:40:50,226 - Epoch: [41][ 110/ 134] Loss 0.307920 Top1 84.137074 Top5 97.741477 -2022-12-06 10:40:50,355 - Epoch: [41][ 120/ 134] Loss 0.308798 Top1 84.127604 Top5 97.760417 -2022-12-06 10:40:50,482 - Epoch: [41][ 130/ 134] Loss 0.309108 Top1 84.158654 Top5 97.761418 -2022-12-06 10:40:50,518 - Epoch: [41][ 134/ 134] Loss 0.308575 Top1 84.160098 Top5 97.773155 -2022-12-06 10:40:50,619 - ==> Top1: 84.160 Top5: 97.773 Loss: 0.309 - -2022-12-06 10:40:50,620 - ==> Confusion: -[[ 877 2 3 2 9 5 1 0 7 60 0 2 2 1 8 3 3 5 2 0 4] - [ 2 934 1 2 10 22 7 12 0 1 4 5 0 2 2 1 5 2 6 0 9] - [ 2 3 977 15 3 1 50 10 0 1 5 5 2 1 3 5 2 1 4 4 9] - [ 2 3 24 929 0 5 2 0 0 0 9 1 4 2 12 0 3 6 11 0 7] - [ 9 4 3 1 938 10 0 1 1 7 1 5 1 2 8 8 12 3 0 1 5] - [ 3 21 3 1 10 980 3 14 1 0 0 12 0 6 3 3 2 1 0 4 2] - [ 1 1 5 0 0 3 1077 6 0 0 2 3 0 1 0 6 0 2 1 9 1] - [ 0 10 4 2 1 26 11 941 0 0 1 8 2 1 2 2 2 1 20 18 2] - [ 8 3 0 1 0 5 1 0 951 41 9 2 2 14 12 2 4 3 2 1 3] - [ 65 0 5 0 3 6 1 3 15 864 1 3 2 16 5 0 0 3 2 0 7] - [ 0 3 5 5 0 5 2 0 8 2 945 3 2 18 4 0 1 0 11 3 2] - [ 2 3 1 0 0 12 3 1 1 0 0 974 20 8 1 7 1 8 0 9 0] - [ 1 1 3 1 1 4 2 0 0 0 0 36 875 2 0 9 1 22 1 4 6] - [ 0 2 0 0 0 22 0 2 5 9 7 5 3 950 0 4 2 0 0 5 7] - [ 5 1 3 17 3 8 1 0 12 2 1 2 5 4 1034 0 4 3 12 1 12] - [ 1 1 1 0 1 3 5 0 0 0 1 9 4 1 0 995 9 7 1 2 2] - [ 0 6 2 2 1 2 2 0 0 0 1 4 3 1 0 13 1023 3 1 3 5] - [ 3 0 0 3 1 3 4 1 1 2 0 6 11 2 0 15 0 984 0 0 0] - [ 5 10 7 15 2 7 2 25 1 1 4 2 3 0 3 0 3 0 909 6 3] - [ 0 3 3 0 0 7 4 0 0 1 0 23 4 6 0 4 4 2 1 1011 7] - [ 124 281 196 114 146 250 117 153 60 69 193 167 373 363 141 167 222 99 162 273 9556]] - -2022-12-06 10:40:51,282 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:40:51,282 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:40:51,288 - - -2022-12-06 10:40:51,288 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:40:52,232 - Epoch: [42][ 10/ 1200] Overall Loss 0.307530 Objective Loss 0.307530 LR 0.001000 Time 0.094361 -2022-12-06 10:40:52,430 - Epoch: [42][ 20/ 1200] Overall Loss 0.305093 Objective Loss 0.305093 LR 0.001000 Time 0.057015 -2022-12-06 10:40:52,627 - Epoch: [42][ 30/ 1200] Overall Loss 0.307223 Objective Loss 0.307223 LR 0.001000 Time 0.044577 -2022-12-06 10:40:52,822 - Epoch: [42][ 40/ 1200] Overall Loss 0.310079 Objective Loss 0.310079 LR 0.001000 Time 0.038302 -2022-12-06 10:40:53,020 - Epoch: [42][ 50/ 1200] Overall Loss 0.305088 Objective Loss 0.305088 LR 0.001000 Time 0.034573 -2022-12-06 10:40:53,214 - Epoch: [42][ 60/ 1200] Overall Loss 0.300522 Objective Loss 0.300522 LR 0.001000 Time 0.032043 -2022-12-06 10:40:53,411 - Epoch: [42][ 70/ 1200] Overall Loss 0.301816 Objective Loss 0.301816 LR 0.001000 Time 0.030272 -2022-12-06 10:40:53,606 - Epoch: [42][ 80/ 1200] Overall Loss 0.300103 Objective Loss 0.300103 LR 0.001000 Time 0.028919 -2022-12-06 10:40:53,803 - Epoch: [42][ 90/ 1200] Overall Loss 0.295590 Objective Loss 0.295590 LR 0.001000 Time 0.027885 -2022-12-06 10:40:53,998 - Epoch: [42][ 100/ 1200] Overall Loss 0.294955 Objective Loss 0.294955 LR 0.001000 Time 0.027044 -2022-12-06 10:40:54,195 - Epoch: [42][ 110/ 1200] Overall Loss 0.295330 Objective Loss 0.295330 LR 0.001000 Time 0.026370 -2022-12-06 10:40:54,389 - Epoch: [42][ 120/ 1200] Overall Loss 0.292305 Objective Loss 0.292305 LR 0.001000 Time 0.025790 -2022-12-06 10:40:54,587 - Epoch: [42][ 130/ 1200] Overall Loss 0.291438 Objective Loss 0.291438 LR 0.001000 Time 0.025320 -2022-12-06 10:40:54,782 - Epoch: [42][ 140/ 1200] Overall Loss 0.289892 Objective Loss 0.289892 LR 0.001000 Time 0.024902 -2022-12-06 10:40:54,979 - Epoch: [42][ 150/ 1200] Overall Loss 0.289032 Objective Loss 0.289032 LR 0.001000 Time 0.024551 -2022-12-06 10:40:55,174 - Epoch: [42][ 160/ 1200] Overall Loss 0.290115 Objective Loss 0.290115 LR 0.001000 Time 0.024232 -2022-12-06 10:40:55,371 - Epoch: [42][ 170/ 1200] Overall Loss 0.290469 Objective Loss 0.290469 LR 0.001000 Time 0.023964 -2022-12-06 10:40:55,568 - Epoch: [42][ 180/ 1200] Overall Loss 0.290925 Objective Loss 0.290925 LR 0.001000 Time 0.023720 -2022-12-06 10:40:55,766 - Epoch: [42][ 190/ 1200] Overall Loss 0.291213 Objective Loss 0.291213 LR 0.001000 Time 0.023516 -2022-12-06 10:40:55,964 - Epoch: [42][ 200/ 1200] Overall Loss 0.290346 Objective Loss 0.290346 LR 0.001000 Time 0.023323 -2022-12-06 10:40:56,163 - Epoch: [42][ 210/ 1200] Overall Loss 0.290109 Objective Loss 0.290109 LR 0.001000 Time 0.023157 -2022-12-06 10:40:56,359 - Epoch: [42][ 220/ 1200] Overall Loss 0.290574 Objective Loss 0.290574 LR 0.001000 Time 0.022993 -2022-12-06 10:40:56,558 - Epoch: [42][ 230/ 1200] Overall Loss 0.290184 Objective Loss 0.290184 LR 0.001000 Time 0.022858 -2022-12-06 10:40:56,754 - Epoch: [42][ 240/ 1200] Overall Loss 0.289801 Objective Loss 0.289801 LR 0.001000 Time 0.022721 -2022-12-06 10:40:56,954 - Epoch: [42][ 250/ 1200] Overall Loss 0.290676 Objective Loss 0.290676 LR 0.001000 Time 0.022607 -2022-12-06 10:40:57,150 - Epoch: [42][ 260/ 1200] Overall Loss 0.291327 Objective Loss 0.291327 LR 0.001000 Time 0.022489 -2022-12-06 10:40:57,349 - Epoch: [42][ 270/ 1200] Overall Loss 0.290837 Objective Loss 0.290837 LR 0.001000 Time 0.022392 -2022-12-06 10:40:57,545 - Epoch: [42][ 280/ 1200] Overall Loss 0.291161 Objective Loss 0.291161 LR 0.001000 Time 0.022292 -2022-12-06 10:40:57,744 - Epoch: [42][ 290/ 1200] Overall Loss 0.291647 Objective Loss 0.291647 LR 0.001000 Time 0.022208 -2022-12-06 10:40:57,941 - Epoch: [42][ 300/ 1200] Overall Loss 0.291732 Objective Loss 0.291732 LR 0.001000 Time 0.022121 -2022-12-06 10:40:58,140 - Epoch: [42][ 310/ 1200] Overall Loss 0.291445 Objective Loss 0.291445 LR 0.001000 Time 0.022049 -2022-12-06 10:40:58,337 - Epoch: [42][ 320/ 1200] Overall Loss 0.290745 Objective Loss 0.290745 LR 0.001000 Time 0.021974 -2022-12-06 10:40:58,537 - Epoch: [42][ 330/ 1200] Overall Loss 0.290465 Objective Loss 0.290465 LR 0.001000 Time 0.021911 -2022-12-06 10:40:58,732 - Epoch: [42][ 340/ 1200] Overall Loss 0.290243 Objective Loss 0.290243 LR 0.001000 Time 0.021840 -2022-12-06 10:40:58,932 - Epoch: [42][ 350/ 1200] Overall Loss 0.290122 Objective Loss 0.290122 LR 0.001000 Time 0.021784 -2022-12-06 10:40:59,128 - Epoch: [42][ 360/ 1200] Overall Loss 0.290208 Objective Loss 0.290208 LR 0.001000 Time 0.021724 -2022-12-06 10:40:59,328 - Epoch: [42][ 370/ 1200] Overall Loss 0.289260 Objective Loss 0.289260 LR 0.001000 Time 0.021675 -2022-12-06 10:40:59,524 - Epoch: [42][ 380/ 1200] Overall Loss 0.290024 Objective Loss 0.290024 LR 0.001000 Time 0.021619 -2022-12-06 10:40:59,723 - Epoch: [42][ 390/ 1200] Overall Loss 0.290341 Objective Loss 0.290341 LR 0.001000 Time 0.021574 -2022-12-06 10:40:59,920 - Epoch: [42][ 400/ 1200] Overall Loss 0.290108 Objective Loss 0.290108 LR 0.001000 Time 0.021525 -2022-12-06 10:41:00,119 - Epoch: [42][ 410/ 1200] Overall Loss 0.289605 Objective Loss 0.289605 LR 0.001000 Time 0.021485 -2022-12-06 10:41:00,316 - Epoch: [42][ 420/ 1200] Overall Loss 0.289677 Objective Loss 0.289677 LR 0.001000 Time 0.021441 -2022-12-06 10:41:00,516 - Epoch: [42][ 430/ 1200] Overall Loss 0.289812 Objective Loss 0.289812 LR 0.001000 Time 0.021406 -2022-12-06 10:41:00,713 - Epoch: [42][ 440/ 1200] Overall Loss 0.289258 Objective Loss 0.289258 LR 0.001000 Time 0.021365 -2022-12-06 10:41:00,912 - Epoch: [42][ 450/ 1200] Overall Loss 0.289419 Objective Loss 0.289419 LR 0.001000 Time 0.021332 -2022-12-06 10:41:01,110 - Epoch: [42][ 460/ 1200] Overall Loss 0.289240 Objective Loss 0.289240 LR 0.001000 Time 0.021297 -2022-12-06 10:41:01,309 - Epoch: [42][ 470/ 1200] Overall Loss 0.289124 Objective Loss 0.289124 LR 0.001000 Time 0.021267 -2022-12-06 10:41:01,507 - Epoch: [42][ 480/ 1200] Overall Loss 0.288990 Objective Loss 0.288990 LR 0.001000 Time 0.021236 -2022-12-06 10:41:01,707 - Epoch: [42][ 490/ 1200] Overall Loss 0.289544 Objective Loss 0.289544 LR 0.001000 Time 0.021209 -2022-12-06 10:41:01,903 - Epoch: [42][ 500/ 1200] Overall Loss 0.289177 Objective Loss 0.289177 LR 0.001000 Time 0.021176 -2022-12-06 10:41:02,103 - Epoch: [42][ 510/ 1200] Overall Loss 0.289216 Objective Loss 0.289216 LR 0.001000 Time 0.021151 -2022-12-06 10:41:02,299 - Epoch: [42][ 520/ 1200] Overall Loss 0.288924 Objective Loss 0.288924 LR 0.001000 Time 0.021121 -2022-12-06 10:41:02,498 - Epoch: [42][ 530/ 1200] Overall Loss 0.289082 Objective Loss 0.289082 LR 0.001000 Time 0.021096 -2022-12-06 10:41:02,695 - Epoch: [42][ 540/ 1200] Overall Loss 0.289209 Objective Loss 0.289209 LR 0.001000 Time 0.021069 -2022-12-06 10:41:02,895 - Epoch: [42][ 550/ 1200] Overall Loss 0.289173 Objective Loss 0.289173 LR 0.001000 Time 0.021048 -2022-12-06 10:41:03,091 - Epoch: [42][ 560/ 1200] Overall Loss 0.289191 Objective Loss 0.289191 LR 0.001000 Time 0.021022 -2022-12-06 10:41:03,290 - Epoch: [42][ 570/ 1200] Overall Loss 0.289429 Objective Loss 0.289429 LR 0.001000 Time 0.021002 -2022-12-06 10:41:03,487 - Epoch: [42][ 580/ 1200] Overall Loss 0.288896 Objective Loss 0.288896 LR 0.001000 Time 0.020978 -2022-12-06 10:41:03,686 - Epoch: [42][ 590/ 1200] Overall Loss 0.288870 Objective Loss 0.288870 LR 0.001000 Time 0.020959 -2022-12-06 10:41:03,883 - Epoch: [42][ 600/ 1200] Overall Loss 0.289059 Objective Loss 0.289059 LR 0.001000 Time 0.020937 -2022-12-06 10:41:04,082 - Epoch: [42][ 610/ 1200] Overall Loss 0.288803 Objective Loss 0.288803 LR 0.001000 Time 0.020920 -2022-12-06 10:41:04,279 - Epoch: [42][ 620/ 1200] Overall Loss 0.288729 Objective Loss 0.288729 LR 0.001000 Time 0.020899 -2022-12-06 10:41:04,479 - Epoch: [42][ 630/ 1200] Overall Loss 0.289108 Objective Loss 0.289108 LR 0.001000 Time 0.020884 -2022-12-06 10:41:04,676 - Epoch: [42][ 640/ 1200] Overall Loss 0.290133 Objective Loss 0.290133 LR 0.001000 Time 0.020864 -2022-12-06 10:41:04,875 - Epoch: [42][ 650/ 1200] Overall Loss 0.289624 Objective Loss 0.289624 LR 0.001000 Time 0.020849 -2022-12-06 10:41:05,072 - Epoch: [42][ 660/ 1200] Overall Loss 0.289348 Objective Loss 0.289348 LR 0.001000 Time 0.020831 -2022-12-06 10:41:05,272 - Epoch: [42][ 670/ 1200] Overall Loss 0.289296 Objective Loss 0.289296 LR 0.001000 Time 0.020817 -2022-12-06 10:41:05,469 - Epoch: [42][ 680/ 1200] Overall Loss 0.289098 Objective Loss 0.289098 LR 0.001000 Time 0.020800 -2022-12-06 10:41:05,668 - Epoch: [42][ 690/ 1200] Overall Loss 0.288583 Objective Loss 0.288583 LR 0.001000 Time 0.020786 -2022-12-06 10:41:05,865 - Epoch: [42][ 700/ 1200] Overall Loss 0.287779 Objective Loss 0.287779 LR 0.001000 Time 0.020769 -2022-12-06 10:41:06,064 - Epoch: [42][ 710/ 1200] Overall Loss 0.287849 Objective Loss 0.287849 LR 0.001000 Time 0.020757 -2022-12-06 10:41:06,260 - Epoch: [42][ 720/ 1200] Overall Loss 0.287548 Objective Loss 0.287548 LR 0.001000 Time 0.020740 -2022-12-06 10:41:06,460 - Epoch: [42][ 730/ 1200] Overall Loss 0.287640 Objective Loss 0.287640 LR 0.001000 Time 0.020729 -2022-12-06 10:41:06,656 - Epoch: [42][ 740/ 1200] Overall Loss 0.287651 Objective Loss 0.287651 LR 0.001000 Time 0.020714 -2022-12-06 10:41:06,856 - Epoch: [42][ 750/ 1200] Overall Loss 0.287445 Objective Loss 0.287445 LR 0.001000 Time 0.020703 -2022-12-06 10:41:07,053 - Epoch: [42][ 760/ 1200] Overall Loss 0.287878 Objective Loss 0.287878 LR 0.001000 Time 0.020688 -2022-12-06 10:41:07,253 - Epoch: [42][ 770/ 1200] Overall Loss 0.288197 Objective Loss 0.288197 LR 0.001000 Time 0.020679 -2022-12-06 10:41:07,449 - Epoch: [42][ 780/ 1200] Overall Loss 0.288382 Objective Loss 0.288382 LR 0.001000 Time 0.020665 -2022-12-06 10:41:07,649 - Epoch: [42][ 790/ 1200] Overall Loss 0.288305 Objective Loss 0.288305 LR 0.001000 Time 0.020655 -2022-12-06 10:41:07,845 - Epoch: [42][ 800/ 1200] Overall Loss 0.288290 Objective Loss 0.288290 LR 0.001000 Time 0.020642 -2022-12-06 10:41:08,045 - Epoch: [42][ 810/ 1200] Overall Loss 0.288311 Objective Loss 0.288311 LR 0.001000 Time 0.020633 -2022-12-06 10:41:08,241 - Epoch: [42][ 820/ 1200] Overall Loss 0.288483 Objective Loss 0.288483 LR 0.001000 Time 0.020619 -2022-12-06 10:41:08,440 - Epoch: [42][ 830/ 1200] Overall Loss 0.288405 Objective Loss 0.288405 LR 0.001000 Time 0.020611 -2022-12-06 10:41:08,637 - Epoch: [42][ 840/ 1200] Overall Loss 0.288325 Objective Loss 0.288325 LR 0.001000 Time 0.020599 -2022-12-06 10:41:08,837 - Epoch: [42][ 850/ 1200] Overall Loss 0.288205 Objective Loss 0.288205 LR 0.001000 Time 0.020591 -2022-12-06 10:41:09,034 - Epoch: [42][ 860/ 1200] Overall Loss 0.288624 Objective Loss 0.288624 LR 0.001000 Time 0.020580 -2022-12-06 10:41:09,233 - Epoch: [42][ 870/ 1200] Overall Loss 0.289057 Objective Loss 0.289057 LR 0.001000 Time 0.020572 -2022-12-06 10:41:09,430 - Epoch: [42][ 880/ 1200] Overall Loss 0.288946 Objective Loss 0.288946 LR 0.001000 Time 0.020561 -2022-12-06 10:41:09,630 - Epoch: [42][ 890/ 1200] Overall Loss 0.288822 Objective Loss 0.288822 LR 0.001000 Time 0.020554 -2022-12-06 10:41:09,826 - Epoch: [42][ 900/ 1200] Overall Loss 0.288851 Objective Loss 0.288851 LR 0.001000 Time 0.020544 -2022-12-06 10:41:10,026 - Epoch: [42][ 910/ 1200] Overall Loss 0.288760 Objective Loss 0.288760 LR 0.001000 Time 0.020536 -2022-12-06 10:41:10,223 - Epoch: [42][ 920/ 1200] Overall Loss 0.288957 Objective Loss 0.288957 LR 0.001000 Time 0.020527 -2022-12-06 10:41:10,422 - Epoch: [42][ 930/ 1200] Overall Loss 0.289042 Objective Loss 0.289042 LR 0.001000 Time 0.020520 -2022-12-06 10:41:10,620 - Epoch: [42][ 940/ 1200] Overall Loss 0.288822 Objective Loss 0.288822 LR 0.001000 Time 0.020511 -2022-12-06 10:41:10,820 - Epoch: [42][ 950/ 1200] Overall Loss 0.288876 Objective Loss 0.288876 LR 0.001000 Time 0.020506 -2022-12-06 10:41:11,017 - Epoch: [42][ 960/ 1200] Overall Loss 0.288660 Objective Loss 0.288660 LR 0.001000 Time 0.020496 -2022-12-06 10:41:11,216 - Epoch: [42][ 970/ 1200] Overall Loss 0.289040 Objective Loss 0.289040 LR 0.001000 Time 0.020490 -2022-12-06 10:41:11,413 - Epoch: [42][ 980/ 1200] Overall Loss 0.288948 Objective Loss 0.288948 LR 0.001000 Time 0.020481 -2022-12-06 10:41:11,612 - Epoch: [42][ 990/ 1200] Overall Loss 0.289343 Objective Loss 0.289343 LR 0.001000 Time 0.020475 -2022-12-06 10:41:11,810 - Epoch: [42][ 1000/ 1200] Overall Loss 0.289552 Objective Loss 0.289552 LR 0.001000 Time 0.020468 -2022-12-06 10:41:12,010 - Epoch: [42][ 1010/ 1200] Overall Loss 0.289663 Objective Loss 0.289663 LR 0.001000 Time 0.020462 -2022-12-06 10:41:12,207 - Epoch: [42][ 1020/ 1200] Overall Loss 0.289465 Objective Loss 0.289465 LR 0.001000 Time 0.020454 -2022-12-06 10:41:12,407 - Epoch: [42][ 1030/ 1200] Overall Loss 0.289574 Objective Loss 0.289574 LR 0.001000 Time 0.020449 -2022-12-06 10:41:12,604 - Epoch: [42][ 1040/ 1200] Overall Loss 0.289928 Objective Loss 0.289928 LR 0.001000 Time 0.020441 -2022-12-06 10:41:12,803 - Epoch: [42][ 1050/ 1200] Overall Loss 0.289846 Objective Loss 0.289846 LR 0.001000 Time 0.020436 -2022-12-06 10:41:13,000 - Epoch: [42][ 1060/ 1200] Overall Loss 0.289913 Objective Loss 0.289913 LR 0.001000 Time 0.020428 -2022-12-06 10:41:13,200 - Epoch: [42][ 1070/ 1200] Overall Loss 0.289819 Objective Loss 0.289819 LR 0.001000 Time 0.020424 -2022-12-06 10:41:13,396 - Epoch: [42][ 1080/ 1200] Overall Loss 0.289915 Objective Loss 0.289915 LR 0.001000 Time 0.020416 -2022-12-06 10:41:13,596 - Epoch: [42][ 1090/ 1200] Overall Loss 0.289673 Objective Loss 0.289673 LR 0.001000 Time 0.020411 -2022-12-06 10:41:13,793 - Epoch: [42][ 1100/ 1200] Overall Loss 0.289504 Objective Loss 0.289504 LR 0.001000 Time 0.020404 -2022-12-06 10:41:13,993 - Epoch: [42][ 1110/ 1200] Overall Loss 0.289634 Objective Loss 0.289634 LR 0.001000 Time 0.020400 -2022-12-06 10:41:14,190 - Epoch: [42][ 1120/ 1200] Overall Loss 0.289510 Objective Loss 0.289510 LR 0.001000 Time 0.020394 -2022-12-06 10:41:14,389 - Epoch: [42][ 1130/ 1200] Overall Loss 0.289692 Objective Loss 0.289692 LR 0.001000 Time 0.020389 -2022-12-06 10:41:14,585 - Epoch: [42][ 1140/ 1200] Overall Loss 0.289587 Objective Loss 0.289587 LR 0.001000 Time 0.020382 -2022-12-06 10:41:14,784 - Epoch: [42][ 1150/ 1200] Overall Loss 0.289715 Objective Loss 0.289715 LR 0.001000 Time 0.020377 -2022-12-06 10:41:14,981 - Epoch: [42][ 1160/ 1200] Overall Loss 0.289839 Objective Loss 0.289839 LR 0.001000 Time 0.020370 -2022-12-06 10:41:15,180 - Epoch: [42][ 1170/ 1200] Overall Loss 0.289361 Objective Loss 0.289361 LR 0.001000 Time 0.020366 -2022-12-06 10:41:15,377 - Epoch: [42][ 1180/ 1200] Overall Loss 0.289262 Objective Loss 0.289262 LR 0.001000 Time 0.020360 -2022-12-06 10:41:15,576 - Epoch: [42][ 1190/ 1200] Overall Loss 0.289594 Objective Loss 0.289594 LR 0.001000 Time 0.020356 -2022-12-06 10:41:15,803 - Epoch: [42][ 1200/ 1200] Overall Loss 0.289465 Objective Loss 0.289465 Top1 87.447699 Top5 98.744770 LR 0.001000 Time 0.020375 -2022-12-06 10:41:15,892 - --- validate (epoch=42)----------- -2022-12-06 10:41:15,892 - 34129 samples (256 per mini-batch) -2022-12-06 10:41:16,336 - Epoch: [42][ 10/ 134] Loss 0.310476 Top1 84.570312 Top5 97.382812 -2022-12-06 10:41:16,472 - Epoch: [42][ 20/ 134] Loss 0.288228 Top1 85.058594 Top5 97.871094 -2022-12-06 10:41:16,606 - Epoch: [42][ 30/ 134] Loss 0.296647 Top1 84.843750 Top5 97.994792 -2022-12-06 10:41:16,748 - Epoch: [42][ 40/ 134] Loss 0.299064 Top1 84.697266 Top5 97.939453 -2022-12-06 10:41:16,876 - Epoch: [42][ 50/ 134] Loss 0.296736 Top1 84.765625 Top5 97.867188 -2022-12-06 10:41:17,007 - Epoch: [42][ 60/ 134] Loss 0.297199 Top1 84.687500 Top5 97.858073 -2022-12-06 10:41:17,141 - Epoch: [42][ 70/ 134] Loss 0.297375 Top1 84.709821 Top5 97.851562 -2022-12-06 10:41:17,271 - Epoch: [42][ 80/ 134] Loss 0.300568 Top1 84.599609 Top5 97.836914 -2022-12-06 10:41:17,398 - Epoch: [42][ 90/ 134] Loss 0.303304 Top1 84.448785 Top5 97.756076 -2022-12-06 10:41:17,524 - Epoch: [42][ 100/ 134] Loss 0.302705 Top1 84.480469 Top5 97.812500 -2022-12-06 10:41:17,649 - Epoch: [42][ 110/ 134] Loss 0.303388 Top1 84.438920 Top5 97.791193 -2022-12-06 10:41:17,783 - Epoch: [42][ 120/ 134] Loss 0.302343 Top1 84.430339 Top5 97.828776 -2022-12-06 10:41:17,914 - Epoch: [42][ 130/ 134] Loss 0.301830 Top1 84.411058 Top5 97.836538 -2022-12-06 10:41:17,952 - Epoch: [42][ 134/ 134] Loss 0.301958 Top1 84.397433 Top5 97.840546 -2022-12-06 10:41:18,042 - ==> Top1: 84.397 Top5: 97.841 Loss: 0.302 - -2022-12-06 10:41:18,043 - ==> Confusion: -[[ 871 2 2 2 7 9 0 1 5 75 0 3 3 0 11 1 0 0 0 0 4] - [ 0 922 1 1 11 34 4 12 1 1 3 3 0 2 4 2 3 1 12 3 7] - [ 5 3 987 21 2 3 39 8 0 2 3 4 0 1 3 6 3 0 3 3 7] - [ 2 1 25 921 0 4 0 0 0 1 14 0 3 5 21 2 1 5 8 0 7] - [ 5 3 1 1 952 6 2 1 0 8 1 2 1 2 11 7 9 1 1 0 6] - [ 2 9 1 1 10 985 3 13 3 2 0 7 2 16 3 2 3 0 0 4 3] - [ 1 3 4 2 0 3 1080 2 0 0 2 2 0 3 0 9 0 0 0 6 1] - [ 0 12 11 2 2 30 17 917 0 0 3 5 0 1 2 1 0 3 23 17 8] - [ 7 3 0 0 0 3 2 0 954 42 8 3 1 10 18 1 2 1 4 2 3] - [ 51 0 3 0 1 4 1 2 22 882 2 1 1 13 5 1 1 1 0 2 8] - [ 1 2 5 6 0 2 2 2 12 2 949 4 2 15 3 0 1 0 4 2 5] - [ 3 1 1 0 1 13 7 2 1 0 2 959 16 11 1 9 2 6 0 13 3] - [ 2 0 2 1 1 2 3 0 0 0 0 39 873 3 0 10 2 18 1 7 5] - [ 0 0 0 0 2 6 0 2 8 13 4 7 2 963 0 2 1 0 0 4 9] - [ 5 2 2 17 4 2 0 1 25 5 1 2 1 5 1041 1 1 3 2 2 8] - [ 2 0 3 1 4 3 3 0 0 0 0 9 5 4 0 983 7 10 1 4 4] - [ 2 5 3 0 3 1 0 1 0 0 0 5 2 4 1 9 1015 3 0 7 11] - [ 3 1 0 3 1 3 0 0 2 2 0 3 20 3 2 29 0 960 1 2 1] - [ 6 4 8 21 0 5 2 26 2 0 9 1 3 2 11 0 2 2 895 4 5] - [ 4 2 2 0 1 9 10 5 0 0 1 9 4 8 0 3 2 2 1 1011 6] - [ 147 246 196 130 126 231 125 148 70 85 197 132 325 376 139 197 176 76 117 306 9681]] - -2022-12-06 10:41:18,625 - ==> Best [Top1: 84.702 Top5: 98.013 Sparsity:0.00 Params: 5376 on epoch: 30] -2022-12-06 10:41:18,625 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:41:18,632 - - -2022-12-06 10:41:18,632 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:41:19,673 - Epoch: [43][ 10/ 1200] Overall Loss 0.295406 Objective Loss 0.295406 LR 0.001000 Time 0.104056 -2022-12-06 10:41:19,873 - Epoch: [43][ 20/ 1200] Overall Loss 0.296031 Objective Loss 0.296031 LR 0.001000 Time 0.062010 -2022-12-06 10:41:20,071 - Epoch: [43][ 30/ 1200] Overall Loss 0.296857 Objective Loss 0.296857 LR 0.001000 Time 0.047909 -2022-12-06 10:41:20,265 - Epoch: [43][ 40/ 1200] Overall Loss 0.287923 Objective Loss 0.287923 LR 0.001000 Time 0.040768 -2022-12-06 10:41:20,462 - Epoch: [43][ 50/ 1200] Overall Loss 0.286475 Objective Loss 0.286475 LR 0.001000 Time 0.036547 -2022-12-06 10:41:20,655 - Epoch: [43][ 60/ 1200] Overall Loss 0.289702 Objective Loss 0.289702 LR 0.001000 Time 0.033665 -2022-12-06 10:41:20,853 - Epoch: [43][ 70/ 1200] Overall Loss 0.285283 Objective Loss 0.285283 LR 0.001000 Time 0.031670 -2022-12-06 10:41:21,047 - Epoch: [43][ 80/ 1200] Overall Loss 0.284982 Objective Loss 0.284982 LR 0.001000 Time 0.030132 -2022-12-06 10:41:21,244 - Epoch: [43][ 90/ 1200] Overall Loss 0.283234 Objective Loss 0.283234 LR 0.001000 Time 0.028966 -2022-12-06 10:41:21,438 - Epoch: [43][ 100/ 1200] Overall Loss 0.280238 Objective Loss 0.280238 LR 0.001000 Time 0.028008 -2022-12-06 10:41:21,635 - Epoch: [43][ 110/ 1200] Overall Loss 0.279723 Objective Loss 0.279723 LR 0.001000 Time 0.027251 -2022-12-06 10:41:21,830 - Epoch: [43][ 120/ 1200] Overall Loss 0.279643 Objective Loss 0.279643 LR 0.001000 Time 0.026595 -2022-12-06 10:41:22,028 - Epoch: [43][ 130/ 1200] Overall Loss 0.278023 Objective Loss 0.278023 LR 0.001000 Time 0.026072 -2022-12-06 10:41:22,223 - Epoch: [43][ 140/ 1200] Overall Loss 0.279576 Objective Loss 0.279576 LR 0.001000 Time 0.025593 -2022-12-06 10:41:22,419 - Epoch: [43][ 150/ 1200] Overall Loss 0.280220 Objective Loss 0.280220 LR 0.001000 Time 0.025194 -2022-12-06 10:41:22,613 - Epoch: [43][ 160/ 1200] Overall Loss 0.282402 Objective Loss 0.282402 LR 0.001000 Time 0.024828 -2022-12-06 10:41:22,810 - Epoch: [43][ 170/ 1200] Overall Loss 0.282341 Objective Loss 0.282341 LR 0.001000 Time 0.024524 -2022-12-06 10:41:23,005 - Epoch: [43][ 180/ 1200] Overall Loss 0.282916 Objective Loss 0.282916 LR 0.001000 Time 0.024239 -2022-12-06 10:41:23,202 - Epoch: [43][ 190/ 1200] Overall Loss 0.284002 Objective Loss 0.284002 LR 0.001000 Time 0.023999 -2022-12-06 10:41:23,396 - Epoch: [43][ 200/ 1200] Overall Loss 0.284433 Objective Loss 0.284433 LR 0.001000 Time 0.023766 -2022-12-06 10:41:23,594 - Epoch: [43][ 210/ 1200] Overall Loss 0.284119 Objective Loss 0.284119 LR 0.001000 Time 0.023573 -2022-12-06 10:41:23,787 - Epoch: [43][ 220/ 1200] Overall Loss 0.283780 Objective Loss 0.283780 LR 0.001000 Time 0.023378 -2022-12-06 10:41:23,985 - Epoch: [43][ 230/ 1200] Overall Loss 0.283774 Objective Loss 0.283774 LR 0.001000 Time 0.023219 -2022-12-06 10:41:24,179 - Epoch: [43][ 240/ 1200] Overall Loss 0.283455 Objective Loss 0.283455 LR 0.001000 Time 0.023058 -2022-12-06 10:41:24,376 - Epoch: [43][ 250/ 1200] Overall Loss 0.283830 Objective Loss 0.283830 LR 0.001000 Time 0.022920 -2022-12-06 10:41:24,570 - Epoch: [43][ 260/ 1200] Overall Loss 0.284163 Objective Loss 0.284163 LR 0.001000 Time 0.022784 -2022-12-06 10:41:24,767 - Epoch: [43][ 270/ 1200] Overall Loss 0.283847 Objective Loss 0.283847 LR 0.001000 Time 0.022669 -2022-12-06 10:41:24,961 - Epoch: [43][ 280/ 1200] Overall Loss 0.284560 Objective Loss 0.284560 LR 0.001000 Time 0.022550 -2022-12-06 10:41:25,159 - Epoch: [43][ 290/ 1200] Overall Loss 0.284407 Objective Loss 0.284407 LR 0.001000 Time 0.022451 -2022-12-06 10:41:25,353 - Epoch: [43][ 300/ 1200] Overall Loss 0.284065 Objective Loss 0.284065 LR 0.001000 Time 0.022349 -2022-12-06 10:41:25,550 - Epoch: [43][ 310/ 1200] Overall Loss 0.284610 Objective Loss 0.284610 LR 0.001000 Time 0.022261 -2022-12-06 10:41:25,744 - Epoch: [43][ 320/ 1200] Overall Loss 0.284526 Objective Loss 0.284526 LR 0.001000 Time 0.022169 -2022-12-06 10:41:25,941 - Epoch: [43][ 330/ 1200] Overall Loss 0.284017 Objective Loss 0.284017 LR 0.001000 Time 0.022093 -2022-12-06 10:41:26,135 - Epoch: [43][ 340/ 1200] Overall Loss 0.283476 Objective Loss 0.283476 LR 0.001000 Time 0.022012 -2022-12-06 10:41:26,332 - Epoch: [43][ 350/ 1200] Overall Loss 0.283408 Objective Loss 0.283408 LR 0.001000 Time 0.021945 -2022-12-06 10:41:26,525 - Epoch: [43][ 360/ 1200] Overall Loss 0.284614 Objective Loss 0.284614 LR 0.001000 Time 0.021871 -2022-12-06 10:41:26,722 - Epoch: [43][ 370/ 1200] Overall Loss 0.284436 Objective Loss 0.284436 LR 0.001000 Time 0.021811 -2022-12-06 10:41:26,916 - Epoch: [43][ 380/ 1200] Overall Loss 0.285052 Objective Loss 0.285052 LR 0.001000 Time 0.021746 -2022-12-06 10:41:27,113 - Epoch: [43][ 390/ 1200] Overall Loss 0.285560 Objective Loss 0.285560 LR 0.001000 Time 0.021693 -2022-12-06 10:41:27,307 - Epoch: [43][ 400/ 1200] Overall Loss 0.286145 Objective Loss 0.286145 LR 0.001000 Time 0.021633 -2022-12-06 10:41:27,504 - Epoch: [43][ 410/ 1200] Overall Loss 0.286509 Objective Loss 0.286509 LR 0.001000 Time 0.021584 -2022-12-06 10:41:27,698 - Epoch: [43][ 420/ 1200] Overall Loss 0.285424 Objective Loss 0.285424 LR 0.001000 Time 0.021531 -2022-12-06 10:41:27,896 - Epoch: [43][ 430/ 1200] Overall Loss 0.286116 Objective Loss 0.286116 LR 0.001000 Time 0.021488 -2022-12-06 10:41:28,090 - Epoch: [43][ 440/ 1200] Overall Loss 0.286052 Objective Loss 0.286052 LR 0.001000 Time 0.021441 -2022-12-06 10:41:28,287 - Epoch: [43][ 450/ 1200] Overall Loss 0.286883 Objective Loss 0.286883 LR 0.001000 Time 0.021402 -2022-12-06 10:41:28,481 - Epoch: [43][ 460/ 1200] Overall Loss 0.286723 Objective Loss 0.286723 LR 0.001000 Time 0.021355 -2022-12-06 10:41:28,678 - Epoch: [43][ 470/ 1200] Overall Loss 0.286768 Objective Loss 0.286768 LR 0.001000 Time 0.021319 -2022-12-06 10:41:28,872 - Epoch: [43][ 480/ 1200] Overall Loss 0.286782 Objective Loss 0.286782 LR 0.001000 Time 0.021278 -2022-12-06 10:41:29,068 - Epoch: [43][ 490/ 1200] Overall Loss 0.286494 Objective Loss 0.286494 LR 0.001000 Time 0.021244 -2022-12-06 10:41:29,263 - Epoch: [43][ 500/ 1200] Overall Loss 0.286192 Objective Loss 0.286192 LR 0.001000 Time 0.021207 -2022-12-06 10:41:29,460 - Epoch: [43][ 510/ 1200] Overall Loss 0.287084 Objective Loss 0.287084 LR 0.001000 Time 0.021177 -2022-12-06 10:41:29,655 - Epoch: [43][ 520/ 1200] Overall Loss 0.287138 Objective Loss 0.287138 LR 0.001000 Time 0.021144 -2022-12-06 10:41:29,852 - Epoch: [43][ 530/ 1200] Overall Loss 0.287251 Objective Loss 0.287251 LR 0.001000 Time 0.021115 -2022-12-06 10:41:30,046 - Epoch: [43][ 540/ 1200] Overall Loss 0.287201 Objective Loss 0.287201 LR 0.001000 Time 0.021083 -2022-12-06 10:41:30,244 - Epoch: [43][ 550/ 1200] Overall Loss 0.287571 Objective Loss 0.287571 LR 0.001000 Time 0.021057 -2022-12-06 10:41:30,437 - Epoch: [43][ 560/ 1200] Overall Loss 0.288032 Objective Loss 0.288032 LR 0.001000 Time 0.021026 -2022-12-06 10:41:30,634 - Epoch: [43][ 570/ 1200] Overall Loss 0.287943 Objective Loss 0.287943 LR 0.001000 Time 0.021002 -2022-12-06 10:41:30,829 - Epoch: [43][ 580/ 1200] Overall Loss 0.288233 Objective Loss 0.288233 LR 0.001000 Time 0.020974 -2022-12-06 10:41:31,026 - Epoch: [43][ 590/ 1200] Overall Loss 0.288334 Objective Loss 0.288334 LR 0.001000 Time 0.020952 -2022-12-06 10:41:31,219 - Epoch: [43][ 600/ 1200] Overall Loss 0.288663 Objective Loss 0.288663 LR 0.001000 Time 0.020924 -2022-12-06 10:41:31,417 - Epoch: [43][ 610/ 1200] Overall Loss 0.288593 Objective Loss 0.288593 LR 0.001000 Time 0.020904 -2022-12-06 10:41:31,611 - Epoch: [43][ 620/ 1200] Overall Loss 0.288956 Objective Loss 0.288956 LR 0.001000 Time 0.020879 -2022-12-06 10:41:31,808 - Epoch: [43][ 630/ 1200] Overall Loss 0.289017 Objective Loss 0.289017 LR 0.001000 Time 0.020860 -2022-12-06 10:41:32,002 - Epoch: [43][ 640/ 1200] Overall Loss 0.289123 Objective Loss 0.289123 LR 0.001000 Time 0.020836 -2022-12-06 10:41:32,200 - Epoch: [43][ 650/ 1200] Overall Loss 0.288777 Objective Loss 0.288777 LR 0.001000 Time 0.020819 -2022-12-06 10:41:32,394 - Epoch: [43][ 660/ 1200] Overall Loss 0.288369 Objective Loss 0.288369 LR 0.001000 Time 0.020797 -2022-12-06 10:41:32,584 - Epoch: [43][ 670/ 1200] Overall Loss 0.288526 Objective Loss 0.288526 LR 0.001000 Time 0.020770 -2022-12-06 10:41:32,774 - Epoch: [43][ 680/ 1200] Overall Loss 0.288112 Objective Loss 0.288112 LR 0.001000 Time 0.020743 -2022-12-06 10:41:32,965 - Epoch: [43][ 690/ 1200] Overall Loss 0.288125 Objective Loss 0.288125 LR 0.001000 Time 0.020717 -2022-12-06 10:41:33,154 - Epoch: [43][ 700/ 1200] Overall Loss 0.288064 Objective Loss 0.288064 LR 0.001000 Time 0.020691 -2022-12-06 10:41:33,344 - Epoch: [43][ 710/ 1200] Overall Loss 0.287858 Objective Loss 0.287858 LR 0.001000 Time 0.020667 -2022-12-06 10:41:33,534 - Epoch: [43][ 720/ 1200] Overall Loss 0.288034 Objective Loss 0.288034 LR 0.001000 Time 0.020642 -2022-12-06 10:41:33,724 - Epoch: [43][ 730/ 1200] Overall Loss 0.288146 Objective Loss 0.288146 LR 0.001000 Time 0.020619 -2022-12-06 10:41:33,914 - Epoch: [43][ 740/ 1200] Overall Loss 0.287733 Objective Loss 0.287733 LR 0.001000 Time 0.020596 -2022-12-06 10:41:34,104 - Epoch: [43][ 750/ 1200] Overall Loss 0.288346 Objective Loss 0.288346 LR 0.001000 Time 0.020574 -2022-12-06 10:41:34,294 - Epoch: [43][ 760/ 1200] Overall Loss 0.288377 Objective Loss 0.288377 LR 0.001000 Time 0.020553 -2022-12-06 10:41:34,484 - Epoch: [43][ 770/ 1200] Overall Loss 0.288584 Objective Loss 0.288584 LR 0.001000 Time 0.020532 -2022-12-06 10:41:34,674 - Epoch: [43][ 780/ 1200] Overall Loss 0.288499 Objective Loss 0.288499 LR 0.001000 Time 0.020511 -2022-12-06 10:41:34,863 - Epoch: [43][ 790/ 1200] Overall Loss 0.288356 Objective Loss 0.288356 LR 0.001000 Time 0.020491 -2022-12-06 10:41:35,053 - Epoch: [43][ 800/ 1200] Overall Loss 0.288271 Objective Loss 0.288271 LR 0.001000 Time 0.020471 -2022-12-06 10:41:35,242 - Epoch: [43][ 810/ 1200] Overall Loss 0.288179 Objective Loss 0.288179 LR 0.001000 Time 0.020452 -2022-12-06 10:41:35,432 - Epoch: [43][ 820/ 1200] Overall Loss 0.288253 Objective Loss 0.288253 LR 0.001000 Time 0.020433 -2022-12-06 10:41:35,621 - Epoch: [43][ 830/ 1200] Overall Loss 0.288262 Objective Loss 0.288262 LR 0.001000 Time 0.020414 -2022-12-06 10:41:35,811 - Epoch: [43][ 840/ 1200] Overall Loss 0.288300 Objective Loss 0.288300 LR 0.001000 Time 0.020397 -2022-12-06 10:41:36,002 - Epoch: [43][ 850/ 1200] Overall Loss 0.288080 Objective Loss 0.288080 LR 0.001000 Time 0.020380 -2022-12-06 10:41:36,191 - Epoch: [43][ 860/ 1200] Overall Loss 0.288453 Objective Loss 0.288453 LR 0.001000 Time 0.020362 -2022-12-06 10:41:36,381 - Epoch: [43][ 870/ 1200] Overall Loss 0.288230 Objective Loss 0.288230 LR 0.001000 Time 0.020347 -2022-12-06 10:41:36,571 - Epoch: [43][ 880/ 1200] Overall Loss 0.288058 Objective Loss 0.288058 LR 0.001000 Time 0.020330 -2022-12-06 10:41:36,761 - Epoch: [43][ 890/ 1200] Overall Loss 0.288177 Objective Loss 0.288177 LR 0.001000 Time 0.020314 -2022-12-06 10:41:36,950 - Epoch: [43][ 900/ 1200] Overall Loss 0.288099 Objective Loss 0.288099 LR 0.001000 Time 0.020299 -2022-12-06 10:41:37,140 - Epoch: [43][ 910/ 1200] Overall Loss 0.288169 Objective Loss 0.288169 LR 0.001000 Time 0.020284 -2022-12-06 10:41:37,330 - Epoch: [43][ 920/ 1200] Overall Loss 0.288193 Objective Loss 0.288193 LR 0.001000 Time 0.020269 -2022-12-06 10:41:37,520 - Epoch: [43][ 930/ 1200] Overall Loss 0.288028 Objective Loss 0.288028 LR 0.001000 Time 0.020254 -2022-12-06 10:41:37,709 - Epoch: [43][ 940/ 1200] Overall Loss 0.288344 Objective Loss 0.288344 LR 0.001000 Time 0.020240 -2022-12-06 10:41:37,899 - Epoch: [43][ 950/ 1200] Overall Loss 0.288455 Objective Loss 0.288455 LR 0.001000 Time 0.020227 -2022-12-06 10:41:38,089 - Epoch: [43][ 960/ 1200] Overall Loss 0.288570 Objective Loss 0.288570 LR 0.001000 Time 0.020213 -2022-12-06 10:41:38,278 - Epoch: [43][ 970/ 1200] Overall Loss 0.288303 Objective Loss 0.288303 LR 0.001000 Time 0.020199 -2022-12-06 10:41:38,468 - Epoch: [43][ 980/ 1200] Overall Loss 0.288273 Objective Loss 0.288273 LR 0.001000 Time 0.020186 -2022-12-06 10:41:38,658 - Epoch: [43][ 990/ 1200] Overall Loss 0.288526 Objective Loss 0.288526 LR 0.001000 Time 0.020173 -2022-12-06 10:41:38,847 - Epoch: [43][ 1000/ 1200] Overall Loss 0.288615 Objective Loss 0.288615 LR 0.001000 Time 0.020160 -2022-12-06 10:41:39,037 - Epoch: [43][ 1010/ 1200] Overall Loss 0.288467 Objective Loss 0.288467 LR 0.001000 Time 0.020148 -2022-12-06 10:41:39,226 - Epoch: [43][ 1020/ 1200] Overall Loss 0.288239 Objective Loss 0.288239 LR 0.001000 Time 0.020136 -2022-12-06 10:41:39,417 - Epoch: [43][ 1030/ 1200] Overall Loss 0.288208 Objective Loss 0.288208 LR 0.001000 Time 0.020125 -2022-12-06 10:41:39,606 - Epoch: [43][ 1040/ 1200] Overall Loss 0.288345 Objective Loss 0.288345 LR 0.001000 Time 0.020113 -2022-12-06 10:41:39,796 - Epoch: [43][ 1050/ 1200] Overall Loss 0.288244 Objective Loss 0.288244 LR 0.001000 Time 0.020102 -2022-12-06 10:41:39,986 - Epoch: [43][ 1060/ 1200] Overall Loss 0.288457 Objective Loss 0.288457 LR 0.001000 Time 0.020090 -2022-12-06 10:41:40,176 - Epoch: [43][ 1070/ 1200] Overall Loss 0.288750 Objective Loss 0.288750 LR 0.001000 Time 0.020080 -2022-12-06 10:41:40,366 - Epoch: [43][ 1080/ 1200] Overall Loss 0.288984 Objective Loss 0.288984 LR 0.001000 Time 0.020069 -2022-12-06 10:41:40,556 - Epoch: [43][ 1090/ 1200] Overall Loss 0.289390 Objective Loss 0.289390 LR 0.001000 Time 0.020059 -2022-12-06 10:41:40,745 - Epoch: [43][ 1100/ 1200] Overall Loss 0.289727 Objective Loss 0.289727 LR 0.001000 Time 0.020048 -2022-12-06 10:41:40,934 - Epoch: [43][ 1110/ 1200] Overall Loss 0.289834 Objective Loss 0.289834 LR 0.001000 Time 0.020038 -2022-12-06 10:41:41,124 - Epoch: [43][ 1120/ 1200] Overall Loss 0.289555 Objective Loss 0.289555 LR 0.001000 Time 0.020027 -2022-12-06 10:41:41,313 - Epoch: [43][ 1130/ 1200] Overall Loss 0.289625 Objective Loss 0.289625 LR 0.001000 Time 0.020017 -2022-12-06 10:41:41,503 - Epoch: [43][ 1140/ 1200] Overall Loss 0.289715 Objective Loss 0.289715 LR 0.001000 Time 0.020007 -2022-12-06 10:41:41,693 - Epoch: [43][ 1150/ 1200] Overall Loss 0.289454 Objective Loss 0.289454 LR 0.001000 Time 0.019998 -2022-12-06 10:41:41,882 - Epoch: [43][ 1160/ 1200] Overall Loss 0.289716 Objective Loss 0.289716 LR 0.001000 Time 0.019988 -2022-12-06 10:41:42,072 - Epoch: [43][ 1170/ 1200] Overall Loss 0.289674 Objective Loss 0.289674 LR 0.001000 Time 0.019979 -2022-12-06 10:41:42,261 - Epoch: [43][ 1180/ 1200] Overall Loss 0.289379 Objective Loss 0.289379 LR 0.001000 Time 0.019970 -2022-12-06 10:41:42,451 - Epoch: [43][ 1190/ 1200] Overall Loss 0.289446 Objective Loss 0.289446 LR 0.001000 Time 0.019961 -2022-12-06 10:41:42,681 - Epoch: [43][ 1200/ 1200] Overall Loss 0.289346 Objective Loss 0.289346 Top1 82.008368 Top5 99.581590 LR 0.001000 Time 0.019986 -2022-12-06 10:41:42,770 - --- validate (epoch=43)----------- -2022-12-06 10:41:42,770 - 34129 samples (256 per mini-batch) -2022-12-06 10:41:43,217 - Epoch: [43][ 10/ 134] Loss 0.306681 Top1 85.390625 Top5 97.812500 -2022-12-06 10:41:43,349 - Epoch: [43][ 20/ 134] Loss 0.302922 Top1 84.843750 Top5 97.851562 -2022-12-06 10:41:43,481 - Epoch: [43][ 30/ 134] Loss 0.285618 Top1 84.921875 Top5 97.903646 -2022-12-06 10:41:43,613 - Epoch: [43][ 40/ 134] Loss 0.294740 Top1 84.755859 Top5 97.871094 -2022-12-06 10:41:43,744 - Epoch: [43][ 50/ 134] Loss 0.300540 Top1 84.773438 Top5 97.875000 -2022-12-06 10:41:43,875 - Epoch: [43][ 60/ 134] Loss 0.298265 Top1 84.609375 Top5 97.890625 -2022-12-06 10:41:44,006 - Epoch: [43][ 70/ 134] Loss 0.300111 Top1 84.575893 Top5 97.868304 -2022-12-06 10:41:44,135 - Epoch: [43][ 80/ 134] Loss 0.299437 Top1 84.687500 Top5 97.900391 -2022-12-06 10:41:44,266 - Epoch: [43][ 90/ 134] Loss 0.298250 Top1 84.704861 Top5 97.903646 -2022-12-06 10:41:44,397 - Epoch: [43][ 100/ 134] Loss 0.297231 Top1 84.707031 Top5 97.906250 -2022-12-06 10:41:44,529 - Epoch: [43][ 110/ 134] Loss 0.296724 Top1 84.673295 Top5 97.890625 -2022-12-06 10:41:44,660 - Epoch: [43][ 120/ 134] Loss 0.299023 Top1 84.661458 Top5 97.825521 -2022-12-06 10:41:44,791 - Epoch: [43][ 130/ 134] Loss 0.297417 Top1 84.705529 Top5 97.803486 -2022-12-06 10:41:44,829 - Epoch: [43][ 134/ 134] Loss 0.297894 Top1 84.716810 Top5 97.817106 -2022-12-06 10:41:44,917 - ==> Top1: 84.717 Top5: 97.817 Loss: 0.298 - -2022-12-06 10:41:44,918 - ==> Confusion: -[[ 911 1 2 1 4 4 1 0 3 43 0 3 3 3 7 2 3 1 2 0 2] - [ 1 924 1 1 12 28 2 17 1 1 1 5 4 1 4 0 2 4 10 1 7] - [ 6 2 984 23 2 6 23 11 0 2 3 5 4 1 3 4 3 1 10 4 6] - [ 2 1 18 916 1 5 0 0 0 0 16 0 4 1 29 2 2 6 8 0 9] - [ 16 5 2 0 950 3 0 1 1 9 1 2 1 4 8 4 7 2 0 0 4] - [ 4 12 1 2 6 964 3 18 2 2 1 20 3 13 2 2 0 2 0 3 9] - [ 1 3 14 4 0 5 1060 3 0 0 6 2 0 2 1 4 1 1 1 6 4] - [ 1 5 2 4 3 35 2 944 1 1 4 8 0 1 1 1 0 1 23 13 4] - [ 6 2 0 0 1 6 0 0 969 39 9 4 1 7 13 1 2 1 0 1 2] - [ 75 1 1 0 2 4 2 0 21 863 1 4 0 12 3 1 1 2 1 0 7] - [ 1 2 1 10 0 4 1 2 10 1 950 3 1 16 3 2 1 0 4 3 4] - [ 5 1 2 1 0 11 1 7 0 0 0 972 27 2 1 2 5 2 1 9 2] - [ 1 1 2 5 1 3 1 0 1 0 0 45 874 0 3 7 1 9 3 4 8] - [ 0 1 2 1 1 13 0 3 10 11 8 5 8 943 2 1 1 0 0 3 10] - [ 18 4 1 10 4 3 0 0 17 3 1 3 1 2 1045 2 3 1 5 1 6] - [ 1 1 0 0 3 2 6 0 0 0 1 10 11 4 0 974 12 7 1 2 8] - [ 5 4 1 0 3 1 0 1 1 0 0 6 2 0 3 6 1021 4 3 4 7] - [ 2 0 0 1 0 3 3 0 0 0 1 15 33 1 2 11 2 958 2 1 1] - [ 4 6 4 10 1 5 1 25 2 1 8 3 3 0 12 1 0 1 914 2 5] - [ 1 2 0 0 1 8 4 7 0 0 2 15 7 3 1 2 2 1 1 1019 4] - [ 153 241 168 111 136 228 97 176 92 89 191 154 384 280 175 96 185 69 180 268 9753]] - -2022-12-06 10:41:45,484 - ==> Best [Top1: 84.717 Top5: 97.817 Sparsity:0.00 Params: 5376 on epoch: 43] -2022-12-06 10:41:45,484 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:41:45,491 - - -2022-12-06 10:41:45,491 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:41:46,526 - Epoch: [44][ 10/ 1200] Overall Loss 0.254318 Objective Loss 0.254318 LR 0.001000 Time 0.103359 -2022-12-06 10:41:46,723 - Epoch: [44][ 20/ 1200] Overall Loss 0.279615 Objective Loss 0.279615 LR 0.001000 Time 0.061509 -2022-12-06 10:41:46,912 - Epoch: [44][ 30/ 1200] Overall Loss 0.278359 Objective Loss 0.278359 LR 0.001000 Time 0.047299 -2022-12-06 10:41:47,101 - Epoch: [44][ 40/ 1200] Overall Loss 0.278659 Objective Loss 0.278659 LR 0.001000 Time 0.040192 -2022-12-06 10:41:47,291 - Epoch: [44][ 50/ 1200] Overall Loss 0.270926 Objective Loss 0.270926 LR 0.001000 Time 0.035931 -2022-12-06 10:41:47,479 - Epoch: [44][ 60/ 1200] Overall Loss 0.270673 Objective Loss 0.270673 LR 0.001000 Time 0.033071 -2022-12-06 10:41:47,668 - Epoch: [44][ 70/ 1200] Overall Loss 0.270673 Objective Loss 0.270673 LR 0.001000 Time 0.031043 -2022-12-06 10:41:47,857 - Epoch: [44][ 80/ 1200] Overall Loss 0.272914 Objective Loss 0.272914 LR 0.001000 Time 0.029519 -2022-12-06 10:41:48,046 - Epoch: [44][ 90/ 1200] Overall Loss 0.276200 Objective Loss 0.276200 LR 0.001000 Time 0.028333 -2022-12-06 10:41:48,235 - Epoch: [44][ 100/ 1200] Overall Loss 0.275248 Objective Loss 0.275248 LR 0.001000 Time 0.027384 -2022-12-06 10:41:48,424 - Epoch: [44][ 110/ 1200] Overall Loss 0.278677 Objective Loss 0.278677 LR 0.001000 Time 0.026608 -2022-12-06 10:41:48,615 - Epoch: [44][ 120/ 1200] Overall Loss 0.277212 Objective Loss 0.277212 LR 0.001000 Time 0.025978 -2022-12-06 10:41:48,806 - Epoch: [44][ 130/ 1200] Overall Loss 0.277821 Objective Loss 0.277821 LR 0.001000 Time 0.025441 -2022-12-06 10:41:48,998 - Epoch: [44][ 140/ 1200] Overall Loss 0.278996 Objective Loss 0.278996 LR 0.001000 Time 0.024990 -2022-12-06 10:41:49,188 - Epoch: [44][ 150/ 1200] Overall Loss 0.279141 Objective Loss 0.279141 LR 0.001000 Time 0.024588 -2022-12-06 10:41:49,379 - Epoch: [44][ 160/ 1200] Overall Loss 0.278425 Objective Loss 0.278425 LR 0.001000 Time 0.024243 -2022-12-06 10:41:49,571 - Epoch: [44][ 170/ 1200] Overall Loss 0.278050 Objective Loss 0.278050 LR 0.001000 Time 0.023941 -2022-12-06 10:41:49,761 - Epoch: [44][ 180/ 1200] Overall Loss 0.277755 Objective Loss 0.277755 LR 0.001000 Time 0.023666 -2022-12-06 10:41:49,951 - Epoch: [44][ 190/ 1200] Overall Loss 0.277461 Objective Loss 0.277461 LR 0.001000 Time 0.023419 -2022-12-06 10:41:50,143 - Epoch: [44][ 200/ 1200] Overall Loss 0.279060 Objective Loss 0.279060 LR 0.001000 Time 0.023202 -2022-12-06 10:41:50,333 - Epoch: [44][ 210/ 1200] Overall Loss 0.279828 Objective Loss 0.279828 LR 0.001000 Time 0.023003 -2022-12-06 10:41:50,524 - Epoch: [44][ 220/ 1200] Overall Loss 0.280316 Objective Loss 0.280316 LR 0.001000 Time 0.022823 -2022-12-06 10:41:50,715 - Epoch: [44][ 230/ 1200] Overall Loss 0.280277 Objective Loss 0.280277 LR 0.001000 Time 0.022658 -2022-12-06 10:41:50,905 - Epoch: [44][ 240/ 1200] Overall Loss 0.280009 Objective Loss 0.280009 LR 0.001000 Time 0.022504 -2022-12-06 10:41:51,095 - Epoch: [44][ 250/ 1200] Overall Loss 0.280179 Objective Loss 0.280179 LR 0.001000 Time 0.022362 -2022-12-06 10:41:51,286 - Epoch: [44][ 260/ 1200] Overall Loss 0.281021 Objective Loss 0.281021 LR 0.001000 Time 0.022234 -2022-12-06 10:41:51,477 - Epoch: [44][ 270/ 1200] Overall Loss 0.281634 Objective Loss 0.281634 LR 0.001000 Time 0.022116 -2022-12-06 10:41:51,668 - Epoch: [44][ 280/ 1200] Overall Loss 0.282152 Objective Loss 0.282152 LR 0.001000 Time 0.022006 -2022-12-06 10:41:51,859 - Epoch: [44][ 290/ 1200] Overall Loss 0.282301 Objective Loss 0.282301 LR 0.001000 Time 0.021904 -2022-12-06 10:41:52,050 - Epoch: [44][ 300/ 1200] Overall Loss 0.281816 Objective Loss 0.281816 LR 0.001000 Time 0.021806 -2022-12-06 10:41:52,241 - Epoch: [44][ 310/ 1200] Overall Loss 0.282147 Objective Loss 0.282147 LR 0.001000 Time 0.021720 -2022-12-06 10:41:52,433 - Epoch: [44][ 320/ 1200] Overall Loss 0.280640 Objective Loss 0.280640 LR 0.001000 Time 0.021639 -2022-12-06 10:41:52,623 - Epoch: [44][ 330/ 1200] Overall Loss 0.279904 Objective Loss 0.279904 LR 0.001000 Time 0.021558 -2022-12-06 10:41:52,814 - Epoch: [44][ 340/ 1200] Overall Loss 0.279421 Objective Loss 0.279421 LR 0.001000 Time 0.021482 -2022-12-06 10:41:53,004 - Epoch: [44][ 350/ 1200] Overall Loss 0.279779 Objective Loss 0.279779 LR 0.001000 Time 0.021410 -2022-12-06 10:41:53,195 - Epoch: [44][ 360/ 1200] Overall Loss 0.279355 Objective Loss 0.279355 LR 0.001000 Time 0.021344 -2022-12-06 10:41:53,386 - Epoch: [44][ 370/ 1200] Overall Loss 0.278772 Objective Loss 0.278772 LR 0.001000 Time 0.021283 -2022-12-06 10:41:53,577 - Epoch: [44][ 380/ 1200] Overall Loss 0.278435 Objective Loss 0.278435 LR 0.001000 Time 0.021223 -2022-12-06 10:41:53,768 - Epoch: [44][ 390/ 1200] Overall Loss 0.278917 Objective Loss 0.278917 LR 0.001000 Time 0.021169 -2022-12-06 10:41:53,959 - Epoch: [44][ 400/ 1200] Overall Loss 0.278787 Objective Loss 0.278787 LR 0.001000 Time 0.021115 -2022-12-06 10:41:54,150 - Epoch: [44][ 410/ 1200] Overall Loss 0.278871 Objective Loss 0.278871 LR 0.001000 Time 0.021065 -2022-12-06 10:41:54,341 - Epoch: [44][ 420/ 1200] Overall Loss 0.279298 Objective Loss 0.279298 LR 0.001000 Time 0.021017 -2022-12-06 10:41:54,532 - Epoch: [44][ 430/ 1200] Overall Loss 0.278664 Objective Loss 0.278664 LR 0.001000 Time 0.020970 -2022-12-06 10:41:54,723 - Epoch: [44][ 440/ 1200] Overall Loss 0.279310 Objective Loss 0.279310 LR 0.001000 Time 0.020927 -2022-12-06 10:41:54,913 - Epoch: [44][ 450/ 1200] Overall Loss 0.278896 Objective Loss 0.278896 LR 0.001000 Time 0.020884 -2022-12-06 10:41:55,105 - Epoch: [44][ 460/ 1200] Overall Loss 0.279304 Objective Loss 0.279304 LR 0.001000 Time 0.020845 -2022-12-06 10:41:55,296 - Epoch: [44][ 470/ 1200] Overall Loss 0.279732 Objective Loss 0.279732 LR 0.001000 Time 0.020806 -2022-12-06 10:41:55,487 - Epoch: [44][ 480/ 1200] Overall Loss 0.281114 Objective Loss 0.281114 LR 0.001000 Time 0.020769 -2022-12-06 10:41:55,678 - Epoch: [44][ 490/ 1200] Overall Loss 0.281287 Objective Loss 0.281287 LR 0.001000 Time 0.020736 -2022-12-06 10:41:55,869 - Epoch: [44][ 500/ 1200] Overall Loss 0.280957 Objective Loss 0.280957 LR 0.001000 Time 0.020702 -2022-12-06 10:41:56,060 - Epoch: [44][ 510/ 1200] Overall Loss 0.280891 Objective Loss 0.280891 LR 0.001000 Time 0.020670 -2022-12-06 10:41:56,251 - Epoch: [44][ 520/ 1200] Overall Loss 0.280863 Objective Loss 0.280863 LR 0.001000 Time 0.020638 -2022-12-06 10:41:56,443 - Epoch: [44][ 530/ 1200] Overall Loss 0.281042 Objective Loss 0.281042 LR 0.001000 Time 0.020610 -2022-12-06 10:41:56,634 - Epoch: [44][ 540/ 1200] Overall Loss 0.281119 Objective Loss 0.281119 LR 0.001000 Time 0.020581 -2022-12-06 10:41:56,825 - Epoch: [44][ 550/ 1200] Overall Loss 0.281362 Objective Loss 0.281362 LR 0.001000 Time 0.020552 -2022-12-06 10:41:57,015 - Epoch: [44][ 560/ 1200] Overall Loss 0.281846 Objective Loss 0.281846 LR 0.001000 Time 0.020525 -2022-12-06 10:41:57,206 - Epoch: [44][ 570/ 1200] Overall Loss 0.281408 Objective Loss 0.281408 LR 0.001000 Time 0.020499 -2022-12-06 10:41:57,398 - Epoch: [44][ 580/ 1200] Overall Loss 0.281035 Objective Loss 0.281035 LR 0.001000 Time 0.020474 -2022-12-06 10:41:57,589 - Epoch: [44][ 590/ 1200] Overall Loss 0.281104 Objective Loss 0.281104 LR 0.001000 Time 0.020450 -2022-12-06 10:41:57,780 - Epoch: [44][ 600/ 1200] Overall Loss 0.281798 Objective Loss 0.281798 LR 0.001000 Time 0.020427 -2022-12-06 10:41:57,971 - Epoch: [44][ 610/ 1200] Overall Loss 0.281897 Objective Loss 0.281897 LR 0.001000 Time 0.020405 -2022-12-06 10:41:58,162 - Epoch: [44][ 620/ 1200] Overall Loss 0.282161 Objective Loss 0.282161 LR 0.001000 Time 0.020383 -2022-12-06 10:41:58,353 - Epoch: [44][ 630/ 1200] Overall Loss 0.282337 Objective Loss 0.282337 LR 0.001000 Time 0.020362 -2022-12-06 10:41:58,545 - Epoch: [44][ 640/ 1200] Overall Loss 0.282834 Objective Loss 0.282834 LR 0.001000 Time 0.020342 -2022-12-06 10:41:58,735 - Epoch: [44][ 650/ 1200] Overall Loss 0.282780 Objective Loss 0.282780 LR 0.001000 Time 0.020322 -2022-12-06 10:41:58,926 - Epoch: [44][ 660/ 1200] Overall Loss 0.282784 Objective Loss 0.282784 LR 0.001000 Time 0.020302 -2022-12-06 10:41:59,116 - Epoch: [44][ 670/ 1200] Overall Loss 0.282837 Objective Loss 0.282837 LR 0.001000 Time 0.020281 -2022-12-06 10:41:59,306 - Epoch: [44][ 680/ 1200] Overall Loss 0.282384 Objective Loss 0.282384 LR 0.001000 Time 0.020262 -2022-12-06 10:41:59,497 - Epoch: [44][ 690/ 1200] Overall Loss 0.282011 Objective Loss 0.282011 LR 0.001000 Time 0.020245 -2022-12-06 10:41:59,688 - Epoch: [44][ 700/ 1200] Overall Loss 0.282097 Objective Loss 0.282097 LR 0.001000 Time 0.020227 -2022-12-06 10:41:59,879 - Epoch: [44][ 710/ 1200] Overall Loss 0.282251 Objective Loss 0.282251 LR 0.001000 Time 0.020210 -2022-12-06 10:42:00,070 - Epoch: [44][ 720/ 1200] Overall Loss 0.282370 Objective Loss 0.282370 LR 0.001000 Time 0.020194 -2022-12-06 10:42:00,260 - Epoch: [44][ 730/ 1200] Overall Loss 0.282201 Objective Loss 0.282201 LR 0.001000 Time 0.020177 -2022-12-06 10:42:00,451 - Epoch: [44][ 740/ 1200] Overall Loss 0.282483 Objective Loss 0.282483 LR 0.001000 Time 0.020161 -2022-12-06 10:42:00,642 - Epoch: [44][ 750/ 1200] Overall Loss 0.282987 Objective Loss 0.282987 LR 0.001000 Time 0.020147 -2022-12-06 10:42:00,832 - Epoch: [44][ 760/ 1200] Overall Loss 0.283309 Objective Loss 0.283309 LR 0.001000 Time 0.020132 -2022-12-06 10:42:01,023 - Epoch: [44][ 770/ 1200] Overall Loss 0.283146 Objective Loss 0.283146 LR 0.001000 Time 0.020117 -2022-12-06 10:42:01,214 - Epoch: [44][ 780/ 1200] Overall Loss 0.283040 Objective Loss 0.283040 LR 0.001000 Time 0.020103 -2022-12-06 10:42:01,406 - Epoch: [44][ 790/ 1200] Overall Loss 0.283055 Objective Loss 0.283055 LR 0.001000 Time 0.020091 -2022-12-06 10:42:01,597 - Epoch: [44][ 800/ 1200] Overall Loss 0.282736 Objective Loss 0.282736 LR 0.001000 Time 0.020078 -2022-12-06 10:42:01,787 - Epoch: [44][ 810/ 1200] Overall Loss 0.282687 Objective Loss 0.282687 LR 0.001000 Time 0.020065 -2022-12-06 10:42:01,978 - Epoch: [44][ 820/ 1200] Overall Loss 0.282703 Objective Loss 0.282703 LR 0.001000 Time 0.020052 -2022-12-06 10:42:02,169 - Epoch: [44][ 830/ 1200] Overall Loss 0.282560 Objective Loss 0.282560 LR 0.001000 Time 0.020040 -2022-12-06 10:42:02,360 - Epoch: [44][ 840/ 1200] Overall Loss 0.282317 Objective Loss 0.282317 LR 0.001000 Time 0.020028 -2022-12-06 10:42:02,551 - Epoch: [44][ 850/ 1200] Overall Loss 0.282564 Objective Loss 0.282564 LR 0.001000 Time 0.020017 -2022-12-06 10:42:02,742 - Epoch: [44][ 860/ 1200] Overall Loss 0.282600 Objective Loss 0.282600 LR 0.001000 Time 0.020005 -2022-12-06 10:42:02,932 - Epoch: [44][ 870/ 1200] Overall Loss 0.282688 Objective Loss 0.282688 LR 0.001000 Time 0.019993 -2022-12-06 10:42:03,123 - Epoch: [44][ 880/ 1200] Overall Loss 0.282447 Objective Loss 0.282447 LR 0.001000 Time 0.019982 -2022-12-06 10:42:03,314 - Epoch: [44][ 890/ 1200] Overall Loss 0.282647 Objective Loss 0.282647 LR 0.001000 Time 0.019971 -2022-12-06 10:42:03,505 - Epoch: [44][ 900/ 1200] Overall Loss 0.282721 Objective Loss 0.282721 LR 0.001000 Time 0.019961 -2022-12-06 10:42:03,695 - Epoch: [44][ 910/ 1200] Overall Loss 0.282406 Objective Loss 0.282406 LR 0.001000 Time 0.019951 -2022-12-06 10:42:03,886 - Epoch: [44][ 920/ 1200] Overall Loss 0.282485 Objective Loss 0.282485 LR 0.001000 Time 0.019941 -2022-12-06 10:42:04,076 - Epoch: [44][ 930/ 1200] Overall Loss 0.282693 Objective Loss 0.282693 LR 0.001000 Time 0.019929 -2022-12-06 10:42:04,266 - Epoch: [44][ 940/ 1200] Overall Loss 0.282740 Objective Loss 0.282740 LR 0.001000 Time 0.019919 -2022-12-06 10:42:04,455 - Epoch: [44][ 950/ 1200] Overall Loss 0.283073 Objective Loss 0.283073 LR 0.001000 Time 0.019908 -2022-12-06 10:42:04,645 - Epoch: [44][ 960/ 1200] Overall Loss 0.282878 Objective Loss 0.282878 LR 0.001000 Time 0.019898 -2022-12-06 10:42:04,834 - Epoch: [44][ 970/ 1200] Overall Loss 0.282688 Objective Loss 0.282688 LR 0.001000 Time 0.019887 -2022-12-06 10:42:05,023 - Epoch: [44][ 980/ 1200] Overall Loss 0.283015 Objective Loss 0.283015 LR 0.001000 Time 0.019877 -2022-12-06 10:42:05,213 - Epoch: [44][ 990/ 1200] Overall Loss 0.283131 Objective Loss 0.283131 LR 0.001000 Time 0.019867 -2022-12-06 10:42:05,402 - Epoch: [44][ 1000/ 1200] Overall Loss 0.282989 Objective Loss 0.282989 LR 0.001000 Time 0.019857 -2022-12-06 10:42:05,592 - Epoch: [44][ 1010/ 1200] Overall Loss 0.282945 Objective Loss 0.282945 LR 0.001000 Time 0.019848 -2022-12-06 10:42:05,781 - Epoch: [44][ 1020/ 1200] Overall Loss 0.283059 Objective Loss 0.283059 LR 0.001000 Time 0.019838 -2022-12-06 10:42:05,971 - Epoch: [44][ 1030/ 1200] Overall Loss 0.283070 Objective Loss 0.283070 LR 0.001000 Time 0.019829 -2022-12-06 10:42:06,160 - Epoch: [44][ 1040/ 1200] Overall Loss 0.282819 Objective Loss 0.282819 LR 0.001000 Time 0.019820 -2022-12-06 10:42:06,350 - Epoch: [44][ 1050/ 1200] Overall Loss 0.282775 Objective Loss 0.282775 LR 0.001000 Time 0.019811 -2022-12-06 10:42:06,539 - Epoch: [44][ 1060/ 1200] Overall Loss 0.282807 Objective Loss 0.282807 LR 0.001000 Time 0.019802 -2022-12-06 10:42:06,728 - Epoch: [44][ 1070/ 1200] Overall Loss 0.282627 Objective Loss 0.282627 LR 0.001000 Time 0.019794 -2022-12-06 10:42:06,917 - Epoch: [44][ 1080/ 1200] Overall Loss 0.282828 Objective Loss 0.282828 LR 0.001000 Time 0.019785 -2022-12-06 10:42:07,107 - Epoch: [44][ 1090/ 1200] Overall Loss 0.282990 Objective Loss 0.282990 LR 0.001000 Time 0.019777 -2022-12-06 10:42:07,296 - Epoch: [44][ 1100/ 1200] Overall Loss 0.283131 Objective Loss 0.283131 LR 0.001000 Time 0.019769 -2022-12-06 10:42:07,485 - Epoch: [44][ 1110/ 1200] Overall Loss 0.283258 Objective Loss 0.283258 LR 0.001000 Time 0.019760 -2022-12-06 10:42:07,674 - Epoch: [44][ 1120/ 1200] Overall Loss 0.283284 Objective Loss 0.283284 LR 0.001000 Time 0.019753 -2022-12-06 10:42:07,864 - Epoch: [44][ 1130/ 1200] Overall Loss 0.283343 Objective Loss 0.283343 LR 0.001000 Time 0.019745 -2022-12-06 10:42:08,054 - Epoch: [44][ 1140/ 1200] Overall Loss 0.283361 Objective Loss 0.283361 LR 0.001000 Time 0.019738 -2022-12-06 10:42:08,244 - Epoch: [44][ 1150/ 1200] Overall Loss 0.283467 Objective Loss 0.283467 LR 0.001000 Time 0.019731 -2022-12-06 10:42:08,433 - Epoch: [44][ 1160/ 1200] Overall Loss 0.283429 Objective Loss 0.283429 LR 0.001000 Time 0.019724 -2022-12-06 10:42:08,623 - Epoch: [44][ 1170/ 1200] Overall Loss 0.283334 Objective Loss 0.283334 LR 0.001000 Time 0.019717 -2022-12-06 10:42:08,813 - Epoch: [44][ 1180/ 1200] Overall Loss 0.283500 Objective Loss 0.283500 LR 0.001000 Time 0.019710 -2022-12-06 10:42:09,003 - Epoch: [44][ 1190/ 1200] Overall Loss 0.283472 Objective Loss 0.283472 LR 0.001000 Time 0.019704 -2022-12-06 10:42:09,226 - Epoch: [44][ 1200/ 1200] Overall Loss 0.283509 Objective Loss 0.283509 Top1 82.426778 Top5 97.907950 LR 0.001000 Time 0.019725 -2022-12-06 10:42:09,323 - --- validate (epoch=44)----------- -2022-12-06 10:42:09,323 - 34129 samples (256 per mini-batch) -2022-12-06 10:42:09,771 - Epoch: [44][ 10/ 134] Loss 0.311349 Top1 82.851562 Top5 96.679688 -2022-12-06 10:42:09,894 - Epoch: [44][ 20/ 134] Loss 0.302582 Top1 83.183594 Top5 97.109375 -2022-12-06 10:42:10,021 - Epoch: [44][ 30/ 134] Loss 0.304093 Top1 83.515625 Top5 97.265625 -2022-12-06 10:42:10,145 - Epoch: [44][ 40/ 134] Loss 0.301584 Top1 83.320312 Top5 97.275391 -2022-12-06 10:42:10,270 - Epoch: [44][ 50/ 134] Loss 0.313970 Top1 83.140625 Top5 97.250000 -2022-12-06 10:42:10,392 - Epoch: [44][ 60/ 134] Loss 0.312959 Top1 83.235677 Top5 97.285156 -2022-12-06 10:42:10,520 - Epoch: [44][ 70/ 134] Loss 0.310589 Top1 83.203125 Top5 97.371652 -2022-12-06 10:42:10,643 - Epoch: [44][ 80/ 134] Loss 0.308720 Top1 83.334961 Top5 97.460938 -2022-12-06 10:42:10,767 - Epoch: [44][ 90/ 134] Loss 0.307717 Top1 83.333333 Top5 97.465278 -2022-12-06 10:42:10,892 - Epoch: [44][ 100/ 134] Loss 0.306504 Top1 83.332031 Top5 97.468750 -2022-12-06 10:42:11,017 - Epoch: [44][ 110/ 134] Loss 0.307907 Top1 83.316761 Top5 97.492898 -2022-12-06 10:42:11,141 - Epoch: [44][ 120/ 134] Loss 0.308763 Top1 83.313802 Top5 97.483724 -2022-12-06 10:42:11,267 - Epoch: [44][ 130/ 134] Loss 0.309824 Top1 83.239183 Top5 97.430889 -2022-12-06 10:42:11,303 - Epoch: [44][ 134/ 134] Loss 0.310099 Top1 83.225409 Top5 97.418618 -2022-12-06 10:42:11,391 - ==> Top1: 83.225 Top5: 97.419 Loss: 0.310 - -2022-12-06 10:42:11,392 - ==> Confusion: -[[ 851 2 3 0 8 5 1 2 6 96 1 2 1 2 9 3 0 1 1 0 2] - [ 3 908 2 1 12 44 1 19 3 1 3 4 1 4 2 0 3 3 9 0 4] - [ 4 5 955 23 1 4 34 19 1 5 5 8 3 3 4 8 2 2 5 2 10] - [ 2 3 18 926 0 8 1 2 0 1 13 0 3 1 20 0 0 3 9 1 9] - [ 6 7 3 0 943 7 1 1 1 10 3 6 0 3 11 7 5 4 0 1 1] - [ 1 12 0 2 8 973 3 19 4 2 0 15 4 17 2 1 0 1 0 2 3] - [ 0 2 5 4 0 4 1066 9 0 0 1 2 4 1 0 9 0 1 1 7 2] - [ 1 9 2 3 2 38 6 943 1 0 4 5 2 4 0 2 1 1 15 13 2] - [ 8 4 0 0 1 2 0 0 972 43 7 4 0 7 9 1 2 1 1 1 1] - [ 35 0 2 0 0 2 0 3 22 919 1 0 0 6 3 1 0 2 1 1 3] - [ 0 0 2 2 1 5 1 2 20 0 952 3 0 17 4 0 0 0 6 2 2] - [ 5 4 2 0 0 7 1 2 0 1 1 983 18 4 1 3 3 4 2 6 4] - [ 1 2 0 6 0 4 0 1 1 0 0 47 873 1 2 7 2 9 2 2 9] - [ 0 4 0 1 0 12 0 3 12 17 2 11 2 943 2 1 2 0 0 3 8] - [ 7 5 3 11 5 6 1 1 22 8 4 0 1 4 1031 1 1 3 7 2 7] - [ 1 0 1 0 0 5 4 0 0 0 1 13 10 2 0 985 6 7 0 3 5] - [ 3 3 0 0 0 5 0 0 2 1 0 5 1 1 2 13 1023 3 0 6 4] - [ 4 3 0 4 0 3 1 0 4 2 0 9 23 3 0 10 5 961 1 1 2] - [ 3 3 4 13 1 8 1 31 2 1 7 3 5 0 14 0 0 0 907 3 2] - [ 3 2 0 1 0 11 5 17 0 0 2 29 9 5 0 4 8 2 0 977 5] - [ 126 230 181 131 144 279 92 187 145 157 209 168 400 406 180 163 213 91 173 248 9303]] - -2022-12-06 10:42:11,960 - ==> Best [Top1: 84.717 Top5: 97.817 Sparsity:0.00 Params: 5376 on epoch: 43] -2022-12-06 10:42:11,960 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:42:11,966 - - -2022-12-06 10:42:11,966 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:42:12,897 - Epoch: [45][ 10/ 1200] Overall Loss 0.285719 Objective Loss 0.285719 LR 0.001000 Time 0.093027 -2022-12-06 10:42:13,096 - Epoch: [45][ 20/ 1200] Overall Loss 0.263943 Objective Loss 0.263943 LR 0.001000 Time 0.056422 -2022-12-06 10:42:13,291 - Epoch: [45][ 30/ 1200] Overall Loss 0.261510 Objective Loss 0.261510 LR 0.001000 Time 0.044117 -2022-12-06 10:42:13,489 - Epoch: [45][ 40/ 1200] Overall Loss 0.266684 Objective Loss 0.266684 LR 0.001000 Time 0.038023 -2022-12-06 10:42:13,685 - Epoch: [45][ 50/ 1200] Overall Loss 0.268303 Objective Loss 0.268303 LR 0.001000 Time 0.034314 -2022-12-06 10:42:13,883 - Epoch: [45][ 60/ 1200] Overall Loss 0.272877 Objective Loss 0.272877 LR 0.001000 Time 0.031885 -2022-12-06 10:42:14,078 - Epoch: [45][ 70/ 1200] Overall Loss 0.275319 Objective Loss 0.275319 LR 0.001000 Time 0.030116 -2022-12-06 10:42:14,276 - Epoch: [45][ 80/ 1200] Overall Loss 0.276451 Objective Loss 0.276451 LR 0.001000 Time 0.028820 -2022-12-06 10:42:14,473 - Epoch: [45][ 90/ 1200] Overall Loss 0.277614 Objective Loss 0.277614 LR 0.001000 Time 0.027798 -2022-12-06 10:42:14,672 - Epoch: [45][ 100/ 1200] Overall Loss 0.276419 Objective Loss 0.276419 LR 0.001000 Time 0.026998 -2022-12-06 10:42:14,867 - Epoch: [45][ 110/ 1200] Overall Loss 0.278234 Objective Loss 0.278234 LR 0.001000 Time 0.026318 -2022-12-06 10:42:15,065 - Epoch: [45][ 120/ 1200] Overall Loss 0.277652 Objective Loss 0.277652 LR 0.001000 Time 0.025767 -2022-12-06 10:42:15,260 - Epoch: [45][ 130/ 1200] Overall Loss 0.276495 Objective Loss 0.276495 LR 0.001000 Time 0.025283 -2022-12-06 10:42:15,458 - Epoch: [45][ 140/ 1200] Overall Loss 0.277784 Objective Loss 0.277784 LR 0.001000 Time 0.024883 -2022-12-06 10:42:15,654 - Epoch: [45][ 150/ 1200] Overall Loss 0.278925 Objective Loss 0.278925 LR 0.001000 Time 0.024529 -2022-12-06 10:42:15,851 - Epoch: [45][ 160/ 1200] Overall Loss 0.278372 Objective Loss 0.278372 LR 0.001000 Time 0.024228 -2022-12-06 10:42:16,048 - Epoch: [45][ 170/ 1200] Overall Loss 0.278128 Objective Loss 0.278128 LR 0.001000 Time 0.023955 -2022-12-06 10:42:16,246 - Epoch: [45][ 180/ 1200] Overall Loss 0.279331 Objective Loss 0.279331 LR 0.001000 Time 0.023719 -2022-12-06 10:42:16,441 - Epoch: [45][ 190/ 1200] Overall Loss 0.280825 Objective Loss 0.280825 LR 0.001000 Time 0.023499 -2022-12-06 10:42:16,640 - Epoch: [45][ 200/ 1200] Overall Loss 0.279580 Objective Loss 0.279580 LR 0.001000 Time 0.023313 -2022-12-06 10:42:16,837 - Epoch: [45][ 210/ 1200] Overall Loss 0.280939 Objective Loss 0.280939 LR 0.001000 Time 0.023139 -2022-12-06 10:42:17,035 - Epoch: [45][ 220/ 1200] Overall Loss 0.280764 Objective Loss 0.280764 LR 0.001000 Time 0.022986 -2022-12-06 10:42:17,232 - Epoch: [45][ 230/ 1200] Overall Loss 0.280218 Objective Loss 0.280218 LR 0.001000 Time 0.022839 -2022-12-06 10:42:17,430 - Epoch: [45][ 240/ 1200] Overall Loss 0.279829 Objective Loss 0.279829 LR 0.001000 Time 0.022710 -2022-12-06 10:42:17,626 - Epoch: [45][ 250/ 1200] Overall Loss 0.280151 Objective Loss 0.280151 LR 0.001000 Time 0.022583 -2022-12-06 10:42:17,824 - Epoch: [45][ 260/ 1200] Overall Loss 0.278605 Objective Loss 0.278605 LR 0.001000 Time 0.022474 -2022-12-06 10:42:18,019 - Epoch: [45][ 270/ 1200] Overall Loss 0.278346 Objective Loss 0.278346 LR 0.001000 Time 0.022365 -2022-12-06 10:42:18,217 - Epoch: [45][ 280/ 1200] Overall Loss 0.278083 Objective Loss 0.278083 LR 0.001000 Time 0.022271 -2022-12-06 10:42:18,413 - Epoch: [45][ 290/ 1200] Overall Loss 0.277983 Objective Loss 0.277983 LR 0.001000 Time 0.022177 -2022-12-06 10:42:18,611 - Epoch: [45][ 300/ 1200] Overall Loss 0.278844 Objective Loss 0.278844 LR 0.001000 Time 0.022095 -2022-12-06 10:42:18,807 - Epoch: [45][ 310/ 1200] Overall Loss 0.278989 Objective Loss 0.278989 LR 0.001000 Time 0.022012 -2022-12-06 10:42:19,005 - Epoch: [45][ 320/ 1200] Overall Loss 0.278392 Objective Loss 0.278392 LR 0.001000 Time 0.021941 -2022-12-06 10:42:19,201 - Epoch: [45][ 330/ 1200] Overall Loss 0.278476 Objective Loss 0.278476 LR 0.001000 Time 0.021869 -2022-12-06 10:42:19,399 - Epoch: [45][ 340/ 1200] Overall Loss 0.278595 Objective Loss 0.278595 LR 0.001000 Time 0.021805 -2022-12-06 10:42:19,594 - Epoch: [45][ 350/ 1200] Overall Loss 0.279525 Objective Loss 0.279525 LR 0.001000 Time 0.021739 -2022-12-06 10:42:19,792 - Epoch: [45][ 360/ 1200] Overall Loss 0.279759 Objective Loss 0.279759 LR 0.001000 Time 0.021684 -2022-12-06 10:42:19,988 - Epoch: [45][ 370/ 1200] Overall Loss 0.280573 Objective Loss 0.280573 LR 0.001000 Time 0.021625 -2022-12-06 10:42:20,185 - Epoch: [45][ 380/ 1200] Overall Loss 0.280692 Objective Loss 0.280692 LR 0.001000 Time 0.021575 -2022-12-06 10:42:20,381 - Epoch: [45][ 390/ 1200] Overall Loss 0.280125 Objective Loss 0.280125 LR 0.001000 Time 0.021523 -2022-12-06 10:42:20,579 - Epoch: [45][ 400/ 1200] Overall Loss 0.280566 Objective Loss 0.280566 LR 0.001000 Time 0.021477 -2022-12-06 10:42:20,775 - Epoch: [45][ 410/ 1200] Overall Loss 0.281035 Objective Loss 0.281035 LR 0.001000 Time 0.021431 -2022-12-06 10:42:20,973 - Epoch: [45][ 420/ 1200] Overall Loss 0.281364 Objective Loss 0.281364 LR 0.001000 Time 0.021390 -2022-12-06 10:42:21,169 - Epoch: [45][ 430/ 1200] Overall Loss 0.281130 Objective Loss 0.281130 LR 0.001000 Time 0.021348 -2022-12-06 10:42:21,368 - Epoch: [45][ 440/ 1200] Overall Loss 0.280988 Objective Loss 0.280988 LR 0.001000 Time 0.021312 -2022-12-06 10:42:21,563 - Epoch: [45][ 450/ 1200] Overall Loss 0.281400 Objective Loss 0.281400 LR 0.001000 Time 0.021272 -2022-12-06 10:42:21,762 - Epoch: [45][ 460/ 1200] Overall Loss 0.281742 Objective Loss 0.281742 LR 0.001000 Time 0.021241 -2022-12-06 10:42:21,959 - Epoch: [45][ 470/ 1200] Overall Loss 0.281509 Objective Loss 0.281509 LR 0.001000 Time 0.021207 -2022-12-06 10:42:22,160 - Epoch: [45][ 480/ 1200] Overall Loss 0.281564 Objective Loss 0.281564 LR 0.001000 Time 0.021182 -2022-12-06 10:42:22,357 - Epoch: [45][ 490/ 1200] Overall Loss 0.281570 Objective Loss 0.281570 LR 0.001000 Time 0.021151 -2022-12-06 10:42:22,557 - Epoch: [45][ 500/ 1200] Overall Loss 0.281777 Objective Loss 0.281777 LR 0.001000 Time 0.021127 -2022-12-06 10:42:22,755 - Epoch: [45][ 510/ 1200] Overall Loss 0.281281 Objective Loss 0.281281 LR 0.001000 Time 0.021099 -2022-12-06 10:42:22,954 - Epoch: [45][ 520/ 1200] Overall Loss 0.281796 Objective Loss 0.281796 LR 0.001000 Time 0.021076 -2022-12-06 10:42:23,151 - Epoch: [45][ 530/ 1200] Overall Loss 0.281915 Objective Loss 0.281915 LR 0.001000 Time 0.021049 -2022-12-06 10:42:23,352 - Epoch: [45][ 540/ 1200] Overall Loss 0.282188 Objective Loss 0.282188 LR 0.001000 Time 0.021030 -2022-12-06 10:42:23,549 - Epoch: [45][ 550/ 1200] Overall Loss 0.282037 Objective Loss 0.282037 LR 0.001000 Time 0.021005 -2022-12-06 10:42:23,749 - Epoch: [45][ 560/ 1200] Overall Loss 0.282208 Objective Loss 0.282208 LR 0.001000 Time 0.020986 -2022-12-06 10:42:23,946 - Epoch: [45][ 570/ 1200] Overall Loss 0.281926 Objective Loss 0.281926 LR 0.001000 Time 0.020963 -2022-12-06 10:42:24,146 - Epoch: [45][ 580/ 1200] Overall Loss 0.282012 Objective Loss 0.282012 LR 0.001000 Time 0.020945 -2022-12-06 10:42:24,343 - Epoch: [45][ 590/ 1200] Overall Loss 0.281839 Objective Loss 0.281839 LR 0.001000 Time 0.020923 -2022-12-06 10:42:24,543 - Epoch: [45][ 600/ 1200] Overall Loss 0.281962 Objective Loss 0.281962 LR 0.001000 Time 0.020906 -2022-12-06 10:42:24,739 - Epoch: [45][ 610/ 1200] Overall Loss 0.281591 Objective Loss 0.281591 LR 0.001000 Time 0.020884 -2022-12-06 10:42:24,939 - Epoch: [45][ 620/ 1200] Overall Loss 0.281201 Objective Loss 0.281201 LR 0.001000 Time 0.020869 -2022-12-06 10:42:25,136 - Epoch: [45][ 630/ 1200] Overall Loss 0.280834 Objective Loss 0.280834 LR 0.001000 Time 0.020849 -2022-12-06 10:42:25,335 - Epoch: [45][ 640/ 1200] Overall Loss 0.281174 Objective Loss 0.281174 LR 0.001000 Time 0.020834 -2022-12-06 10:42:25,532 - Epoch: [45][ 650/ 1200] Overall Loss 0.281403 Objective Loss 0.281403 LR 0.001000 Time 0.020815 -2022-12-06 10:42:25,732 - Epoch: [45][ 660/ 1200] Overall Loss 0.281773 Objective Loss 0.281773 LR 0.001000 Time 0.020802 -2022-12-06 10:42:25,928 - Epoch: [45][ 670/ 1200] Overall Loss 0.281891 Objective Loss 0.281891 LR 0.001000 Time 0.020784 -2022-12-06 10:42:26,128 - Epoch: [45][ 680/ 1200] Overall Loss 0.281888 Objective Loss 0.281888 LR 0.001000 Time 0.020771 -2022-12-06 10:42:26,325 - Epoch: [45][ 690/ 1200] Overall Loss 0.281570 Objective Loss 0.281570 LR 0.001000 Time 0.020755 -2022-12-06 10:42:26,524 - Epoch: [45][ 700/ 1200] Overall Loss 0.281636 Objective Loss 0.281636 LR 0.001000 Time 0.020743 -2022-12-06 10:42:26,722 - Epoch: [45][ 710/ 1200] Overall Loss 0.281891 Objective Loss 0.281891 LR 0.001000 Time 0.020728 -2022-12-06 10:42:26,921 - Epoch: [45][ 720/ 1200] Overall Loss 0.282116 Objective Loss 0.282116 LR 0.001000 Time 0.020716 -2022-12-06 10:42:27,118 - Epoch: [45][ 730/ 1200] Overall Loss 0.282152 Objective Loss 0.282152 LR 0.001000 Time 0.020702 -2022-12-06 10:42:27,318 - Epoch: [45][ 740/ 1200] Overall Loss 0.282066 Objective Loss 0.282066 LR 0.001000 Time 0.020691 -2022-12-06 10:42:27,515 - Epoch: [45][ 750/ 1200] Overall Loss 0.282349 Objective Loss 0.282349 LR 0.001000 Time 0.020677 -2022-12-06 10:42:27,715 - Epoch: [45][ 760/ 1200] Overall Loss 0.281924 Objective Loss 0.281924 LR 0.001000 Time 0.020667 -2022-12-06 10:42:27,911 - Epoch: [45][ 770/ 1200] Overall Loss 0.281577 Objective Loss 0.281577 LR 0.001000 Time 0.020653 -2022-12-06 10:42:28,111 - Epoch: [45][ 780/ 1200] Overall Loss 0.281157 Objective Loss 0.281157 LR 0.001000 Time 0.020643 -2022-12-06 10:42:28,307 - Epoch: [45][ 790/ 1200] Overall Loss 0.281048 Objective Loss 0.281048 LR 0.001000 Time 0.020630 -2022-12-06 10:42:28,507 - Epoch: [45][ 800/ 1200] Overall Loss 0.280844 Objective Loss 0.280844 LR 0.001000 Time 0.020621 -2022-12-06 10:42:28,704 - Epoch: [45][ 810/ 1200] Overall Loss 0.280991 Objective Loss 0.280991 LR 0.001000 Time 0.020609 -2022-12-06 10:42:28,904 - Epoch: [45][ 820/ 1200] Overall Loss 0.281211 Objective Loss 0.281211 LR 0.001000 Time 0.020601 -2022-12-06 10:42:29,101 - Epoch: [45][ 830/ 1200] Overall Loss 0.281062 Objective Loss 0.281062 LR 0.001000 Time 0.020590 -2022-12-06 10:42:29,301 - Epoch: [45][ 840/ 1200] Overall Loss 0.280696 Objective Loss 0.280696 LR 0.001000 Time 0.020582 -2022-12-06 10:42:29,498 - Epoch: [45][ 850/ 1200] Overall Loss 0.280444 Objective Loss 0.280444 LR 0.001000 Time 0.020571 -2022-12-06 10:42:29,697 - Epoch: [45][ 860/ 1200] Overall Loss 0.280016 Objective Loss 0.280016 LR 0.001000 Time 0.020563 -2022-12-06 10:42:29,894 - Epoch: [45][ 870/ 1200] Overall Loss 0.279929 Objective Loss 0.279929 LR 0.001000 Time 0.020552 -2022-12-06 10:42:30,094 - Epoch: [45][ 880/ 1200] Overall Loss 0.280328 Objective Loss 0.280328 LR 0.001000 Time 0.020545 -2022-12-06 10:42:30,291 - Epoch: [45][ 890/ 1200] Overall Loss 0.280391 Objective Loss 0.280391 LR 0.001000 Time 0.020535 -2022-12-06 10:42:30,491 - Epoch: [45][ 900/ 1200] Overall Loss 0.280109 Objective Loss 0.280109 LR 0.001000 Time 0.020528 -2022-12-06 10:42:30,688 - Epoch: [45][ 910/ 1200] Overall Loss 0.280231 Objective Loss 0.280231 LR 0.001000 Time 0.020518 -2022-12-06 10:42:30,888 - Epoch: [45][ 920/ 1200] Overall Loss 0.280720 Objective Loss 0.280720 LR 0.001000 Time 0.020513 -2022-12-06 10:42:31,085 - Epoch: [45][ 930/ 1200] Overall Loss 0.281059 Objective Loss 0.281059 LR 0.001000 Time 0.020503 -2022-12-06 10:42:31,285 - Epoch: [45][ 940/ 1200] Overall Loss 0.281487 Objective Loss 0.281487 LR 0.001000 Time 0.020497 -2022-12-06 10:42:31,482 - Epoch: [45][ 950/ 1200] Overall Loss 0.282122 Objective Loss 0.282122 LR 0.001000 Time 0.020488 -2022-12-06 10:42:31,682 - Epoch: [45][ 960/ 1200] Overall Loss 0.282416 Objective Loss 0.282416 LR 0.001000 Time 0.020482 -2022-12-06 10:42:31,879 - Epoch: [45][ 970/ 1200] Overall Loss 0.282912 Objective Loss 0.282912 LR 0.001000 Time 0.020474 -2022-12-06 10:42:32,079 - Epoch: [45][ 980/ 1200] Overall Loss 0.282679 Objective Loss 0.282679 LR 0.001000 Time 0.020468 -2022-12-06 10:42:32,275 - Epoch: [45][ 990/ 1200] Overall Loss 0.282766 Objective Loss 0.282766 LR 0.001000 Time 0.020460 -2022-12-06 10:42:32,475 - Epoch: [45][ 1000/ 1200] Overall Loss 0.282870 Objective Loss 0.282870 LR 0.001000 Time 0.020454 -2022-12-06 10:42:32,672 - Epoch: [45][ 1010/ 1200] Overall Loss 0.283327 Objective Loss 0.283327 LR 0.001000 Time 0.020445 -2022-12-06 10:42:32,872 - Epoch: [45][ 1020/ 1200] Overall Loss 0.283389 Objective Loss 0.283389 LR 0.001000 Time 0.020441 -2022-12-06 10:42:33,068 - Epoch: [45][ 1030/ 1200] Overall Loss 0.283568 Objective Loss 0.283568 LR 0.001000 Time 0.020432 -2022-12-06 10:42:33,268 - Epoch: [45][ 1040/ 1200] Overall Loss 0.283866 Objective Loss 0.283866 LR 0.001000 Time 0.020428 -2022-12-06 10:42:33,465 - Epoch: [45][ 1050/ 1200] Overall Loss 0.283972 Objective Loss 0.283972 LR 0.001000 Time 0.020420 -2022-12-06 10:42:33,665 - Epoch: [45][ 1060/ 1200] Overall Loss 0.283865 Objective Loss 0.283865 LR 0.001000 Time 0.020416 -2022-12-06 10:42:33,862 - Epoch: [45][ 1070/ 1200] Overall Loss 0.283920 Objective Loss 0.283920 LR 0.001000 Time 0.020408 -2022-12-06 10:42:34,062 - Epoch: [45][ 1080/ 1200] Overall Loss 0.284105 Objective Loss 0.284105 LR 0.001000 Time 0.020404 -2022-12-06 10:42:34,259 - Epoch: [45][ 1090/ 1200] Overall Loss 0.284346 Objective Loss 0.284346 LR 0.001000 Time 0.020397 -2022-12-06 10:42:34,458 - Epoch: [45][ 1100/ 1200] Overall Loss 0.284044 Objective Loss 0.284044 LR 0.001000 Time 0.020392 -2022-12-06 10:42:34,653 - Epoch: [45][ 1110/ 1200] Overall Loss 0.284073 Objective Loss 0.284073 LR 0.001000 Time 0.020384 -2022-12-06 10:42:34,851 - Epoch: [45][ 1120/ 1200] Overall Loss 0.284021 Objective Loss 0.284021 LR 0.001000 Time 0.020378 -2022-12-06 10:42:35,047 - Epoch: [45][ 1130/ 1200] Overall Loss 0.283764 Objective Loss 0.283764 LR 0.001000 Time 0.020370 -2022-12-06 10:42:35,244 - Epoch: [45][ 1140/ 1200] Overall Loss 0.283906 Objective Loss 0.283906 LR 0.001000 Time 0.020364 -2022-12-06 10:42:35,440 - Epoch: [45][ 1150/ 1200] Overall Loss 0.284158 Objective Loss 0.284158 LR 0.001000 Time 0.020357 -2022-12-06 10:42:35,637 - Epoch: [45][ 1160/ 1200] Overall Loss 0.284032 Objective Loss 0.284032 LR 0.001000 Time 0.020351 -2022-12-06 10:42:35,833 - Epoch: [45][ 1170/ 1200] Overall Loss 0.283692 Objective Loss 0.283692 LR 0.001000 Time 0.020344 -2022-12-06 10:42:36,031 - Epoch: [45][ 1180/ 1200] Overall Loss 0.284044 Objective Loss 0.284044 LR 0.001000 Time 0.020339 -2022-12-06 10:42:36,227 - Epoch: [45][ 1190/ 1200] Overall Loss 0.283970 Objective Loss 0.283970 LR 0.001000 Time 0.020332 -2022-12-06 10:42:36,453 - Epoch: [45][ 1200/ 1200] Overall Loss 0.284168 Objective Loss 0.284168 Top1 86.820084 Top5 98.953975 LR 0.001000 Time 0.020351 -2022-12-06 10:42:36,542 - --- validate (epoch=45)----------- -2022-12-06 10:42:36,542 - 34129 samples (256 per mini-batch) -2022-12-06 10:42:37,008 - Epoch: [45][ 10/ 134] Loss 0.306759 Top1 84.609375 Top5 98.398438 -2022-12-06 10:42:37,157 - Epoch: [45][ 20/ 134] Loss 0.311612 Top1 84.550781 Top5 98.007812 -2022-12-06 10:42:37,298 - Epoch: [45][ 30/ 134] Loss 0.313466 Top1 84.713542 Top5 97.916667 -2022-12-06 10:42:37,440 - Epoch: [45][ 40/ 134] Loss 0.321639 Top1 84.306641 Top5 97.841797 -2022-12-06 10:42:37,585 - Epoch: [45][ 50/ 134] Loss 0.315569 Top1 84.304688 Top5 97.765625 -2022-12-06 10:42:37,735 - Epoch: [45][ 60/ 134] Loss 0.313378 Top1 84.420573 Top5 97.792969 -2022-12-06 10:42:37,877 - Epoch: [45][ 70/ 134] Loss 0.315723 Top1 84.151786 Top5 97.650670 -2022-12-06 10:42:38,024 - Epoch: [45][ 80/ 134] Loss 0.314255 Top1 84.130859 Top5 97.661133 -2022-12-06 10:42:38,166 - Epoch: [45][ 90/ 134] Loss 0.314973 Top1 84.027778 Top5 97.625868 -2022-12-06 10:42:38,315 - Epoch: [45][ 100/ 134] Loss 0.315278 Top1 84.031250 Top5 97.640625 -2022-12-06 10:42:38,458 - Epoch: [45][ 110/ 134] Loss 0.313100 Top1 84.002131 Top5 97.663352 -2022-12-06 10:42:38,608 - Epoch: [45][ 120/ 134] Loss 0.310164 Top1 84.013672 Top5 97.698568 -2022-12-06 10:42:38,747 - Epoch: [45][ 130/ 134] Loss 0.310404 Top1 83.999399 Top5 97.695312 -2022-12-06 10:42:38,785 - Epoch: [45][ 134/ 134] Loss 0.310284 Top1 83.975505 Top5 97.699903 -2022-12-06 10:42:38,875 - ==> Top1: 83.976 Top5: 97.700 Loss: 0.310 - -2022-12-06 10:42:38,875 - ==> Confusion: -[[ 902 2 1 0 11 8 1 1 6 41 0 2 1 0 5 4 1 5 2 0 3] - [ 2 925 5 2 11 30 3 15 0 0 5 3 1 1 0 0 5 2 12 0 5] - [ 9 2 1000 17 4 2 22 12 0 1 2 4 1 0 2 3 3 1 8 1 9] - [ 2 2 21 930 0 8 1 2 1 1 6 0 1 0 21 0 1 4 13 2 4] - [ 11 4 1 2 945 9 1 2 1 7 1 3 0 4 9 3 7 3 1 2 4] - [ 3 18 1 1 7 964 5 23 3 2 0 11 5 10 4 1 2 0 1 5 3] - [ 0 1 22 2 0 3 1059 5 0 0 2 2 0 2 0 4 3 1 1 9 2] - [ 1 11 7 2 2 25 6 950 0 0 1 4 0 0 0 2 1 1 23 14 4] - [ 6 5 1 0 0 2 0 0 952 63 5 2 0 6 10 1 3 1 3 1 3] - [ 87 0 1 0 0 5 0 3 13 868 1 1 0 8 0 1 2 1 1 1 8] - [ 1 2 3 13 0 2 1 1 9 2 938 2 0 11 4 1 1 0 17 4 7] - [ 5 1 4 0 0 13 4 5 1 1 0 959 16 5 1 4 6 5 2 18 1] - [ 3 0 3 7 2 4 4 1 1 0 0 55 827 1 5 13 3 18 3 9 10] - [ 1 2 1 0 0 19 0 2 11 18 12 10 2 927 0 1 5 2 1 4 5] - [ 9 4 2 15 4 2 0 2 25 8 0 1 0 4 1026 1 2 1 11 2 11] - [ 2 0 4 2 3 3 5 0 0 0 1 11 4 1 0 980 14 7 0 4 2] - [ 2 4 1 0 3 3 2 0 1 0 0 4 1 3 4 6 1030 2 2 1 3] - [ 7 0 1 3 0 0 4 2 1 1 1 15 4 1 1 13 3 975 0 3 1] - [ 5 6 5 9 0 4 1 28 2 1 6 1 3 1 5 0 1 0 922 6 2] - [ 2 2 3 0 0 7 7 8 0 0 0 12 2 3 0 7 7 0 0 1015 5] - [ 136 282 227 159 129 232 79 201 101 113 131 121 301 330 152 119 253 97 213 286 9564]] - -2022-12-06 10:42:39,545 - ==> Best [Top1: 84.717 Top5: 97.817 Sparsity:0.00 Params: 5376 on epoch: 43] -2022-12-06 10:42:39,546 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:42:39,551 - - -2022-12-06 10:42:39,552 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:42:40,487 - Epoch: [46][ 10/ 1200] Overall Loss 0.323074 Objective Loss 0.323074 LR 0.001000 Time 0.093464 -2022-12-06 10:42:40,687 - Epoch: [46][ 20/ 1200] Overall Loss 0.295242 Objective Loss 0.295242 LR 0.001000 Time 0.056697 -2022-12-06 10:42:40,878 - Epoch: [46][ 30/ 1200] Overall Loss 0.286211 Objective Loss 0.286211 LR 0.001000 Time 0.044136 -2022-12-06 10:42:41,068 - Epoch: [46][ 40/ 1200] Overall Loss 0.277589 Objective Loss 0.277589 LR 0.001000 Time 0.037842 -2022-12-06 10:42:41,258 - Epoch: [46][ 50/ 1200] Overall Loss 0.278772 Objective Loss 0.278772 LR 0.001000 Time 0.034063 -2022-12-06 10:42:41,448 - Epoch: [46][ 60/ 1200] Overall Loss 0.274723 Objective Loss 0.274723 LR 0.001000 Time 0.031541 -2022-12-06 10:42:41,638 - Epoch: [46][ 70/ 1200] Overall Loss 0.275030 Objective Loss 0.275030 LR 0.001000 Time 0.029743 -2022-12-06 10:42:41,828 - Epoch: [46][ 80/ 1200] Overall Loss 0.272553 Objective Loss 0.272553 LR 0.001000 Time 0.028394 -2022-12-06 10:42:42,018 - Epoch: [46][ 90/ 1200] Overall Loss 0.273544 Objective Loss 0.273544 LR 0.001000 Time 0.027348 -2022-12-06 10:42:42,209 - Epoch: [46][ 100/ 1200] Overall Loss 0.274882 Objective Loss 0.274882 LR 0.001000 Time 0.026517 -2022-12-06 10:42:42,400 - Epoch: [46][ 110/ 1200] Overall Loss 0.276572 Objective Loss 0.276572 LR 0.001000 Time 0.025839 -2022-12-06 10:42:42,591 - Epoch: [46][ 120/ 1200] Overall Loss 0.273270 Objective Loss 0.273270 LR 0.001000 Time 0.025271 -2022-12-06 10:42:42,782 - Epoch: [46][ 130/ 1200] Overall Loss 0.273103 Objective Loss 0.273103 LR 0.001000 Time 0.024791 -2022-12-06 10:42:42,973 - Epoch: [46][ 140/ 1200] Overall Loss 0.274330 Objective Loss 0.274330 LR 0.001000 Time 0.024380 -2022-12-06 10:42:43,164 - Epoch: [46][ 150/ 1200] Overall Loss 0.274393 Objective Loss 0.274393 LR 0.001000 Time 0.024024 -2022-12-06 10:42:43,354 - Epoch: [46][ 160/ 1200] Overall Loss 0.276138 Objective Loss 0.276138 LR 0.001000 Time 0.023707 -2022-12-06 10:42:43,544 - Epoch: [46][ 170/ 1200] Overall Loss 0.276688 Objective Loss 0.276688 LR 0.001000 Time 0.023430 -2022-12-06 10:42:43,735 - Epoch: [46][ 180/ 1200] Overall Loss 0.277972 Objective Loss 0.277972 LR 0.001000 Time 0.023184 -2022-12-06 10:42:43,925 - Epoch: [46][ 190/ 1200] Overall Loss 0.276615 Objective Loss 0.276615 LR 0.001000 Time 0.022962 -2022-12-06 10:42:44,116 - Epoch: [46][ 200/ 1200] Overall Loss 0.277157 Objective Loss 0.277157 LR 0.001000 Time 0.022765 -2022-12-06 10:42:44,307 - Epoch: [46][ 210/ 1200] Overall Loss 0.277206 Objective Loss 0.277206 LR 0.001000 Time 0.022587 -2022-12-06 10:42:44,497 - Epoch: [46][ 220/ 1200] Overall Loss 0.276890 Objective Loss 0.276890 LR 0.001000 Time 0.022421 -2022-12-06 10:42:44,688 - Epoch: [46][ 230/ 1200] Overall Loss 0.277620 Objective Loss 0.277620 LR 0.001000 Time 0.022274 -2022-12-06 10:42:44,878 - Epoch: [46][ 240/ 1200] Overall Loss 0.278278 Objective Loss 0.278278 LR 0.001000 Time 0.022139 -2022-12-06 10:42:45,069 - Epoch: [46][ 250/ 1200] Overall Loss 0.278930 Objective Loss 0.278930 LR 0.001000 Time 0.022014 -2022-12-06 10:42:45,259 - Epoch: [46][ 260/ 1200] Overall Loss 0.278563 Objective Loss 0.278563 LR 0.001000 Time 0.021896 -2022-12-06 10:42:45,449 - Epoch: [46][ 270/ 1200] Overall Loss 0.280162 Objective Loss 0.280162 LR 0.001000 Time 0.021787 -2022-12-06 10:42:45,640 - Epoch: [46][ 280/ 1200] Overall Loss 0.280615 Objective Loss 0.280615 LR 0.001000 Time 0.021687 -2022-12-06 10:42:45,830 - Epoch: [46][ 290/ 1200] Overall Loss 0.281534 Objective Loss 0.281534 LR 0.001000 Time 0.021595 -2022-12-06 10:42:46,021 - Epoch: [46][ 300/ 1200] Overall Loss 0.281186 Objective Loss 0.281186 LR 0.001000 Time 0.021510 -2022-12-06 10:42:46,212 - Epoch: [46][ 310/ 1200] Overall Loss 0.279689 Objective Loss 0.279689 LR 0.001000 Time 0.021429 -2022-12-06 10:42:46,402 - Epoch: [46][ 320/ 1200] Overall Loss 0.280035 Objective Loss 0.280035 LR 0.001000 Time 0.021353 -2022-12-06 10:42:46,593 - Epoch: [46][ 330/ 1200] Overall Loss 0.280286 Objective Loss 0.280286 LR 0.001000 Time 0.021282 -2022-12-06 10:42:46,783 - Epoch: [46][ 340/ 1200] Overall Loss 0.279847 Objective Loss 0.279847 LR 0.001000 Time 0.021215 -2022-12-06 10:42:46,974 - Epoch: [46][ 350/ 1200] Overall Loss 0.279411 Objective Loss 0.279411 LR 0.001000 Time 0.021151 -2022-12-06 10:42:47,164 - Epoch: [46][ 360/ 1200] Overall Loss 0.279283 Objective Loss 0.279283 LR 0.001000 Time 0.021089 -2022-12-06 10:42:47,354 - Epoch: [46][ 370/ 1200] Overall Loss 0.278686 Objective Loss 0.278686 LR 0.001000 Time 0.021033 -2022-12-06 10:42:47,544 - Epoch: [46][ 380/ 1200] Overall Loss 0.278583 Objective Loss 0.278583 LR 0.001000 Time 0.020976 -2022-12-06 10:42:47,734 - Epoch: [46][ 390/ 1200] Overall Loss 0.278300 Objective Loss 0.278300 LR 0.001000 Time 0.020926 -2022-12-06 10:42:47,925 - Epoch: [46][ 400/ 1200] Overall Loss 0.279031 Objective Loss 0.279031 LR 0.001000 Time 0.020878 -2022-12-06 10:42:48,116 - Epoch: [46][ 410/ 1200] Overall Loss 0.279351 Objective Loss 0.279351 LR 0.001000 Time 0.020833 -2022-12-06 10:42:48,307 - Epoch: [46][ 420/ 1200] Overall Loss 0.278801 Objective Loss 0.278801 LR 0.001000 Time 0.020790 -2022-12-06 10:42:48,497 - Epoch: [46][ 430/ 1200] Overall Loss 0.278376 Objective Loss 0.278376 LR 0.001000 Time 0.020747 -2022-12-06 10:42:48,687 - Epoch: [46][ 440/ 1200] Overall Loss 0.277931 Objective Loss 0.277931 LR 0.001000 Time 0.020707 -2022-12-06 10:42:48,878 - Epoch: [46][ 450/ 1200] Overall Loss 0.278373 Objective Loss 0.278373 LR 0.001000 Time 0.020669 -2022-12-06 10:42:49,068 - Epoch: [46][ 460/ 1200] Overall Loss 0.277888 Objective Loss 0.277888 LR 0.001000 Time 0.020633 -2022-12-06 10:42:49,259 - Epoch: [46][ 470/ 1200] Overall Loss 0.278162 Objective Loss 0.278162 LR 0.001000 Time 0.020599 -2022-12-06 10:42:49,450 - Epoch: [46][ 480/ 1200] Overall Loss 0.278696 Objective Loss 0.278696 LR 0.001000 Time 0.020566 -2022-12-06 10:42:49,642 - Epoch: [46][ 490/ 1200] Overall Loss 0.279169 Objective Loss 0.279169 LR 0.001000 Time 0.020536 -2022-12-06 10:42:49,832 - Epoch: [46][ 500/ 1200] Overall Loss 0.278913 Objective Loss 0.278913 LR 0.001000 Time 0.020505 -2022-12-06 10:42:50,022 - Epoch: [46][ 510/ 1200] Overall Loss 0.279311 Objective Loss 0.279311 LR 0.001000 Time 0.020475 -2022-12-06 10:42:50,213 - Epoch: [46][ 520/ 1200] Overall Loss 0.279631 Objective Loss 0.279631 LR 0.001000 Time 0.020448 -2022-12-06 10:42:50,404 - Epoch: [46][ 530/ 1200] Overall Loss 0.279718 Objective Loss 0.279718 LR 0.001000 Time 0.020421 -2022-12-06 10:42:50,595 - Epoch: [46][ 540/ 1200] Overall Loss 0.279288 Objective Loss 0.279288 LR 0.001000 Time 0.020396 -2022-12-06 10:42:50,786 - Epoch: [46][ 550/ 1200] Overall Loss 0.279777 Objective Loss 0.279777 LR 0.001000 Time 0.020370 -2022-12-06 10:42:50,976 - Epoch: [46][ 560/ 1200] Overall Loss 0.280059 Objective Loss 0.280059 LR 0.001000 Time 0.020345 -2022-12-06 10:42:51,166 - Epoch: [46][ 570/ 1200] Overall Loss 0.280032 Objective Loss 0.280032 LR 0.001000 Time 0.020322 -2022-12-06 10:42:51,357 - Epoch: [46][ 580/ 1200] Overall Loss 0.280129 Objective Loss 0.280129 LR 0.001000 Time 0.020300 -2022-12-06 10:42:51,548 - Epoch: [46][ 590/ 1200] Overall Loss 0.280381 Objective Loss 0.280381 LR 0.001000 Time 0.020278 -2022-12-06 10:42:51,739 - Epoch: [46][ 600/ 1200] Overall Loss 0.280519 Objective Loss 0.280519 LR 0.001000 Time 0.020257 -2022-12-06 10:42:51,930 - Epoch: [46][ 610/ 1200] Overall Loss 0.280720 Objective Loss 0.280720 LR 0.001000 Time 0.020238 -2022-12-06 10:42:52,121 - Epoch: [46][ 620/ 1200] Overall Loss 0.280067 Objective Loss 0.280067 LR 0.001000 Time 0.020218 -2022-12-06 10:42:52,312 - Epoch: [46][ 630/ 1200] Overall Loss 0.279489 Objective Loss 0.279489 LR 0.001000 Time 0.020200 -2022-12-06 10:42:52,503 - Epoch: [46][ 640/ 1200] Overall Loss 0.280036 Objective Loss 0.280036 LR 0.001000 Time 0.020181 -2022-12-06 10:42:52,693 - Epoch: [46][ 650/ 1200] Overall Loss 0.280326 Objective Loss 0.280326 LR 0.001000 Time 0.020162 -2022-12-06 10:42:52,883 - Epoch: [46][ 660/ 1200] Overall Loss 0.280284 Objective Loss 0.280284 LR 0.001000 Time 0.020144 -2022-12-06 10:42:53,074 - Epoch: [46][ 670/ 1200] Overall Loss 0.280790 Objective Loss 0.280790 LR 0.001000 Time 0.020127 -2022-12-06 10:42:53,264 - Epoch: [46][ 680/ 1200] Overall Loss 0.281302 Objective Loss 0.281302 LR 0.001000 Time 0.020110 -2022-12-06 10:42:53,455 - Epoch: [46][ 690/ 1200] Overall Loss 0.281408 Objective Loss 0.281408 LR 0.001000 Time 0.020094 -2022-12-06 10:42:53,645 - Epoch: [46][ 700/ 1200] Overall Loss 0.281437 Objective Loss 0.281437 LR 0.001000 Time 0.020078 -2022-12-06 10:42:53,835 - Epoch: [46][ 710/ 1200] Overall Loss 0.282085 Objective Loss 0.282085 LR 0.001000 Time 0.020062 -2022-12-06 10:42:54,026 - Epoch: [46][ 720/ 1200] Overall Loss 0.281786 Objective Loss 0.281786 LR 0.001000 Time 0.020048 -2022-12-06 10:42:54,216 - Epoch: [46][ 730/ 1200] Overall Loss 0.281758 Objective Loss 0.281758 LR 0.001000 Time 0.020033 -2022-12-06 10:42:54,407 - Epoch: [46][ 740/ 1200] Overall Loss 0.282140 Objective Loss 0.282140 LR 0.001000 Time 0.020019 -2022-12-06 10:42:54,597 - Epoch: [46][ 750/ 1200] Overall Loss 0.282083 Objective Loss 0.282083 LR 0.001000 Time 0.020005 -2022-12-06 10:42:54,788 - Epoch: [46][ 760/ 1200] Overall Loss 0.281984 Objective Loss 0.281984 LR 0.001000 Time 0.019992 -2022-12-06 10:42:54,978 - Epoch: [46][ 770/ 1200] Overall Loss 0.281763 Objective Loss 0.281763 LR 0.001000 Time 0.019980 -2022-12-06 10:42:55,168 - Epoch: [46][ 780/ 1200] Overall Loss 0.281599 Objective Loss 0.281599 LR 0.001000 Time 0.019966 -2022-12-06 10:42:55,359 - Epoch: [46][ 790/ 1200] Overall Loss 0.281613 Objective Loss 0.281613 LR 0.001000 Time 0.019954 -2022-12-06 10:42:55,549 - Epoch: [46][ 800/ 1200] Overall Loss 0.281474 Objective Loss 0.281474 LR 0.001000 Time 0.019942 -2022-12-06 10:42:55,740 - Epoch: [46][ 810/ 1200] Overall Loss 0.281455 Objective Loss 0.281455 LR 0.001000 Time 0.019930 -2022-12-06 10:42:55,931 - Epoch: [46][ 820/ 1200] Overall Loss 0.281497 Objective Loss 0.281497 LR 0.001000 Time 0.019920 -2022-12-06 10:42:56,122 - Epoch: [46][ 830/ 1200] Overall Loss 0.281207 Objective Loss 0.281207 LR 0.001000 Time 0.019909 -2022-12-06 10:42:56,313 - Epoch: [46][ 840/ 1200] Overall Loss 0.280963 Objective Loss 0.280963 LR 0.001000 Time 0.019899 -2022-12-06 10:42:56,504 - Epoch: [46][ 850/ 1200] Overall Loss 0.280786 Objective Loss 0.280786 LR 0.001000 Time 0.019888 -2022-12-06 10:42:56,694 - Epoch: [46][ 860/ 1200] Overall Loss 0.280823 Objective Loss 0.280823 LR 0.001000 Time 0.019878 -2022-12-06 10:42:56,885 - Epoch: [46][ 870/ 1200] Overall Loss 0.280822 Objective Loss 0.280822 LR 0.001000 Time 0.019868 -2022-12-06 10:42:57,076 - Epoch: [46][ 880/ 1200] Overall Loss 0.281138 Objective Loss 0.281138 LR 0.001000 Time 0.019858 -2022-12-06 10:42:57,267 - Epoch: [46][ 890/ 1200] Overall Loss 0.281182 Objective Loss 0.281182 LR 0.001000 Time 0.019849 -2022-12-06 10:42:57,458 - Epoch: [46][ 900/ 1200] Overall Loss 0.281122 Objective Loss 0.281122 LR 0.001000 Time 0.019840 -2022-12-06 10:42:57,649 - Epoch: [46][ 910/ 1200] Overall Loss 0.280986 Objective Loss 0.280986 LR 0.001000 Time 0.019832 -2022-12-06 10:42:57,840 - Epoch: [46][ 920/ 1200] Overall Loss 0.280931 Objective Loss 0.280931 LR 0.001000 Time 0.019823 -2022-12-06 10:42:58,031 - Epoch: [46][ 930/ 1200] Overall Loss 0.280969 Objective Loss 0.280969 LR 0.001000 Time 0.019815 -2022-12-06 10:42:58,222 - Epoch: [46][ 940/ 1200] Overall Loss 0.280896 Objective Loss 0.280896 LR 0.001000 Time 0.019807 -2022-12-06 10:42:58,413 - Epoch: [46][ 950/ 1200] Overall Loss 0.281089 Objective Loss 0.281089 LR 0.001000 Time 0.019798 -2022-12-06 10:42:58,603 - Epoch: [46][ 960/ 1200] Overall Loss 0.281099 Objective Loss 0.281099 LR 0.001000 Time 0.019790 -2022-12-06 10:42:58,794 - Epoch: [46][ 970/ 1200] Overall Loss 0.281051 Objective Loss 0.281051 LR 0.001000 Time 0.019782 -2022-12-06 10:42:58,985 - Epoch: [46][ 980/ 1200] Overall Loss 0.280982 Objective Loss 0.280982 LR 0.001000 Time 0.019774 -2022-12-06 10:42:59,176 - Epoch: [46][ 990/ 1200] Overall Loss 0.281040 Objective Loss 0.281040 LR 0.001000 Time 0.019767 -2022-12-06 10:42:59,367 - Epoch: [46][ 1000/ 1200] Overall Loss 0.281180 Objective Loss 0.281180 LR 0.001000 Time 0.019760 -2022-12-06 10:42:59,557 - Epoch: [46][ 1010/ 1200] Overall Loss 0.281011 Objective Loss 0.281011 LR 0.001000 Time 0.019752 -2022-12-06 10:42:59,748 - Epoch: [46][ 1020/ 1200] Overall Loss 0.281031 Objective Loss 0.281031 LR 0.001000 Time 0.019745 -2022-12-06 10:42:59,939 - Epoch: [46][ 1030/ 1200] Overall Loss 0.281353 Objective Loss 0.281353 LR 0.001000 Time 0.019738 -2022-12-06 10:43:00,130 - Epoch: [46][ 1040/ 1200] Overall Loss 0.281576 Objective Loss 0.281576 LR 0.001000 Time 0.019732 -2022-12-06 10:43:00,321 - Epoch: [46][ 1050/ 1200] Overall Loss 0.281669 Objective Loss 0.281669 LR 0.001000 Time 0.019725 -2022-12-06 10:43:00,512 - Epoch: [46][ 1060/ 1200] Overall Loss 0.281782 Objective Loss 0.281782 LR 0.001000 Time 0.019718 -2022-12-06 10:43:00,702 - Epoch: [46][ 1070/ 1200] Overall Loss 0.281875 Objective Loss 0.281875 LR 0.001000 Time 0.019711 -2022-12-06 10:43:00,893 - Epoch: [46][ 1080/ 1200] Overall Loss 0.281739 Objective Loss 0.281739 LR 0.001000 Time 0.019705 -2022-12-06 10:43:01,084 - Epoch: [46][ 1090/ 1200] Overall Loss 0.282028 Objective Loss 0.282028 LR 0.001000 Time 0.019699 -2022-12-06 10:43:01,275 - Epoch: [46][ 1100/ 1200] Overall Loss 0.282140 Objective Loss 0.282140 LR 0.001000 Time 0.019693 -2022-12-06 10:43:01,466 - Epoch: [46][ 1110/ 1200] Overall Loss 0.282632 Objective Loss 0.282632 LR 0.001000 Time 0.019688 -2022-12-06 10:43:01,657 - Epoch: [46][ 1120/ 1200] Overall Loss 0.282721 Objective Loss 0.282721 LR 0.001000 Time 0.019682 -2022-12-06 10:43:01,847 - Epoch: [46][ 1130/ 1200] Overall Loss 0.282816 Objective Loss 0.282816 LR 0.001000 Time 0.019675 -2022-12-06 10:43:02,038 - Epoch: [46][ 1140/ 1200] Overall Loss 0.282862 Objective Loss 0.282862 LR 0.001000 Time 0.019669 -2022-12-06 10:43:02,228 - Epoch: [46][ 1150/ 1200] Overall Loss 0.282974 Objective Loss 0.282974 LR 0.001000 Time 0.019663 -2022-12-06 10:43:02,419 - Epoch: [46][ 1160/ 1200] Overall Loss 0.283100 Objective Loss 0.283100 LR 0.001000 Time 0.019658 -2022-12-06 10:43:02,610 - Epoch: [46][ 1170/ 1200] Overall Loss 0.282886 Objective Loss 0.282886 LR 0.001000 Time 0.019653 -2022-12-06 10:43:02,801 - Epoch: [46][ 1180/ 1200] Overall Loss 0.282908 Objective Loss 0.282908 LR 0.001000 Time 0.019647 -2022-12-06 10:43:02,991 - Epoch: [46][ 1190/ 1200] Overall Loss 0.282852 Objective Loss 0.282852 LR 0.001000 Time 0.019642 -2022-12-06 10:43:03,214 - Epoch: [46][ 1200/ 1200] Overall Loss 0.282789 Objective Loss 0.282789 Top1 83.682008 Top5 98.326360 LR 0.001000 Time 0.019663 -2022-12-06 10:43:03,303 - --- validate (epoch=46)----------- -2022-12-06 10:43:03,303 - 34129 samples (256 per mini-batch) -2022-12-06 10:43:03,747 - Epoch: [46][ 10/ 134] Loss 0.322388 Top1 83.945312 Top5 97.539062 -2022-12-06 10:43:03,892 - Epoch: [46][ 20/ 134] Loss 0.310318 Top1 83.808594 Top5 97.734375 -2022-12-06 10:43:04,039 - Epoch: [46][ 30/ 134] Loss 0.296944 Top1 84.049479 Top5 97.656250 -2022-12-06 10:43:04,181 - Epoch: [46][ 40/ 134] Loss 0.296677 Top1 83.916016 Top5 97.685547 -2022-12-06 10:43:04,310 - Epoch: [46][ 50/ 134] Loss 0.302962 Top1 83.914062 Top5 97.734375 -2022-12-06 10:43:04,439 - Epoch: [46][ 60/ 134] Loss 0.300298 Top1 83.951823 Top5 97.753906 -2022-12-06 10:43:04,569 - Epoch: [46][ 70/ 134] Loss 0.303469 Top1 84.001116 Top5 97.712054 -2022-12-06 10:43:04,701 - Epoch: [46][ 80/ 134] Loss 0.303187 Top1 83.891602 Top5 97.695312 -2022-12-06 10:43:04,833 - Epoch: [46][ 90/ 134] Loss 0.300378 Top1 83.953993 Top5 97.660590 -2022-12-06 10:43:04,965 - Epoch: [46][ 100/ 134] Loss 0.295929 Top1 84.097656 Top5 97.675781 -2022-12-06 10:43:05,097 - Epoch: [46][ 110/ 134] Loss 0.300543 Top1 83.984375 Top5 97.656250 -2022-12-06 10:43:05,229 - Epoch: [46][ 120/ 134] Loss 0.302191 Top1 83.942057 Top5 97.636719 -2022-12-06 10:43:05,364 - Epoch: [46][ 130/ 134] Loss 0.300207 Top1 83.993389 Top5 97.692308 -2022-12-06 10:43:05,404 - Epoch: [46][ 134/ 134] Loss 0.300127 Top1 83.975505 Top5 97.691113 -2022-12-06 10:43:05,491 - ==> Top1: 83.976 Top5: 97.691 Loss: 0.300 - -2022-12-06 10:43:05,492 - ==> Confusion: -[[ 910 1 1 2 6 3 0 2 3 53 0 1 2 2 3 3 1 0 0 0 3] - [ 1 899 2 2 6 59 2 17 3 1 5 5 1 2 2 1 4 2 7 0 6] - [ 6 2 981 22 2 8 26 14 1 2 3 5 3 2 1 8 1 1 4 2 9] - [ 3 0 11 939 0 11 0 0 1 0 8 0 5 1 20 1 2 4 10 0 4] - [ 21 6 2 0 932 4 2 1 1 6 1 3 0 2 10 5 11 5 0 1 7] - [ 3 9 1 2 6 985 4 14 2 0 2 5 5 16 4 2 1 0 0 4 4] - [ 1 0 9 1 1 4 1071 6 0 0 5 1 2 2 0 8 1 2 0 3 1] - [ 0 8 4 1 2 43 4 948 0 0 2 5 3 2 0 3 1 1 15 6 6] - [ 7 1 0 0 1 6 0 0 972 45 8 2 1 5 10 1 1 0 2 1 1] - [ 72 0 1 0 2 5 1 1 13 881 1 3 0 12 1 1 0 2 0 0 5] - [ 1 1 2 8 1 4 0 2 11 3 947 4 4 11 3 0 3 0 7 3 4] - [ 4 1 2 0 1 10 3 5 0 0 0 942 42 7 1 6 4 5 0 16 2] - [ 2 0 0 1 1 5 2 0 1 0 0 24 901 2 0 8 2 10 1 4 5] - [ 1 1 0 0 0 11 0 2 10 18 4 4 4 953 1 3 2 1 0 2 6] - [ 10 3 2 19 3 0 2 2 22 4 1 1 4 1 1039 1 2 3 4 0 7] - [ 3 0 0 2 2 4 2 1 0 0 0 4 9 3 0 987 9 13 0 2 2] - [ 4 4 1 2 1 2 0 0 2 0 1 3 3 1 2 7 1030 3 0 3 3] - [ 6 0 2 5 0 2 1 0 1 0 0 4 24 2 2 15 1 970 0 1 0] - [ 3 5 3 15 0 6 1 46 2 1 4 2 4 1 9 1 0 0 901 0 4] - [ 3 3 3 0 0 14 13 15 1 0 2 10 9 5 0 4 4 2 0 988 4] - [ 179 194 150 149 102 304 109 217 84 95 190 122 475 327 153 146 227 92 152 280 9479]] - -2022-12-06 10:43:06,176 - ==> Best [Top1: 84.717 Top5: 97.817 Sparsity:0.00 Params: 5376 on epoch: 43] -2022-12-06 10:43:06,176 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:43:06,182 - - -2022-12-06 10:43:06,182 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:43:07,140 - Epoch: [47][ 10/ 1200] Overall Loss 0.278336 Objective Loss 0.278336 LR 0.001000 Time 0.095665 -2022-12-06 10:43:07,356 - Epoch: [47][ 20/ 1200] Overall Loss 0.268816 Objective Loss 0.268816 LR 0.001000 Time 0.058624 -2022-12-06 10:43:07,559 - Epoch: [47][ 30/ 1200] Overall Loss 0.274558 Objective Loss 0.274558 LR 0.001000 Time 0.045826 -2022-12-06 10:43:07,764 - Epoch: [47][ 40/ 1200] Overall Loss 0.266411 Objective Loss 0.266411 LR 0.001000 Time 0.039491 -2022-12-06 10:43:07,967 - Epoch: [47][ 50/ 1200] Overall Loss 0.266803 Objective Loss 0.266803 LR 0.001000 Time 0.035626 -2022-12-06 10:43:08,172 - Epoch: [47][ 60/ 1200] Overall Loss 0.265377 Objective Loss 0.265377 LR 0.001000 Time 0.033099 -2022-12-06 10:43:08,374 - Epoch: [47][ 70/ 1200] Overall Loss 0.267203 Objective Loss 0.267203 LR 0.001000 Time 0.031243 -2022-12-06 10:43:08,580 - Epoch: [47][ 80/ 1200] Overall Loss 0.270093 Objective Loss 0.270093 LR 0.001000 Time 0.029908 -2022-12-06 10:43:08,782 - Epoch: [47][ 90/ 1200] Overall Loss 0.269537 Objective Loss 0.269537 LR 0.001000 Time 0.028829 -2022-12-06 10:43:08,988 - Epoch: [47][ 100/ 1200] Overall Loss 0.268763 Objective Loss 0.268763 LR 0.001000 Time 0.027998 -2022-12-06 10:43:09,190 - Epoch: [47][ 110/ 1200] Overall Loss 0.268784 Objective Loss 0.268784 LR 0.001000 Time 0.027286 -2022-12-06 10:43:09,396 - Epoch: [47][ 120/ 1200] Overall Loss 0.270076 Objective Loss 0.270076 LR 0.001000 Time 0.026721 -2022-12-06 10:43:09,598 - Epoch: [47][ 130/ 1200] Overall Loss 0.273659 Objective Loss 0.273659 LR 0.001000 Time 0.026215 -2022-12-06 10:43:09,803 - Epoch: [47][ 140/ 1200] Overall Loss 0.273472 Objective Loss 0.273472 LR 0.001000 Time 0.025802 -2022-12-06 10:43:10,005 - Epoch: [47][ 150/ 1200] Overall Loss 0.273886 Objective Loss 0.273886 LR 0.001000 Time 0.025425 -2022-12-06 10:43:10,211 - Epoch: [47][ 160/ 1200] Overall Loss 0.274712 Objective Loss 0.274712 LR 0.001000 Time 0.025116 -2022-12-06 10:43:10,413 - Epoch: [47][ 170/ 1200] Overall Loss 0.274721 Objective Loss 0.274721 LR 0.001000 Time 0.024825 -2022-12-06 10:43:10,619 - Epoch: [47][ 180/ 1200] Overall Loss 0.274294 Objective Loss 0.274294 LR 0.001000 Time 0.024588 -2022-12-06 10:43:10,821 - Epoch: [47][ 190/ 1200] Overall Loss 0.274395 Objective Loss 0.274395 LR 0.001000 Time 0.024354 -2022-12-06 10:43:11,027 - Epoch: [47][ 200/ 1200] Overall Loss 0.275354 Objective Loss 0.275354 LR 0.001000 Time 0.024161 -2022-12-06 10:43:11,230 - Epoch: [47][ 210/ 1200] Overall Loss 0.275220 Objective Loss 0.275220 LR 0.001000 Time 0.023975 -2022-12-06 10:43:11,436 - Epoch: [47][ 220/ 1200] Overall Loss 0.276335 Objective Loss 0.276335 LR 0.001000 Time 0.023821 -2022-12-06 10:43:11,638 - Epoch: [47][ 230/ 1200] Overall Loss 0.276446 Objective Loss 0.276446 LR 0.001000 Time 0.023660 -2022-12-06 10:43:11,844 - Epoch: [47][ 240/ 1200] Overall Loss 0.276080 Objective Loss 0.276080 LR 0.001000 Time 0.023530 -2022-12-06 10:43:12,046 - Epoch: [47][ 250/ 1200] Overall Loss 0.277641 Objective Loss 0.277641 LR 0.001000 Time 0.023395 -2022-12-06 10:43:12,252 - Epoch: [47][ 260/ 1200] Overall Loss 0.276127 Objective Loss 0.276127 LR 0.001000 Time 0.023286 -2022-12-06 10:43:12,454 - Epoch: [47][ 270/ 1200] Overall Loss 0.275163 Objective Loss 0.275163 LR 0.001000 Time 0.023169 -2022-12-06 10:43:12,656 - Epoch: [47][ 280/ 1200] Overall Loss 0.275474 Objective Loss 0.275474 LR 0.001000 Time 0.023060 -2022-12-06 10:43:12,847 - Epoch: [47][ 290/ 1200] Overall Loss 0.275957 Objective Loss 0.275957 LR 0.001000 Time 0.022922 -2022-12-06 10:43:13,039 - Epoch: [47][ 300/ 1200] Overall Loss 0.275918 Objective Loss 0.275918 LR 0.001000 Time 0.022797 -2022-12-06 10:43:13,231 - Epoch: [47][ 310/ 1200] Overall Loss 0.276207 Objective Loss 0.276207 LR 0.001000 Time 0.022677 -2022-12-06 10:43:13,422 - Epoch: [47][ 320/ 1200] Overall Loss 0.276214 Objective Loss 0.276214 LR 0.001000 Time 0.022565 -2022-12-06 10:43:13,613 - Epoch: [47][ 330/ 1200] Overall Loss 0.275980 Objective Loss 0.275980 LR 0.001000 Time 0.022458 -2022-12-06 10:43:13,806 - Epoch: [47][ 340/ 1200] Overall Loss 0.276452 Objective Loss 0.276452 LR 0.001000 Time 0.022363 -2022-12-06 10:43:13,998 - Epoch: [47][ 350/ 1200] Overall Loss 0.276630 Objective Loss 0.276630 LR 0.001000 Time 0.022270 -2022-12-06 10:43:14,190 - Epoch: [47][ 360/ 1200] Overall Loss 0.276280 Objective Loss 0.276280 LR 0.001000 Time 0.022186 -2022-12-06 10:43:14,381 - Epoch: [47][ 370/ 1200] Overall Loss 0.276498 Objective Loss 0.276498 LR 0.001000 Time 0.022101 -2022-12-06 10:43:14,573 - Epoch: [47][ 380/ 1200] Overall Loss 0.275979 Objective Loss 0.275979 LR 0.001000 Time 0.022023 -2022-12-06 10:43:14,765 - Epoch: [47][ 390/ 1200] Overall Loss 0.275971 Objective Loss 0.275971 LR 0.001000 Time 0.021947 -2022-12-06 10:43:14,957 - Epoch: [47][ 400/ 1200] Overall Loss 0.277000 Objective Loss 0.277000 LR 0.001000 Time 0.021878 -2022-12-06 10:43:15,148 - Epoch: [47][ 410/ 1200] Overall Loss 0.277299 Objective Loss 0.277299 LR 0.001000 Time 0.021809 -2022-12-06 10:43:15,340 - Epoch: [47][ 420/ 1200] Overall Loss 0.277066 Objective Loss 0.277066 LR 0.001000 Time 0.021746 -2022-12-06 10:43:15,531 - Epoch: [47][ 430/ 1200] Overall Loss 0.276946 Objective Loss 0.276946 LR 0.001000 Time 0.021683 -2022-12-06 10:43:15,722 - Epoch: [47][ 440/ 1200] Overall Loss 0.276706 Objective Loss 0.276706 LR 0.001000 Time 0.021624 -2022-12-06 10:43:15,913 - Epoch: [47][ 450/ 1200] Overall Loss 0.276313 Objective Loss 0.276313 LR 0.001000 Time 0.021567 -2022-12-06 10:43:16,105 - Epoch: [47][ 460/ 1200] Overall Loss 0.276233 Objective Loss 0.276233 LR 0.001000 Time 0.021514 -2022-12-06 10:43:16,298 - Epoch: [47][ 470/ 1200] Overall Loss 0.275812 Objective Loss 0.275812 LR 0.001000 Time 0.021464 -2022-12-06 10:43:16,490 - Epoch: [47][ 480/ 1200] Overall Loss 0.276219 Objective Loss 0.276219 LR 0.001000 Time 0.021417 -2022-12-06 10:43:16,681 - Epoch: [47][ 490/ 1200] Overall Loss 0.276956 Objective Loss 0.276956 LR 0.001000 Time 0.021369 -2022-12-06 10:43:16,873 - Epoch: [47][ 500/ 1200] Overall Loss 0.276761 Objective Loss 0.276761 LR 0.001000 Time 0.021325 -2022-12-06 10:43:17,064 - Epoch: [47][ 510/ 1200] Overall Loss 0.275720 Objective Loss 0.275720 LR 0.001000 Time 0.021280 -2022-12-06 10:43:17,256 - Epoch: [47][ 520/ 1200] Overall Loss 0.275971 Objective Loss 0.275971 LR 0.001000 Time 0.021239 -2022-12-06 10:43:17,447 - Epoch: [47][ 530/ 1200] Overall Loss 0.275966 Objective Loss 0.275966 LR 0.001000 Time 0.021197 -2022-12-06 10:43:17,639 - Epoch: [47][ 540/ 1200] Overall Loss 0.276337 Objective Loss 0.276337 LR 0.001000 Time 0.021158 -2022-12-06 10:43:17,830 - Epoch: [47][ 550/ 1200] Overall Loss 0.275935 Objective Loss 0.275935 LR 0.001000 Time 0.021120 -2022-12-06 10:43:18,022 - Epoch: [47][ 560/ 1200] Overall Loss 0.275751 Objective Loss 0.275751 LR 0.001000 Time 0.021085 -2022-12-06 10:43:18,213 - Epoch: [47][ 570/ 1200] Overall Loss 0.275802 Objective Loss 0.275802 LR 0.001000 Time 0.021050 -2022-12-06 10:43:18,405 - Epoch: [47][ 580/ 1200] Overall Loss 0.275932 Objective Loss 0.275932 LR 0.001000 Time 0.021016 -2022-12-06 10:43:18,595 - Epoch: [47][ 590/ 1200] Overall Loss 0.276018 Objective Loss 0.276018 LR 0.001000 Time 0.020983 -2022-12-06 10:43:18,787 - Epoch: [47][ 600/ 1200] Overall Loss 0.276438 Objective Loss 0.276438 LR 0.001000 Time 0.020951 -2022-12-06 10:43:18,978 - Epoch: [47][ 610/ 1200] Overall Loss 0.276974 Objective Loss 0.276974 LR 0.001000 Time 0.020920 -2022-12-06 10:43:19,170 - Epoch: [47][ 620/ 1200] Overall Loss 0.276933 Objective Loss 0.276933 LR 0.001000 Time 0.020891 -2022-12-06 10:43:19,361 - Epoch: [47][ 630/ 1200] Overall Loss 0.276952 Objective Loss 0.276952 LR 0.001000 Time 0.020863 -2022-12-06 10:43:19,554 - Epoch: [47][ 640/ 1200] Overall Loss 0.276760 Objective Loss 0.276760 LR 0.001000 Time 0.020836 -2022-12-06 10:43:19,745 - Epoch: [47][ 650/ 1200] Overall Loss 0.277156 Objective Loss 0.277156 LR 0.001000 Time 0.020810 -2022-12-06 10:43:19,937 - Epoch: [47][ 660/ 1200] Overall Loss 0.277458 Objective Loss 0.277458 LR 0.001000 Time 0.020785 -2022-12-06 10:43:20,128 - Epoch: [47][ 670/ 1200] Overall Loss 0.277300 Objective Loss 0.277300 LR 0.001000 Time 0.020759 -2022-12-06 10:43:20,320 - Epoch: [47][ 680/ 1200] Overall Loss 0.277028 Objective Loss 0.277028 LR 0.001000 Time 0.020735 -2022-12-06 10:43:20,511 - Epoch: [47][ 690/ 1200] Overall Loss 0.277119 Objective Loss 0.277119 LR 0.001000 Time 0.020710 -2022-12-06 10:43:20,703 - Epoch: [47][ 700/ 1200] Overall Loss 0.277366 Objective Loss 0.277366 LR 0.001000 Time 0.020687 -2022-12-06 10:43:20,894 - Epoch: [47][ 710/ 1200] Overall Loss 0.276767 Objective Loss 0.276767 LR 0.001000 Time 0.020664 -2022-12-06 10:43:21,086 - Epoch: [47][ 720/ 1200] Overall Loss 0.276807 Objective Loss 0.276807 LR 0.001000 Time 0.020643 -2022-12-06 10:43:21,277 - Epoch: [47][ 730/ 1200] Overall Loss 0.276800 Objective Loss 0.276800 LR 0.001000 Time 0.020622 -2022-12-06 10:43:21,469 - Epoch: [47][ 740/ 1200] Overall Loss 0.276779 Objective Loss 0.276779 LR 0.001000 Time 0.020602 -2022-12-06 10:43:21,660 - Epoch: [47][ 750/ 1200] Overall Loss 0.276428 Objective Loss 0.276428 LR 0.001000 Time 0.020581 -2022-12-06 10:43:21,852 - Epoch: [47][ 760/ 1200] Overall Loss 0.276462 Objective Loss 0.276462 LR 0.001000 Time 0.020562 -2022-12-06 10:43:22,042 - Epoch: [47][ 770/ 1200] Overall Loss 0.276571 Objective Loss 0.276571 LR 0.001000 Time 0.020542 -2022-12-06 10:43:22,235 - Epoch: [47][ 780/ 1200] Overall Loss 0.276559 Objective Loss 0.276559 LR 0.001000 Time 0.020524 -2022-12-06 10:43:22,428 - Epoch: [47][ 790/ 1200] Overall Loss 0.276432 Objective Loss 0.276432 LR 0.001000 Time 0.020508 -2022-12-06 10:43:22,620 - Epoch: [47][ 800/ 1200] Overall Loss 0.276589 Objective Loss 0.276589 LR 0.001000 Time 0.020491 -2022-12-06 10:43:22,810 - Epoch: [47][ 810/ 1200] Overall Loss 0.276603 Objective Loss 0.276603 LR 0.001000 Time 0.020473 -2022-12-06 10:43:23,002 - Epoch: [47][ 820/ 1200] Overall Loss 0.276703 Objective Loss 0.276703 LR 0.001000 Time 0.020456 -2022-12-06 10:43:23,193 - Epoch: [47][ 830/ 1200] Overall Loss 0.277389 Objective Loss 0.277389 LR 0.001000 Time 0.020439 -2022-12-06 10:43:23,385 - Epoch: [47][ 840/ 1200] Overall Loss 0.277410 Objective Loss 0.277410 LR 0.001000 Time 0.020423 -2022-12-06 10:43:23,576 - Epoch: [47][ 850/ 1200] Overall Loss 0.277559 Objective Loss 0.277559 LR 0.001000 Time 0.020407 -2022-12-06 10:43:23,768 - Epoch: [47][ 860/ 1200] Overall Loss 0.277212 Objective Loss 0.277212 LR 0.001000 Time 0.020393 -2022-12-06 10:43:23,959 - Epoch: [47][ 870/ 1200] Overall Loss 0.277022 Objective Loss 0.277022 LR 0.001000 Time 0.020377 -2022-12-06 10:43:24,151 - Epoch: [47][ 880/ 1200] Overall Loss 0.277076 Objective Loss 0.277076 LR 0.001000 Time 0.020364 -2022-12-06 10:43:24,343 - Epoch: [47][ 890/ 1200] Overall Loss 0.277371 Objective Loss 0.277371 LR 0.001000 Time 0.020350 -2022-12-06 10:43:24,535 - Epoch: [47][ 900/ 1200] Overall Loss 0.277563 Objective Loss 0.277563 LR 0.001000 Time 0.020336 -2022-12-06 10:43:24,727 - Epoch: [47][ 910/ 1200] Overall Loss 0.277650 Objective Loss 0.277650 LR 0.001000 Time 0.020323 -2022-12-06 10:43:24,918 - Epoch: [47][ 920/ 1200] Overall Loss 0.277913 Objective Loss 0.277913 LR 0.001000 Time 0.020310 -2022-12-06 10:43:25,110 - Epoch: [47][ 930/ 1200] Overall Loss 0.278119 Objective Loss 0.278119 LR 0.001000 Time 0.020297 -2022-12-06 10:43:25,301 - Epoch: [47][ 940/ 1200] Overall Loss 0.278300 Objective Loss 0.278300 LR 0.001000 Time 0.020284 -2022-12-06 10:43:25,493 - Epoch: [47][ 950/ 1200] Overall Loss 0.278432 Objective Loss 0.278432 LR 0.001000 Time 0.020271 -2022-12-06 10:43:25,685 - Epoch: [47][ 960/ 1200] Overall Loss 0.278694 Objective Loss 0.278694 LR 0.001000 Time 0.020260 -2022-12-06 10:43:25,876 - Epoch: [47][ 970/ 1200] Overall Loss 0.278735 Objective Loss 0.278735 LR 0.001000 Time 0.020248 -2022-12-06 10:43:26,068 - Epoch: [47][ 980/ 1200] Overall Loss 0.278661 Objective Loss 0.278661 LR 0.001000 Time 0.020237 -2022-12-06 10:43:26,259 - Epoch: [47][ 990/ 1200] Overall Loss 0.278532 Objective Loss 0.278532 LR 0.001000 Time 0.020224 -2022-12-06 10:43:26,452 - Epoch: [47][ 1000/ 1200] Overall Loss 0.278416 Objective Loss 0.278416 LR 0.001000 Time 0.020214 -2022-12-06 10:43:26,643 - Epoch: [47][ 1010/ 1200] Overall Loss 0.278201 Objective Loss 0.278201 LR 0.001000 Time 0.020203 -2022-12-06 10:43:26,835 - Epoch: [47][ 1020/ 1200] Overall Loss 0.278403 Objective Loss 0.278403 LR 0.001000 Time 0.020193 -2022-12-06 10:43:27,027 - Epoch: [47][ 1030/ 1200] Overall Loss 0.278236 Objective Loss 0.278236 LR 0.001000 Time 0.020182 -2022-12-06 10:43:27,219 - Epoch: [47][ 1040/ 1200] Overall Loss 0.278428 Objective Loss 0.278428 LR 0.001000 Time 0.020173 -2022-12-06 10:43:27,411 - Epoch: [47][ 1050/ 1200] Overall Loss 0.278535 Objective Loss 0.278535 LR 0.001000 Time 0.020163 -2022-12-06 10:43:27,604 - Epoch: [47][ 1060/ 1200] Overall Loss 0.278586 Objective Loss 0.278586 LR 0.001000 Time 0.020153 -2022-12-06 10:43:27,795 - Epoch: [47][ 1070/ 1200] Overall Loss 0.278584 Objective Loss 0.278584 LR 0.001000 Time 0.020143 -2022-12-06 10:43:27,986 - Epoch: [47][ 1080/ 1200] Overall Loss 0.278853 Objective Loss 0.278853 LR 0.001000 Time 0.020134 -2022-12-06 10:43:28,177 - Epoch: [47][ 1090/ 1200] Overall Loss 0.279191 Objective Loss 0.279191 LR 0.001000 Time 0.020124 -2022-12-06 10:43:28,369 - Epoch: [47][ 1100/ 1200] Overall Loss 0.279384 Objective Loss 0.279384 LR 0.001000 Time 0.020115 -2022-12-06 10:43:28,561 - Epoch: [47][ 1110/ 1200] Overall Loss 0.279656 Objective Loss 0.279656 LR 0.001000 Time 0.020106 -2022-12-06 10:43:28,753 - Epoch: [47][ 1120/ 1200] Overall Loss 0.279654 Objective Loss 0.279654 LR 0.001000 Time 0.020097 -2022-12-06 10:43:28,945 - Epoch: [47][ 1130/ 1200] Overall Loss 0.279644 Objective Loss 0.279644 LR 0.001000 Time 0.020089 -2022-12-06 10:43:29,138 - Epoch: [47][ 1140/ 1200] Overall Loss 0.279735 Objective Loss 0.279735 LR 0.001000 Time 0.020081 -2022-12-06 10:43:29,330 - Epoch: [47][ 1150/ 1200] Overall Loss 0.279706 Objective Loss 0.279706 LR 0.001000 Time 0.020073 -2022-12-06 10:43:29,521 - Epoch: [47][ 1160/ 1200] Overall Loss 0.279722 Objective Loss 0.279722 LR 0.001000 Time 0.020065 -2022-12-06 10:43:29,713 - Epoch: [47][ 1170/ 1200] Overall Loss 0.279745 Objective Loss 0.279745 LR 0.001000 Time 0.020057 -2022-12-06 10:43:29,906 - Epoch: [47][ 1180/ 1200] Overall Loss 0.279743 Objective Loss 0.279743 LR 0.001000 Time 0.020049 -2022-12-06 10:43:30,097 - Epoch: [47][ 1190/ 1200] Overall Loss 0.279913 Objective Loss 0.279913 LR 0.001000 Time 0.020041 -2022-12-06 10:43:30,328 - Epoch: [47][ 1200/ 1200] Overall Loss 0.279552 Objective Loss 0.279552 Top1 88.075314 Top5 98.326360 LR 0.001000 Time 0.020067 -2022-12-06 10:43:30,417 - --- validate (epoch=47)----------- -2022-12-06 10:43:30,417 - 34129 samples (256 per mini-batch) -2022-12-06 10:43:30,862 - Epoch: [47][ 10/ 134] Loss 0.315482 Top1 86.171875 Top5 98.007812 -2022-12-06 10:43:30,988 - Epoch: [47][ 20/ 134] Loss 0.303130 Top1 85.800781 Top5 97.988281 -2022-12-06 10:43:31,118 - Epoch: [47][ 30/ 134] Loss 0.299554 Top1 85.651042 Top5 98.138021 -2022-12-06 10:43:31,246 - Epoch: [47][ 40/ 134] Loss 0.304959 Top1 85.449219 Top5 98.164062 -2022-12-06 10:43:31,376 - Epoch: [47][ 50/ 134] Loss 0.304231 Top1 85.484375 Top5 98.140625 -2022-12-06 10:43:31,504 - Epoch: [47][ 60/ 134] Loss 0.299751 Top1 85.514323 Top5 98.098958 -2022-12-06 10:43:31,631 - Epoch: [47][ 70/ 134] Loss 0.299076 Top1 85.502232 Top5 98.152902 -2022-12-06 10:43:31,759 - Epoch: [47][ 80/ 134] Loss 0.300326 Top1 85.478516 Top5 98.125000 -2022-12-06 10:43:31,886 - Epoch: [47][ 90/ 134] Loss 0.302667 Top1 85.373264 Top5 98.098958 -2022-12-06 10:43:32,014 - Epoch: [47][ 100/ 134] Loss 0.303800 Top1 85.183594 Top5 98.121094 -2022-12-06 10:43:32,142 - Epoch: [47][ 110/ 134] Loss 0.304691 Top1 85.149148 Top5 98.128551 -2022-12-06 10:43:32,271 - Epoch: [47][ 120/ 134] Loss 0.304146 Top1 85.205078 Top5 98.164062 -2022-12-06 10:43:32,401 - Epoch: [47][ 130/ 134] Loss 0.302668 Top1 85.318510 Top5 98.164062 -2022-12-06 10:43:32,439 - Epoch: [47][ 134/ 134] Loss 0.304123 Top1 85.255941 Top5 98.171643 -2022-12-06 10:43:32,527 - ==> Top1: 85.256 Top5: 98.172 Loss: 0.304 - -2022-12-06 10:43:32,528 - ==> Confusion: -[[ 880 3 1 0 14 7 1 0 4 63 0 2 2 5 6 1 1 0 2 0 4] - [ 1 937 0 2 8 11 5 9 2 2 5 4 7 2 1 1 4 1 9 0 16] - [ 4 4 980 10 2 2 38 6 0 1 4 5 5 4 2 9 1 2 7 4 13] - [ 3 4 20 916 1 5 2 0 0 0 9 0 9 0 27 1 0 6 10 0 7] - [ 7 8 2 0 941 2 0 0 0 7 2 2 2 7 10 8 8 3 0 1 10] - [ 2 26 0 3 12 927 4 12 2 3 0 17 9 18 5 3 0 2 6 6 12] - [ 0 3 10 1 0 1 1071 3 0 0 1 3 0 2 0 7 0 3 2 7 4] - [ 0 30 5 2 2 28 6 882 2 0 2 8 4 2 0 1 1 1 51 18 9] - [ 9 4 0 0 0 2 1 0 931 41 19 2 4 16 22 0 3 2 2 2 4] - [ 58 0 1 0 2 1 1 1 20 884 1 1 0 12 9 2 0 1 0 0 7] - [ 0 2 6 1 0 0 2 0 4 0 955 3 3 19 7 0 2 2 6 1 6] - [ 3 2 0 0 0 8 1 3 1 2 1 964 28 9 1 2 6 6 0 9 5] - [ 2 1 0 2 0 2 0 1 0 0 0 40 888 1 0 11 0 10 0 4 7] - [ 0 1 0 0 0 9 1 2 4 12 6 7 7 944 2 4 1 3 0 6 14] - [ 2 4 0 9 8 3 0 1 11 2 2 2 3 2 1061 0 1 3 6 1 9] - [ 0 1 0 0 1 2 6 0 0 0 0 8 12 2 0 990 6 9 0 2 4] - [ 2 2 2 1 3 1 3 0 0 0 0 5 6 2 3 14 1016 1 1 4 6] - [ 4 1 0 1 0 2 0 0 0 3 0 12 20 2 1 12 2 974 0 1 1] - [ 4 5 8 7 2 4 1 8 1 0 6 3 7 2 8 0 0 1 933 2 6] - [ 2 4 3 1 0 7 7 5 0 0 0 21 7 4 0 2 3 4 2 996 12] - [ 101 219 129 84 118 149 122 91 55 72 194 113 446 341 165 162 123 90 175 253 10024]] - -2022-12-06 10:43:33,112 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:43:33,112 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:43:33,119 - - -2022-12-06 10:43:33,119 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:43:34,161 - Epoch: [48][ 10/ 1200] Overall Loss 0.243110 Objective Loss 0.243110 LR 0.001000 Time 0.104167 -2022-12-06 10:43:34,367 - Epoch: [48][ 20/ 1200] Overall Loss 0.257749 Objective Loss 0.257749 LR 0.001000 Time 0.062371 -2022-12-06 10:43:34,568 - Epoch: [48][ 30/ 1200] Overall Loss 0.261948 Objective Loss 0.261948 LR 0.001000 Time 0.048257 -2022-12-06 10:43:34,771 - Epoch: [48][ 40/ 1200] Overall Loss 0.278945 Objective Loss 0.278945 LR 0.001000 Time 0.041247 -2022-12-06 10:43:34,972 - Epoch: [48][ 50/ 1200] Overall Loss 0.275085 Objective Loss 0.275085 LR 0.001000 Time 0.036998 -2022-12-06 10:43:35,175 - Epoch: [48][ 60/ 1200] Overall Loss 0.273679 Objective Loss 0.273679 LR 0.001000 Time 0.034211 -2022-12-06 10:43:35,376 - Epoch: [48][ 70/ 1200] Overall Loss 0.272160 Objective Loss 0.272160 LR 0.001000 Time 0.032183 -2022-12-06 10:43:35,579 - Epoch: [48][ 80/ 1200] Overall Loss 0.276830 Objective Loss 0.276830 LR 0.001000 Time 0.030690 -2022-12-06 10:43:35,779 - Epoch: [48][ 90/ 1200] Overall Loss 0.281366 Objective Loss 0.281366 LR 0.001000 Time 0.029491 -2022-12-06 10:43:35,982 - Epoch: [48][ 100/ 1200] Overall Loss 0.280812 Objective Loss 0.280812 LR 0.001000 Time 0.028573 -2022-12-06 10:43:36,181 - Epoch: [48][ 110/ 1200] Overall Loss 0.282474 Objective Loss 0.282474 LR 0.001000 Time 0.027779 -2022-12-06 10:43:36,385 - Epoch: [48][ 120/ 1200] Overall Loss 0.281485 Objective Loss 0.281485 LR 0.001000 Time 0.027157 -2022-12-06 10:43:36,585 - Epoch: [48][ 130/ 1200] Overall Loss 0.283393 Objective Loss 0.283393 LR 0.001000 Time 0.026601 -2022-12-06 10:43:36,787 - Epoch: [48][ 140/ 1200] Overall Loss 0.283404 Objective Loss 0.283404 LR 0.001000 Time 0.026143 -2022-12-06 10:43:36,987 - Epoch: [48][ 150/ 1200] Overall Loss 0.282020 Objective Loss 0.282020 LR 0.001000 Time 0.025727 -2022-12-06 10:43:37,190 - Epoch: [48][ 160/ 1200] Overall Loss 0.281453 Objective Loss 0.281453 LR 0.001000 Time 0.025381 -2022-12-06 10:43:37,389 - Epoch: [48][ 170/ 1200] Overall Loss 0.280405 Objective Loss 0.280405 LR 0.001000 Time 0.025059 -2022-12-06 10:43:37,593 - Epoch: [48][ 180/ 1200] Overall Loss 0.281416 Objective Loss 0.281416 LR 0.001000 Time 0.024797 -2022-12-06 10:43:37,793 - Epoch: [48][ 190/ 1200] Overall Loss 0.281847 Objective Loss 0.281847 LR 0.001000 Time 0.024541 -2022-12-06 10:43:37,996 - Epoch: [48][ 200/ 1200] Overall Loss 0.281340 Objective Loss 0.281340 LR 0.001000 Time 0.024323 -2022-12-06 10:43:38,195 - Epoch: [48][ 210/ 1200] Overall Loss 0.281419 Objective Loss 0.281419 LR 0.001000 Time 0.024113 -2022-12-06 10:43:38,399 - Epoch: [48][ 220/ 1200] Overall Loss 0.281324 Objective Loss 0.281324 LR 0.001000 Time 0.023941 -2022-12-06 10:43:38,599 - Epoch: [48][ 230/ 1200] Overall Loss 0.280046 Objective Loss 0.280046 LR 0.001000 Time 0.023767 -2022-12-06 10:43:38,802 - Epoch: [48][ 240/ 1200] Overall Loss 0.279058 Objective Loss 0.279058 LR 0.001000 Time 0.023620 -2022-12-06 10:43:39,002 - Epoch: [48][ 250/ 1200] Overall Loss 0.279040 Objective Loss 0.279040 LR 0.001000 Time 0.023470 -2022-12-06 10:43:39,204 - Epoch: [48][ 260/ 1200] Overall Loss 0.277755 Objective Loss 0.277755 LR 0.001000 Time 0.023345 -2022-12-06 10:43:39,404 - Epoch: [48][ 270/ 1200] Overall Loss 0.277261 Objective Loss 0.277261 LR 0.001000 Time 0.023218 -2022-12-06 10:43:39,608 - Epoch: [48][ 280/ 1200] Overall Loss 0.276409 Objective Loss 0.276409 LR 0.001000 Time 0.023115 -2022-12-06 10:43:39,808 - Epoch: [48][ 290/ 1200] Overall Loss 0.275521 Objective Loss 0.275521 LR 0.001000 Time 0.023006 -2022-12-06 10:43:40,012 - Epoch: [48][ 300/ 1200] Overall Loss 0.275227 Objective Loss 0.275227 LR 0.001000 Time 0.022915 -2022-12-06 10:43:40,212 - Epoch: [48][ 310/ 1200] Overall Loss 0.276413 Objective Loss 0.276413 LR 0.001000 Time 0.022819 -2022-12-06 10:43:40,415 - Epoch: [48][ 320/ 1200] Overall Loss 0.276230 Objective Loss 0.276230 LR 0.001000 Time 0.022738 -2022-12-06 10:43:40,616 - Epoch: [48][ 330/ 1200] Overall Loss 0.276261 Objective Loss 0.276261 LR 0.001000 Time 0.022656 -2022-12-06 10:43:40,819 - Epoch: [48][ 340/ 1200] Overall Loss 0.276148 Objective Loss 0.276148 LR 0.001000 Time 0.022586 -2022-12-06 10:43:41,019 - Epoch: [48][ 350/ 1200] Overall Loss 0.277047 Objective Loss 0.277047 LR 0.001000 Time 0.022511 -2022-12-06 10:43:41,222 - Epoch: [48][ 360/ 1200] Overall Loss 0.276883 Objective Loss 0.276883 LR 0.001000 Time 0.022448 -2022-12-06 10:43:41,414 - Epoch: [48][ 370/ 1200] Overall Loss 0.277442 Objective Loss 0.277442 LR 0.001000 Time 0.022357 -2022-12-06 10:43:41,604 - Epoch: [48][ 380/ 1200] Overall Loss 0.277500 Objective Loss 0.277500 LR 0.001000 Time 0.022267 -2022-12-06 10:43:41,793 - Epoch: [48][ 390/ 1200] Overall Loss 0.277586 Objective Loss 0.277586 LR 0.001000 Time 0.022180 -2022-12-06 10:43:41,983 - Epoch: [48][ 400/ 1200] Overall Loss 0.277650 Objective Loss 0.277650 LR 0.001000 Time 0.022098 -2022-12-06 10:43:42,172 - Epoch: [48][ 410/ 1200] Overall Loss 0.278668 Objective Loss 0.278668 LR 0.001000 Time 0.022021 -2022-12-06 10:43:42,361 - Epoch: [48][ 420/ 1200] Overall Loss 0.278235 Objective Loss 0.278235 LR 0.001000 Time 0.021945 -2022-12-06 10:43:42,551 - Epoch: [48][ 430/ 1200] Overall Loss 0.278049 Objective Loss 0.278049 LR 0.001000 Time 0.021875 -2022-12-06 10:43:42,741 - Epoch: [48][ 440/ 1200] Overall Loss 0.278437 Objective Loss 0.278437 LR 0.001000 Time 0.021808 -2022-12-06 10:43:42,931 - Epoch: [48][ 450/ 1200] Overall Loss 0.278632 Objective Loss 0.278632 LR 0.001000 Time 0.021744 -2022-12-06 10:43:43,120 - Epoch: [48][ 460/ 1200] Overall Loss 0.278923 Objective Loss 0.278923 LR 0.001000 Time 0.021682 -2022-12-06 10:43:43,310 - Epoch: [48][ 470/ 1200] Overall Loss 0.278265 Objective Loss 0.278265 LR 0.001000 Time 0.021622 -2022-12-06 10:43:43,501 - Epoch: [48][ 480/ 1200] Overall Loss 0.278009 Objective Loss 0.278009 LR 0.001000 Time 0.021568 -2022-12-06 10:43:43,690 - Epoch: [48][ 490/ 1200] Overall Loss 0.277918 Objective Loss 0.277918 LR 0.001000 Time 0.021514 -2022-12-06 10:43:43,879 - Epoch: [48][ 500/ 1200] Overall Loss 0.278384 Objective Loss 0.278384 LR 0.001000 Time 0.021460 -2022-12-06 10:43:44,068 - Epoch: [48][ 510/ 1200] Overall Loss 0.278352 Objective Loss 0.278352 LR 0.001000 Time 0.021410 -2022-12-06 10:43:44,258 - Epoch: [48][ 520/ 1200] Overall Loss 0.278518 Objective Loss 0.278518 LR 0.001000 Time 0.021362 -2022-12-06 10:43:44,447 - Epoch: [48][ 530/ 1200] Overall Loss 0.278415 Objective Loss 0.278415 LR 0.001000 Time 0.021314 -2022-12-06 10:43:44,637 - Epoch: [48][ 540/ 1200] Overall Loss 0.278468 Objective Loss 0.278468 LR 0.001000 Time 0.021270 -2022-12-06 10:43:44,826 - Epoch: [48][ 550/ 1200] Overall Loss 0.278303 Objective Loss 0.278303 LR 0.001000 Time 0.021227 -2022-12-06 10:43:45,016 - Epoch: [48][ 560/ 1200] Overall Loss 0.278443 Objective Loss 0.278443 LR 0.001000 Time 0.021186 -2022-12-06 10:43:45,205 - Epoch: [48][ 570/ 1200] Overall Loss 0.278739 Objective Loss 0.278739 LR 0.001000 Time 0.021145 -2022-12-06 10:43:45,395 - Epoch: [48][ 580/ 1200] Overall Loss 0.278493 Objective Loss 0.278493 LR 0.001000 Time 0.021107 -2022-12-06 10:43:45,585 - Epoch: [48][ 590/ 1200] Overall Loss 0.278408 Objective Loss 0.278408 LR 0.001000 Time 0.021069 -2022-12-06 10:43:45,775 - Epoch: [48][ 600/ 1200] Overall Loss 0.277908 Objective Loss 0.277908 LR 0.001000 Time 0.021034 -2022-12-06 10:43:45,964 - Epoch: [48][ 610/ 1200] Overall Loss 0.277681 Objective Loss 0.277681 LR 0.001000 Time 0.020999 -2022-12-06 10:43:46,155 - Epoch: [48][ 620/ 1200] Overall Loss 0.277614 Objective Loss 0.277614 LR 0.001000 Time 0.020967 -2022-12-06 10:43:46,345 - Epoch: [48][ 630/ 1200] Overall Loss 0.277791 Objective Loss 0.277791 LR 0.001000 Time 0.020934 -2022-12-06 10:43:46,534 - Epoch: [48][ 640/ 1200] Overall Loss 0.277454 Objective Loss 0.277454 LR 0.001000 Time 0.020902 -2022-12-06 10:43:46,724 - Epoch: [48][ 650/ 1200] Overall Loss 0.277797 Objective Loss 0.277797 LR 0.001000 Time 0.020872 -2022-12-06 10:43:46,913 - Epoch: [48][ 660/ 1200] Overall Loss 0.277257 Objective Loss 0.277257 LR 0.001000 Time 0.020842 -2022-12-06 10:43:47,103 - Epoch: [48][ 670/ 1200] Overall Loss 0.276694 Objective Loss 0.276694 LR 0.001000 Time 0.020813 -2022-12-06 10:43:47,293 - Epoch: [48][ 680/ 1200] Overall Loss 0.276761 Objective Loss 0.276761 LR 0.001000 Time 0.020785 -2022-12-06 10:43:47,483 - Epoch: [48][ 690/ 1200] Overall Loss 0.276627 Objective Loss 0.276627 LR 0.001000 Time 0.020758 -2022-12-06 10:43:47,672 - Epoch: [48][ 700/ 1200] Overall Loss 0.276422 Objective Loss 0.276422 LR 0.001000 Time 0.020731 -2022-12-06 10:43:47,861 - Epoch: [48][ 710/ 1200] Overall Loss 0.276260 Objective Loss 0.276260 LR 0.001000 Time 0.020705 -2022-12-06 10:43:48,051 - Epoch: [48][ 720/ 1200] Overall Loss 0.276205 Objective Loss 0.276205 LR 0.001000 Time 0.020680 -2022-12-06 10:43:48,240 - Epoch: [48][ 730/ 1200] Overall Loss 0.276621 Objective Loss 0.276621 LR 0.001000 Time 0.020655 -2022-12-06 10:43:48,430 - Epoch: [48][ 740/ 1200] Overall Loss 0.277182 Objective Loss 0.277182 LR 0.001000 Time 0.020632 -2022-12-06 10:43:48,619 - Epoch: [48][ 750/ 1200] Overall Loss 0.277397 Objective Loss 0.277397 LR 0.001000 Time 0.020609 -2022-12-06 10:43:48,808 - Epoch: [48][ 760/ 1200] Overall Loss 0.277270 Objective Loss 0.277270 LR 0.001000 Time 0.020586 -2022-12-06 10:43:48,998 - Epoch: [48][ 770/ 1200] Overall Loss 0.276946 Objective Loss 0.276946 LR 0.001000 Time 0.020564 -2022-12-06 10:43:49,187 - Epoch: [48][ 780/ 1200] Overall Loss 0.276361 Objective Loss 0.276361 LR 0.001000 Time 0.020542 -2022-12-06 10:43:49,376 - Epoch: [48][ 790/ 1200] Overall Loss 0.276450 Objective Loss 0.276450 LR 0.001000 Time 0.020520 -2022-12-06 10:43:49,565 - Epoch: [48][ 800/ 1200] Overall Loss 0.276433 Objective Loss 0.276433 LR 0.001000 Time 0.020500 -2022-12-06 10:43:49,755 - Epoch: [48][ 810/ 1200] Overall Loss 0.276764 Objective Loss 0.276764 LR 0.001000 Time 0.020480 -2022-12-06 10:43:49,944 - Epoch: [48][ 820/ 1200] Overall Loss 0.276898 Objective Loss 0.276898 LR 0.001000 Time 0.020461 -2022-12-06 10:43:50,134 - Epoch: [48][ 830/ 1200] Overall Loss 0.276881 Objective Loss 0.276881 LR 0.001000 Time 0.020442 -2022-12-06 10:43:50,324 - Epoch: [48][ 840/ 1200] Overall Loss 0.276978 Objective Loss 0.276978 LR 0.001000 Time 0.020425 -2022-12-06 10:43:50,514 - Epoch: [48][ 850/ 1200] Overall Loss 0.277017 Objective Loss 0.277017 LR 0.001000 Time 0.020407 -2022-12-06 10:43:50,704 - Epoch: [48][ 860/ 1200] Overall Loss 0.277364 Objective Loss 0.277364 LR 0.001000 Time 0.020389 -2022-12-06 10:43:50,894 - Epoch: [48][ 870/ 1200] Overall Loss 0.277522 Objective Loss 0.277522 LR 0.001000 Time 0.020373 -2022-12-06 10:43:51,083 - Epoch: [48][ 880/ 1200] Overall Loss 0.277890 Objective Loss 0.277890 LR 0.001000 Time 0.020356 -2022-12-06 10:43:51,273 - Epoch: [48][ 890/ 1200] Overall Loss 0.277464 Objective Loss 0.277464 LR 0.001000 Time 0.020339 -2022-12-06 10:43:51,462 - Epoch: [48][ 900/ 1200] Overall Loss 0.277824 Objective Loss 0.277824 LR 0.001000 Time 0.020324 -2022-12-06 10:43:51,652 - Epoch: [48][ 910/ 1200] Overall Loss 0.277694 Objective Loss 0.277694 LR 0.001000 Time 0.020308 -2022-12-06 10:43:51,842 - Epoch: [48][ 920/ 1200] Overall Loss 0.277578 Objective Loss 0.277578 LR 0.001000 Time 0.020293 -2022-12-06 10:43:52,031 - Epoch: [48][ 930/ 1200] Overall Loss 0.277942 Objective Loss 0.277942 LR 0.001000 Time 0.020278 -2022-12-06 10:43:52,221 - Epoch: [48][ 940/ 1200] Overall Loss 0.277785 Objective Loss 0.277785 LR 0.001000 Time 0.020264 -2022-12-06 10:43:52,411 - Epoch: [48][ 950/ 1200] Overall Loss 0.278070 Objective Loss 0.278070 LR 0.001000 Time 0.020249 -2022-12-06 10:43:52,601 - Epoch: [48][ 960/ 1200] Overall Loss 0.278158 Objective Loss 0.278158 LR 0.001000 Time 0.020236 -2022-12-06 10:43:52,791 - Epoch: [48][ 970/ 1200] Overall Loss 0.277973 Objective Loss 0.277973 LR 0.001000 Time 0.020223 -2022-12-06 10:43:52,981 - Epoch: [48][ 980/ 1200] Overall Loss 0.277698 Objective Loss 0.277698 LR 0.001000 Time 0.020209 -2022-12-06 10:43:53,170 - Epoch: [48][ 990/ 1200] Overall Loss 0.277495 Objective Loss 0.277495 LR 0.001000 Time 0.020196 -2022-12-06 10:43:53,360 - Epoch: [48][ 1000/ 1200] Overall Loss 0.277423 Objective Loss 0.277423 LR 0.001000 Time 0.020183 -2022-12-06 10:43:53,550 - Epoch: [48][ 1010/ 1200] Overall Loss 0.277828 Objective Loss 0.277828 LR 0.001000 Time 0.020171 -2022-12-06 10:43:53,739 - Epoch: [48][ 1020/ 1200] Overall Loss 0.278262 Objective Loss 0.278262 LR 0.001000 Time 0.020158 -2022-12-06 10:43:53,930 - Epoch: [48][ 1030/ 1200] Overall Loss 0.278324 Objective Loss 0.278324 LR 0.001000 Time 0.020147 -2022-12-06 10:43:54,120 - Epoch: [48][ 1040/ 1200] Overall Loss 0.278203 Objective Loss 0.278203 LR 0.001000 Time 0.020135 -2022-12-06 10:43:54,309 - Epoch: [48][ 1050/ 1200] Overall Loss 0.278462 Objective Loss 0.278462 LR 0.001000 Time 0.020123 -2022-12-06 10:43:54,499 - Epoch: [48][ 1060/ 1200] Overall Loss 0.278460 Objective Loss 0.278460 LR 0.001000 Time 0.020112 -2022-12-06 10:43:54,688 - Epoch: [48][ 1070/ 1200] Overall Loss 0.278795 Objective Loss 0.278795 LR 0.001000 Time 0.020100 -2022-12-06 10:43:54,878 - Epoch: [48][ 1080/ 1200] Overall Loss 0.278924 Objective Loss 0.278924 LR 0.001000 Time 0.020089 -2022-12-06 10:43:55,067 - Epoch: [48][ 1090/ 1200] Overall Loss 0.278951 Objective Loss 0.278951 LR 0.001000 Time 0.020078 -2022-12-06 10:43:55,257 - Epoch: [48][ 1100/ 1200] Overall Loss 0.279065 Objective Loss 0.279065 LR 0.001000 Time 0.020068 -2022-12-06 10:43:55,446 - Epoch: [48][ 1110/ 1200] Overall Loss 0.279413 Objective Loss 0.279413 LR 0.001000 Time 0.020057 -2022-12-06 10:43:55,636 - Epoch: [48][ 1120/ 1200] Overall Loss 0.279429 Objective Loss 0.279429 LR 0.001000 Time 0.020047 -2022-12-06 10:43:55,826 - Epoch: [48][ 1130/ 1200] Overall Loss 0.279562 Objective Loss 0.279562 LR 0.001000 Time 0.020037 -2022-12-06 10:43:56,016 - Epoch: [48][ 1140/ 1200] Overall Loss 0.279606 Objective Loss 0.279606 LR 0.001000 Time 0.020028 -2022-12-06 10:43:56,206 - Epoch: [48][ 1150/ 1200] Overall Loss 0.279885 Objective Loss 0.279885 LR 0.001000 Time 0.020018 -2022-12-06 10:43:56,395 - Epoch: [48][ 1160/ 1200] Overall Loss 0.279771 Objective Loss 0.279771 LR 0.001000 Time 0.020008 -2022-12-06 10:43:56,584 - Epoch: [48][ 1170/ 1200] Overall Loss 0.279764 Objective Loss 0.279764 LR 0.001000 Time 0.019999 -2022-12-06 10:43:56,774 - Epoch: [48][ 1180/ 1200] Overall Loss 0.279515 Objective Loss 0.279515 LR 0.001000 Time 0.019989 -2022-12-06 10:43:56,964 - Epoch: [48][ 1190/ 1200] Overall Loss 0.279466 Objective Loss 0.279466 LR 0.001000 Time 0.019980 -2022-12-06 10:43:57,193 - Epoch: [48][ 1200/ 1200] Overall Loss 0.279515 Objective Loss 0.279515 Top1 86.192469 Top5 99.163180 LR 0.001000 Time 0.020004 -2022-12-06 10:43:57,282 - --- validate (epoch=48)----------- -2022-12-06 10:43:57,282 - 34129 samples (256 per mini-batch) -2022-12-06 10:43:57,744 - Epoch: [48][ 10/ 134] Loss 0.286400 Top1 84.492188 Top5 97.382812 -2022-12-06 10:43:57,868 - Epoch: [48][ 20/ 134] Loss 0.294155 Top1 84.023438 Top5 97.617188 -2022-12-06 10:43:57,995 - Epoch: [48][ 30/ 134] Loss 0.297707 Top1 84.166667 Top5 97.604167 -2022-12-06 10:43:58,130 - Epoch: [48][ 40/ 134] Loss 0.293117 Top1 84.482422 Top5 97.675781 -2022-12-06 10:43:58,275 - Epoch: [48][ 50/ 134] Loss 0.295131 Top1 84.250000 Top5 97.742188 -2022-12-06 10:43:58,418 - Epoch: [48][ 60/ 134] Loss 0.297217 Top1 84.264323 Top5 97.773438 -2022-12-06 10:43:58,567 - Epoch: [48][ 70/ 134] Loss 0.295017 Top1 84.140625 Top5 97.801339 -2022-12-06 10:43:58,709 - Epoch: [48][ 80/ 134] Loss 0.294336 Top1 84.067383 Top5 97.763672 -2022-12-06 10:43:58,858 - Epoch: [48][ 90/ 134] Loss 0.296757 Top1 84.040799 Top5 97.738715 -2022-12-06 10:43:59,000 - Epoch: [48][ 100/ 134] Loss 0.295669 Top1 84.101562 Top5 97.792969 -2022-12-06 10:43:59,148 - Epoch: [48][ 110/ 134] Loss 0.294694 Top1 84.069602 Top5 97.819602 -2022-12-06 10:43:59,293 - Epoch: [48][ 120/ 134] Loss 0.294851 Top1 84.127604 Top5 97.838542 -2022-12-06 10:43:59,431 - Epoch: [48][ 130/ 134] Loss 0.295715 Top1 84.149639 Top5 97.824519 -2022-12-06 10:43:59,469 - Epoch: [48][ 134/ 134] Loss 0.296477 Top1 84.148378 Top5 97.817106 -2022-12-06 10:43:59,556 - ==> Top1: 84.148 Top5: 97.817 Loss: 0.296 - -2022-12-06 10:43:59,557 - ==> Confusion: -[[ 888 2 4 1 4 10 0 3 4 57 1 3 3 4 3 3 0 1 2 2 1] - [ 0 902 0 1 11 49 2 23 1 0 2 1 1 1 1 1 4 1 10 6 10] - [ 7 2 967 8 1 4 53 14 0 3 5 4 1 4 1 5 2 3 4 4 11] - [ 2 0 15 935 0 6 2 1 1 0 16 0 4 1 18 1 0 5 6 1 6] - [ 14 6 2 0 939 12 0 2 1 4 2 3 2 3 11 7 8 1 1 1 1] - [ 2 7 1 2 6 992 6 19 3 1 1 7 4 6 1 0 0 0 1 6 4] - [ 0 0 6 0 0 1 1085 3 1 0 1 1 1 2 0 7 0 0 2 6 2] - [ 0 1 5 2 1 37 6 955 3 0 1 4 0 1 0 1 0 0 22 12 3] - [ 3 1 0 1 1 4 1 1 964 52 8 2 2 7 11 0 2 0 0 3 1] - [ 55 0 0 0 4 5 0 4 24 879 1 2 0 11 7 1 0 1 0 1 6] - [ 1 2 3 8 2 1 3 2 14 2 950 3 1 13 4 0 0 0 6 1 3] - [ 4 0 2 0 0 13 4 6 1 0 1 939 31 9 0 6 6 10 0 15 4] - [ 0 0 1 4 1 3 4 3 0 1 1 31 875 1 0 10 3 17 1 7 6] - [ 0 1 0 0 0 17 0 1 11 20 4 6 4 946 0 3 3 1 0 3 3] - [ 2 3 1 20 4 5 0 1 19 2 2 2 3 4 1048 1 3 1 5 0 4] - [ 1 0 3 2 3 2 2 0 0 0 0 4 8 3 0 991 5 10 0 5 4] - [ 4 4 0 1 1 2 2 1 1 1 0 3 0 4 0 9 1025 2 0 7 5] - [ 3 0 2 0 0 1 3 1 0 3 0 6 11 2 0 10 0 988 0 5 1] - [ 1 3 1 9 1 6 1 29 1 0 8 3 3 2 10 0 0 2 919 7 2] - [ 3 2 0 0 1 7 11 11 0 0 1 16 4 4 1 2 2 1 1 1010 3] - [ 124 150 156 116 111 316 123 211 83 95 208 143 368 357 169 148 184 113 206 327 9518]] - -2022-12-06 10:44:00,124 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:44:00,124 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:44:00,130 - - -2022-12-06 10:44:00,130 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:44:01,058 - Epoch: [49][ 10/ 1200] Overall Loss 0.270072 Objective Loss 0.270072 LR 0.001000 Time 0.092698 -2022-12-06 10:44:01,254 - Epoch: [49][ 20/ 1200] Overall Loss 0.265395 Objective Loss 0.265395 LR 0.001000 Time 0.056144 -2022-12-06 10:44:01,445 - Epoch: [49][ 30/ 1200] Overall Loss 0.260840 Objective Loss 0.260840 LR 0.001000 Time 0.043776 -2022-12-06 10:44:01,636 - Epoch: [49][ 40/ 1200] Overall Loss 0.256123 Objective Loss 0.256123 LR 0.001000 Time 0.037587 -2022-12-06 10:44:01,826 - Epoch: [49][ 50/ 1200] Overall Loss 0.267270 Objective Loss 0.267270 LR 0.001000 Time 0.033861 -2022-12-06 10:44:02,017 - Epoch: [49][ 60/ 1200] Overall Loss 0.265707 Objective Loss 0.265707 LR 0.001000 Time 0.031386 -2022-12-06 10:44:02,206 - Epoch: [49][ 70/ 1200] Overall Loss 0.265893 Objective Loss 0.265893 LR 0.001000 Time 0.029609 -2022-12-06 10:44:02,396 - Epoch: [49][ 80/ 1200] Overall Loss 0.271473 Objective Loss 0.271473 LR 0.001000 Time 0.028276 -2022-12-06 10:44:02,586 - Epoch: [49][ 90/ 1200] Overall Loss 0.269152 Objective Loss 0.269152 LR 0.001000 Time 0.027240 -2022-12-06 10:44:02,776 - Epoch: [49][ 100/ 1200] Overall Loss 0.268847 Objective Loss 0.268847 LR 0.001000 Time 0.026406 -2022-12-06 10:44:02,967 - Epoch: [49][ 110/ 1200] Overall Loss 0.269775 Objective Loss 0.269775 LR 0.001000 Time 0.025733 -2022-12-06 10:44:03,156 - Epoch: [49][ 120/ 1200] Overall Loss 0.274196 Objective Loss 0.274196 LR 0.001000 Time 0.025166 -2022-12-06 10:44:03,348 - Epoch: [49][ 130/ 1200] Overall Loss 0.272894 Objective Loss 0.272894 LR 0.001000 Time 0.024697 -2022-12-06 10:44:03,538 - Epoch: [49][ 140/ 1200] Overall Loss 0.274563 Objective Loss 0.274563 LR 0.001000 Time 0.024288 -2022-12-06 10:44:03,728 - Epoch: [49][ 150/ 1200] Overall Loss 0.275536 Objective Loss 0.275536 LR 0.001000 Time 0.023934 -2022-12-06 10:44:03,918 - Epoch: [49][ 160/ 1200] Overall Loss 0.273983 Objective Loss 0.273983 LR 0.001000 Time 0.023623 -2022-12-06 10:44:04,109 - Epoch: [49][ 170/ 1200] Overall Loss 0.275970 Objective Loss 0.275970 LR 0.001000 Time 0.023352 -2022-12-06 10:44:04,299 - Epoch: [49][ 180/ 1200] Overall Loss 0.275467 Objective Loss 0.275467 LR 0.001000 Time 0.023108 -2022-12-06 10:44:04,490 - Epoch: [49][ 190/ 1200] Overall Loss 0.274771 Objective Loss 0.274771 LR 0.001000 Time 0.022892 -2022-12-06 10:44:04,679 - Epoch: [49][ 200/ 1200] Overall Loss 0.274503 Objective Loss 0.274503 LR 0.001000 Time 0.022692 -2022-12-06 10:44:04,869 - Epoch: [49][ 210/ 1200] Overall Loss 0.274495 Objective Loss 0.274495 LR 0.001000 Time 0.022514 -2022-12-06 10:44:05,059 - Epoch: [49][ 220/ 1200] Overall Loss 0.274458 Objective Loss 0.274458 LR 0.001000 Time 0.022351 -2022-12-06 10:44:05,250 - Epoch: [49][ 230/ 1200] Overall Loss 0.274834 Objective Loss 0.274834 LR 0.001000 Time 0.022206 -2022-12-06 10:44:05,440 - Epoch: [49][ 240/ 1200] Overall Loss 0.273515 Objective Loss 0.273515 LR 0.001000 Time 0.022070 -2022-12-06 10:44:05,630 - Epoch: [49][ 250/ 1200] Overall Loss 0.272441 Objective Loss 0.272441 LR 0.001000 Time 0.021947 -2022-12-06 10:44:05,820 - Epoch: [49][ 260/ 1200] Overall Loss 0.272846 Objective Loss 0.272846 LR 0.001000 Time 0.021832 -2022-12-06 10:44:06,010 - Epoch: [49][ 270/ 1200] Overall Loss 0.272293 Objective Loss 0.272293 LR 0.001000 Time 0.021726 -2022-12-06 10:44:06,201 - Epoch: [49][ 280/ 1200] Overall Loss 0.272582 Objective Loss 0.272582 LR 0.001000 Time 0.021627 -2022-12-06 10:44:06,391 - Epoch: [49][ 290/ 1200] Overall Loss 0.271724 Objective Loss 0.271724 LR 0.001000 Time 0.021534 -2022-12-06 10:44:06,581 - Epoch: [49][ 300/ 1200] Overall Loss 0.272917 Objective Loss 0.272917 LR 0.001000 Time 0.021448 -2022-12-06 10:44:06,771 - Epoch: [49][ 310/ 1200] Overall Loss 0.271829 Objective Loss 0.271829 LR 0.001000 Time 0.021368 -2022-12-06 10:44:06,960 - Epoch: [49][ 320/ 1200] Overall Loss 0.271495 Objective Loss 0.271495 LR 0.001000 Time 0.021291 -2022-12-06 10:44:07,150 - Epoch: [49][ 330/ 1200] Overall Loss 0.272677 Objective Loss 0.272677 LR 0.001000 Time 0.021220 -2022-12-06 10:44:07,342 - Epoch: [49][ 340/ 1200] Overall Loss 0.271893 Objective Loss 0.271893 LR 0.001000 Time 0.021157 -2022-12-06 10:44:07,532 - Epoch: [49][ 350/ 1200] Overall Loss 0.271811 Objective Loss 0.271811 LR 0.001000 Time 0.021095 -2022-12-06 10:44:07,722 - Epoch: [49][ 360/ 1200] Overall Loss 0.271435 Objective Loss 0.271435 LR 0.001000 Time 0.021036 -2022-12-06 10:44:07,913 - Epoch: [49][ 370/ 1200] Overall Loss 0.271359 Objective Loss 0.271359 LR 0.001000 Time 0.020982 -2022-12-06 10:44:08,104 - Epoch: [49][ 380/ 1200] Overall Loss 0.271234 Objective Loss 0.271234 LR 0.001000 Time 0.020930 -2022-12-06 10:44:08,294 - Epoch: [49][ 390/ 1200] Overall Loss 0.270704 Objective Loss 0.270704 LR 0.001000 Time 0.020880 -2022-12-06 10:44:08,484 - Epoch: [49][ 400/ 1200] Overall Loss 0.271756 Objective Loss 0.271756 LR 0.001000 Time 0.020832 -2022-12-06 10:44:08,675 - Epoch: [49][ 410/ 1200] Overall Loss 0.272000 Objective Loss 0.272000 LR 0.001000 Time 0.020788 -2022-12-06 10:44:08,865 - Epoch: [49][ 420/ 1200] Overall Loss 0.271788 Objective Loss 0.271788 LR 0.001000 Time 0.020745 -2022-12-06 10:44:09,056 - Epoch: [49][ 430/ 1200] Overall Loss 0.272282 Objective Loss 0.272282 LR 0.001000 Time 0.020705 -2022-12-06 10:44:09,247 - Epoch: [49][ 440/ 1200] Overall Loss 0.272647 Objective Loss 0.272647 LR 0.001000 Time 0.020667 -2022-12-06 10:44:09,437 - Epoch: [49][ 450/ 1200] Overall Loss 0.272847 Objective Loss 0.272847 LR 0.001000 Time 0.020628 -2022-12-06 10:44:09,627 - Epoch: [49][ 460/ 1200] Overall Loss 0.273379 Objective Loss 0.273379 LR 0.001000 Time 0.020591 -2022-12-06 10:44:09,816 - Epoch: [49][ 470/ 1200] Overall Loss 0.273146 Objective Loss 0.273146 LR 0.001000 Time 0.020556 -2022-12-06 10:44:10,007 - Epoch: [49][ 480/ 1200] Overall Loss 0.273416 Objective Loss 0.273416 LR 0.001000 Time 0.020523 -2022-12-06 10:44:10,197 - Epoch: [49][ 490/ 1200] Overall Loss 0.273257 Objective Loss 0.273257 LR 0.001000 Time 0.020492 -2022-12-06 10:44:10,387 - Epoch: [49][ 500/ 1200] Overall Loss 0.273068 Objective Loss 0.273068 LR 0.001000 Time 0.020461 -2022-12-06 10:44:10,578 - Epoch: [49][ 510/ 1200] Overall Loss 0.272703 Objective Loss 0.272703 LR 0.001000 Time 0.020433 -2022-12-06 10:44:10,769 - Epoch: [49][ 520/ 1200] Overall Loss 0.272333 Objective Loss 0.272333 LR 0.001000 Time 0.020405 -2022-12-06 10:44:10,959 - Epoch: [49][ 530/ 1200] Overall Loss 0.272377 Objective Loss 0.272377 LR 0.001000 Time 0.020379 -2022-12-06 10:44:11,150 - Epoch: [49][ 540/ 1200] Overall Loss 0.272709 Objective Loss 0.272709 LR 0.001000 Time 0.020353 -2022-12-06 10:44:11,341 - Epoch: [49][ 550/ 1200] Overall Loss 0.272845 Objective Loss 0.272845 LR 0.001000 Time 0.020329 -2022-12-06 10:44:11,531 - Epoch: [49][ 560/ 1200] Overall Loss 0.273277 Objective Loss 0.273277 LR 0.001000 Time 0.020305 -2022-12-06 10:44:11,722 - Epoch: [49][ 570/ 1200] Overall Loss 0.273311 Objective Loss 0.273311 LR 0.001000 Time 0.020283 -2022-12-06 10:44:11,912 - Epoch: [49][ 580/ 1200] Overall Loss 0.273248 Objective Loss 0.273248 LR 0.001000 Time 0.020261 -2022-12-06 10:44:12,102 - Epoch: [49][ 590/ 1200] Overall Loss 0.273095 Objective Loss 0.273095 LR 0.001000 Time 0.020238 -2022-12-06 10:44:12,292 - Epoch: [49][ 600/ 1200] Overall Loss 0.272985 Objective Loss 0.272985 LR 0.001000 Time 0.020216 -2022-12-06 10:44:12,482 - Epoch: [49][ 610/ 1200] Overall Loss 0.272708 Objective Loss 0.272708 LR 0.001000 Time 0.020196 -2022-12-06 10:44:12,672 - Epoch: [49][ 620/ 1200] Overall Loss 0.272420 Objective Loss 0.272420 LR 0.001000 Time 0.020176 -2022-12-06 10:44:12,863 - Epoch: [49][ 630/ 1200] Overall Loss 0.272582 Objective Loss 0.272582 LR 0.001000 Time 0.020157 -2022-12-06 10:44:13,054 - Epoch: [49][ 640/ 1200] Overall Loss 0.272739 Objective Loss 0.272739 LR 0.001000 Time 0.020140 -2022-12-06 10:44:13,245 - Epoch: [49][ 650/ 1200] Overall Loss 0.272604 Objective Loss 0.272604 LR 0.001000 Time 0.020123 -2022-12-06 10:44:13,435 - Epoch: [49][ 660/ 1200] Overall Loss 0.272400 Objective Loss 0.272400 LR 0.001000 Time 0.020105 -2022-12-06 10:44:13,625 - Epoch: [49][ 670/ 1200] Overall Loss 0.272732 Objective Loss 0.272732 LR 0.001000 Time 0.020088 -2022-12-06 10:44:13,815 - Epoch: [49][ 680/ 1200] Overall Loss 0.272544 Objective Loss 0.272544 LR 0.001000 Time 0.020071 -2022-12-06 10:44:14,005 - Epoch: [49][ 690/ 1200] Overall Loss 0.272657 Objective Loss 0.272657 LR 0.001000 Time 0.020055 -2022-12-06 10:44:14,194 - Epoch: [49][ 700/ 1200] Overall Loss 0.272489 Objective Loss 0.272489 LR 0.001000 Time 0.020038 -2022-12-06 10:44:14,385 - Epoch: [49][ 710/ 1200] Overall Loss 0.272270 Objective Loss 0.272270 LR 0.001000 Time 0.020024 -2022-12-06 10:44:14,575 - Epoch: [49][ 720/ 1200] Overall Loss 0.272317 Objective Loss 0.272317 LR 0.001000 Time 0.020008 -2022-12-06 10:44:14,765 - Epoch: [49][ 730/ 1200] Overall Loss 0.272056 Objective Loss 0.272056 LR 0.001000 Time 0.019994 -2022-12-06 10:44:14,955 - Epoch: [49][ 740/ 1200] Overall Loss 0.272211 Objective Loss 0.272211 LR 0.001000 Time 0.019980 -2022-12-06 10:44:15,146 - Epoch: [49][ 750/ 1200] Overall Loss 0.272479 Objective Loss 0.272479 LR 0.001000 Time 0.019968 -2022-12-06 10:44:15,336 - Epoch: [49][ 760/ 1200] Overall Loss 0.272315 Objective Loss 0.272315 LR 0.001000 Time 0.019954 -2022-12-06 10:44:15,526 - Epoch: [49][ 770/ 1200] Overall Loss 0.272685 Objective Loss 0.272685 LR 0.001000 Time 0.019942 -2022-12-06 10:44:15,717 - Epoch: [49][ 780/ 1200] Overall Loss 0.273182 Objective Loss 0.273182 LR 0.001000 Time 0.019930 -2022-12-06 10:44:15,908 - Epoch: [49][ 790/ 1200] Overall Loss 0.273260 Objective Loss 0.273260 LR 0.001000 Time 0.019919 -2022-12-06 10:44:16,099 - Epoch: [49][ 800/ 1200] Overall Loss 0.273586 Objective Loss 0.273586 LR 0.001000 Time 0.019908 -2022-12-06 10:44:16,290 - Epoch: [49][ 810/ 1200] Overall Loss 0.273107 Objective Loss 0.273107 LR 0.001000 Time 0.019897 -2022-12-06 10:44:16,480 - Epoch: [49][ 820/ 1200] Overall Loss 0.273082 Objective Loss 0.273082 LR 0.001000 Time 0.019885 -2022-12-06 10:44:16,670 - Epoch: [49][ 830/ 1200] Overall Loss 0.273258 Objective Loss 0.273258 LR 0.001000 Time 0.019874 -2022-12-06 10:44:16,860 - Epoch: [49][ 840/ 1200] Overall Loss 0.273345 Objective Loss 0.273345 LR 0.001000 Time 0.019864 -2022-12-06 10:44:17,050 - Epoch: [49][ 850/ 1200] Overall Loss 0.273620 Objective Loss 0.273620 LR 0.001000 Time 0.019853 -2022-12-06 10:44:17,241 - Epoch: [49][ 860/ 1200] Overall Loss 0.273659 Objective Loss 0.273659 LR 0.001000 Time 0.019843 -2022-12-06 10:44:17,431 - Epoch: [49][ 870/ 1200] Overall Loss 0.273636 Objective Loss 0.273636 LR 0.001000 Time 0.019833 -2022-12-06 10:44:17,621 - Epoch: [49][ 880/ 1200] Overall Loss 0.273731 Objective Loss 0.273731 LR 0.001000 Time 0.019823 -2022-12-06 10:44:17,812 - Epoch: [49][ 890/ 1200] Overall Loss 0.273217 Objective Loss 0.273217 LR 0.001000 Time 0.019814 -2022-12-06 10:44:18,003 - Epoch: [49][ 900/ 1200] Overall Loss 0.273501 Objective Loss 0.273501 LR 0.001000 Time 0.019805 -2022-12-06 10:44:18,194 - Epoch: [49][ 910/ 1200] Overall Loss 0.273860 Objective Loss 0.273860 LR 0.001000 Time 0.019797 -2022-12-06 10:44:18,384 - Epoch: [49][ 920/ 1200] Overall Loss 0.273869 Objective Loss 0.273869 LR 0.001000 Time 0.019788 -2022-12-06 10:44:18,574 - Epoch: [49][ 930/ 1200] Overall Loss 0.273994 Objective Loss 0.273994 LR 0.001000 Time 0.019779 -2022-12-06 10:44:18,765 - Epoch: [49][ 940/ 1200] Overall Loss 0.274036 Objective Loss 0.274036 LR 0.001000 Time 0.019771 -2022-12-06 10:44:18,956 - Epoch: [49][ 950/ 1200] Overall Loss 0.274008 Objective Loss 0.274008 LR 0.001000 Time 0.019763 -2022-12-06 10:44:19,147 - Epoch: [49][ 960/ 1200] Overall Loss 0.273982 Objective Loss 0.273982 LR 0.001000 Time 0.019756 -2022-12-06 10:44:19,338 - Epoch: [49][ 970/ 1200] Overall Loss 0.273958 Objective Loss 0.273958 LR 0.001000 Time 0.019748 -2022-12-06 10:44:19,528 - Epoch: [49][ 980/ 1200] Overall Loss 0.273468 Objective Loss 0.273468 LR 0.001000 Time 0.019740 -2022-12-06 10:44:19,719 - Epoch: [49][ 990/ 1200] Overall Loss 0.273437 Objective Loss 0.273437 LR 0.001000 Time 0.019733 -2022-12-06 10:44:19,909 - Epoch: [49][ 1000/ 1200] Overall Loss 0.273560 Objective Loss 0.273560 LR 0.001000 Time 0.019725 -2022-12-06 10:44:20,099 - Epoch: [49][ 1010/ 1200] Overall Loss 0.273689 Objective Loss 0.273689 LR 0.001000 Time 0.019718 -2022-12-06 10:44:20,291 - Epoch: [49][ 1020/ 1200] Overall Loss 0.273955 Objective Loss 0.273955 LR 0.001000 Time 0.019712 -2022-12-06 10:44:20,481 - Epoch: [49][ 1030/ 1200] Overall Loss 0.274040 Objective Loss 0.274040 LR 0.001000 Time 0.019705 -2022-12-06 10:44:20,671 - Epoch: [49][ 1040/ 1200] Overall Loss 0.273906 Objective Loss 0.273906 LR 0.001000 Time 0.019698 -2022-12-06 10:44:20,862 - Epoch: [49][ 1050/ 1200] Overall Loss 0.274135 Objective Loss 0.274135 LR 0.001000 Time 0.019691 -2022-12-06 10:44:21,053 - Epoch: [49][ 1060/ 1200] Overall Loss 0.274527 Objective Loss 0.274527 LR 0.001000 Time 0.019685 -2022-12-06 10:44:21,244 - Epoch: [49][ 1070/ 1200] Overall Loss 0.274342 Objective Loss 0.274342 LR 0.001000 Time 0.019679 -2022-12-06 10:44:21,434 - Epoch: [49][ 1080/ 1200] Overall Loss 0.273980 Objective Loss 0.273980 LR 0.001000 Time 0.019673 -2022-12-06 10:44:21,625 - Epoch: [49][ 1090/ 1200] Overall Loss 0.273867 Objective Loss 0.273867 LR 0.001000 Time 0.019667 -2022-12-06 10:44:21,816 - Epoch: [49][ 1100/ 1200] Overall Loss 0.273991 Objective Loss 0.273991 LR 0.001000 Time 0.019661 -2022-12-06 10:44:22,008 - Epoch: [49][ 1110/ 1200] Overall Loss 0.274128 Objective Loss 0.274128 LR 0.001000 Time 0.019656 -2022-12-06 10:44:22,198 - Epoch: [49][ 1120/ 1200] Overall Loss 0.274320 Objective Loss 0.274320 LR 0.001000 Time 0.019650 -2022-12-06 10:44:22,389 - Epoch: [49][ 1130/ 1200] Overall Loss 0.274390 Objective Loss 0.274390 LR 0.001000 Time 0.019645 -2022-12-06 10:44:22,579 - Epoch: [49][ 1140/ 1200] Overall Loss 0.274619 Objective Loss 0.274619 LR 0.001000 Time 0.019639 -2022-12-06 10:44:22,770 - Epoch: [49][ 1150/ 1200] Overall Loss 0.274605 Objective Loss 0.274605 LR 0.001000 Time 0.019633 -2022-12-06 10:44:22,961 - Epoch: [49][ 1160/ 1200] Overall Loss 0.274820 Objective Loss 0.274820 LR 0.001000 Time 0.019628 -2022-12-06 10:44:23,152 - Epoch: [49][ 1170/ 1200] Overall Loss 0.275108 Objective Loss 0.275108 LR 0.001000 Time 0.019623 -2022-12-06 10:44:23,343 - Epoch: [49][ 1180/ 1200] Overall Loss 0.275180 Objective Loss 0.275180 LR 0.001000 Time 0.019618 -2022-12-06 10:44:23,533 - Epoch: [49][ 1190/ 1200] Overall Loss 0.275264 Objective Loss 0.275264 LR 0.001000 Time 0.019612 -2022-12-06 10:44:23,766 - Epoch: [49][ 1200/ 1200] Overall Loss 0.275371 Objective Loss 0.275371 Top1 83.263598 Top5 98.117155 LR 0.001000 Time 0.019643 -2022-12-06 10:44:23,860 - --- validate (epoch=49)----------- -2022-12-06 10:44:23,860 - 34129 samples (256 per mini-batch) -2022-12-06 10:44:24,422 - Epoch: [49][ 10/ 134] Loss 0.285555 Top1 83.593750 Top5 97.773438 -2022-12-06 10:44:24,553 - Epoch: [49][ 20/ 134] Loss 0.302525 Top1 83.886719 Top5 97.734375 -2022-12-06 10:44:24,686 - Epoch: [49][ 30/ 134] Loss 0.302270 Top1 83.932292 Top5 97.708333 -2022-12-06 10:44:24,819 - Epoch: [49][ 40/ 134] Loss 0.302771 Top1 83.681641 Top5 97.841797 -2022-12-06 10:44:24,950 - Epoch: [49][ 50/ 134] Loss 0.303160 Top1 83.546875 Top5 97.804688 -2022-12-06 10:44:25,083 - Epoch: [49][ 60/ 134] Loss 0.306565 Top1 83.476562 Top5 97.682292 -2022-12-06 10:44:25,215 - Epoch: [49][ 70/ 134] Loss 0.310067 Top1 83.426339 Top5 97.622768 -2022-12-06 10:44:25,348 - Epoch: [49][ 80/ 134] Loss 0.304766 Top1 83.452148 Top5 97.617188 -2022-12-06 10:44:25,480 - Epoch: [49][ 90/ 134] Loss 0.300499 Top1 83.585069 Top5 97.656250 -2022-12-06 10:44:25,609 - Epoch: [49][ 100/ 134] Loss 0.298377 Top1 83.679688 Top5 97.664062 -2022-12-06 10:44:25,736 - Epoch: [49][ 110/ 134] Loss 0.298793 Top1 83.742898 Top5 97.649148 -2022-12-06 10:44:25,864 - Epoch: [49][ 120/ 134] Loss 0.301318 Top1 83.655599 Top5 97.636719 -2022-12-06 10:44:25,991 - Epoch: [49][ 130/ 134] Loss 0.302368 Top1 83.602764 Top5 97.626202 -2022-12-06 10:44:26,028 - Epoch: [49][ 134/ 134] Loss 0.302830 Top1 83.565296 Top5 97.626652 -2022-12-06 10:44:26,135 - ==> Top1: 83.565 Top5: 97.627 Loss: 0.303 - -2022-12-06 10:44:26,136 - ==> Confusion: -[[ 892 2 3 1 7 6 0 1 3 71 0 1 2 1 1 3 1 0 0 0 1] - [ 1 928 1 3 8 31 3 15 0 2 9 6 0 3 0 2 0 1 5 0 9] - [ 5 5 998 12 1 3 30 11 0 4 4 4 0 6 0 4 2 3 3 1 7] - [ 2 3 31 906 0 9 1 1 4 0 15 0 4 2 18 5 0 3 10 0 6] - [ 14 6 3 1 945 8 2 3 1 8 2 1 1 3 4 5 7 2 0 0 4] - [ 1 18 1 2 3 989 3 11 4 2 2 11 4 8 2 1 0 1 1 1 4] - [ 0 0 12 5 1 4 1067 6 0 0 7 1 0 4 0 6 0 1 0 2 2] - [ 0 14 6 2 3 39 13 936 1 0 3 6 1 1 0 1 0 2 16 8 2] - [ 3 2 0 0 0 3 1 0 962 56 8 2 2 12 7 0 1 0 2 3 0] - [ 72 0 2 0 3 4 1 2 18 883 1 2 0 7 1 1 0 0 0 0 4] - [ 0 2 5 5 1 4 2 0 17 2 946 2 1 20 3 0 1 0 5 2 1] - [ 3 1 1 0 1 14 5 3 2 0 1 962 15 12 0 8 4 7 2 9 1] - [ 2 0 2 3 0 8 4 2 0 0 2 60 845 4 1 8 1 15 1 4 7] - [ 0 1 1 0 2 16 0 4 11 14 5 5 3 952 0 3 2 0 0 1 3] - [ 13 4 5 6 2 4 0 0 23 10 5 1 4 5 1029 1 1 2 7 1 7] - [ 2 0 0 1 2 2 7 1 0 0 0 6 2 6 0 998 3 9 0 3 1] - [ 4 3 1 1 3 3 1 1 2 0 0 4 1 3 2 16 1013 2 0 5 7] - [ 3 0 2 5 0 1 3 1 2 3 1 16 27 2 2 13 1 951 0 1 2] - [ 3 4 6 8 1 5 1 32 2 1 10 1 3 0 8 1 2 0 915 2 3] - [ 1 2 2 1 0 14 6 12 0 0 2 10 4 7 0 4 3 4 0 1004 4] - [ 188 212 261 102 117 294 108 165 97 149 246 152 400 373 149 160 159 68 148 285 9393]] - -2022-12-06 10:44:26,706 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:44:26,706 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:44:26,711 - - -2022-12-06 10:44:26,712 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:44:27,653 - Epoch: [50][ 10/ 1200] Overall Loss 0.273136 Objective Loss 0.273136 LR 0.001000 Time 0.094110 -2022-12-06 10:44:27,873 - Epoch: [50][ 20/ 1200] Overall Loss 0.258791 Objective Loss 0.258791 LR 0.001000 Time 0.058010 -2022-12-06 10:44:28,081 - Epoch: [50][ 30/ 1200] Overall Loss 0.262598 Objective Loss 0.262598 LR 0.001000 Time 0.045602 -2022-12-06 10:44:28,291 - Epoch: [50][ 40/ 1200] Overall Loss 0.261995 Objective Loss 0.261995 LR 0.001000 Time 0.039425 -2022-12-06 10:44:28,493 - Epoch: [50][ 50/ 1200] Overall Loss 0.266596 Objective Loss 0.266596 LR 0.001000 Time 0.035579 -2022-12-06 10:44:28,701 - Epoch: [50][ 60/ 1200] Overall Loss 0.268374 Objective Loss 0.268374 LR 0.001000 Time 0.033094 -2022-12-06 10:44:28,904 - Epoch: [50][ 70/ 1200] Overall Loss 0.265738 Objective Loss 0.265738 LR 0.001000 Time 0.031258 -2022-12-06 10:44:29,111 - Epoch: [50][ 80/ 1200] Overall Loss 0.263061 Objective Loss 0.263061 LR 0.001000 Time 0.029931 -2022-12-06 10:44:29,314 - Epoch: [50][ 90/ 1200] Overall Loss 0.263822 Objective Loss 0.263822 LR 0.001000 Time 0.028855 -2022-12-06 10:44:29,521 - Epoch: [50][ 100/ 1200] Overall Loss 0.268408 Objective Loss 0.268408 LR 0.001000 Time 0.028035 -2022-12-06 10:44:29,723 - Epoch: [50][ 110/ 1200] Overall Loss 0.267518 Objective Loss 0.267518 LR 0.001000 Time 0.027324 -2022-12-06 10:44:29,930 - Epoch: [50][ 120/ 1200] Overall Loss 0.268017 Objective Loss 0.268017 LR 0.001000 Time 0.026761 -2022-12-06 10:44:30,132 - Epoch: [50][ 130/ 1200] Overall Loss 0.268343 Objective Loss 0.268343 LR 0.001000 Time 0.026255 -2022-12-06 10:44:30,339 - Epoch: [50][ 140/ 1200] Overall Loss 0.268851 Objective Loss 0.268851 LR 0.001000 Time 0.025855 -2022-12-06 10:44:30,542 - Epoch: [50][ 150/ 1200] Overall Loss 0.268027 Objective Loss 0.268027 LR 0.001000 Time 0.025480 -2022-12-06 10:44:30,749 - Epoch: [50][ 160/ 1200] Overall Loss 0.268465 Objective Loss 0.268465 LR 0.001000 Time 0.025180 -2022-12-06 10:44:30,952 - Epoch: [50][ 170/ 1200] Overall Loss 0.268420 Objective Loss 0.268420 LR 0.001000 Time 0.024887 -2022-12-06 10:44:31,159 - Epoch: [50][ 180/ 1200] Overall Loss 0.268064 Objective Loss 0.268064 LR 0.001000 Time 0.024654 -2022-12-06 10:44:31,362 - Epoch: [50][ 190/ 1200] Overall Loss 0.268653 Objective Loss 0.268653 LR 0.001000 Time 0.024419 -2022-12-06 10:44:31,569 - Epoch: [50][ 200/ 1200] Overall Loss 0.267356 Objective Loss 0.267356 LR 0.001000 Time 0.024234 -2022-12-06 10:44:31,772 - Epoch: [50][ 210/ 1200] Overall Loss 0.266434 Objective Loss 0.266434 LR 0.001000 Time 0.024041 -2022-12-06 10:44:31,979 - Epoch: [50][ 220/ 1200] Overall Loss 0.266104 Objective Loss 0.266104 LR 0.001000 Time 0.023889 -2022-12-06 10:44:32,182 - Epoch: [50][ 230/ 1200] Overall Loss 0.267507 Objective Loss 0.267507 LR 0.001000 Time 0.023729 -2022-12-06 10:44:32,389 - Epoch: [50][ 240/ 1200] Overall Loss 0.267333 Objective Loss 0.267333 LR 0.001000 Time 0.023600 -2022-12-06 10:44:32,592 - Epoch: [50][ 250/ 1200] Overall Loss 0.266128 Objective Loss 0.266128 LR 0.001000 Time 0.023466 -2022-12-06 10:44:32,799 - Epoch: [50][ 260/ 1200] Overall Loss 0.265903 Objective Loss 0.265903 LR 0.001000 Time 0.023358 -2022-12-06 10:44:33,002 - Epoch: [50][ 270/ 1200] Overall Loss 0.266697 Objective Loss 0.266697 LR 0.001000 Time 0.023243 -2022-12-06 10:44:33,209 - Epoch: [50][ 280/ 1200] Overall Loss 0.267280 Objective Loss 0.267280 LR 0.001000 Time 0.023151 -2022-12-06 10:44:33,412 - Epoch: [50][ 290/ 1200] Overall Loss 0.267640 Objective Loss 0.267640 LR 0.001000 Time 0.023050 -2022-12-06 10:44:33,619 - Epoch: [50][ 300/ 1200] Overall Loss 0.268109 Objective Loss 0.268109 LR 0.001000 Time 0.022971 -2022-12-06 10:44:33,822 - Epoch: [50][ 310/ 1200] Overall Loss 0.267753 Objective Loss 0.267753 LR 0.001000 Time 0.022881 -2022-12-06 10:44:34,029 - Epoch: [50][ 320/ 1200] Overall Loss 0.268850 Objective Loss 0.268850 LR 0.001000 Time 0.022812 -2022-12-06 10:44:34,231 - Epoch: [50][ 330/ 1200] Overall Loss 0.268995 Objective Loss 0.268995 LR 0.001000 Time 0.022732 -2022-12-06 10:44:34,439 - Epoch: [50][ 340/ 1200] Overall Loss 0.268731 Objective Loss 0.268731 LR 0.001000 Time 0.022673 -2022-12-06 10:44:34,642 - Epoch: [50][ 350/ 1200] Overall Loss 0.268596 Objective Loss 0.268596 LR 0.001000 Time 0.022605 -2022-12-06 10:44:34,850 - Epoch: [50][ 360/ 1200] Overall Loss 0.268524 Objective Loss 0.268524 LR 0.001000 Time 0.022551 -2022-12-06 10:44:35,052 - Epoch: [50][ 370/ 1200] Overall Loss 0.268955 Objective Loss 0.268955 LR 0.001000 Time 0.022487 -2022-12-06 10:44:35,260 - Epoch: [50][ 380/ 1200] Overall Loss 0.269685 Objective Loss 0.269685 LR 0.001000 Time 0.022441 -2022-12-06 10:44:35,463 - Epoch: [50][ 390/ 1200] Overall Loss 0.269557 Objective Loss 0.269557 LR 0.001000 Time 0.022384 -2022-12-06 10:44:35,669 - Epoch: [50][ 400/ 1200] Overall Loss 0.269454 Objective Loss 0.269454 LR 0.001000 Time 0.022340 -2022-12-06 10:44:35,873 - Epoch: [50][ 410/ 1200] Overall Loss 0.269789 Objective Loss 0.269789 LR 0.001000 Time 0.022290 -2022-12-06 10:44:36,080 - Epoch: [50][ 420/ 1200] Overall Loss 0.269790 Objective Loss 0.269790 LR 0.001000 Time 0.022251 -2022-12-06 10:44:36,283 - Epoch: [50][ 430/ 1200] Overall Loss 0.269995 Objective Loss 0.269995 LR 0.001000 Time 0.022203 -2022-12-06 10:44:36,489 - Epoch: [50][ 440/ 1200] Overall Loss 0.270361 Objective Loss 0.270361 LR 0.001000 Time 0.022167 -2022-12-06 10:44:36,692 - Epoch: [50][ 450/ 1200] Overall Loss 0.270464 Objective Loss 0.270464 LR 0.001000 Time 0.022124 -2022-12-06 10:44:36,899 - Epoch: [50][ 460/ 1200] Overall Loss 0.270865 Objective Loss 0.270865 LR 0.001000 Time 0.022092 -2022-12-06 10:44:37,101 - Epoch: [50][ 470/ 1200] Overall Loss 0.270342 Objective Loss 0.270342 LR 0.001000 Time 0.022051 -2022-12-06 10:44:37,308 - Epoch: [50][ 480/ 1200] Overall Loss 0.270597 Objective Loss 0.270597 LR 0.001000 Time 0.022020 -2022-12-06 10:44:37,510 - Epoch: [50][ 490/ 1200] Overall Loss 0.270839 Objective Loss 0.270839 LR 0.001000 Time 0.021982 -2022-12-06 10:44:37,717 - Epoch: [50][ 500/ 1200] Overall Loss 0.270815 Objective Loss 0.270815 LR 0.001000 Time 0.021956 -2022-12-06 10:44:37,920 - Epoch: [50][ 510/ 1200] Overall Loss 0.271332 Objective Loss 0.271332 LR 0.001000 Time 0.021921 -2022-12-06 10:44:38,127 - Epoch: [50][ 520/ 1200] Overall Loss 0.271388 Objective Loss 0.271388 LR 0.001000 Time 0.021898 -2022-12-06 10:44:38,330 - Epoch: [50][ 530/ 1200] Overall Loss 0.271905 Objective Loss 0.271905 LR 0.001000 Time 0.021866 -2022-12-06 10:44:38,537 - Epoch: [50][ 540/ 1200] Overall Loss 0.272308 Objective Loss 0.272308 LR 0.001000 Time 0.021843 -2022-12-06 10:44:38,732 - Epoch: [50][ 550/ 1200] Overall Loss 0.272315 Objective Loss 0.272315 LR 0.001000 Time 0.021801 -2022-12-06 10:44:38,937 - Epoch: [50][ 560/ 1200] Overall Loss 0.272621 Objective Loss 0.272621 LR 0.001000 Time 0.021776 -2022-12-06 10:44:39,138 - Epoch: [50][ 570/ 1200] Overall Loss 0.272472 Objective Loss 0.272472 LR 0.001000 Time 0.021746 -2022-12-06 10:44:39,343 - Epoch: [50][ 580/ 1200] Overall Loss 0.272910 Objective Loss 0.272910 LR 0.001000 Time 0.021723 -2022-12-06 10:44:39,545 - Epoch: [50][ 590/ 1200] Overall Loss 0.273271 Objective Loss 0.273271 LR 0.001000 Time 0.021695 -2022-12-06 10:44:39,749 - Epoch: [50][ 600/ 1200] Overall Loss 0.273483 Objective Loss 0.273483 LR 0.001000 Time 0.021674 -2022-12-06 10:44:39,951 - Epoch: [50][ 610/ 1200] Overall Loss 0.273586 Objective Loss 0.273586 LR 0.001000 Time 0.021648 -2022-12-06 10:44:40,155 - Epoch: [50][ 620/ 1200] Overall Loss 0.273379 Objective Loss 0.273379 LR 0.001000 Time 0.021628 -2022-12-06 10:44:40,357 - Epoch: [50][ 630/ 1200] Overall Loss 0.273597 Objective Loss 0.273597 LR 0.001000 Time 0.021604 -2022-12-06 10:44:40,562 - Epoch: [50][ 640/ 1200] Overall Loss 0.273828 Objective Loss 0.273828 LR 0.001000 Time 0.021586 -2022-12-06 10:44:40,764 - Epoch: [50][ 650/ 1200] Overall Loss 0.273940 Objective Loss 0.273940 LR 0.001000 Time 0.021563 -2022-12-06 10:44:40,968 - Epoch: [50][ 660/ 1200] Overall Loss 0.273741 Objective Loss 0.273741 LR 0.001000 Time 0.021545 -2022-12-06 10:44:41,170 - Epoch: [50][ 670/ 1200] Overall Loss 0.273801 Objective Loss 0.273801 LR 0.001000 Time 0.021524 -2022-12-06 10:44:41,374 - Epoch: [50][ 680/ 1200] Overall Loss 0.273933 Objective Loss 0.273933 LR 0.001000 Time 0.021507 -2022-12-06 10:44:41,576 - Epoch: [50][ 690/ 1200] Overall Loss 0.273893 Objective Loss 0.273893 LR 0.001000 Time 0.021487 -2022-12-06 10:44:41,780 - Epoch: [50][ 700/ 1200] Overall Loss 0.274172 Objective Loss 0.274172 LR 0.001000 Time 0.021471 -2022-12-06 10:44:41,971 - Epoch: [50][ 710/ 1200] Overall Loss 0.274100 Objective Loss 0.274100 LR 0.001000 Time 0.021437 -2022-12-06 10:44:42,162 - Epoch: [50][ 720/ 1200] Overall Loss 0.273926 Objective Loss 0.273926 LR 0.001000 Time 0.021403 -2022-12-06 10:44:42,353 - Epoch: [50][ 730/ 1200] Overall Loss 0.273685 Objective Loss 0.273685 LR 0.001000 Time 0.021370 -2022-12-06 10:44:42,544 - Epoch: [50][ 740/ 1200] Overall Loss 0.273328 Objective Loss 0.273328 LR 0.001000 Time 0.021339 -2022-12-06 10:44:42,734 - Epoch: [50][ 750/ 1200] Overall Loss 0.273450 Objective Loss 0.273450 LR 0.001000 Time 0.021307 -2022-12-06 10:44:42,925 - Epoch: [50][ 760/ 1200] Overall Loss 0.273508 Objective Loss 0.273508 LR 0.001000 Time 0.021277 -2022-12-06 10:44:43,115 - Epoch: [50][ 770/ 1200] Overall Loss 0.273520 Objective Loss 0.273520 LR 0.001000 Time 0.021248 -2022-12-06 10:44:43,306 - Epoch: [50][ 780/ 1200] Overall Loss 0.273123 Objective Loss 0.273123 LR 0.001000 Time 0.021219 -2022-12-06 10:44:43,496 - Epoch: [50][ 790/ 1200] Overall Loss 0.273266 Objective Loss 0.273266 LR 0.001000 Time 0.021191 -2022-12-06 10:44:43,687 - Epoch: [50][ 800/ 1200] Overall Loss 0.273380 Objective Loss 0.273380 LR 0.001000 Time 0.021164 -2022-12-06 10:44:43,878 - Epoch: [50][ 810/ 1200] Overall Loss 0.273510 Objective Loss 0.273510 LR 0.001000 Time 0.021138 -2022-12-06 10:44:44,069 - Epoch: [50][ 820/ 1200] Overall Loss 0.273621 Objective Loss 0.273621 LR 0.001000 Time 0.021112 -2022-12-06 10:44:44,259 - Epoch: [50][ 830/ 1200] Overall Loss 0.273383 Objective Loss 0.273383 LR 0.001000 Time 0.021086 -2022-12-06 10:44:44,449 - Epoch: [50][ 840/ 1200] Overall Loss 0.273398 Objective Loss 0.273398 LR 0.001000 Time 0.021060 -2022-12-06 10:44:44,639 - Epoch: [50][ 850/ 1200] Overall Loss 0.273371 Objective Loss 0.273371 LR 0.001000 Time 0.021035 -2022-12-06 10:44:44,830 - Epoch: [50][ 860/ 1200] Overall Loss 0.273367 Objective Loss 0.273367 LR 0.001000 Time 0.021012 -2022-12-06 10:44:45,020 - Epoch: [50][ 870/ 1200] Overall Loss 0.273831 Objective Loss 0.273831 LR 0.001000 Time 0.020989 -2022-12-06 10:44:45,211 - Epoch: [50][ 880/ 1200] Overall Loss 0.273938 Objective Loss 0.273938 LR 0.001000 Time 0.020967 -2022-12-06 10:44:45,402 - Epoch: [50][ 890/ 1200] Overall Loss 0.274405 Objective Loss 0.274405 LR 0.001000 Time 0.020945 -2022-12-06 10:44:45,593 - Epoch: [50][ 900/ 1200] Overall Loss 0.274427 Objective Loss 0.274427 LR 0.001000 Time 0.020924 -2022-12-06 10:44:45,784 - Epoch: [50][ 910/ 1200] Overall Loss 0.274408 Objective Loss 0.274408 LR 0.001000 Time 0.020903 -2022-12-06 10:44:45,974 - Epoch: [50][ 920/ 1200] Overall Loss 0.274765 Objective Loss 0.274765 LR 0.001000 Time 0.020882 -2022-12-06 10:44:46,166 - Epoch: [50][ 930/ 1200] Overall Loss 0.275005 Objective Loss 0.275005 LR 0.001000 Time 0.020863 -2022-12-06 10:44:46,357 - Epoch: [50][ 940/ 1200] Overall Loss 0.274922 Objective Loss 0.274922 LR 0.001000 Time 0.020844 -2022-12-06 10:44:46,547 - Epoch: [50][ 950/ 1200] Overall Loss 0.275037 Objective Loss 0.275037 LR 0.001000 Time 0.020824 -2022-12-06 10:44:46,738 - Epoch: [50][ 960/ 1200] Overall Loss 0.275256 Objective Loss 0.275256 LR 0.001000 Time 0.020806 -2022-12-06 10:44:46,929 - Epoch: [50][ 970/ 1200] Overall Loss 0.275113 Objective Loss 0.275113 LR 0.001000 Time 0.020787 -2022-12-06 10:44:47,120 - Epoch: [50][ 980/ 1200] Overall Loss 0.275094 Objective Loss 0.275094 LR 0.001000 Time 0.020769 -2022-12-06 10:44:47,310 - Epoch: [50][ 990/ 1200] Overall Loss 0.275289 Objective Loss 0.275289 LR 0.001000 Time 0.020752 -2022-12-06 10:44:47,501 - Epoch: [50][ 1000/ 1200] Overall Loss 0.275280 Objective Loss 0.275280 LR 0.001000 Time 0.020735 -2022-12-06 10:44:47,691 - Epoch: [50][ 1010/ 1200] Overall Loss 0.275363 Objective Loss 0.275363 LR 0.001000 Time 0.020717 -2022-12-06 10:44:47,882 - Epoch: [50][ 1020/ 1200] Overall Loss 0.275403 Objective Loss 0.275403 LR 0.001000 Time 0.020700 -2022-12-06 10:44:48,073 - Epoch: [50][ 1030/ 1200] Overall Loss 0.275326 Objective Loss 0.275326 LR 0.001000 Time 0.020684 -2022-12-06 10:44:48,264 - Epoch: [50][ 1040/ 1200] Overall Loss 0.275380 Objective Loss 0.275380 LR 0.001000 Time 0.020669 -2022-12-06 10:44:48,455 - Epoch: [50][ 1050/ 1200] Overall Loss 0.275209 Objective Loss 0.275209 LR 0.001000 Time 0.020653 -2022-12-06 10:44:48,645 - Epoch: [50][ 1060/ 1200] Overall Loss 0.275306 Objective Loss 0.275306 LR 0.001000 Time 0.020637 -2022-12-06 10:44:48,835 - Epoch: [50][ 1070/ 1200] Overall Loss 0.275396 Objective Loss 0.275396 LR 0.001000 Time 0.020621 -2022-12-06 10:44:49,026 - Epoch: [50][ 1080/ 1200] Overall Loss 0.275322 Objective Loss 0.275322 LR 0.001000 Time 0.020606 -2022-12-06 10:44:49,217 - Epoch: [50][ 1090/ 1200] Overall Loss 0.275040 Objective Loss 0.275040 LR 0.001000 Time 0.020592 -2022-12-06 10:44:49,407 - Epoch: [50][ 1100/ 1200] Overall Loss 0.275135 Objective Loss 0.275135 LR 0.001000 Time 0.020577 -2022-12-06 10:44:49,597 - Epoch: [50][ 1110/ 1200] Overall Loss 0.275317 Objective Loss 0.275317 LR 0.001000 Time 0.020563 -2022-12-06 10:44:49,787 - Epoch: [50][ 1120/ 1200] Overall Loss 0.275488 Objective Loss 0.275488 LR 0.001000 Time 0.020548 -2022-12-06 10:44:49,978 - Epoch: [50][ 1130/ 1200] Overall Loss 0.275656 Objective Loss 0.275656 LR 0.001000 Time 0.020535 -2022-12-06 10:44:50,169 - Epoch: [50][ 1140/ 1200] Overall Loss 0.275515 Objective Loss 0.275515 LR 0.001000 Time 0.020522 -2022-12-06 10:44:50,359 - Epoch: [50][ 1150/ 1200] Overall Loss 0.275556 Objective Loss 0.275556 LR 0.001000 Time 0.020508 -2022-12-06 10:44:50,550 - Epoch: [50][ 1160/ 1200] Overall Loss 0.275619 Objective Loss 0.275619 LR 0.001000 Time 0.020495 -2022-12-06 10:44:50,740 - Epoch: [50][ 1170/ 1200] Overall Loss 0.275306 Objective Loss 0.275306 LR 0.001000 Time 0.020482 -2022-12-06 10:44:50,931 - Epoch: [50][ 1180/ 1200] Overall Loss 0.275317 Objective Loss 0.275317 LR 0.001000 Time 0.020470 -2022-12-06 10:44:51,122 - Epoch: [50][ 1190/ 1200] Overall Loss 0.275621 Objective Loss 0.275621 LR 0.001000 Time 0.020458 -2022-12-06 10:44:51,347 - Epoch: [50][ 1200/ 1200] Overall Loss 0.275603 Objective Loss 0.275603 Top1 87.447699 Top5 97.907950 LR 0.001000 Time 0.020474 -2022-12-06 10:44:51,436 - --- validate (epoch=50)----------- -2022-12-06 10:44:51,437 - 34129 samples (256 per mini-batch) -2022-12-06 10:44:51,879 - Epoch: [50][ 10/ 134] Loss 0.288321 Top1 84.609375 Top5 97.968750 -2022-12-06 10:44:52,009 - Epoch: [50][ 20/ 134] Loss 0.289284 Top1 84.296875 Top5 97.578125 -2022-12-06 10:44:52,139 - Epoch: [50][ 30/ 134] Loss 0.295209 Top1 84.127604 Top5 97.500000 -2022-12-06 10:44:52,270 - Epoch: [50][ 40/ 134] Loss 0.295461 Top1 84.091797 Top5 97.480469 -2022-12-06 10:44:52,402 - Epoch: [50][ 50/ 134] Loss 0.294705 Top1 84.179688 Top5 97.578125 -2022-12-06 10:44:52,533 - Epoch: [50][ 60/ 134] Loss 0.301586 Top1 84.127604 Top5 97.649740 -2022-12-06 10:44:52,662 - Epoch: [50][ 70/ 134] Loss 0.300815 Top1 84.051339 Top5 97.706473 -2022-12-06 10:44:52,793 - Epoch: [50][ 80/ 134] Loss 0.304619 Top1 84.057617 Top5 97.666016 -2022-12-06 10:44:52,925 - Epoch: [50][ 90/ 134] Loss 0.303259 Top1 84.123264 Top5 97.721354 -2022-12-06 10:44:53,055 - Epoch: [50][ 100/ 134] Loss 0.302856 Top1 84.226562 Top5 97.742188 -2022-12-06 10:44:53,183 - Epoch: [50][ 110/ 134] Loss 0.304459 Top1 84.275568 Top5 97.695312 -2022-12-06 10:44:53,316 - Epoch: [50][ 120/ 134] Loss 0.305536 Top1 84.371745 Top5 97.695312 -2022-12-06 10:44:53,451 - Epoch: [50][ 130/ 134] Loss 0.307830 Top1 84.311899 Top5 97.683293 -2022-12-06 10:44:53,488 - Epoch: [50][ 134/ 134] Loss 0.308284 Top1 84.327112 Top5 97.688183 -2022-12-06 10:44:53,576 - ==> Top1: 84.327 Top5: 97.688 Loss: 0.308 - -2022-12-06 10:44:53,576 - ==> Confusion: -[[ 881 7 3 2 7 7 0 1 5 63 0 2 3 3 4 1 2 0 1 0 4] - [ 2 896 4 2 9 47 3 14 1 0 3 3 3 2 1 1 5 1 16 2 12] - [ 4 4 989 25 2 2 22 8 0 1 3 6 3 4 2 5 2 1 4 2 14] - [ 1 3 15 925 1 4 0 1 0 0 12 2 9 1 16 0 3 3 19 1 4] - [ 14 7 2 1 924 16 1 0 0 7 2 4 1 4 12 4 11 1 2 1 6] - [ 2 6 0 3 2 983 3 14 2 1 0 18 5 13 3 1 0 2 1 4 6] - [ 0 2 11 4 0 5 1063 3 0 0 2 2 0 2 0 12 1 0 1 5 5] - [ 1 8 8 1 2 42 8 923 0 0 5 9 1 3 0 0 3 0 25 11 4] - [ 4 7 1 0 0 6 0 1 950 44 8 2 4 17 11 1 2 0 5 0 1] - [ 59 0 1 1 2 6 0 3 26 872 0 4 1 15 0 2 0 1 0 0 8] - [ 0 4 2 9 2 1 2 3 10 1 945 3 1 18 4 0 0 0 8 0 6] - [ 3 0 4 0 0 13 2 2 0 0 1 989 14 7 0 4 1 2 3 4 2] - [ 0 1 2 1 1 2 1 1 0 0 0 58 868 2 0 9 4 8 0 3 8] - [ 0 2 0 0 1 13 1 1 11 12 8 7 6 939 3 2 5 0 1 2 9] - [ 5 4 1 12 5 7 0 1 21 4 4 1 2 3 1039 0 3 0 7 2 9] - [ 0 1 2 0 2 2 3 1 0 0 0 10 8 3 0 984 11 10 0 4 2] - [ 0 2 0 1 1 2 0 0 0 1 0 4 1 3 0 6 1037 5 1 4 4] - [ 3 0 1 3 0 1 3 0 1 1 0 19 34 5 1 13 3 945 1 1 1] - [ 1 5 2 8 2 4 0 28 0 1 6 1 3 2 7 2 3 3 924 1 5] - [ 1 1 2 1 0 11 8 8 0 1 2 33 10 4 0 1 5 2 0 983 7] - [ 128 234 158 130 101 300 72 152 90 86 187 164 345 334 182 115 275 55 177 223 9718]] - -2022-12-06 10:44:54,237 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:44:54,237 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:44:54,243 - - -2022-12-06 10:44:54,243 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:44:55,172 - Epoch: [51][ 10/ 1200] Overall Loss 0.297074 Objective Loss 0.297074 LR 0.001000 Time 0.092867 -2022-12-06 10:44:55,365 - Epoch: [51][ 20/ 1200] Overall Loss 0.276517 Objective Loss 0.276517 LR 0.001000 Time 0.056045 -2022-12-06 10:44:55,563 - Epoch: [51][ 30/ 1200] Overall Loss 0.282607 Objective Loss 0.282607 LR 0.001000 Time 0.043922 -2022-12-06 10:44:55,769 - Epoch: [51][ 40/ 1200] Overall Loss 0.280620 Objective Loss 0.280620 LR 0.001000 Time 0.038077 -2022-12-06 10:44:55,972 - Epoch: [51][ 50/ 1200] Overall Loss 0.276021 Objective Loss 0.276021 LR 0.001000 Time 0.034518 -2022-12-06 10:44:56,179 - Epoch: [51][ 60/ 1200] Overall Loss 0.274338 Objective Loss 0.274338 LR 0.001000 Time 0.032196 -2022-12-06 10:44:56,381 - Epoch: [51][ 70/ 1200] Overall Loss 0.270586 Objective Loss 0.270586 LR 0.001000 Time 0.030484 -2022-12-06 10:44:56,587 - Epoch: [51][ 80/ 1200] Overall Loss 0.268815 Objective Loss 0.268815 LR 0.001000 Time 0.029242 -2022-12-06 10:44:56,790 - Epoch: [51][ 90/ 1200] Overall Loss 0.269052 Objective Loss 0.269052 LR 0.001000 Time 0.028235 -2022-12-06 10:44:56,996 - Epoch: [51][ 100/ 1200] Overall Loss 0.271407 Objective Loss 0.271407 LR 0.001000 Time 0.027472 -2022-12-06 10:44:57,199 - Epoch: [51][ 110/ 1200] Overall Loss 0.272845 Objective Loss 0.272845 LR 0.001000 Time 0.026812 -2022-12-06 10:44:57,406 - Epoch: [51][ 120/ 1200] Overall Loss 0.271523 Objective Loss 0.271523 LR 0.001000 Time 0.026296 -2022-12-06 10:44:57,608 - Epoch: [51][ 130/ 1200] Overall Loss 0.272143 Objective Loss 0.272143 LR 0.001000 Time 0.025825 -2022-12-06 10:44:57,815 - Epoch: [51][ 140/ 1200] Overall Loss 0.269180 Objective Loss 0.269180 LR 0.001000 Time 0.025455 -2022-12-06 10:44:58,018 - Epoch: [51][ 150/ 1200] Overall Loss 0.268696 Objective Loss 0.268696 LR 0.001000 Time 0.025105 -2022-12-06 10:44:58,224 - Epoch: [51][ 160/ 1200] Overall Loss 0.267647 Objective Loss 0.267647 LR 0.001000 Time 0.024822 -2022-12-06 10:44:58,427 - Epoch: [51][ 170/ 1200] Overall Loss 0.268057 Objective Loss 0.268057 LR 0.001000 Time 0.024547 -2022-12-06 10:44:58,634 - Epoch: [51][ 180/ 1200] Overall Loss 0.266412 Objective Loss 0.266412 LR 0.001000 Time 0.024330 -2022-12-06 10:44:58,836 - Epoch: [51][ 190/ 1200] Overall Loss 0.264845 Objective Loss 0.264845 LR 0.001000 Time 0.024112 -2022-12-06 10:44:59,043 - Epoch: [51][ 200/ 1200] Overall Loss 0.264105 Objective Loss 0.264105 LR 0.001000 Time 0.023936 -2022-12-06 10:44:59,246 - Epoch: [51][ 210/ 1200] Overall Loss 0.264659 Objective Loss 0.264659 LR 0.001000 Time 0.023763 -2022-12-06 10:44:59,453 - Epoch: [51][ 220/ 1200] Overall Loss 0.266335 Objective Loss 0.266335 LR 0.001000 Time 0.023621 -2022-12-06 10:44:59,656 - Epoch: [51][ 230/ 1200] Overall Loss 0.266401 Objective Loss 0.266401 LR 0.001000 Time 0.023473 -2022-12-06 10:44:59,863 - Epoch: [51][ 240/ 1200] Overall Loss 0.266480 Objective Loss 0.266480 LR 0.001000 Time 0.023353 -2022-12-06 10:45:00,065 - Epoch: [51][ 250/ 1200] Overall Loss 0.265903 Objective Loss 0.265903 LR 0.001000 Time 0.023228 -2022-12-06 10:45:00,271 - Epoch: [51][ 260/ 1200] Overall Loss 0.265716 Objective Loss 0.265716 LR 0.001000 Time 0.023124 -2022-12-06 10:45:00,474 - Epoch: [51][ 270/ 1200] Overall Loss 0.266548 Objective Loss 0.266548 LR 0.001000 Time 0.023014 -2022-12-06 10:45:00,680 - Epoch: [51][ 280/ 1200] Overall Loss 0.266012 Objective Loss 0.266012 LR 0.001000 Time 0.022926 -2022-12-06 10:45:00,881 - Epoch: [51][ 290/ 1200] Overall Loss 0.265965 Objective Loss 0.265965 LR 0.001000 Time 0.022829 -2022-12-06 10:45:01,088 - Epoch: [51][ 300/ 1200] Overall Loss 0.266638 Objective Loss 0.266638 LR 0.001000 Time 0.022754 -2022-12-06 10:45:01,290 - Epoch: [51][ 310/ 1200] Overall Loss 0.267607 Objective Loss 0.267607 LR 0.001000 Time 0.022670 -2022-12-06 10:45:01,496 - Epoch: [51][ 320/ 1200] Overall Loss 0.267852 Objective Loss 0.267852 LR 0.001000 Time 0.022605 -2022-12-06 10:45:01,699 - Epoch: [51][ 330/ 1200] Overall Loss 0.267702 Objective Loss 0.267702 LR 0.001000 Time 0.022530 -2022-12-06 10:45:01,906 - Epoch: [51][ 340/ 1200] Overall Loss 0.267625 Objective Loss 0.267625 LR 0.001000 Time 0.022474 -2022-12-06 10:45:02,107 - Epoch: [51][ 350/ 1200] Overall Loss 0.267745 Objective Loss 0.267745 LR 0.001000 Time 0.022405 -2022-12-06 10:45:02,312 - Epoch: [51][ 360/ 1200] Overall Loss 0.268466 Objective Loss 0.268466 LR 0.001000 Time 0.022351 -2022-12-06 10:45:02,514 - Epoch: [51][ 370/ 1200] Overall Loss 0.268972 Objective Loss 0.268972 LR 0.001000 Time 0.022291 -2022-12-06 10:45:02,720 - Epoch: [51][ 380/ 1200] Overall Loss 0.269173 Objective Loss 0.269173 LR 0.001000 Time 0.022246 -2022-12-06 10:45:02,921 - Epoch: [51][ 390/ 1200] Overall Loss 0.269394 Objective Loss 0.269394 LR 0.001000 Time 0.022190 -2022-12-06 10:45:03,128 - Epoch: [51][ 400/ 1200] Overall Loss 0.269687 Objective Loss 0.269687 LR 0.001000 Time 0.022150 -2022-12-06 10:45:03,330 - Epoch: [51][ 410/ 1200] Overall Loss 0.269445 Objective Loss 0.269445 LR 0.001000 Time 0.022100 -2022-12-06 10:45:03,536 - Epoch: [51][ 420/ 1200] Overall Loss 0.270136 Objective Loss 0.270136 LR 0.001000 Time 0.022063 -2022-12-06 10:45:03,739 - Epoch: [51][ 430/ 1200] Overall Loss 0.270406 Objective Loss 0.270406 LR 0.001000 Time 0.022021 -2022-12-06 10:45:03,946 - Epoch: [51][ 440/ 1200] Overall Loss 0.270731 Objective Loss 0.270731 LR 0.001000 Time 0.021990 -2022-12-06 10:45:04,149 - Epoch: [51][ 450/ 1200] Overall Loss 0.271286 Objective Loss 0.271286 LR 0.001000 Time 0.021950 -2022-12-06 10:45:04,355 - Epoch: [51][ 460/ 1200] Overall Loss 0.271621 Objective Loss 0.271621 LR 0.001000 Time 0.021920 -2022-12-06 10:45:04,557 - Epoch: [51][ 470/ 1200] Overall Loss 0.272473 Objective Loss 0.272473 LR 0.001000 Time 0.021883 -2022-12-06 10:45:04,763 - Epoch: [51][ 480/ 1200] Overall Loss 0.272824 Objective Loss 0.272824 LR 0.001000 Time 0.021855 -2022-12-06 10:45:04,967 - Epoch: [51][ 490/ 1200] Overall Loss 0.272693 Objective Loss 0.272693 LR 0.001000 Time 0.021823 -2022-12-06 10:45:05,173 - Epoch: [51][ 500/ 1200] Overall Loss 0.272273 Objective Loss 0.272273 LR 0.001000 Time 0.021797 -2022-12-06 10:45:05,376 - Epoch: [51][ 510/ 1200] Overall Loss 0.272720 Objective Loss 0.272720 LR 0.001000 Time 0.021767 -2022-12-06 10:45:05,583 - Epoch: [51][ 520/ 1200] Overall Loss 0.272691 Objective Loss 0.272691 LR 0.001000 Time 0.021745 -2022-12-06 10:45:05,785 - Epoch: [51][ 530/ 1200] Overall Loss 0.272490 Objective Loss 0.272490 LR 0.001000 Time 0.021715 -2022-12-06 10:45:05,991 - Epoch: [51][ 540/ 1200] Overall Loss 0.271756 Objective Loss 0.271756 LR 0.001000 Time 0.021693 -2022-12-06 10:45:06,195 - Epoch: [51][ 550/ 1200] Overall Loss 0.271965 Objective Loss 0.271965 LR 0.001000 Time 0.021668 -2022-12-06 10:45:06,401 - Epoch: [51][ 560/ 1200] Overall Loss 0.271650 Objective Loss 0.271650 LR 0.001000 Time 0.021648 -2022-12-06 10:45:06,603 - Epoch: [51][ 570/ 1200] Overall Loss 0.271494 Objective Loss 0.271494 LR 0.001000 Time 0.021622 -2022-12-06 10:45:06,809 - Epoch: [51][ 580/ 1200] Overall Loss 0.271455 Objective Loss 0.271455 LR 0.001000 Time 0.021603 -2022-12-06 10:45:07,012 - Epoch: [51][ 590/ 1200] Overall Loss 0.271266 Objective Loss 0.271266 LR 0.001000 Time 0.021580 -2022-12-06 10:45:07,219 - Epoch: [51][ 600/ 1200] Overall Loss 0.271301 Objective Loss 0.271301 LR 0.001000 Time 0.021564 -2022-12-06 10:45:07,422 - Epoch: [51][ 610/ 1200] Overall Loss 0.271747 Objective Loss 0.271747 LR 0.001000 Time 0.021543 -2022-12-06 10:45:07,629 - Epoch: [51][ 620/ 1200] Overall Loss 0.271890 Objective Loss 0.271890 LR 0.001000 Time 0.021528 -2022-12-06 10:45:07,832 - Epoch: [51][ 630/ 1200] Overall Loss 0.272319 Objective Loss 0.272319 LR 0.001000 Time 0.021507 -2022-12-06 10:45:08,038 - Epoch: [51][ 640/ 1200] Overall Loss 0.271969 Objective Loss 0.271969 LR 0.001000 Time 0.021493 -2022-12-06 10:45:08,240 - Epoch: [51][ 650/ 1200] Overall Loss 0.272353 Objective Loss 0.272353 LR 0.001000 Time 0.021472 -2022-12-06 10:45:08,446 - Epoch: [51][ 660/ 1200] Overall Loss 0.272227 Objective Loss 0.272227 LR 0.001000 Time 0.021458 -2022-12-06 10:45:08,649 - Epoch: [51][ 670/ 1200] Overall Loss 0.272231 Objective Loss 0.272231 LR 0.001000 Time 0.021439 -2022-12-06 10:45:08,855 - Epoch: [51][ 680/ 1200] Overall Loss 0.272317 Objective Loss 0.272317 LR 0.001000 Time 0.021426 -2022-12-06 10:45:09,058 - Epoch: [51][ 690/ 1200] Overall Loss 0.272221 Objective Loss 0.272221 LR 0.001000 Time 0.021409 -2022-12-06 10:45:09,265 - Epoch: [51][ 700/ 1200] Overall Loss 0.272794 Objective Loss 0.272794 LR 0.001000 Time 0.021398 -2022-12-06 10:45:09,468 - Epoch: [51][ 710/ 1200] Overall Loss 0.272833 Objective Loss 0.272833 LR 0.001000 Time 0.021382 -2022-12-06 10:45:09,674 - Epoch: [51][ 720/ 1200] Overall Loss 0.273031 Objective Loss 0.273031 LR 0.001000 Time 0.021369 -2022-12-06 10:45:09,876 - Epoch: [51][ 730/ 1200] Overall Loss 0.273665 Objective Loss 0.273665 LR 0.001000 Time 0.021353 -2022-12-06 10:45:10,083 - Epoch: [51][ 740/ 1200] Overall Loss 0.273929 Objective Loss 0.273929 LR 0.001000 Time 0.021343 -2022-12-06 10:45:10,285 - Epoch: [51][ 750/ 1200] Overall Loss 0.273676 Objective Loss 0.273676 LR 0.001000 Time 0.021327 -2022-12-06 10:45:10,491 - Epoch: [51][ 760/ 1200] Overall Loss 0.273934 Objective Loss 0.273934 LR 0.001000 Time 0.021317 -2022-12-06 10:45:10,695 - Epoch: [51][ 770/ 1200] Overall Loss 0.274158 Objective Loss 0.274158 LR 0.001000 Time 0.021304 -2022-12-06 10:45:10,902 - Epoch: [51][ 780/ 1200] Overall Loss 0.273807 Objective Loss 0.273807 LR 0.001000 Time 0.021295 -2022-12-06 10:45:11,105 - Epoch: [51][ 790/ 1200] Overall Loss 0.273895 Objective Loss 0.273895 LR 0.001000 Time 0.021282 -2022-12-06 10:45:11,311 - Epoch: [51][ 800/ 1200] Overall Loss 0.273785 Objective Loss 0.273785 LR 0.001000 Time 0.021273 -2022-12-06 10:45:11,514 - Epoch: [51][ 810/ 1200] Overall Loss 0.274187 Objective Loss 0.274187 LR 0.001000 Time 0.021260 -2022-12-06 10:45:11,720 - Epoch: [51][ 820/ 1200] Overall Loss 0.274613 Objective Loss 0.274613 LR 0.001000 Time 0.021251 -2022-12-06 10:45:11,923 - Epoch: [51][ 830/ 1200] Overall Loss 0.274427 Objective Loss 0.274427 LR 0.001000 Time 0.021239 -2022-12-06 10:45:12,129 - Epoch: [51][ 840/ 1200] Overall Loss 0.274293 Objective Loss 0.274293 LR 0.001000 Time 0.021231 -2022-12-06 10:45:12,332 - Epoch: [51][ 850/ 1200] Overall Loss 0.274537 Objective Loss 0.274537 LR 0.001000 Time 0.021219 -2022-12-06 10:45:12,538 - Epoch: [51][ 860/ 1200] Overall Loss 0.274757 Objective Loss 0.274757 LR 0.001000 Time 0.021211 -2022-12-06 10:45:12,741 - Epoch: [51][ 870/ 1200] Overall Loss 0.274665 Objective Loss 0.274665 LR 0.001000 Time 0.021200 -2022-12-06 10:45:12,948 - Epoch: [51][ 880/ 1200] Overall Loss 0.274652 Objective Loss 0.274652 LR 0.001000 Time 0.021193 -2022-12-06 10:45:13,151 - Epoch: [51][ 890/ 1200] Overall Loss 0.274506 Objective Loss 0.274506 LR 0.001000 Time 0.021183 -2022-12-06 10:45:13,358 - Epoch: [51][ 900/ 1200] Overall Loss 0.274526 Objective Loss 0.274526 LR 0.001000 Time 0.021176 -2022-12-06 10:45:13,561 - Epoch: [51][ 910/ 1200] Overall Loss 0.274511 Objective Loss 0.274511 LR 0.001000 Time 0.021166 -2022-12-06 10:45:13,767 - Epoch: [51][ 920/ 1200] Overall Loss 0.274477 Objective Loss 0.274477 LR 0.001000 Time 0.021159 -2022-12-06 10:45:13,970 - Epoch: [51][ 930/ 1200] Overall Loss 0.274562 Objective Loss 0.274562 LR 0.001000 Time 0.021150 -2022-12-06 10:45:14,177 - Epoch: [51][ 940/ 1200] Overall Loss 0.274930 Objective Loss 0.274930 LR 0.001000 Time 0.021144 -2022-12-06 10:45:14,380 - Epoch: [51][ 950/ 1200] Overall Loss 0.274996 Objective Loss 0.274996 LR 0.001000 Time 0.021134 -2022-12-06 10:45:14,586 - Epoch: [51][ 960/ 1200] Overall Loss 0.275004 Objective Loss 0.275004 LR 0.001000 Time 0.021128 -2022-12-06 10:45:14,790 - Epoch: [51][ 970/ 1200] Overall Loss 0.274944 Objective Loss 0.274944 LR 0.001000 Time 0.021120 -2022-12-06 10:45:14,995 - Epoch: [51][ 980/ 1200] Overall Loss 0.274959 Objective Loss 0.274959 LR 0.001000 Time 0.021114 -2022-12-06 10:45:15,199 - Epoch: [51][ 990/ 1200] Overall Loss 0.274907 Objective Loss 0.274907 LR 0.001000 Time 0.021105 -2022-12-06 10:45:15,406 - Epoch: [51][ 1000/ 1200] Overall Loss 0.275039 Objective Loss 0.275039 LR 0.001000 Time 0.021100 -2022-12-06 10:45:15,608 - Epoch: [51][ 1010/ 1200] Overall Loss 0.275237 Objective Loss 0.275237 LR 0.001000 Time 0.021091 -2022-12-06 10:45:15,815 - Epoch: [51][ 1020/ 1200] Overall Loss 0.275110 Objective Loss 0.275110 LR 0.001000 Time 0.021086 -2022-12-06 10:45:16,017 - Epoch: [51][ 1030/ 1200] Overall Loss 0.275011 Objective Loss 0.275011 LR 0.001000 Time 0.021078 -2022-12-06 10:45:16,224 - Epoch: [51][ 1040/ 1200] Overall Loss 0.274961 Objective Loss 0.274961 LR 0.001000 Time 0.021073 -2022-12-06 10:45:16,427 - Epoch: [51][ 1050/ 1200] Overall Loss 0.274835 Objective Loss 0.274835 LR 0.001000 Time 0.021065 -2022-12-06 10:45:16,633 - Epoch: [51][ 1060/ 1200] Overall Loss 0.274966 Objective Loss 0.274966 LR 0.001000 Time 0.021061 -2022-12-06 10:45:16,836 - Epoch: [51][ 1070/ 1200] Overall Loss 0.274976 Objective Loss 0.274976 LR 0.001000 Time 0.021053 -2022-12-06 10:45:17,042 - Epoch: [51][ 1080/ 1200] Overall Loss 0.274915 Objective Loss 0.274915 LR 0.001000 Time 0.021048 -2022-12-06 10:45:17,245 - Epoch: [51][ 1090/ 1200] Overall Loss 0.275291 Objective Loss 0.275291 LR 0.001000 Time 0.021040 -2022-12-06 10:45:17,451 - Epoch: [51][ 1100/ 1200] Overall Loss 0.275417 Objective Loss 0.275417 LR 0.001000 Time 0.021036 -2022-12-06 10:45:17,653 - Epoch: [51][ 1110/ 1200] Overall Loss 0.275483 Objective Loss 0.275483 LR 0.001000 Time 0.021028 -2022-12-06 10:45:17,859 - Epoch: [51][ 1120/ 1200] Overall Loss 0.275232 Objective Loss 0.275232 LR 0.001000 Time 0.021024 -2022-12-06 10:45:18,062 - Epoch: [51][ 1130/ 1200] Overall Loss 0.275154 Objective Loss 0.275154 LR 0.001000 Time 0.021017 -2022-12-06 10:45:18,269 - Epoch: [51][ 1140/ 1200] Overall Loss 0.275177 Objective Loss 0.275177 LR 0.001000 Time 0.021013 -2022-12-06 10:45:18,471 - Epoch: [51][ 1150/ 1200] Overall Loss 0.275277 Objective Loss 0.275277 LR 0.001000 Time 0.021006 -2022-12-06 10:45:18,678 - Epoch: [51][ 1160/ 1200] Overall Loss 0.275451 Objective Loss 0.275451 LR 0.001000 Time 0.021002 -2022-12-06 10:45:18,880 - Epoch: [51][ 1170/ 1200] Overall Loss 0.275606 Objective Loss 0.275606 LR 0.001000 Time 0.020995 -2022-12-06 10:45:19,087 - Epoch: [51][ 1180/ 1200] Overall Loss 0.275776 Objective Loss 0.275776 LR 0.001000 Time 0.020991 -2022-12-06 10:45:19,290 - Epoch: [51][ 1190/ 1200] Overall Loss 0.276116 Objective Loss 0.276116 LR 0.001000 Time 0.020985 -2022-12-06 10:45:19,528 - Epoch: [51][ 1200/ 1200] Overall Loss 0.276019 Objective Loss 0.276019 Top1 86.820084 Top5 99.163180 LR 0.001000 Time 0.021008 -2022-12-06 10:45:19,616 - --- validate (epoch=51)----------- -2022-12-06 10:45:19,617 - 34129 samples (256 per mini-batch) -2022-12-06 10:45:20,072 - Epoch: [51][ 10/ 134] Loss 0.312765 Top1 84.492188 Top5 98.164062 -2022-12-06 10:45:20,207 - Epoch: [51][ 20/ 134] Loss 0.302966 Top1 84.277344 Top5 98.183594 -2022-12-06 10:45:20,339 - Epoch: [51][ 30/ 134] Loss 0.289941 Top1 84.726562 Top5 98.190104 -2022-12-06 10:45:20,469 - Epoch: [51][ 40/ 134] Loss 0.291147 Top1 84.599609 Top5 98.095703 -2022-12-06 10:45:20,600 - Epoch: [51][ 50/ 134] Loss 0.294464 Top1 84.515625 Top5 98.117188 -2022-12-06 10:45:20,730 - Epoch: [51][ 60/ 134] Loss 0.297991 Top1 84.394531 Top5 98.059896 -2022-12-06 10:45:20,860 - Epoch: [51][ 70/ 134] Loss 0.302014 Top1 84.380580 Top5 98.074777 -2022-12-06 10:45:20,992 - Epoch: [51][ 80/ 134] Loss 0.298830 Top1 84.448242 Top5 98.081055 -2022-12-06 10:45:21,124 - Epoch: [51][ 90/ 134] Loss 0.301993 Top1 84.361979 Top5 98.020833 -2022-12-06 10:45:21,255 - Epoch: [51][ 100/ 134] Loss 0.299874 Top1 84.410156 Top5 98.046875 -2022-12-06 10:45:21,388 - Epoch: [51][ 110/ 134] Loss 0.296796 Top1 84.570312 Top5 98.064631 -2022-12-06 10:45:21,521 - Epoch: [51][ 120/ 134] Loss 0.296847 Top1 84.567057 Top5 98.043620 -2022-12-06 10:45:21,656 - Epoch: [51][ 130/ 134] Loss 0.296386 Top1 84.498197 Top5 98.052885 -2022-12-06 10:45:21,694 - Epoch: [51][ 134/ 134] Loss 0.296337 Top1 84.497055 Top5 98.066161 -2022-12-06 10:45:21,784 - ==> Top1: 84.497 Top5: 98.066 Loss: 0.296 - -2022-12-06 10:45:21,785 - ==> Confusion: -[[ 894 1 6 3 5 6 1 1 2 57 0 3 1 1 7 1 2 1 0 0 4] - [ 1 923 4 4 5 26 2 14 0 2 6 5 1 1 0 1 4 3 12 2 11] - [ 3 2 1010 15 2 1 26 6 0 2 5 6 0 1 4 2 0 1 8 1 8] - [ 1 2 22 945 1 1 1 0 0 1 9 2 2 2 10 0 2 2 11 0 6] - [ 8 4 3 2 945 6 0 2 0 9 1 2 2 2 9 3 13 1 2 2 4] - [ 1 17 0 3 6 960 6 23 1 2 0 12 6 9 3 0 3 0 3 4 10] - [ 1 2 18 0 0 1 1077 2 0 1 0 2 1 1 0 4 2 0 1 4 1] - [ 0 9 14 3 2 27 4 942 0 0 4 5 1 0 1 0 0 0 24 14 4] - [ 13 2 1 0 0 4 0 0 927 54 15 3 6 11 18 0 3 0 4 0 3] - [ 67 0 3 0 4 3 0 4 19 880 1 2 0 7 3 1 1 1 0 1 4] - [ 1 0 11 4 0 3 6 2 3 2 948 3 3 12 5 0 0 0 8 2 6] - [ 5 1 2 0 1 13 3 7 1 0 0 940 40 5 1 3 4 6 2 13 4] - [ 0 2 2 5 1 2 0 1 0 1 1 25 891 4 0 7 0 15 0 4 8] - [ 1 4 4 0 1 9 0 4 11 14 8 4 4 938 0 2 4 1 0 5 9] - [ 8 1 5 19 3 4 0 1 12 5 1 3 2 1 1049 0 0 0 7 2 7] - [ 2 0 4 2 1 2 7 0 0 0 0 7 9 5 0 981 7 8 0 5 3] - [ 0 3 4 1 1 1 4 1 0 0 0 6 0 1 1 5 1027 3 0 7 7] - [ 2 0 0 4 0 0 2 4 0 4 1 6 14 3 4 11 2 970 3 2 4] - [ 4 2 5 14 2 0 0 26 1 0 6 2 2 1 9 1 0 1 925 2 5] - [ 3 2 2 0 0 7 10 5 0 0 1 11 7 2 2 1 4 5 1 1015 2] - [ 148 236 272 136 124 203 118 173 75 89 162 128 395 300 158 109 218 73 185 277 9647]] - -2022-12-06 10:45:22,456 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:45:22,457 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:45:22,462 - - -2022-12-06 10:45:22,463 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:45:23,404 - Epoch: [52][ 10/ 1200] Overall Loss 0.272947 Objective Loss 0.272947 LR 0.001000 Time 0.094091 -2022-12-06 10:45:23,601 - Epoch: [52][ 20/ 1200] Overall Loss 0.268029 Objective Loss 0.268029 LR 0.001000 Time 0.056862 -2022-12-06 10:45:23,792 - Epoch: [52][ 30/ 1200] Overall Loss 0.269144 Objective Loss 0.269144 LR 0.001000 Time 0.044271 -2022-12-06 10:45:23,983 - Epoch: [52][ 40/ 1200] Overall Loss 0.280486 Objective Loss 0.280486 LR 0.001000 Time 0.037965 -2022-12-06 10:45:24,174 - Epoch: [52][ 50/ 1200] Overall Loss 0.284875 Objective Loss 0.284875 LR 0.001000 Time 0.034173 -2022-12-06 10:45:24,364 - Epoch: [52][ 60/ 1200] Overall Loss 0.283391 Objective Loss 0.283391 LR 0.001000 Time 0.031640 -2022-12-06 10:45:24,556 - Epoch: [52][ 70/ 1200] Overall Loss 0.280903 Objective Loss 0.280903 LR 0.001000 Time 0.029849 -2022-12-06 10:45:24,747 - Epoch: [52][ 80/ 1200] Overall Loss 0.279067 Objective Loss 0.279067 LR 0.001000 Time 0.028495 -2022-12-06 10:45:24,937 - Epoch: [52][ 90/ 1200] Overall Loss 0.275201 Objective Loss 0.275201 LR 0.001000 Time 0.027433 -2022-12-06 10:45:25,127 - Epoch: [52][ 100/ 1200] Overall Loss 0.273298 Objective Loss 0.273298 LR 0.001000 Time 0.026593 -2022-12-06 10:45:25,319 - Epoch: [52][ 110/ 1200] Overall Loss 0.272568 Objective Loss 0.272568 LR 0.001000 Time 0.025912 -2022-12-06 10:45:25,510 - Epoch: [52][ 120/ 1200] Overall Loss 0.272383 Objective Loss 0.272383 LR 0.001000 Time 0.025339 -2022-12-06 10:45:25,701 - Epoch: [52][ 130/ 1200] Overall Loss 0.271729 Objective Loss 0.271729 LR 0.001000 Time 0.024856 -2022-12-06 10:45:25,891 - Epoch: [52][ 140/ 1200] Overall Loss 0.269607 Objective Loss 0.269607 LR 0.001000 Time 0.024437 -2022-12-06 10:45:26,083 - Epoch: [52][ 150/ 1200] Overall Loss 0.271945 Objective Loss 0.271945 LR 0.001000 Time 0.024080 -2022-12-06 10:45:26,273 - Epoch: [52][ 160/ 1200] Overall Loss 0.271447 Objective Loss 0.271447 LR 0.001000 Time 0.023762 -2022-12-06 10:45:26,464 - Epoch: [52][ 170/ 1200] Overall Loss 0.271386 Objective Loss 0.271386 LR 0.001000 Time 0.023480 -2022-12-06 10:45:26,654 - Epoch: [52][ 180/ 1200] Overall Loss 0.272621 Objective Loss 0.272621 LR 0.001000 Time 0.023229 -2022-12-06 10:45:26,845 - Epoch: [52][ 190/ 1200] Overall Loss 0.271948 Objective Loss 0.271948 LR 0.001000 Time 0.023011 -2022-12-06 10:45:27,036 - Epoch: [52][ 200/ 1200] Overall Loss 0.271685 Objective Loss 0.271685 LR 0.001000 Time 0.022814 -2022-12-06 10:45:27,227 - Epoch: [52][ 210/ 1200] Overall Loss 0.270278 Objective Loss 0.270278 LR 0.001000 Time 0.022635 -2022-12-06 10:45:27,419 - Epoch: [52][ 220/ 1200] Overall Loss 0.270848 Objective Loss 0.270848 LR 0.001000 Time 0.022472 -2022-12-06 10:45:27,610 - Epoch: [52][ 230/ 1200] Overall Loss 0.272958 Objective Loss 0.272958 LR 0.001000 Time 0.022325 -2022-12-06 10:45:27,802 - Epoch: [52][ 240/ 1200] Overall Loss 0.272958 Objective Loss 0.272958 LR 0.001000 Time 0.022191 -2022-12-06 10:45:27,993 - Epoch: [52][ 250/ 1200] Overall Loss 0.272537 Objective Loss 0.272537 LR 0.001000 Time 0.022067 -2022-12-06 10:45:28,184 - Epoch: [52][ 260/ 1200] Overall Loss 0.272129 Objective Loss 0.272129 LR 0.001000 Time 0.021951 -2022-12-06 10:45:28,375 - Epoch: [52][ 270/ 1200] Overall Loss 0.272583 Objective Loss 0.272583 LR 0.001000 Time 0.021845 -2022-12-06 10:45:28,567 - Epoch: [52][ 280/ 1200] Overall Loss 0.272816 Objective Loss 0.272816 LR 0.001000 Time 0.021746 -2022-12-06 10:45:28,758 - Epoch: [52][ 290/ 1200] Overall Loss 0.272002 Objective Loss 0.272002 LR 0.001000 Time 0.021654 -2022-12-06 10:45:28,949 - Epoch: [52][ 300/ 1200] Overall Loss 0.271673 Objective Loss 0.271673 LR 0.001000 Time 0.021567 -2022-12-06 10:45:29,141 - Epoch: [52][ 310/ 1200] Overall Loss 0.271816 Objective Loss 0.271816 LR 0.001000 Time 0.021487 -2022-12-06 10:45:29,331 - Epoch: [52][ 320/ 1200] Overall Loss 0.272124 Objective Loss 0.272124 LR 0.001000 Time 0.021410 -2022-12-06 10:45:29,522 - Epoch: [52][ 330/ 1200] Overall Loss 0.272588 Objective Loss 0.272588 LR 0.001000 Time 0.021339 -2022-12-06 10:45:29,713 - Epoch: [52][ 340/ 1200] Overall Loss 0.272282 Objective Loss 0.272282 LR 0.001000 Time 0.021269 -2022-12-06 10:45:29,903 - Epoch: [52][ 350/ 1200] Overall Loss 0.272601 Objective Loss 0.272601 LR 0.001000 Time 0.021204 -2022-12-06 10:45:30,094 - Epoch: [52][ 360/ 1200] Overall Loss 0.272104 Objective Loss 0.272104 LR 0.001000 Time 0.021145 -2022-12-06 10:45:30,285 - Epoch: [52][ 370/ 1200] Overall Loss 0.271292 Objective Loss 0.271292 LR 0.001000 Time 0.021086 -2022-12-06 10:45:30,475 - Epoch: [52][ 380/ 1200] Overall Loss 0.270491 Objective Loss 0.270491 LR 0.001000 Time 0.021031 -2022-12-06 10:45:30,667 - Epoch: [52][ 390/ 1200] Overall Loss 0.271275 Objective Loss 0.271275 LR 0.001000 Time 0.020982 -2022-12-06 10:45:30,858 - Epoch: [52][ 400/ 1200] Overall Loss 0.271700 Objective Loss 0.271700 LR 0.001000 Time 0.020935 -2022-12-06 10:45:31,050 - Epoch: [52][ 410/ 1200] Overall Loss 0.272029 Objective Loss 0.272029 LR 0.001000 Time 0.020889 -2022-12-06 10:45:31,240 - Epoch: [52][ 420/ 1200] Overall Loss 0.272057 Objective Loss 0.272057 LR 0.001000 Time 0.020845 -2022-12-06 10:45:31,431 - Epoch: [52][ 430/ 1200] Overall Loss 0.272259 Objective Loss 0.272259 LR 0.001000 Time 0.020803 -2022-12-06 10:45:31,622 - Epoch: [52][ 440/ 1200] Overall Loss 0.272101 Objective Loss 0.272101 LR 0.001000 Time 0.020763 -2022-12-06 10:45:31,814 - Epoch: [52][ 450/ 1200] Overall Loss 0.273024 Objective Loss 0.273024 LR 0.001000 Time 0.020727 -2022-12-06 10:45:32,005 - Epoch: [52][ 460/ 1200] Overall Loss 0.272990 Objective Loss 0.272990 LR 0.001000 Time 0.020690 -2022-12-06 10:45:32,196 - Epoch: [52][ 470/ 1200] Overall Loss 0.272583 Objective Loss 0.272583 LR 0.001000 Time 0.020655 -2022-12-06 10:45:32,387 - Epoch: [52][ 480/ 1200] Overall Loss 0.272230 Objective Loss 0.272230 LR 0.001000 Time 0.020620 -2022-12-06 10:45:32,578 - Epoch: [52][ 490/ 1200] Overall Loss 0.272225 Objective Loss 0.272225 LR 0.001000 Time 0.020590 -2022-12-06 10:45:32,769 - Epoch: [52][ 500/ 1200] Overall Loss 0.272446 Objective Loss 0.272446 LR 0.001000 Time 0.020559 -2022-12-06 10:45:32,961 - Epoch: [52][ 510/ 1200] Overall Loss 0.272700 Objective Loss 0.272700 LR 0.001000 Time 0.020530 -2022-12-06 10:45:33,152 - Epoch: [52][ 520/ 1200] Overall Loss 0.272755 Objective Loss 0.272755 LR 0.001000 Time 0.020502 -2022-12-06 10:45:33,343 - Epoch: [52][ 530/ 1200] Overall Loss 0.272978 Objective Loss 0.272978 LR 0.001000 Time 0.020475 -2022-12-06 10:45:33,535 - Epoch: [52][ 540/ 1200] Overall Loss 0.273230 Objective Loss 0.273230 LR 0.001000 Time 0.020450 -2022-12-06 10:45:33,726 - Epoch: [52][ 550/ 1200] Overall Loss 0.273179 Objective Loss 0.273179 LR 0.001000 Time 0.020424 -2022-12-06 10:45:33,916 - Epoch: [52][ 560/ 1200] Overall Loss 0.272808 Objective Loss 0.272808 LR 0.001000 Time 0.020399 -2022-12-06 10:45:34,108 - Epoch: [52][ 570/ 1200] Overall Loss 0.272824 Objective Loss 0.272824 LR 0.001000 Time 0.020376 -2022-12-06 10:45:34,299 - Epoch: [52][ 580/ 1200] Overall Loss 0.272533 Objective Loss 0.272533 LR 0.001000 Time 0.020353 -2022-12-06 10:45:34,490 - Epoch: [52][ 590/ 1200] Overall Loss 0.272705 Objective Loss 0.272705 LR 0.001000 Time 0.020331 -2022-12-06 10:45:34,680 - Epoch: [52][ 600/ 1200] Overall Loss 0.272454 Objective Loss 0.272454 LR 0.001000 Time 0.020309 -2022-12-06 10:45:34,871 - Epoch: [52][ 610/ 1200] Overall Loss 0.272438 Objective Loss 0.272438 LR 0.001000 Time 0.020288 -2022-12-06 10:45:35,063 - Epoch: [52][ 620/ 1200] Overall Loss 0.272397 Objective Loss 0.272397 LR 0.001000 Time 0.020268 -2022-12-06 10:45:35,254 - Epoch: [52][ 630/ 1200] Overall Loss 0.272178 Objective Loss 0.272178 LR 0.001000 Time 0.020250 -2022-12-06 10:45:35,446 - Epoch: [52][ 640/ 1200] Overall Loss 0.272074 Objective Loss 0.272074 LR 0.001000 Time 0.020232 -2022-12-06 10:45:35,637 - Epoch: [52][ 650/ 1200] Overall Loss 0.271955 Objective Loss 0.271955 LR 0.001000 Time 0.020215 -2022-12-06 10:45:35,828 - Epoch: [52][ 660/ 1200] Overall Loss 0.272311 Objective Loss 0.272311 LR 0.001000 Time 0.020197 -2022-12-06 10:45:36,019 - Epoch: [52][ 670/ 1200] Overall Loss 0.272280 Objective Loss 0.272280 LR 0.001000 Time 0.020179 -2022-12-06 10:45:36,210 - Epoch: [52][ 680/ 1200] Overall Loss 0.271713 Objective Loss 0.271713 LR 0.001000 Time 0.020162 -2022-12-06 10:45:36,401 - Epoch: [52][ 690/ 1200] Overall Loss 0.271638 Objective Loss 0.271638 LR 0.001000 Time 0.020146 -2022-12-06 10:45:36,592 - Epoch: [52][ 700/ 1200] Overall Loss 0.271859 Objective Loss 0.271859 LR 0.001000 Time 0.020130 -2022-12-06 10:45:36,783 - Epoch: [52][ 710/ 1200] Overall Loss 0.271907 Objective Loss 0.271907 LR 0.001000 Time 0.020116 -2022-12-06 10:45:36,974 - Epoch: [52][ 720/ 1200] Overall Loss 0.271983 Objective Loss 0.271983 LR 0.001000 Time 0.020101 -2022-12-06 10:45:37,166 - Epoch: [52][ 730/ 1200] Overall Loss 0.272432 Objective Loss 0.272432 LR 0.001000 Time 0.020087 -2022-12-06 10:45:37,356 - Epoch: [52][ 740/ 1200] Overall Loss 0.272720 Objective Loss 0.272720 LR 0.001000 Time 0.020072 -2022-12-06 10:45:37,547 - Epoch: [52][ 750/ 1200] Overall Loss 0.272618 Objective Loss 0.272618 LR 0.001000 Time 0.020058 -2022-12-06 10:45:37,737 - Epoch: [52][ 760/ 1200] Overall Loss 0.272836 Objective Loss 0.272836 LR 0.001000 Time 0.020044 -2022-12-06 10:45:37,928 - Epoch: [52][ 770/ 1200] Overall Loss 0.273319 Objective Loss 0.273319 LR 0.001000 Time 0.020031 -2022-12-06 10:45:38,119 - Epoch: [52][ 780/ 1200] Overall Loss 0.273016 Objective Loss 0.273016 LR 0.001000 Time 0.020018 -2022-12-06 10:45:38,311 - Epoch: [52][ 790/ 1200] Overall Loss 0.272856 Objective Loss 0.272856 LR 0.001000 Time 0.020006 -2022-12-06 10:45:38,502 - Epoch: [52][ 800/ 1200] Overall Loss 0.272658 Objective Loss 0.272658 LR 0.001000 Time 0.019995 -2022-12-06 10:45:38,693 - Epoch: [52][ 810/ 1200] Overall Loss 0.272883 Objective Loss 0.272883 LR 0.001000 Time 0.019984 -2022-12-06 10:45:38,884 - Epoch: [52][ 820/ 1200] Overall Loss 0.272784 Objective Loss 0.272784 LR 0.001000 Time 0.019971 -2022-12-06 10:45:39,074 - Epoch: [52][ 830/ 1200] Overall Loss 0.272869 Objective Loss 0.272869 LR 0.001000 Time 0.019960 -2022-12-06 10:45:39,265 - Epoch: [52][ 840/ 1200] Overall Loss 0.272811 Objective Loss 0.272811 LR 0.001000 Time 0.019948 -2022-12-06 10:45:39,456 - Epoch: [52][ 850/ 1200] Overall Loss 0.272861 Objective Loss 0.272861 LR 0.001000 Time 0.019937 -2022-12-06 10:45:39,647 - Epoch: [52][ 860/ 1200] Overall Loss 0.273219 Objective Loss 0.273219 LR 0.001000 Time 0.019927 -2022-12-06 10:45:39,838 - Epoch: [52][ 870/ 1200] Overall Loss 0.273245 Objective Loss 0.273245 LR 0.001000 Time 0.019917 -2022-12-06 10:45:40,029 - Epoch: [52][ 880/ 1200] Overall Loss 0.273237 Objective Loss 0.273237 LR 0.001000 Time 0.019907 -2022-12-06 10:45:40,220 - Epoch: [52][ 890/ 1200] Overall Loss 0.273231 Objective Loss 0.273231 LR 0.001000 Time 0.019898 -2022-12-06 10:45:40,411 - Epoch: [52][ 900/ 1200] Overall Loss 0.273380 Objective Loss 0.273380 LR 0.001000 Time 0.019888 -2022-12-06 10:45:40,602 - Epoch: [52][ 910/ 1200] Overall Loss 0.273110 Objective Loss 0.273110 LR 0.001000 Time 0.019879 -2022-12-06 10:45:40,793 - Epoch: [52][ 920/ 1200] Overall Loss 0.273329 Objective Loss 0.273329 LR 0.001000 Time 0.019870 -2022-12-06 10:45:40,984 - Epoch: [52][ 930/ 1200] Overall Loss 0.273409 Objective Loss 0.273409 LR 0.001000 Time 0.019861 -2022-12-06 10:45:41,175 - Epoch: [52][ 940/ 1200] Overall Loss 0.273505 Objective Loss 0.273505 LR 0.001000 Time 0.019853 -2022-12-06 10:45:41,366 - Epoch: [52][ 950/ 1200] Overall Loss 0.273579 Objective Loss 0.273579 LR 0.001000 Time 0.019844 -2022-12-06 10:45:41,557 - Epoch: [52][ 960/ 1200] Overall Loss 0.273465 Objective Loss 0.273465 LR 0.001000 Time 0.019836 -2022-12-06 10:45:41,747 - Epoch: [52][ 970/ 1200] Overall Loss 0.273492 Objective Loss 0.273492 LR 0.001000 Time 0.019827 -2022-12-06 10:45:41,938 - Epoch: [52][ 980/ 1200] Overall Loss 0.273270 Objective Loss 0.273270 LR 0.001000 Time 0.019819 -2022-12-06 10:45:42,130 - Epoch: [52][ 990/ 1200] Overall Loss 0.273624 Objective Loss 0.273624 LR 0.001000 Time 0.019811 -2022-12-06 10:45:42,321 - Epoch: [52][ 1000/ 1200] Overall Loss 0.273849 Objective Loss 0.273849 LR 0.001000 Time 0.019803 -2022-12-06 10:45:42,511 - Epoch: [52][ 1010/ 1200] Overall Loss 0.273964 Objective Loss 0.273964 LR 0.001000 Time 0.019795 -2022-12-06 10:45:42,702 - Epoch: [52][ 1020/ 1200] Overall Loss 0.274114 Objective Loss 0.274114 LR 0.001000 Time 0.019788 -2022-12-06 10:45:42,893 - Epoch: [52][ 1030/ 1200] Overall Loss 0.274667 Objective Loss 0.274667 LR 0.001000 Time 0.019781 -2022-12-06 10:45:43,083 - Epoch: [52][ 1040/ 1200] Overall Loss 0.274592 Objective Loss 0.274592 LR 0.001000 Time 0.019773 -2022-12-06 10:45:43,274 - Epoch: [52][ 1050/ 1200] Overall Loss 0.274751 Objective Loss 0.274751 LR 0.001000 Time 0.019766 -2022-12-06 10:45:43,465 - Epoch: [52][ 1060/ 1200] Overall Loss 0.274759 Objective Loss 0.274759 LR 0.001000 Time 0.019759 -2022-12-06 10:45:43,656 - Epoch: [52][ 1070/ 1200] Overall Loss 0.275038 Objective Loss 0.275038 LR 0.001000 Time 0.019753 -2022-12-06 10:45:43,847 - Epoch: [52][ 1080/ 1200] Overall Loss 0.275075 Objective Loss 0.275075 LR 0.001000 Time 0.019746 -2022-12-06 10:45:44,038 - Epoch: [52][ 1090/ 1200] Overall Loss 0.275035 Objective Loss 0.275035 LR 0.001000 Time 0.019740 -2022-12-06 10:45:44,229 - Epoch: [52][ 1100/ 1200] Overall Loss 0.274924 Objective Loss 0.274924 LR 0.001000 Time 0.019733 -2022-12-06 10:45:44,420 - Epoch: [52][ 1110/ 1200] Overall Loss 0.275196 Objective Loss 0.275196 LR 0.001000 Time 0.019727 -2022-12-06 10:45:44,611 - Epoch: [52][ 1120/ 1200] Overall Loss 0.275310 Objective Loss 0.275310 LR 0.001000 Time 0.019721 -2022-12-06 10:45:44,802 - Epoch: [52][ 1130/ 1200] Overall Loss 0.275277 Objective Loss 0.275277 LR 0.001000 Time 0.019715 -2022-12-06 10:45:44,993 - Epoch: [52][ 1140/ 1200] Overall Loss 0.275080 Objective Loss 0.275080 LR 0.001000 Time 0.019709 -2022-12-06 10:45:45,183 - Epoch: [52][ 1150/ 1200] Overall Loss 0.275064 Objective Loss 0.275064 LR 0.001000 Time 0.019703 -2022-12-06 10:45:45,374 - Epoch: [52][ 1160/ 1200] Overall Loss 0.275242 Objective Loss 0.275242 LR 0.001000 Time 0.019697 -2022-12-06 10:45:45,566 - Epoch: [52][ 1170/ 1200] Overall Loss 0.275224 Objective Loss 0.275224 LR 0.001000 Time 0.019692 -2022-12-06 10:45:45,757 - Epoch: [52][ 1180/ 1200] Overall Loss 0.275148 Objective Loss 0.275148 LR 0.001000 Time 0.019687 -2022-12-06 10:45:45,948 - Epoch: [52][ 1190/ 1200] Overall Loss 0.275091 Objective Loss 0.275091 LR 0.001000 Time 0.019681 -2022-12-06 10:45:46,177 - Epoch: [52][ 1200/ 1200] Overall Loss 0.275009 Objective Loss 0.275009 Top1 85.146444 Top5 97.071130 LR 0.001000 Time 0.019708 -2022-12-06 10:45:46,266 - --- validate (epoch=52)----------- -2022-12-06 10:45:46,266 - 34129 samples (256 per mini-batch) -2022-12-06 10:45:46,722 - Epoch: [52][ 10/ 134] Loss 0.305876 Top1 84.882812 Top5 97.578125 -2022-12-06 10:45:46,854 - Epoch: [52][ 20/ 134] Loss 0.299662 Top1 85.175781 Top5 97.617188 -2022-12-06 10:45:46,986 - Epoch: [52][ 30/ 134] Loss 0.285884 Top1 85.442708 Top5 97.812500 -2022-12-06 10:45:47,117 - Epoch: [52][ 40/ 134] Loss 0.285457 Top1 85.097656 Top5 97.841797 -2022-12-06 10:45:47,249 - Epoch: [52][ 50/ 134] Loss 0.295242 Top1 84.960938 Top5 97.828125 -2022-12-06 10:45:47,379 - Epoch: [52][ 60/ 134] Loss 0.292067 Top1 85.084635 Top5 97.890625 -2022-12-06 10:45:47,511 - Epoch: [52][ 70/ 134] Loss 0.289648 Top1 85.078125 Top5 97.890625 -2022-12-06 10:45:47,645 - Epoch: [52][ 80/ 134] Loss 0.289478 Top1 85.117188 Top5 98.022461 -2022-12-06 10:45:47,776 - Epoch: [52][ 90/ 134] Loss 0.290753 Top1 85.039062 Top5 97.999132 -2022-12-06 10:45:47,908 - Epoch: [52][ 100/ 134] Loss 0.291721 Top1 84.996094 Top5 97.992188 -2022-12-06 10:45:48,040 - Epoch: [52][ 110/ 134] Loss 0.292500 Top1 85.003551 Top5 97.933239 -2022-12-06 10:45:48,174 - Epoch: [52][ 120/ 134] Loss 0.292578 Top1 85.003255 Top5 97.958984 -2022-12-06 10:45:48,304 - Epoch: [52][ 130/ 134] Loss 0.294703 Top1 84.861779 Top5 97.959736 -2022-12-06 10:45:48,340 - Epoch: [52][ 134/ 134] Loss 0.295649 Top1 84.813502 Top5 97.943098 -2022-12-06 10:45:48,429 - ==> Top1: 84.814 Top5: 97.943 Loss: 0.296 - -2022-12-06 10:45:48,430 - ==> Confusion: -[[ 874 2 2 2 8 5 1 1 13 67 1 0 1 1 9 1 2 0 2 0 4] - [ 1 928 1 4 6 19 2 16 1 1 5 2 1 2 0 1 7 1 16 2 11] - [ 3 4 990 24 1 0 26 10 0 3 4 3 2 2 1 3 3 2 7 1 14] - [ 2 2 10 937 1 0 0 1 1 0 5 0 1 3 23 0 1 2 22 0 9] - [ 10 3 2 2 958 3 1 1 2 4 1 1 1 0 12 5 5 1 1 1 6] - [ 3 21 0 2 10 960 3 22 3 2 0 5 9 9 5 0 2 0 0 6 7] - [ 0 2 9 5 1 1 1067 6 0 0 3 0 1 3 0 7 1 2 2 7 1] - [ 1 8 8 1 3 29 7 922 0 0 1 6 0 3 1 0 0 1 45 14 4] - [ 5 5 0 0 0 2 0 0 971 52 5 0 2 6 9 0 2 0 4 1 0] - [ 52 0 2 0 5 1 0 4 21 900 1 0 0 3 3 2 0 0 2 0 5] - [ 0 2 5 9 3 0 2 2 17 1 939 2 4 12 4 0 2 0 9 2 4] - [ 5 4 1 0 0 18 1 5 3 0 0 951 29 7 0 6 5 3 1 11 1] - [ 1 1 0 5 1 2 2 1 0 0 0 30 873 3 3 11 1 16 4 4 11] - [ 1 2 0 0 2 10 0 3 19 28 6 5 7 922 1 2 4 2 0 1 8] - [ 5 5 2 10 5 2 0 2 19 5 0 1 2 3 1046 0 2 1 14 2 4] - [ 1 0 1 3 4 2 3 2 0 0 0 7 9 3 0 981 9 7 0 6 5] - [ 4 3 1 2 5 2 0 1 1 0 0 2 1 1 1 11 1023 2 0 7 5] - [ 3 1 1 2 0 1 2 2 2 1 0 13 9 4 1 15 2 972 1 2 2] - [ 2 2 3 7 1 2 0 13 1 0 8 3 2 1 11 0 2 0 945 4 1] - [ 2 5 1 0 2 8 8 9 0 0 1 8 7 10 3 1 6 3 3 996 7] - [ 120 209 162 130 118 156 89 162 119 115 127 97 340 326 176 136 239 97 243 280 9785]] - -2022-12-06 10:45:49,001 - ==> Best [Top1: 85.256 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 47] -2022-12-06 10:45:49,001 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:45:49,007 - - -2022-12-06 10:45:49,007 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:45:50,064 - Epoch: [53][ 10/ 1200] Overall Loss 0.287244 Objective Loss 0.287244 LR 0.001000 Time 0.105632 -2022-12-06 10:45:50,285 - Epoch: [53][ 20/ 1200] Overall Loss 0.280995 Objective Loss 0.280995 LR 0.001000 Time 0.063760 -2022-12-06 10:45:50,494 - Epoch: [53][ 30/ 1200] Overall Loss 0.278730 Objective Loss 0.278730 LR 0.001000 Time 0.049456 -2022-12-06 10:45:50,707 - Epoch: [53][ 40/ 1200] Overall Loss 0.269182 Objective Loss 0.269182 LR 0.001000 Time 0.042422 -2022-12-06 10:45:50,915 - Epoch: [53][ 50/ 1200] Overall Loss 0.270857 Objective Loss 0.270857 LR 0.001000 Time 0.038080 -2022-12-06 10:45:51,130 - Epoch: [53][ 60/ 1200] Overall Loss 0.269490 Objective Loss 0.269490 LR 0.001000 Time 0.035297 -2022-12-06 10:45:51,338 - Epoch: [53][ 70/ 1200] Overall Loss 0.266447 Objective Loss 0.266447 LR 0.001000 Time 0.033214 -2022-12-06 10:45:51,538 - Epoch: [53][ 80/ 1200] Overall Loss 0.265674 Objective Loss 0.265674 LR 0.001000 Time 0.031566 -2022-12-06 10:45:51,736 - Epoch: [53][ 90/ 1200] Overall Loss 0.262869 Objective Loss 0.262869 LR 0.001000 Time 0.030244 -2022-12-06 10:45:51,935 - Epoch: [53][ 100/ 1200] Overall Loss 0.259314 Objective Loss 0.259314 LR 0.001000 Time 0.029207 -2022-12-06 10:45:52,132 - Epoch: [53][ 110/ 1200] Overall Loss 0.259516 Objective Loss 0.259516 LR 0.001000 Time 0.028339 -2022-12-06 10:45:52,333 - Epoch: [53][ 120/ 1200] Overall Loss 0.261402 Objective Loss 0.261402 LR 0.001000 Time 0.027646 -2022-12-06 10:45:52,530 - Epoch: [53][ 130/ 1200] Overall Loss 0.262169 Objective Loss 0.262169 LR 0.001000 Time 0.027032 -2022-12-06 10:45:52,730 - Epoch: [53][ 140/ 1200] Overall Loss 0.262789 Objective Loss 0.262789 LR 0.001000 Time 0.026527 -2022-12-06 10:45:52,928 - Epoch: [53][ 150/ 1200] Overall Loss 0.262982 Objective Loss 0.262982 LR 0.001000 Time 0.026074 -2022-12-06 10:45:53,127 - Epoch: [53][ 160/ 1200] Overall Loss 0.264204 Objective Loss 0.264204 LR 0.001000 Time 0.025684 -2022-12-06 10:45:53,324 - Epoch: [53][ 170/ 1200] Overall Loss 0.264318 Objective Loss 0.264318 LR 0.001000 Time 0.025330 -2022-12-06 10:45:53,524 - Epoch: [53][ 180/ 1200] Overall Loss 0.263890 Objective Loss 0.263890 LR 0.001000 Time 0.025033 -2022-12-06 10:45:53,724 - Epoch: [53][ 190/ 1200] Overall Loss 0.264950 Objective Loss 0.264950 LR 0.001000 Time 0.024760 -2022-12-06 10:45:53,923 - Epoch: [53][ 200/ 1200] Overall Loss 0.264656 Objective Loss 0.264656 LR 0.001000 Time 0.024515 -2022-12-06 10:45:54,121 - Epoch: [53][ 210/ 1200] Overall Loss 0.265847 Objective Loss 0.265847 LR 0.001000 Time 0.024289 -2022-12-06 10:45:54,321 - Epoch: [53][ 220/ 1200] Overall Loss 0.263853 Objective Loss 0.263853 LR 0.001000 Time 0.024092 -2022-12-06 10:45:54,518 - Epoch: [53][ 230/ 1200] Overall Loss 0.264769 Objective Loss 0.264769 LR 0.001000 Time 0.023900 -2022-12-06 10:45:54,717 - Epoch: [53][ 240/ 1200] Overall Loss 0.264870 Objective Loss 0.264870 LR 0.001000 Time 0.023729 -2022-12-06 10:45:54,915 - Epoch: [53][ 250/ 1200] Overall Loss 0.264610 Objective Loss 0.264610 LR 0.001000 Time 0.023572 -2022-12-06 10:45:55,115 - Epoch: [53][ 260/ 1200] Overall Loss 0.265554 Objective Loss 0.265554 LR 0.001000 Time 0.023430 -2022-12-06 10:45:55,312 - Epoch: [53][ 270/ 1200] Overall Loss 0.267897 Objective Loss 0.267897 LR 0.001000 Time 0.023292 -2022-12-06 10:45:55,512 - Epoch: [53][ 280/ 1200] Overall Loss 0.268148 Objective Loss 0.268148 LR 0.001000 Time 0.023173 -2022-12-06 10:45:55,711 - Epoch: [53][ 290/ 1200] Overall Loss 0.268457 Objective Loss 0.268457 LR 0.001000 Time 0.023056 -2022-12-06 10:45:55,910 - Epoch: [53][ 300/ 1200] Overall Loss 0.267792 Objective Loss 0.267792 LR 0.001000 Time 0.022950 -2022-12-06 10:45:56,107 - Epoch: [53][ 310/ 1200] Overall Loss 0.268087 Objective Loss 0.268087 LR 0.001000 Time 0.022844 -2022-12-06 10:45:56,307 - Epoch: [53][ 320/ 1200] Overall Loss 0.268076 Objective Loss 0.268076 LR 0.001000 Time 0.022753 -2022-12-06 10:45:56,505 - Epoch: [53][ 330/ 1200] Overall Loss 0.268503 Objective Loss 0.268503 LR 0.001000 Time 0.022660 -2022-12-06 10:45:56,704 - Epoch: [53][ 340/ 1200] Overall Loss 0.268214 Objective Loss 0.268214 LR 0.001000 Time 0.022579 -2022-12-06 10:45:56,901 - Epoch: [53][ 350/ 1200] Overall Loss 0.268597 Objective Loss 0.268597 LR 0.001000 Time 0.022495 -2022-12-06 10:45:57,100 - Epoch: [53][ 360/ 1200] Overall Loss 0.267628 Objective Loss 0.267628 LR 0.001000 Time 0.022422 -2022-12-06 10:45:57,297 - Epoch: [53][ 370/ 1200] Overall Loss 0.268183 Objective Loss 0.268183 LR 0.001000 Time 0.022347 -2022-12-06 10:45:57,496 - Epoch: [53][ 380/ 1200] Overall Loss 0.269228 Objective Loss 0.269228 LR 0.001000 Time 0.022281 -2022-12-06 10:45:57,694 - Epoch: [53][ 390/ 1200] Overall Loss 0.268909 Objective Loss 0.268909 LR 0.001000 Time 0.022215 -2022-12-06 10:45:57,893 - Epoch: [53][ 400/ 1200] Overall Loss 0.268070 Objective Loss 0.268070 LR 0.001000 Time 0.022157 -2022-12-06 10:45:58,091 - Epoch: [53][ 410/ 1200] Overall Loss 0.269631 Objective Loss 0.269631 LR 0.001000 Time 0.022098 -2022-12-06 10:45:58,290 - Epoch: [53][ 420/ 1200] Overall Loss 0.269230 Objective Loss 0.269230 LR 0.001000 Time 0.022045 -2022-12-06 10:45:58,488 - Epoch: [53][ 430/ 1200] Overall Loss 0.269731 Objective Loss 0.269731 LR 0.001000 Time 0.021991 -2022-12-06 10:45:58,689 - Epoch: [53][ 440/ 1200] Overall Loss 0.269862 Objective Loss 0.269862 LR 0.001000 Time 0.021946 -2022-12-06 10:45:58,887 - Epoch: [53][ 450/ 1200] Overall Loss 0.270820 Objective Loss 0.270820 LR 0.001000 Time 0.021897 -2022-12-06 10:45:59,086 - Epoch: [53][ 460/ 1200] Overall Loss 0.270941 Objective Loss 0.270941 LR 0.001000 Time 0.021854 -2022-12-06 10:45:59,284 - Epoch: [53][ 470/ 1200] Overall Loss 0.271161 Objective Loss 0.271161 LR 0.001000 Time 0.021809 -2022-12-06 10:45:59,484 - Epoch: [53][ 480/ 1200] Overall Loss 0.271340 Objective Loss 0.271340 LR 0.001000 Time 0.021769 -2022-12-06 10:45:59,682 - Epoch: [53][ 490/ 1200] Overall Loss 0.270893 Objective Loss 0.270893 LR 0.001000 Time 0.021728 -2022-12-06 10:45:59,882 - Epoch: [53][ 500/ 1200] Overall Loss 0.271317 Objective Loss 0.271317 LR 0.001000 Time 0.021692 -2022-12-06 10:46:00,080 - Epoch: [53][ 510/ 1200] Overall Loss 0.271439 Objective Loss 0.271439 LR 0.001000 Time 0.021654 -2022-12-06 10:46:00,280 - Epoch: [53][ 520/ 1200] Overall Loss 0.271256 Objective Loss 0.271256 LR 0.001000 Time 0.021622 -2022-12-06 10:46:00,478 - Epoch: [53][ 530/ 1200] Overall Loss 0.271228 Objective Loss 0.271228 LR 0.001000 Time 0.021586 -2022-12-06 10:46:00,677 - Epoch: [53][ 540/ 1200] Overall Loss 0.270863 Objective Loss 0.270863 LR 0.001000 Time 0.021554 -2022-12-06 10:46:00,875 - Epoch: [53][ 550/ 1200] Overall Loss 0.270409 Objective Loss 0.270409 LR 0.001000 Time 0.021520 -2022-12-06 10:46:01,075 - Epoch: [53][ 560/ 1200] Overall Loss 0.270572 Objective Loss 0.270572 LR 0.001000 Time 0.021492 -2022-12-06 10:46:01,272 - Epoch: [53][ 570/ 1200] Overall Loss 0.270464 Objective Loss 0.270464 LR 0.001000 Time 0.021460 -2022-12-06 10:46:01,471 - Epoch: [53][ 580/ 1200] Overall Loss 0.269991 Objective Loss 0.269991 LR 0.001000 Time 0.021432 -2022-12-06 10:46:01,669 - Epoch: [53][ 590/ 1200] Overall Loss 0.270178 Objective Loss 0.270178 LR 0.001000 Time 0.021404 -2022-12-06 10:46:01,869 - Epoch: [53][ 600/ 1200] Overall Loss 0.270496 Objective Loss 0.270496 LR 0.001000 Time 0.021379 -2022-12-06 10:46:02,068 - Epoch: [53][ 610/ 1200] Overall Loss 0.271206 Objective Loss 0.271206 LR 0.001000 Time 0.021354 -2022-12-06 10:46:02,269 - Epoch: [53][ 620/ 1200] Overall Loss 0.271076 Objective Loss 0.271076 LR 0.001000 Time 0.021333 -2022-12-06 10:46:02,466 - Epoch: [53][ 630/ 1200] Overall Loss 0.271239 Objective Loss 0.271239 LR 0.001000 Time 0.021307 -2022-12-06 10:46:02,665 - Epoch: [53][ 640/ 1200] Overall Loss 0.271721 Objective Loss 0.271721 LR 0.001000 Time 0.021284 -2022-12-06 10:46:02,863 - Epoch: [53][ 650/ 1200] Overall Loss 0.271695 Objective Loss 0.271695 LR 0.001000 Time 0.021260 -2022-12-06 10:46:03,064 - Epoch: [53][ 660/ 1200] Overall Loss 0.272253 Objective Loss 0.272253 LR 0.001000 Time 0.021241 -2022-12-06 10:46:03,261 - Epoch: [53][ 670/ 1200] Overall Loss 0.272470 Objective Loss 0.272470 LR 0.001000 Time 0.021218 -2022-12-06 10:46:03,461 - Epoch: [53][ 680/ 1200] Overall Loss 0.272327 Objective Loss 0.272327 LR 0.001000 Time 0.021200 -2022-12-06 10:46:03,659 - Epoch: [53][ 690/ 1200] Overall Loss 0.272432 Objective Loss 0.272432 LR 0.001000 Time 0.021178 -2022-12-06 10:46:03,859 - Epoch: [53][ 700/ 1200] Overall Loss 0.272550 Objective Loss 0.272550 LR 0.001000 Time 0.021160 -2022-12-06 10:46:04,057 - Epoch: [53][ 710/ 1200] Overall Loss 0.272392 Objective Loss 0.272392 LR 0.001000 Time 0.021140 -2022-12-06 10:46:04,257 - Epoch: [53][ 720/ 1200] Overall Loss 0.272465 Objective Loss 0.272465 LR 0.001000 Time 0.021123 -2022-12-06 10:46:04,454 - Epoch: [53][ 730/ 1200] Overall Loss 0.272184 Objective Loss 0.272184 LR 0.001000 Time 0.021104 -2022-12-06 10:46:04,654 - Epoch: [53][ 740/ 1200] Overall Loss 0.272296 Objective Loss 0.272296 LR 0.001000 Time 0.021088 -2022-12-06 10:46:04,852 - Epoch: [53][ 750/ 1200] Overall Loss 0.272360 Objective Loss 0.272360 LR 0.001000 Time 0.021070 -2022-12-06 10:46:05,051 - Epoch: [53][ 760/ 1200] Overall Loss 0.272396 Objective Loss 0.272396 LR 0.001000 Time 0.021054 -2022-12-06 10:46:05,249 - Epoch: [53][ 770/ 1200] Overall Loss 0.272156 Objective Loss 0.272156 LR 0.001000 Time 0.021037 -2022-12-06 10:46:05,449 - Epoch: [53][ 780/ 1200] Overall Loss 0.272191 Objective Loss 0.272191 LR 0.001000 Time 0.021023 -2022-12-06 10:46:05,647 - Epoch: [53][ 790/ 1200] Overall Loss 0.272167 Objective Loss 0.272167 LR 0.001000 Time 0.021007 -2022-12-06 10:46:05,847 - Epoch: [53][ 800/ 1200] Overall Loss 0.272258 Objective Loss 0.272258 LR 0.001000 Time 0.020993 -2022-12-06 10:46:06,044 - Epoch: [53][ 810/ 1200] Overall Loss 0.272183 Objective Loss 0.272183 LR 0.001000 Time 0.020977 -2022-12-06 10:46:06,244 - Epoch: [53][ 820/ 1200] Overall Loss 0.272110 Objective Loss 0.272110 LR 0.001000 Time 0.020964 -2022-12-06 10:46:06,441 - Epoch: [53][ 830/ 1200] Overall Loss 0.272117 Objective Loss 0.272117 LR 0.001000 Time 0.020949 -2022-12-06 10:46:06,641 - Epoch: [53][ 840/ 1200] Overall Loss 0.272086 Objective Loss 0.272086 LR 0.001000 Time 0.020936 -2022-12-06 10:46:06,839 - Epoch: [53][ 850/ 1200] Overall Loss 0.272033 Objective Loss 0.272033 LR 0.001000 Time 0.020923 -2022-12-06 10:46:07,040 - Epoch: [53][ 860/ 1200] Overall Loss 0.271662 Objective Loss 0.271662 LR 0.001000 Time 0.020912 -2022-12-06 10:46:07,238 - Epoch: [53][ 870/ 1200] Overall Loss 0.271896 Objective Loss 0.271896 LR 0.001000 Time 0.020898 -2022-12-06 10:46:07,437 - Epoch: [53][ 880/ 1200] Overall Loss 0.271873 Objective Loss 0.271873 LR 0.001000 Time 0.020887 -2022-12-06 10:46:07,635 - Epoch: [53][ 890/ 1200] Overall Loss 0.272126 Objective Loss 0.272126 LR 0.001000 Time 0.020874 -2022-12-06 10:46:07,835 - Epoch: [53][ 900/ 1200] Overall Loss 0.272536 Objective Loss 0.272536 LR 0.001000 Time 0.020863 -2022-12-06 10:46:08,032 - Epoch: [53][ 910/ 1200] Overall Loss 0.272832 Objective Loss 0.272832 LR 0.001000 Time 0.020850 -2022-12-06 10:46:08,232 - Epoch: [53][ 920/ 1200] Overall Loss 0.272923 Objective Loss 0.272923 LR 0.001000 Time 0.020841 -2022-12-06 10:46:08,431 - Epoch: [53][ 930/ 1200] Overall Loss 0.272683 Objective Loss 0.272683 LR 0.001000 Time 0.020830 -2022-12-06 10:46:08,631 - Epoch: [53][ 940/ 1200] Overall Loss 0.272646 Objective Loss 0.272646 LR 0.001000 Time 0.020821 -2022-12-06 10:46:08,829 - Epoch: [53][ 950/ 1200] Overall Loss 0.272706 Objective Loss 0.272706 LR 0.001000 Time 0.020809 -2022-12-06 10:46:09,029 - Epoch: [53][ 960/ 1200] Overall Loss 0.272776 Objective Loss 0.272776 LR 0.001000 Time 0.020800 -2022-12-06 10:46:09,228 - Epoch: [53][ 970/ 1200] Overall Loss 0.272634 Objective Loss 0.272634 LR 0.001000 Time 0.020790 -2022-12-06 10:46:09,427 - Epoch: [53][ 980/ 1200] Overall Loss 0.272659 Objective Loss 0.272659 LR 0.001000 Time 0.020781 -2022-12-06 10:46:09,624 - Epoch: [53][ 990/ 1200] Overall Loss 0.272658 Objective Loss 0.272658 LR 0.001000 Time 0.020769 -2022-12-06 10:46:09,823 - Epoch: [53][ 1000/ 1200] Overall Loss 0.272691 Objective Loss 0.272691 LR 0.001000 Time 0.020760 -2022-12-06 10:46:10,021 - Epoch: [53][ 1010/ 1200] Overall Loss 0.272774 Objective Loss 0.272774 LR 0.001000 Time 0.020750 -2022-12-06 10:46:10,220 - Epoch: [53][ 1020/ 1200] Overall Loss 0.272906 Objective Loss 0.272906 LR 0.001000 Time 0.020741 -2022-12-06 10:46:10,418 - Epoch: [53][ 1030/ 1200] Overall Loss 0.272743 Objective Loss 0.272743 LR 0.001000 Time 0.020731 -2022-12-06 10:46:10,618 - Epoch: [53][ 1040/ 1200] Overall Loss 0.272935 Objective Loss 0.272935 LR 0.001000 Time 0.020723 -2022-12-06 10:46:10,815 - Epoch: [53][ 1050/ 1200] Overall Loss 0.272954 Objective Loss 0.272954 LR 0.001000 Time 0.020714 -2022-12-06 10:46:11,015 - Epoch: [53][ 1060/ 1200] Overall Loss 0.272826 Objective Loss 0.272826 LR 0.001000 Time 0.020707 -2022-12-06 10:46:11,213 - Epoch: [53][ 1070/ 1200] Overall Loss 0.272729 Objective Loss 0.272729 LR 0.001000 Time 0.020698 -2022-12-06 10:46:11,413 - Epoch: [53][ 1080/ 1200] Overall Loss 0.272620 Objective Loss 0.272620 LR 0.001000 Time 0.020691 -2022-12-06 10:46:11,612 - Epoch: [53][ 1090/ 1200] Overall Loss 0.272721 Objective Loss 0.272721 LR 0.001000 Time 0.020683 -2022-12-06 10:46:11,812 - Epoch: [53][ 1100/ 1200] Overall Loss 0.272814 Objective Loss 0.272814 LR 0.001000 Time 0.020676 -2022-12-06 10:46:12,011 - Epoch: [53][ 1110/ 1200] Overall Loss 0.272834 Objective Loss 0.272834 LR 0.001000 Time 0.020668 -2022-12-06 10:46:12,210 - Epoch: [53][ 1120/ 1200] Overall Loss 0.273204 Objective Loss 0.273204 LR 0.001000 Time 0.020661 -2022-12-06 10:46:12,407 - Epoch: [53][ 1130/ 1200] Overall Loss 0.273111 Objective Loss 0.273111 LR 0.001000 Time 0.020653 -2022-12-06 10:46:12,607 - Epoch: [53][ 1140/ 1200] Overall Loss 0.273325 Objective Loss 0.273325 LR 0.001000 Time 0.020646 -2022-12-06 10:46:12,806 - Epoch: [53][ 1150/ 1200] Overall Loss 0.273416 Objective Loss 0.273416 LR 0.001000 Time 0.020639 -2022-12-06 10:46:13,005 - Epoch: [53][ 1160/ 1200] Overall Loss 0.273377 Objective Loss 0.273377 LR 0.001000 Time 0.020633 -2022-12-06 10:46:13,203 - Epoch: [53][ 1170/ 1200] Overall Loss 0.273411 Objective Loss 0.273411 LR 0.001000 Time 0.020625 -2022-12-06 10:46:13,403 - Epoch: [53][ 1180/ 1200] Overall Loss 0.273391 Objective Loss 0.273391 LR 0.001000 Time 0.020619 -2022-12-06 10:46:13,601 - Epoch: [53][ 1190/ 1200] Overall Loss 0.273490 Objective Loss 0.273490 LR 0.001000 Time 0.020612 -2022-12-06 10:46:13,830 - Epoch: [53][ 1200/ 1200] Overall Loss 0.273701 Objective Loss 0.273701 Top1 86.401674 Top5 97.280335 LR 0.001000 Time 0.020630 -2022-12-06 10:46:13,920 - --- validate (epoch=53)----------- -2022-12-06 10:46:13,920 - 34129 samples (256 per mini-batch) -2022-12-06 10:46:14,362 - Epoch: [53][ 10/ 134] Loss 0.290103 Top1 85.859375 Top5 98.164062 -2022-12-06 10:46:14,493 - Epoch: [53][ 20/ 134] Loss 0.298530 Top1 85.644531 Top5 98.378906 -2022-12-06 10:46:14,625 - Epoch: [53][ 30/ 134] Loss 0.299520 Top1 85.859375 Top5 98.385417 -2022-12-06 10:46:14,760 - Epoch: [53][ 40/ 134] Loss 0.290939 Top1 86.201172 Top5 98.300781 -2022-12-06 10:46:14,892 - Epoch: [53][ 50/ 134] Loss 0.292135 Top1 86.132812 Top5 98.210938 -2022-12-06 10:46:15,026 - Epoch: [53][ 60/ 134] Loss 0.292334 Top1 86.087240 Top5 98.209635 -2022-12-06 10:46:15,158 - Epoch: [53][ 70/ 134] Loss 0.294785 Top1 86.149554 Top5 98.214286 -2022-12-06 10:46:15,292 - Epoch: [53][ 80/ 134] Loss 0.292456 Top1 86.240234 Top5 98.183594 -2022-12-06 10:46:15,424 - Epoch: [53][ 90/ 134] Loss 0.290924 Top1 86.189236 Top5 98.172743 -2022-12-06 10:46:15,555 - Epoch: [53][ 100/ 134] Loss 0.292506 Top1 86.113281 Top5 98.125000 -2022-12-06 10:46:15,684 - Epoch: [53][ 110/ 134] Loss 0.291671 Top1 86.036932 Top5 98.132102 -2022-12-06 10:46:15,813 - Epoch: [53][ 120/ 134] Loss 0.290826 Top1 86.093750 Top5 98.167318 -2022-12-06 10:46:15,941 - Epoch: [53][ 130/ 134] Loss 0.292439 Top1 86.042668 Top5 98.170072 -2022-12-06 10:46:15,978 - Epoch: [53][ 134/ 134] Loss 0.294283 Top1 86.029476 Top5 98.168713 -2022-12-06 10:46:16,074 - ==> Top1: 86.029 Top5: 98.169 Loss: 0.294 - -2022-12-06 10:46:16,074 - ==> Confusion: -[[ 903 1 0 1 9 5 1 2 9 50 0 1 1 3 3 2 0 0 0 1 4] - [ 1 904 0 4 9 26 4 21 0 0 7 2 2 4 2 0 5 3 19 4 10] - [ 3 5 994 14 2 3 25 11 0 3 8 3 3 0 4 6 1 1 6 1 10] - [ 5 2 20 922 0 2 0 1 0 0 11 0 6 6 18 0 0 4 15 2 6] - [ 10 7 2 0 947 2 1 2 1 9 1 0 2 2 8 5 9 2 2 1 7] - [ 5 14 1 3 7 964 4 13 2 1 0 13 6 13 2 0 4 0 0 10 7] - [ 0 0 18 2 1 5 1055 3 0 0 5 3 4 0 0 8 1 0 2 7 4] - [ 3 5 5 0 2 33 3 926 0 0 2 9 0 4 2 0 2 0 32 22 4] - [ 4 3 0 0 0 3 1 1 977 36 9 2 3 8 8 0 2 2 3 1 1] - [ 60 0 1 0 3 1 0 7 34 866 1 1 1 12 7 0 0 1 1 1 4] - [ 0 4 4 4 0 1 2 1 11 1 956 4 2 16 2 0 1 0 4 0 6] - [ 3 4 1 1 1 7 3 3 0 0 2 940 44 5 1 5 2 6 0 14 9] - [ 0 2 2 4 2 2 1 1 0 0 0 21 894 2 2 7 1 11 0 5 12] - [ 0 1 1 0 1 8 0 2 12 10 6 9 4 951 1 2 4 0 0 2 9] - [ 8 6 4 17 4 0 0 2 16 7 2 0 4 3 1038 0 0 1 8 1 9] - [ 4 1 1 1 3 3 2 0 0 0 0 7 6 6 0 984 6 7 0 3 9] - [ 3 4 2 1 3 0 0 0 1 1 1 2 3 1 0 6 1022 1 0 9 12] - [ 3 2 1 2 0 0 5 0 4 2 1 4 19 3 0 13 0 973 0 3 1] - [ 2 5 3 8 1 2 0 17 1 0 2 2 7 2 8 0 1 0 936 6 5] - [ 2 3 1 0 0 6 8 4 0 0 1 13 9 5 0 1 7 2 2 1009 7] - [ 156 152 162 82 88 152 60 150 94 78 160 102 340 306 137 99 172 77 203 260 10196]] - -2022-12-06 10:46:16,656 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:46:16,657 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:46:16,664 - - -2022-12-06 10:46:16,664 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:46:17,606 - Epoch: [54][ 10/ 1200] Overall Loss 0.275874 Objective Loss 0.275874 LR 0.001000 Time 0.094135 -2022-12-06 10:46:17,804 - Epoch: [54][ 20/ 1200] Overall Loss 0.261092 Objective Loss 0.261092 LR 0.001000 Time 0.056929 -2022-12-06 10:46:18,004 - Epoch: [54][ 30/ 1200] Overall Loss 0.261438 Objective Loss 0.261438 LR 0.001000 Time 0.044591 -2022-12-06 10:46:18,200 - Epoch: [54][ 40/ 1200] Overall Loss 0.261872 Objective Loss 0.261872 LR 0.001000 Time 0.038330 -2022-12-06 10:46:18,400 - Epoch: [54][ 50/ 1200] Overall Loss 0.261889 Objective Loss 0.261889 LR 0.001000 Time 0.034645 -2022-12-06 10:46:18,596 - Epoch: [54][ 60/ 1200] Overall Loss 0.261241 Objective Loss 0.261241 LR 0.001000 Time 0.032130 -2022-12-06 10:46:18,794 - Epoch: [54][ 70/ 1200] Overall Loss 0.258128 Objective Loss 0.258128 LR 0.001000 Time 0.030367 -2022-12-06 10:46:18,991 - Epoch: [54][ 80/ 1200] Overall Loss 0.255137 Objective Loss 0.255137 LR 0.001000 Time 0.029024 -2022-12-06 10:46:19,191 - Epoch: [54][ 90/ 1200] Overall Loss 0.255681 Objective Loss 0.255681 LR 0.001000 Time 0.028015 -2022-12-06 10:46:19,388 - Epoch: [54][ 100/ 1200] Overall Loss 0.255608 Objective Loss 0.255608 LR 0.001000 Time 0.027178 -2022-12-06 10:46:19,587 - Epoch: [54][ 110/ 1200] Overall Loss 0.254469 Objective Loss 0.254469 LR 0.001000 Time 0.026514 -2022-12-06 10:46:19,783 - Epoch: [54][ 120/ 1200] Overall Loss 0.258052 Objective Loss 0.258052 LR 0.001000 Time 0.025933 -2022-12-06 10:46:19,983 - Epoch: [54][ 130/ 1200] Overall Loss 0.257554 Objective Loss 0.257554 LR 0.001000 Time 0.025470 -2022-12-06 10:46:20,179 - Epoch: [54][ 140/ 1200] Overall Loss 0.256303 Objective Loss 0.256303 LR 0.001000 Time 0.025050 -2022-12-06 10:46:20,379 - Epoch: [54][ 150/ 1200] Overall Loss 0.257968 Objective Loss 0.257968 LR 0.001000 Time 0.024706 -2022-12-06 10:46:20,576 - Epoch: [54][ 160/ 1200] Overall Loss 0.260026 Objective Loss 0.260026 LR 0.001000 Time 0.024391 -2022-12-06 10:46:20,776 - Epoch: [54][ 170/ 1200] Overall Loss 0.260994 Objective Loss 0.260994 LR 0.001000 Time 0.024127 -2022-12-06 10:46:20,973 - Epoch: [54][ 180/ 1200] Overall Loss 0.260866 Objective Loss 0.260866 LR 0.001000 Time 0.023880 -2022-12-06 10:46:21,172 - Epoch: [54][ 190/ 1200] Overall Loss 0.261173 Objective Loss 0.261173 LR 0.001000 Time 0.023666 -2022-12-06 10:46:21,367 - Epoch: [54][ 200/ 1200] Overall Loss 0.261835 Objective Loss 0.261835 LR 0.001000 Time 0.023459 -2022-12-06 10:46:21,567 - Epoch: [54][ 210/ 1200] Overall Loss 0.262428 Objective Loss 0.262428 LR 0.001000 Time 0.023290 -2022-12-06 10:46:21,764 - Epoch: [54][ 220/ 1200] Overall Loss 0.261127 Objective Loss 0.261127 LR 0.001000 Time 0.023126 -2022-12-06 10:46:21,964 - Epoch: [54][ 230/ 1200] Overall Loss 0.262033 Objective Loss 0.262033 LR 0.001000 Time 0.022987 -2022-12-06 10:46:22,161 - Epoch: [54][ 240/ 1200] Overall Loss 0.261640 Objective Loss 0.261640 LR 0.001000 Time 0.022848 -2022-12-06 10:46:22,361 - Epoch: [54][ 250/ 1200] Overall Loss 0.261677 Objective Loss 0.261677 LR 0.001000 Time 0.022731 -2022-12-06 10:46:22,559 - Epoch: [54][ 260/ 1200] Overall Loss 0.261910 Objective Loss 0.261910 LR 0.001000 Time 0.022614 -2022-12-06 10:46:22,759 - Epoch: [54][ 270/ 1200] Overall Loss 0.262235 Objective Loss 0.262235 LR 0.001000 Time 0.022516 -2022-12-06 10:46:22,956 - Epoch: [54][ 280/ 1200] Overall Loss 0.262429 Objective Loss 0.262429 LR 0.001000 Time 0.022414 -2022-12-06 10:46:23,155 - Epoch: [54][ 290/ 1200] Overall Loss 0.263362 Objective Loss 0.263362 LR 0.001000 Time 0.022327 -2022-12-06 10:46:23,352 - Epoch: [54][ 300/ 1200] Overall Loss 0.262840 Objective Loss 0.262840 LR 0.001000 Time 0.022236 -2022-12-06 10:46:23,552 - Epoch: [54][ 310/ 1200] Overall Loss 0.262797 Objective Loss 0.262797 LR 0.001000 Time 0.022162 -2022-12-06 10:46:23,749 - Epoch: [54][ 320/ 1200] Overall Loss 0.262143 Objective Loss 0.262143 LR 0.001000 Time 0.022083 -2022-12-06 10:46:23,949 - Epoch: [54][ 330/ 1200] Overall Loss 0.261934 Objective Loss 0.261934 LR 0.001000 Time 0.022018 -2022-12-06 10:46:24,145 - Epoch: [54][ 340/ 1200] Overall Loss 0.261651 Objective Loss 0.261651 LR 0.001000 Time 0.021947 -2022-12-06 10:46:24,345 - Epoch: [54][ 350/ 1200] Overall Loss 0.261898 Objective Loss 0.261898 LR 0.001000 Time 0.021888 -2022-12-06 10:46:24,541 - Epoch: [54][ 360/ 1200] Overall Loss 0.262505 Objective Loss 0.262505 LR 0.001000 Time 0.021825 -2022-12-06 10:46:24,741 - Epoch: [54][ 370/ 1200] Overall Loss 0.263034 Objective Loss 0.263034 LR 0.001000 Time 0.021773 -2022-12-06 10:46:24,937 - Epoch: [54][ 380/ 1200] Overall Loss 0.263922 Objective Loss 0.263922 LR 0.001000 Time 0.021715 -2022-12-06 10:46:25,137 - Epoch: [54][ 390/ 1200] Overall Loss 0.264196 Objective Loss 0.264196 LR 0.001000 Time 0.021668 -2022-12-06 10:46:25,333 - Epoch: [54][ 400/ 1200] Overall Loss 0.264851 Objective Loss 0.264851 LR 0.001000 Time 0.021616 -2022-12-06 10:46:25,532 - Epoch: [54][ 410/ 1200] Overall Loss 0.264858 Objective Loss 0.264858 LR 0.001000 Time 0.021574 -2022-12-06 10:46:25,729 - Epoch: [54][ 420/ 1200] Overall Loss 0.264976 Objective Loss 0.264976 LR 0.001000 Time 0.021526 -2022-12-06 10:46:25,928 - Epoch: [54][ 430/ 1200] Overall Loss 0.264975 Objective Loss 0.264975 LR 0.001000 Time 0.021488 -2022-12-06 10:46:26,124 - Epoch: [54][ 440/ 1200] Overall Loss 0.265095 Objective Loss 0.265095 LR 0.001000 Time 0.021444 -2022-12-06 10:46:26,324 - Epoch: [54][ 450/ 1200] Overall Loss 0.265517 Objective Loss 0.265517 LR 0.001000 Time 0.021410 -2022-12-06 10:46:26,519 - Epoch: [54][ 460/ 1200] Overall Loss 0.266084 Objective Loss 0.266084 LR 0.001000 Time 0.021368 -2022-12-06 10:46:26,719 - Epoch: [54][ 470/ 1200] Overall Loss 0.266410 Objective Loss 0.266410 LR 0.001000 Time 0.021338 -2022-12-06 10:46:26,917 - Epoch: [54][ 480/ 1200] Overall Loss 0.267317 Objective Loss 0.267317 LR 0.001000 Time 0.021304 -2022-12-06 10:46:27,117 - Epoch: [54][ 490/ 1200] Overall Loss 0.267535 Objective Loss 0.267535 LR 0.001000 Time 0.021276 -2022-12-06 10:46:27,314 - Epoch: [54][ 500/ 1200] Overall Loss 0.267224 Objective Loss 0.267224 LR 0.001000 Time 0.021243 -2022-12-06 10:46:27,514 - Epoch: [54][ 510/ 1200] Overall Loss 0.267199 Objective Loss 0.267199 LR 0.001000 Time 0.021218 -2022-12-06 10:46:27,711 - Epoch: [54][ 520/ 1200] Overall Loss 0.266972 Objective Loss 0.266972 LR 0.001000 Time 0.021189 -2022-12-06 10:46:27,912 - Epoch: [54][ 530/ 1200] Overall Loss 0.267006 Objective Loss 0.267006 LR 0.001000 Time 0.021166 -2022-12-06 10:46:28,108 - Epoch: [54][ 540/ 1200] Overall Loss 0.267238 Objective Loss 0.267238 LR 0.001000 Time 0.021136 -2022-12-06 10:46:28,307 - Epoch: [54][ 550/ 1200] Overall Loss 0.267328 Objective Loss 0.267328 LR 0.001000 Time 0.021113 -2022-12-06 10:46:28,505 - Epoch: [54][ 560/ 1200] Overall Loss 0.267505 Objective Loss 0.267505 LR 0.001000 Time 0.021088 -2022-12-06 10:46:28,705 - Epoch: [54][ 570/ 1200] Overall Loss 0.267671 Objective Loss 0.267671 LR 0.001000 Time 0.021068 -2022-12-06 10:46:28,902 - Epoch: [54][ 580/ 1200] Overall Loss 0.267817 Objective Loss 0.267817 LR 0.001000 Time 0.021044 -2022-12-06 10:46:29,102 - Epoch: [54][ 590/ 1200] Overall Loss 0.268343 Objective Loss 0.268343 LR 0.001000 Time 0.021025 -2022-12-06 10:46:29,299 - Epoch: [54][ 600/ 1200] Overall Loss 0.268727 Objective Loss 0.268727 LR 0.001000 Time 0.021002 -2022-12-06 10:46:29,499 - Epoch: [54][ 610/ 1200] Overall Loss 0.268735 Objective Loss 0.268735 LR 0.001000 Time 0.020985 -2022-12-06 10:46:29,696 - Epoch: [54][ 620/ 1200] Overall Loss 0.268991 Objective Loss 0.268991 LR 0.001000 Time 0.020963 -2022-12-06 10:46:29,896 - Epoch: [54][ 630/ 1200] Overall Loss 0.269118 Objective Loss 0.269118 LR 0.001000 Time 0.020947 -2022-12-06 10:46:30,092 - Epoch: [54][ 640/ 1200] Overall Loss 0.269044 Objective Loss 0.269044 LR 0.001000 Time 0.020925 -2022-12-06 10:46:30,291 - Epoch: [54][ 650/ 1200] Overall Loss 0.269631 Objective Loss 0.269631 LR 0.001000 Time 0.020909 -2022-12-06 10:46:30,488 - Epoch: [54][ 660/ 1200] Overall Loss 0.269901 Objective Loss 0.269901 LR 0.001000 Time 0.020890 -2022-12-06 10:46:30,688 - Epoch: [54][ 670/ 1200] Overall Loss 0.270752 Objective Loss 0.270752 LR 0.001000 Time 0.020875 -2022-12-06 10:46:30,885 - Epoch: [54][ 680/ 1200] Overall Loss 0.270869 Objective Loss 0.270869 LR 0.001000 Time 0.020858 -2022-12-06 10:46:31,085 - Epoch: [54][ 690/ 1200] Overall Loss 0.271063 Objective Loss 0.271063 LR 0.001000 Time 0.020845 -2022-12-06 10:46:31,283 - Epoch: [54][ 700/ 1200] Overall Loss 0.271204 Objective Loss 0.271204 LR 0.001000 Time 0.020828 -2022-12-06 10:46:31,483 - Epoch: [54][ 710/ 1200] Overall Loss 0.271234 Objective Loss 0.271234 LR 0.001000 Time 0.020816 -2022-12-06 10:46:31,680 - Epoch: [54][ 720/ 1200] Overall Loss 0.271368 Objective Loss 0.271368 LR 0.001000 Time 0.020800 -2022-12-06 10:46:31,880 - Epoch: [54][ 730/ 1200] Overall Loss 0.271294 Objective Loss 0.271294 LR 0.001000 Time 0.020788 -2022-12-06 10:46:32,076 - Epoch: [54][ 740/ 1200] Overall Loss 0.271426 Objective Loss 0.271426 LR 0.001000 Time 0.020771 -2022-12-06 10:46:32,276 - Epoch: [54][ 750/ 1200] Overall Loss 0.270957 Objective Loss 0.270957 LR 0.001000 Time 0.020760 -2022-12-06 10:46:32,473 - Epoch: [54][ 760/ 1200] Overall Loss 0.270968 Objective Loss 0.270968 LR 0.001000 Time 0.020745 -2022-12-06 10:46:32,672 - Epoch: [54][ 770/ 1200] Overall Loss 0.270831 Objective Loss 0.270831 LR 0.001000 Time 0.020734 -2022-12-06 10:46:32,869 - Epoch: [54][ 780/ 1200] Overall Loss 0.270647 Objective Loss 0.270647 LR 0.001000 Time 0.020719 -2022-12-06 10:46:33,068 - Epoch: [54][ 790/ 1200] Overall Loss 0.270553 Objective Loss 0.270553 LR 0.001000 Time 0.020709 -2022-12-06 10:46:33,265 - Epoch: [54][ 800/ 1200] Overall Loss 0.270631 Objective Loss 0.270631 LR 0.001000 Time 0.020695 -2022-12-06 10:46:33,464 - Epoch: [54][ 810/ 1200] Overall Loss 0.270578 Objective Loss 0.270578 LR 0.001000 Time 0.020685 -2022-12-06 10:46:33,662 - Epoch: [54][ 820/ 1200] Overall Loss 0.270685 Objective Loss 0.270685 LR 0.001000 Time 0.020673 -2022-12-06 10:46:33,862 - Epoch: [54][ 830/ 1200] Overall Loss 0.271016 Objective Loss 0.271016 LR 0.001000 Time 0.020664 -2022-12-06 10:46:34,058 - Epoch: [54][ 840/ 1200] Overall Loss 0.270739 Objective Loss 0.270739 LR 0.001000 Time 0.020652 -2022-12-06 10:46:34,258 - Epoch: [54][ 850/ 1200] Overall Loss 0.270543 Objective Loss 0.270543 LR 0.001000 Time 0.020643 -2022-12-06 10:46:34,455 - Epoch: [54][ 860/ 1200] Overall Loss 0.270294 Objective Loss 0.270294 LR 0.001000 Time 0.020631 -2022-12-06 10:46:34,655 - Epoch: [54][ 870/ 1200] Overall Loss 0.270854 Objective Loss 0.270854 LR 0.001000 Time 0.020623 -2022-12-06 10:46:34,851 - Epoch: [54][ 880/ 1200] Overall Loss 0.270639 Objective Loss 0.270639 LR 0.001000 Time 0.020611 -2022-12-06 10:46:35,050 - Epoch: [54][ 890/ 1200] Overall Loss 0.270870 Objective Loss 0.270870 LR 0.001000 Time 0.020603 -2022-12-06 10:46:35,246 - Epoch: [54][ 900/ 1200] Overall Loss 0.270917 Objective Loss 0.270917 LR 0.001000 Time 0.020591 -2022-12-06 10:46:35,447 - Epoch: [54][ 910/ 1200] Overall Loss 0.271137 Objective Loss 0.271137 LR 0.001000 Time 0.020584 -2022-12-06 10:46:35,644 - Epoch: [54][ 920/ 1200] Overall Loss 0.271246 Objective Loss 0.271246 LR 0.001000 Time 0.020574 -2022-12-06 10:46:35,843 - Epoch: [54][ 930/ 1200] Overall Loss 0.271032 Objective Loss 0.271032 LR 0.001000 Time 0.020567 -2022-12-06 10:46:36,040 - Epoch: [54][ 940/ 1200] Overall Loss 0.271044 Objective Loss 0.271044 LR 0.001000 Time 0.020557 -2022-12-06 10:46:36,240 - Epoch: [54][ 950/ 1200] Overall Loss 0.270669 Objective Loss 0.270669 LR 0.001000 Time 0.020550 -2022-12-06 10:46:36,437 - Epoch: [54][ 960/ 1200] Overall Loss 0.271031 Objective Loss 0.271031 LR 0.001000 Time 0.020541 -2022-12-06 10:46:36,637 - Epoch: [54][ 970/ 1200] Overall Loss 0.270969 Objective Loss 0.270969 LR 0.001000 Time 0.020535 -2022-12-06 10:46:36,836 - Epoch: [54][ 980/ 1200] Overall Loss 0.271037 Objective Loss 0.271037 LR 0.001000 Time 0.020528 -2022-12-06 10:46:37,038 - Epoch: [54][ 990/ 1200] Overall Loss 0.270905 Objective Loss 0.270905 LR 0.001000 Time 0.020524 -2022-12-06 10:46:37,236 - Epoch: [54][ 1000/ 1200] Overall Loss 0.271133 Objective Loss 0.271133 LR 0.001000 Time 0.020516 -2022-12-06 10:46:37,437 - Epoch: [54][ 1010/ 1200] Overall Loss 0.271328 Objective Loss 0.271328 LR 0.001000 Time 0.020511 -2022-12-06 10:46:37,635 - Epoch: [54][ 1020/ 1200] Overall Loss 0.271519 Objective Loss 0.271519 LR 0.001000 Time 0.020504 -2022-12-06 10:46:37,837 - Epoch: [54][ 1030/ 1200] Overall Loss 0.271311 Objective Loss 0.271311 LR 0.001000 Time 0.020501 -2022-12-06 10:46:38,036 - Epoch: [54][ 1040/ 1200] Overall Loss 0.271345 Objective Loss 0.271345 LR 0.001000 Time 0.020494 -2022-12-06 10:46:38,238 - Epoch: [54][ 1050/ 1200] Overall Loss 0.271231 Objective Loss 0.271231 LR 0.001000 Time 0.020491 -2022-12-06 10:46:38,437 - Epoch: [54][ 1060/ 1200] Overall Loss 0.271588 Objective Loss 0.271588 LR 0.001000 Time 0.020484 -2022-12-06 10:46:38,638 - Epoch: [54][ 1070/ 1200] Overall Loss 0.271835 Objective Loss 0.271835 LR 0.001000 Time 0.020480 -2022-12-06 10:46:38,836 - Epoch: [54][ 1080/ 1200] Overall Loss 0.271789 Objective Loss 0.271789 LR 0.001000 Time 0.020474 -2022-12-06 10:46:39,037 - Epoch: [54][ 1090/ 1200] Overall Loss 0.272128 Objective Loss 0.272128 LR 0.001000 Time 0.020470 -2022-12-06 10:46:39,237 - Epoch: [54][ 1100/ 1200] Overall Loss 0.272476 Objective Loss 0.272476 LR 0.001000 Time 0.020465 -2022-12-06 10:46:39,437 - Epoch: [54][ 1110/ 1200] Overall Loss 0.272497 Objective Loss 0.272497 LR 0.001000 Time 0.020460 -2022-12-06 10:46:39,635 - Epoch: [54][ 1120/ 1200] Overall Loss 0.272384 Objective Loss 0.272384 LR 0.001000 Time 0.020454 -2022-12-06 10:46:39,837 - Epoch: [54][ 1130/ 1200] Overall Loss 0.272666 Objective Loss 0.272666 LR 0.001000 Time 0.020451 -2022-12-06 10:46:40,035 - Epoch: [54][ 1140/ 1200] Overall Loss 0.272827 Objective Loss 0.272827 LR 0.001000 Time 0.020445 -2022-12-06 10:46:40,236 - Epoch: [54][ 1150/ 1200] Overall Loss 0.272914 Objective Loss 0.272914 LR 0.001000 Time 0.020441 -2022-12-06 10:46:40,435 - Epoch: [54][ 1160/ 1200] Overall Loss 0.272723 Objective Loss 0.272723 LR 0.001000 Time 0.020436 -2022-12-06 10:46:40,636 - Epoch: [54][ 1170/ 1200] Overall Loss 0.272798 Objective Loss 0.272798 LR 0.001000 Time 0.020433 -2022-12-06 10:46:40,834 - Epoch: [54][ 1180/ 1200] Overall Loss 0.273100 Objective Loss 0.273100 LR 0.001000 Time 0.020427 -2022-12-06 10:46:41,036 - Epoch: [54][ 1190/ 1200] Overall Loss 0.272928 Objective Loss 0.272928 LR 0.001000 Time 0.020424 -2022-12-06 10:46:41,269 - Epoch: [54][ 1200/ 1200] Overall Loss 0.273087 Objective Loss 0.273087 Top1 84.728033 Top5 98.744770 LR 0.001000 Time 0.020448 -2022-12-06 10:46:41,358 - --- validate (epoch=54)----------- -2022-12-06 10:46:41,358 - 34129 samples (256 per mini-batch) -2022-12-06 10:46:41,817 - Epoch: [54][ 10/ 134] Loss 0.286818 Top1 84.335938 Top5 97.695312 -2022-12-06 10:46:41,950 - Epoch: [54][ 20/ 134] Loss 0.311939 Top1 83.964844 Top5 97.617188 -2022-12-06 10:46:42,086 - Epoch: [54][ 30/ 134] Loss 0.307546 Top1 83.554688 Top5 97.786458 -2022-12-06 10:46:42,223 - Epoch: [54][ 40/ 134] Loss 0.308962 Top1 83.486328 Top5 97.812500 -2022-12-06 10:46:42,362 - Epoch: [54][ 50/ 134] Loss 0.302735 Top1 83.593750 Top5 97.906250 -2022-12-06 10:46:42,498 - Epoch: [54][ 60/ 134] Loss 0.294451 Top1 83.977865 Top5 97.955729 -2022-12-06 10:46:42,637 - Epoch: [54][ 70/ 134] Loss 0.294659 Top1 83.895089 Top5 97.946429 -2022-12-06 10:46:42,771 - Epoch: [54][ 80/ 134] Loss 0.296086 Top1 83.876953 Top5 97.954102 -2022-12-06 10:46:42,911 - Epoch: [54][ 90/ 134] Loss 0.295766 Top1 83.880208 Top5 97.907986 -2022-12-06 10:46:43,045 - Epoch: [54][ 100/ 134] Loss 0.295537 Top1 84.050781 Top5 97.886719 -2022-12-06 10:46:43,184 - Epoch: [54][ 110/ 134] Loss 0.294995 Top1 84.009233 Top5 97.919034 -2022-12-06 10:46:43,321 - Epoch: [54][ 120/ 134] Loss 0.297955 Top1 83.883464 Top5 97.900391 -2022-12-06 10:46:43,457 - Epoch: [54][ 130/ 134] Loss 0.297221 Top1 83.867188 Top5 97.893630 -2022-12-06 10:46:43,495 - Epoch: [54][ 134/ 134] Loss 0.297112 Top1 83.937414 Top5 97.916728 -2022-12-06 10:46:43,583 - ==> Top1: 83.937 Top5: 97.917 Loss: 0.297 - -2022-12-06 10:46:43,584 - ==> Confusion: -[[ 883 1 5 1 5 7 0 0 6 64 0 3 4 2 9 1 2 1 1 0 1] - [ 0 953 1 3 4 18 1 8 2 2 3 4 4 4 3 0 3 1 5 1 7] - [ 3 7 991 10 1 3 37 10 1 1 8 5 1 0 4 3 1 2 3 4 8] - [ 1 3 21 914 0 4 2 0 2 1 10 2 5 3 21 1 1 8 11 1 9] - [ 13 9 2 1 936 3 0 0 2 8 1 3 1 1 16 4 9 2 2 4 3] - [ 3 23 0 2 7 961 2 13 2 1 2 17 2 14 4 1 0 0 1 8 6] - [ 1 6 9 3 0 5 1063 3 0 0 2 5 0 1 1 4 1 0 2 7 5] - [ 0 14 5 1 3 45 4 923 1 0 1 10 1 1 1 0 0 2 21 16 5] - [ 9 3 1 0 0 3 0 0 972 36 7 5 1 6 12 1 3 1 2 1 1] - [ 55 0 1 1 0 3 0 2 36 879 1 3 0 7 8 0 0 0 0 0 5] - [ 1 3 3 4 0 2 1 4 15 1 955 3 3 13 4 0 2 0 1 1 3] - [ 5 1 2 0 0 5 2 2 0 0 1 981 21 7 1 3 3 4 1 8 4] - [ 1 2 3 2 0 2 0 1 0 0 0 46 884 3 0 8 2 6 1 3 5] - [ 0 2 0 0 0 12 0 1 13 19 6 11 1 941 2 1 3 3 0 4 4] - [ 5 9 1 7 3 2 0 2 14 3 0 3 1 3 1066 0 2 1 3 1 4] - [ 1 2 2 0 4 1 2 2 0 0 1 14 4 4 0 974 12 11 0 3 6] - [ 2 5 1 0 1 1 0 0 0 1 1 8 2 2 1 5 1026 3 0 8 5] - [ 3 1 1 2 0 0 0 0 2 1 0 18 16 1 1 13 1 974 0 2 0] - [ 0 12 5 9 0 6 1 23 2 0 9 1 6 2 10 0 2 1 912 4 3] - [ 4 7 1 0 0 5 6 7 0 0 0 21 5 8 1 1 5 4 0 999 6] - [ 130 324 182 103 106 219 81 147 102 115 231 150 411 361 196 120 254 104 157 276 9457]] - -2022-12-06 10:46:44,251 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:46:44,251 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:46:44,257 - - -2022-12-06 10:46:44,257 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:46:45,202 - Epoch: [55][ 10/ 1200] Overall Loss 0.274603 Objective Loss 0.274603 LR 0.001000 Time 0.094443 -2022-12-06 10:46:45,404 - Epoch: [55][ 20/ 1200] Overall Loss 0.266918 Objective Loss 0.266918 LR 0.001000 Time 0.057295 -2022-12-06 10:46:45,597 - Epoch: [55][ 30/ 1200] Overall Loss 0.273444 Objective Loss 0.273444 LR 0.001000 Time 0.044619 -2022-12-06 10:46:45,790 - Epoch: [55][ 40/ 1200] Overall Loss 0.270932 Objective Loss 0.270932 LR 0.001000 Time 0.038270 -2022-12-06 10:46:45,982 - Epoch: [55][ 50/ 1200] Overall Loss 0.269365 Objective Loss 0.269365 LR 0.001000 Time 0.034448 -2022-12-06 10:46:46,175 - Epoch: [55][ 60/ 1200] Overall Loss 0.276736 Objective Loss 0.276736 LR 0.001000 Time 0.031905 -2022-12-06 10:46:46,367 - Epoch: [55][ 70/ 1200] Overall Loss 0.275426 Objective Loss 0.275426 LR 0.001000 Time 0.030085 -2022-12-06 10:46:46,559 - Epoch: [55][ 80/ 1200] Overall Loss 0.271341 Objective Loss 0.271341 LR 0.001000 Time 0.028714 -2022-12-06 10:46:46,751 - Epoch: [55][ 90/ 1200] Overall Loss 0.272482 Objective Loss 0.272482 LR 0.001000 Time 0.027655 -2022-12-06 10:46:46,943 - Epoch: [55][ 100/ 1200] Overall Loss 0.271335 Objective Loss 0.271335 LR 0.001000 Time 0.026807 -2022-12-06 10:46:47,135 - Epoch: [55][ 110/ 1200] Overall Loss 0.268297 Objective Loss 0.268297 LR 0.001000 Time 0.026112 -2022-12-06 10:46:47,327 - Epoch: [55][ 120/ 1200] Overall Loss 0.268315 Objective Loss 0.268315 LR 0.001000 Time 0.025529 -2022-12-06 10:46:47,520 - Epoch: [55][ 130/ 1200] Overall Loss 0.266161 Objective Loss 0.266161 LR 0.001000 Time 0.025040 -2022-12-06 10:46:47,712 - Epoch: [55][ 140/ 1200] Overall Loss 0.265520 Objective Loss 0.265520 LR 0.001000 Time 0.024621 -2022-12-06 10:46:47,904 - Epoch: [55][ 150/ 1200] Overall Loss 0.266374 Objective Loss 0.266374 LR 0.001000 Time 0.024257 -2022-12-06 10:46:48,095 - Epoch: [55][ 160/ 1200] Overall Loss 0.266256 Objective Loss 0.266256 LR 0.001000 Time 0.023934 -2022-12-06 10:46:48,288 - Epoch: [55][ 170/ 1200] Overall Loss 0.266827 Objective Loss 0.266827 LR 0.001000 Time 0.023657 -2022-12-06 10:46:48,481 - Epoch: [55][ 180/ 1200] Overall Loss 0.266079 Objective Loss 0.266079 LR 0.001000 Time 0.023409 -2022-12-06 10:46:48,673 - Epoch: [55][ 190/ 1200] Overall Loss 0.266629 Objective Loss 0.266629 LR 0.001000 Time 0.023186 -2022-12-06 10:46:48,865 - Epoch: [55][ 200/ 1200] Overall Loss 0.266417 Objective Loss 0.266417 LR 0.001000 Time 0.022985 -2022-12-06 10:46:49,058 - Epoch: [55][ 210/ 1200] Overall Loss 0.266244 Objective Loss 0.266244 LR 0.001000 Time 0.022805 -2022-12-06 10:46:49,250 - Epoch: [55][ 220/ 1200] Overall Loss 0.266943 Objective Loss 0.266943 LR 0.001000 Time 0.022640 -2022-12-06 10:46:49,442 - Epoch: [55][ 230/ 1200] Overall Loss 0.266088 Objective Loss 0.266088 LR 0.001000 Time 0.022490 -2022-12-06 10:46:49,635 - Epoch: [55][ 240/ 1200] Overall Loss 0.265581 Objective Loss 0.265581 LR 0.001000 Time 0.022352 -2022-12-06 10:46:49,827 - Epoch: [55][ 250/ 1200] Overall Loss 0.267186 Objective Loss 0.267186 LR 0.001000 Time 0.022224 -2022-12-06 10:46:50,019 - Epoch: [55][ 260/ 1200] Overall Loss 0.267161 Objective Loss 0.267161 LR 0.001000 Time 0.022107 -2022-12-06 10:46:50,212 - Epoch: [55][ 270/ 1200] Overall Loss 0.267508 Objective Loss 0.267508 LR 0.001000 Time 0.022000 -2022-12-06 10:46:50,404 - Epoch: [55][ 280/ 1200] Overall Loss 0.268063 Objective Loss 0.268063 LR 0.001000 Time 0.021899 -2022-12-06 10:46:50,597 - Epoch: [55][ 290/ 1200] Overall Loss 0.267813 Objective Loss 0.267813 LR 0.001000 Time 0.021807 -2022-12-06 10:46:50,790 - Epoch: [55][ 300/ 1200] Overall Loss 0.268602 Objective Loss 0.268602 LR 0.001000 Time 0.021721 -2022-12-06 10:46:50,982 - Epoch: [55][ 310/ 1200] Overall Loss 0.268140 Objective Loss 0.268140 LR 0.001000 Time 0.021639 -2022-12-06 10:46:51,175 - Epoch: [55][ 320/ 1200] Overall Loss 0.267534 Objective Loss 0.267534 LR 0.001000 Time 0.021563 -2022-12-06 10:46:51,367 - Epoch: [55][ 330/ 1200] Overall Loss 0.267333 Objective Loss 0.267333 LR 0.001000 Time 0.021490 -2022-12-06 10:46:51,559 - Epoch: [55][ 340/ 1200] Overall Loss 0.267661 Objective Loss 0.267661 LR 0.001000 Time 0.021421 -2022-12-06 10:46:51,751 - Epoch: [55][ 350/ 1200] Overall Loss 0.267680 Objective Loss 0.267680 LR 0.001000 Time 0.021357 -2022-12-06 10:46:51,943 - Epoch: [55][ 360/ 1200] Overall Loss 0.267161 Objective Loss 0.267161 LR 0.001000 Time 0.021295 -2022-12-06 10:46:52,136 - Epoch: [55][ 370/ 1200] Overall Loss 0.266390 Objective Loss 0.266390 LR 0.001000 Time 0.021238 -2022-12-06 10:46:52,328 - Epoch: [55][ 380/ 1200] Overall Loss 0.266677 Objective Loss 0.266677 LR 0.001000 Time 0.021182 -2022-12-06 10:46:52,520 - Epoch: [55][ 390/ 1200] Overall Loss 0.266932 Objective Loss 0.266932 LR 0.001000 Time 0.021131 -2022-12-06 10:46:52,711 - Epoch: [55][ 400/ 1200] Overall Loss 0.267406 Objective Loss 0.267406 LR 0.001000 Time 0.021080 -2022-12-06 10:46:52,904 - Epoch: [55][ 410/ 1200] Overall Loss 0.267283 Objective Loss 0.267283 LR 0.001000 Time 0.021035 -2022-12-06 10:46:53,096 - Epoch: [55][ 420/ 1200] Overall Loss 0.267261 Objective Loss 0.267261 LR 0.001000 Time 0.020990 -2022-12-06 10:46:53,289 - Epoch: [55][ 430/ 1200] Overall Loss 0.267839 Objective Loss 0.267839 LR 0.001000 Time 0.020949 -2022-12-06 10:46:53,481 - Epoch: [55][ 440/ 1200] Overall Loss 0.267885 Objective Loss 0.267885 LR 0.001000 Time 0.020908 -2022-12-06 10:46:53,674 - Epoch: [55][ 450/ 1200] Overall Loss 0.267651 Objective Loss 0.267651 LR 0.001000 Time 0.020870 -2022-12-06 10:46:53,866 - Epoch: [55][ 460/ 1200] Overall Loss 0.267933 Objective Loss 0.267933 LR 0.001000 Time 0.020833 -2022-12-06 10:46:54,058 - Epoch: [55][ 470/ 1200] Overall Loss 0.267886 Objective Loss 0.267886 LR 0.001000 Time 0.020798 -2022-12-06 10:46:54,250 - Epoch: [55][ 480/ 1200] Overall Loss 0.267857 Objective Loss 0.267857 LR 0.001000 Time 0.020763 -2022-12-06 10:46:54,443 - Epoch: [55][ 490/ 1200] Overall Loss 0.267867 Objective Loss 0.267867 LR 0.001000 Time 0.020732 -2022-12-06 10:46:54,635 - Epoch: [55][ 500/ 1200] Overall Loss 0.267981 Objective Loss 0.267981 LR 0.001000 Time 0.020700 -2022-12-06 10:46:54,827 - Epoch: [55][ 510/ 1200] Overall Loss 0.267893 Objective Loss 0.267893 LR 0.001000 Time 0.020670 -2022-12-06 10:46:55,019 - Epoch: [55][ 520/ 1200] Overall Loss 0.267847 Objective Loss 0.267847 LR 0.001000 Time 0.020640 -2022-12-06 10:46:55,211 - Epoch: [55][ 530/ 1200] Overall Loss 0.267391 Objective Loss 0.267391 LR 0.001000 Time 0.020612 -2022-12-06 10:46:55,404 - Epoch: [55][ 540/ 1200] Overall Loss 0.267168 Objective Loss 0.267168 LR 0.001000 Time 0.020586 -2022-12-06 10:46:55,596 - Epoch: [55][ 550/ 1200] Overall Loss 0.267199 Objective Loss 0.267199 LR 0.001000 Time 0.020561 -2022-12-06 10:46:55,788 - Epoch: [55][ 560/ 1200] Overall Loss 0.266589 Objective Loss 0.266589 LR 0.001000 Time 0.020535 -2022-12-06 10:46:55,980 - Epoch: [55][ 570/ 1200] Overall Loss 0.267160 Objective Loss 0.267160 LR 0.001000 Time 0.020511 -2022-12-06 10:46:56,172 - Epoch: [55][ 580/ 1200] Overall Loss 0.267509 Objective Loss 0.267509 LR 0.001000 Time 0.020488 -2022-12-06 10:46:56,364 - Epoch: [55][ 590/ 1200] Overall Loss 0.267409 Objective Loss 0.267409 LR 0.001000 Time 0.020465 -2022-12-06 10:46:56,557 - Epoch: [55][ 600/ 1200] Overall Loss 0.268092 Objective Loss 0.268092 LR 0.001000 Time 0.020444 -2022-12-06 10:46:56,749 - Epoch: [55][ 610/ 1200] Overall Loss 0.267887 Objective Loss 0.267887 LR 0.001000 Time 0.020423 -2022-12-06 10:46:56,942 - Epoch: [55][ 620/ 1200] Overall Loss 0.268727 Objective Loss 0.268727 LR 0.001000 Time 0.020403 -2022-12-06 10:46:57,134 - Epoch: [55][ 630/ 1200] Overall Loss 0.268872 Objective Loss 0.268872 LR 0.001000 Time 0.020383 -2022-12-06 10:46:57,326 - Epoch: [55][ 640/ 1200] Overall Loss 0.269483 Objective Loss 0.269483 LR 0.001000 Time 0.020364 -2022-12-06 10:46:57,518 - Epoch: [55][ 650/ 1200] Overall Loss 0.270093 Objective Loss 0.270093 LR 0.001000 Time 0.020346 -2022-12-06 10:46:57,710 - Epoch: [55][ 660/ 1200] Overall Loss 0.270543 Objective Loss 0.270543 LR 0.001000 Time 0.020328 -2022-12-06 10:46:57,903 - Epoch: [55][ 670/ 1200] Overall Loss 0.270452 Objective Loss 0.270452 LR 0.001000 Time 0.020311 -2022-12-06 10:46:58,095 - Epoch: [55][ 680/ 1200] Overall Loss 0.270206 Objective Loss 0.270206 LR 0.001000 Time 0.020295 -2022-12-06 10:46:58,288 - Epoch: [55][ 690/ 1200] Overall Loss 0.270216 Objective Loss 0.270216 LR 0.001000 Time 0.020279 -2022-12-06 10:46:58,480 - Epoch: [55][ 700/ 1200] Overall Loss 0.270179 Objective Loss 0.270179 LR 0.001000 Time 0.020263 -2022-12-06 10:46:58,673 - Epoch: [55][ 710/ 1200] Overall Loss 0.269850 Objective Loss 0.269850 LR 0.001000 Time 0.020249 -2022-12-06 10:46:58,865 - Epoch: [55][ 720/ 1200] Overall Loss 0.269875 Objective Loss 0.269875 LR 0.001000 Time 0.020233 -2022-12-06 10:46:59,058 - Epoch: [55][ 730/ 1200] Overall Loss 0.270076 Objective Loss 0.270076 LR 0.001000 Time 0.020219 -2022-12-06 10:46:59,250 - Epoch: [55][ 740/ 1200] Overall Loss 0.270317 Objective Loss 0.270317 LR 0.001000 Time 0.020205 -2022-12-06 10:46:59,442 - Epoch: [55][ 750/ 1200] Overall Loss 0.270301 Objective Loss 0.270301 LR 0.001000 Time 0.020191 -2022-12-06 10:46:59,634 - Epoch: [55][ 760/ 1200] Overall Loss 0.270434 Objective Loss 0.270434 LR 0.001000 Time 0.020177 -2022-12-06 10:46:59,827 - Epoch: [55][ 770/ 1200] Overall Loss 0.270489 Objective Loss 0.270489 LR 0.001000 Time 0.020165 -2022-12-06 10:47:00,019 - Epoch: [55][ 780/ 1200] Overall Loss 0.270579 Objective Loss 0.270579 LR 0.001000 Time 0.020152 -2022-12-06 10:47:00,212 - Epoch: [55][ 790/ 1200] Overall Loss 0.270784 Objective Loss 0.270784 LR 0.001000 Time 0.020140 -2022-12-06 10:47:00,404 - Epoch: [55][ 800/ 1200] Overall Loss 0.270899 Objective Loss 0.270899 LR 0.001000 Time 0.020128 -2022-12-06 10:47:00,596 - Epoch: [55][ 810/ 1200] Overall Loss 0.270821 Objective Loss 0.270821 LR 0.001000 Time 0.020116 -2022-12-06 10:47:00,788 - Epoch: [55][ 820/ 1200] Overall Loss 0.270905 Objective Loss 0.270905 LR 0.001000 Time 0.020104 -2022-12-06 10:47:00,981 - Epoch: [55][ 830/ 1200] Overall Loss 0.270990 Objective Loss 0.270990 LR 0.001000 Time 0.020093 -2022-12-06 10:47:01,173 - Epoch: [55][ 840/ 1200] Overall Loss 0.271310 Objective Loss 0.271310 LR 0.001000 Time 0.020082 -2022-12-06 10:47:01,366 - Epoch: [55][ 850/ 1200] Overall Loss 0.271687 Objective Loss 0.271687 LR 0.001000 Time 0.020072 -2022-12-06 10:47:01,558 - Epoch: [55][ 860/ 1200] Overall Loss 0.271665 Objective Loss 0.271665 LR 0.001000 Time 0.020062 -2022-12-06 10:47:01,751 - Epoch: [55][ 870/ 1200] Overall Loss 0.271977 Objective Loss 0.271977 LR 0.001000 Time 0.020052 -2022-12-06 10:47:01,943 - Epoch: [55][ 880/ 1200] Overall Loss 0.271585 Objective Loss 0.271585 LR 0.001000 Time 0.020042 -2022-12-06 10:47:02,135 - Epoch: [55][ 890/ 1200] Overall Loss 0.271596 Objective Loss 0.271596 LR 0.001000 Time 0.020032 -2022-12-06 10:47:02,328 - Epoch: [55][ 900/ 1200] Overall Loss 0.271551 Objective Loss 0.271551 LR 0.001000 Time 0.020023 -2022-12-06 10:47:02,521 - Epoch: [55][ 910/ 1200] Overall Loss 0.271469 Objective Loss 0.271469 LR 0.001000 Time 0.020015 -2022-12-06 10:47:02,714 - Epoch: [55][ 920/ 1200] Overall Loss 0.271290 Objective Loss 0.271290 LR 0.001000 Time 0.020006 -2022-12-06 10:47:02,906 - Epoch: [55][ 930/ 1200] Overall Loss 0.271159 Objective Loss 0.271159 LR 0.001000 Time 0.019996 -2022-12-06 10:47:03,098 - Epoch: [55][ 940/ 1200] Overall Loss 0.271512 Objective Loss 0.271512 LR 0.001000 Time 0.019987 -2022-12-06 10:47:03,290 - Epoch: [55][ 950/ 1200] Overall Loss 0.271513 Objective Loss 0.271513 LR 0.001000 Time 0.019979 -2022-12-06 10:47:03,482 - Epoch: [55][ 960/ 1200] Overall Loss 0.271198 Objective Loss 0.271198 LR 0.001000 Time 0.019970 -2022-12-06 10:47:03,675 - Epoch: [55][ 970/ 1200] Overall Loss 0.271257 Objective Loss 0.271257 LR 0.001000 Time 0.019963 -2022-12-06 10:47:03,870 - Epoch: [55][ 980/ 1200] Overall Loss 0.271417 Objective Loss 0.271417 LR 0.001000 Time 0.019957 -2022-12-06 10:47:04,069 - Epoch: [55][ 990/ 1200] Overall Loss 0.271661 Objective Loss 0.271661 LR 0.001000 Time 0.019957 -2022-12-06 10:47:04,265 - Epoch: [55][ 1000/ 1200] Overall Loss 0.271479 Objective Loss 0.271479 LR 0.001000 Time 0.019952 -2022-12-06 10:47:04,463 - Epoch: [55][ 1010/ 1200] Overall Loss 0.271376 Objective Loss 0.271376 LR 0.001000 Time 0.019951 -2022-12-06 10:47:04,660 - Epoch: [55][ 1020/ 1200] Overall Loss 0.271660 Objective Loss 0.271660 LR 0.001000 Time 0.019947 -2022-12-06 10:47:04,859 - Epoch: [55][ 1030/ 1200] Overall Loss 0.271854 Objective Loss 0.271854 LR 0.001000 Time 0.019946 -2022-12-06 10:47:05,055 - Epoch: [55][ 1040/ 1200] Overall Loss 0.271962 Objective Loss 0.271962 LR 0.001000 Time 0.019943 -2022-12-06 10:47:05,255 - Epoch: [55][ 1050/ 1200] Overall Loss 0.272223 Objective Loss 0.272223 LR 0.001000 Time 0.019942 -2022-12-06 10:47:05,451 - Epoch: [55][ 1060/ 1200] Overall Loss 0.272550 Objective Loss 0.272550 LR 0.001000 Time 0.019939 -2022-12-06 10:47:05,651 - Epoch: [55][ 1070/ 1200] Overall Loss 0.272474 Objective Loss 0.272474 LR 0.001000 Time 0.019939 -2022-12-06 10:47:05,847 - Epoch: [55][ 1080/ 1200] Overall Loss 0.272615 Objective Loss 0.272615 LR 0.001000 Time 0.019935 -2022-12-06 10:47:06,046 - Epoch: [55][ 1090/ 1200] Overall Loss 0.272665 Objective Loss 0.272665 LR 0.001000 Time 0.019934 -2022-12-06 10:47:06,242 - Epoch: [55][ 1100/ 1200] Overall Loss 0.272437 Objective Loss 0.272437 LR 0.001000 Time 0.019931 -2022-12-06 10:47:06,441 - Epoch: [55][ 1110/ 1200] Overall Loss 0.272654 Objective Loss 0.272654 LR 0.001000 Time 0.019930 -2022-12-06 10:47:06,637 - Epoch: [55][ 1120/ 1200] Overall Loss 0.272618 Objective Loss 0.272618 LR 0.001000 Time 0.019926 -2022-12-06 10:47:06,836 - Epoch: [55][ 1130/ 1200] Overall Loss 0.272705 Objective Loss 0.272705 LR 0.001000 Time 0.019926 -2022-12-06 10:47:07,032 - Epoch: [55][ 1140/ 1200] Overall Loss 0.272665 Objective Loss 0.272665 LR 0.001000 Time 0.019923 -2022-12-06 10:47:07,231 - Epoch: [55][ 1150/ 1200] Overall Loss 0.272616 Objective Loss 0.272616 LR 0.001000 Time 0.019922 -2022-12-06 10:47:07,427 - Epoch: [55][ 1160/ 1200] Overall Loss 0.272744 Objective Loss 0.272744 LR 0.001000 Time 0.019919 -2022-12-06 10:47:07,626 - Epoch: [55][ 1170/ 1200] Overall Loss 0.273215 Objective Loss 0.273215 LR 0.001000 Time 0.019918 -2022-12-06 10:47:07,822 - Epoch: [55][ 1180/ 1200] Overall Loss 0.273330 Objective Loss 0.273330 LR 0.001000 Time 0.019914 -2022-12-06 10:47:08,021 - Epoch: [55][ 1190/ 1200] Overall Loss 0.273600 Objective Loss 0.273600 LR 0.001000 Time 0.019914 -2022-12-06 10:47:08,246 - Epoch: [55][ 1200/ 1200] Overall Loss 0.273749 Objective Loss 0.273749 Top1 85.564854 Top5 98.117155 LR 0.001000 Time 0.019935 -2022-12-06 10:47:08,335 - --- validate (epoch=55)----------- -2022-12-06 10:47:08,335 - 34129 samples (256 per mini-batch) -2022-12-06 10:47:08,781 - Epoch: [55][ 10/ 134] Loss 0.290619 Top1 84.726562 Top5 98.320312 -2022-12-06 10:47:08,908 - Epoch: [55][ 20/ 134] Loss 0.303051 Top1 84.863281 Top5 98.085938 -2022-12-06 10:47:09,049 - Epoch: [55][ 30/ 134] Loss 0.298865 Top1 84.973958 Top5 98.125000 -2022-12-06 10:47:09,190 - Epoch: [55][ 40/ 134] Loss 0.296913 Top1 84.824219 Top5 98.173828 -2022-12-06 10:47:09,330 - Epoch: [55][ 50/ 134] Loss 0.287675 Top1 85.343750 Top5 98.187500 -2022-12-06 10:47:09,467 - Epoch: [55][ 60/ 134] Loss 0.290423 Top1 85.273438 Top5 98.138021 -2022-12-06 10:47:09,597 - Epoch: [55][ 70/ 134] Loss 0.297062 Top1 85.172991 Top5 98.024554 -2022-12-06 10:47:09,727 - Epoch: [55][ 80/ 134] Loss 0.295396 Top1 85.068359 Top5 97.973633 -2022-12-06 10:47:09,857 - Epoch: [55][ 90/ 134] Loss 0.293208 Top1 85.095486 Top5 98.003472 -2022-12-06 10:47:09,988 - Epoch: [55][ 100/ 134] Loss 0.291892 Top1 85.105469 Top5 98.007812 -2022-12-06 10:47:10,120 - Epoch: [55][ 110/ 134] Loss 0.297106 Top1 84.957386 Top5 98.000710 -2022-12-06 10:47:10,252 - Epoch: [55][ 120/ 134] Loss 0.296023 Top1 84.918620 Top5 97.988281 -2022-12-06 10:47:10,383 - Epoch: [55][ 130/ 134] Loss 0.296319 Top1 84.891827 Top5 97.971755 -2022-12-06 10:47:10,420 - Epoch: [55][ 134/ 134] Loss 0.296452 Top1 84.863313 Top5 97.989979 -2022-12-06 10:47:10,507 - ==> Top1: 84.863 Top5: 97.990 Loss: 0.296 - -2022-12-06 10:47:10,508 - ==> Confusion: -[[ 921 4 1 1 5 4 1 2 4 32 0 3 5 1 3 2 1 1 2 1 2] - [ 2 958 1 1 4 14 2 6 2 0 7 2 0 0 1 0 4 3 15 1 4] - [ 6 7 982 19 2 3 24 11 2 1 3 3 0 1 2 4 1 3 14 3 12] - [ 1 2 14 928 1 3 0 1 0 0 10 1 4 3 19 1 0 5 20 0 7] - [ 18 10 3 0 935 3 0 4 2 3 1 4 2 2 13 5 6 1 0 1 7] - [ 4 33 0 2 8 956 4 22 5 4 0 8 4 7 1 0 3 0 0 5 3] - [ 0 2 12 3 0 2 1057 6 0 0 7 1 1 4 0 10 0 0 2 7 4] - [ 0 20 9 0 2 28 2 922 1 1 4 7 1 1 1 0 0 0 38 13 4] - [ 7 3 0 0 0 1 0 1 981 36 7 3 1 3 11 0 3 0 5 1 1] - [ 108 2 0 1 2 1 0 3 30 838 2 0 0 7 2 0 0 0 0 1 4] - [ 0 1 2 6 1 1 2 3 15 1 950 2 3 11 5 0 0 0 8 1 7] - [ 4 1 3 0 0 10 1 1 2 0 1 952 29 5 1 3 10 5 1 17 5] - [ 0 1 2 3 1 3 1 1 1 0 2 44 869 3 0 8 2 11 2 5 10] - [ 0 1 1 0 0 12 0 2 16 23 6 6 5 928 0 3 3 0 2 1 14] - [ 12 4 2 12 2 3 0 1 21 2 1 1 3 3 1042 1 2 1 10 0 7] - [ 0 2 4 1 0 2 2 0 0 0 1 8 9 4 0 975 12 11 1 4 7] - [ 2 4 2 1 0 0 0 0 3 1 2 4 1 1 0 8 1030 2 1 6 4] - [ 2 1 2 4 0 1 2 2 3 4 0 14 27 0 3 15 4 941 3 4 4] - [ 2 6 1 10 0 0 1 19 1 0 5 1 1 2 2 1 0 1 949 2 4] - [ 2 3 1 0 0 4 4 7 2 0 1 10 9 5 1 2 6 3 1 1010 9] - [ 136 324 146 143 100 179 73 139 115 88 143 101 332 284 157 121 213 73 249 272 9838]] - -2022-12-06 10:47:11,173 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:47:11,173 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:47:11,179 - - -2022-12-06 10:47:11,179 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:47:12,131 - Epoch: [56][ 10/ 1200] Overall Loss 0.247994 Objective Loss 0.247994 LR 0.001000 Time 0.095123 -2022-12-06 10:47:12,343 - Epoch: [56][ 20/ 1200] Overall Loss 0.256181 Objective Loss 0.256181 LR 0.001000 Time 0.058131 -2022-12-06 10:47:12,545 - Epoch: [56][ 30/ 1200] Overall Loss 0.253330 Objective Loss 0.253330 LR 0.001000 Time 0.045473 -2022-12-06 10:47:12,751 - Epoch: [56][ 40/ 1200] Overall Loss 0.255871 Objective Loss 0.255871 LR 0.001000 Time 0.039243 -2022-12-06 10:47:12,954 - Epoch: [56][ 50/ 1200] Overall Loss 0.256600 Objective Loss 0.256600 LR 0.001000 Time 0.035424 -2022-12-06 10:47:13,160 - Epoch: [56][ 60/ 1200] Overall Loss 0.257696 Objective Loss 0.257696 LR 0.001000 Time 0.032952 -2022-12-06 10:47:13,362 - Epoch: [56][ 70/ 1200] Overall Loss 0.262318 Objective Loss 0.262318 LR 0.001000 Time 0.031126 -2022-12-06 10:47:13,568 - Epoch: [56][ 80/ 1200] Overall Loss 0.268161 Objective Loss 0.268161 LR 0.001000 Time 0.029797 -2022-12-06 10:47:13,770 - Epoch: [56][ 90/ 1200] Overall Loss 0.269996 Objective Loss 0.269996 LR 0.001000 Time 0.028726 -2022-12-06 10:47:13,977 - Epoch: [56][ 100/ 1200] Overall Loss 0.269187 Objective Loss 0.269187 LR 0.001000 Time 0.027913 -2022-12-06 10:47:14,179 - Epoch: [56][ 110/ 1200] Overall Loss 0.268863 Objective Loss 0.268863 LR 0.001000 Time 0.027206 -2022-12-06 10:47:14,384 - Epoch: [56][ 120/ 1200] Overall Loss 0.267145 Objective Loss 0.267145 LR 0.001000 Time 0.026648 -2022-12-06 10:47:14,587 - Epoch: [56][ 130/ 1200] Overall Loss 0.266535 Objective Loss 0.266535 LR 0.001000 Time 0.026150 -2022-12-06 10:47:14,791 - Epoch: [56][ 140/ 1200] Overall Loss 0.268666 Objective Loss 0.268666 LR 0.001000 Time 0.025742 -2022-12-06 10:47:14,994 - Epoch: [56][ 150/ 1200] Overall Loss 0.268545 Objective Loss 0.268545 LR 0.001000 Time 0.025369 -2022-12-06 10:47:15,200 - Epoch: [56][ 160/ 1200] Overall Loss 0.266212 Objective Loss 0.266212 LR 0.001000 Time 0.025069 -2022-12-06 10:47:15,402 - Epoch: [56][ 170/ 1200] Overall Loss 0.267446 Objective Loss 0.267446 LR 0.001000 Time 0.024781 -2022-12-06 10:47:15,608 - Epoch: [56][ 180/ 1200] Overall Loss 0.268623 Objective Loss 0.268623 LR 0.001000 Time 0.024543 -2022-12-06 10:47:15,810 - Epoch: [56][ 190/ 1200] Overall Loss 0.269019 Objective Loss 0.269019 LR 0.001000 Time 0.024310 -2022-12-06 10:47:16,015 - Epoch: [56][ 200/ 1200] Overall Loss 0.268899 Objective Loss 0.268899 LR 0.001000 Time 0.024120 -2022-12-06 10:47:16,218 - Epoch: [56][ 210/ 1200] Overall Loss 0.268082 Objective Loss 0.268082 LR 0.001000 Time 0.023932 -2022-12-06 10:47:16,423 - Epoch: [56][ 220/ 1200] Overall Loss 0.268449 Objective Loss 0.268449 LR 0.001000 Time 0.023777 -2022-12-06 10:47:16,626 - Epoch: [56][ 230/ 1200] Overall Loss 0.269020 Objective Loss 0.269020 LR 0.001000 Time 0.023622 -2022-12-06 10:47:16,833 - Epoch: [56][ 240/ 1200] Overall Loss 0.267504 Objective Loss 0.267504 LR 0.001000 Time 0.023496 -2022-12-06 10:47:17,035 - Epoch: [56][ 250/ 1200] Overall Loss 0.267420 Objective Loss 0.267420 LR 0.001000 Time 0.023363 -2022-12-06 10:47:17,241 - Epoch: [56][ 260/ 1200] Overall Loss 0.268453 Objective Loss 0.268453 LR 0.001000 Time 0.023255 -2022-12-06 10:47:17,443 - Epoch: [56][ 270/ 1200] Overall Loss 0.268202 Objective Loss 0.268202 LR 0.001000 Time 0.023140 -2022-12-06 10:47:17,650 - Epoch: [56][ 280/ 1200] Overall Loss 0.267101 Objective Loss 0.267101 LR 0.001000 Time 0.023049 -2022-12-06 10:47:17,852 - Epoch: [56][ 290/ 1200] Overall Loss 0.267463 Objective Loss 0.267463 LR 0.001000 Time 0.022950 -2022-12-06 10:47:18,059 - Epoch: [56][ 300/ 1200] Overall Loss 0.267565 Objective Loss 0.267565 LR 0.001000 Time 0.022872 -2022-12-06 10:47:18,261 - Epoch: [56][ 310/ 1200] Overall Loss 0.267673 Objective Loss 0.267673 LR 0.001000 Time 0.022784 -2022-12-06 10:47:18,468 - Epoch: [56][ 320/ 1200] Overall Loss 0.267898 Objective Loss 0.267898 LR 0.001000 Time 0.022716 -2022-12-06 10:47:18,670 - Epoch: [56][ 330/ 1200] Overall Loss 0.267975 Objective Loss 0.267975 LR 0.001000 Time 0.022638 -2022-12-06 10:47:18,876 - Epoch: [56][ 340/ 1200] Overall Loss 0.268023 Objective Loss 0.268023 LR 0.001000 Time 0.022576 -2022-12-06 10:47:19,077 - Epoch: [56][ 350/ 1200] Overall Loss 0.268046 Objective Loss 0.268046 LR 0.001000 Time 0.022506 -2022-12-06 10:47:19,284 - Epoch: [56][ 360/ 1200] Overall Loss 0.268133 Objective Loss 0.268133 LR 0.001000 Time 0.022452 -2022-12-06 10:47:19,487 - Epoch: [56][ 370/ 1200] Overall Loss 0.267789 Objective Loss 0.267789 LR 0.001000 Time 0.022392 -2022-12-06 10:47:19,693 - Epoch: [56][ 380/ 1200] Overall Loss 0.267419 Objective Loss 0.267419 LR 0.001000 Time 0.022345 -2022-12-06 10:47:19,896 - Epoch: [56][ 390/ 1200] Overall Loss 0.267471 Objective Loss 0.267471 LR 0.001000 Time 0.022289 -2022-12-06 10:47:20,101 - Epoch: [56][ 400/ 1200] Overall Loss 0.267791 Objective Loss 0.267791 LR 0.001000 Time 0.022244 -2022-12-06 10:47:20,303 - Epoch: [56][ 410/ 1200] Overall Loss 0.268278 Objective Loss 0.268278 LR 0.001000 Time 0.022193 -2022-12-06 10:47:20,510 - Epoch: [56][ 420/ 1200] Overall Loss 0.268725 Objective Loss 0.268725 LR 0.001000 Time 0.022154 -2022-12-06 10:47:20,711 - Epoch: [56][ 430/ 1200] Overall Loss 0.268920 Objective Loss 0.268920 LR 0.001000 Time 0.022107 -2022-12-06 10:47:20,917 - Epoch: [56][ 440/ 1200] Overall Loss 0.268581 Objective Loss 0.268581 LR 0.001000 Time 0.022071 -2022-12-06 10:47:21,119 - Epoch: [56][ 450/ 1200] Overall Loss 0.268292 Objective Loss 0.268292 LR 0.001000 Time 0.022027 -2022-12-06 10:47:21,324 - Epoch: [56][ 460/ 1200] Overall Loss 0.268513 Objective Loss 0.268513 LR 0.001000 Time 0.021994 -2022-12-06 10:47:21,527 - Epoch: [56][ 470/ 1200] Overall Loss 0.269353 Objective Loss 0.269353 LR 0.001000 Time 0.021955 -2022-12-06 10:47:21,733 - Epoch: [56][ 480/ 1200] Overall Loss 0.268607 Objective Loss 0.268607 LR 0.001000 Time 0.021925 -2022-12-06 10:47:21,935 - Epoch: [56][ 490/ 1200] Overall Loss 0.268558 Objective Loss 0.268558 LR 0.001000 Time 0.021890 -2022-12-06 10:47:22,142 - Epoch: [56][ 500/ 1200] Overall Loss 0.269089 Objective Loss 0.269089 LR 0.001000 Time 0.021864 -2022-12-06 10:47:22,344 - Epoch: [56][ 510/ 1200] Overall Loss 0.269629 Objective Loss 0.269629 LR 0.001000 Time 0.021831 -2022-12-06 10:47:22,550 - Epoch: [56][ 520/ 1200] Overall Loss 0.269740 Objective Loss 0.269740 LR 0.001000 Time 0.021807 -2022-12-06 10:47:22,753 - Epoch: [56][ 530/ 1200] Overall Loss 0.269877 Objective Loss 0.269877 LR 0.001000 Time 0.021776 -2022-12-06 10:47:22,960 - Epoch: [56][ 540/ 1200] Overall Loss 0.270015 Objective Loss 0.270015 LR 0.001000 Time 0.021755 -2022-12-06 10:47:23,162 - Epoch: [56][ 550/ 1200] Overall Loss 0.269664 Objective Loss 0.269664 LR 0.001000 Time 0.021726 -2022-12-06 10:47:23,368 - Epoch: [56][ 560/ 1200] Overall Loss 0.269571 Objective Loss 0.269571 LR 0.001000 Time 0.021705 -2022-12-06 10:47:23,571 - Epoch: [56][ 570/ 1200] Overall Loss 0.270131 Objective Loss 0.270131 LR 0.001000 Time 0.021680 -2022-12-06 10:47:23,777 - Epoch: [56][ 580/ 1200] Overall Loss 0.270008 Objective Loss 0.270008 LR 0.001000 Time 0.021659 -2022-12-06 10:47:23,979 - Epoch: [56][ 590/ 1200] Overall Loss 0.270061 Objective Loss 0.270061 LR 0.001000 Time 0.021634 -2022-12-06 10:47:24,185 - Epoch: [56][ 600/ 1200] Overall Loss 0.270124 Objective Loss 0.270124 LR 0.001000 Time 0.021615 -2022-12-06 10:47:24,387 - Epoch: [56][ 610/ 1200] Overall Loss 0.270335 Objective Loss 0.270335 LR 0.001000 Time 0.021591 -2022-12-06 10:47:24,592 - Epoch: [56][ 620/ 1200] Overall Loss 0.270333 Objective Loss 0.270333 LR 0.001000 Time 0.021573 -2022-12-06 10:47:24,793 - Epoch: [56][ 630/ 1200] Overall Loss 0.270332 Objective Loss 0.270332 LR 0.001000 Time 0.021549 -2022-12-06 10:47:25,000 - Epoch: [56][ 640/ 1200] Overall Loss 0.270315 Objective Loss 0.270315 LR 0.001000 Time 0.021533 -2022-12-06 10:47:25,202 - Epoch: [56][ 650/ 1200] Overall Loss 0.270351 Objective Loss 0.270351 LR 0.001000 Time 0.021513 -2022-12-06 10:47:25,409 - Epoch: [56][ 660/ 1200] Overall Loss 0.269800 Objective Loss 0.269800 LR 0.001000 Time 0.021499 -2022-12-06 10:47:25,611 - Epoch: [56][ 670/ 1200] Overall Loss 0.270077 Objective Loss 0.270077 LR 0.001000 Time 0.021479 -2022-12-06 10:47:25,817 - Epoch: [56][ 680/ 1200] Overall Loss 0.270313 Objective Loss 0.270313 LR 0.001000 Time 0.021466 -2022-12-06 10:47:26,019 - Epoch: [56][ 690/ 1200] Overall Loss 0.270412 Objective Loss 0.270412 LR 0.001000 Time 0.021445 -2022-12-06 10:47:26,225 - Epoch: [56][ 700/ 1200] Overall Loss 0.270808 Objective Loss 0.270808 LR 0.001000 Time 0.021433 -2022-12-06 10:47:26,426 - Epoch: [56][ 710/ 1200] Overall Loss 0.270732 Objective Loss 0.270732 LR 0.001000 Time 0.021413 -2022-12-06 10:47:26,632 - Epoch: [56][ 720/ 1200] Overall Loss 0.270596 Objective Loss 0.270596 LR 0.001000 Time 0.021402 -2022-12-06 10:47:26,834 - Epoch: [56][ 730/ 1200] Overall Loss 0.270643 Objective Loss 0.270643 LR 0.001000 Time 0.021384 -2022-12-06 10:47:27,041 - Epoch: [56][ 740/ 1200] Overall Loss 0.270715 Objective Loss 0.270715 LR 0.001000 Time 0.021374 -2022-12-06 10:47:27,243 - Epoch: [56][ 750/ 1200] Overall Loss 0.270450 Objective Loss 0.270450 LR 0.001000 Time 0.021358 -2022-12-06 10:47:27,451 - Epoch: [56][ 760/ 1200] Overall Loss 0.270429 Objective Loss 0.270429 LR 0.001000 Time 0.021349 -2022-12-06 10:47:27,652 - Epoch: [56][ 770/ 1200] Overall Loss 0.270392 Objective Loss 0.270392 LR 0.001000 Time 0.021332 -2022-12-06 10:47:27,859 - Epoch: [56][ 780/ 1200] Overall Loss 0.270647 Objective Loss 0.270647 LR 0.001000 Time 0.021323 -2022-12-06 10:47:28,060 - Epoch: [56][ 790/ 1200] Overall Loss 0.270542 Objective Loss 0.270542 LR 0.001000 Time 0.021308 -2022-12-06 10:47:28,267 - Epoch: [56][ 800/ 1200] Overall Loss 0.270847 Objective Loss 0.270847 LR 0.001000 Time 0.021298 -2022-12-06 10:47:28,469 - Epoch: [56][ 810/ 1200] Overall Loss 0.271003 Objective Loss 0.271003 LR 0.001000 Time 0.021284 -2022-12-06 10:47:28,674 - Epoch: [56][ 820/ 1200] Overall Loss 0.270921 Objective Loss 0.270921 LR 0.001000 Time 0.021274 -2022-12-06 10:47:28,875 - Epoch: [56][ 830/ 1200] Overall Loss 0.270776 Objective Loss 0.270776 LR 0.001000 Time 0.021259 -2022-12-06 10:47:29,080 - Epoch: [56][ 840/ 1200] Overall Loss 0.270633 Objective Loss 0.270633 LR 0.001000 Time 0.021250 -2022-12-06 10:47:29,281 - Epoch: [56][ 850/ 1200] Overall Loss 0.271090 Objective Loss 0.271090 LR 0.001000 Time 0.021235 -2022-12-06 10:47:29,487 - Epoch: [56][ 860/ 1200] Overall Loss 0.271024 Objective Loss 0.271024 LR 0.001000 Time 0.021227 -2022-12-06 10:47:29,688 - Epoch: [56][ 870/ 1200] Overall Loss 0.271316 Objective Loss 0.271316 LR 0.001000 Time 0.021213 -2022-12-06 10:47:29,893 - Epoch: [56][ 880/ 1200] Overall Loss 0.271247 Objective Loss 0.271247 LR 0.001000 Time 0.021205 -2022-12-06 10:47:30,095 - Epoch: [56][ 890/ 1200] Overall Loss 0.271221 Objective Loss 0.271221 LR 0.001000 Time 0.021193 -2022-12-06 10:47:30,301 - Epoch: [56][ 900/ 1200] Overall Loss 0.271493 Objective Loss 0.271493 LR 0.001000 Time 0.021186 -2022-12-06 10:47:30,502 - Epoch: [56][ 910/ 1200] Overall Loss 0.271576 Objective Loss 0.271576 LR 0.001000 Time 0.021173 -2022-12-06 10:47:30,708 - Epoch: [56][ 920/ 1200] Overall Loss 0.271668 Objective Loss 0.271668 LR 0.001000 Time 0.021165 -2022-12-06 10:47:30,909 - Epoch: [56][ 930/ 1200] Overall Loss 0.271413 Objective Loss 0.271413 LR 0.001000 Time 0.021153 -2022-12-06 10:47:31,114 - Epoch: [56][ 940/ 1200] Overall Loss 0.271345 Objective Loss 0.271345 LR 0.001000 Time 0.021146 -2022-12-06 10:47:31,315 - Epoch: [56][ 950/ 1200] Overall Loss 0.271074 Objective Loss 0.271074 LR 0.001000 Time 0.021135 -2022-12-06 10:47:31,521 - Epoch: [56][ 960/ 1200] Overall Loss 0.271142 Objective Loss 0.271142 LR 0.001000 Time 0.021128 -2022-12-06 10:47:31,723 - Epoch: [56][ 970/ 1200] Overall Loss 0.271227 Objective Loss 0.271227 LR 0.001000 Time 0.021117 -2022-12-06 10:47:31,928 - Epoch: [56][ 980/ 1200] Overall Loss 0.271206 Objective Loss 0.271206 LR 0.001000 Time 0.021111 -2022-12-06 10:47:32,129 - Epoch: [56][ 990/ 1200] Overall Loss 0.271110 Objective Loss 0.271110 LR 0.001000 Time 0.021100 -2022-12-06 10:47:32,335 - Epoch: [56][ 1000/ 1200] Overall Loss 0.271179 Objective Loss 0.271179 LR 0.001000 Time 0.021094 -2022-12-06 10:47:32,536 - Epoch: [56][ 1010/ 1200] Overall Loss 0.271557 Objective Loss 0.271557 LR 0.001000 Time 0.021084 -2022-12-06 10:47:32,741 - Epoch: [56][ 1020/ 1200] Overall Loss 0.271647 Objective Loss 0.271647 LR 0.001000 Time 0.021077 -2022-12-06 10:47:32,942 - Epoch: [56][ 1030/ 1200] Overall Loss 0.271831 Objective Loss 0.271831 LR 0.001000 Time 0.021068 -2022-12-06 10:47:33,148 - Epoch: [56][ 1040/ 1200] Overall Loss 0.272119 Objective Loss 0.272119 LR 0.001000 Time 0.021062 -2022-12-06 10:47:33,349 - Epoch: [56][ 1050/ 1200] Overall Loss 0.272143 Objective Loss 0.272143 LR 0.001000 Time 0.021053 -2022-12-06 10:47:33,554 - Epoch: [56][ 1060/ 1200] Overall Loss 0.272379 Objective Loss 0.272379 LR 0.001000 Time 0.021047 -2022-12-06 10:47:33,755 - Epoch: [56][ 1070/ 1200] Overall Loss 0.272472 Objective Loss 0.272472 LR 0.001000 Time 0.021038 -2022-12-06 10:47:33,960 - Epoch: [56][ 1080/ 1200] Overall Loss 0.272319 Objective Loss 0.272319 LR 0.001000 Time 0.021032 -2022-12-06 10:47:34,162 - Epoch: [56][ 1090/ 1200] Overall Loss 0.272225 Objective Loss 0.272225 LR 0.001000 Time 0.021024 -2022-12-06 10:47:34,367 - Epoch: [56][ 1100/ 1200] Overall Loss 0.271905 Objective Loss 0.271905 LR 0.001000 Time 0.021019 -2022-12-06 10:47:34,568 - Epoch: [56][ 1110/ 1200] Overall Loss 0.271923 Objective Loss 0.271923 LR 0.001000 Time 0.021010 -2022-12-06 10:47:34,774 - Epoch: [56][ 1120/ 1200] Overall Loss 0.271636 Objective Loss 0.271636 LR 0.001000 Time 0.021005 -2022-12-06 10:47:34,975 - Epoch: [56][ 1130/ 1200] Overall Loss 0.271578 Objective Loss 0.271578 LR 0.001000 Time 0.020997 -2022-12-06 10:47:35,181 - Epoch: [56][ 1140/ 1200] Overall Loss 0.271564 Objective Loss 0.271564 LR 0.001000 Time 0.020992 -2022-12-06 10:47:35,382 - Epoch: [56][ 1150/ 1200] Overall Loss 0.271709 Objective Loss 0.271709 LR 0.001000 Time 0.020985 -2022-12-06 10:47:35,588 - Epoch: [56][ 1160/ 1200] Overall Loss 0.271631 Objective Loss 0.271631 LR 0.001000 Time 0.020980 -2022-12-06 10:47:35,789 - Epoch: [56][ 1170/ 1200] Overall Loss 0.271843 Objective Loss 0.271843 LR 0.001000 Time 0.020972 -2022-12-06 10:47:35,995 - Epoch: [56][ 1180/ 1200] Overall Loss 0.271879 Objective Loss 0.271879 LR 0.001000 Time 0.020969 -2022-12-06 10:47:36,196 - Epoch: [56][ 1190/ 1200] Overall Loss 0.272044 Objective Loss 0.272044 LR 0.001000 Time 0.020961 -2022-12-06 10:47:36,432 - Epoch: [56][ 1200/ 1200] Overall Loss 0.272108 Objective Loss 0.272108 Top1 83.682008 Top5 98.326360 LR 0.001000 Time 0.020983 -2022-12-06 10:47:36,525 - --- validate (epoch=56)----------- -2022-12-06 10:47:36,526 - 34129 samples (256 per mini-batch) -2022-12-06 10:47:36,975 - Epoch: [56][ 10/ 134] Loss 0.283726 Top1 85.234375 Top5 98.281250 -2022-12-06 10:47:37,107 - Epoch: [56][ 20/ 134] Loss 0.288743 Top1 84.726562 Top5 98.066406 -2022-12-06 10:47:37,238 - Epoch: [56][ 30/ 134] Loss 0.288328 Top1 84.908854 Top5 98.085938 -2022-12-06 10:47:37,371 - Epoch: [56][ 40/ 134] Loss 0.292405 Top1 84.658203 Top5 97.939453 -2022-12-06 10:47:37,505 - Epoch: [56][ 50/ 134] Loss 0.285814 Top1 84.789062 Top5 97.906250 -2022-12-06 10:47:37,636 - Epoch: [56][ 60/ 134] Loss 0.293649 Top1 84.687500 Top5 97.897135 -2022-12-06 10:47:37,770 - Epoch: [56][ 70/ 134] Loss 0.293470 Top1 84.654018 Top5 97.924107 -2022-12-06 10:47:37,903 - Epoch: [56][ 80/ 134] Loss 0.295583 Top1 84.628906 Top5 97.900391 -2022-12-06 10:47:38,037 - Epoch: [56][ 90/ 134] Loss 0.296281 Top1 84.670139 Top5 97.890625 -2022-12-06 10:47:38,170 - Epoch: [56][ 100/ 134] Loss 0.297538 Top1 84.628906 Top5 97.886719 -2022-12-06 10:47:38,304 - Epoch: [56][ 110/ 134] Loss 0.297177 Top1 84.687500 Top5 97.858665 -2022-12-06 10:47:38,438 - Epoch: [56][ 120/ 134] Loss 0.297196 Top1 84.690755 Top5 97.874349 -2022-12-06 10:47:38,569 - Epoch: [56][ 130/ 134] Loss 0.299068 Top1 84.699519 Top5 97.860577 -2022-12-06 10:47:38,608 - Epoch: [56][ 134/ 134] Loss 0.297718 Top1 84.734390 Top5 97.872777 -2022-12-06 10:47:38,695 - ==> Top1: 84.734 Top5: 97.873 Loss: 0.298 - -2022-12-06 10:47:38,696 - ==> Confusion: -[[ 909 4 1 3 4 6 0 1 2 48 0 1 1 1 3 3 5 1 0 1 2] - [ 1 925 3 2 12 27 2 2 2 0 5 6 2 4 0 1 8 1 13 1 10] - [ 4 2 982 17 7 1 33 5 1 0 4 7 1 1 3 5 2 2 6 4 16] - [ 3 4 21 929 1 5 0 0 2 0 11 0 2 3 16 2 1 5 9 2 4] - [ 15 5 1 1 954 3 0 0 2 3 1 1 1 2 9 4 8 3 0 1 6] - [ 7 15 2 3 7 959 4 20 6 3 3 9 5 12 3 0 1 2 0 1 7] - [ 0 5 10 3 0 3 1065 1 0 0 5 2 2 0 1 9 0 1 2 6 3] - [ 1 13 10 1 2 34 11 907 1 3 5 5 2 2 1 0 1 1 34 14 6] - [ 7 4 1 2 0 0 0 0 959 48 12 2 2 12 10 0 1 0 0 2 2] - [ 90 1 3 2 9 3 1 2 22 844 0 1 0 15 3 0 0 0 0 1 4] - [ 1 1 4 9 2 0 1 2 11 2 956 3 3 10 4 1 2 0 3 0 4] - [ 4 1 2 0 1 10 3 4 4 0 0 973 17 8 1 4 3 7 1 6 2] - [ 0 2 0 6 1 2 2 1 2 0 0 32 879 4 2 8 2 16 0 2 8] - [ 0 1 0 0 3 13 0 2 11 14 11 6 3 935 4 2 1 1 1 5 10] - [ 6 5 1 14 5 1 0 1 12 5 3 1 4 1 1055 0 2 0 5 1 8] - [ 1 1 2 1 1 2 5 0 0 0 0 6 5 7 0 986 10 9 0 3 4] - [ 2 3 0 2 6 0 1 1 2 0 0 4 3 3 1 6 1024 3 0 7 4] - [ 4 1 1 4 0 1 1 0 0 1 1 8 18 3 2 11 1 976 0 2 1] - [ 2 7 3 18 1 5 0 21 1 0 11 5 6 4 10 0 2 0 904 2 6] - [ 2 4 1 0 2 11 9 6 0 0 2 15 8 6 1 2 6 5 1 991 8] - [ 144 214 162 128 147 188 73 113 96 92 193 101 370 324 137 141 268 107 172 252 9804]] - -2022-12-06 10:47:39,262 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:47:39,262 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:47:39,268 - - -2022-12-06 10:47:39,268 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:47:40,317 - Epoch: [57][ 10/ 1200] Overall Loss 0.262240 Objective Loss 0.262240 LR 0.001000 Time 0.104821 -2022-12-06 10:47:40,527 - Epoch: [57][ 20/ 1200] Overall Loss 0.271661 Objective Loss 0.271661 LR 0.001000 Time 0.062842 -2022-12-06 10:47:40,731 - Epoch: [57][ 30/ 1200] Overall Loss 0.269009 Objective Loss 0.269009 LR 0.001000 Time 0.048700 -2022-12-06 10:47:40,932 - Epoch: [57][ 40/ 1200] Overall Loss 0.276343 Objective Loss 0.276343 LR 0.001000 Time 0.041523 -2022-12-06 10:47:41,136 - Epoch: [57][ 50/ 1200] Overall Loss 0.276322 Objective Loss 0.276322 LR 0.001000 Time 0.037285 -2022-12-06 10:47:41,335 - Epoch: [57][ 60/ 1200] Overall Loss 0.278345 Objective Loss 0.278345 LR 0.001000 Time 0.034391 -2022-12-06 10:47:41,540 - Epoch: [57][ 70/ 1200] Overall Loss 0.276778 Objective Loss 0.276778 LR 0.001000 Time 0.032399 -2022-12-06 10:47:41,740 - Epoch: [57][ 80/ 1200] Overall Loss 0.272524 Objective Loss 0.272524 LR 0.001000 Time 0.030831 -2022-12-06 10:47:41,943 - Epoch: [57][ 90/ 1200] Overall Loss 0.274505 Objective Loss 0.274505 LR 0.001000 Time 0.029661 -2022-12-06 10:47:42,143 - Epoch: [57][ 100/ 1200] Overall Loss 0.272987 Objective Loss 0.272987 LR 0.001000 Time 0.028688 -2022-12-06 10:47:42,349 - Epoch: [57][ 110/ 1200] Overall Loss 0.276589 Objective Loss 0.276589 LR 0.001000 Time 0.027946 -2022-12-06 10:47:42,549 - Epoch: [57][ 120/ 1200] Overall Loss 0.275476 Objective Loss 0.275476 LR 0.001000 Time 0.027280 -2022-12-06 10:47:42,754 - Epoch: [57][ 130/ 1200] Overall Loss 0.275446 Objective Loss 0.275446 LR 0.001000 Time 0.026754 -2022-12-06 10:47:42,954 - Epoch: [57][ 140/ 1200] Overall Loss 0.275151 Objective Loss 0.275151 LR 0.001000 Time 0.026266 -2022-12-06 10:47:43,157 - Epoch: [57][ 150/ 1200] Overall Loss 0.276087 Objective Loss 0.276087 LR 0.001000 Time 0.025866 -2022-12-06 10:47:43,358 - Epoch: [57][ 160/ 1200] Overall Loss 0.275368 Objective Loss 0.275368 LR 0.001000 Time 0.025502 -2022-12-06 10:47:43,562 - Epoch: [57][ 170/ 1200] Overall Loss 0.273326 Objective Loss 0.273326 LR 0.001000 Time 0.025198 -2022-12-06 10:47:43,765 - Epoch: [57][ 180/ 1200] Overall Loss 0.272823 Objective Loss 0.272823 LR 0.001000 Time 0.024922 -2022-12-06 10:47:43,971 - Epoch: [57][ 190/ 1200] Overall Loss 0.271977 Objective Loss 0.271977 LR 0.001000 Time 0.024693 -2022-12-06 10:47:44,174 - Epoch: [57][ 200/ 1200] Overall Loss 0.270292 Objective Loss 0.270292 LR 0.001000 Time 0.024469 -2022-12-06 10:47:44,381 - Epoch: [57][ 210/ 1200] Overall Loss 0.269967 Objective Loss 0.269967 LR 0.001000 Time 0.024288 -2022-12-06 10:47:44,584 - Epoch: [57][ 220/ 1200] Overall Loss 0.269784 Objective Loss 0.269784 LR 0.001000 Time 0.024106 -2022-12-06 10:47:44,791 - Epoch: [57][ 230/ 1200] Overall Loss 0.270357 Objective Loss 0.270357 LR 0.001000 Time 0.023953 -2022-12-06 10:47:44,993 - Epoch: [57][ 240/ 1200] Overall Loss 0.268977 Objective Loss 0.268977 LR 0.001000 Time 0.023797 -2022-12-06 10:47:45,201 - Epoch: [57][ 250/ 1200] Overall Loss 0.269376 Objective Loss 0.269376 LR 0.001000 Time 0.023672 -2022-12-06 10:47:45,404 - Epoch: [57][ 260/ 1200] Overall Loss 0.268727 Objective Loss 0.268727 LR 0.001000 Time 0.023542 -2022-12-06 10:47:45,611 - Epoch: [57][ 270/ 1200] Overall Loss 0.269821 Objective Loss 0.269821 LR 0.001000 Time 0.023435 -2022-12-06 10:47:45,814 - Epoch: [57][ 280/ 1200] Overall Loss 0.269884 Objective Loss 0.269884 LR 0.001000 Time 0.023319 -2022-12-06 10:47:46,021 - Epoch: [57][ 290/ 1200] Overall Loss 0.268855 Objective Loss 0.268855 LR 0.001000 Time 0.023229 -2022-12-06 10:47:46,224 - Epoch: [57][ 300/ 1200] Overall Loss 0.268806 Objective Loss 0.268806 LR 0.001000 Time 0.023130 -2022-12-06 10:47:46,431 - Epoch: [57][ 310/ 1200] Overall Loss 0.268505 Objective Loss 0.268505 LR 0.001000 Time 0.023049 -2022-12-06 10:47:46,634 - Epoch: [57][ 320/ 1200] Overall Loss 0.269230 Objective Loss 0.269230 LR 0.001000 Time 0.022960 -2022-12-06 10:47:46,841 - Epoch: [57][ 330/ 1200] Overall Loss 0.268798 Objective Loss 0.268798 LR 0.001000 Time 0.022890 -2022-12-06 10:47:47,044 - Epoch: [57][ 340/ 1200] Overall Loss 0.268925 Objective Loss 0.268925 LR 0.001000 Time 0.022812 -2022-12-06 10:47:47,251 - Epoch: [57][ 350/ 1200] Overall Loss 0.268795 Objective Loss 0.268795 LR 0.001000 Time 0.022752 -2022-12-06 10:47:47,453 - Epoch: [57][ 360/ 1200] Overall Loss 0.269378 Objective Loss 0.269378 LR 0.001000 Time 0.022679 -2022-12-06 10:47:47,661 - Epoch: [57][ 370/ 1200] Overall Loss 0.268962 Objective Loss 0.268962 LR 0.001000 Time 0.022625 -2022-12-06 10:47:47,864 - Epoch: [57][ 380/ 1200] Overall Loss 0.268793 Objective Loss 0.268793 LR 0.001000 Time 0.022564 -2022-12-06 10:47:48,071 - Epoch: [57][ 390/ 1200] Overall Loss 0.269930 Objective Loss 0.269930 LR 0.001000 Time 0.022514 -2022-12-06 10:47:48,274 - Epoch: [57][ 400/ 1200] Overall Loss 0.269827 Objective Loss 0.269827 LR 0.001000 Time 0.022457 -2022-12-06 10:47:48,481 - Epoch: [57][ 410/ 1200] Overall Loss 0.269409 Objective Loss 0.269409 LR 0.001000 Time 0.022413 -2022-12-06 10:47:48,684 - Epoch: [57][ 420/ 1200] Overall Loss 0.269113 Objective Loss 0.269113 LR 0.001000 Time 0.022361 -2022-12-06 10:47:48,890 - Epoch: [57][ 430/ 1200] Overall Loss 0.269138 Objective Loss 0.269138 LR 0.001000 Time 0.022319 -2022-12-06 10:47:49,093 - Epoch: [57][ 440/ 1200] Overall Loss 0.269046 Objective Loss 0.269046 LR 0.001000 Time 0.022273 -2022-12-06 10:47:49,300 - Epoch: [57][ 450/ 1200] Overall Loss 0.269451 Objective Loss 0.269451 LR 0.001000 Time 0.022237 -2022-12-06 10:47:49,504 - Epoch: [57][ 460/ 1200] Overall Loss 0.269101 Objective Loss 0.269101 LR 0.001000 Time 0.022195 -2022-12-06 10:47:49,711 - Epoch: [57][ 470/ 1200] Overall Loss 0.268740 Objective Loss 0.268740 LR 0.001000 Time 0.022162 -2022-12-06 10:47:49,914 - Epoch: [57][ 480/ 1200] Overall Loss 0.268883 Objective Loss 0.268883 LR 0.001000 Time 0.022122 -2022-12-06 10:47:50,120 - Epoch: [57][ 490/ 1200] Overall Loss 0.269057 Objective Loss 0.269057 LR 0.001000 Time 0.022090 -2022-12-06 10:47:50,324 - Epoch: [57][ 500/ 1200] Overall Loss 0.269020 Objective Loss 0.269020 LR 0.001000 Time 0.022055 -2022-12-06 10:47:50,531 - Epoch: [57][ 510/ 1200] Overall Loss 0.269217 Objective Loss 0.269217 LR 0.001000 Time 0.022027 -2022-12-06 10:47:50,733 - Epoch: [57][ 520/ 1200] Overall Loss 0.269424 Objective Loss 0.269424 LR 0.001000 Time 0.021991 -2022-12-06 10:47:50,940 - Epoch: [57][ 530/ 1200] Overall Loss 0.269190 Objective Loss 0.269190 LR 0.001000 Time 0.021966 -2022-12-06 10:47:51,143 - Epoch: [57][ 540/ 1200] Overall Loss 0.268933 Objective Loss 0.268933 LR 0.001000 Time 0.021934 -2022-12-06 10:47:51,350 - Epoch: [57][ 550/ 1200] Overall Loss 0.269103 Objective Loss 0.269103 LR 0.001000 Time 0.021910 -2022-12-06 10:47:51,552 - Epoch: [57][ 560/ 1200] Overall Loss 0.269353 Objective Loss 0.269353 LR 0.001000 Time 0.021878 -2022-12-06 10:47:51,758 - Epoch: [57][ 570/ 1200] Overall Loss 0.269781 Objective Loss 0.269781 LR 0.001000 Time 0.021855 -2022-12-06 10:47:51,961 - Epoch: [57][ 580/ 1200] Overall Loss 0.270682 Objective Loss 0.270682 LR 0.001000 Time 0.021828 -2022-12-06 10:47:52,168 - Epoch: [57][ 590/ 1200] Overall Loss 0.270515 Objective Loss 0.270515 LR 0.001000 Time 0.021808 -2022-12-06 10:47:52,371 - Epoch: [57][ 600/ 1200] Overall Loss 0.270507 Objective Loss 0.270507 LR 0.001000 Time 0.021781 -2022-12-06 10:47:52,578 - Epoch: [57][ 610/ 1200] Overall Loss 0.270137 Objective Loss 0.270137 LR 0.001000 Time 0.021763 -2022-12-06 10:47:52,781 - Epoch: [57][ 620/ 1200] Overall Loss 0.270571 Objective Loss 0.270571 LR 0.001000 Time 0.021737 -2022-12-06 10:47:52,987 - Epoch: [57][ 630/ 1200] Overall Loss 0.270635 Objective Loss 0.270635 LR 0.001000 Time 0.021720 -2022-12-06 10:47:53,190 - Epoch: [57][ 640/ 1200] Overall Loss 0.270394 Objective Loss 0.270394 LR 0.001000 Time 0.021695 -2022-12-06 10:47:53,397 - Epoch: [57][ 650/ 1200] Overall Loss 0.270486 Objective Loss 0.270486 LR 0.001000 Time 0.021680 -2022-12-06 10:47:53,600 - Epoch: [57][ 660/ 1200] Overall Loss 0.270654 Objective Loss 0.270654 LR 0.001000 Time 0.021657 -2022-12-06 10:47:53,807 - Epoch: [57][ 670/ 1200] Overall Loss 0.270566 Objective Loss 0.270566 LR 0.001000 Time 0.021642 -2022-12-06 10:47:54,009 - Epoch: [57][ 680/ 1200] Overall Loss 0.271067 Objective Loss 0.271067 LR 0.001000 Time 0.021620 -2022-12-06 10:47:54,215 - Epoch: [57][ 690/ 1200] Overall Loss 0.271323 Objective Loss 0.271323 LR 0.001000 Time 0.021606 -2022-12-06 10:47:54,418 - Epoch: [57][ 700/ 1200] Overall Loss 0.271670 Objective Loss 0.271670 LR 0.001000 Time 0.021586 -2022-12-06 10:47:54,625 - Epoch: [57][ 710/ 1200] Overall Loss 0.271770 Objective Loss 0.271770 LR 0.001000 Time 0.021573 -2022-12-06 10:47:54,828 - Epoch: [57][ 720/ 1200] Overall Loss 0.271845 Objective Loss 0.271845 LR 0.001000 Time 0.021554 -2022-12-06 10:47:55,035 - Epoch: [57][ 730/ 1200] Overall Loss 0.271326 Objective Loss 0.271326 LR 0.001000 Time 0.021542 -2022-12-06 10:47:55,238 - Epoch: [57][ 740/ 1200] Overall Loss 0.271721 Objective Loss 0.271721 LR 0.001000 Time 0.021524 -2022-12-06 10:47:55,445 - Epoch: [57][ 750/ 1200] Overall Loss 0.271782 Objective Loss 0.271782 LR 0.001000 Time 0.021512 -2022-12-06 10:47:55,649 - Epoch: [57][ 760/ 1200] Overall Loss 0.272038 Objective Loss 0.272038 LR 0.001000 Time 0.021496 -2022-12-06 10:47:55,856 - Epoch: [57][ 770/ 1200] Overall Loss 0.272065 Objective Loss 0.272065 LR 0.001000 Time 0.021485 -2022-12-06 10:47:56,059 - Epoch: [57][ 780/ 1200] Overall Loss 0.272131 Objective Loss 0.272131 LR 0.001000 Time 0.021470 -2022-12-06 10:47:56,266 - Epoch: [57][ 790/ 1200] Overall Loss 0.272112 Objective Loss 0.272112 LR 0.001000 Time 0.021459 -2022-12-06 10:47:56,468 - Epoch: [57][ 800/ 1200] Overall Loss 0.272229 Objective Loss 0.272229 LR 0.001000 Time 0.021443 -2022-12-06 10:47:56,675 - Epoch: [57][ 810/ 1200] Overall Loss 0.271710 Objective Loss 0.271710 LR 0.001000 Time 0.021433 -2022-12-06 10:47:56,878 - Epoch: [57][ 820/ 1200] Overall Loss 0.271555 Objective Loss 0.271555 LR 0.001000 Time 0.021418 -2022-12-06 10:47:57,084 - Epoch: [57][ 830/ 1200] Overall Loss 0.271521 Objective Loss 0.271521 LR 0.001000 Time 0.021408 -2022-12-06 10:47:57,286 - Epoch: [57][ 840/ 1200] Overall Loss 0.271608 Objective Loss 0.271608 LR 0.001000 Time 0.021393 -2022-12-06 10:47:57,493 - Epoch: [57][ 850/ 1200] Overall Loss 0.271702 Objective Loss 0.271702 LR 0.001000 Time 0.021384 -2022-12-06 10:47:57,697 - Epoch: [57][ 860/ 1200] Overall Loss 0.271716 Objective Loss 0.271716 LR 0.001000 Time 0.021372 -2022-12-06 10:47:57,904 - Epoch: [57][ 870/ 1200] Overall Loss 0.271938 Objective Loss 0.271938 LR 0.001000 Time 0.021363 -2022-12-06 10:47:58,107 - Epoch: [57][ 880/ 1200] Overall Loss 0.271764 Objective Loss 0.271764 LR 0.001000 Time 0.021351 -2022-12-06 10:47:58,313 - Epoch: [57][ 890/ 1200] Overall Loss 0.271873 Objective Loss 0.271873 LR 0.001000 Time 0.021342 -2022-12-06 10:47:58,516 - Epoch: [57][ 900/ 1200] Overall Loss 0.272046 Objective Loss 0.272046 LR 0.001000 Time 0.021329 -2022-12-06 10:47:58,722 - Epoch: [57][ 910/ 1200] Overall Loss 0.272035 Objective Loss 0.272035 LR 0.001000 Time 0.021321 -2022-12-06 10:47:58,925 - Epoch: [57][ 920/ 1200] Overall Loss 0.271841 Objective Loss 0.271841 LR 0.001000 Time 0.021310 -2022-12-06 10:47:59,132 - Epoch: [57][ 930/ 1200] Overall Loss 0.271585 Objective Loss 0.271585 LR 0.001000 Time 0.021302 -2022-12-06 10:47:59,334 - Epoch: [57][ 940/ 1200] Overall Loss 0.271767 Objective Loss 0.271767 LR 0.001000 Time 0.021290 -2022-12-06 10:47:59,542 - Epoch: [57][ 950/ 1200] Overall Loss 0.271393 Objective Loss 0.271393 LR 0.001000 Time 0.021284 -2022-12-06 10:47:59,744 - Epoch: [57][ 960/ 1200] Overall Loss 0.271520 Objective Loss 0.271520 LR 0.001000 Time 0.021272 -2022-12-06 10:47:59,951 - Epoch: [57][ 970/ 1200] Overall Loss 0.271477 Objective Loss 0.271477 LR 0.001000 Time 0.021265 -2022-12-06 10:48:00,153 - Epoch: [57][ 980/ 1200] Overall Loss 0.271027 Objective Loss 0.271027 LR 0.001000 Time 0.021254 -2022-12-06 10:48:00,360 - Epoch: [57][ 990/ 1200] Overall Loss 0.270876 Objective Loss 0.270876 LR 0.001000 Time 0.021248 -2022-12-06 10:48:00,563 - Epoch: [57][ 1000/ 1200] Overall Loss 0.270975 Objective Loss 0.270975 LR 0.001000 Time 0.021238 -2022-12-06 10:48:00,769 - Epoch: [57][ 1010/ 1200] Overall Loss 0.270981 Objective Loss 0.270981 LR 0.001000 Time 0.021231 -2022-12-06 10:48:00,972 - Epoch: [57][ 1020/ 1200] Overall Loss 0.271046 Objective Loss 0.271046 LR 0.001000 Time 0.021222 -2022-12-06 10:48:01,179 - Epoch: [57][ 1030/ 1200] Overall Loss 0.271018 Objective Loss 0.271018 LR 0.001000 Time 0.021216 -2022-12-06 10:48:01,382 - Epoch: [57][ 1040/ 1200] Overall Loss 0.271239 Objective Loss 0.271239 LR 0.001000 Time 0.021207 -2022-12-06 10:48:01,588 - Epoch: [57][ 1050/ 1200] Overall Loss 0.271168 Objective Loss 0.271168 LR 0.001000 Time 0.021201 -2022-12-06 10:48:01,792 - Epoch: [57][ 1060/ 1200] Overall Loss 0.271128 Objective Loss 0.271128 LR 0.001000 Time 0.021192 -2022-12-06 10:48:01,998 - Epoch: [57][ 1070/ 1200] Overall Loss 0.270897 Objective Loss 0.270897 LR 0.001000 Time 0.021186 -2022-12-06 10:48:02,202 - Epoch: [57][ 1080/ 1200] Overall Loss 0.270891 Objective Loss 0.270891 LR 0.001000 Time 0.021178 -2022-12-06 10:48:02,407 - Epoch: [57][ 1090/ 1200] Overall Loss 0.270856 Objective Loss 0.270856 LR 0.001000 Time 0.021172 -2022-12-06 10:48:02,610 - Epoch: [57][ 1100/ 1200] Overall Loss 0.270809 Objective Loss 0.270809 LR 0.001000 Time 0.021163 -2022-12-06 10:48:02,818 - Epoch: [57][ 1110/ 1200] Overall Loss 0.270637 Objective Loss 0.270637 LR 0.001000 Time 0.021159 -2022-12-06 10:48:03,022 - Epoch: [57][ 1120/ 1200] Overall Loss 0.270783 Objective Loss 0.270783 LR 0.001000 Time 0.021152 -2022-12-06 10:48:03,229 - Epoch: [57][ 1130/ 1200] Overall Loss 0.270797 Objective Loss 0.270797 LR 0.001000 Time 0.021147 -2022-12-06 10:48:03,432 - Epoch: [57][ 1140/ 1200] Overall Loss 0.270924 Objective Loss 0.270924 LR 0.001000 Time 0.021140 -2022-12-06 10:48:03,639 - Epoch: [57][ 1150/ 1200] Overall Loss 0.270930 Objective Loss 0.270930 LR 0.001000 Time 0.021135 -2022-12-06 10:48:03,842 - Epoch: [57][ 1160/ 1200] Overall Loss 0.270780 Objective Loss 0.270780 LR 0.001000 Time 0.021127 -2022-12-06 10:48:04,048 - Epoch: [57][ 1170/ 1200] Overall Loss 0.270884 Objective Loss 0.270884 LR 0.001000 Time 0.021123 -2022-12-06 10:48:04,251 - Epoch: [57][ 1180/ 1200] Overall Loss 0.271042 Objective Loss 0.271042 LR 0.001000 Time 0.021115 -2022-12-06 10:48:04,458 - Epoch: [57][ 1190/ 1200] Overall Loss 0.271214 Objective Loss 0.271214 LR 0.001000 Time 0.021111 -2022-12-06 10:48:04,690 - Epoch: [57][ 1200/ 1200] Overall Loss 0.271411 Objective Loss 0.271411 Top1 83.682008 Top5 98.535565 LR 0.001000 Time 0.021128 -2022-12-06 10:48:04,778 - --- validate (epoch=57)----------- -2022-12-06 10:48:04,779 - 34129 samples (256 per mini-batch) -2022-12-06 10:48:05,225 - Epoch: [57][ 10/ 134] Loss 0.342548 Top1 84.375000 Top5 97.460938 -2022-12-06 10:48:05,357 - Epoch: [57][ 20/ 134] Loss 0.329653 Top1 84.160156 Top5 97.851562 -2022-12-06 10:48:05,491 - Epoch: [57][ 30/ 134] Loss 0.319936 Top1 84.648438 Top5 98.007812 -2022-12-06 10:48:05,625 - Epoch: [57][ 40/ 134] Loss 0.314766 Top1 84.755859 Top5 98.085938 -2022-12-06 10:48:05,765 - Epoch: [57][ 50/ 134] Loss 0.318980 Top1 84.687500 Top5 98.062500 -2022-12-06 10:48:05,893 - Epoch: [57][ 60/ 134] Loss 0.325164 Top1 84.641927 Top5 98.027344 -2022-12-06 10:48:06,025 - Epoch: [57][ 70/ 134] Loss 0.323383 Top1 84.626116 Top5 98.041295 -2022-12-06 10:48:06,158 - Epoch: [57][ 80/ 134] Loss 0.323120 Top1 84.609375 Top5 97.993164 -2022-12-06 10:48:06,289 - Epoch: [57][ 90/ 134] Loss 0.317605 Top1 84.730903 Top5 97.986111 -2022-12-06 10:48:06,421 - Epoch: [57][ 100/ 134] Loss 0.313700 Top1 84.882812 Top5 98.007812 -2022-12-06 10:48:06,554 - Epoch: [57][ 110/ 134] Loss 0.313907 Top1 84.804688 Top5 98.011364 -2022-12-06 10:48:06,689 - Epoch: [57][ 120/ 134] Loss 0.309365 Top1 84.895833 Top5 98.037109 -2022-12-06 10:48:06,821 - Epoch: [57][ 130/ 134] Loss 0.307250 Top1 84.891827 Top5 98.037861 -2022-12-06 10:48:06,860 - Epoch: [57][ 134/ 134] Loss 0.308607 Top1 84.848662 Top5 98.033930 -2022-12-06 10:48:06,950 - ==> Top1: 84.849 Top5: 98.034 Loss: 0.309 - -2022-12-06 10:48:06,951 - ==> Confusion: -[[ 894 1 0 1 9 4 0 2 8 59 0 3 2 3 1 3 1 1 0 0 4] - [ 3 900 3 4 9 23 3 27 3 0 0 5 5 2 0 1 8 3 15 3 10] - [ 6 1 1018 7 4 3 9 13 1 3 2 2 2 0 1 3 2 4 5 4 13] - [ 2 2 29 924 0 4 0 3 2 1 7 0 10 3 7 2 3 2 11 0 8] - [ 16 6 2 1 940 6 1 3 1 8 3 4 1 3 5 5 4 1 1 1 8] - [ 4 27 2 0 5 945 2 30 2 2 2 20 5 5 2 2 1 3 1 4 5] - [ 2 2 27 3 0 1 1046 7 1 0 4 3 1 1 0 5 1 1 0 10 3] - [ 1 4 4 2 1 22 2 957 0 0 3 10 1 0 1 1 0 1 23 16 5] - [ 5 4 1 0 0 0 0 0 970 48 8 1 3 8 4 2 2 1 2 1 4] - [ 78 1 1 0 5 2 0 3 28 861 1 2 1 8 1 1 0 2 0 1 5] - [ 0 2 4 10 2 2 1 5 13 1 935 3 3 9 2 2 1 0 11 1 12] - [ 4 0 2 1 0 8 1 2 3 2 1 967 26 5 0 5 7 4 1 10 2] - [ 3 1 1 2 0 1 1 1 0 0 1 46 873 3 0 10 1 11 1 2 11] - [ 1 0 2 1 6 15 0 4 18 23 12 12 5 893 2 2 8 1 0 4 14] - [ 11 5 6 17 6 1 0 1 44 5 3 2 6 3 1003 0 1 0 6 0 10] - [ 1 0 3 0 4 1 3 1 0 0 0 10 5 3 0 990 6 9 0 3 4] - [ 2 3 3 3 4 3 0 1 0 1 0 1 2 1 0 14 1015 4 0 7 8] - [ 4 0 2 3 0 2 0 1 2 1 0 9 30 4 0 12 2 961 0 2 1] - [ 3 4 3 10 0 1 0 30 2 1 6 2 1 2 7 0 1 0 927 4 4] - [ 1 1 1 2 1 6 3 10 1 0 0 22 6 5 0 2 7 3 1 1000 8] - [ 152 187 196 95 114 164 71 209 122 126 136 152 353 225 116 147 198 107 183 249 9924]] - -2022-12-06 10:48:07,522 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:48:07,522 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:48:07,528 - - -2022-12-06 10:48:07,528 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:48:08,568 - Epoch: [58][ 10/ 1200] Overall Loss 0.272792 Objective Loss 0.272792 LR 0.001000 Time 0.103943 -2022-12-06 10:48:08,766 - Epoch: [58][ 20/ 1200] Overall Loss 0.263183 Objective Loss 0.263183 LR 0.001000 Time 0.061828 -2022-12-06 10:48:08,966 - Epoch: [58][ 30/ 1200] Overall Loss 0.264724 Objective Loss 0.264724 LR 0.001000 Time 0.047888 -2022-12-06 10:48:09,162 - Epoch: [58][ 40/ 1200] Overall Loss 0.262871 Objective Loss 0.262871 LR 0.001000 Time 0.040789 -2022-12-06 10:48:09,361 - Epoch: [58][ 50/ 1200] Overall Loss 0.264141 Objective Loss 0.264141 LR 0.001000 Time 0.036599 -2022-12-06 10:48:09,557 - Epoch: [58][ 60/ 1200] Overall Loss 0.262097 Objective Loss 0.262097 LR 0.001000 Time 0.033751 -2022-12-06 10:48:09,755 - Epoch: [58][ 70/ 1200] Overall Loss 0.263062 Objective Loss 0.263062 LR 0.001000 Time 0.031753 -2022-12-06 10:48:09,950 - Epoch: [58][ 80/ 1200] Overall Loss 0.263032 Objective Loss 0.263032 LR 0.001000 Time 0.030219 -2022-12-06 10:48:10,148 - Epoch: [58][ 90/ 1200] Overall Loss 0.265821 Objective Loss 0.265821 LR 0.001000 Time 0.029060 -2022-12-06 10:48:10,344 - Epoch: [58][ 100/ 1200] Overall Loss 0.266281 Objective Loss 0.266281 LR 0.001000 Time 0.028107 -2022-12-06 10:48:10,543 - Epoch: [58][ 110/ 1200] Overall Loss 0.265179 Objective Loss 0.265179 LR 0.001000 Time 0.027351 -2022-12-06 10:48:10,738 - Epoch: [58][ 120/ 1200] Overall Loss 0.265662 Objective Loss 0.265662 LR 0.001000 Time 0.026695 -2022-12-06 10:48:10,937 - Epoch: [58][ 130/ 1200] Overall Loss 0.264538 Objective Loss 0.264538 LR 0.001000 Time 0.026165 -2022-12-06 10:48:11,132 - Epoch: [58][ 140/ 1200] Overall Loss 0.265288 Objective Loss 0.265288 LR 0.001000 Time 0.025684 -2022-12-06 10:48:11,330 - Epoch: [58][ 150/ 1200] Overall Loss 0.264401 Objective Loss 0.264401 LR 0.001000 Time 0.025291 -2022-12-06 10:48:11,526 - Epoch: [58][ 160/ 1200] Overall Loss 0.262733 Objective Loss 0.262733 LR 0.001000 Time 0.024929 -2022-12-06 10:48:11,724 - Epoch: [58][ 170/ 1200] Overall Loss 0.263483 Objective Loss 0.263483 LR 0.001000 Time 0.024626 -2022-12-06 10:48:11,920 - Epoch: [58][ 180/ 1200] Overall Loss 0.263123 Objective Loss 0.263123 LR 0.001000 Time 0.024342 -2022-12-06 10:48:12,117 - Epoch: [58][ 190/ 1200] Overall Loss 0.263568 Objective Loss 0.263568 LR 0.001000 Time 0.024098 -2022-12-06 10:48:12,312 - Epoch: [58][ 200/ 1200] Overall Loss 0.261768 Objective Loss 0.261768 LR 0.001000 Time 0.023865 -2022-12-06 10:48:12,511 - Epoch: [58][ 210/ 1200] Overall Loss 0.262013 Objective Loss 0.262013 LR 0.001000 Time 0.023671 -2022-12-06 10:48:12,706 - Epoch: [58][ 220/ 1200] Overall Loss 0.260982 Objective Loss 0.260982 LR 0.001000 Time 0.023482 -2022-12-06 10:48:12,904 - Epoch: [58][ 230/ 1200] Overall Loss 0.261222 Objective Loss 0.261222 LR 0.001000 Time 0.023320 -2022-12-06 10:48:13,099 - Epoch: [58][ 240/ 1200] Overall Loss 0.261181 Objective Loss 0.261181 LR 0.001000 Time 0.023157 -2022-12-06 10:48:13,297 - Epoch: [58][ 250/ 1200] Overall Loss 0.261441 Objective Loss 0.261441 LR 0.001000 Time 0.023019 -2022-12-06 10:48:13,491 - Epoch: [58][ 260/ 1200] Overall Loss 0.260947 Objective Loss 0.260947 LR 0.001000 Time 0.022881 -2022-12-06 10:48:13,690 - Epoch: [58][ 270/ 1200] Overall Loss 0.261613 Objective Loss 0.261613 LR 0.001000 Time 0.022767 -2022-12-06 10:48:13,885 - Epoch: [58][ 280/ 1200] Overall Loss 0.261763 Objective Loss 0.261763 LR 0.001000 Time 0.022647 -2022-12-06 10:48:14,083 - Epoch: [58][ 290/ 1200] Overall Loss 0.262149 Objective Loss 0.262149 LR 0.001000 Time 0.022548 -2022-12-06 10:48:14,279 - Epoch: [58][ 300/ 1200] Overall Loss 0.262810 Objective Loss 0.262810 LR 0.001000 Time 0.022446 -2022-12-06 10:48:14,477 - Epoch: [58][ 310/ 1200] Overall Loss 0.262874 Objective Loss 0.262874 LR 0.001000 Time 0.022361 -2022-12-06 10:48:14,672 - Epoch: [58][ 320/ 1200] Overall Loss 0.262587 Objective Loss 0.262587 LR 0.001000 Time 0.022271 -2022-12-06 10:48:14,871 - Epoch: [58][ 330/ 1200] Overall Loss 0.262547 Objective Loss 0.262547 LR 0.001000 Time 0.022195 -2022-12-06 10:48:15,066 - Epoch: [58][ 340/ 1200] Overall Loss 0.262382 Objective Loss 0.262382 LR 0.001000 Time 0.022114 -2022-12-06 10:48:15,264 - Epoch: [58][ 350/ 1200] Overall Loss 0.262503 Objective Loss 0.262503 LR 0.001000 Time 0.022047 -2022-12-06 10:48:15,459 - Epoch: [58][ 360/ 1200] Overall Loss 0.262915 Objective Loss 0.262915 LR 0.001000 Time 0.021975 -2022-12-06 10:48:15,658 - Epoch: [58][ 370/ 1200] Overall Loss 0.263143 Objective Loss 0.263143 LR 0.001000 Time 0.021917 -2022-12-06 10:48:15,854 - Epoch: [58][ 380/ 1200] Overall Loss 0.263418 Objective Loss 0.263418 LR 0.001000 Time 0.021855 -2022-12-06 10:48:16,052 - Epoch: [58][ 390/ 1200] Overall Loss 0.264015 Objective Loss 0.264015 LR 0.001000 Time 0.021801 -2022-12-06 10:48:16,247 - Epoch: [58][ 400/ 1200] Overall Loss 0.263423 Objective Loss 0.263423 LR 0.001000 Time 0.021741 -2022-12-06 10:48:16,445 - Epoch: [58][ 410/ 1200] Overall Loss 0.263708 Objective Loss 0.263708 LR 0.001000 Time 0.021693 -2022-12-06 10:48:16,640 - Epoch: [58][ 420/ 1200] Overall Loss 0.263858 Objective Loss 0.263858 LR 0.001000 Time 0.021639 -2022-12-06 10:48:16,838 - Epoch: [58][ 430/ 1200] Overall Loss 0.264022 Objective Loss 0.264022 LR 0.001000 Time 0.021596 -2022-12-06 10:48:17,033 - Epoch: [58][ 440/ 1200] Overall Loss 0.264458 Objective Loss 0.264458 LR 0.001000 Time 0.021547 -2022-12-06 10:48:17,231 - Epoch: [58][ 450/ 1200] Overall Loss 0.264281 Objective Loss 0.264281 LR 0.001000 Time 0.021506 -2022-12-06 10:48:17,425 - Epoch: [58][ 460/ 1200] Overall Loss 0.265064 Objective Loss 0.265064 LR 0.001000 Time 0.021461 -2022-12-06 10:48:17,624 - Epoch: [58][ 470/ 1200] Overall Loss 0.265664 Objective Loss 0.265664 LR 0.001000 Time 0.021425 -2022-12-06 10:48:17,818 - Epoch: [58][ 480/ 1200] Overall Loss 0.265941 Objective Loss 0.265941 LR 0.001000 Time 0.021383 -2022-12-06 10:48:18,017 - Epoch: [58][ 490/ 1200] Overall Loss 0.266025 Objective Loss 0.266025 LR 0.001000 Time 0.021350 -2022-12-06 10:48:18,211 - Epoch: [58][ 500/ 1200] Overall Loss 0.266585 Objective Loss 0.266585 LR 0.001000 Time 0.021311 -2022-12-06 10:48:18,410 - Epoch: [58][ 510/ 1200] Overall Loss 0.266196 Objective Loss 0.266196 LR 0.001000 Time 0.021281 -2022-12-06 10:48:18,605 - Epoch: [58][ 520/ 1200] Overall Loss 0.266295 Objective Loss 0.266295 LR 0.001000 Time 0.021246 -2022-12-06 10:48:18,803 - Epoch: [58][ 530/ 1200] Overall Loss 0.266465 Objective Loss 0.266465 LR 0.001000 Time 0.021218 -2022-12-06 10:48:18,998 - Epoch: [58][ 540/ 1200] Overall Loss 0.266566 Objective Loss 0.266566 LR 0.001000 Time 0.021186 -2022-12-06 10:48:19,197 - Epoch: [58][ 550/ 1200] Overall Loss 0.266642 Objective Loss 0.266642 LR 0.001000 Time 0.021160 -2022-12-06 10:48:19,391 - Epoch: [58][ 560/ 1200] Overall Loss 0.266870 Objective Loss 0.266870 LR 0.001000 Time 0.021129 -2022-12-06 10:48:19,590 - Epoch: [58][ 570/ 1200] Overall Loss 0.266943 Objective Loss 0.266943 LR 0.001000 Time 0.021106 -2022-12-06 10:48:19,785 - Epoch: [58][ 580/ 1200] Overall Loss 0.267492 Objective Loss 0.267492 LR 0.001000 Time 0.021077 -2022-12-06 10:48:19,983 - Epoch: [58][ 590/ 1200] Overall Loss 0.268153 Objective Loss 0.268153 LR 0.001000 Time 0.021055 -2022-12-06 10:48:20,179 - Epoch: [58][ 600/ 1200] Overall Loss 0.268835 Objective Loss 0.268835 LR 0.001000 Time 0.021029 -2022-12-06 10:48:20,377 - Epoch: [58][ 610/ 1200] Overall Loss 0.268978 Objective Loss 0.268978 LR 0.001000 Time 0.021009 -2022-12-06 10:48:20,572 - Epoch: [58][ 620/ 1200] Overall Loss 0.269450 Objective Loss 0.269450 LR 0.001000 Time 0.020983 -2022-12-06 10:48:20,770 - Epoch: [58][ 630/ 1200] Overall Loss 0.269408 Objective Loss 0.269408 LR 0.001000 Time 0.020964 -2022-12-06 10:48:20,966 - Epoch: [58][ 640/ 1200] Overall Loss 0.269061 Objective Loss 0.269061 LR 0.001000 Time 0.020941 -2022-12-06 10:48:21,164 - Epoch: [58][ 650/ 1200] Overall Loss 0.268972 Objective Loss 0.268972 LR 0.001000 Time 0.020923 -2022-12-06 10:48:21,359 - Epoch: [58][ 660/ 1200] Overall Loss 0.268770 Objective Loss 0.268770 LR 0.001000 Time 0.020901 -2022-12-06 10:48:21,558 - Epoch: [58][ 670/ 1200] Overall Loss 0.268464 Objective Loss 0.268464 LR 0.001000 Time 0.020885 -2022-12-06 10:48:21,753 - Epoch: [58][ 680/ 1200] Overall Loss 0.268445 Objective Loss 0.268445 LR 0.001000 Time 0.020864 -2022-12-06 10:48:21,951 - Epoch: [58][ 690/ 1200] Overall Loss 0.268370 Objective Loss 0.268370 LR 0.001000 Time 0.020848 -2022-12-06 10:48:22,146 - Epoch: [58][ 700/ 1200] Overall Loss 0.268950 Objective Loss 0.268950 LR 0.001000 Time 0.020828 -2022-12-06 10:48:22,344 - Epoch: [58][ 710/ 1200] Overall Loss 0.268959 Objective Loss 0.268959 LR 0.001000 Time 0.020813 -2022-12-06 10:48:22,540 - Epoch: [58][ 720/ 1200] Overall Loss 0.269312 Objective Loss 0.269312 LR 0.001000 Time 0.020794 -2022-12-06 10:48:22,738 - Epoch: [58][ 730/ 1200] Overall Loss 0.269589 Objective Loss 0.269589 LR 0.001000 Time 0.020780 -2022-12-06 10:48:22,933 - Epoch: [58][ 740/ 1200] Overall Loss 0.269820 Objective Loss 0.269820 LR 0.001000 Time 0.020762 -2022-12-06 10:48:23,131 - Epoch: [58][ 750/ 1200] Overall Loss 0.270154 Objective Loss 0.270154 LR 0.001000 Time 0.020749 -2022-12-06 10:48:23,326 - Epoch: [58][ 760/ 1200] Overall Loss 0.269918 Objective Loss 0.269918 LR 0.001000 Time 0.020732 -2022-12-06 10:48:23,525 - Epoch: [58][ 770/ 1200] Overall Loss 0.270098 Objective Loss 0.270098 LR 0.001000 Time 0.020720 -2022-12-06 10:48:23,721 - Epoch: [58][ 780/ 1200] Overall Loss 0.270012 Objective Loss 0.270012 LR 0.001000 Time 0.020705 -2022-12-06 10:48:23,920 - Epoch: [58][ 790/ 1200] Overall Loss 0.269837 Objective Loss 0.269837 LR 0.001000 Time 0.020694 -2022-12-06 10:48:24,115 - Epoch: [58][ 800/ 1200] Overall Loss 0.269852 Objective Loss 0.269852 LR 0.001000 Time 0.020679 -2022-12-06 10:48:24,314 - Epoch: [58][ 810/ 1200] Overall Loss 0.269577 Objective Loss 0.269577 LR 0.001000 Time 0.020668 -2022-12-06 10:48:24,509 - Epoch: [58][ 820/ 1200] Overall Loss 0.269651 Objective Loss 0.269651 LR 0.001000 Time 0.020654 -2022-12-06 10:48:24,708 - Epoch: [58][ 830/ 1200] Overall Loss 0.269436 Objective Loss 0.269436 LR 0.001000 Time 0.020643 -2022-12-06 10:48:24,903 - Epoch: [58][ 840/ 1200] Overall Loss 0.269718 Objective Loss 0.269718 LR 0.001000 Time 0.020629 -2022-12-06 10:48:25,101 - Epoch: [58][ 850/ 1200] Overall Loss 0.269650 Objective Loss 0.269650 LR 0.001000 Time 0.020618 -2022-12-06 10:48:25,295 - Epoch: [58][ 860/ 1200] Overall Loss 0.269563 Objective Loss 0.269563 LR 0.001000 Time 0.020605 -2022-12-06 10:48:25,494 - Epoch: [58][ 870/ 1200] Overall Loss 0.269574 Objective Loss 0.269574 LR 0.001000 Time 0.020595 -2022-12-06 10:48:25,689 - Epoch: [58][ 880/ 1200] Overall Loss 0.269809 Objective Loss 0.269809 LR 0.001000 Time 0.020582 -2022-12-06 10:48:25,887 - Epoch: [58][ 890/ 1200] Overall Loss 0.269836 Objective Loss 0.269836 LR 0.001000 Time 0.020573 -2022-12-06 10:48:26,082 - Epoch: [58][ 900/ 1200] Overall Loss 0.269790 Objective Loss 0.269790 LR 0.001000 Time 0.020561 -2022-12-06 10:48:26,280 - Epoch: [58][ 910/ 1200] Overall Loss 0.269497 Objective Loss 0.269497 LR 0.001000 Time 0.020551 -2022-12-06 10:48:26,475 - Epoch: [58][ 920/ 1200] Overall Loss 0.269378 Objective Loss 0.269378 LR 0.001000 Time 0.020539 -2022-12-06 10:48:26,673 - Epoch: [58][ 930/ 1200] Overall Loss 0.269352 Objective Loss 0.269352 LR 0.001000 Time 0.020531 -2022-12-06 10:48:26,868 - Epoch: [58][ 940/ 1200] Overall Loss 0.269344 Objective Loss 0.269344 LR 0.001000 Time 0.020520 -2022-12-06 10:48:27,066 - Epoch: [58][ 950/ 1200] Overall Loss 0.269145 Objective Loss 0.269145 LR 0.001000 Time 0.020511 -2022-12-06 10:48:27,262 - Epoch: [58][ 960/ 1200] Overall Loss 0.268839 Objective Loss 0.268839 LR 0.001000 Time 0.020501 -2022-12-06 10:48:27,460 - Epoch: [58][ 970/ 1200] Overall Loss 0.268669 Objective Loss 0.268669 LR 0.001000 Time 0.020493 -2022-12-06 10:48:27,655 - Epoch: [58][ 980/ 1200] Overall Loss 0.268874 Objective Loss 0.268874 LR 0.001000 Time 0.020482 -2022-12-06 10:48:27,853 - Epoch: [58][ 990/ 1200] Overall Loss 0.268978 Objective Loss 0.268978 LR 0.001000 Time 0.020475 -2022-12-06 10:48:28,048 - Epoch: [58][ 1000/ 1200] Overall Loss 0.269071 Objective Loss 0.269071 LR 0.001000 Time 0.020464 -2022-12-06 10:48:28,246 - Epoch: [58][ 1010/ 1200] Overall Loss 0.269358 Objective Loss 0.269358 LR 0.001000 Time 0.020458 -2022-12-06 10:48:28,442 - Epoch: [58][ 1020/ 1200] Overall Loss 0.269134 Objective Loss 0.269134 LR 0.001000 Time 0.020448 -2022-12-06 10:48:28,640 - Epoch: [58][ 1030/ 1200] Overall Loss 0.269325 Objective Loss 0.269325 LR 0.001000 Time 0.020442 -2022-12-06 10:48:28,835 - Epoch: [58][ 1040/ 1200] Overall Loss 0.269490 Objective Loss 0.269490 LR 0.001000 Time 0.020433 -2022-12-06 10:48:29,034 - Epoch: [58][ 1050/ 1200] Overall Loss 0.269726 Objective Loss 0.269726 LR 0.001000 Time 0.020426 -2022-12-06 10:48:29,229 - Epoch: [58][ 1060/ 1200] Overall Loss 0.269531 Objective Loss 0.269531 LR 0.001000 Time 0.020417 -2022-12-06 10:48:29,427 - Epoch: [58][ 1070/ 1200] Overall Loss 0.269443 Objective Loss 0.269443 LR 0.001000 Time 0.020411 -2022-12-06 10:48:29,621 - Epoch: [58][ 1080/ 1200] Overall Loss 0.269861 Objective Loss 0.269861 LR 0.001000 Time 0.020402 -2022-12-06 10:48:29,819 - Epoch: [58][ 1090/ 1200] Overall Loss 0.269773 Objective Loss 0.269773 LR 0.001000 Time 0.020395 -2022-12-06 10:48:30,014 - Epoch: [58][ 1100/ 1200] Overall Loss 0.270081 Objective Loss 0.270081 LR 0.001000 Time 0.020387 -2022-12-06 10:48:30,213 - Epoch: [58][ 1110/ 1200] Overall Loss 0.269885 Objective Loss 0.269885 LR 0.001000 Time 0.020381 -2022-12-06 10:48:30,407 - Epoch: [58][ 1120/ 1200] Overall Loss 0.270018 Objective Loss 0.270018 LR 0.001000 Time 0.020373 -2022-12-06 10:48:30,605 - Epoch: [58][ 1130/ 1200] Overall Loss 0.270014 Objective Loss 0.270014 LR 0.001000 Time 0.020367 -2022-12-06 10:48:30,800 - Epoch: [58][ 1140/ 1200] Overall Loss 0.270212 Objective Loss 0.270212 LR 0.001000 Time 0.020359 -2022-12-06 10:48:30,998 - Epoch: [58][ 1150/ 1200] Overall Loss 0.270225 Objective Loss 0.270225 LR 0.001000 Time 0.020354 -2022-12-06 10:48:31,193 - Epoch: [58][ 1160/ 1200] Overall Loss 0.270294 Objective Loss 0.270294 LR 0.001000 Time 0.020345 -2022-12-06 10:48:31,391 - Epoch: [58][ 1170/ 1200] Overall Loss 0.270115 Objective Loss 0.270115 LR 0.001000 Time 0.020340 -2022-12-06 10:48:31,586 - Epoch: [58][ 1180/ 1200] Overall Loss 0.270308 Objective Loss 0.270308 LR 0.001000 Time 0.020333 -2022-12-06 10:48:31,784 - Epoch: [58][ 1190/ 1200] Overall Loss 0.270477 Objective Loss 0.270477 LR 0.001000 Time 0.020328 -2022-12-06 10:48:32,008 - Epoch: [58][ 1200/ 1200] Overall Loss 0.270554 Objective Loss 0.270554 Top1 86.401674 Top5 98.953975 LR 0.001000 Time 0.020345 -2022-12-06 10:48:32,097 - --- validate (epoch=58)----------- -2022-12-06 10:48:32,098 - 34129 samples (256 per mini-batch) -2022-12-06 10:48:32,536 - Epoch: [58][ 10/ 134] Loss 0.301827 Top1 83.203125 Top5 97.695312 -2022-12-06 10:48:32,665 - Epoch: [58][ 20/ 134] Loss 0.301728 Top1 83.496094 Top5 97.832031 -2022-12-06 10:48:32,793 - Epoch: [58][ 30/ 134] Loss 0.297020 Top1 83.750000 Top5 97.838542 -2022-12-06 10:48:32,925 - Epoch: [58][ 40/ 134] Loss 0.302121 Top1 83.486328 Top5 97.705078 -2022-12-06 10:48:33,058 - Epoch: [58][ 50/ 134] Loss 0.303757 Top1 83.648438 Top5 97.734375 -2022-12-06 10:48:33,190 - Epoch: [58][ 60/ 134] Loss 0.311189 Top1 83.619792 Top5 97.708333 -2022-12-06 10:48:33,322 - Epoch: [58][ 70/ 134] Loss 0.313581 Top1 83.610491 Top5 97.734375 -2022-12-06 10:48:33,456 - Epoch: [58][ 80/ 134] Loss 0.312770 Top1 83.583984 Top5 97.690430 -2022-12-06 10:48:33,589 - Epoch: [58][ 90/ 134] Loss 0.312033 Top1 83.585069 Top5 97.690972 -2022-12-06 10:48:33,722 - Epoch: [58][ 100/ 134] Loss 0.310314 Top1 83.652344 Top5 97.703125 -2022-12-06 10:48:33,854 - Epoch: [58][ 110/ 134] Loss 0.308897 Top1 83.707386 Top5 97.670455 -2022-12-06 10:48:33,987 - Epoch: [58][ 120/ 134] Loss 0.311217 Top1 83.671875 Top5 97.636719 -2022-12-06 10:48:34,122 - Epoch: [58][ 130/ 134] Loss 0.308278 Top1 83.783053 Top5 97.623197 -2022-12-06 10:48:34,162 - Epoch: [58][ 134/ 134] Loss 0.307302 Top1 83.811421 Top5 97.632512 -2022-12-06 10:48:34,250 - ==> Top1: 83.811 Top5: 97.633 Loss: 0.307 - -2022-12-06 10:48:34,251 - ==> Confusion: -[[ 872 3 1 5 4 4 1 0 6 86 0 1 2 3 2 1 1 3 1 0 0] - [ 1 928 3 3 8 27 3 9 5 1 3 4 3 2 1 0 7 3 8 3 5] - [ 8 6 985 14 2 0 22 10 1 6 8 4 2 7 2 3 2 3 8 3 7] - [ 0 3 14 930 0 5 1 2 3 1 15 0 3 5 15 1 1 6 10 0 5] - [ 16 4 4 1 948 4 0 0 2 10 1 0 2 3 6 3 7 3 0 3 3] - [ 4 13 2 0 7 951 6 17 6 2 3 13 2 22 2 2 2 0 1 6 8] - [ 1 2 10 0 0 0 1070 3 0 0 5 1 1 3 1 7 1 1 1 9 2] - [ 1 12 7 3 0 31 11 921 0 3 3 7 3 6 2 0 2 1 26 12 3] - [ 5 4 0 0 0 2 1 0 975 60 4 2 1 4 3 0 1 0 0 1 1] - [ 37 0 1 0 3 2 0 2 19 920 1 2 0 6 1 0 0 2 1 1 3] - [ 0 3 3 4 2 0 0 1 18 4 946 2 3 18 3 0 2 1 5 1 3] - [ 4 2 3 0 0 7 3 2 2 0 2 966 17 11 1 3 4 6 0 13 5] - [ 2 2 1 4 2 6 2 1 1 1 0 50 859 3 1 9 1 15 0 2 7] - [ 0 1 0 1 0 6 0 2 12 25 6 4 1 954 0 1 2 2 0 1 5] - [ 12 5 0 17 6 1 0 1 33 10 2 3 2 3 1018 1 1 3 8 0 4] - [ 2 0 1 5 1 1 7 1 0 1 0 6 5 5 0 980 7 15 0 4 2] - [ 3 2 3 0 4 1 1 0 1 1 0 4 2 2 1 8 1030 0 0 5 4] - [ 1 2 1 1 0 0 1 1 4 3 0 12 17 4 0 9 1 975 1 2 1] - [ 3 4 3 10 2 3 2 23 3 1 5 1 4 4 11 1 1 0 922 3 2] - [ 1 3 0 0 1 12 8 8 0 0 2 14 8 15 0 2 6 4 0 992 4] - [ 123 255 185 103 155 196 91 131 153 175 216 151 339 360 113 156 306 92 185 283 9458]] - -2022-12-06 10:48:34,826 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:48:34,826 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:48:34,832 - - -2022-12-06 10:48:34,832 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:48:35,760 - Epoch: [59][ 10/ 1200] Overall Loss 0.258834 Objective Loss 0.258834 LR 0.001000 Time 0.092659 -2022-12-06 10:48:35,963 - Epoch: [59][ 20/ 1200] Overall Loss 0.273416 Objective Loss 0.273416 LR 0.001000 Time 0.056485 -2022-12-06 10:48:36,159 - Epoch: [59][ 30/ 1200] Overall Loss 0.272738 Objective Loss 0.272738 LR 0.001000 Time 0.044156 -2022-12-06 10:48:36,356 - Epoch: [59][ 40/ 1200] Overall Loss 0.273367 Objective Loss 0.273367 LR 0.001000 Time 0.038041 -2022-12-06 10:48:36,552 - Epoch: [59][ 50/ 1200] Overall Loss 0.276193 Objective Loss 0.276193 LR 0.001000 Time 0.034324 -2022-12-06 10:48:36,749 - Epoch: [59][ 60/ 1200] Overall Loss 0.274733 Objective Loss 0.274733 LR 0.001000 Time 0.031889 -2022-12-06 10:48:36,945 - Epoch: [59][ 70/ 1200] Overall Loss 0.274317 Objective Loss 0.274317 LR 0.001000 Time 0.030123 -2022-12-06 10:48:37,143 - Epoch: [59][ 80/ 1200] Overall Loss 0.269379 Objective Loss 0.269379 LR 0.001000 Time 0.028819 -2022-12-06 10:48:37,337 - Epoch: [59][ 90/ 1200] Overall Loss 0.271331 Objective Loss 0.271331 LR 0.001000 Time 0.027776 -2022-12-06 10:48:37,535 - Epoch: [59][ 100/ 1200] Overall Loss 0.268580 Objective Loss 0.268580 LR 0.001000 Time 0.026973 -2022-12-06 10:48:37,730 - Epoch: [59][ 110/ 1200] Overall Loss 0.270254 Objective Loss 0.270254 LR 0.001000 Time 0.026289 -2022-12-06 10:48:37,928 - Epoch: [59][ 120/ 1200] Overall Loss 0.268768 Objective Loss 0.268768 LR 0.001000 Time 0.025739 -2022-12-06 10:48:38,123 - Epoch: [59][ 130/ 1200] Overall Loss 0.268810 Objective Loss 0.268810 LR 0.001000 Time 0.025255 -2022-12-06 10:48:38,321 - Epoch: [59][ 140/ 1200] Overall Loss 0.270190 Objective Loss 0.270190 LR 0.001000 Time 0.024861 -2022-12-06 10:48:38,516 - Epoch: [59][ 150/ 1200] Overall Loss 0.268532 Objective Loss 0.268532 LR 0.001000 Time 0.024503 -2022-12-06 10:48:38,714 - Epoch: [59][ 160/ 1200] Overall Loss 0.268608 Objective Loss 0.268608 LR 0.001000 Time 0.024202 -2022-12-06 10:48:38,908 - Epoch: [59][ 170/ 1200] Overall Loss 0.267035 Objective Loss 0.267035 LR 0.001000 Time 0.023919 -2022-12-06 10:48:39,106 - Epoch: [59][ 180/ 1200] Overall Loss 0.265395 Objective Loss 0.265395 LR 0.001000 Time 0.023687 -2022-12-06 10:48:39,302 - Epoch: [59][ 190/ 1200] Overall Loss 0.264767 Objective Loss 0.264767 LR 0.001000 Time 0.023466 -2022-12-06 10:48:39,499 - Epoch: [59][ 200/ 1200] Overall Loss 0.265391 Objective Loss 0.265391 LR 0.001000 Time 0.023277 -2022-12-06 10:48:39,694 - Epoch: [59][ 210/ 1200] Overall Loss 0.265825 Objective Loss 0.265825 LR 0.001000 Time 0.023097 -2022-12-06 10:48:39,892 - Epoch: [59][ 220/ 1200] Overall Loss 0.265837 Objective Loss 0.265837 LR 0.001000 Time 0.022943 -2022-12-06 10:48:40,088 - Epoch: [59][ 230/ 1200] Overall Loss 0.264591 Objective Loss 0.264591 LR 0.001000 Time 0.022793 -2022-12-06 10:48:40,285 - Epoch: [59][ 240/ 1200] Overall Loss 0.265463 Objective Loss 0.265463 LR 0.001000 Time 0.022665 -2022-12-06 10:48:40,480 - Epoch: [59][ 250/ 1200] Overall Loss 0.265093 Objective Loss 0.265093 LR 0.001000 Time 0.022536 -2022-12-06 10:48:40,678 - Epoch: [59][ 260/ 1200] Overall Loss 0.265100 Objective Loss 0.265100 LR 0.001000 Time 0.022426 -2022-12-06 10:48:40,873 - Epoch: [59][ 270/ 1200] Overall Loss 0.264509 Objective Loss 0.264509 LR 0.001000 Time 0.022317 -2022-12-06 10:48:41,070 - Epoch: [59][ 280/ 1200] Overall Loss 0.263282 Objective Loss 0.263282 LR 0.001000 Time 0.022223 -2022-12-06 10:48:41,266 - Epoch: [59][ 290/ 1200] Overall Loss 0.262664 Objective Loss 0.262664 LR 0.001000 Time 0.022129 -2022-12-06 10:48:41,463 - Epoch: [59][ 300/ 1200] Overall Loss 0.262992 Objective Loss 0.262992 LR 0.001000 Time 0.022048 -2022-12-06 10:48:41,659 - Epoch: [59][ 310/ 1200] Overall Loss 0.263558 Objective Loss 0.263558 LR 0.001000 Time 0.021965 -2022-12-06 10:48:41,856 - Epoch: [59][ 320/ 1200] Overall Loss 0.264106 Objective Loss 0.264106 LR 0.001000 Time 0.021893 -2022-12-06 10:48:42,051 - Epoch: [59][ 330/ 1200] Overall Loss 0.263651 Objective Loss 0.263651 LR 0.001000 Time 0.021819 -2022-12-06 10:48:42,249 - Epoch: [59][ 340/ 1200] Overall Loss 0.263831 Objective Loss 0.263831 LR 0.001000 Time 0.021758 -2022-12-06 10:48:42,444 - Epoch: [59][ 350/ 1200] Overall Loss 0.264356 Objective Loss 0.264356 LR 0.001000 Time 0.021692 -2022-12-06 10:48:42,642 - Epoch: [59][ 360/ 1200] Overall Loss 0.264961 Objective Loss 0.264961 LR 0.001000 Time 0.021638 -2022-12-06 10:48:42,838 - Epoch: [59][ 370/ 1200] Overall Loss 0.265303 Objective Loss 0.265303 LR 0.001000 Time 0.021581 -2022-12-06 10:48:43,035 - Epoch: [59][ 380/ 1200] Overall Loss 0.264638 Objective Loss 0.264638 LR 0.001000 Time 0.021531 -2022-12-06 10:48:43,231 - Epoch: [59][ 390/ 1200] Overall Loss 0.264420 Objective Loss 0.264420 LR 0.001000 Time 0.021479 -2022-12-06 10:48:43,428 - Epoch: [59][ 400/ 1200] Overall Loss 0.264425 Objective Loss 0.264425 LR 0.001000 Time 0.021435 -2022-12-06 10:48:43,624 - Epoch: [59][ 410/ 1200] Overall Loss 0.265060 Objective Loss 0.265060 LR 0.001000 Time 0.021387 -2022-12-06 10:48:43,821 - Epoch: [59][ 420/ 1200] Overall Loss 0.264493 Objective Loss 0.264493 LR 0.001000 Time 0.021347 -2022-12-06 10:48:44,017 - Epoch: [59][ 430/ 1200] Overall Loss 0.264002 Objective Loss 0.264002 LR 0.001000 Time 0.021303 -2022-12-06 10:48:44,214 - Epoch: [59][ 440/ 1200] Overall Loss 0.264259 Objective Loss 0.264259 LR 0.001000 Time 0.021267 -2022-12-06 10:48:44,410 - Epoch: [59][ 450/ 1200] Overall Loss 0.263175 Objective Loss 0.263175 LR 0.001000 Time 0.021227 -2022-12-06 10:48:44,607 - Epoch: [59][ 460/ 1200] Overall Loss 0.263183 Objective Loss 0.263183 LR 0.001000 Time 0.021193 -2022-12-06 10:48:44,802 - Epoch: [59][ 470/ 1200] Overall Loss 0.264035 Objective Loss 0.264035 LR 0.001000 Time 0.021157 -2022-12-06 10:48:45,000 - Epoch: [59][ 480/ 1200] Overall Loss 0.264619 Objective Loss 0.264619 LR 0.001000 Time 0.021126 -2022-12-06 10:48:45,194 - Epoch: [59][ 490/ 1200] Overall Loss 0.265037 Objective Loss 0.265037 LR 0.001000 Time 0.021091 -2022-12-06 10:48:45,392 - Epoch: [59][ 500/ 1200] Overall Loss 0.264564 Objective Loss 0.264564 LR 0.001000 Time 0.021064 -2022-12-06 10:48:45,587 - Epoch: [59][ 510/ 1200] Overall Loss 0.264615 Objective Loss 0.264615 LR 0.001000 Time 0.021032 -2022-12-06 10:48:45,786 - Epoch: [59][ 520/ 1200] Overall Loss 0.263946 Objective Loss 0.263946 LR 0.001000 Time 0.021008 -2022-12-06 10:48:45,981 - Epoch: [59][ 530/ 1200] Overall Loss 0.263551 Objective Loss 0.263551 LR 0.001000 Time 0.020979 -2022-12-06 10:48:46,178 - Epoch: [59][ 540/ 1200] Overall Loss 0.263790 Objective Loss 0.263790 LR 0.001000 Time 0.020956 -2022-12-06 10:48:46,375 - Epoch: [59][ 550/ 1200] Overall Loss 0.264275 Objective Loss 0.264275 LR 0.001000 Time 0.020931 -2022-12-06 10:48:46,572 - Epoch: [59][ 560/ 1200] Overall Loss 0.263801 Objective Loss 0.263801 LR 0.001000 Time 0.020908 -2022-12-06 10:48:46,767 - Epoch: [59][ 570/ 1200] Overall Loss 0.263921 Objective Loss 0.263921 LR 0.001000 Time 0.020883 -2022-12-06 10:48:46,964 - Epoch: [59][ 580/ 1200] Overall Loss 0.264270 Objective Loss 0.264270 LR 0.001000 Time 0.020862 -2022-12-06 10:48:47,160 - Epoch: [59][ 590/ 1200] Overall Loss 0.264314 Objective Loss 0.264314 LR 0.001000 Time 0.020838 -2022-12-06 10:48:47,357 - Epoch: [59][ 600/ 1200] Overall Loss 0.264627 Objective Loss 0.264627 LR 0.001000 Time 0.020820 -2022-12-06 10:48:47,553 - Epoch: [59][ 610/ 1200] Overall Loss 0.264761 Objective Loss 0.264761 LR 0.001000 Time 0.020797 -2022-12-06 10:48:47,751 - Epoch: [59][ 620/ 1200] Overall Loss 0.265570 Objective Loss 0.265570 LR 0.001000 Time 0.020780 -2022-12-06 10:48:47,946 - Epoch: [59][ 630/ 1200] Overall Loss 0.265537 Objective Loss 0.265537 LR 0.001000 Time 0.020760 -2022-12-06 10:48:48,144 - Epoch: [59][ 640/ 1200] Overall Loss 0.265725 Objective Loss 0.265725 LR 0.001000 Time 0.020744 -2022-12-06 10:48:48,339 - Epoch: [59][ 650/ 1200] Overall Loss 0.265954 Objective Loss 0.265954 LR 0.001000 Time 0.020725 -2022-12-06 10:48:48,537 - Epoch: [59][ 660/ 1200] Overall Loss 0.266081 Objective Loss 0.266081 LR 0.001000 Time 0.020709 -2022-12-06 10:48:48,733 - Epoch: [59][ 670/ 1200] Overall Loss 0.265836 Objective Loss 0.265836 LR 0.001000 Time 0.020692 -2022-12-06 10:48:48,931 - Epoch: [59][ 680/ 1200] Overall Loss 0.265592 Objective Loss 0.265592 LR 0.001000 Time 0.020678 -2022-12-06 10:48:49,126 - Epoch: [59][ 690/ 1200] Overall Loss 0.265909 Objective Loss 0.265909 LR 0.001000 Time 0.020660 -2022-12-06 10:48:49,324 - Epoch: [59][ 700/ 1200] Overall Loss 0.265959 Objective Loss 0.265959 LR 0.001000 Time 0.020647 -2022-12-06 10:48:49,519 - Epoch: [59][ 710/ 1200] Overall Loss 0.265829 Objective Loss 0.265829 LR 0.001000 Time 0.020630 -2022-12-06 10:48:49,716 - Epoch: [59][ 720/ 1200] Overall Loss 0.265761 Objective Loss 0.265761 LR 0.001000 Time 0.020617 -2022-12-06 10:48:49,912 - Epoch: [59][ 730/ 1200] Overall Loss 0.265829 Objective Loss 0.265829 LR 0.001000 Time 0.020601 -2022-12-06 10:48:50,109 - Epoch: [59][ 740/ 1200] Overall Loss 0.266125 Objective Loss 0.266125 LR 0.001000 Time 0.020589 -2022-12-06 10:48:50,305 - Epoch: [59][ 750/ 1200] Overall Loss 0.265943 Objective Loss 0.265943 LR 0.001000 Time 0.020575 -2022-12-06 10:48:50,503 - Epoch: [59][ 760/ 1200] Overall Loss 0.266107 Objective Loss 0.266107 LR 0.001000 Time 0.020564 -2022-12-06 10:48:50,699 - Epoch: [59][ 770/ 1200] Overall Loss 0.266289 Objective Loss 0.266289 LR 0.001000 Time 0.020550 -2022-12-06 10:48:50,896 - Epoch: [59][ 780/ 1200] Overall Loss 0.266507 Objective Loss 0.266507 LR 0.001000 Time 0.020539 -2022-12-06 10:48:51,092 - Epoch: [59][ 790/ 1200] Overall Loss 0.266645 Objective Loss 0.266645 LR 0.001000 Time 0.020526 -2022-12-06 10:48:51,289 - Epoch: [59][ 800/ 1200] Overall Loss 0.266829 Objective Loss 0.266829 LR 0.001000 Time 0.020516 -2022-12-06 10:48:51,485 - Epoch: [59][ 810/ 1200] Overall Loss 0.266984 Objective Loss 0.266984 LR 0.001000 Time 0.020503 -2022-12-06 10:48:51,683 - Epoch: [59][ 820/ 1200] Overall Loss 0.267173 Objective Loss 0.267173 LR 0.001000 Time 0.020494 -2022-12-06 10:48:51,879 - Epoch: [59][ 830/ 1200] Overall Loss 0.267547 Objective Loss 0.267547 LR 0.001000 Time 0.020482 -2022-12-06 10:48:52,076 - Epoch: [59][ 840/ 1200] Overall Loss 0.267402 Objective Loss 0.267402 LR 0.001000 Time 0.020472 -2022-12-06 10:48:52,272 - Epoch: [59][ 850/ 1200] Overall Loss 0.267525 Objective Loss 0.267525 LR 0.001000 Time 0.020462 -2022-12-06 10:48:52,470 - Epoch: [59][ 860/ 1200] Overall Loss 0.267565 Objective Loss 0.267565 LR 0.001000 Time 0.020453 -2022-12-06 10:48:52,665 - Epoch: [59][ 870/ 1200] Overall Loss 0.267468 Objective Loss 0.267468 LR 0.001000 Time 0.020442 -2022-12-06 10:48:52,863 - Epoch: [59][ 880/ 1200] Overall Loss 0.267405 Objective Loss 0.267405 LR 0.001000 Time 0.020434 -2022-12-06 10:48:53,058 - Epoch: [59][ 890/ 1200] Overall Loss 0.267280 Objective Loss 0.267280 LR 0.001000 Time 0.020423 -2022-12-06 10:48:53,257 - Epoch: [59][ 900/ 1200] Overall Loss 0.267494 Objective Loss 0.267494 LR 0.001000 Time 0.020416 -2022-12-06 10:48:53,453 - Epoch: [59][ 910/ 1200] Overall Loss 0.267953 Objective Loss 0.267953 LR 0.001000 Time 0.020406 -2022-12-06 10:48:53,650 - Epoch: [59][ 920/ 1200] Overall Loss 0.268179 Objective Loss 0.268179 LR 0.001000 Time 0.020398 -2022-12-06 10:48:53,846 - Epoch: [59][ 930/ 1200] Overall Loss 0.268509 Objective Loss 0.268509 LR 0.001000 Time 0.020389 -2022-12-06 10:48:54,043 - Epoch: [59][ 940/ 1200] Overall Loss 0.269019 Objective Loss 0.269019 LR 0.001000 Time 0.020382 -2022-12-06 10:48:54,239 - Epoch: [59][ 950/ 1200] Overall Loss 0.269039 Objective Loss 0.269039 LR 0.001000 Time 0.020372 -2022-12-06 10:48:54,437 - Epoch: [59][ 960/ 1200] Overall Loss 0.269522 Objective Loss 0.269522 LR 0.001000 Time 0.020366 -2022-12-06 10:48:54,633 - Epoch: [59][ 970/ 1200] Overall Loss 0.270014 Objective Loss 0.270014 LR 0.001000 Time 0.020357 -2022-12-06 10:48:54,831 - Epoch: [59][ 980/ 1200] Overall Loss 0.270193 Objective Loss 0.270193 LR 0.001000 Time 0.020351 -2022-12-06 10:48:55,026 - Epoch: [59][ 990/ 1200] Overall Loss 0.270175 Objective Loss 0.270175 LR 0.001000 Time 0.020342 -2022-12-06 10:48:55,223 - Epoch: [59][ 1000/ 1200] Overall Loss 0.270187 Objective Loss 0.270187 LR 0.001000 Time 0.020335 -2022-12-06 10:48:55,418 - Epoch: [59][ 1010/ 1200] Overall Loss 0.270117 Objective Loss 0.270117 LR 0.001000 Time 0.020327 -2022-12-06 10:48:55,616 - Epoch: [59][ 1020/ 1200] Overall Loss 0.270228 Objective Loss 0.270228 LR 0.001000 Time 0.020321 -2022-12-06 10:48:55,812 - Epoch: [59][ 1030/ 1200] Overall Loss 0.270327 Objective Loss 0.270327 LR 0.001000 Time 0.020313 -2022-12-06 10:48:56,009 - Epoch: [59][ 1040/ 1200] Overall Loss 0.270155 Objective Loss 0.270155 LR 0.001000 Time 0.020307 -2022-12-06 10:48:56,205 - Epoch: [59][ 1050/ 1200] Overall Loss 0.270363 Objective Loss 0.270363 LR 0.001000 Time 0.020300 -2022-12-06 10:48:56,403 - Epoch: [59][ 1060/ 1200] Overall Loss 0.270932 Objective Loss 0.270932 LR 0.001000 Time 0.020294 -2022-12-06 10:48:56,598 - Epoch: [59][ 1070/ 1200] Overall Loss 0.270928 Objective Loss 0.270928 LR 0.001000 Time 0.020287 -2022-12-06 10:48:56,796 - Epoch: [59][ 1080/ 1200] Overall Loss 0.270986 Objective Loss 0.270986 LR 0.001000 Time 0.020282 -2022-12-06 10:48:56,992 - Epoch: [59][ 1090/ 1200] Overall Loss 0.270786 Objective Loss 0.270786 LR 0.001000 Time 0.020275 -2022-12-06 10:48:57,190 - Epoch: [59][ 1100/ 1200] Overall Loss 0.270750 Objective Loss 0.270750 LR 0.001000 Time 0.020269 -2022-12-06 10:48:57,385 - Epoch: [59][ 1110/ 1200] Overall Loss 0.270675 Objective Loss 0.270675 LR 0.001000 Time 0.020262 -2022-12-06 10:48:57,582 - Epoch: [59][ 1120/ 1200] Overall Loss 0.270606 Objective Loss 0.270606 LR 0.001000 Time 0.020257 -2022-12-06 10:48:57,778 - Epoch: [59][ 1130/ 1200] Overall Loss 0.270510 Objective Loss 0.270510 LR 0.001000 Time 0.020250 -2022-12-06 10:48:57,976 - Epoch: [59][ 1140/ 1200] Overall Loss 0.270491 Objective Loss 0.270491 LR 0.001000 Time 0.020246 -2022-12-06 10:48:58,171 - Epoch: [59][ 1150/ 1200] Overall Loss 0.270422 Objective Loss 0.270422 LR 0.001000 Time 0.020239 -2022-12-06 10:48:58,370 - Epoch: [59][ 1160/ 1200] Overall Loss 0.270578 Objective Loss 0.270578 LR 0.001000 Time 0.020235 -2022-12-06 10:48:58,566 - Epoch: [59][ 1170/ 1200] Overall Loss 0.270657 Objective Loss 0.270657 LR 0.001000 Time 0.020230 -2022-12-06 10:48:58,766 - Epoch: [59][ 1180/ 1200] Overall Loss 0.270279 Objective Loss 0.270279 LR 0.001000 Time 0.020228 -2022-12-06 10:48:58,963 - Epoch: [59][ 1190/ 1200] Overall Loss 0.270101 Objective Loss 0.270101 LR 0.001000 Time 0.020222 -2022-12-06 10:48:59,200 - Epoch: [59][ 1200/ 1200] Overall Loss 0.269984 Objective Loss 0.269984 Top1 84.518828 Top5 98.326360 LR 0.001000 Time 0.020251 -2022-12-06 10:48:59,301 - --- validate (epoch=59)----------- -2022-12-06 10:48:59,302 - 34129 samples (256 per mini-batch) -2022-12-06 10:48:59,756 - Epoch: [59][ 10/ 134] Loss 0.275797 Top1 83.007812 Top5 97.695312 -2022-12-06 10:48:59,887 - Epoch: [59][ 20/ 134] Loss 0.286555 Top1 83.320312 Top5 97.363281 -2022-12-06 10:49:00,017 - Epoch: [59][ 30/ 134] Loss 0.288918 Top1 83.684896 Top5 97.460938 -2022-12-06 10:49:00,144 - Epoch: [59][ 40/ 134] Loss 0.297865 Top1 83.671875 Top5 97.431641 -2022-12-06 10:49:00,274 - Epoch: [59][ 50/ 134] Loss 0.298572 Top1 83.648438 Top5 97.507812 -2022-12-06 10:49:00,401 - Epoch: [59][ 60/ 134] Loss 0.302150 Top1 83.489583 Top5 97.526042 -2022-12-06 10:49:00,527 - Epoch: [59][ 70/ 134] Loss 0.303243 Top1 83.504464 Top5 97.539062 -2022-12-06 10:49:00,652 - Epoch: [59][ 80/ 134] Loss 0.306904 Top1 83.525391 Top5 97.500000 -2022-12-06 10:49:00,779 - Epoch: [59][ 90/ 134] Loss 0.304445 Top1 83.528646 Top5 97.513021 -2022-12-06 10:49:00,906 - Epoch: [59][ 100/ 134] Loss 0.303908 Top1 83.523438 Top5 97.546875 -2022-12-06 10:49:01,031 - Epoch: [59][ 110/ 134] Loss 0.304445 Top1 83.394886 Top5 97.553267 -2022-12-06 10:49:01,158 - Epoch: [59][ 120/ 134] Loss 0.304458 Top1 83.365885 Top5 97.600911 -2022-12-06 10:49:01,285 - Epoch: [59][ 130/ 134] Loss 0.303746 Top1 83.350361 Top5 97.647236 -2022-12-06 10:49:01,322 - Epoch: [59][ 134/ 134] Loss 0.301554 Top1 83.412933 Top5 97.667673 -2022-12-06 10:49:01,412 - ==> Top1: 83.413 Top5: 97.668 Loss: 0.302 - -2022-12-06 10:49:01,413 - ==> Confusion: -[[ 904 2 4 1 6 4 0 0 6 51 0 0 2 3 5 1 1 2 3 0 1] - [ 2 907 0 3 9 20 6 11 2 0 7 5 3 4 4 2 3 3 19 5 12] - [ 6 1 988 20 3 1 37 5 0 3 2 3 6 0 4 3 1 3 8 4 5] - [ 3 1 15 923 0 2 1 2 1 0 10 0 12 6 20 1 0 6 14 0 3] - [ 14 4 4 1 934 4 1 0 0 6 2 3 1 3 12 9 10 3 1 2 6] - [ 5 16 1 2 5 951 6 28 2 3 1 12 9 5 3 2 0 2 4 4 8] - [ 1 0 8 1 0 2 1075 5 1 0 2 2 0 1 0 9 1 1 2 7 0] - [ 0 12 2 3 1 25 12 922 0 2 5 9 1 1 1 2 1 0 38 14 3] - [ 8 0 0 3 0 2 1 0 953 46 7 2 2 9 17 1 2 3 4 1 3] - [ 73 1 3 0 2 5 0 3 20 867 2 1 0 12 3 0 0 1 1 0 7] - [ 2 1 3 11 1 4 4 2 11 0 940 0 7 12 5 0 2 1 9 1 3] - [ 3 0 3 1 0 4 2 8 3 0 0 937 56 4 0 7 1 10 2 10 0] - [ 2 2 1 2 1 0 0 0 0 0 0 23 908 4 0 4 1 10 1 3 7] - [ 0 1 1 0 0 11 1 3 10 20 10 7 6 932 4 1 1 6 2 0 7] - [ 8 2 2 8 1 2 1 1 13 4 0 1 5 2 1068 1 1 1 6 1 2] - [ 2 1 3 0 2 1 5 0 0 0 0 2 12 4 0 993 3 9 0 4 2] - [ 4 1 0 0 2 1 2 0 1 0 1 6 9 2 2 20 1011 2 0 4 4] - [ 5 0 1 3 0 1 2 0 1 1 0 14 31 4 2 13 1 955 1 1 0] - [ 4 2 4 16 1 3 0 13 0 0 6 1 2 2 9 0 2 0 940 3 0] - [ 2 3 2 1 0 5 10 7 0 0 1 19 7 6 1 5 2 4 2 998 5] - [ 153 215 230 164 97 175 120 155 89 97 220 121 549 274 216 186 151 111 281 260 9362]] - -2022-12-06 10:49:02,082 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:49:02,082 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:49:02,088 - - -2022-12-06 10:49:02,088 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:49:03,023 - Epoch: [60][ 10/ 1200] Overall Loss 0.245421 Objective Loss 0.245421 LR 0.001000 Time 0.093422 -2022-12-06 10:49:03,226 - Epoch: [60][ 20/ 1200] Overall Loss 0.260078 Objective Loss 0.260078 LR 0.001000 Time 0.056818 -2022-12-06 10:49:03,424 - Epoch: [60][ 30/ 1200] Overall Loss 0.260592 Objective Loss 0.260592 LR 0.001000 Time 0.044469 -2022-12-06 10:49:03,619 - Epoch: [60][ 40/ 1200] Overall Loss 0.262638 Objective Loss 0.262638 LR 0.001000 Time 0.038224 -2022-12-06 10:49:03,818 - Epoch: [60][ 50/ 1200] Overall Loss 0.258479 Objective Loss 0.258479 LR 0.001000 Time 0.034538 -2022-12-06 10:49:04,013 - Epoch: [60][ 60/ 1200] Overall Loss 0.263400 Objective Loss 0.263400 LR 0.001000 Time 0.032021 -2022-12-06 10:49:04,211 - Epoch: [60][ 70/ 1200] Overall Loss 0.264839 Objective Loss 0.264839 LR 0.001000 Time 0.030271 -2022-12-06 10:49:04,406 - Epoch: [60][ 80/ 1200] Overall Loss 0.264510 Objective Loss 0.264510 LR 0.001000 Time 0.028915 -2022-12-06 10:49:04,604 - Epoch: [60][ 90/ 1200] Overall Loss 0.265573 Objective Loss 0.265573 LR 0.001000 Time 0.027899 -2022-12-06 10:49:04,799 - Epoch: [60][ 100/ 1200] Overall Loss 0.263753 Objective Loss 0.263753 LR 0.001000 Time 0.027051 -2022-12-06 10:49:04,996 - Epoch: [60][ 110/ 1200] Overall Loss 0.265833 Objective Loss 0.265833 LR 0.001000 Time 0.026382 -2022-12-06 10:49:05,191 - Epoch: [60][ 120/ 1200] Overall Loss 0.267700 Objective Loss 0.267700 LR 0.001000 Time 0.025801 -2022-12-06 10:49:05,389 - Epoch: [60][ 130/ 1200] Overall Loss 0.268477 Objective Loss 0.268477 LR 0.001000 Time 0.025335 -2022-12-06 10:49:05,583 - Epoch: [60][ 140/ 1200] Overall Loss 0.267748 Objective Loss 0.267748 LR 0.001000 Time 0.024910 -2022-12-06 10:49:05,781 - Epoch: [60][ 150/ 1200] Overall Loss 0.267863 Objective Loss 0.267863 LR 0.001000 Time 0.024566 -2022-12-06 10:49:05,976 - Epoch: [60][ 160/ 1200] Overall Loss 0.266629 Objective Loss 0.266629 LR 0.001000 Time 0.024243 -2022-12-06 10:49:06,174 - Epoch: [60][ 170/ 1200] Overall Loss 0.266039 Objective Loss 0.266039 LR 0.001000 Time 0.023978 -2022-12-06 10:49:06,368 - Epoch: [60][ 180/ 1200] Overall Loss 0.266196 Objective Loss 0.266196 LR 0.001000 Time 0.023725 -2022-12-06 10:49:06,567 - Epoch: [60][ 190/ 1200] Overall Loss 0.266022 Objective Loss 0.266022 LR 0.001000 Time 0.023517 -2022-12-06 10:49:06,761 - Epoch: [60][ 200/ 1200] Overall Loss 0.265370 Objective Loss 0.265370 LR 0.001000 Time 0.023313 -2022-12-06 10:49:06,959 - Epoch: [60][ 210/ 1200] Overall Loss 0.264699 Objective Loss 0.264699 LR 0.001000 Time 0.023139 -2022-12-06 10:49:07,154 - Epoch: [60][ 220/ 1200] Overall Loss 0.264828 Objective Loss 0.264828 LR 0.001000 Time 0.022972 -2022-12-06 10:49:07,352 - Epoch: [60][ 230/ 1200] Overall Loss 0.264485 Objective Loss 0.264485 LR 0.001000 Time 0.022832 -2022-12-06 10:49:07,547 - Epoch: [60][ 240/ 1200] Overall Loss 0.265141 Objective Loss 0.265141 LR 0.001000 Time 0.022691 -2022-12-06 10:49:07,745 - Epoch: [60][ 250/ 1200] Overall Loss 0.266184 Objective Loss 0.266184 LR 0.001000 Time 0.022575 -2022-12-06 10:49:07,940 - Epoch: [60][ 260/ 1200] Overall Loss 0.265690 Objective Loss 0.265690 LR 0.001000 Time 0.022452 -2022-12-06 10:49:08,138 - Epoch: [60][ 270/ 1200] Overall Loss 0.266461 Objective Loss 0.266461 LR 0.001000 Time 0.022351 -2022-12-06 10:49:08,333 - Epoch: [60][ 280/ 1200] Overall Loss 0.266293 Objective Loss 0.266293 LR 0.001000 Time 0.022248 -2022-12-06 10:49:08,530 - Epoch: [60][ 290/ 1200] Overall Loss 0.266621 Objective Loss 0.266621 LR 0.001000 Time 0.022160 -2022-12-06 10:49:08,725 - Epoch: [60][ 300/ 1200] Overall Loss 0.267636 Objective Loss 0.267636 LR 0.001000 Time 0.022070 -2022-12-06 10:49:08,923 - Epoch: [60][ 310/ 1200] Overall Loss 0.267763 Objective Loss 0.267763 LR 0.001000 Time 0.021995 -2022-12-06 10:49:09,119 - Epoch: [60][ 320/ 1200] Overall Loss 0.267987 Objective Loss 0.267987 LR 0.001000 Time 0.021916 -2022-12-06 10:49:09,317 - Epoch: [60][ 330/ 1200] Overall Loss 0.268720 Objective Loss 0.268720 LR 0.001000 Time 0.021851 -2022-12-06 10:49:09,511 - Epoch: [60][ 340/ 1200] Overall Loss 0.268387 Objective Loss 0.268387 LR 0.001000 Time 0.021778 -2022-12-06 10:49:09,709 - Epoch: [60][ 350/ 1200] Overall Loss 0.268524 Objective Loss 0.268524 LR 0.001000 Time 0.021720 -2022-12-06 10:49:09,904 - Epoch: [60][ 360/ 1200] Overall Loss 0.267800 Objective Loss 0.267800 LR 0.001000 Time 0.021656 -2022-12-06 10:49:10,101 - Epoch: [60][ 370/ 1200] Overall Loss 0.267827 Objective Loss 0.267827 LR 0.001000 Time 0.021602 -2022-12-06 10:49:10,296 - Epoch: [60][ 380/ 1200] Overall Loss 0.267659 Objective Loss 0.267659 LR 0.001000 Time 0.021546 -2022-12-06 10:49:10,494 - Epoch: [60][ 390/ 1200] Overall Loss 0.268104 Objective Loss 0.268104 LR 0.001000 Time 0.021500 -2022-12-06 10:49:10,689 - Epoch: [60][ 400/ 1200] Overall Loss 0.267866 Objective Loss 0.267866 LR 0.001000 Time 0.021449 -2022-12-06 10:49:10,888 - Epoch: [60][ 410/ 1200] Overall Loss 0.268234 Objective Loss 0.268234 LR 0.001000 Time 0.021407 -2022-12-06 10:49:11,082 - Epoch: [60][ 420/ 1200] Overall Loss 0.268133 Objective Loss 0.268133 LR 0.001000 Time 0.021361 -2022-12-06 10:49:11,281 - Epoch: [60][ 430/ 1200] Overall Loss 0.268435 Objective Loss 0.268435 LR 0.001000 Time 0.021324 -2022-12-06 10:49:11,476 - Epoch: [60][ 440/ 1200] Overall Loss 0.268347 Objective Loss 0.268347 LR 0.001000 Time 0.021281 -2022-12-06 10:49:11,674 - Epoch: [60][ 450/ 1200] Overall Loss 0.267832 Objective Loss 0.267832 LR 0.001000 Time 0.021247 -2022-12-06 10:49:11,869 - Epoch: [60][ 460/ 1200] Overall Loss 0.268126 Objective Loss 0.268126 LR 0.001000 Time 0.021208 -2022-12-06 10:49:12,066 - Epoch: [60][ 470/ 1200] Overall Loss 0.268424 Objective Loss 0.268424 LR 0.001000 Time 0.021176 -2022-12-06 10:49:12,262 - Epoch: [60][ 480/ 1200] Overall Loss 0.268767 Objective Loss 0.268767 LR 0.001000 Time 0.021141 -2022-12-06 10:49:12,460 - Epoch: [60][ 490/ 1200] Overall Loss 0.268703 Objective Loss 0.268703 LR 0.001000 Time 0.021113 -2022-12-06 10:49:12,655 - Epoch: [60][ 500/ 1200] Overall Loss 0.268976 Objective Loss 0.268976 LR 0.001000 Time 0.021079 -2022-12-06 10:49:12,852 - Epoch: [60][ 510/ 1200] Overall Loss 0.268470 Objective Loss 0.268470 LR 0.001000 Time 0.021051 -2022-12-06 10:49:13,047 - Epoch: [60][ 520/ 1200] Overall Loss 0.268254 Objective Loss 0.268254 LR 0.001000 Time 0.021020 -2022-12-06 10:49:13,245 - Epoch: [60][ 530/ 1200] Overall Loss 0.267857 Objective Loss 0.267857 LR 0.001000 Time 0.020996 -2022-12-06 10:49:13,439 - Epoch: [60][ 540/ 1200] Overall Loss 0.267567 Objective Loss 0.267567 LR 0.001000 Time 0.020966 -2022-12-06 10:49:13,638 - Epoch: [60][ 550/ 1200] Overall Loss 0.267119 Objective Loss 0.267119 LR 0.001000 Time 0.020945 -2022-12-06 10:49:13,832 - Epoch: [60][ 560/ 1200] Overall Loss 0.267260 Objective Loss 0.267260 LR 0.001000 Time 0.020917 -2022-12-06 10:49:14,030 - Epoch: [60][ 570/ 1200] Overall Loss 0.267745 Objective Loss 0.267745 LR 0.001000 Time 0.020896 -2022-12-06 10:49:14,224 - Epoch: [60][ 580/ 1200] Overall Loss 0.267615 Objective Loss 0.267615 LR 0.001000 Time 0.020870 -2022-12-06 10:49:14,422 - Epoch: [60][ 590/ 1200] Overall Loss 0.267965 Objective Loss 0.267965 LR 0.001000 Time 0.020851 -2022-12-06 10:49:14,617 - Epoch: [60][ 600/ 1200] Overall Loss 0.268046 Objective Loss 0.268046 LR 0.001000 Time 0.020827 -2022-12-06 10:49:14,815 - Epoch: [60][ 610/ 1200] Overall Loss 0.268427 Objective Loss 0.268427 LR 0.001000 Time 0.020810 -2022-12-06 10:49:15,010 - Epoch: [60][ 620/ 1200] Overall Loss 0.268275 Objective Loss 0.268275 LR 0.001000 Time 0.020787 -2022-12-06 10:49:15,209 - Epoch: [60][ 630/ 1200] Overall Loss 0.268108 Objective Loss 0.268108 LR 0.001000 Time 0.020772 -2022-12-06 10:49:15,403 - Epoch: [60][ 640/ 1200] Overall Loss 0.268049 Objective Loss 0.268049 LR 0.001000 Time 0.020750 -2022-12-06 10:49:15,601 - Epoch: [60][ 650/ 1200] Overall Loss 0.268670 Objective Loss 0.268670 LR 0.001000 Time 0.020734 -2022-12-06 10:49:15,796 - Epoch: [60][ 660/ 1200] Overall Loss 0.268791 Objective Loss 0.268791 LR 0.001000 Time 0.020715 -2022-12-06 10:49:15,994 - Epoch: [60][ 670/ 1200] Overall Loss 0.268848 Objective Loss 0.268848 LR 0.001000 Time 0.020700 -2022-12-06 10:49:16,189 - Epoch: [60][ 680/ 1200] Overall Loss 0.268846 Objective Loss 0.268846 LR 0.001000 Time 0.020682 -2022-12-06 10:49:16,387 - Epoch: [60][ 690/ 1200] Overall Loss 0.268236 Objective Loss 0.268236 LR 0.001000 Time 0.020669 -2022-12-06 10:49:16,582 - Epoch: [60][ 700/ 1200] Overall Loss 0.268068 Objective Loss 0.268068 LR 0.001000 Time 0.020651 -2022-12-06 10:49:16,779 - Epoch: [60][ 710/ 1200] Overall Loss 0.268009 Objective Loss 0.268009 LR 0.001000 Time 0.020637 -2022-12-06 10:49:16,974 - Epoch: [60][ 720/ 1200] Overall Loss 0.268254 Objective Loss 0.268254 LR 0.001000 Time 0.020620 -2022-12-06 10:49:17,172 - Epoch: [60][ 730/ 1200] Overall Loss 0.268400 Objective Loss 0.268400 LR 0.001000 Time 0.020608 -2022-12-06 10:49:17,367 - Epoch: [60][ 740/ 1200] Overall Loss 0.268298 Objective Loss 0.268298 LR 0.001000 Time 0.020593 -2022-12-06 10:49:17,565 - Epoch: [60][ 750/ 1200] Overall Loss 0.268149 Objective Loss 0.268149 LR 0.001000 Time 0.020582 -2022-12-06 10:49:17,760 - Epoch: [60][ 760/ 1200] Overall Loss 0.268516 Objective Loss 0.268516 LR 0.001000 Time 0.020567 -2022-12-06 10:49:17,958 - Epoch: [60][ 770/ 1200] Overall Loss 0.268435 Objective Loss 0.268435 LR 0.001000 Time 0.020556 -2022-12-06 10:49:18,153 - Epoch: [60][ 780/ 1200] Overall Loss 0.268423 Objective Loss 0.268423 LR 0.001000 Time 0.020542 -2022-12-06 10:49:18,351 - Epoch: [60][ 790/ 1200] Overall Loss 0.268227 Objective Loss 0.268227 LR 0.001000 Time 0.020531 -2022-12-06 10:49:18,545 - Epoch: [60][ 800/ 1200] Overall Loss 0.268076 Objective Loss 0.268076 LR 0.001000 Time 0.020517 -2022-12-06 10:49:18,743 - Epoch: [60][ 810/ 1200] Overall Loss 0.267992 Objective Loss 0.267992 LR 0.001000 Time 0.020507 -2022-12-06 10:49:18,938 - Epoch: [60][ 820/ 1200] Overall Loss 0.268161 Objective Loss 0.268161 LR 0.001000 Time 0.020494 -2022-12-06 10:49:19,136 - Epoch: [60][ 830/ 1200] Overall Loss 0.268152 Objective Loss 0.268152 LR 0.001000 Time 0.020485 -2022-12-06 10:49:19,331 - Epoch: [60][ 840/ 1200] Overall Loss 0.267937 Objective Loss 0.267937 LR 0.001000 Time 0.020473 -2022-12-06 10:49:19,529 - Epoch: [60][ 850/ 1200] Overall Loss 0.267893 Objective Loss 0.267893 LR 0.001000 Time 0.020464 -2022-12-06 10:49:19,724 - Epoch: [60][ 860/ 1200] Overall Loss 0.267901 Objective Loss 0.267901 LR 0.001000 Time 0.020452 -2022-12-06 10:49:19,922 - Epoch: [60][ 870/ 1200] Overall Loss 0.267968 Objective Loss 0.267968 LR 0.001000 Time 0.020444 -2022-12-06 10:49:20,116 - Epoch: [60][ 880/ 1200] Overall Loss 0.268277 Objective Loss 0.268277 LR 0.001000 Time 0.020432 -2022-12-06 10:49:20,314 - Epoch: [60][ 890/ 1200] Overall Loss 0.268206 Objective Loss 0.268206 LR 0.001000 Time 0.020424 -2022-12-06 10:49:20,509 - Epoch: [60][ 900/ 1200] Overall Loss 0.268456 Objective Loss 0.268456 LR 0.001000 Time 0.020413 -2022-12-06 10:49:20,707 - Epoch: [60][ 910/ 1200] Overall Loss 0.268369 Objective Loss 0.268369 LR 0.001000 Time 0.020406 -2022-12-06 10:49:20,901 - Epoch: [60][ 920/ 1200] Overall Loss 0.268466 Objective Loss 0.268466 LR 0.001000 Time 0.020395 -2022-12-06 10:49:21,099 - Epoch: [60][ 930/ 1200] Overall Loss 0.268184 Objective Loss 0.268184 LR 0.001000 Time 0.020388 -2022-12-06 10:49:21,294 - Epoch: [60][ 940/ 1200] Overall Loss 0.268317 Objective Loss 0.268317 LR 0.001000 Time 0.020378 -2022-12-06 10:49:21,492 - Epoch: [60][ 950/ 1200] Overall Loss 0.268283 Objective Loss 0.268283 LR 0.001000 Time 0.020371 -2022-12-06 10:49:21,687 - Epoch: [60][ 960/ 1200] Overall Loss 0.268263 Objective Loss 0.268263 LR 0.001000 Time 0.020361 -2022-12-06 10:49:21,884 - Epoch: [60][ 970/ 1200] Overall Loss 0.268539 Objective Loss 0.268539 LR 0.001000 Time 0.020354 -2022-12-06 10:49:22,079 - Epoch: [60][ 980/ 1200] Overall Loss 0.268650 Objective Loss 0.268650 LR 0.001000 Time 0.020344 -2022-12-06 10:49:22,276 - Epoch: [60][ 990/ 1200] Overall Loss 0.268726 Objective Loss 0.268726 LR 0.001000 Time 0.020338 -2022-12-06 10:49:22,471 - Epoch: [60][ 1000/ 1200] Overall Loss 0.268744 Objective Loss 0.268744 LR 0.001000 Time 0.020328 -2022-12-06 10:49:22,669 - Epoch: [60][ 1010/ 1200] Overall Loss 0.268974 Objective Loss 0.268974 LR 0.001000 Time 0.020323 -2022-12-06 10:49:22,864 - Epoch: [60][ 1020/ 1200] Overall Loss 0.269368 Objective Loss 0.269368 LR 0.001000 Time 0.020314 -2022-12-06 10:49:23,062 - Epoch: [60][ 1030/ 1200] Overall Loss 0.269678 Objective Loss 0.269678 LR 0.001000 Time 0.020308 -2022-12-06 10:49:23,256 - Epoch: [60][ 1040/ 1200] Overall Loss 0.269693 Objective Loss 0.269693 LR 0.001000 Time 0.020299 -2022-12-06 10:49:23,453 - Epoch: [60][ 1050/ 1200] Overall Loss 0.269498 Objective Loss 0.269498 LR 0.001000 Time 0.020293 -2022-12-06 10:49:23,648 - Epoch: [60][ 1060/ 1200] Overall Loss 0.269675 Objective Loss 0.269675 LR 0.001000 Time 0.020285 -2022-12-06 10:49:23,845 - Epoch: [60][ 1070/ 1200] Overall Loss 0.269780 Objective Loss 0.269780 LR 0.001000 Time 0.020279 -2022-12-06 10:49:24,040 - Epoch: [60][ 1080/ 1200] Overall Loss 0.270042 Objective Loss 0.270042 LR 0.001000 Time 0.020271 -2022-12-06 10:49:24,238 - Epoch: [60][ 1090/ 1200] Overall Loss 0.269834 Objective Loss 0.269834 LR 0.001000 Time 0.020267 -2022-12-06 10:49:24,433 - Epoch: [60][ 1100/ 1200] Overall Loss 0.269922 Objective Loss 0.269922 LR 0.001000 Time 0.020259 -2022-12-06 10:49:24,631 - Epoch: [60][ 1110/ 1200] Overall Loss 0.269982 Objective Loss 0.269982 LR 0.001000 Time 0.020254 -2022-12-06 10:49:24,825 - Epoch: [60][ 1120/ 1200] Overall Loss 0.270129 Objective Loss 0.270129 LR 0.001000 Time 0.020247 -2022-12-06 10:49:25,024 - Epoch: [60][ 1130/ 1200] Overall Loss 0.270280 Objective Loss 0.270280 LR 0.001000 Time 0.020243 -2022-12-06 10:49:25,219 - Epoch: [60][ 1140/ 1200] Overall Loss 0.270269 Objective Loss 0.270269 LR 0.001000 Time 0.020236 -2022-12-06 10:49:25,417 - Epoch: [60][ 1150/ 1200] Overall Loss 0.270339 Objective Loss 0.270339 LR 0.001000 Time 0.020231 -2022-12-06 10:49:25,612 - Epoch: [60][ 1160/ 1200] Overall Loss 0.270234 Objective Loss 0.270234 LR 0.001000 Time 0.020225 -2022-12-06 10:49:25,810 - Epoch: [60][ 1170/ 1200] Overall Loss 0.270253 Objective Loss 0.270253 LR 0.001000 Time 0.020220 -2022-12-06 10:49:26,005 - Epoch: [60][ 1180/ 1200] Overall Loss 0.270404 Objective Loss 0.270404 LR 0.001000 Time 0.020214 -2022-12-06 10:49:26,203 - Epoch: [60][ 1190/ 1200] Overall Loss 0.270507 Objective Loss 0.270507 LR 0.001000 Time 0.020210 -2022-12-06 10:49:26,434 - Epoch: [60][ 1200/ 1200] Overall Loss 0.270377 Objective Loss 0.270377 Top1 87.238494 Top5 98.953975 LR 0.001000 Time 0.020234 -2022-12-06 10:49:26,524 - --- validate (epoch=60)----------- -2022-12-06 10:49:26,524 - 34129 samples (256 per mini-batch) -2022-12-06 10:49:26,968 - Epoch: [60][ 10/ 134] Loss 0.321978 Top1 84.062500 Top5 98.203125 -2022-12-06 10:49:27,101 - Epoch: [60][ 20/ 134] Loss 0.313541 Top1 83.789062 Top5 98.066406 -2022-12-06 10:49:27,233 - Epoch: [60][ 30/ 134] Loss 0.301796 Top1 84.661458 Top5 98.059896 -2022-12-06 10:49:27,365 - Epoch: [60][ 40/ 134] Loss 0.301341 Top1 84.804688 Top5 98.115234 -2022-12-06 10:49:27,497 - Epoch: [60][ 50/ 134] Loss 0.296054 Top1 84.984375 Top5 98.101562 -2022-12-06 10:49:27,627 - Epoch: [60][ 60/ 134] Loss 0.297420 Top1 85.136719 Top5 98.125000 -2022-12-06 10:49:27,762 - Epoch: [60][ 70/ 134] Loss 0.302256 Top1 84.938616 Top5 98.102679 -2022-12-06 10:49:27,896 - Epoch: [60][ 80/ 134] Loss 0.303618 Top1 84.902344 Top5 98.041992 -2022-12-06 10:49:28,029 - Epoch: [60][ 90/ 134] Loss 0.304232 Top1 84.930556 Top5 98.059896 -2022-12-06 10:49:28,163 - Epoch: [60][ 100/ 134] Loss 0.301876 Top1 85.097656 Top5 98.082031 -2022-12-06 10:49:28,297 - Epoch: [60][ 110/ 134] Loss 0.301221 Top1 85.127841 Top5 98.085938 -2022-12-06 10:49:28,432 - Epoch: [60][ 120/ 134] Loss 0.299232 Top1 85.133464 Top5 98.082682 -2022-12-06 10:49:28,566 - Epoch: [60][ 130/ 134] Loss 0.296939 Top1 85.153245 Top5 98.073918 -2022-12-06 10:49:28,605 - Epoch: [60][ 134/ 134] Loss 0.298368 Top1 85.097717 Top5 98.072021 -2022-12-06 10:49:28,692 - ==> Top1: 85.098 Top5: 98.072 Loss: 0.298 - -2022-12-06 10:49:28,693 - ==> Confusion: -[[ 883 0 3 5 3 5 0 0 7 70 0 2 3 2 4 2 2 1 0 0 4] - [ 1 934 2 3 12 26 2 6 2 1 0 4 4 3 1 0 7 3 7 2 7] - [ 9 1 986 12 4 0 25 8 0 2 1 10 2 3 4 2 0 1 6 4 23] - [ 3 1 18 935 1 2 2 0 0 0 6 2 3 3 17 0 3 5 11 1 7] - [ 14 5 2 1 948 6 0 0 1 6 1 5 3 2 11 4 8 0 0 3 0] - [ 2 25 0 1 5 959 2 14 3 1 1 16 6 9 3 0 3 2 1 7 9] - [ 0 4 13 2 0 3 1058 6 0 0 0 3 2 1 0 8 0 3 0 12 3] - [ 2 29 7 6 1 38 6 893 1 1 0 7 1 3 3 1 2 0 35 11 7] - [ 4 3 1 2 0 4 0 0 973 42 3 3 1 7 11 2 0 1 1 1 5] - [ 45 0 0 1 1 3 1 3 27 901 1 1 0 9 4 0 0 1 0 0 3] - [ 0 3 5 9 4 4 1 2 17 2 923 4 0 19 3 0 1 1 14 1 6] - [ 5 1 1 1 0 6 2 2 1 0 0 973 24 6 1 7 1 6 0 11 3] - [ 2 1 0 4 0 2 1 0 1 0 0 47 876 0 2 9 0 10 2 4 8] - [ 0 0 1 0 1 17 0 1 16 16 10 9 4 925 2 3 4 0 2 3 9] - [ 6 3 2 14 2 2 0 0 20 1 0 2 1 5 1061 0 2 2 2 0 5] - [ 0 0 1 2 4 1 6 0 0 0 0 13 4 5 0 978 9 13 0 3 4] - [ 2 4 1 2 3 1 0 0 0 0 0 5 4 3 5 13 1011 3 0 6 9] - [ 5 0 1 2 0 0 0 2 1 2 0 8 22 1 1 12 1 973 2 2 1] - [ 3 7 3 8 1 1 0 18 3 1 1 6 3 3 10 0 0 0 930 5 5] - [ 2 2 0 2 1 13 5 9 1 0 0 23 5 4 1 4 3 2 1 997 5] - [ 106 274 137 149 100 194 60 131 115 113 115 140 366 268 194 128 196 100 186 236 9918]] - -2022-12-06 10:49:29,354 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:49:29,354 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:49:29,360 - - -2022-12-06 10:49:29,360 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:49:30,298 - Epoch: [61][ 10/ 1200] Overall Loss 0.285862 Objective Loss 0.285862 LR 0.001000 Time 0.093696 -2022-12-06 10:49:30,498 - Epoch: [61][ 20/ 1200] Overall Loss 0.272740 Objective Loss 0.272740 LR 0.001000 Time 0.056847 -2022-12-06 10:49:30,690 - Epoch: [61][ 30/ 1200] Overall Loss 0.260106 Objective Loss 0.260106 LR 0.001000 Time 0.044248 -2022-12-06 10:49:30,880 - Epoch: [61][ 40/ 1200] Overall Loss 0.259916 Objective Loss 0.259916 LR 0.001000 Time 0.037943 -2022-12-06 10:49:31,070 - Epoch: [61][ 50/ 1200] Overall Loss 0.263545 Objective Loss 0.263545 LR 0.001000 Time 0.034145 -2022-12-06 10:49:31,260 - Epoch: [61][ 60/ 1200] Overall Loss 0.259641 Objective Loss 0.259641 LR 0.001000 Time 0.031604 -2022-12-06 10:49:31,451 - Epoch: [61][ 70/ 1200] Overall Loss 0.263875 Objective Loss 0.263875 LR 0.001000 Time 0.029805 -2022-12-06 10:49:31,641 - Epoch: [61][ 80/ 1200] Overall Loss 0.260608 Objective Loss 0.260608 LR 0.001000 Time 0.028452 -2022-12-06 10:49:31,831 - Epoch: [61][ 90/ 1200] Overall Loss 0.258494 Objective Loss 0.258494 LR 0.001000 Time 0.027401 -2022-12-06 10:49:32,022 - Epoch: [61][ 100/ 1200] Overall Loss 0.258723 Objective Loss 0.258723 LR 0.001000 Time 0.026558 -2022-12-06 10:49:32,212 - Epoch: [61][ 110/ 1200] Overall Loss 0.260070 Objective Loss 0.260070 LR 0.001000 Time 0.025864 -2022-12-06 10:49:32,401 - Epoch: [61][ 120/ 1200] Overall Loss 0.257792 Objective Loss 0.257792 LR 0.001000 Time 0.025287 -2022-12-06 10:49:32,591 - Epoch: [61][ 130/ 1200] Overall Loss 0.257194 Objective Loss 0.257194 LR 0.001000 Time 0.024800 -2022-12-06 10:49:32,782 - Epoch: [61][ 140/ 1200] Overall Loss 0.257924 Objective Loss 0.257924 LR 0.001000 Time 0.024383 -2022-12-06 10:49:32,972 - Epoch: [61][ 150/ 1200] Overall Loss 0.259604 Objective Loss 0.259604 LR 0.001000 Time 0.024020 -2022-12-06 10:49:33,162 - Epoch: [61][ 160/ 1200] Overall Loss 0.259771 Objective Loss 0.259771 LR 0.001000 Time 0.023705 -2022-12-06 10:49:33,352 - Epoch: [61][ 170/ 1200] Overall Loss 0.259309 Objective Loss 0.259309 LR 0.001000 Time 0.023424 -2022-12-06 10:49:33,542 - Epoch: [61][ 180/ 1200] Overall Loss 0.259887 Objective Loss 0.259887 LR 0.001000 Time 0.023176 -2022-12-06 10:49:33,732 - Epoch: [61][ 190/ 1200] Overall Loss 0.260847 Objective Loss 0.260847 LR 0.001000 Time 0.022955 -2022-12-06 10:49:33,922 - Epoch: [61][ 200/ 1200] Overall Loss 0.260735 Objective Loss 0.260735 LR 0.001000 Time 0.022751 -2022-12-06 10:49:34,112 - Epoch: [61][ 210/ 1200] Overall Loss 0.261313 Objective Loss 0.261313 LR 0.001000 Time 0.022570 -2022-12-06 10:49:34,302 - Epoch: [61][ 220/ 1200] Overall Loss 0.260970 Objective Loss 0.260970 LR 0.001000 Time 0.022405 -2022-12-06 10:49:34,492 - Epoch: [61][ 230/ 1200] Overall Loss 0.261463 Objective Loss 0.261463 LR 0.001000 Time 0.022255 -2022-12-06 10:49:34,682 - Epoch: [61][ 240/ 1200] Overall Loss 0.262141 Objective Loss 0.262141 LR 0.001000 Time 0.022117 -2022-12-06 10:49:34,872 - Epoch: [61][ 250/ 1200] Overall Loss 0.261472 Objective Loss 0.261472 LR 0.001000 Time 0.021990 -2022-12-06 10:49:35,062 - Epoch: [61][ 260/ 1200] Overall Loss 0.261115 Objective Loss 0.261115 LR 0.001000 Time 0.021876 -2022-12-06 10:49:35,253 - Epoch: [61][ 270/ 1200] Overall Loss 0.260591 Objective Loss 0.260591 LR 0.001000 Time 0.021768 -2022-12-06 10:49:35,443 - Epoch: [61][ 280/ 1200] Overall Loss 0.261904 Objective Loss 0.261904 LR 0.001000 Time 0.021668 -2022-12-06 10:49:35,632 - Epoch: [61][ 290/ 1200] Overall Loss 0.261906 Objective Loss 0.261906 LR 0.001000 Time 0.021573 -2022-12-06 10:49:35,823 - Epoch: [61][ 300/ 1200] Overall Loss 0.262075 Objective Loss 0.262075 LR 0.001000 Time 0.021486 -2022-12-06 10:49:36,013 - Epoch: [61][ 310/ 1200] Overall Loss 0.262658 Objective Loss 0.262658 LR 0.001000 Time 0.021404 -2022-12-06 10:49:36,204 - Epoch: [61][ 320/ 1200] Overall Loss 0.262810 Objective Loss 0.262810 LR 0.001000 Time 0.021331 -2022-12-06 10:49:36,394 - Epoch: [61][ 330/ 1200] Overall Loss 0.262948 Objective Loss 0.262948 LR 0.001000 Time 0.021260 -2022-12-06 10:49:36,585 - Epoch: [61][ 340/ 1200] Overall Loss 0.263889 Objective Loss 0.263889 LR 0.001000 Time 0.021194 -2022-12-06 10:49:36,776 - Epoch: [61][ 350/ 1200] Overall Loss 0.264880 Objective Loss 0.264880 LR 0.001000 Time 0.021133 -2022-12-06 10:49:36,966 - Epoch: [61][ 360/ 1200] Overall Loss 0.264902 Objective Loss 0.264902 LR 0.001000 Time 0.021071 -2022-12-06 10:49:37,156 - Epoch: [61][ 370/ 1200] Overall Loss 0.265040 Objective Loss 0.265040 LR 0.001000 Time 0.021013 -2022-12-06 10:49:37,345 - Epoch: [61][ 380/ 1200] Overall Loss 0.265345 Objective Loss 0.265345 LR 0.001000 Time 0.020958 -2022-12-06 10:49:37,535 - Epoch: [61][ 390/ 1200] Overall Loss 0.265852 Objective Loss 0.265852 LR 0.001000 Time 0.020907 -2022-12-06 10:49:37,726 - Epoch: [61][ 400/ 1200] Overall Loss 0.266207 Objective Loss 0.266207 LR 0.001000 Time 0.020858 -2022-12-06 10:49:37,916 - Epoch: [61][ 410/ 1200] Overall Loss 0.266040 Objective Loss 0.266040 LR 0.001000 Time 0.020813 -2022-12-06 10:49:38,107 - Epoch: [61][ 420/ 1200] Overall Loss 0.266201 Objective Loss 0.266201 LR 0.001000 Time 0.020770 -2022-12-06 10:49:38,296 - Epoch: [61][ 430/ 1200] Overall Loss 0.266851 Objective Loss 0.266851 LR 0.001000 Time 0.020727 -2022-12-06 10:49:38,488 - Epoch: [61][ 440/ 1200] Overall Loss 0.267403 Objective Loss 0.267403 LR 0.001000 Time 0.020689 -2022-12-06 10:49:38,678 - Epoch: [61][ 450/ 1200] Overall Loss 0.267140 Objective Loss 0.267140 LR 0.001000 Time 0.020651 -2022-12-06 10:49:38,869 - Epoch: [61][ 460/ 1200] Overall Loss 0.266894 Objective Loss 0.266894 LR 0.001000 Time 0.020617 -2022-12-06 10:49:39,059 - Epoch: [61][ 470/ 1200] Overall Loss 0.266797 Objective Loss 0.266797 LR 0.001000 Time 0.020581 -2022-12-06 10:49:39,250 - Epoch: [61][ 480/ 1200] Overall Loss 0.266710 Objective Loss 0.266710 LR 0.001000 Time 0.020548 -2022-12-06 10:49:39,440 - Epoch: [61][ 490/ 1200] Overall Loss 0.267530 Objective Loss 0.267530 LR 0.001000 Time 0.020515 -2022-12-06 10:49:39,630 - Epoch: [61][ 500/ 1200] Overall Loss 0.267910 Objective Loss 0.267910 LR 0.001000 Time 0.020485 -2022-12-06 10:49:39,820 - Epoch: [61][ 510/ 1200] Overall Loss 0.267297 Objective Loss 0.267297 LR 0.001000 Time 0.020455 -2022-12-06 10:49:40,011 - Epoch: [61][ 520/ 1200] Overall Loss 0.267995 Objective Loss 0.267995 LR 0.001000 Time 0.020427 -2022-12-06 10:49:40,201 - Epoch: [61][ 530/ 1200] Overall Loss 0.268570 Objective Loss 0.268570 LR 0.001000 Time 0.020400 -2022-12-06 10:49:40,392 - Epoch: [61][ 540/ 1200] Overall Loss 0.268899 Objective Loss 0.268899 LR 0.001000 Time 0.020374 -2022-12-06 10:49:40,582 - Epoch: [61][ 550/ 1200] Overall Loss 0.268848 Objective Loss 0.268848 LR 0.001000 Time 0.020348 -2022-12-06 10:49:40,773 - Epoch: [61][ 560/ 1200] Overall Loss 0.268829 Objective Loss 0.268829 LR 0.001000 Time 0.020325 -2022-12-06 10:49:40,963 - Epoch: [61][ 570/ 1200] Overall Loss 0.268948 Objective Loss 0.268948 LR 0.001000 Time 0.020301 -2022-12-06 10:49:41,154 - Epoch: [61][ 580/ 1200] Overall Loss 0.269319 Objective Loss 0.269319 LR 0.001000 Time 0.020279 -2022-12-06 10:49:41,344 - Epoch: [61][ 590/ 1200] Overall Loss 0.269154 Objective Loss 0.269154 LR 0.001000 Time 0.020256 -2022-12-06 10:49:41,534 - Epoch: [61][ 600/ 1200] Overall Loss 0.269152 Objective Loss 0.269152 LR 0.001000 Time 0.020235 -2022-12-06 10:49:41,725 - Epoch: [61][ 610/ 1200] Overall Loss 0.268951 Objective Loss 0.268951 LR 0.001000 Time 0.020215 -2022-12-06 10:49:41,915 - Epoch: [61][ 620/ 1200] Overall Loss 0.269366 Objective Loss 0.269366 LR 0.001000 Time 0.020196 -2022-12-06 10:49:42,106 - Epoch: [61][ 630/ 1200] Overall Loss 0.269344 Objective Loss 0.269344 LR 0.001000 Time 0.020177 -2022-12-06 10:49:42,297 - Epoch: [61][ 640/ 1200] Overall Loss 0.269206 Objective Loss 0.269206 LR 0.001000 Time 0.020159 -2022-12-06 10:49:42,487 - Epoch: [61][ 650/ 1200] Overall Loss 0.269395 Objective Loss 0.269395 LR 0.001000 Time 0.020141 -2022-12-06 10:49:42,678 - Epoch: [61][ 660/ 1200] Overall Loss 0.269128 Objective Loss 0.269128 LR 0.001000 Time 0.020123 -2022-12-06 10:49:42,867 - Epoch: [61][ 670/ 1200] Overall Loss 0.268929 Objective Loss 0.268929 LR 0.001000 Time 0.020105 -2022-12-06 10:49:43,058 - Epoch: [61][ 680/ 1200] Overall Loss 0.268675 Objective Loss 0.268675 LR 0.001000 Time 0.020089 -2022-12-06 10:49:43,248 - Epoch: [61][ 690/ 1200] Overall Loss 0.268672 Objective Loss 0.268672 LR 0.001000 Time 0.020073 -2022-12-06 10:49:43,438 - Epoch: [61][ 700/ 1200] Overall Loss 0.268917 Objective Loss 0.268917 LR 0.001000 Time 0.020057 -2022-12-06 10:49:43,629 - Epoch: [61][ 710/ 1200] Overall Loss 0.269068 Objective Loss 0.269068 LR 0.001000 Time 0.020042 -2022-12-06 10:49:43,820 - Epoch: [61][ 720/ 1200] Overall Loss 0.269148 Objective Loss 0.269148 LR 0.001000 Time 0.020029 -2022-12-06 10:49:44,011 - Epoch: [61][ 730/ 1200] Overall Loss 0.269210 Objective Loss 0.269210 LR 0.001000 Time 0.020015 -2022-12-06 10:49:44,202 - Epoch: [61][ 740/ 1200] Overall Loss 0.269497 Objective Loss 0.269497 LR 0.001000 Time 0.020002 -2022-12-06 10:49:44,392 - Epoch: [61][ 750/ 1200] Overall Loss 0.269660 Objective Loss 0.269660 LR 0.001000 Time 0.019987 -2022-12-06 10:49:44,583 - Epoch: [61][ 760/ 1200] Overall Loss 0.269733 Objective Loss 0.269733 LR 0.001000 Time 0.019975 -2022-12-06 10:49:44,774 - Epoch: [61][ 770/ 1200] Overall Loss 0.270139 Objective Loss 0.270139 LR 0.001000 Time 0.019963 -2022-12-06 10:49:44,964 - Epoch: [61][ 780/ 1200] Overall Loss 0.270435 Objective Loss 0.270435 LR 0.001000 Time 0.019951 -2022-12-06 10:49:45,155 - Epoch: [61][ 790/ 1200] Overall Loss 0.270697 Objective Loss 0.270697 LR 0.001000 Time 0.019938 -2022-12-06 10:49:45,345 - Epoch: [61][ 800/ 1200] Overall Loss 0.270835 Objective Loss 0.270835 LR 0.001000 Time 0.019926 -2022-12-06 10:49:45,535 - Epoch: [61][ 810/ 1200] Overall Loss 0.270932 Objective Loss 0.270932 LR 0.001000 Time 0.019915 -2022-12-06 10:49:45,726 - Epoch: [61][ 820/ 1200] Overall Loss 0.270985 Objective Loss 0.270985 LR 0.001000 Time 0.019904 -2022-12-06 10:49:45,916 - Epoch: [61][ 830/ 1200] Overall Loss 0.270743 Objective Loss 0.270743 LR 0.001000 Time 0.019892 -2022-12-06 10:49:46,107 - Epoch: [61][ 840/ 1200] Overall Loss 0.270616 Objective Loss 0.270616 LR 0.001000 Time 0.019883 -2022-12-06 10:49:46,299 - Epoch: [61][ 850/ 1200] Overall Loss 0.270419 Objective Loss 0.270419 LR 0.001000 Time 0.019873 -2022-12-06 10:49:46,490 - Epoch: [61][ 860/ 1200] Overall Loss 0.270487 Objective Loss 0.270487 LR 0.001000 Time 0.019863 -2022-12-06 10:49:46,681 - Epoch: [61][ 870/ 1200] Overall Loss 0.270486 Objective Loss 0.270486 LR 0.001000 Time 0.019854 -2022-12-06 10:49:46,871 - Epoch: [61][ 880/ 1200] Overall Loss 0.270433 Objective Loss 0.270433 LR 0.001000 Time 0.019845 -2022-12-06 10:49:47,062 - Epoch: [61][ 890/ 1200] Overall Loss 0.270257 Objective Loss 0.270257 LR 0.001000 Time 0.019835 -2022-12-06 10:49:47,253 - Epoch: [61][ 900/ 1200] Overall Loss 0.270217 Objective Loss 0.270217 LR 0.001000 Time 0.019826 -2022-12-06 10:49:47,442 - Epoch: [61][ 910/ 1200] Overall Loss 0.269912 Objective Loss 0.269912 LR 0.001000 Time 0.019816 -2022-12-06 10:49:47,633 - Epoch: [61][ 920/ 1200] Overall Loss 0.269869 Objective Loss 0.269869 LR 0.001000 Time 0.019807 -2022-12-06 10:49:47,823 - Epoch: [61][ 930/ 1200] Overall Loss 0.269938 Objective Loss 0.269938 LR 0.001000 Time 0.019798 -2022-12-06 10:49:48,013 - Epoch: [61][ 940/ 1200] Overall Loss 0.270179 Objective Loss 0.270179 LR 0.001000 Time 0.019789 -2022-12-06 10:49:48,203 - Epoch: [61][ 950/ 1200] Overall Loss 0.270446 Objective Loss 0.270446 LR 0.001000 Time 0.019780 -2022-12-06 10:49:48,393 - Epoch: [61][ 960/ 1200] Overall Loss 0.270478 Objective Loss 0.270478 LR 0.001000 Time 0.019772 -2022-12-06 10:49:48,584 - Epoch: [61][ 970/ 1200] Overall Loss 0.270744 Objective Loss 0.270744 LR 0.001000 Time 0.019763 -2022-12-06 10:49:48,774 - Epoch: [61][ 980/ 1200] Overall Loss 0.270581 Objective Loss 0.270581 LR 0.001000 Time 0.019755 -2022-12-06 10:49:48,964 - Epoch: [61][ 990/ 1200] Overall Loss 0.270454 Objective Loss 0.270454 LR 0.001000 Time 0.019747 -2022-12-06 10:49:49,154 - Epoch: [61][ 1000/ 1200] Overall Loss 0.270294 Objective Loss 0.270294 LR 0.001000 Time 0.019740 -2022-12-06 10:49:49,345 - Epoch: [61][ 1010/ 1200] Overall Loss 0.270179 Objective Loss 0.270179 LR 0.001000 Time 0.019732 -2022-12-06 10:49:49,535 - Epoch: [61][ 1020/ 1200] Overall Loss 0.269907 Objective Loss 0.269907 LR 0.001000 Time 0.019725 -2022-12-06 10:49:49,726 - Epoch: [61][ 1030/ 1200] Overall Loss 0.270040 Objective Loss 0.270040 LR 0.001000 Time 0.019718 -2022-12-06 10:49:49,916 - Epoch: [61][ 1040/ 1200] Overall Loss 0.270005 Objective Loss 0.270005 LR 0.001000 Time 0.019711 -2022-12-06 10:49:50,107 - Epoch: [61][ 1050/ 1200] Overall Loss 0.270203 Objective Loss 0.270203 LR 0.001000 Time 0.019704 -2022-12-06 10:49:50,297 - Epoch: [61][ 1060/ 1200] Overall Loss 0.270237 Objective Loss 0.270237 LR 0.001000 Time 0.019697 -2022-12-06 10:49:50,488 - Epoch: [61][ 1070/ 1200] Overall Loss 0.270167 Objective Loss 0.270167 LR 0.001000 Time 0.019691 -2022-12-06 10:49:50,679 - Epoch: [61][ 1080/ 1200] Overall Loss 0.270483 Objective Loss 0.270483 LR 0.001000 Time 0.019685 -2022-12-06 10:49:50,869 - Epoch: [61][ 1090/ 1200] Overall Loss 0.270487 Objective Loss 0.270487 LR 0.001000 Time 0.019678 -2022-12-06 10:49:51,060 - Epoch: [61][ 1100/ 1200] Overall Loss 0.270426 Objective Loss 0.270426 LR 0.001000 Time 0.019672 -2022-12-06 10:49:51,250 - Epoch: [61][ 1110/ 1200] Overall Loss 0.270403 Objective Loss 0.270403 LR 0.001000 Time 0.019666 -2022-12-06 10:49:51,440 - Epoch: [61][ 1120/ 1200] Overall Loss 0.270472 Objective Loss 0.270472 LR 0.001000 Time 0.019660 -2022-12-06 10:49:51,631 - Epoch: [61][ 1130/ 1200] Overall Loss 0.270589 Objective Loss 0.270589 LR 0.001000 Time 0.019654 -2022-12-06 10:49:51,821 - Epoch: [61][ 1140/ 1200] Overall Loss 0.270602 Objective Loss 0.270602 LR 0.001000 Time 0.019648 -2022-12-06 10:49:52,012 - Epoch: [61][ 1150/ 1200] Overall Loss 0.270948 Objective Loss 0.270948 LR 0.001000 Time 0.019642 -2022-12-06 10:49:52,202 - Epoch: [61][ 1160/ 1200] Overall Loss 0.271066 Objective Loss 0.271066 LR 0.001000 Time 0.019637 -2022-12-06 10:49:52,395 - Epoch: [61][ 1170/ 1200] Overall Loss 0.271137 Objective Loss 0.271137 LR 0.001000 Time 0.019633 -2022-12-06 10:49:52,587 - Epoch: [61][ 1180/ 1200] Overall Loss 0.271055 Objective Loss 0.271055 LR 0.001000 Time 0.019629 -2022-12-06 10:49:52,779 - Epoch: [61][ 1190/ 1200] Overall Loss 0.271015 Objective Loss 0.271015 LR 0.001000 Time 0.019626 -2022-12-06 10:49:53,011 - Epoch: [61][ 1200/ 1200] Overall Loss 0.271233 Objective Loss 0.271233 Top1 83.054393 Top5 98.117155 LR 0.001000 Time 0.019655 -2022-12-06 10:49:53,100 - --- validate (epoch=61)----------- -2022-12-06 10:49:53,100 - 34129 samples (256 per mini-batch) -2022-12-06 10:49:53,548 - Epoch: [61][ 10/ 134] Loss 0.290674 Top1 83.671875 Top5 97.812500 -2022-12-06 10:49:53,677 - Epoch: [61][ 20/ 134] Loss 0.297715 Top1 83.886719 Top5 97.714844 -2022-12-06 10:49:53,805 - Epoch: [61][ 30/ 134] Loss 0.280330 Top1 84.283854 Top5 97.786458 -2022-12-06 10:49:53,932 - Epoch: [61][ 40/ 134] Loss 0.297102 Top1 84.257812 Top5 97.792969 -2022-12-06 10:49:54,059 - Epoch: [61][ 50/ 134] Loss 0.295460 Top1 84.210938 Top5 97.875000 -2022-12-06 10:49:54,186 - Epoch: [61][ 60/ 134] Loss 0.297090 Top1 84.290365 Top5 97.923177 -2022-12-06 10:49:54,312 - Epoch: [61][ 70/ 134] Loss 0.301717 Top1 84.213170 Top5 97.901786 -2022-12-06 10:49:54,438 - Epoch: [61][ 80/ 134] Loss 0.302991 Top1 84.248047 Top5 97.919922 -2022-12-06 10:49:54,565 - Epoch: [61][ 90/ 134] Loss 0.301345 Top1 84.292535 Top5 97.955729 -2022-12-06 10:49:54,693 - Epoch: [61][ 100/ 134] Loss 0.301257 Top1 84.218750 Top5 97.968750 -2022-12-06 10:49:54,820 - Epoch: [61][ 110/ 134] Loss 0.303838 Top1 84.208097 Top5 97.950994 -2022-12-06 10:49:54,947 - Epoch: [61][ 120/ 134] Loss 0.305544 Top1 84.225260 Top5 97.932943 -2022-12-06 10:49:55,074 - Epoch: [61][ 130/ 134] Loss 0.307828 Top1 84.122596 Top5 97.938702 -2022-12-06 10:49:55,110 - Epoch: [61][ 134/ 134] Loss 0.306522 Top1 84.130798 Top5 97.948958 -2022-12-06 10:49:55,198 - ==> Top1: 84.131 Top5: 97.949 Loss: 0.307 - -2022-12-06 10:49:55,198 - ==> Confusion: -[[ 883 2 2 4 16 3 1 1 10 51 0 1 1 2 2 3 6 2 1 0 5] - [ 1 914 5 4 7 26 4 10 1 0 7 10 2 3 2 1 11 2 8 2 7] - [ 8 3 997 8 3 1 24 7 0 0 9 3 1 4 3 4 3 3 4 4 14] - [ 3 1 20 915 0 1 1 1 0 0 17 0 3 5 18 1 2 4 18 0 10] - [ 8 3 2 0 957 5 0 0 1 3 1 7 0 4 8 6 8 2 0 2 3] - [ 2 14 2 1 7 959 4 12 0 1 5 15 4 23 3 0 5 1 0 5 6] - [ 0 0 10 3 0 3 1072 3 0 0 5 4 2 1 0 3 0 2 1 7 2] - [ 0 7 10 2 3 41 15 902 1 1 4 9 1 3 1 0 1 1 27 17 8] - [ 7 3 0 1 0 2 1 0 960 35 15 3 1 13 15 1 2 0 0 2 3] - [ 67 0 1 0 6 3 0 2 30 862 1 1 0 15 2 3 1 2 1 0 4] - [ 0 1 3 6 0 3 2 1 10 0 957 5 1 15 5 1 0 0 6 0 3] - [ 3 1 0 1 2 6 1 2 1 0 0 989 15 7 0 3 6 2 0 9 3] - [ 1 1 1 5 1 3 2 0 0 0 0 81 824 3 3 10 2 11 1 5 15] - [ 0 1 1 0 1 11 0 2 9 5 7 6 2 959 1 2 3 1 1 2 9] - [ 3 4 1 14 4 3 0 1 18 4 0 4 2 2 1054 0 5 0 1 1 9] - [ 0 0 3 1 2 0 5 0 0 0 1 13 3 5 3 982 14 5 0 2 4] - [ 1 4 1 0 0 1 1 0 1 0 0 6 1 3 3 6 1037 2 0 2 3] - [ 3 0 1 2 0 0 0 1 0 3 0 25 18 4 1 17 1 955 0 2 3] - [ 2 6 2 10 1 4 0 21 1 0 13 3 3 1 9 1 4 0 914 7 6] - [ 2 0 1 1 0 8 9 5 1 0 2 24 6 7 2 4 4 3 1 992 8] - [ 106 204 190 85 128 213 95 111 70 104 250 196 271 391 184 131 416 74 135 244 9628]] - -2022-12-06 10:49:55,763 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:49:55,763 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:49:55,769 - - -2022-12-06 10:49:55,769 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:49:56,813 - Epoch: [62][ 10/ 1200] Overall Loss 0.260591 Objective Loss 0.260591 LR 0.001000 Time 0.104295 -2022-12-06 10:49:57,025 - Epoch: [62][ 20/ 1200] Overall Loss 0.254812 Objective Loss 0.254812 LR 0.001000 Time 0.062692 -2022-12-06 10:49:57,226 - Epoch: [62][ 30/ 1200] Overall Loss 0.263450 Objective Loss 0.263450 LR 0.001000 Time 0.048479 -2022-12-06 10:49:57,431 - Epoch: [62][ 40/ 1200] Overall Loss 0.256250 Objective Loss 0.256250 LR 0.001000 Time 0.041482 -2022-12-06 10:49:57,632 - Epoch: [62][ 50/ 1200] Overall Loss 0.260440 Objective Loss 0.260440 LR 0.001000 Time 0.037190 -2022-12-06 10:49:57,837 - Epoch: [62][ 60/ 1200] Overall Loss 0.259043 Objective Loss 0.259043 LR 0.001000 Time 0.034392 -2022-12-06 10:49:58,037 - Epoch: [62][ 70/ 1200] Overall Loss 0.255469 Objective Loss 0.255469 LR 0.001000 Time 0.032332 -2022-12-06 10:49:58,242 - Epoch: [62][ 80/ 1200] Overall Loss 0.253438 Objective Loss 0.253438 LR 0.001000 Time 0.030845 -2022-12-06 10:49:58,442 - Epoch: [62][ 90/ 1200] Overall Loss 0.253779 Objective Loss 0.253779 LR 0.001000 Time 0.029640 -2022-12-06 10:49:58,647 - Epoch: [62][ 100/ 1200] Overall Loss 0.257537 Objective Loss 0.257537 LR 0.001000 Time 0.028716 -2022-12-06 10:49:58,848 - Epoch: [62][ 110/ 1200] Overall Loss 0.258563 Objective Loss 0.258563 LR 0.001000 Time 0.027924 -2022-12-06 10:49:59,052 - Epoch: [62][ 120/ 1200] Overall Loss 0.259509 Objective Loss 0.259509 LR 0.001000 Time 0.027295 -2022-12-06 10:49:59,253 - Epoch: [62][ 130/ 1200] Overall Loss 0.261323 Objective Loss 0.261323 LR 0.001000 Time 0.026733 -2022-12-06 10:49:59,457 - Epoch: [62][ 140/ 1200] Overall Loss 0.258903 Objective Loss 0.258903 LR 0.001000 Time 0.026279 -2022-12-06 10:49:59,657 - Epoch: [62][ 150/ 1200] Overall Loss 0.259362 Objective Loss 0.259362 LR 0.001000 Time 0.025858 -2022-12-06 10:49:59,861 - Epoch: [62][ 160/ 1200] Overall Loss 0.259482 Objective Loss 0.259482 LR 0.001000 Time 0.025515 -2022-12-06 10:50:00,062 - Epoch: [62][ 170/ 1200] Overall Loss 0.259916 Objective Loss 0.259916 LR 0.001000 Time 0.025189 -2022-12-06 10:50:00,267 - Epoch: [62][ 180/ 1200] Overall Loss 0.258954 Objective Loss 0.258954 LR 0.001000 Time 0.024924 -2022-12-06 10:50:00,466 - Epoch: [62][ 190/ 1200] Overall Loss 0.259173 Objective Loss 0.259173 LR 0.001000 Time 0.024660 -2022-12-06 10:50:00,671 - Epoch: [62][ 200/ 1200] Overall Loss 0.259677 Objective Loss 0.259677 LR 0.001000 Time 0.024447 -2022-12-06 10:50:00,872 - Epoch: [62][ 210/ 1200] Overall Loss 0.260004 Objective Loss 0.260004 LR 0.001000 Time 0.024236 -2022-12-06 10:50:01,076 - Epoch: [62][ 220/ 1200] Overall Loss 0.260145 Objective Loss 0.260145 LR 0.001000 Time 0.024060 -2022-12-06 10:50:01,276 - Epoch: [62][ 230/ 1200] Overall Loss 0.259414 Objective Loss 0.259414 LR 0.001000 Time 0.023882 -2022-12-06 10:50:01,481 - Epoch: [62][ 240/ 1200] Overall Loss 0.259898 Objective Loss 0.259898 LR 0.001000 Time 0.023738 -2022-12-06 10:50:01,681 - Epoch: [62][ 250/ 1200] Overall Loss 0.258758 Objective Loss 0.258758 LR 0.001000 Time 0.023588 -2022-12-06 10:50:01,886 - Epoch: [62][ 260/ 1200] Overall Loss 0.259016 Objective Loss 0.259016 LR 0.001000 Time 0.023465 -2022-12-06 10:50:02,086 - Epoch: [62][ 270/ 1200] Overall Loss 0.258309 Objective Loss 0.258309 LR 0.001000 Time 0.023336 -2022-12-06 10:50:02,291 - Epoch: [62][ 280/ 1200] Overall Loss 0.259201 Objective Loss 0.259201 LR 0.001000 Time 0.023231 -2022-12-06 10:50:02,492 - Epoch: [62][ 290/ 1200] Overall Loss 0.259717 Objective Loss 0.259717 LR 0.001000 Time 0.023120 -2022-12-06 10:50:02,695 - Epoch: [62][ 300/ 1200] Overall Loss 0.259452 Objective Loss 0.259452 LR 0.001000 Time 0.023025 -2022-12-06 10:50:02,896 - Epoch: [62][ 310/ 1200] Overall Loss 0.260283 Objective Loss 0.260283 LR 0.001000 Time 0.022927 -2022-12-06 10:50:03,099 - Epoch: [62][ 320/ 1200] Overall Loss 0.261352 Objective Loss 0.261352 LR 0.001000 Time 0.022846 -2022-12-06 10:50:03,299 - Epoch: [62][ 330/ 1200] Overall Loss 0.261118 Objective Loss 0.261118 LR 0.001000 Time 0.022757 -2022-12-06 10:50:03,503 - Epoch: [62][ 340/ 1200] Overall Loss 0.261183 Objective Loss 0.261183 LR 0.001000 Time 0.022687 -2022-12-06 10:50:03,703 - Epoch: [62][ 350/ 1200] Overall Loss 0.261386 Objective Loss 0.261386 LR 0.001000 Time 0.022608 -2022-12-06 10:50:03,908 - Epoch: [62][ 360/ 1200] Overall Loss 0.260243 Objective Loss 0.260243 LR 0.001000 Time 0.022546 -2022-12-06 10:50:04,109 - Epoch: [62][ 370/ 1200] Overall Loss 0.261253 Objective Loss 0.261253 LR 0.001000 Time 0.022479 -2022-12-06 10:50:04,314 - Epoch: [62][ 380/ 1200] Overall Loss 0.260746 Objective Loss 0.260746 LR 0.001000 Time 0.022425 -2022-12-06 10:50:04,514 - Epoch: [62][ 390/ 1200] Overall Loss 0.261380 Objective Loss 0.261380 LR 0.001000 Time 0.022361 -2022-12-06 10:50:04,718 - Epoch: [62][ 400/ 1200] Overall Loss 0.261021 Objective Loss 0.261021 LR 0.001000 Time 0.022312 -2022-12-06 10:50:04,919 - Epoch: [62][ 410/ 1200] Overall Loss 0.261336 Objective Loss 0.261336 LR 0.001000 Time 0.022256 -2022-12-06 10:50:05,124 - Epoch: [62][ 420/ 1200] Overall Loss 0.261244 Objective Loss 0.261244 LR 0.001000 Time 0.022212 -2022-12-06 10:50:05,324 - Epoch: [62][ 430/ 1200] Overall Loss 0.261424 Objective Loss 0.261424 LR 0.001000 Time 0.022158 -2022-12-06 10:50:05,528 - Epoch: [62][ 440/ 1200] Overall Loss 0.261815 Objective Loss 0.261815 LR 0.001000 Time 0.022119 -2022-12-06 10:50:05,729 - Epoch: [62][ 450/ 1200] Overall Loss 0.261761 Objective Loss 0.261761 LR 0.001000 Time 0.022072 -2022-12-06 10:50:05,933 - Epoch: [62][ 460/ 1200] Overall Loss 0.261157 Objective Loss 0.261157 LR 0.001000 Time 0.022034 -2022-12-06 10:50:06,133 - Epoch: [62][ 470/ 1200] Overall Loss 0.261548 Objective Loss 0.261548 LR 0.001000 Time 0.021990 -2022-12-06 10:50:06,337 - Epoch: [62][ 480/ 1200] Overall Loss 0.261773 Objective Loss 0.261773 LR 0.001000 Time 0.021956 -2022-12-06 10:50:06,538 - Epoch: [62][ 490/ 1200] Overall Loss 0.261569 Objective Loss 0.261569 LR 0.001000 Time 0.021916 -2022-12-06 10:50:06,742 - Epoch: [62][ 500/ 1200] Overall Loss 0.261942 Objective Loss 0.261942 LR 0.001000 Time 0.021885 -2022-12-06 10:50:06,942 - Epoch: [62][ 510/ 1200] Overall Loss 0.262381 Objective Loss 0.262381 LR 0.001000 Time 0.021848 -2022-12-06 10:50:07,147 - Epoch: [62][ 520/ 1200] Overall Loss 0.262224 Objective Loss 0.262224 LR 0.001000 Time 0.021820 -2022-12-06 10:50:07,348 - Epoch: [62][ 530/ 1200] Overall Loss 0.263027 Objective Loss 0.263027 LR 0.001000 Time 0.021786 -2022-12-06 10:50:07,553 - Epoch: [62][ 540/ 1200] Overall Loss 0.263259 Objective Loss 0.263259 LR 0.001000 Time 0.021761 -2022-12-06 10:50:07,754 - Epoch: [62][ 550/ 1200] Overall Loss 0.263119 Objective Loss 0.263119 LR 0.001000 Time 0.021729 -2022-12-06 10:50:07,958 - Epoch: [62][ 560/ 1200] Overall Loss 0.262775 Objective Loss 0.262775 LR 0.001000 Time 0.021705 -2022-12-06 10:50:08,158 - Epoch: [62][ 570/ 1200] Overall Loss 0.262360 Objective Loss 0.262360 LR 0.001000 Time 0.021674 -2022-12-06 10:50:08,363 - Epoch: [62][ 580/ 1200] Overall Loss 0.262568 Objective Loss 0.262568 LR 0.001000 Time 0.021652 -2022-12-06 10:50:08,564 - Epoch: [62][ 590/ 1200] Overall Loss 0.262383 Objective Loss 0.262383 LR 0.001000 Time 0.021625 -2022-12-06 10:50:08,768 - Epoch: [62][ 600/ 1200] Overall Loss 0.262953 Objective Loss 0.262953 LR 0.001000 Time 0.021604 -2022-12-06 10:50:08,968 - Epoch: [62][ 610/ 1200] Overall Loss 0.263317 Objective Loss 0.263317 LR 0.001000 Time 0.021577 -2022-12-06 10:50:09,172 - Epoch: [62][ 620/ 1200] Overall Loss 0.263997 Objective Loss 0.263997 LR 0.001000 Time 0.021557 -2022-12-06 10:50:09,372 - Epoch: [62][ 630/ 1200] Overall Loss 0.264120 Objective Loss 0.264120 LR 0.001000 Time 0.021531 -2022-12-06 10:50:09,577 - Epoch: [62][ 640/ 1200] Overall Loss 0.263868 Objective Loss 0.263868 LR 0.001000 Time 0.021513 -2022-12-06 10:50:09,778 - Epoch: [62][ 650/ 1200] Overall Loss 0.264115 Objective Loss 0.264115 LR 0.001000 Time 0.021491 -2022-12-06 10:50:09,983 - Epoch: [62][ 660/ 1200] Overall Loss 0.264445 Objective Loss 0.264445 LR 0.001000 Time 0.021476 -2022-12-06 10:50:10,185 - Epoch: [62][ 670/ 1200] Overall Loss 0.264549 Objective Loss 0.264549 LR 0.001000 Time 0.021455 -2022-12-06 10:50:10,391 - Epoch: [62][ 680/ 1200] Overall Loss 0.265056 Objective Loss 0.265056 LR 0.001000 Time 0.021441 -2022-12-06 10:50:10,592 - Epoch: [62][ 690/ 1200] Overall Loss 0.265298 Objective Loss 0.265298 LR 0.001000 Time 0.021421 -2022-12-06 10:50:10,798 - Epoch: [62][ 700/ 1200] Overall Loss 0.264938 Objective Loss 0.264938 LR 0.001000 Time 0.021408 -2022-12-06 10:50:11,000 - Epoch: [62][ 710/ 1200] Overall Loss 0.265106 Objective Loss 0.265106 LR 0.001000 Time 0.021391 -2022-12-06 10:50:11,205 - Epoch: [62][ 720/ 1200] Overall Loss 0.265619 Objective Loss 0.265619 LR 0.001000 Time 0.021379 -2022-12-06 10:50:11,407 - Epoch: [62][ 730/ 1200] Overall Loss 0.266068 Objective Loss 0.266068 LR 0.001000 Time 0.021361 -2022-12-06 10:50:11,612 - Epoch: [62][ 740/ 1200] Overall Loss 0.266422 Objective Loss 0.266422 LR 0.001000 Time 0.021349 -2022-12-06 10:50:11,814 - Epoch: [62][ 750/ 1200] Overall Loss 0.266633 Objective Loss 0.266633 LR 0.001000 Time 0.021333 -2022-12-06 10:50:12,020 - Epoch: [62][ 760/ 1200] Overall Loss 0.266564 Objective Loss 0.266564 LR 0.001000 Time 0.021322 -2022-12-06 10:50:12,221 - Epoch: [62][ 770/ 1200] Overall Loss 0.266495 Objective Loss 0.266495 LR 0.001000 Time 0.021305 -2022-12-06 10:50:12,427 - Epoch: [62][ 780/ 1200] Overall Loss 0.266359 Objective Loss 0.266359 LR 0.001000 Time 0.021295 -2022-12-06 10:50:12,629 - Epoch: [62][ 790/ 1200] Overall Loss 0.266275 Objective Loss 0.266275 LR 0.001000 Time 0.021280 -2022-12-06 10:50:12,834 - Epoch: [62][ 800/ 1200] Overall Loss 0.266578 Objective Loss 0.266578 LR 0.001000 Time 0.021270 -2022-12-06 10:50:13,036 - Epoch: [62][ 810/ 1200] Overall Loss 0.266552 Objective Loss 0.266552 LR 0.001000 Time 0.021256 -2022-12-06 10:50:13,241 - Epoch: [62][ 820/ 1200] Overall Loss 0.266580 Objective Loss 0.266580 LR 0.001000 Time 0.021247 -2022-12-06 10:50:13,443 - Epoch: [62][ 830/ 1200] Overall Loss 0.266640 Objective Loss 0.266640 LR 0.001000 Time 0.021233 -2022-12-06 10:50:13,649 - Epoch: [62][ 840/ 1200] Overall Loss 0.266866 Objective Loss 0.266866 LR 0.001000 Time 0.021224 -2022-12-06 10:50:13,849 - Epoch: [62][ 850/ 1200] Overall Loss 0.267284 Objective Loss 0.267284 LR 0.001000 Time 0.021210 -2022-12-06 10:50:14,054 - Epoch: [62][ 860/ 1200] Overall Loss 0.267724 Objective Loss 0.267724 LR 0.001000 Time 0.021201 -2022-12-06 10:50:14,256 - Epoch: [62][ 870/ 1200] Overall Loss 0.267702 Objective Loss 0.267702 LR 0.001000 Time 0.021189 -2022-12-06 10:50:14,462 - Epoch: [62][ 880/ 1200] Overall Loss 0.267780 Objective Loss 0.267780 LR 0.001000 Time 0.021181 -2022-12-06 10:50:14,664 - Epoch: [62][ 890/ 1200] Overall Loss 0.267743 Objective Loss 0.267743 LR 0.001000 Time 0.021169 -2022-12-06 10:50:14,870 - Epoch: [62][ 900/ 1200] Overall Loss 0.267512 Objective Loss 0.267512 LR 0.001000 Time 0.021162 -2022-12-06 10:50:15,071 - Epoch: [62][ 910/ 1200] Overall Loss 0.267471 Objective Loss 0.267471 LR 0.001000 Time 0.021150 -2022-12-06 10:50:15,277 - Epoch: [62][ 920/ 1200] Overall Loss 0.267824 Objective Loss 0.267824 LR 0.001000 Time 0.021143 -2022-12-06 10:50:15,479 - Epoch: [62][ 930/ 1200] Overall Loss 0.267953 Objective Loss 0.267953 LR 0.001000 Time 0.021133 -2022-12-06 10:50:15,685 - Epoch: [62][ 940/ 1200] Overall Loss 0.267872 Objective Loss 0.267872 LR 0.001000 Time 0.021126 -2022-12-06 10:50:15,887 - Epoch: [62][ 950/ 1200] Overall Loss 0.267950 Objective Loss 0.267950 LR 0.001000 Time 0.021116 -2022-12-06 10:50:16,093 - Epoch: [62][ 960/ 1200] Overall Loss 0.268149 Objective Loss 0.268149 LR 0.001000 Time 0.021110 -2022-12-06 10:50:16,295 - Epoch: [62][ 970/ 1200] Overall Loss 0.268068 Objective Loss 0.268068 LR 0.001000 Time 0.021100 -2022-12-06 10:50:16,501 - Epoch: [62][ 980/ 1200] Overall Loss 0.268511 Objective Loss 0.268511 LR 0.001000 Time 0.021094 -2022-12-06 10:50:16,703 - Epoch: [62][ 990/ 1200] Overall Loss 0.268422 Objective Loss 0.268422 LR 0.001000 Time 0.021084 -2022-12-06 10:50:16,909 - Epoch: [62][ 1000/ 1200] Overall Loss 0.268714 Objective Loss 0.268714 LR 0.001000 Time 0.021080 -2022-12-06 10:50:17,111 - Epoch: [62][ 1010/ 1200] Overall Loss 0.269020 Objective Loss 0.269020 LR 0.001000 Time 0.021070 -2022-12-06 10:50:17,317 - Epoch: [62][ 1020/ 1200] Overall Loss 0.269150 Objective Loss 0.269150 LR 0.001000 Time 0.021065 -2022-12-06 10:50:17,519 - Epoch: [62][ 1030/ 1200] Overall Loss 0.269311 Objective Loss 0.269311 LR 0.001000 Time 0.021056 -2022-12-06 10:50:17,725 - Epoch: [62][ 1040/ 1200] Overall Loss 0.269393 Objective Loss 0.269393 LR 0.001000 Time 0.021051 -2022-12-06 10:50:17,927 - Epoch: [62][ 1050/ 1200] Overall Loss 0.269365 Objective Loss 0.269365 LR 0.001000 Time 0.021042 -2022-12-06 10:50:18,132 - Epoch: [62][ 1060/ 1200] Overall Loss 0.269495 Objective Loss 0.269495 LR 0.001000 Time 0.021037 -2022-12-06 10:50:18,334 - Epoch: [62][ 1070/ 1200] Overall Loss 0.269276 Objective Loss 0.269276 LR 0.001000 Time 0.021028 -2022-12-06 10:50:18,539 - Epoch: [62][ 1080/ 1200] Overall Loss 0.269326 Objective Loss 0.269326 LR 0.001000 Time 0.021023 -2022-12-06 10:50:18,741 - Epoch: [62][ 1090/ 1200] Overall Loss 0.269549 Objective Loss 0.269549 LR 0.001000 Time 0.021015 -2022-12-06 10:50:18,947 - Epoch: [62][ 1100/ 1200] Overall Loss 0.269714 Objective Loss 0.269714 LR 0.001000 Time 0.021010 -2022-12-06 10:50:19,149 - Epoch: [62][ 1110/ 1200] Overall Loss 0.269906 Objective Loss 0.269906 LR 0.001000 Time 0.021002 -2022-12-06 10:50:19,355 - Epoch: [62][ 1120/ 1200] Overall Loss 0.270161 Objective Loss 0.270161 LR 0.001000 Time 0.020998 -2022-12-06 10:50:19,556 - Epoch: [62][ 1130/ 1200] Overall Loss 0.270047 Objective Loss 0.270047 LR 0.001000 Time 0.020990 -2022-12-06 10:50:19,762 - Epoch: [62][ 1140/ 1200] Overall Loss 0.270010 Objective Loss 0.270010 LR 0.001000 Time 0.020986 -2022-12-06 10:50:19,964 - Epoch: [62][ 1150/ 1200] Overall Loss 0.270120 Objective Loss 0.270120 LR 0.001000 Time 0.020978 -2022-12-06 10:50:20,170 - Epoch: [62][ 1160/ 1200] Overall Loss 0.269954 Objective Loss 0.269954 LR 0.001000 Time 0.020975 -2022-12-06 10:50:20,371 - Epoch: [62][ 1170/ 1200] Overall Loss 0.270102 Objective Loss 0.270102 LR 0.001000 Time 0.020967 -2022-12-06 10:50:20,577 - Epoch: [62][ 1180/ 1200] Overall Loss 0.270226 Objective Loss 0.270226 LR 0.001000 Time 0.020963 -2022-12-06 10:50:20,779 - Epoch: [62][ 1190/ 1200] Overall Loss 0.270430 Objective Loss 0.270430 LR 0.001000 Time 0.020956 -2022-12-06 10:50:21,015 - Epoch: [62][ 1200/ 1200] Overall Loss 0.270662 Objective Loss 0.270662 Top1 85.774059 Top5 98.117155 LR 0.001000 Time 0.020978 -2022-12-06 10:50:21,111 - --- validate (epoch=62)----------- -2022-12-06 10:50:21,111 - 34129 samples (256 per mini-batch) -2022-12-06 10:50:21,570 - Epoch: [62][ 10/ 134] Loss 0.297911 Top1 85.703125 Top5 98.203125 -2022-12-06 10:50:21,709 - Epoch: [62][ 20/ 134] Loss 0.303531 Top1 84.902344 Top5 98.027344 -2022-12-06 10:50:21,854 - Epoch: [62][ 30/ 134] Loss 0.298743 Top1 84.934896 Top5 98.085938 -2022-12-06 10:50:21,984 - Epoch: [62][ 40/ 134] Loss 0.305534 Top1 84.853516 Top5 98.066406 -2022-12-06 10:50:22,111 - Epoch: [62][ 50/ 134] Loss 0.311913 Top1 84.578125 Top5 98.101562 -2022-12-06 10:50:22,238 - Epoch: [62][ 60/ 134] Loss 0.313367 Top1 84.518229 Top5 98.040365 -2022-12-06 10:50:22,364 - Epoch: [62][ 70/ 134] Loss 0.308385 Top1 84.492188 Top5 97.991071 -2022-12-06 10:50:22,491 - Epoch: [62][ 80/ 134] Loss 0.305782 Top1 84.614258 Top5 97.993164 -2022-12-06 10:50:22,620 - Epoch: [62][ 90/ 134] Loss 0.303549 Top1 84.661458 Top5 97.934028 -2022-12-06 10:50:22,746 - Epoch: [62][ 100/ 134] Loss 0.304724 Top1 84.609375 Top5 97.933594 -2022-12-06 10:50:22,875 - Epoch: [62][ 110/ 134] Loss 0.302039 Top1 84.669744 Top5 97.933239 -2022-12-06 10:50:23,001 - Epoch: [62][ 120/ 134] Loss 0.302541 Top1 84.700521 Top5 97.942708 -2022-12-06 10:50:23,130 - Epoch: [62][ 130/ 134] Loss 0.301582 Top1 84.729567 Top5 97.932692 -2022-12-06 10:50:23,167 - Epoch: [62][ 134/ 134] Loss 0.300527 Top1 84.760761 Top5 97.922588 -2022-12-06 10:50:23,262 - ==> Top1: 84.761 Top5: 97.923 Loss: 0.301 - -2022-12-06 10:50:23,263 - ==> Confusion: -[[ 901 2 2 4 11 8 0 0 4 47 0 3 1 2 2 2 0 0 0 0 7] - [ 1 941 3 1 7 20 3 21 2 2 3 5 0 0 1 2 4 0 4 1 6] - [ 9 3 996 7 5 3 28 11 0 1 4 7 1 3 2 3 0 2 4 2 12] - [ 5 6 24 884 3 3 5 2 1 0 12 2 5 2 19 1 0 4 29 1 12] - [ 18 9 4 0 943 6 2 1 0 4 1 5 0 2 7 3 7 2 0 2 4] - [ 3 19 0 1 4 965 2 25 3 1 0 15 5 12 1 1 1 1 4 2 4] - [ 3 4 14 0 0 3 1066 10 0 0 0 3 0 1 0 4 0 0 0 7 3] - [ 4 10 6 0 2 24 8 960 0 0 1 10 0 1 1 1 1 0 12 7 6] - [ 6 4 0 0 1 5 1 2 961 44 7 3 1 10 5 0 3 1 5 0 5] - [ 94 1 2 0 8 3 1 2 19 848 1 3 0 10 0 0 0 2 1 0 6] - [ 1 5 4 4 4 5 5 4 9 0 928 4 1 17 4 0 0 1 12 0 11] - [ 8 5 3 0 0 6 0 3 0 0 1 980 15 8 0 5 2 3 0 9 3] - [ 2 1 2 2 2 4 2 1 1 0 0 47 860 2 3 10 0 15 1 7 7] - [ 1 1 0 0 2 15 0 3 10 14 4 9 4 944 0 3 2 0 0 4 7] - [ 15 5 3 8 12 2 0 3 19 1 0 3 1 5 1026 0 2 2 14 2 7] - [ 0 1 3 3 1 3 4 0 0 0 0 10 6 1 0 979 7 16 0 5 4] - [ 3 9 2 0 3 4 2 0 1 0 0 4 1 2 2 11 1015 2 0 4 7] - [ 3 1 3 2 0 0 0 2 1 3 1 12 11 1 0 13 2 976 1 2 2] - [ 4 12 6 2 2 3 0 40 0 0 3 4 1 2 5 0 0 0 918 3 3] - [ 5 2 1 0 1 9 5 13 1 0 0 22 5 4 0 1 5 7 1 993 5] - [ 139 346 184 57 107 246 88 206 78 82 133 149 332 298 108 126 191 78 165 277 9836]] - -2022-12-06 10:50:23,831 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:50:23,831 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:50:23,837 - - -2022-12-06 10:50:23,837 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:50:24,770 - Epoch: [63][ 10/ 1200] Overall Loss 0.293175 Objective Loss 0.293175 LR 0.001000 Time 0.093248 -2022-12-06 10:50:24,967 - Epoch: [63][ 20/ 1200] Overall Loss 0.279198 Objective Loss 0.279198 LR 0.001000 Time 0.056427 -2022-12-06 10:50:25,166 - Epoch: [63][ 30/ 1200] Overall Loss 0.275270 Objective Loss 0.275270 LR 0.001000 Time 0.044238 -2022-12-06 10:50:25,362 - Epoch: [63][ 40/ 1200] Overall Loss 0.270409 Objective Loss 0.270409 LR 0.001000 Time 0.038072 -2022-12-06 10:50:25,561 - Epoch: [63][ 50/ 1200] Overall Loss 0.262568 Objective Loss 0.262568 LR 0.001000 Time 0.034412 -2022-12-06 10:50:25,757 - Epoch: [63][ 60/ 1200] Overall Loss 0.263257 Objective Loss 0.263257 LR 0.001000 Time 0.031937 -2022-12-06 10:50:25,955 - Epoch: [63][ 70/ 1200] Overall Loss 0.259907 Objective Loss 0.259907 LR 0.001000 Time 0.030199 -2022-12-06 10:50:26,151 - Epoch: [63][ 80/ 1200] Overall Loss 0.257273 Objective Loss 0.257273 LR 0.001000 Time 0.028864 -2022-12-06 10:50:26,349 - Epoch: [63][ 90/ 1200] Overall Loss 0.260236 Objective Loss 0.260236 LR 0.001000 Time 0.027855 -2022-12-06 10:50:26,545 - Epoch: [63][ 100/ 1200] Overall Loss 0.258925 Objective Loss 0.258925 LR 0.001000 Time 0.027026 -2022-12-06 10:50:26,744 - Epoch: [63][ 110/ 1200] Overall Loss 0.257897 Objective Loss 0.257897 LR 0.001000 Time 0.026369 -2022-12-06 10:50:26,940 - Epoch: [63][ 120/ 1200] Overall Loss 0.258482 Objective Loss 0.258482 LR 0.001000 Time 0.025800 -2022-12-06 10:50:27,139 - Epoch: [63][ 130/ 1200] Overall Loss 0.257858 Objective Loss 0.257858 LR 0.001000 Time 0.025340 -2022-12-06 10:50:27,335 - Epoch: [63][ 140/ 1200] Overall Loss 0.257583 Objective Loss 0.257583 LR 0.001000 Time 0.024925 -2022-12-06 10:50:27,533 - Epoch: [63][ 150/ 1200] Overall Loss 0.256997 Objective Loss 0.256997 LR 0.001000 Time 0.024582 -2022-12-06 10:50:27,729 - Epoch: [63][ 160/ 1200] Overall Loss 0.257697 Objective Loss 0.257697 LR 0.001000 Time 0.024267 -2022-12-06 10:50:27,925 - Epoch: [63][ 170/ 1200] Overall Loss 0.255959 Objective Loss 0.255959 LR 0.001000 Time 0.023993 -2022-12-06 10:50:28,117 - Epoch: [63][ 180/ 1200] Overall Loss 0.254813 Objective Loss 0.254813 LR 0.001000 Time 0.023723 -2022-12-06 10:50:28,309 - Epoch: [63][ 190/ 1200] Overall Loss 0.256229 Objective Loss 0.256229 LR 0.001000 Time 0.023479 -2022-12-06 10:50:28,499 - Epoch: [63][ 200/ 1200] Overall Loss 0.257661 Objective Loss 0.257661 LR 0.001000 Time 0.023255 -2022-12-06 10:50:28,690 - Epoch: [63][ 210/ 1200] Overall Loss 0.259154 Objective Loss 0.259154 LR 0.001000 Time 0.023055 -2022-12-06 10:50:28,882 - Epoch: [63][ 220/ 1200] Overall Loss 0.258256 Objective Loss 0.258256 LR 0.001000 Time 0.022877 -2022-12-06 10:50:29,073 - Epoch: [63][ 230/ 1200] Overall Loss 0.258137 Objective Loss 0.258137 LR 0.001000 Time 0.022708 -2022-12-06 10:50:29,263 - Epoch: [63][ 240/ 1200] Overall Loss 0.258496 Objective Loss 0.258496 LR 0.001000 Time 0.022553 -2022-12-06 10:50:29,454 - Epoch: [63][ 250/ 1200] Overall Loss 0.257657 Objective Loss 0.257657 LR 0.001000 Time 0.022412 -2022-12-06 10:50:29,646 - Epoch: [63][ 260/ 1200] Overall Loss 0.256857 Objective Loss 0.256857 LR 0.001000 Time 0.022285 -2022-12-06 10:50:29,837 - Epoch: [63][ 270/ 1200] Overall Loss 0.258497 Objective Loss 0.258497 LR 0.001000 Time 0.022166 -2022-12-06 10:50:30,028 - Epoch: [63][ 280/ 1200] Overall Loss 0.258126 Objective Loss 0.258126 LR 0.001000 Time 0.022054 -2022-12-06 10:50:30,219 - Epoch: [63][ 290/ 1200] Overall Loss 0.258738 Objective Loss 0.258738 LR 0.001000 Time 0.021950 -2022-12-06 10:50:30,411 - Epoch: [63][ 300/ 1200] Overall Loss 0.258875 Objective Loss 0.258875 LR 0.001000 Time 0.021856 -2022-12-06 10:50:30,601 - Epoch: [63][ 310/ 1200] Overall Loss 0.259687 Objective Loss 0.259687 LR 0.001000 Time 0.021764 -2022-12-06 10:50:30,792 - Epoch: [63][ 320/ 1200] Overall Loss 0.260557 Objective Loss 0.260557 LR 0.001000 Time 0.021679 -2022-12-06 10:50:30,984 - Epoch: [63][ 330/ 1200] Overall Loss 0.260843 Objective Loss 0.260843 LR 0.001000 Time 0.021602 -2022-12-06 10:50:31,175 - Epoch: [63][ 340/ 1200] Overall Loss 0.261044 Objective Loss 0.261044 LR 0.001000 Time 0.021526 -2022-12-06 10:50:31,365 - Epoch: [63][ 350/ 1200] Overall Loss 0.261273 Objective Loss 0.261273 LR 0.001000 Time 0.021454 -2022-12-06 10:50:31,556 - Epoch: [63][ 360/ 1200] Overall Loss 0.261041 Objective Loss 0.261041 LR 0.001000 Time 0.021387 -2022-12-06 10:50:31,747 - Epoch: [63][ 370/ 1200] Overall Loss 0.261970 Objective Loss 0.261970 LR 0.001000 Time 0.021323 -2022-12-06 10:50:31,939 - Epoch: [63][ 380/ 1200] Overall Loss 0.261709 Objective Loss 0.261709 LR 0.001000 Time 0.021265 -2022-12-06 10:50:32,130 - Epoch: [63][ 390/ 1200] Overall Loss 0.261742 Objective Loss 0.261742 LR 0.001000 Time 0.021208 -2022-12-06 10:50:32,320 - Epoch: [63][ 400/ 1200] Overall Loss 0.261886 Objective Loss 0.261886 LR 0.001000 Time 0.021153 -2022-12-06 10:50:32,512 - Epoch: [63][ 410/ 1200] Overall Loss 0.261221 Objective Loss 0.261221 LR 0.001000 Time 0.021103 -2022-12-06 10:50:32,702 - Epoch: [63][ 420/ 1200] Overall Loss 0.261507 Objective Loss 0.261507 LR 0.001000 Time 0.021052 -2022-12-06 10:50:32,893 - Epoch: [63][ 430/ 1200] Overall Loss 0.261727 Objective Loss 0.261727 LR 0.001000 Time 0.021005 -2022-12-06 10:50:33,083 - Epoch: [63][ 440/ 1200] Overall Loss 0.261488 Objective Loss 0.261488 LR 0.001000 Time 0.020958 -2022-12-06 10:50:33,274 - Epoch: [63][ 450/ 1200] Overall Loss 0.261146 Objective Loss 0.261146 LR 0.001000 Time 0.020917 -2022-12-06 10:50:33,465 - Epoch: [63][ 460/ 1200] Overall Loss 0.261045 Objective Loss 0.261045 LR 0.001000 Time 0.020876 -2022-12-06 10:50:33,656 - Epoch: [63][ 470/ 1200] Overall Loss 0.260870 Objective Loss 0.260870 LR 0.001000 Time 0.020836 -2022-12-06 10:50:33,847 - Epoch: [63][ 480/ 1200] Overall Loss 0.261046 Objective Loss 0.261046 LR 0.001000 Time 0.020799 -2022-12-06 10:50:34,038 - Epoch: [63][ 490/ 1200] Overall Loss 0.260891 Objective Loss 0.260891 LR 0.001000 Time 0.020762 -2022-12-06 10:50:34,228 - Epoch: [63][ 500/ 1200] Overall Loss 0.260974 Objective Loss 0.260974 LR 0.001000 Time 0.020727 -2022-12-06 10:50:34,419 - Epoch: [63][ 510/ 1200] Overall Loss 0.260916 Objective Loss 0.260916 LR 0.001000 Time 0.020694 -2022-12-06 10:50:34,611 - Epoch: [63][ 520/ 1200] Overall Loss 0.261098 Objective Loss 0.261098 LR 0.001000 Time 0.020663 -2022-12-06 10:50:34,802 - Epoch: [63][ 530/ 1200] Overall Loss 0.261184 Objective Loss 0.261184 LR 0.001000 Time 0.020632 -2022-12-06 10:50:34,992 - Epoch: [63][ 540/ 1200] Overall Loss 0.261795 Objective Loss 0.261795 LR 0.001000 Time 0.020602 -2022-12-06 10:50:35,184 - Epoch: [63][ 550/ 1200] Overall Loss 0.261492 Objective Loss 0.261492 LR 0.001000 Time 0.020576 -2022-12-06 10:50:35,375 - Epoch: [63][ 560/ 1200] Overall Loss 0.261886 Objective Loss 0.261886 LR 0.001000 Time 0.020548 -2022-12-06 10:50:35,565 - Epoch: [63][ 570/ 1200] Overall Loss 0.261759 Objective Loss 0.261759 LR 0.001000 Time 0.020520 -2022-12-06 10:50:35,756 - Epoch: [63][ 580/ 1200] Overall Loss 0.261954 Objective Loss 0.261954 LR 0.001000 Time 0.020494 -2022-12-06 10:50:35,949 - Epoch: [63][ 590/ 1200] Overall Loss 0.261950 Objective Loss 0.261950 LR 0.001000 Time 0.020473 -2022-12-06 10:50:36,139 - Epoch: [63][ 600/ 1200] Overall Loss 0.262293 Objective Loss 0.262293 LR 0.001000 Time 0.020448 -2022-12-06 10:50:36,329 - Epoch: [63][ 610/ 1200] Overall Loss 0.262338 Objective Loss 0.262338 LR 0.001000 Time 0.020423 -2022-12-06 10:50:36,520 - Epoch: [63][ 620/ 1200] Overall Loss 0.262109 Objective Loss 0.262109 LR 0.001000 Time 0.020400 -2022-12-06 10:50:36,711 - Epoch: [63][ 630/ 1200] Overall Loss 0.262178 Objective Loss 0.262178 LR 0.001000 Time 0.020379 -2022-12-06 10:50:36,902 - Epoch: [63][ 640/ 1200] Overall Loss 0.262316 Objective Loss 0.262316 LR 0.001000 Time 0.020358 -2022-12-06 10:50:37,093 - Epoch: [63][ 650/ 1200] Overall Loss 0.262304 Objective Loss 0.262304 LR 0.001000 Time 0.020338 -2022-12-06 10:50:37,283 - Epoch: [63][ 660/ 1200] Overall Loss 0.262622 Objective Loss 0.262622 LR 0.001000 Time 0.020318 -2022-12-06 10:50:37,474 - Epoch: [63][ 670/ 1200] Overall Loss 0.263071 Objective Loss 0.263071 LR 0.001000 Time 0.020299 -2022-12-06 10:50:37,667 - Epoch: [63][ 680/ 1200] Overall Loss 0.262832 Objective Loss 0.262832 LR 0.001000 Time 0.020282 -2022-12-06 10:50:37,857 - Epoch: [63][ 690/ 1200] Overall Loss 0.263006 Objective Loss 0.263006 LR 0.001000 Time 0.020264 -2022-12-06 10:50:38,049 - Epoch: [63][ 700/ 1200] Overall Loss 0.263015 Objective Loss 0.263015 LR 0.001000 Time 0.020247 -2022-12-06 10:50:38,240 - Epoch: [63][ 710/ 1200] Overall Loss 0.262722 Objective Loss 0.262722 LR 0.001000 Time 0.020230 -2022-12-06 10:50:38,432 - Epoch: [63][ 720/ 1200] Overall Loss 0.262795 Objective Loss 0.262795 LR 0.001000 Time 0.020216 -2022-12-06 10:50:38,623 - Epoch: [63][ 730/ 1200] Overall Loss 0.262617 Objective Loss 0.262617 LR 0.001000 Time 0.020199 -2022-12-06 10:50:38,814 - Epoch: [63][ 740/ 1200] Overall Loss 0.262839 Objective Loss 0.262839 LR 0.001000 Time 0.020184 -2022-12-06 10:50:39,005 - Epoch: [63][ 750/ 1200] Overall Loss 0.262971 Objective Loss 0.262971 LR 0.001000 Time 0.020168 -2022-12-06 10:50:39,196 - Epoch: [63][ 760/ 1200] Overall Loss 0.262902 Objective Loss 0.262902 LR 0.001000 Time 0.020154 -2022-12-06 10:50:39,388 - Epoch: [63][ 770/ 1200] Overall Loss 0.262854 Objective Loss 0.262854 LR 0.001000 Time 0.020141 -2022-12-06 10:50:39,579 - Epoch: [63][ 780/ 1200] Overall Loss 0.263190 Objective Loss 0.263190 LR 0.001000 Time 0.020127 -2022-12-06 10:50:39,770 - Epoch: [63][ 790/ 1200] Overall Loss 0.263229 Objective Loss 0.263229 LR 0.001000 Time 0.020113 -2022-12-06 10:50:39,961 - Epoch: [63][ 800/ 1200] Overall Loss 0.263071 Objective Loss 0.263071 LR 0.001000 Time 0.020100 -2022-12-06 10:50:40,151 - Epoch: [63][ 810/ 1200] Overall Loss 0.263213 Objective Loss 0.263213 LR 0.001000 Time 0.020086 -2022-12-06 10:50:40,343 - Epoch: [63][ 820/ 1200] Overall Loss 0.263188 Objective Loss 0.263188 LR 0.001000 Time 0.020073 -2022-12-06 10:50:40,534 - Epoch: [63][ 830/ 1200] Overall Loss 0.263450 Objective Loss 0.263450 LR 0.001000 Time 0.020062 -2022-12-06 10:50:40,726 - Epoch: [63][ 840/ 1200] Overall Loss 0.263834 Objective Loss 0.263834 LR 0.001000 Time 0.020050 -2022-12-06 10:50:40,918 - Epoch: [63][ 850/ 1200] Overall Loss 0.263684 Objective Loss 0.263684 LR 0.001000 Time 0.020039 -2022-12-06 10:50:41,108 - Epoch: [63][ 860/ 1200] Overall Loss 0.263828 Objective Loss 0.263828 LR 0.001000 Time 0.020028 -2022-12-06 10:50:41,301 - Epoch: [63][ 870/ 1200] Overall Loss 0.264225 Objective Loss 0.264225 LR 0.001000 Time 0.020018 -2022-12-06 10:50:41,492 - Epoch: [63][ 880/ 1200] Overall Loss 0.264501 Objective Loss 0.264501 LR 0.001000 Time 0.020008 -2022-12-06 10:50:41,684 - Epoch: [63][ 890/ 1200] Overall Loss 0.264741 Objective Loss 0.264741 LR 0.001000 Time 0.019998 -2022-12-06 10:50:41,876 - Epoch: [63][ 900/ 1200] Overall Loss 0.265138 Objective Loss 0.265138 LR 0.001000 Time 0.019988 -2022-12-06 10:50:42,068 - Epoch: [63][ 910/ 1200] Overall Loss 0.265465 Objective Loss 0.265465 LR 0.001000 Time 0.019978 -2022-12-06 10:50:42,259 - Epoch: [63][ 920/ 1200] Overall Loss 0.265763 Objective Loss 0.265763 LR 0.001000 Time 0.019968 -2022-12-06 10:50:42,449 - Epoch: [63][ 930/ 1200] Overall Loss 0.265859 Objective Loss 0.265859 LR 0.001000 Time 0.019958 -2022-12-06 10:50:42,640 - Epoch: [63][ 940/ 1200] Overall Loss 0.265936 Objective Loss 0.265936 LR 0.001000 Time 0.019948 -2022-12-06 10:50:42,831 - Epoch: [63][ 950/ 1200] Overall Loss 0.266115 Objective Loss 0.266115 LR 0.001000 Time 0.019938 -2022-12-06 10:50:43,022 - Epoch: [63][ 960/ 1200] Overall Loss 0.266056 Objective Loss 0.266056 LR 0.001000 Time 0.019929 -2022-12-06 10:50:43,212 - Epoch: [63][ 970/ 1200] Overall Loss 0.265938 Objective Loss 0.265938 LR 0.001000 Time 0.019919 -2022-12-06 10:50:43,403 - Epoch: [63][ 980/ 1200] Overall Loss 0.265714 Objective Loss 0.265714 LR 0.001000 Time 0.019910 -2022-12-06 10:50:43,594 - Epoch: [63][ 990/ 1200] Overall Loss 0.265866 Objective Loss 0.265866 LR 0.001000 Time 0.019902 -2022-12-06 10:50:43,785 - Epoch: [63][ 1000/ 1200] Overall Loss 0.265983 Objective Loss 0.265983 LR 0.001000 Time 0.019893 -2022-12-06 10:50:43,976 - Epoch: [63][ 1010/ 1200] Overall Loss 0.266079 Objective Loss 0.266079 LR 0.001000 Time 0.019884 -2022-12-06 10:50:44,167 - Epoch: [63][ 1020/ 1200] Overall Loss 0.266089 Objective Loss 0.266089 LR 0.001000 Time 0.019876 -2022-12-06 10:50:44,358 - Epoch: [63][ 1030/ 1200] Overall Loss 0.265964 Objective Loss 0.265964 LR 0.001000 Time 0.019868 -2022-12-06 10:50:44,549 - Epoch: [63][ 1040/ 1200] Overall Loss 0.266315 Objective Loss 0.266315 LR 0.001000 Time 0.019861 -2022-12-06 10:50:44,740 - Epoch: [63][ 1050/ 1200] Overall Loss 0.266554 Objective Loss 0.266554 LR 0.001000 Time 0.019852 -2022-12-06 10:50:44,931 - Epoch: [63][ 1060/ 1200] Overall Loss 0.266686 Objective Loss 0.266686 LR 0.001000 Time 0.019844 -2022-12-06 10:50:45,121 - Epoch: [63][ 1070/ 1200] Overall Loss 0.266533 Objective Loss 0.266533 LR 0.001000 Time 0.019836 -2022-12-06 10:50:45,311 - Epoch: [63][ 1080/ 1200] Overall Loss 0.266807 Objective Loss 0.266807 LR 0.001000 Time 0.019828 -2022-12-06 10:50:45,502 - Epoch: [63][ 1090/ 1200] Overall Loss 0.266718 Objective Loss 0.266718 LR 0.001000 Time 0.019820 -2022-12-06 10:50:45,692 - Epoch: [63][ 1100/ 1200] Overall Loss 0.267058 Objective Loss 0.267058 LR 0.001000 Time 0.019813 -2022-12-06 10:50:45,883 - Epoch: [63][ 1110/ 1200] Overall Loss 0.267086 Objective Loss 0.267086 LR 0.001000 Time 0.019806 -2022-12-06 10:50:46,073 - Epoch: [63][ 1120/ 1200] Overall Loss 0.267178 Objective Loss 0.267178 LR 0.001000 Time 0.019799 -2022-12-06 10:50:46,264 - Epoch: [63][ 1130/ 1200] Overall Loss 0.267135 Objective Loss 0.267135 LR 0.001000 Time 0.019792 -2022-12-06 10:50:46,455 - Epoch: [63][ 1140/ 1200] Overall Loss 0.267165 Objective Loss 0.267165 LR 0.001000 Time 0.019785 -2022-12-06 10:50:46,645 - Epoch: [63][ 1150/ 1200] Overall Loss 0.267068 Objective Loss 0.267068 LR 0.001000 Time 0.019778 -2022-12-06 10:50:46,836 - Epoch: [63][ 1160/ 1200] Overall Loss 0.267289 Objective Loss 0.267289 LR 0.001000 Time 0.019772 -2022-12-06 10:50:47,027 - Epoch: [63][ 1170/ 1200] Overall Loss 0.267418 Objective Loss 0.267418 LR 0.001000 Time 0.019765 -2022-12-06 10:50:47,218 - Epoch: [63][ 1180/ 1200] Overall Loss 0.267478 Objective Loss 0.267478 LR 0.001000 Time 0.019759 -2022-12-06 10:50:47,409 - Epoch: [63][ 1190/ 1200] Overall Loss 0.267356 Objective Loss 0.267356 LR 0.001000 Time 0.019753 -2022-12-06 10:50:47,631 - Epoch: [63][ 1200/ 1200] Overall Loss 0.267549 Objective Loss 0.267549 Top1 83.891213 Top5 97.698745 LR 0.001000 Time 0.019773 -2022-12-06 10:50:47,723 - --- validate (epoch=63)----------- -2022-12-06 10:50:47,723 - 34129 samples (256 per mini-batch) -2022-12-06 10:50:48,310 - Epoch: [63][ 10/ 134] Loss 0.290830 Top1 84.960938 Top5 97.929688 -2022-12-06 10:50:48,436 - Epoch: [63][ 20/ 134] Loss 0.286727 Top1 84.628906 Top5 97.871094 -2022-12-06 10:50:48,566 - Epoch: [63][ 30/ 134] Loss 0.298011 Top1 84.583333 Top5 97.981771 -2022-12-06 10:50:48,696 - Epoch: [63][ 40/ 134] Loss 0.289001 Top1 84.746094 Top5 97.890625 -2022-12-06 10:50:48,828 - Epoch: [63][ 50/ 134] Loss 0.290148 Top1 84.632812 Top5 97.859375 -2022-12-06 10:50:48,960 - Epoch: [63][ 60/ 134] Loss 0.291999 Top1 84.381510 Top5 97.766927 -2022-12-06 10:50:49,086 - Epoch: [63][ 70/ 134] Loss 0.295146 Top1 84.157366 Top5 97.717634 -2022-12-06 10:50:49,214 - Epoch: [63][ 80/ 134] Loss 0.292512 Top1 84.296875 Top5 97.744141 -2022-12-06 10:50:49,342 - Epoch: [63][ 90/ 134] Loss 0.290683 Top1 84.275174 Top5 97.760417 -2022-12-06 10:50:49,469 - Epoch: [63][ 100/ 134] Loss 0.290012 Top1 84.347656 Top5 97.808594 -2022-12-06 10:50:49,598 - Epoch: [63][ 110/ 134] Loss 0.292441 Top1 84.197443 Top5 97.840909 -2022-12-06 10:50:49,726 - Epoch: [63][ 120/ 134] Loss 0.291368 Top1 84.166667 Top5 97.838542 -2022-12-06 10:50:49,854 - Epoch: [63][ 130/ 134] Loss 0.291471 Top1 84.143630 Top5 97.806490 -2022-12-06 10:50:49,890 - Epoch: [63][ 134/ 134] Loss 0.289923 Top1 84.165959 Top5 97.805385 -2022-12-06 10:50:49,978 - ==> Top1: 84.166 Top5: 97.805 Loss: 0.290 - -2022-12-06 10:50:49,978 - ==> Confusion: -[[ 887 4 1 4 15 5 0 0 7 57 0 3 2 1 2 3 2 1 1 0 1] - [ 2 930 3 2 14 26 2 14 2 0 4 8 0 2 2 0 4 3 4 0 5] - [ 6 5 997 17 3 3 28 10 0 4 4 3 0 2 3 6 1 3 2 1 5] - [ 3 2 18 948 0 6 0 0 1 1 6 1 1 2 19 0 0 2 9 0 1] - [ 12 5 0 0 963 6 0 1 1 5 2 3 1 0 7 5 5 2 0 1 1] - [ 1 17 1 5 8 970 6 19 7 1 2 13 3 2 2 2 1 2 1 3 3] - [ 1 2 7 4 0 5 1074 6 0 0 0 6 2 1 1 4 0 0 0 5 0] - [ 1 10 7 2 3 34 9 940 1 0 4 5 0 0 1 1 3 2 22 5 4] - [ 5 3 1 1 1 2 0 1 967 50 10 0 1 7 11 1 1 0 1 0 1] - [ 66 0 0 1 4 7 0 1 26 877 1 2 0 6 2 1 1 1 0 0 5] - [ 0 2 1 13 1 1 2 3 10 3 944 2 2 12 6 1 1 0 8 1 6] - [ 4 2 3 0 2 12 2 5 0 0 0 969 23 2 0 6 1 11 0 6 3] - [ 2 1 0 5 2 4 2 0 0 0 0 31 870 1 3 10 2 28 0 2 6] - [ 1 1 0 0 6 18 1 4 10 24 7 9 3 920 3 2 4 2 0 3 5] - [ 5 6 2 14 9 3 0 2 15 6 2 3 2 2 1051 0 1 0 2 0 5] - [ 3 1 1 0 1 3 4 0 0 0 0 5 5 3 0 994 9 10 0 1 3] - [ 4 4 2 0 2 3 1 0 0 0 0 6 0 1 1 7 1030 3 0 2 6] - [ 4 0 1 1 0 1 3 1 0 3 1 8 14 3 0 14 0 977 0 1 4] - [ 2 4 8 15 0 3 0 27 5 1 6 0 1 2 8 0 2 0 918 2 4] - [ 3 4 1 0 1 13 7 7 0 1 1 27 6 5 1 2 8 6 4 975 8] - [ 180 265 198 152 194 230 88 152 88 115 171 175 358 274 183 172 267 114 148 182 9520]] - -2022-12-06 10:50:50,544 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:50:50,544 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:50:50,549 - - -2022-12-06 10:50:50,550 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:50:51,481 - Epoch: [64][ 10/ 1200] Overall Loss 0.269417 Objective Loss 0.269417 LR 0.001000 Time 0.093083 -2022-12-06 10:50:51,680 - Epoch: [64][ 20/ 1200] Overall Loss 0.276784 Objective Loss 0.276784 LR 0.001000 Time 0.056480 -2022-12-06 10:50:51,877 - Epoch: [64][ 30/ 1200] Overall Loss 0.275033 Objective Loss 0.275033 LR 0.001000 Time 0.044201 -2022-12-06 10:50:52,076 - Epoch: [64][ 40/ 1200] Overall Loss 0.278888 Objective Loss 0.278888 LR 0.001000 Time 0.038103 -2022-12-06 10:50:52,272 - Epoch: [64][ 50/ 1200] Overall Loss 0.277432 Objective Loss 0.277432 LR 0.001000 Time 0.034391 -2022-12-06 10:50:52,471 - Epoch: [64][ 60/ 1200] Overall Loss 0.276490 Objective Loss 0.276490 LR 0.001000 Time 0.031960 -2022-12-06 10:50:52,666 - Epoch: [64][ 70/ 1200] Overall Loss 0.272971 Objective Loss 0.272971 LR 0.001000 Time 0.030179 -2022-12-06 10:50:52,865 - Epoch: [64][ 80/ 1200] Overall Loss 0.270023 Objective Loss 0.270023 LR 0.001000 Time 0.028880 -2022-12-06 10:50:53,060 - Epoch: [64][ 90/ 1200] Overall Loss 0.269353 Objective Loss 0.269353 LR 0.001000 Time 0.027838 -2022-12-06 10:50:53,258 - Epoch: [64][ 100/ 1200] Overall Loss 0.269240 Objective Loss 0.269240 LR 0.001000 Time 0.027032 -2022-12-06 10:50:53,455 - Epoch: [64][ 110/ 1200] Overall Loss 0.268409 Objective Loss 0.268409 LR 0.001000 Time 0.026353 -2022-12-06 10:50:53,653 - Epoch: [64][ 120/ 1200] Overall Loss 0.269435 Objective Loss 0.269435 LR 0.001000 Time 0.025805 -2022-12-06 10:50:53,849 - Epoch: [64][ 130/ 1200] Overall Loss 0.269276 Objective Loss 0.269276 LR 0.001000 Time 0.025327 -2022-12-06 10:50:54,048 - Epoch: [64][ 140/ 1200] Overall Loss 0.268942 Objective Loss 0.268942 LR 0.001000 Time 0.024933 -2022-12-06 10:50:54,244 - Epoch: [64][ 150/ 1200] Overall Loss 0.269025 Objective Loss 0.269025 LR 0.001000 Time 0.024577 -2022-12-06 10:50:54,443 - Epoch: [64][ 160/ 1200] Overall Loss 0.267984 Objective Loss 0.267984 LR 0.001000 Time 0.024279 -2022-12-06 10:50:54,640 - Epoch: [64][ 170/ 1200] Overall Loss 0.266604 Objective Loss 0.266604 LR 0.001000 Time 0.024004 -2022-12-06 10:50:54,839 - Epoch: [64][ 180/ 1200] Overall Loss 0.267437 Objective Loss 0.267437 LR 0.001000 Time 0.023773 -2022-12-06 10:50:55,035 - Epoch: [64][ 190/ 1200] Overall Loss 0.267007 Objective Loss 0.267007 LR 0.001000 Time 0.023550 -2022-12-06 10:50:55,235 - Epoch: [64][ 200/ 1200] Overall Loss 0.268304 Objective Loss 0.268304 LR 0.001000 Time 0.023370 -2022-12-06 10:50:55,432 - Epoch: [64][ 210/ 1200] Overall Loss 0.268345 Objective Loss 0.268345 LR 0.001000 Time 0.023195 -2022-12-06 10:50:55,633 - Epoch: [64][ 220/ 1200] Overall Loss 0.268303 Objective Loss 0.268303 LR 0.001000 Time 0.023052 -2022-12-06 10:50:55,831 - Epoch: [64][ 230/ 1200] Overall Loss 0.268045 Objective Loss 0.268045 LR 0.001000 Time 0.022907 -2022-12-06 10:50:56,032 - Epoch: [64][ 240/ 1200] Overall Loss 0.267913 Objective Loss 0.267913 LR 0.001000 Time 0.022786 -2022-12-06 10:50:56,229 - Epoch: [64][ 250/ 1200] Overall Loss 0.268912 Objective Loss 0.268912 LR 0.001000 Time 0.022664 -2022-12-06 10:50:56,430 - Epoch: [64][ 260/ 1200] Overall Loss 0.269347 Objective Loss 0.269347 LR 0.001000 Time 0.022560 -2022-12-06 10:50:56,628 - Epoch: [64][ 270/ 1200] Overall Loss 0.269089 Objective Loss 0.269089 LR 0.001000 Time 0.022456 -2022-12-06 10:50:56,829 - Epoch: [64][ 280/ 1200] Overall Loss 0.269716 Objective Loss 0.269716 LR 0.001000 Time 0.022369 -2022-12-06 10:50:57,026 - Epoch: [64][ 290/ 1200] Overall Loss 0.269673 Objective Loss 0.269673 LR 0.001000 Time 0.022276 -2022-12-06 10:50:57,227 - Epoch: [64][ 300/ 1200] Overall Loss 0.269041 Objective Loss 0.269041 LR 0.001000 Time 0.022201 -2022-12-06 10:50:57,424 - Epoch: [64][ 310/ 1200] Overall Loss 0.269135 Objective Loss 0.269135 LR 0.001000 Time 0.022121 -2022-12-06 10:50:57,626 - Epoch: [64][ 320/ 1200] Overall Loss 0.268472 Objective Loss 0.268472 LR 0.001000 Time 0.022057 -2022-12-06 10:50:57,823 - Epoch: [64][ 330/ 1200] Overall Loss 0.269057 Objective Loss 0.269057 LR 0.001000 Time 0.021985 -2022-12-06 10:50:58,024 - Epoch: [64][ 340/ 1200] Overall Loss 0.269907 Objective Loss 0.269907 LR 0.001000 Time 0.021928 -2022-12-06 10:50:58,221 - Epoch: [64][ 350/ 1200] Overall Loss 0.270098 Objective Loss 0.270098 LR 0.001000 Time 0.021862 -2022-12-06 10:50:58,422 - Epoch: [64][ 360/ 1200] Overall Loss 0.270432 Objective Loss 0.270432 LR 0.001000 Time 0.021811 -2022-12-06 10:50:58,619 - Epoch: [64][ 370/ 1200] Overall Loss 0.270463 Objective Loss 0.270463 LR 0.001000 Time 0.021753 -2022-12-06 10:50:58,819 - Epoch: [64][ 380/ 1200] Overall Loss 0.271214 Objective Loss 0.271214 LR 0.001000 Time 0.021707 -2022-12-06 10:50:59,017 - Epoch: [64][ 390/ 1200] Overall Loss 0.271486 Objective Loss 0.271486 LR 0.001000 Time 0.021655 -2022-12-06 10:50:59,218 - Epoch: [64][ 400/ 1200] Overall Loss 0.271552 Objective Loss 0.271552 LR 0.001000 Time 0.021614 -2022-12-06 10:50:59,415 - Epoch: [64][ 410/ 1200] Overall Loss 0.271439 Objective Loss 0.271439 LR 0.001000 Time 0.021568 -2022-12-06 10:50:59,616 - Epoch: [64][ 420/ 1200] Overall Loss 0.271496 Objective Loss 0.271496 LR 0.001000 Time 0.021530 -2022-12-06 10:50:59,813 - Epoch: [64][ 430/ 1200] Overall Loss 0.270953 Objective Loss 0.270953 LR 0.001000 Time 0.021488 -2022-12-06 10:51:00,014 - Epoch: [64][ 440/ 1200] Overall Loss 0.271276 Objective Loss 0.271276 LR 0.001000 Time 0.021454 -2022-12-06 10:51:00,211 - Epoch: [64][ 450/ 1200] Overall Loss 0.271143 Objective Loss 0.271143 LR 0.001000 Time 0.021414 -2022-12-06 10:51:00,411 - Epoch: [64][ 460/ 1200] Overall Loss 0.271377 Objective Loss 0.271377 LR 0.001000 Time 0.021382 -2022-12-06 10:51:00,608 - Epoch: [64][ 470/ 1200] Overall Loss 0.270766 Objective Loss 0.270766 LR 0.001000 Time 0.021345 -2022-12-06 10:51:00,810 - Epoch: [64][ 480/ 1200] Overall Loss 0.270439 Objective Loss 0.270439 LR 0.001000 Time 0.021320 -2022-12-06 10:51:01,006 - Epoch: [64][ 490/ 1200] Overall Loss 0.270255 Objective Loss 0.270255 LR 0.001000 Time 0.021285 -2022-12-06 10:51:01,207 - Epoch: [64][ 500/ 1200] Overall Loss 0.269690 Objective Loss 0.269690 LR 0.001000 Time 0.021260 -2022-12-06 10:51:01,405 - Epoch: [64][ 510/ 1200] Overall Loss 0.269563 Objective Loss 0.269563 LR 0.001000 Time 0.021230 -2022-12-06 10:51:01,606 - Epoch: [64][ 520/ 1200] Overall Loss 0.269632 Objective Loss 0.269632 LR 0.001000 Time 0.021206 -2022-12-06 10:51:01,802 - Epoch: [64][ 530/ 1200] Overall Loss 0.269512 Objective Loss 0.269512 LR 0.001000 Time 0.021176 -2022-12-06 10:51:02,003 - Epoch: [64][ 540/ 1200] Overall Loss 0.269064 Objective Loss 0.269064 LR 0.001000 Time 0.021154 -2022-12-06 10:51:02,200 - Epoch: [64][ 550/ 1200] Overall Loss 0.269554 Objective Loss 0.269554 LR 0.001000 Time 0.021128 -2022-12-06 10:51:02,400 - Epoch: [64][ 560/ 1200] Overall Loss 0.269789 Objective Loss 0.269789 LR 0.001000 Time 0.021107 -2022-12-06 10:51:02,599 - Epoch: [64][ 570/ 1200] Overall Loss 0.269846 Objective Loss 0.269846 LR 0.001000 Time 0.021084 -2022-12-06 10:51:02,799 - Epoch: [64][ 580/ 1200] Overall Loss 0.269357 Objective Loss 0.269357 LR 0.001000 Time 0.021064 -2022-12-06 10:51:02,996 - Epoch: [64][ 590/ 1200] Overall Loss 0.269483 Objective Loss 0.269483 LR 0.001000 Time 0.021041 -2022-12-06 10:51:03,197 - Epoch: [64][ 600/ 1200] Overall Loss 0.268784 Objective Loss 0.268784 LR 0.001000 Time 0.021024 -2022-12-06 10:51:03,393 - Epoch: [64][ 610/ 1200] Overall Loss 0.268977 Objective Loss 0.268977 LR 0.001000 Time 0.021000 -2022-12-06 10:51:03,594 - Epoch: [64][ 620/ 1200] Overall Loss 0.268666 Objective Loss 0.268666 LR 0.001000 Time 0.020984 -2022-12-06 10:51:03,791 - Epoch: [64][ 630/ 1200] Overall Loss 0.268459 Objective Loss 0.268459 LR 0.001000 Time 0.020964 -2022-12-06 10:51:03,992 - Epoch: [64][ 640/ 1200] Overall Loss 0.268509 Objective Loss 0.268509 LR 0.001000 Time 0.020949 -2022-12-06 10:51:04,190 - Epoch: [64][ 650/ 1200] Overall Loss 0.268256 Objective Loss 0.268256 LR 0.001000 Time 0.020929 -2022-12-06 10:51:04,390 - Epoch: [64][ 660/ 1200] Overall Loss 0.268321 Objective Loss 0.268321 LR 0.001000 Time 0.020915 -2022-12-06 10:51:04,588 - Epoch: [64][ 670/ 1200] Overall Loss 0.268579 Objective Loss 0.268579 LR 0.001000 Time 0.020898 -2022-12-06 10:51:04,789 - Epoch: [64][ 680/ 1200] Overall Loss 0.268824 Objective Loss 0.268824 LR 0.001000 Time 0.020885 -2022-12-06 10:51:04,987 - Epoch: [64][ 690/ 1200] Overall Loss 0.269307 Objective Loss 0.269307 LR 0.001000 Time 0.020868 -2022-12-06 10:51:05,188 - Epoch: [64][ 700/ 1200] Overall Loss 0.269337 Objective Loss 0.269337 LR 0.001000 Time 0.020857 -2022-12-06 10:51:05,385 - Epoch: [64][ 710/ 1200] Overall Loss 0.268838 Objective Loss 0.268838 LR 0.001000 Time 0.020840 -2022-12-06 10:51:05,586 - Epoch: [64][ 720/ 1200] Overall Loss 0.268945 Objective Loss 0.268945 LR 0.001000 Time 0.020829 -2022-12-06 10:51:05,784 - Epoch: [64][ 730/ 1200] Overall Loss 0.268529 Objective Loss 0.268529 LR 0.001000 Time 0.020813 -2022-12-06 10:51:05,985 - Epoch: [64][ 740/ 1200] Overall Loss 0.268712 Objective Loss 0.268712 LR 0.001000 Time 0.020803 -2022-12-06 10:51:06,182 - Epoch: [64][ 750/ 1200] Overall Loss 0.268371 Objective Loss 0.268371 LR 0.001000 Time 0.020788 -2022-12-06 10:51:06,383 - Epoch: [64][ 760/ 1200] Overall Loss 0.268081 Objective Loss 0.268081 LR 0.001000 Time 0.020778 -2022-12-06 10:51:06,580 - Epoch: [64][ 770/ 1200] Overall Loss 0.267904 Objective Loss 0.267904 LR 0.001000 Time 0.020764 -2022-12-06 10:51:06,782 - Epoch: [64][ 780/ 1200] Overall Loss 0.267790 Objective Loss 0.267790 LR 0.001000 Time 0.020755 -2022-12-06 10:51:06,979 - Epoch: [64][ 790/ 1200] Overall Loss 0.267681 Objective Loss 0.267681 LR 0.001000 Time 0.020742 -2022-12-06 10:51:07,180 - Epoch: [64][ 800/ 1200] Overall Loss 0.267699 Objective Loss 0.267699 LR 0.001000 Time 0.020733 -2022-12-06 10:51:07,378 - Epoch: [64][ 810/ 1200] Overall Loss 0.267863 Objective Loss 0.267863 LR 0.001000 Time 0.020721 -2022-12-06 10:51:07,578 - Epoch: [64][ 820/ 1200] Overall Loss 0.268177 Objective Loss 0.268177 LR 0.001000 Time 0.020711 -2022-12-06 10:51:07,776 - Epoch: [64][ 830/ 1200] Overall Loss 0.268332 Objective Loss 0.268332 LR 0.001000 Time 0.020699 -2022-12-06 10:51:07,978 - Epoch: [64][ 840/ 1200] Overall Loss 0.268584 Objective Loss 0.268584 LR 0.001000 Time 0.020692 -2022-12-06 10:51:08,175 - Epoch: [64][ 850/ 1200] Overall Loss 0.268777 Objective Loss 0.268777 LR 0.001000 Time 0.020681 -2022-12-06 10:51:08,376 - Epoch: [64][ 860/ 1200] Overall Loss 0.268696 Objective Loss 0.268696 LR 0.001000 Time 0.020673 -2022-12-06 10:51:08,574 - Epoch: [64][ 870/ 1200] Overall Loss 0.268494 Objective Loss 0.268494 LR 0.001000 Time 0.020662 -2022-12-06 10:51:08,774 - Epoch: [64][ 880/ 1200] Overall Loss 0.268085 Objective Loss 0.268085 LR 0.001000 Time 0.020655 -2022-12-06 10:51:08,971 - Epoch: [64][ 890/ 1200] Overall Loss 0.267929 Objective Loss 0.267929 LR 0.001000 Time 0.020643 -2022-12-06 10:51:09,172 - Epoch: [64][ 900/ 1200] Overall Loss 0.267801 Objective Loss 0.267801 LR 0.001000 Time 0.020636 -2022-12-06 10:51:09,370 - Epoch: [64][ 910/ 1200] Overall Loss 0.267648 Objective Loss 0.267648 LR 0.001000 Time 0.020627 -2022-12-06 10:51:09,571 - Epoch: [64][ 920/ 1200] Overall Loss 0.267470 Objective Loss 0.267470 LR 0.001000 Time 0.020620 -2022-12-06 10:51:09,768 - Epoch: [64][ 930/ 1200] Overall Loss 0.267714 Objective Loss 0.267714 LR 0.001000 Time 0.020609 -2022-12-06 10:51:09,969 - Epoch: [64][ 940/ 1200] Overall Loss 0.267758 Objective Loss 0.267758 LR 0.001000 Time 0.020604 -2022-12-06 10:51:10,167 - Epoch: [64][ 950/ 1200] Overall Loss 0.267785 Objective Loss 0.267785 LR 0.001000 Time 0.020594 -2022-12-06 10:51:10,366 - Epoch: [64][ 960/ 1200] Overall Loss 0.267618 Objective Loss 0.267618 LR 0.001000 Time 0.020587 -2022-12-06 10:51:10,564 - Epoch: [64][ 970/ 1200] Overall Loss 0.267547 Objective Loss 0.267547 LR 0.001000 Time 0.020579 -2022-12-06 10:51:10,766 - Epoch: [64][ 980/ 1200] Overall Loss 0.267800 Objective Loss 0.267800 LR 0.001000 Time 0.020574 -2022-12-06 10:51:10,964 - Epoch: [64][ 990/ 1200] Overall Loss 0.268028 Objective Loss 0.268028 LR 0.001000 Time 0.020565 -2022-12-06 10:51:11,166 - Epoch: [64][ 1000/ 1200] Overall Loss 0.268022 Objective Loss 0.268022 LR 0.001000 Time 0.020561 -2022-12-06 10:51:11,364 - Epoch: [64][ 1010/ 1200] Overall Loss 0.268177 Objective Loss 0.268177 LR 0.001000 Time 0.020553 -2022-12-06 10:51:11,565 - Epoch: [64][ 1020/ 1200] Overall Loss 0.268490 Objective Loss 0.268490 LR 0.001000 Time 0.020548 -2022-12-06 10:51:11,763 - Epoch: [64][ 1030/ 1200] Overall Loss 0.268463 Objective Loss 0.268463 LR 0.001000 Time 0.020540 -2022-12-06 10:51:11,963 - Epoch: [64][ 1040/ 1200] Overall Loss 0.268227 Objective Loss 0.268227 LR 0.001000 Time 0.020534 -2022-12-06 10:51:12,160 - Epoch: [64][ 1050/ 1200] Overall Loss 0.268489 Objective Loss 0.268489 LR 0.001000 Time 0.020526 -2022-12-06 10:51:12,360 - Epoch: [64][ 1060/ 1200] Overall Loss 0.268152 Objective Loss 0.268152 LR 0.001000 Time 0.020521 -2022-12-06 10:51:12,557 - Epoch: [64][ 1070/ 1200] Overall Loss 0.267914 Objective Loss 0.267914 LR 0.001000 Time 0.020513 -2022-12-06 10:51:12,758 - Epoch: [64][ 1080/ 1200] Overall Loss 0.267963 Objective Loss 0.267963 LR 0.001000 Time 0.020508 -2022-12-06 10:51:12,956 - Epoch: [64][ 1090/ 1200] Overall Loss 0.268138 Objective Loss 0.268138 LR 0.001000 Time 0.020501 -2022-12-06 10:51:13,158 - Epoch: [64][ 1100/ 1200] Overall Loss 0.268272 Objective Loss 0.268272 LR 0.001000 Time 0.020498 -2022-12-06 10:51:13,355 - Epoch: [64][ 1110/ 1200] Overall Loss 0.268322 Objective Loss 0.268322 LR 0.001000 Time 0.020490 -2022-12-06 10:51:13,557 - Epoch: [64][ 1120/ 1200] Overall Loss 0.268253 Objective Loss 0.268253 LR 0.001000 Time 0.020487 -2022-12-06 10:51:13,755 - Epoch: [64][ 1130/ 1200] Overall Loss 0.268448 Objective Loss 0.268448 LR 0.001000 Time 0.020480 -2022-12-06 10:51:13,957 - Epoch: [64][ 1140/ 1200] Overall Loss 0.268288 Objective Loss 0.268288 LR 0.001000 Time 0.020478 -2022-12-06 10:51:14,154 - Epoch: [64][ 1150/ 1200] Overall Loss 0.268204 Objective Loss 0.268204 LR 0.001000 Time 0.020471 -2022-12-06 10:51:14,356 - Epoch: [64][ 1160/ 1200] Overall Loss 0.268202 Objective Loss 0.268202 LR 0.001000 Time 0.020468 -2022-12-06 10:51:14,554 - Epoch: [64][ 1170/ 1200] Overall Loss 0.267997 Objective Loss 0.267997 LR 0.001000 Time 0.020461 -2022-12-06 10:51:14,754 - Epoch: [64][ 1180/ 1200] Overall Loss 0.267876 Objective Loss 0.267876 LR 0.001000 Time 0.020457 -2022-12-06 10:51:14,952 - Epoch: [64][ 1190/ 1200] Overall Loss 0.267932 Objective Loss 0.267932 LR 0.001000 Time 0.020451 -2022-12-06 10:51:15,187 - Epoch: [64][ 1200/ 1200] Overall Loss 0.267904 Objective Loss 0.267904 Top1 86.401674 Top5 97.698745 LR 0.001000 Time 0.020476 -2022-12-06 10:51:15,275 - --- validate (epoch=64)----------- -2022-12-06 10:51:15,276 - 34129 samples (256 per mini-batch) -2022-12-06 10:51:15,721 - Epoch: [64][ 10/ 134] Loss 0.287182 Top1 83.828125 Top5 97.851562 -2022-12-06 10:51:15,856 - Epoch: [64][ 20/ 134] Loss 0.296811 Top1 83.925781 Top5 97.929688 -2022-12-06 10:51:15,992 - Epoch: [64][ 30/ 134] Loss 0.285329 Top1 84.388021 Top5 98.138021 -2022-12-06 10:51:16,124 - Epoch: [64][ 40/ 134] Loss 0.286503 Top1 84.345703 Top5 98.066406 -2022-12-06 10:51:16,256 - Epoch: [64][ 50/ 134] Loss 0.283459 Top1 84.546875 Top5 98.101562 -2022-12-06 10:51:16,383 - Epoch: [64][ 60/ 134] Loss 0.281701 Top1 84.700521 Top5 98.164062 -2022-12-06 10:51:16,514 - Epoch: [64][ 70/ 134] Loss 0.287550 Top1 84.598214 Top5 98.069196 -2022-12-06 10:51:16,645 - Epoch: [64][ 80/ 134] Loss 0.284435 Top1 84.633789 Top5 98.046875 -2022-12-06 10:51:16,777 - Epoch: [64][ 90/ 134] Loss 0.281019 Top1 84.657118 Top5 98.033854 -2022-12-06 10:51:16,907 - Epoch: [64][ 100/ 134] Loss 0.282238 Top1 84.585938 Top5 98.011719 -2022-12-06 10:51:17,040 - Epoch: [64][ 110/ 134] Loss 0.284769 Top1 84.538352 Top5 97.982955 -2022-12-06 10:51:17,170 - Epoch: [64][ 120/ 134] Loss 0.283200 Top1 84.576823 Top5 97.988281 -2022-12-06 10:51:17,305 - Epoch: [64][ 130/ 134] Loss 0.286078 Top1 84.513221 Top5 97.956731 -2022-12-06 10:51:17,344 - Epoch: [64][ 134/ 134] Loss 0.288394 Top1 84.409154 Top5 97.937238 -2022-12-06 10:51:17,431 - ==> Top1: 84.409 Top5: 97.937 Loss: 0.288 - -2022-12-06 10:51:17,432 - ==> Confusion: -[[ 888 2 3 2 4 5 1 2 8 58 1 2 3 3 6 1 2 0 0 2 3] - [ 1 920 2 2 12 27 4 15 0 0 6 7 1 1 1 0 5 2 14 2 5] - [ 1 4 1016 11 6 2 21 6 0 2 5 2 3 4 3 1 3 1 2 2 8] - [ 0 3 22 935 1 3 0 1 1 0 10 0 6 2 12 0 2 4 13 0 5] - [ 10 3 1 0 951 7 0 1 0 6 2 1 0 1 11 3 12 3 1 3 4] - [ 3 12 0 5 4 953 2 27 3 1 1 21 1 20 3 1 4 1 1 4 2] - [ 1 3 20 1 0 4 1060 5 0 0 3 2 1 2 0 5 0 1 0 9 1] - [ 1 9 17 1 3 33 9 923 0 0 4 8 1 2 0 1 1 0 21 16 4] - [ 5 2 0 0 0 3 0 2 969 39 14 2 1 7 12 1 1 1 0 2 3] - [ 70 0 2 0 3 2 0 3 33 867 3 1 0 8 3 2 0 0 0 1 3] - [ 1 1 8 8 0 1 2 3 9 0 954 3 2 10 4 0 1 0 5 2 5] - [ 2 1 2 0 0 9 3 5 0 0 1 977 14 5 1 5 1 10 1 12 2] - [ 1 0 1 2 2 1 0 5 0 0 1 36 872 2 4 9 1 22 0 5 5] - [ 0 4 0 1 2 10 0 3 10 17 5 7 4 947 0 0 3 1 0 3 6] - [ 9 5 4 12 5 2 0 1 20 4 3 3 1 3 1048 0 4 1 3 0 2] - [ 2 0 1 1 3 3 6 0 0 0 0 8 2 4 0 980 10 17 0 2 4] - [ 2 1 1 3 0 4 2 0 1 0 0 4 4 4 2 9 1025 1 0 6 3] - [ 3 1 1 6 0 1 3 2 1 3 0 6 11 2 2 9 1 983 0 0 1] - [ 1 3 3 17 2 5 1 32 2 1 9 2 4 2 7 0 0 0 912 3 2] - [ 1 1 6 2 1 6 7 5 0 0 0 17 7 7 1 4 2 2 1 1006 4] - [ 141 214 269 138 127 197 90 174 96 106 183 115 354 316 164 126 241 100 152 304 9619]] - -2022-12-06 10:51:18,105 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:51:18,106 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:51:18,112 - - -2022-12-06 10:51:18,112 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:51:19,049 - Epoch: [65][ 10/ 1200] Overall Loss 0.240764 Objective Loss 0.240764 LR 0.001000 Time 0.093619 -2022-12-06 10:51:19,250 - Epoch: [65][ 20/ 1200] Overall Loss 0.241706 Objective Loss 0.241706 LR 0.001000 Time 0.056848 -2022-12-06 10:51:19,443 - Epoch: [65][ 30/ 1200] Overall Loss 0.244221 Objective Loss 0.244221 LR 0.001000 Time 0.044307 -2022-12-06 10:51:19,635 - Epoch: [65][ 40/ 1200] Overall Loss 0.244725 Objective Loss 0.244725 LR 0.001000 Time 0.038023 -2022-12-06 10:51:19,827 - Epoch: [65][ 50/ 1200] Overall Loss 0.245119 Objective Loss 0.245119 LR 0.001000 Time 0.034250 -2022-12-06 10:51:20,019 - Epoch: [65][ 60/ 1200] Overall Loss 0.249585 Objective Loss 0.249585 LR 0.001000 Time 0.031730 -2022-12-06 10:51:20,212 - Epoch: [65][ 70/ 1200] Overall Loss 0.252430 Objective Loss 0.252430 LR 0.001000 Time 0.029941 -2022-12-06 10:51:20,404 - Epoch: [65][ 80/ 1200] Overall Loss 0.253525 Objective Loss 0.253525 LR 0.001000 Time 0.028591 -2022-12-06 10:51:20,595 - Epoch: [65][ 90/ 1200] Overall Loss 0.253746 Objective Loss 0.253746 LR 0.001000 Time 0.027539 -2022-12-06 10:51:20,788 - Epoch: [65][ 100/ 1200] Overall Loss 0.258209 Objective Loss 0.258209 LR 0.001000 Time 0.026703 -2022-12-06 10:51:20,980 - Epoch: [65][ 110/ 1200] Overall Loss 0.257521 Objective Loss 0.257521 LR 0.001000 Time 0.026018 -2022-12-06 10:51:21,172 - Epoch: [65][ 120/ 1200] Overall Loss 0.257657 Objective Loss 0.257657 LR 0.001000 Time 0.025446 -2022-12-06 10:51:21,365 - Epoch: [65][ 130/ 1200] Overall Loss 0.258469 Objective Loss 0.258469 LR 0.001000 Time 0.024968 -2022-12-06 10:51:21,557 - Epoch: [65][ 140/ 1200] Overall Loss 0.257315 Objective Loss 0.257315 LR 0.001000 Time 0.024549 -2022-12-06 10:51:21,749 - Epoch: [65][ 150/ 1200] Overall Loss 0.256345 Objective Loss 0.256345 LR 0.001000 Time 0.024190 -2022-12-06 10:51:21,941 - Epoch: [65][ 160/ 1200] Overall Loss 0.257093 Objective Loss 0.257093 LR 0.001000 Time 0.023877 -2022-12-06 10:51:22,134 - Epoch: [65][ 170/ 1200] Overall Loss 0.256954 Objective Loss 0.256954 LR 0.001000 Time 0.023601 -2022-12-06 10:51:22,326 - Epoch: [65][ 180/ 1200] Overall Loss 0.257564 Objective Loss 0.257564 LR 0.001000 Time 0.023355 -2022-12-06 10:51:22,519 - Epoch: [65][ 190/ 1200] Overall Loss 0.258222 Objective Loss 0.258222 LR 0.001000 Time 0.023139 -2022-12-06 10:51:22,711 - Epoch: [65][ 200/ 1200] Overall Loss 0.258165 Objective Loss 0.258165 LR 0.001000 Time 0.022942 -2022-12-06 10:51:22,904 - Epoch: [65][ 210/ 1200] Overall Loss 0.258507 Objective Loss 0.258507 LR 0.001000 Time 0.022762 -2022-12-06 10:51:23,096 - Epoch: [65][ 220/ 1200] Overall Loss 0.258871 Objective Loss 0.258871 LR 0.001000 Time 0.022599 -2022-12-06 10:51:23,288 - Epoch: [65][ 230/ 1200] Overall Loss 0.259634 Objective Loss 0.259634 LR 0.001000 Time 0.022451 -2022-12-06 10:51:23,480 - Epoch: [65][ 240/ 1200] Overall Loss 0.260004 Objective Loss 0.260004 LR 0.001000 Time 0.022312 -2022-12-06 10:51:23,678 - Epoch: [65][ 250/ 1200] Overall Loss 0.260198 Objective Loss 0.260198 LR 0.001000 Time 0.022208 -2022-12-06 10:51:23,874 - Epoch: [65][ 260/ 1200] Overall Loss 0.260978 Objective Loss 0.260978 LR 0.001000 Time 0.022108 -2022-12-06 10:51:24,074 - Epoch: [65][ 270/ 1200] Overall Loss 0.261571 Objective Loss 0.261571 LR 0.001000 Time 0.022025 -2022-12-06 10:51:24,270 - Epoch: [65][ 280/ 1200] Overall Loss 0.260983 Objective Loss 0.260983 LR 0.001000 Time 0.021939 -2022-12-06 10:51:24,470 - Epoch: [65][ 290/ 1200] Overall Loss 0.261901 Objective Loss 0.261901 LR 0.001000 Time 0.021869 -2022-12-06 10:51:24,666 - Epoch: [65][ 300/ 1200] Overall Loss 0.262458 Objective Loss 0.262458 LR 0.001000 Time 0.021791 -2022-12-06 10:51:24,865 - Epoch: [65][ 310/ 1200] Overall Loss 0.262366 Objective Loss 0.262366 LR 0.001000 Time 0.021730 -2022-12-06 10:51:25,062 - Epoch: [65][ 320/ 1200] Overall Loss 0.262061 Objective Loss 0.262061 LR 0.001000 Time 0.021662 -2022-12-06 10:51:25,261 - Epoch: [65][ 330/ 1200] Overall Loss 0.262042 Objective Loss 0.262042 LR 0.001000 Time 0.021609 -2022-12-06 10:51:25,457 - Epoch: [65][ 340/ 1200] Overall Loss 0.261753 Objective Loss 0.261753 LR 0.001000 Time 0.021548 -2022-12-06 10:51:25,656 - Epoch: [65][ 350/ 1200] Overall Loss 0.261748 Objective Loss 0.261748 LR 0.001000 Time 0.021500 -2022-12-06 10:51:25,854 - Epoch: [65][ 360/ 1200] Overall Loss 0.262073 Objective Loss 0.262073 LR 0.001000 Time 0.021449 -2022-12-06 10:51:26,053 - Epoch: [65][ 370/ 1200] Overall Loss 0.262152 Objective Loss 0.262152 LR 0.001000 Time 0.021406 -2022-12-06 10:51:26,249 - Epoch: [65][ 380/ 1200] Overall Loss 0.261751 Objective Loss 0.261751 LR 0.001000 Time 0.021358 -2022-12-06 10:51:26,448 - Epoch: [65][ 390/ 1200] Overall Loss 0.262316 Objective Loss 0.262316 LR 0.001000 Time 0.021319 -2022-12-06 10:51:26,644 - Epoch: [65][ 400/ 1200] Overall Loss 0.262650 Objective Loss 0.262650 LR 0.001000 Time 0.021275 -2022-12-06 10:51:26,844 - Epoch: [65][ 410/ 1200] Overall Loss 0.263068 Objective Loss 0.263068 LR 0.001000 Time 0.021241 -2022-12-06 10:51:27,040 - Epoch: [65][ 420/ 1200] Overall Loss 0.262485 Objective Loss 0.262485 LR 0.001000 Time 0.021203 -2022-12-06 10:51:27,240 - Epoch: [65][ 430/ 1200] Overall Loss 0.262531 Objective Loss 0.262531 LR 0.001000 Time 0.021171 -2022-12-06 10:51:27,435 - Epoch: [65][ 440/ 1200] Overall Loss 0.263999 Objective Loss 0.263999 LR 0.001000 Time 0.021133 -2022-12-06 10:51:27,634 - Epoch: [65][ 450/ 1200] Overall Loss 0.263878 Objective Loss 0.263878 LR 0.001000 Time 0.021105 -2022-12-06 10:51:27,830 - Epoch: [65][ 460/ 1200] Overall Loss 0.263271 Objective Loss 0.263271 LR 0.001000 Time 0.021071 -2022-12-06 10:51:28,030 - Epoch: [65][ 470/ 1200] Overall Loss 0.263396 Objective Loss 0.263396 LR 0.001000 Time 0.021046 -2022-12-06 10:51:28,226 - Epoch: [65][ 480/ 1200] Overall Loss 0.263162 Objective Loss 0.263162 LR 0.001000 Time 0.021015 -2022-12-06 10:51:28,425 - Epoch: [65][ 490/ 1200] Overall Loss 0.263717 Objective Loss 0.263717 LR 0.001000 Time 0.020992 -2022-12-06 10:51:28,621 - Epoch: [65][ 500/ 1200] Overall Loss 0.263835 Objective Loss 0.263835 LR 0.001000 Time 0.020963 -2022-12-06 10:51:28,821 - Epoch: [65][ 510/ 1200] Overall Loss 0.263458 Objective Loss 0.263458 LR 0.001000 Time 0.020942 -2022-12-06 10:51:29,018 - Epoch: [65][ 520/ 1200] Overall Loss 0.264417 Objective Loss 0.264417 LR 0.001000 Time 0.020916 -2022-12-06 10:51:29,216 - Epoch: [65][ 530/ 1200] Overall Loss 0.264555 Objective Loss 0.264555 LR 0.001000 Time 0.020896 -2022-12-06 10:51:29,413 - Epoch: [65][ 540/ 1200] Overall Loss 0.263890 Objective Loss 0.263890 LR 0.001000 Time 0.020871 -2022-12-06 10:51:29,612 - Epoch: [65][ 550/ 1200] Overall Loss 0.263873 Objective Loss 0.263873 LR 0.001000 Time 0.020853 -2022-12-06 10:51:29,808 - Epoch: [65][ 560/ 1200] Overall Loss 0.263969 Objective Loss 0.263969 LR 0.001000 Time 0.020831 -2022-12-06 10:51:30,007 - Epoch: [65][ 570/ 1200] Overall Loss 0.263620 Objective Loss 0.263620 LR 0.001000 Time 0.020813 -2022-12-06 10:51:30,204 - Epoch: [65][ 580/ 1200] Overall Loss 0.263588 Objective Loss 0.263588 LR 0.001000 Time 0.020792 -2022-12-06 10:51:30,403 - Epoch: [65][ 590/ 1200] Overall Loss 0.263866 Objective Loss 0.263866 LR 0.001000 Time 0.020777 -2022-12-06 10:51:30,599 - Epoch: [65][ 600/ 1200] Overall Loss 0.264371 Objective Loss 0.264371 LR 0.001000 Time 0.020756 -2022-12-06 10:51:30,798 - Epoch: [65][ 610/ 1200] Overall Loss 0.264178 Objective Loss 0.264178 LR 0.001000 Time 0.020742 -2022-12-06 10:51:30,994 - Epoch: [65][ 620/ 1200] Overall Loss 0.264178 Objective Loss 0.264178 LR 0.001000 Time 0.020722 -2022-12-06 10:51:31,194 - Epoch: [65][ 630/ 1200] Overall Loss 0.263996 Objective Loss 0.263996 LR 0.001000 Time 0.020709 -2022-12-06 10:51:31,391 - Epoch: [65][ 640/ 1200] Overall Loss 0.263992 Objective Loss 0.263992 LR 0.001000 Time 0.020692 -2022-12-06 10:51:31,591 - Epoch: [65][ 650/ 1200] Overall Loss 0.263697 Objective Loss 0.263697 LR 0.001000 Time 0.020681 -2022-12-06 10:51:31,787 - Epoch: [65][ 660/ 1200] Overall Loss 0.263595 Objective Loss 0.263595 LR 0.001000 Time 0.020664 -2022-12-06 10:51:31,987 - Epoch: [65][ 670/ 1200] Overall Loss 0.263125 Objective Loss 0.263125 LR 0.001000 Time 0.020653 -2022-12-06 10:51:32,183 - Epoch: [65][ 680/ 1200] Overall Loss 0.263272 Objective Loss 0.263272 LR 0.001000 Time 0.020637 -2022-12-06 10:51:32,383 - Epoch: [65][ 690/ 1200] Overall Loss 0.263200 Objective Loss 0.263200 LR 0.001000 Time 0.020627 -2022-12-06 10:51:32,579 - Epoch: [65][ 700/ 1200] Overall Loss 0.263390 Objective Loss 0.263390 LR 0.001000 Time 0.020612 -2022-12-06 10:51:32,778 - Epoch: [65][ 710/ 1200] Overall Loss 0.263330 Objective Loss 0.263330 LR 0.001000 Time 0.020601 -2022-12-06 10:51:32,974 - Epoch: [65][ 720/ 1200] Overall Loss 0.263248 Objective Loss 0.263248 LR 0.001000 Time 0.020586 -2022-12-06 10:51:33,174 - Epoch: [65][ 730/ 1200] Overall Loss 0.263797 Objective Loss 0.263797 LR 0.001000 Time 0.020576 -2022-12-06 10:51:33,370 - Epoch: [65][ 740/ 1200] Overall Loss 0.263774 Objective Loss 0.263774 LR 0.001000 Time 0.020563 -2022-12-06 10:51:33,569 - Epoch: [65][ 750/ 1200] Overall Loss 0.264010 Objective Loss 0.264010 LR 0.001000 Time 0.020554 -2022-12-06 10:51:33,765 - Epoch: [65][ 760/ 1200] Overall Loss 0.264028 Objective Loss 0.264028 LR 0.001000 Time 0.020541 -2022-12-06 10:51:33,964 - Epoch: [65][ 770/ 1200] Overall Loss 0.264156 Objective Loss 0.264156 LR 0.001000 Time 0.020532 -2022-12-06 10:51:34,160 - Epoch: [65][ 780/ 1200] Overall Loss 0.264420 Objective Loss 0.264420 LR 0.001000 Time 0.020519 -2022-12-06 10:51:34,359 - Epoch: [65][ 790/ 1200] Overall Loss 0.264588 Objective Loss 0.264588 LR 0.001000 Time 0.020510 -2022-12-06 10:51:34,556 - Epoch: [65][ 800/ 1200] Overall Loss 0.264546 Objective Loss 0.264546 LR 0.001000 Time 0.020499 -2022-12-06 10:51:34,755 - Epoch: [65][ 810/ 1200] Overall Loss 0.264788 Objective Loss 0.264788 LR 0.001000 Time 0.020491 -2022-12-06 10:51:34,951 - Epoch: [65][ 820/ 1200] Overall Loss 0.265096 Objective Loss 0.265096 LR 0.001000 Time 0.020480 -2022-12-06 10:51:35,150 - Epoch: [65][ 830/ 1200] Overall Loss 0.265262 Objective Loss 0.265262 LR 0.001000 Time 0.020472 -2022-12-06 10:51:35,346 - Epoch: [65][ 840/ 1200] Overall Loss 0.265476 Objective Loss 0.265476 LR 0.001000 Time 0.020461 -2022-12-06 10:51:35,546 - Epoch: [65][ 850/ 1200] Overall Loss 0.265815 Objective Loss 0.265815 LR 0.001000 Time 0.020454 -2022-12-06 10:51:35,741 - Epoch: [65][ 860/ 1200] Overall Loss 0.265687 Objective Loss 0.265687 LR 0.001000 Time 0.020444 -2022-12-06 10:51:35,941 - Epoch: [65][ 870/ 1200] Overall Loss 0.265865 Objective Loss 0.265865 LR 0.001000 Time 0.020437 -2022-12-06 10:51:36,137 - Epoch: [65][ 880/ 1200] Overall Loss 0.266027 Objective Loss 0.266027 LR 0.001000 Time 0.020428 -2022-12-06 10:51:36,337 - Epoch: [65][ 890/ 1200] Overall Loss 0.265846 Objective Loss 0.265846 LR 0.001000 Time 0.020421 -2022-12-06 10:51:36,533 - Epoch: [65][ 900/ 1200] Overall Loss 0.265984 Objective Loss 0.265984 LR 0.001000 Time 0.020412 -2022-12-06 10:51:36,732 - Epoch: [65][ 910/ 1200] Overall Loss 0.266068 Objective Loss 0.266068 LR 0.001000 Time 0.020406 -2022-12-06 10:51:36,928 - Epoch: [65][ 920/ 1200] Overall Loss 0.265983 Objective Loss 0.265983 LR 0.001000 Time 0.020397 -2022-12-06 10:51:37,128 - Epoch: [65][ 930/ 1200] Overall Loss 0.265902 Objective Loss 0.265902 LR 0.001000 Time 0.020391 -2022-12-06 10:51:37,324 - Epoch: [65][ 940/ 1200] Overall Loss 0.265961 Objective Loss 0.265961 LR 0.001000 Time 0.020382 -2022-12-06 10:51:37,523 - Epoch: [65][ 950/ 1200] Overall Loss 0.265964 Objective Loss 0.265964 LR 0.001000 Time 0.020377 -2022-12-06 10:51:37,719 - Epoch: [65][ 960/ 1200] Overall Loss 0.266151 Objective Loss 0.266151 LR 0.001000 Time 0.020368 -2022-12-06 10:51:37,918 - Epoch: [65][ 970/ 1200] Overall Loss 0.266233 Objective Loss 0.266233 LR 0.001000 Time 0.020363 -2022-12-06 10:51:38,114 - Epoch: [65][ 980/ 1200] Overall Loss 0.266405 Objective Loss 0.266405 LR 0.001000 Time 0.020355 -2022-12-06 10:51:38,313 - Epoch: [65][ 990/ 1200] Overall Loss 0.266458 Objective Loss 0.266458 LR 0.001000 Time 0.020350 -2022-12-06 10:51:38,510 - Epoch: [65][ 1000/ 1200] Overall Loss 0.266379 Objective Loss 0.266379 LR 0.001000 Time 0.020342 -2022-12-06 10:51:38,709 - Epoch: [65][ 1010/ 1200] Overall Loss 0.266268 Objective Loss 0.266268 LR 0.001000 Time 0.020337 -2022-12-06 10:51:38,906 - Epoch: [65][ 1020/ 1200] Overall Loss 0.266328 Objective Loss 0.266328 LR 0.001000 Time 0.020330 -2022-12-06 10:51:39,104 - Epoch: [65][ 1030/ 1200] Overall Loss 0.266763 Objective Loss 0.266763 LR 0.001000 Time 0.020325 -2022-12-06 10:51:39,301 - Epoch: [65][ 1040/ 1200] Overall Loss 0.267046 Objective Loss 0.267046 LR 0.001000 Time 0.020318 -2022-12-06 10:51:39,499 - Epoch: [65][ 1050/ 1200] Overall Loss 0.267348 Objective Loss 0.267348 LR 0.001000 Time 0.020313 -2022-12-06 10:51:39,696 - Epoch: [65][ 1060/ 1200] Overall Loss 0.267634 Objective Loss 0.267634 LR 0.001000 Time 0.020306 -2022-12-06 10:51:39,895 - Epoch: [65][ 1070/ 1200] Overall Loss 0.267581 Objective Loss 0.267581 LR 0.001000 Time 0.020303 -2022-12-06 10:51:40,092 - Epoch: [65][ 1080/ 1200] Overall Loss 0.267429 Objective Loss 0.267429 LR 0.001000 Time 0.020296 -2022-12-06 10:51:40,292 - Epoch: [65][ 1090/ 1200] Overall Loss 0.267581 Objective Loss 0.267581 LR 0.001000 Time 0.020292 -2022-12-06 10:51:40,489 - Epoch: [65][ 1100/ 1200] Overall Loss 0.267505 Objective Loss 0.267505 LR 0.001000 Time 0.020287 -2022-12-06 10:51:40,688 - Epoch: [65][ 1110/ 1200] Overall Loss 0.267688 Objective Loss 0.267688 LR 0.001000 Time 0.020283 -2022-12-06 10:51:40,884 - Epoch: [65][ 1120/ 1200] Overall Loss 0.267681 Objective Loss 0.267681 LR 0.001000 Time 0.020277 -2022-12-06 10:51:41,084 - Epoch: [65][ 1130/ 1200] Overall Loss 0.267375 Objective Loss 0.267375 LR 0.001000 Time 0.020273 -2022-12-06 10:51:41,280 - Epoch: [65][ 1140/ 1200] Overall Loss 0.267363 Objective Loss 0.267363 LR 0.001000 Time 0.020267 -2022-12-06 10:51:41,479 - Epoch: [65][ 1150/ 1200] Overall Loss 0.267426 Objective Loss 0.267426 LR 0.001000 Time 0.020264 -2022-12-06 10:51:41,676 - Epoch: [65][ 1160/ 1200] Overall Loss 0.267490 Objective Loss 0.267490 LR 0.001000 Time 0.020258 -2022-12-06 10:51:41,876 - Epoch: [65][ 1170/ 1200] Overall Loss 0.267428 Objective Loss 0.267428 LR 0.001000 Time 0.020255 -2022-12-06 10:51:42,073 - Epoch: [65][ 1180/ 1200] Overall Loss 0.267130 Objective Loss 0.267130 LR 0.001000 Time 0.020250 -2022-12-06 10:51:42,272 - Epoch: [65][ 1190/ 1200] Overall Loss 0.267211 Objective Loss 0.267211 LR 0.001000 Time 0.020247 -2022-12-06 10:51:42,504 - Epoch: [65][ 1200/ 1200] Overall Loss 0.267320 Objective Loss 0.267320 Top1 84.518828 Top5 97.907950 LR 0.001000 Time 0.020271 -2022-12-06 10:51:42,596 - --- validate (epoch=65)----------- -2022-12-06 10:51:42,596 - 34129 samples (256 per mini-batch) -2022-12-06 10:51:43,041 - Epoch: [65][ 10/ 134] Loss 0.335352 Top1 83.515625 Top5 97.109375 -2022-12-06 10:51:43,168 - Epoch: [65][ 20/ 134] Loss 0.327194 Top1 83.808594 Top5 97.558594 -2022-12-06 10:51:43,292 - Epoch: [65][ 30/ 134] Loss 0.311380 Top1 84.062500 Top5 97.773438 -2022-12-06 10:51:43,418 - Epoch: [65][ 40/ 134] Loss 0.317409 Top1 83.945312 Top5 97.607422 -2022-12-06 10:51:43,545 - Epoch: [65][ 50/ 134] Loss 0.315667 Top1 84.070312 Top5 97.656250 -2022-12-06 10:51:43,672 - Epoch: [65][ 60/ 134] Loss 0.309944 Top1 84.251302 Top5 97.740885 -2022-12-06 10:51:43,798 - Epoch: [65][ 70/ 134] Loss 0.303084 Top1 84.386161 Top5 97.790179 -2022-12-06 10:51:43,926 - Epoch: [65][ 80/ 134] Loss 0.302905 Top1 84.467773 Top5 97.827148 -2022-12-06 10:51:44,052 - Epoch: [65][ 90/ 134] Loss 0.299804 Top1 84.531250 Top5 97.838542 -2022-12-06 10:51:44,181 - Epoch: [65][ 100/ 134] Loss 0.297205 Top1 84.488281 Top5 97.851562 -2022-12-06 10:51:44,312 - Epoch: [65][ 110/ 134] Loss 0.293557 Top1 84.531250 Top5 97.862216 -2022-12-06 10:51:44,441 - Epoch: [65][ 120/ 134] Loss 0.292848 Top1 84.524740 Top5 97.900391 -2022-12-06 10:51:44,572 - Epoch: [65][ 130/ 134] Loss 0.294199 Top1 84.477163 Top5 97.890625 -2022-12-06 10:51:44,609 - Epoch: [65][ 134/ 134] Loss 0.292576 Top1 84.499985 Top5 97.887427 -2022-12-06 10:51:44,704 - ==> Top1: 84.500 Top5: 97.887 Loss: 0.293 - -2022-12-06 10:51:44,704 - ==> Confusion: -[[ 921 3 1 3 4 3 0 3 6 37 0 1 2 2 3 2 1 0 2 1 1] - [ 0 911 2 2 7 41 5 15 1 0 6 2 3 1 0 2 5 2 9 3 10] - [ 5 6 977 24 5 3 30 13 0 1 6 4 3 2 0 2 0 0 4 3 15] - [ 2 1 11 934 1 4 0 2 3 0 14 1 7 4 12 0 1 1 16 2 4] - [ 13 5 1 1 940 11 1 4 2 7 1 3 2 2 6 6 9 2 0 2 2] - [ 5 8 0 5 5 975 2 15 2 1 1 8 4 18 2 1 1 2 2 5 7] - [ 0 2 12 2 0 6 1070 2 0 0 4 5 1 0 0 3 1 2 2 5 1] - [ 0 5 3 3 3 41 6 933 0 0 3 7 1 4 2 2 1 1 26 8 5] - [ 6 1 0 1 0 2 0 1 958 47 15 2 3 7 9 0 2 1 6 2 1] - [ 104 0 0 1 6 1 0 3 22 838 0 1 1 13 1 1 0 2 1 0 6] - [ 0 2 3 6 2 1 0 3 9 0 960 2 2 9 3 1 0 0 8 2 6] - [ 4 1 2 0 0 11 6 2 0 0 1 968 19 8 0 4 1 2 2 16 4] - [ 0 2 2 4 0 2 0 1 1 0 1 42 885 0 0 8 0 13 3 3 2] - [ 0 1 0 0 2 10 0 0 8 11 6 8 3 956 1 1 3 2 2 3 6] - [ 13 1 2 23 1 2 1 1 26 3 1 2 3 7 1028 1 2 0 3 1 9] - [ 2 1 3 2 2 0 5 0 0 1 0 12 4 5 0 974 6 16 0 4 6] - [ 1 3 1 0 2 1 1 1 0 0 1 7 3 1 1 13 1020 2 1 9 4] - [ 5 0 1 1 0 0 1 1 1 0 0 16 24 2 3 12 1 962 0 4 2] - [ 2 3 4 11 0 3 1 24 2 1 6 1 1 0 6 1 0 0 932 6 4] - [ 1 3 1 0 0 11 12 6 0 0 1 17 4 7 0 5 2 3 2 1000 5] - [ 154 167 129 151 106 270 98 129 70 87 190 170 475 292 138 119 204 79 209 296 9693]] - -2022-12-06 10:51:45,365 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:51:45,365 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:51:45,371 - - -2022-12-06 10:51:45,371 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:51:46,293 - Epoch: [66][ 10/ 1200] Overall Loss 0.254574 Objective Loss 0.254574 LR 0.001000 Time 0.092123 -2022-12-06 10:51:46,495 - Epoch: [66][ 20/ 1200] Overall Loss 0.271773 Objective Loss 0.271773 LR 0.001000 Time 0.056127 -2022-12-06 10:51:46,687 - Epoch: [66][ 30/ 1200] Overall Loss 0.273194 Objective Loss 0.273194 LR 0.001000 Time 0.043799 -2022-12-06 10:51:46,878 - Epoch: [66][ 40/ 1200] Overall Loss 0.264305 Objective Loss 0.264305 LR 0.001000 Time 0.037622 -2022-12-06 10:51:47,067 - Epoch: [66][ 50/ 1200] Overall Loss 0.264509 Objective Loss 0.264509 LR 0.001000 Time 0.033873 -2022-12-06 10:51:47,258 - Epoch: [66][ 60/ 1200] Overall Loss 0.263639 Objective Loss 0.263639 LR 0.001000 Time 0.031394 -2022-12-06 10:51:47,448 - Epoch: [66][ 70/ 1200] Overall Loss 0.262572 Objective Loss 0.262572 LR 0.001000 Time 0.029623 -2022-12-06 10:51:47,639 - Epoch: [66][ 80/ 1200] Overall Loss 0.260725 Objective Loss 0.260725 LR 0.001000 Time 0.028292 -2022-12-06 10:51:47,829 - Epoch: [66][ 90/ 1200] Overall Loss 0.259519 Objective Loss 0.259519 LR 0.001000 Time 0.027259 -2022-12-06 10:51:48,020 - Epoch: [66][ 100/ 1200] Overall Loss 0.259394 Objective Loss 0.259394 LR 0.001000 Time 0.026442 -2022-12-06 10:51:48,210 - Epoch: [66][ 110/ 1200] Overall Loss 0.257035 Objective Loss 0.257035 LR 0.001000 Time 0.025758 -2022-12-06 10:51:48,401 - Epoch: [66][ 120/ 1200] Overall Loss 0.260923 Objective Loss 0.260923 LR 0.001000 Time 0.025197 -2022-12-06 10:51:48,592 - Epoch: [66][ 130/ 1200] Overall Loss 0.259672 Objective Loss 0.259672 LR 0.001000 Time 0.024725 -2022-12-06 10:51:48,783 - Epoch: [66][ 140/ 1200] Overall Loss 0.258930 Objective Loss 0.258930 LR 0.001000 Time 0.024315 -2022-12-06 10:51:48,973 - Epoch: [66][ 150/ 1200] Overall Loss 0.259070 Objective Loss 0.259070 LR 0.001000 Time 0.023959 -2022-12-06 10:51:49,165 - Epoch: [66][ 160/ 1200] Overall Loss 0.259980 Objective Loss 0.259980 LR 0.001000 Time 0.023657 -2022-12-06 10:51:49,355 - Epoch: [66][ 170/ 1200] Overall Loss 0.261025 Objective Loss 0.261025 LR 0.001000 Time 0.023379 -2022-12-06 10:51:49,544 - Epoch: [66][ 180/ 1200] Overall Loss 0.260583 Objective Loss 0.260583 LR 0.001000 Time 0.023130 -2022-12-06 10:51:49,735 - Epoch: [66][ 190/ 1200] Overall Loss 0.259802 Objective Loss 0.259802 LR 0.001000 Time 0.022913 -2022-12-06 10:51:49,924 - Epoch: [66][ 200/ 1200] Overall Loss 0.261205 Objective Loss 0.261205 LR 0.001000 Time 0.022713 -2022-12-06 10:51:50,115 - Epoch: [66][ 210/ 1200] Overall Loss 0.261078 Objective Loss 0.261078 LR 0.001000 Time 0.022535 -2022-12-06 10:51:50,304 - Epoch: [66][ 220/ 1200] Overall Loss 0.261553 Objective Loss 0.261553 LR 0.001000 Time 0.022371 -2022-12-06 10:51:50,495 - Epoch: [66][ 230/ 1200] Overall Loss 0.261447 Objective Loss 0.261447 LR 0.001000 Time 0.022222 -2022-12-06 10:51:50,684 - Epoch: [66][ 240/ 1200] Overall Loss 0.262388 Objective Loss 0.262388 LR 0.001000 Time 0.022084 -2022-12-06 10:51:50,874 - Epoch: [66][ 250/ 1200] Overall Loss 0.262613 Objective Loss 0.262613 LR 0.001000 Time 0.021957 -2022-12-06 10:51:51,064 - Epoch: [66][ 260/ 1200] Overall Loss 0.262084 Objective Loss 0.262084 LR 0.001000 Time 0.021841 -2022-12-06 10:51:51,254 - Epoch: [66][ 270/ 1200] Overall Loss 0.262072 Objective Loss 0.262072 LR 0.001000 Time 0.021735 -2022-12-06 10:51:51,444 - Epoch: [66][ 280/ 1200] Overall Loss 0.262494 Objective Loss 0.262494 LR 0.001000 Time 0.021637 -2022-12-06 10:51:51,635 - Epoch: [66][ 290/ 1200] Overall Loss 0.262570 Objective Loss 0.262570 LR 0.001000 Time 0.021545 -2022-12-06 10:51:51,824 - Epoch: [66][ 300/ 1200] Overall Loss 0.261935 Objective Loss 0.261935 LR 0.001000 Time 0.021457 -2022-12-06 10:51:52,015 - Epoch: [66][ 310/ 1200] Overall Loss 0.261468 Objective Loss 0.261468 LR 0.001000 Time 0.021377 -2022-12-06 10:51:52,205 - Epoch: [66][ 320/ 1200] Overall Loss 0.262005 Objective Loss 0.262005 LR 0.001000 Time 0.021301 -2022-12-06 10:51:52,394 - Epoch: [66][ 330/ 1200] Overall Loss 0.263190 Objective Loss 0.263190 LR 0.001000 Time 0.021228 -2022-12-06 10:51:52,584 - Epoch: [66][ 340/ 1200] Overall Loss 0.263227 Objective Loss 0.263227 LR 0.001000 Time 0.021161 -2022-12-06 10:51:52,775 - Epoch: [66][ 350/ 1200] Overall Loss 0.262467 Objective Loss 0.262467 LR 0.001000 Time 0.021099 -2022-12-06 10:51:52,965 - Epoch: [66][ 360/ 1200] Overall Loss 0.263545 Objective Loss 0.263545 LR 0.001000 Time 0.021040 -2022-12-06 10:51:53,154 - Epoch: [66][ 370/ 1200] Overall Loss 0.264041 Objective Loss 0.264041 LR 0.001000 Time 0.020982 -2022-12-06 10:51:53,345 - Epoch: [66][ 380/ 1200] Overall Loss 0.264576 Objective Loss 0.264576 LR 0.001000 Time 0.020930 -2022-12-06 10:51:53,535 - Epoch: [66][ 390/ 1200] Overall Loss 0.265848 Objective Loss 0.265848 LR 0.001000 Time 0.020879 -2022-12-06 10:51:53,725 - Epoch: [66][ 400/ 1200] Overall Loss 0.266798 Objective Loss 0.266798 LR 0.001000 Time 0.020830 -2022-12-06 10:51:53,915 - Epoch: [66][ 410/ 1200] Overall Loss 0.267078 Objective Loss 0.267078 LR 0.001000 Time 0.020784 -2022-12-06 10:51:54,105 - Epoch: [66][ 420/ 1200] Overall Loss 0.267114 Objective Loss 0.267114 LR 0.001000 Time 0.020740 -2022-12-06 10:51:54,295 - Epoch: [66][ 430/ 1200] Overall Loss 0.267422 Objective Loss 0.267422 LR 0.001000 Time 0.020699 -2022-12-06 10:51:54,485 - Epoch: [66][ 440/ 1200] Overall Loss 0.266916 Objective Loss 0.266916 LR 0.001000 Time 0.020660 -2022-12-06 10:51:54,676 - Epoch: [66][ 450/ 1200] Overall Loss 0.266603 Objective Loss 0.266603 LR 0.001000 Time 0.020623 -2022-12-06 10:51:54,865 - Epoch: [66][ 460/ 1200] Overall Loss 0.266462 Objective Loss 0.266462 LR 0.001000 Time 0.020585 -2022-12-06 10:51:55,055 - Epoch: [66][ 470/ 1200] Overall Loss 0.266526 Objective Loss 0.266526 LR 0.001000 Time 0.020550 -2022-12-06 10:51:55,245 - Epoch: [66][ 480/ 1200] Overall Loss 0.266137 Objective Loss 0.266137 LR 0.001000 Time 0.020517 -2022-12-06 10:51:55,435 - Epoch: [66][ 490/ 1200] Overall Loss 0.265860 Objective Loss 0.265860 LR 0.001000 Time 0.020484 -2022-12-06 10:51:55,625 - Epoch: [66][ 500/ 1200] Overall Loss 0.265989 Objective Loss 0.265989 LR 0.001000 Time 0.020454 -2022-12-06 10:51:55,815 - Epoch: [66][ 510/ 1200] Overall Loss 0.265711 Objective Loss 0.265711 LR 0.001000 Time 0.020424 -2022-12-06 10:51:56,005 - Epoch: [66][ 520/ 1200] Overall Loss 0.265971 Objective Loss 0.265971 LR 0.001000 Time 0.020396 -2022-12-06 10:51:56,195 - Epoch: [66][ 530/ 1200] Overall Loss 0.266134 Objective Loss 0.266134 LR 0.001000 Time 0.020369 -2022-12-06 10:51:56,386 - Epoch: [66][ 540/ 1200] Overall Loss 0.266135 Objective Loss 0.266135 LR 0.001000 Time 0.020343 -2022-12-06 10:51:56,577 - Epoch: [66][ 550/ 1200] Overall Loss 0.266311 Objective Loss 0.266311 LR 0.001000 Time 0.020321 -2022-12-06 10:51:56,768 - Epoch: [66][ 560/ 1200] Overall Loss 0.266494 Objective Loss 0.266494 LR 0.001000 Time 0.020298 -2022-12-06 10:51:56,959 - Epoch: [66][ 570/ 1200] Overall Loss 0.266452 Objective Loss 0.266452 LR 0.001000 Time 0.020275 -2022-12-06 10:51:57,149 - Epoch: [66][ 580/ 1200] Overall Loss 0.266950 Objective Loss 0.266950 LR 0.001000 Time 0.020252 -2022-12-06 10:51:57,339 - Epoch: [66][ 590/ 1200] Overall Loss 0.266582 Objective Loss 0.266582 LR 0.001000 Time 0.020230 -2022-12-06 10:51:57,529 - Epoch: [66][ 600/ 1200] Overall Loss 0.266196 Objective Loss 0.266196 LR 0.001000 Time 0.020209 -2022-12-06 10:51:57,720 - Epoch: [66][ 610/ 1200] Overall Loss 0.265802 Objective Loss 0.265802 LR 0.001000 Time 0.020190 -2022-12-06 10:51:57,910 - Epoch: [66][ 620/ 1200] Overall Loss 0.265762 Objective Loss 0.265762 LR 0.001000 Time 0.020169 -2022-12-06 10:51:58,100 - Epoch: [66][ 630/ 1200] Overall Loss 0.265964 Objective Loss 0.265964 LR 0.001000 Time 0.020150 -2022-12-06 10:51:58,290 - Epoch: [66][ 640/ 1200] Overall Loss 0.265806 Objective Loss 0.265806 LR 0.001000 Time 0.020131 -2022-12-06 10:51:58,480 - Epoch: [66][ 650/ 1200] Overall Loss 0.265353 Objective Loss 0.265353 LR 0.001000 Time 0.020113 -2022-12-06 10:51:58,670 - Epoch: [66][ 660/ 1200] Overall Loss 0.265313 Objective Loss 0.265313 LR 0.001000 Time 0.020096 -2022-12-06 10:51:58,860 - Epoch: [66][ 670/ 1200] Overall Loss 0.265053 Objective Loss 0.265053 LR 0.001000 Time 0.020078 -2022-12-06 10:51:59,050 - Epoch: [66][ 680/ 1200] Overall Loss 0.264774 Objective Loss 0.264774 LR 0.001000 Time 0.020062 -2022-12-06 10:51:59,240 - Epoch: [66][ 690/ 1200] Overall Loss 0.264458 Objective Loss 0.264458 LR 0.001000 Time 0.020045 -2022-12-06 10:51:59,430 - Epoch: [66][ 700/ 1200] Overall Loss 0.264454 Objective Loss 0.264454 LR 0.001000 Time 0.020030 -2022-12-06 10:51:59,620 - Epoch: [66][ 710/ 1200] Overall Loss 0.264902 Objective Loss 0.264902 LR 0.001000 Time 0.020014 -2022-12-06 10:51:59,811 - Epoch: [66][ 720/ 1200] Overall Loss 0.264551 Objective Loss 0.264551 LR 0.001000 Time 0.020001 -2022-12-06 10:52:00,001 - Epoch: [66][ 730/ 1200] Overall Loss 0.264818 Objective Loss 0.264818 LR 0.001000 Time 0.019987 -2022-12-06 10:52:00,192 - Epoch: [66][ 740/ 1200] Overall Loss 0.264966 Objective Loss 0.264966 LR 0.001000 Time 0.019974 -2022-12-06 10:52:00,383 - Epoch: [66][ 750/ 1200] Overall Loss 0.264823 Objective Loss 0.264823 LR 0.001000 Time 0.019962 -2022-12-06 10:52:00,574 - Epoch: [66][ 760/ 1200] Overall Loss 0.264949 Objective Loss 0.264949 LR 0.001000 Time 0.019949 -2022-12-06 10:52:00,763 - Epoch: [66][ 770/ 1200] Overall Loss 0.264694 Objective Loss 0.264694 LR 0.001000 Time 0.019935 -2022-12-06 10:52:00,952 - Epoch: [66][ 780/ 1200] Overall Loss 0.264670 Objective Loss 0.264670 LR 0.001000 Time 0.019921 -2022-12-06 10:52:01,142 - Epoch: [66][ 790/ 1200] Overall Loss 0.264732 Objective Loss 0.264732 LR 0.001000 Time 0.019909 -2022-12-06 10:52:01,332 - Epoch: [66][ 800/ 1200] Overall Loss 0.264555 Objective Loss 0.264555 LR 0.001000 Time 0.019897 -2022-12-06 10:52:01,522 - Epoch: [66][ 810/ 1200] Overall Loss 0.264316 Objective Loss 0.264316 LR 0.001000 Time 0.019885 -2022-12-06 10:52:01,713 - Epoch: [66][ 820/ 1200] Overall Loss 0.264357 Objective Loss 0.264357 LR 0.001000 Time 0.019874 -2022-12-06 10:52:01,903 - Epoch: [66][ 830/ 1200] Overall Loss 0.264250 Objective Loss 0.264250 LR 0.001000 Time 0.019863 -2022-12-06 10:52:02,092 - Epoch: [66][ 840/ 1200] Overall Loss 0.264465 Objective Loss 0.264465 LR 0.001000 Time 0.019852 -2022-12-06 10:52:02,283 - Epoch: [66][ 850/ 1200] Overall Loss 0.264775 Objective Loss 0.264775 LR 0.001000 Time 0.019841 -2022-12-06 10:52:02,473 - Epoch: [66][ 860/ 1200] Overall Loss 0.264817 Objective Loss 0.264817 LR 0.001000 Time 0.019832 -2022-12-06 10:52:02,663 - Epoch: [66][ 870/ 1200] Overall Loss 0.265426 Objective Loss 0.265426 LR 0.001000 Time 0.019821 -2022-12-06 10:52:02,853 - Epoch: [66][ 880/ 1200] Overall Loss 0.265553 Objective Loss 0.265553 LR 0.001000 Time 0.019811 -2022-12-06 10:52:03,042 - Epoch: [66][ 890/ 1200] Overall Loss 0.265391 Objective Loss 0.265391 LR 0.001000 Time 0.019801 -2022-12-06 10:52:03,233 - Epoch: [66][ 900/ 1200] Overall Loss 0.265371 Objective Loss 0.265371 LR 0.001000 Time 0.019792 -2022-12-06 10:52:03,423 - Epoch: [66][ 910/ 1200] Overall Loss 0.265281 Objective Loss 0.265281 LR 0.001000 Time 0.019782 -2022-12-06 10:52:03,612 - Epoch: [66][ 920/ 1200] Overall Loss 0.265556 Objective Loss 0.265556 LR 0.001000 Time 0.019773 -2022-12-06 10:52:03,803 - Epoch: [66][ 930/ 1200] Overall Loss 0.265463 Objective Loss 0.265463 LR 0.001000 Time 0.019765 -2022-12-06 10:52:03,993 - Epoch: [66][ 940/ 1200] Overall Loss 0.265347 Objective Loss 0.265347 LR 0.001000 Time 0.019756 -2022-12-06 10:52:04,183 - Epoch: [66][ 950/ 1200] Overall Loss 0.265294 Objective Loss 0.265294 LR 0.001000 Time 0.019748 -2022-12-06 10:52:04,373 - Epoch: [66][ 960/ 1200] Overall Loss 0.265286 Objective Loss 0.265286 LR 0.001000 Time 0.019739 -2022-12-06 10:52:04,564 - Epoch: [66][ 970/ 1200] Overall Loss 0.265378 Objective Loss 0.265378 LR 0.001000 Time 0.019732 -2022-12-06 10:52:04,755 - Epoch: [66][ 980/ 1200] Overall Loss 0.265427 Objective Loss 0.265427 LR 0.001000 Time 0.019725 -2022-12-06 10:52:04,945 - Epoch: [66][ 990/ 1200] Overall Loss 0.265149 Objective Loss 0.265149 LR 0.001000 Time 0.019717 -2022-12-06 10:52:05,134 - Epoch: [66][ 1000/ 1200] Overall Loss 0.264881 Objective Loss 0.264881 LR 0.001000 Time 0.019709 -2022-12-06 10:52:05,325 - Epoch: [66][ 1010/ 1200] Overall Loss 0.264726 Objective Loss 0.264726 LR 0.001000 Time 0.019701 -2022-12-06 10:52:05,516 - Epoch: [66][ 1020/ 1200] Overall Loss 0.264636 Objective Loss 0.264636 LR 0.001000 Time 0.019695 -2022-12-06 10:52:05,707 - Epoch: [66][ 1030/ 1200] Overall Loss 0.264712 Objective Loss 0.264712 LR 0.001000 Time 0.019689 -2022-12-06 10:52:05,898 - Epoch: [66][ 1040/ 1200] Overall Loss 0.264555 Objective Loss 0.264555 LR 0.001000 Time 0.019682 -2022-12-06 10:52:06,088 - Epoch: [66][ 1050/ 1200] Overall Loss 0.264671 Objective Loss 0.264671 LR 0.001000 Time 0.019676 -2022-12-06 10:52:06,279 - Epoch: [66][ 1060/ 1200] Overall Loss 0.264731 Objective Loss 0.264731 LR 0.001000 Time 0.019670 -2022-12-06 10:52:06,470 - Epoch: [66][ 1070/ 1200] Overall Loss 0.264868 Objective Loss 0.264868 LR 0.001000 Time 0.019664 -2022-12-06 10:52:06,661 - Epoch: [66][ 1080/ 1200] Overall Loss 0.264758 Objective Loss 0.264758 LR 0.001000 Time 0.019658 -2022-12-06 10:52:06,851 - Epoch: [66][ 1090/ 1200] Overall Loss 0.265094 Objective Loss 0.265094 LR 0.001000 Time 0.019652 -2022-12-06 10:52:07,042 - Epoch: [66][ 1100/ 1200] Overall Loss 0.265010 Objective Loss 0.265010 LR 0.001000 Time 0.019646 -2022-12-06 10:52:07,233 - Epoch: [66][ 1110/ 1200] Overall Loss 0.264875 Objective Loss 0.264875 LR 0.001000 Time 0.019641 -2022-12-06 10:52:07,424 - Epoch: [66][ 1120/ 1200] Overall Loss 0.264882 Objective Loss 0.264882 LR 0.001000 Time 0.019636 -2022-12-06 10:52:07,615 - Epoch: [66][ 1130/ 1200] Overall Loss 0.265089 Objective Loss 0.265089 LR 0.001000 Time 0.019631 -2022-12-06 10:52:07,806 - Epoch: [66][ 1140/ 1200] Overall Loss 0.264898 Objective Loss 0.264898 LR 0.001000 Time 0.019625 -2022-12-06 10:52:07,996 - Epoch: [66][ 1150/ 1200] Overall Loss 0.264870 Objective Loss 0.264870 LR 0.001000 Time 0.019619 -2022-12-06 10:52:08,187 - Epoch: [66][ 1160/ 1200] Overall Loss 0.264804 Objective Loss 0.264804 LR 0.001000 Time 0.019614 -2022-12-06 10:52:08,378 - Epoch: [66][ 1170/ 1200] Overall Loss 0.264844 Objective Loss 0.264844 LR 0.001000 Time 0.019609 -2022-12-06 10:52:08,569 - Epoch: [66][ 1180/ 1200] Overall Loss 0.264665 Objective Loss 0.264665 LR 0.001000 Time 0.019604 -2022-12-06 10:52:08,760 - Epoch: [66][ 1190/ 1200] Overall Loss 0.264710 Objective Loss 0.264710 LR 0.001000 Time 0.019600 -2022-12-06 10:52:08,982 - Epoch: [66][ 1200/ 1200] Overall Loss 0.264922 Objective Loss 0.264922 Top1 84.309623 Top5 98.953975 LR 0.001000 Time 0.019621 -2022-12-06 10:52:09,070 - --- validate (epoch=66)----------- -2022-12-06 10:52:09,070 - 34129 samples (256 per mini-batch) -2022-12-06 10:52:09,521 - Epoch: [66][ 10/ 134] Loss 0.292982 Top1 83.750000 Top5 98.125000 -2022-12-06 10:52:09,654 - Epoch: [66][ 20/ 134] Loss 0.286639 Top1 84.003906 Top5 97.929688 -2022-12-06 10:52:09,786 - Epoch: [66][ 30/ 134] Loss 0.299932 Top1 83.177083 Top5 97.838542 -2022-12-06 10:52:09,919 - Epoch: [66][ 40/ 134] Loss 0.311553 Top1 82.812500 Top5 97.763672 -2022-12-06 10:52:10,051 - Epoch: [66][ 50/ 134] Loss 0.307378 Top1 83.179688 Top5 97.804688 -2022-12-06 10:52:10,183 - Epoch: [66][ 60/ 134] Loss 0.303357 Top1 83.509115 Top5 97.903646 -2022-12-06 10:52:10,315 - Epoch: [66][ 70/ 134] Loss 0.300192 Top1 83.521205 Top5 97.968750 -2022-12-06 10:52:10,447 - Epoch: [66][ 80/ 134] Loss 0.294537 Top1 83.725586 Top5 97.993164 -2022-12-06 10:52:10,579 - Epoch: [66][ 90/ 134] Loss 0.291517 Top1 83.858507 Top5 97.994792 -2022-12-06 10:52:10,712 - Epoch: [66][ 100/ 134] Loss 0.290241 Top1 84.000000 Top5 97.992188 -2022-12-06 10:52:10,843 - Epoch: [66][ 110/ 134] Loss 0.289399 Top1 83.977273 Top5 98.000710 -2022-12-06 10:52:10,977 - Epoch: [66][ 120/ 134] Loss 0.292028 Top1 83.958333 Top5 97.978516 -2022-12-06 10:52:11,110 - Epoch: [66][ 130/ 134] Loss 0.292269 Top1 84.008413 Top5 97.971755 -2022-12-06 10:52:11,147 - Epoch: [66][ 134/ 134] Loss 0.294378 Top1 83.990155 Top5 97.981189 -2022-12-06 10:52:11,235 - ==> Top1: 83.990 Top5: 97.981 Loss: 0.294 - -2022-12-06 10:52:11,236 - ==> Confusion: -[[ 858 3 1 1 7 4 1 0 4 94 0 1 3 4 7 1 1 0 0 0 6] - [ 3 924 2 2 15 22 2 10 2 1 8 6 2 1 1 2 7 2 6 2 7] - [ 4 3 985 14 5 5 24 8 0 4 11 5 6 3 4 6 1 1 2 2 10] - [ 3 1 17 937 0 3 1 0 1 0 18 1 3 5 13 1 1 2 9 1 3] - [ 8 2 2 0 962 4 1 1 0 7 2 3 1 3 10 3 7 2 0 0 2] - [ 5 18 1 5 8 969 3 13 3 1 2 10 4 11 0 0 2 0 0 8 6] - [ 1 3 11 2 1 6 1056 3 0 0 5 4 0 1 0 8 0 3 1 10 3] - [ 1 16 12 2 3 30 6 917 3 0 6 6 1 3 1 0 2 3 26 11 5] - [ 6 3 0 0 0 2 0 1 975 47 6 1 1 9 9 0 2 0 2 0 0] - [ 35 0 0 0 7 6 0 1 17 911 2 1 0 12 3 0 0 0 0 0 6] - [ 1 1 1 5 1 0 1 2 11 0 968 2 1 14 3 0 2 1 2 1 2] - [ 4 0 2 0 0 11 3 5 3 0 1 953 31 5 0 11 5 4 1 9 3] - [ 3 2 0 3 1 2 2 0 0 0 0 40 884 2 0 9 1 8 2 6 4] - [ 0 1 0 1 1 11 0 2 12 16 7 5 1 954 0 2 2 1 0 4 3] - [ 4 3 3 19 5 2 0 0 27 4 4 2 2 5 1040 0 2 0 5 0 3] - [ 3 0 2 0 4 2 4 0 1 0 0 5 7 3 0 993 8 6 0 2 3] - [ 3 3 3 0 3 1 1 0 3 0 0 2 0 4 2 8 1028 0 1 4 6] - [ 4 1 3 2 0 1 2 1 5 2 0 11 33 1 0 23 2 943 0 1 1] - [ 4 3 6 17 3 4 0 26 2 1 13 3 3 0 9 0 2 0 903 5 4] - [ 2 1 2 0 2 8 4 7 0 1 2 13 9 6 1 2 11 7 0 999 3] - [ 147 206 177 143 156 193 93 111 115 107 268 112 407 370 203 160 271 80 155 254 9498]] - -2022-12-06 10:52:11,903 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:52:11,903 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:52:11,909 - - -2022-12-06 10:52:11,909 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:52:12,847 - Epoch: [67][ 10/ 1200] Overall Loss 0.290178 Objective Loss 0.290178 LR 0.001000 Time 0.093648 -2022-12-06 10:52:13,046 - Epoch: [67][ 20/ 1200] Overall Loss 0.270127 Objective Loss 0.270127 LR 0.001000 Time 0.056740 -2022-12-06 10:52:13,238 - Epoch: [67][ 30/ 1200] Overall Loss 0.266861 Objective Loss 0.266861 LR 0.001000 Time 0.044218 -2022-12-06 10:52:13,430 - Epoch: [67][ 40/ 1200] Overall Loss 0.265340 Objective Loss 0.265340 LR 0.001000 Time 0.037951 -2022-12-06 10:52:13,621 - Epoch: [67][ 50/ 1200] Overall Loss 0.266956 Objective Loss 0.266956 LR 0.001000 Time 0.034180 -2022-12-06 10:52:13,813 - Epoch: [67][ 60/ 1200] Overall Loss 0.269384 Objective Loss 0.269384 LR 0.001000 Time 0.031668 -2022-12-06 10:52:14,005 - Epoch: [67][ 70/ 1200] Overall Loss 0.268200 Objective Loss 0.268200 LR 0.001000 Time 0.029880 -2022-12-06 10:52:14,197 - Epoch: [67][ 80/ 1200] Overall Loss 0.266114 Objective Loss 0.266114 LR 0.001000 Time 0.028538 -2022-12-06 10:52:14,389 - Epoch: [67][ 90/ 1200] Overall Loss 0.265992 Objective Loss 0.265992 LR 0.001000 Time 0.027489 -2022-12-06 10:52:14,580 - Epoch: [67][ 100/ 1200] Overall Loss 0.265589 Objective Loss 0.265589 LR 0.001000 Time 0.026648 -2022-12-06 10:52:14,772 - Epoch: [67][ 110/ 1200] Overall Loss 0.265064 Objective Loss 0.265064 LR 0.001000 Time 0.025963 -2022-12-06 10:52:14,964 - Epoch: [67][ 120/ 1200] Overall Loss 0.264874 Objective Loss 0.264874 LR 0.001000 Time 0.025396 -2022-12-06 10:52:15,158 - Epoch: [67][ 130/ 1200] Overall Loss 0.264695 Objective Loss 0.264695 LR 0.001000 Time 0.024931 -2022-12-06 10:52:15,353 - Epoch: [67][ 140/ 1200] Overall Loss 0.264233 Objective Loss 0.264233 LR 0.001000 Time 0.024537 -2022-12-06 10:52:15,547 - Epoch: [67][ 150/ 1200] Overall Loss 0.264866 Objective Loss 0.264866 LR 0.001000 Time 0.024195 -2022-12-06 10:52:15,742 - Epoch: [67][ 160/ 1200] Overall Loss 0.264493 Objective Loss 0.264493 LR 0.001000 Time 0.023894 -2022-12-06 10:52:15,936 - Epoch: [67][ 170/ 1200] Overall Loss 0.265528 Objective Loss 0.265528 LR 0.001000 Time 0.023630 -2022-12-06 10:52:16,131 - Epoch: [67][ 180/ 1200] Overall Loss 0.267431 Objective Loss 0.267431 LR 0.001000 Time 0.023394 -2022-12-06 10:52:16,324 - Epoch: [67][ 190/ 1200] Overall Loss 0.266943 Objective Loss 0.266943 LR 0.001000 Time 0.023181 -2022-12-06 10:52:16,519 - Epoch: [67][ 200/ 1200] Overall Loss 0.268776 Objective Loss 0.268776 LR 0.001000 Time 0.022993 -2022-12-06 10:52:16,714 - Epoch: [67][ 210/ 1200] Overall Loss 0.268769 Objective Loss 0.268769 LR 0.001000 Time 0.022821 -2022-12-06 10:52:16,908 - Epoch: [67][ 220/ 1200] Overall Loss 0.267745 Objective Loss 0.267745 LR 0.001000 Time 0.022666 -2022-12-06 10:52:17,102 - Epoch: [67][ 230/ 1200] Overall Loss 0.265777 Objective Loss 0.265777 LR 0.001000 Time 0.022522 -2022-12-06 10:52:17,297 - Epoch: [67][ 240/ 1200] Overall Loss 0.265825 Objective Loss 0.265825 LR 0.001000 Time 0.022391 -2022-12-06 10:52:17,491 - Epoch: [67][ 250/ 1200] Overall Loss 0.264720 Objective Loss 0.264720 LR 0.001000 Time 0.022270 -2022-12-06 10:52:17,686 - Epoch: [67][ 260/ 1200] Overall Loss 0.265196 Objective Loss 0.265196 LR 0.001000 Time 0.022161 -2022-12-06 10:52:17,880 - Epoch: [67][ 270/ 1200] Overall Loss 0.265592 Objective Loss 0.265592 LR 0.001000 Time 0.022057 -2022-12-06 10:52:18,075 - Epoch: [67][ 280/ 1200] Overall Loss 0.265723 Objective Loss 0.265723 LR 0.001000 Time 0.021965 -2022-12-06 10:52:18,270 - Epoch: [67][ 290/ 1200] Overall Loss 0.266034 Objective Loss 0.266034 LR 0.001000 Time 0.021875 -2022-12-06 10:52:18,464 - Epoch: [67][ 300/ 1200] Overall Loss 0.265541 Objective Loss 0.265541 LR 0.001000 Time 0.021793 -2022-12-06 10:52:18,657 - Epoch: [67][ 310/ 1200] Overall Loss 0.265970 Objective Loss 0.265970 LR 0.001000 Time 0.021709 -2022-12-06 10:52:18,849 - Epoch: [67][ 320/ 1200] Overall Loss 0.265175 Objective Loss 0.265175 LR 0.001000 Time 0.021630 -2022-12-06 10:52:19,041 - Epoch: [67][ 330/ 1200] Overall Loss 0.265726 Objective Loss 0.265726 LR 0.001000 Time 0.021553 -2022-12-06 10:52:19,233 - Epoch: [67][ 340/ 1200] Overall Loss 0.265952 Objective Loss 0.265952 LR 0.001000 Time 0.021484 -2022-12-06 10:52:19,425 - Epoch: [67][ 350/ 1200] Overall Loss 0.265332 Objective Loss 0.265332 LR 0.001000 Time 0.021417 -2022-12-06 10:52:19,617 - Epoch: [67][ 360/ 1200] Overall Loss 0.265371 Objective Loss 0.265371 LR 0.001000 Time 0.021354 -2022-12-06 10:52:19,809 - Epoch: [67][ 370/ 1200] Overall Loss 0.264753 Objective Loss 0.264753 LR 0.001000 Time 0.021295 -2022-12-06 10:52:20,001 - Epoch: [67][ 380/ 1200] Overall Loss 0.264342 Objective Loss 0.264342 LR 0.001000 Time 0.021237 -2022-12-06 10:52:20,192 - Epoch: [67][ 390/ 1200] Overall Loss 0.264832 Objective Loss 0.264832 LR 0.001000 Time 0.021181 -2022-12-06 10:52:20,383 - Epoch: [67][ 400/ 1200] Overall Loss 0.266213 Objective Loss 0.266213 LR 0.001000 Time 0.021128 -2022-12-06 10:52:20,576 - Epoch: [67][ 410/ 1200] Overall Loss 0.266749 Objective Loss 0.266749 LR 0.001000 Time 0.021081 -2022-12-06 10:52:20,767 - Epoch: [67][ 420/ 1200] Overall Loss 0.267136 Objective Loss 0.267136 LR 0.001000 Time 0.021034 -2022-12-06 10:52:20,959 - Epoch: [67][ 430/ 1200] Overall Loss 0.267899 Objective Loss 0.267899 LR 0.001000 Time 0.020989 -2022-12-06 10:52:21,151 - Epoch: [67][ 440/ 1200] Overall Loss 0.267060 Objective Loss 0.267060 LR 0.001000 Time 0.020947 -2022-12-06 10:52:21,342 - Epoch: [67][ 450/ 1200] Overall Loss 0.267177 Objective Loss 0.267177 LR 0.001000 Time 0.020906 -2022-12-06 10:52:21,534 - Epoch: [67][ 460/ 1200] Overall Loss 0.266965 Objective Loss 0.266965 LR 0.001000 Time 0.020867 -2022-12-06 10:52:21,726 - Epoch: [67][ 470/ 1200] Overall Loss 0.267627 Objective Loss 0.267627 LR 0.001000 Time 0.020830 -2022-12-06 10:52:21,918 - Epoch: [67][ 480/ 1200] Overall Loss 0.268215 Objective Loss 0.268215 LR 0.001000 Time 0.020796 -2022-12-06 10:52:22,110 - Epoch: [67][ 490/ 1200] Overall Loss 0.267730 Objective Loss 0.267730 LR 0.001000 Time 0.020762 -2022-12-06 10:52:22,302 - Epoch: [67][ 500/ 1200] Overall Loss 0.267766 Objective Loss 0.267766 LR 0.001000 Time 0.020729 -2022-12-06 10:52:22,494 - Epoch: [67][ 510/ 1200] Overall Loss 0.267507 Objective Loss 0.267507 LR 0.001000 Time 0.020698 -2022-12-06 10:52:22,686 - Epoch: [67][ 520/ 1200] Overall Loss 0.267548 Objective Loss 0.267548 LR 0.001000 Time 0.020668 -2022-12-06 10:52:22,878 - Epoch: [67][ 530/ 1200] Overall Loss 0.267142 Objective Loss 0.267142 LR 0.001000 Time 0.020639 -2022-12-06 10:52:23,069 - Epoch: [67][ 540/ 1200] Overall Loss 0.267364 Objective Loss 0.267364 LR 0.001000 Time 0.020611 -2022-12-06 10:52:23,261 - Epoch: [67][ 550/ 1200] Overall Loss 0.267392 Objective Loss 0.267392 LR 0.001000 Time 0.020583 -2022-12-06 10:52:23,453 - Epoch: [67][ 560/ 1200] Overall Loss 0.267111 Objective Loss 0.267111 LR 0.001000 Time 0.020558 -2022-12-06 10:52:23,645 - Epoch: [67][ 570/ 1200] Overall Loss 0.266963 Objective Loss 0.266963 LR 0.001000 Time 0.020533 -2022-12-06 10:52:23,837 - Epoch: [67][ 580/ 1200] Overall Loss 0.266309 Objective Loss 0.266309 LR 0.001000 Time 0.020509 -2022-12-06 10:52:24,029 - Epoch: [67][ 590/ 1200] Overall Loss 0.266231 Objective Loss 0.266231 LR 0.001000 Time 0.020486 -2022-12-06 10:52:24,221 - Epoch: [67][ 600/ 1200] Overall Loss 0.265804 Objective Loss 0.265804 LR 0.001000 Time 0.020463 -2022-12-06 10:52:24,413 - Epoch: [67][ 610/ 1200] Overall Loss 0.265673 Objective Loss 0.265673 LR 0.001000 Time 0.020442 -2022-12-06 10:52:24,605 - Epoch: [67][ 620/ 1200] Overall Loss 0.265007 Objective Loss 0.265007 LR 0.001000 Time 0.020421 -2022-12-06 10:52:24,797 - Epoch: [67][ 630/ 1200] Overall Loss 0.264709 Objective Loss 0.264709 LR 0.001000 Time 0.020401 -2022-12-06 10:52:24,989 - Epoch: [67][ 640/ 1200] Overall Loss 0.264213 Objective Loss 0.264213 LR 0.001000 Time 0.020381 -2022-12-06 10:52:25,181 - Epoch: [67][ 650/ 1200] Overall Loss 0.264308 Objective Loss 0.264308 LR 0.001000 Time 0.020362 -2022-12-06 10:52:25,373 - Epoch: [67][ 660/ 1200] Overall Loss 0.264850 Objective Loss 0.264850 LR 0.001000 Time 0.020344 -2022-12-06 10:52:25,565 - Epoch: [67][ 670/ 1200] Overall Loss 0.264573 Objective Loss 0.264573 LR 0.001000 Time 0.020326 -2022-12-06 10:52:25,757 - Epoch: [67][ 680/ 1200] Overall Loss 0.264876 Objective Loss 0.264876 LR 0.001000 Time 0.020308 -2022-12-06 10:52:25,949 - Epoch: [67][ 690/ 1200] Overall Loss 0.265382 Objective Loss 0.265382 LR 0.001000 Time 0.020291 -2022-12-06 10:52:26,141 - Epoch: [67][ 700/ 1200] Overall Loss 0.265268 Objective Loss 0.265268 LR 0.001000 Time 0.020274 -2022-12-06 10:52:26,332 - Epoch: [67][ 710/ 1200] Overall Loss 0.265692 Objective Loss 0.265692 LR 0.001000 Time 0.020258 -2022-12-06 10:52:26,524 - Epoch: [67][ 720/ 1200] Overall Loss 0.265507 Objective Loss 0.265507 LR 0.001000 Time 0.020242 -2022-12-06 10:52:26,715 - Epoch: [67][ 730/ 1200] Overall Loss 0.265484 Objective Loss 0.265484 LR 0.001000 Time 0.020226 -2022-12-06 10:52:26,908 - Epoch: [67][ 740/ 1200] Overall Loss 0.265638 Objective Loss 0.265638 LR 0.001000 Time 0.020212 -2022-12-06 10:52:27,100 - Epoch: [67][ 750/ 1200] Overall Loss 0.265637 Objective Loss 0.265637 LR 0.001000 Time 0.020199 -2022-12-06 10:52:27,292 - Epoch: [67][ 760/ 1200] Overall Loss 0.265325 Objective Loss 0.265325 LR 0.001000 Time 0.020184 -2022-12-06 10:52:27,484 - Epoch: [67][ 770/ 1200] Overall Loss 0.265755 Objective Loss 0.265755 LR 0.001000 Time 0.020171 -2022-12-06 10:52:27,676 - Epoch: [67][ 780/ 1200] Overall Loss 0.265899 Objective Loss 0.265899 LR 0.001000 Time 0.020157 -2022-12-06 10:52:27,868 - Epoch: [67][ 790/ 1200] Overall Loss 0.265692 Objective Loss 0.265692 LR 0.001000 Time 0.020145 -2022-12-06 10:52:28,060 - Epoch: [67][ 800/ 1200] Overall Loss 0.265635 Objective Loss 0.265635 LR 0.001000 Time 0.020132 -2022-12-06 10:52:28,252 - Epoch: [67][ 810/ 1200] Overall Loss 0.265538 Objective Loss 0.265538 LR 0.001000 Time 0.020120 -2022-12-06 10:52:28,445 - Epoch: [67][ 820/ 1200] Overall Loss 0.265922 Objective Loss 0.265922 LR 0.001000 Time 0.020110 -2022-12-06 10:52:28,637 - Epoch: [67][ 830/ 1200] Overall Loss 0.265693 Objective Loss 0.265693 LR 0.001000 Time 0.020098 -2022-12-06 10:52:28,830 - Epoch: [67][ 840/ 1200] Overall Loss 0.265624 Objective Loss 0.265624 LR 0.001000 Time 0.020087 -2022-12-06 10:52:29,022 - Epoch: [67][ 850/ 1200] Overall Loss 0.265898 Objective Loss 0.265898 LR 0.001000 Time 0.020076 -2022-12-06 10:52:29,214 - Epoch: [67][ 860/ 1200] Overall Loss 0.266220 Objective Loss 0.266220 LR 0.001000 Time 0.020066 -2022-12-06 10:52:29,406 - Epoch: [67][ 870/ 1200] Overall Loss 0.266352 Objective Loss 0.266352 LR 0.001000 Time 0.020055 -2022-12-06 10:52:29,599 - Epoch: [67][ 880/ 1200] Overall Loss 0.266128 Objective Loss 0.266128 LR 0.001000 Time 0.020046 -2022-12-06 10:52:29,791 - Epoch: [67][ 890/ 1200] Overall Loss 0.266192 Objective Loss 0.266192 LR 0.001000 Time 0.020036 -2022-12-06 10:52:29,983 - Epoch: [67][ 900/ 1200] Overall Loss 0.266057 Objective Loss 0.266057 LR 0.001000 Time 0.020025 -2022-12-06 10:52:30,175 - Epoch: [67][ 910/ 1200] Overall Loss 0.265982 Objective Loss 0.265982 LR 0.001000 Time 0.020016 -2022-12-06 10:52:30,367 - Epoch: [67][ 920/ 1200] Overall Loss 0.266310 Objective Loss 0.266310 LR 0.001000 Time 0.020007 -2022-12-06 10:52:30,559 - Epoch: [67][ 930/ 1200] Overall Loss 0.266341 Objective Loss 0.266341 LR 0.001000 Time 0.019997 -2022-12-06 10:52:30,751 - Epoch: [67][ 940/ 1200] Overall Loss 0.266383 Objective Loss 0.266383 LR 0.001000 Time 0.019988 -2022-12-06 10:52:30,943 - Epoch: [67][ 950/ 1200] Overall Loss 0.266734 Objective Loss 0.266734 LR 0.001000 Time 0.019979 -2022-12-06 10:52:31,136 - Epoch: [67][ 960/ 1200] Overall Loss 0.266837 Objective Loss 0.266837 LR 0.001000 Time 0.019971 -2022-12-06 10:52:31,328 - Epoch: [67][ 970/ 1200] Overall Loss 0.267011 Objective Loss 0.267011 LR 0.001000 Time 0.019963 -2022-12-06 10:52:31,520 - Epoch: [67][ 980/ 1200] Overall Loss 0.267087 Objective Loss 0.267087 LR 0.001000 Time 0.019955 -2022-12-06 10:52:31,712 - Epoch: [67][ 990/ 1200] Overall Loss 0.267111 Objective Loss 0.267111 LR 0.001000 Time 0.019947 -2022-12-06 10:52:31,905 - Epoch: [67][ 1000/ 1200] Overall Loss 0.267087 Objective Loss 0.267087 LR 0.001000 Time 0.019939 -2022-12-06 10:52:32,097 - Epoch: [67][ 1010/ 1200] Overall Loss 0.267314 Objective Loss 0.267314 LR 0.001000 Time 0.019932 -2022-12-06 10:52:32,289 - Epoch: [67][ 1020/ 1200] Overall Loss 0.267498 Objective Loss 0.267498 LR 0.001000 Time 0.019924 -2022-12-06 10:52:32,480 - Epoch: [67][ 1030/ 1200] Overall Loss 0.267408 Objective Loss 0.267408 LR 0.001000 Time 0.019916 -2022-12-06 10:52:32,673 - Epoch: [67][ 1040/ 1200] Overall Loss 0.267443 Objective Loss 0.267443 LR 0.001000 Time 0.019909 -2022-12-06 10:52:32,865 - Epoch: [67][ 1050/ 1200] Overall Loss 0.267459 Objective Loss 0.267459 LR 0.001000 Time 0.019901 -2022-12-06 10:52:33,056 - Epoch: [67][ 1060/ 1200] Overall Loss 0.267518 Objective Loss 0.267518 LR 0.001000 Time 0.019894 -2022-12-06 10:52:33,248 - Epoch: [67][ 1070/ 1200] Overall Loss 0.267742 Objective Loss 0.267742 LR 0.001000 Time 0.019887 -2022-12-06 10:52:33,439 - Epoch: [67][ 1080/ 1200] Overall Loss 0.267625 Objective Loss 0.267625 LR 0.001000 Time 0.019879 -2022-12-06 10:52:33,631 - Epoch: [67][ 1090/ 1200] Overall Loss 0.267633 Objective Loss 0.267633 LR 0.001000 Time 0.019872 -2022-12-06 10:52:33,823 - Epoch: [67][ 1100/ 1200] Overall Loss 0.267742 Objective Loss 0.267742 LR 0.001000 Time 0.019865 -2022-12-06 10:52:34,015 - Epoch: [67][ 1110/ 1200] Overall Loss 0.268051 Objective Loss 0.268051 LR 0.001000 Time 0.019859 -2022-12-06 10:52:34,206 - Epoch: [67][ 1120/ 1200] Overall Loss 0.268039 Objective Loss 0.268039 LR 0.001000 Time 0.019852 -2022-12-06 10:52:34,398 - Epoch: [67][ 1130/ 1200] Overall Loss 0.267930 Objective Loss 0.267930 LR 0.001000 Time 0.019846 -2022-12-06 10:52:34,591 - Epoch: [67][ 1140/ 1200] Overall Loss 0.267991 Objective Loss 0.267991 LR 0.001000 Time 0.019840 -2022-12-06 10:52:34,782 - Epoch: [67][ 1150/ 1200] Overall Loss 0.268105 Objective Loss 0.268105 LR 0.001000 Time 0.019833 -2022-12-06 10:52:34,974 - Epoch: [67][ 1160/ 1200] Overall Loss 0.268138 Objective Loss 0.268138 LR 0.001000 Time 0.019828 -2022-12-06 10:52:35,166 - Epoch: [67][ 1170/ 1200] Overall Loss 0.268042 Objective Loss 0.268042 LR 0.001000 Time 0.019822 -2022-12-06 10:52:35,358 - Epoch: [67][ 1180/ 1200] Overall Loss 0.268391 Objective Loss 0.268391 LR 0.001000 Time 0.019816 -2022-12-06 10:52:35,550 - Epoch: [67][ 1190/ 1200] Overall Loss 0.268406 Objective Loss 0.268406 LR 0.001000 Time 0.019810 -2022-12-06 10:52:35,784 - Epoch: [67][ 1200/ 1200] Overall Loss 0.268427 Objective Loss 0.268427 Top1 84.937238 Top5 98.326360 LR 0.001000 Time 0.019840 -2022-12-06 10:52:35,873 - --- validate (epoch=67)----------- -2022-12-06 10:52:35,873 - 34129 samples (256 per mini-batch) -2022-12-06 10:52:36,324 - Epoch: [67][ 10/ 134] Loss 0.316450 Top1 82.500000 Top5 97.968750 -2022-12-06 10:52:36,455 - Epoch: [67][ 20/ 134] Loss 0.299751 Top1 83.261719 Top5 97.871094 -2022-12-06 10:52:36,591 - Epoch: [67][ 30/ 134] Loss 0.292270 Top1 83.450521 Top5 97.838542 -2022-12-06 10:52:36,719 - Epoch: [67][ 40/ 134] Loss 0.296950 Top1 83.525391 Top5 97.763672 -2022-12-06 10:52:36,856 - Epoch: [67][ 50/ 134] Loss 0.298835 Top1 83.609375 Top5 97.765625 -2022-12-06 10:52:36,987 - Epoch: [67][ 60/ 134] Loss 0.295888 Top1 83.613281 Top5 97.799479 -2022-12-06 10:52:37,124 - Epoch: [67][ 70/ 134] Loss 0.295212 Top1 83.655134 Top5 97.767857 -2022-12-06 10:52:37,250 - Epoch: [67][ 80/ 134] Loss 0.294414 Top1 83.745117 Top5 97.792969 -2022-12-06 10:52:37,382 - Epoch: [67][ 90/ 134] Loss 0.292273 Top1 83.793403 Top5 97.812500 -2022-12-06 10:52:37,512 - Epoch: [67][ 100/ 134] Loss 0.292138 Top1 83.765625 Top5 97.792969 -2022-12-06 10:52:37,645 - Epoch: [67][ 110/ 134] Loss 0.290714 Top1 83.735795 Top5 97.794744 -2022-12-06 10:52:37,776 - Epoch: [67][ 120/ 134] Loss 0.291768 Top1 83.854167 Top5 97.776693 -2022-12-06 10:52:37,910 - Epoch: [67][ 130/ 134] Loss 0.293181 Top1 83.852163 Top5 97.785457 -2022-12-06 10:52:37,949 - Epoch: [67][ 134/ 134] Loss 0.293012 Top1 83.890533 Top5 97.793665 -2022-12-06 10:52:38,039 - ==> Top1: 83.891 Top5: 97.794 Loss: 0.293 - -2022-12-06 10:52:38,040 - ==> Confusion: -[[ 905 2 1 0 5 8 0 4 2 55 0 2 1 3 3 1 2 0 1 0 1] - [ 0 916 3 5 4 22 5 21 0 3 3 6 2 2 0 0 4 3 16 4 8] - [ 6 3 970 25 5 2 33 13 1 2 9 5 2 1 2 2 0 2 8 3 9] - [ 3 2 14 928 2 2 2 1 0 0 13 0 2 2 21 0 1 3 19 2 3] - [ 11 8 3 0 936 5 1 3 0 12 1 5 2 2 13 4 9 1 0 3 1] - [ 4 11 1 4 10 949 5 20 3 1 2 12 5 17 4 0 0 0 2 10 9] - [ 2 0 14 2 0 5 1065 4 0 1 4 2 2 2 0 6 2 1 0 5 1] - [ 3 3 4 2 1 31 10 944 0 0 5 9 2 2 0 0 0 1 26 7 4] - [ 3 3 0 2 1 4 0 2 954 47 16 2 1 8 14 1 1 0 3 1 1] - [ 60 1 0 0 5 2 0 2 22 883 1 5 0 10 3 0 0 1 0 0 6] - [ 1 1 2 9 1 2 0 3 10 0 961 3 1 6 4 0 1 0 9 1 4] - [ 4 2 4 1 1 8 4 5 0 0 1 962 20 7 1 7 4 8 1 9 2] - [ 2 1 3 2 1 1 2 2 0 0 0 40 879 0 2 9 1 13 2 4 5] - [ 0 1 0 1 0 7 0 4 12 15 15 7 1 943 2 3 1 1 1 2 7] - [ 11 2 2 8 4 3 0 2 17 3 2 1 2 4 1055 0 0 1 8 1 4] - [ 2 0 1 2 4 1 4 1 0 0 1 8 2 4 0 994 5 9 0 3 2] - [ 5 7 3 0 2 0 0 0 0 1 2 9 0 2 2 10 1018 0 2 5 4] - [ 3 2 4 4 0 3 3 2 1 1 1 11 9 3 3 13 2 967 1 2 1] - [ 5 6 1 7 0 3 0 31 1 1 9 4 2 0 8 2 0 0 925 2 1] - [ 4 6 1 0 0 9 10 11 0 0 0 20 6 5 1 2 4 2 0 995 4] - [ 199 218 202 134 117 175 117 198 81 91 228 154 409 254 223 180 200 102 201 263 9480]] - -2022-12-06 10:52:38,617 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:52:38,617 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:52:38,623 - - -2022-12-06 10:52:38,623 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:52:39,668 - Epoch: [68][ 10/ 1200] Overall Loss 0.261291 Objective Loss 0.261291 LR 0.001000 Time 0.104436 -2022-12-06 10:52:39,870 - Epoch: [68][ 20/ 1200] Overall Loss 0.267450 Objective Loss 0.267450 LR 0.001000 Time 0.062310 -2022-12-06 10:52:40,063 - Epoch: [68][ 30/ 1200] Overall Loss 0.283835 Objective Loss 0.283835 LR 0.001000 Time 0.047942 -2022-12-06 10:52:40,255 - Epoch: [68][ 40/ 1200] Overall Loss 0.282363 Objective Loss 0.282363 LR 0.001000 Time 0.040748 -2022-12-06 10:52:40,447 - Epoch: [68][ 50/ 1200] Overall Loss 0.275898 Objective Loss 0.275898 LR 0.001000 Time 0.036424 -2022-12-06 10:52:40,639 - Epoch: [68][ 60/ 1200] Overall Loss 0.273719 Objective Loss 0.273719 LR 0.001000 Time 0.033547 -2022-12-06 10:52:40,831 - Epoch: [68][ 70/ 1200] Overall Loss 0.271811 Objective Loss 0.271811 LR 0.001000 Time 0.031485 -2022-12-06 10:52:41,022 - Epoch: [68][ 80/ 1200] Overall Loss 0.275429 Objective Loss 0.275429 LR 0.001000 Time 0.029935 -2022-12-06 10:52:41,214 - Epoch: [68][ 90/ 1200] Overall Loss 0.272139 Objective Loss 0.272139 LR 0.001000 Time 0.028731 -2022-12-06 10:52:41,405 - Epoch: [68][ 100/ 1200] Overall Loss 0.270590 Objective Loss 0.270590 LR 0.001000 Time 0.027768 -2022-12-06 10:52:41,597 - Epoch: [68][ 110/ 1200] Overall Loss 0.269416 Objective Loss 0.269416 LR 0.001000 Time 0.026980 -2022-12-06 10:52:41,789 - Epoch: [68][ 120/ 1200] Overall Loss 0.269593 Objective Loss 0.269593 LR 0.001000 Time 0.026324 -2022-12-06 10:52:41,980 - Epoch: [68][ 130/ 1200] Overall Loss 0.267826 Objective Loss 0.267826 LR 0.001000 Time 0.025765 -2022-12-06 10:52:42,172 - Epoch: [68][ 140/ 1200] Overall Loss 0.264536 Objective Loss 0.264536 LR 0.001000 Time 0.025291 -2022-12-06 10:52:42,363 - Epoch: [68][ 150/ 1200] Overall Loss 0.262745 Objective Loss 0.262745 LR 0.001000 Time 0.024880 -2022-12-06 10:52:42,555 - Epoch: [68][ 160/ 1200] Overall Loss 0.264687 Objective Loss 0.264687 LR 0.001000 Time 0.024518 -2022-12-06 10:52:42,747 - Epoch: [68][ 170/ 1200] Overall Loss 0.265588 Objective Loss 0.265588 LR 0.001000 Time 0.024200 -2022-12-06 10:52:42,938 - Epoch: [68][ 180/ 1200] Overall Loss 0.265071 Objective Loss 0.265071 LR 0.001000 Time 0.023916 -2022-12-06 10:52:43,129 - Epoch: [68][ 190/ 1200] Overall Loss 0.266707 Objective Loss 0.266707 LR 0.001000 Time 0.023662 -2022-12-06 10:52:43,321 - Epoch: [68][ 200/ 1200] Overall Loss 0.266265 Objective Loss 0.266265 LR 0.001000 Time 0.023437 -2022-12-06 10:52:43,513 - Epoch: [68][ 210/ 1200] Overall Loss 0.266446 Objective Loss 0.266446 LR 0.001000 Time 0.023231 -2022-12-06 10:52:43,706 - Epoch: [68][ 220/ 1200] Overall Loss 0.265907 Objective Loss 0.265907 LR 0.001000 Time 0.023050 -2022-12-06 10:52:43,899 - Epoch: [68][ 230/ 1200] Overall Loss 0.265333 Objective Loss 0.265333 LR 0.001000 Time 0.022885 -2022-12-06 10:52:44,093 - Epoch: [68][ 240/ 1200] Overall Loss 0.265620 Objective Loss 0.265620 LR 0.001000 Time 0.022734 -2022-12-06 10:52:44,286 - Epoch: [68][ 250/ 1200] Overall Loss 0.265405 Objective Loss 0.265405 LR 0.001000 Time 0.022597 -2022-12-06 10:52:44,479 - Epoch: [68][ 260/ 1200] Overall Loss 0.264939 Objective Loss 0.264939 LR 0.001000 Time 0.022468 -2022-12-06 10:52:44,673 - Epoch: [68][ 270/ 1200] Overall Loss 0.264883 Objective Loss 0.264883 LR 0.001000 Time 0.022350 -2022-12-06 10:52:44,866 - Epoch: [68][ 280/ 1200] Overall Loss 0.264969 Objective Loss 0.264969 LR 0.001000 Time 0.022240 -2022-12-06 10:52:45,059 - Epoch: [68][ 290/ 1200] Overall Loss 0.265352 Objective Loss 0.265352 LR 0.001000 Time 0.022137 -2022-12-06 10:52:45,252 - Epoch: [68][ 300/ 1200] Overall Loss 0.265935 Objective Loss 0.265935 LR 0.001000 Time 0.022042 -2022-12-06 10:52:45,446 - Epoch: [68][ 310/ 1200] Overall Loss 0.266108 Objective Loss 0.266108 LR 0.001000 Time 0.021955 -2022-12-06 10:52:45,640 - Epoch: [68][ 320/ 1200] Overall Loss 0.266395 Objective Loss 0.266395 LR 0.001000 Time 0.021871 -2022-12-06 10:52:45,832 - Epoch: [68][ 330/ 1200] Overall Loss 0.266443 Objective Loss 0.266443 LR 0.001000 Time 0.021789 -2022-12-06 10:52:46,026 - Epoch: [68][ 340/ 1200] Overall Loss 0.265934 Objective Loss 0.265934 LR 0.001000 Time 0.021716 -2022-12-06 10:52:46,219 - Epoch: [68][ 350/ 1200] Overall Loss 0.265631 Objective Loss 0.265631 LR 0.001000 Time 0.021646 -2022-12-06 10:52:46,412 - Epoch: [68][ 360/ 1200] Overall Loss 0.265636 Objective Loss 0.265636 LR 0.001000 Time 0.021580 -2022-12-06 10:52:46,605 - Epoch: [68][ 370/ 1200] Overall Loss 0.264911 Objective Loss 0.264911 LR 0.001000 Time 0.021516 -2022-12-06 10:52:46,798 - Epoch: [68][ 380/ 1200] Overall Loss 0.265109 Objective Loss 0.265109 LR 0.001000 Time 0.021457 -2022-12-06 10:52:46,990 - Epoch: [68][ 390/ 1200] Overall Loss 0.265606 Objective Loss 0.265606 LR 0.001000 Time 0.021399 -2022-12-06 10:52:47,183 - Epoch: [68][ 400/ 1200] Overall Loss 0.265441 Objective Loss 0.265441 LR 0.001000 Time 0.021345 -2022-12-06 10:52:47,376 - Epoch: [68][ 410/ 1200] Overall Loss 0.265498 Objective Loss 0.265498 LR 0.001000 Time 0.021294 -2022-12-06 10:52:47,570 - Epoch: [68][ 420/ 1200] Overall Loss 0.266162 Objective Loss 0.266162 LR 0.001000 Time 0.021246 -2022-12-06 10:52:47,763 - Epoch: [68][ 430/ 1200] Overall Loss 0.265736 Objective Loss 0.265736 LR 0.001000 Time 0.021200 -2022-12-06 10:52:47,957 - Epoch: [68][ 440/ 1200] Overall Loss 0.265217 Objective Loss 0.265217 LR 0.001000 Time 0.021157 -2022-12-06 10:52:48,150 - Epoch: [68][ 450/ 1200] Overall Loss 0.265481 Objective Loss 0.265481 LR 0.001000 Time 0.021115 -2022-12-06 10:52:48,343 - Epoch: [68][ 460/ 1200] Overall Loss 0.264123 Objective Loss 0.264123 LR 0.001000 Time 0.021075 -2022-12-06 10:52:48,537 - Epoch: [68][ 470/ 1200] Overall Loss 0.264307 Objective Loss 0.264307 LR 0.001000 Time 0.021038 -2022-12-06 10:52:48,730 - Epoch: [68][ 480/ 1200] Overall Loss 0.264477 Objective Loss 0.264477 LR 0.001000 Time 0.021001 -2022-12-06 10:52:48,924 - Epoch: [68][ 490/ 1200] Overall Loss 0.264638 Objective Loss 0.264638 LR 0.001000 Time 0.020966 -2022-12-06 10:52:49,117 - Epoch: [68][ 500/ 1200] Overall Loss 0.264153 Objective Loss 0.264153 LR 0.001000 Time 0.020932 -2022-12-06 10:52:49,310 - Epoch: [68][ 510/ 1200] Overall Loss 0.263715 Objective Loss 0.263715 LR 0.001000 Time 0.020900 -2022-12-06 10:52:49,505 - Epoch: [68][ 520/ 1200] Overall Loss 0.263954 Objective Loss 0.263954 LR 0.001000 Time 0.020870 -2022-12-06 10:52:49,699 - Epoch: [68][ 530/ 1200] Overall Loss 0.263826 Objective Loss 0.263826 LR 0.001000 Time 0.020841 -2022-12-06 10:52:49,892 - Epoch: [68][ 540/ 1200] Overall Loss 0.264052 Objective Loss 0.264052 LR 0.001000 Time 0.020813 -2022-12-06 10:52:50,085 - Epoch: [68][ 550/ 1200] Overall Loss 0.263798 Objective Loss 0.263798 LR 0.001000 Time 0.020784 -2022-12-06 10:52:50,279 - Epoch: [68][ 560/ 1200] Overall Loss 0.263543 Objective Loss 0.263543 LR 0.001000 Time 0.020757 -2022-12-06 10:52:50,472 - Epoch: [68][ 570/ 1200] Overall Loss 0.263888 Objective Loss 0.263888 LR 0.001000 Time 0.020731 -2022-12-06 10:52:50,665 - Epoch: [68][ 580/ 1200] Overall Loss 0.263905 Objective Loss 0.263905 LR 0.001000 Time 0.020706 -2022-12-06 10:52:50,859 - Epoch: [68][ 590/ 1200] Overall Loss 0.264015 Objective Loss 0.264015 LR 0.001000 Time 0.020683 -2022-12-06 10:52:51,052 - Epoch: [68][ 600/ 1200] Overall Loss 0.264333 Objective Loss 0.264333 LR 0.001000 Time 0.020660 -2022-12-06 10:52:51,246 - Epoch: [68][ 610/ 1200] Overall Loss 0.264675 Objective Loss 0.264675 LR 0.001000 Time 0.020638 -2022-12-06 10:52:51,440 - Epoch: [68][ 620/ 1200] Overall Loss 0.265039 Objective Loss 0.265039 LR 0.001000 Time 0.020617 -2022-12-06 10:52:51,634 - Epoch: [68][ 630/ 1200] Overall Loss 0.265265 Objective Loss 0.265265 LR 0.001000 Time 0.020596 -2022-12-06 10:52:51,827 - Epoch: [68][ 640/ 1200] Overall Loss 0.264404 Objective Loss 0.264404 LR 0.001000 Time 0.020575 -2022-12-06 10:52:52,021 - Epoch: [68][ 650/ 1200] Overall Loss 0.264337 Objective Loss 0.264337 LR 0.001000 Time 0.020556 -2022-12-06 10:52:52,214 - Epoch: [68][ 660/ 1200] Overall Loss 0.264714 Objective Loss 0.264714 LR 0.001000 Time 0.020536 -2022-12-06 10:52:52,407 - Epoch: [68][ 670/ 1200] Overall Loss 0.264928 Objective Loss 0.264928 LR 0.001000 Time 0.020517 -2022-12-06 10:52:52,601 - Epoch: [68][ 680/ 1200] Overall Loss 0.265203 Objective Loss 0.265203 LR 0.001000 Time 0.020499 -2022-12-06 10:52:52,794 - Epoch: [68][ 690/ 1200] Overall Loss 0.265181 Objective Loss 0.265181 LR 0.001000 Time 0.020482 -2022-12-06 10:52:52,987 - Epoch: [68][ 700/ 1200] Overall Loss 0.265503 Objective Loss 0.265503 LR 0.001000 Time 0.020464 -2022-12-06 10:52:53,181 - Epoch: [68][ 710/ 1200] Overall Loss 0.265210 Objective Loss 0.265210 LR 0.001000 Time 0.020448 -2022-12-06 10:52:53,374 - Epoch: [68][ 720/ 1200] Overall Loss 0.265438 Objective Loss 0.265438 LR 0.001000 Time 0.020431 -2022-12-06 10:52:53,567 - Epoch: [68][ 730/ 1200] Overall Loss 0.265416 Objective Loss 0.265416 LR 0.001000 Time 0.020416 -2022-12-06 10:52:53,761 - Epoch: [68][ 740/ 1200] Overall Loss 0.265427 Objective Loss 0.265427 LR 0.001000 Time 0.020401 -2022-12-06 10:52:53,954 - Epoch: [68][ 750/ 1200] Overall Loss 0.265405 Objective Loss 0.265405 LR 0.001000 Time 0.020386 -2022-12-06 10:52:54,148 - Epoch: [68][ 760/ 1200] Overall Loss 0.265196 Objective Loss 0.265196 LR 0.001000 Time 0.020372 -2022-12-06 10:52:54,342 - Epoch: [68][ 770/ 1200] Overall Loss 0.265082 Objective Loss 0.265082 LR 0.001000 Time 0.020358 -2022-12-06 10:52:54,534 - Epoch: [68][ 780/ 1200] Overall Loss 0.265335 Objective Loss 0.265335 LR 0.001000 Time 0.020343 -2022-12-06 10:52:54,728 - Epoch: [68][ 790/ 1200] Overall Loss 0.265587 Objective Loss 0.265587 LR 0.001000 Time 0.020331 -2022-12-06 10:52:54,921 - Epoch: [68][ 800/ 1200] Overall Loss 0.265418 Objective Loss 0.265418 LR 0.001000 Time 0.020317 -2022-12-06 10:52:55,115 - Epoch: [68][ 810/ 1200] Overall Loss 0.265674 Objective Loss 0.265674 LR 0.001000 Time 0.020305 -2022-12-06 10:52:55,309 - Epoch: [68][ 820/ 1200] Overall Loss 0.265761 Objective Loss 0.265761 LR 0.001000 Time 0.020292 -2022-12-06 10:52:55,502 - Epoch: [68][ 830/ 1200] Overall Loss 0.266527 Objective Loss 0.266527 LR 0.001000 Time 0.020280 -2022-12-06 10:52:55,696 - Epoch: [68][ 840/ 1200] Overall Loss 0.266446 Objective Loss 0.266446 LR 0.001000 Time 0.020269 -2022-12-06 10:52:55,889 - Epoch: [68][ 850/ 1200] Overall Loss 0.266556 Objective Loss 0.266556 LR 0.001000 Time 0.020257 -2022-12-06 10:52:56,082 - Epoch: [68][ 860/ 1200] Overall Loss 0.266398 Objective Loss 0.266398 LR 0.001000 Time 0.020246 -2022-12-06 10:52:56,276 - Epoch: [68][ 870/ 1200] Overall Loss 0.266416 Objective Loss 0.266416 LR 0.001000 Time 0.020235 -2022-12-06 10:52:56,470 - Epoch: [68][ 880/ 1200] Overall Loss 0.266556 Objective Loss 0.266556 LR 0.001000 Time 0.020224 -2022-12-06 10:52:56,663 - Epoch: [68][ 890/ 1200] Overall Loss 0.266287 Objective Loss 0.266287 LR 0.001000 Time 0.020214 -2022-12-06 10:52:56,857 - Epoch: [68][ 900/ 1200] Overall Loss 0.266014 Objective Loss 0.266014 LR 0.001000 Time 0.020204 -2022-12-06 10:52:57,050 - Epoch: [68][ 910/ 1200] Overall Loss 0.266022 Objective Loss 0.266022 LR 0.001000 Time 0.020193 -2022-12-06 10:52:57,243 - Epoch: [68][ 920/ 1200] Overall Loss 0.266471 Objective Loss 0.266471 LR 0.001000 Time 0.020184 -2022-12-06 10:52:57,436 - Epoch: [68][ 930/ 1200] Overall Loss 0.266763 Objective Loss 0.266763 LR 0.001000 Time 0.020173 -2022-12-06 10:52:57,630 - Epoch: [68][ 940/ 1200] Overall Loss 0.266545 Objective Loss 0.266545 LR 0.001000 Time 0.020164 -2022-12-06 10:52:57,823 - Epoch: [68][ 950/ 1200] Overall Loss 0.266578 Objective Loss 0.266578 LR 0.001000 Time 0.020155 -2022-12-06 10:52:58,017 - Epoch: [68][ 960/ 1200] Overall Loss 0.266457 Objective Loss 0.266457 LR 0.001000 Time 0.020146 -2022-12-06 10:52:58,210 - Epoch: [68][ 970/ 1200] Overall Loss 0.266347 Objective Loss 0.266347 LR 0.001000 Time 0.020137 -2022-12-06 10:52:58,404 - Epoch: [68][ 980/ 1200] Overall Loss 0.266705 Objective Loss 0.266705 LR 0.001000 Time 0.020129 -2022-12-06 10:52:58,597 - Epoch: [68][ 990/ 1200] Overall Loss 0.266837 Objective Loss 0.266837 LR 0.001000 Time 0.020120 -2022-12-06 10:52:58,791 - Epoch: [68][ 1000/ 1200] Overall Loss 0.266904 Objective Loss 0.266904 LR 0.001000 Time 0.020112 -2022-12-06 10:52:58,985 - Epoch: [68][ 1010/ 1200] Overall Loss 0.267008 Objective Loss 0.267008 LR 0.001000 Time 0.020104 -2022-12-06 10:52:59,178 - Epoch: [68][ 1020/ 1200] Overall Loss 0.267123 Objective Loss 0.267123 LR 0.001000 Time 0.020096 -2022-12-06 10:52:59,372 - Epoch: [68][ 1030/ 1200] Overall Loss 0.267049 Objective Loss 0.267049 LR 0.001000 Time 0.020088 -2022-12-06 10:52:59,565 - Epoch: [68][ 1040/ 1200] Overall Loss 0.267293 Objective Loss 0.267293 LR 0.001000 Time 0.020081 -2022-12-06 10:52:59,758 - Epoch: [68][ 1050/ 1200] Overall Loss 0.267485 Objective Loss 0.267485 LR 0.001000 Time 0.020073 -2022-12-06 10:52:59,951 - Epoch: [68][ 1060/ 1200] Overall Loss 0.267474 Objective Loss 0.267474 LR 0.001000 Time 0.020065 -2022-12-06 10:53:00,145 - Epoch: [68][ 1070/ 1200] Overall Loss 0.267495 Objective Loss 0.267495 LR 0.001000 Time 0.020058 -2022-12-06 10:53:00,338 - Epoch: [68][ 1080/ 1200] Overall Loss 0.267424 Objective Loss 0.267424 LR 0.001000 Time 0.020050 -2022-12-06 10:53:00,531 - Epoch: [68][ 1090/ 1200] Overall Loss 0.267398 Objective Loss 0.267398 LR 0.001000 Time 0.020043 -2022-12-06 10:53:00,725 - Epoch: [68][ 1100/ 1200] Overall Loss 0.267427 Objective Loss 0.267427 LR 0.001000 Time 0.020037 -2022-12-06 10:53:00,918 - Epoch: [68][ 1110/ 1200] Overall Loss 0.267515 Objective Loss 0.267515 LR 0.001000 Time 0.020030 -2022-12-06 10:53:01,112 - Epoch: [68][ 1120/ 1200] Overall Loss 0.267714 Objective Loss 0.267714 LR 0.001000 Time 0.020024 -2022-12-06 10:53:01,306 - Epoch: [68][ 1130/ 1200] Overall Loss 0.267766 Objective Loss 0.267766 LR 0.001000 Time 0.020017 -2022-12-06 10:53:01,500 - Epoch: [68][ 1140/ 1200] Overall Loss 0.267869 Objective Loss 0.267869 LR 0.001000 Time 0.020011 -2022-12-06 10:53:01,692 - Epoch: [68][ 1150/ 1200] Overall Loss 0.267657 Objective Loss 0.267657 LR 0.001000 Time 0.020004 -2022-12-06 10:53:01,886 - Epoch: [68][ 1160/ 1200] Overall Loss 0.267696 Objective Loss 0.267696 LR 0.001000 Time 0.019998 -2022-12-06 10:53:02,079 - Epoch: [68][ 1170/ 1200] Overall Loss 0.267930 Objective Loss 0.267930 LR 0.001000 Time 0.019992 -2022-12-06 10:53:02,273 - Epoch: [68][ 1180/ 1200] Overall Loss 0.268147 Objective Loss 0.268147 LR 0.001000 Time 0.019986 -2022-12-06 10:53:02,466 - Epoch: [68][ 1190/ 1200] Overall Loss 0.268030 Objective Loss 0.268030 LR 0.001000 Time 0.019980 -2022-12-06 10:53:02,693 - Epoch: [68][ 1200/ 1200] Overall Loss 0.268053 Objective Loss 0.268053 Top1 87.029289 Top5 97.698745 LR 0.001000 Time 0.020002 -2022-12-06 10:53:02,781 - --- validate (epoch=68)----------- -2022-12-06 10:53:02,781 - 34129 samples (256 per mini-batch) -2022-12-06 10:53:03,231 - Epoch: [68][ 10/ 134] Loss 0.302819 Top1 85.195312 Top5 97.617188 -2022-12-06 10:53:03,360 - Epoch: [68][ 20/ 134] Loss 0.303861 Top1 84.121094 Top5 97.949219 -2022-12-06 10:53:03,491 - Epoch: [68][ 30/ 134] Loss 0.301167 Top1 84.153646 Top5 97.955729 -2022-12-06 10:53:03,624 - Epoch: [68][ 40/ 134] Loss 0.296811 Top1 84.335938 Top5 97.998047 -2022-12-06 10:53:03,753 - Epoch: [68][ 50/ 134] Loss 0.298498 Top1 84.437500 Top5 97.984375 -2022-12-06 10:53:03,884 - Epoch: [68][ 60/ 134] Loss 0.300521 Top1 84.361979 Top5 97.981771 -2022-12-06 10:53:04,013 - Epoch: [68][ 70/ 134] Loss 0.303840 Top1 84.391741 Top5 97.940848 -2022-12-06 10:53:04,144 - Epoch: [68][ 80/ 134] Loss 0.302564 Top1 84.472656 Top5 97.910156 -2022-12-06 10:53:04,272 - Epoch: [68][ 90/ 134] Loss 0.301322 Top1 84.401042 Top5 97.868924 -2022-12-06 10:53:04,402 - Epoch: [68][ 100/ 134] Loss 0.304711 Top1 84.375000 Top5 97.871094 -2022-12-06 10:53:04,533 - Epoch: [68][ 110/ 134] Loss 0.305090 Top1 84.332386 Top5 97.890625 -2022-12-06 10:53:04,665 - Epoch: [68][ 120/ 134] Loss 0.306996 Top1 84.309896 Top5 97.884115 -2022-12-06 10:53:04,797 - Epoch: [68][ 130/ 134] Loss 0.309855 Top1 84.278846 Top5 97.884615 -2022-12-06 10:53:04,835 - Epoch: [68][ 134/ 134] Loss 0.309604 Top1 84.291951 Top5 97.896217 -2022-12-06 10:53:04,928 - ==> Top1: 84.292 Top5: 97.896 Loss: 0.310 - -2022-12-06 10:53:04,929 - ==> Confusion: -[[ 931 0 1 3 4 4 0 2 2 38 0 2 1 1 1 2 0 0 0 4 0] - [ 3 912 2 4 14 24 3 19 2 1 4 7 1 0 1 1 11 2 5 6 5] - [ 10 2 1012 6 2 2 18 12 1 3 4 7 0 1 0 2 3 1 7 3 7] - [ 4 2 25 906 1 4 1 1 1 0 10 1 3 4 18 1 2 4 21 1 10] - [ 23 2 3 1 942 2 1 1 1 7 1 3 0 1 11 4 10 1 0 2 4] - [ 9 11 1 3 11 945 4 22 4 2 1 13 5 8 1 1 5 0 1 10 12] - [ 1 1 26 2 1 3 1048 8 0 0 2 3 0 0 0 8 1 1 2 9 2] - [ 1 8 6 1 1 29 7 950 1 1 0 9 1 2 0 1 2 1 17 12 4] - [ 11 1 1 0 1 4 1 2 950 56 9 4 1 6 5 0 5 1 1 2 3] - [ 115 0 1 0 3 3 0 2 15 848 1 2 0 7 1 1 0 0 0 0 2] - [ 1 1 6 6 1 1 0 6 21 3 934 3 3 14 2 0 2 0 7 2 6] - [ 3 0 3 0 0 10 3 4 0 1 0 974 20 3 0 6 7 4 0 6 7] - [ 1 1 1 4 0 0 2 0 1 0 0 55 856 1 0 12 2 20 0 6 7] - [ 2 1 2 0 2 8 1 4 13 20 7 8 4 928 1 3 7 0 0 5 7] - [ 22 0 1 8 4 1 1 1 26 8 2 2 2 6 1024 0 3 2 7 2 8] - [ 1 0 3 1 0 0 6 0 0 1 0 10 3 3 0 981 11 15 0 5 3] - [ 4 2 2 1 2 0 0 1 1 1 0 6 0 1 0 8 1034 1 1 6 1] - [ 3 0 1 4 0 0 2 1 4 4 0 8 8 3 2 12 2 980 0 1 1] - [ 5 6 6 7 2 1 0 33 1 0 5 2 3 1 9 0 3 1 912 6 5] - [ 2 1 2 1 0 7 9 9 0 1 0 23 5 6 1 4 9 6 1 988 5] - [ 239 234 187 87 172 150 96 178 104 135 149 156 313 276 131 117 306 112 145 233 9706]] - -2022-12-06 10:53:05,500 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:53:05,501 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:53:05,507 - - -2022-12-06 10:53:05,507 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:53:06,430 - Epoch: [69][ 10/ 1200] Overall Loss 0.273556 Objective Loss 0.273556 LR 0.001000 Time 0.092235 -2022-12-06 10:53:06,624 - Epoch: [69][ 20/ 1200] Overall Loss 0.270785 Objective Loss 0.270785 LR 0.001000 Time 0.055812 -2022-12-06 10:53:06,815 - Epoch: [69][ 30/ 1200] Overall Loss 0.267426 Objective Loss 0.267426 LR 0.001000 Time 0.043527 -2022-12-06 10:53:07,005 - Epoch: [69][ 40/ 1200] Overall Loss 0.264259 Objective Loss 0.264259 LR 0.001000 Time 0.037400 -2022-12-06 10:53:07,196 - Epoch: [69][ 50/ 1200] Overall Loss 0.261835 Objective Loss 0.261835 LR 0.001000 Time 0.033718 -2022-12-06 10:53:07,385 - Epoch: [69][ 60/ 1200] Overall Loss 0.259231 Objective Loss 0.259231 LR 0.001000 Time 0.031250 -2022-12-06 10:53:07,575 - Epoch: [69][ 70/ 1200] Overall Loss 0.259365 Objective Loss 0.259365 LR 0.001000 Time 0.029489 -2022-12-06 10:53:07,765 - Epoch: [69][ 80/ 1200] Overall Loss 0.262708 Objective Loss 0.262708 LR 0.001000 Time 0.028170 -2022-12-06 10:53:07,956 - Epoch: [69][ 90/ 1200] Overall Loss 0.262698 Objective Loss 0.262698 LR 0.001000 Time 0.027152 -2022-12-06 10:53:08,146 - Epoch: [69][ 100/ 1200] Overall Loss 0.260710 Objective Loss 0.260710 LR 0.001000 Time 0.026338 -2022-12-06 10:53:08,337 - Epoch: [69][ 110/ 1200] Overall Loss 0.261474 Objective Loss 0.261474 LR 0.001000 Time 0.025669 -2022-12-06 10:53:08,526 - Epoch: [69][ 120/ 1200] Overall Loss 0.259645 Objective Loss 0.259645 LR 0.001000 Time 0.025107 -2022-12-06 10:53:08,717 - Epoch: [69][ 130/ 1200] Overall Loss 0.259601 Objective Loss 0.259601 LR 0.001000 Time 0.024638 -2022-12-06 10:53:08,907 - Epoch: [69][ 140/ 1200] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.001000 Time 0.024233 -2022-12-06 10:53:09,097 - Epoch: [69][ 150/ 1200] Overall Loss 0.259181 Objective Loss 0.259181 LR 0.001000 Time 0.023879 -2022-12-06 10:53:09,287 - Epoch: [69][ 160/ 1200] Overall Loss 0.257968 Objective Loss 0.257968 LR 0.001000 Time 0.023569 -2022-12-06 10:53:09,477 - Epoch: [69][ 170/ 1200] Overall Loss 0.257299 Objective Loss 0.257299 LR 0.001000 Time 0.023298 -2022-12-06 10:53:09,667 - Epoch: [69][ 180/ 1200] Overall Loss 0.258176 Objective Loss 0.258176 LR 0.001000 Time 0.023055 -2022-12-06 10:53:09,857 - Epoch: [69][ 190/ 1200] Overall Loss 0.258289 Objective Loss 0.258289 LR 0.001000 Time 0.022838 -2022-12-06 10:53:10,046 - Epoch: [69][ 200/ 1200] Overall Loss 0.258971 Objective Loss 0.258971 LR 0.001000 Time 0.022642 -2022-12-06 10:53:10,236 - Epoch: [69][ 210/ 1200] Overall Loss 0.259395 Objective Loss 0.259395 LR 0.001000 Time 0.022464 -2022-12-06 10:53:10,426 - Epoch: [69][ 220/ 1200] Overall Loss 0.259544 Objective Loss 0.259544 LR 0.001000 Time 0.022303 -2022-12-06 10:53:10,616 - Epoch: [69][ 230/ 1200] Overall Loss 0.260850 Objective Loss 0.260850 LR 0.001000 Time 0.022159 -2022-12-06 10:53:10,807 - Epoch: [69][ 240/ 1200] Overall Loss 0.260632 Objective Loss 0.260632 LR 0.001000 Time 0.022028 -2022-12-06 10:53:10,997 - Epoch: [69][ 250/ 1200] Overall Loss 0.261572 Objective Loss 0.261572 LR 0.001000 Time 0.021904 -2022-12-06 10:53:11,187 - Epoch: [69][ 260/ 1200] Overall Loss 0.262462 Objective Loss 0.262462 LR 0.001000 Time 0.021791 -2022-12-06 10:53:11,376 - Epoch: [69][ 270/ 1200] Overall Loss 0.262238 Objective Loss 0.262238 LR 0.001000 Time 0.021684 -2022-12-06 10:53:11,566 - Epoch: [69][ 280/ 1200] Overall Loss 0.262887 Objective Loss 0.262887 LR 0.001000 Time 0.021585 -2022-12-06 10:53:11,756 - Epoch: [69][ 290/ 1200] Overall Loss 0.263850 Objective Loss 0.263850 LR 0.001000 Time 0.021494 -2022-12-06 10:53:11,946 - Epoch: [69][ 300/ 1200] Overall Loss 0.263441 Objective Loss 0.263441 LR 0.001000 Time 0.021409 -2022-12-06 10:53:12,137 - Epoch: [69][ 310/ 1200] Overall Loss 0.263793 Objective Loss 0.263793 LR 0.001000 Time 0.021332 -2022-12-06 10:53:12,327 - Epoch: [69][ 320/ 1200] Overall Loss 0.264116 Objective Loss 0.264116 LR 0.001000 Time 0.021258 -2022-12-06 10:53:12,517 - Epoch: [69][ 330/ 1200] Overall Loss 0.263952 Objective Loss 0.263952 LR 0.001000 Time 0.021190 -2022-12-06 10:53:12,707 - Epoch: [69][ 340/ 1200] Overall Loss 0.264529 Objective Loss 0.264529 LR 0.001000 Time 0.021123 -2022-12-06 10:53:12,897 - Epoch: [69][ 350/ 1200] Overall Loss 0.264933 Objective Loss 0.264933 LR 0.001000 Time 0.021061 -2022-12-06 10:53:13,088 - Epoch: [69][ 360/ 1200] Overall Loss 0.265834 Objective Loss 0.265834 LR 0.001000 Time 0.021003 -2022-12-06 10:53:13,278 - Epoch: [69][ 370/ 1200] Overall Loss 0.266071 Objective Loss 0.266071 LR 0.001000 Time 0.020948 -2022-12-06 10:53:13,468 - Epoch: [69][ 380/ 1200] Overall Loss 0.266736 Objective Loss 0.266736 LR 0.001000 Time 0.020895 -2022-12-06 10:53:13,658 - Epoch: [69][ 390/ 1200] Overall Loss 0.267234 Objective Loss 0.267234 LR 0.001000 Time 0.020845 -2022-12-06 10:53:13,847 - Epoch: [69][ 400/ 1200] Overall Loss 0.267123 Objective Loss 0.267123 LR 0.001000 Time 0.020797 -2022-12-06 10:53:14,037 - Epoch: [69][ 410/ 1200] Overall Loss 0.266660 Objective Loss 0.266660 LR 0.001000 Time 0.020752 -2022-12-06 10:53:14,228 - Epoch: [69][ 420/ 1200] Overall Loss 0.267594 Objective Loss 0.267594 LR 0.001000 Time 0.020709 -2022-12-06 10:53:14,418 - Epoch: [69][ 430/ 1200] Overall Loss 0.268121 Objective Loss 0.268121 LR 0.001000 Time 0.020668 -2022-12-06 10:53:14,608 - Epoch: [69][ 440/ 1200] Overall Loss 0.268085 Objective Loss 0.268085 LR 0.001000 Time 0.020629 -2022-12-06 10:53:14,798 - Epoch: [69][ 450/ 1200] Overall Loss 0.267479 Objective Loss 0.267479 LR 0.001000 Time 0.020593 -2022-12-06 10:53:14,989 - Epoch: [69][ 460/ 1200] Overall Loss 0.267280 Objective Loss 0.267280 LR 0.001000 Time 0.020559 -2022-12-06 10:53:15,179 - Epoch: [69][ 470/ 1200] Overall Loss 0.267157 Objective Loss 0.267157 LR 0.001000 Time 0.020525 -2022-12-06 10:53:15,369 - Epoch: [69][ 480/ 1200] Overall Loss 0.266585 Objective Loss 0.266585 LR 0.001000 Time 0.020492 -2022-12-06 10:53:15,560 - Epoch: [69][ 490/ 1200] Overall Loss 0.266451 Objective Loss 0.266451 LR 0.001000 Time 0.020463 -2022-12-06 10:53:15,750 - Epoch: [69][ 500/ 1200] Overall Loss 0.266291 Objective Loss 0.266291 LR 0.001000 Time 0.020432 -2022-12-06 10:53:15,940 - Epoch: [69][ 510/ 1200] Overall Loss 0.266508 Objective Loss 0.266508 LR 0.001000 Time 0.020404 -2022-12-06 10:53:16,131 - Epoch: [69][ 520/ 1200] Overall Loss 0.266357 Objective Loss 0.266357 LR 0.001000 Time 0.020376 -2022-12-06 10:53:16,321 - Epoch: [69][ 530/ 1200] Overall Loss 0.266402 Objective Loss 0.266402 LR 0.001000 Time 0.020350 -2022-12-06 10:53:16,511 - Epoch: [69][ 540/ 1200] Overall Loss 0.266097 Objective Loss 0.266097 LR 0.001000 Time 0.020324 -2022-12-06 10:53:16,702 - Epoch: [69][ 550/ 1200] Overall Loss 0.266000 Objective Loss 0.266000 LR 0.001000 Time 0.020300 -2022-12-06 10:53:16,891 - Epoch: [69][ 560/ 1200] Overall Loss 0.265951 Objective Loss 0.265951 LR 0.001000 Time 0.020275 -2022-12-06 10:53:17,081 - Epoch: [69][ 570/ 1200] Overall Loss 0.266394 Objective Loss 0.266394 LR 0.001000 Time 0.020252 -2022-12-06 10:53:17,271 - Epoch: [69][ 580/ 1200] Overall Loss 0.266604 Objective Loss 0.266604 LR 0.001000 Time 0.020228 -2022-12-06 10:53:17,461 - Epoch: [69][ 590/ 1200] Overall Loss 0.266313 Objective Loss 0.266313 LR 0.001000 Time 0.020207 -2022-12-06 10:53:17,651 - Epoch: [69][ 600/ 1200] Overall Loss 0.266812 Objective Loss 0.266812 LR 0.001000 Time 0.020186 -2022-12-06 10:53:17,841 - Epoch: [69][ 610/ 1200] Overall Loss 0.266760 Objective Loss 0.266760 LR 0.001000 Time 0.020166 -2022-12-06 10:53:18,031 - Epoch: [69][ 620/ 1200] Overall Loss 0.267161 Objective Loss 0.267161 LR 0.001000 Time 0.020146 -2022-12-06 10:53:18,221 - Epoch: [69][ 630/ 1200] Overall Loss 0.267262 Objective Loss 0.267262 LR 0.001000 Time 0.020127 -2022-12-06 10:53:18,411 - Epoch: [69][ 640/ 1200] Overall Loss 0.267068 Objective Loss 0.267068 LR 0.001000 Time 0.020109 -2022-12-06 10:53:18,601 - Epoch: [69][ 650/ 1200] Overall Loss 0.267343 Objective Loss 0.267343 LR 0.001000 Time 0.020091 -2022-12-06 10:53:18,792 - Epoch: [69][ 660/ 1200] Overall Loss 0.267463 Objective Loss 0.267463 LR 0.001000 Time 0.020074 -2022-12-06 10:53:18,981 - Epoch: [69][ 670/ 1200] Overall Loss 0.267459 Objective Loss 0.267459 LR 0.001000 Time 0.020056 -2022-12-06 10:53:19,171 - Epoch: [69][ 680/ 1200] Overall Loss 0.267599 Objective Loss 0.267599 LR 0.001000 Time 0.020040 -2022-12-06 10:53:19,360 - Epoch: [69][ 690/ 1200] Overall Loss 0.267489 Objective Loss 0.267489 LR 0.001000 Time 0.020023 -2022-12-06 10:53:19,550 - Epoch: [69][ 700/ 1200] Overall Loss 0.266881 Objective Loss 0.266881 LR 0.001000 Time 0.020008 -2022-12-06 10:53:19,739 - Epoch: [69][ 710/ 1200] Overall Loss 0.266815 Objective Loss 0.266815 LR 0.001000 Time 0.019991 -2022-12-06 10:53:19,929 - Epoch: [69][ 720/ 1200] Overall Loss 0.266779 Objective Loss 0.266779 LR 0.001000 Time 0.019976 -2022-12-06 10:53:20,119 - Epoch: [69][ 730/ 1200] Overall Loss 0.267216 Objective Loss 0.267216 LR 0.001000 Time 0.019962 -2022-12-06 10:53:20,308 - Epoch: [69][ 740/ 1200] Overall Loss 0.267206 Objective Loss 0.267206 LR 0.001000 Time 0.019947 -2022-12-06 10:53:20,498 - Epoch: [69][ 750/ 1200] Overall Loss 0.267053 Objective Loss 0.267053 LR 0.001000 Time 0.019934 -2022-12-06 10:53:20,688 - Epoch: [69][ 760/ 1200] Overall Loss 0.266851 Objective Loss 0.266851 LR 0.001000 Time 0.019920 -2022-12-06 10:53:20,877 - Epoch: [69][ 770/ 1200] Overall Loss 0.266892 Objective Loss 0.266892 LR 0.001000 Time 0.019907 -2022-12-06 10:53:21,067 - Epoch: [69][ 780/ 1200] Overall Loss 0.266684 Objective Loss 0.266684 LR 0.001000 Time 0.019895 -2022-12-06 10:53:21,257 - Epoch: [69][ 790/ 1200] Overall Loss 0.266866 Objective Loss 0.266866 LR 0.001000 Time 0.019883 -2022-12-06 10:53:21,447 - Epoch: [69][ 800/ 1200] Overall Loss 0.266829 Objective Loss 0.266829 LR 0.001000 Time 0.019871 -2022-12-06 10:53:21,637 - Epoch: [69][ 810/ 1200] Overall Loss 0.266801 Objective Loss 0.266801 LR 0.001000 Time 0.019860 -2022-12-06 10:53:21,828 - Epoch: [69][ 820/ 1200] Overall Loss 0.266883 Objective Loss 0.266883 LR 0.001000 Time 0.019849 -2022-12-06 10:53:22,018 - Epoch: [69][ 830/ 1200] Overall Loss 0.266399 Objective Loss 0.266399 LR 0.001000 Time 0.019838 -2022-12-06 10:53:22,208 - Epoch: [69][ 840/ 1200] Overall Loss 0.266907 Objective Loss 0.266907 LR 0.001000 Time 0.019828 -2022-12-06 10:53:22,397 - Epoch: [69][ 850/ 1200] Overall Loss 0.266632 Objective Loss 0.266632 LR 0.001000 Time 0.019817 -2022-12-06 10:53:22,587 - Epoch: [69][ 860/ 1200] Overall Loss 0.266983 Objective Loss 0.266983 LR 0.001000 Time 0.019806 -2022-12-06 10:53:22,777 - Epoch: [69][ 870/ 1200] Overall Loss 0.267168 Objective Loss 0.267168 LR 0.001000 Time 0.019797 -2022-12-06 10:53:22,967 - Epoch: [69][ 880/ 1200] Overall Loss 0.267460 Objective Loss 0.267460 LR 0.001000 Time 0.019787 -2022-12-06 10:53:23,157 - Epoch: [69][ 890/ 1200] Overall Loss 0.267608 Objective Loss 0.267608 LR 0.001000 Time 0.019777 -2022-12-06 10:53:23,347 - Epoch: [69][ 900/ 1200] Overall Loss 0.267473 Objective Loss 0.267473 LR 0.001000 Time 0.019768 -2022-12-06 10:53:23,537 - Epoch: [69][ 910/ 1200] Overall Loss 0.267605 Objective Loss 0.267605 LR 0.001000 Time 0.019759 -2022-12-06 10:53:23,726 - Epoch: [69][ 920/ 1200] Overall Loss 0.267655 Objective Loss 0.267655 LR 0.001000 Time 0.019750 -2022-12-06 10:53:23,916 - Epoch: [69][ 930/ 1200] Overall Loss 0.267558 Objective Loss 0.267558 LR 0.001000 Time 0.019741 -2022-12-06 10:53:24,106 - Epoch: [69][ 940/ 1200] Overall Loss 0.267565 Objective Loss 0.267565 LR 0.001000 Time 0.019732 -2022-12-06 10:53:24,296 - Epoch: [69][ 950/ 1200] Overall Loss 0.267744 Objective Loss 0.267744 LR 0.001000 Time 0.019724 -2022-12-06 10:53:24,486 - Epoch: [69][ 960/ 1200] Overall Loss 0.267557 Objective Loss 0.267557 LR 0.001000 Time 0.019715 -2022-12-06 10:53:24,676 - Epoch: [69][ 970/ 1200] Overall Loss 0.267441 Objective Loss 0.267441 LR 0.001000 Time 0.019707 -2022-12-06 10:53:24,865 - Epoch: [69][ 980/ 1200] Overall Loss 0.267511 Objective Loss 0.267511 LR 0.001000 Time 0.019699 -2022-12-06 10:53:25,055 - Epoch: [69][ 990/ 1200] Overall Loss 0.267350 Objective Loss 0.267350 LR 0.001000 Time 0.019691 -2022-12-06 10:53:25,245 - Epoch: [69][ 1000/ 1200] Overall Loss 0.267389 Objective Loss 0.267389 LR 0.001000 Time 0.019683 -2022-12-06 10:53:25,434 - Epoch: [69][ 1010/ 1200] Overall Loss 0.267502 Objective Loss 0.267502 LR 0.001000 Time 0.019676 -2022-12-06 10:53:25,624 - Epoch: [69][ 1020/ 1200] Overall Loss 0.267352 Objective Loss 0.267352 LR 0.001000 Time 0.019668 -2022-12-06 10:53:25,814 - Epoch: [69][ 1030/ 1200] Overall Loss 0.267799 Objective Loss 0.267799 LR 0.001000 Time 0.019661 -2022-12-06 10:53:26,003 - Epoch: [69][ 1040/ 1200] Overall Loss 0.267918 Objective Loss 0.267918 LR 0.001000 Time 0.019654 -2022-12-06 10:53:26,193 - Epoch: [69][ 1050/ 1200] Overall Loss 0.268316 Objective Loss 0.268316 LR 0.001000 Time 0.019647 -2022-12-06 10:53:26,383 - Epoch: [69][ 1060/ 1200] Overall Loss 0.268279 Objective Loss 0.268279 LR 0.001000 Time 0.019640 -2022-12-06 10:53:26,573 - Epoch: [69][ 1070/ 1200] Overall Loss 0.268330 Objective Loss 0.268330 LR 0.001000 Time 0.019634 -2022-12-06 10:53:26,764 - Epoch: [69][ 1080/ 1200] Overall Loss 0.268496 Objective Loss 0.268496 LR 0.001000 Time 0.019628 -2022-12-06 10:53:26,954 - Epoch: [69][ 1090/ 1200] Overall Loss 0.268582 Objective Loss 0.268582 LR 0.001000 Time 0.019621 -2022-12-06 10:53:27,144 - Epoch: [69][ 1100/ 1200] Overall Loss 0.268484 Objective Loss 0.268484 LR 0.001000 Time 0.019615 -2022-12-06 10:53:27,333 - Epoch: [69][ 1110/ 1200] Overall Loss 0.268832 Objective Loss 0.268832 LR 0.001000 Time 0.019609 -2022-12-06 10:53:27,523 - Epoch: [69][ 1120/ 1200] Overall Loss 0.268889 Objective Loss 0.268889 LR 0.001000 Time 0.019603 -2022-12-06 10:53:27,713 - Epoch: [69][ 1130/ 1200] Overall Loss 0.269153 Objective Loss 0.269153 LR 0.001000 Time 0.019597 -2022-12-06 10:53:27,903 - Epoch: [69][ 1140/ 1200] Overall Loss 0.269510 Objective Loss 0.269510 LR 0.001000 Time 0.019592 -2022-12-06 10:53:28,093 - Epoch: [69][ 1150/ 1200] Overall Loss 0.269309 Objective Loss 0.269309 LR 0.001000 Time 0.019585 -2022-12-06 10:53:28,282 - Epoch: [69][ 1160/ 1200] Overall Loss 0.269315 Objective Loss 0.269315 LR 0.001000 Time 0.019580 -2022-12-06 10:53:28,472 - Epoch: [69][ 1170/ 1200] Overall Loss 0.269262 Objective Loss 0.269262 LR 0.001000 Time 0.019574 -2022-12-06 10:53:28,662 - Epoch: [69][ 1180/ 1200] Overall Loss 0.269232 Objective Loss 0.269232 LR 0.001000 Time 0.019569 -2022-12-06 10:53:28,852 - Epoch: [69][ 1190/ 1200] Overall Loss 0.269331 Objective Loss 0.269331 LR 0.001000 Time 0.019564 -2022-12-06 10:53:29,077 - Epoch: [69][ 1200/ 1200] Overall Loss 0.269601 Objective Loss 0.269601 Top1 83.682008 Top5 98.117155 LR 0.001000 Time 0.019588 -2022-12-06 10:53:29,165 - --- validate (epoch=69)----------- -2022-12-06 10:53:29,165 - 34129 samples (256 per mini-batch) -2022-12-06 10:53:29,617 - Epoch: [69][ 10/ 134] Loss 0.329274 Top1 82.382812 Top5 98.164062 -2022-12-06 10:53:29,754 - Epoch: [69][ 20/ 134] Loss 0.332575 Top1 82.226562 Top5 97.890625 -2022-12-06 10:53:29,887 - Epoch: [69][ 30/ 134] Loss 0.320111 Top1 82.486979 Top5 97.851562 -2022-12-06 10:53:30,021 - Epoch: [69][ 40/ 134] Loss 0.319585 Top1 82.744141 Top5 97.763672 -2022-12-06 10:53:30,153 - Epoch: [69][ 50/ 134] Loss 0.317367 Top1 82.828125 Top5 97.765625 -2022-12-06 10:53:30,287 - Epoch: [69][ 60/ 134] Loss 0.314148 Top1 83.027344 Top5 97.792969 -2022-12-06 10:53:30,419 - Epoch: [69][ 70/ 134] Loss 0.315610 Top1 83.024554 Top5 97.812500 -2022-12-06 10:53:30,554 - Epoch: [69][ 80/ 134] Loss 0.314399 Top1 83.071289 Top5 97.802734 -2022-12-06 10:53:30,685 - Epoch: [69][ 90/ 134] Loss 0.315243 Top1 83.094618 Top5 97.803819 -2022-12-06 10:53:30,819 - Epoch: [69][ 100/ 134] Loss 0.313569 Top1 83.214844 Top5 97.812500 -2022-12-06 10:53:30,951 - Epoch: [69][ 110/ 134] Loss 0.313845 Top1 83.256392 Top5 97.801847 -2022-12-06 10:53:31,086 - Epoch: [69][ 120/ 134] Loss 0.312578 Top1 83.414714 Top5 97.796224 -2022-12-06 10:53:31,219 - Epoch: [69][ 130/ 134] Loss 0.310182 Top1 83.458534 Top5 97.857572 -2022-12-06 10:53:31,258 - Epoch: [69][ 134/ 134] Loss 0.307683 Top1 83.506695 Top5 97.872777 -2022-12-06 10:53:31,346 - ==> Top1: 83.507 Top5: 97.873 Loss: 0.308 - -2022-12-06 10:53:31,347 - ==> Confusion: -[[ 890 2 2 8 13 7 0 3 9 39 0 2 0 2 8 4 3 0 2 0 2] - [ 0 937 3 4 8 33 1 8 0 0 2 4 3 2 1 0 5 1 4 3 8] - [ 7 2 1004 12 4 1 24 8 0 2 2 5 4 1 3 3 4 0 4 5 8] - [ 4 1 29 915 2 2 2 0 1 0 9 0 7 2 9 1 3 3 22 1 7] - [ 8 8 3 0 948 4 0 1 0 6 2 3 1 2 12 5 11 1 0 2 3] - [ 3 17 1 4 9 970 1 7 2 2 0 18 5 11 2 3 2 0 3 5 4] - [ 0 2 17 4 1 5 1067 0 0 0 0 3 1 0 0 4 1 0 3 8 2] - [ 0 9 12 5 2 38 6 922 0 0 0 10 1 0 1 0 2 1 31 10 4] - [ 5 5 2 0 3 2 0 0 939 46 10 1 3 12 18 1 8 2 4 1 2] - [ 74 0 5 0 6 2 0 2 32 849 2 2 0 16 2 1 0 3 1 1 3] - [ 0 3 4 19 1 2 6 2 7 2 931 4 1 10 3 0 2 1 16 2 3] - [ 2 3 4 0 1 9 3 3 0 0 0 978 23 2 1 5 6 1 2 8 0] - [ 2 2 2 5 1 1 2 1 0 0 0 70 842 1 2 12 1 13 0 4 8] - [ 0 0 1 0 1 12 0 2 12 10 6 8 4 943 0 5 7 1 2 4 5] - [ 10 3 4 14 5 4 0 1 20 5 1 3 6 6 1035 0 1 0 5 2 5] - [ 0 0 2 0 3 2 3 0 0 0 1 14 3 2 0 993 10 5 0 3 2] - [ 3 3 1 1 1 0 0 0 0 1 2 4 0 2 1 18 1030 1 1 3 0] - [ 1 3 2 4 1 0 1 1 0 4 0 23 25 2 0 24 2 937 2 2 2] - [ 2 7 3 11 2 3 0 22 2 0 6 5 1 0 8 0 3 1 926 3 3] - [ 2 4 0 0 2 5 7 1 0 0 0 29 5 3 1 5 4 2 1 1006 3] - [ 124 267 211 132 138 206 73 139 77 94 141 189 405 350 177 150 345 82 198 294 9434]] - -2022-12-06 10:53:32,014 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:53:32,014 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:53:32,020 - - -2022-12-06 10:53:32,020 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:53:32,954 - Epoch: [70][ 10/ 1200] Overall Loss 0.287267 Objective Loss 0.287267 LR 0.001000 Time 0.093271 -2022-12-06 10:53:33,159 - Epoch: [70][ 20/ 1200] Overall Loss 0.261449 Objective Loss 0.261449 LR 0.001000 Time 0.056862 -2022-12-06 10:53:33,360 - Epoch: [70][ 30/ 1200] Overall Loss 0.256225 Objective Loss 0.256225 LR 0.001000 Time 0.044590 -2022-12-06 10:53:33,564 - Epoch: [70][ 40/ 1200] Overall Loss 0.255552 Objective Loss 0.255552 LR 0.001000 Time 0.038513 -2022-12-06 10:53:33,764 - Epoch: [70][ 50/ 1200] Overall Loss 0.264622 Objective Loss 0.264622 LR 0.001000 Time 0.034815 -2022-12-06 10:53:33,969 - Epoch: [70][ 60/ 1200] Overall Loss 0.262311 Objective Loss 0.262311 LR 0.001000 Time 0.032415 -2022-12-06 10:53:34,169 - Epoch: [70][ 70/ 1200] Overall Loss 0.264980 Objective Loss 0.264980 LR 0.001000 Time 0.030636 -2022-12-06 10:53:34,374 - Epoch: [70][ 80/ 1200] Overall Loss 0.265415 Objective Loss 0.265415 LR 0.001000 Time 0.029360 -2022-12-06 10:53:34,574 - Epoch: [70][ 90/ 1200] Overall Loss 0.265390 Objective Loss 0.265390 LR 0.001000 Time 0.028313 -2022-12-06 10:53:34,778 - Epoch: [70][ 100/ 1200] Overall Loss 0.263973 Objective Loss 0.263973 LR 0.001000 Time 0.027511 -2022-12-06 10:53:34,978 - Epoch: [70][ 110/ 1200] Overall Loss 0.263367 Objective Loss 0.263367 LR 0.001000 Time 0.026824 -2022-12-06 10:53:35,183 - Epoch: [70][ 120/ 1200] Overall Loss 0.263432 Objective Loss 0.263432 LR 0.001000 Time 0.026291 -2022-12-06 10:53:35,383 - Epoch: [70][ 130/ 1200] Overall Loss 0.264105 Objective Loss 0.264105 LR 0.001000 Time 0.025803 -2022-12-06 10:53:35,587 - Epoch: [70][ 140/ 1200] Overall Loss 0.263478 Objective Loss 0.263478 LR 0.001000 Time 0.025414 -2022-12-06 10:53:35,783 - Epoch: [70][ 150/ 1200] Overall Loss 0.264903 Objective Loss 0.264903 LR 0.001000 Time 0.025019 -2022-12-06 10:53:35,973 - Epoch: [70][ 160/ 1200] Overall Loss 0.265688 Objective Loss 0.265688 LR 0.001000 Time 0.024645 -2022-12-06 10:53:36,164 - Epoch: [70][ 170/ 1200] Overall Loss 0.266453 Objective Loss 0.266453 LR 0.001000 Time 0.024313 -2022-12-06 10:53:36,355 - Epoch: [70][ 180/ 1200] Overall Loss 0.264897 Objective Loss 0.264897 LR 0.001000 Time 0.024022 -2022-12-06 10:53:36,545 - Epoch: [70][ 190/ 1200] Overall Loss 0.264765 Objective Loss 0.264765 LR 0.001000 Time 0.023755 -2022-12-06 10:53:36,736 - Epoch: [70][ 200/ 1200] Overall Loss 0.266377 Objective Loss 0.266377 LR 0.001000 Time 0.023520 -2022-12-06 10:53:36,928 - Epoch: [70][ 210/ 1200] Overall Loss 0.265926 Objective Loss 0.265926 LR 0.001000 Time 0.023308 -2022-12-06 10:53:37,118 - Epoch: [70][ 220/ 1200] Overall Loss 0.266647 Objective Loss 0.266647 LR 0.001000 Time 0.023113 -2022-12-06 10:53:37,310 - Epoch: [70][ 230/ 1200] Overall Loss 0.266315 Objective Loss 0.266315 LR 0.001000 Time 0.022938 -2022-12-06 10:53:37,502 - Epoch: [70][ 240/ 1200] Overall Loss 0.267204 Objective Loss 0.267204 LR 0.001000 Time 0.022782 -2022-12-06 10:53:37,694 - Epoch: [70][ 250/ 1200] Overall Loss 0.266515 Objective Loss 0.266515 LR 0.001000 Time 0.022634 -2022-12-06 10:53:37,887 - Epoch: [70][ 260/ 1200] Overall Loss 0.266804 Objective Loss 0.266804 LR 0.001000 Time 0.022506 -2022-12-06 10:53:38,078 - Epoch: [70][ 270/ 1200] Overall Loss 0.266196 Objective Loss 0.266196 LR 0.001000 Time 0.022376 -2022-12-06 10:53:38,271 - Epoch: [70][ 280/ 1200] Overall Loss 0.267156 Objective Loss 0.267156 LR 0.001000 Time 0.022265 -2022-12-06 10:53:38,463 - Epoch: [70][ 290/ 1200] Overall Loss 0.268030 Objective Loss 0.268030 LR 0.001000 Time 0.022156 -2022-12-06 10:53:38,656 - Epoch: [70][ 300/ 1200] Overall Loss 0.268004 Objective Loss 0.268004 LR 0.001000 Time 0.022060 -2022-12-06 10:53:38,847 - Epoch: [70][ 310/ 1200] Overall Loss 0.267543 Objective Loss 0.267543 LR 0.001000 Time 0.021964 -2022-12-06 10:53:39,040 - Epoch: [70][ 320/ 1200] Overall Loss 0.267507 Objective Loss 0.267507 LR 0.001000 Time 0.021878 -2022-12-06 10:53:39,231 - Epoch: [70][ 330/ 1200] Overall Loss 0.267235 Objective Loss 0.267235 LR 0.001000 Time 0.021793 -2022-12-06 10:53:39,424 - Epoch: [70][ 340/ 1200] Overall Loss 0.266629 Objective Loss 0.266629 LR 0.001000 Time 0.021716 -2022-12-06 10:53:39,616 - Epoch: [70][ 350/ 1200] Overall Loss 0.266707 Objective Loss 0.266707 LR 0.001000 Time 0.021643 -2022-12-06 10:53:39,809 - Epoch: [70][ 360/ 1200] Overall Loss 0.267411 Objective Loss 0.267411 LR 0.001000 Time 0.021576 -2022-12-06 10:53:40,001 - Epoch: [70][ 370/ 1200] Overall Loss 0.267200 Objective Loss 0.267200 LR 0.001000 Time 0.021510 -2022-12-06 10:53:40,194 - Epoch: [70][ 380/ 1200] Overall Loss 0.266602 Objective Loss 0.266602 LR 0.001000 Time 0.021451 -2022-12-06 10:53:40,385 - Epoch: [70][ 390/ 1200] Overall Loss 0.266385 Objective Loss 0.266385 LR 0.001000 Time 0.021391 -2022-12-06 10:53:40,576 - Epoch: [70][ 400/ 1200] Overall Loss 0.266516 Objective Loss 0.266516 LR 0.001000 Time 0.021333 -2022-12-06 10:53:40,767 - Epoch: [70][ 410/ 1200] Overall Loss 0.267408 Objective Loss 0.267408 LR 0.001000 Time 0.021276 -2022-12-06 10:53:40,958 - Epoch: [70][ 420/ 1200] Overall Loss 0.267076 Objective Loss 0.267076 LR 0.001000 Time 0.021223 -2022-12-06 10:53:41,149 - Epoch: [70][ 430/ 1200] Overall Loss 0.267389 Objective Loss 0.267389 LR 0.001000 Time 0.021172 -2022-12-06 10:53:41,339 - Epoch: [70][ 440/ 1200] Overall Loss 0.267106 Objective Loss 0.267106 LR 0.001000 Time 0.021122 -2022-12-06 10:53:41,529 - Epoch: [70][ 450/ 1200] Overall Loss 0.267335 Objective Loss 0.267335 LR 0.001000 Time 0.021073 -2022-12-06 10:53:41,719 - Epoch: [70][ 460/ 1200] Overall Loss 0.267285 Objective Loss 0.267285 LR 0.001000 Time 0.021028 -2022-12-06 10:53:41,909 - Epoch: [70][ 470/ 1200] Overall Loss 0.266741 Objective Loss 0.266741 LR 0.001000 Time 0.020983 -2022-12-06 10:53:42,101 - Epoch: [70][ 480/ 1200] Overall Loss 0.266921 Objective Loss 0.266921 LR 0.001000 Time 0.020944 -2022-12-06 10:53:42,291 - Epoch: [70][ 490/ 1200] Overall Loss 0.266698 Objective Loss 0.266698 LR 0.001000 Time 0.020904 -2022-12-06 10:53:42,482 - Epoch: [70][ 500/ 1200] Overall Loss 0.265705 Objective Loss 0.265705 LR 0.001000 Time 0.020866 -2022-12-06 10:53:42,672 - Epoch: [70][ 510/ 1200] Overall Loss 0.265049 Objective Loss 0.265049 LR 0.001000 Time 0.020830 -2022-12-06 10:53:42,863 - Epoch: [70][ 520/ 1200] Overall Loss 0.264937 Objective Loss 0.264937 LR 0.001000 Time 0.020795 -2022-12-06 10:53:43,054 - Epoch: [70][ 530/ 1200] Overall Loss 0.264679 Objective Loss 0.264679 LR 0.001000 Time 0.020761 -2022-12-06 10:53:43,245 - Epoch: [70][ 540/ 1200] Overall Loss 0.264795 Objective Loss 0.264795 LR 0.001000 Time 0.020729 -2022-12-06 10:53:43,435 - Epoch: [70][ 550/ 1200] Overall Loss 0.264525 Objective Loss 0.264525 LR 0.001000 Time 0.020698 -2022-12-06 10:53:43,626 - Epoch: [70][ 560/ 1200] Overall Loss 0.264256 Objective Loss 0.264256 LR 0.001000 Time 0.020668 -2022-12-06 10:53:43,816 - Epoch: [70][ 570/ 1200] Overall Loss 0.264119 Objective Loss 0.264119 LR 0.001000 Time 0.020637 -2022-12-06 10:53:44,006 - Epoch: [70][ 580/ 1200] Overall Loss 0.264050 Objective Loss 0.264050 LR 0.001000 Time 0.020608 -2022-12-06 10:53:44,196 - Epoch: [70][ 590/ 1200] Overall Loss 0.263522 Objective Loss 0.263522 LR 0.001000 Time 0.020580 -2022-12-06 10:53:44,387 - Epoch: [70][ 600/ 1200] Overall Loss 0.263568 Objective Loss 0.263568 LR 0.001000 Time 0.020554 -2022-12-06 10:53:44,577 - Epoch: [70][ 610/ 1200] Overall Loss 0.263687 Objective Loss 0.263687 LR 0.001000 Time 0.020528 -2022-12-06 10:53:44,767 - Epoch: [70][ 620/ 1200] Overall Loss 0.263820 Objective Loss 0.263820 LR 0.001000 Time 0.020502 -2022-12-06 10:53:44,956 - Epoch: [70][ 630/ 1200] Overall Loss 0.264547 Objective Loss 0.264547 LR 0.001000 Time 0.020477 -2022-12-06 10:53:45,147 - Epoch: [70][ 640/ 1200] Overall Loss 0.264962 Objective Loss 0.264962 LR 0.001000 Time 0.020454 -2022-12-06 10:53:45,336 - Epoch: [70][ 650/ 1200] Overall Loss 0.265513 Objective Loss 0.265513 LR 0.001000 Time 0.020430 -2022-12-06 10:53:45,527 - Epoch: [70][ 660/ 1200] Overall Loss 0.265787 Objective Loss 0.265787 LR 0.001000 Time 0.020408 -2022-12-06 10:53:45,716 - Epoch: [70][ 670/ 1200] Overall Loss 0.266185 Objective Loss 0.266185 LR 0.001000 Time 0.020386 -2022-12-06 10:53:45,906 - Epoch: [70][ 680/ 1200] Overall Loss 0.266285 Objective Loss 0.266285 LR 0.001000 Time 0.020365 -2022-12-06 10:53:46,096 - Epoch: [70][ 690/ 1200] Overall Loss 0.266813 Objective Loss 0.266813 LR 0.001000 Time 0.020344 -2022-12-06 10:53:46,286 - Epoch: [70][ 700/ 1200] Overall Loss 0.267201 Objective Loss 0.267201 LR 0.001000 Time 0.020323 -2022-12-06 10:53:46,476 - Epoch: [70][ 710/ 1200] Overall Loss 0.267560 Objective Loss 0.267560 LR 0.001000 Time 0.020304 -2022-12-06 10:53:46,666 - Epoch: [70][ 720/ 1200] Overall Loss 0.267838 Objective Loss 0.267838 LR 0.001000 Time 0.020285 -2022-12-06 10:53:46,856 - Epoch: [70][ 730/ 1200] Overall Loss 0.267386 Objective Loss 0.267386 LR 0.001000 Time 0.020267 -2022-12-06 10:53:47,045 - Epoch: [70][ 740/ 1200] Overall Loss 0.267908 Objective Loss 0.267908 LR 0.001000 Time 0.020248 -2022-12-06 10:53:47,235 - Epoch: [70][ 750/ 1200] Overall Loss 0.268017 Objective Loss 0.268017 LR 0.001000 Time 0.020231 -2022-12-06 10:53:47,426 - Epoch: [70][ 760/ 1200] Overall Loss 0.267667 Objective Loss 0.267667 LR 0.001000 Time 0.020215 -2022-12-06 10:53:47,617 - Epoch: [70][ 770/ 1200] Overall Loss 0.267306 Objective Loss 0.267306 LR 0.001000 Time 0.020199 -2022-12-06 10:53:47,807 - Epoch: [70][ 780/ 1200] Overall Loss 0.267483 Objective Loss 0.267483 LR 0.001000 Time 0.020183 -2022-12-06 10:53:47,996 - Epoch: [70][ 790/ 1200] Overall Loss 0.267379 Objective Loss 0.267379 LR 0.001000 Time 0.020167 -2022-12-06 10:53:48,186 - Epoch: [70][ 800/ 1200] Overall Loss 0.267604 Objective Loss 0.267604 LR 0.001000 Time 0.020151 -2022-12-06 10:53:48,376 - Epoch: [70][ 810/ 1200] Overall Loss 0.267908 Objective Loss 0.267908 LR 0.001000 Time 0.020136 -2022-12-06 10:53:48,566 - Epoch: [70][ 820/ 1200] Overall Loss 0.268023 Objective Loss 0.268023 LR 0.001000 Time 0.020122 -2022-12-06 10:53:48,756 - Epoch: [70][ 830/ 1200] Overall Loss 0.268024 Objective Loss 0.268024 LR 0.001000 Time 0.020107 -2022-12-06 10:53:48,946 - Epoch: [70][ 840/ 1200] Overall Loss 0.268117 Objective Loss 0.268117 LR 0.001000 Time 0.020093 -2022-12-06 10:53:49,136 - Epoch: [70][ 850/ 1200] Overall Loss 0.268288 Objective Loss 0.268288 LR 0.001000 Time 0.020080 -2022-12-06 10:53:49,326 - Epoch: [70][ 860/ 1200] Overall Loss 0.267978 Objective Loss 0.267978 LR 0.001000 Time 0.020067 -2022-12-06 10:53:49,515 - Epoch: [70][ 870/ 1200] Overall Loss 0.268359 Objective Loss 0.268359 LR 0.001000 Time 0.020054 -2022-12-06 10:53:49,705 - Epoch: [70][ 880/ 1200] Overall Loss 0.268235 Objective Loss 0.268235 LR 0.001000 Time 0.020041 -2022-12-06 10:53:49,895 - Epoch: [70][ 890/ 1200] Overall Loss 0.268260 Objective Loss 0.268260 LR 0.001000 Time 0.020028 -2022-12-06 10:53:50,085 - Epoch: [70][ 900/ 1200] Overall Loss 0.268474 Objective Loss 0.268474 LR 0.001000 Time 0.020017 -2022-12-06 10:53:50,275 - Epoch: [70][ 910/ 1200] Overall Loss 0.268309 Objective Loss 0.268309 LR 0.001000 Time 0.020005 -2022-12-06 10:53:50,465 - Epoch: [70][ 920/ 1200] Overall Loss 0.268001 Objective Loss 0.268001 LR 0.001000 Time 0.019994 -2022-12-06 10:53:50,656 - Epoch: [70][ 930/ 1200] Overall Loss 0.267953 Objective Loss 0.267953 LR 0.001000 Time 0.019982 -2022-12-06 10:53:50,845 - Epoch: [70][ 940/ 1200] Overall Loss 0.267860 Objective Loss 0.267860 LR 0.001000 Time 0.019971 -2022-12-06 10:53:51,035 - Epoch: [70][ 950/ 1200] Overall Loss 0.268157 Objective Loss 0.268157 LR 0.001000 Time 0.019960 -2022-12-06 10:53:51,226 - Epoch: [70][ 960/ 1200] Overall Loss 0.268400 Objective Loss 0.268400 LR 0.001000 Time 0.019950 -2022-12-06 10:53:51,416 - Epoch: [70][ 970/ 1200] Overall Loss 0.268747 Objective Loss 0.268747 LR 0.001000 Time 0.019940 -2022-12-06 10:53:51,607 - Epoch: [70][ 980/ 1200] Overall Loss 0.268700 Objective Loss 0.268700 LR 0.001000 Time 0.019931 -2022-12-06 10:53:51,797 - Epoch: [70][ 990/ 1200] Overall Loss 0.268726 Objective Loss 0.268726 LR 0.001000 Time 0.019921 -2022-12-06 10:53:51,987 - Epoch: [70][ 1000/ 1200] Overall Loss 0.268836 Objective Loss 0.268836 LR 0.001000 Time 0.019911 -2022-12-06 10:53:52,177 - Epoch: [70][ 1010/ 1200] Overall Loss 0.268776 Objective Loss 0.268776 LR 0.001000 Time 0.019902 -2022-12-06 10:53:52,367 - Epoch: [70][ 1020/ 1200] Overall Loss 0.268584 Objective Loss 0.268584 LR 0.001000 Time 0.019892 -2022-12-06 10:53:52,558 - Epoch: [70][ 1030/ 1200] Overall Loss 0.268425 Objective Loss 0.268425 LR 0.001000 Time 0.019884 -2022-12-06 10:53:52,750 - Epoch: [70][ 1040/ 1200] Overall Loss 0.268371 Objective Loss 0.268371 LR 0.001000 Time 0.019877 -2022-12-06 10:53:52,942 - Epoch: [70][ 1050/ 1200] Overall Loss 0.268361 Objective Loss 0.268361 LR 0.001000 Time 0.019870 -2022-12-06 10:53:53,134 - Epoch: [70][ 1060/ 1200] Overall Loss 0.268307 Objective Loss 0.268307 LR 0.001000 Time 0.019864 -2022-12-06 10:53:53,326 - Epoch: [70][ 1070/ 1200] Overall Loss 0.268434 Objective Loss 0.268434 LR 0.001000 Time 0.019856 -2022-12-06 10:53:53,518 - Epoch: [70][ 1080/ 1200] Overall Loss 0.268479 Objective Loss 0.268479 LR 0.001000 Time 0.019850 -2022-12-06 10:53:53,710 - Epoch: [70][ 1090/ 1200] Overall Loss 0.268595 Objective Loss 0.268595 LR 0.001000 Time 0.019843 -2022-12-06 10:53:53,902 - Epoch: [70][ 1100/ 1200] Overall Loss 0.268675 Objective Loss 0.268675 LR 0.001000 Time 0.019837 -2022-12-06 10:53:54,094 - Epoch: [70][ 1110/ 1200] Overall Loss 0.269028 Objective Loss 0.269028 LR 0.001000 Time 0.019831 -2022-12-06 10:53:54,286 - Epoch: [70][ 1120/ 1200] Overall Loss 0.269105 Objective Loss 0.269105 LR 0.001000 Time 0.019825 -2022-12-06 10:53:54,478 - Epoch: [70][ 1130/ 1200] Overall Loss 0.269264 Objective Loss 0.269264 LR 0.001000 Time 0.019819 -2022-12-06 10:53:54,670 - Epoch: [70][ 1140/ 1200] Overall Loss 0.269505 Objective Loss 0.269505 LR 0.001000 Time 0.019813 -2022-12-06 10:53:54,862 - Epoch: [70][ 1150/ 1200] Overall Loss 0.269608 Objective Loss 0.269608 LR 0.001000 Time 0.019807 -2022-12-06 10:53:55,054 - Epoch: [70][ 1160/ 1200] Overall Loss 0.269523 Objective Loss 0.269523 LR 0.001000 Time 0.019801 -2022-12-06 10:53:55,245 - Epoch: [70][ 1170/ 1200] Overall Loss 0.269586 Objective Loss 0.269586 LR 0.001000 Time 0.019795 -2022-12-06 10:53:55,438 - Epoch: [70][ 1180/ 1200] Overall Loss 0.269614 Objective Loss 0.269614 LR 0.001000 Time 0.019791 -2022-12-06 10:53:55,630 - Epoch: [70][ 1190/ 1200] Overall Loss 0.269576 Objective Loss 0.269576 LR 0.001000 Time 0.019785 -2022-12-06 10:53:55,851 - Epoch: [70][ 1200/ 1200] Overall Loss 0.269600 Objective Loss 0.269600 Top1 85.564854 Top5 97.071130 LR 0.001000 Time 0.019804 -2022-12-06 10:53:55,940 - --- validate (epoch=70)----------- -2022-12-06 10:53:55,940 - 34129 samples (256 per mini-batch) -2022-12-06 10:53:56,382 - Epoch: [70][ 10/ 134] Loss 0.281987 Top1 84.492188 Top5 97.929688 -2022-12-06 10:53:56,516 - Epoch: [70][ 20/ 134] Loss 0.285914 Top1 84.570312 Top5 97.949219 -2022-12-06 10:53:56,647 - Epoch: [70][ 30/ 134] Loss 0.292547 Top1 84.401042 Top5 97.799479 -2022-12-06 10:53:56,781 - Epoch: [70][ 40/ 134] Loss 0.291257 Top1 84.130859 Top5 97.871094 -2022-12-06 10:53:56,913 - Epoch: [70][ 50/ 134] Loss 0.297006 Top1 83.960938 Top5 97.906250 -2022-12-06 10:53:57,045 - Epoch: [70][ 60/ 134] Loss 0.298212 Top1 84.003906 Top5 97.968750 -2022-12-06 10:53:57,177 - Epoch: [70][ 70/ 134] Loss 0.301122 Top1 83.922991 Top5 97.935268 -2022-12-06 10:53:57,306 - Epoch: [70][ 80/ 134] Loss 0.296398 Top1 83.999023 Top5 97.963867 -2022-12-06 10:53:57,436 - Epoch: [70][ 90/ 134] Loss 0.299123 Top1 84.019097 Top5 97.890625 -2022-12-06 10:53:57,569 - Epoch: [70][ 100/ 134] Loss 0.299120 Top1 84.015625 Top5 97.847656 -2022-12-06 10:53:57,702 - Epoch: [70][ 110/ 134] Loss 0.301365 Top1 84.055398 Top5 97.833807 -2022-12-06 10:53:57,830 - Epoch: [70][ 120/ 134] Loss 0.299695 Top1 84.095052 Top5 97.867839 -2022-12-06 10:53:57,961 - Epoch: [70][ 130/ 134] Loss 0.299713 Top1 84.110577 Top5 97.869591 -2022-12-06 10:53:57,999 - Epoch: [70][ 134/ 134] Loss 0.298923 Top1 84.130798 Top5 97.878637 -2022-12-06 10:53:58,095 - ==> Top1: 84.131 Top5: 97.879 Loss: 0.299 - -2022-12-06 10:53:58,096 - ==> Confusion: -[[ 890 0 0 6 3 6 0 0 11 58 0 4 0 3 6 1 2 1 2 0 3] - [ 0 931 0 4 6 8 3 16 2 1 6 7 3 1 2 0 1 3 23 1 9] - [ 7 4 990 12 3 1 24 11 1 3 10 7 2 2 0 2 0 2 10 4 8] - [ 4 0 14 923 0 3 2 3 0 0 14 3 4 3 19 0 2 3 17 0 6] - [ 15 8 5 2 919 6 1 2 3 6 1 3 2 3 18 4 8 2 0 2 10] - [ 2 36 0 2 4 928 4 29 6 1 4 19 5 7 3 0 2 1 3 7 6] - [ 0 1 5 0 0 3 1066 5 0 0 2 3 3 2 0 6 1 4 3 13 1] - [ 0 7 5 5 1 16 6 936 2 0 2 11 2 3 0 1 0 2 37 14 4] - [ 5 3 1 0 0 0 0 3 993 34 6 0 1 5 6 0 1 0 4 1 1] - [ 53 0 1 0 2 1 0 2 65 858 1 3 0 7 4 0 1 0 0 0 3] - [ 1 1 3 4 1 0 2 1 18 1 956 3 1 10 3 1 0 0 9 2 2] - [ 2 1 1 1 0 8 1 2 1 0 1 950 38 3 1 3 6 13 1 9 9] - [ 0 1 1 4 1 0 2 0 2 0 1 22 892 2 1 10 0 25 0 2 3] - [ 1 0 0 0 0 6 0 4 23 14 12 7 5 934 0 1 2 4 2 2 6] - [ 8 4 2 10 1 2 0 0 36 1 4 1 2 5 1042 0 0 1 7 0 4] - [ 3 0 0 1 4 0 5 1 0 0 1 13 6 2 0 968 8 24 0 4 3] - [ 5 4 4 1 2 0 0 1 1 0 2 6 1 1 2 15 1015 2 2 1 7] - [ 2 0 0 3 0 1 0 0 4 3 0 5 18 0 4 7 1 982 2 2 2] - [ 3 5 3 5 1 2 0 18 2 0 10 2 5 0 6 0 1 0 938 2 5] - [ 2 4 1 3 1 7 10 8 0 0 2 15 7 4 1 3 2 7 2 997 4] - [ 105 221 143 127 72 184 77 163 204 105 267 87 395 299 226 97 223 137 268 230 9596]] - -2022-12-06 10:53:58,764 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:53:58,764 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:53:58,770 - - -2022-12-06 10:53:58,770 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:53:59,712 - Epoch: [71][ 10/ 1200] Overall Loss 0.242079 Objective Loss 0.242079 LR 0.001000 Time 0.094117 -2022-12-06 10:53:59,913 - Epoch: [71][ 20/ 1200] Overall Loss 0.259704 Objective Loss 0.259704 LR 0.001000 Time 0.057109 -2022-12-06 10:54:00,105 - Epoch: [71][ 30/ 1200] Overall Loss 0.253542 Objective Loss 0.253542 LR 0.001000 Time 0.044432 -2022-12-06 10:54:00,297 - Epoch: [71][ 40/ 1200] Overall Loss 0.258400 Objective Loss 0.258400 LR 0.001000 Time 0.038109 -2022-12-06 10:54:00,488 - Epoch: [71][ 50/ 1200] Overall Loss 0.261319 Objective Loss 0.261319 LR 0.001000 Time 0.034296 -2022-12-06 10:54:00,678 - Epoch: [71][ 60/ 1200] Overall Loss 0.265121 Objective Loss 0.265121 LR 0.001000 Time 0.031745 -2022-12-06 10:54:00,870 - Epoch: [71][ 70/ 1200] Overall Loss 0.263446 Objective Loss 0.263446 LR 0.001000 Time 0.029948 -2022-12-06 10:54:01,061 - Epoch: [71][ 80/ 1200] Overall Loss 0.267045 Objective Loss 0.267045 LR 0.001000 Time 0.028577 -2022-12-06 10:54:01,252 - Epoch: [71][ 90/ 1200] Overall Loss 0.270702 Objective Loss 0.270702 LR 0.001000 Time 0.027522 -2022-12-06 10:54:01,443 - Epoch: [71][ 100/ 1200] Overall Loss 0.267098 Objective Loss 0.267098 LR 0.001000 Time 0.026670 -2022-12-06 10:54:01,634 - Epoch: [71][ 110/ 1200] Overall Loss 0.266321 Objective Loss 0.266321 LR 0.001000 Time 0.025981 -2022-12-06 10:54:01,826 - Epoch: [71][ 120/ 1200] Overall Loss 0.267455 Objective Loss 0.267455 LR 0.001000 Time 0.025407 -2022-12-06 10:54:02,016 - Epoch: [71][ 130/ 1200] Overall Loss 0.268121 Objective Loss 0.268121 LR 0.001000 Time 0.024910 -2022-12-06 10:54:02,207 - Epoch: [71][ 140/ 1200] Overall Loss 0.268827 Objective Loss 0.268827 LR 0.001000 Time 0.024494 -2022-12-06 10:54:02,398 - Epoch: [71][ 150/ 1200] Overall Loss 0.270012 Objective Loss 0.270012 LR 0.001000 Time 0.024132 -2022-12-06 10:54:02,589 - Epoch: [71][ 160/ 1200] Overall Loss 0.269775 Objective Loss 0.269775 LR 0.001000 Time 0.023815 -2022-12-06 10:54:02,780 - Epoch: [71][ 170/ 1200] Overall Loss 0.269404 Objective Loss 0.269404 LR 0.001000 Time 0.023532 -2022-12-06 10:54:02,971 - Epoch: [71][ 180/ 1200] Overall Loss 0.267962 Objective Loss 0.267962 LR 0.001000 Time 0.023283 -2022-12-06 10:54:03,163 - Epoch: [71][ 190/ 1200] Overall Loss 0.267310 Objective Loss 0.267310 LR 0.001000 Time 0.023063 -2022-12-06 10:54:03,353 - Epoch: [71][ 200/ 1200] Overall Loss 0.268165 Objective Loss 0.268165 LR 0.001000 Time 0.022860 -2022-12-06 10:54:03,545 - Epoch: [71][ 210/ 1200] Overall Loss 0.268395 Objective Loss 0.268395 LR 0.001000 Time 0.022684 -2022-12-06 10:54:03,737 - Epoch: [71][ 220/ 1200] Overall Loss 0.267260 Objective Loss 0.267260 LR 0.001000 Time 0.022523 -2022-12-06 10:54:03,930 - Epoch: [71][ 230/ 1200] Overall Loss 0.267620 Objective Loss 0.267620 LR 0.001000 Time 0.022380 -2022-12-06 10:54:04,122 - Epoch: [71][ 240/ 1200] Overall Loss 0.267776 Objective Loss 0.267776 LR 0.001000 Time 0.022246 -2022-12-06 10:54:04,315 - Epoch: [71][ 250/ 1200] Overall Loss 0.267836 Objective Loss 0.267836 LR 0.001000 Time 0.022124 -2022-12-06 10:54:04,508 - Epoch: [71][ 260/ 1200] Overall Loss 0.267081 Objective Loss 0.267081 LR 0.001000 Time 0.022012 -2022-12-06 10:54:04,701 - Epoch: [71][ 270/ 1200] Overall Loss 0.266154 Objective Loss 0.266154 LR 0.001000 Time 0.021910 -2022-12-06 10:54:04,893 - Epoch: [71][ 280/ 1200] Overall Loss 0.265497 Objective Loss 0.265497 LR 0.001000 Time 0.021813 -2022-12-06 10:54:05,087 - Epoch: [71][ 290/ 1200] Overall Loss 0.265009 Objective Loss 0.265009 LR 0.001000 Time 0.021726 -2022-12-06 10:54:05,279 - Epoch: [71][ 300/ 1200] Overall Loss 0.265920 Objective Loss 0.265920 LR 0.001000 Time 0.021642 -2022-12-06 10:54:05,472 - Epoch: [71][ 310/ 1200] Overall Loss 0.266205 Objective Loss 0.266205 LR 0.001000 Time 0.021564 -2022-12-06 10:54:05,664 - Epoch: [71][ 320/ 1200] Overall Loss 0.265575 Objective Loss 0.265575 LR 0.001000 Time 0.021490 -2022-12-06 10:54:05,857 - Epoch: [71][ 330/ 1200] Overall Loss 0.265790 Objective Loss 0.265790 LR 0.001000 Time 0.021421 -2022-12-06 10:54:06,049 - Epoch: [71][ 340/ 1200] Overall Loss 0.265815 Objective Loss 0.265815 LR 0.001000 Time 0.021355 -2022-12-06 10:54:06,243 - Epoch: [71][ 350/ 1200] Overall Loss 0.266443 Objective Loss 0.266443 LR 0.001000 Time 0.021295 -2022-12-06 10:54:06,435 - Epoch: [71][ 360/ 1200] Overall Loss 0.266311 Objective Loss 0.266311 LR 0.001000 Time 0.021236 -2022-12-06 10:54:06,627 - Epoch: [71][ 370/ 1200] Overall Loss 0.266239 Objective Loss 0.266239 LR 0.001000 Time 0.021181 -2022-12-06 10:54:06,819 - Epoch: [71][ 380/ 1200] Overall Loss 0.265989 Objective Loss 0.265989 LR 0.001000 Time 0.021128 -2022-12-06 10:54:07,012 - Epoch: [71][ 390/ 1200] Overall Loss 0.265248 Objective Loss 0.265248 LR 0.001000 Time 0.021079 -2022-12-06 10:54:07,204 - Epoch: [71][ 400/ 1200] Overall Loss 0.264236 Objective Loss 0.264236 LR 0.001000 Time 0.021031 -2022-12-06 10:54:07,398 - Epoch: [71][ 410/ 1200] Overall Loss 0.264494 Objective Loss 0.264494 LR 0.001000 Time 0.020988 -2022-12-06 10:54:07,590 - Epoch: [71][ 420/ 1200] Overall Loss 0.264307 Objective Loss 0.264307 LR 0.001000 Time 0.020944 -2022-12-06 10:54:07,780 - Epoch: [71][ 430/ 1200] Overall Loss 0.264244 Objective Loss 0.264244 LR 0.001000 Time 0.020897 -2022-12-06 10:54:07,971 - Epoch: [71][ 440/ 1200] Overall Loss 0.264722 Objective Loss 0.264722 LR 0.001000 Time 0.020856 -2022-12-06 10:54:08,162 - Epoch: [71][ 450/ 1200] Overall Loss 0.264623 Objective Loss 0.264623 LR 0.001000 Time 0.020815 -2022-12-06 10:54:08,352 - Epoch: [71][ 460/ 1200] Overall Loss 0.264453 Objective Loss 0.264453 LR 0.001000 Time 0.020776 -2022-12-06 10:54:08,542 - Epoch: [71][ 470/ 1200] Overall Loss 0.264048 Objective Loss 0.264048 LR 0.001000 Time 0.020738 -2022-12-06 10:54:08,733 - Epoch: [71][ 480/ 1200] Overall Loss 0.263538 Objective Loss 0.263538 LR 0.001000 Time 0.020702 -2022-12-06 10:54:08,924 - Epoch: [71][ 490/ 1200] Overall Loss 0.263228 Objective Loss 0.263228 LR 0.001000 Time 0.020667 -2022-12-06 10:54:09,115 - Epoch: [71][ 500/ 1200] Overall Loss 0.263369 Objective Loss 0.263369 LR 0.001000 Time 0.020634 -2022-12-06 10:54:09,306 - Epoch: [71][ 510/ 1200] Overall Loss 0.263257 Objective Loss 0.263257 LR 0.001000 Time 0.020603 -2022-12-06 10:54:09,496 - Epoch: [71][ 520/ 1200] Overall Loss 0.263341 Objective Loss 0.263341 LR 0.001000 Time 0.020572 -2022-12-06 10:54:09,687 - Epoch: [71][ 530/ 1200] Overall Loss 0.263468 Objective Loss 0.263468 LR 0.001000 Time 0.020542 -2022-12-06 10:54:09,878 - Epoch: [71][ 540/ 1200] Overall Loss 0.263741 Objective Loss 0.263741 LR 0.001000 Time 0.020516 -2022-12-06 10:54:10,069 - Epoch: [71][ 550/ 1200] Overall Loss 0.263540 Objective Loss 0.263540 LR 0.001000 Time 0.020488 -2022-12-06 10:54:10,260 - Epoch: [71][ 560/ 1200] Overall Loss 0.263904 Objective Loss 0.263904 LR 0.001000 Time 0.020462 -2022-12-06 10:54:10,450 - Epoch: [71][ 570/ 1200] Overall Loss 0.263649 Objective Loss 0.263649 LR 0.001000 Time 0.020436 -2022-12-06 10:54:10,640 - Epoch: [71][ 580/ 1200] Overall Loss 0.263546 Objective Loss 0.263546 LR 0.001000 Time 0.020411 -2022-12-06 10:54:10,830 - Epoch: [71][ 590/ 1200] Overall Loss 0.263864 Objective Loss 0.263864 LR 0.001000 Time 0.020386 -2022-12-06 10:54:11,021 - Epoch: [71][ 600/ 1200] Overall Loss 0.263749 Objective Loss 0.263749 LR 0.001000 Time 0.020364 -2022-12-06 10:54:11,212 - Epoch: [71][ 610/ 1200] Overall Loss 0.263229 Objective Loss 0.263229 LR 0.001000 Time 0.020341 -2022-12-06 10:54:11,403 - Epoch: [71][ 620/ 1200] Overall Loss 0.263159 Objective Loss 0.263159 LR 0.001000 Time 0.020321 -2022-12-06 10:54:11,593 - Epoch: [71][ 630/ 1200] Overall Loss 0.263356 Objective Loss 0.263356 LR 0.001000 Time 0.020300 -2022-12-06 10:54:11,784 - Epoch: [71][ 640/ 1200] Overall Loss 0.263259 Objective Loss 0.263259 LR 0.001000 Time 0.020279 -2022-12-06 10:54:11,974 - Epoch: [71][ 650/ 1200] Overall Loss 0.263249 Objective Loss 0.263249 LR 0.001000 Time 0.020259 -2022-12-06 10:54:12,165 - Epoch: [71][ 660/ 1200] Overall Loss 0.263346 Objective Loss 0.263346 LR 0.001000 Time 0.020241 -2022-12-06 10:54:12,356 - Epoch: [71][ 670/ 1200] Overall Loss 0.263844 Objective Loss 0.263844 LR 0.001000 Time 0.020222 -2022-12-06 10:54:12,546 - Epoch: [71][ 680/ 1200] Overall Loss 0.263776 Objective Loss 0.263776 LR 0.001000 Time 0.020204 -2022-12-06 10:54:12,736 - Epoch: [71][ 690/ 1200] Overall Loss 0.263617 Objective Loss 0.263617 LR 0.001000 Time 0.020185 -2022-12-06 10:54:12,927 - Epoch: [71][ 700/ 1200] Overall Loss 0.263402 Objective Loss 0.263402 LR 0.001000 Time 0.020169 -2022-12-06 10:54:13,118 - Epoch: [71][ 710/ 1200] Overall Loss 0.263722 Objective Loss 0.263722 LR 0.001000 Time 0.020153 -2022-12-06 10:54:13,308 - Epoch: [71][ 720/ 1200] Overall Loss 0.264388 Objective Loss 0.264388 LR 0.001000 Time 0.020137 -2022-12-06 10:54:13,499 - Epoch: [71][ 730/ 1200] Overall Loss 0.263966 Objective Loss 0.263966 LR 0.001000 Time 0.020122 -2022-12-06 10:54:13,689 - Epoch: [71][ 740/ 1200] Overall Loss 0.264089 Objective Loss 0.264089 LR 0.001000 Time 0.020106 -2022-12-06 10:54:13,879 - Epoch: [71][ 750/ 1200] Overall Loss 0.263764 Objective Loss 0.263764 LR 0.001000 Time 0.020091 -2022-12-06 10:54:14,070 - Epoch: [71][ 760/ 1200] Overall Loss 0.263741 Objective Loss 0.263741 LR 0.001000 Time 0.020076 -2022-12-06 10:54:14,260 - Epoch: [71][ 770/ 1200] Overall Loss 0.263884 Objective Loss 0.263884 LR 0.001000 Time 0.020061 -2022-12-06 10:54:14,450 - Epoch: [71][ 780/ 1200] Overall Loss 0.264036 Objective Loss 0.264036 LR 0.001000 Time 0.020048 -2022-12-06 10:54:14,640 - Epoch: [71][ 790/ 1200] Overall Loss 0.263865 Objective Loss 0.263865 LR 0.001000 Time 0.020034 -2022-12-06 10:54:14,831 - Epoch: [71][ 800/ 1200] Overall Loss 0.263768 Objective Loss 0.263768 LR 0.001000 Time 0.020021 -2022-12-06 10:54:15,021 - Epoch: [71][ 810/ 1200] Overall Loss 0.263744 Objective Loss 0.263744 LR 0.001000 Time 0.020008 -2022-12-06 10:54:15,212 - Epoch: [71][ 820/ 1200] Overall Loss 0.263807 Objective Loss 0.263807 LR 0.001000 Time 0.019996 -2022-12-06 10:54:15,402 - Epoch: [71][ 830/ 1200] Overall Loss 0.263773 Objective Loss 0.263773 LR 0.001000 Time 0.019983 -2022-12-06 10:54:15,592 - Epoch: [71][ 840/ 1200] Overall Loss 0.263911 Objective Loss 0.263911 LR 0.001000 Time 0.019971 -2022-12-06 10:54:15,783 - Epoch: [71][ 850/ 1200] Overall Loss 0.264221 Objective Loss 0.264221 LR 0.001000 Time 0.019960 -2022-12-06 10:54:15,974 - Epoch: [71][ 860/ 1200] Overall Loss 0.264391 Objective Loss 0.264391 LR 0.001000 Time 0.019950 -2022-12-06 10:54:16,166 - Epoch: [71][ 870/ 1200] Overall Loss 0.264447 Objective Loss 0.264447 LR 0.001000 Time 0.019940 -2022-12-06 10:54:16,356 - Epoch: [71][ 880/ 1200] Overall Loss 0.264527 Objective Loss 0.264527 LR 0.001000 Time 0.019929 -2022-12-06 10:54:16,548 - Epoch: [71][ 890/ 1200] Overall Loss 0.264872 Objective Loss 0.264872 LR 0.001000 Time 0.019920 -2022-12-06 10:54:16,738 - Epoch: [71][ 900/ 1200] Overall Loss 0.264808 Objective Loss 0.264808 LR 0.001000 Time 0.019910 -2022-12-06 10:54:16,929 - Epoch: [71][ 910/ 1200] Overall Loss 0.265034 Objective Loss 0.265034 LR 0.001000 Time 0.019900 -2022-12-06 10:54:17,119 - Epoch: [71][ 920/ 1200] Overall Loss 0.265055 Objective Loss 0.265055 LR 0.001000 Time 0.019890 -2022-12-06 10:54:17,310 - Epoch: [71][ 930/ 1200] Overall Loss 0.264784 Objective Loss 0.264784 LR 0.001000 Time 0.019881 -2022-12-06 10:54:17,501 - Epoch: [71][ 940/ 1200] Overall Loss 0.265114 Objective Loss 0.265114 LR 0.001000 Time 0.019872 -2022-12-06 10:54:17,692 - Epoch: [71][ 950/ 1200] Overall Loss 0.265291 Objective Loss 0.265291 LR 0.001000 Time 0.019863 -2022-12-06 10:54:17,883 - Epoch: [71][ 960/ 1200] Overall Loss 0.265185 Objective Loss 0.265185 LR 0.001000 Time 0.019854 -2022-12-06 10:54:18,074 - Epoch: [71][ 970/ 1200] Overall Loss 0.265284 Objective Loss 0.265284 LR 0.001000 Time 0.019846 -2022-12-06 10:54:18,264 - Epoch: [71][ 980/ 1200] Overall Loss 0.265134 Objective Loss 0.265134 LR 0.001000 Time 0.019837 -2022-12-06 10:54:18,455 - Epoch: [71][ 990/ 1200] Overall Loss 0.265282 Objective Loss 0.265282 LR 0.001000 Time 0.019829 -2022-12-06 10:54:18,646 - Epoch: [71][ 1000/ 1200] Overall Loss 0.265282 Objective Loss 0.265282 LR 0.001000 Time 0.019821 -2022-12-06 10:54:18,838 - Epoch: [71][ 1010/ 1200] Overall Loss 0.265390 Objective Loss 0.265390 LR 0.001000 Time 0.019815 -2022-12-06 10:54:19,029 - Epoch: [71][ 1020/ 1200] Overall Loss 0.265444 Objective Loss 0.265444 LR 0.001000 Time 0.019807 -2022-12-06 10:54:19,220 - Epoch: [71][ 1030/ 1200] Overall Loss 0.265560 Objective Loss 0.265560 LR 0.001000 Time 0.019799 -2022-12-06 10:54:19,410 - Epoch: [71][ 1040/ 1200] Overall Loss 0.265668 Objective Loss 0.265668 LR 0.001000 Time 0.019792 -2022-12-06 10:54:19,601 - Epoch: [71][ 1050/ 1200] Overall Loss 0.265559 Objective Loss 0.265559 LR 0.001000 Time 0.019785 -2022-12-06 10:54:19,792 - Epoch: [71][ 1060/ 1200] Overall Loss 0.265577 Objective Loss 0.265577 LR 0.001000 Time 0.019777 -2022-12-06 10:54:19,983 - Epoch: [71][ 1070/ 1200] Overall Loss 0.265579 Objective Loss 0.265579 LR 0.001000 Time 0.019770 -2022-12-06 10:54:20,173 - Epoch: [71][ 1080/ 1200] Overall Loss 0.265786 Objective Loss 0.265786 LR 0.001000 Time 0.019763 -2022-12-06 10:54:20,364 - Epoch: [71][ 1090/ 1200] Overall Loss 0.265790 Objective Loss 0.265790 LR 0.001000 Time 0.019756 -2022-12-06 10:54:20,554 - Epoch: [71][ 1100/ 1200] Overall Loss 0.265746 Objective Loss 0.265746 LR 0.001000 Time 0.019749 -2022-12-06 10:54:20,746 - Epoch: [71][ 1110/ 1200] Overall Loss 0.265816 Objective Loss 0.265816 LR 0.001000 Time 0.019743 -2022-12-06 10:54:20,936 - Epoch: [71][ 1120/ 1200] Overall Loss 0.265639 Objective Loss 0.265639 LR 0.001000 Time 0.019737 -2022-12-06 10:54:21,127 - Epoch: [71][ 1130/ 1200] Overall Loss 0.265741 Objective Loss 0.265741 LR 0.001000 Time 0.019730 -2022-12-06 10:54:21,318 - Epoch: [71][ 1140/ 1200] Overall Loss 0.265754 Objective Loss 0.265754 LR 0.001000 Time 0.019724 -2022-12-06 10:54:21,509 - Epoch: [71][ 1150/ 1200] Overall Loss 0.265680 Objective Loss 0.265680 LR 0.001000 Time 0.019718 -2022-12-06 10:54:21,701 - Epoch: [71][ 1160/ 1200] Overall Loss 0.265496 Objective Loss 0.265496 LR 0.001000 Time 0.019713 -2022-12-06 10:54:21,891 - Epoch: [71][ 1170/ 1200] Overall Loss 0.265450 Objective Loss 0.265450 LR 0.001000 Time 0.019707 -2022-12-06 10:54:22,082 - Epoch: [71][ 1180/ 1200] Overall Loss 0.265480 Objective Loss 0.265480 LR 0.001000 Time 0.019701 -2022-12-06 10:54:22,273 - Epoch: [71][ 1190/ 1200] Overall Loss 0.265749 Objective Loss 0.265749 LR 0.001000 Time 0.019696 -2022-12-06 10:54:22,504 - Epoch: [71][ 1200/ 1200] Overall Loss 0.265844 Objective Loss 0.265844 Top1 87.866109 Top5 98.744770 LR 0.001000 Time 0.019723 -2022-12-06 10:54:22,593 - --- validate (epoch=71)----------- -2022-12-06 10:54:22,593 - 34129 samples (256 per mini-batch) -2022-12-06 10:54:23,042 - Epoch: [71][ 10/ 134] Loss 0.268034 Top1 86.328125 Top5 98.046875 -2022-12-06 10:54:23,184 - Epoch: [71][ 20/ 134] Loss 0.276601 Top1 85.957031 Top5 98.183594 -2022-12-06 10:54:23,327 - Epoch: [71][ 30/ 134] Loss 0.292506 Top1 85.585938 Top5 98.007812 -2022-12-06 10:54:23,474 - Epoch: [71][ 40/ 134] Loss 0.291599 Top1 85.244141 Top5 98.076172 -2022-12-06 10:54:23,616 - Epoch: [71][ 50/ 134] Loss 0.284769 Top1 85.437500 Top5 98.109375 -2022-12-06 10:54:23,762 - Epoch: [71][ 60/ 134] Loss 0.290783 Top1 85.436198 Top5 98.138021 -2022-12-06 10:54:23,907 - Epoch: [71][ 70/ 134] Loss 0.288848 Top1 85.535714 Top5 98.231027 -2022-12-06 10:54:24,052 - Epoch: [71][ 80/ 134] Loss 0.289626 Top1 85.590820 Top5 98.227539 -2022-12-06 10:54:24,197 - Epoch: [71][ 90/ 134] Loss 0.294909 Top1 85.468750 Top5 98.142361 -2022-12-06 10:54:24,344 - Epoch: [71][ 100/ 134] Loss 0.295309 Top1 85.445312 Top5 98.121094 -2022-12-06 10:54:24,489 - Epoch: [71][ 110/ 134] Loss 0.295668 Top1 85.387074 Top5 98.135653 -2022-12-06 10:54:24,638 - Epoch: [71][ 120/ 134] Loss 0.295811 Top1 85.449219 Top5 98.154297 -2022-12-06 10:54:24,777 - Epoch: [71][ 130/ 134] Loss 0.295667 Top1 85.438702 Top5 98.146034 -2022-12-06 10:54:24,815 - Epoch: [71][ 134/ 134] Loss 0.297722 Top1 85.402444 Top5 98.139412 -2022-12-06 10:54:24,903 - ==> Top1: 85.402 Top5: 98.139 Loss: 0.298 - -2022-12-06 10:54:24,904 - ==> Confusion: -[[ 892 2 2 7 11 7 0 1 12 41 0 1 0 1 4 6 1 0 0 3 5] - [ 0 929 3 2 8 26 4 11 1 0 1 5 2 3 3 0 7 4 8 3 7] - [ 7 6 984 13 1 4 25 12 0 1 3 7 2 0 7 3 1 2 3 6 16] - [ 2 4 11 933 0 2 3 0 0 1 12 2 6 0 23 0 0 4 12 0 5] - [ 13 11 2 0 934 5 0 3 2 6 0 3 0 1 9 6 14 1 1 1 8] - [ 2 16 0 4 6 957 2 22 2 1 0 11 5 13 4 1 1 2 3 9 8] - [ 0 5 6 6 0 4 1065 3 0 0 0 2 4 0 0 8 1 2 1 8 3] - [ 0 8 7 3 1 31 9 925 0 0 4 5 2 0 0 1 2 2 29 17 8] - [ 8 3 0 1 0 2 1 1 968 34 5 2 2 8 13 0 5 3 4 1 3] - [ 94 1 3 0 4 0 1 3 36 822 2 2 0 14 11 0 0 2 0 0 6] - [ 0 2 6 9 1 1 5 4 12 0 941 1 4 8 8 0 2 0 9 2 4] - [ 2 0 1 1 1 7 2 4 0 0 0 955 23 9 0 5 6 12 2 9 12] - [ 1 1 3 6 0 1 0 0 0 0 0 38 870 0 0 12 2 21 0 4 10] - [ 0 0 1 0 2 10 0 1 14 14 7 4 6 929 4 4 5 4 1 6 11] - [ 9 3 2 13 4 1 0 0 16 1 0 3 2 2 1058 1 2 0 8 1 4] - [ 1 0 3 0 2 2 1 0 0 0 0 6 2 1 0 994 10 14 1 2 4] - [ 3 4 2 1 1 0 0 0 2 1 0 2 4 0 0 12 1024 2 0 8 6] - [ 4 1 0 4 0 0 0 1 0 1 0 4 12 4 0 19 0 982 1 2 1] - [ 2 4 3 8 0 4 1 31 4 0 4 2 2 1 3 3 1 1 926 5 3] - [ 3 3 2 2 0 6 10 6 1 0 2 17 5 3 0 2 4 4 0 1005 5] - [ 124 237 152 119 103 180 75 147 102 69 146 92 323 253 175 108 224 117 181 250 10049]] - -2022-12-06 10:54:25,472 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:54:25,473 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:54:25,478 - - -2022-12-06 10:54:25,479 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:54:26,520 - Epoch: [72][ 10/ 1200] Overall Loss 0.302389 Objective Loss 0.302389 LR 0.001000 Time 0.104029 -2022-12-06 10:54:26,713 - Epoch: [72][ 20/ 1200] Overall Loss 0.284897 Objective Loss 0.284897 LR 0.001000 Time 0.061656 -2022-12-06 10:54:26,904 - Epoch: [72][ 30/ 1200] Overall Loss 0.274889 Objective Loss 0.274889 LR 0.001000 Time 0.047461 -2022-12-06 10:54:27,095 - Epoch: [72][ 40/ 1200] Overall Loss 0.270345 Objective Loss 0.270345 LR 0.001000 Time 0.040344 -2022-12-06 10:54:27,285 - Epoch: [72][ 50/ 1200] Overall Loss 0.270379 Objective Loss 0.270379 LR 0.001000 Time 0.036063 -2022-12-06 10:54:27,475 - Epoch: [72][ 60/ 1200] Overall Loss 0.270707 Objective Loss 0.270707 LR 0.001000 Time 0.033219 -2022-12-06 10:54:27,666 - Epoch: [72][ 70/ 1200] Overall Loss 0.265588 Objective Loss 0.265588 LR 0.001000 Time 0.031193 -2022-12-06 10:54:27,857 - Epoch: [72][ 80/ 1200] Overall Loss 0.266337 Objective Loss 0.266337 LR 0.001000 Time 0.029670 -2022-12-06 10:54:28,047 - Epoch: [72][ 90/ 1200] Overall Loss 0.264546 Objective Loss 0.264546 LR 0.001000 Time 0.028484 -2022-12-06 10:54:28,238 - Epoch: [72][ 100/ 1200] Overall Loss 0.265674 Objective Loss 0.265674 LR 0.001000 Time 0.027535 -2022-12-06 10:54:28,428 - Epoch: [72][ 110/ 1200] Overall Loss 0.265702 Objective Loss 0.265702 LR 0.001000 Time 0.026759 -2022-12-06 10:54:28,618 - Epoch: [72][ 120/ 1200] Overall Loss 0.265572 Objective Loss 0.265572 LR 0.001000 Time 0.026106 -2022-12-06 10:54:28,809 - Epoch: [72][ 130/ 1200] Overall Loss 0.265106 Objective Loss 0.265106 LR 0.001000 Time 0.025561 -2022-12-06 10:54:28,999 - Epoch: [72][ 140/ 1200] Overall Loss 0.263021 Objective Loss 0.263021 LR 0.001000 Time 0.025092 -2022-12-06 10:54:29,190 - Epoch: [72][ 150/ 1200] Overall Loss 0.262978 Objective Loss 0.262978 LR 0.001000 Time 0.024688 -2022-12-06 10:54:29,381 - Epoch: [72][ 160/ 1200] Overall Loss 0.260793 Objective Loss 0.260793 LR 0.001000 Time 0.024335 -2022-12-06 10:54:29,572 - Epoch: [72][ 170/ 1200] Overall Loss 0.260334 Objective Loss 0.260334 LR 0.001000 Time 0.024021 -2022-12-06 10:54:29,762 - Epoch: [72][ 180/ 1200] Overall Loss 0.260968 Objective Loss 0.260968 LR 0.001000 Time 0.023742 -2022-12-06 10:54:29,953 - Epoch: [72][ 190/ 1200] Overall Loss 0.260003 Objective Loss 0.260003 LR 0.001000 Time 0.023493 -2022-12-06 10:54:30,143 - Epoch: [72][ 200/ 1200] Overall Loss 0.259400 Objective Loss 0.259400 LR 0.001000 Time 0.023269 -2022-12-06 10:54:30,334 - Epoch: [72][ 210/ 1200] Overall Loss 0.259494 Objective Loss 0.259494 LR 0.001000 Time 0.023067 -2022-12-06 10:54:30,525 - Epoch: [72][ 220/ 1200] Overall Loss 0.261234 Objective Loss 0.261234 LR 0.001000 Time 0.022883 -2022-12-06 10:54:30,716 - Epoch: [72][ 230/ 1200] Overall Loss 0.262584 Objective Loss 0.262584 LR 0.001000 Time 0.022714 -2022-12-06 10:54:30,906 - Epoch: [72][ 240/ 1200] Overall Loss 0.262895 Objective Loss 0.262895 LR 0.001000 Time 0.022560 -2022-12-06 10:54:31,097 - Epoch: [72][ 250/ 1200] Overall Loss 0.262881 Objective Loss 0.262881 LR 0.001000 Time 0.022418 -2022-12-06 10:54:31,287 - Epoch: [72][ 260/ 1200] Overall Loss 0.262979 Objective Loss 0.262979 LR 0.001000 Time 0.022287 -2022-12-06 10:54:31,478 - Epoch: [72][ 270/ 1200] Overall Loss 0.263494 Objective Loss 0.263494 LR 0.001000 Time 0.022165 -2022-12-06 10:54:31,668 - Epoch: [72][ 280/ 1200] Overall Loss 0.263448 Objective Loss 0.263448 LR 0.001000 Time 0.022050 -2022-12-06 10:54:31,859 - Epoch: [72][ 290/ 1200] Overall Loss 0.263496 Objective Loss 0.263496 LR 0.001000 Time 0.021945 -2022-12-06 10:54:32,050 - Epoch: [72][ 300/ 1200] Overall Loss 0.263519 Objective Loss 0.263519 LR 0.001000 Time 0.021848 -2022-12-06 10:54:32,240 - Epoch: [72][ 310/ 1200] Overall Loss 0.263926 Objective Loss 0.263926 LR 0.001000 Time 0.021755 -2022-12-06 10:54:32,430 - Epoch: [72][ 320/ 1200] Overall Loss 0.264001 Objective Loss 0.264001 LR 0.001000 Time 0.021669 -2022-12-06 10:54:32,621 - Epoch: [72][ 330/ 1200] Overall Loss 0.264201 Objective Loss 0.264201 LR 0.001000 Time 0.021590 -2022-12-06 10:54:32,812 - Epoch: [72][ 340/ 1200] Overall Loss 0.264511 Objective Loss 0.264511 LR 0.001000 Time 0.021514 -2022-12-06 10:54:33,003 - Epoch: [72][ 350/ 1200] Overall Loss 0.264802 Objective Loss 0.264802 LR 0.001000 Time 0.021443 -2022-12-06 10:54:33,194 - Epoch: [72][ 360/ 1200] Overall Loss 0.264011 Objective Loss 0.264011 LR 0.001000 Time 0.021376 -2022-12-06 10:54:33,384 - Epoch: [72][ 370/ 1200] Overall Loss 0.263692 Objective Loss 0.263692 LR 0.001000 Time 0.021311 -2022-12-06 10:54:33,574 - Epoch: [72][ 380/ 1200] Overall Loss 0.263556 Objective Loss 0.263556 LR 0.001000 Time 0.021250 -2022-12-06 10:54:33,765 - Epoch: [72][ 390/ 1200] Overall Loss 0.263510 Objective Loss 0.263510 LR 0.001000 Time 0.021192 -2022-12-06 10:54:33,956 - Epoch: [72][ 400/ 1200] Overall Loss 0.264773 Objective Loss 0.264773 LR 0.001000 Time 0.021138 -2022-12-06 10:54:34,146 - Epoch: [72][ 410/ 1200] Overall Loss 0.266020 Objective Loss 0.266020 LR 0.001000 Time 0.021086 -2022-12-06 10:54:34,337 - Epoch: [72][ 420/ 1200] Overall Loss 0.266857 Objective Loss 0.266857 LR 0.001000 Time 0.021037 -2022-12-06 10:54:34,528 - Epoch: [72][ 430/ 1200] Overall Loss 0.266974 Objective Loss 0.266974 LR 0.001000 Time 0.020990 -2022-12-06 10:54:34,719 - Epoch: [72][ 440/ 1200] Overall Loss 0.266912 Objective Loss 0.266912 LR 0.001000 Time 0.020946 -2022-12-06 10:54:34,910 - Epoch: [72][ 450/ 1200] Overall Loss 0.267152 Objective Loss 0.267152 LR 0.001000 Time 0.020903 -2022-12-06 10:54:35,100 - Epoch: [72][ 460/ 1200] Overall Loss 0.267823 Objective Loss 0.267823 LR 0.001000 Time 0.020862 -2022-12-06 10:54:35,291 - Epoch: [72][ 470/ 1200] Overall Loss 0.267660 Objective Loss 0.267660 LR 0.001000 Time 0.020822 -2022-12-06 10:54:35,481 - Epoch: [72][ 480/ 1200] Overall Loss 0.267908 Objective Loss 0.267908 LR 0.001000 Time 0.020784 -2022-12-06 10:54:35,672 - Epoch: [72][ 490/ 1200] Overall Loss 0.268135 Objective Loss 0.268135 LR 0.001000 Time 0.020749 -2022-12-06 10:54:35,863 - Epoch: [72][ 500/ 1200] Overall Loss 0.268407 Objective Loss 0.268407 LR 0.001000 Time 0.020715 -2022-12-06 10:54:36,054 - Epoch: [72][ 510/ 1200] Overall Loss 0.268792 Objective Loss 0.268792 LR 0.001000 Time 0.020681 -2022-12-06 10:54:36,244 - Epoch: [72][ 520/ 1200] Overall Loss 0.268514 Objective Loss 0.268514 LR 0.001000 Time 0.020649 -2022-12-06 10:54:36,435 - Epoch: [72][ 530/ 1200] Overall Loss 0.268737 Objective Loss 0.268737 LR 0.001000 Time 0.020617 -2022-12-06 10:54:36,625 - Epoch: [72][ 540/ 1200] Overall Loss 0.268189 Objective Loss 0.268189 LR 0.001000 Time 0.020587 -2022-12-06 10:54:36,816 - Epoch: [72][ 550/ 1200] Overall Loss 0.267244 Objective Loss 0.267244 LR 0.001000 Time 0.020559 -2022-12-06 10:54:37,006 - Epoch: [72][ 560/ 1200] Overall Loss 0.267310 Objective Loss 0.267310 LR 0.001000 Time 0.020531 -2022-12-06 10:54:37,197 - Epoch: [72][ 570/ 1200] Overall Loss 0.266987 Objective Loss 0.266987 LR 0.001000 Time 0.020504 -2022-12-06 10:54:37,388 - Epoch: [72][ 580/ 1200] Overall Loss 0.266888 Objective Loss 0.266888 LR 0.001000 Time 0.020479 -2022-12-06 10:54:37,578 - Epoch: [72][ 590/ 1200] Overall Loss 0.266694 Objective Loss 0.266694 LR 0.001000 Time 0.020454 -2022-12-06 10:54:37,769 - Epoch: [72][ 600/ 1200] Overall Loss 0.266466 Objective Loss 0.266466 LR 0.001000 Time 0.020430 -2022-12-06 10:54:37,960 - Epoch: [72][ 610/ 1200] Overall Loss 0.266019 Objective Loss 0.266019 LR 0.001000 Time 0.020407 -2022-12-06 10:54:38,150 - Epoch: [72][ 620/ 1200] Overall Loss 0.266198 Objective Loss 0.266198 LR 0.001000 Time 0.020384 -2022-12-06 10:54:38,342 - Epoch: [72][ 630/ 1200] Overall Loss 0.266475 Objective Loss 0.266475 LR 0.001000 Time 0.020363 -2022-12-06 10:54:38,532 - Epoch: [72][ 640/ 1200] Overall Loss 0.266997 Objective Loss 0.266997 LR 0.001000 Time 0.020341 -2022-12-06 10:54:38,722 - Epoch: [72][ 650/ 1200] Overall Loss 0.267185 Objective Loss 0.267185 LR 0.001000 Time 0.020321 -2022-12-06 10:54:38,913 - Epoch: [72][ 660/ 1200] Overall Loss 0.266921 Objective Loss 0.266921 LR 0.001000 Time 0.020301 -2022-12-06 10:54:39,104 - Epoch: [72][ 670/ 1200] Overall Loss 0.267449 Objective Loss 0.267449 LR 0.001000 Time 0.020282 -2022-12-06 10:54:39,294 - Epoch: [72][ 680/ 1200] Overall Loss 0.267623 Objective Loss 0.267623 LR 0.001000 Time 0.020262 -2022-12-06 10:54:39,484 - Epoch: [72][ 690/ 1200] Overall Loss 0.267502 Objective Loss 0.267502 LR 0.001000 Time 0.020244 -2022-12-06 10:54:39,675 - Epoch: [72][ 700/ 1200] Overall Loss 0.267668 Objective Loss 0.267668 LR 0.001000 Time 0.020226 -2022-12-06 10:54:39,866 - Epoch: [72][ 710/ 1200] Overall Loss 0.267414 Objective Loss 0.267414 LR 0.001000 Time 0.020209 -2022-12-06 10:54:40,057 - Epoch: [72][ 720/ 1200] Overall Loss 0.267035 Objective Loss 0.267035 LR 0.001000 Time 0.020193 -2022-12-06 10:54:40,248 - Epoch: [72][ 730/ 1200] Overall Loss 0.267245 Objective Loss 0.267245 LR 0.001000 Time 0.020177 -2022-12-06 10:54:40,438 - Epoch: [72][ 740/ 1200] Overall Loss 0.267104 Objective Loss 0.267104 LR 0.001000 Time 0.020161 -2022-12-06 10:54:40,629 - Epoch: [72][ 750/ 1200] Overall Loss 0.267185 Objective Loss 0.267185 LR 0.001000 Time 0.020146 -2022-12-06 10:54:40,820 - Epoch: [72][ 760/ 1200] Overall Loss 0.267351 Objective Loss 0.267351 LR 0.001000 Time 0.020131 -2022-12-06 10:54:41,011 - Epoch: [72][ 770/ 1200] Overall Loss 0.267810 Objective Loss 0.267810 LR 0.001000 Time 0.020117 -2022-12-06 10:54:41,202 - Epoch: [72][ 780/ 1200] Overall Loss 0.267910 Objective Loss 0.267910 LR 0.001000 Time 0.020104 -2022-12-06 10:54:41,392 - Epoch: [72][ 790/ 1200] Overall Loss 0.267750 Objective Loss 0.267750 LR 0.001000 Time 0.020090 -2022-12-06 10:54:41,583 - Epoch: [72][ 800/ 1200] Overall Loss 0.267607 Objective Loss 0.267607 LR 0.001000 Time 0.020076 -2022-12-06 10:54:41,774 - Epoch: [72][ 810/ 1200] Overall Loss 0.267736 Objective Loss 0.267736 LR 0.001000 Time 0.020064 -2022-12-06 10:54:41,964 - Epoch: [72][ 820/ 1200] Overall Loss 0.267848 Objective Loss 0.267848 LR 0.001000 Time 0.020050 -2022-12-06 10:54:42,155 - Epoch: [72][ 830/ 1200] Overall Loss 0.268069 Objective Loss 0.268069 LR 0.001000 Time 0.020038 -2022-12-06 10:54:42,346 - Epoch: [72][ 840/ 1200] Overall Loss 0.268038 Objective Loss 0.268038 LR 0.001000 Time 0.020026 -2022-12-06 10:54:42,537 - Epoch: [72][ 850/ 1200] Overall Loss 0.268151 Objective Loss 0.268151 LR 0.001000 Time 0.020015 -2022-12-06 10:54:42,728 - Epoch: [72][ 860/ 1200] Overall Loss 0.268484 Objective Loss 0.268484 LR 0.001000 Time 0.020003 -2022-12-06 10:54:42,918 - Epoch: [72][ 870/ 1200] Overall Loss 0.268352 Objective Loss 0.268352 LR 0.001000 Time 0.019991 -2022-12-06 10:54:43,109 - Epoch: [72][ 880/ 1200] Overall Loss 0.268290 Objective Loss 0.268290 LR 0.001000 Time 0.019980 -2022-12-06 10:54:43,300 - Epoch: [72][ 890/ 1200] Overall Loss 0.268109 Objective Loss 0.268109 LR 0.001000 Time 0.019969 -2022-12-06 10:54:43,491 - Epoch: [72][ 900/ 1200] Overall Loss 0.267913 Objective Loss 0.267913 LR 0.001000 Time 0.019959 -2022-12-06 10:54:43,682 - Epoch: [72][ 910/ 1200] Overall Loss 0.268061 Objective Loss 0.268061 LR 0.001000 Time 0.019949 -2022-12-06 10:54:43,872 - Epoch: [72][ 920/ 1200] Overall Loss 0.267996 Objective Loss 0.267996 LR 0.001000 Time 0.019939 -2022-12-06 10:54:44,063 - Epoch: [72][ 930/ 1200] Overall Loss 0.267885 Objective Loss 0.267885 LR 0.001000 Time 0.019929 -2022-12-06 10:54:44,254 - Epoch: [72][ 940/ 1200] Overall Loss 0.268365 Objective Loss 0.268365 LR 0.001000 Time 0.019920 -2022-12-06 10:54:44,446 - Epoch: [72][ 950/ 1200] Overall Loss 0.268429 Objective Loss 0.268429 LR 0.001000 Time 0.019911 -2022-12-06 10:54:44,637 - Epoch: [72][ 960/ 1200] Overall Loss 0.268346 Objective Loss 0.268346 LR 0.001000 Time 0.019902 -2022-12-06 10:54:44,828 - Epoch: [72][ 970/ 1200] Overall Loss 0.268331 Objective Loss 0.268331 LR 0.001000 Time 0.019893 -2022-12-06 10:54:45,018 - Epoch: [72][ 980/ 1200] Overall Loss 0.268032 Objective Loss 0.268032 LR 0.001000 Time 0.019884 -2022-12-06 10:54:45,208 - Epoch: [72][ 990/ 1200] Overall Loss 0.268239 Objective Loss 0.268239 LR 0.001000 Time 0.019875 -2022-12-06 10:54:45,399 - Epoch: [72][ 1000/ 1200] Overall Loss 0.268225 Objective Loss 0.268225 LR 0.001000 Time 0.019866 -2022-12-06 10:54:45,590 - Epoch: [72][ 1010/ 1200] Overall Loss 0.268022 Objective Loss 0.268022 LR 0.001000 Time 0.019857 -2022-12-06 10:54:45,781 - Epoch: [72][ 1020/ 1200] Overall Loss 0.267905 Objective Loss 0.267905 LR 0.001000 Time 0.019849 -2022-12-06 10:54:45,971 - Epoch: [72][ 1030/ 1200] Overall Loss 0.267966 Objective Loss 0.267966 LR 0.001000 Time 0.019841 -2022-12-06 10:54:46,162 - Epoch: [72][ 1040/ 1200] Overall Loss 0.267922 Objective Loss 0.267922 LR 0.001000 Time 0.019833 -2022-12-06 10:54:46,352 - Epoch: [72][ 1050/ 1200] Overall Loss 0.268083 Objective Loss 0.268083 LR 0.001000 Time 0.019825 -2022-12-06 10:54:46,542 - Epoch: [72][ 1060/ 1200] Overall Loss 0.268121 Objective Loss 0.268121 LR 0.001000 Time 0.019817 -2022-12-06 10:54:46,733 - Epoch: [72][ 1070/ 1200] Overall Loss 0.267979 Objective Loss 0.267979 LR 0.001000 Time 0.019810 -2022-12-06 10:54:46,924 - Epoch: [72][ 1080/ 1200] Overall Loss 0.268013 Objective Loss 0.268013 LR 0.001000 Time 0.019802 -2022-12-06 10:54:47,114 - Epoch: [72][ 1090/ 1200] Overall Loss 0.267957 Objective Loss 0.267957 LR 0.001000 Time 0.019795 -2022-12-06 10:54:47,305 - Epoch: [72][ 1100/ 1200] Overall Loss 0.267828 Objective Loss 0.267828 LR 0.001000 Time 0.019788 -2022-12-06 10:54:47,495 - Epoch: [72][ 1110/ 1200] Overall Loss 0.267907 Objective Loss 0.267907 LR 0.001000 Time 0.019781 -2022-12-06 10:54:47,686 - Epoch: [72][ 1120/ 1200] Overall Loss 0.268237 Objective Loss 0.268237 LR 0.001000 Time 0.019774 -2022-12-06 10:54:47,877 - Epoch: [72][ 1130/ 1200] Overall Loss 0.267938 Objective Loss 0.267938 LR 0.001000 Time 0.019767 -2022-12-06 10:54:48,068 - Epoch: [72][ 1140/ 1200] Overall Loss 0.267914 Objective Loss 0.267914 LR 0.001000 Time 0.019761 -2022-12-06 10:54:48,258 - Epoch: [72][ 1150/ 1200] Overall Loss 0.267636 Objective Loss 0.267636 LR 0.001000 Time 0.019754 -2022-12-06 10:54:48,449 - Epoch: [72][ 1160/ 1200] Overall Loss 0.267570 Objective Loss 0.267570 LR 0.001000 Time 0.019748 -2022-12-06 10:54:48,640 - Epoch: [72][ 1170/ 1200] Overall Loss 0.267647 Objective Loss 0.267647 LR 0.001000 Time 0.019741 -2022-12-06 10:54:48,830 - Epoch: [72][ 1180/ 1200] Overall Loss 0.267768 Objective Loss 0.267768 LR 0.001000 Time 0.019735 -2022-12-06 10:54:49,020 - Epoch: [72][ 1190/ 1200] Overall Loss 0.267905 Objective Loss 0.267905 LR 0.001000 Time 0.019729 -2022-12-06 10:54:49,251 - Epoch: [72][ 1200/ 1200] Overall Loss 0.268073 Objective Loss 0.268073 Top1 84.309623 Top5 98.535565 LR 0.001000 Time 0.019756 -2022-12-06 10:54:49,340 - --- validate (epoch=72)----------- -2022-12-06 10:54:49,340 - 34129 samples (256 per mini-batch) -2022-12-06 10:54:49,787 - Epoch: [72][ 10/ 134] Loss 0.283937 Top1 84.257812 Top5 97.929688 -2022-12-06 10:54:49,920 - Epoch: [72][ 20/ 134] Loss 0.304654 Top1 84.218750 Top5 98.085938 -2022-12-06 10:54:50,052 - Epoch: [72][ 30/ 134] Loss 0.296251 Top1 84.570312 Top5 97.890625 -2022-12-06 10:54:50,182 - Epoch: [72][ 40/ 134] Loss 0.291699 Top1 84.404297 Top5 98.066406 -2022-12-06 10:54:50,316 - Epoch: [72][ 50/ 134] Loss 0.300304 Top1 84.273438 Top5 97.992188 -2022-12-06 10:54:50,446 - Epoch: [72][ 60/ 134] Loss 0.307502 Top1 83.951823 Top5 97.955729 -2022-12-06 10:54:50,580 - Epoch: [72][ 70/ 134] Loss 0.305308 Top1 84.095982 Top5 97.946429 -2022-12-06 10:54:50,710 - Epoch: [72][ 80/ 134] Loss 0.304762 Top1 84.091797 Top5 97.939453 -2022-12-06 10:54:50,840 - Epoch: [72][ 90/ 134] Loss 0.305318 Top1 83.958333 Top5 97.899306 -2022-12-06 10:54:50,967 - Epoch: [72][ 100/ 134] Loss 0.303932 Top1 83.976562 Top5 97.902344 -2022-12-06 10:54:51,100 - Epoch: [72][ 110/ 134] Loss 0.304565 Top1 84.019886 Top5 97.929688 -2022-12-06 10:54:51,228 - Epoch: [72][ 120/ 134] Loss 0.303307 Top1 84.143880 Top5 97.923177 -2022-12-06 10:54:51,357 - Epoch: [72][ 130/ 134] Loss 0.301604 Top1 84.110577 Top5 97.875601 -2022-12-06 10:54:51,395 - Epoch: [72][ 134/ 134] Loss 0.301929 Top1 84.089777 Top5 97.890357 -2022-12-06 10:54:51,485 - ==> Top1: 84.090 Top5: 97.890 Loss: 0.302 - -2022-12-06 10:54:51,486 - ==> Confusion: -[[ 913 0 0 5 4 5 1 0 2 45 0 1 1 4 4 2 2 0 1 1 5] - [ 2 893 3 4 12 29 0 22 0 2 0 7 2 4 7 1 8 2 20 3 6] - [ 9 2 993 14 5 3 24 14 1 3 4 5 1 1 3 3 1 1 6 1 9] - [ 3 0 21 922 1 2 1 1 1 0 8 1 7 5 15 0 2 3 23 0 4] - [ 17 4 2 0 945 6 0 0 1 8 0 3 2 2 9 4 9 5 0 1 2] - [ 2 10 0 2 7 955 3 26 4 3 3 14 3 17 1 1 1 0 2 4 11] - [ 0 2 13 4 0 2 1064 8 0 1 2 2 0 2 0 6 0 0 3 6 3] - [ 1 4 7 1 1 35 3 945 0 0 2 8 3 2 0 0 1 1 30 7 3] - [ 7 2 1 1 0 2 0 0 968 43 3 3 1 11 13 0 4 2 2 1 0] - [ 96 0 0 1 2 1 0 4 25 846 1 3 0 13 2 0 1 1 1 1 3] - [ 0 3 7 14 2 3 1 3 8 0 930 1 2 19 6 1 1 1 12 0 5] - [ 3 1 2 0 0 7 3 4 0 0 1 965 22 6 1 6 8 7 2 11 2] - [ 0 0 4 2 2 4 1 2 1 0 0 39 868 5 1 10 2 16 1 3 8] - [ 1 0 0 0 1 6 0 4 15 11 7 5 5 955 1 1 4 1 0 0 6] - [ 11 1 0 17 6 3 0 1 25 4 1 3 3 3 1038 0 2 2 8 0 2] - [ 1 0 2 1 3 1 4 1 0 0 0 9 3 3 0 981 14 16 0 2 2] - [ 1 2 4 1 2 0 2 2 0 1 0 2 1 4 4 9 1024 3 2 5 3] - [ 0 0 2 3 0 1 1 1 0 1 0 16 18 5 2 15 1 966 1 1 2] - [ 4 5 3 9 1 1 0 27 2 1 1 2 3 2 9 1 1 0 935 1 0] - [ 1 6 3 0 2 3 8 12 1 1 1 14 7 8 1 3 6 9 5 984 5] - [ 153 188 177 112 152 215 72 177 115 86 128 138 376 350 193 140 258 109 265 218 9604]] - -2022-12-06 10:54:52,058 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:54:52,058 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:54:52,064 - - -2022-12-06 10:54:52,064 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:54:52,985 - Epoch: [73][ 10/ 1200] Overall Loss 0.268960 Objective Loss 0.268960 LR 0.001000 Time 0.091955 -2022-12-06 10:54:53,185 - Epoch: [73][ 20/ 1200] Overall Loss 0.261951 Objective Loss 0.261951 LR 0.001000 Time 0.055984 -2022-12-06 10:54:53,383 - Epoch: [73][ 30/ 1200] Overall Loss 0.261655 Objective Loss 0.261655 LR 0.001000 Time 0.043881 -2022-12-06 10:54:53,582 - Epoch: [73][ 40/ 1200] Overall Loss 0.259668 Objective Loss 0.259668 LR 0.001000 Time 0.037869 -2022-12-06 10:54:53,778 - Epoch: [73][ 50/ 1200] Overall Loss 0.262644 Objective Loss 0.262644 LR 0.001000 Time 0.034219 -2022-12-06 10:54:53,977 - Epoch: [73][ 60/ 1200] Overall Loss 0.261051 Objective Loss 0.261051 LR 0.001000 Time 0.031824 -2022-12-06 10:54:54,174 - Epoch: [73][ 70/ 1200] Overall Loss 0.259994 Objective Loss 0.259994 LR 0.001000 Time 0.030077 -2022-12-06 10:54:54,373 - Epoch: [73][ 80/ 1200] Overall Loss 0.258223 Objective Loss 0.258223 LR 0.001000 Time 0.028802 -2022-12-06 10:54:54,570 - Epoch: [73][ 90/ 1200] Overall Loss 0.257304 Objective Loss 0.257304 LR 0.001000 Time 0.027781 -2022-12-06 10:54:54,768 - Epoch: [73][ 100/ 1200] Overall Loss 0.256755 Objective Loss 0.256755 LR 0.001000 Time 0.026982 -2022-12-06 10:54:54,965 - Epoch: [73][ 110/ 1200] Overall Loss 0.258752 Objective Loss 0.258752 LR 0.001000 Time 0.026311 -2022-12-06 10:54:55,164 - Epoch: [73][ 120/ 1200] Overall Loss 0.258319 Objective Loss 0.258319 LR 0.001000 Time 0.025769 -2022-12-06 10:54:55,360 - Epoch: [73][ 130/ 1200] Overall Loss 0.260867 Objective Loss 0.260867 LR 0.001000 Time 0.025290 -2022-12-06 10:54:55,559 - Epoch: [73][ 140/ 1200] Overall Loss 0.259792 Objective Loss 0.259792 LR 0.001000 Time 0.024903 -2022-12-06 10:54:55,756 - Epoch: [73][ 150/ 1200] Overall Loss 0.263739 Objective Loss 0.263739 LR 0.001000 Time 0.024552 -2022-12-06 10:54:55,955 - Epoch: [73][ 160/ 1200] Overall Loss 0.262103 Objective Loss 0.262103 LR 0.001000 Time 0.024257 -2022-12-06 10:54:56,151 - Epoch: [73][ 170/ 1200] Overall Loss 0.262716 Objective Loss 0.262716 LR 0.001000 Time 0.023982 -2022-12-06 10:54:56,350 - Epoch: [73][ 180/ 1200] Overall Loss 0.261427 Objective Loss 0.261427 LR 0.001000 Time 0.023753 -2022-12-06 10:54:56,547 - Epoch: [73][ 190/ 1200] Overall Loss 0.261559 Objective Loss 0.261559 LR 0.001000 Time 0.023536 -2022-12-06 10:54:56,746 - Epoch: [73][ 200/ 1200] Overall Loss 0.260916 Objective Loss 0.260916 LR 0.001000 Time 0.023350 -2022-12-06 10:54:56,942 - Epoch: [73][ 210/ 1200] Overall Loss 0.260648 Objective Loss 0.260648 LR 0.001000 Time 0.023171 -2022-12-06 10:54:57,142 - Epoch: [73][ 220/ 1200] Overall Loss 0.260805 Objective Loss 0.260805 LR 0.001000 Time 0.023022 -2022-12-06 10:54:57,339 - Epoch: [73][ 230/ 1200] Overall Loss 0.261954 Objective Loss 0.261954 LR 0.001000 Time 0.022875 -2022-12-06 10:54:57,538 - Epoch: [73][ 240/ 1200] Overall Loss 0.261903 Objective Loss 0.261903 LR 0.001000 Time 0.022750 -2022-12-06 10:54:57,734 - Epoch: [73][ 250/ 1200] Overall Loss 0.262688 Objective Loss 0.262688 LR 0.001000 Time 0.022624 -2022-12-06 10:54:57,934 - Epoch: [73][ 260/ 1200] Overall Loss 0.262609 Objective Loss 0.262609 LR 0.001000 Time 0.022521 -2022-12-06 10:54:58,131 - Epoch: [73][ 270/ 1200] Overall Loss 0.261182 Objective Loss 0.261182 LR 0.001000 Time 0.022413 -2022-12-06 10:54:58,330 - Epoch: [73][ 280/ 1200] Overall Loss 0.260717 Objective Loss 0.260717 LR 0.001000 Time 0.022320 -2022-12-06 10:54:58,527 - Epoch: [73][ 290/ 1200] Overall Loss 0.261636 Objective Loss 0.261636 LR 0.001000 Time 0.022228 -2022-12-06 10:54:58,726 - Epoch: [73][ 300/ 1200] Overall Loss 0.262528 Objective Loss 0.262528 LR 0.001000 Time 0.022149 -2022-12-06 10:54:58,923 - Epoch: [73][ 310/ 1200] Overall Loss 0.262739 Objective Loss 0.262739 LR 0.001000 Time 0.022067 -2022-12-06 10:54:59,121 - Epoch: [73][ 320/ 1200] Overall Loss 0.262403 Objective Loss 0.262403 LR 0.001000 Time 0.021996 -2022-12-06 10:54:59,318 - Epoch: [73][ 330/ 1200] Overall Loss 0.262919 Objective Loss 0.262919 LR 0.001000 Time 0.021923 -2022-12-06 10:54:59,517 - Epoch: [73][ 340/ 1200] Overall Loss 0.262621 Objective Loss 0.262621 LR 0.001000 Time 0.021863 -2022-12-06 10:54:59,714 - Epoch: [73][ 350/ 1200] Overall Loss 0.262141 Objective Loss 0.262141 LR 0.001000 Time 0.021800 -2022-12-06 10:54:59,914 - Epoch: [73][ 360/ 1200] Overall Loss 0.261145 Objective Loss 0.261145 LR 0.001000 Time 0.021747 -2022-12-06 10:55:00,110 - Epoch: [73][ 370/ 1200] Overall Loss 0.260394 Objective Loss 0.260394 LR 0.001000 Time 0.021690 -2022-12-06 10:55:00,310 - Epoch: [73][ 380/ 1200] Overall Loss 0.259988 Objective Loss 0.259988 LR 0.001000 Time 0.021642 -2022-12-06 10:55:00,505 - Epoch: [73][ 390/ 1200] Overall Loss 0.260408 Objective Loss 0.260408 LR 0.001000 Time 0.021588 -2022-12-06 10:55:00,705 - Epoch: [73][ 400/ 1200] Overall Loss 0.260616 Objective Loss 0.260616 LR 0.001000 Time 0.021545 -2022-12-06 10:55:00,902 - Epoch: [73][ 410/ 1200] Overall Loss 0.260614 Objective Loss 0.260614 LR 0.001000 Time 0.021499 -2022-12-06 10:55:01,102 - Epoch: [73][ 420/ 1200] Overall Loss 0.260980 Objective Loss 0.260980 LR 0.001000 Time 0.021461 -2022-12-06 10:55:01,298 - Epoch: [73][ 430/ 1200] Overall Loss 0.261168 Objective Loss 0.261168 LR 0.001000 Time 0.021418 -2022-12-06 10:55:01,497 - Epoch: [73][ 440/ 1200] Overall Loss 0.261828 Objective Loss 0.261828 LR 0.001000 Time 0.021383 -2022-12-06 10:55:01,694 - Epoch: [73][ 450/ 1200] Overall Loss 0.261990 Objective Loss 0.261990 LR 0.001000 Time 0.021343 -2022-12-06 10:55:01,894 - Epoch: [73][ 460/ 1200] Overall Loss 0.262120 Objective Loss 0.262120 LR 0.001000 Time 0.021312 -2022-12-06 10:55:02,090 - Epoch: [73][ 470/ 1200] Overall Loss 0.262357 Objective Loss 0.262357 LR 0.001000 Time 0.021275 -2022-12-06 10:55:02,290 - Epoch: [73][ 480/ 1200] Overall Loss 0.262017 Objective Loss 0.262017 LR 0.001000 Time 0.021247 -2022-12-06 10:55:02,486 - Epoch: [73][ 490/ 1200] Overall Loss 0.261303 Objective Loss 0.261303 LR 0.001000 Time 0.021214 -2022-12-06 10:55:02,686 - Epoch: [73][ 500/ 1200] Overall Loss 0.261157 Objective Loss 0.261157 LR 0.001000 Time 0.021187 -2022-12-06 10:55:02,882 - Epoch: [73][ 510/ 1200] Overall Loss 0.261280 Objective Loss 0.261280 LR 0.001000 Time 0.021155 -2022-12-06 10:55:03,082 - Epoch: [73][ 520/ 1200] Overall Loss 0.261345 Objective Loss 0.261345 LR 0.001000 Time 0.021131 -2022-12-06 10:55:03,279 - Epoch: [73][ 530/ 1200] Overall Loss 0.261287 Objective Loss 0.261287 LR 0.001000 Time 0.021103 -2022-12-06 10:55:03,478 - Epoch: [73][ 540/ 1200] Overall Loss 0.261010 Objective Loss 0.261010 LR 0.001000 Time 0.021080 -2022-12-06 10:55:03,674 - Epoch: [73][ 550/ 1200] Overall Loss 0.260891 Objective Loss 0.260891 LR 0.001000 Time 0.021053 -2022-12-06 10:55:03,873 - Epoch: [73][ 560/ 1200] Overall Loss 0.260676 Objective Loss 0.260676 LR 0.001000 Time 0.021031 -2022-12-06 10:55:04,070 - Epoch: [73][ 570/ 1200] Overall Loss 0.260818 Objective Loss 0.260818 LR 0.001000 Time 0.021006 -2022-12-06 10:55:04,268 - Epoch: [73][ 580/ 1200] Overall Loss 0.260543 Objective Loss 0.260543 LR 0.001000 Time 0.020986 -2022-12-06 10:55:04,465 - Epoch: [73][ 590/ 1200] Overall Loss 0.260959 Objective Loss 0.260959 LR 0.001000 Time 0.020963 -2022-12-06 10:55:04,664 - Epoch: [73][ 600/ 1200] Overall Loss 0.261060 Objective Loss 0.261060 LR 0.001000 Time 0.020944 -2022-12-06 10:55:04,861 - Epoch: [73][ 610/ 1200] Overall Loss 0.261371 Objective Loss 0.261371 LR 0.001000 Time 0.020922 -2022-12-06 10:55:05,060 - Epoch: [73][ 620/ 1200] Overall Loss 0.261360 Objective Loss 0.261360 LR 0.001000 Time 0.020905 -2022-12-06 10:55:05,256 - Epoch: [73][ 630/ 1200] Overall Loss 0.261384 Objective Loss 0.261384 LR 0.001000 Time 0.020884 -2022-12-06 10:55:05,455 - Epoch: [73][ 640/ 1200] Overall Loss 0.261750 Objective Loss 0.261750 LR 0.001000 Time 0.020867 -2022-12-06 10:55:05,652 - Epoch: [73][ 650/ 1200] Overall Loss 0.261666 Objective Loss 0.261666 LR 0.001000 Time 0.020848 -2022-12-06 10:55:05,850 - Epoch: [73][ 660/ 1200] Overall Loss 0.261543 Objective Loss 0.261543 LR 0.001000 Time 0.020832 -2022-12-06 10:55:06,047 - Epoch: [73][ 670/ 1200] Overall Loss 0.261485 Objective Loss 0.261485 LR 0.001000 Time 0.020814 -2022-12-06 10:55:06,245 - Epoch: [73][ 680/ 1200] Overall Loss 0.261692 Objective Loss 0.261692 LR 0.001000 Time 0.020799 -2022-12-06 10:55:06,442 - Epoch: [73][ 690/ 1200] Overall Loss 0.261580 Objective Loss 0.261580 LR 0.001000 Time 0.020782 -2022-12-06 10:55:06,642 - Epoch: [73][ 700/ 1200] Overall Loss 0.261543 Objective Loss 0.261543 LR 0.001000 Time 0.020769 -2022-12-06 10:55:06,839 - Epoch: [73][ 710/ 1200] Overall Loss 0.261267 Objective Loss 0.261267 LR 0.001000 Time 0.020753 -2022-12-06 10:55:07,038 - Epoch: [73][ 720/ 1200] Overall Loss 0.260945 Objective Loss 0.260945 LR 0.001000 Time 0.020741 -2022-12-06 10:55:07,234 - Epoch: [73][ 730/ 1200] Overall Loss 0.260971 Objective Loss 0.260971 LR 0.001000 Time 0.020725 -2022-12-06 10:55:07,433 - Epoch: [73][ 740/ 1200] Overall Loss 0.261450 Objective Loss 0.261450 LR 0.001000 Time 0.020712 -2022-12-06 10:55:07,629 - Epoch: [73][ 750/ 1200] Overall Loss 0.261479 Objective Loss 0.261479 LR 0.001000 Time 0.020697 -2022-12-06 10:55:07,828 - Epoch: [73][ 760/ 1200] Overall Loss 0.261453 Objective Loss 0.261453 LR 0.001000 Time 0.020686 -2022-12-06 10:55:08,024 - Epoch: [73][ 770/ 1200] Overall Loss 0.261221 Objective Loss 0.261221 LR 0.001000 Time 0.020671 -2022-12-06 10:55:08,223 - Epoch: [73][ 780/ 1200] Overall Loss 0.261426 Objective Loss 0.261426 LR 0.001000 Time 0.020660 -2022-12-06 10:55:08,420 - Epoch: [73][ 790/ 1200] Overall Loss 0.261060 Objective Loss 0.261060 LR 0.001000 Time 0.020647 -2022-12-06 10:55:08,619 - Epoch: [73][ 800/ 1200] Overall Loss 0.261252 Objective Loss 0.261252 LR 0.001000 Time 0.020637 -2022-12-06 10:55:08,816 - Epoch: [73][ 810/ 1200] Overall Loss 0.261307 Objective Loss 0.261307 LR 0.001000 Time 0.020625 -2022-12-06 10:55:09,014 - Epoch: [73][ 820/ 1200] Overall Loss 0.261606 Objective Loss 0.261606 LR 0.001000 Time 0.020615 -2022-12-06 10:55:09,210 - Epoch: [73][ 830/ 1200] Overall Loss 0.261557 Objective Loss 0.261557 LR 0.001000 Time 0.020602 -2022-12-06 10:55:09,409 - Epoch: [73][ 840/ 1200] Overall Loss 0.261747 Objective Loss 0.261747 LR 0.001000 Time 0.020593 -2022-12-06 10:55:09,606 - Epoch: [73][ 850/ 1200] Overall Loss 0.262129 Objective Loss 0.262129 LR 0.001000 Time 0.020582 -2022-12-06 10:55:09,805 - Epoch: [73][ 860/ 1200] Overall Loss 0.262553 Objective Loss 0.262553 LR 0.001000 Time 0.020573 -2022-12-06 10:55:10,001 - Epoch: [73][ 870/ 1200] Overall Loss 0.262489 Objective Loss 0.262489 LR 0.001000 Time 0.020562 -2022-12-06 10:55:10,201 - Epoch: [73][ 880/ 1200] Overall Loss 0.263064 Objective Loss 0.263064 LR 0.001000 Time 0.020554 -2022-12-06 10:55:10,397 - Epoch: [73][ 890/ 1200] Overall Loss 0.263238 Objective Loss 0.263238 LR 0.001000 Time 0.020543 -2022-12-06 10:55:10,597 - Epoch: [73][ 900/ 1200] Overall Loss 0.263111 Objective Loss 0.263111 LR 0.001000 Time 0.020536 -2022-12-06 10:55:10,794 - Epoch: [73][ 910/ 1200] Overall Loss 0.263129 Objective Loss 0.263129 LR 0.001000 Time 0.020526 -2022-12-06 10:55:10,993 - Epoch: [73][ 920/ 1200] Overall Loss 0.263392 Objective Loss 0.263392 LR 0.001000 Time 0.020519 -2022-12-06 10:55:11,190 - Epoch: [73][ 930/ 1200] Overall Loss 0.263328 Objective Loss 0.263328 LR 0.001000 Time 0.020509 -2022-12-06 10:55:11,388 - Epoch: [73][ 940/ 1200] Overall Loss 0.263629 Objective Loss 0.263629 LR 0.001000 Time 0.020502 -2022-12-06 10:55:11,585 - Epoch: [73][ 950/ 1200] Overall Loss 0.263947 Objective Loss 0.263947 LR 0.001000 Time 0.020492 -2022-12-06 10:55:11,783 - Epoch: [73][ 960/ 1200] Overall Loss 0.264292 Objective Loss 0.264292 LR 0.001000 Time 0.020485 -2022-12-06 10:55:11,980 - Epoch: [73][ 970/ 1200] Overall Loss 0.264498 Objective Loss 0.264498 LR 0.001000 Time 0.020476 -2022-12-06 10:55:12,178 - Epoch: [73][ 980/ 1200] Overall Loss 0.264153 Objective Loss 0.264153 LR 0.001000 Time 0.020469 -2022-12-06 10:55:12,374 - Epoch: [73][ 990/ 1200] Overall Loss 0.264454 Objective Loss 0.264454 LR 0.001000 Time 0.020460 -2022-12-06 10:55:12,573 - Epoch: [73][ 1000/ 1200] Overall Loss 0.264618 Objective Loss 0.264618 LR 0.001000 Time 0.020453 -2022-12-06 10:55:12,769 - Epoch: [73][ 1010/ 1200] Overall Loss 0.264801 Objective Loss 0.264801 LR 0.001000 Time 0.020444 -2022-12-06 10:55:12,968 - Epoch: [73][ 1020/ 1200] Overall Loss 0.264707 Objective Loss 0.264707 LR 0.001000 Time 0.020439 -2022-12-06 10:55:13,164 - Epoch: [73][ 1030/ 1200] Overall Loss 0.264697 Objective Loss 0.264697 LR 0.001000 Time 0.020430 -2022-12-06 10:55:13,363 - Epoch: [73][ 1040/ 1200] Overall Loss 0.264585 Objective Loss 0.264585 LR 0.001000 Time 0.020424 -2022-12-06 10:55:13,560 - Epoch: [73][ 1050/ 1200] Overall Loss 0.264763 Objective Loss 0.264763 LR 0.001000 Time 0.020417 -2022-12-06 10:55:13,759 - Epoch: [73][ 1060/ 1200] Overall Loss 0.264770 Objective Loss 0.264770 LR 0.001000 Time 0.020411 -2022-12-06 10:55:13,956 - Epoch: [73][ 1070/ 1200] Overall Loss 0.264706 Objective Loss 0.264706 LR 0.001000 Time 0.020404 -2022-12-06 10:55:14,154 - Epoch: [73][ 1080/ 1200] Overall Loss 0.264739 Objective Loss 0.264739 LR 0.001000 Time 0.020398 -2022-12-06 10:55:14,350 - Epoch: [73][ 1090/ 1200] Overall Loss 0.264658 Objective Loss 0.264658 LR 0.001000 Time 0.020390 -2022-12-06 10:55:14,549 - Epoch: [73][ 1100/ 1200] Overall Loss 0.264441 Objective Loss 0.264441 LR 0.001000 Time 0.020385 -2022-12-06 10:55:14,745 - Epoch: [73][ 1110/ 1200] Overall Loss 0.264779 Objective Loss 0.264779 LR 0.001000 Time 0.020377 -2022-12-06 10:55:14,943 - Epoch: [73][ 1120/ 1200] Overall Loss 0.264707 Objective Loss 0.264707 LR 0.001000 Time 0.020372 -2022-12-06 10:55:15,140 - Epoch: [73][ 1130/ 1200] Overall Loss 0.264603 Objective Loss 0.264603 LR 0.001000 Time 0.020365 -2022-12-06 10:55:15,338 - Epoch: [73][ 1140/ 1200] Overall Loss 0.264662 Objective Loss 0.264662 LR 0.001000 Time 0.020360 -2022-12-06 10:55:15,535 - Epoch: [73][ 1150/ 1200] Overall Loss 0.264506 Objective Loss 0.264506 LR 0.001000 Time 0.020354 -2022-12-06 10:55:15,733 - Epoch: [73][ 1160/ 1200] Overall Loss 0.264465 Objective Loss 0.264465 LR 0.001000 Time 0.020349 -2022-12-06 10:55:15,930 - Epoch: [73][ 1170/ 1200] Overall Loss 0.264623 Objective Loss 0.264623 LR 0.001000 Time 0.020343 -2022-12-06 10:55:16,129 - Epoch: [73][ 1180/ 1200] Overall Loss 0.264599 Objective Loss 0.264599 LR 0.001000 Time 0.020338 -2022-12-06 10:55:16,325 - Epoch: [73][ 1190/ 1200] Overall Loss 0.264592 Objective Loss 0.264592 LR 0.001000 Time 0.020332 -2022-12-06 10:55:16,561 - Epoch: [73][ 1200/ 1200] Overall Loss 0.264455 Objective Loss 0.264455 Top1 84.100418 Top5 97.907950 LR 0.001000 Time 0.020358 -2022-12-06 10:55:16,649 - --- validate (epoch=73)----------- -2022-12-06 10:55:16,649 - 34129 samples (256 per mini-batch) -2022-12-06 10:55:17,217 - Epoch: [73][ 10/ 134] Loss 0.276224 Top1 84.882812 Top5 98.046875 -2022-12-06 10:55:17,356 - Epoch: [73][ 20/ 134] Loss 0.289552 Top1 85.019531 Top5 98.183594 -2022-12-06 10:55:17,495 - Epoch: [73][ 30/ 134] Loss 0.279712 Top1 85.481771 Top5 98.229167 -2022-12-06 10:55:17,637 - Epoch: [73][ 40/ 134] Loss 0.281711 Top1 85.156250 Top5 98.251953 -2022-12-06 10:55:17,764 - Epoch: [73][ 50/ 134] Loss 0.280411 Top1 85.187500 Top5 98.187500 -2022-12-06 10:55:17,895 - Epoch: [73][ 60/ 134] Loss 0.280061 Top1 85.149740 Top5 98.268229 -2022-12-06 10:55:18,039 - Epoch: [73][ 70/ 134] Loss 0.279320 Top1 85.150670 Top5 98.231027 -2022-12-06 10:55:18,178 - Epoch: [73][ 80/ 134] Loss 0.278666 Top1 85.102539 Top5 98.222656 -2022-12-06 10:55:18,298 - Epoch: [73][ 90/ 134] Loss 0.279580 Top1 85.043403 Top5 98.151042 -2022-12-06 10:55:18,427 - Epoch: [73][ 100/ 134] Loss 0.282352 Top1 84.906250 Top5 98.109375 -2022-12-06 10:55:18,554 - Epoch: [73][ 110/ 134] Loss 0.285511 Top1 84.808239 Top5 98.064631 -2022-12-06 10:55:18,681 - Epoch: [73][ 120/ 134] Loss 0.288748 Top1 84.749349 Top5 98.063151 -2022-12-06 10:55:18,813 - Epoch: [73][ 130/ 134] Loss 0.289301 Top1 84.762620 Top5 98.031851 -2022-12-06 10:55:18,851 - Epoch: [73][ 134/ 134] Loss 0.290452 Top1 84.804712 Top5 98.048580 -2022-12-06 10:55:18,938 - ==> Top1: 84.805 Top5: 98.049 Loss: 0.290 - -2022-12-06 10:55:18,939 - ==> Confusion: -[[ 901 2 3 6 5 8 0 1 5 43 1 3 1 4 5 2 1 2 2 0 1] - [ 2 926 2 4 5 15 3 17 0 0 4 8 1 2 2 1 4 2 18 3 8] - [ 6 5 987 15 3 3 25 11 2 0 4 5 3 4 3 3 2 3 6 4 9] - [ 1 1 27 920 2 2 2 1 1 0 10 0 7 5 13 1 2 3 18 0 4] - [ 12 3 1 1 944 7 1 2 1 6 1 3 1 7 6 9 6 3 0 3 3] - [ 1 18 0 3 6 927 4 31 3 1 1 24 5 25 1 0 1 0 4 7 7] - [ 1 3 10 1 1 2 1068 10 0 0 2 1 1 1 1 5 0 1 1 6 3] - [ 2 9 4 3 1 18 7 961 0 0 3 5 1 2 0 0 0 1 26 9 2] - [ 4 3 0 3 1 4 0 1 946 43 8 1 4 23 13 0 0 2 5 0 3] - [ 71 2 2 2 3 2 0 5 27 853 1 1 0 20 5 0 0 0 0 1 6] - [ 0 1 4 7 1 3 0 1 6 1 948 0 3 21 4 0 2 1 9 2 5] - [ 1 0 2 1 0 6 3 3 1 0 2 967 27 9 0 5 1 7 2 11 3] - [ 0 1 0 5 0 1 0 4 0 0 0 42 876 2 0 10 0 16 1 5 6] - [ 0 0 0 0 0 5 0 3 6 8 5 5 5 969 2 2 3 2 0 1 7] - [ 10 3 2 16 6 3 0 1 24 2 4 2 4 5 1032 0 0 1 8 2 5] - [ 1 0 3 2 3 2 3 1 0 0 0 9 3 7 0 983 4 15 0 3 4] - [ 4 4 2 1 2 0 1 0 0 0 0 3 2 5 3 23 1004 2 1 8 7] - [ 2 1 2 0 0 0 2 1 1 3 0 7 16 7 0 9 0 979 1 2 3] - [ 1 4 5 7 2 2 0 31 1 0 8 2 2 0 8 1 0 0 928 3 3] - [ 1 4 3 3 0 4 5 12 1 0 3 11 6 7 1 2 2 5 1 1005 4] - [ 104 229 207 138 112 162 91 164 73 76 196 127 345 361 152 149 138 132 224 234 9812]] - -2022-12-06 10:55:19,506 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:55:19,507 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:55:19,512 - - -2022-12-06 10:55:19,512 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:55:20,442 - Epoch: [74][ 10/ 1200] Overall Loss 0.268431 Objective Loss 0.268431 LR 0.001000 Time 0.092887 -2022-12-06 10:55:20,640 - Epoch: [74][ 20/ 1200] Overall Loss 0.265673 Objective Loss 0.265673 LR 0.001000 Time 0.056318 -2022-12-06 10:55:20,840 - Epoch: [74][ 30/ 1200] Overall Loss 0.258820 Objective Loss 0.258820 LR 0.001000 Time 0.044188 -2022-12-06 10:55:21,035 - Epoch: [74][ 40/ 1200] Overall Loss 0.260802 Objective Loss 0.260802 LR 0.001000 Time 0.038017 -2022-12-06 10:55:21,234 - Epoch: [74][ 50/ 1200] Overall Loss 0.260981 Objective Loss 0.260981 LR 0.001000 Time 0.034380 -2022-12-06 10:55:21,430 - Epoch: [74][ 60/ 1200] Overall Loss 0.260663 Objective Loss 0.260663 LR 0.001000 Time 0.031907 -2022-12-06 10:55:21,630 - Epoch: [74][ 70/ 1200] Overall Loss 0.263678 Objective Loss 0.263678 LR 0.001000 Time 0.030188 -2022-12-06 10:55:21,825 - Epoch: [74][ 80/ 1200] Overall Loss 0.264746 Objective Loss 0.264746 LR 0.001000 Time 0.028856 -2022-12-06 10:55:22,024 - Epoch: [74][ 90/ 1200] Overall Loss 0.264720 Objective Loss 0.264720 LR 0.001000 Time 0.027854 -2022-12-06 10:55:22,221 - Epoch: [74][ 100/ 1200] Overall Loss 0.262322 Objective Loss 0.262322 LR 0.001000 Time 0.027025 -2022-12-06 10:55:22,420 - Epoch: [74][ 110/ 1200] Overall Loss 0.263370 Objective Loss 0.263370 LR 0.001000 Time 0.026374 -2022-12-06 10:55:22,617 - Epoch: [74][ 120/ 1200] Overall Loss 0.263411 Objective Loss 0.263411 LR 0.001000 Time 0.025818 -2022-12-06 10:55:22,819 - Epoch: [74][ 130/ 1200] Overall Loss 0.264924 Objective Loss 0.264924 LR 0.001000 Time 0.025377 -2022-12-06 10:55:23,018 - Epoch: [74][ 140/ 1200] Overall Loss 0.264755 Objective Loss 0.264755 LR 0.001000 Time 0.024982 -2022-12-06 10:55:23,219 - Epoch: [74][ 150/ 1200] Overall Loss 0.266228 Objective Loss 0.266228 LR 0.001000 Time 0.024655 -2022-12-06 10:55:23,417 - Epoch: [74][ 160/ 1200] Overall Loss 0.265615 Objective Loss 0.265615 LR 0.001000 Time 0.024349 -2022-12-06 10:55:23,619 - Epoch: [74][ 170/ 1200] Overall Loss 0.263596 Objective Loss 0.263596 LR 0.001000 Time 0.024099 -2022-12-06 10:55:23,818 - Epoch: [74][ 180/ 1200] Overall Loss 0.263539 Objective Loss 0.263539 LR 0.001000 Time 0.023862 -2022-12-06 10:55:24,019 - Epoch: [74][ 190/ 1200] Overall Loss 0.262650 Objective Loss 0.262650 LR 0.001000 Time 0.023663 -2022-12-06 10:55:24,217 - Epoch: [74][ 200/ 1200] Overall Loss 0.261219 Objective Loss 0.261219 LR 0.001000 Time 0.023466 -2022-12-06 10:55:24,417 - Epoch: [74][ 210/ 1200] Overall Loss 0.261653 Objective Loss 0.261653 LR 0.001000 Time 0.023300 -2022-12-06 10:55:24,615 - Epoch: [74][ 220/ 1200] Overall Loss 0.260013 Objective Loss 0.260013 LR 0.001000 Time 0.023139 -2022-12-06 10:55:24,816 - Epoch: [74][ 230/ 1200] Overall Loss 0.260669 Objective Loss 0.260669 LR 0.001000 Time 0.023002 -2022-12-06 10:55:25,014 - Epoch: [74][ 240/ 1200] Overall Loss 0.260646 Objective Loss 0.260646 LR 0.001000 Time 0.022868 -2022-12-06 10:55:25,215 - Epoch: [74][ 250/ 1200] Overall Loss 0.259579 Objective Loss 0.259579 LR 0.001000 Time 0.022755 -2022-12-06 10:55:25,412 - Epoch: [74][ 260/ 1200] Overall Loss 0.261484 Objective Loss 0.261484 LR 0.001000 Time 0.022637 -2022-12-06 10:55:25,613 - Epoch: [74][ 270/ 1200] Overall Loss 0.261515 Objective Loss 0.261515 LR 0.001000 Time 0.022539 -2022-12-06 10:55:25,810 - Epoch: [74][ 280/ 1200] Overall Loss 0.261996 Objective Loss 0.261996 LR 0.001000 Time 0.022438 -2022-12-06 10:55:26,012 - Epoch: [74][ 290/ 1200] Overall Loss 0.262585 Objective Loss 0.262585 LR 0.001000 Time 0.022356 -2022-12-06 10:55:26,209 - Epoch: [74][ 300/ 1200] Overall Loss 0.262148 Objective Loss 0.262148 LR 0.001000 Time 0.022266 -2022-12-06 10:55:26,410 - Epoch: [74][ 310/ 1200] Overall Loss 0.261993 Objective Loss 0.261993 LR 0.001000 Time 0.022196 -2022-12-06 10:55:26,608 - Epoch: [74][ 320/ 1200] Overall Loss 0.262168 Objective Loss 0.262168 LR 0.001000 Time 0.022119 -2022-12-06 10:55:26,808 - Epoch: [74][ 330/ 1200] Overall Loss 0.261992 Objective Loss 0.261992 LR 0.001000 Time 0.022054 -2022-12-06 10:55:27,007 - Epoch: [74][ 340/ 1200] Overall Loss 0.262301 Objective Loss 0.262301 LR 0.001000 Time 0.021987 -2022-12-06 10:55:27,208 - Epoch: [74][ 350/ 1200] Overall Loss 0.262081 Objective Loss 0.262081 LR 0.001000 Time 0.021931 -2022-12-06 10:55:27,406 - Epoch: [74][ 360/ 1200] Overall Loss 0.261502 Objective Loss 0.261502 LR 0.001000 Time 0.021872 -2022-12-06 10:55:27,607 - Epoch: [74][ 370/ 1200] Overall Loss 0.261623 Objective Loss 0.261623 LR 0.001000 Time 0.021822 -2022-12-06 10:55:27,806 - Epoch: [74][ 380/ 1200] Overall Loss 0.261208 Objective Loss 0.261208 LR 0.001000 Time 0.021769 -2022-12-06 10:55:28,007 - Epoch: [74][ 390/ 1200] Overall Loss 0.261549 Objective Loss 0.261549 LR 0.001000 Time 0.021726 -2022-12-06 10:55:28,205 - Epoch: [74][ 400/ 1200] Overall Loss 0.261048 Objective Loss 0.261048 LR 0.001000 Time 0.021678 -2022-12-06 10:55:28,407 - Epoch: [74][ 410/ 1200] Overall Loss 0.261730 Objective Loss 0.261730 LR 0.001000 Time 0.021639 -2022-12-06 10:55:28,606 - Epoch: [74][ 420/ 1200] Overall Loss 0.261940 Objective Loss 0.261940 LR 0.001000 Time 0.021595 -2022-12-06 10:55:28,807 - Epoch: [74][ 430/ 1200] Overall Loss 0.261917 Objective Loss 0.261917 LR 0.001000 Time 0.021559 -2022-12-06 10:55:29,005 - Epoch: [74][ 440/ 1200] Overall Loss 0.262264 Objective Loss 0.262264 LR 0.001000 Time 0.021520 -2022-12-06 10:55:29,206 - Epoch: [74][ 450/ 1200] Overall Loss 0.262250 Objective Loss 0.262250 LR 0.001000 Time 0.021486 -2022-12-06 10:55:29,405 - Epoch: [74][ 460/ 1200] Overall Loss 0.262697 Objective Loss 0.262697 LR 0.001000 Time 0.021450 -2022-12-06 10:55:29,606 - Epoch: [74][ 470/ 1200] Overall Loss 0.262082 Objective Loss 0.262082 LR 0.001000 Time 0.021420 -2022-12-06 10:55:29,805 - Epoch: [74][ 480/ 1200] Overall Loss 0.262725 Objective Loss 0.262725 LR 0.001000 Time 0.021387 -2022-12-06 10:55:30,006 - Epoch: [74][ 490/ 1200] Overall Loss 0.262578 Objective Loss 0.262578 LR 0.001000 Time 0.021361 -2022-12-06 10:55:30,205 - Epoch: [74][ 500/ 1200] Overall Loss 0.262351 Objective Loss 0.262351 LR 0.001000 Time 0.021330 -2022-12-06 10:55:30,407 - Epoch: [74][ 510/ 1200] Overall Loss 0.263734 Objective Loss 0.263734 LR 0.001000 Time 0.021306 -2022-12-06 10:55:30,606 - Epoch: [74][ 520/ 1200] Overall Loss 0.264875 Objective Loss 0.264875 LR 0.001000 Time 0.021279 -2022-12-06 10:55:30,808 - Epoch: [74][ 530/ 1200] Overall Loss 0.265037 Objective Loss 0.265037 LR 0.001000 Time 0.021258 -2022-12-06 10:55:31,006 - Epoch: [74][ 540/ 1200] Overall Loss 0.264650 Objective Loss 0.264650 LR 0.001000 Time 0.021230 -2022-12-06 10:55:31,208 - Epoch: [74][ 550/ 1200] Overall Loss 0.265229 Objective Loss 0.265229 LR 0.001000 Time 0.021210 -2022-12-06 10:55:31,406 - Epoch: [74][ 560/ 1200] Overall Loss 0.264874 Objective Loss 0.264874 LR 0.001000 Time 0.021183 -2022-12-06 10:55:31,608 - Epoch: [74][ 570/ 1200] Overall Loss 0.265184 Objective Loss 0.265184 LR 0.001000 Time 0.021165 -2022-12-06 10:55:31,806 - Epoch: [74][ 580/ 1200] Overall Loss 0.264907 Objective Loss 0.264907 LR 0.001000 Time 0.021141 -2022-12-06 10:55:32,007 - Epoch: [74][ 590/ 1200] Overall Loss 0.264685 Objective Loss 0.264685 LR 0.001000 Time 0.021123 -2022-12-06 10:55:32,206 - Epoch: [74][ 600/ 1200] Overall Loss 0.264367 Objective Loss 0.264367 LR 0.001000 Time 0.021101 -2022-12-06 10:55:32,408 - Epoch: [74][ 610/ 1200] Overall Loss 0.263809 Objective Loss 0.263809 LR 0.001000 Time 0.021086 -2022-12-06 10:55:32,606 - Epoch: [74][ 620/ 1200] Overall Loss 0.263273 Objective Loss 0.263273 LR 0.001000 Time 0.021064 -2022-12-06 10:55:32,806 - Epoch: [74][ 630/ 1200] Overall Loss 0.263297 Objective Loss 0.263297 LR 0.001000 Time 0.021046 -2022-12-06 10:55:33,003 - Epoch: [74][ 640/ 1200] Overall Loss 0.263352 Objective Loss 0.263352 LR 0.001000 Time 0.021024 -2022-12-06 10:55:33,202 - Epoch: [74][ 650/ 1200] Overall Loss 0.263277 Objective Loss 0.263277 LR 0.001000 Time 0.021007 -2022-12-06 10:55:33,399 - Epoch: [74][ 660/ 1200] Overall Loss 0.263449 Objective Loss 0.263449 LR 0.001000 Time 0.020986 -2022-12-06 10:55:33,599 - Epoch: [74][ 670/ 1200] Overall Loss 0.263081 Objective Loss 0.263081 LR 0.001000 Time 0.020970 -2022-12-06 10:55:33,797 - Epoch: [74][ 680/ 1200] Overall Loss 0.262976 Objective Loss 0.262976 LR 0.001000 Time 0.020952 -2022-12-06 10:55:33,997 - Epoch: [74][ 690/ 1200] Overall Loss 0.263138 Objective Loss 0.263138 LR 0.001000 Time 0.020937 -2022-12-06 10:55:34,192 - Epoch: [74][ 700/ 1200] Overall Loss 0.263362 Objective Loss 0.263362 LR 0.001000 Time 0.020916 -2022-12-06 10:55:34,392 - Epoch: [74][ 710/ 1200] Overall Loss 0.263756 Objective Loss 0.263756 LR 0.001000 Time 0.020902 -2022-12-06 10:55:34,589 - Epoch: [74][ 720/ 1200] Overall Loss 0.263824 Objective Loss 0.263824 LR 0.001000 Time 0.020885 -2022-12-06 10:55:34,789 - Epoch: [74][ 730/ 1200] Overall Loss 0.263593 Objective Loss 0.263593 LR 0.001000 Time 0.020872 -2022-12-06 10:55:34,985 - Epoch: [74][ 740/ 1200] Overall Loss 0.263837 Objective Loss 0.263837 LR 0.001000 Time 0.020854 -2022-12-06 10:55:35,184 - Epoch: [74][ 750/ 1200] Overall Loss 0.263912 Objective Loss 0.263912 LR 0.001000 Time 0.020841 -2022-12-06 10:55:35,380 - Epoch: [74][ 760/ 1200] Overall Loss 0.264551 Objective Loss 0.264551 LR 0.001000 Time 0.020823 -2022-12-06 10:55:35,579 - Epoch: [74][ 770/ 1200] Overall Loss 0.265224 Objective Loss 0.265224 LR 0.001000 Time 0.020810 -2022-12-06 10:55:35,775 - Epoch: [74][ 780/ 1200] Overall Loss 0.265524 Objective Loss 0.265524 LR 0.001000 Time 0.020795 -2022-12-06 10:55:35,975 - Epoch: [74][ 790/ 1200] Overall Loss 0.265662 Objective Loss 0.265662 LR 0.001000 Time 0.020784 -2022-12-06 10:55:36,171 - Epoch: [74][ 800/ 1200] Overall Loss 0.265591 Objective Loss 0.265591 LR 0.001000 Time 0.020768 -2022-12-06 10:55:36,371 - Epoch: [74][ 810/ 1200] Overall Loss 0.266024 Objective Loss 0.266024 LR 0.001000 Time 0.020758 -2022-12-06 10:55:36,568 - Epoch: [74][ 820/ 1200] Overall Loss 0.265908 Objective Loss 0.265908 LR 0.001000 Time 0.020744 -2022-12-06 10:55:36,767 - Epoch: [74][ 830/ 1200] Overall Loss 0.266343 Objective Loss 0.266343 LR 0.001000 Time 0.020734 -2022-12-06 10:55:36,964 - Epoch: [74][ 840/ 1200] Overall Loss 0.266489 Objective Loss 0.266489 LR 0.001000 Time 0.020720 -2022-12-06 10:55:37,163 - Epoch: [74][ 850/ 1200] Overall Loss 0.266422 Objective Loss 0.266422 LR 0.001000 Time 0.020710 -2022-12-06 10:55:37,359 - Epoch: [74][ 860/ 1200] Overall Loss 0.266535 Objective Loss 0.266535 LR 0.001000 Time 0.020697 -2022-12-06 10:55:37,559 - Epoch: [74][ 870/ 1200] Overall Loss 0.266334 Objective Loss 0.266334 LR 0.001000 Time 0.020688 -2022-12-06 10:55:37,755 - Epoch: [74][ 880/ 1200] Overall Loss 0.266207 Objective Loss 0.266207 LR 0.001000 Time 0.020675 -2022-12-06 10:55:37,955 - Epoch: [74][ 890/ 1200] Overall Loss 0.266019 Objective Loss 0.266019 LR 0.001000 Time 0.020667 -2022-12-06 10:55:38,151 - Epoch: [74][ 900/ 1200] Overall Loss 0.266065 Objective Loss 0.266065 LR 0.001000 Time 0.020655 -2022-12-06 10:55:38,350 - Epoch: [74][ 910/ 1200] Overall Loss 0.266213 Objective Loss 0.266213 LR 0.001000 Time 0.020646 -2022-12-06 10:55:38,546 - Epoch: [74][ 920/ 1200] Overall Loss 0.266539 Objective Loss 0.266539 LR 0.001000 Time 0.020634 -2022-12-06 10:55:38,746 - Epoch: [74][ 930/ 1200] Overall Loss 0.266759 Objective Loss 0.266759 LR 0.001000 Time 0.020627 -2022-12-06 10:55:38,943 - Epoch: [74][ 940/ 1200] Overall Loss 0.266446 Objective Loss 0.266446 LR 0.001000 Time 0.020616 -2022-12-06 10:55:39,143 - Epoch: [74][ 950/ 1200] Overall Loss 0.267025 Objective Loss 0.267025 LR 0.001000 Time 0.020609 -2022-12-06 10:55:39,339 - Epoch: [74][ 960/ 1200] Overall Loss 0.267528 Objective Loss 0.267528 LR 0.001000 Time 0.020598 -2022-12-06 10:55:39,539 - Epoch: [74][ 970/ 1200] Overall Loss 0.267983 Objective Loss 0.267983 LR 0.001000 Time 0.020591 -2022-12-06 10:55:39,735 - Epoch: [74][ 980/ 1200] Overall Loss 0.268049 Objective Loss 0.268049 LR 0.001000 Time 0.020580 -2022-12-06 10:55:39,935 - Epoch: [74][ 990/ 1200] Overall Loss 0.268031 Objective Loss 0.268031 LR 0.001000 Time 0.020574 -2022-12-06 10:55:40,131 - Epoch: [74][ 1000/ 1200] Overall Loss 0.268215 Objective Loss 0.268215 LR 0.001000 Time 0.020564 -2022-12-06 10:55:40,331 - Epoch: [74][ 1010/ 1200] Overall Loss 0.268322 Objective Loss 0.268322 LR 0.001000 Time 0.020557 -2022-12-06 10:55:40,528 - Epoch: [74][ 1020/ 1200] Overall Loss 0.268522 Objective Loss 0.268522 LR 0.001000 Time 0.020549 -2022-12-06 10:55:40,728 - Epoch: [74][ 1030/ 1200] Overall Loss 0.268477 Objective Loss 0.268477 LR 0.001000 Time 0.020543 -2022-12-06 10:55:40,924 - Epoch: [74][ 1040/ 1200] Overall Loss 0.268479 Objective Loss 0.268479 LR 0.001000 Time 0.020533 -2022-12-06 10:55:41,123 - Epoch: [74][ 1050/ 1200] Overall Loss 0.268433 Objective Loss 0.268433 LR 0.001000 Time 0.020527 -2022-12-06 10:55:41,320 - Epoch: [74][ 1060/ 1200] Overall Loss 0.268294 Objective Loss 0.268294 LR 0.001000 Time 0.020518 -2022-12-06 10:55:41,519 - Epoch: [74][ 1070/ 1200] Overall Loss 0.268324 Objective Loss 0.268324 LR 0.001000 Time 0.020512 -2022-12-06 10:55:41,716 - Epoch: [74][ 1080/ 1200] Overall Loss 0.268492 Objective Loss 0.268492 LR 0.001000 Time 0.020504 -2022-12-06 10:55:41,915 - Epoch: [74][ 1090/ 1200] Overall Loss 0.268711 Objective Loss 0.268711 LR 0.001000 Time 0.020498 -2022-12-06 10:55:42,111 - Epoch: [74][ 1100/ 1200] Overall Loss 0.269175 Objective Loss 0.269175 LR 0.001000 Time 0.020489 -2022-12-06 10:55:42,311 - Epoch: [74][ 1110/ 1200] Overall Loss 0.269057 Objective Loss 0.269057 LR 0.001000 Time 0.020484 -2022-12-06 10:55:42,508 - Epoch: [74][ 1120/ 1200] Overall Loss 0.269361 Objective Loss 0.269361 LR 0.001000 Time 0.020476 -2022-12-06 10:55:42,707 - Epoch: [74][ 1130/ 1200] Overall Loss 0.269438 Objective Loss 0.269438 LR 0.001000 Time 0.020471 -2022-12-06 10:55:42,904 - Epoch: [74][ 1140/ 1200] Overall Loss 0.269583 Objective Loss 0.269583 LR 0.001000 Time 0.020464 -2022-12-06 10:55:43,103 - Epoch: [74][ 1150/ 1200] Overall Loss 0.269850 Objective Loss 0.269850 LR 0.001000 Time 0.020459 -2022-12-06 10:55:43,299 - Epoch: [74][ 1160/ 1200] Overall Loss 0.269833 Objective Loss 0.269833 LR 0.001000 Time 0.020451 -2022-12-06 10:55:43,499 - Epoch: [74][ 1170/ 1200] Overall Loss 0.269744 Objective Loss 0.269744 LR 0.001000 Time 0.020446 -2022-12-06 10:55:43,695 - Epoch: [74][ 1180/ 1200] Overall Loss 0.269709 Objective Loss 0.269709 LR 0.001000 Time 0.020439 -2022-12-06 10:55:43,895 - Epoch: [74][ 1190/ 1200] Overall Loss 0.269866 Objective Loss 0.269866 LR 0.001000 Time 0.020435 -2022-12-06 10:55:44,126 - Epoch: [74][ 1200/ 1200] Overall Loss 0.270027 Objective Loss 0.270027 Top1 83.891213 Top5 98.326360 LR 0.001000 Time 0.020456 -2022-12-06 10:55:44,214 - --- validate (epoch=74)----------- -2022-12-06 10:55:44,214 - 34129 samples (256 per mini-batch) -2022-12-06 10:55:44,656 - Epoch: [74][ 10/ 134] Loss 0.268528 Top1 84.062500 Top5 98.164062 -2022-12-06 10:55:44,786 - Epoch: [74][ 20/ 134] Loss 0.294106 Top1 83.808594 Top5 98.183594 -2022-12-06 10:55:44,925 - Epoch: [74][ 30/ 134] Loss 0.310685 Top1 83.489583 Top5 97.929688 -2022-12-06 10:55:45,070 - Epoch: [74][ 40/ 134] Loss 0.311818 Top1 83.623047 Top5 97.880859 -2022-12-06 10:55:45,211 - Epoch: [74][ 50/ 134] Loss 0.306652 Top1 83.812500 Top5 97.906250 -2022-12-06 10:55:45,354 - Epoch: [74][ 60/ 134] Loss 0.298403 Top1 84.003906 Top5 97.897135 -2022-12-06 10:55:45,496 - Epoch: [74][ 70/ 134] Loss 0.301223 Top1 83.906250 Top5 97.885045 -2022-12-06 10:55:45,638 - Epoch: [74][ 80/ 134] Loss 0.300616 Top1 83.833008 Top5 97.841797 -2022-12-06 10:55:45,777 - Epoch: [74][ 90/ 134] Loss 0.300977 Top1 83.923611 Top5 97.782118 -2022-12-06 10:55:45,919 - Epoch: [74][ 100/ 134] Loss 0.300598 Top1 83.835938 Top5 97.734375 -2022-12-06 10:55:46,059 - Epoch: [74][ 110/ 134] Loss 0.302806 Top1 83.778409 Top5 97.730824 -2022-12-06 10:55:46,188 - Epoch: [74][ 120/ 134] Loss 0.302676 Top1 83.880208 Top5 97.727865 -2022-12-06 10:55:46,319 - Epoch: [74][ 130/ 134] Loss 0.300907 Top1 83.942308 Top5 97.746394 -2022-12-06 10:55:46,355 - Epoch: [74][ 134/ 134] Loss 0.300555 Top1 83.978435 Top5 97.737994 -2022-12-06 10:55:46,443 - ==> Top1: 83.978 Top5: 97.738 Loss: 0.301 - -2022-12-06 10:55:46,444 - ==> Confusion: -[[ 863 1 0 4 10 8 1 1 10 78 0 3 3 2 4 2 1 3 0 0 2] - [ 1 913 1 1 8 33 6 21 2 1 0 4 1 1 3 1 4 3 16 2 5] - [ 6 2 983 16 2 5 27 17 2 3 5 5 1 4 5 1 1 0 5 3 10] - [ 2 0 18 917 0 6 0 1 1 1 7 1 7 3 28 0 1 4 20 0 3] - [ 8 4 3 1 939 11 0 3 1 8 0 6 0 5 8 6 9 3 2 1 2] - [ 2 9 0 1 6 983 1 25 4 1 0 12 4 8 2 1 2 0 2 2 4] - [ 1 1 12 0 0 4 1071 5 0 1 1 3 0 1 2 4 0 2 1 6 3] - [ 3 2 1 1 1 29 4 978 0 2 2 6 0 1 0 1 0 1 14 5 3] - [ 2 3 0 1 0 4 0 2 941 58 8 2 2 10 20 0 3 3 3 1 1] - [ 49 0 1 0 8 3 0 6 21 887 2 0 0 14 2 0 1 2 0 1 4] - [ 0 1 3 8 1 4 2 6 15 1 930 3 1 15 7 0 1 0 17 0 4] - [ 4 2 2 0 0 11 1 4 2 0 0 973 28 6 0 5 3 1 0 7 2] - [ 0 1 2 3 0 4 1 1 2 0 2 39 878 2 2 5 2 7 4 4 10] - [ 1 0 0 0 2 11 0 5 11 18 5 5 4 942 1 3 2 0 1 3 9] - [ 10 3 2 7 7 3 1 1 15 7 0 2 2 3 1055 0 2 0 7 0 3] - [ 2 1 7 2 1 7 2 0 0 0 0 14 4 1 0 983 8 7 0 2 2] - [ 4 3 3 0 0 4 0 1 1 2 0 0 4 3 2 12 1017 1 1 4 10] - [ 1 1 2 2 0 1 1 3 2 1 0 16 21 3 3 11 0 964 0 1 3] - [ 1 6 2 4 0 2 1 40 3 1 1 1 5 0 8 0 0 0 929 2 2] - [ 2 4 1 0 0 13 6 16 2 0 0 12 7 6 0 4 3 3 0 994 7] - [ 146 230 156 105 94 289 95 214 92 128 152 130 397 354 220 103 171 65 249 314 9522]] - -2022-12-06 10:55:47,110 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:55:47,110 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:55:47,116 - - -2022-12-06 10:55:47,116 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:55:48,060 - Epoch: [75][ 10/ 1200] Overall Loss 0.270309 Objective Loss 0.270309 LR 0.001000 Time 0.094377 -2022-12-06 10:55:48,264 - Epoch: [75][ 20/ 1200] Overall Loss 0.284034 Objective Loss 0.284034 LR 0.001000 Time 0.057352 -2022-12-06 10:55:48,461 - Epoch: [75][ 30/ 1200] Overall Loss 0.280789 Objective Loss 0.280789 LR 0.001000 Time 0.044787 -2022-12-06 10:55:48,660 - Epoch: [75][ 40/ 1200] Overall Loss 0.273792 Objective Loss 0.273792 LR 0.001000 Time 0.038531 -2022-12-06 10:55:48,856 - Epoch: [75][ 50/ 1200] Overall Loss 0.277690 Objective Loss 0.277690 LR 0.001000 Time 0.034742 -2022-12-06 10:55:49,054 - Epoch: [75][ 60/ 1200] Overall Loss 0.274068 Objective Loss 0.274068 LR 0.001000 Time 0.032240 -2022-12-06 10:55:49,251 - Epoch: [75][ 70/ 1200] Overall Loss 0.274571 Objective Loss 0.274571 LR 0.001000 Time 0.030436 -2022-12-06 10:55:49,449 - Epoch: [75][ 80/ 1200] Overall Loss 0.275796 Objective Loss 0.275796 LR 0.001000 Time 0.029104 -2022-12-06 10:55:49,645 - Epoch: [75][ 90/ 1200] Overall Loss 0.271310 Objective Loss 0.271310 LR 0.001000 Time 0.028039 -2022-12-06 10:55:49,843 - Epoch: [75][ 100/ 1200] Overall Loss 0.271694 Objective Loss 0.271694 LR 0.001000 Time 0.027210 -2022-12-06 10:55:50,039 - Epoch: [75][ 110/ 1200] Overall Loss 0.269746 Objective Loss 0.269746 LR 0.001000 Time 0.026514 -2022-12-06 10:55:50,236 - Epoch: [75][ 120/ 1200] Overall Loss 0.271890 Objective Loss 0.271890 LR 0.001000 Time 0.025947 -2022-12-06 10:55:50,433 - Epoch: [75][ 130/ 1200] Overall Loss 0.270470 Objective Loss 0.270470 LR 0.001000 Time 0.025456 -2022-12-06 10:55:50,630 - Epoch: [75][ 140/ 1200] Overall Loss 0.270478 Objective Loss 0.270478 LR 0.001000 Time 0.025046 -2022-12-06 10:55:50,827 - Epoch: [75][ 150/ 1200] Overall Loss 0.268193 Objective Loss 0.268193 LR 0.001000 Time 0.024681 -2022-12-06 10:55:51,025 - Epoch: [75][ 160/ 1200] Overall Loss 0.268993 Objective Loss 0.268993 LR 0.001000 Time 0.024373 -2022-12-06 10:55:51,221 - Epoch: [75][ 170/ 1200] Overall Loss 0.267736 Objective Loss 0.267736 LR 0.001000 Time 0.024089 -2022-12-06 10:55:51,419 - Epoch: [75][ 180/ 1200] Overall Loss 0.267077 Objective Loss 0.267077 LR 0.001000 Time 0.023848 -2022-12-06 10:55:51,615 - Epoch: [75][ 190/ 1200] Overall Loss 0.266180 Objective Loss 0.266180 LR 0.001000 Time 0.023621 -2022-12-06 10:55:51,813 - Epoch: [75][ 200/ 1200] Overall Loss 0.267488 Objective Loss 0.267488 LR 0.001000 Time 0.023430 -2022-12-06 10:55:52,009 - Epoch: [75][ 210/ 1200] Overall Loss 0.265940 Objective Loss 0.265940 LR 0.001000 Time 0.023244 -2022-12-06 10:55:52,207 - Epoch: [75][ 220/ 1200] Overall Loss 0.265554 Objective Loss 0.265554 LR 0.001000 Time 0.023085 -2022-12-06 10:55:52,404 - Epoch: [75][ 230/ 1200] Overall Loss 0.265793 Objective Loss 0.265793 LR 0.001000 Time 0.022936 -2022-12-06 10:55:52,604 - Epoch: [75][ 240/ 1200] Overall Loss 0.267078 Objective Loss 0.267078 LR 0.001000 Time 0.022812 -2022-12-06 10:55:52,801 - Epoch: [75][ 250/ 1200] Overall Loss 0.267308 Objective Loss 0.267308 LR 0.001000 Time 0.022686 -2022-12-06 10:55:53,000 - Epoch: [75][ 260/ 1200] Overall Loss 0.266819 Objective Loss 0.266819 LR 0.001000 Time 0.022573 -2022-12-06 10:55:53,196 - Epoch: [75][ 270/ 1200] Overall Loss 0.267151 Objective Loss 0.267151 LR 0.001000 Time 0.022462 -2022-12-06 10:55:53,394 - Epoch: [75][ 280/ 1200] Overall Loss 0.267051 Objective Loss 0.267051 LR 0.001000 Time 0.022364 -2022-12-06 10:55:53,591 - Epoch: [75][ 290/ 1200] Overall Loss 0.266723 Objective Loss 0.266723 LR 0.001000 Time 0.022270 -2022-12-06 10:55:53,788 - Epoch: [75][ 300/ 1200] Overall Loss 0.266743 Objective Loss 0.266743 LR 0.001000 Time 0.022186 -2022-12-06 10:55:53,984 - Epoch: [75][ 310/ 1200] Overall Loss 0.266267 Objective Loss 0.266267 LR 0.001000 Time 0.022101 -2022-12-06 10:55:54,183 - Epoch: [75][ 320/ 1200] Overall Loss 0.266204 Objective Loss 0.266204 LR 0.001000 Time 0.022029 -2022-12-06 10:55:54,380 - Epoch: [75][ 330/ 1200] Overall Loss 0.264998 Objective Loss 0.264998 LR 0.001000 Time 0.021956 -2022-12-06 10:55:54,579 - Epoch: [75][ 340/ 1200] Overall Loss 0.265634 Objective Loss 0.265634 LR 0.001000 Time 0.021894 -2022-12-06 10:55:54,774 - Epoch: [75][ 350/ 1200] Overall Loss 0.265973 Objective Loss 0.265973 LR 0.001000 Time 0.021825 -2022-12-06 10:55:54,973 - Epoch: [75][ 360/ 1200] Overall Loss 0.265441 Objective Loss 0.265441 LR 0.001000 Time 0.021770 -2022-12-06 10:55:55,170 - Epoch: [75][ 370/ 1200] Overall Loss 0.265656 Objective Loss 0.265656 LR 0.001000 Time 0.021711 -2022-12-06 10:55:55,369 - Epoch: [75][ 380/ 1200] Overall Loss 0.265591 Objective Loss 0.265591 LR 0.001000 Time 0.021662 -2022-12-06 10:55:55,564 - Epoch: [75][ 390/ 1200] Overall Loss 0.265949 Objective Loss 0.265949 LR 0.001000 Time 0.021607 -2022-12-06 10:55:55,763 - Epoch: [75][ 400/ 1200] Overall Loss 0.265706 Objective Loss 0.265706 LR 0.001000 Time 0.021562 -2022-12-06 10:55:55,959 - Epoch: [75][ 410/ 1200] Overall Loss 0.266288 Objective Loss 0.266288 LR 0.001000 Time 0.021513 -2022-12-06 10:55:56,157 - Epoch: [75][ 420/ 1200] Overall Loss 0.266275 Objective Loss 0.266275 LR 0.001000 Time 0.021472 -2022-12-06 10:55:56,354 - Epoch: [75][ 430/ 1200] Overall Loss 0.266335 Objective Loss 0.266335 LR 0.001000 Time 0.021428 -2022-12-06 10:55:56,552 - Epoch: [75][ 440/ 1200] Overall Loss 0.266729 Objective Loss 0.266729 LR 0.001000 Time 0.021391 -2022-12-06 10:55:56,748 - Epoch: [75][ 450/ 1200] Overall Loss 0.266819 Objective Loss 0.266819 LR 0.001000 Time 0.021350 -2022-12-06 10:55:56,947 - Epoch: [75][ 460/ 1200] Overall Loss 0.267111 Objective Loss 0.267111 LR 0.001000 Time 0.021316 -2022-12-06 10:55:57,144 - Epoch: [75][ 470/ 1200] Overall Loss 0.266524 Objective Loss 0.266524 LR 0.001000 Time 0.021280 -2022-12-06 10:55:57,342 - Epoch: [75][ 480/ 1200] Overall Loss 0.266522 Objective Loss 0.266522 LR 0.001000 Time 0.021249 -2022-12-06 10:55:57,539 - Epoch: [75][ 490/ 1200] Overall Loss 0.267119 Objective Loss 0.267119 LR 0.001000 Time 0.021217 -2022-12-06 10:55:57,737 - Epoch: [75][ 500/ 1200] Overall Loss 0.266550 Objective Loss 0.266550 LR 0.001000 Time 0.021187 -2022-12-06 10:55:57,933 - Epoch: [75][ 510/ 1200] Overall Loss 0.266456 Objective Loss 0.266456 LR 0.001000 Time 0.021155 -2022-12-06 10:55:58,131 - Epoch: [75][ 520/ 1200] Overall Loss 0.266203 Objective Loss 0.266203 LR 0.001000 Time 0.021127 -2022-12-06 10:55:58,328 - Epoch: [75][ 530/ 1200] Overall Loss 0.266217 Objective Loss 0.266217 LR 0.001000 Time 0.021099 -2022-12-06 10:55:58,526 - Epoch: [75][ 540/ 1200] Overall Loss 0.266286 Objective Loss 0.266286 LR 0.001000 Time 0.021074 -2022-12-06 10:55:58,722 - Epoch: [75][ 550/ 1200] Overall Loss 0.265825 Objective Loss 0.265825 LR 0.001000 Time 0.021047 -2022-12-06 10:55:58,920 - Epoch: [75][ 560/ 1200] Overall Loss 0.265612 Objective Loss 0.265612 LR 0.001000 Time 0.021024 -2022-12-06 10:55:59,116 - Epoch: [75][ 570/ 1200] Overall Loss 0.265936 Objective Loss 0.265936 LR 0.001000 Time 0.020998 -2022-12-06 10:55:59,314 - Epoch: [75][ 580/ 1200] Overall Loss 0.266093 Objective Loss 0.266093 LR 0.001000 Time 0.020976 -2022-12-06 10:55:59,510 - Epoch: [75][ 590/ 1200] Overall Loss 0.265882 Objective Loss 0.265882 LR 0.001000 Time 0.020952 -2022-12-06 10:55:59,709 - Epoch: [75][ 600/ 1200] Overall Loss 0.265834 Objective Loss 0.265834 LR 0.001000 Time 0.020933 -2022-12-06 10:55:59,906 - Epoch: [75][ 610/ 1200] Overall Loss 0.265608 Objective Loss 0.265608 LR 0.001000 Time 0.020911 -2022-12-06 10:56:00,104 - Epoch: [75][ 620/ 1200] Overall Loss 0.265611 Objective Loss 0.265611 LR 0.001000 Time 0.020894 -2022-12-06 10:56:00,300 - Epoch: [75][ 630/ 1200] Overall Loss 0.265810 Objective Loss 0.265810 LR 0.001000 Time 0.020872 -2022-12-06 10:56:00,499 - Epoch: [75][ 640/ 1200] Overall Loss 0.265672 Objective Loss 0.265672 LR 0.001000 Time 0.020855 -2022-12-06 10:56:00,695 - Epoch: [75][ 650/ 1200] Overall Loss 0.265829 Objective Loss 0.265829 LR 0.001000 Time 0.020835 -2022-12-06 10:56:00,894 - Epoch: [75][ 660/ 1200] Overall Loss 0.265959 Objective Loss 0.265959 LR 0.001000 Time 0.020820 -2022-12-06 10:56:01,089 - Epoch: [75][ 670/ 1200] Overall Loss 0.266492 Objective Loss 0.266492 LR 0.001000 Time 0.020800 -2022-12-06 10:56:01,288 - Epoch: [75][ 680/ 1200] Overall Loss 0.266542 Objective Loss 0.266542 LR 0.001000 Time 0.020786 -2022-12-06 10:56:01,485 - Epoch: [75][ 690/ 1200] Overall Loss 0.266566 Objective Loss 0.266566 LR 0.001000 Time 0.020769 -2022-12-06 10:56:01,683 - Epoch: [75][ 700/ 1200] Overall Loss 0.266732 Objective Loss 0.266732 LR 0.001000 Time 0.020755 -2022-12-06 10:56:01,880 - Epoch: [75][ 710/ 1200] Overall Loss 0.267198 Objective Loss 0.267198 LR 0.001000 Time 0.020739 -2022-12-06 10:56:02,078 - Epoch: [75][ 720/ 1200] Overall Loss 0.267281 Objective Loss 0.267281 LR 0.001000 Time 0.020725 -2022-12-06 10:56:02,273 - Epoch: [75][ 730/ 1200] Overall Loss 0.267259 Objective Loss 0.267259 LR 0.001000 Time 0.020708 -2022-12-06 10:56:02,472 - Epoch: [75][ 740/ 1200] Overall Loss 0.267466 Objective Loss 0.267466 LR 0.001000 Time 0.020696 -2022-12-06 10:56:02,668 - Epoch: [75][ 750/ 1200] Overall Loss 0.267245 Objective Loss 0.267245 LR 0.001000 Time 0.020681 -2022-12-06 10:56:02,866 - Epoch: [75][ 760/ 1200] Overall Loss 0.267275 Objective Loss 0.267275 LR 0.001000 Time 0.020668 -2022-12-06 10:56:03,062 - Epoch: [75][ 770/ 1200] Overall Loss 0.267132 Objective Loss 0.267132 LR 0.001000 Time 0.020654 -2022-12-06 10:56:03,260 - Epoch: [75][ 780/ 1200] Overall Loss 0.266898 Objective Loss 0.266898 LR 0.001000 Time 0.020643 -2022-12-06 10:56:03,457 - Epoch: [75][ 790/ 1200] Overall Loss 0.266664 Objective Loss 0.266664 LR 0.001000 Time 0.020629 -2022-12-06 10:56:03,655 - Epoch: [75][ 800/ 1200] Overall Loss 0.266490 Objective Loss 0.266490 LR 0.001000 Time 0.020619 -2022-12-06 10:56:03,852 - Epoch: [75][ 810/ 1200] Overall Loss 0.266210 Objective Loss 0.266210 LR 0.001000 Time 0.020606 -2022-12-06 10:56:04,050 - Epoch: [75][ 820/ 1200] Overall Loss 0.266339 Objective Loss 0.266339 LR 0.001000 Time 0.020596 -2022-12-06 10:56:04,247 - Epoch: [75][ 830/ 1200] Overall Loss 0.266525 Objective Loss 0.266525 LR 0.001000 Time 0.020584 -2022-12-06 10:56:04,445 - Epoch: [75][ 840/ 1200] Overall Loss 0.266837 Objective Loss 0.266837 LR 0.001000 Time 0.020575 -2022-12-06 10:56:04,641 - Epoch: [75][ 850/ 1200] Overall Loss 0.266700 Objective Loss 0.266700 LR 0.001000 Time 0.020562 -2022-12-06 10:56:04,840 - Epoch: [75][ 860/ 1200] Overall Loss 0.266895 Objective Loss 0.266895 LR 0.001000 Time 0.020554 -2022-12-06 10:56:05,036 - Epoch: [75][ 870/ 1200] Overall Loss 0.266819 Objective Loss 0.266819 LR 0.001000 Time 0.020543 -2022-12-06 10:56:05,235 - Epoch: [75][ 880/ 1200] Overall Loss 0.267071 Objective Loss 0.267071 LR 0.001000 Time 0.020534 -2022-12-06 10:56:05,432 - Epoch: [75][ 890/ 1200] Overall Loss 0.267150 Objective Loss 0.267150 LR 0.001000 Time 0.020524 -2022-12-06 10:56:05,630 - Epoch: [75][ 900/ 1200] Overall Loss 0.267456 Objective Loss 0.267456 LR 0.001000 Time 0.020516 -2022-12-06 10:56:05,827 - Epoch: [75][ 910/ 1200] Overall Loss 0.267422 Objective Loss 0.267422 LR 0.001000 Time 0.020507 -2022-12-06 10:56:06,026 - Epoch: [75][ 920/ 1200] Overall Loss 0.267580 Objective Loss 0.267580 LR 0.001000 Time 0.020499 -2022-12-06 10:56:06,223 - Epoch: [75][ 930/ 1200] Overall Loss 0.267772 Objective Loss 0.267772 LR 0.001000 Time 0.020489 -2022-12-06 10:56:06,421 - Epoch: [75][ 940/ 1200] Overall Loss 0.268113 Objective Loss 0.268113 LR 0.001000 Time 0.020482 -2022-12-06 10:56:06,617 - Epoch: [75][ 950/ 1200] Overall Loss 0.268236 Objective Loss 0.268236 LR 0.001000 Time 0.020472 -2022-12-06 10:56:06,816 - Epoch: [75][ 960/ 1200] Overall Loss 0.268312 Objective Loss 0.268312 LR 0.001000 Time 0.020465 -2022-12-06 10:56:07,012 - Epoch: [75][ 970/ 1200] Overall Loss 0.268348 Objective Loss 0.268348 LR 0.001000 Time 0.020456 -2022-12-06 10:56:07,211 - Epoch: [75][ 980/ 1200] Overall Loss 0.268112 Objective Loss 0.268112 LR 0.001000 Time 0.020450 -2022-12-06 10:56:07,408 - Epoch: [75][ 990/ 1200] Overall Loss 0.268058 Objective Loss 0.268058 LR 0.001000 Time 0.020442 -2022-12-06 10:56:07,607 - Epoch: [75][ 1000/ 1200] Overall Loss 0.268071 Objective Loss 0.268071 LR 0.001000 Time 0.020435 -2022-12-06 10:56:07,803 - Epoch: [75][ 1010/ 1200] Overall Loss 0.267900 Objective Loss 0.267900 LR 0.001000 Time 0.020427 -2022-12-06 10:56:08,001 - Epoch: [75][ 1020/ 1200] Overall Loss 0.267654 Objective Loss 0.267654 LR 0.001000 Time 0.020421 -2022-12-06 10:56:08,197 - Epoch: [75][ 1030/ 1200] Overall Loss 0.267746 Objective Loss 0.267746 LR 0.001000 Time 0.020412 -2022-12-06 10:56:08,396 - Epoch: [75][ 1040/ 1200] Overall Loss 0.268064 Objective Loss 0.268064 LR 0.001000 Time 0.020406 -2022-12-06 10:56:08,592 - Epoch: [75][ 1050/ 1200] Overall Loss 0.268275 Objective Loss 0.268275 LR 0.001000 Time 0.020398 -2022-12-06 10:56:08,790 - Epoch: [75][ 1060/ 1200] Overall Loss 0.268131 Objective Loss 0.268131 LR 0.001000 Time 0.020391 -2022-12-06 10:56:08,985 - Epoch: [75][ 1070/ 1200] Overall Loss 0.268324 Objective Loss 0.268324 LR 0.001000 Time 0.020383 -2022-12-06 10:56:09,184 - Epoch: [75][ 1080/ 1200] Overall Loss 0.268319 Objective Loss 0.268319 LR 0.001000 Time 0.020378 -2022-12-06 10:56:09,381 - Epoch: [75][ 1090/ 1200] Overall Loss 0.268384 Objective Loss 0.268384 LR 0.001000 Time 0.020371 -2022-12-06 10:56:09,580 - Epoch: [75][ 1100/ 1200] Overall Loss 0.268284 Objective Loss 0.268284 LR 0.001000 Time 0.020366 -2022-12-06 10:56:09,776 - Epoch: [75][ 1110/ 1200] Overall Loss 0.268400 Objective Loss 0.268400 LR 0.001000 Time 0.020359 -2022-12-06 10:56:09,974 - Epoch: [75][ 1120/ 1200] Overall Loss 0.268107 Objective Loss 0.268107 LR 0.001000 Time 0.020353 -2022-12-06 10:56:10,171 - Epoch: [75][ 1130/ 1200] Overall Loss 0.267945 Objective Loss 0.267945 LR 0.001000 Time 0.020347 -2022-12-06 10:56:10,369 - Epoch: [75][ 1140/ 1200] Overall Loss 0.268185 Objective Loss 0.268185 LR 0.001000 Time 0.020342 -2022-12-06 10:56:10,565 - Epoch: [75][ 1150/ 1200] Overall Loss 0.268272 Objective Loss 0.268272 LR 0.001000 Time 0.020335 -2022-12-06 10:56:10,763 - Epoch: [75][ 1160/ 1200] Overall Loss 0.268311 Objective Loss 0.268311 LR 0.001000 Time 0.020330 -2022-12-06 10:56:10,960 - Epoch: [75][ 1170/ 1200] Overall Loss 0.268203 Objective Loss 0.268203 LR 0.001000 Time 0.020324 -2022-12-06 10:56:11,158 - Epoch: [75][ 1180/ 1200] Overall Loss 0.268367 Objective Loss 0.268367 LR 0.001000 Time 0.020319 -2022-12-06 10:56:11,355 - Epoch: [75][ 1190/ 1200] Overall Loss 0.268424 Objective Loss 0.268424 LR 0.001000 Time 0.020313 -2022-12-06 10:56:11,589 - Epoch: [75][ 1200/ 1200] Overall Loss 0.268641 Objective Loss 0.268641 Top1 86.401674 Top5 98.953975 LR 0.001000 Time 0.020339 -2022-12-06 10:56:11,678 - --- validate (epoch=75)----------- -2022-12-06 10:56:11,678 - 34129 samples (256 per mini-batch) -2022-12-06 10:56:12,122 - Epoch: [75][ 10/ 134] Loss 0.313603 Top1 83.828125 Top5 98.085938 -2022-12-06 10:56:12,250 - Epoch: [75][ 20/ 134] Loss 0.312224 Top1 84.609375 Top5 98.164062 -2022-12-06 10:56:12,376 - Epoch: [75][ 30/ 134] Loss 0.301775 Top1 84.596354 Top5 98.151042 -2022-12-06 10:56:12,501 - Epoch: [75][ 40/ 134] Loss 0.296533 Top1 85.097656 Top5 98.193359 -2022-12-06 10:56:12,634 - Epoch: [75][ 50/ 134] Loss 0.300107 Top1 85.312500 Top5 98.234375 -2022-12-06 10:56:12,780 - Epoch: [75][ 60/ 134] Loss 0.302317 Top1 85.188802 Top5 98.092448 -2022-12-06 10:56:12,919 - Epoch: [75][ 70/ 134] Loss 0.297552 Top1 85.362723 Top5 98.074777 -2022-12-06 10:56:13,064 - Epoch: [75][ 80/ 134] Loss 0.296674 Top1 85.395508 Top5 98.071289 -2022-12-06 10:56:13,202 - Epoch: [75][ 90/ 134] Loss 0.296726 Top1 85.394965 Top5 98.003472 -2022-12-06 10:56:13,346 - Epoch: [75][ 100/ 134] Loss 0.297229 Top1 85.414062 Top5 97.988281 -2022-12-06 10:56:13,486 - Epoch: [75][ 110/ 134] Loss 0.295796 Top1 85.358665 Top5 97.997159 -2022-12-06 10:56:13,633 - Epoch: [75][ 120/ 134] Loss 0.297669 Top1 85.361328 Top5 97.968750 -2022-12-06 10:56:13,768 - Epoch: [75][ 130/ 134] Loss 0.299017 Top1 85.261418 Top5 97.959736 -2022-12-06 10:56:13,805 - Epoch: [75][ 134/ 134] Loss 0.299996 Top1 85.217850 Top5 97.951888 -2022-12-06 10:56:13,895 - ==> Top1: 85.218 Top5: 97.952 Loss: 0.300 - -2022-12-06 10:56:13,895 - ==> Confusion: -[[ 914 1 1 5 4 4 0 0 5 50 0 2 0 4 1 1 2 0 1 0 1] - [ 2 897 1 4 13 35 5 11 1 1 5 6 0 2 4 0 8 3 13 5 11] - [ 6 6 1003 16 4 0 17 9 0 0 4 5 0 2 2 6 4 4 3 3 9] - [ 4 2 17 926 6 1 1 2 1 0 15 1 8 0 16 0 0 0 10 0 10] - [ 10 4 1 0 944 4 0 2 0 8 3 4 0 3 9 10 7 2 0 3 6] - [ 5 9 2 3 9 958 2 13 3 2 3 14 1 17 0 1 2 3 1 10 11] - [ 1 1 14 3 1 5 1056 5 0 0 0 3 1 1 0 11 2 1 1 11 1] - [ 2 9 9 3 2 35 5 923 0 0 1 5 1 3 0 2 0 2 25 17 10] - [ 7 3 1 1 2 3 0 0 953 45 11 1 1 15 13 1 1 0 3 1 2] - [ 65 0 1 1 7 2 0 1 25 871 1 2 1 12 3 2 0 1 1 2 3] - [ 0 2 6 10 1 2 0 1 8 0 954 3 1 18 3 0 0 0 2 2 6] - [ 5 1 3 0 0 9 1 2 5 1 0 963 13 11 0 9 7 2 1 13 5] - [ 0 1 1 6 1 2 0 1 1 0 2 53 858 1 2 9 1 8 1 7 14] - [ 2 1 2 0 0 14 0 2 4 15 6 7 3 950 0 2 3 0 0 3 9] - [ 14 2 2 7 6 5 0 1 23 1 2 2 4 5 1040 1 3 0 3 2 7] - [ 2 1 3 0 2 1 1 0 1 1 1 5 9 3 0 988 9 11 0 2 3] - [ 5 2 2 1 5 1 1 0 0 1 0 1 2 2 1 16 1022 1 0 5 4] - [ 5 0 3 0 1 3 3 1 3 2 0 11 18 2 1 23 1 956 0 1 2] - [ 1 4 8 12 6 2 0 27 4 1 8 1 3 1 12 1 2 0 899 2 14] - [ 3 1 3 2 1 3 8 6 0 0 1 13 5 8 0 3 4 3 1 1006 9] - [ 181 162 190 90 140 178 74 132 76 97 182 134 308 357 125 145 237 60 109 252 9997]] - -2022-12-06 10:56:14,572 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:56:14,572 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:56:14,578 - - -2022-12-06 10:56:14,578 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:56:15,526 - Epoch: [76][ 10/ 1200] Overall Loss 0.270352 Objective Loss 0.270352 LR 0.001000 Time 0.094706 -2022-12-06 10:56:15,730 - Epoch: [76][ 20/ 1200] Overall Loss 0.268696 Objective Loss 0.268696 LR 0.001000 Time 0.057525 -2022-12-06 10:56:15,930 - Epoch: [76][ 30/ 1200] Overall Loss 0.264567 Objective Loss 0.264567 LR 0.001000 Time 0.045001 -2022-12-06 10:56:16,127 - Epoch: [76][ 40/ 1200] Overall Loss 0.262436 Objective Loss 0.262436 LR 0.001000 Time 0.038668 -2022-12-06 10:56:16,327 - Epoch: [76][ 50/ 1200] Overall Loss 0.259704 Objective Loss 0.259704 LR 0.001000 Time 0.034914 -2022-12-06 10:56:16,524 - Epoch: [76][ 60/ 1200] Overall Loss 0.261168 Objective Loss 0.261168 LR 0.001000 Time 0.032363 -2022-12-06 10:56:16,723 - Epoch: [76][ 70/ 1200] Overall Loss 0.264134 Objective Loss 0.264134 LR 0.001000 Time 0.030575 -2022-12-06 10:56:16,920 - Epoch: [76][ 80/ 1200] Overall Loss 0.262587 Objective Loss 0.262587 LR 0.001000 Time 0.029209 -2022-12-06 10:56:17,119 - Epoch: [76][ 90/ 1200] Overall Loss 0.260688 Objective Loss 0.260688 LR 0.001000 Time 0.028174 -2022-12-06 10:56:17,317 - Epoch: [76][ 100/ 1200] Overall Loss 0.259519 Objective Loss 0.259519 LR 0.001000 Time 0.027324 -2022-12-06 10:56:17,516 - Epoch: [76][ 110/ 1200] Overall Loss 0.259061 Objective Loss 0.259061 LR 0.001000 Time 0.026651 -2022-12-06 10:56:17,713 - Epoch: [76][ 120/ 1200] Overall Loss 0.258555 Objective Loss 0.258555 LR 0.001000 Time 0.026069 -2022-12-06 10:56:17,913 - Epoch: [76][ 130/ 1200] Overall Loss 0.257819 Objective Loss 0.257819 LR 0.001000 Time 0.025591 -2022-12-06 10:56:18,110 - Epoch: [76][ 140/ 1200] Overall Loss 0.258998 Objective Loss 0.258998 LR 0.001000 Time 0.025170 -2022-12-06 10:56:18,310 - Epoch: [76][ 150/ 1200] Overall Loss 0.260656 Objective Loss 0.260656 LR 0.001000 Time 0.024819 -2022-12-06 10:56:18,507 - Epoch: [76][ 160/ 1200] Overall Loss 0.260477 Objective Loss 0.260477 LR 0.001000 Time 0.024497 -2022-12-06 10:56:18,707 - Epoch: [76][ 170/ 1200] Overall Loss 0.260753 Objective Loss 0.260753 LR 0.001000 Time 0.024227 -2022-12-06 10:56:18,903 - Epoch: [76][ 180/ 1200] Overall Loss 0.260693 Objective Loss 0.260693 LR 0.001000 Time 0.023970 -2022-12-06 10:56:19,103 - Epoch: [76][ 190/ 1200] Overall Loss 0.262139 Objective Loss 0.262139 LR 0.001000 Time 0.023757 -2022-12-06 10:56:19,300 - Epoch: [76][ 200/ 1200] Overall Loss 0.261544 Objective Loss 0.261544 LR 0.001000 Time 0.023552 -2022-12-06 10:56:19,500 - Epoch: [76][ 210/ 1200] Overall Loss 0.262502 Objective Loss 0.262502 LR 0.001000 Time 0.023379 -2022-12-06 10:56:19,697 - Epoch: [76][ 220/ 1200] Overall Loss 0.262054 Objective Loss 0.262054 LR 0.001000 Time 0.023211 -2022-12-06 10:56:19,897 - Epoch: [76][ 230/ 1200] Overall Loss 0.262247 Objective Loss 0.262247 LR 0.001000 Time 0.023067 -2022-12-06 10:56:20,094 - Epoch: [76][ 240/ 1200] Overall Loss 0.262372 Objective Loss 0.262372 LR 0.001000 Time 0.022925 -2022-12-06 10:56:20,293 - Epoch: [76][ 250/ 1200] Overall Loss 0.263076 Objective Loss 0.263076 LR 0.001000 Time 0.022800 -2022-12-06 10:56:20,489 - Epoch: [76][ 260/ 1200] Overall Loss 0.264205 Objective Loss 0.264205 LR 0.001000 Time 0.022678 -2022-12-06 10:56:20,689 - Epoch: [76][ 270/ 1200] Overall Loss 0.264738 Objective Loss 0.264738 LR 0.001000 Time 0.022574 -2022-12-06 10:56:20,886 - Epoch: [76][ 280/ 1200] Overall Loss 0.265166 Objective Loss 0.265166 LR 0.001000 Time 0.022469 -2022-12-06 10:56:21,085 - Epoch: [76][ 290/ 1200] Overall Loss 0.264973 Objective Loss 0.264973 LR 0.001000 Time 0.022379 -2022-12-06 10:56:21,282 - Epoch: [76][ 300/ 1200] Overall Loss 0.264350 Objective Loss 0.264350 LR 0.001000 Time 0.022288 -2022-12-06 10:56:21,481 - Epoch: [76][ 310/ 1200] Overall Loss 0.265646 Objective Loss 0.265646 LR 0.001000 Time 0.022211 -2022-12-06 10:56:21,679 - Epoch: [76][ 320/ 1200] Overall Loss 0.265807 Objective Loss 0.265807 LR 0.001000 Time 0.022132 -2022-12-06 10:56:21,878 - Epoch: [76][ 330/ 1200] Overall Loss 0.265505 Objective Loss 0.265505 LR 0.001000 Time 0.022064 -2022-12-06 10:56:22,075 - Epoch: [76][ 340/ 1200] Overall Loss 0.264589 Objective Loss 0.264589 LR 0.001000 Time 0.021991 -2022-12-06 10:56:22,275 - Epoch: [76][ 350/ 1200] Overall Loss 0.264728 Objective Loss 0.264728 LR 0.001000 Time 0.021932 -2022-12-06 10:56:22,471 - Epoch: [76][ 360/ 1200] Overall Loss 0.264513 Objective Loss 0.264513 LR 0.001000 Time 0.021867 -2022-12-06 10:56:22,670 - Epoch: [76][ 370/ 1200] Overall Loss 0.263788 Objective Loss 0.263788 LR 0.001000 Time 0.021813 -2022-12-06 10:56:22,867 - Epoch: [76][ 380/ 1200] Overall Loss 0.263747 Objective Loss 0.263747 LR 0.001000 Time 0.021756 -2022-12-06 10:56:23,067 - Epoch: [76][ 390/ 1200] Overall Loss 0.263860 Objective Loss 0.263860 LR 0.001000 Time 0.021709 -2022-12-06 10:56:23,264 - Epoch: [76][ 400/ 1200] Overall Loss 0.265423 Objective Loss 0.265423 LR 0.001000 Time 0.021658 -2022-12-06 10:56:23,463 - Epoch: [76][ 410/ 1200] Overall Loss 0.265890 Objective Loss 0.265890 LR 0.001000 Time 0.021612 -2022-12-06 10:56:23,660 - Epoch: [76][ 420/ 1200] Overall Loss 0.265769 Objective Loss 0.265769 LR 0.001000 Time 0.021566 -2022-12-06 10:56:23,859 - Epoch: [76][ 430/ 1200] Overall Loss 0.265917 Objective Loss 0.265917 LR 0.001000 Time 0.021526 -2022-12-06 10:56:24,056 - Epoch: [76][ 440/ 1200] Overall Loss 0.266311 Objective Loss 0.266311 LR 0.001000 Time 0.021484 -2022-12-06 10:56:24,256 - Epoch: [76][ 450/ 1200] Overall Loss 0.266407 Objective Loss 0.266407 LR 0.001000 Time 0.021449 -2022-12-06 10:56:24,453 - Epoch: [76][ 460/ 1200] Overall Loss 0.266830 Objective Loss 0.266830 LR 0.001000 Time 0.021409 -2022-12-06 10:56:24,652 - Epoch: [76][ 470/ 1200] Overall Loss 0.266967 Objective Loss 0.266967 LR 0.001000 Time 0.021376 -2022-12-06 10:56:24,849 - Epoch: [76][ 480/ 1200] Overall Loss 0.266610 Objective Loss 0.266610 LR 0.001000 Time 0.021341 -2022-12-06 10:56:25,049 - Epoch: [76][ 490/ 1200] Overall Loss 0.266927 Objective Loss 0.266927 LR 0.001000 Time 0.021311 -2022-12-06 10:56:25,245 - Epoch: [76][ 500/ 1200] Overall Loss 0.266553 Objective Loss 0.266553 LR 0.001000 Time 0.021277 -2022-12-06 10:56:25,445 - Epoch: [76][ 510/ 1200] Overall Loss 0.266735 Objective Loss 0.266735 LR 0.001000 Time 0.021250 -2022-12-06 10:56:25,642 - Epoch: [76][ 520/ 1200] Overall Loss 0.266247 Objective Loss 0.266247 LR 0.001000 Time 0.021220 -2022-12-06 10:56:25,842 - Epoch: [76][ 530/ 1200] Overall Loss 0.266751 Objective Loss 0.266751 LR 0.001000 Time 0.021196 -2022-12-06 10:56:26,040 - Epoch: [76][ 540/ 1200] Overall Loss 0.267220 Objective Loss 0.267220 LR 0.001000 Time 0.021168 -2022-12-06 10:56:26,240 - Epoch: [76][ 550/ 1200] Overall Loss 0.267239 Objective Loss 0.267239 LR 0.001000 Time 0.021146 -2022-12-06 10:56:26,437 - Epoch: [76][ 560/ 1200] Overall Loss 0.267130 Objective Loss 0.267130 LR 0.001000 Time 0.021120 -2022-12-06 10:56:26,637 - Epoch: [76][ 570/ 1200] Overall Loss 0.266660 Objective Loss 0.266660 LR 0.001000 Time 0.021098 -2022-12-06 10:56:26,834 - Epoch: [76][ 580/ 1200] Overall Loss 0.266450 Objective Loss 0.266450 LR 0.001000 Time 0.021074 -2022-12-06 10:56:27,033 - Epoch: [76][ 590/ 1200] Overall Loss 0.265899 Objective Loss 0.265899 LR 0.001000 Time 0.021052 -2022-12-06 10:56:27,229 - Epoch: [76][ 600/ 1200] Overall Loss 0.265718 Objective Loss 0.265718 LR 0.001000 Time 0.021027 -2022-12-06 10:56:27,428 - Epoch: [76][ 610/ 1200] Overall Loss 0.265725 Objective Loss 0.265725 LR 0.001000 Time 0.021009 -2022-12-06 10:56:27,625 - Epoch: [76][ 620/ 1200] Overall Loss 0.265877 Objective Loss 0.265877 LR 0.001000 Time 0.020986 -2022-12-06 10:56:27,823 - Epoch: [76][ 630/ 1200] Overall Loss 0.265651 Objective Loss 0.265651 LR 0.001000 Time 0.020967 -2022-12-06 10:56:28,020 - Epoch: [76][ 640/ 1200] Overall Loss 0.265124 Objective Loss 0.265124 LR 0.001000 Time 0.020946 -2022-12-06 10:56:28,220 - Epoch: [76][ 650/ 1200] Overall Loss 0.265574 Objective Loss 0.265574 LR 0.001000 Time 0.020930 -2022-12-06 10:56:28,417 - Epoch: [76][ 660/ 1200] Overall Loss 0.265048 Objective Loss 0.265048 LR 0.001000 Time 0.020911 -2022-12-06 10:56:28,617 - Epoch: [76][ 670/ 1200] Overall Loss 0.265338 Objective Loss 0.265338 LR 0.001000 Time 0.020896 -2022-12-06 10:56:28,814 - Epoch: [76][ 680/ 1200] Overall Loss 0.265375 Objective Loss 0.265375 LR 0.001000 Time 0.020878 -2022-12-06 10:56:29,013 - Epoch: [76][ 690/ 1200] Overall Loss 0.265370 Objective Loss 0.265370 LR 0.001000 Time 0.020863 -2022-12-06 10:56:29,209 - Epoch: [76][ 700/ 1200] Overall Loss 0.265171 Objective Loss 0.265171 LR 0.001000 Time 0.020845 -2022-12-06 10:56:29,409 - Epoch: [76][ 710/ 1200] Overall Loss 0.265553 Objective Loss 0.265553 LR 0.001000 Time 0.020831 -2022-12-06 10:56:29,606 - Epoch: [76][ 720/ 1200] Overall Loss 0.265925 Objective Loss 0.265925 LR 0.001000 Time 0.020814 -2022-12-06 10:56:29,805 - Epoch: [76][ 730/ 1200] Overall Loss 0.265960 Objective Loss 0.265960 LR 0.001000 Time 0.020801 -2022-12-06 10:56:30,003 - Epoch: [76][ 740/ 1200] Overall Loss 0.266888 Objective Loss 0.266888 LR 0.001000 Time 0.020787 -2022-12-06 10:56:30,202 - Epoch: [76][ 750/ 1200] Overall Loss 0.267179 Objective Loss 0.267179 LR 0.001000 Time 0.020774 -2022-12-06 10:56:30,399 - Epoch: [76][ 760/ 1200] Overall Loss 0.267371 Objective Loss 0.267371 LR 0.001000 Time 0.020760 -2022-12-06 10:56:30,599 - Epoch: [76][ 770/ 1200] Overall Loss 0.267359 Objective Loss 0.267359 LR 0.001000 Time 0.020749 -2022-12-06 10:56:30,796 - Epoch: [76][ 780/ 1200] Overall Loss 0.267291 Objective Loss 0.267291 LR 0.001000 Time 0.020734 -2022-12-06 10:56:30,995 - Epoch: [76][ 790/ 1200] Overall Loss 0.267179 Objective Loss 0.267179 LR 0.001000 Time 0.020724 -2022-12-06 10:56:31,192 - Epoch: [76][ 800/ 1200] Overall Loss 0.266980 Objective Loss 0.266980 LR 0.001000 Time 0.020711 -2022-12-06 10:56:31,392 - Epoch: [76][ 810/ 1200] Overall Loss 0.266896 Objective Loss 0.266896 LR 0.001000 Time 0.020701 -2022-12-06 10:56:31,589 - Epoch: [76][ 820/ 1200] Overall Loss 0.266879 Objective Loss 0.266879 LR 0.001000 Time 0.020688 -2022-12-06 10:56:31,789 - Epoch: [76][ 830/ 1200] Overall Loss 0.266892 Objective Loss 0.266892 LR 0.001000 Time 0.020678 -2022-12-06 10:56:31,988 - Epoch: [76][ 840/ 1200] Overall Loss 0.266958 Objective Loss 0.266958 LR 0.001000 Time 0.020669 -2022-12-06 10:56:32,190 - Epoch: [76][ 850/ 1200] Overall Loss 0.267132 Objective Loss 0.267132 LR 0.001000 Time 0.020662 -2022-12-06 10:56:32,388 - Epoch: [76][ 860/ 1200] Overall Loss 0.267113 Objective Loss 0.267113 LR 0.001000 Time 0.020652 -2022-12-06 10:56:32,590 - Epoch: [76][ 870/ 1200] Overall Loss 0.267191 Objective Loss 0.267191 LR 0.001000 Time 0.020646 -2022-12-06 10:56:32,790 - Epoch: [76][ 880/ 1200] Overall Loss 0.267227 Objective Loss 0.267227 LR 0.001000 Time 0.020638 -2022-12-06 10:56:32,991 - Epoch: [76][ 890/ 1200] Overall Loss 0.267308 Objective Loss 0.267308 LR 0.001000 Time 0.020632 -2022-12-06 10:56:33,191 - Epoch: [76][ 900/ 1200] Overall Loss 0.267148 Objective Loss 0.267148 LR 0.001000 Time 0.020624 -2022-12-06 10:56:33,393 - Epoch: [76][ 910/ 1200] Overall Loss 0.267025 Objective Loss 0.267025 LR 0.001000 Time 0.020619 -2022-12-06 10:56:33,593 - Epoch: [76][ 920/ 1200] Overall Loss 0.266840 Objective Loss 0.266840 LR 0.001000 Time 0.020611 -2022-12-06 10:56:33,795 - Epoch: [76][ 930/ 1200] Overall Loss 0.267149 Objective Loss 0.267149 LR 0.001000 Time 0.020606 -2022-12-06 10:56:33,994 - Epoch: [76][ 940/ 1200] Overall Loss 0.267368 Objective Loss 0.267368 LR 0.001000 Time 0.020598 -2022-12-06 10:56:34,196 - Epoch: [76][ 950/ 1200] Overall Loss 0.267465 Objective Loss 0.267465 LR 0.001000 Time 0.020593 -2022-12-06 10:56:34,395 - Epoch: [76][ 960/ 1200] Overall Loss 0.267655 Objective Loss 0.267655 LR 0.001000 Time 0.020585 -2022-12-06 10:56:34,598 - Epoch: [76][ 970/ 1200] Overall Loss 0.267513 Objective Loss 0.267513 LR 0.001000 Time 0.020582 -2022-12-06 10:56:34,796 - Epoch: [76][ 980/ 1200] Overall Loss 0.267254 Objective Loss 0.267254 LR 0.001000 Time 0.020574 -2022-12-06 10:56:34,999 - Epoch: [76][ 990/ 1200] Overall Loss 0.267680 Objective Loss 0.267680 LR 0.001000 Time 0.020570 -2022-12-06 10:56:35,198 - Epoch: [76][ 1000/ 1200] Overall Loss 0.267832 Objective Loss 0.267832 LR 0.001000 Time 0.020563 -2022-12-06 10:56:35,400 - Epoch: [76][ 1010/ 1200] Overall Loss 0.267954 Objective Loss 0.267954 LR 0.001000 Time 0.020559 -2022-12-06 10:56:35,599 - Epoch: [76][ 1020/ 1200] Overall Loss 0.268167 Objective Loss 0.268167 LR 0.001000 Time 0.020552 -2022-12-06 10:56:35,801 - Epoch: [76][ 1030/ 1200] Overall Loss 0.268156 Objective Loss 0.268156 LR 0.001000 Time 0.020547 -2022-12-06 10:56:36,000 - Epoch: [76][ 1040/ 1200] Overall Loss 0.268199 Objective Loss 0.268199 LR 0.001000 Time 0.020541 -2022-12-06 10:56:36,202 - Epoch: [76][ 1050/ 1200] Overall Loss 0.268078 Objective Loss 0.268078 LR 0.001000 Time 0.020537 -2022-12-06 10:56:36,400 - Epoch: [76][ 1060/ 1200] Overall Loss 0.267908 Objective Loss 0.267908 LR 0.001000 Time 0.020530 -2022-12-06 10:56:36,602 - Epoch: [76][ 1070/ 1200] Overall Loss 0.267856 Objective Loss 0.267856 LR 0.001000 Time 0.020526 -2022-12-06 10:56:36,800 - Epoch: [76][ 1080/ 1200] Overall Loss 0.267660 Objective Loss 0.267660 LR 0.001000 Time 0.020518 -2022-12-06 10:56:37,002 - Epoch: [76][ 1090/ 1200] Overall Loss 0.267669 Objective Loss 0.267669 LR 0.001000 Time 0.020515 -2022-12-06 10:56:37,200 - Epoch: [76][ 1100/ 1200] Overall Loss 0.267736 Objective Loss 0.267736 LR 0.001000 Time 0.020508 -2022-12-06 10:56:37,402 - Epoch: [76][ 1110/ 1200] Overall Loss 0.267574 Objective Loss 0.267574 LR 0.001000 Time 0.020504 -2022-12-06 10:56:37,600 - Epoch: [76][ 1120/ 1200] Overall Loss 0.267624 Objective Loss 0.267624 LR 0.001000 Time 0.020498 -2022-12-06 10:56:37,802 - Epoch: [76][ 1130/ 1200] Overall Loss 0.267634 Objective Loss 0.267634 LR 0.001000 Time 0.020495 -2022-12-06 10:56:38,001 - Epoch: [76][ 1140/ 1200] Overall Loss 0.268127 Objective Loss 0.268127 LR 0.001000 Time 0.020489 -2022-12-06 10:56:38,202 - Epoch: [76][ 1150/ 1200] Overall Loss 0.268304 Objective Loss 0.268304 LR 0.001000 Time 0.020485 -2022-12-06 10:56:38,401 - Epoch: [76][ 1160/ 1200] Overall Loss 0.268353 Objective Loss 0.268353 LR 0.001000 Time 0.020480 -2022-12-06 10:56:38,603 - Epoch: [76][ 1170/ 1200] Overall Loss 0.268137 Objective Loss 0.268137 LR 0.001000 Time 0.020477 -2022-12-06 10:56:38,803 - Epoch: [76][ 1180/ 1200] Overall Loss 0.268381 Objective Loss 0.268381 LR 0.001000 Time 0.020472 -2022-12-06 10:56:39,005 - Epoch: [76][ 1190/ 1200] Overall Loss 0.268421 Objective Loss 0.268421 LR 0.001000 Time 0.020470 -2022-12-06 10:56:39,231 - Epoch: [76][ 1200/ 1200] Overall Loss 0.268552 Objective Loss 0.268552 Top1 80.125523 Top5 97.280335 LR 0.001000 Time 0.020487 -2022-12-06 10:56:39,320 - --- validate (epoch=76)----------- -2022-12-06 10:56:39,320 - 34129 samples (256 per mini-batch) -2022-12-06 10:56:39,777 - Epoch: [76][ 10/ 134] Loss 0.283491 Top1 84.687500 Top5 98.281250 -2022-12-06 10:56:39,906 - Epoch: [76][ 20/ 134] Loss 0.274314 Top1 84.785156 Top5 98.164062 -2022-12-06 10:56:40,038 - Epoch: [76][ 30/ 134] Loss 0.276929 Top1 84.882812 Top5 98.085938 -2022-12-06 10:56:40,168 - Epoch: [76][ 40/ 134] Loss 0.277367 Top1 84.863281 Top5 98.144531 -2022-12-06 10:56:40,299 - Epoch: [76][ 50/ 134] Loss 0.281435 Top1 84.796875 Top5 98.070312 -2022-12-06 10:56:40,431 - Epoch: [76][ 60/ 134] Loss 0.288496 Top1 84.609375 Top5 97.975260 -2022-12-06 10:56:40,565 - Epoch: [76][ 70/ 134] Loss 0.293454 Top1 84.687500 Top5 97.946429 -2022-12-06 10:56:40,697 - Epoch: [76][ 80/ 134] Loss 0.296519 Top1 84.648438 Top5 97.895508 -2022-12-06 10:56:40,828 - Epoch: [76][ 90/ 134] Loss 0.297073 Top1 84.596354 Top5 97.929688 -2022-12-06 10:56:40,963 - Epoch: [76][ 100/ 134] Loss 0.293400 Top1 84.812500 Top5 97.949219 -2022-12-06 10:56:41,098 - Epoch: [76][ 110/ 134] Loss 0.294414 Top1 84.769176 Top5 97.929688 -2022-12-06 10:56:41,230 - Epoch: [76][ 120/ 134] Loss 0.294806 Top1 84.729818 Top5 97.929688 -2022-12-06 10:56:41,365 - Epoch: [76][ 130/ 134] Loss 0.294803 Top1 84.735577 Top5 97.932692 -2022-12-06 10:56:41,403 - Epoch: [76][ 134/ 134] Loss 0.295236 Top1 84.740250 Top5 97.943098 -2022-12-06 10:56:41,493 - ==> Top1: 84.740 Top5: 97.943 Loss: 0.295 - -2022-12-06 10:56:41,494 - ==> Confusion: -[[ 884 1 0 3 7 6 0 2 9 67 1 2 2 3 5 2 0 1 0 0 1] - [ 2 944 2 3 5 12 5 8 3 1 4 2 4 3 1 1 5 2 11 2 7] - [ 10 4 968 33 4 4 19 13 1 2 8 4 2 2 4 5 1 2 3 4 10] - [ 3 2 7 939 1 2 0 1 4 1 11 0 2 2 17 0 2 2 20 1 3] - [ 14 3 2 3 953 4 0 1 1 7 2 2 1 1 8 6 5 2 1 0 4] - [ 5 26 1 6 8 954 0 15 2 3 2 9 3 11 0 3 2 1 4 5 9] - [ 1 4 16 6 2 2 1050 6 0 0 4 5 0 0 1 6 0 2 1 11 1] - [ 0 12 4 4 2 35 3 924 0 0 4 9 3 2 0 1 2 0 32 12 5] - [ 6 1 0 2 0 2 0 1 985 33 12 1 2 5 7 0 1 0 2 1 3] - [ 36 1 2 1 9 3 0 1 33 892 1 3 0 10 3 0 0 1 0 0 5] - [ 3 2 1 7 0 1 0 2 16 1 956 0 2 14 3 1 1 0 4 0 5] - [ 6 4 3 1 0 11 1 3 2 1 0 955 19 9 0 7 10 3 0 12 4] - [ 1 1 1 4 0 3 1 2 0 0 0 32 874 0 1 8 2 19 3 4 13] - [ 1 1 0 1 1 9 0 4 17 18 7 4 3 944 1 3 1 1 0 2 5] - [ 9 5 0 15 3 2 0 0 28 4 7 0 1 4 1029 1 0 1 12 1 8] - [ 1 0 2 1 2 2 1 0 0 1 0 5 10 5 0 983 10 14 0 2 4] - [ 4 1 2 2 4 2 1 0 0 1 0 2 1 3 2 12 1025 2 0 2 6] - [ 2 1 1 3 1 1 0 1 1 2 1 8 18 3 0 14 1 974 0 3 1] - [ 3 7 2 12 0 3 0 19 3 1 7 1 0 2 8 0 0 1 936 2 1] - [ 3 2 3 1 1 6 5 10 2 0 3 15 7 4 0 7 5 5 1 990 10] - [ 154 265 138 162 122 179 72 138 125 129 229 113 372 278 155 102 204 92 223 213 9761]] - -2022-12-06 10:56:42,075 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:56:42,075 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:56:42,081 - - -2022-12-06 10:56:42,081 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:56:43,113 - Epoch: [77][ 10/ 1200] Overall Loss 0.265670 Objective Loss 0.265670 LR 0.001000 Time 0.103147 -2022-12-06 10:56:43,313 - Epoch: [77][ 20/ 1200] Overall Loss 0.254267 Objective Loss 0.254267 LR 0.001000 Time 0.061549 -2022-12-06 10:56:43,511 - Epoch: [77][ 30/ 1200] Overall Loss 0.256333 Objective Loss 0.256333 LR 0.001000 Time 0.047599 -2022-12-06 10:56:43,708 - Epoch: [77][ 40/ 1200] Overall Loss 0.253605 Objective Loss 0.253605 LR 0.001000 Time 0.040619 -2022-12-06 10:56:43,904 - Epoch: [77][ 50/ 1200] Overall Loss 0.256232 Objective Loss 0.256232 LR 0.001000 Time 0.036398 -2022-12-06 10:56:44,101 - Epoch: [77][ 60/ 1200] Overall Loss 0.259148 Objective Loss 0.259148 LR 0.001000 Time 0.033616 -2022-12-06 10:56:44,297 - Epoch: [77][ 70/ 1200] Overall Loss 0.259005 Objective Loss 0.259005 LR 0.001000 Time 0.031597 -2022-12-06 10:56:44,494 - Epoch: [77][ 80/ 1200] Overall Loss 0.257609 Objective Loss 0.257609 LR 0.001000 Time 0.030109 -2022-12-06 10:56:44,690 - Epoch: [77][ 90/ 1200] Overall Loss 0.255697 Objective Loss 0.255697 LR 0.001000 Time 0.028937 -2022-12-06 10:56:44,888 - Epoch: [77][ 100/ 1200] Overall Loss 0.253028 Objective Loss 0.253028 LR 0.001000 Time 0.028016 -2022-12-06 10:56:45,084 - Epoch: [77][ 110/ 1200] Overall Loss 0.252999 Objective Loss 0.252999 LR 0.001000 Time 0.027246 -2022-12-06 10:56:45,282 - Epoch: [77][ 120/ 1200] Overall Loss 0.252120 Objective Loss 0.252120 LR 0.001000 Time 0.026618 -2022-12-06 10:56:45,478 - Epoch: [77][ 130/ 1200] Overall Loss 0.252897 Objective Loss 0.252897 LR 0.001000 Time 0.026072 -2022-12-06 10:56:45,675 - Epoch: [77][ 140/ 1200] Overall Loss 0.253047 Objective Loss 0.253047 LR 0.001000 Time 0.025616 -2022-12-06 10:56:45,871 - Epoch: [77][ 150/ 1200] Overall Loss 0.255098 Objective Loss 0.255098 LR 0.001000 Time 0.025208 -2022-12-06 10:56:46,068 - Epoch: [77][ 160/ 1200] Overall Loss 0.257462 Objective Loss 0.257462 LR 0.001000 Time 0.024863 -2022-12-06 10:56:46,264 - Epoch: [77][ 170/ 1200] Overall Loss 0.259449 Objective Loss 0.259449 LR 0.001000 Time 0.024549 -2022-12-06 10:56:46,461 - Epoch: [77][ 180/ 1200] Overall Loss 0.259452 Objective Loss 0.259452 LR 0.001000 Time 0.024280 -2022-12-06 10:56:46,658 - Epoch: [77][ 190/ 1200] Overall Loss 0.259210 Objective Loss 0.259210 LR 0.001000 Time 0.024032 -2022-12-06 10:56:46,856 - Epoch: [77][ 200/ 1200] Overall Loss 0.260802 Objective Loss 0.260802 LR 0.001000 Time 0.023818 -2022-12-06 10:56:47,052 - Epoch: [77][ 210/ 1200] Overall Loss 0.260357 Objective Loss 0.260357 LR 0.001000 Time 0.023615 -2022-12-06 10:56:47,251 - Epoch: [77][ 220/ 1200] Overall Loss 0.261005 Objective Loss 0.261005 LR 0.001000 Time 0.023444 -2022-12-06 10:56:47,447 - Epoch: [77][ 230/ 1200] Overall Loss 0.260158 Objective Loss 0.260158 LR 0.001000 Time 0.023276 -2022-12-06 10:56:47,644 - Epoch: [77][ 240/ 1200] Overall Loss 0.260475 Objective Loss 0.260475 LR 0.001000 Time 0.023125 -2022-12-06 10:56:47,840 - Epoch: [77][ 250/ 1200] Overall Loss 0.260912 Objective Loss 0.260912 LR 0.001000 Time 0.022981 -2022-12-06 10:56:48,038 - Epoch: [77][ 260/ 1200] Overall Loss 0.261476 Objective Loss 0.261476 LR 0.001000 Time 0.022854 -2022-12-06 10:56:48,234 - Epoch: [77][ 270/ 1200] Overall Loss 0.263037 Objective Loss 0.263037 LR 0.001000 Time 0.022732 -2022-12-06 10:56:48,431 - Epoch: [77][ 280/ 1200] Overall Loss 0.263649 Objective Loss 0.263649 LR 0.001000 Time 0.022625 -2022-12-06 10:56:48,628 - Epoch: [77][ 290/ 1200] Overall Loss 0.263982 Objective Loss 0.263982 LR 0.001000 Time 0.022521 -2022-12-06 10:56:48,826 - Epoch: [77][ 300/ 1200] Overall Loss 0.265400 Objective Loss 0.265400 LR 0.001000 Time 0.022427 -2022-12-06 10:56:49,022 - Epoch: [77][ 310/ 1200] Overall Loss 0.266804 Objective Loss 0.266804 LR 0.001000 Time 0.022334 -2022-12-06 10:56:49,220 - Epoch: [77][ 320/ 1200] Overall Loss 0.267442 Objective Loss 0.267442 LR 0.001000 Time 0.022253 -2022-12-06 10:56:49,416 - Epoch: [77][ 330/ 1200] Overall Loss 0.266943 Objective Loss 0.266943 LR 0.001000 Time 0.022172 -2022-12-06 10:56:49,615 - Epoch: [77][ 340/ 1200] Overall Loss 0.266957 Objective Loss 0.266957 LR 0.001000 Time 0.022103 -2022-12-06 10:56:49,810 - Epoch: [77][ 350/ 1200] Overall Loss 0.267044 Objective Loss 0.267044 LR 0.001000 Time 0.022029 -2022-12-06 10:56:50,008 - Epoch: [77][ 360/ 1200] Overall Loss 0.267444 Objective Loss 0.267444 LR 0.001000 Time 0.021964 -2022-12-06 10:56:50,204 - Epoch: [77][ 370/ 1200] Overall Loss 0.267456 Objective Loss 0.267456 LR 0.001000 Time 0.021898 -2022-12-06 10:56:50,402 - Epoch: [77][ 380/ 1200] Overall Loss 0.267594 Objective Loss 0.267594 LR 0.001000 Time 0.021842 -2022-12-06 10:56:50,598 - Epoch: [77][ 390/ 1200] Overall Loss 0.268506 Objective Loss 0.268506 LR 0.001000 Time 0.021783 -2022-12-06 10:56:50,796 - Epoch: [77][ 400/ 1200] Overall Loss 0.268666 Objective Loss 0.268666 LR 0.001000 Time 0.021732 -2022-12-06 10:56:50,991 - Epoch: [77][ 410/ 1200] Overall Loss 0.268024 Objective Loss 0.268024 LR 0.001000 Time 0.021678 -2022-12-06 10:56:51,190 - Epoch: [77][ 420/ 1200] Overall Loss 0.267481 Objective Loss 0.267481 LR 0.001000 Time 0.021632 -2022-12-06 10:56:51,386 - Epoch: [77][ 430/ 1200] Overall Loss 0.267534 Objective Loss 0.267534 LR 0.001000 Time 0.021583 -2022-12-06 10:56:51,583 - Epoch: [77][ 440/ 1200] Overall Loss 0.267787 Objective Loss 0.267787 LR 0.001000 Time 0.021540 -2022-12-06 10:56:51,779 - Epoch: [77][ 450/ 1200] Overall Loss 0.268348 Objective Loss 0.268348 LR 0.001000 Time 0.021496 -2022-12-06 10:56:51,977 - Epoch: [77][ 460/ 1200] Overall Loss 0.267997 Objective Loss 0.267997 LR 0.001000 Time 0.021458 -2022-12-06 10:56:52,174 - Epoch: [77][ 470/ 1200] Overall Loss 0.267443 Objective Loss 0.267443 LR 0.001000 Time 0.021418 -2022-12-06 10:56:52,371 - Epoch: [77][ 480/ 1200] Overall Loss 0.267096 Objective Loss 0.267096 LR 0.001000 Time 0.021383 -2022-12-06 10:56:52,567 - Epoch: [77][ 490/ 1200] Overall Loss 0.267181 Objective Loss 0.267181 LR 0.001000 Time 0.021345 -2022-12-06 10:56:52,773 - Epoch: [77][ 500/ 1200] Overall Loss 0.267073 Objective Loss 0.267073 LR 0.001000 Time 0.021329 -2022-12-06 10:56:52,982 - Epoch: [77][ 510/ 1200] Overall Loss 0.267458 Objective Loss 0.267458 LR 0.001000 Time 0.021318 -2022-12-06 10:56:53,188 - Epoch: [77][ 520/ 1200] Overall Loss 0.267424 Objective Loss 0.267424 LR 0.001000 Time 0.021304 -2022-12-06 10:56:53,395 - Epoch: [77][ 530/ 1200] Overall Loss 0.267248 Objective Loss 0.267248 LR 0.001000 Time 0.021292 -2022-12-06 10:56:53,602 - Epoch: [77][ 540/ 1200] Overall Loss 0.266911 Objective Loss 0.266911 LR 0.001000 Time 0.021279 -2022-12-06 10:56:53,810 - Epoch: [77][ 550/ 1200] Overall Loss 0.266798 Objective Loss 0.266798 LR 0.001000 Time 0.021271 -2022-12-06 10:56:54,016 - Epoch: [77][ 560/ 1200] Overall Loss 0.266791 Objective Loss 0.266791 LR 0.001000 Time 0.021257 -2022-12-06 10:56:54,225 - Epoch: [77][ 570/ 1200] Overall Loss 0.266305 Objective Loss 0.266305 LR 0.001000 Time 0.021250 -2022-12-06 10:56:54,431 - Epoch: [77][ 580/ 1200] Overall Loss 0.266555 Objective Loss 0.266555 LR 0.001000 Time 0.021238 -2022-12-06 10:56:54,639 - Epoch: [77][ 590/ 1200] Overall Loss 0.266056 Objective Loss 0.266056 LR 0.001000 Time 0.021230 -2022-12-06 10:56:54,845 - Epoch: [77][ 600/ 1200] Overall Loss 0.265781 Objective Loss 0.265781 LR 0.001000 Time 0.021218 -2022-12-06 10:56:55,054 - Epoch: [77][ 610/ 1200] Overall Loss 0.265481 Objective Loss 0.265481 LR 0.001000 Time 0.021212 -2022-12-06 10:56:55,260 - Epoch: [77][ 620/ 1200] Overall Loss 0.265401 Objective Loss 0.265401 LR 0.001000 Time 0.021201 -2022-12-06 10:56:55,468 - Epoch: [77][ 630/ 1200] Overall Loss 0.265498 Objective Loss 0.265498 LR 0.001000 Time 0.021194 -2022-12-06 10:56:55,673 - Epoch: [77][ 640/ 1200] Overall Loss 0.265377 Objective Loss 0.265377 LR 0.001000 Time 0.021183 -2022-12-06 10:56:55,882 - Epoch: [77][ 650/ 1200] Overall Loss 0.265449 Objective Loss 0.265449 LR 0.001000 Time 0.021177 -2022-12-06 10:56:56,086 - Epoch: [77][ 660/ 1200] Overall Loss 0.265489 Objective Loss 0.265489 LR 0.001000 Time 0.021165 -2022-12-06 10:56:56,295 - Epoch: [77][ 670/ 1200] Overall Loss 0.265612 Objective Loss 0.265612 LR 0.001000 Time 0.021159 -2022-12-06 10:56:56,501 - Epoch: [77][ 680/ 1200] Overall Loss 0.265774 Objective Loss 0.265774 LR 0.001000 Time 0.021150 -2022-12-06 10:56:56,707 - Epoch: [77][ 690/ 1200] Overall Loss 0.266077 Objective Loss 0.266077 LR 0.001000 Time 0.021142 -2022-12-06 10:56:56,909 - Epoch: [77][ 700/ 1200] Overall Loss 0.266305 Objective Loss 0.266305 LR 0.001000 Time 0.021128 -2022-12-06 10:56:57,111 - Epoch: [77][ 710/ 1200] Overall Loss 0.266355 Objective Loss 0.266355 LR 0.001000 Time 0.021114 -2022-12-06 10:56:57,313 - Epoch: [77][ 720/ 1200] Overall Loss 0.266431 Objective Loss 0.266431 LR 0.001000 Time 0.021100 -2022-12-06 10:56:57,515 - Epoch: [77][ 730/ 1200] Overall Loss 0.266195 Objective Loss 0.266195 LR 0.001000 Time 0.021087 -2022-12-06 10:56:57,716 - Epoch: [77][ 740/ 1200] Overall Loss 0.266080 Objective Loss 0.266080 LR 0.001000 Time 0.021073 -2022-12-06 10:56:57,918 - Epoch: [77][ 750/ 1200] Overall Loss 0.266441 Objective Loss 0.266441 LR 0.001000 Time 0.021061 -2022-12-06 10:56:58,120 - Epoch: [77][ 760/ 1200] Overall Loss 0.266547 Objective Loss 0.266547 LR 0.001000 Time 0.021048 -2022-12-06 10:56:58,322 - Epoch: [77][ 770/ 1200] Overall Loss 0.266629 Objective Loss 0.266629 LR 0.001000 Time 0.021036 -2022-12-06 10:56:58,524 - Epoch: [77][ 780/ 1200] Overall Loss 0.266688 Objective Loss 0.266688 LR 0.001000 Time 0.021025 -2022-12-06 10:56:58,725 - Epoch: [77][ 790/ 1200] Overall Loss 0.267089 Objective Loss 0.267089 LR 0.001000 Time 0.021013 -2022-12-06 10:56:58,928 - Epoch: [77][ 800/ 1200] Overall Loss 0.267163 Objective Loss 0.267163 LR 0.001000 Time 0.021003 -2022-12-06 10:56:59,129 - Epoch: [77][ 810/ 1200] Overall Loss 0.267437 Objective Loss 0.267437 LR 0.001000 Time 0.020991 -2022-12-06 10:56:59,331 - Epoch: [77][ 820/ 1200] Overall Loss 0.267495 Objective Loss 0.267495 LR 0.001000 Time 0.020982 -2022-12-06 10:56:59,533 - Epoch: [77][ 830/ 1200] Overall Loss 0.267848 Objective Loss 0.267848 LR 0.001000 Time 0.020971 -2022-12-06 10:56:59,735 - Epoch: [77][ 840/ 1200] Overall Loss 0.268236 Objective Loss 0.268236 LR 0.001000 Time 0.020961 -2022-12-06 10:56:59,938 - Epoch: [77][ 850/ 1200] Overall Loss 0.268345 Objective Loss 0.268345 LR 0.001000 Time 0.020953 -2022-12-06 10:57:00,140 - Epoch: [77][ 860/ 1200] Overall Loss 0.268725 Objective Loss 0.268725 LR 0.001000 Time 0.020944 -2022-12-06 10:57:00,344 - Epoch: [77][ 870/ 1200] Overall Loss 0.269040 Objective Loss 0.269040 LR 0.001000 Time 0.020936 -2022-12-06 10:57:00,547 - Epoch: [77][ 880/ 1200] Overall Loss 0.269364 Objective Loss 0.269364 LR 0.001000 Time 0.020929 -2022-12-06 10:57:00,751 - Epoch: [77][ 890/ 1200] Overall Loss 0.269711 Objective Loss 0.269711 LR 0.001000 Time 0.020922 -2022-12-06 10:57:00,955 - Epoch: [77][ 900/ 1200] Overall Loss 0.269445 Objective Loss 0.269445 LR 0.001000 Time 0.020915 -2022-12-06 10:57:01,157 - Epoch: [77][ 910/ 1200] Overall Loss 0.269406 Objective Loss 0.269406 LR 0.001000 Time 0.020908 -2022-12-06 10:57:01,361 - Epoch: [77][ 920/ 1200] Overall Loss 0.269282 Objective Loss 0.269282 LR 0.001000 Time 0.020902 -2022-12-06 10:57:01,565 - Epoch: [77][ 930/ 1200] Overall Loss 0.269341 Objective Loss 0.269341 LR 0.001000 Time 0.020895 -2022-12-06 10:57:01,769 - Epoch: [77][ 940/ 1200] Overall Loss 0.269115 Objective Loss 0.269115 LR 0.001000 Time 0.020889 -2022-12-06 10:57:01,972 - Epoch: [77][ 950/ 1200] Overall Loss 0.269365 Objective Loss 0.269365 LR 0.001000 Time 0.020882 -2022-12-06 10:57:02,176 - Epoch: [77][ 960/ 1200] Overall Loss 0.269271 Objective Loss 0.269271 LR 0.001000 Time 0.020877 -2022-12-06 10:57:02,380 - Epoch: [77][ 970/ 1200] Overall Loss 0.269356 Objective Loss 0.269356 LR 0.001000 Time 0.020871 -2022-12-06 10:57:02,584 - Epoch: [77][ 980/ 1200] Overall Loss 0.269478 Objective Loss 0.269478 LR 0.001000 Time 0.020866 -2022-12-06 10:57:02,787 - Epoch: [77][ 990/ 1200] Overall Loss 0.269320 Objective Loss 0.269320 LR 0.001000 Time 0.020860 -2022-12-06 10:57:02,990 - Epoch: [77][ 1000/ 1200] Overall Loss 0.269763 Objective Loss 0.269763 LR 0.001000 Time 0.020854 -2022-12-06 10:57:03,193 - Epoch: [77][ 1010/ 1200] Overall Loss 0.269660 Objective Loss 0.269660 LR 0.001000 Time 0.020848 -2022-12-06 10:57:03,398 - Epoch: [77][ 1020/ 1200] Overall Loss 0.269653 Objective Loss 0.269653 LR 0.001000 Time 0.020844 -2022-12-06 10:57:03,601 - Epoch: [77][ 1030/ 1200] Overall Loss 0.270070 Objective Loss 0.270070 LR 0.001000 Time 0.020838 -2022-12-06 10:57:03,805 - Epoch: [77][ 1040/ 1200] Overall Loss 0.269983 Objective Loss 0.269983 LR 0.001000 Time 0.020833 -2022-12-06 10:57:04,008 - Epoch: [77][ 1050/ 1200] Overall Loss 0.270043 Objective Loss 0.270043 LR 0.001000 Time 0.020828 -2022-12-06 10:57:04,212 - Epoch: [77][ 1060/ 1200] Overall Loss 0.270279 Objective Loss 0.270279 LR 0.001000 Time 0.020823 -2022-12-06 10:57:04,415 - Epoch: [77][ 1070/ 1200] Overall Loss 0.270352 Objective Loss 0.270352 LR 0.001000 Time 0.020818 -2022-12-06 10:57:04,619 - Epoch: [77][ 1080/ 1200] Overall Loss 0.270710 Objective Loss 0.270710 LR 0.001000 Time 0.020813 -2022-12-06 10:57:04,823 - Epoch: [77][ 1090/ 1200] Overall Loss 0.270504 Objective Loss 0.270504 LR 0.001000 Time 0.020808 -2022-12-06 10:57:05,026 - Epoch: [77][ 1100/ 1200] Overall Loss 0.270553 Objective Loss 0.270553 LR 0.001000 Time 0.020803 -2022-12-06 10:57:05,230 - Epoch: [77][ 1110/ 1200] Overall Loss 0.270898 Objective Loss 0.270898 LR 0.001000 Time 0.020799 -2022-12-06 10:57:05,434 - Epoch: [77][ 1120/ 1200] Overall Loss 0.271045 Objective Loss 0.271045 LR 0.001000 Time 0.020795 -2022-12-06 10:57:05,637 - Epoch: [77][ 1130/ 1200] Overall Loss 0.270843 Objective Loss 0.270843 LR 0.001000 Time 0.020790 -2022-12-06 10:57:05,841 - Epoch: [77][ 1140/ 1200] Overall Loss 0.270918 Objective Loss 0.270918 LR 0.001000 Time 0.020786 -2022-12-06 10:57:06,044 - Epoch: [77][ 1150/ 1200] Overall Loss 0.270907 Objective Loss 0.270907 LR 0.001000 Time 0.020782 -2022-12-06 10:57:06,248 - Epoch: [77][ 1160/ 1200] Overall Loss 0.271105 Objective Loss 0.271105 LR 0.001000 Time 0.020778 -2022-12-06 10:57:06,452 - Epoch: [77][ 1170/ 1200] Overall Loss 0.271180 Objective Loss 0.271180 LR 0.001000 Time 0.020774 -2022-12-06 10:57:06,655 - Epoch: [77][ 1180/ 1200] Overall Loss 0.271136 Objective Loss 0.271136 LR 0.001000 Time 0.020769 -2022-12-06 10:57:06,858 - Epoch: [77][ 1190/ 1200] Overall Loss 0.271089 Objective Loss 0.271089 LR 0.001000 Time 0.020766 -2022-12-06 10:57:07,085 - Epoch: [77][ 1200/ 1200] Overall Loss 0.271094 Objective Loss 0.271094 Top1 85.774059 Top5 98.535565 LR 0.001000 Time 0.020781 -2022-12-06 10:57:07,174 - --- validate (epoch=77)----------- -2022-12-06 10:57:07,174 - 34129 samples (256 per mini-batch) -2022-12-06 10:57:07,629 - Epoch: [77][ 10/ 134] Loss 0.299314 Top1 83.515625 Top5 97.656250 -2022-12-06 10:57:07,763 - Epoch: [77][ 20/ 134] Loss 0.291859 Top1 84.433594 Top5 97.695312 -2022-12-06 10:57:07,895 - Epoch: [77][ 30/ 134] Loss 0.304103 Top1 84.322917 Top5 97.786458 -2022-12-06 10:57:08,025 - Epoch: [77][ 40/ 134] Loss 0.307461 Top1 84.228516 Top5 97.675781 -2022-12-06 10:57:08,154 - Epoch: [77][ 50/ 134] Loss 0.296005 Top1 84.375000 Top5 97.820312 -2022-12-06 10:57:08,284 - Epoch: [77][ 60/ 134] Loss 0.292694 Top1 84.394531 Top5 97.910156 -2022-12-06 10:57:08,414 - Epoch: [77][ 70/ 134] Loss 0.292507 Top1 84.386161 Top5 97.885045 -2022-12-06 10:57:08,545 - Epoch: [77][ 80/ 134] Loss 0.293754 Top1 84.477539 Top5 97.900391 -2022-12-06 10:57:08,678 - Epoch: [77][ 90/ 134] Loss 0.296454 Top1 84.301215 Top5 97.842882 -2022-12-06 10:57:08,811 - Epoch: [77][ 100/ 134] Loss 0.301853 Top1 84.214844 Top5 97.820312 -2022-12-06 10:57:08,943 - Epoch: [77][ 110/ 134] Loss 0.301979 Top1 84.296875 Top5 97.833807 -2022-12-06 10:57:09,077 - Epoch: [77][ 120/ 134] Loss 0.299618 Top1 84.296875 Top5 97.845052 -2022-12-06 10:57:09,211 - Epoch: [77][ 130/ 134] Loss 0.299387 Top1 84.359976 Top5 97.836538 -2022-12-06 10:57:09,250 - Epoch: [77][ 134/ 134] Loss 0.299295 Top1 84.382783 Top5 97.846406 -2022-12-06 10:57:09,343 - ==> Top1: 84.383 Top5: 97.846 Loss: 0.299 - -2022-12-06 10:57:09,344 - ==> Confusion: -[[ 909 1 4 2 9 1 0 2 2 44 0 1 2 4 5 1 4 1 0 0 4] - [ 2 931 2 2 11 16 0 15 2 0 3 8 0 0 2 2 7 2 14 2 6] - [ 7 2 993 18 6 1 28 13 1 0 4 5 1 2 2 1 1 2 3 3 10] - [ 3 1 14 930 4 2 0 1 1 0 10 1 4 3 13 3 1 2 22 0 5] - [ 12 4 2 1 949 1 0 2 0 5 1 0 1 2 14 6 12 2 0 3 3] - [ 2 21 0 6 7 930 2 37 3 2 1 18 2 22 0 1 2 0 1 7 5] - [ 0 3 6 1 1 5 1064 6 0 0 2 2 0 4 0 11 1 1 3 7 1] - [ 1 6 3 3 2 24 6 964 0 0 2 8 0 3 1 1 2 1 16 11 0] - [ 5 3 0 3 1 1 0 3 924 61 13 2 1 16 16 1 2 0 6 2 4] - [ 77 1 2 0 5 1 0 2 12 878 2 1 0 8 2 0 0 2 2 1 5] - [ 1 1 4 8 3 1 3 6 11 0 937 3 2 14 8 0 1 0 8 0 8] - [ 3 2 1 0 0 11 2 3 3 1 0 984 8 5 0 6 7 2 0 11 2] - [ 0 0 2 6 0 7 1 1 1 0 0 70 844 1 0 8 2 12 0 4 10] - [ 2 1 0 0 1 4 0 3 5 15 9 6 1 957 3 2 3 2 1 2 6] - [ 9 3 3 16 6 2 0 1 15 2 0 2 4 5 1039 1 5 4 7 1 5] - [ 1 1 0 1 3 3 2 0 0 0 1 17 4 4 0 982 8 10 0 2 4] - [ 4 2 2 3 1 0 0 0 2 1 0 6 0 2 1 10 1031 0 0 4 3] - [ 2 0 1 3 0 1 1 2 4 3 0 25 14 2 1 22 0 952 1 1 1] - [ 2 8 3 8 2 2 1 45 0 1 5 1 1 1 8 1 1 0 913 2 3] - [ 1 3 2 2 1 7 8 12 1 0 1 18 7 6 1 3 2 2 1 996 6] - [ 142 255 185 112 113 173 90 233 75 106 164 169 294 349 169 143 253 95 187 234 9685]] - -2022-12-06 10:57:09,912 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:57:09,912 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:57:09,917 - - -2022-12-06 10:57:09,918 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:57:10,839 - Epoch: [78][ 10/ 1200] Overall Loss 0.280037 Objective Loss 0.280037 LR 0.001000 Time 0.092060 -2022-12-06 10:57:11,041 - Epoch: [78][ 20/ 1200] Overall Loss 0.276988 Objective Loss 0.276988 LR 0.001000 Time 0.056133 -2022-12-06 10:57:11,233 - Epoch: [78][ 30/ 1200] Overall Loss 0.267216 Objective Loss 0.267216 LR 0.001000 Time 0.043782 -2022-12-06 10:57:11,424 - Epoch: [78][ 40/ 1200] Overall Loss 0.262293 Objective Loss 0.262293 LR 0.001000 Time 0.037597 -2022-12-06 10:57:11,614 - Epoch: [78][ 50/ 1200] Overall Loss 0.257956 Objective Loss 0.257956 LR 0.001000 Time 0.033877 -2022-12-06 10:57:11,804 - Epoch: [78][ 60/ 1200] Overall Loss 0.258580 Objective Loss 0.258580 LR 0.001000 Time 0.031389 -2022-12-06 10:57:11,994 - Epoch: [78][ 70/ 1200] Overall Loss 0.258075 Objective Loss 0.258075 LR 0.001000 Time 0.029610 -2022-12-06 10:57:12,185 - Epoch: [78][ 80/ 1200] Overall Loss 0.258246 Objective Loss 0.258246 LR 0.001000 Time 0.028290 -2022-12-06 10:57:12,376 - Epoch: [78][ 90/ 1200] Overall Loss 0.256208 Objective Loss 0.256208 LR 0.001000 Time 0.027256 -2022-12-06 10:57:12,566 - Epoch: [78][ 100/ 1200] Overall Loss 0.259108 Objective Loss 0.259108 LR 0.001000 Time 0.026427 -2022-12-06 10:57:12,756 - Epoch: [78][ 110/ 1200] Overall Loss 0.257480 Objective Loss 0.257480 LR 0.001000 Time 0.025751 -2022-12-06 10:57:12,946 - Epoch: [78][ 120/ 1200] Overall Loss 0.257331 Objective Loss 0.257331 LR 0.001000 Time 0.025185 -2022-12-06 10:57:13,137 - Epoch: [78][ 130/ 1200] Overall Loss 0.258171 Objective Loss 0.258171 LR 0.001000 Time 0.024713 -2022-12-06 10:57:13,328 - Epoch: [78][ 140/ 1200] Overall Loss 0.260319 Objective Loss 0.260319 LR 0.001000 Time 0.024303 -2022-12-06 10:57:13,518 - Epoch: [78][ 150/ 1200] Overall Loss 0.259401 Objective Loss 0.259401 LR 0.001000 Time 0.023947 -2022-12-06 10:57:13,708 - Epoch: [78][ 160/ 1200] Overall Loss 0.259592 Objective Loss 0.259592 LR 0.001000 Time 0.023637 -2022-12-06 10:57:13,899 - Epoch: [78][ 170/ 1200] Overall Loss 0.258901 Objective Loss 0.258901 LR 0.001000 Time 0.023365 -2022-12-06 10:57:14,090 - Epoch: [78][ 180/ 1200] Overall Loss 0.258148 Objective Loss 0.258148 LR 0.001000 Time 0.023124 -2022-12-06 10:57:14,280 - Epoch: [78][ 190/ 1200] Overall Loss 0.258647 Objective Loss 0.258647 LR 0.001000 Time 0.022905 -2022-12-06 10:57:14,470 - Epoch: [78][ 200/ 1200] Overall Loss 0.259825 Objective Loss 0.259825 LR 0.001000 Time 0.022709 -2022-12-06 10:57:14,661 - Epoch: [78][ 210/ 1200] Overall Loss 0.261255 Objective Loss 0.261255 LR 0.001000 Time 0.022534 -2022-12-06 10:57:14,852 - Epoch: [78][ 220/ 1200] Overall Loss 0.259853 Objective Loss 0.259853 LR 0.001000 Time 0.022374 -2022-12-06 10:57:15,042 - Epoch: [78][ 230/ 1200] Overall Loss 0.258954 Objective Loss 0.258954 LR 0.001000 Time 0.022227 -2022-12-06 10:57:15,232 - Epoch: [78][ 240/ 1200] Overall Loss 0.259649 Objective Loss 0.259649 LR 0.001000 Time 0.022091 -2022-12-06 10:57:15,422 - Epoch: [78][ 250/ 1200] Overall Loss 0.259597 Objective Loss 0.259597 LR 0.001000 Time 0.021964 -2022-12-06 10:57:15,613 - Epoch: [78][ 260/ 1200] Overall Loss 0.259414 Objective Loss 0.259414 LR 0.001000 Time 0.021850 -2022-12-06 10:57:15,803 - Epoch: [78][ 270/ 1200] Overall Loss 0.259354 Objective Loss 0.259354 LR 0.001000 Time 0.021744 -2022-12-06 10:57:15,994 - Epoch: [78][ 280/ 1200] Overall Loss 0.260182 Objective Loss 0.260182 LR 0.001000 Time 0.021646 -2022-12-06 10:57:16,184 - Epoch: [78][ 290/ 1200] Overall Loss 0.262023 Objective Loss 0.262023 LR 0.001000 Time 0.021552 -2022-12-06 10:57:16,374 - Epoch: [78][ 300/ 1200] Overall Loss 0.261813 Objective Loss 0.261813 LR 0.001000 Time 0.021468 -2022-12-06 10:57:16,565 - Epoch: [78][ 310/ 1200] Overall Loss 0.261926 Objective Loss 0.261926 LR 0.001000 Time 0.021389 -2022-12-06 10:57:16,755 - Epoch: [78][ 320/ 1200] Overall Loss 0.262006 Objective Loss 0.262006 LR 0.001000 Time 0.021314 -2022-12-06 10:57:16,946 - Epoch: [78][ 330/ 1200] Overall Loss 0.263005 Objective Loss 0.263005 LR 0.001000 Time 0.021242 -2022-12-06 10:57:17,136 - Epoch: [78][ 340/ 1200] Overall Loss 0.263790 Objective Loss 0.263790 LR 0.001000 Time 0.021176 -2022-12-06 10:57:17,326 - Epoch: [78][ 350/ 1200] Overall Loss 0.264096 Objective Loss 0.264096 LR 0.001000 Time 0.021113 -2022-12-06 10:57:17,517 - Epoch: [78][ 360/ 1200] Overall Loss 0.263325 Objective Loss 0.263325 LR 0.001000 Time 0.021055 -2022-12-06 10:57:17,708 - Epoch: [78][ 370/ 1200] Overall Loss 0.263407 Objective Loss 0.263407 LR 0.001000 Time 0.021000 -2022-12-06 10:57:17,898 - Epoch: [78][ 380/ 1200] Overall Loss 0.263879 Objective Loss 0.263879 LR 0.001000 Time 0.020946 -2022-12-06 10:57:18,088 - Epoch: [78][ 390/ 1200] Overall Loss 0.264180 Objective Loss 0.264180 LR 0.001000 Time 0.020895 -2022-12-06 10:57:18,278 - Epoch: [78][ 400/ 1200] Overall Loss 0.264536 Objective Loss 0.264536 LR 0.001000 Time 0.020848 -2022-12-06 10:57:18,469 - Epoch: [78][ 410/ 1200] Overall Loss 0.264595 Objective Loss 0.264595 LR 0.001000 Time 0.020803 -2022-12-06 10:57:18,660 - Epoch: [78][ 420/ 1200] Overall Loss 0.264779 Objective Loss 0.264779 LR 0.001000 Time 0.020760 -2022-12-06 10:57:18,850 - Epoch: [78][ 430/ 1200] Overall Loss 0.264838 Objective Loss 0.264838 LR 0.001000 Time 0.020718 -2022-12-06 10:57:19,040 - Epoch: [78][ 440/ 1200] Overall Loss 0.265197 Objective Loss 0.265197 LR 0.001000 Time 0.020679 -2022-12-06 10:57:19,230 - Epoch: [78][ 450/ 1200] Overall Loss 0.264743 Objective Loss 0.264743 LR 0.001000 Time 0.020640 -2022-12-06 10:57:19,421 - Epoch: [78][ 460/ 1200] Overall Loss 0.265308 Objective Loss 0.265308 LR 0.001000 Time 0.020604 -2022-12-06 10:57:19,611 - Epoch: [78][ 470/ 1200] Overall Loss 0.265508 Objective Loss 0.265508 LR 0.001000 Time 0.020570 -2022-12-06 10:57:19,801 - Epoch: [78][ 480/ 1200] Overall Loss 0.265558 Objective Loss 0.265558 LR 0.001000 Time 0.020537 -2022-12-06 10:57:19,992 - Epoch: [78][ 490/ 1200] Overall Loss 0.265795 Objective Loss 0.265795 LR 0.001000 Time 0.020505 -2022-12-06 10:57:20,182 - Epoch: [78][ 500/ 1200] Overall Loss 0.265958 Objective Loss 0.265958 LR 0.001000 Time 0.020475 -2022-12-06 10:57:20,372 - Epoch: [78][ 510/ 1200] Overall Loss 0.265345 Objective Loss 0.265345 LR 0.001000 Time 0.020445 -2022-12-06 10:57:20,563 - Epoch: [78][ 520/ 1200] Overall Loss 0.265633 Objective Loss 0.265633 LR 0.001000 Time 0.020417 -2022-12-06 10:57:20,753 - Epoch: [78][ 530/ 1200] Overall Loss 0.266252 Objective Loss 0.266252 LR 0.001000 Time 0.020390 -2022-12-06 10:57:20,944 - Epoch: [78][ 540/ 1200] Overall Loss 0.266712 Objective Loss 0.266712 LR 0.001000 Time 0.020364 -2022-12-06 10:57:21,134 - Epoch: [78][ 550/ 1200] Overall Loss 0.267609 Objective Loss 0.267609 LR 0.001000 Time 0.020339 -2022-12-06 10:57:21,324 - Epoch: [78][ 560/ 1200] Overall Loss 0.266977 Objective Loss 0.266977 LR 0.001000 Time 0.020314 -2022-12-06 10:57:21,515 - Epoch: [78][ 570/ 1200] Overall Loss 0.267149 Objective Loss 0.267149 LR 0.001000 Time 0.020292 -2022-12-06 10:57:21,705 - Epoch: [78][ 580/ 1200] Overall Loss 0.266952 Objective Loss 0.266952 LR 0.001000 Time 0.020270 -2022-12-06 10:57:21,896 - Epoch: [78][ 590/ 1200] Overall Loss 0.267073 Objective Loss 0.267073 LR 0.001000 Time 0.020248 -2022-12-06 10:57:22,086 - Epoch: [78][ 600/ 1200] Overall Loss 0.267598 Objective Loss 0.267598 LR 0.001000 Time 0.020226 -2022-12-06 10:57:22,277 - Epoch: [78][ 610/ 1200] Overall Loss 0.267834 Objective Loss 0.267834 LR 0.001000 Time 0.020207 -2022-12-06 10:57:22,467 - Epoch: [78][ 620/ 1200] Overall Loss 0.267973 Objective Loss 0.267973 LR 0.001000 Time 0.020187 -2022-12-06 10:57:22,658 - Epoch: [78][ 630/ 1200] Overall Loss 0.268076 Objective Loss 0.268076 LR 0.001000 Time 0.020169 -2022-12-06 10:57:22,848 - Epoch: [78][ 640/ 1200] Overall Loss 0.268790 Objective Loss 0.268790 LR 0.001000 Time 0.020150 -2022-12-06 10:57:23,039 - Epoch: [78][ 650/ 1200] Overall Loss 0.268530 Objective Loss 0.268530 LR 0.001000 Time 0.020133 -2022-12-06 10:57:23,230 - Epoch: [78][ 660/ 1200] Overall Loss 0.268664 Objective Loss 0.268664 LR 0.001000 Time 0.020116 -2022-12-06 10:57:23,420 - Epoch: [78][ 670/ 1200] Overall Loss 0.268538 Objective Loss 0.268538 LR 0.001000 Time 0.020098 -2022-12-06 10:57:23,611 - Epoch: [78][ 680/ 1200] Overall Loss 0.268460 Objective Loss 0.268460 LR 0.001000 Time 0.020083 -2022-12-06 10:57:23,801 - Epoch: [78][ 690/ 1200] Overall Loss 0.268575 Objective Loss 0.268575 LR 0.001000 Time 0.020067 -2022-12-06 10:57:23,992 - Epoch: [78][ 700/ 1200] Overall Loss 0.268865 Objective Loss 0.268865 LR 0.001000 Time 0.020051 -2022-12-06 10:57:24,181 - Epoch: [78][ 710/ 1200] Overall Loss 0.268675 Objective Loss 0.268675 LR 0.001000 Time 0.020036 -2022-12-06 10:57:24,372 - Epoch: [78][ 720/ 1200] Overall Loss 0.268634 Objective Loss 0.268634 LR 0.001000 Time 0.020021 -2022-12-06 10:57:24,562 - Epoch: [78][ 730/ 1200] Overall Loss 0.268888 Objective Loss 0.268888 LR 0.001000 Time 0.020007 -2022-12-06 10:57:24,753 - Epoch: [78][ 740/ 1200] Overall Loss 0.268717 Objective Loss 0.268717 LR 0.001000 Time 0.019993 -2022-12-06 10:57:24,943 - Epoch: [78][ 750/ 1200] Overall Loss 0.268903 Objective Loss 0.268903 LR 0.001000 Time 0.019980 -2022-12-06 10:57:25,134 - Epoch: [78][ 760/ 1200] Overall Loss 0.269075 Objective Loss 0.269075 LR 0.001000 Time 0.019967 -2022-12-06 10:57:25,324 - Epoch: [78][ 770/ 1200] Overall Loss 0.269129 Objective Loss 0.269129 LR 0.001000 Time 0.019954 -2022-12-06 10:57:25,514 - Epoch: [78][ 780/ 1200] Overall Loss 0.269451 Objective Loss 0.269451 LR 0.001000 Time 0.019941 -2022-12-06 10:57:25,704 - Epoch: [78][ 790/ 1200] Overall Loss 0.269341 Objective Loss 0.269341 LR 0.001000 Time 0.019928 -2022-12-06 10:57:25,894 - Epoch: [78][ 800/ 1200] Overall Loss 0.269946 Objective Loss 0.269946 LR 0.001000 Time 0.019916 -2022-12-06 10:57:26,084 - Epoch: [78][ 810/ 1200] Overall Loss 0.269625 Objective Loss 0.269625 LR 0.001000 Time 0.019905 -2022-12-06 10:57:26,275 - Epoch: [78][ 820/ 1200] Overall Loss 0.269626 Objective Loss 0.269626 LR 0.001000 Time 0.019894 -2022-12-06 10:57:26,466 - Epoch: [78][ 830/ 1200] Overall Loss 0.269539 Objective Loss 0.269539 LR 0.001000 Time 0.019884 -2022-12-06 10:57:26,657 - Epoch: [78][ 840/ 1200] Overall Loss 0.269522 Objective Loss 0.269522 LR 0.001000 Time 0.019874 -2022-12-06 10:57:26,848 - Epoch: [78][ 850/ 1200] Overall Loss 0.269420 Objective Loss 0.269420 LR 0.001000 Time 0.019864 -2022-12-06 10:57:27,038 - Epoch: [78][ 860/ 1200] Overall Loss 0.269046 Objective Loss 0.269046 LR 0.001000 Time 0.019854 -2022-12-06 10:57:27,229 - Epoch: [78][ 870/ 1200] Overall Loss 0.269035 Objective Loss 0.269035 LR 0.001000 Time 0.019844 -2022-12-06 10:57:27,419 - Epoch: [78][ 880/ 1200] Overall Loss 0.268995 Objective Loss 0.268995 LR 0.001000 Time 0.019834 -2022-12-06 10:57:27,610 - Epoch: [78][ 890/ 1200] Overall Loss 0.269065 Objective Loss 0.269065 LR 0.001000 Time 0.019825 -2022-12-06 10:57:27,800 - Epoch: [78][ 900/ 1200] Overall Loss 0.269007 Objective Loss 0.269007 LR 0.001000 Time 0.019815 -2022-12-06 10:57:27,990 - Epoch: [78][ 910/ 1200] Overall Loss 0.268873 Objective Loss 0.268873 LR 0.001000 Time 0.019806 -2022-12-06 10:57:28,182 - Epoch: [78][ 920/ 1200] Overall Loss 0.269112 Objective Loss 0.269112 LR 0.001000 Time 0.019798 -2022-12-06 10:57:28,372 - Epoch: [78][ 930/ 1200] Overall Loss 0.268958 Objective Loss 0.268958 LR 0.001000 Time 0.019789 -2022-12-06 10:57:28,562 - Epoch: [78][ 940/ 1200] Overall Loss 0.269321 Objective Loss 0.269321 LR 0.001000 Time 0.019780 -2022-12-06 10:57:28,752 - Epoch: [78][ 950/ 1200] Overall Loss 0.269315 Objective Loss 0.269315 LR 0.001000 Time 0.019771 -2022-12-06 10:57:28,942 - Epoch: [78][ 960/ 1200] Overall Loss 0.269569 Objective Loss 0.269569 LR 0.001000 Time 0.019763 -2022-12-06 10:57:29,133 - Epoch: [78][ 970/ 1200] Overall Loss 0.269565 Objective Loss 0.269565 LR 0.001000 Time 0.019755 -2022-12-06 10:57:29,323 - Epoch: [78][ 980/ 1200] Overall Loss 0.269981 Objective Loss 0.269981 LR 0.001000 Time 0.019747 -2022-12-06 10:57:29,514 - Epoch: [78][ 990/ 1200] Overall Loss 0.270151 Objective Loss 0.270151 LR 0.001000 Time 0.019740 -2022-12-06 10:57:29,704 - Epoch: [78][ 1000/ 1200] Overall Loss 0.270167 Objective Loss 0.270167 LR 0.001000 Time 0.019733 -2022-12-06 10:57:29,895 - Epoch: [78][ 1010/ 1200] Overall Loss 0.270527 Objective Loss 0.270527 LR 0.001000 Time 0.019726 -2022-12-06 10:57:30,086 - Epoch: [78][ 1020/ 1200] Overall Loss 0.270939 Objective Loss 0.270939 LR 0.001000 Time 0.019719 -2022-12-06 10:57:30,276 - Epoch: [78][ 1030/ 1200] Overall Loss 0.271047 Objective Loss 0.271047 LR 0.001000 Time 0.019711 -2022-12-06 10:57:30,466 - Epoch: [78][ 1040/ 1200] Overall Loss 0.271292 Objective Loss 0.271292 LR 0.001000 Time 0.019704 -2022-12-06 10:57:30,656 - Epoch: [78][ 1050/ 1200] Overall Loss 0.271352 Objective Loss 0.271352 LR 0.001000 Time 0.019697 -2022-12-06 10:57:30,847 - Epoch: [78][ 1060/ 1200] Overall Loss 0.271252 Objective Loss 0.271252 LR 0.001000 Time 0.019691 -2022-12-06 10:57:31,038 - Epoch: [78][ 1070/ 1200] Overall Loss 0.271148 Objective Loss 0.271148 LR 0.001000 Time 0.019685 -2022-12-06 10:57:31,229 - Epoch: [78][ 1080/ 1200] Overall Loss 0.271053 Objective Loss 0.271053 LR 0.001000 Time 0.019679 -2022-12-06 10:57:31,420 - Epoch: [78][ 1090/ 1200] Overall Loss 0.271100 Objective Loss 0.271100 LR 0.001000 Time 0.019673 -2022-12-06 10:57:31,610 - Epoch: [78][ 1100/ 1200] Overall Loss 0.271125 Objective Loss 0.271125 LR 0.001000 Time 0.019667 -2022-12-06 10:57:31,801 - Epoch: [78][ 1110/ 1200] Overall Loss 0.270824 Objective Loss 0.270824 LR 0.001000 Time 0.019661 -2022-12-06 10:57:31,992 - Epoch: [78][ 1120/ 1200] Overall Loss 0.270806 Objective Loss 0.270806 LR 0.001000 Time 0.019655 -2022-12-06 10:57:32,182 - Epoch: [78][ 1130/ 1200] Overall Loss 0.270743 Objective Loss 0.270743 LR 0.001000 Time 0.019649 -2022-12-06 10:57:32,373 - Epoch: [78][ 1140/ 1200] Overall Loss 0.270488 Objective Loss 0.270488 LR 0.001000 Time 0.019644 -2022-12-06 10:57:32,564 - Epoch: [78][ 1150/ 1200] Overall Loss 0.270328 Objective Loss 0.270328 LR 0.001000 Time 0.019638 -2022-12-06 10:57:32,754 - Epoch: [78][ 1160/ 1200] Overall Loss 0.270581 Objective Loss 0.270581 LR 0.001000 Time 0.019632 -2022-12-06 10:57:32,944 - Epoch: [78][ 1170/ 1200] Overall Loss 0.270324 Objective Loss 0.270324 LR 0.001000 Time 0.019627 -2022-12-06 10:57:33,134 - Epoch: [78][ 1180/ 1200] Overall Loss 0.270052 Objective Loss 0.270052 LR 0.001000 Time 0.019621 -2022-12-06 10:57:33,324 - Epoch: [78][ 1190/ 1200] Overall Loss 0.270162 Objective Loss 0.270162 LR 0.001000 Time 0.019615 -2022-12-06 10:57:33,555 - Epoch: [78][ 1200/ 1200] Overall Loss 0.270211 Objective Loss 0.270211 Top1 86.820084 Top5 97.698745 LR 0.001000 Time 0.019644 -2022-12-06 10:57:33,659 - --- validate (epoch=78)----------- -2022-12-06 10:57:33,660 - 34129 samples (256 per mini-batch) -2022-12-06 10:57:34,218 - Epoch: [78][ 10/ 134] Loss 0.325876 Top1 85.000000 Top5 97.695312 -2022-12-06 10:57:34,347 - Epoch: [78][ 20/ 134] Loss 0.308105 Top1 84.746094 Top5 98.027344 -2022-12-06 10:57:34,478 - Epoch: [78][ 30/ 134] Loss 0.299630 Top1 84.921875 Top5 98.125000 -2022-12-06 10:57:34,607 - Epoch: [78][ 40/ 134] Loss 0.292448 Top1 84.873047 Top5 98.115234 -2022-12-06 10:57:34,735 - Epoch: [78][ 50/ 134] Loss 0.294912 Top1 84.578125 Top5 98.031250 -2022-12-06 10:57:34,867 - Epoch: [78][ 60/ 134] Loss 0.304906 Top1 84.576823 Top5 97.942708 -2022-12-06 10:57:35,004 - Epoch: [78][ 70/ 134] Loss 0.303708 Top1 84.637277 Top5 97.918527 -2022-12-06 10:57:35,146 - Epoch: [78][ 80/ 134] Loss 0.301088 Top1 84.638672 Top5 97.949219 -2022-12-06 10:57:35,270 - Epoch: [78][ 90/ 134] Loss 0.298937 Top1 84.691840 Top5 97.960069 -2022-12-06 10:57:35,404 - Epoch: [78][ 100/ 134] Loss 0.297901 Top1 84.808594 Top5 97.941406 -2022-12-06 10:57:35,536 - Epoch: [78][ 110/ 134] Loss 0.297419 Top1 84.783381 Top5 97.926136 -2022-12-06 10:57:35,670 - Epoch: [78][ 120/ 134] Loss 0.297125 Top1 84.775391 Top5 97.939453 -2022-12-06 10:57:35,800 - Epoch: [78][ 130/ 134] Loss 0.297789 Top1 84.768630 Top5 97.932692 -2022-12-06 10:57:35,837 - Epoch: [78][ 134/ 134] Loss 0.298880 Top1 84.746110 Top5 97.907938 -2022-12-06 10:57:35,925 - ==> Top1: 84.746 Top5: 97.908 Loss: 0.299 - -2022-12-06 10:57:35,926 - ==> Confusion: -[[ 902 0 2 7 8 4 0 1 2 51 0 2 2 3 5 2 2 0 0 0 3] - [ 1 923 2 4 14 22 1 20 2 1 4 4 1 4 4 1 8 2 4 2 3] - [ 8 3 998 13 5 1 25 12 2 2 3 4 2 1 5 0 2 1 6 2 8] - [ 2 2 19 930 1 3 0 1 1 0 10 2 1 3 20 0 0 5 14 1 5] - [ 15 1 1 0 949 4 1 3 0 7 0 3 0 1 12 6 8 3 0 3 3] - [ 1 18 0 2 5 952 2 24 5 3 5 18 4 11 1 2 2 0 0 5 9] - [ 1 1 13 3 0 5 1060 4 1 0 5 1 1 1 0 6 0 2 3 8 3] - [ 1 18 8 3 1 32 7 920 0 2 4 5 0 0 0 1 0 1 32 13 6] - [ 6 2 2 1 0 3 0 0 954 47 12 0 2 8 18 0 3 2 3 0 1] - [ 62 1 1 0 6 4 0 3 20 880 1 2 0 15 1 0 1 0 0 1 3] - [ 1 1 7 10 2 2 0 1 7 0 943 2 2 13 7 1 2 0 8 2 8] - [ 3 1 1 0 0 15 6 5 2 2 1 953 24 1 1 9 4 7 1 10 5] - [ 3 1 2 3 1 6 0 1 0 0 0 36 874 1 1 8 2 12 1 7 10] - [ 1 0 1 0 0 12 0 4 10 19 12 2 3 941 1 1 3 1 1 2 9] - [ 8 4 2 12 7 2 0 2 20 5 0 2 1 4 1046 1 0 2 7 1 4] - [ 0 0 2 2 4 2 7 0 1 0 0 7 4 3 0 980 12 12 0 3 4] - [ 2 4 2 2 2 2 1 0 1 1 0 2 3 1 2 16 1023 0 0 4 4] - [ 2 1 2 4 0 1 2 0 2 2 0 17 19 3 3 12 1 958 2 3 2] - [ 3 2 7 14 5 4 2 20 1 1 5 1 2 1 18 0 1 0 918 1 2] - [ 0 3 6 4 1 6 6 13 0 0 1 17 7 5 0 5 2 3 2 990 9] - [ 171 198 179 130 141 196 81 151 86 115 168 93 323 320 186 107 185 93 221 259 9823]] - -2022-12-06 10:57:36,485 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:57:36,485 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:57:36,491 - - -2022-12-06 10:57:36,491 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:57:37,440 - Epoch: [79][ 10/ 1200] Overall Loss 0.252358 Objective Loss 0.252358 LR 0.001000 Time 0.094850 -2022-12-06 10:57:37,639 - Epoch: [79][ 20/ 1200] Overall Loss 0.264264 Objective Loss 0.264264 LR 0.001000 Time 0.057368 -2022-12-06 10:57:37,841 - Epoch: [79][ 30/ 1200] Overall Loss 0.269596 Objective Loss 0.269596 LR 0.001000 Time 0.044958 -2022-12-06 10:57:38,039 - Epoch: [79][ 40/ 1200] Overall Loss 0.267592 Objective Loss 0.267592 LR 0.001000 Time 0.038639 -2022-12-06 10:57:38,239 - Epoch: [79][ 50/ 1200] Overall Loss 0.265760 Objective Loss 0.265760 LR 0.001000 Time 0.034911 -2022-12-06 10:57:38,436 - Epoch: [79][ 60/ 1200] Overall Loss 0.272244 Objective Loss 0.272244 LR 0.001000 Time 0.032368 -2022-12-06 10:57:38,637 - Epoch: [79][ 70/ 1200] Overall Loss 0.267006 Objective Loss 0.267006 LR 0.001000 Time 0.030598 -2022-12-06 10:57:38,834 - Epoch: [79][ 80/ 1200] Overall Loss 0.267837 Objective Loss 0.267837 LR 0.001000 Time 0.029233 -2022-12-06 10:57:39,036 - Epoch: [79][ 90/ 1200] Overall Loss 0.269096 Objective Loss 0.269096 LR 0.001000 Time 0.028219 -2022-12-06 10:57:39,233 - Epoch: [79][ 100/ 1200] Overall Loss 0.273419 Objective Loss 0.273419 LR 0.001000 Time 0.027366 -2022-12-06 10:57:39,435 - Epoch: [79][ 110/ 1200] Overall Loss 0.272872 Objective Loss 0.272872 LR 0.001000 Time 0.026705 -2022-12-06 10:57:39,632 - Epoch: [79][ 120/ 1200] Overall Loss 0.271232 Objective Loss 0.271232 LR 0.001000 Time 0.026123 -2022-12-06 10:57:39,833 - Epoch: [79][ 130/ 1200] Overall Loss 0.270866 Objective Loss 0.270866 LR 0.001000 Time 0.025654 -2022-12-06 10:57:40,030 - Epoch: [79][ 140/ 1200] Overall Loss 0.272127 Objective Loss 0.272127 LR 0.001000 Time 0.025226 -2022-12-06 10:57:40,232 - Epoch: [79][ 150/ 1200] Overall Loss 0.272252 Objective Loss 0.272252 LR 0.001000 Time 0.024883 -2022-12-06 10:57:40,429 - Epoch: [79][ 160/ 1200] Overall Loss 0.271191 Objective Loss 0.271191 LR 0.001000 Time 0.024559 -2022-12-06 10:57:40,630 - Epoch: [79][ 170/ 1200] Overall Loss 0.271725 Objective Loss 0.271725 LR 0.001000 Time 0.024291 -2022-12-06 10:57:40,827 - Epoch: [79][ 180/ 1200] Overall Loss 0.271276 Objective Loss 0.271276 LR 0.001000 Time 0.024035 -2022-12-06 10:57:41,028 - Epoch: [79][ 190/ 1200] Overall Loss 0.271683 Objective Loss 0.271683 LR 0.001000 Time 0.023824 -2022-12-06 10:57:41,226 - Epoch: [79][ 200/ 1200] Overall Loss 0.273168 Objective Loss 0.273168 LR 0.001000 Time 0.023618 -2022-12-06 10:57:41,427 - Epoch: [79][ 210/ 1200] Overall Loss 0.273960 Objective Loss 0.273960 LR 0.001000 Time 0.023448 -2022-12-06 10:57:41,624 - Epoch: [79][ 220/ 1200] Overall Loss 0.272858 Objective Loss 0.272858 LR 0.001000 Time 0.023276 -2022-12-06 10:57:41,825 - Epoch: [79][ 230/ 1200] Overall Loss 0.273576 Objective Loss 0.273576 LR 0.001000 Time 0.023138 -2022-12-06 10:57:42,022 - Epoch: [79][ 240/ 1200] Overall Loss 0.275305 Objective Loss 0.275305 LR 0.001000 Time 0.022990 -2022-12-06 10:57:42,223 - Epoch: [79][ 250/ 1200] Overall Loss 0.275874 Objective Loss 0.275874 LR 0.001000 Time 0.022874 -2022-12-06 10:57:42,421 - Epoch: [79][ 260/ 1200] Overall Loss 0.276210 Objective Loss 0.276210 LR 0.001000 Time 0.022753 -2022-12-06 10:57:42,622 - Epoch: [79][ 270/ 1200] Overall Loss 0.276466 Objective Loss 0.276466 LR 0.001000 Time 0.022652 -2022-12-06 10:57:42,819 - Epoch: [79][ 280/ 1200] Overall Loss 0.275305 Objective Loss 0.275305 LR 0.001000 Time 0.022547 -2022-12-06 10:57:43,020 - Epoch: [79][ 290/ 1200] Overall Loss 0.274213 Objective Loss 0.274213 LR 0.001000 Time 0.022459 -2022-12-06 10:57:43,217 - Epoch: [79][ 300/ 1200] Overall Loss 0.274172 Objective Loss 0.274172 LR 0.001000 Time 0.022365 -2022-12-06 10:57:43,418 - Epoch: [79][ 310/ 1200] Overall Loss 0.274389 Objective Loss 0.274389 LR 0.001000 Time 0.022290 -2022-12-06 10:57:43,614 - Epoch: [79][ 320/ 1200] Overall Loss 0.274340 Objective Loss 0.274340 LR 0.001000 Time 0.022203 -2022-12-06 10:57:43,812 - Epoch: [79][ 330/ 1200] Overall Loss 0.273936 Objective Loss 0.273936 LR 0.001000 Time 0.022131 -2022-12-06 10:57:44,009 - Epoch: [79][ 340/ 1200] Overall Loss 0.274359 Objective Loss 0.274359 LR 0.001000 Time 0.022056 -2022-12-06 10:57:44,208 - Epoch: [79][ 350/ 1200] Overall Loss 0.274161 Objective Loss 0.274161 LR 0.001000 Time 0.021993 -2022-12-06 10:57:44,404 - Epoch: [79][ 360/ 1200] Overall Loss 0.273439 Objective Loss 0.273439 LR 0.001000 Time 0.021924 -2022-12-06 10:57:44,602 - Epoch: [79][ 370/ 1200] Overall Loss 0.273132 Objective Loss 0.273132 LR 0.001000 Time 0.021867 -2022-12-06 10:57:44,798 - Epoch: [79][ 380/ 1200] Overall Loss 0.272762 Objective Loss 0.272762 LR 0.001000 Time 0.021806 -2022-12-06 10:57:44,997 - Epoch: [79][ 390/ 1200] Overall Loss 0.272737 Objective Loss 0.272737 LR 0.001000 Time 0.021757 -2022-12-06 10:57:45,194 - Epoch: [79][ 400/ 1200] Overall Loss 0.272855 Objective Loss 0.272855 LR 0.001000 Time 0.021702 -2022-12-06 10:57:45,393 - Epoch: [79][ 410/ 1200] Overall Loss 0.272840 Objective Loss 0.272840 LR 0.001000 Time 0.021657 -2022-12-06 10:57:45,589 - Epoch: [79][ 420/ 1200] Overall Loss 0.272492 Objective Loss 0.272492 LR 0.001000 Time 0.021606 -2022-12-06 10:57:45,788 - Epoch: [79][ 430/ 1200] Overall Loss 0.272208 Objective Loss 0.272208 LR 0.001000 Time 0.021565 -2022-12-06 10:57:45,983 - Epoch: [79][ 440/ 1200] Overall Loss 0.272437 Objective Loss 0.272437 LR 0.001000 Time 0.021519 -2022-12-06 10:57:46,182 - Epoch: [79][ 450/ 1200] Overall Loss 0.272112 Objective Loss 0.272112 LR 0.001000 Time 0.021481 -2022-12-06 10:57:46,379 - Epoch: [79][ 460/ 1200] Overall Loss 0.272243 Objective Loss 0.272243 LR 0.001000 Time 0.021440 -2022-12-06 10:57:46,578 - Epoch: [79][ 470/ 1200] Overall Loss 0.272684 Objective Loss 0.272684 LR 0.001000 Time 0.021407 -2022-12-06 10:57:46,774 - Epoch: [79][ 480/ 1200] Overall Loss 0.273053 Objective Loss 0.273053 LR 0.001000 Time 0.021368 -2022-12-06 10:57:46,973 - Epoch: [79][ 490/ 1200] Overall Loss 0.272896 Objective Loss 0.272896 LR 0.001000 Time 0.021337 -2022-12-06 10:57:47,169 - Epoch: [79][ 500/ 1200] Overall Loss 0.272547 Objective Loss 0.272547 LR 0.001000 Time 0.021301 -2022-12-06 10:57:47,368 - Epoch: [79][ 510/ 1200] Overall Loss 0.272650 Objective Loss 0.272650 LR 0.001000 Time 0.021273 -2022-12-06 10:57:47,564 - Epoch: [79][ 520/ 1200] Overall Loss 0.273313 Objective Loss 0.273313 LR 0.001000 Time 0.021240 -2022-12-06 10:57:47,763 - Epoch: [79][ 530/ 1200] Overall Loss 0.273545 Objective Loss 0.273545 LR 0.001000 Time 0.021213 -2022-12-06 10:57:47,958 - Epoch: [79][ 540/ 1200] Overall Loss 0.273545 Objective Loss 0.273545 LR 0.001000 Time 0.021181 -2022-12-06 10:57:48,157 - Epoch: [79][ 550/ 1200] Overall Loss 0.273669 Objective Loss 0.273669 LR 0.001000 Time 0.021156 -2022-12-06 10:57:48,353 - Epoch: [79][ 560/ 1200] Overall Loss 0.273572 Objective Loss 0.273572 LR 0.001000 Time 0.021127 -2022-12-06 10:57:48,552 - Epoch: [79][ 570/ 1200] Overall Loss 0.273157 Objective Loss 0.273157 LR 0.001000 Time 0.021105 -2022-12-06 10:57:48,747 - Epoch: [79][ 580/ 1200] Overall Loss 0.273420 Objective Loss 0.273420 LR 0.001000 Time 0.021077 -2022-12-06 10:57:48,946 - Epoch: [79][ 590/ 1200] Overall Loss 0.273311 Objective Loss 0.273311 LR 0.001000 Time 0.021055 -2022-12-06 10:57:49,141 - Epoch: [79][ 600/ 1200] Overall Loss 0.273115 Objective Loss 0.273115 LR 0.001000 Time 0.021028 -2022-12-06 10:57:49,339 - Epoch: [79][ 610/ 1200] Overall Loss 0.272800 Objective Loss 0.272800 LR 0.001000 Time 0.021008 -2022-12-06 10:57:49,535 - Epoch: [79][ 620/ 1200] Overall Loss 0.272915 Objective Loss 0.272915 LR 0.001000 Time 0.020984 -2022-12-06 10:57:49,733 - Epoch: [79][ 630/ 1200] Overall Loss 0.273001 Objective Loss 0.273001 LR 0.001000 Time 0.020965 -2022-12-06 10:57:49,929 - Epoch: [79][ 640/ 1200] Overall Loss 0.273316 Objective Loss 0.273316 LR 0.001000 Time 0.020942 -2022-12-06 10:57:50,127 - Epoch: [79][ 650/ 1200] Overall Loss 0.273061 Objective Loss 0.273061 LR 0.001000 Time 0.020924 -2022-12-06 10:57:50,323 - Epoch: [79][ 660/ 1200] Overall Loss 0.273171 Objective Loss 0.273171 LR 0.001000 Time 0.020903 -2022-12-06 10:57:50,521 - Epoch: [79][ 670/ 1200] Overall Loss 0.273169 Objective Loss 0.273169 LR 0.001000 Time 0.020886 -2022-12-06 10:57:50,717 - Epoch: [79][ 680/ 1200] Overall Loss 0.272795 Objective Loss 0.272795 LR 0.001000 Time 0.020866 -2022-12-06 10:57:50,916 - Epoch: [79][ 690/ 1200] Overall Loss 0.272825 Objective Loss 0.272825 LR 0.001000 Time 0.020851 -2022-12-06 10:57:51,111 - Epoch: [79][ 700/ 1200] Overall Loss 0.272927 Objective Loss 0.272927 LR 0.001000 Time 0.020831 -2022-12-06 10:57:51,310 - Epoch: [79][ 710/ 1200] Overall Loss 0.272582 Objective Loss 0.272582 LR 0.001000 Time 0.020817 -2022-12-06 10:57:51,506 - Epoch: [79][ 720/ 1200] Overall Loss 0.272325 Objective Loss 0.272325 LR 0.001000 Time 0.020799 -2022-12-06 10:57:51,704 - Epoch: [79][ 730/ 1200] Overall Loss 0.272831 Objective Loss 0.272831 LR 0.001000 Time 0.020785 -2022-12-06 10:57:51,900 - Epoch: [79][ 740/ 1200] Overall Loss 0.272792 Objective Loss 0.272792 LR 0.001000 Time 0.020768 -2022-12-06 10:57:52,099 - Epoch: [79][ 750/ 1200] Overall Loss 0.272680 Objective Loss 0.272680 LR 0.001000 Time 0.020756 -2022-12-06 10:57:52,295 - Epoch: [79][ 760/ 1200] Overall Loss 0.272705 Objective Loss 0.272705 LR 0.001000 Time 0.020739 -2022-12-06 10:57:52,493 - Epoch: [79][ 770/ 1200] Overall Loss 0.272590 Objective Loss 0.272590 LR 0.001000 Time 0.020727 -2022-12-06 10:57:52,690 - Epoch: [79][ 780/ 1200] Overall Loss 0.272308 Objective Loss 0.272308 LR 0.001000 Time 0.020713 -2022-12-06 10:57:52,889 - Epoch: [79][ 790/ 1200] Overall Loss 0.272056 Objective Loss 0.272056 LR 0.001000 Time 0.020702 -2022-12-06 10:57:53,085 - Epoch: [79][ 800/ 1200] Overall Loss 0.271907 Objective Loss 0.271907 LR 0.001000 Time 0.020687 -2022-12-06 10:57:53,283 - Epoch: [79][ 810/ 1200] Overall Loss 0.271658 Objective Loss 0.271658 LR 0.001000 Time 0.020676 -2022-12-06 10:57:53,479 - Epoch: [79][ 820/ 1200] Overall Loss 0.271652 Objective Loss 0.271652 LR 0.001000 Time 0.020663 -2022-12-06 10:57:53,678 - Epoch: [79][ 830/ 1200] Overall Loss 0.271211 Objective Loss 0.271211 LR 0.001000 Time 0.020653 -2022-12-06 10:57:53,875 - Epoch: [79][ 840/ 1200] Overall Loss 0.270861 Objective Loss 0.270861 LR 0.001000 Time 0.020640 -2022-12-06 10:57:54,073 - Epoch: [79][ 850/ 1200] Overall Loss 0.270421 Objective Loss 0.270421 LR 0.001000 Time 0.020630 -2022-12-06 10:57:54,269 - Epoch: [79][ 860/ 1200] Overall Loss 0.270405 Objective Loss 0.270405 LR 0.001000 Time 0.020617 -2022-12-06 10:57:54,467 - Epoch: [79][ 870/ 1200] Overall Loss 0.270298 Objective Loss 0.270298 LR 0.001000 Time 0.020607 -2022-12-06 10:57:54,663 - Epoch: [79][ 880/ 1200] Overall Loss 0.270075 Objective Loss 0.270075 LR 0.001000 Time 0.020595 -2022-12-06 10:57:54,862 - Epoch: [79][ 890/ 1200] Overall Loss 0.269910 Objective Loss 0.269910 LR 0.001000 Time 0.020587 -2022-12-06 10:57:55,058 - Epoch: [79][ 900/ 1200] Overall Loss 0.269725 Objective Loss 0.269725 LR 0.001000 Time 0.020575 -2022-12-06 10:57:55,257 - Epoch: [79][ 910/ 1200] Overall Loss 0.269797 Objective Loss 0.269797 LR 0.001000 Time 0.020567 -2022-12-06 10:57:55,453 - Epoch: [79][ 920/ 1200] Overall Loss 0.269805 Objective Loss 0.269805 LR 0.001000 Time 0.020556 -2022-12-06 10:57:55,653 - Epoch: [79][ 930/ 1200] Overall Loss 0.269646 Objective Loss 0.269646 LR 0.001000 Time 0.020549 -2022-12-06 10:57:55,849 - Epoch: [79][ 940/ 1200] Overall Loss 0.270025 Objective Loss 0.270025 LR 0.001000 Time 0.020539 -2022-12-06 10:57:56,048 - Epoch: [79][ 950/ 1200] Overall Loss 0.269838 Objective Loss 0.269838 LR 0.001000 Time 0.020531 -2022-12-06 10:57:56,244 - Epoch: [79][ 960/ 1200] Overall Loss 0.269798 Objective Loss 0.269798 LR 0.001000 Time 0.020522 -2022-12-06 10:57:56,443 - Epoch: [79][ 970/ 1200] Overall Loss 0.269895 Objective Loss 0.269895 LR 0.001000 Time 0.020514 -2022-12-06 10:57:56,639 - Epoch: [79][ 980/ 1200] Overall Loss 0.269912 Objective Loss 0.269912 LR 0.001000 Time 0.020504 -2022-12-06 10:57:56,838 - Epoch: [79][ 990/ 1200] Overall Loss 0.269668 Objective Loss 0.269668 LR 0.001000 Time 0.020498 -2022-12-06 10:57:57,034 - Epoch: [79][ 1000/ 1200] Overall Loss 0.269283 Objective Loss 0.269283 LR 0.001000 Time 0.020488 -2022-12-06 10:57:57,233 - Epoch: [79][ 1010/ 1200] Overall Loss 0.269461 Objective Loss 0.269461 LR 0.001000 Time 0.020482 -2022-12-06 10:57:57,429 - Epoch: [79][ 1020/ 1200] Overall Loss 0.269109 Objective Loss 0.269109 LR 0.001000 Time 0.020473 -2022-12-06 10:57:57,628 - Epoch: [79][ 1030/ 1200] Overall Loss 0.269183 Objective Loss 0.269183 LR 0.001000 Time 0.020467 -2022-12-06 10:57:57,825 - Epoch: [79][ 1040/ 1200] Overall Loss 0.269243 Objective Loss 0.269243 LR 0.001000 Time 0.020458 -2022-12-06 10:57:58,023 - Epoch: [79][ 1050/ 1200] Overall Loss 0.269349 Objective Loss 0.269349 LR 0.001000 Time 0.020452 -2022-12-06 10:57:58,220 - Epoch: [79][ 1060/ 1200] Overall Loss 0.269470 Objective Loss 0.269470 LR 0.001000 Time 0.020444 -2022-12-06 10:57:58,419 - Epoch: [79][ 1070/ 1200] Overall Loss 0.269786 Objective Loss 0.269786 LR 0.001000 Time 0.020438 -2022-12-06 10:57:58,615 - Epoch: [79][ 1080/ 1200] Overall Loss 0.269749 Objective Loss 0.269749 LR 0.001000 Time 0.020430 -2022-12-06 10:57:58,814 - Epoch: [79][ 1090/ 1200] Overall Loss 0.270063 Objective Loss 0.270063 LR 0.001000 Time 0.020425 -2022-12-06 10:57:59,010 - Epoch: [79][ 1100/ 1200] Overall Loss 0.270031 Objective Loss 0.270031 LR 0.001000 Time 0.020417 -2022-12-06 10:57:59,209 - Epoch: [79][ 1110/ 1200] Overall Loss 0.269742 Objective Loss 0.269742 LR 0.001000 Time 0.020412 -2022-12-06 10:57:59,405 - Epoch: [79][ 1120/ 1200] Overall Loss 0.269654 Objective Loss 0.269654 LR 0.001000 Time 0.020404 -2022-12-06 10:57:59,604 - Epoch: [79][ 1130/ 1200] Overall Loss 0.269869 Objective Loss 0.269869 LR 0.001000 Time 0.020400 -2022-12-06 10:57:59,801 - Epoch: [79][ 1140/ 1200] Overall Loss 0.269806 Objective Loss 0.269806 LR 0.001000 Time 0.020392 -2022-12-06 10:57:59,999 - Epoch: [79][ 1150/ 1200] Overall Loss 0.269880 Objective Loss 0.269880 LR 0.001000 Time 0.020387 -2022-12-06 10:58:00,195 - Epoch: [79][ 1160/ 1200] Overall Loss 0.269905 Objective Loss 0.269905 LR 0.001000 Time 0.020380 -2022-12-06 10:58:00,395 - Epoch: [79][ 1170/ 1200] Overall Loss 0.270145 Objective Loss 0.270145 LR 0.001000 Time 0.020376 -2022-12-06 10:58:00,590 - Epoch: [79][ 1180/ 1200] Overall Loss 0.270563 Objective Loss 0.270563 LR 0.001000 Time 0.020368 -2022-12-06 10:58:00,789 - Epoch: [79][ 1190/ 1200] Overall Loss 0.270441 Objective Loss 0.270441 LR 0.001000 Time 0.020364 -2022-12-06 10:58:01,019 - Epoch: [79][ 1200/ 1200] Overall Loss 0.270485 Objective Loss 0.270485 Top1 85.355649 Top5 98.117155 LR 0.001000 Time 0.020385 -2022-12-06 10:58:01,112 - --- validate (epoch=79)----------- -2022-12-06 10:58:01,112 - 34129 samples (256 per mini-batch) -2022-12-06 10:58:01,554 - Epoch: [79][ 10/ 134] Loss 0.310056 Top1 84.609375 Top5 97.812500 -2022-12-06 10:58:01,690 - Epoch: [79][ 20/ 134] Loss 0.301596 Top1 85.058594 Top5 97.968750 -2022-12-06 10:58:01,819 - Epoch: [79][ 30/ 134] Loss 0.302256 Top1 84.947917 Top5 97.955729 -2022-12-06 10:58:01,944 - Epoch: [79][ 40/ 134] Loss 0.297584 Top1 84.873047 Top5 98.027344 -2022-12-06 10:58:02,071 - Epoch: [79][ 50/ 134] Loss 0.297395 Top1 84.906250 Top5 98.046875 -2022-12-06 10:58:02,199 - Epoch: [79][ 60/ 134] Loss 0.290548 Top1 85.058594 Top5 98.007812 -2022-12-06 10:58:02,326 - Epoch: [79][ 70/ 134] Loss 0.296830 Top1 84.877232 Top5 98.013393 -2022-12-06 10:58:02,453 - Epoch: [79][ 80/ 134] Loss 0.294188 Top1 84.965820 Top5 98.046875 -2022-12-06 10:58:02,581 - Epoch: [79][ 90/ 134] Loss 0.296521 Top1 84.787326 Top5 98.016493 -2022-12-06 10:58:02,714 - Epoch: [79][ 100/ 134] Loss 0.296499 Top1 84.796875 Top5 97.988281 -2022-12-06 10:58:02,847 - Epoch: [79][ 110/ 134] Loss 0.294562 Top1 84.911222 Top5 98.029119 -2022-12-06 10:58:02,983 - Epoch: [79][ 120/ 134] Loss 0.292982 Top1 84.954427 Top5 98.037109 -2022-12-06 10:58:03,114 - Epoch: [79][ 130/ 134] Loss 0.293963 Top1 84.957933 Top5 98.058894 -2022-12-06 10:58:03,151 - Epoch: [79][ 134/ 134] Loss 0.293305 Top1 84.983445 Top5 98.045650 -2022-12-06 10:58:03,240 - ==> Top1: 84.983 Top5: 98.046 Loss: 0.293 - -2022-12-06 10:58:03,241 - ==> Confusion: -[[ 871 0 1 2 11 4 2 2 10 68 0 3 2 4 11 1 3 0 0 1 0] - [ 1 918 1 3 10 35 5 11 3 0 3 6 2 0 1 1 4 3 7 1 12] - [ 11 1 990 14 3 4 22 9 2 1 5 8 4 2 4 2 2 4 3 1 11] - [ 1 2 19 928 2 3 1 1 3 0 7 2 6 1 18 0 1 5 10 0 10] - [ 14 4 0 0 933 5 0 3 0 7 0 3 1 3 24 5 9 2 0 0 7] - [ 2 19 0 3 6 957 2 13 3 2 1 23 5 18 3 0 2 0 1 6 3] - [ 1 3 12 6 1 0 1058 6 0 0 0 5 3 1 0 6 0 1 0 10 5] - [ 2 13 4 3 3 35 2 943 1 0 2 9 1 1 0 2 0 1 15 12 5] - [ 5 1 0 0 0 2 0 0 962 40 4 4 2 14 16 1 1 2 2 1 7] - [ 49 2 0 0 5 1 0 2 34 870 1 0 0 22 4 0 1 1 1 0 8] - [ 1 1 3 9 2 1 0 1 15 0 939 5 3 20 4 0 1 0 5 1 8] - [ 2 1 1 1 0 11 1 4 0 0 0 984 17 6 1 5 4 5 1 7 0] - [ 2 0 0 4 1 1 0 2 0 0 0 55 873 1 0 7 1 13 1 3 5] - [ 1 0 0 0 0 6 0 4 19 10 3 4 4 948 0 1 2 0 0 2 19] - [ 8 1 1 14 4 0 0 0 18 2 1 2 4 5 1058 0 1 1 3 2 5] - [ 0 0 2 0 6 1 5 0 0 0 1 14 2 5 0 987 7 7 0 2 4] - [ 2 5 3 2 2 4 1 0 0 0 0 6 1 2 1 10 1018 1 0 4 10] - [ 1 0 2 3 0 2 1 1 0 2 0 15 32 2 3 9 0 961 0 1 1] - [ 2 7 6 13 1 3 0 32 3 1 6 4 5 1 15 0 0 1 903 0 5] - [ 2 2 2 0 0 12 6 5 0 0 1 22 8 6 0 3 2 5 0 996 8] - [ 143 184 169 117 107 200 73 173 110 81 151 152 402 321 199 113 204 89 131 208 9899]] - -2022-12-06 10:58:03,905 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:58:03,906 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:58:03,911 - - -2022-12-06 10:58:03,911 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:58:04,847 - Epoch: [80][ 10/ 1200] Overall Loss 0.245021 Objective Loss 0.245021 LR 0.001000 Time 0.093504 -2022-12-06 10:58:05,047 - Epoch: [80][ 20/ 1200] Overall Loss 0.257025 Objective Loss 0.257025 LR 0.001000 Time 0.056731 -2022-12-06 10:58:05,240 - Epoch: [80][ 30/ 1200] Overall Loss 0.255034 Objective Loss 0.255034 LR 0.001000 Time 0.044241 -2022-12-06 10:58:05,433 - Epoch: [80][ 40/ 1200] Overall Loss 0.258465 Objective Loss 0.258465 LR 0.001000 Time 0.037968 -2022-12-06 10:58:05,625 - Epoch: [80][ 50/ 1200] Overall Loss 0.252580 Objective Loss 0.252580 LR 0.001000 Time 0.034213 -2022-12-06 10:58:05,817 - Epoch: [80][ 60/ 1200] Overall Loss 0.254432 Objective Loss 0.254432 LR 0.001000 Time 0.031701 -2022-12-06 10:58:06,009 - Epoch: [80][ 70/ 1200] Overall Loss 0.254672 Objective Loss 0.254672 LR 0.001000 Time 0.029914 -2022-12-06 10:58:06,201 - Epoch: [80][ 80/ 1200] Overall Loss 0.254579 Objective Loss 0.254579 LR 0.001000 Time 0.028564 -2022-12-06 10:58:06,394 - Epoch: [80][ 90/ 1200] Overall Loss 0.258577 Objective Loss 0.258577 LR 0.001000 Time 0.027522 -2022-12-06 10:58:06,585 - Epoch: [80][ 100/ 1200] Overall Loss 0.261305 Objective Loss 0.261305 LR 0.001000 Time 0.026682 -2022-12-06 10:58:06,778 - Epoch: [80][ 110/ 1200] Overall Loss 0.263475 Objective Loss 0.263475 LR 0.001000 Time 0.026001 -2022-12-06 10:58:06,970 - Epoch: [80][ 120/ 1200] Overall Loss 0.264479 Objective Loss 0.264479 LR 0.001000 Time 0.025428 -2022-12-06 10:58:07,162 - Epoch: [80][ 130/ 1200] Overall Loss 0.264237 Objective Loss 0.264237 LR 0.001000 Time 0.024945 -2022-12-06 10:58:07,354 - Epoch: [80][ 140/ 1200] Overall Loss 0.263613 Objective Loss 0.263613 LR 0.001000 Time 0.024531 -2022-12-06 10:58:07,546 - Epoch: [80][ 150/ 1200] Overall Loss 0.266062 Objective Loss 0.266062 LR 0.001000 Time 0.024173 -2022-12-06 10:58:07,738 - Epoch: [80][ 160/ 1200] Overall Loss 0.266221 Objective Loss 0.266221 LR 0.001000 Time 0.023859 -2022-12-06 10:58:07,930 - Epoch: [80][ 170/ 1200] Overall Loss 0.266325 Objective Loss 0.266325 LR 0.001000 Time 0.023585 -2022-12-06 10:58:08,122 - Epoch: [80][ 180/ 1200] Overall Loss 0.268864 Objective Loss 0.268864 LR 0.001000 Time 0.023337 -2022-12-06 10:58:08,315 - Epoch: [80][ 190/ 1200] Overall Loss 0.269404 Objective Loss 0.269404 LR 0.001000 Time 0.023119 -2022-12-06 10:58:08,506 - Epoch: [80][ 200/ 1200] Overall Loss 0.269009 Objective Loss 0.269009 LR 0.001000 Time 0.022919 -2022-12-06 10:58:08,699 - Epoch: [80][ 210/ 1200] Overall Loss 0.268891 Objective Loss 0.268891 LR 0.001000 Time 0.022741 -2022-12-06 10:58:08,891 - Epoch: [80][ 220/ 1200] Overall Loss 0.268204 Objective Loss 0.268204 LR 0.001000 Time 0.022580 -2022-12-06 10:58:09,084 - Epoch: [80][ 230/ 1200] Overall Loss 0.267556 Objective Loss 0.267556 LR 0.001000 Time 0.022434 -2022-12-06 10:58:09,276 - Epoch: [80][ 240/ 1200] Overall Loss 0.267056 Objective Loss 0.267056 LR 0.001000 Time 0.022296 -2022-12-06 10:58:09,468 - Epoch: [80][ 250/ 1200] Overall Loss 0.267503 Objective Loss 0.267503 LR 0.001000 Time 0.022170 -2022-12-06 10:58:09,660 - Epoch: [80][ 260/ 1200] Overall Loss 0.266414 Objective Loss 0.266414 LR 0.001000 Time 0.022054 -2022-12-06 10:58:09,854 - Epoch: [80][ 270/ 1200] Overall Loss 0.266413 Objective Loss 0.266413 LR 0.001000 Time 0.021956 -2022-12-06 10:58:10,048 - Epoch: [80][ 280/ 1200] Overall Loss 0.266581 Objective Loss 0.266581 LR 0.001000 Time 0.021862 -2022-12-06 10:58:10,243 - Epoch: [80][ 290/ 1200] Overall Loss 0.267000 Objective Loss 0.267000 LR 0.001000 Time 0.021777 -2022-12-06 10:58:10,437 - Epoch: [80][ 300/ 1200] Overall Loss 0.267026 Objective Loss 0.267026 LR 0.001000 Time 0.021696 -2022-12-06 10:58:10,632 - Epoch: [80][ 310/ 1200] Overall Loss 0.266885 Objective Loss 0.266885 LR 0.001000 Time 0.021623 -2022-12-06 10:58:10,825 - Epoch: [80][ 320/ 1200] Overall Loss 0.267237 Objective Loss 0.267237 LR 0.001000 Time 0.021551 -2022-12-06 10:58:11,020 - Epoch: [80][ 330/ 1200] Overall Loss 0.267175 Objective Loss 0.267175 LR 0.001000 Time 0.021486 -2022-12-06 10:58:11,214 - Epoch: [80][ 340/ 1200] Overall Loss 0.267680 Objective Loss 0.267680 LR 0.001000 Time 0.021423 -2022-12-06 10:58:11,409 - Epoch: [80][ 350/ 1200] Overall Loss 0.267615 Objective Loss 0.267615 LR 0.001000 Time 0.021367 -2022-12-06 10:58:11,603 - Epoch: [80][ 360/ 1200] Overall Loss 0.266988 Objective Loss 0.266988 LR 0.001000 Time 0.021310 -2022-12-06 10:58:11,798 - Epoch: [80][ 370/ 1200] Overall Loss 0.267124 Objective Loss 0.267124 LR 0.001000 Time 0.021259 -2022-12-06 10:58:11,992 - Epoch: [80][ 380/ 1200] Overall Loss 0.266353 Objective Loss 0.266353 LR 0.001000 Time 0.021208 -2022-12-06 10:58:12,186 - Epoch: [80][ 390/ 1200] Overall Loss 0.266639 Objective Loss 0.266639 LR 0.001000 Time 0.021162 -2022-12-06 10:58:12,380 - Epoch: [80][ 400/ 1200] Overall Loss 0.266645 Objective Loss 0.266645 LR 0.001000 Time 0.021117 -2022-12-06 10:58:12,574 - Epoch: [80][ 410/ 1200] Overall Loss 0.267005 Objective Loss 0.267005 LR 0.001000 Time 0.021074 -2022-12-06 10:58:12,768 - Epoch: [80][ 420/ 1200] Overall Loss 0.266882 Objective Loss 0.266882 LR 0.001000 Time 0.021032 -2022-12-06 10:58:12,963 - Epoch: [80][ 430/ 1200] Overall Loss 0.266669 Objective Loss 0.266669 LR 0.001000 Time 0.020995 -2022-12-06 10:58:13,156 - Epoch: [80][ 440/ 1200] Overall Loss 0.267318 Objective Loss 0.267318 LR 0.001000 Time 0.020956 -2022-12-06 10:58:13,351 - Epoch: [80][ 450/ 1200] Overall Loss 0.267232 Objective Loss 0.267232 LR 0.001000 Time 0.020921 -2022-12-06 10:58:13,544 - Epoch: [80][ 460/ 1200] Overall Loss 0.268174 Objective Loss 0.268174 LR 0.001000 Time 0.020886 -2022-12-06 10:58:13,740 - Epoch: [80][ 470/ 1200] Overall Loss 0.269011 Objective Loss 0.269011 LR 0.001000 Time 0.020857 -2022-12-06 10:58:13,934 - Epoch: [80][ 480/ 1200] Overall Loss 0.268879 Objective Loss 0.268879 LR 0.001000 Time 0.020826 -2022-12-06 10:58:14,130 - Epoch: [80][ 490/ 1200] Overall Loss 0.269221 Objective Loss 0.269221 LR 0.001000 Time 0.020799 -2022-12-06 10:58:14,324 - Epoch: [80][ 500/ 1200] Overall Loss 0.269448 Objective Loss 0.269448 LR 0.001000 Time 0.020769 -2022-12-06 10:58:14,519 - Epoch: [80][ 510/ 1200] Overall Loss 0.270218 Objective Loss 0.270218 LR 0.001000 Time 0.020744 -2022-12-06 10:58:14,713 - Epoch: [80][ 520/ 1200] Overall Loss 0.269803 Objective Loss 0.269803 LR 0.001000 Time 0.020717 -2022-12-06 10:58:14,907 - Epoch: [80][ 530/ 1200] Overall Loss 0.269975 Objective Loss 0.269975 LR 0.001000 Time 0.020692 -2022-12-06 10:58:15,101 - Epoch: [80][ 540/ 1200] Overall Loss 0.270128 Objective Loss 0.270128 LR 0.001000 Time 0.020667 -2022-12-06 10:58:15,296 - Epoch: [80][ 550/ 1200] Overall Loss 0.269857 Objective Loss 0.269857 LR 0.001000 Time 0.020645 -2022-12-06 10:58:15,490 - Epoch: [80][ 560/ 1200] Overall Loss 0.269205 Objective Loss 0.269205 LR 0.001000 Time 0.020622 -2022-12-06 10:58:15,685 - Epoch: [80][ 570/ 1200] Overall Loss 0.269426 Objective Loss 0.269426 LR 0.001000 Time 0.020601 -2022-12-06 10:58:15,880 - Epoch: [80][ 580/ 1200] Overall Loss 0.269109 Objective Loss 0.269109 LR 0.001000 Time 0.020580 -2022-12-06 10:58:16,075 - Epoch: [80][ 590/ 1200] Overall Loss 0.269176 Objective Loss 0.269176 LR 0.001000 Time 0.020561 -2022-12-06 10:58:16,269 - Epoch: [80][ 600/ 1200] Overall Loss 0.269544 Objective Loss 0.269544 LR 0.001000 Time 0.020541 -2022-12-06 10:58:16,465 - Epoch: [80][ 610/ 1200] Overall Loss 0.269301 Objective Loss 0.269301 LR 0.001000 Time 0.020524 -2022-12-06 10:58:16,659 - Epoch: [80][ 620/ 1200] Overall Loss 0.269094 Objective Loss 0.269094 LR 0.001000 Time 0.020505 -2022-12-06 10:58:16,854 - Epoch: [80][ 630/ 1200] Overall Loss 0.269063 Objective Loss 0.269063 LR 0.001000 Time 0.020488 -2022-12-06 10:58:17,048 - Epoch: [80][ 640/ 1200] Overall Loss 0.268771 Objective Loss 0.268771 LR 0.001000 Time 0.020471 -2022-12-06 10:58:17,243 - Epoch: [80][ 650/ 1200] Overall Loss 0.268902 Objective Loss 0.268902 LR 0.001000 Time 0.020455 -2022-12-06 10:58:17,437 - Epoch: [80][ 660/ 1200] Overall Loss 0.268968 Objective Loss 0.268968 LR 0.001000 Time 0.020438 -2022-12-06 10:58:17,632 - Epoch: [80][ 670/ 1200] Overall Loss 0.268977 Objective Loss 0.268977 LR 0.001000 Time 0.020424 -2022-12-06 10:58:17,826 - Epoch: [80][ 680/ 1200] Overall Loss 0.268862 Objective Loss 0.268862 LR 0.001000 Time 0.020408 -2022-12-06 10:58:18,021 - Epoch: [80][ 690/ 1200] Overall Loss 0.269293 Objective Loss 0.269293 LR 0.001000 Time 0.020394 -2022-12-06 10:58:18,215 - Epoch: [80][ 700/ 1200] Overall Loss 0.269106 Objective Loss 0.269106 LR 0.001000 Time 0.020379 -2022-12-06 10:58:18,409 - Epoch: [80][ 710/ 1200] Overall Loss 0.269299 Objective Loss 0.269299 LR 0.001000 Time 0.020364 -2022-12-06 10:58:18,603 - Epoch: [80][ 720/ 1200] Overall Loss 0.268679 Objective Loss 0.268679 LR 0.001000 Time 0.020350 -2022-12-06 10:58:18,798 - Epoch: [80][ 730/ 1200] Overall Loss 0.268584 Objective Loss 0.268584 LR 0.001000 Time 0.020338 -2022-12-06 10:58:18,992 - Epoch: [80][ 740/ 1200] Overall Loss 0.268916 Objective Loss 0.268916 LR 0.001000 Time 0.020324 -2022-12-06 10:58:19,187 - Epoch: [80][ 750/ 1200] Overall Loss 0.268622 Objective Loss 0.268622 LR 0.001000 Time 0.020312 -2022-12-06 10:58:19,381 - Epoch: [80][ 760/ 1200] Overall Loss 0.268445 Objective Loss 0.268445 LR 0.001000 Time 0.020299 -2022-12-06 10:58:19,575 - Epoch: [80][ 770/ 1200] Overall Loss 0.268584 Objective Loss 0.268584 LR 0.001000 Time 0.020287 -2022-12-06 10:58:19,769 - Epoch: [80][ 780/ 1200] Overall Loss 0.268968 Objective Loss 0.268968 LR 0.001000 Time 0.020275 -2022-12-06 10:58:19,964 - Epoch: [80][ 790/ 1200] Overall Loss 0.268716 Objective Loss 0.268716 LR 0.001000 Time 0.020265 -2022-12-06 10:58:20,158 - Epoch: [80][ 800/ 1200] Overall Loss 0.268666 Objective Loss 0.268666 LR 0.001000 Time 0.020253 -2022-12-06 10:58:20,352 - Epoch: [80][ 810/ 1200] Overall Loss 0.268672 Objective Loss 0.268672 LR 0.001000 Time 0.020242 -2022-12-06 10:58:20,546 - Epoch: [80][ 820/ 1200] Overall Loss 0.268558 Objective Loss 0.268558 LR 0.001000 Time 0.020231 -2022-12-06 10:58:20,741 - Epoch: [80][ 830/ 1200] Overall Loss 0.268802 Objective Loss 0.268802 LR 0.001000 Time 0.020222 -2022-12-06 10:58:20,935 - Epoch: [80][ 840/ 1200] Overall Loss 0.268621 Objective Loss 0.268621 LR 0.001000 Time 0.020211 -2022-12-06 10:58:21,129 - Epoch: [80][ 850/ 1200] Overall Loss 0.268653 Objective Loss 0.268653 LR 0.001000 Time 0.020202 -2022-12-06 10:58:21,323 - Epoch: [80][ 860/ 1200] Overall Loss 0.268595 Objective Loss 0.268595 LR 0.001000 Time 0.020191 -2022-12-06 10:58:21,517 - Epoch: [80][ 870/ 1200] Overall Loss 0.268385 Objective Loss 0.268385 LR 0.001000 Time 0.020182 -2022-12-06 10:58:21,711 - Epoch: [80][ 880/ 1200] Overall Loss 0.268679 Objective Loss 0.268679 LR 0.001000 Time 0.020172 -2022-12-06 10:58:21,905 - Epoch: [80][ 890/ 1200] Overall Loss 0.269028 Objective Loss 0.269028 LR 0.001000 Time 0.020163 -2022-12-06 10:58:22,099 - Epoch: [80][ 900/ 1200] Overall Loss 0.268726 Objective Loss 0.268726 LR 0.001000 Time 0.020153 -2022-12-06 10:58:22,294 - Epoch: [80][ 910/ 1200] Overall Loss 0.268924 Objective Loss 0.268924 LR 0.001000 Time 0.020146 -2022-12-06 10:58:22,488 - Epoch: [80][ 920/ 1200] Overall Loss 0.269360 Objective Loss 0.269360 LR 0.001000 Time 0.020137 -2022-12-06 10:58:22,683 - Epoch: [80][ 930/ 1200] Overall Loss 0.269246 Objective Loss 0.269246 LR 0.001000 Time 0.020130 -2022-12-06 10:58:22,877 - Epoch: [80][ 940/ 1200] Overall Loss 0.269139 Objective Loss 0.269139 LR 0.001000 Time 0.020122 -2022-12-06 10:58:23,072 - Epoch: [80][ 950/ 1200] Overall Loss 0.269250 Objective Loss 0.269250 LR 0.001000 Time 0.020115 -2022-12-06 10:58:23,266 - Epoch: [80][ 960/ 1200] Overall Loss 0.269269 Objective Loss 0.269269 LR 0.001000 Time 0.020106 -2022-12-06 10:58:23,461 - Epoch: [80][ 970/ 1200] Overall Loss 0.269336 Objective Loss 0.269336 LR 0.001000 Time 0.020099 -2022-12-06 10:58:23,654 - Epoch: [80][ 980/ 1200] Overall Loss 0.269153 Objective Loss 0.269153 LR 0.001000 Time 0.020091 -2022-12-06 10:58:23,849 - Epoch: [80][ 990/ 1200] Overall Loss 0.269231 Objective Loss 0.269231 LR 0.001000 Time 0.020084 -2022-12-06 10:58:24,043 - Epoch: [80][ 1000/ 1200] Overall Loss 0.269239 Objective Loss 0.269239 LR 0.001000 Time 0.020077 -2022-12-06 10:58:24,238 - Epoch: [80][ 1010/ 1200] Overall Loss 0.269470 Objective Loss 0.269470 LR 0.001000 Time 0.020070 -2022-12-06 10:58:24,432 - Epoch: [80][ 1020/ 1200] Overall Loss 0.269673 Objective Loss 0.269673 LR 0.001000 Time 0.020063 -2022-12-06 10:58:24,627 - Epoch: [80][ 1030/ 1200] Overall Loss 0.269659 Objective Loss 0.269659 LR 0.001000 Time 0.020057 -2022-12-06 10:58:24,821 - Epoch: [80][ 1040/ 1200] Overall Loss 0.269609 Objective Loss 0.269609 LR 0.001000 Time 0.020051 -2022-12-06 10:58:25,017 - Epoch: [80][ 1050/ 1200] Overall Loss 0.269659 Objective Loss 0.269659 LR 0.001000 Time 0.020045 -2022-12-06 10:58:25,210 - Epoch: [80][ 1060/ 1200] Overall Loss 0.269884 Objective Loss 0.269884 LR 0.001000 Time 0.020038 -2022-12-06 10:58:25,405 - Epoch: [80][ 1070/ 1200] Overall Loss 0.269828 Objective Loss 0.269828 LR 0.001000 Time 0.020032 -2022-12-06 10:58:25,599 - Epoch: [80][ 1080/ 1200] Overall Loss 0.269810 Objective Loss 0.269810 LR 0.001000 Time 0.020026 -2022-12-06 10:58:25,794 - Epoch: [80][ 1090/ 1200] Overall Loss 0.269866 Objective Loss 0.269866 LR 0.001000 Time 0.020021 -2022-12-06 10:58:25,987 - Epoch: [80][ 1100/ 1200] Overall Loss 0.269924 Objective Loss 0.269924 LR 0.001000 Time 0.020014 -2022-12-06 10:58:26,183 - Epoch: [80][ 1110/ 1200] Overall Loss 0.269961 Objective Loss 0.269961 LR 0.001000 Time 0.020010 -2022-12-06 10:58:26,377 - Epoch: [80][ 1120/ 1200] Overall Loss 0.270105 Objective Loss 0.270105 LR 0.001000 Time 0.020004 -2022-12-06 10:58:26,571 - Epoch: [80][ 1130/ 1200] Overall Loss 0.269940 Objective Loss 0.269940 LR 0.001000 Time 0.019998 -2022-12-06 10:58:26,765 - Epoch: [80][ 1140/ 1200] Overall Loss 0.269916 Objective Loss 0.269916 LR 0.001000 Time 0.019992 -2022-12-06 10:58:26,961 - Epoch: [80][ 1150/ 1200] Overall Loss 0.270018 Objective Loss 0.270018 LR 0.001000 Time 0.019988 -2022-12-06 10:58:27,154 - Epoch: [80][ 1160/ 1200] Overall Loss 0.269999 Objective Loss 0.269999 LR 0.001000 Time 0.019982 -2022-12-06 10:58:27,349 - Epoch: [80][ 1170/ 1200] Overall Loss 0.269859 Objective Loss 0.269859 LR 0.001000 Time 0.019977 -2022-12-06 10:58:27,542 - Epoch: [80][ 1180/ 1200] Overall Loss 0.269830 Objective Loss 0.269830 LR 0.001000 Time 0.019971 -2022-12-06 10:58:27,737 - Epoch: [80][ 1190/ 1200] Overall Loss 0.270047 Objective Loss 0.270047 LR 0.001000 Time 0.019967 -2022-12-06 10:58:27,969 - Epoch: [80][ 1200/ 1200] Overall Loss 0.270282 Objective Loss 0.270282 Top1 84.728033 Top5 98.535565 LR 0.001000 Time 0.019993 -2022-12-06 10:58:28,059 - --- validate (epoch=80)----------- -2022-12-06 10:58:28,060 - 34129 samples (256 per mini-batch) -2022-12-06 10:58:28,510 - Epoch: [80][ 10/ 134] Loss 0.309792 Top1 83.281250 Top5 97.812500 -2022-12-06 10:58:28,649 - Epoch: [80][ 20/ 134] Loss 0.291916 Top1 83.378906 Top5 98.085938 -2022-12-06 10:58:28,782 - Epoch: [80][ 30/ 134] Loss 0.296135 Top1 83.619792 Top5 98.190104 -2022-12-06 10:58:28,913 - Epoch: [80][ 40/ 134] Loss 0.297123 Top1 83.876953 Top5 98.066406 -2022-12-06 10:58:29,041 - Epoch: [80][ 50/ 134] Loss 0.293080 Top1 83.929688 Top5 98.070312 -2022-12-06 10:58:29,170 - Epoch: [80][ 60/ 134] Loss 0.299664 Top1 83.854167 Top5 98.040365 -2022-12-06 10:58:29,302 - Epoch: [80][ 70/ 134] Loss 0.292904 Top1 83.878348 Top5 98.018973 -2022-12-06 10:58:29,431 - Epoch: [80][ 80/ 134] Loss 0.295449 Top1 83.798828 Top5 97.949219 -2022-12-06 10:58:29,561 - Epoch: [80][ 90/ 134] Loss 0.298115 Top1 83.732639 Top5 97.955729 -2022-12-06 10:58:29,689 - Epoch: [80][ 100/ 134] Loss 0.298924 Top1 83.742188 Top5 97.957031 -2022-12-06 10:58:29,818 - Epoch: [80][ 110/ 134] Loss 0.302032 Top1 83.636364 Top5 97.908381 -2022-12-06 10:58:29,947 - Epoch: [80][ 120/ 134] Loss 0.300903 Top1 83.645833 Top5 97.906901 -2022-12-06 10:58:30,078 - Epoch: [80][ 130/ 134] Loss 0.301903 Top1 83.641827 Top5 97.899639 -2022-12-06 10:58:30,116 - Epoch: [80][ 134/ 134] Loss 0.302705 Top1 83.641478 Top5 97.887427 -2022-12-06 10:58:30,211 - ==> Top1: 83.641 Top5: 97.887 Loss: 0.303 - -2022-12-06 10:58:30,212 - ==> Confusion: -[[ 896 1 1 2 12 5 1 2 4 48 0 3 2 5 6 1 1 3 0 0 3] - [ 3 915 3 6 11 31 4 8 1 1 2 4 2 1 5 2 4 0 13 7 4] - [ 7 3 986 25 3 2 25 5 1 2 7 6 4 0 2 6 0 2 2 3 12] - [ 5 2 17 942 1 1 0 1 0 0 9 0 6 2 12 1 1 4 12 1 3] - [ 12 6 4 1 933 6 1 0 1 7 0 4 0 4 15 8 8 3 0 4 3] - [ 4 12 1 5 2 958 0 19 3 2 2 17 6 19 2 1 1 1 1 5 8] - [ 0 4 7 1 0 4 1065 1 0 0 2 6 4 0 0 9 0 1 0 11 3] - [ 1 16 6 1 3 40 6 911 0 1 2 9 2 3 0 1 0 1 23 23 5] - [ 8 3 0 2 1 3 0 3 946 42 8 1 3 16 19 0 1 4 3 0 1] - [ 70 0 2 0 5 1 2 1 25 862 1 3 0 13 5 0 1 5 0 0 5] - [ 1 2 10 5 0 2 2 2 10 1 939 5 3 16 4 0 3 1 5 2 6] - [ 4 0 1 0 0 7 0 3 1 0 0 985 16 9 0 8 2 6 0 5 4] - [ 1 0 2 2 1 2 0 0 0 0 0 52 867 1 2 7 0 20 0 7 5] - [ 3 0 0 1 1 5 0 5 11 13 3 8 5 952 3 4 5 0 0 0 4] - [ 10 4 3 19 4 2 0 0 13 1 0 1 2 1 1056 0 2 1 6 1 4] - [ 3 0 2 2 2 1 1 0 0 0 0 9 3 4 0 993 1 18 0 2 2] - [ 3 6 1 0 2 0 1 0 1 0 0 4 7 1 1 23 1010 0 0 6 6] - [ 3 0 2 3 1 0 0 1 0 2 1 13 11 2 0 10 0 983 0 2 2] - [ 6 4 5 25 2 3 0 29 1 0 5 4 7 0 13 0 1 0 898 3 2] - [ 4 4 3 2 0 7 4 2 1 0 0 35 6 6 0 2 5 5 0 991 3] - [ 175 203 181 158 93 231 58 147 82 115 166 207 430 364 202 187 220 128 152 267 9460]] - -2022-12-06 10:58:30,885 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:58:30,886 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:58:30,892 - - -2022-12-06 10:58:30,892 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:58:31,819 - Epoch: [81][ 10/ 1200] Overall Loss 0.217769 Objective Loss 0.217769 LR 0.001000 Time 0.092693 -2022-12-06 10:58:32,012 - Epoch: [81][ 20/ 1200] Overall Loss 0.246586 Objective Loss 0.246586 LR 0.001000 Time 0.055955 -2022-12-06 10:58:32,204 - Epoch: [81][ 30/ 1200] Overall Loss 0.254503 Objective Loss 0.254503 LR 0.001000 Time 0.043692 -2022-12-06 10:58:32,396 - Epoch: [81][ 40/ 1200] Overall Loss 0.249012 Objective Loss 0.249012 LR 0.001000 Time 0.037550 -2022-12-06 10:58:32,588 - Epoch: [81][ 50/ 1200] Overall Loss 0.247630 Objective Loss 0.247630 LR 0.001000 Time 0.033869 -2022-12-06 10:58:32,780 - Epoch: [81][ 60/ 1200] Overall Loss 0.254120 Objective Loss 0.254120 LR 0.001000 Time 0.031409 -2022-12-06 10:58:32,971 - Epoch: [81][ 70/ 1200] Overall Loss 0.255516 Objective Loss 0.255516 LR 0.001000 Time 0.029653 -2022-12-06 10:58:33,163 - Epoch: [81][ 80/ 1200] Overall Loss 0.255733 Objective Loss 0.255733 LR 0.001000 Time 0.028330 -2022-12-06 10:58:33,355 - Epoch: [81][ 90/ 1200] Overall Loss 0.256435 Objective Loss 0.256435 LR 0.001000 Time 0.027310 -2022-12-06 10:58:33,547 - Epoch: [81][ 100/ 1200] Overall Loss 0.253211 Objective Loss 0.253211 LR 0.001000 Time 0.026494 -2022-12-06 10:58:33,738 - Epoch: [81][ 110/ 1200] Overall Loss 0.251191 Objective Loss 0.251191 LR 0.001000 Time 0.025820 -2022-12-06 10:58:33,930 - Epoch: [81][ 120/ 1200] Overall Loss 0.252647 Objective Loss 0.252647 LR 0.001000 Time 0.025262 -2022-12-06 10:58:34,122 - Epoch: [81][ 130/ 1200] Overall Loss 0.250950 Objective Loss 0.250950 LR 0.001000 Time 0.024789 -2022-12-06 10:58:34,313 - Epoch: [81][ 140/ 1200] Overall Loss 0.249983 Objective Loss 0.249983 LR 0.001000 Time 0.024385 -2022-12-06 10:58:34,505 - Epoch: [81][ 150/ 1200] Overall Loss 0.249356 Objective Loss 0.249356 LR 0.001000 Time 0.024033 -2022-12-06 10:58:34,697 - Epoch: [81][ 160/ 1200] Overall Loss 0.248462 Objective Loss 0.248462 LR 0.001000 Time 0.023725 -2022-12-06 10:58:34,888 - Epoch: [81][ 170/ 1200] Overall Loss 0.248869 Objective Loss 0.248869 LR 0.001000 Time 0.023454 -2022-12-06 10:58:35,080 - Epoch: [81][ 180/ 1200] Overall Loss 0.247031 Objective Loss 0.247031 LR 0.001000 Time 0.023212 -2022-12-06 10:58:35,272 - Epoch: [81][ 190/ 1200] Overall Loss 0.248624 Objective Loss 0.248624 LR 0.001000 Time 0.022999 -2022-12-06 10:58:35,464 - Epoch: [81][ 200/ 1200] Overall Loss 0.248712 Objective Loss 0.248712 LR 0.001000 Time 0.022805 -2022-12-06 10:58:35,655 - Epoch: [81][ 210/ 1200] Overall Loss 0.250140 Objective Loss 0.250140 LR 0.001000 Time 0.022629 -2022-12-06 10:58:35,848 - Epoch: [81][ 220/ 1200] Overall Loss 0.250059 Objective Loss 0.250059 LR 0.001000 Time 0.022471 -2022-12-06 10:58:36,039 - Epoch: [81][ 230/ 1200] Overall Loss 0.250698 Objective Loss 0.250698 LR 0.001000 Time 0.022325 -2022-12-06 10:58:36,231 - Epoch: [81][ 240/ 1200] Overall Loss 0.249953 Objective Loss 0.249953 LR 0.001000 Time 0.022192 -2022-12-06 10:58:36,423 - Epoch: [81][ 250/ 1200] Overall Loss 0.250102 Objective Loss 0.250102 LR 0.001000 Time 0.022068 -2022-12-06 10:58:36,614 - Epoch: [81][ 260/ 1200] Overall Loss 0.250978 Objective Loss 0.250978 LR 0.001000 Time 0.021955 -2022-12-06 10:58:36,806 - Epoch: [81][ 270/ 1200] Overall Loss 0.251450 Objective Loss 0.251450 LR 0.001000 Time 0.021850 -2022-12-06 10:58:36,998 - Epoch: [81][ 280/ 1200] Overall Loss 0.252722 Objective Loss 0.252722 LR 0.001000 Time 0.021753 -2022-12-06 10:58:37,190 - Epoch: [81][ 290/ 1200] Overall Loss 0.252725 Objective Loss 0.252725 LR 0.001000 Time 0.021662 -2022-12-06 10:58:37,382 - Epoch: [81][ 300/ 1200] Overall Loss 0.252890 Objective Loss 0.252890 LR 0.001000 Time 0.021577 -2022-12-06 10:58:37,573 - Epoch: [81][ 310/ 1200] Overall Loss 0.253600 Objective Loss 0.253600 LR 0.001000 Time 0.021498 -2022-12-06 10:58:37,765 - Epoch: [81][ 320/ 1200] Overall Loss 0.254630 Objective Loss 0.254630 LR 0.001000 Time 0.021424 -2022-12-06 10:58:37,956 - Epoch: [81][ 330/ 1200] Overall Loss 0.255205 Objective Loss 0.255205 LR 0.001000 Time 0.021353 -2022-12-06 10:58:38,148 - Epoch: [81][ 340/ 1200] Overall Loss 0.256034 Objective Loss 0.256034 LR 0.001000 Time 0.021287 -2022-12-06 10:58:38,340 - Epoch: [81][ 350/ 1200] Overall Loss 0.256106 Objective Loss 0.256106 LR 0.001000 Time 0.021226 -2022-12-06 10:58:38,531 - Epoch: [81][ 360/ 1200] Overall Loss 0.256218 Objective Loss 0.256218 LR 0.001000 Time 0.021166 -2022-12-06 10:58:38,723 - Epoch: [81][ 370/ 1200] Overall Loss 0.257392 Objective Loss 0.257392 LR 0.001000 Time 0.021110 -2022-12-06 10:58:38,915 - Epoch: [81][ 380/ 1200] Overall Loss 0.257365 Objective Loss 0.257365 LR 0.001000 Time 0.021058 -2022-12-06 10:58:39,106 - Epoch: [81][ 390/ 1200] Overall Loss 0.257347 Objective Loss 0.257347 LR 0.001000 Time 0.021007 -2022-12-06 10:58:39,298 - Epoch: [81][ 400/ 1200] Overall Loss 0.257279 Objective Loss 0.257279 LR 0.001000 Time 0.020959 -2022-12-06 10:58:39,489 - Epoch: [81][ 410/ 1200] Overall Loss 0.257181 Objective Loss 0.257181 LR 0.001000 Time 0.020915 -2022-12-06 10:58:39,680 - Epoch: [81][ 420/ 1200] Overall Loss 0.257125 Objective Loss 0.257125 LR 0.001000 Time 0.020870 -2022-12-06 10:58:39,872 - Epoch: [81][ 430/ 1200] Overall Loss 0.256553 Objective Loss 0.256553 LR 0.001000 Time 0.020830 -2022-12-06 10:58:40,064 - Epoch: [81][ 440/ 1200] Overall Loss 0.257618 Objective Loss 0.257618 LR 0.001000 Time 0.020790 -2022-12-06 10:58:40,255 - Epoch: [81][ 450/ 1200] Overall Loss 0.257416 Objective Loss 0.257416 LR 0.001000 Time 0.020753 -2022-12-06 10:58:40,447 - Epoch: [81][ 460/ 1200] Overall Loss 0.257501 Objective Loss 0.257501 LR 0.001000 Time 0.020717 -2022-12-06 10:58:40,639 - Epoch: [81][ 470/ 1200] Overall Loss 0.257994 Objective Loss 0.257994 LR 0.001000 Time 0.020684 -2022-12-06 10:58:40,831 - Epoch: [81][ 480/ 1200] Overall Loss 0.258181 Objective Loss 0.258181 LR 0.001000 Time 0.020651 -2022-12-06 10:58:41,022 - Epoch: [81][ 490/ 1200] Overall Loss 0.259101 Objective Loss 0.259101 LR 0.001000 Time 0.020619 -2022-12-06 10:58:41,214 - Epoch: [81][ 500/ 1200] Overall Loss 0.259390 Objective Loss 0.259390 LR 0.001000 Time 0.020589 -2022-12-06 10:58:41,405 - Epoch: [81][ 510/ 1200] Overall Loss 0.259401 Objective Loss 0.259401 LR 0.001000 Time 0.020560 -2022-12-06 10:58:41,597 - Epoch: [81][ 520/ 1200] Overall Loss 0.259855 Objective Loss 0.259855 LR 0.001000 Time 0.020531 -2022-12-06 10:58:41,789 - Epoch: [81][ 530/ 1200] Overall Loss 0.259622 Objective Loss 0.259622 LR 0.001000 Time 0.020506 -2022-12-06 10:58:41,980 - Epoch: [81][ 540/ 1200] Overall Loss 0.259821 Objective Loss 0.259821 LR 0.001000 Time 0.020479 -2022-12-06 10:58:42,172 - Epoch: [81][ 550/ 1200] Overall Loss 0.260102 Objective Loss 0.260102 LR 0.001000 Time 0.020455 -2022-12-06 10:58:42,364 - Epoch: [81][ 560/ 1200] Overall Loss 0.260463 Objective Loss 0.260463 LR 0.001000 Time 0.020431 -2022-12-06 10:58:42,556 - Epoch: [81][ 570/ 1200] Overall Loss 0.260559 Objective Loss 0.260559 LR 0.001000 Time 0.020408 -2022-12-06 10:58:42,747 - Epoch: [81][ 580/ 1200] Overall Loss 0.260908 Objective Loss 0.260908 LR 0.001000 Time 0.020386 -2022-12-06 10:58:42,940 - Epoch: [81][ 590/ 1200] Overall Loss 0.261637 Objective Loss 0.261637 LR 0.001000 Time 0.020365 -2022-12-06 10:58:43,131 - Epoch: [81][ 600/ 1200] Overall Loss 0.261569 Objective Loss 0.261569 LR 0.001000 Time 0.020344 -2022-12-06 10:58:43,323 - Epoch: [81][ 610/ 1200] Overall Loss 0.261694 Objective Loss 0.261694 LR 0.001000 Time 0.020323 -2022-12-06 10:58:43,514 - Epoch: [81][ 620/ 1200] Overall Loss 0.261920 Objective Loss 0.261920 LR 0.001000 Time 0.020304 -2022-12-06 10:58:43,706 - Epoch: [81][ 630/ 1200] Overall Loss 0.262271 Objective Loss 0.262271 LR 0.001000 Time 0.020286 -2022-12-06 10:58:43,899 - Epoch: [81][ 640/ 1200] Overall Loss 0.262676 Objective Loss 0.262676 LR 0.001000 Time 0.020268 -2022-12-06 10:58:44,091 - Epoch: [81][ 650/ 1200] Overall Loss 0.262653 Objective Loss 0.262653 LR 0.001000 Time 0.020251 -2022-12-06 10:58:44,283 - Epoch: [81][ 660/ 1200] Overall Loss 0.262640 Objective Loss 0.262640 LR 0.001000 Time 0.020234 -2022-12-06 10:58:44,475 - Epoch: [81][ 670/ 1200] Overall Loss 0.262642 Objective Loss 0.262642 LR 0.001000 Time 0.020218 -2022-12-06 10:58:44,666 - Epoch: [81][ 680/ 1200] Overall Loss 0.262595 Objective Loss 0.262595 LR 0.001000 Time 0.020202 -2022-12-06 10:58:44,858 - Epoch: [81][ 690/ 1200] Overall Loss 0.262862 Objective Loss 0.262862 LR 0.001000 Time 0.020186 -2022-12-06 10:58:45,050 - Epoch: [81][ 700/ 1200] Overall Loss 0.262851 Objective Loss 0.262851 LR 0.001000 Time 0.020171 -2022-12-06 10:58:45,242 - Epoch: [81][ 710/ 1200] Overall Loss 0.263187 Objective Loss 0.263187 LR 0.001000 Time 0.020157 -2022-12-06 10:58:45,434 - Epoch: [81][ 720/ 1200] Overall Loss 0.262841 Objective Loss 0.262841 LR 0.001000 Time 0.020142 -2022-12-06 10:58:45,626 - Epoch: [81][ 730/ 1200] Overall Loss 0.263047 Objective Loss 0.263047 LR 0.001000 Time 0.020129 -2022-12-06 10:58:45,818 - Epoch: [81][ 740/ 1200] Overall Loss 0.263348 Objective Loss 0.263348 LR 0.001000 Time 0.020116 -2022-12-06 10:58:46,010 - Epoch: [81][ 750/ 1200] Overall Loss 0.263592 Objective Loss 0.263592 LR 0.001000 Time 0.020103 -2022-12-06 10:58:46,203 - Epoch: [81][ 760/ 1200] Overall Loss 0.263640 Objective Loss 0.263640 LR 0.001000 Time 0.020091 -2022-12-06 10:58:46,394 - Epoch: [81][ 770/ 1200] Overall Loss 0.264199 Objective Loss 0.264199 LR 0.001000 Time 0.020078 -2022-12-06 10:58:46,586 - Epoch: [81][ 780/ 1200] Overall Loss 0.264494 Objective Loss 0.264494 LR 0.001000 Time 0.020066 -2022-12-06 10:58:46,778 - Epoch: [81][ 790/ 1200] Overall Loss 0.264455 Objective Loss 0.264455 LR 0.001000 Time 0.020054 -2022-12-06 10:58:46,970 - Epoch: [81][ 800/ 1200] Overall Loss 0.264580 Objective Loss 0.264580 LR 0.001000 Time 0.020042 -2022-12-06 10:58:47,161 - Epoch: [81][ 810/ 1200] Overall Loss 0.264650 Objective Loss 0.264650 LR 0.001000 Time 0.020031 -2022-12-06 10:58:47,353 - Epoch: [81][ 820/ 1200] Overall Loss 0.264752 Objective Loss 0.264752 LR 0.001000 Time 0.020020 -2022-12-06 10:58:47,545 - Epoch: [81][ 830/ 1200] Overall Loss 0.264863 Objective Loss 0.264863 LR 0.001000 Time 0.020009 -2022-12-06 10:58:47,737 - Epoch: [81][ 840/ 1200] Overall Loss 0.265284 Objective Loss 0.265284 LR 0.001000 Time 0.019999 -2022-12-06 10:58:47,929 - Epoch: [81][ 850/ 1200] Overall Loss 0.265387 Objective Loss 0.265387 LR 0.001000 Time 0.019989 -2022-12-06 10:58:48,121 - Epoch: [81][ 860/ 1200] Overall Loss 0.265440 Objective Loss 0.265440 LR 0.001000 Time 0.019979 -2022-12-06 10:58:48,313 - Epoch: [81][ 870/ 1200] Overall Loss 0.265755 Objective Loss 0.265755 LR 0.001000 Time 0.019969 -2022-12-06 10:58:48,504 - Epoch: [81][ 880/ 1200] Overall Loss 0.265582 Objective Loss 0.265582 LR 0.001000 Time 0.019959 -2022-12-06 10:58:48,696 - Epoch: [81][ 890/ 1200] Overall Loss 0.265814 Objective Loss 0.265814 LR 0.001000 Time 0.019950 -2022-12-06 10:58:48,888 - Epoch: [81][ 900/ 1200] Overall Loss 0.265944 Objective Loss 0.265944 LR 0.001000 Time 0.019941 -2022-12-06 10:58:49,080 - Epoch: [81][ 910/ 1200] Overall Loss 0.265659 Objective Loss 0.265659 LR 0.001000 Time 0.019932 -2022-12-06 10:58:49,272 - Epoch: [81][ 920/ 1200] Overall Loss 0.265551 Objective Loss 0.265551 LR 0.001000 Time 0.019924 -2022-12-06 10:58:49,464 - Epoch: [81][ 930/ 1200] Overall Loss 0.265643 Objective Loss 0.265643 LR 0.001000 Time 0.019915 -2022-12-06 10:58:49,656 - Epoch: [81][ 940/ 1200] Overall Loss 0.265853 Objective Loss 0.265853 LR 0.001000 Time 0.019907 -2022-12-06 10:58:49,848 - Epoch: [81][ 950/ 1200] Overall Loss 0.265883 Objective Loss 0.265883 LR 0.001000 Time 0.019899 -2022-12-06 10:58:50,040 - Epoch: [81][ 960/ 1200] Overall Loss 0.265923 Objective Loss 0.265923 LR 0.001000 Time 0.019891 -2022-12-06 10:58:50,232 - Epoch: [81][ 970/ 1200] Overall Loss 0.266029 Objective Loss 0.266029 LR 0.001000 Time 0.019883 -2022-12-06 10:58:50,424 - Epoch: [81][ 980/ 1200] Overall Loss 0.266143 Objective Loss 0.266143 LR 0.001000 Time 0.019875 -2022-12-06 10:58:50,616 - Epoch: [81][ 990/ 1200] Overall Loss 0.265993 Objective Loss 0.265993 LR 0.001000 Time 0.019868 -2022-12-06 10:58:50,808 - Epoch: [81][ 1000/ 1200] Overall Loss 0.266253 Objective Loss 0.266253 LR 0.001000 Time 0.019861 -2022-12-06 10:58:50,999 - Epoch: [81][ 1010/ 1200] Overall Loss 0.266573 Objective Loss 0.266573 LR 0.001000 Time 0.019854 -2022-12-06 10:58:51,192 - Epoch: [81][ 1020/ 1200] Overall Loss 0.266583 Objective Loss 0.266583 LR 0.001000 Time 0.019847 -2022-12-06 10:58:51,384 - Epoch: [81][ 1030/ 1200] Overall Loss 0.266435 Objective Loss 0.266435 LR 0.001000 Time 0.019840 -2022-12-06 10:58:51,575 - Epoch: [81][ 1040/ 1200] Overall Loss 0.266339 Objective Loss 0.266339 LR 0.001000 Time 0.019833 -2022-12-06 10:58:51,768 - Epoch: [81][ 1050/ 1200] Overall Loss 0.266206 Objective Loss 0.266206 LR 0.001000 Time 0.019827 -2022-12-06 10:58:51,960 - Epoch: [81][ 1060/ 1200] Overall Loss 0.266392 Objective Loss 0.266392 LR 0.001000 Time 0.019820 -2022-12-06 10:58:52,151 - Epoch: [81][ 1070/ 1200] Overall Loss 0.266505 Objective Loss 0.266505 LR 0.001000 Time 0.019813 -2022-12-06 10:58:52,343 - Epoch: [81][ 1080/ 1200] Overall Loss 0.266625 Objective Loss 0.266625 LR 0.001000 Time 0.019807 -2022-12-06 10:58:52,535 - Epoch: [81][ 1090/ 1200] Overall Loss 0.266638 Objective Loss 0.266638 LR 0.001000 Time 0.019801 -2022-12-06 10:58:52,727 - Epoch: [81][ 1100/ 1200] Overall Loss 0.266824 Objective Loss 0.266824 LR 0.001000 Time 0.019795 -2022-12-06 10:58:52,919 - Epoch: [81][ 1110/ 1200] Overall Loss 0.266945 Objective Loss 0.266945 LR 0.001000 Time 0.019789 -2022-12-06 10:58:53,110 - Epoch: [81][ 1120/ 1200] Overall Loss 0.267144 Objective Loss 0.267144 LR 0.001000 Time 0.019783 -2022-12-06 10:58:53,302 - Epoch: [81][ 1130/ 1200] Overall Loss 0.267513 Objective Loss 0.267513 LR 0.001000 Time 0.019777 -2022-12-06 10:58:53,494 - Epoch: [81][ 1140/ 1200] Overall Loss 0.267602 Objective Loss 0.267602 LR 0.001000 Time 0.019771 -2022-12-06 10:58:53,686 - Epoch: [81][ 1150/ 1200] Overall Loss 0.267709 Objective Loss 0.267709 LR 0.001000 Time 0.019766 -2022-12-06 10:58:53,877 - Epoch: [81][ 1160/ 1200] Overall Loss 0.267935 Objective Loss 0.267935 LR 0.001000 Time 0.019760 -2022-12-06 10:58:54,069 - Epoch: [81][ 1170/ 1200] Overall Loss 0.267903 Objective Loss 0.267903 LR 0.001000 Time 0.019754 -2022-12-06 10:58:54,260 - Epoch: [81][ 1180/ 1200] Overall Loss 0.268158 Objective Loss 0.268158 LR 0.001000 Time 0.019749 -2022-12-06 10:58:54,452 - Epoch: [81][ 1190/ 1200] Overall Loss 0.268390 Objective Loss 0.268390 LR 0.001000 Time 0.019743 -2022-12-06 10:58:54,677 - Epoch: [81][ 1200/ 1200] Overall Loss 0.268434 Objective Loss 0.268434 Top1 89.539749 Top5 98.744770 LR 0.001000 Time 0.019766 -2022-12-06 10:58:54,767 - --- validate (epoch=81)----------- -2022-12-06 10:58:54,767 - 34129 samples (256 per mini-batch) -2022-12-06 10:58:55,223 - Epoch: [81][ 10/ 134] Loss 0.293956 Top1 84.062500 Top5 98.125000 -2022-12-06 10:58:55,369 - Epoch: [81][ 20/ 134] Loss 0.297150 Top1 84.492188 Top5 98.320312 -2022-12-06 10:58:55,504 - Epoch: [81][ 30/ 134] Loss 0.307419 Top1 84.518229 Top5 98.111979 -2022-12-06 10:58:55,646 - Epoch: [81][ 40/ 134] Loss 0.310461 Top1 84.609375 Top5 98.027344 -2022-12-06 10:58:55,791 - Epoch: [81][ 50/ 134] Loss 0.314492 Top1 84.554688 Top5 97.992188 -2022-12-06 10:58:55,920 - Epoch: [81][ 60/ 134] Loss 0.312136 Top1 84.609375 Top5 98.040365 -2022-12-06 10:58:56,049 - Epoch: [81][ 70/ 134] Loss 0.314846 Top1 84.536830 Top5 97.946429 -2022-12-06 10:58:56,176 - Epoch: [81][ 80/ 134] Loss 0.314243 Top1 84.511719 Top5 97.954102 -2022-12-06 10:58:56,302 - Epoch: [81][ 90/ 134] Loss 0.311164 Top1 84.639757 Top5 98.016493 -2022-12-06 10:58:56,428 - Epoch: [81][ 100/ 134] Loss 0.317589 Top1 84.511719 Top5 97.968750 -2022-12-06 10:58:56,557 - Epoch: [81][ 110/ 134] Loss 0.318062 Top1 84.502841 Top5 97.922585 -2022-12-06 10:58:56,685 - Epoch: [81][ 120/ 134] Loss 0.316019 Top1 84.537760 Top5 97.932943 -2022-12-06 10:58:56,814 - Epoch: [81][ 130/ 134] Loss 0.316427 Top1 84.480168 Top5 97.947716 -2022-12-06 10:58:56,851 - Epoch: [81][ 134/ 134] Loss 0.315979 Top1 84.461895 Top5 97.951888 -2022-12-06 10:58:56,952 - ==> Top1: 84.462 Top5: 97.952 Loss: 0.316 - -2022-12-06 10:58:56,952 - ==> Confusion: -[[ 906 2 1 3 9 8 0 0 5 40 0 3 2 3 4 1 1 1 1 1 5] - [ 2 900 2 3 12 43 5 8 0 0 4 4 4 0 1 1 6 0 15 8 9] - [ 9 2 955 17 6 2 40 10 1 2 9 7 1 2 1 8 3 2 4 5 17] - [ 2 3 16 943 2 5 1 0 2 0 9 1 3 2 13 0 0 1 10 1 6] - [ 13 2 1 1 947 8 0 0 1 5 1 6 1 2 5 8 8 2 0 4 5] - [ 1 9 0 1 7 966 3 11 3 1 2 20 3 12 2 4 2 3 2 10 7] - [ 0 3 6 2 1 4 1065 2 0 0 0 2 3 0 0 10 1 3 2 9 5] - [ 0 8 9 3 1 42 16 886 1 0 5 7 4 2 0 1 3 1 33 27 5] - [ 10 5 0 2 2 1 0 0 951 38 20 4 2 13 8 0 2 0 2 1 3] - [ 88 1 0 0 6 4 1 3 33 828 1 3 0 13 2 0 0 4 1 2 11] - [ 2 2 8 6 0 2 2 1 7 1 942 3 2 14 4 2 1 0 7 2 11] - [ 5 1 0 0 0 4 3 2 1 0 0 981 22 7 0 4 2 3 2 11 3] - [ 0 0 1 2 0 1 0 1 1 0 0 42 881 0 1 9 0 14 0 3 13] - [ 0 0 0 0 0 13 0 3 8 14 13 7 4 937 0 3 1 2 0 5 13] - [ 11 2 3 17 6 4 0 0 23 7 3 2 3 3 1023 2 2 0 9 1 9] - [ 0 0 1 0 2 3 4 0 0 0 0 8 5 2 0 999 3 7 1 3 5] - [ 2 1 2 3 3 2 0 0 0 0 0 5 3 3 2 19 1006 1 0 8 12] - [ 3 1 1 3 0 0 2 2 0 2 0 18 21 3 1 23 2 951 0 1 2] - [ 1 6 3 17 2 4 0 20 3 1 7 4 4 1 6 0 1 0 923 2 3] - [ 2 0 1 2 0 4 6 4 1 0 1 20 6 4 0 3 3 6 2 1006 9] - [ 167 190 150 129 160 240 84 122 84 65 186 140 352 298 132 179 159 94 167 304 9824]] - -2022-12-06 10:58:57,630 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:58:57,630 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:58:57,636 - - -2022-12-06 10:58:57,636 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:58:58,582 - Epoch: [82][ 10/ 1200] Overall Loss 0.247163 Objective Loss 0.247163 LR 0.001000 Time 0.094501 -2022-12-06 10:58:58,798 - Epoch: [82][ 20/ 1200] Overall Loss 0.269200 Objective Loss 0.269200 LR 0.001000 Time 0.058008 -2022-12-06 10:58:59,001 - Epoch: [82][ 30/ 1200] Overall Loss 0.272049 Objective Loss 0.272049 LR 0.001000 Time 0.045406 -2022-12-06 10:58:59,206 - Epoch: [82][ 40/ 1200] Overall Loss 0.282754 Objective Loss 0.282754 LR 0.001000 Time 0.039168 -2022-12-06 10:58:59,408 - Epoch: [82][ 50/ 1200] Overall Loss 0.277278 Objective Loss 0.277278 LR 0.001000 Time 0.035374 -2022-12-06 10:58:59,614 - Epoch: [82][ 60/ 1200] Overall Loss 0.274919 Objective Loss 0.274919 LR 0.001000 Time 0.032894 -2022-12-06 10:58:59,816 - Epoch: [82][ 70/ 1200] Overall Loss 0.270326 Objective Loss 0.270326 LR 0.001000 Time 0.031074 -2022-12-06 10:59:00,021 - Epoch: [82][ 80/ 1200] Overall Loss 0.268902 Objective Loss 0.268902 LR 0.001000 Time 0.029741 -2022-12-06 10:59:00,223 - Epoch: [82][ 90/ 1200] Overall Loss 0.271268 Objective Loss 0.271268 LR 0.001000 Time 0.028675 -2022-12-06 10:59:00,428 - Epoch: [82][ 100/ 1200] Overall Loss 0.273746 Objective Loss 0.273746 LR 0.001000 Time 0.027856 -2022-12-06 10:59:00,630 - Epoch: [82][ 110/ 1200] Overall Loss 0.274519 Objective Loss 0.274519 LR 0.001000 Time 0.027151 -2022-12-06 10:59:00,835 - Epoch: [82][ 120/ 1200] Overall Loss 0.270642 Objective Loss 0.270642 LR 0.001000 Time 0.026597 -2022-12-06 10:59:01,037 - Epoch: [82][ 130/ 1200] Overall Loss 0.269528 Objective Loss 0.269528 LR 0.001000 Time 0.026099 -2022-12-06 10:59:01,243 - Epoch: [82][ 140/ 1200] Overall Loss 0.267930 Objective Loss 0.267930 LR 0.001000 Time 0.025699 -2022-12-06 10:59:01,445 - Epoch: [82][ 150/ 1200] Overall Loss 0.268870 Objective Loss 0.268870 LR 0.001000 Time 0.025330 -2022-12-06 10:59:01,652 - Epoch: [82][ 160/ 1200] Overall Loss 0.270658 Objective Loss 0.270658 LR 0.001000 Time 0.025035 -2022-12-06 10:59:01,854 - Epoch: [82][ 170/ 1200] Overall Loss 0.272096 Objective Loss 0.272096 LR 0.001000 Time 0.024749 -2022-12-06 10:59:02,060 - Epoch: [82][ 180/ 1200] Overall Loss 0.270857 Objective Loss 0.270857 LR 0.001000 Time 0.024511 -2022-12-06 10:59:02,262 - Epoch: [82][ 190/ 1200] Overall Loss 0.270966 Objective Loss 0.270966 LR 0.001000 Time 0.024282 -2022-12-06 10:59:02,467 - Epoch: [82][ 200/ 1200] Overall Loss 0.271732 Objective Loss 0.271732 LR 0.001000 Time 0.024091 -2022-12-06 10:59:02,669 - Epoch: [82][ 210/ 1200] Overall Loss 0.270760 Objective Loss 0.270760 LR 0.001000 Time 0.023905 -2022-12-06 10:59:02,875 - Epoch: [82][ 220/ 1200] Overall Loss 0.269860 Objective Loss 0.269860 LR 0.001000 Time 0.023750 -2022-12-06 10:59:03,078 - Epoch: [82][ 230/ 1200] Overall Loss 0.268762 Objective Loss 0.268762 LR 0.001000 Time 0.023596 -2022-12-06 10:59:03,284 - Epoch: [82][ 240/ 1200] Overall Loss 0.268994 Objective Loss 0.268994 LR 0.001000 Time 0.023471 -2022-12-06 10:59:03,486 - Epoch: [82][ 250/ 1200] Overall Loss 0.269030 Objective Loss 0.269030 LR 0.001000 Time 0.023337 -2022-12-06 10:59:03,692 - Epoch: [82][ 260/ 1200] Overall Loss 0.268055 Objective Loss 0.268055 LR 0.001000 Time 0.023230 -2022-12-06 10:59:03,894 - Epoch: [82][ 270/ 1200] Overall Loss 0.268498 Objective Loss 0.268498 LR 0.001000 Time 0.023116 -2022-12-06 10:59:04,100 - Epoch: [82][ 280/ 1200] Overall Loss 0.267678 Objective Loss 0.267678 LR 0.001000 Time 0.023022 -2022-12-06 10:59:04,302 - Epoch: [82][ 290/ 1200] Overall Loss 0.268118 Objective Loss 0.268118 LR 0.001000 Time 0.022922 -2022-12-06 10:59:04,508 - Epoch: [82][ 300/ 1200] Overall Loss 0.267990 Objective Loss 0.267990 LR 0.001000 Time 0.022842 -2022-12-06 10:59:04,710 - Epoch: [82][ 310/ 1200] Overall Loss 0.268629 Objective Loss 0.268629 LR 0.001000 Time 0.022756 -2022-12-06 10:59:04,916 - Epoch: [82][ 320/ 1200] Overall Loss 0.268211 Objective Loss 0.268211 LR 0.001000 Time 0.022686 -2022-12-06 10:59:05,118 - Epoch: [82][ 330/ 1200] Overall Loss 0.269029 Objective Loss 0.269029 LR 0.001000 Time 0.022608 -2022-12-06 10:59:05,325 - Epoch: [82][ 340/ 1200] Overall Loss 0.269753 Objective Loss 0.269753 LR 0.001000 Time 0.022552 -2022-12-06 10:59:05,528 - Epoch: [82][ 350/ 1200] Overall Loss 0.269425 Objective Loss 0.269425 LR 0.001000 Time 0.022486 -2022-12-06 10:59:05,734 - Epoch: [82][ 360/ 1200] Overall Loss 0.269295 Objective Loss 0.269295 LR 0.001000 Time 0.022431 -2022-12-06 10:59:05,937 - Epoch: [82][ 370/ 1200] Overall Loss 0.269158 Objective Loss 0.269158 LR 0.001000 Time 0.022371 -2022-12-06 10:59:06,142 - Epoch: [82][ 380/ 1200] Overall Loss 0.268504 Objective Loss 0.268504 LR 0.001000 Time 0.022322 -2022-12-06 10:59:06,344 - Epoch: [82][ 390/ 1200] Overall Loss 0.268452 Objective Loss 0.268452 LR 0.001000 Time 0.022266 -2022-12-06 10:59:06,551 - Epoch: [82][ 400/ 1200] Overall Loss 0.268824 Objective Loss 0.268824 LR 0.001000 Time 0.022224 -2022-12-06 10:59:06,753 - Epoch: [82][ 410/ 1200] Overall Loss 0.269042 Objective Loss 0.269042 LR 0.001000 Time 0.022173 -2022-12-06 10:59:06,959 - Epoch: [82][ 420/ 1200] Overall Loss 0.268862 Objective Loss 0.268862 LR 0.001000 Time 0.022135 -2022-12-06 10:59:07,162 - Epoch: [82][ 430/ 1200] Overall Loss 0.268349 Objective Loss 0.268349 LR 0.001000 Time 0.022092 -2022-12-06 10:59:07,368 - Epoch: [82][ 440/ 1200] Overall Loss 0.268714 Objective Loss 0.268714 LR 0.001000 Time 0.022056 -2022-12-06 10:59:07,571 - Epoch: [82][ 450/ 1200] Overall Loss 0.269029 Objective Loss 0.269029 LR 0.001000 Time 0.022015 -2022-12-06 10:59:07,777 - Epoch: [82][ 460/ 1200] Overall Loss 0.268121 Objective Loss 0.268121 LR 0.001000 Time 0.021983 -2022-12-06 10:59:07,984 - Epoch: [82][ 470/ 1200] Overall Loss 0.267860 Objective Loss 0.267860 LR 0.001000 Time 0.021954 -2022-12-06 10:59:08,196 - Epoch: [82][ 480/ 1200] Overall Loss 0.267902 Objective Loss 0.267902 LR 0.001000 Time 0.021938 -2022-12-06 10:59:08,403 - Epoch: [82][ 490/ 1200] Overall Loss 0.267736 Objective Loss 0.267736 LR 0.001000 Time 0.021911 -2022-12-06 10:59:08,615 - Epoch: [82][ 500/ 1200] Overall Loss 0.267940 Objective Loss 0.267940 LR 0.001000 Time 0.021895 -2022-12-06 10:59:08,818 - Epoch: [82][ 510/ 1200] Overall Loss 0.267987 Objective Loss 0.267987 LR 0.001000 Time 0.021864 -2022-12-06 10:59:09,024 - Epoch: [82][ 520/ 1200] Overall Loss 0.267865 Objective Loss 0.267865 LR 0.001000 Time 0.021838 -2022-12-06 10:59:09,226 - Epoch: [82][ 530/ 1200] Overall Loss 0.267974 Objective Loss 0.267974 LR 0.001000 Time 0.021806 -2022-12-06 10:59:09,433 - Epoch: [82][ 540/ 1200] Overall Loss 0.267667 Objective Loss 0.267667 LR 0.001000 Time 0.021785 -2022-12-06 10:59:09,636 - Epoch: [82][ 550/ 1200] Overall Loss 0.268044 Objective Loss 0.268044 LR 0.001000 Time 0.021756 -2022-12-06 10:59:09,842 - Epoch: [82][ 560/ 1200] Overall Loss 0.267595 Objective Loss 0.267595 LR 0.001000 Time 0.021733 -2022-12-06 10:59:10,044 - Epoch: [82][ 570/ 1200] Overall Loss 0.267446 Objective Loss 0.267446 LR 0.001000 Time 0.021707 -2022-12-06 10:59:10,251 - Epoch: [82][ 580/ 1200] Overall Loss 0.267525 Objective Loss 0.267525 LR 0.001000 Time 0.021687 -2022-12-06 10:59:10,453 - Epoch: [82][ 590/ 1200] Overall Loss 0.267330 Objective Loss 0.267330 LR 0.001000 Time 0.021662 -2022-12-06 10:59:10,659 - Epoch: [82][ 600/ 1200] Overall Loss 0.267061 Objective Loss 0.267061 LR 0.001000 Time 0.021643 -2022-12-06 10:59:10,862 - Epoch: [82][ 610/ 1200] Overall Loss 0.267118 Objective Loss 0.267118 LR 0.001000 Time 0.021619 -2022-12-06 10:59:11,068 - Epoch: [82][ 620/ 1200] Overall Loss 0.267270 Objective Loss 0.267270 LR 0.001000 Time 0.021602 -2022-12-06 10:59:11,270 - Epoch: [82][ 630/ 1200] Overall Loss 0.267152 Objective Loss 0.267152 LR 0.001000 Time 0.021580 -2022-12-06 10:59:11,476 - Epoch: [82][ 640/ 1200] Overall Loss 0.267148 Objective Loss 0.267148 LR 0.001000 Time 0.021563 -2022-12-06 10:59:11,678 - Epoch: [82][ 650/ 1200] Overall Loss 0.267507 Objective Loss 0.267507 LR 0.001000 Time 0.021541 -2022-12-06 10:59:11,885 - Epoch: [82][ 660/ 1200] Overall Loss 0.267425 Objective Loss 0.267425 LR 0.001000 Time 0.021527 -2022-12-06 10:59:12,087 - Epoch: [82][ 670/ 1200] Overall Loss 0.267546 Objective Loss 0.267546 LR 0.001000 Time 0.021506 -2022-12-06 10:59:12,294 - Epoch: [82][ 680/ 1200] Overall Loss 0.267885 Objective Loss 0.267885 LR 0.001000 Time 0.021493 -2022-12-06 10:59:12,496 - Epoch: [82][ 690/ 1200] Overall Loss 0.268218 Objective Loss 0.268218 LR 0.001000 Time 0.021475 -2022-12-06 10:59:12,703 - Epoch: [82][ 700/ 1200] Overall Loss 0.267908 Objective Loss 0.267908 LR 0.001000 Time 0.021462 -2022-12-06 10:59:12,905 - Epoch: [82][ 710/ 1200] Overall Loss 0.268479 Objective Loss 0.268479 LR 0.001000 Time 0.021443 -2022-12-06 10:59:13,112 - Epoch: [82][ 720/ 1200] Overall Loss 0.268505 Objective Loss 0.268505 LR 0.001000 Time 0.021432 -2022-12-06 10:59:13,314 - Epoch: [82][ 730/ 1200] Overall Loss 0.268627 Objective Loss 0.268627 LR 0.001000 Time 0.021415 -2022-12-06 10:59:13,520 - Epoch: [82][ 740/ 1200] Overall Loss 0.268847 Objective Loss 0.268847 LR 0.001000 Time 0.021403 -2022-12-06 10:59:13,722 - Epoch: [82][ 750/ 1200] Overall Loss 0.268900 Objective Loss 0.268900 LR 0.001000 Time 0.021386 -2022-12-06 10:59:13,928 - Epoch: [82][ 760/ 1200] Overall Loss 0.269150 Objective Loss 0.269150 LR 0.001000 Time 0.021375 -2022-12-06 10:59:14,130 - Epoch: [82][ 770/ 1200] Overall Loss 0.269455 Objective Loss 0.269455 LR 0.001000 Time 0.021359 -2022-12-06 10:59:14,337 - Epoch: [82][ 780/ 1200] Overall Loss 0.269042 Objective Loss 0.269042 LR 0.001000 Time 0.021350 -2022-12-06 10:59:14,540 - Epoch: [82][ 790/ 1200] Overall Loss 0.269272 Objective Loss 0.269272 LR 0.001000 Time 0.021336 -2022-12-06 10:59:14,747 - Epoch: [82][ 800/ 1200] Overall Loss 0.268954 Objective Loss 0.268954 LR 0.001000 Time 0.021326 -2022-12-06 10:59:14,950 - Epoch: [82][ 810/ 1200] Overall Loss 0.270193 Objective Loss 0.270193 LR 0.001000 Time 0.021313 -2022-12-06 10:59:15,158 - Epoch: [82][ 820/ 1200] Overall Loss 0.272349 Objective Loss 0.272349 LR 0.001000 Time 0.021306 -2022-12-06 10:59:15,360 - Epoch: [82][ 830/ 1200] Overall Loss 0.274201 Objective Loss 0.274201 LR 0.001000 Time 0.021292 -2022-12-06 10:59:15,567 - Epoch: [82][ 840/ 1200] Overall Loss 0.275842 Objective Loss 0.275842 LR 0.001000 Time 0.021284 -2022-12-06 10:59:15,770 - Epoch: [82][ 850/ 1200] Overall Loss 0.277312 Objective Loss 0.277312 LR 0.001000 Time 0.021272 -2022-12-06 10:59:15,977 - Epoch: [82][ 860/ 1200] Overall Loss 0.278422 Objective Loss 0.278422 LR 0.001000 Time 0.021264 -2022-12-06 10:59:16,179 - Epoch: [82][ 870/ 1200] Overall Loss 0.279545 Objective Loss 0.279545 LR 0.001000 Time 0.021252 -2022-12-06 10:59:16,385 - Epoch: [82][ 880/ 1200] Overall Loss 0.280757 Objective Loss 0.280757 LR 0.001000 Time 0.021244 -2022-12-06 10:59:16,588 - Epoch: [82][ 890/ 1200] Overall Loss 0.281994 Objective Loss 0.281994 LR 0.001000 Time 0.021233 -2022-12-06 10:59:16,795 - Epoch: [82][ 900/ 1200] Overall Loss 0.283318 Objective Loss 0.283318 LR 0.001000 Time 0.021226 -2022-12-06 10:59:16,997 - Epoch: [82][ 910/ 1200] Overall Loss 0.284632 Objective Loss 0.284632 LR 0.001000 Time 0.021214 -2022-12-06 10:59:17,203 - Epoch: [82][ 920/ 1200] Overall Loss 0.285606 Objective Loss 0.285606 LR 0.001000 Time 0.021206 -2022-12-06 10:59:17,406 - Epoch: [82][ 930/ 1200] Overall Loss 0.286597 Objective Loss 0.286597 LR 0.001000 Time 0.021196 -2022-12-06 10:59:17,612 - Epoch: [82][ 940/ 1200] Overall Loss 0.287474 Objective Loss 0.287474 LR 0.001000 Time 0.021189 -2022-12-06 10:59:17,815 - Epoch: [82][ 950/ 1200] Overall Loss 0.288440 Objective Loss 0.288440 LR 0.001000 Time 0.021179 -2022-12-06 10:59:18,021 - Epoch: [82][ 960/ 1200] Overall Loss 0.289010 Objective Loss 0.289010 LR 0.001000 Time 0.021173 -2022-12-06 10:59:18,223 - Epoch: [82][ 970/ 1200] Overall Loss 0.289624 Objective Loss 0.289624 LR 0.001000 Time 0.021162 -2022-12-06 10:59:18,429 - Epoch: [82][ 980/ 1200] Overall Loss 0.290020 Objective Loss 0.290020 LR 0.001000 Time 0.021155 -2022-12-06 10:59:18,632 - Epoch: [82][ 990/ 1200] Overall Loss 0.290883 Objective Loss 0.290883 LR 0.001000 Time 0.021146 -2022-12-06 10:59:18,838 - Epoch: [82][ 1000/ 1200] Overall Loss 0.291615 Objective Loss 0.291615 LR 0.001000 Time 0.021140 -2022-12-06 10:59:19,041 - Epoch: [82][ 1010/ 1200] Overall Loss 0.292162 Objective Loss 0.292162 LR 0.001000 Time 0.021131 -2022-12-06 10:59:19,247 - Epoch: [82][ 1020/ 1200] Overall Loss 0.292669 Objective Loss 0.292669 LR 0.001000 Time 0.021126 -2022-12-06 10:59:19,450 - Epoch: [82][ 1030/ 1200] Overall Loss 0.293221 Objective Loss 0.293221 LR 0.001000 Time 0.021117 -2022-12-06 10:59:19,657 - Epoch: [82][ 1040/ 1200] Overall Loss 0.293635 Objective Loss 0.293635 LR 0.001000 Time 0.021112 -2022-12-06 10:59:19,860 - Epoch: [82][ 1050/ 1200] Overall Loss 0.294079 Objective Loss 0.294079 LR 0.001000 Time 0.021104 -2022-12-06 10:59:20,066 - Epoch: [82][ 1060/ 1200] Overall Loss 0.294674 Objective Loss 0.294674 LR 0.001000 Time 0.021099 -2022-12-06 10:59:20,269 - Epoch: [82][ 1070/ 1200] Overall Loss 0.295310 Objective Loss 0.295310 LR 0.001000 Time 0.021090 -2022-12-06 10:59:20,476 - Epoch: [82][ 1080/ 1200] Overall Loss 0.295797 Objective Loss 0.295797 LR 0.001000 Time 0.021086 -2022-12-06 10:59:20,678 - Epoch: [82][ 1090/ 1200] Overall Loss 0.296221 Objective Loss 0.296221 LR 0.001000 Time 0.021078 -2022-12-06 10:59:20,885 - Epoch: [82][ 1100/ 1200] Overall Loss 0.296743 Objective Loss 0.296743 LR 0.001000 Time 0.021074 -2022-12-06 10:59:21,089 - Epoch: [82][ 1110/ 1200] Overall Loss 0.297210 Objective Loss 0.297210 LR 0.001000 Time 0.021067 -2022-12-06 10:59:21,295 - Epoch: [82][ 1120/ 1200] Overall Loss 0.297516 Objective Loss 0.297516 LR 0.001000 Time 0.021062 -2022-12-06 10:59:21,497 - Epoch: [82][ 1130/ 1200] Overall Loss 0.298061 Objective Loss 0.298061 LR 0.001000 Time 0.021054 -2022-12-06 10:59:21,703 - Epoch: [82][ 1140/ 1200] Overall Loss 0.298538 Objective Loss 0.298538 LR 0.001000 Time 0.021049 -2022-12-06 10:59:21,906 - Epoch: [82][ 1150/ 1200] Overall Loss 0.298837 Objective Loss 0.298837 LR 0.001000 Time 0.021042 -2022-12-06 10:59:22,112 - Epoch: [82][ 1160/ 1200] Overall Loss 0.299321 Objective Loss 0.299321 LR 0.001000 Time 0.021038 -2022-12-06 10:59:22,314 - Epoch: [82][ 1170/ 1200] Overall Loss 0.299711 Objective Loss 0.299711 LR 0.001000 Time 0.021031 -2022-12-06 10:59:22,521 - Epoch: [82][ 1180/ 1200] Overall Loss 0.300064 Objective Loss 0.300064 LR 0.001000 Time 0.021027 -2022-12-06 10:59:22,724 - Epoch: [82][ 1190/ 1200] Overall Loss 0.300508 Objective Loss 0.300508 LR 0.001000 Time 0.021020 -2022-12-06 10:59:22,959 - Epoch: [82][ 1200/ 1200] Overall Loss 0.300966 Objective Loss 0.300966 Top1 84.518828 Top5 98.535565 LR 0.001000 Time 0.021041 -2022-12-06 10:59:23,047 - --- validate (epoch=82)----------- -2022-12-06 10:59:23,047 - 34129 samples (256 per mini-batch) -2022-12-06 10:59:23,498 - Epoch: [82][ 10/ 134] Loss 0.368001 Top1 82.734375 Top5 97.304688 -2022-12-06 10:59:23,631 - Epoch: [82][ 20/ 134] Loss 0.362325 Top1 82.890625 Top5 97.460938 -2022-12-06 10:59:23,762 - Epoch: [82][ 30/ 134] Loss 0.359847 Top1 83.085938 Top5 97.591146 -2022-12-06 10:59:23,893 - Epoch: [82][ 40/ 134] Loss 0.360711 Top1 83.085938 Top5 97.685547 -2022-12-06 10:59:24,023 - Epoch: [82][ 50/ 134] Loss 0.357880 Top1 83.101562 Top5 97.726562 -2022-12-06 10:59:24,153 - Epoch: [82][ 60/ 134] Loss 0.358271 Top1 83.131510 Top5 97.734375 -2022-12-06 10:59:24,282 - Epoch: [82][ 70/ 134] Loss 0.359571 Top1 83.052455 Top5 97.756696 -2022-12-06 10:59:24,412 - Epoch: [82][ 80/ 134] Loss 0.359856 Top1 83.027344 Top5 97.758789 -2022-12-06 10:59:24,542 - Epoch: [82][ 90/ 134] Loss 0.360105 Top1 83.012153 Top5 97.725694 -2022-12-06 10:59:24,672 - Epoch: [82][ 100/ 134] Loss 0.361423 Top1 83.117188 Top5 97.750000 -2022-12-06 10:59:24,802 - Epoch: [82][ 110/ 134] Loss 0.360570 Top1 83.249290 Top5 97.720170 -2022-12-06 10:59:24,932 - Epoch: [82][ 120/ 134] Loss 0.358117 Top1 83.346354 Top5 97.744141 -2022-12-06 10:59:25,066 - Epoch: [82][ 130/ 134] Loss 0.357681 Top1 83.398438 Top5 97.743389 -2022-12-06 10:59:25,105 - Epoch: [82][ 134/ 134] Loss 0.359180 Top1 83.345542 Top5 97.717484 -2022-12-06 10:59:25,192 - ==> Top1: 83.346 Top5: 97.717 Loss: 0.359 - -2022-12-06 10:59:25,193 - ==> Confusion: -[[ 906 1 1 5 5 2 1 1 5 44 0 3 3 6 6 2 0 1 0 3 1] - [ 1 909 2 2 8 27 1 20 3 0 4 11 4 4 0 0 5 2 14 4 6] - [ 9 1 998 19 3 2 19 6 1 1 3 5 4 3 3 5 0 1 6 5 9] - [ 3 1 22 937 0 3 0 3 0 1 9 0 3 2 11 0 0 3 17 0 5] - [ 13 4 1 1 947 10 1 2 1 7 1 6 1 2 8 5 4 1 0 1 4] - [ 2 12 0 4 4 965 2 24 2 1 0 15 3 18 1 1 2 1 2 6 4] - [ 0 2 14 2 2 5 1062 7 0 0 0 3 2 2 0 6 1 0 1 6 3] - [ 1 5 6 4 1 33 7 937 1 1 4 9 0 0 0 0 0 1 31 11 2] - [ 6 3 1 1 0 1 0 1 967 49 6 5 1 12 6 0 3 0 2 0 0] - [ 74 0 0 0 3 4 0 2 17 877 1 1 0 13 3 0 0 1 1 0 4] - [ 1 2 2 6 2 2 2 4 10 3 951 3 2 13 3 0 0 0 9 0 4] - [ 1 1 0 0 0 5 2 2 2 1 3 982 23 10 1 5 2 5 0 4 2] - [ 0 0 1 2 1 1 2 0 1 0 0 29 890 3 2 10 2 15 1 4 5] - [ 1 2 1 0 1 10 0 0 7 12 3 1 2 968 3 1 2 1 2 2 4] - [ 12 2 3 9 4 0 0 1 14 7 0 2 2 4 1059 2 0 0 5 0 4] - [ 1 0 2 0 1 3 4 0 1 0 1 5 7 4 0 987 8 9 0 5 5] - [ 3 2 2 1 4 1 1 0 1 2 0 3 4 4 2 13 1012 1 1 9 6] - [ 2 0 2 5 1 0 1 1 2 0 0 7 18 2 0 11 0 979 1 4 0] - [ 1 4 4 14 1 1 0 22 4 1 6 1 1 1 11 1 0 3 925 4 3] - [ 1 2 2 1 1 6 6 8 0 0 3 22 9 5 1 5 2 2 0 1002 2] - [ 189 235 220 159 142 207 79 170 110 110 188 142 503 430 204 138 161 112 241 303 9183]] - -2022-12-06 10:59:25,741 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:59:25,742 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:59:25,747 - - -2022-12-06 10:59:25,747 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:59:26,780 - Epoch: [83][ 10/ 1200] Overall Loss 0.334684 Objective Loss 0.334684 LR 0.001000 Time 0.103232 -2022-12-06 10:59:26,975 - Epoch: [83][ 20/ 1200] Overall Loss 0.329179 Objective Loss 0.329179 LR 0.001000 Time 0.061311 -2022-12-06 10:59:27,167 - Epoch: [83][ 30/ 1200] Overall Loss 0.336182 Objective Loss 0.336182 LR 0.001000 Time 0.047240 -2022-12-06 10:59:27,358 - Epoch: [83][ 40/ 1200] Overall Loss 0.332291 Objective Loss 0.332291 LR 0.001000 Time 0.040191 -2022-12-06 10:59:27,549 - Epoch: [83][ 50/ 1200] Overall Loss 0.335978 Objective Loss 0.335978 LR 0.001000 Time 0.035964 -2022-12-06 10:59:27,740 - Epoch: [83][ 60/ 1200] Overall Loss 0.336227 Objective Loss 0.336227 LR 0.001000 Time 0.033144 -2022-12-06 10:59:27,930 - Epoch: [83][ 70/ 1200] Overall Loss 0.334083 Objective Loss 0.334083 LR 0.001000 Time 0.031127 -2022-12-06 10:59:28,121 - Epoch: [83][ 80/ 1200] Overall Loss 0.334162 Objective Loss 0.334162 LR 0.001000 Time 0.029614 -2022-12-06 10:59:28,311 - Epoch: [83][ 90/ 1200] Overall Loss 0.336047 Objective Loss 0.336047 LR 0.001000 Time 0.028430 -2022-12-06 10:59:28,502 - Epoch: [83][ 100/ 1200] Overall Loss 0.334412 Objective Loss 0.334412 LR 0.001000 Time 0.027490 -2022-12-06 10:59:28,693 - Epoch: [83][ 110/ 1200] Overall Loss 0.334122 Objective Loss 0.334122 LR 0.001000 Time 0.026722 -2022-12-06 10:59:28,884 - Epoch: [83][ 120/ 1200] Overall Loss 0.334167 Objective Loss 0.334167 LR 0.001000 Time 0.026083 -2022-12-06 10:59:29,075 - Epoch: [83][ 130/ 1200] Overall Loss 0.334246 Objective Loss 0.334246 LR 0.001000 Time 0.025542 -2022-12-06 10:59:29,266 - Epoch: [83][ 140/ 1200] Overall Loss 0.332830 Objective Loss 0.332830 LR 0.001000 Time 0.025076 -2022-12-06 10:59:29,457 - Epoch: [83][ 150/ 1200] Overall Loss 0.332632 Objective Loss 0.332632 LR 0.001000 Time 0.024673 -2022-12-06 10:59:29,648 - Epoch: [83][ 160/ 1200] Overall Loss 0.331557 Objective Loss 0.331557 LR 0.001000 Time 0.024321 -2022-12-06 10:59:29,839 - Epoch: [83][ 170/ 1200] Overall Loss 0.330503 Objective Loss 0.330503 LR 0.001000 Time 0.024009 -2022-12-06 10:59:30,030 - Epoch: [83][ 180/ 1200] Overall Loss 0.330740 Objective Loss 0.330740 LR 0.001000 Time 0.023734 -2022-12-06 10:59:30,220 - Epoch: [83][ 190/ 1200] Overall Loss 0.331392 Objective Loss 0.331392 LR 0.001000 Time 0.023484 -2022-12-06 10:59:30,411 - Epoch: [83][ 200/ 1200] Overall Loss 0.331915 Objective Loss 0.331915 LR 0.001000 Time 0.023263 -2022-12-06 10:59:30,602 - Epoch: [83][ 210/ 1200] Overall Loss 0.331360 Objective Loss 0.331360 LR 0.001000 Time 0.023062 -2022-12-06 10:59:30,793 - Epoch: [83][ 220/ 1200] Overall Loss 0.330176 Objective Loss 0.330176 LR 0.001000 Time 0.022879 -2022-12-06 10:59:30,984 - Epoch: [83][ 230/ 1200] Overall Loss 0.330095 Objective Loss 0.330095 LR 0.001000 Time 0.022712 -2022-12-06 10:59:31,175 - Epoch: [83][ 240/ 1200] Overall Loss 0.329652 Objective Loss 0.329652 LR 0.001000 Time 0.022558 -2022-12-06 10:59:31,366 - Epoch: [83][ 250/ 1200] Overall Loss 0.329070 Objective Loss 0.329070 LR 0.001000 Time 0.022418 -2022-12-06 10:59:31,557 - Epoch: [83][ 260/ 1200] Overall Loss 0.328762 Objective Loss 0.328762 LR 0.001000 Time 0.022287 -2022-12-06 10:59:31,748 - Epoch: [83][ 270/ 1200] Overall Loss 0.327827 Objective Loss 0.327827 LR 0.001000 Time 0.022168 -2022-12-06 10:59:31,940 - Epoch: [83][ 280/ 1200] Overall Loss 0.327880 Objective Loss 0.327880 LR 0.001000 Time 0.022059 -2022-12-06 10:59:32,131 - Epoch: [83][ 290/ 1200] Overall Loss 0.328491 Objective Loss 0.328491 LR 0.001000 Time 0.021955 -2022-12-06 10:59:32,322 - Epoch: [83][ 300/ 1200] Overall Loss 0.328401 Objective Loss 0.328401 LR 0.001000 Time 0.021858 -2022-12-06 10:59:32,513 - Epoch: [83][ 310/ 1200] Overall Loss 0.328613 Objective Loss 0.328613 LR 0.001000 Time 0.021770 -2022-12-06 10:59:32,705 - Epoch: [83][ 320/ 1200] Overall Loss 0.329176 Objective Loss 0.329176 LR 0.001000 Time 0.021687 -2022-12-06 10:59:32,896 - Epoch: [83][ 330/ 1200] Overall Loss 0.328882 Objective Loss 0.328882 LR 0.001000 Time 0.021607 -2022-12-06 10:59:33,088 - Epoch: [83][ 340/ 1200] Overall Loss 0.328731 Objective Loss 0.328731 LR 0.001000 Time 0.021533 -2022-12-06 10:59:33,279 - Epoch: [83][ 350/ 1200] Overall Loss 0.328338 Objective Loss 0.328338 LR 0.001000 Time 0.021463 -2022-12-06 10:59:33,470 - Epoch: [83][ 360/ 1200] Overall Loss 0.328118 Objective Loss 0.328118 LR 0.001000 Time 0.021396 -2022-12-06 10:59:33,662 - Epoch: [83][ 370/ 1200] Overall Loss 0.327898 Objective Loss 0.327898 LR 0.001000 Time 0.021336 -2022-12-06 10:59:33,853 - Epoch: [83][ 380/ 1200] Overall Loss 0.328428 Objective Loss 0.328428 LR 0.001000 Time 0.021276 -2022-12-06 10:59:34,044 - Epoch: [83][ 390/ 1200] Overall Loss 0.327520 Objective Loss 0.327520 LR 0.001000 Time 0.021217 -2022-12-06 10:59:34,235 - Epoch: [83][ 400/ 1200] Overall Loss 0.327179 Objective Loss 0.327179 LR 0.001000 Time 0.021164 -2022-12-06 10:59:34,426 - Epoch: [83][ 410/ 1200] Overall Loss 0.326920 Objective Loss 0.326920 LR 0.001000 Time 0.021112 -2022-12-06 10:59:34,618 - Epoch: [83][ 420/ 1200] Overall Loss 0.327058 Objective Loss 0.327058 LR 0.001000 Time 0.021064 -2022-12-06 10:59:34,809 - Epoch: [83][ 430/ 1200] Overall Loss 0.326089 Objective Loss 0.326089 LR 0.001000 Time 0.021019 -2022-12-06 10:59:35,001 - Epoch: [83][ 440/ 1200] Overall Loss 0.325772 Objective Loss 0.325772 LR 0.001000 Time 0.020975 -2022-12-06 10:59:35,192 - Epoch: [83][ 450/ 1200] Overall Loss 0.325997 Objective Loss 0.325997 LR 0.001000 Time 0.020933 -2022-12-06 10:59:35,383 - Epoch: [83][ 460/ 1200] Overall Loss 0.326052 Objective Loss 0.326052 LR 0.001000 Time 0.020891 -2022-12-06 10:59:35,574 - Epoch: [83][ 470/ 1200] Overall Loss 0.326304 Objective Loss 0.326304 LR 0.001000 Time 0.020852 -2022-12-06 10:59:35,766 - Epoch: [83][ 480/ 1200] Overall Loss 0.325899 Objective Loss 0.325899 LR 0.001000 Time 0.020816 -2022-12-06 10:59:35,957 - Epoch: [83][ 490/ 1200] Overall Loss 0.326041 Objective Loss 0.326041 LR 0.001000 Time 0.020780 -2022-12-06 10:59:36,149 - Epoch: [83][ 500/ 1200] Overall Loss 0.325638 Objective Loss 0.325638 LR 0.001000 Time 0.020747 -2022-12-06 10:59:36,340 - Epoch: [83][ 510/ 1200] Overall Loss 0.325660 Objective Loss 0.325660 LR 0.001000 Time 0.020714 -2022-12-06 10:59:36,531 - Epoch: [83][ 520/ 1200] Overall Loss 0.325451 Objective Loss 0.325451 LR 0.001000 Time 0.020683 -2022-12-06 10:59:36,723 - Epoch: [83][ 530/ 1200] Overall Loss 0.325168 Objective Loss 0.325168 LR 0.001000 Time 0.020653 -2022-12-06 10:59:36,914 - Epoch: [83][ 540/ 1200] Overall Loss 0.325199 Objective Loss 0.325199 LR 0.001000 Time 0.020623 -2022-12-06 10:59:37,106 - Epoch: [83][ 550/ 1200] Overall Loss 0.325941 Objective Loss 0.325941 LR 0.001000 Time 0.020596 -2022-12-06 10:59:37,297 - Epoch: [83][ 560/ 1200] Overall Loss 0.325468 Objective Loss 0.325468 LR 0.001000 Time 0.020569 -2022-12-06 10:59:37,488 - Epoch: [83][ 570/ 1200] Overall Loss 0.325334 Objective Loss 0.325334 LR 0.001000 Time 0.020543 -2022-12-06 10:59:37,680 - Epoch: [83][ 580/ 1200] Overall Loss 0.325119 Objective Loss 0.325119 LR 0.001000 Time 0.020518 -2022-12-06 10:59:37,871 - Epoch: [83][ 590/ 1200] Overall Loss 0.324902 Objective Loss 0.324902 LR 0.001000 Time 0.020493 -2022-12-06 10:59:38,062 - Epoch: [83][ 600/ 1200] Overall Loss 0.324666 Objective Loss 0.324666 LR 0.001000 Time 0.020469 -2022-12-06 10:59:38,253 - Epoch: [83][ 610/ 1200] Overall Loss 0.324539 Objective Loss 0.324539 LR 0.001000 Time 0.020446 -2022-12-06 10:59:38,445 - Epoch: [83][ 620/ 1200] Overall Loss 0.324789 Objective Loss 0.324789 LR 0.001000 Time 0.020424 -2022-12-06 10:59:38,637 - Epoch: [83][ 630/ 1200] Overall Loss 0.324374 Objective Loss 0.324374 LR 0.001000 Time 0.020403 -2022-12-06 10:59:38,828 - Epoch: [83][ 640/ 1200] Overall Loss 0.324157 Objective Loss 0.324157 LR 0.001000 Time 0.020383 -2022-12-06 10:59:39,020 - Epoch: [83][ 650/ 1200] Overall Loss 0.323516 Objective Loss 0.323516 LR 0.001000 Time 0.020364 -2022-12-06 10:59:39,211 - Epoch: [83][ 660/ 1200] Overall Loss 0.323891 Objective Loss 0.323891 LR 0.001000 Time 0.020344 -2022-12-06 10:59:39,402 - Epoch: [83][ 670/ 1200] Overall Loss 0.323503 Objective Loss 0.323503 LR 0.001000 Time 0.020325 -2022-12-06 10:59:39,593 - Epoch: [83][ 680/ 1200] Overall Loss 0.323612 Objective Loss 0.323612 LR 0.001000 Time 0.020306 -2022-12-06 10:59:39,785 - Epoch: [83][ 690/ 1200] Overall Loss 0.323251 Objective Loss 0.323251 LR 0.001000 Time 0.020289 -2022-12-06 10:59:39,977 - Epoch: [83][ 700/ 1200] Overall Loss 0.323247 Objective Loss 0.323247 LR 0.001000 Time 0.020272 -2022-12-06 10:59:40,168 - Epoch: [83][ 710/ 1200] Overall Loss 0.323418 Objective Loss 0.323418 LR 0.001000 Time 0.020255 -2022-12-06 10:59:40,359 - Epoch: [83][ 720/ 1200] Overall Loss 0.323189 Objective Loss 0.323189 LR 0.001000 Time 0.020238 -2022-12-06 10:59:40,549 - Epoch: [83][ 730/ 1200] Overall Loss 0.323040 Objective Loss 0.323040 LR 0.001000 Time 0.020221 -2022-12-06 10:59:40,740 - Epoch: [83][ 740/ 1200] Overall Loss 0.322895 Objective Loss 0.322895 LR 0.001000 Time 0.020205 -2022-12-06 10:59:40,931 - Epoch: [83][ 750/ 1200] Overall Loss 0.323054 Objective Loss 0.323054 LR 0.001000 Time 0.020190 -2022-12-06 10:59:41,123 - Epoch: [83][ 760/ 1200] Overall Loss 0.322965 Objective Loss 0.322965 LR 0.001000 Time 0.020176 -2022-12-06 10:59:41,314 - Epoch: [83][ 770/ 1200] Overall Loss 0.322825 Objective Loss 0.322825 LR 0.001000 Time 0.020161 -2022-12-06 10:59:41,504 - Epoch: [83][ 780/ 1200] Overall Loss 0.322909 Objective Loss 0.322909 LR 0.001000 Time 0.020146 -2022-12-06 10:59:41,696 - Epoch: [83][ 790/ 1200] Overall Loss 0.322826 Objective Loss 0.322826 LR 0.001000 Time 0.020132 -2022-12-06 10:59:41,887 - Epoch: [83][ 800/ 1200] Overall Loss 0.322524 Objective Loss 0.322524 LR 0.001000 Time 0.020119 -2022-12-06 10:59:42,078 - Epoch: [83][ 810/ 1200] Overall Loss 0.322474 Objective Loss 0.322474 LR 0.001000 Time 0.020106 -2022-12-06 10:59:42,270 - Epoch: [83][ 820/ 1200] Overall Loss 0.322249 Objective Loss 0.322249 LR 0.001000 Time 0.020094 -2022-12-06 10:59:42,461 - Epoch: [83][ 830/ 1200] Overall Loss 0.322038 Objective Loss 0.322038 LR 0.001000 Time 0.020082 -2022-12-06 10:59:42,653 - Epoch: [83][ 840/ 1200] Overall Loss 0.322011 Objective Loss 0.322011 LR 0.001000 Time 0.020070 -2022-12-06 10:59:42,844 - Epoch: [83][ 850/ 1200] Overall Loss 0.322065 Objective Loss 0.322065 LR 0.001000 Time 0.020058 -2022-12-06 10:59:43,035 - Epoch: [83][ 860/ 1200] Overall Loss 0.322079 Objective Loss 0.322079 LR 0.001000 Time 0.020047 -2022-12-06 10:59:43,227 - Epoch: [83][ 870/ 1200] Overall Loss 0.321982 Objective Loss 0.321982 LR 0.001000 Time 0.020036 -2022-12-06 10:59:43,418 - Epoch: [83][ 880/ 1200] Overall Loss 0.321875 Objective Loss 0.321875 LR 0.001000 Time 0.020025 -2022-12-06 10:59:43,610 - Epoch: [83][ 890/ 1200] Overall Loss 0.321582 Objective Loss 0.321582 LR 0.001000 Time 0.020014 -2022-12-06 10:59:43,801 - Epoch: [83][ 900/ 1200] Overall Loss 0.321488 Objective Loss 0.321488 LR 0.001000 Time 0.020004 -2022-12-06 10:59:43,992 - Epoch: [83][ 910/ 1200] Overall Loss 0.321516 Objective Loss 0.321516 LR 0.001000 Time 0.019993 -2022-12-06 10:59:44,183 - Epoch: [83][ 920/ 1200] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.019984 -2022-12-06 10:59:44,375 - Epoch: [83][ 930/ 1200] Overall Loss 0.321539 Objective Loss 0.321539 LR 0.001000 Time 0.019974 -2022-12-06 10:59:44,566 - Epoch: [83][ 940/ 1200] Overall Loss 0.321510 Objective Loss 0.321510 LR 0.001000 Time 0.019964 -2022-12-06 10:59:44,757 - Epoch: [83][ 950/ 1200] Overall Loss 0.321183 Objective Loss 0.321183 LR 0.001000 Time 0.019955 -2022-12-06 10:59:44,948 - Epoch: [83][ 960/ 1200] Overall Loss 0.320942 Objective Loss 0.320942 LR 0.001000 Time 0.019945 -2022-12-06 10:59:45,140 - Epoch: [83][ 970/ 1200] Overall Loss 0.320670 Objective Loss 0.320670 LR 0.001000 Time 0.019937 -2022-12-06 10:59:45,331 - Epoch: [83][ 980/ 1200] Overall Loss 0.320575 Objective Loss 0.320575 LR 0.001000 Time 0.019928 -2022-12-06 10:59:45,522 - Epoch: [83][ 990/ 1200] Overall Loss 0.320491 Objective Loss 0.320491 LR 0.001000 Time 0.019919 -2022-12-06 10:59:45,714 - Epoch: [83][ 1000/ 1200] Overall Loss 0.320513 Objective Loss 0.320513 LR 0.001000 Time 0.019911 -2022-12-06 10:59:45,905 - Epoch: [83][ 1010/ 1200] Overall Loss 0.320674 Objective Loss 0.320674 LR 0.001000 Time 0.019903 -2022-12-06 10:59:46,096 - Epoch: [83][ 1020/ 1200] Overall Loss 0.320745 Objective Loss 0.320745 LR 0.001000 Time 0.019894 -2022-12-06 10:59:46,288 - Epoch: [83][ 1030/ 1200] Overall Loss 0.320805 Objective Loss 0.320805 LR 0.001000 Time 0.019887 -2022-12-06 10:59:46,479 - Epoch: [83][ 1040/ 1200] Overall Loss 0.320666 Objective Loss 0.320666 LR 0.001000 Time 0.019879 -2022-12-06 10:59:46,670 - Epoch: [83][ 1050/ 1200] Overall Loss 0.320459 Objective Loss 0.320459 LR 0.001000 Time 0.019871 -2022-12-06 10:59:46,862 - Epoch: [83][ 1060/ 1200] Overall Loss 0.320206 Objective Loss 0.320206 LR 0.001000 Time 0.019864 -2022-12-06 10:59:47,054 - Epoch: [83][ 1070/ 1200] Overall Loss 0.319979 Objective Loss 0.319979 LR 0.001000 Time 0.019857 -2022-12-06 10:59:47,246 - Epoch: [83][ 1080/ 1200] Overall Loss 0.319810 Objective Loss 0.319810 LR 0.001000 Time 0.019851 -2022-12-06 10:59:47,438 - Epoch: [83][ 1090/ 1200] Overall Loss 0.319908 Objective Loss 0.319908 LR 0.001000 Time 0.019844 -2022-12-06 10:59:47,630 - Epoch: [83][ 1100/ 1200] Overall Loss 0.319806 Objective Loss 0.319806 LR 0.001000 Time 0.019837 -2022-12-06 10:59:47,821 - Epoch: [83][ 1110/ 1200] Overall Loss 0.319606 Objective Loss 0.319606 LR 0.001000 Time 0.019830 -2022-12-06 10:59:48,013 - Epoch: [83][ 1120/ 1200] Overall Loss 0.319407 Objective Loss 0.319407 LR 0.001000 Time 0.019825 -2022-12-06 10:59:48,205 - Epoch: [83][ 1130/ 1200] Overall Loss 0.319489 Objective Loss 0.319489 LR 0.001000 Time 0.019818 -2022-12-06 10:59:48,397 - Epoch: [83][ 1140/ 1200] Overall Loss 0.319633 Objective Loss 0.319633 LR 0.001000 Time 0.019812 -2022-12-06 10:59:48,588 - Epoch: [83][ 1150/ 1200] Overall Loss 0.319453 Objective Loss 0.319453 LR 0.001000 Time 0.019806 -2022-12-06 10:59:48,780 - Epoch: [83][ 1160/ 1200] Overall Loss 0.319464 Objective Loss 0.319464 LR 0.001000 Time 0.019800 -2022-12-06 10:59:48,972 - Epoch: [83][ 1170/ 1200] Overall Loss 0.319423 Objective Loss 0.319423 LR 0.001000 Time 0.019794 -2022-12-06 10:59:49,163 - Epoch: [83][ 1180/ 1200] Overall Loss 0.319425 Objective Loss 0.319425 LR 0.001000 Time 0.019788 -2022-12-06 10:59:49,354 - Epoch: [83][ 1190/ 1200] Overall Loss 0.319425 Objective Loss 0.319425 LR 0.001000 Time 0.019782 -2022-12-06 10:59:49,588 - Epoch: [83][ 1200/ 1200] Overall Loss 0.319333 Objective Loss 0.319333 Top1 86.610879 Top5 98.953975 LR 0.001000 Time 0.019812 -2022-12-06 10:59:49,679 - --- validate (epoch=83)----------- -2022-12-06 10:59:49,679 - 34129 samples (256 per mini-batch) -2022-12-06 10:59:50,123 - Epoch: [83][ 10/ 134] Loss 0.336032 Top1 83.046875 Top5 97.812500 -2022-12-06 10:59:50,249 - Epoch: [83][ 20/ 134] Loss 0.349942 Top1 82.871094 Top5 97.695312 -2022-12-06 10:59:50,378 - Epoch: [83][ 30/ 134] Loss 0.336724 Top1 83.294271 Top5 97.708333 -2022-12-06 10:59:50,513 - Epoch: [83][ 40/ 134] Loss 0.341857 Top1 83.476562 Top5 97.705078 -2022-12-06 10:59:50,657 - Epoch: [83][ 50/ 134] Loss 0.330000 Top1 83.812500 Top5 97.835938 -2022-12-06 10:59:50,781 - Epoch: [83][ 60/ 134] Loss 0.329818 Top1 83.704427 Top5 97.805990 -2022-12-06 10:59:50,909 - Epoch: [83][ 70/ 134] Loss 0.330630 Top1 83.683036 Top5 97.829241 -2022-12-06 10:59:51,034 - Epoch: [83][ 80/ 134] Loss 0.331561 Top1 83.652344 Top5 97.802734 -2022-12-06 10:59:51,160 - Epoch: [83][ 90/ 134] Loss 0.331639 Top1 83.823785 Top5 97.769097 -2022-12-06 10:59:51,287 - Epoch: [83][ 100/ 134] Loss 0.328249 Top1 84.003906 Top5 97.808594 -2022-12-06 10:59:51,411 - Epoch: [83][ 110/ 134] Loss 0.329122 Top1 84.041193 Top5 97.794744 -2022-12-06 10:59:51,536 - Epoch: [83][ 120/ 134] Loss 0.327407 Top1 84.127604 Top5 97.805990 -2022-12-06 10:59:51,661 - Epoch: [83][ 130/ 134] Loss 0.327459 Top1 84.170673 Top5 97.809495 -2022-12-06 10:59:51,697 - Epoch: [83][ 134/ 134] Loss 0.327418 Top1 84.171819 Top5 97.817106 -2022-12-06 10:59:51,784 - ==> Top1: 84.172 Top5: 97.817 Loss: 0.327 - -2022-12-06 10:59:51,785 - ==> Confusion: -[[ 892 1 3 3 5 6 0 0 7 62 0 1 2 1 6 2 1 2 1 0 1] - [ 1 934 2 5 4 27 1 8 2 0 0 6 1 1 5 2 6 1 11 2 8] - [ 5 3 986 17 3 4 27 6 1 4 6 7 2 2 3 6 1 2 5 4 9] - [ 2 2 13 934 1 6 1 0 0 0 9 0 6 0 24 1 1 3 10 0 7] - [ 11 3 2 0 952 6 0 0 1 6 1 6 0 2 12 6 5 3 1 0 3] - [ 3 13 1 2 7 968 2 14 2 1 2 18 1 15 1 3 4 1 2 6 3] - [ 1 1 9 2 0 2 1072 3 0 0 0 3 1 1 0 7 0 2 0 9 5] - [ 0 12 11 3 1 29 8 919 0 2 3 13 0 1 2 2 2 1 29 13 3] - [ 3 2 0 1 1 3 2 1 978 42 7 1 1 9 9 0 1 0 2 0 1] - [ 61 1 0 0 6 2 0 1 27 877 1 1 1 11 3 0 0 3 0 0 6] - [ 1 3 3 5 1 1 2 2 11 3 938 3 4 21 6 0 1 0 8 2 4] - [ 1 0 2 1 0 9 4 1 1 2 1 987 14 3 1 11 2 4 0 6 1] - [ 1 1 1 3 0 3 0 0 0 0 0 61 864 0 0 12 1 13 0 3 6] - [ 0 0 1 0 0 3 0 1 13 15 5 7 4 956 3 3 2 1 0 2 7] - [ 5 4 2 9 4 2 0 0 11 5 0 4 1 4 1065 2 1 0 4 0 7] - [ 0 1 1 1 2 3 1 0 0 0 1 12 4 0 0 1004 1 4 0 5 3] - [ 2 3 2 2 2 1 1 0 0 1 0 6 1 1 0 17 1018 1 2 7 5] - [ 2 0 1 3 0 2 1 1 0 3 0 15 13 2 2 32 1 954 0 1 3] - [ 4 5 4 13 0 3 2 20 2 0 7 1 5 3 16 2 0 1 913 3 4] - [ 1 7 2 1 0 3 9 8 0 0 0 24 6 5 1 8 3 6 0 991 5] - [ 140 249 185 96 93 200 99 139 113 98 188 176 431 351 207 192 209 79 156 304 9521]] - -2022-12-06 10:59:52,352 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 10:59:52,352 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 10:59:52,358 - - -2022-12-06 10:59:52,358 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 10:59:53,274 - Epoch: [84][ 10/ 1200] Overall Loss 0.303121 Objective Loss 0.303121 LR 0.001000 Time 0.091471 -2022-12-06 10:59:53,465 - Epoch: [84][ 20/ 1200] Overall Loss 0.309592 Objective Loss 0.309592 LR 0.001000 Time 0.055252 -2022-12-06 10:59:53,655 - Epoch: [84][ 30/ 1200] Overall Loss 0.317411 Objective Loss 0.317411 LR 0.001000 Time 0.043167 -2022-12-06 10:59:53,844 - Epoch: [84][ 40/ 1200] Overall Loss 0.316846 Objective Loss 0.316846 LR 0.001000 Time 0.037089 -2022-12-06 10:59:54,035 - Epoch: [84][ 50/ 1200] Overall Loss 0.313836 Objective Loss 0.313836 LR 0.001000 Time 0.033476 -2022-12-06 10:59:54,225 - Epoch: [84][ 60/ 1200] Overall Loss 0.313107 Objective Loss 0.313107 LR 0.001000 Time 0.031048 -2022-12-06 10:59:54,415 - Epoch: [84][ 70/ 1200] Overall Loss 0.309178 Objective Loss 0.309178 LR 0.001000 Time 0.029317 -2022-12-06 10:59:54,604 - Epoch: [84][ 80/ 1200] Overall Loss 0.308257 Objective Loss 0.308257 LR 0.001000 Time 0.028013 -2022-12-06 10:59:54,794 - Epoch: [84][ 90/ 1200] Overall Loss 0.307062 Objective Loss 0.307062 LR 0.001000 Time 0.027009 -2022-12-06 10:59:54,985 - Epoch: [84][ 100/ 1200] Overall Loss 0.307652 Objective Loss 0.307652 LR 0.001000 Time 0.026208 -2022-12-06 10:59:55,174 - Epoch: [84][ 110/ 1200] Overall Loss 0.306777 Objective Loss 0.306777 LR 0.001000 Time 0.025542 -2022-12-06 10:59:55,365 - Epoch: [84][ 120/ 1200] Overall Loss 0.306496 Objective Loss 0.306496 LR 0.001000 Time 0.024994 -2022-12-06 10:59:55,554 - Epoch: [84][ 130/ 1200] Overall Loss 0.304671 Objective Loss 0.304671 LR 0.001000 Time 0.024524 -2022-12-06 10:59:55,743 - Epoch: [84][ 140/ 1200] Overall Loss 0.305260 Objective Loss 0.305260 LR 0.001000 Time 0.024122 -2022-12-06 10:59:55,932 - Epoch: [84][ 150/ 1200] Overall Loss 0.304405 Objective Loss 0.304405 LR 0.001000 Time 0.023770 -2022-12-06 10:59:56,122 - Epoch: [84][ 160/ 1200] Overall Loss 0.304828 Objective Loss 0.304828 LR 0.001000 Time 0.023467 -2022-12-06 10:59:56,312 - Epoch: [84][ 170/ 1200] Overall Loss 0.303517 Objective Loss 0.303517 LR 0.001000 Time 0.023202 -2022-12-06 10:59:56,502 - Epoch: [84][ 180/ 1200] Overall Loss 0.304173 Objective Loss 0.304173 LR 0.001000 Time 0.022963 -2022-12-06 10:59:56,693 - Epoch: [84][ 190/ 1200] Overall Loss 0.304137 Objective Loss 0.304137 LR 0.001000 Time 0.022755 -2022-12-06 10:59:56,882 - Epoch: [84][ 200/ 1200] Overall Loss 0.303692 Objective Loss 0.303692 LR 0.001000 Time 0.022561 -2022-12-06 10:59:57,072 - Epoch: [84][ 210/ 1200] Overall Loss 0.304577 Objective Loss 0.304577 LR 0.001000 Time 0.022388 -2022-12-06 10:59:57,262 - Epoch: [84][ 220/ 1200] Overall Loss 0.304518 Objective Loss 0.304518 LR 0.001000 Time 0.022231 -2022-12-06 10:59:57,451 - Epoch: [84][ 230/ 1200] Overall Loss 0.305084 Objective Loss 0.305084 LR 0.001000 Time 0.022084 -2022-12-06 10:59:57,640 - Epoch: [84][ 240/ 1200] Overall Loss 0.303346 Objective Loss 0.303346 LR 0.001000 Time 0.021952 -2022-12-06 10:59:57,830 - Epoch: [84][ 250/ 1200] Overall Loss 0.303509 Objective Loss 0.303509 LR 0.001000 Time 0.021831 -2022-12-06 10:59:58,019 - Epoch: [84][ 260/ 1200] Overall Loss 0.303644 Objective Loss 0.303644 LR 0.001000 Time 0.021717 -2022-12-06 10:59:58,210 - Epoch: [84][ 270/ 1200] Overall Loss 0.303870 Objective Loss 0.303870 LR 0.001000 Time 0.021616 -2022-12-06 10:59:58,401 - Epoch: [84][ 280/ 1200] Overall Loss 0.304353 Objective Loss 0.304353 LR 0.001000 Time 0.021523 -2022-12-06 10:59:58,591 - Epoch: [84][ 290/ 1200] Overall Loss 0.304091 Objective Loss 0.304091 LR 0.001000 Time 0.021434 -2022-12-06 10:59:58,780 - Epoch: [84][ 300/ 1200] Overall Loss 0.303485 Objective Loss 0.303485 LR 0.001000 Time 0.021348 -2022-12-06 10:59:58,970 - Epoch: [84][ 310/ 1200] Overall Loss 0.303325 Objective Loss 0.303325 LR 0.001000 Time 0.021272 -2022-12-06 10:59:59,160 - Epoch: [84][ 320/ 1200] Overall Loss 0.303276 Objective Loss 0.303276 LR 0.001000 Time 0.021197 -2022-12-06 10:59:59,350 - Epoch: [84][ 330/ 1200] Overall Loss 0.303444 Objective Loss 0.303444 LR 0.001000 Time 0.021129 -2022-12-06 10:59:59,539 - Epoch: [84][ 340/ 1200] Overall Loss 0.302998 Objective Loss 0.302998 LR 0.001000 Time 0.021063 -2022-12-06 10:59:59,730 - Epoch: [84][ 350/ 1200] Overall Loss 0.302538 Objective Loss 0.302538 LR 0.001000 Time 0.021005 -2022-12-06 10:59:59,920 - Epoch: [84][ 360/ 1200] Overall Loss 0.303304 Objective Loss 0.303304 LR 0.001000 Time 0.020948 -2022-12-06 11:00:00,110 - Epoch: [84][ 370/ 1200] Overall Loss 0.303575 Objective Loss 0.303575 LR 0.001000 Time 0.020894 -2022-12-06 11:00:00,301 - Epoch: [84][ 380/ 1200] Overall Loss 0.303673 Objective Loss 0.303673 LR 0.001000 Time 0.020845 -2022-12-06 11:00:00,490 - Epoch: [84][ 390/ 1200] Overall Loss 0.304103 Objective Loss 0.304103 LR 0.001000 Time 0.020795 -2022-12-06 11:00:00,680 - Epoch: [84][ 400/ 1200] Overall Loss 0.304270 Objective Loss 0.304270 LR 0.001000 Time 0.020747 -2022-12-06 11:00:00,870 - Epoch: [84][ 410/ 1200] Overall Loss 0.304048 Objective Loss 0.304048 LR 0.001000 Time 0.020702 -2022-12-06 11:00:01,059 - Epoch: [84][ 420/ 1200] Overall Loss 0.304258 Objective Loss 0.304258 LR 0.001000 Time 0.020659 -2022-12-06 11:00:01,248 - Epoch: [84][ 430/ 1200] Overall Loss 0.304576 Objective Loss 0.304576 LR 0.001000 Time 0.020617 -2022-12-06 11:00:01,437 - Epoch: [84][ 440/ 1200] Overall Loss 0.304620 Objective Loss 0.304620 LR 0.001000 Time 0.020578 -2022-12-06 11:00:01,627 - Epoch: [84][ 450/ 1200] Overall Loss 0.304271 Objective Loss 0.304271 LR 0.001000 Time 0.020542 -2022-12-06 11:00:01,818 - Epoch: [84][ 460/ 1200] Overall Loss 0.304501 Objective Loss 0.304501 LR 0.001000 Time 0.020507 -2022-12-06 11:00:02,008 - Epoch: [84][ 470/ 1200] Overall Loss 0.304765 Objective Loss 0.304765 LR 0.001000 Time 0.020475 -2022-12-06 11:00:02,198 - Epoch: [84][ 480/ 1200] Overall Loss 0.304814 Objective Loss 0.304814 LR 0.001000 Time 0.020443 -2022-12-06 11:00:02,389 - Epoch: [84][ 490/ 1200] Overall Loss 0.304850 Objective Loss 0.304850 LR 0.001000 Time 0.020414 -2022-12-06 11:00:02,579 - Epoch: [84][ 500/ 1200] Overall Loss 0.304863 Objective Loss 0.304863 LR 0.001000 Time 0.020384 -2022-12-06 11:00:02,769 - Epoch: [84][ 510/ 1200] Overall Loss 0.304984 Objective Loss 0.304984 LR 0.001000 Time 0.020356 -2022-12-06 11:00:02,958 - Epoch: [84][ 520/ 1200] Overall Loss 0.305466 Objective Loss 0.305466 LR 0.001000 Time 0.020328 -2022-12-06 11:00:03,148 - Epoch: [84][ 530/ 1200] Overall Loss 0.305813 Objective Loss 0.305813 LR 0.001000 Time 0.020302 -2022-12-06 11:00:03,339 - Epoch: [84][ 540/ 1200] Overall Loss 0.305950 Objective Loss 0.305950 LR 0.001000 Time 0.020278 -2022-12-06 11:00:03,529 - Epoch: [84][ 550/ 1200] Overall Loss 0.306124 Objective Loss 0.306124 LR 0.001000 Time 0.020254 -2022-12-06 11:00:03,719 - Epoch: [84][ 560/ 1200] Overall Loss 0.305881 Objective Loss 0.305881 LR 0.001000 Time 0.020231 -2022-12-06 11:00:03,910 - Epoch: [84][ 570/ 1200] Overall Loss 0.305519 Objective Loss 0.305519 LR 0.001000 Time 0.020209 -2022-12-06 11:00:04,099 - Epoch: [84][ 580/ 1200] Overall Loss 0.305626 Objective Loss 0.305626 LR 0.001000 Time 0.020186 -2022-12-06 11:00:04,290 - Epoch: [84][ 590/ 1200] Overall Loss 0.305854 Objective Loss 0.305854 LR 0.001000 Time 0.020166 -2022-12-06 11:00:04,479 - Epoch: [84][ 600/ 1200] Overall Loss 0.306253 Objective Loss 0.306253 LR 0.001000 Time 0.020144 -2022-12-06 11:00:04,668 - Epoch: [84][ 610/ 1200] Overall Loss 0.306283 Objective Loss 0.306283 LR 0.001000 Time 0.020124 -2022-12-06 11:00:04,858 - Epoch: [84][ 620/ 1200] Overall Loss 0.306259 Objective Loss 0.306259 LR 0.001000 Time 0.020104 -2022-12-06 11:00:05,048 - Epoch: [84][ 630/ 1200] Overall Loss 0.306068 Objective Loss 0.306068 LR 0.001000 Time 0.020086 -2022-12-06 11:00:05,238 - Epoch: [84][ 640/ 1200] Overall Loss 0.306011 Objective Loss 0.306011 LR 0.001000 Time 0.020067 -2022-12-06 11:00:05,428 - Epoch: [84][ 650/ 1200] Overall Loss 0.305542 Objective Loss 0.305542 LR 0.001000 Time 0.020051 -2022-12-06 11:00:05,619 - Epoch: [84][ 660/ 1200] Overall Loss 0.305003 Objective Loss 0.305003 LR 0.001000 Time 0.020035 -2022-12-06 11:00:05,808 - Epoch: [84][ 670/ 1200] Overall Loss 0.305401 Objective Loss 0.305401 LR 0.001000 Time 0.020018 -2022-12-06 11:00:05,998 - Epoch: [84][ 680/ 1200] Overall Loss 0.305186 Objective Loss 0.305186 LR 0.001000 Time 0.020001 -2022-12-06 11:00:06,188 - Epoch: [84][ 690/ 1200] Overall Loss 0.304995 Objective Loss 0.304995 LR 0.001000 Time 0.019986 -2022-12-06 11:00:06,377 - Epoch: [84][ 700/ 1200] Overall Loss 0.304794 Objective Loss 0.304794 LR 0.001000 Time 0.019971 -2022-12-06 11:00:06,568 - Epoch: [84][ 710/ 1200] Overall Loss 0.304573 Objective Loss 0.304573 LR 0.001000 Time 0.019957 -2022-12-06 11:00:06,758 - Epoch: [84][ 720/ 1200] Overall Loss 0.304884 Objective Loss 0.304884 LR 0.001000 Time 0.019943 -2022-12-06 11:00:06,948 - Epoch: [84][ 730/ 1200] Overall Loss 0.305000 Objective Loss 0.305000 LR 0.001000 Time 0.019929 -2022-12-06 11:00:07,138 - Epoch: [84][ 740/ 1200] Overall Loss 0.305003 Objective Loss 0.305003 LR 0.001000 Time 0.019916 -2022-12-06 11:00:07,330 - Epoch: [84][ 750/ 1200] Overall Loss 0.305157 Objective Loss 0.305157 LR 0.001000 Time 0.019905 -2022-12-06 11:00:07,519 - Epoch: [84][ 760/ 1200] Overall Loss 0.305463 Objective Loss 0.305463 LR 0.001000 Time 0.019892 -2022-12-06 11:00:07,710 - Epoch: [84][ 770/ 1200] Overall Loss 0.305726 Objective Loss 0.305726 LR 0.001000 Time 0.019880 -2022-12-06 11:00:07,899 - Epoch: [84][ 780/ 1200] Overall Loss 0.305580 Objective Loss 0.305580 LR 0.001000 Time 0.019868 -2022-12-06 11:00:08,088 - Epoch: [84][ 790/ 1200] Overall Loss 0.305064 Objective Loss 0.305064 LR 0.001000 Time 0.019855 -2022-12-06 11:00:08,279 - Epoch: [84][ 800/ 1200] Overall Loss 0.305117 Objective Loss 0.305117 LR 0.001000 Time 0.019844 -2022-12-06 11:00:08,468 - Epoch: [84][ 810/ 1200] Overall Loss 0.305092 Objective Loss 0.305092 LR 0.001000 Time 0.019832 -2022-12-06 11:00:08,658 - Epoch: [84][ 820/ 1200] Overall Loss 0.304932 Objective Loss 0.304932 LR 0.001000 Time 0.019821 -2022-12-06 11:00:08,848 - Epoch: [84][ 830/ 1200] Overall Loss 0.304780 Objective Loss 0.304780 LR 0.001000 Time 0.019810 -2022-12-06 11:00:09,037 - Epoch: [84][ 840/ 1200] Overall Loss 0.304440 Objective Loss 0.304440 LR 0.001000 Time 0.019799 -2022-12-06 11:00:09,227 - Epoch: [84][ 850/ 1200] Overall Loss 0.304424 Objective Loss 0.304424 LR 0.001000 Time 0.019789 -2022-12-06 11:00:09,417 - Epoch: [84][ 860/ 1200] Overall Loss 0.304302 Objective Loss 0.304302 LR 0.001000 Time 0.019779 -2022-12-06 11:00:09,607 - Epoch: [84][ 870/ 1200] Overall Loss 0.304199 Objective Loss 0.304199 LR 0.001000 Time 0.019770 -2022-12-06 11:00:09,796 - Epoch: [84][ 880/ 1200] Overall Loss 0.304280 Objective Loss 0.304280 LR 0.001000 Time 0.019759 -2022-12-06 11:00:09,986 - Epoch: [84][ 890/ 1200] Overall Loss 0.304160 Objective Loss 0.304160 LR 0.001000 Time 0.019750 -2022-12-06 11:00:10,175 - Epoch: [84][ 900/ 1200] Overall Loss 0.304069 Objective Loss 0.304069 LR 0.001000 Time 0.019740 -2022-12-06 11:00:10,366 - Epoch: [84][ 910/ 1200] Overall Loss 0.304048 Objective Loss 0.304048 LR 0.001000 Time 0.019732 -2022-12-06 11:00:10,555 - Epoch: [84][ 920/ 1200] Overall Loss 0.304228 Objective Loss 0.304228 LR 0.001000 Time 0.019722 -2022-12-06 11:00:10,744 - Epoch: [84][ 930/ 1200] Overall Loss 0.304268 Objective Loss 0.304268 LR 0.001000 Time 0.019713 -2022-12-06 11:00:10,934 - Epoch: [84][ 940/ 1200] Overall Loss 0.304445 Objective Loss 0.304445 LR 0.001000 Time 0.019705 -2022-12-06 11:00:11,124 - Epoch: [84][ 950/ 1200] Overall Loss 0.304832 Objective Loss 0.304832 LR 0.001000 Time 0.019697 -2022-12-06 11:00:11,315 - Epoch: [84][ 960/ 1200] Overall Loss 0.305019 Objective Loss 0.305019 LR 0.001000 Time 0.019690 -2022-12-06 11:00:11,505 - Epoch: [84][ 970/ 1200] Overall Loss 0.304790 Objective Loss 0.304790 LR 0.001000 Time 0.019682 -2022-12-06 11:00:11,694 - Epoch: [84][ 980/ 1200] Overall Loss 0.304768 Objective Loss 0.304768 LR 0.001000 Time 0.019674 -2022-12-06 11:00:11,885 - Epoch: [84][ 990/ 1200] Overall Loss 0.304632 Objective Loss 0.304632 LR 0.001000 Time 0.019667 -2022-12-06 11:00:12,074 - Epoch: [84][ 1000/ 1200] Overall Loss 0.304419 Objective Loss 0.304419 LR 0.001000 Time 0.019659 -2022-12-06 11:00:12,265 - Epoch: [84][ 1010/ 1200] Overall Loss 0.304569 Objective Loss 0.304569 LR 0.001000 Time 0.019653 -2022-12-06 11:00:12,454 - Epoch: [84][ 1020/ 1200] Overall Loss 0.304774 Objective Loss 0.304774 LR 0.001000 Time 0.019645 -2022-12-06 11:00:12,644 - Epoch: [84][ 1030/ 1200] Overall Loss 0.305110 Objective Loss 0.305110 LR 0.001000 Time 0.019639 -2022-12-06 11:00:12,834 - Epoch: [84][ 1040/ 1200] Overall Loss 0.305222 Objective Loss 0.305222 LR 0.001000 Time 0.019632 -2022-12-06 11:00:13,025 - Epoch: [84][ 1050/ 1200] Overall Loss 0.305476 Objective Loss 0.305476 LR 0.001000 Time 0.019626 -2022-12-06 11:00:13,214 - Epoch: [84][ 1060/ 1200] Overall Loss 0.305733 Objective Loss 0.305733 LR 0.001000 Time 0.019619 -2022-12-06 11:00:13,405 - Epoch: [84][ 1070/ 1200] Overall Loss 0.305572 Objective Loss 0.305572 LR 0.001000 Time 0.019613 -2022-12-06 11:00:13,595 - Epoch: [84][ 1080/ 1200] Overall Loss 0.305791 Objective Loss 0.305791 LR 0.001000 Time 0.019607 -2022-12-06 11:00:13,785 - Epoch: [84][ 1090/ 1200] Overall Loss 0.305967 Objective Loss 0.305967 LR 0.001000 Time 0.019601 -2022-12-06 11:00:13,974 - Epoch: [84][ 1100/ 1200] Overall Loss 0.305695 Objective Loss 0.305695 LR 0.001000 Time 0.019594 -2022-12-06 11:00:14,164 - Epoch: [84][ 1110/ 1200] Overall Loss 0.305730 Objective Loss 0.305730 LR 0.001000 Time 0.019588 -2022-12-06 11:00:14,354 - Epoch: [84][ 1120/ 1200] Overall Loss 0.305694 Objective Loss 0.305694 LR 0.001000 Time 0.019583 -2022-12-06 11:00:14,544 - Epoch: [84][ 1130/ 1200] Overall Loss 0.305727 Objective Loss 0.305727 LR 0.001000 Time 0.019577 -2022-12-06 11:00:14,734 - Epoch: [84][ 1140/ 1200] Overall Loss 0.305726 Objective Loss 0.305726 LR 0.001000 Time 0.019572 -2022-12-06 11:00:14,925 - Epoch: [84][ 1150/ 1200] Overall Loss 0.305677 Objective Loss 0.305677 LR 0.001000 Time 0.019567 -2022-12-06 11:00:15,115 - Epoch: [84][ 1160/ 1200] Overall Loss 0.305765 Objective Loss 0.305765 LR 0.001000 Time 0.019561 -2022-12-06 11:00:15,306 - Epoch: [84][ 1170/ 1200] Overall Loss 0.305355 Objective Loss 0.305355 LR 0.001000 Time 0.019557 -2022-12-06 11:00:15,497 - Epoch: [84][ 1180/ 1200] Overall Loss 0.305422 Objective Loss 0.305422 LR 0.001000 Time 0.019553 -2022-12-06 11:00:15,687 - Epoch: [84][ 1190/ 1200] Overall Loss 0.305420 Objective Loss 0.305420 LR 0.001000 Time 0.019547 -2022-12-06 11:00:15,918 - Epoch: [84][ 1200/ 1200] Overall Loss 0.305224 Objective Loss 0.305224 Top1 83.472803 Top5 98.535565 LR 0.001000 Time 0.019577 -2022-12-06 11:00:16,007 - --- validate (epoch=84)----------- -2022-12-06 11:00:16,007 - 34129 samples (256 per mini-batch) -2022-12-06 11:00:16,454 - Epoch: [84][ 10/ 134] Loss 0.291191 Top1 84.648438 Top5 97.929688 -2022-12-06 11:00:16,597 - Epoch: [84][ 20/ 134] Loss 0.307556 Top1 84.804688 Top5 97.929688 -2022-12-06 11:00:16,739 - Epoch: [84][ 30/ 134] Loss 0.301354 Top1 84.895833 Top5 98.033854 -2022-12-06 11:00:16,868 - Epoch: [84][ 40/ 134] Loss 0.304810 Top1 84.580078 Top5 97.900391 -2022-12-06 11:00:16,999 - Epoch: [84][ 50/ 134] Loss 0.311938 Top1 84.500000 Top5 97.835938 -2022-12-06 11:00:17,145 - Epoch: [84][ 60/ 134] Loss 0.315585 Top1 84.375000 Top5 97.812500 -2022-12-06 11:00:17,281 - Epoch: [84][ 70/ 134] Loss 0.315311 Top1 84.441964 Top5 97.812500 -2022-12-06 11:00:17,427 - Epoch: [84][ 80/ 134] Loss 0.315186 Top1 84.448242 Top5 97.822266 -2022-12-06 11:00:17,566 - Epoch: [84][ 90/ 134] Loss 0.313611 Top1 84.466146 Top5 97.868924 -2022-12-06 11:00:17,713 - Epoch: [84][ 100/ 134] Loss 0.314795 Top1 84.492188 Top5 97.886719 -2022-12-06 11:00:17,852 - Epoch: [84][ 110/ 134] Loss 0.313165 Top1 84.566761 Top5 97.940341 -2022-12-06 11:00:18,000 - Epoch: [84][ 120/ 134] Loss 0.316385 Top1 84.531250 Top5 97.900391 -2022-12-06 11:00:18,133 - Epoch: [84][ 130/ 134] Loss 0.314058 Top1 84.609375 Top5 97.905649 -2022-12-06 11:00:18,169 - Epoch: [84][ 134/ 134] Loss 0.315557 Top1 84.535146 Top5 97.893287 -2022-12-06 11:00:18,257 - ==> Top1: 84.535 Top5: 97.893 Loss: 0.316 - -2022-12-06 11:00:18,258 - ==> Confusion: -[[ 912 0 3 2 2 5 0 0 3 49 0 2 2 4 5 2 1 1 0 0 3] - [ 4 902 0 2 9 38 3 18 3 0 4 6 3 4 1 2 5 1 12 2 8] - [ 9 3 967 20 1 1 31 11 0 2 8 4 3 5 4 7 3 5 2 3 14] - [ 6 2 12 928 1 4 0 1 2 0 16 1 5 3 15 2 2 2 13 0 5] - [ 11 4 2 0 953 3 2 0 0 7 1 4 1 2 11 6 8 2 0 1 2] - [ 7 9 0 2 5 982 5 12 2 4 2 9 2 12 3 1 2 1 1 4 4] - [ 1 1 3 0 0 1 1072 3 0 1 2 4 1 2 0 8 1 4 1 12 1] - [ 1 5 8 2 2 38 13 929 0 1 2 10 1 3 1 3 0 2 20 9 4] - [ 7 2 0 1 0 3 1 0 958 45 8 1 1 19 8 1 2 1 3 0 3] - [ 66 2 0 0 2 2 0 0 18 886 2 2 0 8 2 1 0 1 1 1 7] - [ 1 0 7 3 2 1 0 1 15 1 958 3 2 13 3 1 0 0 4 2 2] - [ 3 0 1 1 0 7 1 2 1 0 0 949 39 6 0 6 3 12 0 16 4] - [ 1 0 0 6 0 2 0 0 0 0 0 23 898 3 0 8 1 19 0 3 5] - [ 0 1 1 0 0 7 0 3 10 12 3 4 3 965 2 2 2 0 0 4 4] - [ 12 3 1 9 4 1 0 1 26 7 3 3 2 5 1033 2 4 2 5 0 7] - [ 1 0 1 0 2 2 4 0 0 0 0 7 7 3 0 988 6 16 0 4 2] - [ 5 1 0 1 2 2 2 0 2 0 0 4 2 3 1 14 1020 1 0 8 4] - [ 3 0 1 3 1 0 1 0 2 0 1 5 21 5 1 10 1 977 1 0 3] - [ 3 5 5 18 0 3 2 28 2 0 12 2 1 2 12 1 0 0 906 2 4] - [ 3 2 2 1 1 4 5 4 0 0 0 10 6 7 0 6 3 5 1 1017 3] - [ 142 177 158 118 89 232 88 134 98 96 250 111 437 407 134 160 200 115 141 295 9644]] - -2022-12-06 11:00:18,916 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:00:18,916 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:00:18,922 - - -2022-12-06 11:00:18,922 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:00:19,873 - Epoch: [85][ 10/ 1200] Overall Loss 0.294760 Objective Loss 0.294760 LR 0.001000 Time 0.094984 -2022-12-06 11:00:20,073 - Epoch: [85][ 20/ 1200] Overall Loss 0.288112 Objective Loss 0.288112 LR 0.001000 Time 0.057484 -2022-12-06 11:00:20,268 - Epoch: [85][ 30/ 1200] Overall Loss 0.288521 Objective Loss 0.288521 LR 0.001000 Time 0.044787 -2022-12-06 11:00:20,462 - Epoch: [85][ 40/ 1200] Overall Loss 0.292444 Objective Loss 0.292444 LR 0.001000 Time 0.038444 -2022-12-06 11:00:20,656 - Epoch: [85][ 50/ 1200] Overall Loss 0.289671 Objective Loss 0.289671 LR 0.001000 Time 0.034620 -2022-12-06 11:00:20,851 - Epoch: [85][ 60/ 1200] Overall Loss 0.288360 Objective Loss 0.288360 LR 0.001000 Time 0.032080 -2022-12-06 11:00:21,045 - Epoch: [85][ 70/ 1200] Overall Loss 0.292275 Objective Loss 0.292275 LR 0.001000 Time 0.030266 -2022-12-06 11:00:21,239 - Epoch: [85][ 80/ 1200] Overall Loss 0.290687 Objective Loss 0.290687 LR 0.001000 Time 0.028903 -2022-12-06 11:00:21,433 - Epoch: [85][ 90/ 1200] Overall Loss 0.292282 Objective Loss 0.292282 LR 0.001000 Time 0.027845 -2022-12-06 11:00:21,628 - Epoch: [85][ 100/ 1200] Overall Loss 0.292841 Objective Loss 0.292841 LR 0.001000 Time 0.026998 -2022-12-06 11:00:21,822 - Epoch: [85][ 110/ 1200] Overall Loss 0.291982 Objective Loss 0.291982 LR 0.001000 Time 0.026302 -2022-12-06 11:00:22,015 - Epoch: [85][ 120/ 1200] Overall Loss 0.292411 Objective Loss 0.292411 LR 0.001000 Time 0.025715 -2022-12-06 11:00:22,209 - Epoch: [85][ 130/ 1200] Overall Loss 0.293059 Objective Loss 0.293059 LR 0.001000 Time 0.025228 -2022-12-06 11:00:22,403 - Epoch: [85][ 140/ 1200] Overall Loss 0.292968 Objective Loss 0.292968 LR 0.001000 Time 0.024808 -2022-12-06 11:00:22,598 - Epoch: [85][ 150/ 1200] Overall Loss 0.291961 Objective Loss 0.291961 LR 0.001000 Time 0.024446 -2022-12-06 11:00:22,792 - Epoch: [85][ 160/ 1200] Overall Loss 0.292089 Objective Loss 0.292089 LR 0.001000 Time 0.024126 -2022-12-06 11:00:22,986 - Epoch: [85][ 170/ 1200] Overall Loss 0.292234 Objective Loss 0.292234 LR 0.001000 Time 0.023849 -2022-12-06 11:00:23,180 - Epoch: [85][ 180/ 1200] Overall Loss 0.291925 Objective Loss 0.291925 LR 0.001000 Time 0.023596 -2022-12-06 11:00:23,373 - Epoch: [85][ 190/ 1200] Overall Loss 0.292469 Objective Loss 0.292469 LR 0.001000 Time 0.023370 -2022-12-06 11:00:23,567 - Epoch: [85][ 200/ 1200] Overall Loss 0.290985 Objective Loss 0.290985 LR 0.001000 Time 0.023168 -2022-12-06 11:00:23,761 - Epoch: [85][ 210/ 1200] Overall Loss 0.290846 Objective Loss 0.290846 LR 0.001000 Time 0.022986 -2022-12-06 11:00:23,955 - Epoch: [85][ 220/ 1200] Overall Loss 0.290009 Objective Loss 0.290009 LR 0.001000 Time 0.022818 -2022-12-06 11:00:24,149 - Epoch: [85][ 230/ 1200] Overall Loss 0.290661 Objective Loss 0.290661 LR 0.001000 Time 0.022667 -2022-12-06 11:00:24,342 - Epoch: [85][ 240/ 1200] Overall Loss 0.290370 Objective Loss 0.290370 LR 0.001000 Time 0.022527 -2022-12-06 11:00:24,536 - Epoch: [85][ 250/ 1200] Overall Loss 0.290628 Objective Loss 0.290628 LR 0.001000 Time 0.022400 -2022-12-06 11:00:24,730 - Epoch: [85][ 260/ 1200] Overall Loss 0.291498 Objective Loss 0.291498 LR 0.001000 Time 0.022282 -2022-12-06 11:00:24,924 - Epoch: [85][ 270/ 1200] Overall Loss 0.291354 Objective Loss 0.291354 LR 0.001000 Time 0.022174 -2022-12-06 11:00:25,118 - Epoch: [85][ 280/ 1200] Overall Loss 0.290767 Objective Loss 0.290767 LR 0.001000 Time 0.022070 -2022-12-06 11:00:25,312 - Epoch: [85][ 290/ 1200] Overall Loss 0.290830 Objective Loss 0.290830 LR 0.001000 Time 0.021979 -2022-12-06 11:00:25,506 - Epoch: [85][ 300/ 1200] Overall Loss 0.290877 Objective Loss 0.290877 LR 0.001000 Time 0.021890 -2022-12-06 11:00:25,700 - Epoch: [85][ 310/ 1200] Overall Loss 0.290832 Objective Loss 0.290832 LR 0.001000 Time 0.021809 -2022-12-06 11:00:25,894 - Epoch: [85][ 320/ 1200] Overall Loss 0.291285 Objective Loss 0.291285 LR 0.001000 Time 0.021729 -2022-12-06 11:00:26,088 - Epoch: [85][ 330/ 1200] Overall Loss 0.291902 Objective Loss 0.291902 LR 0.001000 Time 0.021657 -2022-12-06 11:00:26,281 - Epoch: [85][ 340/ 1200] Overall Loss 0.292627 Objective Loss 0.292627 LR 0.001000 Time 0.021587 -2022-12-06 11:00:26,476 - Epoch: [85][ 350/ 1200] Overall Loss 0.292119 Objective Loss 0.292119 LR 0.001000 Time 0.021525 -2022-12-06 11:00:26,669 - Epoch: [85][ 360/ 1200] Overall Loss 0.292306 Objective Loss 0.292306 LR 0.001000 Time 0.021464 -2022-12-06 11:00:26,863 - Epoch: [85][ 370/ 1200] Overall Loss 0.292463 Objective Loss 0.292463 LR 0.001000 Time 0.021406 -2022-12-06 11:00:27,057 - Epoch: [85][ 380/ 1200] Overall Loss 0.291932 Objective Loss 0.291932 LR 0.001000 Time 0.021352 -2022-12-06 11:00:27,252 - Epoch: [85][ 390/ 1200] Overall Loss 0.292438 Objective Loss 0.292438 LR 0.001000 Time 0.021301 -2022-12-06 11:00:27,446 - Epoch: [85][ 400/ 1200] Overall Loss 0.292456 Objective Loss 0.292456 LR 0.001000 Time 0.021252 -2022-12-06 11:00:27,640 - Epoch: [85][ 410/ 1200] Overall Loss 0.292596 Objective Loss 0.292596 LR 0.001000 Time 0.021207 -2022-12-06 11:00:27,834 - Epoch: [85][ 420/ 1200] Overall Loss 0.292933 Objective Loss 0.292933 LR 0.001000 Time 0.021162 -2022-12-06 11:00:28,028 - Epoch: [85][ 430/ 1200] Overall Loss 0.292748 Objective Loss 0.292748 LR 0.001000 Time 0.021120 -2022-12-06 11:00:28,222 - Epoch: [85][ 440/ 1200] Overall Loss 0.292193 Objective Loss 0.292193 LR 0.001000 Time 0.021079 -2022-12-06 11:00:28,417 - Epoch: [85][ 450/ 1200] Overall Loss 0.292960 Objective Loss 0.292960 LR 0.001000 Time 0.021043 -2022-12-06 11:00:28,611 - Epoch: [85][ 460/ 1200] Overall Loss 0.293173 Objective Loss 0.293173 LR 0.001000 Time 0.021006 -2022-12-06 11:00:28,805 - Epoch: [85][ 470/ 1200] Overall Loss 0.292827 Objective Loss 0.292827 LR 0.001000 Time 0.020971 -2022-12-06 11:00:28,999 - Epoch: [85][ 480/ 1200] Overall Loss 0.292542 Objective Loss 0.292542 LR 0.001000 Time 0.020938 -2022-12-06 11:00:29,193 - Epoch: [85][ 490/ 1200] Overall Loss 0.291943 Objective Loss 0.291943 LR 0.001000 Time 0.020905 -2022-12-06 11:00:29,387 - Epoch: [85][ 500/ 1200] Overall Loss 0.291939 Objective Loss 0.291939 LR 0.001000 Time 0.020874 -2022-12-06 11:00:29,581 - Epoch: [85][ 510/ 1200] Overall Loss 0.292413 Objective Loss 0.292413 LR 0.001000 Time 0.020843 -2022-12-06 11:00:29,776 - Epoch: [85][ 520/ 1200] Overall Loss 0.292556 Objective Loss 0.292556 LR 0.001000 Time 0.020817 -2022-12-06 11:00:29,971 - Epoch: [85][ 530/ 1200] Overall Loss 0.292674 Objective Loss 0.292674 LR 0.001000 Time 0.020791 -2022-12-06 11:00:30,165 - Epoch: [85][ 540/ 1200] Overall Loss 0.292596 Objective Loss 0.292596 LR 0.001000 Time 0.020764 -2022-12-06 11:00:30,359 - Epoch: [85][ 550/ 1200] Overall Loss 0.292782 Objective Loss 0.292782 LR 0.001000 Time 0.020738 -2022-12-06 11:00:30,553 - Epoch: [85][ 560/ 1200] Overall Loss 0.292643 Objective Loss 0.292643 LR 0.001000 Time 0.020713 -2022-12-06 11:00:30,748 - Epoch: [85][ 570/ 1200] Overall Loss 0.292233 Objective Loss 0.292233 LR 0.001000 Time 0.020690 -2022-12-06 11:00:30,941 - Epoch: [85][ 580/ 1200] Overall Loss 0.292188 Objective Loss 0.292188 LR 0.001000 Time 0.020666 -2022-12-06 11:00:31,136 - Epoch: [85][ 590/ 1200] Overall Loss 0.291936 Objective Loss 0.291936 LR 0.001000 Time 0.020645 -2022-12-06 11:00:31,330 - Epoch: [85][ 600/ 1200] Overall Loss 0.292205 Objective Loss 0.292205 LR 0.001000 Time 0.020623 -2022-12-06 11:00:31,525 - Epoch: [85][ 610/ 1200] Overall Loss 0.292162 Objective Loss 0.292162 LR 0.001000 Time 0.020604 -2022-12-06 11:00:31,718 - Epoch: [85][ 620/ 1200] Overall Loss 0.292262 Objective Loss 0.292262 LR 0.001000 Time 0.020583 -2022-12-06 11:00:31,912 - Epoch: [85][ 630/ 1200] Overall Loss 0.292013 Objective Loss 0.292013 LR 0.001000 Time 0.020563 -2022-12-06 11:00:32,106 - Epoch: [85][ 640/ 1200] Overall Loss 0.292125 Objective Loss 0.292125 LR 0.001000 Time 0.020543 -2022-12-06 11:00:32,300 - Epoch: [85][ 650/ 1200] Overall Loss 0.291875 Objective Loss 0.291875 LR 0.001000 Time 0.020526 -2022-12-06 11:00:32,494 - Epoch: [85][ 660/ 1200] Overall Loss 0.291964 Objective Loss 0.291964 LR 0.001000 Time 0.020508 -2022-12-06 11:00:32,689 - Epoch: [85][ 670/ 1200] Overall Loss 0.292325 Objective Loss 0.292325 LR 0.001000 Time 0.020492 -2022-12-06 11:00:32,882 - Epoch: [85][ 680/ 1200] Overall Loss 0.292313 Objective Loss 0.292313 LR 0.001000 Time 0.020473 -2022-12-06 11:00:33,077 - Epoch: [85][ 690/ 1200] Overall Loss 0.292419 Objective Loss 0.292419 LR 0.001000 Time 0.020459 -2022-12-06 11:00:33,271 - Epoch: [85][ 700/ 1200] Overall Loss 0.292755 Objective Loss 0.292755 LR 0.001000 Time 0.020443 -2022-12-06 11:00:33,465 - Epoch: [85][ 710/ 1200] Overall Loss 0.292834 Objective Loss 0.292834 LR 0.001000 Time 0.020427 -2022-12-06 11:00:33,660 - Epoch: [85][ 720/ 1200] Overall Loss 0.293106 Objective Loss 0.293106 LR 0.001000 Time 0.020413 -2022-12-06 11:00:33,854 - Epoch: [85][ 730/ 1200] Overall Loss 0.293264 Objective Loss 0.293264 LR 0.001000 Time 0.020398 -2022-12-06 11:00:34,047 - Epoch: [85][ 740/ 1200] Overall Loss 0.293442 Objective Loss 0.293442 LR 0.001000 Time 0.020383 -2022-12-06 11:00:34,241 - Epoch: [85][ 750/ 1200] Overall Loss 0.293837 Objective Loss 0.293837 LR 0.001000 Time 0.020369 -2022-12-06 11:00:34,435 - Epoch: [85][ 760/ 1200] Overall Loss 0.293892 Objective Loss 0.293892 LR 0.001000 Time 0.020355 -2022-12-06 11:00:34,630 - Epoch: [85][ 770/ 1200] Overall Loss 0.294220 Objective Loss 0.294220 LR 0.001000 Time 0.020343 -2022-12-06 11:00:34,823 - Epoch: [85][ 780/ 1200] Overall Loss 0.294387 Objective Loss 0.294387 LR 0.001000 Time 0.020330 -2022-12-06 11:00:35,017 - Epoch: [85][ 790/ 1200] Overall Loss 0.294507 Objective Loss 0.294507 LR 0.001000 Time 0.020317 -2022-12-06 11:00:35,211 - Epoch: [85][ 800/ 1200] Overall Loss 0.294626 Objective Loss 0.294626 LR 0.001000 Time 0.020305 -2022-12-06 11:00:35,407 - Epoch: [85][ 810/ 1200] Overall Loss 0.294820 Objective Loss 0.294820 LR 0.001000 Time 0.020295 -2022-12-06 11:00:35,600 - Epoch: [85][ 820/ 1200] Overall Loss 0.295265 Objective Loss 0.295265 LR 0.001000 Time 0.020283 -2022-12-06 11:00:35,795 - Epoch: [85][ 830/ 1200] Overall Loss 0.295356 Objective Loss 0.295356 LR 0.001000 Time 0.020272 -2022-12-06 11:00:35,989 - Epoch: [85][ 840/ 1200] Overall Loss 0.295428 Objective Loss 0.295428 LR 0.001000 Time 0.020262 -2022-12-06 11:00:36,183 - Epoch: [85][ 850/ 1200] Overall Loss 0.295329 Objective Loss 0.295329 LR 0.001000 Time 0.020251 -2022-12-06 11:00:36,378 - Epoch: [85][ 860/ 1200] Overall Loss 0.295678 Objective Loss 0.295678 LR 0.001000 Time 0.020241 -2022-12-06 11:00:36,572 - Epoch: [85][ 870/ 1200] Overall Loss 0.295914 Objective Loss 0.295914 LR 0.001000 Time 0.020231 -2022-12-06 11:00:36,766 - Epoch: [85][ 880/ 1200] Overall Loss 0.296122 Objective Loss 0.296122 LR 0.001000 Time 0.020220 -2022-12-06 11:00:36,961 - Epoch: [85][ 890/ 1200] Overall Loss 0.296582 Objective Loss 0.296582 LR 0.001000 Time 0.020212 -2022-12-06 11:00:37,155 - Epoch: [85][ 900/ 1200] Overall Loss 0.296416 Objective Loss 0.296416 LR 0.001000 Time 0.020202 -2022-12-06 11:00:37,349 - Epoch: [85][ 910/ 1200] Overall Loss 0.296472 Objective Loss 0.296472 LR 0.001000 Time 0.020194 -2022-12-06 11:00:37,543 - Epoch: [85][ 920/ 1200] Overall Loss 0.296601 Objective Loss 0.296601 LR 0.001000 Time 0.020184 -2022-12-06 11:00:37,738 - Epoch: [85][ 930/ 1200] Overall Loss 0.296785 Objective Loss 0.296785 LR 0.001000 Time 0.020176 -2022-12-06 11:00:37,932 - Epoch: [85][ 940/ 1200] Overall Loss 0.297070 Objective Loss 0.297070 LR 0.001000 Time 0.020167 -2022-12-06 11:00:38,127 - Epoch: [85][ 950/ 1200] Overall Loss 0.297124 Objective Loss 0.297124 LR 0.001000 Time 0.020159 -2022-12-06 11:00:38,320 - Epoch: [85][ 960/ 1200] Overall Loss 0.297291 Objective Loss 0.297291 LR 0.001000 Time 0.020150 -2022-12-06 11:00:38,515 - Epoch: [85][ 970/ 1200] Overall Loss 0.297545 Objective Loss 0.297545 LR 0.001000 Time 0.020143 -2022-12-06 11:00:38,708 - Epoch: [85][ 980/ 1200] Overall Loss 0.297540 Objective Loss 0.297540 LR 0.001000 Time 0.020134 -2022-12-06 11:00:38,903 - Epoch: [85][ 990/ 1200] Overall Loss 0.297829 Objective Loss 0.297829 LR 0.001000 Time 0.020127 -2022-12-06 11:00:39,097 - Epoch: [85][ 1000/ 1200] Overall Loss 0.297841 Objective Loss 0.297841 LR 0.001000 Time 0.020119 -2022-12-06 11:00:39,291 - Epoch: [85][ 1010/ 1200] Overall Loss 0.298021 Objective Loss 0.298021 LR 0.001000 Time 0.020111 -2022-12-06 11:00:39,486 - Epoch: [85][ 1020/ 1200] Overall Loss 0.298230 Objective Loss 0.298230 LR 0.001000 Time 0.020104 -2022-12-06 11:00:39,680 - Epoch: [85][ 1030/ 1200] Overall Loss 0.298361 Objective Loss 0.298361 LR 0.001000 Time 0.020097 -2022-12-06 11:00:39,874 - Epoch: [85][ 1040/ 1200] Overall Loss 0.298354 Objective Loss 0.298354 LR 0.001000 Time 0.020089 -2022-12-06 11:00:40,068 - Epoch: [85][ 1050/ 1200] Overall Loss 0.298658 Objective Loss 0.298658 LR 0.001000 Time 0.020082 -2022-12-06 11:00:40,261 - Epoch: [85][ 1060/ 1200] Overall Loss 0.299142 Objective Loss 0.299142 LR 0.001000 Time 0.020075 -2022-12-06 11:00:40,456 - Epoch: [85][ 1070/ 1200] Overall Loss 0.299231 Objective Loss 0.299231 LR 0.001000 Time 0.020068 -2022-12-06 11:00:40,649 - Epoch: [85][ 1080/ 1200] Overall Loss 0.299237 Objective Loss 0.299237 LR 0.001000 Time 0.020061 -2022-12-06 11:00:40,844 - Epoch: [85][ 1090/ 1200] Overall Loss 0.299388 Objective Loss 0.299388 LR 0.001000 Time 0.020055 -2022-12-06 11:00:41,038 - Epoch: [85][ 1100/ 1200] Overall Loss 0.299529 Objective Loss 0.299529 LR 0.001000 Time 0.020049 -2022-12-06 11:00:41,232 - Epoch: [85][ 1110/ 1200] Overall Loss 0.299398 Objective Loss 0.299398 LR 0.001000 Time 0.020042 -2022-12-06 11:00:41,426 - Epoch: [85][ 1120/ 1200] Overall Loss 0.299240 Objective Loss 0.299240 LR 0.001000 Time 0.020036 -2022-12-06 11:00:41,620 - Epoch: [85][ 1130/ 1200] Overall Loss 0.299351 Objective Loss 0.299351 LR 0.001000 Time 0.020030 -2022-12-06 11:00:41,814 - Epoch: [85][ 1140/ 1200] Overall Loss 0.299439 Objective Loss 0.299439 LR 0.001000 Time 0.020024 -2022-12-06 11:00:42,008 - Epoch: [85][ 1150/ 1200] Overall Loss 0.299412 Objective Loss 0.299412 LR 0.001000 Time 0.020018 -2022-12-06 11:00:42,202 - Epoch: [85][ 1160/ 1200] Overall Loss 0.299535 Objective Loss 0.299535 LR 0.001000 Time 0.020012 -2022-12-06 11:00:42,396 - Epoch: [85][ 1170/ 1200] Overall Loss 0.299490 Objective Loss 0.299490 LR 0.001000 Time 0.020007 -2022-12-06 11:00:42,590 - Epoch: [85][ 1180/ 1200] Overall Loss 0.299481 Objective Loss 0.299481 LR 0.001000 Time 0.020001 -2022-12-06 11:00:42,784 - Epoch: [85][ 1190/ 1200] Overall Loss 0.299463 Objective Loss 0.299463 LR 0.001000 Time 0.019996 -2022-12-06 11:00:43,017 - Epoch: [85][ 1200/ 1200] Overall Loss 0.299624 Objective Loss 0.299624 Top1 86.401674 Top5 98.744770 LR 0.001000 Time 0.020023 -2022-12-06 11:00:43,106 - --- validate (epoch=85)----------- -2022-12-06 11:00:43,106 - 34129 samples (256 per mini-batch) -2022-12-06 11:00:43,565 - Epoch: [85][ 10/ 134] Loss 0.305348 Top1 85.429688 Top5 97.851562 -2022-12-06 11:00:43,712 - Epoch: [85][ 20/ 134] Loss 0.314715 Top1 84.785156 Top5 98.007812 -2022-12-06 11:00:43,860 - Epoch: [85][ 30/ 134] Loss 0.308584 Top1 84.934896 Top5 97.994792 -2022-12-06 11:00:44,005 - Epoch: [85][ 40/ 134] Loss 0.312920 Top1 84.687500 Top5 97.841797 -2022-12-06 11:00:44,153 - Epoch: [85][ 50/ 134] Loss 0.312170 Top1 84.695312 Top5 97.859375 -2022-12-06 11:00:44,299 - Epoch: [85][ 60/ 134] Loss 0.313820 Top1 84.641927 Top5 97.871094 -2022-12-06 11:00:44,444 - Epoch: [85][ 70/ 134] Loss 0.317189 Top1 84.659598 Top5 97.912946 -2022-12-06 11:00:44,589 - Epoch: [85][ 80/ 134] Loss 0.317797 Top1 84.599609 Top5 97.910156 -2022-12-06 11:00:44,736 - Epoch: [85][ 90/ 134] Loss 0.316209 Top1 84.548611 Top5 97.877604 -2022-12-06 11:00:44,882 - Epoch: [85][ 100/ 134] Loss 0.315476 Top1 84.460938 Top5 97.890625 -2022-12-06 11:00:45,030 - Epoch: [85][ 110/ 134] Loss 0.317603 Top1 84.417614 Top5 97.897727 -2022-12-06 11:00:45,177 - Epoch: [85][ 120/ 134] Loss 0.315998 Top1 84.440104 Top5 97.880859 -2022-12-06 11:00:45,313 - Epoch: [85][ 130/ 134] Loss 0.317426 Top1 84.393029 Top5 97.872596 -2022-12-06 11:00:45,351 - Epoch: [85][ 134/ 134] Loss 0.316868 Top1 84.403293 Top5 97.866917 -2022-12-06 11:00:45,441 - ==> Top1: 84.403 Top5: 97.867 Loss: 0.317 - -2022-12-06 11:00:45,442 - ==> Confusion: -[[ 909 0 2 1 1 8 0 0 2 45 0 4 2 5 8 1 0 1 2 1 4] - [ 2 884 2 2 7 50 3 34 0 0 3 1 2 2 2 2 9 2 9 5 6] - [ 5 3 992 19 4 4 24 16 0 2 5 4 2 2 2 2 0 1 0 3 13] - [ 3 0 21 944 1 4 0 1 0 0 9 3 4 2 15 0 1 2 5 0 5] - [ 15 1 3 0 946 15 1 1 0 6 1 2 0 3 7 5 7 1 1 1 4] - [ 2 3 0 2 3 999 4 15 2 3 4 10 2 9 1 1 0 1 1 5 2] - [ 1 2 2 2 1 5 1077 3 1 0 0 3 0 2 0 7 1 1 1 6 3] - [ 2 1 5 2 1 41 13 950 0 0 2 6 3 3 0 2 0 0 10 9 4] - [ 9 5 0 1 1 8 0 0 929 50 10 1 3 27 11 1 2 2 1 1 2] - [ 80 0 0 1 8 9 0 1 14 847 1 1 1 19 5 1 1 2 0 1 9] - [ 1 1 4 6 2 5 2 4 7 1 940 3 2 20 4 0 2 0 10 1 4] - [ 1 3 0 2 1 26 2 3 2 0 1 945 24 8 0 9 6 3 0 11 4] - [ 1 2 1 5 1 3 1 0 0 0 2 29 888 3 1 10 0 12 0 3 7] - [ 0 2 0 1 0 15 1 3 6 9 3 2 2 965 0 1 3 2 0 2 6] - [ 9 1 3 13 7 4 0 1 9 3 1 2 0 1 1058 2 2 1 5 2 6] - [ 2 0 2 2 5 5 1 0 0 0 1 10 3 4 0 985 7 12 0 2 2] - [ 4 6 1 0 4 1 2 0 0 0 0 1 0 4 1 13 1022 0 0 6 7] - [ 3 0 1 3 2 1 2 0 2 0 0 9 22 3 1 10 1 973 1 1 1] - [ 3 3 6 18 1 5 3 35 2 1 7 1 4 1 11 1 0 0 899 3 4] - [ 4 2 1 0 2 11 8 4 0 0 0 12 4 7 0 1 3 2 0 1017 2] - [ 150 181 195 125 100 355 108 177 54 93 158 121 383 393 182 117 179 93 115 319 9628]] - -2022-12-06 11:00:46,122 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:00:46,122 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:00:46,128 - - -2022-12-06 11:00:46,128 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:00:47,068 - Epoch: [86][ 10/ 1200] Overall Loss 0.280713 Objective Loss 0.280713 LR 0.001000 Time 0.093920 -2022-12-06 11:00:47,265 - Epoch: [86][ 20/ 1200] Overall Loss 0.281848 Objective Loss 0.281848 LR 0.001000 Time 0.056766 -2022-12-06 11:00:47,458 - Epoch: [86][ 30/ 1200] Overall Loss 0.281640 Objective Loss 0.281640 LR 0.001000 Time 0.044266 -2022-12-06 11:00:47,651 - Epoch: [86][ 40/ 1200] Overall Loss 0.283622 Objective Loss 0.283622 LR 0.001000 Time 0.037997 -2022-12-06 11:00:47,843 - Epoch: [86][ 50/ 1200] Overall Loss 0.282823 Objective Loss 0.282823 LR 0.001000 Time 0.034229 -2022-12-06 11:00:48,034 - Epoch: [86][ 60/ 1200] Overall Loss 0.283845 Objective Loss 0.283845 LR 0.001000 Time 0.031701 -2022-12-06 11:00:48,226 - Epoch: [86][ 70/ 1200] Overall Loss 0.280255 Objective Loss 0.280255 LR 0.001000 Time 0.029904 -2022-12-06 11:00:48,417 - Epoch: [86][ 80/ 1200] Overall Loss 0.283283 Objective Loss 0.283283 LR 0.001000 Time 0.028545 -2022-12-06 11:00:48,609 - Epoch: [86][ 90/ 1200] Overall Loss 0.285742 Objective Loss 0.285742 LR 0.001000 Time 0.027499 -2022-12-06 11:00:48,800 - Epoch: [86][ 100/ 1200] Overall Loss 0.284648 Objective Loss 0.284648 LR 0.001000 Time 0.026661 -2022-12-06 11:00:48,992 - Epoch: [86][ 110/ 1200] Overall Loss 0.281358 Objective Loss 0.281358 LR 0.001000 Time 0.025978 -2022-12-06 11:00:49,184 - Epoch: [86][ 120/ 1200] Overall Loss 0.283102 Objective Loss 0.283102 LR 0.001000 Time 0.025402 -2022-12-06 11:00:49,375 - Epoch: [86][ 130/ 1200] Overall Loss 0.281305 Objective Loss 0.281305 LR 0.001000 Time 0.024917 -2022-12-06 11:00:49,566 - Epoch: [86][ 140/ 1200] Overall Loss 0.281440 Objective Loss 0.281440 LR 0.001000 Time 0.024500 -2022-12-06 11:00:49,759 - Epoch: [86][ 150/ 1200] Overall Loss 0.281320 Objective Loss 0.281320 LR 0.001000 Time 0.024144 -2022-12-06 11:00:49,950 - Epoch: [86][ 160/ 1200] Overall Loss 0.280039 Objective Loss 0.280039 LR 0.001000 Time 0.023827 -2022-12-06 11:00:50,142 - Epoch: [86][ 170/ 1200] Overall Loss 0.280932 Objective Loss 0.280932 LR 0.001000 Time 0.023552 -2022-12-06 11:00:50,333 - Epoch: [86][ 180/ 1200] Overall Loss 0.280636 Objective Loss 0.280636 LR 0.001000 Time 0.023304 -2022-12-06 11:00:50,525 - Epoch: [86][ 190/ 1200] Overall Loss 0.281251 Objective Loss 0.281251 LR 0.001000 Time 0.023085 -2022-12-06 11:00:50,717 - Epoch: [86][ 200/ 1200] Overall Loss 0.281286 Objective Loss 0.281286 LR 0.001000 Time 0.022886 -2022-12-06 11:00:50,909 - Epoch: [86][ 210/ 1200] Overall Loss 0.281136 Objective Loss 0.281136 LR 0.001000 Time 0.022709 -2022-12-06 11:00:51,101 - Epoch: [86][ 220/ 1200] Overall Loss 0.282053 Objective Loss 0.282053 LR 0.001000 Time 0.022547 -2022-12-06 11:00:51,293 - Epoch: [86][ 230/ 1200] Overall Loss 0.281508 Objective Loss 0.281508 LR 0.001000 Time 0.022399 -2022-12-06 11:00:51,485 - Epoch: [86][ 240/ 1200] Overall Loss 0.281543 Objective Loss 0.281543 LR 0.001000 Time 0.022263 -2022-12-06 11:00:51,677 - Epoch: [86][ 250/ 1200] Overall Loss 0.282118 Objective Loss 0.282118 LR 0.001000 Time 0.022139 -2022-12-06 11:00:51,869 - Epoch: [86][ 260/ 1200] Overall Loss 0.282081 Objective Loss 0.282081 LR 0.001000 Time 0.022022 -2022-12-06 11:00:52,060 - Epoch: [86][ 270/ 1200] Overall Loss 0.283917 Objective Loss 0.283917 LR 0.001000 Time 0.021914 -2022-12-06 11:00:52,252 - Epoch: [86][ 280/ 1200] Overall Loss 0.283553 Objective Loss 0.283553 LR 0.001000 Time 0.021815 -2022-12-06 11:00:52,444 - Epoch: [86][ 290/ 1200] Overall Loss 0.283720 Objective Loss 0.283720 LR 0.001000 Time 0.021722 -2022-12-06 11:00:52,636 - Epoch: [86][ 300/ 1200] Overall Loss 0.283844 Objective Loss 0.283844 LR 0.001000 Time 0.021634 -2022-12-06 11:00:52,827 - Epoch: [86][ 310/ 1200] Overall Loss 0.283758 Objective Loss 0.283758 LR 0.001000 Time 0.021553 -2022-12-06 11:00:53,019 - Epoch: [86][ 320/ 1200] Overall Loss 0.284636 Objective Loss 0.284636 LR 0.001000 Time 0.021477 -2022-12-06 11:00:53,211 - Epoch: [86][ 330/ 1200] Overall Loss 0.285151 Objective Loss 0.285151 LR 0.001000 Time 0.021406 -2022-12-06 11:00:53,403 - Epoch: [86][ 340/ 1200] Overall Loss 0.285406 Objective Loss 0.285406 LR 0.001000 Time 0.021339 -2022-12-06 11:00:53,595 - Epoch: [86][ 350/ 1200] Overall Loss 0.286363 Objective Loss 0.286363 LR 0.001000 Time 0.021278 -2022-12-06 11:00:53,787 - Epoch: [86][ 360/ 1200] Overall Loss 0.287094 Objective Loss 0.287094 LR 0.001000 Time 0.021218 -2022-12-06 11:00:53,979 - Epoch: [86][ 370/ 1200] Overall Loss 0.287303 Objective Loss 0.287303 LR 0.001000 Time 0.021161 -2022-12-06 11:00:54,170 - Epoch: [86][ 380/ 1200] Overall Loss 0.286843 Objective Loss 0.286843 LR 0.001000 Time 0.021108 -2022-12-06 11:00:54,362 - Epoch: [86][ 390/ 1200] Overall Loss 0.287014 Objective Loss 0.287014 LR 0.001000 Time 0.021056 -2022-12-06 11:00:54,554 - Epoch: [86][ 400/ 1200] Overall Loss 0.287006 Objective Loss 0.287006 LR 0.001000 Time 0.021008 -2022-12-06 11:00:54,746 - Epoch: [86][ 410/ 1200] Overall Loss 0.286709 Objective Loss 0.286709 LR 0.001000 Time 0.020963 -2022-12-06 11:00:54,938 - Epoch: [86][ 420/ 1200] Overall Loss 0.287016 Objective Loss 0.287016 LR 0.001000 Time 0.020919 -2022-12-06 11:00:55,129 - Epoch: [86][ 430/ 1200] Overall Loss 0.287308 Objective Loss 0.287308 LR 0.001000 Time 0.020877 -2022-12-06 11:00:55,321 - Epoch: [86][ 440/ 1200] Overall Loss 0.287631 Objective Loss 0.287631 LR 0.001000 Time 0.020837 -2022-12-06 11:00:55,513 - Epoch: [86][ 450/ 1200] Overall Loss 0.287760 Objective Loss 0.287760 LR 0.001000 Time 0.020799 -2022-12-06 11:00:55,705 - Epoch: [86][ 460/ 1200] Overall Loss 0.287665 Objective Loss 0.287665 LR 0.001000 Time 0.020763 -2022-12-06 11:00:55,897 - Epoch: [86][ 470/ 1200] Overall Loss 0.288229 Objective Loss 0.288229 LR 0.001000 Time 0.020729 -2022-12-06 11:00:56,089 - Epoch: [86][ 480/ 1200] Overall Loss 0.288440 Objective Loss 0.288440 LR 0.001000 Time 0.020696 -2022-12-06 11:00:56,281 - Epoch: [86][ 490/ 1200] Overall Loss 0.288317 Objective Loss 0.288317 LR 0.001000 Time 0.020664 -2022-12-06 11:00:56,472 - Epoch: [86][ 500/ 1200] Overall Loss 0.288517 Objective Loss 0.288517 LR 0.001000 Time 0.020633 -2022-12-06 11:00:56,664 - Epoch: [86][ 510/ 1200] Overall Loss 0.288576 Objective Loss 0.288576 LR 0.001000 Time 0.020603 -2022-12-06 11:00:56,857 - Epoch: [86][ 520/ 1200] Overall Loss 0.288438 Objective Loss 0.288438 LR 0.001000 Time 0.020577 -2022-12-06 11:00:57,054 - Epoch: [86][ 530/ 1200] Overall Loss 0.289264 Objective Loss 0.289264 LR 0.001000 Time 0.020559 -2022-12-06 11:00:57,251 - Epoch: [86][ 540/ 1200] Overall Loss 0.289599 Objective Loss 0.289599 LR 0.001000 Time 0.020542 -2022-12-06 11:00:57,447 - Epoch: [86][ 550/ 1200] Overall Loss 0.289774 Objective Loss 0.289774 LR 0.001000 Time 0.020524 -2022-12-06 11:00:57,643 - Epoch: [86][ 560/ 1200] Overall Loss 0.289896 Objective Loss 0.289896 LR 0.001000 Time 0.020507 -2022-12-06 11:00:57,840 - Epoch: [86][ 570/ 1200] Overall Loss 0.290027 Objective Loss 0.290027 LR 0.001000 Time 0.020491 -2022-12-06 11:00:58,036 - Epoch: [86][ 580/ 1200] Overall Loss 0.289707 Objective Loss 0.289707 LR 0.001000 Time 0.020475 -2022-12-06 11:00:58,232 - Epoch: [86][ 590/ 1200] Overall Loss 0.289867 Objective Loss 0.289867 LR 0.001000 Time 0.020459 -2022-12-06 11:00:58,428 - Epoch: [86][ 600/ 1200] Overall Loss 0.289739 Objective Loss 0.289739 LR 0.001000 Time 0.020444 -2022-12-06 11:00:58,625 - Epoch: [86][ 610/ 1200] Overall Loss 0.289690 Objective Loss 0.289690 LR 0.001000 Time 0.020430 -2022-12-06 11:00:58,820 - Epoch: [86][ 620/ 1200] Overall Loss 0.290229 Objective Loss 0.290229 LR 0.001000 Time 0.020415 -2022-12-06 11:00:59,016 - Epoch: [86][ 630/ 1200] Overall Loss 0.290310 Objective Loss 0.290310 LR 0.001000 Time 0.020401 -2022-12-06 11:00:59,212 - Epoch: [86][ 640/ 1200] Overall Loss 0.290007 Objective Loss 0.290007 LR 0.001000 Time 0.020387 -2022-12-06 11:00:59,408 - Epoch: [86][ 650/ 1200] Overall Loss 0.289952 Objective Loss 0.289952 LR 0.001000 Time 0.020374 -2022-12-06 11:00:59,603 - Epoch: [86][ 660/ 1200] Overall Loss 0.289875 Objective Loss 0.289875 LR 0.001000 Time 0.020361 -2022-12-06 11:00:59,799 - Epoch: [86][ 670/ 1200] Overall Loss 0.289944 Objective Loss 0.289944 LR 0.001000 Time 0.020349 -2022-12-06 11:00:59,995 - Epoch: [86][ 680/ 1200] Overall Loss 0.290143 Objective Loss 0.290143 LR 0.001000 Time 0.020337 -2022-12-06 11:01:00,192 - Epoch: [86][ 690/ 1200] Overall Loss 0.290392 Objective Loss 0.290392 LR 0.001000 Time 0.020326 -2022-12-06 11:01:00,388 - Epoch: [86][ 700/ 1200] Overall Loss 0.290230 Objective Loss 0.290230 LR 0.001000 Time 0.020315 -2022-12-06 11:01:00,584 - Epoch: [86][ 710/ 1200] Overall Loss 0.289830 Objective Loss 0.289830 LR 0.001000 Time 0.020304 -2022-12-06 11:01:00,779 - Epoch: [86][ 720/ 1200] Overall Loss 0.289964 Objective Loss 0.289964 LR 0.001000 Time 0.020293 -2022-12-06 11:01:00,975 - Epoch: [86][ 730/ 1200] Overall Loss 0.290187 Objective Loss 0.290187 LR 0.001000 Time 0.020283 -2022-12-06 11:01:01,171 - Epoch: [86][ 740/ 1200] Overall Loss 0.289981 Objective Loss 0.289981 LR 0.001000 Time 0.020273 -2022-12-06 11:01:01,367 - Epoch: [86][ 750/ 1200] Overall Loss 0.289918 Objective Loss 0.289918 LR 0.001000 Time 0.020263 -2022-12-06 11:01:01,563 - Epoch: [86][ 760/ 1200] Overall Loss 0.289791 Objective Loss 0.289791 LR 0.001000 Time 0.020253 -2022-12-06 11:01:01,759 - Epoch: [86][ 770/ 1200] Overall Loss 0.289700 Objective Loss 0.289700 LR 0.001000 Time 0.020244 -2022-12-06 11:01:01,955 - Epoch: [86][ 780/ 1200] Overall Loss 0.289898 Objective Loss 0.289898 LR 0.001000 Time 0.020235 -2022-12-06 11:01:02,151 - Epoch: [86][ 790/ 1200] Overall Loss 0.289815 Objective Loss 0.289815 LR 0.001000 Time 0.020226 -2022-12-06 11:01:02,346 - Epoch: [86][ 800/ 1200] Overall Loss 0.290123 Objective Loss 0.290123 LR 0.001000 Time 0.020217 -2022-12-06 11:01:02,543 - Epoch: [86][ 810/ 1200] Overall Loss 0.290158 Objective Loss 0.290158 LR 0.001000 Time 0.020209 -2022-12-06 11:01:02,739 - Epoch: [86][ 820/ 1200] Overall Loss 0.290131 Objective Loss 0.290131 LR 0.001000 Time 0.020202 -2022-12-06 11:01:02,935 - Epoch: [86][ 830/ 1200] Overall Loss 0.290376 Objective Loss 0.290376 LR 0.001000 Time 0.020194 -2022-12-06 11:01:03,132 - Epoch: [86][ 840/ 1200] Overall Loss 0.290578 Objective Loss 0.290578 LR 0.001000 Time 0.020187 -2022-12-06 11:01:03,328 - Epoch: [86][ 850/ 1200] Overall Loss 0.290641 Objective Loss 0.290641 LR 0.001000 Time 0.020180 -2022-12-06 11:01:03,524 - Epoch: [86][ 860/ 1200] Overall Loss 0.290616 Objective Loss 0.290616 LR 0.001000 Time 0.020172 -2022-12-06 11:01:03,721 - Epoch: [86][ 870/ 1200] Overall Loss 0.291016 Objective Loss 0.291016 LR 0.001000 Time 0.020166 -2022-12-06 11:01:03,917 - Epoch: [86][ 880/ 1200] Overall Loss 0.290721 Objective Loss 0.290721 LR 0.001000 Time 0.020159 -2022-12-06 11:01:04,113 - Epoch: [86][ 890/ 1200] Overall Loss 0.290610 Objective Loss 0.290610 LR 0.001000 Time 0.020153 -2022-12-06 11:01:04,309 - Epoch: [86][ 900/ 1200] Overall Loss 0.290542 Objective Loss 0.290542 LR 0.001000 Time 0.020146 -2022-12-06 11:01:04,506 - Epoch: [86][ 910/ 1200] Overall Loss 0.290490 Objective Loss 0.290490 LR 0.001000 Time 0.020140 -2022-12-06 11:01:04,703 - Epoch: [86][ 920/ 1200] Overall Loss 0.290843 Objective Loss 0.290843 LR 0.001000 Time 0.020134 -2022-12-06 11:01:04,900 - Epoch: [86][ 930/ 1200] Overall Loss 0.290739 Objective Loss 0.290739 LR 0.001000 Time 0.020129 -2022-12-06 11:01:05,097 - Epoch: [86][ 940/ 1200] Overall Loss 0.290613 Objective Loss 0.290613 LR 0.001000 Time 0.020123 -2022-12-06 11:01:05,293 - Epoch: [86][ 950/ 1200] Overall Loss 0.290859 Objective Loss 0.290859 LR 0.001000 Time 0.020118 -2022-12-06 11:01:05,490 - Epoch: [86][ 960/ 1200] Overall Loss 0.290611 Objective Loss 0.290611 LR 0.001000 Time 0.020113 -2022-12-06 11:01:05,686 - Epoch: [86][ 970/ 1200] Overall Loss 0.290741 Objective Loss 0.290741 LR 0.001000 Time 0.020107 -2022-12-06 11:01:05,883 - Epoch: [86][ 980/ 1200] Overall Loss 0.290863 Objective Loss 0.290863 LR 0.001000 Time 0.020102 -2022-12-06 11:01:06,080 - Epoch: [86][ 990/ 1200] Overall Loss 0.290624 Objective Loss 0.290624 LR 0.001000 Time 0.020097 -2022-12-06 11:01:06,275 - Epoch: [86][ 1000/ 1200] Overall Loss 0.290582 Objective Loss 0.290582 LR 0.001000 Time 0.020091 -2022-12-06 11:01:06,471 - Epoch: [86][ 1010/ 1200] Overall Loss 0.290661 Objective Loss 0.290661 LR 0.001000 Time 0.020085 -2022-12-06 11:01:06,667 - Epoch: [86][ 1020/ 1200] Overall Loss 0.290795 Objective Loss 0.290795 LR 0.001000 Time 0.020080 -2022-12-06 11:01:06,862 - Epoch: [86][ 1030/ 1200] Overall Loss 0.290965 Objective Loss 0.290965 LR 0.001000 Time 0.020074 -2022-12-06 11:01:07,058 - Epoch: [86][ 1040/ 1200] Overall Loss 0.290904 Objective Loss 0.290904 LR 0.001000 Time 0.020069 -2022-12-06 11:01:07,254 - Epoch: [86][ 1050/ 1200] Overall Loss 0.291300 Objective Loss 0.291300 LR 0.001000 Time 0.020064 -2022-12-06 11:01:07,449 - Epoch: [86][ 1060/ 1200] Overall Loss 0.291085 Objective Loss 0.291085 LR 0.001000 Time 0.020059 -2022-12-06 11:01:07,645 - Epoch: [86][ 1070/ 1200] Overall Loss 0.290814 Objective Loss 0.290814 LR 0.001000 Time 0.020054 -2022-12-06 11:01:07,841 - Epoch: [86][ 1080/ 1200] Overall Loss 0.291112 Objective Loss 0.291112 LR 0.001000 Time 0.020049 -2022-12-06 11:01:08,037 - Epoch: [86][ 1090/ 1200] Overall Loss 0.291389 Objective Loss 0.291389 LR 0.001000 Time 0.020044 -2022-12-06 11:01:08,232 - Epoch: [86][ 1100/ 1200] Overall Loss 0.291437 Objective Loss 0.291437 LR 0.001000 Time 0.020039 -2022-12-06 11:01:08,428 - Epoch: [86][ 1110/ 1200] Overall Loss 0.291171 Objective Loss 0.291171 LR 0.001000 Time 0.020035 -2022-12-06 11:01:08,624 - Epoch: [86][ 1120/ 1200] Overall Loss 0.291323 Objective Loss 0.291323 LR 0.001000 Time 0.020030 -2022-12-06 11:01:08,820 - Epoch: [86][ 1130/ 1200] Overall Loss 0.291286 Objective Loss 0.291286 LR 0.001000 Time 0.020025 -2022-12-06 11:01:09,015 - Epoch: [86][ 1140/ 1200] Overall Loss 0.291139 Objective Loss 0.291139 LR 0.001000 Time 0.020021 -2022-12-06 11:01:09,211 - Epoch: [86][ 1150/ 1200] Overall Loss 0.291208 Objective Loss 0.291208 LR 0.001000 Time 0.020017 -2022-12-06 11:01:09,407 - Epoch: [86][ 1160/ 1200] Overall Loss 0.291262 Objective Loss 0.291262 LR 0.001000 Time 0.020013 -2022-12-06 11:01:09,603 - Epoch: [86][ 1170/ 1200] Overall Loss 0.290992 Objective Loss 0.290992 LR 0.001000 Time 0.020008 -2022-12-06 11:01:09,798 - Epoch: [86][ 1180/ 1200] Overall Loss 0.290838 Objective Loss 0.290838 LR 0.001000 Time 0.020004 -2022-12-06 11:01:09,994 - Epoch: [86][ 1190/ 1200] Overall Loss 0.291177 Objective Loss 0.291177 LR 0.001000 Time 0.020000 -2022-12-06 11:01:10,226 - Epoch: [86][ 1200/ 1200] Overall Loss 0.291022 Objective Loss 0.291022 Top1 85.564854 Top5 98.326360 LR 0.001000 Time 0.020026 -2022-12-06 11:01:10,316 - --- validate (epoch=86)----------- -2022-12-06 11:01:10,316 - 34129 samples (256 per mini-batch) -2022-12-06 11:01:10,766 - Epoch: [86][ 10/ 134] Loss 0.307096 Top1 84.570312 Top5 97.968750 -2022-12-06 11:01:10,897 - Epoch: [86][ 20/ 134] Loss 0.299329 Top1 85.058594 Top5 97.871094 -2022-12-06 11:01:11,045 - Epoch: [86][ 30/ 134] Loss 0.299340 Top1 84.648438 Top5 97.929688 -2022-12-06 11:01:11,181 - Epoch: [86][ 40/ 134] Loss 0.306012 Top1 84.511719 Top5 97.851562 -2022-12-06 11:01:11,312 - Epoch: [86][ 50/ 134] Loss 0.310382 Top1 84.578125 Top5 97.750000 -2022-12-06 11:01:11,441 - Epoch: [86][ 60/ 134] Loss 0.309188 Top1 84.602865 Top5 97.760417 -2022-12-06 11:01:11,572 - Epoch: [86][ 70/ 134] Loss 0.315851 Top1 84.224330 Top5 97.700893 -2022-12-06 11:01:11,703 - Epoch: [86][ 80/ 134] Loss 0.314335 Top1 84.179688 Top5 97.739258 -2022-12-06 11:01:11,833 - Epoch: [86][ 90/ 134] Loss 0.312391 Top1 84.197049 Top5 97.756076 -2022-12-06 11:01:11,962 - Epoch: [86][ 100/ 134] Loss 0.313618 Top1 84.136719 Top5 97.699219 -2022-12-06 11:01:12,091 - Epoch: [86][ 110/ 134] Loss 0.315335 Top1 84.154830 Top5 97.709517 -2022-12-06 11:01:12,223 - Epoch: [86][ 120/ 134] Loss 0.314014 Top1 84.176432 Top5 97.711589 -2022-12-06 11:01:12,351 - Epoch: [86][ 130/ 134] Loss 0.314480 Top1 84.143630 Top5 97.728365 -2022-12-06 11:01:12,389 - Epoch: [86][ 134/ 134] Loss 0.314411 Top1 84.142518 Top5 97.711624 -2022-12-06 11:01:12,481 - ==> Top1: 84.143 Top5: 97.712 Loss: 0.314 - -2022-12-06 11:01:12,482 - ==> Confusion: -[[ 895 4 2 0 10 5 0 0 10 49 0 3 2 3 5 3 1 0 1 0 3] - [ 0 917 1 2 7 26 3 16 0 1 8 5 5 2 1 2 3 2 13 4 9] - [ 5 2 978 16 3 2 33 7 0 3 5 5 5 2 5 3 1 1 8 3 16] - [ 3 3 19 928 1 4 0 1 1 1 8 0 10 2 19 1 0 3 11 0 5] - [ 7 3 2 0 949 9 1 0 1 8 1 1 1 2 14 6 7 2 0 2 4] - [ 7 14 1 2 6 965 4 18 2 2 3 9 7 13 2 3 2 0 1 4 4] - [ 1 3 9 0 0 4 1072 3 0 1 3 3 0 1 0 4 0 3 2 7 2] - [ 2 15 7 3 2 31 6 922 1 3 0 7 4 1 2 3 1 0 26 11 7] - [ 3 2 0 1 1 1 1 0 953 49 8 3 1 8 22 1 4 0 4 1 1] - [ 50 2 2 0 4 3 0 4 28 884 1 0 1 11 2 2 0 0 0 1 6] - [ 2 1 7 3 0 0 2 3 11 2 949 1 4 12 5 1 0 0 9 2 5] - [ 3 1 2 1 1 9 3 4 0 1 2 953 34 7 1 6 4 3 3 11 2] - [ 1 0 0 4 0 4 1 0 0 1 1 30 899 1 1 8 1 6 0 5 6] - [ 0 0 2 0 1 12 0 3 15 19 8 7 3 925 0 2 6 0 2 4 14] - [ 8 2 1 13 4 1 1 2 16 5 0 2 5 2 1051 1 0 1 9 1 5] - [ 1 0 2 0 0 2 3 0 1 0 0 9 14 4 1 982 6 13 0 4 1] - [ 3 3 1 1 1 1 0 0 0 0 0 5 5 1 3 16 1019 0 1 4 8] - [ 2 0 2 2 0 1 1 1 0 3 0 8 27 4 3 13 1 962 1 2 3] - [ 4 5 3 6 0 4 2 21 5 0 4 2 3 1 10 1 1 0 931 2 3] - [ 2 0 1 1 0 6 10 4 0 0 1 13 10 4 1 3 4 3 1 1009 7] - [ 176 210 192 123 136 215 94 115 89 119 163 113 490 278 200 166 196 96 183 299 9573]] - -2022-12-06 11:01:13,060 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:01:13,060 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:01:13,066 - - -2022-12-06 11:01:13,066 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:01:14,104 - Epoch: [87][ 10/ 1200] Overall Loss 0.275923 Objective Loss 0.275923 LR 0.001000 Time 0.103744 -2022-12-06 11:01:14,298 - Epoch: [87][ 20/ 1200] Overall Loss 0.271738 Objective Loss 0.271738 LR 0.001000 Time 0.061523 -2022-12-06 11:01:14,490 - Epoch: [87][ 30/ 1200] Overall Loss 0.273981 Objective Loss 0.273981 LR 0.001000 Time 0.047411 -2022-12-06 11:01:14,682 - Epoch: [87][ 40/ 1200] Overall Loss 0.273799 Objective Loss 0.273799 LR 0.001000 Time 0.040333 -2022-12-06 11:01:14,873 - Epoch: [87][ 50/ 1200] Overall Loss 0.272800 Objective Loss 0.272800 LR 0.001000 Time 0.036084 -2022-12-06 11:01:15,066 - Epoch: [87][ 60/ 1200] Overall Loss 0.270406 Objective Loss 0.270406 LR 0.001000 Time 0.033264 -2022-12-06 11:01:15,257 - Epoch: [87][ 70/ 1200] Overall Loss 0.274615 Objective Loss 0.274615 LR 0.001000 Time 0.031238 -2022-12-06 11:01:15,449 - Epoch: [87][ 80/ 1200] Overall Loss 0.272832 Objective Loss 0.272832 LR 0.001000 Time 0.029725 -2022-12-06 11:01:15,640 - Epoch: [87][ 90/ 1200] Overall Loss 0.273512 Objective Loss 0.273512 LR 0.001000 Time 0.028539 -2022-12-06 11:01:15,832 - Epoch: [87][ 100/ 1200] Overall Loss 0.273493 Objective Loss 0.273493 LR 0.001000 Time 0.027597 -2022-12-06 11:01:16,023 - Epoch: [87][ 110/ 1200] Overall Loss 0.270549 Objective Loss 0.270549 LR 0.001000 Time 0.026825 -2022-12-06 11:01:16,215 - Epoch: [87][ 120/ 1200] Overall Loss 0.271353 Objective Loss 0.271353 LR 0.001000 Time 0.026183 -2022-12-06 11:01:16,406 - Epoch: [87][ 130/ 1200] Overall Loss 0.274967 Objective Loss 0.274967 LR 0.001000 Time 0.025637 -2022-12-06 11:01:16,598 - Epoch: [87][ 140/ 1200] Overall Loss 0.276002 Objective Loss 0.276002 LR 0.001000 Time 0.025168 -2022-12-06 11:01:16,788 - Epoch: [87][ 150/ 1200] Overall Loss 0.275979 Objective Loss 0.275979 LR 0.001000 Time 0.024759 -2022-12-06 11:01:16,980 - Epoch: [87][ 160/ 1200] Overall Loss 0.278708 Objective Loss 0.278708 LR 0.001000 Time 0.024405 -2022-12-06 11:01:17,171 - Epoch: [87][ 170/ 1200] Overall Loss 0.278345 Objective Loss 0.278345 LR 0.001000 Time 0.024092 -2022-12-06 11:01:17,363 - Epoch: [87][ 180/ 1200] Overall Loss 0.280151 Objective Loss 0.280151 LR 0.001000 Time 0.023813 -2022-12-06 11:01:17,554 - Epoch: [87][ 190/ 1200] Overall Loss 0.281620 Objective Loss 0.281620 LR 0.001000 Time 0.023563 -2022-12-06 11:01:17,745 - Epoch: [87][ 200/ 1200] Overall Loss 0.281641 Objective Loss 0.281641 LR 0.001000 Time 0.023338 -2022-12-06 11:01:17,936 - Epoch: [87][ 210/ 1200] Overall Loss 0.282638 Objective Loss 0.282638 LR 0.001000 Time 0.023134 -2022-12-06 11:01:18,127 - Epoch: [87][ 220/ 1200] Overall Loss 0.283320 Objective Loss 0.283320 LR 0.001000 Time 0.022950 -2022-12-06 11:01:18,319 - Epoch: [87][ 230/ 1200] Overall Loss 0.285307 Objective Loss 0.285307 LR 0.001000 Time 0.022783 -2022-12-06 11:01:18,510 - Epoch: [87][ 240/ 1200] Overall Loss 0.285314 Objective Loss 0.285314 LR 0.001000 Time 0.022628 -2022-12-06 11:01:18,701 - Epoch: [87][ 250/ 1200] Overall Loss 0.285366 Objective Loss 0.285366 LR 0.001000 Time 0.022485 -2022-12-06 11:01:18,893 - Epoch: [87][ 260/ 1200] Overall Loss 0.285865 Objective Loss 0.285865 LR 0.001000 Time 0.022355 -2022-12-06 11:01:19,084 - Epoch: [87][ 270/ 1200] Overall Loss 0.287374 Objective Loss 0.287374 LR 0.001000 Time 0.022232 -2022-12-06 11:01:19,275 - Epoch: [87][ 280/ 1200] Overall Loss 0.287612 Objective Loss 0.287612 LR 0.001000 Time 0.022120 -2022-12-06 11:01:19,467 - Epoch: [87][ 290/ 1200] Overall Loss 0.287920 Objective Loss 0.287920 LR 0.001000 Time 0.022015 -2022-12-06 11:01:19,659 - Epoch: [87][ 300/ 1200] Overall Loss 0.287857 Objective Loss 0.287857 LR 0.001000 Time 0.021920 -2022-12-06 11:01:19,850 - Epoch: [87][ 310/ 1200] Overall Loss 0.286848 Objective Loss 0.286848 LR 0.001000 Time 0.021827 -2022-12-06 11:01:20,042 - Epoch: [87][ 320/ 1200] Overall Loss 0.287305 Objective Loss 0.287305 LR 0.001000 Time 0.021743 -2022-12-06 11:01:20,232 - Epoch: [87][ 330/ 1200] Overall Loss 0.286954 Objective Loss 0.286954 LR 0.001000 Time 0.021661 -2022-12-06 11:01:20,424 - Epoch: [87][ 340/ 1200] Overall Loss 0.286429 Objective Loss 0.286429 LR 0.001000 Time 0.021584 -2022-12-06 11:01:20,615 - Epoch: [87][ 350/ 1200] Overall Loss 0.286517 Objective Loss 0.286517 LR 0.001000 Time 0.021512 -2022-12-06 11:01:20,807 - Epoch: [87][ 360/ 1200] Overall Loss 0.285871 Objective Loss 0.285871 LR 0.001000 Time 0.021446 -2022-12-06 11:01:20,998 - Epoch: [87][ 370/ 1200] Overall Loss 0.285150 Objective Loss 0.285150 LR 0.001000 Time 0.021382 -2022-12-06 11:01:21,189 - Epoch: [87][ 380/ 1200] Overall Loss 0.284956 Objective Loss 0.284956 LR 0.001000 Time 0.021322 -2022-12-06 11:01:21,381 - Epoch: [87][ 390/ 1200] Overall Loss 0.285255 Objective Loss 0.285255 LR 0.001000 Time 0.021265 -2022-12-06 11:01:21,573 - Epoch: [87][ 400/ 1200] Overall Loss 0.284343 Objective Loss 0.284343 LR 0.001000 Time 0.021212 -2022-12-06 11:01:21,764 - Epoch: [87][ 410/ 1200] Overall Loss 0.284620 Objective Loss 0.284620 LR 0.001000 Time 0.021159 -2022-12-06 11:01:21,955 - Epoch: [87][ 420/ 1200] Overall Loss 0.284428 Objective Loss 0.284428 LR 0.001000 Time 0.021109 -2022-12-06 11:01:22,146 - Epoch: [87][ 430/ 1200] Overall Loss 0.284792 Objective Loss 0.284792 LR 0.001000 Time 0.021060 -2022-12-06 11:01:22,338 - Epoch: [87][ 440/ 1200] Overall Loss 0.285051 Objective Loss 0.285051 LR 0.001000 Time 0.021016 -2022-12-06 11:01:22,529 - Epoch: [87][ 450/ 1200] Overall Loss 0.284498 Objective Loss 0.284498 LR 0.001000 Time 0.020974 -2022-12-06 11:01:22,721 - Epoch: [87][ 460/ 1200] Overall Loss 0.283794 Objective Loss 0.283794 LR 0.001000 Time 0.020934 -2022-12-06 11:01:22,913 - Epoch: [87][ 470/ 1200] Overall Loss 0.283783 Objective Loss 0.283783 LR 0.001000 Time 0.020896 -2022-12-06 11:01:23,105 - Epoch: [87][ 480/ 1200] Overall Loss 0.284035 Objective Loss 0.284035 LR 0.001000 Time 0.020859 -2022-12-06 11:01:23,297 - Epoch: [87][ 490/ 1200] Overall Loss 0.283947 Objective Loss 0.283947 LR 0.001000 Time 0.020823 -2022-12-06 11:01:23,488 - Epoch: [87][ 500/ 1200] Overall Loss 0.284323 Objective Loss 0.284323 LR 0.001000 Time 0.020788 -2022-12-06 11:01:23,679 - Epoch: [87][ 510/ 1200] Overall Loss 0.284543 Objective Loss 0.284543 LR 0.001000 Time 0.020755 -2022-12-06 11:01:23,871 - Epoch: [87][ 520/ 1200] Overall Loss 0.284455 Objective Loss 0.284455 LR 0.001000 Time 0.020723 -2022-12-06 11:01:24,062 - Epoch: [87][ 530/ 1200] Overall Loss 0.284156 Objective Loss 0.284156 LR 0.001000 Time 0.020691 -2022-12-06 11:01:24,253 - Epoch: [87][ 540/ 1200] Overall Loss 0.284446 Objective Loss 0.284446 LR 0.001000 Time 0.020661 -2022-12-06 11:01:24,444 - Epoch: [87][ 550/ 1200] Overall Loss 0.284531 Objective Loss 0.284531 LR 0.001000 Time 0.020631 -2022-12-06 11:01:24,635 - Epoch: [87][ 560/ 1200] Overall Loss 0.284383 Objective Loss 0.284383 LR 0.001000 Time 0.020604 -2022-12-06 11:01:24,826 - Epoch: [87][ 570/ 1200] Overall Loss 0.284282 Objective Loss 0.284282 LR 0.001000 Time 0.020577 -2022-12-06 11:01:25,018 - Epoch: [87][ 580/ 1200] Overall Loss 0.283676 Objective Loss 0.283676 LR 0.001000 Time 0.020552 -2022-12-06 11:01:25,210 - Epoch: [87][ 590/ 1200] Overall Loss 0.283887 Objective Loss 0.283887 LR 0.001000 Time 0.020528 -2022-12-06 11:01:25,401 - Epoch: [87][ 600/ 1200] Overall Loss 0.283725 Objective Loss 0.283725 LR 0.001000 Time 0.020503 -2022-12-06 11:01:25,593 - Epoch: [87][ 610/ 1200] Overall Loss 0.283863 Objective Loss 0.283863 LR 0.001000 Time 0.020481 -2022-12-06 11:01:25,785 - Epoch: [87][ 620/ 1200] Overall Loss 0.283743 Objective Loss 0.283743 LR 0.001000 Time 0.020459 -2022-12-06 11:01:25,976 - Epoch: [87][ 630/ 1200] Overall Loss 0.283629 Objective Loss 0.283629 LR 0.001000 Time 0.020437 -2022-12-06 11:01:26,169 - Epoch: [87][ 640/ 1200] Overall Loss 0.283743 Objective Loss 0.283743 LR 0.001000 Time 0.020418 -2022-12-06 11:01:26,360 - Epoch: [87][ 650/ 1200] Overall Loss 0.284049 Objective Loss 0.284049 LR 0.001000 Time 0.020397 -2022-12-06 11:01:26,551 - Epoch: [87][ 660/ 1200] Overall Loss 0.284459 Objective Loss 0.284459 LR 0.001000 Time 0.020377 -2022-12-06 11:01:26,742 - Epoch: [87][ 670/ 1200] Overall Loss 0.284524 Objective Loss 0.284524 LR 0.001000 Time 0.020357 -2022-12-06 11:01:26,934 - Epoch: [87][ 680/ 1200] Overall Loss 0.284540 Objective Loss 0.284540 LR 0.001000 Time 0.020339 -2022-12-06 11:01:27,125 - Epoch: [87][ 690/ 1200] Overall Loss 0.284291 Objective Loss 0.284291 LR 0.001000 Time 0.020321 -2022-12-06 11:01:27,317 - Epoch: [87][ 700/ 1200] Overall Loss 0.283959 Objective Loss 0.283959 LR 0.001000 Time 0.020304 -2022-12-06 11:01:27,509 - Epoch: [87][ 710/ 1200] Overall Loss 0.283655 Objective Loss 0.283655 LR 0.001000 Time 0.020287 -2022-12-06 11:01:27,700 - Epoch: [87][ 720/ 1200] Overall Loss 0.283238 Objective Loss 0.283238 LR 0.001000 Time 0.020270 -2022-12-06 11:01:27,892 - Epoch: [87][ 730/ 1200] Overall Loss 0.283506 Objective Loss 0.283506 LR 0.001000 Time 0.020254 -2022-12-06 11:01:28,083 - Epoch: [87][ 740/ 1200] Overall Loss 0.283483 Objective Loss 0.283483 LR 0.001000 Time 0.020238 -2022-12-06 11:01:28,274 - Epoch: [87][ 750/ 1200] Overall Loss 0.283986 Objective Loss 0.283986 LR 0.001000 Time 0.020222 -2022-12-06 11:01:28,466 - Epoch: [87][ 760/ 1200] Overall Loss 0.284443 Objective Loss 0.284443 LR 0.001000 Time 0.020208 -2022-12-06 11:01:28,658 - Epoch: [87][ 770/ 1200] Overall Loss 0.284489 Objective Loss 0.284489 LR 0.001000 Time 0.020194 -2022-12-06 11:01:28,849 - Epoch: [87][ 780/ 1200] Overall Loss 0.284464 Objective Loss 0.284464 LR 0.001000 Time 0.020179 -2022-12-06 11:01:29,041 - Epoch: [87][ 790/ 1200] Overall Loss 0.284738 Objective Loss 0.284738 LR 0.001000 Time 0.020166 -2022-12-06 11:01:29,232 - Epoch: [87][ 800/ 1200] Overall Loss 0.285036 Objective Loss 0.285036 LR 0.001000 Time 0.020153 -2022-12-06 11:01:29,424 - Epoch: [87][ 810/ 1200] Overall Loss 0.285113 Objective Loss 0.285113 LR 0.001000 Time 0.020140 -2022-12-06 11:01:29,615 - Epoch: [87][ 820/ 1200] Overall Loss 0.285127 Objective Loss 0.285127 LR 0.001000 Time 0.020127 -2022-12-06 11:01:29,807 - Epoch: [87][ 830/ 1200] Overall Loss 0.285470 Objective Loss 0.285470 LR 0.001000 Time 0.020115 -2022-12-06 11:01:29,998 - Epoch: [87][ 840/ 1200] Overall Loss 0.285701 Objective Loss 0.285701 LR 0.001000 Time 0.020102 -2022-12-06 11:01:30,190 - Epoch: [87][ 850/ 1200] Overall Loss 0.285790 Objective Loss 0.285790 LR 0.001000 Time 0.020090 -2022-12-06 11:01:30,381 - Epoch: [87][ 860/ 1200] Overall Loss 0.285850 Objective Loss 0.285850 LR 0.001000 Time 0.020079 -2022-12-06 11:01:30,572 - Epoch: [87][ 870/ 1200] Overall Loss 0.286036 Objective Loss 0.286036 LR 0.001000 Time 0.020067 -2022-12-06 11:01:30,763 - Epoch: [87][ 880/ 1200] Overall Loss 0.286168 Objective Loss 0.286168 LR 0.001000 Time 0.020055 -2022-12-06 11:01:30,955 - Epoch: [87][ 890/ 1200] Overall Loss 0.285993 Objective Loss 0.285993 LR 0.001000 Time 0.020045 -2022-12-06 11:01:31,146 - Epoch: [87][ 900/ 1200] Overall Loss 0.286215 Objective Loss 0.286215 LR 0.001000 Time 0.020034 -2022-12-06 11:01:31,338 - Epoch: [87][ 910/ 1200] Overall Loss 0.286508 Objective Loss 0.286508 LR 0.001000 Time 0.020024 -2022-12-06 11:01:31,530 - Epoch: [87][ 920/ 1200] Overall Loss 0.286713 Objective Loss 0.286713 LR 0.001000 Time 0.020014 -2022-12-06 11:01:31,721 - Epoch: [87][ 930/ 1200] Overall Loss 0.286573 Objective Loss 0.286573 LR 0.001000 Time 0.020004 -2022-12-06 11:01:31,913 - Epoch: [87][ 940/ 1200] Overall Loss 0.286626 Objective Loss 0.286626 LR 0.001000 Time 0.019994 -2022-12-06 11:01:32,104 - Epoch: [87][ 950/ 1200] Overall Loss 0.286711 Objective Loss 0.286711 LR 0.001000 Time 0.019985 -2022-12-06 11:01:32,295 - Epoch: [87][ 960/ 1200] Overall Loss 0.286666 Objective Loss 0.286666 LR 0.001000 Time 0.019975 -2022-12-06 11:01:32,486 - Epoch: [87][ 970/ 1200] Overall Loss 0.286733 Objective Loss 0.286733 LR 0.001000 Time 0.019966 -2022-12-06 11:01:32,678 - Epoch: [87][ 980/ 1200] Overall Loss 0.286866 Objective Loss 0.286866 LR 0.001000 Time 0.019957 -2022-12-06 11:01:32,869 - Epoch: [87][ 990/ 1200] Overall Loss 0.287218 Objective Loss 0.287218 LR 0.001000 Time 0.019948 -2022-12-06 11:01:33,061 - Epoch: [87][ 1000/ 1200] Overall Loss 0.287577 Objective Loss 0.287577 LR 0.001000 Time 0.019940 -2022-12-06 11:01:33,252 - Epoch: [87][ 1010/ 1200] Overall Loss 0.287651 Objective Loss 0.287651 LR 0.001000 Time 0.019931 -2022-12-06 11:01:33,443 - Epoch: [87][ 1020/ 1200] Overall Loss 0.287708 Objective Loss 0.287708 LR 0.001000 Time 0.019923 -2022-12-06 11:01:33,635 - Epoch: [87][ 1030/ 1200] Overall Loss 0.287525 Objective Loss 0.287525 LR 0.001000 Time 0.019915 -2022-12-06 11:01:33,827 - Epoch: [87][ 1040/ 1200] Overall Loss 0.287603 Objective Loss 0.287603 LR 0.001000 Time 0.019908 -2022-12-06 11:01:34,018 - Epoch: [87][ 1050/ 1200] Overall Loss 0.287598 Objective Loss 0.287598 LR 0.001000 Time 0.019900 -2022-12-06 11:01:34,209 - Epoch: [87][ 1060/ 1200] Overall Loss 0.287528 Objective Loss 0.287528 LR 0.001000 Time 0.019892 -2022-12-06 11:01:34,400 - Epoch: [87][ 1070/ 1200] Overall Loss 0.287519 Objective Loss 0.287519 LR 0.001000 Time 0.019883 -2022-12-06 11:01:34,592 - Epoch: [87][ 1080/ 1200] Overall Loss 0.287521 Objective Loss 0.287521 LR 0.001000 Time 0.019876 -2022-12-06 11:01:34,783 - Epoch: [87][ 1090/ 1200] Overall Loss 0.287609 Objective Loss 0.287609 LR 0.001000 Time 0.019869 -2022-12-06 11:01:34,975 - Epoch: [87][ 1100/ 1200] Overall Loss 0.287527 Objective Loss 0.287527 LR 0.001000 Time 0.019862 -2022-12-06 11:01:35,165 - Epoch: [87][ 1110/ 1200] Overall Loss 0.287438 Objective Loss 0.287438 LR 0.001000 Time 0.019854 -2022-12-06 11:01:35,357 - Epoch: [87][ 1120/ 1200] Overall Loss 0.287565 Objective Loss 0.287565 LR 0.001000 Time 0.019847 -2022-12-06 11:01:35,548 - Epoch: [87][ 1130/ 1200] Overall Loss 0.287957 Objective Loss 0.287957 LR 0.001000 Time 0.019841 -2022-12-06 11:01:35,739 - Epoch: [87][ 1140/ 1200] Overall Loss 0.288020 Objective Loss 0.288020 LR 0.001000 Time 0.019834 -2022-12-06 11:01:35,930 - Epoch: [87][ 1150/ 1200] Overall Loss 0.287849 Objective Loss 0.287849 LR 0.001000 Time 0.019827 -2022-12-06 11:01:36,122 - Epoch: [87][ 1160/ 1200] Overall Loss 0.288276 Objective Loss 0.288276 LR 0.001000 Time 0.019821 -2022-12-06 11:01:36,313 - Epoch: [87][ 1170/ 1200] Overall Loss 0.288488 Objective Loss 0.288488 LR 0.001000 Time 0.019814 -2022-12-06 11:01:36,504 - Epoch: [87][ 1180/ 1200] Overall Loss 0.288630 Objective Loss 0.288630 LR 0.001000 Time 0.019808 -2022-12-06 11:01:36,695 - Epoch: [87][ 1190/ 1200] Overall Loss 0.288471 Objective Loss 0.288471 LR 0.001000 Time 0.019801 -2022-12-06 11:01:36,927 - Epoch: [87][ 1200/ 1200] Overall Loss 0.288431 Objective Loss 0.288431 Top1 85.774059 Top5 98.326360 LR 0.001000 Time 0.019829 -2022-12-06 11:01:37,015 - --- validate (epoch=87)----------- -2022-12-06 11:01:37,016 - 34129 samples (256 per mini-batch) -2022-12-06 11:01:37,465 - Epoch: [87][ 10/ 134] Loss 0.306488 Top1 83.906250 Top5 97.617188 -2022-12-06 11:01:37,599 - Epoch: [87][ 20/ 134] Loss 0.304960 Top1 84.765625 Top5 97.519531 -2022-12-06 11:01:37,733 - Epoch: [87][ 30/ 134] Loss 0.304507 Top1 84.570312 Top5 97.486979 -2022-12-06 11:01:37,866 - Epoch: [87][ 40/ 134] Loss 0.305496 Top1 84.687500 Top5 97.666016 -2022-12-06 11:01:37,999 - Epoch: [87][ 50/ 134] Loss 0.303875 Top1 84.640625 Top5 97.687500 -2022-12-06 11:01:38,132 - Epoch: [87][ 60/ 134] Loss 0.301846 Top1 84.602865 Top5 97.701823 -2022-12-06 11:01:38,263 - Epoch: [87][ 70/ 134] Loss 0.299962 Top1 84.598214 Top5 97.751116 -2022-12-06 11:01:38,396 - Epoch: [87][ 80/ 134] Loss 0.301924 Top1 84.580078 Top5 97.758789 -2022-12-06 11:01:38,527 - Epoch: [87][ 90/ 134] Loss 0.306546 Top1 84.470486 Top5 97.803819 -2022-12-06 11:01:38,659 - Epoch: [87][ 100/ 134] Loss 0.307398 Top1 84.437500 Top5 97.796875 -2022-12-06 11:01:38,794 - Epoch: [87][ 110/ 134] Loss 0.305519 Top1 84.517045 Top5 97.823153 -2022-12-06 11:01:38,926 - Epoch: [87][ 120/ 134] Loss 0.304237 Top1 84.524740 Top5 97.835286 -2022-12-06 11:01:39,060 - Epoch: [87][ 130/ 134] Loss 0.307232 Top1 84.525240 Top5 97.821514 -2022-12-06 11:01:39,100 - Epoch: [87][ 134/ 134] Loss 0.307558 Top1 84.576167 Top5 97.814176 -2022-12-06 11:01:39,188 - ==> Top1: 84.576 Top5: 97.814 Loss: 0.308 - -2022-12-06 11:01:39,189 - ==> Confusion: -[[ 890 0 1 4 9 4 0 2 1 71 1 1 2 4 2 1 1 0 0 0 2] - [ 3 915 4 3 15 25 6 19 0 0 2 5 1 1 2 1 3 0 13 3 6] - [ 7 1 1003 6 4 1 26 12 2 4 5 6 1 1 2 8 1 2 2 0 9] - [ 3 1 25 926 2 2 1 1 3 1 7 2 4 1 21 0 0 1 12 0 7] - [ 13 5 3 0 960 4 0 1 1 6 0 2 1 1 8 9 3 0 0 1 2] - [ 4 13 0 3 8 955 6 23 3 2 1 15 2 13 3 1 4 0 1 5 7] - [ 1 2 12 0 0 2 1076 6 0 0 1 4 0 1 0 7 1 0 1 3 1] - [ 4 5 11 2 1 28 7 947 0 0 2 7 2 1 1 1 0 0 16 11 8] - [ 6 3 0 1 1 1 1 1 952 54 8 2 2 10 15 1 1 1 2 1 1] - [ 33 0 4 1 4 1 0 1 21 917 1 1 0 4 3 1 1 2 1 0 5] - [ 3 1 8 4 2 2 2 7 6 1 949 2 1 10 4 1 2 0 9 1 4] - [ 3 2 1 0 1 8 4 4 4 0 0 974 19 5 0 7 2 4 1 7 5] - [ 0 1 2 6 1 4 2 1 0 0 0 43 869 2 5 11 0 12 0 2 8] - [ 0 0 3 0 2 11 0 3 13 15 14 7 3 926 3 2 4 1 1 5 10] - [ 9 2 2 5 5 1 0 0 17 7 0 3 1 4 1052 2 0 0 7 2 11] - [ 4 0 1 0 4 0 8 0 0 0 1 11 3 1 0 997 2 5 0 3 3] - [ 4 3 2 0 2 5 3 1 0 0 1 3 3 1 1 18 1017 0 0 3 5] - [ 3 0 1 3 0 0 3 1 0 3 0 16 15 2 2 17 0 966 0 2 2] - [ 2 2 3 15 1 2 0 35 2 1 7 1 2 1 9 1 2 0 918 1 3] - [ 5 4 1 0 0 9 9 6 0 0 0 18 8 3 0 4 3 2 0 1002 6] - [ 176 199 228 100 167 203 113 203 94 123 162 152 364 289 189 145 189 51 158 269 9652]] - -2022-12-06 11:01:39,764 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:01:39,764 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:01:39,770 - - -2022-12-06 11:01:39,770 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:01:40,695 - Epoch: [88][ 10/ 1200] Overall Loss 0.266370 Objective Loss 0.266370 LR 0.001000 Time 0.092436 -2022-12-06 11:01:40,902 - Epoch: [88][ 20/ 1200] Overall Loss 0.273087 Objective Loss 0.273087 LR 0.001000 Time 0.056524 -2022-12-06 11:01:41,098 - Epoch: [88][ 30/ 1200] Overall Loss 0.269634 Objective Loss 0.269634 LR 0.001000 Time 0.044207 -2022-12-06 11:01:41,296 - Epoch: [88][ 40/ 1200] Overall Loss 0.274750 Objective Loss 0.274750 LR 0.001000 Time 0.038088 -2022-12-06 11:01:41,492 - Epoch: [88][ 50/ 1200] Overall Loss 0.283176 Objective Loss 0.283176 LR 0.001000 Time 0.034371 -2022-12-06 11:01:41,690 - Epoch: [88][ 60/ 1200] Overall Loss 0.280994 Objective Loss 0.280994 LR 0.001000 Time 0.031936 -2022-12-06 11:01:41,885 - Epoch: [88][ 70/ 1200] Overall Loss 0.280110 Objective Loss 0.280110 LR 0.001000 Time 0.030154 -2022-12-06 11:01:42,083 - Epoch: [88][ 80/ 1200] Overall Loss 0.280504 Objective Loss 0.280504 LR 0.001000 Time 0.028850 -2022-12-06 11:01:42,278 - Epoch: [88][ 90/ 1200] Overall Loss 0.278983 Objective Loss 0.278983 LR 0.001000 Time 0.027808 -2022-12-06 11:01:42,476 - Epoch: [88][ 100/ 1200] Overall Loss 0.275474 Objective Loss 0.275474 LR 0.001000 Time 0.026999 -2022-12-06 11:01:42,671 - Epoch: [88][ 110/ 1200] Overall Loss 0.274602 Objective Loss 0.274602 LR 0.001000 Time 0.026318 -2022-12-06 11:01:42,868 - Epoch: [88][ 120/ 1200] Overall Loss 0.275859 Objective Loss 0.275859 LR 0.001000 Time 0.025763 -2022-12-06 11:01:43,064 - Epoch: [88][ 130/ 1200] Overall Loss 0.277816 Objective Loss 0.277816 LR 0.001000 Time 0.025281 -2022-12-06 11:01:43,262 - Epoch: [88][ 140/ 1200] Overall Loss 0.276234 Objective Loss 0.276234 LR 0.001000 Time 0.024885 -2022-12-06 11:01:43,457 - Epoch: [88][ 150/ 1200] Overall Loss 0.277165 Objective Loss 0.277165 LR 0.001000 Time 0.024526 -2022-12-06 11:01:43,655 - Epoch: [88][ 160/ 1200] Overall Loss 0.277769 Objective Loss 0.277769 LR 0.001000 Time 0.024226 -2022-12-06 11:01:43,851 - Epoch: [88][ 170/ 1200] Overall Loss 0.280232 Objective Loss 0.280232 LR 0.001000 Time 0.023947 -2022-12-06 11:01:44,048 - Epoch: [88][ 180/ 1200] Overall Loss 0.281716 Objective Loss 0.281716 LR 0.001000 Time 0.023711 -2022-12-06 11:01:44,244 - Epoch: [88][ 190/ 1200] Overall Loss 0.281230 Objective Loss 0.281230 LR 0.001000 Time 0.023489 -2022-12-06 11:01:44,442 - Epoch: [88][ 200/ 1200] Overall Loss 0.282287 Objective Loss 0.282287 LR 0.001000 Time 0.023301 -2022-12-06 11:01:44,637 - Epoch: [88][ 210/ 1200] Overall Loss 0.282064 Objective Loss 0.282064 LR 0.001000 Time 0.023120 -2022-12-06 11:01:44,834 - Epoch: [88][ 220/ 1200] Overall Loss 0.282492 Objective Loss 0.282492 LR 0.001000 Time 0.022964 -2022-12-06 11:01:45,029 - Epoch: [88][ 230/ 1200] Overall Loss 0.281859 Objective Loss 0.281859 LR 0.001000 Time 0.022810 -2022-12-06 11:01:45,227 - Epoch: [88][ 240/ 1200] Overall Loss 0.281809 Objective Loss 0.281809 LR 0.001000 Time 0.022680 -2022-12-06 11:01:45,422 - Epoch: [88][ 250/ 1200] Overall Loss 0.282547 Objective Loss 0.282547 LR 0.001000 Time 0.022552 -2022-12-06 11:01:45,621 - Epoch: [88][ 260/ 1200] Overall Loss 0.284961 Objective Loss 0.284961 LR 0.001000 Time 0.022446 -2022-12-06 11:01:45,817 - Epoch: [88][ 270/ 1200] Overall Loss 0.284427 Objective Loss 0.284427 LR 0.001000 Time 0.022339 -2022-12-06 11:01:46,014 - Epoch: [88][ 280/ 1200] Overall Loss 0.285758 Objective Loss 0.285758 LR 0.001000 Time 0.022245 -2022-12-06 11:01:46,210 - Epoch: [88][ 290/ 1200] Overall Loss 0.286585 Objective Loss 0.286585 LR 0.001000 Time 0.022150 -2022-12-06 11:01:46,408 - Epoch: [88][ 300/ 1200] Overall Loss 0.286887 Objective Loss 0.286887 LR 0.001000 Time 0.022070 -2022-12-06 11:01:46,603 - Epoch: [88][ 310/ 1200] Overall Loss 0.286815 Objective Loss 0.286815 LR 0.001000 Time 0.021987 -2022-12-06 11:01:46,801 - Epoch: [88][ 320/ 1200] Overall Loss 0.285511 Objective Loss 0.285511 LR 0.001000 Time 0.021914 -2022-12-06 11:01:46,996 - Epoch: [88][ 330/ 1200] Overall Loss 0.286089 Objective Loss 0.286089 LR 0.001000 Time 0.021842 -2022-12-06 11:01:47,195 - Epoch: [88][ 340/ 1200] Overall Loss 0.285847 Objective Loss 0.285847 LR 0.001000 Time 0.021781 -2022-12-06 11:01:47,390 - Epoch: [88][ 350/ 1200] Overall Loss 0.286196 Objective Loss 0.286196 LR 0.001000 Time 0.021716 -2022-12-06 11:01:47,588 - Epoch: [88][ 360/ 1200] Overall Loss 0.286928 Objective Loss 0.286928 LR 0.001000 Time 0.021661 -2022-12-06 11:01:47,784 - Epoch: [88][ 370/ 1200] Overall Loss 0.287319 Objective Loss 0.287319 LR 0.001000 Time 0.021603 -2022-12-06 11:01:47,982 - Epoch: [88][ 380/ 1200] Overall Loss 0.287276 Objective Loss 0.287276 LR 0.001000 Time 0.021556 -2022-12-06 11:01:48,178 - Epoch: [88][ 390/ 1200] Overall Loss 0.287039 Objective Loss 0.287039 LR 0.001000 Time 0.021502 -2022-12-06 11:01:48,376 - Epoch: [88][ 400/ 1200] Overall Loss 0.286340 Objective Loss 0.286340 LR 0.001000 Time 0.021458 -2022-12-06 11:01:48,572 - Epoch: [88][ 410/ 1200] Overall Loss 0.286552 Objective Loss 0.286552 LR 0.001000 Time 0.021411 -2022-12-06 11:01:48,770 - Epoch: [88][ 420/ 1200] Overall Loss 0.286578 Objective Loss 0.286578 LR 0.001000 Time 0.021372 -2022-12-06 11:01:48,966 - Epoch: [88][ 430/ 1200] Overall Loss 0.285785 Objective Loss 0.285785 LR 0.001000 Time 0.021329 -2022-12-06 11:01:49,164 - Epoch: [88][ 440/ 1200] Overall Loss 0.285547 Objective Loss 0.285547 LR 0.001000 Time 0.021294 -2022-12-06 11:01:49,359 - Epoch: [88][ 450/ 1200] Overall Loss 0.285404 Objective Loss 0.285404 LR 0.001000 Time 0.021254 -2022-12-06 11:01:49,558 - Epoch: [88][ 460/ 1200] Overall Loss 0.285266 Objective Loss 0.285266 LR 0.001000 Time 0.021222 -2022-12-06 11:01:49,754 - Epoch: [88][ 470/ 1200] Overall Loss 0.285356 Objective Loss 0.285356 LR 0.001000 Time 0.021186 -2022-12-06 11:01:49,952 - Epoch: [88][ 480/ 1200] Overall Loss 0.285104 Objective Loss 0.285104 LR 0.001000 Time 0.021156 -2022-12-06 11:01:50,148 - Epoch: [88][ 490/ 1200] Overall Loss 0.284995 Objective Loss 0.284995 LR 0.001000 Time 0.021123 -2022-12-06 11:01:50,346 - Epoch: [88][ 500/ 1200] Overall Loss 0.284894 Objective Loss 0.284894 LR 0.001000 Time 0.021096 -2022-12-06 11:01:50,542 - Epoch: [88][ 510/ 1200] Overall Loss 0.284883 Objective Loss 0.284883 LR 0.001000 Time 0.021065 -2022-12-06 11:01:50,740 - Epoch: [88][ 520/ 1200] Overall Loss 0.284850 Objective Loss 0.284850 LR 0.001000 Time 0.021040 -2022-12-06 11:01:50,936 - Epoch: [88][ 530/ 1200] Overall Loss 0.285217 Objective Loss 0.285217 LR 0.001000 Time 0.021012 -2022-12-06 11:01:51,134 - Epoch: [88][ 540/ 1200] Overall Loss 0.285317 Objective Loss 0.285317 LR 0.001000 Time 0.020988 -2022-12-06 11:01:51,330 - Epoch: [88][ 550/ 1200] Overall Loss 0.284936 Objective Loss 0.284936 LR 0.001000 Time 0.020962 -2022-12-06 11:01:51,528 - Epoch: [88][ 560/ 1200] Overall Loss 0.284471 Objective Loss 0.284471 LR 0.001000 Time 0.020940 -2022-12-06 11:01:51,726 - Epoch: [88][ 570/ 1200] Overall Loss 0.284603 Objective Loss 0.284603 LR 0.001000 Time 0.020919 -2022-12-06 11:01:51,925 - Epoch: [88][ 580/ 1200] Overall Loss 0.284348 Objective Loss 0.284348 LR 0.001000 Time 0.020901 -2022-12-06 11:01:52,123 - Epoch: [88][ 590/ 1200] Overall Loss 0.284776 Objective Loss 0.284776 LR 0.001000 Time 0.020881 -2022-12-06 11:01:52,323 - Epoch: [88][ 600/ 1200] Overall Loss 0.285221 Objective Loss 0.285221 LR 0.001000 Time 0.020865 -2022-12-06 11:01:52,520 - Epoch: [88][ 610/ 1200] Overall Loss 0.285449 Objective Loss 0.285449 LR 0.001000 Time 0.020846 -2022-12-06 11:01:52,720 - Epoch: [88][ 620/ 1200] Overall Loss 0.286068 Objective Loss 0.286068 LR 0.001000 Time 0.020831 -2022-12-06 11:01:52,918 - Epoch: [88][ 630/ 1200] Overall Loss 0.286008 Objective Loss 0.286008 LR 0.001000 Time 0.020814 -2022-12-06 11:01:53,117 - Epoch: [88][ 640/ 1200] Overall Loss 0.286356 Objective Loss 0.286356 LR 0.001000 Time 0.020799 -2022-12-06 11:01:53,314 - Epoch: [88][ 650/ 1200] Overall Loss 0.286508 Objective Loss 0.286508 LR 0.001000 Time 0.020782 -2022-12-06 11:01:53,514 - Epoch: [88][ 660/ 1200] Overall Loss 0.286671 Objective Loss 0.286671 LR 0.001000 Time 0.020768 -2022-12-06 11:01:53,712 - Epoch: [88][ 670/ 1200] Overall Loss 0.286906 Objective Loss 0.286906 LR 0.001000 Time 0.020753 -2022-12-06 11:01:53,911 - Epoch: [88][ 680/ 1200] Overall Loss 0.286396 Objective Loss 0.286396 LR 0.001000 Time 0.020740 -2022-12-06 11:01:54,109 - Epoch: [88][ 690/ 1200] Overall Loss 0.286199 Objective Loss 0.286199 LR 0.001000 Time 0.020725 -2022-12-06 11:01:54,308 - Epoch: [88][ 700/ 1200] Overall Loss 0.286105 Objective Loss 0.286105 LR 0.001000 Time 0.020713 -2022-12-06 11:01:54,506 - Epoch: [88][ 710/ 1200] Overall Loss 0.286083 Objective Loss 0.286083 LR 0.001000 Time 0.020699 -2022-12-06 11:01:54,706 - Epoch: [88][ 720/ 1200] Overall Loss 0.285875 Objective Loss 0.285875 LR 0.001000 Time 0.020688 -2022-12-06 11:01:54,903 - Epoch: [88][ 730/ 1200] Overall Loss 0.285567 Objective Loss 0.285567 LR 0.001000 Time 0.020674 -2022-12-06 11:01:55,102 - Epoch: [88][ 740/ 1200] Overall Loss 0.285312 Objective Loss 0.285312 LR 0.001000 Time 0.020663 -2022-12-06 11:01:55,299 - Epoch: [88][ 750/ 1200] Overall Loss 0.285557 Objective Loss 0.285557 LR 0.001000 Time 0.020650 -2022-12-06 11:01:55,498 - Epoch: [88][ 760/ 1200] Overall Loss 0.285622 Objective Loss 0.285622 LR 0.001000 Time 0.020640 -2022-12-06 11:01:55,696 - Epoch: [88][ 770/ 1200] Overall Loss 0.285883 Objective Loss 0.285883 LR 0.001000 Time 0.020627 -2022-12-06 11:01:55,895 - Epoch: [88][ 780/ 1200] Overall Loss 0.285618 Objective Loss 0.285618 LR 0.001000 Time 0.020618 -2022-12-06 11:01:56,093 - Epoch: [88][ 790/ 1200] Overall Loss 0.285818 Objective Loss 0.285818 LR 0.001000 Time 0.020606 -2022-12-06 11:01:56,292 - Epoch: [88][ 800/ 1200] Overall Loss 0.285923 Objective Loss 0.285923 LR 0.001000 Time 0.020597 -2022-12-06 11:01:56,490 - Epoch: [88][ 810/ 1200] Overall Loss 0.286016 Objective Loss 0.286016 LR 0.001000 Time 0.020586 -2022-12-06 11:01:56,690 - Epoch: [88][ 820/ 1200] Overall Loss 0.286157 Objective Loss 0.286157 LR 0.001000 Time 0.020578 -2022-12-06 11:01:56,887 - Epoch: [88][ 830/ 1200] Overall Loss 0.286316 Objective Loss 0.286316 LR 0.001000 Time 0.020568 -2022-12-06 11:01:57,087 - Epoch: [88][ 840/ 1200] Overall Loss 0.286393 Objective Loss 0.286393 LR 0.001000 Time 0.020560 -2022-12-06 11:01:57,285 - Epoch: [88][ 850/ 1200] Overall Loss 0.286556 Objective Loss 0.286556 LR 0.001000 Time 0.020550 -2022-12-06 11:01:57,485 - Epoch: [88][ 860/ 1200] Overall Loss 0.286545 Objective Loss 0.286545 LR 0.001000 Time 0.020543 -2022-12-06 11:01:57,682 - Epoch: [88][ 870/ 1200] Overall Loss 0.286596 Objective Loss 0.286596 LR 0.001000 Time 0.020533 -2022-12-06 11:01:57,883 - Epoch: [88][ 880/ 1200] Overall Loss 0.286722 Objective Loss 0.286722 LR 0.001000 Time 0.020527 -2022-12-06 11:01:58,080 - Epoch: [88][ 890/ 1200] Overall Loss 0.286700 Objective Loss 0.286700 LR 0.001000 Time 0.020518 -2022-12-06 11:01:58,280 - Epoch: [88][ 900/ 1200] Overall Loss 0.286641 Objective Loss 0.286641 LR 0.001000 Time 0.020511 -2022-12-06 11:01:58,477 - Epoch: [88][ 910/ 1200] Overall Loss 0.286542 Objective Loss 0.286542 LR 0.001000 Time 0.020502 -2022-12-06 11:01:58,677 - Epoch: [88][ 920/ 1200] Overall Loss 0.286807 Objective Loss 0.286807 LR 0.001000 Time 0.020496 -2022-12-06 11:01:58,875 - Epoch: [88][ 930/ 1200] Overall Loss 0.286955 Objective Loss 0.286955 LR 0.001000 Time 0.020487 -2022-12-06 11:01:59,074 - Epoch: [88][ 940/ 1200] Overall Loss 0.287033 Objective Loss 0.287033 LR 0.001000 Time 0.020481 -2022-12-06 11:01:59,272 - Epoch: [88][ 950/ 1200] Overall Loss 0.286903 Objective Loss 0.286903 LR 0.001000 Time 0.020473 -2022-12-06 11:01:59,472 - Epoch: [88][ 960/ 1200] Overall Loss 0.286888 Objective Loss 0.286888 LR 0.001000 Time 0.020467 -2022-12-06 11:01:59,669 - Epoch: [88][ 970/ 1200] Overall Loss 0.286971 Objective Loss 0.286971 LR 0.001000 Time 0.020459 -2022-12-06 11:01:59,869 - Epoch: [88][ 980/ 1200] Overall Loss 0.286938 Objective Loss 0.286938 LR 0.001000 Time 0.020453 -2022-12-06 11:02:00,066 - Epoch: [88][ 990/ 1200] Overall Loss 0.286982 Objective Loss 0.286982 LR 0.001000 Time 0.020446 -2022-12-06 11:02:00,266 - Epoch: [88][ 1000/ 1200] Overall Loss 0.286844 Objective Loss 0.286844 LR 0.001000 Time 0.020440 -2022-12-06 11:02:00,463 - Epoch: [88][ 1010/ 1200] Overall Loss 0.286424 Objective Loss 0.286424 LR 0.001000 Time 0.020433 -2022-12-06 11:02:00,663 - Epoch: [88][ 1020/ 1200] Overall Loss 0.286493 Objective Loss 0.286493 LR 0.001000 Time 0.020427 -2022-12-06 11:02:00,860 - Epoch: [88][ 1030/ 1200] Overall Loss 0.286234 Objective Loss 0.286234 LR 0.001000 Time 0.020420 -2022-12-06 11:02:01,059 - Epoch: [88][ 1040/ 1200] Overall Loss 0.286479 Objective Loss 0.286479 LR 0.001000 Time 0.020415 -2022-12-06 11:02:01,256 - Epoch: [88][ 1050/ 1200] Overall Loss 0.286406 Objective Loss 0.286406 LR 0.001000 Time 0.020408 -2022-12-06 11:02:01,456 - Epoch: [88][ 1060/ 1200] Overall Loss 0.286239 Objective Loss 0.286239 LR 0.001000 Time 0.020403 -2022-12-06 11:02:01,654 - Epoch: [88][ 1070/ 1200] Overall Loss 0.286259 Objective Loss 0.286259 LR 0.001000 Time 0.020396 -2022-12-06 11:02:01,853 - Epoch: [88][ 1080/ 1200] Overall Loss 0.286140 Objective Loss 0.286140 LR 0.001000 Time 0.020392 -2022-12-06 11:02:02,051 - Epoch: [88][ 1090/ 1200] Overall Loss 0.286039 Objective Loss 0.286039 LR 0.001000 Time 0.020386 -2022-12-06 11:02:02,250 - Epoch: [88][ 1100/ 1200] Overall Loss 0.285931 Objective Loss 0.285931 LR 0.001000 Time 0.020381 -2022-12-06 11:02:02,447 - Epoch: [88][ 1110/ 1200] Overall Loss 0.285970 Objective Loss 0.285970 LR 0.001000 Time 0.020375 -2022-12-06 11:02:02,647 - Epoch: [88][ 1120/ 1200] Overall Loss 0.285825 Objective Loss 0.285825 LR 0.001000 Time 0.020370 -2022-12-06 11:02:02,844 - Epoch: [88][ 1130/ 1200] Overall Loss 0.285699 Objective Loss 0.285699 LR 0.001000 Time 0.020364 -2022-12-06 11:02:03,043 - Epoch: [88][ 1140/ 1200] Overall Loss 0.285530 Objective Loss 0.285530 LR 0.001000 Time 0.020360 -2022-12-06 11:02:03,240 - Epoch: [88][ 1150/ 1200] Overall Loss 0.285587 Objective Loss 0.285587 LR 0.001000 Time 0.020353 -2022-12-06 11:02:03,439 - Epoch: [88][ 1160/ 1200] Overall Loss 0.285720 Objective Loss 0.285720 LR 0.001000 Time 0.020349 -2022-12-06 11:02:03,636 - Epoch: [88][ 1170/ 1200] Overall Loss 0.285635 Objective Loss 0.285635 LR 0.001000 Time 0.020343 -2022-12-06 11:02:03,835 - Epoch: [88][ 1180/ 1200] Overall Loss 0.285502 Objective Loss 0.285502 LR 0.001000 Time 0.020339 -2022-12-06 11:02:04,032 - Epoch: [88][ 1190/ 1200] Overall Loss 0.285412 Objective Loss 0.285412 LR 0.001000 Time 0.020333 -2022-12-06 11:02:04,268 - Epoch: [88][ 1200/ 1200] Overall Loss 0.285679 Objective Loss 0.285679 Top1 82.426778 Top5 97.698745 LR 0.001000 Time 0.020360 -2022-12-06 11:02:04,362 - --- validate (epoch=88)----------- -2022-12-06 11:02:04,362 - 34129 samples (256 per mini-batch) -2022-12-06 11:02:04,925 - Epoch: [88][ 10/ 134] Loss 0.321243 Top1 84.609375 Top5 98.046875 -2022-12-06 11:02:05,060 - Epoch: [88][ 20/ 134] Loss 0.312137 Top1 84.453125 Top5 97.988281 -2022-12-06 11:02:05,195 - Epoch: [88][ 30/ 134] Loss 0.314115 Top1 84.518229 Top5 97.799479 -2022-12-06 11:02:05,328 - Epoch: [88][ 40/ 134] Loss 0.312772 Top1 84.638672 Top5 97.753906 -2022-12-06 11:02:05,461 - Epoch: [88][ 50/ 134] Loss 0.309194 Top1 84.539062 Top5 97.742188 -2022-12-06 11:02:05,592 - Epoch: [88][ 60/ 134] Loss 0.304962 Top1 84.726562 Top5 97.799479 -2022-12-06 11:02:05,725 - Epoch: [88][ 70/ 134] Loss 0.302589 Top1 84.765625 Top5 97.823661 -2022-12-06 11:02:05,856 - Epoch: [88][ 80/ 134] Loss 0.299511 Top1 84.750977 Top5 97.866211 -2022-12-06 11:02:05,989 - Epoch: [88][ 90/ 134] Loss 0.299098 Top1 84.765625 Top5 97.851562 -2022-12-06 11:02:06,120 - Epoch: [88][ 100/ 134] Loss 0.300421 Top1 84.679688 Top5 97.886719 -2022-12-06 11:02:06,253 - Epoch: [88][ 110/ 134] Loss 0.303879 Top1 84.644886 Top5 97.837358 -2022-12-06 11:02:06,383 - Epoch: [88][ 120/ 134] Loss 0.303032 Top1 84.671224 Top5 97.832031 -2022-12-06 11:02:06,517 - Epoch: [88][ 130/ 134] Loss 0.304293 Top1 84.639423 Top5 97.827524 -2022-12-06 11:02:06,557 - Epoch: [88][ 134/ 134] Loss 0.303000 Top1 84.705090 Top5 97.849336 -2022-12-06 11:02:06,658 - ==> Top1: 84.705 Top5: 97.849 Loss: 0.303 - -2022-12-06 11:02:06,660 - ==> Confusion: -[[ 887 0 1 1 12 9 0 2 7 57 1 2 2 3 5 2 0 0 2 0 3] - [ 0 924 1 3 11 26 4 15 0 1 4 7 2 2 1 2 5 2 8 2 7] - [ 5 1 981 10 6 3 32 10 0 2 9 5 2 0 4 2 0 2 10 1 18] - [ 2 3 13 942 0 5 2 2 0 0 10 4 6 2 12 0 0 2 11 0 4] - [ 11 5 1 0 955 6 1 2 2 4 1 5 1 0 10 3 4 1 2 2 4] - [ 3 15 0 1 12 962 1 19 2 3 0 15 6 12 0 2 2 0 6 4 4] - [ 1 3 7 0 2 7 1063 7 0 0 2 4 1 3 0 0 0 1 1 8 8] - [ 2 10 5 2 2 27 6 955 0 0 4 8 0 2 1 0 2 0 23 1 4] - [ 7 8 0 0 0 3 0 0 949 48 13 5 1 10 14 2 1 0 3 0 0] - [ 47 0 4 0 9 3 0 3 23 890 1 1 1 11 0 1 0 1 0 0 6] - [ 1 2 3 5 1 0 1 8 8 3 947 1 1 15 5 1 0 0 10 0 7] - [ 2 1 0 0 1 8 5 3 2 0 0 978 29 1 0 3 1 4 2 10 1] - [ 2 1 3 2 1 3 2 2 0 0 0 35 878 2 1 7 0 17 1 4 8] - [ 0 0 0 2 4 11 0 3 16 17 2 12 4 929 1 1 5 1 1 3 11] - [ 8 2 1 11 6 1 0 1 21 3 5 4 4 3 1042 0 0 3 3 0 12] - [ 1 0 3 0 5 2 2 0 1 0 1 13 5 3 0 976 5 16 0 7 3] - [ 2 6 1 0 6 1 4 0 1 0 1 5 1 2 1 7 1021 1 1 3 8] - [ 3 0 1 5 0 0 1 3 1 2 0 13 15 2 2 10 1 972 0 1 4] - [ 2 3 3 12 2 2 2 22 5 1 5 0 1 2 6 1 2 0 933 1 3] - [ 4 4 1 0 0 5 11 14 0 0 1 27 6 8 0 4 4 3 0 982 6] - [ 144 200 143 128 172 205 92 178 101 98 194 176 395 309 143 104 161 93 208 251 9731]] - -2022-12-06 11:02:07,241 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:02:07,242 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:02:07,247 - - -2022-12-06 11:02:07,248 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:02:08,191 - Epoch: [89][ 10/ 1200] Overall Loss 0.258951 Objective Loss 0.258951 LR 0.001000 Time 0.094249 -2022-12-06 11:02:08,391 - Epoch: [89][ 20/ 1200] Overall Loss 0.266352 Objective Loss 0.266352 LR 0.001000 Time 0.057112 -2022-12-06 11:02:08,583 - Epoch: [89][ 30/ 1200] Overall Loss 0.261562 Objective Loss 0.261562 LR 0.001000 Time 0.044455 -2022-12-06 11:02:08,775 - Epoch: [89][ 40/ 1200] Overall Loss 0.270104 Objective Loss 0.270104 LR 0.001000 Time 0.038123 -2022-12-06 11:02:08,967 - Epoch: [89][ 50/ 1200] Overall Loss 0.276171 Objective Loss 0.276171 LR 0.001000 Time 0.034322 -2022-12-06 11:02:09,159 - Epoch: [89][ 60/ 1200] Overall Loss 0.276045 Objective Loss 0.276045 LR 0.001000 Time 0.031792 -2022-12-06 11:02:09,350 - Epoch: [89][ 70/ 1200] Overall Loss 0.280186 Objective Loss 0.280186 LR 0.001000 Time 0.029970 -2022-12-06 11:02:09,542 - Epoch: [89][ 80/ 1200] Overall Loss 0.278847 Objective Loss 0.278847 LR 0.001000 Time 0.028618 -2022-12-06 11:02:09,733 - Epoch: [89][ 90/ 1200] Overall Loss 0.278281 Objective Loss 0.278281 LR 0.001000 Time 0.027558 -2022-12-06 11:02:09,925 - Epoch: [89][ 100/ 1200] Overall Loss 0.278843 Objective Loss 0.278843 LR 0.001000 Time 0.026712 -2022-12-06 11:02:10,116 - Epoch: [89][ 110/ 1200] Overall Loss 0.279364 Objective Loss 0.279364 LR 0.001000 Time 0.026020 -2022-12-06 11:02:10,308 - Epoch: [89][ 120/ 1200] Overall Loss 0.277524 Objective Loss 0.277524 LR 0.001000 Time 0.025443 -2022-12-06 11:02:10,499 - Epoch: [89][ 130/ 1200] Overall Loss 0.275956 Objective Loss 0.275956 LR 0.001000 Time 0.024954 -2022-12-06 11:02:10,691 - Epoch: [89][ 140/ 1200] Overall Loss 0.273764 Objective Loss 0.273764 LR 0.001000 Time 0.024538 -2022-12-06 11:02:10,882 - Epoch: [89][ 150/ 1200] Overall Loss 0.274775 Objective Loss 0.274775 LR 0.001000 Time 0.024171 -2022-12-06 11:02:11,074 - Epoch: [89][ 160/ 1200] Overall Loss 0.275986 Objective Loss 0.275986 LR 0.001000 Time 0.023855 -2022-12-06 11:02:11,265 - Epoch: [89][ 170/ 1200] Overall Loss 0.277432 Objective Loss 0.277432 LR 0.001000 Time 0.023575 -2022-12-06 11:02:11,457 - Epoch: [89][ 180/ 1200] Overall Loss 0.277750 Objective Loss 0.277750 LR 0.001000 Time 0.023326 -2022-12-06 11:02:11,647 - Epoch: [89][ 190/ 1200] Overall Loss 0.277119 Objective Loss 0.277119 LR 0.001000 Time 0.023098 -2022-12-06 11:02:11,838 - Epoch: [89][ 200/ 1200] Overall Loss 0.277234 Objective Loss 0.277234 LR 0.001000 Time 0.022897 -2022-12-06 11:02:12,030 - Epoch: [89][ 210/ 1200] Overall Loss 0.276867 Objective Loss 0.276867 LR 0.001000 Time 0.022715 -2022-12-06 11:02:12,221 - Epoch: [89][ 220/ 1200] Overall Loss 0.275334 Objective Loss 0.275334 LR 0.001000 Time 0.022551 -2022-12-06 11:02:12,413 - Epoch: [89][ 230/ 1200] Overall Loss 0.273752 Objective Loss 0.273752 LR 0.001000 Time 0.022399 -2022-12-06 11:02:12,604 - Epoch: [89][ 240/ 1200] Overall Loss 0.273275 Objective Loss 0.273275 LR 0.001000 Time 0.022261 -2022-12-06 11:02:12,795 - Epoch: [89][ 250/ 1200] Overall Loss 0.273921 Objective Loss 0.273921 LR 0.001000 Time 0.022133 -2022-12-06 11:02:12,987 - Epoch: [89][ 260/ 1200] Overall Loss 0.274338 Objective Loss 0.274338 LR 0.001000 Time 0.022016 -2022-12-06 11:02:13,178 - Epoch: [89][ 270/ 1200] Overall Loss 0.274815 Objective Loss 0.274815 LR 0.001000 Time 0.021907 -2022-12-06 11:02:13,369 - Epoch: [89][ 280/ 1200] Overall Loss 0.274877 Objective Loss 0.274877 LR 0.001000 Time 0.021806 -2022-12-06 11:02:13,561 - Epoch: [89][ 290/ 1200] Overall Loss 0.275924 Objective Loss 0.275924 LR 0.001000 Time 0.021712 -2022-12-06 11:02:13,753 - Epoch: [89][ 300/ 1200] Overall Loss 0.276038 Objective Loss 0.276038 LR 0.001000 Time 0.021626 -2022-12-06 11:02:13,943 - Epoch: [89][ 310/ 1200] Overall Loss 0.276210 Objective Loss 0.276210 LR 0.001000 Time 0.021542 -2022-12-06 11:02:14,135 - Epoch: [89][ 320/ 1200] Overall Loss 0.276361 Objective Loss 0.276361 LR 0.001000 Time 0.021466 -2022-12-06 11:02:14,326 - Epoch: [89][ 330/ 1200] Overall Loss 0.276413 Objective Loss 0.276413 LR 0.001000 Time 0.021394 -2022-12-06 11:02:14,518 - Epoch: [89][ 340/ 1200] Overall Loss 0.276877 Objective Loss 0.276877 LR 0.001000 Time 0.021326 -2022-12-06 11:02:14,709 - Epoch: [89][ 350/ 1200] Overall Loss 0.277690 Objective Loss 0.277690 LR 0.001000 Time 0.021260 -2022-12-06 11:02:14,900 - Epoch: [89][ 360/ 1200] Overall Loss 0.277140 Objective Loss 0.277140 LR 0.001000 Time 0.021200 -2022-12-06 11:02:15,092 - Epoch: [89][ 370/ 1200] Overall Loss 0.276733 Objective Loss 0.276733 LR 0.001000 Time 0.021143 -2022-12-06 11:02:15,283 - Epoch: [89][ 380/ 1200] Overall Loss 0.276814 Objective Loss 0.276814 LR 0.001000 Time 0.021089 -2022-12-06 11:02:15,474 - Epoch: [89][ 390/ 1200] Overall Loss 0.276709 Objective Loss 0.276709 LR 0.001000 Time 0.021037 -2022-12-06 11:02:15,666 - Epoch: [89][ 400/ 1200] Overall Loss 0.277189 Objective Loss 0.277189 LR 0.001000 Time 0.020989 -2022-12-06 11:02:15,858 - Epoch: [89][ 410/ 1200] Overall Loss 0.277353 Objective Loss 0.277353 LR 0.001000 Time 0.020943 -2022-12-06 11:02:16,050 - Epoch: [89][ 420/ 1200] Overall Loss 0.276698 Objective Loss 0.276698 LR 0.001000 Time 0.020900 -2022-12-06 11:02:16,241 - Epoch: [89][ 430/ 1200] Overall Loss 0.275964 Objective Loss 0.275964 LR 0.001000 Time 0.020858 -2022-12-06 11:02:16,432 - Epoch: [89][ 440/ 1200] Overall Loss 0.275963 Objective Loss 0.275963 LR 0.001000 Time 0.020818 -2022-12-06 11:02:16,623 - Epoch: [89][ 450/ 1200] Overall Loss 0.275785 Objective Loss 0.275785 LR 0.001000 Time 0.020778 -2022-12-06 11:02:16,815 - Epoch: [89][ 460/ 1200] Overall Loss 0.276013 Objective Loss 0.276013 LR 0.001000 Time 0.020742 -2022-12-06 11:02:17,007 - Epoch: [89][ 470/ 1200] Overall Loss 0.276520 Objective Loss 0.276520 LR 0.001000 Time 0.020707 -2022-12-06 11:02:17,199 - Epoch: [89][ 480/ 1200] Overall Loss 0.277140 Objective Loss 0.277140 LR 0.001000 Time 0.020675 -2022-12-06 11:02:17,390 - Epoch: [89][ 490/ 1200] Overall Loss 0.277937 Objective Loss 0.277937 LR 0.001000 Time 0.020642 -2022-12-06 11:02:17,582 - Epoch: [89][ 500/ 1200] Overall Loss 0.278553 Objective Loss 0.278553 LR 0.001000 Time 0.020612 -2022-12-06 11:02:17,773 - Epoch: [89][ 510/ 1200] Overall Loss 0.278895 Objective Loss 0.278895 LR 0.001000 Time 0.020582 -2022-12-06 11:02:17,965 - Epoch: [89][ 520/ 1200] Overall Loss 0.278874 Objective Loss 0.278874 LR 0.001000 Time 0.020553 -2022-12-06 11:02:18,157 - Epoch: [89][ 530/ 1200] Overall Loss 0.278968 Objective Loss 0.278968 LR 0.001000 Time 0.020526 -2022-12-06 11:02:18,348 - Epoch: [89][ 540/ 1200] Overall Loss 0.278621 Objective Loss 0.278621 LR 0.001000 Time 0.020499 -2022-12-06 11:02:18,539 - Epoch: [89][ 550/ 1200] Overall Loss 0.279006 Objective Loss 0.279006 LR 0.001000 Time 0.020473 -2022-12-06 11:02:18,731 - Epoch: [89][ 560/ 1200] Overall Loss 0.278710 Objective Loss 0.278710 LR 0.001000 Time 0.020449 -2022-12-06 11:02:18,922 - Epoch: [89][ 570/ 1200] Overall Loss 0.278221 Objective Loss 0.278221 LR 0.001000 Time 0.020425 -2022-12-06 11:02:19,114 - Epoch: [89][ 580/ 1200] Overall Loss 0.278591 Objective Loss 0.278591 LR 0.001000 Time 0.020402 -2022-12-06 11:02:19,305 - Epoch: [89][ 590/ 1200] Overall Loss 0.278702 Objective Loss 0.278702 LR 0.001000 Time 0.020380 -2022-12-06 11:02:19,497 - Epoch: [89][ 600/ 1200] Overall Loss 0.278721 Objective Loss 0.278721 LR 0.001000 Time 0.020359 -2022-12-06 11:02:19,689 - Epoch: [89][ 610/ 1200] Overall Loss 0.279047 Objective Loss 0.279047 LR 0.001000 Time 0.020339 -2022-12-06 11:02:19,879 - Epoch: [89][ 620/ 1200] Overall Loss 0.279229 Objective Loss 0.279229 LR 0.001000 Time 0.020317 -2022-12-06 11:02:20,070 - Epoch: [89][ 630/ 1200] Overall Loss 0.278933 Objective Loss 0.278933 LR 0.001000 Time 0.020296 -2022-12-06 11:02:20,260 - Epoch: [89][ 640/ 1200] Overall Loss 0.278628 Objective Loss 0.278628 LR 0.001000 Time 0.020276 -2022-12-06 11:02:20,451 - Epoch: [89][ 650/ 1200] Overall Loss 0.278917 Objective Loss 0.278917 LR 0.001000 Time 0.020257 -2022-12-06 11:02:20,642 - Epoch: [89][ 660/ 1200] Overall Loss 0.279032 Objective Loss 0.279032 LR 0.001000 Time 0.020238 -2022-12-06 11:02:20,833 - Epoch: [89][ 670/ 1200] Overall Loss 0.278951 Objective Loss 0.278951 LR 0.001000 Time 0.020220 -2022-12-06 11:02:21,024 - Epoch: [89][ 680/ 1200] Overall Loss 0.278753 Objective Loss 0.278753 LR 0.001000 Time 0.020203 -2022-12-06 11:02:21,215 - Epoch: [89][ 690/ 1200] Overall Loss 0.278828 Objective Loss 0.278828 LR 0.001000 Time 0.020186 -2022-12-06 11:02:21,406 - Epoch: [89][ 700/ 1200] Overall Loss 0.278922 Objective Loss 0.278922 LR 0.001000 Time 0.020170 -2022-12-06 11:02:21,597 - Epoch: [89][ 710/ 1200] Overall Loss 0.279054 Objective Loss 0.279054 LR 0.001000 Time 0.020153 -2022-12-06 11:02:21,787 - Epoch: [89][ 720/ 1200] Overall Loss 0.279006 Objective Loss 0.279006 LR 0.001000 Time 0.020138 -2022-12-06 11:02:21,978 - Epoch: [89][ 730/ 1200] Overall Loss 0.279149 Objective Loss 0.279149 LR 0.001000 Time 0.020122 -2022-12-06 11:02:22,169 - Epoch: [89][ 740/ 1200] Overall Loss 0.279042 Objective Loss 0.279042 LR 0.001000 Time 0.020107 -2022-12-06 11:02:22,359 - Epoch: [89][ 750/ 1200] Overall Loss 0.279243 Objective Loss 0.279243 LR 0.001000 Time 0.020093 -2022-12-06 11:02:22,551 - Epoch: [89][ 760/ 1200] Overall Loss 0.279355 Objective Loss 0.279355 LR 0.001000 Time 0.020079 -2022-12-06 11:02:22,741 - Epoch: [89][ 770/ 1200] Overall Loss 0.279565 Objective Loss 0.279565 LR 0.001000 Time 0.020065 -2022-12-06 11:02:22,932 - Epoch: [89][ 780/ 1200] Overall Loss 0.279631 Objective Loss 0.279631 LR 0.001000 Time 0.020052 -2022-12-06 11:02:23,123 - Epoch: [89][ 790/ 1200] Overall Loss 0.279453 Objective Loss 0.279453 LR 0.001000 Time 0.020039 -2022-12-06 11:02:23,313 - Epoch: [89][ 800/ 1200] Overall Loss 0.279603 Objective Loss 0.279603 LR 0.001000 Time 0.020026 -2022-12-06 11:02:23,504 - Epoch: [89][ 810/ 1200] Overall Loss 0.279792 Objective Loss 0.279792 LR 0.001000 Time 0.020014 -2022-12-06 11:02:23,695 - Epoch: [89][ 820/ 1200] Overall Loss 0.279907 Objective Loss 0.279907 LR 0.001000 Time 0.020002 -2022-12-06 11:02:23,886 - Epoch: [89][ 830/ 1200] Overall Loss 0.280128 Objective Loss 0.280128 LR 0.001000 Time 0.019990 -2022-12-06 11:02:24,077 - Epoch: [89][ 840/ 1200] Overall Loss 0.280201 Objective Loss 0.280201 LR 0.001000 Time 0.019979 -2022-12-06 11:02:24,268 - Epoch: [89][ 850/ 1200] Overall Loss 0.280355 Objective Loss 0.280355 LR 0.001000 Time 0.019967 -2022-12-06 11:02:24,459 - Epoch: [89][ 860/ 1200] Overall Loss 0.280411 Objective Loss 0.280411 LR 0.001000 Time 0.019957 -2022-12-06 11:02:24,649 - Epoch: [89][ 870/ 1200] Overall Loss 0.280753 Objective Loss 0.280753 LR 0.001000 Time 0.019946 -2022-12-06 11:02:24,840 - Epoch: [89][ 880/ 1200] Overall Loss 0.280903 Objective Loss 0.280903 LR 0.001000 Time 0.019935 -2022-12-06 11:02:25,031 - Epoch: [89][ 890/ 1200] Overall Loss 0.281020 Objective Loss 0.281020 LR 0.001000 Time 0.019925 -2022-12-06 11:02:25,222 - Epoch: [89][ 900/ 1200] Overall Loss 0.281161 Objective Loss 0.281161 LR 0.001000 Time 0.019915 -2022-12-06 11:02:25,412 - Epoch: [89][ 910/ 1200] Overall Loss 0.281223 Objective Loss 0.281223 LR 0.001000 Time 0.019905 -2022-12-06 11:02:25,602 - Epoch: [89][ 920/ 1200] Overall Loss 0.281181 Objective Loss 0.281181 LR 0.001000 Time 0.019895 -2022-12-06 11:02:25,793 - Epoch: [89][ 930/ 1200] Overall Loss 0.281005 Objective Loss 0.281005 LR 0.001000 Time 0.019885 -2022-12-06 11:02:25,984 - Epoch: [89][ 940/ 1200] Overall Loss 0.281015 Objective Loss 0.281015 LR 0.001000 Time 0.019876 -2022-12-06 11:02:26,174 - Epoch: [89][ 950/ 1200] Overall Loss 0.280909 Objective Loss 0.280909 LR 0.001000 Time 0.019867 -2022-12-06 11:02:26,366 - Epoch: [89][ 960/ 1200] Overall Loss 0.281016 Objective Loss 0.281016 LR 0.001000 Time 0.019859 -2022-12-06 11:02:26,556 - Epoch: [89][ 970/ 1200] Overall Loss 0.280958 Objective Loss 0.280958 LR 0.001000 Time 0.019850 -2022-12-06 11:02:26,747 - Epoch: [89][ 980/ 1200] Overall Loss 0.280766 Objective Loss 0.280766 LR 0.001000 Time 0.019842 -2022-12-06 11:02:26,938 - Epoch: [89][ 990/ 1200] Overall Loss 0.280967 Objective Loss 0.280967 LR 0.001000 Time 0.019834 -2022-12-06 11:02:27,130 - Epoch: [89][ 1000/ 1200] Overall Loss 0.280674 Objective Loss 0.280674 LR 0.001000 Time 0.019826 -2022-12-06 11:02:27,320 - Epoch: [89][ 1010/ 1200] Overall Loss 0.280573 Objective Loss 0.280573 LR 0.001000 Time 0.019818 -2022-12-06 11:02:27,511 - Epoch: [89][ 1020/ 1200] Overall Loss 0.280655 Objective Loss 0.280655 LR 0.001000 Time 0.019810 -2022-12-06 11:02:27,702 - Epoch: [89][ 1030/ 1200] Overall Loss 0.280985 Objective Loss 0.280985 LR 0.001000 Time 0.019803 -2022-12-06 11:02:27,893 - Epoch: [89][ 1040/ 1200] Overall Loss 0.281219 Objective Loss 0.281219 LR 0.001000 Time 0.019796 -2022-12-06 11:02:28,084 - Epoch: [89][ 1050/ 1200] Overall Loss 0.281326 Objective Loss 0.281326 LR 0.001000 Time 0.019789 -2022-12-06 11:02:28,275 - Epoch: [89][ 1060/ 1200] Overall Loss 0.281568 Objective Loss 0.281568 LR 0.001000 Time 0.019781 -2022-12-06 11:02:28,466 - Epoch: [89][ 1070/ 1200] Overall Loss 0.281806 Objective Loss 0.281806 LR 0.001000 Time 0.019774 -2022-12-06 11:02:28,657 - Epoch: [89][ 1080/ 1200] Overall Loss 0.282095 Objective Loss 0.282095 LR 0.001000 Time 0.019767 -2022-12-06 11:02:28,847 - Epoch: [89][ 1090/ 1200] Overall Loss 0.281864 Objective Loss 0.281864 LR 0.001000 Time 0.019760 -2022-12-06 11:02:29,038 - Epoch: [89][ 1100/ 1200] Overall Loss 0.282080 Objective Loss 0.282080 LR 0.001000 Time 0.019754 -2022-12-06 11:02:29,229 - Epoch: [89][ 1110/ 1200] Overall Loss 0.282033 Objective Loss 0.282033 LR 0.001000 Time 0.019748 -2022-12-06 11:02:29,421 - Epoch: [89][ 1120/ 1200] Overall Loss 0.281980 Objective Loss 0.281980 LR 0.001000 Time 0.019741 -2022-12-06 11:02:29,611 - Epoch: [89][ 1130/ 1200] Overall Loss 0.282102 Objective Loss 0.282102 LR 0.001000 Time 0.019735 -2022-12-06 11:02:29,803 - Epoch: [89][ 1140/ 1200] Overall Loss 0.282164 Objective Loss 0.282164 LR 0.001000 Time 0.019729 -2022-12-06 11:02:29,994 - Epoch: [89][ 1150/ 1200] Overall Loss 0.282107 Objective Loss 0.282107 LR 0.001000 Time 0.019723 -2022-12-06 11:02:30,185 - Epoch: [89][ 1160/ 1200] Overall Loss 0.282192 Objective Loss 0.282192 LR 0.001000 Time 0.019718 -2022-12-06 11:02:30,376 - Epoch: [89][ 1170/ 1200] Overall Loss 0.282288 Objective Loss 0.282288 LR 0.001000 Time 0.019712 -2022-12-06 11:02:30,567 - Epoch: [89][ 1180/ 1200] Overall Loss 0.282339 Objective Loss 0.282339 LR 0.001000 Time 0.019706 -2022-12-06 11:02:30,757 - Epoch: [89][ 1190/ 1200] Overall Loss 0.282688 Objective Loss 0.282688 LR 0.001000 Time 0.019700 -2022-12-06 11:02:30,980 - Epoch: [89][ 1200/ 1200] Overall Loss 0.282906 Objective Loss 0.282906 Top1 83.891213 Top5 98.535565 LR 0.001000 Time 0.019721 -2022-12-06 11:02:31,068 - --- validate (epoch=89)----------- -2022-12-06 11:02:31,068 - 34129 samples (256 per mini-batch) -2022-12-06 11:02:31,513 - Epoch: [89][ 10/ 134] Loss 0.317350 Top1 83.593750 Top5 97.656250 -2022-12-06 11:02:31,647 - Epoch: [89][ 20/ 134] Loss 0.300777 Top1 84.160156 Top5 97.480469 -2022-12-06 11:02:31,777 - Epoch: [89][ 30/ 134] Loss 0.294209 Top1 83.815104 Top5 97.617188 -2022-12-06 11:02:31,918 - Epoch: [89][ 40/ 134] Loss 0.302505 Top1 83.818359 Top5 97.705078 -2022-12-06 11:02:32,062 - Epoch: [89][ 50/ 134] Loss 0.306427 Top1 83.726562 Top5 97.718750 -2022-12-06 11:02:32,207 - Epoch: [89][ 60/ 134] Loss 0.304129 Top1 83.906250 Top5 97.701823 -2022-12-06 11:02:32,339 - Epoch: [89][ 70/ 134] Loss 0.307075 Top1 83.733259 Top5 97.695312 -2022-12-06 11:02:32,486 - Epoch: [89][ 80/ 134] Loss 0.305130 Top1 83.735352 Top5 97.666016 -2022-12-06 11:02:32,626 - Epoch: [89][ 90/ 134] Loss 0.308489 Top1 83.632812 Top5 97.638889 -2022-12-06 11:02:32,765 - Epoch: [89][ 100/ 134] Loss 0.307957 Top1 83.742188 Top5 97.605469 -2022-12-06 11:02:32,908 - Epoch: [89][ 110/ 134] Loss 0.308698 Top1 83.657670 Top5 97.602983 -2022-12-06 11:02:33,053 - Epoch: [89][ 120/ 134] Loss 0.308594 Top1 83.642578 Top5 97.600911 -2022-12-06 11:02:33,189 - Epoch: [89][ 130/ 134] Loss 0.307302 Top1 83.734976 Top5 97.632212 -2022-12-06 11:02:33,227 - Epoch: [89][ 134/ 134] Loss 0.307308 Top1 83.720590 Top5 97.632512 -2022-12-06 11:02:33,315 - ==> Top1: 83.721 Top5: 97.633 Loss: 0.307 - -2022-12-06 11:02:33,316 - ==> Confusion: -[[ 901 1 3 1 8 7 0 2 3 49 0 2 2 3 5 3 2 2 0 0 2] - [ 3 928 4 1 6 27 0 16 0 1 6 5 5 0 1 1 6 2 9 2 4] - [ 4 4 1011 9 4 0 25 7 1 1 3 5 1 1 0 2 2 3 5 3 12] - [ 7 5 25 918 0 0 1 0 1 0 7 0 4 3 16 1 1 5 13 2 11] - [ 13 8 4 0 945 6 2 1 1 8 1 2 1 0 8 6 6 3 3 1 1] - [ 5 16 1 2 8 962 1 19 3 2 0 9 6 10 0 2 2 2 2 12 5] - [ 0 5 14 1 0 2 1069 2 1 0 0 1 0 2 0 6 2 2 1 9 1] - [ 2 10 12 2 1 33 7 940 0 0 3 8 4 1 0 0 0 0 16 13 2] - [ 10 4 0 1 1 5 0 0 958 42 7 1 2 10 13 1 3 3 2 1 0] - [ 67 1 6 0 3 4 0 2 26 869 1 2 0 8 2 1 0 1 1 1 6] - [ 1 2 9 11 1 3 3 4 12 1 923 2 3 11 3 1 4 0 17 1 7] - [ 4 2 3 0 1 16 2 6 1 0 0 921 41 4 0 9 7 15 0 15 4] - [ 2 2 2 6 0 4 2 0 1 0 0 22 883 1 2 12 2 19 0 2 7] - [ 2 3 1 0 0 10 0 1 15 12 5 5 4 946 1 1 5 4 1 1 6] - [ 10 1 3 13 6 3 0 1 26 5 1 3 4 8 1025 0 2 2 11 0 6] - [ 1 0 2 0 3 0 6 0 2 1 0 4 3 2 0 994 5 15 0 2 3] - [ 4 2 2 1 4 1 2 0 0 1 0 0 0 1 0 14 1026 3 0 3 8] - [ 4 0 0 1 0 1 0 1 3 2 1 4 18 1 1 15 0 981 0 0 3] - [ 4 7 5 9 0 4 0 32 2 2 2 1 4 3 6 0 2 0 922 1 2] - [ 5 4 2 0 1 5 3 6 0 0 0 9 7 4 0 4 5 3 1 1017 4] - [ 171 334 225 106 116 224 120 197 84 117 136 84 383 361 171 150 255 108 184 275 9425]] - -2022-12-06 11:02:33,980 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:02:33,981 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:02:33,986 - - -2022-12-06 11:02:33,986 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:02:34,923 - Epoch: [90][ 10/ 1200] Overall Loss 0.256245 Objective Loss 0.256245 LR 0.001000 Time 0.093600 -2022-12-06 11:02:35,117 - Epoch: [90][ 20/ 1200] Overall Loss 0.271186 Objective Loss 0.271186 LR 0.001000 Time 0.056474 -2022-12-06 11:02:35,309 - Epoch: [90][ 30/ 1200] Overall Loss 0.266749 Objective Loss 0.266749 LR 0.001000 Time 0.044020 -2022-12-06 11:02:35,500 - Epoch: [90][ 40/ 1200] Overall Loss 0.264520 Objective Loss 0.264520 LR 0.001000 Time 0.037774 -2022-12-06 11:02:35,691 - Epoch: [90][ 50/ 1200] Overall Loss 0.268319 Objective Loss 0.268319 LR 0.001000 Time 0.034032 -2022-12-06 11:02:35,881 - Epoch: [90][ 60/ 1200] Overall Loss 0.271727 Objective Loss 0.271727 LR 0.001000 Time 0.031517 -2022-12-06 11:02:36,072 - Epoch: [90][ 70/ 1200] Overall Loss 0.271714 Objective Loss 0.271714 LR 0.001000 Time 0.029741 -2022-12-06 11:02:36,263 - Epoch: [90][ 80/ 1200] Overall Loss 0.278082 Objective Loss 0.278082 LR 0.001000 Time 0.028397 -2022-12-06 11:02:36,453 - Epoch: [90][ 90/ 1200] Overall Loss 0.276763 Objective Loss 0.276763 LR 0.001000 Time 0.027349 -2022-12-06 11:02:36,643 - Epoch: [90][ 100/ 1200] Overall Loss 0.274027 Objective Loss 0.274027 LR 0.001000 Time 0.026506 -2022-12-06 11:02:36,833 - Epoch: [90][ 110/ 1200] Overall Loss 0.271634 Objective Loss 0.271634 LR 0.001000 Time 0.025824 -2022-12-06 11:02:37,024 - Epoch: [90][ 120/ 1200] Overall Loss 0.270371 Objective Loss 0.270371 LR 0.001000 Time 0.025259 -2022-12-06 11:02:37,215 - Epoch: [90][ 130/ 1200] Overall Loss 0.268678 Objective Loss 0.268678 LR 0.001000 Time 0.024778 -2022-12-06 11:02:37,406 - Epoch: [90][ 140/ 1200] Overall Loss 0.270488 Objective Loss 0.270488 LR 0.001000 Time 0.024371 -2022-12-06 11:02:37,597 - Epoch: [90][ 150/ 1200] Overall Loss 0.269847 Objective Loss 0.269847 LR 0.001000 Time 0.024016 -2022-12-06 11:02:37,787 - Epoch: [90][ 160/ 1200] Overall Loss 0.269278 Objective Loss 0.269278 LR 0.001000 Time 0.023700 -2022-12-06 11:02:37,979 - Epoch: [90][ 170/ 1200] Overall Loss 0.271054 Objective Loss 0.271054 LR 0.001000 Time 0.023429 -2022-12-06 11:02:38,170 - Epoch: [90][ 180/ 1200] Overall Loss 0.270971 Objective Loss 0.270971 LR 0.001000 Time 0.023184 -2022-12-06 11:02:38,361 - Epoch: [90][ 190/ 1200] Overall Loss 0.271913 Objective Loss 0.271913 LR 0.001000 Time 0.022966 -2022-12-06 11:02:38,552 - Epoch: [90][ 200/ 1200] Overall Loss 0.272830 Objective Loss 0.272830 LR 0.001000 Time 0.022769 -2022-12-06 11:02:38,742 - Epoch: [90][ 210/ 1200] Overall Loss 0.273137 Objective Loss 0.273137 LR 0.001000 Time 0.022588 -2022-12-06 11:02:38,933 - Epoch: [90][ 220/ 1200] Overall Loss 0.273484 Objective Loss 0.273484 LR 0.001000 Time 0.022427 -2022-12-06 11:02:39,124 - Epoch: [90][ 230/ 1200] Overall Loss 0.273229 Objective Loss 0.273229 LR 0.001000 Time 0.022279 -2022-12-06 11:02:39,314 - Epoch: [90][ 240/ 1200] Overall Loss 0.273737 Objective Loss 0.273737 LR 0.001000 Time 0.022143 -2022-12-06 11:02:39,505 - Epoch: [90][ 250/ 1200] Overall Loss 0.273568 Objective Loss 0.273568 LR 0.001000 Time 0.022017 -2022-12-06 11:02:39,695 - Epoch: [90][ 260/ 1200] Overall Loss 0.274111 Objective Loss 0.274111 LR 0.001000 Time 0.021900 -2022-12-06 11:02:39,885 - Epoch: [90][ 270/ 1200] Overall Loss 0.273482 Objective Loss 0.273482 LR 0.001000 Time 0.021789 -2022-12-06 11:02:40,075 - Epoch: [90][ 280/ 1200] Overall Loss 0.272910 Objective Loss 0.272910 LR 0.001000 Time 0.021688 -2022-12-06 11:02:40,265 - Epoch: [90][ 290/ 1200] Overall Loss 0.274184 Objective Loss 0.274184 LR 0.001000 Time 0.021595 -2022-12-06 11:02:40,456 - Epoch: [90][ 300/ 1200] Overall Loss 0.274027 Objective Loss 0.274027 LR 0.001000 Time 0.021508 -2022-12-06 11:02:40,646 - Epoch: [90][ 310/ 1200] Overall Loss 0.274113 Objective Loss 0.274113 LR 0.001000 Time 0.021426 -2022-12-06 11:02:40,837 - Epoch: [90][ 320/ 1200] Overall Loss 0.274019 Objective Loss 0.274019 LR 0.001000 Time 0.021353 -2022-12-06 11:02:41,027 - Epoch: [90][ 330/ 1200] Overall Loss 0.274351 Objective Loss 0.274351 LR 0.001000 Time 0.021279 -2022-12-06 11:02:41,217 - Epoch: [90][ 340/ 1200] Overall Loss 0.274753 Objective Loss 0.274753 LR 0.001000 Time 0.021211 -2022-12-06 11:02:41,407 - Epoch: [90][ 350/ 1200] Overall Loss 0.275335 Objective Loss 0.275335 LR 0.001000 Time 0.021146 -2022-12-06 11:02:41,597 - Epoch: [90][ 360/ 1200] Overall Loss 0.276278 Objective Loss 0.276278 LR 0.001000 Time 0.021085 -2022-12-06 11:02:41,787 - Epoch: [90][ 370/ 1200] Overall Loss 0.276096 Objective Loss 0.276096 LR 0.001000 Time 0.021028 -2022-12-06 11:02:41,978 - Epoch: [90][ 380/ 1200] Overall Loss 0.275867 Objective Loss 0.275867 LR 0.001000 Time 0.020973 -2022-12-06 11:02:42,168 - Epoch: [90][ 390/ 1200] Overall Loss 0.276836 Objective Loss 0.276836 LR 0.001000 Time 0.020922 -2022-12-06 11:02:42,358 - Epoch: [90][ 400/ 1200] Overall Loss 0.276162 Objective Loss 0.276162 LR 0.001000 Time 0.020874 -2022-12-06 11:02:42,548 - Epoch: [90][ 410/ 1200] Overall Loss 0.275569 Objective Loss 0.275569 LR 0.001000 Time 0.020827 -2022-12-06 11:02:42,739 - Epoch: [90][ 420/ 1200] Overall Loss 0.274911 Objective Loss 0.274911 LR 0.001000 Time 0.020785 -2022-12-06 11:02:42,929 - Epoch: [90][ 430/ 1200] Overall Loss 0.275247 Objective Loss 0.275247 LR 0.001000 Time 0.020741 -2022-12-06 11:02:43,119 - Epoch: [90][ 440/ 1200] Overall Loss 0.276079 Objective Loss 0.276079 LR 0.001000 Time 0.020701 -2022-12-06 11:02:43,310 - Epoch: [90][ 450/ 1200] Overall Loss 0.276208 Objective Loss 0.276208 LR 0.001000 Time 0.020663 -2022-12-06 11:02:43,501 - Epoch: [90][ 460/ 1200] Overall Loss 0.276619 Objective Loss 0.276619 LR 0.001000 Time 0.020627 -2022-12-06 11:02:43,691 - Epoch: [90][ 470/ 1200] Overall Loss 0.276712 Objective Loss 0.276712 LR 0.001000 Time 0.020593 -2022-12-06 11:02:43,882 - Epoch: [90][ 480/ 1200] Overall Loss 0.276814 Objective Loss 0.276814 LR 0.001000 Time 0.020560 -2022-12-06 11:02:44,072 - Epoch: [90][ 490/ 1200] Overall Loss 0.277016 Objective Loss 0.277016 LR 0.001000 Time 0.020528 -2022-12-06 11:02:44,262 - Epoch: [90][ 500/ 1200] Overall Loss 0.276868 Objective Loss 0.276868 LR 0.001000 Time 0.020496 -2022-12-06 11:02:44,452 - Epoch: [90][ 510/ 1200] Overall Loss 0.276573 Objective Loss 0.276573 LR 0.001000 Time 0.020466 -2022-12-06 11:02:44,643 - Epoch: [90][ 520/ 1200] Overall Loss 0.275868 Objective Loss 0.275868 LR 0.001000 Time 0.020438 -2022-12-06 11:02:44,833 - Epoch: [90][ 530/ 1200] Overall Loss 0.275827 Objective Loss 0.275827 LR 0.001000 Time 0.020411 -2022-12-06 11:02:45,023 - Epoch: [90][ 540/ 1200] Overall Loss 0.276162 Objective Loss 0.276162 LR 0.001000 Time 0.020384 -2022-12-06 11:02:45,214 - Epoch: [90][ 550/ 1200] Overall Loss 0.276005 Objective Loss 0.276005 LR 0.001000 Time 0.020358 -2022-12-06 11:02:45,404 - Epoch: [90][ 560/ 1200] Overall Loss 0.275955 Objective Loss 0.275955 LR 0.001000 Time 0.020333 -2022-12-06 11:02:45,595 - Epoch: [90][ 570/ 1200] Overall Loss 0.276256 Objective Loss 0.276256 LR 0.001000 Time 0.020310 -2022-12-06 11:02:45,785 - Epoch: [90][ 580/ 1200] Overall Loss 0.275909 Objective Loss 0.275909 LR 0.001000 Time 0.020287 -2022-12-06 11:02:45,976 - Epoch: [90][ 590/ 1200] Overall Loss 0.275711 Objective Loss 0.275711 LR 0.001000 Time 0.020265 -2022-12-06 11:02:46,166 - Epoch: [90][ 600/ 1200] Overall Loss 0.275855 Objective Loss 0.275855 LR 0.001000 Time 0.020244 -2022-12-06 11:02:46,356 - Epoch: [90][ 610/ 1200] Overall Loss 0.275931 Objective Loss 0.275931 LR 0.001000 Time 0.020223 -2022-12-06 11:02:46,547 - Epoch: [90][ 620/ 1200] Overall Loss 0.275903 Objective Loss 0.275903 LR 0.001000 Time 0.020203 -2022-12-06 11:02:46,738 - Epoch: [90][ 630/ 1200] Overall Loss 0.276236 Objective Loss 0.276236 LR 0.001000 Time 0.020185 -2022-12-06 11:02:46,928 - Epoch: [90][ 640/ 1200] Overall Loss 0.276268 Objective Loss 0.276268 LR 0.001000 Time 0.020166 -2022-12-06 11:02:47,119 - Epoch: [90][ 650/ 1200] Overall Loss 0.276379 Objective Loss 0.276379 LR 0.001000 Time 0.020148 -2022-12-06 11:02:47,310 - Epoch: [90][ 660/ 1200] Overall Loss 0.276411 Objective Loss 0.276411 LR 0.001000 Time 0.020131 -2022-12-06 11:02:47,500 - Epoch: [90][ 670/ 1200] Overall Loss 0.276539 Objective Loss 0.276539 LR 0.001000 Time 0.020114 -2022-12-06 11:02:47,691 - Epoch: [90][ 680/ 1200] Overall Loss 0.276942 Objective Loss 0.276942 LR 0.001000 Time 0.020098 -2022-12-06 11:02:47,882 - Epoch: [90][ 690/ 1200] Overall Loss 0.277093 Objective Loss 0.277093 LR 0.001000 Time 0.020084 -2022-12-06 11:02:48,072 - Epoch: [90][ 700/ 1200] Overall Loss 0.276628 Objective Loss 0.276628 LR 0.001000 Time 0.020067 -2022-12-06 11:02:48,264 - Epoch: [90][ 710/ 1200] Overall Loss 0.276949 Objective Loss 0.276949 LR 0.001000 Time 0.020053 -2022-12-06 11:02:48,454 - Epoch: [90][ 720/ 1200] Overall Loss 0.277452 Objective Loss 0.277452 LR 0.001000 Time 0.020038 -2022-12-06 11:02:48,645 - Epoch: [90][ 730/ 1200] Overall Loss 0.277691 Objective Loss 0.277691 LR 0.001000 Time 0.020025 -2022-12-06 11:02:48,835 - Epoch: [90][ 740/ 1200] Overall Loss 0.277643 Objective Loss 0.277643 LR 0.001000 Time 0.020011 -2022-12-06 11:02:49,026 - Epoch: [90][ 750/ 1200] Overall Loss 0.277770 Objective Loss 0.277770 LR 0.001000 Time 0.019998 -2022-12-06 11:02:49,217 - Epoch: [90][ 760/ 1200] Overall Loss 0.277655 Objective Loss 0.277655 LR 0.001000 Time 0.019986 -2022-12-06 11:02:49,407 - Epoch: [90][ 770/ 1200] Overall Loss 0.277593 Objective Loss 0.277593 LR 0.001000 Time 0.019972 -2022-12-06 11:02:49,598 - Epoch: [90][ 780/ 1200] Overall Loss 0.278105 Objective Loss 0.278105 LR 0.001000 Time 0.019960 -2022-12-06 11:02:49,789 - Epoch: [90][ 790/ 1200] Overall Loss 0.278208 Objective Loss 0.278208 LR 0.001000 Time 0.019948 -2022-12-06 11:02:49,979 - Epoch: [90][ 800/ 1200] Overall Loss 0.278313 Objective Loss 0.278313 LR 0.001000 Time 0.019936 -2022-12-06 11:02:50,171 - Epoch: [90][ 810/ 1200] Overall Loss 0.278461 Objective Loss 0.278461 LR 0.001000 Time 0.019926 -2022-12-06 11:02:50,363 - Epoch: [90][ 820/ 1200] Overall Loss 0.278370 Objective Loss 0.278370 LR 0.001000 Time 0.019916 -2022-12-06 11:02:50,554 - Epoch: [90][ 830/ 1200] Overall Loss 0.278468 Objective Loss 0.278468 LR 0.001000 Time 0.019906 -2022-12-06 11:02:50,745 - Epoch: [90][ 840/ 1200] Overall Loss 0.278262 Objective Loss 0.278262 LR 0.001000 Time 0.019896 -2022-12-06 11:02:50,936 - Epoch: [90][ 850/ 1200] Overall Loss 0.278048 Objective Loss 0.278048 LR 0.001000 Time 0.019886 -2022-12-06 11:02:51,127 - Epoch: [90][ 860/ 1200] Overall Loss 0.277963 Objective Loss 0.277963 LR 0.001000 Time 0.019876 -2022-12-06 11:02:51,318 - Epoch: [90][ 870/ 1200] Overall Loss 0.278243 Objective Loss 0.278243 LR 0.001000 Time 0.019866 -2022-12-06 11:02:51,509 - Epoch: [90][ 880/ 1200] Overall Loss 0.278143 Objective Loss 0.278143 LR 0.001000 Time 0.019857 -2022-12-06 11:02:51,700 - Epoch: [90][ 890/ 1200] Overall Loss 0.278537 Objective Loss 0.278537 LR 0.001000 Time 0.019848 -2022-12-06 11:02:51,891 - Epoch: [90][ 900/ 1200] Overall Loss 0.278550 Objective Loss 0.278550 LR 0.001000 Time 0.019838 -2022-12-06 11:02:52,082 - Epoch: [90][ 910/ 1200] Overall Loss 0.278541 Objective Loss 0.278541 LR 0.001000 Time 0.019829 -2022-12-06 11:02:52,272 - Epoch: [90][ 920/ 1200] Overall Loss 0.278460 Objective Loss 0.278460 LR 0.001000 Time 0.019820 -2022-12-06 11:02:52,463 - Epoch: [90][ 930/ 1200] Overall Loss 0.278431 Objective Loss 0.278431 LR 0.001000 Time 0.019812 -2022-12-06 11:02:52,653 - Epoch: [90][ 940/ 1200] Overall Loss 0.278578 Objective Loss 0.278578 LR 0.001000 Time 0.019803 -2022-12-06 11:02:52,844 - Epoch: [90][ 950/ 1200] Overall Loss 0.278500 Objective Loss 0.278500 LR 0.001000 Time 0.019795 -2022-12-06 11:02:53,035 - Epoch: [90][ 960/ 1200] Overall Loss 0.278413 Objective Loss 0.278413 LR 0.001000 Time 0.019787 -2022-12-06 11:02:53,225 - Epoch: [90][ 970/ 1200] Overall Loss 0.278281 Objective Loss 0.278281 LR 0.001000 Time 0.019778 -2022-12-06 11:02:53,415 - Epoch: [90][ 980/ 1200] Overall Loss 0.278419 Objective Loss 0.278419 LR 0.001000 Time 0.019770 -2022-12-06 11:02:53,606 - Epoch: [90][ 990/ 1200] Overall Loss 0.278575 Objective Loss 0.278575 LR 0.001000 Time 0.019762 -2022-12-06 11:02:53,796 - Epoch: [90][ 1000/ 1200] Overall Loss 0.278760 Objective Loss 0.278760 LR 0.001000 Time 0.019755 -2022-12-06 11:02:53,987 - Epoch: [90][ 1010/ 1200] Overall Loss 0.278762 Objective Loss 0.278762 LR 0.001000 Time 0.019747 -2022-12-06 11:02:54,178 - Epoch: [90][ 1020/ 1200] Overall Loss 0.278855 Objective Loss 0.278855 LR 0.001000 Time 0.019741 -2022-12-06 11:02:54,369 - Epoch: [90][ 1030/ 1200] Overall Loss 0.278952 Objective Loss 0.278952 LR 0.001000 Time 0.019734 -2022-12-06 11:02:54,560 - Epoch: [90][ 1040/ 1200] Overall Loss 0.279242 Objective Loss 0.279242 LR 0.001000 Time 0.019727 -2022-12-06 11:02:54,751 - Epoch: [90][ 1050/ 1200] Overall Loss 0.279418 Objective Loss 0.279418 LR 0.001000 Time 0.019721 -2022-12-06 11:02:54,942 - Epoch: [90][ 1060/ 1200] Overall Loss 0.279466 Objective Loss 0.279466 LR 0.001000 Time 0.019715 -2022-12-06 11:02:55,133 - Epoch: [90][ 1070/ 1200] Overall Loss 0.279678 Objective Loss 0.279678 LR 0.001000 Time 0.019708 -2022-12-06 11:02:55,324 - Epoch: [90][ 1080/ 1200] Overall Loss 0.279571 Objective Loss 0.279571 LR 0.001000 Time 0.019702 -2022-12-06 11:02:55,515 - Epoch: [90][ 1090/ 1200] Overall Loss 0.279652 Objective Loss 0.279652 LR 0.001000 Time 0.019696 -2022-12-06 11:02:55,706 - Epoch: [90][ 1100/ 1200] Overall Loss 0.279929 Objective Loss 0.279929 LR 0.001000 Time 0.019690 -2022-12-06 11:02:55,897 - Epoch: [90][ 1110/ 1200] Overall Loss 0.280141 Objective Loss 0.280141 LR 0.001000 Time 0.019684 -2022-12-06 11:02:56,087 - Epoch: [90][ 1120/ 1200] Overall Loss 0.280399 Objective Loss 0.280399 LR 0.001000 Time 0.019678 -2022-12-06 11:02:56,279 - Epoch: [90][ 1130/ 1200] Overall Loss 0.280483 Objective Loss 0.280483 LR 0.001000 Time 0.019673 -2022-12-06 11:02:56,469 - Epoch: [90][ 1140/ 1200] Overall Loss 0.280431 Objective Loss 0.280431 LR 0.001000 Time 0.019667 -2022-12-06 11:02:56,661 - Epoch: [90][ 1150/ 1200] Overall Loss 0.280128 Objective Loss 0.280128 LR 0.001000 Time 0.019662 -2022-12-06 11:02:56,851 - Epoch: [90][ 1160/ 1200] Overall Loss 0.280174 Objective Loss 0.280174 LR 0.001000 Time 0.019656 -2022-12-06 11:02:57,043 - Epoch: [90][ 1170/ 1200] Overall Loss 0.280267 Objective Loss 0.280267 LR 0.001000 Time 0.019651 -2022-12-06 11:02:57,233 - Epoch: [90][ 1180/ 1200] Overall Loss 0.280528 Objective Loss 0.280528 LR 0.001000 Time 0.019646 -2022-12-06 11:02:57,424 - Epoch: [90][ 1190/ 1200] Overall Loss 0.280600 Objective Loss 0.280600 LR 0.001000 Time 0.019640 -2022-12-06 11:02:57,654 - Epoch: [90][ 1200/ 1200] Overall Loss 0.280681 Objective Loss 0.280681 Top1 84.728033 Top5 98.117155 LR 0.001000 Time 0.019668 -2022-12-06 11:02:57,743 - --- validate (epoch=90)----------- -2022-12-06 11:02:57,743 - 34129 samples (256 per mini-batch) -2022-12-06 11:02:58,186 - Epoch: [90][ 10/ 134] Loss 0.304011 Top1 85.234375 Top5 97.656250 -2022-12-06 11:02:58,318 - Epoch: [90][ 20/ 134] Loss 0.299640 Top1 85.195312 Top5 97.792969 -2022-12-06 11:02:58,449 - Epoch: [90][ 30/ 134] Loss 0.301309 Top1 84.687500 Top5 97.734375 -2022-12-06 11:02:58,581 - Epoch: [90][ 40/ 134] Loss 0.300715 Top1 84.658203 Top5 97.783203 -2022-12-06 11:02:58,713 - Epoch: [90][ 50/ 134] Loss 0.304100 Top1 84.703125 Top5 97.812500 -2022-12-06 11:02:58,845 - Epoch: [90][ 60/ 134] Loss 0.301765 Top1 84.667969 Top5 97.877604 -2022-12-06 11:02:58,977 - Epoch: [90][ 70/ 134] Loss 0.303013 Top1 84.587054 Top5 97.845982 -2022-12-06 11:02:59,108 - Epoch: [90][ 80/ 134] Loss 0.299850 Top1 84.653320 Top5 97.792969 -2022-12-06 11:02:59,239 - Epoch: [90][ 90/ 134] Loss 0.299756 Top1 84.748264 Top5 97.795139 -2022-12-06 11:02:59,371 - Epoch: [90][ 100/ 134] Loss 0.303175 Top1 84.710938 Top5 97.804688 -2022-12-06 11:02:59,504 - Epoch: [90][ 110/ 134] Loss 0.301678 Top1 84.687500 Top5 97.840909 -2022-12-06 11:02:59,636 - Epoch: [90][ 120/ 134] Loss 0.302038 Top1 84.746094 Top5 97.858073 -2022-12-06 11:02:59,770 - Epoch: [90][ 130/ 134] Loss 0.302413 Top1 84.747596 Top5 97.851562 -2022-12-06 11:02:59,808 - Epoch: [90][ 134/ 134] Loss 0.300954 Top1 84.743180 Top5 97.881567 -2022-12-06 11:02:59,895 - ==> Top1: 84.743 Top5: 97.882 Loss: 0.301 - -2022-12-06 11:02:59,896 - ==> Confusion: -[[ 907 0 2 3 5 5 1 1 0 57 0 3 2 0 1 2 2 0 2 2 1] - [ 2 911 1 2 12 22 2 23 1 0 8 6 3 2 2 1 3 1 10 7 8] - [ 6 4 991 7 0 3 37 8 0 4 5 4 3 1 1 3 3 2 3 7 11] - [ 2 2 25 931 1 2 1 0 0 1 8 1 3 4 13 0 0 2 14 1 9] - [ 12 4 2 0 945 4 1 2 1 7 3 4 0 4 8 5 9 2 1 2 4] - [ 5 16 1 4 5 947 6 25 2 5 1 14 2 19 1 2 5 0 1 6 2] - [ 2 1 11 0 0 2 1074 2 0 0 0 3 0 0 0 6 0 2 2 12 1] - [ 2 8 9 2 0 27 9 938 0 0 6 6 1 2 0 1 1 0 23 10 9] - [ 11 3 0 0 0 1 1 0 969 41 12 2 1 5 10 2 0 1 1 1 3] - [ 52 0 2 0 5 1 1 5 16 893 2 1 0 11 3 1 0 1 0 2 5] - [ 2 0 10 5 1 2 2 5 5 3 939 3 0 15 4 1 2 0 9 6 5] - [ 3 2 0 0 1 11 1 5 0 1 1 958 30 3 0 9 7 3 1 13 2] - [ 1 0 1 4 0 2 2 1 0 0 1 38 870 3 1 11 5 14 1 8 6] - [ 1 0 1 0 1 6 0 3 15 16 7 10 3 942 3 1 5 0 0 2 7] - [ 8 3 2 13 8 3 0 1 17 6 1 0 2 3 1040 1 0 0 10 4 8] - [ 1 0 1 0 2 0 4 0 0 0 0 9 2 2 2 992 9 9 0 6 4] - [ 3 1 2 1 3 4 2 0 1 0 0 0 0 2 1 10 1028 1 1 4 8] - [ 2 0 0 2 0 0 0 1 0 3 0 15 27 3 0 16 2 961 1 1 2] - [ 2 6 2 16 1 2 1 26 2 1 8 2 3 1 9 1 2 2 915 5 1] - [ 2 5 3 0 0 4 8 7 0 0 0 15 3 6 0 7 4 4 0 1003 9] - [ 158 226 226 109 105 143 102 178 94 101 188 124 369 302 133 143 224 69 195 276 9761]] - -2022-12-06 11:03:00,571 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:03:00,571 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:03:00,577 - - -2022-12-06 11:03:00,577 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:03:01,510 - Epoch: [91][ 10/ 1200] Overall Loss 0.307071 Objective Loss 0.307071 LR 0.001000 Time 0.093139 -2022-12-06 11:03:01,715 - Epoch: [91][ 20/ 1200] Overall Loss 0.287853 Objective Loss 0.287853 LR 0.001000 Time 0.056814 -2022-12-06 11:03:01,914 - Epoch: [91][ 30/ 1200] Overall Loss 0.276522 Objective Loss 0.276522 LR 0.001000 Time 0.044486 -2022-12-06 11:03:02,111 - Epoch: [91][ 40/ 1200] Overall Loss 0.274456 Objective Loss 0.274456 LR 0.001000 Time 0.038278 -2022-12-06 11:03:02,309 - Epoch: [91][ 50/ 1200] Overall Loss 0.273656 Objective Loss 0.273656 LR 0.001000 Time 0.034581 -2022-12-06 11:03:02,507 - Epoch: [91][ 60/ 1200] Overall Loss 0.273370 Objective Loss 0.273370 LR 0.001000 Time 0.032092 -2022-12-06 11:03:02,705 - Epoch: [91][ 70/ 1200] Overall Loss 0.270693 Objective Loss 0.270693 LR 0.001000 Time 0.030333 -2022-12-06 11:03:02,901 - Epoch: [91][ 80/ 1200] Overall Loss 0.267849 Objective Loss 0.267849 LR 0.001000 Time 0.028990 -2022-12-06 11:03:03,100 - Epoch: [91][ 90/ 1200] Overall Loss 0.271956 Objective Loss 0.271956 LR 0.001000 Time 0.027970 -2022-12-06 11:03:03,296 - Epoch: [91][ 100/ 1200] Overall Loss 0.272462 Objective Loss 0.272462 LR 0.001000 Time 0.027132 -2022-12-06 11:03:03,496 - Epoch: [91][ 110/ 1200] Overall Loss 0.271754 Objective Loss 0.271754 LR 0.001000 Time 0.026473 -2022-12-06 11:03:03,692 - Epoch: [91][ 120/ 1200] Overall Loss 0.273265 Objective Loss 0.273265 LR 0.001000 Time 0.025899 -2022-12-06 11:03:03,892 - Epoch: [91][ 130/ 1200] Overall Loss 0.272679 Objective Loss 0.272679 LR 0.001000 Time 0.025438 -2022-12-06 11:03:04,088 - Epoch: [91][ 140/ 1200] Overall Loss 0.272638 Objective Loss 0.272638 LR 0.001000 Time 0.025016 -2022-12-06 11:03:04,287 - Epoch: [91][ 150/ 1200] Overall Loss 0.271750 Objective Loss 0.271750 LR 0.001000 Time 0.024672 -2022-12-06 11:03:04,484 - Epoch: [91][ 160/ 1200] Overall Loss 0.273564 Objective Loss 0.273564 LR 0.001000 Time 0.024357 -2022-12-06 11:03:04,682 - Epoch: [91][ 170/ 1200] Overall Loss 0.272977 Objective Loss 0.272977 LR 0.001000 Time 0.024090 -2022-12-06 11:03:04,879 - Epoch: [91][ 180/ 1200] Overall Loss 0.272020 Objective Loss 0.272020 LR 0.001000 Time 0.023839 -2022-12-06 11:03:05,078 - Epoch: [91][ 190/ 1200] Overall Loss 0.272953 Objective Loss 0.272953 LR 0.001000 Time 0.023630 -2022-12-06 11:03:05,275 - Epoch: [91][ 200/ 1200] Overall Loss 0.273862 Objective Loss 0.273862 LR 0.001000 Time 0.023430 -2022-12-06 11:03:05,473 - Epoch: [91][ 210/ 1200] Overall Loss 0.275005 Objective Loss 0.275005 LR 0.001000 Time 0.023255 -2022-12-06 11:03:05,669 - Epoch: [91][ 220/ 1200] Overall Loss 0.275870 Objective Loss 0.275870 LR 0.001000 Time 0.023089 -2022-12-06 11:03:05,868 - Epoch: [91][ 230/ 1200] Overall Loss 0.274813 Objective Loss 0.274813 LR 0.001000 Time 0.022948 -2022-12-06 11:03:06,065 - Epoch: [91][ 240/ 1200] Overall Loss 0.275430 Objective Loss 0.275430 LR 0.001000 Time 0.022809 -2022-12-06 11:03:06,264 - Epoch: [91][ 250/ 1200] Overall Loss 0.275848 Objective Loss 0.275848 LR 0.001000 Time 0.022690 -2022-12-06 11:03:06,460 - Epoch: [91][ 260/ 1200] Overall Loss 0.276725 Objective Loss 0.276725 LR 0.001000 Time 0.022569 -2022-12-06 11:03:06,659 - Epoch: [91][ 270/ 1200] Overall Loss 0.277086 Objective Loss 0.277086 LR 0.001000 Time 0.022469 -2022-12-06 11:03:06,856 - Epoch: [91][ 280/ 1200] Overall Loss 0.277543 Objective Loss 0.277543 LR 0.001000 Time 0.022366 -2022-12-06 11:03:07,054 - Epoch: [91][ 290/ 1200] Overall Loss 0.276076 Objective Loss 0.276076 LR 0.001000 Time 0.022276 -2022-12-06 11:03:07,251 - Epoch: [91][ 300/ 1200] Overall Loss 0.276048 Objective Loss 0.276048 LR 0.001000 Time 0.022187 -2022-12-06 11:03:07,449 - Epoch: [91][ 310/ 1200] Overall Loss 0.276207 Objective Loss 0.276207 LR 0.001000 Time 0.022110 -2022-12-06 11:03:07,645 - Epoch: [91][ 320/ 1200] Overall Loss 0.276343 Objective Loss 0.276343 LR 0.001000 Time 0.022031 -2022-12-06 11:03:07,844 - Epoch: [91][ 330/ 1200] Overall Loss 0.275835 Objective Loss 0.275835 LR 0.001000 Time 0.021964 -2022-12-06 11:03:08,041 - Epoch: [91][ 340/ 1200] Overall Loss 0.276792 Objective Loss 0.276792 LR 0.001000 Time 0.021894 -2022-12-06 11:03:08,240 - Epoch: [91][ 350/ 1200] Overall Loss 0.276403 Objective Loss 0.276403 LR 0.001000 Time 0.021836 -2022-12-06 11:03:08,436 - Epoch: [91][ 360/ 1200] Overall Loss 0.276924 Objective Loss 0.276924 LR 0.001000 Time 0.021773 -2022-12-06 11:03:08,635 - Epoch: [91][ 370/ 1200] Overall Loss 0.277417 Objective Loss 0.277417 LR 0.001000 Time 0.021720 -2022-12-06 11:03:08,831 - Epoch: [91][ 380/ 1200] Overall Loss 0.277541 Objective Loss 0.277541 LR 0.001000 Time 0.021664 -2022-12-06 11:03:09,030 - Epoch: [91][ 390/ 1200] Overall Loss 0.277453 Objective Loss 0.277453 LR 0.001000 Time 0.021617 -2022-12-06 11:03:09,226 - Epoch: [91][ 400/ 1200] Overall Loss 0.277316 Objective Loss 0.277316 LR 0.001000 Time 0.021565 -2022-12-06 11:03:09,424 - Epoch: [91][ 410/ 1200] Overall Loss 0.277765 Objective Loss 0.277765 LR 0.001000 Time 0.021522 -2022-12-06 11:03:09,621 - Epoch: [91][ 420/ 1200] Overall Loss 0.278086 Objective Loss 0.278086 LR 0.001000 Time 0.021476 -2022-12-06 11:03:09,820 - Epoch: [91][ 430/ 1200] Overall Loss 0.277810 Objective Loss 0.277810 LR 0.001000 Time 0.021438 -2022-12-06 11:03:10,016 - Epoch: [91][ 440/ 1200] Overall Loss 0.278068 Objective Loss 0.278068 LR 0.001000 Time 0.021396 -2022-12-06 11:03:10,215 - Epoch: [91][ 450/ 1200] Overall Loss 0.277987 Objective Loss 0.277987 LR 0.001000 Time 0.021360 -2022-12-06 11:03:10,411 - Epoch: [91][ 460/ 1200] Overall Loss 0.278475 Objective Loss 0.278475 LR 0.001000 Time 0.021322 -2022-12-06 11:03:10,610 - Epoch: [91][ 470/ 1200] Overall Loss 0.278569 Objective Loss 0.278569 LR 0.001000 Time 0.021289 -2022-12-06 11:03:10,806 - Epoch: [91][ 480/ 1200] Overall Loss 0.278225 Objective Loss 0.278225 LR 0.001000 Time 0.021255 -2022-12-06 11:03:11,005 - Epoch: [91][ 490/ 1200] Overall Loss 0.278221 Objective Loss 0.278221 LR 0.001000 Time 0.021225 -2022-12-06 11:03:11,202 - Epoch: [91][ 500/ 1200] Overall Loss 0.278286 Objective Loss 0.278286 LR 0.001000 Time 0.021192 -2022-12-06 11:03:11,400 - Epoch: [91][ 510/ 1200] Overall Loss 0.278265 Objective Loss 0.278265 LR 0.001000 Time 0.021165 -2022-12-06 11:03:11,597 - Epoch: [91][ 520/ 1200] Overall Loss 0.277948 Objective Loss 0.277948 LR 0.001000 Time 0.021135 -2022-12-06 11:03:11,796 - Epoch: [91][ 530/ 1200] Overall Loss 0.278052 Objective Loss 0.278052 LR 0.001000 Time 0.021111 -2022-12-06 11:03:11,992 - Epoch: [91][ 540/ 1200] Overall Loss 0.277953 Objective Loss 0.277953 LR 0.001000 Time 0.021083 -2022-12-06 11:03:12,191 - Epoch: [91][ 550/ 1200] Overall Loss 0.278250 Objective Loss 0.278250 LR 0.001000 Time 0.021061 -2022-12-06 11:03:12,388 - Epoch: [91][ 560/ 1200] Overall Loss 0.278885 Objective Loss 0.278885 LR 0.001000 Time 0.021034 -2022-12-06 11:03:12,587 - Epoch: [91][ 570/ 1200] Overall Loss 0.279176 Objective Loss 0.279176 LR 0.001000 Time 0.021014 -2022-12-06 11:03:12,783 - Epoch: [91][ 580/ 1200] Overall Loss 0.279979 Objective Loss 0.279979 LR 0.001000 Time 0.020989 -2022-12-06 11:03:12,982 - Epoch: [91][ 590/ 1200] Overall Loss 0.279923 Objective Loss 0.279923 LR 0.001000 Time 0.020969 -2022-12-06 11:03:13,179 - Epoch: [91][ 600/ 1200] Overall Loss 0.280405 Objective Loss 0.280405 LR 0.001000 Time 0.020947 -2022-12-06 11:03:13,378 - Epoch: [91][ 610/ 1200] Overall Loss 0.280533 Objective Loss 0.280533 LR 0.001000 Time 0.020928 -2022-12-06 11:03:13,575 - Epoch: [91][ 620/ 1200] Overall Loss 0.280461 Objective Loss 0.280461 LR 0.001000 Time 0.020908 -2022-12-06 11:03:13,774 - Epoch: [91][ 630/ 1200] Overall Loss 0.280731 Objective Loss 0.280731 LR 0.001000 Time 0.020891 -2022-12-06 11:03:13,970 - Epoch: [91][ 640/ 1200] Overall Loss 0.280329 Objective Loss 0.280329 LR 0.001000 Time 0.020871 -2022-12-06 11:03:14,169 - Epoch: [91][ 650/ 1200] Overall Loss 0.280560 Objective Loss 0.280560 LR 0.001000 Time 0.020855 -2022-12-06 11:03:14,366 - Epoch: [91][ 660/ 1200] Overall Loss 0.280579 Objective Loss 0.280579 LR 0.001000 Time 0.020836 -2022-12-06 11:03:14,565 - Epoch: [91][ 670/ 1200] Overall Loss 0.281160 Objective Loss 0.281160 LR 0.001000 Time 0.020821 -2022-12-06 11:03:14,762 - Epoch: [91][ 680/ 1200] Overall Loss 0.281076 Objective Loss 0.281076 LR 0.001000 Time 0.020803 -2022-12-06 11:03:14,961 - Epoch: [91][ 690/ 1200] Overall Loss 0.280706 Objective Loss 0.280706 LR 0.001000 Time 0.020790 -2022-12-06 11:03:15,158 - Epoch: [91][ 700/ 1200] Overall Loss 0.280462 Objective Loss 0.280462 LR 0.001000 Time 0.020774 -2022-12-06 11:03:15,358 - Epoch: [91][ 710/ 1200] Overall Loss 0.280777 Objective Loss 0.280777 LR 0.001000 Time 0.020761 -2022-12-06 11:03:15,555 - Epoch: [91][ 720/ 1200] Overall Loss 0.280706 Objective Loss 0.280706 LR 0.001000 Time 0.020746 -2022-12-06 11:03:15,753 - Epoch: [91][ 730/ 1200] Overall Loss 0.280785 Objective Loss 0.280785 LR 0.001000 Time 0.020733 -2022-12-06 11:03:15,950 - Epoch: [91][ 740/ 1200] Overall Loss 0.280948 Objective Loss 0.280948 LR 0.001000 Time 0.020718 -2022-12-06 11:03:16,149 - Epoch: [91][ 750/ 1200] Overall Loss 0.280895 Objective Loss 0.280895 LR 0.001000 Time 0.020707 -2022-12-06 11:03:16,345 - Epoch: [91][ 760/ 1200] Overall Loss 0.280936 Objective Loss 0.280936 LR 0.001000 Time 0.020691 -2022-12-06 11:03:16,544 - Epoch: [91][ 770/ 1200] Overall Loss 0.280926 Objective Loss 0.280926 LR 0.001000 Time 0.020680 -2022-12-06 11:03:16,741 - Epoch: [91][ 780/ 1200] Overall Loss 0.280968 Objective Loss 0.280968 LR 0.001000 Time 0.020667 -2022-12-06 11:03:16,940 - Epoch: [91][ 790/ 1200] Overall Loss 0.281315 Objective Loss 0.281315 LR 0.001000 Time 0.020657 -2022-12-06 11:03:17,137 - Epoch: [91][ 800/ 1200] Overall Loss 0.281417 Objective Loss 0.281417 LR 0.001000 Time 0.020644 -2022-12-06 11:03:17,337 - Epoch: [91][ 810/ 1200] Overall Loss 0.281472 Objective Loss 0.281472 LR 0.001000 Time 0.020635 -2022-12-06 11:03:17,534 - Epoch: [91][ 820/ 1200] Overall Loss 0.281716 Objective Loss 0.281716 LR 0.001000 Time 0.020623 -2022-12-06 11:03:17,733 - Epoch: [91][ 830/ 1200] Overall Loss 0.281693 Objective Loss 0.281693 LR 0.001000 Time 0.020613 -2022-12-06 11:03:17,929 - Epoch: [91][ 840/ 1200] Overall Loss 0.282213 Objective Loss 0.282213 LR 0.001000 Time 0.020601 -2022-12-06 11:03:18,128 - Epoch: [91][ 850/ 1200] Overall Loss 0.282312 Objective Loss 0.282312 LR 0.001000 Time 0.020593 -2022-12-06 11:03:18,325 - Epoch: [91][ 860/ 1200] Overall Loss 0.282353 Objective Loss 0.282353 LR 0.001000 Time 0.020581 -2022-12-06 11:03:18,524 - Epoch: [91][ 870/ 1200] Overall Loss 0.282445 Objective Loss 0.282445 LR 0.001000 Time 0.020573 -2022-12-06 11:03:18,721 - Epoch: [91][ 880/ 1200] Overall Loss 0.282595 Objective Loss 0.282595 LR 0.001000 Time 0.020562 -2022-12-06 11:03:18,920 - Epoch: [91][ 890/ 1200] Overall Loss 0.282862 Objective Loss 0.282862 LR 0.001000 Time 0.020554 -2022-12-06 11:03:19,116 - Epoch: [91][ 900/ 1200] Overall Loss 0.282716 Objective Loss 0.282716 LR 0.001000 Time 0.020543 -2022-12-06 11:03:19,315 - Epoch: [91][ 910/ 1200] Overall Loss 0.282652 Objective Loss 0.282652 LR 0.001000 Time 0.020535 -2022-12-06 11:03:19,511 - Epoch: [91][ 920/ 1200] Overall Loss 0.282767 Objective Loss 0.282767 LR 0.001000 Time 0.020524 -2022-12-06 11:03:19,710 - Epoch: [91][ 930/ 1200] Overall Loss 0.282938 Objective Loss 0.282938 LR 0.001000 Time 0.020517 -2022-12-06 11:03:19,908 - Epoch: [91][ 940/ 1200] Overall Loss 0.283390 Objective Loss 0.283390 LR 0.001000 Time 0.020508 -2022-12-06 11:03:20,107 - Epoch: [91][ 950/ 1200] Overall Loss 0.283480 Objective Loss 0.283480 LR 0.001000 Time 0.020501 -2022-12-06 11:03:20,304 - Epoch: [91][ 960/ 1200] Overall Loss 0.283478 Objective Loss 0.283478 LR 0.001000 Time 0.020492 -2022-12-06 11:03:20,503 - Epoch: [91][ 970/ 1200] Overall Loss 0.283470 Objective Loss 0.283470 LR 0.001000 Time 0.020486 -2022-12-06 11:03:20,700 - Epoch: [91][ 980/ 1200] Overall Loss 0.283586 Objective Loss 0.283586 LR 0.001000 Time 0.020477 -2022-12-06 11:03:20,898 - Epoch: [91][ 990/ 1200] Overall Loss 0.283330 Objective Loss 0.283330 LR 0.001000 Time 0.020470 -2022-12-06 11:03:21,095 - Epoch: [91][ 1000/ 1200] Overall Loss 0.283143 Objective Loss 0.283143 LR 0.001000 Time 0.020462 -2022-12-06 11:03:21,294 - Epoch: [91][ 1010/ 1200] Overall Loss 0.282952 Objective Loss 0.282952 LR 0.001000 Time 0.020455 -2022-12-06 11:03:21,491 - Epoch: [91][ 1020/ 1200] Overall Loss 0.283262 Objective Loss 0.283262 LR 0.001000 Time 0.020447 -2022-12-06 11:03:21,690 - Epoch: [91][ 1030/ 1200] Overall Loss 0.283225 Objective Loss 0.283225 LR 0.001000 Time 0.020442 -2022-12-06 11:03:21,887 - Epoch: [91][ 1040/ 1200] Overall Loss 0.283236 Objective Loss 0.283236 LR 0.001000 Time 0.020434 -2022-12-06 11:03:22,086 - Epoch: [91][ 1050/ 1200] Overall Loss 0.283408 Objective Loss 0.283408 LR 0.001000 Time 0.020429 -2022-12-06 11:03:22,282 - Epoch: [91][ 1060/ 1200] Overall Loss 0.283433 Objective Loss 0.283433 LR 0.001000 Time 0.020421 -2022-12-06 11:03:22,481 - Epoch: [91][ 1070/ 1200] Overall Loss 0.283364 Objective Loss 0.283364 LR 0.001000 Time 0.020415 -2022-12-06 11:03:22,677 - Epoch: [91][ 1080/ 1200] Overall Loss 0.283418 Objective Loss 0.283418 LR 0.001000 Time 0.020407 -2022-12-06 11:03:22,876 - Epoch: [91][ 1090/ 1200] Overall Loss 0.283490 Objective Loss 0.283490 LR 0.001000 Time 0.020402 -2022-12-06 11:03:23,073 - Epoch: [91][ 1100/ 1200] Overall Loss 0.283395 Objective Loss 0.283395 LR 0.001000 Time 0.020395 -2022-12-06 11:03:23,272 - Epoch: [91][ 1110/ 1200] Overall Loss 0.283444 Objective Loss 0.283444 LR 0.001000 Time 0.020390 -2022-12-06 11:03:23,469 - Epoch: [91][ 1120/ 1200] Overall Loss 0.283324 Objective Loss 0.283324 LR 0.001000 Time 0.020384 -2022-12-06 11:03:23,668 - Epoch: [91][ 1130/ 1200] Overall Loss 0.283270 Objective Loss 0.283270 LR 0.001000 Time 0.020379 -2022-12-06 11:03:23,865 - Epoch: [91][ 1140/ 1200] Overall Loss 0.283032 Objective Loss 0.283032 LR 0.001000 Time 0.020372 -2022-12-06 11:03:24,064 - Epoch: [91][ 1150/ 1200] Overall Loss 0.282958 Objective Loss 0.282958 LR 0.001000 Time 0.020367 -2022-12-06 11:03:24,260 - Epoch: [91][ 1160/ 1200] Overall Loss 0.283261 Objective Loss 0.283261 LR 0.001000 Time 0.020360 -2022-12-06 11:03:24,462 - Epoch: [91][ 1170/ 1200] Overall Loss 0.283448 Objective Loss 0.283448 LR 0.001000 Time 0.020359 -2022-12-06 11:03:24,662 - Epoch: [91][ 1180/ 1200] Overall Loss 0.283444 Objective Loss 0.283444 LR 0.001000 Time 0.020355 -2022-12-06 11:03:24,866 - Epoch: [91][ 1190/ 1200] Overall Loss 0.283174 Objective Loss 0.283174 LR 0.001000 Time 0.020355 -2022-12-06 11:03:25,095 - Epoch: [91][ 1200/ 1200] Overall Loss 0.283184 Objective Loss 0.283184 Top1 84.518828 Top5 97.907950 LR 0.001000 Time 0.020376 -2022-12-06 11:03:25,183 - --- validate (epoch=91)----------- -2022-12-06 11:03:25,184 - 34129 samples (256 per mini-batch) -2022-12-06 11:03:25,630 - Epoch: [91][ 10/ 134] Loss 0.304283 Top1 84.414062 Top5 98.007812 -2022-12-06 11:03:25,760 - Epoch: [91][ 20/ 134] Loss 0.292205 Top1 84.472656 Top5 97.792969 -2022-12-06 11:03:25,893 - Epoch: [91][ 30/ 134] Loss 0.301854 Top1 84.348958 Top5 97.851562 -2022-12-06 11:03:26,025 - Epoch: [91][ 40/ 134] Loss 0.300751 Top1 84.462891 Top5 97.763672 -2022-12-06 11:03:26,158 - Epoch: [91][ 50/ 134] Loss 0.294455 Top1 84.476562 Top5 97.804688 -2022-12-06 11:03:26,290 - Epoch: [91][ 60/ 134] Loss 0.292411 Top1 84.407552 Top5 97.832031 -2022-12-06 11:03:26,423 - Epoch: [91][ 70/ 134] Loss 0.291752 Top1 84.224330 Top5 97.734375 -2022-12-06 11:03:26,555 - Epoch: [91][ 80/ 134] Loss 0.295221 Top1 84.125977 Top5 97.666016 -2022-12-06 11:03:26,687 - Epoch: [91][ 90/ 134] Loss 0.295890 Top1 84.140625 Top5 97.677951 -2022-12-06 11:03:26,820 - Epoch: [91][ 100/ 134] Loss 0.296814 Top1 84.125000 Top5 97.714844 -2022-12-06 11:03:26,953 - Epoch: [91][ 110/ 134] Loss 0.296405 Top1 84.030540 Top5 97.695312 -2022-12-06 11:03:27,086 - Epoch: [91][ 120/ 134] Loss 0.296498 Top1 83.974609 Top5 97.711589 -2022-12-06 11:03:27,219 - Epoch: [91][ 130/ 134] Loss 0.298595 Top1 83.903245 Top5 97.647236 -2022-12-06 11:03:27,258 - Epoch: [91][ 134/ 134] Loss 0.301175 Top1 83.861232 Top5 97.644232 -2022-12-06 11:03:27,346 - ==> Top1: 83.861 Top5: 97.644 Loss: 0.301 - -2022-12-06 11:03:27,347 - ==> Confusion: -[[ 906 0 0 2 3 8 0 1 6 54 0 2 1 5 2 1 1 1 1 0 2] - [ 4 912 1 4 4 39 1 20 3 0 2 4 1 2 2 3 2 1 10 4 8] - [ 7 4 999 20 3 3 18 12 2 2 3 10 0 1 1 2 0 3 3 3 7] - [ 4 2 16 943 1 2 0 1 2 0 8 3 5 0 12 0 0 4 14 0 3] - [ 13 3 1 2 941 9 1 4 0 7 1 4 0 2 10 7 8 2 2 1 2] - [ 6 11 0 3 6 973 4 21 1 1 0 13 1 11 1 1 5 2 1 6 2] - [ 0 5 13 2 0 2 1059 6 0 0 2 4 2 0 0 6 0 4 4 8 1] - [ 3 8 6 3 1 33 7 939 1 0 4 9 0 1 0 0 1 2 24 8 4] - [ 8 5 1 1 0 6 0 1 950 50 11 2 1 12 8 1 1 2 2 1 1] - [ 64 0 1 1 4 3 0 4 23 871 1 3 0 16 2 1 0 1 1 0 5] - [ 2 0 6 10 1 3 3 5 8 0 943 3 0 16 1 1 1 0 10 0 6] - [ 2 0 2 2 1 16 5 5 0 0 0 974 17 2 0 7 4 5 0 7 2] - [ 4 0 1 3 0 3 3 2 0 0 0 53 862 3 0 8 2 15 1 5 4] - [ 0 1 1 0 0 5 0 2 12 11 7 10 3 958 0 3 4 0 0 0 6] - [ 12 1 2 16 1 2 0 1 29 4 1 1 4 2 1031 1 0 2 14 1 5] - [ 1 1 2 1 5 1 1 1 0 1 0 10 5 3 0 982 11 13 0 2 3] - [ 3 3 1 0 2 1 4 2 0 0 0 4 1 1 0 13 1031 2 0 1 3] - [ 5 0 2 4 0 1 1 3 2 1 0 18 20 1 1 13 3 958 0 2 1] - [ 8 6 3 12 0 3 0 21 1 1 3 5 2 0 5 0 0 1 934 1 2] - [ 5 4 1 0 0 6 5 6 0 0 0 23 4 6 0 4 6 4 1 1002 3] - [ 178 226 195 140 85 255 86 195 107 97 155 163 389 339 144 154 298 82 224 271 9443]] - -2022-12-06 11:03:27,913 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:03:27,913 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:03:27,919 - - -2022-12-06 11:03:27,919 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:03:28,942 - Epoch: [92][ 10/ 1200] Overall Loss 0.261036 Objective Loss 0.261036 LR 0.001000 Time 0.102205 -2022-12-06 11:03:29,135 - Epoch: [92][ 20/ 1200] Overall Loss 0.265805 Objective Loss 0.265805 LR 0.001000 Time 0.060712 -2022-12-06 11:03:29,326 - Epoch: [92][ 30/ 1200] Overall Loss 0.269885 Objective Loss 0.269885 LR 0.001000 Time 0.046821 -2022-12-06 11:03:29,517 - Epoch: [92][ 40/ 1200] Overall Loss 0.266256 Objective Loss 0.266256 LR 0.001000 Time 0.039876 -2022-12-06 11:03:29,707 - Epoch: [92][ 50/ 1200] Overall Loss 0.266296 Objective Loss 0.266296 LR 0.001000 Time 0.035698 -2022-12-06 11:03:29,897 - Epoch: [92][ 60/ 1200] Overall Loss 0.262289 Objective Loss 0.262289 LR 0.001000 Time 0.032903 -2022-12-06 11:03:30,087 - Epoch: [92][ 70/ 1200] Overall Loss 0.261772 Objective Loss 0.261772 LR 0.001000 Time 0.030913 -2022-12-06 11:03:30,277 - Epoch: [92][ 80/ 1200] Overall Loss 0.264530 Objective Loss 0.264530 LR 0.001000 Time 0.029421 -2022-12-06 11:03:30,469 - Epoch: [92][ 90/ 1200] Overall Loss 0.265632 Objective Loss 0.265632 LR 0.001000 Time 0.028269 -2022-12-06 11:03:30,659 - Epoch: [92][ 100/ 1200] Overall Loss 0.268317 Objective Loss 0.268317 LR 0.001000 Time 0.027340 -2022-12-06 11:03:30,849 - Epoch: [92][ 110/ 1200] Overall Loss 0.268551 Objective Loss 0.268551 LR 0.001000 Time 0.026578 -2022-12-06 11:03:31,039 - Epoch: [92][ 120/ 1200] Overall Loss 0.271056 Objective Loss 0.271056 LR 0.001000 Time 0.025940 -2022-12-06 11:03:31,229 - Epoch: [92][ 130/ 1200] Overall Loss 0.272470 Objective Loss 0.272470 LR 0.001000 Time 0.025407 -2022-12-06 11:03:31,419 - Epoch: [92][ 140/ 1200] Overall Loss 0.272645 Objective Loss 0.272645 LR 0.001000 Time 0.024944 -2022-12-06 11:03:31,610 - Epoch: [92][ 150/ 1200] Overall Loss 0.272120 Objective Loss 0.272120 LR 0.001000 Time 0.024546 -2022-12-06 11:03:31,801 - Epoch: [92][ 160/ 1200] Overall Loss 0.272483 Objective Loss 0.272483 LR 0.001000 Time 0.024203 -2022-12-06 11:03:31,991 - Epoch: [92][ 170/ 1200] Overall Loss 0.272983 Objective Loss 0.272983 LR 0.001000 Time 0.023895 -2022-12-06 11:03:32,181 - Epoch: [92][ 180/ 1200] Overall Loss 0.273222 Objective Loss 0.273222 LR 0.001000 Time 0.023620 -2022-12-06 11:03:32,372 - Epoch: [92][ 190/ 1200] Overall Loss 0.274241 Objective Loss 0.274241 LR 0.001000 Time 0.023379 -2022-12-06 11:03:32,562 - Epoch: [92][ 200/ 1200] Overall Loss 0.275266 Objective Loss 0.275266 LR 0.001000 Time 0.023159 -2022-12-06 11:03:32,753 - Epoch: [92][ 210/ 1200] Overall Loss 0.275925 Objective Loss 0.275925 LR 0.001000 Time 0.022963 -2022-12-06 11:03:32,943 - Epoch: [92][ 220/ 1200] Overall Loss 0.275537 Objective Loss 0.275537 LR 0.001000 Time 0.022782 -2022-12-06 11:03:33,134 - Epoch: [92][ 230/ 1200] Overall Loss 0.276308 Objective Loss 0.276308 LR 0.001000 Time 0.022616 -2022-12-06 11:03:33,324 - Epoch: [92][ 240/ 1200] Overall Loss 0.275938 Objective Loss 0.275938 LR 0.001000 Time 0.022463 -2022-12-06 11:03:33,514 - Epoch: [92][ 250/ 1200] Overall Loss 0.275816 Objective Loss 0.275816 LR 0.001000 Time 0.022322 -2022-12-06 11:03:33,704 - Epoch: [92][ 260/ 1200] Overall Loss 0.275952 Objective Loss 0.275952 LR 0.001000 Time 0.022195 -2022-12-06 11:03:33,895 - Epoch: [92][ 270/ 1200] Overall Loss 0.276999 Objective Loss 0.276999 LR 0.001000 Time 0.022075 -2022-12-06 11:03:34,085 - Epoch: [92][ 280/ 1200] Overall Loss 0.275963 Objective Loss 0.275963 LR 0.001000 Time 0.021963 -2022-12-06 11:03:34,276 - Epoch: [92][ 290/ 1200] Overall Loss 0.276627 Objective Loss 0.276627 LR 0.001000 Time 0.021864 -2022-12-06 11:03:34,467 - Epoch: [92][ 300/ 1200] Overall Loss 0.276238 Objective Loss 0.276238 LR 0.001000 Time 0.021770 -2022-12-06 11:03:34,657 - Epoch: [92][ 310/ 1200] Overall Loss 0.276814 Objective Loss 0.276814 LR 0.001000 Time 0.021679 -2022-12-06 11:03:34,847 - Epoch: [92][ 320/ 1200] Overall Loss 0.275500 Objective Loss 0.275500 LR 0.001000 Time 0.021594 -2022-12-06 11:03:35,038 - Epoch: [92][ 330/ 1200] Overall Loss 0.275825 Objective Loss 0.275825 LR 0.001000 Time 0.021517 -2022-12-06 11:03:35,229 - Epoch: [92][ 340/ 1200] Overall Loss 0.276117 Objective Loss 0.276117 LR 0.001000 Time 0.021443 -2022-12-06 11:03:35,419 - Epoch: [92][ 350/ 1200] Overall Loss 0.275912 Objective Loss 0.275912 LR 0.001000 Time 0.021372 -2022-12-06 11:03:35,610 - Epoch: [92][ 360/ 1200] Overall Loss 0.276953 Objective Loss 0.276953 LR 0.001000 Time 0.021306 -2022-12-06 11:03:35,800 - Epoch: [92][ 370/ 1200] Overall Loss 0.276803 Objective Loss 0.276803 LR 0.001000 Time 0.021243 -2022-12-06 11:03:35,991 - Epoch: [92][ 380/ 1200] Overall Loss 0.276745 Objective Loss 0.276745 LR 0.001000 Time 0.021185 -2022-12-06 11:03:36,181 - Epoch: [92][ 390/ 1200] Overall Loss 0.276263 Objective Loss 0.276263 LR 0.001000 Time 0.021128 -2022-12-06 11:03:36,371 - Epoch: [92][ 400/ 1200] Overall Loss 0.276672 Objective Loss 0.276672 LR 0.001000 Time 0.021074 -2022-12-06 11:03:36,563 - Epoch: [92][ 410/ 1200] Overall Loss 0.276959 Objective Loss 0.276959 LR 0.001000 Time 0.021027 -2022-12-06 11:03:36,756 - Epoch: [92][ 420/ 1200] Overall Loss 0.277021 Objective Loss 0.277021 LR 0.001000 Time 0.020984 -2022-12-06 11:03:36,949 - Epoch: [92][ 430/ 1200] Overall Loss 0.276905 Objective Loss 0.276905 LR 0.001000 Time 0.020943 -2022-12-06 11:03:37,141 - Epoch: [92][ 440/ 1200] Overall Loss 0.277522 Objective Loss 0.277522 LR 0.001000 Time 0.020904 -2022-12-06 11:03:37,334 - Epoch: [92][ 450/ 1200] Overall Loss 0.277942 Objective Loss 0.277942 LR 0.001000 Time 0.020866 -2022-12-06 11:03:37,526 - Epoch: [92][ 460/ 1200] Overall Loss 0.278515 Objective Loss 0.278515 LR 0.001000 Time 0.020829 -2022-12-06 11:03:37,719 - Epoch: [92][ 470/ 1200] Overall Loss 0.278613 Objective Loss 0.278613 LR 0.001000 Time 0.020796 -2022-12-06 11:03:37,912 - Epoch: [92][ 480/ 1200] Overall Loss 0.279030 Objective Loss 0.279030 LR 0.001000 Time 0.020763 -2022-12-06 11:03:38,106 - Epoch: [92][ 490/ 1200] Overall Loss 0.279459 Objective Loss 0.279459 LR 0.001000 Time 0.020733 -2022-12-06 11:03:38,298 - Epoch: [92][ 500/ 1200] Overall Loss 0.279570 Objective Loss 0.279570 LR 0.001000 Time 0.020702 -2022-12-06 11:03:38,490 - Epoch: [92][ 510/ 1200] Overall Loss 0.279150 Objective Loss 0.279150 LR 0.001000 Time 0.020672 -2022-12-06 11:03:38,683 - Epoch: [92][ 520/ 1200] Overall Loss 0.278998 Objective Loss 0.278998 LR 0.001000 Time 0.020644 -2022-12-06 11:03:38,876 - Epoch: [92][ 530/ 1200] Overall Loss 0.279114 Objective Loss 0.279114 LR 0.001000 Time 0.020618 -2022-12-06 11:03:39,069 - Epoch: [92][ 540/ 1200] Overall Loss 0.278944 Objective Loss 0.278944 LR 0.001000 Time 0.020592 -2022-12-06 11:03:39,263 - Epoch: [92][ 550/ 1200] Overall Loss 0.278443 Objective Loss 0.278443 LR 0.001000 Time 0.020569 -2022-12-06 11:03:39,455 - Epoch: [92][ 560/ 1200] Overall Loss 0.278817 Objective Loss 0.278817 LR 0.001000 Time 0.020545 -2022-12-06 11:03:39,649 - Epoch: [92][ 570/ 1200] Overall Loss 0.278785 Objective Loss 0.278785 LR 0.001000 Time 0.020522 -2022-12-06 11:03:39,841 - Epoch: [92][ 580/ 1200] Overall Loss 0.278669 Objective Loss 0.278669 LR 0.001000 Time 0.020500 -2022-12-06 11:03:40,035 - Epoch: [92][ 590/ 1200] Overall Loss 0.278298 Objective Loss 0.278298 LR 0.001000 Time 0.020479 -2022-12-06 11:03:40,227 - Epoch: [92][ 600/ 1200] Overall Loss 0.278001 Objective Loss 0.278001 LR 0.001000 Time 0.020458 -2022-12-06 11:03:40,420 - Epoch: [92][ 610/ 1200] Overall Loss 0.278301 Objective Loss 0.278301 LR 0.001000 Time 0.020438 -2022-12-06 11:03:40,613 - Epoch: [92][ 620/ 1200] Overall Loss 0.278573 Objective Loss 0.278573 LR 0.001000 Time 0.020418 -2022-12-06 11:03:40,806 - Epoch: [92][ 630/ 1200] Overall Loss 0.278579 Objective Loss 0.278579 LR 0.001000 Time 0.020400 -2022-12-06 11:03:40,998 - Epoch: [92][ 640/ 1200] Overall Loss 0.279188 Objective Loss 0.279188 LR 0.001000 Time 0.020380 -2022-12-06 11:03:41,192 - Epoch: [92][ 650/ 1200] Overall Loss 0.279078 Objective Loss 0.279078 LR 0.001000 Time 0.020363 -2022-12-06 11:03:41,384 - Epoch: [92][ 660/ 1200] Overall Loss 0.279161 Objective Loss 0.279161 LR 0.001000 Time 0.020345 -2022-12-06 11:03:41,577 - Epoch: [92][ 670/ 1200] Overall Loss 0.279872 Objective Loss 0.279872 LR 0.001000 Time 0.020329 -2022-12-06 11:03:41,770 - Epoch: [92][ 680/ 1200] Overall Loss 0.279937 Objective Loss 0.279937 LR 0.001000 Time 0.020313 -2022-12-06 11:03:41,963 - Epoch: [92][ 690/ 1200] Overall Loss 0.279958 Objective Loss 0.279958 LR 0.001000 Time 0.020297 -2022-12-06 11:03:42,156 - Epoch: [92][ 700/ 1200] Overall Loss 0.280258 Objective Loss 0.280258 LR 0.001000 Time 0.020282 -2022-12-06 11:03:42,349 - Epoch: [92][ 710/ 1200] Overall Loss 0.280437 Objective Loss 0.280437 LR 0.001000 Time 0.020268 -2022-12-06 11:03:42,540 - Epoch: [92][ 720/ 1200] Overall Loss 0.280518 Objective Loss 0.280518 LR 0.001000 Time 0.020252 -2022-12-06 11:03:42,734 - Epoch: [92][ 730/ 1200] Overall Loss 0.280284 Objective Loss 0.280284 LR 0.001000 Time 0.020239 -2022-12-06 11:03:42,927 - Epoch: [92][ 740/ 1200] Overall Loss 0.280383 Objective Loss 0.280383 LR 0.001000 Time 0.020225 -2022-12-06 11:03:43,120 - Epoch: [92][ 750/ 1200] Overall Loss 0.280196 Objective Loss 0.280196 LR 0.001000 Time 0.020212 -2022-12-06 11:03:43,311 - Epoch: [92][ 760/ 1200] Overall Loss 0.280341 Objective Loss 0.280341 LR 0.001000 Time 0.020197 -2022-12-06 11:03:43,504 - Epoch: [92][ 770/ 1200] Overall Loss 0.280768 Objective Loss 0.280768 LR 0.001000 Time 0.020184 -2022-12-06 11:03:43,696 - Epoch: [92][ 780/ 1200] Overall Loss 0.280804 Objective Loss 0.280804 LR 0.001000 Time 0.020171 -2022-12-06 11:03:43,888 - Epoch: [92][ 790/ 1200] Overall Loss 0.281083 Objective Loss 0.281083 LR 0.001000 Time 0.020159 -2022-12-06 11:03:44,081 - Epoch: [92][ 800/ 1200] Overall Loss 0.281034 Objective Loss 0.281034 LR 0.001000 Time 0.020146 -2022-12-06 11:03:44,273 - Epoch: [92][ 810/ 1200] Overall Loss 0.281225 Objective Loss 0.281225 LR 0.001000 Time 0.020135 -2022-12-06 11:03:44,466 - Epoch: [92][ 820/ 1200] Overall Loss 0.281264 Objective Loss 0.281264 LR 0.001000 Time 0.020123 -2022-12-06 11:03:44,659 - Epoch: [92][ 830/ 1200] Overall Loss 0.281104 Objective Loss 0.281104 LR 0.001000 Time 0.020113 -2022-12-06 11:03:44,851 - Epoch: [92][ 840/ 1200] Overall Loss 0.281045 Objective Loss 0.281045 LR 0.001000 Time 0.020101 -2022-12-06 11:03:45,043 - Epoch: [92][ 850/ 1200] Overall Loss 0.280975 Objective Loss 0.280975 LR 0.001000 Time 0.020090 -2022-12-06 11:03:45,236 - Epoch: [92][ 860/ 1200] Overall Loss 0.280787 Objective Loss 0.280787 LR 0.001000 Time 0.020080 -2022-12-06 11:03:45,428 - Epoch: [92][ 870/ 1200] Overall Loss 0.280828 Objective Loss 0.280828 LR 0.001000 Time 0.020070 -2022-12-06 11:03:45,621 - Epoch: [92][ 880/ 1200] Overall Loss 0.280615 Objective Loss 0.280615 LR 0.001000 Time 0.020061 -2022-12-06 11:03:45,814 - Epoch: [92][ 890/ 1200] Overall Loss 0.280327 Objective Loss 0.280327 LR 0.001000 Time 0.020051 -2022-12-06 11:03:46,006 - Epoch: [92][ 900/ 1200] Overall Loss 0.280608 Objective Loss 0.280608 LR 0.001000 Time 0.020041 -2022-12-06 11:03:46,198 - Epoch: [92][ 910/ 1200] Overall Loss 0.280716 Objective Loss 0.280716 LR 0.001000 Time 0.020032 -2022-12-06 11:03:46,391 - Epoch: [92][ 920/ 1200] Overall Loss 0.280766 Objective Loss 0.280766 LR 0.001000 Time 0.020023 -2022-12-06 11:03:46,584 - Epoch: [92][ 930/ 1200] Overall Loss 0.280578 Objective Loss 0.280578 LR 0.001000 Time 0.020015 -2022-12-06 11:03:46,778 - Epoch: [92][ 940/ 1200] Overall Loss 0.280413 Objective Loss 0.280413 LR 0.001000 Time 0.020007 -2022-12-06 11:03:46,971 - Epoch: [92][ 950/ 1200] Overall Loss 0.280074 Objective Loss 0.280074 LR 0.001000 Time 0.020000 -2022-12-06 11:03:47,164 - Epoch: [92][ 960/ 1200] Overall Loss 0.280031 Objective Loss 0.280031 LR 0.001000 Time 0.019991 -2022-12-06 11:03:47,358 - Epoch: [92][ 970/ 1200] Overall Loss 0.279603 Objective Loss 0.279603 LR 0.001000 Time 0.019985 -2022-12-06 11:03:47,551 - Epoch: [92][ 980/ 1200] Overall Loss 0.279472 Objective Loss 0.279472 LR 0.001000 Time 0.019977 -2022-12-06 11:03:47,745 - Epoch: [92][ 990/ 1200] Overall Loss 0.279266 Objective Loss 0.279266 LR 0.001000 Time 0.019970 -2022-12-06 11:03:47,938 - Epoch: [92][ 1000/ 1200] Overall Loss 0.279134 Objective Loss 0.279134 LR 0.001000 Time 0.019964 -2022-12-06 11:03:48,133 - Epoch: [92][ 1010/ 1200] Overall Loss 0.279119 Objective Loss 0.279119 LR 0.001000 Time 0.019958 -2022-12-06 11:03:48,326 - Epoch: [92][ 1020/ 1200] Overall Loss 0.278994 Objective Loss 0.278994 LR 0.001000 Time 0.019951 -2022-12-06 11:03:48,520 - Epoch: [92][ 1030/ 1200] Overall Loss 0.279110 Objective Loss 0.279110 LR 0.001000 Time 0.019945 -2022-12-06 11:03:48,713 - Epoch: [92][ 1040/ 1200] Overall Loss 0.278934 Objective Loss 0.278934 LR 0.001000 Time 0.019939 -2022-12-06 11:03:48,907 - Epoch: [92][ 1050/ 1200] Overall Loss 0.278871 Objective Loss 0.278871 LR 0.001000 Time 0.019933 -2022-12-06 11:03:49,100 - Epoch: [92][ 1060/ 1200] Overall Loss 0.278668 Objective Loss 0.278668 LR 0.001000 Time 0.019927 -2022-12-06 11:03:49,294 - Epoch: [92][ 1070/ 1200] Overall Loss 0.279052 Objective Loss 0.279052 LR 0.001000 Time 0.019921 -2022-12-06 11:03:49,487 - Epoch: [92][ 1080/ 1200] Overall Loss 0.278998 Objective Loss 0.278998 LR 0.001000 Time 0.019915 -2022-12-06 11:03:49,680 - Epoch: [92][ 1090/ 1200] Overall Loss 0.278994 Objective Loss 0.278994 LR 0.001000 Time 0.019909 -2022-12-06 11:03:49,874 - Epoch: [92][ 1100/ 1200] Overall Loss 0.279169 Objective Loss 0.279169 LR 0.001000 Time 0.019904 -2022-12-06 11:03:50,068 - Epoch: [92][ 1110/ 1200] Overall Loss 0.279329 Objective Loss 0.279329 LR 0.001000 Time 0.019899 -2022-12-06 11:03:50,262 - Epoch: [92][ 1120/ 1200] Overall Loss 0.279351 Objective Loss 0.279351 LR 0.001000 Time 0.019893 -2022-12-06 11:03:50,455 - Epoch: [92][ 1130/ 1200] Overall Loss 0.279220 Objective Loss 0.279220 LR 0.001000 Time 0.019888 -2022-12-06 11:03:50,648 - Epoch: [92][ 1140/ 1200] Overall Loss 0.279371 Objective Loss 0.279371 LR 0.001000 Time 0.019882 -2022-12-06 11:03:50,842 - Epoch: [92][ 1150/ 1200] Overall Loss 0.279354 Objective Loss 0.279354 LR 0.001000 Time 0.019878 -2022-12-06 11:03:51,036 - Epoch: [92][ 1160/ 1200] Overall Loss 0.279510 Objective Loss 0.279510 LR 0.001000 Time 0.019873 -2022-12-06 11:03:51,230 - Epoch: [92][ 1170/ 1200] Overall Loss 0.279823 Objective Loss 0.279823 LR 0.001000 Time 0.019868 -2022-12-06 11:03:51,423 - Epoch: [92][ 1180/ 1200] Overall Loss 0.280095 Objective Loss 0.280095 LR 0.001000 Time 0.019863 -2022-12-06 11:03:51,617 - Epoch: [92][ 1190/ 1200] Overall Loss 0.280192 Objective Loss 0.280192 LR 0.001000 Time 0.019858 -2022-12-06 11:03:51,843 - Epoch: [92][ 1200/ 1200] Overall Loss 0.280002 Objective Loss 0.280002 Top1 85.564854 Top5 98.117155 LR 0.001000 Time 0.019881 -2022-12-06 11:03:51,931 - --- validate (epoch=92)----------- -2022-12-06 11:03:51,931 - 34129 samples (256 per mini-batch) -2022-12-06 11:03:52,382 - Epoch: [92][ 10/ 134] Loss 0.276139 Top1 84.375000 Top5 97.812500 -2022-12-06 11:03:52,516 - Epoch: [92][ 20/ 134] Loss 0.288764 Top1 84.687500 Top5 97.675781 -2022-12-06 11:03:52,647 - Epoch: [92][ 30/ 134] Loss 0.284561 Top1 84.778646 Top5 97.773438 -2022-12-06 11:03:52,780 - Epoch: [92][ 40/ 134] Loss 0.289837 Top1 84.492188 Top5 97.822266 -2022-12-06 11:03:52,913 - Epoch: [92][ 50/ 134] Loss 0.292511 Top1 84.554688 Top5 97.796875 -2022-12-06 11:03:53,046 - Epoch: [92][ 60/ 134] Loss 0.294483 Top1 84.537760 Top5 97.812500 -2022-12-06 11:03:53,178 - Epoch: [92][ 70/ 134] Loss 0.292951 Top1 84.363839 Top5 97.795759 -2022-12-06 11:03:53,310 - Epoch: [92][ 80/ 134] Loss 0.292254 Top1 84.233398 Top5 97.832031 -2022-12-06 11:03:53,443 - Epoch: [92][ 90/ 134] Loss 0.296756 Top1 84.175347 Top5 97.816840 -2022-12-06 11:03:53,575 - Epoch: [92][ 100/ 134] Loss 0.295345 Top1 84.261719 Top5 97.824219 -2022-12-06 11:03:53,707 - Epoch: [92][ 110/ 134] Loss 0.298536 Top1 84.176136 Top5 97.808949 -2022-12-06 11:03:53,840 - Epoch: [92][ 120/ 134] Loss 0.299033 Top1 84.179688 Top5 97.835286 -2022-12-06 11:03:53,968 - Epoch: [92][ 130/ 134] Loss 0.298602 Top1 84.272837 Top5 97.857572 -2022-12-06 11:03:54,006 - Epoch: [92][ 134/ 134] Loss 0.301988 Top1 84.221630 Top5 97.843476 -2022-12-06 11:03:54,096 - ==> Top1: 84.222 Top5: 97.843 Loss: 0.302 - -2022-12-06 11:03:54,096 - ==> Confusion: -[[ 896 1 2 2 3 9 0 2 5 53 0 2 1 5 5 3 1 2 0 0 4] - [ 1 915 0 3 8 32 4 13 4 1 4 5 2 1 2 1 4 0 16 0 11] - [ 6 5 981 11 1 2 35 10 0 1 13 5 1 2 5 2 2 3 5 4 9] - [ 5 1 23 927 0 5 1 1 2 0 11 0 6 0 17 0 1 5 8 0 7] - [ 12 6 3 0 947 8 2 1 0 4 1 3 1 0 9 6 11 1 2 2 1] - [ 2 24 0 2 6 963 4 14 1 5 0 13 4 12 0 0 1 1 2 12 3] - [ 0 3 7 1 0 4 1073 4 1 0 2 3 0 2 0 3 0 2 2 8 3] - [ 2 10 10 2 2 42 13 905 0 2 4 9 2 1 1 1 3 0 33 9 3] - [ 5 4 0 0 1 6 1 1 939 54 10 1 1 9 17 2 2 2 5 1 3] - [ 52 1 1 0 5 3 1 4 20 893 1 1 0 8 1 1 1 1 0 0 7] - [ 2 2 4 6 2 2 0 6 10 1 948 0 2 10 3 0 0 0 10 1 10] - [ 3 2 1 1 0 16 5 4 0 0 0 954 26 2 0 8 4 8 1 12 4] - [ 0 0 0 2 0 3 2 1 0 0 0 39 874 1 3 12 3 17 0 6 6] - [ 0 1 0 0 0 9 0 4 11 17 10 6 2 940 4 2 3 0 1 1 12] - [ 8 3 1 12 4 2 0 1 21 7 0 3 2 2 1047 1 0 1 2 1 12] - [ 2 1 2 1 1 2 9 1 0 0 0 5 4 2 0 989 2 13 0 5 4] - [ 5 3 1 1 2 0 1 2 1 0 0 4 0 2 1 16 1016 3 0 7 7] - [ 1 1 0 5 0 1 2 2 0 4 0 5 19 3 1 14 1 973 0 2 2] - [ 1 2 10 14 0 6 0 23 6 1 5 3 5 0 8 1 2 2 911 2 6] - [ 2 4 4 0 0 3 5 10 0 0 2 16 4 6 0 2 4 3 1 1011 3] - [ 135 203 182 111 105 218 119 144 91 115 188 136 381 299 180 178 236 107 190 276 9632]] - -2022-12-06 11:03:54,677 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:03:54,677 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:03:54,683 - - -2022-12-06 11:03:54,683 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:03:55,602 - Epoch: [93][ 10/ 1200] Overall Loss 0.223248 Objective Loss 0.223248 LR 0.001000 Time 0.091806 -2022-12-06 11:03:55,797 - Epoch: [93][ 20/ 1200] Overall Loss 0.253907 Objective Loss 0.253907 LR 0.001000 Time 0.055638 -2022-12-06 11:03:55,990 - Epoch: [93][ 30/ 1200] Overall Loss 0.259187 Objective Loss 0.259187 LR 0.001000 Time 0.043502 -2022-12-06 11:03:56,183 - Epoch: [93][ 40/ 1200] Overall Loss 0.267427 Objective Loss 0.267427 LR 0.001000 Time 0.037433 -2022-12-06 11:03:56,374 - Epoch: [93][ 50/ 1200] Overall Loss 0.270382 Objective Loss 0.270382 LR 0.001000 Time 0.033760 -2022-12-06 11:03:56,566 - Epoch: [93][ 60/ 1200] Overall Loss 0.265148 Objective Loss 0.265148 LR 0.001000 Time 0.031325 -2022-12-06 11:03:56,758 - Epoch: [93][ 70/ 1200] Overall Loss 0.267704 Objective Loss 0.267704 LR 0.001000 Time 0.029576 -2022-12-06 11:03:56,949 - Epoch: [93][ 80/ 1200] Overall Loss 0.266105 Objective Loss 0.266105 LR 0.001000 Time 0.028269 -2022-12-06 11:03:57,141 - Epoch: [93][ 90/ 1200] Overall Loss 0.267145 Objective Loss 0.267145 LR 0.001000 Time 0.027250 -2022-12-06 11:03:57,333 - Epoch: [93][ 100/ 1200] Overall Loss 0.265789 Objective Loss 0.265789 LR 0.001000 Time 0.026441 -2022-12-06 11:03:57,525 - Epoch: [93][ 110/ 1200] Overall Loss 0.265428 Objective Loss 0.265428 LR 0.001000 Time 0.025779 -2022-12-06 11:03:57,717 - Epoch: [93][ 120/ 1200] Overall Loss 0.266565 Objective Loss 0.266565 LR 0.001000 Time 0.025229 -2022-12-06 11:03:57,909 - Epoch: [93][ 130/ 1200] Overall Loss 0.268535 Objective Loss 0.268535 LR 0.001000 Time 0.024758 -2022-12-06 11:03:58,101 - Epoch: [93][ 140/ 1200] Overall Loss 0.270526 Objective Loss 0.270526 LR 0.001000 Time 0.024358 -2022-12-06 11:03:58,293 - Epoch: [93][ 150/ 1200] Overall Loss 0.269093 Objective Loss 0.269093 LR 0.001000 Time 0.024006 -2022-12-06 11:03:58,484 - Epoch: [93][ 160/ 1200] Overall Loss 0.268691 Objective Loss 0.268691 LR 0.001000 Time 0.023701 -2022-12-06 11:03:58,676 - Epoch: [93][ 170/ 1200] Overall Loss 0.269319 Objective Loss 0.269319 LR 0.001000 Time 0.023432 -2022-12-06 11:03:58,868 - Epoch: [93][ 180/ 1200] Overall Loss 0.270422 Objective Loss 0.270422 LR 0.001000 Time 0.023195 -2022-12-06 11:03:59,060 - Epoch: [93][ 190/ 1200] Overall Loss 0.270360 Objective Loss 0.270360 LR 0.001000 Time 0.022981 -2022-12-06 11:03:59,251 - Epoch: [93][ 200/ 1200] Overall Loss 0.269976 Objective Loss 0.269976 LR 0.001000 Time 0.022784 -2022-12-06 11:03:59,442 - Epoch: [93][ 210/ 1200] Overall Loss 0.268902 Objective Loss 0.268902 LR 0.001000 Time 0.022607 -2022-12-06 11:03:59,635 - Epoch: [93][ 220/ 1200] Overall Loss 0.268909 Objective Loss 0.268909 LR 0.001000 Time 0.022452 -2022-12-06 11:03:59,826 - Epoch: [93][ 230/ 1200] Overall Loss 0.269280 Objective Loss 0.269280 LR 0.001000 Time 0.022305 -2022-12-06 11:04:00,018 - Epoch: [93][ 240/ 1200] Overall Loss 0.268890 Objective Loss 0.268890 LR 0.001000 Time 0.022171 -2022-12-06 11:04:00,209 - Epoch: [93][ 250/ 1200] Overall Loss 0.269573 Objective Loss 0.269573 LR 0.001000 Time 0.022048 -2022-12-06 11:04:00,402 - Epoch: [93][ 260/ 1200] Overall Loss 0.270013 Objective Loss 0.270013 LR 0.001000 Time 0.021937 -2022-12-06 11:04:00,593 - Epoch: [93][ 270/ 1200] Overall Loss 0.270971 Objective Loss 0.270971 LR 0.001000 Time 0.021833 -2022-12-06 11:04:00,786 - Epoch: [93][ 280/ 1200] Overall Loss 0.271295 Objective Loss 0.271295 LR 0.001000 Time 0.021737 -2022-12-06 11:04:00,977 - Epoch: [93][ 290/ 1200] Overall Loss 0.271109 Objective Loss 0.271109 LR 0.001000 Time 0.021647 -2022-12-06 11:04:01,169 - Epoch: [93][ 300/ 1200] Overall Loss 0.271179 Objective Loss 0.271179 LR 0.001000 Time 0.021563 -2022-12-06 11:04:01,361 - Epoch: [93][ 310/ 1200] Overall Loss 0.271423 Objective Loss 0.271423 LR 0.001000 Time 0.021484 -2022-12-06 11:04:01,554 - Epoch: [93][ 320/ 1200] Overall Loss 0.269991 Objective Loss 0.269991 LR 0.001000 Time 0.021413 -2022-12-06 11:04:01,745 - Epoch: [93][ 330/ 1200] Overall Loss 0.269791 Objective Loss 0.269791 LR 0.001000 Time 0.021344 -2022-12-06 11:04:01,937 - Epoch: [93][ 340/ 1200] Overall Loss 0.269589 Objective Loss 0.269589 LR 0.001000 Time 0.021278 -2022-12-06 11:04:02,129 - Epoch: [93][ 350/ 1200] Overall Loss 0.269061 Objective Loss 0.269061 LR 0.001000 Time 0.021217 -2022-12-06 11:04:02,321 - Epoch: [93][ 360/ 1200] Overall Loss 0.269497 Objective Loss 0.269497 LR 0.001000 Time 0.021159 -2022-12-06 11:04:02,513 - Epoch: [93][ 370/ 1200] Overall Loss 0.269368 Objective Loss 0.269368 LR 0.001000 Time 0.021105 -2022-12-06 11:04:02,705 - Epoch: [93][ 380/ 1200] Overall Loss 0.270440 Objective Loss 0.270440 LR 0.001000 Time 0.021054 -2022-12-06 11:04:02,896 - Epoch: [93][ 390/ 1200] Overall Loss 0.271181 Objective Loss 0.271181 LR 0.001000 Time 0.021002 -2022-12-06 11:04:03,088 - Epoch: [93][ 400/ 1200] Overall Loss 0.270980 Objective Loss 0.270980 LR 0.001000 Time 0.020954 -2022-12-06 11:04:03,279 - Epoch: [93][ 410/ 1200] Overall Loss 0.270971 Objective Loss 0.270971 LR 0.001000 Time 0.020909 -2022-12-06 11:04:03,472 - Epoch: [93][ 420/ 1200] Overall Loss 0.271158 Objective Loss 0.271158 LR 0.001000 Time 0.020869 -2022-12-06 11:04:03,664 - Epoch: [93][ 430/ 1200] Overall Loss 0.271255 Objective Loss 0.271255 LR 0.001000 Time 0.020827 -2022-12-06 11:04:03,855 - Epoch: [93][ 440/ 1200] Overall Loss 0.271581 Objective Loss 0.271581 LR 0.001000 Time 0.020788 -2022-12-06 11:04:04,046 - Epoch: [93][ 450/ 1200] Overall Loss 0.271575 Objective Loss 0.271575 LR 0.001000 Time 0.020750 -2022-12-06 11:04:04,239 - Epoch: [93][ 460/ 1200] Overall Loss 0.271512 Objective Loss 0.271512 LR 0.001000 Time 0.020716 -2022-12-06 11:04:04,430 - Epoch: [93][ 470/ 1200] Overall Loss 0.271988 Objective Loss 0.271988 LR 0.001000 Time 0.020681 -2022-12-06 11:04:04,623 - Epoch: [93][ 480/ 1200] Overall Loss 0.272280 Objective Loss 0.272280 LR 0.001000 Time 0.020651 -2022-12-06 11:04:04,815 - Epoch: [93][ 490/ 1200] Overall Loss 0.272684 Objective Loss 0.272684 LR 0.001000 Time 0.020619 -2022-12-06 11:04:05,008 - Epoch: [93][ 500/ 1200] Overall Loss 0.273122 Objective Loss 0.273122 LR 0.001000 Time 0.020592 -2022-12-06 11:04:05,199 - Epoch: [93][ 510/ 1200] Overall Loss 0.273562 Objective Loss 0.273562 LR 0.001000 Time 0.020562 -2022-12-06 11:04:05,392 - Epoch: [93][ 520/ 1200] Overall Loss 0.273826 Objective Loss 0.273826 LR 0.001000 Time 0.020536 -2022-12-06 11:04:05,582 - Epoch: [93][ 530/ 1200] Overall Loss 0.274574 Objective Loss 0.274574 LR 0.001000 Time 0.020508 -2022-12-06 11:04:05,775 - Epoch: [93][ 540/ 1200] Overall Loss 0.274871 Objective Loss 0.274871 LR 0.001000 Time 0.020484 -2022-12-06 11:04:05,967 - Epoch: [93][ 550/ 1200] Overall Loss 0.275549 Objective Loss 0.275549 LR 0.001000 Time 0.020459 -2022-12-06 11:04:06,159 - Epoch: [93][ 560/ 1200] Overall Loss 0.275471 Objective Loss 0.275471 LR 0.001000 Time 0.020435 -2022-12-06 11:04:06,350 - Epoch: [93][ 570/ 1200] Overall Loss 0.275510 Objective Loss 0.275510 LR 0.001000 Time 0.020411 -2022-12-06 11:04:06,542 - Epoch: [93][ 580/ 1200] Overall Loss 0.275912 Objective Loss 0.275912 LR 0.001000 Time 0.020389 -2022-12-06 11:04:06,734 - Epoch: [93][ 590/ 1200] Overall Loss 0.276044 Objective Loss 0.276044 LR 0.001000 Time 0.020368 -2022-12-06 11:04:06,927 - Epoch: [93][ 600/ 1200] Overall Loss 0.276136 Objective Loss 0.276136 LR 0.001000 Time 0.020349 -2022-12-06 11:04:07,118 - Epoch: [93][ 610/ 1200] Overall Loss 0.276526 Objective Loss 0.276526 LR 0.001000 Time 0.020327 -2022-12-06 11:04:07,310 - Epoch: [93][ 620/ 1200] Overall Loss 0.276344 Objective Loss 0.276344 LR 0.001000 Time 0.020309 -2022-12-06 11:04:07,502 - Epoch: [93][ 630/ 1200] Overall Loss 0.276951 Objective Loss 0.276951 LR 0.001000 Time 0.020291 -2022-12-06 11:04:07,695 - Epoch: [93][ 640/ 1200] Overall Loss 0.277358 Objective Loss 0.277358 LR 0.001000 Time 0.020273 -2022-12-06 11:04:07,886 - Epoch: [93][ 650/ 1200] Overall Loss 0.277875 Objective Loss 0.277875 LR 0.001000 Time 0.020255 -2022-12-06 11:04:08,078 - Epoch: [93][ 660/ 1200] Overall Loss 0.277965 Objective Loss 0.277965 LR 0.001000 Time 0.020238 -2022-12-06 11:04:08,270 - Epoch: [93][ 670/ 1200] Overall Loss 0.277967 Objective Loss 0.277967 LR 0.001000 Time 0.020222 -2022-12-06 11:04:08,462 - Epoch: [93][ 680/ 1200] Overall Loss 0.277975 Objective Loss 0.277975 LR 0.001000 Time 0.020206 -2022-12-06 11:04:08,655 - Epoch: [93][ 690/ 1200] Overall Loss 0.278042 Objective Loss 0.278042 LR 0.001000 Time 0.020191 -2022-12-06 11:04:08,847 - Epoch: [93][ 700/ 1200] Overall Loss 0.278161 Objective Loss 0.278161 LR 0.001000 Time 0.020177 -2022-12-06 11:04:09,039 - Epoch: [93][ 710/ 1200] Overall Loss 0.278187 Objective Loss 0.278187 LR 0.001000 Time 0.020162 -2022-12-06 11:04:09,232 - Epoch: [93][ 720/ 1200] Overall Loss 0.278095 Objective Loss 0.278095 LR 0.001000 Time 0.020149 -2022-12-06 11:04:09,424 - Epoch: [93][ 730/ 1200] Overall Loss 0.278097 Objective Loss 0.278097 LR 0.001000 Time 0.020135 -2022-12-06 11:04:09,616 - Epoch: [93][ 740/ 1200] Overall Loss 0.277894 Objective Loss 0.277894 LR 0.001000 Time 0.020122 -2022-12-06 11:04:09,808 - Epoch: [93][ 750/ 1200] Overall Loss 0.277984 Objective Loss 0.277984 LR 0.001000 Time 0.020109 -2022-12-06 11:04:10,000 - Epoch: [93][ 760/ 1200] Overall Loss 0.278234 Objective Loss 0.278234 LR 0.001000 Time 0.020096 -2022-12-06 11:04:10,191 - Epoch: [93][ 770/ 1200] Overall Loss 0.278293 Objective Loss 0.278293 LR 0.001000 Time 0.020083 -2022-12-06 11:04:10,383 - Epoch: [93][ 780/ 1200] Overall Loss 0.278484 Objective Loss 0.278484 LR 0.001000 Time 0.020071 -2022-12-06 11:04:10,575 - Epoch: [93][ 790/ 1200] Overall Loss 0.278208 Objective Loss 0.278208 LR 0.001000 Time 0.020059 -2022-12-06 11:04:10,767 - Epoch: [93][ 800/ 1200] Overall Loss 0.278066 Objective Loss 0.278066 LR 0.001000 Time 0.020048 -2022-12-06 11:04:10,959 - Epoch: [93][ 810/ 1200] Overall Loss 0.278310 Objective Loss 0.278310 LR 0.001000 Time 0.020036 -2022-12-06 11:04:11,151 - Epoch: [93][ 820/ 1200] Overall Loss 0.278215 Objective Loss 0.278215 LR 0.001000 Time 0.020025 -2022-12-06 11:04:11,343 - Epoch: [93][ 830/ 1200] Overall Loss 0.278488 Objective Loss 0.278488 LR 0.001000 Time 0.020015 -2022-12-06 11:04:11,536 - Epoch: [93][ 840/ 1200] Overall Loss 0.278302 Objective Loss 0.278302 LR 0.001000 Time 0.020005 -2022-12-06 11:04:11,728 - Epoch: [93][ 850/ 1200] Overall Loss 0.278654 Objective Loss 0.278654 LR 0.001000 Time 0.019996 -2022-12-06 11:04:11,921 - Epoch: [93][ 860/ 1200] Overall Loss 0.278593 Objective Loss 0.278593 LR 0.001000 Time 0.019987 -2022-12-06 11:04:12,113 - Epoch: [93][ 870/ 1200] Overall Loss 0.278929 Objective Loss 0.278929 LR 0.001000 Time 0.019977 -2022-12-06 11:04:12,305 - Epoch: [93][ 880/ 1200] Overall Loss 0.279145 Objective Loss 0.279145 LR 0.001000 Time 0.019967 -2022-12-06 11:04:12,497 - Epoch: [93][ 890/ 1200] Overall Loss 0.279182 Objective Loss 0.279182 LR 0.001000 Time 0.019958 -2022-12-06 11:04:12,690 - Epoch: [93][ 900/ 1200] Overall Loss 0.279781 Objective Loss 0.279781 LR 0.001000 Time 0.019950 -2022-12-06 11:04:12,881 - Epoch: [93][ 910/ 1200] Overall Loss 0.279653 Objective Loss 0.279653 LR 0.001000 Time 0.019941 -2022-12-06 11:04:13,074 - Epoch: [93][ 920/ 1200] Overall Loss 0.279801 Objective Loss 0.279801 LR 0.001000 Time 0.019932 -2022-12-06 11:04:13,266 - Epoch: [93][ 930/ 1200] Overall Loss 0.279853 Objective Loss 0.279853 LR 0.001000 Time 0.019924 -2022-12-06 11:04:13,458 - Epoch: [93][ 940/ 1200] Overall Loss 0.279766 Objective Loss 0.279766 LR 0.001000 Time 0.019916 -2022-12-06 11:04:13,650 - Epoch: [93][ 950/ 1200] Overall Loss 0.279755 Objective Loss 0.279755 LR 0.001000 Time 0.019908 -2022-12-06 11:04:13,842 - Epoch: [93][ 960/ 1200] Overall Loss 0.279502 Objective Loss 0.279502 LR 0.001000 Time 0.019900 -2022-12-06 11:04:14,035 - Epoch: [93][ 970/ 1200] Overall Loss 0.279524 Objective Loss 0.279524 LR 0.001000 Time 0.019893 -2022-12-06 11:04:14,225 - Epoch: [93][ 980/ 1200] Overall Loss 0.279512 Objective Loss 0.279512 LR 0.001000 Time 0.019884 -2022-12-06 11:04:14,417 - Epoch: [93][ 990/ 1200] Overall Loss 0.279752 Objective Loss 0.279752 LR 0.001000 Time 0.019875 -2022-12-06 11:04:14,608 - Epoch: [93][ 1000/ 1200] Overall Loss 0.280052 Objective Loss 0.280052 LR 0.001000 Time 0.019867 -2022-12-06 11:04:14,799 - Epoch: [93][ 1010/ 1200] Overall Loss 0.280208 Objective Loss 0.280208 LR 0.001000 Time 0.019860 -2022-12-06 11:04:14,990 - Epoch: [93][ 1020/ 1200] Overall Loss 0.280382 Objective Loss 0.280382 LR 0.001000 Time 0.019851 -2022-12-06 11:04:15,181 - Epoch: [93][ 1030/ 1200] Overall Loss 0.280221 Objective Loss 0.280221 LR 0.001000 Time 0.019844 -2022-12-06 11:04:15,373 - Epoch: [93][ 1040/ 1200] Overall Loss 0.280316 Objective Loss 0.280316 LR 0.001000 Time 0.019836 -2022-12-06 11:04:15,564 - Epoch: [93][ 1050/ 1200] Overall Loss 0.280439 Objective Loss 0.280439 LR 0.001000 Time 0.019829 -2022-12-06 11:04:15,755 - Epoch: [93][ 1060/ 1200] Overall Loss 0.280434 Objective Loss 0.280434 LR 0.001000 Time 0.019822 -2022-12-06 11:04:15,947 - Epoch: [93][ 1070/ 1200] Overall Loss 0.280787 Objective Loss 0.280787 LR 0.001000 Time 0.019815 -2022-12-06 11:04:16,138 - Epoch: [93][ 1080/ 1200] Overall Loss 0.280734 Objective Loss 0.280734 LR 0.001000 Time 0.019808 -2022-12-06 11:04:16,329 - Epoch: [93][ 1090/ 1200] Overall Loss 0.280947 Objective Loss 0.280947 LR 0.001000 Time 0.019802 -2022-12-06 11:04:16,520 - Epoch: [93][ 1100/ 1200] Overall Loss 0.281175 Objective Loss 0.281175 LR 0.001000 Time 0.019795 -2022-12-06 11:04:16,712 - Epoch: [93][ 1110/ 1200] Overall Loss 0.281239 Objective Loss 0.281239 LR 0.001000 Time 0.019788 -2022-12-06 11:04:16,904 - Epoch: [93][ 1120/ 1200] Overall Loss 0.281214 Objective Loss 0.281214 LR 0.001000 Time 0.019782 -2022-12-06 11:04:17,095 - Epoch: [93][ 1130/ 1200] Overall Loss 0.280973 Objective Loss 0.280973 LR 0.001000 Time 0.019776 -2022-12-06 11:04:17,286 - Epoch: [93][ 1140/ 1200] Overall Loss 0.280986 Objective Loss 0.280986 LR 0.001000 Time 0.019770 -2022-12-06 11:04:17,478 - Epoch: [93][ 1150/ 1200] Overall Loss 0.281139 Objective Loss 0.281139 LR 0.001000 Time 0.019764 -2022-12-06 11:04:17,668 - Epoch: [93][ 1160/ 1200] Overall Loss 0.280981 Objective Loss 0.280981 LR 0.001000 Time 0.019757 -2022-12-06 11:04:17,860 - Epoch: [93][ 1170/ 1200] Overall Loss 0.281080 Objective Loss 0.281080 LR 0.001000 Time 0.019752 -2022-12-06 11:04:18,050 - Epoch: [93][ 1180/ 1200] Overall Loss 0.281251 Objective Loss 0.281251 LR 0.001000 Time 0.019745 -2022-12-06 11:04:18,241 - Epoch: [93][ 1190/ 1200] Overall Loss 0.281337 Objective Loss 0.281337 LR 0.001000 Time 0.019739 -2022-12-06 11:04:18,472 - Epoch: [93][ 1200/ 1200] Overall Loss 0.281455 Objective Loss 0.281455 Top1 81.799163 Top5 97.489540 LR 0.001000 Time 0.019767 -2022-12-06 11:04:18,561 - --- validate (epoch=93)----------- -2022-12-06 11:04:18,562 - 34129 samples (256 per mini-batch) -2022-12-06 11:04:19,115 - Epoch: [93][ 10/ 134] Loss 0.306122 Top1 84.101562 Top5 97.812500 -2022-12-06 11:04:19,246 - Epoch: [93][ 20/ 134] Loss 0.305556 Top1 84.980469 Top5 97.753906 -2022-12-06 11:04:19,378 - Epoch: [93][ 30/ 134] Loss 0.300534 Top1 84.856771 Top5 97.825521 -2022-12-06 11:04:19,510 - Epoch: [93][ 40/ 134] Loss 0.288687 Top1 84.980469 Top5 97.880859 -2022-12-06 11:04:19,641 - Epoch: [93][ 50/ 134] Loss 0.288560 Top1 84.835938 Top5 97.835938 -2022-12-06 11:04:19,773 - Epoch: [93][ 60/ 134] Loss 0.290402 Top1 84.850260 Top5 97.897135 -2022-12-06 11:04:19,902 - Epoch: [93][ 70/ 134] Loss 0.292764 Top1 84.832589 Top5 97.868304 -2022-12-06 11:04:20,027 - Epoch: [93][ 80/ 134] Loss 0.291821 Top1 84.960938 Top5 97.910156 -2022-12-06 11:04:20,164 - Epoch: [93][ 90/ 134] Loss 0.290790 Top1 84.895833 Top5 97.894965 -2022-12-06 11:04:20,291 - Epoch: [93][ 100/ 134] Loss 0.292375 Top1 84.867188 Top5 97.878906 -2022-12-06 11:04:20,429 - Epoch: [93][ 110/ 134] Loss 0.291911 Top1 84.840199 Top5 97.901278 -2022-12-06 11:04:20,574 - Epoch: [93][ 120/ 134] Loss 0.294513 Top1 84.794922 Top5 97.916667 -2022-12-06 11:04:20,709 - Epoch: [93][ 130/ 134] Loss 0.295793 Top1 84.813702 Top5 97.893630 -2022-12-06 11:04:20,746 - Epoch: [93][ 134/ 134] Loss 0.297287 Top1 84.810572 Top5 97.907938 -2022-12-06 11:04:20,836 - ==> Top1: 84.811 Top5: 97.908 Loss: 0.297 - -2022-12-06 11:04:20,836 - ==> Confusion: -[[ 888 3 2 2 12 5 0 0 5 56 0 3 3 5 7 2 0 0 0 0 3] - [ 0 918 2 2 11 22 2 19 2 1 4 3 2 3 3 0 7 1 17 2 6] - [ 5 2 983 20 4 5 21 8 0 2 6 3 5 4 7 2 1 0 6 2 17] - [ 2 1 24 921 1 0 2 0 1 0 11 0 5 4 21 1 2 3 11 0 10] - [ 9 3 1 0 956 7 0 1 0 7 1 4 1 2 8 6 8 2 1 0 3] - [ 2 24 2 1 8 944 2 24 3 4 1 12 5 13 0 1 1 1 1 9 11] - [ 0 2 12 1 0 4 1062 2 1 0 3 1 4 2 1 8 0 2 2 9 2] - [ 2 9 7 2 2 21 5 943 1 2 4 8 5 1 1 2 1 0 23 9 6] - [ 4 3 0 0 1 2 1 0 973 46 5 1 1 5 15 2 2 0 1 1 1] - [ 60 0 0 0 3 1 0 2 24 886 2 2 2 9 4 0 0 0 1 0 5] - [ 3 1 2 4 3 0 2 4 11 1 954 2 5 15 3 0 0 0 3 1 5] - [ 2 5 1 0 0 7 3 3 2 1 1 960 32 8 0 8 2 3 0 9 4] - [ 1 0 1 3 1 3 0 2 0 1 0 30 890 2 3 10 2 12 0 2 6] - [ 0 1 1 0 1 7 0 4 12 17 6 4 4 945 4 2 3 0 1 1 10] - [ 4 5 1 14 3 3 0 0 17 8 1 2 1 2 1053 0 0 0 4 2 10] - [ 1 0 1 0 3 0 5 0 2 1 0 8 10 1 0 983 10 11 0 1 6] - [ 2 3 0 1 5 1 1 0 2 0 0 3 3 3 1 10 1030 0 0 2 5] - [ 4 0 1 3 0 1 1 0 3 0 1 8 25 4 1 11 5 965 0 2 1] - [ 2 6 4 8 1 1 0 20 2 1 12 2 5 1 8 1 2 1 928 1 2] - [ 3 3 2 1 1 3 9 4 1 1 1 18 9 10 0 5 4 2 0 994 9] - [ 150 223 155 108 145 126 85 148 104 122 177 108 414 340 188 144 227 82 165 256 9759]] - -2022-12-06 11:04:21,402 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:04:21,402 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:04:21,408 - - -2022-12-06 11:04:21,408 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:04:22,324 - Epoch: [94][ 10/ 1200] Overall Loss 0.270101 Objective Loss 0.270101 LR 0.001000 Time 0.091596 -2022-12-06 11:04:22,520 - Epoch: [94][ 20/ 1200] Overall Loss 0.279986 Objective Loss 0.279986 LR 0.001000 Time 0.055551 -2022-12-06 11:04:22,711 - Epoch: [94][ 30/ 1200] Overall Loss 0.284290 Objective Loss 0.284290 LR 0.001000 Time 0.043372 -2022-12-06 11:04:22,901 - Epoch: [94][ 40/ 1200] Overall Loss 0.283598 Objective Loss 0.283598 LR 0.001000 Time 0.037266 -2022-12-06 11:04:23,091 - Epoch: [94][ 50/ 1200] Overall Loss 0.284383 Objective Loss 0.284383 LR 0.001000 Time 0.033602 -2022-12-06 11:04:23,281 - Epoch: [94][ 60/ 1200] Overall Loss 0.277815 Objective Loss 0.277815 LR 0.001000 Time 0.031170 -2022-12-06 11:04:23,471 - Epoch: [94][ 70/ 1200] Overall Loss 0.275448 Objective Loss 0.275448 LR 0.001000 Time 0.029417 -2022-12-06 11:04:23,661 - Epoch: [94][ 80/ 1200] Overall Loss 0.275871 Objective Loss 0.275871 LR 0.001000 Time 0.028104 -2022-12-06 11:04:23,851 - Epoch: [94][ 90/ 1200] Overall Loss 0.275159 Objective Loss 0.275159 LR 0.001000 Time 0.027091 -2022-12-06 11:04:24,041 - Epoch: [94][ 100/ 1200] Overall Loss 0.277340 Objective Loss 0.277340 LR 0.001000 Time 0.026277 -2022-12-06 11:04:24,232 - Epoch: [94][ 110/ 1200] Overall Loss 0.275371 Objective Loss 0.275371 LR 0.001000 Time 0.025614 -2022-12-06 11:04:24,422 - Epoch: [94][ 120/ 1200] Overall Loss 0.276434 Objective Loss 0.276434 LR 0.001000 Time 0.025063 -2022-12-06 11:04:24,613 - Epoch: [94][ 130/ 1200] Overall Loss 0.275155 Objective Loss 0.275155 LR 0.001000 Time 0.024597 -2022-12-06 11:04:24,803 - Epoch: [94][ 140/ 1200] Overall Loss 0.275544 Objective Loss 0.275544 LR 0.001000 Time 0.024197 -2022-12-06 11:04:24,994 - Epoch: [94][ 150/ 1200] Overall Loss 0.275264 Objective Loss 0.275264 LR 0.001000 Time 0.023852 -2022-12-06 11:04:25,184 - Epoch: [94][ 160/ 1200] Overall Loss 0.275567 Objective Loss 0.275567 LR 0.001000 Time 0.023548 -2022-12-06 11:04:25,375 - Epoch: [94][ 170/ 1200] Overall Loss 0.275272 Objective Loss 0.275272 LR 0.001000 Time 0.023280 -2022-12-06 11:04:25,565 - Epoch: [94][ 180/ 1200] Overall Loss 0.274738 Objective Loss 0.274738 LR 0.001000 Time 0.023038 -2022-12-06 11:04:25,755 - Epoch: [94][ 190/ 1200] Overall Loss 0.273444 Objective Loss 0.273444 LR 0.001000 Time 0.022824 -2022-12-06 11:04:25,945 - Epoch: [94][ 200/ 1200] Overall Loss 0.273156 Objective Loss 0.273156 LR 0.001000 Time 0.022628 -2022-12-06 11:04:26,136 - Epoch: [94][ 210/ 1200] Overall Loss 0.272852 Objective Loss 0.272852 LR 0.001000 Time 0.022458 -2022-12-06 11:04:26,326 - Epoch: [94][ 220/ 1200] Overall Loss 0.272912 Objective Loss 0.272912 LR 0.001000 Time 0.022298 -2022-12-06 11:04:26,517 - Epoch: [94][ 230/ 1200] Overall Loss 0.273697 Objective Loss 0.273697 LR 0.001000 Time 0.022159 -2022-12-06 11:04:26,707 - Epoch: [94][ 240/ 1200] Overall Loss 0.274534 Objective Loss 0.274534 LR 0.001000 Time 0.022023 -2022-12-06 11:04:26,897 - Epoch: [94][ 250/ 1200] Overall Loss 0.273727 Objective Loss 0.273727 LR 0.001000 Time 0.021901 -2022-12-06 11:04:27,087 - Epoch: [94][ 260/ 1200] Overall Loss 0.273706 Objective Loss 0.273706 LR 0.001000 Time 0.021787 -2022-12-06 11:04:27,278 - Epoch: [94][ 270/ 1200] Overall Loss 0.273267 Objective Loss 0.273267 LR 0.001000 Time 0.021688 -2022-12-06 11:04:27,469 - Epoch: [94][ 280/ 1200] Overall Loss 0.273345 Objective Loss 0.273345 LR 0.001000 Time 0.021591 -2022-12-06 11:04:27,660 - Epoch: [94][ 290/ 1200] Overall Loss 0.272868 Objective Loss 0.272868 LR 0.001000 Time 0.021504 -2022-12-06 11:04:27,850 - Epoch: [94][ 300/ 1200] Overall Loss 0.272584 Objective Loss 0.272584 LR 0.001000 Time 0.021420 -2022-12-06 11:04:28,040 - Epoch: [94][ 310/ 1200] Overall Loss 0.273071 Objective Loss 0.273071 LR 0.001000 Time 0.021340 -2022-12-06 11:04:28,231 - Epoch: [94][ 320/ 1200] Overall Loss 0.273303 Objective Loss 0.273303 LR 0.001000 Time 0.021267 -2022-12-06 11:04:28,421 - Epoch: [94][ 330/ 1200] Overall Loss 0.273167 Objective Loss 0.273167 LR 0.001000 Time 0.021197 -2022-12-06 11:04:28,611 - Epoch: [94][ 340/ 1200] Overall Loss 0.273143 Objective Loss 0.273143 LR 0.001000 Time 0.021130 -2022-12-06 11:04:28,801 - Epoch: [94][ 350/ 1200] Overall Loss 0.272602 Objective Loss 0.272602 LR 0.001000 Time 0.021069 -2022-12-06 11:04:28,991 - Epoch: [94][ 360/ 1200] Overall Loss 0.272278 Objective Loss 0.272278 LR 0.001000 Time 0.021010 -2022-12-06 11:04:29,182 - Epoch: [94][ 370/ 1200] Overall Loss 0.272099 Objective Loss 0.272099 LR 0.001000 Time 0.020955 -2022-12-06 11:04:29,371 - Epoch: [94][ 380/ 1200] Overall Loss 0.272035 Objective Loss 0.272035 LR 0.001000 Time 0.020902 -2022-12-06 11:04:29,562 - Epoch: [94][ 390/ 1200] Overall Loss 0.272264 Objective Loss 0.272264 LR 0.001000 Time 0.020853 -2022-12-06 11:04:29,752 - Epoch: [94][ 400/ 1200] Overall Loss 0.272393 Objective Loss 0.272393 LR 0.001000 Time 0.020804 -2022-12-06 11:04:29,942 - Epoch: [94][ 410/ 1200] Overall Loss 0.272799 Objective Loss 0.272799 LR 0.001000 Time 0.020761 -2022-12-06 11:04:30,133 - Epoch: [94][ 420/ 1200] Overall Loss 0.273067 Objective Loss 0.273067 LR 0.001000 Time 0.020719 -2022-12-06 11:04:30,323 - Epoch: [94][ 430/ 1200] Overall Loss 0.273201 Objective Loss 0.273201 LR 0.001000 Time 0.020679 -2022-12-06 11:04:30,514 - Epoch: [94][ 440/ 1200] Overall Loss 0.273095 Objective Loss 0.273095 LR 0.001000 Time 0.020640 -2022-12-06 11:04:30,704 - Epoch: [94][ 450/ 1200] Overall Loss 0.273780 Objective Loss 0.273780 LR 0.001000 Time 0.020604 -2022-12-06 11:04:30,894 - Epoch: [94][ 460/ 1200] Overall Loss 0.275126 Objective Loss 0.275126 LR 0.001000 Time 0.020568 -2022-12-06 11:04:31,084 - Epoch: [94][ 470/ 1200] Overall Loss 0.275012 Objective Loss 0.275012 LR 0.001000 Time 0.020534 -2022-12-06 11:04:31,275 - Epoch: [94][ 480/ 1200] Overall Loss 0.275069 Objective Loss 0.275069 LR 0.001000 Time 0.020501 -2022-12-06 11:04:31,466 - Epoch: [94][ 490/ 1200] Overall Loss 0.274991 Objective Loss 0.274991 LR 0.001000 Time 0.020471 -2022-12-06 11:04:31,655 - Epoch: [94][ 500/ 1200] Overall Loss 0.274519 Objective Loss 0.274519 LR 0.001000 Time 0.020439 -2022-12-06 11:04:31,846 - Epoch: [94][ 510/ 1200] Overall Loss 0.274510 Objective Loss 0.274510 LR 0.001000 Time 0.020412 -2022-12-06 11:04:32,037 - Epoch: [94][ 520/ 1200] Overall Loss 0.274356 Objective Loss 0.274356 LR 0.001000 Time 0.020386 -2022-12-06 11:04:32,228 - Epoch: [94][ 530/ 1200] Overall Loss 0.274233 Objective Loss 0.274233 LR 0.001000 Time 0.020360 -2022-12-06 11:04:32,418 - Epoch: [94][ 540/ 1200] Overall Loss 0.274138 Objective Loss 0.274138 LR 0.001000 Time 0.020334 -2022-12-06 11:04:32,608 - Epoch: [94][ 550/ 1200] Overall Loss 0.274046 Objective Loss 0.274046 LR 0.001000 Time 0.020310 -2022-12-06 11:04:32,798 - Epoch: [94][ 560/ 1200] Overall Loss 0.273603 Objective Loss 0.273603 LR 0.001000 Time 0.020285 -2022-12-06 11:04:32,989 - Epoch: [94][ 570/ 1200] Overall Loss 0.273842 Objective Loss 0.273842 LR 0.001000 Time 0.020263 -2022-12-06 11:04:33,180 - Epoch: [94][ 580/ 1200] Overall Loss 0.273942 Objective Loss 0.273942 LR 0.001000 Time 0.020241 -2022-12-06 11:04:33,369 - Epoch: [94][ 590/ 1200] Overall Loss 0.274369 Objective Loss 0.274369 LR 0.001000 Time 0.020219 -2022-12-06 11:04:33,560 - Epoch: [94][ 600/ 1200] Overall Loss 0.274539 Objective Loss 0.274539 LR 0.001000 Time 0.020199 -2022-12-06 11:04:33,751 - Epoch: [94][ 610/ 1200] Overall Loss 0.274947 Objective Loss 0.274947 LR 0.001000 Time 0.020181 -2022-12-06 11:04:33,942 - Epoch: [94][ 620/ 1200] Overall Loss 0.274926 Objective Loss 0.274926 LR 0.001000 Time 0.020162 -2022-12-06 11:04:34,132 - Epoch: [94][ 630/ 1200] Overall Loss 0.275285 Objective Loss 0.275285 LR 0.001000 Time 0.020142 -2022-12-06 11:04:34,322 - Epoch: [94][ 640/ 1200] Overall Loss 0.275527 Objective Loss 0.275527 LR 0.001000 Time 0.020124 -2022-12-06 11:04:34,513 - Epoch: [94][ 650/ 1200] Overall Loss 0.275911 Objective Loss 0.275911 LR 0.001000 Time 0.020107 -2022-12-06 11:04:34,703 - Epoch: [94][ 660/ 1200] Overall Loss 0.275226 Objective Loss 0.275226 LR 0.001000 Time 0.020090 -2022-12-06 11:04:34,895 - Epoch: [94][ 670/ 1200] Overall Loss 0.274762 Objective Loss 0.274762 LR 0.001000 Time 0.020075 -2022-12-06 11:04:35,084 - Epoch: [94][ 680/ 1200] Overall Loss 0.274754 Objective Loss 0.274754 LR 0.001000 Time 0.020058 -2022-12-06 11:04:35,274 - Epoch: [94][ 690/ 1200] Overall Loss 0.274519 Objective Loss 0.274519 LR 0.001000 Time 0.020042 -2022-12-06 11:04:35,465 - Epoch: [94][ 700/ 1200] Overall Loss 0.274741 Objective Loss 0.274741 LR 0.001000 Time 0.020026 -2022-12-06 11:04:35,654 - Epoch: [94][ 710/ 1200] Overall Loss 0.274446 Objective Loss 0.274446 LR 0.001000 Time 0.020010 -2022-12-06 11:04:35,844 - Epoch: [94][ 720/ 1200] Overall Loss 0.274295 Objective Loss 0.274295 LR 0.001000 Time 0.019996 -2022-12-06 11:04:36,035 - Epoch: [94][ 730/ 1200] Overall Loss 0.274036 Objective Loss 0.274036 LR 0.001000 Time 0.019982 -2022-12-06 11:04:36,225 - Epoch: [94][ 740/ 1200] Overall Loss 0.274184 Objective Loss 0.274184 LR 0.001000 Time 0.019968 -2022-12-06 11:04:36,415 - Epoch: [94][ 750/ 1200] Overall Loss 0.274266 Objective Loss 0.274266 LR 0.001000 Time 0.019955 -2022-12-06 11:04:36,606 - Epoch: [94][ 760/ 1200] Overall Loss 0.274097 Objective Loss 0.274097 LR 0.001000 Time 0.019943 -2022-12-06 11:04:36,797 - Epoch: [94][ 770/ 1200] Overall Loss 0.274327 Objective Loss 0.274327 LR 0.001000 Time 0.019931 -2022-12-06 11:04:36,987 - Epoch: [94][ 780/ 1200] Overall Loss 0.274512 Objective Loss 0.274512 LR 0.001000 Time 0.019919 -2022-12-06 11:04:37,178 - Epoch: [94][ 790/ 1200] Overall Loss 0.274430 Objective Loss 0.274430 LR 0.001000 Time 0.019908 -2022-12-06 11:04:37,368 - Epoch: [94][ 800/ 1200] Overall Loss 0.274141 Objective Loss 0.274141 LR 0.001000 Time 0.019895 -2022-12-06 11:04:37,559 - Epoch: [94][ 810/ 1200] Overall Loss 0.274069 Objective Loss 0.274069 LR 0.001000 Time 0.019885 -2022-12-06 11:04:37,749 - Epoch: [94][ 820/ 1200] Overall Loss 0.274067 Objective Loss 0.274067 LR 0.001000 Time 0.019873 -2022-12-06 11:04:37,940 - Epoch: [94][ 830/ 1200] Overall Loss 0.274607 Objective Loss 0.274607 LR 0.001000 Time 0.019863 -2022-12-06 11:04:38,130 - Epoch: [94][ 840/ 1200] Overall Loss 0.274442 Objective Loss 0.274442 LR 0.001000 Time 0.019853 -2022-12-06 11:04:38,321 - Epoch: [94][ 850/ 1200] Overall Loss 0.274206 Objective Loss 0.274206 LR 0.001000 Time 0.019843 -2022-12-06 11:04:38,511 - Epoch: [94][ 860/ 1200] Overall Loss 0.274312 Objective Loss 0.274312 LR 0.001000 Time 0.019833 -2022-12-06 11:04:38,701 - Epoch: [94][ 870/ 1200] Overall Loss 0.274356 Objective Loss 0.274356 LR 0.001000 Time 0.019823 -2022-12-06 11:04:38,892 - Epoch: [94][ 880/ 1200] Overall Loss 0.274261 Objective Loss 0.274261 LR 0.001000 Time 0.019813 -2022-12-06 11:04:39,081 - Epoch: [94][ 890/ 1200] Overall Loss 0.274214 Objective Loss 0.274214 LR 0.001000 Time 0.019803 -2022-12-06 11:04:39,272 - Epoch: [94][ 900/ 1200] Overall Loss 0.274702 Objective Loss 0.274702 LR 0.001000 Time 0.019795 -2022-12-06 11:04:39,463 - Epoch: [94][ 910/ 1200] Overall Loss 0.275156 Objective Loss 0.275156 LR 0.001000 Time 0.019786 -2022-12-06 11:04:39,653 - Epoch: [94][ 920/ 1200] Overall Loss 0.275042 Objective Loss 0.275042 LR 0.001000 Time 0.019777 -2022-12-06 11:04:39,843 - Epoch: [94][ 930/ 1200] Overall Loss 0.274856 Objective Loss 0.274856 LR 0.001000 Time 0.019768 -2022-12-06 11:04:40,033 - Epoch: [94][ 940/ 1200] Overall Loss 0.274969 Objective Loss 0.274969 LR 0.001000 Time 0.019759 -2022-12-06 11:04:40,223 - Epoch: [94][ 950/ 1200] Overall Loss 0.274889 Objective Loss 0.274889 LR 0.001000 Time 0.019751 -2022-12-06 11:04:40,414 - Epoch: [94][ 960/ 1200] Overall Loss 0.274668 Objective Loss 0.274668 LR 0.001000 Time 0.019743 -2022-12-06 11:04:40,604 - Epoch: [94][ 970/ 1200] Overall Loss 0.274670 Objective Loss 0.274670 LR 0.001000 Time 0.019735 -2022-12-06 11:04:40,795 - Epoch: [94][ 980/ 1200] Overall Loss 0.274886 Objective Loss 0.274886 LR 0.001000 Time 0.019728 -2022-12-06 11:04:40,985 - Epoch: [94][ 990/ 1200] Overall Loss 0.274759 Objective Loss 0.274759 LR 0.001000 Time 0.019720 -2022-12-06 11:04:41,175 - Epoch: [94][ 1000/ 1200] Overall Loss 0.274748 Objective Loss 0.274748 LR 0.001000 Time 0.019713 -2022-12-06 11:04:41,366 - Epoch: [94][ 1010/ 1200] Overall Loss 0.274700 Objective Loss 0.274700 LR 0.001000 Time 0.019706 -2022-12-06 11:04:41,557 - Epoch: [94][ 1020/ 1200] Overall Loss 0.274841 Objective Loss 0.274841 LR 0.001000 Time 0.019699 -2022-12-06 11:04:41,747 - Epoch: [94][ 1030/ 1200] Overall Loss 0.274871 Objective Loss 0.274871 LR 0.001000 Time 0.019692 -2022-12-06 11:04:41,938 - Epoch: [94][ 1040/ 1200] Overall Loss 0.274918 Objective Loss 0.274918 LR 0.001000 Time 0.019686 -2022-12-06 11:04:42,129 - Epoch: [94][ 1050/ 1200] Overall Loss 0.274768 Objective Loss 0.274768 LR 0.001000 Time 0.019680 -2022-12-06 11:04:42,319 - Epoch: [94][ 1060/ 1200] Overall Loss 0.274874 Objective Loss 0.274874 LR 0.001000 Time 0.019673 -2022-12-06 11:04:42,510 - Epoch: [94][ 1070/ 1200] Overall Loss 0.274729 Objective Loss 0.274729 LR 0.001000 Time 0.019667 -2022-12-06 11:04:42,701 - Epoch: [94][ 1080/ 1200] Overall Loss 0.274623 Objective Loss 0.274623 LR 0.001000 Time 0.019661 -2022-12-06 11:04:42,891 - Epoch: [94][ 1090/ 1200] Overall Loss 0.274746 Objective Loss 0.274746 LR 0.001000 Time 0.019655 -2022-12-06 11:04:43,082 - Epoch: [94][ 1100/ 1200] Overall Loss 0.274898 Objective Loss 0.274898 LR 0.001000 Time 0.019649 -2022-12-06 11:04:43,272 - Epoch: [94][ 1110/ 1200] Overall Loss 0.275002 Objective Loss 0.275002 LR 0.001000 Time 0.019643 -2022-12-06 11:04:43,463 - Epoch: [94][ 1120/ 1200] Overall Loss 0.274877 Objective Loss 0.274877 LR 0.001000 Time 0.019637 -2022-12-06 11:04:43,653 - Epoch: [94][ 1130/ 1200] Overall Loss 0.275006 Objective Loss 0.275006 LR 0.001000 Time 0.019631 -2022-12-06 11:04:43,844 - Epoch: [94][ 1140/ 1200] Overall Loss 0.275010 Objective Loss 0.275010 LR 0.001000 Time 0.019626 -2022-12-06 11:04:44,034 - Epoch: [94][ 1150/ 1200] Overall Loss 0.274899 Objective Loss 0.274899 LR 0.001000 Time 0.019620 -2022-12-06 11:04:44,224 - Epoch: [94][ 1160/ 1200] Overall Loss 0.274864 Objective Loss 0.274864 LR 0.001000 Time 0.019615 -2022-12-06 11:04:44,415 - Epoch: [94][ 1170/ 1200] Overall Loss 0.275036 Objective Loss 0.275036 LR 0.001000 Time 0.019610 -2022-12-06 11:04:44,607 - Epoch: [94][ 1180/ 1200] Overall Loss 0.275118 Objective Loss 0.275118 LR 0.001000 Time 0.019605 -2022-12-06 11:04:44,799 - Epoch: [94][ 1190/ 1200] Overall Loss 0.275305 Objective Loss 0.275305 LR 0.001000 Time 0.019601 -2022-12-06 11:04:45,019 - Epoch: [94][ 1200/ 1200] Overall Loss 0.275416 Objective Loss 0.275416 Top1 85.983264 Top5 98.326360 LR 0.001000 Time 0.019621 -2022-12-06 11:04:45,107 - --- validate (epoch=94)----------- -2022-12-06 11:04:45,108 - 34129 samples (256 per mini-batch) -2022-12-06 11:04:45,550 - Epoch: [94][ 10/ 134] Loss 0.293573 Top1 83.710938 Top5 97.265625 -2022-12-06 11:04:45,683 - Epoch: [94][ 20/ 134] Loss 0.297221 Top1 83.457031 Top5 97.519531 -2022-12-06 11:04:45,815 - Epoch: [94][ 30/ 134] Loss 0.307595 Top1 83.307292 Top5 97.617188 -2022-12-06 11:04:45,947 - Epoch: [94][ 40/ 134] Loss 0.305301 Top1 83.203125 Top5 97.685547 -2022-12-06 11:04:46,078 - Epoch: [94][ 50/ 134] Loss 0.304072 Top1 83.343750 Top5 97.687500 -2022-12-06 11:04:46,214 - Epoch: [94][ 60/ 134] Loss 0.302343 Top1 83.333333 Top5 97.682292 -2022-12-06 11:04:46,357 - Epoch: [94][ 70/ 134] Loss 0.303674 Top1 83.164062 Top5 97.639509 -2022-12-06 11:04:46,503 - Epoch: [94][ 80/ 134] Loss 0.306865 Top1 83.149414 Top5 97.685547 -2022-12-06 11:04:46,647 - Epoch: [94][ 90/ 134] Loss 0.306363 Top1 83.233507 Top5 97.634549 -2022-12-06 11:04:46,792 - Epoch: [94][ 100/ 134] Loss 0.307276 Top1 83.097656 Top5 97.566406 -2022-12-06 11:04:46,936 - Epoch: [94][ 110/ 134] Loss 0.306623 Top1 83.011364 Top5 97.524858 -2022-12-06 11:04:47,082 - Epoch: [94][ 120/ 134] Loss 0.304631 Top1 83.079427 Top5 97.568359 -2022-12-06 11:04:47,218 - Epoch: [94][ 130/ 134] Loss 0.305302 Top1 83.155048 Top5 97.602163 -2022-12-06 11:04:47,255 - Epoch: [94][ 134/ 134] Loss 0.305650 Top1 83.163878 Top5 97.612002 -2022-12-06 11:04:47,342 - ==> Top1: 83.164 Top5: 97.612 Loss: 0.306 - -2022-12-06 11:04:47,343 - ==> Confusion: -[[ 893 0 3 1 3 6 0 0 6 65 0 1 2 4 6 1 0 1 0 0 4] - [ 3 908 1 4 10 35 2 14 3 1 4 8 2 2 2 2 6 1 18 0 1] - [ 9 3 983 15 4 2 22 10 2 5 4 6 2 1 5 6 3 3 6 5 7] - [ 4 1 13 938 0 2 0 0 0 1 9 2 5 2 21 1 1 4 12 0 4] - [ 19 6 1 0 944 6 0 2 1 6 4 1 0 2 11 6 5 2 1 1 2] - [ 4 9 1 3 10 965 2 17 4 2 1 11 5 13 1 0 6 2 4 5 4] - [ 1 2 11 2 1 4 1057 5 0 3 1 3 2 0 0 5 2 1 1 15 2] - [ 2 3 3 2 2 32 3 932 1 0 3 7 1 5 0 0 2 1 37 15 3] - [ 9 2 0 0 0 3 0 0 971 38 5 5 2 7 16 0 2 0 2 1 1] - [ 62 0 0 0 4 1 0 1 19 891 1 2 0 10 5 1 0 1 0 0 3] - [ 3 1 2 8 2 2 2 4 20 2 928 1 2 19 4 0 2 0 12 3 2] - [ 5 1 2 1 0 7 1 3 2 1 1 962 27 7 0 4 3 6 1 12 5] - [ 2 0 1 0 0 2 0 2 1 0 0 39 884 2 1 9 3 12 0 6 5] - [ 1 0 0 1 0 5 0 3 15 19 6 2 3 946 2 0 4 1 1 6 8] - [ 14 3 0 5 3 1 0 0 11 7 0 3 1 3 1062 0 1 2 5 3 6] - [ 2 0 2 3 2 1 1 0 1 0 0 8 8 2 0 987 6 13 0 5 2] - [ 3 3 0 2 3 1 1 0 1 2 0 3 2 2 1 8 1028 0 1 8 3] - [ 2 0 2 3 0 0 0 0 2 4 1 20 8 1 0 10 1 979 0 1 2] - [ 4 4 2 21 0 5 0 24 3 0 4 2 4 1 14 0 1 0 915 2 2] - [ 4 3 0 0 1 6 5 6 0 0 1 15 6 5 1 3 7 2 1 1009 5] - [ 200 192 161 157 129 210 69 178 130 138 179 140 413 411 231 120 258 91 245 379 9195]] - -2022-12-06 11:04:48,008 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:04:48,008 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:04:48,014 - - -2022-12-06 11:04:48,014 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:04:48,942 - Epoch: [95][ 10/ 1200] Overall Loss 0.277902 Objective Loss 0.277902 LR 0.001000 Time 0.092694 -2022-12-06 11:04:49,144 - Epoch: [95][ 20/ 1200] Overall Loss 0.282757 Objective Loss 0.282757 LR 0.001000 Time 0.056403 -2022-12-06 11:04:49,340 - Epoch: [95][ 30/ 1200] Overall Loss 0.283356 Objective Loss 0.283356 LR 0.001000 Time 0.044142 -2022-12-06 11:04:49,539 - Epoch: [95][ 40/ 1200] Overall Loss 0.277958 Objective Loss 0.277958 LR 0.001000 Time 0.038059 -2022-12-06 11:04:49,736 - Epoch: [95][ 50/ 1200] Overall Loss 0.275258 Objective Loss 0.275258 LR 0.001000 Time 0.034367 -2022-12-06 11:04:49,934 - Epoch: [95][ 60/ 1200] Overall Loss 0.278136 Objective Loss 0.278136 LR 0.001000 Time 0.031940 -2022-12-06 11:04:50,130 - Epoch: [95][ 70/ 1200] Overall Loss 0.278811 Objective Loss 0.278811 LR 0.001000 Time 0.030167 -2022-12-06 11:04:50,329 - Epoch: [95][ 80/ 1200] Overall Loss 0.277352 Objective Loss 0.277352 LR 0.001000 Time 0.028870 -2022-12-06 11:04:50,525 - Epoch: [95][ 90/ 1200] Overall Loss 0.280417 Objective Loss 0.280417 LR 0.001000 Time 0.027834 -2022-12-06 11:04:50,723 - Epoch: [95][ 100/ 1200] Overall Loss 0.279785 Objective Loss 0.279785 LR 0.001000 Time 0.027026 -2022-12-06 11:04:50,918 - Epoch: [95][ 110/ 1200] Overall Loss 0.277946 Objective Loss 0.277946 LR 0.001000 Time 0.026342 -2022-12-06 11:04:51,117 - Epoch: [95][ 120/ 1200] Overall Loss 0.277510 Objective Loss 0.277510 LR 0.001000 Time 0.025799 -2022-12-06 11:04:51,313 - Epoch: [95][ 130/ 1200] Overall Loss 0.278653 Objective Loss 0.278653 LR 0.001000 Time 0.025321 -2022-12-06 11:04:51,511 - Epoch: [95][ 140/ 1200] Overall Loss 0.279001 Objective Loss 0.279001 LR 0.001000 Time 0.024923 -2022-12-06 11:04:51,708 - Epoch: [95][ 150/ 1200] Overall Loss 0.277987 Objective Loss 0.277987 LR 0.001000 Time 0.024568 -2022-12-06 11:04:51,907 - Epoch: [95][ 160/ 1200] Overall Loss 0.276322 Objective Loss 0.276322 LR 0.001000 Time 0.024272 -2022-12-06 11:04:52,104 - Epoch: [95][ 170/ 1200] Overall Loss 0.277714 Objective Loss 0.277714 LR 0.001000 Time 0.023998 -2022-12-06 11:04:52,302 - Epoch: [95][ 180/ 1200] Overall Loss 0.277141 Objective Loss 0.277141 LR 0.001000 Time 0.023766 -2022-12-06 11:04:52,499 - Epoch: [95][ 190/ 1200] Overall Loss 0.275983 Objective Loss 0.275983 LR 0.001000 Time 0.023545 -2022-12-06 11:04:52,697 - Epoch: [95][ 200/ 1200] Overall Loss 0.274289 Objective Loss 0.274289 LR 0.001000 Time 0.023357 -2022-12-06 11:04:52,893 - Epoch: [95][ 210/ 1200] Overall Loss 0.273105 Objective Loss 0.273105 LR 0.001000 Time 0.023177 -2022-12-06 11:04:53,091 - Epoch: [95][ 220/ 1200] Overall Loss 0.272132 Objective Loss 0.272132 LR 0.001000 Time 0.023021 -2022-12-06 11:04:53,287 - Epoch: [95][ 230/ 1200] Overall Loss 0.271885 Objective Loss 0.271885 LR 0.001000 Time 0.022870 -2022-12-06 11:04:53,486 - Epoch: [95][ 240/ 1200] Overall Loss 0.271826 Objective Loss 0.271826 LR 0.001000 Time 0.022743 -2022-12-06 11:04:53,683 - Epoch: [95][ 250/ 1200] Overall Loss 0.270826 Objective Loss 0.270826 LR 0.001000 Time 0.022617 -2022-12-06 11:04:53,881 - Epoch: [95][ 260/ 1200] Overall Loss 0.272388 Objective Loss 0.272388 LR 0.001000 Time 0.022507 -2022-12-06 11:04:54,077 - Epoch: [95][ 270/ 1200] Overall Loss 0.273017 Objective Loss 0.273017 LR 0.001000 Time 0.022398 -2022-12-06 11:04:54,275 - Epoch: [95][ 280/ 1200] Overall Loss 0.273415 Objective Loss 0.273415 LR 0.001000 Time 0.022303 -2022-12-06 11:04:54,471 - Epoch: [95][ 290/ 1200] Overall Loss 0.273146 Objective Loss 0.273146 LR 0.001000 Time 0.022209 -2022-12-06 11:04:54,670 - Epoch: [95][ 300/ 1200] Overall Loss 0.273897 Objective Loss 0.273897 LR 0.001000 Time 0.022129 -2022-12-06 11:04:54,867 - Epoch: [95][ 310/ 1200] Overall Loss 0.274639 Objective Loss 0.274639 LR 0.001000 Time 0.022048 -2022-12-06 11:04:55,065 - Epoch: [95][ 320/ 1200] Overall Loss 0.274580 Objective Loss 0.274580 LR 0.001000 Time 0.021978 -2022-12-06 11:04:55,261 - Epoch: [95][ 330/ 1200] Overall Loss 0.274933 Objective Loss 0.274933 LR 0.001000 Time 0.021904 -2022-12-06 11:04:55,459 - Epoch: [95][ 340/ 1200] Overall Loss 0.274693 Objective Loss 0.274693 LR 0.001000 Time 0.021841 -2022-12-06 11:04:55,655 - Epoch: [95][ 350/ 1200] Overall Loss 0.274588 Objective Loss 0.274588 LR 0.001000 Time 0.021776 -2022-12-06 11:04:55,854 - Epoch: [95][ 360/ 1200] Overall Loss 0.275209 Objective Loss 0.275209 LR 0.001000 Time 0.021721 -2022-12-06 11:04:56,050 - Epoch: [95][ 370/ 1200] Overall Loss 0.275019 Objective Loss 0.275019 LR 0.001000 Time 0.021663 -2022-12-06 11:04:56,249 - Epoch: [95][ 380/ 1200] Overall Loss 0.275447 Objective Loss 0.275447 LR 0.001000 Time 0.021614 -2022-12-06 11:04:56,445 - Epoch: [95][ 390/ 1200] Overall Loss 0.274903 Objective Loss 0.274903 LR 0.001000 Time 0.021562 -2022-12-06 11:04:56,644 - Epoch: [95][ 400/ 1200] Overall Loss 0.274657 Objective Loss 0.274657 LR 0.001000 Time 0.021518 -2022-12-06 11:04:56,840 - Epoch: [95][ 410/ 1200] Overall Loss 0.274164 Objective Loss 0.274164 LR 0.001000 Time 0.021470 -2022-12-06 11:04:57,038 - Epoch: [95][ 420/ 1200] Overall Loss 0.274630 Objective Loss 0.274630 LR 0.001000 Time 0.021429 -2022-12-06 11:04:57,235 - Epoch: [95][ 430/ 1200] Overall Loss 0.274229 Objective Loss 0.274229 LR 0.001000 Time 0.021387 -2022-12-06 11:04:57,433 - Epoch: [95][ 440/ 1200] Overall Loss 0.273844 Objective Loss 0.273844 LR 0.001000 Time 0.021350 -2022-12-06 11:04:57,629 - Epoch: [95][ 450/ 1200] Overall Loss 0.274436 Objective Loss 0.274436 LR 0.001000 Time 0.021311 -2022-12-06 11:04:57,828 - Epoch: [95][ 460/ 1200] Overall Loss 0.274665 Objective Loss 0.274665 LR 0.001000 Time 0.021278 -2022-12-06 11:04:58,024 - Epoch: [95][ 470/ 1200] Overall Loss 0.274835 Objective Loss 0.274835 LR 0.001000 Time 0.021242 -2022-12-06 11:04:58,223 - Epoch: [95][ 480/ 1200] Overall Loss 0.275543 Objective Loss 0.275543 LR 0.001000 Time 0.021212 -2022-12-06 11:04:58,419 - Epoch: [95][ 490/ 1200] Overall Loss 0.275409 Objective Loss 0.275409 LR 0.001000 Time 0.021178 -2022-12-06 11:04:58,617 - Epoch: [95][ 500/ 1200] Overall Loss 0.276312 Objective Loss 0.276312 LR 0.001000 Time 0.021149 -2022-12-06 11:04:58,813 - Epoch: [95][ 510/ 1200] Overall Loss 0.276333 Objective Loss 0.276333 LR 0.001000 Time 0.021119 -2022-12-06 11:04:59,012 - Epoch: [95][ 520/ 1200] Overall Loss 0.276389 Objective Loss 0.276389 LR 0.001000 Time 0.021095 -2022-12-06 11:04:59,209 - Epoch: [95][ 530/ 1200] Overall Loss 0.275997 Objective Loss 0.275997 LR 0.001000 Time 0.021066 -2022-12-06 11:04:59,408 - Epoch: [95][ 540/ 1200] Overall Loss 0.276203 Objective Loss 0.276203 LR 0.001000 Time 0.021044 -2022-12-06 11:04:59,604 - Epoch: [95][ 550/ 1200] Overall Loss 0.276060 Objective Loss 0.276060 LR 0.001000 Time 0.021017 -2022-12-06 11:04:59,803 - Epoch: [95][ 560/ 1200] Overall Loss 0.275758 Objective Loss 0.275758 LR 0.001000 Time 0.020996 -2022-12-06 11:05:00,000 - Epoch: [95][ 570/ 1200] Overall Loss 0.275258 Objective Loss 0.275258 LR 0.001000 Time 0.020971 -2022-12-06 11:05:00,199 - Epoch: [95][ 580/ 1200] Overall Loss 0.274950 Objective Loss 0.274950 LR 0.001000 Time 0.020952 -2022-12-06 11:05:00,395 - Epoch: [95][ 590/ 1200] Overall Loss 0.275612 Objective Loss 0.275612 LR 0.001000 Time 0.020929 -2022-12-06 11:05:00,594 - Epoch: [95][ 600/ 1200] Overall Loss 0.275428 Objective Loss 0.275428 LR 0.001000 Time 0.020911 -2022-12-06 11:05:00,791 - Epoch: [95][ 610/ 1200] Overall Loss 0.275418 Objective Loss 0.275418 LR 0.001000 Time 0.020890 -2022-12-06 11:05:00,990 - Epoch: [95][ 620/ 1200] Overall Loss 0.275686 Objective Loss 0.275686 LR 0.001000 Time 0.020872 -2022-12-06 11:05:01,186 - Epoch: [95][ 630/ 1200] Overall Loss 0.275936 Objective Loss 0.275936 LR 0.001000 Time 0.020852 -2022-12-06 11:05:01,385 - Epoch: [95][ 640/ 1200] Overall Loss 0.275927 Objective Loss 0.275927 LR 0.001000 Time 0.020836 -2022-12-06 11:05:01,581 - Epoch: [95][ 650/ 1200] Overall Loss 0.275981 Objective Loss 0.275981 LR 0.001000 Time 0.020816 -2022-12-06 11:05:01,779 - Epoch: [95][ 660/ 1200] Overall Loss 0.275772 Objective Loss 0.275772 LR 0.001000 Time 0.020801 -2022-12-06 11:05:01,976 - Epoch: [95][ 670/ 1200] Overall Loss 0.275509 Objective Loss 0.275509 LR 0.001000 Time 0.020783 -2022-12-06 11:05:02,175 - Epoch: [95][ 680/ 1200] Overall Loss 0.275485 Objective Loss 0.275485 LR 0.001000 Time 0.020769 -2022-12-06 11:05:02,371 - Epoch: [95][ 690/ 1200] Overall Loss 0.275428 Objective Loss 0.275428 LR 0.001000 Time 0.020752 -2022-12-06 11:05:02,571 - Epoch: [95][ 700/ 1200] Overall Loss 0.275552 Objective Loss 0.275552 LR 0.001000 Time 0.020739 -2022-12-06 11:05:02,767 - Epoch: [95][ 710/ 1200] Overall Loss 0.275929 Objective Loss 0.275929 LR 0.001000 Time 0.020722 -2022-12-06 11:05:02,965 - Epoch: [95][ 720/ 1200] Overall Loss 0.275815 Objective Loss 0.275815 LR 0.001000 Time 0.020710 -2022-12-06 11:05:03,162 - Epoch: [95][ 730/ 1200] Overall Loss 0.275827 Objective Loss 0.275827 LR 0.001000 Time 0.020694 -2022-12-06 11:05:03,360 - Epoch: [95][ 740/ 1200] Overall Loss 0.275703 Objective Loss 0.275703 LR 0.001000 Time 0.020682 -2022-12-06 11:05:03,556 - Epoch: [95][ 750/ 1200] Overall Loss 0.275687 Objective Loss 0.275687 LR 0.001000 Time 0.020667 -2022-12-06 11:05:03,755 - Epoch: [95][ 760/ 1200] Overall Loss 0.275596 Objective Loss 0.275596 LR 0.001000 Time 0.020656 -2022-12-06 11:05:03,951 - Epoch: [95][ 770/ 1200] Overall Loss 0.275443 Objective Loss 0.275443 LR 0.001000 Time 0.020641 -2022-12-06 11:05:04,149 - Epoch: [95][ 780/ 1200] Overall Loss 0.275536 Objective Loss 0.275536 LR 0.001000 Time 0.020631 -2022-12-06 11:05:04,346 - Epoch: [95][ 790/ 1200] Overall Loss 0.275586 Objective Loss 0.275586 LR 0.001000 Time 0.020617 -2022-12-06 11:05:04,544 - Epoch: [95][ 800/ 1200] Overall Loss 0.275626 Objective Loss 0.275626 LR 0.001000 Time 0.020607 -2022-12-06 11:05:04,740 - Epoch: [95][ 810/ 1200] Overall Loss 0.275524 Objective Loss 0.275524 LR 0.001000 Time 0.020594 -2022-12-06 11:05:04,938 - Epoch: [95][ 820/ 1200] Overall Loss 0.275548 Objective Loss 0.275548 LR 0.001000 Time 0.020584 -2022-12-06 11:05:05,134 - Epoch: [95][ 830/ 1200] Overall Loss 0.275461 Objective Loss 0.275461 LR 0.001000 Time 0.020571 -2022-12-06 11:05:05,332 - Epoch: [95][ 840/ 1200] Overall Loss 0.275264 Objective Loss 0.275264 LR 0.001000 Time 0.020561 -2022-12-06 11:05:05,528 - Epoch: [95][ 850/ 1200] Overall Loss 0.275444 Objective Loss 0.275444 LR 0.001000 Time 0.020549 -2022-12-06 11:05:05,727 - Epoch: [95][ 860/ 1200] Overall Loss 0.275360 Objective Loss 0.275360 LR 0.001000 Time 0.020541 -2022-12-06 11:05:05,923 - Epoch: [95][ 870/ 1200] Overall Loss 0.275034 Objective Loss 0.275034 LR 0.001000 Time 0.020530 -2022-12-06 11:05:06,123 - Epoch: [95][ 880/ 1200] Overall Loss 0.275232 Objective Loss 0.275232 LR 0.001000 Time 0.020522 -2022-12-06 11:05:06,319 - Epoch: [95][ 890/ 1200] Overall Loss 0.275157 Objective Loss 0.275157 LR 0.001000 Time 0.020511 -2022-12-06 11:05:06,517 - Epoch: [95][ 900/ 1200] Overall Loss 0.275325 Objective Loss 0.275325 LR 0.001000 Time 0.020503 -2022-12-06 11:05:06,713 - Epoch: [95][ 910/ 1200] Overall Loss 0.275261 Objective Loss 0.275261 LR 0.001000 Time 0.020493 -2022-12-06 11:05:06,911 - Epoch: [95][ 920/ 1200] Overall Loss 0.275575 Objective Loss 0.275575 LR 0.001000 Time 0.020485 -2022-12-06 11:05:07,108 - Epoch: [95][ 930/ 1200] Overall Loss 0.275321 Objective Loss 0.275321 LR 0.001000 Time 0.020475 -2022-12-06 11:05:07,306 - Epoch: [95][ 940/ 1200] Overall Loss 0.275079 Objective Loss 0.275079 LR 0.001000 Time 0.020468 -2022-12-06 11:05:07,503 - Epoch: [95][ 950/ 1200] Overall Loss 0.275086 Objective Loss 0.275086 LR 0.001000 Time 0.020459 -2022-12-06 11:05:07,702 - Epoch: [95][ 960/ 1200] Overall Loss 0.274863 Objective Loss 0.274863 LR 0.001000 Time 0.020452 -2022-12-06 11:05:07,898 - Epoch: [95][ 970/ 1200] Overall Loss 0.274995 Objective Loss 0.274995 LR 0.001000 Time 0.020443 -2022-12-06 11:05:08,097 - Epoch: [95][ 980/ 1200] Overall Loss 0.275001 Objective Loss 0.275001 LR 0.001000 Time 0.020437 -2022-12-06 11:05:08,293 - Epoch: [95][ 990/ 1200] Overall Loss 0.275158 Objective Loss 0.275158 LR 0.001000 Time 0.020428 -2022-12-06 11:05:08,491 - Epoch: [95][ 1000/ 1200] Overall Loss 0.275032 Objective Loss 0.275032 LR 0.001000 Time 0.020421 -2022-12-06 11:05:08,688 - Epoch: [95][ 1010/ 1200] Overall Loss 0.275232 Objective Loss 0.275232 LR 0.001000 Time 0.020413 -2022-12-06 11:05:08,886 - Epoch: [95][ 1020/ 1200] Overall Loss 0.275043 Objective Loss 0.275043 LR 0.001000 Time 0.020407 -2022-12-06 11:05:09,083 - Epoch: [95][ 1030/ 1200] Overall Loss 0.275002 Objective Loss 0.275002 LR 0.001000 Time 0.020399 -2022-12-06 11:05:09,281 - Epoch: [95][ 1040/ 1200] Overall Loss 0.274903 Objective Loss 0.274903 LR 0.001000 Time 0.020393 -2022-12-06 11:05:09,478 - Epoch: [95][ 1050/ 1200] Overall Loss 0.274713 Objective Loss 0.274713 LR 0.001000 Time 0.020386 -2022-12-06 11:05:09,677 - Epoch: [95][ 1060/ 1200] Overall Loss 0.274633 Objective Loss 0.274633 LR 0.001000 Time 0.020381 -2022-12-06 11:05:09,873 - Epoch: [95][ 1070/ 1200] Overall Loss 0.274656 Objective Loss 0.274656 LR 0.001000 Time 0.020373 -2022-12-06 11:05:10,071 - Epoch: [95][ 1080/ 1200] Overall Loss 0.275011 Objective Loss 0.275011 LR 0.001000 Time 0.020368 -2022-12-06 11:05:10,267 - Epoch: [95][ 1090/ 1200] Overall Loss 0.274870 Objective Loss 0.274870 LR 0.001000 Time 0.020360 -2022-12-06 11:05:10,466 - Epoch: [95][ 1100/ 1200] Overall Loss 0.274934 Objective Loss 0.274934 LR 0.001000 Time 0.020355 -2022-12-06 11:05:10,662 - Epoch: [95][ 1110/ 1200] Overall Loss 0.274915 Objective Loss 0.274915 LR 0.001000 Time 0.020348 -2022-12-06 11:05:10,861 - Epoch: [95][ 1120/ 1200] Overall Loss 0.275408 Objective Loss 0.275408 LR 0.001000 Time 0.020343 -2022-12-06 11:05:11,057 - Epoch: [95][ 1130/ 1200] Overall Loss 0.275200 Objective Loss 0.275200 LR 0.001000 Time 0.020336 -2022-12-06 11:05:11,255 - Epoch: [95][ 1140/ 1200] Overall Loss 0.275145 Objective Loss 0.275145 LR 0.001000 Time 0.020332 -2022-12-06 11:05:11,451 - Epoch: [95][ 1150/ 1200] Overall Loss 0.275047 Objective Loss 0.275047 LR 0.001000 Time 0.020324 -2022-12-06 11:05:11,649 - Epoch: [95][ 1160/ 1200] Overall Loss 0.275054 Objective Loss 0.275054 LR 0.001000 Time 0.020320 -2022-12-06 11:05:11,845 - Epoch: [95][ 1170/ 1200] Overall Loss 0.274812 Objective Loss 0.274812 LR 0.001000 Time 0.020313 -2022-12-06 11:05:12,044 - Epoch: [95][ 1180/ 1200] Overall Loss 0.274907 Objective Loss 0.274907 LR 0.001000 Time 0.020309 -2022-12-06 11:05:12,240 - Epoch: [95][ 1190/ 1200] Overall Loss 0.274894 Objective Loss 0.274894 LR 0.001000 Time 0.020302 -2022-12-06 11:05:12,472 - Epoch: [95][ 1200/ 1200] Overall Loss 0.274726 Objective Loss 0.274726 Top1 87.238494 Top5 98.535565 LR 0.001000 Time 0.020326 -2022-12-06 11:05:12,561 - --- validate (epoch=95)----------- -2022-12-06 11:05:12,561 - 34129 samples (256 per mini-batch) -2022-12-06 11:05:13,012 - Epoch: [95][ 10/ 134] Loss 0.313202 Top1 84.804688 Top5 97.656250 -2022-12-06 11:05:13,144 - Epoch: [95][ 20/ 134] Loss 0.295595 Top1 85.253906 Top5 97.890625 -2022-12-06 11:05:13,276 - Epoch: [95][ 30/ 134] Loss 0.290712 Top1 85.494792 Top5 97.994792 -2022-12-06 11:05:13,404 - Epoch: [95][ 40/ 134] Loss 0.300631 Top1 85.214844 Top5 97.958984 -2022-12-06 11:05:13,535 - Epoch: [95][ 50/ 134] Loss 0.293940 Top1 85.195312 Top5 97.992188 -2022-12-06 11:05:13,664 - Epoch: [95][ 60/ 134] Loss 0.289236 Top1 85.305990 Top5 98.033854 -2022-12-06 11:05:13,795 - Epoch: [95][ 70/ 134] Loss 0.292629 Top1 85.267857 Top5 98.052455 -2022-12-06 11:05:13,924 - Epoch: [95][ 80/ 134] Loss 0.294710 Top1 85.224609 Top5 97.998047 -2022-12-06 11:05:14,055 - Epoch: [95][ 90/ 134] Loss 0.297191 Top1 85.130208 Top5 98.033854 -2022-12-06 11:05:14,185 - Epoch: [95][ 100/ 134] Loss 0.298052 Top1 85.101562 Top5 98.039062 -2022-12-06 11:05:14,316 - Epoch: [95][ 110/ 134] Loss 0.297137 Top1 85.223722 Top5 98.057528 -2022-12-06 11:05:14,448 - Epoch: [95][ 120/ 134] Loss 0.296277 Top1 85.211589 Top5 98.056641 -2022-12-06 11:05:14,581 - Epoch: [95][ 130/ 134] Loss 0.296215 Top1 85.291466 Top5 98.064904 -2022-12-06 11:05:14,620 - Epoch: [95][ 134/ 134] Loss 0.296343 Top1 85.264731 Top5 98.051510 -2022-12-06 11:05:14,709 - ==> Top1: 85.265 Top5: 98.052 Loss: 0.296 - -2022-12-06 11:05:14,710 - ==> Confusion: -[[ 919 2 1 3 1 5 0 2 8 36 0 1 3 2 4 1 1 0 2 1 4] - [ 3 898 2 3 6 38 2 19 1 2 3 4 4 3 1 3 5 1 19 3 7] - [ 10 4 999 15 1 2 21 9 2 1 3 5 2 3 1 3 2 3 4 2 11] - [ 5 1 26 915 1 4 0 1 1 0 7 0 5 1 19 1 2 4 15 0 12] - [ 12 2 2 0 951 8 0 3 1 6 1 4 0 4 9 4 6 1 0 1 5] - [ 4 10 1 4 5 975 2 19 2 1 1 12 3 15 1 1 2 0 1 7 3] - [ 0 1 9 4 0 2 1074 4 0 0 1 1 2 1 0 4 1 1 2 7 4] - [ 2 5 7 2 2 34 9 935 0 1 3 5 4 0 1 1 0 1 25 14 3] - [ 7 6 0 0 1 5 0 0 954 43 9 2 2 14 12 1 1 1 4 2 0] - [ 73 0 1 1 8 2 1 2 23 858 1 2 2 13 3 1 0 1 0 1 8] - [ 2 3 10 2 3 1 0 4 11 2 942 3 2 11 2 0 0 1 8 1 11] - [ 5 1 1 1 0 12 1 2 0 0 1 977 22 4 0 6 4 3 0 5 6] - [ 0 1 1 3 0 3 0 1 0 0 0 40 880 2 1 7 1 10 0 6 13] - [ 2 0 1 0 0 9 0 2 7 9 10 7 3 961 0 1 2 0 0 4 5] - [ 12 4 2 6 4 1 0 0 25 2 2 2 1 4 1044 1 0 1 9 1 9] - [ 2 0 4 2 4 0 3 0 0 0 0 15 5 3 0 973 13 8 1 6 4] - [ 5 0 2 1 3 1 1 1 2 0 0 3 0 2 3 10 1024 0 1 5 8] - [ 2 0 1 2 0 0 1 0 0 3 1 16 26 3 1 10 0 964 0 3 3] - [ 4 7 7 12 0 1 1 19 2 0 3 5 3 1 7 0 2 2 927 1 4] - [ 3 2 0 1 0 3 7 11 0 0 2 17 7 5 1 3 5 5 1 1000 7] - [ 150 181 197 100 93 229 91 164 76 99 128 131 343 340 142 99 266 71 159 246 9921]] - -2022-12-06 11:05:15,368 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:05:15,368 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:05:15,374 - - -2022-12-06 11:05:15,374 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:05:16,304 - Epoch: [96][ 10/ 1200] Overall Loss 0.273410 Objective Loss 0.273410 LR 0.001000 Time 0.092965 -2022-12-06 11:05:16,507 - Epoch: [96][ 20/ 1200] Overall Loss 0.261762 Objective Loss 0.261762 LR 0.001000 Time 0.056595 -2022-12-06 11:05:16,699 - Epoch: [96][ 30/ 1200] Overall Loss 0.259440 Objective Loss 0.259440 LR 0.001000 Time 0.044096 -2022-12-06 11:05:16,891 - Epoch: [96][ 40/ 1200] Overall Loss 0.268145 Objective Loss 0.268145 LR 0.001000 Time 0.037863 -2022-12-06 11:05:17,083 - Epoch: [96][ 50/ 1200] Overall Loss 0.270145 Objective Loss 0.270145 LR 0.001000 Time 0.034123 -2022-12-06 11:05:17,275 - Epoch: [96][ 60/ 1200] Overall Loss 0.271408 Objective Loss 0.271408 LR 0.001000 Time 0.031630 -2022-12-06 11:05:17,467 - Epoch: [96][ 70/ 1200] Overall Loss 0.273530 Objective Loss 0.273530 LR 0.001000 Time 0.029849 -2022-12-06 11:05:17,659 - Epoch: [96][ 80/ 1200] Overall Loss 0.276503 Objective Loss 0.276503 LR 0.001000 Time 0.028503 -2022-12-06 11:05:17,850 - Epoch: [96][ 90/ 1200] Overall Loss 0.273906 Objective Loss 0.273906 LR 0.001000 Time 0.027458 -2022-12-06 11:05:18,043 - Epoch: [96][ 100/ 1200] Overall Loss 0.274221 Objective Loss 0.274221 LR 0.001000 Time 0.026630 -2022-12-06 11:05:18,234 - Epoch: [96][ 110/ 1200] Overall Loss 0.272704 Objective Loss 0.272704 LR 0.001000 Time 0.025945 -2022-12-06 11:05:18,427 - Epoch: [96][ 120/ 1200] Overall Loss 0.272462 Objective Loss 0.272462 LR 0.001000 Time 0.025381 -2022-12-06 11:05:18,618 - Epoch: [96][ 130/ 1200] Overall Loss 0.270470 Objective Loss 0.270470 LR 0.001000 Time 0.024896 -2022-12-06 11:05:18,810 - Epoch: [96][ 140/ 1200] Overall Loss 0.271236 Objective Loss 0.271236 LR 0.001000 Time 0.024489 -2022-12-06 11:05:19,003 - Epoch: [96][ 150/ 1200] Overall Loss 0.269901 Objective Loss 0.269901 LR 0.001000 Time 0.024134 -2022-12-06 11:05:19,194 - Epoch: [96][ 160/ 1200] Overall Loss 0.270672 Objective Loss 0.270672 LR 0.001000 Time 0.023820 -2022-12-06 11:05:19,386 - Epoch: [96][ 170/ 1200] Overall Loss 0.270305 Objective Loss 0.270305 LR 0.001000 Time 0.023546 -2022-12-06 11:05:19,579 - Epoch: [96][ 180/ 1200] Overall Loss 0.269660 Objective Loss 0.269660 LR 0.001000 Time 0.023305 -2022-12-06 11:05:19,771 - Epoch: [96][ 190/ 1200] Overall Loss 0.268575 Objective Loss 0.268575 LR 0.001000 Time 0.023085 -2022-12-06 11:05:19,963 - Epoch: [96][ 200/ 1200] Overall Loss 0.268515 Objective Loss 0.268515 LR 0.001000 Time 0.022890 -2022-12-06 11:05:20,156 - Epoch: [96][ 210/ 1200] Overall Loss 0.267564 Objective Loss 0.267564 LR 0.001000 Time 0.022714 -2022-12-06 11:05:20,347 - Epoch: [96][ 220/ 1200] Overall Loss 0.266693 Objective Loss 0.266693 LR 0.001000 Time 0.022551 -2022-12-06 11:05:20,540 - Epoch: [96][ 230/ 1200] Overall Loss 0.267099 Objective Loss 0.267099 LR 0.001000 Time 0.022403 -2022-12-06 11:05:20,732 - Epoch: [96][ 240/ 1200] Overall Loss 0.267314 Objective Loss 0.267314 LR 0.001000 Time 0.022268 -2022-12-06 11:05:20,924 - Epoch: [96][ 250/ 1200] Overall Loss 0.267123 Objective Loss 0.267123 LR 0.001000 Time 0.022144 -2022-12-06 11:05:21,116 - Epoch: [96][ 260/ 1200] Overall Loss 0.266393 Objective Loss 0.266393 LR 0.001000 Time 0.022029 -2022-12-06 11:05:21,308 - Epoch: [96][ 270/ 1200] Overall Loss 0.266519 Objective Loss 0.266519 LR 0.001000 Time 0.021922 -2022-12-06 11:05:21,500 - Epoch: [96][ 280/ 1200] Overall Loss 0.267249 Objective Loss 0.267249 LR 0.001000 Time 0.021823 -2022-12-06 11:05:21,692 - Epoch: [96][ 290/ 1200] Overall Loss 0.266532 Objective Loss 0.266532 LR 0.001000 Time 0.021730 -2022-12-06 11:05:21,884 - Epoch: [96][ 300/ 1200] Overall Loss 0.266840 Objective Loss 0.266840 LR 0.001000 Time 0.021644 -2022-12-06 11:05:22,076 - Epoch: [96][ 310/ 1200] Overall Loss 0.266686 Objective Loss 0.266686 LR 0.001000 Time 0.021564 -2022-12-06 11:05:22,268 - Epoch: [96][ 320/ 1200] Overall Loss 0.266521 Objective Loss 0.266521 LR 0.001000 Time 0.021489 -2022-12-06 11:05:22,460 - Epoch: [96][ 330/ 1200] Overall Loss 0.266262 Objective Loss 0.266262 LR 0.001000 Time 0.021418 -2022-12-06 11:05:22,652 - Epoch: [96][ 340/ 1200] Overall Loss 0.267341 Objective Loss 0.267341 LR 0.001000 Time 0.021352 -2022-12-06 11:05:22,844 - Epoch: [96][ 350/ 1200] Overall Loss 0.267268 Objective Loss 0.267268 LR 0.001000 Time 0.021289 -2022-12-06 11:05:23,036 - Epoch: [96][ 360/ 1200] Overall Loss 0.268073 Objective Loss 0.268073 LR 0.001000 Time 0.021228 -2022-12-06 11:05:23,228 - Epoch: [96][ 370/ 1200] Overall Loss 0.268428 Objective Loss 0.268428 LR 0.001000 Time 0.021171 -2022-12-06 11:05:23,420 - Epoch: [96][ 380/ 1200] Overall Loss 0.268397 Objective Loss 0.268397 LR 0.001000 Time 0.021117 -2022-12-06 11:05:23,611 - Epoch: [96][ 390/ 1200] Overall Loss 0.269070 Objective Loss 0.269070 LR 0.001000 Time 0.021066 -2022-12-06 11:05:23,803 - Epoch: [96][ 400/ 1200] Overall Loss 0.268394 Objective Loss 0.268394 LR 0.001000 Time 0.021017 -2022-12-06 11:05:23,995 - Epoch: [96][ 410/ 1200] Overall Loss 0.268738 Objective Loss 0.268738 LR 0.001000 Time 0.020971 -2022-12-06 11:05:24,187 - Epoch: [96][ 420/ 1200] Overall Loss 0.268665 Objective Loss 0.268665 LR 0.001000 Time 0.020929 -2022-12-06 11:05:24,380 - Epoch: [96][ 430/ 1200] Overall Loss 0.268836 Objective Loss 0.268836 LR 0.001000 Time 0.020888 -2022-12-06 11:05:24,572 - Epoch: [96][ 440/ 1200] Overall Loss 0.269417 Objective Loss 0.269417 LR 0.001000 Time 0.020849 -2022-12-06 11:05:24,764 - Epoch: [96][ 450/ 1200] Overall Loss 0.269351 Objective Loss 0.269351 LR 0.001000 Time 0.020812 -2022-12-06 11:05:24,956 - Epoch: [96][ 460/ 1200] Overall Loss 0.269715 Objective Loss 0.269715 LR 0.001000 Time 0.020775 -2022-12-06 11:05:25,148 - Epoch: [96][ 470/ 1200] Overall Loss 0.270132 Objective Loss 0.270132 LR 0.001000 Time 0.020740 -2022-12-06 11:05:25,339 - Epoch: [96][ 480/ 1200] Overall Loss 0.271173 Objective Loss 0.271173 LR 0.001000 Time 0.020706 -2022-12-06 11:05:25,532 - Epoch: [96][ 490/ 1200] Overall Loss 0.271889 Objective Loss 0.271889 LR 0.001000 Time 0.020674 -2022-12-06 11:05:25,724 - Epoch: [96][ 500/ 1200] Overall Loss 0.272650 Objective Loss 0.272650 LR 0.001000 Time 0.020644 -2022-12-06 11:05:25,916 - Epoch: [96][ 510/ 1200] Overall Loss 0.273383 Objective Loss 0.273383 LR 0.001000 Time 0.020615 -2022-12-06 11:05:26,108 - Epoch: [96][ 520/ 1200] Overall Loss 0.273478 Objective Loss 0.273478 LR 0.001000 Time 0.020587 -2022-12-06 11:05:26,299 - Epoch: [96][ 530/ 1200] Overall Loss 0.274020 Objective Loss 0.274020 LR 0.001000 Time 0.020558 -2022-12-06 11:05:26,492 - Epoch: [96][ 540/ 1200] Overall Loss 0.274571 Objective Loss 0.274571 LR 0.001000 Time 0.020533 -2022-12-06 11:05:26,684 - Epoch: [96][ 550/ 1200] Overall Loss 0.274866 Objective Loss 0.274866 LR 0.001000 Time 0.020508 -2022-12-06 11:05:26,876 - Epoch: [96][ 560/ 1200] Overall Loss 0.274616 Objective Loss 0.274616 LR 0.001000 Time 0.020484 -2022-12-06 11:05:27,068 - Epoch: [96][ 570/ 1200] Overall Loss 0.274795 Objective Loss 0.274795 LR 0.001000 Time 0.020460 -2022-12-06 11:05:27,260 - Epoch: [96][ 580/ 1200] Overall Loss 0.275136 Objective Loss 0.275136 LR 0.001000 Time 0.020438 -2022-12-06 11:05:27,452 - Epoch: [96][ 590/ 1200] Overall Loss 0.275662 Objective Loss 0.275662 LR 0.001000 Time 0.020415 -2022-12-06 11:05:27,644 - Epoch: [96][ 600/ 1200] Overall Loss 0.275411 Objective Loss 0.275411 LR 0.001000 Time 0.020394 -2022-12-06 11:05:27,836 - Epoch: [96][ 610/ 1200] Overall Loss 0.275668 Objective Loss 0.275668 LR 0.001000 Time 0.020374 -2022-12-06 11:05:28,028 - Epoch: [96][ 620/ 1200] Overall Loss 0.275660 Objective Loss 0.275660 LR 0.001000 Time 0.020354 -2022-12-06 11:05:28,220 - Epoch: [96][ 630/ 1200] Overall Loss 0.275674 Objective Loss 0.275674 LR 0.001000 Time 0.020335 -2022-12-06 11:05:28,412 - Epoch: [96][ 640/ 1200] Overall Loss 0.275708 Objective Loss 0.275708 LR 0.001000 Time 0.020316 -2022-12-06 11:05:28,603 - Epoch: [96][ 650/ 1200] Overall Loss 0.276114 Objective Loss 0.276114 LR 0.001000 Time 0.020297 -2022-12-06 11:05:28,795 - Epoch: [96][ 660/ 1200] Overall Loss 0.275885 Objective Loss 0.275885 LR 0.001000 Time 0.020279 -2022-12-06 11:05:28,987 - Epoch: [96][ 670/ 1200] Overall Loss 0.275767 Objective Loss 0.275767 LR 0.001000 Time 0.020262 -2022-12-06 11:05:29,179 - Epoch: [96][ 680/ 1200] Overall Loss 0.275916 Objective Loss 0.275916 LR 0.001000 Time 0.020246 -2022-12-06 11:05:29,370 - Epoch: [96][ 690/ 1200] Overall Loss 0.275874 Objective Loss 0.275874 LR 0.001000 Time 0.020229 -2022-12-06 11:05:29,562 - Epoch: [96][ 700/ 1200] Overall Loss 0.276208 Objective Loss 0.276208 LR 0.001000 Time 0.020214 -2022-12-06 11:05:29,754 - Epoch: [96][ 710/ 1200] Overall Loss 0.276027 Objective Loss 0.276027 LR 0.001000 Time 0.020199 -2022-12-06 11:05:29,946 - Epoch: [96][ 720/ 1200] Overall Loss 0.275250 Objective Loss 0.275250 LR 0.001000 Time 0.020184 -2022-12-06 11:05:30,138 - Epoch: [96][ 730/ 1200] Overall Loss 0.275244 Objective Loss 0.275244 LR 0.001000 Time 0.020170 -2022-12-06 11:05:30,331 - Epoch: [96][ 740/ 1200] Overall Loss 0.275298 Objective Loss 0.275298 LR 0.001000 Time 0.020156 -2022-12-06 11:05:30,522 - Epoch: [96][ 750/ 1200] Overall Loss 0.275061 Objective Loss 0.275061 LR 0.001000 Time 0.020142 -2022-12-06 11:05:30,714 - Epoch: [96][ 760/ 1200] Overall Loss 0.275463 Objective Loss 0.275463 LR 0.001000 Time 0.020130 -2022-12-06 11:05:30,906 - Epoch: [96][ 770/ 1200] Overall Loss 0.275350 Objective Loss 0.275350 LR 0.001000 Time 0.020116 -2022-12-06 11:05:31,098 - Epoch: [96][ 780/ 1200] Overall Loss 0.275735 Objective Loss 0.275735 LR 0.001000 Time 0.020104 -2022-12-06 11:05:31,291 - Epoch: [96][ 790/ 1200] Overall Loss 0.275811 Objective Loss 0.275811 LR 0.001000 Time 0.020093 -2022-12-06 11:05:31,483 - Epoch: [96][ 800/ 1200] Overall Loss 0.275668 Objective Loss 0.275668 LR 0.001000 Time 0.020081 -2022-12-06 11:05:31,676 - Epoch: [96][ 810/ 1200] Overall Loss 0.275470 Objective Loss 0.275470 LR 0.001000 Time 0.020070 -2022-12-06 11:05:31,868 - Epoch: [96][ 820/ 1200] Overall Loss 0.275504 Objective Loss 0.275504 LR 0.001000 Time 0.020059 -2022-12-06 11:05:32,060 - Epoch: [96][ 830/ 1200] Overall Loss 0.275716 Objective Loss 0.275716 LR 0.001000 Time 0.020048 -2022-12-06 11:05:32,252 - Epoch: [96][ 840/ 1200] Overall Loss 0.275463 Objective Loss 0.275463 LR 0.001000 Time 0.020038 -2022-12-06 11:05:32,444 - Epoch: [96][ 850/ 1200] Overall Loss 0.275148 Objective Loss 0.275148 LR 0.001000 Time 0.020027 -2022-12-06 11:05:32,636 - Epoch: [96][ 860/ 1200] Overall Loss 0.275387 Objective Loss 0.275387 LR 0.001000 Time 0.020017 -2022-12-06 11:05:32,829 - Epoch: [96][ 870/ 1200] Overall Loss 0.275500 Objective Loss 0.275500 LR 0.001000 Time 0.020008 -2022-12-06 11:05:33,021 - Epoch: [96][ 880/ 1200] Overall Loss 0.275553 Objective Loss 0.275553 LR 0.001000 Time 0.019998 -2022-12-06 11:05:33,213 - Epoch: [96][ 890/ 1200] Overall Loss 0.275573 Objective Loss 0.275573 LR 0.001000 Time 0.019989 -2022-12-06 11:05:33,405 - Epoch: [96][ 900/ 1200] Overall Loss 0.275645 Objective Loss 0.275645 LR 0.001000 Time 0.019980 -2022-12-06 11:05:33,598 - Epoch: [96][ 910/ 1200] Overall Loss 0.275761 Objective Loss 0.275761 LR 0.001000 Time 0.019971 -2022-12-06 11:05:33,790 - Epoch: [96][ 920/ 1200] Overall Loss 0.275756 Objective Loss 0.275756 LR 0.001000 Time 0.019963 -2022-12-06 11:05:33,983 - Epoch: [96][ 930/ 1200] Overall Loss 0.276041 Objective Loss 0.276041 LR 0.001000 Time 0.019954 -2022-12-06 11:05:34,175 - Epoch: [96][ 940/ 1200] Overall Loss 0.276223 Objective Loss 0.276223 LR 0.001000 Time 0.019946 -2022-12-06 11:05:34,367 - Epoch: [96][ 950/ 1200] Overall Loss 0.276501 Objective Loss 0.276501 LR 0.001000 Time 0.019937 -2022-12-06 11:05:34,559 - Epoch: [96][ 960/ 1200] Overall Loss 0.276778 Objective Loss 0.276778 LR 0.001000 Time 0.019929 -2022-12-06 11:05:34,751 - Epoch: [96][ 970/ 1200] Overall Loss 0.276780 Objective Loss 0.276780 LR 0.001000 Time 0.019921 -2022-12-06 11:05:34,944 - Epoch: [96][ 980/ 1200] Overall Loss 0.276963 Objective Loss 0.276963 LR 0.001000 Time 0.019914 -2022-12-06 11:05:35,136 - Epoch: [96][ 990/ 1200] Overall Loss 0.277083 Objective Loss 0.277083 LR 0.001000 Time 0.019906 -2022-12-06 11:05:35,328 - Epoch: [96][ 1000/ 1200] Overall Loss 0.277097 Objective Loss 0.277097 LR 0.001000 Time 0.019898 -2022-12-06 11:05:35,520 - Epoch: [96][ 1010/ 1200] Overall Loss 0.277139 Objective Loss 0.277139 LR 0.001000 Time 0.019891 -2022-12-06 11:05:35,713 - Epoch: [96][ 1020/ 1200] Overall Loss 0.276978 Objective Loss 0.276978 LR 0.001000 Time 0.019884 -2022-12-06 11:05:35,905 - Epoch: [96][ 1030/ 1200] Overall Loss 0.276756 Objective Loss 0.276756 LR 0.001000 Time 0.019878 -2022-12-06 11:05:36,098 - Epoch: [96][ 1040/ 1200] Overall Loss 0.276785 Objective Loss 0.276785 LR 0.001000 Time 0.019871 -2022-12-06 11:05:36,290 - Epoch: [96][ 1050/ 1200] Overall Loss 0.276648 Objective Loss 0.276648 LR 0.001000 Time 0.019865 -2022-12-06 11:05:36,483 - Epoch: [96][ 1060/ 1200] Overall Loss 0.276764 Objective Loss 0.276764 LR 0.001000 Time 0.019859 -2022-12-06 11:05:36,675 - Epoch: [96][ 1070/ 1200] Overall Loss 0.276687 Objective Loss 0.276687 LR 0.001000 Time 0.019852 -2022-12-06 11:05:36,867 - Epoch: [96][ 1080/ 1200] Overall Loss 0.276707 Objective Loss 0.276707 LR 0.001000 Time 0.019846 -2022-12-06 11:05:37,060 - Epoch: [96][ 1090/ 1200] Overall Loss 0.277160 Objective Loss 0.277160 LR 0.001000 Time 0.019840 -2022-12-06 11:05:37,252 - Epoch: [96][ 1100/ 1200] Overall Loss 0.277018 Objective Loss 0.277018 LR 0.001000 Time 0.019833 -2022-12-06 11:05:37,444 - Epoch: [96][ 1110/ 1200] Overall Loss 0.277131 Objective Loss 0.277131 LR 0.001000 Time 0.019827 -2022-12-06 11:05:37,636 - Epoch: [96][ 1120/ 1200] Overall Loss 0.277124 Objective Loss 0.277124 LR 0.001000 Time 0.019822 -2022-12-06 11:05:37,829 - Epoch: [96][ 1130/ 1200] Overall Loss 0.277637 Objective Loss 0.277637 LR 0.001000 Time 0.019816 -2022-12-06 11:05:38,021 - Epoch: [96][ 1140/ 1200] Overall Loss 0.277780 Objective Loss 0.277780 LR 0.001000 Time 0.019811 -2022-12-06 11:05:38,214 - Epoch: [96][ 1150/ 1200] Overall Loss 0.277881 Objective Loss 0.277881 LR 0.001000 Time 0.019805 -2022-12-06 11:05:38,406 - Epoch: [96][ 1160/ 1200] Overall Loss 0.277655 Objective Loss 0.277655 LR 0.001000 Time 0.019800 -2022-12-06 11:05:38,599 - Epoch: [96][ 1170/ 1200] Overall Loss 0.277700 Objective Loss 0.277700 LR 0.001000 Time 0.019795 -2022-12-06 11:05:38,791 - Epoch: [96][ 1180/ 1200] Overall Loss 0.277927 Objective Loss 0.277927 LR 0.001000 Time 0.019790 -2022-12-06 11:05:38,984 - Epoch: [96][ 1190/ 1200] Overall Loss 0.277956 Objective Loss 0.277956 LR 0.001000 Time 0.019785 -2022-12-06 11:05:39,210 - Epoch: [96][ 1200/ 1200] Overall Loss 0.278137 Objective Loss 0.278137 Top1 84.518828 Top5 97.907950 LR 0.001000 Time 0.019808 -2022-12-06 11:05:39,298 - --- validate (epoch=96)----------- -2022-12-06 11:05:39,299 - 34129 samples (256 per mini-batch) -2022-12-06 11:05:39,752 - Epoch: [96][ 10/ 134] Loss 0.297167 Top1 84.023438 Top5 97.851562 -2022-12-06 11:05:39,881 - Epoch: [96][ 20/ 134] Loss 0.312043 Top1 83.710938 Top5 97.851562 -2022-12-06 11:05:40,009 - Epoch: [96][ 30/ 134] Loss 0.316542 Top1 83.671875 Top5 97.669271 -2022-12-06 11:05:40,138 - Epoch: [96][ 40/ 134] Loss 0.321669 Top1 83.828125 Top5 97.617188 -2022-12-06 11:05:40,267 - Epoch: [96][ 50/ 134] Loss 0.321341 Top1 83.812500 Top5 97.593750 -2022-12-06 11:05:40,395 - Epoch: [96][ 60/ 134] Loss 0.319070 Top1 83.893229 Top5 97.571615 -2022-12-06 11:05:40,522 - Epoch: [96][ 70/ 134] Loss 0.317679 Top1 83.895089 Top5 97.639509 -2022-12-06 11:05:40,649 - Epoch: [96][ 80/ 134] Loss 0.314260 Top1 83.999023 Top5 97.670898 -2022-12-06 11:05:40,778 - Epoch: [96][ 90/ 134] Loss 0.314767 Top1 84.001736 Top5 97.621528 -2022-12-06 11:05:40,908 - Epoch: [96][ 100/ 134] Loss 0.311355 Top1 84.148438 Top5 97.695312 -2022-12-06 11:05:41,037 - Epoch: [96][ 110/ 134] Loss 0.312515 Top1 84.108665 Top5 97.681108 -2022-12-06 11:05:41,167 - Epoch: [96][ 120/ 134] Loss 0.310577 Top1 84.117839 Top5 97.685547 -2022-12-06 11:05:41,296 - Epoch: [96][ 130/ 134] Loss 0.309237 Top1 84.098558 Top5 97.734375 -2022-12-06 11:05:41,333 - Epoch: [96][ 134/ 134] Loss 0.307850 Top1 84.145448 Top5 97.740924 -2022-12-06 11:05:41,424 - ==> Top1: 84.145 Top5: 97.741 Loss: 0.308 - -2022-12-06 11:05:41,424 - ==> Confusion: -[[ 880 3 2 1 6 6 0 0 6 63 1 1 5 4 4 2 1 1 2 0 8] - [ 2 916 2 4 5 21 1 17 2 0 6 4 2 1 3 1 4 1 21 4 10] - [ 6 4 975 23 2 1 22 10 1 2 11 5 4 4 4 1 2 0 7 4 15] - [ 7 2 15 933 0 1 0 0 0 0 13 0 4 1 13 1 1 0 23 0 6] - [ 9 13 1 0 924 11 0 1 3 9 2 5 0 4 17 6 8 0 1 1 5] - [ 2 18 2 5 3 961 3 21 3 2 1 12 4 12 2 0 2 1 2 8 5] - [ 1 1 9 3 0 3 1061 4 0 0 3 0 4 1 0 8 2 4 1 8 5] - [ 0 10 9 2 2 32 7 908 0 0 8 6 4 1 0 0 2 0 44 13 6] - [ 4 4 0 0 0 2 0 2 986 34 5 1 0 10 8 0 2 1 2 1 2] - [ 55 0 0 1 3 2 0 2 31 884 1 3 0 10 2 0 0 2 0 0 5] - [ 1 1 2 5 1 0 3 3 17 2 950 2 4 15 1 0 0 0 8 0 4] - [ 1 2 0 0 0 10 2 3 1 0 1 952 41 5 0 6 6 7 1 8 5] - [ 1 0 0 2 2 3 1 1 0 0 0 21 899 4 0 11 2 14 1 4 3] - [ 0 1 0 1 0 15 0 2 18 15 8 6 6 937 0 4 3 0 0 2 5] - [ 9 2 0 18 3 4 0 0 36 4 0 1 1 5 1022 1 3 1 11 2 7] - [ 1 0 0 3 2 1 4 0 0 0 1 7 11 2 0 986 8 10 0 4 3] - [ 2 3 2 2 2 2 1 0 1 0 0 4 4 4 1 16 1016 0 0 4 8] - [ 2 0 0 2 0 0 0 0 0 2 1 5 35 5 0 12 1 966 1 3 1] - [ 3 6 3 10 1 2 0 21 2 0 7 3 5 1 5 1 1 2 929 3 3] - [ 2 4 1 0 0 6 7 8 0 0 2 18 14 7 1 2 10 2 1 991 4] - [ 102 231 135 120 76 211 93 150 135 112 204 114 470 351 157 105 208 111 247 257 9637]] - -2022-12-06 11:05:42,104 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:05:42,105 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:05:42,110 - - -2022-12-06 11:05:42,111 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:05:43,073 - Epoch: [97][ 10/ 1200] Overall Loss 0.248398 Objective Loss 0.248398 LR 0.001000 Time 0.096190 -2022-12-06 11:05:43,274 - Epoch: [97][ 20/ 1200] Overall Loss 0.248666 Objective Loss 0.248666 LR 0.001000 Time 0.058104 -2022-12-06 11:05:43,475 - Epoch: [97][ 30/ 1200] Overall Loss 0.252458 Objective Loss 0.252458 LR 0.001000 Time 0.045411 -2022-12-06 11:05:43,671 - Epoch: [97][ 40/ 1200] Overall Loss 0.257799 Objective Loss 0.257799 LR 0.001000 Time 0.038958 -2022-12-06 11:05:43,872 - Epoch: [97][ 50/ 1200] Overall Loss 0.257462 Objective Loss 0.257462 LR 0.001000 Time 0.035177 -2022-12-06 11:05:44,068 - Epoch: [97][ 60/ 1200] Overall Loss 0.264308 Objective Loss 0.264308 LR 0.001000 Time 0.032571 -2022-12-06 11:05:44,268 - Epoch: [97][ 70/ 1200] Overall Loss 0.264994 Objective Loss 0.264994 LR 0.001000 Time 0.030761 -2022-12-06 11:05:44,464 - Epoch: [97][ 80/ 1200] Overall Loss 0.264596 Objective Loss 0.264596 LR 0.001000 Time 0.029360 -2022-12-06 11:05:44,665 - Epoch: [97][ 90/ 1200] Overall Loss 0.262181 Objective Loss 0.262181 LR 0.001000 Time 0.028323 -2022-12-06 11:05:44,861 - Epoch: [97][ 100/ 1200] Overall Loss 0.261271 Objective Loss 0.261271 LR 0.001000 Time 0.027443 -2022-12-06 11:05:45,061 - Epoch: [97][ 110/ 1200] Overall Loss 0.258874 Objective Loss 0.258874 LR 0.001000 Time 0.026765 -2022-12-06 11:05:45,258 - Epoch: [97][ 120/ 1200] Overall Loss 0.261139 Objective Loss 0.261139 LR 0.001000 Time 0.026168 -2022-12-06 11:05:45,458 - Epoch: [97][ 130/ 1200] Overall Loss 0.262816 Objective Loss 0.262816 LR 0.001000 Time 0.025692 -2022-12-06 11:05:45,654 - Epoch: [97][ 140/ 1200] Overall Loss 0.261978 Objective Loss 0.261978 LR 0.001000 Time 0.025254 -2022-12-06 11:05:45,854 - Epoch: [97][ 150/ 1200] Overall Loss 0.260812 Objective Loss 0.260812 LR 0.001000 Time 0.024895 -2022-12-06 11:05:46,050 - Epoch: [97][ 160/ 1200] Overall Loss 0.261388 Objective Loss 0.261388 LR 0.001000 Time 0.024563 -2022-12-06 11:05:46,250 - Epoch: [97][ 170/ 1200] Overall Loss 0.263225 Objective Loss 0.263225 LR 0.001000 Time 0.024290 -2022-12-06 11:05:46,446 - Epoch: [97][ 180/ 1200] Overall Loss 0.265028 Objective Loss 0.265028 LR 0.001000 Time 0.024030 -2022-12-06 11:05:46,646 - Epoch: [97][ 190/ 1200] Overall Loss 0.264324 Objective Loss 0.264324 LR 0.001000 Time 0.023814 -2022-12-06 11:05:46,843 - Epoch: [97][ 200/ 1200] Overall Loss 0.264414 Objective Loss 0.264414 LR 0.001000 Time 0.023606 -2022-12-06 11:05:47,043 - Epoch: [97][ 210/ 1200] Overall Loss 0.265197 Objective Loss 0.265197 LR 0.001000 Time 0.023431 -2022-12-06 11:05:47,240 - Epoch: [97][ 220/ 1200] Overall Loss 0.264646 Objective Loss 0.264646 LR 0.001000 Time 0.023256 -2022-12-06 11:05:47,440 - Epoch: [97][ 230/ 1200] Overall Loss 0.266351 Objective Loss 0.266351 LR 0.001000 Time 0.023113 -2022-12-06 11:05:47,635 - Epoch: [97][ 240/ 1200] Overall Loss 0.266571 Objective Loss 0.266571 LR 0.001000 Time 0.022963 -2022-12-06 11:05:47,835 - Epoch: [97][ 250/ 1200] Overall Loss 0.266046 Objective Loss 0.266046 LR 0.001000 Time 0.022841 -2022-12-06 11:05:48,032 - Epoch: [97][ 260/ 1200] Overall Loss 0.264957 Objective Loss 0.264957 LR 0.001000 Time 0.022716 -2022-12-06 11:05:48,231 - Epoch: [97][ 270/ 1200] Overall Loss 0.264815 Objective Loss 0.264815 LR 0.001000 Time 0.022611 -2022-12-06 11:05:48,427 - Epoch: [97][ 280/ 1200] Overall Loss 0.265457 Objective Loss 0.265457 LR 0.001000 Time 0.022502 -2022-12-06 11:05:48,627 - Epoch: [97][ 290/ 1200] Overall Loss 0.265586 Objective Loss 0.265586 LR 0.001000 Time 0.022411 -2022-12-06 11:05:48,822 - Epoch: [97][ 300/ 1200] Overall Loss 0.265861 Objective Loss 0.265861 LR 0.001000 Time 0.022315 -2022-12-06 11:05:49,022 - Epoch: [97][ 310/ 1200] Overall Loss 0.265457 Objective Loss 0.265457 LR 0.001000 Time 0.022237 -2022-12-06 11:05:49,219 - Epoch: [97][ 320/ 1200] Overall Loss 0.264733 Objective Loss 0.264733 LR 0.001000 Time 0.022157 -2022-12-06 11:05:49,419 - Epoch: [97][ 330/ 1200] Overall Loss 0.264279 Objective Loss 0.264279 LR 0.001000 Time 0.022090 -2022-12-06 11:05:49,615 - Epoch: [97][ 340/ 1200] Overall Loss 0.264506 Objective Loss 0.264506 LR 0.001000 Time 0.022016 -2022-12-06 11:05:49,816 - Epoch: [97][ 350/ 1200] Overall Loss 0.264328 Objective Loss 0.264328 LR 0.001000 Time 0.021957 -2022-12-06 11:05:50,012 - Epoch: [97][ 360/ 1200] Overall Loss 0.264560 Objective Loss 0.264560 LR 0.001000 Time 0.021890 -2022-12-06 11:05:50,211 - Epoch: [97][ 370/ 1200] Overall Loss 0.263768 Objective Loss 0.263768 LR 0.001000 Time 0.021836 -2022-12-06 11:05:50,407 - Epoch: [97][ 380/ 1200] Overall Loss 0.263882 Objective Loss 0.263882 LR 0.001000 Time 0.021774 -2022-12-06 11:05:50,606 - Epoch: [97][ 390/ 1200] Overall Loss 0.265204 Objective Loss 0.265204 LR 0.001000 Time 0.021726 -2022-12-06 11:05:50,803 - Epoch: [97][ 400/ 1200] Overall Loss 0.265288 Objective Loss 0.265288 LR 0.001000 Time 0.021673 -2022-12-06 11:05:51,002 - Epoch: [97][ 410/ 1200] Overall Loss 0.265760 Objective Loss 0.265760 LR 0.001000 Time 0.021629 -2022-12-06 11:05:51,199 - Epoch: [97][ 420/ 1200] Overall Loss 0.266029 Objective Loss 0.266029 LR 0.001000 Time 0.021582 -2022-12-06 11:05:51,399 - Epoch: [97][ 430/ 1200] Overall Loss 0.265917 Objective Loss 0.265917 LR 0.001000 Time 0.021543 -2022-12-06 11:05:51,595 - Epoch: [97][ 440/ 1200] Overall Loss 0.266428 Objective Loss 0.266428 LR 0.001000 Time 0.021499 -2022-12-06 11:05:51,795 - Epoch: [97][ 450/ 1200] Overall Loss 0.266657 Objective Loss 0.266657 LR 0.001000 Time 0.021464 -2022-12-06 11:05:51,991 - Epoch: [97][ 460/ 1200] Overall Loss 0.266596 Objective Loss 0.266596 LR 0.001000 Time 0.021421 -2022-12-06 11:05:52,191 - Epoch: [97][ 470/ 1200] Overall Loss 0.266991 Objective Loss 0.266991 LR 0.001000 Time 0.021390 -2022-12-06 11:05:52,387 - Epoch: [97][ 480/ 1200] Overall Loss 0.266619 Objective Loss 0.266619 LR 0.001000 Time 0.021353 -2022-12-06 11:05:52,587 - Epoch: [97][ 490/ 1200] Overall Loss 0.266331 Objective Loss 0.266331 LR 0.001000 Time 0.021324 -2022-12-06 11:05:52,784 - Epoch: [97][ 500/ 1200] Overall Loss 0.266427 Objective Loss 0.266427 LR 0.001000 Time 0.021290 -2022-12-06 11:05:52,985 - Epoch: [97][ 510/ 1200] Overall Loss 0.266838 Objective Loss 0.266838 LR 0.001000 Time 0.021265 -2022-12-06 11:05:53,182 - Epoch: [97][ 520/ 1200] Overall Loss 0.266939 Objective Loss 0.266939 LR 0.001000 Time 0.021234 -2022-12-06 11:05:53,382 - Epoch: [97][ 530/ 1200] Overall Loss 0.266825 Objective Loss 0.266825 LR 0.001000 Time 0.021210 -2022-12-06 11:05:53,579 - Epoch: [97][ 540/ 1200] Overall Loss 0.266132 Objective Loss 0.266132 LR 0.001000 Time 0.021180 -2022-12-06 11:05:53,778 - Epoch: [97][ 550/ 1200] Overall Loss 0.266499 Objective Loss 0.266499 LR 0.001000 Time 0.021156 -2022-12-06 11:05:53,975 - Epoch: [97][ 560/ 1200] Overall Loss 0.266550 Objective Loss 0.266550 LR 0.001000 Time 0.021128 -2022-12-06 11:05:54,175 - Epoch: [97][ 570/ 1200] Overall Loss 0.266159 Objective Loss 0.266159 LR 0.001000 Time 0.021108 -2022-12-06 11:05:54,373 - Epoch: [97][ 580/ 1200] Overall Loss 0.266306 Objective Loss 0.266306 LR 0.001000 Time 0.021084 -2022-12-06 11:05:54,573 - Epoch: [97][ 590/ 1200] Overall Loss 0.266854 Objective Loss 0.266854 LR 0.001000 Time 0.021065 -2022-12-06 11:05:54,770 - Epoch: [97][ 600/ 1200] Overall Loss 0.267163 Objective Loss 0.267163 LR 0.001000 Time 0.021041 -2022-12-06 11:05:54,969 - Epoch: [97][ 610/ 1200] Overall Loss 0.266788 Objective Loss 0.266788 LR 0.001000 Time 0.021022 -2022-12-06 11:05:55,165 - Epoch: [97][ 620/ 1200] Overall Loss 0.266814 Objective Loss 0.266814 LR 0.001000 Time 0.020998 -2022-12-06 11:05:55,365 - Epoch: [97][ 630/ 1200] Overall Loss 0.266875 Objective Loss 0.266875 LR 0.001000 Time 0.020981 -2022-12-06 11:05:55,562 - Epoch: [97][ 640/ 1200] Overall Loss 0.267135 Objective Loss 0.267135 LR 0.001000 Time 0.020960 -2022-12-06 11:05:55,762 - Epoch: [97][ 650/ 1200] Overall Loss 0.267966 Objective Loss 0.267966 LR 0.001000 Time 0.020944 -2022-12-06 11:05:55,958 - Epoch: [97][ 660/ 1200] Overall Loss 0.268084 Objective Loss 0.268084 LR 0.001000 Time 0.020924 -2022-12-06 11:05:56,158 - Epoch: [97][ 670/ 1200] Overall Loss 0.268117 Objective Loss 0.268117 LR 0.001000 Time 0.020910 -2022-12-06 11:05:56,354 - Epoch: [97][ 680/ 1200] Overall Loss 0.268144 Objective Loss 0.268144 LR 0.001000 Time 0.020890 -2022-12-06 11:05:56,554 - Epoch: [97][ 690/ 1200] Overall Loss 0.268062 Objective Loss 0.268062 LR 0.001000 Time 0.020875 -2022-12-06 11:05:56,751 - Epoch: [97][ 700/ 1200] Overall Loss 0.268180 Objective Loss 0.268180 LR 0.001000 Time 0.020857 -2022-12-06 11:05:56,951 - Epoch: [97][ 710/ 1200] Overall Loss 0.268214 Objective Loss 0.268214 LR 0.001000 Time 0.020844 -2022-12-06 11:05:57,147 - Epoch: [97][ 720/ 1200] Overall Loss 0.267906 Objective Loss 0.267906 LR 0.001000 Time 0.020827 -2022-12-06 11:05:57,347 - Epoch: [97][ 730/ 1200] Overall Loss 0.267778 Objective Loss 0.267778 LR 0.001000 Time 0.020814 -2022-12-06 11:05:57,544 - Epoch: [97][ 740/ 1200] Overall Loss 0.267691 Objective Loss 0.267691 LR 0.001000 Time 0.020798 -2022-12-06 11:05:57,743 - Epoch: [97][ 750/ 1200] Overall Loss 0.267549 Objective Loss 0.267549 LR 0.001000 Time 0.020786 -2022-12-06 11:05:57,939 - Epoch: [97][ 760/ 1200] Overall Loss 0.267482 Objective Loss 0.267482 LR 0.001000 Time 0.020770 -2022-12-06 11:05:58,139 - Epoch: [97][ 770/ 1200] Overall Loss 0.267811 Objective Loss 0.267811 LR 0.001000 Time 0.020759 -2022-12-06 11:05:58,336 - Epoch: [97][ 780/ 1200] Overall Loss 0.267931 Objective Loss 0.267931 LR 0.001000 Time 0.020744 -2022-12-06 11:05:58,536 - Epoch: [97][ 790/ 1200] Overall Loss 0.267835 Objective Loss 0.267835 LR 0.001000 Time 0.020734 -2022-12-06 11:05:58,732 - Epoch: [97][ 800/ 1200] Overall Loss 0.267795 Objective Loss 0.267795 LR 0.001000 Time 0.020719 -2022-12-06 11:05:58,931 - Epoch: [97][ 810/ 1200] Overall Loss 0.267886 Objective Loss 0.267886 LR 0.001000 Time 0.020709 -2022-12-06 11:05:59,128 - Epoch: [97][ 820/ 1200] Overall Loss 0.267983 Objective Loss 0.267983 LR 0.001000 Time 0.020695 -2022-12-06 11:05:59,327 - Epoch: [97][ 830/ 1200] Overall Loss 0.268604 Objective Loss 0.268604 LR 0.001000 Time 0.020686 -2022-12-06 11:05:59,524 - Epoch: [97][ 840/ 1200] Overall Loss 0.269152 Objective Loss 0.269152 LR 0.001000 Time 0.020672 -2022-12-06 11:05:59,724 - Epoch: [97][ 850/ 1200] Overall Loss 0.269241 Objective Loss 0.269241 LR 0.001000 Time 0.020665 -2022-12-06 11:05:59,922 - Epoch: [97][ 860/ 1200] Overall Loss 0.269831 Objective Loss 0.269831 LR 0.001000 Time 0.020654 -2022-12-06 11:06:00,121 - Epoch: [97][ 870/ 1200] Overall Loss 0.269799 Objective Loss 0.269799 LR 0.001000 Time 0.020645 -2022-12-06 11:06:00,317 - Epoch: [97][ 880/ 1200] Overall Loss 0.270038 Objective Loss 0.270038 LR 0.001000 Time 0.020632 -2022-12-06 11:06:00,517 - Epoch: [97][ 890/ 1200] Overall Loss 0.269728 Objective Loss 0.269728 LR 0.001000 Time 0.020624 -2022-12-06 11:06:00,715 - Epoch: [97][ 900/ 1200] Overall Loss 0.270025 Objective Loss 0.270025 LR 0.001000 Time 0.020614 -2022-12-06 11:06:00,914 - Epoch: [97][ 910/ 1200] Overall Loss 0.270749 Objective Loss 0.270749 LR 0.001000 Time 0.020606 -2022-12-06 11:06:01,111 - Epoch: [97][ 920/ 1200] Overall Loss 0.271030 Objective Loss 0.271030 LR 0.001000 Time 0.020595 -2022-12-06 11:06:01,311 - Epoch: [97][ 930/ 1200] Overall Loss 0.271119 Objective Loss 0.271119 LR 0.001000 Time 0.020588 -2022-12-06 11:06:01,508 - Epoch: [97][ 940/ 1200] Overall Loss 0.271092 Objective Loss 0.271092 LR 0.001000 Time 0.020578 -2022-12-06 11:06:01,708 - Epoch: [97][ 950/ 1200] Overall Loss 0.271248 Objective Loss 0.271248 LR 0.001000 Time 0.020571 -2022-12-06 11:06:01,903 - Epoch: [97][ 960/ 1200] Overall Loss 0.271104 Objective Loss 0.271104 LR 0.001000 Time 0.020560 -2022-12-06 11:06:02,104 - Epoch: [97][ 970/ 1200] Overall Loss 0.271058 Objective Loss 0.271058 LR 0.001000 Time 0.020554 -2022-12-06 11:06:02,301 - Epoch: [97][ 980/ 1200] Overall Loss 0.271205 Objective Loss 0.271205 LR 0.001000 Time 0.020545 -2022-12-06 11:06:02,500 - Epoch: [97][ 990/ 1200] Overall Loss 0.271268 Objective Loss 0.271268 LR 0.001000 Time 0.020538 -2022-12-06 11:06:02,698 - Epoch: [97][ 1000/ 1200] Overall Loss 0.271391 Objective Loss 0.271391 LR 0.001000 Time 0.020529 -2022-12-06 11:06:02,897 - Epoch: [97][ 1010/ 1200] Overall Loss 0.271420 Objective Loss 0.271420 LR 0.001000 Time 0.020523 -2022-12-06 11:06:03,095 - Epoch: [97][ 1020/ 1200] Overall Loss 0.271710 Objective Loss 0.271710 LR 0.001000 Time 0.020515 -2022-12-06 11:06:03,294 - Epoch: [97][ 1030/ 1200] Overall Loss 0.271979 Objective Loss 0.271979 LR 0.001000 Time 0.020509 -2022-12-06 11:06:03,491 - Epoch: [97][ 1040/ 1200] Overall Loss 0.272237 Objective Loss 0.272237 LR 0.001000 Time 0.020500 -2022-12-06 11:06:03,690 - Epoch: [97][ 1050/ 1200] Overall Loss 0.272499 Objective Loss 0.272499 LR 0.001000 Time 0.020495 -2022-12-06 11:06:03,888 - Epoch: [97][ 1060/ 1200] Overall Loss 0.272605 Objective Loss 0.272605 LR 0.001000 Time 0.020487 -2022-12-06 11:06:04,088 - Epoch: [97][ 1070/ 1200] Overall Loss 0.272640 Objective Loss 0.272640 LR 0.001000 Time 0.020482 -2022-12-06 11:06:04,284 - Epoch: [97][ 1080/ 1200] Overall Loss 0.272463 Objective Loss 0.272463 LR 0.001000 Time 0.020474 -2022-12-06 11:06:04,485 - Epoch: [97][ 1090/ 1200] Overall Loss 0.272252 Objective Loss 0.272252 LR 0.001000 Time 0.020469 -2022-12-06 11:06:04,681 - Epoch: [97][ 1100/ 1200] Overall Loss 0.271963 Objective Loss 0.271963 LR 0.001000 Time 0.020461 -2022-12-06 11:06:04,881 - Epoch: [97][ 1110/ 1200] Overall Loss 0.271913 Objective Loss 0.271913 LR 0.001000 Time 0.020456 -2022-12-06 11:06:05,077 - Epoch: [97][ 1120/ 1200] Overall Loss 0.272065 Objective Loss 0.272065 LR 0.001000 Time 0.020448 -2022-12-06 11:06:05,277 - Epoch: [97][ 1130/ 1200] Overall Loss 0.272256 Objective Loss 0.272256 LR 0.001000 Time 0.020444 -2022-12-06 11:06:05,474 - Epoch: [97][ 1140/ 1200] Overall Loss 0.272348 Objective Loss 0.272348 LR 0.001000 Time 0.020437 -2022-12-06 11:06:05,674 - Epoch: [97][ 1150/ 1200] Overall Loss 0.272242 Objective Loss 0.272242 LR 0.001000 Time 0.020433 -2022-12-06 11:06:05,870 - Epoch: [97][ 1160/ 1200] Overall Loss 0.272404 Objective Loss 0.272404 LR 0.001000 Time 0.020425 -2022-12-06 11:06:06,070 - Epoch: [97][ 1170/ 1200] Overall Loss 0.272399 Objective Loss 0.272399 LR 0.001000 Time 0.020421 -2022-12-06 11:06:06,267 - Epoch: [97][ 1180/ 1200] Overall Loss 0.272240 Objective Loss 0.272240 LR 0.001000 Time 0.020414 -2022-12-06 11:06:06,466 - Epoch: [97][ 1190/ 1200] Overall Loss 0.272446 Objective Loss 0.272446 LR 0.001000 Time 0.020410 -2022-12-06 11:06:06,695 - Epoch: [97][ 1200/ 1200] Overall Loss 0.272336 Objective Loss 0.272336 Top1 85.564854 Top5 97.280335 LR 0.001000 Time 0.020430 -2022-12-06 11:06:06,783 - --- validate (epoch=97)----------- -2022-12-06 11:06:06,784 - 34129 samples (256 per mini-batch) -2022-12-06 11:06:07,228 - Epoch: [97][ 10/ 134] Loss 0.302907 Top1 84.296875 Top5 97.890625 -2022-12-06 11:06:07,359 - Epoch: [97][ 20/ 134] Loss 0.308969 Top1 83.750000 Top5 97.871094 -2022-12-06 11:06:07,492 - Epoch: [97][ 30/ 134] Loss 0.309139 Top1 83.984375 Top5 97.786458 -2022-12-06 11:06:07,623 - Epoch: [97][ 40/ 134] Loss 0.309464 Top1 83.847656 Top5 97.832031 -2022-12-06 11:06:07,755 - Epoch: [97][ 50/ 134] Loss 0.310829 Top1 83.984375 Top5 97.867188 -2022-12-06 11:06:07,885 - Epoch: [97][ 60/ 134] Loss 0.314207 Top1 83.958333 Top5 97.819010 -2022-12-06 11:06:08,016 - Epoch: [97][ 70/ 134] Loss 0.306141 Top1 84.146205 Top5 97.845982 -2022-12-06 11:06:08,147 - Epoch: [97][ 80/ 134] Loss 0.301478 Top1 84.262695 Top5 97.817383 -2022-12-06 11:06:08,278 - Epoch: [97][ 90/ 134] Loss 0.304442 Top1 84.032118 Top5 97.786458 -2022-12-06 11:06:08,409 - Epoch: [97][ 100/ 134] Loss 0.306814 Top1 83.972656 Top5 97.738281 -2022-12-06 11:06:08,541 - Epoch: [97][ 110/ 134] Loss 0.305215 Top1 84.048295 Top5 97.766335 -2022-12-06 11:06:08,673 - Epoch: [97][ 120/ 134] Loss 0.303007 Top1 84.127604 Top5 97.760417 -2022-12-06 11:06:08,803 - Epoch: [97][ 130/ 134] Loss 0.304298 Top1 84.047476 Top5 97.767428 -2022-12-06 11:06:08,840 - Epoch: [97][ 134/ 134] Loss 0.304905 Top1 84.048756 Top5 97.746784 -2022-12-06 11:06:08,933 - ==> Top1: 84.049 Top5: 97.747 Loss: 0.305 - -2022-12-06 11:06:08,934 - ==> Confusion: -[[ 874 2 2 2 14 4 0 0 9 60 0 4 2 4 5 4 1 1 2 1 5] - [ 4 933 0 2 4 24 2 16 4 0 10 4 2 1 0 1 7 0 5 2 6] - [ 6 6 961 31 4 1 26 10 0 2 12 3 2 5 6 4 1 0 5 2 16] - [ 3 2 6 947 1 1 0 1 1 0 10 1 4 0 19 3 2 4 10 0 5] - [ 6 12 1 0 945 7 0 2 0 7 2 2 3 2 12 8 7 1 1 1 1] - [ 4 18 0 2 5 976 3 11 1 3 1 14 6 14 2 1 1 0 2 2 3] - [ 2 2 20 2 0 2 1056 5 0 0 6 2 3 2 0 6 1 2 2 2 3] - [ 1 19 6 2 2 33 8 928 1 0 4 11 1 2 0 2 2 0 18 7 7] - [ 5 5 0 2 0 3 0 1 969 33 10 1 4 10 13 0 3 1 2 1 1] - [ 65 0 1 0 7 0 0 3 36 851 2 3 0 16 4 1 1 2 0 2 7] - [ 1 1 1 7 1 0 1 2 8 2 963 1 3 16 3 1 0 0 5 1 2] - [ 1 1 3 1 0 9 4 2 1 0 0 977 25 3 0 7 1 8 0 6 2] - [ 1 1 1 3 1 3 0 0 1 0 0 48 883 0 0 5 1 11 1 3 6] - [ 0 2 1 1 1 14 0 2 12 12 11 4 3 941 1 1 5 3 2 0 7] - [ 8 2 1 13 3 4 0 0 25 0 2 3 1 4 1051 1 2 0 6 0 4] - [ 1 0 1 1 3 1 3 0 1 1 0 16 11 1 0 974 10 13 0 3 3] - [ 3 6 0 2 2 2 0 0 2 0 2 5 2 2 1 11 1020 1 0 2 9] - [ 2 1 1 3 1 2 1 1 1 2 0 16 18 2 1 9 3 969 0 1 2] - [ 4 4 1 20 1 4 1 22 5 1 13 1 3 2 7 1 2 1 909 1 5] - [ 2 4 2 0 0 7 12 6 0 1 1 26 9 6 0 8 8 2 4 976 6] - [ 121 290 166 148 125 190 84 140 107 93 218 165 411 309 213 108 287 88 171 210 9582]] - -2022-12-06 11:06:09,518 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:06:09,518 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:06:09,524 - - -2022-12-06 11:06:09,524 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:06:10,569 - Epoch: [98][ 10/ 1200] Overall Loss 0.270388 Objective Loss 0.270388 LR 0.001000 Time 0.104484 -2022-12-06 11:06:10,773 - Epoch: [98][ 20/ 1200] Overall Loss 0.270660 Objective Loss 0.270660 LR 0.001000 Time 0.062369 -2022-12-06 11:06:10,973 - Epoch: [98][ 30/ 1200] Overall Loss 0.263102 Objective Loss 0.263102 LR 0.001000 Time 0.048230 -2022-12-06 11:06:11,171 - Epoch: [98][ 40/ 1200] Overall Loss 0.261007 Objective Loss 0.261007 LR 0.001000 Time 0.041128 -2022-12-06 11:06:11,371 - Epoch: [98][ 50/ 1200] Overall Loss 0.256703 Objective Loss 0.256703 LR 0.001000 Time 0.036889 -2022-12-06 11:06:11,569 - Epoch: [98][ 60/ 1200] Overall Loss 0.257967 Objective Loss 0.257967 LR 0.001000 Time 0.034024 -2022-12-06 11:06:11,769 - Epoch: [98][ 70/ 1200] Overall Loss 0.259369 Objective Loss 0.259369 LR 0.001000 Time 0.032009 -2022-12-06 11:06:11,967 - Epoch: [98][ 80/ 1200] Overall Loss 0.259034 Objective Loss 0.259034 LR 0.001000 Time 0.030479 -2022-12-06 11:06:12,167 - Epoch: [98][ 90/ 1200] Overall Loss 0.259238 Objective Loss 0.259238 LR 0.001000 Time 0.029305 -2022-12-06 11:06:12,365 - Epoch: [98][ 100/ 1200] Overall Loss 0.260159 Objective Loss 0.260159 LR 0.001000 Time 0.028356 -2022-12-06 11:06:12,565 - Epoch: [98][ 110/ 1200] Overall Loss 0.262543 Objective Loss 0.262543 LR 0.001000 Time 0.027591 -2022-12-06 11:06:12,764 - Epoch: [98][ 120/ 1200] Overall Loss 0.262829 Objective Loss 0.262829 LR 0.001000 Time 0.026943 -2022-12-06 11:06:12,964 - Epoch: [98][ 130/ 1200] Overall Loss 0.260980 Objective Loss 0.260980 LR 0.001000 Time 0.026404 -2022-12-06 11:06:13,163 - Epoch: [98][ 140/ 1200] Overall Loss 0.260904 Objective Loss 0.260904 LR 0.001000 Time 0.025934 -2022-12-06 11:06:13,363 - Epoch: [98][ 150/ 1200] Overall Loss 0.262773 Objective Loss 0.262773 LR 0.001000 Time 0.025536 -2022-12-06 11:06:13,561 - Epoch: [98][ 160/ 1200] Overall Loss 0.262709 Objective Loss 0.262709 LR 0.001000 Time 0.025177 -2022-12-06 11:06:13,762 - Epoch: [98][ 170/ 1200] Overall Loss 0.263564 Objective Loss 0.263564 LR 0.001000 Time 0.024871 -2022-12-06 11:06:13,960 - Epoch: [98][ 180/ 1200] Overall Loss 0.263297 Objective Loss 0.263297 LR 0.001000 Time 0.024589 -2022-12-06 11:06:14,160 - Epoch: [98][ 190/ 1200] Overall Loss 0.264458 Objective Loss 0.264458 LR 0.001000 Time 0.024345 -2022-12-06 11:06:14,359 - Epoch: [98][ 200/ 1200] Overall Loss 0.265047 Objective Loss 0.265047 LR 0.001000 Time 0.024117 -2022-12-06 11:06:14,559 - Epoch: [98][ 210/ 1200] Overall Loss 0.264917 Objective Loss 0.264917 LR 0.001000 Time 0.023919 -2022-12-06 11:06:14,758 - Epoch: [98][ 220/ 1200] Overall Loss 0.265457 Objective Loss 0.265457 LR 0.001000 Time 0.023735 -2022-12-06 11:06:14,958 - Epoch: [98][ 230/ 1200] Overall Loss 0.265831 Objective Loss 0.265831 LR 0.001000 Time 0.023570 -2022-12-06 11:06:15,157 - Epoch: [98][ 240/ 1200] Overall Loss 0.265508 Objective Loss 0.265508 LR 0.001000 Time 0.023413 -2022-12-06 11:06:15,357 - Epoch: [98][ 250/ 1200] Overall Loss 0.266021 Objective Loss 0.266021 LR 0.001000 Time 0.023275 -2022-12-06 11:06:15,555 - Epoch: [98][ 260/ 1200] Overall Loss 0.267027 Objective Loss 0.267027 LR 0.001000 Time 0.023141 -2022-12-06 11:06:15,755 - Epoch: [98][ 270/ 1200] Overall Loss 0.265807 Objective Loss 0.265807 LR 0.001000 Time 0.023023 -2022-12-06 11:06:15,954 - Epoch: [98][ 280/ 1200] Overall Loss 0.266119 Objective Loss 0.266119 LR 0.001000 Time 0.022908 -2022-12-06 11:06:16,154 - Epoch: [98][ 290/ 1200] Overall Loss 0.265614 Objective Loss 0.265614 LR 0.001000 Time 0.022806 -2022-12-06 11:06:16,352 - Epoch: [98][ 300/ 1200] Overall Loss 0.265374 Objective Loss 0.265374 LR 0.001000 Time 0.022706 -2022-12-06 11:06:16,553 - Epoch: [98][ 310/ 1200] Overall Loss 0.265835 Objective Loss 0.265835 LR 0.001000 Time 0.022619 -2022-12-06 11:06:16,752 - Epoch: [98][ 320/ 1200] Overall Loss 0.265768 Objective Loss 0.265768 LR 0.001000 Time 0.022531 -2022-12-06 11:06:16,952 - Epoch: [98][ 330/ 1200] Overall Loss 0.265580 Objective Loss 0.265580 LR 0.001000 Time 0.022454 -2022-12-06 11:06:17,150 - Epoch: [98][ 340/ 1200] Overall Loss 0.264538 Objective Loss 0.264538 LR 0.001000 Time 0.022375 -2022-12-06 11:06:17,350 - Epoch: [98][ 350/ 1200] Overall Loss 0.265311 Objective Loss 0.265311 LR 0.001000 Time 0.022305 -2022-12-06 11:06:17,546 - Epoch: [98][ 360/ 1200] Overall Loss 0.266109 Objective Loss 0.266109 LR 0.001000 Time 0.022229 -2022-12-06 11:06:17,745 - Epoch: [98][ 370/ 1200] Overall Loss 0.266211 Objective Loss 0.266211 LR 0.001000 Time 0.022162 -2022-12-06 11:06:17,941 - Epoch: [98][ 380/ 1200] Overall Loss 0.266760 Objective Loss 0.266760 LR 0.001000 Time 0.022095 -2022-12-06 11:06:18,140 - Epoch: [98][ 390/ 1200] Overall Loss 0.267663 Objective Loss 0.267663 LR 0.001000 Time 0.022036 -2022-12-06 11:06:18,336 - Epoch: [98][ 400/ 1200] Overall Loss 0.267954 Objective Loss 0.267954 LR 0.001000 Time 0.021975 -2022-12-06 11:06:18,534 - Epoch: [98][ 410/ 1200] Overall Loss 0.267757 Objective Loss 0.267757 LR 0.001000 Time 0.021921 -2022-12-06 11:06:18,730 - Epoch: [98][ 420/ 1200] Overall Loss 0.268400 Objective Loss 0.268400 LR 0.001000 Time 0.021864 -2022-12-06 11:06:18,929 - Epoch: [98][ 430/ 1200] Overall Loss 0.268521 Objective Loss 0.268521 LR 0.001000 Time 0.021817 -2022-12-06 11:06:19,126 - Epoch: [98][ 440/ 1200] Overall Loss 0.269379 Objective Loss 0.269379 LR 0.001000 Time 0.021767 -2022-12-06 11:06:19,324 - Epoch: [98][ 450/ 1200] Overall Loss 0.269014 Objective Loss 0.269014 LR 0.001000 Time 0.021722 -2022-12-06 11:06:19,520 - Epoch: [98][ 460/ 1200] Overall Loss 0.268434 Objective Loss 0.268434 LR 0.001000 Time 0.021675 -2022-12-06 11:06:19,718 - Epoch: [98][ 470/ 1200] Overall Loss 0.268984 Objective Loss 0.268984 LR 0.001000 Time 0.021634 -2022-12-06 11:06:19,915 - Epoch: [98][ 480/ 1200] Overall Loss 0.269395 Objective Loss 0.269395 LR 0.001000 Time 0.021592 -2022-12-06 11:06:20,112 - Epoch: [98][ 490/ 1200] Overall Loss 0.270082 Objective Loss 0.270082 LR 0.001000 Time 0.021553 -2022-12-06 11:06:20,309 - Epoch: [98][ 500/ 1200] Overall Loss 0.270213 Objective Loss 0.270213 LR 0.001000 Time 0.021515 -2022-12-06 11:06:20,508 - Epoch: [98][ 510/ 1200] Overall Loss 0.271002 Objective Loss 0.271002 LR 0.001000 Time 0.021482 -2022-12-06 11:06:20,705 - Epoch: [98][ 520/ 1200] Overall Loss 0.271130 Objective Loss 0.271130 LR 0.001000 Time 0.021447 -2022-12-06 11:06:20,904 - Epoch: [98][ 530/ 1200] Overall Loss 0.271686 Objective Loss 0.271686 LR 0.001000 Time 0.021416 -2022-12-06 11:06:21,100 - Epoch: [98][ 540/ 1200] Overall Loss 0.271519 Objective Loss 0.271519 LR 0.001000 Time 0.021381 -2022-12-06 11:06:21,298 - Epoch: [98][ 550/ 1200] Overall Loss 0.272119 Objective Loss 0.272119 LR 0.001000 Time 0.021351 -2022-12-06 11:06:21,494 - Epoch: [98][ 560/ 1200] Overall Loss 0.272125 Objective Loss 0.272125 LR 0.001000 Time 0.021320 -2022-12-06 11:06:21,692 - Epoch: [98][ 570/ 1200] Overall Loss 0.272667 Objective Loss 0.272667 LR 0.001000 Time 0.021293 -2022-12-06 11:06:21,888 - Epoch: [98][ 580/ 1200] Overall Loss 0.272263 Objective Loss 0.272263 LR 0.001000 Time 0.021263 -2022-12-06 11:06:22,087 - Epoch: [98][ 590/ 1200] Overall Loss 0.272253 Objective Loss 0.272253 LR 0.001000 Time 0.021237 -2022-12-06 11:06:22,283 - Epoch: [98][ 600/ 1200] Overall Loss 0.272728 Objective Loss 0.272728 LR 0.001000 Time 0.021210 -2022-12-06 11:06:22,481 - Epoch: [98][ 610/ 1200] Overall Loss 0.273127 Objective Loss 0.273127 LR 0.001000 Time 0.021186 -2022-12-06 11:06:22,678 - Epoch: [98][ 620/ 1200] Overall Loss 0.273332 Objective Loss 0.273332 LR 0.001000 Time 0.021161 -2022-12-06 11:06:22,876 - Epoch: [98][ 630/ 1200] Overall Loss 0.273193 Objective Loss 0.273193 LR 0.001000 Time 0.021138 -2022-12-06 11:06:23,072 - Epoch: [98][ 640/ 1200] Overall Loss 0.272875 Objective Loss 0.272875 LR 0.001000 Time 0.021113 -2022-12-06 11:06:23,270 - Epoch: [98][ 650/ 1200] Overall Loss 0.272967 Objective Loss 0.272967 LR 0.001000 Time 0.021093 -2022-12-06 11:06:23,467 - Epoch: [98][ 660/ 1200] Overall Loss 0.272975 Objective Loss 0.272975 LR 0.001000 Time 0.021070 -2022-12-06 11:06:23,665 - Epoch: [98][ 670/ 1200] Overall Loss 0.273301 Objective Loss 0.273301 LR 0.001000 Time 0.021050 -2022-12-06 11:06:23,862 - Epoch: [98][ 680/ 1200] Overall Loss 0.273429 Objective Loss 0.273429 LR 0.001000 Time 0.021029 -2022-12-06 11:06:24,060 - Epoch: [98][ 690/ 1200] Overall Loss 0.273758 Objective Loss 0.273758 LR 0.001000 Time 0.021011 -2022-12-06 11:06:24,257 - Epoch: [98][ 700/ 1200] Overall Loss 0.274119 Objective Loss 0.274119 LR 0.001000 Time 0.020991 -2022-12-06 11:06:24,456 - Epoch: [98][ 710/ 1200] Overall Loss 0.273878 Objective Loss 0.273878 LR 0.001000 Time 0.020975 -2022-12-06 11:06:24,652 - Epoch: [98][ 720/ 1200] Overall Loss 0.273809 Objective Loss 0.273809 LR 0.001000 Time 0.020956 -2022-12-06 11:06:24,851 - Epoch: [98][ 730/ 1200] Overall Loss 0.273570 Objective Loss 0.273570 LR 0.001000 Time 0.020941 -2022-12-06 11:06:25,048 - Epoch: [98][ 740/ 1200] Overall Loss 0.273953 Objective Loss 0.273953 LR 0.001000 Time 0.020923 -2022-12-06 11:06:25,247 - Epoch: [98][ 750/ 1200] Overall Loss 0.273785 Objective Loss 0.273785 LR 0.001000 Time 0.020908 -2022-12-06 11:06:25,443 - Epoch: [98][ 760/ 1200] Overall Loss 0.273213 Objective Loss 0.273213 LR 0.001000 Time 0.020891 -2022-12-06 11:06:25,642 - Epoch: [98][ 770/ 1200] Overall Loss 0.273258 Objective Loss 0.273258 LR 0.001000 Time 0.020877 -2022-12-06 11:06:25,839 - Epoch: [98][ 780/ 1200] Overall Loss 0.273102 Objective Loss 0.273102 LR 0.001000 Time 0.020861 -2022-12-06 11:06:26,037 - Epoch: [98][ 790/ 1200] Overall Loss 0.272926 Objective Loss 0.272926 LR 0.001000 Time 0.020847 -2022-12-06 11:06:26,233 - Epoch: [98][ 800/ 1200] Overall Loss 0.272735 Objective Loss 0.272735 LR 0.001000 Time 0.020831 -2022-12-06 11:06:26,432 - Epoch: [98][ 810/ 1200] Overall Loss 0.272829 Objective Loss 0.272829 LR 0.001000 Time 0.020818 -2022-12-06 11:06:26,629 - Epoch: [98][ 820/ 1200] Overall Loss 0.272785 Objective Loss 0.272785 LR 0.001000 Time 0.020804 -2022-12-06 11:06:26,828 - Epoch: [98][ 830/ 1200] Overall Loss 0.273108 Objective Loss 0.273108 LR 0.001000 Time 0.020792 -2022-12-06 11:06:27,025 - Epoch: [98][ 840/ 1200] Overall Loss 0.273512 Objective Loss 0.273512 LR 0.001000 Time 0.020779 -2022-12-06 11:06:27,223 - Epoch: [98][ 850/ 1200] Overall Loss 0.273432 Objective Loss 0.273432 LR 0.001000 Time 0.020767 -2022-12-06 11:06:27,420 - Epoch: [98][ 860/ 1200] Overall Loss 0.273487 Objective Loss 0.273487 LR 0.001000 Time 0.020754 -2022-12-06 11:06:27,619 - Epoch: [98][ 870/ 1200] Overall Loss 0.273577 Objective Loss 0.273577 LR 0.001000 Time 0.020744 -2022-12-06 11:06:27,816 - Epoch: [98][ 880/ 1200] Overall Loss 0.273960 Objective Loss 0.273960 LR 0.001000 Time 0.020731 -2022-12-06 11:06:28,015 - Epoch: [98][ 890/ 1200] Overall Loss 0.273931 Objective Loss 0.273931 LR 0.001000 Time 0.020721 -2022-12-06 11:06:28,211 - Epoch: [98][ 900/ 1200] Overall Loss 0.273764 Objective Loss 0.273764 LR 0.001000 Time 0.020708 -2022-12-06 11:06:28,410 - Epoch: [98][ 910/ 1200] Overall Loss 0.273753 Objective Loss 0.273753 LR 0.001000 Time 0.020699 -2022-12-06 11:06:28,607 - Epoch: [98][ 920/ 1200] Overall Loss 0.273737 Objective Loss 0.273737 LR 0.001000 Time 0.020687 -2022-12-06 11:06:28,805 - Epoch: [98][ 930/ 1200] Overall Loss 0.273942 Objective Loss 0.273942 LR 0.001000 Time 0.020677 -2022-12-06 11:06:29,002 - Epoch: [98][ 940/ 1200] Overall Loss 0.273687 Objective Loss 0.273687 LR 0.001000 Time 0.020666 -2022-12-06 11:06:29,202 - Epoch: [98][ 950/ 1200] Overall Loss 0.273496 Objective Loss 0.273496 LR 0.001000 Time 0.020658 -2022-12-06 11:06:29,399 - Epoch: [98][ 960/ 1200] Overall Loss 0.273620 Objective Loss 0.273620 LR 0.001000 Time 0.020648 -2022-12-06 11:06:29,598 - Epoch: [98][ 970/ 1200] Overall Loss 0.273636 Objective Loss 0.273636 LR 0.001000 Time 0.020639 -2022-12-06 11:06:29,795 - Epoch: [98][ 980/ 1200] Overall Loss 0.273739 Objective Loss 0.273739 LR 0.001000 Time 0.020629 -2022-12-06 11:06:29,994 - Epoch: [98][ 990/ 1200] Overall Loss 0.273706 Objective Loss 0.273706 LR 0.001000 Time 0.020621 -2022-12-06 11:06:30,190 - Epoch: [98][ 1000/ 1200] Overall Loss 0.273803 Objective Loss 0.273803 LR 0.001000 Time 0.020611 -2022-12-06 11:06:30,389 - Epoch: [98][ 1010/ 1200] Overall Loss 0.273413 Objective Loss 0.273413 LR 0.001000 Time 0.020603 -2022-12-06 11:06:30,586 - Epoch: [98][ 1020/ 1200] Overall Loss 0.273398 Objective Loss 0.273398 LR 0.001000 Time 0.020593 -2022-12-06 11:06:30,785 - Epoch: [98][ 1030/ 1200] Overall Loss 0.273356 Objective Loss 0.273356 LR 0.001000 Time 0.020586 -2022-12-06 11:06:30,981 - Epoch: [98][ 1040/ 1200] Overall Loss 0.273267 Objective Loss 0.273267 LR 0.001000 Time 0.020576 -2022-12-06 11:06:31,180 - Epoch: [98][ 1050/ 1200] Overall Loss 0.272952 Objective Loss 0.272952 LR 0.001000 Time 0.020570 -2022-12-06 11:06:31,378 - Epoch: [98][ 1060/ 1200] Overall Loss 0.272976 Objective Loss 0.272976 LR 0.001000 Time 0.020561 -2022-12-06 11:06:31,576 - Epoch: [98][ 1070/ 1200] Overall Loss 0.273107 Objective Loss 0.273107 LR 0.001000 Time 0.020554 -2022-12-06 11:06:31,773 - Epoch: [98][ 1080/ 1200] Overall Loss 0.273236 Objective Loss 0.273236 LR 0.001000 Time 0.020546 -2022-12-06 11:06:31,972 - Epoch: [98][ 1090/ 1200] Overall Loss 0.272999 Objective Loss 0.272999 LR 0.001000 Time 0.020539 -2022-12-06 11:06:32,168 - Epoch: [98][ 1100/ 1200] Overall Loss 0.273241 Objective Loss 0.273241 LR 0.001000 Time 0.020530 -2022-12-06 11:06:32,367 - Epoch: [98][ 1110/ 1200] Overall Loss 0.273213 Objective Loss 0.273213 LR 0.001000 Time 0.020524 -2022-12-06 11:06:32,563 - Epoch: [98][ 1120/ 1200] Overall Loss 0.273216 Objective Loss 0.273216 LR 0.001000 Time 0.020515 -2022-12-06 11:06:32,762 - Epoch: [98][ 1130/ 1200] Overall Loss 0.273165 Objective Loss 0.273165 LR 0.001000 Time 0.020509 -2022-12-06 11:06:32,959 - Epoch: [98][ 1140/ 1200] Overall Loss 0.273408 Objective Loss 0.273408 LR 0.001000 Time 0.020501 -2022-12-06 11:06:33,158 - Epoch: [98][ 1150/ 1200] Overall Loss 0.273349 Objective Loss 0.273349 LR 0.001000 Time 0.020496 -2022-12-06 11:06:33,354 - Epoch: [98][ 1160/ 1200] Overall Loss 0.273417 Objective Loss 0.273417 LR 0.001000 Time 0.020488 -2022-12-06 11:06:33,553 - Epoch: [98][ 1170/ 1200] Overall Loss 0.273570 Objective Loss 0.273570 LR 0.001000 Time 0.020482 -2022-12-06 11:06:33,750 - Epoch: [98][ 1180/ 1200] Overall Loss 0.273579 Objective Loss 0.273579 LR 0.001000 Time 0.020475 -2022-12-06 11:06:33,950 - Epoch: [98][ 1190/ 1200] Overall Loss 0.273376 Objective Loss 0.273376 LR 0.001000 Time 0.020470 -2022-12-06 11:06:34,181 - Epoch: [98][ 1200/ 1200] Overall Loss 0.273596 Objective Loss 0.273596 Top1 82.845188 Top5 97.489540 LR 0.001000 Time 0.020492 -2022-12-06 11:06:34,270 - --- validate (epoch=98)----------- -2022-12-06 11:06:34,270 - 34129 samples (256 per mini-batch) -2022-12-06 11:06:34,713 - Epoch: [98][ 10/ 134] Loss 0.304038 Top1 85.546875 Top5 97.734375 -2022-12-06 11:06:34,846 - Epoch: [98][ 20/ 134] Loss 0.300096 Top1 85.136719 Top5 97.832031 -2022-12-06 11:06:34,971 - Epoch: [98][ 30/ 134] Loss 0.298365 Top1 85.052083 Top5 97.877604 -2022-12-06 11:06:35,095 - Epoch: [98][ 40/ 134] Loss 0.299544 Top1 85.117188 Top5 97.939453 -2022-12-06 11:06:35,223 - Epoch: [98][ 50/ 134] Loss 0.300234 Top1 85.031250 Top5 97.921875 -2022-12-06 11:06:35,350 - Epoch: [98][ 60/ 134] Loss 0.296550 Top1 85.123698 Top5 97.897135 -2022-12-06 11:06:35,475 - Epoch: [98][ 70/ 134] Loss 0.298519 Top1 85.217634 Top5 97.924107 -2022-12-06 11:06:35,603 - Epoch: [98][ 80/ 134] Loss 0.297460 Top1 85.185547 Top5 97.939453 -2022-12-06 11:06:35,730 - Epoch: [98][ 90/ 134] Loss 0.296894 Top1 85.286458 Top5 97.964410 -2022-12-06 11:06:35,856 - Epoch: [98][ 100/ 134] Loss 0.301025 Top1 85.304688 Top5 97.902344 -2022-12-06 11:06:35,983 - Epoch: [98][ 110/ 134] Loss 0.299215 Top1 85.340909 Top5 97.965199 -2022-12-06 11:06:36,111 - Epoch: [98][ 120/ 134] Loss 0.299643 Top1 85.208333 Top5 97.926432 -2022-12-06 11:06:36,238 - Epoch: [98][ 130/ 134] Loss 0.301774 Top1 85.213341 Top5 97.902644 -2022-12-06 11:06:36,274 - Epoch: [98][ 134/ 134] Loss 0.302823 Top1 85.203200 Top5 97.907938 -2022-12-06 11:06:36,362 - ==> Top1: 85.203 Top5: 97.908 Loss: 0.303 - -2022-12-06 11:06:36,363 - ==> Confusion: -[[ 867 2 2 3 10 10 0 1 9 73 0 2 0 1 6 3 0 0 1 0 6] - [ 0 937 1 1 6 17 3 14 1 0 4 7 1 2 3 0 6 4 9 3 8] - [ 5 3 977 19 3 1 23 12 0 5 8 5 2 5 4 6 0 2 4 5 14] - [ 4 0 13 949 1 0 0 1 3 0 11 1 5 2 8 1 3 2 9 1 6] - [ 8 6 2 0 944 5 1 1 0 8 2 5 2 3 9 6 6 3 0 2 7] - [ 3 18 0 1 4 955 5 14 1 2 2 19 4 15 2 4 2 1 2 7 8] - [ 0 3 8 2 0 2 1057 3 0 0 1 4 1 3 0 9 0 3 0 15 7] - [ 1 11 9 0 1 34 10 910 0 1 5 13 0 1 0 1 1 0 24 27 5] - [ 4 6 0 0 1 2 0 0 966 37 12 3 6 9 5 0 4 1 5 2 1] - [ 48 0 2 2 7 2 0 2 35 867 1 3 0 16 4 2 0 1 1 1 7] - [ 1 3 1 4 3 1 2 3 5 0 963 5 5 11 1 0 1 0 4 1 5] - [ 2 0 1 0 0 11 4 3 2 1 1 980 13 2 1 4 7 4 1 9 5] - [ 2 1 3 1 0 3 0 0 0 0 0 54 873 1 0 9 2 12 0 2 6] - [ 1 0 0 0 1 19 0 1 12 17 9 11 4 926 0 4 5 2 0 2 9] - [ 5 3 2 25 4 3 0 1 27 6 4 3 1 4 1019 0 3 0 8 1 11] - [ 0 0 3 0 1 0 1 0 1 0 0 10 9 3 0 991 6 10 0 4 4] - [ 3 5 1 1 1 2 1 0 0 0 0 4 2 1 1 19 1019 1 0 5 6] - [ 2 0 1 2 0 1 1 0 1 3 1 12 22 2 1 14 2 965 1 3 2] - [ 2 5 3 16 1 2 2 24 3 0 8 8 4 1 7 0 0 1 912 1 8] - [ 1 3 1 0 0 3 6 6 1 0 0 12 6 10 1 7 7 4 0 1009 3] - [ 100 232 140 110 100 197 61 138 98 76 208 152 365 275 144 136 233 83 158 230 9990]] - -2022-12-06 11:06:36,934 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:06:36,935 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:06:36,941 - - -2022-12-06 11:06:36,941 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:06:37,870 - Epoch: [99][ 10/ 1200] Overall Loss 0.307002 Objective Loss 0.307002 LR 0.001000 Time 0.092889 -2022-12-06 11:06:38,070 - Epoch: [99][ 20/ 1200] Overall Loss 0.294675 Objective Loss 0.294675 LR 0.001000 Time 0.056407 -2022-12-06 11:06:38,267 - Epoch: [99][ 30/ 1200] Overall Loss 0.288903 Objective Loss 0.288903 LR 0.001000 Time 0.044130 -2022-12-06 11:06:38,464 - Epoch: [99][ 40/ 1200] Overall Loss 0.286362 Objective Loss 0.286362 LR 0.001000 Time 0.038029 -2022-12-06 11:06:38,660 - Epoch: [99][ 50/ 1200] Overall Loss 0.284186 Objective Loss 0.284186 LR 0.001000 Time 0.034331 -2022-12-06 11:06:38,858 - Epoch: [99][ 60/ 1200] Overall Loss 0.285270 Objective Loss 0.285270 LR 0.001000 Time 0.031898 -2022-12-06 11:06:39,054 - Epoch: [99][ 70/ 1200] Overall Loss 0.280587 Objective Loss 0.280587 LR 0.001000 Time 0.030129 -2022-12-06 11:06:39,251 - Epoch: [99][ 80/ 1200] Overall Loss 0.282296 Objective Loss 0.282296 LR 0.001000 Time 0.028819 -2022-12-06 11:06:39,446 - Epoch: [99][ 90/ 1200] Overall Loss 0.282446 Objective Loss 0.282446 LR 0.001000 Time 0.027781 -2022-12-06 11:06:39,644 - Epoch: [99][ 100/ 1200] Overall Loss 0.280018 Objective Loss 0.280018 LR 0.001000 Time 0.026972 -2022-12-06 11:06:39,838 - Epoch: [99][ 110/ 1200] Overall Loss 0.277359 Objective Loss 0.277359 LR 0.001000 Time 0.026283 -2022-12-06 11:06:40,036 - Epoch: [99][ 120/ 1200] Overall Loss 0.278210 Objective Loss 0.278210 LR 0.001000 Time 0.025734 -2022-12-06 11:06:40,231 - Epoch: [99][ 130/ 1200] Overall Loss 0.279566 Objective Loss 0.279566 LR 0.001000 Time 0.025253 -2022-12-06 11:06:40,428 - Epoch: [99][ 140/ 1200] Overall Loss 0.279883 Objective Loss 0.279883 LR 0.001000 Time 0.024853 -2022-12-06 11:06:40,623 - Epoch: [99][ 150/ 1200] Overall Loss 0.279646 Objective Loss 0.279646 LR 0.001000 Time 0.024492 -2022-12-06 11:06:40,820 - Epoch: [99][ 160/ 1200] Overall Loss 0.278345 Objective Loss 0.278345 LR 0.001000 Time 0.024189 -2022-12-06 11:06:41,016 - Epoch: [99][ 170/ 1200] Overall Loss 0.279455 Objective Loss 0.279455 LR 0.001000 Time 0.023913 -2022-12-06 11:06:41,213 - Epoch: [99][ 180/ 1200] Overall Loss 0.279277 Objective Loss 0.279277 LR 0.001000 Time 0.023676 -2022-12-06 11:06:41,409 - Epoch: [99][ 190/ 1200] Overall Loss 0.278733 Objective Loss 0.278733 LR 0.001000 Time 0.023457 -2022-12-06 11:06:41,606 - Epoch: [99][ 200/ 1200] Overall Loss 0.279020 Objective Loss 0.279020 LR 0.001000 Time 0.023267 -2022-12-06 11:06:41,801 - Epoch: [99][ 210/ 1200] Overall Loss 0.277040 Objective Loss 0.277040 LR 0.001000 Time 0.023088 -2022-12-06 11:06:41,999 - Epoch: [99][ 220/ 1200] Overall Loss 0.276667 Objective Loss 0.276667 LR 0.001000 Time 0.022934 -2022-12-06 11:06:42,194 - Epoch: [99][ 230/ 1200] Overall Loss 0.275626 Objective Loss 0.275626 LR 0.001000 Time 0.022783 -2022-12-06 11:06:42,392 - Epoch: [99][ 240/ 1200] Overall Loss 0.276070 Objective Loss 0.276070 LR 0.001000 Time 0.022655 -2022-12-06 11:06:42,588 - Epoch: [99][ 250/ 1200] Overall Loss 0.276811 Objective Loss 0.276811 LR 0.001000 Time 0.022529 -2022-12-06 11:06:42,784 - Epoch: [99][ 260/ 1200] Overall Loss 0.277154 Objective Loss 0.277154 LR 0.001000 Time 0.022417 -2022-12-06 11:06:42,979 - Epoch: [99][ 270/ 1200] Overall Loss 0.277994 Objective Loss 0.277994 LR 0.001000 Time 0.022307 -2022-12-06 11:06:43,176 - Epoch: [99][ 280/ 1200] Overall Loss 0.278075 Objective Loss 0.278075 LR 0.001000 Time 0.022211 -2022-12-06 11:06:43,372 - Epoch: [99][ 290/ 1200] Overall Loss 0.277370 Objective Loss 0.277370 LR 0.001000 Time 0.022118 -2022-12-06 11:06:43,569 - Epoch: [99][ 300/ 1200] Overall Loss 0.275915 Objective Loss 0.275915 LR 0.001000 Time 0.022037 -2022-12-06 11:06:43,765 - Epoch: [99][ 310/ 1200] Overall Loss 0.276785 Objective Loss 0.276785 LR 0.001000 Time 0.021956 -2022-12-06 11:06:43,958 - Epoch: [99][ 320/ 1200] Overall Loss 0.276971 Objective Loss 0.276971 LR 0.001000 Time 0.021873 -2022-12-06 11:06:44,149 - Epoch: [99][ 330/ 1200] Overall Loss 0.277401 Objective Loss 0.277401 LR 0.001000 Time 0.021787 -2022-12-06 11:06:44,340 - Epoch: [99][ 340/ 1200] Overall Loss 0.277112 Objective Loss 0.277112 LR 0.001000 Time 0.021705 -2022-12-06 11:06:44,531 - Epoch: [99][ 350/ 1200] Overall Loss 0.276658 Objective Loss 0.276658 LR 0.001000 Time 0.021628 -2022-12-06 11:06:44,722 - Epoch: [99][ 360/ 1200] Overall Loss 0.276277 Objective Loss 0.276277 LR 0.001000 Time 0.021556 -2022-12-06 11:06:44,913 - Epoch: [99][ 370/ 1200] Overall Loss 0.276946 Objective Loss 0.276946 LR 0.001000 Time 0.021489 -2022-12-06 11:06:45,104 - Epoch: [99][ 380/ 1200] Overall Loss 0.276591 Objective Loss 0.276591 LR 0.001000 Time 0.021424 -2022-12-06 11:06:45,295 - Epoch: [99][ 390/ 1200] Overall Loss 0.277227 Objective Loss 0.277227 LR 0.001000 Time 0.021363 -2022-12-06 11:06:45,486 - Epoch: [99][ 400/ 1200] Overall Loss 0.277632 Objective Loss 0.277632 LR 0.001000 Time 0.021305 -2022-12-06 11:06:45,677 - Epoch: [99][ 410/ 1200] Overall Loss 0.277442 Objective Loss 0.277442 LR 0.001000 Time 0.021251 -2022-12-06 11:06:45,868 - Epoch: [99][ 420/ 1200] Overall Loss 0.277535 Objective Loss 0.277535 LR 0.001000 Time 0.021198 -2022-12-06 11:06:46,059 - Epoch: [99][ 430/ 1200] Overall Loss 0.276970 Objective Loss 0.276970 LR 0.001000 Time 0.021149 -2022-12-06 11:06:46,251 - Epoch: [99][ 440/ 1200] Overall Loss 0.277080 Objective Loss 0.277080 LR 0.001000 Time 0.021102 -2022-12-06 11:06:46,442 - Epoch: [99][ 450/ 1200] Overall Loss 0.277745 Objective Loss 0.277745 LR 0.001000 Time 0.021056 -2022-12-06 11:06:46,633 - Epoch: [99][ 460/ 1200] Overall Loss 0.277146 Objective Loss 0.277146 LR 0.001000 Time 0.021013 -2022-12-06 11:06:46,824 - Epoch: [99][ 470/ 1200] Overall Loss 0.277477 Objective Loss 0.277477 LR 0.001000 Time 0.020972 -2022-12-06 11:06:47,015 - Epoch: [99][ 480/ 1200] Overall Loss 0.277909 Objective Loss 0.277909 LR 0.001000 Time 0.020931 -2022-12-06 11:06:47,206 - Epoch: [99][ 490/ 1200] Overall Loss 0.278169 Objective Loss 0.278169 LR 0.001000 Time 0.020893 -2022-12-06 11:06:47,398 - Epoch: [99][ 500/ 1200] Overall Loss 0.278161 Objective Loss 0.278161 LR 0.001000 Time 0.020857 -2022-12-06 11:06:47,589 - Epoch: [99][ 510/ 1200] Overall Loss 0.277699 Objective Loss 0.277699 LR 0.001000 Time 0.020822 -2022-12-06 11:06:47,780 - Epoch: [99][ 520/ 1200] Overall Loss 0.277856 Objective Loss 0.277856 LR 0.001000 Time 0.020788 -2022-12-06 11:06:47,971 - Epoch: [99][ 530/ 1200] Overall Loss 0.277835 Objective Loss 0.277835 LR 0.001000 Time 0.020755 -2022-12-06 11:06:48,162 - Epoch: [99][ 540/ 1200] Overall Loss 0.278088 Objective Loss 0.278088 LR 0.001000 Time 0.020724 -2022-12-06 11:06:48,353 - Epoch: [99][ 550/ 1200] Overall Loss 0.278074 Objective Loss 0.278074 LR 0.001000 Time 0.020692 -2022-12-06 11:06:48,544 - Epoch: [99][ 560/ 1200] Overall Loss 0.278729 Objective Loss 0.278729 LR 0.001000 Time 0.020663 -2022-12-06 11:06:48,736 - Epoch: [99][ 570/ 1200] Overall Loss 0.278642 Objective Loss 0.278642 LR 0.001000 Time 0.020636 -2022-12-06 11:06:48,927 - Epoch: [99][ 580/ 1200] Overall Loss 0.278647 Objective Loss 0.278647 LR 0.001000 Time 0.020610 -2022-12-06 11:06:49,119 - Epoch: [99][ 590/ 1200] Overall Loss 0.278135 Objective Loss 0.278135 LR 0.001000 Time 0.020584 -2022-12-06 11:06:49,310 - Epoch: [99][ 600/ 1200] Overall Loss 0.278377 Objective Loss 0.278377 LR 0.001000 Time 0.020559 -2022-12-06 11:06:49,501 - Epoch: [99][ 610/ 1200] Overall Loss 0.278260 Objective Loss 0.278260 LR 0.001000 Time 0.020534 -2022-12-06 11:06:49,692 - Epoch: [99][ 620/ 1200] Overall Loss 0.278227 Objective Loss 0.278227 LR 0.001000 Time 0.020510 -2022-12-06 11:06:49,884 - Epoch: [99][ 630/ 1200] Overall Loss 0.277712 Objective Loss 0.277712 LR 0.001000 Time 0.020488 -2022-12-06 11:06:50,075 - Epoch: [99][ 640/ 1200] Overall Loss 0.277535 Objective Loss 0.277535 LR 0.001000 Time 0.020466 -2022-12-06 11:06:50,265 - Epoch: [99][ 650/ 1200] Overall Loss 0.277241 Objective Loss 0.277241 LR 0.001000 Time 0.020443 -2022-12-06 11:06:50,456 - Epoch: [99][ 660/ 1200] Overall Loss 0.276908 Objective Loss 0.276908 LR 0.001000 Time 0.020421 -2022-12-06 11:06:50,647 - Epoch: [99][ 670/ 1200] Overall Loss 0.276769 Objective Loss 0.276769 LR 0.001000 Time 0.020401 -2022-12-06 11:06:50,839 - Epoch: [99][ 680/ 1200] Overall Loss 0.276796 Objective Loss 0.276796 LR 0.001000 Time 0.020383 -2022-12-06 11:06:51,030 - Epoch: [99][ 690/ 1200] Overall Loss 0.276493 Objective Loss 0.276493 LR 0.001000 Time 0.020363 -2022-12-06 11:06:51,223 - Epoch: [99][ 700/ 1200] Overall Loss 0.276818 Objective Loss 0.276818 LR 0.001000 Time 0.020346 -2022-12-06 11:06:51,413 - Epoch: [99][ 710/ 1200] Overall Loss 0.277207 Objective Loss 0.277207 LR 0.001000 Time 0.020327 -2022-12-06 11:06:51,605 - Epoch: [99][ 720/ 1200] Overall Loss 0.277644 Objective Loss 0.277644 LR 0.001000 Time 0.020310 -2022-12-06 11:06:51,795 - Epoch: [99][ 730/ 1200] Overall Loss 0.277535 Objective Loss 0.277535 LR 0.001000 Time 0.020292 -2022-12-06 11:06:51,987 - Epoch: [99][ 740/ 1200] Overall Loss 0.277133 Objective Loss 0.277133 LR 0.001000 Time 0.020277 -2022-12-06 11:06:52,178 - Epoch: [99][ 750/ 1200] Overall Loss 0.277746 Objective Loss 0.277746 LR 0.001000 Time 0.020260 -2022-12-06 11:06:52,370 - Epoch: [99][ 760/ 1200] Overall Loss 0.277605 Objective Loss 0.277605 LR 0.001000 Time 0.020245 -2022-12-06 11:06:52,561 - Epoch: [99][ 770/ 1200] Overall Loss 0.277353 Objective Loss 0.277353 LR 0.001000 Time 0.020230 -2022-12-06 11:06:52,753 - Epoch: [99][ 780/ 1200] Overall Loss 0.277254 Objective Loss 0.277254 LR 0.001000 Time 0.020215 -2022-12-06 11:06:52,944 - Epoch: [99][ 790/ 1200] Overall Loss 0.277451 Objective Loss 0.277451 LR 0.001000 Time 0.020200 -2022-12-06 11:06:53,135 - Epoch: [99][ 800/ 1200] Overall Loss 0.277603 Objective Loss 0.277603 LR 0.001000 Time 0.020186 -2022-12-06 11:06:53,326 - Epoch: [99][ 810/ 1200] Overall Loss 0.277727 Objective Loss 0.277727 LR 0.001000 Time 0.020172 -2022-12-06 11:06:53,517 - Epoch: [99][ 820/ 1200] Overall Loss 0.277774 Objective Loss 0.277774 LR 0.001000 Time 0.020159 -2022-12-06 11:06:53,709 - Epoch: [99][ 830/ 1200] Overall Loss 0.277832 Objective Loss 0.277832 LR 0.001000 Time 0.020146 -2022-12-06 11:06:53,900 - Epoch: [99][ 840/ 1200] Overall Loss 0.277888 Objective Loss 0.277888 LR 0.001000 Time 0.020134 -2022-12-06 11:06:54,092 - Epoch: [99][ 850/ 1200] Overall Loss 0.277698 Objective Loss 0.277698 LR 0.001000 Time 0.020122 -2022-12-06 11:06:54,284 - Epoch: [99][ 860/ 1200] Overall Loss 0.277839 Objective Loss 0.277839 LR 0.001000 Time 0.020110 -2022-12-06 11:06:54,475 - Epoch: [99][ 870/ 1200] Overall Loss 0.277936 Objective Loss 0.277936 LR 0.001000 Time 0.020098 -2022-12-06 11:06:54,666 - Epoch: [99][ 880/ 1200] Overall Loss 0.278238 Objective Loss 0.278238 LR 0.001000 Time 0.020086 -2022-12-06 11:06:54,857 - Epoch: [99][ 890/ 1200] Overall Loss 0.278118 Objective Loss 0.278118 LR 0.001000 Time 0.020075 -2022-12-06 11:06:55,049 - Epoch: [99][ 900/ 1200] Overall Loss 0.278359 Objective Loss 0.278359 LR 0.001000 Time 0.020064 -2022-12-06 11:06:55,241 - Epoch: [99][ 910/ 1200] Overall Loss 0.278637 Objective Loss 0.278637 LR 0.001000 Time 0.020053 -2022-12-06 11:06:55,431 - Epoch: [99][ 920/ 1200] Overall Loss 0.278372 Objective Loss 0.278372 LR 0.001000 Time 0.020042 -2022-12-06 11:06:55,623 - Epoch: [99][ 930/ 1200] Overall Loss 0.278517 Objective Loss 0.278517 LR 0.001000 Time 0.020032 -2022-12-06 11:06:55,815 - Epoch: [99][ 940/ 1200] Overall Loss 0.278293 Objective Loss 0.278293 LR 0.001000 Time 0.020023 -2022-12-06 11:06:56,005 - Epoch: [99][ 950/ 1200] Overall Loss 0.278459 Objective Loss 0.278459 LR 0.001000 Time 0.020012 -2022-12-06 11:06:56,197 - Epoch: [99][ 960/ 1200] Overall Loss 0.278451 Objective Loss 0.278451 LR 0.001000 Time 0.020003 -2022-12-06 11:06:56,389 - Epoch: [99][ 970/ 1200] Overall Loss 0.278414 Objective Loss 0.278414 LR 0.001000 Time 0.019993 -2022-12-06 11:06:56,580 - Epoch: [99][ 980/ 1200] Overall Loss 0.278286 Objective Loss 0.278286 LR 0.001000 Time 0.019984 -2022-12-06 11:06:56,772 - Epoch: [99][ 990/ 1200] Overall Loss 0.278043 Objective Loss 0.278043 LR 0.001000 Time 0.019975 -2022-12-06 11:06:56,964 - Epoch: [99][ 1000/ 1200] Overall Loss 0.277991 Objective Loss 0.277991 LR 0.001000 Time 0.019967 -2022-12-06 11:06:57,156 - Epoch: [99][ 1010/ 1200] Overall Loss 0.277820 Objective Loss 0.277820 LR 0.001000 Time 0.019959 -2022-12-06 11:06:57,348 - Epoch: [99][ 1020/ 1200] Overall Loss 0.277785 Objective Loss 0.277785 LR 0.001000 Time 0.019951 -2022-12-06 11:06:57,538 - Epoch: [99][ 1030/ 1200] Overall Loss 0.277753 Objective Loss 0.277753 LR 0.001000 Time 0.019942 -2022-12-06 11:06:57,730 - Epoch: [99][ 1040/ 1200] Overall Loss 0.277636 Objective Loss 0.277636 LR 0.001000 Time 0.019934 -2022-12-06 11:06:57,921 - Epoch: [99][ 1050/ 1200] Overall Loss 0.277215 Objective Loss 0.277215 LR 0.001000 Time 0.019925 -2022-12-06 11:06:58,113 - Epoch: [99][ 1060/ 1200] Overall Loss 0.276706 Objective Loss 0.276706 LR 0.001000 Time 0.019917 -2022-12-06 11:06:58,304 - Epoch: [99][ 1070/ 1200] Overall Loss 0.276672 Objective Loss 0.276672 LR 0.001000 Time 0.019909 -2022-12-06 11:06:58,495 - Epoch: [99][ 1080/ 1200] Overall Loss 0.276786 Objective Loss 0.276786 LR 0.001000 Time 0.019901 -2022-12-06 11:06:58,686 - Epoch: [99][ 1090/ 1200] Overall Loss 0.276606 Objective Loss 0.276606 LR 0.001000 Time 0.019894 -2022-12-06 11:06:58,877 - Epoch: [99][ 1100/ 1200] Overall Loss 0.276509 Objective Loss 0.276509 LR 0.001000 Time 0.019886 -2022-12-06 11:06:59,069 - Epoch: [99][ 1110/ 1200] Overall Loss 0.276417 Objective Loss 0.276417 LR 0.001000 Time 0.019879 -2022-12-06 11:06:59,260 - Epoch: [99][ 1120/ 1200] Overall Loss 0.276389 Objective Loss 0.276389 LR 0.001000 Time 0.019872 -2022-12-06 11:06:59,451 - Epoch: [99][ 1130/ 1200] Overall Loss 0.276593 Objective Loss 0.276593 LR 0.001000 Time 0.019865 -2022-12-06 11:06:59,643 - Epoch: [99][ 1140/ 1200] Overall Loss 0.276524 Objective Loss 0.276524 LR 0.001000 Time 0.019858 -2022-12-06 11:06:59,834 - Epoch: [99][ 1150/ 1200] Overall Loss 0.276666 Objective Loss 0.276666 LR 0.001000 Time 0.019851 -2022-12-06 11:07:00,026 - Epoch: [99][ 1160/ 1200] Overall Loss 0.276324 Objective Loss 0.276324 LR 0.001000 Time 0.019845 -2022-12-06 11:07:00,217 - Epoch: [99][ 1170/ 1200] Overall Loss 0.276371 Objective Loss 0.276371 LR 0.001000 Time 0.019838 -2022-12-06 11:07:00,409 - Epoch: [99][ 1180/ 1200] Overall Loss 0.276507 Objective Loss 0.276507 LR 0.001000 Time 0.019832 -2022-12-06 11:07:00,600 - Epoch: [99][ 1190/ 1200] Overall Loss 0.276645 Objective Loss 0.276645 LR 0.001000 Time 0.019826 -2022-12-06 11:07:00,829 - Epoch: [99][ 1200/ 1200] Overall Loss 0.276519 Objective Loss 0.276519 Top1 85.564854 Top5 98.326360 LR 0.001000 Time 0.019851 -2022-12-06 11:07:00,919 - --- validate (epoch=99)----------- -2022-12-06 11:07:00,919 - 34129 samples (256 per mini-batch) -2022-12-06 11:07:01,369 - Epoch: [99][ 10/ 134] Loss 0.308863 Top1 84.179688 Top5 97.890625 -2022-12-06 11:07:01,499 - Epoch: [99][ 20/ 134] Loss 0.313741 Top1 83.730469 Top5 97.832031 -2022-12-06 11:07:01,628 - Epoch: [99][ 30/ 134] Loss 0.309959 Top1 83.932292 Top5 97.760417 -2022-12-06 11:07:01,757 - Epoch: [99][ 40/ 134] Loss 0.298534 Top1 84.199219 Top5 97.910156 -2022-12-06 11:07:01,885 - Epoch: [99][ 50/ 134] Loss 0.296884 Top1 84.101562 Top5 97.875000 -2022-12-06 11:07:02,015 - Epoch: [99][ 60/ 134] Loss 0.293104 Top1 84.316406 Top5 97.864583 -2022-12-06 11:07:02,141 - Epoch: [99][ 70/ 134] Loss 0.294131 Top1 84.140625 Top5 97.868304 -2022-12-06 11:07:02,269 - Epoch: [99][ 80/ 134] Loss 0.298334 Top1 84.042969 Top5 97.768555 -2022-12-06 11:07:02,395 - Epoch: [99][ 90/ 134] Loss 0.296966 Top1 84.088542 Top5 97.782118 -2022-12-06 11:07:02,535 - Epoch: [99][ 100/ 134] Loss 0.299928 Top1 84.074219 Top5 97.765625 -2022-12-06 11:07:02,674 - Epoch: [99][ 110/ 134] Loss 0.302256 Top1 84.062500 Top5 97.730824 -2022-12-06 11:07:02,820 - Epoch: [99][ 120/ 134] Loss 0.301972 Top1 84.046224 Top5 97.708333 -2022-12-06 11:07:02,953 - Epoch: [99][ 130/ 134] Loss 0.301558 Top1 84.083534 Top5 97.719351 -2022-12-06 11:07:02,990 - Epoch: [99][ 134/ 134] Loss 0.301650 Top1 84.092707 Top5 97.735064 -2022-12-06 11:07:03,077 - ==> Top1: 84.093 Top5: 97.735 Loss: 0.302 - -2022-12-06 11:07:03,078 - ==> Confusion: -[[ 921 2 1 4 7 4 0 0 4 39 0 1 0 3 4 1 0 0 2 1 2] - [ 2 921 2 2 9 21 2 20 1 0 1 6 2 2 4 2 4 2 9 5 10] - [ 9 1 1004 8 3 1 21 10 0 3 6 6 1 1 3 2 1 0 4 3 16] - [ 3 2 23 930 0 4 0 2 1 1 7 0 4 3 17 1 0 4 11 0 7] - [ 13 9 2 0 943 9 1 1 0 5 2 1 0 5 9 8 6 3 0 1 2] - [ 2 19 1 2 7 962 6 21 2 3 0 12 3 18 2 2 1 0 0 4 2] - [ 3 4 16 3 0 2 1059 5 0 0 1 0 0 2 0 7 0 0 3 9 4] - [ 2 13 5 2 2 29 6 943 0 0 4 5 1 1 0 1 0 0 19 15 6] - [ 9 6 0 1 1 3 0 0 906 75 9 2 1 14 27 1 3 0 2 1 3] - [ 90 0 2 0 6 4 0 1 12 856 1 1 0 15 3 3 1 0 0 1 5] - [ 1 1 5 8 5 2 1 3 11 2 925 1 2 22 10 0 0 0 10 5 5] - [ 5 1 2 0 1 17 3 5 2 0 0 966 14 5 1 7 4 3 0 15 0] - [ 0 2 3 6 0 3 1 0 0 0 1 41 880 2 0 11 2 7 0 4 6] - [ 0 2 0 0 1 12 0 5 5 13 5 5 3 958 2 2 1 0 0 4 5] - [ 10 3 3 14 4 3 0 0 14 2 0 5 3 2 1050 0 3 0 6 1 7] - [ 1 1 2 0 2 2 3 1 0 0 0 8 3 4 1 994 9 5 0 4 3] - [ 5 4 1 2 3 1 1 0 0 0 0 1 0 1 0 12 1035 0 0 4 2] - [ 6 0 3 3 0 1 1 2 0 2 1 8 29 6 1 25 1 938 1 5 3] - [ 4 6 4 13 3 1 2 22 1 0 5 1 5 0 10 1 2 0 921 3 4] - [ 3 4 1 1 1 5 6 9 0 0 2 13 6 6 0 1 5 2 0 1011 4] - [ 189 189 213 120 161 239 94 159 72 107 144 111 363 349 195 154 259 50 165 325 9568]] - -2022-12-06 11:07:03,748 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:07:03,748 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:07:03,753 - - -2022-12-06 11:07:03,754 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:07:04,700 - Epoch: [100][ 10/ 1200] Overall Loss 0.250618 Objective Loss 0.250618 LR 0.000500 Time 0.094622 -2022-12-06 11:07:04,899 - Epoch: [100][ 20/ 1200] Overall Loss 0.283330 Objective Loss 0.283330 LR 0.000500 Time 0.057186 -2022-12-06 11:07:05,090 - Epoch: [100][ 30/ 1200] Overall Loss 0.283977 Objective Loss 0.283977 LR 0.000500 Time 0.044483 -2022-12-06 11:07:05,281 - Epoch: [100][ 40/ 1200] Overall Loss 0.273411 Objective Loss 0.273411 LR 0.000500 Time 0.038129 -2022-12-06 11:07:05,471 - Epoch: [100][ 50/ 1200] Overall Loss 0.267221 Objective Loss 0.267221 LR 0.000500 Time 0.034293 -2022-12-06 11:07:05,662 - Epoch: [100][ 60/ 1200] Overall Loss 0.267765 Objective Loss 0.267765 LR 0.000500 Time 0.031745 -2022-12-06 11:07:05,851 - Epoch: [100][ 70/ 1200] Overall Loss 0.264947 Objective Loss 0.264947 LR 0.000500 Time 0.029908 -2022-12-06 11:07:06,041 - Epoch: [100][ 80/ 1200] Overall Loss 0.263396 Objective Loss 0.263396 LR 0.000500 Time 0.028537 -2022-12-06 11:07:06,231 - Epoch: [100][ 90/ 1200] Overall Loss 0.264825 Objective Loss 0.264825 LR 0.000500 Time 0.027469 -2022-12-06 11:07:06,421 - Epoch: [100][ 100/ 1200] Overall Loss 0.260572 Objective Loss 0.260572 LR 0.000500 Time 0.026620 -2022-12-06 11:07:06,612 - Epoch: [100][ 110/ 1200] Overall Loss 0.258494 Objective Loss 0.258494 LR 0.000500 Time 0.025927 -2022-12-06 11:07:06,802 - Epoch: [100][ 120/ 1200] Overall Loss 0.259209 Objective Loss 0.259209 LR 0.000500 Time 0.025349 -2022-12-06 11:07:06,992 - Epoch: [100][ 130/ 1200] Overall Loss 0.256713 Objective Loss 0.256713 LR 0.000500 Time 0.024857 -2022-12-06 11:07:07,184 - Epoch: [100][ 140/ 1200] Overall Loss 0.255095 Objective Loss 0.255095 LR 0.000500 Time 0.024445 -2022-12-06 11:07:07,374 - Epoch: [100][ 150/ 1200] Overall Loss 0.255708 Objective Loss 0.255708 LR 0.000500 Time 0.024080 -2022-12-06 11:07:07,564 - Epoch: [100][ 160/ 1200] Overall Loss 0.255885 Objective Loss 0.255885 LR 0.000500 Time 0.023761 -2022-12-06 11:07:07,756 - Epoch: [100][ 170/ 1200] Overall Loss 0.255729 Objective Loss 0.255729 LR 0.000500 Time 0.023486 -2022-12-06 11:07:07,946 - Epoch: [100][ 180/ 1200] Overall Loss 0.253992 Objective Loss 0.253992 LR 0.000500 Time 0.023235 -2022-12-06 11:07:08,136 - Epoch: [100][ 190/ 1200] Overall Loss 0.253261 Objective Loss 0.253261 LR 0.000500 Time 0.023010 -2022-12-06 11:07:08,327 - Epoch: [100][ 200/ 1200] Overall Loss 0.253034 Objective Loss 0.253034 LR 0.000500 Time 0.022810 -2022-12-06 11:07:08,517 - Epoch: [100][ 210/ 1200] Overall Loss 0.252609 Objective Loss 0.252609 LR 0.000500 Time 0.022629 -2022-12-06 11:07:08,707 - Epoch: [100][ 220/ 1200] Overall Loss 0.250957 Objective Loss 0.250957 LR 0.000500 Time 0.022462 -2022-12-06 11:07:08,897 - Epoch: [100][ 230/ 1200] Overall Loss 0.249687 Objective Loss 0.249687 LR 0.000500 Time 0.022310 -2022-12-06 11:07:09,088 - Epoch: [100][ 240/ 1200] Overall Loss 0.249159 Objective Loss 0.249159 LR 0.000500 Time 0.022171 -2022-12-06 11:07:09,279 - Epoch: [100][ 250/ 1200] Overall Loss 0.248393 Objective Loss 0.248393 LR 0.000500 Time 0.022046 -2022-12-06 11:07:09,470 - Epoch: [100][ 260/ 1200] Overall Loss 0.248142 Objective Loss 0.248142 LR 0.000500 Time 0.021929 -2022-12-06 11:07:09,660 - Epoch: [100][ 270/ 1200] Overall Loss 0.247604 Objective Loss 0.247604 LR 0.000500 Time 0.021820 -2022-12-06 11:07:09,850 - Epoch: [100][ 280/ 1200] Overall Loss 0.247592 Objective Loss 0.247592 LR 0.000500 Time 0.021718 -2022-12-06 11:07:10,040 - Epoch: [100][ 290/ 1200] Overall Loss 0.246945 Objective Loss 0.246945 LR 0.000500 Time 0.021622 -2022-12-06 11:07:10,231 - Epoch: [100][ 300/ 1200] Overall Loss 0.247077 Objective Loss 0.247077 LR 0.000500 Time 0.021535 -2022-12-06 11:07:10,421 - Epoch: [100][ 310/ 1200] Overall Loss 0.247612 Objective Loss 0.247612 LR 0.000500 Time 0.021453 -2022-12-06 11:07:10,611 - Epoch: [100][ 320/ 1200] Overall Loss 0.248239 Objective Loss 0.248239 LR 0.000500 Time 0.021375 -2022-12-06 11:07:10,802 - Epoch: [100][ 330/ 1200] Overall Loss 0.247788 Objective Loss 0.247788 LR 0.000500 Time 0.021303 -2022-12-06 11:07:10,992 - Epoch: [100][ 340/ 1200] Overall Loss 0.246722 Objective Loss 0.246722 LR 0.000500 Time 0.021235 -2022-12-06 11:07:11,183 - Epoch: [100][ 350/ 1200] Overall Loss 0.246358 Objective Loss 0.246358 LR 0.000500 Time 0.021171 -2022-12-06 11:07:11,373 - Epoch: [100][ 360/ 1200] Overall Loss 0.246273 Objective Loss 0.246273 LR 0.000500 Time 0.021110 -2022-12-06 11:07:11,564 - Epoch: [100][ 370/ 1200] Overall Loss 0.246555 Objective Loss 0.246555 LR 0.000500 Time 0.021053 -2022-12-06 11:07:11,754 - Epoch: [100][ 380/ 1200] Overall Loss 0.247028 Objective Loss 0.247028 LR 0.000500 Time 0.020998 -2022-12-06 11:07:11,944 - Epoch: [100][ 390/ 1200] Overall Loss 0.246459 Objective Loss 0.246459 LR 0.000500 Time 0.020947 -2022-12-06 11:07:12,135 - Epoch: [100][ 400/ 1200] Overall Loss 0.245751 Objective Loss 0.245751 LR 0.000500 Time 0.020898 -2022-12-06 11:07:12,325 - Epoch: [100][ 410/ 1200] Overall Loss 0.245687 Objective Loss 0.245687 LR 0.000500 Time 0.020851 -2022-12-06 11:07:12,515 - Epoch: [100][ 420/ 1200] Overall Loss 0.245981 Objective Loss 0.245981 LR 0.000500 Time 0.020807 -2022-12-06 11:07:12,706 - Epoch: [100][ 430/ 1200] Overall Loss 0.245974 Objective Loss 0.245974 LR 0.000500 Time 0.020765 -2022-12-06 11:07:12,897 - Epoch: [100][ 440/ 1200] Overall Loss 0.245841 Objective Loss 0.245841 LR 0.000500 Time 0.020724 -2022-12-06 11:07:13,087 - Epoch: [100][ 450/ 1200] Overall Loss 0.245401 Objective Loss 0.245401 LR 0.000500 Time 0.020686 -2022-12-06 11:07:13,277 - Epoch: [100][ 460/ 1200] Overall Loss 0.244566 Objective Loss 0.244566 LR 0.000500 Time 0.020648 -2022-12-06 11:07:13,468 - Epoch: [100][ 470/ 1200] Overall Loss 0.244407 Objective Loss 0.244407 LR 0.000500 Time 0.020613 -2022-12-06 11:07:13,658 - Epoch: [100][ 480/ 1200] Overall Loss 0.244225 Objective Loss 0.244225 LR 0.000500 Time 0.020580 -2022-12-06 11:07:13,848 - Epoch: [100][ 490/ 1200] Overall Loss 0.244572 Objective Loss 0.244572 LR 0.000500 Time 0.020546 -2022-12-06 11:07:14,039 - Epoch: [100][ 500/ 1200] Overall Loss 0.245116 Objective Loss 0.245116 LR 0.000500 Time 0.020515 -2022-12-06 11:07:14,230 - Epoch: [100][ 510/ 1200] Overall Loss 0.244557 Objective Loss 0.244557 LR 0.000500 Time 0.020486 -2022-12-06 11:07:14,421 - Epoch: [100][ 520/ 1200] Overall Loss 0.244692 Objective Loss 0.244692 LR 0.000500 Time 0.020458 -2022-12-06 11:07:14,611 - Epoch: [100][ 530/ 1200] Overall Loss 0.244568 Objective Loss 0.244568 LR 0.000500 Time 0.020430 -2022-12-06 11:07:14,801 - Epoch: [100][ 540/ 1200] Overall Loss 0.243785 Objective Loss 0.243785 LR 0.000500 Time 0.020404 -2022-12-06 11:07:14,992 - Epoch: [100][ 550/ 1200] Overall Loss 0.243577 Objective Loss 0.243577 LR 0.000500 Time 0.020378 -2022-12-06 11:07:15,182 - Epoch: [100][ 560/ 1200] Overall Loss 0.243468 Objective Loss 0.243468 LR 0.000500 Time 0.020353 -2022-12-06 11:07:15,373 - Epoch: [100][ 570/ 1200] Overall Loss 0.243503 Objective Loss 0.243503 LR 0.000500 Time 0.020330 -2022-12-06 11:07:15,563 - Epoch: [100][ 580/ 1200] Overall Loss 0.243457 Objective Loss 0.243457 LR 0.000500 Time 0.020307 -2022-12-06 11:07:15,754 - Epoch: [100][ 590/ 1200] Overall Loss 0.242993 Objective Loss 0.242993 LR 0.000500 Time 0.020284 -2022-12-06 11:07:15,944 - Epoch: [100][ 600/ 1200] Overall Loss 0.242237 Objective Loss 0.242237 LR 0.000500 Time 0.020262 -2022-12-06 11:07:16,134 - Epoch: [100][ 610/ 1200] Overall Loss 0.242283 Objective Loss 0.242283 LR 0.000500 Time 0.020241 -2022-12-06 11:07:16,324 - Epoch: [100][ 620/ 1200] Overall Loss 0.241794 Objective Loss 0.241794 LR 0.000500 Time 0.020220 -2022-12-06 11:07:16,515 - Epoch: [100][ 630/ 1200] Overall Loss 0.241488 Objective Loss 0.241488 LR 0.000500 Time 0.020201 -2022-12-06 11:07:16,705 - Epoch: [100][ 640/ 1200] Overall Loss 0.241649 Objective Loss 0.241649 LR 0.000500 Time 0.020181 -2022-12-06 11:07:16,895 - Epoch: [100][ 650/ 1200] Overall Loss 0.241883 Objective Loss 0.241883 LR 0.000500 Time 0.020162 -2022-12-06 11:07:17,086 - Epoch: [100][ 660/ 1200] Overall Loss 0.241910 Objective Loss 0.241910 LR 0.000500 Time 0.020145 -2022-12-06 11:07:17,276 - Epoch: [100][ 670/ 1200] Overall Loss 0.241602 Objective Loss 0.241602 LR 0.000500 Time 0.020128 -2022-12-06 11:07:17,467 - Epoch: [100][ 680/ 1200] Overall Loss 0.241267 Objective Loss 0.241267 LR 0.000500 Time 0.020111 -2022-12-06 11:07:17,657 - Epoch: [100][ 690/ 1200] Overall Loss 0.241084 Objective Loss 0.241084 LR 0.000500 Time 0.020095 -2022-12-06 11:07:17,847 - Epoch: [100][ 700/ 1200] Overall Loss 0.241044 Objective Loss 0.241044 LR 0.000500 Time 0.020079 -2022-12-06 11:07:18,038 - Epoch: [100][ 710/ 1200] Overall Loss 0.241012 Objective Loss 0.241012 LR 0.000500 Time 0.020064 -2022-12-06 11:07:18,228 - Epoch: [100][ 720/ 1200] Overall Loss 0.240785 Objective Loss 0.240785 LR 0.000500 Time 0.020048 -2022-12-06 11:07:18,418 - Epoch: [100][ 730/ 1200] Overall Loss 0.240782 Objective Loss 0.240782 LR 0.000500 Time 0.020034 -2022-12-06 11:07:18,608 - Epoch: [100][ 740/ 1200] Overall Loss 0.240775 Objective Loss 0.240775 LR 0.000500 Time 0.020019 -2022-12-06 11:07:18,799 - Epoch: [100][ 750/ 1200] Overall Loss 0.240373 Objective Loss 0.240373 LR 0.000500 Time 0.020005 -2022-12-06 11:07:18,989 - Epoch: [100][ 760/ 1200] Overall Loss 0.240555 Objective Loss 0.240555 LR 0.000500 Time 0.019992 -2022-12-06 11:07:19,179 - Epoch: [100][ 770/ 1200] Overall Loss 0.240291 Objective Loss 0.240291 LR 0.000500 Time 0.019978 -2022-12-06 11:07:19,369 - Epoch: [100][ 780/ 1200] Overall Loss 0.240315 Objective Loss 0.240315 LR 0.000500 Time 0.019965 -2022-12-06 11:07:19,560 - Epoch: [100][ 790/ 1200] Overall Loss 0.239991 Objective Loss 0.239991 LR 0.000500 Time 0.019953 -2022-12-06 11:07:19,750 - Epoch: [100][ 800/ 1200] Overall Loss 0.239996 Objective Loss 0.239996 LR 0.000500 Time 0.019940 -2022-12-06 11:07:19,940 - Epoch: [100][ 810/ 1200] Overall Loss 0.239987 Objective Loss 0.239987 LR 0.000500 Time 0.019929 -2022-12-06 11:07:20,131 - Epoch: [100][ 820/ 1200] Overall Loss 0.239515 Objective Loss 0.239515 LR 0.000500 Time 0.019917 -2022-12-06 11:07:20,321 - Epoch: [100][ 830/ 1200] Overall Loss 0.239800 Objective Loss 0.239800 LR 0.000500 Time 0.019906 -2022-12-06 11:07:20,512 - Epoch: [100][ 840/ 1200] Overall Loss 0.239882 Objective Loss 0.239882 LR 0.000500 Time 0.019895 -2022-12-06 11:07:20,702 - Epoch: [100][ 850/ 1200] Overall Loss 0.239532 Objective Loss 0.239532 LR 0.000500 Time 0.019885 -2022-12-06 11:07:20,893 - Epoch: [100][ 860/ 1200] Overall Loss 0.239581 Objective Loss 0.239581 LR 0.000500 Time 0.019875 -2022-12-06 11:07:21,084 - Epoch: [100][ 870/ 1200] Overall Loss 0.239821 Objective Loss 0.239821 LR 0.000500 Time 0.019864 -2022-12-06 11:07:21,274 - Epoch: [100][ 880/ 1200] Overall Loss 0.240114 Objective Loss 0.240114 LR 0.000500 Time 0.019854 -2022-12-06 11:07:21,464 - Epoch: [100][ 890/ 1200] Overall Loss 0.240179 Objective Loss 0.240179 LR 0.000500 Time 0.019844 -2022-12-06 11:07:21,654 - Epoch: [100][ 900/ 1200] Overall Loss 0.240062 Objective Loss 0.240062 LR 0.000500 Time 0.019835 -2022-12-06 11:07:21,844 - Epoch: [100][ 910/ 1200] Overall Loss 0.240228 Objective Loss 0.240228 LR 0.000500 Time 0.019825 -2022-12-06 11:07:22,035 - Epoch: [100][ 920/ 1200] Overall Loss 0.239894 Objective Loss 0.239894 LR 0.000500 Time 0.019816 -2022-12-06 11:07:22,226 - Epoch: [100][ 930/ 1200] Overall Loss 0.239853 Objective Loss 0.239853 LR 0.000500 Time 0.019807 -2022-12-06 11:07:22,416 - Epoch: [100][ 940/ 1200] Overall Loss 0.239591 Objective Loss 0.239591 LR 0.000500 Time 0.019798 -2022-12-06 11:07:22,607 - Epoch: [100][ 950/ 1200] Overall Loss 0.239055 Objective Loss 0.239055 LR 0.000500 Time 0.019791 -2022-12-06 11:07:22,798 - Epoch: [100][ 960/ 1200] Overall Loss 0.239007 Objective Loss 0.239007 LR 0.000500 Time 0.019782 -2022-12-06 11:07:22,988 - Epoch: [100][ 970/ 1200] Overall Loss 0.238943 Objective Loss 0.238943 LR 0.000500 Time 0.019774 -2022-12-06 11:07:23,178 - Epoch: [100][ 980/ 1200] Overall Loss 0.238880 Objective Loss 0.238880 LR 0.000500 Time 0.019766 -2022-12-06 11:07:23,368 - Epoch: [100][ 990/ 1200] Overall Loss 0.238959 Objective Loss 0.238959 LR 0.000500 Time 0.019758 -2022-12-06 11:07:23,558 - Epoch: [100][ 1000/ 1200] Overall Loss 0.238847 Objective Loss 0.238847 LR 0.000500 Time 0.019750 -2022-12-06 11:07:23,748 - Epoch: [100][ 1010/ 1200] Overall Loss 0.238809 Objective Loss 0.238809 LR 0.000500 Time 0.019742 -2022-12-06 11:07:23,940 - Epoch: [100][ 1020/ 1200] Overall Loss 0.238807 Objective Loss 0.238807 LR 0.000500 Time 0.019735 -2022-12-06 11:07:24,130 - Epoch: [100][ 1030/ 1200] Overall Loss 0.238804 Objective Loss 0.238804 LR 0.000500 Time 0.019728 -2022-12-06 11:07:24,321 - Epoch: [100][ 1040/ 1200] Overall Loss 0.238727 Objective Loss 0.238727 LR 0.000500 Time 0.019721 -2022-12-06 11:07:24,512 - Epoch: [100][ 1050/ 1200] Overall Loss 0.238478 Objective Loss 0.238478 LR 0.000500 Time 0.019714 -2022-12-06 11:07:24,702 - Epoch: [100][ 1060/ 1200] Overall Loss 0.238312 Objective Loss 0.238312 LR 0.000500 Time 0.019707 -2022-12-06 11:07:24,893 - Epoch: [100][ 1070/ 1200] Overall Loss 0.238259 Objective Loss 0.238259 LR 0.000500 Time 0.019701 -2022-12-06 11:07:25,083 - Epoch: [100][ 1080/ 1200] Overall Loss 0.238337 Objective Loss 0.238337 LR 0.000500 Time 0.019695 -2022-12-06 11:07:25,274 - Epoch: [100][ 1090/ 1200] Overall Loss 0.238352 Objective Loss 0.238352 LR 0.000500 Time 0.019688 -2022-12-06 11:07:25,465 - Epoch: [100][ 1100/ 1200] Overall Loss 0.238278 Objective Loss 0.238278 LR 0.000500 Time 0.019683 -2022-12-06 11:07:25,656 - Epoch: [100][ 1110/ 1200] Overall Loss 0.238219 Objective Loss 0.238219 LR 0.000500 Time 0.019677 -2022-12-06 11:07:25,846 - Epoch: [100][ 1120/ 1200] Overall Loss 0.237956 Objective Loss 0.237956 LR 0.000500 Time 0.019670 -2022-12-06 11:07:26,037 - Epoch: [100][ 1130/ 1200] Overall Loss 0.237985 Objective Loss 0.237985 LR 0.000500 Time 0.019664 -2022-12-06 11:07:26,227 - Epoch: [100][ 1140/ 1200] Overall Loss 0.238115 Objective Loss 0.238115 LR 0.000500 Time 0.019659 -2022-12-06 11:07:26,418 - Epoch: [100][ 1150/ 1200] Overall Loss 0.238198 Objective Loss 0.238198 LR 0.000500 Time 0.019653 -2022-12-06 11:07:26,608 - Epoch: [100][ 1160/ 1200] Overall Loss 0.238014 Objective Loss 0.238014 LR 0.000500 Time 0.019647 -2022-12-06 11:07:26,799 - Epoch: [100][ 1170/ 1200] Overall Loss 0.237875 Objective Loss 0.237875 LR 0.000500 Time 0.019642 -2022-12-06 11:07:26,990 - Epoch: [100][ 1180/ 1200] Overall Loss 0.238141 Objective Loss 0.238141 LR 0.000500 Time 0.019636 -2022-12-06 11:07:27,179 - Epoch: [100][ 1190/ 1200] Overall Loss 0.238286 Objective Loss 0.238286 LR 0.000500 Time 0.019630 -2022-12-06 11:07:27,402 - Epoch: [100][ 1200/ 1200] Overall Loss 0.238243 Objective Loss 0.238243 Top1 87.656904 Top5 97.280335 LR 0.000500 Time 0.019652 -2022-12-06 11:07:27,496 - --- validate (epoch=100)----------- -2022-12-06 11:07:27,496 - 34129 samples (256 per mini-batch) -2022-12-06 11:07:27,943 - Epoch: [100][ 10/ 134] Loss 0.278609 Top1 84.570312 Top5 98.359375 -2022-12-06 11:07:28,086 - Epoch: [100][ 20/ 134] Loss 0.284745 Top1 84.687500 Top5 98.261719 -2022-12-06 11:07:28,235 - Epoch: [100][ 30/ 134] Loss 0.285818 Top1 84.765625 Top5 98.125000 -2022-12-06 11:07:28,377 - Epoch: [100][ 40/ 134] Loss 0.276674 Top1 85.009766 Top5 98.046875 -2022-12-06 11:07:28,526 - Epoch: [100][ 50/ 134] Loss 0.274576 Top1 85.007812 Top5 98.093750 -2022-12-06 11:07:28,667 - Epoch: [100][ 60/ 134] Loss 0.273930 Top1 85.162760 Top5 98.111979 -2022-12-06 11:07:28,816 - Epoch: [100][ 70/ 134] Loss 0.271388 Top1 85.357143 Top5 98.119420 -2022-12-06 11:07:28,955 - Epoch: [100][ 80/ 134] Loss 0.273395 Top1 85.312500 Top5 98.085938 -2022-12-06 11:07:29,104 - Epoch: [100][ 90/ 134] Loss 0.272732 Top1 85.403646 Top5 98.081597 -2022-12-06 11:07:29,245 - Epoch: [100][ 100/ 134] Loss 0.272942 Top1 85.425781 Top5 98.062500 -2022-12-06 11:07:29,395 - Epoch: [100][ 110/ 134] Loss 0.272846 Top1 85.379972 Top5 98.029119 -2022-12-06 11:07:29,536 - Epoch: [100][ 120/ 134] Loss 0.272085 Top1 85.322266 Top5 98.043620 -2022-12-06 11:07:29,679 - Epoch: [100][ 130/ 134] Loss 0.273368 Top1 85.252404 Top5 98.049880 -2022-12-06 11:07:29,716 - Epoch: [100][ 134/ 134] Loss 0.272886 Top1 85.223710 Top5 98.057371 -2022-12-06 11:07:29,806 - ==> Top1: 85.224 Top5: 98.057 Loss: 0.273 - -2022-12-06 11:07:29,807 - ==> Confusion: -[[ 902 0 3 4 5 4 0 4 7 52 0 0 1 4 4 2 1 0 2 0 1] - [ 2 930 1 3 9 18 3 25 0 1 2 4 1 3 0 3 2 1 13 0 6] - [ 6 5 1006 13 3 1 18 10 0 1 8 5 2 0 4 4 0 0 4 3 10] - [ 4 1 18 950 0 3 0 1 0 0 8 1 4 1 10 1 4 2 8 0 4] - [ 10 8 2 0 953 4 1 3 1 4 1 0 0 3 13 9 3 2 1 1 1] - [ 5 18 0 0 4 958 5 32 2 2 1 7 4 17 1 1 0 0 2 6 4] - [ 1 3 16 3 0 2 1066 5 0 0 3 1 0 4 0 3 0 0 1 7 3] - [ 2 5 11 2 1 18 3 973 0 1 3 6 1 2 0 0 0 0 13 9 4] - [ 6 6 0 2 0 4 0 3 960 38 16 1 2 12 7 1 1 0 3 1 1] - [ 68 0 1 0 3 3 0 2 27 869 1 1 0 11 3 1 0 2 0 1 8] - [ 1 0 3 8 2 2 1 3 9 0 963 1 1 11 2 0 1 0 6 1 4] - [ 2 1 1 0 1 15 5 2 2 0 1 976 15 3 0 6 3 3 1 12 2] - [ 2 2 4 3 1 2 2 0 0 0 0 41 886 0 0 9 1 7 0 4 5] - [ 0 1 0 0 0 6 0 3 10 12 7 3 2 962 0 2 2 1 1 4 7] - [ 8 5 1 15 2 2 0 1 20 4 1 5 4 5 1038 0 2 1 10 0 6] - [ 1 0 3 0 1 2 3 0 0 0 0 14 5 4 0 987 7 9 0 4 3] - [ 3 3 0 2 2 0 1 1 2 0 0 1 1 1 1 16 1025 0 1 5 7] - [ 2 0 2 2 1 0 1 2 1 4 0 8 23 2 0 10 0 972 1 1 4] - [ 5 7 4 15 1 0 2 27 1 2 2 1 2 0 7 1 0 0 927 0 4] - [ 3 3 3 2 2 4 7 9 0 1 2 9 6 10 0 1 5 2 1 1006 4] - [ 150 213 254 138 105 179 84 203 85 107 190 110 376 354 137 117 184 74 135 259 9772]] - -2022-12-06 11:07:30,487 - ==> Best [Top1: 86.029 Top5: 98.169 Sparsity:0.00 Params: 5376 on epoch: 53] -2022-12-06 11:07:30,487 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:07:30,493 - - -2022-12-06 11:07:30,493 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:07:31,433 - Epoch: [101][ 10/ 1200] Overall Loss 0.224369 Objective Loss 0.224369 LR 0.000500 Time 0.093848 -2022-12-06 11:07:31,625 - Epoch: [101][ 20/ 1200] Overall Loss 0.226679 Objective Loss 0.226679 LR 0.000500 Time 0.056524 -2022-12-06 11:07:31,816 - Epoch: [101][ 30/ 1200] Overall Loss 0.220323 Objective Loss 0.220323 LR 0.000500 Time 0.044021 -2022-12-06 11:07:32,007 - Epoch: [101][ 40/ 1200] Overall Loss 0.217473 Objective Loss 0.217473 LR 0.000500 Time 0.037772 -2022-12-06 11:07:32,197 - Epoch: [101][ 50/ 1200] Overall Loss 0.217071 Objective Loss 0.217071 LR 0.000500 Time 0.034007 -2022-12-06 11:07:32,387 - Epoch: [101][ 60/ 1200] Overall Loss 0.220984 Objective Loss 0.220984 LR 0.000500 Time 0.031502 -2022-12-06 11:07:32,576 - Epoch: [101][ 70/ 1200] Overall Loss 0.225349 Objective Loss 0.225349 LR 0.000500 Time 0.029700 -2022-12-06 11:07:32,767 - Epoch: [101][ 80/ 1200] Overall Loss 0.226531 Objective Loss 0.226531 LR 0.000500 Time 0.028362 -2022-12-06 11:07:32,958 - Epoch: [101][ 90/ 1200] Overall Loss 0.226398 Objective Loss 0.226398 LR 0.000500 Time 0.027326 -2022-12-06 11:07:33,148 - Epoch: [101][ 100/ 1200] Overall Loss 0.229387 Objective Loss 0.229387 LR 0.000500 Time 0.026493 -2022-12-06 11:07:33,339 - Epoch: [101][ 110/ 1200] Overall Loss 0.228678 Objective Loss 0.228678 LR 0.000500 Time 0.025814 -2022-12-06 11:07:33,530 - Epoch: [101][ 120/ 1200] Overall Loss 0.229710 Objective Loss 0.229710 LR 0.000500 Time 0.025245 -2022-12-06 11:07:33,720 - Epoch: [101][ 130/ 1200] Overall Loss 0.230286 Objective Loss 0.230286 LR 0.000500 Time 0.024762 -2022-12-06 11:07:33,910 - Epoch: [101][ 140/ 1200] Overall Loss 0.230941 Objective Loss 0.230941 LR 0.000500 Time 0.024350 -2022-12-06 11:07:34,100 - Epoch: [101][ 150/ 1200] Overall Loss 0.230968 Objective Loss 0.230968 LR 0.000500 Time 0.023987 -2022-12-06 11:07:34,291 - Epoch: [101][ 160/ 1200] Overall Loss 0.231475 Objective Loss 0.231475 LR 0.000500 Time 0.023678 -2022-12-06 11:07:34,481 - Epoch: [101][ 170/ 1200] Overall Loss 0.230823 Objective Loss 0.230823 LR 0.000500 Time 0.023402 -2022-12-06 11:07:34,671 - Epoch: [101][ 180/ 1200] Overall Loss 0.229908 Objective Loss 0.229908 LR 0.000500 Time 0.023151 -2022-12-06 11:07:34,860 - Epoch: [101][ 190/ 1200] Overall Loss 0.230776 Objective Loss 0.230776 LR 0.000500 Time 0.022928 -2022-12-06 11:07:35,051 - Epoch: [101][ 200/ 1200] Overall Loss 0.229489 Objective Loss 0.229489 LR 0.000500 Time 0.022731 -2022-12-06 11:07:35,241 - Epoch: [101][ 210/ 1200] Overall Loss 0.228783 Objective Loss 0.228783 LR 0.000500 Time 0.022553 -2022-12-06 11:07:35,431 - Epoch: [101][ 220/ 1200] Overall Loss 0.228439 Objective Loss 0.228439 LR 0.000500 Time 0.022387 -2022-12-06 11:07:35,622 - Epoch: [101][ 230/ 1200] Overall Loss 0.228066 Objective Loss 0.228066 LR 0.000500 Time 0.022242 -2022-12-06 11:07:35,812 - Epoch: [101][ 240/ 1200] Overall Loss 0.228147 Objective Loss 0.228147 LR 0.000500 Time 0.022105 -2022-12-06 11:07:36,002 - Epoch: [101][ 250/ 1200] Overall Loss 0.227190 Objective Loss 0.227190 LR 0.000500 Time 0.021979 -2022-12-06 11:07:36,193 - Epoch: [101][ 260/ 1200] Overall Loss 0.226906 Objective Loss 0.226906 LR 0.000500 Time 0.021865 -2022-12-06 11:07:36,382 - Epoch: [101][ 270/ 1200] Overall Loss 0.225830 Objective Loss 0.225830 LR 0.000500 Time 0.021756 -2022-12-06 11:07:36,572 - Epoch: [101][ 280/ 1200] Overall Loss 0.226107 Objective Loss 0.226107 LR 0.000500 Time 0.021654 -2022-12-06 11:07:36,762 - Epoch: [101][ 290/ 1200] Overall Loss 0.226586 Objective Loss 0.226586 LR 0.000500 Time 0.021560 -2022-12-06 11:07:36,952 - Epoch: [101][ 300/ 1200] Overall Loss 0.226715 Objective Loss 0.226715 LR 0.000500 Time 0.021473 -2022-12-06 11:07:37,142 - Epoch: [101][ 310/ 1200] Overall Loss 0.226421 Objective Loss 0.226421 LR 0.000500 Time 0.021393 -2022-12-06 11:07:37,332 - Epoch: [101][ 320/ 1200] Overall Loss 0.226876 Objective Loss 0.226876 LR 0.000500 Time 0.021317 -2022-12-06 11:07:37,523 - Epoch: [101][ 330/ 1200] Overall Loss 0.227258 Objective Loss 0.227258 LR 0.000500 Time 0.021246 -2022-12-06 11:07:37,713 - Epoch: [101][ 340/ 1200] Overall Loss 0.228234 Objective Loss 0.228234 LR 0.000500 Time 0.021179 -2022-12-06 11:07:37,904 - Epoch: [101][ 350/ 1200] Overall Loss 0.228090 Objective Loss 0.228090 LR 0.000500 Time 0.021117 -2022-12-06 11:07:38,093 - Epoch: [101][ 360/ 1200] Overall Loss 0.228167 Objective Loss 0.228167 LR 0.000500 Time 0.021054 -2022-12-06 11:07:38,283 - Epoch: [101][ 370/ 1200] Overall Loss 0.227947 Objective Loss 0.227947 LR 0.000500 Time 0.020998 -2022-12-06 11:07:38,473 - Epoch: [101][ 380/ 1200] Overall Loss 0.228739 Objective Loss 0.228739 LR 0.000500 Time 0.020945 -2022-12-06 11:07:38,665 - Epoch: [101][ 390/ 1200] Overall Loss 0.229299 Objective Loss 0.229299 LR 0.000500 Time 0.020896 -2022-12-06 11:07:38,855 - Epoch: [101][ 400/ 1200] Overall Loss 0.229640 Objective Loss 0.229640 LR 0.000500 Time 0.020849 -2022-12-06 11:07:39,046 - Epoch: [101][ 410/ 1200] Overall Loss 0.229630 Objective Loss 0.229630 LR 0.000500 Time 0.020804 -2022-12-06 11:07:39,236 - Epoch: [101][ 420/ 1200] Overall Loss 0.229844 Objective Loss 0.229844 LR 0.000500 Time 0.020761 -2022-12-06 11:07:39,427 - Epoch: [101][ 430/ 1200] Overall Loss 0.230472 Objective Loss 0.230472 LR 0.000500 Time 0.020721 -2022-12-06 11:07:39,618 - Epoch: [101][ 440/ 1200] Overall Loss 0.230520 Objective Loss 0.230520 LR 0.000500 Time 0.020682 -2022-12-06 11:07:39,809 - Epoch: [101][ 450/ 1200] Overall Loss 0.230197 Objective Loss 0.230197 LR 0.000500 Time 0.020645 -2022-12-06 11:07:39,999 - Epoch: [101][ 460/ 1200] Overall Loss 0.230368 Objective Loss 0.230368 LR 0.000500 Time 0.020609 -2022-12-06 11:07:40,190 - Epoch: [101][ 470/ 1200] Overall Loss 0.230760 Objective Loss 0.230760 LR 0.000500 Time 0.020575 -2022-12-06 11:07:40,380 - Epoch: [101][ 480/ 1200] Overall Loss 0.230824 Objective Loss 0.230824 LR 0.000500 Time 0.020542 -2022-12-06 11:07:40,570 - Epoch: [101][ 490/ 1200] Overall Loss 0.231001 Objective Loss 0.231001 LR 0.000500 Time 0.020510 -2022-12-06 11:07:40,761 - Epoch: [101][ 500/ 1200] Overall Loss 0.230841 Objective Loss 0.230841 LR 0.000500 Time 0.020480 -2022-12-06 11:07:40,952 - Epoch: [101][ 510/ 1200] Overall Loss 0.230788 Objective Loss 0.230788 LR 0.000500 Time 0.020451 -2022-12-06 11:07:41,142 - Epoch: [101][ 520/ 1200] Overall Loss 0.230817 Objective Loss 0.230817 LR 0.000500 Time 0.020423 -2022-12-06 11:07:41,333 - Epoch: [101][ 530/ 1200] Overall Loss 0.230624 Objective Loss 0.230624 LR 0.000500 Time 0.020397 -2022-12-06 11:07:41,522 - Epoch: [101][ 540/ 1200] Overall Loss 0.230924 Objective Loss 0.230924 LR 0.000500 Time 0.020369 -2022-12-06 11:07:41,712 - Epoch: [101][ 550/ 1200] Overall Loss 0.230936 Objective Loss 0.230936 LR 0.000500 Time 0.020343 -2022-12-06 11:07:41,903 - Epoch: [101][ 560/ 1200] Overall Loss 0.230541 Objective Loss 0.230541 LR 0.000500 Time 0.020319 -2022-12-06 11:07:42,093 - Epoch: [101][ 570/ 1200] Overall Loss 0.230286 Objective Loss 0.230286 LR 0.000500 Time 0.020295 -2022-12-06 11:07:42,283 - Epoch: [101][ 580/ 1200] Overall Loss 0.230228 Objective Loss 0.230228 LR 0.000500 Time 0.020272 -2022-12-06 11:07:42,473 - Epoch: [101][ 590/ 1200] Overall Loss 0.230023 Objective Loss 0.230023 LR 0.000500 Time 0.020250 -2022-12-06 11:07:42,664 - Epoch: [101][ 600/ 1200] Overall Loss 0.229805 Objective Loss 0.229805 LR 0.000500 Time 0.020229 -2022-12-06 11:07:42,855 - Epoch: [101][ 610/ 1200] Overall Loss 0.229689 Objective Loss 0.229689 LR 0.000500 Time 0.020210 -2022-12-06 11:07:43,046 - Epoch: [101][ 620/ 1200] Overall Loss 0.229148 Objective Loss 0.229148 LR 0.000500 Time 0.020191 -2022-12-06 11:07:43,237 - Epoch: [101][ 630/ 1200] Overall Loss 0.229403 Objective Loss 0.229403 LR 0.000500 Time 0.020172 -2022-12-06 11:07:43,427 - Epoch: [101][ 640/ 1200] Overall Loss 0.229313 Objective Loss 0.229313 LR 0.000500 Time 0.020154 -2022-12-06 11:07:43,619 - Epoch: [101][ 650/ 1200] Overall Loss 0.229511 Objective Loss 0.229511 LR 0.000500 Time 0.020138 -2022-12-06 11:07:43,810 - Epoch: [101][ 660/ 1200] Overall Loss 0.229572 Objective Loss 0.229572 LR 0.000500 Time 0.020122 -2022-12-06 11:07:44,001 - Epoch: [101][ 670/ 1200] Overall Loss 0.229642 Objective Loss 0.229642 LR 0.000500 Time 0.020106 -2022-12-06 11:07:44,192 - Epoch: [101][ 680/ 1200] Overall Loss 0.229631 Objective Loss 0.229631 LR 0.000500 Time 0.020089 -2022-12-06 11:07:44,382 - Epoch: [101][ 690/ 1200] Overall Loss 0.229703 Objective Loss 0.229703 LR 0.000500 Time 0.020073 -2022-12-06 11:07:44,572 - Epoch: [101][ 700/ 1200] Overall Loss 0.229772 Objective Loss 0.229772 LR 0.000500 Time 0.020057 -2022-12-06 11:07:44,763 - Epoch: [101][ 710/ 1200] Overall Loss 0.229725 Objective Loss 0.229725 LR 0.000500 Time 0.020043 -2022-12-06 11:07:44,954 - Epoch: [101][ 720/ 1200] Overall Loss 0.229857 Objective Loss 0.229857 LR 0.000500 Time 0.020029 -2022-12-06 11:07:45,144 - Epoch: [101][ 730/ 1200] Overall Loss 0.230166 Objective Loss 0.230166 LR 0.000500 Time 0.020014 -2022-12-06 11:07:45,334 - Epoch: [101][ 740/ 1200] Overall Loss 0.229935 Objective Loss 0.229935 LR 0.000500 Time 0.020000 -2022-12-06 11:07:45,525 - Epoch: [101][ 750/ 1200] Overall Loss 0.229848 Objective Loss 0.229848 LR 0.000500 Time 0.019987 -2022-12-06 11:07:45,715 - Epoch: [101][ 760/ 1200] Overall Loss 0.230233 Objective Loss 0.230233 LR 0.000500 Time 0.019974 -2022-12-06 11:07:45,905 - Epoch: [101][ 770/ 1200] Overall Loss 0.229965 Objective Loss 0.229965 LR 0.000500 Time 0.019961 -2022-12-06 11:07:46,095 - Epoch: [101][ 780/ 1200] Overall Loss 0.229532 Objective Loss 0.229532 LR 0.000500 Time 0.019947 -2022-12-06 11:07:46,286 - Epoch: [101][ 790/ 1200] Overall Loss 0.229731 Objective Loss 0.229731 LR 0.000500 Time 0.019936 -2022-12-06 11:07:46,477 - Epoch: [101][ 800/ 1200] Overall Loss 0.229544 Objective Loss 0.229544 LR 0.000500 Time 0.019924 -2022-12-06 11:07:46,667 - Epoch: [101][ 810/ 1200] Overall Loss 0.229665 Objective Loss 0.229665 LR 0.000500 Time 0.019912 -2022-12-06 11:07:46,857 - Epoch: [101][ 820/ 1200] Overall Loss 0.229640 Objective Loss 0.229640 LR 0.000500 Time 0.019901 -2022-12-06 11:07:47,048 - Epoch: [101][ 830/ 1200] Overall Loss 0.229473 Objective Loss 0.229473 LR 0.000500 Time 0.019890 -2022-12-06 11:07:47,239 - Epoch: [101][ 840/ 1200] Overall Loss 0.229103 Objective Loss 0.229103 LR 0.000500 Time 0.019880 -2022-12-06 11:07:47,429 - Epoch: [101][ 850/ 1200] Overall Loss 0.228950 Objective Loss 0.228950 LR 0.000500 Time 0.019869 -2022-12-06 11:07:47,619 - Epoch: [101][ 860/ 1200] Overall Loss 0.229411 Objective Loss 0.229411 LR 0.000500 Time 0.019859 -2022-12-06 11:07:47,810 - Epoch: [101][ 870/ 1200] Overall Loss 0.229650 Objective Loss 0.229650 LR 0.000500 Time 0.019849 -2022-12-06 11:07:48,000 - Epoch: [101][ 880/ 1200] Overall Loss 0.229289 Objective Loss 0.229289 LR 0.000500 Time 0.019839 -2022-12-06 11:07:48,191 - Epoch: [101][ 890/ 1200] Overall Loss 0.229226 Objective Loss 0.229226 LR 0.000500 Time 0.019830 -2022-12-06 11:07:48,382 - Epoch: [101][ 900/ 1200] Overall Loss 0.229295 Objective Loss 0.229295 LR 0.000500 Time 0.019822 -2022-12-06 11:07:48,573 - Epoch: [101][ 910/ 1200] Overall Loss 0.229527 Objective Loss 0.229527 LR 0.000500 Time 0.019813 -2022-12-06 11:07:48,764 - Epoch: [101][ 920/ 1200] Overall Loss 0.229338 Objective Loss 0.229338 LR 0.000500 Time 0.019805 -2022-12-06 11:07:48,955 - Epoch: [101][ 930/ 1200] Overall Loss 0.229339 Objective Loss 0.229339 LR 0.000500 Time 0.019797 -2022-12-06 11:07:49,146 - Epoch: [101][ 940/ 1200] Overall Loss 0.229186 Objective Loss 0.229186 LR 0.000500 Time 0.019788 -2022-12-06 11:07:49,336 - Epoch: [101][ 950/ 1200] Overall Loss 0.229022 Objective Loss 0.229022 LR 0.000500 Time 0.019779 -2022-12-06 11:07:49,526 - Epoch: [101][ 960/ 1200] Overall Loss 0.228935 Objective Loss 0.228935 LR 0.000500 Time 0.019771 -2022-12-06 11:07:49,717 - Epoch: [101][ 970/ 1200] Overall Loss 0.229051 Objective Loss 0.229051 LR 0.000500 Time 0.019763 -2022-12-06 11:07:49,907 - Epoch: [101][ 980/ 1200] Overall Loss 0.229017 Objective Loss 0.229017 LR 0.000500 Time 0.019755 -2022-12-06 11:07:50,098 - Epoch: [101][ 990/ 1200] Overall Loss 0.228936 Objective Loss 0.228936 LR 0.000500 Time 0.019747 -2022-12-06 11:07:50,288 - Epoch: [101][ 1000/ 1200] Overall Loss 0.228923 Objective Loss 0.228923 LR 0.000500 Time 0.019740 -2022-12-06 11:07:50,479 - Epoch: [101][ 1010/ 1200] Overall Loss 0.228928 Objective Loss 0.228928 LR 0.000500 Time 0.019732 -2022-12-06 11:07:50,669 - Epoch: [101][ 1020/ 1200] Overall Loss 0.228721 Objective Loss 0.228721 LR 0.000500 Time 0.019725 -2022-12-06 11:07:50,861 - Epoch: [101][ 1030/ 1200] Overall Loss 0.228835 Objective Loss 0.228835 LR 0.000500 Time 0.019719 -2022-12-06 11:07:51,051 - Epoch: [101][ 1040/ 1200] Overall Loss 0.228939 Objective Loss 0.228939 LR 0.000500 Time 0.019712 -2022-12-06 11:07:51,242 - Epoch: [101][ 1050/ 1200] Overall Loss 0.228919 Objective Loss 0.228919 LR 0.000500 Time 0.019706 -2022-12-06 11:07:51,433 - Epoch: [101][ 1060/ 1200] Overall Loss 0.228883 Objective Loss 0.228883 LR 0.000500 Time 0.019700 -2022-12-06 11:07:51,624 - Epoch: [101][ 1070/ 1200] Overall Loss 0.228801 Objective Loss 0.228801 LR 0.000500 Time 0.019693 -2022-12-06 11:07:51,815 - Epoch: [101][ 1080/ 1200] Overall Loss 0.228799 Objective Loss 0.228799 LR 0.000500 Time 0.019687 -2022-12-06 11:07:52,005 - Epoch: [101][ 1090/ 1200] Overall Loss 0.228771 Objective Loss 0.228771 LR 0.000500 Time 0.019681 -2022-12-06 11:07:52,197 - Epoch: [101][ 1100/ 1200] Overall Loss 0.228682 Objective Loss 0.228682 LR 0.000500 Time 0.019676 -2022-12-06 11:07:52,387 - Epoch: [101][ 1110/ 1200] Overall Loss 0.228690 Objective Loss 0.228690 LR 0.000500 Time 0.019669 -2022-12-06 11:07:52,578 - Epoch: [101][ 1120/ 1200] Overall Loss 0.228805 Objective Loss 0.228805 LR 0.000500 Time 0.019664 -2022-12-06 11:07:52,770 - Epoch: [101][ 1130/ 1200] Overall Loss 0.228891 Objective Loss 0.228891 LR 0.000500 Time 0.019659 -2022-12-06 11:07:52,960 - Epoch: [101][ 1140/ 1200] Overall Loss 0.228756 Objective Loss 0.228756 LR 0.000500 Time 0.019653 -2022-12-06 11:07:53,150 - Epoch: [101][ 1150/ 1200] Overall Loss 0.228786 Objective Loss 0.228786 LR 0.000500 Time 0.019647 -2022-12-06 11:07:53,341 - Epoch: [101][ 1160/ 1200] Overall Loss 0.228745 Objective Loss 0.228745 LR 0.000500 Time 0.019642 -2022-12-06 11:07:53,532 - Epoch: [101][ 1170/ 1200] Overall Loss 0.228827 Objective Loss 0.228827 LR 0.000500 Time 0.019636 -2022-12-06 11:07:53,723 - Epoch: [101][ 1180/ 1200] Overall Loss 0.228622 Objective Loss 0.228622 LR 0.000500 Time 0.019631 -2022-12-06 11:07:53,913 - Epoch: [101][ 1190/ 1200] Overall Loss 0.228733 Objective Loss 0.228733 LR 0.000500 Time 0.019625 -2022-12-06 11:07:54,134 - Epoch: [101][ 1200/ 1200] Overall Loss 0.228824 Objective Loss 0.228824 Top1 88.493724 Top5 98.953975 LR 0.000500 Time 0.019645 -2022-12-06 11:07:54,226 - --- validate (epoch=101)----------- -2022-12-06 11:07:54,226 - 34129 samples (256 per mini-batch) -2022-12-06 11:07:54,681 - Epoch: [101][ 10/ 134] Loss 0.231479 Top1 87.226562 Top5 98.203125 -2022-12-06 11:07:54,829 - Epoch: [101][ 20/ 134] Loss 0.241311 Top1 86.875000 Top5 98.222656 -2022-12-06 11:07:54,970 - Epoch: [101][ 30/ 134] Loss 0.262637 Top1 86.197917 Top5 98.033854 -2022-12-06 11:07:55,117 - Epoch: [101][ 40/ 134] Loss 0.275493 Top1 86.093750 Top5 98.046875 -2022-12-06 11:07:55,257 - Epoch: [101][ 50/ 134] Loss 0.274172 Top1 86.070312 Top5 98.062500 -2022-12-06 11:07:55,402 - Epoch: [101][ 60/ 134] Loss 0.276491 Top1 85.944010 Top5 98.014323 -2022-12-06 11:07:55,541 - Epoch: [101][ 70/ 134] Loss 0.274375 Top1 85.998884 Top5 98.041295 -2022-12-06 11:07:55,688 - Epoch: [101][ 80/ 134] Loss 0.271080 Top1 86.000977 Top5 98.071289 -2022-12-06 11:07:55,828 - Epoch: [101][ 90/ 134] Loss 0.272304 Top1 85.937500 Top5 98.068576 -2022-12-06 11:07:55,974 - Epoch: [101][ 100/ 134] Loss 0.271628 Top1 86.082031 Top5 98.113281 -2022-12-06 11:07:56,115 - Epoch: [101][ 110/ 134] Loss 0.270695 Top1 86.104403 Top5 98.103693 -2022-12-06 11:07:56,262 - Epoch: [101][ 120/ 134] Loss 0.270677 Top1 86.103516 Top5 98.134766 -2022-12-06 11:07:56,396 - Epoch: [101][ 130/ 134] Loss 0.270533 Top1 86.210938 Top5 98.121995 -2022-12-06 11:07:56,433 - Epoch: [101][ 134/ 134] Loss 0.271088 Top1 86.208210 Top5 98.101321 -2022-12-06 11:07:56,522 - ==> Top1: 86.208 Top5: 98.101 Loss: 0.271 - -2022-12-06 11:07:56,523 - ==> Confusion: -[[ 909 3 1 2 5 5 0 2 6 41 0 2 1 3 9 2 1 0 2 1 1] - [ 1 926 1 3 8 31 2 17 1 0 2 6 3 2 0 1 2 2 5 1 13] - [ 4 1 987 18 3 3 28 11 1 3 7 4 3 2 4 3 0 2 5 3 11] - [ 3 0 12 946 0 3 1 1 0 1 9 1 4 2 12 0 2 4 10 1 8] - [ 11 5 1 0 951 6 0 1 0 6 0 4 1 2 11 8 3 1 0 4 5] - [ 0 12 0 2 5 985 3 13 3 2 0 17 4 12 2 0 0 1 2 5 1] - [ 0 2 6 1 1 2 1079 4 0 0 3 0 2 2 0 4 1 1 0 7 3] - [ 0 8 8 1 1 26 7 944 2 2 3 8 1 3 0 0 1 1 16 16 6] - [ 5 4 0 4 0 3 1 1 963 29 13 1 5 13 14 0 0 1 2 2 3] - [ 59 1 1 0 8 2 0 3 35 863 1 3 0 11 4 1 0 1 0 1 7] - [ 1 2 3 6 2 1 2 1 8 0 956 3 1 15 4 0 1 1 6 3 3] - [ 3 0 0 0 0 8 4 2 1 1 2 982 20 1 0 5 3 6 1 8 4] - [ 2 1 1 1 1 2 0 0 0 0 0 31 896 2 2 7 1 13 0 5 4] - [ 0 0 1 0 2 4 0 2 10 7 5 3 3 965 2 1 1 3 0 6 8] - [ 6 2 2 13 3 2 0 1 20 1 0 4 3 4 1052 0 2 1 8 0 6] - [ 0 0 1 0 2 3 2 0 0 0 0 9 6 2 0 988 8 13 0 5 4] - [ 2 3 1 3 1 0 2 0 0 0 1 4 3 2 2 12 1029 1 0 4 2] - [ 2 0 1 2 0 1 1 2 1 2 0 7 12 2 0 9 2 990 0 1 1] - [ 5 5 3 11 2 3 0 25 2 0 2 2 4 1 5 0 0 0 935 2 1] - [ 4 3 1 1 1 5 11 10 0 0 1 11 6 7 0 2 4 5 0 1005 3] - [ 110 193 154 111 83 210 90 173 76 71 178 114 366 339 139 83 162 89 166 254 10065]] - -2022-12-06 11:07:57,088 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:07:57,088 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:07:57,125 - - -2022-12-06 11:07:57,125 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:07:58,157 - Epoch: [102][ 10/ 1200] Overall Loss 0.238876 Objective Loss 0.238876 LR 0.000500 Time 0.103131 -2022-12-06 11:07:58,359 - Epoch: [102][ 20/ 1200] Overall Loss 0.220134 Objective Loss 0.220134 LR 0.000500 Time 0.061628 -2022-12-06 11:07:58,556 - Epoch: [102][ 30/ 1200] Overall Loss 0.227854 Objective Loss 0.227854 LR 0.000500 Time 0.047648 -2022-12-06 11:07:58,756 - Epoch: [102][ 40/ 1200] Overall Loss 0.219051 Objective Loss 0.219051 LR 0.000500 Time 0.040711 -2022-12-06 11:07:58,953 - Epoch: [102][ 50/ 1200] Overall Loss 0.216807 Objective Loss 0.216807 LR 0.000500 Time 0.036507 -2022-12-06 11:07:59,153 - Epoch: [102][ 60/ 1200] Overall Loss 0.215812 Objective Loss 0.215812 LR 0.000500 Time 0.033736 -2022-12-06 11:07:59,350 - Epoch: [102][ 70/ 1200] Overall Loss 0.217675 Objective Loss 0.217675 LR 0.000500 Time 0.031725 -2022-12-06 11:07:59,549 - Epoch: [102][ 80/ 1200] Overall Loss 0.222850 Objective Loss 0.222850 LR 0.000500 Time 0.030240 -2022-12-06 11:07:59,746 - Epoch: [102][ 90/ 1200] Overall Loss 0.221806 Objective Loss 0.221806 LR 0.000500 Time 0.029066 -2022-12-06 11:07:59,946 - Epoch: [102][ 100/ 1200] Overall Loss 0.220456 Objective Loss 0.220456 LR 0.000500 Time 0.028151 -2022-12-06 11:08:00,144 - Epoch: [102][ 110/ 1200] Overall Loss 0.220860 Objective Loss 0.220860 LR 0.000500 Time 0.027387 -2022-12-06 11:08:00,343 - Epoch: [102][ 120/ 1200] Overall Loss 0.224424 Objective Loss 0.224424 LR 0.000500 Time 0.026758 -2022-12-06 11:08:00,541 - Epoch: [102][ 130/ 1200] Overall Loss 0.224708 Objective Loss 0.224708 LR 0.000500 Time 0.026217 -2022-12-06 11:08:00,739 - Epoch: [102][ 140/ 1200] Overall Loss 0.223917 Objective Loss 0.223917 LR 0.000500 Time 0.025761 -2022-12-06 11:08:00,937 - Epoch: [102][ 150/ 1200] Overall Loss 0.224215 Objective Loss 0.224215 LR 0.000500 Time 0.025356 -2022-12-06 11:08:01,136 - Epoch: [102][ 160/ 1200] Overall Loss 0.225507 Objective Loss 0.225507 LR 0.000500 Time 0.025011 -2022-12-06 11:08:01,333 - Epoch: [102][ 170/ 1200] Overall Loss 0.224275 Objective Loss 0.224275 LR 0.000500 Time 0.024698 -2022-12-06 11:08:01,532 - Epoch: [102][ 180/ 1200] Overall Loss 0.224528 Objective Loss 0.224528 LR 0.000500 Time 0.024429 -2022-12-06 11:08:01,730 - Epoch: [102][ 190/ 1200] Overall Loss 0.225274 Objective Loss 0.225274 LR 0.000500 Time 0.024181 -2022-12-06 11:08:01,929 - Epoch: [102][ 200/ 1200] Overall Loss 0.224630 Objective Loss 0.224630 LR 0.000500 Time 0.023965 -2022-12-06 11:08:02,126 - Epoch: [102][ 210/ 1200] Overall Loss 0.223528 Objective Loss 0.223528 LR 0.000500 Time 0.023760 -2022-12-06 11:08:02,325 - Epoch: [102][ 220/ 1200] Overall Loss 0.222833 Objective Loss 0.222833 LR 0.000500 Time 0.023582 -2022-12-06 11:08:02,523 - Epoch: [102][ 230/ 1200] Overall Loss 0.222292 Objective Loss 0.222292 LR 0.000500 Time 0.023413 -2022-12-06 11:08:02,722 - Epoch: [102][ 240/ 1200] Overall Loss 0.222244 Objective Loss 0.222244 LR 0.000500 Time 0.023264 -2022-12-06 11:08:02,920 - Epoch: [102][ 250/ 1200] Overall Loss 0.221221 Objective Loss 0.221221 LR 0.000500 Time 0.023124 -2022-12-06 11:08:03,119 - Epoch: [102][ 260/ 1200] Overall Loss 0.220791 Objective Loss 0.220791 LR 0.000500 Time 0.022998 -2022-12-06 11:08:03,316 - Epoch: [102][ 270/ 1200] Overall Loss 0.221036 Objective Loss 0.221036 LR 0.000500 Time 0.022876 -2022-12-06 11:08:03,516 - Epoch: [102][ 280/ 1200] Overall Loss 0.220573 Objective Loss 0.220573 LR 0.000500 Time 0.022768 -2022-12-06 11:08:03,714 - Epoch: [102][ 290/ 1200] Overall Loss 0.220323 Objective Loss 0.220323 LR 0.000500 Time 0.022663 -2022-12-06 11:08:03,913 - Epoch: [102][ 300/ 1200] Overall Loss 0.220691 Objective Loss 0.220691 LR 0.000500 Time 0.022571 -2022-12-06 11:08:04,110 - Epoch: [102][ 310/ 1200] Overall Loss 0.221044 Objective Loss 0.221044 LR 0.000500 Time 0.022477 -2022-12-06 11:08:04,309 - Epoch: [102][ 320/ 1200] Overall Loss 0.221051 Objective Loss 0.221051 LR 0.000500 Time 0.022395 -2022-12-06 11:08:04,507 - Epoch: [102][ 330/ 1200] Overall Loss 0.220623 Objective Loss 0.220623 LR 0.000500 Time 0.022315 -2022-12-06 11:08:04,706 - Epoch: [102][ 340/ 1200] Overall Loss 0.219851 Objective Loss 0.219851 LR 0.000500 Time 0.022243 -2022-12-06 11:08:04,904 - Epoch: [102][ 350/ 1200] Overall Loss 0.219848 Objective Loss 0.219848 LR 0.000500 Time 0.022171 -2022-12-06 11:08:05,104 - Epoch: [102][ 360/ 1200] Overall Loss 0.219464 Objective Loss 0.219464 LR 0.000500 Time 0.022108 -2022-12-06 11:08:05,302 - Epoch: [102][ 370/ 1200] Overall Loss 0.219742 Objective Loss 0.219742 LR 0.000500 Time 0.022044 -2022-12-06 11:08:05,501 - Epoch: [102][ 380/ 1200] Overall Loss 0.219847 Objective Loss 0.219847 LR 0.000500 Time 0.021987 -2022-12-06 11:08:05,699 - Epoch: [102][ 390/ 1200] Overall Loss 0.220330 Objective Loss 0.220330 LR 0.000500 Time 0.021929 -2022-12-06 11:08:05,898 - Epoch: [102][ 400/ 1200] Overall Loss 0.220452 Objective Loss 0.220452 LR 0.000500 Time 0.021878 -2022-12-06 11:08:06,096 - Epoch: [102][ 410/ 1200] Overall Loss 0.220412 Objective Loss 0.220412 LR 0.000500 Time 0.021825 -2022-12-06 11:08:06,295 - Epoch: [102][ 420/ 1200] Overall Loss 0.220733 Objective Loss 0.220733 LR 0.000500 Time 0.021779 -2022-12-06 11:08:06,493 - Epoch: [102][ 430/ 1200] Overall Loss 0.221079 Objective Loss 0.221079 LR 0.000500 Time 0.021730 -2022-12-06 11:08:06,693 - Epoch: [102][ 440/ 1200] Overall Loss 0.221267 Objective Loss 0.221267 LR 0.000500 Time 0.021689 -2022-12-06 11:08:06,890 - Epoch: [102][ 450/ 1200] Overall Loss 0.221548 Objective Loss 0.221548 LR 0.000500 Time 0.021644 -2022-12-06 11:08:07,089 - Epoch: [102][ 460/ 1200] Overall Loss 0.221137 Objective Loss 0.221137 LR 0.000500 Time 0.021606 -2022-12-06 11:08:07,287 - Epoch: [102][ 470/ 1200] Overall Loss 0.222132 Objective Loss 0.222132 LR 0.000500 Time 0.021567 -2022-12-06 11:08:07,487 - Epoch: [102][ 480/ 1200] Overall Loss 0.222670 Objective Loss 0.222670 LR 0.000500 Time 0.021532 -2022-12-06 11:08:07,685 - Epoch: [102][ 490/ 1200] Overall Loss 0.222638 Objective Loss 0.222638 LR 0.000500 Time 0.021495 -2022-12-06 11:08:07,885 - Epoch: [102][ 500/ 1200] Overall Loss 0.222605 Objective Loss 0.222605 LR 0.000500 Time 0.021465 -2022-12-06 11:08:08,082 - Epoch: [102][ 510/ 1200] Overall Loss 0.223005 Objective Loss 0.223005 LR 0.000500 Time 0.021430 -2022-12-06 11:08:08,282 - Epoch: [102][ 520/ 1200] Overall Loss 0.223261 Objective Loss 0.223261 LR 0.000500 Time 0.021400 -2022-12-06 11:08:08,480 - Epoch: [102][ 530/ 1200] Overall Loss 0.223500 Objective Loss 0.223500 LR 0.000500 Time 0.021370 -2022-12-06 11:08:08,680 - Epoch: [102][ 540/ 1200] Overall Loss 0.223507 Objective Loss 0.223507 LR 0.000500 Time 0.021342 -2022-12-06 11:08:08,877 - Epoch: [102][ 550/ 1200] Overall Loss 0.223774 Objective Loss 0.223774 LR 0.000500 Time 0.021312 -2022-12-06 11:08:09,077 - Epoch: [102][ 560/ 1200] Overall Loss 0.224040 Objective Loss 0.224040 LR 0.000500 Time 0.021287 -2022-12-06 11:08:09,274 - Epoch: [102][ 570/ 1200] Overall Loss 0.224263 Objective Loss 0.224263 LR 0.000500 Time 0.021259 -2022-12-06 11:08:09,474 - Epoch: [102][ 580/ 1200] Overall Loss 0.223995 Objective Loss 0.223995 LR 0.000500 Time 0.021236 -2022-12-06 11:08:09,672 - Epoch: [102][ 590/ 1200] Overall Loss 0.223949 Objective Loss 0.223949 LR 0.000500 Time 0.021211 -2022-12-06 11:08:09,871 - Epoch: [102][ 600/ 1200] Overall Loss 0.224280 Objective Loss 0.224280 LR 0.000500 Time 0.021189 -2022-12-06 11:08:10,069 - Epoch: [102][ 610/ 1200] Overall Loss 0.224162 Objective Loss 0.224162 LR 0.000500 Time 0.021164 -2022-12-06 11:08:10,269 - Epoch: [102][ 620/ 1200] Overall Loss 0.224279 Objective Loss 0.224279 LR 0.000500 Time 0.021144 -2022-12-06 11:08:10,466 - Epoch: [102][ 630/ 1200] Overall Loss 0.224242 Objective Loss 0.224242 LR 0.000500 Time 0.021121 -2022-12-06 11:08:10,665 - Epoch: [102][ 640/ 1200] Overall Loss 0.224051 Objective Loss 0.224051 LR 0.000500 Time 0.021102 -2022-12-06 11:08:10,863 - Epoch: [102][ 650/ 1200] Overall Loss 0.224559 Objective Loss 0.224559 LR 0.000500 Time 0.021080 -2022-12-06 11:08:11,062 - Epoch: [102][ 660/ 1200] Overall Loss 0.224419 Objective Loss 0.224419 LR 0.000500 Time 0.021062 -2022-12-06 11:08:11,260 - Epoch: [102][ 670/ 1200] Overall Loss 0.224036 Objective Loss 0.224036 LR 0.000500 Time 0.021042 -2022-12-06 11:08:11,459 - Epoch: [102][ 680/ 1200] Overall Loss 0.224442 Objective Loss 0.224442 LR 0.000500 Time 0.021024 -2022-12-06 11:08:11,656 - Epoch: [102][ 690/ 1200] Overall Loss 0.224415 Objective Loss 0.224415 LR 0.000500 Time 0.021004 -2022-12-06 11:08:11,853 - Epoch: [102][ 700/ 1200] Overall Loss 0.224242 Objective Loss 0.224242 LR 0.000500 Time 0.020984 -2022-12-06 11:08:12,048 - Epoch: [102][ 710/ 1200] Overall Loss 0.224333 Objective Loss 0.224333 LR 0.000500 Time 0.020963 -2022-12-06 11:08:12,246 - Epoch: [102][ 720/ 1200] Overall Loss 0.224722 Objective Loss 0.224722 LR 0.000500 Time 0.020946 -2022-12-06 11:08:12,441 - Epoch: [102][ 730/ 1200] Overall Loss 0.224793 Objective Loss 0.224793 LR 0.000500 Time 0.020926 -2022-12-06 11:08:12,639 - Epoch: [102][ 740/ 1200] Overall Loss 0.224591 Objective Loss 0.224591 LR 0.000500 Time 0.020910 -2022-12-06 11:08:12,835 - Epoch: [102][ 750/ 1200] Overall Loss 0.224263 Objective Loss 0.224263 LR 0.000500 Time 0.020891 -2022-12-06 11:08:13,032 - Epoch: [102][ 760/ 1200] Overall Loss 0.224233 Objective Loss 0.224233 LR 0.000500 Time 0.020875 -2022-12-06 11:08:13,228 - Epoch: [102][ 770/ 1200] Overall Loss 0.224049 Objective Loss 0.224049 LR 0.000500 Time 0.020857 -2022-12-06 11:08:13,425 - Epoch: [102][ 780/ 1200] Overall Loss 0.223930 Objective Loss 0.223930 LR 0.000500 Time 0.020843 -2022-12-06 11:08:13,621 - Epoch: [102][ 790/ 1200] Overall Loss 0.223824 Objective Loss 0.223824 LR 0.000500 Time 0.020826 -2022-12-06 11:08:13,819 - Epoch: [102][ 800/ 1200] Overall Loss 0.224126 Objective Loss 0.224126 LR 0.000500 Time 0.020812 -2022-12-06 11:08:14,014 - Epoch: [102][ 810/ 1200] Overall Loss 0.224144 Objective Loss 0.224144 LR 0.000500 Time 0.020796 -2022-12-06 11:08:14,212 - Epoch: [102][ 820/ 1200] Overall Loss 0.224387 Objective Loss 0.224387 LR 0.000500 Time 0.020782 -2022-12-06 11:08:14,407 - Epoch: [102][ 830/ 1200] Overall Loss 0.224684 Objective Loss 0.224684 LR 0.000500 Time 0.020767 -2022-12-06 11:08:14,606 - Epoch: [102][ 840/ 1200] Overall Loss 0.225009 Objective Loss 0.225009 LR 0.000500 Time 0.020755 -2022-12-06 11:08:14,801 - Epoch: [102][ 850/ 1200] Overall Loss 0.225024 Objective Loss 0.225024 LR 0.000500 Time 0.020740 -2022-12-06 11:08:14,999 - Epoch: [102][ 860/ 1200] Overall Loss 0.224854 Objective Loss 0.224854 LR 0.000500 Time 0.020728 -2022-12-06 11:08:15,194 - Epoch: [102][ 870/ 1200] Overall Loss 0.224843 Objective Loss 0.224843 LR 0.000500 Time 0.020714 -2022-12-06 11:08:15,392 - Epoch: [102][ 880/ 1200] Overall Loss 0.224862 Objective Loss 0.224862 LR 0.000500 Time 0.020703 -2022-12-06 11:08:15,588 - Epoch: [102][ 890/ 1200] Overall Loss 0.224716 Objective Loss 0.224716 LR 0.000500 Time 0.020689 -2022-12-06 11:08:15,785 - Epoch: [102][ 900/ 1200] Overall Loss 0.224876 Objective Loss 0.224876 LR 0.000500 Time 0.020678 -2022-12-06 11:08:15,980 - Epoch: [102][ 910/ 1200] Overall Loss 0.224862 Objective Loss 0.224862 LR 0.000500 Time 0.020665 -2022-12-06 11:08:16,179 - Epoch: [102][ 920/ 1200] Overall Loss 0.225042 Objective Loss 0.225042 LR 0.000500 Time 0.020655 -2022-12-06 11:08:16,374 - Epoch: [102][ 930/ 1200] Overall Loss 0.225033 Objective Loss 0.225033 LR 0.000500 Time 0.020643 -2022-12-06 11:08:16,572 - Epoch: [102][ 940/ 1200] Overall Loss 0.225084 Objective Loss 0.225084 LR 0.000500 Time 0.020633 -2022-12-06 11:08:16,768 - Epoch: [102][ 950/ 1200] Overall Loss 0.225124 Objective Loss 0.225124 LR 0.000500 Time 0.020621 -2022-12-06 11:08:16,965 - Epoch: [102][ 960/ 1200] Overall Loss 0.225332 Objective Loss 0.225332 LR 0.000500 Time 0.020611 -2022-12-06 11:08:17,160 - Epoch: [102][ 970/ 1200] Overall Loss 0.225346 Objective Loss 0.225346 LR 0.000500 Time 0.020600 -2022-12-06 11:08:17,358 - Epoch: [102][ 980/ 1200] Overall Loss 0.225538 Objective Loss 0.225538 LR 0.000500 Time 0.020590 -2022-12-06 11:08:17,553 - Epoch: [102][ 990/ 1200] Overall Loss 0.225613 Objective Loss 0.225613 LR 0.000500 Time 0.020579 -2022-12-06 11:08:17,750 - Epoch: [102][ 1000/ 1200] Overall Loss 0.225792 Objective Loss 0.225792 LR 0.000500 Time 0.020570 -2022-12-06 11:08:17,946 - Epoch: [102][ 1010/ 1200] Overall Loss 0.226097 Objective Loss 0.226097 LR 0.000500 Time 0.020560 -2022-12-06 11:08:18,143 - Epoch: [102][ 1020/ 1200] Overall Loss 0.226044 Objective Loss 0.226044 LR 0.000500 Time 0.020551 -2022-12-06 11:08:18,339 - Epoch: [102][ 1030/ 1200] Overall Loss 0.226050 Objective Loss 0.226050 LR 0.000500 Time 0.020541 -2022-12-06 11:08:18,536 - Epoch: [102][ 1040/ 1200] Overall Loss 0.225851 Objective Loss 0.225851 LR 0.000500 Time 0.020532 -2022-12-06 11:08:18,732 - Epoch: [102][ 1050/ 1200] Overall Loss 0.225693 Objective Loss 0.225693 LR 0.000500 Time 0.020523 -2022-12-06 11:08:18,929 - Epoch: [102][ 1060/ 1200] Overall Loss 0.225603 Objective Loss 0.225603 LR 0.000500 Time 0.020515 -2022-12-06 11:08:19,125 - Epoch: [102][ 1070/ 1200] Overall Loss 0.225333 Objective Loss 0.225333 LR 0.000500 Time 0.020505 -2022-12-06 11:08:19,322 - Epoch: [102][ 1080/ 1200] Overall Loss 0.225473 Objective Loss 0.225473 LR 0.000500 Time 0.020498 -2022-12-06 11:08:19,518 - Epoch: [102][ 1090/ 1200] Overall Loss 0.225295 Objective Loss 0.225295 LR 0.000500 Time 0.020489 -2022-12-06 11:08:19,715 - Epoch: [102][ 1100/ 1200] Overall Loss 0.225314 Objective Loss 0.225314 LR 0.000500 Time 0.020481 -2022-12-06 11:08:19,911 - Epoch: [102][ 1110/ 1200] Overall Loss 0.225043 Objective Loss 0.225043 LR 0.000500 Time 0.020472 -2022-12-06 11:08:20,108 - Epoch: [102][ 1120/ 1200] Overall Loss 0.224919 Objective Loss 0.224919 LR 0.000500 Time 0.020465 -2022-12-06 11:08:20,303 - Epoch: [102][ 1130/ 1200] Overall Loss 0.224811 Objective Loss 0.224811 LR 0.000500 Time 0.020456 -2022-12-06 11:08:20,501 - Epoch: [102][ 1140/ 1200] Overall Loss 0.224675 Objective Loss 0.224675 LR 0.000500 Time 0.020450 -2022-12-06 11:08:20,696 - Epoch: [102][ 1150/ 1200] Overall Loss 0.224958 Objective Loss 0.224958 LR 0.000500 Time 0.020441 -2022-12-06 11:08:20,894 - Epoch: [102][ 1160/ 1200] Overall Loss 0.225025 Objective Loss 0.225025 LR 0.000500 Time 0.020435 -2022-12-06 11:08:21,089 - Epoch: [102][ 1170/ 1200] Overall Loss 0.225205 Objective Loss 0.225205 LR 0.000500 Time 0.020427 -2022-12-06 11:08:21,288 - Epoch: [102][ 1180/ 1200] Overall Loss 0.225169 Objective Loss 0.225169 LR 0.000500 Time 0.020422 -2022-12-06 11:08:21,483 - Epoch: [102][ 1190/ 1200] Overall Loss 0.225265 Objective Loss 0.225265 LR 0.000500 Time 0.020414 -2022-12-06 11:08:21,715 - Epoch: [102][ 1200/ 1200] Overall Loss 0.225256 Objective Loss 0.225256 Top1 88.912134 Top5 98.744770 LR 0.000500 Time 0.020437 -2022-12-06 11:08:21,803 - --- validate (epoch=102)----------- -2022-12-06 11:08:21,804 - 34129 samples (256 per mini-batch) -2022-12-06 11:08:22,282 - Epoch: [102][ 10/ 134] Loss 0.271406 Top1 84.882812 Top5 97.890625 -2022-12-06 11:08:22,442 - Epoch: [102][ 20/ 134] Loss 0.267976 Top1 85.449219 Top5 98.164062 -2022-12-06 11:08:22,573 - Epoch: [102][ 30/ 134] Loss 0.276902 Top1 85.273438 Top5 98.059896 -2022-12-06 11:08:22,704 - Epoch: [102][ 40/ 134] Loss 0.264396 Top1 85.751953 Top5 98.115234 -2022-12-06 11:08:22,836 - Epoch: [102][ 50/ 134] Loss 0.270173 Top1 85.476562 Top5 97.976562 -2022-12-06 11:08:22,965 - Epoch: [102][ 60/ 134] Loss 0.271014 Top1 85.442708 Top5 97.949219 -2022-12-06 11:08:23,094 - Epoch: [102][ 70/ 134] Loss 0.272642 Top1 85.401786 Top5 97.963170 -2022-12-06 11:08:23,223 - Epoch: [102][ 80/ 134] Loss 0.272818 Top1 85.498047 Top5 97.993164 -2022-12-06 11:08:23,356 - Epoch: [102][ 90/ 134] Loss 0.270995 Top1 85.529514 Top5 98.012153 -2022-12-06 11:08:23,486 - Epoch: [102][ 100/ 134] Loss 0.266984 Top1 85.640625 Top5 98.031250 -2022-12-06 11:08:23,615 - Epoch: [102][ 110/ 134] Loss 0.268715 Top1 85.646307 Top5 98.043324 -2022-12-06 11:08:23,748 - Epoch: [102][ 120/ 134] Loss 0.269048 Top1 85.618490 Top5 98.056641 -2022-12-06 11:08:23,882 - Epoch: [102][ 130/ 134] Loss 0.269226 Top1 85.564904 Top5 98.043870 -2022-12-06 11:08:23,921 - Epoch: [102][ 134/ 134] Loss 0.270243 Top1 85.510856 Top5 98.028070 -2022-12-06 11:08:24,011 - ==> Top1: 85.511 Top5: 98.028 Loss: 0.270 - -2022-12-06 11:08:24,011 - ==> Confusion: -[[ 922 0 1 2 6 3 0 1 7 37 0 4 1 2 3 2 2 0 1 0 2] - [ 2 954 0 2 9 19 1 8 3 0 2 6 1 0 0 1 3 0 9 2 5] - [ 6 2 989 14 4 0 17 7 2 1 13 8 2 1 3 6 1 0 9 0 18] - [ 2 1 9 940 0 1 0 1 1 1 12 0 4 1 20 2 0 2 15 1 7] - [ 12 5 1 1 958 3 0 0 2 4 1 4 0 1 8 6 6 1 0 2 5] - [ 4 23 0 3 8 957 3 19 5 1 0 14 3 8 0 1 4 1 3 6 6] - [ 2 3 8 2 0 1 1073 3 0 0 3 2 1 1 0 5 1 4 2 7 0] - [ 2 10 9 1 1 27 5 935 0 2 3 10 1 1 1 2 0 0 22 17 5] - [ 6 4 0 0 0 3 0 1 977 36 10 2 1 3 12 1 0 2 3 1 2] - [ 70 0 4 0 10 0 1 1 34 853 1 3 0 8 3 1 1 1 1 1 8] - [ 3 2 1 4 2 1 2 0 4 0 972 4 0 9 3 1 1 0 3 3 4] - [ 2 0 0 0 0 12 1 1 1 0 1 978 23 3 1 7 5 2 1 9 4] - [ 3 1 0 2 1 2 0 0 1 0 0 27 897 2 2 10 2 11 0 4 4] - [ 0 0 1 0 1 8 0 2 22 17 13 5 5 924 4 2 5 2 1 3 8] - [ 9 4 1 7 4 1 0 1 23 0 2 3 1 2 1056 0 0 2 8 1 5] - [ 1 0 1 0 2 2 3 0 0 0 0 7 3 2 0 995 6 13 1 4 3] - [ 2 2 1 1 2 0 2 0 0 0 0 2 1 1 2 18 1030 0 2 3 3] - [ 2 0 1 1 0 2 1 1 1 2 0 4 16 2 2 14 1 983 0 1 2] - [ 4 5 1 14 0 1 1 19 1 1 5 3 3 1 5 0 1 0 938 2 3] - [ 4 3 1 0 0 5 3 5 0 0 1 23 7 6 0 10 5 4 1 994 8] - [ 166 266 154 119 110 152 98 122 133 88 221 112 367 256 176 117 240 91 182 204 9852]] - -2022-12-06 11:08:24,579 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:08:24,579 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:08:24,585 - - -2022-12-06 11:08:24,585 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:08:25,505 - Epoch: [103][ 10/ 1200] Overall Loss 0.227797 Objective Loss 0.227797 LR 0.000500 Time 0.091906 -2022-12-06 11:08:25,706 - Epoch: [103][ 20/ 1200] Overall Loss 0.222380 Objective Loss 0.222380 LR 0.000500 Time 0.055988 -2022-12-06 11:08:25,898 - Epoch: [103][ 30/ 1200] Overall Loss 0.221457 Objective Loss 0.221457 LR 0.000500 Time 0.043702 -2022-12-06 11:08:26,089 - Epoch: [103][ 40/ 1200] Overall Loss 0.221745 Objective Loss 0.221745 LR 0.000500 Time 0.037532 -2022-12-06 11:08:26,280 - Epoch: [103][ 50/ 1200] Overall Loss 0.220611 Objective Loss 0.220611 LR 0.000500 Time 0.033835 -2022-12-06 11:08:26,470 - Epoch: [103][ 60/ 1200] Overall Loss 0.217851 Objective Loss 0.217851 LR 0.000500 Time 0.031362 -2022-12-06 11:08:26,661 - Epoch: [103][ 70/ 1200] Overall Loss 0.217953 Objective Loss 0.217953 LR 0.000500 Time 0.029605 -2022-12-06 11:08:26,852 - Epoch: [103][ 80/ 1200] Overall Loss 0.216901 Objective Loss 0.216901 LR 0.000500 Time 0.028277 -2022-12-06 11:08:27,042 - Epoch: [103][ 90/ 1200] Overall Loss 0.217031 Objective Loss 0.217031 LR 0.000500 Time 0.027246 -2022-12-06 11:08:27,233 - Epoch: [103][ 100/ 1200] Overall Loss 0.217276 Objective Loss 0.217276 LR 0.000500 Time 0.026420 -2022-12-06 11:08:27,424 - Epoch: [103][ 110/ 1200] Overall Loss 0.218260 Objective Loss 0.218260 LR 0.000500 Time 0.025747 -2022-12-06 11:08:27,614 - Epoch: [103][ 120/ 1200] Overall Loss 0.219242 Objective Loss 0.219242 LR 0.000500 Time 0.025186 -2022-12-06 11:08:27,805 - Epoch: [103][ 130/ 1200] Overall Loss 0.221406 Objective Loss 0.221406 LR 0.000500 Time 0.024714 -2022-12-06 11:08:27,996 - Epoch: [103][ 140/ 1200] Overall Loss 0.221038 Objective Loss 0.221038 LR 0.000500 Time 0.024307 -2022-12-06 11:08:28,187 - Epoch: [103][ 150/ 1200] Overall Loss 0.222231 Objective Loss 0.222231 LR 0.000500 Time 0.023954 -2022-12-06 11:08:28,378 - Epoch: [103][ 160/ 1200] Overall Loss 0.222489 Objective Loss 0.222489 LR 0.000500 Time 0.023647 -2022-12-06 11:08:28,569 - Epoch: [103][ 170/ 1200] Overall Loss 0.224099 Objective Loss 0.224099 LR 0.000500 Time 0.023376 -2022-12-06 11:08:28,759 - Epoch: [103][ 180/ 1200] Overall Loss 0.221986 Objective Loss 0.221986 LR 0.000500 Time 0.023133 -2022-12-06 11:08:28,950 - Epoch: [103][ 190/ 1200] Overall Loss 0.220256 Objective Loss 0.220256 LR 0.000500 Time 0.022916 -2022-12-06 11:08:29,140 - Epoch: [103][ 200/ 1200] Overall Loss 0.220636 Objective Loss 0.220636 LR 0.000500 Time 0.022718 -2022-12-06 11:08:29,330 - Epoch: [103][ 210/ 1200] Overall Loss 0.220227 Objective Loss 0.220227 LR 0.000500 Time 0.022540 -2022-12-06 11:08:29,520 - Epoch: [103][ 220/ 1200] Overall Loss 0.219808 Objective Loss 0.219808 LR 0.000500 Time 0.022376 -2022-12-06 11:08:29,710 - Epoch: [103][ 230/ 1200] Overall Loss 0.218933 Objective Loss 0.218933 LR 0.000500 Time 0.022228 -2022-12-06 11:08:29,901 - Epoch: [103][ 240/ 1200] Overall Loss 0.219671 Objective Loss 0.219671 LR 0.000500 Time 0.022092 -2022-12-06 11:08:30,091 - Epoch: [103][ 250/ 1200] Overall Loss 0.219598 Objective Loss 0.219598 LR 0.000500 Time 0.021969 -2022-12-06 11:08:30,282 - Epoch: [103][ 260/ 1200] Overall Loss 0.220136 Objective Loss 0.220136 LR 0.000500 Time 0.021854 -2022-12-06 11:08:30,472 - Epoch: [103][ 270/ 1200] Overall Loss 0.220864 Objective Loss 0.220864 LR 0.000500 Time 0.021749 -2022-12-06 11:08:30,663 - Epoch: [103][ 280/ 1200] Overall Loss 0.221119 Objective Loss 0.221119 LR 0.000500 Time 0.021651 -2022-12-06 11:08:30,854 - Epoch: [103][ 290/ 1200] Overall Loss 0.221667 Objective Loss 0.221667 LR 0.000500 Time 0.021561 -2022-12-06 11:08:31,044 - Epoch: [103][ 300/ 1200] Overall Loss 0.221527 Objective Loss 0.221527 LR 0.000500 Time 0.021475 -2022-12-06 11:08:31,235 - Epoch: [103][ 310/ 1200] Overall Loss 0.220262 Objective Loss 0.220262 LR 0.000500 Time 0.021396 -2022-12-06 11:08:31,426 - Epoch: [103][ 320/ 1200] Overall Loss 0.220793 Objective Loss 0.220793 LR 0.000500 Time 0.021322 -2022-12-06 11:08:31,628 - Epoch: [103][ 330/ 1200] Overall Loss 0.220640 Objective Loss 0.220640 LR 0.000500 Time 0.021285 -2022-12-06 11:08:31,826 - Epoch: [103][ 340/ 1200] Overall Loss 0.220169 Objective Loss 0.220169 LR 0.000500 Time 0.021240 -2022-12-06 11:08:32,027 - Epoch: [103][ 350/ 1200] Overall Loss 0.220157 Objective Loss 0.220157 LR 0.000500 Time 0.021206 -2022-12-06 11:08:32,226 - Epoch: [103][ 360/ 1200] Overall Loss 0.220317 Objective Loss 0.220317 LR 0.000500 Time 0.021169 -2022-12-06 11:08:32,427 - Epoch: [103][ 370/ 1200] Overall Loss 0.220471 Objective Loss 0.220471 LR 0.000500 Time 0.021138 -2022-12-06 11:08:32,625 - Epoch: [103][ 380/ 1200] Overall Loss 0.220029 Objective Loss 0.220029 LR 0.000500 Time 0.021102 -2022-12-06 11:08:32,827 - Epoch: [103][ 390/ 1200] Overall Loss 0.219685 Objective Loss 0.219685 LR 0.000500 Time 0.021077 -2022-12-06 11:08:33,026 - Epoch: [103][ 400/ 1200] Overall Loss 0.219088 Objective Loss 0.219088 LR 0.000500 Time 0.021046 -2022-12-06 11:08:33,227 - Epoch: [103][ 410/ 1200] Overall Loss 0.218882 Objective Loss 0.218882 LR 0.000500 Time 0.021023 -2022-12-06 11:08:33,427 - Epoch: [103][ 420/ 1200] Overall Loss 0.218946 Objective Loss 0.218946 LR 0.000500 Time 0.020996 -2022-12-06 11:08:33,629 - Epoch: [103][ 430/ 1200] Overall Loss 0.218829 Objective Loss 0.218829 LR 0.000500 Time 0.020976 -2022-12-06 11:08:33,827 - Epoch: [103][ 440/ 1200] Overall Loss 0.219200 Objective Loss 0.219200 LR 0.000500 Time 0.020949 -2022-12-06 11:08:34,029 - Epoch: [103][ 450/ 1200] Overall Loss 0.219413 Objective Loss 0.219413 LR 0.000500 Time 0.020931 -2022-12-06 11:08:34,227 - Epoch: [103][ 460/ 1200] Overall Loss 0.219023 Objective Loss 0.219023 LR 0.000500 Time 0.020906 -2022-12-06 11:08:34,429 - Epoch: [103][ 470/ 1200] Overall Loss 0.218939 Objective Loss 0.218939 LR 0.000500 Time 0.020888 -2022-12-06 11:08:34,627 - Epoch: [103][ 480/ 1200] Overall Loss 0.218865 Objective Loss 0.218865 LR 0.000500 Time 0.020865 -2022-12-06 11:08:34,828 - Epoch: [103][ 490/ 1200] Overall Loss 0.218497 Objective Loss 0.218497 LR 0.000500 Time 0.020849 -2022-12-06 11:08:35,027 - Epoch: [103][ 500/ 1200] Overall Loss 0.218704 Objective Loss 0.218704 LR 0.000500 Time 0.020827 -2022-12-06 11:08:35,228 - Epoch: [103][ 510/ 1200] Overall Loss 0.219100 Objective Loss 0.219100 LR 0.000500 Time 0.020813 -2022-12-06 11:08:35,427 - Epoch: [103][ 520/ 1200] Overall Loss 0.219413 Objective Loss 0.219413 LR 0.000500 Time 0.020794 -2022-12-06 11:08:35,629 - Epoch: [103][ 530/ 1200] Overall Loss 0.219448 Objective Loss 0.219448 LR 0.000500 Time 0.020782 -2022-12-06 11:08:35,829 - Epoch: [103][ 540/ 1200] Overall Loss 0.219387 Objective Loss 0.219387 LR 0.000500 Time 0.020766 -2022-12-06 11:08:36,031 - Epoch: [103][ 550/ 1200] Overall Loss 0.219699 Objective Loss 0.219699 LR 0.000500 Time 0.020755 -2022-12-06 11:08:36,230 - Epoch: [103][ 560/ 1200] Overall Loss 0.219804 Objective Loss 0.219804 LR 0.000500 Time 0.020739 -2022-12-06 11:08:36,432 - Epoch: [103][ 570/ 1200] Overall Loss 0.219963 Objective Loss 0.219963 LR 0.000500 Time 0.020729 -2022-12-06 11:08:36,632 - Epoch: [103][ 580/ 1200] Overall Loss 0.219717 Objective Loss 0.219717 LR 0.000500 Time 0.020715 -2022-12-06 11:08:36,834 - Epoch: [103][ 590/ 1200] Overall Loss 0.219611 Objective Loss 0.219611 LR 0.000500 Time 0.020705 -2022-12-06 11:08:37,033 - Epoch: [103][ 600/ 1200] Overall Loss 0.219522 Objective Loss 0.219522 LR 0.000500 Time 0.020690 -2022-12-06 11:08:37,234 - Epoch: [103][ 610/ 1200] Overall Loss 0.219078 Objective Loss 0.219078 LR 0.000500 Time 0.020681 -2022-12-06 11:08:37,433 - Epoch: [103][ 620/ 1200] Overall Loss 0.219033 Objective Loss 0.219033 LR 0.000500 Time 0.020666 -2022-12-06 11:08:37,633 - Epoch: [103][ 630/ 1200] Overall Loss 0.219264 Objective Loss 0.219264 LR 0.000500 Time 0.020655 -2022-12-06 11:08:37,830 - Epoch: [103][ 640/ 1200] Overall Loss 0.219406 Objective Loss 0.219406 LR 0.000500 Time 0.020638 -2022-12-06 11:08:38,029 - Epoch: [103][ 650/ 1200] Overall Loss 0.219480 Objective Loss 0.219480 LR 0.000500 Time 0.020627 -2022-12-06 11:08:38,226 - Epoch: [103][ 660/ 1200] Overall Loss 0.219518 Objective Loss 0.219518 LR 0.000500 Time 0.020612 -2022-12-06 11:08:38,426 - Epoch: [103][ 670/ 1200] Overall Loss 0.219753 Objective Loss 0.219753 LR 0.000500 Time 0.020602 -2022-12-06 11:08:38,622 - Epoch: [103][ 680/ 1200] Overall Loss 0.220005 Objective Loss 0.220005 LR 0.000500 Time 0.020587 -2022-12-06 11:08:38,822 - Epoch: [103][ 690/ 1200] Overall Loss 0.220287 Objective Loss 0.220287 LR 0.000500 Time 0.020577 -2022-12-06 11:08:39,019 - Epoch: [103][ 700/ 1200] Overall Loss 0.220195 Objective Loss 0.220195 LR 0.000500 Time 0.020564 -2022-12-06 11:08:39,218 - Epoch: [103][ 710/ 1200] Overall Loss 0.220401 Objective Loss 0.220401 LR 0.000500 Time 0.020554 -2022-12-06 11:08:39,415 - Epoch: [103][ 720/ 1200] Overall Loss 0.220520 Objective Loss 0.220520 LR 0.000500 Time 0.020541 -2022-12-06 11:08:39,615 - Epoch: [103][ 730/ 1200] Overall Loss 0.220275 Objective Loss 0.220275 LR 0.000500 Time 0.020532 -2022-12-06 11:08:39,811 - Epoch: [103][ 740/ 1200] Overall Loss 0.220411 Objective Loss 0.220411 LR 0.000500 Time 0.020520 -2022-12-06 11:08:40,010 - Epoch: [103][ 750/ 1200] Overall Loss 0.220338 Objective Loss 0.220338 LR 0.000500 Time 0.020511 -2022-12-06 11:08:40,208 - Epoch: [103][ 760/ 1200] Overall Loss 0.220049 Objective Loss 0.220049 LR 0.000500 Time 0.020500 -2022-12-06 11:08:40,407 - Epoch: [103][ 770/ 1200] Overall Loss 0.220188 Objective Loss 0.220188 LR 0.000500 Time 0.020492 -2022-12-06 11:08:40,604 - Epoch: [103][ 780/ 1200] Overall Loss 0.220334 Objective Loss 0.220334 LR 0.000500 Time 0.020481 -2022-12-06 11:08:40,804 - Epoch: [103][ 790/ 1200] Overall Loss 0.220437 Objective Loss 0.220437 LR 0.000500 Time 0.020474 -2022-12-06 11:08:41,001 - Epoch: [103][ 800/ 1200] Overall Loss 0.220218 Objective Loss 0.220218 LR 0.000500 Time 0.020464 -2022-12-06 11:08:41,201 - Epoch: [103][ 810/ 1200] Overall Loss 0.220158 Objective Loss 0.220158 LR 0.000500 Time 0.020458 -2022-12-06 11:08:41,399 - Epoch: [103][ 820/ 1200] Overall Loss 0.220404 Objective Loss 0.220404 LR 0.000500 Time 0.020448 -2022-12-06 11:08:41,599 - Epoch: [103][ 830/ 1200] Overall Loss 0.220899 Objective Loss 0.220899 LR 0.000500 Time 0.020442 -2022-12-06 11:08:41,796 - Epoch: [103][ 840/ 1200] Overall Loss 0.220816 Objective Loss 0.220816 LR 0.000500 Time 0.020433 -2022-12-06 11:08:41,996 - Epoch: [103][ 850/ 1200] Overall Loss 0.220964 Objective Loss 0.220964 LR 0.000500 Time 0.020427 -2022-12-06 11:08:42,192 - Epoch: [103][ 860/ 1200] Overall Loss 0.220665 Objective Loss 0.220665 LR 0.000500 Time 0.020418 -2022-12-06 11:08:42,392 - Epoch: [103][ 870/ 1200] Overall Loss 0.220774 Objective Loss 0.220774 LR 0.000500 Time 0.020411 -2022-12-06 11:08:42,588 - Epoch: [103][ 880/ 1200] Overall Loss 0.220672 Objective Loss 0.220672 LR 0.000500 Time 0.020402 -2022-12-06 11:08:42,787 - Epoch: [103][ 890/ 1200] Overall Loss 0.220760 Objective Loss 0.220760 LR 0.000500 Time 0.020396 -2022-12-06 11:08:42,985 - Epoch: [103][ 900/ 1200] Overall Loss 0.221038 Objective Loss 0.221038 LR 0.000500 Time 0.020388 -2022-12-06 11:08:43,186 - Epoch: [103][ 910/ 1200] Overall Loss 0.220961 Objective Loss 0.220961 LR 0.000500 Time 0.020384 -2022-12-06 11:08:43,381 - Epoch: [103][ 920/ 1200] Overall Loss 0.221173 Objective Loss 0.221173 LR 0.000500 Time 0.020375 -2022-12-06 11:08:43,581 - Epoch: [103][ 930/ 1200] Overall Loss 0.221147 Objective Loss 0.221147 LR 0.000500 Time 0.020370 -2022-12-06 11:08:43,778 - Epoch: [103][ 940/ 1200] Overall Loss 0.221044 Objective Loss 0.221044 LR 0.000500 Time 0.020362 -2022-12-06 11:08:43,978 - Epoch: [103][ 950/ 1200] Overall Loss 0.221120 Objective Loss 0.221120 LR 0.000500 Time 0.020357 -2022-12-06 11:08:44,175 - Epoch: [103][ 960/ 1200] Overall Loss 0.221245 Objective Loss 0.221245 LR 0.000500 Time 0.020350 -2022-12-06 11:08:44,375 - Epoch: [103][ 970/ 1200] Overall Loss 0.221383 Objective Loss 0.221383 LR 0.000500 Time 0.020346 -2022-12-06 11:08:44,572 - Epoch: [103][ 980/ 1200] Overall Loss 0.221558 Objective Loss 0.221558 LR 0.000500 Time 0.020338 -2022-12-06 11:08:44,772 - Epoch: [103][ 990/ 1200] Overall Loss 0.221482 Objective Loss 0.221482 LR 0.000500 Time 0.020334 -2022-12-06 11:08:44,968 - Epoch: [103][ 1000/ 1200] Overall Loss 0.221437 Objective Loss 0.221437 LR 0.000500 Time 0.020327 -2022-12-06 11:08:45,168 - Epoch: [103][ 1010/ 1200] Overall Loss 0.221406 Objective Loss 0.221406 LR 0.000500 Time 0.020323 -2022-12-06 11:08:45,365 - Epoch: [103][ 1020/ 1200] Overall Loss 0.221395 Objective Loss 0.221395 LR 0.000500 Time 0.020316 -2022-12-06 11:08:45,564 - Epoch: [103][ 1030/ 1200] Overall Loss 0.221558 Objective Loss 0.221558 LR 0.000500 Time 0.020312 -2022-12-06 11:08:45,761 - Epoch: [103][ 1040/ 1200] Overall Loss 0.221267 Objective Loss 0.221267 LR 0.000500 Time 0.020306 -2022-12-06 11:08:45,961 - Epoch: [103][ 1050/ 1200] Overall Loss 0.221383 Objective Loss 0.221383 LR 0.000500 Time 0.020302 -2022-12-06 11:08:46,158 - Epoch: [103][ 1060/ 1200] Overall Loss 0.221332 Objective Loss 0.221332 LR 0.000500 Time 0.020296 -2022-12-06 11:08:46,358 - Epoch: [103][ 1070/ 1200] Overall Loss 0.221491 Objective Loss 0.221491 LR 0.000500 Time 0.020292 -2022-12-06 11:08:46,554 - Epoch: [103][ 1080/ 1200] Overall Loss 0.221510 Objective Loss 0.221510 LR 0.000500 Time 0.020286 -2022-12-06 11:08:46,754 - Epoch: [103][ 1090/ 1200] Overall Loss 0.221619 Objective Loss 0.221619 LR 0.000500 Time 0.020283 -2022-12-06 11:08:46,951 - Epoch: [103][ 1100/ 1200] Overall Loss 0.221645 Objective Loss 0.221645 LR 0.000500 Time 0.020277 -2022-12-06 11:08:47,151 - Epoch: [103][ 1110/ 1200] Overall Loss 0.221696 Objective Loss 0.221696 LR 0.000500 Time 0.020273 -2022-12-06 11:08:47,348 - Epoch: [103][ 1120/ 1200] Overall Loss 0.221582 Objective Loss 0.221582 LR 0.000500 Time 0.020268 -2022-12-06 11:08:47,548 - Epoch: [103][ 1130/ 1200] Overall Loss 0.221694 Objective Loss 0.221694 LR 0.000500 Time 0.020265 -2022-12-06 11:08:47,744 - Epoch: [103][ 1140/ 1200] Overall Loss 0.221678 Objective Loss 0.221678 LR 0.000500 Time 0.020259 -2022-12-06 11:08:47,944 - Epoch: [103][ 1150/ 1200] Overall Loss 0.221812 Objective Loss 0.221812 LR 0.000500 Time 0.020256 -2022-12-06 11:08:48,141 - Epoch: [103][ 1160/ 1200] Overall Loss 0.221952 Objective Loss 0.221952 LR 0.000500 Time 0.020251 -2022-12-06 11:08:48,341 - Epoch: [103][ 1170/ 1200] Overall Loss 0.222099 Objective Loss 0.222099 LR 0.000500 Time 0.020248 -2022-12-06 11:08:48,538 - Epoch: [103][ 1180/ 1200] Overall Loss 0.221928 Objective Loss 0.221928 LR 0.000500 Time 0.020243 -2022-12-06 11:08:48,738 - Epoch: [103][ 1190/ 1200] Overall Loss 0.221847 Objective Loss 0.221847 LR 0.000500 Time 0.020240 -2022-12-06 11:08:48,965 - Epoch: [103][ 1200/ 1200] Overall Loss 0.221763 Objective Loss 0.221763 Top1 85.355649 Top5 97.907950 LR 0.000500 Time 0.020260 -2022-12-06 11:08:49,054 - --- validate (epoch=103)----------- -2022-12-06 11:08:49,054 - 34129 samples (256 per mini-batch) -2022-12-06 11:08:49,620 - Epoch: [103][ 10/ 134] Loss 0.246572 Top1 86.015625 Top5 98.320312 -2022-12-06 11:08:49,751 - Epoch: [103][ 20/ 134] Loss 0.255586 Top1 85.957031 Top5 98.222656 -2022-12-06 11:08:49,881 - Epoch: [103][ 30/ 134] Loss 0.257953 Top1 86.341146 Top5 98.294271 -2022-12-06 11:08:50,009 - Epoch: [103][ 40/ 134] Loss 0.257970 Top1 86.259766 Top5 98.310547 -2022-12-06 11:08:50,137 - Epoch: [103][ 50/ 134] Loss 0.264699 Top1 86.054688 Top5 98.203125 -2022-12-06 11:08:50,266 - Epoch: [103][ 60/ 134] Loss 0.270708 Top1 85.800781 Top5 98.138021 -2022-12-06 11:08:50,394 - Epoch: [103][ 70/ 134] Loss 0.267116 Top1 85.982143 Top5 98.158482 -2022-12-06 11:08:50,522 - Epoch: [103][ 80/ 134] Loss 0.266510 Top1 85.942383 Top5 98.129883 -2022-12-06 11:08:50,650 - Epoch: [103][ 90/ 134] Loss 0.266623 Top1 86.019965 Top5 98.155382 -2022-12-06 11:08:50,778 - Epoch: [103][ 100/ 134] Loss 0.264538 Top1 86.015625 Top5 98.175781 -2022-12-06 11:08:50,908 - Epoch: [103][ 110/ 134] Loss 0.263525 Top1 86.058239 Top5 98.203125 -2022-12-06 11:08:51,035 - Epoch: [103][ 120/ 134] Loss 0.264148 Top1 86.129557 Top5 98.199870 -2022-12-06 11:08:51,165 - Epoch: [103][ 130/ 134] Loss 0.265799 Top1 86.078726 Top5 98.206130 -2022-12-06 11:08:51,202 - Epoch: [103][ 134/ 134] Loss 0.265529 Top1 86.091008 Top5 98.221454 -2022-12-06 11:08:51,289 - ==> Top1: 86.091 Top5: 98.221 Loss: 0.266 - -2022-12-06 11:08:51,290 - ==> Confusion: -[[ 911 1 2 3 9 6 0 1 3 39 0 2 2 3 5 2 2 0 1 0 4] - [ 3 943 3 2 5 24 3 11 2 0 3 6 0 0 1 1 3 0 11 1 5] - [ 7 1 998 14 2 1 26 14 2 2 3 5 3 0 1 5 1 1 1 3 13] - [ 4 2 15 944 2 2 0 0 0 0 7 0 4 1 12 0 3 2 16 0 6] - [ 4 7 1 1 956 4 2 1 1 6 1 4 0 2 8 7 5 2 2 1 5] - [ 2 19 0 3 4 977 1 17 4 3 0 10 3 14 2 1 2 0 0 5 2] - [ 1 2 5 2 0 1 1081 4 0 0 3 1 3 3 0 2 0 1 1 6 2] - [ 2 6 10 2 2 31 6 947 0 0 1 5 2 1 1 0 0 0 17 14 7] - [ 5 5 0 2 0 3 0 0 978 37 9 1 3 7 8 1 1 1 2 1 0] - [ 64 0 3 2 6 2 0 1 39 859 1 1 0 11 1 1 2 1 0 1 6] - [ 0 3 2 6 1 3 1 3 7 0 959 3 1 9 5 0 0 0 6 5 5] - [ 3 1 1 0 0 12 2 2 1 1 1 973 26 3 0 7 2 6 0 6 4] - [ 3 0 0 1 2 1 2 1 0 0 0 24 901 2 0 8 1 16 0 3 4] - [ 2 0 1 0 0 12 0 1 13 9 7 4 3 953 0 1 3 1 1 5 7] - [ 4 3 1 12 1 1 0 1 21 3 0 5 3 3 1049 1 2 1 14 0 5] - [ 1 0 1 2 1 3 2 0 1 1 0 13 7 0 0 992 5 7 1 3 3] - [ 2 1 0 1 0 1 1 0 1 0 1 3 1 2 1 15 1035 0 0 5 2] - [ 3 0 1 6 0 0 0 1 1 1 0 8 16 2 1 14 1 976 0 2 3] - [ 4 5 0 9 1 3 2 24 2 0 6 3 4 1 3 1 0 0 935 1 4] - [ 4 2 2 0 0 9 4 7 0 0 1 15 6 7 0 5 7 6 0 1000 5] - [ 133 248 161 110 85 230 92 135 94 92 147 114 382 276 106 105 215 94 166 225 10016]] - -2022-12-06 11:08:51,870 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:08:51,870 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:08:51,876 - - -2022-12-06 11:08:51,876 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:08:52,814 - Epoch: [104][ 10/ 1200] Overall Loss 0.220089 Objective Loss 0.220089 LR 0.000500 Time 0.093746 -2022-12-06 11:08:53,022 - Epoch: [104][ 20/ 1200] Overall Loss 0.223059 Objective Loss 0.223059 LR 0.000500 Time 0.057255 -2022-12-06 11:08:53,227 - Epoch: [104][ 30/ 1200] Overall Loss 0.216550 Objective Loss 0.216550 LR 0.000500 Time 0.044974 -2022-12-06 11:08:53,430 - Epoch: [104][ 40/ 1200] Overall Loss 0.215509 Objective Loss 0.215509 LR 0.000500 Time 0.038784 -2022-12-06 11:08:53,634 - Epoch: [104][ 50/ 1200] Overall Loss 0.218009 Objective Loss 0.218009 LR 0.000500 Time 0.035106 -2022-12-06 11:08:53,836 - Epoch: [104][ 60/ 1200] Overall Loss 0.218443 Objective Loss 0.218443 LR 0.000500 Time 0.032607 -2022-12-06 11:08:54,035 - Epoch: [104][ 70/ 1200] Overall Loss 0.216757 Objective Loss 0.216757 LR 0.000500 Time 0.030789 -2022-12-06 11:08:54,232 - Epoch: [104][ 80/ 1200] Overall Loss 0.215341 Objective Loss 0.215341 LR 0.000500 Time 0.029397 -2022-12-06 11:08:54,432 - Epoch: [104][ 90/ 1200] Overall Loss 0.215694 Objective Loss 0.215694 LR 0.000500 Time 0.028349 -2022-12-06 11:08:54,629 - Epoch: [104][ 100/ 1200] Overall Loss 0.216862 Objective Loss 0.216862 LR 0.000500 Time 0.027478 -2022-12-06 11:08:54,828 - Epoch: [104][ 110/ 1200] Overall Loss 0.217587 Objective Loss 0.217587 LR 0.000500 Time 0.026784 -2022-12-06 11:08:55,026 - Epoch: [104][ 120/ 1200] Overall Loss 0.216452 Objective Loss 0.216452 LR 0.000500 Time 0.026197 -2022-12-06 11:08:55,225 - Epoch: [104][ 130/ 1200] Overall Loss 0.215937 Objective Loss 0.215937 LR 0.000500 Time 0.025709 -2022-12-06 11:08:55,423 - Epoch: [104][ 140/ 1200] Overall Loss 0.213746 Objective Loss 0.213746 LR 0.000500 Time 0.025279 -2022-12-06 11:08:55,622 - Epoch: [104][ 150/ 1200] Overall Loss 0.214495 Objective Loss 0.214495 LR 0.000500 Time 0.024917 -2022-12-06 11:08:55,819 - Epoch: [104][ 160/ 1200] Overall Loss 0.214928 Objective Loss 0.214928 LR 0.000500 Time 0.024589 -2022-12-06 11:08:56,017 - Epoch: [104][ 170/ 1200] Overall Loss 0.214551 Objective Loss 0.214551 LR 0.000500 Time 0.024305 -2022-12-06 11:08:56,215 - Epoch: [104][ 180/ 1200] Overall Loss 0.216092 Objective Loss 0.216092 LR 0.000500 Time 0.024051 -2022-12-06 11:08:56,415 - Epoch: [104][ 190/ 1200] Overall Loss 0.216193 Objective Loss 0.216193 LR 0.000500 Time 0.023836 -2022-12-06 11:08:56,612 - Epoch: [104][ 200/ 1200] Overall Loss 0.218141 Objective Loss 0.218141 LR 0.000500 Time 0.023624 -2022-12-06 11:08:56,812 - Epoch: [104][ 210/ 1200] Overall Loss 0.218534 Objective Loss 0.218534 LR 0.000500 Time 0.023447 -2022-12-06 11:08:57,008 - Epoch: [104][ 220/ 1200] Overall Loss 0.217875 Objective Loss 0.217875 LR 0.000500 Time 0.023273 -2022-12-06 11:08:57,208 - Epoch: [104][ 230/ 1200] Overall Loss 0.217727 Objective Loss 0.217727 LR 0.000500 Time 0.023128 -2022-12-06 11:08:57,405 - Epoch: [104][ 240/ 1200] Overall Loss 0.217569 Objective Loss 0.217569 LR 0.000500 Time 0.022984 -2022-12-06 11:08:57,605 - Epoch: [104][ 250/ 1200] Overall Loss 0.218295 Objective Loss 0.218295 LR 0.000500 Time 0.022861 -2022-12-06 11:08:57,803 - Epoch: [104][ 260/ 1200] Overall Loss 0.217556 Objective Loss 0.217556 LR 0.000500 Time 0.022740 -2022-12-06 11:08:58,002 - Epoch: [104][ 270/ 1200] Overall Loss 0.218855 Objective Loss 0.218855 LR 0.000500 Time 0.022634 -2022-12-06 11:08:58,198 - Epoch: [104][ 280/ 1200] Overall Loss 0.218441 Objective Loss 0.218441 LR 0.000500 Time 0.022524 -2022-12-06 11:08:58,397 - Epoch: [104][ 290/ 1200] Overall Loss 0.218451 Objective Loss 0.218451 LR 0.000500 Time 0.022431 -2022-12-06 11:08:58,594 - Epoch: [104][ 300/ 1200] Overall Loss 0.218018 Objective Loss 0.218018 LR 0.000500 Time 0.022338 -2022-12-06 11:08:58,793 - Epoch: [104][ 310/ 1200] Overall Loss 0.217120 Objective Loss 0.217120 LR 0.000500 Time 0.022256 -2022-12-06 11:08:58,989 - Epoch: [104][ 320/ 1200] Overall Loss 0.216168 Objective Loss 0.216168 LR 0.000500 Time 0.022173 -2022-12-06 11:08:59,189 - Epoch: [104][ 330/ 1200] Overall Loss 0.216379 Objective Loss 0.216379 LR 0.000500 Time 0.022104 -2022-12-06 11:08:59,387 - Epoch: [104][ 340/ 1200] Overall Loss 0.217000 Objective Loss 0.217000 LR 0.000500 Time 0.022034 -2022-12-06 11:08:59,586 - Epoch: [104][ 350/ 1200] Overall Loss 0.216885 Objective Loss 0.216885 LR 0.000500 Time 0.021972 -2022-12-06 11:08:59,783 - Epoch: [104][ 360/ 1200] Overall Loss 0.216467 Objective Loss 0.216467 LR 0.000500 Time 0.021908 -2022-12-06 11:08:59,982 - Epoch: [104][ 370/ 1200] Overall Loss 0.215883 Objective Loss 0.215883 LR 0.000500 Time 0.021852 -2022-12-06 11:09:00,179 - Epoch: [104][ 380/ 1200] Overall Loss 0.215802 Objective Loss 0.215802 LR 0.000500 Time 0.021795 -2022-12-06 11:09:00,380 - Epoch: [104][ 390/ 1200] Overall Loss 0.215625 Objective Loss 0.215625 LR 0.000500 Time 0.021748 -2022-12-06 11:09:00,578 - Epoch: [104][ 400/ 1200] Overall Loss 0.216087 Objective Loss 0.216087 LR 0.000500 Time 0.021698 -2022-12-06 11:09:00,777 - Epoch: [104][ 410/ 1200] Overall Loss 0.216824 Objective Loss 0.216824 LR 0.000500 Time 0.021654 -2022-12-06 11:09:00,975 - Epoch: [104][ 420/ 1200] Overall Loss 0.216879 Objective Loss 0.216879 LR 0.000500 Time 0.021608 -2022-12-06 11:09:01,175 - Epoch: [104][ 430/ 1200] Overall Loss 0.217318 Objective Loss 0.217318 LR 0.000500 Time 0.021569 -2022-12-06 11:09:01,373 - Epoch: [104][ 440/ 1200] Overall Loss 0.217072 Objective Loss 0.217072 LR 0.000500 Time 0.021529 -2022-12-06 11:09:01,573 - Epoch: [104][ 450/ 1200] Overall Loss 0.217609 Objective Loss 0.217609 LR 0.000500 Time 0.021492 -2022-12-06 11:09:01,771 - Epoch: [104][ 460/ 1200] Overall Loss 0.217944 Objective Loss 0.217944 LR 0.000500 Time 0.021454 -2022-12-06 11:09:01,971 - Epoch: [104][ 470/ 1200] Overall Loss 0.217629 Objective Loss 0.217629 LR 0.000500 Time 0.021421 -2022-12-06 11:09:02,170 - Epoch: [104][ 480/ 1200] Overall Loss 0.217166 Objective Loss 0.217166 LR 0.000500 Time 0.021389 -2022-12-06 11:09:02,368 - Epoch: [104][ 490/ 1200] Overall Loss 0.216972 Objective Loss 0.216972 LR 0.000500 Time 0.021356 -2022-12-06 11:09:02,566 - Epoch: [104][ 500/ 1200] Overall Loss 0.217138 Objective Loss 0.217138 LR 0.000500 Time 0.021323 -2022-12-06 11:09:02,766 - Epoch: [104][ 510/ 1200] Overall Loss 0.217398 Objective Loss 0.217398 LR 0.000500 Time 0.021296 -2022-12-06 11:09:02,964 - Epoch: [104][ 520/ 1200] Overall Loss 0.217285 Objective Loss 0.217285 LR 0.000500 Time 0.021266 -2022-12-06 11:09:03,163 - Epoch: [104][ 530/ 1200] Overall Loss 0.217180 Objective Loss 0.217180 LR 0.000500 Time 0.021239 -2022-12-06 11:09:03,360 - Epoch: [104][ 540/ 1200] Overall Loss 0.217192 Objective Loss 0.217192 LR 0.000500 Time 0.021210 -2022-12-06 11:09:03,560 - Epoch: [104][ 550/ 1200] Overall Loss 0.217268 Objective Loss 0.217268 LR 0.000500 Time 0.021187 -2022-12-06 11:09:03,756 - Epoch: [104][ 560/ 1200] Overall Loss 0.217520 Objective Loss 0.217520 LR 0.000500 Time 0.021159 -2022-12-06 11:09:03,956 - Epoch: [104][ 570/ 1200] Overall Loss 0.217886 Objective Loss 0.217886 LR 0.000500 Time 0.021137 -2022-12-06 11:09:04,152 - Epoch: [104][ 580/ 1200] Overall Loss 0.218043 Objective Loss 0.218043 LR 0.000500 Time 0.021110 -2022-12-06 11:09:04,352 - Epoch: [104][ 590/ 1200] Overall Loss 0.217658 Objective Loss 0.217658 LR 0.000500 Time 0.021090 -2022-12-06 11:09:04,550 - Epoch: [104][ 600/ 1200] Overall Loss 0.218005 Objective Loss 0.218005 LR 0.000500 Time 0.021068 -2022-12-06 11:09:04,751 - Epoch: [104][ 610/ 1200] Overall Loss 0.217777 Objective Loss 0.217777 LR 0.000500 Time 0.021050 -2022-12-06 11:09:04,948 - Epoch: [104][ 620/ 1200] Overall Loss 0.217467 Objective Loss 0.217467 LR 0.000500 Time 0.021028 -2022-12-06 11:09:05,148 - Epoch: [104][ 630/ 1200] Overall Loss 0.217575 Objective Loss 0.217575 LR 0.000500 Time 0.021010 -2022-12-06 11:09:05,346 - Epoch: [104][ 640/ 1200] Overall Loss 0.217344 Objective Loss 0.217344 LR 0.000500 Time 0.020990 -2022-12-06 11:09:05,546 - Epoch: [104][ 650/ 1200] Overall Loss 0.217237 Objective Loss 0.217237 LR 0.000500 Time 0.020975 -2022-12-06 11:09:05,744 - Epoch: [104][ 660/ 1200] Overall Loss 0.217333 Objective Loss 0.217333 LR 0.000500 Time 0.020955 -2022-12-06 11:09:05,944 - Epoch: [104][ 670/ 1200] Overall Loss 0.217690 Objective Loss 0.217690 LR 0.000500 Time 0.020940 -2022-12-06 11:09:06,141 - Epoch: [104][ 680/ 1200] Overall Loss 0.217469 Objective Loss 0.217469 LR 0.000500 Time 0.020922 -2022-12-06 11:09:06,340 - Epoch: [104][ 690/ 1200] Overall Loss 0.217502 Objective Loss 0.217502 LR 0.000500 Time 0.020906 -2022-12-06 11:09:06,536 - Epoch: [104][ 700/ 1200] Overall Loss 0.217898 Objective Loss 0.217898 LR 0.000500 Time 0.020887 -2022-12-06 11:09:06,736 - Epoch: [104][ 710/ 1200] Overall Loss 0.217877 Objective Loss 0.217877 LR 0.000500 Time 0.020873 -2022-12-06 11:09:06,934 - Epoch: [104][ 720/ 1200] Overall Loss 0.218161 Objective Loss 0.218161 LR 0.000500 Time 0.020858 -2022-12-06 11:09:07,134 - Epoch: [104][ 730/ 1200] Overall Loss 0.218452 Objective Loss 0.218452 LR 0.000500 Time 0.020846 -2022-12-06 11:09:07,332 - Epoch: [104][ 740/ 1200] Overall Loss 0.217854 Objective Loss 0.217854 LR 0.000500 Time 0.020829 -2022-12-06 11:09:07,532 - Epoch: [104][ 750/ 1200] Overall Loss 0.217596 Objective Loss 0.217596 LR 0.000500 Time 0.020818 -2022-12-06 11:09:07,728 - Epoch: [104][ 760/ 1200] Overall Loss 0.217634 Objective Loss 0.217634 LR 0.000500 Time 0.020802 -2022-12-06 11:09:07,928 - Epoch: [104][ 770/ 1200] Overall Loss 0.217549 Objective Loss 0.217549 LR 0.000500 Time 0.020790 -2022-12-06 11:09:08,125 - Epoch: [104][ 780/ 1200] Overall Loss 0.217562 Objective Loss 0.217562 LR 0.000500 Time 0.020776 -2022-12-06 11:09:08,324 - Epoch: [104][ 790/ 1200] Overall Loss 0.217579 Objective Loss 0.217579 LR 0.000500 Time 0.020764 -2022-12-06 11:09:08,522 - Epoch: [104][ 800/ 1200] Overall Loss 0.217845 Objective Loss 0.217845 LR 0.000500 Time 0.020751 -2022-12-06 11:09:08,721 - Epoch: [104][ 810/ 1200] Overall Loss 0.218134 Objective Loss 0.218134 LR 0.000500 Time 0.020740 -2022-12-06 11:09:08,918 - Epoch: [104][ 820/ 1200] Overall Loss 0.217925 Objective Loss 0.217925 LR 0.000500 Time 0.020726 -2022-12-06 11:09:09,117 - Epoch: [104][ 830/ 1200] Overall Loss 0.217798 Objective Loss 0.217798 LR 0.000500 Time 0.020716 -2022-12-06 11:09:09,314 - Epoch: [104][ 840/ 1200] Overall Loss 0.217402 Objective Loss 0.217402 LR 0.000500 Time 0.020703 -2022-12-06 11:09:09,514 - Epoch: [104][ 850/ 1200] Overall Loss 0.217450 Objective Loss 0.217450 LR 0.000500 Time 0.020694 -2022-12-06 11:09:09,711 - Epoch: [104][ 860/ 1200] Overall Loss 0.217552 Objective Loss 0.217552 LR 0.000500 Time 0.020681 -2022-12-06 11:09:09,909 - Epoch: [104][ 870/ 1200] Overall Loss 0.217505 Objective Loss 0.217505 LR 0.000500 Time 0.020671 -2022-12-06 11:09:10,107 - Epoch: [104][ 880/ 1200] Overall Loss 0.217537 Objective Loss 0.217537 LR 0.000500 Time 0.020661 -2022-12-06 11:09:10,306 - Epoch: [104][ 890/ 1200] Overall Loss 0.217690 Objective Loss 0.217690 LR 0.000500 Time 0.020652 -2022-12-06 11:09:10,503 - Epoch: [104][ 900/ 1200] Overall Loss 0.217576 Objective Loss 0.217576 LR 0.000500 Time 0.020641 -2022-12-06 11:09:10,703 - Epoch: [104][ 910/ 1200] Overall Loss 0.217687 Objective Loss 0.217687 LR 0.000500 Time 0.020632 -2022-12-06 11:09:10,900 - Epoch: [104][ 920/ 1200] Overall Loss 0.217631 Objective Loss 0.217631 LR 0.000500 Time 0.020621 -2022-12-06 11:09:11,098 - Epoch: [104][ 930/ 1200] Overall Loss 0.217711 Objective Loss 0.217711 LR 0.000500 Time 0.020612 -2022-12-06 11:09:11,295 - Epoch: [104][ 940/ 1200] Overall Loss 0.217631 Objective Loss 0.217631 LR 0.000500 Time 0.020602 -2022-12-06 11:09:11,494 - Epoch: [104][ 950/ 1200] Overall Loss 0.217773 Objective Loss 0.217773 LR 0.000500 Time 0.020594 -2022-12-06 11:09:11,691 - Epoch: [104][ 960/ 1200] Overall Loss 0.217644 Objective Loss 0.217644 LR 0.000500 Time 0.020584 -2022-12-06 11:09:11,890 - Epoch: [104][ 970/ 1200] Overall Loss 0.217775 Objective Loss 0.217775 LR 0.000500 Time 0.020576 -2022-12-06 11:09:12,087 - Epoch: [104][ 980/ 1200] Overall Loss 0.218127 Objective Loss 0.218127 LR 0.000500 Time 0.020567 -2022-12-06 11:09:12,286 - Epoch: [104][ 990/ 1200] Overall Loss 0.218274 Objective Loss 0.218274 LR 0.000500 Time 0.020560 -2022-12-06 11:09:12,483 - Epoch: [104][ 1000/ 1200] Overall Loss 0.218056 Objective Loss 0.218056 LR 0.000500 Time 0.020551 -2022-12-06 11:09:12,683 - Epoch: [104][ 1010/ 1200] Overall Loss 0.218083 Objective Loss 0.218083 LR 0.000500 Time 0.020545 -2022-12-06 11:09:12,880 - Epoch: [104][ 1020/ 1200] Overall Loss 0.217979 Objective Loss 0.217979 LR 0.000500 Time 0.020536 -2022-12-06 11:09:13,079 - Epoch: [104][ 1030/ 1200] Overall Loss 0.218130 Objective Loss 0.218130 LR 0.000500 Time 0.020529 -2022-12-06 11:09:13,276 - Epoch: [104][ 1040/ 1200] Overall Loss 0.218082 Objective Loss 0.218082 LR 0.000500 Time 0.020520 -2022-12-06 11:09:13,475 - Epoch: [104][ 1050/ 1200] Overall Loss 0.218170 Objective Loss 0.218170 LR 0.000500 Time 0.020514 -2022-12-06 11:09:13,672 - Epoch: [104][ 1060/ 1200] Overall Loss 0.218209 Objective Loss 0.218209 LR 0.000500 Time 0.020506 -2022-12-06 11:09:13,872 - Epoch: [104][ 1070/ 1200] Overall Loss 0.218375 Objective Loss 0.218375 LR 0.000500 Time 0.020500 -2022-12-06 11:09:14,069 - Epoch: [104][ 1080/ 1200] Overall Loss 0.218445 Objective Loss 0.218445 LR 0.000500 Time 0.020493 -2022-12-06 11:09:14,269 - Epoch: [104][ 1090/ 1200] Overall Loss 0.218400 Objective Loss 0.218400 LR 0.000500 Time 0.020487 -2022-12-06 11:09:14,467 - Epoch: [104][ 1100/ 1200] Overall Loss 0.218636 Objective Loss 0.218636 LR 0.000500 Time 0.020481 -2022-12-06 11:09:14,666 - Epoch: [104][ 1110/ 1200] Overall Loss 0.218821 Objective Loss 0.218821 LR 0.000500 Time 0.020475 -2022-12-06 11:09:14,863 - Epoch: [104][ 1120/ 1200] Overall Loss 0.218699 Objective Loss 0.218699 LR 0.000500 Time 0.020467 -2022-12-06 11:09:15,062 - Epoch: [104][ 1130/ 1200] Overall Loss 0.218924 Objective Loss 0.218924 LR 0.000500 Time 0.020462 -2022-12-06 11:09:15,259 - Epoch: [104][ 1140/ 1200] Overall Loss 0.218693 Objective Loss 0.218693 LR 0.000500 Time 0.020454 -2022-12-06 11:09:15,458 - Epoch: [104][ 1150/ 1200] Overall Loss 0.218581 Objective Loss 0.218581 LR 0.000500 Time 0.020449 -2022-12-06 11:09:15,655 - Epoch: [104][ 1160/ 1200] Overall Loss 0.218394 Objective Loss 0.218394 LR 0.000500 Time 0.020442 -2022-12-06 11:09:15,854 - Epoch: [104][ 1170/ 1200] Overall Loss 0.218443 Objective Loss 0.218443 LR 0.000500 Time 0.020437 -2022-12-06 11:09:16,050 - Epoch: [104][ 1180/ 1200] Overall Loss 0.218517 Objective Loss 0.218517 LR 0.000500 Time 0.020430 -2022-12-06 11:09:16,250 - Epoch: [104][ 1190/ 1200] Overall Loss 0.218458 Objective Loss 0.218458 LR 0.000500 Time 0.020425 -2022-12-06 11:09:16,476 - Epoch: [104][ 1200/ 1200] Overall Loss 0.218512 Objective Loss 0.218512 Top1 87.866109 Top5 98.535565 LR 0.000500 Time 0.020443 -2022-12-06 11:09:16,570 - --- validate (epoch=104)----------- -2022-12-06 11:09:16,570 - 34129 samples (256 per mini-batch) -2022-12-06 11:09:17,018 - Epoch: [104][ 10/ 134] Loss 0.254587 Top1 86.367188 Top5 97.968750 -2022-12-06 11:09:17,148 - Epoch: [104][ 20/ 134] Loss 0.268246 Top1 85.839844 Top5 98.105469 -2022-12-06 11:09:17,284 - Epoch: [104][ 30/ 134] Loss 0.276013 Top1 85.690104 Top5 98.072917 -2022-12-06 11:09:17,420 - Epoch: [104][ 40/ 134] Loss 0.269654 Top1 85.976562 Top5 98.115234 -2022-12-06 11:09:17,551 - Epoch: [104][ 50/ 134] Loss 0.273830 Top1 86.000000 Top5 98.187500 -2022-12-06 11:09:17,682 - Epoch: [104][ 60/ 134] Loss 0.271350 Top1 85.996094 Top5 98.216146 -2022-12-06 11:09:17,808 - Epoch: [104][ 70/ 134] Loss 0.271870 Top1 86.004464 Top5 98.236607 -2022-12-06 11:09:17,936 - Epoch: [104][ 80/ 134] Loss 0.272630 Top1 86.049805 Top5 98.173828 -2022-12-06 11:09:18,064 - Epoch: [104][ 90/ 134] Loss 0.271398 Top1 86.019965 Top5 98.172743 -2022-12-06 11:09:18,194 - Epoch: [104][ 100/ 134] Loss 0.270015 Top1 86.097656 Top5 98.187500 -2022-12-06 11:09:18,321 - Epoch: [104][ 110/ 134] Loss 0.269216 Top1 86.072443 Top5 98.210227 -2022-12-06 11:09:18,452 - Epoch: [104][ 120/ 134] Loss 0.267877 Top1 85.999349 Top5 98.206380 -2022-12-06 11:09:18,581 - Epoch: [104][ 130/ 134] Loss 0.269426 Top1 85.961538 Top5 98.194111 -2022-12-06 11:09:18,619 - Epoch: [104][ 134/ 134] Loss 0.270570 Top1 85.929854 Top5 98.189223 -2022-12-06 11:09:18,707 - ==> Top1: 85.930 Top5: 98.189 Loss: 0.271 - -2022-12-06 11:09:18,707 - ==> Confusion: -[[ 918 1 2 3 6 4 1 0 3 39 0 2 2 1 6 2 2 0 1 0 3] - [ 4 927 0 2 11 23 3 13 0 1 5 6 0 1 2 0 7 2 8 4 8] - [ 3 1 1004 10 3 0 27 11 1 3 8 4 0 4 1 6 1 3 3 1 9] - [ 5 0 19 934 0 4 0 0 0 2 17 1 5 2 5 1 2 2 13 0 8] - [ 10 4 1 1 955 2 1 1 1 7 3 2 0 3 7 10 7 1 0 1 3] - [ 6 19 0 3 7 956 1 23 3 1 1 15 3 15 0 2 3 1 1 5 4] - [ 1 2 4 1 0 0 1083 4 0 0 2 1 0 3 0 5 0 2 2 4 4] - [ 2 9 9 3 2 25 12 933 0 1 3 6 1 0 0 2 0 0 22 17 7] - [ 8 3 0 2 1 1 1 0 971 39 12 1 0 11 6 1 1 2 2 1 1] - [ 87 0 1 1 8 2 0 2 22 854 1 2 0 9 2 1 2 0 1 0 6] - [ 0 0 3 2 1 1 2 4 6 1 972 2 1 9 5 0 3 0 2 1 4] - [ 4 2 3 2 2 8 6 3 2 0 1 948 28 1 0 11 6 6 0 15 3] - [ 2 1 4 4 1 1 2 1 0 1 0 33 878 0 1 11 3 19 0 3 4] - [ 0 1 2 1 0 6 0 1 7 18 11 6 2 939 1 4 1 2 1 5 15] - [ 11 2 1 16 8 3 1 1 18 2 7 2 3 3 1031 1 5 2 8 0 5] - [ 0 0 2 1 1 1 1 0 1 1 0 7 2 2 0 1000 6 10 1 4 3] - [ 4 1 0 1 2 0 0 1 0 0 0 1 1 2 1 16 1027 1 0 7 7] - [ 2 0 1 1 0 1 0 1 1 2 0 6 15 3 1 22 1 975 0 1 3] - [ 5 3 2 10 1 1 0 19 1 0 7 3 2 0 6 1 1 2 938 1 5] - [ 3 0 1 2 1 3 9 4 0 0 1 17 5 6 0 4 4 1 0 1012 7] - [ 158 200 157 108 132 147 105 147 75 80 221 100 319 254 118 158 220 87 140 237 10063]] - -2022-12-06 11:09:19,377 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:09:19,377 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:09:19,383 - - -2022-12-06 11:09:19,383 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:09:20,314 - Epoch: [105][ 10/ 1200] Overall Loss 0.220150 Objective Loss 0.220150 LR 0.000500 Time 0.093068 -2022-12-06 11:09:20,513 - Epoch: [105][ 20/ 1200] Overall Loss 0.220563 Objective Loss 0.220563 LR 0.000500 Time 0.056416 -2022-12-06 11:09:20,706 - Epoch: [105][ 30/ 1200] Overall Loss 0.210388 Objective Loss 0.210388 LR 0.000500 Time 0.044018 -2022-12-06 11:09:20,898 - Epoch: [105][ 40/ 1200] Overall Loss 0.214708 Objective Loss 0.214708 LR 0.000500 Time 0.037811 -2022-12-06 11:09:21,089 - Epoch: [105][ 50/ 1200] Overall Loss 0.210707 Objective Loss 0.210707 LR 0.000500 Time 0.034062 -2022-12-06 11:09:21,282 - Epoch: [105][ 60/ 1200] Overall Loss 0.216676 Objective Loss 0.216676 LR 0.000500 Time 0.031578 -2022-12-06 11:09:21,472 - Epoch: [105][ 70/ 1200] Overall Loss 0.215481 Objective Loss 0.215481 LR 0.000500 Time 0.029787 -2022-12-06 11:09:21,665 - Epoch: [105][ 80/ 1200] Overall Loss 0.213088 Objective Loss 0.213088 LR 0.000500 Time 0.028464 -2022-12-06 11:09:21,857 - Epoch: [105][ 90/ 1200] Overall Loss 0.213391 Objective Loss 0.213391 LR 0.000500 Time 0.027430 -2022-12-06 11:09:22,050 - Epoch: [105][ 100/ 1200] Overall Loss 0.211595 Objective Loss 0.211595 LR 0.000500 Time 0.026605 -2022-12-06 11:09:22,242 - Epoch: [105][ 110/ 1200] Overall Loss 0.211179 Objective Loss 0.211179 LR 0.000500 Time 0.025928 -2022-12-06 11:09:22,434 - Epoch: [105][ 120/ 1200] Overall Loss 0.210957 Objective Loss 0.210957 LR 0.000500 Time 0.025367 -2022-12-06 11:09:22,626 - Epoch: [105][ 130/ 1200] Overall Loss 0.211983 Objective Loss 0.211983 LR 0.000500 Time 0.024888 -2022-12-06 11:09:22,818 - Epoch: [105][ 140/ 1200] Overall Loss 0.214442 Objective Loss 0.214442 LR 0.000500 Time 0.024479 -2022-12-06 11:09:23,011 - Epoch: [105][ 150/ 1200] Overall Loss 0.212549 Objective Loss 0.212549 LR 0.000500 Time 0.024124 -2022-12-06 11:09:23,203 - Epoch: [105][ 160/ 1200] Overall Loss 0.212995 Objective Loss 0.212995 LR 0.000500 Time 0.023813 -2022-12-06 11:09:23,395 - Epoch: [105][ 170/ 1200] Overall Loss 0.214076 Objective Loss 0.214076 LR 0.000500 Time 0.023538 -2022-12-06 11:09:23,587 - Epoch: [105][ 180/ 1200] Overall Loss 0.214463 Objective Loss 0.214463 LR 0.000500 Time 0.023296 -2022-12-06 11:09:23,779 - Epoch: [105][ 190/ 1200] Overall Loss 0.213842 Objective Loss 0.213842 LR 0.000500 Time 0.023078 -2022-12-06 11:09:23,971 - Epoch: [105][ 200/ 1200] Overall Loss 0.213001 Objective Loss 0.213001 LR 0.000500 Time 0.022882 -2022-12-06 11:09:24,164 - Epoch: [105][ 210/ 1200] Overall Loss 0.211716 Objective Loss 0.211716 LR 0.000500 Time 0.022706 -2022-12-06 11:09:24,356 - Epoch: [105][ 220/ 1200] Overall Loss 0.212256 Objective Loss 0.212256 LR 0.000500 Time 0.022546 -2022-12-06 11:09:24,548 - Epoch: [105][ 230/ 1200] Overall Loss 0.212892 Objective Loss 0.212892 LR 0.000500 Time 0.022399 -2022-12-06 11:09:24,740 - Epoch: [105][ 240/ 1200] Overall Loss 0.213716 Objective Loss 0.213716 LR 0.000500 Time 0.022264 -2022-12-06 11:09:24,933 - Epoch: [105][ 250/ 1200] Overall Loss 0.213628 Objective Loss 0.213628 LR 0.000500 Time 0.022139 -2022-12-06 11:09:25,124 - Epoch: [105][ 260/ 1200] Overall Loss 0.212921 Objective Loss 0.212921 LR 0.000500 Time 0.022023 -2022-12-06 11:09:25,316 - Epoch: [105][ 270/ 1200] Overall Loss 0.213354 Objective Loss 0.213354 LR 0.000500 Time 0.021916 -2022-12-06 11:09:25,508 - Epoch: [105][ 280/ 1200] Overall Loss 0.212835 Objective Loss 0.212835 LR 0.000500 Time 0.021818 -2022-12-06 11:09:25,700 - Epoch: [105][ 290/ 1200] Overall Loss 0.213204 Objective Loss 0.213204 LR 0.000500 Time 0.021726 -2022-12-06 11:09:25,892 - Epoch: [105][ 300/ 1200] Overall Loss 0.213159 Objective Loss 0.213159 LR 0.000500 Time 0.021639 -2022-12-06 11:09:26,084 - Epoch: [105][ 310/ 1200] Overall Loss 0.212595 Objective Loss 0.212595 LR 0.000500 Time 0.021559 -2022-12-06 11:09:26,276 - Epoch: [105][ 320/ 1200] Overall Loss 0.212254 Objective Loss 0.212254 LR 0.000500 Time 0.021483 -2022-12-06 11:09:26,468 - Epoch: [105][ 330/ 1200] Overall Loss 0.211720 Objective Loss 0.211720 LR 0.000500 Time 0.021412 -2022-12-06 11:09:26,660 - Epoch: [105][ 340/ 1200] Overall Loss 0.211507 Objective Loss 0.211507 LR 0.000500 Time 0.021345 -2022-12-06 11:09:26,852 - Epoch: [105][ 350/ 1200] Overall Loss 0.211593 Objective Loss 0.211593 LR 0.000500 Time 0.021283 -2022-12-06 11:09:27,045 - Epoch: [105][ 360/ 1200] Overall Loss 0.211542 Objective Loss 0.211542 LR 0.000500 Time 0.021225 -2022-12-06 11:09:27,237 - Epoch: [105][ 370/ 1200] Overall Loss 0.211960 Objective Loss 0.211960 LR 0.000500 Time 0.021168 -2022-12-06 11:09:27,429 - Epoch: [105][ 380/ 1200] Overall Loss 0.211953 Objective Loss 0.211953 LR 0.000500 Time 0.021116 -2022-12-06 11:09:27,621 - Epoch: [105][ 390/ 1200] Overall Loss 0.212182 Objective Loss 0.212182 LR 0.000500 Time 0.021066 -2022-12-06 11:09:27,813 - Epoch: [105][ 400/ 1200] Overall Loss 0.212639 Objective Loss 0.212639 LR 0.000500 Time 0.021018 -2022-12-06 11:09:28,006 - Epoch: [105][ 410/ 1200] Overall Loss 0.212780 Objective Loss 0.212780 LR 0.000500 Time 0.020974 -2022-12-06 11:09:28,199 - Epoch: [105][ 420/ 1200] Overall Loss 0.213116 Objective Loss 0.213116 LR 0.000500 Time 0.020932 -2022-12-06 11:09:28,391 - Epoch: [105][ 430/ 1200] Overall Loss 0.213344 Objective Loss 0.213344 LR 0.000500 Time 0.020891 -2022-12-06 11:09:28,583 - Epoch: [105][ 440/ 1200] Overall Loss 0.213281 Objective Loss 0.213281 LR 0.000500 Time 0.020851 -2022-12-06 11:09:28,775 - Epoch: [105][ 450/ 1200] Overall Loss 0.213301 Objective Loss 0.213301 LR 0.000500 Time 0.020813 -2022-12-06 11:09:28,967 - Epoch: [105][ 460/ 1200] Overall Loss 0.213480 Objective Loss 0.213480 LR 0.000500 Time 0.020776 -2022-12-06 11:09:29,159 - Epoch: [105][ 470/ 1200] Overall Loss 0.213647 Objective Loss 0.213647 LR 0.000500 Time 0.020742 -2022-12-06 11:09:29,351 - Epoch: [105][ 480/ 1200] Overall Loss 0.213484 Objective Loss 0.213484 LR 0.000500 Time 0.020709 -2022-12-06 11:09:29,543 - Epoch: [105][ 490/ 1200] Overall Loss 0.213379 Objective Loss 0.213379 LR 0.000500 Time 0.020678 -2022-12-06 11:09:29,736 - Epoch: [105][ 500/ 1200] Overall Loss 0.213357 Objective Loss 0.213357 LR 0.000500 Time 0.020648 -2022-12-06 11:09:29,928 - Epoch: [105][ 510/ 1200] Overall Loss 0.213554 Objective Loss 0.213554 LR 0.000500 Time 0.020619 -2022-12-06 11:09:30,120 - Epoch: [105][ 520/ 1200] Overall Loss 0.213637 Objective Loss 0.213637 LR 0.000500 Time 0.020591 -2022-12-06 11:09:30,312 - Epoch: [105][ 530/ 1200] Overall Loss 0.213946 Objective Loss 0.213946 LR 0.000500 Time 0.020564 -2022-12-06 11:09:30,505 - Epoch: [105][ 540/ 1200] Overall Loss 0.213806 Objective Loss 0.213806 LR 0.000500 Time 0.020538 -2022-12-06 11:09:30,697 - Epoch: [105][ 550/ 1200] Overall Loss 0.213733 Objective Loss 0.213733 LR 0.000500 Time 0.020513 -2022-12-06 11:09:30,889 - Epoch: [105][ 560/ 1200] Overall Loss 0.213880 Objective Loss 0.213880 LR 0.000500 Time 0.020489 -2022-12-06 11:09:31,080 - Epoch: [105][ 570/ 1200] Overall Loss 0.213761 Objective Loss 0.213761 LR 0.000500 Time 0.020464 -2022-12-06 11:09:31,272 - Epoch: [105][ 580/ 1200] Overall Loss 0.213907 Objective Loss 0.213907 LR 0.000500 Time 0.020441 -2022-12-06 11:09:31,464 - Epoch: [105][ 590/ 1200] Overall Loss 0.214287 Objective Loss 0.214287 LR 0.000500 Time 0.020419 -2022-12-06 11:09:31,656 - Epoch: [105][ 600/ 1200] Overall Loss 0.214296 Objective Loss 0.214296 LR 0.000500 Time 0.020398 -2022-12-06 11:09:31,848 - Epoch: [105][ 610/ 1200] Overall Loss 0.215116 Objective Loss 0.215116 LR 0.000500 Time 0.020377 -2022-12-06 11:09:32,040 - Epoch: [105][ 620/ 1200] Overall Loss 0.215150 Objective Loss 0.215150 LR 0.000500 Time 0.020356 -2022-12-06 11:09:32,232 - Epoch: [105][ 630/ 1200] Overall Loss 0.215160 Objective Loss 0.215160 LR 0.000500 Time 0.020337 -2022-12-06 11:09:32,423 - Epoch: [105][ 640/ 1200] Overall Loss 0.215047 Objective Loss 0.215047 LR 0.000500 Time 0.020318 -2022-12-06 11:09:32,615 - Epoch: [105][ 650/ 1200] Overall Loss 0.215014 Objective Loss 0.215014 LR 0.000500 Time 0.020299 -2022-12-06 11:09:32,806 - Epoch: [105][ 660/ 1200] Overall Loss 0.215461 Objective Loss 0.215461 LR 0.000500 Time 0.020280 -2022-12-06 11:09:32,998 - Epoch: [105][ 670/ 1200] Overall Loss 0.215091 Objective Loss 0.215091 LR 0.000500 Time 0.020263 -2022-12-06 11:09:33,190 - Epoch: [105][ 680/ 1200] Overall Loss 0.215413 Objective Loss 0.215413 LR 0.000500 Time 0.020247 -2022-12-06 11:09:33,382 - Epoch: [105][ 690/ 1200] Overall Loss 0.215215 Objective Loss 0.215215 LR 0.000500 Time 0.020231 -2022-12-06 11:09:33,574 - Epoch: [105][ 700/ 1200] Overall Loss 0.215614 Objective Loss 0.215614 LR 0.000500 Time 0.020216 -2022-12-06 11:09:33,766 - Epoch: [105][ 710/ 1200] Overall Loss 0.216126 Objective Loss 0.216126 LR 0.000500 Time 0.020201 -2022-12-06 11:09:33,958 - Epoch: [105][ 720/ 1200] Overall Loss 0.216201 Objective Loss 0.216201 LR 0.000500 Time 0.020186 -2022-12-06 11:09:34,150 - Epoch: [105][ 730/ 1200] Overall Loss 0.216122 Objective Loss 0.216122 LR 0.000500 Time 0.020171 -2022-12-06 11:09:34,342 - Epoch: [105][ 740/ 1200] Overall Loss 0.216118 Objective Loss 0.216118 LR 0.000500 Time 0.020157 -2022-12-06 11:09:34,534 - Epoch: [105][ 750/ 1200] Overall Loss 0.215982 Objective Loss 0.215982 LR 0.000500 Time 0.020144 -2022-12-06 11:09:34,726 - Epoch: [105][ 760/ 1200] Overall Loss 0.215812 Objective Loss 0.215812 LR 0.000500 Time 0.020131 -2022-12-06 11:09:34,918 - Epoch: [105][ 770/ 1200] Overall Loss 0.215896 Objective Loss 0.215896 LR 0.000500 Time 0.020118 -2022-12-06 11:09:35,110 - Epoch: [105][ 780/ 1200] Overall Loss 0.215690 Objective Loss 0.215690 LR 0.000500 Time 0.020105 -2022-12-06 11:09:35,302 - Epoch: [105][ 790/ 1200] Overall Loss 0.215675 Objective Loss 0.215675 LR 0.000500 Time 0.020093 -2022-12-06 11:09:35,494 - Epoch: [105][ 800/ 1200] Overall Loss 0.215283 Objective Loss 0.215283 LR 0.000500 Time 0.020081 -2022-12-06 11:09:35,686 - Epoch: [105][ 810/ 1200] Overall Loss 0.215380 Objective Loss 0.215380 LR 0.000500 Time 0.020069 -2022-12-06 11:09:35,878 - Epoch: [105][ 820/ 1200] Overall Loss 0.215374 Objective Loss 0.215374 LR 0.000500 Time 0.020058 -2022-12-06 11:09:36,069 - Epoch: [105][ 830/ 1200] Overall Loss 0.215412 Objective Loss 0.215412 LR 0.000500 Time 0.020046 -2022-12-06 11:09:36,261 - Epoch: [105][ 840/ 1200] Overall Loss 0.215358 Objective Loss 0.215358 LR 0.000500 Time 0.020036 -2022-12-06 11:09:36,453 - Epoch: [105][ 850/ 1200] Overall Loss 0.214948 Objective Loss 0.214948 LR 0.000500 Time 0.020025 -2022-12-06 11:09:36,646 - Epoch: [105][ 860/ 1200] Overall Loss 0.214935 Objective Loss 0.214935 LR 0.000500 Time 0.020016 -2022-12-06 11:09:36,838 - Epoch: [105][ 870/ 1200] Overall Loss 0.214854 Objective Loss 0.214854 LR 0.000500 Time 0.020006 -2022-12-06 11:09:37,030 - Epoch: [105][ 880/ 1200] Overall Loss 0.214838 Objective Loss 0.214838 LR 0.000500 Time 0.019996 -2022-12-06 11:09:37,223 - Epoch: [105][ 890/ 1200] Overall Loss 0.215099 Objective Loss 0.215099 LR 0.000500 Time 0.019987 -2022-12-06 11:09:37,414 - Epoch: [105][ 900/ 1200] Overall Loss 0.215204 Objective Loss 0.215204 LR 0.000500 Time 0.019977 -2022-12-06 11:09:37,606 - Epoch: [105][ 910/ 1200] Overall Loss 0.214755 Objective Loss 0.214755 LR 0.000500 Time 0.019968 -2022-12-06 11:09:37,798 - Epoch: [105][ 920/ 1200] Overall Loss 0.214587 Objective Loss 0.214587 LR 0.000500 Time 0.019959 -2022-12-06 11:09:37,989 - Epoch: [105][ 930/ 1200] Overall Loss 0.214562 Objective Loss 0.214562 LR 0.000500 Time 0.019949 -2022-12-06 11:09:38,182 - Epoch: [105][ 940/ 1200] Overall Loss 0.214356 Objective Loss 0.214356 LR 0.000500 Time 0.019941 -2022-12-06 11:09:38,374 - Epoch: [105][ 950/ 1200] Overall Loss 0.214501 Objective Loss 0.214501 LR 0.000500 Time 0.019933 -2022-12-06 11:09:38,566 - Epoch: [105][ 960/ 1200] Overall Loss 0.214367 Objective Loss 0.214367 LR 0.000500 Time 0.019925 -2022-12-06 11:09:38,758 - Epoch: [105][ 970/ 1200] Overall Loss 0.214492 Objective Loss 0.214492 LR 0.000500 Time 0.019917 -2022-12-06 11:09:38,950 - Epoch: [105][ 980/ 1200] Overall Loss 0.214711 Objective Loss 0.214711 LR 0.000500 Time 0.019909 -2022-12-06 11:09:39,142 - Epoch: [105][ 990/ 1200] Overall Loss 0.214842 Objective Loss 0.214842 LR 0.000500 Time 0.019901 -2022-12-06 11:09:39,334 - Epoch: [105][ 1000/ 1200] Overall Loss 0.214883 Objective Loss 0.214883 LR 0.000500 Time 0.019893 -2022-12-06 11:09:39,526 - Epoch: [105][ 1010/ 1200] Overall Loss 0.215055 Objective Loss 0.215055 LR 0.000500 Time 0.019886 -2022-12-06 11:09:39,718 - Epoch: [105][ 1020/ 1200] Overall Loss 0.215117 Objective Loss 0.215117 LR 0.000500 Time 0.019879 -2022-12-06 11:09:39,910 - Epoch: [105][ 1030/ 1200] Overall Loss 0.215143 Objective Loss 0.215143 LR 0.000500 Time 0.019871 -2022-12-06 11:09:40,103 - Epoch: [105][ 1040/ 1200] Overall Loss 0.215231 Objective Loss 0.215231 LR 0.000500 Time 0.019865 -2022-12-06 11:09:40,295 - Epoch: [105][ 1050/ 1200] Overall Loss 0.215072 Objective Loss 0.215072 LR 0.000500 Time 0.019859 -2022-12-06 11:09:40,487 - Epoch: [105][ 1060/ 1200] Overall Loss 0.214991 Objective Loss 0.214991 LR 0.000500 Time 0.019852 -2022-12-06 11:09:40,680 - Epoch: [105][ 1070/ 1200] Overall Loss 0.215311 Objective Loss 0.215311 LR 0.000500 Time 0.019846 -2022-12-06 11:09:40,872 - Epoch: [105][ 1080/ 1200] Overall Loss 0.215347 Objective Loss 0.215347 LR 0.000500 Time 0.019840 -2022-12-06 11:09:41,064 - Epoch: [105][ 1090/ 1200] Overall Loss 0.215434 Objective Loss 0.215434 LR 0.000500 Time 0.019833 -2022-12-06 11:09:41,255 - Epoch: [105][ 1100/ 1200] Overall Loss 0.215539 Objective Loss 0.215539 LR 0.000500 Time 0.019826 -2022-12-06 11:09:41,448 - Epoch: [105][ 1110/ 1200] Overall Loss 0.215757 Objective Loss 0.215757 LR 0.000500 Time 0.019820 -2022-12-06 11:09:41,639 - Epoch: [105][ 1120/ 1200] Overall Loss 0.215809 Objective Loss 0.215809 LR 0.000500 Time 0.019814 -2022-12-06 11:09:41,831 - Epoch: [105][ 1130/ 1200] Overall Loss 0.215863 Objective Loss 0.215863 LR 0.000500 Time 0.019808 -2022-12-06 11:09:42,023 - Epoch: [105][ 1140/ 1200] Overall Loss 0.215700 Objective Loss 0.215700 LR 0.000500 Time 0.019802 -2022-12-06 11:09:42,215 - Epoch: [105][ 1150/ 1200] Overall Loss 0.215844 Objective Loss 0.215844 LR 0.000500 Time 0.019796 -2022-12-06 11:09:42,406 - Epoch: [105][ 1160/ 1200] Overall Loss 0.215942 Objective Loss 0.215942 LR 0.000500 Time 0.019790 -2022-12-06 11:09:42,598 - Epoch: [105][ 1170/ 1200] Overall Loss 0.215835 Objective Loss 0.215835 LR 0.000500 Time 0.019785 -2022-12-06 11:09:42,790 - Epoch: [105][ 1180/ 1200] Overall Loss 0.215964 Objective Loss 0.215964 LR 0.000500 Time 0.019779 -2022-12-06 11:09:42,982 - Epoch: [105][ 1190/ 1200] Overall Loss 0.215923 Objective Loss 0.215923 LR 0.000500 Time 0.019774 -2022-12-06 11:09:43,209 - Epoch: [105][ 1200/ 1200] Overall Loss 0.215904 Objective Loss 0.215904 Top1 86.401674 Top5 97.280335 LR 0.000500 Time 0.019798 -2022-12-06 11:09:43,297 - --- validate (epoch=105)----------- -2022-12-06 11:09:43,297 - 34129 samples (256 per mini-batch) -2022-12-06 11:09:43,737 - Epoch: [105][ 10/ 134] Loss 0.252087 Top1 85.664062 Top5 98.164062 -2022-12-06 11:09:43,869 - Epoch: [105][ 20/ 134] Loss 0.264685 Top1 85.605469 Top5 98.027344 -2022-12-06 11:09:44,000 - Epoch: [105][ 30/ 134] Loss 0.271342 Top1 85.429688 Top5 97.994792 -2022-12-06 11:09:44,130 - Epoch: [105][ 40/ 134] Loss 0.266529 Top1 85.410156 Top5 98.007812 -2022-12-06 11:09:44,265 - Epoch: [105][ 50/ 134] Loss 0.263303 Top1 85.492188 Top5 97.984375 -2022-12-06 11:09:44,410 - Epoch: [105][ 60/ 134] Loss 0.263244 Top1 85.442708 Top5 98.066406 -2022-12-06 11:09:44,551 - Epoch: [105][ 70/ 134] Loss 0.259250 Top1 85.463170 Top5 98.130580 -2022-12-06 11:09:44,696 - Epoch: [105][ 80/ 134] Loss 0.262707 Top1 85.429688 Top5 98.085938 -2022-12-06 11:09:44,837 - Epoch: [105][ 90/ 134] Loss 0.263444 Top1 85.390625 Top5 98.107639 -2022-12-06 11:09:44,981 - Epoch: [105][ 100/ 134] Loss 0.265098 Top1 85.382812 Top5 98.097656 -2022-12-06 11:09:45,122 - Epoch: [105][ 110/ 134] Loss 0.266228 Top1 85.404830 Top5 98.096591 -2022-12-06 11:09:45,267 - Epoch: [105][ 120/ 134] Loss 0.266105 Top1 85.494792 Top5 98.069661 -2022-12-06 11:09:45,403 - Epoch: [105][ 130/ 134] Loss 0.265699 Top1 85.465745 Top5 98.076923 -2022-12-06 11:09:45,439 - Epoch: [105][ 134/ 134] Loss 0.265196 Top1 85.461045 Top5 98.063231 -2022-12-06 11:09:45,529 - ==> Top1: 85.461 Top5: 98.063 Loss: 0.265 - -2022-12-06 11:09:45,530 - ==> Confusion: -[[ 912 0 2 1 3 6 0 0 3 53 0 1 3 3 6 1 0 0 1 0 1] - [ 1 919 1 1 6 34 4 11 3 1 6 6 1 1 0 1 6 0 13 1 11] - [ 5 2 995 8 4 0 29 11 0 3 4 4 2 3 3 4 1 2 9 3 11] - [ 0 3 16 934 2 0 1 0 1 0 6 0 3 2 14 2 0 6 20 1 9] - [ 9 5 1 0 955 4 1 1 2 6 1 4 0 3 8 8 7 1 1 1 2] - [ 1 18 0 2 2 981 3 13 5 2 0 10 6 13 2 1 1 0 0 7 2] - [ 0 1 7 2 0 3 1079 3 1 0 1 1 1 3 0 5 2 1 0 6 2] - [ 1 9 5 2 2 24 9 949 0 0 2 9 0 1 1 1 0 0 22 13 4] - [ 6 1 0 0 0 2 0 0 991 36 5 0 0 4 12 2 2 1 1 1 0] - [ 62 0 0 0 2 0 0 2 27 886 2 2 0 9 2 1 0 2 0 0 4] - [ 2 1 3 6 0 1 1 3 9 2 952 4 2 13 3 0 1 0 9 2 5] - [ 3 0 1 0 1 9 5 6 1 0 0 958 27 4 0 9 4 11 1 10 1] - [ 0 1 1 1 1 1 0 0 1 0 0 24 906 0 0 13 1 11 0 3 5] - [ 0 1 2 0 0 11 0 2 18 15 3 4 5 937 2 2 1 2 0 5 13] - [ 7 3 1 16 1 2 0 0 22 3 1 2 2 2 1050 1 2 1 10 0 4] - [ 1 0 0 1 2 0 2 0 1 1 1 7 10 2 0 993 7 10 0 3 2] - [ 1 5 1 2 2 1 0 0 0 1 0 1 3 3 1 12 1025 0 1 5 8] - [ 2 0 0 2 1 0 0 1 1 3 0 4 17 1 3 16 0 981 0 3 1] - [ 3 4 3 13 2 1 0 25 2 1 2 2 1 0 6 1 0 0 936 2 4] - [ 1 2 2 0 0 5 6 3 0 0 1 18 8 6 1 5 2 2 1 1012 5] - [ 174 187 161 98 86 228 73 147 130 110 158 101 392 306 158 154 172 95 219 261 9816]] - -2022-12-06 11:09:46,192 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:09:46,192 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:09:46,198 - - -2022-12-06 11:09:46,198 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:09:47,139 - Epoch: [106][ 10/ 1200] Overall Loss 0.221741 Objective Loss 0.221741 LR 0.000500 Time 0.094066 -2022-12-06 11:09:47,342 - Epoch: [106][ 20/ 1200] Overall Loss 0.219176 Objective Loss 0.219176 LR 0.000500 Time 0.057127 -2022-12-06 11:09:47,533 - Epoch: [106][ 30/ 1200] Overall Loss 0.214947 Objective Loss 0.214947 LR 0.000500 Time 0.044458 -2022-12-06 11:09:47,724 - Epoch: [106][ 40/ 1200] Overall Loss 0.211912 Objective Loss 0.211912 LR 0.000500 Time 0.038095 -2022-12-06 11:09:47,914 - Epoch: [106][ 50/ 1200] Overall Loss 0.208502 Objective Loss 0.208502 LR 0.000500 Time 0.034262 -2022-12-06 11:09:48,104 - Epoch: [106][ 60/ 1200] Overall Loss 0.207210 Objective Loss 0.207210 LR 0.000500 Time 0.031714 -2022-12-06 11:09:48,295 - Epoch: [106][ 70/ 1200] Overall Loss 0.210756 Objective Loss 0.210756 LR 0.000500 Time 0.029901 -2022-12-06 11:09:48,486 - Epoch: [106][ 80/ 1200] Overall Loss 0.212138 Objective Loss 0.212138 LR 0.000500 Time 0.028538 -2022-12-06 11:09:48,676 - Epoch: [106][ 90/ 1200] Overall Loss 0.210571 Objective Loss 0.210571 LR 0.000500 Time 0.027472 -2022-12-06 11:09:48,866 - Epoch: [106][ 100/ 1200] Overall Loss 0.212288 Objective Loss 0.212288 LR 0.000500 Time 0.026620 -2022-12-06 11:09:49,056 - Epoch: [106][ 110/ 1200] Overall Loss 0.209292 Objective Loss 0.209292 LR 0.000500 Time 0.025924 -2022-12-06 11:09:49,247 - Epoch: [106][ 120/ 1200] Overall Loss 0.210612 Objective Loss 0.210612 LR 0.000500 Time 0.025348 -2022-12-06 11:09:49,436 - Epoch: [106][ 130/ 1200] Overall Loss 0.210420 Objective Loss 0.210420 LR 0.000500 Time 0.024855 -2022-12-06 11:09:49,627 - Epoch: [106][ 140/ 1200] Overall Loss 0.210405 Objective Loss 0.210405 LR 0.000500 Time 0.024434 -2022-12-06 11:09:49,816 - Epoch: [106][ 150/ 1200] Overall Loss 0.210684 Objective Loss 0.210684 LR 0.000500 Time 0.024066 -2022-12-06 11:09:50,007 - Epoch: [106][ 160/ 1200] Overall Loss 0.209710 Objective Loss 0.209710 LR 0.000500 Time 0.023748 -2022-12-06 11:09:50,199 - Epoch: [106][ 170/ 1200] Overall Loss 0.210621 Objective Loss 0.210621 LR 0.000500 Time 0.023479 -2022-12-06 11:09:50,390 - Epoch: [106][ 180/ 1200] Overall Loss 0.211115 Objective Loss 0.211115 LR 0.000500 Time 0.023232 -2022-12-06 11:09:50,580 - Epoch: [106][ 190/ 1200] Overall Loss 0.210938 Objective Loss 0.210938 LR 0.000500 Time 0.023005 -2022-12-06 11:09:50,770 - Epoch: [106][ 200/ 1200] Overall Loss 0.210757 Objective Loss 0.210757 LR 0.000500 Time 0.022803 -2022-12-06 11:09:50,960 - Epoch: [106][ 210/ 1200] Overall Loss 0.211583 Objective Loss 0.211583 LR 0.000500 Time 0.022619 -2022-12-06 11:09:51,150 - Epoch: [106][ 220/ 1200] Overall Loss 0.210551 Objective Loss 0.210551 LR 0.000500 Time 0.022453 -2022-12-06 11:09:51,340 - Epoch: [106][ 230/ 1200] Overall Loss 0.210294 Objective Loss 0.210294 LR 0.000500 Time 0.022300 -2022-12-06 11:09:51,531 - Epoch: [106][ 240/ 1200] Overall Loss 0.210257 Objective Loss 0.210257 LR 0.000500 Time 0.022163 -2022-12-06 11:09:51,721 - Epoch: [106][ 250/ 1200] Overall Loss 0.210692 Objective Loss 0.210692 LR 0.000500 Time 0.022036 -2022-12-06 11:09:51,912 - Epoch: [106][ 260/ 1200] Overall Loss 0.211429 Objective Loss 0.211429 LR 0.000500 Time 0.021920 -2022-12-06 11:09:52,102 - Epoch: [106][ 270/ 1200] Overall Loss 0.211113 Objective Loss 0.211113 LR 0.000500 Time 0.021810 -2022-12-06 11:09:52,292 - Epoch: [106][ 280/ 1200] Overall Loss 0.211262 Objective Loss 0.211262 LR 0.000500 Time 0.021708 -2022-12-06 11:09:52,482 - Epoch: [106][ 290/ 1200] Overall Loss 0.211273 Objective Loss 0.211273 LR 0.000500 Time 0.021613 -2022-12-06 11:09:52,672 - Epoch: [106][ 300/ 1200] Overall Loss 0.211931 Objective Loss 0.211931 LR 0.000500 Time 0.021524 -2022-12-06 11:09:52,862 - Epoch: [106][ 310/ 1200] Overall Loss 0.211678 Objective Loss 0.211678 LR 0.000500 Time 0.021440 -2022-12-06 11:09:53,052 - Epoch: [106][ 320/ 1200] Overall Loss 0.212367 Objective Loss 0.212367 LR 0.000500 Time 0.021362 -2022-12-06 11:09:53,242 - Epoch: [106][ 330/ 1200] Overall Loss 0.212405 Objective Loss 0.212405 LR 0.000500 Time 0.021289 -2022-12-06 11:09:53,432 - Epoch: [106][ 340/ 1200] Overall Loss 0.212643 Objective Loss 0.212643 LR 0.000500 Time 0.021220 -2022-12-06 11:09:53,622 - Epoch: [106][ 350/ 1200] Overall Loss 0.212737 Objective Loss 0.212737 LR 0.000500 Time 0.021155 -2022-12-06 11:09:53,812 - Epoch: [106][ 360/ 1200] Overall Loss 0.212915 Objective Loss 0.212915 LR 0.000500 Time 0.021094 -2022-12-06 11:09:54,002 - Epoch: [106][ 370/ 1200] Overall Loss 0.212305 Objective Loss 0.212305 LR 0.000500 Time 0.021036 -2022-12-06 11:09:54,193 - Epoch: [106][ 380/ 1200] Overall Loss 0.211923 Objective Loss 0.211923 LR 0.000500 Time 0.020983 -2022-12-06 11:09:54,383 - Epoch: [106][ 390/ 1200] Overall Loss 0.212492 Objective Loss 0.212492 LR 0.000500 Time 0.020931 -2022-12-06 11:09:54,574 - Epoch: [106][ 400/ 1200] Overall Loss 0.212186 Objective Loss 0.212186 LR 0.000500 Time 0.020883 -2022-12-06 11:09:54,764 - Epoch: [106][ 410/ 1200] Overall Loss 0.211568 Objective Loss 0.211568 LR 0.000500 Time 0.020837 -2022-12-06 11:09:54,954 - Epoch: [106][ 420/ 1200] Overall Loss 0.211407 Objective Loss 0.211407 LR 0.000500 Time 0.020792 -2022-12-06 11:09:55,145 - Epoch: [106][ 430/ 1200] Overall Loss 0.211453 Objective Loss 0.211453 LR 0.000500 Time 0.020752 -2022-12-06 11:09:55,335 - Epoch: [106][ 440/ 1200] Overall Loss 0.211602 Objective Loss 0.211602 LR 0.000500 Time 0.020711 -2022-12-06 11:09:55,525 - Epoch: [106][ 450/ 1200] Overall Loss 0.211672 Objective Loss 0.211672 LR 0.000500 Time 0.020672 -2022-12-06 11:09:55,717 - Epoch: [106][ 460/ 1200] Overall Loss 0.211481 Objective Loss 0.211481 LR 0.000500 Time 0.020638 -2022-12-06 11:09:55,907 - Epoch: [106][ 470/ 1200] Overall Loss 0.211521 Objective Loss 0.211521 LR 0.000500 Time 0.020603 -2022-12-06 11:09:56,098 - Epoch: [106][ 480/ 1200] Overall Loss 0.211422 Objective Loss 0.211422 LR 0.000500 Time 0.020570 -2022-12-06 11:09:56,288 - Epoch: [106][ 490/ 1200] Overall Loss 0.211744 Objective Loss 0.211744 LR 0.000500 Time 0.020536 -2022-12-06 11:09:56,478 - Epoch: [106][ 500/ 1200] Overall Loss 0.212362 Objective Loss 0.212362 LR 0.000500 Time 0.020505 -2022-12-06 11:09:56,668 - Epoch: [106][ 510/ 1200] Overall Loss 0.212310 Objective Loss 0.212310 LR 0.000500 Time 0.020474 -2022-12-06 11:09:56,859 - Epoch: [106][ 520/ 1200] Overall Loss 0.212626 Objective Loss 0.212626 LR 0.000500 Time 0.020446 -2022-12-06 11:09:57,048 - Epoch: [106][ 530/ 1200] Overall Loss 0.212641 Objective Loss 0.212641 LR 0.000500 Time 0.020417 -2022-12-06 11:09:57,239 - Epoch: [106][ 540/ 1200] Overall Loss 0.212256 Objective Loss 0.212256 LR 0.000500 Time 0.020390 -2022-12-06 11:09:57,429 - Epoch: [106][ 550/ 1200] Overall Loss 0.212107 Objective Loss 0.212107 LR 0.000500 Time 0.020364 -2022-12-06 11:09:57,619 - Epoch: [106][ 560/ 1200] Overall Loss 0.212258 Objective Loss 0.212258 LR 0.000500 Time 0.020339 -2022-12-06 11:09:57,809 - Epoch: [106][ 570/ 1200] Overall Loss 0.212306 Objective Loss 0.212306 LR 0.000500 Time 0.020314 -2022-12-06 11:09:57,999 - Epoch: [106][ 580/ 1200] Overall Loss 0.212003 Objective Loss 0.212003 LR 0.000500 Time 0.020291 -2022-12-06 11:09:58,190 - Epoch: [106][ 590/ 1200] Overall Loss 0.212465 Objective Loss 0.212465 LR 0.000500 Time 0.020269 -2022-12-06 11:09:58,380 - Epoch: [106][ 600/ 1200] Overall Loss 0.212691 Objective Loss 0.212691 LR 0.000500 Time 0.020247 -2022-12-06 11:09:58,569 - Epoch: [106][ 610/ 1200] Overall Loss 0.212618 Objective Loss 0.212618 LR 0.000500 Time 0.020226 -2022-12-06 11:09:58,760 - Epoch: [106][ 620/ 1200] Overall Loss 0.212421 Objective Loss 0.212421 LR 0.000500 Time 0.020205 -2022-12-06 11:09:58,949 - Epoch: [106][ 630/ 1200] Overall Loss 0.212415 Objective Loss 0.212415 LR 0.000500 Time 0.020184 -2022-12-06 11:09:59,139 - Epoch: [106][ 640/ 1200] Overall Loss 0.212209 Objective Loss 0.212209 LR 0.000500 Time 0.020165 -2022-12-06 11:09:59,329 - Epoch: [106][ 650/ 1200] Overall Loss 0.212328 Objective Loss 0.212328 LR 0.000500 Time 0.020147 -2022-12-06 11:09:59,519 - Epoch: [106][ 660/ 1200] Overall Loss 0.212614 Objective Loss 0.212614 LR 0.000500 Time 0.020128 -2022-12-06 11:09:59,710 - Epoch: [106][ 670/ 1200] Overall Loss 0.212699 Objective Loss 0.212699 LR 0.000500 Time 0.020111 -2022-12-06 11:09:59,901 - Epoch: [106][ 680/ 1200] Overall Loss 0.212745 Objective Loss 0.212745 LR 0.000500 Time 0.020096 -2022-12-06 11:10:00,092 - Epoch: [106][ 690/ 1200] Overall Loss 0.212679 Objective Loss 0.212679 LR 0.000500 Time 0.020080 -2022-12-06 11:10:00,282 - Epoch: [106][ 700/ 1200] Overall Loss 0.212491 Objective Loss 0.212491 LR 0.000500 Time 0.020065 -2022-12-06 11:10:00,473 - Epoch: [106][ 710/ 1200] Overall Loss 0.212464 Objective Loss 0.212464 LR 0.000500 Time 0.020050 -2022-12-06 11:10:00,664 - Epoch: [106][ 720/ 1200] Overall Loss 0.212506 Objective Loss 0.212506 LR 0.000500 Time 0.020036 -2022-12-06 11:10:00,854 - Epoch: [106][ 730/ 1200] Overall Loss 0.212514 Objective Loss 0.212514 LR 0.000500 Time 0.020021 -2022-12-06 11:10:01,045 - Epoch: [106][ 740/ 1200] Overall Loss 0.212455 Objective Loss 0.212455 LR 0.000500 Time 0.020007 -2022-12-06 11:10:01,236 - Epoch: [106][ 750/ 1200] Overall Loss 0.212107 Objective Loss 0.212107 LR 0.000500 Time 0.019994 -2022-12-06 11:10:01,426 - Epoch: [106][ 760/ 1200] Overall Loss 0.212064 Objective Loss 0.212064 LR 0.000500 Time 0.019981 -2022-12-06 11:10:01,616 - Epoch: [106][ 770/ 1200] Overall Loss 0.212197 Objective Loss 0.212197 LR 0.000500 Time 0.019968 -2022-12-06 11:10:01,806 - Epoch: [106][ 780/ 1200] Overall Loss 0.212234 Objective Loss 0.212234 LR 0.000500 Time 0.019955 -2022-12-06 11:10:01,997 - Epoch: [106][ 790/ 1200] Overall Loss 0.212590 Objective Loss 0.212590 LR 0.000500 Time 0.019943 -2022-12-06 11:10:02,189 - Epoch: [106][ 800/ 1200] Overall Loss 0.212358 Objective Loss 0.212358 LR 0.000500 Time 0.019933 -2022-12-06 11:10:02,380 - Epoch: [106][ 810/ 1200] Overall Loss 0.212146 Objective Loss 0.212146 LR 0.000500 Time 0.019922 -2022-12-06 11:10:02,570 - Epoch: [106][ 820/ 1200] Overall Loss 0.212422 Objective Loss 0.212422 LR 0.000500 Time 0.019910 -2022-12-06 11:10:02,765 - Epoch: [106][ 830/ 1200] Overall Loss 0.212342 Objective Loss 0.212342 LR 0.000500 Time 0.019904 -2022-12-06 11:10:02,965 - Epoch: [106][ 840/ 1200] Overall Loss 0.212097 Objective Loss 0.212097 LR 0.000500 Time 0.019905 -2022-12-06 11:10:03,169 - Epoch: [106][ 850/ 1200] Overall Loss 0.212182 Objective Loss 0.212182 LR 0.000500 Time 0.019910 -2022-12-06 11:10:03,369 - Epoch: [106][ 860/ 1200] Overall Loss 0.212397 Objective Loss 0.212397 LR 0.000500 Time 0.019910 -2022-12-06 11:10:03,573 - Epoch: [106][ 870/ 1200] Overall Loss 0.212574 Objective Loss 0.212574 LR 0.000500 Time 0.019915 -2022-12-06 11:10:03,773 - Epoch: [106][ 880/ 1200] Overall Loss 0.212621 Objective Loss 0.212621 LR 0.000500 Time 0.019915 -2022-12-06 11:10:03,977 - Epoch: [106][ 890/ 1200] Overall Loss 0.212530 Objective Loss 0.212530 LR 0.000500 Time 0.019920 -2022-12-06 11:10:04,177 - Epoch: [106][ 900/ 1200] Overall Loss 0.212905 Objective Loss 0.212905 LR 0.000500 Time 0.019920 -2022-12-06 11:10:04,381 - Epoch: [106][ 910/ 1200] Overall Loss 0.213084 Objective Loss 0.213084 LR 0.000500 Time 0.019924 -2022-12-06 11:10:04,582 - Epoch: [106][ 920/ 1200] Overall Loss 0.212989 Objective Loss 0.212989 LR 0.000500 Time 0.019925 -2022-12-06 11:10:04,785 - Epoch: [106][ 930/ 1200] Overall Loss 0.212785 Objective Loss 0.212785 LR 0.000500 Time 0.019930 -2022-12-06 11:10:04,985 - Epoch: [106][ 940/ 1200] Overall Loss 0.212803 Objective Loss 0.212803 LR 0.000500 Time 0.019930 -2022-12-06 11:10:05,189 - Epoch: [106][ 950/ 1200] Overall Loss 0.212549 Objective Loss 0.212549 LR 0.000500 Time 0.019934 -2022-12-06 11:10:05,389 - Epoch: [106][ 960/ 1200] Overall Loss 0.212755 Objective Loss 0.212755 LR 0.000500 Time 0.019934 -2022-12-06 11:10:05,593 - Epoch: [106][ 970/ 1200] Overall Loss 0.212894 Objective Loss 0.212894 LR 0.000500 Time 0.019938 -2022-12-06 11:10:05,793 - Epoch: [106][ 980/ 1200] Overall Loss 0.212904 Objective Loss 0.212904 LR 0.000500 Time 0.019938 -2022-12-06 11:10:05,997 - Epoch: [106][ 990/ 1200] Overall Loss 0.213412 Objective Loss 0.213412 LR 0.000500 Time 0.019942 -2022-12-06 11:10:06,197 - Epoch: [106][ 1000/ 1200] Overall Loss 0.213324 Objective Loss 0.213324 LR 0.000500 Time 0.019942 -2022-12-06 11:10:06,402 - Epoch: [106][ 1010/ 1200] Overall Loss 0.213225 Objective Loss 0.213225 LR 0.000500 Time 0.019947 -2022-12-06 11:10:06,604 - Epoch: [106][ 1020/ 1200] Overall Loss 0.213453 Objective Loss 0.213453 LR 0.000500 Time 0.019948 -2022-12-06 11:10:06,807 - Epoch: [106][ 1030/ 1200] Overall Loss 0.213329 Objective Loss 0.213329 LR 0.000500 Time 0.019952 -2022-12-06 11:10:07,008 - Epoch: [106][ 1040/ 1200] Overall Loss 0.213256 Objective Loss 0.213256 LR 0.000500 Time 0.019953 -2022-12-06 11:10:07,212 - Epoch: [106][ 1050/ 1200] Overall Loss 0.213277 Objective Loss 0.213277 LR 0.000500 Time 0.019957 -2022-12-06 11:10:07,414 - Epoch: [106][ 1060/ 1200] Overall Loss 0.213503 Objective Loss 0.213503 LR 0.000500 Time 0.019958 -2022-12-06 11:10:07,617 - Epoch: [106][ 1070/ 1200] Overall Loss 0.213666 Objective Loss 0.213666 LR 0.000500 Time 0.019961 -2022-12-06 11:10:07,818 - Epoch: [106][ 1080/ 1200] Overall Loss 0.213660 Objective Loss 0.213660 LR 0.000500 Time 0.019961 -2022-12-06 11:10:08,023 - Epoch: [106][ 1090/ 1200] Overall Loss 0.213648 Objective Loss 0.213648 LR 0.000500 Time 0.019965 -2022-12-06 11:10:08,223 - Epoch: [106][ 1100/ 1200] Overall Loss 0.213623 Objective Loss 0.213623 LR 0.000500 Time 0.019966 -2022-12-06 11:10:08,427 - Epoch: [106][ 1110/ 1200] Overall Loss 0.213771 Objective Loss 0.213771 LR 0.000500 Time 0.019969 -2022-12-06 11:10:08,628 - Epoch: [106][ 1120/ 1200] Overall Loss 0.213954 Objective Loss 0.213954 LR 0.000500 Time 0.019969 -2022-12-06 11:10:08,833 - Epoch: [106][ 1130/ 1200] Overall Loss 0.213997 Objective Loss 0.213997 LR 0.000500 Time 0.019973 -2022-12-06 11:10:09,034 - Epoch: [106][ 1140/ 1200] Overall Loss 0.213850 Objective Loss 0.213850 LR 0.000500 Time 0.019974 -2022-12-06 11:10:09,238 - Epoch: [106][ 1150/ 1200] Overall Loss 0.213858 Objective Loss 0.213858 LR 0.000500 Time 0.019977 -2022-12-06 11:10:09,438 - Epoch: [106][ 1160/ 1200] Overall Loss 0.213733 Objective Loss 0.213733 LR 0.000500 Time 0.019977 -2022-12-06 11:10:09,642 - Epoch: [106][ 1170/ 1200] Overall Loss 0.213795 Objective Loss 0.213795 LR 0.000500 Time 0.019980 -2022-12-06 11:10:09,842 - Epoch: [106][ 1180/ 1200] Overall Loss 0.213792 Objective Loss 0.213792 LR 0.000500 Time 0.019980 -2022-12-06 11:10:10,048 - Epoch: [106][ 1190/ 1200] Overall Loss 0.213906 Objective Loss 0.213906 LR 0.000500 Time 0.019984 -2022-12-06 11:10:10,276 - Epoch: [106][ 1200/ 1200] Overall Loss 0.213828 Objective Loss 0.213828 Top1 85.983264 Top5 99.163180 LR 0.000500 Time 0.020007 -2022-12-06 11:10:10,364 - --- validate (epoch=106)----------- -2022-12-06 11:10:10,365 - 34129 samples (256 per mini-batch) -2022-12-06 11:10:10,818 - Epoch: [106][ 10/ 134] Loss 0.241036 Top1 85.976562 Top5 98.281250 -2022-12-06 11:10:10,946 - Epoch: [106][ 20/ 134] Loss 0.241403 Top1 85.781250 Top5 98.281250 -2022-12-06 11:10:11,071 - Epoch: [106][ 30/ 134] Loss 0.252846 Top1 85.494792 Top5 98.255208 -2022-12-06 11:10:11,200 - Epoch: [106][ 40/ 134] Loss 0.250426 Top1 85.625000 Top5 98.291016 -2022-12-06 11:10:11,328 - Epoch: [106][ 50/ 134] Loss 0.248560 Top1 85.875000 Top5 98.250000 -2022-12-06 11:10:11,452 - Epoch: [106][ 60/ 134] Loss 0.255223 Top1 85.839844 Top5 98.222656 -2022-12-06 11:10:11,578 - Epoch: [106][ 70/ 134] Loss 0.262087 Top1 85.647321 Top5 98.091518 -2022-12-06 11:10:11,705 - Epoch: [106][ 80/ 134] Loss 0.263282 Top1 85.673828 Top5 98.085938 -2022-12-06 11:10:11,829 - Epoch: [106][ 90/ 134] Loss 0.260218 Top1 85.802951 Top5 98.081597 -2022-12-06 11:10:11,956 - Epoch: [106][ 100/ 134] Loss 0.262001 Top1 85.730469 Top5 98.105469 -2022-12-06 11:10:12,082 - Epoch: [106][ 110/ 134] Loss 0.260848 Top1 85.799006 Top5 98.135653 -2022-12-06 11:10:12,209 - Epoch: [106][ 120/ 134] Loss 0.259696 Top1 85.846354 Top5 98.138021 -2022-12-06 11:10:12,337 - Epoch: [106][ 130/ 134] Loss 0.258726 Top1 85.871394 Top5 98.185096 -2022-12-06 11:10:12,373 - Epoch: [106][ 134/ 134] Loss 0.259285 Top1 85.844883 Top5 98.189223 -2022-12-06 11:10:12,460 - ==> Top1: 85.845 Top5: 98.189 Loss: 0.259 - -2022-12-06 11:10:12,461 - ==> Confusion: -[[ 933 2 0 2 5 3 0 1 7 30 0 0 1 2 4 2 1 0 1 0 2] - [ 3 945 0 1 5 23 1 11 0 1 5 5 0 2 2 1 2 1 11 2 6] - [ 7 1 995 15 2 0 18 10 3 2 5 6 3 1 7 2 1 0 6 5 14] - [ 2 2 15 949 2 3 0 1 0 0 7 0 5 2 14 0 1 4 9 0 4] - [ 12 5 2 0 951 4 0 2 1 6 0 3 0 1 8 6 10 2 2 0 5] - [ 1 18 1 3 6 970 3 19 1 1 1 11 5 14 1 1 0 1 2 6 4] - [ 2 3 11 4 0 3 1064 4 0 1 0 3 0 1 0 6 0 0 2 12 2] - [ 1 9 9 3 2 26 5 950 0 0 2 5 1 0 0 0 1 2 24 10 4] - [ 7 3 0 0 1 3 0 2 975 33 7 1 2 6 15 1 2 2 3 1 0] - [ 87 0 3 0 8 2 0 4 17 852 2 2 0 8 4 1 2 3 0 0 6] - [ 2 2 1 8 2 0 0 3 8 0 953 3 2 12 5 0 1 1 8 3 5] - [ 3 0 3 0 1 8 3 3 1 0 0 979 18 4 0 5 3 4 3 10 3] - [ 0 1 3 3 0 1 0 0 0 1 0 33 881 1 3 13 2 14 2 3 8] - [ 0 0 1 0 0 10 0 2 9 15 2 4 2 955 2 2 3 2 0 5 9] - [ 6 4 1 12 3 1 0 0 13 2 0 3 3 3 1063 0 0 3 7 2 4] - [ 1 1 0 0 2 1 1 0 0 1 0 7 9 1 0 998 7 9 0 3 2] - [ 5 3 1 1 3 0 1 0 0 0 0 3 1 1 2 12 1032 0 0 4 3] - [ 4 0 2 3 1 1 0 2 1 3 0 6 16 1 2 17 1 975 0 0 1] - [ 4 3 3 17 1 4 0 25 1 0 2 2 2 0 5 0 1 2 931 2 3] - [ 3 2 0 0 1 5 2 5 0 1 0 16 7 5 0 6 6 1 2 1013 5] - [ 141 216 165 137 112 196 67 166 87 77 138 120 343 272 170 136 226 90 177 263 9927]] - -2022-12-06 11:10:13,030 - ==> Best [Top1: 86.208 Top5: 98.101 Sparsity:0.00 Params: 5376 on epoch: 101] -2022-12-06 11:10:13,031 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:10:13,037 - - -2022-12-06 11:10:13,037 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:10:14,059 - Epoch: [107][ 10/ 1200] Overall Loss 0.203838 Objective Loss 0.203838 LR 0.000500 Time 0.102198 -2022-12-06 11:10:14,259 - Epoch: [107][ 20/ 1200] Overall Loss 0.204608 Objective Loss 0.204608 LR 0.000500 Time 0.061030 -2022-12-06 11:10:14,451 - Epoch: [107][ 30/ 1200] Overall Loss 0.207354 Objective Loss 0.207354 LR 0.000500 Time 0.047072 -2022-12-06 11:10:14,641 - Epoch: [107][ 40/ 1200] Overall Loss 0.204433 Objective Loss 0.204433 LR 0.000500 Time 0.040057 -2022-12-06 11:10:14,832 - Epoch: [107][ 50/ 1200] Overall Loss 0.202922 Objective Loss 0.202922 LR 0.000500 Time 0.035852 -2022-12-06 11:10:15,023 - Epoch: [107][ 60/ 1200] Overall Loss 0.201556 Objective Loss 0.201556 LR 0.000500 Time 0.033047 -2022-12-06 11:10:15,214 - Epoch: [107][ 70/ 1200] Overall Loss 0.200474 Objective Loss 0.200474 LR 0.000500 Time 0.031042 -2022-12-06 11:10:15,405 - Epoch: [107][ 80/ 1200] Overall Loss 0.202749 Objective Loss 0.202749 LR 0.000500 Time 0.029541 -2022-12-06 11:10:15,596 - Epoch: [107][ 90/ 1200] Overall Loss 0.201563 Objective Loss 0.201563 LR 0.000500 Time 0.028374 -2022-12-06 11:10:15,787 - Epoch: [107][ 100/ 1200] Overall Loss 0.202471 Objective Loss 0.202471 LR 0.000500 Time 0.027444 -2022-12-06 11:10:15,978 - Epoch: [107][ 110/ 1200] Overall Loss 0.202585 Objective Loss 0.202585 LR 0.000500 Time 0.026678 -2022-12-06 11:10:16,169 - Epoch: [107][ 120/ 1200] Overall Loss 0.203564 Objective Loss 0.203564 LR 0.000500 Time 0.026045 -2022-12-06 11:10:16,360 - Epoch: [107][ 130/ 1200] Overall Loss 0.203064 Objective Loss 0.203064 LR 0.000500 Time 0.025507 -2022-12-06 11:10:16,550 - Epoch: [107][ 140/ 1200] Overall Loss 0.204942 Objective Loss 0.204942 LR 0.000500 Time 0.025041 -2022-12-06 11:10:16,741 - Epoch: [107][ 150/ 1200] Overall Loss 0.205056 Objective Loss 0.205056 LR 0.000500 Time 0.024641 -2022-12-06 11:10:16,932 - Epoch: [107][ 160/ 1200] Overall Loss 0.205420 Objective Loss 0.205420 LR 0.000500 Time 0.024288 -2022-12-06 11:10:17,122 - Epoch: [107][ 170/ 1200] Overall Loss 0.205485 Objective Loss 0.205485 LR 0.000500 Time 0.023977 -2022-12-06 11:10:17,314 - Epoch: [107][ 180/ 1200] Overall Loss 0.206031 Objective Loss 0.206031 LR 0.000500 Time 0.023704 -2022-12-06 11:10:17,504 - Epoch: [107][ 190/ 1200] Overall Loss 0.208309 Objective Loss 0.208309 LR 0.000500 Time 0.023456 -2022-12-06 11:10:17,695 - Epoch: [107][ 200/ 1200] Overall Loss 0.209359 Objective Loss 0.209359 LR 0.000500 Time 0.023235 -2022-12-06 11:10:17,885 - Epoch: [107][ 210/ 1200] Overall Loss 0.208109 Objective Loss 0.208109 LR 0.000500 Time 0.023033 -2022-12-06 11:10:18,076 - Epoch: [107][ 220/ 1200] Overall Loss 0.208428 Objective Loss 0.208428 LR 0.000500 Time 0.022852 -2022-12-06 11:10:18,268 - Epoch: [107][ 230/ 1200] Overall Loss 0.208854 Objective Loss 0.208854 LR 0.000500 Time 0.022687 -2022-12-06 11:10:18,458 - Epoch: [107][ 240/ 1200] Overall Loss 0.209467 Objective Loss 0.209467 LR 0.000500 Time 0.022532 -2022-12-06 11:10:18,649 - Epoch: [107][ 250/ 1200] Overall Loss 0.209820 Objective Loss 0.209820 LR 0.000500 Time 0.022393 -2022-12-06 11:10:18,839 - Epoch: [107][ 260/ 1200] Overall Loss 0.211341 Objective Loss 0.211341 LR 0.000500 Time 0.022262 -2022-12-06 11:10:19,030 - Epoch: [107][ 270/ 1200] Overall Loss 0.210309 Objective Loss 0.210309 LR 0.000500 Time 0.022141 -2022-12-06 11:10:19,221 - Epoch: [107][ 280/ 1200] Overall Loss 0.210592 Objective Loss 0.210592 LR 0.000500 Time 0.022030 -2022-12-06 11:10:19,412 - Epoch: [107][ 290/ 1200] Overall Loss 0.211197 Objective Loss 0.211197 LR 0.000500 Time 0.021928 -2022-12-06 11:10:19,603 - Epoch: [107][ 300/ 1200] Overall Loss 0.210622 Objective Loss 0.210622 LR 0.000500 Time 0.021831 -2022-12-06 11:10:19,793 - Epoch: [107][ 310/ 1200] Overall Loss 0.210727 Objective Loss 0.210727 LR 0.000500 Time 0.021740 -2022-12-06 11:10:19,984 - Epoch: [107][ 320/ 1200] Overall Loss 0.210386 Objective Loss 0.210386 LR 0.000500 Time 0.021656 -2022-12-06 11:10:20,176 - Epoch: [107][ 330/ 1200] Overall Loss 0.210873 Objective Loss 0.210873 LR 0.000500 Time 0.021578 -2022-12-06 11:10:20,367 - Epoch: [107][ 340/ 1200] Overall Loss 0.210687 Objective Loss 0.210687 LR 0.000500 Time 0.021503 -2022-12-06 11:10:20,557 - Epoch: [107][ 350/ 1200] Overall Loss 0.210979 Objective Loss 0.210979 LR 0.000500 Time 0.021432 -2022-12-06 11:10:20,749 - Epoch: [107][ 360/ 1200] Overall Loss 0.211398 Objective Loss 0.211398 LR 0.000500 Time 0.021367 -2022-12-06 11:10:20,939 - Epoch: [107][ 370/ 1200] Overall Loss 0.211595 Objective Loss 0.211595 LR 0.000500 Time 0.021303 -2022-12-06 11:10:21,130 - Epoch: [107][ 380/ 1200] Overall Loss 0.212049 Objective Loss 0.212049 LR 0.000500 Time 0.021244 -2022-12-06 11:10:21,322 - Epoch: [107][ 390/ 1200] Overall Loss 0.211993 Objective Loss 0.211993 LR 0.000500 Time 0.021188 -2022-12-06 11:10:21,513 - Epoch: [107][ 400/ 1200] Overall Loss 0.211752 Objective Loss 0.211752 LR 0.000500 Time 0.021136 -2022-12-06 11:10:21,704 - Epoch: [107][ 410/ 1200] Overall Loss 0.211064 Objective Loss 0.211064 LR 0.000500 Time 0.021085 -2022-12-06 11:10:21,895 - Epoch: [107][ 420/ 1200] Overall Loss 0.211249 Objective Loss 0.211249 LR 0.000500 Time 0.021037 -2022-12-06 11:10:22,086 - Epoch: [107][ 430/ 1200] Overall Loss 0.211535 Objective Loss 0.211535 LR 0.000500 Time 0.020990 -2022-12-06 11:10:22,278 - Epoch: [107][ 440/ 1200] Overall Loss 0.211580 Objective Loss 0.211580 LR 0.000500 Time 0.020946 -2022-12-06 11:10:22,469 - Epoch: [107][ 450/ 1200] Overall Loss 0.211707 Objective Loss 0.211707 LR 0.000500 Time 0.020905 -2022-12-06 11:10:22,659 - Epoch: [107][ 460/ 1200] Overall Loss 0.211790 Objective Loss 0.211790 LR 0.000500 Time 0.020863 -2022-12-06 11:10:22,850 - Epoch: [107][ 470/ 1200] Overall Loss 0.212006 Objective Loss 0.212006 LR 0.000500 Time 0.020825 -2022-12-06 11:10:23,041 - Epoch: [107][ 480/ 1200] Overall Loss 0.211870 Objective Loss 0.211870 LR 0.000500 Time 0.020787 -2022-12-06 11:10:23,232 - Epoch: [107][ 490/ 1200] Overall Loss 0.211522 Objective Loss 0.211522 LR 0.000500 Time 0.020752 -2022-12-06 11:10:23,423 - Epoch: [107][ 500/ 1200] Overall Loss 0.211176 Objective Loss 0.211176 LR 0.000500 Time 0.020717 -2022-12-06 11:10:23,615 - Epoch: [107][ 510/ 1200] Overall Loss 0.211310 Objective Loss 0.211310 LR 0.000500 Time 0.020686 -2022-12-06 11:10:23,806 - Epoch: [107][ 520/ 1200] Overall Loss 0.211551 Objective Loss 0.211551 LR 0.000500 Time 0.020655 -2022-12-06 11:10:23,997 - Epoch: [107][ 530/ 1200] Overall Loss 0.212089 Objective Loss 0.212089 LR 0.000500 Time 0.020624 -2022-12-06 11:10:24,188 - Epoch: [107][ 540/ 1200] Overall Loss 0.212263 Objective Loss 0.212263 LR 0.000500 Time 0.020595 -2022-12-06 11:10:24,379 - Epoch: [107][ 550/ 1200] Overall Loss 0.212089 Objective Loss 0.212089 LR 0.000500 Time 0.020567 -2022-12-06 11:10:24,570 - Epoch: [107][ 560/ 1200] Overall Loss 0.211978 Objective Loss 0.211978 LR 0.000500 Time 0.020540 -2022-12-06 11:10:24,762 - Epoch: [107][ 570/ 1200] Overall Loss 0.211630 Objective Loss 0.211630 LR 0.000500 Time 0.020515 -2022-12-06 11:10:24,954 - Epoch: [107][ 580/ 1200] Overall Loss 0.211592 Objective Loss 0.211592 LR 0.000500 Time 0.020493 -2022-12-06 11:10:25,147 - Epoch: [107][ 590/ 1200] Overall Loss 0.211641 Objective Loss 0.211641 LR 0.000500 Time 0.020471 -2022-12-06 11:10:25,340 - Epoch: [107][ 600/ 1200] Overall Loss 0.211859 Objective Loss 0.211859 LR 0.000500 Time 0.020450 -2022-12-06 11:10:25,532 - Epoch: [107][ 610/ 1200] Overall Loss 0.211683 Objective Loss 0.211683 LR 0.000500 Time 0.020429 -2022-12-06 11:10:25,725 - Epoch: [107][ 620/ 1200] Overall Loss 0.212152 Objective Loss 0.212152 LR 0.000500 Time 0.020409 -2022-12-06 11:10:25,918 - Epoch: [107][ 630/ 1200] Overall Loss 0.211839 Objective Loss 0.211839 LR 0.000500 Time 0.020391 -2022-12-06 11:10:26,110 - Epoch: [107][ 640/ 1200] Overall Loss 0.211928 Objective Loss 0.211928 LR 0.000500 Time 0.020372 -2022-12-06 11:10:26,303 - Epoch: [107][ 650/ 1200] Overall Loss 0.212177 Objective Loss 0.212177 LR 0.000500 Time 0.020355 -2022-12-06 11:10:26,497 - Epoch: [107][ 660/ 1200] Overall Loss 0.212104 Objective Loss 0.212104 LR 0.000500 Time 0.020339 -2022-12-06 11:10:26,690 - Epoch: [107][ 670/ 1200] Overall Loss 0.211741 Objective Loss 0.211741 LR 0.000500 Time 0.020322 -2022-12-06 11:10:26,881 - Epoch: [107][ 680/ 1200] Overall Loss 0.211804 Objective Loss 0.211804 LR 0.000500 Time 0.020304 -2022-12-06 11:10:27,072 - Epoch: [107][ 690/ 1200] Overall Loss 0.211649 Objective Loss 0.211649 LR 0.000500 Time 0.020285 -2022-12-06 11:10:27,263 - Epoch: [107][ 700/ 1200] Overall Loss 0.211541 Objective Loss 0.211541 LR 0.000500 Time 0.020268 -2022-12-06 11:10:27,455 - Epoch: [107][ 710/ 1200] Overall Loss 0.211446 Objective Loss 0.211446 LR 0.000500 Time 0.020252 -2022-12-06 11:10:27,646 - Epoch: [107][ 720/ 1200] Overall Loss 0.211300 Objective Loss 0.211300 LR 0.000500 Time 0.020235 -2022-12-06 11:10:27,837 - Epoch: [107][ 730/ 1200] Overall Loss 0.211185 Objective Loss 0.211185 LR 0.000500 Time 0.020219 -2022-12-06 11:10:28,028 - Epoch: [107][ 740/ 1200] Overall Loss 0.211379 Objective Loss 0.211379 LR 0.000500 Time 0.020203 -2022-12-06 11:10:28,219 - Epoch: [107][ 750/ 1200] Overall Loss 0.211499 Objective Loss 0.211499 LR 0.000500 Time 0.020188 -2022-12-06 11:10:28,410 - Epoch: [107][ 760/ 1200] Overall Loss 0.211284 Objective Loss 0.211284 LR 0.000500 Time 0.020173 -2022-12-06 11:10:28,602 - Epoch: [107][ 770/ 1200] Overall Loss 0.211535 Objective Loss 0.211535 LR 0.000500 Time 0.020159 -2022-12-06 11:10:28,793 - Epoch: [107][ 780/ 1200] Overall Loss 0.211661 Objective Loss 0.211661 LR 0.000500 Time 0.020145 -2022-12-06 11:10:28,984 - Epoch: [107][ 790/ 1200] Overall Loss 0.211840 Objective Loss 0.211840 LR 0.000500 Time 0.020131 -2022-12-06 11:10:29,175 - Epoch: [107][ 800/ 1200] Overall Loss 0.211990 Objective Loss 0.211990 LR 0.000500 Time 0.020117 -2022-12-06 11:10:29,366 - Epoch: [107][ 810/ 1200] Overall Loss 0.212404 Objective Loss 0.212404 LR 0.000500 Time 0.020104 -2022-12-06 11:10:29,557 - Epoch: [107][ 820/ 1200] Overall Loss 0.212232 Objective Loss 0.212232 LR 0.000500 Time 0.020091 -2022-12-06 11:10:29,747 - Epoch: [107][ 830/ 1200] Overall Loss 0.212599 Objective Loss 0.212599 LR 0.000500 Time 0.020078 -2022-12-06 11:10:29,939 - Epoch: [107][ 840/ 1200] Overall Loss 0.212735 Objective Loss 0.212735 LR 0.000500 Time 0.020066 -2022-12-06 11:10:30,130 - Epoch: [107][ 850/ 1200] Overall Loss 0.212999 Objective Loss 0.212999 LR 0.000500 Time 0.020055 -2022-12-06 11:10:30,321 - Epoch: [107][ 860/ 1200] Overall Loss 0.212869 Objective Loss 0.212869 LR 0.000500 Time 0.020043 -2022-12-06 11:10:30,513 - Epoch: [107][ 870/ 1200] Overall Loss 0.212754 Objective Loss 0.212754 LR 0.000500 Time 0.020032 -2022-12-06 11:10:30,705 - Epoch: [107][ 880/ 1200] Overall Loss 0.212803 Objective Loss 0.212803 LR 0.000500 Time 0.020022 -2022-12-06 11:10:30,896 - Epoch: [107][ 890/ 1200] Overall Loss 0.212868 Objective Loss 0.212868 LR 0.000500 Time 0.020011 -2022-12-06 11:10:31,086 - Epoch: [107][ 900/ 1200] Overall Loss 0.212786 Objective Loss 0.212786 LR 0.000500 Time 0.020000 -2022-12-06 11:10:31,279 - Epoch: [107][ 910/ 1200] Overall Loss 0.212995 Objective Loss 0.212995 LR 0.000500 Time 0.019991 -2022-12-06 11:10:31,470 - Epoch: [107][ 920/ 1200] Overall Loss 0.212810 Objective Loss 0.212810 LR 0.000500 Time 0.019981 -2022-12-06 11:10:31,662 - Epoch: [107][ 930/ 1200] Overall Loss 0.212851 Objective Loss 0.212851 LR 0.000500 Time 0.019972 -2022-12-06 11:10:31,854 - Epoch: [107][ 940/ 1200] Overall Loss 0.212884 Objective Loss 0.212884 LR 0.000500 Time 0.019963 -2022-12-06 11:10:32,045 - Epoch: [107][ 950/ 1200] Overall Loss 0.213021 Objective Loss 0.213021 LR 0.000500 Time 0.019954 -2022-12-06 11:10:32,237 - Epoch: [107][ 960/ 1200] Overall Loss 0.213135 Objective Loss 0.213135 LR 0.000500 Time 0.019945 -2022-12-06 11:10:32,428 - Epoch: [107][ 970/ 1200] Overall Loss 0.212938 Objective Loss 0.212938 LR 0.000500 Time 0.019936 -2022-12-06 11:10:32,619 - Epoch: [107][ 980/ 1200] Overall Loss 0.212919 Objective Loss 0.212919 LR 0.000500 Time 0.019927 -2022-12-06 11:10:32,811 - Epoch: [107][ 990/ 1200] Overall Loss 0.212613 Objective Loss 0.212613 LR 0.000500 Time 0.019919 -2022-12-06 11:10:33,002 - Epoch: [107][ 1000/ 1200] Overall Loss 0.212724 Objective Loss 0.212724 LR 0.000500 Time 0.019910 -2022-12-06 11:10:33,193 - Epoch: [107][ 1010/ 1200] Overall Loss 0.212718 Objective Loss 0.212718 LR 0.000500 Time 0.019902 -2022-12-06 11:10:33,385 - Epoch: [107][ 1020/ 1200] Overall Loss 0.212542 Objective Loss 0.212542 LR 0.000500 Time 0.019893 -2022-12-06 11:10:33,576 - Epoch: [107][ 1030/ 1200] Overall Loss 0.212676 Objective Loss 0.212676 LR 0.000500 Time 0.019885 -2022-12-06 11:10:33,767 - Epoch: [107][ 1040/ 1200] Overall Loss 0.212822 Objective Loss 0.212822 LR 0.000500 Time 0.019878 -2022-12-06 11:10:33,959 - Epoch: [107][ 1050/ 1200] Overall Loss 0.212736 Objective Loss 0.212736 LR 0.000500 Time 0.019870 -2022-12-06 11:10:34,150 - Epoch: [107][ 1060/ 1200] Overall Loss 0.212893 Objective Loss 0.212893 LR 0.000500 Time 0.019863 -2022-12-06 11:10:34,341 - Epoch: [107][ 1070/ 1200] Overall Loss 0.213139 Objective Loss 0.213139 LR 0.000500 Time 0.019855 -2022-12-06 11:10:34,532 - Epoch: [107][ 1080/ 1200] Overall Loss 0.213230 Objective Loss 0.213230 LR 0.000500 Time 0.019848 -2022-12-06 11:10:34,724 - Epoch: [107][ 1090/ 1200] Overall Loss 0.213053 Objective Loss 0.213053 LR 0.000500 Time 0.019841 -2022-12-06 11:10:34,915 - Epoch: [107][ 1100/ 1200] Overall Loss 0.212852 Objective Loss 0.212852 LR 0.000500 Time 0.019834 -2022-12-06 11:10:35,107 - Epoch: [107][ 1110/ 1200] Overall Loss 0.212760 Objective Loss 0.212760 LR 0.000500 Time 0.019827 -2022-12-06 11:10:35,298 - Epoch: [107][ 1120/ 1200] Overall Loss 0.212777 Objective Loss 0.212777 LR 0.000500 Time 0.019820 -2022-12-06 11:10:35,489 - Epoch: [107][ 1130/ 1200] Overall Loss 0.212868 Objective Loss 0.212868 LR 0.000500 Time 0.019814 -2022-12-06 11:10:35,681 - Epoch: [107][ 1140/ 1200] Overall Loss 0.212873 Objective Loss 0.212873 LR 0.000500 Time 0.019808 -2022-12-06 11:10:35,872 - Epoch: [107][ 1150/ 1200] Overall Loss 0.212847 Objective Loss 0.212847 LR 0.000500 Time 0.019801 -2022-12-06 11:10:36,064 - Epoch: [107][ 1160/ 1200] Overall Loss 0.212768 Objective Loss 0.212768 LR 0.000500 Time 0.019795 -2022-12-06 11:10:36,255 - Epoch: [107][ 1170/ 1200] Overall Loss 0.212713 Objective Loss 0.212713 LR 0.000500 Time 0.019790 -2022-12-06 11:10:36,447 - Epoch: [107][ 1180/ 1200] Overall Loss 0.212882 Objective Loss 0.212882 LR 0.000500 Time 0.019783 -2022-12-06 11:10:36,638 - Epoch: [107][ 1190/ 1200] Overall Loss 0.212763 Objective Loss 0.212763 LR 0.000500 Time 0.019777 -2022-12-06 11:10:36,870 - Epoch: [107][ 1200/ 1200] Overall Loss 0.212737 Objective Loss 0.212737 Top1 86.192469 Top5 98.117155 LR 0.000500 Time 0.019806 -2022-12-06 11:10:36,966 - --- validate (epoch=107)----------- -2022-12-06 11:10:36,966 - 34129 samples (256 per mini-batch) -2022-12-06 11:10:37,412 - Epoch: [107][ 10/ 134] Loss 0.249434 Top1 86.601562 Top5 98.398438 -2022-12-06 11:10:37,545 - Epoch: [107][ 20/ 134] Loss 0.250767 Top1 86.347656 Top5 98.359375 -2022-12-06 11:10:37,670 - Epoch: [107][ 30/ 134] Loss 0.257477 Top1 86.380208 Top5 98.255208 -2022-12-06 11:10:37,795 - Epoch: [107][ 40/ 134] Loss 0.255576 Top1 86.474609 Top5 98.281250 -2022-12-06 11:10:37,925 - Epoch: [107][ 50/ 134] Loss 0.257810 Top1 86.218750 Top5 98.343750 -2022-12-06 11:10:38,054 - Epoch: [107][ 60/ 134] Loss 0.257778 Top1 86.295573 Top5 98.281250 -2022-12-06 11:10:38,183 - Epoch: [107][ 70/ 134] Loss 0.264124 Top1 86.261161 Top5 98.281250 -2022-12-06 11:10:38,311 - Epoch: [107][ 80/ 134] Loss 0.264949 Top1 86.196289 Top5 98.208008 -2022-12-06 11:10:38,440 - Epoch: [107][ 90/ 134] Loss 0.259106 Top1 86.354167 Top5 98.224826 -2022-12-06 11:10:38,569 - Epoch: [107][ 100/ 134] Loss 0.260475 Top1 86.277344 Top5 98.203125 -2022-12-06 11:10:38,697 - Epoch: [107][ 110/ 134] Loss 0.262901 Top1 86.271307 Top5 98.178267 -2022-12-06 11:10:38,826 - Epoch: [107][ 120/ 134] Loss 0.266204 Top1 86.233724 Top5 98.121745 -2022-12-06 11:10:38,953 - Epoch: [107][ 130/ 134] Loss 0.266473 Top1 86.246995 Top5 98.131010 -2022-12-06 11:10:38,989 - Epoch: [107][ 134/ 134] Loss 0.265288 Top1 86.255091 Top5 98.139412 -2022-12-06 11:10:39,076 - ==> Top1: 86.255 Top5: 98.139 Loss: 0.265 - -2022-12-06 11:10:39,077 - ==> Confusion: -[[ 914 1 1 3 1 3 1 1 7 49 0 0 1 4 5 2 1 1 0 0 1] - [ 4 938 4 3 6 17 4 11 5 0 4 5 1 0 0 1 2 2 13 0 7] - [ 4 2 997 10 4 1 27 4 1 2 8 4 2 3 6 1 2 4 8 2 11] - [ 3 1 15 936 2 3 0 0 0 1 10 0 5 1 18 1 1 3 15 0 5] - [ 14 5 1 0 941 3 1 1 3 8 1 2 2 2 14 4 8 2 0 1 7] - [ 2 25 0 2 4 956 2 24 2 3 1 11 5 15 1 2 1 2 2 4 5] - [ 1 2 7 2 1 1 1079 2 0 0 0 0 3 3 0 5 1 2 0 7 2] - [ 1 5 6 3 1 22 7 947 0 0 2 7 2 1 1 2 0 0 29 11 7] - [ 6 1 0 1 2 2 1 0 999 32 2 1 1 3 6 2 1 0 2 1 1] - [ 64 0 1 1 4 0 0 3 23 880 1 1 0 6 5 1 1 3 0 0 7] - [ 1 2 1 5 3 0 1 4 14 1 961 0 2 10 3 1 0 0 2 2 6] - [ 2 1 2 0 2 11 3 3 1 0 0 967 24 7 1 10 4 5 0 5 3] - [ 0 1 2 4 1 1 2 0 0 0 0 20 908 1 2 11 1 7 0 4 4] - [ 0 0 1 0 2 8 0 1 17 20 8 3 2 942 2 1 4 2 1 1 8] - [ 8 3 0 6 1 2 1 0 21 8 0 3 1 3 1056 1 2 0 9 1 4] - [ 1 0 1 0 2 2 3 0 0 0 0 7 9 2 0 996 6 10 1 2 1] - [ 2 4 0 3 1 2 0 0 0 0 0 2 0 2 1 11 1034 0 1 2 7] - [ 2 0 1 4 1 1 1 1 0 9 0 5 17 0 3 17 0 970 1 2 1] - [ 4 1 3 13 3 4 1 25 1 0 6 2 1 0 9 0 2 1 929 0 3] - [ 1 4 3 1 1 4 6 8 0 1 0 9 7 6 0 4 5 3 1 1009 7] - [ 163 199 177 112 82 136 94 130 119 102 162 110 372 274 191 109 172 78 182 194 10068]] - -2022-12-06 11:10:39,645 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:10:39,645 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:10:39,652 - - -2022-12-06 11:10:39,652 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:10:40,576 - Epoch: [108][ 10/ 1200] Overall Loss 0.183302 Objective Loss 0.183302 LR 0.000500 Time 0.092303 -2022-12-06 11:10:40,772 - Epoch: [108][ 20/ 1200] Overall Loss 0.189456 Objective Loss 0.189456 LR 0.000500 Time 0.055923 -2022-12-06 11:10:40,964 - Epoch: [108][ 30/ 1200] Overall Loss 0.198808 Objective Loss 0.198808 LR 0.000500 Time 0.043652 -2022-12-06 11:10:41,154 - Epoch: [108][ 40/ 1200] Overall Loss 0.207108 Objective Loss 0.207108 LR 0.000500 Time 0.037487 -2022-12-06 11:10:41,345 - Epoch: [108][ 50/ 1200] Overall Loss 0.205310 Objective Loss 0.205310 LR 0.000500 Time 0.033795 -2022-12-06 11:10:41,536 - Epoch: [108][ 60/ 1200] Overall Loss 0.205827 Objective Loss 0.205827 LR 0.000500 Time 0.031328 -2022-12-06 11:10:41,726 - Epoch: [108][ 70/ 1200] Overall Loss 0.205422 Objective Loss 0.205422 LR 0.000500 Time 0.029568 -2022-12-06 11:10:41,916 - Epoch: [108][ 80/ 1200] Overall Loss 0.202714 Objective Loss 0.202714 LR 0.000500 Time 0.028231 -2022-12-06 11:10:42,106 - Epoch: [108][ 90/ 1200] Overall Loss 0.202775 Objective Loss 0.202775 LR 0.000500 Time 0.027201 -2022-12-06 11:10:42,296 - Epoch: [108][ 100/ 1200] Overall Loss 0.202869 Objective Loss 0.202869 LR 0.000500 Time 0.026380 -2022-12-06 11:10:42,486 - Epoch: [108][ 110/ 1200] Overall Loss 0.201362 Objective Loss 0.201362 LR 0.000500 Time 0.025706 -2022-12-06 11:10:42,676 - Epoch: [108][ 120/ 1200] Overall Loss 0.199450 Objective Loss 0.199450 LR 0.000500 Time 0.025143 -2022-12-06 11:10:42,866 - Epoch: [108][ 130/ 1200] Overall Loss 0.201304 Objective Loss 0.201304 LR 0.000500 Time 0.024666 -2022-12-06 11:10:43,057 - Epoch: [108][ 140/ 1200] Overall Loss 0.200582 Objective Loss 0.200582 LR 0.000500 Time 0.024260 -2022-12-06 11:10:43,248 - Epoch: [108][ 150/ 1200] Overall Loss 0.201528 Objective Loss 0.201528 LR 0.000500 Time 0.023912 -2022-12-06 11:10:43,438 - Epoch: [108][ 160/ 1200] Overall Loss 0.203044 Objective Loss 0.203044 LR 0.000500 Time 0.023601 -2022-12-06 11:10:43,628 - Epoch: [108][ 170/ 1200] Overall Loss 0.202651 Objective Loss 0.202651 LR 0.000500 Time 0.023327 -2022-12-06 11:10:43,818 - Epoch: [108][ 180/ 1200] Overall Loss 0.203366 Objective Loss 0.203366 LR 0.000500 Time 0.023085 -2022-12-06 11:10:44,008 - Epoch: [108][ 190/ 1200] Overall Loss 0.203907 Objective Loss 0.203907 LR 0.000500 Time 0.022868 -2022-12-06 11:10:44,199 - Epoch: [108][ 200/ 1200] Overall Loss 0.203622 Objective Loss 0.203622 LR 0.000500 Time 0.022676 -2022-12-06 11:10:44,389 - Epoch: [108][ 210/ 1200] Overall Loss 0.203983 Objective Loss 0.203983 LR 0.000500 Time 0.022501 -2022-12-06 11:10:44,580 - Epoch: [108][ 220/ 1200] Overall Loss 0.204066 Objective Loss 0.204066 LR 0.000500 Time 0.022342 -2022-12-06 11:10:44,770 - Epoch: [108][ 230/ 1200] Overall Loss 0.204123 Objective Loss 0.204123 LR 0.000500 Time 0.022193 -2022-12-06 11:10:44,960 - Epoch: [108][ 240/ 1200] Overall Loss 0.203644 Objective Loss 0.203644 LR 0.000500 Time 0.022057 -2022-12-06 11:10:45,150 - Epoch: [108][ 250/ 1200] Overall Loss 0.204366 Objective Loss 0.204366 LR 0.000500 Time 0.021933 -2022-12-06 11:10:45,339 - Epoch: [108][ 260/ 1200] Overall Loss 0.204735 Objective Loss 0.204735 LR 0.000500 Time 0.021818 -2022-12-06 11:10:45,530 - Epoch: [108][ 270/ 1200] Overall Loss 0.205502 Objective Loss 0.205502 LR 0.000500 Time 0.021713 -2022-12-06 11:10:45,720 - Epoch: [108][ 280/ 1200] Overall Loss 0.205865 Objective Loss 0.205865 LR 0.000500 Time 0.021613 -2022-12-06 11:10:45,910 - Epoch: [108][ 290/ 1200] Overall Loss 0.206580 Objective Loss 0.206580 LR 0.000500 Time 0.021523 -2022-12-06 11:10:46,100 - Epoch: [108][ 300/ 1200] Overall Loss 0.207028 Objective Loss 0.207028 LR 0.000500 Time 0.021435 -2022-12-06 11:10:46,290 - Epoch: [108][ 310/ 1200] Overall Loss 0.206308 Objective Loss 0.206308 LR 0.000500 Time 0.021356 -2022-12-06 11:10:46,480 - Epoch: [108][ 320/ 1200] Overall Loss 0.206746 Objective Loss 0.206746 LR 0.000500 Time 0.021281 -2022-12-06 11:10:46,671 - Epoch: [108][ 330/ 1200] Overall Loss 0.206632 Objective Loss 0.206632 LR 0.000500 Time 0.021213 -2022-12-06 11:10:46,862 - Epoch: [108][ 340/ 1200] Overall Loss 0.206875 Objective Loss 0.206875 LR 0.000500 Time 0.021148 -2022-12-06 11:10:47,052 - Epoch: [108][ 350/ 1200] Overall Loss 0.206503 Objective Loss 0.206503 LR 0.000500 Time 0.021086 -2022-12-06 11:10:47,242 - Epoch: [108][ 360/ 1200] Overall Loss 0.206428 Objective Loss 0.206428 LR 0.000500 Time 0.021027 -2022-12-06 11:10:47,433 - Epoch: [108][ 370/ 1200] Overall Loss 0.206231 Objective Loss 0.206231 LR 0.000500 Time 0.020972 -2022-12-06 11:10:47,623 - Epoch: [108][ 380/ 1200] Overall Loss 0.205484 Objective Loss 0.205484 LR 0.000500 Time 0.020920 -2022-12-06 11:10:47,814 - Epoch: [108][ 390/ 1200] Overall Loss 0.206175 Objective Loss 0.206175 LR 0.000500 Time 0.020871 -2022-12-06 11:10:48,004 - Epoch: [108][ 400/ 1200] Overall Loss 0.206190 Objective Loss 0.206190 LR 0.000500 Time 0.020825 -2022-12-06 11:10:48,194 - Epoch: [108][ 410/ 1200] Overall Loss 0.206193 Objective Loss 0.206193 LR 0.000500 Time 0.020778 -2022-12-06 11:10:48,384 - Epoch: [108][ 420/ 1200] Overall Loss 0.206949 Objective Loss 0.206949 LR 0.000500 Time 0.020735 -2022-12-06 11:10:48,576 - Epoch: [108][ 430/ 1200] Overall Loss 0.207031 Objective Loss 0.207031 LR 0.000500 Time 0.020697 -2022-12-06 11:10:48,766 - Epoch: [108][ 440/ 1200] Overall Loss 0.207152 Objective Loss 0.207152 LR 0.000500 Time 0.020658 -2022-12-06 11:10:48,957 - Epoch: [108][ 450/ 1200] Overall Loss 0.207546 Objective Loss 0.207546 LR 0.000500 Time 0.020621 -2022-12-06 11:10:49,147 - Epoch: [108][ 460/ 1200] Overall Loss 0.208181 Objective Loss 0.208181 LR 0.000500 Time 0.020585 -2022-12-06 11:10:49,338 - Epoch: [108][ 470/ 1200] Overall Loss 0.207977 Objective Loss 0.207977 LR 0.000500 Time 0.020553 -2022-12-06 11:10:49,529 - Epoch: [108][ 480/ 1200] Overall Loss 0.208955 Objective Loss 0.208955 LR 0.000500 Time 0.020520 -2022-12-06 11:10:49,720 - Epoch: [108][ 490/ 1200] Overall Loss 0.209219 Objective Loss 0.209219 LR 0.000500 Time 0.020490 -2022-12-06 11:10:49,910 - Epoch: [108][ 500/ 1200] Overall Loss 0.209286 Objective Loss 0.209286 LR 0.000500 Time 0.020460 -2022-12-06 11:10:50,101 - Epoch: [108][ 510/ 1200] Overall Loss 0.209091 Objective Loss 0.209091 LR 0.000500 Time 0.020432 -2022-12-06 11:10:50,292 - Epoch: [108][ 520/ 1200] Overall Loss 0.209158 Objective Loss 0.209158 LR 0.000500 Time 0.020405 -2022-12-06 11:10:50,483 - Epoch: [108][ 530/ 1200] Overall Loss 0.209379 Objective Loss 0.209379 LR 0.000500 Time 0.020380 -2022-12-06 11:10:50,673 - Epoch: [108][ 540/ 1200] Overall Loss 0.209323 Objective Loss 0.209323 LR 0.000500 Time 0.020354 -2022-12-06 11:10:50,864 - Epoch: [108][ 550/ 1200] Overall Loss 0.209271 Objective Loss 0.209271 LR 0.000500 Time 0.020329 -2022-12-06 11:10:51,055 - Epoch: [108][ 560/ 1200] Overall Loss 0.209274 Objective Loss 0.209274 LR 0.000500 Time 0.020306 -2022-12-06 11:10:51,245 - Epoch: [108][ 570/ 1200] Overall Loss 0.209522 Objective Loss 0.209522 LR 0.000500 Time 0.020283 -2022-12-06 11:10:51,435 - Epoch: [108][ 580/ 1200] Overall Loss 0.210053 Objective Loss 0.210053 LR 0.000500 Time 0.020260 -2022-12-06 11:10:51,626 - Epoch: [108][ 590/ 1200] Overall Loss 0.209742 Objective Loss 0.209742 LR 0.000500 Time 0.020239 -2022-12-06 11:10:51,819 - Epoch: [108][ 600/ 1200] Overall Loss 0.209691 Objective Loss 0.209691 LR 0.000500 Time 0.020222 -2022-12-06 11:10:52,011 - Epoch: [108][ 610/ 1200] Overall Loss 0.209419 Objective Loss 0.209419 LR 0.000500 Time 0.020205 -2022-12-06 11:10:52,203 - Epoch: [108][ 620/ 1200] Overall Loss 0.209633 Objective Loss 0.209633 LR 0.000500 Time 0.020188 -2022-12-06 11:10:52,396 - Epoch: [108][ 630/ 1200] Overall Loss 0.209962 Objective Loss 0.209962 LR 0.000500 Time 0.020172 -2022-12-06 11:10:52,588 - Epoch: [108][ 640/ 1200] Overall Loss 0.210234 Objective Loss 0.210234 LR 0.000500 Time 0.020157 -2022-12-06 11:10:52,781 - Epoch: [108][ 650/ 1200] Overall Loss 0.210434 Objective Loss 0.210434 LR 0.000500 Time 0.020142 -2022-12-06 11:10:52,973 - Epoch: [108][ 660/ 1200] Overall Loss 0.210446 Objective Loss 0.210446 LR 0.000500 Time 0.020128 -2022-12-06 11:10:53,165 - Epoch: [108][ 670/ 1200] Overall Loss 0.210518 Objective Loss 0.210518 LR 0.000500 Time 0.020113 -2022-12-06 11:10:53,358 - Epoch: [108][ 680/ 1200] Overall Loss 0.210429 Objective Loss 0.210429 LR 0.000500 Time 0.020099 -2022-12-06 11:10:53,550 - Epoch: [108][ 690/ 1200] Overall Loss 0.210416 Objective Loss 0.210416 LR 0.000500 Time 0.020087 -2022-12-06 11:10:53,743 - Epoch: [108][ 700/ 1200] Overall Loss 0.210259 Objective Loss 0.210259 LR 0.000500 Time 0.020074 -2022-12-06 11:10:53,935 - Epoch: [108][ 710/ 1200] Overall Loss 0.210041 Objective Loss 0.210041 LR 0.000500 Time 0.020061 -2022-12-06 11:10:54,127 - Epoch: [108][ 720/ 1200] Overall Loss 0.210004 Objective Loss 0.210004 LR 0.000500 Time 0.020048 -2022-12-06 11:10:54,319 - Epoch: [108][ 730/ 1200] Overall Loss 0.209942 Objective Loss 0.209942 LR 0.000500 Time 0.020036 -2022-12-06 11:10:54,511 - Epoch: [108][ 740/ 1200] Overall Loss 0.209588 Objective Loss 0.209588 LR 0.000500 Time 0.020024 -2022-12-06 11:10:54,704 - Epoch: [108][ 750/ 1200] Overall Loss 0.209501 Objective Loss 0.209501 LR 0.000500 Time 0.020014 -2022-12-06 11:10:54,897 - Epoch: [108][ 760/ 1200] Overall Loss 0.209568 Objective Loss 0.209568 LR 0.000500 Time 0.020003 -2022-12-06 11:10:55,089 - Epoch: [108][ 770/ 1200] Overall Loss 0.209418 Objective Loss 0.209418 LR 0.000500 Time 0.019992 -2022-12-06 11:10:55,281 - Epoch: [108][ 780/ 1200] Overall Loss 0.209676 Objective Loss 0.209676 LR 0.000500 Time 0.019981 -2022-12-06 11:10:55,473 - Epoch: [108][ 790/ 1200] Overall Loss 0.210073 Objective Loss 0.210073 LR 0.000500 Time 0.019971 -2022-12-06 11:10:55,666 - Epoch: [108][ 800/ 1200] Overall Loss 0.209845 Objective Loss 0.209845 LR 0.000500 Time 0.019962 -2022-12-06 11:10:55,859 - Epoch: [108][ 810/ 1200] Overall Loss 0.209849 Objective Loss 0.209849 LR 0.000500 Time 0.019952 -2022-12-06 11:10:56,051 - Epoch: [108][ 820/ 1200] Overall Loss 0.209542 Objective Loss 0.209542 LR 0.000500 Time 0.019943 -2022-12-06 11:10:56,243 - Epoch: [108][ 830/ 1200] Overall Loss 0.209469 Objective Loss 0.209469 LR 0.000500 Time 0.019933 -2022-12-06 11:10:56,436 - Epoch: [108][ 840/ 1200] Overall Loss 0.209445 Objective Loss 0.209445 LR 0.000500 Time 0.019925 -2022-12-06 11:10:56,628 - Epoch: [108][ 850/ 1200] Overall Loss 0.209412 Objective Loss 0.209412 LR 0.000500 Time 0.019916 -2022-12-06 11:10:56,820 - Epoch: [108][ 860/ 1200] Overall Loss 0.209441 Objective Loss 0.209441 LR 0.000500 Time 0.019908 -2022-12-06 11:10:57,012 - Epoch: [108][ 870/ 1200] Overall Loss 0.209448 Objective Loss 0.209448 LR 0.000500 Time 0.019899 -2022-12-06 11:10:57,205 - Epoch: [108][ 880/ 1200] Overall Loss 0.209574 Objective Loss 0.209574 LR 0.000500 Time 0.019891 -2022-12-06 11:10:57,397 - Epoch: [108][ 890/ 1200] Overall Loss 0.209643 Objective Loss 0.209643 LR 0.000500 Time 0.019882 -2022-12-06 11:10:57,589 - Epoch: [108][ 900/ 1200] Overall Loss 0.209416 Objective Loss 0.209416 LR 0.000500 Time 0.019875 -2022-12-06 11:10:57,781 - Epoch: [108][ 910/ 1200] Overall Loss 0.209372 Objective Loss 0.209372 LR 0.000500 Time 0.019867 -2022-12-06 11:10:57,974 - Epoch: [108][ 920/ 1200] Overall Loss 0.209715 Objective Loss 0.209715 LR 0.000500 Time 0.019859 -2022-12-06 11:10:58,166 - Epoch: [108][ 930/ 1200] Overall Loss 0.209866 Objective Loss 0.209866 LR 0.000500 Time 0.019852 -2022-12-06 11:10:58,359 - Epoch: [108][ 940/ 1200] Overall Loss 0.209591 Objective Loss 0.209591 LR 0.000500 Time 0.019845 -2022-12-06 11:10:58,551 - Epoch: [108][ 950/ 1200] Overall Loss 0.209440 Objective Loss 0.209440 LR 0.000500 Time 0.019838 -2022-12-06 11:10:58,744 - Epoch: [108][ 960/ 1200] Overall Loss 0.209321 Objective Loss 0.209321 LR 0.000500 Time 0.019831 -2022-12-06 11:10:58,936 - Epoch: [108][ 970/ 1200] Overall Loss 0.209596 Objective Loss 0.209596 LR 0.000500 Time 0.019825 -2022-12-06 11:10:59,128 - Epoch: [108][ 980/ 1200] Overall Loss 0.209512 Objective Loss 0.209512 LR 0.000500 Time 0.019818 -2022-12-06 11:10:59,320 - Epoch: [108][ 990/ 1200] Overall Loss 0.209293 Objective Loss 0.209293 LR 0.000500 Time 0.019812 -2022-12-06 11:10:59,513 - Epoch: [108][ 1000/ 1200] Overall Loss 0.209100 Objective Loss 0.209100 LR 0.000500 Time 0.019805 -2022-12-06 11:10:59,706 - Epoch: [108][ 1010/ 1200] Overall Loss 0.209058 Objective Loss 0.209058 LR 0.000500 Time 0.019799 -2022-12-06 11:10:59,899 - Epoch: [108][ 1020/ 1200] Overall Loss 0.209454 Objective Loss 0.209454 LR 0.000500 Time 0.019795 -2022-12-06 11:11:00,092 - Epoch: [108][ 1030/ 1200] Overall Loss 0.209519 Objective Loss 0.209519 LR 0.000500 Time 0.019789 -2022-12-06 11:11:00,284 - Epoch: [108][ 1040/ 1200] Overall Loss 0.209733 Objective Loss 0.209733 LR 0.000500 Time 0.019783 -2022-12-06 11:11:00,477 - Epoch: [108][ 1050/ 1200] Overall Loss 0.209635 Objective Loss 0.209635 LR 0.000500 Time 0.019778 -2022-12-06 11:11:00,669 - Epoch: [108][ 1060/ 1200] Overall Loss 0.209516 Objective Loss 0.209516 LR 0.000500 Time 0.019772 -2022-12-06 11:11:00,862 - Epoch: [108][ 1070/ 1200] Overall Loss 0.209556 Objective Loss 0.209556 LR 0.000500 Time 0.019767 -2022-12-06 11:11:01,054 - Epoch: [108][ 1080/ 1200] Overall Loss 0.209736 Objective Loss 0.209736 LR 0.000500 Time 0.019761 -2022-12-06 11:11:01,246 - Epoch: [108][ 1090/ 1200] Overall Loss 0.209799 Objective Loss 0.209799 LR 0.000500 Time 0.019756 -2022-12-06 11:11:01,439 - Epoch: [108][ 1100/ 1200] Overall Loss 0.209653 Objective Loss 0.209653 LR 0.000500 Time 0.019751 -2022-12-06 11:11:01,631 - Epoch: [108][ 1110/ 1200] Overall Loss 0.209596 Objective Loss 0.209596 LR 0.000500 Time 0.019746 -2022-12-06 11:11:01,823 - Epoch: [108][ 1120/ 1200] Overall Loss 0.209668 Objective Loss 0.209668 LR 0.000500 Time 0.019740 -2022-12-06 11:11:02,016 - Epoch: [108][ 1130/ 1200] Overall Loss 0.209525 Objective Loss 0.209525 LR 0.000500 Time 0.019735 -2022-12-06 11:11:02,208 - Epoch: [108][ 1140/ 1200] Overall Loss 0.209773 Objective Loss 0.209773 LR 0.000500 Time 0.019731 -2022-12-06 11:11:02,402 - Epoch: [108][ 1150/ 1200] Overall Loss 0.209655 Objective Loss 0.209655 LR 0.000500 Time 0.019727 -2022-12-06 11:11:02,594 - Epoch: [108][ 1160/ 1200] Overall Loss 0.209649 Objective Loss 0.209649 LR 0.000500 Time 0.019722 -2022-12-06 11:11:02,786 - Epoch: [108][ 1170/ 1200] Overall Loss 0.209607 Objective Loss 0.209607 LR 0.000500 Time 0.019718 -2022-12-06 11:11:02,979 - Epoch: [108][ 1180/ 1200] Overall Loss 0.209509 Objective Loss 0.209509 LR 0.000500 Time 0.019713 -2022-12-06 11:11:03,171 - Epoch: [108][ 1190/ 1200] Overall Loss 0.209599 Objective Loss 0.209599 LR 0.000500 Time 0.019709 -2022-12-06 11:11:03,400 - Epoch: [108][ 1200/ 1200] Overall Loss 0.209596 Objective Loss 0.209596 Top1 89.121339 Top5 98.744770 LR 0.000500 Time 0.019735 -2022-12-06 11:11:03,507 - --- validate (epoch=108)----------- -2022-12-06 11:11:03,507 - 34129 samples (256 per mini-batch) -2022-12-06 11:11:04,070 - Epoch: [108][ 10/ 134] Loss 0.221452 Top1 86.250000 Top5 98.554688 -2022-12-06 11:11:04,201 - Epoch: [108][ 20/ 134] Loss 0.234696 Top1 86.757812 Top5 98.378906 -2022-12-06 11:11:04,333 - Epoch: [108][ 30/ 134] Loss 0.249215 Top1 86.250000 Top5 98.203125 -2022-12-06 11:11:04,463 - Epoch: [108][ 40/ 134] Loss 0.258208 Top1 86.015625 Top5 98.154297 -2022-12-06 11:11:04,595 - Epoch: [108][ 50/ 134] Loss 0.255492 Top1 86.117188 Top5 98.171875 -2022-12-06 11:11:04,724 - Epoch: [108][ 60/ 134] Loss 0.255210 Top1 86.015625 Top5 98.125000 -2022-12-06 11:11:04,856 - Epoch: [108][ 70/ 134] Loss 0.256555 Top1 85.998884 Top5 98.108259 -2022-12-06 11:11:04,988 - Epoch: [108][ 80/ 134] Loss 0.256561 Top1 86.005859 Top5 98.168945 -2022-12-06 11:11:05,122 - Epoch: [108][ 90/ 134] Loss 0.255650 Top1 86.015625 Top5 98.194444 -2022-12-06 11:11:05,250 - Epoch: [108][ 100/ 134] Loss 0.255218 Top1 86.000000 Top5 98.175781 -2022-12-06 11:11:05,384 - Epoch: [108][ 110/ 134] Loss 0.255763 Top1 86.001420 Top5 98.135653 -2022-12-06 11:11:05,515 - Epoch: [108][ 120/ 134] Loss 0.255563 Top1 85.992839 Top5 98.167318 -2022-12-06 11:11:05,643 - Epoch: [108][ 130/ 134] Loss 0.257660 Top1 85.949519 Top5 98.143029 -2022-12-06 11:11:05,680 - Epoch: [108][ 134/ 134] Loss 0.256455 Top1 85.932784 Top5 98.145272 -2022-12-06 11:11:05,773 - ==> Top1: 85.933 Top5: 98.145 Loss: 0.256 - -2022-12-06 11:11:05,774 - ==> Confusion: -[[ 915 3 1 3 2 4 0 2 3 48 0 1 1 5 3 1 1 0 0 0 3] - [ 2 931 1 2 7 27 5 12 1 1 8 3 2 1 0 2 5 1 7 1 8] - [ 5 2 996 18 2 0 24 11 0 2 6 3 1 3 1 3 2 3 4 2 15] - [ 3 2 18 936 0 3 0 1 0 2 11 0 7 2 15 1 1 3 11 0 4] - [ 5 4 2 0 959 3 1 4 2 4 3 2 1 2 9 6 7 2 1 1 2] - [ 1 12 0 2 5 979 3 21 3 2 1 9 8 12 2 1 2 1 0 2 3] - [ 0 4 10 2 0 1 1069 4 0 1 3 2 3 2 0 5 0 0 2 6 4] - [ 1 6 5 1 2 24 6 961 0 0 4 9 1 1 0 0 2 0 15 7 9] - [ 6 3 0 0 1 3 0 0 976 43 9 1 2 6 7 1 1 2 2 1 0] - [ 71 0 0 0 4 1 0 5 17 884 2 0 0 9 1 1 0 0 0 0 6] - [ 0 2 0 2 2 1 1 3 8 0 974 1 3 9 4 0 0 0 1 2 6] - [ 2 0 0 1 0 11 4 5 0 2 0 949 36 7 2 5 5 5 2 11 4] - [ 0 1 2 4 0 1 1 0 0 0 0 18 909 1 0 7 2 10 0 6 7] - [ 0 1 1 1 0 5 0 1 16 14 6 2 4 960 0 2 2 1 1 1 5] - [ 7 4 1 11 10 3 1 0 22 5 0 1 3 4 1044 0 1 1 6 2 4] - [ 0 0 1 1 1 1 2 0 0 1 1 9 11 3 0 984 6 13 0 5 4] - [ 1 3 0 2 1 1 1 0 1 1 1 1 2 1 2 8 1035 0 1 4 6] - [ 3 0 1 4 0 1 1 0 1 2 0 4 18 3 0 11 1 984 1 0 1] - [ 3 2 1 5 1 4 0 23 3 0 12 2 3 2 7 0 1 0 937 0 2] - [ 1 3 4 1 0 7 6 4 0 0 2 6 8 7 0 3 3 2 1 1019 3] - [ 146 198 150 93 144 192 96 139 97 111 213 84 376 321 157 82 211 90 155 244 9927]] - -2022-12-06 11:11:06,348 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:11:06,348 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:11:06,354 - - -2022-12-06 11:11:06,354 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:11:07,290 - Epoch: [109][ 10/ 1200] Overall Loss 0.201404 Objective Loss 0.201404 LR 0.000500 Time 0.093527 -2022-12-06 11:11:07,484 - Epoch: [109][ 20/ 1200] Overall Loss 0.183074 Objective Loss 0.183074 LR 0.000500 Time 0.056437 -2022-12-06 11:11:07,676 - Epoch: [109][ 30/ 1200] Overall Loss 0.182073 Objective Loss 0.182073 LR 0.000500 Time 0.044012 -2022-12-06 11:11:07,869 - Epoch: [109][ 40/ 1200] Overall Loss 0.183637 Objective Loss 0.183637 LR 0.000500 Time 0.037812 -2022-12-06 11:11:08,061 - Epoch: [109][ 50/ 1200] Overall Loss 0.187082 Objective Loss 0.187082 LR 0.000500 Time 0.034087 -2022-12-06 11:11:08,254 - Epoch: [109][ 60/ 1200] Overall Loss 0.183881 Objective Loss 0.183881 LR 0.000500 Time 0.031600 -2022-12-06 11:11:08,446 - Epoch: [109][ 70/ 1200] Overall Loss 0.189885 Objective Loss 0.189885 LR 0.000500 Time 0.029825 -2022-12-06 11:11:08,638 - Epoch: [109][ 80/ 1200] Overall Loss 0.191686 Objective Loss 0.191686 LR 0.000500 Time 0.028490 -2022-12-06 11:11:08,830 - Epoch: [109][ 90/ 1200] Overall Loss 0.192230 Objective Loss 0.192230 LR 0.000500 Time 0.027449 -2022-12-06 11:11:09,022 - Epoch: [109][ 100/ 1200] Overall Loss 0.194465 Objective Loss 0.194465 LR 0.000500 Time 0.026619 -2022-12-06 11:11:09,214 - Epoch: [109][ 110/ 1200] Overall Loss 0.194500 Objective Loss 0.194500 LR 0.000500 Time 0.025941 -2022-12-06 11:11:09,406 - Epoch: [109][ 120/ 1200] Overall Loss 0.195865 Objective Loss 0.195865 LR 0.000500 Time 0.025379 -2022-12-06 11:11:09,599 - Epoch: [109][ 130/ 1200] Overall Loss 0.196696 Objective Loss 0.196696 LR 0.000500 Time 0.024901 -2022-12-06 11:11:09,790 - Epoch: [109][ 140/ 1200] Overall Loss 0.197026 Objective Loss 0.197026 LR 0.000500 Time 0.024484 -2022-12-06 11:11:09,982 - Epoch: [109][ 150/ 1200] Overall Loss 0.196662 Objective Loss 0.196662 LR 0.000500 Time 0.024126 -2022-12-06 11:11:10,174 - Epoch: [109][ 160/ 1200] Overall Loss 0.197598 Objective Loss 0.197598 LR 0.000500 Time 0.023816 -2022-12-06 11:11:10,366 - Epoch: [109][ 170/ 1200] Overall Loss 0.199250 Objective Loss 0.199250 LR 0.000500 Time 0.023540 -2022-12-06 11:11:10,558 - Epoch: [109][ 180/ 1200] Overall Loss 0.198579 Objective Loss 0.198579 LR 0.000500 Time 0.023299 -2022-12-06 11:11:10,750 - Epoch: [109][ 190/ 1200] Overall Loss 0.199257 Objective Loss 0.199257 LR 0.000500 Time 0.023078 -2022-12-06 11:11:10,942 - Epoch: [109][ 200/ 1200] Overall Loss 0.199750 Objective Loss 0.199750 LR 0.000500 Time 0.022882 -2022-12-06 11:11:11,134 - Epoch: [109][ 210/ 1200] Overall Loss 0.199995 Objective Loss 0.199995 LR 0.000500 Time 0.022706 -2022-12-06 11:11:11,326 - Epoch: [109][ 220/ 1200] Overall Loss 0.199761 Objective Loss 0.199761 LR 0.000500 Time 0.022542 -2022-12-06 11:11:11,518 - Epoch: [109][ 230/ 1200] Overall Loss 0.201042 Objective Loss 0.201042 LR 0.000500 Time 0.022394 -2022-12-06 11:11:11,710 - Epoch: [109][ 240/ 1200] Overall Loss 0.201059 Objective Loss 0.201059 LR 0.000500 Time 0.022261 -2022-12-06 11:11:11,902 - Epoch: [109][ 250/ 1200] Overall Loss 0.201084 Objective Loss 0.201084 LR 0.000500 Time 0.022137 -2022-12-06 11:11:12,094 - Epoch: [109][ 260/ 1200] Overall Loss 0.201120 Objective Loss 0.201120 LR 0.000500 Time 0.022022 -2022-12-06 11:11:12,287 - Epoch: [109][ 270/ 1200] Overall Loss 0.201410 Objective Loss 0.201410 LR 0.000500 Time 0.021916 -2022-12-06 11:11:12,479 - Epoch: [109][ 280/ 1200] Overall Loss 0.201211 Objective Loss 0.201211 LR 0.000500 Time 0.021820 -2022-12-06 11:11:12,671 - Epoch: [109][ 290/ 1200] Overall Loss 0.201941 Objective Loss 0.201941 LR 0.000500 Time 0.021727 -2022-12-06 11:11:12,864 - Epoch: [109][ 300/ 1200] Overall Loss 0.201680 Objective Loss 0.201680 LR 0.000500 Time 0.021643 -2022-12-06 11:11:13,056 - Epoch: [109][ 310/ 1200] Overall Loss 0.201804 Objective Loss 0.201804 LR 0.000500 Time 0.021563 -2022-12-06 11:11:13,248 - Epoch: [109][ 320/ 1200] Overall Loss 0.201267 Objective Loss 0.201267 LR 0.000500 Time 0.021488 -2022-12-06 11:11:13,441 - Epoch: [109][ 330/ 1200] Overall Loss 0.201177 Objective Loss 0.201177 LR 0.000500 Time 0.021419 -2022-12-06 11:11:13,633 - Epoch: [109][ 340/ 1200] Overall Loss 0.201497 Objective Loss 0.201497 LR 0.000500 Time 0.021353 -2022-12-06 11:11:13,825 - Epoch: [109][ 350/ 1200] Overall Loss 0.201396 Objective Loss 0.201396 LR 0.000500 Time 0.021290 -2022-12-06 11:11:14,018 - Epoch: [109][ 360/ 1200] Overall Loss 0.202316 Objective Loss 0.202316 LR 0.000500 Time 0.021232 -2022-12-06 11:11:14,210 - Epoch: [109][ 370/ 1200] Overall Loss 0.202740 Objective Loss 0.202740 LR 0.000500 Time 0.021177 -2022-12-06 11:11:14,402 - Epoch: [109][ 380/ 1200] Overall Loss 0.202504 Objective Loss 0.202504 LR 0.000500 Time 0.021123 -2022-12-06 11:11:14,594 - Epoch: [109][ 390/ 1200] Overall Loss 0.202975 Objective Loss 0.202975 LR 0.000500 Time 0.021072 -2022-12-06 11:11:14,787 - Epoch: [109][ 400/ 1200] Overall Loss 0.202425 Objective Loss 0.202425 LR 0.000500 Time 0.021025 -2022-12-06 11:11:14,979 - Epoch: [109][ 410/ 1200] Overall Loss 0.202907 Objective Loss 0.202907 LR 0.000500 Time 0.020981 -2022-12-06 11:11:15,172 - Epoch: [109][ 420/ 1200] Overall Loss 0.203025 Objective Loss 0.203025 LR 0.000500 Time 0.020938 -2022-12-06 11:11:15,364 - Epoch: [109][ 430/ 1200] Overall Loss 0.203320 Objective Loss 0.203320 LR 0.000500 Time 0.020898 -2022-12-06 11:11:15,556 - Epoch: [109][ 440/ 1200] Overall Loss 0.203496 Objective Loss 0.203496 LR 0.000500 Time 0.020858 -2022-12-06 11:11:15,749 - Epoch: [109][ 450/ 1200] Overall Loss 0.203411 Objective Loss 0.203411 LR 0.000500 Time 0.020821 -2022-12-06 11:11:15,941 - Epoch: [109][ 460/ 1200] Overall Loss 0.203240 Objective Loss 0.203240 LR 0.000500 Time 0.020785 -2022-12-06 11:11:16,134 - Epoch: [109][ 470/ 1200] Overall Loss 0.202959 Objective Loss 0.202959 LR 0.000500 Time 0.020752 -2022-12-06 11:11:16,326 - Epoch: [109][ 480/ 1200] Overall Loss 0.203442 Objective Loss 0.203442 LR 0.000500 Time 0.020718 -2022-12-06 11:11:16,519 - Epoch: [109][ 490/ 1200] Overall Loss 0.203413 Objective Loss 0.203413 LR 0.000500 Time 0.020688 -2022-12-06 11:11:16,711 - Epoch: [109][ 500/ 1200] Overall Loss 0.202734 Objective Loss 0.202734 LR 0.000500 Time 0.020658 -2022-12-06 11:11:16,904 - Epoch: [109][ 510/ 1200] Overall Loss 0.202667 Objective Loss 0.202667 LR 0.000500 Time 0.020630 -2022-12-06 11:11:17,096 - Epoch: [109][ 520/ 1200] Overall Loss 0.203040 Objective Loss 0.203040 LR 0.000500 Time 0.020602 -2022-12-06 11:11:17,289 - Epoch: [109][ 530/ 1200] Overall Loss 0.203692 Objective Loss 0.203692 LR 0.000500 Time 0.020575 -2022-12-06 11:11:17,480 - Epoch: [109][ 540/ 1200] Overall Loss 0.203621 Objective Loss 0.203621 LR 0.000500 Time 0.020548 -2022-12-06 11:11:17,673 - Epoch: [109][ 550/ 1200] Overall Loss 0.204145 Objective Loss 0.204145 LR 0.000500 Time 0.020523 -2022-12-06 11:11:17,864 - Epoch: [109][ 560/ 1200] Overall Loss 0.204341 Objective Loss 0.204341 LR 0.000500 Time 0.020498 -2022-12-06 11:11:18,057 - Epoch: [109][ 570/ 1200] Overall Loss 0.204470 Objective Loss 0.204470 LR 0.000500 Time 0.020476 -2022-12-06 11:11:18,249 - Epoch: [109][ 580/ 1200] Overall Loss 0.204486 Objective Loss 0.204486 LR 0.000500 Time 0.020453 -2022-12-06 11:11:18,441 - Epoch: [109][ 590/ 1200] Overall Loss 0.204552 Objective Loss 0.204552 LR 0.000500 Time 0.020430 -2022-12-06 11:11:18,633 - Epoch: [109][ 600/ 1200] Overall Loss 0.204366 Objective Loss 0.204366 LR 0.000500 Time 0.020410 -2022-12-06 11:11:18,826 - Epoch: [109][ 610/ 1200] Overall Loss 0.204446 Objective Loss 0.204446 LR 0.000500 Time 0.020389 -2022-12-06 11:11:19,017 - Epoch: [109][ 620/ 1200] Overall Loss 0.204283 Objective Loss 0.204283 LR 0.000500 Time 0.020368 -2022-12-06 11:11:19,209 - Epoch: [109][ 630/ 1200] Overall Loss 0.204278 Objective Loss 0.204278 LR 0.000500 Time 0.020349 -2022-12-06 11:11:19,402 - Epoch: [109][ 640/ 1200] Overall Loss 0.204988 Objective Loss 0.204988 LR 0.000500 Time 0.020331 -2022-12-06 11:11:19,594 - Epoch: [109][ 650/ 1200] Overall Loss 0.205054 Objective Loss 0.205054 LR 0.000500 Time 0.020313 -2022-12-06 11:11:19,786 - Epoch: [109][ 660/ 1200] Overall Loss 0.204906 Objective Loss 0.204906 LR 0.000500 Time 0.020296 -2022-12-06 11:11:19,979 - Epoch: [109][ 670/ 1200] Overall Loss 0.204786 Objective Loss 0.204786 LR 0.000500 Time 0.020279 -2022-12-06 11:11:20,170 - Epoch: [109][ 680/ 1200] Overall Loss 0.204803 Objective Loss 0.204803 LR 0.000500 Time 0.020262 -2022-12-06 11:11:20,363 - Epoch: [109][ 690/ 1200] Overall Loss 0.204654 Objective Loss 0.204654 LR 0.000500 Time 0.020247 -2022-12-06 11:11:20,555 - Epoch: [109][ 700/ 1200] Overall Loss 0.204796 Objective Loss 0.204796 LR 0.000500 Time 0.020231 -2022-12-06 11:11:20,747 - Epoch: [109][ 710/ 1200] Overall Loss 0.204763 Objective Loss 0.204763 LR 0.000500 Time 0.020216 -2022-12-06 11:11:20,939 - Epoch: [109][ 720/ 1200] Overall Loss 0.204398 Objective Loss 0.204398 LR 0.000500 Time 0.020201 -2022-12-06 11:11:21,131 - Epoch: [109][ 730/ 1200] Overall Loss 0.204473 Objective Loss 0.204473 LR 0.000500 Time 0.020187 -2022-12-06 11:11:21,324 - Epoch: [109][ 740/ 1200] Overall Loss 0.205246 Objective Loss 0.205246 LR 0.000500 Time 0.020174 -2022-12-06 11:11:21,516 - Epoch: [109][ 750/ 1200] Overall Loss 0.205321 Objective Loss 0.205321 LR 0.000500 Time 0.020160 -2022-12-06 11:11:21,708 - Epoch: [109][ 760/ 1200] Overall Loss 0.205337 Objective Loss 0.205337 LR 0.000500 Time 0.020147 -2022-12-06 11:11:21,900 - Epoch: [109][ 770/ 1200] Overall Loss 0.205571 Objective Loss 0.205571 LR 0.000500 Time 0.020134 -2022-12-06 11:11:22,092 - Epoch: [109][ 780/ 1200] Overall Loss 0.205535 Objective Loss 0.205535 LR 0.000500 Time 0.020121 -2022-12-06 11:11:22,284 - Epoch: [109][ 790/ 1200] Overall Loss 0.205622 Objective Loss 0.205622 LR 0.000500 Time 0.020109 -2022-12-06 11:11:22,476 - Epoch: [109][ 800/ 1200] Overall Loss 0.205895 Objective Loss 0.205895 LR 0.000500 Time 0.020097 -2022-12-06 11:11:22,669 - Epoch: [109][ 810/ 1200] Overall Loss 0.205952 Objective Loss 0.205952 LR 0.000500 Time 0.020086 -2022-12-06 11:11:22,861 - Epoch: [109][ 820/ 1200] Overall Loss 0.206069 Objective Loss 0.206069 LR 0.000500 Time 0.020074 -2022-12-06 11:11:23,053 - Epoch: [109][ 830/ 1200] Overall Loss 0.205940 Objective Loss 0.205940 LR 0.000500 Time 0.020064 -2022-12-06 11:11:23,245 - Epoch: [109][ 840/ 1200] Overall Loss 0.206012 Objective Loss 0.206012 LR 0.000500 Time 0.020053 -2022-12-06 11:11:23,438 - Epoch: [109][ 850/ 1200] Overall Loss 0.206035 Objective Loss 0.206035 LR 0.000500 Time 0.020043 -2022-12-06 11:11:23,630 - Epoch: [109][ 860/ 1200] Overall Loss 0.205940 Objective Loss 0.205940 LR 0.000500 Time 0.020032 -2022-12-06 11:11:23,822 - Epoch: [109][ 870/ 1200] Overall Loss 0.205888 Objective Loss 0.205888 LR 0.000500 Time 0.020022 -2022-12-06 11:11:24,014 - Epoch: [109][ 880/ 1200] Overall Loss 0.205983 Objective Loss 0.205983 LR 0.000500 Time 0.020013 -2022-12-06 11:11:24,207 - Epoch: [109][ 890/ 1200] Overall Loss 0.206286 Objective Loss 0.206286 LR 0.000500 Time 0.020003 -2022-12-06 11:11:24,399 - Epoch: [109][ 900/ 1200] Overall Loss 0.206446 Objective Loss 0.206446 LR 0.000500 Time 0.019994 -2022-12-06 11:11:24,590 - Epoch: [109][ 910/ 1200] Overall Loss 0.206684 Objective Loss 0.206684 LR 0.000500 Time 0.019984 -2022-12-06 11:11:24,782 - Epoch: [109][ 920/ 1200] Overall Loss 0.206801 Objective Loss 0.206801 LR 0.000500 Time 0.019975 -2022-12-06 11:11:24,974 - Epoch: [109][ 930/ 1200] Overall Loss 0.206869 Objective Loss 0.206869 LR 0.000500 Time 0.019966 -2022-12-06 11:11:25,166 - Epoch: [109][ 940/ 1200] Overall Loss 0.206911 Objective Loss 0.206911 LR 0.000500 Time 0.019957 -2022-12-06 11:11:25,358 - Epoch: [109][ 950/ 1200] Overall Loss 0.207269 Objective Loss 0.207269 LR 0.000500 Time 0.019949 -2022-12-06 11:11:25,551 - Epoch: [109][ 960/ 1200] Overall Loss 0.207318 Objective Loss 0.207318 LR 0.000500 Time 0.019941 -2022-12-06 11:11:25,743 - Epoch: [109][ 970/ 1200] Overall Loss 0.207471 Objective Loss 0.207471 LR 0.000500 Time 0.019933 -2022-12-06 11:11:25,935 - Epoch: [109][ 980/ 1200] Overall Loss 0.207386 Objective Loss 0.207386 LR 0.000500 Time 0.019925 -2022-12-06 11:11:26,127 - Epoch: [109][ 990/ 1200] Overall Loss 0.207330 Objective Loss 0.207330 LR 0.000500 Time 0.019917 -2022-12-06 11:11:26,319 - Epoch: [109][ 1000/ 1200] Overall Loss 0.207246 Objective Loss 0.207246 LR 0.000500 Time 0.019909 -2022-12-06 11:11:26,512 - Epoch: [109][ 1010/ 1200] Overall Loss 0.207367 Objective Loss 0.207367 LR 0.000500 Time 0.019903 -2022-12-06 11:11:26,704 - Epoch: [109][ 1020/ 1200] Overall Loss 0.207304 Objective Loss 0.207304 LR 0.000500 Time 0.019896 -2022-12-06 11:11:26,898 - Epoch: [109][ 1030/ 1200] Overall Loss 0.207298 Objective Loss 0.207298 LR 0.000500 Time 0.019889 -2022-12-06 11:11:27,090 - Epoch: [109][ 1040/ 1200] Overall Loss 0.207228 Objective Loss 0.207228 LR 0.000500 Time 0.019883 -2022-12-06 11:11:27,283 - Epoch: [109][ 1050/ 1200] Overall Loss 0.207085 Objective Loss 0.207085 LR 0.000500 Time 0.019876 -2022-12-06 11:11:27,475 - Epoch: [109][ 1060/ 1200] Overall Loss 0.207250 Objective Loss 0.207250 LR 0.000500 Time 0.019870 -2022-12-06 11:11:27,668 - Epoch: [109][ 1070/ 1200] Overall Loss 0.207061 Objective Loss 0.207061 LR 0.000500 Time 0.019864 -2022-12-06 11:11:27,860 - Epoch: [109][ 1080/ 1200] Overall Loss 0.207132 Objective Loss 0.207132 LR 0.000500 Time 0.019857 -2022-12-06 11:11:28,052 - Epoch: [109][ 1090/ 1200] Overall Loss 0.207392 Objective Loss 0.207392 LR 0.000500 Time 0.019851 -2022-12-06 11:11:28,244 - Epoch: [109][ 1100/ 1200] Overall Loss 0.207488 Objective Loss 0.207488 LR 0.000500 Time 0.019844 -2022-12-06 11:11:28,436 - Epoch: [109][ 1110/ 1200] Overall Loss 0.207667 Objective Loss 0.207667 LR 0.000500 Time 0.019838 -2022-12-06 11:11:28,628 - Epoch: [109][ 1120/ 1200] Overall Loss 0.207789 Objective Loss 0.207789 LR 0.000500 Time 0.019832 -2022-12-06 11:11:28,821 - Epoch: [109][ 1130/ 1200] Overall Loss 0.207846 Objective Loss 0.207846 LR 0.000500 Time 0.019827 -2022-12-06 11:11:29,012 - Epoch: [109][ 1140/ 1200] Overall Loss 0.207886 Objective Loss 0.207886 LR 0.000500 Time 0.019820 -2022-12-06 11:11:29,204 - Epoch: [109][ 1150/ 1200] Overall Loss 0.207837 Objective Loss 0.207837 LR 0.000500 Time 0.019814 -2022-12-06 11:11:29,397 - Epoch: [109][ 1160/ 1200] Overall Loss 0.207940 Objective Loss 0.207940 LR 0.000500 Time 0.019809 -2022-12-06 11:11:29,589 - Epoch: [109][ 1170/ 1200] Overall Loss 0.207951 Objective Loss 0.207951 LR 0.000500 Time 0.019803 -2022-12-06 11:11:29,782 - Epoch: [109][ 1180/ 1200] Overall Loss 0.208198 Objective Loss 0.208198 LR 0.000500 Time 0.019798 -2022-12-06 11:11:29,974 - Epoch: [109][ 1190/ 1200] Overall Loss 0.208353 Objective Loss 0.208353 LR 0.000500 Time 0.019793 -2022-12-06 11:11:30,199 - Epoch: [109][ 1200/ 1200] Overall Loss 0.208295 Objective Loss 0.208295 Top1 88.702929 Top5 99.372385 LR 0.000500 Time 0.019815 -2022-12-06 11:11:30,287 - --- validate (epoch=109)----------- -2022-12-06 11:11:30,287 - 34129 samples (256 per mini-batch) -2022-12-06 11:11:30,741 - Epoch: [109][ 10/ 134] Loss 0.249291 Top1 86.796875 Top5 98.203125 -2022-12-06 11:11:30,873 - Epoch: [109][ 20/ 134] Loss 0.249704 Top1 86.503906 Top5 98.320312 -2022-12-06 11:11:31,004 - Epoch: [109][ 30/ 134] Loss 0.250748 Top1 86.289062 Top5 98.229167 -2022-12-06 11:11:31,135 - Epoch: [109][ 40/ 134] Loss 0.258429 Top1 86.083984 Top5 98.066406 -2022-12-06 11:11:31,276 - Epoch: [109][ 50/ 134] Loss 0.263053 Top1 86.054688 Top5 98.046875 -2022-12-06 11:11:31,420 - Epoch: [109][ 60/ 134] Loss 0.258521 Top1 85.996094 Top5 98.020833 -2022-12-06 11:11:31,564 - Epoch: [109][ 70/ 134] Loss 0.258906 Top1 86.010045 Top5 98.074777 -2022-12-06 11:11:31,706 - Epoch: [109][ 80/ 134] Loss 0.260940 Top1 85.947266 Top5 98.066406 -2022-12-06 11:11:31,847 - Epoch: [109][ 90/ 134] Loss 0.264182 Top1 85.924479 Top5 98.038194 -2022-12-06 11:11:31,996 - Epoch: [109][ 100/ 134] Loss 0.264937 Top1 85.867188 Top5 98.042969 -2022-12-06 11:11:32,140 - Epoch: [109][ 110/ 134] Loss 0.265086 Top1 85.877131 Top5 98.082386 -2022-12-06 11:11:32,277 - Epoch: [109][ 120/ 134] Loss 0.266611 Top1 85.921224 Top5 98.105469 -2022-12-06 11:11:32,407 - Epoch: [109][ 130/ 134] Loss 0.263919 Top1 85.907452 Top5 98.134014 -2022-12-06 11:11:32,445 - Epoch: [109][ 134/ 134] Loss 0.262593 Top1 85.929854 Top5 98.118902 -2022-12-06 11:11:32,560 - ==> Top1: 85.930 Top5: 98.119 Loss: 0.263 - -2022-12-06 11:11:32,560 - ==> Confusion: -[[ 922 1 2 4 8 7 0 1 4 34 0 3 2 1 3 0 1 0 1 0 2] - [ 1 951 0 3 10 12 5 13 4 0 5 5 0 0 0 1 2 1 8 2 4] - [ 5 2 1006 12 4 2 21 12 0 5 5 2 2 1 4 3 2 1 3 3 8] - [ 4 3 17 941 1 2 0 1 0 1 11 0 6 2 14 0 2 2 8 0 5] - [ 7 5 2 0 963 2 0 0 1 6 2 4 1 2 8 7 4 3 0 1 2] - [ 4 20 2 3 5 969 2 18 1 1 1 9 6 13 3 1 1 1 0 4 5] - [ 1 3 8 0 0 1 1072 3 1 1 3 1 2 2 0 5 1 1 2 8 3] - [ 1 15 3 2 1 31 9 937 0 0 6 5 1 1 0 1 0 1 20 11 9] - [ 6 3 0 0 0 2 0 1 986 35 5 0 3 6 11 1 1 1 1 1 1] - [ 71 0 0 0 5 2 0 2 24 871 1 2 0 13 4 1 0 0 0 0 5] - [ 1 0 1 3 2 1 0 5 12 1 967 3 4 8 3 0 1 0 2 0 5] - [ 3 1 0 0 0 8 6 3 1 0 2 982 23 2 0 6 5 5 0 3 1] - [ 2 1 2 0 0 3 1 0 0 0 0 33 893 1 0 10 0 14 1 2 6] - [ 1 1 1 1 0 8 0 3 11 16 5 6 3 949 0 3 4 2 0 1 8] - [ 5 1 1 8 5 3 0 0 17 8 1 1 2 4 1061 0 1 1 7 1 3] - [ 0 0 1 1 2 2 2 0 0 0 1 7 8 1 0 994 6 12 0 1 5] - [ 3 1 0 0 3 1 2 0 1 0 0 1 3 1 3 11 1032 2 0 4 4] - [ 3 0 0 4 0 1 2 2 1 1 1 7 14 4 1 11 1 982 0 0 1] - [ 7 4 4 8 2 2 0 19 1 0 8 2 3 1 12 1 1 1 930 1 1] - [ 4 3 0 1 1 4 6 9 0 0 1 21 9 7 0 5 3 2 0 996 8] - [ 154 260 175 117 131 161 85 131 105 94 217 114 380 273 166 116 195 92 141 200 9919]] - -2022-12-06 11:11:33,223 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:11:33,224 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:11:33,229 - - -2022-12-06 11:11:33,230 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:11:34,207 - Epoch: [110][ 10/ 1200] Overall Loss 0.226125 Objective Loss 0.226125 LR 0.000500 Time 0.097644 -2022-12-06 11:11:34,400 - Epoch: [110][ 20/ 1200] Overall Loss 0.216341 Objective Loss 0.216341 LR 0.000500 Time 0.058478 -2022-12-06 11:11:34,592 - Epoch: [110][ 30/ 1200] Overall Loss 0.219250 Objective Loss 0.219250 LR 0.000500 Time 0.045344 -2022-12-06 11:11:34,783 - Epoch: [110][ 40/ 1200] Overall Loss 0.219312 Objective Loss 0.219312 LR 0.000500 Time 0.038784 -2022-12-06 11:11:34,974 - Epoch: [110][ 50/ 1200] Overall Loss 0.214717 Objective Loss 0.214717 LR 0.000500 Time 0.034831 -2022-12-06 11:11:35,165 - Epoch: [110][ 60/ 1200] Overall Loss 0.207539 Objective Loss 0.207539 LR 0.000500 Time 0.032207 -2022-12-06 11:11:35,357 - Epoch: [110][ 70/ 1200] Overall Loss 0.204863 Objective Loss 0.204863 LR 0.000500 Time 0.030328 -2022-12-06 11:11:35,548 - Epoch: [110][ 80/ 1200] Overall Loss 0.204207 Objective Loss 0.204207 LR 0.000500 Time 0.028925 -2022-12-06 11:11:35,739 - Epoch: [110][ 90/ 1200] Overall Loss 0.204852 Objective Loss 0.204852 LR 0.000500 Time 0.027830 -2022-12-06 11:11:35,931 - Epoch: [110][ 100/ 1200] Overall Loss 0.204135 Objective Loss 0.204135 LR 0.000500 Time 0.026960 -2022-12-06 11:11:36,122 - Epoch: [110][ 110/ 1200] Overall Loss 0.203300 Objective Loss 0.203300 LR 0.000500 Time 0.026243 -2022-12-06 11:11:36,314 - Epoch: [110][ 120/ 1200] Overall Loss 0.202734 Objective Loss 0.202734 LR 0.000500 Time 0.025646 -2022-12-06 11:11:36,506 - Epoch: [110][ 130/ 1200] Overall Loss 0.203205 Objective Loss 0.203205 LR 0.000500 Time 0.025145 -2022-12-06 11:11:36,698 - Epoch: [110][ 140/ 1200] Overall Loss 0.201793 Objective Loss 0.201793 LR 0.000500 Time 0.024718 -2022-12-06 11:11:36,890 - Epoch: [110][ 150/ 1200] Overall Loss 0.202250 Objective Loss 0.202250 LR 0.000500 Time 0.024348 -2022-12-06 11:11:37,081 - Epoch: [110][ 160/ 1200] Overall Loss 0.202978 Objective Loss 0.202978 LR 0.000500 Time 0.024019 -2022-12-06 11:11:37,273 - Epoch: [110][ 170/ 1200] Overall Loss 0.202458 Objective Loss 0.202458 LR 0.000500 Time 0.023728 -2022-12-06 11:11:37,464 - Epoch: [110][ 180/ 1200] Overall Loss 0.202489 Objective Loss 0.202489 LR 0.000500 Time 0.023471 -2022-12-06 11:11:37,656 - Epoch: [110][ 190/ 1200] Overall Loss 0.202484 Objective Loss 0.202484 LR 0.000500 Time 0.023240 -2022-12-06 11:11:37,847 - Epoch: [110][ 200/ 1200] Overall Loss 0.203949 Objective Loss 0.203949 LR 0.000500 Time 0.023033 -2022-12-06 11:11:38,039 - Epoch: [110][ 210/ 1200] Overall Loss 0.202775 Objective Loss 0.202775 LR 0.000500 Time 0.022845 -2022-12-06 11:11:38,230 - Epoch: [110][ 220/ 1200] Overall Loss 0.201989 Objective Loss 0.201989 LR 0.000500 Time 0.022676 -2022-12-06 11:11:38,422 - Epoch: [110][ 230/ 1200] Overall Loss 0.202215 Objective Loss 0.202215 LR 0.000500 Time 0.022522 -2022-12-06 11:11:38,614 - Epoch: [110][ 240/ 1200] Overall Loss 0.202976 Objective Loss 0.202976 LR 0.000500 Time 0.022380 -2022-12-06 11:11:38,806 - Epoch: [110][ 250/ 1200] Overall Loss 0.202361 Objective Loss 0.202361 LR 0.000500 Time 0.022250 -2022-12-06 11:11:38,997 - Epoch: [110][ 260/ 1200] Overall Loss 0.202403 Objective Loss 0.202403 LR 0.000500 Time 0.022128 -2022-12-06 11:11:39,188 - Epoch: [110][ 270/ 1200] Overall Loss 0.203141 Objective Loss 0.203141 LR 0.000500 Time 0.022013 -2022-12-06 11:11:39,380 - Epoch: [110][ 280/ 1200] Overall Loss 0.203385 Objective Loss 0.203385 LR 0.000500 Time 0.021911 -2022-12-06 11:11:39,571 - Epoch: [110][ 290/ 1200] Overall Loss 0.202898 Objective Loss 0.202898 LR 0.000500 Time 0.021813 -2022-12-06 11:11:39,763 - Epoch: [110][ 300/ 1200] Overall Loss 0.202950 Objective Loss 0.202950 LR 0.000500 Time 0.021723 -2022-12-06 11:11:39,954 - Epoch: [110][ 310/ 1200] Overall Loss 0.203876 Objective Loss 0.203876 LR 0.000500 Time 0.021638 -2022-12-06 11:11:40,146 - Epoch: [110][ 320/ 1200] Overall Loss 0.204173 Objective Loss 0.204173 LR 0.000500 Time 0.021560 -2022-12-06 11:11:40,338 - Epoch: [110][ 330/ 1200] Overall Loss 0.204458 Objective Loss 0.204458 LR 0.000500 Time 0.021486 -2022-12-06 11:11:40,530 - Epoch: [110][ 340/ 1200] Overall Loss 0.204552 Objective Loss 0.204552 LR 0.000500 Time 0.021417 -2022-12-06 11:11:40,722 - Epoch: [110][ 350/ 1200] Overall Loss 0.204116 Objective Loss 0.204116 LR 0.000500 Time 0.021352 -2022-12-06 11:11:40,914 - Epoch: [110][ 360/ 1200] Overall Loss 0.204291 Objective Loss 0.204291 LR 0.000500 Time 0.021289 -2022-12-06 11:11:41,104 - Epoch: [110][ 370/ 1200] Overall Loss 0.204356 Objective Loss 0.204356 LR 0.000500 Time 0.021228 -2022-12-06 11:11:41,296 - Epoch: [110][ 380/ 1200] Overall Loss 0.204428 Objective Loss 0.204428 LR 0.000500 Time 0.021172 -2022-12-06 11:11:41,487 - Epoch: [110][ 390/ 1200] Overall Loss 0.204777 Objective Loss 0.204777 LR 0.000500 Time 0.021119 -2022-12-06 11:11:41,678 - Epoch: [110][ 400/ 1200] Overall Loss 0.204723 Objective Loss 0.204723 LR 0.000500 Time 0.021067 -2022-12-06 11:11:41,870 - Epoch: [110][ 410/ 1200] Overall Loss 0.204217 Objective Loss 0.204217 LR 0.000500 Time 0.021019 -2022-12-06 11:11:42,062 - Epoch: [110][ 420/ 1200] Overall Loss 0.204040 Objective Loss 0.204040 LR 0.000500 Time 0.020974 -2022-12-06 11:11:42,254 - Epoch: [110][ 430/ 1200] Overall Loss 0.204164 Objective Loss 0.204164 LR 0.000500 Time 0.020932 -2022-12-06 11:11:42,446 - Epoch: [110][ 440/ 1200] Overall Loss 0.204161 Objective Loss 0.204161 LR 0.000500 Time 0.020891 -2022-12-06 11:11:42,637 - Epoch: [110][ 450/ 1200] Overall Loss 0.203833 Objective Loss 0.203833 LR 0.000500 Time 0.020851 -2022-12-06 11:11:42,829 - Epoch: [110][ 460/ 1200] Overall Loss 0.203975 Objective Loss 0.203975 LR 0.000500 Time 0.020814 -2022-12-06 11:11:43,020 - Epoch: [110][ 470/ 1200] Overall Loss 0.203548 Objective Loss 0.203548 LR 0.000500 Time 0.020776 -2022-12-06 11:11:43,212 - Epoch: [110][ 480/ 1200] Overall Loss 0.203959 Objective Loss 0.203959 LR 0.000500 Time 0.020741 -2022-12-06 11:11:43,403 - Epoch: [110][ 490/ 1200] Overall Loss 0.203655 Objective Loss 0.203655 LR 0.000500 Time 0.020706 -2022-12-06 11:11:43,594 - Epoch: [110][ 500/ 1200] Overall Loss 0.204073 Objective Loss 0.204073 LR 0.000500 Time 0.020673 -2022-12-06 11:11:43,785 - Epoch: [110][ 510/ 1200] Overall Loss 0.204693 Objective Loss 0.204693 LR 0.000500 Time 0.020642 -2022-12-06 11:11:43,977 - Epoch: [110][ 520/ 1200] Overall Loss 0.204577 Objective Loss 0.204577 LR 0.000500 Time 0.020613 -2022-12-06 11:11:44,168 - Epoch: [110][ 530/ 1200] Overall Loss 0.204977 Objective Loss 0.204977 LR 0.000500 Time 0.020584 -2022-12-06 11:11:44,360 - Epoch: [110][ 540/ 1200] Overall Loss 0.204876 Objective Loss 0.204876 LR 0.000500 Time 0.020557 -2022-12-06 11:11:44,551 - Epoch: [110][ 550/ 1200] Overall Loss 0.205058 Objective Loss 0.205058 LR 0.000500 Time 0.020530 -2022-12-06 11:11:44,743 - Epoch: [110][ 560/ 1200] Overall Loss 0.205087 Objective Loss 0.205087 LR 0.000500 Time 0.020505 -2022-12-06 11:11:44,935 - Epoch: [110][ 570/ 1200] Overall Loss 0.204344 Objective Loss 0.204344 LR 0.000500 Time 0.020481 -2022-12-06 11:11:45,126 - Epoch: [110][ 580/ 1200] Overall Loss 0.204393 Objective Loss 0.204393 LR 0.000500 Time 0.020457 -2022-12-06 11:11:45,318 - Epoch: [110][ 590/ 1200] Overall Loss 0.204538 Objective Loss 0.204538 LR 0.000500 Time 0.020433 -2022-12-06 11:11:45,509 - Epoch: [110][ 600/ 1200] Overall Loss 0.204341 Objective Loss 0.204341 LR 0.000500 Time 0.020410 -2022-12-06 11:11:45,699 - Epoch: [110][ 610/ 1200] Overall Loss 0.204349 Objective Loss 0.204349 LR 0.000500 Time 0.020387 -2022-12-06 11:11:45,890 - Epoch: [110][ 620/ 1200] Overall Loss 0.204061 Objective Loss 0.204061 LR 0.000500 Time 0.020365 -2022-12-06 11:11:46,080 - Epoch: [110][ 630/ 1200] Overall Loss 0.204138 Objective Loss 0.204138 LR 0.000500 Time 0.020343 -2022-12-06 11:11:46,271 - Epoch: [110][ 640/ 1200] Overall Loss 0.204271 Objective Loss 0.204271 LR 0.000500 Time 0.020323 -2022-12-06 11:11:46,462 - Epoch: [110][ 650/ 1200] Overall Loss 0.204122 Objective Loss 0.204122 LR 0.000500 Time 0.020303 -2022-12-06 11:11:46,654 - Epoch: [110][ 660/ 1200] Overall Loss 0.204187 Objective Loss 0.204187 LR 0.000500 Time 0.020285 -2022-12-06 11:11:46,845 - Epoch: [110][ 670/ 1200] Overall Loss 0.204041 Objective Loss 0.204041 LR 0.000500 Time 0.020267 -2022-12-06 11:11:47,036 - Epoch: [110][ 680/ 1200] Overall Loss 0.204090 Objective Loss 0.204090 LR 0.000500 Time 0.020249 -2022-12-06 11:11:47,228 - Epoch: [110][ 690/ 1200] Overall Loss 0.204098 Objective Loss 0.204098 LR 0.000500 Time 0.020232 -2022-12-06 11:11:47,419 - Epoch: [110][ 700/ 1200] Overall Loss 0.204019 Objective Loss 0.204019 LR 0.000500 Time 0.020215 -2022-12-06 11:11:47,610 - Epoch: [110][ 710/ 1200] Overall Loss 0.204250 Objective Loss 0.204250 LR 0.000500 Time 0.020199 -2022-12-06 11:11:47,801 - Epoch: [110][ 720/ 1200] Overall Loss 0.204154 Objective Loss 0.204154 LR 0.000500 Time 0.020184 -2022-12-06 11:11:47,993 - Epoch: [110][ 730/ 1200] Overall Loss 0.204317 Objective Loss 0.204317 LR 0.000500 Time 0.020168 -2022-12-06 11:11:48,184 - Epoch: [110][ 740/ 1200] Overall Loss 0.204148 Objective Loss 0.204148 LR 0.000500 Time 0.020154 -2022-12-06 11:11:48,375 - Epoch: [110][ 750/ 1200] Overall Loss 0.204133 Objective Loss 0.204133 LR 0.000500 Time 0.020139 -2022-12-06 11:11:48,567 - Epoch: [110][ 760/ 1200] Overall Loss 0.204103 Objective Loss 0.204103 LR 0.000500 Time 0.020125 -2022-12-06 11:11:48,757 - Epoch: [110][ 770/ 1200] Overall Loss 0.204394 Objective Loss 0.204394 LR 0.000500 Time 0.020111 -2022-12-06 11:11:48,948 - Epoch: [110][ 780/ 1200] Overall Loss 0.204492 Objective Loss 0.204492 LR 0.000500 Time 0.020097 -2022-12-06 11:11:49,139 - Epoch: [110][ 790/ 1200] Overall Loss 0.204458 Objective Loss 0.204458 LR 0.000500 Time 0.020084 -2022-12-06 11:11:49,330 - Epoch: [110][ 800/ 1200] Overall Loss 0.204449 Objective Loss 0.204449 LR 0.000500 Time 0.020071 -2022-12-06 11:11:49,522 - Epoch: [110][ 810/ 1200] Overall Loss 0.204329 Objective Loss 0.204329 LR 0.000500 Time 0.020059 -2022-12-06 11:11:49,713 - Epoch: [110][ 820/ 1200] Overall Loss 0.204443 Objective Loss 0.204443 LR 0.000500 Time 0.020048 -2022-12-06 11:11:49,905 - Epoch: [110][ 830/ 1200] Overall Loss 0.204237 Objective Loss 0.204237 LR 0.000500 Time 0.020036 -2022-12-06 11:11:50,097 - Epoch: [110][ 840/ 1200] Overall Loss 0.204397 Objective Loss 0.204397 LR 0.000500 Time 0.020026 -2022-12-06 11:11:50,288 - Epoch: [110][ 850/ 1200] Overall Loss 0.204331 Objective Loss 0.204331 LR 0.000500 Time 0.020014 -2022-12-06 11:11:50,480 - Epoch: [110][ 860/ 1200] Overall Loss 0.204663 Objective Loss 0.204663 LR 0.000500 Time 0.020004 -2022-12-06 11:11:50,671 - Epoch: [110][ 870/ 1200] Overall Loss 0.204605 Objective Loss 0.204605 LR 0.000500 Time 0.019993 -2022-12-06 11:11:50,863 - Epoch: [110][ 880/ 1200] Overall Loss 0.204461 Objective Loss 0.204461 LR 0.000500 Time 0.019983 -2022-12-06 11:11:51,054 - Epoch: [110][ 890/ 1200] Overall Loss 0.204629 Objective Loss 0.204629 LR 0.000500 Time 0.019973 -2022-12-06 11:11:51,246 - Epoch: [110][ 900/ 1200] Overall Loss 0.204675 Objective Loss 0.204675 LR 0.000500 Time 0.019964 -2022-12-06 11:11:51,438 - Epoch: [110][ 910/ 1200] Overall Loss 0.204956 Objective Loss 0.204956 LR 0.000500 Time 0.019954 -2022-12-06 11:11:51,630 - Epoch: [110][ 920/ 1200] Overall Loss 0.204970 Objective Loss 0.204970 LR 0.000500 Time 0.019946 -2022-12-06 11:11:51,822 - Epoch: [110][ 930/ 1200] Overall Loss 0.205108 Objective Loss 0.205108 LR 0.000500 Time 0.019937 -2022-12-06 11:11:52,013 - Epoch: [110][ 940/ 1200] Overall Loss 0.205366 Objective Loss 0.205366 LR 0.000500 Time 0.019928 -2022-12-06 11:11:52,204 - Epoch: [110][ 950/ 1200] Overall Loss 0.205714 Objective Loss 0.205714 LR 0.000500 Time 0.019918 -2022-12-06 11:11:52,395 - Epoch: [110][ 960/ 1200] Overall Loss 0.205872 Objective Loss 0.205872 LR 0.000500 Time 0.019910 -2022-12-06 11:11:52,586 - Epoch: [110][ 970/ 1200] Overall Loss 0.205953 Objective Loss 0.205953 LR 0.000500 Time 0.019901 -2022-12-06 11:11:52,778 - Epoch: [110][ 980/ 1200] Overall Loss 0.205968 Objective Loss 0.205968 LR 0.000500 Time 0.019893 -2022-12-06 11:11:52,970 - Epoch: [110][ 990/ 1200] Overall Loss 0.206017 Objective Loss 0.206017 LR 0.000500 Time 0.019885 -2022-12-06 11:11:53,162 - Epoch: [110][ 1000/ 1200] Overall Loss 0.206119 Objective Loss 0.206119 LR 0.000500 Time 0.019878 -2022-12-06 11:11:53,353 - Epoch: [110][ 1010/ 1200] Overall Loss 0.206319 Objective Loss 0.206319 LR 0.000500 Time 0.019870 -2022-12-06 11:11:53,545 - Epoch: [110][ 1020/ 1200] Overall Loss 0.206446 Objective Loss 0.206446 LR 0.000500 Time 0.019862 -2022-12-06 11:11:53,736 - Epoch: [110][ 1030/ 1200] Overall Loss 0.206409 Objective Loss 0.206409 LR 0.000500 Time 0.019854 -2022-12-06 11:11:53,928 - Epoch: [110][ 1040/ 1200] Overall Loss 0.206282 Objective Loss 0.206282 LR 0.000500 Time 0.019847 -2022-12-06 11:11:54,119 - Epoch: [110][ 1050/ 1200] Overall Loss 0.206106 Objective Loss 0.206106 LR 0.000500 Time 0.019840 -2022-12-06 11:11:54,311 - Epoch: [110][ 1060/ 1200] Overall Loss 0.205946 Objective Loss 0.205946 LR 0.000500 Time 0.019833 -2022-12-06 11:11:54,502 - Epoch: [110][ 1070/ 1200] Overall Loss 0.205799 Objective Loss 0.205799 LR 0.000500 Time 0.019826 -2022-12-06 11:11:54,693 - Epoch: [110][ 1080/ 1200] Overall Loss 0.205604 Objective Loss 0.205604 LR 0.000500 Time 0.019819 -2022-12-06 11:11:54,885 - Epoch: [110][ 1090/ 1200] Overall Loss 0.205564 Objective Loss 0.205564 LR 0.000500 Time 0.019812 -2022-12-06 11:11:55,076 - Epoch: [110][ 1100/ 1200] Overall Loss 0.205268 Objective Loss 0.205268 LR 0.000500 Time 0.019806 -2022-12-06 11:11:55,268 - Epoch: [110][ 1110/ 1200] Overall Loss 0.205207 Objective Loss 0.205207 LR 0.000500 Time 0.019800 -2022-12-06 11:11:55,460 - Epoch: [110][ 1120/ 1200] Overall Loss 0.205164 Objective Loss 0.205164 LR 0.000500 Time 0.019794 -2022-12-06 11:11:55,652 - Epoch: [110][ 1130/ 1200] Overall Loss 0.205161 Objective Loss 0.205161 LR 0.000500 Time 0.019788 -2022-12-06 11:11:55,844 - Epoch: [110][ 1140/ 1200] Overall Loss 0.205286 Objective Loss 0.205286 LR 0.000500 Time 0.019782 -2022-12-06 11:11:56,035 - Epoch: [110][ 1150/ 1200] Overall Loss 0.205292 Objective Loss 0.205292 LR 0.000500 Time 0.019776 -2022-12-06 11:11:56,227 - Epoch: [110][ 1160/ 1200] Overall Loss 0.205153 Objective Loss 0.205153 LR 0.000500 Time 0.019771 -2022-12-06 11:11:56,418 - Epoch: [110][ 1170/ 1200] Overall Loss 0.205354 Objective Loss 0.205354 LR 0.000500 Time 0.019765 -2022-12-06 11:11:56,610 - Epoch: [110][ 1180/ 1200] Overall Loss 0.205459 Objective Loss 0.205459 LR 0.000500 Time 0.019759 -2022-12-06 11:11:56,801 - Epoch: [110][ 1190/ 1200] Overall Loss 0.205632 Objective Loss 0.205632 LR 0.000500 Time 0.019753 -2022-12-06 11:11:57,024 - Epoch: [110][ 1200/ 1200] Overall Loss 0.205785 Objective Loss 0.205785 Top1 88.912134 Top5 98.117155 LR 0.000500 Time 0.019774 -2022-12-06 11:11:57,128 - --- validate (epoch=110)----------- -2022-12-06 11:11:57,128 - 34129 samples (256 per mini-batch) -2022-12-06 11:11:57,571 - Epoch: [110][ 10/ 134] Loss 0.229339 Top1 85.820312 Top5 98.515625 -2022-12-06 11:11:57,701 - Epoch: [110][ 20/ 134] Loss 0.265827 Top1 85.722656 Top5 98.359375 -2022-12-06 11:11:57,825 - Epoch: [110][ 30/ 134] Loss 0.285609 Top1 85.729167 Top5 98.125000 -2022-12-06 11:11:57,952 - Epoch: [110][ 40/ 134] Loss 0.277127 Top1 85.898438 Top5 98.173828 -2022-12-06 11:11:58,076 - Epoch: [110][ 50/ 134] Loss 0.269849 Top1 86.304688 Top5 98.132812 -2022-12-06 11:11:58,203 - Epoch: [110][ 60/ 134] Loss 0.263674 Top1 86.399740 Top5 98.222656 -2022-12-06 11:11:58,334 - Epoch: [110][ 70/ 134] Loss 0.261356 Top1 86.205357 Top5 98.219866 -2022-12-06 11:11:58,463 - Epoch: [110][ 80/ 134] Loss 0.260957 Top1 86.230469 Top5 98.198242 -2022-12-06 11:11:58,593 - Epoch: [110][ 90/ 134] Loss 0.262945 Top1 86.241319 Top5 98.198785 -2022-12-06 11:11:58,722 - Epoch: [110][ 100/ 134] Loss 0.262464 Top1 86.214844 Top5 98.195312 -2022-12-06 11:11:58,851 - Epoch: [110][ 110/ 134] Loss 0.263754 Top1 86.189631 Top5 98.153409 -2022-12-06 11:11:58,980 - Epoch: [110][ 120/ 134] Loss 0.264077 Top1 86.230469 Top5 98.186849 -2022-12-06 11:11:59,109 - Epoch: [110][ 130/ 134] Loss 0.263826 Top1 86.219952 Top5 98.176082 -2022-12-06 11:11:59,146 - Epoch: [110][ 134/ 134] Loss 0.262158 Top1 86.219930 Top5 98.177503 -2022-12-06 11:11:59,247 - ==> Top1: 86.220 Top5: 98.178 Loss: 0.262 - -2022-12-06 11:11:59,247 - ==> Confusion: -[[ 916 1 2 3 3 9 1 1 4 42 0 3 0 1 5 1 1 1 0 0 2] - [ 1 932 2 2 8 22 3 20 2 1 2 5 0 0 2 1 0 1 12 3 8] - [ 4 2 1011 8 4 2 17 13 0 4 3 7 2 1 3 2 2 1 4 2 11] - [ 4 0 11 951 0 1 1 1 0 1 10 0 6 2 8 0 1 3 17 0 3] - [ 13 4 1 0 957 10 1 1 1 7 3 2 0 1 5 5 3 2 0 1 3] - [ 5 10 0 2 3 999 2 14 2 2 1 7 5 8 3 1 1 0 1 1 2] - [ 2 4 9 3 1 0 1070 10 0 1 1 2 0 3 0 6 0 0 0 3 3] - [ 0 5 9 3 1 25 4 966 1 0 2 8 1 0 0 1 0 0 18 7 3] - [ 4 2 0 0 1 3 0 2 971 41 8 1 1 8 8 1 2 1 5 2 3] - [ 74 0 1 0 11 3 0 2 16 867 1 2 0 12 2 1 0 1 1 0 7] - [ 1 2 4 3 1 2 2 3 7 1 957 4 2 11 2 0 2 0 6 0 9] - [ 5 1 3 0 1 10 3 3 0 0 1 980 14 1 0 6 2 4 0 13 4] - [ 1 0 0 2 1 3 2 1 0 0 0 34 895 1 2 7 2 5 3 6 4] - [ 0 0 2 0 0 11 0 3 8 12 3 6 5 964 1 2 0 0 0 2 4] - [ 7 3 2 21 2 2 0 0 20 4 1 2 3 3 1039 1 0 0 13 1 6] - [ 1 0 1 2 1 3 1 0 0 0 0 12 4 2 0 995 6 9 0 3 3] - [ 2 1 1 1 2 2 0 0 0 0 0 5 1 4 1 13 1023 0 0 7 9] - [ 2 0 1 2 0 2 2 1 0 1 0 11 18 4 1 11 1 974 1 2 2] - [ 3 4 2 7 3 5 1 26 1 1 3 2 0 0 4 0 0 0 941 2 3] - [ 4 2 2 0 0 6 5 16 0 0 1 13 5 8 0 4 3 3 3 1003 2] - [ 151 187 220 126 96 239 85 167 71 76 129 133 350 294 129 123 145 69 197 228 10011]] - -2022-12-06 11:11:59,913 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:11:59,913 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:11:59,919 - - -2022-12-06 11:11:59,919 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:12:00,846 - Epoch: [111][ 10/ 1200] Overall Loss 0.203216 Objective Loss 0.203216 LR 0.000500 Time 0.092581 -2022-12-06 11:12:01,040 - Epoch: [111][ 20/ 1200] Overall Loss 0.210447 Objective Loss 0.210447 LR 0.000500 Time 0.055946 -2022-12-06 11:12:01,231 - Epoch: [111][ 30/ 1200] Overall Loss 0.207148 Objective Loss 0.207148 LR 0.000500 Time 0.043649 -2022-12-06 11:12:01,422 - Epoch: [111][ 40/ 1200] Overall Loss 0.207795 Objective Loss 0.207795 LR 0.000500 Time 0.037497 -2022-12-06 11:12:01,612 - Epoch: [111][ 50/ 1200] Overall Loss 0.206413 Objective Loss 0.206413 LR 0.000500 Time 0.033793 -2022-12-06 11:12:01,802 - Epoch: [111][ 60/ 1200] Overall Loss 0.206439 Objective Loss 0.206439 LR 0.000500 Time 0.031321 -2022-12-06 11:12:01,993 - Epoch: [111][ 70/ 1200] Overall Loss 0.209929 Objective Loss 0.209929 LR 0.000500 Time 0.029559 -2022-12-06 11:12:02,183 - Epoch: [111][ 80/ 1200] Overall Loss 0.205921 Objective Loss 0.205921 LR 0.000500 Time 0.028237 -2022-12-06 11:12:02,373 - Epoch: [111][ 90/ 1200] Overall Loss 0.205243 Objective Loss 0.205243 LR 0.000500 Time 0.027206 -2022-12-06 11:12:02,564 - Epoch: [111][ 100/ 1200] Overall Loss 0.204066 Objective Loss 0.204066 LR 0.000500 Time 0.026387 -2022-12-06 11:12:02,754 - Epoch: [111][ 110/ 1200] Overall Loss 0.202577 Objective Loss 0.202577 LR 0.000500 Time 0.025711 -2022-12-06 11:12:02,944 - Epoch: [111][ 120/ 1200] Overall Loss 0.201517 Objective Loss 0.201517 LR 0.000500 Time 0.025153 -2022-12-06 11:12:03,135 - Epoch: [111][ 130/ 1200] Overall Loss 0.200529 Objective Loss 0.200529 LR 0.000500 Time 0.024679 -2022-12-06 11:12:03,325 - Epoch: [111][ 140/ 1200] Overall Loss 0.201599 Objective Loss 0.201599 LR 0.000500 Time 0.024270 -2022-12-06 11:12:03,515 - Epoch: [111][ 150/ 1200] Overall Loss 0.202908 Objective Loss 0.202908 LR 0.000500 Time 0.023915 -2022-12-06 11:12:03,705 - Epoch: [111][ 160/ 1200] Overall Loss 0.201818 Objective Loss 0.201818 LR 0.000500 Time 0.023606 -2022-12-06 11:12:03,896 - Epoch: [111][ 170/ 1200] Overall Loss 0.201323 Objective Loss 0.201323 LR 0.000500 Time 0.023334 -2022-12-06 11:12:04,087 - Epoch: [111][ 180/ 1200] Overall Loss 0.201977 Objective Loss 0.201977 LR 0.000500 Time 0.023096 -2022-12-06 11:12:04,277 - Epoch: [111][ 190/ 1200] Overall Loss 0.202928 Objective Loss 0.202928 LR 0.000500 Time 0.022878 -2022-12-06 11:12:04,468 - Epoch: [111][ 200/ 1200] Overall Loss 0.202493 Objective Loss 0.202493 LR 0.000500 Time 0.022685 -2022-12-06 11:12:04,657 - Epoch: [111][ 210/ 1200] Overall Loss 0.203176 Objective Loss 0.203176 LR 0.000500 Time 0.022507 -2022-12-06 11:12:04,848 - Epoch: [111][ 220/ 1200] Overall Loss 0.203755 Objective Loss 0.203755 LR 0.000500 Time 0.022348 -2022-12-06 11:12:05,038 - Epoch: [111][ 230/ 1200] Overall Loss 0.204273 Objective Loss 0.204273 LR 0.000500 Time 0.022201 -2022-12-06 11:12:05,229 - Epoch: [111][ 240/ 1200] Overall Loss 0.203228 Objective Loss 0.203228 LR 0.000500 Time 0.022068 -2022-12-06 11:12:05,419 - Epoch: [111][ 250/ 1200] Overall Loss 0.203414 Objective Loss 0.203414 LR 0.000500 Time 0.021944 -2022-12-06 11:12:05,610 - Epoch: [111][ 260/ 1200] Overall Loss 0.203888 Objective Loss 0.203888 LR 0.000500 Time 0.021832 -2022-12-06 11:12:05,800 - Epoch: [111][ 270/ 1200] Overall Loss 0.203943 Objective Loss 0.203943 LR 0.000500 Time 0.021725 -2022-12-06 11:12:05,991 - Epoch: [111][ 280/ 1200] Overall Loss 0.203154 Objective Loss 0.203154 LR 0.000500 Time 0.021630 -2022-12-06 11:12:06,182 - Epoch: [111][ 290/ 1200] Overall Loss 0.202569 Objective Loss 0.202569 LR 0.000500 Time 0.021539 -2022-12-06 11:12:06,372 - Epoch: [111][ 300/ 1200] Overall Loss 0.202568 Objective Loss 0.202568 LR 0.000500 Time 0.021454 -2022-12-06 11:12:06,563 - Epoch: [111][ 310/ 1200] Overall Loss 0.202127 Objective Loss 0.202127 LR 0.000500 Time 0.021374 -2022-12-06 11:12:06,753 - Epoch: [111][ 320/ 1200] Overall Loss 0.202660 Objective Loss 0.202660 LR 0.000500 Time 0.021302 -2022-12-06 11:12:06,943 - Epoch: [111][ 330/ 1200] Overall Loss 0.203068 Objective Loss 0.203068 LR 0.000500 Time 0.021229 -2022-12-06 11:12:07,134 - Epoch: [111][ 340/ 1200] Overall Loss 0.203054 Objective Loss 0.203054 LR 0.000500 Time 0.021163 -2022-12-06 11:12:07,324 - Epoch: [111][ 350/ 1200] Overall Loss 0.202532 Objective Loss 0.202532 LR 0.000500 Time 0.021100 -2022-12-06 11:12:07,514 - Epoch: [111][ 360/ 1200] Overall Loss 0.202424 Objective Loss 0.202424 LR 0.000500 Time 0.021041 -2022-12-06 11:12:07,704 - Epoch: [111][ 370/ 1200] Overall Loss 0.202630 Objective Loss 0.202630 LR 0.000500 Time 0.020986 -2022-12-06 11:12:07,895 - Epoch: [111][ 380/ 1200] Overall Loss 0.202735 Objective Loss 0.202735 LR 0.000500 Time 0.020934 -2022-12-06 11:12:08,086 - Epoch: [111][ 390/ 1200] Overall Loss 0.202726 Objective Loss 0.202726 LR 0.000500 Time 0.020884 -2022-12-06 11:12:08,276 - Epoch: [111][ 400/ 1200] Overall Loss 0.202805 Objective Loss 0.202805 LR 0.000500 Time 0.020837 -2022-12-06 11:12:08,467 - Epoch: [111][ 410/ 1200] Overall Loss 0.203014 Objective Loss 0.203014 LR 0.000500 Time 0.020792 -2022-12-06 11:12:08,657 - Epoch: [111][ 420/ 1200] Overall Loss 0.203006 Objective Loss 0.203006 LR 0.000500 Time 0.020750 -2022-12-06 11:12:08,848 - Epoch: [111][ 430/ 1200] Overall Loss 0.202350 Objective Loss 0.202350 LR 0.000500 Time 0.020710 -2022-12-06 11:12:09,039 - Epoch: [111][ 440/ 1200] Overall Loss 0.202648 Objective Loss 0.202648 LR 0.000500 Time 0.020671 -2022-12-06 11:12:09,230 - Epoch: [111][ 450/ 1200] Overall Loss 0.203553 Objective Loss 0.203553 LR 0.000500 Time 0.020635 -2022-12-06 11:12:09,420 - Epoch: [111][ 460/ 1200] Overall Loss 0.204177 Objective Loss 0.204177 LR 0.000500 Time 0.020600 -2022-12-06 11:12:09,612 - Epoch: [111][ 470/ 1200] Overall Loss 0.204498 Objective Loss 0.204498 LR 0.000500 Time 0.020567 -2022-12-06 11:12:09,803 - Epoch: [111][ 480/ 1200] Overall Loss 0.204507 Objective Loss 0.204507 LR 0.000500 Time 0.020536 -2022-12-06 11:12:09,993 - Epoch: [111][ 490/ 1200] Overall Loss 0.204513 Objective Loss 0.204513 LR 0.000500 Time 0.020505 -2022-12-06 11:12:10,184 - Epoch: [111][ 500/ 1200] Overall Loss 0.204576 Objective Loss 0.204576 LR 0.000500 Time 0.020475 -2022-12-06 11:12:10,375 - Epoch: [111][ 510/ 1200] Overall Loss 0.204711 Objective Loss 0.204711 LR 0.000500 Time 0.020446 -2022-12-06 11:12:10,566 - Epoch: [111][ 520/ 1200] Overall Loss 0.204559 Objective Loss 0.204559 LR 0.000500 Time 0.020419 -2022-12-06 11:12:10,757 - Epoch: [111][ 530/ 1200] Overall Loss 0.205087 Objective Loss 0.205087 LR 0.000500 Time 0.020393 -2022-12-06 11:12:10,948 - Epoch: [111][ 540/ 1200] Overall Loss 0.204852 Objective Loss 0.204852 LR 0.000500 Time 0.020368 -2022-12-06 11:12:11,138 - Epoch: [111][ 550/ 1200] Overall Loss 0.205059 Objective Loss 0.205059 LR 0.000500 Time 0.020343 -2022-12-06 11:12:11,329 - Epoch: [111][ 560/ 1200] Overall Loss 0.204960 Objective Loss 0.204960 LR 0.000500 Time 0.020320 -2022-12-06 11:12:11,520 - Epoch: [111][ 570/ 1200] Overall Loss 0.204646 Objective Loss 0.204646 LR 0.000500 Time 0.020297 -2022-12-06 11:12:11,711 - Epoch: [111][ 580/ 1200] Overall Loss 0.204531 Objective Loss 0.204531 LR 0.000500 Time 0.020275 -2022-12-06 11:12:11,902 - Epoch: [111][ 590/ 1200] Overall Loss 0.204344 Objective Loss 0.204344 LR 0.000500 Time 0.020254 -2022-12-06 11:12:12,092 - Epoch: [111][ 600/ 1200] Overall Loss 0.204210 Objective Loss 0.204210 LR 0.000500 Time 0.020233 -2022-12-06 11:12:12,283 - Epoch: [111][ 610/ 1200] Overall Loss 0.203908 Objective Loss 0.203908 LR 0.000500 Time 0.020213 -2022-12-06 11:12:12,474 - Epoch: [111][ 620/ 1200] Overall Loss 0.203969 Objective Loss 0.203969 LR 0.000500 Time 0.020195 -2022-12-06 11:12:12,665 - Epoch: [111][ 630/ 1200] Overall Loss 0.204121 Objective Loss 0.204121 LR 0.000500 Time 0.020177 -2022-12-06 11:12:12,856 - Epoch: [111][ 640/ 1200] Overall Loss 0.203875 Objective Loss 0.203875 LR 0.000500 Time 0.020159 -2022-12-06 11:12:13,047 - Epoch: [111][ 650/ 1200] Overall Loss 0.203962 Objective Loss 0.203962 LR 0.000500 Time 0.020142 -2022-12-06 11:12:13,237 - Epoch: [111][ 660/ 1200] Overall Loss 0.204509 Objective Loss 0.204509 LR 0.000500 Time 0.020124 -2022-12-06 11:12:13,428 - Epoch: [111][ 670/ 1200] Overall Loss 0.204501 Objective Loss 0.204501 LR 0.000500 Time 0.020107 -2022-12-06 11:12:13,620 - Epoch: [111][ 680/ 1200] Overall Loss 0.204659 Objective Loss 0.204659 LR 0.000500 Time 0.020093 -2022-12-06 11:12:13,810 - Epoch: [111][ 690/ 1200] Overall Loss 0.204428 Objective Loss 0.204428 LR 0.000500 Time 0.020077 -2022-12-06 11:12:14,001 - Epoch: [111][ 700/ 1200] Overall Loss 0.204225 Objective Loss 0.204225 LR 0.000500 Time 0.020062 -2022-12-06 11:12:14,192 - Epoch: [111][ 710/ 1200] Overall Loss 0.204646 Objective Loss 0.204646 LR 0.000500 Time 0.020047 -2022-12-06 11:12:14,383 - Epoch: [111][ 720/ 1200] Overall Loss 0.204868 Objective Loss 0.204868 LR 0.000500 Time 0.020033 -2022-12-06 11:12:14,574 - Epoch: [111][ 730/ 1200] Overall Loss 0.204911 Objective Loss 0.204911 LR 0.000500 Time 0.020019 -2022-12-06 11:12:14,765 - Epoch: [111][ 740/ 1200] Overall Loss 0.205400 Objective Loss 0.205400 LR 0.000500 Time 0.020007 -2022-12-06 11:12:14,956 - Epoch: [111][ 750/ 1200] Overall Loss 0.205327 Objective Loss 0.205327 LR 0.000500 Time 0.019994 -2022-12-06 11:12:15,147 - Epoch: [111][ 760/ 1200] Overall Loss 0.205845 Objective Loss 0.205845 LR 0.000500 Time 0.019981 -2022-12-06 11:12:15,337 - Epoch: [111][ 770/ 1200] Overall Loss 0.205934 Objective Loss 0.205934 LR 0.000500 Time 0.019968 -2022-12-06 11:12:15,528 - Epoch: [111][ 780/ 1200] Overall Loss 0.205930 Objective Loss 0.205930 LR 0.000500 Time 0.019957 -2022-12-06 11:12:15,720 - Epoch: [111][ 790/ 1200] Overall Loss 0.206054 Objective Loss 0.206054 LR 0.000500 Time 0.019946 -2022-12-06 11:12:15,910 - Epoch: [111][ 800/ 1200] Overall Loss 0.206314 Objective Loss 0.206314 LR 0.000500 Time 0.019934 -2022-12-06 11:12:16,101 - Epoch: [111][ 810/ 1200] Overall Loss 0.206422 Objective Loss 0.206422 LR 0.000500 Time 0.019923 -2022-12-06 11:12:16,291 - Epoch: [111][ 820/ 1200] Overall Loss 0.206262 Objective Loss 0.206262 LR 0.000500 Time 0.019911 -2022-12-06 11:12:16,483 - Epoch: [111][ 830/ 1200] Overall Loss 0.206328 Objective Loss 0.206328 LR 0.000500 Time 0.019901 -2022-12-06 11:12:16,674 - Epoch: [111][ 840/ 1200] Overall Loss 0.206349 Objective Loss 0.206349 LR 0.000500 Time 0.019891 -2022-12-06 11:12:16,864 - Epoch: [111][ 850/ 1200] Overall Loss 0.206094 Objective Loss 0.206094 LR 0.000500 Time 0.019880 -2022-12-06 11:12:17,055 - Epoch: [111][ 860/ 1200] Overall Loss 0.205900 Objective Loss 0.205900 LR 0.000500 Time 0.019870 -2022-12-06 11:12:17,246 - Epoch: [111][ 870/ 1200] Overall Loss 0.205576 Objective Loss 0.205576 LR 0.000500 Time 0.019860 -2022-12-06 11:12:17,437 - Epoch: [111][ 880/ 1200] Overall Loss 0.205553 Objective Loss 0.205553 LR 0.000500 Time 0.019852 -2022-12-06 11:12:17,628 - Epoch: [111][ 890/ 1200] Overall Loss 0.205787 Objective Loss 0.205787 LR 0.000500 Time 0.019842 -2022-12-06 11:12:17,819 - Epoch: [111][ 900/ 1200] Overall Loss 0.206050 Objective Loss 0.206050 LR 0.000500 Time 0.019833 -2022-12-06 11:12:18,009 - Epoch: [111][ 910/ 1200] Overall Loss 0.206058 Objective Loss 0.206058 LR 0.000500 Time 0.019824 -2022-12-06 11:12:18,200 - Epoch: [111][ 920/ 1200] Overall Loss 0.206315 Objective Loss 0.206315 LR 0.000500 Time 0.019815 -2022-12-06 11:12:18,391 - Epoch: [111][ 930/ 1200] Overall Loss 0.206072 Objective Loss 0.206072 LR 0.000500 Time 0.019807 -2022-12-06 11:12:18,582 - Epoch: [111][ 940/ 1200] Overall Loss 0.206210 Objective Loss 0.206210 LR 0.000500 Time 0.019799 -2022-12-06 11:12:18,773 - Epoch: [111][ 950/ 1200] Overall Loss 0.206101 Objective Loss 0.206101 LR 0.000500 Time 0.019791 -2022-12-06 11:12:18,963 - Epoch: [111][ 960/ 1200] Overall Loss 0.206166 Objective Loss 0.206166 LR 0.000500 Time 0.019782 -2022-12-06 11:12:19,154 - Epoch: [111][ 970/ 1200] Overall Loss 0.206223 Objective Loss 0.206223 LR 0.000500 Time 0.019774 -2022-12-06 11:12:19,345 - Epoch: [111][ 980/ 1200] Overall Loss 0.206037 Objective Loss 0.206037 LR 0.000500 Time 0.019767 -2022-12-06 11:12:19,535 - Epoch: [111][ 990/ 1200] Overall Loss 0.206014 Objective Loss 0.206014 LR 0.000500 Time 0.019759 -2022-12-06 11:12:19,727 - Epoch: [111][ 1000/ 1200] Overall Loss 0.206090 Objective Loss 0.206090 LR 0.000500 Time 0.019752 -2022-12-06 11:12:19,918 - Epoch: [111][ 1010/ 1200] Overall Loss 0.205915 Objective Loss 0.205915 LR 0.000500 Time 0.019745 -2022-12-06 11:12:20,109 - Epoch: [111][ 1020/ 1200] Overall Loss 0.205842 Objective Loss 0.205842 LR 0.000500 Time 0.019739 -2022-12-06 11:12:20,299 - Epoch: [111][ 1030/ 1200] Overall Loss 0.206133 Objective Loss 0.206133 LR 0.000500 Time 0.019731 -2022-12-06 11:12:20,491 - Epoch: [111][ 1040/ 1200] Overall Loss 0.206119 Objective Loss 0.206119 LR 0.000500 Time 0.019725 -2022-12-06 11:12:20,682 - Epoch: [111][ 1050/ 1200] Overall Loss 0.206088 Objective Loss 0.206088 LR 0.000500 Time 0.019718 -2022-12-06 11:12:20,873 - Epoch: [111][ 1060/ 1200] Overall Loss 0.206011 Objective Loss 0.206011 LR 0.000500 Time 0.019712 -2022-12-06 11:12:21,064 - Epoch: [111][ 1070/ 1200] Overall Loss 0.206226 Objective Loss 0.206226 LR 0.000500 Time 0.019706 -2022-12-06 11:12:21,254 - Epoch: [111][ 1080/ 1200] Overall Loss 0.206194 Objective Loss 0.206194 LR 0.000500 Time 0.019699 -2022-12-06 11:12:21,445 - Epoch: [111][ 1090/ 1200] Overall Loss 0.206166 Objective Loss 0.206166 LR 0.000500 Time 0.019693 -2022-12-06 11:12:21,637 - Epoch: [111][ 1100/ 1200] Overall Loss 0.206213 Objective Loss 0.206213 LR 0.000500 Time 0.019688 -2022-12-06 11:12:21,827 - Epoch: [111][ 1110/ 1200] Overall Loss 0.206431 Objective Loss 0.206431 LR 0.000500 Time 0.019681 -2022-12-06 11:12:22,018 - Epoch: [111][ 1120/ 1200] Overall Loss 0.206334 Objective Loss 0.206334 LR 0.000500 Time 0.019676 -2022-12-06 11:12:22,209 - Epoch: [111][ 1130/ 1200] Overall Loss 0.206324 Objective Loss 0.206324 LR 0.000500 Time 0.019670 -2022-12-06 11:12:22,400 - Epoch: [111][ 1140/ 1200] Overall Loss 0.206418 Objective Loss 0.206418 LR 0.000500 Time 0.019665 -2022-12-06 11:12:22,590 - Epoch: [111][ 1150/ 1200] Overall Loss 0.206484 Objective Loss 0.206484 LR 0.000500 Time 0.019659 -2022-12-06 11:12:22,781 - Epoch: [111][ 1160/ 1200] Overall Loss 0.206498 Objective Loss 0.206498 LR 0.000500 Time 0.019653 -2022-12-06 11:12:22,972 - Epoch: [111][ 1170/ 1200] Overall Loss 0.206537 Objective Loss 0.206537 LR 0.000500 Time 0.019648 -2022-12-06 11:12:23,162 - Epoch: [111][ 1180/ 1200] Overall Loss 0.206727 Objective Loss 0.206727 LR 0.000500 Time 0.019642 -2022-12-06 11:12:23,353 - Epoch: [111][ 1190/ 1200] Overall Loss 0.206866 Objective Loss 0.206866 LR 0.000500 Time 0.019637 -2022-12-06 11:12:23,581 - Epoch: [111][ 1200/ 1200] Overall Loss 0.206837 Objective Loss 0.206837 Top1 87.238494 Top5 98.744770 LR 0.000500 Time 0.019663 -2022-12-06 11:12:23,671 - --- validate (epoch=111)----------- -2022-12-06 11:12:23,671 - 34129 samples (256 per mini-batch) -2022-12-06 11:12:24,116 - Epoch: [111][ 10/ 134] Loss 0.250477 Top1 86.406250 Top5 98.281250 -2022-12-06 11:12:24,251 - Epoch: [111][ 20/ 134] Loss 0.273548 Top1 85.546875 Top5 98.144531 -2022-12-06 11:12:24,397 - Epoch: [111][ 30/ 134] Loss 0.264738 Top1 85.638021 Top5 98.177083 -2022-12-06 11:12:24,537 - Epoch: [111][ 40/ 134] Loss 0.264216 Top1 85.761719 Top5 98.203125 -2022-12-06 11:12:24,683 - Epoch: [111][ 50/ 134] Loss 0.259750 Top1 85.750000 Top5 98.250000 -2022-12-06 11:12:24,821 - Epoch: [111][ 60/ 134] Loss 0.258865 Top1 85.885417 Top5 98.300781 -2022-12-06 11:12:24,969 - Epoch: [111][ 70/ 134] Loss 0.264451 Top1 85.786830 Top5 98.297991 -2022-12-06 11:12:25,107 - Epoch: [111][ 80/ 134] Loss 0.265372 Top1 85.839844 Top5 98.251953 -2022-12-06 11:12:25,241 - Epoch: [111][ 90/ 134] Loss 0.265858 Top1 85.915799 Top5 98.268229 -2022-12-06 11:12:25,374 - Epoch: [111][ 100/ 134] Loss 0.264013 Top1 86.042969 Top5 98.304688 -2022-12-06 11:12:25,505 - Epoch: [111][ 110/ 134] Loss 0.263492 Top1 85.994318 Top5 98.306108 -2022-12-06 11:12:25,642 - Epoch: [111][ 120/ 134] Loss 0.263904 Top1 86.012370 Top5 98.307292 -2022-12-06 11:12:25,779 - Epoch: [111][ 130/ 134] Loss 0.263267 Top1 86.045673 Top5 98.308293 -2022-12-06 11:12:25,816 - Epoch: [111][ 134/ 134] Loss 0.262631 Top1 86.023616 Top5 98.294705 -2022-12-06 11:12:25,903 - ==> Top1: 86.024 Top5: 98.295 Loss: 0.263 - -2022-12-06 11:12:25,904 - ==> Confusion: -[[ 885 0 4 3 7 5 1 0 7 62 0 2 3 2 4 2 0 0 3 1 5] - [ 2 950 3 2 8 9 4 11 4 0 5 4 1 1 0 2 3 1 7 4 6] - [ 5 2 1019 8 4 1 23 5 0 2 5 5 4 1 3 2 0 0 4 2 8] - [ 1 0 20 942 0 2 0 1 1 0 9 0 8 3 13 0 2 1 10 1 6] - [ 7 3 2 0 960 2 0 1 1 6 1 4 2 3 7 7 6 2 1 1 4] - [ 0 18 1 2 4 964 2 16 4 3 1 16 3 16 4 1 2 0 1 6 5] - [ 0 4 10 2 0 0 1075 3 2 1 0 2 1 2 0 5 1 1 1 6 2] - [ 0 10 14 2 1 23 6 938 0 0 3 7 0 2 0 2 0 0 21 20 5] - [ 6 2 0 0 1 2 1 1 963 44 12 3 0 12 9 0 2 2 2 2 0] - [ 33 1 4 0 5 1 0 1 19 911 1 1 0 12 3 1 0 1 0 1 6] - [ 0 1 3 7 1 2 1 4 10 1 960 1 0 10 3 0 1 0 3 2 9] - [ 2 0 1 0 0 7 5 3 1 0 1 983 15 5 2 5 2 4 1 10 4] - [ 0 1 1 0 1 1 0 0 0 0 0 40 894 2 1 11 2 5 0 6 4] - [ 0 1 1 0 1 5 0 2 13 10 5 4 4 969 0 0 2 0 0 1 5] - [ 5 3 3 10 4 2 0 1 28 4 0 2 1 3 1042 0 1 2 9 2 8] - [ 1 0 0 0 1 0 5 0 0 0 1 13 3 2 0 994 4 11 0 3 5] - [ 0 2 1 2 2 1 1 0 2 0 0 3 1 4 2 11 1029 1 1 4 5] - [ 4 0 1 2 0 1 1 2 2 0 1 14 18 3 1 14 1 969 0 0 2] - [ 4 6 7 6 2 2 1 19 0 1 4 4 4 2 8 0 0 0 931 1 6] - [ 4 1 1 0 0 4 8 4 0 0 1 16 10 7 0 5 5 1 1 1009 3] - [ 108 211 232 104 128 146 103 102 99 101 153 125 360 333 122 121 192 83 157 277 9969]] - -2022-12-06 11:12:26,567 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:12:26,567 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:12:26,573 - - -2022-12-06 11:12:26,573 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:12:27,502 - Epoch: [112][ 10/ 1200] Overall Loss 0.203741 Objective Loss 0.203741 LR 0.000500 Time 0.092819 -2022-12-06 11:12:27,698 - Epoch: [112][ 20/ 1200] Overall Loss 0.192979 Objective Loss 0.192979 LR 0.000500 Time 0.056167 -2022-12-06 11:12:27,889 - Epoch: [112][ 30/ 1200] Overall Loss 0.186601 Objective Loss 0.186601 LR 0.000500 Time 0.043791 -2022-12-06 11:12:28,080 - Epoch: [112][ 40/ 1200] Overall Loss 0.186943 Objective Loss 0.186943 LR 0.000500 Time 0.037607 -2022-12-06 11:12:28,270 - Epoch: [112][ 50/ 1200] Overall Loss 0.191820 Objective Loss 0.191820 LR 0.000500 Time 0.033890 -2022-12-06 11:12:28,462 - Epoch: [112][ 60/ 1200] Overall Loss 0.193413 Objective Loss 0.193413 LR 0.000500 Time 0.031422 -2022-12-06 11:12:28,652 - Epoch: [112][ 70/ 1200] Overall Loss 0.195618 Objective Loss 0.195618 LR 0.000500 Time 0.029649 -2022-12-06 11:12:28,844 - Epoch: [112][ 80/ 1200] Overall Loss 0.199024 Objective Loss 0.199024 LR 0.000500 Time 0.028325 -2022-12-06 11:12:29,034 - Epoch: [112][ 90/ 1200] Overall Loss 0.199930 Objective Loss 0.199930 LR 0.000500 Time 0.027291 -2022-12-06 11:12:29,226 - Epoch: [112][ 100/ 1200] Overall Loss 0.198296 Objective Loss 0.198296 LR 0.000500 Time 0.026469 -2022-12-06 11:12:29,416 - Epoch: [112][ 110/ 1200] Overall Loss 0.199319 Objective Loss 0.199319 LR 0.000500 Time 0.025788 -2022-12-06 11:12:29,606 - Epoch: [112][ 120/ 1200] Overall Loss 0.200499 Objective Loss 0.200499 LR 0.000500 Time 0.025222 -2022-12-06 11:12:29,797 - Epoch: [112][ 130/ 1200] Overall Loss 0.202774 Objective Loss 0.202774 LR 0.000500 Time 0.024745 -2022-12-06 11:12:29,988 - Epoch: [112][ 140/ 1200] Overall Loss 0.203818 Objective Loss 0.203818 LR 0.000500 Time 0.024338 -2022-12-06 11:12:30,179 - Epoch: [112][ 150/ 1200] Overall Loss 0.203217 Objective Loss 0.203217 LR 0.000500 Time 0.023986 -2022-12-06 11:12:30,370 - Epoch: [112][ 160/ 1200] Overall Loss 0.203659 Objective Loss 0.203659 LR 0.000500 Time 0.023673 -2022-12-06 11:12:30,560 - Epoch: [112][ 170/ 1200] Overall Loss 0.204407 Objective Loss 0.204407 LR 0.000500 Time 0.023398 -2022-12-06 11:12:30,751 - Epoch: [112][ 180/ 1200] Overall Loss 0.205272 Objective Loss 0.205272 LR 0.000500 Time 0.023156 -2022-12-06 11:12:30,941 - Epoch: [112][ 190/ 1200] Overall Loss 0.204877 Objective Loss 0.204877 LR 0.000500 Time 0.022933 -2022-12-06 11:12:31,132 - Epoch: [112][ 200/ 1200] Overall Loss 0.204593 Objective Loss 0.204593 LR 0.000500 Time 0.022739 -2022-12-06 11:12:31,323 - Epoch: [112][ 210/ 1200] Overall Loss 0.205365 Objective Loss 0.205365 LR 0.000500 Time 0.022563 -2022-12-06 11:12:31,514 - Epoch: [112][ 220/ 1200] Overall Loss 0.205833 Objective Loss 0.205833 LR 0.000500 Time 0.022403 -2022-12-06 11:12:31,704 - Epoch: [112][ 230/ 1200] Overall Loss 0.206867 Objective Loss 0.206867 LR 0.000500 Time 0.022253 -2022-12-06 11:12:31,895 - Epoch: [112][ 240/ 1200] Overall Loss 0.207615 Objective Loss 0.207615 LR 0.000500 Time 0.022118 -2022-12-06 11:12:32,085 - Epoch: [112][ 250/ 1200] Overall Loss 0.207646 Objective Loss 0.207646 LR 0.000500 Time 0.021994 -2022-12-06 11:12:32,276 - Epoch: [112][ 260/ 1200] Overall Loss 0.207782 Objective Loss 0.207782 LR 0.000500 Time 0.021880 -2022-12-06 11:12:32,467 - Epoch: [112][ 270/ 1200] Overall Loss 0.206620 Objective Loss 0.206620 LR 0.000500 Time 0.021774 -2022-12-06 11:12:32,658 - Epoch: [112][ 280/ 1200] Overall Loss 0.206451 Objective Loss 0.206451 LR 0.000500 Time 0.021677 -2022-12-06 11:12:32,848 - Epoch: [112][ 290/ 1200] Overall Loss 0.206669 Objective Loss 0.206669 LR 0.000500 Time 0.021584 -2022-12-06 11:12:33,039 - Epoch: [112][ 300/ 1200] Overall Loss 0.207193 Objective Loss 0.207193 LR 0.000500 Time 0.021499 -2022-12-06 11:12:33,230 - Epoch: [112][ 310/ 1200] Overall Loss 0.206792 Objective Loss 0.206792 LR 0.000500 Time 0.021420 -2022-12-06 11:12:33,421 - Epoch: [112][ 320/ 1200] Overall Loss 0.206258 Objective Loss 0.206258 LR 0.000500 Time 0.021345 -2022-12-06 11:12:33,612 - Epoch: [112][ 330/ 1200] Overall Loss 0.205712 Objective Loss 0.205712 LR 0.000500 Time 0.021275 -2022-12-06 11:12:33,802 - Epoch: [112][ 340/ 1200] Overall Loss 0.206013 Objective Loss 0.206013 LR 0.000500 Time 0.021207 -2022-12-06 11:12:33,993 - Epoch: [112][ 350/ 1200] Overall Loss 0.205500 Objective Loss 0.205500 LR 0.000500 Time 0.021144 -2022-12-06 11:12:34,183 - Epoch: [112][ 360/ 1200] Overall Loss 0.206226 Objective Loss 0.206226 LR 0.000500 Time 0.021084 -2022-12-06 11:12:34,374 - Epoch: [112][ 370/ 1200] Overall Loss 0.205936 Objective Loss 0.205936 LR 0.000500 Time 0.021028 -2022-12-06 11:12:34,564 - Epoch: [112][ 380/ 1200] Overall Loss 0.205818 Objective Loss 0.205818 LR 0.000500 Time 0.020974 -2022-12-06 11:12:34,755 - Epoch: [112][ 390/ 1200] Overall Loss 0.205279 Objective Loss 0.205279 LR 0.000500 Time 0.020923 -2022-12-06 11:12:34,946 - Epoch: [112][ 400/ 1200] Overall Loss 0.205232 Objective Loss 0.205232 LR 0.000500 Time 0.020877 -2022-12-06 11:12:35,136 - Epoch: [112][ 410/ 1200] Overall Loss 0.204766 Objective Loss 0.204766 LR 0.000500 Time 0.020830 -2022-12-06 11:12:35,327 - Epoch: [112][ 420/ 1200] Overall Loss 0.204265 Objective Loss 0.204265 LR 0.000500 Time 0.020787 -2022-12-06 11:12:35,517 - Epoch: [112][ 430/ 1200] Overall Loss 0.204191 Objective Loss 0.204191 LR 0.000500 Time 0.020745 -2022-12-06 11:12:35,708 - Epoch: [112][ 440/ 1200] Overall Loss 0.203889 Objective Loss 0.203889 LR 0.000500 Time 0.020706 -2022-12-06 11:12:35,899 - Epoch: [112][ 450/ 1200] Overall Loss 0.204427 Objective Loss 0.204427 LR 0.000500 Time 0.020669 -2022-12-06 11:12:36,090 - Epoch: [112][ 460/ 1200] Overall Loss 0.204800 Objective Loss 0.204800 LR 0.000500 Time 0.020633 -2022-12-06 11:12:36,280 - Epoch: [112][ 470/ 1200] Overall Loss 0.204378 Objective Loss 0.204378 LR 0.000500 Time 0.020598 -2022-12-06 11:12:36,471 - Epoch: [112][ 480/ 1200] Overall Loss 0.204262 Objective Loss 0.204262 LR 0.000500 Time 0.020565 -2022-12-06 11:12:36,661 - Epoch: [112][ 490/ 1200] Overall Loss 0.203944 Objective Loss 0.203944 LR 0.000500 Time 0.020533 -2022-12-06 11:12:36,852 - Epoch: [112][ 500/ 1200] Overall Loss 0.203855 Objective Loss 0.203855 LR 0.000500 Time 0.020502 -2022-12-06 11:12:37,042 - Epoch: [112][ 510/ 1200] Overall Loss 0.204223 Objective Loss 0.204223 LR 0.000500 Time 0.020473 -2022-12-06 11:12:37,233 - Epoch: [112][ 520/ 1200] Overall Loss 0.203970 Objective Loss 0.203970 LR 0.000500 Time 0.020446 -2022-12-06 11:12:37,424 - Epoch: [112][ 530/ 1200] Overall Loss 0.204279 Objective Loss 0.204279 LR 0.000500 Time 0.020418 -2022-12-06 11:12:37,614 - Epoch: [112][ 540/ 1200] Overall Loss 0.204220 Objective Loss 0.204220 LR 0.000500 Time 0.020392 -2022-12-06 11:12:37,805 - Epoch: [112][ 550/ 1200] Overall Loss 0.203957 Objective Loss 0.203957 LR 0.000500 Time 0.020367 -2022-12-06 11:12:37,995 - Epoch: [112][ 560/ 1200] Overall Loss 0.204005 Objective Loss 0.204005 LR 0.000500 Time 0.020342 -2022-12-06 11:12:38,186 - Epoch: [112][ 570/ 1200] Overall Loss 0.203949 Objective Loss 0.203949 LR 0.000500 Time 0.020319 -2022-12-06 11:12:38,377 - Epoch: [112][ 580/ 1200] Overall Loss 0.204261 Objective Loss 0.204261 LR 0.000500 Time 0.020296 -2022-12-06 11:12:38,567 - Epoch: [112][ 590/ 1200] Overall Loss 0.204822 Objective Loss 0.204822 LR 0.000500 Time 0.020274 -2022-12-06 11:12:38,758 - Epoch: [112][ 600/ 1200] Overall Loss 0.204846 Objective Loss 0.204846 LR 0.000500 Time 0.020254 -2022-12-06 11:12:38,949 - Epoch: [112][ 610/ 1200] Overall Loss 0.204846 Objective Loss 0.204846 LR 0.000500 Time 0.020233 -2022-12-06 11:12:39,140 - Epoch: [112][ 620/ 1200] Overall Loss 0.204953 Objective Loss 0.204953 LR 0.000500 Time 0.020214 -2022-12-06 11:12:39,330 - Epoch: [112][ 630/ 1200] Overall Loss 0.205166 Objective Loss 0.205166 LR 0.000500 Time 0.020194 -2022-12-06 11:12:39,520 - Epoch: [112][ 640/ 1200] Overall Loss 0.205391 Objective Loss 0.205391 LR 0.000500 Time 0.020175 -2022-12-06 11:12:39,711 - Epoch: [112][ 650/ 1200] Overall Loss 0.205359 Objective Loss 0.205359 LR 0.000500 Time 0.020157 -2022-12-06 11:12:39,901 - Epoch: [112][ 660/ 1200] Overall Loss 0.205162 Objective Loss 0.205162 LR 0.000500 Time 0.020140 -2022-12-06 11:12:40,092 - Epoch: [112][ 670/ 1200] Overall Loss 0.205110 Objective Loss 0.205110 LR 0.000500 Time 0.020122 -2022-12-06 11:12:40,282 - Epoch: [112][ 680/ 1200] Overall Loss 0.205217 Objective Loss 0.205217 LR 0.000500 Time 0.020106 -2022-12-06 11:12:40,473 - Epoch: [112][ 690/ 1200] Overall Loss 0.205261 Objective Loss 0.205261 LR 0.000500 Time 0.020089 -2022-12-06 11:12:40,663 - Epoch: [112][ 700/ 1200] Overall Loss 0.205390 Objective Loss 0.205390 LR 0.000500 Time 0.020074 -2022-12-06 11:12:40,853 - Epoch: [112][ 710/ 1200] Overall Loss 0.205352 Objective Loss 0.205352 LR 0.000500 Time 0.020057 -2022-12-06 11:12:41,043 - Epoch: [112][ 720/ 1200] Overall Loss 0.205764 Objective Loss 0.205764 LR 0.000500 Time 0.020042 -2022-12-06 11:12:41,233 - Epoch: [112][ 730/ 1200] Overall Loss 0.205657 Objective Loss 0.205657 LR 0.000500 Time 0.020027 -2022-12-06 11:12:41,423 - Epoch: [112][ 740/ 1200] Overall Loss 0.205916 Objective Loss 0.205916 LR 0.000500 Time 0.020012 -2022-12-06 11:12:41,613 - Epoch: [112][ 750/ 1200] Overall Loss 0.205835 Objective Loss 0.205835 LR 0.000500 Time 0.019999 -2022-12-06 11:12:41,803 - Epoch: [112][ 760/ 1200] Overall Loss 0.205606 Objective Loss 0.205606 LR 0.000500 Time 0.019985 -2022-12-06 11:12:41,994 - Epoch: [112][ 770/ 1200] Overall Loss 0.205516 Objective Loss 0.205516 LR 0.000500 Time 0.019972 -2022-12-06 11:12:42,184 - Epoch: [112][ 780/ 1200] Overall Loss 0.205744 Objective Loss 0.205744 LR 0.000500 Time 0.019959 -2022-12-06 11:12:42,374 - Epoch: [112][ 790/ 1200] Overall Loss 0.205605 Objective Loss 0.205605 LR 0.000500 Time 0.019947 -2022-12-06 11:12:42,564 - Epoch: [112][ 800/ 1200] Overall Loss 0.205763 Objective Loss 0.205763 LR 0.000500 Time 0.019934 -2022-12-06 11:12:42,754 - Epoch: [112][ 810/ 1200] Overall Loss 0.205771 Objective Loss 0.205771 LR 0.000500 Time 0.019922 -2022-12-06 11:12:42,945 - Epoch: [112][ 820/ 1200] Overall Loss 0.205726 Objective Loss 0.205726 LR 0.000500 Time 0.019911 -2022-12-06 11:12:43,136 - Epoch: [112][ 830/ 1200] Overall Loss 0.205428 Objective Loss 0.205428 LR 0.000500 Time 0.019900 -2022-12-06 11:12:43,326 - Epoch: [112][ 840/ 1200] Overall Loss 0.205127 Objective Loss 0.205127 LR 0.000500 Time 0.019889 -2022-12-06 11:12:43,516 - Epoch: [112][ 850/ 1200] Overall Loss 0.205358 Objective Loss 0.205358 LR 0.000500 Time 0.019878 -2022-12-06 11:12:43,706 - Epoch: [112][ 860/ 1200] Overall Loss 0.205424 Objective Loss 0.205424 LR 0.000500 Time 0.019868 -2022-12-06 11:12:43,897 - Epoch: [112][ 870/ 1200] Overall Loss 0.205372 Objective Loss 0.205372 LR 0.000500 Time 0.019857 -2022-12-06 11:12:44,087 - Epoch: [112][ 880/ 1200] Overall Loss 0.205285 Objective Loss 0.205285 LR 0.000500 Time 0.019848 -2022-12-06 11:12:44,278 - Epoch: [112][ 890/ 1200] Overall Loss 0.205436 Objective Loss 0.205436 LR 0.000500 Time 0.019838 -2022-12-06 11:12:44,469 - Epoch: [112][ 900/ 1200] Overall Loss 0.205535 Objective Loss 0.205535 LR 0.000500 Time 0.019830 -2022-12-06 11:12:44,660 - Epoch: [112][ 910/ 1200] Overall Loss 0.205711 Objective Loss 0.205711 LR 0.000500 Time 0.019821 -2022-12-06 11:12:44,851 - Epoch: [112][ 920/ 1200] Overall Loss 0.206355 Objective Loss 0.206355 LR 0.000500 Time 0.019812 -2022-12-06 11:12:45,041 - Epoch: [112][ 930/ 1200] Overall Loss 0.206431 Objective Loss 0.206431 LR 0.000500 Time 0.019803 -2022-12-06 11:12:45,232 - Epoch: [112][ 940/ 1200] Overall Loss 0.206467 Objective Loss 0.206467 LR 0.000500 Time 0.019795 -2022-12-06 11:12:45,423 - Epoch: [112][ 950/ 1200] Overall Loss 0.206625 Objective Loss 0.206625 LR 0.000500 Time 0.019787 -2022-12-06 11:12:45,614 - Epoch: [112][ 960/ 1200] Overall Loss 0.206631 Objective Loss 0.206631 LR 0.000500 Time 0.019779 -2022-12-06 11:12:45,804 - Epoch: [112][ 970/ 1200] Overall Loss 0.207003 Objective Loss 0.207003 LR 0.000500 Time 0.019771 -2022-12-06 11:12:45,996 - Epoch: [112][ 980/ 1200] Overall Loss 0.207016 Objective Loss 0.207016 LR 0.000500 Time 0.019764 -2022-12-06 11:12:46,187 - Epoch: [112][ 990/ 1200] Overall Loss 0.206825 Objective Loss 0.206825 LR 0.000500 Time 0.019757 -2022-12-06 11:12:46,378 - Epoch: [112][ 1000/ 1200] Overall Loss 0.206760 Objective Loss 0.206760 LR 0.000500 Time 0.019750 -2022-12-06 11:12:46,568 - Epoch: [112][ 1010/ 1200] Overall Loss 0.206620 Objective Loss 0.206620 LR 0.000500 Time 0.019742 -2022-12-06 11:12:46,760 - Epoch: [112][ 1020/ 1200] Overall Loss 0.206605 Objective Loss 0.206605 LR 0.000500 Time 0.019736 -2022-12-06 11:12:46,950 - Epoch: [112][ 1030/ 1200] Overall Loss 0.206454 Objective Loss 0.206454 LR 0.000500 Time 0.019729 -2022-12-06 11:12:47,141 - Epoch: [112][ 1040/ 1200] Overall Loss 0.206697 Objective Loss 0.206697 LR 0.000500 Time 0.019722 -2022-12-06 11:12:47,332 - Epoch: [112][ 1050/ 1200] Overall Loss 0.206554 Objective Loss 0.206554 LR 0.000500 Time 0.019716 -2022-12-06 11:12:47,523 - Epoch: [112][ 1060/ 1200] Overall Loss 0.206542 Objective Loss 0.206542 LR 0.000500 Time 0.019710 -2022-12-06 11:12:47,715 - Epoch: [112][ 1070/ 1200] Overall Loss 0.206455 Objective Loss 0.206455 LR 0.000500 Time 0.019704 -2022-12-06 11:12:47,906 - Epoch: [112][ 1080/ 1200] Overall Loss 0.206522 Objective Loss 0.206522 LR 0.000500 Time 0.019697 -2022-12-06 11:12:48,096 - Epoch: [112][ 1090/ 1200] Overall Loss 0.206577 Objective Loss 0.206577 LR 0.000500 Time 0.019691 -2022-12-06 11:12:48,287 - Epoch: [112][ 1100/ 1200] Overall Loss 0.206606 Objective Loss 0.206606 LR 0.000500 Time 0.019685 -2022-12-06 11:12:48,477 - Epoch: [112][ 1110/ 1200] Overall Loss 0.206545 Objective Loss 0.206545 LR 0.000500 Time 0.019679 -2022-12-06 11:12:48,668 - Epoch: [112][ 1120/ 1200] Overall Loss 0.206876 Objective Loss 0.206876 LR 0.000500 Time 0.019673 -2022-12-06 11:12:48,858 - Epoch: [112][ 1130/ 1200] Overall Loss 0.206881 Objective Loss 0.206881 LR 0.000500 Time 0.019667 -2022-12-06 11:12:49,049 - Epoch: [112][ 1140/ 1200] Overall Loss 0.207052 Objective Loss 0.207052 LR 0.000500 Time 0.019661 -2022-12-06 11:12:49,240 - Epoch: [112][ 1150/ 1200] Overall Loss 0.207113 Objective Loss 0.207113 LR 0.000500 Time 0.019655 -2022-12-06 11:12:49,431 - Epoch: [112][ 1160/ 1200] Overall Loss 0.207311 Objective Loss 0.207311 LR 0.000500 Time 0.019650 -2022-12-06 11:12:49,621 - Epoch: [112][ 1170/ 1200] Overall Loss 0.207435 Objective Loss 0.207435 LR 0.000500 Time 0.019645 -2022-12-06 11:12:49,812 - Epoch: [112][ 1180/ 1200] Overall Loss 0.207154 Objective Loss 0.207154 LR 0.000500 Time 0.019639 -2022-12-06 11:12:50,002 - Epoch: [112][ 1190/ 1200] Overall Loss 0.207053 Objective Loss 0.207053 LR 0.000500 Time 0.019634 -2022-12-06 11:12:50,233 - Epoch: [112][ 1200/ 1200] Overall Loss 0.206907 Objective Loss 0.206907 Top1 89.121339 Top5 97.907950 LR 0.000500 Time 0.019662 -2022-12-06 11:12:50,320 - --- validate (epoch=112)----------- -2022-12-06 11:12:50,321 - 34129 samples (256 per mini-batch) -2022-12-06 11:12:50,765 - Epoch: [112][ 10/ 134] Loss 0.254254 Top1 86.640625 Top5 98.007812 -2022-12-06 11:12:50,899 - Epoch: [112][ 20/ 134] Loss 0.246196 Top1 86.308594 Top5 98.183594 -2022-12-06 11:12:51,032 - Epoch: [112][ 30/ 134] Loss 0.256322 Top1 85.885417 Top5 98.098958 -2022-12-06 11:12:51,165 - Epoch: [112][ 40/ 134] Loss 0.258332 Top1 85.976562 Top5 98.164062 -2022-12-06 11:12:51,299 - Epoch: [112][ 50/ 134] Loss 0.259070 Top1 86.156250 Top5 98.187500 -2022-12-06 11:12:51,430 - Epoch: [112][ 60/ 134] Loss 0.259225 Top1 86.028646 Top5 98.190104 -2022-12-06 11:12:51,563 - Epoch: [112][ 70/ 134] Loss 0.258657 Top1 85.943080 Top5 98.180804 -2022-12-06 11:12:51,694 - Epoch: [112][ 80/ 134] Loss 0.261811 Top1 85.908203 Top5 98.164062 -2022-12-06 11:12:51,827 - Epoch: [112][ 90/ 134] Loss 0.265153 Top1 85.798611 Top5 98.142361 -2022-12-06 11:12:51,958 - Epoch: [112][ 100/ 134] Loss 0.266482 Top1 85.656250 Top5 98.101562 -2022-12-06 11:12:52,092 - Epoch: [112][ 110/ 134] Loss 0.264693 Top1 85.781250 Top5 98.156960 -2022-12-06 11:12:52,222 - Epoch: [112][ 120/ 134] Loss 0.264015 Top1 85.797526 Top5 98.125000 -2022-12-06 11:12:52,355 - Epoch: [112][ 130/ 134] Loss 0.264972 Top1 85.871394 Top5 98.149038 -2022-12-06 11:12:52,394 - Epoch: [112][ 134/ 134] Loss 0.264932 Top1 85.912274 Top5 98.162853 -2022-12-06 11:12:52,482 - ==> Top1: 85.912 Top5: 98.163 Loss: 0.265 - -2022-12-06 11:12:52,483 - ==> Confusion: -[[ 909 1 2 3 9 6 1 0 6 45 0 1 0 1 6 2 0 0 1 0 3] - [ 2 947 1 2 12 22 2 7 4 0 3 5 1 1 0 0 1 1 6 5 5] - [ 4 0 988 17 6 2 29 8 1 2 5 7 2 2 5 3 1 2 7 2 10] - [ 3 2 12 946 0 4 1 1 0 0 9 0 7 5 14 0 0 2 11 0 3] - [ 8 6 2 0 964 2 1 1 0 4 2 2 0 1 11 8 3 1 0 1 3] - [ 3 13 0 1 3 975 2 19 2 2 1 8 6 14 3 2 1 1 1 8 4] - [ 0 4 8 4 1 2 1071 3 0 0 1 2 1 2 0 6 0 2 1 7 3] - [ 0 10 8 3 1 24 5 943 0 0 3 5 1 3 1 0 0 0 24 17 6] - [ 8 3 0 0 0 3 0 0 974 41 6 1 4 7 9 1 2 1 1 1 2] - [ 51 0 1 1 7 1 0 2 28 880 1 1 0 13 5 1 0 3 0 1 5] - [ 0 3 4 5 1 2 1 3 8 0 957 2 0 13 5 0 1 1 5 3 5] - [ 4 2 1 0 1 15 3 3 0 0 0 975 20 0 0 5 3 4 0 12 3] - [ 1 0 2 1 0 1 0 1 2 0 0 35 893 1 1 9 1 7 1 6 7] - [ 1 1 1 0 0 8 0 2 13 18 3 3 5 952 2 2 2 2 2 2 4] - [ 8 3 2 9 5 1 0 0 21 4 0 1 2 2 1060 0 0 0 6 1 5] - [ 0 0 0 0 2 3 7 0 0 1 1 8 8 2 0 984 8 12 0 2 5] - [ 2 1 1 0 3 3 1 0 0 0 0 3 3 2 0 11 1031 1 0 5 5] - [ 2 1 1 4 1 1 2 1 1 0 0 12 16 1 1 16 2 969 2 0 3] - [ 5 7 3 8 2 1 1 20 4 0 6 3 7 0 12 1 0 1 923 0 4] - [ 4 3 2 0 0 7 5 7 0 0 1 11 8 6 0 3 5 1 1 1013 3] - [ 147 216 164 113 129 189 91 148 124 83 164 114 387 283 157 106 182 67 139 261 9962]] - -2022-12-06 11:12:53,051 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:12:53,051 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:12:53,057 - - -2022-12-06 11:12:53,057 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:12:54,087 - Epoch: [113][ 10/ 1200] Overall Loss 0.202543 Objective Loss 0.202543 LR 0.000500 Time 0.102931 -2022-12-06 11:12:54,287 - Epoch: [113][ 20/ 1200] Overall Loss 0.192286 Objective Loss 0.192286 LR 0.000500 Time 0.061441 -2022-12-06 11:12:54,479 - Epoch: [113][ 30/ 1200] Overall Loss 0.202590 Objective Loss 0.202590 LR 0.000500 Time 0.047324 -2022-12-06 11:12:54,670 - Epoch: [113][ 40/ 1200] Overall Loss 0.197500 Objective Loss 0.197500 LR 0.000500 Time 0.040256 -2022-12-06 11:12:54,861 - Epoch: [113][ 50/ 1200] Overall Loss 0.194641 Objective Loss 0.194641 LR 0.000500 Time 0.036023 -2022-12-06 11:12:55,052 - Epoch: [113][ 60/ 1200] Overall Loss 0.199759 Objective Loss 0.199759 LR 0.000500 Time 0.033191 -2022-12-06 11:12:55,243 - Epoch: [113][ 70/ 1200] Overall Loss 0.199516 Objective Loss 0.199516 LR 0.000500 Time 0.031161 -2022-12-06 11:12:55,434 - Epoch: [113][ 80/ 1200] Overall Loss 0.201436 Objective Loss 0.201436 LR 0.000500 Time 0.029647 -2022-12-06 11:12:55,625 - Epoch: [113][ 90/ 1200] Overall Loss 0.201347 Objective Loss 0.201347 LR 0.000500 Time 0.028471 -2022-12-06 11:12:55,816 - Epoch: [113][ 100/ 1200] Overall Loss 0.199323 Objective Loss 0.199323 LR 0.000500 Time 0.027526 -2022-12-06 11:12:56,006 - Epoch: [113][ 110/ 1200] Overall Loss 0.198552 Objective Loss 0.198552 LR 0.000500 Time 0.026751 -2022-12-06 11:12:56,198 - Epoch: [113][ 120/ 1200] Overall Loss 0.200830 Objective Loss 0.200830 LR 0.000500 Time 0.026112 -2022-12-06 11:12:56,388 - Epoch: [113][ 130/ 1200] Overall Loss 0.201867 Objective Loss 0.201867 LR 0.000500 Time 0.025566 -2022-12-06 11:12:56,579 - Epoch: [113][ 140/ 1200] Overall Loss 0.202920 Objective Loss 0.202920 LR 0.000500 Time 0.025102 -2022-12-06 11:12:56,770 - Epoch: [113][ 150/ 1200] Overall Loss 0.202476 Objective Loss 0.202476 LR 0.000500 Time 0.024694 -2022-12-06 11:12:56,961 - Epoch: [113][ 160/ 1200] Overall Loss 0.201307 Objective Loss 0.201307 LR 0.000500 Time 0.024340 -2022-12-06 11:12:57,151 - Epoch: [113][ 170/ 1200] Overall Loss 0.201475 Objective Loss 0.201475 LR 0.000500 Time 0.024026 -2022-12-06 11:12:57,342 - Epoch: [113][ 180/ 1200] Overall Loss 0.201891 Objective Loss 0.201891 LR 0.000500 Time 0.023749 -2022-12-06 11:12:57,533 - Epoch: [113][ 190/ 1200] Overall Loss 0.202084 Objective Loss 0.202084 LR 0.000500 Time 0.023499 -2022-12-06 11:12:57,724 - Epoch: [113][ 200/ 1200] Overall Loss 0.203614 Objective Loss 0.203614 LR 0.000500 Time 0.023275 -2022-12-06 11:12:57,915 - Epoch: [113][ 210/ 1200] Overall Loss 0.202890 Objective Loss 0.202890 LR 0.000500 Time 0.023077 -2022-12-06 11:12:58,106 - Epoch: [113][ 220/ 1200] Overall Loss 0.202741 Objective Loss 0.202741 LR 0.000500 Time 0.022893 -2022-12-06 11:12:58,296 - Epoch: [113][ 230/ 1200] Overall Loss 0.202369 Objective Loss 0.202369 LR 0.000500 Time 0.022721 -2022-12-06 11:12:58,487 - Epoch: [113][ 240/ 1200] Overall Loss 0.200701 Objective Loss 0.200701 LR 0.000500 Time 0.022568 -2022-12-06 11:12:58,677 - Epoch: [113][ 250/ 1200] Overall Loss 0.200776 Objective Loss 0.200776 LR 0.000500 Time 0.022424 -2022-12-06 11:12:58,869 - Epoch: [113][ 260/ 1200] Overall Loss 0.201587 Objective Loss 0.201587 LR 0.000500 Time 0.022296 -2022-12-06 11:12:59,059 - Epoch: [113][ 270/ 1200] Overall Loss 0.201729 Objective Loss 0.201729 LR 0.000500 Time 0.022175 -2022-12-06 11:12:59,250 - Epoch: [113][ 280/ 1200] Overall Loss 0.202200 Objective Loss 0.202200 LR 0.000500 Time 0.022063 -2022-12-06 11:12:59,441 - Epoch: [113][ 290/ 1200] Overall Loss 0.201686 Objective Loss 0.201686 LR 0.000500 Time 0.021958 -2022-12-06 11:12:59,632 - Epoch: [113][ 300/ 1200] Overall Loss 0.201387 Objective Loss 0.201387 LR 0.000500 Time 0.021861 -2022-12-06 11:12:59,824 - Epoch: [113][ 310/ 1200] Overall Loss 0.201179 Objective Loss 0.201179 LR 0.000500 Time 0.021771 -2022-12-06 11:13:00,014 - Epoch: [113][ 320/ 1200] Overall Loss 0.202086 Objective Loss 0.202086 LR 0.000500 Time 0.021686 -2022-12-06 11:13:00,206 - Epoch: [113][ 330/ 1200] Overall Loss 0.201719 Objective Loss 0.201719 LR 0.000500 Time 0.021607 -2022-12-06 11:13:00,397 - Epoch: [113][ 340/ 1200] Overall Loss 0.201742 Objective Loss 0.201742 LR 0.000500 Time 0.021531 -2022-12-06 11:13:00,587 - Epoch: [113][ 350/ 1200] Overall Loss 0.202139 Objective Loss 0.202139 LR 0.000500 Time 0.021458 -2022-12-06 11:13:00,778 - Epoch: [113][ 360/ 1200] Overall Loss 0.202136 Objective Loss 0.202136 LR 0.000500 Time 0.021390 -2022-12-06 11:13:00,968 - Epoch: [113][ 370/ 1200] Overall Loss 0.202351 Objective Loss 0.202351 LR 0.000500 Time 0.021326 -2022-12-06 11:13:01,159 - Epoch: [113][ 380/ 1200] Overall Loss 0.202132 Objective Loss 0.202132 LR 0.000500 Time 0.021265 -2022-12-06 11:13:01,350 - Epoch: [113][ 390/ 1200] Overall Loss 0.201974 Objective Loss 0.201974 LR 0.000500 Time 0.021208 -2022-12-06 11:13:01,540 - Epoch: [113][ 400/ 1200] Overall Loss 0.201672 Objective Loss 0.201672 LR 0.000500 Time 0.021153 -2022-12-06 11:13:01,731 - Epoch: [113][ 410/ 1200] Overall Loss 0.201726 Objective Loss 0.201726 LR 0.000500 Time 0.021101 -2022-12-06 11:13:01,922 - Epoch: [113][ 420/ 1200] Overall Loss 0.201692 Objective Loss 0.201692 LR 0.000500 Time 0.021052 -2022-12-06 11:13:02,113 - Epoch: [113][ 430/ 1200] Overall Loss 0.201602 Objective Loss 0.201602 LR 0.000500 Time 0.021004 -2022-12-06 11:13:02,303 - Epoch: [113][ 440/ 1200] Overall Loss 0.201683 Objective Loss 0.201683 LR 0.000500 Time 0.020959 -2022-12-06 11:13:02,494 - Epoch: [113][ 450/ 1200] Overall Loss 0.201926 Objective Loss 0.201926 LR 0.000500 Time 0.020915 -2022-12-06 11:13:02,685 - Epoch: [113][ 460/ 1200] Overall Loss 0.201888 Objective Loss 0.201888 LR 0.000500 Time 0.020875 -2022-12-06 11:13:02,876 - Epoch: [113][ 470/ 1200] Overall Loss 0.201820 Objective Loss 0.201820 LR 0.000500 Time 0.020836 -2022-12-06 11:13:03,067 - Epoch: [113][ 480/ 1200] Overall Loss 0.202355 Objective Loss 0.202355 LR 0.000500 Time 0.020798 -2022-12-06 11:13:03,257 - Epoch: [113][ 490/ 1200] Overall Loss 0.202379 Objective Loss 0.202379 LR 0.000500 Time 0.020761 -2022-12-06 11:13:03,448 - Epoch: [113][ 500/ 1200] Overall Loss 0.202572 Objective Loss 0.202572 LR 0.000500 Time 0.020727 -2022-12-06 11:13:03,640 - Epoch: [113][ 510/ 1200] Overall Loss 0.202689 Objective Loss 0.202689 LR 0.000500 Time 0.020695 -2022-12-06 11:13:03,831 - Epoch: [113][ 520/ 1200] Overall Loss 0.202737 Objective Loss 0.202737 LR 0.000500 Time 0.020663 -2022-12-06 11:13:04,022 - Epoch: [113][ 530/ 1200] Overall Loss 0.202816 Objective Loss 0.202816 LR 0.000500 Time 0.020632 -2022-12-06 11:13:04,213 - Epoch: [113][ 540/ 1200] Overall Loss 0.203520 Objective Loss 0.203520 LR 0.000500 Time 0.020603 -2022-12-06 11:13:04,403 - Epoch: [113][ 550/ 1200] Overall Loss 0.203863 Objective Loss 0.203863 LR 0.000500 Time 0.020574 -2022-12-06 11:13:04,595 - Epoch: [113][ 560/ 1200] Overall Loss 0.203721 Objective Loss 0.203721 LR 0.000500 Time 0.020548 -2022-12-06 11:13:04,785 - Epoch: [113][ 570/ 1200] Overall Loss 0.203750 Objective Loss 0.203750 LR 0.000500 Time 0.020521 -2022-12-06 11:13:04,977 - Epoch: [113][ 580/ 1200] Overall Loss 0.204103 Objective Loss 0.204103 LR 0.000500 Time 0.020496 -2022-12-06 11:13:05,167 - Epoch: [113][ 590/ 1200] Overall Loss 0.203886 Objective Loss 0.203886 LR 0.000500 Time 0.020470 -2022-12-06 11:13:05,358 - Epoch: [113][ 600/ 1200] Overall Loss 0.204397 Objective Loss 0.204397 LR 0.000500 Time 0.020446 -2022-12-06 11:13:05,549 - Epoch: [113][ 610/ 1200] Overall Loss 0.204142 Objective Loss 0.204142 LR 0.000500 Time 0.020424 -2022-12-06 11:13:05,740 - Epoch: [113][ 620/ 1200] Overall Loss 0.204182 Objective Loss 0.204182 LR 0.000500 Time 0.020401 -2022-12-06 11:13:05,931 - Epoch: [113][ 630/ 1200] Overall Loss 0.203899 Objective Loss 0.203899 LR 0.000500 Time 0.020380 -2022-12-06 11:13:06,122 - Epoch: [113][ 640/ 1200] Overall Loss 0.203663 Objective Loss 0.203663 LR 0.000500 Time 0.020359 -2022-12-06 11:13:06,313 - Epoch: [113][ 650/ 1200] Overall Loss 0.203778 Objective Loss 0.203778 LR 0.000500 Time 0.020339 -2022-12-06 11:13:06,503 - Epoch: [113][ 660/ 1200] Overall Loss 0.203754 Objective Loss 0.203754 LR 0.000500 Time 0.020318 -2022-12-06 11:13:06,694 - Epoch: [113][ 670/ 1200] Overall Loss 0.203892 Objective Loss 0.203892 LR 0.000500 Time 0.020298 -2022-12-06 11:13:06,885 - Epoch: [113][ 680/ 1200] Overall Loss 0.204319 Objective Loss 0.204319 LR 0.000500 Time 0.020279 -2022-12-06 11:13:07,076 - Epoch: [113][ 690/ 1200] Overall Loss 0.204141 Objective Loss 0.204141 LR 0.000500 Time 0.020261 -2022-12-06 11:13:07,266 - Epoch: [113][ 700/ 1200] Overall Loss 0.204291 Objective Loss 0.204291 LR 0.000500 Time 0.020243 -2022-12-06 11:13:07,457 - Epoch: [113][ 710/ 1200] Overall Loss 0.204547 Objective Loss 0.204547 LR 0.000500 Time 0.020227 -2022-12-06 11:13:07,648 - Epoch: [113][ 720/ 1200] Overall Loss 0.204457 Objective Loss 0.204457 LR 0.000500 Time 0.020210 -2022-12-06 11:13:07,839 - Epoch: [113][ 730/ 1200] Overall Loss 0.204292 Objective Loss 0.204292 LR 0.000500 Time 0.020194 -2022-12-06 11:13:08,030 - Epoch: [113][ 740/ 1200] Overall Loss 0.204229 Objective Loss 0.204229 LR 0.000500 Time 0.020178 -2022-12-06 11:13:08,220 - Epoch: [113][ 750/ 1200] Overall Loss 0.204228 Objective Loss 0.204228 LR 0.000500 Time 0.020162 -2022-12-06 11:13:08,410 - Epoch: [113][ 760/ 1200] Overall Loss 0.204036 Objective Loss 0.204036 LR 0.000500 Time 0.020147 -2022-12-06 11:13:08,601 - Epoch: [113][ 770/ 1200] Overall Loss 0.204087 Objective Loss 0.204087 LR 0.000500 Time 0.020132 -2022-12-06 11:13:08,791 - Epoch: [113][ 780/ 1200] Overall Loss 0.204797 Objective Loss 0.204797 LR 0.000500 Time 0.020117 -2022-12-06 11:13:08,982 - Epoch: [113][ 790/ 1200] Overall Loss 0.204792 Objective Loss 0.204792 LR 0.000500 Time 0.020103 -2022-12-06 11:13:09,173 - Epoch: [113][ 800/ 1200] Overall Loss 0.204721 Objective Loss 0.204721 LR 0.000500 Time 0.020090 -2022-12-06 11:13:09,364 - Epoch: [113][ 810/ 1200] Overall Loss 0.204460 Objective Loss 0.204460 LR 0.000500 Time 0.020077 -2022-12-06 11:13:09,555 - Epoch: [113][ 820/ 1200] Overall Loss 0.204595 Objective Loss 0.204595 LR 0.000500 Time 0.020064 -2022-12-06 11:13:09,745 - Epoch: [113][ 830/ 1200] Overall Loss 0.204550 Objective Loss 0.204550 LR 0.000500 Time 0.020051 -2022-12-06 11:13:09,936 - Epoch: [113][ 840/ 1200] Overall Loss 0.204642 Objective Loss 0.204642 LR 0.000500 Time 0.020038 -2022-12-06 11:13:10,127 - Epoch: [113][ 850/ 1200] Overall Loss 0.204613 Objective Loss 0.204613 LR 0.000500 Time 0.020027 -2022-12-06 11:13:10,318 - Epoch: [113][ 860/ 1200] Overall Loss 0.204498 Objective Loss 0.204498 LR 0.000500 Time 0.020016 -2022-12-06 11:13:10,509 - Epoch: [113][ 870/ 1200] Overall Loss 0.204364 Objective Loss 0.204364 LR 0.000500 Time 0.020004 -2022-12-06 11:13:10,699 - Epoch: [113][ 880/ 1200] Overall Loss 0.204244 Objective Loss 0.204244 LR 0.000500 Time 0.019993 -2022-12-06 11:13:10,891 - Epoch: [113][ 890/ 1200] Overall Loss 0.204308 Objective Loss 0.204308 LR 0.000500 Time 0.019983 -2022-12-06 11:13:11,082 - Epoch: [113][ 900/ 1200] Overall Loss 0.204421 Objective Loss 0.204421 LR 0.000500 Time 0.019972 -2022-12-06 11:13:11,273 - Epoch: [113][ 910/ 1200] Overall Loss 0.204406 Objective Loss 0.204406 LR 0.000500 Time 0.019962 -2022-12-06 11:13:11,463 - Epoch: [113][ 920/ 1200] Overall Loss 0.204445 Objective Loss 0.204445 LR 0.000500 Time 0.019951 -2022-12-06 11:13:11,655 - Epoch: [113][ 930/ 1200] Overall Loss 0.204425 Objective Loss 0.204425 LR 0.000500 Time 0.019942 -2022-12-06 11:13:11,846 - Epoch: [113][ 940/ 1200] Overall Loss 0.204281 Objective Loss 0.204281 LR 0.000500 Time 0.019933 -2022-12-06 11:13:12,037 - Epoch: [113][ 950/ 1200] Overall Loss 0.204365 Objective Loss 0.204365 LR 0.000500 Time 0.019923 -2022-12-06 11:13:12,227 - Epoch: [113][ 960/ 1200] Overall Loss 0.204478 Objective Loss 0.204478 LR 0.000500 Time 0.019913 -2022-12-06 11:13:12,417 - Epoch: [113][ 970/ 1200] Overall Loss 0.204520 Objective Loss 0.204520 LR 0.000500 Time 0.019904 -2022-12-06 11:13:12,608 - Epoch: [113][ 980/ 1200] Overall Loss 0.204342 Objective Loss 0.204342 LR 0.000500 Time 0.019895 -2022-12-06 11:13:12,799 - Epoch: [113][ 990/ 1200] Overall Loss 0.204211 Objective Loss 0.204211 LR 0.000500 Time 0.019886 -2022-12-06 11:13:12,991 - Epoch: [113][ 1000/ 1200] Overall Loss 0.204032 Objective Loss 0.204032 LR 0.000500 Time 0.019878 -2022-12-06 11:13:13,181 - Epoch: [113][ 1010/ 1200] Overall Loss 0.203778 Objective Loss 0.203778 LR 0.000500 Time 0.019869 -2022-12-06 11:13:13,372 - Epoch: [113][ 1020/ 1200] Overall Loss 0.203765 Objective Loss 0.203765 LR 0.000500 Time 0.019861 -2022-12-06 11:13:13,563 - Epoch: [113][ 1030/ 1200] Overall Loss 0.203831 Objective Loss 0.203831 LR 0.000500 Time 0.019854 -2022-12-06 11:13:13,754 - Epoch: [113][ 1040/ 1200] Overall Loss 0.204109 Objective Loss 0.204109 LR 0.000500 Time 0.019846 -2022-12-06 11:13:13,945 - Epoch: [113][ 1050/ 1200] Overall Loss 0.204216 Objective Loss 0.204216 LR 0.000500 Time 0.019838 -2022-12-06 11:13:14,136 - Epoch: [113][ 1060/ 1200] Overall Loss 0.204129 Objective Loss 0.204129 LR 0.000500 Time 0.019831 -2022-12-06 11:13:14,327 - Epoch: [113][ 1070/ 1200] Overall Loss 0.204106 Objective Loss 0.204106 LR 0.000500 Time 0.019823 -2022-12-06 11:13:14,518 - Epoch: [113][ 1080/ 1200] Overall Loss 0.204155 Objective Loss 0.204155 LR 0.000500 Time 0.019816 -2022-12-06 11:13:14,709 - Epoch: [113][ 1090/ 1200] Overall Loss 0.203924 Objective Loss 0.203924 LR 0.000500 Time 0.019809 -2022-12-06 11:13:14,900 - Epoch: [113][ 1100/ 1200] Overall Loss 0.203899 Objective Loss 0.203899 LR 0.000500 Time 0.019802 -2022-12-06 11:13:15,091 - Epoch: [113][ 1110/ 1200] Overall Loss 0.204026 Objective Loss 0.204026 LR 0.000500 Time 0.019795 -2022-12-06 11:13:15,281 - Epoch: [113][ 1120/ 1200] Overall Loss 0.204042 Objective Loss 0.204042 LR 0.000500 Time 0.019788 -2022-12-06 11:13:15,472 - Epoch: [113][ 1130/ 1200] Overall Loss 0.204122 Objective Loss 0.204122 LR 0.000500 Time 0.019781 -2022-12-06 11:13:15,663 - Epoch: [113][ 1140/ 1200] Overall Loss 0.203989 Objective Loss 0.203989 LR 0.000500 Time 0.019774 -2022-12-06 11:13:15,854 - Epoch: [113][ 1150/ 1200] Overall Loss 0.203886 Objective Loss 0.203886 LR 0.000500 Time 0.019768 -2022-12-06 11:13:16,045 - Epoch: [113][ 1160/ 1200] Overall Loss 0.204044 Objective Loss 0.204044 LR 0.000500 Time 0.019762 -2022-12-06 11:13:16,236 - Epoch: [113][ 1170/ 1200] Overall Loss 0.204069 Objective Loss 0.204069 LR 0.000500 Time 0.019756 -2022-12-06 11:13:16,427 - Epoch: [113][ 1180/ 1200] Overall Loss 0.204223 Objective Loss 0.204223 LR 0.000500 Time 0.019750 -2022-12-06 11:13:16,618 - Epoch: [113][ 1190/ 1200] Overall Loss 0.204502 Objective Loss 0.204502 LR 0.000500 Time 0.019744 -2022-12-06 11:13:16,847 - Epoch: [113][ 1200/ 1200] Overall Loss 0.204592 Objective Loss 0.204592 Top1 86.610879 Top5 98.117155 LR 0.000500 Time 0.019770 -2022-12-06 11:13:16,936 - --- validate (epoch=113)----------- -2022-12-06 11:13:16,937 - 34129 samples (256 per mini-batch) -2022-12-06 11:13:17,383 - Epoch: [113][ 10/ 134] Loss 0.246825 Top1 86.914062 Top5 97.812500 -2022-12-06 11:13:17,510 - Epoch: [113][ 20/ 134] Loss 0.246373 Top1 86.640625 Top5 97.929688 -2022-12-06 11:13:17,639 - Epoch: [113][ 30/ 134] Loss 0.249328 Top1 86.197917 Top5 97.981771 -2022-12-06 11:13:17,768 - Epoch: [113][ 40/ 134] Loss 0.257934 Top1 86.201172 Top5 98.017578 -2022-12-06 11:13:17,900 - Epoch: [113][ 50/ 134] Loss 0.258123 Top1 86.078125 Top5 98.007812 -2022-12-06 11:13:18,033 - Epoch: [113][ 60/ 134] Loss 0.257430 Top1 86.113281 Top5 97.968750 -2022-12-06 11:13:18,166 - Epoch: [113][ 70/ 134] Loss 0.261516 Top1 86.043527 Top5 98.035714 -2022-12-06 11:13:18,300 - Epoch: [113][ 80/ 134] Loss 0.261769 Top1 86.025391 Top5 98.066406 -2022-12-06 11:13:18,433 - Epoch: [113][ 90/ 134] Loss 0.257816 Top1 86.137153 Top5 98.111979 -2022-12-06 11:13:18,566 - Epoch: [113][ 100/ 134] Loss 0.261993 Top1 86.054688 Top5 98.097656 -2022-12-06 11:13:18,699 - Epoch: [113][ 110/ 134] Loss 0.262332 Top1 86.026278 Top5 98.096591 -2022-12-06 11:13:18,833 - Epoch: [113][ 120/ 134] Loss 0.262516 Top1 86.080729 Top5 98.105469 -2022-12-06 11:13:18,968 - Epoch: [113][ 130/ 134] Loss 0.263525 Top1 86.045673 Top5 98.082933 -2022-12-06 11:13:19,008 - Epoch: [113][ 134/ 134] Loss 0.264676 Top1 85.982595 Top5 98.072021 -2022-12-06 11:13:19,096 - ==> Top1: 85.983 Top5: 98.072 Loss: 0.265 - -2022-12-06 11:13:19,096 - ==> Confusion: -[[ 905 1 3 3 5 4 1 2 4 49 0 2 1 8 3 1 0 0 1 0 3] - [ 1 933 2 1 4 22 7 15 2 1 2 5 1 1 0 1 2 0 15 3 9] - [ 2 3 1003 11 2 1 28 7 0 2 7 5 3 5 4 4 0 0 4 4 8] - [ 4 2 16 938 0 1 0 1 0 1 10 0 5 3 13 0 2 2 14 1 7] - [ 4 4 3 1 958 3 0 2 0 8 0 3 2 4 8 8 5 1 0 4 2] - [ 1 17 0 3 5 976 2 21 2 1 0 11 5 14 0 1 0 1 0 6 3] - [ 1 1 6 1 0 2 1079 2 0 0 1 3 1 1 0 6 0 2 1 8 3] - [ 2 7 6 2 2 26 13 940 1 0 0 6 0 3 0 1 1 0 28 13 3] - [ 6 2 0 0 0 4 1 0 964 39 9 2 3 15 14 1 0 0 3 0 1] - [ 53 1 1 1 9 6 0 2 21 877 1 1 0 19 2 2 0 1 1 0 3] - [ 0 3 1 6 0 4 1 2 9 1 957 3 2 15 4 0 1 0 7 1 2] - [ 5 0 1 0 0 12 3 3 1 0 1 960 32 3 0 6 3 3 0 15 3] - [ 2 0 1 2 1 1 1 0 0 0 1 24 908 2 2 7 1 3 1 6 6] - [ 0 1 1 0 0 3 0 1 8 11 2 3 4 972 0 1 2 1 1 3 9] - [ 3 3 3 8 1 3 0 0 15 2 1 1 4 4 1066 0 0 0 9 1 6] - [ 1 0 0 0 3 1 4 0 0 1 0 7 6 1 0 997 4 8 0 4 6] - [ 2 2 1 0 1 1 1 1 1 0 0 2 3 3 0 19 1022 0 1 5 7] - [ 3 0 1 3 0 2 3 0 3 1 0 10 21 4 3 13 0 962 2 3 2] - [ 4 5 3 9 1 2 2 24 3 1 5 0 2 1 7 1 0 0 933 2 3] - [ 4 5 2 1 1 4 3 9 0 0 0 7 6 7 0 2 2 1 0 1021 5] - [ 119 198 166 111 118 198 97 149 70 88 186 92 355 334 177 123 149 59 199 266 9972]] - -2022-12-06 11:13:19,674 - ==> Best [Top1: 86.255 Top5: 98.139 Sparsity:0.00 Params: 5376 on epoch: 107] -2022-12-06 11:13:19,674 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:13:19,681 - - -2022-12-06 11:13:19,681 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:13:20,624 - Epoch: [114][ 10/ 1200] Overall Loss 0.199524 Objective Loss 0.199524 LR 0.000500 Time 0.094284 -2022-12-06 11:13:20,828 - Epoch: [114][ 20/ 1200] Overall Loss 0.203469 Objective Loss 0.203469 LR 0.000500 Time 0.057287 -2022-12-06 11:13:21,028 - Epoch: [114][ 30/ 1200] Overall Loss 0.212339 Objective Loss 0.212339 LR 0.000500 Time 0.044845 -2022-12-06 11:13:21,224 - Epoch: [114][ 40/ 1200] Overall Loss 0.209923 Objective Loss 0.209923 LR 0.000500 Time 0.038520 -2022-12-06 11:13:21,423 - Epoch: [114][ 50/ 1200] Overall Loss 0.206365 Objective Loss 0.206365 LR 0.000500 Time 0.034783 -2022-12-06 11:13:21,619 - Epoch: [114][ 60/ 1200] Overall Loss 0.205579 Objective Loss 0.205579 LR 0.000500 Time 0.032248 -2022-12-06 11:13:21,813 - Epoch: [114][ 70/ 1200] Overall Loss 0.204248 Objective Loss 0.204248 LR 0.000500 Time 0.030396 -2022-12-06 11:13:22,004 - Epoch: [114][ 80/ 1200] Overall Loss 0.204717 Objective Loss 0.204717 LR 0.000500 Time 0.028985 -2022-12-06 11:13:22,196 - Epoch: [114][ 90/ 1200] Overall Loss 0.200769 Objective Loss 0.200769 LR 0.000500 Time 0.027892 -2022-12-06 11:13:22,388 - Epoch: [114][ 100/ 1200] Overall Loss 0.199585 Objective Loss 0.199585 LR 0.000500 Time 0.027014 -2022-12-06 11:13:22,579 - Epoch: [114][ 110/ 1200] Overall Loss 0.201433 Objective Loss 0.201433 LR 0.000500 Time 0.026294 -2022-12-06 11:13:22,771 - Epoch: [114][ 120/ 1200] Overall Loss 0.201470 Objective Loss 0.201470 LR 0.000500 Time 0.025695 -2022-12-06 11:13:22,963 - Epoch: [114][ 130/ 1200] Overall Loss 0.201528 Objective Loss 0.201528 LR 0.000500 Time 0.025189 -2022-12-06 11:13:23,155 - Epoch: [114][ 140/ 1200] Overall Loss 0.200440 Objective Loss 0.200440 LR 0.000500 Time 0.024758 -2022-12-06 11:13:23,347 - Epoch: [114][ 150/ 1200] Overall Loss 0.200638 Objective Loss 0.200638 LR 0.000500 Time 0.024384 -2022-12-06 11:13:23,538 - Epoch: [114][ 160/ 1200] Overall Loss 0.200838 Objective Loss 0.200838 LR 0.000500 Time 0.024053 -2022-12-06 11:13:23,731 - Epoch: [114][ 170/ 1200] Overall Loss 0.200977 Objective Loss 0.200977 LR 0.000500 Time 0.023766 -2022-12-06 11:13:23,922 - Epoch: [114][ 180/ 1200] Overall Loss 0.200691 Objective Loss 0.200691 LR 0.000500 Time 0.023508 -2022-12-06 11:13:24,114 - Epoch: [114][ 190/ 1200] Overall Loss 0.200576 Objective Loss 0.200576 LR 0.000500 Time 0.023276 -2022-12-06 11:13:24,305 - Epoch: [114][ 200/ 1200] Overall Loss 0.200356 Objective Loss 0.200356 LR 0.000500 Time 0.023067 -2022-12-06 11:13:24,497 - Epoch: [114][ 210/ 1200] Overall Loss 0.201096 Objective Loss 0.201096 LR 0.000500 Time 0.022877 -2022-12-06 11:13:24,688 - Epoch: [114][ 220/ 1200] Overall Loss 0.201527 Objective Loss 0.201527 LR 0.000500 Time 0.022705 -2022-12-06 11:13:24,880 - Epoch: [114][ 230/ 1200] Overall Loss 0.202411 Objective Loss 0.202411 LR 0.000500 Time 0.022550 -2022-12-06 11:13:25,072 - Epoch: [114][ 240/ 1200] Overall Loss 0.202500 Objective Loss 0.202500 LR 0.000500 Time 0.022409 -2022-12-06 11:13:25,264 - Epoch: [114][ 250/ 1200] Overall Loss 0.201689 Objective Loss 0.201689 LR 0.000500 Time 0.022277 -2022-12-06 11:13:25,455 - Epoch: [114][ 260/ 1200] Overall Loss 0.202103 Objective Loss 0.202103 LR 0.000500 Time 0.022155 -2022-12-06 11:13:25,647 - Epoch: [114][ 270/ 1200] Overall Loss 0.201714 Objective Loss 0.201714 LR 0.000500 Time 0.022042 -2022-12-06 11:13:25,839 - Epoch: [114][ 280/ 1200] Overall Loss 0.202119 Objective Loss 0.202119 LR 0.000500 Time 0.021938 -2022-12-06 11:13:26,030 - Epoch: [114][ 290/ 1200] Overall Loss 0.202622 Objective Loss 0.202622 LR 0.000500 Time 0.021839 -2022-12-06 11:13:26,222 - Epoch: [114][ 300/ 1200] Overall Loss 0.202116 Objective Loss 0.202116 LR 0.000500 Time 0.021748 -2022-12-06 11:13:26,413 - Epoch: [114][ 310/ 1200] Overall Loss 0.201746 Objective Loss 0.201746 LR 0.000500 Time 0.021662 -2022-12-06 11:13:26,605 - Epoch: [114][ 320/ 1200] Overall Loss 0.201353 Objective Loss 0.201353 LR 0.000500 Time 0.021582 -2022-12-06 11:13:26,797 - Epoch: [114][ 330/ 1200] Overall Loss 0.200945 Objective Loss 0.200945 LR 0.000500 Time 0.021509 -2022-12-06 11:13:26,988 - Epoch: [114][ 340/ 1200] Overall Loss 0.201262 Objective Loss 0.201262 LR 0.000500 Time 0.021438 -2022-12-06 11:13:27,180 - Epoch: [114][ 350/ 1200] Overall Loss 0.202348 Objective Loss 0.202348 LR 0.000500 Time 0.021371 -2022-12-06 11:13:27,372 - Epoch: [114][ 360/ 1200] Overall Loss 0.202098 Objective Loss 0.202098 LR 0.000500 Time 0.021309 -2022-12-06 11:13:27,564 - Epoch: [114][ 370/ 1200] Overall Loss 0.201955 Objective Loss 0.201955 LR 0.000500 Time 0.021250 -2022-12-06 11:13:27,755 - Epoch: [114][ 380/ 1200] Overall Loss 0.201981 Objective Loss 0.201981 LR 0.000500 Time 0.021194 -2022-12-06 11:13:27,947 - Epoch: [114][ 390/ 1200] Overall Loss 0.202512 Objective Loss 0.202512 LR 0.000500 Time 0.021141 -2022-12-06 11:13:28,139 - Epoch: [114][ 400/ 1200] Overall Loss 0.202490 Objective Loss 0.202490 LR 0.000500 Time 0.021090 -2022-12-06 11:13:28,330 - Epoch: [114][ 410/ 1200] Overall Loss 0.203029 Objective Loss 0.203029 LR 0.000500 Time 0.021040 -2022-12-06 11:13:28,521 - Epoch: [114][ 420/ 1200] Overall Loss 0.203157 Objective Loss 0.203157 LR 0.000500 Time 0.020994 -2022-12-06 11:13:28,713 - Epoch: [114][ 430/ 1200] Overall Loss 0.203458 Objective Loss 0.203458 LR 0.000500 Time 0.020951 -2022-12-06 11:13:28,905 - Epoch: [114][ 440/ 1200] Overall Loss 0.203110 Objective Loss 0.203110 LR 0.000500 Time 0.020908 -2022-12-06 11:13:29,096 - Epoch: [114][ 450/ 1200] Overall Loss 0.202761 Objective Loss 0.202761 LR 0.000500 Time 0.020869 -2022-12-06 11:13:29,288 - Epoch: [114][ 460/ 1200] Overall Loss 0.202814 Objective Loss 0.202814 LR 0.000500 Time 0.020830 -2022-12-06 11:13:29,479 - Epoch: [114][ 470/ 1200] Overall Loss 0.203036 Objective Loss 0.203036 LR 0.000500 Time 0.020793 -2022-12-06 11:13:29,671 - Epoch: [114][ 480/ 1200] Overall Loss 0.202973 Objective Loss 0.202973 LR 0.000500 Time 0.020757 -2022-12-06 11:13:29,863 - Epoch: [114][ 490/ 1200] Overall Loss 0.202809 Objective Loss 0.202809 LR 0.000500 Time 0.020724 -2022-12-06 11:13:30,054 - Epoch: [114][ 500/ 1200] Overall Loss 0.202299 Objective Loss 0.202299 LR 0.000500 Time 0.020690 -2022-12-06 11:13:30,245 - Epoch: [114][ 510/ 1200] Overall Loss 0.202274 Objective Loss 0.202274 LR 0.000500 Time 0.020659 -2022-12-06 11:13:30,436 - Epoch: [114][ 520/ 1200] Overall Loss 0.202224 Objective Loss 0.202224 LR 0.000500 Time 0.020628 -2022-12-06 11:13:30,627 - Epoch: [114][ 530/ 1200] Overall Loss 0.202673 Objective Loss 0.202673 LR 0.000500 Time 0.020599 -2022-12-06 11:13:30,819 - Epoch: [114][ 540/ 1200] Overall Loss 0.203164 Objective Loss 0.203164 LR 0.000500 Time 0.020571 -2022-12-06 11:13:31,010 - Epoch: [114][ 550/ 1200] Overall Loss 0.203058 Objective Loss 0.203058 LR 0.000500 Time 0.020544 -2022-12-06 11:13:31,201 - Epoch: [114][ 560/ 1200] Overall Loss 0.202951 Objective Loss 0.202951 LR 0.000500 Time 0.020517 -2022-12-06 11:13:31,393 - Epoch: [114][ 570/ 1200] Overall Loss 0.203144 Objective Loss 0.203144 LR 0.000500 Time 0.020493 -2022-12-06 11:13:31,584 - Epoch: [114][ 580/ 1200] Overall Loss 0.202785 Objective Loss 0.202785 LR 0.000500 Time 0.020467 -2022-12-06 11:13:31,776 - Epoch: [114][ 590/ 1200] Overall Loss 0.202898 Objective Loss 0.202898 LR 0.000500 Time 0.020444 -2022-12-06 11:13:31,967 - Epoch: [114][ 600/ 1200] Overall Loss 0.202741 Objective Loss 0.202741 LR 0.000500 Time 0.020422 -2022-12-06 11:13:32,159 - Epoch: [114][ 610/ 1200] Overall Loss 0.202812 Objective Loss 0.202812 LR 0.000500 Time 0.020400 -2022-12-06 11:13:32,350 - Epoch: [114][ 620/ 1200] Overall Loss 0.202529 Objective Loss 0.202529 LR 0.000500 Time 0.020379 -2022-12-06 11:13:32,542 - Epoch: [114][ 630/ 1200] Overall Loss 0.202525 Objective Loss 0.202525 LR 0.000500 Time 0.020360 -2022-12-06 11:13:32,733 - Epoch: [114][ 640/ 1200] Overall Loss 0.202654 Objective Loss 0.202654 LR 0.000500 Time 0.020339 -2022-12-06 11:13:32,925 - Epoch: [114][ 650/ 1200] Overall Loss 0.202815 Objective Loss 0.202815 LR 0.000500 Time 0.020321 -2022-12-06 11:13:33,116 - Epoch: [114][ 660/ 1200] Overall Loss 0.203070 Objective Loss 0.203070 LR 0.000500 Time 0.020301 -2022-12-06 11:13:33,308 - Epoch: [114][ 670/ 1200] Overall Loss 0.202646 Objective Loss 0.202646 LR 0.000500 Time 0.020284 -2022-12-06 11:13:33,499 - Epoch: [114][ 680/ 1200] Overall Loss 0.202614 Objective Loss 0.202614 LR 0.000500 Time 0.020266 -2022-12-06 11:13:33,691 - Epoch: [114][ 690/ 1200] Overall Loss 0.202689 Objective Loss 0.202689 LR 0.000500 Time 0.020250 -2022-12-06 11:13:33,883 - Epoch: [114][ 700/ 1200] Overall Loss 0.202757 Objective Loss 0.202757 LR 0.000500 Time 0.020233 -2022-12-06 11:13:34,074 - Epoch: [114][ 710/ 1200] Overall Loss 0.202468 Objective Loss 0.202468 LR 0.000500 Time 0.020217 -2022-12-06 11:13:34,265 - Epoch: [114][ 720/ 1200] Overall Loss 0.202616 Objective Loss 0.202616 LR 0.000500 Time 0.020200 -2022-12-06 11:13:34,456 - Epoch: [114][ 730/ 1200] Overall Loss 0.202443 Objective Loss 0.202443 LR 0.000500 Time 0.020185 -2022-12-06 11:13:34,648 - Epoch: [114][ 740/ 1200] Overall Loss 0.202198 Objective Loss 0.202198 LR 0.000500 Time 0.020170 -2022-12-06 11:13:34,839 - Epoch: [114][ 750/ 1200] Overall Loss 0.202315 Objective Loss 0.202315 LR 0.000500 Time 0.020156 -2022-12-06 11:13:35,030 - Epoch: [114][ 760/ 1200] Overall Loss 0.202102 Objective Loss 0.202102 LR 0.000500 Time 0.020141 -2022-12-06 11:13:35,222 - Epoch: [114][ 770/ 1200] Overall Loss 0.201893 Objective Loss 0.201893 LR 0.000500 Time 0.020128 -2022-12-06 11:13:35,413 - Epoch: [114][ 780/ 1200] Overall Loss 0.201993 Objective Loss 0.201993 LR 0.000500 Time 0.020114 -2022-12-06 11:13:35,605 - Epoch: [114][ 790/ 1200] Overall Loss 0.202268 Objective Loss 0.202268 LR 0.000500 Time 0.020102 -2022-12-06 11:13:35,796 - Epoch: [114][ 800/ 1200] Overall Loss 0.202222 Objective Loss 0.202222 LR 0.000500 Time 0.020089 -2022-12-06 11:13:35,988 - Epoch: [114][ 810/ 1200] Overall Loss 0.202132 Objective Loss 0.202132 LR 0.000500 Time 0.020077 -2022-12-06 11:13:36,179 - Epoch: [114][ 820/ 1200] Overall Loss 0.202176 Objective Loss 0.202176 LR 0.000500 Time 0.020065 -2022-12-06 11:13:36,371 - Epoch: [114][ 830/ 1200] Overall Loss 0.201975 Objective Loss 0.201975 LR 0.000500 Time 0.020053 -2022-12-06 11:13:36,562 - Epoch: [114][ 840/ 1200] Overall Loss 0.202293 Objective Loss 0.202293 LR 0.000500 Time 0.020041 -2022-12-06 11:13:36,753 - Epoch: [114][ 850/ 1200] Overall Loss 0.202558 Objective Loss 0.202558 LR 0.000500 Time 0.020030 -2022-12-06 11:13:36,944 - Epoch: [114][ 860/ 1200] Overall Loss 0.202401 Objective Loss 0.202401 LR 0.000500 Time 0.020019 -2022-12-06 11:13:37,136 - Epoch: [114][ 870/ 1200] Overall Loss 0.202590 Objective Loss 0.202590 LR 0.000500 Time 0.020008 -2022-12-06 11:13:37,327 - Epoch: [114][ 880/ 1200] Overall Loss 0.202557 Objective Loss 0.202557 LR 0.000500 Time 0.019998 -2022-12-06 11:13:37,519 - Epoch: [114][ 890/ 1200] Overall Loss 0.202723 Objective Loss 0.202723 LR 0.000500 Time 0.019988 -2022-12-06 11:13:37,710 - Epoch: [114][ 900/ 1200] Overall Loss 0.202564 Objective Loss 0.202564 LR 0.000500 Time 0.019977 -2022-12-06 11:13:37,901 - Epoch: [114][ 910/ 1200] Overall Loss 0.202447 Objective Loss 0.202447 LR 0.000500 Time 0.019968 -2022-12-06 11:13:38,092 - Epoch: [114][ 920/ 1200] Overall Loss 0.202681 Objective Loss 0.202681 LR 0.000500 Time 0.019957 -2022-12-06 11:13:38,284 - Epoch: [114][ 930/ 1200] Overall Loss 0.202797 Objective Loss 0.202797 LR 0.000500 Time 0.019948 -2022-12-06 11:13:38,475 - Epoch: [114][ 940/ 1200] Overall Loss 0.202896 Objective Loss 0.202896 LR 0.000500 Time 0.019939 -2022-12-06 11:13:38,666 - Epoch: [114][ 950/ 1200] Overall Loss 0.202832 Objective Loss 0.202832 LR 0.000500 Time 0.019930 -2022-12-06 11:13:38,857 - Epoch: [114][ 960/ 1200] Overall Loss 0.202831 Objective Loss 0.202831 LR 0.000500 Time 0.019921 -2022-12-06 11:13:39,049 - Epoch: [114][ 970/ 1200] Overall Loss 0.202871 Objective Loss 0.202871 LR 0.000500 Time 0.019913 -2022-12-06 11:13:39,240 - Epoch: [114][ 980/ 1200] Overall Loss 0.202855 Objective Loss 0.202855 LR 0.000500 Time 0.019904 -2022-12-06 11:13:39,432 - Epoch: [114][ 990/ 1200] Overall Loss 0.202914 Objective Loss 0.202914 LR 0.000500 Time 0.019896 -2022-12-06 11:13:39,624 - Epoch: [114][ 1000/ 1200] Overall Loss 0.202970 Objective Loss 0.202970 LR 0.000500 Time 0.019888 -2022-12-06 11:13:39,816 - Epoch: [114][ 1010/ 1200] Overall Loss 0.203136 Objective Loss 0.203136 LR 0.000500 Time 0.019881 -2022-12-06 11:13:40,008 - Epoch: [114][ 1020/ 1200] Overall Loss 0.203313 Objective Loss 0.203313 LR 0.000500 Time 0.019874 -2022-12-06 11:13:40,200 - Epoch: [114][ 1030/ 1200] Overall Loss 0.203072 Objective Loss 0.203072 LR 0.000500 Time 0.019866 -2022-12-06 11:13:40,392 - Epoch: [114][ 1040/ 1200] Overall Loss 0.203115 Objective Loss 0.203115 LR 0.000500 Time 0.019860 -2022-12-06 11:13:40,584 - Epoch: [114][ 1050/ 1200] Overall Loss 0.203060 Objective Loss 0.203060 LR 0.000500 Time 0.019853 -2022-12-06 11:13:40,776 - Epoch: [114][ 1060/ 1200] Overall Loss 0.203157 Objective Loss 0.203157 LR 0.000500 Time 0.019846 -2022-12-06 11:13:40,968 - Epoch: [114][ 1070/ 1200] Overall Loss 0.202964 Objective Loss 0.202964 LR 0.000500 Time 0.019839 -2022-12-06 11:13:41,160 - Epoch: [114][ 1080/ 1200] Overall Loss 0.202899 Objective Loss 0.202899 LR 0.000500 Time 0.019833 -2022-12-06 11:13:41,352 - Epoch: [114][ 1090/ 1200] Overall Loss 0.203033 Objective Loss 0.203033 LR 0.000500 Time 0.019827 -2022-12-06 11:13:41,543 - Epoch: [114][ 1100/ 1200] Overall Loss 0.203173 Objective Loss 0.203173 LR 0.000500 Time 0.019820 -2022-12-06 11:13:41,736 - Epoch: [114][ 1110/ 1200] Overall Loss 0.203145 Objective Loss 0.203145 LR 0.000500 Time 0.019814 -2022-12-06 11:13:41,928 - Epoch: [114][ 1120/ 1200] Overall Loss 0.203384 Objective Loss 0.203384 LR 0.000500 Time 0.019809 -2022-12-06 11:13:42,120 - Epoch: [114][ 1130/ 1200] Overall Loss 0.203510 Objective Loss 0.203510 LR 0.000500 Time 0.019803 -2022-12-06 11:13:42,312 - Epoch: [114][ 1140/ 1200] Overall Loss 0.203554 Objective Loss 0.203554 LR 0.000500 Time 0.019797 -2022-12-06 11:13:42,504 - Epoch: [114][ 1150/ 1200] Overall Loss 0.203572 Objective Loss 0.203572 LR 0.000500 Time 0.019791 -2022-12-06 11:13:42,695 - Epoch: [114][ 1160/ 1200] Overall Loss 0.203402 Objective Loss 0.203402 LR 0.000500 Time 0.019785 -2022-12-06 11:13:42,887 - Epoch: [114][ 1170/ 1200] Overall Loss 0.203295 Objective Loss 0.203295 LR 0.000500 Time 0.019780 -2022-12-06 11:13:43,079 - Epoch: [114][ 1180/ 1200] Overall Loss 0.203346 Objective Loss 0.203346 LR 0.000500 Time 0.019774 -2022-12-06 11:13:43,272 - Epoch: [114][ 1190/ 1200] Overall Loss 0.203307 Objective Loss 0.203307 LR 0.000500 Time 0.019769 -2022-12-06 11:13:43,498 - Epoch: [114][ 1200/ 1200] Overall Loss 0.203522 Objective Loss 0.203522 Top1 85.564854 Top5 98.744770 LR 0.000500 Time 0.019792 -2022-12-06 11:13:43,586 - --- validate (epoch=114)----------- -2022-12-06 11:13:43,587 - 34129 samples (256 per mini-batch) -2022-12-06 11:13:44,039 - Epoch: [114][ 10/ 134] Loss 0.242467 Top1 86.171875 Top5 97.773438 -2022-12-06 11:13:44,175 - Epoch: [114][ 20/ 134] Loss 0.254036 Top1 86.035156 Top5 98.027344 -2022-12-06 11:13:44,310 - Epoch: [114][ 30/ 134] Loss 0.256888 Top1 85.911458 Top5 98.007812 -2022-12-06 11:13:44,442 - Epoch: [114][ 40/ 134] Loss 0.251747 Top1 86.367188 Top5 98.144531 -2022-12-06 11:13:44,577 - Epoch: [114][ 50/ 134] Loss 0.250873 Top1 86.382812 Top5 98.140625 -2022-12-06 11:13:44,716 - Epoch: [114][ 60/ 134] Loss 0.252648 Top1 86.328125 Top5 98.138021 -2022-12-06 11:13:44,849 - Epoch: [114][ 70/ 134] Loss 0.255661 Top1 86.428571 Top5 98.035714 -2022-12-06 11:13:44,984 - Epoch: [114][ 80/ 134] Loss 0.258381 Top1 86.318359 Top5 98.071289 -2022-12-06 11:13:45,118 - Epoch: [114][ 90/ 134] Loss 0.257106 Top1 86.267361 Top5 98.064236 -2022-12-06 11:13:45,254 - Epoch: [114][ 100/ 134] Loss 0.255072 Top1 86.324219 Top5 98.117188 -2022-12-06 11:13:45,387 - Epoch: [114][ 110/ 134] Loss 0.254835 Top1 86.289062 Top5 98.139205 -2022-12-06 11:13:45,526 - Epoch: [114][ 120/ 134] Loss 0.257619 Top1 86.328125 Top5 98.147786 -2022-12-06 11:13:45,658 - Epoch: [114][ 130/ 134] Loss 0.257227 Top1 86.400240 Top5 98.164062 -2022-12-06 11:13:45,695 - Epoch: [114][ 134/ 134] Loss 0.258129 Top1 86.448475 Top5 98.171643 -2022-12-06 11:13:45,785 - ==> Top1: 86.448 Top5: 98.172 Loss: 0.258 - -2022-12-06 11:13:45,786 - ==> Confusion: -[[ 907 1 1 2 3 4 0 0 5 57 0 4 0 1 3 3 1 0 0 1 3] - [ 1 933 2 3 8 30 1 15 2 1 0 5 1 2 1 1 4 1 5 2 9] - [ 6 4 1002 12 4 2 19 7 0 5 5 9 0 2 5 2 2 1 3 1 12] - [ 4 1 13 953 1 0 0 3 0 2 7 0 4 1 10 2 2 2 11 0 4] - [ 9 6 1 0 948 8 0 0 1 9 1 2 0 2 11 9 7 2 0 1 3] - [ 4 12 0 1 5 980 1 18 2 3 0 15 3 10 1 1 1 3 0 5 4] - [ 1 6 8 4 1 4 1058 5 0 0 1 1 3 1 0 8 1 0 0 12 4] - [ 2 2 10 3 3 35 4 952 0 0 1 6 0 1 0 2 1 0 19 11 2] - [ 4 3 0 1 1 1 1 1 983 47 4 1 2 4 7 1 2 0 0 1 0] - [ 42 0 2 0 4 0 0 1 16 912 1 2 0 12 2 2 0 0 0 1 4] - [ 1 1 2 4 1 1 1 4 11 1 959 1 2 15 3 1 0 0 3 2 6] - [ 3 2 3 1 1 12 2 2 1 0 1 985 13 2 0 6 4 3 0 9 1] - [ 2 1 1 1 0 0 0 0 0 1 0 46 884 2 2 12 0 8 0 3 6] - [ 0 1 0 0 0 6 0 2 12 24 3 4 2 958 1 1 4 0 0 1 4] - [ 9 2 3 6 4 2 0 0 19 2 1 2 2 4 1060 0 1 1 5 0 7] - [ 2 1 0 0 2 1 1 0 0 0 0 13 1 2 0 1000 5 8 0 3 4] - [ 3 3 0 2 4 3 2 0 0 0 0 2 1 3 1 11 1031 0 0 4 2] - [ 2 1 1 4 0 2 1 2 3 3 0 18 13 2 1 20 2 958 0 0 3] - [ 4 6 5 7 1 3 1 25 1 2 8 1 2 2 9 1 0 0 923 2 5] - [ 3 4 2 0 1 4 4 6 0 0 0 14 8 4 0 6 5 3 0 1012 4] - [ 136 205 143 88 122 206 64 138 101 112 170 137 309 293 141 142 203 62 123 235 10096]] - -2022-12-06 11:13:46,458 - ==> Best [Top1: 86.448 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 114] -2022-12-06 11:13:46,458 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:13:46,465 - - -2022-12-06 11:13:46,465 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:13:47,396 - Epoch: [115][ 10/ 1200] Overall Loss 0.228093 Objective Loss 0.228093 LR 0.000500 Time 0.093085 -2022-12-06 11:13:47,588 - Epoch: [115][ 20/ 1200] Overall Loss 0.215168 Objective Loss 0.215168 LR 0.000500 Time 0.056082 -2022-12-06 11:13:47,780 - Epoch: [115][ 30/ 1200] Overall Loss 0.215105 Objective Loss 0.215105 LR 0.000500 Time 0.043768 -2022-12-06 11:13:47,971 - Epoch: [115][ 40/ 1200] Overall Loss 0.211821 Objective Loss 0.211821 LR 0.000500 Time 0.037605 -2022-12-06 11:13:48,162 - Epoch: [115][ 50/ 1200] Overall Loss 0.208497 Objective Loss 0.208497 LR 0.000500 Time 0.033895 -2022-12-06 11:13:48,354 - Epoch: [115][ 60/ 1200] Overall Loss 0.205404 Objective Loss 0.205404 LR 0.000500 Time 0.031423 -2022-12-06 11:13:48,544 - Epoch: [115][ 70/ 1200] Overall Loss 0.199845 Objective Loss 0.199845 LR 0.000500 Time 0.029645 -2022-12-06 11:13:48,736 - Epoch: [115][ 80/ 1200] Overall Loss 0.199762 Objective Loss 0.199762 LR 0.000500 Time 0.028327 -2022-12-06 11:13:48,926 - Epoch: [115][ 90/ 1200] Overall Loss 0.199840 Objective Loss 0.199840 LR 0.000500 Time 0.027292 -2022-12-06 11:13:49,117 - Epoch: [115][ 100/ 1200] Overall Loss 0.199379 Objective Loss 0.199379 LR 0.000500 Time 0.026466 -2022-12-06 11:13:49,308 - Epoch: [115][ 110/ 1200] Overall Loss 0.200061 Objective Loss 0.200061 LR 0.000500 Time 0.025787 -2022-12-06 11:13:49,499 - Epoch: [115][ 120/ 1200] Overall Loss 0.200647 Objective Loss 0.200647 LR 0.000500 Time 0.025225 -2022-12-06 11:13:49,690 - Epoch: [115][ 130/ 1200] Overall Loss 0.200971 Objective Loss 0.200971 LR 0.000500 Time 0.024750 -2022-12-06 11:13:49,881 - Epoch: [115][ 140/ 1200] Overall Loss 0.201784 Objective Loss 0.201784 LR 0.000500 Time 0.024344 -2022-12-06 11:13:50,071 - Epoch: [115][ 150/ 1200] Overall Loss 0.201560 Objective Loss 0.201560 LR 0.000500 Time 0.023985 -2022-12-06 11:13:50,261 - Epoch: [115][ 160/ 1200] Overall Loss 0.200755 Objective Loss 0.200755 LR 0.000500 Time 0.023672 -2022-12-06 11:13:50,453 - Epoch: [115][ 170/ 1200] Overall Loss 0.200306 Objective Loss 0.200306 LR 0.000500 Time 0.023404 -2022-12-06 11:13:50,644 - Epoch: [115][ 180/ 1200] Overall Loss 0.199936 Objective Loss 0.199936 LR 0.000500 Time 0.023161 -2022-12-06 11:13:50,834 - Epoch: [115][ 190/ 1200] Overall Loss 0.200627 Objective Loss 0.200627 LR 0.000500 Time 0.022941 -2022-12-06 11:13:51,025 - Epoch: [115][ 200/ 1200] Overall Loss 0.201442 Objective Loss 0.201442 LR 0.000500 Time 0.022748 -2022-12-06 11:13:51,216 - Epoch: [115][ 210/ 1200] Overall Loss 0.202208 Objective Loss 0.202208 LR 0.000500 Time 0.022569 -2022-12-06 11:13:51,407 - Epoch: [115][ 220/ 1200] Overall Loss 0.202530 Objective Loss 0.202530 LR 0.000500 Time 0.022410 -2022-12-06 11:13:51,598 - Epoch: [115][ 230/ 1200] Overall Loss 0.202845 Objective Loss 0.202845 LR 0.000500 Time 0.022263 -2022-12-06 11:13:51,789 - Epoch: [115][ 240/ 1200] Overall Loss 0.202598 Objective Loss 0.202598 LR 0.000500 Time 0.022129 -2022-12-06 11:13:51,980 - Epoch: [115][ 250/ 1200] Overall Loss 0.202008 Objective Loss 0.202008 LR 0.000500 Time 0.022005 -2022-12-06 11:13:52,171 - Epoch: [115][ 260/ 1200] Overall Loss 0.200950 Objective Loss 0.200950 LR 0.000500 Time 0.021890 -2022-12-06 11:13:52,361 - Epoch: [115][ 270/ 1200] Overall Loss 0.201154 Objective Loss 0.201154 LR 0.000500 Time 0.021784 -2022-12-06 11:13:52,552 - Epoch: [115][ 280/ 1200] Overall Loss 0.201386 Objective Loss 0.201386 LR 0.000500 Time 0.021687 -2022-12-06 11:13:52,744 - Epoch: [115][ 290/ 1200] Overall Loss 0.202052 Objective Loss 0.202052 LR 0.000500 Time 0.021597 -2022-12-06 11:13:52,935 - Epoch: [115][ 300/ 1200] Overall Loss 0.201731 Objective Loss 0.201731 LR 0.000500 Time 0.021512 -2022-12-06 11:13:53,126 - Epoch: [115][ 310/ 1200] Overall Loss 0.201813 Objective Loss 0.201813 LR 0.000500 Time 0.021432 -2022-12-06 11:13:53,316 - Epoch: [115][ 320/ 1200] Overall Loss 0.202614 Objective Loss 0.202614 LR 0.000500 Time 0.021356 -2022-12-06 11:13:53,507 - Epoch: [115][ 330/ 1200] Overall Loss 0.202142 Objective Loss 0.202142 LR 0.000500 Time 0.021287 -2022-12-06 11:13:53,700 - Epoch: [115][ 340/ 1200] Overall Loss 0.201991 Objective Loss 0.201991 LR 0.000500 Time 0.021227 -2022-12-06 11:13:53,893 - Epoch: [115][ 350/ 1200] Overall Loss 0.201317 Objective Loss 0.201317 LR 0.000500 Time 0.021167 -2022-12-06 11:13:54,086 - Epoch: [115][ 360/ 1200] Overall Loss 0.200527 Objective Loss 0.200527 LR 0.000500 Time 0.021115 -2022-12-06 11:13:54,279 - Epoch: [115][ 370/ 1200] Overall Loss 0.200797 Objective Loss 0.200797 LR 0.000500 Time 0.021064 -2022-12-06 11:13:54,472 - Epoch: [115][ 380/ 1200] Overall Loss 0.201003 Objective Loss 0.201003 LR 0.000500 Time 0.021016 -2022-12-06 11:13:54,664 - Epoch: [115][ 390/ 1200] Overall Loss 0.200981 Objective Loss 0.200981 LR 0.000500 Time 0.020969 -2022-12-06 11:13:54,857 - Epoch: [115][ 400/ 1200] Overall Loss 0.200461 Objective Loss 0.200461 LR 0.000500 Time 0.020926 -2022-12-06 11:13:55,050 - Epoch: [115][ 410/ 1200] Overall Loss 0.199852 Objective Loss 0.199852 LR 0.000500 Time 0.020885 -2022-12-06 11:13:55,243 - Epoch: [115][ 420/ 1200] Overall Loss 0.199938 Objective Loss 0.199938 LR 0.000500 Time 0.020846 -2022-12-06 11:13:55,436 - Epoch: [115][ 430/ 1200] Overall Loss 0.200172 Objective Loss 0.200172 LR 0.000500 Time 0.020808 -2022-12-06 11:13:55,628 - Epoch: [115][ 440/ 1200] Overall Loss 0.200592 Objective Loss 0.200592 LR 0.000500 Time 0.020772 -2022-12-06 11:13:55,821 - Epoch: [115][ 450/ 1200] Overall Loss 0.200128 Objective Loss 0.200128 LR 0.000500 Time 0.020738 -2022-12-06 11:13:56,015 - Epoch: [115][ 460/ 1200] Overall Loss 0.199787 Objective Loss 0.199787 LR 0.000500 Time 0.020706 -2022-12-06 11:13:56,208 - Epoch: [115][ 470/ 1200] Overall Loss 0.199615 Objective Loss 0.199615 LR 0.000500 Time 0.020675 -2022-12-06 11:13:56,401 - Epoch: [115][ 480/ 1200] Overall Loss 0.199527 Objective Loss 0.199527 LR 0.000500 Time 0.020645 -2022-12-06 11:13:56,594 - Epoch: [115][ 490/ 1200] Overall Loss 0.199452 Objective Loss 0.199452 LR 0.000500 Time 0.020617 -2022-12-06 11:13:56,787 - Epoch: [115][ 500/ 1200] Overall Loss 0.199729 Objective Loss 0.199729 LR 0.000500 Time 0.020589 -2022-12-06 11:13:56,979 - Epoch: [115][ 510/ 1200] Overall Loss 0.199612 Objective Loss 0.199612 LR 0.000500 Time 0.020563 -2022-12-06 11:13:57,172 - Epoch: [115][ 520/ 1200] Overall Loss 0.199258 Objective Loss 0.199258 LR 0.000500 Time 0.020537 -2022-12-06 11:13:57,365 - Epoch: [115][ 530/ 1200] Overall Loss 0.199841 Objective Loss 0.199841 LR 0.000500 Time 0.020512 -2022-12-06 11:13:57,558 - Epoch: [115][ 540/ 1200] Overall Loss 0.199675 Objective Loss 0.199675 LR 0.000500 Time 0.020488 -2022-12-06 11:13:57,751 - Epoch: [115][ 550/ 1200] Overall Loss 0.200082 Objective Loss 0.200082 LR 0.000500 Time 0.020465 -2022-12-06 11:13:57,944 - Epoch: [115][ 560/ 1200] Overall Loss 0.200034 Objective Loss 0.200034 LR 0.000500 Time 0.020444 -2022-12-06 11:13:58,136 - Epoch: [115][ 570/ 1200] Overall Loss 0.199863 Objective Loss 0.199863 LR 0.000500 Time 0.020421 -2022-12-06 11:13:58,330 - Epoch: [115][ 580/ 1200] Overall Loss 0.199795 Objective Loss 0.199795 LR 0.000500 Time 0.020402 -2022-12-06 11:13:58,522 - Epoch: [115][ 590/ 1200] Overall Loss 0.199813 Objective Loss 0.199813 LR 0.000500 Time 0.020381 -2022-12-06 11:13:58,715 - Epoch: [115][ 600/ 1200] Overall Loss 0.199983 Objective Loss 0.199983 LR 0.000500 Time 0.020363 -2022-12-06 11:13:58,908 - Epoch: [115][ 610/ 1200] Overall Loss 0.200063 Objective Loss 0.200063 LR 0.000500 Time 0.020344 -2022-12-06 11:13:59,101 - Epoch: [115][ 620/ 1200] Overall Loss 0.200529 Objective Loss 0.200529 LR 0.000500 Time 0.020326 -2022-12-06 11:13:59,294 - Epoch: [115][ 630/ 1200] Overall Loss 0.200812 Objective Loss 0.200812 LR 0.000500 Time 0.020309 -2022-12-06 11:13:59,486 - Epoch: [115][ 640/ 1200] Overall Loss 0.201072 Objective Loss 0.201072 LR 0.000500 Time 0.020292 -2022-12-06 11:13:59,679 - Epoch: [115][ 650/ 1200] Overall Loss 0.201098 Objective Loss 0.201098 LR 0.000500 Time 0.020274 -2022-12-06 11:13:59,871 - Epoch: [115][ 660/ 1200] Overall Loss 0.201080 Objective Loss 0.201080 LR 0.000500 Time 0.020258 -2022-12-06 11:14:00,064 - Epoch: [115][ 670/ 1200] Overall Loss 0.200897 Objective Loss 0.200897 LR 0.000500 Time 0.020243 -2022-12-06 11:14:00,257 - Epoch: [115][ 680/ 1200] Overall Loss 0.200762 Objective Loss 0.200762 LR 0.000500 Time 0.020227 -2022-12-06 11:14:00,449 - Epoch: [115][ 690/ 1200] Overall Loss 0.200673 Objective Loss 0.200673 LR 0.000500 Time 0.020213 -2022-12-06 11:14:00,642 - Epoch: [115][ 700/ 1200] Overall Loss 0.200901 Objective Loss 0.200901 LR 0.000500 Time 0.020199 -2022-12-06 11:14:00,835 - Epoch: [115][ 710/ 1200] Overall Loss 0.200780 Objective Loss 0.200780 LR 0.000500 Time 0.020185 -2022-12-06 11:14:01,028 - Epoch: [115][ 720/ 1200] Overall Loss 0.200704 Objective Loss 0.200704 LR 0.000500 Time 0.020172 -2022-12-06 11:14:01,221 - Epoch: [115][ 730/ 1200] Overall Loss 0.200657 Objective Loss 0.200657 LR 0.000500 Time 0.020159 -2022-12-06 11:14:01,414 - Epoch: [115][ 740/ 1200] Overall Loss 0.200475 Objective Loss 0.200475 LR 0.000500 Time 0.020147 -2022-12-06 11:14:01,607 - Epoch: [115][ 750/ 1200] Overall Loss 0.200395 Objective Loss 0.200395 LR 0.000500 Time 0.020135 -2022-12-06 11:14:01,800 - Epoch: [115][ 760/ 1200] Overall Loss 0.200188 Objective Loss 0.200188 LR 0.000500 Time 0.020123 -2022-12-06 11:14:01,993 - Epoch: [115][ 770/ 1200] Overall Loss 0.200346 Objective Loss 0.200346 LR 0.000500 Time 0.020111 -2022-12-06 11:14:02,186 - Epoch: [115][ 780/ 1200] Overall Loss 0.200535 Objective Loss 0.200535 LR 0.000500 Time 0.020100 -2022-12-06 11:14:02,379 - Epoch: [115][ 790/ 1200] Overall Loss 0.200583 Objective Loss 0.200583 LR 0.000500 Time 0.020090 -2022-12-06 11:14:02,572 - Epoch: [115][ 800/ 1200] Overall Loss 0.200604 Objective Loss 0.200604 LR 0.000500 Time 0.020079 -2022-12-06 11:14:02,765 - Epoch: [115][ 810/ 1200] Overall Loss 0.200654 Objective Loss 0.200654 LR 0.000500 Time 0.020069 -2022-12-06 11:14:02,958 - Epoch: [115][ 820/ 1200] Overall Loss 0.201189 Objective Loss 0.201189 LR 0.000500 Time 0.020059 -2022-12-06 11:14:03,150 - Epoch: [115][ 830/ 1200] Overall Loss 0.201558 Objective Loss 0.201558 LR 0.000500 Time 0.020049 -2022-12-06 11:14:03,344 - Epoch: [115][ 840/ 1200] Overall Loss 0.201488 Objective Loss 0.201488 LR 0.000500 Time 0.020039 -2022-12-06 11:14:03,536 - Epoch: [115][ 850/ 1200] Overall Loss 0.201595 Objective Loss 0.201595 LR 0.000500 Time 0.020029 -2022-12-06 11:14:03,729 - Epoch: [115][ 860/ 1200] Overall Loss 0.201911 Objective Loss 0.201911 LR 0.000500 Time 0.020020 -2022-12-06 11:14:03,922 - Epoch: [115][ 870/ 1200] Overall Loss 0.201673 Objective Loss 0.201673 LR 0.000500 Time 0.020011 -2022-12-06 11:14:04,114 - Epoch: [115][ 880/ 1200] Overall Loss 0.201889 Objective Loss 0.201889 LR 0.000500 Time 0.020002 -2022-12-06 11:14:04,307 - Epoch: [115][ 890/ 1200] Overall Loss 0.202026 Objective Loss 0.202026 LR 0.000500 Time 0.019993 -2022-12-06 11:14:04,501 - Epoch: [115][ 900/ 1200] Overall Loss 0.201923 Objective Loss 0.201923 LR 0.000500 Time 0.019985 -2022-12-06 11:14:04,693 - Epoch: [115][ 910/ 1200] Overall Loss 0.201999 Objective Loss 0.201999 LR 0.000500 Time 0.019976 -2022-12-06 11:14:04,886 - Epoch: [115][ 920/ 1200] Overall Loss 0.202149 Objective Loss 0.202149 LR 0.000500 Time 0.019969 -2022-12-06 11:14:05,079 - Epoch: [115][ 930/ 1200] Overall Loss 0.202196 Objective Loss 0.202196 LR 0.000500 Time 0.019960 -2022-12-06 11:14:05,272 - Epoch: [115][ 940/ 1200] Overall Loss 0.202246 Objective Loss 0.202246 LR 0.000500 Time 0.019953 -2022-12-06 11:14:05,465 - Epoch: [115][ 950/ 1200] Overall Loss 0.202357 Objective Loss 0.202357 LR 0.000500 Time 0.019945 -2022-12-06 11:14:05,658 - Epoch: [115][ 960/ 1200] Overall Loss 0.202300 Objective Loss 0.202300 LR 0.000500 Time 0.019938 -2022-12-06 11:14:05,850 - Epoch: [115][ 970/ 1200] Overall Loss 0.202238 Objective Loss 0.202238 LR 0.000500 Time 0.019930 -2022-12-06 11:14:06,043 - Epoch: [115][ 980/ 1200] Overall Loss 0.202376 Objective Loss 0.202376 LR 0.000500 Time 0.019923 -2022-12-06 11:14:06,236 - Epoch: [115][ 990/ 1200] Overall Loss 0.202565 Objective Loss 0.202565 LR 0.000500 Time 0.019917 -2022-12-06 11:14:06,430 - Epoch: [115][ 1000/ 1200] Overall Loss 0.202397 Objective Loss 0.202397 LR 0.000500 Time 0.019910 -2022-12-06 11:14:06,622 - Epoch: [115][ 1010/ 1200] Overall Loss 0.202161 Objective Loss 0.202161 LR 0.000500 Time 0.019903 -2022-12-06 11:14:06,816 - Epoch: [115][ 1020/ 1200] Overall Loss 0.202243 Objective Loss 0.202243 LR 0.000500 Time 0.019898 -2022-12-06 11:14:07,008 - Epoch: [115][ 1030/ 1200] Overall Loss 0.202250 Objective Loss 0.202250 LR 0.000500 Time 0.019890 -2022-12-06 11:14:07,200 - Epoch: [115][ 1040/ 1200] Overall Loss 0.202123 Objective Loss 0.202123 LR 0.000500 Time 0.019884 -2022-12-06 11:14:07,392 - Epoch: [115][ 1050/ 1200] Overall Loss 0.201997 Objective Loss 0.201997 LR 0.000500 Time 0.019877 -2022-12-06 11:14:07,586 - Epoch: [115][ 1060/ 1200] Overall Loss 0.202003 Objective Loss 0.202003 LR 0.000500 Time 0.019871 -2022-12-06 11:14:07,778 - Epoch: [115][ 1070/ 1200] Overall Loss 0.202022 Objective Loss 0.202022 LR 0.000500 Time 0.019864 -2022-12-06 11:14:07,972 - Epoch: [115][ 1080/ 1200] Overall Loss 0.201988 Objective Loss 0.201988 LR 0.000500 Time 0.019859 -2022-12-06 11:14:08,164 - Epoch: [115][ 1090/ 1200] Overall Loss 0.202053 Objective Loss 0.202053 LR 0.000500 Time 0.019853 -2022-12-06 11:14:08,357 - Epoch: [115][ 1100/ 1200] Overall Loss 0.202127 Objective Loss 0.202127 LR 0.000500 Time 0.019848 -2022-12-06 11:14:08,550 - Epoch: [115][ 1110/ 1200] Overall Loss 0.202413 Objective Loss 0.202413 LR 0.000500 Time 0.019842 -2022-12-06 11:14:08,743 - Epoch: [115][ 1120/ 1200] Overall Loss 0.202615 Objective Loss 0.202615 LR 0.000500 Time 0.019837 -2022-12-06 11:14:08,936 - Epoch: [115][ 1130/ 1200] Overall Loss 0.202614 Objective Loss 0.202614 LR 0.000500 Time 0.019831 -2022-12-06 11:14:09,129 - Epoch: [115][ 1140/ 1200] Overall Loss 0.202690 Objective Loss 0.202690 LR 0.000500 Time 0.019827 -2022-12-06 11:14:09,323 - Epoch: [115][ 1150/ 1200] Overall Loss 0.202768 Objective Loss 0.202768 LR 0.000500 Time 0.019822 -2022-12-06 11:14:09,516 - Epoch: [115][ 1160/ 1200] Overall Loss 0.202653 Objective Loss 0.202653 LR 0.000500 Time 0.019817 -2022-12-06 11:14:09,708 - Epoch: [115][ 1170/ 1200] Overall Loss 0.202817 Objective Loss 0.202817 LR 0.000500 Time 0.019811 -2022-12-06 11:14:09,901 - Epoch: [115][ 1180/ 1200] Overall Loss 0.202789 Objective Loss 0.202789 LR 0.000500 Time 0.019807 -2022-12-06 11:14:10,094 - Epoch: [115][ 1190/ 1200] Overall Loss 0.202850 Objective Loss 0.202850 LR 0.000500 Time 0.019802 -2022-12-06 11:14:10,317 - Epoch: [115][ 1200/ 1200] Overall Loss 0.202945 Objective Loss 0.202945 Top1 85.774059 Top5 98.535565 LR 0.000500 Time 0.019823 -2022-12-06 11:14:10,412 - --- validate (epoch=115)----------- -2022-12-06 11:14:10,413 - 34129 samples (256 per mini-batch) -2022-12-06 11:14:10,855 - Epoch: [115][ 10/ 134] Loss 0.237974 Top1 86.406250 Top5 98.437500 -2022-12-06 11:14:10,985 - Epoch: [115][ 20/ 134] Loss 0.264915 Top1 86.074219 Top5 98.515625 -2022-12-06 11:14:11,115 - Epoch: [115][ 30/ 134] Loss 0.265593 Top1 85.937500 Top5 98.385417 -2022-12-06 11:14:11,244 - Epoch: [115][ 40/ 134] Loss 0.260571 Top1 86.298828 Top5 98.447266 -2022-12-06 11:14:11,373 - Epoch: [115][ 50/ 134] Loss 0.263767 Top1 86.218750 Top5 98.328125 -2022-12-06 11:14:11,504 - Epoch: [115][ 60/ 134] Loss 0.268704 Top1 86.210938 Top5 98.294271 -2022-12-06 11:14:11,634 - Epoch: [115][ 70/ 134] Loss 0.260685 Top1 86.406250 Top5 98.275670 -2022-12-06 11:14:11,765 - Epoch: [115][ 80/ 134] Loss 0.261282 Top1 86.372070 Top5 98.237305 -2022-12-06 11:14:11,894 - Epoch: [115][ 90/ 134] Loss 0.261896 Top1 86.410590 Top5 98.242188 -2022-12-06 11:14:12,024 - Epoch: [115][ 100/ 134] Loss 0.262928 Top1 86.320312 Top5 98.242188 -2022-12-06 11:14:12,154 - Epoch: [115][ 110/ 134] Loss 0.266819 Top1 86.317472 Top5 98.231534 -2022-12-06 11:14:12,286 - Epoch: [115][ 120/ 134] Loss 0.263876 Top1 86.425781 Top5 98.258464 -2022-12-06 11:14:12,419 - Epoch: [115][ 130/ 134] Loss 0.262877 Top1 86.370192 Top5 98.266226 -2022-12-06 11:14:12,457 - Epoch: [115][ 134/ 134] Loss 0.261755 Top1 86.398664 Top5 98.271265 -2022-12-06 11:14:12,545 - ==> Top1: 86.399 Top5: 98.271 Loss: 0.262 - -2022-12-06 11:14:12,546 - ==> Confusion: -[[ 914 1 2 1 5 7 0 1 7 45 0 1 0 4 4 1 0 1 0 0 2] - [ 1 935 2 2 5 22 1 22 2 2 4 4 0 2 0 0 2 2 12 1 6] - [ 6 4 994 13 4 2 23 12 0 4 4 5 3 2 3 3 2 0 6 2 11] - [ 5 1 11 944 1 2 1 2 1 0 9 0 5 3 9 0 1 2 17 0 6] - [ 9 6 2 0 948 9 1 1 1 5 1 2 0 2 11 4 9 2 0 0 7] - [ 1 15 0 2 3 987 1 17 3 2 1 11 6 11 0 1 1 0 0 5 2] - [ 0 3 7 3 0 4 1068 7 0 0 1 2 3 1 0 4 2 1 0 9 3] - [ 0 4 4 2 3 23 5 958 0 0 2 5 0 3 1 0 0 0 25 15 4] - [ 5 6 0 1 0 0 1 2 980 41 3 2 1 4 8 1 1 1 5 1 1] - [ 52 0 0 0 4 3 0 4 29 881 1 1 0 14 2 1 0 2 0 0 7] - [ 1 2 1 8 1 1 1 4 12 0 954 2 2 15 2 0 0 0 9 2 2] - [ 1 0 1 0 0 13 3 4 0 1 0 976 27 3 0 5 5 4 0 7 1] - [ 0 0 1 2 0 1 1 1 0 0 0 20 916 2 0 6 1 9 0 4 5] - [ 0 0 0 0 0 8 0 1 17 10 4 3 4 955 1 2 4 0 1 2 11] - [ 5 2 2 12 4 4 0 0 29 3 0 1 4 2 1040 0 1 1 12 0 8] - [ 0 0 0 2 2 1 3 0 0 0 0 10 6 3 0 986 9 14 1 2 4] - [ 4 4 0 1 1 1 1 2 0 0 0 5 3 3 0 7 1030 0 1 3 6] - [ 2 1 1 5 0 0 1 1 0 3 1 9 18 3 2 13 1 969 1 1 4] - [ 3 3 1 8 1 4 1 21 1 1 4 2 1 0 4 1 0 2 946 1 3] - [ 2 3 2 0 3 10 4 9 1 0 1 15 10 5 0 2 4 3 1 998 7] - [ 111 226 153 122 85 214 76 173 100 99 130 99 368 262 137 102 162 70 224 211 10102]] - -2022-12-06 11:14:13,114 - ==> Best [Top1: 86.448 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 114] -2022-12-06 11:14:13,114 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:14:13,120 - - -2022-12-06 11:14:13,120 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:14:14,163 - Epoch: [116][ 10/ 1200] Overall Loss 0.197233 Objective Loss 0.197233 LR 0.000500 Time 0.104283 -2022-12-06 11:14:14,372 - Epoch: [116][ 20/ 1200] Overall Loss 0.202450 Objective Loss 0.202450 LR 0.000500 Time 0.062542 -2022-12-06 11:14:14,569 - Epoch: [116][ 30/ 1200] Overall Loss 0.206249 Objective Loss 0.206249 LR 0.000500 Time 0.048236 -2022-12-06 11:14:14,767 - Epoch: [116][ 40/ 1200] Overall Loss 0.202574 Objective Loss 0.202574 LR 0.000500 Time 0.041129 -2022-12-06 11:14:14,964 - Epoch: [116][ 50/ 1200] Overall Loss 0.198430 Objective Loss 0.198430 LR 0.000500 Time 0.036823 -2022-12-06 11:14:15,163 - Epoch: [116][ 60/ 1200] Overall Loss 0.200582 Objective Loss 0.200582 LR 0.000500 Time 0.033989 -2022-12-06 11:14:15,359 - Epoch: [116][ 70/ 1200] Overall Loss 0.203918 Objective Loss 0.203918 LR 0.000500 Time 0.031925 -2022-12-06 11:14:15,558 - Epoch: [116][ 80/ 1200] Overall Loss 0.198752 Objective Loss 0.198752 LR 0.000500 Time 0.030424 -2022-12-06 11:14:15,755 - Epoch: [116][ 90/ 1200] Overall Loss 0.197589 Objective Loss 0.197589 LR 0.000500 Time 0.029221 -2022-12-06 11:14:15,955 - Epoch: [116][ 100/ 1200] Overall Loss 0.196028 Objective Loss 0.196028 LR 0.000500 Time 0.028290 -2022-12-06 11:14:16,150 - Epoch: [116][ 110/ 1200] Overall Loss 0.194659 Objective Loss 0.194659 LR 0.000500 Time 0.027490 -2022-12-06 11:14:16,349 - Epoch: [116][ 120/ 1200] Overall Loss 0.194781 Objective Loss 0.194781 LR 0.000500 Time 0.026850 -2022-12-06 11:14:16,545 - Epoch: [116][ 130/ 1200] Overall Loss 0.195061 Objective Loss 0.195061 LR 0.000500 Time 0.026287 -2022-12-06 11:14:16,743 - Epoch: [116][ 140/ 1200] Overall Loss 0.195474 Objective Loss 0.195474 LR 0.000500 Time 0.025826 -2022-12-06 11:14:16,940 - Epoch: [116][ 150/ 1200] Overall Loss 0.194306 Objective Loss 0.194306 LR 0.000500 Time 0.025408 -2022-12-06 11:14:17,138 - Epoch: [116][ 160/ 1200] Overall Loss 0.194798 Objective Loss 0.194798 LR 0.000500 Time 0.025057 -2022-12-06 11:14:17,334 - Epoch: [116][ 170/ 1200] Overall Loss 0.193956 Objective Loss 0.193956 LR 0.000500 Time 0.024734 -2022-12-06 11:14:17,534 - Epoch: [116][ 180/ 1200] Overall Loss 0.194606 Objective Loss 0.194606 LR 0.000500 Time 0.024465 -2022-12-06 11:14:17,730 - Epoch: [116][ 190/ 1200] Overall Loss 0.194530 Objective Loss 0.194530 LR 0.000500 Time 0.024205 -2022-12-06 11:14:17,929 - Epoch: [116][ 200/ 1200] Overall Loss 0.195738 Objective Loss 0.195738 LR 0.000500 Time 0.023988 -2022-12-06 11:14:18,125 - Epoch: [116][ 210/ 1200] Overall Loss 0.196354 Objective Loss 0.196354 LR 0.000500 Time 0.023776 -2022-12-06 11:14:18,323 - Epoch: [116][ 220/ 1200] Overall Loss 0.196591 Objective Loss 0.196591 LR 0.000500 Time 0.023596 -2022-12-06 11:14:18,520 - Epoch: [116][ 230/ 1200] Overall Loss 0.195982 Objective Loss 0.195982 LR 0.000500 Time 0.023422 -2022-12-06 11:14:18,719 - Epoch: [116][ 240/ 1200] Overall Loss 0.195922 Objective Loss 0.195922 LR 0.000500 Time 0.023273 -2022-12-06 11:14:18,914 - Epoch: [116][ 250/ 1200] Overall Loss 0.195518 Objective Loss 0.195518 LR 0.000500 Time 0.023122 -2022-12-06 11:14:19,113 - Epoch: [116][ 260/ 1200] Overall Loss 0.195660 Objective Loss 0.195660 LR 0.000500 Time 0.022993 -2022-12-06 11:14:19,308 - Epoch: [116][ 270/ 1200] Overall Loss 0.195899 Objective Loss 0.195899 LR 0.000500 Time 0.022863 -2022-12-06 11:14:19,507 - Epoch: [116][ 280/ 1200] Overall Loss 0.196250 Objective Loss 0.196250 LR 0.000500 Time 0.022756 -2022-12-06 11:14:19,703 - Epoch: [116][ 290/ 1200] Overall Loss 0.196064 Objective Loss 0.196064 LR 0.000500 Time 0.022644 -2022-12-06 11:14:19,902 - Epoch: [116][ 300/ 1200] Overall Loss 0.196414 Objective Loss 0.196414 LR 0.000500 Time 0.022550 -2022-12-06 11:14:20,097 - Epoch: [116][ 310/ 1200] Overall Loss 0.196085 Objective Loss 0.196085 LR 0.000500 Time 0.022452 -2022-12-06 11:14:20,297 - Epoch: [116][ 320/ 1200] Overall Loss 0.196954 Objective Loss 0.196954 LR 0.000500 Time 0.022371 -2022-12-06 11:14:20,493 - Epoch: [116][ 330/ 1200] Overall Loss 0.196575 Objective Loss 0.196575 LR 0.000500 Time 0.022286 -2022-12-06 11:14:20,692 - Epoch: [116][ 340/ 1200] Overall Loss 0.196283 Objective Loss 0.196283 LR 0.000500 Time 0.022214 -2022-12-06 11:14:20,888 - Epoch: [116][ 350/ 1200] Overall Loss 0.196577 Objective Loss 0.196577 LR 0.000500 Time 0.022137 -2022-12-06 11:14:21,087 - Epoch: [116][ 360/ 1200] Overall Loss 0.197939 Objective Loss 0.197939 LR 0.000500 Time 0.022074 -2022-12-06 11:14:21,282 - Epoch: [116][ 370/ 1200] Overall Loss 0.198297 Objective Loss 0.198297 LR 0.000500 Time 0.022004 -2022-12-06 11:14:21,481 - Epoch: [116][ 380/ 1200] Overall Loss 0.197947 Objective Loss 0.197947 LR 0.000500 Time 0.021946 -2022-12-06 11:14:21,678 - Epoch: [116][ 390/ 1200] Overall Loss 0.198344 Objective Loss 0.198344 LR 0.000500 Time 0.021887 -2022-12-06 11:14:21,877 - Epoch: [116][ 400/ 1200] Overall Loss 0.198425 Objective Loss 0.198425 LR 0.000500 Time 0.021835 -2022-12-06 11:14:22,073 - Epoch: [116][ 410/ 1200] Overall Loss 0.198246 Objective Loss 0.198246 LR 0.000500 Time 0.021780 -2022-12-06 11:14:22,271 - Epoch: [116][ 420/ 1200] Overall Loss 0.198713 Objective Loss 0.198713 LR 0.000500 Time 0.021733 -2022-12-06 11:14:22,468 - Epoch: [116][ 430/ 1200] Overall Loss 0.198727 Objective Loss 0.198727 LR 0.000500 Time 0.021683 -2022-12-06 11:14:22,667 - Epoch: [116][ 440/ 1200] Overall Loss 0.198307 Objective Loss 0.198307 LR 0.000500 Time 0.021641 -2022-12-06 11:14:22,863 - Epoch: [116][ 450/ 1200] Overall Loss 0.198075 Objective Loss 0.198075 LR 0.000500 Time 0.021595 -2022-12-06 11:14:23,062 - Epoch: [116][ 460/ 1200] Overall Loss 0.198183 Objective Loss 0.198183 LR 0.000500 Time 0.021557 -2022-12-06 11:14:23,258 - Epoch: [116][ 470/ 1200] Overall Loss 0.198588 Objective Loss 0.198588 LR 0.000500 Time 0.021515 -2022-12-06 11:14:23,458 - Epoch: [116][ 480/ 1200] Overall Loss 0.198721 Objective Loss 0.198721 LR 0.000500 Time 0.021481 -2022-12-06 11:14:23,654 - Epoch: [116][ 490/ 1200] Overall Loss 0.198441 Objective Loss 0.198441 LR 0.000500 Time 0.021442 -2022-12-06 11:14:23,853 - Epoch: [116][ 500/ 1200] Overall Loss 0.198425 Objective Loss 0.198425 LR 0.000500 Time 0.021411 -2022-12-06 11:14:24,050 - Epoch: [116][ 510/ 1200] Overall Loss 0.198549 Objective Loss 0.198549 LR 0.000500 Time 0.021375 -2022-12-06 11:14:24,249 - Epoch: [116][ 520/ 1200] Overall Loss 0.198444 Objective Loss 0.198444 LR 0.000500 Time 0.021346 -2022-12-06 11:14:24,445 - Epoch: [116][ 530/ 1200] Overall Loss 0.199277 Objective Loss 0.199277 LR 0.000500 Time 0.021312 -2022-12-06 11:14:24,644 - Epoch: [116][ 540/ 1200] Overall Loss 0.199340 Objective Loss 0.199340 LR 0.000500 Time 0.021285 -2022-12-06 11:14:24,840 - Epoch: [116][ 550/ 1200] Overall Loss 0.199245 Objective Loss 0.199245 LR 0.000500 Time 0.021254 -2022-12-06 11:14:25,039 - Epoch: [116][ 560/ 1200] Overall Loss 0.198978 Objective Loss 0.198978 LR 0.000500 Time 0.021229 -2022-12-06 11:14:25,235 - Epoch: [116][ 570/ 1200] Overall Loss 0.198872 Objective Loss 0.198872 LR 0.000500 Time 0.021199 -2022-12-06 11:14:25,434 - Epoch: [116][ 580/ 1200] Overall Loss 0.198690 Objective Loss 0.198690 LR 0.000500 Time 0.021176 -2022-12-06 11:14:25,630 - Epoch: [116][ 590/ 1200] Overall Loss 0.198967 Objective Loss 0.198967 LR 0.000500 Time 0.021148 -2022-12-06 11:14:25,829 - Epoch: [116][ 600/ 1200] Overall Loss 0.198990 Objective Loss 0.198990 LR 0.000500 Time 0.021126 -2022-12-06 11:14:26,026 - Epoch: [116][ 610/ 1200] Overall Loss 0.198721 Objective Loss 0.198721 LR 0.000500 Time 0.021101 -2022-12-06 11:14:26,224 - Epoch: [116][ 620/ 1200] Overall Loss 0.198697 Objective Loss 0.198697 LR 0.000500 Time 0.021081 -2022-12-06 11:14:26,420 - Epoch: [116][ 630/ 1200] Overall Loss 0.198831 Objective Loss 0.198831 LR 0.000500 Time 0.021056 -2022-12-06 11:14:26,619 - Epoch: [116][ 640/ 1200] Overall Loss 0.199009 Objective Loss 0.199009 LR 0.000500 Time 0.021037 -2022-12-06 11:14:26,815 - Epoch: [116][ 650/ 1200] Overall Loss 0.199379 Objective Loss 0.199379 LR 0.000500 Time 0.021014 -2022-12-06 11:14:27,014 - Epoch: [116][ 660/ 1200] Overall Loss 0.199637 Objective Loss 0.199637 LR 0.000500 Time 0.020996 -2022-12-06 11:14:27,210 - Epoch: [116][ 670/ 1200] Overall Loss 0.199486 Objective Loss 0.199486 LR 0.000500 Time 0.020974 -2022-12-06 11:14:27,409 - Epoch: [116][ 680/ 1200] Overall Loss 0.199350 Objective Loss 0.199350 LR 0.000500 Time 0.020957 -2022-12-06 11:14:27,605 - Epoch: [116][ 690/ 1200] Overall Loss 0.199256 Objective Loss 0.199256 LR 0.000500 Time 0.020937 -2022-12-06 11:14:27,803 - Epoch: [116][ 700/ 1200] Overall Loss 0.199334 Objective Loss 0.199334 LR 0.000500 Time 0.020921 -2022-12-06 11:14:28,000 - Epoch: [116][ 710/ 1200] Overall Loss 0.199239 Objective Loss 0.199239 LR 0.000500 Time 0.020902 -2022-12-06 11:14:28,199 - Epoch: [116][ 720/ 1200] Overall Loss 0.199265 Objective Loss 0.199265 LR 0.000500 Time 0.020888 -2022-12-06 11:14:28,396 - Epoch: [116][ 730/ 1200] Overall Loss 0.199213 Objective Loss 0.199213 LR 0.000500 Time 0.020871 -2022-12-06 11:14:28,595 - Epoch: [116][ 740/ 1200] Overall Loss 0.199045 Objective Loss 0.199045 LR 0.000500 Time 0.020857 -2022-12-06 11:14:28,792 - Epoch: [116][ 750/ 1200] Overall Loss 0.198888 Objective Loss 0.198888 LR 0.000500 Time 0.020840 -2022-12-06 11:14:28,991 - Epoch: [116][ 760/ 1200] Overall Loss 0.198783 Objective Loss 0.198783 LR 0.000500 Time 0.020827 -2022-12-06 11:14:29,188 - Epoch: [116][ 770/ 1200] Overall Loss 0.198650 Objective Loss 0.198650 LR 0.000500 Time 0.020813 -2022-12-06 11:14:29,387 - Epoch: [116][ 780/ 1200] Overall Loss 0.198288 Objective Loss 0.198288 LR 0.000500 Time 0.020800 -2022-12-06 11:14:29,584 - Epoch: [116][ 790/ 1200] Overall Loss 0.198618 Objective Loss 0.198618 LR 0.000500 Time 0.020785 -2022-12-06 11:14:29,783 - Epoch: [116][ 800/ 1200] Overall Loss 0.198372 Objective Loss 0.198372 LR 0.000500 Time 0.020773 -2022-12-06 11:14:29,979 - Epoch: [116][ 810/ 1200] Overall Loss 0.198240 Objective Loss 0.198240 LR 0.000500 Time 0.020758 -2022-12-06 11:14:30,177 - Epoch: [116][ 820/ 1200] Overall Loss 0.198151 Objective Loss 0.198151 LR 0.000500 Time 0.020746 -2022-12-06 11:14:30,374 - Epoch: [116][ 830/ 1200] Overall Loss 0.198023 Objective Loss 0.198023 LR 0.000500 Time 0.020732 -2022-12-06 11:14:30,573 - Epoch: [116][ 840/ 1200] Overall Loss 0.198104 Objective Loss 0.198104 LR 0.000500 Time 0.020722 -2022-12-06 11:14:30,769 - Epoch: [116][ 850/ 1200] Overall Loss 0.198550 Objective Loss 0.198550 LR 0.000500 Time 0.020708 -2022-12-06 11:14:30,968 - Epoch: [116][ 860/ 1200] Overall Loss 0.198659 Objective Loss 0.198659 LR 0.000500 Time 0.020698 -2022-12-06 11:14:31,163 - Epoch: [116][ 870/ 1200] Overall Loss 0.198739 Objective Loss 0.198739 LR 0.000500 Time 0.020684 -2022-12-06 11:14:31,362 - Epoch: [116][ 880/ 1200] Overall Loss 0.198934 Objective Loss 0.198934 LR 0.000500 Time 0.020674 -2022-12-06 11:14:31,558 - Epoch: [116][ 890/ 1200] Overall Loss 0.198685 Objective Loss 0.198685 LR 0.000500 Time 0.020662 -2022-12-06 11:14:31,757 - Epoch: [116][ 900/ 1200] Overall Loss 0.198739 Objective Loss 0.198739 LR 0.000500 Time 0.020652 -2022-12-06 11:14:31,954 - Epoch: [116][ 910/ 1200] Overall Loss 0.198856 Objective Loss 0.198856 LR 0.000500 Time 0.020641 -2022-12-06 11:14:32,153 - Epoch: [116][ 920/ 1200] Overall Loss 0.198677 Objective Loss 0.198677 LR 0.000500 Time 0.020633 -2022-12-06 11:14:32,349 - Epoch: [116][ 930/ 1200] Overall Loss 0.198826 Objective Loss 0.198826 LR 0.000500 Time 0.020621 -2022-12-06 11:14:32,548 - Epoch: [116][ 940/ 1200] Overall Loss 0.199056 Objective Loss 0.199056 LR 0.000500 Time 0.020613 -2022-12-06 11:14:32,744 - Epoch: [116][ 950/ 1200] Overall Loss 0.199253 Objective Loss 0.199253 LR 0.000500 Time 0.020601 -2022-12-06 11:14:32,943 - Epoch: [116][ 960/ 1200] Overall Loss 0.199312 Objective Loss 0.199312 LR 0.000500 Time 0.020594 -2022-12-06 11:14:33,139 - Epoch: [116][ 970/ 1200] Overall Loss 0.199190 Objective Loss 0.199190 LR 0.000500 Time 0.020583 -2022-12-06 11:14:33,338 - Epoch: [116][ 980/ 1200] Overall Loss 0.199476 Objective Loss 0.199476 LR 0.000500 Time 0.020575 -2022-12-06 11:14:33,534 - Epoch: [116][ 990/ 1200] Overall Loss 0.199399 Objective Loss 0.199399 LR 0.000500 Time 0.020565 -2022-12-06 11:14:33,733 - Epoch: [116][ 1000/ 1200] Overall Loss 0.199403 Objective Loss 0.199403 LR 0.000500 Time 0.020558 -2022-12-06 11:14:33,929 - Epoch: [116][ 1010/ 1200] Overall Loss 0.199571 Objective Loss 0.199571 LR 0.000500 Time 0.020548 -2022-12-06 11:14:34,127 - Epoch: [116][ 1020/ 1200] Overall Loss 0.199640 Objective Loss 0.199640 LR 0.000500 Time 0.020540 -2022-12-06 11:14:34,324 - Epoch: [116][ 1030/ 1200] Overall Loss 0.199756 Objective Loss 0.199756 LR 0.000500 Time 0.020531 -2022-12-06 11:14:34,523 - Epoch: [116][ 1040/ 1200] Overall Loss 0.199563 Objective Loss 0.199563 LR 0.000500 Time 0.020525 -2022-12-06 11:14:34,720 - Epoch: [116][ 1050/ 1200] Overall Loss 0.199724 Objective Loss 0.199724 LR 0.000500 Time 0.020516 -2022-12-06 11:14:34,918 - Epoch: [116][ 1060/ 1200] Overall Loss 0.199637 Objective Loss 0.199637 LR 0.000500 Time 0.020509 -2022-12-06 11:14:35,114 - Epoch: [116][ 1070/ 1200] Overall Loss 0.199651 Objective Loss 0.199651 LR 0.000500 Time 0.020500 -2022-12-06 11:14:35,313 - Epoch: [116][ 1080/ 1200] Overall Loss 0.199779 Objective Loss 0.199779 LR 0.000500 Time 0.020494 -2022-12-06 11:14:35,509 - Epoch: [116][ 1090/ 1200] Overall Loss 0.199823 Objective Loss 0.199823 LR 0.000500 Time 0.020485 -2022-12-06 11:14:35,707 - Epoch: [116][ 1100/ 1200] Overall Loss 0.199946 Objective Loss 0.199946 LR 0.000500 Time 0.020479 -2022-12-06 11:14:35,904 - Epoch: [116][ 1110/ 1200] Overall Loss 0.200135 Objective Loss 0.200135 LR 0.000500 Time 0.020471 -2022-12-06 11:14:36,103 - Epoch: [116][ 1120/ 1200] Overall Loss 0.200102 Objective Loss 0.200102 LR 0.000500 Time 0.020465 -2022-12-06 11:14:36,299 - Epoch: [116][ 1130/ 1200] Overall Loss 0.200084 Objective Loss 0.200084 LR 0.000500 Time 0.020457 -2022-12-06 11:14:36,498 - Epoch: [116][ 1140/ 1200] Overall Loss 0.200228 Objective Loss 0.200228 LR 0.000500 Time 0.020452 -2022-12-06 11:14:36,694 - Epoch: [116][ 1150/ 1200] Overall Loss 0.200323 Objective Loss 0.200323 LR 0.000500 Time 0.020444 -2022-12-06 11:14:36,893 - Epoch: [116][ 1160/ 1200] Overall Loss 0.200326 Objective Loss 0.200326 LR 0.000500 Time 0.020439 -2022-12-06 11:14:37,090 - Epoch: [116][ 1170/ 1200] Overall Loss 0.200235 Objective Loss 0.200235 LR 0.000500 Time 0.020432 -2022-12-06 11:14:37,288 - Epoch: [116][ 1180/ 1200] Overall Loss 0.200311 Objective Loss 0.200311 LR 0.000500 Time 0.020426 -2022-12-06 11:14:37,484 - Epoch: [116][ 1190/ 1200] Overall Loss 0.200274 Objective Loss 0.200274 LR 0.000500 Time 0.020419 -2022-12-06 11:14:37,716 - Epoch: [116][ 1200/ 1200] Overall Loss 0.200516 Objective Loss 0.200516 Top1 85.146444 Top5 98.953975 LR 0.000500 Time 0.020441 -2022-12-06 11:14:37,804 - --- validate (epoch=116)----------- -2022-12-06 11:14:37,804 - 34129 samples (256 per mini-batch) -2022-12-06 11:14:38,247 - Epoch: [116][ 10/ 134] Loss 0.245554 Top1 86.093750 Top5 98.476562 -2022-12-06 11:14:38,379 - Epoch: [116][ 20/ 134] Loss 0.247344 Top1 85.839844 Top5 98.320312 -2022-12-06 11:14:38,510 - Epoch: [116][ 30/ 134] Loss 0.250278 Top1 86.119792 Top5 98.138021 -2022-12-06 11:14:38,640 - Epoch: [116][ 40/ 134] Loss 0.249332 Top1 86.181641 Top5 98.203125 -2022-12-06 11:14:38,769 - Epoch: [116][ 50/ 134] Loss 0.251129 Top1 86.140625 Top5 98.179688 -2022-12-06 11:14:38,901 - Epoch: [116][ 60/ 134] Loss 0.249173 Top1 86.276042 Top5 98.287760 -2022-12-06 11:14:39,032 - Epoch: [116][ 70/ 134] Loss 0.251714 Top1 86.065848 Top5 98.292411 -2022-12-06 11:14:39,163 - Epoch: [116][ 80/ 134] Loss 0.253708 Top1 85.976562 Top5 98.310547 -2022-12-06 11:14:39,294 - Epoch: [116][ 90/ 134] Loss 0.254331 Top1 85.963542 Top5 98.324653 -2022-12-06 11:14:39,424 - Epoch: [116][ 100/ 134] Loss 0.258137 Top1 85.851562 Top5 98.250000 -2022-12-06 11:14:39,551 - Epoch: [116][ 110/ 134] Loss 0.254577 Top1 85.919744 Top5 98.259943 -2022-12-06 11:14:39,680 - Epoch: [116][ 120/ 134] Loss 0.255909 Top1 85.826823 Top5 98.261719 -2022-12-06 11:14:39,808 - Epoch: [116][ 130/ 134] Loss 0.255754 Top1 85.910457 Top5 98.239183 -2022-12-06 11:14:39,845 - Epoch: [116][ 134/ 134] Loss 0.258852 Top1 85.894694 Top5 98.221454 -2022-12-06 11:14:39,933 - ==> Top1: 85.895 Top5: 98.221 Loss: 0.259 - -2022-12-06 11:14:39,934 - ==> Confusion: -[[ 896 2 0 2 10 9 0 0 5 55 0 4 1 1 4 1 0 1 2 0 3] - [ 3 952 2 2 5 20 1 11 0 1 4 4 1 0 0 1 1 1 10 4 4] - [ 5 2 1005 7 6 3 23 11 0 2 4 4 1 1 3 3 1 2 9 2 9] - [ 3 2 16 942 2 1 0 1 1 1 10 0 2 2 12 0 0 1 21 0 3] - [ 8 8 3 0 951 5 0 0 0 5 3 5 1 1 10 7 5 0 2 1 5] - [ 1 13 0 1 6 983 1 20 2 3 0 10 4 7 1 1 2 2 1 7 4] - [ 0 2 13 1 0 2 1069 3 0 0 3 1 0 3 0 2 1 3 2 13 0] - [ 3 6 4 0 3 26 7 948 0 0 0 3 0 1 0 0 1 0 36 13 3] - [ 5 3 0 1 0 5 0 1 968 49 6 3 2 4 9 1 2 1 3 1 0] - [ 51 0 2 0 6 7 0 2 16 896 1 1 0 8 3 0 0 2 1 0 5] - [ 1 2 2 2 2 1 1 2 6 1 968 3 1 10 2 1 1 0 9 1 3] - [ 4 0 0 0 1 9 4 1 3 0 0 974 29 0 1 6 4 4 2 8 1] - [ 0 1 1 3 0 2 2 0 1 0 1 24 909 1 0 5 1 9 1 4 4] - [ 2 1 0 0 1 8 0 2 14 17 4 6 3 946 1 2 4 0 2 1 9] - [ 4 3 2 5 3 4 0 0 19 7 0 2 3 2 1055 0 2 0 17 0 2] - [ 1 0 0 0 1 3 1 0 1 0 0 10 8 0 0 993 6 13 0 4 2] - [ 4 2 0 0 4 0 2 0 0 0 0 4 0 3 3 15 1026 0 0 6 3] - [ 2 1 1 5 0 2 0 1 1 2 0 6 20 2 4 15 0 970 1 1 2] - [ 1 9 2 6 2 3 1 17 1 1 5 1 1 1 9 1 0 1 943 0 3] - [ 3 3 0 0 0 4 7 5 1 0 4 11 6 2 0 2 4 4 0 1018 6] - [ 118 228 163 99 122 211 91 157 87 98 178 97 357 261 158 132 195 86 224 267 9897]] - -2022-12-06 11:14:40,512 - ==> Best [Top1: 86.448 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 114] -2022-12-06 11:14:40,512 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:14:40,518 - - -2022-12-06 11:14:40,518 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:14:41,458 - Epoch: [117][ 10/ 1200] Overall Loss 0.205791 Objective Loss 0.205791 LR 0.000500 Time 0.093947 -2022-12-06 11:14:41,661 - Epoch: [117][ 20/ 1200] Overall Loss 0.199577 Objective Loss 0.199577 LR 0.000500 Time 0.057072 -2022-12-06 11:14:41,860 - Epoch: [117][ 30/ 1200] Overall Loss 0.187915 Objective Loss 0.187915 LR 0.000500 Time 0.044680 -2022-12-06 11:14:42,056 - Epoch: [117][ 40/ 1200] Overall Loss 0.188106 Objective Loss 0.188106 LR 0.000500 Time 0.038383 -2022-12-06 11:14:42,256 - Epoch: [117][ 50/ 1200] Overall Loss 0.190056 Objective Loss 0.190056 LR 0.000500 Time 0.034693 -2022-12-06 11:14:42,451 - Epoch: [117][ 60/ 1200] Overall Loss 0.190097 Objective Loss 0.190097 LR 0.000500 Time 0.032156 -2022-12-06 11:14:42,650 - Epoch: [117][ 70/ 1200] Overall Loss 0.190283 Objective Loss 0.190283 LR 0.000500 Time 0.030398 -2022-12-06 11:14:42,845 - Epoch: [117][ 80/ 1200] Overall Loss 0.191033 Objective Loss 0.191033 LR 0.000500 Time 0.029034 -2022-12-06 11:14:43,044 - Epoch: [117][ 90/ 1200] Overall Loss 0.191621 Objective Loss 0.191621 LR 0.000500 Time 0.028006 -2022-12-06 11:14:43,239 - Epoch: [117][ 100/ 1200] Overall Loss 0.189813 Objective Loss 0.189813 LR 0.000500 Time 0.027155 -2022-12-06 11:14:43,439 - Epoch: [117][ 110/ 1200] Overall Loss 0.187640 Objective Loss 0.187640 LR 0.000500 Time 0.026493 -2022-12-06 11:14:43,634 - Epoch: [117][ 120/ 1200] Overall Loss 0.186616 Objective Loss 0.186616 LR 0.000500 Time 0.025909 -2022-12-06 11:14:43,833 - Epoch: [117][ 130/ 1200] Overall Loss 0.187146 Objective Loss 0.187146 LR 0.000500 Time 0.025443 -2022-12-06 11:14:44,029 - Epoch: [117][ 140/ 1200] Overall Loss 0.186630 Objective Loss 0.186630 LR 0.000500 Time 0.025020 -2022-12-06 11:14:44,227 - Epoch: [117][ 150/ 1200] Overall Loss 0.187062 Objective Loss 0.187062 LR 0.000500 Time 0.024671 -2022-12-06 11:14:44,422 - Epoch: [117][ 160/ 1200] Overall Loss 0.187138 Objective Loss 0.187138 LR 0.000500 Time 0.024343 -2022-12-06 11:14:44,620 - Epoch: [117][ 170/ 1200] Overall Loss 0.187601 Objective Loss 0.187601 LR 0.000500 Time 0.024074 -2022-12-06 11:14:44,816 - Epoch: [117][ 180/ 1200] Overall Loss 0.187628 Objective Loss 0.187628 LR 0.000500 Time 0.023821 -2022-12-06 11:14:45,015 - Epoch: [117][ 190/ 1200] Overall Loss 0.189232 Objective Loss 0.189232 LR 0.000500 Time 0.023611 -2022-12-06 11:14:45,210 - Epoch: [117][ 200/ 1200] Overall Loss 0.188405 Objective Loss 0.188405 LR 0.000500 Time 0.023402 -2022-12-06 11:14:45,408 - Epoch: [117][ 210/ 1200] Overall Loss 0.188840 Objective Loss 0.188840 LR 0.000500 Time 0.023231 -2022-12-06 11:14:45,604 - Epoch: [117][ 220/ 1200] Overall Loss 0.188641 Objective Loss 0.188641 LR 0.000500 Time 0.023059 -2022-12-06 11:14:45,802 - Epoch: [117][ 230/ 1200] Overall Loss 0.190060 Objective Loss 0.190060 LR 0.000500 Time 0.022919 -2022-12-06 11:14:45,998 - Epoch: [117][ 240/ 1200] Overall Loss 0.189908 Objective Loss 0.189908 LR 0.000500 Time 0.022777 -2022-12-06 11:14:46,197 - Epoch: [117][ 250/ 1200] Overall Loss 0.190116 Objective Loss 0.190116 LR 0.000500 Time 0.022659 -2022-12-06 11:14:46,393 - Epoch: [117][ 260/ 1200] Overall Loss 0.190699 Objective Loss 0.190699 LR 0.000500 Time 0.022541 -2022-12-06 11:14:46,592 - Epoch: [117][ 270/ 1200] Overall Loss 0.190356 Objective Loss 0.190356 LR 0.000500 Time 0.022441 -2022-12-06 11:14:46,787 - Epoch: [117][ 280/ 1200] Overall Loss 0.191157 Objective Loss 0.191157 LR 0.000500 Time 0.022334 -2022-12-06 11:14:46,986 - Epoch: [117][ 290/ 1200] Overall Loss 0.191480 Objective Loss 0.191480 LR 0.000500 Time 0.022247 -2022-12-06 11:14:47,182 - Epoch: [117][ 300/ 1200] Overall Loss 0.192207 Objective Loss 0.192207 LR 0.000500 Time 0.022156 -2022-12-06 11:14:47,381 - Epoch: [117][ 310/ 1200] Overall Loss 0.191736 Objective Loss 0.191736 LR 0.000500 Time 0.022081 -2022-12-06 11:14:47,576 - Epoch: [117][ 320/ 1200] Overall Loss 0.192568 Objective Loss 0.192568 LR 0.000500 Time 0.022000 -2022-12-06 11:14:47,775 - Epoch: [117][ 330/ 1200] Overall Loss 0.192310 Objective Loss 0.192310 LR 0.000500 Time 0.021935 -2022-12-06 11:14:47,971 - Epoch: [117][ 340/ 1200] Overall Loss 0.193042 Objective Loss 0.193042 LR 0.000500 Time 0.021865 -2022-12-06 11:14:48,170 - Epoch: [117][ 350/ 1200] Overall Loss 0.193440 Objective Loss 0.193440 LR 0.000500 Time 0.021806 -2022-12-06 11:14:48,366 - Epoch: [117][ 360/ 1200] Overall Loss 0.193335 Objective Loss 0.193335 LR 0.000500 Time 0.021742 -2022-12-06 11:14:48,565 - Epoch: [117][ 370/ 1200] Overall Loss 0.193404 Objective Loss 0.193404 LR 0.000500 Time 0.021693 -2022-12-06 11:14:48,760 - Epoch: [117][ 380/ 1200] Overall Loss 0.193994 Objective Loss 0.193994 LR 0.000500 Time 0.021632 -2022-12-06 11:14:48,958 - Epoch: [117][ 390/ 1200] Overall Loss 0.193904 Objective Loss 0.193904 LR 0.000500 Time 0.021585 -2022-12-06 11:14:49,154 - Epoch: [117][ 400/ 1200] Overall Loss 0.194699 Objective Loss 0.194699 LR 0.000500 Time 0.021532 -2022-12-06 11:14:49,352 - Epoch: [117][ 410/ 1200] Overall Loss 0.194681 Objective Loss 0.194681 LR 0.000500 Time 0.021490 -2022-12-06 11:14:49,547 - Epoch: [117][ 420/ 1200] Overall Loss 0.194840 Objective Loss 0.194840 LR 0.000500 Time 0.021441 -2022-12-06 11:14:49,746 - Epoch: [117][ 430/ 1200] Overall Loss 0.194138 Objective Loss 0.194138 LR 0.000500 Time 0.021403 -2022-12-06 11:14:49,941 - Epoch: [117][ 440/ 1200] Overall Loss 0.194781 Objective Loss 0.194781 LR 0.000500 Time 0.021359 -2022-12-06 11:14:50,140 - Epoch: [117][ 450/ 1200] Overall Loss 0.195212 Objective Loss 0.195212 LR 0.000500 Time 0.021325 -2022-12-06 11:14:50,335 - Epoch: [117][ 460/ 1200] Overall Loss 0.195124 Objective Loss 0.195124 LR 0.000500 Time 0.021284 -2022-12-06 11:14:50,534 - Epoch: [117][ 470/ 1200] Overall Loss 0.195435 Objective Loss 0.195435 LR 0.000500 Time 0.021253 -2022-12-06 11:14:50,729 - Epoch: [117][ 480/ 1200] Overall Loss 0.195328 Objective Loss 0.195328 LR 0.000500 Time 0.021216 -2022-12-06 11:14:50,927 - Epoch: [117][ 490/ 1200] Overall Loss 0.195480 Objective Loss 0.195480 LR 0.000500 Time 0.021187 -2022-12-06 11:14:51,123 - Epoch: [117][ 500/ 1200] Overall Loss 0.196117 Objective Loss 0.196117 LR 0.000500 Time 0.021153 -2022-12-06 11:14:51,322 - Epoch: [117][ 510/ 1200] Overall Loss 0.196419 Objective Loss 0.196419 LR 0.000500 Time 0.021128 -2022-12-06 11:14:51,517 - Epoch: [117][ 520/ 1200] Overall Loss 0.196623 Objective Loss 0.196623 LR 0.000500 Time 0.021096 -2022-12-06 11:14:51,716 - Epoch: [117][ 530/ 1200] Overall Loss 0.196983 Objective Loss 0.196983 LR 0.000500 Time 0.021072 -2022-12-06 11:14:51,911 - Epoch: [117][ 540/ 1200] Overall Loss 0.196807 Objective Loss 0.196807 LR 0.000500 Time 0.021042 -2022-12-06 11:14:52,110 - Epoch: [117][ 550/ 1200] Overall Loss 0.197111 Objective Loss 0.197111 LR 0.000500 Time 0.021020 -2022-12-06 11:14:52,305 - Epoch: [117][ 560/ 1200] Overall Loss 0.197367 Objective Loss 0.197367 LR 0.000500 Time 0.020992 -2022-12-06 11:14:52,504 - Epoch: [117][ 570/ 1200] Overall Loss 0.197552 Objective Loss 0.197552 LR 0.000500 Time 0.020971 -2022-12-06 11:14:52,699 - Epoch: [117][ 580/ 1200] Overall Loss 0.197436 Objective Loss 0.197436 LR 0.000500 Time 0.020945 -2022-12-06 11:14:52,897 - Epoch: [117][ 590/ 1200] Overall Loss 0.197722 Objective Loss 0.197722 LR 0.000500 Time 0.020925 -2022-12-06 11:14:53,092 - Epoch: [117][ 600/ 1200] Overall Loss 0.197305 Objective Loss 0.197305 LR 0.000500 Time 0.020900 -2022-12-06 11:14:53,290 - Epoch: [117][ 610/ 1200] Overall Loss 0.197154 Objective Loss 0.197154 LR 0.000500 Time 0.020882 -2022-12-06 11:14:53,485 - Epoch: [117][ 620/ 1200] Overall Loss 0.197645 Objective Loss 0.197645 LR 0.000500 Time 0.020859 -2022-12-06 11:14:53,684 - Epoch: [117][ 630/ 1200] Overall Loss 0.197744 Objective Loss 0.197744 LR 0.000500 Time 0.020842 -2022-12-06 11:14:53,879 - Epoch: [117][ 640/ 1200] Overall Loss 0.197535 Objective Loss 0.197535 LR 0.000500 Time 0.020821 -2022-12-06 11:14:54,078 - Epoch: [117][ 650/ 1200] Overall Loss 0.197439 Objective Loss 0.197439 LR 0.000500 Time 0.020805 -2022-12-06 11:14:54,274 - Epoch: [117][ 660/ 1200] Overall Loss 0.197562 Objective Loss 0.197562 LR 0.000500 Time 0.020785 -2022-12-06 11:14:54,472 - Epoch: [117][ 670/ 1200] Overall Loss 0.197316 Objective Loss 0.197316 LR 0.000500 Time 0.020770 -2022-12-06 11:14:54,667 - Epoch: [117][ 680/ 1200] Overall Loss 0.197202 Objective Loss 0.197202 LR 0.000500 Time 0.020751 -2022-12-06 11:14:54,866 - Epoch: [117][ 690/ 1200] Overall Loss 0.197320 Objective Loss 0.197320 LR 0.000500 Time 0.020738 -2022-12-06 11:14:55,061 - Epoch: [117][ 700/ 1200] Overall Loss 0.197482 Objective Loss 0.197482 LR 0.000500 Time 0.020720 -2022-12-06 11:14:55,259 - Epoch: [117][ 710/ 1200] Overall Loss 0.197532 Objective Loss 0.197532 LR 0.000500 Time 0.020706 -2022-12-06 11:14:55,455 - Epoch: [117][ 720/ 1200] Overall Loss 0.197453 Objective Loss 0.197453 LR 0.000500 Time 0.020690 -2022-12-06 11:14:55,654 - Epoch: [117][ 730/ 1200] Overall Loss 0.197308 Objective Loss 0.197308 LR 0.000500 Time 0.020677 -2022-12-06 11:14:55,848 - Epoch: [117][ 740/ 1200] Overall Loss 0.197614 Objective Loss 0.197614 LR 0.000500 Time 0.020660 -2022-12-06 11:14:56,047 - Epoch: [117][ 750/ 1200] Overall Loss 0.197747 Objective Loss 0.197747 LR 0.000500 Time 0.020649 -2022-12-06 11:14:56,242 - Epoch: [117][ 760/ 1200] Overall Loss 0.197972 Objective Loss 0.197972 LR 0.000500 Time 0.020633 -2022-12-06 11:14:56,441 - Epoch: [117][ 770/ 1200] Overall Loss 0.198232 Objective Loss 0.198232 LR 0.000500 Time 0.020623 -2022-12-06 11:14:56,635 - Epoch: [117][ 780/ 1200] Overall Loss 0.198069 Objective Loss 0.198069 LR 0.000500 Time 0.020607 -2022-12-06 11:14:56,834 - Epoch: [117][ 790/ 1200] Overall Loss 0.198231 Objective Loss 0.198231 LR 0.000500 Time 0.020597 -2022-12-06 11:14:57,030 - Epoch: [117][ 800/ 1200] Overall Loss 0.198115 Objective Loss 0.198115 LR 0.000500 Time 0.020583 -2022-12-06 11:14:57,229 - Epoch: [117][ 810/ 1200] Overall Loss 0.198069 Objective Loss 0.198069 LR 0.000500 Time 0.020574 -2022-12-06 11:14:57,423 - Epoch: [117][ 820/ 1200] Overall Loss 0.198271 Objective Loss 0.198271 LR 0.000500 Time 0.020560 -2022-12-06 11:14:57,622 - Epoch: [117][ 830/ 1200] Overall Loss 0.198191 Objective Loss 0.198191 LR 0.000500 Time 0.020551 -2022-12-06 11:14:57,817 - Epoch: [117][ 840/ 1200] Overall Loss 0.198232 Objective Loss 0.198232 LR 0.000500 Time 0.020538 -2022-12-06 11:14:58,016 - Epoch: [117][ 850/ 1200] Overall Loss 0.198045 Objective Loss 0.198045 LR 0.000500 Time 0.020530 -2022-12-06 11:14:58,211 - Epoch: [117][ 860/ 1200] Overall Loss 0.198060 Objective Loss 0.198060 LR 0.000500 Time 0.020517 -2022-12-06 11:14:58,410 - Epoch: [117][ 870/ 1200] Overall Loss 0.198169 Objective Loss 0.198169 LR 0.000500 Time 0.020510 -2022-12-06 11:14:58,606 - Epoch: [117][ 880/ 1200] Overall Loss 0.197842 Objective Loss 0.197842 LR 0.000500 Time 0.020499 -2022-12-06 11:14:58,805 - Epoch: [117][ 890/ 1200] Overall Loss 0.197720 Objective Loss 0.197720 LR 0.000500 Time 0.020491 -2022-12-06 11:14:59,000 - Epoch: [117][ 900/ 1200] Overall Loss 0.197480 Objective Loss 0.197480 LR 0.000500 Time 0.020479 -2022-12-06 11:14:59,199 - Epoch: [117][ 910/ 1200] Overall Loss 0.197490 Objective Loss 0.197490 LR 0.000500 Time 0.020472 -2022-12-06 11:14:59,395 - Epoch: [117][ 920/ 1200] Overall Loss 0.197556 Objective Loss 0.197556 LR 0.000500 Time 0.020462 -2022-12-06 11:14:59,593 - Epoch: [117][ 930/ 1200] Overall Loss 0.197412 Objective Loss 0.197412 LR 0.000500 Time 0.020455 -2022-12-06 11:14:59,789 - Epoch: [117][ 940/ 1200] Overall Loss 0.197019 Objective Loss 0.197019 LR 0.000500 Time 0.020445 -2022-12-06 11:14:59,987 - Epoch: [117][ 950/ 1200] Overall Loss 0.197190 Objective Loss 0.197190 LR 0.000500 Time 0.020438 -2022-12-06 11:15:00,182 - Epoch: [117][ 960/ 1200] Overall Loss 0.197103 Objective Loss 0.197103 LR 0.000500 Time 0.020427 -2022-12-06 11:15:00,381 - Epoch: [117][ 970/ 1200] Overall Loss 0.197093 Objective Loss 0.197093 LR 0.000500 Time 0.020421 -2022-12-06 11:15:00,576 - Epoch: [117][ 980/ 1200] Overall Loss 0.197290 Objective Loss 0.197290 LR 0.000500 Time 0.020411 -2022-12-06 11:15:00,775 - Epoch: [117][ 990/ 1200] Overall Loss 0.197235 Objective Loss 0.197235 LR 0.000500 Time 0.020405 -2022-12-06 11:15:00,971 - Epoch: [117][ 1000/ 1200] Overall Loss 0.197365 Objective Loss 0.197365 LR 0.000500 Time 0.020397 -2022-12-06 11:15:01,170 - Epoch: [117][ 1010/ 1200] Overall Loss 0.197706 Objective Loss 0.197706 LR 0.000500 Time 0.020391 -2022-12-06 11:15:01,365 - Epoch: [117][ 1020/ 1200] Overall Loss 0.197952 Objective Loss 0.197952 LR 0.000500 Time 0.020382 -2022-12-06 11:15:01,564 - Epoch: [117][ 1030/ 1200] Overall Loss 0.197920 Objective Loss 0.197920 LR 0.000500 Time 0.020376 -2022-12-06 11:15:01,759 - Epoch: [117][ 1040/ 1200] Overall Loss 0.197770 Objective Loss 0.197770 LR 0.000500 Time 0.020367 -2022-12-06 11:15:01,957 - Epoch: [117][ 1050/ 1200] Overall Loss 0.197807 Objective Loss 0.197807 LR 0.000500 Time 0.020362 -2022-12-06 11:15:02,152 - Epoch: [117][ 1060/ 1200] Overall Loss 0.198036 Objective Loss 0.198036 LR 0.000500 Time 0.020353 -2022-12-06 11:15:02,350 - Epoch: [117][ 1070/ 1200] Overall Loss 0.198194 Objective Loss 0.198194 LR 0.000500 Time 0.020348 -2022-12-06 11:15:02,545 - Epoch: [117][ 1080/ 1200] Overall Loss 0.198490 Objective Loss 0.198490 LR 0.000500 Time 0.020339 -2022-12-06 11:15:02,743 - Epoch: [117][ 1090/ 1200] Overall Loss 0.198424 Objective Loss 0.198424 LR 0.000500 Time 0.020334 -2022-12-06 11:15:02,938 - Epoch: [117][ 1100/ 1200] Overall Loss 0.198576 Objective Loss 0.198576 LR 0.000500 Time 0.020325 -2022-12-06 11:15:03,135 - Epoch: [117][ 1110/ 1200] Overall Loss 0.198527 Objective Loss 0.198527 LR 0.000500 Time 0.020320 -2022-12-06 11:15:03,330 - Epoch: [117][ 1120/ 1200] Overall Loss 0.198286 Objective Loss 0.198286 LR 0.000500 Time 0.020312 -2022-12-06 11:15:03,529 - Epoch: [117][ 1130/ 1200] Overall Loss 0.198335 Objective Loss 0.198335 LR 0.000500 Time 0.020307 -2022-12-06 11:15:03,723 - Epoch: [117][ 1140/ 1200] Overall Loss 0.198368 Objective Loss 0.198368 LR 0.000500 Time 0.020299 -2022-12-06 11:15:03,921 - Epoch: [117][ 1150/ 1200] Overall Loss 0.198316 Objective Loss 0.198316 LR 0.000500 Time 0.020294 -2022-12-06 11:15:04,115 - Epoch: [117][ 1160/ 1200] Overall Loss 0.198423 Objective Loss 0.198423 LR 0.000500 Time 0.020286 -2022-12-06 11:15:04,314 - Epoch: [117][ 1170/ 1200] Overall Loss 0.198459 Objective Loss 0.198459 LR 0.000500 Time 0.020282 -2022-12-06 11:15:04,507 - Epoch: [117][ 1180/ 1200] Overall Loss 0.198585 Objective Loss 0.198585 LR 0.000500 Time 0.020274 -2022-12-06 11:15:04,705 - Epoch: [117][ 1190/ 1200] Overall Loss 0.198659 Objective Loss 0.198659 LR 0.000500 Time 0.020269 -2022-12-06 11:15:04,935 - Epoch: [117][ 1200/ 1200] Overall Loss 0.198640 Objective Loss 0.198640 Top1 86.401674 Top5 97.698745 LR 0.000500 Time 0.020291 -2022-12-06 11:15:05,023 - --- validate (epoch=117)----------- -2022-12-06 11:15:05,024 - 34129 samples (256 per mini-batch) -2022-12-06 11:15:05,475 - Epoch: [117][ 10/ 134] Loss 0.230516 Top1 85.898438 Top5 98.750000 -2022-12-06 11:15:05,607 - Epoch: [117][ 20/ 134] Loss 0.225817 Top1 86.367188 Top5 98.554688 -2022-12-06 11:15:05,738 - Epoch: [117][ 30/ 134] Loss 0.243839 Top1 86.601562 Top5 98.541667 -2022-12-06 11:15:05,867 - Epoch: [117][ 40/ 134] Loss 0.250425 Top1 86.289062 Top5 98.486328 -2022-12-06 11:15:05,993 - Epoch: [117][ 50/ 134] Loss 0.252008 Top1 86.460938 Top5 98.500000 -2022-12-06 11:15:06,131 - Epoch: [117][ 60/ 134] Loss 0.250837 Top1 86.425781 Top5 98.476562 -2022-12-06 11:15:06,280 - Epoch: [117][ 70/ 134] Loss 0.252481 Top1 86.378348 Top5 98.415179 -2022-12-06 11:15:06,421 - Epoch: [117][ 80/ 134] Loss 0.253989 Top1 86.313477 Top5 98.403320 -2022-12-06 11:15:06,568 - Epoch: [117][ 90/ 134] Loss 0.253073 Top1 86.323785 Top5 98.411458 -2022-12-06 11:15:06,709 - Epoch: [117][ 100/ 134] Loss 0.253399 Top1 86.312500 Top5 98.351562 -2022-12-06 11:15:06,858 - Epoch: [117][ 110/ 134] Loss 0.254038 Top1 86.310369 Top5 98.323864 -2022-12-06 11:15:06,999 - Epoch: [117][ 120/ 134] Loss 0.256321 Top1 86.354167 Top5 98.294271 -2022-12-06 11:15:07,141 - Epoch: [117][ 130/ 134] Loss 0.256009 Top1 86.424279 Top5 98.299279 -2022-12-06 11:15:07,178 - Epoch: [117][ 134/ 134] Loss 0.255974 Top1 86.425034 Top5 98.318146 -2022-12-06 11:15:07,269 - ==> Top1: 86.425 Top5: 98.318 Loss: 0.256 - -2022-12-06 11:15:07,269 - ==> Confusion: -[[ 910 1 1 1 9 6 0 0 8 46 0 1 0 4 2 2 1 0 0 0 4] - [ 3 929 2 2 9 26 3 16 2 1 1 4 1 0 0 0 6 1 7 4 10] - [ 5 2 1009 13 4 2 21 9 0 2 6 5 2 3 3 3 1 1 3 2 7] - [ 4 1 10 941 3 2 0 0 0 0 11 1 6 1 15 2 0 3 15 0 5] - [ 9 5 2 0 959 5 0 0 0 5 1 4 0 1 10 8 5 2 1 1 2] - [ 3 12 0 1 9 968 2 22 2 2 1 13 7 10 3 1 0 0 1 6 6] - [ 3 1 7 1 0 2 1074 4 0 0 2 2 3 1 0 8 0 2 0 7 1] - [ 3 2 10 1 2 31 4 955 0 0 3 8 0 0 0 0 1 0 19 11 4] - [ 6 3 0 0 0 3 0 1 987 35 6 1 3 3 9 1 1 0 1 1 3] - [ 56 0 1 0 4 2 0 2 25 883 1 3 0 12 3 1 0 1 0 0 7] - [ 1 1 1 9 1 2 1 2 13 0 954 3 1 15 2 1 1 0 5 1 5] - [ 3 1 0 1 1 8 2 2 2 0 1 987 20 0 0 4 5 1 0 7 6] - [ 1 1 1 3 0 4 0 1 1 0 0 40 892 2 0 9 1 6 0 2 5] - [ 0 1 1 0 0 6 0 2 9 13 3 2 3 973 0 2 1 0 0 1 6] - [ 6 5 1 8 1 2 0 0 28 8 1 1 4 3 1046 0 0 1 9 0 6] - [ 1 0 1 0 2 0 2 0 0 1 0 10 4 3 0 1003 3 9 0 2 2] - [ 3 1 1 1 3 0 1 0 0 0 0 1 1 2 1 11 1033 0 0 5 8] - [ 2 1 1 2 0 1 1 2 1 3 0 9 31 2 1 18 0 957 0 0 4] - [ 5 5 7 12 2 2 0 27 1 1 5 2 3 0 9 1 0 2 918 2 4] - [ 3 4 1 0 0 4 6 6 0 0 0 14 11 6 1 4 4 1 3 1001 11] - [ 142 184 192 110 119 170 87 135 101 98 157 117 333 320 139 141 174 60 137 194 10116]] - -2022-12-06 11:15:07,943 - ==> Best [Top1: 86.448 Top5: 98.172 Sparsity:0.00 Params: 5376 on epoch: 114] -2022-12-06 11:15:07,943 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:15:07,949 - - -2022-12-06 11:15:07,949 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:15:08,875 - Epoch: [118][ 10/ 1200] Overall Loss 0.171499 Objective Loss 0.171499 LR 0.000500 Time 0.092561 -2022-12-06 11:15:09,071 - Epoch: [118][ 20/ 1200] Overall Loss 0.187842 Objective Loss 0.187842 LR 0.000500 Time 0.056031 -2022-12-06 11:15:09,263 - Epoch: [118][ 30/ 1200] Overall Loss 0.194775 Objective Loss 0.194775 LR 0.000500 Time 0.043725 -2022-12-06 11:15:09,453 - Epoch: [118][ 40/ 1200] Overall Loss 0.194335 Objective Loss 0.194335 LR 0.000500 Time 0.037539 -2022-12-06 11:15:09,644 - Epoch: [118][ 50/ 1200] Overall Loss 0.194134 Objective Loss 0.194134 LR 0.000500 Time 0.033832 -2022-12-06 11:15:09,834 - Epoch: [118][ 60/ 1200] Overall Loss 0.192029 Objective Loss 0.192029 LR 0.000500 Time 0.031359 -2022-12-06 11:15:10,025 - Epoch: [118][ 70/ 1200] Overall Loss 0.192395 Objective Loss 0.192395 LR 0.000500 Time 0.029594 -2022-12-06 11:15:10,215 - Epoch: [118][ 80/ 1200] Overall Loss 0.194260 Objective Loss 0.194260 LR 0.000500 Time 0.028269 -2022-12-06 11:15:10,406 - Epoch: [118][ 90/ 1200] Overall Loss 0.193891 Objective Loss 0.193891 LR 0.000500 Time 0.027239 -2022-12-06 11:15:10,596 - Epoch: [118][ 100/ 1200] Overall Loss 0.196425 Objective Loss 0.196425 LR 0.000500 Time 0.026409 -2022-12-06 11:15:10,786 - Epoch: [118][ 110/ 1200] Overall Loss 0.195605 Objective Loss 0.195605 LR 0.000500 Time 0.025734 -2022-12-06 11:15:10,977 - Epoch: [118][ 120/ 1200] Overall Loss 0.195453 Objective Loss 0.195453 LR 0.000500 Time 0.025171 -2022-12-06 11:15:11,167 - Epoch: [118][ 130/ 1200] Overall Loss 0.194982 Objective Loss 0.194982 LR 0.000500 Time 0.024697 -2022-12-06 11:15:11,357 - Epoch: [118][ 140/ 1200] Overall Loss 0.195750 Objective Loss 0.195750 LR 0.000500 Time 0.024286 -2022-12-06 11:15:11,547 - Epoch: [118][ 150/ 1200] Overall Loss 0.194321 Objective Loss 0.194321 LR 0.000500 Time 0.023930 -2022-12-06 11:15:11,737 - Epoch: [118][ 160/ 1200] Overall Loss 0.195550 Objective Loss 0.195550 LR 0.000500 Time 0.023615 -2022-12-06 11:15:11,927 - Epoch: [118][ 170/ 1200] Overall Loss 0.194776 Objective Loss 0.194776 LR 0.000500 Time 0.023344 -2022-12-06 11:15:12,117 - Epoch: [118][ 180/ 1200] Overall Loss 0.195698 Objective Loss 0.195698 LR 0.000500 Time 0.023099 -2022-12-06 11:15:12,307 - Epoch: [118][ 190/ 1200] Overall Loss 0.195677 Objective Loss 0.195677 LR 0.000500 Time 0.022879 -2022-12-06 11:15:12,497 - Epoch: [118][ 200/ 1200] Overall Loss 0.195665 Objective Loss 0.195665 LR 0.000500 Time 0.022682 -2022-12-06 11:15:12,687 - Epoch: [118][ 210/ 1200] Overall Loss 0.195902 Objective Loss 0.195902 LR 0.000500 Time 0.022505 -2022-12-06 11:15:12,878 - Epoch: [118][ 220/ 1200] Overall Loss 0.196434 Objective Loss 0.196434 LR 0.000500 Time 0.022345 -2022-12-06 11:15:13,069 - Epoch: [118][ 230/ 1200] Overall Loss 0.195286 Objective Loss 0.195286 LR 0.000500 Time 0.022203 -2022-12-06 11:15:13,260 - Epoch: [118][ 240/ 1200] Overall Loss 0.195933 Objective Loss 0.195933 LR 0.000500 Time 0.022069 -2022-12-06 11:15:13,450 - Epoch: [118][ 250/ 1200] Overall Loss 0.195732 Objective Loss 0.195732 LR 0.000500 Time 0.021945 -2022-12-06 11:15:13,640 - Epoch: [118][ 260/ 1200] Overall Loss 0.195927 Objective Loss 0.195927 LR 0.000500 Time 0.021830 -2022-12-06 11:15:13,831 - Epoch: [118][ 270/ 1200] Overall Loss 0.196176 Objective Loss 0.196176 LR 0.000500 Time 0.021727 -2022-12-06 11:15:14,021 - Epoch: [118][ 280/ 1200] Overall Loss 0.195961 Objective Loss 0.195961 LR 0.000500 Time 0.021627 -2022-12-06 11:15:14,211 - Epoch: [118][ 290/ 1200] Overall Loss 0.195917 Objective Loss 0.195917 LR 0.000500 Time 0.021535 -2022-12-06 11:15:14,401 - Epoch: [118][ 300/ 1200] Overall Loss 0.195984 Objective Loss 0.195984 LR 0.000500 Time 0.021449 -2022-12-06 11:15:14,591 - Epoch: [118][ 310/ 1200] Overall Loss 0.195854 Objective Loss 0.195854 LR 0.000500 Time 0.021367 -2022-12-06 11:15:14,781 - Epoch: [118][ 320/ 1200] Overall Loss 0.195867 Objective Loss 0.195867 LR 0.000500 Time 0.021292 -2022-12-06 11:15:14,971 - Epoch: [118][ 330/ 1200] Overall Loss 0.196160 Objective Loss 0.196160 LR 0.000500 Time 0.021222 -2022-12-06 11:15:15,162 - Epoch: [118][ 340/ 1200] Overall Loss 0.196422 Objective Loss 0.196422 LR 0.000500 Time 0.021156 -2022-12-06 11:15:15,353 - Epoch: [118][ 350/ 1200] Overall Loss 0.196499 Objective Loss 0.196499 LR 0.000500 Time 0.021095 -2022-12-06 11:15:15,544 - Epoch: [118][ 360/ 1200] Overall Loss 0.196082 Objective Loss 0.196082 LR 0.000500 Time 0.021038 -2022-12-06 11:15:15,734 - Epoch: [118][ 370/ 1200] Overall Loss 0.196103 Objective Loss 0.196103 LR 0.000500 Time 0.020983 -2022-12-06 11:15:15,924 - Epoch: [118][ 380/ 1200] Overall Loss 0.196182 Objective Loss 0.196182 LR 0.000500 Time 0.020930 -2022-12-06 11:15:16,115 - Epoch: [118][ 390/ 1200] Overall Loss 0.195909 Objective Loss 0.195909 LR 0.000500 Time 0.020881 -2022-12-06 11:15:16,305 - Epoch: [118][ 400/ 1200] Overall Loss 0.196425 Objective Loss 0.196425 LR 0.000500 Time 0.020833 -2022-12-06 11:15:16,496 - Epoch: [118][ 410/ 1200] Overall Loss 0.195975 Objective Loss 0.195975 LR 0.000500 Time 0.020789 -2022-12-06 11:15:16,687 - Epoch: [118][ 420/ 1200] Overall Loss 0.196451 Objective Loss 0.196451 LR 0.000500 Time 0.020747 -2022-12-06 11:15:16,878 - Epoch: [118][ 430/ 1200] Overall Loss 0.195961 Objective Loss 0.195961 LR 0.000500 Time 0.020708 -2022-12-06 11:15:17,069 - Epoch: [118][ 440/ 1200] Overall Loss 0.195833 Objective Loss 0.195833 LR 0.000500 Time 0.020670 -2022-12-06 11:15:17,260 - Epoch: [118][ 450/ 1200] Overall Loss 0.195931 Objective Loss 0.195931 LR 0.000500 Time 0.020632 -2022-12-06 11:15:17,450 - Epoch: [118][ 460/ 1200] Overall Loss 0.195639 Objective Loss 0.195639 LR 0.000500 Time 0.020596 -2022-12-06 11:15:17,641 - Epoch: [118][ 470/ 1200] Overall Loss 0.195690 Objective Loss 0.195690 LR 0.000500 Time 0.020564 -2022-12-06 11:15:17,832 - Epoch: [118][ 480/ 1200] Overall Loss 0.195916 Objective Loss 0.195916 LR 0.000500 Time 0.020531 -2022-12-06 11:15:18,023 - Epoch: [118][ 490/ 1200] Overall Loss 0.195590 Objective Loss 0.195590 LR 0.000500 Time 0.020502 -2022-12-06 11:15:18,213 - Epoch: [118][ 500/ 1200] Overall Loss 0.195281 Objective Loss 0.195281 LR 0.000500 Time 0.020470 -2022-12-06 11:15:18,404 - Epoch: [118][ 510/ 1200] Overall Loss 0.195249 Objective Loss 0.195249 LR 0.000500 Time 0.020441 -2022-12-06 11:15:18,594 - Epoch: [118][ 520/ 1200] Overall Loss 0.195236 Objective Loss 0.195236 LR 0.000500 Time 0.020413 -2022-12-06 11:15:18,785 - Epoch: [118][ 530/ 1200] Overall Loss 0.195492 Objective Loss 0.195492 LR 0.000500 Time 0.020388 -2022-12-06 11:15:18,976 - Epoch: [118][ 540/ 1200] Overall Loss 0.195834 Objective Loss 0.195834 LR 0.000500 Time 0.020362 -2022-12-06 11:15:19,167 - Epoch: [118][ 550/ 1200] Overall Loss 0.195894 Objective Loss 0.195894 LR 0.000500 Time 0.020339 -2022-12-06 11:15:19,357 - Epoch: [118][ 560/ 1200] Overall Loss 0.195954 Objective Loss 0.195954 LR 0.000500 Time 0.020313 -2022-12-06 11:15:19,547 - Epoch: [118][ 570/ 1200] Overall Loss 0.195847 Objective Loss 0.195847 LR 0.000500 Time 0.020290 -2022-12-06 11:15:19,737 - Epoch: [118][ 580/ 1200] Overall Loss 0.196182 Objective Loss 0.196182 LR 0.000500 Time 0.020266 -2022-12-06 11:15:19,927 - Epoch: [118][ 590/ 1200] Overall Loss 0.196497 Objective Loss 0.196497 LR 0.000500 Time 0.020244 -2022-12-06 11:15:20,117 - Epoch: [118][ 600/ 1200] Overall Loss 0.196378 Objective Loss 0.196378 LR 0.000500 Time 0.020222 -2022-12-06 11:15:20,307 - Epoch: [118][ 610/ 1200] Overall Loss 0.196519 Objective Loss 0.196519 LR 0.000500 Time 0.020202 -2022-12-06 11:15:20,497 - Epoch: [118][ 620/ 1200] Overall Loss 0.196420 Objective Loss 0.196420 LR 0.000500 Time 0.020182 -2022-12-06 11:15:20,687 - Epoch: [118][ 630/ 1200] Overall Loss 0.196287 Objective Loss 0.196287 LR 0.000500 Time 0.020162 -2022-12-06 11:15:20,877 - Epoch: [118][ 640/ 1200] Overall Loss 0.196430 Objective Loss 0.196430 LR 0.000500 Time 0.020142 -2022-12-06 11:15:21,068 - Epoch: [118][ 650/ 1200] Overall Loss 0.196024 Objective Loss 0.196024 LR 0.000500 Time 0.020125 -2022-12-06 11:15:21,259 - Epoch: [118][ 660/ 1200] Overall Loss 0.196231 Objective Loss 0.196231 LR 0.000500 Time 0.020108 -2022-12-06 11:15:21,449 - Epoch: [118][ 670/ 1200] Overall Loss 0.196379 Objective Loss 0.196379 LR 0.000500 Time 0.020092 -2022-12-06 11:15:21,640 - Epoch: [118][ 680/ 1200] Overall Loss 0.196797 Objective Loss 0.196797 LR 0.000500 Time 0.020075 -2022-12-06 11:15:21,831 - Epoch: [118][ 690/ 1200] Overall Loss 0.196912 Objective Loss 0.196912 LR 0.000500 Time 0.020061 -2022-12-06 11:15:22,021 - Epoch: [118][ 700/ 1200] Overall Loss 0.197024 Objective Loss 0.197024 LR 0.000500 Time 0.020045 -2022-12-06 11:15:22,212 - Epoch: [118][ 710/ 1200] Overall Loss 0.196820 Objective Loss 0.196820 LR 0.000500 Time 0.020030 -2022-12-06 11:15:22,401 - Epoch: [118][ 720/ 1200] Overall Loss 0.196592 Objective Loss 0.196592 LR 0.000500 Time 0.020015 -2022-12-06 11:15:22,591 - Epoch: [118][ 730/ 1200] Overall Loss 0.196888 Objective Loss 0.196888 LR 0.000500 Time 0.020000 -2022-12-06 11:15:22,781 - Epoch: [118][ 740/ 1200] Overall Loss 0.196808 Objective Loss 0.196808 LR 0.000500 Time 0.019985 -2022-12-06 11:15:22,971 - Epoch: [118][ 750/ 1200] Overall Loss 0.196886 Objective Loss 0.196886 LR 0.000500 Time 0.019972 -2022-12-06 11:15:23,161 - Epoch: [118][ 760/ 1200] Overall Loss 0.197096 Objective Loss 0.197096 LR 0.000500 Time 0.019958 -2022-12-06 11:15:23,351 - Epoch: [118][ 770/ 1200] Overall Loss 0.196931 Objective Loss 0.196931 LR 0.000500 Time 0.019945 -2022-12-06 11:15:23,541 - Epoch: [118][ 780/ 1200] Overall Loss 0.197007 Objective Loss 0.197007 LR 0.000500 Time 0.019933 -2022-12-06 11:15:23,731 - Epoch: [118][ 790/ 1200] Overall Loss 0.197113 Objective Loss 0.197113 LR 0.000500 Time 0.019920 -2022-12-06 11:15:23,922 - Epoch: [118][ 800/ 1200] Overall Loss 0.197292 Objective Loss 0.197292 LR 0.000500 Time 0.019908 -2022-12-06 11:15:24,113 - Epoch: [118][ 810/ 1200] Overall Loss 0.197044 Objective Loss 0.197044 LR 0.000500 Time 0.019898 -2022-12-06 11:15:24,304 - Epoch: [118][ 820/ 1200] Overall Loss 0.197148 Objective Loss 0.197148 LR 0.000500 Time 0.019887 -2022-12-06 11:15:24,495 - Epoch: [118][ 830/ 1200] Overall Loss 0.197105 Objective Loss 0.197105 LR 0.000500 Time 0.019877 -2022-12-06 11:15:24,685 - Epoch: [118][ 840/ 1200] Overall Loss 0.196991 Objective Loss 0.196991 LR 0.000500 Time 0.019867 -2022-12-06 11:15:24,876 - Epoch: [118][ 850/ 1200] Overall Loss 0.197260 Objective Loss 0.197260 LR 0.000500 Time 0.019856 -2022-12-06 11:15:25,066 - Epoch: [118][ 860/ 1200] Overall Loss 0.197495 Objective Loss 0.197495 LR 0.000500 Time 0.019846 -2022-12-06 11:15:25,257 - Epoch: [118][ 870/ 1200] Overall Loss 0.197644 Objective Loss 0.197644 LR 0.000500 Time 0.019836 -2022-12-06 11:15:25,447 - Epoch: [118][ 880/ 1200] Overall Loss 0.197481 Objective Loss 0.197481 LR 0.000500 Time 0.019826 -2022-12-06 11:15:25,637 - Epoch: [118][ 890/ 1200] Overall Loss 0.197247 Objective Loss 0.197247 LR 0.000500 Time 0.019817 -2022-12-06 11:15:25,828 - Epoch: [118][ 900/ 1200] Overall Loss 0.197427 Objective Loss 0.197427 LR 0.000500 Time 0.019807 -2022-12-06 11:15:26,019 - Epoch: [118][ 910/ 1200] Overall Loss 0.197568 Objective Loss 0.197568 LR 0.000500 Time 0.019799 -2022-12-06 11:15:26,209 - Epoch: [118][ 920/ 1200] Overall Loss 0.197612 Objective Loss 0.197612 LR 0.000500 Time 0.019790 -2022-12-06 11:15:26,399 - Epoch: [118][ 930/ 1200] Overall Loss 0.197508 Objective Loss 0.197508 LR 0.000500 Time 0.019781 -2022-12-06 11:15:26,589 - Epoch: [118][ 940/ 1200] Overall Loss 0.197388 Objective Loss 0.197388 LR 0.000500 Time 0.019772 -2022-12-06 11:15:26,779 - Epoch: [118][ 950/ 1200] Overall Loss 0.197504 Objective Loss 0.197504 LR 0.000500 Time 0.019764 -2022-12-06 11:15:26,970 - Epoch: [118][ 960/ 1200] Overall Loss 0.197686 Objective Loss 0.197686 LR 0.000500 Time 0.019756 -2022-12-06 11:15:27,160 - Epoch: [118][ 970/ 1200] Overall Loss 0.197886 Objective Loss 0.197886 LR 0.000500 Time 0.019748 -2022-12-06 11:15:27,351 - Epoch: [118][ 980/ 1200] Overall Loss 0.197972 Objective Loss 0.197972 LR 0.000500 Time 0.019740 -2022-12-06 11:15:27,541 - Epoch: [118][ 990/ 1200] Overall Loss 0.197857 Objective Loss 0.197857 LR 0.000500 Time 0.019732 -2022-12-06 11:15:27,731 - Epoch: [118][ 1000/ 1200] Overall Loss 0.197751 Objective Loss 0.197751 LR 0.000500 Time 0.019724 -2022-12-06 11:15:27,920 - Epoch: [118][ 1010/ 1200] Overall Loss 0.197730 Objective Loss 0.197730 LR 0.000500 Time 0.019716 -2022-12-06 11:15:28,110 - Epoch: [118][ 1020/ 1200] Overall Loss 0.197833 Objective Loss 0.197833 LR 0.000500 Time 0.019709 -2022-12-06 11:15:28,301 - Epoch: [118][ 1030/ 1200] Overall Loss 0.197734 Objective Loss 0.197734 LR 0.000500 Time 0.019702 -2022-12-06 11:15:28,491 - Epoch: [118][ 1040/ 1200] Overall Loss 0.197684 Objective Loss 0.197684 LR 0.000500 Time 0.019695 -2022-12-06 11:15:28,681 - Epoch: [118][ 1050/ 1200] Overall Loss 0.197782 Objective Loss 0.197782 LR 0.000500 Time 0.019687 -2022-12-06 11:15:28,871 - Epoch: [118][ 1060/ 1200] Overall Loss 0.197857 Objective Loss 0.197857 LR 0.000500 Time 0.019680 -2022-12-06 11:15:29,061 - Epoch: [118][ 1070/ 1200] Overall Loss 0.198033 Objective Loss 0.198033 LR 0.000500 Time 0.019674 -2022-12-06 11:15:29,252 - Epoch: [118][ 1080/ 1200] Overall Loss 0.197913 Objective Loss 0.197913 LR 0.000500 Time 0.019668 -2022-12-06 11:15:29,442 - Epoch: [118][ 1090/ 1200] Overall Loss 0.198013 Objective Loss 0.198013 LR 0.000500 Time 0.019661 -2022-12-06 11:15:29,632 - Epoch: [118][ 1100/ 1200] Overall Loss 0.198003 Objective Loss 0.198003 LR 0.000500 Time 0.019655 -2022-12-06 11:15:29,823 - Epoch: [118][ 1110/ 1200] Overall Loss 0.198219 Objective Loss 0.198219 LR 0.000500 Time 0.019649 -2022-12-06 11:15:30,014 - Epoch: [118][ 1120/ 1200] Overall Loss 0.198345 Objective Loss 0.198345 LR 0.000500 Time 0.019644 -2022-12-06 11:15:30,205 - Epoch: [118][ 1130/ 1200] Overall Loss 0.198384 Objective Loss 0.198384 LR 0.000500 Time 0.019638 -2022-12-06 11:15:30,395 - Epoch: [118][ 1140/ 1200] Overall Loss 0.198557 Objective Loss 0.198557 LR 0.000500 Time 0.019633 -2022-12-06 11:15:30,585 - Epoch: [118][ 1150/ 1200] Overall Loss 0.198484 Objective Loss 0.198484 LR 0.000500 Time 0.019626 -2022-12-06 11:15:30,775 - Epoch: [118][ 1160/ 1200] Overall Loss 0.198583 Objective Loss 0.198583 LR 0.000500 Time 0.019621 -2022-12-06 11:15:30,965 - Epoch: [118][ 1170/ 1200] Overall Loss 0.198773 Objective Loss 0.198773 LR 0.000500 Time 0.019615 -2022-12-06 11:15:31,156 - Epoch: [118][ 1180/ 1200] Overall Loss 0.198625 Objective Loss 0.198625 LR 0.000500 Time 0.019610 -2022-12-06 11:15:31,347 - Epoch: [118][ 1190/ 1200] Overall Loss 0.198596 Objective Loss 0.198596 LR 0.000500 Time 0.019605 -2022-12-06 11:15:31,567 - Epoch: [118][ 1200/ 1200] Overall Loss 0.198729 Objective Loss 0.198729 Top1 88.075314 Top5 98.953975 LR 0.000500 Time 0.019625 -2022-12-06 11:15:31,662 - --- validate (epoch=118)----------- -2022-12-06 11:15:31,662 - 34129 samples (256 per mini-batch) -2022-12-06 11:15:32,107 - Epoch: [118][ 10/ 134] Loss 0.267261 Top1 87.226562 Top5 98.046875 -2022-12-06 11:15:32,237 - Epoch: [118][ 20/ 134] Loss 0.246060 Top1 87.539062 Top5 98.437500 -2022-12-06 11:15:32,367 - Epoch: [118][ 30/ 134] Loss 0.251963 Top1 87.239583 Top5 98.346354 -2022-12-06 11:15:32,497 - Epoch: [118][ 40/ 134] Loss 0.253067 Top1 86.992188 Top5 98.476562 -2022-12-06 11:15:32,631 - Epoch: [118][ 50/ 134] Loss 0.247918 Top1 86.929688 Top5 98.390625 -2022-12-06 11:15:32,759 - Epoch: [118][ 60/ 134] Loss 0.253521 Top1 86.790365 Top5 98.398438 -2022-12-06 11:15:32,885 - Epoch: [118][ 70/ 134] Loss 0.254817 Top1 86.813616 Top5 98.303571 -2022-12-06 11:15:33,015 - Epoch: [118][ 80/ 134] Loss 0.258126 Top1 86.791992 Top5 98.281250 -2022-12-06 11:15:33,147 - Epoch: [118][ 90/ 134] Loss 0.256048 Top1 86.805556 Top5 98.311632 -2022-12-06 11:15:33,276 - Epoch: [118][ 100/ 134] Loss 0.256019 Top1 86.820312 Top5 98.316406 -2022-12-06 11:15:33,407 - Epoch: [118][ 110/ 134] Loss 0.258274 Top1 86.736506 Top5 98.291903 -2022-12-06 11:15:33,538 - Epoch: [118][ 120/ 134] Loss 0.257244 Top1 86.767578 Top5 98.300781 -2022-12-06 11:15:33,671 - Epoch: [118][ 130/ 134] Loss 0.257050 Top1 86.769832 Top5 98.299279 -2022-12-06 11:15:33,710 - Epoch: [118][ 134/ 134] Loss 0.257555 Top1 86.744411 Top5 98.300566 -2022-12-06 11:15:33,801 - ==> Top1: 86.744 Top5: 98.301 Loss: 0.258 - -2022-12-06 11:15:33,802 - ==> Confusion: -[[ 916 0 0 2 6 5 0 0 5 44 0 3 1 4 3 1 1 2 1 0 2] - [ 1 933 2 3 9 24 5 8 2 0 3 4 2 2 1 1 3 4 11 2 7] - [ 4 2 1003 17 7 3 17 7 0 3 6 4 3 1 4 1 0 3 4 3 11] - [ 4 1 18 939 3 5 0 0 0 0 11 1 6 1 11 0 0 3 11 0 6] - [ 11 5 3 0 949 6 1 0 0 5 1 6 1 2 9 7 4 1 1 0 8] - [ 2 9 0 1 9 974 2 16 4 4 1 15 4 12 4 1 1 0 1 4 5] - [ 2 1 10 3 0 2 1072 2 0 0 0 4 0 1 0 7 2 2 1 7 2] - [ 0 7 6 2 2 24 11 944 0 1 3 7 0 1 1 1 0 1 21 14 8] - [ 3 2 0 0 2 2 2 1 979 40 6 1 3 5 11 1 1 1 2 1 1] - [ 48 0 1 0 6 1 0 2 38 881 1 2 0 11 3 1 0 0 0 0 6] - [ 0 0 2 5 1 3 1 3 13 2 962 1 0 9 3 1 1 0 5 3 4] - [ 0 1 1 0 1 7 5 2 0 0 1 985 19 5 1 5 3 4 0 10 1] - [ 0 1 1 2 2 1 0 2 1 0 0 37 890 1 0 6 1 11 0 5 8] - [ 0 0 0 0 0 9 0 4 8 13 7 7 4 957 3 0 1 1 0 1 8] - [ 5 4 2 8 3 4 0 0 20 2 0 2 1 2 1066 0 1 1 5 0 4] - [ 0 0 1 2 4 3 4 0 2 0 0 7 7 2 0 985 5 12 0 4 5] - [ 0 3 1 1 4 1 2 0 0 0 1 1 3 1 0 14 1024 0 0 6 10] - [ 1 0 2 3 1 2 0 1 1 1 0 12 13 3 1 13 1 978 0 2 1] - [ 4 3 5 5 2 2 0 19 2 1 4 1 4 0 11 1 0 2 935 3 4] - [ 4 3 2 0 1 3 7 6 0 0 1 18 5 3 0 3 3 2 1 1011 7] - [ 115 183 169 131 121 162 71 131 93 107 166 123 326 290 155 87 124 88 149 219 10216]] - -2022-12-06 11:15:34,468 - ==> Best [Top1: 86.744 Top5: 98.301 Sparsity:0.00 Params: 5376 on epoch: 118] -2022-12-06 11:15:34,468 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:15:34,475 - - -2022-12-06 11:15:34,475 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:15:35,414 - Epoch: [119][ 10/ 1200] Overall Loss 0.192910 Objective Loss 0.192910 LR 0.000500 Time 0.093808 -2022-12-06 11:15:35,617 - Epoch: [119][ 20/ 1200] Overall Loss 0.199276 Objective Loss 0.199276 LR 0.000500 Time 0.057041 -2022-12-06 11:15:35,816 - Epoch: [119][ 30/ 1200] Overall Loss 0.191750 Objective Loss 0.191750 LR 0.000500 Time 0.044627 -2022-12-06 11:15:36,012 - Epoch: [119][ 40/ 1200] Overall Loss 0.192617 Objective Loss 0.192617 LR 0.000500 Time 0.038347 -2022-12-06 11:15:36,211 - Epoch: [119][ 50/ 1200] Overall Loss 0.192494 Objective Loss 0.192494 LR 0.000500 Time 0.034645 -2022-12-06 11:15:36,405 - Epoch: [119][ 60/ 1200] Overall Loss 0.195596 Objective Loss 0.195596 LR 0.000500 Time 0.032111 -2022-12-06 11:15:36,604 - Epoch: [119][ 70/ 1200] Overall Loss 0.194162 Objective Loss 0.194162 LR 0.000500 Time 0.030352 -2022-12-06 11:15:36,799 - Epoch: [119][ 80/ 1200] Overall Loss 0.193219 Objective Loss 0.193219 LR 0.000500 Time 0.028983 -2022-12-06 11:15:36,997 - Epoch: [119][ 90/ 1200] Overall Loss 0.192407 Objective Loss 0.192407 LR 0.000500 Time 0.027960 -2022-12-06 11:15:37,193 - Epoch: [119][ 100/ 1200] Overall Loss 0.191875 Objective Loss 0.191875 LR 0.000500 Time 0.027115 -2022-12-06 11:15:37,391 - Epoch: [119][ 110/ 1200] Overall Loss 0.190944 Objective Loss 0.190944 LR 0.000500 Time 0.026449 -2022-12-06 11:15:37,587 - Epoch: [119][ 120/ 1200] Overall Loss 0.191891 Objective Loss 0.191891 LR 0.000500 Time 0.025874 -2022-12-06 11:15:37,784 - Epoch: [119][ 130/ 1200] Overall Loss 0.194383 Objective Loss 0.194383 LR 0.000500 Time 0.025398 -2022-12-06 11:15:37,979 - Epoch: [119][ 140/ 1200] Overall Loss 0.194216 Objective Loss 0.194216 LR 0.000500 Time 0.024973 -2022-12-06 11:15:38,178 - Epoch: [119][ 150/ 1200] Overall Loss 0.194843 Objective Loss 0.194843 LR 0.000500 Time 0.024627 -2022-12-06 11:15:38,372 - Epoch: [119][ 160/ 1200] Overall Loss 0.195582 Objective Loss 0.195582 LR 0.000500 Time 0.024300 -2022-12-06 11:15:38,571 - Epoch: [119][ 170/ 1200] Overall Loss 0.196988 Objective Loss 0.196988 LR 0.000500 Time 0.024034 -2022-12-06 11:15:38,766 - Epoch: [119][ 180/ 1200] Overall Loss 0.197678 Objective Loss 0.197678 LR 0.000500 Time 0.023780 -2022-12-06 11:15:38,964 - Epoch: [119][ 190/ 1200] Overall Loss 0.197236 Objective Loss 0.197236 LR 0.000500 Time 0.023569 -2022-12-06 11:15:39,160 - Epoch: [119][ 200/ 1200] Overall Loss 0.197844 Objective Loss 0.197844 LR 0.000500 Time 0.023365 -2022-12-06 11:15:39,358 - Epoch: [119][ 210/ 1200] Overall Loss 0.197619 Objective Loss 0.197619 LR 0.000500 Time 0.023194 -2022-12-06 11:15:39,553 - Epoch: [119][ 220/ 1200] Overall Loss 0.196996 Objective Loss 0.196996 LR 0.000500 Time 0.023026 -2022-12-06 11:15:39,752 - Epoch: [119][ 230/ 1200] Overall Loss 0.197077 Objective Loss 0.197077 LR 0.000500 Time 0.022886 -2022-12-06 11:15:39,947 - Epoch: [119][ 240/ 1200] Overall Loss 0.195845 Objective Loss 0.195845 LR 0.000500 Time 0.022744 -2022-12-06 11:15:40,145 - Epoch: [119][ 250/ 1200] Overall Loss 0.195803 Objective Loss 0.195803 LR 0.000500 Time 0.022623 -2022-12-06 11:15:40,339 - Epoch: [119][ 260/ 1200] Overall Loss 0.194256 Objective Loss 0.194256 LR 0.000500 Time 0.022498 -2022-12-06 11:15:40,538 - Epoch: [119][ 270/ 1200] Overall Loss 0.194277 Objective Loss 0.194277 LR 0.000500 Time 0.022398 -2022-12-06 11:15:40,733 - Epoch: [119][ 280/ 1200] Overall Loss 0.194406 Objective Loss 0.194406 LR 0.000500 Time 0.022291 -2022-12-06 11:15:40,931 - Epoch: [119][ 290/ 1200] Overall Loss 0.194606 Objective Loss 0.194606 LR 0.000500 Time 0.022205 -2022-12-06 11:15:41,126 - Epoch: [119][ 300/ 1200] Overall Loss 0.194661 Objective Loss 0.194661 LR 0.000500 Time 0.022114 -2022-12-06 11:15:41,325 - Epoch: [119][ 310/ 1200] Overall Loss 0.195167 Objective Loss 0.195167 LR 0.000500 Time 0.022039 -2022-12-06 11:15:41,521 - Epoch: [119][ 320/ 1200] Overall Loss 0.195462 Objective Loss 0.195462 LR 0.000500 Time 0.021962 -2022-12-06 11:15:41,719 - Epoch: [119][ 330/ 1200] Overall Loss 0.195474 Objective Loss 0.195474 LR 0.000500 Time 0.021895 -2022-12-06 11:15:41,913 - Epoch: [119][ 340/ 1200] Overall Loss 0.196224 Objective Loss 0.196224 LR 0.000500 Time 0.021820 -2022-12-06 11:15:42,112 - Epoch: [119][ 350/ 1200] Overall Loss 0.196617 Objective Loss 0.196617 LR 0.000500 Time 0.021762 -2022-12-06 11:15:42,307 - Epoch: [119][ 360/ 1200] Overall Loss 0.196225 Objective Loss 0.196225 LR 0.000500 Time 0.021697 -2022-12-06 11:15:42,505 - Epoch: [119][ 370/ 1200] Overall Loss 0.196625 Objective Loss 0.196625 LR 0.000500 Time 0.021645 -2022-12-06 11:15:42,699 - Epoch: [119][ 380/ 1200] Overall Loss 0.196838 Objective Loss 0.196838 LR 0.000500 Time 0.021586 -2022-12-06 11:15:42,896 - Epoch: [119][ 390/ 1200] Overall Loss 0.196711 Objective Loss 0.196711 LR 0.000500 Time 0.021537 -2022-12-06 11:15:43,091 - Epoch: [119][ 400/ 1200] Overall Loss 0.196702 Objective Loss 0.196702 LR 0.000500 Time 0.021484 -2022-12-06 11:15:43,290 - Epoch: [119][ 410/ 1200] Overall Loss 0.196670 Objective Loss 0.196670 LR 0.000500 Time 0.021443 -2022-12-06 11:15:43,485 - Epoch: [119][ 420/ 1200] Overall Loss 0.196576 Objective Loss 0.196576 LR 0.000500 Time 0.021395 -2022-12-06 11:15:43,682 - Epoch: [119][ 430/ 1200] Overall Loss 0.196113 Objective Loss 0.196113 LR 0.000500 Time 0.021355 -2022-12-06 11:15:43,877 - Epoch: [119][ 440/ 1200] Overall Loss 0.196580 Objective Loss 0.196580 LR 0.000500 Time 0.021312 -2022-12-06 11:15:44,076 - Epoch: [119][ 450/ 1200] Overall Loss 0.196705 Objective Loss 0.196705 LR 0.000500 Time 0.021279 -2022-12-06 11:15:44,266 - Epoch: [119][ 460/ 1200] Overall Loss 0.196715 Objective Loss 0.196715 LR 0.000500 Time 0.021229 -2022-12-06 11:15:44,457 - Epoch: [119][ 470/ 1200] Overall Loss 0.196848 Objective Loss 0.196848 LR 0.000500 Time 0.021183 -2022-12-06 11:15:44,649 - Epoch: [119][ 480/ 1200] Overall Loss 0.197085 Objective Loss 0.197085 LR 0.000500 Time 0.021140 -2022-12-06 11:15:44,840 - Epoch: [119][ 490/ 1200] Overall Loss 0.196864 Objective Loss 0.196864 LR 0.000500 Time 0.021096 -2022-12-06 11:15:45,031 - Epoch: [119][ 500/ 1200] Overall Loss 0.196547 Objective Loss 0.196547 LR 0.000500 Time 0.021056 -2022-12-06 11:15:45,222 - Epoch: [119][ 510/ 1200] Overall Loss 0.196534 Objective Loss 0.196534 LR 0.000500 Time 0.021016 -2022-12-06 11:15:45,414 - Epoch: [119][ 520/ 1200] Overall Loss 0.196267 Objective Loss 0.196267 LR 0.000500 Time 0.020979 -2022-12-06 11:15:45,605 - Epoch: [119][ 530/ 1200] Overall Loss 0.196540 Objective Loss 0.196540 LR 0.000500 Time 0.020944 -2022-12-06 11:15:45,797 - Epoch: [119][ 540/ 1200] Overall Loss 0.196478 Objective Loss 0.196478 LR 0.000500 Time 0.020909 -2022-12-06 11:15:45,987 - Epoch: [119][ 550/ 1200] Overall Loss 0.196540 Objective Loss 0.196540 LR 0.000500 Time 0.020874 -2022-12-06 11:15:46,178 - Epoch: [119][ 560/ 1200] Overall Loss 0.196417 Objective Loss 0.196417 LR 0.000500 Time 0.020842 -2022-12-06 11:15:46,369 - Epoch: [119][ 570/ 1200] Overall Loss 0.196689 Objective Loss 0.196689 LR 0.000500 Time 0.020810 -2022-12-06 11:15:46,559 - Epoch: [119][ 580/ 1200] Overall Loss 0.197091 Objective Loss 0.197091 LR 0.000500 Time 0.020778 -2022-12-06 11:15:46,749 - Epoch: [119][ 590/ 1200] Overall Loss 0.196706 Objective Loss 0.196706 LR 0.000500 Time 0.020747 -2022-12-06 11:15:46,940 - Epoch: [119][ 600/ 1200] Overall Loss 0.196771 Objective Loss 0.196771 LR 0.000500 Time 0.020719 -2022-12-06 11:15:47,132 - Epoch: [119][ 610/ 1200] Overall Loss 0.197138 Objective Loss 0.197138 LR 0.000500 Time 0.020692 -2022-12-06 11:15:47,322 - Epoch: [119][ 620/ 1200] Overall Loss 0.197406 Objective Loss 0.197406 LR 0.000500 Time 0.020665 -2022-12-06 11:15:47,514 - Epoch: [119][ 630/ 1200] Overall Loss 0.197319 Objective Loss 0.197319 LR 0.000500 Time 0.020640 -2022-12-06 11:15:47,704 - Epoch: [119][ 640/ 1200] Overall Loss 0.197331 Objective Loss 0.197331 LR 0.000500 Time 0.020613 -2022-12-06 11:15:47,894 - Epoch: [119][ 650/ 1200] Overall Loss 0.197368 Objective Loss 0.197368 LR 0.000500 Time 0.020589 -2022-12-06 11:15:48,086 - Epoch: [119][ 660/ 1200] Overall Loss 0.197137 Objective Loss 0.197137 LR 0.000500 Time 0.020566 -2022-12-06 11:15:48,276 - Epoch: [119][ 670/ 1200] Overall Loss 0.196931 Objective Loss 0.196931 LR 0.000500 Time 0.020542 -2022-12-06 11:15:48,467 - Epoch: [119][ 680/ 1200] Overall Loss 0.197185 Objective Loss 0.197185 LR 0.000500 Time 0.020520 -2022-12-06 11:15:48,657 - Epoch: [119][ 690/ 1200] Overall Loss 0.196929 Objective Loss 0.196929 LR 0.000500 Time 0.020498 -2022-12-06 11:15:48,848 - Epoch: [119][ 700/ 1200] Overall Loss 0.196749 Objective Loss 0.196749 LR 0.000500 Time 0.020477 -2022-12-06 11:15:49,039 - Epoch: [119][ 710/ 1200] Overall Loss 0.196843 Objective Loss 0.196843 LR 0.000500 Time 0.020456 -2022-12-06 11:15:49,231 - Epoch: [119][ 720/ 1200] Overall Loss 0.196811 Objective Loss 0.196811 LR 0.000500 Time 0.020438 -2022-12-06 11:15:49,422 - Epoch: [119][ 730/ 1200] Overall Loss 0.196859 Objective Loss 0.196859 LR 0.000500 Time 0.020418 -2022-12-06 11:15:49,612 - Epoch: [119][ 740/ 1200] Overall Loss 0.196816 Objective Loss 0.196816 LR 0.000500 Time 0.020399 -2022-12-06 11:15:49,803 - Epoch: [119][ 750/ 1200] Overall Loss 0.196701 Objective Loss 0.196701 LR 0.000500 Time 0.020381 -2022-12-06 11:15:49,993 - Epoch: [119][ 760/ 1200] Overall Loss 0.196979 Objective Loss 0.196979 LR 0.000500 Time 0.020363 -2022-12-06 11:15:50,184 - Epoch: [119][ 770/ 1200] Overall Loss 0.196916 Objective Loss 0.196916 LR 0.000500 Time 0.020345 -2022-12-06 11:15:50,376 - Epoch: [119][ 780/ 1200] Overall Loss 0.196893 Objective Loss 0.196893 LR 0.000500 Time 0.020329 -2022-12-06 11:15:50,567 - Epoch: [119][ 790/ 1200] Overall Loss 0.196914 Objective Loss 0.196914 LR 0.000500 Time 0.020313 -2022-12-06 11:15:50,758 - Epoch: [119][ 800/ 1200] Overall Loss 0.197209 Objective Loss 0.197209 LR 0.000500 Time 0.020297 -2022-12-06 11:15:50,948 - Epoch: [119][ 810/ 1200] Overall Loss 0.197209 Objective Loss 0.197209 LR 0.000500 Time 0.020281 -2022-12-06 11:15:51,139 - Epoch: [119][ 820/ 1200] Overall Loss 0.196977 Objective Loss 0.196977 LR 0.000500 Time 0.020265 -2022-12-06 11:15:51,329 - Epoch: [119][ 830/ 1200] Overall Loss 0.196949 Objective Loss 0.196949 LR 0.000500 Time 0.020250 -2022-12-06 11:15:51,520 - Epoch: [119][ 840/ 1200] Overall Loss 0.196844 Objective Loss 0.196844 LR 0.000500 Time 0.020235 -2022-12-06 11:15:51,711 - Epoch: [119][ 850/ 1200] Overall Loss 0.196821 Objective Loss 0.196821 LR 0.000500 Time 0.020221 -2022-12-06 11:15:51,902 - Epoch: [119][ 860/ 1200] Overall Loss 0.197101 Objective Loss 0.197101 LR 0.000500 Time 0.020208 -2022-12-06 11:15:52,095 - Epoch: [119][ 870/ 1200] Overall Loss 0.197138 Objective Loss 0.197138 LR 0.000500 Time 0.020197 -2022-12-06 11:15:52,288 - Epoch: [119][ 880/ 1200] Overall Loss 0.197160 Objective Loss 0.197160 LR 0.000500 Time 0.020186 -2022-12-06 11:15:52,482 - Epoch: [119][ 890/ 1200] Overall Loss 0.197000 Objective Loss 0.197000 LR 0.000500 Time 0.020176 -2022-12-06 11:15:52,674 - Epoch: [119][ 900/ 1200] Overall Loss 0.197057 Objective Loss 0.197057 LR 0.000500 Time 0.020165 -2022-12-06 11:15:52,868 - Epoch: [119][ 910/ 1200] Overall Loss 0.197074 Objective Loss 0.197074 LR 0.000500 Time 0.020156 -2022-12-06 11:15:53,060 - Epoch: [119][ 920/ 1200] Overall Loss 0.197172 Objective Loss 0.197172 LR 0.000500 Time 0.020145 -2022-12-06 11:15:53,254 - Epoch: [119][ 930/ 1200] Overall Loss 0.197377 Objective Loss 0.197377 LR 0.000500 Time 0.020136 -2022-12-06 11:15:53,446 - Epoch: [119][ 940/ 1200] Overall Loss 0.197239 Objective Loss 0.197239 LR 0.000500 Time 0.020126 -2022-12-06 11:15:53,640 - Epoch: [119][ 950/ 1200] Overall Loss 0.197414 Objective Loss 0.197414 LR 0.000500 Time 0.020117 -2022-12-06 11:15:53,832 - Epoch: [119][ 960/ 1200] Overall Loss 0.197345 Objective Loss 0.197345 LR 0.000500 Time 0.020108 -2022-12-06 11:15:54,025 - Epoch: [119][ 970/ 1200] Overall Loss 0.197336 Objective Loss 0.197336 LR 0.000500 Time 0.020099 -2022-12-06 11:15:54,218 - Epoch: [119][ 980/ 1200] Overall Loss 0.197324 Objective Loss 0.197324 LR 0.000500 Time 0.020090 -2022-12-06 11:15:54,412 - Epoch: [119][ 990/ 1200] Overall Loss 0.197267 Objective Loss 0.197267 LR 0.000500 Time 0.020082 -2022-12-06 11:15:54,604 - Epoch: [119][ 1000/ 1200] Overall Loss 0.197289 Objective Loss 0.197289 LR 0.000500 Time 0.020073 -2022-12-06 11:15:54,798 - Epoch: [119][ 1010/ 1200] Overall Loss 0.197112 Objective Loss 0.197112 LR 0.000500 Time 0.020066 -2022-12-06 11:15:54,992 - Epoch: [119][ 1020/ 1200] Overall Loss 0.197259 Objective Loss 0.197259 LR 0.000500 Time 0.020058 -2022-12-06 11:15:55,185 - Epoch: [119][ 1030/ 1200] Overall Loss 0.197518 Objective Loss 0.197518 LR 0.000500 Time 0.020050 -2022-12-06 11:15:55,377 - Epoch: [119][ 1040/ 1200] Overall Loss 0.197537 Objective Loss 0.197537 LR 0.000500 Time 0.020042 -2022-12-06 11:15:55,571 - Epoch: [119][ 1050/ 1200] Overall Loss 0.197470 Objective Loss 0.197470 LR 0.000500 Time 0.020035 -2022-12-06 11:15:55,763 - Epoch: [119][ 1060/ 1200] Overall Loss 0.197375 Objective Loss 0.197375 LR 0.000500 Time 0.020027 -2022-12-06 11:15:55,957 - Epoch: [119][ 1070/ 1200] Overall Loss 0.197341 Objective Loss 0.197341 LR 0.000500 Time 0.020020 -2022-12-06 11:15:56,149 - Epoch: [119][ 1080/ 1200] Overall Loss 0.197343 Objective Loss 0.197343 LR 0.000500 Time 0.020013 -2022-12-06 11:15:56,343 - Epoch: [119][ 1090/ 1200] Overall Loss 0.197319 Objective Loss 0.197319 LR 0.000500 Time 0.020006 -2022-12-06 11:15:56,536 - Epoch: [119][ 1100/ 1200] Overall Loss 0.197582 Objective Loss 0.197582 LR 0.000500 Time 0.019999 -2022-12-06 11:15:56,729 - Epoch: [119][ 1110/ 1200] Overall Loss 0.197848 Objective Loss 0.197848 LR 0.000500 Time 0.019993 -2022-12-06 11:15:56,922 - Epoch: [119][ 1120/ 1200] Overall Loss 0.197957 Objective Loss 0.197957 LR 0.000500 Time 0.019986 -2022-12-06 11:15:57,116 - Epoch: [119][ 1130/ 1200] Overall Loss 0.197812 Objective Loss 0.197812 LR 0.000500 Time 0.019980 -2022-12-06 11:15:57,309 - Epoch: [119][ 1140/ 1200] Overall Loss 0.197463 Objective Loss 0.197463 LR 0.000500 Time 0.019973 -2022-12-06 11:15:57,502 - Epoch: [119][ 1150/ 1200] Overall Loss 0.197546 Objective Loss 0.197546 LR 0.000500 Time 0.019967 -2022-12-06 11:15:57,694 - Epoch: [119][ 1160/ 1200] Overall Loss 0.197461 Objective Loss 0.197461 LR 0.000500 Time 0.019961 -2022-12-06 11:15:57,888 - Epoch: [119][ 1170/ 1200] Overall Loss 0.197517 Objective Loss 0.197517 LR 0.000500 Time 0.019955 -2022-12-06 11:15:58,080 - Epoch: [119][ 1180/ 1200] Overall Loss 0.197623 Objective Loss 0.197623 LR 0.000500 Time 0.019949 -2022-12-06 11:15:58,274 - Epoch: [119][ 1190/ 1200] Overall Loss 0.197764 Objective Loss 0.197764 LR 0.000500 Time 0.019943 -2022-12-06 11:15:58,505 - Epoch: [119][ 1200/ 1200] Overall Loss 0.197888 Objective Loss 0.197888 Top1 91.213389 Top5 98.953975 LR 0.000500 Time 0.019969 -2022-12-06 11:15:58,594 - --- validate (epoch=119)----------- -2022-12-06 11:15:58,594 - 34129 samples (256 per mini-batch) -2022-12-06 11:15:59,037 - Epoch: [119][ 10/ 134] Loss 0.280517 Top1 86.132812 Top5 98.398438 -2022-12-06 11:15:59,165 - Epoch: [119][ 20/ 134] Loss 0.279226 Top1 86.250000 Top5 98.574219 -2022-12-06 11:15:59,299 - Epoch: [119][ 30/ 134] Loss 0.264065 Top1 86.549479 Top5 98.567708 -2022-12-06 11:15:59,442 - Epoch: [119][ 40/ 134] Loss 0.261926 Top1 86.376953 Top5 98.437500 -2022-12-06 11:15:59,570 - Epoch: [119][ 50/ 134] Loss 0.262586 Top1 86.179688 Top5 98.390625 -2022-12-06 11:15:59,697 - Epoch: [119][ 60/ 134] Loss 0.265407 Top1 86.269531 Top5 98.385417 -2022-12-06 11:15:59,841 - Epoch: [119][ 70/ 134] Loss 0.260578 Top1 86.478795 Top5 98.454241 -2022-12-06 11:15:59,982 - Epoch: [119][ 80/ 134] Loss 0.260658 Top1 86.425781 Top5 98.476562 -2022-12-06 11:16:00,107 - Epoch: [119][ 90/ 134] Loss 0.259466 Top1 86.545139 Top5 98.472222 -2022-12-06 11:16:00,231 - Epoch: [119][ 100/ 134] Loss 0.255990 Top1 86.746094 Top5 98.441406 -2022-12-06 11:16:00,356 - Epoch: [119][ 110/ 134] Loss 0.257508 Top1 86.722301 Top5 98.433949 -2022-12-06 11:16:00,480 - Epoch: [119][ 120/ 134] Loss 0.257543 Top1 86.829427 Top5 98.408203 -2022-12-06 11:16:00,607 - Epoch: [119][ 130/ 134] Loss 0.256388 Top1 86.847957 Top5 98.398438 -2022-12-06 11:16:00,643 - Epoch: [119][ 134/ 134] Loss 0.256859 Top1 86.823523 Top5 98.373817 -2022-12-06 11:16:00,730 - ==> Top1: 86.824 Top5: 98.374 Loss: 0.257 - -2022-12-06 11:16:00,731 - ==> Confusion: -[[ 919 0 1 2 5 6 0 0 9 39 0 1 1 3 3 2 0 0 0 0 5] - [ 3 946 2 2 7 15 4 8 1 1 4 5 1 1 0 1 1 0 11 2 12] - [ 4 3 1011 8 5 0 24 4 0 0 7 3 2 1 4 4 1 1 6 3 12] - [ 4 0 15 925 2 2 2 2 2 1 15 0 7 1 15 1 0 1 15 0 10] - [ 11 6 2 0 951 5 1 1 1 6 1 3 1 1 11 8 6 0 0 1 4] - [ 2 24 0 4 7 973 1 17 2 4 0 8 6 8 2 1 2 1 1 2 4] - [ 1 0 7 2 0 2 1083 3 0 0 2 2 0 2 0 5 0 1 1 6 1] - [ 1 7 6 1 2 19 10 944 0 1 3 5 2 1 0 1 2 0 33 7 9] - [ 4 3 0 0 0 4 0 0 986 34 8 1 3 3 12 1 1 0 2 1 1] - [ 64 0 3 0 7 3 0 3 30 870 1 1 0 5 8 1 0 1 0 0 4] - [ 1 1 2 0 2 2 2 1 7 1 974 2 0 8 2 0 2 0 7 2 3] - [ 1 0 0 0 2 15 5 1 1 0 3 966 26 2 0 9 2 3 0 12 3] - [ 0 0 2 2 0 2 1 1 1 0 0 35 896 1 0 7 2 9 0 4 6] - [ 1 0 0 0 0 7 0 2 16 13 10 6 5 943 1 1 5 1 0 2 10] - [ 5 2 2 7 3 2 0 0 24 1 1 1 2 1 1064 1 0 0 6 1 7] - [ 1 0 1 1 3 0 6 0 0 0 0 6 3 4 1 995 5 9 0 4 4] - [ 3 2 1 2 2 0 1 0 1 0 1 1 1 1 0 13 1030 0 0 8 5] - [ 2 1 3 2 0 0 1 0 0 3 0 9 23 3 4 17 2 959 2 2 3] - [ 4 4 4 7 2 2 1 16 3 1 11 0 2 0 6 1 0 0 938 1 5] - [ 1 2 2 2 3 5 9 7 0 0 3 8 7 3 0 4 4 2 0 1010 8] - [ 122 194 191 96 103 134 99 131 93 94 198 100 326 239 151 107 179 71 149 206 10243]] - -2022-12-06 11:16:01,297 - ==> Best [Top1: 86.824 Top5: 98.374 Sparsity:0.00 Params: 5376 on epoch: 119] -2022-12-06 11:16:01,297 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:16:01,339 - - -2022-12-06 11:16:01,339 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:16:02,389 - Epoch: [120][ 10/ 1200] Overall Loss 0.204445 Objective Loss 0.204445 LR 0.000500 Time 0.104936 -2022-12-06 11:16:02,593 - Epoch: [120][ 20/ 1200] Overall Loss 0.193536 Objective Loss 0.193536 LR 0.000500 Time 0.062665 -2022-12-06 11:16:02,794 - Epoch: [120][ 30/ 1200] Overall Loss 0.203690 Objective Loss 0.203690 LR 0.000500 Time 0.048450 -2022-12-06 11:16:02,999 - Epoch: [120][ 40/ 1200] Overall Loss 0.197584 Objective Loss 0.197584 LR 0.000500 Time 0.041442 -2022-12-06 11:16:03,199 - Epoch: [120][ 50/ 1200] Overall Loss 0.198032 Objective Loss 0.198032 LR 0.000500 Time 0.037144 -2022-12-06 11:16:03,404 - Epoch: [120][ 60/ 1200] Overall Loss 0.198253 Objective Loss 0.198253 LR 0.000500 Time 0.034364 -2022-12-06 11:16:03,605 - Epoch: [120][ 70/ 1200] Overall Loss 0.197103 Objective Loss 0.197103 LR 0.000500 Time 0.032312 -2022-12-06 11:16:03,810 - Epoch: [120][ 80/ 1200] Overall Loss 0.197365 Objective Loss 0.197365 LR 0.000500 Time 0.030823 -2022-12-06 11:16:04,011 - Epoch: [120][ 90/ 1200] Overall Loss 0.196868 Objective Loss 0.196868 LR 0.000500 Time 0.029629 -2022-12-06 11:16:04,216 - Epoch: [120][ 100/ 1200] Overall Loss 0.195827 Objective Loss 0.195827 LR 0.000500 Time 0.028707 -2022-12-06 11:16:04,417 - Epoch: [120][ 110/ 1200] Overall Loss 0.194061 Objective Loss 0.194061 LR 0.000500 Time 0.027924 -2022-12-06 11:16:04,622 - Epoch: [120][ 120/ 1200] Overall Loss 0.193613 Objective Loss 0.193613 LR 0.000500 Time 0.027297 -2022-12-06 11:16:04,823 - Epoch: [120][ 130/ 1200] Overall Loss 0.195246 Objective Loss 0.195246 LR 0.000500 Time 0.026739 -2022-12-06 11:16:05,028 - Epoch: [120][ 140/ 1200] Overall Loss 0.194718 Objective Loss 0.194718 LR 0.000500 Time 0.026290 -2022-12-06 11:16:05,229 - Epoch: [120][ 150/ 1200] Overall Loss 0.194392 Objective Loss 0.194392 LR 0.000500 Time 0.025872 -2022-12-06 11:16:05,433 - Epoch: [120][ 160/ 1200] Overall Loss 0.195385 Objective Loss 0.195385 LR 0.000500 Time 0.025529 -2022-12-06 11:16:05,633 - Epoch: [120][ 170/ 1200] Overall Loss 0.195687 Objective Loss 0.195687 LR 0.000500 Time 0.025201 -2022-12-06 11:16:05,838 - Epoch: [120][ 180/ 1200] Overall Loss 0.194260 Objective Loss 0.194260 LR 0.000500 Time 0.024937 -2022-12-06 11:16:06,039 - Epoch: [120][ 190/ 1200] Overall Loss 0.192616 Objective Loss 0.192616 LR 0.000500 Time 0.024675 -2022-12-06 11:16:06,243 - Epoch: [120][ 200/ 1200] Overall Loss 0.193933 Objective Loss 0.193933 LR 0.000500 Time 0.024458 -2022-12-06 11:16:06,444 - Epoch: [120][ 210/ 1200] Overall Loss 0.193935 Objective Loss 0.193935 LR 0.000500 Time 0.024249 -2022-12-06 11:16:06,649 - Epoch: [120][ 220/ 1200] Overall Loss 0.194201 Objective Loss 0.194201 LR 0.000500 Time 0.024075 -2022-12-06 11:16:06,849 - Epoch: [120][ 230/ 1200] Overall Loss 0.193282 Objective Loss 0.193282 LR 0.000500 Time 0.023897 -2022-12-06 11:16:07,054 - Epoch: [120][ 240/ 1200] Overall Loss 0.192683 Objective Loss 0.192683 LR 0.000500 Time 0.023752 -2022-12-06 11:16:07,255 - Epoch: [120][ 250/ 1200] Overall Loss 0.192352 Objective Loss 0.192352 LR 0.000500 Time 0.023605 -2022-12-06 11:16:07,460 - Epoch: [120][ 260/ 1200] Overall Loss 0.192680 Objective Loss 0.192680 LR 0.000500 Time 0.023483 -2022-12-06 11:16:07,662 - Epoch: [120][ 270/ 1200] Overall Loss 0.193031 Objective Loss 0.193031 LR 0.000500 Time 0.023357 -2022-12-06 11:16:07,867 - Epoch: [120][ 280/ 1200] Overall Loss 0.192917 Objective Loss 0.192917 LR 0.000500 Time 0.023253 -2022-12-06 11:16:08,067 - Epoch: [120][ 290/ 1200] Overall Loss 0.193506 Objective Loss 0.193506 LR 0.000500 Time 0.023138 -2022-12-06 11:16:08,271 - Epoch: [120][ 300/ 1200] Overall Loss 0.193657 Objective Loss 0.193657 LR 0.000500 Time 0.023045 -2022-12-06 11:16:08,471 - Epoch: [120][ 310/ 1200] Overall Loss 0.193671 Objective Loss 0.193671 LR 0.000500 Time 0.022946 -2022-12-06 11:16:08,675 - Epoch: [120][ 320/ 1200] Overall Loss 0.193879 Objective Loss 0.193879 LR 0.000500 Time 0.022864 -2022-12-06 11:16:08,876 - Epoch: [120][ 330/ 1200] Overall Loss 0.194386 Objective Loss 0.194386 LR 0.000500 Time 0.022779 -2022-12-06 11:16:09,081 - Epoch: [120][ 340/ 1200] Overall Loss 0.194417 Objective Loss 0.194417 LR 0.000500 Time 0.022710 -2022-12-06 11:16:09,281 - Epoch: [120][ 350/ 1200] Overall Loss 0.194374 Objective Loss 0.194374 LR 0.000500 Time 0.022631 -2022-12-06 11:16:09,485 - Epoch: [120][ 360/ 1200] Overall Loss 0.194404 Objective Loss 0.194404 LR 0.000500 Time 0.022569 -2022-12-06 11:16:09,686 - Epoch: [120][ 370/ 1200] Overall Loss 0.194227 Objective Loss 0.194227 LR 0.000500 Time 0.022500 -2022-12-06 11:16:09,890 - Epoch: [120][ 380/ 1200] Overall Loss 0.194437 Objective Loss 0.194437 LR 0.000500 Time 0.022443 -2022-12-06 11:16:10,092 - Epoch: [120][ 390/ 1200] Overall Loss 0.194625 Objective Loss 0.194625 LR 0.000500 Time 0.022382 -2022-12-06 11:16:10,296 - Epoch: [120][ 400/ 1200] Overall Loss 0.194510 Objective Loss 0.194510 LR 0.000500 Time 0.022333 -2022-12-06 11:16:10,498 - Epoch: [120][ 410/ 1200] Overall Loss 0.194829 Objective Loss 0.194829 LR 0.000500 Time 0.022278 -2022-12-06 11:16:10,703 - Epoch: [120][ 420/ 1200] Overall Loss 0.194817 Objective Loss 0.194817 LR 0.000500 Time 0.022234 -2022-12-06 11:16:10,903 - Epoch: [120][ 430/ 1200] Overall Loss 0.195209 Objective Loss 0.195209 LR 0.000500 Time 0.022182 -2022-12-06 11:16:11,109 - Epoch: [120][ 440/ 1200] Overall Loss 0.194892 Objective Loss 0.194892 LR 0.000500 Time 0.022144 -2022-12-06 11:16:11,310 - Epoch: [120][ 450/ 1200] Overall Loss 0.194962 Objective Loss 0.194962 LR 0.000500 Time 0.022098 -2022-12-06 11:16:11,516 - Epoch: [120][ 460/ 1200] Overall Loss 0.195146 Objective Loss 0.195146 LR 0.000500 Time 0.022063 -2022-12-06 11:16:11,716 - Epoch: [120][ 470/ 1200] Overall Loss 0.195231 Objective Loss 0.195231 LR 0.000500 Time 0.022019 -2022-12-06 11:16:11,921 - Epoch: [120][ 480/ 1200] Overall Loss 0.195675 Objective Loss 0.195675 LR 0.000500 Time 0.021985 -2022-12-06 11:16:12,120 - Epoch: [120][ 490/ 1200] Overall Loss 0.195645 Objective Loss 0.195645 LR 0.000500 Time 0.021942 -2022-12-06 11:16:12,325 - Epoch: [120][ 500/ 1200] Overall Loss 0.195921 Objective Loss 0.195921 LR 0.000500 Time 0.021912 -2022-12-06 11:16:12,525 - Epoch: [120][ 510/ 1200] Overall Loss 0.196030 Objective Loss 0.196030 LR 0.000500 Time 0.021874 -2022-12-06 11:16:12,730 - Epoch: [120][ 520/ 1200] Overall Loss 0.195723 Objective Loss 0.195723 LR 0.000500 Time 0.021846 -2022-12-06 11:16:12,930 - Epoch: [120][ 530/ 1200] Overall Loss 0.195752 Objective Loss 0.195752 LR 0.000500 Time 0.021809 -2022-12-06 11:16:13,134 - Epoch: [120][ 540/ 1200] Overall Loss 0.195564 Objective Loss 0.195564 LR 0.000500 Time 0.021782 -2022-12-06 11:16:13,335 - Epoch: [120][ 550/ 1200] Overall Loss 0.196003 Objective Loss 0.196003 LR 0.000500 Time 0.021750 -2022-12-06 11:16:13,528 - Epoch: [120][ 560/ 1200] Overall Loss 0.196566 Objective Loss 0.196566 LR 0.000500 Time 0.021706 -2022-12-06 11:16:13,718 - Epoch: [120][ 570/ 1200] Overall Loss 0.196664 Objective Loss 0.196664 LR 0.000500 Time 0.021658 -2022-12-06 11:16:13,909 - Epoch: [120][ 580/ 1200] Overall Loss 0.197238 Objective Loss 0.197238 LR 0.000500 Time 0.021612 -2022-12-06 11:16:14,100 - Epoch: [120][ 590/ 1200] Overall Loss 0.197279 Objective Loss 0.197279 LR 0.000500 Time 0.021569 -2022-12-06 11:16:14,291 - Epoch: [120][ 600/ 1200] Overall Loss 0.197579 Objective Loss 0.197579 LR 0.000500 Time 0.021527 -2022-12-06 11:16:14,482 - Epoch: [120][ 610/ 1200] Overall Loss 0.197706 Objective Loss 0.197706 LR 0.000500 Time 0.021486 -2022-12-06 11:16:14,672 - Epoch: [120][ 620/ 1200] Overall Loss 0.198005 Objective Loss 0.198005 LR 0.000500 Time 0.021445 -2022-12-06 11:16:14,862 - Epoch: [120][ 630/ 1200] Overall Loss 0.197890 Objective Loss 0.197890 LR 0.000500 Time 0.021406 -2022-12-06 11:16:15,052 - Epoch: [120][ 640/ 1200] Overall Loss 0.197752 Objective Loss 0.197752 LR 0.000500 Time 0.021367 -2022-12-06 11:16:15,243 - Epoch: [120][ 650/ 1200] Overall Loss 0.197702 Objective Loss 0.197702 LR 0.000500 Time 0.021331 -2022-12-06 11:16:15,434 - Epoch: [120][ 660/ 1200] Overall Loss 0.197951 Objective Loss 0.197951 LR 0.000500 Time 0.021297 -2022-12-06 11:16:15,626 - Epoch: [120][ 670/ 1200] Overall Loss 0.197478 Objective Loss 0.197478 LR 0.000500 Time 0.021264 -2022-12-06 11:16:15,816 - Epoch: [120][ 680/ 1200] Overall Loss 0.197664 Objective Loss 0.197664 LR 0.000500 Time 0.021231 -2022-12-06 11:16:16,007 - Epoch: [120][ 690/ 1200] Overall Loss 0.197937 Objective Loss 0.197937 LR 0.000500 Time 0.021199 -2022-12-06 11:16:16,197 - Epoch: [120][ 700/ 1200] Overall Loss 0.197809 Objective Loss 0.197809 LR 0.000500 Time 0.021167 -2022-12-06 11:16:16,389 - Epoch: [120][ 710/ 1200] Overall Loss 0.197857 Objective Loss 0.197857 LR 0.000500 Time 0.021138 -2022-12-06 11:16:16,580 - Epoch: [120][ 720/ 1200] Overall Loss 0.197738 Objective Loss 0.197738 LR 0.000500 Time 0.021108 -2022-12-06 11:16:16,770 - Epoch: [120][ 730/ 1200] Overall Loss 0.197806 Objective Loss 0.197806 LR 0.000500 Time 0.021080 -2022-12-06 11:16:16,961 - Epoch: [120][ 740/ 1200] Overall Loss 0.197852 Objective Loss 0.197852 LR 0.000500 Time 0.021052 -2022-12-06 11:16:17,152 - Epoch: [120][ 750/ 1200] Overall Loss 0.197786 Objective Loss 0.197786 LR 0.000500 Time 0.021025 -2022-12-06 11:16:17,343 - Epoch: [120][ 760/ 1200] Overall Loss 0.197454 Objective Loss 0.197454 LR 0.000500 Time 0.020998 -2022-12-06 11:16:17,534 - Epoch: [120][ 770/ 1200] Overall Loss 0.197166 Objective Loss 0.197166 LR 0.000500 Time 0.020973 -2022-12-06 11:16:17,724 - Epoch: [120][ 780/ 1200] Overall Loss 0.197226 Objective Loss 0.197226 LR 0.000500 Time 0.020948 -2022-12-06 11:16:17,914 - Epoch: [120][ 790/ 1200] Overall Loss 0.197168 Objective Loss 0.197168 LR 0.000500 Time 0.020922 -2022-12-06 11:16:18,105 - Epoch: [120][ 800/ 1200] Overall Loss 0.197495 Objective Loss 0.197495 LR 0.000500 Time 0.020898 -2022-12-06 11:16:18,295 - Epoch: [120][ 810/ 1200] Overall Loss 0.197674 Objective Loss 0.197674 LR 0.000500 Time 0.020875 -2022-12-06 11:16:18,487 - Epoch: [120][ 820/ 1200] Overall Loss 0.197391 Objective Loss 0.197391 LR 0.000500 Time 0.020853 -2022-12-06 11:16:18,678 - Epoch: [120][ 830/ 1200] Overall Loss 0.197354 Objective Loss 0.197354 LR 0.000500 Time 0.020831 -2022-12-06 11:16:18,868 - Epoch: [120][ 840/ 1200] Overall Loss 0.197660 Objective Loss 0.197660 LR 0.000500 Time 0.020809 -2022-12-06 11:16:19,059 - Epoch: [120][ 850/ 1200] Overall Loss 0.197648 Objective Loss 0.197648 LR 0.000500 Time 0.020788 -2022-12-06 11:16:19,249 - Epoch: [120][ 860/ 1200] Overall Loss 0.197523 Objective Loss 0.197523 LR 0.000500 Time 0.020767 -2022-12-06 11:16:19,440 - Epoch: [120][ 870/ 1200] Overall Loss 0.197478 Objective Loss 0.197478 LR 0.000500 Time 0.020748 -2022-12-06 11:16:19,631 - Epoch: [120][ 880/ 1200] Overall Loss 0.197672 Objective Loss 0.197672 LR 0.000500 Time 0.020728 -2022-12-06 11:16:19,822 - Epoch: [120][ 890/ 1200] Overall Loss 0.197713 Objective Loss 0.197713 LR 0.000500 Time 0.020709 -2022-12-06 11:16:20,012 - Epoch: [120][ 900/ 1200] Overall Loss 0.197564 Objective Loss 0.197564 LR 0.000500 Time 0.020690 -2022-12-06 11:16:20,203 - Epoch: [120][ 910/ 1200] Overall Loss 0.197708 Objective Loss 0.197708 LR 0.000500 Time 0.020672 -2022-12-06 11:16:20,394 - Epoch: [120][ 920/ 1200] Overall Loss 0.197656 Objective Loss 0.197656 LR 0.000500 Time 0.020654 -2022-12-06 11:16:20,584 - Epoch: [120][ 930/ 1200] Overall Loss 0.197883 Objective Loss 0.197883 LR 0.000500 Time 0.020636 -2022-12-06 11:16:20,776 - Epoch: [120][ 940/ 1200] Overall Loss 0.197872 Objective Loss 0.197872 LR 0.000500 Time 0.020619 -2022-12-06 11:16:20,966 - Epoch: [120][ 950/ 1200] Overall Loss 0.198280 Objective Loss 0.198280 LR 0.000500 Time 0.020602 -2022-12-06 11:16:21,156 - Epoch: [120][ 960/ 1200] Overall Loss 0.198063 Objective Loss 0.198063 LR 0.000500 Time 0.020585 -2022-12-06 11:16:21,347 - Epoch: [120][ 970/ 1200] Overall Loss 0.198230 Objective Loss 0.198230 LR 0.000500 Time 0.020569 -2022-12-06 11:16:21,537 - Epoch: [120][ 980/ 1200] Overall Loss 0.198385 Objective Loss 0.198385 LR 0.000500 Time 0.020553 -2022-12-06 11:16:21,728 - Epoch: [120][ 990/ 1200] Overall Loss 0.198269 Objective Loss 0.198269 LR 0.000500 Time 0.020537 -2022-12-06 11:16:21,918 - Epoch: [120][ 1000/ 1200] Overall Loss 0.198442 Objective Loss 0.198442 LR 0.000500 Time 0.020521 -2022-12-06 11:16:22,108 - Epoch: [120][ 1010/ 1200] Overall Loss 0.198309 Objective Loss 0.198309 LR 0.000500 Time 0.020505 -2022-12-06 11:16:22,299 - Epoch: [120][ 1020/ 1200] Overall Loss 0.198431 Objective Loss 0.198431 LR 0.000500 Time 0.020491 -2022-12-06 11:16:22,489 - Epoch: [120][ 1030/ 1200] Overall Loss 0.198171 Objective Loss 0.198171 LR 0.000500 Time 0.020477 -2022-12-06 11:16:22,679 - Epoch: [120][ 1040/ 1200] Overall Loss 0.198378 Objective Loss 0.198378 LR 0.000500 Time 0.020462 -2022-12-06 11:16:22,870 - Epoch: [120][ 1050/ 1200] Overall Loss 0.198541 Objective Loss 0.198541 LR 0.000500 Time 0.020448 -2022-12-06 11:16:23,060 - Epoch: [120][ 1060/ 1200] Overall Loss 0.198712 Objective Loss 0.198712 LR 0.000500 Time 0.020434 -2022-12-06 11:16:23,250 - Epoch: [120][ 1070/ 1200] Overall Loss 0.198634 Objective Loss 0.198634 LR 0.000500 Time 0.020420 -2022-12-06 11:16:23,440 - Epoch: [120][ 1080/ 1200] Overall Loss 0.198511 Objective Loss 0.198511 LR 0.000500 Time 0.020406 -2022-12-06 11:16:23,630 - Epoch: [120][ 1090/ 1200] Overall Loss 0.198517 Objective Loss 0.198517 LR 0.000500 Time 0.020393 -2022-12-06 11:16:23,820 - Epoch: [120][ 1100/ 1200] Overall Loss 0.198591 Objective Loss 0.198591 LR 0.000500 Time 0.020380 -2022-12-06 11:16:24,011 - Epoch: [120][ 1110/ 1200] Overall Loss 0.198455 Objective Loss 0.198455 LR 0.000500 Time 0.020367 -2022-12-06 11:16:24,201 - Epoch: [120][ 1120/ 1200] Overall Loss 0.198383 Objective Loss 0.198383 LR 0.000500 Time 0.020355 -2022-12-06 11:16:24,392 - Epoch: [120][ 1130/ 1200] Overall Loss 0.198423 Objective Loss 0.198423 LR 0.000500 Time 0.020343 -2022-12-06 11:16:24,583 - Epoch: [120][ 1140/ 1200] Overall Loss 0.198551 Objective Loss 0.198551 LR 0.000500 Time 0.020332 -2022-12-06 11:16:24,773 - Epoch: [120][ 1150/ 1200] Overall Loss 0.198833 Objective Loss 0.198833 LR 0.000500 Time 0.020320 -2022-12-06 11:16:24,964 - Epoch: [120][ 1160/ 1200] Overall Loss 0.198827 Objective Loss 0.198827 LR 0.000500 Time 0.020309 -2022-12-06 11:16:25,154 - Epoch: [120][ 1170/ 1200] Overall Loss 0.198796 Objective Loss 0.198796 LR 0.000500 Time 0.020297 -2022-12-06 11:16:25,344 - Epoch: [120][ 1180/ 1200] Overall Loss 0.198698 Objective Loss 0.198698 LR 0.000500 Time 0.020286 -2022-12-06 11:16:25,534 - Epoch: [120][ 1190/ 1200] Overall Loss 0.198416 Objective Loss 0.198416 LR 0.000500 Time 0.020274 -2022-12-06 11:16:25,759 - Epoch: [120][ 1200/ 1200] Overall Loss 0.198483 Objective Loss 0.198483 Top1 86.820084 Top5 99.372385 LR 0.000500 Time 0.020293 -2022-12-06 11:16:25,847 - --- validate (epoch=120)----------- -2022-12-06 11:16:25,847 - 34129 samples (256 per mini-batch) -2022-12-06 11:16:26,290 - Epoch: [120][ 10/ 134] Loss 0.270055 Top1 86.601562 Top5 97.968750 -2022-12-06 11:16:26,419 - Epoch: [120][ 20/ 134] Loss 0.245319 Top1 86.523438 Top5 98.476562 -2022-12-06 11:16:26,545 - Epoch: [120][ 30/ 134] Loss 0.243292 Top1 86.875000 Top5 98.528646 -2022-12-06 11:16:26,671 - Epoch: [120][ 40/ 134] Loss 0.251429 Top1 86.591797 Top5 98.427734 -2022-12-06 11:16:26,797 - Epoch: [120][ 50/ 134] Loss 0.251170 Top1 86.429688 Top5 98.367188 -2022-12-06 11:16:26,923 - Epoch: [120][ 60/ 134] Loss 0.250438 Top1 86.549479 Top5 98.372396 -2022-12-06 11:16:27,048 - Epoch: [120][ 70/ 134] Loss 0.253834 Top1 86.506696 Top5 98.292411 -2022-12-06 11:16:27,174 - Epoch: [120][ 80/ 134] Loss 0.258725 Top1 86.406250 Top5 98.295898 -2022-12-06 11:16:27,300 - Epoch: [120][ 90/ 134] Loss 0.258476 Top1 86.471354 Top5 98.281250 -2022-12-06 11:16:27,427 - Epoch: [120][ 100/ 134] Loss 0.255394 Top1 86.515625 Top5 98.316406 -2022-12-06 11:16:27,551 - Epoch: [120][ 110/ 134] Loss 0.255669 Top1 86.445312 Top5 98.316761 -2022-12-06 11:16:27,677 - Epoch: [120][ 120/ 134] Loss 0.255524 Top1 86.490885 Top5 98.326823 -2022-12-06 11:16:27,802 - Epoch: [120][ 130/ 134] Loss 0.254943 Top1 86.541466 Top5 98.308293 -2022-12-06 11:16:27,838 - Epoch: [120][ 134/ 134] Loss 0.255975 Top1 86.486566 Top5 98.285915 -2022-12-06 11:16:27,933 - ==> Top1: 86.487 Top5: 98.286 Loss: 0.256 - -2022-12-06 11:16:27,934 - ==> Confusion: -[[ 897 2 1 3 4 4 0 0 9 56 0 1 1 3 6 1 1 0 2 0 5] - [ 2 948 1 2 7 22 4 9 1 1 2 6 1 0 0 1 1 0 10 3 6] - [ 3 4 1015 10 6 2 15 9 0 3 6 4 2 2 4 3 0 2 5 0 8] - [ 2 1 12 957 2 4 0 0 0 1 8 0 5 0 8 0 2 1 11 0 6] - [ 12 8 3 0 942 7 1 1 1 8 1 1 0 4 12 6 6 1 0 2 4] - [ 1 15 0 2 7 983 3 15 3 4 1 6 5 14 1 2 1 0 1 1 4] - [ 0 5 8 0 2 2 1068 5 0 0 1 2 2 3 0 7 1 2 1 7 2] - [ 1 15 9 2 2 33 8 930 1 1 0 4 1 1 1 0 2 0 26 11 6] - [ 5 3 0 0 0 3 0 1 965 46 3 1 5 8 18 0 1 0 3 1 1] - [ 50 0 1 1 4 2 0 2 25 894 1 1 0 12 2 0 0 1 0 0 5] - [ 1 1 2 12 2 3 1 3 13 0 943 1 1 12 7 0 1 0 9 2 5] - [ 5 0 2 1 0 14 1 2 0 0 0 967 27 6 0 4 3 8 0 10 1] - [ 0 1 0 3 1 3 0 1 1 0 0 21 916 3 1 3 1 6 0 2 6] - [ 0 1 0 0 1 11 0 3 6 9 7 1 5 962 2 1 4 0 0 2 8] - [ 6 3 2 16 2 3 0 0 13 3 0 1 6 5 1054 0 0 1 9 1 5] - [ 0 0 1 1 2 4 2 0 0 0 0 3 11 2 0 988 6 10 1 5 7] - [ 2 4 1 2 2 2 1 0 1 0 0 1 4 1 1 10 1026 0 0 6 8] - [ 2 0 1 3 1 1 1 1 1 1 0 7 17 2 1 11 0 980 1 3 2] - [ 3 5 2 8 3 5 1 18 3 1 7 2 2 0 7 2 0 1 936 1 1] - [ 4 2 1 1 1 7 6 8 0 0 1 12 7 6 0 4 4 1 0 1008 7] - [ 118 220 167 135 97 195 78 149 81 77 130 103 384 276 168 98 166 85 174 187 10138]] - -2022-12-06 11:16:28,503 - ==> Best [Top1: 86.824 Top5: 98.374 Sparsity:0.00 Params: 5376 on epoch: 119] -2022-12-06 11:16:28,503 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:16:28,510 - - -2022-12-06 11:16:28,510 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:16:29,439 - Epoch: [121][ 10/ 1200] Overall Loss 0.182638 Objective Loss 0.182638 LR 0.000500 Time 0.092898 -2022-12-06 11:16:29,636 - Epoch: [121][ 20/ 1200] Overall Loss 0.189203 Objective Loss 0.189203 LR 0.000500 Time 0.056249 -2022-12-06 11:16:29,828 - Epoch: [121][ 30/ 1200] Overall Loss 0.193126 Objective Loss 0.193126 LR 0.000500 Time 0.043901 -2022-12-06 11:16:30,022 - Epoch: [121][ 40/ 1200] Overall Loss 0.193388 Objective Loss 0.193388 LR 0.000500 Time 0.037739 -2022-12-06 11:16:30,213 - Epoch: [121][ 50/ 1200] Overall Loss 0.194890 Objective Loss 0.194890 LR 0.000500 Time 0.034012 -2022-12-06 11:16:30,405 - Epoch: [121][ 60/ 1200] Overall Loss 0.197146 Objective Loss 0.197146 LR 0.000500 Time 0.031536 -2022-12-06 11:16:30,597 - Epoch: [121][ 70/ 1200] Overall Loss 0.198721 Objective Loss 0.198721 LR 0.000500 Time 0.029761 -2022-12-06 11:16:30,789 - Epoch: [121][ 80/ 1200] Overall Loss 0.199177 Objective Loss 0.199177 LR 0.000500 Time 0.028436 -2022-12-06 11:16:30,980 - Epoch: [121][ 90/ 1200] Overall Loss 0.201272 Objective Loss 0.201272 LR 0.000500 Time 0.027397 -2022-12-06 11:16:31,173 - Epoch: [121][ 100/ 1200] Overall Loss 0.201126 Objective Loss 0.201126 LR 0.000500 Time 0.026576 -2022-12-06 11:16:31,363 - Epoch: [121][ 110/ 1200] Overall Loss 0.200295 Objective Loss 0.200295 LR 0.000500 Time 0.025885 -2022-12-06 11:16:31,556 - Epoch: [121][ 120/ 1200] Overall Loss 0.199006 Objective Loss 0.199006 LR 0.000500 Time 0.025332 -2022-12-06 11:16:31,748 - Epoch: [121][ 130/ 1200] Overall Loss 0.198393 Objective Loss 0.198393 LR 0.000500 Time 0.024853 -2022-12-06 11:16:31,940 - Epoch: [121][ 140/ 1200] Overall Loss 0.199922 Objective Loss 0.199922 LR 0.000500 Time 0.024445 -2022-12-06 11:16:32,131 - Epoch: [121][ 150/ 1200] Overall Loss 0.197739 Objective Loss 0.197739 LR 0.000500 Time 0.024089 -2022-12-06 11:16:32,323 - Epoch: [121][ 160/ 1200] Overall Loss 0.197094 Objective Loss 0.197094 LR 0.000500 Time 0.023779 -2022-12-06 11:16:32,515 - Epoch: [121][ 170/ 1200] Overall Loss 0.196725 Objective Loss 0.196725 LR 0.000500 Time 0.023505 -2022-12-06 11:16:32,707 - Epoch: [121][ 180/ 1200] Overall Loss 0.197761 Objective Loss 0.197761 LR 0.000500 Time 0.023264 -2022-12-06 11:16:32,899 - Epoch: [121][ 190/ 1200] Overall Loss 0.197035 Objective Loss 0.197035 LR 0.000500 Time 0.023045 -2022-12-06 11:16:33,091 - Epoch: [121][ 200/ 1200] Overall Loss 0.196200 Objective Loss 0.196200 LR 0.000500 Time 0.022852 -2022-12-06 11:16:33,283 - Epoch: [121][ 210/ 1200] Overall Loss 0.196275 Objective Loss 0.196275 LR 0.000500 Time 0.022673 -2022-12-06 11:16:33,475 - Epoch: [121][ 220/ 1200] Overall Loss 0.196317 Objective Loss 0.196317 LR 0.000500 Time 0.022515 -2022-12-06 11:16:33,667 - Epoch: [121][ 230/ 1200] Overall Loss 0.196187 Objective Loss 0.196187 LR 0.000500 Time 0.022368 -2022-12-06 11:16:33,859 - Epoch: [121][ 240/ 1200] Overall Loss 0.195944 Objective Loss 0.195944 LR 0.000500 Time 0.022234 -2022-12-06 11:16:34,051 - Epoch: [121][ 250/ 1200] Overall Loss 0.195097 Objective Loss 0.195097 LR 0.000500 Time 0.022108 -2022-12-06 11:16:34,243 - Epoch: [121][ 260/ 1200] Overall Loss 0.194575 Objective Loss 0.194575 LR 0.000500 Time 0.021995 -2022-12-06 11:16:34,435 - Epoch: [121][ 270/ 1200] Overall Loss 0.195302 Objective Loss 0.195302 LR 0.000500 Time 0.021889 -2022-12-06 11:16:34,627 - Epoch: [121][ 280/ 1200] Overall Loss 0.195444 Objective Loss 0.195444 LR 0.000500 Time 0.021790 -2022-12-06 11:16:34,819 - Epoch: [121][ 290/ 1200] Overall Loss 0.195796 Objective Loss 0.195796 LR 0.000500 Time 0.021698 -2022-12-06 11:16:35,011 - Epoch: [121][ 300/ 1200] Overall Loss 0.196525 Objective Loss 0.196525 LR 0.000500 Time 0.021613 -2022-12-06 11:16:35,203 - Epoch: [121][ 310/ 1200] Overall Loss 0.196448 Objective Loss 0.196448 LR 0.000500 Time 0.021535 -2022-12-06 11:16:35,395 - Epoch: [121][ 320/ 1200] Overall Loss 0.196183 Objective Loss 0.196183 LR 0.000500 Time 0.021459 -2022-12-06 11:16:35,586 - Epoch: [121][ 330/ 1200] Overall Loss 0.195414 Objective Loss 0.195414 LR 0.000500 Time 0.021387 -2022-12-06 11:16:35,778 - Epoch: [121][ 340/ 1200] Overall Loss 0.195045 Objective Loss 0.195045 LR 0.000500 Time 0.021322 -2022-12-06 11:16:35,970 - Epoch: [121][ 350/ 1200] Overall Loss 0.194220 Objective Loss 0.194220 LR 0.000500 Time 0.021259 -2022-12-06 11:16:36,163 - Epoch: [121][ 360/ 1200] Overall Loss 0.194229 Objective Loss 0.194229 LR 0.000500 Time 0.021202 -2022-12-06 11:16:36,354 - Epoch: [121][ 370/ 1200] Overall Loss 0.194370 Objective Loss 0.194370 LR 0.000500 Time 0.021145 -2022-12-06 11:16:36,547 - Epoch: [121][ 380/ 1200] Overall Loss 0.194216 Objective Loss 0.194216 LR 0.000500 Time 0.021094 -2022-12-06 11:16:36,739 - Epoch: [121][ 390/ 1200] Overall Loss 0.193809 Objective Loss 0.193809 LR 0.000500 Time 0.021044 -2022-12-06 11:16:36,932 - Epoch: [121][ 400/ 1200] Overall Loss 0.193346 Objective Loss 0.193346 LR 0.000500 Time 0.020999 -2022-12-06 11:16:37,123 - Epoch: [121][ 410/ 1200] Overall Loss 0.193703 Objective Loss 0.193703 LR 0.000500 Time 0.020953 -2022-12-06 11:16:37,316 - Epoch: [121][ 420/ 1200] Overall Loss 0.193265 Objective Loss 0.193265 LR 0.000500 Time 0.020910 -2022-12-06 11:16:37,508 - Epoch: [121][ 430/ 1200] Overall Loss 0.193998 Objective Loss 0.193998 LR 0.000500 Time 0.020869 -2022-12-06 11:16:37,700 - Epoch: [121][ 440/ 1200] Overall Loss 0.194073 Objective Loss 0.194073 LR 0.000500 Time 0.020832 -2022-12-06 11:16:37,892 - Epoch: [121][ 450/ 1200] Overall Loss 0.194196 Objective Loss 0.194196 LR 0.000500 Time 0.020792 -2022-12-06 11:16:38,084 - Epoch: [121][ 460/ 1200] Overall Loss 0.193890 Objective Loss 0.193890 LR 0.000500 Time 0.020757 -2022-12-06 11:16:38,275 - Epoch: [121][ 470/ 1200] Overall Loss 0.193991 Objective Loss 0.193991 LR 0.000500 Time 0.020721 -2022-12-06 11:16:38,468 - Epoch: [121][ 480/ 1200] Overall Loss 0.193545 Objective Loss 0.193545 LR 0.000500 Time 0.020690 -2022-12-06 11:16:38,660 - Epoch: [121][ 490/ 1200] Overall Loss 0.193505 Objective Loss 0.193505 LR 0.000500 Time 0.020659 -2022-12-06 11:16:38,853 - Epoch: [121][ 500/ 1200] Overall Loss 0.193392 Objective Loss 0.193392 LR 0.000500 Time 0.020629 -2022-12-06 11:16:39,045 - Epoch: [121][ 510/ 1200] Overall Loss 0.194001 Objective Loss 0.194001 LR 0.000500 Time 0.020601 -2022-12-06 11:16:39,237 - Epoch: [121][ 520/ 1200] Overall Loss 0.193976 Objective Loss 0.193976 LR 0.000500 Time 0.020574 -2022-12-06 11:16:39,429 - Epoch: [121][ 530/ 1200] Overall Loss 0.194098 Objective Loss 0.194098 LR 0.000500 Time 0.020546 -2022-12-06 11:16:39,622 - Epoch: [121][ 540/ 1200] Overall Loss 0.194171 Objective Loss 0.194171 LR 0.000500 Time 0.020521 -2022-12-06 11:16:39,814 - Epoch: [121][ 550/ 1200] Overall Loss 0.194083 Objective Loss 0.194083 LR 0.000500 Time 0.020497 -2022-12-06 11:16:40,006 - Epoch: [121][ 560/ 1200] Overall Loss 0.194273 Objective Loss 0.194273 LR 0.000500 Time 0.020473 -2022-12-06 11:16:40,198 - Epoch: [121][ 570/ 1200] Overall Loss 0.194055 Objective Loss 0.194055 LR 0.000500 Time 0.020449 -2022-12-06 11:16:40,390 - Epoch: [121][ 580/ 1200] Overall Loss 0.193861 Objective Loss 0.193861 LR 0.000500 Time 0.020427 -2022-12-06 11:16:40,582 - Epoch: [121][ 590/ 1200] Overall Loss 0.194007 Objective Loss 0.194007 LR 0.000500 Time 0.020405 -2022-12-06 11:16:40,774 - Epoch: [121][ 600/ 1200] Overall Loss 0.193727 Objective Loss 0.193727 LR 0.000500 Time 0.020384 -2022-12-06 11:16:40,966 - Epoch: [121][ 610/ 1200] Overall Loss 0.193875 Objective Loss 0.193875 LR 0.000500 Time 0.020364 -2022-12-06 11:16:41,158 - Epoch: [121][ 620/ 1200] Overall Loss 0.193869 Objective Loss 0.193869 LR 0.000500 Time 0.020344 -2022-12-06 11:16:41,350 - Epoch: [121][ 630/ 1200] Overall Loss 0.193995 Objective Loss 0.193995 LR 0.000500 Time 0.020325 -2022-12-06 11:16:41,543 - Epoch: [121][ 640/ 1200] Overall Loss 0.193965 Objective Loss 0.193965 LR 0.000500 Time 0.020308 -2022-12-06 11:16:41,735 - Epoch: [121][ 650/ 1200] Overall Loss 0.194008 Objective Loss 0.194008 LR 0.000500 Time 0.020290 -2022-12-06 11:16:41,927 - Epoch: [121][ 660/ 1200] Overall Loss 0.194035 Objective Loss 0.194035 LR 0.000500 Time 0.020273 -2022-12-06 11:16:42,118 - Epoch: [121][ 670/ 1200] Overall Loss 0.194668 Objective Loss 0.194668 LR 0.000500 Time 0.020255 -2022-12-06 11:16:42,310 - Epoch: [121][ 680/ 1200] Overall Loss 0.194786 Objective Loss 0.194786 LR 0.000500 Time 0.020239 -2022-12-06 11:16:42,502 - Epoch: [121][ 690/ 1200] Overall Loss 0.194997 Objective Loss 0.194997 LR 0.000500 Time 0.020222 -2022-12-06 11:16:42,694 - Epoch: [121][ 700/ 1200] Overall Loss 0.195016 Objective Loss 0.195016 LR 0.000500 Time 0.020207 -2022-12-06 11:16:42,886 - Epoch: [121][ 710/ 1200] Overall Loss 0.195243 Objective Loss 0.195243 LR 0.000500 Time 0.020192 -2022-12-06 11:16:43,083 - Epoch: [121][ 720/ 1200] Overall Loss 0.195570 Objective Loss 0.195570 LR 0.000500 Time 0.020185 -2022-12-06 11:16:43,286 - Epoch: [121][ 730/ 1200] Overall Loss 0.195344 Objective Loss 0.195344 LR 0.000500 Time 0.020185 -2022-12-06 11:16:43,491 - Epoch: [121][ 740/ 1200] Overall Loss 0.195366 Objective Loss 0.195366 LR 0.000500 Time 0.020188 -2022-12-06 11:16:43,693 - Epoch: [121][ 750/ 1200] Overall Loss 0.194880 Objective Loss 0.194880 LR 0.000500 Time 0.020188 -2022-12-06 11:16:43,899 - Epoch: [121][ 760/ 1200] Overall Loss 0.195019 Objective Loss 0.195019 LR 0.000500 Time 0.020192 -2022-12-06 11:16:44,100 - Epoch: [121][ 770/ 1200] Overall Loss 0.195001 Objective Loss 0.195001 LR 0.000500 Time 0.020191 -2022-12-06 11:16:44,305 - Epoch: [121][ 780/ 1200] Overall Loss 0.195281 Objective Loss 0.195281 LR 0.000500 Time 0.020194 -2022-12-06 11:16:44,507 - Epoch: [121][ 790/ 1200] Overall Loss 0.195409 Objective Loss 0.195409 LR 0.000500 Time 0.020193 -2022-12-06 11:16:44,712 - Epoch: [121][ 800/ 1200] Overall Loss 0.195474 Objective Loss 0.195474 LR 0.000500 Time 0.020196 -2022-12-06 11:16:44,914 - Epoch: [121][ 810/ 1200] Overall Loss 0.195504 Objective Loss 0.195504 LR 0.000500 Time 0.020195 -2022-12-06 11:16:45,119 - Epoch: [121][ 820/ 1200] Overall Loss 0.195509 Objective Loss 0.195509 LR 0.000500 Time 0.020198 -2022-12-06 11:16:45,322 - Epoch: [121][ 830/ 1200] Overall Loss 0.195621 Objective Loss 0.195621 LR 0.000500 Time 0.020198 -2022-12-06 11:16:45,526 - Epoch: [121][ 840/ 1200] Overall Loss 0.195907 Objective Loss 0.195907 LR 0.000500 Time 0.020201 -2022-12-06 11:16:45,728 - Epoch: [121][ 850/ 1200] Overall Loss 0.195760 Objective Loss 0.195760 LR 0.000500 Time 0.020200 -2022-12-06 11:16:45,932 - Epoch: [121][ 860/ 1200] Overall Loss 0.196047 Objective Loss 0.196047 LR 0.000500 Time 0.020201 -2022-12-06 11:16:46,133 - Epoch: [121][ 870/ 1200] Overall Loss 0.196312 Objective Loss 0.196312 LR 0.000500 Time 0.020200 -2022-12-06 11:16:46,339 - Epoch: [121][ 880/ 1200] Overall Loss 0.196134 Objective Loss 0.196134 LR 0.000500 Time 0.020204 -2022-12-06 11:16:46,542 - Epoch: [121][ 890/ 1200] Overall Loss 0.196203 Objective Loss 0.196203 LR 0.000500 Time 0.020204 -2022-12-06 11:16:46,747 - Epoch: [121][ 900/ 1200] Overall Loss 0.196143 Objective Loss 0.196143 LR 0.000500 Time 0.020207 -2022-12-06 11:16:46,949 - Epoch: [121][ 910/ 1200] Overall Loss 0.196393 Objective Loss 0.196393 LR 0.000500 Time 0.020205 -2022-12-06 11:16:47,153 - Epoch: [121][ 920/ 1200] Overall Loss 0.196049 Objective Loss 0.196049 LR 0.000500 Time 0.020207 -2022-12-06 11:16:47,355 - Epoch: [121][ 930/ 1200] Overall Loss 0.196221 Objective Loss 0.196221 LR 0.000500 Time 0.020206 -2022-12-06 11:16:47,560 - Epoch: [121][ 940/ 1200] Overall Loss 0.196160 Objective Loss 0.196160 LR 0.000500 Time 0.020209 -2022-12-06 11:16:47,763 - Epoch: [121][ 950/ 1200] Overall Loss 0.196341 Objective Loss 0.196341 LR 0.000500 Time 0.020209 -2022-12-06 11:16:47,969 - Epoch: [121][ 960/ 1200] Overall Loss 0.196315 Objective Loss 0.196315 LR 0.000500 Time 0.020212 -2022-12-06 11:16:48,170 - Epoch: [121][ 970/ 1200] Overall Loss 0.196325 Objective Loss 0.196325 LR 0.000500 Time 0.020211 -2022-12-06 11:16:48,375 - Epoch: [121][ 980/ 1200] Overall Loss 0.196234 Objective Loss 0.196234 LR 0.000500 Time 0.020213 -2022-12-06 11:16:48,578 - Epoch: [121][ 990/ 1200] Overall Loss 0.196097 Objective Loss 0.196097 LR 0.000500 Time 0.020213 -2022-12-06 11:16:48,783 - Epoch: [121][ 1000/ 1200] Overall Loss 0.196086 Objective Loss 0.196086 LR 0.000500 Time 0.020216 -2022-12-06 11:16:48,985 - Epoch: [121][ 1010/ 1200] Overall Loss 0.196130 Objective Loss 0.196130 LR 0.000500 Time 0.020215 -2022-12-06 11:16:49,191 - Epoch: [121][ 1020/ 1200] Overall Loss 0.196092 Objective Loss 0.196092 LR 0.000500 Time 0.020218 -2022-12-06 11:16:49,393 - Epoch: [121][ 1030/ 1200] Overall Loss 0.196264 Objective Loss 0.196264 LR 0.000500 Time 0.020217 -2022-12-06 11:16:49,598 - Epoch: [121][ 1040/ 1200] Overall Loss 0.196450 Objective Loss 0.196450 LR 0.000500 Time 0.020220 -2022-12-06 11:16:49,801 - Epoch: [121][ 1050/ 1200] Overall Loss 0.196613 Objective Loss 0.196613 LR 0.000500 Time 0.020220 -2022-12-06 11:16:50,007 - Epoch: [121][ 1060/ 1200] Overall Loss 0.196688 Objective Loss 0.196688 LR 0.000500 Time 0.020222 -2022-12-06 11:16:50,209 - Epoch: [121][ 1070/ 1200] Overall Loss 0.196731 Objective Loss 0.196731 LR 0.000500 Time 0.020222 -2022-12-06 11:16:50,415 - Epoch: [121][ 1080/ 1200] Overall Loss 0.196446 Objective Loss 0.196446 LR 0.000500 Time 0.020225 -2022-12-06 11:16:50,616 - Epoch: [121][ 1090/ 1200] Overall Loss 0.196499 Objective Loss 0.196499 LR 0.000500 Time 0.020223 -2022-12-06 11:16:50,821 - Epoch: [121][ 1100/ 1200] Overall Loss 0.196533 Objective Loss 0.196533 LR 0.000500 Time 0.020225 -2022-12-06 11:16:51,022 - Epoch: [121][ 1110/ 1200] Overall Loss 0.196590 Objective Loss 0.196590 LR 0.000500 Time 0.020224 -2022-12-06 11:16:51,227 - Epoch: [121][ 1120/ 1200] Overall Loss 0.196709 Objective Loss 0.196709 LR 0.000500 Time 0.020226 -2022-12-06 11:16:51,429 - Epoch: [121][ 1130/ 1200] Overall Loss 0.196746 Objective Loss 0.196746 LR 0.000500 Time 0.020225 -2022-12-06 11:16:51,633 - Epoch: [121][ 1140/ 1200] Overall Loss 0.196800 Objective Loss 0.196800 LR 0.000500 Time 0.020226 -2022-12-06 11:16:51,836 - Epoch: [121][ 1150/ 1200] Overall Loss 0.196630 Objective Loss 0.196630 LR 0.000500 Time 0.020225 -2022-12-06 11:16:52,041 - Epoch: [121][ 1160/ 1200] Overall Loss 0.196819 Objective Loss 0.196819 LR 0.000500 Time 0.020228 -2022-12-06 11:16:52,243 - Epoch: [121][ 1170/ 1200] Overall Loss 0.196853 Objective Loss 0.196853 LR 0.000500 Time 0.020226 -2022-12-06 11:16:52,448 - Epoch: [121][ 1180/ 1200] Overall Loss 0.196871 Objective Loss 0.196871 LR 0.000500 Time 0.020228 -2022-12-06 11:16:52,648 - Epoch: [121][ 1190/ 1200] Overall Loss 0.196974 Objective Loss 0.196974 LR 0.000500 Time 0.020226 -2022-12-06 11:16:52,883 - Epoch: [121][ 1200/ 1200] Overall Loss 0.196758 Objective Loss 0.196758 Top1 90.794979 Top5 99.581590 LR 0.000500 Time 0.020253 -2022-12-06 11:16:52,972 - --- validate (epoch=121)----------- -2022-12-06 11:16:52,972 - 34129 samples (256 per mini-batch) -2022-12-06 11:16:53,425 - Epoch: [121][ 10/ 134] Loss 0.254368 Top1 87.109375 Top5 98.125000 -2022-12-06 11:16:53,554 - Epoch: [121][ 20/ 134] Loss 0.255201 Top1 86.796875 Top5 98.144531 -2022-12-06 11:16:53,682 - Epoch: [121][ 30/ 134] Loss 0.250681 Top1 86.927083 Top5 98.216146 -2022-12-06 11:16:53,826 - Epoch: [121][ 40/ 134] Loss 0.252869 Top1 86.992188 Top5 98.203125 -2022-12-06 11:16:53,965 - Epoch: [121][ 50/ 134] Loss 0.252220 Top1 86.882812 Top5 98.218750 -2022-12-06 11:16:54,098 - Epoch: [121][ 60/ 134] Loss 0.249268 Top1 86.972656 Top5 98.242188 -2022-12-06 11:16:54,230 - Epoch: [121][ 70/ 134] Loss 0.252445 Top1 86.902902 Top5 98.208705 -2022-12-06 11:16:54,361 - Epoch: [121][ 80/ 134] Loss 0.250196 Top1 87.036133 Top5 98.281250 -2022-12-06 11:16:54,490 - Epoch: [121][ 90/ 134] Loss 0.252591 Top1 86.970486 Top5 98.311632 -2022-12-06 11:16:54,621 - Epoch: [121][ 100/ 134] Loss 0.254235 Top1 86.996094 Top5 98.277344 -2022-12-06 11:16:54,752 - Epoch: [121][ 110/ 134] Loss 0.253794 Top1 86.981534 Top5 98.281250 -2022-12-06 11:16:54,885 - Epoch: [121][ 120/ 134] Loss 0.253493 Top1 86.992188 Top5 98.271484 -2022-12-06 11:16:55,016 - Epoch: [121][ 130/ 134] Loss 0.252981 Top1 87.064303 Top5 98.290264 -2022-12-06 11:16:55,054 - Epoch: [121][ 134/ 134] Loss 0.253650 Top1 87.040347 Top5 98.288845 -2022-12-06 11:16:55,142 - ==> Top1: 87.040 Top5: 98.289 Loss: 0.254 - -2022-12-06 11:16:55,142 - ==> Confusion: -[[ 900 2 1 4 4 5 0 0 8 58 0 2 0 2 4 2 0 0 0 0 4] - [ 1 941 1 3 6 20 3 11 2 2 7 4 0 1 0 1 2 1 13 1 7] - [ 3 3 1017 5 3 2 14 6 0 3 9 3 3 2 7 2 0 2 2 3 14] - [ 4 1 21 930 0 4 0 0 1 0 12 0 2 3 15 0 2 2 16 0 7] - [ 14 7 2 0 951 5 1 0 1 6 2 2 0 3 9 3 7 1 2 1 3] - [ 3 23 0 3 5 961 1 15 6 2 0 9 5 19 2 0 3 0 1 4 7] - [ 2 2 10 1 1 2 1079 1 0 0 1 3 1 2 0 3 0 0 2 4 4] - [ 1 16 7 1 2 26 3 947 0 0 3 6 1 3 0 1 0 1 23 8 5] - [ 4 2 0 1 0 3 0 0 986 38 8 2 2 2 10 1 1 0 2 1 1] - [ 46 0 1 0 5 2 0 1 26 898 1 1 0 9 3 0 0 2 0 1 5] - [ 1 1 3 1 1 1 0 3 11 1 971 0 1 8 3 0 0 0 9 1 3] - [ 2 1 2 0 1 12 2 1 2 0 1 976 23 6 0 5 3 6 0 6 2] - [ 1 1 1 3 1 4 0 0 0 0 0 25 901 2 0 7 2 10 0 3 8] - [ 0 1 0 0 0 9 0 3 17 16 9 1 1 949 3 1 1 0 1 2 9] - [ 11 6 2 5 2 4 0 0 14 2 1 2 1 2 1064 0 1 2 4 2 5] - [ 3 0 3 1 5 1 2 0 0 0 0 10 6 1 0 990 6 7 0 3 5] - [ 3 2 0 2 3 1 1 0 1 0 0 3 1 1 1 7 1037 0 0 3 6] - [ 3 0 2 4 0 1 0 0 1 2 0 5 17 2 1 11 1 981 2 1 2] - [ 2 4 5 11 1 3 0 24 2 1 6 0 3 0 10 0 0 1 931 0 4] - [ 4 4 1 1 0 6 7 8 0 1 1 13 8 4 0 3 3 4 3 1000 9] - [ 134 207 182 111 102 147 77 126 116 89 201 100 289 236 145 101 168 83 153 168 10291]] - -2022-12-06 11:16:55,818 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:16:55,818 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:16:55,825 - - -2022-12-06 11:16:55,825 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:16:56,758 - Epoch: [122][ 10/ 1200] Overall Loss 0.187178 Objective Loss 0.187178 LR 0.000500 Time 0.093172 -2022-12-06 11:16:56,950 - Epoch: [122][ 20/ 1200] Overall Loss 0.203138 Objective Loss 0.203138 LR 0.000500 Time 0.056147 -2022-12-06 11:16:57,141 - Epoch: [122][ 30/ 1200] Overall Loss 0.194277 Objective Loss 0.194277 LR 0.000500 Time 0.043797 -2022-12-06 11:16:57,332 - Epoch: [122][ 40/ 1200] Overall Loss 0.191921 Objective Loss 0.191921 LR 0.000500 Time 0.037618 -2022-12-06 11:16:57,524 - Epoch: [122][ 50/ 1200] Overall Loss 0.190027 Objective Loss 0.190027 LR 0.000500 Time 0.033908 -2022-12-06 11:16:57,715 - Epoch: [122][ 60/ 1200] Overall Loss 0.189901 Objective Loss 0.189901 LR 0.000500 Time 0.031436 -2022-12-06 11:16:57,905 - Epoch: [122][ 70/ 1200] Overall Loss 0.191736 Objective Loss 0.191736 LR 0.000500 Time 0.029658 -2022-12-06 11:16:58,096 - Epoch: [122][ 80/ 1200] Overall Loss 0.189619 Objective Loss 0.189619 LR 0.000500 Time 0.028330 -2022-12-06 11:16:58,288 - Epoch: [122][ 90/ 1200] Overall Loss 0.191081 Objective Loss 0.191081 LR 0.000500 Time 0.027301 -2022-12-06 11:16:58,478 - Epoch: [122][ 100/ 1200] Overall Loss 0.190423 Objective Loss 0.190423 LR 0.000500 Time 0.026473 -2022-12-06 11:16:58,670 - Epoch: [122][ 110/ 1200] Overall Loss 0.190183 Objective Loss 0.190183 LR 0.000500 Time 0.025805 -2022-12-06 11:16:58,862 - Epoch: [122][ 120/ 1200] Overall Loss 0.190706 Objective Loss 0.190706 LR 0.000500 Time 0.025249 -2022-12-06 11:16:59,053 - Epoch: [122][ 130/ 1200] Overall Loss 0.190429 Objective Loss 0.190429 LR 0.000500 Time 0.024774 -2022-12-06 11:16:59,244 - Epoch: [122][ 140/ 1200] Overall Loss 0.190040 Objective Loss 0.190040 LR 0.000500 Time 0.024363 -2022-12-06 11:16:59,435 - Epoch: [122][ 150/ 1200] Overall Loss 0.189514 Objective Loss 0.189514 LR 0.000500 Time 0.024009 -2022-12-06 11:16:59,626 - Epoch: [122][ 160/ 1200] Overall Loss 0.190615 Objective Loss 0.190615 LR 0.000500 Time 0.023700 -2022-12-06 11:16:59,817 - Epoch: [122][ 170/ 1200] Overall Loss 0.191263 Objective Loss 0.191263 LR 0.000500 Time 0.023426 -2022-12-06 11:17:00,007 - Epoch: [122][ 180/ 1200] Overall Loss 0.189806 Objective Loss 0.189806 LR 0.000500 Time 0.023176 -2022-12-06 11:17:00,198 - Epoch: [122][ 190/ 1200] Overall Loss 0.190026 Objective Loss 0.190026 LR 0.000500 Time 0.022956 -2022-12-06 11:17:00,389 - Epoch: [122][ 200/ 1200] Overall Loss 0.190661 Objective Loss 0.190661 LR 0.000500 Time 0.022761 -2022-12-06 11:17:00,580 - Epoch: [122][ 210/ 1200] Overall Loss 0.191375 Objective Loss 0.191375 LR 0.000500 Time 0.022587 -2022-12-06 11:17:00,771 - Epoch: [122][ 220/ 1200] Overall Loss 0.191000 Objective Loss 0.191000 LR 0.000500 Time 0.022426 -2022-12-06 11:17:00,963 - Epoch: [122][ 230/ 1200] Overall Loss 0.190551 Objective Loss 0.190551 LR 0.000500 Time 0.022280 -2022-12-06 11:17:01,154 - Epoch: [122][ 240/ 1200] Overall Loss 0.190404 Objective Loss 0.190404 LR 0.000500 Time 0.022146 -2022-12-06 11:17:01,344 - Epoch: [122][ 250/ 1200] Overall Loss 0.190683 Objective Loss 0.190683 LR 0.000500 Time 0.022021 -2022-12-06 11:17:01,535 - Epoch: [122][ 260/ 1200] Overall Loss 0.191140 Objective Loss 0.191140 LR 0.000500 Time 0.021905 -2022-12-06 11:17:01,727 - Epoch: [122][ 270/ 1200] Overall Loss 0.192100 Objective Loss 0.192100 LR 0.000500 Time 0.021802 -2022-12-06 11:17:01,918 - Epoch: [122][ 280/ 1200] Overall Loss 0.192038 Objective Loss 0.192038 LR 0.000500 Time 0.021703 -2022-12-06 11:17:02,108 - Epoch: [122][ 290/ 1200] Overall Loss 0.192998 Objective Loss 0.192998 LR 0.000500 Time 0.021610 -2022-12-06 11:17:02,299 - Epoch: [122][ 300/ 1200] Overall Loss 0.193635 Objective Loss 0.193635 LR 0.000500 Time 0.021525 -2022-12-06 11:17:02,490 - Epoch: [122][ 310/ 1200] Overall Loss 0.194049 Objective Loss 0.194049 LR 0.000500 Time 0.021444 -2022-12-06 11:17:02,681 - Epoch: [122][ 320/ 1200] Overall Loss 0.193855 Objective Loss 0.193855 LR 0.000500 Time 0.021370 -2022-12-06 11:17:02,873 - Epoch: [122][ 330/ 1200] Overall Loss 0.193907 Objective Loss 0.193907 LR 0.000500 Time 0.021301 -2022-12-06 11:17:03,064 - Epoch: [122][ 340/ 1200] Overall Loss 0.194339 Objective Loss 0.194339 LR 0.000500 Time 0.021234 -2022-12-06 11:17:03,254 - Epoch: [122][ 350/ 1200] Overall Loss 0.194002 Objective Loss 0.194002 LR 0.000500 Time 0.021171 -2022-12-06 11:17:03,446 - Epoch: [122][ 360/ 1200] Overall Loss 0.194017 Objective Loss 0.194017 LR 0.000500 Time 0.021112 -2022-12-06 11:17:03,637 - Epoch: [122][ 370/ 1200] Overall Loss 0.194357 Objective Loss 0.194357 LR 0.000500 Time 0.021057 -2022-12-06 11:17:03,828 - Epoch: [122][ 380/ 1200] Overall Loss 0.194158 Objective Loss 0.194158 LR 0.000500 Time 0.021003 -2022-12-06 11:17:04,019 - Epoch: [122][ 390/ 1200] Overall Loss 0.194096 Objective Loss 0.194096 LR 0.000500 Time 0.020954 -2022-12-06 11:17:04,210 - Epoch: [122][ 400/ 1200] Overall Loss 0.194272 Objective Loss 0.194272 LR 0.000500 Time 0.020906 -2022-12-06 11:17:04,401 - Epoch: [122][ 410/ 1200] Overall Loss 0.194478 Objective Loss 0.194478 LR 0.000500 Time 0.020861 -2022-12-06 11:17:04,592 - Epoch: [122][ 420/ 1200] Overall Loss 0.194356 Objective Loss 0.194356 LR 0.000500 Time 0.020817 -2022-12-06 11:17:04,783 - Epoch: [122][ 430/ 1200] Overall Loss 0.194144 Objective Loss 0.194144 LR 0.000500 Time 0.020776 -2022-12-06 11:17:04,974 - Epoch: [122][ 440/ 1200] Overall Loss 0.194071 Objective Loss 0.194071 LR 0.000500 Time 0.020738 -2022-12-06 11:17:05,165 - Epoch: [122][ 450/ 1200] Overall Loss 0.194067 Objective Loss 0.194067 LR 0.000500 Time 0.020701 -2022-12-06 11:17:05,357 - Epoch: [122][ 460/ 1200] Overall Loss 0.194495 Objective Loss 0.194495 LR 0.000500 Time 0.020666 -2022-12-06 11:17:05,548 - Epoch: [122][ 470/ 1200] Overall Loss 0.194404 Objective Loss 0.194404 LR 0.000500 Time 0.020631 -2022-12-06 11:17:05,739 - Epoch: [122][ 480/ 1200] Overall Loss 0.194305 Objective Loss 0.194305 LR 0.000500 Time 0.020599 -2022-12-06 11:17:05,930 - Epoch: [122][ 490/ 1200] Overall Loss 0.193971 Objective Loss 0.193971 LR 0.000500 Time 0.020566 -2022-12-06 11:17:06,121 - Epoch: [122][ 500/ 1200] Overall Loss 0.193795 Objective Loss 0.193795 LR 0.000500 Time 0.020537 -2022-12-06 11:17:06,313 - Epoch: [122][ 510/ 1200] Overall Loss 0.194388 Objective Loss 0.194388 LR 0.000500 Time 0.020508 -2022-12-06 11:17:06,504 - Epoch: [122][ 520/ 1200] Overall Loss 0.193984 Objective Loss 0.193984 LR 0.000500 Time 0.020481 -2022-12-06 11:17:06,695 - Epoch: [122][ 530/ 1200] Overall Loss 0.193885 Objective Loss 0.193885 LR 0.000500 Time 0.020454 -2022-12-06 11:17:06,886 - Epoch: [122][ 540/ 1200] Overall Loss 0.193568 Objective Loss 0.193568 LR 0.000500 Time 0.020428 -2022-12-06 11:17:07,077 - Epoch: [122][ 550/ 1200] Overall Loss 0.193222 Objective Loss 0.193222 LR 0.000500 Time 0.020403 -2022-12-06 11:17:07,269 - Epoch: [122][ 560/ 1200] Overall Loss 0.192869 Objective Loss 0.192869 LR 0.000500 Time 0.020379 -2022-12-06 11:17:07,459 - Epoch: [122][ 570/ 1200] Overall Loss 0.193175 Objective Loss 0.193175 LR 0.000500 Time 0.020355 -2022-12-06 11:17:07,650 - Epoch: [122][ 580/ 1200] Overall Loss 0.193285 Objective Loss 0.193285 LR 0.000500 Time 0.020333 -2022-12-06 11:17:07,842 - Epoch: [122][ 590/ 1200] Overall Loss 0.193572 Objective Loss 0.193572 LR 0.000500 Time 0.020312 -2022-12-06 11:17:08,032 - Epoch: [122][ 600/ 1200] Overall Loss 0.193676 Objective Loss 0.193676 LR 0.000500 Time 0.020290 -2022-12-06 11:17:08,224 - Epoch: [122][ 610/ 1200] Overall Loss 0.193541 Objective Loss 0.193541 LR 0.000500 Time 0.020270 -2022-12-06 11:17:08,414 - Epoch: [122][ 620/ 1200] Overall Loss 0.193322 Objective Loss 0.193322 LR 0.000500 Time 0.020250 -2022-12-06 11:17:08,605 - Epoch: [122][ 630/ 1200] Overall Loss 0.193396 Objective Loss 0.193396 LR 0.000500 Time 0.020231 -2022-12-06 11:17:08,795 - Epoch: [122][ 640/ 1200] Overall Loss 0.193651 Objective Loss 0.193651 LR 0.000500 Time 0.020211 -2022-12-06 11:17:08,987 - Epoch: [122][ 650/ 1200] Overall Loss 0.193362 Objective Loss 0.193362 LR 0.000500 Time 0.020193 -2022-12-06 11:17:09,177 - Epoch: [122][ 660/ 1200] Overall Loss 0.193118 Objective Loss 0.193118 LR 0.000500 Time 0.020175 -2022-12-06 11:17:09,368 - Epoch: [122][ 670/ 1200] Overall Loss 0.193187 Objective Loss 0.193187 LR 0.000500 Time 0.020158 -2022-12-06 11:17:09,559 - Epoch: [122][ 680/ 1200] Overall Loss 0.193014 Objective Loss 0.193014 LR 0.000500 Time 0.020141 -2022-12-06 11:17:09,750 - Epoch: [122][ 690/ 1200] Overall Loss 0.193200 Objective Loss 0.193200 LR 0.000500 Time 0.020126 -2022-12-06 11:17:09,941 - Epoch: [122][ 700/ 1200] Overall Loss 0.193193 Objective Loss 0.193193 LR 0.000500 Time 0.020110 -2022-12-06 11:17:10,132 - Epoch: [122][ 710/ 1200] Overall Loss 0.193271 Objective Loss 0.193271 LR 0.000500 Time 0.020096 -2022-12-06 11:17:10,323 - Epoch: [122][ 720/ 1200] Overall Loss 0.193006 Objective Loss 0.193006 LR 0.000500 Time 0.020081 -2022-12-06 11:17:10,514 - Epoch: [122][ 730/ 1200] Overall Loss 0.192840 Objective Loss 0.192840 LR 0.000500 Time 0.020066 -2022-12-06 11:17:10,705 - Epoch: [122][ 740/ 1200] Overall Loss 0.192875 Objective Loss 0.192875 LR 0.000500 Time 0.020053 -2022-12-06 11:17:10,897 - Epoch: [122][ 750/ 1200] Overall Loss 0.192812 Objective Loss 0.192812 LR 0.000500 Time 0.020040 -2022-12-06 11:17:11,087 - Epoch: [122][ 760/ 1200] Overall Loss 0.192704 Objective Loss 0.192704 LR 0.000500 Time 0.020026 -2022-12-06 11:17:11,278 - Epoch: [122][ 770/ 1200] Overall Loss 0.192634 Objective Loss 0.192634 LR 0.000500 Time 0.020013 -2022-12-06 11:17:11,468 - Epoch: [122][ 780/ 1200] Overall Loss 0.193066 Objective Loss 0.193066 LR 0.000500 Time 0.020000 -2022-12-06 11:17:11,660 - Epoch: [122][ 790/ 1200] Overall Loss 0.192944 Objective Loss 0.192944 LR 0.000500 Time 0.019989 -2022-12-06 11:17:11,851 - Epoch: [122][ 800/ 1200] Overall Loss 0.193079 Objective Loss 0.193079 LR 0.000500 Time 0.019977 -2022-12-06 11:17:12,041 - Epoch: [122][ 810/ 1200] Overall Loss 0.192853 Objective Loss 0.192853 LR 0.000500 Time 0.019965 -2022-12-06 11:17:12,232 - Epoch: [122][ 820/ 1200] Overall Loss 0.192923 Objective Loss 0.192923 LR 0.000500 Time 0.019954 -2022-12-06 11:17:12,425 - Epoch: [122][ 830/ 1200] Overall Loss 0.192790 Objective Loss 0.192790 LR 0.000500 Time 0.019945 -2022-12-06 11:17:12,617 - Epoch: [122][ 840/ 1200] Overall Loss 0.193003 Objective Loss 0.193003 LR 0.000500 Time 0.019936 -2022-12-06 11:17:12,811 - Epoch: [122][ 850/ 1200] Overall Loss 0.193106 Objective Loss 0.193106 LR 0.000500 Time 0.019929 -2022-12-06 11:17:13,004 - Epoch: [122][ 860/ 1200] Overall Loss 0.192706 Objective Loss 0.192706 LR 0.000500 Time 0.019921 -2022-12-06 11:17:13,197 - Epoch: [122][ 870/ 1200] Overall Loss 0.192926 Objective Loss 0.192926 LR 0.000500 Time 0.019913 -2022-12-06 11:17:13,390 - Epoch: [122][ 880/ 1200] Overall Loss 0.192997 Objective Loss 0.192997 LR 0.000500 Time 0.019905 -2022-12-06 11:17:13,584 - Epoch: [122][ 890/ 1200] Overall Loss 0.193016 Objective Loss 0.193016 LR 0.000500 Time 0.019899 -2022-12-06 11:17:13,777 - Epoch: [122][ 900/ 1200] Overall Loss 0.193193 Objective Loss 0.193193 LR 0.000500 Time 0.019892 -2022-12-06 11:17:13,971 - Epoch: [122][ 910/ 1200] Overall Loss 0.193212 Objective Loss 0.193212 LR 0.000500 Time 0.019885 -2022-12-06 11:17:14,164 - Epoch: [122][ 920/ 1200] Overall Loss 0.193066 Objective Loss 0.193066 LR 0.000500 Time 0.019878 -2022-12-06 11:17:14,358 - Epoch: [122][ 930/ 1200] Overall Loss 0.193281 Objective Loss 0.193281 LR 0.000500 Time 0.019873 -2022-12-06 11:17:14,551 - Epoch: [122][ 940/ 1200] Overall Loss 0.192976 Objective Loss 0.192976 LR 0.000500 Time 0.019866 -2022-12-06 11:17:14,745 - Epoch: [122][ 950/ 1200] Overall Loss 0.193113 Objective Loss 0.193113 LR 0.000500 Time 0.019861 -2022-12-06 11:17:14,938 - Epoch: [122][ 960/ 1200] Overall Loss 0.193264 Objective Loss 0.193264 LR 0.000500 Time 0.019854 -2022-12-06 11:17:15,131 - Epoch: [122][ 970/ 1200] Overall Loss 0.193358 Objective Loss 0.193358 LR 0.000500 Time 0.019848 -2022-12-06 11:17:15,324 - Epoch: [122][ 980/ 1200] Overall Loss 0.193456 Objective Loss 0.193456 LR 0.000500 Time 0.019842 -2022-12-06 11:17:15,518 - Epoch: [122][ 990/ 1200] Overall Loss 0.193377 Objective Loss 0.193377 LR 0.000500 Time 0.019837 -2022-12-06 11:17:15,712 - Epoch: [122][ 1000/ 1200] Overall Loss 0.193334 Objective Loss 0.193334 LR 0.000500 Time 0.019832 -2022-12-06 11:17:15,905 - Epoch: [122][ 1010/ 1200] Overall Loss 0.193461 Objective Loss 0.193461 LR 0.000500 Time 0.019827 -2022-12-06 11:17:16,098 - Epoch: [122][ 1020/ 1200] Overall Loss 0.193409 Objective Loss 0.193409 LR 0.000500 Time 0.019821 -2022-12-06 11:17:16,292 - Epoch: [122][ 1030/ 1200] Overall Loss 0.193392 Objective Loss 0.193392 LR 0.000500 Time 0.019816 -2022-12-06 11:17:16,485 - Epoch: [122][ 1040/ 1200] Overall Loss 0.193630 Objective Loss 0.193630 LR 0.000500 Time 0.019810 -2022-12-06 11:17:16,678 - Epoch: [122][ 1050/ 1200] Overall Loss 0.193776 Objective Loss 0.193776 LR 0.000500 Time 0.019805 -2022-12-06 11:17:16,872 - Epoch: [122][ 1060/ 1200] Overall Loss 0.193707 Objective Loss 0.193707 LR 0.000500 Time 0.019801 -2022-12-06 11:17:17,065 - Epoch: [122][ 1070/ 1200] Overall Loss 0.193761 Objective Loss 0.193761 LR 0.000500 Time 0.019796 -2022-12-06 11:17:17,258 - Epoch: [122][ 1080/ 1200] Overall Loss 0.193740 Objective Loss 0.193740 LR 0.000500 Time 0.019791 -2022-12-06 11:17:17,451 - Epoch: [122][ 1090/ 1200] Overall Loss 0.193659 Objective Loss 0.193659 LR 0.000500 Time 0.019786 -2022-12-06 11:17:17,644 - Epoch: [122][ 1100/ 1200] Overall Loss 0.193835 Objective Loss 0.193835 LR 0.000500 Time 0.019780 -2022-12-06 11:17:17,838 - Epoch: [122][ 1110/ 1200] Overall Loss 0.193795 Objective Loss 0.193795 LR 0.000500 Time 0.019776 -2022-12-06 11:17:18,031 - Epoch: [122][ 1120/ 1200] Overall Loss 0.193708 Objective Loss 0.193708 LR 0.000500 Time 0.019771 -2022-12-06 11:17:18,224 - Epoch: [122][ 1130/ 1200] Overall Loss 0.193840 Objective Loss 0.193840 LR 0.000500 Time 0.019767 -2022-12-06 11:17:18,418 - Epoch: [122][ 1140/ 1200] Overall Loss 0.193944 Objective Loss 0.193944 LR 0.000500 Time 0.019763 -2022-12-06 11:17:18,611 - Epoch: [122][ 1150/ 1200] Overall Loss 0.194182 Objective Loss 0.194182 LR 0.000500 Time 0.019759 -2022-12-06 11:17:18,804 - Epoch: [122][ 1160/ 1200] Overall Loss 0.194187 Objective Loss 0.194187 LR 0.000500 Time 0.019754 -2022-12-06 11:17:18,998 - Epoch: [122][ 1170/ 1200] Overall Loss 0.194163 Objective Loss 0.194163 LR 0.000500 Time 0.019751 -2022-12-06 11:17:19,191 - Epoch: [122][ 1180/ 1200] Overall Loss 0.194221 Objective Loss 0.194221 LR 0.000500 Time 0.019746 -2022-12-06 11:17:19,384 - Epoch: [122][ 1190/ 1200] Overall Loss 0.194244 Objective Loss 0.194244 LR 0.000500 Time 0.019743 -2022-12-06 11:17:19,615 - Epoch: [122][ 1200/ 1200] Overall Loss 0.194194 Objective Loss 0.194194 Top1 89.121339 Top5 99.163180 LR 0.000500 Time 0.019770 -2022-12-06 11:17:19,704 - --- validate (epoch=122)----------- -2022-12-06 11:17:19,704 - 34129 samples (256 per mini-batch) -2022-12-06 11:17:20,152 - Epoch: [122][ 10/ 134] Loss 0.247939 Top1 86.367188 Top5 98.398438 -2022-12-06 11:17:20,283 - Epoch: [122][ 20/ 134] Loss 0.241238 Top1 86.757812 Top5 98.593750 -2022-12-06 11:17:20,416 - Epoch: [122][ 30/ 134] Loss 0.252766 Top1 86.263021 Top5 98.372396 -2022-12-06 11:17:20,545 - Epoch: [122][ 40/ 134] Loss 0.250564 Top1 86.416016 Top5 98.291016 -2022-12-06 11:17:20,675 - Epoch: [122][ 50/ 134] Loss 0.254663 Top1 86.335938 Top5 98.250000 -2022-12-06 11:17:20,804 - Epoch: [122][ 60/ 134] Loss 0.249766 Top1 86.308594 Top5 98.333333 -2022-12-06 11:17:20,943 - Epoch: [122][ 70/ 134] Loss 0.253468 Top1 86.255580 Top5 98.359375 -2022-12-06 11:17:21,076 - Epoch: [122][ 80/ 134] Loss 0.253474 Top1 86.147461 Top5 98.330078 -2022-12-06 11:17:21,211 - Epoch: [122][ 90/ 134] Loss 0.255538 Top1 86.176215 Top5 98.324653 -2022-12-06 11:17:21,343 - Epoch: [122][ 100/ 134] Loss 0.255727 Top1 86.070312 Top5 98.328125 -2022-12-06 11:17:21,480 - Epoch: [122][ 110/ 134] Loss 0.254599 Top1 86.125710 Top5 98.288352 -2022-12-06 11:17:21,611 - Epoch: [122][ 120/ 134] Loss 0.255392 Top1 86.223958 Top5 98.277995 -2022-12-06 11:17:21,744 - Epoch: [122][ 130/ 134] Loss 0.256618 Top1 86.207933 Top5 98.251202 -2022-12-06 11:17:21,785 - Epoch: [122][ 134/ 134] Loss 0.257175 Top1 86.211140 Top5 98.265405 -2022-12-06 11:17:21,872 - ==> Top1: 86.211 Top5: 98.265 Loss: 0.257 - -2022-12-06 11:17:21,873 - ==> Confusion: -[[ 897 2 2 3 7 3 0 2 5 58 0 1 0 5 5 1 1 1 2 0 1] - [ 1 951 2 2 9 14 2 10 0 0 3 3 2 2 0 1 1 0 11 6 7] - [ 2 3 1010 13 2 0 19 12 0 5 6 3 4 2 4 1 2 0 3 4 8] - [ 1 1 15 937 0 2 0 1 0 1 15 0 7 2 18 0 0 1 13 0 6] - [ 5 6 2 1 960 1 0 2 1 5 1 2 1 1 12 5 9 2 1 1 2] - [ 2 25 0 4 12 932 3 23 1 3 1 12 5 24 3 1 1 0 2 7 8] - [ 1 1 15 1 0 1 1071 1 1 0 1 1 3 1 0 4 1 3 3 6 3] - [ 2 9 7 3 2 19 10 944 0 1 5 3 2 2 0 0 0 0 32 10 3] - [ 5 4 0 0 0 1 0 1 978 37 8 0 3 7 15 1 1 0 1 1 1] - [ 54 0 0 0 5 1 0 3 26 889 2 1 0 9 4 1 1 0 0 0 5] - [ 0 1 5 3 2 1 2 0 9 1 960 1 1 10 6 0 2 0 8 3 4] - [ 3 0 2 0 0 8 4 3 3 1 0 979 19 8 0 0 4 3 1 10 3] - [ 0 1 1 4 0 1 0 0 0 0 1 26 902 3 2 6 1 7 0 10 4] - [ 0 0 0 0 1 5 0 1 7 14 7 2 3 966 3 1 2 0 1 2 8] - [ 7 0 2 7 2 2 0 1 14 4 1 3 2 5 1066 0 0 1 6 2 5] - [ 0 0 1 2 2 0 3 0 0 0 0 9 9 3 0 988 9 9 0 4 4] - [ 2 2 1 1 4 0 1 0 0 0 0 6 3 2 1 7 1032 0 0 7 3] - [ 1 1 1 0 0 1 0 0 0 4 1 10 21 1 3 14 1 973 1 1 2] - [ 2 4 7 5 1 3 1 16 1 1 7 4 2 1 8 0 0 0 942 1 2] - [ 1 5 2 1 0 5 6 9 1 0 2 6 6 8 1 4 2 4 0 1014 3] - [ 98 195 200 120 133 134 83 153 86 95 183 102 360 287 173 117 190 63 187 240 10027]] - -2022-12-06 11:17:22,452 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:17:22,452 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:17:22,458 - - -2022-12-06 11:17:22,458 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:17:23,495 - Epoch: [123][ 10/ 1200] Overall Loss 0.179780 Objective Loss 0.179780 LR 0.000500 Time 0.103672 -2022-12-06 11:17:23,700 - Epoch: [123][ 20/ 1200] Overall Loss 0.171873 Objective Loss 0.171873 LR 0.000500 Time 0.062040 -2022-12-06 11:17:23,892 - Epoch: [123][ 30/ 1200] Overall Loss 0.180395 Objective Loss 0.180395 LR 0.000500 Time 0.047763 -2022-12-06 11:17:24,085 - Epoch: [123][ 40/ 1200] Overall Loss 0.178947 Objective Loss 0.178947 LR 0.000500 Time 0.040613 -2022-12-06 11:17:24,277 - Epoch: [123][ 50/ 1200] Overall Loss 0.185194 Objective Loss 0.185194 LR 0.000500 Time 0.036333 -2022-12-06 11:17:24,470 - Epoch: [123][ 60/ 1200] Overall Loss 0.183120 Objective Loss 0.183120 LR 0.000500 Time 0.033474 -2022-12-06 11:17:24,662 - Epoch: [123][ 70/ 1200] Overall Loss 0.180552 Objective Loss 0.180552 LR 0.000500 Time 0.031438 -2022-12-06 11:17:24,855 - Epoch: [123][ 80/ 1200] Overall Loss 0.181822 Objective Loss 0.181822 LR 0.000500 Time 0.029906 -2022-12-06 11:17:25,047 - Epoch: [123][ 90/ 1200] Overall Loss 0.182902 Objective Loss 0.182902 LR 0.000500 Time 0.028708 -2022-12-06 11:17:25,239 - Epoch: [123][ 100/ 1200] Overall Loss 0.184820 Objective Loss 0.184820 LR 0.000500 Time 0.027755 -2022-12-06 11:17:25,431 - Epoch: [123][ 110/ 1200] Overall Loss 0.184949 Objective Loss 0.184949 LR 0.000500 Time 0.026972 -2022-12-06 11:17:25,623 - Epoch: [123][ 120/ 1200] Overall Loss 0.186947 Objective Loss 0.186947 LR 0.000500 Time 0.026321 -2022-12-06 11:17:25,815 - Epoch: [123][ 130/ 1200] Overall Loss 0.186179 Objective Loss 0.186179 LR 0.000500 Time 0.025768 -2022-12-06 11:17:26,007 - Epoch: [123][ 140/ 1200] Overall Loss 0.185438 Objective Loss 0.185438 LR 0.000500 Time 0.025294 -2022-12-06 11:17:26,199 - Epoch: [123][ 150/ 1200] Overall Loss 0.185261 Objective Loss 0.185261 LR 0.000500 Time 0.024884 -2022-12-06 11:17:26,391 - Epoch: [123][ 160/ 1200] Overall Loss 0.187320 Objective Loss 0.187320 LR 0.000500 Time 0.024526 -2022-12-06 11:17:26,583 - Epoch: [123][ 170/ 1200] Overall Loss 0.186943 Objective Loss 0.186943 LR 0.000500 Time 0.024209 -2022-12-06 11:17:26,776 - Epoch: [123][ 180/ 1200] Overall Loss 0.187028 Objective Loss 0.187028 LR 0.000500 Time 0.023932 -2022-12-06 11:17:26,968 - Epoch: [123][ 190/ 1200] Overall Loss 0.187216 Objective Loss 0.187216 LR 0.000500 Time 0.023680 -2022-12-06 11:17:27,160 - Epoch: [123][ 200/ 1200] Overall Loss 0.187217 Objective Loss 0.187217 LR 0.000500 Time 0.023454 -2022-12-06 11:17:27,352 - Epoch: [123][ 210/ 1200] Overall Loss 0.186767 Objective Loss 0.186767 LR 0.000500 Time 0.023248 -2022-12-06 11:17:27,544 - Epoch: [123][ 220/ 1200] Overall Loss 0.187297 Objective Loss 0.187297 LR 0.000500 Time 0.023062 -2022-12-06 11:17:27,736 - Epoch: [123][ 230/ 1200] Overall Loss 0.187919 Objective Loss 0.187919 LR 0.000500 Time 0.022893 -2022-12-06 11:17:27,928 - Epoch: [123][ 240/ 1200] Overall Loss 0.188631 Objective Loss 0.188631 LR 0.000500 Time 0.022736 -2022-12-06 11:17:28,120 - Epoch: [123][ 250/ 1200] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000500 Time 0.022592 -2022-12-06 11:17:28,312 - Epoch: [123][ 260/ 1200] Overall Loss 0.189381 Objective Loss 0.189381 LR 0.000500 Time 0.022461 -2022-12-06 11:17:28,504 - Epoch: [123][ 270/ 1200] Overall Loss 0.189779 Objective Loss 0.189779 LR 0.000500 Time 0.022340 -2022-12-06 11:17:28,696 - Epoch: [123][ 280/ 1200] Overall Loss 0.188984 Objective Loss 0.188984 LR 0.000500 Time 0.022225 -2022-12-06 11:17:28,888 - Epoch: [123][ 290/ 1200] Overall Loss 0.189189 Objective Loss 0.189189 LR 0.000500 Time 0.022118 -2022-12-06 11:17:29,080 - Epoch: [123][ 300/ 1200] Overall Loss 0.189386 Objective Loss 0.189386 LR 0.000500 Time 0.022017 -2022-12-06 11:17:29,272 - Epoch: [123][ 310/ 1200] Overall Loss 0.189845 Objective Loss 0.189845 LR 0.000500 Time 0.021925 -2022-12-06 11:17:29,464 - Epoch: [123][ 320/ 1200] Overall Loss 0.190010 Objective Loss 0.190010 LR 0.000500 Time 0.021839 -2022-12-06 11:17:29,656 - Epoch: [123][ 330/ 1200] Overall Loss 0.190702 Objective Loss 0.190702 LR 0.000500 Time 0.021756 -2022-12-06 11:17:29,848 - Epoch: [123][ 340/ 1200] Overall Loss 0.191122 Objective Loss 0.191122 LR 0.000500 Time 0.021680 -2022-12-06 11:17:30,039 - Epoch: [123][ 350/ 1200] Overall Loss 0.191425 Objective Loss 0.191425 LR 0.000500 Time 0.021606 -2022-12-06 11:17:30,231 - Epoch: [123][ 360/ 1200] Overall Loss 0.191378 Objective Loss 0.191378 LR 0.000500 Time 0.021537 -2022-12-06 11:17:30,423 - Epoch: [123][ 370/ 1200] Overall Loss 0.191171 Objective Loss 0.191171 LR 0.000500 Time 0.021472 -2022-12-06 11:17:30,616 - Epoch: [123][ 380/ 1200] Overall Loss 0.190968 Objective Loss 0.190968 LR 0.000500 Time 0.021414 -2022-12-06 11:17:30,808 - Epoch: [123][ 390/ 1200] Overall Loss 0.191036 Objective Loss 0.191036 LR 0.000500 Time 0.021356 -2022-12-06 11:17:31,000 - Epoch: [123][ 400/ 1200] Overall Loss 0.191091 Objective Loss 0.191091 LR 0.000500 Time 0.021301 -2022-12-06 11:17:31,193 - Epoch: [123][ 410/ 1200] Overall Loss 0.190972 Objective Loss 0.190972 LR 0.000500 Time 0.021249 -2022-12-06 11:17:31,385 - Epoch: [123][ 420/ 1200] Overall Loss 0.190836 Objective Loss 0.190836 LR 0.000500 Time 0.021199 -2022-12-06 11:17:31,576 - Epoch: [123][ 430/ 1200] Overall Loss 0.191143 Objective Loss 0.191143 LR 0.000500 Time 0.021149 -2022-12-06 11:17:31,769 - Epoch: [123][ 440/ 1200] Overall Loss 0.191141 Objective Loss 0.191141 LR 0.000500 Time 0.021106 -2022-12-06 11:17:31,961 - Epoch: [123][ 450/ 1200] Overall Loss 0.191327 Objective Loss 0.191327 LR 0.000500 Time 0.021063 -2022-12-06 11:17:32,154 - Epoch: [123][ 460/ 1200] Overall Loss 0.191133 Objective Loss 0.191133 LR 0.000500 Time 0.021023 -2022-12-06 11:17:32,347 - Epoch: [123][ 470/ 1200] Overall Loss 0.191350 Objective Loss 0.191350 LR 0.000500 Time 0.020985 -2022-12-06 11:17:32,540 - Epoch: [123][ 480/ 1200] Overall Loss 0.191183 Objective Loss 0.191183 LR 0.000500 Time 0.020949 -2022-12-06 11:17:32,733 - Epoch: [123][ 490/ 1200] Overall Loss 0.191036 Objective Loss 0.191036 LR 0.000500 Time 0.020914 -2022-12-06 11:17:32,926 - Epoch: [123][ 500/ 1200] Overall Loss 0.190982 Objective Loss 0.190982 LR 0.000500 Time 0.020881 -2022-12-06 11:17:33,118 - Epoch: [123][ 510/ 1200] Overall Loss 0.190714 Objective Loss 0.190714 LR 0.000500 Time 0.020847 -2022-12-06 11:17:33,310 - Epoch: [123][ 520/ 1200] Overall Loss 0.191225 Objective Loss 0.191225 LR 0.000500 Time 0.020815 -2022-12-06 11:17:33,503 - Epoch: [123][ 530/ 1200] Overall Loss 0.191335 Objective Loss 0.191335 LR 0.000500 Time 0.020784 -2022-12-06 11:17:33,695 - Epoch: [123][ 540/ 1200] Overall Loss 0.191445 Objective Loss 0.191445 LR 0.000500 Time 0.020755 -2022-12-06 11:17:33,888 - Epoch: [123][ 550/ 1200] Overall Loss 0.191645 Objective Loss 0.191645 LR 0.000500 Time 0.020727 -2022-12-06 11:17:34,081 - Epoch: [123][ 560/ 1200] Overall Loss 0.191671 Objective Loss 0.191671 LR 0.000500 Time 0.020700 -2022-12-06 11:17:34,272 - Epoch: [123][ 570/ 1200] Overall Loss 0.191747 Objective Loss 0.191747 LR 0.000500 Time 0.020672 -2022-12-06 11:17:34,465 - Epoch: [123][ 580/ 1200] Overall Loss 0.192513 Objective Loss 0.192513 LR 0.000500 Time 0.020647 -2022-12-06 11:17:34,657 - Epoch: [123][ 590/ 1200] Overall Loss 0.192696 Objective Loss 0.192696 LR 0.000500 Time 0.020622 -2022-12-06 11:17:34,849 - Epoch: [123][ 600/ 1200] Overall Loss 0.192856 Objective Loss 0.192856 LR 0.000500 Time 0.020597 -2022-12-06 11:17:35,042 - Epoch: [123][ 610/ 1200] Overall Loss 0.192751 Objective Loss 0.192751 LR 0.000500 Time 0.020574 -2022-12-06 11:17:35,234 - Epoch: [123][ 620/ 1200] Overall Loss 0.192505 Objective Loss 0.192505 LR 0.000500 Time 0.020552 -2022-12-06 11:17:35,426 - Epoch: [123][ 630/ 1200] Overall Loss 0.192525 Objective Loss 0.192525 LR 0.000500 Time 0.020530 -2022-12-06 11:17:35,619 - Epoch: [123][ 640/ 1200] Overall Loss 0.192304 Objective Loss 0.192304 LR 0.000500 Time 0.020508 -2022-12-06 11:17:35,811 - Epoch: [123][ 650/ 1200] Overall Loss 0.192113 Objective Loss 0.192113 LR 0.000500 Time 0.020487 -2022-12-06 11:17:36,003 - Epoch: [123][ 660/ 1200] Overall Loss 0.192312 Objective Loss 0.192312 LR 0.000500 Time 0.020467 -2022-12-06 11:17:36,195 - Epoch: [123][ 670/ 1200] Overall Loss 0.192356 Objective Loss 0.192356 LR 0.000500 Time 0.020448 -2022-12-06 11:17:36,388 - Epoch: [123][ 680/ 1200] Overall Loss 0.192651 Objective Loss 0.192651 LR 0.000500 Time 0.020430 -2022-12-06 11:17:36,580 - Epoch: [123][ 690/ 1200] Overall Loss 0.192612 Objective Loss 0.192612 LR 0.000500 Time 0.020412 -2022-12-06 11:17:36,772 - Epoch: [123][ 700/ 1200] Overall Loss 0.192892 Objective Loss 0.192892 LR 0.000500 Time 0.020394 -2022-12-06 11:17:36,964 - Epoch: [123][ 710/ 1200] Overall Loss 0.192483 Objective Loss 0.192483 LR 0.000500 Time 0.020376 -2022-12-06 11:17:37,157 - Epoch: [123][ 720/ 1200] Overall Loss 0.192848 Objective Loss 0.192848 LR 0.000500 Time 0.020360 -2022-12-06 11:17:37,349 - Epoch: [123][ 730/ 1200] Overall Loss 0.192946 Objective Loss 0.192946 LR 0.000500 Time 0.020344 -2022-12-06 11:17:37,542 - Epoch: [123][ 740/ 1200] Overall Loss 0.192995 Objective Loss 0.192995 LR 0.000500 Time 0.020328 -2022-12-06 11:17:37,734 - Epoch: [123][ 750/ 1200] Overall Loss 0.193356 Objective Loss 0.193356 LR 0.000500 Time 0.020313 -2022-12-06 11:17:37,927 - Epoch: [123][ 760/ 1200] Overall Loss 0.193844 Objective Loss 0.193844 LR 0.000500 Time 0.020298 -2022-12-06 11:17:38,119 - Epoch: [123][ 770/ 1200] Overall Loss 0.193608 Objective Loss 0.193608 LR 0.000500 Time 0.020283 -2022-12-06 11:17:38,311 - Epoch: [123][ 780/ 1200] Overall Loss 0.193826 Objective Loss 0.193826 LR 0.000500 Time 0.020269 -2022-12-06 11:17:38,503 - Epoch: [123][ 790/ 1200] Overall Loss 0.194066 Objective Loss 0.194066 LR 0.000500 Time 0.020255 -2022-12-06 11:17:38,695 - Epoch: [123][ 800/ 1200] Overall Loss 0.194079 Objective Loss 0.194079 LR 0.000500 Time 0.020241 -2022-12-06 11:17:38,887 - Epoch: [123][ 810/ 1200] Overall Loss 0.193736 Objective Loss 0.193736 LR 0.000500 Time 0.020228 -2022-12-06 11:17:39,080 - Epoch: [123][ 820/ 1200] Overall Loss 0.193480 Objective Loss 0.193480 LR 0.000500 Time 0.020215 -2022-12-06 11:17:39,272 - Epoch: [123][ 830/ 1200] Overall Loss 0.193359 Objective Loss 0.193359 LR 0.000500 Time 0.020203 -2022-12-06 11:17:39,464 - Epoch: [123][ 840/ 1200] Overall Loss 0.193347 Objective Loss 0.193347 LR 0.000500 Time 0.020190 -2022-12-06 11:17:39,656 - Epoch: [123][ 850/ 1200] Overall Loss 0.193459 Objective Loss 0.193459 LR 0.000500 Time 0.020178 -2022-12-06 11:17:39,849 - Epoch: [123][ 860/ 1200] Overall Loss 0.193424 Objective Loss 0.193424 LR 0.000500 Time 0.020166 -2022-12-06 11:17:40,042 - Epoch: [123][ 870/ 1200] Overall Loss 0.193553 Objective Loss 0.193553 LR 0.000500 Time 0.020156 -2022-12-06 11:17:40,235 - Epoch: [123][ 880/ 1200] Overall Loss 0.193478 Objective Loss 0.193478 LR 0.000500 Time 0.020146 -2022-12-06 11:17:40,427 - Epoch: [123][ 890/ 1200] Overall Loss 0.193457 Objective Loss 0.193457 LR 0.000500 Time 0.020135 -2022-12-06 11:17:40,619 - Epoch: [123][ 900/ 1200] Overall Loss 0.193555 Objective Loss 0.193555 LR 0.000500 Time 0.020124 -2022-12-06 11:17:40,812 - Epoch: [123][ 910/ 1200] Overall Loss 0.193548 Objective Loss 0.193548 LR 0.000500 Time 0.020114 -2022-12-06 11:17:41,004 - Epoch: [123][ 920/ 1200] Overall Loss 0.193642 Objective Loss 0.193642 LR 0.000500 Time 0.020103 -2022-12-06 11:17:41,196 - Epoch: [123][ 930/ 1200] Overall Loss 0.193561 Objective Loss 0.193561 LR 0.000500 Time 0.020093 -2022-12-06 11:17:41,388 - Epoch: [123][ 940/ 1200] Overall Loss 0.193712 Objective Loss 0.193712 LR 0.000500 Time 0.020083 -2022-12-06 11:17:41,581 - Epoch: [123][ 950/ 1200] Overall Loss 0.193713 Objective Loss 0.193713 LR 0.000500 Time 0.020074 -2022-12-06 11:17:41,773 - Epoch: [123][ 960/ 1200] Overall Loss 0.193725 Objective Loss 0.193725 LR 0.000500 Time 0.020064 -2022-12-06 11:17:41,965 - Epoch: [123][ 970/ 1200] Overall Loss 0.193890 Objective Loss 0.193890 LR 0.000500 Time 0.020055 -2022-12-06 11:17:42,157 - Epoch: [123][ 980/ 1200] Overall Loss 0.193766 Objective Loss 0.193766 LR 0.000500 Time 0.020046 -2022-12-06 11:17:42,349 - Epoch: [123][ 990/ 1200] Overall Loss 0.193811 Objective Loss 0.193811 LR 0.000500 Time 0.020037 -2022-12-06 11:17:42,542 - Epoch: [123][ 1000/ 1200] Overall Loss 0.193729 Objective Loss 0.193729 LR 0.000500 Time 0.020029 -2022-12-06 11:17:42,734 - Epoch: [123][ 1010/ 1200] Overall Loss 0.193529 Objective Loss 0.193529 LR 0.000500 Time 0.020020 -2022-12-06 11:17:42,927 - Epoch: [123][ 1020/ 1200] Overall Loss 0.193590 Objective Loss 0.193590 LR 0.000500 Time 0.020012 -2022-12-06 11:17:43,119 - Epoch: [123][ 1030/ 1200] Overall Loss 0.193410 Objective Loss 0.193410 LR 0.000500 Time 0.020004 -2022-12-06 11:17:43,311 - Epoch: [123][ 1040/ 1200] Overall Loss 0.193527 Objective Loss 0.193527 LR 0.000500 Time 0.019996 -2022-12-06 11:17:43,504 - Epoch: [123][ 1050/ 1200] Overall Loss 0.193627 Objective Loss 0.193627 LR 0.000500 Time 0.019988 -2022-12-06 11:17:43,696 - Epoch: [123][ 1060/ 1200] Overall Loss 0.193700 Objective Loss 0.193700 LR 0.000500 Time 0.019980 -2022-12-06 11:17:43,888 - Epoch: [123][ 1070/ 1200] Overall Loss 0.193852 Objective Loss 0.193852 LR 0.000500 Time 0.019973 -2022-12-06 11:17:44,081 - Epoch: [123][ 1080/ 1200] Overall Loss 0.193994 Objective Loss 0.193994 LR 0.000500 Time 0.019966 -2022-12-06 11:17:44,273 - Epoch: [123][ 1090/ 1200] Overall Loss 0.194048 Objective Loss 0.194048 LR 0.000500 Time 0.019959 -2022-12-06 11:17:44,465 - Epoch: [123][ 1100/ 1200] Overall Loss 0.193978 Objective Loss 0.193978 LR 0.000500 Time 0.019951 -2022-12-06 11:17:44,657 - Epoch: [123][ 1110/ 1200] Overall Loss 0.193976 Objective Loss 0.193976 LR 0.000500 Time 0.019944 -2022-12-06 11:17:44,849 - Epoch: [123][ 1120/ 1200] Overall Loss 0.194124 Objective Loss 0.194124 LR 0.000500 Time 0.019937 -2022-12-06 11:17:45,042 - Epoch: [123][ 1130/ 1200] Overall Loss 0.194018 Objective Loss 0.194018 LR 0.000500 Time 0.019931 -2022-12-06 11:17:45,234 - Epoch: [123][ 1140/ 1200] Overall Loss 0.193983 Objective Loss 0.193983 LR 0.000500 Time 0.019924 -2022-12-06 11:17:45,427 - Epoch: [123][ 1150/ 1200] Overall Loss 0.193901 Objective Loss 0.193901 LR 0.000500 Time 0.019918 -2022-12-06 11:17:45,619 - Epoch: [123][ 1160/ 1200] Overall Loss 0.193761 Objective Loss 0.193761 LR 0.000500 Time 0.019911 -2022-12-06 11:17:45,811 - Epoch: [123][ 1170/ 1200] Overall Loss 0.193821 Objective Loss 0.193821 LR 0.000500 Time 0.019905 -2022-12-06 11:17:46,003 - Epoch: [123][ 1180/ 1200] Overall Loss 0.193733 Objective Loss 0.193733 LR 0.000500 Time 0.019898 -2022-12-06 11:17:46,196 - Epoch: [123][ 1190/ 1200] Overall Loss 0.193624 Objective Loss 0.193624 LR 0.000500 Time 0.019892 -2022-12-06 11:17:46,421 - Epoch: [123][ 1200/ 1200] Overall Loss 0.193613 Objective Loss 0.193613 Top1 89.121339 Top5 99.372385 LR 0.000500 Time 0.019914 -2022-12-06 11:17:46,509 - --- validate (epoch=123)----------- -2022-12-06 11:17:46,509 - 34129 samples (256 per mini-batch) -2022-12-06 11:17:46,964 - Epoch: [123][ 10/ 134] Loss 0.267920 Top1 86.445312 Top5 97.968750 -2022-12-06 11:17:47,096 - Epoch: [123][ 20/ 134] Loss 0.262746 Top1 86.171875 Top5 98.164062 -2022-12-06 11:17:47,224 - Epoch: [123][ 30/ 134] Loss 0.263548 Top1 86.289062 Top5 98.164062 -2022-12-06 11:17:47,353 - Epoch: [123][ 40/ 134] Loss 0.266187 Top1 86.357422 Top5 98.154297 -2022-12-06 11:17:47,481 - Epoch: [123][ 50/ 134] Loss 0.257065 Top1 86.531250 Top5 98.320312 -2022-12-06 11:17:47,609 - Epoch: [123][ 60/ 134] Loss 0.255213 Top1 86.562500 Top5 98.326823 -2022-12-06 11:17:47,735 - Epoch: [123][ 70/ 134] Loss 0.256084 Top1 86.556920 Top5 98.314732 -2022-12-06 11:17:47,876 - Epoch: [123][ 80/ 134] Loss 0.257026 Top1 86.582031 Top5 98.305664 -2022-12-06 11:17:48,004 - Epoch: [123][ 90/ 134] Loss 0.255399 Top1 86.575521 Top5 98.376736 -2022-12-06 11:17:48,133 - Epoch: [123][ 100/ 134] Loss 0.252657 Top1 86.605469 Top5 98.382812 -2022-12-06 11:17:48,261 - Epoch: [123][ 110/ 134] Loss 0.253037 Top1 86.608665 Top5 98.380682 -2022-12-06 11:17:48,388 - Epoch: [123][ 120/ 134] Loss 0.253687 Top1 86.695964 Top5 98.395182 -2022-12-06 11:17:48,515 - Epoch: [123][ 130/ 134] Loss 0.254074 Top1 86.673678 Top5 98.359375 -2022-12-06 11:17:48,552 - Epoch: [123][ 134/ 134] Loss 0.255357 Top1 86.644789 Top5 98.347446 -2022-12-06 11:17:48,653 - ==> Top1: 86.645 Top5: 98.347 Loss: 0.255 - -2022-12-06 11:17:48,654 - ==> Confusion: -[[ 907 0 3 0 6 4 1 1 6 53 0 1 1 4 2 2 0 1 1 0 3] - [ 2 937 3 1 10 24 6 11 1 1 2 4 0 2 0 3 3 1 7 0 9] - [ 5 2 1006 7 5 2 19 8 0 4 9 5 2 5 1 3 1 2 5 2 10] - [ 2 2 17 942 2 1 0 2 0 0 13 0 5 2 14 1 1 1 10 0 5] - [ 4 6 1 1 957 5 1 1 2 7 1 1 1 3 8 6 9 2 1 1 2] - [ 4 15 2 0 8 975 2 14 2 1 1 12 4 17 1 1 1 1 1 2 5] - [ 1 3 9 2 0 2 1069 4 0 0 2 3 1 1 0 7 1 3 0 7 3] - [ 3 8 4 2 3 32 8 943 0 1 3 6 1 2 0 2 0 1 19 11 5] - [ 3 3 0 1 0 3 0 1 968 48 9 2 0 12 7 1 2 0 3 0 1] - [ 46 0 1 0 5 1 0 2 21 900 1 1 0 17 1 2 0 0 0 0 3] - [ 1 0 1 3 1 3 2 2 7 1 964 2 0 21 1 1 1 0 4 0 4] - [ 2 0 2 0 0 9 3 3 2 2 0 981 14 4 1 6 6 4 0 10 2] - [ 1 1 3 1 0 2 0 0 0 0 0 38 889 1 2 9 2 9 0 4 7] - [ 1 1 1 0 1 8 0 3 4 7 1 3 2 980 0 1 1 1 0 2 6] - [ 7 2 2 4 3 3 0 0 20 7 3 2 1 9 1050 0 2 2 7 1 5] - [ 1 0 2 1 2 0 2 0 0 0 0 11 5 1 0 991 7 15 0 1 4] - [ 5 4 0 0 3 1 0 0 0 0 0 2 2 1 0 12 1039 0 0 2 1] - [ 2 1 1 4 0 1 1 2 1 4 0 8 13 1 1 13 1 979 0 0 3] - [ 3 7 4 7 1 3 0 19 2 1 8 2 2 0 8 1 1 0 934 1 4] - [ 3 4 1 2 0 6 5 5 0 0 1 15 9 5 0 3 4 5 1 1006 5] - [ 121 178 174 97 126 208 92 122 73 90 208 103 303 356 112 104 207 79 131 193 10149]] - -2022-12-06 11:17:49,221 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:17:49,221 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:17:49,227 - - -2022-12-06 11:17:49,227 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:17:50,158 - Epoch: [124][ 10/ 1200] Overall Loss 0.217769 Objective Loss 0.217769 LR 0.000500 Time 0.092943 -2022-12-06 11:17:50,365 - Epoch: [124][ 20/ 1200] Overall Loss 0.197346 Objective Loss 0.197346 LR 0.000500 Time 0.056810 -2022-12-06 11:17:50,563 - Epoch: [124][ 30/ 1200] Overall Loss 0.190940 Objective Loss 0.190940 LR 0.000500 Time 0.044449 -2022-12-06 11:17:50,763 - Epoch: [124][ 40/ 1200] Overall Loss 0.193140 Objective Loss 0.193140 LR 0.000500 Time 0.038325 -2022-12-06 11:17:50,961 - Epoch: [124][ 50/ 1200] Overall Loss 0.192499 Objective Loss 0.192499 LR 0.000500 Time 0.034604 -2022-12-06 11:17:51,161 - Epoch: [124][ 60/ 1200] Overall Loss 0.194037 Objective Loss 0.194037 LR 0.000500 Time 0.032164 -2022-12-06 11:17:51,358 - Epoch: [124][ 70/ 1200] Overall Loss 0.191711 Objective Loss 0.191711 LR 0.000500 Time 0.030385 -2022-12-06 11:17:51,559 - Epoch: [124][ 80/ 1200] Overall Loss 0.194306 Objective Loss 0.194306 LR 0.000500 Time 0.029086 -2022-12-06 11:17:51,756 - Epoch: [124][ 90/ 1200] Overall Loss 0.195395 Objective Loss 0.195395 LR 0.000500 Time 0.028040 -2022-12-06 11:17:51,957 - Epoch: [124][ 100/ 1200] Overall Loss 0.195723 Objective Loss 0.195723 LR 0.000500 Time 0.027242 -2022-12-06 11:17:52,154 - Epoch: [124][ 110/ 1200] Overall Loss 0.195936 Objective Loss 0.195936 LR 0.000500 Time 0.026553 -2022-12-06 11:17:52,355 - Epoch: [124][ 120/ 1200] Overall Loss 0.197157 Objective Loss 0.197157 LR 0.000500 Time 0.026011 -2022-12-06 11:17:52,554 - Epoch: [124][ 130/ 1200] Overall Loss 0.195038 Objective Loss 0.195038 LR 0.000500 Time 0.025531 -2022-12-06 11:17:52,754 - Epoch: [124][ 140/ 1200] Overall Loss 0.195274 Objective Loss 0.195274 LR 0.000500 Time 0.025135 -2022-12-06 11:17:52,952 - Epoch: [124][ 150/ 1200] Overall Loss 0.193893 Objective Loss 0.193893 LR 0.000500 Time 0.024772 -2022-12-06 11:17:53,153 - Epoch: [124][ 160/ 1200] Overall Loss 0.193184 Objective Loss 0.193184 LR 0.000500 Time 0.024477 -2022-12-06 11:17:53,351 - Epoch: [124][ 170/ 1200] Overall Loss 0.194933 Objective Loss 0.194933 LR 0.000500 Time 0.024203 -2022-12-06 11:17:53,552 - Epoch: [124][ 180/ 1200] Overall Loss 0.194693 Objective Loss 0.194693 LR 0.000500 Time 0.023971 -2022-12-06 11:17:53,750 - Epoch: [124][ 190/ 1200] Overall Loss 0.194339 Objective Loss 0.194339 LR 0.000500 Time 0.023747 -2022-12-06 11:17:53,951 - Epoch: [124][ 200/ 1200] Overall Loss 0.192821 Objective Loss 0.192821 LR 0.000500 Time 0.023561 -2022-12-06 11:17:54,149 - Epoch: [124][ 210/ 1200] Overall Loss 0.193864 Objective Loss 0.193864 LR 0.000500 Time 0.023381 -2022-12-06 11:17:54,350 - Epoch: [124][ 220/ 1200] Overall Loss 0.192648 Objective Loss 0.192648 LR 0.000500 Time 0.023230 -2022-12-06 11:17:54,548 - Epoch: [124][ 230/ 1200] Overall Loss 0.191897 Objective Loss 0.191897 LR 0.000500 Time 0.023078 -2022-12-06 11:17:54,749 - Epoch: [124][ 240/ 1200] Overall Loss 0.192491 Objective Loss 0.192491 LR 0.000500 Time 0.022952 -2022-12-06 11:17:54,947 - Epoch: [124][ 250/ 1200] Overall Loss 0.193041 Objective Loss 0.193041 LR 0.000500 Time 0.022823 -2022-12-06 11:17:55,147 - Epoch: [124][ 260/ 1200] Overall Loss 0.194096 Objective Loss 0.194096 LR 0.000500 Time 0.022714 -2022-12-06 11:17:55,345 - Epoch: [124][ 270/ 1200] Overall Loss 0.193596 Objective Loss 0.193596 LR 0.000500 Time 0.022603 -2022-12-06 11:17:55,546 - Epoch: [124][ 280/ 1200] Overall Loss 0.193140 Objective Loss 0.193140 LR 0.000500 Time 0.022512 -2022-12-06 11:17:55,743 - Epoch: [124][ 290/ 1200] Overall Loss 0.192923 Objective Loss 0.192923 LR 0.000500 Time 0.022413 -2022-12-06 11:17:55,944 - Epoch: [124][ 300/ 1200] Overall Loss 0.193073 Objective Loss 0.193073 LR 0.000500 Time 0.022332 -2022-12-06 11:17:56,141 - Epoch: [124][ 310/ 1200] Overall Loss 0.192871 Objective Loss 0.192871 LR 0.000500 Time 0.022248 -2022-12-06 11:17:56,343 - Epoch: [124][ 320/ 1200] Overall Loss 0.192878 Objective Loss 0.192878 LR 0.000500 Time 0.022180 -2022-12-06 11:17:56,540 - Epoch: [124][ 330/ 1200] Overall Loss 0.192553 Objective Loss 0.192553 LR 0.000500 Time 0.022106 -2022-12-06 11:17:56,741 - Epoch: [124][ 340/ 1200] Overall Loss 0.192246 Objective Loss 0.192246 LR 0.000500 Time 0.022044 -2022-12-06 11:17:56,939 - Epoch: [124][ 350/ 1200] Overall Loss 0.192687 Objective Loss 0.192687 LR 0.000500 Time 0.021979 -2022-12-06 11:17:57,141 - Epoch: [124][ 360/ 1200] Overall Loss 0.192362 Objective Loss 0.192362 LR 0.000500 Time 0.021926 -2022-12-06 11:17:57,339 - Epoch: [124][ 370/ 1200] Overall Loss 0.192304 Objective Loss 0.192304 LR 0.000500 Time 0.021868 -2022-12-06 11:17:57,540 - Epoch: [124][ 380/ 1200] Overall Loss 0.192138 Objective Loss 0.192138 LR 0.000500 Time 0.021820 -2022-12-06 11:17:57,738 - Epoch: [124][ 390/ 1200] Overall Loss 0.192187 Objective Loss 0.192187 LR 0.000500 Time 0.021767 -2022-12-06 11:17:57,938 - Epoch: [124][ 400/ 1200] Overall Loss 0.192899 Objective Loss 0.192899 LR 0.000500 Time 0.021723 -2022-12-06 11:17:58,136 - Epoch: [124][ 410/ 1200] Overall Loss 0.193213 Objective Loss 0.193213 LR 0.000500 Time 0.021674 -2022-12-06 11:17:58,337 - Epoch: [124][ 420/ 1200] Overall Loss 0.193255 Objective Loss 0.193255 LR 0.000500 Time 0.021634 -2022-12-06 11:17:58,535 - Epoch: [124][ 430/ 1200] Overall Loss 0.193252 Objective Loss 0.193252 LR 0.000500 Time 0.021591 -2022-12-06 11:17:58,736 - Epoch: [124][ 440/ 1200] Overall Loss 0.193156 Objective Loss 0.193156 LR 0.000500 Time 0.021555 -2022-12-06 11:17:58,934 - Epoch: [124][ 450/ 1200] Overall Loss 0.192985 Objective Loss 0.192985 LR 0.000500 Time 0.021515 -2022-12-06 11:17:59,135 - Epoch: [124][ 460/ 1200] Overall Loss 0.193030 Objective Loss 0.193030 LR 0.000500 Time 0.021483 -2022-12-06 11:17:59,333 - Epoch: [124][ 470/ 1200] Overall Loss 0.192957 Objective Loss 0.192957 LR 0.000500 Time 0.021445 -2022-12-06 11:17:59,534 - Epoch: [124][ 480/ 1200] Overall Loss 0.193068 Objective Loss 0.193068 LR 0.000500 Time 0.021417 -2022-12-06 11:17:59,733 - Epoch: [124][ 490/ 1200] Overall Loss 0.193079 Objective Loss 0.193079 LR 0.000500 Time 0.021384 -2022-12-06 11:17:59,934 - Epoch: [124][ 500/ 1200] Overall Loss 0.192911 Objective Loss 0.192911 LR 0.000500 Time 0.021358 -2022-12-06 11:18:00,132 - Epoch: [124][ 510/ 1200] Overall Loss 0.192834 Objective Loss 0.192834 LR 0.000500 Time 0.021326 -2022-12-06 11:18:00,333 - Epoch: [124][ 520/ 1200] Overall Loss 0.192721 Objective Loss 0.192721 LR 0.000500 Time 0.021301 -2022-12-06 11:18:00,531 - Epoch: [124][ 530/ 1200] Overall Loss 0.192604 Objective Loss 0.192604 LR 0.000500 Time 0.021272 -2022-12-06 11:18:00,730 - Epoch: [124][ 540/ 1200] Overall Loss 0.192926 Objective Loss 0.192926 LR 0.000500 Time 0.021246 -2022-12-06 11:18:00,926 - Epoch: [124][ 550/ 1200] Overall Loss 0.193022 Objective Loss 0.193022 LR 0.000500 Time 0.021215 -2022-12-06 11:18:01,126 - Epoch: [124][ 560/ 1200] Overall Loss 0.192626 Objective Loss 0.192626 LR 0.000500 Time 0.021192 -2022-12-06 11:18:01,322 - Epoch: [124][ 570/ 1200] Overall Loss 0.192563 Objective Loss 0.192563 LR 0.000500 Time 0.021163 -2022-12-06 11:18:01,521 - Epoch: [124][ 580/ 1200] Overall Loss 0.192862 Objective Loss 0.192862 LR 0.000500 Time 0.021140 -2022-12-06 11:18:01,716 - Epoch: [124][ 590/ 1200] Overall Loss 0.192699 Objective Loss 0.192699 LR 0.000500 Time 0.021113 -2022-12-06 11:18:01,915 - Epoch: [124][ 600/ 1200] Overall Loss 0.192727 Objective Loss 0.192727 LR 0.000500 Time 0.021092 -2022-12-06 11:18:02,111 - Epoch: [124][ 610/ 1200] Overall Loss 0.192898 Objective Loss 0.192898 LR 0.000500 Time 0.021066 -2022-12-06 11:18:02,309 - Epoch: [124][ 620/ 1200] Overall Loss 0.193206 Objective Loss 0.193206 LR 0.000500 Time 0.021045 -2022-12-06 11:18:02,505 - Epoch: [124][ 630/ 1200] Overall Loss 0.193195 Objective Loss 0.193195 LR 0.000500 Time 0.021020 -2022-12-06 11:18:02,704 - Epoch: [124][ 640/ 1200] Overall Loss 0.192939 Objective Loss 0.192939 LR 0.000500 Time 0.021003 -2022-12-06 11:18:02,900 - Epoch: [124][ 650/ 1200] Overall Loss 0.193222 Objective Loss 0.193222 LR 0.000500 Time 0.020979 -2022-12-06 11:18:03,098 - Epoch: [124][ 660/ 1200] Overall Loss 0.193209 Objective Loss 0.193209 LR 0.000500 Time 0.020962 -2022-12-06 11:18:03,294 - Epoch: [124][ 670/ 1200] Overall Loss 0.193282 Objective Loss 0.193282 LR 0.000500 Time 0.020941 -2022-12-06 11:18:03,494 - Epoch: [124][ 680/ 1200] Overall Loss 0.193787 Objective Loss 0.193787 LR 0.000500 Time 0.020925 -2022-12-06 11:18:03,690 - Epoch: [124][ 690/ 1200] Overall Loss 0.193986 Objective Loss 0.193986 LR 0.000500 Time 0.020906 -2022-12-06 11:18:03,889 - Epoch: [124][ 700/ 1200] Overall Loss 0.193636 Objective Loss 0.193636 LR 0.000500 Time 0.020891 -2022-12-06 11:18:04,085 - Epoch: [124][ 710/ 1200] Overall Loss 0.193922 Objective Loss 0.193922 LR 0.000500 Time 0.020871 -2022-12-06 11:18:04,284 - Epoch: [124][ 720/ 1200] Overall Loss 0.193838 Objective Loss 0.193838 LR 0.000500 Time 0.020857 -2022-12-06 11:18:04,479 - Epoch: [124][ 730/ 1200] Overall Loss 0.193579 Objective Loss 0.193579 LR 0.000500 Time 0.020838 -2022-12-06 11:18:04,678 - Epoch: [124][ 740/ 1200] Overall Loss 0.193471 Objective Loss 0.193471 LR 0.000500 Time 0.020825 -2022-12-06 11:18:04,874 - Epoch: [124][ 750/ 1200] Overall Loss 0.193674 Objective Loss 0.193674 LR 0.000500 Time 0.020807 -2022-12-06 11:18:05,073 - Epoch: [124][ 760/ 1200] Overall Loss 0.193748 Objective Loss 0.193748 LR 0.000500 Time 0.020795 -2022-12-06 11:18:05,269 - Epoch: [124][ 770/ 1200] Overall Loss 0.193818 Objective Loss 0.193818 LR 0.000500 Time 0.020779 -2022-12-06 11:18:05,467 - Epoch: [124][ 780/ 1200] Overall Loss 0.193662 Objective Loss 0.193662 LR 0.000500 Time 0.020765 -2022-12-06 11:18:05,663 - Epoch: [124][ 790/ 1200] Overall Loss 0.193866 Objective Loss 0.193866 LR 0.000500 Time 0.020750 -2022-12-06 11:18:05,863 - Epoch: [124][ 800/ 1200] Overall Loss 0.194204 Objective Loss 0.194204 LR 0.000500 Time 0.020740 -2022-12-06 11:18:06,058 - Epoch: [124][ 810/ 1200] Overall Loss 0.194029 Objective Loss 0.194029 LR 0.000500 Time 0.020724 -2022-12-06 11:18:06,257 - Epoch: [124][ 820/ 1200] Overall Loss 0.194175 Objective Loss 0.194175 LR 0.000500 Time 0.020713 -2022-12-06 11:18:06,453 - Epoch: [124][ 830/ 1200] Overall Loss 0.194369 Objective Loss 0.194369 LR 0.000500 Time 0.020698 -2022-12-06 11:18:06,651 - Epoch: [124][ 840/ 1200] Overall Loss 0.194330 Objective Loss 0.194330 LR 0.000500 Time 0.020688 -2022-12-06 11:18:06,847 - Epoch: [124][ 850/ 1200] Overall Loss 0.194454 Objective Loss 0.194454 LR 0.000500 Time 0.020674 -2022-12-06 11:18:07,046 - Epoch: [124][ 860/ 1200] Overall Loss 0.194368 Objective Loss 0.194368 LR 0.000500 Time 0.020664 -2022-12-06 11:18:07,242 - Epoch: [124][ 870/ 1200] Overall Loss 0.194458 Objective Loss 0.194458 LR 0.000500 Time 0.020651 -2022-12-06 11:18:07,441 - Epoch: [124][ 880/ 1200] Overall Loss 0.194606 Objective Loss 0.194606 LR 0.000500 Time 0.020642 -2022-12-06 11:18:07,636 - Epoch: [124][ 890/ 1200] Overall Loss 0.194767 Objective Loss 0.194767 LR 0.000500 Time 0.020630 -2022-12-06 11:18:07,836 - Epoch: [124][ 900/ 1200] Overall Loss 0.194846 Objective Loss 0.194846 LR 0.000500 Time 0.020621 -2022-12-06 11:18:08,031 - Epoch: [124][ 910/ 1200] Overall Loss 0.194728 Objective Loss 0.194728 LR 0.000500 Time 0.020609 -2022-12-06 11:18:08,230 - Epoch: [124][ 920/ 1200] Overall Loss 0.195024 Objective Loss 0.195024 LR 0.000500 Time 0.020600 -2022-12-06 11:18:08,425 - Epoch: [124][ 930/ 1200] Overall Loss 0.195231 Objective Loss 0.195231 LR 0.000500 Time 0.020588 -2022-12-06 11:18:08,624 - Epoch: [124][ 940/ 1200] Overall Loss 0.195246 Objective Loss 0.195246 LR 0.000500 Time 0.020580 -2022-12-06 11:18:08,820 - Epoch: [124][ 950/ 1200] Overall Loss 0.195356 Objective Loss 0.195356 LR 0.000500 Time 0.020569 -2022-12-06 11:18:09,019 - Epoch: [124][ 960/ 1200] Overall Loss 0.195419 Objective Loss 0.195419 LR 0.000500 Time 0.020561 -2022-12-06 11:18:09,214 - Epoch: [124][ 970/ 1200] Overall Loss 0.195459 Objective Loss 0.195459 LR 0.000500 Time 0.020550 -2022-12-06 11:18:09,413 - Epoch: [124][ 980/ 1200] Overall Loss 0.195515 Objective Loss 0.195515 LR 0.000500 Time 0.020542 -2022-12-06 11:18:09,609 - Epoch: [124][ 990/ 1200] Overall Loss 0.195693 Objective Loss 0.195693 LR 0.000500 Time 0.020533 -2022-12-06 11:18:09,809 - Epoch: [124][ 1000/ 1200] Overall Loss 0.195540 Objective Loss 0.195540 LR 0.000500 Time 0.020526 -2022-12-06 11:18:10,004 - Epoch: [124][ 1010/ 1200] Overall Loss 0.195424 Objective Loss 0.195424 LR 0.000500 Time 0.020516 -2022-12-06 11:18:10,204 - Epoch: [124][ 1020/ 1200] Overall Loss 0.195455 Objective Loss 0.195455 LR 0.000500 Time 0.020510 -2022-12-06 11:18:10,399 - Epoch: [124][ 1030/ 1200] Overall Loss 0.195340 Objective Loss 0.195340 LR 0.000500 Time 0.020501 -2022-12-06 11:18:10,598 - Epoch: [124][ 1040/ 1200] Overall Loss 0.195359 Objective Loss 0.195359 LR 0.000500 Time 0.020494 -2022-12-06 11:18:10,794 - Epoch: [124][ 1050/ 1200] Overall Loss 0.195419 Objective Loss 0.195419 LR 0.000500 Time 0.020485 -2022-12-06 11:18:10,993 - Epoch: [124][ 1060/ 1200] Overall Loss 0.195479 Objective Loss 0.195479 LR 0.000500 Time 0.020479 -2022-12-06 11:18:11,188 - Epoch: [124][ 1070/ 1200] Overall Loss 0.195380 Objective Loss 0.195380 LR 0.000500 Time 0.020469 -2022-12-06 11:18:11,387 - Epoch: [124][ 1080/ 1200] Overall Loss 0.195323 Objective Loss 0.195323 LR 0.000500 Time 0.020464 -2022-12-06 11:18:11,583 - Epoch: [124][ 1090/ 1200] Overall Loss 0.195087 Objective Loss 0.195087 LR 0.000500 Time 0.020455 -2022-12-06 11:18:11,781 - Epoch: [124][ 1100/ 1200] Overall Loss 0.195169 Objective Loss 0.195169 LR 0.000500 Time 0.020448 -2022-12-06 11:18:11,978 - Epoch: [124][ 1110/ 1200] Overall Loss 0.195096 Objective Loss 0.195096 LR 0.000500 Time 0.020441 -2022-12-06 11:18:12,177 - Epoch: [124][ 1120/ 1200] Overall Loss 0.195026 Objective Loss 0.195026 LR 0.000500 Time 0.020436 -2022-12-06 11:18:12,373 - Epoch: [124][ 1130/ 1200] Overall Loss 0.195107 Objective Loss 0.195107 LR 0.000500 Time 0.020428 -2022-12-06 11:18:12,572 - Epoch: [124][ 1140/ 1200] Overall Loss 0.195173 Objective Loss 0.195173 LR 0.000500 Time 0.020423 -2022-12-06 11:18:12,768 - Epoch: [124][ 1150/ 1200] Overall Loss 0.194960 Objective Loss 0.194960 LR 0.000500 Time 0.020415 -2022-12-06 11:18:12,967 - Epoch: [124][ 1160/ 1200] Overall Loss 0.194734 Objective Loss 0.194734 LR 0.000500 Time 0.020410 -2022-12-06 11:18:13,163 - Epoch: [124][ 1170/ 1200] Overall Loss 0.194917 Objective Loss 0.194917 LR 0.000500 Time 0.020403 -2022-12-06 11:18:13,361 - Epoch: [124][ 1180/ 1200] Overall Loss 0.194911 Objective Loss 0.194911 LR 0.000500 Time 0.020397 -2022-12-06 11:18:13,557 - Epoch: [124][ 1190/ 1200] Overall Loss 0.194923 Objective Loss 0.194923 LR 0.000500 Time 0.020390 -2022-12-06 11:18:13,779 - Epoch: [124][ 1200/ 1200] Overall Loss 0.194924 Objective Loss 0.194924 Top1 88.912134 Top5 98.535565 LR 0.000500 Time 0.020405 -2022-12-06 11:18:13,868 - --- validate (epoch=124)----------- -2022-12-06 11:18:13,868 - 34129 samples (256 per mini-batch) -2022-12-06 11:18:14,315 - Epoch: [124][ 10/ 134] Loss 0.257450 Top1 86.054688 Top5 97.578125 -2022-12-06 11:18:14,448 - Epoch: [124][ 20/ 134] Loss 0.254667 Top1 86.464844 Top5 98.085938 -2022-12-06 11:18:14,578 - Epoch: [124][ 30/ 134] Loss 0.256777 Top1 86.653646 Top5 98.229167 -2022-12-06 11:18:14,710 - Epoch: [124][ 40/ 134] Loss 0.255120 Top1 86.845703 Top5 98.183594 -2022-12-06 11:18:14,839 - Epoch: [124][ 50/ 134] Loss 0.257355 Top1 86.648438 Top5 98.226562 -2022-12-06 11:18:14,969 - Epoch: [124][ 60/ 134] Loss 0.253875 Top1 86.757812 Top5 98.281250 -2022-12-06 11:18:15,099 - Epoch: [124][ 70/ 134] Loss 0.254078 Top1 86.785714 Top5 98.297991 -2022-12-06 11:18:15,229 - Epoch: [124][ 80/ 134] Loss 0.254371 Top1 86.860352 Top5 98.305664 -2022-12-06 11:18:15,361 - Epoch: [124][ 90/ 134] Loss 0.254145 Top1 86.875000 Top5 98.320312 -2022-12-06 11:18:15,490 - Epoch: [124][ 100/ 134] Loss 0.255785 Top1 86.894531 Top5 98.316406 -2022-12-06 11:18:15,620 - Epoch: [124][ 110/ 134] Loss 0.257171 Top1 86.850142 Top5 98.338068 -2022-12-06 11:18:15,750 - Epoch: [124][ 120/ 134] Loss 0.258667 Top1 86.692708 Top5 98.343099 -2022-12-06 11:18:15,879 - Epoch: [124][ 130/ 134] Loss 0.259466 Top1 86.628606 Top5 98.317308 -2022-12-06 11:18:15,916 - Epoch: [124][ 134/ 134] Loss 0.257643 Top1 86.644789 Top5 98.332796 -2022-12-06 11:18:16,003 - ==> Top1: 86.645 Top5: 98.333 Loss: 0.258 - -2022-12-06 11:18:16,004 - ==> Confusion: -[[ 906 1 3 3 5 5 0 0 4 48 1 3 2 3 2 1 1 1 3 0 4] - [ 1 927 1 2 10 17 5 19 1 1 5 4 0 2 0 1 1 0 15 3 12] - [ 4 4 997 8 3 2 34 9 0 5 6 4 2 1 4 1 0 1 4 4 10] - [ 3 2 13 944 1 2 2 2 1 0 8 0 5 2 10 0 0 4 15 0 6] - [ 8 5 1 0 949 6 1 1 1 7 1 2 1 2 10 7 8 1 1 3 5] - [ 3 11 0 2 5 976 3 19 3 2 1 12 5 12 0 1 3 0 1 5 5] - [ 1 2 5 0 0 0 1083 2 1 0 2 1 0 2 1 4 1 0 2 9 2] - [ 1 5 8 1 1 21 14 952 1 1 2 4 2 1 0 1 1 0 17 16 5] - [ 3 4 0 0 0 2 1 1 957 43 17 1 2 6 15 0 3 1 3 1 4] - [ 64 0 2 0 1 1 0 4 25 879 1 0 0 13 3 1 0 0 1 0 6] - [ 1 1 2 5 1 0 4 3 10 2 959 0 1 8 4 1 0 1 9 2 5] - [ 3 1 4 0 0 12 5 2 0 1 0 965 26 4 0 3 3 9 0 11 2] - [ 1 0 0 2 0 2 0 0 1 0 0 22 904 2 1 7 1 10 0 7 9] - [ 0 0 0 0 0 7 0 3 16 14 11 5 5 944 2 1 1 2 0 5 7] - [ 8 2 1 7 6 2 0 1 8 5 0 3 4 2 1060 1 1 1 12 0 6] - [ 1 0 1 0 1 2 5 0 0 0 1 6 7 2 0 988 5 12 1 4 7] - [ 1 4 0 1 0 1 3 0 1 0 1 4 0 2 1 15 1023 1 2 7 5] - [ 1 1 0 2 0 1 0 0 0 1 0 7 16 1 1 18 1 981 1 1 3] - [ 2 3 3 10 2 4 0 17 2 1 4 2 3 0 5 0 0 0 942 2 6] - [ 1 3 1 1 0 7 9 7 0 1 3 10 6 7 0 3 1 2 0 1014 4] - [ 129 180 149 99 89 198 117 166 80 75 182 87 349 254 131 94 134 83 164 251 10215]] - -2022-12-06 11:18:16,670 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:18:16,670 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:18:16,676 - - -2022-12-06 11:18:16,676 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:18:17,609 - Epoch: [125][ 10/ 1200] Overall Loss 0.193920 Objective Loss 0.193920 LR 0.000500 Time 0.093249 -2022-12-06 11:18:17,806 - Epoch: [125][ 20/ 1200] Overall Loss 0.186349 Objective Loss 0.186349 LR 0.000500 Time 0.056411 -2022-12-06 11:18:18,001 - Epoch: [125][ 30/ 1200] Overall Loss 0.189510 Objective Loss 0.189510 LR 0.000500 Time 0.044106 -2022-12-06 11:18:18,196 - Epoch: [125][ 40/ 1200] Overall Loss 0.193920 Objective Loss 0.193920 LR 0.000500 Time 0.037938 -2022-12-06 11:18:18,391 - Epoch: [125][ 50/ 1200] Overall Loss 0.196259 Objective Loss 0.196259 LR 0.000500 Time 0.034241 -2022-12-06 11:18:18,584 - Epoch: [125][ 60/ 1200] Overall Loss 0.197719 Objective Loss 0.197719 LR 0.000500 Time 0.031740 -2022-12-06 11:18:18,778 - Epoch: [125][ 70/ 1200] Overall Loss 0.197613 Objective Loss 0.197613 LR 0.000500 Time 0.029976 -2022-12-06 11:18:18,973 - Epoch: [125][ 80/ 1200] Overall Loss 0.199423 Objective Loss 0.199423 LR 0.000500 Time 0.028650 -2022-12-06 11:18:19,168 - Epoch: [125][ 90/ 1200] Overall Loss 0.200264 Objective Loss 0.200264 LR 0.000500 Time 0.027628 -2022-12-06 11:18:19,362 - Epoch: [125][ 100/ 1200] Overall Loss 0.197143 Objective Loss 0.197143 LR 0.000500 Time 0.026803 -2022-12-06 11:18:19,557 - Epoch: [125][ 110/ 1200] Overall Loss 0.197121 Objective Loss 0.197121 LR 0.000500 Time 0.026135 -2022-12-06 11:18:19,752 - Epoch: [125][ 120/ 1200] Overall Loss 0.196845 Objective Loss 0.196845 LR 0.000500 Time 0.025575 -2022-12-06 11:18:19,946 - Epoch: [125][ 130/ 1200] Overall Loss 0.197609 Objective Loss 0.197609 LR 0.000500 Time 0.025101 -2022-12-06 11:18:20,140 - Epoch: [125][ 140/ 1200] Overall Loss 0.197755 Objective Loss 0.197755 LR 0.000500 Time 0.024685 -2022-12-06 11:18:20,335 - Epoch: [125][ 150/ 1200] Overall Loss 0.198005 Objective Loss 0.198005 LR 0.000500 Time 0.024335 -2022-12-06 11:18:20,528 - Epoch: [125][ 160/ 1200] Overall Loss 0.196986 Objective Loss 0.196986 LR 0.000500 Time 0.024021 -2022-12-06 11:18:20,723 - Epoch: [125][ 170/ 1200] Overall Loss 0.196001 Objective Loss 0.196001 LR 0.000500 Time 0.023747 -2022-12-06 11:18:20,916 - Epoch: [125][ 180/ 1200] Overall Loss 0.195052 Objective Loss 0.195052 LR 0.000500 Time 0.023499 -2022-12-06 11:18:21,109 - Epoch: [125][ 190/ 1200] Overall Loss 0.195131 Objective Loss 0.195131 LR 0.000500 Time 0.023278 -2022-12-06 11:18:21,304 - Epoch: [125][ 200/ 1200] Overall Loss 0.194834 Objective Loss 0.194834 LR 0.000500 Time 0.023082 -2022-12-06 11:18:21,499 - Epoch: [125][ 210/ 1200] Overall Loss 0.194986 Objective Loss 0.194986 LR 0.000500 Time 0.022910 -2022-12-06 11:18:21,693 - Epoch: [125][ 220/ 1200] Overall Loss 0.194389 Objective Loss 0.194389 LR 0.000500 Time 0.022751 -2022-12-06 11:18:21,887 - Epoch: [125][ 230/ 1200] Overall Loss 0.193624 Objective Loss 0.193624 LR 0.000500 Time 0.022602 -2022-12-06 11:18:22,082 - Epoch: [125][ 240/ 1200] Overall Loss 0.193184 Objective Loss 0.193184 LR 0.000500 Time 0.022468 -2022-12-06 11:18:22,276 - Epoch: [125][ 250/ 1200] Overall Loss 0.192898 Objective Loss 0.192898 LR 0.000500 Time 0.022343 -2022-12-06 11:18:22,470 - Epoch: [125][ 260/ 1200] Overall Loss 0.191964 Objective Loss 0.191964 LR 0.000500 Time 0.022228 -2022-12-06 11:18:22,665 - Epoch: [125][ 270/ 1200] Overall Loss 0.192684 Objective Loss 0.192684 LR 0.000500 Time 0.022127 -2022-12-06 11:18:22,861 - Epoch: [125][ 280/ 1200] Overall Loss 0.193599 Objective Loss 0.193599 LR 0.000500 Time 0.022032 -2022-12-06 11:18:23,055 - Epoch: [125][ 290/ 1200] Overall Loss 0.193237 Objective Loss 0.193237 LR 0.000500 Time 0.021940 -2022-12-06 11:18:23,248 - Epoch: [125][ 300/ 1200] Overall Loss 0.193527 Objective Loss 0.193527 LR 0.000500 Time 0.021852 -2022-12-06 11:18:23,443 - Epoch: [125][ 310/ 1200] Overall Loss 0.193789 Objective Loss 0.193789 LR 0.000500 Time 0.021773 -2022-12-06 11:18:23,637 - Epoch: [125][ 320/ 1200] Overall Loss 0.194176 Objective Loss 0.194176 LR 0.000500 Time 0.021696 -2022-12-06 11:18:23,831 - Epoch: [125][ 330/ 1200] Overall Loss 0.193767 Objective Loss 0.193767 LR 0.000500 Time 0.021627 -2022-12-06 11:18:24,025 - Epoch: [125][ 340/ 1200] Overall Loss 0.193060 Objective Loss 0.193060 LR 0.000500 Time 0.021559 -2022-12-06 11:18:24,219 - Epoch: [125][ 350/ 1200] Overall Loss 0.193099 Objective Loss 0.193099 LR 0.000500 Time 0.021497 -2022-12-06 11:18:24,414 - Epoch: [125][ 360/ 1200] Overall Loss 0.192434 Objective Loss 0.192434 LR 0.000500 Time 0.021438 -2022-12-06 11:18:24,608 - Epoch: [125][ 370/ 1200] Overall Loss 0.192064 Objective Loss 0.192064 LR 0.000500 Time 0.021382 -2022-12-06 11:18:24,802 - Epoch: [125][ 380/ 1200] Overall Loss 0.192130 Objective Loss 0.192130 LR 0.000500 Time 0.021329 -2022-12-06 11:18:24,997 - Epoch: [125][ 390/ 1200] Overall Loss 0.192675 Objective Loss 0.192675 LR 0.000500 Time 0.021281 -2022-12-06 11:18:25,192 - Epoch: [125][ 400/ 1200] Overall Loss 0.193132 Objective Loss 0.193132 LR 0.000500 Time 0.021235 -2022-12-06 11:18:25,387 - Epoch: [125][ 410/ 1200] Overall Loss 0.192862 Objective Loss 0.192862 LR 0.000500 Time 0.021190 -2022-12-06 11:18:25,581 - Epoch: [125][ 420/ 1200] Overall Loss 0.192847 Objective Loss 0.192847 LR 0.000500 Time 0.021147 -2022-12-06 11:18:25,776 - Epoch: [125][ 430/ 1200] Overall Loss 0.192417 Objective Loss 0.192417 LR 0.000500 Time 0.021107 -2022-12-06 11:18:25,970 - Epoch: [125][ 440/ 1200] Overall Loss 0.192159 Objective Loss 0.192159 LR 0.000500 Time 0.021067 -2022-12-06 11:18:26,164 - Epoch: [125][ 450/ 1200] Overall Loss 0.192056 Objective Loss 0.192056 LR 0.000500 Time 0.021030 -2022-12-06 11:18:26,359 - Epoch: [125][ 460/ 1200] Overall Loss 0.192150 Objective Loss 0.192150 LR 0.000500 Time 0.020994 -2022-12-06 11:18:26,554 - Epoch: [125][ 470/ 1200] Overall Loss 0.192153 Objective Loss 0.192153 LR 0.000500 Time 0.020962 -2022-12-06 11:18:26,748 - Epoch: [125][ 480/ 1200] Overall Loss 0.192124 Objective Loss 0.192124 LR 0.000500 Time 0.020928 -2022-12-06 11:18:26,943 - Epoch: [125][ 490/ 1200] Overall Loss 0.192392 Objective Loss 0.192392 LR 0.000500 Time 0.020898 -2022-12-06 11:18:27,137 - Epoch: [125][ 500/ 1200] Overall Loss 0.191807 Objective Loss 0.191807 LR 0.000500 Time 0.020866 -2022-12-06 11:18:27,331 - Epoch: [125][ 510/ 1200] Overall Loss 0.191727 Objective Loss 0.191727 LR 0.000500 Time 0.020837 -2022-12-06 11:18:27,526 - Epoch: [125][ 520/ 1200] Overall Loss 0.191922 Objective Loss 0.191922 LR 0.000500 Time 0.020811 -2022-12-06 11:18:27,720 - Epoch: [125][ 530/ 1200] Overall Loss 0.191899 Objective Loss 0.191899 LR 0.000500 Time 0.020783 -2022-12-06 11:18:27,914 - Epoch: [125][ 540/ 1200] Overall Loss 0.191730 Objective Loss 0.191730 LR 0.000500 Time 0.020756 -2022-12-06 11:18:28,108 - Epoch: [125][ 550/ 1200] Overall Loss 0.191666 Objective Loss 0.191666 LR 0.000500 Time 0.020731 -2022-12-06 11:18:28,302 - Epoch: [125][ 560/ 1200] Overall Loss 0.191667 Objective Loss 0.191667 LR 0.000500 Time 0.020706 -2022-12-06 11:18:28,497 - Epoch: [125][ 570/ 1200] Overall Loss 0.191392 Objective Loss 0.191392 LR 0.000500 Time 0.020683 -2022-12-06 11:18:28,691 - Epoch: [125][ 580/ 1200] Overall Loss 0.191018 Objective Loss 0.191018 LR 0.000500 Time 0.020661 -2022-12-06 11:18:28,885 - Epoch: [125][ 590/ 1200] Overall Loss 0.190771 Objective Loss 0.190771 LR 0.000500 Time 0.020638 -2022-12-06 11:18:29,077 - Epoch: [125][ 600/ 1200] Overall Loss 0.190833 Objective Loss 0.190833 LR 0.000500 Time 0.020613 -2022-12-06 11:18:29,270 - Epoch: [125][ 610/ 1200] Overall Loss 0.190655 Objective Loss 0.190655 LR 0.000500 Time 0.020590 -2022-12-06 11:18:29,463 - Epoch: [125][ 620/ 1200] Overall Loss 0.190687 Objective Loss 0.190687 LR 0.000500 Time 0.020569 -2022-12-06 11:18:29,654 - Epoch: [125][ 630/ 1200] Overall Loss 0.190712 Objective Loss 0.190712 LR 0.000500 Time 0.020545 -2022-12-06 11:18:29,847 - Epoch: [125][ 640/ 1200] Overall Loss 0.190596 Objective Loss 0.190596 LR 0.000500 Time 0.020524 -2022-12-06 11:18:30,038 - Epoch: [125][ 650/ 1200] Overall Loss 0.190416 Objective Loss 0.190416 LR 0.000500 Time 0.020502 -2022-12-06 11:18:30,231 - Epoch: [125][ 660/ 1200] Overall Loss 0.190371 Objective Loss 0.190371 LR 0.000500 Time 0.020482 -2022-12-06 11:18:30,422 - Epoch: [125][ 670/ 1200] Overall Loss 0.190237 Objective Loss 0.190237 LR 0.000500 Time 0.020462 -2022-12-06 11:18:30,614 - Epoch: [125][ 680/ 1200] Overall Loss 0.190041 Objective Loss 0.190041 LR 0.000500 Time 0.020442 -2022-12-06 11:18:30,805 - Epoch: [125][ 690/ 1200] Overall Loss 0.189961 Objective Loss 0.189961 LR 0.000500 Time 0.020422 -2022-12-06 11:18:30,997 - Epoch: [125][ 700/ 1200] Overall Loss 0.189950 Objective Loss 0.189950 LR 0.000500 Time 0.020403 -2022-12-06 11:18:31,189 - Epoch: [125][ 710/ 1200] Overall Loss 0.190062 Objective Loss 0.190062 LR 0.000500 Time 0.020386 -2022-12-06 11:18:31,381 - Epoch: [125][ 720/ 1200] Overall Loss 0.190073 Objective Loss 0.190073 LR 0.000500 Time 0.020369 -2022-12-06 11:18:31,573 - Epoch: [125][ 730/ 1200] Overall Loss 0.190323 Objective Loss 0.190323 LR 0.000500 Time 0.020352 -2022-12-06 11:18:31,765 - Epoch: [125][ 740/ 1200] Overall Loss 0.190146 Objective Loss 0.190146 LR 0.000500 Time 0.020336 -2022-12-06 11:18:31,957 - Epoch: [125][ 750/ 1200] Overall Loss 0.190195 Objective Loss 0.190195 LR 0.000500 Time 0.020319 -2022-12-06 11:18:32,150 - Epoch: [125][ 760/ 1200] Overall Loss 0.190043 Objective Loss 0.190043 LR 0.000500 Time 0.020305 -2022-12-06 11:18:32,342 - Epoch: [125][ 770/ 1200] Overall Loss 0.190434 Objective Loss 0.190434 LR 0.000500 Time 0.020290 -2022-12-06 11:18:32,534 - Epoch: [125][ 780/ 1200] Overall Loss 0.190810 Objective Loss 0.190810 LR 0.000500 Time 0.020276 -2022-12-06 11:18:32,726 - Epoch: [125][ 790/ 1200] Overall Loss 0.190350 Objective Loss 0.190350 LR 0.000500 Time 0.020262 -2022-12-06 11:18:32,918 - Epoch: [125][ 800/ 1200] Overall Loss 0.190151 Objective Loss 0.190151 LR 0.000500 Time 0.020248 -2022-12-06 11:18:33,110 - Epoch: [125][ 810/ 1200] Overall Loss 0.190157 Objective Loss 0.190157 LR 0.000500 Time 0.020234 -2022-12-06 11:18:33,302 - Epoch: [125][ 820/ 1200] Overall Loss 0.190056 Objective Loss 0.190056 LR 0.000500 Time 0.020220 -2022-12-06 11:18:33,493 - Epoch: [125][ 830/ 1200] Overall Loss 0.189762 Objective Loss 0.189762 LR 0.000500 Time 0.020207 -2022-12-06 11:18:33,685 - Epoch: [125][ 840/ 1200] Overall Loss 0.190021 Objective Loss 0.190021 LR 0.000500 Time 0.020194 -2022-12-06 11:18:33,877 - Epoch: [125][ 850/ 1200] Overall Loss 0.190291 Objective Loss 0.190291 LR 0.000500 Time 0.020181 -2022-12-06 11:18:34,068 - Epoch: [125][ 860/ 1200] Overall Loss 0.190253 Objective Loss 0.190253 LR 0.000500 Time 0.020169 -2022-12-06 11:18:34,261 - Epoch: [125][ 870/ 1200] Overall Loss 0.190521 Objective Loss 0.190521 LR 0.000500 Time 0.020157 -2022-12-06 11:18:34,453 - Epoch: [125][ 880/ 1200] Overall Loss 0.190538 Objective Loss 0.190538 LR 0.000500 Time 0.020147 -2022-12-06 11:18:34,645 - Epoch: [125][ 890/ 1200] Overall Loss 0.190693 Objective Loss 0.190693 LR 0.000500 Time 0.020135 -2022-12-06 11:18:34,837 - Epoch: [125][ 900/ 1200] Overall Loss 0.190864 Objective Loss 0.190864 LR 0.000500 Time 0.020123 -2022-12-06 11:18:35,029 - Epoch: [125][ 910/ 1200] Overall Loss 0.190809 Objective Loss 0.190809 LR 0.000500 Time 0.020113 -2022-12-06 11:18:35,221 - Epoch: [125][ 920/ 1200] Overall Loss 0.190696 Objective Loss 0.190696 LR 0.000500 Time 0.020102 -2022-12-06 11:18:35,413 - Epoch: [125][ 930/ 1200] Overall Loss 0.190979 Objective Loss 0.190979 LR 0.000500 Time 0.020092 -2022-12-06 11:18:35,604 - Epoch: [125][ 940/ 1200] Overall Loss 0.190932 Objective Loss 0.190932 LR 0.000500 Time 0.020082 -2022-12-06 11:18:35,797 - Epoch: [125][ 950/ 1200] Overall Loss 0.190957 Objective Loss 0.190957 LR 0.000500 Time 0.020073 -2022-12-06 11:18:35,990 - Epoch: [125][ 960/ 1200] Overall Loss 0.190901 Objective Loss 0.190901 LR 0.000500 Time 0.020063 -2022-12-06 11:18:36,182 - Epoch: [125][ 970/ 1200] Overall Loss 0.191035 Objective Loss 0.191035 LR 0.000500 Time 0.020054 -2022-12-06 11:18:36,374 - Epoch: [125][ 980/ 1200] Overall Loss 0.190865 Objective Loss 0.190865 LR 0.000500 Time 0.020045 -2022-12-06 11:18:36,565 - Epoch: [125][ 990/ 1200] Overall Loss 0.191129 Objective Loss 0.191129 LR 0.000500 Time 0.020035 -2022-12-06 11:18:36,758 - Epoch: [125][ 1000/ 1200] Overall Loss 0.191051 Objective Loss 0.191051 LR 0.000500 Time 0.020026 -2022-12-06 11:18:36,950 - Epoch: [125][ 1010/ 1200] Overall Loss 0.190833 Objective Loss 0.190833 LR 0.000500 Time 0.020018 -2022-12-06 11:18:37,141 - Epoch: [125][ 1020/ 1200] Overall Loss 0.191045 Objective Loss 0.191045 LR 0.000500 Time 0.020009 -2022-12-06 11:18:37,334 - Epoch: [125][ 1030/ 1200] Overall Loss 0.191080 Objective Loss 0.191080 LR 0.000500 Time 0.020001 -2022-12-06 11:18:37,527 - Epoch: [125][ 1040/ 1200] Overall Loss 0.191211 Objective Loss 0.191211 LR 0.000500 Time 0.019994 -2022-12-06 11:18:37,719 - Epoch: [125][ 1050/ 1200] Overall Loss 0.191334 Objective Loss 0.191334 LR 0.000500 Time 0.019986 -2022-12-06 11:18:37,911 - Epoch: [125][ 1060/ 1200] Overall Loss 0.191347 Objective Loss 0.191347 LR 0.000500 Time 0.019978 -2022-12-06 11:18:38,104 - Epoch: [125][ 1070/ 1200] Overall Loss 0.191267 Objective Loss 0.191267 LR 0.000500 Time 0.019971 -2022-12-06 11:18:38,295 - Epoch: [125][ 1080/ 1200] Overall Loss 0.191307 Objective Loss 0.191307 LR 0.000500 Time 0.019962 -2022-12-06 11:18:38,487 - Epoch: [125][ 1090/ 1200] Overall Loss 0.191620 Objective Loss 0.191620 LR 0.000500 Time 0.019955 -2022-12-06 11:18:38,679 - Epoch: [125][ 1100/ 1200] Overall Loss 0.191712 Objective Loss 0.191712 LR 0.000500 Time 0.019947 -2022-12-06 11:18:38,871 - Epoch: [125][ 1110/ 1200] Overall Loss 0.191697 Objective Loss 0.191697 LR 0.000500 Time 0.019940 -2022-12-06 11:18:39,063 - Epoch: [125][ 1120/ 1200] Overall Loss 0.191525 Objective Loss 0.191525 LR 0.000500 Time 0.019933 -2022-12-06 11:18:39,254 - Epoch: [125][ 1130/ 1200] Overall Loss 0.191529 Objective Loss 0.191529 LR 0.000500 Time 0.019926 -2022-12-06 11:18:39,446 - Epoch: [125][ 1140/ 1200] Overall Loss 0.191518 Objective Loss 0.191518 LR 0.000500 Time 0.019919 -2022-12-06 11:18:39,638 - Epoch: [125][ 1150/ 1200] Overall Loss 0.191643 Objective Loss 0.191643 LR 0.000500 Time 0.019912 -2022-12-06 11:18:39,831 - Epoch: [125][ 1160/ 1200] Overall Loss 0.192090 Objective Loss 0.192090 LR 0.000500 Time 0.019906 -2022-12-06 11:18:40,024 - Epoch: [125][ 1170/ 1200] Overall Loss 0.192117 Objective Loss 0.192117 LR 0.000500 Time 0.019900 -2022-12-06 11:18:40,216 - Epoch: [125][ 1180/ 1200] Overall Loss 0.192093 Objective Loss 0.192093 LR 0.000500 Time 0.019894 -2022-12-06 11:18:40,409 - Epoch: [125][ 1190/ 1200] Overall Loss 0.192112 Objective Loss 0.192112 LR 0.000500 Time 0.019889 -2022-12-06 11:18:40,631 - Epoch: [125][ 1200/ 1200] Overall Loss 0.192087 Objective Loss 0.192087 Top1 89.748954 Top5 98.953975 LR 0.000500 Time 0.019907 -2022-12-06 11:18:40,720 - --- validate (epoch=125)----------- -2022-12-06 11:18:40,720 - 34129 samples (256 per mini-batch) -2022-12-06 11:18:41,191 - Epoch: [125][ 10/ 134] Loss 0.249922 Top1 87.460938 Top5 98.242188 -2022-12-06 11:18:41,329 - Epoch: [125][ 20/ 134] Loss 0.261306 Top1 87.382812 Top5 98.183594 -2022-12-06 11:18:41,464 - Epoch: [125][ 30/ 134] Loss 0.251293 Top1 87.330729 Top5 98.476562 -2022-12-06 11:18:41,600 - Epoch: [125][ 40/ 134] Loss 0.253791 Top1 87.294922 Top5 98.505859 -2022-12-06 11:18:41,728 - Epoch: [125][ 50/ 134] Loss 0.251445 Top1 87.320312 Top5 98.437500 -2022-12-06 11:18:41,860 - Epoch: [125][ 60/ 134] Loss 0.252484 Top1 87.259115 Top5 98.470052 -2022-12-06 11:18:41,989 - Epoch: [125][ 70/ 134] Loss 0.250035 Top1 87.187500 Top5 98.426339 -2022-12-06 11:18:42,120 - Epoch: [125][ 80/ 134] Loss 0.251036 Top1 87.128906 Top5 98.447266 -2022-12-06 11:18:42,250 - Epoch: [125][ 90/ 134] Loss 0.253531 Top1 86.996528 Top5 98.441840 -2022-12-06 11:18:42,380 - Epoch: [125][ 100/ 134] Loss 0.254564 Top1 86.937500 Top5 98.410156 -2022-12-06 11:18:42,514 - Epoch: [125][ 110/ 134] Loss 0.255138 Top1 86.960227 Top5 98.423295 -2022-12-06 11:18:42,648 - Epoch: [125][ 120/ 134] Loss 0.253991 Top1 86.936849 Top5 98.447266 -2022-12-06 11:18:42,782 - Epoch: [125][ 130/ 134] Loss 0.252501 Top1 86.992188 Top5 98.470553 -2022-12-06 11:18:42,820 - Epoch: [125][ 134/ 134] Loss 0.251844 Top1 86.981746 Top5 98.458789 -2022-12-06 11:18:42,910 - ==> Top1: 86.982 Top5: 98.459 Loss: 0.252 - -2022-12-06 11:18:42,911 - ==> Confusion: -[[ 929 0 0 2 4 8 1 0 2 33 0 1 1 3 4 2 3 0 0 0 3] - [ 0 938 4 5 6 23 2 9 0 1 6 3 1 1 0 1 1 0 17 4 5] - [ 5 2 1019 7 5 4 13 10 0 4 5 4 0 3 2 2 1 1 4 4 8] - [ 2 1 18 942 3 3 0 1 0 0 9 0 4 2 14 0 1 1 14 0 5] - [ 8 6 2 0 958 4 1 0 1 5 1 0 1 1 9 7 8 2 1 2 3] - [ 2 14 1 3 6 964 3 23 3 2 1 11 3 17 3 1 1 0 1 7 3] - [ 1 1 9 2 2 2 1081 3 0 0 0 1 0 1 0 2 1 2 2 7 1] - [ 2 4 7 1 1 20 11 956 0 0 2 5 0 0 0 2 0 1 22 14 6] - [ 3 4 0 0 1 1 0 1 966 43 14 2 4 9 10 0 1 1 2 1 1] - [ 92 0 1 0 5 2 0 3 28 852 2 1 0 7 2 1 0 0 1 0 4] - [ 1 0 5 5 0 0 1 3 5 0 967 1 2 9 6 0 1 0 5 2 6] - [ 6 0 1 0 1 12 3 3 2 0 1 968 24 3 0 6 4 6 0 8 3] - [ 3 0 1 2 0 1 1 1 0 0 0 23 904 1 1 5 2 10 0 5 9] - [ 1 0 0 0 0 4 0 4 9 10 5 2 5 964 1 2 3 0 1 3 9] - [ 10 3 4 7 3 1 0 1 14 2 1 1 3 1 1063 1 1 0 10 0 4] - [ 0 0 2 1 1 1 2 1 0 1 0 5 8 3 0 996 6 7 0 6 3] - [ 3 2 2 2 2 0 0 0 0 0 0 2 1 1 0 15 1033 0 0 5 4] - [ 2 0 1 7 1 1 2 0 1 2 0 11 16 2 1 15 3 964 1 2 4] - [ 4 5 6 9 2 2 0 20 1 1 5 2 2 1 8 1 0 1 929 1 8] - [ 2 4 2 0 1 5 8 3 0 1 1 12 6 5 1 4 3 3 1 1011 7] - [ 130 190 207 120 107 161 93 151 64 74 145 90 337 244 135 113 154 58 162 213 10278]] - -2022-12-06 11:18:43,570 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:18:43,570 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:18:43,576 - - -2022-12-06 11:18:43,576 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:18:44,509 - Epoch: [126][ 10/ 1200] Overall Loss 0.212944 Objective Loss 0.212944 LR 0.000500 Time 0.093227 -2022-12-06 11:18:44,707 - Epoch: [126][ 20/ 1200] Overall Loss 0.199061 Objective Loss 0.199061 LR 0.000500 Time 0.056518 -2022-12-06 11:18:44,899 - Epoch: [126][ 30/ 1200] Overall Loss 0.200989 Objective Loss 0.200989 LR 0.000500 Time 0.044044 -2022-12-06 11:18:45,090 - Epoch: [126][ 40/ 1200] Overall Loss 0.200547 Objective Loss 0.200547 LR 0.000500 Time 0.037798 -2022-12-06 11:18:45,282 - Epoch: [126][ 50/ 1200] Overall Loss 0.196294 Objective Loss 0.196294 LR 0.000500 Time 0.034064 -2022-12-06 11:18:45,473 - Epoch: [126][ 60/ 1200] Overall Loss 0.193996 Objective Loss 0.193996 LR 0.000500 Time 0.031561 -2022-12-06 11:18:45,663 - Epoch: [126][ 70/ 1200] Overall Loss 0.196291 Objective Loss 0.196291 LR 0.000500 Time 0.029761 -2022-12-06 11:18:45,854 - Epoch: [126][ 80/ 1200] Overall Loss 0.196758 Objective Loss 0.196758 LR 0.000500 Time 0.028414 -2022-12-06 11:18:46,045 - Epoch: [126][ 90/ 1200] Overall Loss 0.196153 Objective Loss 0.196153 LR 0.000500 Time 0.027372 -2022-12-06 11:18:46,235 - Epoch: [126][ 100/ 1200] Overall Loss 0.192219 Objective Loss 0.192219 LR 0.000500 Time 0.026533 -2022-12-06 11:18:46,426 - Epoch: [126][ 110/ 1200] Overall Loss 0.193006 Objective Loss 0.193006 LR 0.000500 Time 0.025857 -2022-12-06 11:18:46,617 - Epoch: [126][ 120/ 1200] Overall Loss 0.193398 Objective Loss 0.193398 LR 0.000500 Time 0.025286 -2022-12-06 11:18:46,808 - Epoch: [126][ 130/ 1200] Overall Loss 0.192259 Objective Loss 0.192259 LR 0.000500 Time 0.024804 -2022-12-06 11:18:46,999 - Epoch: [126][ 140/ 1200] Overall Loss 0.193104 Objective Loss 0.193104 LR 0.000500 Time 0.024392 -2022-12-06 11:18:47,190 - Epoch: [126][ 150/ 1200] Overall Loss 0.192783 Objective Loss 0.192783 LR 0.000500 Time 0.024038 -2022-12-06 11:18:47,381 - Epoch: [126][ 160/ 1200] Overall Loss 0.192478 Objective Loss 0.192478 LR 0.000500 Time 0.023726 -2022-12-06 11:18:47,571 - Epoch: [126][ 170/ 1200] Overall Loss 0.192585 Objective Loss 0.192585 LR 0.000500 Time 0.023446 -2022-12-06 11:18:47,762 - Epoch: [126][ 180/ 1200] Overall Loss 0.192071 Objective Loss 0.192071 LR 0.000500 Time 0.023202 -2022-12-06 11:18:47,953 - Epoch: [126][ 190/ 1200] Overall Loss 0.192217 Objective Loss 0.192217 LR 0.000500 Time 0.022983 -2022-12-06 11:18:48,145 - Epoch: [126][ 200/ 1200] Overall Loss 0.192474 Objective Loss 0.192474 LR 0.000500 Time 0.022791 -2022-12-06 11:18:48,335 - Epoch: [126][ 210/ 1200] Overall Loss 0.192331 Objective Loss 0.192331 LR 0.000500 Time 0.022609 -2022-12-06 11:18:48,527 - Epoch: [126][ 220/ 1200] Overall Loss 0.191656 Objective Loss 0.191656 LR 0.000500 Time 0.022447 -2022-12-06 11:18:48,718 - Epoch: [126][ 230/ 1200] Overall Loss 0.191014 Objective Loss 0.191014 LR 0.000500 Time 0.022300 -2022-12-06 11:18:48,908 - Epoch: [126][ 240/ 1200] Overall Loss 0.191288 Objective Loss 0.191288 LR 0.000500 Time 0.022163 -2022-12-06 11:18:49,099 - Epoch: [126][ 250/ 1200] Overall Loss 0.191301 Objective Loss 0.191301 LR 0.000500 Time 0.022036 -2022-12-06 11:18:49,289 - Epoch: [126][ 260/ 1200] Overall Loss 0.191198 Objective Loss 0.191198 LR 0.000500 Time 0.021918 -2022-12-06 11:18:49,479 - Epoch: [126][ 270/ 1200] Overall Loss 0.189734 Objective Loss 0.189734 LR 0.000500 Time 0.021808 -2022-12-06 11:18:49,670 - Epoch: [126][ 280/ 1200] Overall Loss 0.189872 Objective Loss 0.189872 LR 0.000500 Time 0.021709 -2022-12-06 11:18:49,861 - Epoch: [126][ 290/ 1200] Overall Loss 0.189756 Objective Loss 0.189756 LR 0.000500 Time 0.021615 -2022-12-06 11:18:50,052 - Epoch: [126][ 300/ 1200] Overall Loss 0.190134 Objective Loss 0.190134 LR 0.000500 Time 0.021530 -2022-12-06 11:18:50,243 - Epoch: [126][ 310/ 1200] Overall Loss 0.189025 Objective Loss 0.189025 LR 0.000500 Time 0.021450 -2022-12-06 11:18:50,434 - Epoch: [126][ 320/ 1200] Overall Loss 0.189045 Objective Loss 0.189045 LR 0.000500 Time 0.021376 -2022-12-06 11:18:50,625 - Epoch: [126][ 330/ 1200] Overall Loss 0.189608 Objective Loss 0.189608 LR 0.000500 Time 0.021304 -2022-12-06 11:18:50,815 - Epoch: [126][ 340/ 1200] Overall Loss 0.188734 Objective Loss 0.188734 LR 0.000500 Time 0.021236 -2022-12-06 11:18:51,006 - Epoch: [126][ 350/ 1200] Overall Loss 0.187966 Objective Loss 0.187966 LR 0.000500 Time 0.021173 -2022-12-06 11:18:51,196 - Epoch: [126][ 360/ 1200] Overall Loss 0.187908 Objective Loss 0.187908 LR 0.000500 Time 0.021113 -2022-12-06 11:18:51,387 - Epoch: [126][ 370/ 1200] Overall Loss 0.188132 Objective Loss 0.188132 LR 0.000500 Time 0.021054 -2022-12-06 11:18:51,577 - Epoch: [126][ 380/ 1200] Overall Loss 0.188439 Objective Loss 0.188439 LR 0.000500 Time 0.020999 -2022-12-06 11:18:51,767 - Epoch: [126][ 390/ 1200] Overall Loss 0.188941 Objective Loss 0.188941 LR 0.000500 Time 0.020948 -2022-12-06 11:18:51,958 - Epoch: [126][ 400/ 1200] Overall Loss 0.188832 Objective Loss 0.188832 LR 0.000500 Time 0.020899 -2022-12-06 11:18:52,149 - Epoch: [126][ 410/ 1200] Overall Loss 0.188810 Objective Loss 0.188810 LR 0.000500 Time 0.020853 -2022-12-06 11:18:52,340 - Epoch: [126][ 420/ 1200] Overall Loss 0.188901 Objective Loss 0.188901 LR 0.000500 Time 0.020811 -2022-12-06 11:18:52,530 - Epoch: [126][ 430/ 1200] Overall Loss 0.189104 Objective Loss 0.189104 LR 0.000500 Time 0.020768 -2022-12-06 11:18:52,721 - Epoch: [126][ 440/ 1200] Overall Loss 0.190080 Objective Loss 0.190080 LR 0.000500 Time 0.020729 -2022-12-06 11:18:52,912 - Epoch: [126][ 450/ 1200] Overall Loss 0.190224 Objective Loss 0.190224 LR 0.000500 Time 0.020691 -2022-12-06 11:18:53,102 - Epoch: [126][ 460/ 1200] Overall Loss 0.190116 Objective Loss 0.190116 LR 0.000500 Time 0.020654 -2022-12-06 11:18:53,292 - Epoch: [126][ 470/ 1200] Overall Loss 0.190424 Objective Loss 0.190424 LR 0.000500 Time 0.020618 -2022-12-06 11:18:53,483 - Epoch: [126][ 480/ 1200] Overall Loss 0.190478 Objective Loss 0.190478 LR 0.000500 Time 0.020583 -2022-12-06 11:18:53,674 - Epoch: [126][ 490/ 1200] Overall Loss 0.190800 Objective Loss 0.190800 LR 0.000500 Time 0.020553 -2022-12-06 11:18:53,865 - Epoch: [126][ 500/ 1200] Overall Loss 0.190721 Objective Loss 0.190721 LR 0.000500 Time 0.020522 -2022-12-06 11:18:54,055 - Epoch: [126][ 510/ 1200] Overall Loss 0.190883 Objective Loss 0.190883 LR 0.000500 Time 0.020491 -2022-12-06 11:18:54,246 - Epoch: [126][ 520/ 1200] Overall Loss 0.190426 Objective Loss 0.190426 LR 0.000500 Time 0.020464 -2022-12-06 11:18:54,437 - Epoch: [126][ 530/ 1200] Overall Loss 0.190379 Objective Loss 0.190379 LR 0.000500 Time 0.020437 -2022-12-06 11:18:54,628 - Epoch: [126][ 540/ 1200] Overall Loss 0.190346 Objective Loss 0.190346 LR 0.000500 Time 0.020411 -2022-12-06 11:18:54,818 - Epoch: [126][ 550/ 1200] Overall Loss 0.190045 Objective Loss 0.190045 LR 0.000500 Time 0.020384 -2022-12-06 11:18:55,008 - Epoch: [126][ 560/ 1200] Overall Loss 0.190639 Objective Loss 0.190639 LR 0.000500 Time 0.020359 -2022-12-06 11:18:55,199 - Epoch: [126][ 570/ 1200] Overall Loss 0.190090 Objective Loss 0.190090 LR 0.000500 Time 0.020336 -2022-12-06 11:18:55,390 - Epoch: [126][ 580/ 1200] Overall Loss 0.190049 Objective Loss 0.190049 LR 0.000500 Time 0.020313 -2022-12-06 11:18:55,580 - Epoch: [126][ 590/ 1200] Overall Loss 0.189504 Objective Loss 0.189504 LR 0.000500 Time 0.020291 -2022-12-06 11:18:55,772 - Epoch: [126][ 600/ 1200] Overall Loss 0.189988 Objective Loss 0.189988 LR 0.000500 Time 0.020271 -2022-12-06 11:18:55,963 - Epoch: [126][ 610/ 1200] Overall Loss 0.189690 Objective Loss 0.189690 LR 0.000500 Time 0.020251 -2022-12-06 11:18:56,154 - Epoch: [126][ 620/ 1200] Overall Loss 0.189865 Objective Loss 0.189865 LR 0.000500 Time 0.020231 -2022-12-06 11:18:56,346 - Epoch: [126][ 630/ 1200] Overall Loss 0.189908 Objective Loss 0.189908 LR 0.000500 Time 0.020214 -2022-12-06 11:18:56,536 - Epoch: [126][ 640/ 1200] Overall Loss 0.190134 Objective Loss 0.190134 LR 0.000500 Time 0.020195 -2022-12-06 11:18:56,727 - Epoch: [126][ 650/ 1200] Overall Loss 0.190171 Objective Loss 0.190171 LR 0.000500 Time 0.020177 -2022-12-06 11:18:56,918 - Epoch: [126][ 660/ 1200] Overall Loss 0.190269 Objective Loss 0.190269 LR 0.000500 Time 0.020159 -2022-12-06 11:18:57,108 - Epoch: [126][ 670/ 1200] Overall Loss 0.190458 Objective Loss 0.190458 LR 0.000500 Time 0.020142 -2022-12-06 11:18:57,299 - Epoch: [126][ 680/ 1200] Overall Loss 0.190937 Objective Loss 0.190937 LR 0.000500 Time 0.020126 -2022-12-06 11:18:57,490 - Epoch: [126][ 690/ 1200] Overall Loss 0.190910 Objective Loss 0.190910 LR 0.000500 Time 0.020110 -2022-12-06 11:18:57,681 - Epoch: [126][ 700/ 1200] Overall Loss 0.191137 Objective Loss 0.191137 LR 0.000500 Time 0.020094 -2022-12-06 11:18:57,871 - Epoch: [126][ 710/ 1200] Overall Loss 0.191002 Objective Loss 0.191002 LR 0.000500 Time 0.020078 -2022-12-06 11:18:58,061 - Epoch: [126][ 720/ 1200] Overall Loss 0.191063 Objective Loss 0.191063 LR 0.000500 Time 0.020063 -2022-12-06 11:18:58,252 - Epoch: [126][ 730/ 1200] Overall Loss 0.190940 Objective Loss 0.190940 LR 0.000500 Time 0.020048 -2022-12-06 11:18:58,443 - Epoch: [126][ 740/ 1200] Overall Loss 0.190731 Objective Loss 0.190731 LR 0.000500 Time 0.020035 -2022-12-06 11:18:58,635 - Epoch: [126][ 750/ 1200] Overall Loss 0.191049 Objective Loss 0.191049 LR 0.000500 Time 0.020023 -2022-12-06 11:18:58,825 - Epoch: [126][ 760/ 1200] Overall Loss 0.191254 Objective Loss 0.191254 LR 0.000500 Time 0.020009 -2022-12-06 11:18:59,015 - Epoch: [126][ 770/ 1200] Overall Loss 0.191263 Objective Loss 0.191263 LR 0.000500 Time 0.019995 -2022-12-06 11:18:59,205 - Epoch: [126][ 780/ 1200] Overall Loss 0.191135 Objective Loss 0.191135 LR 0.000500 Time 0.019982 -2022-12-06 11:18:59,397 - Epoch: [126][ 790/ 1200] Overall Loss 0.191318 Objective Loss 0.191318 LR 0.000500 Time 0.019971 -2022-12-06 11:18:59,588 - Epoch: [126][ 800/ 1200] Overall Loss 0.191167 Objective Loss 0.191167 LR 0.000500 Time 0.019959 -2022-12-06 11:18:59,778 - Epoch: [126][ 810/ 1200] Overall Loss 0.191163 Objective Loss 0.191163 LR 0.000500 Time 0.019947 -2022-12-06 11:18:59,969 - Epoch: [126][ 820/ 1200] Overall Loss 0.191257 Objective Loss 0.191257 LR 0.000500 Time 0.019936 -2022-12-06 11:19:00,159 - Epoch: [126][ 830/ 1200] Overall Loss 0.191268 Objective Loss 0.191268 LR 0.000500 Time 0.019925 -2022-12-06 11:19:00,351 - Epoch: [126][ 840/ 1200] Overall Loss 0.191349 Objective Loss 0.191349 LR 0.000500 Time 0.019915 -2022-12-06 11:19:00,542 - Epoch: [126][ 850/ 1200] Overall Loss 0.191139 Objective Loss 0.191139 LR 0.000500 Time 0.019905 -2022-12-06 11:19:00,733 - Epoch: [126][ 860/ 1200] Overall Loss 0.191195 Objective Loss 0.191195 LR 0.000500 Time 0.019894 -2022-12-06 11:19:00,923 - Epoch: [126][ 870/ 1200] Overall Loss 0.191164 Objective Loss 0.191164 LR 0.000500 Time 0.019883 -2022-12-06 11:19:01,113 - Epoch: [126][ 880/ 1200] Overall Loss 0.191153 Objective Loss 0.191153 LR 0.000500 Time 0.019873 -2022-12-06 11:19:01,304 - Epoch: [126][ 890/ 1200] Overall Loss 0.191252 Objective Loss 0.191252 LR 0.000500 Time 0.019864 -2022-12-06 11:19:01,494 - Epoch: [126][ 900/ 1200] Overall Loss 0.191174 Objective Loss 0.191174 LR 0.000500 Time 0.019854 -2022-12-06 11:19:01,686 - Epoch: [126][ 910/ 1200] Overall Loss 0.191167 Objective Loss 0.191167 LR 0.000500 Time 0.019845 -2022-12-06 11:19:01,877 - Epoch: [126][ 920/ 1200] Overall Loss 0.191365 Objective Loss 0.191365 LR 0.000500 Time 0.019836 -2022-12-06 11:19:02,067 - Epoch: [126][ 930/ 1200] Overall Loss 0.191333 Objective Loss 0.191333 LR 0.000500 Time 0.019827 -2022-12-06 11:19:02,257 - Epoch: [126][ 940/ 1200] Overall Loss 0.191325 Objective Loss 0.191325 LR 0.000500 Time 0.019818 -2022-12-06 11:19:02,449 - Epoch: [126][ 950/ 1200] Overall Loss 0.191408 Objective Loss 0.191408 LR 0.000500 Time 0.019811 -2022-12-06 11:19:02,640 - Epoch: [126][ 960/ 1200] Overall Loss 0.191452 Objective Loss 0.191452 LR 0.000500 Time 0.019802 -2022-12-06 11:19:02,831 - Epoch: [126][ 970/ 1200] Overall Loss 0.191548 Objective Loss 0.191548 LR 0.000500 Time 0.019795 -2022-12-06 11:19:03,022 - Epoch: [126][ 980/ 1200] Overall Loss 0.191466 Objective Loss 0.191466 LR 0.000500 Time 0.019788 -2022-12-06 11:19:03,214 - Epoch: [126][ 990/ 1200] Overall Loss 0.191602 Objective Loss 0.191602 LR 0.000500 Time 0.019781 -2022-12-06 11:19:03,405 - Epoch: [126][ 1000/ 1200] Overall Loss 0.191688 Objective Loss 0.191688 LR 0.000500 Time 0.019773 -2022-12-06 11:19:03,595 - Epoch: [126][ 1010/ 1200] Overall Loss 0.191640 Objective Loss 0.191640 LR 0.000500 Time 0.019765 -2022-12-06 11:19:03,786 - Epoch: [126][ 1020/ 1200] Overall Loss 0.191737 Objective Loss 0.191737 LR 0.000500 Time 0.019758 -2022-12-06 11:19:03,975 - Epoch: [126][ 1030/ 1200] Overall Loss 0.191710 Objective Loss 0.191710 LR 0.000500 Time 0.019750 -2022-12-06 11:19:04,166 - Epoch: [126][ 1040/ 1200] Overall Loss 0.191753 Objective Loss 0.191753 LR 0.000500 Time 0.019742 -2022-12-06 11:19:04,357 - Epoch: [126][ 1050/ 1200] Overall Loss 0.191790 Objective Loss 0.191790 LR 0.000500 Time 0.019736 -2022-12-06 11:19:04,548 - Epoch: [126][ 1060/ 1200] Overall Loss 0.191700 Objective Loss 0.191700 LR 0.000500 Time 0.019729 -2022-12-06 11:19:04,738 - Epoch: [126][ 1070/ 1200] Overall Loss 0.191756 Objective Loss 0.191756 LR 0.000500 Time 0.019722 -2022-12-06 11:19:04,929 - Epoch: [126][ 1080/ 1200] Overall Loss 0.191960 Objective Loss 0.191960 LR 0.000500 Time 0.019716 -2022-12-06 11:19:05,119 - Epoch: [126][ 1090/ 1200] Overall Loss 0.191878 Objective Loss 0.191878 LR 0.000500 Time 0.019708 -2022-12-06 11:19:05,309 - Epoch: [126][ 1100/ 1200] Overall Loss 0.191801 Objective Loss 0.191801 LR 0.000500 Time 0.019702 -2022-12-06 11:19:05,499 - Epoch: [126][ 1110/ 1200] Overall Loss 0.191954 Objective Loss 0.191954 LR 0.000500 Time 0.019695 -2022-12-06 11:19:05,689 - Epoch: [126][ 1120/ 1200] Overall Loss 0.192141 Objective Loss 0.192141 LR 0.000500 Time 0.019689 -2022-12-06 11:19:05,880 - Epoch: [126][ 1130/ 1200] Overall Loss 0.192112 Objective Loss 0.192112 LR 0.000500 Time 0.019683 -2022-12-06 11:19:06,070 - Epoch: [126][ 1140/ 1200] Overall Loss 0.192343 Objective Loss 0.192343 LR 0.000500 Time 0.019676 -2022-12-06 11:19:06,260 - Epoch: [126][ 1150/ 1200] Overall Loss 0.192173 Objective Loss 0.192173 LR 0.000500 Time 0.019670 -2022-12-06 11:19:06,451 - Epoch: [126][ 1160/ 1200] Overall Loss 0.192151 Objective Loss 0.192151 LR 0.000500 Time 0.019664 -2022-12-06 11:19:06,642 - Epoch: [126][ 1170/ 1200] Overall Loss 0.192235 Objective Loss 0.192235 LR 0.000500 Time 0.019659 -2022-12-06 11:19:06,832 - Epoch: [126][ 1180/ 1200] Overall Loss 0.192371 Objective Loss 0.192371 LR 0.000500 Time 0.019653 -2022-12-06 11:19:07,023 - Epoch: [126][ 1190/ 1200] Overall Loss 0.192533 Objective Loss 0.192533 LR 0.000500 Time 0.019648 -2022-12-06 11:19:07,244 - Epoch: [126][ 1200/ 1200] Overall Loss 0.192392 Objective Loss 0.192392 Top1 89.121339 Top5 99.581590 LR 0.000500 Time 0.019668 -2022-12-06 11:19:07,333 - --- validate (epoch=126)----------- -2022-12-06 11:19:07,333 - 34129 samples (256 per mini-batch) -2022-12-06 11:19:07,789 - Epoch: [126][ 10/ 134] Loss 0.259364 Top1 87.031250 Top5 98.242188 -2022-12-06 11:19:07,920 - Epoch: [126][ 20/ 134] Loss 0.246590 Top1 87.558594 Top5 98.359375 -2022-12-06 11:19:08,052 - Epoch: [126][ 30/ 134] Loss 0.242486 Top1 87.265625 Top5 98.359375 -2022-12-06 11:19:08,187 - Epoch: [126][ 40/ 134] Loss 0.241096 Top1 87.080078 Top5 98.300781 -2022-12-06 11:19:08,318 - Epoch: [126][ 50/ 134] Loss 0.242752 Top1 87.031250 Top5 98.375000 -2022-12-06 11:19:08,451 - Epoch: [126][ 60/ 134] Loss 0.251914 Top1 86.940104 Top5 98.378906 -2022-12-06 11:19:08,583 - Epoch: [126][ 70/ 134] Loss 0.254421 Top1 86.925223 Top5 98.348214 -2022-12-06 11:19:08,716 - Epoch: [126][ 80/ 134] Loss 0.257216 Top1 86.835938 Top5 98.320312 -2022-12-06 11:19:08,847 - Epoch: [126][ 90/ 134] Loss 0.256649 Top1 86.931424 Top5 98.346354 -2022-12-06 11:19:08,980 - Epoch: [126][ 100/ 134] Loss 0.250652 Top1 86.933594 Top5 98.355469 -2022-12-06 11:19:09,111 - Epoch: [126][ 110/ 134] Loss 0.252261 Top1 86.875000 Top5 98.330966 -2022-12-06 11:19:09,245 - Epoch: [126][ 120/ 134] Loss 0.253969 Top1 86.858724 Top5 98.330078 -2022-12-06 11:19:09,377 - Epoch: [126][ 130/ 134] Loss 0.251250 Top1 86.893029 Top5 98.338341 -2022-12-06 11:19:09,415 - Epoch: [126][ 134/ 134] Loss 0.252485 Top1 86.870403 Top5 98.341586 -2022-12-06 11:19:09,518 - ==> Top1: 86.870 Top5: 98.342 Loss: 0.252 - -2022-12-06 11:19:09,519 - ==> Confusion: -[[ 913 1 1 2 9 7 0 1 7 41 0 2 1 3 2 2 0 0 0 0 4] - [ 2 923 3 3 9 23 3 17 2 1 3 4 1 1 1 2 3 0 13 3 10] - [ 5 3 1003 17 6 2 18 8 0 3 7 3 2 0 3 2 1 1 7 2 10] - [ 2 1 12 952 3 5 1 0 0 0 9 0 4 1 12 0 0 2 10 0 6] - [ 9 6 1 0 953 7 0 0 1 6 2 5 0 1 14 7 5 0 1 0 2] - [ 4 18 0 3 3 981 1 17 3 2 1 9 4 10 1 1 3 0 1 4 3] - [ 1 5 9 2 0 1 1068 6 0 0 1 2 1 1 0 5 1 4 2 7 2] - [ 3 8 5 1 2 30 8 947 0 1 4 4 1 1 2 0 1 0 21 7 8] - [ 3 2 0 0 1 2 0 0 974 39 12 1 1 7 13 1 3 2 2 1 0] - [ 72 0 1 0 3 5 0 3 31 862 1 1 0 10 4 1 0 0 0 0 7] - [ 0 1 2 6 1 1 1 3 10 0 964 1 3 7 5 1 2 0 5 2 4] - [ 2 0 1 0 1 13 1 4 0 1 0 972 27 3 0 4 3 5 0 11 3] - [ 2 0 2 1 0 1 0 2 2 1 0 20 914 2 1 3 0 8 0 4 6] - [ 1 0 0 0 0 7 0 3 8 10 6 5 5 960 4 1 4 0 0 2 7] - [ 9 3 1 9 3 2 0 0 15 2 1 2 2 3 1063 0 0 1 10 1 3] - [ 1 0 2 0 1 0 3 1 0 0 0 7 13 3 0 988 4 11 0 5 4] - [ 2 4 0 1 3 1 2 0 0 0 0 1 1 1 0 13 1031 0 0 5 7] - [ 5 0 1 2 0 1 0 1 3 2 0 5 22 2 1 12 0 972 2 2 3] - [ 3 4 4 6 3 2 0 19 2 1 8 1 2 1 8 1 1 0 936 2 4] - [ 2 3 1 0 2 4 5 12 2 0 2 9 4 9 0 4 6 3 2 1004 6] - [ 129 171 162 112 114 215 101 126 83 68 164 88 344 263 165 101 157 73 151 177 10262]] - -2022-12-06 11:19:10,105 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:19:10,105 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:19:10,111 - - -2022-12-06 11:19:10,111 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:19:11,153 - Epoch: [127][ 10/ 1200] Overall Loss 0.176354 Objective Loss 0.176354 LR 0.000500 Time 0.104114 -2022-12-06 11:19:11,348 - Epoch: [127][ 20/ 1200] Overall Loss 0.169849 Objective Loss 0.169849 LR 0.000500 Time 0.061770 -2022-12-06 11:19:11,543 - Epoch: [127][ 30/ 1200] Overall Loss 0.185482 Objective Loss 0.185482 LR 0.000500 Time 0.047673 -2022-12-06 11:19:11,737 - Epoch: [127][ 40/ 1200] Overall Loss 0.185221 Objective Loss 0.185221 LR 0.000500 Time 0.040599 -2022-12-06 11:19:11,932 - Epoch: [127][ 50/ 1200] Overall Loss 0.185357 Objective Loss 0.185357 LR 0.000500 Time 0.036363 -2022-12-06 11:19:12,126 - Epoch: [127][ 60/ 1200] Overall Loss 0.183775 Objective Loss 0.183775 LR 0.000500 Time 0.033527 -2022-12-06 11:19:12,321 - Epoch: [127][ 70/ 1200] Overall Loss 0.185778 Objective Loss 0.185778 LR 0.000500 Time 0.031510 -2022-12-06 11:19:12,515 - Epoch: [127][ 80/ 1200] Overall Loss 0.184478 Objective Loss 0.184478 LR 0.000500 Time 0.029989 -2022-12-06 11:19:12,709 - Epoch: [127][ 90/ 1200] Overall Loss 0.183199 Objective Loss 0.183199 LR 0.000500 Time 0.028814 -2022-12-06 11:19:12,903 - Epoch: [127][ 100/ 1200] Overall Loss 0.183032 Objective Loss 0.183032 LR 0.000500 Time 0.027860 -2022-12-06 11:19:13,097 - Epoch: [127][ 110/ 1200] Overall Loss 0.184626 Objective Loss 0.184626 LR 0.000500 Time 0.027089 -2022-12-06 11:19:13,290 - Epoch: [127][ 120/ 1200] Overall Loss 0.184930 Objective Loss 0.184930 LR 0.000500 Time 0.026439 -2022-12-06 11:19:13,485 - Epoch: [127][ 130/ 1200] Overall Loss 0.184897 Objective Loss 0.184897 LR 0.000500 Time 0.025898 -2022-12-06 11:19:13,678 - Epoch: [127][ 140/ 1200] Overall Loss 0.184015 Objective Loss 0.184015 LR 0.000500 Time 0.025425 -2022-12-06 11:19:13,873 - Epoch: [127][ 150/ 1200] Overall Loss 0.184377 Objective Loss 0.184377 LR 0.000500 Time 0.025023 -2022-12-06 11:19:14,067 - Epoch: [127][ 160/ 1200] Overall Loss 0.185226 Objective Loss 0.185226 LR 0.000500 Time 0.024671 -2022-12-06 11:19:14,262 - Epoch: [127][ 170/ 1200] Overall Loss 0.185043 Objective Loss 0.185043 LR 0.000500 Time 0.024365 -2022-12-06 11:19:14,456 - Epoch: [127][ 180/ 1200] Overall Loss 0.185793 Objective Loss 0.185793 LR 0.000500 Time 0.024083 -2022-12-06 11:19:14,649 - Epoch: [127][ 190/ 1200] Overall Loss 0.186971 Objective Loss 0.186971 LR 0.000500 Time 0.023829 -2022-12-06 11:19:14,842 - Epoch: [127][ 200/ 1200] Overall Loss 0.187100 Objective Loss 0.187100 LR 0.000500 Time 0.023598 -2022-12-06 11:19:15,033 - Epoch: [127][ 210/ 1200] Overall Loss 0.186455 Objective Loss 0.186455 LR 0.000500 Time 0.023385 -2022-12-06 11:19:15,226 - Epoch: [127][ 220/ 1200] Overall Loss 0.187021 Objective Loss 0.187021 LR 0.000500 Time 0.023192 -2022-12-06 11:19:15,418 - Epoch: [127][ 230/ 1200] Overall Loss 0.186885 Objective Loss 0.186885 LR 0.000500 Time 0.023016 -2022-12-06 11:19:15,609 - Epoch: [127][ 240/ 1200] Overall Loss 0.186893 Objective Loss 0.186893 LR 0.000500 Time 0.022854 -2022-12-06 11:19:15,801 - Epoch: [127][ 250/ 1200] Overall Loss 0.186283 Objective Loss 0.186283 LR 0.000500 Time 0.022705 -2022-12-06 11:19:15,993 - Epoch: [127][ 260/ 1200] Overall Loss 0.186159 Objective Loss 0.186159 LR 0.000500 Time 0.022568 -2022-12-06 11:19:16,186 - Epoch: [127][ 270/ 1200] Overall Loss 0.186403 Objective Loss 0.186403 LR 0.000500 Time 0.022443 -2022-12-06 11:19:16,378 - Epoch: [127][ 280/ 1200] Overall Loss 0.186132 Objective Loss 0.186132 LR 0.000500 Time 0.022326 -2022-12-06 11:19:16,570 - Epoch: [127][ 290/ 1200] Overall Loss 0.186181 Objective Loss 0.186181 LR 0.000500 Time 0.022218 -2022-12-06 11:19:16,763 - Epoch: [127][ 300/ 1200] Overall Loss 0.186698 Objective Loss 0.186698 LR 0.000500 Time 0.022118 -2022-12-06 11:19:16,955 - Epoch: [127][ 310/ 1200] Overall Loss 0.185705 Objective Loss 0.185705 LR 0.000500 Time 0.022022 -2022-12-06 11:19:17,147 - Epoch: [127][ 320/ 1200] Overall Loss 0.186085 Objective Loss 0.186085 LR 0.000500 Time 0.021932 -2022-12-06 11:19:17,339 - Epoch: [127][ 330/ 1200] Overall Loss 0.185700 Objective Loss 0.185700 LR 0.000500 Time 0.021848 -2022-12-06 11:19:17,531 - Epoch: [127][ 340/ 1200] Overall Loss 0.185411 Objective Loss 0.185411 LR 0.000500 Time 0.021768 -2022-12-06 11:19:17,723 - Epoch: [127][ 350/ 1200] Overall Loss 0.186339 Objective Loss 0.186339 LR 0.000500 Time 0.021694 -2022-12-06 11:19:17,916 - Epoch: [127][ 360/ 1200] Overall Loss 0.186202 Objective Loss 0.186202 LR 0.000500 Time 0.021624 -2022-12-06 11:19:18,108 - Epoch: [127][ 370/ 1200] Overall Loss 0.186453 Objective Loss 0.186453 LR 0.000500 Time 0.021556 -2022-12-06 11:19:18,300 - Epoch: [127][ 380/ 1200] Overall Loss 0.186691 Objective Loss 0.186691 LR 0.000500 Time 0.021494 -2022-12-06 11:19:18,492 - Epoch: [127][ 390/ 1200] Overall Loss 0.186820 Objective Loss 0.186820 LR 0.000500 Time 0.021433 -2022-12-06 11:19:18,684 - Epoch: [127][ 400/ 1200] Overall Loss 0.186558 Objective Loss 0.186558 LR 0.000500 Time 0.021376 -2022-12-06 11:19:18,876 - Epoch: [127][ 410/ 1200] Overall Loss 0.186437 Objective Loss 0.186437 LR 0.000500 Time 0.021322 -2022-12-06 11:19:19,067 - Epoch: [127][ 420/ 1200] Overall Loss 0.186773 Objective Loss 0.186773 LR 0.000500 Time 0.021269 -2022-12-06 11:19:19,259 - Epoch: [127][ 430/ 1200] Overall Loss 0.187588 Objective Loss 0.187588 LR 0.000500 Time 0.021218 -2022-12-06 11:19:19,451 - Epoch: [127][ 440/ 1200] Overall Loss 0.187426 Objective Loss 0.187426 LR 0.000500 Time 0.021171 -2022-12-06 11:19:19,643 - Epoch: [127][ 450/ 1200] Overall Loss 0.187128 Objective Loss 0.187128 LR 0.000500 Time 0.021126 -2022-12-06 11:19:19,835 - Epoch: [127][ 460/ 1200] Overall Loss 0.186903 Objective Loss 0.186903 LR 0.000500 Time 0.021083 -2022-12-06 11:19:20,027 - Epoch: [127][ 470/ 1200] Overall Loss 0.186295 Objective Loss 0.186295 LR 0.000500 Time 0.021042 -2022-12-06 11:19:20,219 - Epoch: [127][ 480/ 1200] Overall Loss 0.186614 Objective Loss 0.186614 LR 0.000500 Time 0.021002 -2022-12-06 11:19:20,411 - Epoch: [127][ 490/ 1200] Overall Loss 0.186861 Objective Loss 0.186861 LR 0.000500 Time 0.020964 -2022-12-06 11:19:20,603 - Epoch: [127][ 500/ 1200] Overall Loss 0.187203 Objective Loss 0.187203 LR 0.000500 Time 0.020928 -2022-12-06 11:19:20,795 - Epoch: [127][ 510/ 1200] Overall Loss 0.187551 Objective Loss 0.187551 LR 0.000500 Time 0.020893 -2022-12-06 11:19:20,988 - Epoch: [127][ 520/ 1200] Overall Loss 0.187767 Objective Loss 0.187767 LR 0.000500 Time 0.020861 -2022-12-06 11:19:21,180 - Epoch: [127][ 530/ 1200] Overall Loss 0.187575 Objective Loss 0.187575 LR 0.000500 Time 0.020829 -2022-12-06 11:19:21,372 - Epoch: [127][ 540/ 1200] Overall Loss 0.187964 Objective Loss 0.187964 LR 0.000500 Time 0.020798 -2022-12-06 11:19:21,564 - Epoch: [127][ 550/ 1200] Overall Loss 0.187949 Objective Loss 0.187949 LR 0.000500 Time 0.020769 -2022-12-06 11:19:21,757 - Epoch: [127][ 560/ 1200] Overall Loss 0.188027 Objective Loss 0.188027 LR 0.000500 Time 0.020740 -2022-12-06 11:19:21,949 - Epoch: [127][ 570/ 1200] Overall Loss 0.188531 Objective Loss 0.188531 LR 0.000500 Time 0.020713 -2022-12-06 11:19:22,141 - Epoch: [127][ 580/ 1200] Overall Loss 0.188910 Objective Loss 0.188910 LR 0.000500 Time 0.020686 -2022-12-06 11:19:22,333 - Epoch: [127][ 590/ 1200] Overall Loss 0.189039 Objective Loss 0.189039 LR 0.000500 Time 0.020660 -2022-12-06 11:19:22,526 - Epoch: [127][ 600/ 1200] Overall Loss 0.189187 Objective Loss 0.189187 LR 0.000500 Time 0.020635 -2022-12-06 11:19:22,720 - Epoch: [127][ 610/ 1200] Overall Loss 0.188943 Objective Loss 0.188943 LR 0.000500 Time 0.020615 -2022-12-06 11:19:22,914 - Epoch: [127][ 620/ 1200] Overall Loss 0.189077 Objective Loss 0.189077 LR 0.000500 Time 0.020594 -2022-12-06 11:19:23,108 - Epoch: [127][ 630/ 1200] Overall Loss 0.188954 Objective Loss 0.188954 LR 0.000500 Time 0.020575 -2022-12-06 11:19:23,302 - Epoch: [127][ 640/ 1200] Overall Loss 0.188932 Objective Loss 0.188932 LR 0.000500 Time 0.020556 -2022-12-06 11:19:23,497 - Epoch: [127][ 650/ 1200] Overall Loss 0.188898 Objective Loss 0.188898 LR 0.000500 Time 0.020538 -2022-12-06 11:19:23,691 - Epoch: [127][ 660/ 1200] Overall Loss 0.189167 Objective Loss 0.189167 LR 0.000500 Time 0.020520 -2022-12-06 11:19:23,886 - Epoch: [127][ 670/ 1200] Overall Loss 0.189384 Objective Loss 0.189384 LR 0.000500 Time 0.020504 -2022-12-06 11:19:24,079 - Epoch: [127][ 680/ 1200] Overall Loss 0.189208 Objective Loss 0.189208 LR 0.000500 Time 0.020486 -2022-12-06 11:19:24,274 - Epoch: [127][ 690/ 1200] Overall Loss 0.189461 Objective Loss 0.189461 LR 0.000500 Time 0.020470 -2022-12-06 11:19:24,467 - Epoch: [127][ 700/ 1200] Overall Loss 0.189674 Objective Loss 0.189674 LR 0.000500 Time 0.020454 -2022-12-06 11:19:24,662 - Epoch: [127][ 710/ 1200] Overall Loss 0.189486 Objective Loss 0.189486 LR 0.000500 Time 0.020438 -2022-12-06 11:19:24,856 - Epoch: [127][ 720/ 1200] Overall Loss 0.189555 Objective Loss 0.189555 LR 0.000500 Time 0.020423 -2022-12-06 11:19:25,050 - Epoch: [127][ 730/ 1200] Overall Loss 0.189574 Objective Loss 0.189574 LR 0.000500 Time 0.020409 -2022-12-06 11:19:25,244 - Epoch: [127][ 740/ 1200] Overall Loss 0.189620 Objective Loss 0.189620 LR 0.000500 Time 0.020394 -2022-12-06 11:19:25,438 - Epoch: [127][ 750/ 1200] Overall Loss 0.189762 Objective Loss 0.189762 LR 0.000500 Time 0.020380 -2022-12-06 11:19:25,631 - Epoch: [127][ 760/ 1200] Overall Loss 0.189502 Objective Loss 0.189502 LR 0.000500 Time 0.020366 -2022-12-06 11:19:25,826 - Epoch: [127][ 770/ 1200] Overall Loss 0.189332 Objective Loss 0.189332 LR 0.000500 Time 0.020353 -2022-12-06 11:19:26,020 - Epoch: [127][ 780/ 1200] Overall Loss 0.189029 Objective Loss 0.189029 LR 0.000500 Time 0.020340 -2022-12-06 11:19:26,215 - Epoch: [127][ 790/ 1200] Overall Loss 0.189141 Objective Loss 0.189141 LR 0.000500 Time 0.020329 -2022-12-06 11:19:26,408 - Epoch: [127][ 800/ 1200] Overall Loss 0.189219 Objective Loss 0.189219 LR 0.000500 Time 0.020316 -2022-12-06 11:19:26,603 - Epoch: [127][ 810/ 1200] Overall Loss 0.189303 Objective Loss 0.189303 LR 0.000500 Time 0.020305 -2022-12-06 11:19:26,796 - Epoch: [127][ 820/ 1200] Overall Loss 0.189298 Objective Loss 0.189298 LR 0.000500 Time 0.020292 -2022-12-06 11:19:26,991 - Epoch: [127][ 830/ 1200] Overall Loss 0.189331 Objective Loss 0.189331 LR 0.000500 Time 0.020282 -2022-12-06 11:19:27,184 - Epoch: [127][ 840/ 1200] Overall Loss 0.189566 Objective Loss 0.189566 LR 0.000500 Time 0.020270 -2022-12-06 11:19:27,380 - Epoch: [127][ 850/ 1200] Overall Loss 0.189912 Objective Loss 0.189912 LR 0.000500 Time 0.020261 -2022-12-06 11:19:27,573 - Epoch: [127][ 860/ 1200] Overall Loss 0.189846 Objective Loss 0.189846 LR 0.000500 Time 0.020249 -2022-12-06 11:19:27,767 - Epoch: [127][ 870/ 1200] Overall Loss 0.190040 Objective Loss 0.190040 LR 0.000500 Time 0.020239 -2022-12-06 11:19:27,961 - Epoch: [127][ 880/ 1200] Overall Loss 0.190076 Objective Loss 0.190076 LR 0.000500 Time 0.020229 -2022-12-06 11:19:28,156 - Epoch: [127][ 890/ 1200] Overall Loss 0.190189 Objective Loss 0.190189 LR 0.000500 Time 0.020219 -2022-12-06 11:19:28,349 - Epoch: [127][ 900/ 1200] Overall Loss 0.190405 Objective Loss 0.190405 LR 0.000500 Time 0.020209 -2022-12-06 11:19:28,543 - Epoch: [127][ 910/ 1200] Overall Loss 0.190589 Objective Loss 0.190589 LR 0.000500 Time 0.020199 -2022-12-06 11:19:28,736 - Epoch: [127][ 920/ 1200] Overall Loss 0.190852 Objective Loss 0.190852 LR 0.000500 Time 0.020189 -2022-12-06 11:19:28,931 - Epoch: [127][ 930/ 1200] Overall Loss 0.190886 Objective Loss 0.190886 LR 0.000500 Time 0.020181 -2022-12-06 11:19:29,124 - Epoch: [127][ 940/ 1200] Overall Loss 0.190912 Objective Loss 0.190912 LR 0.000500 Time 0.020171 -2022-12-06 11:19:29,319 - Epoch: [127][ 950/ 1200] Overall Loss 0.190638 Objective Loss 0.190638 LR 0.000500 Time 0.020164 -2022-12-06 11:19:29,513 - Epoch: [127][ 960/ 1200] Overall Loss 0.190776 Objective Loss 0.190776 LR 0.000500 Time 0.020155 -2022-12-06 11:19:29,708 - Epoch: [127][ 970/ 1200] Overall Loss 0.190686 Objective Loss 0.190686 LR 0.000500 Time 0.020147 -2022-12-06 11:19:29,902 - Epoch: [127][ 980/ 1200] Overall Loss 0.190860 Objective Loss 0.190860 LR 0.000500 Time 0.020140 -2022-12-06 11:19:30,098 - Epoch: [127][ 990/ 1200] Overall Loss 0.190989 Objective Loss 0.190989 LR 0.000500 Time 0.020134 -2022-12-06 11:19:30,292 - Epoch: [127][ 1000/ 1200] Overall Loss 0.190982 Objective Loss 0.190982 LR 0.000500 Time 0.020125 -2022-12-06 11:19:30,486 - Epoch: [127][ 1010/ 1200] Overall Loss 0.190830 Objective Loss 0.190830 LR 0.000500 Time 0.020118 -2022-12-06 11:19:30,680 - Epoch: [127][ 1020/ 1200] Overall Loss 0.191045 Objective Loss 0.191045 LR 0.000500 Time 0.020110 -2022-12-06 11:19:30,875 - Epoch: [127][ 1030/ 1200] Overall Loss 0.191361 Objective Loss 0.191361 LR 0.000500 Time 0.020104 -2022-12-06 11:19:31,069 - Epoch: [127][ 1040/ 1200] Overall Loss 0.191293 Objective Loss 0.191293 LR 0.000500 Time 0.020096 -2022-12-06 11:19:31,264 - Epoch: [127][ 1050/ 1200] Overall Loss 0.191550 Objective Loss 0.191550 LR 0.000500 Time 0.020090 -2022-12-06 11:19:31,457 - Epoch: [127][ 1060/ 1200] Overall Loss 0.191257 Objective Loss 0.191257 LR 0.000500 Time 0.020082 -2022-12-06 11:19:31,652 - Epoch: [127][ 1070/ 1200] Overall Loss 0.191092 Objective Loss 0.191092 LR 0.000500 Time 0.020076 -2022-12-06 11:19:31,846 - Epoch: [127][ 1080/ 1200] Overall Loss 0.190870 Objective Loss 0.190870 LR 0.000500 Time 0.020069 -2022-12-06 11:19:32,040 - Epoch: [127][ 1090/ 1200] Overall Loss 0.191060 Objective Loss 0.191060 LR 0.000500 Time 0.020063 -2022-12-06 11:19:32,234 - Epoch: [127][ 1100/ 1200] Overall Loss 0.190964 Objective Loss 0.190964 LR 0.000500 Time 0.020056 -2022-12-06 11:19:32,430 - Epoch: [127][ 1110/ 1200] Overall Loss 0.190892 Objective Loss 0.190892 LR 0.000500 Time 0.020051 -2022-12-06 11:19:32,624 - Epoch: [127][ 1120/ 1200] Overall Loss 0.191051 Objective Loss 0.191051 LR 0.000500 Time 0.020045 -2022-12-06 11:19:32,818 - Epoch: [127][ 1130/ 1200] Overall Loss 0.191061 Objective Loss 0.191061 LR 0.000500 Time 0.020039 -2022-12-06 11:19:33,012 - Epoch: [127][ 1140/ 1200] Overall Loss 0.191093 Objective Loss 0.191093 LR 0.000500 Time 0.020033 -2022-12-06 11:19:33,205 - Epoch: [127][ 1150/ 1200] Overall Loss 0.191186 Objective Loss 0.191186 LR 0.000500 Time 0.020026 -2022-12-06 11:19:33,398 - Epoch: [127][ 1160/ 1200] Overall Loss 0.191213 Objective Loss 0.191213 LR 0.000500 Time 0.020020 -2022-12-06 11:19:33,592 - Epoch: [127][ 1170/ 1200] Overall Loss 0.191220 Objective Loss 0.191220 LR 0.000500 Time 0.020014 -2022-12-06 11:19:33,786 - Epoch: [127][ 1180/ 1200] Overall Loss 0.191299 Objective Loss 0.191299 LR 0.000500 Time 0.020008 -2022-12-06 11:19:33,980 - Epoch: [127][ 1190/ 1200] Overall Loss 0.191288 Objective Loss 0.191288 LR 0.000500 Time 0.020003 -2022-12-06 11:19:34,205 - Epoch: [127][ 1200/ 1200] Overall Loss 0.191412 Objective Loss 0.191412 Top1 87.029289 Top5 99.163180 LR 0.000500 Time 0.020023 -2022-12-06 11:19:34,294 - --- validate (epoch=127)----------- -2022-12-06 11:19:34,294 - 34129 samples (256 per mini-batch) -2022-12-06 11:19:34,757 - Epoch: [127][ 10/ 134] Loss 0.239142 Top1 86.953125 Top5 98.476562 -2022-12-06 11:19:34,894 - Epoch: [127][ 20/ 134] Loss 0.247793 Top1 86.953125 Top5 98.457031 -2022-12-06 11:19:35,027 - Epoch: [127][ 30/ 134] Loss 0.249656 Top1 87.122396 Top5 98.515625 -2022-12-06 11:19:35,159 - Epoch: [127][ 40/ 134] Loss 0.249669 Top1 87.167969 Top5 98.496094 -2022-12-06 11:19:35,292 - Epoch: [127][ 50/ 134] Loss 0.251709 Top1 87.203125 Top5 98.539062 -2022-12-06 11:19:35,425 - Epoch: [127][ 60/ 134] Loss 0.253749 Top1 86.979167 Top5 98.463542 -2022-12-06 11:19:35,558 - Epoch: [127][ 70/ 134] Loss 0.258699 Top1 86.785714 Top5 98.415179 -2022-12-06 11:19:35,689 - Epoch: [127][ 80/ 134] Loss 0.257269 Top1 86.884766 Top5 98.413086 -2022-12-06 11:19:35,822 - Epoch: [127][ 90/ 134] Loss 0.259225 Top1 86.792535 Top5 98.398438 -2022-12-06 11:19:35,955 - Epoch: [127][ 100/ 134] Loss 0.258664 Top1 86.746094 Top5 98.421875 -2022-12-06 11:19:36,088 - Epoch: [127][ 110/ 134] Loss 0.257563 Top1 86.740057 Top5 98.380682 -2022-12-06 11:19:36,218 - Epoch: [127][ 120/ 134] Loss 0.255774 Top1 86.783854 Top5 98.401693 -2022-12-06 11:19:36,353 - Epoch: [127][ 130/ 134] Loss 0.254861 Top1 86.841947 Top5 98.416466 -2022-12-06 11:19:36,391 - Epoch: [127][ 134/ 134] Loss 0.254464 Top1 86.852823 Top5 98.403118 -2022-12-06 11:19:36,479 - ==> Top1: 86.853 Top5: 98.403 Loss: 0.254 - -2022-12-06 11:19:36,479 - ==> Confusion: -[[ 903 2 0 3 2 4 0 0 5 59 0 1 1 4 5 1 1 0 1 0 4] - [ 2 927 2 2 11 22 3 13 0 1 3 3 2 0 5 1 4 1 11 2 12] - [ 6 1 1012 12 5 0 12 8 0 2 6 4 3 4 6 3 0 1 4 2 12] - [ 1 1 15 947 2 4 0 0 0 0 8 0 3 3 17 0 0 4 10 0 5] - [ 6 4 1 0 959 4 1 0 1 6 1 2 2 1 17 8 4 1 0 0 2] - [ 5 16 0 1 8 957 2 16 2 4 2 14 6 15 2 1 2 0 1 8 7] - [ 0 3 14 4 1 1 1059 2 1 0 3 1 1 2 0 8 1 1 1 14 1] - [ 1 10 7 2 2 25 12 942 0 0 0 7 1 2 0 0 1 1 17 17 7] - [ 3 0 1 0 2 3 0 1 975 38 7 2 1 12 10 0 2 2 2 1 2] - [ 51 0 2 0 5 2 0 3 21 893 2 1 0 7 3 2 0 1 0 0 8] - [ 1 0 2 5 2 0 1 4 9 1 957 4 3 13 5 1 1 0 3 1 6] - [ 2 0 0 0 0 6 4 2 0 0 1 983 26 3 0 2 3 3 0 13 3] - [ 0 0 1 2 1 2 0 1 0 0 0 36 900 1 0 6 2 9 1 4 3] - [ 0 1 0 0 1 3 0 2 8 15 4 6 4 958 2 2 4 1 0 3 9] - [ 7 3 2 10 2 0 0 0 12 7 1 3 1 1 1069 0 1 1 7 1 2] - [ 1 0 2 0 0 0 3 0 0 0 2 5 11 2 0 995 3 12 0 3 4] - [ 2 1 2 0 4 1 2 0 0 0 1 2 1 2 0 12 1031 0 0 4 7] - [ 1 0 2 1 0 0 1 0 1 2 0 13 17 2 3 10 2 976 0 5 0] - [ 5 2 7 13 1 3 0 15 2 1 3 1 2 0 8 2 1 0 935 2 5] - [ 1 3 1 1 1 4 5 3 2 0 2 15 7 5 0 2 5 1 1 1017 4] - [ 128 137 190 123 126 136 62 119 66 89 130 106 347 268 168 109 149 99 156 280 10238]] - -2022-12-06 11:19:37,060 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:19:37,060 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:19:37,067 - - -2022-12-06 11:19:37,067 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:19:38,007 - Epoch: [128][ 10/ 1200] Overall Loss 0.167719 Objective Loss 0.167719 LR 0.000500 Time 0.093910 -2022-12-06 11:19:38,202 - Epoch: [128][ 20/ 1200] Overall Loss 0.183493 Objective Loss 0.183493 LR 0.000500 Time 0.056678 -2022-12-06 11:19:38,396 - Epoch: [128][ 30/ 1200] Overall Loss 0.187301 Objective Loss 0.187301 LR 0.000500 Time 0.044240 -2022-12-06 11:19:38,588 - Epoch: [128][ 40/ 1200] Overall Loss 0.179653 Objective Loss 0.179653 LR 0.000500 Time 0.037969 -2022-12-06 11:19:38,781 - Epoch: [128][ 50/ 1200] Overall Loss 0.178838 Objective Loss 0.178838 LR 0.000500 Time 0.034226 -2022-12-06 11:19:38,974 - Epoch: [128][ 60/ 1200] Overall Loss 0.180684 Objective Loss 0.180684 LR 0.000500 Time 0.031721 -2022-12-06 11:19:39,167 - Epoch: [128][ 70/ 1200] Overall Loss 0.180567 Objective Loss 0.180567 LR 0.000500 Time 0.029936 -2022-12-06 11:19:39,358 - Epoch: [128][ 80/ 1200] Overall Loss 0.183214 Objective Loss 0.183214 LR 0.000500 Time 0.028582 -2022-12-06 11:19:39,551 - Epoch: [128][ 90/ 1200] Overall Loss 0.185898 Objective Loss 0.185898 LR 0.000500 Time 0.027543 -2022-12-06 11:19:39,743 - Epoch: [128][ 100/ 1200] Overall Loss 0.187432 Objective Loss 0.187432 LR 0.000500 Time 0.026706 -2022-12-06 11:19:39,936 - Epoch: [128][ 110/ 1200] Overall Loss 0.186127 Objective Loss 0.186127 LR 0.000500 Time 0.026027 -2022-12-06 11:19:40,128 - Epoch: [128][ 120/ 1200] Overall Loss 0.186052 Objective Loss 0.186052 LR 0.000500 Time 0.025453 -2022-12-06 11:19:40,321 - Epoch: [128][ 130/ 1200] Overall Loss 0.186173 Objective Loss 0.186173 LR 0.000500 Time 0.024971 -2022-12-06 11:19:40,512 - Epoch: [128][ 140/ 1200] Overall Loss 0.184233 Objective Loss 0.184233 LR 0.000500 Time 0.024553 -2022-12-06 11:19:40,704 - Epoch: [128][ 150/ 1200] Overall Loss 0.184445 Objective Loss 0.184445 LR 0.000500 Time 0.024193 -2022-12-06 11:19:40,897 - Epoch: [128][ 160/ 1200] Overall Loss 0.185539 Objective Loss 0.185539 LR 0.000500 Time 0.023882 -2022-12-06 11:19:41,089 - Epoch: [128][ 170/ 1200] Overall Loss 0.186831 Objective Loss 0.186831 LR 0.000500 Time 0.023603 -2022-12-06 11:19:41,281 - Epoch: [128][ 180/ 1200] Overall Loss 0.186613 Objective Loss 0.186613 LR 0.000500 Time 0.023356 -2022-12-06 11:19:41,473 - Epoch: [128][ 190/ 1200] Overall Loss 0.187016 Objective Loss 0.187016 LR 0.000500 Time 0.023134 -2022-12-06 11:19:41,666 - Epoch: [128][ 200/ 1200] Overall Loss 0.187332 Objective Loss 0.187332 LR 0.000500 Time 0.022937 -2022-12-06 11:19:41,858 - Epoch: [128][ 210/ 1200] Overall Loss 0.187296 Objective Loss 0.187296 LR 0.000500 Time 0.022759 -2022-12-06 11:19:42,051 - Epoch: [128][ 220/ 1200] Overall Loss 0.186328 Objective Loss 0.186328 LR 0.000500 Time 0.022598 -2022-12-06 11:19:42,242 - Epoch: [128][ 230/ 1200] Overall Loss 0.186304 Objective Loss 0.186304 LR 0.000500 Time 0.022446 -2022-12-06 11:19:42,435 - Epoch: [128][ 240/ 1200] Overall Loss 0.187299 Objective Loss 0.187299 LR 0.000500 Time 0.022311 -2022-12-06 11:19:42,627 - Epoch: [128][ 250/ 1200] Overall Loss 0.187426 Objective Loss 0.187426 LR 0.000500 Time 0.022184 -2022-12-06 11:19:42,820 - Epoch: [128][ 260/ 1200] Overall Loss 0.186919 Objective Loss 0.186919 LR 0.000500 Time 0.022070 -2022-12-06 11:19:43,012 - Epoch: [128][ 270/ 1200] Overall Loss 0.186980 Objective Loss 0.186980 LR 0.000500 Time 0.021963 -2022-12-06 11:19:43,204 - Epoch: [128][ 280/ 1200] Overall Loss 0.186806 Objective Loss 0.186806 LR 0.000500 Time 0.021861 -2022-12-06 11:19:43,396 - Epoch: [128][ 290/ 1200] Overall Loss 0.187143 Objective Loss 0.187143 LR 0.000500 Time 0.021768 -2022-12-06 11:19:43,588 - Epoch: [128][ 300/ 1200] Overall Loss 0.187206 Objective Loss 0.187206 LR 0.000500 Time 0.021680 -2022-12-06 11:19:43,780 - Epoch: [128][ 310/ 1200] Overall Loss 0.186865 Objective Loss 0.186865 LR 0.000500 Time 0.021600 -2022-12-06 11:19:43,972 - Epoch: [128][ 320/ 1200] Overall Loss 0.186091 Objective Loss 0.186091 LR 0.000500 Time 0.021522 -2022-12-06 11:19:44,164 - Epoch: [128][ 330/ 1200] Overall Loss 0.186264 Objective Loss 0.186264 LR 0.000500 Time 0.021450 -2022-12-06 11:19:44,357 - Epoch: [128][ 340/ 1200] Overall Loss 0.186125 Objective Loss 0.186125 LR 0.000500 Time 0.021387 -2022-12-06 11:19:44,549 - Epoch: [128][ 350/ 1200] Overall Loss 0.186487 Objective Loss 0.186487 LR 0.000500 Time 0.021323 -2022-12-06 11:19:44,741 - Epoch: [128][ 360/ 1200] Overall Loss 0.186812 Objective Loss 0.186812 LR 0.000500 Time 0.021262 -2022-12-06 11:19:44,934 - Epoch: [128][ 370/ 1200] Overall Loss 0.187340 Objective Loss 0.187340 LR 0.000500 Time 0.021205 -2022-12-06 11:19:45,126 - Epoch: [128][ 380/ 1200] Overall Loss 0.188056 Objective Loss 0.188056 LR 0.000500 Time 0.021151 -2022-12-06 11:19:45,318 - Epoch: [128][ 390/ 1200] Overall Loss 0.188036 Objective Loss 0.188036 LR 0.000500 Time 0.021100 -2022-12-06 11:19:45,510 - Epoch: [128][ 400/ 1200] Overall Loss 0.188245 Objective Loss 0.188245 LR 0.000500 Time 0.021051 -2022-12-06 11:19:45,703 - Epoch: [128][ 410/ 1200] Overall Loss 0.188100 Objective Loss 0.188100 LR 0.000500 Time 0.021007 -2022-12-06 11:19:45,895 - Epoch: [128][ 420/ 1200] Overall Loss 0.188265 Objective Loss 0.188265 LR 0.000500 Time 0.020963 -2022-12-06 11:19:46,087 - Epoch: [128][ 430/ 1200] Overall Loss 0.188580 Objective Loss 0.188580 LR 0.000500 Time 0.020921 -2022-12-06 11:19:46,280 - Epoch: [128][ 440/ 1200] Overall Loss 0.188021 Objective Loss 0.188021 LR 0.000500 Time 0.020883 -2022-12-06 11:19:46,472 - Epoch: [128][ 450/ 1200] Overall Loss 0.188288 Objective Loss 0.188288 LR 0.000500 Time 0.020845 -2022-12-06 11:19:46,664 - Epoch: [128][ 460/ 1200] Overall Loss 0.188671 Objective Loss 0.188671 LR 0.000500 Time 0.020807 -2022-12-06 11:19:46,856 - Epoch: [128][ 470/ 1200] Overall Loss 0.188541 Objective Loss 0.188541 LR 0.000500 Time 0.020773 -2022-12-06 11:19:47,048 - Epoch: [128][ 480/ 1200] Overall Loss 0.188271 Objective Loss 0.188271 LR 0.000500 Time 0.020739 -2022-12-06 11:19:47,242 - Epoch: [128][ 490/ 1200] Overall Loss 0.187975 Objective Loss 0.187975 LR 0.000500 Time 0.020710 -2022-12-06 11:19:47,434 - Epoch: [128][ 500/ 1200] Overall Loss 0.187782 Objective Loss 0.187782 LR 0.000500 Time 0.020678 -2022-12-06 11:19:47,627 - Epoch: [128][ 510/ 1200] Overall Loss 0.188090 Objective Loss 0.188090 LR 0.000500 Time 0.020650 -2022-12-06 11:19:47,819 - Epoch: [128][ 520/ 1200] Overall Loss 0.188025 Objective Loss 0.188025 LR 0.000500 Time 0.020622 -2022-12-06 11:19:48,012 - Epoch: [128][ 530/ 1200] Overall Loss 0.187827 Objective Loss 0.187827 LR 0.000500 Time 0.020595 -2022-12-06 11:19:48,205 - Epoch: [128][ 540/ 1200] Overall Loss 0.187990 Objective Loss 0.187990 LR 0.000500 Time 0.020570 -2022-12-06 11:19:48,397 - Epoch: [128][ 550/ 1200] Overall Loss 0.188029 Objective Loss 0.188029 LR 0.000500 Time 0.020545 -2022-12-06 11:19:48,590 - Epoch: [128][ 560/ 1200] Overall Loss 0.187761 Objective Loss 0.187761 LR 0.000500 Time 0.020520 -2022-12-06 11:19:48,782 - Epoch: [128][ 570/ 1200] Overall Loss 0.187708 Objective Loss 0.187708 LR 0.000500 Time 0.020496 -2022-12-06 11:19:48,974 - Epoch: [128][ 580/ 1200] Overall Loss 0.188092 Objective Loss 0.188092 LR 0.000500 Time 0.020474 -2022-12-06 11:19:49,167 - Epoch: [128][ 590/ 1200] Overall Loss 0.187968 Objective Loss 0.187968 LR 0.000500 Time 0.020452 -2022-12-06 11:19:49,359 - Epoch: [128][ 600/ 1200] Overall Loss 0.188284 Objective Loss 0.188284 LR 0.000500 Time 0.020431 -2022-12-06 11:19:49,551 - Epoch: [128][ 610/ 1200] Overall Loss 0.188331 Objective Loss 0.188331 LR 0.000500 Time 0.020410 -2022-12-06 11:19:49,743 - Epoch: [128][ 620/ 1200] Overall Loss 0.188562 Objective Loss 0.188562 LR 0.000500 Time 0.020390 -2022-12-06 11:19:49,936 - Epoch: [128][ 630/ 1200] Overall Loss 0.188630 Objective Loss 0.188630 LR 0.000500 Time 0.020371 -2022-12-06 11:19:50,128 - Epoch: [128][ 640/ 1200] Overall Loss 0.188504 Objective Loss 0.188504 LR 0.000500 Time 0.020352 -2022-12-06 11:19:50,320 - Epoch: [128][ 650/ 1200] Overall Loss 0.188833 Objective Loss 0.188833 LR 0.000500 Time 0.020334 -2022-12-06 11:19:50,512 - Epoch: [128][ 660/ 1200] Overall Loss 0.189088 Objective Loss 0.189088 LR 0.000500 Time 0.020316 -2022-12-06 11:19:50,704 - Epoch: [128][ 670/ 1200] Overall Loss 0.189119 Objective Loss 0.189119 LR 0.000500 Time 0.020298 -2022-12-06 11:19:50,896 - Epoch: [128][ 680/ 1200] Overall Loss 0.188898 Objective Loss 0.188898 LR 0.000500 Time 0.020282 -2022-12-06 11:19:51,088 - Epoch: [128][ 690/ 1200] Overall Loss 0.188835 Objective Loss 0.188835 LR 0.000500 Time 0.020265 -2022-12-06 11:19:51,282 - Epoch: [128][ 700/ 1200] Overall Loss 0.189152 Objective Loss 0.189152 LR 0.000500 Time 0.020252 -2022-12-06 11:19:51,475 - Epoch: [128][ 710/ 1200] Overall Loss 0.189225 Objective Loss 0.189225 LR 0.000500 Time 0.020237 -2022-12-06 11:19:51,667 - Epoch: [128][ 720/ 1200] Overall Loss 0.189166 Objective Loss 0.189166 LR 0.000500 Time 0.020222 -2022-12-06 11:19:51,860 - Epoch: [128][ 730/ 1200] Overall Loss 0.188823 Objective Loss 0.188823 LR 0.000500 Time 0.020208 -2022-12-06 11:19:52,053 - Epoch: [128][ 740/ 1200] Overall Loss 0.188792 Objective Loss 0.188792 LR 0.000500 Time 0.020196 -2022-12-06 11:19:52,246 - Epoch: [128][ 750/ 1200] Overall Loss 0.188650 Objective Loss 0.188650 LR 0.000500 Time 0.020183 -2022-12-06 11:19:52,438 - Epoch: [128][ 760/ 1200] Overall Loss 0.188643 Objective Loss 0.188643 LR 0.000500 Time 0.020170 -2022-12-06 11:19:52,631 - Epoch: [128][ 770/ 1200] Overall Loss 0.188687 Objective Loss 0.188687 LR 0.000500 Time 0.020157 -2022-12-06 11:19:52,823 - Epoch: [128][ 780/ 1200] Overall Loss 0.188768 Objective Loss 0.188768 LR 0.000500 Time 0.020144 -2022-12-06 11:19:53,015 - Epoch: [128][ 790/ 1200] Overall Loss 0.188616 Objective Loss 0.188616 LR 0.000500 Time 0.020132 -2022-12-06 11:19:53,207 - Epoch: [128][ 800/ 1200] Overall Loss 0.189033 Objective Loss 0.189033 LR 0.000500 Time 0.020120 -2022-12-06 11:19:53,400 - Epoch: [128][ 810/ 1200] Overall Loss 0.189140 Objective Loss 0.189140 LR 0.000500 Time 0.020109 -2022-12-06 11:19:53,593 - Epoch: [128][ 820/ 1200] Overall Loss 0.189281 Objective Loss 0.189281 LR 0.000500 Time 0.020098 -2022-12-06 11:19:53,786 - Epoch: [128][ 830/ 1200] Overall Loss 0.189197 Objective Loss 0.189197 LR 0.000500 Time 0.020087 -2022-12-06 11:19:53,978 - Epoch: [128][ 840/ 1200] Overall Loss 0.189320 Objective Loss 0.189320 LR 0.000500 Time 0.020077 -2022-12-06 11:19:54,171 - Epoch: [128][ 850/ 1200] Overall Loss 0.189261 Objective Loss 0.189261 LR 0.000500 Time 0.020066 -2022-12-06 11:19:54,363 - Epoch: [128][ 860/ 1200] Overall Loss 0.189575 Objective Loss 0.189575 LR 0.000500 Time 0.020057 -2022-12-06 11:19:54,557 - Epoch: [128][ 870/ 1200] Overall Loss 0.189598 Objective Loss 0.189598 LR 0.000500 Time 0.020048 -2022-12-06 11:19:54,749 - Epoch: [128][ 880/ 1200] Overall Loss 0.189477 Objective Loss 0.189477 LR 0.000500 Time 0.020038 -2022-12-06 11:19:54,942 - Epoch: [128][ 890/ 1200] Overall Loss 0.189323 Objective Loss 0.189323 LR 0.000500 Time 0.020029 -2022-12-06 11:19:55,134 - Epoch: [128][ 900/ 1200] Overall Loss 0.189379 Objective Loss 0.189379 LR 0.000500 Time 0.020019 -2022-12-06 11:19:55,326 - Epoch: [128][ 910/ 1200] Overall Loss 0.189503 Objective Loss 0.189503 LR 0.000500 Time 0.020010 -2022-12-06 11:19:55,518 - Epoch: [128][ 920/ 1200] Overall Loss 0.189555 Objective Loss 0.189555 LR 0.000500 Time 0.020000 -2022-12-06 11:19:55,712 - Epoch: [128][ 930/ 1200] Overall Loss 0.189899 Objective Loss 0.189899 LR 0.000500 Time 0.019992 -2022-12-06 11:19:55,904 - Epoch: [128][ 940/ 1200] Overall Loss 0.189991 Objective Loss 0.189991 LR 0.000500 Time 0.019984 -2022-12-06 11:19:56,096 - Epoch: [128][ 950/ 1200] Overall Loss 0.190113 Objective Loss 0.190113 LR 0.000500 Time 0.019975 -2022-12-06 11:19:56,289 - Epoch: [128][ 960/ 1200] Overall Loss 0.189902 Objective Loss 0.189902 LR 0.000500 Time 0.019967 -2022-12-06 11:19:56,481 - Epoch: [128][ 970/ 1200] Overall Loss 0.189780 Objective Loss 0.189780 LR 0.000500 Time 0.019959 -2022-12-06 11:19:56,673 - Epoch: [128][ 980/ 1200] Overall Loss 0.189707 Objective Loss 0.189707 LR 0.000500 Time 0.019951 -2022-12-06 11:19:56,866 - Epoch: [128][ 990/ 1200] Overall Loss 0.189604 Objective Loss 0.189604 LR 0.000500 Time 0.019943 -2022-12-06 11:19:57,058 - Epoch: [128][ 1000/ 1200] Overall Loss 0.189601 Objective Loss 0.189601 LR 0.000500 Time 0.019936 -2022-12-06 11:19:57,250 - Epoch: [128][ 1010/ 1200] Overall Loss 0.189407 Objective Loss 0.189407 LR 0.000500 Time 0.019928 -2022-12-06 11:19:57,442 - Epoch: [128][ 1020/ 1200] Overall Loss 0.189430 Objective Loss 0.189430 LR 0.000500 Time 0.019920 -2022-12-06 11:19:57,635 - Epoch: [128][ 1030/ 1200] Overall Loss 0.189191 Objective Loss 0.189191 LR 0.000500 Time 0.019913 -2022-12-06 11:19:57,827 - Epoch: [128][ 1040/ 1200] Overall Loss 0.189311 Objective Loss 0.189311 LR 0.000500 Time 0.019906 -2022-12-06 11:19:58,019 - Epoch: [128][ 1050/ 1200] Overall Loss 0.189514 Objective Loss 0.189514 LR 0.000500 Time 0.019898 -2022-12-06 11:19:58,211 - Epoch: [128][ 1060/ 1200] Overall Loss 0.189366 Objective Loss 0.189366 LR 0.000500 Time 0.019891 -2022-12-06 11:19:58,404 - Epoch: [128][ 1070/ 1200] Overall Loss 0.189503 Objective Loss 0.189503 LR 0.000500 Time 0.019885 -2022-12-06 11:19:58,596 - Epoch: [128][ 1080/ 1200] Overall Loss 0.189496 Objective Loss 0.189496 LR 0.000500 Time 0.019878 -2022-12-06 11:19:58,788 - Epoch: [128][ 1090/ 1200] Overall Loss 0.189495 Objective Loss 0.189495 LR 0.000500 Time 0.019872 -2022-12-06 11:19:58,981 - Epoch: [128][ 1100/ 1200] Overall Loss 0.189557 Objective Loss 0.189557 LR 0.000500 Time 0.019866 -2022-12-06 11:19:59,173 - Epoch: [128][ 1110/ 1200] Overall Loss 0.189361 Objective Loss 0.189361 LR 0.000500 Time 0.019860 -2022-12-06 11:19:59,365 - Epoch: [128][ 1120/ 1200] Overall Loss 0.189373 Objective Loss 0.189373 LR 0.000500 Time 0.019854 -2022-12-06 11:19:59,558 - Epoch: [128][ 1130/ 1200] Overall Loss 0.189277 Objective Loss 0.189277 LR 0.000500 Time 0.019848 -2022-12-06 11:19:59,751 - Epoch: [128][ 1140/ 1200] Overall Loss 0.189376 Objective Loss 0.189376 LR 0.000500 Time 0.019842 -2022-12-06 11:19:59,943 - Epoch: [128][ 1150/ 1200] Overall Loss 0.189421 Objective Loss 0.189421 LR 0.000500 Time 0.019837 -2022-12-06 11:20:00,135 - Epoch: [128][ 1160/ 1200] Overall Loss 0.189573 Objective Loss 0.189573 LR 0.000500 Time 0.019831 -2022-12-06 11:20:00,328 - Epoch: [128][ 1170/ 1200] Overall Loss 0.189728 Objective Loss 0.189728 LR 0.000500 Time 0.019826 -2022-12-06 11:20:00,520 - Epoch: [128][ 1180/ 1200] Overall Loss 0.189902 Objective Loss 0.189902 LR 0.000500 Time 0.019820 -2022-12-06 11:20:00,712 - Epoch: [128][ 1190/ 1200] Overall Loss 0.189875 Objective Loss 0.189875 LR 0.000500 Time 0.019815 -2022-12-06 11:20:00,938 - Epoch: [128][ 1200/ 1200] Overall Loss 0.189916 Objective Loss 0.189916 Top1 87.447699 Top5 98.535565 LR 0.000500 Time 0.019837 -2022-12-06 11:20:01,027 - --- validate (epoch=128)----------- -2022-12-06 11:20:01,027 - 34129 samples (256 per mini-batch) -2022-12-06 11:20:01,484 - Epoch: [128][ 10/ 134] Loss 0.268515 Top1 86.406250 Top5 98.671875 -2022-12-06 11:20:01,619 - Epoch: [128][ 20/ 134] Loss 0.256658 Top1 86.542969 Top5 98.437500 -2022-12-06 11:20:01,753 - Epoch: [128][ 30/ 134] Loss 0.255978 Top1 86.406250 Top5 98.372396 -2022-12-06 11:20:01,886 - Epoch: [128][ 40/ 134] Loss 0.254237 Top1 86.425781 Top5 98.320312 -2022-12-06 11:20:02,022 - Epoch: [128][ 50/ 134] Loss 0.260692 Top1 86.296875 Top5 98.226562 -2022-12-06 11:20:02,156 - Epoch: [128][ 60/ 134] Loss 0.261772 Top1 86.360677 Top5 98.274740 -2022-12-06 11:20:02,290 - Epoch: [128][ 70/ 134] Loss 0.258958 Top1 86.367188 Top5 98.376116 -2022-12-06 11:20:02,423 - Epoch: [128][ 80/ 134] Loss 0.254240 Top1 86.479492 Top5 98.422852 -2022-12-06 11:20:02,558 - Epoch: [128][ 90/ 134] Loss 0.253794 Top1 86.545139 Top5 98.433160 -2022-12-06 11:20:02,691 - Epoch: [128][ 100/ 134] Loss 0.254964 Top1 86.609375 Top5 98.382812 -2022-12-06 11:20:02,825 - Epoch: [128][ 110/ 134] Loss 0.254389 Top1 86.651278 Top5 98.401989 -2022-12-06 11:20:02,959 - Epoch: [128][ 120/ 134] Loss 0.254118 Top1 86.656901 Top5 98.362630 -2022-12-06 11:20:03,093 - Epoch: [128][ 130/ 134] Loss 0.252961 Top1 86.676683 Top5 98.371394 -2022-12-06 11:20:03,131 - Epoch: [128][ 134/ 134] Loss 0.252397 Top1 86.694600 Top5 98.373817 -2022-12-06 11:20:03,222 - ==> Top1: 86.695 Top5: 98.374 Loss: 0.252 - -2022-12-06 11:20:03,222 - ==> Confusion: -[[ 877 1 1 1 9 5 0 0 4 80 0 2 1 5 5 1 2 0 0 0 2] - [ 0 951 5 2 8 17 3 10 1 0 2 4 1 2 0 1 1 1 12 2 4] - [ 4 3 1011 9 5 1 26 9 0 4 4 4 0 0 5 2 0 0 3 4 9] - [ 2 2 22 944 3 2 1 3 0 0 6 0 6 0 11 0 0 3 10 1 4] - [ 9 4 2 0 958 3 1 0 0 7 2 1 1 2 12 5 7 2 0 1 3] - [ 5 22 1 3 3 965 1 21 3 2 1 10 3 16 2 3 1 0 2 4 1] - [ 2 2 9 2 1 2 1077 3 0 0 2 3 0 1 0 4 0 2 2 5 1] - [ 2 10 8 3 2 22 7 949 0 1 0 4 0 2 0 2 0 1 24 11 6] - [ 4 2 0 0 1 1 1 1 958 51 11 0 1 7 17 0 1 1 3 2 2] - [ 27 0 2 0 4 3 0 4 14 926 1 1 0 8 5 2 0 1 0 1 2] - [ 1 1 3 5 1 0 1 2 10 2 953 2 2 14 5 1 1 0 8 1 6] - [ 2 0 0 0 0 16 4 4 0 0 0 976 23 4 1 3 3 5 0 6 4] - [ 0 0 2 4 0 2 0 0 0 0 0 21 909 2 1 7 1 9 0 6 5] - [ 2 1 1 0 1 5 0 2 9 12 4 3 3 969 0 0 2 1 0 3 5] - [ 4 1 2 5 3 2 0 0 11 4 0 2 1 4 1073 0 1 0 10 1 6] - [ 0 0 4 0 0 1 4 0 0 1 1 5 6 3 0 990 8 14 0 3 3] - [ 3 3 1 1 5 0 1 0 2 0 0 1 0 2 1 11 1029 0 0 4 8] - [ 4 0 1 2 1 1 2 0 1 4 0 9 17 4 3 13 0 970 2 1 1] - [ 2 7 7 6 4 2 0 20 2 1 4 0 3 1 9 0 1 0 935 1 3] - [ 3 2 1 1 2 3 6 8 0 0 2 14 6 6 0 4 4 1 1 1011 5] - [ 130 191 193 114 124 151 78 133 86 116 151 96 313 306 181 82 145 77 166 236 10157]] - -2022-12-06 11:20:03,897 - ==> Best [Top1: 87.040 Top5: 98.289 Sparsity:0.00 Params: 5376 on epoch: 121] -2022-12-06 11:20:03,897 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:20:03,903 - - -2022-12-06 11:20:03,903 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:20:04,839 - Epoch: [129][ 10/ 1200] Overall Loss 0.196007 Objective Loss 0.196007 LR 0.000500 Time 0.093547 -2022-12-06 11:20:05,038 - Epoch: [129][ 20/ 1200] Overall Loss 0.205179 Objective Loss 0.205179 LR 0.000500 Time 0.056662 -2022-12-06 11:20:05,232 - Epoch: [129][ 30/ 1200] Overall Loss 0.192500 Objective Loss 0.192500 LR 0.000500 Time 0.044219 -2022-12-06 11:20:05,425 - Epoch: [129][ 40/ 1200] Overall Loss 0.189751 Objective Loss 0.189751 LR 0.000500 Time 0.037991 -2022-12-06 11:20:05,618 - Epoch: [129][ 50/ 1200] Overall Loss 0.189205 Objective Loss 0.189205 LR 0.000500 Time 0.034234 -2022-12-06 11:20:05,811 - Epoch: [129][ 60/ 1200] Overall Loss 0.188200 Objective Loss 0.188200 LR 0.000500 Time 0.031733 -2022-12-06 11:20:06,005 - Epoch: [129][ 70/ 1200] Overall Loss 0.186303 Objective Loss 0.186303 LR 0.000500 Time 0.029965 -2022-12-06 11:20:06,200 - Epoch: [129][ 80/ 1200] Overall Loss 0.188391 Objective Loss 0.188391 LR 0.000500 Time 0.028656 -2022-12-06 11:20:06,395 - Epoch: [129][ 90/ 1200] Overall Loss 0.186764 Objective Loss 0.186764 LR 0.000500 Time 0.027621 -2022-12-06 11:20:06,589 - Epoch: [129][ 100/ 1200] Overall Loss 0.186121 Objective Loss 0.186121 LR 0.000500 Time 0.026800 -2022-12-06 11:20:06,783 - Epoch: [129][ 110/ 1200] Overall Loss 0.185269 Objective Loss 0.185269 LR 0.000500 Time 0.026123 -2022-12-06 11:20:06,978 - Epoch: [129][ 120/ 1200] Overall Loss 0.184513 Objective Loss 0.184513 LR 0.000500 Time 0.025565 -2022-12-06 11:20:07,172 - Epoch: [129][ 130/ 1200] Overall Loss 0.185393 Objective Loss 0.185393 LR 0.000500 Time 0.025083 -2022-12-06 11:20:07,367 - Epoch: [129][ 140/ 1200] Overall Loss 0.187856 Objective Loss 0.187856 LR 0.000500 Time 0.024681 -2022-12-06 11:20:07,561 - Epoch: [129][ 150/ 1200] Overall Loss 0.188593 Objective Loss 0.188593 LR 0.000500 Time 0.024324 -2022-12-06 11:20:07,754 - Epoch: [129][ 160/ 1200] Overall Loss 0.190359 Objective Loss 0.190359 LR 0.000500 Time 0.024007 -2022-12-06 11:20:07,947 - Epoch: [129][ 170/ 1200] Overall Loss 0.189590 Objective Loss 0.189590 LR 0.000500 Time 0.023726 -2022-12-06 11:20:08,140 - Epoch: [129][ 180/ 1200] Overall Loss 0.188616 Objective Loss 0.188616 LR 0.000500 Time 0.023479 -2022-12-06 11:20:08,334 - Epoch: [129][ 190/ 1200] Overall Loss 0.188878 Objective Loss 0.188878 LR 0.000500 Time 0.023260 -2022-12-06 11:20:08,528 - Epoch: [129][ 200/ 1200] Overall Loss 0.187900 Objective Loss 0.187900 LR 0.000500 Time 0.023066 -2022-12-06 11:20:08,722 - Epoch: [129][ 210/ 1200] Overall Loss 0.189340 Objective Loss 0.189340 LR 0.000500 Time 0.022891 -2022-12-06 11:20:08,917 - Epoch: [129][ 220/ 1200] Overall Loss 0.187939 Objective Loss 0.187939 LR 0.000500 Time 0.022731 -2022-12-06 11:20:09,111 - Epoch: [129][ 230/ 1200] Overall Loss 0.188309 Objective Loss 0.188309 LR 0.000500 Time 0.022586 -2022-12-06 11:20:09,304 - Epoch: [129][ 240/ 1200] Overall Loss 0.188741 Objective Loss 0.188741 LR 0.000500 Time 0.022447 -2022-12-06 11:20:09,496 - Epoch: [129][ 250/ 1200] Overall Loss 0.190039 Objective Loss 0.190039 LR 0.000500 Time 0.022315 -2022-12-06 11:20:09,689 - Epoch: [129][ 260/ 1200] Overall Loss 0.189199 Objective Loss 0.189199 LR 0.000500 Time 0.022197 -2022-12-06 11:20:09,882 - Epoch: [129][ 270/ 1200] Overall Loss 0.189305 Objective Loss 0.189305 LR 0.000500 Time 0.022086 -2022-12-06 11:20:10,075 - Epoch: [129][ 280/ 1200] Overall Loss 0.189647 Objective Loss 0.189647 LR 0.000500 Time 0.021985 -2022-12-06 11:20:10,267 - Epoch: [129][ 290/ 1200] Overall Loss 0.189089 Objective Loss 0.189089 LR 0.000500 Time 0.021887 -2022-12-06 11:20:10,460 - Epoch: [129][ 300/ 1200] Overall Loss 0.188444 Objective Loss 0.188444 LR 0.000500 Time 0.021799 -2022-12-06 11:20:10,652 - Epoch: [129][ 310/ 1200] Overall Loss 0.188529 Objective Loss 0.188529 LR 0.000500 Time 0.021715 -2022-12-06 11:20:10,846 - Epoch: [129][ 320/ 1200] Overall Loss 0.188442 Objective Loss 0.188442 LR 0.000500 Time 0.021638 -2022-12-06 11:20:11,038 - Epoch: [129][ 330/ 1200] Overall Loss 0.188507 Objective Loss 0.188507 LR 0.000500 Time 0.021562 -2022-12-06 11:20:11,230 - Epoch: [129][ 340/ 1200] Overall Loss 0.187750 Objective Loss 0.187750 LR 0.000500 Time 0.021494 -2022-12-06 11:20:11,423 - Epoch: [129][ 350/ 1200] Overall Loss 0.186884 Objective Loss 0.186884 LR 0.000500 Time 0.021427 -2022-12-06 11:20:11,616 - Epoch: [129][ 360/ 1200] Overall Loss 0.186679 Objective Loss 0.186679 LR 0.000500 Time 0.021367 -2022-12-06 11:20:11,808 - Epoch: [129][ 370/ 1200] Overall Loss 0.186963 Objective Loss 0.186963 LR 0.000500 Time 0.021308 -2022-12-06 11:20:12,001 - Epoch: [129][ 380/ 1200] Overall Loss 0.186702 Objective Loss 0.186702 LR 0.000500 Time 0.021254 -2022-12-06 11:20:12,193 - Epoch: [129][ 390/ 1200] Overall Loss 0.186399 Objective Loss 0.186399 LR 0.000500 Time 0.021200 -2022-12-06 11:20:12,386 - Epoch: [129][ 400/ 1200] Overall Loss 0.186852 Objective Loss 0.186852 LR 0.000500 Time 0.021150 -2022-12-06 11:20:12,578 - Epoch: [129][ 410/ 1200] Overall Loss 0.186790 Objective Loss 0.186790 LR 0.000500 Time 0.021102 -2022-12-06 11:20:12,771 - Epoch: [129][ 420/ 1200] Overall Loss 0.186795 Objective Loss 0.186795 LR 0.000500 Time 0.021058 -2022-12-06 11:20:12,964 - Epoch: [129][ 430/ 1200] Overall Loss 0.187305 Objective Loss 0.187305 LR 0.000500 Time 0.021014 -2022-12-06 11:20:13,157 - Epoch: [129][ 440/ 1200] Overall Loss 0.187508 Objective Loss 0.187508 LR 0.000500 Time 0.020974 -2022-12-06 11:20:13,349 - Epoch: [129][ 450/ 1200] Overall Loss 0.187862 Objective Loss 0.187862 LR 0.000500 Time 0.020934 -2022-12-06 11:20:13,542 - Epoch: [129][ 460/ 1200] Overall Loss 0.187763 Objective Loss 0.187763 LR 0.000500 Time 0.020897 -2022-12-06 11:20:13,735 - Epoch: [129][ 470/ 1200] Overall Loss 0.187895 Objective Loss 0.187895 LR 0.000500 Time 0.020862 -2022-12-06 11:20:13,928 - Epoch: [129][ 480/ 1200] Overall Loss 0.188020 Objective Loss 0.188020 LR 0.000500 Time 0.020829 -2022-12-06 11:20:14,121 - Epoch: [129][ 490/ 1200] Overall Loss 0.187842 Objective Loss 0.187842 LR 0.000500 Time 0.020795 -2022-12-06 11:20:14,314 - Epoch: [129][ 500/ 1200] Overall Loss 0.187757 Objective Loss 0.187757 LR 0.000500 Time 0.020765 -2022-12-06 11:20:14,507 - Epoch: [129][ 510/ 1200] Overall Loss 0.187743 Objective Loss 0.187743 LR 0.000500 Time 0.020735 -2022-12-06 11:20:14,700 - Epoch: [129][ 520/ 1200] Overall Loss 0.187372 Objective Loss 0.187372 LR 0.000500 Time 0.020707 -2022-12-06 11:20:14,892 - Epoch: [129][ 530/ 1200] Overall Loss 0.187617 Objective Loss 0.187617 LR 0.000500 Time 0.020678 -2022-12-06 11:20:15,085 - Epoch: [129][ 540/ 1200] Overall Loss 0.187809 Objective Loss 0.187809 LR 0.000500 Time 0.020651 -2022-12-06 11:20:15,278 - Epoch: [129][ 550/ 1200] Overall Loss 0.187792 Objective Loss 0.187792 LR 0.000500 Time 0.020625 -2022-12-06 11:20:15,471 - Epoch: [129][ 560/ 1200] Overall Loss 0.188071 Objective Loss 0.188071 LR 0.000500 Time 0.020601 -2022-12-06 11:20:15,664 - Epoch: [129][ 570/ 1200] Overall Loss 0.188014 Objective Loss 0.188014 LR 0.000500 Time 0.020576 -2022-12-06 11:20:15,857 - Epoch: [129][ 580/ 1200] Overall Loss 0.188418 Objective Loss 0.188418 LR 0.000500 Time 0.020554 -2022-12-06 11:20:16,049 - Epoch: [129][ 590/ 1200] Overall Loss 0.188771 Objective Loss 0.188771 LR 0.000500 Time 0.020531 -2022-12-06 11:20:16,243 - Epoch: [129][ 600/ 1200] Overall Loss 0.188807 Objective Loss 0.188807 LR 0.000500 Time 0.020509 -2022-12-06 11:20:16,435 - Epoch: [129][ 610/ 1200] Overall Loss 0.188870 Objective Loss 0.188870 LR 0.000500 Time 0.020488 -2022-12-06 11:20:16,628 - Epoch: [129][ 620/ 1200] Overall Loss 0.189138 Objective Loss 0.189138 LR 0.000500 Time 0.020468 -2022-12-06 11:20:16,820 - Epoch: [129][ 630/ 1200] Overall Loss 0.188767 Objective Loss 0.188767 LR 0.000500 Time 0.020447 -2022-12-06 11:20:17,013 - Epoch: [129][ 640/ 1200] Overall Loss 0.189142 Objective Loss 0.189142 LR 0.000500 Time 0.020428 -2022-12-06 11:20:17,206 - Epoch: [129][ 650/ 1200] Overall Loss 0.188999 Objective Loss 0.188999 LR 0.000500 Time 0.020409 -2022-12-06 11:20:17,398 - Epoch: [129][ 660/ 1200] Overall Loss 0.189537 Objective Loss 0.189537 LR 0.000500 Time 0.020391 -2022-12-06 11:20:17,591 - Epoch: [129][ 670/ 1200] Overall Loss 0.189469 Objective Loss 0.189469 LR 0.000500 Time 0.020374 -2022-12-06 11:20:17,784 - Epoch: [129][ 680/ 1200] Overall Loss 0.189725 Objective Loss 0.189725 LR 0.000500 Time 0.020357 -2022-12-06 11:20:17,977 - Epoch: [129][ 690/ 1200] Overall Loss 0.189871 Objective Loss 0.189871 LR 0.000500 Time 0.020340 -2022-12-06 11:20:18,169 - Epoch: [129][ 700/ 1200] Overall Loss 0.189724 Objective Loss 0.189724 LR 0.000500 Time 0.020324 -2022-12-06 11:20:18,361 - Epoch: [129][ 710/ 1200] Overall Loss 0.190181 Objective Loss 0.190181 LR 0.000500 Time 0.020307 -2022-12-06 11:20:18,554 - Epoch: [129][ 720/ 1200] Overall Loss 0.190014 Objective Loss 0.190014 LR 0.000500 Time 0.020292 -2022-12-06 11:20:18,746 - Epoch: [129][ 730/ 1200] Overall Loss 0.189891 Objective Loss 0.189891 LR 0.000500 Time 0.020277 -2022-12-06 11:20:18,939 - Epoch: [129][ 740/ 1200] Overall Loss 0.189979 Objective Loss 0.189979 LR 0.000500 Time 0.020262 -2022-12-06 11:20:19,132 - Epoch: [129][ 750/ 1200] Overall Loss 0.190030 Objective Loss 0.190030 LR 0.000500 Time 0.020248 -2022-12-06 11:20:19,324 - Epoch: [129][ 760/ 1200] Overall Loss 0.189982 Objective Loss 0.189982 LR 0.000500 Time 0.020234 -2022-12-06 11:20:19,516 - Epoch: [129][ 770/ 1200] Overall Loss 0.190069 Objective Loss 0.190069 LR 0.000500 Time 0.020220 -2022-12-06 11:20:19,709 - Epoch: [129][ 780/ 1200] Overall Loss 0.190032 Objective Loss 0.190032 LR 0.000500 Time 0.020207 -2022-12-06 11:20:19,901 - Epoch: [129][ 790/ 1200] Overall Loss 0.190251 Objective Loss 0.190251 LR 0.000500 Time 0.020194 -2022-12-06 11:20:20,093 - Epoch: [129][ 800/ 1200] Overall Loss 0.190540 Objective Loss 0.190540 LR 0.000500 Time 0.020182 -2022-12-06 11:20:20,286 - Epoch: [129][ 810/ 1200] Overall Loss 0.190102 Objective Loss 0.190102 LR 0.000500 Time 0.020170 -2022-12-06 11:20:20,478 - Epoch: [129][ 820/ 1200] Overall Loss 0.190193 Objective Loss 0.190193 LR 0.000500 Time 0.020158 -2022-12-06 11:20:20,671 - Epoch: [129][ 830/ 1200] Overall Loss 0.190212 Objective Loss 0.190212 LR 0.000500 Time 0.020146 -2022-12-06 11:20:20,863 - Epoch: [129][ 840/ 1200] Overall Loss 0.190089 Objective Loss 0.190089 LR 0.000500 Time 0.020134 -2022-12-06 11:20:21,055 - Epoch: [129][ 850/ 1200] Overall Loss 0.190174 Objective Loss 0.190174 LR 0.000500 Time 0.020122 -2022-12-06 11:20:21,247 - Epoch: [129][ 860/ 1200] Overall Loss 0.189952 Objective Loss 0.189952 LR 0.000500 Time 0.020111 -2022-12-06 11:20:21,440 - Epoch: [129][ 870/ 1200] Overall Loss 0.190166 Objective Loss 0.190166 LR 0.000500 Time 0.020101 -2022-12-06 11:20:21,632 - Epoch: [129][ 880/ 1200] Overall Loss 0.190226 Objective Loss 0.190226 LR 0.000500 Time 0.020091 -2022-12-06 11:20:21,824 - Epoch: [129][ 890/ 1200] Overall Loss 0.190083 Objective Loss 0.190083 LR 0.000500 Time 0.020080 -2022-12-06 11:20:22,016 - Epoch: [129][ 900/ 1200] Overall Loss 0.189786 Objective Loss 0.189786 LR 0.000500 Time 0.020069 -2022-12-06 11:20:22,208 - Epoch: [129][ 910/ 1200] Overall Loss 0.189939 Objective Loss 0.189939 LR 0.000500 Time 0.020059 -2022-12-06 11:20:22,401 - Epoch: [129][ 920/ 1200] Overall Loss 0.189914 Objective Loss 0.189914 LR 0.000500 Time 0.020051 -2022-12-06 11:20:22,593 - Epoch: [129][ 930/ 1200] Overall Loss 0.190021 Objective Loss 0.190021 LR 0.000500 Time 0.020041 -2022-12-06 11:20:22,786 - Epoch: [129][ 940/ 1200] Overall Loss 0.189808 Objective Loss 0.189808 LR 0.000500 Time 0.020032 -2022-12-06 11:20:22,979 - Epoch: [129][ 950/ 1200] Overall Loss 0.189685 Objective Loss 0.189685 LR 0.000500 Time 0.020024 -2022-12-06 11:20:23,171 - Epoch: [129][ 960/ 1200] Overall Loss 0.189723 Objective Loss 0.189723 LR 0.000500 Time 0.020015 -2022-12-06 11:20:23,364 - Epoch: [129][ 970/ 1200] Overall Loss 0.189423 Objective Loss 0.189423 LR 0.000500 Time 0.020007 -2022-12-06 11:20:23,556 - Epoch: [129][ 980/ 1200] Overall Loss 0.189512 Objective Loss 0.189512 LR 0.000500 Time 0.019998 -2022-12-06 11:20:23,748 - Epoch: [129][ 990/ 1200] Overall Loss 0.189425 Objective Loss 0.189425 LR 0.000500 Time 0.019990 -2022-12-06 11:20:23,941 - Epoch: [129][ 1000/ 1200] Overall Loss 0.189647 Objective Loss 0.189647 LR 0.000500 Time 0.019982 -2022-12-06 11:20:24,133 - Epoch: [129][ 1010/ 1200] Overall Loss 0.190208 Objective Loss 0.190208 LR 0.000500 Time 0.019974 -2022-12-06 11:20:24,326 - Epoch: [129][ 1020/ 1200] Overall Loss 0.190190 Objective Loss 0.190190 LR 0.000500 Time 0.019966 -2022-12-06 11:20:24,519 - Epoch: [129][ 1030/ 1200] Overall Loss 0.190115 Objective Loss 0.190115 LR 0.000500 Time 0.019959 -2022-12-06 11:20:24,711 - Epoch: [129][ 1040/ 1200] Overall Loss 0.190252 Objective Loss 0.190252 LR 0.000500 Time 0.019952 -2022-12-06 11:20:24,904 - Epoch: [129][ 1050/ 1200] Overall Loss 0.190292 Objective Loss 0.190292 LR 0.000500 Time 0.019944 -2022-12-06 11:20:25,096 - Epoch: [129][ 1060/ 1200] Overall Loss 0.190459 Objective Loss 0.190459 LR 0.000500 Time 0.019937 -2022-12-06 11:20:25,289 - Epoch: [129][ 1070/ 1200] Overall Loss 0.190560 Objective Loss 0.190560 LR 0.000500 Time 0.019930 -2022-12-06 11:20:25,481 - Epoch: [129][ 1080/ 1200] Overall Loss 0.190662 Objective Loss 0.190662 LR 0.000500 Time 0.019924 -2022-12-06 11:20:25,674 - Epoch: [129][ 1090/ 1200] Overall Loss 0.190746 Objective Loss 0.190746 LR 0.000500 Time 0.019917 -2022-12-06 11:20:25,867 - Epoch: [129][ 1100/ 1200] Overall Loss 0.190686 Objective Loss 0.190686 LR 0.000500 Time 0.019911 -2022-12-06 11:20:26,059 - Epoch: [129][ 1110/ 1200] Overall Loss 0.190338 Objective Loss 0.190338 LR 0.000500 Time 0.019904 -2022-12-06 11:20:26,251 - Epoch: [129][ 1120/ 1200] Overall Loss 0.190532 Objective Loss 0.190532 LR 0.000500 Time 0.019898 -2022-12-06 11:20:26,444 - Epoch: [129][ 1130/ 1200] Overall Loss 0.190714 Objective Loss 0.190714 LR 0.000500 Time 0.019891 -2022-12-06 11:20:26,636 - Epoch: [129][ 1140/ 1200] Overall Loss 0.190842 Objective Loss 0.190842 LR 0.000500 Time 0.019885 -2022-12-06 11:20:26,828 - Epoch: [129][ 1150/ 1200] Overall Loss 0.190618 Objective Loss 0.190618 LR 0.000500 Time 0.019879 -2022-12-06 11:20:27,020 - Epoch: [129][ 1160/ 1200] Overall Loss 0.190587 Objective Loss 0.190587 LR 0.000500 Time 0.019872 -2022-12-06 11:20:27,212 - Epoch: [129][ 1170/ 1200] Overall Loss 0.190599 Objective Loss 0.190599 LR 0.000500 Time 0.019867 -2022-12-06 11:20:27,405 - Epoch: [129][ 1180/ 1200] Overall Loss 0.190761 Objective Loss 0.190761 LR 0.000500 Time 0.019861 -2022-12-06 11:20:27,598 - Epoch: [129][ 1190/ 1200] Overall Loss 0.190789 Objective Loss 0.190789 LR 0.000500 Time 0.019855 -2022-12-06 11:20:27,829 - Epoch: [129][ 1200/ 1200] Overall Loss 0.190831 Objective Loss 0.190831 Top1 89.958159 Top5 98.953975 LR 0.000500 Time 0.019882 -2022-12-06 11:20:27,920 - --- validate (epoch=129)----------- -2022-12-06 11:20:27,921 - 34129 samples (256 per mini-batch) -2022-12-06 11:20:28,366 - Epoch: [129][ 10/ 134] Loss 0.234730 Top1 88.437500 Top5 98.515625 -2022-12-06 11:20:28,498 - Epoch: [129][ 20/ 134] Loss 0.239120 Top1 87.929688 Top5 98.554688 -2022-12-06 11:20:28,629 - Epoch: [129][ 30/ 134] Loss 0.251481 Top1 87.526042 Top5 98.528646 -2022-12-06 11:20:28,754 - Epoch: [129][ 40/ 134] Loss 0.258896 Top1 87.333984 Top5 98.330078 -2022-12-06 11:20:28,891 - Epoch: [129][ 50/ 134] Loss 0.258201 Top1 87.328125 Top5 98.289062 -2022-12-06 11:20:29,029 - Epoch: [129][ 60/ 134] Loss 0.257477 Top1 87.467448 Top5 98.235677 -2022-12-06 11:20:29,174 - Epoch: [129][ 70/ 134] Loss 0.255443 Top1 87.522321 Top5 98.258929 -2022-12-06 11:20:29,307 - Epoch: [129][ 80/ 134] Loss 0.250338 Top1 87.607422 Top5 98.310547 -2022-12-06 11:20:29,438 - Epoch: [129][ 90/ 134] Loss 0.251700 Top1 87.547743 Top5 98.281250 -2022-12-06 11:20:29,573 - Epoch: [129][ 100/ 134] Loss 0.254828 Top1 87.406250 Top5 98.289062 -2022-12-06 11:20:29,702 - Epoch: [129][ 110/ 134] Loss 0.254492 Top1 87.414773 Top5 98.338068 -2022-12-06 11:20:29,835 - Epoch: [129][ 120/ 134] Loss 0.252628 Top1 87.441406 Top5 98.359375 -2022-12-06 11:20:29,979 - Epoch: [129][ 130/ 134] Loss 0.254241 Top1 87.376803 Top5 98.347356 -2022-12-06 11:20:30,016 - Epoch: [129][ 134/ 134] Loss 0.254132 Top1 87.350933 Top5 98.350377 -2022-12-06 11:20:30,106 - ==> Top1: 87.351 Top5: 98.350 Loss: 0.254 - -2022-12-06 11:20:30,107 - ==> Confusion: -[[ 938 2 1 3 2 3 0 0 3 29 0 1 1 4 4 1 1 1 0 0 2] - [ 4 942 3 2 10 18 4 12 2 0 2 6 0 1 1 2 7 0 4 2 5] - [ 8 3 1021 6 6 0 19 9 0 2 4 4 1 0 2 5 0 2 2 2 7] - [ 2 3 19 938 3 2 0 1 0 0 10 0 1 2 16 0 1 2 12 0 8] - [ 11 3 1 0 963 5 0 1 1 5 2 2 0 1 9 5 4 2 2 0 3] - [ 4 12 2 3 5 984 1 16 2 2 1 11 2 11 0 2 2 0 1 3 5] - [ 2 1 9 0 3 3 1077 1 0 0 2 4 2 0 0 4 0 1 0 7 2] - [ 3 10 10 1 2 29 13 948 0 0 1 6 0 0 1 0 0 1 14 11 4] - [ 10 3 0 0 1 4 1 0 969 41 7 1 2 12 7 0 2 2 1 1 0] - [ 85 0 0 0 4 6 0 1 17 862 1 5 0 10 0 1 1 2 0 0 6] - [ 3 0 4 3 1 3 0 4 5 2 966 1 0 11 5 0 1 0 2 3 5] - [ 5 1 3 0 1 12 4 2 0 0 1 984 14 4 0 4 2 2 1 8 3] - [ 1 1 1 2 1 5 0 0 0 0 0 38 887 0 1 8 3 6 1 3 11] - [ 0 0 0 0 0 5 1 3 10 10 4 8 3 961 1 2 4 2 0 1 8] - [ 9 3 3 10 6 4 0 0 15 3 0 2 1 3 1059 0 3 0 5 0 4] - [ 0 0 1 3 5 0 5 0 0 0 1 6 2 1 0 996 5 12 0 3 3] - [ 1 1 0 0 3 0 0 0 0 0 0 2 1 3 0 10 1041 0 0 4 6] - [ 3 0 2 1 0 2 2 1 0 2 0 12 14 2 1 19 2 970 0 1 2] - [ 4 8 2 8 1 1 0 21 1 1 5 4 3 2 7 1 1 0 929 2 7] - [ 2 3 1 0 1 5 9 8 0 1 3 13 3 4 1 3 5 0 2 1006 10] - [ 160 207 169 77 123 181 68 140 78 81 157 111 262 242 127 113 201 65 107 188 10369]] - -2022-12-06 11:20:30,680 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:20:30,680 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:20:30,687 - - -2022-12-06 11:20:30,687 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:20:31,741 - Epoch: [130][ 10/ 1200] Overall Loss 0.195982 Objective Loss 0.195982 LR 0.000500 Time 0.105328 -2022-12-06 11:20:31,942 - Epoch: [130][ 20/ 1200] Overall Loss 0.192223 Objective Loss 0.192223 LR 0.000500 Time 0.062669 -2022-12-06 11:20:32,143 - Epoch: [130][ 30/ 1200] Overall Loss 0.196949 Objective Loss 0.196949 LR 0.000500 Time 0.048467 -2022-12-06 11:20:32,341 - Epoch: [130][ 40/ 1200] Overall Loss 0.192641 Objective Loss 0.192641 LR 0.000500 Time 0.041275 -2022-12-06 11:20:32,542 - Epoch: [130][ 50/ 1200] Overall Loss 0.189281 Objective Loss 0.189281 LR 0.000500 Time 0.037039 -2022-12-06 11:20:32,740 - Epoch: [130][ 60/ 1200] Overall Loss 0.190976 Objective Loss 0.190976 LR 0.000500 Time 0.034155 -2022-12-06 11:20:32,940 - Epoch: [130][ 70/ 1200] Overall Loss 0.192093 Objective Loss 0.192093 LR 0.000500 Time 0.032130 -2022-12-06 11:20:33,138 - Epoch: [130][ 80/ 1200] Overall Loss 0.193262 Objective Loss 0.193262 LR 0.000500 Time 0.030575 -2022-12-06 11:20:33,338 - Epoch: [130][ 90/ 1200] Overall Loss 0.192468 Objective Loss 0.192468 LR 0.000500 Time 0.029403 -2022-12-06 11:20:33,537 - Epoch: [130][ 100/ 1200] Overall Loss 0.192263 Objective Loss 0.192263 LR 0.000500 Time 0.028438 -2022-12-06 11:20:33,737 - Epoch: [130][ 110/ 1200] Overall Loss 0.192405 Objective Loss 0.192405 LR 0.000500 Time 0.027673 -2022-12-06 11:20:33,935 - Epoch: [130][ 120/ 1200] Overall Loss 0.189931 Objective Loss 0.189931 LR 0.000500 Time 0.027006 -2022-12-06 11:20:34,135 - Epoch: [130][ 130/ 1200] Overall Loss 0.189787 Objective Loss 0.189787 LR 0.000500 Time 0.026468 -2022-12-06 11:20:34,333 - Epoch: [130][ 140/ 1200] Overall Loss 0.190564 Objective Loss 0.190564 LR 0.000500 Time 0.025984 -2022-12-06 11:20:34,534 - Epoch: [130][ 150/ 1200] Overall Loss 0.189548 Objective Loss 0.189548 LR 0.000500 Time 0.025590 -2022-12-06 11:20:34,731 - Epoch: [130][ 160/ 1200] Overall Loss 0.189107 Objective Loss 0.189107 LR 0.000500 Time 0.025217 -2022-12-06 11:20:34,932 - Epoch: [130][ 170/ 1200] Overall Loss 0.189320 Objective Loss 0.189320 LR 0.000500 Time 0.024912 -2022-12-06 11:20:35,130 - Epoch: [130][ 180/ 1200] Overall Loss 0.188230 Objective Loss 0.188230 LR 0.000500 Time 0.024624 -2022-12-06 11:20:35,330 - Epoch: [130][ 190/ 1200] Overall Loss 0.188323 Objective Loss 0.188323 LR 0.000500 Time 0.024381 -2022-12-06 11:20:35,528 - Epoch: [130][ 200/ 1200] Overall Loss 0.188361 Objective Loss 0.188361 LR 0.000500 Time 0.024148 -2022-12-06 11:20:35,728 - Epoch: [130][ 210/ 1200] Overall Loss 0.187387 Objective Loss 0.187387 LR 0.000500 Time 0.023949 -2022-12-06 11:20:35,925 - Epoch: [130][ 220/ 1200] Overall Loss 0.188126 Objective Loss 0.188126 LR 0.000500 Time 0.023754 -2022-12-06 11:20:36,126 - Epoch: [130][ 230/ 1200] Overall Loss 0.189315 Objective Loss 0.189315 LR 0.000500 Time 0.023592 -2022-12-06 11:20:36,324 - Epoch: [130][ 240/ 1200] Overall Loss 0.189148 Objective Loss 0.189148 LR 0.000500 Time 0.023431 -2022-12-06 11:20:36,524 - Epoch: [130][ 250/ 1200] Overall Loss 0.188221 Objective Loss 0.188221 LR 0.000500 Time 0.023293 -2022-12-06 11:20:36,722 - Epoch: [130][ 260/ 1200] Overall Loss 0.187245 Objective Loss 0.187245 LR 0.000500 Time 0.023154 -2022-12-06 11:20:36,922 - Epoch: [130][ 270/ 1200] Overall Loss 0.187304 Objective Loss 0.187304 LR 0.000500 Time 0.023036 -2022-12-06 11:20:37,120 - Epoch: [130][ 280/ 1200] Overall Loss 0.187929 Objective Loss 0.187929 LR 0.000500 Time 0.022919 -2022-12-06 11:20:37,320 - Epoch: [130][ 290/ 1200] Overall Loss 0.187774 Objective Loss 0.187774 LR 0.000500 Time 0.022815 -2022-12-06 11:20:37,518 - Epoch: [130][ 300/ 1200] Overall Loss 0.187094 Objective Loss 0.187094 LR 0.000500 Time 0.022713 -2022-12-06 11:20:37,719 - Epoch: [130][ 310/ 1200] Overall Loss 0.186572 Objective Loss 0.186572 LR 0.000500 Time 0.022628 -2022-12-06 11:20:37,916 - Epoch: [130][ 320/ 1200] Overall Loss 0.186646 Objective Loss 0.186646 LR 0.000500 Time 0.022535 -2022-12-06 11:20:38,116 - Epoch: [130][ 330/ 1200] Overall Loss 0.186991 Objective Loss 0.186991 LR 0.000500 Time 0.022456 -2022-12-06 11:20:38,313 - Epoch: [130][ 340/ 1200] Overall Loss 0.187198 Objective Loss 0.187198 LR 0.000500 Time 0.022374 -2022-12-06 11:20:38,514 - Epoch: [130][ 350/ 1200] Overall Loss 0.187350 Objective Loss 0.187350 LR 0.000500 Time 0.022306 -2022-12-06 11:20:38,723 - Epoch: [130][ 360/ 1200] Overall Loss 0.187217 Objective Loss 0.187217 LR 0.000500 Time 0.022267 -2022-12-06 11:20:38,932 - Epoch: [130][ 370/ 1200] Overall Loss 0.187129 Objective Loss 0.187129 LR 0.000500 Time 0.022228 -2022-12-06 11:20:39,141 - Epoch: [130][ 380/ 1200] Overall Loss 0.187525 Objective Loss 0.187525 LR 0.000500 Time 0.022192 -2022-12-06 11:20:39,350 - Epoch: [130][ 390/ 1200] Overall Loss 0.187514 Objective Loss 0.187514 LR 0.000500 Time 0.022156 -2022-12-06 11:20:39,559 - Epoch: [130][ 400/ 1200] Overall Loss 0.187229 Objective Loss 0.187229 LR 0.000500 Time 0.022125 -2022-12-06 11:20:39,769 - Epoch: [130][ 410/ 1200] Overall Loss 0.187511 Objective Loss 0.187511 LR 0.000500 Time 0.022094 -2022-12-06 11:20:39,978 - Epoch: [130][ 420/ 1200] Overall Loss 0.187076 Objective Loss 0.187076 LR 0.000500 Time 0.022066 -2022-12-06 11:20:40,186 - Epoch: [130][ 430/ 1200] Overall Loss 0.187634 Objective Loss 0.187634 LR 0.000500 Time 0.022035 -2022-12-06 11:20:40,396 - Epoch: [130][ 440/ 1200] Overall Loss 0.188027 Objective Loss 0.188027 LR 0.000500 Time 0.022010 -2022-12-06 11:20:40,604 - Epoch: [130][ 450/ 1200] Overall Loss 0.188070 Objective Loss 0.188070 LR 0.000500 Time 0.021983 -2022-12-06 11:20:40,814 - Epoch: [130][ 460/ 1200] Overall Loss 0.188326 Objective Loss 0.188326 LR 0.000500 Time 0.021960 -2022-12-06 11:20:41,022 - Epoch: [130][ 470/ 1200] Overall Loss 0.188388 Objective Loss 0.188388 LR 0.000500 Time 0.021934 -2022-12-06 11:20:41,232 - Epoch: [130][ 480/ 1200] Overall Loss 0.187874 Objective Loss 0.187874 LR 0.000500 Time 0.021914 -2022-12-06 11:20:41,442 - Epoch: [130][ 490/ 1200] Overall Loss 0.188284 Objective Loss 0.188284 LR 0.000500 Time 0.021892 -2022-12-06 11:20:41,651 - Epoch: [130][ 500/ 1200] Overall Loss 0.188218 Objective Loss 0.188218 LR 0.000500 Time 0.021873 -2022-12-06 11:20:41,861 - Epoch: [130][ 510/ 1200] Overall Loss 0.187752 Objective Loss 0.187752 LR 0.000500 Time 0.021853 -2022-12-06 11:20:42,070 - Epoch: [130][ 520/ 1200] Overall Loss 0.187775 Objective Loss 0.187775 LR 0.000500 Time 0.021835 -2022-12-06 11:20:42,278 - Epoch: [130][ 530/ 1200] Overall Loss 0.187754 Objective Loss 0.187754 LR 0.000500 Time 0.021815 -2022-12-06 11:20:42,488 - Epoch: [130][ 540/ 1200] Overall Loss 0.188082 Objective Loss 0.188082 LR 0.000500 Time 0.021799 -2022-12-06 11:20:42,697 - Epoch: [130][ 550/ 1200] Overall Loss 0.188367 Objective Loss 0.188367 LR 0.000500 Time 0.021781 -2022-12-06 11:20:42,907 - Epoch: [130][ 560/ 1200] Overall Loss 0.188525 Objective Loss 0.188525 LR 0.000500 Time 0.021766 -2022-12-06 11:20:43,116 - Epoch: [130][ 570/ 1200] Overall Loss 0.188191 Objective Loss 0.188191 LR 0.000500 Time 0.021749 -2022-12-06 11:20:43,326 - Epoch: [130][ 580/ 1200] Overall Loss 0.188233 Objective Loss 0.188233 LR 0.000500 Time 0.021735 -2022-12-06 11:20:43,535 - Epoch: [130][ 590/ 1200] Overall Loss 0.188444 Objective Loss 0.188444 LR 0.000500 Time 0.021720 -2022-12-06 11:20:43,744 - Epoch: [130][ 600/ 1200] Overall Loss 0.188381 Objective Loss 0.188381 LR 0.000500 Time 0.021706 -2022-12-06 11:20:43,953 - Epoch: [130][ 610/ 1200] Overall Loss 0.188625 Objective Loss 0.188625 LR 0.000500 Time 0.021691 -2022-12-06 11:20:44,162 - Epoch: [130][ 620/ 1200] Overall Loss 0.188675 Objective Loss 0.188675 LR 0.000500 Time 0.021679 -2022-12-06 11:20:44,370 - Epoch: [130][ 630/ 1200] Overall Loss 0.188682 Objective Loss 0.188682 LR 0.000500 Time 0.021664 -2022-12-06 11:20:44,581 - Epoch: [130][ 640/ 1200] Overall Loss 0.188002 Objective Loss 0.188002 LR 0.000500 Time 0.021654 -2022-12-06 11:20:44,790 - Epoch: [130][ 650/ 1200] Overall Loss 0.187694 Objective Loss 0.187694 LR 0.000500 Time 0.021642 -2022-12-06 11:20:45,000 - Epoch: [130][ 660/ 1200] Overall Loss 0.187686 Objective Loss 0.187686 LR 0.000500 Time 0.021631 -2022-12-06 11:20:45,210 - Epoch: [130][ 670/ 1200] Overall Loss 0.187659 Objective Loss 0.187659 LR 0.000500 Time 0.021620 -2022-12-06 11:20:45,420 - Epoch: [130][ 680/ 1200] Overall Loss 0.187692 Objective Loss 0.187692 LR 0.000500 Time 0.021610 -2022-12-06 11:20:45,628 - Epoch: [130][ 690/ 1200] Overall Loss 0.187544 Objective Loss 0.187544 LR 0.000500 Time 0.021599 -2022-12-06 11:20:45,838 - Epoch: [130][ 700/ 1200] Overall Loss 0.187556 Objective Loss 0.187556 LR 0.000500 Time 0.021589 -2022-12-06 11:20:46,047 - Epoch: [130][ 710/ 1200] Overall Loss 0.187456 Objective Loss 0.187456 LR 0.000500 Time 0.021578 -2022-12-06 11:20:46,256 - Epoch: [130][ 720/ 1200] Overall Loss 0.187570 Objective Loss 0.187570 LR 0.000500 Time 0.021569 -2022-12-06 11:20:46,465 - Epoch: [130][ 730/ 1200] Overall Loss 0.187807 Objective Loss 0.187807 LR 0.000500 Time 0.021558 -2022-12-06 11:20:46,675 - Epoch: [130][ 740/ 1200] Overall Loss 0.188010 Objective Loss 0.188010 LR 0.000500 Time 0.021549 -2022-12-06 11:20:46,883 - Epoch: [130][ 750/ 1200] Overall Loss 0.187990 Objective Loss 0.187990 LR 0.000500 Time 0.021540 -2022-12-06 11:20:47,093 - Epoch: [130][ 760/ 1200] Overall Loss 0.187941 Objective Loss 0.187941 LR 0.000500 Time 0.021532 -2022-12-06 11:20:47,302 - Epoch: [130][ 770/ 1200] Overall Loss 0.188104 Objective Loss 0.188104 LR 0.000500 Time 0.021522 -2022-12-06 11:20:47,512 - Epoch: [130][ 780/ 1200] Overall Loss 0.188047 Objective Loss 0.188047 LR 0.000500 Time 0.021515 -2022-12-06 11:20:47,721 - Epoch: [130][ 790/ 1200] Overall Loss 0.188013 Objective Loss 0.188013 LR 0.000500 Time 0.021507 -2022-12-06 11:20:47,931 - Epoch: [130][ 800/ 1200] Overall Loss 0.187990 Objective Loss 0.187990 LR 0.000500 Time 0.021500 -2022-12-06 11:20:48,139 - Epoch: [130][ 810/ 1200] Overall Loss 0.188222 Objective Loss 0.188222 LR 0.000500 Time 0.021491 -2022-12-06 11:20:48,349 - Epoch: [130][ 820/ 1200] Overall Loss 0.188462 Objective Loss 0.188462 LR 0.000500 Time 0.021484 -2022-12-06 11:20:48,558 - Epoch: [130][ 830/ 1200] Overall Loss 0.188907 Objective Loss 0.188907 LR 0.000500 Time 0.021475 -2022-12-06 11:20:48,768 - Epoch: [130][ 840/ 1200] Overall Loss 0.189067 Objective Loss 0.189067 LR 0.000500 Time 0.021469 -2022-12-06 11:20:48,976 - Epoch: [130][ 850/ 1200] Overall Loss 0.189096 Objective Loss 0.189096 LR 0.000500 Time 0.021461 -2022-12-06 11:20:49,186 - Epoch: [130][ 860/ 1200] Overall Loss 0.189138 Objective Loss 0.189138 LR 0.000500 Time 0.021455 -2022-12-06 11:20:49,394 - Epoch: [130][ 870/ 1200] Overall Loss 0.188845 Objective Loss 0.188845 LR 0.000500 Time 0.021447 -2022-12-06 11:20:49,604 - Epoch: [130][ 880/ 1200] Overall Loss 0.188536 Objective Loss 0.188536 LR 0.000500 Time 0.021441 -2022-12-06 11:20:49,813 - Epoch: [130][ 890/ 1200] Overall Loss 0.188314 Objective Loss 0.188314 LR 0.000500 Time 0.021434 -2022-12-06 11:20:50,022 - Epoch: [130][ 900/ 1200] Overall Loss 0.188301 Objective Loss 0.188301 LR 0.000500 Time 0.021428 -2022-12-06 11:20:50,230 - Epoch: [130][ 910/ 1200] Overall Loss 0.188269 Objective Loss 0.188269 LR 0.000500 Time 0.021421 -2022-12-06 11:20:50,440 - Epoch: [130][ 920/ 1200] Overall Loss 0.188203 Objective Loss 0.188203 LR 0.000500 Time 0.021415 -2022-12-06 11:20:50,648 - Epoch: [130][ 930/ 1200] Overall Loss 0.188435 Objective Loss 0.188435 LR 0.000500 Time 0.021408 -2022-12-06 11:20:50,858 - Epoch: [130][ 940/ 1200] Overall Loss 0.188364 Objective Loss 0.188364 LR 0.000500 Time 0.021403 -2022-12-06 11:20:51,067 - Epoch: [130][ 950/ 1200] Overall Loss 0.188184 Objective Loss 0.188184 LR 0.000500 Time 0.021397 -2022-12-06 11:20:51,278 - Epoch: [130][ 960/ 1200] Overall Loss 0.188493 Objective Loss 0.188493 LR 0.000500 Time 0.021393 -2022-12-06 11:20:51,487 - Epoch: [130][ 970/ 1200] Overall Loss 0.188444 Objective Loss 0.188444 LR 0.000500 Time 0.021388 -2022-12-06 11:20:51,697 - Epoch: [130][ 980/ 1200] Overall Loss 0.188265 Objective Loss 0.188265 LR 0.000500 Time 0.021384 -2022-12-06 11:20:51,906 - Epoch: [130][ 990/ 1200] Overall Loss 0.188372 Objective Loss 0.188372 LR 0.000500 Time 0.021378 -2022-12-06 11:20:52,116 - Epoch: [130][ 1000/ 1200] Overall Loss 0.188353 Objective Loss 0.188353 LR 0.000500 Time 0.021373 -2022-12-06 11:20:52,325 - Epoch: [130][ 1010/ 1200] Overall Loss 0.188149 Objective Loss 0.188149 LR 0.000500 Time 0.021368 -2022-12-06 11:20:52,535 - Epoch: [130][ 1020/ 1200] Overall Loss 0.188142 Objective Loss 0.188142 LR 0.000500 Time 0.021364 -2022-12-06 11:20:52,744 - Epoch: [130][ 1030/ 1200] Overall Loss 0.188344 Objective Loss 0.188344 LR 0.000500 Time 0.021359 -2022-12-06 11:20:52,953 - Epoch: [130][ 1040/ 1200] Overall Loss 0.188483 Objective Loss 0.188483 LR 0.000500 Time 0.021354 -2022-12-06 11:20:53,162 - Epoch: [130][ 1050/ 1200] Overall Loss 0.188556 Objective Loss 0.188556 LR 0.000500 Time 0.021349 -2022-12-06 11:20:53,371 - Epoch: [130][ 1060/ 1200] Overall Loss 0.188675 Objective Loss 0.188675 LR 0.000500 Time 0.021345 -2022-12-06 11:20:53,580 - Epoch: [130][ 1070/ 1200] Overall Loss 0.188783 Objective Loss 0.188783 LR 0.000500 Time 0.021340 -2022-12-06 11:20:53,789 - Epoch: [130][ 1080/ 1200] Overall Loss 0.188886 Objective Loss 0.188886 LR 0.000500 Time 0.021336 -2022-12-06 11:20:53,998 - Epoch: [130][ 1090/ 1200] Overall Loss 0.188887 Objective Loss 0.188887 LR 0.000500 Time 0.021331 -2022-12-06 11:20:54,208 - Epoch: [130][ 1100/ 1200] Overall Loss 0.189215 Objective Loss 0.189215 LR 0.000500 Time 0.021327 -2022-12-06 11:20:54,417 - Epoch: [130][ 1110/ 1200] Overall Loss 0.189180 Objective Loss 0.189180 LR 0.000500 Time 0.021322 -2022-12-06 11:20:54,626 - Epoch: [130][ 1120/ 1200] Overall Loss 0.189227 Objective Loss 0.189227 LR 0.000500 Time 0.021319 -2022-12-06 11:20:54,835 - Epoch: [130][ 1130/ 1200] Overall Loss 0.189287 Objective Loss 0.189287 LR 0.000500 Time 0.021314 -2022-12-06 11:20:55,044 - Epoch: [130][ 1140/ 1200] Overall Loss 0.189318 Objective Loss 0.189318 LR 0.000500 Time 0.021310 -2022-12-06 11:20:55,253 - Epoch: [130][ 1150/ 1200] Overall Loss 0.189467 Objective Loss 0.189467 LR 0.000500 Time 0.021306 -2022-12-06 11:20:55,462 - Epoch: [130][ 1160/ 1200] Overall Loss 0.189602 Objective Loss 0.189602 LR 0.000500 Time 0.021302 -2022-12-06 11:20:55,671 - Epoch: [130][ 1170/ 1200] Overall Loss 0.189491 Objective Loss 0.189491 LR 0.000500 Time 0.021298 -2022-12-06 11:20:55,880 - Epoch: [130][ 1180/ 1200] Overall Loss 0.189366 Objective Loss 0.189366 LR 0.000500 Time 0.021294 -2022-12-06 11:20:56,089 - Epoch: [130][ 1190/ 1200] Overall Loss 0.189531 Objective Loss 0.189531 LR 0.000500 Time 0.021290 -2022-12-06 11:20:56,317 - Epoch: [130][ 1200/ 1200] Overall Loss 0.189782 Objective Loss 0.189782 Top1 86.401674 Top5 98.326360 LR 0.000500 Time 0.021303 -2022-12-06 11:20:56,413 - --- validate (epoch=130)----------- -2022-12-06 11:20:56,414 - 34129 samples (256 per mini-batch) -2022-12-06 11:20:56,856 - Epoch: [130][ 10/ 134] Loss 0.262682 Top1 87.382812 Top5 98.359375 -2022-12-06 11:20:56,986 - Epoch: [130][ 20/ 134] Loss 0.258617 Top1 86.835938 Top5 98.300781 -2022-12-06 11:20:57,115 - Epoch: [130][ 30/ 134] Loss 0.263191 Top1 86.510417 Top5 98.268229 -2022-12-06 11:20:57,245 - Epoch: [130][ 40/ 134] Loss 0.258050 Top1 86.669922 Top5 98.349609 -2022-12-06 11:20:57,374 - Epoch: [130][ 50/ 134] Loss 0.261466 Top1 86.796875 Top5 98.351562 -2022-12-06 11:20:57,503 - Epoch: [130][ 60/ 134] Loss 0.264762 Top1 86.634115 Top5 98.346354 -2022-12-06 11:20:57,630 - Epoch: [130][ 70/ 134] Loss 0.261369 Top1 86.657366 Top5 98.337054 -2022-12-06 11:20:57,759 - Epoch: [130][ 80/ 134] Loss 0.259840 Top1 86.616211 Top5 98.354492 -2022-12-06 11:20:57,888 - Epoch: [130][ 90/ 134] Loss 0.257740 Top1 86.701389 Top5 98.337674 -2022-12-06 11:20:58,018 - Epoch: [130][ 100/ 134] Loss 0.258829 Top1 86.746094 Top5 98.355469 -2022-12-06 11:20:58,148 - Epoch: [130][ 110/ 134] Loss 0.258065 Top1 86.740057 Top5 98.394886 -2022-12-06 11:20:58,278 - Epoch: [130][ 120/ 134] Loss 0.260018 Top1 86.699219 Top5 98.398438 -2022-12-06 11:20:58,409 - Epoch: [130][ 130/ 134] Loss 0.258684 Top1 86.790865 Top5 98.380409 -2022-12-06 11:20:58,446 - Epoch: [130][ 134/ 134] Loss 0.255541 Top1 86.829383 Top5 98.394327 -2022-12-06 11:20:58,536 - ==> Top1: 86.829 Top5: 98.394 Loss: 0.256 - -2022-12-06 11:20:58,536 - ==> Confusion: -[[ 919 1 1 4 5 6 0 1 2 45 0 2 0 2 3 1 1 0 0 0 3] - [ 2 936 3 4 6 18 3 14 0 1 3 6 2 0 0 0 4 1 13 4 7] - [ 3 1 1018 14 5 1 18 9 0 1 3 2 2 0 2 3 2 3 5 3 8] - [ 0 1 17 956 1 1 0 0 0 2 8 0 4 2 10 0 0 1 12 0 5] - [ 12 5 1 0 951 6 0 1 0 6 4 3 0 2 13 4 6 1 2 1 2] - [ 1 8 0 1 5 1000 2 14 1 0 3 8 5 8 1 1 2 1 3 4 1] - [ 0 2 10 4 2 1 1078 2 0 0 0 2 0 1 0 5 0 1 2 8 0] - [ 0 6 12 2 2 35 10 938 1 0 0 6 0 2 0 1 0 0 21 15 3] - [ 8 3 0 1 0 3 1 1 947 51 18 2 1 8 13 0 2 1 1 1 2] - [ 56 0 2 0 3 5 0 6 20 884 1 3 0 8 2 1 0 2 0 1 7] - [ 1 2 4 9 2 2 2 2 6 1 956 1 2 8 2 1 2 0 7 3 6] - [ 2 0 0 1 1 19 3 5 0 1 0 968 22 3 1 5 1 6 0 10 3] - [ 0 0 1 1 0 4 0 1 0 2 0 28 903 0 0 9 0 10 0 6 4] - [ 2 0 2 0 0 11 0 4 7 19 4 3 4 946 1 2 2 0 0 5 11] - [ 3 4 1 26 3 5 0 0 10 4 1 1 3 4 1046 1 0 0 13 1 4] - [ 1 0 1 0 1 3 2 0 1 0 0 8 9 1 0 994 3 10 1 5 3] - [ 3 2 2 1 2 1 0 0 0 1 0 3 4 2 1 11 1031 0 0 5 3] - [ 2 0 1 3 1 2 0 2 0 5 0 6 14 1 1 13 1 980 0 2 2] - [ 5 2 1 8 1 3 0 20 1 1 6 3 3 0 4 1 0 0 941 2 6] - [ 0 4 1 1 1 7 4 3 0 1 2 15 6 5 1 3 0 3 2 1014 7] - [ 107 185 221 168 94 209 77 130 63 63 153 101 314 238 118 108 170 86 157 241 10223]] - -2022-12-06 11:20:59,116 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:20:59,116 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:20:59,122 - - -2022-12-06 11:20:59,122 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:21:00,060 - Epoch: [131][ 10/ 1200] Overall Loss 0.196747 Objective Loss 0.196747 LR 0.000500 Time 0.093696 -2022-12-06 11:21:00,254 - Epoch: [131][ 20/ 1200] Overall Loss 0.189589 Objective Loss 0.189589 LR 0.000500 Time 0.056507 -2022-12-06 11:21:00,446 - Epoch: [131][ 30/ 1200] Overall Loss 0.191551 Objective Loss 0.191551 LR 0.000500 Time 0.044051 -2022-12-06 11:21:00,637 - Epoch: [131][ 40/ 1200] Overall Loss 0.191688 Objective Loss 0.191688 LR 0.000500 Time 0.037799 -2022-12-06 11:21:00,827 - Epoch: [131][ 50/ 1200] Overall Loss 0.192589 Objective Loss 0.192589 LR 0.000500 Time 0.034046 -2022-12-06 11:21:01,019 - Epoch: [131][ 60/ 1200] Overall Loss 0.194345 Objective Loss 0.194345 LR 0.000500 Time 0.031555 -2022-12-06 11:21:01,211 - Epoch: [131][ 70/ 1200] Overall Loss 0.193713 Objective Loss 0.193713 LR 0.000500 Time 0.029776 -2022-12-06 11:21:01,402 - Epoch: [131][ 80/ 1200] Overall Loss 0.195307 Objective Loss 0.195307 LR 0.000500 Time 0.028435 -2022-12-06 11:21:01,593 - Epoch: [131][ 90/ 1200] Overall Loss 0.196227 Objective Loss 0.196227 LR 0.000500 Time 0.027397 -2022-12-06 11:21:01,784 - Epoch: [131][ 100/ 1200] Overall Loss 0.194365 Objective Loss 0.194365 LR 0.000500 Time 0.026560 -2022-12-06 11:21:01,976 - Epoch: [131][ 110/ 1200] Overall Loss 0.194132 Objective Loss 0.194132 LR 0.000500 Time 0.025882 -2022-12-06 11:21:02,166 - Epoch: [131][ 120/ 1200] Overall Loss 0.192874 Objective Loss 0.192874 LR 0.000500 Time 0.025311 -2022-12-06 11:21:02,358 - Epoch: [131][ 130/ 1200] Overall Loss 0.193703 Objective Loss 0.193703 LR 0.000500 Time 0.024831 -2022-12-06 11:21:02,549 - Epoch: [131][ 140/ 1200] Overall Loss 0.193701 Objective Loss 0.193701 LR 0.000500 Time 0.024419 -2022-12-06 11:21:02,739 - Epoch: [131][ 150/ 1200] Overall Loss 0.194213 Objective Loss 0.194213 LR 0.000500 Time 0.024058 -2022-12-06 11:21:02,931 - Epoch: [131][ 160/ 1200] Overall Loss 0.193725 Objective Loss 0.193725 LR 0.000500 Time 0.023750 -2022-12-06 11:21:03,123 - Epoch: [131][ 170/ 1200] Overall Loss 0.192654 Objective Loss 0.192654 LR 0.000500 Time 0.023475 -2022-12-06 11:21:03,313 - Epoch: [131][ 180/ 1200] Overall Loss 0.192801 Objective Loss 0.192801 LR 0.000500 Time 0.023227 -2022-12-06 11:21:03,504 - Epoch: [131][ 190/ 1200] Overall Loss 0.193053 Objective Loss 0.193053 LR 0.000500 Time 0.023006 -2022-12-06 11:21:03,695 - Epoch: [131][ 200/ 1200] Overall Loss 0.194053 Objective Loss 0.194053 LR 0.000500 Time 0.022809 -2022-12-06 11:21:03,887 - Epoch: [131][ 210/ 1200] Overall Loss 0.193610 Objective Loss 0.193610 LR 0.000500 Time 0.022631 -2022-12-06 11:21:04,078 - Epoch: [131][ 220/ 1200] Overall Loss 0.193522 Objective Loss 0.193522 LR 0.000500 Time 0.022469 -2022-12-06 11:21:04,269 - Epoch: [131][ 230/ 1200] Overall Loss 0.193400 Objective Loss 0.193400 LR 0.000500 Time 0.022322 -2022-12-06 11:21:04,460 - Epoch: [131][ 240/ 1200] Overall Loss 0.192967 Objective Loss 0.192967 LR 0.000500 Time 0.022184 -2022-12-06 11:21:04,651 - Epoch: [131][ 250/ 1200] Overall Loss 0.193157 Objective Loss 0.193157 LR 0.000500 Time 0.022058 -2022-12-06 11:21:04,841 - Epoch: [131][ 260/ 1200] Overall Loss 0.192096 Objective Loss 0.192096 LR 0.000500 Time 0.021940 -2022-12-06 11:21:05,032 - Epoch: [131][ 270/ 1200] Overall Loss 0.191680 Objective Loss 0.191680 LR 0.000500 Time 0.021834 -2022-12-06 11:21:05,224 - Epoch: [131][ 280/ 1200] Overall Loss 0.191415 Objective Loss 0.191415 LR 0.000500 Time 0.021737 -2022-12-06 11:21:05,416 - Epoch: [131][ 290/ 1200] Overall Loss 0.191169 Objective Loss 0.191169 LR 0.000500 Time 0.021647 -2022-12-06 11:21:05,608 - Epoch: [131][ 300/ 1200] Overall Loss 0.190703 Objective Loss 0.190703 LR 0.000500 Time 0.021563 -2022-12-06 11:21:05,799 - Epoch: [131][ 310/ 1200] Overall Loss 0.190588 Objective Loss 0.190588 LR 0.000500 Time 0.021482 -2022-12-06 11:21:05,990 - Epoch: [131][ 320/ 1200] Overall Loss 0.190716 Objective Loss 0.190716 LR 0.000500 Time 0.021407 -2022-12-06 11:21:06,182 - Epoch: [131][ 330/ 1200] Overall Loss 0.191094 Objective Loss 0.191094 LR 0.000500 Time 0.021337 -2022-12-06 11:21:06,372 - Epoch: [131][ 340/ 1200] Overall Loss 0.191506 Objective Loss 0.191506 LR 0.000500 Time 0.021268 -2022-12-06 11:21:06,564 - Epoch: [131][ 350/ 1200] Overall Loss 0.191141 Objective Loss 0.191141 LR 0.000500 Time 0.021207 -2022-12-06 11:21:06,756 - Epoch: [131][ 360/ 1200] Overall Loss 0.190587 Objective Loss 0.190587 LR 0.000500 Time 0.021149 -2022-12-06 11:21:06,947 - Epoch: [131][ 370/ 1200] Overall Loss 0.189754 Objective Loss 0.189754 LR 0.000500 Time 0.021092 -2022-12-06 11:21:07,138 - Epoch: [131][ 380/ 1200] Overall Loss 0.189483 Objective Loss 0.189483 LR 0.000500 Time 0.021040 -2022-12-06 11:21:07,330 - Epoch: [131][ 390/ 1200] Overall Loss 0.189174 Objective Loss 0.189174 LR 0.000500 Time 0.020991 -2022-12-06 11:21:07,521 - Epoch: [131][ 400/ 1200] Overall Loss 0.188931 Objective Loss 0.188931 LR 0.000500 Time 0.020942 -2022-12-06 11:21:07,712 - Epoch: [131][ 410/ 1200] Overall Loss 0.188954 Objective Loss 0.188954 LR 0.000500 Time 0.020896 -2022-12-06 11:21:07,904 - Epoch: [131][ 420/ 1200] Overall Loss 0.188859 Objective Loss 0.188859 LR 0.000500 Time 0.020853 -2022-12-06 11:21:08,095 - Epoch: [131][ 430/ 1200] Overall Loss 0.189125 Objective Loss 0.189125 LR 0.000500 Time 0.020812 -2022-12-06 11:21:08,287 - Epoch: [131][ 440/ 1200] Overall Loss 0.189147 Objective Loss 0.189147 LR 0.000500 Time 0.020774 -2022-12-06 11:21:08,479 - Epoch: [131][ 450/ 1200] Overall Loss 0.188930 Objective Loss 0.188930 LR 0.000500 Time 0.020737 -2022-12-06 11:21:08,671 - Epoch: [131][ 460/ 1200] Overall Loss 0.188862 Objective Loss 0.188862 LR 0.000500 Time 0.020702 -2022-12-06 11:21:08,862 - Epoch: [131][ 470/ 1200] Overall Loss 0.188431 Objective Loss 0.188431 LR 0.000500 Time 0.020668 -2022-12-06 11:21:09,055 - Epoch: [131][ 480/ 1200] Overall Loss 0.188674 Objective Loss 0.188674 LR 0.000500 Time 0.020637 -2022-12-06 11:21:09,247 - Epoch: [131][ 490/ 1200] Overall Loss 0.188795 Objective Loss 0.188795 LR 0.000500 Time 0.020607 -2022-12-06 11:21:09,439 - Epoch: [131][ 500/ 1200] Overall Loss 0.189159 Objective Loss 0.189159 LR 0.000500 Time 0.020577 -2022-12-06 11:21:09,630 - Epoch: [131][ 510/ 1200] Overall Loss 0.189135 Objective Loss 0.189135 LR 0.000500 Time 0.020548 -2022-12-06 11:21:09,822 - Epoch: [131][ 520/ 1200] Overall Loss 0.189057 Objective Loss 0.189057 LR 0.000500 Time 0.020520 -2022-12-06 11:21:10,013 - Epoch: [131][ 530/ 1200] Overall Loss 0.189343 Objective Loss 0.189343 LR 0.000500 Time 0.020492 -2022-12-06 11:21:10,204 - Epoch: [131][ 540/ 1200] Overall Loss 0.189522 Objective Loss 0.189522 LR 0.000500 Time 0.020466 -2022-12-06 11:21:10,396 - Epoch: [131][ 550/ 1200] Overall Loss 0.190083 Objective Loss 0.190083 LR 0.000500 Time 0.020442 -2022-12-06 11:21:10,587 - Epoch: [131][ 560/ 1200] Overall Loss 0.190462 Objective Loss 0.190462 LR 0.000500 Time 0.020417 -2022-12-06 11:21:10,779 - Epoch: [131][ 570/ 1200] Overall Loss 0.190695 Objective Loss 0.190695 LR 0.000500 Time 0.020394 -2022-12-06 11:21:10,970 - Epoch: [131][ 580/ 1200] Overall Loss 0.190942 Objective Loss 0.190942 LR 0.000500 Time 0.020371 -2022-12-06 11:21:11,161 - Epoch: [131][ 590/ 1200] Overall Loss 0.190645 Objective Loss 0.190645 LR 0.000500 Time 0.020349 -2022-12-06 11:21:11,352 - Epoch: [131][ 600/ 1200] Overall Loss 0.190676 Objective Loss 0.190676 LR 0.000500 Time 0.020327 -2022-12-06 11:21:11,543 - Epoch: [131][ 610/ 1200] Overall Loss 0.190758 Objective Loss 0.190758 LR 0.000500 Time 0.020306 -2022-12-06 11:21:11,735 - Epoch: [131][ 620/ 1200] Overall Loss 0.190883 Objective Loss 0.190883 LR 0.000500 Time 0.020287 -2022-12-06 11:21:11,927 - Epoch: [131][ 630/ 1200] Overall Loss 0.190392 Objective Loss 0.190392 LR 0.000500 Time 0.020269 -2022-12-06 11:21:12,118 - Epoch: [131][ 640/ 1200] Overall Loss 0.190348 Objective Loss 0.190348 LR 0.000500 Time 0.020250 -2022-12-06 11:21:12,309 - Epoch: [131][ 650/ 1200] Overall Loss 0.190220 Objective Loss 0.190220 LR 0.000500 Time 0.020232 -2022-12-06 11:21:12,500 - Epoch: [131][ 660/ 1200] Overall Loss 0.190201 Objective Loss 0.190201 LR 0.000500 Time 0.020214 -2022-12-06 11:21:12,692 - Epoch: [131][ 670/ 1200] Overall Loss 0.190273 Objective Loss 0.190273 LR 0.000500 Time 0.020198 -2022-12-06 11:21:12,884 - Epoch: [131][ 680/ 1200] Overall Loss 0.190532 Objective Loss 0.190532 LR 0.000500 Time 0.020182 -2022-12-06 11:21:13,075 - Epoch: [131][ 690/ 1200] Overall Loss 0.190680 Objective Loss 0.190680 LR 0.000500 Time 0.020166 -2022-12-06 11:21:13,267 - Epoch: [131][ 700/ 1200] Overall Loss 0.190840 Objective Loss 0.190840 LR 0.000500 Time 0.020152 -2022-12-06 11:21:13,458 - Epoch: [131][ 710/ 1200] Overall Loss 0.190760 Objective Loss 0.190760 LR 0.000500 Time 0.020136 -2022-12-06 11:21:13,650 - Epoch: [131][ 720/ 1200] Overall Loss 0.190769 Objective Loss 0.190769 LR 0.000500 Time 0.020122 -2022-12-06 11:21:13,842 - Epoch: [131][ 730/ 1200] Overall Loss 0.190731 Objective Loss 0.190731 LR 0.000500 Time 0.020108 -2022-12-06 11:21:14,033 - Epoch: [131][ 740/ 1200] Overall Loss 0.190809 Objective Loss 0.190809 LR 0.000500 Time 0.020094 -2022-12-06 11:21:14,224 - Epoch: [131][ 750/ 1200] Overall Loss 0.190546 Objective Loss 0.190546 LR 0.000500 Time 0.020080 -2022-12-06 11:21:14,415 - Epoch: [131][ 760/ 1200] Overall Loss 0.190587 Objective Loss 0.190587 LR 0.000500 Time 0.020067 -2022-12-06 11:21:14,606 - Epoch: [131][ 770/ 1200] Overall Loss 0.190769 Objective Loss 0.190769 LR 0.000500 Time 0.020054 -2022-12-06 11:21:14,798 - Epoch: [131][ 780/ 1200] Overall Loss 0.190429 Objective Loss 0.190429 LR 0.000500 Time 0.020042 -2022-12-06 11:21:14,990 - Epoch: [131][ 790/ 1200] Overall Loss 0.190714 Objective Loss 0.190714 LR 0.000500 Time 0.020030 -2022-12-06 11:21:15,181 - Epoch: [131][ 800/ 1200] Overall Loss 0.190802 Objective Loss 0.190802 LR 0.000500 Time 0.020018 -2022-12-06 11:21:15,373 - Epoch: [131][ 810/ 1200] Overall Loss 0.190802 Objective Loss 0.190802 LR 0.000500 Time 0.020007 -2022-12-06 11:21:15,564 - Epoch: [131][ 820/ 1200] Overall Loss 0.190596 Objective Loss 0.190596 LR 0.000500 Time 0.019996 -2022-12-06 11:21:15,757 - Epoch: [131][ 830/ 1200] Overall Loss 0.190498 Objective Loss 0.190498 LR 0.000500 Time 0.019986 -2022-12-06 11:21:15,948 - Epoch: [131][ 840/ 1200] Overall Loss 0.190298 Objective Loss 0.190298 LR 0.000500 Time 0.019975 -2022-12-06 11:21:16,141 - Epoch: [131][ 850/ 1200] Overall Loss 0.190087 Objective Loss 0.190087 LR 0.000500 Time 0.019966 -2022-12-06 11:21:16,332 - Epoch: [131][ 860/ 1200] Overall Loss 0.189965 Objective Loss 0.189965 LR 0.000500 Time 0.019956 -2022-12-06 11:21:16,524 - Epoch: [131][ 870/ 1200] Overall Loss 0.189738 Objective Loss 0.189738 LR 0.000500 Time 0.019947 -2022-12-06 11:21:16,716 - Epoch: [131][ 880/ 1200] Overall Loss 0.190157 Objective Loss 0.190157 LR 0.000500 Time 0.019938 -2022-12-06 11:21:16,909 - Epoch: [131][ 890/ 1200] Overall Loss 0.190229 Objective Loss 0.190229 LR 0.000500 Time 0.019930 -2022-12-06 11:21:17,101 - Epoch: [131][ 900/ 1200] Overall Loss 0.190105 Objective Loss 0.190105 LR 0.000500 Time 0.019920 -2022-12-06 11:21:17,292 - Epoch: [131][ 910/ 1200] Overall Loss 0.190209 Objective Loss 0.190209 LR 0.000500 Time 0.019912 -2022-12-06 11:21:17,484 - Epoch: [131][ 920/ 1200] Overall Loss 0.189918 Objective Loss 0.189918 LR 0.000500 Time 0.019903 -2022-12-06 11:21:17,676 - Epoch: [131][ 930/ 1200] Overall Loss 0.189862 Objective Loss 0.189862 LR 0.000500 Time 0.019895 -2022-12-06 11:21:17,868 - Epoch: [131][ 940/ 1200] Overall Loss 0.189635 Objective Loss 0.189635 LR 0.000500 Time 0.019887 -2022-12-06 11:21:18,059 - Epoch: [131][ 950/ 1200] Overall Loss 0.189667 Objective Loss 0.189667 LR 0.000500 Time 0.019878 -2022-12-06 11:21:18,250 - Epoch: [131][ 960/ 1200] Overall Loss 0.189642 Objective Loss 0.189642 LR 0.000500 Time 0.019870 -2022-12-06 11:21:18,442 - Epoch: [131][ 970/ 1200] Overall Loss 0.189645 Objective Loss 0.189645 LR 0.000500 Time 0.019862 -2022-12-06 11:21:18,634 - Epoch: [131][ 980/ 1200] Overall Loss 0.189432 Objective Loss 0.189432 LR 0.000500 Time 0.019855 -2022-12-06 11:21:18,826 - Epoch: [131][ 990/ 1200] Overall Loss 0.189691 Objective Loss 0.189691 LR 0.000500 Time 0.019847 -2022-12-06 11:21:19,017 - Epoch: [131][ 1000/ 1200] Overall Loss 0.189808 Objective Loss 0.189808 LR 0.000500 Time 0.019840 -2022-12-06 11:21:19,209 - Epoch: [131][ 1010/ 1200] Overall Loss 0.189867 Objective Loss 0.189867 LR 0.000500 Time 0.019833 -2022-12-06 11:21:19,400 - Epoch: [131][ 1020/ 1200] Overall Loss 0.189610 Objective Loss 0.189610 LR 0.000500 Time 0.019825 -2022-12-06 11:21:19,592 - Epoch: [131][ 1030/ 1200] Overall Loss 0.189444 Objective Loss 0.189444 LR 0.000500 Time 0.019818 -2022-12-06 11:21:19,783 - Epoch: [131][ 1040/ 1200] Overall Loss 0.189455 Objective Loss 0.189455 LR 0.000500 Time 0.019811 -2022-12-06 11:21:19,975 - Epoch: [131][ 1050/ 1200] Overall Loss 0.189375 Objective Loss 0.189375 LR 0.000500 Time 0.019804 -2022-12-06 11:21:20,167 - Epoch: [131][ 1060/ 1200] Overall Loss 0.189532 Objective Loss 0.189532 LR 0.000500 Time 0.019798 -2022-12-06 11:21:20,359 - Epoch: [131][ 1070/ 1200] Overall Loss 0.189551 Objective Loss 0.189551 LR 0.000500 Time 0.019792 -2022-12-06 11:21:20,550 - Epoch: [131][ 1080/ 1200] Overall Loss 0.189463 Objective Loss 0.189463 LR 0.000500 Time 0.019785 -2022-12-06 11:21:20,741 - Epoch: [131][ 1090/ 1200] Overall Loss 0.189373 Objective Loss 0.189373 LR 0.000500 Time 0.019779 -2022-12-06 11:21:20,932 - Epoch: [131][ 1100/ 1200] Overall Loss 0.189375 Objective Loss 0.189375 LR 0.000500 Time 0.019772 -2022-12-06 11:21:21,124 - Epoch: [131][ 1110/ 1200] Overall Loss 0.189599 Objective Loss 0.189599 LR 0.000500 Time 0.019766 -2022-12-06 11:21:21,316 - Epoch: [131][ 1120/ 1200] Overall Loss 0.189782 Objective Loss 0.189782 LR 0.000500 Time 0.019760 -2022-12-06 11:21:21,507 - Epoch: [131][ 1130/ 1200] Overall Loss 0.190053 Objective Loss 0.190053 LR 0.000500 Time 0.019754 -2022-12-06 11:21:21,698 - Epoch: [131][ 1140/ 1200] Overall Loss 0.189950 Objective Loss 0.189950 LR 0.000500 Time 0.019748 -2022-12-06 11:21:21,890 - Epoch: [131][ 1150/ 1200] Overall Loss 0.189876 Objective Loss 0.189876 LR 0.000500 Time 0.019742 -2022-12-06 11:21:22,081 - Epoch: [131][ 1160/ 1200] Overall Loss 0.189840 Objective Loss 0.189840 LR 0.000500 Time 0.019736 -2022-12-06 11:21:22,272 - Epoch: [131][ 1170/ 1200] Overall Loss 0.189912 Objective Loss 0.189912 LR 0.000500 Time 0.019731 -2022-12-06 11:21:22,463 - Epoch: [131][ 1180/ 1200] Overall Loss 0.189924 Objective Loss 0.189924 LR 0.000500 Time 0.019725 -2022-12-06 11:21:22,655 - Epoch: [131][ 1190/ 1200] Overall Loss 0.189939 Objective Loss 0.189939 LR 0.000500 Time 0.019720 -2022-12-06 11:21:22,879 - Epoch: [131][ 1200/ 1200] Overall Loss 0.190102 Objective Loss 0.190102 Top1 89.958159 Top5 99.372385 LR 0.000500 Time 0.019742 -2022-12-06 11:21:22,968 - --- validate (epoch=131)----------- -2022-12-06 11:21:22,969 - 34129 samples (256 per mini-batch) -2022-12-06 11:21:23,410 - Epoch: [131][ 10/ 134] Loss 0.256310 Top1 87.382812 Top5 98.437500 -2022-12-06 11:21:23,541 - Epoch: [131][ 20/ 134] Loss 0.279358 Top1 86.777344 Top5 98.027344 -2022-12-06 11:21:23,672 - Epoch: [131][ 30/ 134] Loss 0.254148 Top1 87.148438 Top5 98.125000 -2022-12-06 11:21:23,800 - Epoch: [131][ 40/ 134] Loss 0.261510 Top1 87.050781 Top5 98.007812 -2022-12-06 11:21:23,928 - Epoch: [131][ 50/ 134] Loss 0.260865 Top1 87.046875 Top5 98.148438 -2022-12-06 11:21:24,059 - Epoch: [131][ 60/ 134] Loss 0.259054 Top1 87.109375 Top5 98.209635 -2022-12-06 11:21:24,188 - Epoch: [131][ 70/ 134] Loss 0.254648 Top1 87.064732 Top5 98.286830 -2022-12-06 11:21:24,316 - Epoch: [131][ 80/ 134] Loss 0.256601 Top1 87.016602 Top5 98.271484 -2022-12-06 11:21:24,444 - Epoch: [131][ 90/ 134] Loss 0.260317 Top1 87.026910 Top5 98.216146 -2022-12-06 11:21:24,573 - Epoch: [131][ 100/ 134] Loss 0.258188 Top1 86.945312 Top5 98.234375 -2022-12-06 11:21:24,701 - Epoch: [131][ 110/ 134] Loss 0.256479 Top1 87.006392 Top5 98.263494 -2022-12-06 11:21:24,829 - Epoch: [131][ 120/ 134] Loss 0.256096 Top1 86.956380 Top5 98.310547 -2022-12-06 11:21:24,959 - Epoch: [131][ 130/ 134] Loss 0.252870 Top1 86.995192 Top5 98.326322 -2022-12-06 11:21:24,995 - Epoch: [131][ 134/ 134] Loss 0.252520 Top1 86.981746 Top5 98.332796 -2022-12-06 11:21:25,083 - ==> Top1: 86.982 Top5: 98.333 Loss: 0.253 - -2022-12-06 11:21:25,083 - ==> Confusion: -[[ 920 2 0 3 7 8 0 2 1 39 0 2 2 2 2 2 0 1 1 0 2] - [ 1 944 2 1 10 12 1 17 1 1 2 3 2 2 1 1 1 3 11 3 8] - [ 5 1 1006 8 7 2 17 13 1 2 7 6 1 1 5 1 0 2 4 4 10] - [ 2 2 15 941 3 1 0 0 0 0 12 0 6 0 12 0 2 6 8 0 10] - [ 12 4 1 0 958 2 1 0 1 5 1 2 2 3 9 6 6 1 0 0 6] - [ 3 15 1 3 12 958 1 19 1 2 2 17 2 15 2 1 2 0 4 4 5] - [ 1 3 8 5 2 1 1075 1 0 0 1 1 0 1 0 6 0 0 2 8 3] - [ 1 3 7 2 1 26 10 957 0 0 2 7 1 2 0 1 0 0 14 14 6] - [ 9 1 0 1 0 4 1 1 966 46 9 2 3 4 9 0 3 0 2 1 2] - [ 73 0 2 0 3 1 0 1 17 887 2 2 0 4 2 3 0 1 0 0 3] - [ 1 1 1 5 3 0 2 3 8 1 960 1 1 9 6 1 0 1 5 3 7] - [ 4 0 1 0 1 6 4 3 0 0 1 967 28 2 0 10 2 8 0 10 4] - [ 0 0 2 1 1 2 0 0 0 0 0 19 901 2 1 9 1 18 0 3 9] - [ 1 0 1 1 0 7 0 1 13 24 5 4 5 941 3 3 3 1 0 1 9] - [ 8 2 2 6 5 1 0 0 14 5 0 3 2 3 1067 1 0 2 6 0 3] - [ 0 0 2 1 2 0 3 0 0 0 0 5 9 2 0 1003 3 10 0 1 2] - [ 3 1 1 1 2 1 0 1 2 0 0 1 2 1 1 17 1026 0 0 5 7] - [ 2 1 2 1 2 1 0 0 0 3 0 2 11 1 1 15 1 990 0 2 1] - [ 4 4 0 9 0 2 1 25 1 1 2 1 6 0 9 0 1 2 931 3 6] - [ 1 4 3 2 0 4 2 6 0 0 2 16 7 4 1 4 3 4 0 1008 9] - [ 141 189 157 99 145 122 76 144 87 104 153 78 315 258 155 123 129 104 138 231 10278]] - -2022-12-06 11:21:25,741 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:21:25,741 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:21:25,747 - - -2022-12-06 11:21:25,747 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:21:26,678 - Epoch: [132][ 10/ 1200] Overall Loss 0.170769 Objective Loss 0.170769 LR 0.000500 Time 0.092990 -2022-12-06 11:21:26,876 - Epoch: [132][ 20/ 1200] Overall Loss 0.185142 Objective Loss 0.185142 LR 0.000500 Time 0.056336 -2022-12-06 11:21:27,066 - Epoch: [132][ 30/ 1200] Overall Loss 0.187377 Objective Loss 0.187377 LR 0.000500 Time 0.043900 -2022-12-06 11:21:27,258 - Epoch: [132][ 40/ 1200] Overall Loss 0.188981 Objective Loss 0.188981 LR 0.000500 Time 0.037692 -2022-12-06 11:21:27,448 - Epoch: [132][ 50/ 1200] Overall Loss 0.187504 Objective Loss 0.187504 LR 0.000500 Time 0.033953 -2022-12-06 11:21:27,639 - Epoch: [132][ 60/ 1200] Overall Loss 0.189579 Objective Loss 0.189579 LR 0.000500 Time 0.031466 -2022-12-06 11:21:27,830 - Epoch: [132][ 70/ 1200] Overall Loss 0.189174 Objective Loss 0.189174 LR 0.000500 Time 0.029686 -2022-12-06 11:21:28,021 - Epoch: [132][ 80/ 1200] Overall Loss 0.187961 Objective Loss 0.187961 LR 0.000500 Time 0.028361 -2022-12-06 11:21:28,211 - Epoch: [132][ 90/ 1200] Overall Loss 0.188656 Objective Loss 0.188656 LR 0.000500 Time 0.027318 -2022-12-06 11:21:28,402 - Epoch: [132][ 100/ 1200] Overall Loss 0.189610 Objective Loss 0.189610 LR 0.000500 Time 0.026492 -2022-12-06 11:21:28,593 - Epoch: [132][ 110/ 1200] Overall Loss 0.189522 Objective Loss 0.189522 LR 0.000500 Time 0.025813 -2022-12-06 11:21:28,785 - Epoch: [132][ 120/ 1200] Overall Loss 0.186293 Objective Loss 0.186293 LR 0.000500 Time 0.025252 -2022-12-06 11:21:28,976 - Epoch: [132][ 130/ 1200] Overall Loss 0.186323 Objective Loss 0.186323 LR 0.000500 Time 0.024776 -2022-12-06 11:21:29,166 - Epoch: [132][ 140/ 1200] Overall Loss 0.187980 Objective Loss 0.187980 LR 0.000500 Time 0.024361 -2022-12-06 11:21:29,357 - Epoch: [132][ 150/ 1200] Overall Loss 0.188553 Objective Loss 0.188553 LR 0.000500 Time 0.024004 -2022-12-06 11:21:29,547 - Epoch: [132][ 160/ 1200] Overall Loss 0.189519 Objective Loss 0.189519 LR 0.000500 Time 0.023691 -2022-12-06 11:21:29,737 - Epoch: [132][ 170/ 1200] Overall Loss 0.189014 Objective Loss 0.189014 LR 0.000500 Time 0.023413 -2022-12-06 11:21:29,928 - Epoch: [132][ 180/ 1200] Overall Loss 0.188184 Objective Loss 0.188184 LR 0.000500 Time 0.023167 -2022-12-06 11:21:30,118 - Epoch: [132][ 190/ 1200] Overall Loss 0.187875 Objective Loss 0.187875 LR 0.000500 Time 0.022944 -2022-12-06 11:21:30,308 - Epoch: [132][ 200/ 1200] Overall Loss 0.186982 Objective Loss 0.186982 LR 0.000500 Time 0.022749 -2022-12-06 11:21:30,499 - Epoch: [132][ 210/ 1200] Overall Loss 0.187245 Objective Loss 0.187245 LR 0.000500 Time 0.022573 -2022-12-06 11:21:30,690 - Epoch: [132][ 220/ 1200] Overall Loss 0.187959 Objective Loss 0.187959 LR 0.000500 Time 0.022411 -2022-12-06 11:21:30,881 - Epoch: [132][ 230/ 1200] Overall Loss 0.189149 Objective Loss 0.189149 LR 0.000500 Time 0.022263 -2022-12-06 11:21:31,073 - Epoch: [132][ 240/ 1200] Overall Loss 0.189452 Objective Loss 0.189452 LR 0.000500 Time 0.022132 -2022-12-06 11:21:31,263 - Epoch: [132][ 250/ 1200] Overall Loss 0.188741 Objective Loss 0.188741 LR 0.000500 Time 0.022008 -2022-12-06 11:21:31,455 - Epoch: [132][ 260/ 1200] Overall Loss 0.189210 Objective Loss 0.189210 LR 0.000500 Time 0.021895 -2022-12-06 11:21:31,645 - Epoch: [132][ 270/ 1200] Overall Loss 0.188366 Objective Loss 0.188366 LR 0.000500 Time 0.021788 -2022-12-06 11:21:31,836 - Epoch: [132][ 280/ 1200] Overall Loss 0.189241 Objective Loss 0.189241 LR 0.000500 Time 0.021690 -2022-12-06 11:21:32,027 - Epoch: [132][ 290/ 1200] Overall Loss 0.189541 Objective Loss 0.189541 LR 0.000500 Time 0.021596 -2022-12-06 11:21:32,218 - Epoch: [132][ 300/ 1200] Overall Loss 0.189302 Objective Loss 0.189302 LR 0.000500 Time 0.021513 -2022-12-06 11:21:32,409 - Epoch: [132][ 310/ 1200] Overall Loss 0.189023 Objective Loss 0.189023 LR 0.000500 Time 0.021432 -2022-12-06 11:21:32,600 - Epoch: [132][ 320/ 1200] Overall Loss 0.189137 Objective Loss 0.189137 LR 0.000500 Time 0.021358 -2022-12-06 11:21:32,791 - Epoch: [132][ 330/ 1200] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000500 Time 0.021288 -2022-12-06 11:21:32,982 - Epoch: [132][ 340/ 1200] Overall Loss 0.188845 Objective Loss 0.188845 LR 0.000500 Time 0.021221 -2022-12-06 11:21:33,172 - Epoch: [132][ 350/ 1200] Overall Loss 0.189147 Objective Loss 0.189147 LR 0.000500 Time 0.021158 -2022-12-06 11:21:33,363 - Epoch: [132][ 360/ 1200] Overall Loss 0.189092 Objective Loss 0.189092 LR 0.000500 Time 0.021098 -2022-12-06 11:21:33,553 - Epoch: [132][ 370/ 1200] Overall Loss 0.189204 Objective Loss 0.189204 LR 0.000500 Time 0.021041 -2022-12-06 11:21:33,744 - Epoch: [132][ 380/ 1200] Overall Loss 0.189544 Objective Loss 0.189544 LR 0.000500 Time 0.020989 -2022-12-06 11:21:33,935 - Epoch: [132][ 390/ 1200] Overall Loss 0.189499 Objective Loss 0.189499 LR 0.000500 Time 0.020939 -2022-12-06 11:21:34,127 - Epoch: [132][ 400/ 1200] Overall Loss 0.189793 Objective Loss 0.189793 LR 0.000500 Time 0.020893 -2022-12-06 11:21:34,317 - Epoch: [132][ 410/ 1200] Overall Loss 0.189690 Objective Loss 0.189690 LR 0.000500 Time 0.020846 -2022-12-06 11:21:34,507 - Epoch: [132][ 420/ 1200] Overall Loss 0.189670 Objective Loss 0.189670 LR 0.000500 Time 0.020801 -2022-12-06 11:21:34,697 - Epoch: [132][ 430/ 1200] Overall Loss 0.189180 Objective Loss 0.189180 LR 0.000500 Time 0.020758 -2022-12-06 11:21:34,888 - Epoch: [132][ 440/ 1200] Overall Loss 0.188902 Objective Loss 0.188902 LR 0.000500 Time 0.020719 -2022-12-06 11:21:35,079 - Epoch: [132][ 450/ 1200] Overall Loss 0.188695 Objective Loss 0.188695 LR 0.000500 Time 0.020682 -2022-12-06 11:21:35,270 - Epoch: [132][ 460/ 1200] Overall Loss 0.187983 Objective Loss 0.187983 LR 0.000500 Time 0.020647 -2022-12-06 11:21:35,461 - Epoch: [132][ 470/ 1200] Overall Loss 0.187620 Objective Loss 0.187620 LR 0.000500 Time 0.020612 -2022-12-06 11:21:35,653 - Epoch: [132][ 480/ 1200] Overall Loss 0.187647 Objective Loss 0.187647 LR 0.000500 Time 0.020581 -2022-12-06 11:21:35,844 - Epoch: [132][ 490/ 1200] Overall Loss 0.187886 Objective Loss 0.187886 LR 0.000500 Time 0.020550 -2022-12-06 11:21:36,035 - Epoch: [132][ 500/ 1200] Overall Loss 0.187814 Objective Loss 0.187814 LR 0.000500 Time 0.020519 -2022-12-06 11:21:36,225 - Epoch: [132][ 510/ 1200] Overall Loss 0.187727 Objective Loss 0.187727 LR 0.000500 Time 0.020489 -2022-12-06 11:21:36,416 - Epoch: [132][ 520/ 1200] Overall Loss 0.188123 Objective Loss 0.188123 LR 0.000500 Time 0.020461 -2022-12-06 11:21:36,607 - Epoch: [132][ 530/ 1200] Overall Loss 0.188532 Objective Loss 0.188532 LR 0.000500 Time 0.020434 -2022-12-06 11:21:36,798 - Epoch: [132][ 540/ 1200] Overall Loss 0.188414 Objective Loss 0.188414 LR 0.000500 Time 0.020409 -2022-12-06 11:21:36,989 - Epoch: [132][ 550/ 1200] Overall Loss 0.188628 Objective Loss 0.188628 LR 0.000500 Time 0.020384 -2022-12-06 11:21:37,180 - Epoch: [132][ 560/ 1200] Overall Loss 0.188858 Objective Loss 0.188858 LR 0.000500 Time 0.020360 -2022-12-06 11:21:37,371 - Epoch: [132][ 570/ 1200] Overall Loss 0.189082 Objective Loss 0.189082 LR 0.000500 Time 0.020336 -2022-12-06 11:21:37,562 - Epoch: [132][ 580/ 1200] Overall Loss 0.189533 Objective Loss 0.189533 LR 0.000500 Time 0.020315 -2022-12-06 11:21:37,753 - Epoch: [132][ 590/ 1200] Overall Loss 0.189180 Objective Loss 0.189180 LR 0.000500 Time 0.020294 -2022-12-06 11:21:37,945 - Epoch: [132][ 600/ 1200] Overall Loss 0.189140 Objective Loss 0.189140 LR 0.000500 Time 0.020274 -2022-12-06 11:21:38,136 - Epoch: [132][ 610/ 1200] Overall Loss 0.189268 Objective Loss 0.189268 LR 0.000500 Time 0.020253 -2022-12-06 11:21:38,327 - Epoch: [132][ 620/ 1200] Overall Loss 0.189302 Objective Loss 0.189302 LR 0.000500 Time 0.020234 -2022-12-06 11:21:38,517 - Epoch: [132][ 630/ 1200] Overall Loss 0.189113 Objective Loss 0.189113 LR 0.000500 Time 0.020214 -2022-12-06 11:21:38,707 - Epoch: [132][ 640/ 1200] Overall Loss 0.188931 Objective Loss 0.188931 LR 0.000500 Time 0.020194 -2022-12-06 11:21:38,897 - Epoch: [132][ 650/ 1200] Overall Loss 0.189022 Objective Loss 0.189022 LR 0.000500 Time 0.020175 -2022-12-06 11:21:39,088 - Epoch: [132][ 660/ 1200] Overall Loss 0.189204 Objective Loss 0.189204 LR 0.000500 Time 0.020158 -2022-12-06 11:21:39,278 - Epoch: [132][ 670/ 1200] Overall Loss 0.189556 Objective Loss 0.189556 LR 0.000500 Time 0.020140 -2022-12-06 11:21:39,469 - Epoch: [132][ 680/ 1200] Overall Loss 0.189774 Objective Loss 0.189774 LR 0.000500 Time 0.020124 -2022-12-06 11:21:39,660 - Epoch: [132][ 690/ 1200] Overall Loss 0.190324 Objective Loss 0.190324 LR 0.000500 Time 0.020108 -2022-12-06 11:21:39,851 - Epoch: [132][ 700/ 1200] Overall Loss 0.189701 Objective Loss 0.189701 LR 0.000500 Time 0.020093 -2022-12-06 11:21:40,041 - Epoch: [132][ 710/ 1200] Overall Loss 0.189817 Objective Loss 0.189817 LR 0.000500 Time 0.020077 -2022-12-06 11:21:40,232 - Epoch: [132][ 720/ 1200] Overall Loss 0.189733 Objective Loss 0.189733 LR 0.000500 Time 0.020062 -2022-12-06 11:21:40,423 - Epoch: [132][ 730/ 1200] Overall Loss 0.189407 Objective Loss 0.189407 LR 0.000500 Time 0.020048 -2022-12-06 11:21:40,615 - Epoch: [132][ 740/ 1200] Overall Loss 0.189779 Objective Loss 0.189779 LR 0.000500 Time 0.020036 -2022-12-06 11:21:40,805 - Epoch: [132][ 750/ 1200] Overall Loss 0.189982 Objective Loss 0.189982 LR 0.000500 Time 0.020022 -2022-12-06 11:21:40,997 - Epoch: [132][ 760/ 1200] Overall Loss 0.189985 Objective Loss 0.189985 LR 0.000500 Time 0.020010 -2022-12-06 11:21:41,187 - Epoch: [132][ 770/ 1200] Overall Loss 0.189601 Objective Loss 0.189601 LR 0.000500 Time 0.019997 -2022-12-06 11:21:41,379 - Epoch: [132][ 780/ 1200] Overall Loss 0.189536 Objective Loss 0.189536 LR 0.000500 Time 0.019985 -2022-12-06 11:21:41,569 - Epoch: [132][ 790/ 1200] Overall Loss 0.189847 Objective Loss 0.189847 LR 0.000500 Time 0.019972 -2022-12-06 11:21:41,760 - Epoch: [132][ 800/ 1200] Overall Loss 0.189884 Objective Loss 0.189884 LR 0.000500 Time 0.019960 -2022-12-06 11:21:41,951 - Epoch: [132][ 810/ 1200] Overall Loss 0.189829 Objective Loss 0.189829 LR 0.000500 Time 0.019950 -2022-12-06 11:21:42,142 - Epoch: [132][ 820/ 1200] Overall Loss 0.189952 Objective Loss 0.189952 LR 0.000500 Time 0.019939 -2022-12-06 11:21:42,332 - Epoch: [132][ 830/ 1200] Overall Loss 0.190376 Objective Loss 0.190376 LR 0.000500 Time 0.019926 -2022-12-06 11:21:42,523 - Epoch: [132][ 840/ 1200] Overall Loss 0.190353 Objective Loss 0.190353 LR 0.000500 Time 0.019916 -2022-12-06 11:21:42,714 - Epoch: [132][ 850/ 1200] Overall Loss 0.190131 Objective Loss 0.190131 LR 0.000500 Time 0.019905 -2022-12-06 11:21:42,905 - Epoch: [132][ 860/ 1200] Overall Loss 0.190202 Objective Loss 0.190202 LR 0.000500 Time 0.019895 -2022-12-06 11:21:43,096 - Epoch: [132][ 870/ 1200] Overall Loss 0.190174 Objective Loss 0.190174 LR 0.000500 Time 0.019885 -2022-12-06 11:21:43,287 - Epoch: [132][ 880/ 1200] Overall Loss 0.190195 Objective Loss 0.190195 LR 0.000500 Time 0.019876 -2022-12-06 11:21:43,477 - Epoch: [132][ 890/ 1200] Overall Loss 0.189789 Objective Loss 0.189789 LR 0.000500 Time 0.019866 -2022-12-06 11:21:43,668 - Epoch: [132][ 900/ 1200] Overall Loss 0.190006 Objective Loss 0.190006 LR 0.000500 Time 0.019857 -2022-12-06 11:21:43,859 - Epoch: [132][ 910/ 1200] Overall Loss 0.190284 Objective Loss 0.190284 LR 0.000500 Time 0.019848 -2022-12-06 11:21:44,051 - Epoch: [132][ 920/ 1200] Overall Loss 0.190330 Objective Loss 0.190330 LR 0.000500 Time 0.019840 -2022-12-06 11:21:44,242 - Epoch: [132][ 930/ 1200] Overall Loss 0.190267 Objective Loss 0.190267 LR 0.000500 Time 0.019831 -2022-12-06 11:21:44,432 - Epoch: [132][ 940/ 1200] Overall Loss 0.190630 Objective Loss 0.190630 LR 0.000500 Time 0.019822 -2022-12-06 11:21:44,623 - Epoch: [132][ 950/ 1200] Overall Loss 0.190783 Objective Loss 0.190783 LR 0.000500 Time 0.019814 -2022-12-06 11:21:44,814 - Epoch: [132][ 960/ 1200] Overall Loss 0.190744 Objective Loss 0.190744 LR 0.000500 Time 0.019806 -2022-12-06 11:21:45,005 - Epoch: [132][ 970/ 1200] Overall Loss 0.190667 Objective Loss 0.190667 LR 0.000500 Time 0.019798 -2022-12-06 11:21:45,196 - Epoch: [132][ 980/ 1200] Overall Loss 0.190723 Objective Loss 0.190723 LR 0.000500 Time 0.019791 -2022-12-06 11:21:45,388 - Epoch: [132][ 990/ 1200] Overall Loss 0.190612 Objective Loss 0.190612 LR 0.000500 Time 0.019784 -2022-12-06 11:21:45,579 - Epoch: [132][ 1000/ 1200] Overall Loss 0.190733 Objective Loss 0.190733 LR 0.000500 Time 0.019776 -2022-12-06 11:21:45,770 - Epoch: [132][ 1010/ 1200] Overall Loss 0.190650 Objective Loss 0.190650 LR 0.000500 Time 0.019769 -2022-12-06 11:21:45,961 - Epoch: [132][ 1020/ 1200] Overall Loss 0.190773 Objective Loss 0.190773 LR 0.000500 Time 0.019762 -2022-12-06 11:21:46,152 - Epoch: [132][ 1030/ 1200] Overall Loss 0.190679 Objective Loss 0.190679 LR 0.000500 Time 0.019756 -2022-12-06 11:21:46,344 - Epoch: [132][ 1040/ 1200] Overall Loss 0.190632 Objective Loss 0.190632 LR 0.000500 Time 0.019749 -2022-12-06 11:21:46,535 - Epoch: [132][ 1050/ 1200] Overall Loss 0.190613 Objective Loss 0.190613 LR 0.000500 Time 0.019743 -2022-12-06 11:21:46,726 - Epoch: [132][ 1060/ 1200] Overall Loss 0.190468 Objective Loss 0.190468 LR 0.000500 Time 0.019736 -2022-12-06 11:21:46,917 - Epoch: [132][ 1070/ 1200] Overall Loss 0.190528 Objective Loss 0.190528 LR 0.000500 Time 0.019730 -2022-12-06 11:21:47,108 - Epoch: [132][ 1080/ 1200] Overall Loss 0.190602 Objective Loss 0.190602 LR 0.000500 Time 0.019723 -2022-12-06 11:21:47,299 - Epoch: [132][ 1090/ 1200] Overall Loss 0.190483 Objective Loss 0.190483 LR 0.000500 Time 0.019717 -2022-12-06 11:21:47,490 - Epoch: [132][ 1100/ 1200] Overall Loss 0.190319 Objective Loss 0.190319 LR 0.000500 Time 0.019711 -2022-12-06 11:21:47,681 - Epoch: [132][ 1110/ 1200] Overall Loss 0.190180 Objective Loss 0.190180 LR 0.000500 Time 0.019705 -2022-12-06 11:21:47,873 - Epoch: [132][ 1120/ 1200] Overall Loss 0.190203 Objective Loss 0.190203 LR 0.000500 Time 0.019700 -2022-12-06 11:21:48,064 - Epoch: [132][ 1130/ 1200] Overall Loss 0.190272 Objective Loss 0.190272 LR 0.000500 Time 0.019694 -2022-12-06 11:21:48,255 - Epoch: [132][ 1140/ 1200] Overall Loss 0.190343 Objective Loss 0.190343 LR 0.000500 Time 0.019689 -2022-12-06 11:21:48,446 - Epoch: [132][ 1150/ 1200] Overall Loss 0.190225 Objective Loss 0.190225 LR 0.000500 Time 0.019682 -2022-12-06 11:21:48,637 - Epoch: [132][ 1160/ 1200] Overall Loss 0.190273 Objective Loss 0.190273 LR 0.000500 Time 0.019677 -2022-12-06 11:21:48,827 - Epoch: [132][ 1170/ 1200] Overall Loss 0.190322 Objective Loss 0.190322 LR 0.000500 Time 0.019671 -2022-12-06 11:21:49,019 - Epoch: [132][ 1180/ 1200] Overall Loss 0.190225 Objective Loss 0.190225 LR 0.000500 Time 0.019666 -2022-12-06 11:21:49,209 - Epoch: [132][ 1190/ 1200] Overall Loss 0.190066 Objective Loss 0.190066 LR 0.000500 Time 0.019661 -2022-12-06 11:21:49,439 - Epoch: [132][ 1200/ 1200] Overall Loss 0.190207 Objective Loss 0.190207 Top1 87.029289 Top5 99.372385 LR 0.000500 Time 0.019688 -2022-12-06 11:21:49,527 - --- validate (epoch=132)----------- -2022-12-06 11:21:49,527 - 34129 samples (256 per mini-batch) -2022-12-06 11:21:49,981 - Epoch: [132][ 10/ 134] Loss 0.220703 Top1 87.656250 Top5 98.398438 -2022-12-06 11:21:50,125 - Epoch: [132][ 20/ 134] Loss 0.233339 Top1 87.207031 Top5 98.417969 -2022-12-06 11:21:50,253 - Epoch: [132][ 30/ 134] Loss 0.235806 Top1 86.992188 Top5 98.515625 -2022-12-06 11:21:50,385 - Epoch: [132][ 40/ 134] Loss 0.240752 Top1 86.826172 Top5 98.505859 -2022-12-06 11:21:50,516 - Epoch: [132][ 50/ 134] Loss 0.244205 Top1 86.914062 Top5 98.445312 -2022-12-06 11:21:50,648 - Epoch: [132][ 60/ 134] Loss 0.242610 Top1 86.861979 Top5 98.463542 -2022-12-06 11:21:50,777 - Epoch: [132][ 70/ 134] Loss 0.244151 Top1 86.813616 Top5 98.415179 -2022-12-06 11:21:50,909 - Epoch: [132][ 80/ 134] Loss 0.248127 Top1 86.899414 Top5 98.388672 -2022-12-06 11:21:51,040 - Epoch: [132][ 90/ 134] Loss 0.251535 Top1 86.822917 Top5 98.385417 -2022-12-06 11:21:51,173 - Epoch: [132][ 100/ 134] Loss 0.253839 Top1 86.824219 Top5 98.332031 -2022-12-06 11:21:51,303 - Epoch: [132][ 110/ 134] Loss 0.252572 Top1 86.871449 Top5 98.348722 -2022-12-06 11:21:51,437 - Epoch: [132][ 120/ 134] Loss 0.251525 Top1 86.897786 Top5 98.352865 -2022-12-06 11:21:51,570 - Epoch: [132][ 130/ 134] Loss 0.250948 Top1 86.884014 Top5 98.359375 -2022-12-06 11:21:51,610 - Epoch: [132][ 134/ 134] Loss 0.250063 Top1 86.905564 Top5 98.356237 -2022-12-06 11:21:51,698 - ==> Top1: 86.906 Top5: 98.356 Loss: 0.250 - -2022-12-06 11:21:51,698 - ==> Confusion: -[[ 903 3 0 4 6 4 1 0 1 51 0 2 1 4 6 3 1 0 1 0 5] - [ 0 940 3 2 6 19 3 15 2 0 3 4 1 2 0 1 3 2 14 2 5] - [ 1 3 1013 11 6 2 14 13 1 2 6 5 3 4 3 2 0 0 3 1 10] - [ 2 3 15 946 1 2 0 0 0 1 11 0 5 3 9 0 0 1 15 0 6] - [ 8 7 2 0 957 4 0 0 2 7 1 3 0 1 8 6 6 2 1 1 4] - [ 3 11 0 3 6 978 3 14 5 2 1 10 4 16 3 1 1 0 2 4 2] - [ 0 3 12 2 0 4 1072 1 0 0 0 3 1 0 0 4 0 4 1 10 1] - [ 1 8 5 1 3 23 12 950 0 0 2 6 2 3 0 1 0 0 17 15 5] - [ 6 3 0 0 0 5 1 0 969 39 13 2 3 5 9 0 1 2 3 1 2] - [ 54 1 1 0 2 3 0 3 22 894 1 2 1 6 2 2 0 1 0 0 6] - [ 0 1 4 4 2 1 2 2 5 1 966 2 1 15 4 0 0 0 6 1 2] - [ 2 0 2 0 0 9 2 2 0 0 1 970 27 7 1 5 3 5 0 10 5] - [ 1 1 1 3 0 4 0 1 1 0 0 29 906 1 0 4 1 9 1 3 3] - [ 0 1 0 0 0 8 0 1 11 13 7 9 5 952 1 1 3 0 1 2 8] - [ 3 6 1 12 2 4 0 0 13 2 2 1 5 4 1059 0 2 1 10 0 3] - [ 0 1 2 0 3 0 3 0 1 0 1 6 7 2 0 999 3 11 0 2 2] - [ 1 2 2 1 3 0 0 1 1 0 0 5 3 0 1 21 1018 0 0 6 7] - [ 2 1 3 0 0 1 0 1 1 2 0 5 26 1 2 10 0 975 1 2 3] - [ 3 6 2 8 1 3 0 20 2 1 2 2 4 0 6 0 0 1 941 2 4] - [ 2 5 2 1 0 3 4 4 1 1 2 15 10 6 0 5 1 1 0 1009 8] - [ 89 216 173 131 91 178 86 141 78 85 161 98 348 286 137 134 108 69 179 200 10238]] - -2022-12-06 11:21:52,377 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:21:52,378 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:21:52,384 - - -2022-12-06 11:21:52,384 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:21:53,336 - Epoch: [133][ 10/ 1200] Overall Loss 0.172567 Objective Loss 0.172567 LR 0.000500 Time 0.095191 -2022-12-06 11:21:53,540 - Epoch: [133][ 20/ 1200] Overall Loss 0.191522 Objective Loss 0.191522 LR 0.000500 Time 0.057772 -2022-12-06 11:21:53,735 - Epoch: [133][ 30/ 1200] Overall Loss 0.186274 Objective Loss 0.186274 LR 0.000500 Time 0.044998 -2022-12-06 11:21:53,930 - Epoch: [133][ 40/ 1200] Overall Loss 0.186240 Objective Loss 0.186240 LR 0.000500 Time 0.038599 -2022-12-06 11:21:54,125 - Epoch: [133][ 50/ 1200] Overall Loss 0.184248 Objective Loss 0.184248 LR 0.000500 Time 0.034775 -2022-12-06 11:21:54,320 - Epoch: [133][ 60/ 1200] Overall Loss 0.180727 Objective Loss 0.180727 LR 0.000500 Time 0.032211 -2022-12-06 11:21:54,514 - Epoch: [133][ 70/ 1200] Overall Loss 0.181310 Objective Loss 0.181310 LR 0.000500 Time 0.030377 -2022-12-06 11:21:54,709 - Epoch: [133][ 80/ 1200] Overall Loss 0.181766 Objective Loss 0.181766 LR 0.000500 Time 0.029012 -2022-12-06 11:21:54,904 - Epoch: [133][ 90/ 1200] Overall Loss 0.180535 Objective Loss 0.180535 LR 0.000500 Time 0.027943 -2022-12-06 11:21:55,098 - Epoch: [133][ 100/ 1200] Overall Loss 0.179873 Objective Loss 0.179873 LR 0.000500 Time 0.027088 -2022-12-06 11:21:55,293 - Epoch: [133][ 110/ 1200] Overall Loss 0.178974 Objective Loss 0.178974 LR 0.000500 Time 0.026390 -2022-12-06 11:21:55,488 - Epoch: [133][ 120/ 1200] Overall Loss 0.181376 Objective Loss 0.181376 LR 0.000500 Time 0.025811 -2022-12-06 11:21:55,683 - Epoch: [133][ 130/ 1200] Overall Loss 0.182298 Objective Loss 0.182298 LR 0.000500 Time 0.025322 -2022-12-06 11:21:55,878 - Epoch: [133][ 140/ 1200] Overall Loss 0.182534 Objective Loss 0.182534 LR 0.000500 Time 0.024902 -2022-12-06 11:21:56,072 - Epoch: [133][ 150/ 1200] Overall Loss 0.183918 Objective Loss 0.183918 LR 0.000500 Time 0.024536 -2022-12-06 11:21:56,267 - Epoch: [133][ 160/ 1200] Overall Loss 0.185494 Objective Loss 0.185494 LR 0.000500 Time 0.024214 -2022-12-06 11:21:56,461 - Epoch: [133][ 170/ 1200] Overall Loss 0.184983 Objective Loss 0.184983 LR 0.000500 Time 0.023931 -2022-12-06 11:21:56,656 - Epoch: [133][ 180/ 1200] Overall Loss 0.185665 Objective Loss 0.185665 LR 0.000500 Time 0.023680 -2022-12-06 11:21:56,851 - Epoch: [133][ 190/ 1200] Overall Loss 0.185265 Objective Loss 0.185265 LR 0.000500 Time 0.023454 -2022-12-06 11:21:57,045 - Epoch: [133][ 200/ 1200] Overall Loss 0.185381 Objective Loss 0.185381 LR 0.000500 Time 0.023251 -2022-12-06 11:21:57,240 - Epoch: [133][ 210/ 1200] Overall Loss 0.185926 Objective Loss 0.185926 LR 0.000500 Time 0.023068 -2022-12-06 11:21:57,435 - Epoch: [133][ 220/ 1200] Overall Loss 0.187486 Objective Loss 0.187486 LR 0.000500 Time 0.022904 -2022-12-06 11:21:57,629 - Epoch: [133][ 230/ 1200] Overall Loss 0.187364 Objective Loss 0.187364 LR 0.000500 Time 0.022749 -2022-12-06 11:21:57,823 - Epoch: [133][ 240/ 1200] Overall Loss 0.187579 Objective Loss 0.187579 LR 0.000500 Time 0.022610 -2022-12-06 11:21:58,018 - Epoch: [133][ 250/ 1200] Overall Loss 0.187336 Objective Loss 0.187336 LR 0.000500 Time 0.022480 -2022-12-06 11:21:58,212 - Epoch: [133][ 260/ 1200] Overall Loss 0.187131 Objective Loss 0.187131 LR 0.000500 Time 0.022362 -2022-12-06 11:21:58,407 - Epoch: [133][ 270/ 1200] Overall Loss 0.187497 Objective Loss 0.187497 LR 0.000500 Time 0.022252 -2022-12-06 11:21:58,602 - Epoch: [133][ 280/ 1200] Overall Loss 0.187801 Objective Loss 0.187801 LR 0.000500 Time 0.022152 -2022-12-06 11:21:58,796 - Epoch: [133][ 290/ 1200] Overall Loss 0.187295 Objective Loss 0.187295 LR 0.000500 Time 0.022057 -2022-12-06 11:21:58,991 - Epoch: [133][ 300/ 1200] Overall Loss 0.187559 Objective Loss 0.187559 LR 0.000500 Time 0.021968 -2022-12-06 11:21:59,185 - Epoch: [133][ 310/ 1200] Overall Loss 0.187049 Objective Loss 0.187049 LR 0.000500 Time 0.021885 -2022-12-06 11:21:59,379 - Epoch: [133][ 320/ 1200] Overall Loss 0.186738 Objective Loss 0.186738 LR 0.000500 Time 0.021806 -2022-12-06 11:21:59,574 - Epoch: [133][ 330/ 1200] Overall Loss 0.187058 Objective Loss 0.187058 LR 0.000500 Time 0.021733 -2022-12-06 11:21:59,769 - Epoch: [133][ 340/ 1200] Overall Loss 0.187173 Objective Loss 0.187173 LR 0.000500 Time 0.021665 -2022-12-06 11:21:59,963 - Epoch: [133][ 350/ 1200] Overall Loss 0.186952 Objective Loss 0.186952 LR 0.000500 Time 0.021599 -2022-12-06 11:22:00,157 - Epoch: [133][ 360/ 1200] Overall Loss 0.187116 Objective Loss 0.187116 LR 0.000500 Time 0.021537 -2022-12-06 11:22:00,351 - Epoch: [133][ 370/ 1200] Overall Loss 0.186493 Objective Loss 0.186493 LR 0.000500 Time 0.021479 -2022-12-06 11:22:00,546 - Epoch: [133][ 380/ 1200] Overall Loss 0.186122 Objective Loss 0.186122 LR 0.000500 Time 0.021424 -2022-12-06 11:22:00,740 - Epoch: [133][ 390/ 1200] Overall Loss 0.186417 Objective Loss 0.186417 LR 0.000500 Time 0.021371 -2022-12-06 11:22:00,935 - Epoch: [133][ 400/ 1200] Overall Loss 0.186766 Objective Loss 0.186766 LR 0.000500 Time 0.021323 -2022-12-06 11:22:01,130 - Epoch: [133][ 410/ 1200] Overall Loss 0.186767 Objective Loss 0.186767 LR 0.000500 Time 0.021276 -2022-12-06 11:22:01,324 - Epoch: [133][ 420/ 1200] Overall Loss 0.186801 Objective Loss 0.186801 LR 0.000500 Time 0.021230 -2022-12-06 11:22:01,519 - Epoch: [133][ 430/ 1200] Overall Loss 0.186748 Objective Loss 0.186748 LR 0.000500 Time 0.021188 -2022-12-06 11:22:01,712 - Epoch: [133][ 440/ 1200] Overall Loss 0.186518 Objective Loss 0.186518 LR 0.000500 Time 0.021145 -2022-12-06 11:22:01,905 - Epoch: [133][ 450/ 1200] Overall Loss 0.186373 Objective Loss 0.186373 LR 0.000500 Time 0.021103 -2022-12-06 11:22:02,098 - Epoch: [133][ 460/ 1200] Overall Loss 0.186081 Objective Loss 0.186081 LR 0.000500 Time 0.021063 -2022-12-06 11:22:02,292 - Epoch: [133][ 470/ 1200] Overall Loss 0.186339 Objective Loss 0.186339 LR 0.000500 Time 0.021025 -2022-12-06 11:22:02,484 - Epoch: [133][ 480/ 1200] Overall Loss 0.186222 Objective Loss 0.186222 LR 0.000500 Time 0.020987 -2022-12-06 11:22:02,677 - Epoch: [133][ 490/ 1200] Overall Loss 0.186029 Objective Loss 0.186029 LR 0.000500 Time 0.020952 -2022-12-06 11:22:02,870 - Epoch: [133][ 500/ 1200] Overall Loss 0.186043 Objective Loss 0.186043 LR 0.000500 Time 0.020918 -2022-12-06 11:22:03,063 - Epoch: [133][ 510/ 1200] Overall Loss 0.186317 Objective Loss 0.186317 LR 0.000500 Time 0.020884 -2022-12-06 11:22:03,256 - Epoch: [133][ 520/ 1200] Overall Loss 0.186769 Objective Loss 0.186769 LR 0.000500 Time 0.020853 -2022-12-06 11:22:03,449 - Epoch: [133][ 530/ 1200] Overall Loss 0.186941 Objective Loss 0.186941 LR 0.000500 Time 0.020823 -2022-12-06 11:22:03,643 - Epoch: [133][ 540/ 1200] Overall Loss 0.186661 Objective Loss 0.186661 LR 0.000500 Time 0.020795 -2022-12-06 11:22:03,835 - Epoch: [133][ 550/ 1200] Overall Loss 0.186409 Objective Loss 0.186409 LR 0.000500 Time 0.020765 -2022-12-06 11:22:04,029 - Epoch: [133][ 560/ 1200] Overall Loss 0.186251 Objective Loss 0.186251 LR 0.000500 Time 0.020739 -2022-12-06 11:22:04,222 - Epoch: [133][ 570/ 1200] Overall Loss 0.186575 Objective Loss 0.186575 LR 0.000500 Time 0.020713 -2022-12-06 11:22:04,414 - Epoch: [133][ 580/ 1200] Overall Loss 0.187147 Objective Loss 0.187147 LR 0.000500 Time 0.020687 -2022-12-06 11:22:04,607 - Epoch: [133][ 590/ 1200] Overall Loss 0.186713 Objective Loss 0.186713 LR 0.000500 Time 0.020662 -2022-12-06 11:22:04,803 - Epoch: [133][ 600/ 1200] Overall Loss 0.186537 Objective Loss 0.186537 LR 0.000500 Time 0.020643 -2022-12-06 11:22:04,998 - Epoch: [133][ 610/ 1200] Overall Loss 0.186952 Objective Loss 0.186952 LR 0.000500 Time 0.020623 -2022-12-06 11:22:05,193 - Epoch: [133][ 620/ 1200] Overall Loss 0.186781 Objective Loss 0.186781 LR 0.000500 Time 0.020604 -2022-12-06 11:22:05,388 - Epoch: [133][ 630/ 1200] Overall Loss 0.186694 Objective Loss 0.186694 LR 0.000500 Time 0.020586 -2022-12-06 11:22:05,583 - Epoch: [133][ 640/ 1200] Overall Loss 0.186434 Objective Loss 0.186434 LR 0.000500 Time 0.020568 -2022-12-06 11:22:05,778 - Epoch: [133][ 650/ 1200] Overall Loss 0.186401 Objective Loss 0.186401 LR 0.000500 Time 0.020551 -2022-12-06 11:22:05,973 - Epoch: [133][ 660/ 1200] Overall Loss 0.186242 Objective Loss 0.186242 LR 0.000500 Time 0.020535 -2022-12-06 11:22:06,167 - Epoch: [133][ 670/ 1200] Overall Loss 0.186065 Objective Loss 0.186065 LR 0.000500 Time 0.020517 -2022-12-06 11:22:06,362 - Epoch: [133][ 680/ 1200] Overall Loss 0.185950 Objective Loss 0.185950 LR 0.000500 Time 0.020501 -2022-12-06 11:22:06,557 - Epoch: [133][ 690/ 1200] Overall Loss 0.185912 Objective Loss 0.185912 LR 0.000500 Time 0.020486 -2022-12-06 11:22:06,752 - Epoch: [133][ 700/ 1200] Overall Loss 0.185948 Objective Loss 0.185948 LR 0.000500 Time 0.020470 -2022-12-06 11:22:06,946 - Epoch: [133][ 710/ 1200] Overall Loss 0.185703 Objective Loss 0.185703 LR 0.000500 Time 0.020455 -2022-12-06 11:22:07,141 - Epoch: [133][ 720/ 1200] Overall Loss 0.185562 Objective Loss 0.185562 LR 0.000500 Time 0.020441 -2022-12-06 11:22:07,336 - Epoch: [133][ 730/ 1200] Overall Loss 0.185364 Objective Loss 0.185364 LR 0.000500 Time 0.020427 -2022-12-06 11:22:07,531 - Epoch: [133][ 740/ 1200] Overall Loss 0.185645 Objective Loss 0.185645 LR 0.000500 Time 0.020413 -2022-12-06 11:22:07,725 - Epoch: [133][ 750/ 1200] Overall Loss 0.185624 Objective Loss 0.185624 LR 0.000500 Time 0.020400 -2022-12-06 11:22:07,920 - Epoch: [133][ 760/ 1200] Overall Loss 0.185736 Objective Loss 0.185736 LR 0.000500 Time 0.020387 -2022-12-06 11:22:08,115 - Epoch: [133][ 770/ 1200] Overall Loss 0.185846 Objective Loss 0.185846 LR 0.000500 Time 0.020375 -2022-12-06 11:22:08,310 - Epoch: [133][ 780/ 1200] Overall Loss 0.185733 Objective Loss 0.185733 LR 0.000500 Time 0.020363 -2022-12-06 11:22:08,505 - Epoch: [133][ 790/ 1200] Overall Loss 0.186011 Objective Loss 0.186011 LR 0.000500 Time 0.020351 -2022-12-06 11:22:08,700 - Epoch: [133][ 800/ 1200] Overall Loss 0.186119 Objective Loss 0.186119 LR 0.000500 Time 0.020340 -2022-12-06 11:22:08,895 - Epoch: [133][ 810/ 1200] Overall Loss 0.186127 Objective Loss 0.186127 LR 0.000500 Time 0.020329 -2022-12-06 11:22:09,090 - Epoch: [133][ 820/ 1200] Overall Loss 0.186080 Objective Loss 0.186080 LR 0.000500 Time 0.020318 -2022-12-06 11:22:09,285 - Epoch: [133][ 830/ 1200] Overall Loss 0.186136 Objective Loss 0.186136 LR 0.000500 Time 0.020307 -2022-12-06 11:22:09,480 - Epoch: [133][ 840/ 1200] Overall Loss 0.186027 Objective Loss 0.186027 LR 0.000500 Time 0.020297 -2022-12-06 11:22:09,675 - Epoch: [133][ 850/ 1200] Overall Loss 0.186064 Objective Loss 0.186064 LR 0.000500 Time 0.020287 -2022-12-06 11:22:09,869 - Epoch: [133][ 860/ 1200] Overall Loss 0.186087 Objective Loss 0.186087 LR 0.000500 Time 0.020276 -2022-12-06 11:22:10,064 - Epoch: [133][ 870/ 1200] Overall Loss 0.186428 Objective Loss 0.186428 LR 0.000500 Time 0.020267 -2022-12-06 11:22:10,259 - Epoch: [133][ 880/ 1200] Overall Loss 0.186390 Objective Loss 0.186390 LR 0.000500 Time 0.020257 -2022-12-06 11:22:10,453 - Epoch: [133][ 890/ 1200] Overall Loss 0.186417 Objective Loss 0.186417 LR 0.000500 Time 0.020247 -2022-12-06 11:22:10,649 - Epoch: [133][ 900/ 1200] Overall Loss 0.186434 Objective Loss 0.186434 LR 0.000500 Time 0.020239 -2022-12-06 11:22:10,844 - Epoch: [133][ 910/ 1200] Overall Loss 0.186611 Objective Loss 0.186611 LR 0.000500 Time 0.020230 -2022-12-06 11:22:11,039 - Epoch: [133][ 920/ 1200] Overall Loss 0.186452 Objective Loss 0.186452 LR 0.000500 Time 0.020222 -2022-12-06 11:22:11,234 - Epoch: [133][ 930/ 1200] Overall Loss 0.186635 Objective Loss 0.186635 LR 0.000500 Time 0.020214 -2022-12-06 11:22:11,429 - Epoch: [133][ 940/ 1200] Overall Loss 0.186742 Objective Loss 0.186742 LR 0.000500 Time 0.020206 -2022-12-06 11:22:11,624 - Epoch: [133][ 950/ 1200] Overall Loss 0.186655 Objective Loss 0.186655 LR 0.000500 Time 0.020197 -2022-12-06 11:22:11,819 - Epoch: [133][ 960/ 1200] Overall Loss 0.186601 Objective Loss 0.186601 LR 0.000500 Time 0.020190 -2022-12-06 11:22:12,014 - Epoch: [133][ 970/ 1200] Overall Loss 0.186461 Objective Loss 0.186461 LR 0.000500 Time 0.020182 -2022-12-06 11:22:12,209 - Epoch: [133][ 980/ 1200] Overall Loss 0.186496 Objective Loss 0.186496 LR 0.000500 Time 0.020175 -2022-12-06 11:22:12,404 - Epoch: [133][ 990/ 1200] Overall Loss 0.186932 Objective Loss 0.186932 LR 0.000500 Time 0.020167 -2022-12-06 11:22:12,599 - Epoch: [133][ 1000/ 1200] Overall Loss 0.187002 Objective Loss 0.187002 LR 0.000500 Time 0.020160 -2022-12-06 11:22:12,794 - Epoch: [133][ 1010/ 1200] Overall Loss 0.187314 Objective Loss 0.187314 LR 0.000500 Time 0.020153 -2022-12-06 11:22:12,989 - Epoch: [133][ 1020/ 1200] Overall Loss 0.187239 Objective Loss 0.187239 LR 0.000500 Time 0.020146 -2022-12-06 11:22:13,184 - Epoch: [133][ 1030/ 1200] Overall Loss 0.187326 Objective Loss 0.187326 LR 0.000500 Time 0.020139 -2022-12-06 11:22:13,379 - Epoch: [133][ 1040/ 1200] Overall Loss 0.187216 Objective Loss 0.187216 LR 0.000500 Time 0.020132 -2022-12-06 11:22:13,574 - Epoch: [133][ 1050/ 1200] Overall Loss 0.187358 Objective Loss 0.187358 LR 0.000500 Time 0.020125 -2022-12-06 11:22:13,769 - Epoch: [133][ 1060/ 1200] Overall Loss 0.187511 Objective Loss 0.187511 LR 0.000500 Time 0.020119 -2022-12-06 11:22:13,964 - Epoch: [133][ 1070/ 1200] Overall Loss 0.187530 Objective Loss 0.187530 LR 0.000500 Time 0.020113 -2022-12-06 11:22:14,159 - Epoch: [133][ 1080/ 1200] Overall Loss 0.187585 Objective Loss 0.187585 LR 0.000500 Time 0.020107 -2022-12-06 11:22:14,354 - Epoch: [133][ 1090/ 1200] Overall Loss 0.187933 Objective Loss 0.187933 LR 0.000500 Time 0.020100 -2022-12-06 11:22:14,549 - Epoch: [133][ 1100/ 1200] Overall Loss 0.187945 Objective Loss 0.187945 LR 0.000500 Time 0.020095 -2022-12-06 11:22:14,744 - Epoch: [133][ 1110/ 1200] Overall Loss 0.187936 Objective Loss 0.187936 LR 0.000500 Time 0.020089 -2022-12-06 11:22:14,939 - Epoch: [133][ 1120/ 1200] Overall Loss 0.187978 Objective Loss 0.187978 LR 0.000500 Time 0.020083 -2022-12-06 11:22:15,132 - Epoch: [133][ 1130/ 1200] Overall Loss 0.188211 Objective Loss 0.188211 LR 0.000500 Time 0.020076 -2022-12-06 11:22:15,325 - Epoch: [133][ 1140/ 1200] Overall Loss 0.188154 Objective Loss 0.188154 LR 0.000500 Time 0.020068 -2022-12-06 11:22:15,517 - Epoch: [133][ 1150/ 1200] Overall Loss 0.188209 Objective Loss 0.188209 LR 0.000500 Time 0.020060 -2022-12-06 11:22:15,709 - Epoch: [133][ 1160/ 1200] Overall Loss 0.188050 Objective Loss 0.188050 LR 0.000500 Time 0.020052 -2022-12-06 11:22:15,902 - Epoch: [133][ 1170/ 1200] Overall Loss 0.188012 Objective Loss 0.188012 LR 0.000500 Time 0.020045 -2022-12-06 11:22:16,094 - Epoch: [133][ 1180/ 1200] Overall Loss 0.188208 Objective Loss 0.188208 LR 0.000500 Time 0.020038 -2022-12-06 11:22:16,287 - Epoch: [133][ 1190/ 1200] Overall Loss 0.188218 Objective Loss 0.188218 LR 0.000500 Time 0.020031 -2022-12-06 11:22:16,520 - Epoch: [133][ 1200/ 1200] Overall Loss 0.188238 Objective Loss 0.188238 Top1 88.702929 Top5 98.535565 LR 0.000500 Time 0.020058 -2022-12-06 11:22:16,608 - --- validate (epoch=133)----------- -2022-12-06 11:22:16,608 - 34129 samples (256 per mini-batch) -2022-12-06 11:22:17,062 - Epoch: [133][ 10/ 134] Loss 0.250649 Top1 86.679688 Top5 98.359375 -2022-12-06 11:22:17,200 - Epoch: [133][ 20/ 134] Loss 0.247856 Top1 86.250000 Top5 98.261719 -2022-12-06 11:22:17,332 - Epoch: [133][ 30/ 134] Loss 0.255632 Top1 86.210938 Top5 98.281250 -2022-12-06 11:22:17,468 - Epoch: [133][ 40/ 134] Loss 0.259688 Top1 86.240234 Top5 98.300781 -2022-12-06 11:22:17,602 - Epoch: [133][ 50/ 134] Loss 0.253967 Top1 86.257812 Top5 98.328125 -2022-12-06 11:22:17,738 - Epoch: [133][ 60/ 134] Loss 0.259239 Top1 86.093750 Top5 98.339844 -2022-12-06 11:22:17,872 - Epoch: [133][ 70/ 134] Loss 0.260422 Top1 86.010045 Top5 98.331473 -2022-12-06 11:22:18,008 - Epoch: [133][ 80/ 134] Loss 0.256185 Top1 86.132812 Top5 98.330078 -2022-12-06 11:22:18,142 - Epoch: [133][ 90/ 134] Loss 0.258933 Top1 86.115451 Top5 98.324653 -2022-12-06 11:22:18,278 - Epoch: [133][ 100/ 134] Loss 0.258396 Top1 86.187500 Top5 98.332031 -2022-12-06 11:22:18,413 - Epoch: [133][ 110/ 134] Loss 0.258952 Top1 86.072443 Top5 98.320312 -2022-12-06 11:22:18,549 - Epoch: [133][ 120/ 134] Loss 0.261214 Top1 86.074219 Top5 98.287760 -2022-12-06 11:22:18,684 - Epoch: [133][ 130/ 134] Loss 0.264789 Top1 85.976562 Top5 98.254207 -2022-12-06 11:22:18,723 - Epoch: [133][ 134/ 134] Loss 0.263943 Top1 86.000176 Top5 98.256615 -2022-12-06 11:22:18,824 - ==> Top1: 86.000 Top5: 98.257 Loss: 0.264 - -2022-12-06 11:22:18,825 - ==> Confusion: -[[ 908 3 2 4 3 6 0 1 7 46 0 2 2 3 4 1 0 0 2 0 2] - [ 2 945 1 2 6 20 1 14 1 1 4 4 1 2 0 1 5 0 8 5 4] - [ 7 2 989 10 6 3 24 11 1 3 9 5 3 3 4 2 1 1 6 4 9] - [ 2 3 19 914 2 1 1 3 0 1 11 1 5 2 21 0 3 2 20 0 9] - [ 13 6 2 0 941 10 0 2 3 6 1 1 0 4 11 5 10 2 1 1 1] - [ 4 12 0 2 3 976 1 23 3 4 0 10 3 12 2 1 0 0 2 9 2] - [ 1 1 4 0 0 3 1077 4 0 0 1 3 2 1 0 4 0 2 1 11 3] - [ 0 7 6 2 3 25 8 959 0 0 2 4 3 2 0 0 0 0 13 13 7] - [ 6 1 0 0 1 0 0 1 996 29 11 0 0 7 4 0 2 1 1 2 2] - [ 52 0 1 1 3 1 0 2 34 880 1 2 0 15 3 1 0 1 1 0 3] - [ 1 0 2 6 0 1 1 3 14 3 953 1 1 14 5 1 0 0 6 1 6] - [ 3 0 1 1 1 9 6 3 0 0 1 969 25 4 2 4 3 4 1 13 1] - [ 1 0 3 1 1 3 0 1 0 0 1 27 897 5 2 6 3 6 1 7 4] - [ 0 1 0 0 0 5 0 1 14 7 3 2 3 975 0 1 3 0 2 4 2] - [ 10 4 0 2 2 4 0 0 18 8 0 3 1 3 1060 0 3 0 5 1 6] - [ 1 1 0 1 3 0 3 0 0 0 1 5 9 5 0 988 8 8 2 3 5] - [ 4 1 0 0 2 2 2 0 2 0 0 3 3 2 0 6 1034 0 0 6 5] - [ 2 1 1 1 1 2 1 2 3 3 0 16 18 3 4 13 2 957 0 4 2] - [ 1 2 3 4 2 4 0 20 2 1 4 2 4 1 9 0 0 1 945 2 1] - [ 2 4 3 2 1 3 6 5 1 0 0 15 8 5 0 2 3 1 2 1014 3] - [ 137 218 135 90 74 191 96 155 125 100 161 116 320 346 156 120 191 72 160 294 9969]] - -2022-12-06 11:22:19,406 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:22:19,407 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:22:19,413 - - -2022-12-06 11:22:19,413 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:22:20,468 - Epoch: [134][ 10/ 1200] Overall Loss 0.216728 Objective Loss 0.216728 LR 0.000500 Time 0.105465 -2022-12-06 11:22:20,670 - Epoch: [134][ 20/ 1200] Overall Loss 0.188136 Objective Loss 0.188136 LR 0.000500 Time 0.062823 -2022-12-06 11:22:20,871 - Epoch: [134][ 30/ 1200] Overall Loss 0.192154 Objective Loss 0.192154 LR 0.000500 Time 0.048538 -2022-12-06 11:22:21,068 - Epoch: [134][ 40/ 1200] Overall Loss 0.191854 Objective Loss 0.191854 LR 0.000500 Time 0.041333 -2022-12-06 11:22:21,268 - Epoch: [134][ 50/ 1200] Overall Loss 0.188514 Objective Loss 0.188514 LR 0.000500 Time 0.037051 -2022-12-06 11:22:21,465 - Epoch: [134][ 60/ 1200] Overall Loss 0.186516 Objective Loss 0.186516 LR 0.000500 Time 0.034153 -2022-12-06 11:22:21,665 - Epoch: [134][ 70/ 1200] Overall Loss 0.186936 Objective Loss 0.186936 LR 0.000500 Time 0.032117 -2022-12-06 11:22:21,862 - Epoch: [134][ 80/ 1200] Overall Loss 0.184755 Objective Loss 0.184755 LR 0.000500 Time 0.030560 -2022-12-06 11:22:22,061 - Epoch: [134][ 90/ 1200] Overall Loss 0.183128 Objective Loss 0.183128 LR 0.000500 Time 0.029371 -2022-12-06 11:22:22,258 - Epoch: [134][ 100/ 1200] Overall Loss 0.180432 Objective Loss 0.180432 LR 0.000500 Time 0.028395 -2022-12-06 11:22:22,457 - Epoch: [134][ 110/ 1200] Overall Loss 0.181164 Objective Loss 0.181164 LR 0.000500 Time 0.027621 -2022-12-06 11:22:22,654 - Epoch: [134][ 120/ 1200] Overall Loss 0.182930 Objective Loss 0.182930 LR 0.000500 Time 0.026956 -2022-12-06 11:22:22,854 - Epoch: [134][ 130/ 1200] Overall Loss 0.184384 Objective Loss 0.184384 LR 0.000500 Time 0.026412 -2022-12-06 11:22:23,052 - Epoch: [134][ 140/ 1200] Overall Loss 0.183063 Objective Loss 0.183063 LR 0.000500 Time 0.025940 -2022-12-06 11:22:23,251 - Epoch: [134][ 150/ 1200] Overall Loss 0.184387 Objective Loss 0.184387 LR 0.000500 Time 0.025531 -2022-12-06 11:22:23,448 - Epoch: [134][ 160/ 1200] Overall Loss 0.183713 Objective Loss 0.183713 LR 0.000500 Time 0.025164 -2022-12-06 11:22:23,647 - Epoch: [134][ 170/ 1200] Overall Loss 0.184988 Objective Loss 0.184988 LR 0.000500 Time 0.024851 -2022-12-06 11:22:23,844 - Epoch: [134][ 180/ 1200] Overall Loss 0.186570 Objective Loss 0.186570 LR 0.000500 Time 0.024562 -2022-12-06 11:22:24,043 - Epoch: [134][ 190/ 1200] Overall Loss 0.186045 Objective Loss 0.186045 LR 0.000500 Time 0.024312 -2022-12-06 11:22:24,239 - Epoch: [134][ 200/ 1200] Overall Loss 0.186077 Objective Loss 0.186077 LR 0.000500 Time 0.024073 -2022-12-06 11:22:24,438 - Epoch: [134][ 210/ 1200] Overall Loss 0.185888 Objective Loss 0.185888 LR 0.000500 Time 0.023872 -2022-12-06 11:22:24,635 - Epoch: [134][ 220/ 1200] Overall Loss 0.185147 Objective Loss 0.185147 LR 0.000500 Time 0.023681 -2022-12-06 11:22:24,835 - Epoch: [134][ 230/ 1200] Overall Loss 0.185117 Objective Loss 0.185117 LR 0.000500 Time 0.023517 -2022-12-06 11:22:25,032 - Epoch: [134][ 240/ 1200] Overall Loss 0.185411 Objective Loss 0.185411 LR 0.000500 Time 0.023358 -2022-12-06 11:22:25,232 - Epoch: [134][ 250/ 1200] Overall Loss 0.186125 Objective Loss 0.186125 LR 0.000500 Time 0.023220 -2022-12-06 11:22:25,429 - Epoch: [134][ 260/ 1200] Overall Loss 0.185805 Objective Loss 0.185805 LR 0.000500 Time 0.023082 -2022-12-06 11:22:25,628 - Epoch: [134][ 270/ 1200] Overall Loss 0.184702 Objective Loss 0.184702 LR 0.000500 Time 0.022963 -2022-12-06 11:22:25,825 - Epoch: [134][ 280/ 1200] Overall Loss 0.184758 Objective Loss 0.184758 LR 0.000500 Time 0.022845 -2022-12-06 11:22:26,024 - Epoch: [134][ 290/ 1200] Overall Loss 0.185219 Objective Loss 0.185219 LR 0.000500 Time 0.022741 -2022-12-06 11:22:26,221 - Epoch: [134][ 300/ 1200] Overall Loss 0.185913 Objective Loss 0.185913 LR 0.000500 Time 0.022637 -2022-12-06 11:22:26,420 - Epoch: [134][ 310/ 1200] Overall Loss 0.186020 Objective Loss 0.186020 LR 0.000500 Time 0.022547 -2022-12-06 11:22:26,618 - Epoch: [134][ 320/ 1200] Overall Loss 0.186664 Objective Loss 0.186664 LR 0.000500 Time 0.022460 -2022-12-06 11:22:26,819 - Epoch: [134][ 330/ 1200] Overall Loss 0.187475 Objective Loss 0.187475 LR 0.000500 Time 0.022386 -2022-12-06 11:22:27,016 - Epoch: [134][ 340/ 1200] Overall Loss 0.187338 Objective Loss 0.187338 LR 0.000500 Time 0.022307 -2022-12-06 11:22:27,216 - Epoch: [134][ 350/ 1200] Overall Loss 0.187134 Objective Loss 0.187134 LR 0.000500 Time 0.022238 -2022-12-06 11:22:27,414 - Epoch: [134][ 360/ 1200] Overall Loss 0.187066 Objective Loss 0.187066 LR 0.000500 Time 0.022168 -2022-12-06 11:22:27,613 - Epoch: [134][ 370/ 1200] Overall Loss 0.186867 Objective Loss 0.186867 LR 0.000500 Time 0.022106 -2022-12-06 11:22:27,810 - Epoch: [134][ 380/ 1200] Overall Loss 0.186203 Objective Loss 0.186203 LR 0.000500 Time 0.022042 -2022-12-06 11:22:28,009 - Epoch: [134][ 390/ 1200] Overall Loss 0.186935 Objective Loss 0.186935 LR 0.000500 Time 0.021986 -2022-12-06 11:22:28,206 - Epoch: [134][ 400/ 1200] Overall Loss 0.187252 Objective Loss 0.187252 LR 0.000500 Time 0.021927 -2022-12-06 11:22:28,405 - Epoch: [134][ 410/ 1200] Overall Loss 0.186982 Objective Loss 0.186982 LR 0.000500 Time 0.021877 -2022-12-06 11:22:28,602 - Epoch: [134][ 420/ 1200] Overall Loss 0.186848 Objective Loss 0.186848 LR 0.000500 Time 0.021824 -2022-12-06 11:22:28,801 - Epoch: [134][ 430/ 1200] Overall Loss 0.186909 Objective Loss 0.186909 LR 0.000500 Time 0.021777 -2022-12-06 11:22:28,998 - Epoch: [134][ 440/ 1200] Overall Loss 0.186634 Objective Loss 0.186634 LR 0.000500 Time 0.021729 -2022-12-06 11:22:29,197 - Epoch: [134][ 450/ 1200] Overall Loss 0.185917 Objective Loss 0.185917 LR 0.000500 Time 0.021687 -2022-12-06 11:22:29,394 - Epoch: [134][ 460/ 1200] Overall Loss 0.185865 Objective Loss 0.185865 LR 0.000500 Time 0.021642 -2022-12-06 11:22:29,593 - Epoch: [134][ 470/ 1200] Overall Loss 0.186078 Objective Loss 0.186078 LR 0.000500 Time 0.021605 -2022-12-06 11:22:29,791 - Epoch: [134][ 480/ 1200] Overall Loss 0.186074 Objective Loss 0.186074 LR 0.000500 Time 0.021565 -2022-12-06 11:22:29,991 - Epoch: [134][ 490/ 1200] Overall Loss 0.186220 Objective Loss 0.186220 LR 0.000500 Time 0.021532 -2022-12-06 11:22:30,188 - Epoch: [134][ 500/ 1200] Overall Loss 0.186733 Objective Loss 0.186733 LR 0.000500 Time 0.021495 -2022-12-06 11:22:30,388 - Epoch: [134][ 510/ 1200] Overall Loss 0.186513 Objective Loss 0.186513 LR 0.000500 Time 0.021463 -2022-12-06 11:22:30,586 - Epoch: [134][ 520/ 1200] Overall Loss 0.186018 Objective Loss 0.186018 LR 0.000500 Time 0.021430 -2022-12-06 11:22:30,785 - Epoch: [134][ 530/ 1200] Overall Loss 0.186298 Objective Loss 0.186298 LR 0.000500 Time 0.021401 -2022-12-06 11:22:30,982 - Epoch: [134][ 540/ 1200] Overall Loss 0.186583 Objective Loss 0.186583 LR 0.000500 Time 0.021369 -2022-12-06 11:22:31,181 - Epoch: [134][ 550/ 1200] Overall Loss 0.187231 Objective Loss 0.187231 LR 0.000500 Time 0.021342 -2022-12-06 11:22:31,379 - Epoch: [134][ 560/ 1200] Overall Loss 0.186735 Objective Loss 0.186735 LR 0.000500 Time 0.021312 -2022-12-06 11:22:31,579 - Epoch: [134][ 570/ 1200] Overall Loss 0.186606 Objective Loss 0.186606 LR 0.000500 Time 0.021288 -2022-12-06 11:22:31,776 - Epoch: [134][ 580/ 1200] Overall Loss 0.186688 Objective Loss 0.186688 LR 0.000500 Time 0.021260 -2022-12-06 11:22:31,975 - Epoch: [134][ 590/ 1200] Overall Loss 0.187026 Objective Loss 0.187026 LR 0.000500 Time 0.021236 -2022-12-06 11:22:32,172 - Epoch: [134][ 600/ 1200] Overall Loss 0.187166 Objective Loss 0.187166 LR 0.000500 Time 0.021210 -2022-12-06 11:22:32,371 - Epoch: [134][ 610/ 1200] Overall Loss 0.187540 Objective Loss 0.187540 LR 0.000500 Time 0.021187 -2022-12-06 11:22:32,568 - Epoch: [134][ 620/ 1200] Overall Loss 0.187870 Objective Loss 0.187870 LR 0.000500 Time 0.021163 -2022-12-06 11:22:32,768 - Epoch: [134][ 630/ 1200] Overall Loss 0.187827 Objective Loss 0.187827 LR 0.000500 Time 0.021143 -2022-12-06 11:22:32,965 - Epoch: [134][ 640/ 1200] Overall Loss 0.188309 Objective Loss 0.188309 LR 0.000500 Time 0.021119 -2022-12-06 11:22:33,164 - Epoch: [134][ 650/ 1200] Overall Loss 0.188282 Objective Loss 0.188282 LR 0.000500 Time 0.021099 -2022-12-06 11:22:33,361 - Epoch: [134][ 660/ 1200] Overall Loss 0.188299 Objective Loss 0.188299 LR 0.000500 Time 0.021077 -2022-12-06 11:22:33,560 - Epoch: [134][ 670/ 1200] Overall Loss 0.188359 Objective Loss 0.188359 LR 0.000500 Time 0.021059 -2022-12-06 11:22:33,757 - Epoch: [134][ 680/ 1200] Overall Loss 0.188405 Objective Loss 0.188405 LR 0.000500 Time 0.021039 -2022-12-06 11:22:33,956 - Epoch: [134][ 690/ 1200] Overall Loss 0.188738 Objective Loss 0.188738 LR 0.000500 Time 0.021021 -2022-12-06 11:22:34,153 - Epoch: [134][ 700/ 1200] Overall Loss 0.188949 Objective Loss 0.188949 LR 0.000500 Time 0.021002 -2022-12-06 11:22:34,352 - Epoch: [134][ 710/ 1200] Overall Loss 0.188991 Objective Loss 0.188991 LR 0.000500 Time 0.020985 -2022-12-06 11:22:34,548 - Epoch: [134][ 720/ 1200] Overall Loss 0.189089 Objective Loss 0.189089 LR 0.000500 Time 0.020966 -2022-12-06 11:22:34,747 - Epoch: [134][ 730/ 1200] Overall Loss 0.188902 Objective Loss 0.188902 LR 0.000500 Time 0.020950 -2022-12-06 11:22:34,945 - Epoch: [134][ 740/ 1200] Overall Loss 0.188824 Objective Loss 0.188824 LR 0.000500 Time 0.020934 -2022-12-06 11:22:35,144 - Epoch: [134][ 750/ 1200] Overall Loss 0.188885 Objective Loss 0.188885 LR 0.000500 Time 0.020918 -2022-12-06 11:22:35,341 - Epoch: [134][ 760/ 1200] Overall Loss 0.189162 Objective Loss 0.189162 LR 0.000500 Time 0.020902 -2022-12-06 11:22:35,540 - Epoch: [134][ 770/ 1200] Overall Loss 0.189022 Objective Loss 0.189022 LR 0.000500 Time 0.020889 -2022-12-06 11:22:35,738 - Epoch: [134][ 780/ 1200] Overall Loss 0.189024 Objective Loss 0.189024 LR 0.000500 Time 0.020874 -2022-12-06 11:22:35,938 - Epoch: [134][ 790/ 1200] Overall Loss 0.188987 Objective Loss 0.188987 LR 0.000500 Time 0.020862 -2022-12-06 11:22:36,135 - Epoch: [134][ 800/ 1200] Overall Loss 0.189010 Objective Loss 0.189010 LR 0.000500 Time 0.020847 -2022-12-06 11:22:36,335 - Epoch: [134][ 810/ 1200] Overall Loss 0.188877 Objective Loss 0.188877 LR 0.000500 Time 0.020836 -2022-12-06 11:22:36,532 - Epoch: [134][ 820/ 1200] Overall Loss 0.188796 Objective Loss 0.188796 LR 0.000500 Time 0.020822 -2022-12-06 11:22:36,732 - Epoch: [134][ 830/ 1200] Overall Loss 0.188675 Objective Loss 0.188675 LR 0.000500 Time 0.020811 -2022-12-06 11:22:36,929 - Epoch: [134][ 840/ 1200] Overall Loss 0.188842 Objective Loss 0.188842 LR 0.000500 Time 0.020797 -2022-12-06 11:22:37,128 - Epoch: [134][ 850/ 1200] Overall Loss 0.188796 Objective Loss 0.188796 LR 0.000500 Time 0.020785 -2022-12-06 11:22:37,325 - Epoch: [134][ 860/ 1200] Overall Loss 0.188686 Objective Loss 0.188686 LR 0.000500 Time 0.020772 -2022-12-06 11:22:37,524 - Epoch: [134][ 870/ 1200] Overall Loss 0.188423 Objective Loss 0.188423 LR 0.000500 Time 0.020762 -2022-12-06 11:22:37,721 - Epoch: [134][ 880/ 1200] Overall Loss 0.188477 Objective Loss 0.188477 LR 0.000500 Time 0.020749 -2022-12-06 11:22:37,921 - Epoch: [134][ 890/ 1200] Overall Loss 0.188510 Objective Loss 0.188510 LR 0.000500 Time 0.020740 -2022-12-06 11:22:38,118 - Epoch: [134][ 900/ 1200] Overall Loss 0.188642 Objective Loss 0.188642 LR 0.000500 Time 0.020728 -2022-12-06 11:22:38,318 - Epoch: [134][ 910/ 1200] Overall Loss 0.188746 Objective Loss 0.188746 LR 0.000500 Time 0.020719 -2022-12-06 11:22:38,516 - Epoch: [134][ 920/ 1200] Overall Loss 0.188937 Objective Loss 0.188937 LR 0.000500 Time 0.020708 -2022-12-06 11:22:38,715 - Epoch: [134][ 930/ 1200] Overall Loss 0.188864 Objective Loss 0.188864 LR 0.000500 Time 0.020699 -2022-12-06 11:22:38,913 - Epoch: [134][ 940/ 1200] Overall Loss 0.188890 Objective Loss 0.188890 LR 0.000500 Time 0.020689 -2022-12-06 11:22:39,112 - Epoch: [134][ 950/ 1200] Overall Loss 0.188789 Objective Loss 0.188789 LR 0.000500 Time 0.020680 -2022-12-06 11:22:39,309 - Epoch: [134][ 960/ 1200] Overall Loss 0.188522 Objective Loss 0.188522 LR 0.000500 Time 0.020669 -2022-12-06 11:22:39,509 - Epoch: [134][ 970/ 1200] Overall Loss 0.188749 Objective Loss 0.188749 LR 0.000500 Time 0.020662 -2022-12-06 11:22:39,706 - Epoch: [134][ 980/ 1200] Overall Loss 0.188414 Objective Loss 0.188414 LR 0.000500 Time 0.020652 -2022-12-06 11:22:39,906 - Epoch: [134][ 990/ 1200] Overall Loss 0.188441 Objective Loss 0.188441 LR 0.000500 Time 0.020645 -2022-12-06 11:22:40,104 - Epoch: [134][ 1000/ 1200] Overall Loss 0.188459 Objective Loss 0.188459 LR 0.000500 Time 0.020635 -2022-12-06 11:22:40,304 - Epoch: [134][ 1010/ 1200] Overall Loss 0.188600 Objective Loss 0.188600 LR 0.000500 Time 0.020628 -2022-12-06 11:22:40,501 - Epoch: [134][ 1020/ 1200] Overall Loss 0.188608 Objective Loss 0.188608 LR 0.000500 Time 0.020619 -2022-12-06 11:22:40,701 - Epoch: [134][ 1030/ 1200] Overall Loss 0.188935 Objective Loss 0.188935 LR 0.000500 Time 0.020612 -2022-12-06 11:22:40,897 - Epoch: [134][ 1040/ 1200] Overall Loss 0.188802 Objective Loss 0.188802 LR 0.000500 Time 0.020602 -2022-12-06 11:22:41,090 - Epoch: [134][ 1050/ 1200] Overall Loss 0.188898 Objective Loss 0.188898 LR 0.000500 Time 0.020590 -2022-12-06 11:22:41,284 - Epoch: [134][ 1060/ 1200] Overall Loss 0.189070 Objective Loss 0.189070 LR 0.000500 Time 0.020577 -2022-12-06 11:22:41,480 - Epoch: [134][ 1070/ 1200] Overall Loss 0.189029 Objective Loss 0.189029 LR 0.000500 Time 0.020567 -2022-12-06 11:22:41,674 - Epoch: [134][ 1080/ 1200] Overall Loss 0.188955 Objective Loss 0.188955 LR 0.000500 Time 0.020556 -2022-12-06 11:22:41,868 - Epoch: [134][ 1090/ 1200] Overall Loss 0.188915 Objective Loss 0.188915 LR 0.000500 Time 0.020545 -2022-12-06 11:22:42,061 - Epoch: [134][ 1100/ 1200] Overall Loss 0.188982 Objective Loss 0.188982 LR 0.000500 Time 0.020533 -2022-12-06 11:22:42,254 - Epoch: [134][ 1110/ 1200] Overall Loss 0.189027 Objective Loss 0.189027 LR 0.000500 Time 0.020522 -2022-12-06 11:22:42,448 - Epoch: [134][ 1120/ 1200] Overall Loss 0.189078 Objective Loss 0.189078 LR 0.000500 Time 0.020511 -2022-12-06 11:22:42,643 - Epoch: [134][ 1130/ 1200] Overall Loss 0.189115 Objective Loss 0.189115 LR 0.000500 Time 0.020502 -2022-12-06 11:22:42,837 - Epoch: [134][ 1140/ 1200] Overall Loss 0.189083 Objective Loss 0.189083 LR 0.000500 Time 0.020491 -2022-12-06 11:22:43,030 - Epoch: [134][ 1150/ 1200] Overall Loss 0.189009 Objective Loss 0.189009 LR 0.000500 Time 0.020481 -2022-12-06 11:22:43,223 - Epoch: [134][ 1160/ 1200] Overall Loss 0.188993 Objective Loss 0.188993 LR 0.000500 Time 0.020470 -2022-12-06 11:22:43,416 - Epoch: [134][ 1170/ 1200] Overall Loss 0.189028 Objective Loss 0.189028 LR 0.000500 Time 0.020460 -2022-12-06 11:22:43,609 - Epoch: [134][ 1180/ 1200] Overall Loss 0.188927 Objective Loss 0.188927 LR 0.000500 Time 0.020449 -2022-12-06 11:22:43,803 - Epoch: [134][ 1190/ 1200] Overall Loss 0.188931 Objective Loss 0.188931 LR 0.000500 Time 0.020440 -2022-12-06 11:22:44,026 - Epoch: [134][ 1200/ 1200] Overall Loss 0.189059 Objective Loss 0.189059 Top1 86.401674 Top5 98.535565 LR 0.000500 Time 0.020456 -2022-12-06 11:22:44,135 - --- validate (epoch=134)----------- -2022-12-06 11:22:44,135 - 34129 samples (256 per mini-batch) -2022-12-06 11:22:44,594 - Epoch: [134][ 10/ 134] Loss 0.275731 Top1 86.132812 Top5 98.281250 -2022-12-06 11:22:44,727 - Epoch: [134][ 20/ 134] Loss 0.254364 Top1 86.425781 Top5 98.222656 -2022-12-06 11:22:44,859 - Epoch: [134][ 30/ 134] Loss 0.255472 Top1 86.289062 Top5 98.268229 -2022-12-06 11:22:44,990 - Epoch: [134][ 40/ 134] Loss 0.253208 Top1 86.552734 Top5 98.310547 -2022-12-06 11:22:45,124 - Epoch: [134][ 50/ 134] Loss 0.250363 Top1 86.546875 Top5 98.234375 -2022-12-06 11:22:45,254 - Epoch: [134][ 60/ 134] Loss 0.252388 Top1 86.471354 Top5 98.248698 -2022-12-06 11:22:45,385 - Epoch: [134][ 70/ 134] Loss 0.251438 Top1 86.512277 Top5 98.281250 -2022-12-06 11:22:45,516 - Epoch: [134][ 80/ 134] Loss 0.252030 Top1 86.484375 Top5 98.310547 -2022-12-06 11:22:45,648 - Epoch: [134][ 90/ 134] Loss 0.254276 Top1 86.427951 Top5 98.311632 -2022-12-06 11:22:45,776 - Epoch: [134][ 100/ 134] Loss 0.256938 Top1 86.402344 Top5 98.277344 -2022-12-06 11:22:45,907 - Epoch: [134][ 110/ 134] Loss 0.257406 Top1 86.399148 Top5 98.284801 -2022-12-06 11:22:46,040 - Epoch: [134][ 120/ 134] Loss 0.258683 Top1 86.380208 Top5 98.258464 -2022-12-06 11:22:46,173 - Epoch: [134][ 130/ 134] Loss 0.258764 Top1 86.358173 Top5 98.245192 -2022-12-06 11:22:46,213 - Epoch: [134][ 134/ 134] Loss 0.256587 Top1 86.404524 Top5 98.262475 -2022-12-06 11:22:46,300 - ==> Top1: 86.405 Top5: 98.262 Loss: 0.257 - -2022-12-06 11:22:46,301 - ==> Confusion: -[[ 925 1 1 1 2 8 0 1 7 42 0 3 1 1 1 1 0 0 0 0 1] - [ 1 950 1 1 10 14 1 14 1 4 6 4 0 1 0 1 3 1 5 2 7] - [ 4 2 1000 11 5 4 16 13 0 3 6 7 0 0 3 5 1 1 5 4 13] - [ 3 1 13 935 2 2 1 3 2 1 13 0 4 2 17 0 1 2 12 0 6] - [ 15 6 0 1 950 3 0 1 1 9 1 2 1 1 10 5 5 2 1 1 5] - [ 4 17 0 2 3 988 2 15 4 2 0 11 4 5 2 0 1 0 3 2 4] - [ 1 6 12 2 2 2 1066 3 0 0 1 3 3 0 0 5 0 2 1 7 2] - [ 3 7 8 0 1 29 8 951 0 1 0 4 2 3 1 1 0 0 20 10 5] - [ 7 0 0 0 0 2 0 0 981 47 7 2 2 5 3 1 2 1 1 0 3] - [ 73 0 0 0 3 1 0 2 23 884 1 1 0 7 0 0 1 1 0 0 4] - [ 0 2 2 2 1 1 0 2 17 3 957 1 0 12 6 1 0 1 4 2 5] - [ 3 1 0 0 1 13 4 4 2 0 0 975 24 1 0 4 4 3 1 8 3] - [ 0 1 0 3 2 2 0 1 1 0 0 18 910 3 2 7 3 4 0 4 8] - [ 0 1 1 0 0 13 0 2 21 16 4 8 5 939 1 0 5 0 0 1 6] - [ 10 4 2 4 1 2 0 0 21 11 0 2 4 2 1057 0 1 0 6 1 2] - [ 0 1 2 1 2 1 3 0 0 0 1 5 4 3 0 995 5 15 0 2 3] - [ 3 1 0 1 4 2 1 0 1 0 0 3 2 2 1 12 1030 0 0 5 4] - [ 2 2 1 1 0 2 0 0 2 3 0 8 18 1 0 14 0 977 1 2 2] - [ 6 4 2 7 1 2 0 26 3 1 6 1 4 1 12 1 0 0 929 1 1] - [ 2 5 1 1 0 5 5 10 1 0 1 11 7 7 1 2 4 4 1 1007 5] - [ 153 245 164 105 87 182 63 186 124 119 139 102 386 249 147 102 181 88 134 193 10077]] - -2022-12-06 11:22:46,869 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:22:46,869 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:22:46,876 - - -2022-12-06 11:22:46,876 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:22:47,797 - Epoch: [135][ 10/ 1200] Overall Loss 0.195996 Objective Loss 0.195996 LR 0.000500 Time 0.092010 -2022-12-06 11:22:47,988 - Epoch: [135][ 20/ 1200] Overall Loss 0.190862 Objective Loss 0.190862 LR 0.000500 Time 0.055533 -2022-12-06 11:22:48,178 - Epoch: [135][ 30/ 1200] Overall Loss 0.185511 Objective Loss 0.185511 LR 0.000500 Time 0.043354 -2022-12-06 11:22:48,368 - Epoch: [135][ 40/ 1200] Overall Loss 0.184242 Objective Loss 0.184242 LR 0.000500 Time 0.037244 -2022-12-06 11:22:48,558 - Epoch: [135][ 50/ 1200] Overall Loss 0.185758 Objective Loss 0.185758 LR 0.000500 Time 0.033589 -2022-12-06 11:22:48,748 - Epoch: [135][ 60/ 1200] Overall Loss 0.182182 Objective Loss 0.182182 LR 0.000500 Time 0.031148 -2022-12-06 11:22:48,938 - Epoch: [135][ 70/ 1200] Overall Loss 0.181265 Objective Loss 0.181265 LR 0.000500 Time 0.029403 -2022-12-06 11:22:49,129 - Epoch: [135][ 80/ 1200] Overall Loss 0.182269 Objective Loss 0.182269 LR 0.000500 Time 0.028108 -2022-12-06 11:22:49,319 - Epoch: [135][ 90/ 1200] Overall Loss 0.181689 Objective Loss 0.181689 LR 0.000500 Time 0.027086 -2022-12-06 11:22:49,508 - Epoch: [135][ 100/ 1200] Overall Loss 0.182424 Objective Loss 0.182424 LR 0.000500 Time 0.026263 -2022-12-06 11:22:49,698 - Epoch: [135][ 110/ 1200] Overall Loss 0.183016 Objective Loss 0.183016 LR 0.000500 Time 0.025600 -2022-12-06 11:22:49,887 - Epoch: [135][ 120/ 1200] Overall Loss 0.182222 Objective Loss 0.182222 LR 0.000500 Time 0.025040 -2022-12-06 11:22:50,078 - Epoch: [135][ 130/ 1200] Overall Loss 0.182733 Objective Loss 0.182733 LR 0.000500 Time 0.024574 -2022-12-06 11:22:50,266 - Epoch: [135][ 140/ 1200] Overall Loss 0.183284 Objective Loss 0.183284 LR 0.000500 Time 0.024163 -2022-12-06 11:22:50,456 - Epoch: [135][ 150/ 1200] Overall Loss 0.183657 Objective Loss 0.183657 LR 0.000500 Time 0.023813 -2022-12-06 11:22:50,646 - Epoch: [135][ 160/ 1200] Overall Loss 0.182368 Objective Loss 0.182368 LR 0.000500 Time 0.023508 -2022-12-06 11:22:50,836 - Epoch: [135][ 170/ 1200] Overall Loss 0.182852 Objective Loss 0.182852 LR 0.000500 Time 0.023238 -2022-12-06 11:22:51,025 - Epoch: [135][ 180/ 1200] Overall Loss 0.183907 Objective Loss 0.183907 LR 0.000500 Time 0.022998 -2022-12-06 11:22:51,215 - Epoch: [135][ 190/ 1200] Overall Loss 0.183812 Objective Loss 0.183812 LR 0.000500 Time 0.022785 -2022-12-06 11:22:51,406 - Epoch: [135][ 200/ 1200] Overall Loss 0.183294 Objective Loss 0.183294 LR 0.000500 Time 0.022594 -2022-12-06 11:22:51,595 - Epoch: [135][ 210/ 1200] Overall Loss 0.182961 Objective Loss 0.182961 LR 0.000500 Time 0.022417 -2022-12-06 11:22:51,784 - Epoch: [135][ 220/ 1200] Overall Loss 0.183037 Objective Loss 0.183037 LR 0.000500 Time 0.022256 -2022-12-06 11:22:51,974 - Epoch: [135][ 230/ 1200] Overall Loss 0.182982 Objective Loss 0.182982 LR 0.000500 Time 0.022110 -2022-12-06 11:22:52,164 - Epoch: [135][ 240/ 1200] Overall Loss 0.183294 Objective Loss 0.183294 LR 0.000500 Time 0.021977 -2022-12-06 11:22:52,353 - Epoch: [135][ 250/ 1200] Overall Loss 0.182349 Objective Loss 0.182349 LR 0.000500 Time 0.021854 -2022-12-06 11:22:52,542 - Epoch: [135][ 260/ 1200] Overall Loss 0.181807 Objective Loss 0.181807 LR 0.000500 Time 0.021740 -2022-12-06 11:22:52,732 - Epoch: [135][ 270/ 1200] Overall Loss 0.181748 Objective Loss 0.181748 LR 0.000500 Time 0.021635 -2022-12-06 11:22:52,922 - Epoch: [135][ 280/ 1200] Overall Loss 0.181708 Objective Loss 0.181708 LR 0.000500 Time 0.021538 -2022-12-06 11:22:53,111 - Epoch: [135][ 290/ 1200] Overall Loss 0.182041 Objective Loss 0.182041 LR 0.000500 Time 0.021446 -2022-12-06 11:22:53,301 - Epoch: [135][ 300/ 1200] Overall Loss 0.182214 Objective Loss 0.182214 LR 0.000500 Time 0.021361 -2022-12-06 11:22:53,491 - Epoch: [135][ 310/ 1200] Overall Loss 0.182750 Objective Loss 0.182750 LR 0.000500 Time 0.021283 -2022-12-06 11:22:53,680 - Epoch: [135][ 320/ 1200] Overall Loss 0.183603 Objective Loss 0.183603 LR 0.000500 Time 0.021209 -2022-12-06 11:22:53,869 - Epoch: [135][ 330/ 1200] Overall Loss 0.184025 Objective Loss 0.184025 LR 0.000500 Time 0.021138 -2022-12-06 11:22:54,059 - Epoch: [135][ 340/ 1200] Overall Loss 0.184027 Objective Loss 0.184027 LR 0.000500 Time 0.021074 -2022-12-06 11:22:54,249 - Epoch: [135][ 350/ 1200] Overall Loss 0.183316 Objective Loss 0.183316 LR 0.000500 Time 0.021012 -2022-12-06 11:22:54,439 - Epoch: [135][ 360/ 1200] Overall Loss 0.183560 Objective Loss 0.183560 LR 0.000500 Time 0.020954 -2022-12-06 11:22:54,629 - Epoch: [135][ 370/ 1200] Overall Loss 0.183829 Objective Loss 0.183829 LR 0.000500 Time 0.020899 -2022-12-06 11:22:54,818 - Epoch: [135][ 380/ 1200] Overall Loss 0.184095 Objective Loss 0.184095 LR 0.000500 Time 0.020846 -2022-12-06 11:22:55,008 - Epoch: [135][ 390/ 1200] Overall Loss 0.183810 Objective Loss 0.183810 LR 0.000500 Time 0.020797 -2022-12-06 11:22:55,198 - Epoch: [135][ 400/ 1200] Overall Loss 0.183375 Objective Loss 0.183375 LR 0.000500 Time 0.020750 -2022-12-06 11:22:55,387 - Epoch: [135][ 410/ 1200] Overall Loss 0.183344 Objective Loss 0.183344 LR 0.000500 Time 0.020705 -2022-12-06 11:22:55,577 - Epoch: [135][ 420/ 1200] Overall Loss 0.183610 Objective Loss 0.183610 LR 0.000500 Time 0.020663 -2022-12-06 11:22:55,767 - Epoch: [135][ 430/ 1200] Overall Loss 0.183139 Objective Loss 0.183139 LR 0.000500 Time 0.020623 -2022-12-06 11:22:55,958 - Epoch: [135][ 440/ 1200] Overall Loss 0.183269 Objective Loss 0.183269 LR 0.000500 Time 0.020586 -2022-12-06 11:22:56,147 - Epoch: [135][ 450/ 1200] Overall Loss 0.183492 Objective Loss 0.183492 LR 0.000500 Time 0.020549 -2022-12-06 11:22:56,337 - Epoch: [135][ 460/ 1200] Overall Loss 0.183335 Objective Loss 0.183335 LR 0.000500 Time 0.020514 -2022-12-06 11:22:56,527 - Epoch: [135][ 470/ 1200] Overall Loss 0.183549 Objective Loss 0.183549 LR 0.000500 Time 0.020481 -2022-12-06 11:22:56,718 - Epoch: [135][ 480/ 1200] Overall Loss 0.183312 Objective Loss 0.183312 LR 0.000500 Time 0.020449 -2022-12-06 11:22:56,907 - Epoch: [135][ 490/ 1200] Overall Loss 0.182669 Objective Loss 0.182669 LR 0.000500 Time 0.020418 -2022-12-06 11:22:57,098 - Epoch: [135][ 500/ 1200] Overall Loss 0.182521 Objective Loss 0.182521 LR 0.000500 Time 0.020389 -2022-12-06 11:22:57,288 - Epoch: [135][ 510/ 1200] Overall Loss 0.182111 Objective Loss 0.182111 LR 0.000500 Time 0.020361 -2022-12-06 11:22:57,477 - Epoch: [135][ 520/ 1200] Overall Loss 0.181609 Objective Loss 0.181609 LR 0.000500 Time 0.020333 -2022-12-06 11:22:57,668 - Epoch: [135][ 530/ 1200] Overall Loss 0.181493 Objective Loss 0.181493 LR 0.000500 Time 0.020307 -2022-12-06 11:22:57,857 - Epoch: [135][ 540/ 1200] Overall Loss 0.181700 Objective Loss 0.181700 LR 0.000500 Time 0.020281 -2022-12-06 11:22:58,047 - Epoch: [135][ 550/ 1200] Overall Loss 0.182004 Objective Loss 0.182004 LR 0.000500 Time 0.020257 -2022-12-06 11:22:58,237 - Epoch: [135][ 560/ 1200] Overall Loss 0.181784 Objective Loss 0.181784 LR 0.000500 Time 0.020232 -2022-12-06 11:22:58,426 - Epoch: [135][ 570/ 1200] Overall Loss 0.181941 Objective Loss 0.181941 LR 0.000500 Time 0.020209 -2022-12-06 11:22:58,616 - Epoch: [135][ 580/ 1200] Overall Loss 0.182127 Objective Loss 0.182127 LR 0.000500 Time 0.020187 -2022-12-06 11:22:58,806 - Epoch: [135][ 590/ 1200] Overall Loss 0.182343 Objective Loss 0.182343 LR 0.000500 Time 0.020166 -2022-12-06 11:22:58,996 - Epoch: [135][ 600/ 1200] Overall Loss 0.182508 Objective Loss 0.182508 LR 0.000500 Time 0.020146 -2022-12-06 11:22:59,186 - Epoch: [135][ 610/ 1200] Overall Loss 0.182365 Objective Loss 0.182365 LR 0.000500 Time 0.020126 -2022-12-06 11:22:59,375 - Epoch: [135][ 620/ 1200] Overall Loss 0.182553 Objective Loss 0.182553 LR 0.000500 Time 0.020106 -2022-12-06 11:22:59,565 - Epoch: [135][ 630/ 1200] Overall Loss 0.182512 Objective Loss 0.182512 LR 0.000500 Time 0.020088 -2022-12-06 11:22:59,755 - Epoch: [135][ 640/ 1200] Overall Loss 0.182499 Objective Loss 0.182499 LR 0.000500 Time 0.020069 -2022-12-06 11:22:59,944 - Epoch: [135][ 650/ 1200] Overall Loss 0.182488 Objective Loss 0.182488 LR 0.000500 Time 0.020051 -2022-12-06 11:23:00,134 - Epoch: [135][ 660/ 1200] Overall Loss 0.182615 Objective Loss 0.182615 LR 0.000500 Time 0.020033 -2022-12-06 11:23:00,323 - Epoch: [135][ 670/ 1200] Overall Loss 0.182517 Objective Loss 0.182517 LR 0.000500 Time 0.020016 -2022-12-06 11:23:00,513 - Epoch: [135][ 680/ 1200] Overall Loss 0.182768 Objective Loss 0.182768 LR 0.000500 Time 0.020000 -2022-12-06 11:23:00,703 - Epoch: [135][ 690/ 1200] Overall Loss 0.182401 Objective Loss 0.182401 LR 0.000500 Time 0.019985 -2022-12-06 11:23:00,892 - Epoch: [135][ 700/ 1200] Overall Loss 0.182729 Objective Loss 0.182729 LR 0.000500 Time 0.019969 -2022-12-06 11:23:01,082 - Epoch: [135][ 710/ 1200] Overall Loss 0.182804 Objective Loss 0.182804 LR 0.000500 Time 0.019954 -2022-12-06 11:23:01,272 - Epoch: [135][ 720/ 1200] Overall Loss 0.182990 Objective Loss 0.182990 LR 0.000500 Time 0.019940 -2022-12-06 11:23:01,461 - Epoch: [135][ 730/ 1200] Overall Loss 0.182864 Objective Loss 0.182864 LR 0.000500 Time 0.019926 -2022-12-06 11:23:01,651 - Epoch: [135][ 740/ 1200] Overall Loss 0.183179 Objective Loss 0.183179 LR 0.000500 Time 0.019912 -2022-12-06 11:23:01,841 - Epoch: [135][ 750/ 1200] Overall Loss 0.183696 Objective Loss 0.183696 LR 0.000500 Time 0.019899 -2022-12-06 11:23:02,030 - Epoch: [135][ 760/ 1200] Overall Loss 0.183775 Objective Loss 0.183775 LR 0.000500 Time 0.019885 -2022-12-06 11:23:02,220 - Epoch: [135][ 770/ 1200] Overall Loss 0.183861 Objective Loss 0.183861 LR 0.000500 Time 0.019873 -2022-12-06 11:23:02,409 - Epoch: [135][ 780/ 1200] Overall Loss 0.183905 Objective Loss 0.183905 LR 0.000500 Time 0.019860 -2022-12-06 11:23:02,599 - Epoch: [135][ 790/ 1200] Overall Loss 0.183872 Objective Loss 0.183872 LR 0.000500 Time 0.019849 -2022-12-06 11:23:02,789 - Epoch: [135][ 800/ 1200] Overall Loss 0.184053 Objective Loss 0.184053 LR 0.000500 Time 0.019837 -2022-12-06 11:23:02,978 - Epoch: [135][ 810/ 1200] Overall Loss 0.183989 Objective Loss 0.183989 LR 0.000500 Time 0.019825 -2022-12-06 11:23:03,168 - Epoch: [135][ 820/ 1200] Overall Loss 0.183879 Objective Loss 0.183879 LR 0.000500 Time 0.019814 -2022-12-06 11:23:03,358 - Epoch: [135][ 830/ 1200] Overall Loss 0.184072 Objective Loss 0.184072 LR 0.000500 Time 0.019804 -2022-12-06 11:23:03,549 - Epoch: [135][ 840/ 1200] Overall Loss 0.184017 Objective Loss 0.184017 LR 0.000500 Time 0.019794 -2022-12-06 11:23:03,738 - Epoch: [135][ 850/ 1200] Overall Loss 0.183969 Objective Loss 0.183969 LR 0.000500 Time 0.019784 -2022-12-06 11:23:03,928 - Epoch: [135][ 860/ 1200] Overall Loss 0.184294 Objective Loss 0.184294 LR 0.000500 Time 0.019774 -2022-12-06 11:23:04,119 - Epoch: [135][ 870/ 1200] Overall Loss 0.184509 Objective Loss 0.184509 LR 0.000500 Time 0.019765 -2022-12-06 11:23:04,309 - Epoch: [135][ 880/ 1200] Overall Loss 0.184073 Objective Loss 0.184073 LR 0.000500 Time 0.019755 -2022-12-06 11:23:04,500 - Epoch: [135][ 890/ 1200] Overall Loss 0.184043 Objective Loss 0.184043 LR 0.000500 Time 0.019747 -2022-12-06 11:23:04,689 - Epoch: [135][ 900/ 1200] Overall Loss 0.184055 Objective Loss 0.184055 LR 0.000500 Time 0.019738 -2022-12-06 11:23:04,879 - Epoch: [135][ 910/ 1200] Overall Loss 0.183960 Objective Loss 0.183960 LR 0.000500 Time 0.019729 -2022-12-06 11:23:05,069 - Epoch: [135][ 920/ 1200] Overall Loss 0.183936 Objective Loss 0.183936 LR 0.000500 Time 0.019721 -2022-12-06 11:23:05,258 - Epoch: [135][ 930/ 1200] Overall Loss 0.184028 Objective Loss 0.184028 LR 0.000500 Time 0.019712 -2022-12-06 11:23:05,448 - Epoch: [135][ 940/ 1200] Overall Loss 0.183856 Objective Loss 0.183856 LR 0.000500 Time 0.019703 -2022-12-06 11:23:05,639 - Epoch: [135][ 950/ 1200] Overall Loss 0.183962 Objective Loss 0.183962 LR 0.000500 Time 0.019696 -2022-12-06 11:23:05,828 - Epoch: [135][ 960/ 1200] Overall Loss 0.184243 Objective Loss 0.184243 LR 0.000500 Time 0.019688 -2022-12-06 11:23:06,018 - Epoch: [135][ 970/ 1200] Overall Loss 0.184152 Objective Loss 0.184152 LR 0.000500 Time 0.019679 -2022-12-06 11:23:06,207 - Epoch: [135][ 980/ 1200] Overall Loss 0.184403 Objective Loss 0.184403 LR 0.000500 Time 0.019671 -2022-12-06 11:23:06,397 - Epoch: [135][ 990/ 1200] Overall Loss 0.184369 Objective Loss 0.184369 LR 0.000500 Time 0.019664 -2022-12-06 11:23:06,587 - Epoch: [135][ 1000/ 1200] Overall Loss 0.184435 Objective Loss 0.184435 LR 0.000500 Time 0.019657 -2022-12-06 11:23:06,776 - Epoch: [135][ 1010/ 1200] Overall Loss 0.184746 Objective Loss 0.184746 LR 0.000500 Time 0.019649 -2022-12-06 11:23:06,966 - Epoch: [135][ 1020/ 1200] Overall Loss 0.184797 Objective Loss 0.184797 LR 0.000500 Time 0.019642 -2022-12-06 11:23:07,157 - Epoch: [135][ 1030/ 1200] Overall Loss 0.184671 Objective Loss 0.184671 LR 0.000500 Time 0.019635 -2022-12-06 11:23:07,346 - Epoch: [135][ 1040/ 1200] Overall Loss 0.185118 Objective Loss 0.185118 LR 0.000500 Time 0.019628 -2022-12-06 11:23:07,535 - Epoch: [135][ 1050/ 1200] Overall Loss 0.185168 Objective Loss 0.185168 LR 0.000500 Time 0.019621 -2022-12-06 11:23:07,725 - Epoch: [135][ 1060/ 1200] Overall Loss 0.185031 Objective Loss 0.185031 LR 0.000500 Time 0.019614 -2022-12-06 11:23:07,915 - Epoch: [135][ 1070/ 1200] Overall Loss 0.184992 Objective Loss 0.184992 LR 0.000500 Time 0.019608 -2022-12-06 11:23:08,105 - Epoch: [135][ 1080/ 1200] Overall Loss 0.185130 Objective Loss 0.185130 LR 0.000500 Time 0.019602 -2022-12-06 11:23:08,295 - Epoch: [135][ 1090/ 1200] Overall Loss 0.185002 Objective Loss 0.185002 LR 0.000500 Time 0.019596 -2022-12-06 11:23:08,485 - Epoch: [135][ 1100/ 1200] Overall Loss 0.185191 Objective Loss 0.185191 LR 0.000500 Time 0.019590 -2022-12-06 11:23:08,674 - Epoch: [135][ 1110/ 1200] Overall Loss 0.185082 Objective Loss 0.185082 LR 0.000500 Time 0.019583 -2022-12-06 11:23:08,864 - Epoch: [135][ 1120/ 1200] Overall Loss 0.185161 Objective Loss 0.185161 LR 0.000500 Time 0.019578 -2022-12-06 11:23:09,054 - Epoch: [135][ 1130/ 1200] Overall Loss 0.185197 Objective Loss 0.185197 LR 0.000500 Time 0.019572 -2022-12-06 11:23:09,245 - Epoch: [135][ 1140/ 1200] Overall Loss 0.185212 Objective Loss 0.185212 LR 0.000500 Time 0.019567 -2022-12-06 11:23:09,434 - Epoch: [135][ 1150/ 1200] Overall Loss 0.185158 Objective Loss 0.185158 LR 0.000500 Time 0.019561 -2022-12-06 11:23:09,624 - Epoch: [135][ 1160/ 1200] Overall Loss 0.185201 Objective Loss 0.185201 LR 0.000500 Time 0.019556 -2022-12-06 11:23:09,814 - Epoch: [135][ 1170/ 1200] Overall Loss 0.185314 Objective Loss 0.185314 LR 0.000500 Time 0.019551 -2022-12-06 11:23:10,004 - Epoch: [135][ 1180/ 1200] Overall Loss 0.185424 Objective Loss 0.185424 LR 0.000500 Time 0.019546 -2022-12-06 11:23:10,195 - Epoch: [135][ 1190/ 1200] Overall Loss 0.185265 Objective Loss 0.185265 LR 0.000500 Time 0.019541 -2022-12-06 11:23:10,417 - Epoch: [135][ 1200/ 1200] Overall Loss 0.185147 Objective Loss 0.185147 Top1 90.794979 Top5 99.163180 LR 0.000500 Time 0.019563 -2022-12-06 11:23:10,505 - --- validate (epoch=135)----------- -2022-12-06 11:23:10,505 - 34129 samples (256 per mini-batch) -2022-12-06 11:23:10,950 - Epoch: [135][ 10/ 134] Loss 0.272424 Top1 87.851562 Top5 98.359375 -2022-12-06 11:23:11,085 - Epoch: [135][ 20/ 134] Loss 0.260885 Top1 87.578125 Top5 98.476562 -2022-12-06 11:23:11,216 - Epoch: [135][ 30/ 134] Loss 0.249875 Top1 87.695312 Top5 98.463542 -2022-12-06 11:23:11,349 - Epoch: [135][ 40/ 134] Loss 0.248112 Top1 87.500000 Top5 98.486328 -2022-12-06 11:23:11,480 - Epoch: [135][ 50/ 134] Loss 0.244302 Top1 87.414062 Top5 98.562500 -2022-12-06 11:23:11,612 - Epoch: [135][ 60/ 134] Loss 0.245160 Top1 87.265625 Top5 98.548177 -2022-12-06 11:23:11,742 - Epoch: [135][ 70/ 134] Loss 0.241287 Top1 87.226562 Top5 98.571429 -2022-12-06 11:23:11,874 - Epoch: [135][ 80/ 134] Loss 0.248585 Top1 87.041016 Top5 98.535156 -2022-12-06 11:23:12,005 - Epoch: [135][ 90/ 134] Loss 0.249250 Top1 86.948785 Top5 98.511285 -2022-12-06 11:23:12,137 - Epoch: [135][ 100/ 134] Loss 0.253217 Top1 86.875000 Top5 98.445312 -2022-12-06 11:23:12,266 - Epoch: [135][ 110/ 134] Loss 0.252262 Top1 86.935369 Top5 98.430398 -2022-12-06 11:23:12,398 - Epoch: [135][ 120/ 134] Loss 0.254219 Top1 86.878255 Top5 98.398438 -2022-12-06 11:23:12,531 - Epoch: [135][ 130/ 134] Loss 0.253074 Top1 86.926082 Top5 98.389423 -2022-12-06 11:23:12,569 - Epoch: [135][ 134/ 134] Loss 0.253334 Top1 86.887984 Top5 98.370887 -2022-12-06 11:23:12,657 - ==> Top1: 86.888 Top5: 98.371 Loss: 0.253 - -2022-12-06 11:23:12,658 - ==> Confusion: -[[ 898 1 1 3 7 7 0 0 4 53 0 2 1 3 6 1 0 0 2 1 6] - [ 1 941 2 3 4 21 3 16 2 0 1 4 1 2 0 0 5 1 12 3 5] - [ 5 2 995 15 5 3 21 13 0 2 5 4 2 4 1 2 3 2 4 1 14] - [ 2 1 18 938 1 3 0 2 0 3 8 1 3 1 13 0 2 2 16 1 5] - [ 6 8 2 0 953 7 1 1 0 5 1 3 1 3 14 4 5 0 1 0 5] - [ 2 10 1 2 2 987 3 18 3 3 3 11 1 12 0 2 3 0 1 2 3] - [ 0 2 6 2 1 4 1078 5 1 0 0 0 0 1 0 2 2 3 0 9 2] - [ 1 5 6 1 0 22 5 981 0 0 1 5 0 2 1 1 0 0 12 7 4] - [ 5 4 0 0 0 3 1 2 960 43 15 2 0 8 11 1 1 0 5 1 2] - [ 44 0 1 0 7 4 0 4 22 889 1 3 0 14 4 3 0 1 0 0 4] - [ 1 1 2 5 3 0 3 3 5 1 964 2 2 9 3 0 0 0 6 2 7] - [ 1 0 2 0 0 9 3 5 0 1 4 987 13 5 1 3 5 5 0 6 1] - [ 0 1 1 2 0 2 0 1 1 0 1 35 885 4 3 4 1 11 1 7 9] - [ 0 1 1 0 0 6 0 4 7 8 3 5 1 977 0 0 3 0 0 2 5] - [ 8 2 3 9 3 4 0 0 9 2 1 3 1 4 1063 0 1 0 11 1 5] - [ 0 1 1 0 1 1 4 0 0 0 2 13 2 2 0 992 6 10 0 3 5] - [ 2 1 0 1 2 1 1 1 0 0 0 6 2 3 1 10 1030 0 1 6 4] - [ 1 1 0 1 0 0 1 1 1 2 1 11 16 5 2 9 0 977 1 2 4] - [ 0 4 5 5 1 3 0 31 0 1 2 3 2 0 4 0 0 2 941 0 4] - [ 3 4 1 1 0 6 6 15 0 0 3 14 6 8 0 5 5 2 1 991 9] - [ 83 239 145 123 93 196 87 189 68 78 167 125 301 273 135 92 172 73 186 182 10219]] - -2022-12-06 11:23:13,330 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:23:13,330 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:23:13,336 - - -2022-12-06 11:23:13,336 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:23:14,279 - Epoch: [136][ 10/ 1200] Overall Loss 0.185197 Objective Loss 0.185197 LR 0.000500 Time 0.094287 -2022-12-06 11:23:14,490 - Epoch: [136][ 20/ 1200] Overall Loss 0.194246 Objective Loss 0.194246 LR 0.000500 Time 0.057632 -2022-12-06 11:23:14,699 - Epoch: [136][ 30/ 1200] Overall Loss 0.185948 Objective Loss 0.185948 LR 0.000500 Time 0.045372 -2022-12-06 11:23:14,909 - Epoch: [136][ 40/ 1200] Overall Loss 0.184844 Objective Loss 0.184844 LR 0.000500 Time 0.039252 -2022-12-06 11:23:15,117 - Epoch: [136][ 50/ 1200] Overall Loss 0.185873 Objective Loss 0.185873 LR 0.000500 Time 0.035551 -2022-12-06 11:23:15,326 - Epoch: [136][ 60/ 1200] Overall Loss 0.186469 Objective Loss 0.186469 LR 0.000500 Time 0.033098 -2022-12-06 11:23:15,535 - Epoch: [136][ 70/ 1200] Overall Loss 0.184638 Objective Loss 0.184638 LR 0.000500 Time 0.031354 -2022-12-06 11:23:15,745 - Epoch: [136][ 80/ 1200] Overall Loss 0.186029 Objective Loss 0.186029 LR 0.000500 Time 0.030053 -2022-12-06 11:23:15,954 - Epoch: [136][ 90/ 1200] Overall Loss 0.185537 Objective Loss 0.185537 LR 0.000500 Time 0.029031 -2022-12-06 11:23:16,162 - Epoch: [136][ 100/ 1200] Overall Loss 0.187092 Objective Loss 0.187092 LR 0.000500 Time 0.028201 -2022-12-06 11:23:16,371 - Epoch: [136][ 110/ 1200] Overall Loss 0.185161 Objective Loss 0.185161 LR 0.000500 Time 0.027526 -2022-12-06 11:23:16,580 - Epoch: [136][ 120/ 1200] Overall Loss 0.185179 Objective Loss 0.185179 LR 0.000500 Time 0.026971 -2022-12-06 11:23:16,789 - Epoch: [136][ 130/ 1200] Overall Loss 0.185016 Objective Loss 0.185016 LR 0.000500 Time 0.026497 -2022-12-06 11:23:16,997 - Epoch: [136][ 140/ 1200] Overall Loss 0.184488 Objective Loss 0.184488 LR 0.000500 Time 0.026090 -2022-12-06 11:23:17,206 - Epoch: [136][ 150/ 1200] Overall Loss 0.184931 Objective Loss 0.184931 LR 0.000500 Time 0.025736 -2022-12-06 11:23:17,414 - Epoch: [136][ 160/ 1200] Overall Loss 0.186053 Objective Loss 0.186053 LR 0.000500 Time 0.025425 -2022-12-06 11:23:17,622 - Epoch: [136][ 170/ 1200] Overall Loss 0.187439 Objective Loss 0.187439 LR 0.000500 Time 0.025151 -2022-12-06 11:23:17,830 - Epoch: [136][ 180/ 1200] Overall Loss 0.187152 Objective Loss 0.187152 LR 0.000500 Time 0.024906 -2022-12-06 11:23:18,038 - Epoch: [136][ 190/ 1200] Overall Loss 0.187720 Objective Loss 0.187720 LR 0.000500 Time 0.024688 -2022-12-06 11:23:18,247 - Epoch: [136][ 200/ 1200] Overall Loss 0.188165 Objective Loss 0.188165 LR 0.000500 Time 0.024492 -2022-12-06 11:23:18,454 - Epoch: [136][ 210/ 1200] Overall Loss 0.188163 Objective Loss 0.188163 LR 0.000500 Time 0.024310 -2022-12-06 11:23:18,663 - Epoch: [136][ 220/ 1200] Overall Loss 0.188122 Objective Loss 0.188122 LR 0.000500 Time 0.024152 -2022-12-06 11:23:18,871 - Epoch: [136][ 230/ 1200] Overall Loss 0.186230 Objective Loss 0.186230 LR 0.000500 Time 0.024003 -2022-12-06 11:23:19,080 - Epoch: [136][ 240/ 1200] Overall Loss 0.186221 Objective Loss 0.186221 LR 0.000500 Time 0.023872 -2022-12-06 11:23:19,288 - Epoch: [136][ 250/ 1200] Overall Loss 0.186671 Objective Loss 0.186671 LR 0.000500 Time 0.023745 -2022-12-06 11:23:19,497 - Epoch: [136][ 260/ 1200] Overall Loss 0.185089 Objective Loss 0.185089 LR 0.000500 Time 0.023635 -2022-12-06 11:23:19,706 - Epoch: [136][ 270/ 1200] Overall Loss 0.184257 Objective Loss 0.184257 LR 0.000500 Time 0.023530 -2022-12-06 11:23:19,914 - Epoch: [136][ 280/ 1200] Overall Loss 0.183879 Objective Loss 0.183879 LR 0.000500 Time 0.023433 -2022-12-06 11:23:20,121 - Epoch: [136][ 290/ 1200] Overall Loss 0.184220 Objective Loss 0.184220 LR 0.000500 Time 0.023338 -2022-12-06 11:23:20,330 - Epoch: [136][ 300/ 1200] Overall Loss 0.184224 Objective Loss 0.184224 LR 0.000500 Time 0.023253 -2022-12-06 11:23:20,533 - Epoch: [136][ 310/ 1200] Overall Loss 0.184376 Objective Loss 0.184376 LR 0.000500 Time 0.023154 -2022-12-06 11:23:20,737 - Epoch: [136][ 320/ 1200] Overall Loss 0.184343 Objective Loss 0.184343 LR 0.000500 Time 0.023067 -2022-12-06 11:23:20,937 - Epoch: [136][ 330/ 1200] Overall Loss 0.184279 Objective Loss 0.184279 LR 0.000500 Time 0.022974 -2022-12-06 11:23:21,142 - Epoch: [136][ 340/ 1200] Overall Loss 0.184804 Objective Loss 0.184804 LR 0.000500 Time 0.022899 -2022-12-06 11:23:21,342 - Epoch: [136][ 350/ 1200] Overall Loss 0.185320 Objective Loss 0.185320 LR 0.000500 Time 0.022814 -2022-12-06 11:23:21,546 - Epoch: [136][ 360/ 1200] Overall Loss 0.185660 Objective Loss 0.185660 LR 0.000500 Time 0.022744 -2022-12-06 11:23:21,746 - Epoch: [136][ 370/ 1200] Overall Loss 0.185367 Objective Loss 0.185367 LR 0.000500 Time 0.022668 -2022-12-06 11:23:21,950 - Epoch: [136][ 380/ 1200] Overall Loss 0.185407 Objective Loss 0.185407 LR 0.000500 Time 0.022608 -2022-12-06 11:23:22,151 - Epoch: [136][ 390/ 1200] Overall Loss 0.185720 Objective Loss 0.185720 LR 0.000500 Time 0.022541 -2022-12-06 11:23:22,355 - Epoch: [136][ 400/ 1200] Overall Loss 0.185665 Objective Loss 0.185665 LR 0.000500 Time 0.022486 -2022-12-06 11:23:22,555 - Epoch: [136][ 410/ 1200] Overall Loss 0.186505 Objective Loss 0.186505 LR 0.000500 Time 0.022425 -2022-12-06 11:23:22,759 - Epoch: [136][ 420/ 1200] Overall Loss 0.186687 Objective Loss 0.186687 LR 0.000500 Time 0.022374 -2022-12-06 11:23:22,960 - Epoch: [136][ 430/ 1200] Overall Loss 0.186343 Objective Loss 0.186343 LR 0.000500 Time 0.022320 -2022-12-06 11:23:23,163 - Epoch: [136][ 440/ 1200] Overall Loss 0.185882 Objective Loss 0.185882 LR 0.000500 Time 0.022274 -2022-12-06 11:23:23,363 - Epoch: [136][ 450/ 1200] Overall Loss 0.185746 Objective Loss 0.185746 LR 0.000500 Time 0.022221 -2022-12-06 11:23:23,567 - Epoch: [136][ 460/ 1200] Overall Loss 0.185675 Objective Loss 0.185675 LR 0.000500 Time 0.022181 -2022-12-06 11:23:23,768 - Epoch: [136][ 470/ 1200] Overall Loss 0.185842 Objective Loss 0.185842 LR 0.000500 Time 0.022134 -2022-12-06 11:23:23,972 - Epoch: [136][ 480/ 1200] Overall Loss 0.185921 Objective Loss 0.185921 LR 0.000500 Time 0.022098 -2022-12-06 11:23:24,173 - Epoch: [136][ 490/ 1200] Overall Loss 0.185691 Objective Loss 0.185691 LR 0.000500 Time 0.022056 -2022-12-06 11:23:24,378 - Epoch: [136][ 500/ 1200] Overall Loss 0.185623 Objective Loss 0.185623 LR 0.000500 Time 0.022024 -2022-12-06 11:23:24,580 - Epoch: [136][ 510/ 1200] Overall Loss 0.185663 Objective Loss 0.185663 LR 0.000500 Time 0.021986 -2022-12-06 11:23:24,785 - Epoch: [136][ 520/ 1200] Overall Loss 0.185187 Objective Loss 0.185187 LR 0.000500 Time 0.021957 -2022-12-06 11:23:24,986 - Epoch: [136][ 530/ 1200] Overall Loss 0.185212 Objective Loss 0.185212 LR 0.000500 Time 0.021920 -2022-12-06 11:23:25,190 - Epoch: [136][ 540/ 1200] Overall Loss 0.184838 Objective Loss 0.184838 LR 0.000500 Time 0.021892 -2022-12-06 11:23:25,391 - Epoch: [136][ 550/ 1200] Overall Loss 0.184847 Objective Loss 0.184847 LR 0.000500 Time 0.021858 -2022-12-06 11:23:25,595 - Epoch: [136][ 560/ 1200] Overall Loss 0.185031 Objective Loss 0.185031 LR 0.000500 Time 0.021831 -2022-12-06 11:23:25,796 - Epoch: [136][ 570/ 1200] Overall Loss 0.185536 Objective Loss 0.185536 LR 0.000500 Time 0.021798 -2022-12-06 11:23:26,000 - Epoch: [136][ 580/ 1200] Overall Loss 0.185364 Objective Loss 0.185364 LR 0.000500 Time 0.021773 -2022-12-06 11:23:26,200 - Epoch: [136][ 590/ 1200] Overall Loss 0.185639 Objective Loss 0.185639 LR 0.000500 Time 0.021742 -2022-12-06 11:23:26,404 - Epoch: [136][ 600/ 1200] Overall Loss 0.185792 Objective Loss 0.185792 LR 0.000500 Time 0.021719 -2022-12-06 11:23:26,604 - Epoch: [136][ 610/ 1200] Overall Loss 0.185325 Objective Loss 0.185325 LR 0.000500 Time 0.021691 -2022-12-06 11:23:26,809 - Epoch: [136][ 620/ 1200] Overall Loss 0.185242 Objective Loss 0.185242 LR 0.000500 Time 0.021670 -2022-12-06 11:23:27,009 - Epoch: [136][ 630/ 1200] Overall Loss 0.185339 Objective Loss 0.185339 LR 0.000500 Time 0.021643 -2022-12-06 11:23:27,213 - Epoch: [136][ 640/ 1200] Overall Loss 0.185650 Objective Loss 0.185650 LR 0.000500 Time 0.021623 -2022-12-06 11:23:27,413 - Epoch: [136][ 650/ 1200] Overall Loss 0.185459 Objective Loss 0.185459 LR 0.000500 Time 0.021597 -2022-12-06 11:23:27,618 - Epoch: [136][ 660/ 1200] Overall Loss 0.185290 Objective Loss 0.185290 LR 0.000500 Time 0.021579 -2022-12-06 11:23:27,818 - Epoch: [136][ 670/ 1200] Overall Loss 0.185538 Objective Loss 0.185538 LR 0.000500 Time 0.021555 -2022-12-06 11:23:28,022 - Epoch: [136][ 680/ 1200] Overall Loss 0.185291 Objective Loss 0.185291 LR 0.000500 Time 0.021537 -2022-12-06 11:23:28,222 - Epoch: [136][ 690/ 1200] Overall Loss 0.185325 Objective Loss 0.185325 LR 0.000500 Time 0.021513 -2022-12-06 11:23:28,426 - Epoch: [136][ 700/ 1200] Overall Loss 0.184924 Objective Loss 0.184924 LR 0.000500 Time 0.021497 -2022-12-06 11:23:28,628 - Epoch: [136][ 710/ 1200] Overall Loss 0.184971 Objective Loss 0.184971 LR 0.000500 Time 0.021477 -2022-12-06 11:23:28,831 - Epoch: [136][ 720/ 1200] Overall Loss 0.185578 Objective Loss 0.185578 LR 0.000500 Time 0.021461 -2022-12-06 11:23:29,031 - Epoch: [136][ 730/ 1200] Overall Loss 0.185582 Objective Loss 0.185582 LR 0.000500 Time 0.021440 -2022-12-06 11:23:29,235 - Epoch: [136][ 740/ 1200] Overall Loss 0.185403 Objective Loss 0.185403 LR 0.000500 Time 0.021425 -2022-12-06 11:23:29,436 - Epoch: [136][ 750/ 1200] Overall Loss 0.185287 Objective Loss 0.185287 LR 0.000500 Time 0.021405 -2022-12-06 11:23:29,640 - Epoch: [136][ 760/ 1200] Overall Loss 0.185349 Objective Loss 0.185349 LR 0.000500 Time 0.021392 -2022-12-06 11:23:29,840 - Epoch: [136][ 770/ 1200] Overall Loss 0.185232 Objective Loss 0.185232 LR 0.000500 Time 0.021373 -2022-12-06 11:23:30,044 - Epoch: [136][ 780/ 1200] Overall Loss 0.185209 Objective Loss 0.185209 LR 0.000500 Time 0.021360 -2022-12-06 11:23:30,244 - Epoch: [136][ 790/ 1200] Overall Loss 0.185296 Objective Loss 0.185296 LR 0.000500 Time 0.021342 -2022-12-06 11:23:30,448 - Epoch: [136][ 800/ 1200] Overall Loss 0.185152 Objective Loss 0.185152 LR 0.000500 Time 0.021329 -2022-12-06 11:23:30,649 - Epoch: [136][ 810/ 1200] Overall Loss 0.185028 Objective Loss 0.185028 LR 0.000500 Time 0.021314 -2022-12-06 11:23:30,854 - Epoch: [136][ 820/ 1200] Overall Loss 0.184911 Objective Loss 0.184911 LR 0.000500 Time 0.021302 -2022-12-06 11:23:31,054 - Epoch: [136][ 830/ 1200] Overall Loss 0.184922 Objective Loss 0.184922 LR 0.000500 Time 0.021287 -2022-12-06 11:23:31,258 - Epoch: [136][ 840/ 1200] Overall Loss 0.184970 Objective Loss 0.184970 LR 0.000500 Time 0.021276 -2022-12-06 11:23:31,459 - Epoch: [136][ 850/ 1200] Overall Loss 0.184712 Objective Loss 0.184712 LR 0.000500 Time 0.021260 -2022-12-06 11:23:31,663 - Epoch: [136][ 860/ 1200] Overall Loss 0.184699 Objective Loss 0.184699 LR 0.000500 Time 0.021249 -2022-12-06 11:23:31,862 - Epoch: [136][ 870/ 1200] Overall Loss 0.184994 Objective Loss 0.184994 LR 0.000500 Time 0.021234 -2022-12-06 11:23:32,067 - Epoch: [136][ 880/ 1200] Overall Loss 0.184851 Objective Loss 0.184851 LR 0.000500 Time 0.021224 -2022-12-06 11:23:32,268 - Epoch: [136][ 890/ 1200] Overall Loss 0.184992 Objective Loss 0.184992 LR 0.000500 Time 0.021211 -2022-12-06 11:23:32,471 - Epoch: [136][ 900/ 1200] Overall Loss 0.185010 Objective Loss 0.185010 LR 0.000500 Time 0.021200 -2022-12-06 11:23:32,671 - Epoch: [136][ 910/ 1200] Overall Loss 0.185389 Objective Loss 0.185389 LR 0.000500 Time 0.021187 -2022-12-06 11:23:32,876 - Epoch: [136][ 920/ 1200] Overall Loss 0.185401 Objective Loss 0.185401 LR 0.000500 Time 0.021178 -2022-12-06 11:23:33,076 - Epoch: [136][ 930/ 1200] Overall Loss 0.185419 Objective Loss 0.185419 LR 0.000500 Time 0.021165 -2022-12-06 11:23:33,281 - Epoch: [136][ 940/ 1200] Overall Loss 0.185502 Objective Loss 0.185502 LR 0.000500 Time 0.021157 -2022-12-06 11:23:33,481 - Epoch: [136][ 950/ 1200] Overall Loss 0.185503 Objective Loss 0.185503 LR 0.000500 Time 0.021145 -2022-12-06 11:23:33,686 - Epoch: [136][ 960/ 1200] Overall Loss 0.185590 Objective Loss 0.185590 LR 0.000500 Time 0.021137 -2022-12-06 11:23:33,887 - Epoch: [136][ 970/ 1200] Overall Loss 0.185677 Objective Loss 0.185677 LR 0.000500 Time 0.021126 -2022-12-06 11:23:34,091 - Epoch: [136][ 980/ 1200] Overall Loss 0.185619 Objective Loss 0.185619 LR 0.000500 Time 0.021118 -2022-12-06 11:23:34,290 - Epoch: [136][ 990/ 1200] Overall Loss 0.185422 Objective Loss 0.185422 LR 0.000500 Time 0.021106 -2022-12-06 11:23:34,495 - Epoch: [136][ 1000/ 1200] Overall Loss 0.185324 Objective Loss 0.185324 LR 0.000500 Time 0.021098 -2022-12-06 11:23:34,695 - Epoch: [136][ 1010/ 1200] Overall Loss 0.185591 Objective Loss 0.185591 LR 0.000500 Time 0.021087 -2022-12-06 11:23:34,899 - Epoch: [136][ 1020/ 1200] Overall Loss 0.185680 Objective Loss 0.185680 LR 0.000500 Time 0.021080 -2022-12-06 11:23:35,100 - Epoch: [136][ 1030/ 1200] Overall Loss 0.185884 Objective Loss 0.185884 LR 0.000500 Time 0.021069 -2022-12-06 11:23:35,304 - Epoch: [136][ 1040/ 1200] Overall Loss 0.186042 Objective Loss 0.186042 LR 0.000500 Time 0.021063 -2022-12-06 11:23:35,506 - Epoch: [136][ 1050/ 1200] Overall Loss 0.186109 Objective Loss 0.186109 LR 0.000500 Time 0.021054 -2022-12-06 11:23:35,710 - Epoch: [136][ 1060/ 1200] Overall Loss 0.186082 Objective Loss 0.186082 LR 0.000500 Time 0.021047 -2022-12-06 11:23:35,911 - Epoch: [136][ 1070/ 1200] Overall Loss 0.186105 Objective Loss 0.186105 LR 0.000500 Time 0.021038 -2022-12-06 11:23:36,115 - Epoch: [136][ 1080/ 1200] Overall Loss 0.185988 Objective Loss 0.185988 LR 0.000500 Time 0.021031 -2022-12-06 11:23:36,316 - Epoch: [136][ 1090/ 1200] Overall Loss 0.185788 Objective Loss 0.185788 LR 0.000500 Time 0.021022 -2022-12-06 11:23:36,519 - Epoch: [136][ 1100/ 1200] Overall Loss 0.185850 Objective Loss 0.185850 LR 0.000500 Time 0.021015 -2022-12-06 11:23:36,720 - Epoch: [136][ 1110/ 1200] Overall Loss 0.185760 Objective Loss 0.185760 LR 0.000500 Time 0.021006 -2022-12-06 11:23:36,924 - Epoch: [136][ 1120/ 1200] Overall Loss 0.185754 Objective Loss 0.185754 LR 0.000500 Time 0.021000 -2022-12-06 11:23:37,124 - Epoch: [136][ 1130/ 1200] Overall Loss 0.185685 Objective Loss 0.185685 LR 0.000500 Time 0.020991 -2022-12-06 11:23:37,328 - Epoch: [136][ 1140/ 1200] Overall Loss 0.185795 Objective Loss 0.185795 LR 0.000500 Time 0.020985 -2022-12-06 11:23:37,528 - Epoch: [136][ 1150/ 1200] Overall Loss 0.185880 Objective Loss 0.185880 LR 0.000500 Time 0.020976 -2022-12-06 11:23:37,732 - Epoch: [136][ 1160/ 1200] Overall Loss 0.185888 Objective Loss 0.185888 LR 0.000500 Time 0.020971 -2022-12-06 11:23:37,933 - Epoch: [136][ 1170/ 1200] Overall Loss 0.185909 Objective Loss 0.185909 LR 0.000500 Time 0.020962 -2022-12-06 11:23:38,137 - Epoch: [136][ 1180/ 1200] Overall Loss 0.186143 Objective Loss 0.186143 LR 0.000500 Time 0.020957 -2022-12-06 11:23:38,337 - Epoch: [136][ 1190/ 1200] Overall Loss 0.186304 Objective Loss 0.186304 LR 0.000500 Time 0.020949 -2022-12-06 11:23:38,571 - Epoch: [136][ 1200/ 1200] Overall Loss 0.186415 Objective Loss 0.186415 Top1 87.866109 Top5 99.581590 LR 0.000500 Time 0.020969 -2022-12-06 11:23:38,659 - --- validate (epoch=136)----------- -2022-12-06 11:23:38,660 - 34129 samples (256 per mini-batch) -2022-12-06 11:23:39,100 - Epoch: [136][ 10/ 134] Loss 0.253957 Top1 85.546875 Top5 98.125000 -2022-12-06 11:23:39,231 - Epoch: [136][ 20/ 134] Loss 0.266617 Top1 85.976562 Top5 98.007812 -2022-12-06 11:23:39,364 - Epoch: [136][ 30/ 134] Loss 0.258673 Top1 85.885417 Top5 98.072917 -2022-12-06 11:23:39,497 - Epoch: [136][ 40/ 134] Loss 0.258653 Top1 85.888672 Top5 98.027344 -2022-12-06 11:23:39,631 - Epoch: [136][ 50/ 134] Loss 0.253516 Top1 86.039062 Top5 98.093750 -2022-12-06 11:23:39,764 - Epoch: [136][ 60/ 134] Loss 0.253980 Top1 86.002604 Top5 98.118490 -2022-12-06 11:23:39,896 - Epoch: [136][ 70/ 134] Loss 0.257245 Top1 86.071429 Top5 98.130580 -2022-12-06 11:23:40,027 - Epoch: [136][ 80/ 134] Loss 0.254109 Top1 86.103516 Top5 98.134766 -2022-12-06 11:23:40,157 - Epoch: [136][ 90/ 134] Loss 0.252158 Top1 86.184896 Top5 98.129340 -2022-12-06 11:23:40,287 - Epoch: [136][ 100/ 134] Loss 0.255098 Top1 86.160156 Top5 98.121094 -2022-12-06 11:23:40,417 - Epoch: [136][ 110/ 134] Loss 0.252725 Top1 86.239347 Top5 98.160511 -2022-12-06 11:23:40,549 - Epoch: [136][ 120/ 134] Loss 0.253458 Top1 86.298828 Top5 98.177083 -2022-12-06 11:23:40,681 - Epoch: [136][ 130/ 134] Loss 0.253456 Top1 86.213942 Top5 98.161058 -2022-12-06 11:23:40,720 - Epoch: [136][ 134/ 134] Loss 0.253711 Top1 86.199420 Top5 98.165783 -2022-12-06 11:23:40,810 - ==> Top1: 86.199 Top5: 98.166 Loss: 0.254 - -2022-12-06 11:23:40,811 - ==> Confusion: -[[ 918 3 0 0 7 4 0 0 4 46 0 0 2 4 4 1 0 1 0 0 2] - [ 3 944 0 1 7 19 3 12 3 0 2 3 0 1 0 3 3 1 8 5 9] - [ 4 2 993 15 5 3 25 8 1 2 10 3 1 5 1 1 1 2 8 5 8] - [ 6 1 14 940 2 3 1 1 1 0 13 0 4 3 7 0 1 1 14 1 7] - [ 15 6 0 0 952 5 0 1 0 4 0 3 1 5 6 8 8 1 1 1 3] - [ 3 9 0 2 6 977 4 23 1 1 1 15 5 13 2 1 0 1 0 4 1] - [ 1 2 5 2 0 1 1082 5 0 0 2 1 1 1 0 3 1 1 1 8 1] - [ 1 5 5 1 2 21 9 954 0 0 0 5 0 4 1 2 0 1 32 8 3] - [ 5 0 0 0 0 1 1 0 985 32 13 0 0 11 7 1 2 1 3 2 0] - [ 60 1 1 0 3 3 0 1 27 879 2 1 0 16 2 1 0 2 0 1 1] - [ 1 1 2 4 1 1 1 2 6 3 970 2 2 9 4 0 2 0 3 1 4] - [ 2 3 0 0 0 12 3 3 1 0 0 975 20 5 0 7 4 3 1 11 1] - [ 3 0 1 1 0 2 0 1 1 0 1 30 896 2 0 5 1 13 1 5 6] - [ 1 1 1 0 0 7 0 2 6 10 3 4 2 971 1 1 3 0 0 3 7] - [ 11 3 1 10 4 3 0 0 18 0 1 2 4 6 1050 0 1 1 7 1 7] - [ 0 1 1 0 2 1 3 0 1 0 0 11 7 2 0 994 5 10 0 3 2] - [ 2 2 1 1 2 1 4 1 1 0 0 3 3 3 0 7 1030 0 1 6 4] - [ 2 0 1 1 1 1 2 1 0 3 0 7 9 2 1 13 0 987 0 1 4] - [ 4 5 3 4 2 0 1 21 2 1 2 1 3 1 5 2 0 0 948 1 2] - [ 2 3 2 0 0 5 6 9 0 1 2 14 8 6 0 2 2 2 2 1010 4] - [ 154 217 161 106 82 188 107 155 96 97 191 103 355 307 140 103 172 81 183 265 9963]] - -2022-12-06 11:23:41,481 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:23:41,481 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:23:41,487 - - -2022-12-06 11:23:41,487 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:23:42,420 - Epoch: [137][ 10/ 1200] Overall Loss 0.201940 Objective Loss 0.201940 LR 0.000500 Time 0.093249 -2022-12-06 11:23:42,623 - Epoch: [137][ 20/ 1200] Overall Loss 0.197425 Objective Loss 0.197425 LR 0.000500 Time 0.056743 -2022-12-06 11:23:42,815 - Epoch: [137][ 30/ 1200] Overall Loss 0.195774 Objective Loss 0.195774 LR 0.000500 Time 0.044206 -2022-12-06 11:23:43,006 - Epoch: [137][ 40/ 1200] Overall Loss 0.193092 Objective Loss 0.193092 LR 0.000500 Time 0.037933 -2022-12-06 11:23:43,198 - Epoch: [137][ 50/ 1200] Overall Loss 0.194089 Objective Loss 0.194089 LR 0.000500 Time 0.034168 -2022-12-06 11:23:43,389 - Epoch: [137][ 60/ 1200] Overall Loss 0.195839 Objective Loss 0.195839 LR 0.000500 Time 0.031655 -2022-12-06 11:23:43,580 - Epoch: [137][ 70/ 1200] Overall Loss 0.193315 Objective Loss 0.193315 LR 0.000500 Time 0.029852 -2022-12-06 11:23:43,772 - Epoch: [137][ 80/ 1200] Overall Loss 0.189896 Objective Loss 0.189896 LR 0.000500 Time 0.028511 -2022-12-06 11:23:43,964 - Epoch: [137][ 90/ 1200] Overall Loss 0.189116 Objective Loss 0.189116 LR 0.000500 Time 0.027466 -2022-12-06 11:23:44,155 - Epoch: [137][ 100/ 1200] Overall Loss 0.188902 Objective Loss 0.188902 LR 0.000500 Time 0.026630 -2022-12-06 11:23:44,347 - Epoch: [137][ 110/ 1200] Overall Loss 0.188044 Objective Loss 0.188044 LR 0.000500 Time 0.025947 -2022-12-06 11:23:44,539 - Epoch: [137][ 120/ 1200] Overall Loss 0.187584 Objective Loss 0.187584 LR 0.000500 Time 0.025382 -2022-12-06 11:23:44,730 - Epoch: [137][ 130/ 1200] Overall Loss 0.185950 Objective Loss 0.185950 LR 0.000500 Time 0.024896 -2022-12-06 11:23:44,922 - Epoch: [137][ 140/ 1200] Overall Loss 0.184973 Objective Loss 0.184973 LR 0.000500 Time 0.024479 -2022-12-06 11:23:45,113 - Epoch: [137][ 150/ 1200] Overall Loss 0.187348 Objective Loss 0.187348 LR 0.000500 Time 0.024120 -2022-12-06 11:23:45,304 - Epoch: [137][ 160/ 1200] Overall Loss 0.185280 Objective Loss 0.185280 LR 0.000500 Time 0.023804 -2022-12-06 11:23:45,496 - Epoch: [137][ 170/ 1200] Overall Loss 0.183909 Objective Loss 0.183909 LR 0.000500 Time 0.023526 -2022-12-06 11:23:45,687 - Epoch: [137][ 180/ 1200] Overall Loss 0.182197 Objective Loss 0.182197 LR 0.000500 Time 0.023278 -2022-12-06 11:23:45,878 - Epoch: [137][ 190/ 1200] Overall Loss 0.182577 Objective Loss 0.182577 LR 0.000500 Time 0.023058 -2022-12-06 11:23:46,070 - Epoch: [137][ 200/ 1200] Overall Loss 0.183038 Objective Loss 0.183038 LR 0.000500 Time 0.022860 -2022-12-06 11:23:46,261 - Epoch: [137][ 210/ 1200] Overall Loss 0.182784 Objective Loss 0.182784 LR 0.000500 Time 0.022678 -2022-12-06 11:23:46,452 - Epoch: [137][ 220/ 1200] Overall Loss 0.183016 Objective Loss 0.183016 LR 0.000500 Time 0.022515 -2022-12-06 11:23:46,643 - Epoch: [137][ 230/ 1200] Overall Loss 0.182063 Objective Loss 0.182063 LR 0.000500 Time 0.022364 -2022-12-06 11:23:46,834 - Epoch: [137][ 240/ 1200] Overall Loss 0.182127 Objective Loss 0.182127 LR 0.000500 Time 0.022226 -2022-12-06 11:23:47,026 - Epoch: [137][ 250/ 1200] Overall Loss 0.182033 Objective Loss 0.182033 LR 0.000500 Time 0.022100 -2022-12-06 11:23:47,217 - Epoch: [137][ 260/ 1200] Overall Loss 0.181154 Objective Loss 0.181154 LR 0.000500 Time 0.021985 -2022-12-06 11:23:47,408 - Epoch: [137][ 270/ 1200] Overall Loss 0.180690 Objective Loss 0.180690 LR 0.000500 Time 0.021876 -2022-12-06 11:23:47,600 - Epoch: [137][ 280/ 1200] Overall Loss 0.181198 Objective Loss 0.181198 LR 0.000500 Time 0.021778 -2022-12-06 11:23:47,791 - Epoch: [137][ 290/ 1200] Overall Loss 0.181887 Objective Loss 0.181887 LR 0.000500 Time 0.021685 -2022-12-06 11:23:47,983 - Epoch: [137][ 300/ 1200] Overall Loss 0.182390 Objective Loss 0.182390 LR 0.000500 Time 0.021598 -2022-12-06 11:23:48,174 - Epoch: [137][ 310/ 1200] Overall Loss 0.182043 Objective Loss 0.182043 LR 0.000500 Time 0.021518 -2022-12-06 11:23:48,365 - Epoch: [137][ 320/ 1200] Overall Loss 0.181844 Objective Loss 0.181844 LR 0.000500 Time 0.021441 -2022-12-06 11:23:48,557 - Epoch: [137][ 330/ 1200] Overall Loss 0.181991 Objective Loss 0.181991 LR 0.000500 Time 0.021369 -2022-12-06 11:23:48,748 - Epoch: [137][ 340/ 1200] Overall Loss 0.182478 Objective Loss 0.182478 LR 0.000500 Time 0.021301 -2022-12-06 11:23:48,939 - Epoch: [137][ 350/ 1200] Overall Loss 0.182844 Objective Loss 0.182844 LR 0.000500 Time 0.021236 -2022-12-06 11:23:49,130 - Epoch: [137][ 360/ 1200] Overall Loss 0.183965 Objective Loss 0.183965 LR 0.000500 Time 0.021178 -2022-12-06 11:23:49,322 - Epoch: [137][ 370/ 1200] Overall Loss 0.183408 Objective Loss 0.183408 LR 0.000500 Time 0.021121 -2022-12-06 11:23:49,513 - Epoch: [137][ 380/ 1200] Overall Loss 0.183319 Objective Loss 0.183319 LR 0.000500 Time 0.021067 -2022-12-06 11:23:49,704 - Epoch: [137][ 390/ 1200] Overall Loss 0.184006 Objective Loss 0.184006 LR 0.000500 Time 0.021016 -2022-12-06 11:23:49,896 - Epoch: [137][ 400/ 1200] Overall Loss 0.184444 Objective Loss 0.184444 LR 0.000500 Time 0.020968 -2022-12-06 11:23:50,087 - Epoch: [137][ 410/ 1200] Overall Loss 0.184997 Objective Loss 0.184997 LR 0.000500 Time 0.020921 -2022-12-06 11:23:50,279 - Epoch: [137][ 420/ 1200] Overall Loss 0.185217 Objective Loss 0.185217 LR 0.000500 Time 0.020878 -2022-12-06 11:23:50,470 - Epoch: [137][ 430/ 1200] Overall Loss 0.185310 Objective Loss 0.185310 LR 0.000500 Time 0.020837 -2022-12-06 11:23:50,662 - Epoch: [137][ 440/ 1200] Overall Loss 0.185261 Objective Loss 0.185261 LR 0.000500 Time 0.020798 -2022-12-06 11:23:50,853 - Epoch: [137][ 450/ 1200] Overall Loss 0.185289 Objective Loss 0.185289 LR 0.000500 Time 0.020760 -2022-12-06 11:23:51,045 - Epoch: [137][ 460/ 1200] Overall Loss 0.185216 Objective Loss 0.185216 LR 0.000500 Time 0.020725 -2022-12-06 11:23:51,236 - Epoch: [137][ 470/ 1200] Overall Loss 0.185036 Objective Loss 0.185036 LR 0.000500 Time 0.020689 -2022-12-06 11:23:51,428 - Epoch: [137][ 480/ 1200] Overall Loss 0.185620 Objective Loss 0.185620 LR 0.000500 Time 0.020656 -2022-12-06 11:23:51,619 - Epoch: [137][ 490/ 1200] Overall Loss 0.185418 Objective Loss 0.185418 LR 0.000500 Time 0.020624 -2022-12-06 11:23:51,811 - Epoch: [137][ 500/ 1200] Overall Loss 0.185289 Objective Loss 0.185289 LR 0.000500 Time 0.020594 -2022-12-06 11:23:52,003 - Epoch: [137][ 510/ 1200] Overall Loss 0.184936 Objective Loss 0.184936 LR 0.000500 Time 0.020565 -2022-12-06 11:23:52,195 - Epoch: [137][ 520/ 1200] Overall Loss 0.184753 Objective Loss 0.184753 LR 0.000500 Time 0.020537 -2022-12-06 11:23:52,386 - Epoch: [137][ 530/ 1200] Overall Loss 0.184748 Objective Loss 0.184748 LR 0.000500 Time 0.020511 -2022-12-06 11:23:52,578 - Epoch: [137][ 540/ 1200] Overall Loss 0.184906 Objective Loss 0.184906 LR 0.000500 Time 0.020486 -2022-12-06 11:23:52,770 - Epoch: [137][ 550/ 1200] Overall Loss 0.184743 Objective Loss 0.184743 LR 0.000500 Time 0.020461 -2022-12-06 11:23:52,962 - Epoch: [137][ 560/ 1200] Overall Loss 0.184720 Objective Loss 0.184720 LR 0.000500 Time 0.020437 -2022-12-06 11:23:53,154 - Epoch: [137][ 570/ 1200] Overall Loss 0.185225 Objective Loss 0.185225 LR 0.000500 Time 0.020414 -2022-12-06 11:23:53,346 - Epoch: [137][ 580/ 1200] Overall Loss 0.185075 Objective Loss 0.185075 LR 0.000500 Time 0.020392 -2022-12-06 11:23:53,538 - Epoch: [137][ 590/ 1200] Overall Loss 0.185006 Objective Loss 0.185006 LR 0.000500 Time 0.020371 -2022-12-06 11:23:53,730 - Epoch: [137][ 600/ 1200] Overall Loss 0.185292 Objective Loss 0.185292 LR 0.000500 Time 0.020350 -2022-12-06 11:23:53,921 - Epoch: [137][ 610/ 1200] Overall Loss 0.185632 Objective Loss 0.185632 LR 0.000500 Time 0.020330 -2022-12-06 11:23:54,113 - Epoch: [137][ 620/ 1200] Overall Loss 0.185526 Objective Loss 0.185526 LR 0.000500 Time 0.020310 -2022-12-06 11:23:54,304 - Epoch: [137][ 630/ 1200] Overall Loss 0.185251 Objective Loss 0.185251 LR 0.000500 Time 0.020291 -2022-12-06 11:23:54,496 - Epoch: [137][ 640/ 1200] Overall Loss 0.185570 Objective Loss 0.185570 LR 0.000500 Time 0.020273 -2022-12-06 11:23:54,688 - Epoch: [137][ 650/ 1200] Overall Loss 0.185776 Objective Loss 0.185776 LR 0.000500 Time 0.020255 -2022-12-06 11:23:54,880 - Epoch: [137][ 660/ 1200] Overall Loss 0.185554 Objective Loss 0.185554 LR 0.000500 Time 0.020238 -2022-12-06 11:23:55,072 - Epoch: [137][ 670/ 1200] Overall Loss 0.185159 Objective Loss 0.185159 LR 0.000500 Time 0.020222 -2022-12-06 11:23:55,264 - Epoch: [137][ 680/ 1200] Overall Loss 0.185299 Objective Loss 0.185299 LR 0.000500 Time 0.020206 -2022-12-06 11:23:55,456 - Epoch: [137][ 690/ 1200] Overall Loss 0.185272 Objective Loss 0.185272 LR 0.000500 Time 0.020190 -2022-12-06 11:23:55,647 - Epoch: [137][ 700/ 1200] Overall Loss 0.185169 Objective Loss 0.185169 LR 0.000500 Time 0.020175 -2022-12-06 11:23:55,839 - Epoch: [137][ 710/ 1200] Overall Loss 0.185079 Objective Loss 0.185079 LR 0.000500 Time 0.020160 -2022-12-06 11:23:56,031 - Epoch: [137][ 720/ 1200] Overall Loss 0.184956 Objective Loss 0.184956 LR 0.000500 Time 0.020146 -2022-12-06 11:23:56,223 - Epoch: [137][ 730/ 1200] Overall Loss 0.185027 Objective Loss 0.185027 LR 0.000500 Time 0.020132 -2022-12-06 11:23:56,414 - Epoch: [137][ 740/ 1200] Overall Loss 0.185163 Objective Loss 0.185163 LR 0.000500 Time 0.020118 -2022-12-06 11:23:56,606 - Epoch: [137][ 750/ 1200] Overall Loss 0.185222 Objective Loss 0.185222 LR 0.000500 Time 0.020105 -2022-12-06 11:23:56,797 - Epoch: [137][ 760/ 1200] Overall Loss 0.185001 Objective Loss 0.185001 LR 0.000500 Time 0.020091 -2022-12-06 11:23:56,989 - Epoch: [137][ 770/ 1200] Overall Loss 0.185159 Objective Loss 0.185159 LR 0.000500 Time 0.020078 -2022-12-06 11:23:57,181 - Epoch: [137][ 780/ 1200] Overall Loss 0.184736 Objective Loss 0.184736 LR 0.000500 Time 0.020066 -2022-12-06 11:23:57,373 - Epoch: [137][ 790/ 1200] Overall Loss 0.184786 Objective Loss 0.184786 LR 0.000500 Time 0.020054 -2022-12-06 11:23:57,564 - Epoch: [137][ 800/ 1200] Overall Loss 0.184579 Objective Loss 0.184579 LR 0.000500 Time 0.020042 -2022-12-06 11:23:57,756 - Epoch: [137][ 810/ 1200] Overall Loss 0.184659 Objective Loss 0.184659 LR 0.000500 Time 0.020031 -2022-12-06 11:23:57,948 - Epoch: [137][ 820/ 1200] Overall Loss 0.184712 Objective Loss 0.184712 LR 0.000500 Time 0.020020 -2022-12-06 11:23:58,139 - Epoch: [137][ 830/ 1200] Overall Loss 0.184748 Objective Loss 0.184748 LR 0.000500 Time 0.020009 -2022-12-06 11:23:58,331 - Epoch: [137][ 840/ 1200] Overall Loss 0.184653 Objective Loss 0.184653 LR 0.000500 Time 0.019998 -2022-12-06 11:23:58,523 - Epoch: [137][ 850/ 1200] Overall Loss 0.184716 Objective Loss 0.184716 LR 0.000500 Time 0.019988 -2022-12-06 11:23:58,714 - Epoch: [137][ 860/ 1200] Overall Loss 0.184530 Objective Loss 0.184530 LR 0.000500 Time 0.019977 -2022-12-06 11:23:58,906 - Epoch: [137][ 870/ 1200] Overall Loss 0.184531 Objective Loss 0.184531 LR 0.000500 Time 0.019967 -2022-12-06 11:23:59,097 - Epoch: [137][ 880/ 1200] Overall Loss 0.184451 Objective Loss 0.184451 LR 0.000500 Time 0.019957 -2022-12-06 11:23:59,288 - Epoch: [137][ 890/ 1200] Overall Loss 0.184475 Objective Loss 0.184475 LR 0.000500 Time 0.019946 -2022-12-06 11:23:59,479 - Epoch: [137][ 900/ 1200] Overall Loss 0.184630 Objective Loss 0.184630 LR 0.000500 Time 0.019937 -2022-12-06 11:23:59,670 - Epoch: [137][ 910/ 1200] Overall Loss 0.184750 Objective Loss 0.184750 LR 0.000500 Time 0.019927 -2022-12-06 11:23:59,862 - Epoch: [137][ 920/ 1200] Overall Loss 0.184734 Objective Loss 0.184734 LR 0.000500 Time 0.019919 -2022-12-06 11:24:00,054 - Epoch: [137][ 930/ 1200] Overall Loss 0.184855 Objective Loss 0.184855 LR 0.000500 Time 0.019910 -2022-12-06 11:24:00,246 - Epoch: [137][ 940/ 1200] Overall Loss 0.184547 Objective Loss 0.184547 LR 0.000500 Time 0.019902 -2022-12-06 11:24:00,438 - Epoch: [137][ 950/ 1200] Overall Loss 0.184448 Objective Loss 0.184448 LR 0.000500 Time 0.019894 -2022-12-06 11:24:00,629 - Epoch: [137][ 960/ 1200] Overall Loss 0.184310 Objective Loss 0.184310 LR 0.000500 Time 0.019885 -2022-12-06 11:24:00,820 - Epoch: [137][ 970/ 1200] Overall Loss 0.184446 Objective Loss 0.184446 LR 0.000500 Time 0.019877 -2022-12-06 11:24:01,012 - Epoch: [137][ 980/ 1200] Overall Loss 0.184453 Objective Loss 0.184453 LR 0.000500 Time 0.019869 -2022-12-06 11:24:01,203 - Epoch: [137][ 990/ 1200] Overall Loss 0.184432 Objective Loss 0.184432 LR 0.000500 Time 0.019860 -2022-12-06 11:24:01,394 - Epoch: [137][ 1000/ 1200] Overall Loss 0.184644 Objective Loss 0.184644 LR 0.000500 Time 0.019852 -2022-12-06 11:24:01,585 - Epoch: [137][ 1010/ 1200] Overall Loss 0.184672 Objective Loss 0.184672 LR 0.000500 Time 0.019845 -2022-12-06 11:24:01,777 - Epoch: [137][ 1020/ 1200] Overall Loss 0.184659 Objective Loss 0.184659 LR 0.000500 Time 0.019837 -2022-12-06 11:24:01,968 - Epoch: [137][ 1030/ 1200] Overall Loss 0.184637 Objective Loss 0.184637 LR 0.000500 Time 0.019830 -2022-12-06 11:24:02,160 - Epoch: [137][ 1040/ 1200] Overall Loss 0.184633 Objective Loss 0.184633 LR 0.000500 Time 0.019823 -2022-12-06 11:24:02,351 - Epoch: [137][ 1050/ 1200] Overall Loss 0.184784 Objective Loss 0.184784 LR 0.000500 Time 0.019816 -2022-12-06 11:24:02,543 - Epoch: [137][ 1060/ 1200] Overall Loss 0.185000 Objective Loss 0.185000 LR 0.000500 Time 0.019810 -2022-12-06 11:24:02,735 - Epoch: [137][ 1070/ 1200] Overall Loss 0.184994 Objective Loss 0.184994 LR 0.000500 Time 0.019803 -2022-12-06 11:24:02,927 - Epoch: [137][ 1080/ 1200] Overall Loss 0.184670 Objective Loss 0.184670 LR 0.000500 Time 0.019797 -2022-12-06 11:24:03,118 - Epoch: [137][ 1090/ 1200] Overall Loss 0.184479 Objective Loss 0.184479 LR 0.000500 Time 0.019790 -2022-12-06 11:24:03,309 - Epoch: [137][ 1100/ 1200] Overall Loss 0.184597 Objective Loss 0.184597 LR 0.000500 Time 0.019784 -2022-12-06 11:24:03,501 - Epoch: [137][ 1110/ 1200] Overall Loss 0.184692 Objective Loss 0.184692 LR 0.000500 Time 0.019778 -2022-12-06 11:24:03,692 - Epoch: [137][ 1120/ 1200] Overall Loss 0.184611 Objective Loss 0.184611 LR 0.000500 Time 0.019772 -2022-12-06 11:24:03,884 - Epoch: [137][ 1130/ 1200] Overall Loss 0.184514 Objective Loss 0.184514 LR 0.000500 Time 0.019766 -2022-12-06 11:24:04,076 - Epoch: [137][ 1140/ 1200] Overall Loss 0.184590 Objective Loss 0.184590 LR 0.000500 Time 0.019760 -2022-12-06 11:24:04,267 - Epoch: [137][ 1150/ 1200] Overall Loss 0.184695 Objective Loss 0.184695 LR 0.000500 Time 0.019755 -2022-12-06 11:24:04,459 - Epoch: [137][ 1160/ 1200] Overall Loss 0.184903 Objective Loss 0.184903 LR 0.000500 Time 0.019749 -2022-12-06 11:24:04,651 - Epoch: [137][ 1170/ 1200] Overall Loss 0.184777 Objective Loss 0.184777 LR 0.000500 Time 0.019744 -2022-12-06 11:24:04,842 - Epoch: [137][ 1180/ 1200] Overall Loss 0.184868 Objective Loss 0.184868 LR 0.000500 Time 0.019738 -2022-12-06 11:24:05,033 - Epoch: [137][ 1190/ 1200] Overall Loss 0.184845 Objective Loss 0.184845 LR 0.000500 Time 0.019732 -2022-12-06 11:24:05,266 - Epoch: [137][ 1200/ 1200] Overall Loss 0.184814 Objective Loss 0.184814 Top1 85.983264 Top5 99.581590 LR 0.000500 Time 0.019762 -2022-12-06 11:24:05,355 - --- validate (epoch=137)----------- -2022-12-06 11:24:05,355 - 34129 samples (256 per mini-batch) -2022-12-06 11:24:05,791 - Epoch: [137][ 10/ 134] Loss 0.242583 Top1 86.367188 Top5 98.320312 -2022-12-06 11:24:05,921 - Epoch: [137][ 20/ 134] Loss 0.235752 Top1 86.757812 Top5 98.457031 -2022-12-06 11:24:06,051 - Epoch: [137][ 30/ 134] Loss 0.240088 Top1 86.875000 Top5 98.437500 -2022-12-06 11:24:06,183 - Epoch: [137][ 40/ 134] Loss 0.244532 Top1 86.708984 Top5 98.369141 -2022-12-06 11:24:06,318 - Epoch: [137][ 50/ 134] Loss 0.248971 Top1 86.656250 Top5 98.359375 -2022-12-06 11:24:06,454 - Epoch: [137][ 60/ 134] Loss 0.249538 Top1 86.634115 Top5 98.352865 -2022-12-06 11:24:06,587 - Epoch: [137][ 70/ 134] Loss 0.250222 Top1 86.635045 Top5 98.337054 -2022-12-06 11:24:06,722 - Epoch: [137][ 80/ 134] Loss 0.251056 Top1 86.704102 Top5 98.334961 -2022-12-06 11:24:06,858 - Epoch: [137][ 90/ 134] Loss 0.251049 Top1 86.688368 Top5 98.376736 -2022-12-06 11:24:06,992 - Epoch: [137][ 100/ 134] Loss 0.252211 Top1 86.699219 Top5 98.414062 -2022-12-06 11:24:07,131 - Epoch: [137][ 110/ 134] Loss 0.253561 Top1 86.722301 Top5 98.401989 -2022-12-06 11:24:07,267 - Epoch: [137][ 120/ 134] Loss 0.253540 Top1 86.718750 Top5 98.401693 -2022-12-06 11:24:07,401 - Epoch: [137][ 130/ 134] Loss 0.253385 Top1 86.727764 Top5 98.383413 -2022-12-06 11:24:07,439 - Epoch: [137][ 134/ 134] Loss 0.250937 Top1 86.776642 Top5 98.408978 -2022-12-06 11:24:07,529 - ==> Top1: 86.777 Top5: 98.409 Loss: 0.251 - -2022-12-06 11:24:07,530 - ==> Confusion: -[[ 928 0 0 1 4 7 0 0 8 35 0 2 0 2 3 2 1 0 1 0 2] - [ 1 945 1 2 6 15 4 15 1 0 1 3 0 0 0 2 5 3 11 3 9] - [ 8 1 1014 12 4 2 12 6 0 4 7 4 1 1 0 4 0 4 3 4 12] - [ 3 1 14 947 4 1 0 2 0 0 11 1 3 1 8 0 1 2 15 0 6] - [ 12 4 1 0 957 5 0 1 1 7 1 1 1 2 7 4 9 1 0 2 4] - [ 2 14 1 4 9 976 3 18 2 1 0 11 3 12 2 0 1 0 0 7 3] - [ 2 2 12 7 1 3 1071 3 0 0 0 1 1 1 0 4 0 2 1 7 0] - [ 0 3 7 3 2 23 8 966 0 0 0 6 0 1 1 2 1 1 17 9 4] - [ 6 2 0 0 0 2 0 2 984 35 10 1 2 7 6 0 1 0 2 1 3] - [ 68 0 2 0 4 3 0 0 24 879 1 1 0 10 1 1 0 1 0 1 5] - [ 2 1 1 2 1 3 1 5 9 1 962 1 1 15 3 0 1 0 5 1 4] - [ 2 0 1 0 0 15 4 4 0 0 1 962 29 4 0 5 4 6 0 11 3] - [ 2 0 2 2 1 2 1 0 0 0 0 22 903 3 4 7 0 8 0 5 7] - [ 2 0 0 0 3 6 0 2 9 14 5 5 4 955 1 2 4 0 0 4 7] - [ 9 2 2 17 2 5 0 1 18 3 0 1 1 2 1053 0 0 1 5 1 7] - [ 0 1 1 1 5 0 1 0 0 0 1 11 3 1 0 995 7 8 0 5 3] - [ 2 3 1 1 3 0 0 1 1 1 0 1 1 2 0 10 1032 0 1 6 6] - [ 4 0 2 3 2 1 0 2 1 2 0 1 20 1 1 15 0 975 0 2 4] - [ 5 2 1 8 1 2 1 26 1 1 1 0 5 0 6 0 0 1 941 1 5] - [ 1 3 4 1 2 4 6 5 0 1 1 9 7 4 1 2 4 2 2 1016 5] - [ 161 179 180 135 114 150 71 161 80 89 176 94 325 297 114 89 202 70 144 242 10153]] - -2022-12-06 11:24:08,080 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:24:08,080 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:24:08,086 - - -2022-12-06 11:24:08,086 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:24:09,128 - Epoch: [138][ 10/ 1200] Overall Loss 0.172128 Objective Loss 0.172128 LR 0.000500 Time 0.104078 -2022-12-06 11:24:09,336 - Epoch: [138][ 20/ 1200] Overall Loss 0.184778 Objective Loss 0.184778 LR 0.000500 Time 0.062400 -2022-12-06 11:24:09,532 - Epoch: [138][ 30/ 1200] Overall Loss 0.180521 Objective Loss 0.180521 LR 0.000500 Time 0.048119 -2022-12-06 11:24:09,730 - Epoch: [138][ 40/ 1200] Overall Loss 0.180178 Objective Loss 0.180178 LR 0.000500 Time 0.041044 -2022-12-06 11:24:09,926 - Epoch: [138][ 50/ 1200] Overall Loss 0.178266 Objective Loss 0.178266 LR 0.000500 Time 0.036740 -2022-12-06 11:24:10,125 - Epoch: [138][ 60/ 1200] Overall Loss 0.182684 Objective Loss 0.182684 LR 0.000500 Time 0.033913 -2022-12-06 11:24:10,319 - Epoch: [138][ 70/ 1200] Overall Loss 0.179731 Objective Loss 0.179731 LR 0.000500 Time 0.031843 -2022-12-06 11:24:10,518 - Epoch: [138][ 80/ 1200] Overall Loss 0.178916 Objective Loss 0.178916 LR 0.000500 Time 0.030337 -2022-12-06 11:24:10,713 - Epoch: [138][ 90/ 1200] Overall Loss 0.176819 Objective Loss 0.176819 LR 0.000500 Time 0.029124 -2022-12-06 11:24:10,910 - Epoch: [138][ 100/ 1200] Overall Loss 0.177482 Objective Loss 0.177482 LR 0.000500 Time 0.028179 -2022-12-06 11:24:11,107 - Epoch: [138][ 110/ 1200] Overall Loss 0.179071 Objective Loss 0.179071 LR 0.000500 Time 0.027402 -2022-12-06 11:24:11,306 - Epoch: [138][ 120/ 1200] Overall Loss 0.179475 Objective Loss 0.179475 LR 0.000500 Time 0.026773 -2022-12-06 11:24:11,501 - Epoch: [138][ 130/ 1200] Overall Loss 0.180625 Objective Loss 0.180625 LR 0.000500 Time 0.026214 -2022-12-06 11:24:11,700 - Epoch: [138][ 140/ 1200] Overall Loss 0.181066 Objective Loss 0.181066 LR 0.000500 Time 0.025758 -2022-12-06 11:24:11,896 - Epoch: [138][ 150/ 1200] Overall Loss 0.181922 Objective Loss 0.181922 LR 0.000500 Time 0.025343 -2022-12-06 11:24:12,095 - Epoch: [138][ 160/ 1200] Overall Loss 0.182342 Objective Loss 0.182342 LR 0.000500 Time 0.024997 -2022-12-06 11:24:12,291 - Epoch: [138][ 170/ 1200] Overall Loss 0.182435 Objective Loss 0.182435 LR 0.000500 Time 0.024676 -2022-12-06 11:24:12,489 - Epoch: [138][ 180/ 1200] Overall Loss 0.183223 Objective Loss 0.183223 LR 0.000500 Time 0.024402 -2022-12-06 11:24:12,684 - Epoch: [138][ 190/ 1200] Overall Loss 0.182046 Objective Loss 0.182046 LR 0.000500 Time 0.024141 -2022-12-06 11:24:12,882 - Epoch: [138][ 200/ 1200] Overall Loss 0.182564 Objective Loss 0.182564 LR 0.000500 Time 0.023924 -2022-12-06 11:24:13,077 - Epoch: [138][ 210/ 1200] Overall Loss 0.182970 Objective Loss 0.182970 LR 0.000500 Time 0.023711 -2022-12-06 11:24:13,275 - Epoch: [138][ 220/ 1200] Overall Loss 0.184051 Objective Loss 0.184051 LR 0.000500 Time 0.023532 -2022-12-06 11:24:13,471 - Epoch: [138][ 230/ 1200] Overall Loss 0.184101 Objective Loss 0.184101 LR 0.000500 Time 0.023357 -2022-12-06 11:24:13,669 - Epoch: [138][ 240/ 1200] Overall Loss 0.184047 Objective Loss 0.184047 LR 0.000500 Time 0.023207 -2022-12-06 11:24:13,865 - Epoch: [138][ 250/ 1200] Overall Loss 0.183820 Objective Loss 0.183820 LR 0.000500 Time 0.023058 -2022-12-06 11:24:14,063 - Epoch: [138][ 260/ 1200] Overall Loss 0.182885 Objective Loss 0.182885 LR 0.000500 Time 0.022931 -2022-12-06 11:24:14,258 - Epoch: [138][ 270/ 1200] Overall Loss 0.182803 Objective Loss 0.182803 LR 0.000500 Time 0.022802 -2022-12-06 11:24:14,457 - Epoch: [138][ 280/ 1200] Overall Loss 0.182974 Objective Loss 0.182974 LR 0.000500 Time 0.022696 -2022-12-06 11:24:14,652 - Epoch: [138][ 290/ 1200] Overall Loss 0.183532 Objective Loss 0.183532 LR 0.000500 Time 0.022584 -2022-12-06 11:24:14,850 - Epoch: [138][ 300/ 1200] Overall Loss 0.183911 Objective Loss 0.183911 LR 0.000500 Time 0.022490 -2022-12-06 11:24:15,044 - Epoch: [138][ 310/ 1200] Overall Loss 0.183962 Objective Loss 0.183962 LR 0.000500 Time 0.022390 -2022-12-06 11:24:15,243 - Epoch: [138][ 320/ 1200] Overall Loss 0.183168 Objective Loss 0.183168 LR 0.000500 Time 0.022308 -2022-12-06 11:24:15,438 - Epoch: [138][ 330/ 1200] Overall Loss 0.183237 Objective Loss 0.183237 LR 0.000500 Time 0.022223 -2022-12-06 11:24:15,636 - Epoch: [138][ 340/ 1200] Overall Loss 0.183491 Objective Loss 0.183491 LR 0.000500 Time 0.022151 -2022-12-06 11:24:15,832 - Epoch: [138][ 350/ 1200] Overall Loss 0.183641 Objective Loss 0.183641 LR 0.000500 Time 0.022074 -2022-12-06 11:24:16,030 - Epoch: [138][ 360/ 1200] Overall Loss 0.183930 Objective Loss 0.183930 LR 0.000500 Time 0.022010 -2022-12-06 11:24:16,226 - Epoch: [138][ 370/ 1200] Overall Loss 0.183813 Objective Loss 0.183813 LR 0.000500 Time 0.021943 -2022-12-06 11:24:16,424 - Epoch: [138][ 380/ 1200] Overall Loss 0.184066 Objective Loss 0.184066 LR 0.000500 Time 0.021886 -2022-12-06 11:24:16,620 - Epoch: [138][ 390/ 1200] Overall Loss 0.183979 Objective Loss 0.183979 LR 0.000500 Time 0.021825 -2022-12-06 11:24:16,817 - Epoch: [138][ 400/ 1200] Overall Loss 0.184498 Objective Loss 0.184498 LR 0.000500 Time 0.021773 -2022-12-06 11:24:17,012 - Epoch: [138][ 410/ 1200] Overall Loss 0.184516 Objective Loss 0.184516 LR 0.000500 Time 0.021716 -2022-12-06 11:24:17,210 - Epoch: [138][ 420/ 1200] Overall Loss 0.184287 Objective Loss 0.184287 LR 0.000500 Time 0.021669 -2022-12-06 11:24:17,407 - Epoch: [138][ 430/ 1200] Overall Loss 0.184063 Objective Loss 0.184063 LR 0.000500 Time 0.021620 -2022-12-06 11:24:17,605 - Epoch: [138][ 440/ 1200] Overall Loss 0.184284 Objective Loss 0.184284 LR 0.000500 Time 0.021578 -2022-12-06 11:24:17,802 - Epoch: [138][ 450/ 1200] Overall Loss 0.184653 Objective Loss 0.184653 LR 0.000500 Time 0.021535 -2022-12-06 11:24:18,001 - Epoch: [138][ 460/ 1200] Overall Loss 0.185015 Objective Loss 0.185015 LR 0.000500 Time 0.021498 -2022-12-06 11:24:18,197 - Epoch: [138][ 470/ 1200] Overall Loss 0.184544 Objective Loss 0.184544 LR 0.000500 Time 0.021456 -2022-12-06 11:24:18,395 - Epoch: [138][ 480/ 1200] Overall Loss 0.184966 Objective Loss 0.184966 LR 0.000500 Time 0.021421 -2022-12-06 11:24:18,591 - Epoch: [138][ 490/ 1200] Overall Loss 0.185236 Objective Loss 0.185236 LR 0.000500 Time 0.021382 -2022-12-06 11:24:18,789 - Epoch: [138][ 500/ 1200] Overall Loss 0.184785 Objective Loss 0.184785 LR 0.000500 Time 0.021351 -2022-12-06 11:24:18,985 - Epoch: [138][ 510/ 1200] Overall Loss 0.184768 Objective Loss 0.184768 LR 0.000500 Time 0.021316 -2022-12-06 11:24:19,184 - Epoch: [138][ 520/ 1200] Overall Loss 0.184639 Objective Loss 0.184639 LR 0.000500 Time 0.021287 -2022-12-06 11:24:19,379 - Epoch: [138][ 530/ 1200] Overall Loss 0.184457 Objective Loss 0.184457 LR 0.000500 Time 0.021252 -2022-12-06 11:24:19,577 - Epoch: [138][ 540/ 1200] Overall Loss 0.184900 Objective Loss 0.184900 LR 0.000500 Time 0.021225 -2022-12-06 11:24:19,772 - Epoch: [138][ 550/ 1200] Overall Loss 0.184985 Objective Loss 0.184985 LR 0.000500 Time 0.021193 -2022-12-06 11:24:19,970 - Epoch: [138][ 560/ 1200] Overall Loss 0.184960 Objective Loss 0.184960 LR 0.000500 Time 0.021167 -2022-12-06 11:24:20,165 - Epoch: [138][ 570/ 1200] Overall Loss 0.185608 Objective Loss 0.185608 LR 0.000500 Time 0.021137 -2022-12-06 11:24:20,363 - Epoch: [138][ 580/ 1200] Overall Loss 0.185307 Objective Loss 0.185307 LR 0.000500 Time 0.021113 -2022-12-06 11:24:20,559 - Epoch: [138][ 590/ 1200] Overall Loss 0.185626 Objective Loss 0.185626 LR 0.000500 Time 0.021085 -2022-12-06 11:24:20,757 - Epoch: [138][ 600/ 1200] Overall Loss 0.185310 Objective Loss 0.185310 LR 0.000500 Time 0.021063 -2022-12-06 11:24:20,952 - Epoch: [138][ 610/ 1200] Overall Loss 0.185438 Objective Loss 0.185438 LR 0.000500 Time 0.021037 -2022-12-06 11:24:21,150 - Epoch: [138][ 620/ 1200] Overall Loss 0.185524 Objective Loss 0.185524 LR 0.000500 Time 0.021016 -2022-12-06 11:24:21,345 - Epoch: [138][ 630/ 1200] Overall Loss 0.185155 Objective Loss 0.185155 LR 0.000500 Time 0.020992 -2022-12-06 11:24:21,544 - Epoch: [138][ 640/ 1200] Overall Loss 0.184927 Objective Loss 0.184927 LR 0.000500 Time 0.020973 -2022-12-06 11:24:21,739 - Epoch: [138][ 650/ 1200] Overall Loss 0.185141 Objective Loss 0.185141 LR 0.000500 Time 0.020949 -2022-12-06 11:24:21,937 - Epoch: [138][ 660/ 1200] Overall Loss 0.185195 Objective Loss 0.185195 LR 0.000500 Time 0.020931 -2022-12-06 11:24:22,131 - Epoch: [138][ 670/ 1200] Overall Loss 0.185088 Objective Loss 0.185088 LR 0.000500 Time 0.020908 -2022-12-06 11:24:22,330 - Epoch: [138][ 680/ 1200] Overall Loss 0.185337 Objective Loss 0.185337 LR 0.000500 Time 0.020891 -2022-12-06 11:24:22,525 - Epoch: [138][ 690/ 1200] Overall Loss 0.185645 Objective Loss 0.185645 LR 0.000500 Time 0.020870 -2022-12-06 11:24:22,723 - Epoch: [138][ 700/ 1200] Overall Loss 0.185793 Objective Loss 0.185793 LR 0.000500 Time 0.020855 -2022-12-06 11:24:22,919 - Epoch: [138][ 710/ 1200] Overall Loss 0.185828 Objective Loss 0.185828 LR 0.000500 Time 0.020837 -2022-12-06 11:24:23,118 - Epoch: [138][ 720/ 1200] Overall Loss 0.186018 Objective Loss 0.186018 LR 0.000500 Time 0.020822 -2022-12-06 11:24:23,312 - Epoch: [138][ 730/ 1200] Overall Loss 0.186197 Objective Loss 0.186197 LR 0.000500 Time 0.020803 -2022-12-06 11:24:23,511 - Epoch: [138][ 740/ 1200] Overall Loss 0.186296 Objective Loss 0.186296 LR 0.000500 Time 0.020789 -2022-12-06 11:24:23,706 - Epoch: [138][ 750/ 1200] Overall Loss 0.186477 Objective Loss 0.186477 LR 0.000500 Time 0.020772 -2022-12-06 11:24:23,905 - Epoch: [138][ 760/ 1200] Overall Loss 0.186494 Objective Loss 0.186494 LR 0.000500 Time 0.020759 -2022-12-06 11:24:24,100 - Epoch: [138][ 770/ 1200] Overall Loss 0.186682 Objective Loss 0.186682 LR 0.000500 Time 0.020743 -2022-12-06 11:24:24,299 - Epoch: [138][ 780/ 1200] Overall Loss 0.186787 Objective Loss 0.186787 LR 0.000500 Time 0.020730 -2022-12-06 11:24:24,494 - Epoch: [138][ 790/ 1200] Overall Loss 0.186566 Objective Loss 0.186566 LR 0.000500 Time 0.020715 -2022-12-06 11:24:24,693 - Epoch: [138][ 800/ 1200] Overall Loss 0.186732 Objective Loss 0.186732 LR 0.000500 Time 0.020703 -2022-12-06 11:24:24,888 - Epoch: [138][ 810/ 1200] Overall Loss 0.186799 Objective Loss 0.186799 LR 0.000500 Time 0.020688 -2022-12-06 11:24:25,087 - Epoch: [138][ 820/ 1200] Overall Loss 0.187051 Objective Loss 0.187051 LR 0.000500 Time 0.020677 -2022-12-06 11:24:25,282 - Epoch: [138][ 830/ 1200] Overall Loss 0.187187 Objective Loss 0.187187 LR 0.000500 Time 0.020663 -2022-12-06 11:24:25,481 - Epoch: [138][ 840/ 1200] Overall Loss 0.187446 Objective Loss 0.187446 LR 0.000500 Time 0.020653 -2022-12-06 11:24:25,676 - Epoch: [138][ 850/ 1200] Overall Loss 0.187413 Objective Loss 0.187413 LR 0.000500 Time 0.020639 -2022-12-06 11:24:25,874 - Epoch: [138][ 860/ 1200] Overall Loss 0.187530 Objective Loss 0.187530 LR 0.000500 Time 0.020629 -2022-12-06 11:24:26,070 - Epoch: [138][ 870/ 1200] Overall Loss 0.187428 Objective Loss 0.187428 LR 0.000500 Time 0.020616 -2022-12-06 11:24:26,269 - Epoch: [138][ 880/ 1200] Overall Loss 0.187543 Objective Loss 0.187543 LR 0.000500 Time 0.020607 -2022-12-06 11:24:26,465 - Epoch: [138][ 890/ 1200] Overall Loss 0.187695 Objective Loss 0.187695 LR 0.000500 Time 0.020595 -2022-12-06 11:24:26,663 - Epoch: [138][ 900/ 1200] Overall Loss 0.187722 Objective Loss 0.187722 LR 0.000500 Time 0.020586 -2022-12-06 11:24:26,859 - Epoch: [138][ 910/ 1200] Overall Loss 0.187433 Objective Loss 0.187433 LR 0.000500 Time 0.020575 -2022-12-06 11:24:27,058 - Epoch: [138][ 920/ 1200] Overall Loss 0.187277 Objective Loss 0.187277 LR 0.000500 Time 0.020566 -2022-12-06 11:24:27,253 - Epoch: [138][ 930/ 1200] Overall Loss 0.187193 Objective Loss 0.187193 LR 0.000500 Time 0.020555 -2022-12-06 11:24:27,453 - Epoch: [138][ 940/ 1200] Overall Loss 0.187261 Objective Loss 0.187261 LR 0.000500 Time 0.020548 -2022-12-06 11:24:27,649 - Epoch: [138][ 950/ 1200] Overall Loss 0.187155 Objective Loss 0.187155 LR 0.000500 Time 0.020537 -2022-12-06 11:24:27,847 - Epoch: [138][ 960/ 1200] Overall Loss 0.187002 Objective Loss 0.187002 LR 0.000500 Time 0.020529 -2022-12-06 11:24:28,041 - Epoch: [138][ 970/ 1200] Overall Loss 0.186973 Objective Loss 0.186973 LR 0.000500 Time 0.020517 -2022-12-06 11:24:28,238 - Epoch: [138][ 980/ 1200] Overall Loss 0.187006 Objective Loss 0.187006 LR 0.000500 Time 0.020509 -2022-12-06 11:24:28,434 - Epoch: [138][ 990/ 1200] Overall Loss 0.187175 Objective Loss 0.187175 LR 0.000500 Time 0.020499 -2022-12-06 11:24:28,633 - Epoch: [138][ 1000/ 1200] Overall Loss 0.187083 Objective Loss 0.187083 LR 0.000500 Time 0.020492 -2022-12-06 11:24:28,829 - Epoch: [138][ 1010/ 1200] Overall Loss 0.186785 Objective Loss 0.186785 LR 0.000500 Time 0.020483 -2022-12-06 11:24:29,027 - Epoch: [138][ 1020/ 1200] Overall Loss 0.186605 Objective Loss 0.186605 LR 0.000500 Time 0.020475 -2022-12-06 11:24:29,223 - Epoch: [138][ 1030/ 1200] Overall Loss 0.186951 Objective Loss 0.186951 LR 0.000500 Time 0.020466 -2022-12-06 11:24:29,421 - Epoch: [138][ 1040/ 1200] Overall Loss 0.186919 Objective Loss 0.186919 LR 0.000500 Time 0.020459 -2022-12-06 11:24:29,616 - Epoch: [138][ 1050/ 1200] Overall Loss 0.186874 Objective Loss 0.186874 LR 0.000500 Time 0.020450 -2022-12-06 11:24:29,815 - Epoch: [138][ 1060/ 1200] Overall Loss 0.186921 Objective Loss 0.186921 LR 0.000500 Time 0.020444 -2022-12-06 11:24:30,010 - Epoch: [138][ 1070/ 1200] Overall Loss 0.186967 Objective Loss 0.186967 LR 0.000500 Time 0.020435 -2022-12-06 11:24:30,207 - Epoch: [138][ 1080/ 1200] Overall Loss 0.187116 Objective Loss 0.187116 LR 0.000500 Time 0.020428 -2022-12-06 11:24:30,403 - Epoch: [138][ 1090/ 1200] Overall Loss 0.187295 Objective Loss 0.187295 LR 0.000500 Time 0.020419 -2022-12-06 11:24:30,601 - Epoch: [138][ 1100/ 1200] Overall Loss 0.187228 Objective Loss 0.187228 LR 0.000500 Time 0.020413 -2022-12-06 11:24:30,796 - Epoch: [138][ 1110/ 1200] Overall Loss 0.187295 Objective Loss 0.187295 LR 0.000500 Time 0.020404 -2022-12-06 11:24:30,994 - Epoch: [138][ 1120/ 1200] Overall Loss 0.187418 Objective Loss 0.187418 LR 0.000500 Time 0.020399 -2022-12-06 11:24:31,190 - Epoch: [138][ 1130/ 1200] Overall Loss 0.187347 Objective Loss 0.187347 LR 0.000500 Time 0.020391 -2022-12-06 11:24:31,388 - Epoch: [138][ 1140/ 1200] Overall Loss 0.187342 Objective Loss 0.187342 LR 0.000500 Time 0.020386 -2022-12-06 11:24:31,583 - Epoch: [138][ 1150/ 1200] Overall Loss 0.187319 Objective Loss 0.187319 LR 0.000500 Time 0.020378 -2022-12-06 11:24:31,782 - Epoch: [138][ 1160/ 1200] Overall Loss 0.187242 Objective Loss 0.187242 LR 0.000500 Time 0.020372 -2022-12-06 11:24:31,977 - Epoch: [138][ 1170/ 1200] Overall Loss 0.187158 Objective Loss 0.187158 LR 0.000500 Time 0.020365 -2022-12-06 11:24:32,175 - Epoch: [138][ 1180/ 1200] Overall Loss 0.187054 Objective Loss 0.187054 LR 0.000500 Time 0.020359 -2022-12-06 11:24:32,370 - Epoch: [138][ 1190/ 1200] Overall Loss 0.187214 Objective Loss 0.187214 LR 0.000500 Time 0.020352 -2022-12-06 11:24:32,600 - Epoch: [138][ 1200/ 1200] Overall Loss 0.187058 Objective Loss 0.187058 Top1 88.493724 Top5 99.581590 LR 0.000500 Time 0.020373 -2022-12-06 11:24:32,688 - --- validate (epoch=138)----------- -2022-12-06 11:24:32,688 - 34129 samples (256 per mini-batch) -2022-12-06 11:24:33,138 - Epoch: [138][ 10/ 134] Loss 0.221850 Top1 87.578125 Top5 98.554688 -2022-12-06 11:24:33,281 - Epoch: [138][ 20/ 134] Loss 0.243110 Top1 86.621094 Top5 98.281250 -2022-12-06 11:24:33,424 - Epoch: [138][ 30/ 134] Loss 0.244551 Top1 86.770833 Top5 98.281250 -2022-12-06 11:24:33,568 - Epoch: [138][ 40/ 134] Loss 0.247630 Top1 86.572266 Top5 98.271484 -2022-12-06 11:24:33,707 - Epoch: [138][ 50/ 134] Loss 0.253146 Top1 86.250000 Top5 98.273438 -2022-12-06 11:24:33,838 - Epoch: [138][ 60/ 134] Loss 0.258319 Top1 85.970052 Top5 98.203125 -2022-12-06 11:24:33,971 - Epoch: [138][ 70/ 134] Loss 0.257209 Top1 85.915179 Top5 98.180804 -2022-12-06 11:24:34,115 - Epoch: [138][ 80/ 134] Loss 0.255469 Top1 85.976562 Top5 98.203125 -2022-12-06 11:24:34,254 - Epoch: [138][ 90/ 134] Loss 0.253258 Top1 86.072049 Top5 98.242188 -2022-12-06 11:24:34,398 - Epoch: [138][ 100/ 134] Loss 0.252617 Top1 86.066406 Top5 98.214844 -2022-12-06 11:24:34,537 - Epoch: [138][ 110/ 134] Loss 0.250633 Top1 86.235795 Top5 98.267045 -2022-12-06 11:24:34,670 - Epoch: [138][ 120/ 134] Loss 0.249356 Top1 86.292318 Top5 98.284505 -2022-12-06 11:24:34,804 - Epoch: [138][ 130/ 134] Loss 0.249205 Top1 86.295072 Top5 98.281250 -2022-12-06 11:24:34,842 - Epoch: [138][ 134/ 134] Loss 0.248850 Top1 86.313692 Top5 98.280055 -2022-12-06 11:24:34,932 - ==> Top1: 86.314 Top5: 98.280 Loss: 0.249 - -2022-12-06 11:24:34,933 - ==> Confusion: -[[ 915 1 3 4 1 6 0 0 0 53 0 1 1 2 4 1 1 0 0 1 2] - [ 3 936 2 2 12 18 3 13 1 0 0 4 1 1 1 3 5 0 11 5 6] - [ 4 3 1028 9 4 1 9 5 0 3 4 5 1 3 3 6 0 2 2 2 9] - [ 2 0 25 942 1 2 1 1 0 1 9 0 7 0 10 1 1 3 7 1 6] - [ 13 4 1 0 958 1 0 0 0 8 2 2 1 4 7 3 7 2 1 2 4] - [ 4 7 0 4 4 977 2 13 3 2 1 10 1 22 0 1 2 0 2 10 4] - [ 0 3 11 3 2 2 1068 3 0 0 0 1 1 2 0 8 0 3 1 8 2] - [ 0 4 13 2 1 35 10 941 0 0 2 4 0 3 1 0 0 1 22 12 3] - [ 8 1 0 2 0 0 0 0 967 49 8 1 1 13 7 0 2 0 3 2 0] - [ 51 0 3 0 2 1 0 2 18 898 1 2 0 11 2 2 0 2 0 1 5] - [ 1 1 6 3 1 1 2 3 8 2 954 2 1 17 3 0 0 0 4 2 8] - [ 3 1 4 0 0 8 4 2 2 1 0 975 17 8 0 9 2 3 0 11 1] - [ 0 1 2 2 0 4 0 1 0 0 0 28 894 2 2 10 1 13 0 3 6] - [ 0 1 0 0 0 1 0 2 9 11 0 3 4 976 1 2 3 1 1 2 6] - [ 11 3 1 6 3 3 0 0 13 7 0 1 2 4 1062 0 1 1 5 2 5] - [ 0 0 2 0 2 0 0 0 0 1 0 10 2 3 0 1001 3 13 0 5 1] - [ 2 2 0 2 5 2 1 0 1 0 0 3 0 2 0 21 1019 0 0 5 7] - [ 2 1 2 2 1 1 0 1 1 1 0 5 12 1 0 17 2 982 0 1 4] - [ 1 3 4 11 0 3 0 22 2 1 4 1 4 0 9 1 1 3 931 3 4] - [ 2 0 2 2 0 3 5 3 0 1 1 17 5 5 0 2 5 5 1 1019 2] - [ 141 166 224 108 127 169 73 130 70 96 146 119 299 341 165 183 173 81 160 240 10015]] - -2022-12-06 11:24:35,500 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:24:35,500 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:24:35,506 - - -2022-12-06 11:24:35,506 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:24:36,436 - Epoch: [139][ 10/ 1200] Overall Loss 0.165718 Objective Loss 0.165718 LR 0.000500 Time 0.092929 -2022-12-06 11:24:36,642 - Epoch: [139][ 20/ 1200] Overall Loss 0.172325 Objective Loss 0.172325 LR 0.000500 Time 0.056715 -2022-12-06 11:24:36,843 - Epoch: [139][ 30/ 1200] Overall Loss 0.170901 Objective Loss 0.170901 LR 0.000500 Time 0.044499 -2022-12-06 11:24:37,043 - Epoch: [139][ 40/ 1200] Overall Loss 0.178875 Objective Loss 0.178875 LR 0.000500 Time 0.038348 -2022-12-06 11:24:37,244 - Epoch: [139][ 50/ 1200] Overall Loss 0.179036 Objective Loss 0.179036 LR 0.000500 Time 0.034703 -2022-12-06 11:24:37,449 - Epoch: [139][ 60/ 1200] Overall Loss 0.180907 Objective Loss 0.180907 LR 0.000500 Time 0.032311 -2022-12-06 11:24:37,665 - Epoch: [139][ 70/ 1200] Overall Loss 0.183082 Objective Loss 0.183082 LR 0.000500 Time 0.030778 -2022-12-06 11:24:37,874 - Epoch: [139][ 80/ 1200] Overall Loss 0.180816 Objective Loss 0.180816 LR 0.000500 Time 0.029539 -2022-12-06 11:24:38,090 - Epoch: [139][ 90/ 1200] Overall Loss 0.181557 Objective Loss 0.181557 LR 0.000500 Time 0.028648 -2022-12-06 11:24:38,299 - Epoch: [139][ 100/ 1200] Overall Loss 0.180410 Objective Loss 0.180410 LR 0.000500 Time 0.027867 -2022-12-06 11:24:38,515 - Epoch: [139][ 110/ 1200] Overall Loss 0.179775 Objective Loss 0.179775 LR 0.000500 Time 0.027295 -2022-12-06 11:24:38,724 - Epoch: [139][ 120/ 1200] Overall Loss 0.183039 Objective Loss 0.183039 LR 0.000500 Time 0.026755 -2022-12-06 11:24:38,938 - Epoch: [139][ 130/ 1200] Overall Loss 0.182385 Objective Loss 0.182385 LR 0.000500 Time 0.026338 -2022-12-06 11:24:39,142 - Epoch: [139][ 140/ 1200] Overall Loss 0.181976 Objective Loss 0.181976 LR 0.000500 Time 0.025909 -2022-12-06 11:24:39,350 - Epoch: [139][ 150/ 1200] Overall Loss 0.180362 Objective Loss 0.180362 LR 0.000500 Time 0.025569 -2022-12-06 11:24:39,554 - Epoch: [139][ 160/ 1200] Overall Loss 0.178860 Objective Loss 0.178860 LR 0.000500 Time 0.025238 -2022-12-06 11:24:39,762 - Epoch: [139][ 170/ 1200] Overall Loss 0.178538 Objective Loss 0.178538 LR 0.000500 Time 0.024977 -2022-12-06 11:24:39,966 - Epoch: [139][ 180/ 1200] Overall Loss 0.178192 Objective Loss 0.178192 LR 0.000500 Time 0.024720 -2022-12-06 11:24:40,174 - Epoch: [139][ 190/ 1200] Overall Loss 0.176746 Objective Loss 0.176746 LR 0.000500 Time 0.024510 -2022-12-06 11:24:40,377 - Epoch: [139][ 200/ 1200] Overall Loss 0.176707 Objective Loss 0.176707 LR 0.000500 Time 0.024296 -2022-12-06 11:24:40,585 - Epoch: [139][ 210/ 1200] Overall Loss 0.176694 Objective Loss 0.176694 LR 0.000500 Time 0.024128 -2022-12-06 11:24:40,789 - Epoch: [139][ 220/ 1200] Overall Loss 0.176268 Objective Loss 0.176268 LR 0.000500 Time 0.023953 -2022-12-06 11:24:40,997 - Epoch: [139][ 230/ 1200] Overall Loss 0.177254 Objective Loss 0.177254 LR 0.000500 Time 0.023814 -2022-12-06 11:24:41,201 - Epoch: [139][ 240/ 1200] Overall Loss 0.177428 Objective Loss 0.177428 LR 0.000500 Time 0.023671 -2022-12-06 11:24:41,410 - Epoch: [139][ 250/ 1200] Overall Loss 0.178404 Objective Loss 0.178404 LR 0.000500 Time 0.023556 -2022-12-06 11:24:41,613 - Epoch: [139][ 260/ 1200] Overall Loss 0.178106 Objective Loss 0.178106 LR 0.000500 Time 0.023429 -2022-12-06 11:24:41,821 - Epoch: [139][ 270/ 1200] Overall Loss 0.177570 Objective Loss 0.177570 LR 0.000500 Time 0.023331 -2022-12-06 11:24:42,026 - Epoch: [139][ 280/ 1200] Overall Loss 0.176861 Objective Loss 0.176861 LR 0.000500 Time 0.023227 -2022-12-06 11:24:42,235 - Epoch: [139][ 290/ 1200] Overall Loss 0.177257 Objective Loss 0.177257 LR 0.000500 Time 0.023144 -2022-12-06 11:24:42,439 - Epoch: [139][ 300/ 1200] Overall Loss 0.177892 Objective Loss 0.177892 LR 0.000500 Time 0.023050 -2022-12-06 11:24:42,633 - Epoch: [139][ 310/ 1200] Overall Loss 0.178125 Objective Loss 0.178125 LR 0.000500 Time 0.022932 -2022-12-06 11:24:42,828 - Epoch: [139][ 320/ 1200] Overall Loss 0.178404 Objective Loss 0.178404 LR 0.000500 Time 0.022822 -2022-12-06 11:24:43,023 - Epoch: [139][ 330/ 1200] Overall Loss 0.178480 Objective Loss 0.178480 LR 0.000500 Time 0.022720 -2022-12-06 11:24:43,218 - Epoch: [139][ 340/ 1200] Overall Loss 0.178754 Objective Loss 0.178754 LR 0.000500 Time 0.022624 -2022-12-06 11:24:43,413 - Epoch: [139][ 350/ 1200] Overall Loss 0.179560 Objective Loss 0.179560 LR 0.000500 Time 0.022533 -2022-12-06 11:24:43,608 - Epoch: [139][ 360/ 1200] Overall Loss 0.179525 Objective Loss 0.179525 LR 0.000500 Time 0.022448 -2022-12-06 11:24:43,803 - Epoch: [139][ 370/ 1200] Overall Loss 0.179678 Objective Loss 0.179678 LR 0.000500 Time 0.022367 -2022-12-06 11:24:43,998 - Epoch: [139][ 380/ 1200] Overall Loss 0.180187 Objective Loss 0.180187 LR 0.000500 Time 0.022291 -2022-12-06 11:24:44,193 - Epoch: [139][ 390/ 1200] Overall Loss 0.180345 Objective Loss 0.180345 LR 0.000500 Time 0.022216 -2022-12-06 11:24:44,388 - Epoch: [139][ 400/ 1200] Overall Loss 0.179892 Objective Loss 0.179892 LR 0.000500 Time 0.022148 -2022-12-06 11:24:44,583 - Epoch: [139][ 410/ 1200] Overall Loss 0.179965 Objective Loss 0.179965 LR 0.000500 Time 0.022082 -2022-12-06 11:24:44,779 - Epoch: [139][ 420/ 1200] Overall Loss 0.180377 Objective Loss 0.180377 LR 0.000500 Time 0.022020 -2022-12-06 11:24:44,974 - Epoch: [139][ 430/ 1200] Overall Loss 0.180843 Objective Loss 0.180843 LR 0.000500 Time 0.021960 -2022-12-06 11:24:45,169 - Epoch: [139][ 440/ 1200] Overall Loss 0.180915 Objective Loss 0.180915 LR 0.000500 Time 0.021904 -2022-12-06 11:24:45,364 - Epoch: [139][ 450/ 1200] Overall Loss 0.181588 Objective Loss 0.181588 LR 0.000500 Time 0.021848 -2022-12-06 11:24:45,559 - Epoch: [139][ 460/ 1200] Overall Loss 0.181285 Objective Loss 0.181285 LR 0.000500 Time 0.021796 -2022-12-06 11:24:45,752 - Epoch: [139][ 470/ 1200] Overall Loss 0.181373 Objective Loss 0.181373 LR 0.000500 Time 0.021741 -2022-12-06 11:24:45,944 - Epoch: [139][ 480/ 1200] Overall Loss 0.181460 Objective Loss 0.181460 LR 0.000500 Time 0.021688 -2022-12-06 11:24:46,137 - Epoch: [139][ 490/ 1200] Overall Loss 0.181963 Objective Loss 0.181963 LR 0.000500 Time 0.021638 -2022-12-06 11:24:46,329 - Epoch: [139][ 500/ 1200] Overall Loss 0.181878 Objective Loss 0.181878 LR 0.000500 Time 0.021589 -2022-12-06 11:24:46,529 - Epoch: [139][ 510/ 1200] Overall Loss 0.181780 Objective Loss 0.181780 LR 0.000500 Time 0.021557 -2022-12-06 11:24:46,728 - Epoch: [139][ 520/ 1200] Overall Loss 0.181533 Objective Loss 0.181533 LR 0.000500 Time 0.021523 -2022-12-06 11:24:46,928 - Epoch: [139][ 530/ 1200] Overall Loss 0.181697 Objective Loss 0.181697 LR 0.000500 Time 0.021494 -2022-12-06 11:24:47,127 - Epoch: [139][ 540/ 1200] Overall Loss 0.182137 Objective Loss 0.182137 LR 0.000500 Time 0.021463 -2022-12-06 11:24:47,327 - Epoch: [139][ 550/ 1200] Overall Loss 0.182095 Objective Loss 0.182095 LR 0.000500 Time 0.021436 -2022-12-06 11:24:47,526 - Epoch: [139][ 560/ 1200] Overall Loss 0.181968 Objective Loss 0.181968 LR 0.000500 Time 0.021407 -2022-12-06 11:24:47,726 - Epoch: [139][ 570/ 1200] Overall Loss 0.182346 Objective Loss 0.182346 LR 0.000500 Time 0.021382 -2022-12-06 11:24:47,924 - Epoch: [139][ 580/ 1200] Overall Loss 0.182277 Objective Loss 0.182277 LR 0.000500 Time 0.021353 -2022-12-06 11:24:48,124 - Epoch: [139][ 590/ 1200] Overall Loss 0.182525 Objective Loss 0.182525 LR 0.000500 Time 0.021329 -2022-12-06 11:24:48,321 - Epoch: [139][ 600/ 1200] Overall Loss 0.182786 Objective Loss 0.182786 LR 0.000500 Time 0.021302 -2022-12-06 11:24:48,521 - Epoch: [139][ 610/ 1200] Overall Loss 0.183002 Objective Loss 0.183002 LR 0.000500 Time 0.021279 -2022-12-06 11:24:48,720 - Epoch: [139][ 620/ 1200] Overall Loss 0.182950 Objective Loss 0.182950 LR 0.000500 Time 0.021256 -2022-12-06 11:24:48,921 - Epoch: [139][ 630/ 1200] Overall Loss 0.182797 Objective Loss 0.182797 LR 0.000500 Time 0.021236 -2022-12-06 11:24:49,119 - Epoch: [139][ 640/ 1200] Overall Loss 0.182699 Objective Loss 0.182699 LR 0.000500 Time 0.021213 -2022-12-06 11:24:49,311 - Epoch: [139][ 650/ 1200] Overall Loss 0.182769 Objective Loss 0.182769 LR 0.000500 Time 0.021181 -2022-12-06 11:24:49,503 - Epoch: [139][ 660/ 1200] Overall Loss 0.183320 Objective Loss 0.183320 LR 0.000500 Time 0.021150 -2022-12-06 11:24:49,695 - Epoch: [139][ 670/ 1200] Overall Loss 0.183559 Objective Loss 0.183559 LR 0.000500 Time 0.021120 -2022-12-06 11:24:49,887 - Epoch: [139][ 680/ 1200] Overall Loss 0.183720 Objective Loss 0.183720 LR 0.000500 Time 0.021092 -2022-12-06 11:24:50,079 - Epoch: [139][ 690/ 1200] Overall Loss 0.183898 Objective Loss 0.183898 LR 0.000500 Time 0.021063 -2022-12-06 11:24:50,272 - Epoch: [139][ 700/ 1200] Overall Loss 0.183922 Objective Loss 0.183922 LR 0.000500 Time 0.021037 -2022-12-06 11:24:50,464 - Epoch: [139][ 710/ 1200] Overall Loss 0.183831 Objective Loss 0.183831 LR 0.000500 Time 0.021010 -2022-12-06 11:24:50,656 - Epoch: [139][ 720/ 1200] Overall Loss 0.183895 Objective Loss 0.183895 LR 0.000500 Time 0.020984 -2022-12-06 11:24:50,848 - Epoch: [139][ 730/ 1200] Overall Loss 0.183596 Objective Loss 0.183596 LR 0.000500 Time 0.020959 -2022-12-06 11:24:51,040 - Epoch: [139][ 740/ 1200] Overall Loss 0.183837 Objective Loss 0.183837 LR 0.000500 Time 0.020935 -2022-12-06 11:24:51,232 - Epoch: [139][ 750/ 1200] Overall Loss 0.183848 Objective Loss 0.183848 LR 0.000500 Time 0.020912 -2022-12-06 11:24:51,425 - Epoch: [139][ 760/ 1200] Overall Loss 0.183725 Objective Loss 0.183725 LR 0.000500 Time 0.020889 -2022-12-06 11:24:51,618 - Epoch: [139][ 770/ 1200] Overall Loss 0.183569 Objective Loss 0.183569 LR 0.000500 Time 0.020867 -2022-12-06 11:24:51,810 - Epoch: [139][ 780/ 1200] Overall Loss 0.183556 Objective Loss 0.183556 LR 0.000500 Time 0.020845 -2022-12-06 11:24:52,002 - Epoch: [139][ 790/ 1200] Overall Loss 0.183517 Objective Loss 0.183517 LR 0.000500 Time 0.020824 -2022-12-06 11:24:52,194 - Epoch: [139][ 800/ 1200] Overall Loss 0.183390 Objective Loss 0.183390 LR 0.000500 Time 0.020803 -2022-12-06 11:24:52,386 - Epoch: [139][ 810/ 1200] Overall Loss 0.183446 Objective Loss 0.183446 LR 0.000500 Time 0.020783 -2022-12-06 11:24:52,579 - Epoch: [139][ 820/ 1200] Overall Loss 0.183301 Objective Loss 0.183301 LR 0.000500 Time 0.020763 -2022-12-06 11:24:52,771 - Epoch: [139][ 830/ 1200] Overall Loss 0.183576 Objective Loss 0.183576 LR 0.000500 Time 0.020744 -2022-12-06 11:24:52,962 - Epoch: [139][ 840/ 1200] Overall Loss 0.183685 Objective Loss 0.183685 LR 0.000500 Time 0.020724 -2022-12-06 11:24:53,155 - Epoch: [139][ 850/ 1200] Overall Loss 0.183570 Objective Loss 0.183570 LR 0.000500 Time 0.020706 -2022-12-06 11:24:53,347 - Epoch: [139][ 860/ 1200] Overall Loss 0.183845 Objective Loss 0.183845 LR 0.000500 Time 0.020689 -2022-12-06 11:24:53,539 - Epoch: [139][ 870/ 1200] Overall Loss 0.183806 Objective Loss 0.183806 LR 0.000500 Time 0.020671 -2022-12-06 11:24:53,731 - Epoch: [139][ 880/ 1200] Overall Loss 0.183756 Objective Loss 0.183756 LR 0.000500 Time 0.020653 -2022-12-06 11:24:53,924 - Epoch: [139][ 890/ 1200] Overall Loss 0.183832 Objective Loss 0.183832 LR 0.000500 Time 0.020637 -2022-12-06 11:24:54,116 - Epoch: [139][ 900/ 1200] Overall Loss 0.183840 Objective Loss 0.183840 LR 0.000500 Time 0.020621 -2022-12-06 11:24:54,308 - Epoch: [139][ 910/ 1200] Overall Loss 0.183598 Objective Loss 0.183598 LR 0.000500 Time 0.020605 -2022-12-06 11:24:54,501 - Epoch: [139][ 920/ 1200] Overall Loss 0.183618 Objective Loss 0.183618 LR 0.000500 Time 0.020589 -2022-12-06 11:24:54,693 - Epoch: [139][ 930/ 1200] Overall Loss 0.183780 Objective Loss 0.183780 LR 0.000500 Time 0.020574 -2022-12-06 11:24:54,885 - Epoch: [139][ 940/ 1200] Overall Loss 0.183835 Objective Loss 0.183835 LR 0.000500 Time 0.020559 -2022-12-06 11:24:55,077 - Epoch: [139][ 950/ 1200] Overall Loss 0.184264 Objective Loss 0.184264 LR 0.000500 Time 0.020545 -2022-12-06 11:24:55,270 - Epoch: [139][ 960/ 1200] Overall Loss 0.184559 Objective Loss 0.184559 LR 0.000500 Time 0.020530 -2022-12-06 11:24:55,462 - Epoch: [139][ 970/ 1200] Overall Loss 0.184614 Objective Loss 0.184614 LR 0.000500 Time 0.020516 -2022-12-06 11:24:55,655 - Epoch: [139][ 980/ 1200] Overall Loss 0.184631 Objective Loss 0.184631 LR 0.000500 Time 0.020503 -2022-12-06 11:24:55,848 - Epoch: [139][ 990/ 1200] Overall Loss 0.184527 Objective Loss 0.184527 LR 0.000500 Time 0.020490 -2022-12-06 11:24:56,040 - Epoch: [139][ 1000/ 1200] Overall Loss 0.184644 Objective Loss 0.184644 LR 0.000500 Time 0.020477 -2022-12-06 11:24:56,232 - Epoch: [139][ 1010/ 1200] Overall Loss 0.184490 Objective Loss 0.184490 LR 0.000500 Time 0.020464 -2022-12-06 11:24:56,424 - Epoch: [139][ 1020/ 1200] Overall Loss 0.184568 Objective Loss 0.184568 LR 0.000500 Time 0.020451 -2022-12-06 11:24:56,617 - Epoch: [139][ 1030/ 1200] Overall Loss 0.184651 Objective Loss 0.184651 LR 0.000500 Time 0.020439 -2022-12-06 11:24:56,809 - Epoch: [139][ 1040/ 1200] Overall Loss 0.184688 Objective Loss 0.184688 LR 0.000500 Time 0.020427 -2022-12-06 11:24:57,002 - Epoch: [139][ 1050/ 1200] Overall Loss 0.184788 Objective Loss 0.184788 LR 0.000500 Time 0.020415 -2022-12-06 11:24:57,194 - Epoch: [139][ 1060/ 1200] Overall Loss 0.184954 Objective Loss 0.184954 LR 0.000500 Time 0.020404 -2022-12-06 11:24:57,387 - Epoch: [139][ 1070/ 1200] Overall Loss 0.185055 Objective Loss 0.185055 LR 0.000500 Time 0.020393 -2022-12-06 11:24:57,579 - Epoch: [139][ 1080/ 1200] Overall Loss 0.185258 Objective Loss 0.185258 LR 0.000500 Time 0.020381 -2022-12-06 11:24:57,771 - Epoch: [139][ 1090/ 1200] Overall Loss 0.185321 Objective Loss 0.185321 LR 0.000500 Time 0.020370 -2022-12-06 11:24:57,964 - Epoch: [139][ 1100/ 1200] Overall Loss 0.185617 Objective Loss 0.185617 LR 0.000500 Time 0.020360 -2022-12-06 11:24:58,156 - Epoch: [139][ 1110/ 1200] Overall Loss 0.185615 Objective Loss 0.185615 LR 0.000500 Time 0.020349 -2022-12-06 11:24:58,348 - Epoch: [139][ 1120/ 1200] Overall Loss 0.185626 Objective Loss 0.185626 LR 0.000500 Time 0.020338 -2022-12-06 11:24:58,540 - Epoch: [139][ 1130/ 1200] Overall Loss 0.185452 Objective Loss 0.185452 LR 0.000500 Time 0.020328 -2022-12-06 11:24:58,732 - Epoch: [139][ 1140/ 1200] Overall Loss 0.185583 Objective Loss 0.185583 LR 0.000500 Time 0.020317 -2022-12-06 11:24:58,924 - Epoch: [139][ 1150/ 1200] Overall Loss 0.185717 Objective Loss 0.185717 LR 0.000500 Time 0.020307 -2022-12-06 11:24:59,117 - Epoch: [139][ 1160/ 1200] Overall Loss 0.185717 Objective Loss 0.185717 LR 0.000500 Time 0.020298 -2022-12-06 11:24:59,309 - Epoch: [139][ 1170/ 1200] Overall Loss 0.185843 Objective Loss 0.185843 LR 0.000500 Time 0.020288 -2022-12-06 11:24:59,501 - Epoch: [139][ 1180/ 1200] Overall Loss 0.185756 Objective Loss 0.185756 LR 0.000500 Time 0.020278 -2022-12-06 11:24:59,694 - Epoch: [139][ 1190/ 1200] Overall Loss 0.185931 Objective Loss 0.185931 LR 0.000500 Time 0.020269 -2022-12-06 11:24:59,924 - Epoch: [139][ 1200/ 1200] Overall Loss 0.186024 Objective Loss 0.186024 Top1 87.866109 Top5 97.907950 LR 0.000500 Time 0.020292 -2022-12-06 11:25:00,012 - --- validate (epoch=139)----------- -2022-12-06 11:25:00,013 - 34129 samples (256 per mini-batch) -2022-12-06 11:25:00,587 - Epoch: [139][ 10/ 134] Loss 0.252365 Top1 87.148438 Top5 98.164062 -2022-12-06 11:25:00,718 - Epoch: [139][ 20/ 134] Loss 0.243421 Top1 87.011719 Top5 98.281250 -2022-12-06 11:25:00,853 - Epoch: [139][ 30/ 134] Loss 0.247232 Top1 86.888021 Top5 98.216146 -2022-12-06 11:25:00,987 - Epoch: [139][ 40/ 134] Loss 0.254570 Top1 86.806641 Top5 98.291016 -2022-12-06 11:25:01,121 - Epoch: [139][ 50/ 134] Loss 0.252938 Top1 86.679688 Top5 98.289062 -2022-12-06 11:25:01,255 - Epoch: [139][ 60/ 134] Loss 0.250228 Top1 86.705729 Top5 98.307292 -2022-12-06 11:25:01,386 - Epoch: [139][ 70/ 134] Loss 0.249217 Top1 86.568080 Top5 98.297991 -2022-12-06 11:25:01,518 - Epoch: [139][ 80/ 134] Loss 0.253762 Top1 86.582031 Top5 98.256836 -2022-12-06 11:25:01,648 - Epoch: [139][ 90/ 134] Loss 0.251124 Top1 86.701389 Top5 98.268229 -2022-12-06 11:25:01,779 - Epoch: [139][ 100/ 134] Loss 0.252155 Top1 86.632812 Top5 98.289062 -2022-12-06 11:25:01,911 - Epoch: [139][ 110/ 134] Loss 0.251064 Top1 86.619318 Top5 98.281250 -2022-12-06 11:25:02,045 - Epoch: [139][ 120/ 134] Loss 0.252186 Top1 86.614583 Top5 98.297526 -2022-12-06 11:25:02,180 - Epoch: [139][ 130/ 134] Loss 0.252430 Top1 86.631611 Top5 98.293269 -2022-12-06 11:25:02,220 - Epoch: [139][ 134/ 134] Loss 0.251643 Top1 86.682880 Top5 98.300566 -2022-12-06 11:25:02,308 - ==> Top1: 86.683 Top5: 98.301 Loss: 0.252 - -2022-12-06 11:25:02,309 - ==> Confusion: -[[ 903 0 3 2 4 7 1 1 5 49 0 4 0 4 5 1 1 0 1 0 5] - [ 1 940 2 2 7 23 3 13 1 0 2 5 2 1 0 1 3 1 9 2 9] - [ 4 1 1017 13 4 1 16 8 0 2 4 3 1 3 4 4 2 3 3 3 7] - [ 2 0 19 949 2 2 1 0 0 1 8 0 5 0 10 1 0 2 15 0 3] - [ 11 4 2 0 952 2 1 1 0 6 2 2 1 3 13 6 6 3 0 1 4] - [ 1 12 0 3 4 978 1 21 2 3 1 9 4 15 2 1 1 1 3 5 2] - [ 2 3 10 3 0 1 1071 5 0 0 2 1 1 3 0 6 0 0 1 9 0] - [ 0 8 10 2 1 26 10 945 0 0 3 4 0 3 0 1 0 0 24 14 3] - [ 4 2 0 0 0 1 1 0 964 51 11 2 1 9 10 0 2 0 3 2 1] - [ 52 0 2 1 5 1 0 1 16 894 2 1 0 16 2 3 0 1 1 0 3] - [ 1 2 6 6 1 0 1 3 5 1 961 2 3 10 4 1 0 0 2 2 8] - [ 2 1 2 1 0 11 4 2 0 0 1 964 33 2 0 8 3 5 0 11 1] - [ 0 0 2 1 0 3 0 1 0 0 0 21 914 1 0 8 2 5 1 4 6] - [ 1 1 0 0 0 6 1 2 13 8 2 4 2 971 0 1 2 0 0 3 6] - [ 6 1 3 15 4 3 0 0 16 3 0 1 1 2 1059 0 1 0 10 1 4] - [ 0 0 1 1 1 1 6 0 1 0 0 6 7 2 0 997 4 11 0 4 1] - [ 2 0 0 1 3 1 1 0 1 0 0 4 1 2 1 15 1024 0 1 7 8] - [ 0 2 2 4 1 1 2 1 1 4 0 11 27 2 1 15 1 957 0 0 4] - [ 3 2 4 10 1 2 0 25 2 0 3 3 3 2 6 0 0 3 933 3 3] - [ 2 3 5 1 0 4 2 5 0 0 3 13 7 6 0 2 1 3 1 1018 4] - [ 98 178 196 126 89 172 95 154 75 81 140 92 359 287 148 129 172 59 155 250 10171]] - -2022-12-06 11:25:02,886 - ==> Best [Top1: 87.351 Top5: 98.350 Sparsity:0.00 Params: 5376 on epoch: 129] -2022-12-06 11:25:02,886 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:25:02,892 - - -2022-12-06 11:25:02,892 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:25:03,834 - Epoch: [140][ 10/ 1200] Overall Loss 0.166454 Objective Loss 0.166454 LR 0.000250 Time 0.094126 -2022-12-06 11:25:04,029 - Epoch: [140][ 20/ 1200] Overall Loss 0.167739 Objective Loss 0.167739 LR 0.000250 Time 0.056811 -2022-12-06 11:25:04,223 - Epoch: [140][ 30/ 1200] Overall Loss 0.168274 Objective Loss 0.168274 LR 0.000250 Time 0.044303 -2022-12-06 11:25:04,416 - Epoch: [140][ 40/ 1200] Overall Loss 0.168598 Objective Loss 0.168598 LR 0.000250 Time 0.038040 -2022-12-06 11:25:04,610 - Epoch: [140][ 50/ 1200] Overall Loss 0.171833 Objective Loss 0.171833 LR 0.000250 Time 0.034293 -2022-12-06 11:25:04,803 - Epoch: [140][ 60/ 1200] Overall Loss 0.170020 Objective Loss 0.170020 LR 0.000250 Time 0.031795 -2022-12-06 11:25:04,996 - Epoch: [140][ 70/ 1200] Overall Loss 0.166298 Objective Loss 0.166298 LR 0.000250 Time 0.030004 -2022-12-06 11:25:05,189 - Epoch: [140][ 80/ 1200] Overall Loss 0.163823 Objective Loss 0.163823 LR 0.000250 Time 0.028660 -2022-12-06 11:25:05,382 - Epoch: [140][ 90/ 1200] Overall Loss 0.165390 Objective Loss 0.165390 LR 0.000250 Time 0.027608 -2022-12-06 11:25:05,577 - Epoch: [140][ 100/ 1200] Overall Loss 0.166695 Objective Loss 0.166695 LR 0.000250 Time 0.026788 -2022-12-06 11:25:05,770 - Epoch: [140][ 110/ 1200] Overall Loss 0.166124 Objective Loss 0.166124 LR 0.000250 Time 0.026109 -2022-12-06 11:25:05,964 - Epoch: [140][ 120/ 1200] Overall Loss 0.164746 Objective Loss 0.164746 LR 0.000250 Time 0.025543 -2022-12-06 11:25:06,157 - Epoch: [140][ 130/ 1200] Overall Loss 0.165301 Objective Loss 0.165301 LR 0.000250 Time 0.025060 -2022-12-06 11:25:06,351 - Epoch: [140][ 140/ 1200] Overall Loss 0.166105 Objective Loss 0.166105 LR 0.000250 Time 0.024652 -2022-12-06 11:25:06,544 - Epoch: [140][ 150/ 1200] Overall Loss 0.167013 Objective Loss 0.167013 LR 0.000250 Time 0.024289 -2022-12-06 11:25:06,738 - Epoch: [140][ 160/ 1200] Overall Loss 0.167501 Objective Loss 0.167501 LR 0.000250 Time 0.023978 -2022-12-06 11:25:06,931 - Epoch: [140][ 170/ 1200] Overall Loss 0.167896 Objective Loss 0.167896 LR 0.000250 Time 0.023701 -2022-12-06 11:25:07,124 - Epoch: [140][ 180/ 1200] Overall Loss 0.167693 Objective Loss 0.167693 LR 0.000250 Time 0.023456 -2022-12-06 11:25:07,318 - Epoch: [140][ 190/ 1200] Overall Loss 0.168813 Objective Loss 0.168813 LR 0.000250 Time 0.023237 -2022-12-06 11:25:07,512 - Epoch: [140][ 200/ 1200] Overall Loss 0.169787 Objective Loss 0.169787 LR 0.000250 Time 0.023041 -2022-12-06 11:25:07,703 - Epoch: [140][ 210/ 1200] Overall Loss 0.169020 Objective Loss 0.169020 LR 0.000250 Time 0.022853 -2022-12-06 11:25:07,895 - Epoch: [140][ 220/ 1200] Overall Loss 0.168896 Objective Loss 0.168896 LR 0.000250 Time 0.022682 -2022-12-06 11:25:08,087 - Epoch: [140][ 230/ 1200] Overall Loss 0.168926 Objective Loss 0.168926 LR 0.000250 Time 0.022528 -2022-12-06 11:25:08,279 - Epoch: [140][ 240/ 1200] Overall Loss 0.168109 Objective Loss 0.168109 LR 0.000250 Time 0.022388 -2022-12-06 11:25:08,471 - Epoch: [140][ 250/ 1200] Overall Loss 0.168485 Objective Loss 0.168485 LR 0.000250 Time 0.022257 -2022-12-06 11:25:08,661 - Epoch: [140][ 260/ 1200] Overall Loss 0.167924 Objective Loss 0.167924 LR 0.000250 Time 0.022133 -2022-12-06 11:25:08,853 - Epoch: [140][ 270/ 1200] Overall Loss 0.167924 Objective Loss 0.167924 LR 0.000250 Time 0.022020 -2022-12-06 11:25:09,044 - Epoch: [140][ 280/ 1200] Overall Loss 0.167694 Objective Loss 0.167694 LR 0.000250 Time 0.021915 -2022-12-06 11:25:09,239 - Epoch: [140][ 290/ 1200] Overall Loss 0.168515 Objective Loss 0.168515 LR 0.000250 Time 0.021827 -2022-12-06 11:25:09,432 - Epoch: [140][ 300/ 1200] Overall Loss 0.168388 Objective Loss 0.168388 LR 0.000250 Time 0.021742 -2022-12-06 11:25:09,626 - Epoch: [140][ 310/ 1200] Overall Loss 0.168506 Objective Loss 0.168506 LR 0.000250 Time 0.021666 -2022-12-06 11:25:09,820 - Epoch: [140][ 320/ 1200] Overall Loss 0.168665 Objective Loss 0.168665 LR 0.000250 Time 0.021593 -2022-12-06 11:25:10,014 - Epoch: [140][ 330/ 1200] Overall Loss 0.168311 Objective Loss 0.168311 LR 0.000250 Time 0.021525 -2022-12-06 11:25:10,208 - Epoch: [140][ 340/ 1200] Overall Loss 0.168089 Objective Loss 0.168089 LR 0.000250 Time 0.021459 -2022-12-06 11:25:10,402 - Epoch: [140][ 350/ 1200] Overall Loss 0.167874 Objective Loss 0.167874 LR 0.000250 Time 0.021399 -2022-12-06 11:25:10,595 - Epoch: [140][ 360/ 1200] Overall Loss 0.167943 Objective Loss 0.167943 LR 0.000250 Time 0.021339 -2022-12-06 11:25:10,790 - Epoch: [140][ 370/ 1200] Overall Loss 0.167731 Objective Loss 0.167731 LR 0.000250 Time 0.021290 -2022-12-06 11:25:10,984 - Epoch: [140][ 380/ 1200] Overall Loss 0.168108 Objective Loss 0.168108 LR 0.000250 Time 0.021237 -2022-12-06 11:25:11,178 - Epoch: [140][ 390/ 1200] Overall Loss 0.168180 Objective Loss 0.168180 LR 0.000250 Time 0.021188 -2022-12-06 11:25:11,371 - Epoch: [140][ 400/ 1200] Overall Loss 0.168446 Objective Loss 0.168446 LR 0.000250 Time 0.021140 -2022-12-06 11:25:11,566 - Epoch: [140][ 410/ 1200] Overall Loss 0.168261 Objective Loss 0.168261 LR 0.000250 Time 0.021098 -2022-12-06 11:25:11,759 - Epoch: [140][ 420/ 1200] Overall Loss 0.168104 Objective Loss 0.168104 LR 0.000250 Time 0.021055 -2022-12-06 11:25:11,954 - Epoch: [140][ 430/ 1200] Overall Loss 0.168601 Objective Loss 0.168601 LR 0.000250 Time 0.021017 -2022-12-06 11:25:12,147 - Epoch: [140][ 440/ 1200] Overall Loss 0.168678 Objective Loss 0.168678 LR 0.000250 Time 0.020978 -2022-12-06 11:25:12,342 - Epoch: [140][ 450/ 1200] Overall Loss 0.168957 Objective Loss 0.168957 LR 0.000250 Time 0.020942 -2022-12-06 11:25:12,536 - Epoch: [140][ 460/ 1200] Overall Loss 0.168873 Objective Loss 0.168873 LR 0.000250 Time 0.020907 -2022-12-06 11:25:12,730 - Epoch: [140][ 470/ 1200] Overall Loss 0.168600 Objective Loss 0.168600 LR 0.000250 Time 0.020876 -2022-12-06 11:25:12,924 - Epoch: [140][ 480/ 1200] Overall Loss 0.168833 Objective Loss 0.168833 LR 0.000250 Time 0.020843 -2022-12-06 11:25:13,118 - Epoch: [140][ 490/ 1200] Overall Loss 0.168850 Objective Loss 0.168850 LR 0.000250 Time 0.020813 -2022-12-06 11:25:13,311 - Epoch: [140][ 500/ 1200] Overall Loss 0.168528 Objective Loss 0.168528 LR 0.000250 Time 0.020781 -2022-12-06 11:25:13,506 - Epoch: [140][ 510/ 1200] Overall Loss 0.168912 Objective Loss 0.168912 LR 0.000250 Time 0.020754 -2022-12-06 11:25:13,700 - Epoch: [140][ 520/ 1200] Overall Loss 0.169271 Objective Loss 0.169271 LR 0.000250 Time 0.020727 -2022-12-06 11:25:13,894 - Epoch: [140][ 530/ 1200] Overall Loss 0.169399 Objective Loss 0.169399 LR 0.000250 Time 0.020702 -2022-12-06 11:25:14,088 - Epoch: [140][ 540/ 1200] Overall Loss 0.169422 Objective Loss 0.169422 LR 0.000250 Time 0.020677 -2022-12-06 11:25:14,283 - Epoch: [140][ 550/ 1200] Overall Loss 0.169203 Objective Loss 0.169203 LR 0.000250 Time 0.020654 -2022-12-06 11:25:14,476 - Epoch: [140][ 560/ 1200] Overall Loss 0.169149 Objective Loss 0.169149 LR 0.000250 Time 0.020630 -2022-12-06 11:25:14,671 - Epoch: [140][ 570/ 1200] Overall Loss 0.169223 Objective Loss 0.169223 LR 0.000250 Time 0.020608 -2022-12-06 11:25:14,864 - Epoch: [140][ 580/ 1200] Overall Loss 0.169553 Objective Loss 0.169553 LR 0.000250 Time 0.020585 -2022-12-06 11:25:15,058 - Epoch: [140][ 590/ 1200] Overall Loss 0.169561 Objective Loss 0.169561 LR 0.000250 Time 0.020564 -2022-12-06 11:25:15,251 - Epoch: [140][ 600/ 1200] Overall Loss 0.169516 Objective Loss 0.169516 LR 0.000250 Time 0.020542 -2022-12-06 11:25:15,446 - Epoch: [140][ 610/ 1200] Overall Loss 0.170027 Objective Loss 0.170027 LR 0.000250 Time 0.020523 -2022-12-06 11:25:15,639 - Epoch: [140][ 620/ 1200] Overall Loss 0.170196 Objective Loss 0.170196 LR 0.000250 Time 0.020503 -2022-12-06 11:25:15,834 - Epoch: [140][ 630/ 1200] Overall Loss 0.170043 Objective Loss 0.170043 LR 0.000250 Time 0.020486 -2022-12-06 11:25:16,027 - Epoch: [140][ 640/ 1200] Overall Loss 0.170114 Objective Loss 0.170114 LR 0.000250 Time 0.020467 -2022-12-06 11:25:16,222 - Epoch: [140][ 650/ 1200] Overall Loss 0.170007 Objective Loss 0.170007 LR 0.000250 Time 0.020451 -2022-12-06 11:25:16,416 - Epoch: [140][ 660/ 1200] Overall Loss 0.169648 Objective Loss 0.169648 LR 0.000250 Time 0.020434 -2022-12-06 11:25:16,610 - Epoch: [140][ 670/ 1200] Overall Loss 0.169357 Objective Loss 0.169357 LR 0.000250 Time 0.020418 -2022-12-06 11:25:16,803 - Epoch: [140][ 680/ 1200] Overall Loss 0.169668 Objective Loss 0.169668 LR 0.000250 Time 0.020401 -2022-12-06 11:25:16,997 - Epoch: [140][ 690/ 1200] Overall Loss 0.169579 Objective Loss 0.169579 LR 0.000250 Time 0.020386 -2022-12-06 11:25:17,191 - Epoch: [140][ 700/ 1200] Overall Loss 0.169503 Objective Loss 0.169503 LR 0.000250 Time 0.020371 -2022-12-06 11:25:17,385 - Epoch: [140][ 710/ 1200] Overall Loss 0.169427 Objective Loss 0.169427 LR 0.000250 Time 0.020356 -2022-12-06 11:25:17,578 - Epoch: [140][ 720/ 1200] Overall Loss 0.169512 Objective Loss 0.169512 LR 0.000250 Time 0.020341 -2022-12-06 11:25:17,773 - Epoch: [140][ 730/ 1200] Overall Loss 0.169535 Objective Loss 0.169535 LR 0.000250 Time 0.020328 -2022-12-06 11:25:17,967 - Epoch: [140][ 740/ 1200] Overall Loss 0.169633 Objective Loss 0.169633 LR 0.000250 Time 0.020315 -2022-12-06 11:25:18,161 - Epoch: [140][ 750/ 1200] Overall Loss 0.169612 Objective Loss 0.169612 LR 0.000250 Time 0.020302 -2022-12-06 11:25:18,354 - Epoch: [140][ 760/ 1200] Overall Loss 0.169221 Objective Loss 0.169221 LR 0.000250 Time 0.020289 -2022-12-06 11:25:18,548 - Epoch: [140][ 770/ 1200] Overall Loss 0.169071 Objective Loss 0.169071 LR 0.000250 Time 0.020276 -2022-12-06 11:25:18,741 - Epoch: [140][ 780/ 1200] Overall Loss 0.169193 Objective Loss 0.169193 LR 0.000250 Time 0.020263 -2022-12-06 11:25:18,936 - Epoch: [140][ 790/ 1200] Overall Loss 0.169168 Objective Loss 0.169168 LR 0.000250 Time 0.020252 -2022-12-06 11:25:19,129 - Epoch: [140][ 800/ 1200] Overall Loss 0.169216 Objective Loss 0.169216 LR 0.000250 Time 0.020239 -2022-12-06 11:25:19,323 - Epoch: [140][ 810/ 1200] Overall Loss 0.169261 Objective Loss 0.169261 LR 0.000250 Time 0.020229 -2022-12-06 11:25:19,517 - Epoch: [140][ 820/ 1200] Overall Loss 0.169473 Objective Loss 0.169473 LR 0.000250 Time 0.020218 -2022-12-06 11:25:19,712 - Epoch: [140][ 830/ 1200] Overall Loss 0.169639 Objective Loss 0.169639 LR 0.000250 Time 0.020208 -2022-12-06 11:25:19,904 - Epoch: [140][ 840/ 1200] Overall Loss 0.169636 Objective Loss 0.169636 LR 0.000250 Time 0.020196 -2022-12-06 11:25:20,095 - Epoch: [140][ 850/ 1200] Overall Loss 0.169544 Objective Loss 0.169544 LR 0.000250 Time 0.020182 -2022-12-06 11:25:20,286 - Epoch: [140][ 860/ 1200] Overall Loss 0.169437 Objective Loss 0.169437 LR 0.000250 Time 0.020169 -2022-12-06 11:25:20,477 - Epoch: [140][ 870/ 1200] Overall Loss 0.169395 Objective Loss 0.169395 LR 0.000250 Time 0.020156 -2022-12-06 11:25:20,668 - Epoch: [140][ 880/ 1200] Overall Loss 0.169541 Objective Loss 0.169541 LR 0.000250 Time 0.020143 -2022-12-06 11:25:20,858 - Epoch: [140][ 890/ 1200] Overall Loss 0.169254 Objective Loss 0.169254 LR 0.000250 Time 0.020130 -2022-12-06 11:25:21,050 - Epoch: [140][ 900/ 1200] Overall Loss 0.168996 Objective Loss 0.168996 LR 0.000250 Time 0.020119 -2022-12-06 11:25:21,240 - Epoch: [140][ 910/ 1200] Overall Loss 0.168924 Objective Loss 0.168924 LR 0.000250 Time 0.020107 -2022-12-06 11:25:21,432 - Epoch: [140][ 920/ 1200] Overall Loss 0.168717 Objective Loss 0.168717 LR 0.000250 Time 0.020095 -2022-12-06 11:25:21,622 - Epoch: [140][ 930/ 1200] Overall Loss 0.168703 Objective Loss 0.168703 LR 0.000250 Time 0.020084 -2022-12-06 11:25:21,814 - Epoch: [140][ 940/ 1200] Overall Loss 0.168616 Objective Loss 0.168616 LR 0.000250 Time 0.020073 -2022-12-06 11:25:22,004 - Epoch: [140][ 950/ 1200] Overall Loss 0.168765 Objective Loss 0.168765 LR 0.000250 Time 0.020062 -2022-12-06 11:25:22,195 - Epoch: [140][ 960/ 1200] Overall Loss 0.168661 Objective Loss 0.168661 LR 0.000250 Time 0.020051 -2022-12-06 11:25:22,386 - Epoch: [140][ 970/ 1200] Overall Loss 0.168664 Objective Loss 0.168664 LR 0.000250 Time 0.020041 -2022-12-06 11:25:22,577 - Epoch: [140][ 980/ 1200] Overall Loss 0.168657 Objective Loss 0.168657 LR 0.000250 Time 0.020030 -2022-12-06 11:25:22,768 - Epoch: [140][ 990/ 1200] Overall Loss 0.168619 Objective Loss 0.168619 LR 0.000250 Time 0.020020 -2022-12-06 11:25:22,959 - Epoch: [140][ 1000/ 1200] Overall Loss 0.168502 Objective Loss 0.168502 LR 0.000250 Time 0.020011 -2022-12-06 11:25:23,151 - Epoch: [140][ 1010/ 1200] Overall Loss 0.168681 Objective Loss 0.168681 LR 0.000250 Time 0.020002 -2022-12-06 11:25:23,342 - Epoch: [140][ 1020/ 1200] Overall Loss 0.168864 Objective Loss 0.168864 LR 0.000250 Time 0.019992 -2022-12-06 11:25:23,533 - Epoch: [140][ 1030/ 1200] Overall Loss 0.168776 Objective Loss 0.168776 LR 0.000250 Time 0.019983 -2022-12-06 11:25:23,724 - Epoch: [140][ 1040/ 1200] Overall Loss 0.168774 Objective Loss 0.168774 LR 0.000250 Time 0.019974 -2022-12-06 11:25:23,915 - Epoch: [140][ 1050/ 1200] Overall Loss 0.168674 Objective Loss 0.168674 LR 0.000250 Time 0.019965 -2022-12-06 11:25:24,106 - Epoch: [140][ 1060/ 1200] Overall Loss 0.168642 Objective Loss 0.168642 LR 0.000250 Time 0.019957 -2022-12-06 11:25:24,297 - Epoch: [140][ 1070/ 1200] Overall Loss 0.168475 Objective Loss 0.168475 LR 0.000250 Time 0.019948 -2022-12-06 11:25:24,488 - Epoch: [140][ 1080/ 1200] Overall Loss 0.168543 Objective Loss 0.168543 LR 0.000250 Time 0.019940 -2022-12-06 11:25:24,679 - Epoch: [140][ 1090/ 1200] Overall Loss 0.168607 Objective Loss 0.168607 LR 0.000250 Time 0.019932 -2022-12-06 11:25:24,870 - Epoch: [140][ 1100/ 1200] Overall Loss 0.168554 Objective Loss 0.168554 LR 0.000250 Time 0.019924 -2022-12-06 11:25:25,062 - Epoch: [140][ 1110/ 1200] Overall Loss 0.168362 Objective Loss 0.168362 LR 0.000250 Time 0.019916 -2022-12-06 11:25:25,253 - Epoch: [140][ 1120/ 1200] Overall Loss 0.168187 Objective Loss 0.168187 LR 0.000250 Time 0.019909 -2022-12-06 11:25:25,444 - Epoch: [140][ 1130/ 1200] Overall Loss 0.168244 Objective Loss 0.168244 LR 0.000250 Time 0.019901 -2022-12-06 11:25:25,635 - Epoch: [140][ 1140/ 1200] Overall Loss 0.168097 Objective Loss 0.168097 LR 0.000250 Time 0.019894 -2022-12-06 11:25:25,826 - Epoch: [140][ 1150/ 1200] Overall Loss 0.168085 Objective Loss 0.168085 LR 0.000250 Time 0.019886 -2022-12-06 11:25:26,017 - Epoch: [140][ 1160/ 1200] Overall Loss 0.168223 Objective Loss 0.168223 LR 0.000250 Time 0.019879 -2022-12-06 11:25:26,208 - Epoch: [140][ 1170/ 1200] Overall Loss 0.168137 Objective Loss 0.168137 LR 0.000250 Time 0.019872 -2022-12-06 11:25:26,400 - Epoch: [140][ 1180/ 1200] Overall Loss 0.168136 Objective Loss 0.168136 LR 0.000250 Time 0.019865 -2022-12-06 11:25:26,590 - Epoch: [140][ 1190/ 1200] Overall Loss 0.167992 Objective Loss 0.167992 LR 0.000250 Time 0.019858 -2022-12-06 11:25:26,815 - Epoch: [140][ 1200/ 1200] Overall Loss 0.168135 Objective Loss 0.168135 Top1 87.238494 Top5 99.163180 LR 0.000250 Time 0.019880 -2022-12-06 11:25:26,904 - --- validate (epoch=140)----------- -2022-12-06 11:25:26,905 - 34129 samples (256 per mini-batch) -2022-12-06 11:25:27,350 - Epoch: [140][ 10/ 134] Loss 0.241269 Top1 87.265625 Top5 98.437500 -2022-12-06 11:25:27,479 - Epoch: [140][ 20/ 134] Loss 0.241395 Top1 87.089844 Top5 98.359375 -2022-12-06 11:25:27,607 - Epoch: [140][ 30/ 134] Loss 0.247696 Top1 87.044271 Top5 98.398438 -2022-12-06 11:25:27,737 - Epoch: [140][ 40/ 134] Loss 0.240705 Top1 87.314453 Top5 98.457031 -2022-12-06 11:25:27,866 - Epoch: [140][ 50/ 134] Loss 0.236824 Top1 87.390625 Top5 98.468750 -2022-12-06 11:25:27,995 - Epoch: [140][ 60/ 134] Loss 0.233461 Top1 87.441406 Top5 98.457031 -2022-12-06 11:25:28,125 - Epoch: [140][ 70/ 134] Loss 0.236365 Top1 87.399554 Top5 98.392857 -2022-12-06 11:25:28,259 - Epoch: [140][ 80/ 134] Loss 0.241783 Top1 87.387695 Top5 98.403320 -2022-12-06 11:25:28,393 - Epoch: [140][ 90/ 134] Loss 0.241096 Top1 87.382812 Top5 98.467882 -2022-12-06 11:25:28,524 - Epoch: [140][ 100/ 134] Loss 0.241363 Top1 87.375000 Top5 98.468750 -2022-12-06 11:25:28,650 - Epoch: [140][ 110/ 134] Loss 0.242416 Top1 87.350852 Top5 98.490767 -2022-12-06 11:25:28,779 - Epoch: [140][ 120/ 134] Loss 0.242382 Top1 87.386068 Top5 98.444010 -2022-12-06 11:25:28,907 - Epoch: [140][ 130/ 134] Loss 0.242105 Top1 87.439904 Top5 98.458534 -2022-12-06 11:25:28,944 - Epoch: [140][ 134/ 134] Loss 0.241665 Top1 87.403674 Top5 98.464649 -2022-12-06 11:25:29,031 - ==> Top1: 87.404 Top5: 98.465 Loss: 0.242 - -2022-12-06 11:25:29,031 - ==> Confusion: -[[ 908 0 0 2 4 7 1 1 5 46 0 1 1 3 6 1 1 1 1 0 7] - [ 1 944 1 2 9 19 2 6 0 1 2 4 0 0 1 2 5 0 16 6 6] - [ 4 2 1018 12 7 1 22 8 0 1 4 3 2 0 2 2 0 0 2 3 10] - [ 1 1 24 950 1 1 1 0 1 0 9 0 3 2 10 1 1 1 9 1 3] - [ 6 3 0 0 956 5 1 0 0 4 2 2 1 2 16 5 8 2 3 1 3] - [ 2 18 0 2 5 976 1 18 2 2 1 9 4 12 2 1 1 1 2 6 4] - [ 2 2 11 0 1 2 1077 6 0 0 0 2 0 2 0 3 0 0 2 7 1] - [ 2 11 5 2 2 22 12 944 0 1 2 4 1 1 0 0 1 0 24 10 10] - [ 5 3 0 0 1 0 1 0 967 42 14 2 1 7 15 0 1 0 2 2 1] - [ 46 0 1 0 7 1 0 4 25 894 2 1 0 8 4 2 0 0 0 0 6] - [ 1 2 5 5 0 0 1 6 4 2 968 0 2 8 3 1 0 0 3 3 5] - [ 2 0 2 0 0 15 5 2 0 0 0 970 18 6 1 7 4 5 0 11 3] - [ 1 1 3 5 1 4 0 1 0 0 0 24 897 0 1 8 2 11 0 4 6] - [ 1 0 0 0 1 9 0 2 11 10 8 2 3 956 1 3 5 1 0 3 7] - [ 4 2 1 14 3 4 0 1 8 3 1 2 1 4 1069 0 1 1 5 1 5] - [ 1 0 3 1 1 0 5 0 2 0 1 7 4 1 0 994 6 12 0 4 1] - [ 2 2 0 2 2 1 1 0 0 1 0 2 2 1 1 10 1034 0 1 5 5] - [ 3 0 2 3 1 1 0 0 1 1 0 7 11 1 0 14 0 988 1 0 2] - [ 5 4 2 10 2 3 0 14 2 0 3 3 4 1 5 0 0 2 943 1 4] - [ 0 4 2 1 1 4 4 3 0 0 3 12 6 7 2 3 2 2 0 1016 8] - [ 115 199 205 123 105 143 95 122 74 70 162 90 296 214 147 116 193 70 137 196 10354]] - -2022-12-06 11:25:29,690 - ==> Best [Top1: 87.404 Top5: 98.465 Sparsity:0.00 Params: 5376 on epoch: 140] -2022-12-06 11:25:29,690 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:25:29,697 - - -2022-12-06 11:25:29,697 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:25:30,622 - Epoch: [141][ 10/ 1200] Overall Loss 0.162821 Objective Loss 0.162821 LR 0.000250 Time 0.092482 -2022-12-06 11:25:30,827 - Epoch: [141][ 20/ 1200] Overall Loss 0.175969 Objective Loss 0.175969 LR 0.000250 Time 0.056415 -2022-12-06 11:25:31,022 - Epoch: [141][ 30/ 1200] Overall Loss 0.170909 Objective Loss 0.170909 LR 0.000250 Time 0.044103 -2022-12-06 11:25:31,220 - Epoch: [141][ 40/ 1200] Overall Loss 0.169857 Objective Loss 0.169857 LR 0.000250 Time 0.038017 -2022-12-06 11:25:31,415 - Epoch: [141][ 50/ 1200] Overall Loss 0.168067 Objective Loss 0.168067 LR 0.000250 Time 0.034301 -2022-12-06 11:25:31,613 - Epoch: [141][ 60/ 1200] Overall Loss 0.164955 Objective Loss 0.164955 LR 0.000250 Time 0.031881 -2022-12-06 11:25:31,808 - Epoch: [141][ 70/ 1200] Overall Loss 0.161716 Objective Loss 0.161716 LR 0.000250 Time 0.030106 -2022-12-06 11:25:32,007 - Epoch: [141][ 80/ 1200] Overall Loss 0.162292 Objective Loss 0.162292 LR 0.000250 Time 0.028815 -2022-12-06 11:25:32,202 - Epoch: [141][ 90/ 1200] Overall Loss 0.161340 Objective Loss 0.161340 LR 0.000250 Time 0.027771 -2022-12-06 11:25:32,400 - Epoch: [141][ 100/ 1200] Overall Loss 0.163086 Objective Loss 0.163086 LR 0.000250 Time 0.026970 -2022-12-06 11:25:32,594 - Epoch: [141][ 110/ 1200] Overall Loss 0.163050 Objective Loss 0.163050 LR 0.000250 Time 0.026284 -2022-12-06 11:25:32,793 - Epoch: [141][ 120/ 1200] Overall Loss 0.161055 Objective Loss 0.161055 LR 0.000250 Time 0.025741 -2022-12-06 11:25:32,988 - Epoch: [141][ 130/ 1200] Overall Loss 0.160340 Objective Loss 0.160340 LR 0.000250 Time 0.025261 -2022-12-06 11:25:33,187 - Epoch: [141][ 140/ 1200] Overall Loss 0.162614 Objective Loss 0.162614 LR 0.000250 Time 0.024870 -2022-12-06 11:25:33,383 - Epoch: [141][ 150/ 1200] Overall Loss 0.163439 Objective Loss 0.163439 LR 0.000250 Time 0.024514 -2022-12-06 11:25:33,581 - Epoch: [141][ 160/ 1200] Overall Loss 0.162680 Objective Loss 0.162680 LR 0.000250 Time 0.024218 -2022-12-06 11:25:33,775 - Epoch: [141][ 170/ 1200] Overall Loss 0.162590 Objective Loss 0.162590 LR 0.000250 Time 0.023934 -2022-12-06 11:25:33,973 - Epoch: [141][ 180/ 1200] Overall Loss 0.163437 Objective Loss 0.163437 LR 0.000250 Time 0.023702 -2022-12-06 11:25:34,169 - Epoch: [141][ 190/ 1200] Overall Loss 0.163567 Objective Loss 0.163567 LR 0.000250 Time 0.023479 -2022-12-06 11:25:34,367 - Epoch: [141][ 200/ 1200] Overall Loss 0.163557 Objective Loss 0.163557 LR 0.000250 Time 0.023294 -2022-12-06 11:25:34,562 - Epoch: [141][ 210/ 1200] Overall Loss 0.163229 Objective Loss 0.163229 LR 0.000250 Time 0.023113 -2022-12-06 11:25:34,761 - Epoch: [141][ 220/ 1200] Overall Loss 0.163361 Objective Loss 0.163361 LR 0.000250 Time 0.022961 -2022-12-06 11:25:34,955 - Epoch: [141][ 230/ 1200] Overall Loss 0.163457 Objective Loss 0.163457 LR 0.000250 Time 0.022806 -2022-12-06 11:25:35,154 - Epoch: [141][ 240/ 1200] Overall Loss 0.162346 Objective Loss 0.162346 LR 0.000250 Time 0.022682 -2022-12-06 11:25:35,349 - Epoch: [141][ 250/ 1200] Overall Loss 0.162564 Objective Loss 0.162564 LR 0.000250 Time 0.022551 -2022-12-06 11:25:35,547 - Epoch: [141][ 260/ 1200] Overall Loss 0.162389 Objective Loss 0.162389 LR 0.000250 Time 0.022446 -2022-12-06 11:25:35,742 - Epoch: [141][ 270/ 1200] Overall Loss 0.163118 Objective Loss 0.163118 LR 0.000250 Time 0.022333 -2022-12-06 11:25:35,941 - Epoch: [141][ 280/ 1200] Overall Loss 0.163030 Objective Loss 0.163030 LR 0.000250 Time 0.022243 -2022-12-06 11:25:36,135 - Epoch: [141][ 290/ 1200] Overall Loss 0.162775 Objective Loss 0.162775 LR 0.000250 Time 0.022146 -2022-12-06 11:25:36,334 - Epoch: [141][ 300/ 1200] Overall Loss 0.162585 Objective Loss 0.162585 LR 0.000250 Time 0.022068 -2022-12-06 11:25:36,530 - Epoch: [141][ 310/ 1200] Overall Loss 0.162999 Objective Loss 0.162999 LR 0.000250 Time 0.021985 -2022-12-06 11:25:36,728 - Epoch: [141][ 320/ 1200] Overall Loss 0.163055 Objective Loss 0.163055 LR 0.000250 Time 0.021917 -2022-12-06 11:25:36,924 - Epoch: [141][ 330/ 1200] Overall Loss 0.163097 Objective Loss 0.163097 LR 0.000250 Time 0.021843 -2022-12-06 11:25:37,122 - Epoch: [141][ 340/ 1200] Overall Loss 0.162856 Objective Loss 0.162856 LR 0.000250 Time 0.021784 -2022-12-06 11:25:37,318 - Epoch: [141][ 350/ 1200] Overall Loss 0.162670 Objective Loss 0.162670 LR 0.000250 Time 0.021720 -2022-12-06 11:25:37,517 - Epoch: [141][ 360/ 1200] Overall Loss 0.162152 Objective Loss 0.162152 LR 0.000250 Time 0.021667 -2022-12-06 11:25:37,712 - Epoch: [141][ 370/ 1200] Overall Loss 0.161361 Objective Loss 0.161361 LR 0.000250 Time 0.021607 -2022-12-06 11:25:37,911 - Epoch: [141][ 380/ 1200] Overall Loss 0.161957 Objective Loss 0.161957 LR 0.000250 Time 0.021560 -2022-12-06 11:25:38,106 - Epoch: [141][ 390/ 1200] Overall Loss 0.161981 Objective Loss 0.161981 LR 0.000250 Time 0.021507 -2022-12-06 11:25:38,305 - Epoch: [141][ 400/ 1200] Overall Loss 0.162039 Objective Loss 0.162039 LR 0.000250 Time 0.021464 -2022-12-06 11:25:38,500 - Epoch: [141][ 410/ 1200] Overall Loss 0.162094 Objective Loss 0.162094 LR 0.000250 Time 0.021416 -2022-12-06 11:25:38,699 - Epoch: [141][ 420/ 1200] Overall Loss 0.162615 Objective Loss 0.162615 LR 0.000250 Time 0.021377 -2022-12-06 11:25:38,894 - Epoch: [141][ 430/ 1200] Overall Loss 0.162631 Objective Loss 0.162631 LR 0.000250 Time 0.021333 -2022-12-06 11:25:39,092 - Epoch: [141][ 440/ 1200] Overall Loss 0.162440 Objective Loss 0.162440 LR 0.000250 Time 0.021298 -2022-12-06 11:25:39,288 - Epoch: [141][ 450/ 1200] Overall Loss 0.162478 Objective Loss 0.162478 LR 0.000250 Time 0.021257 -2022-12-06 11:25:39,486 - Epoch: [141][ 460/ 1200] Overall Loss 0.162324 Objective Loss 0.162324 LR 0.000250 Time 0.021225 -2022-12-06 11:25:39,682 - Epoch: [141][ 470/ 1200] Overall Loss 0.162756 Objective Loss 0.162756 LR 0.000250 Time 0.021190 -2022-12-06 11:25:39,882 - Epoch: [141][ 480/ 1200] Overall Loss 0.162943 Objective Loss 0.162943 LR 0.000250 Time 0.021163 -2022-12-06 11:25:40,077 - Epoch: [141][ 490/ 1200] Overall Loss 0.162668 Objective Loss 0.162668 LR 0.000250 Time 0.021129 -2022-12-06 11:25:40,276 - Epoch: [141][ 500/ 1200] Overall Loss 0.162759 Objective Loss 0.162759 LR 0.000250 Time 0.021102 -2022-12-06 11:25:40,471 - Epoch: [141][ 510/ 1200] Overall Loss 0.163058 Objective Loss 0.163058 LR 0.000250 Time 0.021071 -2022-12-06 11:25:40,670 - Epoch: [141][ 520/ 1200] Overall Loss 0.163105 Objective Loss 0.163105 LR 0.000250 Time 0.021047 -2022-12-06 11:25:40,866 - Epoch: [141][ 530/ 1200] Overall Loss 0.162874 Objective Loss 0.162874 LR 0.000250 Time 0.021018 -2022-12-06 11:25:41,064 - Epoch: [141][ 540/ 1200] Overall Loss 0.163061 Objective Loss 0.163061 LR 0.000250 Time 0.020995 -2022-12-06 11:25:41,259 - Epoch: [141][ 550/ 1200] Overall Loss 0.163021 Objective Loss 0.163021 LR 0.000250 Time 0.020967 -2022-12-06 11:25:41,458 - Epoch: [141][ 560/ 1200] Overall Loss 0.163087 Objective Loss 0.163087 LR 0.000250 Time 0.020947 -2022-12-06 11:25:41,654 - Epoch: [141][ 570/ 1200] Overall Loss 0.163173 Objective Loss 0.163173 LR 0.000250 Time 0.020922 -2022-12-06 11:25:41,853 - Epoch: [141][ 580/ 1200] Overall Loss 0.163466 Objective Loss 0.163466 LR 0.000250 Time 0.020903 -2022-12-06 11:25:42,048 - Epoch: [141][ 590/ 1200] Overall Loss 0.163341 Objective Loss 0.163341 LR 0.000250 Time 0.020879 -2022-12-06 11:25:42,247 - Epoch: [141][ 600/ 1200] Overall Loss 0.163246 Objective Loss 0.163246 LR 0.000250 Time 0.020861 -2022-12-06 11:25:42,443 - Epoch: [141][ 610/ 1200] Overall Loss 0.162615 Objective Loss 0.162615 LR 0.000250 Time 0.020840 -2022-12-06 11:25:42,642 - Epoch: [141][ 620/ 1200] Overall Loss 0.162715 Objective Loss 0.162715 LR 0.000250 Time 0.020824 -2022-12-06 11:25:42,837 - Epoch: [141][ 630/ 1200] Overall Loss 0.162851 Objective Loss 0.162851 LR 0.000250 Time 0.020802 -2022-12-06 11:25:43,036 - Epoch: [141][ 640/ 1200] Overall Loss 0.162761 Objective Loss 0.162761 LR 0.000250 Time 0.020788 -2022-12-06 11:25:43,232 - Epoch: [141][ 650/ 1200] Overall Loss 0.162702 Objective Loss 0.162702 LR 0.000250 Time 0.020767 -2022-12-06 11:25:43,430 - Epoch: [141][ 660/ 1200] Overall Loss 0.162785 Objective Loss 0.162785 LR 0.000250 Time 0.020752 -2022-12-06 11:25:43,626 - Epoch: [141][ 670/ 1200] Overall Loss 0.162750 Objective Loss 0.162750 LR 0.000250 Time 0.020734 -2022-12-06 11:25:43,825 - Epoch: [141][ 680/ 1200] Overall Loss 0.162760 Objective Loss 0.162760 LR 0.000250 Time 0.020721 -2022-12-06 11:25:44,020 - Epoch: [141][ 690/ 1200] Overall Loss 0.162723 Objective Loss 0.162723 LR 0.000250 Time 0.020703 -2022-12-06 11:25:44,219 - Epoch: [141][ 700/ 1200] Overall Loss 0.162632 Objective Loss 0.162632 LR 0.000250 Time 0.020691 -2022-12-06 11:25:44,415 - Epoch: [141][ 710/ 1200] Overall Loss 0.162779 Objective Loss 0.162779 LR 0.000250 Time 0.020675 -2022-12-06 11:25:44,614 - Epoch: [141][ 720/ 1200] Overall Loss 0.162860 Objective Loss 0.162860 LR 0.000250 Time 0.020663 -2022-12-06 11:25:44,810 - Epoch: [141][ 730/ 1200] Overall Loss 0.162926 Objective Loss 0.162926 LR 0.000250 Time 0.020648 -2022-12-06 11:25:45,008 - Epoch: [141][ 740/ 1200] Overall Loss 0.163145 Objective Loss 0.163145 LR 0.000250 Time 0.020636 -2022-12-06 11:25:45,204 - Epoch: [141][ 750/ 1200] Overall Loss 0.163119 Objective Loss 0.163119 LR 0.000250 Time 0.020620 -2022-12-06 11:25:45,402 - Epoch: [141][ 760/ 1200] Overall Loss 0.163109 Objective Loss 0.163109 LR 0.000250 Time 0.020609 -2022-12-06 11:25:45,597 - Epoch: [141][ 770/ 1200] Overall Loss 0.162877 Objective Loss 0.162877 LR 0.000250 Time 0.020594 -2022-12-06 11:25:45,795 - Epoch: [141][ 780/ 1200] Overall Loss 0.163007 Objective Loss 0.163007 LR 0.000250 Time 0.020583 -2022-12-06 11:25:45,991 - Epoch: [141][ 790/ 1200] Overall Loss 0.162864 Objective Loss 0.162864 LR 0.000250 Time 0.020570 -2022-12-06 11:25:46,189 - Epoch: [141][ 800/ 1200] Overall Loss 0.162863 Objective Loss 0.162863 LR 0.000250 Time 0.020560 -2022-12-06 11:25:46,384 - Epoch: [141][ 810/ 1200] Overall Loss 0.162937 Objective Loss 0.162937 LR 0.000250 Time 0.020546 -2022-12-06 11:25:46,583 - Epoch: [141][ 820/ 1200] Overall Loss 0.162936 Objective Loss 0.162936 LR 0.000250 Time 0.020537 -2022-12-06 11:25:46,777 - Epoch: [141][ 830/ 1200] Overall Loss 0.162618 Objective Loss 0.162618 LR 0.000250 Time 0.020524 -2022-12-06 11:25:46,976 - Epoch: [141][ 840/ 1200] Overall Loss 0.162380 Objective Loss 0.162380 LR 0.000250 Time 0.020515 -2022-12-06 11:25:47,171 - Epoch: [141][ 850/ 1200] Overall Loss 0.162360 Objective Loss 0.162360 LR 0.000250 Time 0.020503 -2022-12-06 11:25:47,370 - Epoch: [141][ 860/ 1200] Overall Loss 0.162495 Objective Loss 0.162495 LR 0.000250 Time 0.020495 -2022-12-06 11:25:47,565 - Epoch: [141][ 870/ 1200] Overall Loss 0.162438 Objective Loss 0.162438 LR 0.000250 Time 0.020483 -2022-12-06 11:25:47,764 - Epoch: [141][ 880/ 1200] Overall Loss 0.162328 Objective Loss 0.162328 LR 0.000250 Time 0.020475 -2022-12-06 11:25:47,959 - Epoch: [141][ 890/ 1200] Overall Loss 0.162435 Objective Loss 0.162435 LR 0.000250 Time 0.020464 -2022-12-06 11:25:48,158 - Epoch: [141][ 900/ 1200] Overall Loss 0.162560 Objective Loss 0.162560 LR 0.000250 Time 0.020457 -2022-12-06 11:25:48,354 - Epoch: [141][ 910/ 1200] Overall Loss 0.162663 Objective Loss 0.162663 LR 0.000250 Time 0.020447 -2022-12-06 11:25:48,552 - Epoch: [141][ 920/ 1200] Overall Loss 0.162623 Objective Loss 0.162623 LR 0.000250 Time 0.020440 -2022-12-06 11:25:48,747 - Epoch: [141][ 930/ 1200] Overall Loss 0.162605 Objective Loss 0.162605 LR 0.000250 Time 0.020429 -2022-12-06 11:25:48,946 - Epoch: [141][ 940/ 1200] Overall Loss 0.162506 Objective Loss 0.162506 LR 0.000250 Time 0.020423 -2022-12-06 11:25:49,142 - Epoch: [141][ 950/ 1200] Overall Loss 0.162470 Objective Loss 0.162470 LR 0.000250 Time 0.020413 -2022-12-06 11:25:49,340 - Epoch: [141][ 960/ 1200] Overall Loss 0.162420 Objective Loss 0.162420 LR 0.000250 Time 0.020407 -2022-12-06 11:25:49,536 - Epoch: [141][ 970/ 1200] Overall Loss 0.162421 Objective Loss 0.162421 LR 0.000250 Time 0.020397 -2022-12-06 11:25:49,734 - Epoch: [141][ 980/ 1200] Overall Loss 0.162546 Objective Loss 0.162546 LR 0.000250 Time 0.020391 -2022-12-06 11:25:49,930 - Epoch: [141][ 990/ 1200] Overall Loss 0.162438 Objective Loss 0.162438 LR 0.000250 Time 0.020382 -2022-12-06 11:25:50,128 - Epoch: [141][ 1000/ 1200] Overall Loss 0.162484 Objective Loss 0.162484 LR 0.000250 Time 0.020376 -2022-12-06 11:25:50,323 - Epoch: [141][ 1010/ 1200] Overall Loss 0.162471 Objective Loss 0.162471 LR 0.000250 Time 0.020367 -2022-12-06 11:25:50,521 - Epoch: [141][ 1020/ 1200] Overall Loss 0.162505 Objective Loss 0.162505 LR 0.000250 Time 0.020361 -2022-12-06 11:25:50,717 - Epoch: [141][ 1030/ 1200] Overall Loss 0.162404 Objective Loss 0.162404 LR 0.000250 Time 0.020353 -2022-12-06 11:25:50,915 - Epoch: [141][ 1040/ 1200] Overall Loss 0.162712 Objective Loss 0.162712 LR 0.000250 Time 0.020347 -2022-12-06 11:25:51,111 - Epoch: [141][ 1050/ 1200] Overall Loss 0.163064 Objective Loss 0.163064 LR 0.000250 Time 0.020339 -2022-12-06 11:25:51,309 - Epoch: [141][ 1060/ 1200] Overall Loss 0.163202 Objective Loss 0.163202 LR 0.000250 Time 0.020334 -2022-12-06 11:25:51,504 - Epoch: [141][ 1070/ 1200] Overall Loss 0.163116 Objective Loss 0.163116 LR 0.000250 Time 0.020325 -2022-12-06 11:25:51,703 - Epoch: [141][ 1080/ 1200] Overall Loss 0.163050 Objective Loss 0.163050 LR 0.000250 Time 0.020321 -2022-12-06 11:25:51,898 - Epoch: [141][ 1090/ 1200] Overall Loss 0.163089 Objective Loss 0.163089 LR 0.000250 Time 0.020313 -2022-12-06 11:25:52,097 - Epoch: [141][ 1100/ 1200] Overall Loss 0.163092 Objective Loss 0.163092 LR 0.000250 Time 0.020308 -2022-12-06 11:25:52,292 - Epoch: [141][ 1110/ 1200] Overall Loss 0.163187 Objective Loss 0.163187 LR 0.000250 Time 0.020301 -2022-12-06 11:25:52,490 - Epoch: [141][ 1120/ 1200] Overall Loss 0.163358 Objective Loss 0.163358 LR 0.000250 Time 0.020296 -2022-12-06 11:25:52,686 - Epoch: [141][ 1130/ 1200] Overall Loss 0.163270 Objective Loss 0.163270 LR 0.000250 Time 0.020289 -2022-12-06 11:25:52,884 - Epoch: [141][ 1140/ 1200] Overall Loss 0.163247 Objective Loss 0.163247 LR 0.000250 Time 0.020285 -2022-12-06 11:25:53,079 - Epoch: [141][ 1150/ 1200] Overall Loss 0.163431 Objective Loss 0.163431 LR 0.000250 Time 0.020278 -2022-12-06 11:25:53,278 - Epoch: [141][ 1160/ 1200] Overall Loss 0.163398 Objective Loss 0.163398 LR 0.000250 Time 0.020274 -2022-12-06 11:25:53,473 - Epoch: [141][ 1170/ 1200] Overall Loss 0.163468 Objective Loss 0.163468 LR 0.000250 Time 0.020267 -2022-12-06 11:25:53,672 - Epoch: [141][ 1180/ 1200] Overall Loss 0.163501 Objective Loss 0.163501 LR 0.000250 Time 0.020262 -2022-12-06 11:25:53,867 - Epoch: [141][ 1190/ 1200] Overall Loss 0.163685 Objective Loss 0.163685 LR 0.000250 Time 0.020256 -2022-12-06 11:25:54,088 - Epoch: [141][ 1200/ 1200] Overall Loss 0.163770 Objective Loss 0.163770 Top1 88.493724 Top5 98.744770 LR 0.000250 Time 0.020271 -2022-12-06 11:25:54,177 - --- validate (epoch=141)----------- -2022-12-06 11:25:54,177 - 34129 samples (256 per mini-batch) -2022-12-06 11:25:54,623 - Epoch: [141][ 10/ 134] Loss 0.236837 Top1 87.265625 Top5 98.398438 -2022-12-06 11:25:54,751 - Epoch: [141][ 20/ 134] Loss 0.232701 Top1 87.675781 Top5 98.320312 -2022-12-06 11:25:54,876 - Epoch: [141][ 30/ 134] Loss 0.235719 Top1 87.747396 Top5 98.307292 -2022-12-06 11:25:55,002 - Epoch: [141][ 40/ 134] Loss 0.244621 Top1 87.519531 Top5 98.222656 -2022-12-06 11:25:55,126 - Epoch: [141][ 50/ 134] Loss 0.251308 Top1 87.570312 Top5 98.312500 -2022-12-06 11:25:55,252 - Epoch: [141][ 60/ 134] Loss 0.246690 Top1 87.649740 Top5 98.320312 -2022-12-06 11:25:55,380 - Epoch: [141][ 70/ 134] Loss 0.246436 Top1 87.712054 Top5 98.370536 -2022-12-06 11:25:55,508 - Epoch: [141][ 80/ 134] Loss 0.246573 Top1 87.622070 Top5 98.359375 -2022-12-06 11:25:55,636 - Epoch: [141][ 90/ 134] Loss 0.243853 Top1 87.717014 Top5 98.372396 -2022-12-06 11:25:55,762 - Epoch: [141][ 100/ 134] Loss 0.239820 Top1 87.765625 Top5 98.390625 -2022-12-06 11:25:55,889 - Epoch: [141][ 110/ 134] Loss 0.238727 Top1 87.816051 Top5 98.416193 -2022-12-06 11:25:56,014 - Epoch: [141][ 120/ 134] Loss 0.239154 Top1 87.792969 Top5 98.437500 -2022-12-06 11:25:56,143 - Epoch: [141][ 130/ 134] Loss 0.238634 Top1 87.833534 Top5 98.461538 -2022-12-06 11:25:56,179 - Epoch: [141][ 134/ 134] Loss 0.238201 Top1 87.890064 Top5 98.479299 -2022-12-06 11:25:56,266 - ==> Top1: 87.890 Top5: 98.479 Loss: 0.238 - -2022-12-06 11:25:56,267 - ==> Confusion: -[[ 915 0 2 3 6 5 1 0 3 42 0 1 0 3 6 1 1 1 1 0 5] - [ 1 944 1 2 9 19 2 11 0 1 3 5 0 1 2 3 3 0 10 5 5] - [ 3 3 1014 8 6 3 20 9 0 2 4 3 1 1 1 1 2 2 4 5 11] - [ 2 0 19 954 1 2 0 1 0 1 9 1 3 3 8 0 1 1 9 0 5] - [ 8 4 2 0 963 1 1 2 0 7 0 2 1 2 6 4 6 1 0 3 7] - [ 1 13 0 2 6 967 4 21 3 4 1 12 2 18 0 1 2 1 1 6 4] - [ 0 4 7 1 1 0 1075 4 0 1 1 1 0 1 0 6 2 1 2 8 3] - [ 1 6 10 3 2 20 9 964 0 0 0 5 0 3 0 1 0 1 12 9 8] - [ 4 1 0 0 0 1 1 0 978 41 9 1 1 6 10 1 1 0 3 1 5] - [ 61 0 1 0 3 2 1 1 26 880 2 2 0 13 1 1 0 1 0 0 6] - [ 2 1 5 3 2 0 1 3 6 1 973 0 0 9 2 2 1 0 2 1 5] - [ 3 0 2 0 0 9 5 2 0 0 0 980 18 4 1 7 3 7 0 7 3] - [ 0 1 1 2 0 3 1 1 0 0 0 29 902 2 0 7 1 9 0 4 6] - [ 1 1 0 0 0 9 0 2 9 11 4 3 3 964 1 3 2 1 0 1 8] - [ 7 2 2 12 3 1 0 0 15 3 2 4 2 3 1059 0 0 2 5 0 8] - [ 0 0 1 2 1 0 1 0 0 0 1 8 3 4 0 1003 2 9 1 3 4] - [ 1 1 0 1 1 1 2 1 1 1 0 6 1 1 1 11 1032 0 0 2 8] - [ 2 0 1 3 1 1 0 1 0 5 0 6 8 1 2 12 0 988 0 2 3] - [ 3 3 4 6 2 1 1 19 1 1 1 3 4 0 4 1 0 0 947 1 6] - [ 2 2 2 1 0 5 5 6 0 1 3 18 5 6 0 2 3 2 1 1010 6] - [ 108 179 159 106 89 127 72 131 76 78 136 108 297 274 116 118 169 76 142 186 10479]] - -2022-12-06 11:25:56,927 - ==> Best [Top1: 87.890 Top5: 98.479 Sparsity:0.00 Params: 5376 on epoch: 141] -2022-12-06 11:25:56,927 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:25:56,934 - - -2022-12-06 11:25:56,934 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:25:57,871 - Epoch: [142][ 10/ 1200] Overall Loss 0.180759 Objective Loss 0.180759 LR 0.000250 Time 0.093606 -2022-12-06 11:25:58,078 - Epoch: [142][ 20/ 1200] Overall Loss 0.166109 Objective Loss 0.166109 LR 0.000250 Time 0.057122 -2022-12-06 11:25:58,275 - Epoch: [142][ 30/ 1200] Overall Loss 0.169672 Objective Loss 0.169672 LR 0.000250 Time 0.044628 -2022-12-06 11:25:58,474 - Epoch: [142][ 40/ 1200] Overall Loss 0.164106 Objective Loss 0.164106 LR 0.000250 Time 0.038439 -2022-12-06 11:25:58,670 - Epoch: [142][ 50/ 1200] Overall Loss 0.165794 Objective Loss 0.165794 LR 0.000250 Time 0.034653 -2022-12-06 11:25:58,869 - Epoch: [142][ 60/ 1200] Overall Loss 0.163527 Objective Loss 0.163527 LR 0.000250 Time 0.032180 -2022-12-06 11:25:59,064 - Epoch: [142][ 70/ 1200] Overall Loss 0.167469 Objective Loss 0.167469 LR 0.000250 Time 0.030362 -2022-12-06 11:25:59,262 - Epoch: [142][ 80/ 1200] Overall Loss 0.166354 Objective Loss 0.166354 LR 0.000250 Time 0.029043 -2022-12-06 11:25:59,458 - Epoch: [142][ 90/ 1200] Overall Loss 0.165344 Objective Loss 0.165344 LR 0.000250 Time 0.027985 -2022-12-06 11:25:59,658 - Epoch: [142][ 100/ 1200] Overall Loss 0.165230 Objective Loss 0.165230 LR 0.000250 Time 0.027177 -2022-12-06 11:25:59,853 - Epoch: [142][ 110/ 1200] Overall Loss 0.164419 Objective Loss 0.164419 LR 0.000250 Time 0.026480 -2022-12-06 11:26:00,052 - Epoch: [142][ 120/ 1200] Overall Loss 0.164077 Objective Loss 0.164077 LR 0.000250 Time 0.025926 -2022-12-06 11:26:00,247 - Epoch: [142][ 130/ 1200] Overall Loss 0.163531 Objective Loss 0.163531 LR 0.000250 Time 0.025430 -2022-12-06 11:26:00,446 - Epoch: [142][ 140/ 1200] Overall Loss 0.163552 Objective Loss 0.163552 LR 0.000250 Time 0.025025 -2022-12-06 11:26:00,641 - Epoch: [142][ 150/ 1200] Overall Loss 0.162760 Objective Loss 0.162760 LR 0.000250 Time 0.024657 -2022-12-06 11:26:00,841 - Epoch: [142][ 160/ 1200] Overall Loss 0.162549 Objective Loss 0.162549 LR 0.000250 Time 0.024358 -2022-12-06 11:26:01,036 - Epoch: [142][ 170/ 1200] Overall Loss 0.163167 Objective Loss 0.163167 LR 0.000250 Time 0.024071 -2022-12-06 11:26:01,236 - Epoch: [142][ 180/ 1200] Overall Loss 0.163123 Objective Loss 0.163123 LR 0.000250 Time 0.023846 -2022-12-06 11:26:01,433 - Epoch: [142][ 190/ 1200] Overall Loss 0.162179 Objective Loss 0.162179 LR 0.000250 Time 0.023623 -2022-12-06 11:26:01,634 - Epoch: [142][ 200/ 1200] Overall Loss 0.161430 Objective Loss 0.161430 LR 0.000250 Time 0.023444 -2022-12-06 11:26:01,832 - Epoch: [142][ 210/ 1200] Overall Loss 0.162350 Objective Loss 0.162350 LR 0.000250 Time 0.023265 -2022-12-06 11:26:02,033 - Epoch: [142][ 220/ 1200] Overall Loss 0.161843 Objective Loss 0.161843 LR 0.000250 Time 0.023119 -2022-12-06 11:26:02,229 - Epoch: [142][ 230/ 1200] Overall Loss 0.162242 Objective Loss 0.162242 LR 0.000250 Time 0.022966 -2022-12-06 11:26:02,429 - Epoch: [142][ 240/ 1200] Overall Loss 0.162689 Objective Loss 0.162689 LR 0.000250 Time 0.022839 -2022-12-06 11:26:02,627 - Epoch: [142][ 250/ 1200] Overall Loss 0.162262 Objective Loss 0.162262 LR 0.000250 Time 0.022714 -2022-12-06 11:26:02,826 - Epoch: [142][ 260/ 1200] Overall Loss 0.162394 Objective Loss 0.162394 LR 0.000250 Time 0.022606 -2022-12-06 11:26:03,024 - Epoch: [142][ 270/ 1200] Overall Loss 0.162175 Objective Loss 0.162175 LR 0.000250 Time 0.022498 -2022-12-06 11:26:03,224 - Epoch: [142][ 280/ 1200] Overall Loss 0.161403 Objective Loss 0.161403 LR 0.000250 Time 0.022406 -2022-12-06 11:26:03,421 - Epoch: [142][ 290/ 1200] Overall Loss 0.160968 Objective Loss 0.160968 LR 0.000250 Time 0.022312 -2022-12-06 11:26:03,621 - Epoch: [142][ 300/ 1200] Overall Loss 0.161635 Objective Loss 0.161635 LR 0.000250 Time 0.022232 -2022-12-06 11:26:03,818 - Epoch: [142][ 310/ 1200] Overall Loss 0.161789 Objective Loss 0.161789 LR 0.000250 Time 0.022149 -2022-12-06 11:26:04,018 - Epoch: [142][ 320/ 1200] Overall Loss 0.161757 Objective Loss 0.161757 LR 0.000250 Time 0.022080 -2022-12-06 11:26:04,214 - Epoch: [142][ 330/ 1200] Overall Loss 0.161090 Objective Loss 0.161090 LR 0.000250 Time 0.022006 -2022-12-06 11:26:04,414 - Epoch: [142][ 340/ 1200] Overall Loss 0.160905 Objective Loss 0.160905 LR 0.000250 Time 0.021945 -2022-12-06 11:26:04,611 - Epoch: [142][ 350/ 1200] Overall Loss 0.160691 Objective Loss 0.160691 LR 0.000250 Time 0.021879 -2022-12-06 11:26:04,813 - Epoch: [142][ 360/ 1200] Overall Loss 0.161041 Objective Loss 0.161041 LR 0.000250 Time 0.021829 -2022-12-06 11:26:05,010 - Epoch: [142][ 370/ 1200] Overall Loss 0.161392 Objective Loss 0.161392 LR 0.000250 Time 0.021771 -2022-12-06 11:26:05,210 - Epoch: [142][ 380/ 1200] Overall Loss 0.161122 Objective Loss 0.161122 LR 0.000250 Time 0.021722 -2022-12-06 11:26:05,407 - Epoch: [142][ 390/ 1200] Overall Loss 0.160816 Objective Loss 0.160816 LR 0.000250 Time 0.021671 -2022-12-06 11:26:05,608 - Epoch: [142][ 400/ 1200] Overall Loss 0.160741 Objective Loss 0.160741 LR 0.000250 Time 0.021628 -2022-12-06 11:26:05,805 - Epoch: [142][ 410/ 1200] Overall Loss 0.160822 Objective Loss 0.160822 LR 0.000250 Time 0.021580 -2022-12-06 11:26:06,005 - Epoch: [142][ 420/ 1200] Overall Loss 0.160254 Objective Loss 0.160254 LR 0.000250 Time 0.021541 -2022-12-06 11:26:06,202 - Epoch: [142][ 430/ 1200] Overall Loss 0.160452 Objective Loss 0.160452 LR 0.000250 Time 0.021497 -2022-12-06 11:26:06,401 - Epoch: [142][ 440/ 1200] Overall Loss 0.160202 Objective Loss 0.160202 LR 0.000250 Time 0.021460 -2022-12-06 11:26:06,596 - Epoch: [142][ 450/ 1200] Overall Loss 0.160246 Objective Loss 0.160246 LR 0.000250 Time 0.021416 -2022-12-06 11:26:06,795 - Epoch: [142][ 460/ 1200] Overall Loss 0.159827 Objective Loss 0.159827 LR 0.000250 Time 0.021382 -2022-12-06 11:26:06,991 - Epoch: [142][ 470/ 1200] Overall Loss 0.159245 Objective Loss 0.159245 LR 0.000250 Time 0.021341 -2022-12-06 11:26:07,190 - Epoch: [142][ 480/ 1200] Overall Loss 0.159117 Objective Loss 0.159117 LR 0.000250 Time 0.021310 -2022-12-06 11:26:07,385 - Epoch: [142][ 490/ 1200] Overall Loss 0.159056 Objective Loss 0.159056 LR 0.000250 Time 0.021272 -2022-12-06 11:26:07,583 - Epoch: [142][ 500/ 1200] Overall Loss 0.159143 Objective Loss 0.159143 LR 0.000250 Time 0.021242 -2022-12-06 11:26:07,778 - Epoch: [142][ 510/ 1200] Overall Loss 0.159450 Objective Loss 0.159450 LR 0.000250 Time 0.021207 -2022-12-06 11:26:07,977 - Epoch: [142][ 520/ 1200] Overall Loss 0.159661 Objective Loss 0.159661 LR 0.000250 Time 0.021181 -2022-12-06 11:26:08,172 - Epoch: [142][ 530/ 1200] Overall Loss 0.160003 Objective Loss 0.160003 LR 0.000250 Time 0.021149 -2022-12-06 11:26:08,371 - Epoch: [142][ 540/ 1200] Overall Loss 0.159797 Objective Loss 0.159797 LR 0.000250 Time 0.021125 -2022-12-06 11:26:08,566 - Epoch: [142][ 550/ 1200] Overall Loss 0.159935 Objective Loss 0.159935 LR 0.000250 Time 0.021094 -2022-12-06 11:26:08,765 - Epoch: [142][ 560/ 1200] Overall Loss 0.160076 Objective Loss 0.160076 LR 0.000250 Time 0.021072 -2022-12-06 11:26:08,961 - Epoch: [142][ 570/ 1200] Overall Loss 0.159996 Objective Loss 0.159996 LR 0.000250 Time 0.021045 -2022-12-06 11:26:09,161 - Epoch: [142][ 580/ 1200] Overall Loss 0.159700 Objective Loss 0.159700 LR 0.000250 Time 0.021025 -2022-12-06 11:26:09,356 - Epoch: [142][ 590/ 1200] Overall Loss 0.159735 Objective Loss 0.159735 LR 0.000250 Time 0.020999 -2022-12-06 11:26:09,555 - Epoch: [142][ 600/ 1200] Overall Loss 0.159500 Objective Loss 0.159500 LR 0.000250 Time 0.020980 -2022-12-06 11:26:09,751 - Epoch: [142][ 610/ 1200] Overall Loss 0.159948 Objective Loss 0.159948 LR 0.000250 Time 0.020956 -2022-12-06 11:26:09,950 - Epoch: [142][ 620/ 1200] Overall Loss 0.160202 Objective Loss 0.160202 LR 0.000250 Time 0.020937 -2022-12-06 11:26:10,145 - Epoch: [142][ 630/ 1200] Overall Loss 0.159994 Objective Loss 0.159994 LR 0.000250 Time 0.020914 -2022-12-06 11:26:10,344 - Epoch: [142][ 640/ 1200] Overall Loss 0.159855 Objective Loss 0.159855 LR 0.000250 Time 0.020897 -2022-12-06 11:26:10,538 - Epoch: [142][ 650/ 1200] Overall Loss 0.160096 Objective Loss 0.160096 LR 0.000250 Time 0.020874 -2022-12-06 11:26:10,737 - Epoch: [142][ 660/ 1200] Overall Loss 0.159978 Objective Loss 0.159978 LR 0.000250 Time 0.020858 -2022-12-06 11:26:10,932 - Epoch: [142][ 670/ 1200] Overall Loss 0.159988 Objective Loss 0.159988 LR 0.000250 Time 0.020837 -2022-12-06 11:26:11,131 - Epoch: [142][ 680/ 1200] Overall Loss 0.160128 Objective Loss 0.160128 LR 0.000250 Time 0.020822 -2022-12-06 11:26:11,326 - Epoch: [142][ 690/ 1200] Overall Loss 0.160182 Objective Loss 0.160182 LR 0.000250 Time 0.020802 -2022-12-06 11:26:11,524 - Epoch: [142][ 700/ 1200] Overall Loss 0.160012 Objective Loss 0.160012 LR 0.000250 Time 0.020788 -2022-12-06 11:26:11,719 - Epoch: [142][ 710/ 1200] Overall Loss 0.160109 Objective Loss 0.160109 LR 0.000250 Time 0.020769 -2022-12-06 11:26:11,919 - Epoch: [142][ 720/ 1200] Overall Loss 0.160447 Objective Loss 0.160447 LR 0.000250 Time 0.020757 -2022-12-06 11:26:12,114 - Epoch: [142][ 730/ 1200] Overall Loss 0.160420 Objective Loss 0.160420 LR 0.000250 Time 0.020739 -2022-12-06 11:26:12,312 - Epoch: [142][ 740/ 1200] Overall Loss 0.160589 Objective Loss 0.160589 LR 0.000250 Time 0.020726 -2022-12-06 11:26:12,508 - Epoch: [142][ 750/ 1200] Overall Loss 0.160646 Objective Loss 0.160646 LR 0.000250 Time 0.020710 -2022-12-06 11:26:12,706 - Epoch: [142][ 760/ 1200] Overall Loss 0.160356 Objective Loss 0.160356 LR 0.000250 Time 0.020698 -2022-12-06 11:26:12,901 - Epoch: [142][ 770/ 1200] Overall Loss 0.160300 Objective Loss 0.160300 LR 0.000250 Time 0.020681 -2022-12-06 11:26:13,099 - Epoch: [142][ 780/ 1200] Overall Loss 0.160167 Objective Loss 0.160167 LR 0.000250 Time 0.020669 -2022-12-06 11:26:13,294 - Epoch: [142][ 790/ 1200] Overall Loss 0.160066 Objective Loss 0.160066 LR 0.000250 Time 0.020654 -2022-12-06 11:26:13,493 - Epoch: [142][ 800/ 1200] Overall Loss 0.159969 Objective Loss 0.159969 LR 0.000250 Time 0.020644 -2022-12-06 11:26:13,688 - Epoch: [142][ 810/ 1200] Overall Loss 0.160225 Objective Loss 0.160225 LR 0.000250 Time 0.020629 -2022-12-06 11:26:13,886 - Epoch: [142][ 820/ 1200] Overall Loss 0.160516 Objective Loss 0.160516 LR 0.000250 Time 0.020619 -2022-12-06 11:26:14,082 - Epoch: [142][ 830/ 1200] Overall Loss 0.160492 Objective Loss 0.160492 LR 0.000250 Time 0.020605 -2022-12-06 11:26:14,281 - Epoch: [142][ 840/ 1200] Overall Loss 0.160423 Objective Loss 0.160423 LR 0.000250 Time 0.020596 -2022-12-06 11:26:14,476 - Epoch: [142][ 850/ 1200] Overall Loss 0.160412 Objective Loss 0.160412 LR 0.000250 Time 0.020582 -2022-12-06 11:26:14,674 - Epoch: [142][ 860/ 1200] Overall Loss 0.160567 Objective Loss 0.160567 LR 0.000250 Time 0.020573 -2022-12-06 11:26:14,869 - Epoch: [142][ 870/ 1200] Overall Loss 0.160661 Objective Loss 0.160661 LR 0.000250 Time 0.020560 -2022-12-06 11:26:15,068 - Epoch: [142][ 880/ 1200] Overall Loss 0.160727 Objective Loss 0.160727 LR 0.000250 Time 0.020552 -2022-12-06 11:26:15,263 - Epoch: [142][ 890/ 1200] Overall Loss 0.160830 Objective Loss 0.160830 LR 0.000250 Time 0.020540 -2022-12-06 11:26:15,462 - Epoch: [142][ 900/ 1200] Overall Loss 0.160871 Objective Loss 0.160871 LR 0.000250 Time 0.020532 -2022-12-06 11:26:15,657 - Epoch: [142][ 910/ 1200] Overall Loss 0.160860 Objective Loss 0.160860 LR 0.000250 Time 0.020520 -2022-12-06 11:26:15,857 - Epoch: [142][ 920/ 1200] Overall Loss 0.160814 Objective Loss 0.160814 LR 0.000250 Time 0.020513 -2022-12-06 11:26:16,053 - Epoch: [142][ 930/ 1200] Overall Loss 0.160888 Objective Loss 0.160888 LR 0.000250 Time 0.020503 -2022-12-06 11:26:16,252 - Epoch: [142][ 940/ 1200] Overall Loss 0.160799 Objective Loss 0.160799 LR 0.000250 Time 0.020496 -2022-12-06 11:26:16,447 - Epoch: [142][ 950/ 1200] Overall Loss 0.160835 Objective Loss 0.160835 LR 0.000250 Time 0.020485 -2022-12-06 11:26:16,646 - Epoch: [142][ 960/ 1200] Overall Loss 0.161026 Objective Loss 0.161026 LR 0.000250 Time 0.020478 -2022-12-06 11:26:16,842 - Epoch: [142][ 970/ 1200] Overall Loss 0.160963 Objective Loss 0.160963 LR 0.000250 Time 0.020468 -2022-12-06 11:26:17,040 - Epoch: [142][ 980/ 1200] Overall Loss 0.160798 Objective Loss 0.160798 LR 0.000250 Time 0.020462 -2022-12-06 11:26:17,236 - Epoch: [142][ 990/ 1200] Overall Loss 0.160758 Objective Loss 0.160758 LR 0.000250 Time 0.020452 -2022-12-06 11:26:17,435 - Epoch: [142][ 1000/ 1200] Overall Loss 0.160876 Objective Loss 0.160876 LR 0.000250 Time 0.020446 -2022-12-06 11:26:17,630 - Epoch: [142][ 1010/ 1200] Overall Loss 0.161076 Objective Loss 0.161076 LR 0.000250 Time 0.020436 -2022-12-06 11:26:17,829 - Epoch: [142][ 1020/ 1200] Overall Loss 0.161073 Objective Loss 0.161073 LR 0.000250 Time 0.020430 -2022-12-06 11:26:18,024 - Epoch: [142][ 1030/ 1200] Overall Loss 0.161108 Objective Loss 0.161108 LR 0.000250 Time 0.020421 -2022-12-06 11:26:18,223 - Epoch: [142][ 1040/ 1200] Overall Loss 0.161168 Objective Loss 0.161168 LR 0.000250 Time 0.020415 -2022-12-06 11:26:18,419 - Epoch: [142][ 1050/ 1200] Overall Loss 0.161090 Objective Loss 0.161090 LR 0.000250 Time 0.020407 -2022-12-06 11:26:18,618 - Epoch: [142][ 1060/ 1200] Overall Loss 0.161168 Objective Loss 0.161168 LR 0.000250 Time 0.020402 -2022-12-06 11:26:18,814 - Epoch: [142][ 1070/ 1200] Overall Loss 0.161286 Objective Loss 0.161286 LR 0.000250 Time 0.020393 -2022-12-06 11:26:19,012 - Epoch: [142][ 1080/ 1200] Overall Loss 0.161185 Objective Loss 0.161185 LR 0.000250 Time 0.020387 -2022-12-06 11:26:19,208 - Epoch: [142][ 1090/ 1200] Overall Loss 0.161251 Objective Loss 0.161251 LR 0.000250 Time 0.020380 -2022-12-06 11:26:19,406 - Epoch: [142][ 1100/ 1200] Overall Loss 0.161247 Objective Loss 0.161247 LR 0.000250 Time 0.020374 -2022-12-06 11:26:19,601 - Epoch: [142][ 1110/ 1200] Overall Loss 0.161296 Objective Loss 0.161296 LR 0.000250 Time 0.020366 -2022-12-06 11:26:19,800 - Epoch: [142][ 1120/ 1200] Overall Loss 0.161169 Objective Loss 0.161169 LR 0.000250 Time 0.020361 -2022-12-06 11:26:19,995 - Epoch: [142][ 1130/ 1200] Overall Loss 0.161375 Objective Loss 0.161375 LR 0.000250 Time 0.020353 -2022-12-06 11:26:20,194 - Epoch: [142][ 1140/ 1200] Overall Loss 0.161483 Objective Loss 0.161483 LR 0.000250 Time 0.020348 -2022-12-06 11:26:20,389 - Epoch: [142][ 1150/ 1200] Overall Loss 0.161529 Objective Loss 0.161529 LR 0.000250 Time 0.020341 -2022-12-06 11:26:20,588 - Epoch: [142][ 1160/ 1200] Overall Loss 0.161695 Objective Loss 0.161695 LR 0.000250 Time 0.020336 -2022-12-06 11:26:20,783 - Epoch: [142][ 1170/ 1200] Overall Loss 0.161566 Objective Loss 0.161566 LR 0.000250 Time 0.020329 -2022-12-06 11:26:20,983 - Epoch: [142][ 1180/ 1200] Overall Loss 0.161788 Objective Loss 0.161788 LR 0.000250 Time 0.020326 -2022-12-06 11:26:21,178 - Epoch: [142][ 1190/ 1200] Overall Loss 0.161859 Objective Loss 0.161859 LR 0.000250 Time 0.020318 -2022-12-06 11:26:21,408 - Epoch: [142][ 1200/ 1200] Overall Loss 0.161656 Objective Loss 0.161656 Top1 90.376569 Top5 99.163180 LR 0.000250 Time 0.020340 -2022-12-06 11:26:21,497 - --- validate (epoch=142)----------- -2022-12-06 11:26:21,497 - 34129 samples (256 per mini-batch) -2022-12-06 11:26:21,941 - Epoch: [142][ 10/ 134] Loss 0.263333 Top1 87.343750 Top5 98.203125 -2022-12-06 11:26:22,073 - Epoch: [142][ 20/ 134] Loss 0.260025 Top1 87.207031 Top5 98.281250 -2022-12-06 11:26:22,209 - Epoch: [142][ 30/ 134] Loss 0.252366 Top1 86.888021 Top5 98.398438 -2022-12-06 11:26:22,341 - Epoch: [142][ 40/ 134] Loss 0.241170 Top1 87.255859 Top5 98.486328 -2022-12-06 11:26:22,471 - Epoch: [142][ 50/ 134] Loss 0.238262 Top1 87.343750 Top5 98.476562 -2022-12-06 11:26:22,613 - Epoch: [142][ 60/ 134] Loss 0.240632 Top1 87.369792 Top5 98.509115 -2022-12-06 11:26:22,738 - Epoch: [142][ 70/ 134] Loss 0.239987 Top1 87.332589 Top5 98.521205 -2022-12-06 11:26:22,864 - Epoch: [142][ 80/ 134] Loss 0.238598 Top1 87.348633 Top5 98.530273 -2022-12-06 11:26:22,990 - Epoch: [142][ 90/ 134] Loss 0.238290 Top1 87.460938 Top5 98.485243 -2022-12-06 11:26:23,116 - Epoch: [142][ 100/ 134] Loss 0.235681 Top1 87.562500 Top5 98.480469 -2022-12-06 11:26:23,242 - Epoch: [142][ 110/ 134] Loss 0.233703 Top1 87.606534 Top5 98.480114 -2022-12-06 11:26:23,369 - Epoch: [142][ 120/ 134] Loss 0.233070 Top1 87.652995 Top5 98.476562 -2022-12-06 11:26:23,494 - Epoch: [142][ 130/ 134] Loss 0.235712 Top1 87.542067 Top5 98.446514 -2022-12-06 11:26:23,531 - Epoch: [142][ 134/ 134] Loss 0.236685 Top1 87.526737 Top5 98.447068 -2022-12-06 11:26:23,626 - ==> Top1: 87.527 Top5: 98.447 Loss: 0.237 - -2022-12-06 11:26:23,627 - ==> Confusion: -[[ 923 2 2 0 6 5 1 0 2 44 0 1 0 2 2 1 2 0 0 0 3] - [ 1 943 2 2 8 18 2 12 1 2 1 4 1 1 0 1 5 1 7 4 11] - [ 4 3 1020 14 5 3 14 5 0 1 7 3 2 0 1 3 1 1 3 4 9] - [ 1 1 17 960 2 2 0 1 0 0 10 1 2 2 7 0 0 1 9 0 4] - [ 11 5 1 0 963 2 1 3 1 6 1 2 0 2 5 6 5 2 0 2 2] - [ 2 15 0 2 4 984 1 18 2 2 1 7 4 11 1 0 1 1 1 7 5] - [ 2 4 12 2 0 0 1078 1 0 1 0 1 1 2 0 2 1 0 2 7 2] - [ 2 4 11 2 4 26 10 948 0 0 2 6 0 1 0 0 1 0 21 9 7] - [ 5 3 0 0 0 2 0 0 982 37 12 2 2 5 8 0 2 0 2 0 2] - [ 55 1 1 0 5 4 0 2 21 892 1 1 0 6 2 1 0 4 0 0 5] - [ 1 2 1 5 1 0 1 3 8 3 970 0 1 8 2 1 0 0 2 3 7] - [ 3 1 1 0 1 16 5 3 2 0 1 958 32 1 0 7 5 5 0 7 3] - [ 0 1 3 2 0 3 0 1 0 1 0 17 917 1 0 9 2 4 0 2 6] - [ 1 0 1 0 2 7 0 2 16 12 5 1 6 957 1 1 2 0 0 4 5] - [ 8 6 1 12 4 0 0 0 11 3 0 2 4 3 1061 0 0 2 5 0 8] - [ 0 0 1 0 2 0 1 0 2 1 0 7 5 2 0 1000 6 10 0 2 4] - [ 3 2 0 1 2 0 1 1 0 1 2 1 0 1 0 11 1034 0 0 4 8] - [ 2 1 1 3 1 1 1 1 0 4 0 5 24 1 2 14 0 971 0 1 3] - [ 2 5 6 10 0 1 0 20 2 1 3 4 3 0 7 1 0 3 936 1 3] - [ 2 5 3 1 0 3 6 7 0 1 3 11 7 4 1 5 5 1 1 1006 8] - [ 115 192 190 119 105 166 77 135 69 77 143 75 341 235 132 135 181 70 137 168 10364]] - -2022-12-06 11:26:24,193 - ==> Best [Top1: 87.890 Top5: 98.479 Sparsity:0.00 Params: 5376 on epoch: 141] -2022-12-06 11:26:24,193 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:26:24,199 - - -2022-12-06 11:26:24,199 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:26:25,231 - Epoch: [143][ 10/ 1200] Overall Loss 0.148356 Objective Loss 0.148356 LR 0.000250 Time 0.103110 -2022-12-06 11:26:25,427 - Epoch: [143][ 20/ 1200] Overall Loss 0.157075 Objective Loss 0.157075 LR 0.000250 Time 0.061301 -2022-12-06 11:26:25,619 - Epoch: [143][ 30/ 1200] Overall Loss 0.162261 Objective Loss 0.162261 LR 0.000250 Time 0.047277 -2022-12-06 11:26:25,812 - Epoch: [143][ 40/ 1200] Overall Loss 0.162901 Objective Loss 0.162901 LR 0.000250 Time 0.040261 -2022-12-06 11:26:26,005 - Epoch: [143][ 50/ 1200] Overall Loss 0.163505 Objective Loss 0.163505 LR 0.000250 Time 0.036054 -2022-12-06 11:26:26,197 - Epoch: [143][ 60/ 1200] Overall Loss 0.163261 Objective Loss 0.163261 LR 0.000250 Time 0.033234 -2022-12-06 11:26:26,390 - Epoch: [143][ 70/ 1200] Overall Loss 0.162015 Objective Loss 0.162015 LR 0.000250 Time 0.031232 -2022-12-06 11:26:26,583 - Epoch: [143][ 80/ 1200] Overall Loss 0.162722 Objective Loss 0.162722 LR 0.000250 Time 0.029732 -2022-12-06 11:26:26,775 - Epoch: [143][ 90/ 1200] Overall Loss 0.165535 Objective Loss 0.165535 LR 0.000250 Time 0.028556 -2022-12-06 11:26:26,967 - Epoch: [143][ 100/ 1200] Overall Loss 0.163907 Objective Loss 0.163907 LR 0.000250 Time 0.027621 -2022-12-06 11:26:27,160 - Epoch: [143][ 110/ 1200] Overall Loss 0.163512 Objective Loss 0.163512 LR 0.000250 Time 0.026859 -2022-12-06 11:26:27,353 - Epoch: [143][ 120/ 1200] Overall Loss 0.162105 Objective Loss 0.162105 LR 0.000250 Time 0.026220 -2022-12-06 11:26:27,545 - Epoch: [143][ 130/ 1200] Overall Loss 0.161261 Objective Loss 0.161261 LR 0.000250 Time 0.025681 -2022-12-06 11:26:27,737 - Epoch: [143][ 140/ 1200] Overall Loss 0.160257 Objective Loss 0.160257 LR 0.000250 Time 0.025215 -2022-12-06 11:26:27,930 - Epoch: [143][ 150/ 1200] Overall Loss 0.159368 Objective Loss 0.159368 LR 0.000250 Time 0.024815 -2022-12-06 11:26:28,123 - Epoch: [143][ 160/ 1200] Overall Loss 0.159833 Objective Loss 0.159833 LR 0.000250 Time 0.024465 -2022-12-06 11:26:28,316 - Epoch: [143][ 170/ 1200] Overall Loss 0.159711 Objective Loss 0.159711 LR 0.000250 Time 0.024159 -2022-12-06 11:26:28,508 - Epoch: [143][ 180/ 1200] Overall Loss 0.159696 Objective Loss 0.159696 LR 0.000250 Time 0.023883 -2022-12-06 11:26:28,701 - Epoch: [143][ 190/ 1200] Overall Loss 0.159704 Objective Loss 0.159704 LR 0.000250 Time 0.023636 -2022-12-06 11:26:28,893 - Epoch: [143][ 200/ 1200] Overall Loss 0.160372 Objective Loss 0.160372 LR 0.000250 Time 0.023410 -2022-12-06 11:26:29,085 - Epoch: [143][ 210/ 1200] Overall Loss 0.160381 Objective Loss 0.160381 LR 0.000250 Time 0.023210 -2022-12-06 11:26:29,277 - Epoch: [143][ 220/ 1200] Overall Loss 0.161030 Objective Loss 0.161030 LR 0.000250 Time 0.023025 -2022-12-06 11:26:29,470 - Epoch: [143][ 230/ 1200] Overall Loss 0.161293 Objective Loss 0.161293 LR 0.000250 Time 0.022860 -2022-12-06 11:26:29,663 - Epoch: [143][ 240/ 1200] Overall Loss 0.160957 Objective Loss 0.160957 LR 0.000250 Time 0.022707 -2022-12-06 11:26:29,855 - Epoch: [143][ 250/ 1200] Overall Loss 0.160678 Objective Loss 0.160678 LR 0.000250 Time 0.022565 -2022-12-06 11:26:30,047 - Epoch: [143][ 260/ 1200] Overall Loss 0.160817 Objective Loss 0.160817 LR 0.000250 Time 0.022435 -2022-12-06 11:26:30,240 - Epoch: [143][ 270/ 1200] Overall Loss 0.160973 Objective Loss 0.160973 LR 0.000250 Time 0.022315 -2022-12-06 11:26:30,432 - Epoch: [143][ 280/ 1200] Overall Loss 0.160828 Objective Loss 0.160828 LR 0.000250 Time 0.022205 -2022-12-06 11:26:30,625 - Epoch: [143][ 290/ 1200] Overall Loss 0.160435 Objective Loss 0.160435 LR 0.000250 Time 0.022100 -2022-12-06 11:26:30,818 - Epoch: [143][ 300/ 1200] Overall Loss 0.160496 Objective Loss 0.160496 LR 0.000250 Time 0.022005 -2022-12-06 11:26:31,010 - Epoch: [143][ 310/ 1200] Overall Loss 0.160460 Objective Loss 0.160460 LR 0.000250 Time 0.021914 -2022-12-06 11:26:31,202 - Epoch: [143][ 320/ 1200] Overall Loss 0.159883 Objective Loss 0.159883 LR 0.000250 Time 0.021828 -2022-12-06 11:26:31,395 - Epoch: [143][ 330/ 1200] Overall Loss 0.159950 Objective Loss 0.159950 LR 0.000250 Time 0.021748 -2022-12-06 11:26:31,587 - Epoch: [143][ 340/ 1200] Overall Loss 0.159610 Objective Loss 0.159610 LR 0.000250 Time 0.021672 -2022-12-06 11:26:31,779 - Epoch: [143][ 350/ 1200] Overall Loss 0.159847 Objective Loss 0.159847 LR 0.000250 Time 0.021601 -2022-12-06 11:26:31,972 - Epoch: [143][ 360/ 1200] Overall Loss 0.159640 Objective Loss 0.159640 LR 0.000250 Time 0.021534 -2022-12-06 11:26:32,164 - Epoch: [143][ 370/ 1200] Overall Loss 0.159314 Objective Loss 0.159314 LR 0.000250 Time 0.021469 -2022-12-06 11:26:32,356 - Epoch: [143][ 380/ 1200] Overall Loss 0.159623 Objective Loss 0.159623 LR 0.000250 Time 0.021409 -2022-12-06 11:26:32,548 - Epoch: [143][ 390/ 1200] Overall Loss 0.160117 Objective Loss 0.160117 LR 0.000250 Time 0.021352 -2022-12-06 11:26:32,741 - Epoch: [143][ 400/ 1200] Overall Loss 0.160234 Objective Loss 0.160234 LR 0.000250 Time 0.021298 -2022-12-06 11:26:32,934 - Epoch: [143][ 410/ 1200] Overall Loss 0.160192 Objective Loss 0.160192 LR 0.000250 Time 0.021248 -2022-12-06 11:26:33,127 - Epoch: [143][ 420/ 1200] Overall Loss 0.160852 Objective Loss 0.160852 LR 0.000250 Time 0.021199 -2022-12-06 11:26:33,319 - Epoch: [143][ 430/ 1200] Overall Loss 0.160777 Objective Loss 0.160777 LR 0.000250 Time 0.021152 -2022-12-06 11:26:33,512 - Epoch: [143][ 440/ 1200] Overall Loss 0.160897 Objective Loss 0.160897 LR 0.000250 Time 0.021108 -2022-12-06 11:26:33,704 - Epoch: [143][ 450/ 1200] Overall Loss 0.160701 Objective Loss 0.160701 LR 0.000250 Time 0.021065 -2022-12-06 11:26:33,896 - Epoch: [143][ 460/ 1200] Overall Loss 0.160637 Objective Loss 0.160637 LR 0.000250 Time 0.021024 -2022-12-06 11:26:34,089 - Epoch: [143][ 470/ 1200] Overall Loss 0.160619 Objective Loss 0.160619 LR 0.000250 Time 0.020986 -2022-12-06 11:26:34,282 - Epoch: [143][ 480/ 1200] Overall Loss 0.160565 Objective Loss 0.160565 LR 0.000250 Time 0.020949 -2022-12-06 11:26:34,474 - Epoch: [143][ 490/ 1200] Overall Loss 0.160439 Objective Loss 0.160439 LR 0.000250 Time 0.020913 -2022-12-06 11:26:34,667 - Epoch: [143][ 500/ 1200] Overall Loss 0.160348 Objective Loss 0.160348 LR 0.000250 Time 0.020880 -2022-12-06 11:26:34,860 - Epoch: [143][ 510/ 1200] Overall Loss 0.160152 Objective Loss 0.160152 LR 0.000250 Time 0.020847 -2022-12-06 11:26:35,053 - Epoch: [143][ 520/ 1200] Overall Loss 0.160313 Objective Loss 0.160313 LR 0.000250 Time 0.020815 -2022-12-06 11:26:35,245 - Epoch: [143][ 530/ 1200] Overall Loss 0.160495 Objective Loss 0.160495 LR 0.000250 Time 0.020785 -2022-12-06 11:26:35,438 - Epoch: [143][ 540/ 1200] Overall Loss 0.160490 Objective Loss 0.160490 LR 0.000250 Time 0.020756 -2022-12-06 11:26:35,630 - Epoch: [143][ 550/ 1200] Overall Loss 0.160425 Objective Loss 0.160425 LR 0.000250 Time 0.020728 -2022-12-06 11:26:35,823 - Epoch: [143][ 560/ 1200] Overall Loss 0.160544 Objective Loss 0.160544 LR 0.000250 Time 0.020701 -2022-12-06 11:26:36,016 - Epoch: [143][ 570/ 1200] Overall Loss 0.160677 Objective Loss 0.160677 LR 0.000250 Time 0.020675 -2022-12-06 11:26:36,209 - Epoch: [143][ 580/ 1200] Overall Loss 0.160480 Objective Loss 0.160480 LR 0.000250 Time 0.020649 -2022-12-06 11:26:36,401 - Epoch: [143][ 590/ 1200] Overall Loss 0.160585 Objective Loss 0.160585 LR 0.000250 Time 0.020624 -2022-12-06 11:26:36,593 - Epoch: [143][ 600/ 1200] Overall Loss 0.160993 Objective Loss 0.160993 LR 0.000250 Time 0.020601 -2022-12-06 11:26:36,785 - Epoch: [143][ 610/ 1200] Overall Loss 0.160988 Objective Loss 0.160988 LR 0.000250 Time 0.020577 -2022-12-06 11:26:36,978 - Epoch: [143][ 620/ 1200] Overall Loss 0.160697 Objective Loss 0.160697 LR 0.000250 Time 0.020554 -2022-12-06 11:26:37,171 - Epoch: [143][ 630/ 1200] Overall Loss 0.160760 Objective Loss 0.160760 LR 0.000250 Time 0.020534 -2022-12-06 11:26:37,366 - Epoch: [143][ 640/ 1200] Overall Loss 0.160784 Objective Loss 0.160784 LR 0.000250 Time 0.020517 -2022-12-06 11:26:37,560 - Epoch: [143][ 650/ 1200] Overall Loss 0.160799 Objective Loss 0.160799 LR 0.000250 Time 0.020499 -2022-12-06 11:26:37,755 - Epoch: [143][ 660/ 1200] Overall Loss 0.161295 Objective Loss 0.161295 LR 0.000250 Time 0.020483 -2022-12-06 11:26:37,949 - Epoch: [143][ 670/ 1200] Overall Loss 0.161195 Objective Loss 0.161195 LR 0.000250 Time 0.020466 -2022-12-06 11:26:38,144 - Epoch: [143][ 680/ 1200] Overall Loss 0.161420 Objective Loss 0.161420 LR 0.000250 Time 0.020450 -2022-12-06 11:26:38,338 - Epoch: [143][ 690/ 1200] Overall Loss 0.160975 Objective Loss 0.160975 LR 0.000250 Time 0.020435 -2022-12-06 11:26:38,532 - Epoch: [143][ 700/ 1200] Overall Loss 0.161073 Objective Loss 0.161073 LR 0.000250 Time 0.020419 -2022-12-06 11:26:38,727 - Epoch: [143][ 710/ 1200] Overall Loss 0.161073 Objective Loss 0.161073 LR 0.000250 Time 0.020405 -2022-12-06 11:26:38,921 - Epoch: [143][ 720/ 1200] Overall Loss 0.160922 Objective Loss 0.160922 LR 0.000250 Time 0.020391 -2022-12-06 11:26:39,115 - Epoch: [143][ 730/ 1200] Overall Loss 0.160743 Objective Loss 0.160743 LR 0.000250 Time 0.020377 -2022-12-06 11:26:39,310 - Epoch: [143][ 740/ 1200] Overall Loss 0.160580 Objective Loss 0.160580 LR 0.000250 Time 0.020363 -2022-12-06 11:26:39,505 - Epoch: [143][ 750/ 1200] Overall Loss 0.160655 Objective Loss 0.160655 LR 0.000250 Time 0.020351 -2022-12-06 11:26:39,699 - Epoch: [143][ 760/ 1200] Overall Loss 0.160890 Objective Loss 0.160890 LR 0.000250 Time 0.020339 -2022-12-06 11:26:39,894 - Epoch: [143][ 770/ 1200] Overall Loss 0.160535 Objective Loss 0.160535 LR 0.000250 Time 0.020326 -2022-12-06 11:26:40,088 - Epoch: [143][ 780/ 1200] Overall Loss 0.160748 Objective Loss 0.160748 LR 0.000250 Time 0.020315 -2022-12-06 11:26:40,283 - Epoch: [143][ 790/ 1200] Overall Loss 0.160667 Objective Loss 0.160667 LR 0.000250 Time 0.020303 -2022-12-06 11:26:40,476 - Epoch: [143][ 800/ 1200] Overall Loss 0.160970 Objective Loss 0.160970 LR 0.000250 Time 0.020290 -2022-12-06 11:26:40,671 - Epoch: [143][ 810/ 1200] Overall Loss 0.160888 Objective Loss 0.160888 LR 0.000250 Time 0.020280 -2022-12-06 11:26:40,866 - Epoch: [143][ 820/ 1200] Overall Loss 0.160913 Objective Loss 0.160913 LR 0.000250 Time 0.020269 -2022-12-06 11:26:41,060 - Epoch: [143][ 830/ 1200] Overall Loss 0.160806 Objective Loss 0.160806 LR 0.000250 Time 0.020258 -2022-12-06 11:26:41,254 - Epoch: [143][ 840/ 1200] Overall Loss 0.160850 Objective Loss 0.160850 LR 0.000250 Time 0.020247 -2022-12-06 11:26:41,448 - Epoch: [143][ 850/ 1200] Overall Loss 0.160980 Objective Loss 0.160980 LR 0.000250 Time 0.020236 -2022-12-06 11:26:41,643 - Epoch: [143][ 860/ 1200] Overall Loss 0.160770 Objective Loss 0.160770 LR 0.000250 Time 0.020227 -2022-12-06 11:26:41,837 - Epoch: [143][ 870/ 1200] Overall Loss 0.160555 Objective Loss 0.160555 LR 0.000250 Time 0.020217 -2022-12-06 11:26:42,032 - Epoch: [143][ 880/ 1200] Overall Loss 0.160685 Objective Loss 0.160685 LR 0.000250 Time 0.020209 -2022-12-06 11:26:42,227 - Epoch: [143][ 890/ 1200] Overall Loss 0.160700 Objective Loss 0.160700 LR 0.000250 Time 0.020200 -2022-12-06 11:26:42,422 - Epoch: [143][ 900/ 1200] Overall Loss 0.160996 Objective Loss 0.160996 LR 0.000250 Time 0.020191 -2022-12-06 11:26:42,615 - Epoch: [143][ 910/ 1200] Overall Loss 0.160883 Objective Loss 0.160883 LR 0.000250 Time 0.020181 -2022-12-06 11:26:42,810 - Epoch: [143][ 920/ 1200] Overall Loss 0.160613 Objective Loss 0.160613 LR 0.000250 Time 0.020173 -2022-12-06 11:26:43,003 - Epoch: [143][ 930/ 1200] Overall Loss 0.160515 Objective Loss 0.160515 LR 0.000250 Time 0.020163 -2022-12-06 11:26:43,198 - Epoch: [143][ 940/ 1200] Overall Loss 0.160577 Objective Loss 0.160577 LR 0.000250 Time 0.020156 -2022-12-06 11:26:43,392 - Epoch: [143][ 950/ 1200] Overall Loss 0.160381 Objective Loss 0.160381 LR 0.000250 Time 0.020147 -2022-12-06 11:26:43,587 - Epoch: [143][ 960/ 1200] Overall Loss 0.160422 Objective Loss 0.160422 LR 0.000250 Time 0.020139 -2022-12-06 11:26:43,781 - Epoch: [143][ 970/ 1200] Overall Loss 0.160252 Objective Loss 0.160252 LR 0.000250 Time 0.020131 -2022-12-06 11:26:43,975 - Epoch: [143][ 980/ 1200] Overall Loss 0.160169 Objective Loss 0.160169 LR 0.000250 Time 0.020123 -2022-12-06 11:26:44,170 - Epoch: [143][ 990/ 1200] Overall Loss 0.160225 Objective Loss 0.160225 LR 0.000250 Time 0.020116 -2022-12-06 11:26:44,364 - Epoch: [143][ 1000/ 1200] Overall Loss 0.160170 Objective Loss 0.160170 LR 0.000250 Time 0.020109 -2022-12-06 11:26:44,559 - Epoch: [143][ 1010/ 1200] Overall Loss 0.160376 Objective Loss 0.160376 LR 0.000250 Time 0.020102 -2022-12-06 11:26:44,753 - Epoch: [143][ 1020/ 1200] Overall Loss 0.160241 Objective Loss 0.160241 LR 0.000250 Time 0.020095 -2022-12-06 11:26:44,948 - Epoch: [143][ 1030/ 1200] Overall Loss 0.160329 Objective Loss 0.160329 LR 0.000250 Time 0.020088 -2022-12-06 11:26:45,142 - Epoch: [143][ 1040/ 1200] Overall Loss 0.160480 Objective Loss 0.160480 LR 0.000250 Time 0.020082 -2022-12-06 11:26:45,338 - Epoch: [143][ 1050/ 1200] Overall Loss 0.160604 Objective Loss 0.160604 LR 0.000250 Time 0.020076 -2022-12-06 11:26:45,531 - Epoch: [143][ 1060/ 1200] Overall Loss 0.160854 Objective Loss 0.160854 LR 0.000250 Time 0.020069 -2022-12-06 11:26:45,725 - Epoch: [143][ 1070/ 1200] Overall Loss 0.160765 Objective Loss 0.160765 LR 0.000250 Time 0.020062 -2022-12-06 11:26:45,920 - Epoch: [143][ 1080/ 1200] Overall Loss 0.160659 Objective Loss 0.160659 LR 0.000250 Time 0.020055 -2022-12-06 11:26:46,114 - Epoch: [143][ 1090/ 1200] Overall Loss 0.160701 Objective Loss 0.160701 LR 0.000250 Time 0.020049 -2022-12-06 11:26:46,309 - Epoch: [143][ 1100/ 1200] Overall Loss 0.160747 Objective Loss 0.160747 LR 0.000250 Time 0.020044 -2022-12-06 11:26:46,503 - Epoch: [143][ 1110/ 1200] Overall Loss 0.160881 Objective Loss 0.160881 LR 0.000250 Time 0.020038 -2022-12-06 11:26:46,698 - Epoch: [143][ 1120/ 1200] Overall Loss 0.160802 Objective Loss 0.160802 LR 0.000250 Time 0.020032 -2022-12-06 11:26:46,892 - Epoch: [143][ 1130/ 1200] Overall Loss 0.160930 Objective Loss 0.160930 LR 0.000250 Time 0.020026 -2022-12-06 11:26:47,087 - Epoch: [143][ 1140/ 1200] Overall Loss 0.160987 Objective Loss 0.160987 LR 0.000250 Time 0.020021 -2022-12-06 11:26:47,281 - Epoch: [143][ 1150/ 1200] Overall Loss 0.161030 Objective Loss 0.161030 LR 0.000250 Time 0.020015 -2022-12-06 11:26:47,476 - Epoch: [143][ 1160/ 1200] Overall Loss 0.160991 Objective Loss 0.160991 LR 0.000250 Time 0.020010 -2022-12-06 11:26:47,670 - Epoch: [143][ 1170/ 1200] Overall Loss 0.161066 Objective Loss 0.161066 LR 0.000250 Time 0.020005 -2022-12-06 11:26:47,864 - Epoch: [143][ 1180/ 1200] Overall Loss 0.160823 Objective Loss 0.160823 LR 0.000250 Time 0.019999 -2022-12-06 11:26:48,058 - Epoch: [143][ 1190/ 1200] Overall Loss 0.160667 Objective Loss 0.160667 LR 0.000250 Time 0.019993 -2022-12-06 11:26:48,282 - Epoch: [143][ 1200/ 1200] Overall Loss 0.160624 Objective Loss 0.160624 Top1 91.631799 Top5 99.790795 LR 0.000250 Time 0.020013 -2022-12-06 11:26:48,371 - --- validate (epoch=143)----------- -2022-12-06 11:26:48,371 - 34129 samples (256 per mini-batch) -2022-12-06 11:26:48,820 - Epoch: [143][ 10/ 134] Loss 0.268636 Top1 87.656250 Top5 98.125000 -2022-12-06 11:26:48,956 - Epoch: [143][ 20/ 134] Loss 0.253175 Top1 87.988281 Top5 98.359375 -2022-12-06 11:26:49,089 - Epoch: [143][ 30/ 134] Loss 0.253413 Top1 87.812500 Top5 98.450521 -2022-12-06 11:26:49,224 - Epoch: [143][ 40/ 134] Loss 0.252521 Top1 87.812500 Top5 98.447266 -2022-12-06 11:26:49,357 - Epoch: [143][ 50/ 134] Loss 0.250344 Top1 87.820312 Top5 98.484375 -2022-12-06 11:26:49,491 - Epoch: [143][ 60/ 134] Loss 0.244369 Top1 87.903646 Top5 98.509115 -2022-12-06 11:26:49,622 - Epoch: [143][ 70/ 134] Loss 0.243522 Top1 87.812500 Top5 98.537946 -2022-12-06 11:26:49,755 - Epoch: [143][ 80/ 134] Loss 0.246476 Top1 87.695312 Top5 98.525391 -2022-12-06 11:26:49,888 - Epoch: [143][ 90/ 134] Loss 0.241344 Top1 87.899306 Top5 98.554688 -2022-12-06 11:26:50,019 - Epoch: [143][ 100/ 134] Loss 0.240430 Top1 87.910156 Top5 98.535156 -2022-12-06 11:26:50,151 - Epoch: [143][ 110/ 134] Loss 0.239887 Top1 87.933239 Top5 98.494318 -2022-12-06 11:26:50,285 - Epoch: [143][ 120/ 134] Loss 0.240136 Top1 87.874349 Top5 98.512370 -2022-12-06 11:26:50,419 - Epoch: [143][ 130/ 134] Loss 0.239086 Top1 87.905649 Top5 98.509615 -2022-12-06 11:26:50,460 - Epoch: [143][ 134/ 134] Loss 0.238199 Top1 87.910575 Top5 98.526180 -2022-12-06 11:26:50,550 - ==> Top1: 87.911 Top5: 98.526 Loss: 0.238 - -2022-12-06 11:26:50,551 - ==> Confusion: -[[ 911 1 1 4 1 6 1 0 6 47 0 2 0 3 5 1 1 1 0 0 5] - [ 1 939 2 2 5 24 2 15 0 1 2 4 3 1 0 1 4 1 12 2 6] - [ 3 3 1019 15 6 2 14 9 0 3 3 2 1 1 4 1 0 1 2 2 12] - [ 1 0 17 958 1 1 1 0 0 2 7 0 4 2 10 0 0 2 9 0 5] - [ 10 4 1 0 962 4 1 1 0 5 2 2 1 2 9 3 6 2 1 0 4] - [ 1 11 0 4 4 994 1 21 2 1 1 7 3 8 1 1 2 0 0 4 3] - [ 0 3 11 4 0 3 1077 2 0 0 0 1 0 2 0 3 1 2 2 6 1] - [ 1 5 7 4 3 25 7 956 1 0 2 3 0 3 1 0 2 1 20 8 5] - [ 5 1 0 0 1 3 0 1 984 40 8 1 1 5 8 0 0 1 1 1 3] - [ 50 0 1 0 4 2 0 3 22 896 1 2 0 9 3 1 0 2 1 0 4] - [ 0 1 3 4 0 1 3 3 7 3 965 0 1 9 1 1 0 0 7 2 8] - [ 3 0 2 0 0 11 4 3 0 0 0 963 31 6 0 7 3 6 0 8 4] - [ 2 1 0 2 0 1 0 0 0 1 0 14 918 1 1 6 1 10 0 5 6] - [ 0 1 1 0 0 11 0 3 11 10 4 3 2 962 1 2 3 0 0 2 7] - [ 6 2 1 5 4 1 0 0 15 4 0 3 2 3 1072 0 0 1 7 0 4] - [ 1 0 1 0 2 1 3 0 0 1 1 4 7 2 0 996 5 11 0 4 4] - [ 1 0 1 1 3 0 1 1 1 1 1 3 2 1 1 11 1024 0 1 4 14] - [ 2 1 1 2 1 1 1 1 0 4 0 5 17 1 4 10 0 981 0 1 3] - [ 2 3 1 12 0 1 0 22 1 1 2 3 4 0 8 0 0 1 942 1 4] - [ 1 4 0 2 0 6 4 5 0 0 2 11 6 8 0 6 3 2 1 1012 7] - [ 106 173 155 105 91 165 74 146 79 88 142 78 314 237 156 106 138 82 149 179 10463]] - -2022-12-06 11:26:51,124 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:26:51,124 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:26:51,131 - - -2022-12-06 11:26:51,131 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:26:52,068 - Epoch: [144][ 10/ 1200] Overall Loss 0.154879 Objective Loss 0.154879 LR 0.000250 Time 0.093605 -2022-12-06 11:26:52,269 - Epoch: [144][ 20/ 1200] Overall Loss 0.144057 Objective Loss 0.144057 LR 0.000250 Time 0.056815 -2022-12-06 11:26:52,468 - Epoch: [144][ 30/ 1200] Overall Loss 0.142523 Objective Loss 0.142523 LR 0.000250 Time 0.044483 -2022-12-06 11:26:52,663 - Epoch: [144][ 40/ 1200] Overall Loss 0.146759 Objective Loss 0.146759 LR 0.000250 Time 0.038225 -2022-12-06 11:26:52,862 - Epoch: [144][ 50/ 1200] Overall Loss 0.148725 Objective Loss 0.148725 LR 0.000250 Time 0.034539 -2022-12-06 11:26:53,057 - Epoch: [144][ 60/ 1200] Overall Loss 0.153246 Objective Loss 0.153246 LR 0.000250 Time 0.032033 -2022-12-06 11:26:53,255 - Epoch: [144][ 70/ 1200] Overall Loss 0.152175 Objective Loss 0.152175 LR 0.000250 Time 0.030280 -2022-12-06 11:26:53,451 - Epoch: [144][ 80/ 1200] Overall Loss 0.153656 Objective Loss 0.153656 LR 0.000250 Time 0.028930 -2022-12-06 11:26:53,649 - Epoch: [144][ 90/ 1200] Overall Loss 0.156858 Objective Loss 0.156858 LR 0.000250 Time 0.027912 -2022-12-06 11:26:53,844 - Epoch: [144][ 100/ 1200] Overall Loss 0.156830 Objective Loss 0.156830 LR 0.000250 Time 0.027065 -2022-12-06 11:26:54,042 - Epoch: [144][ 110/ 1200] Overall Loss 0.156711 Objective Loss 0.156711 LR 0.000250 Time 0.026405 -2022-12-06 11:26:54,238 - Epoch: [144][ 120/ 1200] Overall Loss 0.156546 Objective Loss 0.156546 LR 0.000250 Time 0.025827 -2022-12-06 11:26:54,436 - Epoch: [144][ 130/ 1200] Overall Loss 0.156927 Objective Loss 0.156927 LR 0.000250 Time 0.025365 -2022-12-06 11:26:54,631 - Epoch: [144][ 140/ 1200] Overall Loss 0.157367 Objective Loss 0.157367 LR 0.000250 Time 0.024942 -2022-12-06 11:26:54,830 - Epoch: [144][ 150/ 1200] Overall Loss 0.156353 Objective Loss 0.156353 LR 0.000250 Time 0.024599 -2022-12-06 11:26:55,025 - Epoch: [144][ 160/ 1200] Overall Loss 0.157263 Objective Loss 0.157263 LR 0.000250 Time 0.024280 -2022-12-06 11:26:55,224 - Epoch: [144][ 170/ 1200] Overall Loss 0.155935 Objective Loss 0.155935 LR 0.000250 Time 0.024019 -2022-12-06 11:26:55,420 - Epoch: [144][ 180/ 1200] Overall Loss 0.157024 Objective Loss 0.157024 LR 0.000250 Time 0.023766 -2022-12-06 11:26:55,618 - Epoch: [144][ 190/ 1200] Overall Loss 0.156491 Objective Loss 0.156491 LR 0.000250 Time 0.023557 -2022-12-06 11:26:55,814 - Epoch: [144][ 200/ 1200] Overall Loss 0.156361 Objective Loss 0.156361 LR 0.000250 Time 0.023353 -2022-12-06 11:26:56,012 - Epoch: [144][ 210/ 1200] Overall Loss 0.156049 Objective Loss 0.156049 LR 0.000250 Time 0.023182 -2022-12-06 11:26:56,207 - Epoch: [144][ 220/ 1200] Overall Loss 0.156902 Objective Loss 0.156902 LR 0.000250 Time 0.023015 -2022-12-06 11:26:56,406 - Epoch: [144][ 230/ 1200] Overall Loss 0.156863 Objective Loss 0.156863 LR 0.000250 Time 0.022875 -2022-12-06 11:26:56,600 - Epoch: [144][ 240/ 1200] Overall Loss 0.156601 Objective Loss 0.156601 LR 0.000250 Time 0.022731 -2022-12-06 11:26:56,799 - Epoch: [144][ 250/ 1200] Overall Loss 0.156661 Objective Loss 0.156661 LR 0.000250 Time 0.022614 -2022-12-06 11:26:56,994 - Epoch: [144][ 260/ 1200] Overall Loss 0.157352 Objective Loss 0.157352 LR 0.000250 Time 0.022492 -2022-12-06 11:26:57,193 - Epoch: [144][ 270/ 1200] Overall Loss 0.156754 Objective Loss 0.156754 LR 0.000250 Time 0.022393 -2022-12-06 11:26:57,388 - Epoch: [144][ 280/ 1200] Overall Loss 0.157333 Objective Loss 0.157333 LR 0.000250 Time 0.022287 -2022-12-06 11:26:57,586 - Epoch: [144][ 290/ 1200] Overall Loss 0.157725 Objective Loss 0.157725 LR 0.000250 Time 0.022203 -2022-12-06 11:26:57,782 - Epoch: [144][ 300/ 1200] Overall Loss 0.158067 Objective Loss 0.158067 LR 0.000250 Time 0.022112 -2022-12-06 11:26:57,980 - Epoch: [144][ 310/ 1200] Overall Loss 0.157776 Objective Loss 0.157776 LR 0.000250 Time 0.022036 -2022-12-06 11:26:58,175 - Epoch: [144][ 320/ 1200] Overall Loss 0.157903 Objective Loss 0.157903 LR 0.000250 Time 0.021955 -2022-12-06 11:26:58,374 - Epoch: [144][ 330/ 1200] Overall Loss 0.157341 Objective Loss 0.157341 LR 0.000250 Time 0.021891 -2022-12-06 11:26:58,570 - Epoch: [144][ 340/ 1200] Overall Loss 0.157285 Objective Loss 0.157285 LR 0.000250 Time 0.021822 -2022-12-06 11:26:58,770 - Epoch: [144][ 350/ 1200] Overall Loss 0.157473 Objective Loss 0.157473 LR 0.000250 Time 0.021770 -2022-12-06 11:26:58,968 - Epoch: [144][ 360/ 1200] Overall Loss 0.158147 Objective Loss 0.158147 LR 0.000250 Time 0.021712 -2022-12-06 11:26:59,168 - Epoch: [144][ 370/ 1200] Overall Loss 0.157554 Objective Loss 0.157554 LR 0.000250 Time 0.021665 -2022-12-06 11:26:59,366 - Epoch: [144][ 380/ 1200] Overall Loss 0.157312 Objective Loss 0.157312 LR 0.000250 Time 0.021614 -2022-12-06 11:26:59,567 - Epoch: [144][ 390/ 1200] Overall Loss 0.157297 Objective Loss 0.157297 LR 0.000250 Time 0.021574 -2022-12-06 11:26:59,764 - Epoch: [144][ 400/ 1200] Overall Loss 0.157113 Objective Loss 0.157113 LR 0.000250 Time 0.021526 -2022-12-06 11:26:59,965 - Epoch: [144][ 410/ 1200] Overall Loss 0.156871 Objective Loss 0.156871 LR 0.000250 Time 0.021489 -2022-12-06 11:27:00,162 - Epoch: [144][ 420/ 1200] Overall Loss 0.157203 Objective Loss 0.157203 LR 0.000250 Time 0.021446 -2022-12-06 11:27:00,362 - Epoch: [144][ 430/ 1200] Overall Loss 0.156888 Objective Loss 0.156888 LR 0.000250 Time 0.021411 -2022-12-06 11:27:00,559 - Epoch: [144][ 440/ 1200] Overall Loss 0.157241 Objective Loss 0.157241 LR 0.000250 Time 0.021371 -2022-12-06 11:27:00,760 - Epoch: [144][ 450/ 1200] Overall Loss 0.157924 Objective Loss 0.157924 LR 0.000250 Time 0.021341 -2022-12-06 11:27:00,958 - Epoch: [144][ 460/ 1200] Overall Loss 0.158214 Objective Loss 0.158214 LR 0.000250 Time 0.021306 -2022-12-06 11:27:01,159 - Epoch: [144][ 470/ 1200] Overall Loss 0.158599 Objective Loss 0.158599 LR 0.000250 Time 0.021278 -2022-12-06 11:27:01,356 - Epoch: [144][ 480/ 1200] Overall Loss 0.158384 Objective Loss 0.158384 LR 0.000250 Time 0.021246 -2022-12-06 11:27:01,558 - Epoch: [144][ 490/ 1200] Overall Loss 0.158138 Objective Loss 0.158138 LR 0.000250 Time 0.021222 -2022-12-06 11:27:01,755 - Epoch: [144][ 500/ 1200] Overall Loss 0.157885 Objective Loss 0.157885 LR 0.000250 Time 0.021190 -2022-12-06 11:27:01,955 - Epoch: [144][ 510/ 1200] Overall Loss 0.157510 Objective Loss 0.157510 LR 0.000250 Time 0.021167 -2022-12-06 11:27:02,153 - Epoch: [144][ 520/ 1200] Overall Loss 0.157533 Objective Loss 0.157533 LR 0.000250 Time 0.021140 -2022-12-06 11:27:02,354 - Epoch: [144][ 530/ 1200] Overall Loss 0.157481 Objective Loss 0.157481 LR 0.000250 Time 0.021119 -2022-12-06 11:27:02,551 - Epoch: [144][ 540/ 1200] Overall Loss 0.157636 Objective Loss 0.157636 LR 0.000250 Time 0.021092 -2022-12-06 11:27:02,752 - Epoch: [144][ 550/ 1200] Overall Loss 0.157501 Objective Loss 0.157501 LR 0.000250 Time 0.021073 -2022-12-06 11:27:02,950 - Epoch: [144][ 560/ 1200] Overall Loss 0.157494 Objective Loss 0.157494 LR 0.000250 Time 0.021048 -2022-12-06 11:27:03,151 - Epoch: [144][ 570/ 1200] Overall Loss 0.157299 Objective Loss 0.157299 LR 0.000250 Time 0.021031 -2022-12-06 11:27:03,350 - Epoch: [144][ 580/ 1200] Overall Loss 0.157589 Objective Loss 0.157589 LR 0.000250 Time 0.021010 -2022-12-06 11:27:03,551 - Epoch: [144][ 590/ 1200] Overall Loss 0.157543 Objective Loss 0.157543 LR 0.000250 Time 0.020994 -2022-12-06 11:27:03,749 - Epoch: [144][ 600/ 1200] Overall Loss 0.157932 Objective Loss 0.157932 LR 0.000250 Time 0.020973 -2022-12-06 11:27:03,950 - Epoch: [144][ 610/ 1200] Overall Loss 0.158321 Objective Loss 0.158321 LR 0.000250 Time 0.020958 -2022-12-06 11:27:04,147 - Epoch: [144][ 620/ 1200] Overall Loss 0.158165 Objective Loss 0.158165 LR 0.000250 Time 0.020937 -2022-12-06 11:27:04,348 - Epoch: [144][ 630/ 1200] Overall Loss 0.158261 Objective Loss 0.158261 LR 0.000250 Time 0.020923 -2022-12-06 11:27:04,546 - Epoch: [144][ 640/ 1200] Overall Loss 0.158109 Objective Loss 0.158109 LR 0.000250 Time 0.020904 -2022-12-06 11:27:04,747 - Epoch: [144][ 650/ 1200] Overall Loss 0.158049 Objective Loss 0.158049 LR 0.000250 Time 0.020890 -2022-12-06 11:27:04,945 - Epoch: [144][ 660/ 1200] Overall Loss 0.157889 Objective Loss 0.157889 LR 0.000250 Time 0.020873 -2022-12-06 11:27:05,146 - Epoch: [144][ 670/ 1200] Overall Loss 0.157884 Objective Loss 0.157884 LR 0.000250 Time 0.020861 -2022-12-06 11:27:05,344 - Epoch: [144][ 680/ 1200] Overall Loss 0.157820 Objective Loss 0.157820 LR 0.000250 Time 0.020845 -2022-12-06 11:27:05,545 - Epoch: [144][ 690/ 1200] Overall Loss 0.157798 Objective Loss 0.157798 LR 0.000250 Time 0.020833 -2022-12-06 11:27:05,741 - Epoch: [144][ 700/ 1200] Overall Loss 0.158042 Objective Loss 0.158042 LR 0.000250 Time 0.020815 -2022-12-06 11:27:05,943 - Epoch: [144][ 710/ 1200] Overall Loss 0.158207 Objective Loss 0.158207 LR 0.000250 Time 0.020804 -2022-12-06 11:27:06,140 - Epoch: [144][ 720/ 1200] Overall Loss 0.158261 Objective Loss 0.158261 LR 0.000250 Time 0.020789 -2022-12-06 11:27:06,341 - Epoch: [144][ 730/ 1200] Overall Loss 0.158208 Objective Loss 0.158208 LR 0.000250 Time 0.020778 -2022-12-06 11:27:06,538 - Epoch: [144][ 740/ 1200] Overall Loss 0.158290 Objective Loss 0.158290 LR 0.000250 Time 0.020764 -2022-12-06 11:27:06,740 - Epoch: [144][ 750/ 1200] Overall Loss 0.158356 Objective Loss 0.158356 LR 0.000250 Time 0.020754 -2022-12-06 11:27:06,937 - Epoch: [144][ 760/ 1200] Overall Loss 0.158598 Objective Loss 0.158598 LR 0.000250 Time 0.020741 -2022-12-06 11:27:07,138 - Epoch: [144][ 770/ 1200] Overall Loss 0.158638 Objective Loss 0.158638 LR 0.000250 Time 0.020731 -2022-12-06 11:27:07,336 - Epoch: [144][ 780/ 1200] Overall Loss 0.158937 Objective Loss 0.158937 LR 0.000250 Time 0.020718 -2022-12-06 11:27:07,536 - Epoch: [144][ 790/ 1200] Overall Loss 0.158872 Objective Loss 0.158872 LR 0.000250 Time 0.020709 -2022-12-06 11:27:07,733 - Epoch: [144][ 800/ 1200] Overall Loss 0.158677 Objective Loss 0.158677 LR 0.000250 Time 0.020696 -2022-12-06 11:27:07,934 - Epoch: [144][ 810/ 1200] Overall Loss 0.158857 Objective Loss 0.158857 LR 0.000250 Time 0.020688 -2022-12-06 11:27:08,132 - Epoch: [144][ 820/ 1200] Overall Loss 0.158745 Objective Loss 0.158745 LR 0.000250 Time 0.020677 -2022-12-06 11:27:08,333 - Epoch: [144][ 830/ 1200] Overall Loss 0.159042 Objective Loss 0.159042 LR 0.000250 Time 0.020669 -2022-12-06 11:27:08,531 - Epoch: [144][ 840/ 1200] Overall Loss 0.159006 Objective Loss 0.159006 LR 0.000250 Time 0.020657 -2022-12-06 11:27:08,730 - Epoch: [144][ 850/ 1200] Overall Loss 0.159000 Objective Loss 0.159000 LR 0.000250 Time 0.020648 -2022-12-06 11:27:08,926 - Epoch: [144][ 860/ 1200] Overall Loss 0.158913 Objective Loss 0.158913 LR 0.000250 Time 0.020635 -2022-12-06 11:27:09,124 - Epoch: [144][ 870/ 1200] Overall Loss 0.158990 Objective Loss 0.158990 LR 0.000250 Time 0.020625 -2022-12-06 11:27:09,320 - Epoch: [144][ 880/ 1200] Overall Loss 0.158867 Objective Loss 0.158867 LR 0.000250 Time 0.020612 -2022-12-06 11:27:09,519 - Epoch: [144][ 890/ 1200] Overall Loss 0.158905 Objective Loss 0.158905 LR 0.000250 Time 0.020603 -2022-12-06 11:27:09,714 - Epoch: [144][ 900/ 1200] Overall Loss 0.159065 Objective Loss 0.159065 LR 0.000250 Time 0.020591 -2022-12-06 11:27:09,912 - Epoch: [144][ 910/ 1200] Overall Loss 0.158877 Objective Loss 0.158877 LR 0.000250 Time 0.020582 -2022-12-06 11:27:10,108 - Epoch: [144][ 920/ 1200] Overall Loss 0.159128 Objective Loss 0.159128 LR 0.000250 Time 0.020570 -2022-12-06 11:27:10,306 - Epoch: [144][ 930/ 1200] Overall Loss 0.159278 Objective Loss 0.159278 LR 0.000250 Time 0.020562 -2022-12-06 11:27:10,502 - Epoch: [144][ 940/ 1200] Overall Loss 0.159203 Objective Loss 0.159203 LR 0.000250 Time 0.020551 -2022-12-06 11:27:10,700 - Epoch: [144][ 950/ 1200] Overall Loss 0.159200 Objective Loss 0.159200 LR 0.000250 Time 0.020542 -2022-12-06 11:27:10,896 - Epoch: [144][ 960/ 1200] Overall Loss 0.159289 Objective Loss 0.159289 LR 0.000250 Time 0.020532 -2022-12-06 11:27:11,094 - Epoch: [144][ 970/ 1200] Overall Loss 0.159314 Objective Loss 0.159314 LR 0.000250 Time 0.020524 -2022-12-06 11:27:11,290 - Epoch: [144][ 980/ 1200] Overall Loss 0.159234 Objective Loss 0.159234 LR 0.000250 Time 0.020514 -2022-12-06 11:27:11,488 - Epoch: [144][ 990/ 1200] Overall Loss 0.159258 Objective Loss 0.159258 LR 0.000250 Time 0.020506 -2022-12-06 11:27:11,683 - Epoch: [144][ 1000/ 1200] Overall Loss 0.159274 Objective Loss 0.159274 LR 0.000250 Time 0.020496 -2022-12-06 11:27:11,882 - Epoch: [144][ 1010/ 1200] Overall Loss 0.159274 Objective Loss 0.159274 LR 0.000250 Time 0.020489 -2022-12-06 11:27:12,077 - Epoch: [144][ 1020/ 1200] Overall Loss 0.159332 Objective Loss 0.159332 LR 0.000250 Time 0.020479 -2022-12-06 11:27:12,276 - Epoch: [144][ 1030/ 1200] Overall Loss 0.159377 Objective Loss 0.159377 LR 0.000250 Time 0.020472 -2022-12-06 11:27:12,472 - Epoch: [144][ 1040/ 1200] Overall Loss 0.159407 Objective Loss 0.159407 LR 0.000250 Time 0.020464 -2022-12-06 11:27:12,671 - Epoch: [144][ 1050/ 1200] Overall Loss 0.159312 Objective Loss 0.159312 LR 0.000250 Time 0.020457 -2022-12-06 11:27:12,866 - Epoch: [144][ 1060/ 1200] Overall Loss 0.159219 Objective Loss 0.159219 LR 0.000250 Time 0.020448 -2022-12-06 11:27:13,065 - Epoch: [144][ 1070/ 1200] Overall Loss 0.159165 Objective Loss 0.159165 LR 0.000250 Time 0.020442 -2022-12-06 11:27:13,260 - Epoch: [144][ 1080/ 1200] Overall Loss 0.159068 Objective Loss 0.159068 LR 0.000250 Time 0.020433 -2022-12-06 11:27:13,458 - Epoch: [144][ 1090/ 1200] Overall Loss 0.159062 Objective Loss 0.159062 LR 0.000250 Time 0.020427 -2022-12-06 11:27:13,654 - Epoch: [144][ 1100/ 1200] Overall Loss 0.158925 Objective Loss 0.158925 LR 0.000250 Time 0.020419 -2022-12-06 11:27:13,852 - Epoch: [144][ 1110/ 1200] Overall Loss 0.159006 Objective Loss 0.159006 LR 0.000250 Time 0.020413 -2022-12-06 11:27:14,047 - Epoch: [144][ 1120/ 1200] Overall Loss 0.159032 Objective Loss 0.159032 LR 0.000250 Time 0.020404 -2022-12-06 11:27:14,246 - Epoch: [144][ 1130/ 1200] Overall Loss 0.159034 Objective Loss 0.159034 LR 0.000250 Time 0.020399 -2022-12-06 11:27:14,441 - Epoch: [144][ 1140/ 1200] Overall Loss 0.159333 Objective Loss 0.159333 LR 0.000250 Time 0.020391 -2022-12-06 11:27:14,639 - Epoch: [144][ 1150/ 1200] Overall Loss 0.159354 Objective Loss 0.159354 LR 0.000250 Time 0.020386 -2022-12-06 11:27:14,834 - Epoch: [144][ 1160/ 1200] Overall Loss 0.159456 Objective Loss 0.159456 LR 0.000250 Time 0.020378 -2022-12-06 11:27:15,033 - Epoch: [144][ 1170/ 1200] Overall Loss 0.159498 Objective Loss 0.159498 LR 0.000250 Time 0.020373 -2022-12-06 11:27:15,228 - Epoch: [144][ 1180/ 1200] Overall Loss 0.159687 Objective Loss 0.159687 LR 0.000250 Time 0.020365 -2022-12-06 11:27:15,426 - Epoch: [144][ 1190/ 1200] Overall Loss 0.159700 Objective Loss 0.159700 LR 0.000250 Time 0.020360 -2022-12-06 11:27:15,650 - Epoch: [144][ 1200/ 1200] Overall Loss 0.159550 Objective Loss 0.159550 Top1 91.213389 Top5 99.372385 LR 0.000250 Time 0.020376 -2022-12-06 11:27:15,739 - --- validate (epoch=144)----------- -2022-12-06 11:27:15,739 - 34129 samples (256 per mini-batch) -2022-12-06 11:27:16,302 - Epoch: [144][ 10/ 134] Loss 0.223540 Top1 87.421875 Top5 98.398438 -2022-12-06 11:27:16,432 - Epoch: [144][ 20/ 134] Loss 0.226654 Top1 87.910156 Top5 98.515625 -2022-12-06 11:27:16,562 - Epoch: [144][ 30/ 134] Loss 0.234768 Top1 87.643229 Top5 98.463542 -2022-12-06 11:27:16,692 - Epoch: [144][ 40/ 134] Loss 0.236722 Top1 87.519531 Top5 98.505859 -2022-12-06 11:27:16,822 - Epoch: [144][ 50/ 134] Loss 0.243613 Top1 87.375000 Top5 98.445312 -2022-12-06 11:27:16,949 - Epoch: [144][ 60/ 134] Loss 0.242753 Top1 87.395833 Top5 98.489583 -2022-12-06 11:27:17,079 - Epoch: [144][ 70/ 134] Loss 0.246243 Top1 87.393973 Top5 98.515625 -2022-12-06 11:27:17,205 - Epoch: [144][ 80/ 134] Loss 0.242009 Top1 87.416992 Top5 98.540039 -2022-12-06 11:27:17,332 - Epoch: [144][ 90/ 134] Loss 0.237969 Top1 87.521701 Top5 98.563368 -2022-12-06 11:27:17,461 - Epoch: [144][ 100/ 134] Loss 0.239106 Top1 87.539062 Top5 98.546875 -2022-12-06 11:27:17,591 - Epoch: [144][ 110/ 134] Loss 0.238977 Top1 87.567472 Top5 98.529830 -2022-12-06 11:27:17,720 - Epoch: [144][ 120/ 134] Loss 0.240337 Top1 87.578125 Top5 98.548177 -2022-12-06 11:27:17,851 - Epoch: [144][ 130/ 134] Loss 0.239198 Top1 87.608173 Top5 98.578726 -2022-12-06 11:27:17,889 - Epoch: [144][ 134/ 134] Loss 0.240365 Top1 87.597058 Top5 98.575991 -2022-12-06 11:27:17,975 - ==> Top1: 87.597 Top5: 98.576 Loss: 0.240 - -2022-12-06 11:27:17,976 - ==> Confusion: -[[ 919 1 2 0 5 4 0 0 5 42 0 1 2 3 7 1 0 0 1 0 3] - [ 1 952 3 2 10 12 3 6 2 1 2 4 1 0 0 1 3 0 13 2 9] - [ 3 5 1016 13 5 1 14 7 0 2 5 3 2 2 4 2 1 2 4 5 7] - [ 3 1 20 940 1 1 0 0 0 0 12 0 5 1 10 0 1 2 16 1 6] - [ 7 9 1 1 958 2 1 0 1 6 2 0 1 2 8 7 4 2 2 1 5] - [ 1 17 0 2 6 977 4 18 2 2 1 8 3 12 2 1 1 1 1 6 4] - [ 2 4 6 1 0 0 1081 3 0 0 0 1 1 2 0 3 3 1 0 9 1] - [ 2 9 10 2 2 22 10 942 0 0 2 3 0 1 1 1 0 0 30 11 6] - [ 5 2 0 0 0 1 1 1 981 35 11 1 3 6 13 0 1 0 1 0 2] - [ 54 0 2 0 6 3 0 1 29 884 1 2 0 10 2 1 0 0 0 0 6] - [ 1 2 2 3 1 0 2 3 6 0 969 0 2 10 3 0 0 0 6 3 6] - [ 2 1 2 0 1 12 4 3 1 0 0 966 22 5 1 9 2 6 0 9 5] - [ 0 1 1 4 0 2 0 1 0 0 0 19 913 0 2 8 1 6 0 4 7] - [ 0 1 1 0 1 9 0 2 12 14 4 3 5 954 2 3 2 0 1 2 7] - [ 4 1 1 8 5 1 0 0 9 2 1 2 2 3 1072 0 1 2 9 1 6] - [ 1 0 1 0 1 1 3 0 1 1 0 6 3 3 0 1005 5 6 0 3 3] - [ 3 2 1 0 3 0 1 0 0 0 1 2 2 1 0 12 1028 1 0 7 8] - [ 2 1 1 1 0 1 2 0 0 5 0 5 23 1 1 19 2 968 0 1 3] - [ 2 3 5 5 1 2 0 21 2 1 2 2 4 0 8 0 0 2 943 1 4] - [ 1 4 1 0 1 3 5 4 0 1 4 15 8 4 1 3 2 3 0 1013 7] - [ 100 222 172 98 107 128 82 122 77 74 162 79 306 244 154 113 169 68 157 183 10409]] - -2022-12-06 11:27:18,543 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:27:18,543 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:27:18,548 - - -2022-12-06 11:27:18,549 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:27:19,470 - Epoch: [145][ 10/ 1200] Overall Loss 0.146257 Objective Loss 0.146257 LR 0.000250 Time 0.092044 -2022-12-06 11:27:19,671 - Epoch: [145][ 20/ 1200] Overall Loss 0.154556 Objective Loss 0.154556 LR 0.000250 Time 0.056034 -2022-12-06 11:27:19,863 - Epoch: [145][ 30/ 1200] Overall Loss 0.157654 Objective Loss 0.157654 LR 0.000250 Time 0.043749 -2022-12-06 11:27:20,053 - Epoch: [145][ 40/ 1200] Overall Loss 0.158347 Objective Loss 0.158347 LR 0.000250 Time 0.037555 -2022-12-06 11:27:20,243 - Epoch: [145][ 50/ 1200] Overall Loss 0.158931 Objective Loss 0.158931 LR 0.000250 Time 0.033838 -2022-12-06 11:27:20,434 - Epoch: [145][ 60/ 1200] Overall Loss 0.157567 Objective Loss 0.157567 LR 0.000250 Time 0.031364 -2022-12-06 11:27:20,624 - Epoch: [145][ 70/ 1200] Overall Loss 0.153317 Objective Loss 0.153317 LR 0.000250 Time 0.029590 -2022-12-06 11:27:20,814 - Epoch: [145][ 80/ 1200] Overall Loss 0.152284 Objective Loss 0.152284 LR 0.000250 Time 0.028260 -2022-12-06 11:27:21,004 - Epoch: [145][ 90/ 1200] Overall Loss 0.152231 Objective Loss 0.152231 LR 0.000250 Time 0.027227 -2022-12-06 11:27:21,195 - Epoch: [145][ 100/ 1200] Overall Loss 0.150352 Objective Loss 0.150352 LR 0.000250 Time 0.026406 -2022-12-06 11:27:21,385 - Epoch: [145][ 110/ 1200] Overall Loss 0.151111 Objective Loss 0.151111 LR 0.000250 Time 0.025729 -2022-12-06 11:27:21,576 - Epoch: [145][ 120/ 1200] Overall Loss 0.151749 Objective Loss 0.151749 LR 0.000250 Time 0.025169 -2022-12-06 11:27:21,766 - Epoch: [145][ 130/ 1200] Overall Loss 0.153281 Objective Loss 0.153281 LR 0.000250 Time 0.024695 -2022-12-06 11:27:21,957 - Epoch: [145][ 140/ 1200] Overall Loss 0.153274 Objective Loss 0.153274 LR 0.000250 Time 0.024288 -2022-12-06 11:27:22,147 - Epoch: [145][ 150/ 1200] Overall Loss 0.153103 Objective Loss 0.153103 LR 0.000250 Time 0.023932 -2022-12-06 11:27:22,337 - Epoch: [145][ 160/ 1200] Overall Loss 0.153282 Objective Loss 0.153282 LR 0.000250 Time 0.023623 -2022-12-06 11:27:22,528 - Epoch: [145][ 170/ 1200] Overall Loss 0.153646 Objective Loss 0.153646 LR 0.000250 Time 0.023350 -2022-12-06 11:27:22,718 - Epoch: [145][ 180/ 1200] Overall Loss 0.153086 Objective Loss 0.153086 LR 0.000250 Time 0.023107 -2022-12-06 11:27:22,909 - Epoch: [145][ 190/ 1200] Overall Loss 0.153729 Objective Loss 0.153729 LR 0.000250 Time 0.022891 -2022-12-06 11:27:23,100 - Epoch: [145][ 200/ 1200] Overall Loss 0.154178 Objective Loss 0.154178 LR 0.000250 Time 0.022699 -2022-12-06 11:27:23,290 - Epoch: [145][ 210/ 1200] Overall Loss 0.153656 Objective Loss 0.153656 LR 0.000250 Time 0.022523 -2022-12-06 11:27:23,481 - Epoch: [145][ 220/ 1200] Overall Loss 0.154217 Objective Loss 0.154217 LR 0.000250 Time 0.022365 -2022-12-06 11:27:23,672 - Epoch: [145][ 230/ 1200] Overall Loss 0.154770 Objective Loss 0.154770 LR 0.000250 Time 0.022218 -2022-12-06 11:27:23,862 - Epoch: [145][ 240/ 1200] Overall Loss 0.154511 Objective Loss 0.154511 LR 0.000250 Time 0.022083 -2022-12-06 11:27:24,052 - Epoch: [145][ 250/ 1200] Overall Loss 0.154942 Objective Loss 0.154942 LR 0.000250 Time 0.021958 -2022-12-06 11:27:24,243 - Epoch: [145][ 260/ 1200] Overall Loss 0.155119 Objective Loss 0.155119 LR 0.000250 Time 0.021844 -2022-12-06 11:27:24,433 - Epoch: [145][ 270/ 1200] Overall Loss 0.154883 Objective Loss 0.154883 LR 0.000250 Time 0.021738 -2022-12-06 11:27:24,624 - Epoch: [145][ 280/ 1200] Overall Loss 0.155161 Objective Loss 0.155161 LR 0.000250 Time 0.021641 -2022-12-06 11:27:24,814 - Epoch: [145][ 290/ 1200] Overall Loss 0.154471 Objective Loss 0.154471 LR 0.000250 Time 0.021549 -2022-12-06 11:27:25,005 - Epoch: [145][ 300/ 1200] Overall Loss 0.154230 Objective Loss 0.154230 LR 0.000250 Time 0.021465 -2022-12-06 11:27:25,195 - Epoch: [145][ 310/ 1200] Overall Loss 0.154825 Objective Loss 0.154825 LR 0.000250 Time 0.021384 -2022-12-06 11:27:25,386 - Epoch: [145][ 320/ 1200] Overall Loss 0.155219 Objective Loss 0.155219 LR 0.000250 Time 0.021311 -2022-12-06 11:27:25,576 - Epoch: [145][ 330/ 1200] Overall Loss 0.155711 Objective Loss 0.155711 LR 0.000250 Time 0.021238 -2022-12-06 11:27:25,766 - Epoch: [145][ 340/ 1200] Overall Loss 0.155343 Objective Loss 0.155343 LR 0.000250 Time 0.021172 -2022-12-06 11:27:25,956 - Epoch: [145][ 350/ 1200] Overall Loss 0.155256 Objective Loss 0.155256 LR 0.000250 Time 0.021109 -2022-12-06 11:27:26,148 - Epoch: [145][ 360/ 1200] Overall Loss 0.155083 Objective Loss 0.155083 LR 0.000250 Time 0.021052 -2022-12-06 11:27:26,337 - Epoch: [145][ 370/ 1200] Overall Loss 0.155068 Objective Loss 0.155068 LR 0.000250 Time 0.020995 -2022-12-06 11:27:26,528 - Epoch: [145][ 380/ 1200] Overall Loss 0.155930 Objective Loss 0.155930 LR 0.000250 Time 0.020943 -2022-12-06 11:27:26,719 - Epoch: [145][ 390/ 1200] Overall Loss 0.155815 Objective Loss 0.155815 LR 0.000250 Time 0.020894 -2022-12-06 11:27:26,910 - Epoch: [145][ 400/ 1200] Overall Loss 0.155395 Objective Loss 0.155395 LR 0.000250 Time 0.020847 -2022-12-06 11:27:27,100 - Epoch: [145][ 410/ 1200] Overall Loss 0.155948 Objective Loss 0.155948 LR 0.000250 Time 0.020801 -2022-12-06 11:27:27,290 - Epoch: [145][ 420/ 1200] Overall Loss 0.156458 Objective Loss 0.156458 LR 0.000250 Time 0.020757 -2022-12-06 11:27:27,480 - Epoch: [145][ 430/ 1200] Overall Loss 0.156354 Objective Loss 0.156354 LR 0.000250 Time 0.020714 -2022-12-06 11:27:27,670 - Epoch: [145][ 440/ 1200] Overall Loss 0.155989 Objective Loss 0.155989 LR 0.000250 Time 0.020675 -2022-12-06 11:27:27,860 - Epoch: [145][ 450/ 1200] Overall Loss 0.156473 Objective Loss 0.156473 LR 0.000250 Time 0.020636 -2022-12-06 11:27:28,051 - Epoch: [145][ 460/ 1200] Overall Loss 0.156292 Objective Loss 0.156292 LR 0.000250 Time 0.020601 -2022-12-06 11:27:28,241 - Epoch: [145][ 470/ 1200] Overall Loss 0.156067 Objective Loss 0.156067 LR 0.000250 Time 0.020566 -2022-12-06 11:27:28,432 - Epoch: [145][ 480/ 1200] Overall Loss 0.156374 Objective Loss 0.156374 LR 0.000250 Time 0.020534 -2022-12-06 11:27:28,622 - Epoch: [145][ 490/ 1200] Overall Loss 0.156401 Objective Loss 0.156401 LR 0.000250 Time 0.020503 -2022-12-06 11:27:28,812 - Epoch: [145][ 500/ 1200] Overall Loss 0.156486 Objective Loss 0.156486 LR 0.000250 Time 0.020472 -2022-12-06 11:27:29,002 - Epoch: [145][ 510/ 1200] Overall Loss 0.156653 Objective Loss 0.156653 LR 0.000250 Time 0.020442 -2022-12-06 11:27:29,193 - Epoch: [145][ 520/ 1200] Overall Loss 0.156132 Objective Loss 0.156132 LR 0.000250 Time 0.020413 -2022-12-06 11:27:29,382 - Epoch: [145][ 530/ 1200] Overall Loss 0.156280 Objective Loss 0.156280 LR 0.000250 Time 0.020385 -2022-12-06 11:27:29,572 - Epoch: [145][ 540/ 1200] Overall Loss 0.156796 Objective Loss 0.156796 LR 0.000250 Time 0.020359 -2022-12-06 11:27:29,763 - Epoch: [145][ 550/ 1200] Overall Loss 0.156722 Objective Loss 0.156722 LR 0.000250 Time 0.020333 -2022-12-06 11:27:29,954 - Epoch: [145][ 560/ 1200] Overall Loss 0.156596 Objective Loss 0.156596 LR 0.000250 Time 0.020311 -2022-12-06 11:27:30,144 - Epoch: [145][ 570/ 1200] Overall Loss 0.156713 Objective Loss 0.156713 LR 0.000250 Time 0.020287 -2022-12-06 11:27:30,335 - Epoch: [145][ 580/ 1200] Overall Loss 0.156552 Objective Loss 0.156552 LR 0.000250 Time 0.020265 -2022-12-06 11:27:30,525 - Epoch: [145][ 590/ 1200] Overall Loss 0.156560 Objective Loss 0.156560 LR 0.000250 Time 0.020244 -2022-12-06 11:27:30,716 - Epoch: [145][ 600/ 1200] Overall Loss 0.156466 Objective Loss 0.156466 LR 0.000250 Time 0.020223 -2022-12-06 11:27:30,906 - Epoch: [145][ 610/ 1200] Overall Loss 0.156275 Objective Loss 0.156275 LR 0.000250 Time 0.020203 -2022-12-06 11:27:31,097 - Epoch: [145][ 620/ 1200] Overall Loss 0.156253 Objective Loss 0.156253 LR 0.000250 Time 0.020183 -2022-12-06 11:27:31,287 - Epoch: [145][ 630/ 1200] Overall Loss 0.156541 Objective Loss 0.156541 LR 0.000250 Time 0.020164 -2022-12-06 11:27:31,478 - Epoch: [145][ 640/ 1200] Overall Loss 0.156743 Objective Loss 0.156743 LR 0.000250 Time 0.020146 -2022-12-06 11:27:31,668 - Epoch: [145][ 650/ 1200] Overall Loss 0.156960 Objective Loss 0.156960 LR 0.000250 Time 0.020128 -2022-12-06 11:27:31,858 - Epoch: [145][ 660/ 1200] Overall Loss 0.157248 Objective Loss 0.157248 LR 0.000250 Time 0.020110 -2022-12-06 11:27:32,049 - Epoch: [145][ 670/ 1200] Overall Loss 0.157210 Objective Loss 0.157210 LR 0.000250 Time 0.020094 -2022-12-06 11:27:32,240 - Epoch: [145][ 680/ 1200] Overall Loss 0.157347 Objective Loss 0.157347 LR 0.000250 Time 0.020078 -2022-12-06 11:27:32,429 - Epoch: [145][ 690/ 1200] Overall Loss 0.157215 Objective Loss 0.157215 LR 0.000250 Time 0.020061 -2022-12-06 11:27:32,620 - Epoch: [145][ 700/ 1200] Overall Loss 0.157643 Objective Loss 0.157643 LR 0.000250 Time 0.020046 -2022-12-06 11:27:32,810 - Epoch: [145][ 710/ 1200] Overall Loss 0.157806 Objective Loss 0.157806 LR 0.000250 Time 0.020030 -2022-12-06 11:27:33,000 - Epoch: [145][ 720/ 1200] Overall Loss 0.157758 Objective Loss 0.157758 LR 0.000250 Time 0.020016 -2022-12-06 11:27:33,190 - Epoch: [145][ 730/ 1200] Overall Loss 0.157572 Objective Loss 0.157572 LR 0.000250 Time 0.020001 -2022-12-06 11:27:33,380 - Epoch: [145][ 740/ 1200] Overall Loss 0.157554 Objective Loss 0.157554 LR 0.000250 Time 0.019987 -2022-12-06 11:27:33,570 - Epoch: [145][ 750/ 1200] Overall Loss 0.157283 Objective Loss 0.157283 LR 0.000250 Time 0.019972 -2022-12-06 11:27:33,760 - Epoch: [145][ 760/ 1200] Overall Loss 0.157072 Objective Loss 0.157072 LR 0.000250 Time 0.019959 -2022-12-06 11:27:33,951 - Epoch: [145][ 770/ 1200] Overall Loss 0.157064 Objective Loss 0.157064 LR 0.000250 Time 0.019947 -2022-12-06 11:27:34,141 - Epoch: [145][ 780/ 1200] Overall Loss 0.156905 Objective Loss 0.156905 LR 0.000250 Time 0.019935 -2022-12-06 11:27:34,331 - Epoch: [145][ 790/ 1200] Overall Loss 0.156847 Objective Loss 0.156847 LR 0.000250 Time 0.019922 -2022-12-06 11:27:34,521 - Epoch: [145][ 800/ 1200] Overall Loss 0.156821 Objective Loss 0.156821 LR 0.000250 Time 0.019910 -2022-12-06 11:27:34,711 - Epoch: [145][ 810/ 1200] Overall Loss 0.156776 Objective Loss 0.156776 LR 0.000250 Time 0.019898 -2022-12-06 11:27:34,902 - Epoch: [145][ 820/ 1200] Overall Loss 0.156847 Objective Loss 0.156847 LR 0.000250 Time 0.019887 -2022-12-06 11:27:35,092 - Epoch: [145][ 830/ 1200] Overall Loss 0.156982 Objective Loss 0.156982 LR 0.000250 Time 0.019876 -2022-12-06 11:27:35,283 - Epoch: [145][ 840/ 1200] Overall Loss 0.156762 Objective Loss 0.156762 LR 0.000250 Time 0.019866 -2022-12-06 11:27:35,473 - Epoch: [145][ 850/ 1200] Overall Loss 0.156902 Objective Loss 0.156902 LR 0.000250 Time 0.019856 -2022-12-06 11:27:35,664 - Epoch: [145][ 860/ 1200] Overall Loss 0.156844 Objective Loss 0.156844 LR 0.000250 Time 0.019845 -2022-12-06 11:27:35,854 - Epoch: [145][ 870/ 1200] Overall Loss 0.156962 Objective Loss 0.156962 LR 0.000250 Time 0.019835 -2022-12-06 11:27:36,044 - Epoch: [145][ 880/ 1200] Overall Loss 0.157112 Objective Loss 0.157112 LR 0.000250 Time 0.019826 -2022-12-06 11:27:36,234 - Epoch: [145][ 890/ 1200] Overall Loss 0.156883 Objective Loss 0.156883 LR 0.000250 Time 0.019816 -2022-12-06 11:27:36,425 - Epoch: [145][ 900/ 1200] Overall Loss 0.156818 Objective Loss 0.156818 LR 0.000250 Time 0.019807 -2022-12-06 11:27:36,615 - Epoch: [145][ 910/ 1200] Overall Loss 0.156962 Objective Loss 0.156962 LR 0.000250 Time 0.019798 -2022-12-06 11:27:36,806 - Epoch: [145][ 920/ 1200] Overall Loss 0.156940 Objective Loss 0.156940 LR 0.000250 Time 0.019789 -2022-12-06 11:27:36,996 - Epoch: [145][ 930/ 1200] Overall Loss 0.156877 Objective Loss 0.156877 LR 0.000250 Time 0.019780 -2022-12-06 11:27:37,187 - Epoch: [145][ 940/ 1200] Overall Loss 0.156976 Objective Loss 0.156976 LR 0.000250 Time 0.019772 -2022-12-06 11:27:37,377 - Epoch: [145][ 950/ 1200] Overall Loss 0.157185 Objective Loss 0.157185 LR 0.000250 Time 0.019764 -2022-12-06 11:27:37,568 - Epoch: [145][ 960/ 1200] Overall Loss 0.157229 Objective Loss 0.157229 LR 0.000250 Time 0.019756 -2022-12-06 11:27:37,759 - Epoch: [145][ 970/ 1200] Overall Loss 0.157270 Objective Loss 0.157270 LR 0.000250 Time 0.019749 -2022-12-06 11:27:37,949 - Epoch: [145][ 980/ 1200] Overall Loss 0.157146 Objective Loss 0.157146 LR 0.000250 Time 0.019741 -2022-12-06 11:27:38,139 - Epoch: [145][ 990/ 1200] Overall Loss 0.157454 Objective Loss 0.157454 LR 0.000250 Time 0.019733 -2022-12-06 11:27:38,330 - Epoch: [145][ 1000/ 1200] Overall Loss 0.157460 Objective Loss 0.157460 LR 0.000250 Time 0.019726 -2022-12-06 11:27:38,520 - Epoch: [145][ 1010/ 1200] Overall Loss 0.157515 Objective Loss 0.157515 LR 0.000250 Time 0.019719 -2022-12-06 11:27:38,712 - Epoch: [145][ 1020/ 1200] Overall Loss 0.157413 Objective Loss 0.157413 LR 0.000250 Time 0.019712 -2022-12-06 11:27:38,902 - Epoch: [145][ 1030/ 1200] Overall Loss 0.157229 Objective Loss 0.157229 LR 0.000250 Time 0.019705 -2022-12-06 11:27:39,092 - Epoch: [145][ 1040/ 1200] Overall Loss 0.157245 Objective Loss 0.157245 LR 0.000250 Time 0.019698 -2022-12-06 11:27:39,283 - Epoch: [145][ 1050/ 1200] Overall Loss 0.157182 Objective Loss 0.157182 LR 0.000250 Time 0.019691 -2022-12-06 11:27:39,474 - Epoch: [145][ 1060/ 1200] Overall Loss 0.157183 Objective Loss 0.157183 LR 0.000250 Time 0.019685 -2022-12-06 11:27:39,664 - Epoch: [145][ 1070/ 1200] Overall Loss 0.157139 Objective Loss 0.157139 LR 0.000250 Time 0.019678 -2022-12-06 11:27:39,854 - Epoch: [145][ 1080/ 1200] Overall Loss 0.157294 Objective Loss 0.157294 LR 0.000250 Time 0.019672 -2022-12-06 11:27:40,045 - Epoch: [145][ 1090/ 1200] Overall Loss 0.157163 Objective Loss 0.157163 LR 0.000250 Time 0.019666 -2022-12-06 11:27:40,235 - Epoch: [145][ 1100/ 1200] Overall Loss 0.157034 Objective Loss 0.157034 LR 0.000250 Time 0.019660 -2022-12-06 11:27:40,426 - Epoch: [145][ 1110/ 1200] Overall Loss 0.156995 Objective Loss 0.156995 LR 0.000250 Time 0.019653 -2022-12-06 11:27:40,616 - Epoch: [145][ 1120/ 1200] Overall Loss 0.157162 Objective Loss 0.157162 LR 0.000250 Time 0.019647 -2022-12-06 11:27:40,806 - Epoch: [145][ 1130/ 1200] Overall Loss 0.157286 Objective Loss 0.157286 LR 0.000250 Time 0.019642 -2022-12-06 11:27:40,997 - Epoch: [145][ 1140/ 1200] Overall Loss 0.157240 Objective Loss 0.157240 LR 0.000250 Time 0.019636 -2022-12-06 11:27:41,188 - Epoch: [145][ 1150/ 1200] Overall Loss 0.157398 Objective Loss 0.157398 LR 0.000250 Time 0.019631 -2022-12-06 11:27:41,378 - Epoch: [145][ 1160/ 1200] Overall Loss 0.157287 Objective Loss 0.157287 LR 0.000250 Time 0.019625 -2022-12-06 11:27:41,569 - Epoch: [145][ 1170/ 1200] Overall Loss 0.157266 Objective Loss 0.157266 LR 0.000250 Time 0.019620 -2022-12-06 11:27:41,760 - Epoch: [145][ 1180/ 1200] Overall Loss 0.157429 Objective Loss 0.157429 LR 0.000250 Time 0.019615 -2022-12-06 11:27:41,950 - Epoch: [145][ 1190/ 1200] Overall Loss 0.157417 Objective Loss 0.157417 LR 0.000250 Time 0.019610 -2022-12-06 11:27:42,176 - Epoch: [145][ 1200/ 1200] Overall Loss 0.157384 Objective Loss 0.157384 Top1 88.075314 Top5 99.372385 LR 0.000250 Time 0.019634 -2022-12-06 11:27:42,264 - --- validate (epoch=145)----------- -2022-12-06 11:27:42,265 - 34129 samples (256 per mini-batch) -2022-12-06 11:27:42,708 - Epoch: [145][ 10/ 134] Loss 0.247664 Top1 87.929688 Top5 98.515625 -2022-12-06 11:27:42,839 - Epoch: [145][ 20/ 134] Loss 0.245516 Top1 87.988281 Top5 98.535156 -2022-12-06 11:27:42,971 - Epoch: [145][ 30/ 134] Loss 0.232518 Top1 87.903646 Top5 98.632812 -2022-12-06 11:27:43,101 - Epoch: [145][ 40/ 134] Loss 0.231962 Top1 87.958984 Top5 98.710938 -2022-12-06 11:27:43,232 - Epoch: [145][ 50/ 134] Loss 0.229376 Top1 88.054688 Top5 98.671875 -2022-12-06 11:27:43,364 - Epoch: [145][ 60/ 134] Loss 0.233100 Top1 87.903646 Top5 98.561198 -2022-12-06 11:27:43,496 - Epoch: [145][ 70/ 134] Loss 0.231209 Top1 87.929688 Top5 98.565848 -2022-12-06 11:27:43,627 - Epoch: [145][ 80/ 134] Loss 0.232317 Top1 87.709961 Top5 98.525391 -2022-12-06 11:27:43,759 - Epoch: [145][ 90/ 134] Loss 0.229651 Top1 87.855903 Top5 98.532986 -2022-12-06 11:27:43,888 - Epoch: [145][ 100/ 134] Loss 0.232005 Top1 87.792969 Top5 98.500000 -2022-12-06 11:27:44,023 - Epoch: [145][ 110/ 134] Loss 0.232913 Top1 87.737926 Top5 98.462358 -2022-12-06 11:27:44,154 - Epoch: [145][ 120/ 134] Loss 0.235610 Top1 87.682292 Top5 98.453776 -2022-12-06 11:27:44,287 - Epoch: [145][ 130/ 134] Loss 0.235915 Top1 87.677284 Top5 98.428486 -2022-12-06 11:27:44,326 - Epoch: [145][ 134/ 134] Loss 0.237168 Top1 87.682030 Top5 98.417768 -2022-12-06 11:27:44,413 - ==> Top1: 87.682 Top5: 98.418 Loss: 0.237 - -2022-12-06 11:27:44,414 - ==> Confusion: -[[ 931 0 2 2 3 6 0 1 3 39 0 1 0 3 2 1 1 0 0 0 1] - [ 1 945 3 2 9 19 2 14 2 0 4 3 0 0 0 0 4 2 9 2 6] - [ 7 2 1013 15 4 2 14 10 0 3 5 2 1 1 2 1 1 1 3 4 12] - [ 3 0 13 954 1 2 0 1 0 2 10 0 3 2 9 0 1 3 12 0 4] - [ 12 5 2 0 958 1 0 3 1 6 1 2 0 3 8 6 5 2 0 1 4] - [ 2 13 0 3 5 990 1 16 3 1 0 11 3 11 0 1 0 2 1 4 2] - [ 1 5 7 2 0 0 1075 5 0 0 0 0 0 3 0 6 0 2 2 7 3] - [ 3 5 5 2 2 25 11 973 0 0 1 4 1 2 1 1 0 0 12 2 4] - [ 6 2 0 1 0 1 0 0 997 32 6 1 1 5 5 0 2 0 1 2 2] - [ 63 0 1 0 2 3 0 2 29 885 1 2 0 3 2 1 0 1 1 0 5] - [ 1 1 5 1 0 0 1 2 10 4 971 1 0 9 2 1 0 0 4 1 5] - [ 2 0 1 0 2 9 4 6 1 1 1 976 19 3 1 7 2 4 0 10 2] - [ 1 0 1 4 0 1 1 1 0 1 0 30 900 2 0 8 2 5 0 5 7] - [ 1 0 0 0 1 7 0 2 17 12 4 2 2 966 0 0 1 0 0 3 5] - [ 9 4 1 5 4 0 0 0 22 1 2 4 2 4 1056 0 0 2 8 0 6] - [ 0 0 3 0 2 0 2 0 2 1 0 8 7 1 0 993 8 9 0 4 3] - [ 3 2 1 0 3 0 1 0 2 0 0 3 1 2 1 10 1030 0 0 6 7] - [ 4 0 1 0 0 1 0 0 0 5 0 11 11 1 1 11 0 985 0 2 3] - [ 4 2 1 7 1 2 0 26 3 1 2 1 3 0 6 0 0 1 946 1 1] - [ 4 4 1 0 0 4 5 6 2 1 2 12 6 6 0 2 4 3 0 1015 3] - [ 147 210 135 103 88 133 78 168 97 85 141 94 292 258 129 100 161 76 140 229 10362]] - -2022-12-06 11:27:45,074 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:27:45,074 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:27:45,080 - - -2022-12-06 11:27:45,080 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:27:46,009 - Epoch: [146][ 10/ 1200] Overall Loss 0.154074 Objective Loss 0.154074 LR 0.000250 Time 0.092814 -2022-12-06 11:27:46,210 - Epoch: [146][ 20/ 1200] Overall Loss 0.167154 Objective Loss 0.167154 LR 0.000250 Time 0.056405 -2022-12-06 11:27:46,402 - Epoch: [146][ 30/ 1200] Overall Loss 0.160967 Objective Loss 0.160967 LR 0.000250 Time 0.043992 -2022-12-06 11:27:46,595 - Epoch: [146][ 40/ 1200] Overall Loss 0.163853 Objective Loss 0.163853 LR 0.000250 Time 0.037819 -2022-12-06 11:27:46,788 - Epoch: [146][ 50/ 1200] Overall Loss 0.162758 Objective Loss 0.162758 LR 0.000250 Time 0.034103 -2022-12-06 11:27:46,981 - Epoch: [146][ 60/ 1200] Overall Loss 0.162423 Objective Loss 0.162423 LR 0.000250 Time 0.031625 -2022-12-06 11:27:47,174 - Epoch: [146][ 70/ 1200] Overall Loss 0.158696 Objective Loss 0.158696 LR 0.000250 Time 0.029849 -2022-12-06 11:27:47,366 - Epoch: [146][ 80/ 1200] Overall Loss 0.155811 Objective Loss 0.155811 LR 0.000250 Time 0.028521 -2022-12-06 11:27:47,558 - Epoch: [146][ 90/ 1200] Overall Loss 0.155639 Objective Loss 0.155639 LR 0.000250 Time 0.027470 -2022-12-06 11:27:47,748 - Epoch: [146][ 100/ 1200] Overall Loss 0.155322 Objective Loss 0.155322 LR 0.000250 Time 0.026620 -2022-12-06 11:27:47,938 - Epoch: [146][ 110/ 1200] Overall Loss 0.153736 Objective Loss 0.153736 LR 0.000250 Time 0.025928 -2022-12-06 11:27:48,128 - Epoch: [146][ 120/ 1200] Overall Loss 0.153931 Objective Loss 0.153931 LR 0.000250 Time 0.025344 -2022-12-06 11:27:48,319 - Epoch: [146][ 130/ 1200] Overall Loss 0.153224 Objective Loss 0.153224 LR 0.000250 Time 0.024857 -2022-12-06 11:27:48,510 - Epoch: [146][ 140/ 1200] Overall Loss 0.153640 Objective Loss 0.153640 LR 0.000250 Time 0.024442 -2022-12-06 11:27:48,701 - Epoch: [146][ 150/ 1200] Overall Loss 0.153264 Objective Loss 0.153264 LR 0.000250 Time 0.024081 -2022-12-06 11:27:48,891 - Epoch: [146][ 160/ 1200] Overall Loss 0.154131 Objective Loss 0.154131 LR 0.000250 Time 0.023763 -2022-12-06 11:27:49,082 - Epoch: [146][ 170/ 1200] Overall Loss 0.154820 Objective Loss 0.154820 LR 0.000250 Time 0.023485 -2022-12-06 11:27:49,272 - Epoch: [146][ 180/ 1200] Overall Loss 0.154113 Objective Loss 0.154113 LR 0.000250 Time 0.023232 -2022-12-06 11:27:49,463 - Epoch: [146][ 190/ 1200] Overall Loss 0.154529 Objective Loss 0.154529 LR 0.000250 Time 0.023012 -2022-12-06 11:27:49,654 - Epoch: [146][ 200/ 1200] Overall Loss 0.155261 Objective Loss 0.155261 LR 0.000250 Time 0.022812 -2022-12-06 11:27:49,845 - Epoch: [146][ 210/ 1200] Overall Loss 0.155616 Objective Loss 0.155616 LR 0.000250 Time 0.022633 -2022-12-06 11:27:50,035 - Epoch: [146][ 220/ 1200] Overall Loss 0.155181 Objective Loss 0.155181 LR 0.000250 Time 0.022468 -2022-12-06 11:27:50,226 - Epoch: [146][ 230/ 1200] Overall Loss 0.155746 Objective Loss 0.155746 LR 0.000250 Time 0.022316 -2022-12-06 11:27:50,416 - Epoch: [146][ 240/ 1200] Overall Loss 0.155714 Objective Loss 0.155714 LR 0.000250 Time 0.022175 -2022-12-06 11:27:50,607 - Epoch: [146][ 250/ 1200] Overall Loss 0.155677 Objective Loss 0.155677 LR 0.000250 Time 0.022050 -2022-12-06 11:27:50,798 - Epoch: [146][ 260/ 1200] Overall Loss 0.155549 Objective Loss 0.155549 LR 0.000250 Time 0.021935 -2022-12-06 11:27:50,988 - Epoch: [146][ 270/ 1200] Overall Loss 0.154875 Objective Loss 0.154875 LR 0.000250 Time 0.021825 -2022-12-06 11:27:51,178 - Epoch: [146][ 280/ 1200] Overall Loss 0.154760 Objective Loss 0.154760 LR 0.000250 Time 0.021724 -2022-12-06 11:27:51,368 - Epoch: [146][ 290/ 1200] Overall Loss 0.154927 Objective Loss 0.154927 LR 0.000250 Time 0.021627 -2022-12-06 11:27:51,558 - Epoch: [146][ 300/ 1200] Overall Loss 0.154787 Objective Loss 0.154787 LR 0.000250 Time 0.021539 -2022-12-06 11:27:51,749 - Epoch: [146][ 310/ 1200] Overall Loss 0.154423 Objective Loss 0.154423 LR 0.000250 Time 0.021458 -2022-12-06 11:27:51,940 - Epoch: [146][ 320/ 1200] Overall Loss 0.153799 Objective Loss 0.153799 LR 0.000250 Time 0.021383 -2022-12-06 11:27:52,130 - Epoch: [146][ 330/ 1200] Overall Loss 0.153754 Objective Loss 0.153754 LR 0.000250 Time 0.021309 -2022-12-06 11:27:52,320 - Epoch: [146][ 340/ 1200] Overall Loss 0.154096 Objective Loss 0.154096 LR 0.000250 Time 0.021239 -2022-12-06 11:27:52,512 - Epoch: [146][ 350/ 1200] Overall Loss 0.154232 Objective Loss 0.154232 LR 0.000250 Time 0.021178 -2022-12-06 11:27:52,702 - Epoch: [146][ 360/ 1200] Overall Loss 0.154396 Objective Loss 0.154396 LR 0.000250 Time 0.021117 -2022-12-06 11:27:52,893 - Epoch: [146][ 370/ 1200] Overall Loss 0.154485 Objective Loss 0.154485 LR 0.000250 Time 0.021060 -2022-12-06 11:27:53,083 - Epoch: [146][ 380/ 1200] Overall Loss 0.154827 Objective Loss 0.154827 LR 0.000250 Time 0.021005 -2022-12-06 11:27:53,274 - Epoch: [146][ 390/ 1200] Overall Loss 0.154689 Objective Loss 0.154689 LR 0.000250 Time 0.020954 -2022-12-06 11:27:53,465 - Epoch: [146][ 400/ 1200] Overall Loss 0.154607 Objective Loss 0.154607 LR 0.000250 Time 0.020907 -2022-12-06 11:27:53,656 - Epoch: [146][ 410/ 1200] Overall Loss 0.154737 Objective Loss 0.154737 LR 0.000250 Time 0.020862 -2022-12-06 11:27:53,847 - Epoch: [146][ 420/ 1200] Overall Loss 0.154736 Objective Loss 0.154736 LR 0.000250 Time 0.020819 -2022-12-06 11:27:54,038 - Epoch: [146][ 430/ 1200] Overall Loss 0.155191 Objective Loss 0.155191 LR 0.000250 Time 0.020778 -2022-12-06 11:27:54,229 - Epoch: [146][ 440/ 1200] Overall Loss 0.155484 Objective Loss 0.155484 LR 0.000250 Time 0.020738 -2022-12-06 11:27:54,420 - Epoch: [146][ 450/ 1200] Overall Loss 0.155161 Objective Loss 0.155161 LR 0.000250 Time 0.020700 -2022-12-06 11:27:54,610 - Epoch: [146][ 460/ 1200] Overall Loss 0.154989 Objective Loss 0.154989 LR 0.000250 Time 0.020663 -2022-12-06 11:27:54,801 - Epoch: [146][ 470/ 1200] Overall Loss 0.154866 Objective Loss 0.154866 LR 0.000250 Time 0.020628 -2022-12-06 11:27:54,993 - Epoch: [146][ 480/ 1200] Overall Loss 0.154966 Objective Loss 0.154966 LR 0.000250 Time 0.020596 -2022-12-06 11:27:55,184 - Epoch: [146][ 490/ 1200] Overall Loss 0.155057 Objective Loss 0.155057 LR 0.000250 Time 0.020565 -2022-12-06 11:27:55,375 - Epoch: [146][ 500/ 1200] Overall Loss 0.155089 Objective Loss 0.155089 LR 0.000250 Time 0.020534 -2022-12-06 11:27:55,566 - Epoch: [146][ 510/ 1200] Overall Loss 0.155029 Objective Loss 0.155029 LR 0.000250 Time 0.020505 -2022-12-06 11:27:55,756 - Epoch: [146][ 520/ 1200] Overall Loss 0.155183 Objective Loss 0.155183 LR 0.000250 Time 0.020476 -2022-12-06 11:27:55,948 - Epoch: [146][ 530/ 1200] Overall Loss 0.155153 Objective Loss 0.155153 LR 0.000250 Time 0.020449 -2022-12-06 11:27:56,138 - Epoch: [146][ 540/ 1200] Overall Loss 0.155065 Objective Loss 0.155065 LR 0.000250 Time 0.020423 -2022-12-06 11:27:56,329 - Epoch: [146][ 550/ 1200] Overall Loss 0.155405 Objective Loss 0.155405 LR 0.000250 Time 0.020398 -2022-12-06 11:27:56,520 - Epoch: [146][ 560/ 1200] Overall Loss 0.155411 Objective Loss 0.155411 LR 0.000250 Time 0.020373 -2022-12-06 11:27:56,710 - Epoch: [146][ 570/ 1200] Overall Loss 0.155547 Objective Loss 0.155547 LR 0.000250 Time 0.020348 -2022-12-06 11:27:56,901 - Epoch: [146][ 580/ 1200] Overall Loss 0.155870 Objective Loss 0.155870 LR 0.000250 Time 0.020326 -2022-12-06 11:27:57,092 - Epoch: [146][ 590/ 1200] Overall Loss 0.155694 Objective Loss 0.155694 LR 0.000250 Time 0.020304 -2022-12-06 11:27:57,283 - Epoch: [146][ 600/ 1200] Overall Loss 0.155371 Objective Loss 0.155371 LR 0.000250 Time 0.020283 -2022-12-06 11:27:57,474 - Epoch: [146][ 610/ 1200] Overall Loss 0.155677 Objective Loss 0.155677 LR 0.000250 Time 0.020262 -2022-12-06 11:27:57,664 - Epoch: [146][ 620/ 1200] Overall Loss 0.155831 Objective Loss 0.155831 LR 0.000250 Time 0.020242 -2022-12-06 11:27:57,855 - Epoch: [146][ 630/ 1200] Overall Loss 0.155809 Objective Loss 0.155809 LR 0.000250 Time 0.020223 -2022-12-06 11:27:58,046 - Epoch: [146][ 640/ 1200] Overall Loss 0.156113 Objective Loss 0.156113 LR 0.000250 Time 0.020204 -2022-12-06 11:27:58,237 - Epoch: [146][ 650/ 1200] Overall Loss 0.156142 Objective Loss 0.156142 LR 0.000250 Time 0.020186 -2022-12-06 11:27:58,428 - Epoch: [146][ 660/ 1200] Overall Loss 0.155986 Objective Loss 0.155986 LR 0.000250 Time 0.020169 -2022-12-06 11:27:58,618 - Epoch: [146][ 670/ 1200] Overall Loss 0.156226 Objective Loss 0.156226 LR 0.000250 Time 0.020151 -2022-12-06 11:27:58,809 - Epoch: [146][ 680/ 1200] Overall Loss 0.156062 Objective Loss 0.156062 LR 0.000250 Time 0.020135 -2022-12-06 11:27:59,000 - Epoch: [146][ 690/ 1200] Overall Loss 0.155863 Objective Loss 0.155863 LR 0.000250 Time 0.020119 -2022-12-06 11:27:59,191 - Epoch: [146][ 700/ 1200] Overall Loss 0.155818 Objective Loss 0.155818 LR 0.000250 Time 0.020103 -2022-12-06 11:27:59,382 - Epoch: [146][ 710/ 1200] Overall Loss 0.155683 Objective Loss 0.155683 LR 0.000250 Time 0.020088 -2022-12-06 11:27:59,572 - Epoch: [146][ 720/ 1200] Overall Loss 0.155553 Objective Loss 0.155553 LR 0.000250 Time 0.020073 -2022-12-06 11:27:59,763 - Epoch: [146][ 730/ 1200] Overall Loss 0.155481 Objective Loss 0.155481 LR 0.000250 Time 0.020059 -2022-12-06 11:27:59,954 - Epoch: [146][ 740/ 1200] Overall Loss 0.155856 Objective Loss 0.155856 LR 0.000250 Time 0.020045 -2022-12-06 11:28:00,144 - Epoch: [146][ 750/ 1200] Overall Loss 0.155866 Objective Loss 0.155866 LR 0.000250 Time 0.020030 -2022-12-06 11:28:00,335 - Epoch: [146][ 760/ 1200] Overall Loss 0.155909 Objective Loss 0.155909 LR 0.000250 Time 0.020017 -2022-12-06 11:28:00,526 - Epoch: [146][ 770/ 1200] Overall Loss 0.155887 Objective Loss 0.155887 LR 0.000250 Time 0.020005 -2022-12-06 11:28:00,717 - Epoch: [146][ 780/ 1200] Overall Loss 0.155815 Objective Loss 0.155815 LR 0.000250 Time 0.019992 -2022-12-06 11:28:00,907 - Epoch: [146][ 790/ 1200] Overall Loss 0.155686 Objective Loss 0.155686 LR 0.000250 Time 0.019980 -2022-12-06 11:28:01,098 - Epoch: [146][ 800/ 1200] Overall Loss 0.155701 Objective Loss 0.155701 LR 0.000250 Time 0.019967 -2022-12-06 11:28:01,289 - Epoch: [146][ 810/ 1200] Overall Loss 0.156071 Objective Loss 0.156071 LR 0.000250 Time 0.019956 -2022-12-06 11:28:01,480 - Epoch: [146][ 820/ 1200] Overall Loss 0.155984 Objective Loss 0.155984 LR 0.000250 Time 0.019944 -2022-12-06 11:28:01,670 - Epoch: [146][ 830/ 1200] Overall Loss 0.155945 Objective Loss 0.155945 LR 0.000250 Time 0.019933 -2022-12-06 11:28:01,860 - Epoch: [146][ 840/ 1200] Overall Loss 0.155956 Objective Loss 0.155956 LR 0.000250 Time 0.019921 -2022-12-06 11:28:02,051 - Epoch: [146][ 850/ 1200] Overall Loss 0.155933 Objective Loss 0.155933 LR 0.000250 Time 0.019911 -2022-12-06 11:28:02,242 - Epoch: [146][ 860/ 1200] Overall Loss 0.155853 Objective Loss 0.155853 LR 0.000250 Time 0.019901 -2022-12-06 11:28:02,432 - Epoch: [146][ 870/ 1200] Overall Loss 0.155926 Objective Loss 0.155926 LR 0.000250 Time 0.019890 -2022-12-06 11:28:02,623 - Epoch: [146][ 880/ 1200] Overall Loss 0.155863 Objective Loss 0.155863 LR 0.000250 Time 0.019880 -2022-12-06 11:28:02,814 - Epoch: [146][ 890/ 1200] Overall Loss 0.155836 Objective Loss 0.155836 LR 0.000250 Time 0.019870 -2022-12-06 11:28:03,004 - Epoch: [146][ 900/ 1200] Overall Loss 0.155747 Objective Loss 0.155747 LR 0.000250 Time 0.019860 -2022-12-06 11:28:03,194 - Epoch: [146][ 910/ 1200] Overall Loss 0.155967 Objective Loss 0.155967 LR 0.000250 Time 0.019851 -2022-12-06 11:28:03,385 - Epoch: [146][ 920/ 1200] Overall Loss 0.155952 Objective Loss 0.155952 LR 0.000250 Time 0.019841 -2022-12-06 11:28:03,575 - Epoch: [146][ 930/ 1200] Overall Loss 0.156137 Objective Loss 0.156137 LR 0.000250 Time 0.019832 -2022-12-06 11:28:03,765 - Epoch: [146][ 940/ 1200] Overall Loss 0.156231 Objective Loss 0.156231 LR 0.000250 Time 0.019823 -2022-12-06 11:28:03,957 - Epoch: [146][ 950/ 1200] Overall Loss 0.156114 Objective Loss 0.156114 LR 0.000250 Time 0.019815 -2022-12-06 11:28:04,147 - Epoch: [146][ 960/ 1200] Overall Loss 0.156078 Objective Loss 0.156078 LR 0.000250 Time 0.019807 -2022-12-06 11:28:04,338 - Epoch: [146][ 970/ 1200] Overall Loss 0.156203 Objective Loss 0.156203 LR 0.000250 Time 0.019799 -2022-12-06 11:28:04,530 - Epoch: [146][ 980/ 1200] Overall Loss 0.156285 Objective Loss 0.156285 LR 0.000250 Time 0.019791 -2022-12-06 11:28:04,720 - Epoch: [146][ 990/ 1200] Overall Loss 0.156166 Objective Loss 0.156166 LR 0.000250 Time 0.019784 -2022-12-06 11:28:04,912 - Epoch: [146][ 1000/ 1200] Overall Loss 0.156171 Objective Loss 0.156171 LR 0.000250 Time 0.019777 -2022-12-06 11:28:05,102 - Epoch: [146][ 1010/ 1200] Overall Loss 0.156458 Objective Loss 0.156458 LR 0.000250 Time 0.019769 -2022-12-06 11:28:05,293 - Epoch: [146][ 1020/ 1200] Overall Loss 0.156450 Objective Loss 0.156450 LR 0.000250 Time 0.019762 -2022-12-06 11:28:05,484 - Epoch: [146][ 1030/ 1200] Overall Loss 0.156375 Objective Loss 0.156375 LR 0.000250 Time 0.019754 -2022-12-06 11:28:05,676 - Epoch: [146][ 1040/ 1200] Overall Loss 0.156247 Objective Loss 0.156247 LR 0.000250 Time 0.019748 -2022-12-06 11:28:05,867 - Epoch: [146][ 1050/ 1200] Overall Loss 0.156325 Objective Loss 0.156325 LR 0.000250 Time 0.019742 -2022-12-06 11:28:06,058 - Epoch: [146][ 1060/ 1200] Overall Loss 0.156394 Objective Loss 0.156394 LR 0.000250 Time 0.019736 -2022-12-06 11:28:06,250 - Epoch: [146][ 1070/ 1200] Overall Loss 0.156430 Objective Loss 0.156430 LR 0.000250 Time 0.019730 -2022-12-06 11:28:06,441 - Epoch: [146][ 1080/ 1200] Overall Loss 0.156577 Objective Loss 0.156577 LR 0.000250 Time 0.019724 -2022-12-06 11:28:06,631 - Epoch: [146][ 1090/ 1200] Overall Loss 0.156471 Objective Loss 0.156471 LR 0.000250 Time 0.019717 -2022-12-06 11:28:06,823 - Epoch: [146][ 1100/ 1200] Overall Loss 0.156541 Objective Loss 0.156541 LR 0.000250 Time 0.019711 -2022-12-06 11:28:07,014 - Epoch: [146][ 1110/ 1200] Overall Loss 0.156511 Objective Loss 0.156511 LR 0.000250 Time 0.019705 -2022-12-06 11:28:07,204 - Epoch: [146][ 1120/ 1200] Overall Loss 0.156398 Objective Loss 0.156398 LR 0.000250 Time 0.019699 -2022-12-06 11:28:07,396 - Epoch: [146][ 1130/ 1200] Overall Loss 0.156528 Objective Loss 0.156528 LR 0.000250 Time 0.019693 -2022-12-06 11:28:07,587 - Epoch: [146][ 1140/ 1200] Overall Loss 0.156482 Objective Loss 0.156482 LR 0.000250 Time 0.019688 -2022-12-06 11:28:07,778 - Epoch: [146][ 1150/ 1200] Overall Loss 0.156445 Objective Loss 0.156445 LR 0.000250 Time 0.019682 -2022-12-06 11:28:07,969 - Epoch: [146][ 1160/ 1200] Overall Loss 0.156369 Objective Loss 0.156369 LR 0.000250 Time 0.019677 -2022-12-06 11:28:08,160 - Epoch: [146][ 1170/ 1200] Overall Loss 0.156373 Objective Loss 0.156373 LR 0.000250 Time 0.019671 -2022-12-06 11:28:08,351 - Epoch: [146][ 1180/ 1200] Overall Loss 0.156374 Objective Loss 0.156374 LR 0.000250 Time 0.019666 -2022-12-06 11:28:08,541 - Epoch: [146][ 1190/ 1200] Overall Loss 0.156297 Objective Loss 0.156297 LR 0.000250 Time 0.019660 -2022-12-06 11:28:08,771 - Epoch: [146][ 1200/ 1200] Overall Loss 0.156259 Objective Loss 0.156259 Top1 89.958159 Top5 99.163180 LR 0.000250 Time 0.019687 -2022-12-06 11:28:08,862 - --- validate (epoch=146)----------- -2022-12-06 11:28:08,862 - 34129 samples (256 per mini-batch) -2022-12-06 11:28:09,310 - Epoch: [146][ 10/ 134] Loss 0.214563 Top1 88.515625 Top5 98.437500 -2022-12-06 11:28:09,446 - Epoch: [146][ 20/ 134] Loss 0.225916 Top1 88.281250 Top5 98.593750 -2022-12-06 11:28:09,579 - Epoch: [146][ 30/ 134] Loss 0.231529 Top1 88.072917 Top5 98.541667 -2022-12-06 11:28:09,713 - Epoch: [146][ 40/ 134] Loss 0.230977 Top1 87.919922 Top5 98.623047 -2022-12-06 11:28:09,844 - Epoch: [146][ 50/ 134] Loss 0.234623 Top1 87.890625 Top5 98.554688 -2022-12-06 11:28:09,978 - Epoch: [146][ 60/ 134] Loss 0.234966 Top1 87.832031 Top5 98.548177 -2022-12-06 11:28:10,110 - Epoch: [146][ 70/ 134] Loss 0.234909 Top1 87.952009 Top5 98.543527 -2022-12-06 11:28:10,243 - Epoch: [146][ 80/ 134] Loss 0.235295 Top1 87.905273 Top5 98.500977 -2022-12-06 11:28:10,375 - Epoch: [146][ 90/ 134] Loss 0.235613 Top1 87.821181 Top5 98.502604 -2022-12-06 11:28:10,509 - Epoch: [146][ 100/ 134] Loss 0.236703 Top1 87.785156 Top5 98.457031 -2022-12-06 11:28:10,642 - Epoch: [146][ 110/ 134] Loss 0.237025 Top1 87.837358 Top5 98.469460 -2022-12-06 11:28:10,776 - Epoch: [146][ 120/ 134] Loss 0.236592 Top1 87.825521 Top5 98.486328 -2022-12-06 11:28:10,913 - Epoch: [146][ 130/ 134] Loss 0.237857 Top1 87.827524 Top5 98.494591 -2022-12-06 11:28:10,952 - Epoch: [146][ 134/ 134] Loss 0.237703 Top1 87.825603 Top5 98.508600 -2022-12-06 11:28:11,042 - ==> Top1: 87.826 Top5: 98.509 Loss: 0.238 - -2022-12-06 11:28:11,043 - ==> Confusion: -[[ 915 0 2 1 3 7 0 0 5 47 0 3 0 2 3 2 1 0 0 0 5] - [ 1 935 1 2 8 21 6 11 3 1 3 7 1 1 0 1 4 1 12 2 6] - [ 5 1 1015 12 4 3 18 7 0 4 5 4 1 0 2 1 1 2 2 4 12] - [ 2 0 18 947 1 2 1 1 0 2 6 0 4 1 10 0 0 4 13 1 7] - [ 10 4 2 0 960 2 0 1 2 6 2 3 0 2 9 5 6 2 0 1 3] - [ 1 13 0 1 5 991 2 18 3 2 1 10 4 10 0 1 0 0 1 5 1] - [ 1 1 9 1 0 0 1076 3 1 0 0 1 1 2 0 5 1 3 2 8 3] - [ 1 2 11 1 2 26 12 950 0 0 4 4 2 0 0 0 0 0 19 12 8] - [ 7 1 0 0 0 3 1 0 985 35 12 1 0 4 6 1 0 0 1 2 5] - [ 53 0 2 0 1 3 0 4 30 887 1 2 0 7 3 1 0 2 0 0 5] - [ 0 1 3 6 0 0 2 3 10 3 965 1 0 7 1 0 0 0 7 2 8] - [ 2 0 0 0 0 11 5 5 0 0 0 973 28 4 0 5 2 5 0 8 3] - [ 0 0 3 0 0 1 0 1 1 1 2 27 906 1 1 7 0 6 0 4 8] - [ 1 0 1 0 1 7 0 3 13 7 4 3 4 963 1 1 2 0 0 3 9] - [ 9 3 1 7 2 2 0 0 13 1 0 3 2 6 1065 0 0 1 7 1 7] - [ 1 0 2 0 1 0 3 1 0 0 2 11 4 1 0 992 6 11 0 6 2] - [ 3 2 0 1 1 0 2 1 1 0 0 5 4 3 0 8 1030 0 0 5 6] - [ 3 0 1 2 0 1 0 2 0 3 0 9 17 1 1 7 1 985 0 1 2] - [ 3 4 4 9 0 2 0 20 1 1 3 2 3 0 6 0 0 1 946 1 2] - [ 0 2 1 2 1 4 5 3 1 0 4 11 8 7 0 2 4 3 1 1016 5] - [ 112 177 143 97 84 156 72 128 90 82 150 91 311 248 124 80 136 82 168 229 10466]] - -2022-12-06 11:28:11,707 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:28:11,707 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:28:11,713 - - -2022-12-06 11:28:11,713 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:28:12,646 - Epoch: [147][ 10/ 1200] Overall Loss 0.149750 Objective Loss 0.149750 LR 0.000250 Time 0.093243 -2022-12-06 11:28:12,848 - Epoch: [147][ 20/ 1200] Overall Loss 0.151405 Objective Loss 0.151405 LR 0.000250 Time 0.056674 -2022-12-06 11:28:13,038 - Epoch: [147][ 30/ 1200] Overall Loss 0.154491 Objective Loss 0.154491 LR 0.000250 Time 0.044101 -2022-12-06 11:28:13,228 - Epoch: [147][ 40/ 1200] Overall Loss 0.153539 Objective Loss 0.153539 LR 0.000250 Time 0.037815 -2022-12-06 11:28:13,418 - Epoch: [147][ 50/ 1200] Overall Loss 0.152854 Objective Loss 0.152854 LR 0.000250 Time 0.034033 -2022-12-06 11:28:13,607 - Epoch: [147][ 60/ 1200] Overall Loss 0.154513 Objective Loss 0.154513 LR 0.000250 Time 0.031516 -2022-12-06 11:28:13,799 - Epoch: [147][ 70/ 1200] Overall Loss 0.154584 Objective Loss 0.154584 LR 0.000250 Time 0.029740 -2022-12-06 11:28:13,989 - Epoch: [147][ 80/ 1200] Overall Loss 0.157955 Objective Loss 0.157955 LR 0.000250 Time 0.028388 -2022-12-06 11:28:14,179 - Epoch: [147][ 90/ 1200] Overall Loss 0.157570 Objective Loss 0.157570 LR 0.000250 Time 0.027342 -2022-12-06 11:28:14,369 - Epoch: [147][ 100/ 1200] Overall Loss 0.153643 Objective Loss 0.153643 LR 0.000250 Time 0.026503 -2022-12-06 11:28:14,559 - Epoch: [147][ 110/ 1200] Overall Loss 0.153366 Objective Loss 0.153366 LR 0.000250 Time 0.025817 -2022-12-06 11:28:14,749 - Epoch: [147][ 120/ 1200] Overall Loss 0.153501 Objective Loss 0.153501 LR 0.000250 Time 0.025242 -2022-12-06 11:28:14,938 - Epoch: [147][ 130/ 1200] Overall Loss 0.155059 Objective Loss 0.155059 LR 0.000250 Time 0.024754 -2022-12-06 11:28:15,128 - Epoch: [147][ 140/ 1200] Overall Loss 0.155403 Objective Loss 0.155403 LR 0.000250 Time 0.024338 -2022-12-06 11:28:15,319 - Epoch: [147][ 150/ 1200] Overall Loss 0.155513 Objective Loss 0.155513 LR 0.000250 Time 0.023983 -2022-12-06 11:28:15,508 - Epoch: [147][ 160/ 1200] Overall Loss 0.156198 Objective Loss 0.156198 LR 0.000250 Time 0.023663 -2022-12-06 11:28:15,699 - Epoch: [147][ 170/ 1200] Overall Loss 0.157395 Objective Loss 0.157395 LR 0.000250 Time 0.023389 -2022-12-06 11:28:15,888 - Epoch: [147][ 180/ 1200] Overall Loss 0.157279 Objective Loss 0.157279 LR 0.000250 Time 0.023140 -2022-12-06 11:28:16,079 - Epoch: [147][ 190/ 1200] Overall Loss 0.156693 Objective Loss 0.156693 LR 0.000250 Time 0.022922 -2022-12-06 11:28:16,268 - Epoch: [147][ 200/ 1200] Overall Loss 0.157056 Objective Loss 0.157056 LR 0.000250 Time 0.022721 -2022-12-06 11:28:16,458 - Epoch: [147][ 210/ 1200] Overall Loss 0.157750 Objective Loss 0.157750 LR 0.000250 Time 0.022541 -2022-12-06 11:28:16,648 - Epoch: [147][ 220/ 1200] Overall Loss 0.157462 Objective Loss 0.157462 LR 0.000250 Time 0.022378 -2022-12-06 11:28:16,839 - Epoch: [147][ 230/ 1200] Overall Loss 0.156270 Objective Loss 0.156270 LR 0.000250 Time 0.022232 -2022-12-06 11:28:17,028 - Epoch: [147][ 240/ 1200] Overall Loss 0.155649 Objective Loss 0.155649 LR 0.000250 Time 0.022093 -2022-12-06 11:28:17,217 - Epoch: [147][ 250/ 1200] Overall Loss 0.155107 Objective Loss 0.155107 LR 0.000250 Time 0.021963 -2022-12-06 11:28:17,407 - Epoch: [147][ 260/ 1200] Overall Loss 0.155088 Objective Loss 0.155088 LR 0.000250 Time 0.021847 -2022-12-06 11:28:17,597 - Epoch: [147][ 270/ 1200] Overall Loss 0.154826 Objective Loss 0.154826 LR 0.000250 Time 0.021739 -2022-12-06 11:28:17,787 - Epoch: [147][ 280/ 1200] Overall Loss 0.154605 Objective Loss 0.154605 LR 0.000250 Time 0.021637 -2022-12-06 11:28:17,975 - Epoch: [147][ 290/ 1200] Overall Loss 0.154710 Objective Loss 0.154710 LR 0.000250 Time 0.021540 -2022-12-06 11:28:18,165 - Epoch: [147][ 300/ 1200] Overall Loss 0.154969 Objective Loss 0.154969 LR 0.000250 Time 0.021452 -2022-12-06 11:28:18,355 - Epoch: [147][ 310/ 1200] Overall Loss 0.155204 Objective Loss 0.155204 LR 0.000250 Time 0.021371 -2022-12-06 11:28:18,544 - Epoch: [147][ 320/ 1200] Overall Loss 0.154815 Objective Loss 0.154815 LR 0.000250 Time 0.021293 -2022-12-06 11:28:18,735 - Epoch: [147][ 330/ 1200] Overall Loss 0.154534 Objective Loss 0.154534 LR 0.000250 Time 0.021224 -2022-12-06 11:28:18,925 - Epoch: [147][ 340/ 1200] Overall Loss 0.154797 Objective Loss 0.154797 LR 0.000250 Time 0.021155 -2022-12-06 11:28:19,114 - Epoch: [147][ 350/ 1200] Overall Loss 0.154398 Objective Loss 0.154398 LR 0.000250 Time 0.021091 -2022-12-06 11:28:19,304 - Epoch: [147][ 360/ 1200] Overall Loss 0.154916 Objective Loss 0.154916 LR 0.000250 Time 0.021031 -2022-12-06 11:28:19,494 - Epoch: [147][ 370/ 1200] Overall Loss 0.154448 Objective Loss 0.154448 LR 0.000250 Time 0.020973 -2022-12-06 11:28:19,683 - Epoch: [147][ 380/ 1200] Overall Loss 0.154069 Objective Loss 0.154069 LR 0.000250 Time 0.020918 -2022-12-06 11:28:19,873 - Epoch: [147][ 390/ 1200] Overall Loss 0.153408 Objective Loss 0.153408 LR 0.000250 Time 0.020867 -2022-12-06 11:28:20,062 - Epoch: [147][ 400/ 1200] Overall Loss 0.153518 Objective Loss 0.153518 LR 0.000250 Time 0.020818 -2022-12-06 11:28:20,252 - Epoch: [147][ 410/ 1200] Overall Loss 0.153762 Objective Loss 0.153762 LR 0.000250 Time 0.020772 -2022-12-06 11:28:20,442 - Epoch: [147][ 420/ 1200] Overall Loss 0.154053 Objective Loss 0.154053 LR 0.000250 Time 0.020727 -2022-12-06 11:28:20,632 - Epoch: [147][ 430/ 1200] Overall Loss 0.154206 Objective Loss 0.154206 LR 0.000250 Time 0.020687 -2022-12-06 11:28:20,821 - Epoch: [147][ 440/ 1200] Overall Loss 0.154303 Objective Loss 0.154303 LR 0.000250 Time 0.020645 -2022-12-06 11:28:21,011 - Epoch: [147][ 450/ 1200] Overall Loss 0.153949 Objective Loss 0.153949 LR 0.000250 Time 0.020607 -2022-12-06 11:28:21,201 - Epoch: [147][ 460/ 1200] Overall Loss 0.154142 Objective Loss 0.154142 LR 0.000250 Time 0.020570 -2022-12-06 11:28:21,391 - Epoch: [147][ 470/ 1200] Overall Loss 0.154016 Objective Loss 0.154016 LR 0.000250 Time 0.020536 -2022-12-06 11:28:21,581 - Epoch: [147][ 480/ 1200] Overall Loss 0.153976 Objective Loss 0.153976 LR 0.000250 Time 0.020502 -2022-12-06 11:28:21,771 - Epoch: [147][ 490/ 1200] Overall Loss 0.153967 Objective Loss 0.153967 LR 0.000250 Time 0.020471 -2022-12-06 11:28:21,960 - Epoch: [147][ 500/ 1200] Overall Loss 0.154102 Objective Loss 0.154102 LR 0.000250 Time 0.020440 -2022-12-06 11:28:22,150 - Epoch: [147][ 510/ 1200] Overall Loss 0.154149 Objective Loss 0.154149 LR 0.000250 Time 0.020410 -2022-12-06 11:28:22,340 - Epoch: [147][ 520/ 1200] Overall Loss 0.154220 Objective Loss 0.154220 LR 0.000250 Time 0.020382 -2022-12-06 11:28:22,531 - Epoch: [147][ 530/ 1200] Overall Loss 0.154284 Objective Loss 0.154284 LR 0.000250 Time 0.020356 -2022-12-06 11:28:22,720 - Epoch: [147][ 540/ 1200] Overall Loss 0.154372 Objective Loss 0.154372 LR 0.000250 Time 0.020329 -2022-12-06 11:28:22,910 - Epoch: [147][ 550/ 1200] Overall Loss 0.154701 Objective Loss 0.154701 LR 0.000250 Time 0.020304 -2022-12-06 11:28:23,100 - Epoch: [147][ 560/ 1200] Overall Loss 0.155075 Objective Loss 0.155075 LR 0.000250 Time 0.020280 -2022-12-06 11:28:23,290 - Epoch: [147][ 570/ 1200] Overall Loss 0.155100 Objective Loss 0.155100 LR 0.000250 Time 0.020255 -2022-12-06 11:28:23,480 - Epoch: [147][ 580/ 1200] Overall Loss 0.155060 Objective Loss 0.155060 LR 0.000250 Time 0.020233 -2022-12-06 11:28:23,670 - Epoch: [147][ 590/ 1200] Overall Loss 0.155071 Objective Loss 0.155071 LR 0.000250 Time 0.020211 -2022-12-06 11:28:23,860 - Epoch: [147][ 600/ 1200] Overall Loss 0.154837 Objective Loss 0.154837 LR 0.000250 Time 0.020190 -2022-12-06 11:28:24,050 - Epoch: [147][ 610/ 1200] Overall Loss 0.154597 Objective Loss 0.154597 LR 0.000250 Time 0.020170 -2022-12-06 11:28:24,240 - Epoch: [147][ 620/ 1200] Overall Loss 0.154425 Objective Loss 0.154425 LR 0.000250 Time 0.020150 -2022-12-06 11:28:24,430 - Epoch: [147][ 630/ 1200] Overall Loss 0.154177 Objective Loss 0.154177 LR 0.000250 Time 0.020131 -2022-12-06 11:28:24,620 - Epoch: [147][ 640/ 1200] Overall Loss 0.154098 Objective Loss 0.154098 LR 0.000250 Time 0.020112 -2022-12-06 11:28:24,810 - Epoch: [147][ 650/ 1200] Overall Loss 0.154137 Objective Loss 0.154137 LR 0.000250 Time 0.020094 -2022-12-06 11:28:25,000 - Epoch: [147][ 660/ 1200] Overall Loss 0.153985 Objective Loss 0.153985 LR 0.000250 Time 0.020077 -2022-12-06 11:28:25,190 - Epoch: [147][ 670/ 1200] Overall Loss 0.154096 Objective Loss 0.154096 LR 0.000250 Time 0.020059 -2022-12-06 11:28:25,379 - Epoch: [147][ 680/ 1200] Overall Loss 0.154168 Objective Loss 0.154168 LR 0.000250 Time 0.020042 -2022-12-06 11:28:25,569 - Epoch: [147][ 690/ 1200] Overall Loss 0.154256 Objective Loss 0.154256 LR 0.000250 Time 0.020027 -2022-12-06 11:28:25,759 - Epoch: [147][ 700/ 1200] Overall Loss 0.153983 Objective Loss 0.153983 LR 0.000250 Time 0.020011 -2022-12-06 11:28:25,950 - Epoch: [147][ 710/ 1200] Overall Loss 0.154089 Objective Loss 0.154089 LR 0.000250 Time 0.019997 -2022-12-06 11:28:26,140 - Epoch: [147][ 720/ 1200] Overall Loss 0.153978 Objective Loss 0.153978 LR 0.000250 Time 0.019982 -2022-12-06 11:28:26,330 - Epoch: [147][ 730/ 1200] Overall Loss 0.153860 Objective Loss 0.153860 LR 0.000250 Time 0.019968 -2022-12-06 11:28:26,519 - Epoch: [147][ 740/ 1200] Overall Loss 0.153904 Objective Loss 0.153904 LR 0.000250 Time 0.019954 -2022-12-06 11:28:26,709 - Epoch: [147][ 750/ 1200] Overall Loss 0.153858 Objective Loss 0.153858 LR 0.000250 Time 0.019940 -2022-12-06 11:28:26,899 - Epoch: [147][ 760/ 1200] Overall Loss 0.153721 Objective Loss 0.153721 LR 0.000250 Time 0.019927 -2022-12-06 11:28:27,089 - Epoch: [147][ 770/ 1200] Overall Loss 0.154008 Objective Loss 0.154008 LR 0.000250 Time 0.019915 -2022-12-06 11:28:27,279 - Epoch: [147][ 780/ 1200] Overall Loss 0.153981 Objective Loss 0.153981 LR 0.000250 Time 0.019901 -2022-12-06 11:28:27,469 - Epoch: [147][ 790/ 1200] Overall Loss 0.154066 Objective Loss 0.154066 LR 0.000250 Time 0.019889 -2022-12-06 11:28:27,658 - Epoch: [147][ 800/ 1200] Overall Loss 0.154166 Objective Loss 0.154166 LR 0.000250 Time 0.019876 -2022-12-06 11:28:27,848 - Epoch: [147][ 810/ 1200] Overall Loss 0.154161 Objective Loss 0.154161 LR 0.000250 Time 0.019865 -2022-12-06 11:28:28,038 - Epoch: [147][ 820/ 1200] Overall Loss 0.154282 Objective Loss 0.154282 LR 0.000250 Time 0.019853 -2022-12-06 11:28:28,229 - Epoch: [147][ 830/ 1200] Overall Loss 0.154302 Objective Loss 0.154302 LR 0.000250 Time 0.019844 -2022-12-06 11:28:28,418 - Epoch: [147][ 840/ 1200] Overall Loss 0.154391 Objective Loss 0.154391 LR 0.000250 Time 0.019832 -2022-12-06 11:28:28,609 - Epoch: [147][ 850/ 1200] Overall Loss 0.154471 Objective Loss 0.154471 LR 0.000250 Time 0.019822 -2022-12-06 11:28:28,798 - Epoch: [147][ 860/ 1200] Overall Loss 0.154658 Objective Loss 0.154658 LR 0.000250 Time 0.019812 -2022-12-06 11:28:28,989 - Epoch: [147][ 870/ 1200] Overall Loss 0.154604 Objective Loss 0.154604 LR 0.000250 Time 0.019802 -2022-12-06 11:28:29,178 - Epoch: [147][ 880/ 1200] Overall Loss 0.154785 Objective Loss 0.154785 LR 0.000250 Time 0.019792 -2022-12-06 11:28:29,369 - Epoch: [147][ 890/ 1200] Overall Loss 0.154773 Objective Loss 0.154773 LR 0.000250 Time 0.019783 -2022-12-06 11:28:29,559 - Epoch: [147][ 900/ 1200] Overall Loss 0.154913 Objective Loss 0.154913 LR 0.000250 Time 0.019774 -2022-12-06 11:28:29,750 - Epoch: [147][ 910/ 1200] Overall Loss 0.154911 Objective Loss 0.154911 LR 0.000250 Time 0.019766 -2022-12-06 11:28:29,939 - Epoch: [147][ 920/ 1200] Overall Loss 0.154910 Objective Loss 0.154910 LR 0.000250 Time 0.019756 -2022-12-06 11:28:30,130 - Epoch: [147][ 930/ 1200] Overall Loss 0.154960 Objective Loss 0.154960 LR 0.000250 Time 0.019748 -2022-12-06 11:28:30,319 - Epoch: [147][ 940/ 1200] Overall Loss 0.154823 Objective Loss 0.154823 LR 0.000250 Time 0.019739 -2022-12-06 11:28:30,510 - Epoch: [147][ 950/ 1200] Overall Loss 0.154919 Objective Loss 0.154919 LR 0.000250 Time 0.019731 -2022-12-06 11:28:30,699 - Epoch: [147][ 960/ 1200] Overall Loss 0.155078 Objective Loss 0.155078 LR 0.000250 Time 0.019722 -2022-12-06 11:28:30,889 - Epoch: [147][ 970/ 1200] Overall Loss 0.155081 Objective Loss 0.155081 LR 0.000250 Time 0.019714 -2022-12-06 11:28:31,079 - Epoch: [147][ 980/ 1200] Overall Loss 0.155350 Objective Loss 0.155350 LR 0.000250 Time 0.019706 -2022-12-06 11:28:31,269 - Epoch: [147][ 990/ 1200] Overall Loss 0.155413 Objective Loss 0.155413 LR 0.000250 Time 0.019699 -2022-12-06 11:28:31,459 - Epoch: [147][ 1000/ 1200] Overall Loss 0.155549 Objective Loss 0.155549 LR 0.000250 Time 0.019691 -2022-12-06 11:28:31,649 - Epoch: [147][ 1010/ 1200] Overall Loss 0.155601 Objective Loss 0.155601 LR 0.000250 Time 0.019684 -2022-12-06 11:28:31,839 - Epoch: [147][ 1020/ 1200] Overall Loss 0.155706 Objective Loss 0.155706 LR 0.000250 Time 0.019677 -2022-12-06 11:28:32,030 - Epoch: [147][ 1030/ 1200] Overall Loss 0.155735 Objective Loss 0.155735 LR 0.000250 Time 0.019670 -2022-12-06 11:28:32,219 - Epoch: [147][ 1040/ 1200] Overall Loss 0.155751 Objective Loss 0.155751 LR 0.000250 Time 0.019663 -2022-12-06 11:28:32,409 - Epoch: [147][ 1050/ 1200] Overall Loss 0.155719 Objective Loss 0.155719 LR 0.000250 Time 0.019656 -2022-12-06 11:28:32,600 - Epoch: [147][ 1060/ 1200] Overall Loss 0.155804 Objective Loss 0.155804 LR 0.000250 Time 0.019649 -2022-12-06 11:28:32,790 - Epoch: [147][ 1070/ 1200] Overall Loss 0.155771 Objective Loss 0.155771 LR 0.000250 Time 0.019644 -2022-12-06 11:28:32,980 - Epoch: [147][ 1080/ 1200] Overall Loss 0.155843 Objective Loss 0.155843 LR 0.000250 Time 0.019637 -2022-12-06 11:28:33,171 - Epoch: [147][ 1090/ 1200] Overall Loss 0.156146 Objective Loss 0.156146 LR 0.000250 Time 0.019631 -2022-12-06 11:28:33,361 - Epoch: [147][ 1100/ 1200] Overall Loss 0.156133 Objective Loss 0.156133 LR 0.000250 Time 0.019625 -2022-12-06 11:28:33,551 - Epoch: [147][ 1110/ 1200] Overall Loss 0.156268 Objective Loss 0.156268 LR 0.000250 Time 0.019619 -2022-12-06 11:28:33,742 - Epoch: [147][ 1120/ 1200] Overall Loss 0.156413 Objective Loss 0.156413 LR 0.000250 Time 0.019613 -2022-12-06 11:28:33,932 - Epoch: [147][ 1130/ 1200] Overall Loss 0.156320 Objective Loss 0.156320 LR 0.000250 Time 0.019608 -2022-12-06 11:28:34,122 - Epoch: [147][ 1140/ 1200] Overall Loss 0.156462 Objective Loss 0.156462 LR 0.000250 Time 0.019602 -2022-12-06 11:28:34,313 - Epoch: [147][ 1150/ 1200] Overall Loss 0.156598 Objective Loss 0.156598 LR 0.000250 Time 0.019597 -2022-12-06 11:28:34,503 - Epoch: [147][ 1160/ 1200] Overall Loss 0.156358 Objective Loss 0.156358 LR 0.000250 Time 0.019592 -2022-12-06 11:28:34,694 - Epoch: [147][ 1170/ 1200] Overall Loss 0.156459 Objective Loss 0.156459 LR 0.000250 Time 0.019587 -2022-12-06 11:28:34,884 - Epoch: [147][ 1180/ 1200] Overall Loss 0.156218 Objective Loss 0.156218 LR 0.000250 Time 0.019582 -2022-12-06 11:28:35,075 - Epoch: [147][ 1190/ 1200] Overall Loss 0.156079 Objective Loss 0.156079 LR 0.000250 Time 0.019577 -2022-12-06 11:28:35,305 - Epoch: [147][ 1200/ 1200] Overall Loss 0.155974 Objective Loss 0.155974 Top1 88.702929 Top5 99.163180 LR 0.000250 Time 0.019605 -2022-12-06 11:28:35,393 - --- validate (epoch=147)----------- -2022-12-06 11:28:35,393 - 34129 samples (256 per mini-batch) -2022-12-06 11:28:35,837 - Epoch: [147][ 10/ 134] Loss 0.204243 Top1 88.476562 Top5 98.984375 -2022-12-06 11:28:35,978 - Epoch: [147][ 20/ 134] Loss 0.217252 Top1 88.125000 Top5 98.789062 -2022-12-06 11:28:36,102 - Epoch: [147][ 30/ 134] Loss 0.231317 Top1 87.708333 Top5 98.515625 -2022-12-06 11:28:36,256 - Epoch: [147][ 40/ 134] Loss 0.230775 Top1 87.636719 Top5 98.564453 -2022-12-06 11:28:36,384 - Epoch: [147][ 50/ 134] Loss 0.235055 Top1 87.492188 Top5 98.414062 -2022-12-06 11:28:36,510 - Epoch: [147][ 60/ 134] Loss 0.236499 Top1 87.571615 Top5 98.404948 -2022-12-06 11:28:36,640 - Epoch: [147][ 70/ 134] Loss 0.235793 Top1 87.667411 Top5 98.454241 -2022-12-06 11:28:36,787 - Epoch: [147][ 80/ 134] Loss 0.239423 Top1 87.475586 Top5 98.471680 -2022-12-06 11:28:36,912 - Epoch: [147][ 90/ 134] Loss 0.238925 Top1 87.491319 Top5 98.467882 -2022-12-06 11:28:37,043 - Epoch: [147][ 100/ 134] Loss 0.237946 Top1 87.464844 Top5 98.476562 -2022-12-06 11:28:37,176 - Epoch: [147][ 110/ 134] Loss 0.237584 Top1 87.482244 Top5 98.494318 -2022-12-06 11:28:37,308 - Epoch: [147][ 120/ 134] Loss 0.240017 Top1 87.522786 Top5 98.492839 -2022-12-06 11:28:37,439 - Epoch: [147][ 130/ 134] Loss 0.239680 Top1 87.533053 Top5 98.518630 -2022-12-06 11:28:37,476 - Epoch: [147][ 134/ 134] Loss 0.237944 Top1 87.532597 Top5 98.529110 -2022-12-06 11:28:37,569 - ==> Top1: 87.533 Top5: 98.529 Loss: 0.238 - -2022-12-06 11:28:37,570 - ==> Confusion: -[[ 912 1 1 4 3 3 1 1 4 47 0 2 0 3 7 1 2 0 1 0 3] - [ 1 936 1 2 10 21 3 12 0 0 5 4 2 1 0 2 3 2 11 3 8] - [ 4 2 1013 13 5 2 16 11 0 3 5 5 2 2 0 2 1 1 3 2 11] - [ 1 1 18 954 2 1 1 1 1 1 7 0 2 1 10 0 1 1 10 1 6] - [ 8 5 0 0 970 2 0 1 1 5 1 1 1 2 9 4 4 0 1 0 5] - [ 0 11 0 2 9 988 1 15 1 4 2 13 3 10 1 2 1 2 1 1 2] - [ 0 3 5 1 0 2 1079 4 0 0 0 1 0 2 0 5 0 4 2 8 2] - [ 2 6 7 2 2 25 10 961 0 0 1 4 0 1 0 1 1 0 14 11 6] - [ 6 3 0 1 0 2 2 0 972 38 13 2 0 12 8 1 0 0 1 1 2] - [ 48 0 1 0 7 2 0 1 18 896 1 2 0 14 4 1 0 1 1 0 4] - [ 0 2 4 4 1 0 2 5 4 1 968 0 1 12 2 1 1 0 3 1 7] - [ 4 0 2 0 0 12 5 2 0 0 0 974 18 4 1 7 3 6 0 12 1] - [ 0 0 0 4 1 2 1 1 1 0 0 27 905 2 1 7 1 6 0 3 7] - [ 2 0 0 0 2 7 0 3 6 6 2 3 2 979 2 2 2 0 0 1 4] - [ 6 3 1 11 4 1 0 0 14 2 1 2 2 5 1071 0 0 1 3 1 2] - [ 0 0 1 1 2 0 2 0 0 1 1 6 9 5 0 993 5 9 0 6 2] - [ 1 0 0 1 3 0 1 1 2 1 0 3 2 3 1 11 1030 0 0 5 7] - [ 5 0 2 3 1 1 2 1 0 3 0 8 20 3 1 21 0 959 0 2 4] - [ 2 3 3 12 1 2 1 29 1 1 3 1 3 0 8 1 0 2 931 1 3] - [ 3 4 1 1 0 5 3 5 0 1 3 15 6 6 1 3 2 2 0 1013 6] - [ 118 185 163 132 109 151 82 123 72 89 158 98 308 285 142 107 147 73 111 211 10362]] - -2022-12-06 11:28:38,231 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:28:38,232 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:28:38,237 - - -2022-12-06 11:28:38,237 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:28:39,182 - Epoch: [148][ 10/ 1200] Overall Loss 0.182051 Objective Loss 0.182051 LR 0.000250 Time 0.094411 -2022-12-06 11:28:39,382 - Epoch: [148][ 20/ 1200] Overall Loss 0.159623 Objective Loss 0.159623 LR 0.000250 Time 0.057168 -2022-12-06 11:28:39,580 - Epoch: [148][ 30/ 1200] Overall Loss 0.165505 Objective Loss 0.165505 LR 0.000250 Time 0.044679 -2022-12-06 11:28:39,774 - Epoch: [148][ 40/ 1200] Overall Loss 0.163496 Objective Loss 0.163496 LR 0.000250 Time 0.038365 -2022-12-06 11:28:39,972 - Epoch: [148][ 50/ 1200] Overall Loss 0.162669 Objective Loss 0.162669 LR 0.000250 Time 0.034637 -2022-12-06 11:28:40,167 - Epoch: [148][ 60/ 1200] Overall Loss 0.160393 Objective Loss 0.160393 LR 0.000250 Time 0.032094 -2022-12-06 11:28:40,364 - Epoch: [148][ 70/ 1200] Overall Loss 0.161089 Objective Loss 0.161089 LR 0.000250 Time 0.030329 -2022-12-06 11:28:40,559 - Epoch: [148][ 80/ 1200] Overall Loss 0.158038 Objective Loss 0.158038 LR 0.000250 Time 0.028959 -2022-12-06 11:28:40,756 - Epoch: [148][ 90/ 1200] Overall Loss 0.158373 Objective Loss 0.158373 LR 0.000250 Time 0.027928 -2022-12-06 11:28:40,950 - Epoch: [148][ 100/ 1200] Overall Loss 0.159902 Objective Loss 0.159902 LR 0.000250 Time 0.027068 -2022-12-06 11:28:41,148 - Epoch: [148][ 110/ 1200] Overall Loss 0.160369 Objective Loss 0.160369 LR 0.000250 Time 0.026404 -2022-12-06 11:28:41,342 - Epoch: [148][ 120/ 1200] Overall Loss 0.160198 Objective Loss 0.160198 LR 0.000250 Time 0.025816 -2022-12-06 11:28:41,540 - Epoch: [148][ 130/ 1200] Overall Loss 0.158709 Objective Loss 0.158709 LR 0.000250 Time 0.025348 -2022-12-06 11:28:41,734 - Epoch: [148][ 140/ 1200] Overall Loss 0.158866 Objective Loss 0.158866 LR 0.000250 Time 0.024919 -2022-12-06 11:28:41,932 - Epoch: [148][ 150/ 1200] Overall Loss 0.158815 Objective Loss 0.158815 LR 0.000250 Time 0.024578 -2022-12-06 11:28:42,129 - Epoch: [148][ 160/ 1200] Overall Loss 0.158044 Objective Loss 0.158044 LR 0.000250 Time 0.024264 -2022-12-06 11:28:42,328 - Epoch: [148][ 170/ 1200] Overall Loss 0.157971 Objective Loss 0.157971 LR 0.000250 Time 0.024008 -2022-12-06 11:28:42,525 - Epoch: [148][ 180/ 1200] Overall Loss 0.157645 Objective Loss 0.157645 LR 0.000250 Time 0.023766 -2022-12-06 11:28:42,725 - Epoch: [148][ 190/ 1200] Overall Loss 0.156154 Objective Loss 0.156154 LR 0.000250 Time 0.023565 -2022-12-06 11:28:42,921 - Epoch: [148][ 200/ 1200] Overall Loss 0.156675 Objective Loss 0.156675 LR 0.000250 Time 0.023365 -2022-12-06 11:28:43,121 - Epoch: [148][ 210/ 1200] Overall Loss 0.156035 Objective Loss 0.156035 LR 0.000250 Time 0.023198 -2022-12-06 11:28:43,317 - Epoch: [148][ 220/ 1200] Overall Loss 0.155922 Objective Loss 0.155922 LR 0.000250 Time 0.023035 -2022-12-06 11:28:43,517 - Epoch: [148][ 230/ 1200] Overall Loss 0.155126 Objective Loss 0.155126 LR 0.000250 Time 0.022900 -2022-12-06 11:28:43,714 - Epoch: [148][ 240/ 1200] Overall Loss 0.155730 Objective Loss 0.155730 LR 0.000250 Time 0.022764 -2022-12-06 11:28:43,914 - Epoch: [148][ 250/ 1200] Overall Loss 0.155400 Objective Loss 0.155400 LR 0.000250 Time 0.022650 -2022-12-06 11:28:44,110 - Epoch: [148][ 260/ 1200] Overall Loss 0.155077 Objective Loss 0.155077 LR 0.000250 Time 0.022531 -2022-12-06 11:28:44,309 - Epoch: [148][ 270/ 1200] Overall Loss 0.154434 Objective Loss 0.154434 LR 0.000250 Time 0.022434 -2022-12-06 11:28:44,505 - Epoch: [148][ 280/ 1200] Overall Loss 0.154488 Objective Loss 0.154488 LR 0.000250 Time 0.022331 -2022-12-06 11:28:44,705 - Epoch: [148][ 290/ 1200] Overall Loss 0.154687 Objective Loss 0.154687 LR 0.000250 Time 0.022246 -2022-12-06 11:28:44,901 - Epoch: [148][ 300/ 1200] Overall Loss 0.154897 Objective Loss 0.154897 LR 0.000250 Time 0.022158 -2022-12-06 11:28:45,101 - Epoch: [148][ 310/ 1200] Overall Loss 0.154915 Objective Loss 0.154915 LR 0.000250 Time 0.022086 -2022-12-06 11:28:45,297 - Epoch: [148][ 320/ 1200] Overall Loss 0.155271 Objective Loss 0.155271 LR 0.000250 Time 0.022006 -2022-12-06 11:28:45,496 - Epoch: [148][ 330/ 1200] Overall Loss 0.155199 Objective Loss 0.155199 LR 0.000250 Time 0.021940 -2022-12-06 11:28:45,692 - Epoch: [148][ 340/ 1200] Overall Loss 0.155288 Objective Loss 0.155288 LR 0.000250 Time 0.021870 -2022-12-06 11:28:45,892 - Epoch: [148][ 350/ 1200] Overall Loss 0.155698 Objective Loss 0.155698 LR 0.000250 Time 0.021817 -2022-12-06 11:28:46,090 - Epoch: [148][ 360/ 1200] Overall Loss 0.155537 Objective Loss 0.155537 LR 0.000250 Time 0.021757 -2022-12-06 11:28:46,289 - Epoch: [148][ 370/ 1200] Overall Loss 0.155847 Objective Loss 0.155847 LR 0.000250 Time 0.021706 -2022-12-06 11:28:46,486 - Epoch: [148][ 380/ 1200] Overall Loss 0.155642 Objective Loss 0.155642 LR 0.000250 Time 0.021652 -2022-12-06 11:28:46,686 - Epoch: [148][ 390/ 1200] Overall Loss 0.155012 Objective Loss 0.155012 LR 0.000250 Time 0.021609 -2022-12-06 11:28:46,883 - Epoch: [148][ 400/ 1200] Overall Loss 0.155095 Objective Loss 0.155095 LR 0.000250 Time 0.021558 -2022-12-06 11:28:47,082 - Epoch: [148][ 410/ 1200] Overall Loss 0.154857 Objective Loss 0.154857 LR 0.000250 Time 0.021517 -2022-12-06 11:28:47,278 - Epoch: [148][ 420/ 1200] Overall Loss 0.155081 Objective Loss 0.155081 LR 0.000250 Time 0.021471 -2022-12-06 11:28:47,478 - Epoch: [148][ 430/ 1200] Overall Loss 0.155177 Objective Loss 0.155177 LR 0.000250 Time 0.021435 -2022-12-06 11:28:47,675 - Epoch: [148][ 440/ 1200] Overall Loss 0.155129 Objective Loss 0.155129 LR 0.000250 Time 0.021395 -2022-12-06 11:28:47,874 - Epoch: [148][ 450/ 1200] Overall Loss 0.155025 Objective Loss 0.155025 LR 0.000250 Time 0.021360 -2022-12-06 11:28:48,072 - Epoch: [148][ 460/ 1200] Overall Loss 0.155560 Objective Loss 0.155560 LR 0.000250 Time 0.021324 -2022-12-06 11:28:48,271 - Epoch: [148][ 470/ 1200] Overall Loss 0.155342 Objective Loss 0.155342 LR 0.000250 Time 0.021294 -2022-12-06 11:28:48,469 - Epoch: [148][ 480/ 1200] Overall Loss 0.155565 Objective Loss 0.155565 LR 0.000250 Time 0.021260 -2022-12-06 11:28:48,668 - Epoch: [148][ 490/ 1200] Overall Loss 0.155532 Objective Loss 0.155532 LR 0.000250 Time 0.021232 -2022-12-06 11:28:48,864 - Epoch: [148][ 500/ 1200] Overall Loss 0.155493 Objective Loss 0.155493 LR 0.000250 Time 0.021199 -2022-12-06 11:28:49,065 - Epoch: [148][ 510/ 1200] Overall Loss 0.155765 Objective Loss 0.155765 LR 0.000250 Time 0.021176 -2022-12-06 11:28:49,262 - Epoch: [148][ 520/ 1200] Overall Loss 0.155366 Objective Loss 0.155366 LR 0.000250 Time 0.021146 -2022-12-06 11:28:49,460 - Epoch: [148][ 530/ 1200] Overall Loss 0.155815 Objective Loss 0.155815 LR 0.000250 Time 0.021120 -2022-12-06 11:28:49,657 - Epoch: [148][ 540/ 1200] Overall Loss 0.155801 Objective Loss 0.155801 LR 0.000250 Time 0.021093 -2022-12-06 11:28:49,857 - Epoch: [148][ 550/ 1200] Overall Loss 0.156050 Objective Loss 0.156050 LR 0.000250 Time 0.021072 -2022-12-06 11:28:50,054 - Epoch: [148][ 560/ 1200] Overall Loss 0.155980 Objective Loss 0.155980 LR 0.000250 Time 0.021047 -2022-12-06 11:28:50,254 - Epoch: [148][ 570/ 1200] Overall Loss 0.155922 Objective Loss 0.155922 LR 0.000250 Time 0.021027 -2022-12-06 11:28:50,450 - Epoch: [148][ 580/ 1200] Overall Loss 0.155790 Objective Loss 0.155790 LR 0.000250 Time 0.021002 -2022-12-06 11:28:50,650 - Epoch: [148][ 590/ 1200] Overall Loss 0.155725 Objective Loss 0.155725 LR 0.000250 Time 0.020983 -2022-12-06 11:28:50,847 - Epoch: [148][ 600/ 1200] Overall Loss 0.156041 Objective Loss 0.156041 LR 0.000250 Time 0.020961 -2022-12-06 11:28:51,046 - Epoch: [148][ 610/ 1200] Overall Loss 0.156038 Objective Loss 0.156038 LR 0.000250 Time 0.020944 -2022-12-06 11:28:51,244 - Epoch: [148][ 620/ 1200] Overall Loss 0.155867 Objective Loss 0.155867 LR 0.000250 Time 0.020923 -2022-12-06 11:28:51,443 - Epoch: [148][ 630/ 1200] Overall Loss 0.155903 Objective Loss 0.155903 LR 0.000250 Time 0.020906 -2022-12-06 11:28:51,640 - Epoch: [148][ 640/ 1200] Overall Loss 0.155773 Objective Loss 0.155773 LR 0.000250 Time 0.020887 -2022-12-06 11:28:51,839 - Epoch: [148][ 650/ 1200] Overall Loss 0.155618 Objective Loss 0.155618 LR 0.000250 Time 0.020871 -2022-12-06 11:28:52,036 - Epoch: [148][ 660/ 1200] Overall Loss 0.155587 Objective Loss 0.155587 LR 0.000250 Time 0.020852 -2022-12-06 11:28:52,235 - Epoch: [148][ 670/ 1200] Overall Loss 0.155433 Objective Loss 0.155433 LR 0.000250 Time 0.020837 -2022-12-06 11:28:52,432 - Epoch: [148][ 680/ 1200] Overall Loss 0.155648 Objective Loss 0.155648 LR 0.000250 Time 0.020819 -2022-12-06 11:28:52,631 - Epoch: [148][ 690/ 1200] Overall Loss 0.155311 Objective Loss 0.155311 LR 0.000250 Time 0.020805 -2022-12-06 11:28:52,828 - Epoch: [148][ 700/ 1200] Overall Loss 0.154974 Objective Loss 0.154974 LR 0.000250 Time 0.020788 -2022-12-06 11:28:53,026 - Epoch: [148][ 710/ 1200] Overall Loss 0.155159 Objective Loss 0.155159 LR 0.000250 Time 0.020774 -2022-12-06 11:28:53,223 - Epoch: [148][ 720/ 1200] Overall Loss 0.155151 Objective Loss 0.155151 LR 0.000250 Time 0.020758 -2022-12-06 11:28:53,422 - Epoch: [148][ 730/ 1200] Overall Loss 0.155315 Objective Loss 0.155315 LR 0.000250 Time 0.020746 -2022-12-06 11:28:53,618 - Epoch: [148][ 740/ 1200] Overall Loss 0.155364 Objective Loss 0.155364 LR 0.000250 Time 0.020730 -2022-12-06 11:28:53,817 - Epoch: [148][ 750/ 1200] Overall Loss 0.155232 Objective Loss 0.155232 LR 0.000250 Time 0.020718 -2022-12-06 11:28:54,014 - Epoch: [148][ 760/ 1200] Overall Loss 0.155286 Objective Loss 0.155286 LR 0.000250 Time 0.020704 -2022-12-06 11:28:54,214 - Epoch: [148][ 770/ 1200] Overall Loss 0.155427 Objective Loss 0.155427 LR 0.000250 Time 0.020694 -2022-12-06 11:28:54,411 - Epoch: [148][ 780/ 1200] Overall Loss 0.155283 Objective Loss 0.155283 LR 0.000250 Time 0.020681 -2022-12-06 11:28:54,611 - Epoch: [148][ 790/ 1200] Overall Loss 0.155188 Objective Loss 0.155188 LR 0.000250 Time 0.020671 -2022-12-06 11:28:54,808 - Epoch: [148][ 800/ 1200] Overall Loss 0.155307 Objective Loss 0.155307 LR 0.000250 Time 0.020659 -2022-12-06 11:28:55,007 - Epoch: [148][ 810/ 1200] Overall Loss 0.155332 Objective Loss 0.155332 LR 0.000250 Time 0.020649 -2022-12-06 11:28:55,204 - Epoch: [148][ 820/ 1200] Overall Loss 0.155168 Objective Loss 0.155168 LR 0.000250 Time 0.020636 -2022-12-06 11:28:55,403 - Epoch: [148][ 830/ 1200] Overall Loss 0.154986 Objective Loss 0.154986 LR 0.000250 Time 0.020627 -2022-12-06 11:28:55,601 - Epoch: [148][ 840/ 1200] Overall Loss 0.154878 Objective Loss 0.154878 LR 0.000250 Time 0.020616 -2022-12-06 11:28:55,800 - Epoch: [148][ 850/ 1200] Overall Loss 0.154723 Objective Loss 0.154723 LR 0.000250 Time 0.020607 -2022-12-06 11:28:55,996 - Epoch: [148][ 860/ 1200] Overall Loss 0.154435 Objective Loss 0.154435 LR 0.000250 Time 0.020595 -2022-12-06 11:28:56,196 - Epoch: [148][ 870/ 1200] Overall Loss 0.154495 Objective Loss 0.154495 LR 0.000250 Time 0.020587 -2022-12-06 11:28:56,394 - Epoch: [148][ 880/ 1200] Overall Loss 0.154655 Objective Loss 0.154655 LR 0.000250 Time 0.020577 -2022-12-06 11:28:56,593 - Epoch: [148][ 890/ 1200] Overall Loss 0.154750 Objective Loss 0.154750 LR 0.000250 Time 0.020569 -2022-12-06 11:28:56,789 - Epoch: [148][ 900/ 1200] Overall Loss 0.154712 Objective Loss 0.154712 LR 0.000250 Time 0.020558 -2022-12-06 11:28:56,989 - Epoch: [148][ 910/ 1200] Overall Loss 0.154826 Objective Loss 0.154826 LR 0.000250 Time 0.020552 -2022-12-06 11:28:57,186 - Epoch: [148][ 920/ 1200] Overall Loss 0.154916 Objective Loss 0.154916 LR 0.000250 Time 0.020542 -2022-12-06 11:28:57,386 - Epoch: [148][ 930/ 1200] Overall Loss 0.154727 Objective Loss 0.154727 LR 0.000250 Time 0.020536 -2022-12-06 11:28:57,584 - Epoch: [148][ 940/ 1200] Overall Loss 0.154844 Objective Loss 0.154844 LR 0.000250 Time 0.020527 -2022-12-06 11:28:57,784 - Epoch: [148][ 950/ 1200] Overall Loss 0.154791 Objective Loss 0.154791 LR 0.000250 Time 0.020521 -2022-12-06 11:28:57,981 - Epoch: [148][ 960/ 1200] Overall Loss 0.154856 Objective Loss 0.154856 LR 0.000250 Time 0.020512 -2022-12-06 11:28:58,182 - Epoch: [148][ 970/ 1200] Overall Loss 0.154871 Objective Loss 0.154871 LR 0.000250 Time 0.020506 -2022-12-06 11:28:58,379 - Epoch: [148][ 980/ 1200] Overall Loss 0.154766 Objective Loss 0.154766 LR 0.000250 Time 0.020498 -2022-12-06 11:28:58,579 - Epoch: [148][ 990/ 1200] Overall Loss 0.154939 Objective Loss 0.154939 LR 0.000250 Time 0.020492 -2022-12-06 11:28:58,775 - Epoch: [148][ 1000/ 1200] Overall Loss 0.154765 Objective Loss 0.154765 LR 0.000250 Time 0.020482 -2022-12-06 11:28:58,972 - Epoch: [148][ 1010/ 1200] Overall Loss 0.154826 Objective Loss 0.154826 LR 0.000250 Time 0.020475 -2022-12-06 11:28:59,167 - Epoch: [148][ 1020/ 1200] Overall Loss 0.154810 Objective Loss 0.154810 LR 0.000250 Time 0.020464 -2022-12-06 11:28:59,365 - Epoch: [148][ 1030/ 1200] Overall Loss 0.154672 Objective Loss 0.154672 LR 0.000250 Time 0.020458 -2022-12-06 11:28:59,561 - Epoch: [148][ 1040/ 1200] Overall Loss 0.154599 Objective Loss 0.154599 LR 0.000250 Time 0.020448 -2022-12-06 11:28:59,761 - Epoch: [148][ 1050/ 1200] Overall Loss 0.154615 Objective Loss 0.154615 LR 0.000250 Time 0.020444 -2022-12-06 11:28:59,959 - Epoch: [148][ 1060/ 1200] Overall Loss 0.154693 Objective Loss 0.154693 LR 0.000250 Time 0.020437 -2022-12-06 11:29:00,158 - Epoch: [148][ 1070/ 1200] Overall Loss 0.154669 Objective Loss 0.154669 LR 0.000250 Time 0.020432 -2022-12-06 11:29:00,356 - Epoch: [148][ 1080/ 1200] Overall Loss 0.154809 Objective Loss 0.154809 LR 0.000250 Time 0.020425 -2022-12-06 11:29:00,555 - Epoch: [148][ 1090/ 1200] Overall Loss 0.154918 Objective Loss 0.154918 LR 0.000250 Time 0.020420 -2022-12-06 11:29:00,752 - Epoch: [148][ 1100/ 1200] Overall Loss 0.154913 Objective Loss 0.154913 LR 0.000250 Time 0.020413 -2022-12-06 11:29:00,952 - Epoch: [148][ 1110/ 1200] Overall Loss 0.154956 Objective Loss 0.154956 LR 0.000250 Time 0.020409 -2022-12-06 11:29:01,149 - Epoch: [148][ 1120/ 1200] Overall Loss 0.154952 Objective Loss 0.154952 LR 0.000250 Time 0.020402 -2022-12-06 11:29:01,349 - Epoch: [148][ 1130/ 1200] Overall Loss 0.155235 Objective Loss 0.155235 LR 0.000250 Time 0.020398 -2022-12-06 11:29:01,546 - Epoch: [148][ 1140/ 1200] Overall Loss 0.155393 Objective Loss 0.155393 LR 0.000250 Time 0.020392 -2022-12-06 11:29:01,746 - Epoch: [148][ 1150/ 1200] Overall Loss 0.155309 Objective Loss 0.155309 LR 0.000250 Time 0.020387 -2022-12-06 11:29:01,943 - Epoch: [148][ 1160/ 1200] Overall Loss 0.155496 Objective Loss 0.155496 LR 0.000250 Time 0.020381 -2022-12-06 11:29:02,142 - Epoch: [148][ 1170/ 1200] Overall Loss 0.155529 Objective Loss 0.155529 LR 0.000250 Time 0.020377 -2022-12-06 11:29:02,340 - Epoch: [148][ 1180/ 1200] Overall Loss 0.155425 Objective Loss 0.155425 LR 0.000250 Time 0.020371 -2022-12-06 11:29:02,541 - Epoch: [148][ 1190/ 1200] Overall Loss 0.155462 Objective Loss 0.155462 LR 0.000250 Time 0.020368 -2022-12-06 11:29:02,765 - Epoch: [148][ 1200/ 1200] Overall Loss 0.155362 Objective Loss 0.155362 Top1 89.330544 Top5 98.535565 LR 0.000250 Time 0.020385 -2022-12-06 11:29:02,853 - --- validate (epoch=148)----------- -2022-12-06 11:29:02,854 - 34129 samples (256 per mini-batch) -2022-12-06 11:29:03,307 - Epoch: [148][ 10/ 134] Loss 0.230193 Top1 87.617188 Top5 98.671875 -2022-12-06 11:29:03,442 - Epoch: [148][ 20/ 134] Loss 0.237916 Top1 87.421875 Top5 98.750000 -2022-12-06 11:29:03,574 - Epoch: [148][ 30/ 134] Loss 0.237944 Top1 87.174479 Top5 98.750000 -2022-12-06 11:29:03,705 - Epoch: [148][ 40/ 134] Loss 0.232411 Top1 87.148438 Top5 98.740234 -2022-12-06 11:29:03,838 - Epoch: [148][ 50/ 134] Loss 0.232453 Top1 87.289062 Top5 98.687500 -2022-12-06 11:29:03,968 - Epoch: [148][ 60/ 134] Loss 0.234758 Top1 87.356771 Top5 98.626302 -2022-12-06 11:29:04,100 - Epoch: [148][ 70/ 134] Loss 0.235674 Top1 87.349330 Top5 98.565848 -2022-12-06 11:29:04,231 - Epoch: [148][ 80/ 134] Loss 0.235417 Top1 87.382812 Top5 98.579102 -2022-12-06 11:29:04,361 - Epoch: [148][ 90/ 134] Loss 0.238514 Top1 87.291667 Top5 98.524306 -2022-12-06 11:29:04,489 - Epoch: [148][ 100/ 134] Loss 0.242018 Top1 87.148438 Top5 98.484375 -2022-12-06 11:29:04,619 - Epoch: [148][ 110/ 134] Loss 0.241031 Top1 87.169744 Top5 98.437500 -2022-12-06 11:29:04,750 - Epoch: [148][ 120/ 134] Loss 0.241126 Top1 87.177734 Top5 98.453776 -2022-12-06 11:29:04,882 - Epoch: [148][ 130/ 134] Loss 0.239406 Top1 87.211538 Top5 98.473558 -2022-12-06 11:29:04,920 - Epoch: [148][ 134/ 134] Loss 0.239044 Top1 87.222011 Top5 98.458789 -2022-12-06 11:29:05,021 - ==> Top1: 87.222 Top5: 98.459 Loss: 0.239 - -2022-12-06 11:29:05,022 - ==> Confusion: -[[ 921 0 0 3 2 7 1 0 3 45 0 1 0 3 2 1 2 1 1 0 3] - [ 2 940 2 2 11 22 2 14 1 0 1 3 0 1 0 2 4 0 8 2 10] - [ 6 3 1011 11 2 2 15 13 0 4 5 3 1 2 2 4 0 4 3 2 10] - [ 0 1 15 949 0 2 1 0 0 0 12 0 3 0 12 0 0 2 16 1 6] - [ 9 4 2 1 961 4 2 1 1 6 2 2 0 2 8 4 4 3 1 0 3] - [ 2 13 0 3 7 989 3 16 1 4 1 10 3 12 0 1 0 0 0 3 1] - [ 1 2 8 2 0 0 1083 4 0 0 1 0 0 3 0 2 1 2 1 7 1] - [ 3 4 9 2 2 26 9 962 0 0 2 3 1 2 0 1 1 0 20 4 3] - [ 4 2 0 0 0 2 1 0 988 38 9 1 1 5 10 0 0 0 1 1 1] - [ 49 0 2 1 5 1 0 3 24 899 1 2 0 6 3 2 0 1 0 0 2] - [ 0 2 3 3 0 0 2 4 12 3 966 0 0 10 4 1 0 0 6 0 3] - [ 3 2 2 0 0 12 2 3 0 1 0 979 22 6 1 5 2 4 0 6 1] - [ 0 1 1 3 0 2 0 1 1 1 0 24 908 2 0 4 1 9 1 3 7] - [ 0 1 1 0 2 7 0 2 17 12 2 6 3 955 2 3 3 0 0 2 5] - [ 9 3 2 7 5 1 0 0 13 3 0 3 3 4 1067 0 0 1 7 0 2] - [ 1 0 1 0 1 0 3 0 0 1 0 9 10 2 0 993 4 11 0 4 3] - [ 2 2 2 1 3 1 1 1 1 0 0 2 2 2 2 12 1029 0 0 2 7] - [ 2 1 2 3 0 1 1 1 0 4 0 9 9 1 1 14 2 979 0 3 3] - [ 1 4 1 6 2 3 0 16 1 1 5 1 3 0 7 2 0 2 948 2 3] - [ 0 3 2 1 1 6 8 8 0 1 3 14 7 7 1 4 3 4 0 1002 5] - [ 129 199 155 114 112 171 87 137 103 105 167 104 306 247 157 111 147 81 165 195 10234]] - -2022-12-06 11:29:05,591 - ==> Best [Top1: 87.911 Top5: 98.526 Sparsity:0.00 Params: 5376 on epoch: 143] -2022-12-06 11:29:05,591 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:29:05,597 - - -2022-12-06 11:29:05,597 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:29:06,644 - Epoch: [149][ 10/ 1200] Overall Loss 0.139321 Objective Loss 0.139321 LR 0.000250 Time 0.104582 -2022-12-06 11:29:06,844 - Epoch: [149][ 20/ 1200] Overall Loss 0.163525 Objective Loss 0.163525 LR 0.000250 Time 0.062272 -2022-12-06 11:29:07,035 - Epoch: [149][ 30/ 1200] Overall Loss 0.157118 Objective Loss 0.157118 LR 0.000250 Time 0.047847 -2022-12-06 11:29:07,225 - Epoch: [149][ 40/ 1200] Overall Loss 0.157636 Objective Loss 0.157636 LR 0.000250 Time 0.040641 -2022-12-06 11:29:07,415 - Epoch: [149][ 50/ 1200] Overall Loss 0.158395 Objective Loss 0.158395 LR 0.000250 Time 0.036303 -2022-12-06 11:29:07,605 - Epoch: [149][ 60/ 1200] Overall Loss 0.157962 Objective Loss 0.157962 LR 0.000250 Time 0.033411 -2022-12-06 11:29:07,795 - Epoch: [149][ 70/ 1200] Overall Loss 0.156107 Objective Loss 0.156107 LR 0.000250 Time 0.031343 -2022-12-06 11:29:07,986 - Epoch: [149][ 80/ 1200] Overall Loss 0.155084 Objective Loss 0.155084 LR 0.000250 Time 0.029805 -2022-12-06 11:29:08,177 - Epoch: [149][ 90/ 1200] Overall Loss 0.155143 Objective Loss 0.155143 LR 0.000250 Time 0.028601 -2022-12-06 11:29:08,367 - Epoch: [149][ 100/ 1200] Overall Loss 0.154097 Objective Loss 0.154097 LR 0.000250 Time 0.027636 -2022-12-06 11:29:08,557 - Epoch: [149][ 110/ 1200] Overall Loss 0.154527 Objective Loss 0.154527 LR 0.000250 Time 0.026850 -2022-12-06 11:29:08,748 - Epoch: [149][ 120/ 1200] Overall Loss 0.156300 Objective Loss 0.156300 LR 0.000250 Time 0.026198 -2022-12-06 11:29:08,938 - Epoch: [149][ 130/ 1200] Overall Loss 0.156482 Objective Loss 0.156482 LR 0.000250 Time 0.025643 -2022-12-06 11:29:09,129 - Epoch: [149][ 140/ 1200] Overall Loss 0.153931 Objective Loss 0.153931 LR 0.000250 Time 0.025169 -2022-12-06 11:29:09,319 - Epoch: [149][ 150/ 1200] Overall Loss 0.154648 Objective Loss 0.154648 LR 0.000250 Time 0.024752 -2022-12-06 11:29:09,510 - Epoch: [149][ 160/ 1200] Overall Loss 0.155766 Objective Loss 0.155766 LR 0.000250 Time 0.024395 -2022-12-06 11:29:09,699 - Epoch: [149][ 170/ 1200] Overall Loss 0.155817 Objective Loss 0.155817 LR 0.000250 Time 0.024071 -2022-12-06 11:29:09,889 - Epoch: [149][ 180/ 1200] Overall Loss 0.155589 Objective Loss 0.155589 LR 0.000250 Time 0.023788 -2022-12-06 11:29:10,079 - Epoch: [149][ 190/ 1200] Overall Loss 0.155621 Objective Loss 0.155621 LR 0.000250 Time 0.023532 -2022-12-06 11:29:10,270 - Epoch: [149][ 200/ 1200] Overall Loss 0.155449 Objective Loss 0.155449 LR 0.000250 Time 0.023307 -2022-12-06 11:29:10,462 - Epoch: [149][ 210/ 1200] Overall Loss 0.154900 Objective Loss 0.154900 LR 0.000250 Time 0.023109 -2022-12-06 11:29:10,654 - Epoch: [149][ 220/ 1200] Overall Loss 0.153591 Objective Loss 0.153591 LR 0.000250 Time 0.022928 -2022-12-06 11:29:10,845 - Epoch: [149][ 230/ 1200] Overall Loss 0.153843 Objective Loss 0.153843 LR 0.000250 Time 0.022762 -2022-12-06 11:29:11,038 - Epoch: [149][ 240/ 1200] Overall Loss 0.153631 Objective Loss 0.153631 LR 0.000250 Time 0.022612 -2022-12-06 11:29:11,230 - Epoch: [149][ 250/ 1200] Overall Loss 0.153576 Objective Loss 0.153576 LR 0.000250 Time 0.022473 -2022-12-06 11:29:11,422 - Epoch: [149][ 260/ 1200] Overall Loss 0.154092 Objective Loss 0.154092 LR 0.000250 Time 0.022345 -2022-12-06 11:29:11,613 - Epoch: [149][ 270/ 1200] Overall Loss 0.153545 Objective Loss 0.153545 LR 0.000250 Time 0.022226 -2022-12-06 11:29:11,805 - Epoch: [149][ 280/ 1200] Overall Loss 0.153302 Objective Loss 0.153302 LR 0.000250 Time 0.022113 -2022-12-06 11:29:11,997 - Epoch: [149][ 290/ 1200] Overall Loss 0.153156 Objective Loss 0.153156 LR 0.000250 Time 0.022011 -2022-12-06 11:29:12,189 - Epoch: [149][ 300/ 1200] Overall Loss 0.153468 Objective Loss 0.153468 LR 0.000250 Time 0.021915 -2022-12-06 11:29:12,381 - Epoch: [149][ 310/ 1200] Overall Loss 0.153269 Objective Loss 0.153269 LR 0.000250 Time 0.021826 -2022-12-06 11:29:12,573 - Epoch: [149][ 320/ 1200] Overall Loss 0.153350 Objective Loss 0.153350 LR 0.000250 Time 0.021742 -2022-12-06 11:29:12,765 - Epoch: [149][ 330/ 1200] Overall Loss 0.153564 Objective Loss 0.153564 LR 0.000250 Time 0.021663 -2022-12-06 11:29:12,957 - Epoch: [149][ 340/ 1200] Overall Loss 0.153465 Objective Loss 0.153465 LR 0.000250 Time 0.021591 -2022-12-06 11:29:13,149 - Epoch: [149][ 350/ 1200] Overall Loss 0.153277 Objective Loss 0.153277 LR 0.000250 Time 0.021520 -2022-12-06 11:29:13,342 - Epoch: [149][ 360/ 1200] Overall Loss 0.153364 Objective Loss 0.153364 LR 0.000250 Time 0.021456 -2022-12-06 11:29:13,534 - Epoch: [149][ 370/ 1200] Overall Loss 0.152886 Objective Loss 0.152886 LR 0.000250 Time 0.021394 -2022-12-06 11:29:13,726 - Epoch: [149][ 380/ 1200] Overall Loss 0.153365 Objective Loss 0.153365 LR 0.000250 Time 0.021335 -2022-12-06 11:29:13,918 - Epoch: [149][ 390/ 1200] Overall Loss 0.153479 Objective Loss 0.153479 LR 0.000250 Time 0.021280 -2022-12-06 11:29:14,110 - Epoch: [149][ 400/ 1200] Overall Loss 0.153583 Objective Loss 0.153583 LR 0.000250 Time 0.021226 -2022-12-06 11:29:14,302 - Epoch: [149][ 410/ 1200] Overall Loss 0.153607 Objective Loss 0.153607 LR 0.000250 Time 0.021176 -2022-12-06 11:29:14,493 - Epoch: [149][ 420/ 1200] Overall Loss 0.153830 Objective Loss 0.153830 LR 0.000250 Time 0.021125 -2022-12-06 11:29:14,685 - Epoch: [149][ 430/ 1200] Overall Loss 0.154179 Objective Loss 0.154179 LR 0.000250 Time 0.021079 -2022-12-06 11:29:14,878 - Epoch: [149][ 440/ 1200] Overall Loss 0.154104 Objective Loss 0.154104 LR 0.000250 Time 0.021036 -2022-12-06 11:29:15,069 - Epoch: [149][ 450/ 1200] Overall Loss 0.153719 Objective Loss 0.153719 LR 0.000250 Time 0.020993 -2022-12-06 11:29:15,261 - Epoch: [149][ 460/ 1200] Overall Loss 0.153527 Objective Loss 0.153527 LR 0.000250 Time 0.020953 -2022-12-06 11:29:15,453 - Epoch: [149][ 470/ 1200] Overall Loss 0.153495 Objective Loss 0.153495 LR 0.000250 Time 0.020914 -2022-12-06 11:29:15,645 - Epoch: [149][ 480/ 1200] Overall Loss 0.153506 Objective Loss 0.153506 LR 0.000250 Time 0.020877 -2022-12-06 11:29:15,837 - Epoch: [149][ 490/ 1200] Overall Loss 0.153596 Objective Loss 0.153596 LR 0.000250 Time 0.020841 -2022-12-06 11:29:16,029 - Epoch: [149][ 500/ 1200] Overall Loss 0.153346 Objective Loss 0.153346 LR 0.000250 Time 0.020809 -2022-12-06 11:29:16,222 - Epoch: [149][ 510/ 1200] Overall Loss 0.153503 Objective Loss 0.153503 LR 0.000250 Time 0.020776 -2022-12-06 11:29:16,414 - Epoch: [149][ 520/ 1200] Overall Loss 0.153781 Objective Loss 0.153781 LR 0.000250 Time 0.020746 -2022-12-06 11:29:16,606 - Epoch: [149][ 530/ 1200] Overall Loss 0.153911 Objective Loss 0.153911 LR 0.000250 Time 0.020715 -2022-12-06 11:29:16,798 - Epoch: [149][ 540/ 1200] Overall Loss 0.154005 Objective Loss 0.154005 LR 0.000250 Time 0.020686 -2022-12-06 11:29:16,990 - Epoch: [149][ 550/ 1200] Overall Loss 0.154354 Objective Loss 0.154354 LR 0.000250 Time 0.020658 -2022-12-06 11:29:17,182 - Epoch: [149][ 560/ 1200] Overall Loss 0.154316 Objective Loss 0.154316 LR 0.000250 Time 0.020631 -2022-12-06 11:29:17,374 - Epoch: [149][ 570/ 1200] Overall Loss 0.154200 Objective Loss 0.154200 LR 0.000250 Time 0.020605 -2022-12-06 11:29:17,567 - Epoch: [149][ 580/ 1200] Overall Loss 0.153868 Objective Loss 0.153868 LR 0.000250 Time 0.020581 -2022-12-06 11:29:17,759 - Epoch: [149][ 590/ 1200] Overall Loss 0.154038 Objective Loss 0.154038 LR 0.000250 Time 0.020557 -2022-12-06 11:29:17,951 - Epoch: [149][ 600/ 1200] Overall Loss 0.154296 Objective Loss 0.154296 LR 0.000250 Time 0.020534 -2022-12-06 11:29:18,143 - Epoch: [149][ 610/ 1200] Overall Loss 0.154354 Objective Loss 0.154354 LR 0.000250 Time 0.020512 -2022-12-06 11:29:18,335 - Epoch: [149][ 620/ 1200] Overall Loss 0.154643 Objective Loss 0.154643 LR 0.000250 Time 0.020490 -2022-12-06 11:29:18,528 - Epoch: [149][ 630/ 1200] Overall Loss 0.154566 Objective Loss 0.154566 LR 0.000250 Time 0.020469 -2022-12-06 11:29:18,720 - Epoch: [149][ 640/ 1200] Overall Loss 0.154437 Objective Loss 0.154437 LR 0.000250 Time 0.020449 -2022-12-06 11:29:18,912 - Epoch: [149][ 650/ 1200] Overall Loss 0.153910 Objective Loss 0.153910 LR 0.000250 Time 0.020429 -2022-12-06 11:29:19,104 - Epoch: [149][ 660/ 1200] Overall Loss 0.154119 Objective Loss 0.154119 LR 0.000250 Time 0.020410 -2022-12-06 11:29:19,296 - Epoch: [149][ 670/ 1200] Overall Loss 0.154163 Objective Loss 0.154163 LR 0.000250 Time 0.020391 -2022-12-06 11:29:19,489 - Epoch: [149][ 680/ 1200] Overall Loss 0.153847 Objective Loss 0.153847 LR 0.000250 Time 0.020373 -2022-12-06 11:29:19,681 - Epoch: [149][ 690/ 1200] Overall Loss 0.154047 Objective Loss 0.154047 LR 0.000250 Time 0.020356 -2022-12-06 11:29:19,873 - Epoch: [149][ 700/ 1200] Overall Loss 0.153878 Objective Loss 0.153878 LR 0.000250 Time 0.020339 -2022-12-06 11:29:20,065 - Epoch: [149][ 710/ 1200] Overall Loss 0.154279 Objective Loss 0.154279 LR 0.000250 Time 0.020321 -2022-12-06 11:29:20,257 - Epoch: [149][ 720/ 1200] Overall Loss 0.154328 Objective Loss 0.154328 LR 0.000250 Time 0.020305 -2022-12-06 11:29:20,448 - Epoch: [149][ 730/ 1200] Overall Loss 0.154513 Objective Loss 0.154513 LR 0.000250 Time 0.020289 -2022-12-06 11:29:20,641 - Epoch: [149][ 740/ 1200] Overall Loss 0.154786 Objective Loss 0.154786 LR 0.000250 Time 0.020273 -2022-12-06 11:29:20,833 - Epoch: [149][ 750/ 1200] Overall Loss 0.155211 Objective Loss 0.155211 LR 0.000250 Time 0.020259 -2022-12-06 11:29:21,025 - Epoch: [149][ 760/ 1200] Overall Loss 0.155278 Objective Loss 0.155278 LR 0.000250 Time 0.020245 -2022-12-06 11:29:21,217 - Epoch: [149][ 770/ 1200] Overall Loss 0.155643 Objective Loss 0.155643 LR 0.000250 Time 0.020230 -2022-12-06 11:29:21,409 - Epoch: [149][ 780/ 1200] Overall Loss 0.155748 Objective Loss 0.155748 LR 0.000250 Time 0.020216 -2022-12-06 11:29:21,601 - Epoch: [149][ 790/ 1200] Overall Loss 0.155876 Objective Loss 0.155876 LR 0.000250 Time 0.020202 -2022-12-06 11:29:21,793 - Epoch: [149][ 800/ 1200] Overall Loss 0.155992 Objective Loss 0.155992 LR 0.000250 Time 0.020189 -2022-12-06 11:29:21,985 - Epoch: [149][ 810/ 1200] Overall Loss 0.156179 Objective Loss 0.156179 LR 0.000250 Time 0.020176 -2022-12-06 11:29:22,176 - Epoch: [149][ 820/ 1200] Overall Loss 0.156288 Objective Loss 0.156288 LR 0.000250 Time 0.020162 -2022-12-06 11:29:22,368 - Epoch: [149][ 830/ 1200] Overall Loss 0.156370 Objective Loss 0.156370 LR 0.000250 Time 0.020150 -2022-12-06 11:29:22,560 - Epoch: [149][ 840/ 1200] Overall Loss 0.156508 Objective Loss 0.156508 LR 0.000250 Time 0.020138 -2022-12-06 11:29:22,751 - Epoch: [149][ 850/ 1200] Overall Loss 0.156474 Objective Loss 0.156474 LR 0.000250 Time 0.020126 -2022-12-06 11:29:22,944 - Epoch: [149][ 860/ 1200] Overall Loss 0.156505 Objective Loss 0.156505 LR 0.000250 Time 0.020115 -2022-12-06 11:29:23,136 - Epoch: [149][ 870/ 1200] Overall Loss 0.156425 Objective Loss 0.156425 LR 0.000250 Time 0.020104 -2022-12-06 11:29:23,328 - Epoch: [149][ 880/ 1200] Overall Loss 0.156551 Objective Loss 0.156551 LR 0.000250 Time 0.020093 -2022-12-06 11:29:23,520 - Epoch: [149][ 890/ 1200] Overall Loss 0.156613 Objective Loss 0.156613 LR 0.000250 Time 0.020082 -2022-12-06 11:29:23,712 - Epoch: [149][ 900/ 1200] Overall Loss 0.156564 Objective Loss 0.156564 LR 0.000250 Time 0.020072 -2022-12-06 11:29:23,904 - Epoch: [149][ 910/ 1200] Overall Loss 0.156540 Objective Loss 0.156540 LR 0.000250 Time 0.020062 -2022-12-06 11:29:24,096 - Epoch: [149][ 920/ 1200] Overall Loss 0.156678 Objective Loss 0.156678 LR 0.000250 Time 0.020052 -2022-12-06 11:29:24,288 - Epoch: [149][ 930/ 1200] Overall Loss 0.156483 Objective Loss 0.156483 LR 0.000250 Time 0.020042 -2022-12-06 11:29:24,480 - Epoch: [149][ 940/ 1200] Overall Loss 0.156448 Objective Loss 0.156448 LR 0.000250 Time 0.020032 -2022-12-06 11:29:24,672 - Epoch: [149][ 950/ 1200] Overall Loss 0.156452 Objective Loss 0.156452 LR 0.000250 Time 0.020024 -2022-12-06 11:29:24,864 - Epoch: [149][ 960/ 1200] Overall Loss 0.156370 Objective Loss 0.156370 LR 0.000250 Time 0.020015 -2022-12-06 11:29:25,056 - Epoch: [149][ 970/ 1200] Overall Loss 0.156178 Objective Loss 0.156178 LR 0.000250 Time 0.020005 -2022-12-06 11:29:25,248 - Epoch: [149][ 980/ 1200] Overall Loss 0.156259 Objective Loss 0.156259 LR 0.000250 Time 0.019997 -2022-12-06 11:29:25,441 - Epoch: [149][ 990/ 1200] Overall Loss 0.156226 Objective Loss 0.156226 LR 0.000250 Time 0.019989 -2022-12-06 11:29:25,633 - Epoch: [149][ 1000/ 1200] Overall Loss 0.156402 Objective Loss 0.156402 LR 0.000250 Time 0.019981 -2022-12-06 11:29:25,826 - Epoch: [149][ 1010/ 1200] Overall Loss 0.156350 Objective Loss 0.156350 LR 0.000250 Time 0.019973 -2022-12-06 11:29:26,018 - Epoch: [149][ 1020/ 1200] Overall Loss 0.156071 Objective Loss 0.156071 LR 0.000250 Time 0.019965 -2022-12-06 11:29:26,210 - Epoch: [149][ 1030/ 1200] Overall Loss 0.156114 Objective Loss 0.156114 LR 0.000250 Time 0.019957 -2022-12-06 11:29:26,402 - Epoch: [149][ 1040/ 1200] Overall Loss 0.156077 Objective Loss 0.156077 LR 0.000250 Time 0.019949 -2022-12-06 11:29:26,595 - Epoch: [149][ 1050/ 1200] Overall Loss 0.156027 Objective Loss 0.156027 LR 0.000250 Time 0.019942 -2022-12-06 11:29:26,787 - Epoch: [149][ 1060/ 1200] Overall Loss 0.155913 Objective Loss 0.155913 LR 0.000250 Time 0.019935 -2022-12-06 11:29:26,980 - Epoch: [149][ 1070/ 1200] Overall Loss 0.155939 Objective Loss 0.155939 LR 0.000250 Time 0.019928 -2022-12-06 11:29:27,172 - Epoch: [149][ 1080/ 1200] Overall Loss 0.156033 Objective Loss 0.156033 LR 0.000250 Time 0.019921 -2022-12-06 11:29:27,364 - Epoch: [149][ 1090/ 1200] Overall Loss 0.156030 Objective Loss 0.156030 LR 0.000250 Time 0.019914 -2022-12-06 11:29:27,556 - Epoch: [149][ 1100/ 1200] Overall Loss 0.156164 Objective Loss 0.156164 LR 0.000250 Time 0.019907 -2022-12-06 11:29:27,747 - Epoch: [149][ 1110/ 1200] Overall Loss 0.156249 Objective Loss 0.156249 LR 0.000250 Time 0.019900 -2022-12-06 11:29:27,940 - Epoch: [149][ 1120/ 1200] Overall Loss 0.156261 Objective Loss 0.156261 LR 0.000250 Time 0.019893 -2022-12-06 11:29:28,131 - Epoch: [149][ 1130/ 1200] Overall Loss 0.156137 Objective Loss 0.156137 LR 0.000250 Time 0.019886 -2022-12-06 11:29:28,324 - Epoch: [149][ 1140/ 1200] Overall Loss 0.156147 Objective Loss 0.156147 LR 0.000250 Time 0.019880 -2022-12-06 11:29:28,515 - Epoch: [149][ 1150/ 1200] Overall Loss 0.156220 Objective Loss 0.156220 LR 0.000250 Time 0.019873 -2022-12-06 11:29:28,708 - Epoch: [149][ 1160/ 1200] Overall Loss 0.156313 Objective Loss 0.156313 LR 0.000250 Time 0.019867 -2022-12-06 11:29:28,899 - Epoch: [149][ 1170/ 1200] Overall Loss 0.156328 Objective Loss 0.156328 LR 0.000250 Time 0.019860 -2022-12-06 11:29:29,090 - Epoch: [149][ 1180/ 1200] Overall Loss 0.156235 Objective Loss 0.156235 LR 0.000250 Time 0.019854 -2022-12-06 11:29:29,282 - Epoch: [149][ 1190/ 1200] Overall Loss 0.156310 Objective Loss 0.156310 LR 0.000250 Time 0.019847 -2022-12-06 11:29:29,515 - Epoch: [149][ 1200/ 1200] Overall Loss 0.156217 Objective Loss 0.156217 Top1 90.167364 Top5 99.163180 LR 0.000250 Time 0.019876 -2022-12-06 11:29:29,603 - --- validate (epoch=149)----------- -2022-12-06 11:29:29,604 - 34129 samples (256 per mini-batch) -2022-12-06 11:29:30,051 - Epoch: [149][ 10/ 134] Loss 0.234647 Top1 87.968750 Top5 98.554688 -2022-12-06 11:29:30,182 - Epoch: [149][ 20/ 134] Loss 0.247896 Top1 88.046875 Top5 98.515625 -2022-12-06 11:29:30,313 - Epoch: [149][ 30/ 134] Loss 0.244341 Top1 88.203125 Top5 98.502604 -2022-12-06 11:29:30,446 - Epoch: [149][ 40/ 134] Loss 0.247916 Top1 88.027344 Top5 98.427734 -2022-12-06 11:29:30,573 - Epoch: [149][ 50/ 134] Loss 0.242340 Top1 88.203125 Top5 98.476562 -2022-12-06 11:29:30,713 - Epoch: [149][ 60/ 134] Loss 0.240776 Top1 88.261719 Top5 98.444010 -2022-12-06 11:29:30,841 - Epoch: [149][ 70/ 134] Loss 0.238333 Top1 88.337054 Top5 98.431920 -2022-12-06 11:29:30,971 - Epoch: [149][ 80/ 134] Loss 0.237228 Top1 88.261719 Top5 98.457031 -2022-12-06 11:29:31,104 - Epoch: [149][ 90/ 134] Loss 0.236725 Top1 88.181424 Top5 98.459201 -2022-12-06 11:29:31,235 - Epoch: [149][ 100/ 134] Loss 0.238829 Top1 88.046875 Top5 98.464844 -2022-12-06 11:29:31,367 - Epoch: [149][ 110/ 134] Loss 0.237694 Top1 87.997159 Top5 98.451705 -2022-12-06 11:29:31,498 - Epoch: [149][ 120/ 134] Loss 0.238692 Top1 87.945964 Top5 98.453776 -2022-12-06 11:29:31,632 - Epoch: [149][ 130/ 134] Loss 0.236495 Top1 87.989784 Top5 98.491587 -2022-12-06 11:29:31,671 - Epoch: [149][ 134/ 134] Loss 0.236830 Top1 88.010197 Top5 98.499810 -2022-12-06 11:29:31,758 - ==> Top1: 88.010 Top5: 98.500 Loss: 0.237 - -2022-12-06 11:29:31,759 - ==> Confusion: -[[ 897 0 1 2 4 3 0 0 7 63 0 3 1 3 4 1 1 0 1 0 5] - [ 0 939 3 2 8 16 1 12 1 0 5 5 1 2 1 0 6 2 10 4 9] - [ 3 4 1032 9 2 1 9 7 0 3 4 5 1 0 3 2 1 1 2 4 10] - [ 1 2 23 946 1 2 0 1 1 1 7 0 2 3 10 1 0 1 11 0 7] - [ 6 5 1 0 962 1 1 2 0 7 2 2 0 3 8 3 7 2 0 2 6] - [ 0 16 0 2 6 970 3 22 1 3 1 14 4 13 1 1 0 1 0 5 6] - [ 1 1 11 2 2 1 1071 2 0 0 0 1 0 0 0 8 0 1 3 12 2] - [ 0 6 8 2 1 22 5 958 0 0 0 5 0 2 1 0 0 0 22 13 9] - [ 6 2 0 0 0 2 1 0 987 32 12 1 1 8 6 0 2 1 1 2 0] - [ 48 0 2 0 3 1 0 2 27 894 1 2 0 11 3 1 0 1 0 0 5] - [ 1 0 6 5 0 0 0 4 8 3 964 2 0 9 2 1 1 0 3 2 8] - [ 2 0 1 0 0 7 4 3 1 0 0 986 17 5 0 5 3 4 0 10 3] - [ 0 1 3 3 0 2 0 1 0 0 0 32 897 1 0 5 3 11 0 4 6] - [ 2 1 1 0 1 5 0 3 10 12 3 4 3 963 2 0 2 0 0 3 8] - [ 5 2 2 11 5 0 0 1 18 6 1 4 2 3 1057 0 0 0 7 0 6] - [ 0 1 1 1 1 0 3 0 0 1 0 8 6 3 0 994 7 12 1 2 2] - [ 3 2 1 0 2 1 0 1 0 0 0 5 1 3 1 12 1026 0 0 8 6] - [ 4 0 1 2 1 1 1 0 0 2 0 5 13 1 2 11 0 986 0 2 4] - [ 1 6 5 8 0 2 0 20 1 1 4 2 3 1 7 0 1 3 939 1 3] - [ 1 3 1 1 0 3 3 5 0 0 3 13 5 7 0 4 3 2 0 1023 3] - [ 116 163 169 85 98 124 63 130 83 89 120 102 290 241 135 93 141 78 128 236 10542]] - -2022-12-06 11:29:32,339 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:29:32,339 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:29:32,346 - - -2022-12-06 11:29:32,346 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:29:33,282 - Epoch: [150][ 10/ 1200] Overall Loss 0.158332 Objective Loss 0.158332 LR 0.000250 Time 0.093500 -2022-12-06 11:29:33,489 - Epoch: [150][ 20/ 1200] Overall Loss 0.157038 Objective Loss 0.157038 LR 0.000250 Time 0.057061 -2022-12-06 11:29:33,690 - Epoch: [150][ 30/ 1200] Overall Loss 0.162256 Objective Loss 0.162256 LR 0.000250 Time 0.044717 -2022-12-06 11:29:33,887 - Epoch: [150][ 40/ 1200] Overall Loss 0.156314 Objective Loss 0.156314 LR 0.000250 Time 0.038450 -2022-12-06 11:29:34,086 - Epoch: [150][ 50/ 1200] Overall Loss 0.150504 Objective Loss 0.150504 LR 0.000250 Time 0.034742 -2022-12-06 11:29:34,283 - Epoch: [150][ 60/ 1200] Overall Loss 0.151083 Objective Loss 0.151083 LR 0.000250 Time 0.032224 -2022-12-06 11:29:34,482 - Epoch: [150][ 70/ 1200] Overall Loss 0.149282 Objective Loss 0.149282 LR 0.000250 Time 0.030460 -2022-12-06 11:29:34,679 - Epoch: [150][ 80/ 1200] Overall Loss 0.151068 Objective Loss 0.151068 LR 0.000250 Time 0.029098 -2022-12-06 11:29:34,878 - Epoch: [150][ 90/ 1200] Overall Loss 0.150304 Objective Loss 0.150304 LR 0.000250 Time 0.028075 -2022-12-06 11:29:35,076 - Epoch: [150][ 100/ 1200] Overall Loss 0.150916 Objective Loss 0.150916 LR 0.000250 Time 0.027242 -2022-12-06 11:29:35,279 - Epoch: [150][ 110/ 1200] Overall Loss 0.149688 Objective Loss 0.149688 LR 0.000250 Time 0.026607 -2022-12-06 11:29:35,477 - Epoch: [150][ 120/ 1200] Overall Loss 0.148993 Objective Loss 0.148993 LR 0.000250 Time 0.026036 -2022-12-06 11:29:35,679 - Epoch: [150][ 130/ 1200] Overall Loss 0.149130 Objective Loss 0.149130 LR 0.000250 Time 0.025583 -2022-12-06 11:29:35,878 - Epoch: [150][ 140/ 1200] Overall Loss 0.148384 Objective Loss 0.148384 LR 0.000250 Time 0.025171 -2022-12-06 11:29:36,080 - Epoch: [150][ 150/ 1200] Overall Loss 0.147299 Objective Loss 0.147299 LR 0.000250 Time 0.024833 -2022-12-06 11:29:36,277 - Epoch: [150][ 160/ 1200] Overall Loss 0.146548 Objective Loss 0.146548 LR 0.000250 Time 0.024508 -2022-12-06 11:29:36,477 - Epoch: [150][ 170/ 1200] Overall Loss 0.146998 Objective Loss 0.146998 LR 0.000250 Time 0.024241 -2022-12-06 11:29:36,673 - Epoch: [150][ 180/ 1200] Overall Loss 0.146422 Objective Loss 0.146422 LR 0.000250 Time 0.023981 -2022-12-06 11:29:36,872 - Epoch: [150][ 190/ 1200] Overall Loss 0.147058 Objective Loss 0.147058 LR 0.000250 Time 0.023765 -2022-12-06 11:29:37,069 - Epoch: [150][ 200/ 1200] Overall Loss 0.147090 Objective Loss 0.147090 LR 0.000250 Time 0.023556 -2022-12-06 11:29:37,268 - Epoch: [150][ 210/ 1200] Overall Loss 0.147216 Objective Loss 0.147216 LR 0.000250 Time 0.023380 -2022-12-06 11:29:37,464 - Epoch: [150][ 220/ 1200] Overall Loss 0.146601 Objective Loss 0.146601 LR 0.000250 Time 0.023207 -2022-12-06 11:29:37,664 - Epoch: [150][ 230/ 1200] Overall Loss 0.146444 Objective Loss 0.146444 LR 0.000250 Time 0.023064 -2022-12-06 11:29:37,860 - Epoch: [150][ 240/ 1200] Overall Loss 0.146956 Objective Loss 0.146956 LR 0.000250 Time 0.022919 -2022-12-06 11:29:38,059 - Epoch: [150][ 250/ 1200] Overall Loss 0.147211 Objective Loss 0.147211 LR 0.000250 Time 0.022796 -2022-12-06 11:29:38,256 - Epoch: [150][ 260/ 1200] Overall Loss 0.147815 Objective Loss 0.147815 LR 0.000250 Time 0.022674 -2022-12-06 11:29:38,456 - Epoch: [150][ 270/ 1200] Overall Loss 0.147717 Objective Loss 0.147717 LR 0.000250 Time 0.022572 -2022-12-06 11:29:38,651 - Epoch: [150][ 280/ 1200] Overall Loss 0.148557 Objective Loss 0.148557 LR 0.000250 Time 0.022462 -2022-12-06 11:29:38,851 - Epoch: [150][ 290/ 1200] Overall Loss 0.148558 Objective Loss 0.148558 LR 0.000250 Time 0.022374 -2022-12-06 11:29:39,047 - Epoch: [150][ 300/ 1200] Overall Loss 0.148640 Objective Loss 0.148640 LR 0.000250 Time 0.022281 -2022-12-06 11:29:39,246 - Epoch: [150][ 310/ 1200] Overall Loss 0.148987 Objective Loss 0.148987 LR 0.000250 Time 0.022201 -2022-12-06 11:29:39,442 - Epoch: [150][ 320/ 1200] Overall Loss 0.148285 Objective Loss 0.148285 LR 0.000250 Time 0.022118 -2022-12-06 11:29:39,641 - Epoch: [150][ 330/ 1200] Overall Loss 0.148321 Objective Loss 0.148321 LR 0.000250 Time 0.022049 -2022-12-06 11:29:39,838 - Epoch: [150][ 340/ 1200] Overall Loss 0.148305 Objective Loss 0.148305 LR 0.000250 Time 0.021977 -2022-12-06 11:29:40,037 - Epoch: [150][ 350/ 1200] Overall Loss 0.148040 Objective Loss 0.148040 LR 0.000250 Time 0.021918 -2022-12-06 11:29:40,233 - Epoch: [150][ 360/ 1200] Overall Loss 0.148051 Objective Loss 0.148051 LR 0.000250 Time 0.021852 -2022-12-06 11:29:40,433 - Epoch: [150][ 370/ 1200] Overall Loss 0.147610 Objective Loss 0.147610 LR 0.000250 Time 0.021800 -2022-12-06 11:29:40,629 - Epoch: [150][ 380/ 1200] Overall Loss 0.147312 Objective Loss 0.147312 LR 0.000250 Time 0.021742 -2022-12-06 11:29:40,829 - Epoch: [150][ 390/ 1200] Overall Loss 0.148341 Objective Loss 0.148341 LR 0.000250 Time 0.021695 -2022-12-06 11:29:41,026 - Epoch: [150][ 400/ 1200] Overall Loss 0.148118 Objective Loss 0.148118 LR 0.000250 Time 0.021642 -2022-12-06 11:29:41,225 - Epoch: [150][ 410/ 1200] Overall Loss 0.148321 Objective Loss 0.148321 LR 0.000250 Time 0.021601 -2022-12-06 11:29:41,421 - Epoch: [150][ 420/ 1200] Overall Loss 0.148372 Objective Loss 0.148372 LR 0.000250 Time 0.021552 -2022-12-06 11:29:41,622 - Epoch: [150][ 430/ 1200] Overall Loss 0.148404 Objective Loss 0.148404 LR 0.000250 Time 0.021515 -2022-12-06 11:29:41,818 - Epoch: [150][ 440/ 1200] Overall Loss 0.148460 Objective Loss 0.148460 LR 0.000250 Time 0.021470 -2022-12-06 11:29:42,018 - Epoch: [150][ 450/ 1200] Overall Loss 0.148616 Objective Loss 0.148616 LR 0.000250 Time 0.021436 -2022-12-06 11:29:42,214 - Epoch: [150][ 460/ 1200] Overall Loss 0.148823 Objective Loss 0.148823 LR 0.000250 Time 0.021396 -2022-12-06 11:29:42,414 - Epoch: [150][ 470/ 1200] Overall Loss 0.148784 Objective Loss 0.148784 LR 0.000250 Time 0.021364 -2022-12-06 11:29:42,610 - Epoch: [150][ 480/ 1200] Overall Loss 0.148864 Objective Loss 0.148864 LR 0.000250 Time 0.021327 -2022-12-06 11:29:42,810 - Epoch: [150][ 490/ 1200] Overall Loss 0.149102 Objective Loss 0.149102 LR 0.000250 Time 0.021298 -2022-12-06 11:29:43,006 - Epoch: [150][ 500/ 1200] Overall Loss 0.149148 Objective Loss 0.149148 LR 0.000250 Time 0.021264 -2022-12-06 11:29:43,205 - Epoch: [150][ 510/ 1200] Overall Loss 0.149201 Objective Loss 0.149201 LR 0.000250 Time 0.021237 -2022-12-06 11:29:43,402 - Epoch: [150][ 520/ 1200] Overall Loss 0.149427 Objective Loss 0.149427 LR 0.000250 Time 0.021205 -2022-12-06 11:29:43,602 - Epoch: [150][ 530/ 1200] Overall Loss 0.149834 Objective Loss 0.149834 LR 0.000250 Time 0.021181 -2022-12-06 11:29:43,798 - Epoch: [150][ 540/ 1200] Overall Loss 0.149505 Objective Loss 0.149505 LR 0.000250 Time 0.021151 -2022-12-06 11:29:43,997 - Epoch: [150][ 550/ 1200] Overall Loss 0.149257 Objective Loss 0.149257 LR 0.000250 Time 0.021127 -2022-12-06 11:29:44,192 - Epoch: [150][ 560/ 1200] Overall Loss 0.149115 Objective Loss 0.149115 LR 0.000250 Time 0.021098 -2022-12-06 11:29:44,392 - Epoch: [150][ 570/ 1200] Overall Loss 0.149083 Objective Loss 0.149083 LR 0.000250 Time 0.021076 -2022-12-06 11:29:44,588 - Epoch: [150][ 580/ 1200] Overall Loss 0.149287 Objective Loss 0.149287 LR 0.000250 Time 0.021051 -2022-12-06 11:29:44,787 - Epoch: [150][ 590/ 1200] Overall Loss 0.149425 Objective Loss 0.149425 LR 0.000250 Time 0.021030 -2022-12-06 11:29:44,983 - Epoch: [150][ 600/ 1200] Overall Loss 0.149596 Objective Loss 0.149596 LR 0.000250 Time 0.021006 -2022-12-06 11:29:45,183 - Epoch: [150][ 610/ 1200] Overall Loss 0.149624 Objective Loss 0.149624 LR 0.000250 Time 0.020988 -2022-12-06 11:29:45,379 - Epoch: [150][ 620/ 1200] Overall Loss 0.149371 Objective Loss 0.149371 LR 0.000250 Time 0.020964 -2022-12-06 11:29:45,578 - Epoch: [150][ 630/ 1200] Overall Loss 0.149346 Objective Loss 0.149346 LR 0.000250 Time 0.020947 -2022-12-06 11:29:45,774 - Epoch: [150][ 640/ 1200] Overall Loss 0.149347 Objective Loss 0.149347 LR 0.000250 Time 0.020925 -2022-12-06 11:29:45,972 - Epoch: [150][ 650/ 1200] Overall Loss 0.149584 Objective Loss 0.149584 LR 0.000250 Time 0.020908 -2022-12-06 11:29:46,169 - Epoch: [150][ 660/ 1200] Overall Loss 0.149667 Objective Loss 0.149667 LR 0.000250 Time 0.020887 -2022-12-06 11:29:46,368 - Epoch: [150][ 670/ 1200] Overall Loss 0.149645 Objective Loss 0.149645 LR 0.000250 Time 0.020872 -2022-12-06 11:29:46,564 - Epoch: [150][ 680/ 1200] Overall Loss 0.149366 Objective Loss 0.149366 LR 0.000250 Time 0.020852 -2022-12-06 11:29:46,763 - Epoch: [150][ 690/ 1200] Overall Loss 0.149431 Objective Loss 0.149431 LR 0.000250 Time 0.020838 -2022-12-06 11:29:46,959 - Epoch: [150][ 700/ 1200] Overall Loss 0.149378 Objective Loss 0.149378 LR 0.000250 Time 0.020820 -2022-12-06 11:29:47,158 - Epoch: [150][ 710/ 1200] Overall Loss 0.149783 Objective Loss 0.149783 LR 0.000250 Time 0.020806 -2022-12-06 11:29:47,354 - Epoch: [150][ 720/ 1200] Overall Loss 0.149827 Objective Loss 0.149827 LR 0.000250 Time 0.020788 -2022-12-06 11:29:47,554 - Epoch: [150][ 730/ 1200] Overall Loss 0.149977 Objective Loss 0.149977 LR 0.000250 Time 0.020777 -2022-12-06 11:29:47,750 - Epoch: [150][ 740/ 1200] Overall Loss 0.150020 Objective Loss 0.150020 LR 0.000250 Time 0.020760 -2022-12-06 11:29:47,949 - Epoch: [150][ 750/ 1200] Overall Loss 0.150217 Objective Loss 0.150217 LR 0.000250 Time 0.020748 -2022-12-06 11:29:48,145 - Epoch: [150][ 760/ 1200] Overall Loss 0.150428 Objective Loss 0.150428 LR 0.000250 Time 0.020732 -2022-12-06 11:29:48,344 - Epoch: [150][ 770/ 1200] Overall Loss 0.150458 Objective Loss 0.150458 LR 0.000250 Time 0.020720 -2022-12-06 11:29:48,540 - Epoch: [150][ 780/ 1200] Overall Loss 0.150365 Objective Loss 0.150365 LR 0.000250 Time 0.020705 -2022-12-06 11:29:48,739 - Epoch: [150][ 790/ 1200] Overall Loss 0.150422 Objective Loss 0.150422 LR 0.000250 Time 0.020694 -2022-12-06 11:29:48,935 - Epoch: [150][ 800/ 1200] Overall Loss 0.150510 Objective Loss 0.150510 LR 0.000250 Time 0.020680 -2022-12-06 11:29:49,133 - Epoch: [150][ 810/ 1200] Overall Loss 0.150633 Objective Loss 0.150633 LR 0.000250 Time 0.020669 -2022-12-06 11:29:49,329 - Epoch: [150][ 820/ 1200] Overall Loss 0.150810 Objective Loss 0.150810 LR 0.000250 Time 0.020655 -2022-12-06 11:29:49,528 - Epoch: [150][ 830/ 1200] Overall Loss 0.150999 Objective Loss 0.150999 LR 0.000250 Time 0.020645 -2022-12-06 11:29:49,724 - Epoch: [150][ 840/ 1200] Overall Loss 0.151101 Objective Loss 0.151101 LR 0.000250 Time 0.020632 -2022-12-06 11:29:49,923 - Epoch: [150][ 850/ 1200] Overall Loss 0.151169 Objective Loss 0.151169 LR 0.000250 Time 0.020623 -2022-12-06 11:29:50,119 - Epoch: [150][ 860/ 1200] Overall Loss 0.151181 Objective Loss 0.151181 LR 0.000250 Time 0.020610 -2022-12-06 11:29:50,318 - Epoch: [150][ 870/ 1200] Overall Loss 0.151197 Objective Loss 0.151197 LR 0.000250 Time 0.020602 -2022-12-06 11:29:50,514 - Epoch: [150][ 880/ 1200] Overall Loss 0.151373 Objective Loss 0.151373 LR 0.000250 Time 0.020590 -2022-12-06 11:29:50,713 - Epoch: [150][ 890/ 1200] Overall Loss 0.151510 Objective Loss 0.151510 LR 0.000250 Time 0.020581 -2022-12-06 11:29:50,909 - Epoch: [150][ 900/ 1200] Overall Loss 0.151515 Objective Loss 0.151515 LR 0.000250 Time 0.020569 -2022-12-06 11:29:51,108 - Epoch: [150][ 910/ 1200] Overall Loss 0.151721 Objective Loss 0.151721 LR 0.000250 Time 0.020562 -2022-12-06 11:29:51,304 - Epoch: [150][ 920/ 1200] Overall Loss 0.151939 Objective Loss 0.151939 LR 0.000250 Time 0.020550 -2022-12-06 11:29:51,503 - Epoch: [150][ 930/ 1200] Overall Loss 0.151987 Objective Loss 0.151987 LR 0.000250 Time 0.020543 -2022-12-06 11:29:51,699 - Epoch: [150][ 940/ 1200] Overall Loss 0.152265 Objective Loss 0.152265 LR 0.000250 Time 0.020532 -2022-12-06 11:29:51,898 - Epoch: [150][ 950/ 1200] Overall Loss 0.152675 Objective Loss 0.152675 LR 0.000250 Time 0.020526 -2022-12-06 11:29:52,094 - Epoch: [150][ 960/ 1200] Overall Loss 0.152589 Objective Loss 0.152589 LR 0.000250 Time 0.020515 -2022-12-06 11:29:52,293 - Epoch: [150][ 970/ 1200] Overall Loss 0.152577 Objective Loss 0.152577 LR 0.000250 Time 0.020508 -2022-12-06 11:29:52,489 - Epoch: [150][ 980/ 1200] Overall Loss 0.152851 Objective Loss 0.152851 LR 0.000250 Time 0.020498 -2022-12-06 11:29:52,688 - Epoch: [150][ 990/ 1200] Overall Loss 0.153052 Objective Loss 0.153052 LR 0.000250 Time 0.020492 -2022-12-06 11:29:52,884 - Epoch: [150][ 1000/ 1200] Overall Loss 0.153069 Objective Loss 0.153069 LR 0.000250 Time 0.020482 -2022-12-06 11:29:53,083 - Epoch: [150][ 1010/ 1200] Overall Loss 0.152955 Objective Loss 0.152955 LR 0.000250 Time 0.020476 -2022-12-06 11:29:53,279 - Epoch: [150][ 1020/ 1200] Overall Loss 0.153123 Objective Loss 0.153123 LR 0.000250 Time 0.020467 -2022-12-06 11:29:53,478 - Epoch: [150][ 1030/ 1200] Overall Loss 0.153094 Objective Loss 0.153094 LR 0.000250 Time 0.020461 -2022-12-06 11:29:53,674 - Epoch: [150][ 1040/ 1200] Overall Loss 0.153369 Objective Loss 0.153369 LR 0.000250 Time 0.020452 -2022-12-06 11:29:53,874 - Epoch: [150][ 1050/ 1200] Overall Loss 0.153412 Objective Loss 0.153412 LR 0.000250 Time 0.020447 -2022-12-06 11:29:54,069 - Epoch: [150][ 1060/ 1200] Overall Loss 0.153335 Objective Loss 0.153335 LR 0.000250 Time 0.020438 -2022-12-06 11:29:54,268 - Epoch: [150][ 1070/ 1200] Overall Loss 0.153068 Objective Loss 0.153068 LR 0.000250 Time 0.020432 -2022-12-06 11:29:54,465 - Epoch: [150][ 1080/ 1200] Overall Loss 0.152981 Objective Loss 0.152981 LR 0.000250 Time 0.020424 -2022-12-06 11:29:54,664 - Epoch: [150][ 1090/ 1200] Overall Loss 0.152998 Objective Loss 0.152998 LR 0.000250 Time 0.020419 -2022-12-06 11:29:54,860 - Epoch: [150][ 1100/ 1200] Overall Loss 0.153125 Objective Loss 0.153125 LR 0.000250 Time 0.020411 -2022-12-06 11:29:55,059 - Epoch: [150][ 1110/ 1200] Overall Loss 0.153079 Objective Loss 0.153079 LR 0.000250 Time 0.020406 -2022-12-06 11:29:55,254 - Epoch: [150][ 1120/ 1200] Overall Loss 0.153086 Objective Loss 0.153086 LR 0.000250 Time 0.020398 -2022-12-06 11:29:55,452 - Epoch: [150][ 1130/ 1200] Overall Loss 0.153227 Objective Loss 0.153227 LR 0.000250 Time 0.020392 -2022-12-06 11:29:55,649 - Epoch: [150][ 1140/ 1200] Overall Loss 0.153206 Objective Loss 0.153206 LR 0.000250 Time 0.020385 -2022-12-06 11:29:55,847 - Epoch: [150][ 1150/ 1200] Overall Loss 0.152979 Objective Loss 0.152979 LR 0.000250 Time 0.020380 -2022-12-06 11:29:56,043 - Epoch: [150][ 1160/ 1200] Overall Loss 0.152943 Objective Loss 0.152943 LR 0.000250 Time 0.020372 -2022-12-06 11:29:56,243 - Epoch: [150][ 1170/ 1200] Overall Loss 0.152875 Objective Loss 0.152875 LR 0.000250 Time 0.020369 -2022-12-06 11:29:56,438 - Epoch: [150][ 1180/ 1200] Overall Loss 0.152938 Objective Loss 0.152938 LR 0.000250 Time 0.020362 -2022-12-06 11:29:56,638 - Epoch: [150][ 1190/ 1200] Overall Loss 0.152871 Objective Loss 0.152871 LR 0.000250 Time 0.020357 -2022-12-06 11:29:56,867 - Epoch: [150][ 1200/ 1200] Overall Loss 0.153004 Objective Loss 0.153004 Top1 88.075314 Top5 98.953975 LR 0.000250 Time 0.020379 -2022-12-06 11:29:56,956 - --- validate (epoch=150)----------- -2022-12-06 11:29:56,956 - 34129 samples (256 per mini-batch) -2022-12-06 11:29:57,406 - Epoch: [150][ 10/ 134] Loss 0.235478 Top1 88.046875 Top5 98.593750 -2022-12-06 11:29:57,536 - Epoch: [150][ 20/ 134] Loss 0.243240 Top1 87.539062 Top5 98.437500 -2022-12-06 11:29:57,664 - Epoch: [150][ 30/ 134] Loss 0.229663 Top1 87.890625 Top5 98.697917 -2022-12-06 11:29:57,798 - Epoch: [150][ 40/ 134] Loss 0.223876 Top1 87.978516 Top5 98.554688 -2022-12-06 11:29:57,931 - Epoch: [150][ 50/ 134] Loss 0.224074 Top1 88.062500 Top5 98.539062 -2022-12-06 11:29:58,064 - Epoch: [150][ 60/ 134] Loss 0.228887 Top1 87.955729 Top5 98.600260 -2022-12-06 11:29:58,197 - Epoch: [150][ 70/ 134] Loss 0.234340 Top1 87.834821 Top5 98.565848 -2022-12-06 11:29:58,329 - Epoch: [150][ 80/ 134] Loss 0.236455 Top1 87.749023 Top5 98.535156 -2022-12-06 11:29:58,463 - Epoch: [150][ 90/ 134] Loss 0.236575 Top1 87.756076 Top5 98.546007 -2022-12-06 11:29:58,596 - Epoch: [150][ 100/ 134] Loss 0.236566 Top1 87.769531 Top5 98.535156 -2022-12-06 11:29:58,730 - Epoch: [150][ 110/ 134] Loss 0.235170 Top1 87.727273 Top5 98.533381 -2022-12-06 11:29:58,863 - Epoch: [150][ 120/ 134] Loss 0.236455 Top1 87.672526 Top5 98.551432 -2022-12-06 11:29:58,997 - Epoch: [150][ 130/ 134] Loss 0.234103 Top1 87.773438 Top5 98.560697 -2022-12-06 11:29:59,036 - Epoch: [150][ 134/ 134] Loss 0.234004 Top1 87.790442 Top5 98.573061 -2022-12-06 11:29:59,126 - ==> Top1: 87.790 Top5: 98.573 Loss: 0.234 - -2022-12-06 11:29:59,127 - ==> Confusion: -[[ 914 0 0 1 2 7 1 1 4 46 0 2 1 4 3 1 1 0 1 0 7] - [ 1 941 1 2 7 24 1 12 1 2 3 4 1 1 0 1 3 2 8 3 9] - [ 3 3 1005 13 5 3 14 13 0 6 9 5 0 1 2 1 0 3 1 5 11] - [ 2 0 13 957 2 2 1 0 0 1 10 1 3 0 9 0 1 4 8 0 6] - [ 9 7 1 0 954 3 0 3 1 8 1 1 2 2 10 2 7 3 0 0 6] - [ 1 12 0 1 2 1001 2 15 1 3 1 7 2 12 0 1 0 0 0 5 3] - [ 2 2 8 1 0 3 1076 2 0 0 1 2 0 0 0 5 1 2 2 8 3] - [ 2 6 6 2 1 34 7 962 0 1 0 5 0 0 0 0 0 0 14 11 3] - [ 7 3 0 0 0 3 0 0 985 36 8 1 1 5 10 1 1 0 2 1 0] - [ 51 0 0 0 5 4 0 3 17 901 1 1 0 6 4 0 0 2 0 0 6] - [ 1 1 2 5 1 0 1 3 9 0 968 1 0 12 2 1 0 0 4 2 6] - [ 3 0 1 0 1 13 3 2 0 0 1 980 17 5 0 6 2 5 0 9 3] - [ 0 1 1 3 1 3 0 1 0 0 0 30 897 1 1 6 0 13 0 4 7] - [ 0 1 0 0 1 7 0 3 10 14 1 2 5 964 2 0 3 1 0 3 6] - [ 6 6 3 12 4 3 0 0 18 6 1 4 1 2 1050 0 0 1 6 0 7] - [ 0 0 1 0 1 2 1 0 0 0 1 9 3 1 0 1001 6 9 0 5 3] - [ 2 2 1 1 2 1 1 0 2 0 1 5 2 2 0 11 1027 1 0 5 6] - [ 0 2 1 0 1 1 0 0 0 5 0 6 11 1 1 15 0 987 0 1 4] - [ 1 2 0 12 2 4 0 24 1 1 4 4 2 0 6 1 0 1 936 3 4] - [ 1 2 0 2 1 4 6 4 0 0 3 16 6 6 0 2 3 2 1 1015 6] - [ 112 184 132 106 86 181 77 160 77 98 142 96 292 253 124 102 142 86 135 201 10440]] - -2022-12-06 11:29:59,797 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:29:59,797 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:29:59,803 - - -2022-12-06 11:29:59,803 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:30:00,739 - Epoch: [151][ 10/ 1200] Overall Loss 0.145637 Objective Loss 0.145637 LR 0.000250 Time 0.093581 -2022-12-06 11:30:00,941 - Epoch: [151][ 20/ 1200] Overall Loss 0.148552 Objective Loss 0.148552 LR 0.000250 Time 0.056826 -2022-12-06 11:30:01,134 - Epoch: [151][ 30/ 1200] Overall Loss 0.150354 Objective Loss 0.150354 LR 0.000250 Time 0.044300 -2022-12-06 11:30:01,326 - Epoch: [151][ 40/ 1200] Overall Loss 0.151914 Objective Loss 0.151914 LR 0.000250 Time 0.038024 -2022-12-06 11:30:01,518 - Epoch: [151][ 50/ 1200] Overall Loss 0.152580 Objective Loss 0.152580 LR 0.000250 Time 0.034250 -2022-12-06 11:30:01,711 - Epoch: [151][ 60/ 1200] Overall Loss 0.151901 Objective Loss 0.151901 LR 0.000250 Time 0.031739 -2022-12-06 11:30:01,903 - Epoch: [151][ 70/ 1200] Overall Loss 0.148600 Objective Loss 0.148600 LR 0.000250 Time 0.029935 -2022-12-06 11:30:02,094 - Epoch: [151][ 80/ 1200] Overall Loss 0.148493 Objective Loss 0.148493 LR 0.000250 Time 0.028578 -2022-12-06 11:30:02,286 - Epoch: [151][ 90/ 1200] Overall Loss 0.149264 Objective Loss 0.149264 LR 0.000250 Time 0.027528 -2022-12-06 11:30:02,478 - Epoch: [151][ 100/ 1200] Overall Loss 0.148949 Objective Loss 0.148949 LR 0.000250 Time 0.026691 -2022-12-06 11:30:02,669 - Epoch: [151][ 110/ 1200] Overall Loss 0.150571 Objective Loss 0.150571 LR 0.000250 Time 0.026002 -2022-12-06 11:30:02,862 - Epoch: [151][ 120/ 1200] Overall Loss 0.149893 Objective Loss 0.149893 LR 0.000250 Time 0.025432 -2022-12-06 11:30:03,053 - Epoch: [151][ 130/ 1200] Overall Loss 0.150195 Objective Loss 0.150195 LR 0.000250 Time 0.024946 -2022-12-06 11:30:03,245 - Epoch: [151][ 140/ 1200] Overall Loss 0.150441 Objective Loss 0.150441 LR 0.000250 Time 0.024530 -2022-12-06 11:30:03,437 - Epoch: [151][ 150/ 1200] Overall Loss 0.150127 Objective Loss 0.150127 LR 0.000250 Time 0.024168 -2022-12-06 11:30:03,628 - Epoch: [151][ 160/ 1200] Overall Loss 0.151537 Objective Loss 0.151537 LR 0.000250 Time 0.023851 -2022-12-06 11:30:03,820 - Epoch: [151][ 170/ 1200] Overall Loss 0.150388 Objective Loss 0.150388 LR 0.000250 Time 0.023574 -2022-12-06 11:30:04,012 - Epoch: [151][ 180/ 1200] Overall Loss 0.151598 Objective Loss 0.151598 LR 0.000250 Time 0.023328 -2022-12-06 11:30:04,205 - Epoch: [151][ 190/ 1200] Overall Loss 0.151001 Objective Loss 0.151001 LR 0.000250 Time 0.023111 -2022-12-06 11:30:04,396 - Epoch: [151][ 200/ 1200] Overall Loss 0.149907 Objective Loss 0.149907 LR 0.000250 Time 0.022910 -2022-12-06 11:30:04,588 - Epoch: [151][ 210/ 1200] Overall Loss 0.148898 Objective Loss 0.148898 LR 0.000250 Time 0.022730 -2022-12-06 11:30:04,780 - Epoch: [151][ 220/ 1200] Overall Loss 0.147757 Objective Loss 0.147757 LR 0.000250 Time 0.022565 -2022-12-06 11:30:04,971 - Epoch: [151][ 230/ 1200] Overall Loss 0.147894 Objective Loss 0.147894 LR 0.000250 Time 0.022415 -2022-12-06 11:30:05,162 - Epoch: [151][ 240/ 1200] Overall Loss 0.147862 Objective Loss 0.147862 LR 0.000250 Time 0.022275 -2022-12-06 11:30:05,354 - Epoch: [151][ 250/ 1200] Overall Loss 0.147232 Objective Loss 0.147232 LR 0.000250 Time 0.022148 -2022-12-06 11:30:05,546 - Epoch: [151][ 260/ 1200] Overall Loss 0.147157 Objective Loss 0.147157 LR 0.000250 Time 0.022031 -2022-12-06 11:30:05,738 - Epoch: [151][ 270/ 1200] Overall Loss 0.147238 Objective Loss 0.147238 LR 0.000250 Time 0.021925 -2022-12-06 11:30:05,929 - Epoch: [151][ 280/ 1200] Overall Loss 0.146921 Objective Loss 0.146921 LR 0.000250 Time 0.021824 -2022-12-06 11:30:06,121 - Epoch: [151][ 290/ 1200] Overall Loss 0.147757 Objective Loss 0.147757 LR 0.000250 Time 0.021730 -2022-12-06 11:30:06,313 - Epoch: [151][ 300/ 1200] Overall Loss 0.147708 Objective Loss 0.147708 LR 0.000250 Time 0.021646 -2022-12-06 11:30:06,505 - Epoch: [151][ 310/ 1200] Overall Loss 0.148768 Objective Loss 0.148768 LR 0.000250 Time 0.021563 -2022-12-06 11:30:06,696 - Epoch: [151][ 320/ 1200] Overall Loss 0.148374 Objective Loss 0.148374 LR 0.000250 Time 0.021486 -2022-12-06 11:30:06,888 - Epoch: [151][ 330/ 1200] Overall Loss 0.147785 Objective Loss 0.147785 LR 0.000250 Time 0.021414 -2022-12-06 11:30:07,080 - Epoch: [151][ 340/ 1200] Overall Loss 0.147831 Objective Loss 0.147831 LR 0.000250 Time 0.021348 -2022-12-06 11:30:07,273 - Epoch: [151][ 350/ 1200] Overall Loss 0.148024 Objective Loss 0.148024 LR 0.000250 Time 0.021286 -2022-12-06 11:30:07,464 - Epoch: [151][ 360/ 1200] Overall Loss 0.148007 Objective Loss 0.148007 LR 0.000250 Time 0.021226 -2022-12-06 11:30:07,657 - Epoch: [151][ 370/ 1200] Overall Loss 0.148131 Objective Loss 0.148131 LR 0.000250 Time 0.021170 -2022-12-06 11:30:07,848 - Epoch: [151][ 380/ 1200] Overall Loss 0.148251 Objective Loss 0.148251 LR 0.000250 Time 0.021116 -2022-12-06 11:30:08,040 - Epoch: [151][ 390/ 1200] Overall Loss 0.148800 Objective Loss 0.148800 LR 0.000250 Time 0.021064 -2022-12-06 11:30:08,233 - Epoch: [151][ 400/ 1200] Overall Loss 0.148922 Objective Loss 0.148922 LR 0.000250 Time 0.021018 -2022-12-06 11:30:08,425 - Epoch: [151][ 410/ 1200] Overall Loss 0.148957 Objective Loss 0.148957 LR 0.000250 Time 0.020974 -2022-12-06 11:30:08,617 - Epoch: [151][ 420/ 1200] Overall Loss 0.148878 Objective Loss 0.148878 LR 0.000250 Time 0.020930 -2022-12-06 11:30:08,808 - Epoch: [151][ 430/ 1200] Overall Loss 0.149437 Objective Loss 0.149437 LR 0.000250 Time 0.020886 -2022-12-06 11:30:09,000 - Epoch: [151][ 440/ 1200] Overall Loss 0.149111 Objective Loss 0.149111 LR 0.000250 Time 0.020846 -2022-12-06 11:30:09,191 - Epoch: [151][ 450/ 1200] Overall Loss 0.148967 Objective Loss 0.148967 LR 0.000250 Time 0.020807 -2022-12-06 11:30:09,383 - Epoch: [151][ 460/ 1200] Overall Loss 0.149192 Objective Loss 0.149192 LR 0.000250 Time 0.020770 -2022-12-06 11:30:09,574 - Epoch: [151][ 470/ 1200] Overall Loss 0.149472 Objective Loss 0.149472 LR 0.000250 Time 0.020735 -2022-12-06 11:30:09,767 - Epoch: [151][ 480/ 1200] Overall Loss 0.149589 Objective Loss 0.149589 LR 0.000250 Time 0.020702 -2022-12-06 11:30:09,958 - Epoch: [151][ 490/ 1200] Overall Loss 0.149945 Objective Loss 0.149945 LR 0.000250 Time 0.020668 -2022-12-06 11:30:10,150 - Epoch: [151][ 500/ 1200] Overall Loss 0.150015 Objective Loss 0.150015 LR 0.000250 Time 0.020638 -2022-12-06 11:30:10,341 - Epoch: [151][ 510/ 1200] Overall Loss 0.150025 Objective Loss 0.150025 LR 0.000250 Time 0.020608 -2022-12-06 11:30:10,533 - Epoch: [151][ 520/ 1200] Overall Loss 0.150030 Objective Loss 0.150030 LR 0.000250 Time 0.020579 -2022-12-06 11:30:10,725 - Epoch: [151][ 530/ 1200] Overall Loss 0.150114 Objective Loss 0.150114 LR 0.000250 Time 0.020552 -2022-12-06 11:30:10,916 - Epoch: [151][ 540/ 1200] Overall Loss 0.150435 Objective Loss 0.150435 LR 0.000250 Time 0.020525 -2022-12-06 11:30:11,108 - Epoch: [151][ 550/ 1200] Overall Loss 0.150708 Objective Loss 0.150708 LR 0.000250 Time 0.020499 -2022-12-06 11:30:11,300 - Epoch: [151][ 560/ 1200] Overall Loss 0.150640 Objective Loss 0.150640 LR 0.000250 Time 0.020475 -2022-12-06 11:30:11,492 - Epoch: [151][ 570/ 1200] Overall Loss 0.150560 Objective Loss 0.150560 LR 0.000250 Time 0.020451 -2022-12-06 11:30:11,683 - Epoch: [151][ 580/ 1200] Overall Loss 0.150682 Objective Loss 0.150682 LR 0.000250 Time 0.020428 -2022-12-06 11:30:11,876 - Epoch: [151][ 590/ 1200] Overall Loss 0.150374 Objective Loss 0.150374 LR 0.000250 Time 0.020407 -2022-12-06 11:30:12,067 - Epoch: [151][ 600/ 1200] Overall Loss 0.150509 Objective Loss 0.150509 LR 0.000250 Time 0.020385 -2022-12-06 11:30:12,259 - Epoch: [151][ 610/ 1200] Overall Loss 0.150476 Objective Loss 0.150476 LR 0.000250 Time 0.020365 -2022-12-06 11:30:12,451 - Epoch: [151][ 620/ 1200] Overall Loss 0.150548 Objective Loss 0.150548 LR 0.000250 Time 0.020345 -2022-12-06 11:30:12,643 - Epoch: [151][ 630/ 1200] Overall Loss 0.150619 Objective Loss 0.150619 LR 0.000250 Time 0.020326 -2022-12-06 11:30:12,835 - Epoch: [151][ 640/ 1200] Overall Loss 0.150892 Objective Loss 0.150892 LR 0.000250 Time 0.020307 -2022-12-06 11:30:13,026 - Epoch: [151][ 650/ 1200] Overall Loss 0.150762 Objective Loss 0.150762 LR 0.000250 Time 0.020288 -2022-12-06 11:30:13,218 - Epoch: [151][ 660/ 1200] Overall Loss 0.150854 Objective Loss 0.150854 LR 0.000250 Time 0.020271 -2022-12-06 11:30:13,410 - Epoch: [151][ 670/ 1200] Overall Loss 0.150807 Objective Loss 0.150807 LR 0.000250 Time 0.020254 -2022-12-06 11:30:13,601 - Epoch: [151][ 680/ 1200] Overall Loss 0.150533 Objective Loss 0.150533 LR 0.000250 Time 0.020236 -2022-12-06 11:30:13,792 - Epoch: [151][ 690/ 1200] Overall Loss 0.150317 Objective Loss 0.150317 LR 0.000250 Time 0.020219 -2022-12-06 11:30:13,984 - Epoch: [151][ 700/ 1200] Overall Loss 0.150579 Objective Loss 0.150579 LR 0.000250 Time 0.020203 -2022-12-06 11:30:14,176 - Epoch: [151][ 710/ 1200] Overall Loss 0.150475 Objective Loss 0.150475 LR 0.000250 Time 0.020188 -2022-12-06 11:30:14,368 - Epoch: [151][ 720/ 1200] Overall Loss 0.150433 Objective Loss 0.150433 LR 0.000250 Time 0.020174 -2022-12-06 11:30:14,560 - Epoch: [151][ 730/ 1200] Overall Loss 0.150501 Objective Loss 0.150501 LR 0.000250 Time 0.020159 -2022-12-06 11:30:14,752 - Epoch: [151][ 740/ 1200] Overall Loss 0.150387 Objective Loss 0.150387 LR 0.000250 Time 0.020146 -2022-12-06 11:30:14,943 - Epoch: [151][ 750/ 1200] Overall Loss 0.150459 Objective Loss 0.150459 LR 0.000250 Time 0.020132 -2022-12-06 11:30:15,135 - Epoch: [151][ 760/ 1200] Overall Loss 0.150554 Objective Loss 0.150554 LR 0.000250 Time 0.020119 -2022-12-06 11:30:15,327 - Epoch: [151][ 770/ 1200] Overall Loss 0.150795 Objective Loss 0.150795 LR 0.000250 Time 0.020105 -2022-12-06 11:30:15,518 - Epoch: [151][ 780/ 1200] Overall Loss 0.150707 Objective Loss 0.150707 LR 0.000250 Time 0.020092 -2022-12-06 11:30:15,711 - Epoch: [151][ 790/ 1200] Overall Loss 0.151044 Objective Loss 0.151044 LR 0.000250 Time 0.020081 -2022-12-06 11:30:15,902 - Epoch: [151][ 800/ 1200] Overall Loss 0.151139 Objective Loss 0.151139 LR 0.000250 Time 0.020069 -2022-12-06 11:30:16,095 - Epoch: [151][ 810/ 1200] Overall Loss 0.151566 Objective Loss 0.151566 LR 0.000250 Time 0.020058 -2022-12-06 11:30:16,287 - Epoch: [151][ 820/ 1200] Overall Loss 0.151592 Objective Loss 0.151592 LR 0.000250 Time 0.020047 -2022-12-06 11:30:16,479 - Epoch: [151][ 830/ 1200] Overall Loss 0.151652 Objective Loss 0.151652 LR 0.000250 Time 0.020036 -2022-12-06 11:30:16,671 - Epoch: [151][ 840/ 1200] Overall Loss 0.151611 Objective Loss 0.151611 LR 0.000250 Time 0.020026 -2022-12-06 11:30:16,863 - Epoch: [151][ 850/ 1200] Overall Loss 0.151517 Objective Loss 0.151517 LR 0.000250 Time 0.020016 -2022-12-06 11:30:17,055 - Epoch: [151][ 860/ 1200] Overall Loss 0.151604 Objective Loss 0.151604 LR 0.000250 Time 0.020005 -2022-12-06 11:30:17,246 - Epoch: [151][ 870/ 1200] Overall Loss 0.151585 Objective Loss 0.151585 LR 0.000250 Time 0.019995 -2022-12-06 11:30:17,439 - Epoch: [151][ 880/ 1200] Overall Loss 0.151530 Objective Loss 0.151530 LR 0.000250 Time 0.019985 -2022-12-06 11:30:17,631 - Epoch: [151][ 890/ 1200] Overall Loss 0.151263 Objective Loss 0.151263 LR 0.000250 Time 0.019976 -2022-12-06 11:30:17,822 - Epoch: [151][ 900/ 1200] Overall Loss 0.151331 Objective Loss 0.151331 LR 0.000250 Time 0.019966 -2022-12-06 11:30:18,014 - Epoch: [151][ 910/ 1200] Overall Loss 0.151383 Objective Loss 0.151383 LR 0.000250 Time 0.019957 -2022-12-06 11:30:18,206 - Epoch: [151][ 920/ 1200] Overall Loss 0.151417 Objective Loss 0.151417 LR 0.000250 Time 0.019948 -2022-12-06 11:30:18,397 - Epoch: [151][ 930/ 1200] Overall Loss 0.151328 Objective Loss 0.151328 LR 0.000250 Time 0.019939 -2022-12-06 11:30:18,589 - Epoch: [151][ 940/ 1200] Overall Loss 0.151422 Objective Loss 0.151422 LR 0.000250 Time 0.019930 -2022-12-06 11:30:18,781 - Epoch: [151][ 950/ 1200] Overall Loss 0.151712 Objective Loss 0.151712 LR 0.000250 Time 0.019921 -2022-12-06 11:30:18,972 - Epoch: [151][ 960/ 1200] Overall Loss 0.151717 Objective Loss 0.151717 LR 0.000250 Time 0.019913 -2022-12-06 11:30:19,164 - Epoch: [151][ 970/ 1200] Overall Loss 0.151711 Objective Loss 0.151711 LR 0.000250 Time 0.019905 -2022-12-06 11:30:19,356 - Epoch: [151][ 980/ 1200] Overall Loss 0.151700 Objective Loss 0.151700 LR 0.000250 Time 0.019896 -2022-12-06 11:30:19,548 - Epoch: [151][ 990/ 1200] Overall Loss 0.151841 Objective Loss 0.151841 LR 0.000250 Time 0.019889 -2022-12-06 11:30:19,739 - Epoch: [151][ 1000/ 1200] Overall Loss 0.151663 Objective Loss 0.151663 LR 0.000250 Time 0.019881 -2022-12-06 11:30:19,931 - Epoch: [151][ 1010/ 1200] Overall Loss 0.151677 Objective Loss 0.151677 LR 0.000250 Time 0.019874 -2022-12-06 11:30:20,124 - Epoch: [151][ 1020/ 1200] Overall Loss 0.151733 Objective Loss 0.151733 LR 0.000250 Time 0.019867 -2022-12-06 11:30:20,315 - Epoch: [151][ 1030/ 1200] Overall Loss 0.151764 Objective Loss 0.151764 LR 0.000250 Time 0.019859 -2022-12-06 11:30:20,507 - Epoch: [151][ 1040/ 1200] Overall Loss 0.151842 Objective Loss 0.151842 LR 0.000250 Time 0.019852 -2022-12-06 11:30:20,699 - Epoch: [151][ 1050/ 1200] Overall Loss 0.151813 Objective Loss 0.151813 LR 0.000250 Time 0.019846 -2022-12-06 11:30:20,892 - Epoch: [151][ 1060/ 1200] Overall Loss 0.151830 Objective Loss 0.151830 LR 0.000250 Time 0.019839 -2022-12-06 11:30:21,084 - Epoch: [151][ 1070/ 1200] Overall Loss 0.151936 Objective Loss 0.151936 LR 0.000250 Time 0.019833 -2022-12-06 11:30:21,276 - Epoch: [151][ 1080/ 1200] Overall Loss 0.152156 Objective Loss 0.152156 LR 0.000250 Time 0.019827 -2022-12-06 11:30:21,468 - Epoch: [151][ 1090/ 1200] Overall Loss 0.152214 Objective Loss 0.152214 LR 0.000250 Time 0.019821 -2022-12-06 11:30:21,660 - Epoch: [151][ 1100/ 1200] Overall Loss 0.152300 Objective Loss 0.152300 LR 0.000250 Time 0.019814 -2022-12-06 11:30:21,851 - Epoch: [151][ 1110/ 1200] Overall Loss 0.152230 Objective Loss 0.152230 LR 0.000250 Time 0.019808 -2022-12-06 11:30:22,042 - Epoch: [151][ 1120/ 1200] Overall Loss 0.152278 Objective Loss 0.152278 LR 0.000250 Time 0.019801 -2022-12-06 11:30:22,234 - Epoch: [151][ 1130/ 1200] Overall Loss 0.152236 Objective Loss 0.152236 LR 0.000250 Time 0.019795 -2022-12-06 11:30:22,426 - Epoch: [151][ 1140/ 1200] Overall Loss 0.152297 Objective Loss 0.152297 LR 0.000250 Time 0.019790 -2022-12-06 11:30:22,619 - Epoch: [151][ 1150/ 1200] Overall Loss 0.152383 Objective Loss 0.152383 LR 0.000250 Time 0.019784 -2022-12-06 11:30:22,811 - Epoch: [151][ 1160/ 1200] Overall Loss 0.152481 Objective Loss 0.152481 LR 0.000250 Time 0.019779 -2022-12-06 11:30:23,003 - Epoch: [151][ 1170/ 1200] Overall Loss 0.152454 Objective Loss 0.152454 LR 0.000250 Time 0.019773 -2022-12-06 11:30:23,195 - Epoch: [151][ 1180/ 1200] Overall Loss 0.152244 Objective Loss 0.152244 LR 0.000250 Time 0.019768 -2022-12-06 11:30:23,386 - Epoch: [151][ 1190/ 1200] Overall Loss 0.152422 Objective Loss 0.152422 LR 0.000250 Time 0.019762 -2022-12-06 11:30:23,620 - Epoch: [151][ 1200/ 1200] Overall Loss 0.152434 Objective Loss 0.152434 Top1 92.677824 Top5 99.581590 LR 0.000250 Time 0.019792 -2022-12-06 11:30:23,708 - --- validate (epoch=151)----------- -2022-12-06 11:30:23,709 - 34129 samples (256 per mini-batch) -2022-12-06 11:30:24,153 - Epoch: [151][ 10/ 134] Loss 0.275106 Top1 87.187500 Top5 98.789062 -2022-12-06 11:30:24,287 - Epoch: [151][ 20/ 134] Loss 0.256177 Top1 87.539062 Top5 98.515625 -2022-12-06 11:30:24,419 - Epoch: [151][ 30/ 134] Loss 0.244596 Top1 87.656250 Top5 98.489583 -2022-12-06 11:30:24,551 - Epoch: [151][ 40/ 134] Loss 0.237463 Top1 87.812500 Top5 98.554688 -2022-12-06 11:30:24,686 - Epoch: [151][ 50/ 134] Loss 0.235606 Top1 87.773438 Top5 98.578125 -2022-12-06 11:30:24,826 - Epoch: [151][ 60/ 134] Loss 0.236437 Top1 87.714844 Top5 98.561198 -2022-12-06 11:30:24,965 - Epoch: [151][ 70/ 134] Loss 0.234084 Top1 87.656250 Top5 98.549107 -2022-12-06 11:30:25,112 - Epoch: [151][ 80/ 134] Loss 0.234160 Top1 87.675781 Top5 98.554688 -2022-12-06 11:30:25,251 - Epoch: [151][ 90/ 134] Loss 0.232590 Top1 87.803819 Top5 98.559028 -2022-12-06 11:30:25,397 - Epoch: [151][ 100/ 134] Loss 0.234705 Top1 87.726562 Top5 98.519531 -2022-12-06 11:30:25,536 - Epoch: [151][ 110/ 134] Loss 0.238182 Top1 87.663352 Top5 98.494318 -2022-12-06 11:30:25,682 - Epoch: [151][ 120/ 134] Loss 0.235743 Top1 87.750651 Top5 98.509115 -2022-12-06 11:30:25,817 - Epoch: [151][ 130/ 134] Loss 0.235413 Top1 87.755409 Top5 98.521635 -2022-12-06 11:30:25,854 - Epoch: [151][ 134/ 134] Loss 0.235007 Top1 87.737701 Top5 98.511530 -2022-12-06 11:30:25,942 - ==> Top1: 87.738 Top5: 98.512 Loss: 0.235 - -2022-12-06 11:30:25,942 - ==> Confusion: -[[ 920 0 1 0 5 6 1 0 5 38 0 2 0 3 5 2 2 0 2 0 4] - [ 2 937 3 2 8 21 2 16 0 0 4 5 0 1 0 1 3 1 9 5 7] - [ 5 2 1014 9 5 3 15 10 0 5 6 5 3 1 1 1 0 1 3 5 9] - [ 2 2 15 947 1 1 2 1 1 1 10 1 3 0 10 0 1 3 11 1 7] - [ 8 7 2 0 960 1 2 2 1 6 1 2 0 2 5 4 7 1 2 1 6] - [ 0 9 0 1 5 998 2 15 1 2 1 11 5 7 0 1 0 1 1 8 1] - [ 1 1 6 4 0 0 1079 2 0 1 0 2 0 1 0 6 0 2 2 10 1] - [ 1 5 5 2 1 26 5 966 0 1 0 3 0 2 0 0 0 0 10 18 9] - [ 3 4 0 0 0 3 1 0 972 42 11 1 1 12 10 0 2 0 1 1 0] - [ 66 0 0 0 7 4 1 3 23 869 1 2 0 13 3 1 0 1 0 0 7] - [ 1 0 4 2 1 0 1 4 7 0 965 2 0 12 4 1 0 0 5 2 8] - [ 3 0 0 0 1 6 3 1 1 0 1 984 27 4 0 5 3 2 0 7 3] - [ 0 1 0 0 0 2 1 1 0 1 0 26 912 1 0 8 1 5 1 3 6] - [ 1 0 1 0 0 7 0 3 9 13 4 3 3 966 1 2 2 0 0 1 7] - [ 7 4 4 11 4 4 0 0 16 2 0 2 3 4 1056 0 0 1 6 1 5] - [ 1 0 1 0 2 1 1 0 1 0 0 7 9 1 0 999 5 9 0 4 2] - [ 2 0 0 0 2 2 1 0 1 0 0 5 2 2 1 11 1034 0 0 4 5] - [ 4 0 1 1 0 1 0 0 0 2 0 7 12 2 2 18 0 982 0 1 3] - [ 2 3 4 8 1 2 1 24 1 1 1 5 3 0 7 1 0 2 937 1 4] - [ 1 2 1 2 0 3 7 1 0 0 2 15 5 5 0 3 3 3 3 1014 10] - [ 120 182 157 79 113 184 77 146 70 66 151 94 279 278 117 116 164 85 129 188 10431]] - -2022-12-06 11:30:26,515 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:30:26,515 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:30:26,521 - - -2022-12-06 11:30:26,522 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:30:27,575 - Epoch: [152][ 10/ 1200] Overall Loss 0.134590 Objective Loss 0.134590 LR 0.000250 Time 0.105229 -2022-12-06 11:30:27,776 - Epoch: [152][ 20/ 1200] Overall Loss 0.137889 Objective Loss 0.137889 LR 0.000250 Time 0.062634 -2022-12-06 11:30:27,973 - Epoch: [152][ 30/ 1200] Overall Loss 0.139331 Objective Loss 0.139331 LR 0.000250 Time 0.048316 -2022-12-06 11:30:28,173 - Epoch: [152][ 40/ 1200] Overall Loss 0.143198 Objective Loss 0.143198 LR 0.000250 Time 0.041233 -2022-12-06 11:30:28,370 - Epoch: [152][ 50/ 1200] Overall Loss 0.148212 Objective Loss 0.148212 LR 0.000250 Time 0.036901 -2022-12-06 11:30:28,570 - Epoch: [152][ 60/ 1200] Overall Loss 0.148446 Objective Loss 0.148446 LR 0.000250 Time 0.034078 -2022-12-06 11:30:28,766 - Epoch: [152][ 70/ 1200] Overall Loss 0.147893 Objective Loss 0.147893 LR 0.000250 Time 0.032008 -2022-12-06 11:30:28,966 - Epoch: [152][ 80/ 1200] Overall Loss 0.149031 Objective Loss 0.149031 LR 0.000250 Time 0.030492 -2022-12-06 11:30:29,162 - Epoch: [152][ 90/ 1200] Overall Loss 0.148614 Objective Loss 0.148614 LR 0.000250 Time 0.029282 -2022-12-06 11:30:29,362 - Epoch: [152][ 100/ 1200] Overall Loss 0.145445 Objective Loss 0.145445 LR 0.000250 Time 0.028342 -2022-12-06 11:30:29,558 - Epoch: [152][ 110/ 1200] Overall Loss 0.144526 Objective Loss 0.144526 LR 0.000250 Time 0.027549 -2022-12-06 11:30:29,758 - Epoch: [152][ 120/ 1200] Overall Loss 0.145567 Objective Loss 0.145567 LR 0.000250 Time 0.026910 -2022-12-06 11:30:29,954 - Epoch: [152][ 130/ 1200] Overall Loss 0.144555 Objective Loss 0.144555 LR 0.000250 Time 0.026345 -2022-12-06 11:30:30,153 - Epoch: [152][ 140/ 1200] Overall Loss 0.144966 Objective Loss 0.144966 LR 0.000250 Time 0.025884 -2022-12-06 11:30:30,350 - Epoch: [152][ 150/ 1200] Overall Loss 0.145744 Objective Loss 0.145744 LR 0.000250 Time 0.025465 -2022-12-06 11:30:30,550 - Epoch: [152][ 160/ 1200] Overall Loss 0.146463 Objective Loss 0.146463 LR 0.000250 Time 0.025118 -2022-12-06 11:30:30,746 - Epoch: [152][ 170/ 1200] Overall Loss 0.146400 Objective Loss 0.146400 LR 0.000250 Time 0.024795 -2022-12-06 11:30:30,949 - Epoch: [152][ 180/ 1200] Overall Loss 0.145588 Objective Loss 0.145588 LR 0.000250 Time 0.024538 -2022-12-06 11:30:31,146 - Epoch: [152][ 190/ 1200] Overall Loss 0.145291 Objective Loss 0.145291 LR 0.000250 Time 0.024285 -2022-12-06 11:30:31,348 - Epoch: [152][ 200/ 1200] Overall Loss 0.145727 Objective Loss 0.145727 LR 0.000250 Time 0.024073 -2022-12-06 11:30:31,545 - Epoch: [152][ 210/ 1200] Overall Loss 0.145776 Objective Loss 0.145776 LR 0.000250 Time 0.023866 -2022-12-06 11:30:31,746 - Epoch: [152][ 220/ 1200] Overall Loss 0.146196 Objective Loss 0.146196 LR 0.000250 Time 0.023691 -2022-12-06 11:30:31,944 - Epoch: [152][ 230/ 1200] Overall Loss 0.146885 Objective Loss 0.146885 LR 0.000250 Time 0.023518 -2022-12-06 11:30:32,144 - Epoch: [152][ 240/ 1200] Overall Loss 0.146136 Objective Loss 0.146136 LR 0.000250 Time 0.023372 -2022-12-06 11:30:32,342 - Epoch: [152][ 250/ 1200] Overall Loss 0.146821 Objective Loss 0.146821 LR 0.000250 Time 0.023226 -2022-12-06 11:30:32,544 - Epoch: [152][ 260/ 1200] Overall Loss 0.147063 Objective Loss 0.147063 LR 0.000250 Time 0.023108 -2022-12-06 11:30:32,742 - Epoch: [152][ 270/ 1200] Overall Loss 0.146672 Objective Loss 0.146672 LR 0.000250 Time 0.022981 -2022-12-06 11:30:32,944 - Epoch: [152][ 280/ 1200] Overall Loss 0.146833 Objective Loss 0.146833 LR 0.000250 Time 0.022880 -2022-12-06 11:30:33,142 - Epoch: [152][ 290/ 1200] Overall Loss 0.147812 Objective Loss 0.147812 LR 0.000250 Time 0.022773 -2022-12-06 11:30:33,343 - Epoch: [152][ 300/ 1200] Overall Loss 0.148266 Objective Loss 0.148266 LR 0.000250 Time 0.022683 -2022-12-06 11:30:33,541 - Epoch: [152][ 310/ 1200] Overall Loss 0.147854 Objective Loss 0.147854 LR 0.000250 Time 0.022588 -2022-12-06 11:30:33,743 - Epoch: [152][ 320/ 1200] Overall Loss 0.147510 Objective Loss 0.147510 LR 0.000250 Time 0.022509 -2022-12-06 11:30:33,941 - Epoch: [152][ 330/ 1200] Overall Loss 0.147623 Objective Loss 0.147623 LR 0.000250 Time 0.022427 -2022-12-06 11:30:34,142 - Epoch: [152][ 340/ 1200] Overall Loss 0.147795 Objective Loss 0.147795 LR 0.000250 Time 0.022358 -2022-12-06 11:30:34,341 - Epoch: [152][ 350/ 1200] Overall Loss 0.148235 Objective Loss 0.148235 LR 0.000250 Time 0.022285 -2022-12-06 11:30:34,542 - Epoch: [152][ 360/ 1200] Overall Loss 0.148599 Objective Loss 0.148599 LR 0.000250 Time 0.022224 -2022-12-06 11:30:34,741 - Epoch: [152][ 370/ 1200] Overall Loss 0.148621 Objective Loss 0.148621 LR 0.000250 Time 0.022157 -2022-12-06 11:30:34,942 - Epoch: [152][ 380/ 1200] Overall Loss 0.148565 Objective Loss 0.148565 LR 0.000250 Time 0.022102 -2022-12-06 11:30:35,140 - Epoch: [152][ 390/ 1200] Overall Loss 0.148462 Objective Loss 0.148462 LR 0.000250 Time 0.022043 -2022-12-06 11:30:35,342 - Epoch: [152][ 400/ 1200] Overall Loss 0.148769 Objective Loss 0.148769 LR 0.000250 Time 0.021994 -2022-12-06 11:30:35,540 - Epoch: [152][ 410/ 1200] Overall Loss 0.149082 Objective Loss 0.149082 LR 0.000250 Time 0.021940 -2022-12-06 11:30:35,742 - Epoch: [152][ 420/ 1200] Overall Loss 0.148893 Objective Loss 0.148893 LR 0.000250 Time 0.021897 -2022-12-06 11:30:35,941 - Epoch: [152][ 430/ 1200] Overall Loss 0.149394 Objective Loss 0.149394 LR 0.000250 Time 0.021849 -2022-12-06 11:30:36,142 - Epoch: [152][ 440/ 1200] Overall Loss 0.149357 Objective Loss 0.149357 LR 0.000250 Time 0.021809 -2022-12-06 11:30:36,341 - Epoch: [152][ 450/ 1200] Overall Loss 0.149464 Objective Loss 0.149464 LR 0.000250 Time 0.021764 -2022-12-06 11:30:36,542 - Epoch: [152][ 460/ 1200] Overall Loss 0.149397 Objective Loss 0.149397 LR 0.000250 Time 0.021727 -2022-12-06 11:30:36,741 - Epoch: [152][ 470/ 1200] Overall Loss 0.149534 Objective Loss 0.149534 LR 0.000250 Time 0.021686 -2022-12-06 11:30:36,942 - Epoch: [152][ 480/ 1200] Overall Loss 0.149551 Objective Loss 0.149551 LR 0.000250 Time 0.021654 -2022-12-06 11:30:37,141 - Epoch: [152][ 490/ 1200] Overall Loss 0.149771 Objective Loss 0.149771 LR 0.000250 Time 0.021615 -2022-12-06 11:30:37,342 - Epoch: [152][ 500/ 1200] Overall Loss 0.149951 Objective Loss 0.149951 LR 0.000250 Time 0.021585 -2022-12-06 11:30:37,540 - Epoch: [152][ 510/ 1200] Overall Loss 0.149672 Objective Loss 0.149672 LR 0.000250 Time 0.021549 -2022-12-06 11:30:37,741 - Epoch: [152][ 520/ 1200] Overall Loss 0.149822 Objective Loss 0.149822 LR 0.000250 Time 0.021520 -2022-12-06 11:30:37,940 - Epoch: [152][ 530/ 1200] Overall Loss 0.150116 Objective Loss 0.150116 LR 0.000250 Time 0.021488 -2022-12-06 11:30:38,141 - Epoch: [152][ 540/ 1200] Overall Loss 0.150433 Objective Loss 0.150433 LR 0.000250 Time 0.021462 -2022-12-06 11:30:38,340 - Epoch: [152][ 550/ 1200] Overall Loss 0.150303 Objective Loss 0.150303 LR 0.000250 Time 0.021431 -2022-12-06 11:30:38,541 - Epoch: [152][ 560/ 1200] Overall Loss 0.150295 Objective Loss 0.150295 LR 0.000250 Time 0.021407 -2022-12-06 11:30:38,739 - Epoch: [152][ 570/ 1200] Overall Loss 0.150401 Objective Loss 0.150401 LR 0.000250 Time 0.021379 -2022-12-06 11:30:38,941 - Epoch: [152][ 580/ 1200] Overall Loss 0.150609 Objective Loss 0.150609 LR 0.000250 Time 0.021357 -2022-12-06 11:30:39,139 - Epoch: [152][ 590/ 1200] Overall Loss 0.150598 Objective Loss 0.150598 LR 0.000250 Time 0.021330 -2022-12-06 11:30:39,341 - Epoch: [152][ 600/ 1200] Overall Loss 0.150656 Objective Loss 0.150656 LR 0.000250 Time 0.021309 -2022-12-06 11:30:39,539 - Epoch: [152][ 610/ 1200] Overall Loss 0.150965 Objective Loss 0.150965 LR 0.000250 Time 0.021284 -2022-12-06 11:30:39,741 - Epoch: [152][ 620/ 1200] Overall Loss 0.151045 Objective Loss 0.151045 LR 0.000250 Time 0.021265 -2022-12-06 11:30:39,939 - Epoch: [152][ 630/ 1200] Overall Loss 0.151053 Objective Loss 0.151053 LR 0.000250 Time 0.021242 -2022-12-06 11:30:40,141 - Epoch: [152][ 640/ 1200] Overall Loss 0.151402 Objective Loss 0.151402 LR 0.000250 Time 0.021224 -2022-12-06 11:30:40,340 - Epoch: [152][ 650/ 1200] Overall Loss 0.151606 Objective Loss 0.151606 LR 0.000250 Time 0.021203 -2022-12-06 11:30:40,541 - Epoch: [152][ 660/ 1200] Overall Loss 0.151580 Objective Loss 0.151580 LR 0.000250 Time 0.021186 -2022-12-06 11:30:40,740 - Epoch: [152][ 670/ 1200] Overall Loss 0.151538 Objective Loss 0.151538 LR 0.000250 Time 0.021165 -2022-12-06 11:30:40,942 - Epoch: [152][ 680/ 1200] Overall Loss 0.151575 Objective Loss 0.151575 LR 0.000250 Time 0.021150 -2022-12-06 11:30:41,141 - Epoch: [152][ 690/ 1200] Overall Loss 0.151831 Objective Loss 0.151831 LR 0.000250 Time 0.021131 -2022-12-06 11:30:41,342 - Epoch: [152][ 700/ 1200] Overall Loss 0.151717 Objective Loss 0.151717 LR 0.000250 Time 0.021116 -2022-12-06 11:30:41,540 - Epoch: [152][ 710/ 1200] Overall Loss 0.151851 Objective Loss 0.151851 LR 0.000250 Time 0.021097 -2022-12-06 11:30:41,741 - Epoch: [152][ 720/ 1200] Overall Loss 0.151931 Objective Loss 0.151931 LR 0.000250 Time 0.021082 -2022-12-06 11:30:41,939 - Epoch: [152][ 730/ 1200] Overall Loss 0.151822 Objective Loss 0.151822 LR 0.000250 Time 0.021064 -2022-12-06 11:30:42,141 - Epoch: [152][ 740/ 1200] Overall Loss 0.151602 Objective Loss 0.151602 LR 0.000250 Time 0.021051 -2022-12-06 11:30:42,339 - Epoch: [152][ 750/ 1200] Overall Loss 0.151716 Objective Loss 0.151716 LR 0.000250 Time 0.021033 -2022-12-06 11:30:42,540 - Epoch: [152][ 760/ 1200] Overall Loss 0.151518 Objective Loss 0.151518 LR 0.000250 Time 0.021021 -2022-12-06 11:30:42,738 - Epoch: [152][ 770/ 1200] Overall Loss 0.151571 Objective Loss 0.151571 LR 0.000250 Time 0.021004 -2022-12-06 11:30:42,939 - Epoch: [152][ 780/ 1200] Overall Loss 0.151548 Objective Loss 0.151548 LR 0.000250 Time 0.020992 -2022-12-06 11:30:43,138 - Epoch: [152][ 790/ 1200] Overall Loss 0.151775 Objective Loss 0.151775 LR 0.000250 Time 0.020977 -2022-12-06 11:30:43,338 - Epoch: [152][ 800/ 1200] Overall Loss 0.151645 Objective Loss 0.151645 LR 0.000250 Time 0.020965 -2022-12-06 11:30:43,537 - Epoch: [152][ 810/ 1200] Overall Loss 0.151670 Objective Loss 0.151670 LR 0.000250 Time 0.020950 -2022-12-06 11:30:43,738 - Epoch: [152][ 820/ 1200] Overall Loss 0.151593 Objective Loss 0.151593 LR 0.000250 Time 0.020939 -2022-12-06 11:30:43,936 - Epoch: [152][ 830/ 1200] Overall Loss 0.151508 Objective Loss 0.151508 LR 0.000250 Time 0.020926 -2022-12-06 11:30:44,138 - Epoch: [152][ 840/ 1200] Overall Loss 0.151466 Objective Loss 0.151466 LR 0.000250 Time 0.020916 -2022-12-06 11:30:44,336 - Epoch: [152][ 850/ 1200] Overall Loss 0.151362 Objective Loss 0.151362 LR 0.000250 Time 0.020902 -2022-12-06 11:30:44,536 - Epoch: [152][ 860/ 1200] Overall Loss 0.151424 Objective Loss 0.151424 LR 0.000250 Time 0.020891 -2022-12-06 11:30:44,734 - Epoch: [152][ 870/ 1200] Overall Loss 0.151403 Objective Loss 0.151403 LR 0.000250 Time 0.020878 -2022-12-06 11:30:44,935 - Epoch: [152][ 880/ 1200] Overall Loss 0.151556 Objective Loss 0.151556 LR 0.000250 Time 0.020868 -2022-12-06 11:30:45,133 - Epoch: [152][ 890/ 1200] Overall Loss 0.151478 Objective Loss 0.151478 LR 0.000250 Time 0.020856 -2022-12-06 11:30:45,335 - Epoch: [152][ 900/ 1200] Overall Loss 0.151505 Objective Loss 0.151505 LR 0.000250 Time 0.020847 -2022-12-06 11:30:45,533 - Epoch: [152][ 910/ 1200] Overall Loss 0.151665 Objective Loss 0.151665 LR 0.000250 Time 0.020835 -2022-12-06 11:30:45,734 - Epoch: [152][ 920/ 1200] Overall Loss 0.151782 Objective Loss 0.151782 LR 0.000250 Time 0.020827 -2022-12-06 11:30:45,933 - Epoch: [152][ 930/ 1200] Overall Loss 0.151777 Objective Loss 0.151777 LR 0.000250 Time 0.020816 -2022-12-06 11:30:46,134 - Epoch: [152][ 940/ 1200] Overall Loss 0.152047 Objective Loss 0.152047 LR 0.000250 Time 0.020808 -2022-12-06 11:30:46,332 - Epoch: [152][ 950/ 1200] Overall Loss 0.151899 Objective Loss 0.151899 LR 0.000250 Time 0.020797 -2022-12-06 11:30:46,533 - Epoch: [152][ 960/ 1200] Overall Loss 0.151837 Objective Loss 0.151837 LR 0.000250 Time 0.020789 -2022-12-06 11:30:46,731 - Epoch: [152][ 970/ 1200] Overall Loss 0.151638 Objective Loss 0.151638 LR 0.000250 Time 0.020778 -2022-12-06 11:30:46,932 - Epoch: [152][ 980/ 1200] Overall Loss 0.151633 Objective Loss 0.151633 LR 0.000250 Time 0.020771 -2022-12-06 11:30:47,131 - Epoch: [152][ 990/ 1200] Overall Loss 0.151694 Objective Loss 0.151694 LR 0.000250 Time 0.020761 -2022-12-06 11:30:47,332 - Epoch: [152][ 1000/ 1200] Overall Loss 0.151654 Objective Loss 0.151654 LR 0.000250 Time 0.020754 -2022-12-06 11:30:47,530 - Epoch: [152][ 1010/ 1200] Overall Loss 0.151474 Objective Loss 0.151474 LR 0.000250 Time 0.020744 -2022-12-06 11:30:47,731 - Epoch: [152][ 1020/ 1200] Overall Loss 0.151684 Objective Loss 0.151684 LR 0.000250 Time 0.020738 -2022-12-06 11:30:47,929 - Epoch: [152][ 1030/ 1200] Overall Loss 0.151669 Objective Loss 0.151669 LR 0.000250 Time 0.020728 -2022-12-06 11:30:48,131 - Epoch: [152][ 1040/ 1200] Overall Loss 0.151805 Objective Loss 0.151805 LR 0.000250 Time 0.020722 -2022-12-06 11:30:48,329 - Epoch: [152][ 1050/ 1200] Overall Loss 0.151956 Objective Loss 0.151956 LR 0.000250 Time 0.020713 -2022-12-06 11:30:48,530 - Epoch: [152][ 1060/ 1200] Overall Loss 0.151991 Objective Loss 0.151991 LR 0.000250 Time 0.020707 -2022-12-06 11:30:48,729 - Epoch: [152][ 1070/ 1200] Overall Loss 0.152038 Objective Loss 0.152038 LR 0.000250 Time 0.020698 -2022-12-06 11:30:48,930 - Epoch: [152][ 1080/ 1200] Overall Loss 0.152001 Objective Loss 0.152001 LR 0.000250 Time 0.020692 -2022-12-06 11:30:49,128 - Epoch: [152][ 1090/ 1200] Overall Loss 0.152060 Objective Loss 0.152060 LR 0.000250 Time 0.020684 -2022-12-06 11:30:49,329 - Epoch: [152][ 1100/ 1200] Overall Loss 0.152352 Objective Loss 0.152352 LR 0.000250 Time 0.020678 -2022-12-06 11:30:49,527 - Epoch: [152][ 1110/ 1200] Overall Loss 0.152225 Objective Loss 0.152225 LR 0.000250 Time 0.020670 -2022-12-06 11:30:49,728 - Epoch: [152][ 1120/ 1200] Overall Loss 0.152220 Objective Loss 0.152220 LR 0.000250 Time 0.020664 -2022-12-06 11:30:49,927 - Epoch: [152][ 1130/ 1200] Overall Loss 0.152316 Objective Loss 0.152316 LR 0.000250 Time 0.020656 -2022-12-06 11:30:50,128 - Epoch: [152][ 1140/ 1200] Overall Loss 0.152373 Objective Loss 0.152373 LR 0.000250 Time 0.020651 -2022-12-06 11:30:50,325 - Epoch: [152][ 1150/ 1200] Overall Loss 0.152297 Objective Loss 0.152297 LR 0.000250 Time 0.020643 -2022-12-06 11:30:50,527 - Epoch: [152][ 1160/ 1200] Overall Loss 0.152226 Objective Loss 0.152226 LR 0.000250 Time 0.020638 -2022-12-06 11:30:50,726 - Epoch: [152][ 1170/ 1200] Overall Loss 0.152208 Objective Loss 0.152208 LR 0.000250 Time 0.020631 -2022-12-06 11:30:50,927 - Epoch: [152][ 1180/ 1200] Overall Loss 0.152443 Objective Loss 0.152443 LR 0.000250 Time 0.020626 -2022-12-06 11:30:51,125 - Epoch: [152][ 1190/ 1200] Overall Loss 0.152427 Objective Loss 0.152427 LR 0.000250 Time 0.020619 -2022-12-06 11:30:51,351 - Epoch: [152][ 1200/ 1200] Overall Loss 0.152519 Objective Loss 0.152519 Top1 90.376569 Top5 99.372385 LR 0.000250 Time 0.020635 -2022-12-06 11:30:51,439 - --- validate (epoch=152)----------- -2022-12-06 11:30:51,439 - 34129 samples (256 per mini-batch) -2022-12-06 11:30:51,885 - Epoch: [152][ 10/ 134] Loss 0.258355 Top1 87.617188 Top5 98.437500 -2022-12-06 11:30:52,018 - Epoch: [152][ 20/ 134] Loss 0.229998 Top1 88.066406 Top5 98.730469 -2022-12-06 11:30:52,149 - Epoch: [152][ 30/ 134] Loss 0.221459 Top1 88.203125 Top5 98.736979 -2022-12-06 11:30:52,279 - Epoch: [152][ 40/ 134] Loss 0.224442 Top1 88.183594 Top5 98.623047 -2022-12-06 11:30:52,410 - Epoch: [152][ 50/ 134] Loss 0.223604 Top1 88.015625 Top5 98.632812 -2022-12-06 11:30:52,539 - Epoch: [152][ 60/ 134] Loss 0.225654 Top1 87.936198 Top5 98.593750 -2022-12-06 11:30:52,670 - Epoch: [152][ 70/ 134] Loss 0.225999 Top1 87.862723 Top5 98.610491 -2022-12-06 11:30:52,801 - Epoch: [152][ 80/ 134] Loss 0.226047 Top1 87.900391 Top5 98.608398 -2022-12-06 11:30:52,932 - Epoch: [152][ 90/ 134] Loss 0.227714 Top1 87.864583 Top5 98.541667 -2022-12-06 11:30:53,062 - Epoch: [152][ 100/ 134] Loss 0.228299 Top1 87.914062 Top5 98.519531 -2022-12-06 11:30:53,192 - Epoch: [152][ 110/ 134] Loss 0.231631 Top1 87.844460 Top5 98.522727 -2022-12-06 11:30:53,323 - Epoch: [152][ 120/ 134] Loss 0.232701 Top1 87.727865 Top5 98.512370 -2022-12-06 11:30:53,459 - Epoch: [152][ 130/ 134] Loss 0.232923 Top1 87.815505 Top5 98.545673 -2022-12-06 11:30:53,499 - Epoch: [152][ 134/ 134] Loss 0.233533 Top1 87.810953 Top5 98.552551 -2022-12-06 11:30:53,589 - ==> Top1: 87.811 Top5: 98.553 Loss: 0.234 - -2022-12-06 11:30:53,590 - ==> Confusion: -[[ 901 0 3 2 6 6 0 0 7 52 0 1 0 3 8 2 0 2 1 0 2] - [ 1 948 1 2 5 21 2 8 4 0 2 3 0 0 0 1 4 0 11 5 9] - [ 6 2 1002 9 3 3 23 11 0 4 8 6 3 0 0 2 2 1 3 4 11] - [ 2 3 16 953 0 2 1 0 1 0 7 0 4 1 9 0 1 3 8 1 8] - [ 9 5 0 0 963 3 1 1 1 5 1 2 1 3 9 4 6 2 0 0 4] - [ 1 11 0 1 4 998 1 14 3 2 2 12 4 9 0 1 1 0 1 3 1] - [ 2 1 5 0 1 0 1083 2 0 0 0 1 1 1 1 3 1 3 1 11 1] - [ 0 3 2 2 2 29 10 959 1 1 2 3 0 2 0 0 0 0 20 13 5] - [ 2 4 0 0 0 2 1 1 985 37 8 1 2 7 8 1 1 0 2 1 1] - [ 45 0 1 0 6 5 0 2 20 899 1 2 0 10 2 0 1 2 1 0 4] - [ 0 2 4 1 2 0 2 2 8 0 972 0 1 11 2 0 0 0 8 1 3] - [ 3 0 0 0 0 7 4 2 1 0 1 989 22 5 1 4 1 4 0 5 2] - [ 0 1 0 0 1 2 1 1 1 1 0 27 906 2 0 7 2 6 0 3 8] - [ 1 1 1 0 0 8 0 2 12 13 5 3 4 957 2 1 4 1 0 2 6] - [ 7 3 3 14 3 3 0 1 17 0 0 2 3 3 1057 0 0 1 8 0 5] - [ 1 0 1 0 3 0 5 0 0 0 1 7 5 4 0 996 3 8 0 4 5] - [ 0 2 2 1 3 0 2 0 1 0 0 6 1 2 0 6 1033 0 1 4 8] - [ 2 0 1 1 0 1 1 0 0 4 0 6 19 2 1 14 2 978 0 2 2] - [ 2 4 2 8 2 2 0 20 1 1 2 3 5 0 6 0 0 2 940 4 4] - [ 2 3 1 0 1 5 5 4 0 0 2 15 7 5 0 2 1 1 3 1018 5] - [ 107 192 136 86 103 167 99 122 81 71 168 81 289 263 129 123 163 75 138 207 10426]] - -2022-12-06 11:30:54,169 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:30:54,170 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:30:54,176 - - -2022-12-06 11:30:54,176 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:30:55,104 - Epoch: [153][ 10/ 1200] Overall Loss 0.155745 Objective Loss 0.155745 LR 0.000250 Time 0.092720 -2022-12-06 11:30:55,305 - Epoch: [153][ 20/ 1200] Overall Loss 0.152870 Objective Loss 0.152870 LR 0.000250 Time 0.056402 -2022-12-06 11:30:55,498 - Epoch: [153][ 30/ 1200] Overall Loss 0.146427 Objective Loss 0.146427 LR 0.000250 Time 0.043997 -2022-12-06 11:30:55,690 - Epoch: [153][ 40/ 1200] Overall Loss 0.146722 Objective Loss 0.146722 LR 0.000250 Time 0.037787 -2022-12-06 11:30:55,882 - Epoch: [153][ 50/ 1200] Overall Loss 0.144677 Objective Loss 0.144677 LR 0.000250 Time 0.034060 -2022-12-06 11:30:56,073 - Epoch: [153][ 60/ 1200] Overall Loss 0.147514 Objective Loss 0.147514 LR 0.000250 Time 0.031563 -2022-12-06 11:30:56,265 - Epoch: [153][ 70/ 1200] Overall Loss 0.148000 Objective Loss 0.148000 LR 0.000250 Time 0.029786 -2022-12-06 11:30:56,457 - Epoch: [153][ 80/ 1200] Overall Loss 0.149262 Objective Loss 0.149262 LR 0.000250 Time 0.028453 -2022-12-06 11:30:56,649 - Epoch: [153][ 90/ 1200] Overall Loss 0.148100 Objective Loss 0.148100 LR 0.000250 Time 0.027424 -2022-12-06 11:30:56,841 - Epoch: [153][ 100/ 1200] Overall Loss 0.150647 Objective Loss 0.150647 LR 0.000250 Time 0.026597 -2022-12-06 11:30:57,033 - Epoch: [153][ 110/ 1200] Overall Loss 0.150823 Objective Loss 0.150823 LR 0.000250 Time 0.025921 -2022-12-06 11:30:57,225 - Epoch: [153][ 120/ 1200] Overall Loss 0.152918 Objective Loss 0.152918 LR 0.000250 Time 0.025354 -2022-12-06 11:30:57,417 - Epoch: [153][ 130/ 1200] Overall Loss 0.152710 Objective Loss 0.152710 LR 0.000250 Time 0.024878 -2022-12-06 11:30:57,609 - Epoch: [153][ 140/ 1200] Overall Loss 0.152563 Objective Loss 0.152563 LR 0.000250 Time 0.024463 -2022-12-06 11:30:57,801 - Epoch: [153][ 150/ 1200] Overall Loss 0.151167 Objective Loss 0.151167 LR 0.000250 Time 0.024109 -2022-12-06 11:30:57,993 - Epoch: [153][ 160/ 1200] Overall Loss 0.149268 Objective Loss 0.149268 LR 0.000250 Time 0.023798 -2022-12-06 11:30:58,185 - Epoch: [153][ 170/ 1200] Overall Loss 0.149965 Objective Loss 0.149965 LR 0.000250 Time 0.023525 -2022-12-06 11:30:58,376 - Epoch: [153][ 180/ 1200] Overall Loss 0.151389 Objective Loss 0.151389 LR 0.000250 Time 0.023279 -2022-12-06 11:30:58,568 - Epoch: [153][ 190/ 1200] Overall Loss 0.152696 Objective Loss 0.152696 LR 0.000250 Time 0.023060 -2022-12-06 11:30:58,761 - Epoch: [153][ 200/ 1200] Overall Loss 0.152526 Objective Loss 0.152526 LR 0.000250 Time 0.022867 -2022-12-06 11:30:58,953 - Epoch: [153][ 210/ 1200] Overall Loss 0.152365 Objective Loss 0.152365 LR 0.000250 Time 0.022691 -2022-12-06 11:30:59,145 - Epoch: [153][ 220/ 1200] Overall Loss 0.151818 Objective Loss 0.151818 LR 0.000250 Time 0.022529 -2022-12-06 11:30:59,337 - Epoch: [153][ 230/ 1200] Overall Loss 0.151694 Objective Loss 0.151694 LR 0.000250 Time 0.022382 -2022-12-06 11:30:59,528 - Epoch: [153][ 240/ 1200] Overall Loss 0.152432 Objective Loss 0.152432 LR 0.000250 Time 0.022245 -2022-12-06 11:30:59,720 - Epoch: [153][ 250/ 1200] Overall Loss 0.151929 Objective Loss 0.151929 LR 0.000250 Time 0.022121 -2022-12-06 11:30:59,912 - Epoch: [153][ 260/ 1200] Overall Loss 0.151638 Objective Loss 0.151638 LR 0.000250 Time 0.022005 -2022-12-06 11:31:00,104 - Epoch: [153][ 270/ 1200] Overall Loss 0.151669 Objective Loss 0.151669 LR 0.000250 Time 0.021899 -2022-12-06 11:31:00,296 - Epoch: [153][ 280/ 1200] Overall Loss 0.151520 Objective Loss 0.151520 LR 0.000250 Time 0.021801 -2022-12-06 11:31:00,488 - Epoch: [153][ 290/ 1200] Overall Loss 0.151504 Objective Loss 0.151504 LR 0.000250 Time 0.021710 -2022-12-06 11:31:00,680 - Epoch: [153][ 300/ 1200] Overall Loss 0.151165 Objective Loss 0.151165 LR 0.000250 Time 0.021623 -2022-12-06 11:31:00,872 - Epoch: [153][ 310/ 1200] Overall Loss 0.151759 Objective Loss 0.151759 LR 0.000250 Time 0.021545 -2022-12-06 11:31:01,064 - Epoch: [153][ 320/ 1200] Overall Loss 0.151492 Objective Loss 0.151492 LR 0.000250 Time 0.021470 -2022-12-06 11:31:01,256 - Epoch: [153][ 330/ 1200] Overall Loss 0.151424 Objective Loss 0.151424 LR 0.000250 Time 0.021399 -2022-12-06 11:31:01,447 - Epoch: [153][ 340/ 1200] Overall Loss 0.150922 Objective Loss 0.150922 LR 0.000250 Time 0.021330 -2022-12-06 11:31:01,639 - Epoch: [153][ 350/ 1200] Overall Loss 0.151254 Objective Loss 0.151254 LR 0.000250 Time 0.021268 -2022-12-06 11:31:01,831 - Epoch: [153][ 360/ 1200] Overall Loss 0.151691 Objective Loss 0.151691 LR 0.000250 Time 0.021207 -2022-12-06 11:31:02,023 - Epoch: [153][ 370/ 1200] Overall Loss 0.151353 Objective Loss 0.151353 LR 0.000250 Time 0.021151 -2022-12-06 11:31:02,215 - Epoch: [153][ 380/ 1200] Overall Loss 0.151231 Objective Loss 0.151231 LR 0.000250 Time 0.021098 -2022-12-06 11:31:02,407 - Epoch: [153][ 390/ 1200] Overall Loss 0.151080 Objective Loss 0.151080 LR 0.000250 Time 0.021049 -2022-12-06 11:31:02,599 - Epoch: [153][ 400/ 1200] Overall Loss 0.151433 Objective Loss 0.151433 LR 0.000250 Time 0.021001 -2022-12-06 11:31:02,791 - Epoch: [153][ 410/ 1200] Overall Loss 0.151513 Objective Loss 0.151513 LR 0.000250 Time 0.020957 -2022-12-06 11:31:02,983 - Epoch: [153][ 420/ 1200] Overall Loss 0.151368 Objective Loss 0.151368 LR 0.000250 Time 0.020913 -2022-12-06 11:31:03,175 - Epoch: [153][ 430/ 1200] Overall Loss 0.151342 Objective Loss 0.151342 LR 0.000250 Time 0.020871 -2022-12-06 11:31:03,366 - Epoch: [153][ 440/ 1200] Overall Loss 0.151390 Objective Loss 0.151390 LR 0.000250 Time 0.020830 -2022-12-06 11:31:03,558 - Epoch: [153][ 450/ 1200] Overall Loss 0.151507 Objective Loss 0.151507 LR 0.000250 Time 0.020793 -2022-12-06 11:31:03,749 - Epoch: [153][ 460/ 1200] Overall Loss 0.151124 Objective Loss 0.151124 LR 0.000250 Time 0.020756 -2022-12-06 11:31:03,941 - Epoch: [153][ 470/ 1200] Overall Loss 0.151067 Objective Loss 0.151067 LR 0.000250 Time 0.020721 -2022-12-06 11:31:04,133 - Epoch: [153][ 480/ 1200] Overall Loss 0.151060 Objective Loss 0.151060 LR 0.000250 Time 0.020688 -2022-12-06 11:31:04,325 - Epoch: [153][ 490/ 1200] Overall Loss 0.151017 Objective Loss 0.151017 LR 0.000250 Time 0.020656 -2022-12-06 11:31:04,516 - Epoch: [153][ 500/ 1200] Overall Loss 0.151339 Objective Loss 0.151339 LR 0.000250 Time 0.020624 -2022-12-06 11:31:04,708 - Epoch: [153][ 510/ 1200] Overall Loss 0.151359 Objective Loss 0.151359 LR 0.000250 Time 0.020594 -2022-12-06 11:31:04,900 - Epoch: [153][ 520/ 1200] Overall Loss 0.151769 Objective Loss 0.151769 LR 0.000250 Time 0.020567 -2022-12-06 11:31:05,092 - Epoch: [153][ 530/ 1200] Overall Loss 0.151632 Objective Loss 0.151632 LR 0.000250 Time 0.020540 -2022-12-06 11:31:05,283 - Epoch: [153][ 540/ 1200] Overall Loss 0.151787 Objective Loss 0.151787 LR 0.000250 Time 0.020513 -2022-12-06 11:31:05,475 - Epoch: [153][ 550/ 1200] Overall Loss 0.151449 Objective Loss 0.151449 LR 0.000250 Time 0.020488 -2022-12-06 11:31:05,668 - Epoch: [153][ 560/ 1200] Overall Loss 0.151671 Objective Loss 0.151671 LR 0.000250 Time 0.020465 -2022-12-06 11:31:05,860 - Epoch: [153][ 570/ 1200] Overall Loss 0.151517 Objective Loss 0.151517 LR 0.000250 Time 0.020442 -2022-12-06 11:31:06,052 - Epoch: [153][ 580/ 1200] Overall Loss 0.151449 Objective Loss 0.151449 LR 0.000250 Time 0.020420 -2022-12-06 11:31:06,244 - Epoch: [153][ 590/ 1200] Overall Loss 0.151500 Objective Loss 0.151500 LR 0.000250 Time 0.020399 -2022-12-06 11:31:06,436 - Epoch: [153][ 600/ 1200] Overall Loss 0.151715 Objective Loss 0.151715 LR 0.000250 Time 0.020378 -2022-12-06 11:31:06,628 - Epoch: [153][ 610/ 1200] Overall Loss 0.151582 Objective Loss 0.151582 LR 0.000250 Time 0.020357 -2022-12-06 11:31:06,820 - Epoch: [153][ 620/ 1200] Overall Loss 0.151962 Objective Loss 0.151962 LR 0.000250 Time 0.020337 -2022-12-06 11:31:07,012 - Epoch: [153][ 630/ 1200] Overall Loss 0.152197 Objective Loss 0.152197 LR 0.000250 Time 0.020319 -2022-12-06 11:31:07,204 - Epoch: [153][ 640/ 1200] Overall Loss 0.152088 Objective Loss 0.152088 LR 0.000250 Time 0.020300 -2022-12-06 11:31:07,396 - Epoch: [153][ 650/ 1200] Overall Loss 0.152218 Objective Loss 0.152218 LR 0.000250 Time 0.020283 -2022-12-06 11:31:07,588 - Epoch: [153][ 660/ 1200] Overall Loss 0.152029 Objective Loss 0.152029 LR 0.000250 Time 0.020266 -2022-12-06 11:31:07,781 - Epoch: [153][ 670/ 1200] Overall Loss 0.152243 Objective Loss 0.152243 LR 0.000250 Time 0.020249 -2022-12-06 11:31:07,973 - Epoch: [153][ 680/ 1200] Overall Loss 0.152285 Objective Loss 0.152285 LR 0.000250 Time 0.020233 -2022-12-06 11:31:08,165 - Epoch: [153][ 690/ 1200] Overall Loss 0.152431 Objective Loss 0.152431 LR 0.000250 Time 0.020218 -2022-12-06 11:31:08,357 - Epoch: [153][ 700/ 1200] Overall Loss 0.152360 Objective Loss 0.152360 LR 0.000250 Time 0.020202 -2022-12-06 11:31:08,549 - Epoch: [153][ 710/ 1200] Overall Loss 0.152321 Objective Loss 0.152321 LR 0.000250 Time 0.020187 -2022-12-06 11:31:08,740 - Epoch: [153][ 720/ 1200] Overall Loss 0.152399 Objective Loss 0.152399 LR 0.000250 Time 0.020172 -2022-12-06 11:31:08,934 - Epoch: [153][ 730/ 1200] Overall Loss 0.152607 Objective Loss 0.152607 LR 0.000250 Time 0.020161 -2022-12-06 11:31:09,130 - Epoch: [153][ 740/ 1200] Overall Loss 0.152482 Objective Loss 0.152482 LR 0.000250 Time 0.020152 -2022-12-06 11:31:09,328 - Epoch: [153][ 750/ 1200] Overall Loss 0.152379 Objective Loss 0.152379 LR 0.000250 Time 0.020147 -2022-12-06 11:31:09,524 - Epoch: [153][ 760/ 1200] Overall Loss 0.152118 Objective Loss 0.152118 LR 0.000250 Time 0.020139 -2022-12-06 11:31:09,723 - Epoch: [153][ 770/ 1200] Overall Loss 0.152175 Objective Loss 0.152175 LR 0.000250 Time 0.020134 -2022-12-06 11:31:09,918 - Epoch: [153][ 780/ 1200] Overall Loss 0.151890 Objective Loss 0.151890 LR 0.000250 Time 0.020126 -2022-12-06 11:31:10,117 - Epoch: [153][ 790/ 1200] Overall Loss 0.151645 Objective Loss 0.151645 LR 0.000250 Time 0.020123 -2022-12-06 11:31:10,314 - Epoch: [153][ 800/ 1200] Overall Loss 0.151531 Objective Loss 0.151531 LR 0.000250 Time 0.020116 -2022-12-06 11:31:10,512 - Epoch: [153][ 810/ 1200] Overall Loss 0.151586 Objective Loss 0.151586 LR 0.000250 Time 0.020112 -2022-12-06 11:31:10,708 - Epoch: [153][ 820/ 1200] Overall Loss 0.151532 Objective Loss 0.151532 LR 0.000250 Time 0.020104 -2022-12-06 11:31:10,906 - Epoch: [153][ 830/ 1200] Overall Loss 0.151880 Objective Loss 0.151880 LR 0.000250 Time 0.020101 -2022-12-06 11:31:11,102 - Epoch: [153][ 840/ 1200] Overall Loss 0.151803 Objective Loss 0.151803 LR 0.000250 Time 0.020095 -2022-12-06 11:31:11,300 - Epoch: [153][ 850/ 1200] Overall Loss 0.151848 Objective Loss 0.151848 LR 0.000250 Time 0.020091 -2022-12-06 11:31:11,496 - Epoch: [153][ 860/ 1200] Overall Loss 0.152040 Objective Loss 0.152040 LR 0.000250 Time 0.020083 -2022-12-06 11:31:11,694 - Epoch: [153][ 870/ 1200] Overall Loss 0.151881 Objective Loss 0.151881 LR 0.000250 Time 0.020080 -2022-12-06 11:31:11,891 - Epoch: [153][ 880/ 1200] Overall Loss 0.151919 Objective Loss 0.151919 LR 0.000250 Time 0.020074 -2022-12-06 11:31:12,088 - Epoch: [153][ 890/ 1200] Overall Loss 0.151850 Objective Loss 0.151850 LR 0.000250 Time 0.020070 -2022-12-06 11:31:12,284 - Epoch: [153][ 900/ 1200] Overall Loss 0.151752 Objective Loss 0.151752 LR 0.000250 Time 0.020064 -2022-12-06 11:31:12,481 - Epoch: [153][ 910/ 1200] Overall Loss 0.151787 Objective Loss 0.151787 LR 0.000250 Time 0.020060 -2022-12-06 11:31:12,677 - Epoch: [153][ 920/ 1200] Overall Loss 0.151955 Objective Loss 0.151955 LR 0.000250 Time 0.020054 -2022-12-06 11:31:12,875 - Epoch: [153][ 930/ 1200] Overall Loss 0.151809 Objective Loss 0.151809 LR 0.000250 Time 0.020050 -2022-12-06 11:31:13,070 - Epoch: [153][ 940/ 1200] Overall Loss 0.151749 Objective Loss 0.151749 LR 0.000250 Time 0.020044 -2022-12-06 11:31:13,268 - Epoch: [153][ 950/ 1200] Overall Loss 0.151803 Objective Loss 0.151803 LR 0.000250 Time 0.020041 -2022-12-06 11:31:13,464 - Epoch: [153][ 960/ 1200] Overall Loss 0.151824 Objective Loss 0.151824 LR 0.000250 Time 0.020035 -2022-12-06 11:31:13,662 - Epoch: [153][ 970/ 1200] Overall Loss 0.151880 Objective Loss 0.151880 LR 0.000250 Time 0.020033 -2022-12-06 11:31:13,858 - Epoch: [153][ 980/ 1200] Overall Loss 0.151893 Objective Loss 0.151893 LR 0.000250 Time 0.020028 -2022-12-06 11:31:14,056 - Epoch: [153][ 990/ 1200] Overall Loss 0.151835 Objective Loss 0.151835 LR 0.000250 Time 0.020025 -2022-12-06 11:31:14,253 - Epoch: [153][ 1000/ 1200] Overall Loss 0.151928 Objective Loss 0.151928 LR 0.000250 Time 0.020021 -2022-12-06 11:31:14,451 - Epoch: [153][ 1010/ 1200] Overall Loss 0.151874 Objective Loss 0.151874 LR 0.000250 Time 0.020018 -2022-12-06 11:31:14,647 - Epoch: [153][ 1020/ 1200] Overall Loss 0.151897 Objective Loss 0.151897 LR 0.000250 Time 0.020014 -2022-12-06 11:31:14,845 - Epoch: [153][ 1030/ 1200] Overall Loss 0.151958 Objective Loss 0.151958 LR 0.000250 Time 0.020012 -2022-12-06 11:31:15,041 - Epoch: [153][ 1040/ 1200] Overall Loss 0.151855 Objective Loss 0.151855 LR 0.000250 Time 0.020007 -2022-12-06 11:31:15,239 - Epoch: [153][ 1050/ 1200] Overall Loss 0.152037 Objective Loss 0.152037 LR 0.000250 Time 0.020004 -2022-12-06 11:31:15,435 - Epoch: [153][ 1060/ 1200] Overall Loss 0.151866 Objective Loss 0.151866 LR 0.000250 Time 0.020000 -2022-12-06 11:31:15,633 - Epoch: [153][ 1070/ 1200] Overall Loss 0.151941 Objective Loss 0.151941 LR 0.000250 Time 0.019998 -2022-12-06 11:31:15,829 - Epoch: [153][ 1080/ 1200] Overall Loss 0.151931 Objective Loss 0.151931 LR 0.000250 Time 0.019993 -2022-12-06 11:31:16,027 - Epoch: [153][ 1090/ 1200] Overall Loss 0.151903 Objective Loss 0.151903 LR 0.000250 Time 0.019991 -2022-12-06 11:31:16,223 - Epoch: [153][ 1100/ 1200] Overall Loss 0.151832 Objective Loss 0.151832 LR 0.000250 Time 0.019986 -2022-12-06 11:31:16,420 - Epoch: [153][ 1110/ 1200] Overall Loss 0.151822 Objective Loss 0.151822 LR 0.000250 Time 0.019984 -2022-12-06 11:31:16,616 - Epoch: [153][ 1120/ 1200] Overall Loss 0.151835 Objective Loss 0.151835 LR 0.000250 Time 0.019980 -2022-12-06 11:31:16,814 - Epoch: [153][ 1130/ 1200] Overall Loss 0.151773 Objective Loss 0.151773 LR 0.000250 Time 0.019978 -2022-12-06 11:31:17,010 - Epoch: [153][ 1140/ 1200] Overall Loss 0.151771 Objective Loss 0.151771 LR 0.000250 Time 0.019974 -2022-12-06 11:31:17,208 - Epoch: [153][ 1150/ 1200] Overall Loss 0.151759 Objective Loss 0.151759 LR 0.000250 Time 0.019972 -2022-12-06 11:31:17,403 - Epoch: [153][ 1160/ 1200] Overall Loss 0.151761 Objective Loss 0.151761 LR 0.000250 Time 0.019968 -2022-12-06 11:31:17,601 - Epoch: [153][ 1170/ 1200] Overall Loss 0.151747 Objective Loss 0.151747 LR 0.000250 Time 0.019966 -2022-12-06 11:31:17,798 - Epoch: [153][ 1180/ 1200] Overall Loss 0.151872 Objective Loss 0.151872 LR 0.000250 Time 0.019962 -2022-12-06 11:31:17,996 - Epoch: [153][ 1190/ 1200] Overall Loss 0.151719 Objective Loss 0.151719 LR 0.000250 Time 0.019961 -2022-12-06 11:31:18,226 - Epoch: [153][ 1200/ 1200] Overall Loss 0.151807 Objective Loss 0.151807 Top1 91.841004 Top5 99.372385 LR 0.000250 Time 0.019986 -2022-12-06 11:31:18,315 - --- validate (epoch=153)----------- -2022-12-06 11:31:18,315 - 34129 samples (256 per mini-batch) -2022-12-06 11:31:18,773 - Epoch: [153][ 10/ 134] Loss 0.221419 Top1 88.125000 Top5 98.359375 -2022-12-06 11:31:18,917 - Epoch: [153][ 20/ 134] Loss 0.230394 Top1 88.320312 Top5 98.457031 -2022-12-06 11:31:19,058 - Epoch: [153][ 30/ 134] Loss 0.234032 Top1 87.799479 Top5 98.515625 -2022-12-06 11:31:19,193 - Epoch: [153][ 40/ 134] Loss 0.233660 Top1 87.646484 Top5 98.593750 -2022-12-06 11:31:19,334 - Epoch: [153][ 50/ 134] Loss 0.234240 Top1 87.523438 Top5 98.500000 -2022-12-06 11:31:19,476 - Epoch: [153][ 60/ 134] Loss 0.229437 Top1 87.545573 Top5 98.483073 -2022-12-06 11:31:19,620 - Epoch: [153][ 70/ 134] Loss 0.229970 Top1 87.494420 Top5 98.470982 -2022-12-06 11:31:19,752 - Epoch: [153][ 80/ 134] Loss 0.232572 Top1 87.436523 Top5 98.461914 -2022-12-06 11:31:19,886 - Epoch: [153][ 90/ 134] Loss 0.231440 Top1 87.395833 Top5 98.498264 -2022-12-06 11:31:20,018 - Epoch: [153][ 100/ 134] Loss 0.233524 Top1 87.441406 Top5 98.496094 -2022-12-06 11:31:20,152 - Epoch: [153][ 110/ 134] Loss 0.232693 Top1 87.492898 Top5 98.501420 -2022-12-06 11:31:20,295 - Epoch: [153][ 120/ 134] Loss 0.234069 Top1 87.503255 Top5 98.538411 -2022-12-06 11:31:20,432 - Epoch: [153][ 130/ 134] Loss 0.233971 Top1 87.463942 Top5 98.551683 -2022-12-06 11:31:20,469 - Epoch: [153][ 134/ 134] Loss 0.233634 Top1 87.482786 Top5 98.555481 -2022-12-06 11:31:20,558 - ==> Top1: 87.483 Top5: 98.555 Loss: 0.234 - -2022-12-06 11:31:20,559 - ==> Confusion: -[[ 908 0 1 1 3 4 0 1 9 52 0 2 0 3 5 2 1 0 1 0 3] - [ 1 942 2 3 7 16 2 10 2 0 3 4 2 1 0 1 5 0 12 5 9] - [ 3 3 1015 9 4 2 21 7 0 3 7 3 1 0 0 2 1 2 4 3 13] - [ 1 2 15 955 1 1 0 1 1 1 8 0 3 1 7 0 2 2 13 0 6] - [ 7 7 1 0 960 3 0 2 1 7 2 1 1 2 10 3 5 2 1 1 4] - [ 1 12 0 3 7 972 3 20 2 3 1 10 3 13 1 1 2 3 1 6 5] - [ 0 1 6 4 3 0 1082 1 0 0 0 2 0 1 0 3 1 1 2 9 2] - [ 1 10 5 3 3 18 6 966 1 0 1 4 1 2 0 0 0 0 19 10 4] - [ 4 2 0 0 1 1 1 0 985 34 11 1 1 8 6 0 1 0 3 2 3] - [ 41 1 2 0 2 1 0 1 25 908 1 2 1 10 1 0 0 0 1 0 4] - [ 1 1 2 2 2 1 0 3 9 1 978 0 0 8 2 1 0 1 2 2 3] - [ 3 1 1 0 0 7 5 1 2 0 0 979 19 6 0 1 4 5 0 14 3] - [ 0 1 0 2 0 2 0 2 1 1 0 34 890 1 2 9 1 13 0 4 6] - [ 1 0 0 0 2 4 0 2 11 14 6 2 2 964 1 1 2 1 0 4 6] - [ 6 3 3 12 3 2 0 0 21 2 0 3 2 4 1053 0 0 1 10 0 5] - [ 1 0 1 0 3 1 3 0 0 1 1 9 6 1 0 992 5 13 0 4 2] - [ 2 2 1 1 3 0 2 0 1 0 0 3 4 2 0 5 1038 0 1 2 5] - [ 3 0 1 3 0 0 1 1 1 3 0 7 7 2 2 12 0 986 0 3 4] - [ 2 2 2 7 1 4 0 20 1 1 2 1 3 0 6 0 0 1 951 1 3] - [ 2 1 2 2 1 3 2 5 0 0 4 10 3 7 1 3 2 2 0 1025 5] - [ 102 182 166 111 98 122 91 131 91 95 185 102 287 295 130 92 179 82 157 224 10304]] - -2022-12-06 11:31:21,216 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:31:21,216 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:31:21,222 - - -2022-12-06 11:31:21,222 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:31:22,156 - Epoch: [154][ 10/ 1200] Overall Loss 0.152013 Objective Loss 0.152013 LR 0.000250 Time 0.093289 -2022-12-06 11:31:22,356 - Epoch: [154][ 20/ 1200] Overall Loss 0.151146 Objective Loss 0.151146 LR 0.000250 Time 0.056636 -2022-12-06 11:31:22,553 - Epoch: [154][ 30/ 1200] Overall Loss 0.156209 Objective Loss 0.156209 LR 0.000250 Time 0.044306 -2022-12-06 11:31:22,750 - Epoch: [154][ 40/ 1200] Overall Loss 0.151181 Objective Loss 0.151181 LR 0.000250 Time 0.038137 -2022-12-06 11:31:22,943 - Epoch: [154][ 50/ 1200] Overall Loss 0.152879 Objective Loss 0.152879 LR 0.000250 Time 0.034342 -2022-12-06 11:31:23,133 - Epoch: [154][ 60/ 1200] Overall Loss 0.156632 Objective Loss 0.156632 LR 0.000250 Time 0.031779 -2022-12-06 11:31:23,323 - Epoch: [154][ 70/ 1200] Overall Loss 0.156831 Objective Loss 0.156831 LR 0.000250 Time 0.029952 -2022-12-06 11:31:23,514 - Epoch: [154][ 80/ 1200] Overall Loss 0.156895 Objective Loss 0.156895 LR 0.000250 Time 0.028583 -2022-12-06 11:31:23,705 - Epoch: [154][ 90/ 1200] Overall Loss 0.155153 Objective Loss 0.155153 LR 0.000250 Time 0.027523 -2022-12-06 11:31:23,896 - Epoch: [154][ 100/ 1200] Overall Loss 0.152407 Objective Loss 0.152407 LR 0.000250 Time 0.026677 -2022-12-06 11:31:24,086 - Epoch: [154][ 110/ 1200] Overall Loss 0.150617 Objective Loss 0.150617 LR 0.000250 Time 0.025979 -2022-12-06 11:31:24,277 - Epoch: [154][ 120/ 1200] Overall Loss 0.151902 Objective Loss 0.151902 LR 0.000250 Time 0.025398 -2022-12-06 11:31:24,467 - Epoch: [154][ 130/ 1200] Overall Loss 0.151387 Objective Loss 0.151387 LR 0.000250 Time 0.024904 -2022-12-06 11:31:24,657 - Epoch: [154][ 140/ 1200] Overall Loss 0.150432 Objective Loss 0.150432 LR 0.000250 Time 0.024480 -2022-12-06 11:31:24,848 - Epoch: [154][ 150/ 1200] Overall Loss 0.150895 Objective Loss 0.150895 LR 0.000250 Time 0.024116 -2022-12-06 11:31:25,039 - Epoch: [154][ 160/ 1200] Overall Loss 0.151524 Objective Loss 0.151524 LR 0.000250 Time 0.023798 -2022-12-06 11:31:25,230 - Epoch: [154][ 170/ 1200] Overall Loss 0.151406 Objective Loss 0.151406 LR 0.000250 Time 0.023519 -2022-12-06 11:31:25,421 - Epoch: [154][ 180/ 1200] Overall Loss 0.151647 Objective Loss 0.151647 LR 0.000250 Time 0.023268 -2022-12-06 11:31:25,611 - Epoch: [154][ 190/ 1200] Overall Loss 0.151097 Objective Loss 0.151097 LR 0.000250 Time 0.023044 -2022-12-06 11:31:25,802 - Epoch: [154][ 200/ 1200] Overall Loss 0.150281 Objective Loss 0.150281 LR 0.000250 Time 0.022844 -2022-12-06 11:31:25,993 - Epoch: [154][ 210/ 1200] Overall Loss 0.149848 Objective Loss 0.149848 LR 0.000250 Time 0.022660 -2022-12-06 11:31:26,184 - Epoch: [154][ 220/ 1200] Overall Loss 0.148443 Objective Loss 0.148443 LR 0.000250 Time 0.022497 -2022-12-06 11:31:26,374 - Epoch: [154][ 230/ 1200] Overall Loss 0.148454 Objective Loss 0.148454 LR 0.000250 Time 0.022344 -2022-12-06 11:31:26,565 - Epoch: [154][ 240/ 1200] Overall Loss 0.147419 Objective Loss 0.147419 LR 0.000250 Time 0.022204 -2022-12-06 11:31:26,755 - Epoch: [154][ 250/ 1200] Overall Loss 0.148193 Objective Loss 0.148193 LR 0.000250 Time 0.022075 -2022-12-06 11:31:26,945 - Epoch: [154][ 260/ 1200] Overall Loss 0.148122 Objective Loss 0.148122 LR 0.000250 Time 0.021957 -2022-12-06 11:31:27,135 - Epoch: [154][ 270/ 1200] Overall Loss 0.148012 Objective Loss 0.148012 LR 0.000250 Time 0.021845 -2022-12-06 11:31:27,325 - Epoch: [154][ 280/ 1200] Overall Loss 0.147539 Objective Loss 0.147539 LR 0.000250 Time 0.021742 -2022-12-06 11:31:27,515 - Epoch: [154][ 290/ 1200] Overall Loss 0.147131 Objective Loss 0.147131 LR 0.000250 Time 0.021645 -2022-12-06 11:31:27,706 - Epoch: [154][ 300/ 1200] Overall Loss 0.147238 Objective Loss 0.147238 LR 0.000250 Time 0.021556 -2022-12-06 11:31:27,896 - Epoch: [154][ 310/ 1200] Overall Loss 0.147399 Objective Loss 0.147399 LR 0.000250 Time 0.021472 -2022-12-06 11:31:28,086 - Epoch: [154][ 320/ 1200] Overall Loss 0.147418 Objective Loss 0.147418 LR 0.000250 Time 0.021393 -2022-12-06 11:31:28,276 - Epoch: [154][ 330/ 1200] Overall Loss 0.147844 Objective Loss 0.147844 LR 0.000250 Time 0.021319 -2022-12-06 11:31:28,466 - Epoch: [154][ 340/ 1200] Overall Loss 0.148251 Objective Loss 0.148251 LR 0.000250 Time 0.021249 -2022-12-06 11:31:28,656 - Epoch: [154][ 350/ 1200] Overall Loss 0.147839 Objective Loss 0.147839 LR 0.000250 Time 0.021183 -2022-12-06 11:31:28,846 - Epoch: [154][ 360/ 1200] Overall Loss 0.148167 Objective Loss 0.148167 LR 0.000250 Time 0.021122 -2022-12-06 11:31:29,037 - Epoch: [154][ 370/ 1200] Overall Loss 0.148638 Objective Loss 0.148638 LR 0.000250 Time 0.021065 -2022-12-06 11:31:29,227 - Epoch: [154][ 380/ 1200] Overall Loss 0.148609 Objective Loss 0.148609 LR 0.000250 Time 0.021010 -2022-12-06 11:31:29,417 - Epoch: [154][ 390/ 1200] Overall Loss 0.148583 Objective Loss 0.148583 LR 0.000250 Time 0.020959 -2022-12-06 11:31:29,609 - Epoch: [154][ 400/ 1200] Overall Loss 0.148672 Objective Loss 0.148672 LR 0.000250 Time 0.020911 -2022-12-06 11:31:29,799 - Epoch: [154][ 410/ 1200] Overall Loss 0.148957 Objective Loss 0.148957 LR 0.000250 Time 0.020864 -2022-12-06 11:31:29,989 - Epoch: [154][ 420/ 1200] Overall Loss 0.148708 Objective Loss 0.148708 LR 0.000250 Time 0.020819 -2022-12-06 11:31:30,179 - Epoch: [154][ 430/ 1200] Overall Loss 0.148527 Objective Loss 0.148527 LR 0.000250 Time 0.020775 -2022-12-06 11:31:30,369 - Epoch: [154][ 440/ 1200] Overall Loss 0.148607 Objective Loss 0.148607 LR 0.000250 Time 0.020734 -2022-12-06 11:31:30,559 - Epoch: [154][ 450/ 1200] Overall Loss 0.148810 Objective Loss 0.148810 LR 0.000250 Time 0.020694 -2022-12-06 11:31:30,750 - Epoch: [154][ 460/ 1200] Overall Loss 0.149194 Objective Loss 0.149194 LR 0.000250 Time 0.020658 -2022-12-06 11:31:30,940 - Epoch: [154][ 470/ 1200] Overall Loss 0.149450 Objective Loss 0.149450 LR 0.000250 Time 0.020621 -2022-12-06 11:31:31,130 - Epoch: [154][ 480/ 1200] Overall Loss 0.149851 Objective Loss 0.149851 LR 0.000250 Time 0.020587 -2022-12-06 11:31:31,321 - Epoch: [154][ 490/ 1200] Overall Loss 0.150036 Objective Loss 0.150036 LR 0.000250 Time 0.020554 -2022-12-06 11:31:31,511 - Epoch: [154][ 500/ 1200] Overall Loss 0.150008 Objective Loss 0.150008 LR 0.000250 Time 0.020523 -2022-12-06 11:31:31,702 - Epoch: [154][ 510/ 1200] Overall Loss 0.150234 Objective Loss 0.150234 LR 0.000250 Time 0.020493 -2022-12-06 11:31:31,892 - Epoch: [154][ 520/ 1200] Overall Loss 0.150389 Objective Loss 0.150389 LR 0.000250 Time 0.020465 -2022-12-06 11:31:32,083 - Epoch: [154][ 530/ 1200] Overall Loss 0.150868 Objective Loss 0.150868 LR 0.000250 Time 0.020437 -2022-12-06 11:31:32,273 - Epoch: [154][ 540/ 1200] Overall Loss 0.150987 Objective Loss 0.150987 LR 0.000250 Time 0.020410 -2022-12-06 11:31:32,463 - Epoch: [154][ 550/ 1200] Overall Loss 0.151155 Objective Loss 0.151155 LR 0.000250 Time 0.020383 -2022-12-06 11:31:32,654 - Epoch: [154][ 560/ 1200] Overall Loss 0.151231 Objective Loss 0.151231 LR 0.000250 Time 0.020359 -2022-12-06 11:31:32,844 - Epoch: [154][ 570/ 1200] Overall Loss 0.151575 Objective Loss 0.151575 LR 0.000250 Time 0.020334 -2022-12-06 11:31:33,035 - Epoch: [154][ 580/ 1200] Overall Loss 0.151475 Objective Loss 0.151475 LR 0.000250 Time 0.020312 -2022-12-06 11:31:33,226 - Epoch: [154][ 590/ 1200] Overall Loss 0.151068 Objective Loss 0.151068 LR 0.000250 Time 0.020290 -2022-12-06 11:31:33,417 - Epoch: [154][ 600/ 1200] Overall Loss 0.151133 Objective Loss 0.151133 LR 0.000250 Time 0.020270 -2022-12-06 11:31:33,607 - Epoch: [154][ 610/ 1200] Overall Loss 0.150843 Objective Loss 0.150843 LR 0.000250 Time 0.020249 -2022-12-06 11:31:33,799 - Epoch: [154][ 620/ 1200] Overall Loss 0.150732 Objective Loss 0.150732 LR 0.000250 Time 0.020231 -2022-12-06 11:31:33,990 - Epoch: [154][ 630/ 1200] Overall Loss 0.150731 Objective Loss 0.150731 LR 0.000250 Time 0.020212 -2022-12-06 11:31:34,180 - Epoch: [154][ 640/ 1200] Overall Loss 0.150724 Objective Loss 0.150724 LR 0.000250 Time 0.020193 -2022-12-06 11:31:34,371 - Epoch: [154][ 650/ 1200] Overall Loss 0.150805 Objective Loss 0.150805 LR 0.000250 Time 0.020174 -2022-12-06 11:31:34,561 - Epoch: [154][ 660/ 1200] Overall Loss 0.150841 Objective Loss 0.150841 LR 0.000250 Time 0.020157 -2022-12-06 11:31:34,752 - Epoch: [154][ 670/ 1200] Overall Loss 0.151041 Objective Loss 0.151041 LR 0.000250 Time 0.020139 -2022-12-06 11:31:34,943 - Epoch: [154][ 680/ 1200] Overall Loss 0.151045 Objective Loss 0.151045 LR 0.000250 Time 0.020123 -2022-12-06 11:31:35,133 - Epoch: [154][ 690/ 1200] Overall Loss 0.151119 Objective Loss 0.151119 LR 0.000250 Time 0.020107 -2022-12-06 11:31:35,325 - Epoch: [154][ 700/ 1200] Overall Loss 0.151480 Objective Loss 0.151480 LR 0.000250 Time 0.020092 -2022-12-06 11:31:35,515 - Epoch: [154][ 710/ 1200] Overall Loss 0.151546 Objective Loss 0.151546 LR 0.000250 Time 0.020077 -2022-12-06 11:31:35,705 - Epoch: [154][ 720/ 1200] Overall Loss 0.151712 Objective Loss 0.151712 LR 0.000250 Time 0.020061 -2022-12-06 11:31:35,896 - Epoch: [154][ 730/ 1200] Overall Loss 0.151618 Objective Loss 0.151618 LR 0.000250 Time 0.020046 -2022-12-06 11:31:36,086 - Epoch: [154][ 740/ 1200] Overall Loss 0.151700 Objective Loss 0.151700 LR 0.000250 Time 0.020032 -2022-12-06 11:31:36,277 - Epoch: [154][ 750/ 1200] Overall Loss 0.151771 Objective Loss 0.151771 LR 0.000250 Time 0.020019 -2022-12-06 11:31:36,468 - Epoch: [154][ 760/ 1200] Overall Loss 0.151660 Objective Loss 0.151660 LR 0.000250 Time 0.020006 -2022-12-06 11:31:36,659 - Epoch: [154][ 770/ 1200] Overall Loss 0.151395 Objective Loss 0.151395 LR 0.000250 Time 0.019993 -2022-12-06 11:31:36,849 - Epoch: [154][ 780/ 1200] Overall Loss 0.151287 Objective Loss 0.151287 LR 0.000250 Time 0.019981 -2022-12-06 11:31:37,039 - Epoch: [154][ 790/ 1200] Overall Loss 0.151518 Objective Loss 0.151518 LR 0.000250 Time 0.019967 -2022-12-06 11:31:37,230 - Epoch: [154][ 800/ 1200] Overall Loss 0.151585 Objective Loss 0.151585 LR 0.000250 Time 0.019956 -2022-12-06 11:31:37,421 - Epoch: [154][ 810/ 1200] Overall Loss 0.151760 Objective Loss 0.151760 LR 0.000250 Time 0.019944 -2022-12-06 11:31:37,612 - Epoch: [154][ 820/ 1200] Overall Loss 0.152098 Objective Loss 0.152098 LR 0.000250 Time 0.019933 -2022-12-06 11:31:37,803 - Epoch: [154][ 830/ 1200] Overall Loss 0.152120 Objective Loss 0.152120 LR 0.000250 Time 0.019922 -2022-12-06 11:31:37,994 - Epoch: [154][ 840/ 1200] Overall Loss 0.152419 Objective Loss 0.152419 LR 0.000250 Time 0.019912 -2022-12-06 11:31:38,184 - Epoch: [154][ 850/ 1200] Overall Loss 0.152528 Objective Loss 0.152528 LR 0.000250 Time 0.019901 -2022-12-06 11:31:38,375 - Epoch: [154][ 860/ 1200] Overall Loss 0.152716 Objective Loss 0.152716 LR 0.000250 Time 0.019890 -2022-12-06 11:31:38,565 - Epoch: [154][ 870/ 1200] Overall Loss 0.152622 Objective Loss 0.152622 LR 0.000250 Time 0.019880 -2022-12-06 11:31:38,756 - Epoch: [154][ 880/ 1200] Overall Loss 0.152579 Objective Loss 0.152579 LR 0.000250 Time 0.019871 -2022-12-06 11:31:38,947 - Epoch: [154][ 890/ 1200] Overall Loss 0.152651 Objective Loss 0.152651 LR 0.000250 Time 0.019861 -2022-12-06 11:31:39,138 - Epoch: [154][ 900/ 1200] Overall Loss 0.152680 Objective Loss 0.152680 LR 0.000250 Time 0.019852 -2022-12-06 11:31:39,329 - Epoch: [154][ 910/ 1200] Overall Loss 0.152756 Objective Loss 0.152756 LR 0.000250 Time 0.019843 -2022-12-06 11:31:39,520 - Epoch: [154][ 920/ 1200] Overall Loss 0.152721 Objective Loss 0.152721 LR 0.000250 Time 0.019835 -2022-12-06 11:31:39,711 - Epoch: [154][ 930/ 1200] Overall Loss 0.152830 Objective Loss 0.152830 LR 0.000250 Time 0.019826 -2022-12-06 11:31:39,901 - Epoch: [154][ 940/ 1200] Overall Loss 0.152648 Objective Loss 0.152648 LR 0.000250 Time 0.019817 -2022-12-06 11:31:40,091 - Epoch: [154][ 950/ 1200] Overall Loss 0.152761 Objective Loss 0.152761 LR 0.000250 Time 0.019808 -2022-12-06 11:31:40,282 - Epoch: [154][ 960/ 1200] Overall Loss 0.152640 Objective Loss 0.152640 LR 0.000250 Time 0.019799 -2022-12-06 11:31:40,472 - Epoch: [154][ 970/ 1200] Overall Loss 0.152506 Objective Loss 0.152506 LR 0.000250 Time 0.019791 -2022-12-06 11:31:40,663 - Epoch: [154][ 980/ 1200] Overall Loss 0.152505 Objective Loss 0.152505 LR 0.000250 Time 0.019783 -2022-12-06 11:31:40,853 - Epoch: [154][ 990/ 1200] Overall Loss 0.152539 Objective Loss 0.152539 LR 0.000250 Time 0.019775 -2022-12-06 11:31:41,044 - Epoch: [154][ 1000/ 1200] Overall Loss 0.152519 Objective Loss 0.152519 LR 0.000250 Time 0.019768 -2022-12-06 11:31:41,235 - Epoch: [154][ 1010/ 1200] Overall Loss 0.152549 Objective Loss 0.152549 LR 0.000250 Time 0.019760 -2022-12-06 11:31:41,426 - Epoch: [154][ 1020/ 1200] Overall Loss 0.152594 Objective Loss 0.152594 LR 0.000250 Time 0.019753 -2022-12-06 11:31:41,616 - Epoch: [154][ 1030/ 1200] Overall Loss 0.152426 Objective Loss 0.152426 LR 0.000250 Time 0.019746 -2022-12-06 11:31:41,807 - Epoch: [154][ 1040/ 1200] Overall Loss 0.152327 Objective Loss 0.152327 LR 0.000250 Time 0.019739 -2022-12-06 11:31:41,997 - Epoch: [154][ 1050/ 1200] Overall Loss 0.152306 Objective Loss 0.152306 LR 0.000250 Time 0.019732 -2022-12-06 11:31:42,188 - Epoch: [154][ 1060/ 1200] Overall Loss 0.152126 Objective Loss 0.152126 LR 0.000250 Time 0.019725 -2022-12-06 11:31:42,378 - Epoch: [154][ 1070/ 1200] Overall Loss 0.152099 Objective Loss 0.152099 LR 0.000250 Time 0.019718 -2022-12-06 11:31:42,569 - Epoch: [154][ 1080/ 1200] Overall Loss 0.152089 Objective Loss 0.152089 LR 0.000250 Time 0.019712 -2022-12-06 11:31:42,759 - Epoch: [154][ 1090/ 1200] Overall Loss 0.151977 Objective Loss 0.151977 LR 0.000250 Time 0.019705 -2022-12-06 11:31:42,950 - Epoch: [154][ 1100/ 1200] Overall Loss 0.151864 Objective Loss 0.151864 LR 0.000250 Time 0.019698 -2022-12-06 11:31:43,141 - Epoch: [154][ 1110/ 1200] Overall Loss 0.151736 Objective Loss 0.151736 LR 0.000250 Time 0.019693 -2022-12-06 11:31:43,331 - Epoch: [154][ 1120/ 1200] Overall Loss 0.151681 Objective Loss 0.151681 LR 0.000250 Time 0.019686 -2022-12-06 11:31:43,522 - Epoch: [154][ 1130/ 1200] Overall Loss 0.151548 Objective Loss 0.151548 LR 0.000250 Time 0.019680 -2022-12-06 11:31:43,713 - Epoch: [154][ 1140/ 1200] Overall Loss 0.151618 Objective Loss 0.151618 LR 0.000250 Time 0.019675 -2022-12-06 11:31:43,903 - Epoch: [154][ 1150/ 1200] Overall Loss 0.151603 Objective Loss 0.151603 LR 0.000250 Time 0.019669 -2022-12-06 11:31:44,095 - Epoch: [154][ 1160/ 1200] Overall Loss 0.151759 Objective Loss 0.151759 LR 0.000250 Time 0.019664 -2022-12-06 11:31:44,286 - Epoch: [154][ 1170/ 1200] Overall Loss 0.151654 Objective Loss 0.151654 LR 0.000250 Time 0.019659 -2022-12-06 11:31:44,477 - Epoch: [154][ 1180/ 1200] Overall Loss 0.151777 Objective Loss 0.151777 LR 0.000250 Time 0.019654 -2022-12-06 11:31:44,668 - Epoch: [154][ 1190/ 1200] Overall Loss 0.151794 Objective Loss 0.151794 LR 0.000250 Time 0.019648 -2022-12-06 11:31:44,897 - Epoch: [154][ 1200/ 1200] Overall Loss 0.151844 Objective Loss 0.151844 Top1 90.585774 Top5 99.790795 LR 0.000250 Time 0.019675 -2022-12-06 11:31:44,986 - --- validate (epoch=154)----------- -2022-12-06 11:31:44,987 - 34129 samples (256 per mini-batch) -2022-12-06 11:31:45,433 - Epoch: [154][ 10/ 134] Loss 0.254095 Top1 87.304688 Top5 98.242188 -2022-12-06 11:31:45,564 - Epoch: [154][ 20/ 134] Loss 0.237036 Top1 87.343750 Top5 98.515625 -2022-12-06 11:31:45,697 - Epoch: [154][ 30/ 134] Loss 0.245973 Top1 86.966146 Top5 98.541667 -2022-12-06 11:31:45,829 - Epoch: [154][ 40/ 134] Loss 0.249000 Top1 86.972656 Top5 98.525391 -2022-12-06 11:31:45,961 - Epoch: [154][ 50/ 134] Loss 0.241499 Top1 87.171875 Top5 98.554688 -2022-12-06 11:31:46,092 - Epoch: [154][ 60/ 134] Loss 0.242661 Top1 87.089844 Top5 98.509115 -2022-12-06 11:31:46,226 - Epoch: [154][ 70/ 134] Loss 0.243928 Top1 87.075893 Top5 98.459821 -2022-12-06 11:31:46,355 - Epoch: [154][ 80/ 134] Loss 0.239458 Top1 87.192383 Top5 98.486328 -2022-12-06 11:31:46,487 - Epoch: [154][ 90/ 134] Loss 0.237280 Top1 87.296007 Top5 98.515625 -2022-12-06 11:31:46,619 - Epoch: [154][ 100/ 134] Loss 0.236139 Top1 87.281250 Top5 98.496094 -2022-12-06 11:31:46,752 - Epoch: [154][ 110/ 134] Loss 0.236426 Top1 87.325994 Top5 98.497869 -2022-12-06 11:31:46,883 - Epoch: [154][ 120/ 134] Loss 0.236663 Top1 87.353516 Top5 98.525391 -2022-12-06 11:31:47,014 - Epoch: [154][ 130/ 134] Loss 0.234287 Top1 87.415865 Top5 98.542668 -2022-12-06 11:31:47,051 - Epoch: [154][ 134/ 134] Loss 0.233330 Top1 87.397814 Top5 98.537900 -2022-12-06 11:31:47,141 - ==> Top1: 87.398 Top5: 98.538 Loss: 0.233 - -2022-12-06 11:31:47,142 - ==> Confusion: -[[ 919 1 1 2 6 6 0 0 2 40 0 3 0 3 4 1 0 0 2 0 6] - [ 1 952 3 2 10 17 0 9 2 0 2 4 1 0 0 3 3 0 9 4 5] - [ 5 2 1028 8 4 2 12 6 0 2 4 4 1 2 0 2 0 2 5 3 11] - [ 1 0 22 943 1 2 1 0 1 3 10 0 2 0 13 0 1 1 13 1 5] - [ 7 3 1 0 965 1 1 1 0 7 1 2 0 2 11 5 5 3 0 2 3] - [ 1 17 0 4 7 969 1 24 2 2 1 11 4 13 2 1 1 2 0 3 4] - [ 0 3 13 4 0 1 1072 1 0 0 1 1 0 1 0 4 0 3 2 10 2] - [ 2 7 4 2 5 19 3 970 1 0 1 5 0 2 0 1 0 0 23 4 5] - [ 9 3 0 0 0 2 1 0 978 44 7 1 1 7 4 1 0 0 1 1 4] - [ 62 0 0 0 2 2 0 3 18 894 1 2 0 9 1 1 0 0 0 1 5] - [ 2 1 6 0 1 0 2 5 10 1 961 0 1 11 4 1 0 0 5 0 8] - [ 4 0 2 1 1 10 4 1 0 0 0 981 18 3 2 4 4 5 0 8 3] - [ 0 1 1 2 0 2 0 2 1 2 0 34 895 2 2 10 1 7 1 3 3] - [ 2 0 0 0 1 5 0 3 14 16 3 3 2 959 2 1 3 0 0 1 8] - [ 6 1 3 8 3 2 0 1 21 4 1 1 2 5 1061 0 1 2 5 0 3] - [ 0 0 1 0 2 0 4 0 0 1 1 5 5 2 0 999 7 10 0 3 3] - [ 3 0 2 1 1 0 1 0 2 0 0 2 3 3 0 9 1037 0 0 3 5] - [ 3 0 1 3 1 1 2 1 1 5 0 7 9 1 5 16 1 973 0 2 4] - [ 1 4 2 7 1 1 1 21 1 0 2 3 3 0 9 0 0 2 946 0 4] - [ 1 4 1 1 0 6 4 11 2 1 3 15 6 7 0 4 4 2 1 1001 6] - [ 123 220 193 76 132 137 76 139 93 82 159 86 287 270 128 96 164 70 169 206 10320]] - -2022-12-06 11:31:47,810 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:31:47,811 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:31:47,816 - - -2022-12-06 11:31:47,816 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:31:48,761 - Epoch: [155][ 10/ 1200] Overall Loss 0.151318 Objective Loss 0.151318 LR 0.000250 Time 0.094340 -2022-12-06 11:31:48,959 - Epoch: [155][ 20/ 1200] Overall Loss 0.159443 Objective Loss 0.159443 LR 0.000250 Time 0.057065 -2022-12-06 11:31:49,153 - Epoch: [155][ 30/ 1200] Overall Loss 0.158515 Objective Loss 0.158515 LR 0.000250 Time 0.044489 -2022-12-06 11:31:49,347 - Epoch: [155][ 40/ 1200] Overall Loss 0.159434 Objective Loss 0.159434 LR 0.000250 Time 0.038199 -2022-12-06 11:31:49,540 - Epoch: [155][ 50/ 1200] Overall Loss 0.156255 Objective Loss 0.156255 LR 0.000250 Time 0.034414 -2022-12-06 11:31:49,735 - Epoch: [155][ 60/ 1200] Overall Loss 0.154204 Objective Loss 0.154204 LR 0.000250 Time 0.031918 -2022-12-06 11:31:49,929 - Epoch: [155][ 70/ 1200] Overall Loss 0.158041 Objective Loss 0.158041 LR 0.000250 Time 0.030116 -2022-12-06 11:31:50,123 - Epoch: [155][ 80/ 1200] Overall Loss 0.159202 Objective Loss 0.159202 LR 0.000250 Time 0.028775 -2022-12-06 11:31:50,316 - Epoch: [155][ 90/ 1200] Overall Loss 0.157263 Objective Loss 0.157263 LR 0.000250 Time 0.027722 -2022-12-06 11:31:50,511 - Epoch: [155][ 100/ 1200] Overall Loss 0.157188 Objective Loss 0.157188 LR 0.000250 Time 0.026888 -2022-12-06 11:31:50,704 - Epoch: [155][ 110/ 1200] Overall Loss 0.155961 Objective Loss 0.155961 LR 0.000250 Time 0.026197 -2022-12-06 11:31:50,898 - Epoch: [155][ 120/ 1200] Overall Loss 0.155733 Objective Loss 0.155733 LR 0.000250 Time 0.025627 -2022-12-06 11:31:51,092 - Epoch: [155][ 130/ 1200] Overall Loss 0.155890 Objective Loss 0.155890 LR 0.000250 Time 0.025139 -2022-12-06 11:31:51,286 - Epoch: [155][ 140/ 1200] Overall Loss 0.156585 Objective Loss 0.156585 LR 0.000250 Time 0.024724 -2022-12-06 11:31:51,479 - Epoch: [155][ 150/ 1200] Overall Loss 0.155408 Objective Loss 0.155408 LR 0.000250 Time 0.024361 -2022-12-06 11:31:51,673 - Epoch: [155][ 160/ 1200] Overall Loss 0.154533 Objective Loss 0.154533 LR 0.000250 Time 0.024048 -2022-12-06 11:31:51,866 - Epoch: [155][ 170/ 1200] Overall Loss 0.154517 Objective Loss 0.154517 LR 0.000250 Time 0.023768 -2022-12-06 11:31:52,061 - Epoch: [155][ 180/ 1200] Overall Loss 0.154844 Objective Loss 0.154844 LR 0.000250 Time 0.023524 -2022-12-06 11:31:52,255 - Epoch: [155][ 190/ 1200] Overall Loss 0.154704 Objective Loss 0.154704 LR 0.000250 Time 0.023307 -2022-12-06 11:31:52,450 - Epoch: [155][ 200/ 1200] Overall Loss 0.154459 Objective Loss 0.154459 LR 0.000250 Time 0.023110 -2022-12-06 11:31:52,642 - Epoch: [155][ 210/ 1200] Overall Loss 0.154064 Objective Loss 0.154064 LR 0.000250 Time 0.022922 -2022-12-06 11:31:52,836 - Epoch: [155][ 220/ 1200] Overall Loss 0.154312 Objective Loss 0.154312 LR 0.000250 Time 0.022759 -2022-12-06 11:31:53,029 - Epoch: [155][ 230/ 1200] Overall Loss 0.153937 Objective Loss 0.153937 LR 0.000250 Time 0.022610 -2022-12-06 11:31:53,224 - Epoch: [155][ 240/ 1200] Overall Loss 0.153611 Objective Loss 0.153611 LR 0.000250 Time 0.022475 -2022-12-06 11:31:53,417 - Epoch: [155][ 250/ 1200] Overall Loss 0.153583 Objective Loss 0.153583 LR 0.000250 Time 0.022346 -2022-12-06 11:31:53,611 - Epoch: [155][ 260/ 1200] Overall Loss 0.153707 Objective Loss 0.153707 LR 0.000250 Time 0.022231 -2022-12-06 11:31:53,804 - Epoch: [155][ 270/ 1200] Overall Loss 0.152986 Objective Loss 0.152986 LR 0.000250 Time 0.022121 -2022-12-06 11:31:54,000 - Epoch: [155][ 280/ 1200] Overall Loss 0.152977 Objective Loss 0.152977 LR 0.000250 Time 0.022028 -2022-12-06 11:31:54,193 - Epoch: [155][ 290/ 1200] Overall Loss 0.152879 Objective Loss 0.152879 LR 0.000250 Time 0.021934 -2022-12-06 11:31:54,387 - Epoch: [155][ 300/ 1200] Overall Loss 0.152765 Objective Loss 0.152765 LR 0.000250 Time 0.021848 -2022-12-06 11:31:54,581 - Epoch: [155][ 310/ 1200] Overall Loss 0.152815 Objective Loss 0.152815 LR 0.000250 Time 0.021766 -2022-12-06 11:31:54,775 - Epoch: [155][ 320/ 1200] Overall Loss 0.152767 Objective Loss 0.152767 LR 0.000250 Time 0.021690 -2022-12-06 11:31:54,968 - Epoch: [155][ 330/ 1200] Overall Loss 0.152415 Objective Loss 0.152415 LR 0.000250 Time 0.021618 -2022-12-06 11:31:55,162 - Epoch: [155][ 340/ 1200] Overall Loss 0.153018 Objective Loss 0.153018 LR 0.000250 Time 0.021550 -2022-12-06 11:31:55,356 - Epoch: [155][ 350/ 1200] Overall Loss 0.153019 Objective Loss 0.153019 LR 0.000250 Time 0.021485 -2022-12-06 11:31:55,549 - Epoch: [155][ 360/ 1200] Overall Loss 0.152869 Objective Loss 0.152869 LR 0.000250 Time 0.021425 -2022-12-06 11:31:55,743 - Epoch: [155][ 370/ 1200] Overall Loss 0.152643 Objective Loss 0.152643 LR 0.000250 Time 0.021368 -2022-12-06 11:31:55,937 - Epoch: [155][ 380/ 1200] Overall Loss 0.152501 Objective Loss 0.152501 LR 0.000250 Time 0.021314 -2022-12-06 11:31:56,130 - Epoch: [155][ 390/ 1200] Overall Loss 0.152304 Objective Loss 0.152304 LR 0.000250 Time 0.021261 -2022-12-06 11:31:56,325 - Epoch: [155][ 400/ 1200] Overall Loss 0.152124 Objective Loss 0.152124 LR 0.000250 Time 0.021217 -2022-12-06 11:31:56,519 - Epoch: [155][ 410/ 1200] Overall Loss 0.152117 Objective Loss 0.152117 LR 0.000250 Time 0.021170 -2022-12-06 11:31:56,713 - Epoch: [155][ 420/ 1200] Overall Loss 0.152599 Objective Loss 0.152599 LR 0.000250 Time 0.021127 -2022-12-06 11:31:56,906 - Epoch: [155][ 430/ 1200] Overall Loss 0.152437 Objective Loss 0.152437 LR 0.000250 Time 0.021084 -2022-12-06 11:31:57,100 - Epoch: [155][ 440/ 1200] Overall Loss 0.152491 Objective Loss 0.152491 LR 0.000250 Time 0.021044 -2022-12-06 11:31:57,294 - Epoch: [155][ 450/ 1200] Overall Loss 0.152747 Objective Loss 0.152747 LR 0.000250 Time 0.021006 -2022-12-06 11:31:57,488 - Epoch: [155][ 460/ 1200] Overall Loss 0.152531 Objective Loss 0.152531 LR 0.000250 Time 0.020971 -2022-12-06 11:31:57,682 - Epoch: [155][ 470/ 1200] Overall Loss 0.152222 Objective Loss 0.152222 LR 0.000250 Time 0.020936 -2022-12-06 11:31:57,877 - Epoch: [155][ 480/ 1200] Overall Loss 0.152310 Objective Loss 0.152310 LR 0.000250 Time 0.020904 -2022-12-06 11:31:58,070 - Epoch: [155][ 490/ 1200] Overall Loss 0.152129 Objective Loss 0.152129 LR 0.000250 Time 0.020870 -2022-12-06 11:31:58,264 - Epoch: [155][ 500/ 1200] Overall Loss 0.152132 Objective Loss 0.152132 LR 0.000250 Time 0.020841 -2022-12-06 11:31:58,458 - Epoch: [155][ 510/ 1200] Overall Loss 0.151924 Objective Loss 0.151924 LR 0.000250 Time 0.020810 -2022-12-06 11:31:58,652 - Epoch: [155][ 520/ 1200] Overall Loss 0.151790 Objective Loss 0.151790 LR 0.000250 Time 0.020784 -2022-12-06 11:31:58,846 - Epoch: [155][ 530/ 1200] Overall Loss 0.151553 Objective Loss 0.151553 LR 0.000250 Time 0.020756 -2022-12-06 11:31:59,041 - Epoch: [155][ 540/ 1200] Overall Loss 0.151365 Objective Loss 0.151365 LR 0.000250 Time 0.020731 -2022-12-06 11:31:59,234 - Epoch: [155][ 550/ 1200] Overall Loss 0.151330 Objective Loss 0.151330 LR 0.000250 Time 0.020704 -2022-12-06 11:31:59,428 - Epoch: [155][ 560/ 1200] Overall Loss 0.151003 Objective Loss 0.151003 LR 0.000250 Time 0.020680 -2022-12-06 11:31:59,621 - Epoch: [155][ 570/ 1200] Overall Loss 0.150882 Objective Loss 0.150882 LR 0.000250 Time 0.020655 -2022-12-06 11:31:59,816 - Epoch: [155][ 580/ 1200] Overall Loss 0.150798 Objective Loss 0.150798 LR 0.000250 Time 0.020633 -2022-12-06 11:32:00,009 - Epoch: [155][ 590/ 1200] Overall Loss 0.150955 Objective Loss 0.150955 LR 0.000250 Time 0.020610 -2022-12-06 11:32:00,202 - Epoch: [155][ 600/ 1200] Overall Loss 0.151006 Objective Loss 0.151006 LR 0.000250 Time 0.020588 -2022-12-06 11:32:00,396 - Epoch: [155][ 610/ 1200] Overall Loss 0.150897 Objective Loss 0.150897 LR 0.000250 Time 0.020567 -2022-12-06 11:32:00,591 - Epoch: [155][ 620/ 1200] Overall Loss 0.150541 Objective Loss 0.150541 LR 0.000250 Time 0.020549 -2022-12-06 11:32:00,784 - Epoch: [155][ 630/ 1200] Overall Loss 0.150560 Objective Loss 0.150560 LR 0.000250 Time 0.020529 -2022-12-06 11:32:00,978 - Epoch: [155][ 640/ 1200] Overall Loss 0.150786 Objective Loss 0.150786 LR 0.000250 Time 0.020511 -2022-12-06 11:32:01,172 - Epoch: [155][ 650/ 1200] Overall Loss 0.150714 Objective Loss 0.150714 LR 0.000250 Time 0.020493 -2022-12-06 11:32:01,367 - Epoch: [155][ 660/ 1200] Overall Loss 0.150951 Objective Loss 0.150951 LR 0.000250 Time 0.020476 -2022-12-06 11:32:01,560 - Epoch: [155][ 670/ 1200] Overall Loss 0.150844 Objective Loss 0.150844 LR 0.000250 Time 0.020458 -2022-12-06 11:32:01,754 - Epoch: [155][ 680/ 1200] Overall Loss 0.150953 Objective Loss 0.150953 LR 0.000250 Time 0.020441 -2022-12-06 11:32:01,947 - Epoch: [155][ 690/ 1200] Overall Loss 0.151182 Objective Loss 0.151182 LR 0.000250 Time 0.020424 -2022-12-06 11:32:02,141 - Epoch: [155][ 700/ 1200] Overall Loss 0.151010 Objective Loss 0.151010 LR 0.000250 Time 0.020409 -2022-12-06 11:32:02,334 - Epoch: [155][ 710/ 1200] Overall Loss 0.151180 Objective Loss 0.151180 LR 0.000250 Time 0.020393 -2022-12-06 11:32:02,529 - Epoch: [155][ 720/ 1200] Overall Loss 0.151265 Objective Loss 0.151265 LR 0.000250 Time 0.020380 -2022-12-06 11:32:02,723 - Epoch: [155][ 730/ 1200] Overall Loss 0.151456 Objective Loss 0.151456 LR 0.000250 Time 0.020365 -2022-12-06 11:32:02,918 - Epoch: [155][ 740/ 1200] Overall Loss 0.151788 Objective Loss 0.151788 LR 0.000250 Time 0.020353 -2022-12-06 11:32:03,111 - Epoch: [155][ 750/ 1200] Overall Loss 0.151499 Objective Loss 0.151499 LR 0.000250 Time 0.020339 -2022-12-06 11:32:03,306 - Epoch: [155][ 760/ 1200] Overall Loss 0.151455 Objective Loss 0.151455 LR 0.000250 Time 0.020326 -2022-12-06 11:32:03,499 - Epoch: [155][ 770/ 1200] Overall Loss 0.151377 Objective Loss 0.151377 LR 0.000250 Time 0.020313 -2022-12-06 11:32:03,693 - Epoch: [155][ 780/ 1200] Overall Loss 0.151303 Objective Loss 0.151303 LR 0.000250 Time 0.020300 -2022-12-06 11:32:03,887 - Epoch: [155][ 790/ 1200] Overall Loss 0.151327 Objective Loss 0.151327 LR 0.000250 Time 0.020287 -2022-12-06 11:32:04,080 - Epoch: [155][ 800/ 1200] Overall Loss 0.151268 Objective Loss 0.151268 LR 0.000250 Time 0.020275 -2022-12-06 11:32:04,275 - Epoch: [155][ 810/ 1200] Overall Loss 0.151282 Objective Loss 0.151282 LR 0.000250 Time 0.020264 -2022-12-06 11:32:04,469 - Epoch: [155][ 820/ 1200] Overall Loss 0.151224 Objective Loss 0.151224 LR 0.000250 Time 0.020253 -2022-12-06 11:32:04,662 - Epoch: [155][ 830/ 1200] Overall Loss 0.151392 Objective Loss 0.151392 LR 0.000250 Time 0.020240 -2022-12-06 11:32:04,855 - Epoch: [155][ 840/ 1200] Overall Loss 0.151198 Objective Loss 0.151198 LR 0.000250 Time 0.020230 -2022-12-06 11:32:05,049 - Epoch: [155][ 850/ 1200] Overall Loss 0.151182 Objective Loss 0.151182 LR 0.000250 Time 0.020218 -2022-12-06 11:32:05,242 - Epoch: [155][ 860/ 1200] Overall Loss 0.151431 Objective Loss 0.151431 LR 0.000250 Time 0.020208 -2022-12-06 11:32:05,436 - Epoch: [155][ 870/ 1200] Overall Loss 0.151328 Objective Loss 0.151328 LR 0.000250 Time 0.020197 -2022-12-06 11:32:05,630 - Epoch: [155][ 880/ 1200] Overall Loss 0.151392 Objective Loss 0.151392 LR 0.000250 Time 0.020188 -2022-12-06 11:32:05,823 - Epoch: [155][ 890/ 1200] Overall Loss 0.151155 Objective Loss 0.151155 LR 0.000250 Time 0.020178 -2022-12-06 11:32:06,018 - Epoch: [155][ 900/ 1200] Overall Loss 0.151141 Objective Loss 0.151141 LR 0.000250 Time 0.020169 -2022-12-06 11:32:06,211 - Epoch: [155][ 910/ 1200] Overall Loss 0.150812 Objective Loss 0.150812 LR 0.000250 Time 0.020159 -2022-12-06 11:32:06,405 - Epoch: [155][ 920/ 1200] Overall Loss 0.150853 Objective Loss 0.150853 LR 0.000250 Time 0.020150 -2022-12-06 11:32:06,598 - Epoch: [155][ 930/ 1200] Overall Loss 0.150788 Objective Loss 0.150788 LR 0.000250 Time 0.020140 -2022-12-06 11:32:06,792 - Epoch: [155][ 940/ 1200] Overall Loss 0.150716 Objective Loss 0.150716 LR 0.000250 Time 0.020132 -2022-12-06 11:32:06,986 - Epoch: [155][ 950/ 1200] Overall Loss 0.150714 Objective Loss 0.150714 LR 0.000250 Time 0.020124 -2022-12-06 11:32:07,181 - Epoch: [155][ 960/ 1200] Overall Loss 0.150665 Objective Loss 0.150665 LR 0.000250 Time 0.020117 -2022-12-06 11:32:07,374 - Epoch: [155][ 970/ 1200] Overall Loss 0.150790 Objective Loss 0.150790 LR 0.000250 Time 0.020107 -2022-12-06 11:32:07,568 - Epoch: [155][ 980/ 1200] Overall Loss 0.150875 Objective Loss 0.150875 LR 0.000250 Time 0.020100 -2022-12-06 11:32:07,761 - Epoch: [155][ 990/ 1200] Overall Loss 0.151013 Objective Loss 0.151013 LR 0.000250 Time 0.020092 -2022-12-06 11:32:07,956 - Epoch: [155][ 1000/ 1200] Overall Loss 0.150859 Objective Loss 0.150859 LR 0.000250 Time 0.020084 -2022-12-06 11:32:08,149 - Epoch: [155][ 1010/ 1200] Overall Loss 0.150974 Objective Loss 0.150974 LR 0.000250 Time 0.020077 -2022-12-06 11:32:08,343 - Epoch: [155][ 1020/ 1200] Overall Loss 0.151067 Objective Loss 0.151067 LR 0.000250 Time 0.020069 -2022-12-06 11:32:08,538 - Epoch: [155][ 1030/ 1200] Overall Loss 0.151348 Objective Loss 0.151348 LR 0.000250 Time 0.020063 -2022-12-06 11:32:08,732 - Epoch: [155][ 1040/ 1200] Overall Loss 0.151310 Objective Loss 0.151310 LR 0.000250 Time 0.020056 -2022-12-06 11:32:08,924 - Epoch: [155][ 1050/ 1200] Overall Loss 0.151377 Objective Loss 0.151377 LR 0.000250 Time 0.020048 -2022-12-06 11:32:09,120 - Epoch: [155][ 1060/ 1200] Overall Loss 0.151243 Objective Loss 0.151243 LR 0.000250 Time 0.020042 -2022-12-06 11:32:09,313 - Epoch: [155][ 1070/ 1200] Overall Loss 0.151379 Objective Loss 0.151379 LR 0.000250 Time 0.020035 -2022-12-06 11:32:09,507 - Epoch: [155][ 1080/ 1200] Overall Loss 0.151350 Objective Loss 0.151350 LR 0.000250 Time 0.020029 -2022-12-06 11:32:09,700 - Epoch: [155][ 1090/ 1200] Overall Loss 0.151240 Objective Loss 0.151240 LR 0.000250 Time 0.020022 -2022-12-06 11:32:09,895 - Epoch: [155][ 1100/ 1200] Overall Loss 0.151215 Objective Loss 0.151215 LR 0.000250 Time 0.020016 -2022-12-06 11:32:10,088 - Epoch: [155][ 1110/ 1200] Overall Loss 0.151017 Objective Loss 0.151017 LR 0.000250 Time 0.020010 -2022-12-06 11:32:10,283 - Epoch: [155][ 1120/ 1200] Overall Loss 0.150836 Objective Loss 0.150836 LR 0.000250 Time 0.020004 -2022-12-06 11:32:10,477 - Epoch: [155][ 1130/ 1200] Overall Loss 0.150758 Objective Loss 0.150758 LR 0.000250 Time 0.019998 -2022-12-06 11:32:10,671 - Epoch: [155][ 1140/ 1200] Overall Loss 0.150873 Objective Loss 0.150873 LR 0.000250 Time 0.019993 -2022-12-06 11:32:10,864 - Epoch: [155][ 1150/ 1200] Overall Loss 0.150835 Objective Loss 0.150835 LR 0.000250 Time 0.019987 -2022-12-06 11:32:11,058 - Epoch: [155][ 1160/ 1200] Overall Loss 0.150663 Objective Loss 0.150663 LR 0.000250 Time 0.019981 -2022-12-06 11:32:11,252 - Epoch: [155][ 1170/ 1200] Overall Loss 0.150593 Objective Loss 0.150593 LR 0.000250 Time 0.019976 -2022-12-06 11:32:11,446 - Epoch: [155][ 1180/ 1200] Overall Loss 0.150564 Objective Loss 0.150564 LR 0.000250 Time 0.019970 -2022-12-06 11:32:11,640 - Epoch: [155][ 1190/ 1200] Overall Loss 0.150535 Objective Loss 0.150535 LR 0.000250 Time 0.019965 -2022-12-06 11:32:11,864 - Epoch: [155][ 1200/ 1200] Overall Loss 0.150686 Objective Loss 0.150686 Top1 86.192469 Top5 98.744770 LR 0.000250 Time 0.019985 -2022-12-06 11:32:11,953 - --- validate (epoch=155)----------- -2022-12-06 11:32:11,953 - 34129 samples (256 per mini-batch) -2022-12-06 11:32:12,412 - Epoch: [155][ 10/ 134] Loss 0.271020 Top1 86.953125 Top5 98.515625 -2022-12-06 11:32:12,555 - Epoch: [155][ 20/ 134] Loss 0.251200 Top1 87.246094 Top5 98.554688 -2022-12-06 11:32:12,689 - Epoch: [155][ 30/ 134] Loss 0.243291 Top1 87.565104 Top5 98.476562 -2022-12-06 11:32:12,817 - Epoch: [155][ 40/ 134] Loss 0.235059 Top1 87.783203 Top5 98.603516 -2022-12-06 11:32:12,943 - Epoch: [155][ 50/ 134] Loss 0.234741 Top1 87.804688 Top5 98.546875 -2022-12-06 11:32:13,072 - Epoch: [155][ 60/ 134] Loss 0.230964 Top1 87.877604 Top5 98.554688 -2022-12-06 11:32:13,199 - Epoch: [155][ 70/ 134] Loss 0.231450 Top1 87.918527 Top5 98.577009 -2022-12-06 11:32:13,326 - Epoch: [155][ 80/ 134] Loss 0.232032 Top1 87.851562 Top5 98.588867 -2022-12-06 11:32:13,449 - Epoch: [155][ 90/ 134] Loss 0.231780 Top1 87.808160 Top5 98.559028 -2022-12-06 11:32:13,577 - Epoch: [155][ 100/ 134] Loss 0.232381 Top1 87.847656 Top5 98.558594 -2022-12-06 11:32:13,702 - Epoch: [155][ 110/ 134] Loss 0.231873 Top1 87.879972 Top5 98.565341 -2022-12-06 11:32:13,827 - Epoch: [155][ 120/ 134] Loss 0.235074 Top1 87.796224 Top5 98.528646 -2022-12-06 11:32:13,952 - Epoch: [155][ 130/ 134] Loss 0.231720 Top1 87.827524 Top5 98.563702 -2022-12-06 11:32:13,989 - Epoch: [155][ 134/ 134] Loss 0.230938 Top1 87.822673 Top5 98.555481 -2022-12-06 11:32:14,076 - ==> Top1: 87.823 Top5: 98.555 Loss: 0.231 - -2022-12-06 11:32:14,077 - ==> Confusion: -[[ 925 1 2 4 3 5 0 0 2 38 0 1 1 4 2 1 0 0 3 0 4] - [ 1 939 3 2 9 17 1 9 4 3 2 5 1 0 0 1 5 0 13 3 9] - [ 3 3 1021 8 4 2 13 7 0 3 8 6 0 1 1 1 1 3 3 3 12] - [ 0 0 16 953 1 2 1 0 1 3 9 0 4 2 11 0 1 1 9 0 6] - [ 8 5 1 0 958 1 1 2 0 6 1 3 1 3 12 6 4 3 1 0 4] - [ 1 15 0 4 5 974 1 21 2 2 1 12 2 13 1 3 2 1 0 5 4] - [ 1 1 12 1 1 0 1078 4 0 0 2 1 0 3 0 4 0 1 2 6 1] - [ 1 6 6 2 3 28 6 952 0 2 2 7 1 1 0 1 0 1 22 6 7] - [ 4 1 0 0 0 3 1 1 992 29 12 1 1 8 6 0 0 1 1 1 2] - [ 57 0 1 0 4 1 0 2 25 885 1 2 0 11 3 2 0 0 1 0 6] - [ 0 1 2 1 1 3 2 1 7 0 976 2 0 9 1 1 0 0 4 1 7] - [ 2 0 3 0 0 8 8 2 0 0 0 981 15 6 0 8 3 5 0 8 2] - [ 1 1 1 2 0 1 0 2 0 0 0 28 903 3 1 9 0 9 0 3 5] - [ 1 1 1 0 1 7 0 2 12 10 2 5 3 956 2 2 2 0 0 2 14] - [ 8 2 3 9 4 2 0 0 17 2 1 3 2 4 1062 0 0 3 6 0 2] - [ 0 0 2 1 0 1 2 0 0 0 1 6 6 1 0 1000 5 11 0 4 3] - [ 2 0 1 1 4 1 1 1 1 0 0 3 4 2 0 9 1031 0 0 3 8] - [ 2 0 1 2 0 1 0 0 1 4 0 5 10 1 1 14 0 990 1 0 3] - [ 1 4 2 3 3 0 1 21 3 1 5 3 2 0 6 0 0 0 946 2 5] - [ 1 2 1 2 1 5 4 4 1 0 3 16 5 5 0 5 2 4 0 1015 4] - [ 120 189 164 96 110 135 81 132 73 89 170 103 285 272 134 108 126 79 154 173 10433]] - -2022-12-06 11:32:14,645 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:32:14,645 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:32:14,651 - - -2022-12-06 11:32:14,651 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:32:15,698 - Epoch: [156][ 10/ 1200] Overall Loss 0.138760 Objective Loss 0.138760 LR 0.000250 Time 0.104628 -2022-12-06 11:32:15,891 - Epoch: [156][ 20/ 1200] Overall Loss 0.142248 Objective Loss 0.142248 LR 0.000250 Time 0.061942 -2022-12-06 11:32:16,083 - Epoch: [156][ 30/ 1200] Overall Loss 0.152434 Objective Loss 0.152434 LR 0.000250 Time 0.047661 -2022-12-06 11:32:16,275 - Epoch: [156][ 40/ 1200] Overall Loss 0.148314 Objective Loss 0.148314 LR 0.000250 Time 0.040538 -2022-12-06 11:32:16,467 - Epoch: [156][ 50/ 1200] Overall Loss 0.146246 Objective Loss 0.146246 LR 0.000250 Time 0.036250 -2022-12-06 11:32:16,658 - Epoch: [156][ 60/ 1200] Overall Loss 0.146270 Objective Loss 0.146270 LR 0.000250 Time 0.033389 -2022-12-06 11:32:16,850 - Epoch: [156][ 70/ 1200] Overall Loss 0.148048 Objective Loss 0.148048 LR 0.000250 Time 0.031351 -2022-12-06 11:32:17,042 - Epoch: [156][ 80/ 1200] Overall Loss 0.146524 Objective Loss 0.146524 LR 0.000250 Time 0.029826 -2022-12-06 11:32:17,233 - Epoch: [156][ 90/ 1200] Overall Loss 0.145725 Objective Loss 0.145725 LR 0.000250 Time 0.028624 -2022-12-06 11:32:17,423 - Epoch: [156][ 100/ 1200] Overall Loss 0.145157 Objective Loss 0.145157 LR 0.000250 Time 0.027661 -2022-12-06 11:32:17,614 - Epoch: [156][ 110/ 1200] Overall Loss 0.144736 Objective Loss 0.144736 LR 0.000250 Time 0.026879 -2022-12-06 11:32:17,806 - Epoch: [156][ 120/ 1200] Overall Loss 0.145282 Objective Loss 0.145282 LR 0.000250 Time 0.026233 -2022-12-06 11:32:17,997 - Epoch: [156][ 130/ 1200] Overall Loss 0.145037 Objective Loss 0.145037 LR 0.000250 Time 0.025679 -2022-12-06 11:32:18,188 - Epoch: [156][ 140/ 1200] Overall Loss 0.144898 Objective Loss 0.144898 LR 0.000250 Time 0.025207 -2022-12-06 11:32:18,380 - Epoch: [156][ 150/ 1200] Overall Loss 0.144582 Objective Loss 0.144582 LR 0.000250 Time 0.024798 -2022-12-06 11:32:18,571 - Epoch: [156][ 160/ 1200] Overall Loss 0.143783 Objective Loss 0.143783 LR 0.000250 Time 0.024441 -2022-12-06 11:32:18,762 - Epoch: [156][ 170/ 1200] Overall Loss 0.143932 Objective Loss 0.143932 LR 0.000250 Time 0.024125 -2022-12-06 11:32:18,953 - Epoch: [156][ 180/ 1200] Overall Loss 0.143261 Objective Loss 0.143261 LR 0.000250 Time 0.023844 -2022-12-06 11:32:19,145 - Epoch: [156][ 190/ 1200] Overall Loss 0.142731 Objective Loss 0.142731 LR 0.000250 Time 0.023592 -2022-12-06 11:32:19,336 - Epoch: [156][ 200/ 1200] Overall Loss 0.142777 Objective Loss 0.142777 LR 0.000250 Time 0.023366 -2022-12-06 11:32:19,527 - Epoch: [156][ 210/ 1200] Overall Loss 0.142367 Objective Loss 0.142367 LR 0.000250 Time 0.023161 -2022-12-06 11:32:19,719 - Epoch: [156][ 220/ 1200] Overall Loss 0.142662 Objective Loss 0.142662 LR 0.000250 Time 0.022978 -2022-12-06 11:32:19,910 - Epoch: [156][ 230/ 1200] Overall Loss 0.142964 Objective Loss 0.142964 LR 0.000250 Time 0.022807 -2022-12-06 11:32:20,101 - Epoch: [156][ 240/ 1200] Overall Loss 0.143642 Objective Loss 0.143642 LR 0.000250 Time 0.022651 -2022-12-06 11:32:20,293 - Epoch: [156][ 250/ 1200] Overall Loss 0.143606 Objective Loss 0.143606 LR 0.000250 Time 0.022511 -2022-12-06 11:32:20,484 - Epoch: [156][ 260/ 1200] Overall Loss 0.143928 Objective Loss 0.143928 LR 0.000250 Time 0.022378 -2022-12-06 11:32:20,677 - Epoch: [156][ 270/ 1200] Overall Loss 0.144248 Objective Loss 0.144248 LR 0.000250 Time 0.022260 -2022-12-06 11:32:20,868 - Epoch: [156][ 280/ 1200] Overall Loss 0.144095 Objective Loss 0.144095 LR 0.000250 Time 0.022145 -2022-12-06 11:32:21,059 - Epoch: [156][ 290/ 1200] Overall Loss 0.144459 Objective Loss 0.144459 LR 0.000250 Time 0.022039 -2022-12-06 11:32:21,250 - Epoch: [156][ 300/ 1200] Overall Loss 0.144522 Objective Loss 0.144522 LR 0.000250 Time 0.021941 -2022-12-06 11:32:21,441 - Epoch: [156][ 310/ 1200] Overall Loss 0.145044 Objective Loss 0.145044 LR 0.000250 Time 0.021846 -2022-12-06 11:32:21,632 - Epoch: [156][ 320/ 1200] Overall Loss 0.145090 Objective Loss 0.145090 LR 0.000250 Time 0.021759 -2022-12-06 11:32:21,823 - Epoch: [156][ 330/ 1200] Overall Loss 0.145784 Objective Loss 0.145784 LR 0.000250 Time 0.021677 -2022-12-06 11:32:22,016 - Epoch: [156][ 340/ 1200] Overall Loss 0.145967 Objective Loss 0.145967 LR 0.000250 Time 0.021603 -2022-12-06 11:32:22,206 - Epoch: [156][ 350/ 1200] Overall Loss 0.145529 Objective Loss 0.145529 LR 0.000250 Time 0.021529 -2022-12-06 11:32:22,396 - Epoch: [156][ 360/ 1200] Overall Loss 0.145391 Objective Loss 0.145391 LR 0.000250 Time 0.021458 -2022-12-06 11:32:22,587 - Epoch: [156][ 370/ 1200] Overall Loss 0.145337 Objective Loss 0.145337 LR 0.000250 Time 0.021391 -2022-12-06 11:32:22,778 - Epoch: [156][ 380/ 1200] Overall Loss 0.145392 Objective Loss 0.145392 LR 0.000250 Time 0.021331 -2022-12-06 11:32:22,970 - Epoch: [156][ 390/ 1200] Overall Loss 0.144832 Objective Loss 0.144832 LR 0.000250 Time 0.021273 -2022-12-06 11:32:23,161 - Epoch: [156][ 400/ 1200] Overall Loss 0.145002 Objective Loss 0.145002 LR 0.000250 Time 0.021218 -2022-12-06 11:32:23,354 - Epoch: [156][ 410/ 1200] Overall Loss 0.145802 Objective Loss 0.145802 LR 0.000250 Time 0.021170 -2022-12-06 11:32:23,547 - Epoch: [156][ 420/ 1200] Overall Loss 0.146453 Objective Loss 0.146453 LR 0.000250 Time 0.021125 -2022-12-06 11:32:23,740 - Epoch: [156][ 430/ 1200] Overall Loss 0.146517 Objective Loss 0.146517 LR 0.000250 Time 0.021082 -2022-12-06 11:32:23,934 - Epoch: [156][ 440/ 1200] Overall Loss 0.146918 Objective Loss 0.146918 LR 0.000250 Time 0.021040 -2022-12-06 11:32:24,126 - Epoch: [156][ 450/ 1200] Overall Loss 0.146602 Objective Loss 0.146602 LR 0.000250 Time 0.021000 -2022-12-06 11:32:24,321 - Epoch: [156][ 460/ 1200] Overall Loss 0.146689 Objective Loss 0.146689 LR 0.000250 Time 0.020965 -2022-12-06 11:32:24,515 - Epoch: [156][ 470/ 1200] Overall Loss 0.146874 Objective Loss 0.146874 LR 0.000250 Time 0.020930 -2022-12-06 11:32:24,708 - Epoch: [156][ 480/ 1200] Overall Loss 0.146586 Objective Loss 0.146586 LR 0.000250 Time 0.020896 -2022-12-06 11:32:24,901 - Epoch: [156][ 490/ 1200] Overall Loss 0.146423 Objective Loss 0.146423 LR 0.000250 Time 0.020862 -2022-12-06 11:32:25,095 - Epoch: [156][ 500/ 1200] Overall Loss 0.146756 Objective Loss 0.146756 LR 0.000250 Time 0.020831 -2022-12-06 11:32:25,287 - Epoch: [156][ 510/ 1200] Overall Loss 0.146474 Objective Loss 0.146474 LR 0.000250 Time 0.020799 -2022-12-06 11:32:25,481 - Epoch: [156][ 520/ 1200] Overall Loss 0.146352 Objective Loss 0.146352 LR 0.000250 Time 0.020771 -2022-12-06 11:32:25,674 - Epoch: [156][ 530/ 1200] Overall Loss 0.146901 Objective Loss 0.146901 LR 0.000250 Time 0.020743 -2022-12-06 11:32:25,867 - Epoch: [156][ 540/ 1200] Overall Loss 0.147225 Objective Loss 0.147225 LR 0.000250 Time 0.020715 -2022-12-06 11:32:26,060 - Epoch: [156][ 550/ 1200] Overall Loss 0.147206 Objective Loss 0.147206 LR 0.000250 Time 0.020688 -2022-12-06 11:32:26,254 - Epoch: [156][ 560/ 1200] Overall Loss 0.147297 Objective Loss 0.147297 LR 0.000250 Time 0.020663 -2022-12-06 11:32:26,447 - Epoch: [156][ 570/ 1200] Overall Loss 0.147306 Objective Loss 0.147306 LR 0.000250 Time 0.020640 -2022-12-06 11:32:26,640 - Epoch: [156][ 580/ 1200] Overall Loss 0.147446 Objective Loss 0.147446 LR 0.000250 Time 0.020615 -2022-12-06 11:32:26,833 - Epoch: [156][ 590/ 1200] Overall Loss 0.147522 Objective Loss 0.147522 LR 0.000250 Time 0.020591 -2022-12-06 11:32:27,026 - Epoch: [156][ 600/ 1200] Overall Loss 0.147620 Objective Loss 0.147620 LR 0.000250 Time 0.020569 -2022-12-06 11:32:27,219 - Epoch: [156][ 610/ 1200] Overall Loss 0.147325 Objective Loss 0.147325 LR 0.000250 Time 0.020547 -2022-12-06 11:32:27,413 - Epoch: [156][ 620/ 1200] Overall Loss 0.147349 Objective Loss 0.147349 LR 0.000250 Time 0.020528 -2022-12-06 11:32:27,606 - Epoch: [156][ 630/ 1200] Overall Loss 0.147622 Objective Loss 0.147622 LR 0.000250 Time 0.020508 -2022-12-06 11:32:27,799 - Epoch: [156][ 640/ 1200] Overall Loss 0.147680 Objective Loss 0.147680 LR 0.000250 Time 0.020488 -2022-12-06 11:32:27,992 - Epoch: [156][ 650/ 1200] Overall Loss 0.147848 Objective Loss 0.147848 LR 0.000250 Time 0.020469 -2022-12-06 11:32:28,186 - Epoch: [156][ 660/ 1200] Overall Loss 0.147826 Objective Loss 0.147826 LR 0.000250 Time 0.020452 -2022-12-06 11:32:28,379 - Epoch: [156][ 670/ 1200] Overall Loss 0.147559 Objective Loss 0.147559 LR 0.000250 Time 0.020433 -2022-12-06 11:32:28,572 - Epoch: [156][ 680/ 1200] Overall Loss 0.147475 Objective Loss 0.147475 LR 0.000250 Time 0.020417 -2022-12-06 11:32:28,765 - Epoch: [156][ 690/ 1200] Overall Loss 0.147600 Objective Loss 0.147600 LR 0.000250 Time 0.020399 -2022-12-06 11:32:28,959 - Epoch: [156][ 700/ 1200] Overall Loss 0.147824 Objective Loss 0.147824 LR 0.000250 Time 0.020384 -2022-12-06 11:32:29,151 - Epoch: [156][ 710/ 1200] Overall Loss 0.147572 Objective Loss 0.147572 LR 0.000250 Time 0.020366 -2022-12-06 11:32:29,344 - Epoch: [156][ 720/ 1200] Overall Loss 0.147801 Objective Loss 0.147801 LR 0.000250 Time 0.020351 -2022-12-06 11:32:29,537 - Epoch: [156][ 730/ 1200] Overall Loss 0.147381 Objective Loss 0.147381 LR 0.000250 Time 0.020336 -2022-12-06 11:32:29,731 - Epoch: [156][ 740/ 1200] Overall Loss 0.147439 Objective Loss 0.147439 LR 0.000250 Time 0.020323 -2022-12-06 11:32:29,923 - Epoch: [156][ 750/ 1200] Overall Loss 0.147397 Objective Loss 0.147397 LR 0.000250 Time 0.020307 -2022-12-06 11:32:30,117 - Epoch: [156][ 760/ 1200] Overall Loss 0.147406 Objective Loss 0.147406 LR 0.000250 Time 0.020295 -2022-12-06 11:32:30,310 - Epoch: [156][ 770/ 1200] Overall Loss 0.147497 Objective Loss 0.147497 LR 0.000250 Time 0.020280 -2022-12-06 11:32:30,504 - Epoch: [156][ 780/ 1200] Overall Loss 0.147646 Objective Loss 0.147646 LR 0.000250 Time 0.020268 -2022-12-06 11:32:30,697 - Epoch: [156][ 790/ 1200] Overall Loss 0.147683 Objective Loss 0.147683 LR 0.000250 Time 0.020255 -2022-12-06 11:32:30,890 - Epoch: [156][ 800/ 1200] Overall Loss 0.147476 Objective Loss 0.147476 LR 0.000250 Time 0.020242 -2022-12-06 11:32:31,083 - Epoch: [156][ 810/ 1200] Overall Loss 0.147609 Objective Loss 0.147609 LR 0.000250 Time 0.020231 -2022-12-06 11:32:31,277 - Epoch: [156][ 820/ 1200] Overall Loss 0.147609 Objective Loss 0.147609 LR 0.000250 Time 0.020219 -2022-12-06 11:32:31,469 - Epoch: [156][ 830/ 1200] Overall Loss 0.147536 Objective Loss 0.147536 LR 0.000250 Time 0.020207 -2022-12-06 11:32:31,663 - Epoch: [156][ 840/ 1200] Overall Loss 0.147540 Objective Loss 0.147540 LR 0.000250 Time 0.020196 -2022-12-06 11:32:31,856 - Epoch: [156][ 850/ 1200] Overall Loss 0.147730 Objective Loss 0.147730 LR 0.000250 Time 0.020185 -2022-12-06 11:32:32,048 - Epoch: [156][ 860/ 1200] Overall Loss 0.147603 Objective Loss 0.147603 LR 0.000250 Time 0.020173 -2022-12-06 11:32:32,240 - Epoch: [156][ 870/ 1200] Overall Loss 0.147409 Objective Loss 0.147409 LR 0.000250 Time 0.020161 -2022-12-06 11:32:32,434 - Epoch: [156][ 880/ 1200] Overall Loss 0.147414 Objective Loss 0.147414 LR 0.000250 Time 0.020152 -2022-12-06 11:32:32,627 - Epoch: [156][ 890/ 1200] Overall Loss 0.147645 Objective Loss 0.147645 LR 0.000250 Time 0.020142 -2022-12-06 11:32:32,820 - Epoch: [156][ 900/ 1200] Overall Loss 0.147558 Objective Loss 0.147558 LR 0.000250 Time 0.020132 -2022-12-06 11:32:33,013 - Epoch: [156][ 910/ 1200] Overall Loss 0.147482 Objective Loss 0.147482 LR 0.000250 Time 0.020122 -2022-12-06 11:32:33,207 - Epoch: [156][ 920/ 1200] Overall Loss 0.147652 Objective Loss 0.147652 LR 0.000250 Time 0.020113 -2022-12-06 11:32:33,399 - Epoch: [156][ 930/ 1200] Overall Loss 0.147651 Objective Loss 0.147651 LR 0.000250 Time 0.020104 -2022-12-06 11:32:33,593 - Epoch: [156][ 940/ 1200] Overall Loss 0.147432 Objective Loss 0.147432 LR 0.000250 Time 0.020095 -2022-12-06 11:32:33,786 - Epoch: [156][ 950/ 1200] Overall Loss 0.147516 Objective Loss 0.147516 LR 0.000250 Time 0.020086 -2022-12-06 11:32:33,979 - Epoch: [156][ 960/ 1200] Overall Loss 0.147434 Objective Loss 0.147434 LR 0.000250 Time 0.020077 -2022-12-06 11:32:34,173 - Epoch: [156][ 970/ 1200] Overall Loss 0.147402 Objective Loss 0.147402 LR 0.000250 Time 0.020070 -2022-12-06 11:32:34,367 - Epoch: [156][ 980/ 1200] Overall Loss 0.147488 Objective Loss 0.147488 LR 0.000250 Time 0.020062 -2022-12-06 11:32:34,560 - Epoch: [156][ 990/ 1200] Overall Loss 0.147475 Objective Loss 0.147475 LR 0.000250 Time 0.020055 -2022-12-06 11:32:34,754 - Epoch: [156][ 1000/ 1200] Overall Loss 0.147431 Objective Loss 0.147431 LR 0.000250 Time 0.020047 -2022-12-06 11:32:34,947 - Epoch: [156][ 1010/ 1200] Overall Loss 0.147589 Objective Loss 0.147589 LR 0.000250 Time 0.020040 -2022-12-06 11:32:35,140 - Epoch: [156][ 1020/ 1200] Overall Loss 0.147637 Objective Loss 0.147637 LR 0.000250 Time 0.020032 -2022-12-06 11:32:35,333 - Epoch: [156][ 1030/ 1200] Overall Loss 0.147828 Objective Loss 0.147828 LR 0.000250 Time 0.020024 -2022-12-06 11:32:35,526 - Epoch: [156][ 1040/ 1200] Overall Loss 0.147838 Objective Loss 0.147838 LR 0.000250 Time 0.020017 -2022-12-06 11:32:35,719 - Epoch: [156][ 1050/ 1200] Overall Loss 0.147674 Objective Loss 0.147674 LR 0.000250 Time 0.020009 -2022-12-06 11:32:35,913 - Epoch: [156][ 1060/ 1200] Overall Loss 0.147557 Objective Loss 0.147557 LR 0.000250 Time 0.020003 -2022-12-06 11:32:36,106 - Epoch: [156][ 1070/ 1200] Overall Loss 0.147649 Objective Loss 0.147649 LR 0.000250 Time 0.019996 -2022-12-06 11:32:36,300 - Epoch: [156][ 1080/ 1200] Overall Loss 0.147740 Objective Loss 0.147740 LR 0.000250 Time 0.019990 -2022-12-06 11:32:36,493 - Epoch: [156][ 1090/ 1200] Overall Loss 0.147757 Objective Loss 0.147757 LR 0.000250 Time 0.019983 -2022-12-06 11:32:36,687 - Epoch: [156][ 1100/ 1200] Overall Loss 0.147803 Objective Loss 0.147803 LR 0.000250 Time 0.019977 -2022-12-06 11:32:36,880 - Epoch: [156][ 1110/ 1200] Overall Loss 0.147684 Objective Loss 0.147684 LR 0.000250 Time 0.019971 -2022-12-06 11:32:37,074 - Epoch: [156][ 1120/ 1200] Overall Loss 0.147798 Objective Loss 0.147798 LR 0.000250 Time 0.019965 -2022-12-06 11:32:37,267 - Epoch: [156][ 1130/ 1200] Overall Loss 0.147773 Objective Loss 0.147773 LR 0.000250 Time 0.019959 -2022-12-06 11:32:37,460 - Epoch: [156][ 1140/ 1200] Overall Loss 0.147784 Objective Loss 0.147784 LR 0.000250 Time 0.019953 -2022-12-06 11:32:37,653 - Epoch: [156][ 1150/ 1200] Overall Loss 0.147674 Objective Loss 0.147674 LR 0.000250 Time 0.019946 -2022-12-06 11:32:37,846 - Epoch: [156][ 1160/ 1200] Overall Loss 0.147759 Objective Loss 0.147759 LR 0.000250 Time 0.019941 -2022-12-06 11:32:38,039 - Epoch: [156][ 1170/ 1200] Overall Loss 0.147955 Objective Loss 0.147955 LR 0.000250 Time 0.019934 -2022-12-06 11:32:38,233 - Epoch: [156][ 1180/ 1200] Overall Loss 0.148060 Objective Loss 0.148060 LR 0.000250 Time 0.019929 -2022-12-06 11:32:38,425 - Epoch: [156][ 1190/ 1200] Overall Loss 0.148066 Objective Loss 0.148066 LR 0.000250 Time 0.019923 -2022-12-06 11:32:38,647 - Epoch: [156][ 1200/ 1200] Overall Loss 0.148111 Objective Loss 0.148111 Top1 89.330544 Top5 99.163180 LR 0.000250 Time 0.019942 -2022-12-06 11:32:38,736 - --- validate (epoch=156)----------- -2022-12-06 11:32:38,737 - 34129 samples (256 per mini-batch) -2022-12-06 11:32:39,191 - Epoch: [156][ 10/ 134] Loss 0.230221 Top1 87.460938 Top5 98.203125 -2022-12-06 11:32:39,339 - Epoch: [156][ 20/ 134] Loss 0.226301 Top1 87.851562 Top5 98.535156 -2022-12-06 11:32:39,480 - Epoch: [156][ 30/ 134] Loss 0.230150 Top1 87.851562 Top5 98.554688 -2022-12-06 11:32:39,618 - Epoch: [156][ 40/ 134] Loss 0.236490 Top1 87.812500 Top5 98.496094 -2022-12-06 11:32:39,746 - Epoch: [156][ 50/ 134] Loss 0.236712 Top1 87.921875 Top5 98.429688 -2022-12-06 11:32:39,871 - Epoch: [156][ 60/ 134] Loss 0.233525 Top1 87.988281 Top5 98.476562 -2022-12-06 11:32:39,997 - Epoch: [156][ 70/ 134] Loss 0.235437 Top1 87.957589 Top5 98.498884 -2022-12-06 11:32:40,125 - Epoch: [156][ 80/ 134] Loss 0.233424 Top1 87.993164 Top5 98.554688 -2022-12-06 11:32:40,254 - Epoch: [156][ 90/ 134] Loss 0.234348 Top1 87.994792 Top5 98.515625 -2022-12-06 11:32:40,382 - Epoch: [156][ 100/ 134] Loss 0.232833 Top1 88.042969 Top5 98.531250 -2022-12-06 11:32:40,512 - Epoch: [156][ 110/ 134] Loss 0.234369 Top1 88.039773 Top5 98.565341 -2022-12-06 11:32:40,641 - Epoch: [156][ 120/ 134] Loss 0.234322 Top1 87.945964 Top5 98.554688 -2022-12-06 11:32:40,769 - Epoch: [156][ 130/ 134] Loss 0.235056 Top1 87.971755 Top5 98.536659 -2022-12-06 11:32:40,806 - Epoch: [156][ 134/ 134] Loss 0.236226 Top1 87.887134 Top5 98.532040 -2022-12-06 11:32:40,893 - ==> Top1: 87.887 Top5: 98.532 Loss: 0.236 - -2022-12-06 11:32:40,894 - ==> Confusion: -[[ 924 1 1 2 6 4 0 0 6 30 0 1 1 3 5 2 1 0 2 0 7] - [ 1 943 3 3 7 20 2 11 1 1 3 4 1 0 0 1 3 1 8 5 9] - [ 2 2 1018 11 5 2 11 9 1 5 6 3 2 2 0 2 1 1 3 4 13] - [ 0 0 17 954 0 2 1 1 0 2 9 0 5 0 9 1 0 2 10 0 7] - [ 9 4 2 0 960 1 0 2 0 5 2 2 1 2 10 4 7 1 0 1 7] - [ 1 18 0 3 6 980 0 19 1 2 1 9 5 9 1 2 2 2 0 4 4] - [ 0 5 12 3 1 2 1072 2 1 0 0 2 0 1 0 3 0 2 0 9 3] - [ 1 7 5 2 2 28 11 953 0 0 3 5 1 2 1 1 0 0 16 10 6] - [ 10 2 0 0 0 0 1 0 983 33 13 1 3 6 6 0 1 1 2 2 0] - [ 59 0 0 0 3 2 0 3 19 887 2 2 1 11 3 1 0 0 1 0 7] - [ 2 2 5 5 1 1 0 4 7 0 959 1 0 11 3 1 1 0 6 1 9] - [ 1 2 2 0 0 8 4 1 1 0 0 975 26 4 0 10 2 4 0 6 5] - [ 1 1 2 1 1 2 0 2 0 0 0 21 909 0 1 5 1 13 0 4 5] - [ 1 1 1 0 1 8 0 3 13 12 2 1 4 957 0 2 3 1 0 1 12] - [ 10 4 1 7 5 1 0 0 13 3 0 1 2 3 1068 0 0 1 5 2 4] - [ 0 0 1 0 1 1 2 0 0 0 2 7 10 1 0 997 5 10 0 3 3] - [ 2 2 2 1 2 0 0 0 1 0 0 2 4 2 0 9 1033 0 0 5 7] - [ 4 0 2 1 0 0 0 0 0 2 0 6 14 2 1 16 0 983 0 2 3] - [ 2 4 2 5 2 2 2 20 2 0 2 1 4 0 10 1 0 2 940 2 5] - [ 2 3 1 1 0 5 5 1 0 1 2 18 6 8 0 3 3 2 1 1011 7] - [ 112 203 144 94 99 127 69 120 74 76 143 83 331 251 119 109 162 85 139 204 10482]] - -2022-12-06 11:32:41,465 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:32:41,466 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:32:41,472 - - -2022-12-06 11:32:41,472 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:32:42,403 - Epoch: [157][ 10/ 1200] Overall Loss 0.152433 Objective Loss 0.152433 LR 0.000250 Time 0.093013 -2022-12-06 11:32:42,596 - Epoch: [157][ 20/ 1200] Overall Loss 0.149333 Objective Loss 0.149333 LR 0.000250 Time 0.056149 -2022-12-06 11:32:42,788 - Epoch: [157][ 30/ 1200] Overall Loss 0.150288 Objective Loss 0.150288 LR 0.000250 Time 0.043804 -2022-12-06 11:32:42,979 - Epoch: [157][ 40/ 1200] Overall Loss 0.153950 Objective Loss 0.153950 LR 0.000250 Time 0.037611 -2022-12-06 11:32:43,169 - Epoch: [157][ 50/ 1200] Overall Loss 0.151756 Objective Loss 0.151756 LR 0.000250 Time 0.033879 -2022-12-06 11:32:43,359 - Epoch: [157][ 60/ 1200] Overall Loss 0.147480 Objective Loss 0.147480 LR 0.000250 Time 0.031397 -2022-12-06 11:32:43,550 - Epoch: [157][ 70/ 1200] Overall Loss 0.143636 Objective Loss 0.143636 LR 0.000250 Time 0.029626 -2022-12-06 11:32:43,741 - Epoch: [157][ 80/ 1200] Overall Loss 0.145607 Objective Loss 0.145607 LR 0.000250 Time 0.028303 -2022-12-06 11:32:43,931 - Epoch: [157][ 90/ 1200] Overall Loss 0.144314 Objective Loss 0.144314 LR 0.000250 Time 0.027267 -2022-12-06 11:32:44,122 - Epoch: [157][ 100/ 1200] Overall Loss 0.143490 Objective Loss 0.143490 LR 0.000250 Time 0.026443 -2022-12-06 11:32:44,313 - Epoch: [157][ 110/ 1200] Overall Loss 0.142705 Objective Loss 0.142705 LR 0.000250 Time 0.025771 -2022-12-06 11:32:44,503 - Epoch: [157][ 120/ 1200] Overall Loss 0.141518 Objective Loss 0.141518 LR 0.000250 Time 0.025203 -2022-12-06 11:32:44,695 - Epoch: [157][ 130/ 1200] Overall Loss 0.141719 Objective Loss 0.141719 LR 0.000250 Time 0.024734 -2022-12-06 11:32:44,886 - Epoch: [157][ 140/ 1200] Overall Loss 0.141244 Objective Loss 0.141244 LR 0.000250 Time 0.024329 -2022-12-06 11:32:45,077 - Epoch: [157][ 150/ 1200] Overall Loss 0.140047 Objective Loss 0.140047 LR 0.000250 Time 0.023976 -2022-12-06 11:32:45,267 - Epoch: [157][ 160/ 1200] Overall Loss 0.140837 Objective Loss 0.140837 LR 0.000250 Time 0.023662 -2022-12-06 11:32:45,457 - Epoch: [157][ 170/ 1200] Overall Loss 0.141164 Objective Loss 0.141164 LR 0.000250 Time 0.023388 -2022-12-06 11:32:45,649 - Epoch: [157][ 180/ 1200] Overall Loss 0.141609 Objective Loss 0.141609 LR 0.000250 Time 0.023148 -2022-12-06 11:32:45,839 - Epoch: [157][ 190/ 1200] Overall Loss 0.143282 Objective Loss 0.143282 LR 0.000250 Time 0.022929 -2022-12-06 11:32:46,030 - Epoch: [157][ 200/ 1200] Overall Loss 0.144408 Objective Loss 0.144408 LR 0.000250 Time 0.022733 -2022-12-06 11:32:46,220 - Epoch: [157][ 210/ 1200] Overall Loss 0.143821 Objective Loss 0.143821 LR 0.000250 Time 0.022555 -2022-12-06 11:32:46,411 - Epoch: [157][ 220/ 1200] Overall Loss 0.144994 Objective Loss 0.144994 LR 0.000250 Time 0.022393 -2022-12-06 11:32:46,601 - Epoch: [157][ 230/ 1200] Overall Loss 0.144647 Objective Loss 0.144647 LR 0.000250 Time 0.022244 -2022-12-06 11:32:46,791 - Epoch: [157][ 240/ 1200] Overall Loss 0.144631 Objective Loss 0.144631 LR 0.000250 Time 0.022108 -2022-12-06 11:32:46,981 - Epoch: [157][ 250/ 1200] Overall Loss 0.144557 Objective Loss 0.144557 LR 0.000250 Time 0.021982 -2022-12-06 11:32:47,172 - Epoch: [157][ 260/ 1200] Overall Loss 0.144762 Objective Loss 0.144762 LR 0.000250 Time 0.021866 -2022-12-06 11:32:47,362 - Epoch: [157][ 270/ 1200] Overall Loss 0.145863 Objective Loss 0.145863 LR 0.000250 Time 0.021758 -2022-12-06 11:32:47,552 - Epoch: [157][ 280/ 1200] Overall Loss 0.145621 Objective Loss 0.145621 LR 0.000250 Time 0.021661 -2022-12-06 11:32:47,743 - Epoch: [157][ 290/ 1200] Overall Loss 0.145809 Objective Loss 0.145809 LR 0.000250 Time 0.021570 -2022-12-06 11:32:47,934 - Epoch: [157][ 300/ 1200] Overall Loss 0.145928 Objective Loss 0.145928 LR 0.000250 Time 0.021484 -2022-12-06 11:32:48,125 - Epoch: [157][ 310/ 1200] Overall Loss 0.145708 Objective Loss 0.145708 LR 0.000250 Time 0.021405 -2022-12-06 11:32:48,315 - Epoch: [157][ 320/ 1200] Overall Loss 0.145780 Objective Loss 0.145780 LR 0.000250 Time 0.021328 -2022-12-06 11:32:48,505 - Epoch: [157][ 330/ 1200] Overall Loss 0.146360 Objective Loss 0.146360 LR 0.000250 Time 0.021255 -2022-12-06 11:32:48,695 - Epoch: [157][ 340/ 1200] Overall Loss 0.146518 Objective Loss 0.146518 LR 0.000250 Time 0.021188 -2022-12-06 11:32:48,885 - Epoch: [157][ 350/ 1200] Overall Loss 0.146955 Objective Loss 0.146955 LR 0.000250 Time 0.021125 -2022-12-06 11:32:49,077 - Epoch: [157][ 360/ 1200] Overall Loss 0.146940 Objective Loss 0.146940 LR 0.000250 Time 0.021070 -2022-12-06 11:32:49,268 - Epoch: [157][ 370/ 1200] Overall Loss 0.147013 Objective Loss 0.147013 LR 0.000250 Time 0.021014 -2022-12-06 11:32:49,458 - Epoch: [157][ 380/ 1200] Overall Loss 0.146737 Objective Loss 0.146737 LR 0.000250 Time 0.020960 -2022-12-06 11:32:49,649 - Epoch: [157][ 390/ 1200] Overall Loss 0.146249 Objective Loss 0.146249 LR 0.000250 Time 0.020911 -2022-12-06 11:32:49,839 - Epoch: [157][ 400/ 1200] Overall Loss 0.146592 Objective Loss 0.146592 LR 0.000250 Time 0.020861 -2022-12-06 11:32:50,029 - Epoch: [157][ 410/ 1200] Overall Loss 0.146301 Objective Loss 0.146301 LR 0.000250 Time 0.020816 -2022-12-06 11:32:50,220 - Epoch: [157][ 420/ 1200] Overall Loss 0.146164 Objective Loss 0.146164 LR 0.000250 Time 0.020773 -2022-12-06 11:32:50,410 - Epoch: [157][ 430/ 1200] Overall Loss 0.146450 Objective Loss 0.146450 LR 0.000250 Time 0.020731 -2022-12-06 11:32:50,601 - Epoch: [157][ 440/ 1200] Overall Loss 0.146368 Objective Loss 0.146368 LR 0.000250 Time 0.020692 -2022-12-06 11:32:50,791 - Epoch: [157][ 450/ 1200] Overall Loss 0.146355 Objective Loss 0.146355 LR 0.000250 Time 0.020654 -2022-12-06 11:32:50,982 - Epoch: [157][ 460/ 1200] Overall Loss 0.146483 Objective Loss 0.146483 LR 0.000250 Time 0.020618 -2022-12-06 11:32:51,172 - Epoch: [157][ 470/ 1200] Overall Loss 0.146414 Objective Loss 0.146414 LR 0.000250 Time 0.020583 -2022-12-06 11:32:51,362 - Epoch: [157][ 480/ 1200] Overall Loss 0.146669 Objective Loss 0.146669 LR 0.000250 Time 0.020550 -2022-12-06 11:32:51,553 - Epoch: [157][ 490/ 1200] Overall Loss 0.147039 Objective Loss 0.147039 LR 0.000250 Time 0.020518 -2022-12-06 11:32:51,743 - Epoch: [157][ 500/ 1200] Overall Loss 0.147472 Objective Loss 0.147472 LR 0.000250 Time 0.020487 -2022-12-06 11:32:51,935 - Epoch: [157][ 510/ 1200] Overall Loss 0.147511 Objective Loss 0.147511 LR 0.000250 Time 0.020460 -2022-12-06 11:32:52,128 - Epoch: [157][ 520/ 1200] Overall Loss 0.147880 Objective Loss 0.147880 LR 0.000250 Time 0.020436 -2022-12-06 11:32:52,320 - Epoch: [157][ 530/ 1200] Overall Loss 0.147675 Objective Loss 0.147675 LR 0.000250 Time 0.020413 -2022-12-06 11:32:52,512 - Epoch: [157][ 540/ 1200] Overall Loss 0.147741 Objective Loss 0.147741 LR 0.000250 Time 0.020389 -2022-12-06 11:32:52,705 - Epoch: [157][ 550/ 1200] Overall Loss 0.147705 Objective Loss 0.147705 LR 0.000250 Time 0.020367 -2022-12-06 11:32:52,897 - Epoch: [157][ 560/ 1200] Overall Loss 0.147606 Objective Loss 0.147606 LR 0.000250 Time 0.020347 -2022-12-06 11:32:53,090 - Epoch: [157][ 570/ 1200] Overall Loss 0.147741 Objective Loss 0.147741 LR 0.000250 Time 0.020327 -2022-12-06 11:32:53,283 - Epoch: [157][ 580/ 1200] Overall Loss 0.147957 Objective Loss 0.147957 LR 0.000250 Time 0.020308 -2022-12-06 11:32:53,476 - Epoch: [157][ 590/ 1200] Overall Loss 0.148094 Objective Loss 0.148094 LR 0.000250 Time 0.020291 -2022-12-06 11:32:53,668 - Epoch: [157][ 600/ 1200] Overall Loss 0.148081 Objective Loss 0.148081 LR 0.000250 Time 0.020272 -2022-12-06 11:32:53,861 - Epoch: [157][ 610/ 1200] Overall Loss 0.148418 Objective Loss 0.148418 LR 0.000250 Time 0.020254 -2022-12-06 11:32:54,053 - Epoch: [157][ 620/ 1200] Overall Loss 0.148670 Objective Loss 0.148670 LR 0.000250 Time 0.020237 -2022-12-06 11:32:54,246 - Epoch: [157][ 630/ 1200] Overall Loss 0.148489 Objective Loss 0.148489 LR 0.000250 Time 0.020220 -2022-12-06 11:32:54,439 - Epoch: [157][ 640/ 1200] Overall Loss 0.149013 Objective Loss 0.149013 LR 0.000250 Time 0.020205 -2022-12-06 11:32:54,631 - Epoch: [157][ 650/ 1200] Overall Loss 0.148790 Objective Loss 0.148790 LR 0.000250 Time 0.020189 -2022-12-06 11:32:54,823 - Epoch: [157][ 660/ 1200] Overall Loss 0.148808 Objective Loss 0.148808 LR 0.000250 Time 0.020174 -2022-12-06 11:32:55,016 - Epoch: [157][ 670/ 1200] Overall Loss 0.148756 Objective Loss 0.148756 LR 0.000250 Time 0.020160 -2022-12-06 11:32:55,208 - Epoch: [157][ 680/ 1200] Overall Loss 0.149049 Objective Loss 0.149049 LR 0.000250 Time 0.020145 -2022-12-06 11:32:55,401 - Epoch: [157][ 690/ 1200] Overall Loss 0.148937 Objective Loss 0.148937 LR 0.000250 Time 0.020131 -2022-12-06 11:32:55,593 - Epoch: [157][ 700/ 1200] Overall Loss 0.148986 Objective Loss 0.148986 LR 0.000250 Time 0.020117 -2022-12-06 11:32:55,786 - Epoch: [157][ 710/ 1200] Overall Loss 0.149109 Objective Loss 0.149109 LR 0.000250 Time 0.020104 -2022-12-06 11:32:55,977 - Epoch: [157][ 720/ 1200] Overall Loss 0.149055 Objective Loss 0.149055 LR 0.000250 Time 0.020091 -2022-12-06 11:32:56,169 - Epoch: [157][ 730/ 1200] Overall Loss 0.148905 Objective Loss 0.148905 LR 0.000250 Time 0.020078 -2022-12-06 11:32:56,360 - Epoch: [157][ 740/ 1200] Overall Loss 0.148928 Objective Loss 0.148928 LR 0.000250 Time 0.020063 -2022-12-06 11:32:56,549 - Epoch: [157][ 750/ 1200] Overall Loss 0.148951 Objective Loss 0.148951 LR 0.000250 Time 0.020047 -2022-12-06 11:32:56,739 - Epoch: [157][ 760/ 1200] Overall Loss 0.148830 Objective Loss 0.148830 LR 0.000250 Time 0.020032 -2022-12-06 11:32:56,929 - Epoch: [157][ 770/ 1200] Overall Loss 0.148767 Objective Loss 0.148767 LR 0.000250 Time 0.020018 -2022-12-06 11:32:57,119 - Epoch: [157][ 780/ 1200] Overall Loss 0.148336 Objective Loss 0.148336 LR 0.000250 Time 0.020005 -2022-12-06 11:32:57,309 - Epoch: [157][ 790/ 1200] Overall Loss 0.148146 Objective Loss 0.148146 LR 0.000250 Time 0.019992 -2022-12-06 11:32:57,499 - Epoch: [157][ 800/ 1200] Overall Loss 0.148322 Objective Loss 0.148322 LR 0.000250 Time 0.019979 -2022-12-06 11:32:57,689 - Epoch: [157][ 810/ 1200] Overall Loss 0.148440 Objective Loss 0.148440 LR 0.000250 Time 0.019966 -2022-12-06 11:32:57,879 - Epoch: [157][ 820/ 1200] Overall Loss 0.148502 Objective Loss 0.148502 LR 0.000250 Time 0.019953 -2022-12-06 11:32:58,069 - Epoch: [157][ 830/ 1200] Overall Loss 0.148445 Objective Loss 0.148445 LR 0.000250 Time 0.019941 -2022-12-06 11:32:58,258 - Epoch: [157][ 840/ 1200] Overall Loss 0.148238 Objective Loss 0.148238 LR 0.000250 Time 0.019929 -2022-12-06 11:32:58,448 - Epoch: [157][ 850/ 1200] Overall Loss 0.148020 Objective Loss 0.148020 LR 0.000250 Time 0.019917 -2022-12-06 11:32:58,639 - Epoch: [157][ 860/ 1200] Overall Loss 0.148163 Objective Loss 0.148163 LR 0.000250 Time 0.019906 -2022-12-06 11:32:58,829 - Epoch: [157][ 870/ 1200] Overall Loss 0.148340 Objective Loss 0.148340 LR 0.000250 Time 0.019895 -2022-12-06 11:32:59,019 - Epoch: [157][ 880/ 1200] Overall Loss 0.148438 Objective Loss 0.148438 LR 0.000250 Time 0.019884 -2022-12-06 11:32:59,208 - Epoch: [157][ 890/ 1200] Overall Loss 0.148511 Objective Loss 0.148511 LR 0.000250 Time 0.019873 -2022-12-06 11:32:59,398 - Epoch: [157][ 900/ 1200] Overall Loss 0.148495 Objective Loss 0.148495 LR 0.000250 Time 0.019863 -2022-12-06 11:32:59,588 - Epoch: [157][ 910/ 1200] Overall Loss 0.148605 Objective Loss 0.148605 LR 0.000250 Time 0.019852 -2022-12-06 11:32:59,778 - Epoch: [157][ 920/ 1200] Overall Loss 0.148538 Objective Loss 0.148538 LR 0.000250 Time 0.019843 -2022-12-06 11:32:59,969 - Epoch: [157][ 930/ 1200] Overall Loss 0.148891 Objective Loss 0.148891 LR 0.000250 Time 0.019834 -2022-12-06 11:33:00,159 - Epoch: [157][ 940/ 1200] Overall Loss 0.149039 Objective Loss 0.149039 LR 0.000250 Time 0.019825 -2022-12-06 11:33:00,350 - Epoch: [157][ 950/ 1200] Overall Loss 0.149254 Objective Loss 0.149254 LR 0.000250 Time 0.019817 -2022-12-06 11:33:00,541 - Epoch: [157][ 960/ 1200] Overall Loss 0.149224 Objective Loss 0.149224 LR 0.000250 Time 0.019808 -2022-12-06 11:33:00,731 - Epoch: [157][ 970/ 1200] Overall Loss 0.149270 Objective Loss 0.149270 LR 0.000250 Time 0.019800 -2022-12-06 11:33:00,922 - Epoch: [157][ 980/ 1200] Overall Loss 0.149153 Objective Loss 0.149153 LR 0.000250 Time 0.019791 -2022-12-06 11:33:01,112 - Epoch: [157][ 990/ 1200] Overall Loss 0.149258 Objective Loss 0.149258 LR 0.000250 Time 0.019784 -2022-12-06 11:33:01,303 - Epoch: [157][ 1000/ 1200] Overall Loss 0.149066 Objective Loss 0.149066 LR 0.000250 Time 0.019776 -2022-12-06 11:33:01,493 - Epoch: [157][ 1010/ 1200] Overall Loss 0.148979 Objective Loss 0.148979 LR 0.000250 Time 0.019768 -2022-12-06 11:33:01,683 - Epoch: [157][ 1020/ 1200] Overall Loss 0.149209 Objective Loss 0.149209 LR 0.000250 Time 0.019760 -2022-12-06 11:33:01,874 - Epoch: [157][ 1030/ 1200] Overall Loss 0.149324 Objective Loss 0.149324 LR 0.000250 Time 0.019752 -2022-12-06 11:33:02,065 - Epoch: [157][ 1040/ 1200] Overall Loss 0.149416 Objective Loss 0.149416 LR 0.000250 Time 0.019746 -2022-12-06 11:33:02,256 - Epoch: [157][ 1050/ 1200] Overall Loss 0.149490 Objective Loss 0.149490 LR 0.000250 Time 0.019739 -2022-12-06 11:33:02,446 - Epoch: [157][ 1060/ 1200] Overall Loss 0.149322 Objective Loss 0.149322 LR 0.000250 Time 0.019732 -2022-12-06 11:33:02,636 - Epoch: [157][ 1070/ 1200] Overall Loss 0.149527 Objective Loss 0.149527 LR 0.000250 Time 0.019724 -2022-12-06 11:33:02,827 - Epoch: [157][ 1080/ 1200] Overall Loss 0.149616 Objective Loss 0.149616 LR 0.000250 Time 0.019717 -2022-12-06 11:33:03,016 - Epoch: [157][ 1090/ 1200] Overall Loss 0.149606 Objective Loss 0.149606 LR 0.000250 Time 0.019710 -2022-12-06 11:33:03,208 - Epoch: [157][ 1100/ 1200] Overall Loss 0.149651 Objective Loss 0.149651 LR 0.000250 Time 0.019704 -2022-12-06 11:33:03,398 - Epoch: [157][ 1110/ 1200] Overall Loss 0.149809 Objective Loss 0.149809 LR 0.000250 Time 0.019697 -2022-12-06 11:33:03,589 - Epoch: [157][ 1120/ 1200] Overall Loss 0.149853 Objective Loss 0.149853 LR 0.000250 Time 0.019692 -2022-12-06 11:33:03,780 - Epoch: [157][ 1130/ 1200] Overall Loss 0.149892 Objective Loss 0.149892 LR 0.000250 Time 0.019686 -2022-12-06 11:33:03,970 - Epoch: [157][ 1140/ 1200] Overall Loss 0.149802 Objective Loss 0.149802 LR 0.000250 Time 0.019680 -2022-12-06 11:33:04,161 - Epoch: [157][ 1150/ 1200] Overall Loss 0.149879 Objective Loss 0.149879 LR 0.000250 Time 0.019674 -2022-12-06 11:33:04,351 - Epoch: [157][ 1160/ 1200] Overall Loss 0.149769 Objective Loss 0.149769 LR 0.000250 Time 0.019668 -2022-12-06 11:33:04,541 - Epoch: [157][ 1170/ 1200] Overall Loss 0.149745 Objective Loss 0.149745 LR 0.000250 Time 0.019662 -2022-12-06 11:33:04,732 - Epoch: [157][ 1180/ 1200] Overall Loss 0.149665 Objective Loss 0.149665 LR 0.000250 Time 0.019656 -2022-12-06 11:33:04,923 - Epoch: [157][ 1190/ 1200] Overall Loss 0.149672 Objective Loss 0.149672 LR 0.000250 Time 0.019651 -2022-12-06 11:33:05,144 - Epoch: [157][ 1200/ 1200] Overall Loss 0.149686 Objective Loss 0.149686 Top1 89.958159 Top5 99.163180 LR 0.000250 Time 0.019671 -2022-12-06 11:33:05,232 - --- validate (epoch=157)----------- -2022-12-06 11:33:05,232 - 34129 samples (256 per mini-batch) -2022-12-06 11:33:05,682 - Epoch: [157][ 10/ 134] Loss 0.235184 Top1 87.695312 Top5 98.984375 -2022-12-06 11:33:05,812 - Epoch: [157][ 20/ 134] Loss 0.233259 Top1 87.675781 Top5 98.730469 -2022-12-06 11:33:05,939 - Epoch: [157][ 30/ 134] Loss 0.234693 Top1 87.591146 Top5 98.710938 -2022-12-06 11:33:06,068 - Epoch: [157][ 40/ 134] Loss 0.238878 Top1 87.744141 Top5 98.593750 -2022-12-06 11:33:06,196 - Epoch: [157][ 50/ 134] Loss 0.237014 Top1 87.687500 Top5 98.585938 -2022-12-06 11:33:06,324 - Epoch: [157][ 60/ 134] Loss 0.237884 Top1 87.617188 Top5 98.541667 -2022-12-06 11:33:06,452 - Epoch: [157][ 70/ 134] Loss 0.240656 Top1 87.639509 Top5 98.526786 -2022-12-06 11:33:06,580 - Epoch: [157][ 80/ 134] Loss 0.237222 Top1 87.651367 Top5 98.530273 -2022-12-06 11:33:06,708 - Epoch: [157][ 90/ 134] Loss 0.233810 Top1 87.669271 Top5 98.546007 -2022-12-06 11:33:06,837 - Epoch: [157][ 100/ 134] Loss 0.235327 Top1 87.542969 Top5 98.488281 -2022-12-06 11:33:06,964 - Epoch: [157][ 110/ 134] Loss 0.233790 Top1 87.528409 Top5 98.497869 -2022-12-06 11:33:07,094 - Epoch: [157][ 120/ 134] Loss 0.233829 Top1 87.587891 Top5 98.505859 -2022-12-06 11:33:07,223 - Epoch: [157][ 130/ 134] Loss 0.233296 Top1 87.557091 Top5 98.512620 -2022-12-06 11:33:07,260 - Epoch: [157][ 134/ 134] Loss 0.233212 Top1 87.588268 Top5 98.520320 -2022-12-06 11:33:07,362 - ==> Top1: 87.588 Top5: 98.520 Loss: 0.233 - -2022-12-06 11:33:07,363 - ==> Confusion: -[[ 924 0 0 1 3 6 0 0 5 38 0 1 1 3 5 1 1 1 2 0 4] - [ 1 944 0 3 7 18 4 14 2 0 5 5 0 1 0 1 3 1 10 3 5] - [ 4 3 1006 13 4 3 18 10 0 3 8 2 3 3 1 1 1 2 4 3 11] - [ 0 3 5 969 0 2 0 1 1 1 8 0 3 2 9 1 0 2 8 0 5] - [ 10 5 0 0 958 4 0 2 0 5 1 3 1 4 8 4 7 2 3 1 2] - [ 1 12 0 5 5 991 2 19 1 3 0 9 3 10 2 1 1 0 0 3 1] - [ 2 4 6 1 0 2 1077 6 0 0 1 0 0 2 0 2 0 4 1 9 1] - [ 1 3 4 2 3 18 6 977 1 1 2 3 1 2 1 0 1 0 16 8 4] - [ 7 2 0 1 0 1 1 0 983 34 14 1 1 4 9 0 1 0 2 2 1] - [ 49 0 0 0 4 3 0 2 22 899 1 1 0 9 3 0 0 1 1 0 6] - [ 2 1 0 5 2 1 0 2 4 1 971 1 1 10 4 0 0 0 3 2 9] - [ 2 1 1 0 0 12 2 2 1 0 0 988 15 4 1 4 3 4 0 8 3] - [ 0 1 2 2 0 2 0 1 0 1 0 31 902 1 1 5 1 10 0 3 6] - [ 0 1 0 0 3 5 0 2 12 11 6 3 3 962 2 1 1 0 0 2 9] - [ 6 3 1 12 4 2 0 0 18 3 0 3 2 2 1063 0 0 1 6 1 3] - [ 1 0 1 0 1 0 2 0 1 0 1 9 8 3 0 991 6 9 1 4 5] - [ 4 2 0 0 1 1 1 2 1 0 0 3 4 2 0 8 1034 0 1 3 5] - [ 3 0 1 1 1 1 1 2 1 4 0 4 17 1 1 13 0 983 0 0 2] - [ 2 4 0 9 1 2 0 27 1 1 5 2 2 0 9 0 0 1 937 1 4] - [ 3 2 2 1 0 3 6 6 0 0 3 14 5 5 0 2 3 4 1 1016 4] - [ 116 202 127 118 90 175 71 149 83 79 163 107 311 248 155 95 158 88 151 225 10315]] - -2022-12-06 11:33:08,037 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:33:08,037 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:33:08,043 - - -2022-12-06 11:33:08,043 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:33:08,996 - Epoch: [158][ 10/ 1200] Overall Loss 0.155904 Objective Loss 0.155904 LR 0.000250 Time 0.095216 -2022-12-06 11:33:09,192 - Epoch: [158][ 20/ 1200] Overall Loss 0.152874 Objective Loss 0.152874 LR 0.000250 Time 0.057397 -2022-12-06 11:33:09,385 - Epoch: [158][ 30/ 1200] Overall Loss 0.148532 Objective Loss 0.148532 LR 0.000250 Time 0.044684 -2022-12-06 11:33:09,577 - Epoch: [158][ 40/ 1200] Overall Loss 0.146807 Objective Loss 0.146807 LR 0.000250 Time 0.038295 -2022-12-06 11:33:09,770 - Epoch: [158][ 50/ 1200] Overall Loss 0.145291 Objective Loss 0.145291 LR 0.000250 Time 0.034489 -2022-12-06 11:33:09,962 - Epoch: [158][ 60/ 1200] Overall Loss 0.146948 Objective Loss 0.146948 LR 0.000250 Time 0.031930 -2022-12-06 11:33:10,155 - Epoch: [158][ 70/ 1200] Overall Loss 0.145213 Objective Loss 0.145213 LR 0.000250 Time 0.030110 -2022-12-06 11:33:10,346 - Epoch: [158][ 80/ 1200] Overall Loss 0.145665 Objective Loss 0.145665 LR 0.000250 Time 0.028738 -2022-12-06 11:33:10,539 - Epoch: [158][ 90/ 1200] Overall Loss 0.144905 Objective Loss 0.144905 LR 0.000250 Time 0.027677 -2022-12-06 11:33:10,731 - Epoch: [158][ 100/ 1200] Overall Loss 0.143790 Objective Loss 0.143790 LR 0.000250 Time 0.026826 -2022-12-06 11:33:10,923 - Epoch: [158][ 110/ 1200] Overall Loss 0.143415 Objective Loss 0.143415 LR 0.000250 Time 0.026131 -2022-12-06 11:33:11,115 - Epoch: [158][ 120/ 1200] Overall Loss 0.143097 Objective Loss 0.143097 LR 0.000250 Time 0.025549 -2022-12-06 11:33:11,309 - Epoch: [158][ 130/ 1200] Overall Loss 0.143026 Objective Loss 0.143026 LR 0.000250 Time 0.025068 -2022-12-06 11:33:11,501 - Epoch: [158][ 140/ 1200] Overall Loss 0.143133 Objective Loss 0.143133 LR 0.000250 Time 0.024646 -2022-12-06 11:33:11,693 - Epoch: [158][ 150/ 1200] Overall Loss 0.145574 Objective Loss 0.145574 LR 0.000250 Time 0.024281 -2022-12-06 11:33:11,885 - Epoch: [158][ 160/ 1200] Overall Loss 0.146074 Objective Loss 0.146074 LR 0.000250 Time 0.023958 -2022-12-06 11:33:12,078 - Epoch: [158][ 170/ 1200] Overall Loss 0.145403 Objective Loss 0.145403 LR 0.000250 Time 0.023679 -2022-12-06 11:33:12,270 - Epoch: [158][ 180/ 1200] Overall Loss 0.144810 Objective Loss 0.144810 LR 0.000250 Time 0.023428 -2022-12-06 11:33:12,462 - Epoch: [158][ 190/ 1200] Overall Loss 0.145297 Objective Loss 0.145297 LR 0.000250 Time 0.023205 -2022-12-06 11:33:12,655 - Epoch: [158][ 200/ 1200] Overall Loss 0.145818 Objective Loss 0.145818 LR 0.000250 Time 0.023004 -2022-12-06 11:33:12,847 - Epoch: [158][ 210/ 1200] Overall Loss 0.146008 Objective Loss 0.146008 LR 0.000250 Time 0.022820 -2022-12-06 11:33:13,038 - Epoch: [158][ 220/ 1200] Overall Loss 0.144897 Objective Loss 0.144897 LR 0.000250 Time 0.022649 -2022-12-06 11:33:13,231 - Epoch: [158][ 230/ 1200] Overall Loss 0.145870 Objective Loss 0.145870 LR 0.000250 Time 0.022501 -2022-12-06 11:33:13,422 - Epoch: [158][ 240/ 1200] Overall Loss 0.145606 Objective Loss 0.145606 LR 0.000250 Time 0.022359 -2022-12-06 11:33:13,615 - Epoch: [158][ 250/ 1200] Overall Loss 0.146691 Objective Loss 0.146691 LR 0.000250 Time 0.022236 -2022-12-06 11:33:13,807 - Epoch: [158][ 260/ 1200] Overall Loss 0.146882 Objective Loss 0.146882 LR 0.000250 Time 0.022117 -2022-12-06 11:33:14,000 - Epoch: [158][ 270/ 1200] Overall Loss 0.146901 Objective Loss 0.146901 LR 0.000250 Time 0.022009 -2022-12-06 11:33:14,192 - Epoch: [158][ 280/ 1200] Overall Loss 0.147201 Objective Loss 0.147201 LR 0.000250 Time 0.021906 -2022-12-06 11:33:14,384 - Epoch: [158][ 290/ 1200] Overall Loss 0.146992 Objective Loss 0.146992 LR 0.000250 Time 0.021813 -2022-12-06 11:33:14,577 - Epoch: [158][ 300/ 1200] Overall Loss 0.147766 Objective Loss 0.147766 LR 0.000250 Time 0.021725 -2022-12-06 11:33:14,770 - Epoch: [158][ 310/ 1200] Overall Loss 0.147478 Objective Loss 0.147478 LR 0.000250 Time 0.021644 -2022-12-06 11:33:14,961 - Epoch: [158][ 320/ 1200] Overall Loss 0.147089 Objective Loss 0.147089 LR 0.000250 Time 0.021564 -2022-12-06 11:33:15,154 - Epoch: [158][ 330/ 1200] Overall Loss 0.146647 Objective Loss 0.146647 LR 0.000250 Time 0.021494 -2022-12-06 11:33:15,346 - Epoch: [158][ 340/ 1200] Overall Loss 0.145891 Objective Loss 0.145891 LR 0.000250 Time 0.021424 -2022-12-06 11:33:15,539 - Epoch: [158][ 350/ 1200] Overall Loss 0.145515 Objective Loss 0.145515 LR 0.000250 Time 0.021362 -2022-12-06 11:33:15,731 - Epoch: [158][ 360/ 1200] Overall Loss 0.145351 Objective Loss 0.145351 LR 0.000250 Time 0.021302 -2022-12-06 11:33:15,924 - Epoch: [158][ 370/ 1200] Overall Loss 0.144982 Objective Loss 0.144982 LR 0.000250 Time 0.021245 -2022-12-06 11:33:16,116 - Epoch: [158][ 380/ 1200] Overall Loss 0.144886 Objective Loss 0.144886 LR 0.000250 Time 0.021190 -2022-12-06 11:33:16,309 - Epoch: [158][ 390/ 1200] Overall Loss 0.144861 Objective Loss 0.144861 LR 0.000250 Time 0.021140 -2022-12-06 11:33:16,500 - Epoch: [158][ 400/ 1200] Overall Loss 0.144664 Objective Loss 0.144664 LR 0.000250 Time 0.021089 -2022-12-06 11:33:16,693 - Epoch: [158][ 410/ 1200] Overall Loss 0.145155 Objective Loss 0.145155 LR 0.000250 Time 0.021043 -2022-12-06 11:33:16,885 - Epoch: [158][ 420/ 1200] Overall Loss 0.145004 Objective Loss 0.145004 LR 0.000250 Time 0.020998 -2022-12-06 11:33:17,078 - Epoch: [158][ 430/ 1200] Overall Loss 0.144783 Objective Loss 0.144783 LR 0.000250 Time 0.020957 -2022-12-06 11:33:17,270 - Epoch: [158][ 440/ 1200] Overall Loss 0.145226 Objective Loss 0.145226 LR 0.000250 Time 0.020917 -2022-12-06 11:33:17,463 - Epoch: [158][ 450/ 1200] Overall Loss 0.145080 Objective Loss 0.145080 LR 0.000250 Time 0.020880 -2022-12-06 11:33:17,656 - Epoch: [158][ 460/ 1200] Overall Loss 0.145202 Objective Loss 0.145202 LR 0.000250 Time 0.020843 -2022-12-06 11:33:17,849 - Epoch: [158][ 470/ 1200] Overall Loss 0.145345 Objective Loss 0.145345 LR 0.000250 Time 0.020810 -2022-12-06 11:33:18,042 - Epoch: [158][ 480/ 1200] Overall Loss 0.145152 Objective Loss 0.145152 LR 0.000250 Time 0.020776 -2022-12-06 11:33:18,234 - Epoch: [158][ 490/ 1200] Overall Loss 0.145581 Objective Loss 0.145581 LR 0.000250 Time 0.020744 -2022-12-06 11:33:18,427 - Epoch: [158][ 500/ 1200] Overall Loss 0.145056 Objective Loss 0.145056 LR 0.000250 Time 0.020713 -2022-12-06 11:33:18,620 - Epoch: [158][ 510/ 1200] Overall Loss 0.145248 Objective Loss 0.145248 LR 0.000250 Time 0.020685 -2022-12-06 11:33:18,812 - Epoch: [158][ 520/ 1200] Overall Loss 0.145186 Objective Loss 0.145186 LR 0.000250 Time 0.020655 -2022-12-06 11:33:19,005 - Epoch: [158][ 530/ 1200] Overall Loss 0.144956 Objective Loss 0.144956 LR 0.000250 Time 0.020629 -2022-12-06 11:33:19,197 - Epoch: [158][ 540/ 1200] Overall Loss 0.145033 Objective Loss 0.145033 LR 0.000250 Time 0.020602 -2022-12-06 11:33:19,390 - Epoch: [158][ 550/ 1200] Overall Loss 0.145244 Objective Loss 0.145244 LR 0.000250 Time 0.020577 -2022-12-06 11:33:19,581 - Epoch: [158][ 560/ 1200] Overall Loss 0.145200 Objective Loss 0.145200 LR 0.000250 Time 0.020549 -2022-12-06 11:33:19,772 - Epoch: [158][ 570/ 1200] Overall Loss 0.144761 Objective Loss 0.144761 LR 0.000250 Time 0.020523 -2022-12-06 11:33:19,963 - Epoch: [158][ 580/ 1200] Overall Loss 0.144688 Objective Loss 0.144688 LR 0.000250 Time 0.020497 -2022-12-06 11:33:20,153 - Epoch: [158][ 590/ 1200] Overall Loss 0.144741 Objective Loss 0.144741 LR 0.000250 Time 0.020472 -2022-12-06 11:33:20,344 - Epoch: [158][ 600/ 1200] Overall Loss 0.144751 Objective Loss 0.144751 LR 0.000250 Time 0.020448 -2022-12-06 11:33:20,535 - Epoch: [158][ 610/ 1200] Overall Loss 0.144767 Objective Loss 0.144767 LR 0.000250 Time 0.020424 -2022-12-06 11:33:20,726 - Epoch: [158][ 620/ 1200] Overall Loss 0.144735 Objective Loss 0.144735 LR 0.000250 Time 0.020402 -2022-12-06 11:33:20,917 - Epoch: [158][ 630/ 1200] Overall Loss 0.144723 Objective Loss 0.144723 LR 0.000250 Time 0.020380 -2022-12-06 11:33:21,108 - Epoch: [158][ 640/ 1200] Overall Loss 0.144653 Objective Loss 0.144653 LR 0.000250 Time 0.020359 -2022-12-06 11:33:21,298 - Epoch: [158][ 650/ 1200] Overall Loss 0.144839 Objective Loss 0.144839 LR 0.000250 Time 0.020339 -2022-12-06 11:33:21,489 - Epoch: [158][ 660/ 1200] Overall Loss 0.144905 Objective Loss 0.144905 LR 0.000250 Time 0.020319 -2022-12-06 11:33:21,680 - Epoch: [158][ 670/ 1200] Overall Loss 0.144849 Objective Loss 0.144849 LR 0.000250 Time 0.020300 -2022-12-06 11:33:21,870 - Epoch: [158][ 680/ 1200] Overall Loss 0.144859 Objective Loss 0.144859 LR 0.000250 Time 0.020281 -2022-12-06 11:33:22,061 - Epoch: [158][ 690/ 1200] Overall Loss 0.144832 Objective Loss 0.144832 LR 0.000250 Time 0.020263 -2022-12-06 11:33:22,251 - Epoch: [158][ 700/ 1200] Overall Loss 0.145015 Objective Loss 0.145015 LR 0.000250 Time 0.020244 -2022-12-06 11:33:22,442 - Epoch: [158][ 710/ 1200] Overall Loss 0.144887 Objective Loss 0.144887 LR 0.000250 Time 0.020226 -2022-12-06 11:33:22,633 - Epoch: [158][ 720/ 1200] Overall Loss 0.145184 Objective Loss 0.145184 LR 0.000250 Time 0.020210 -2022-12-06 11:33:22,824 - Epoch: [158][ 730/ 1200] Overall Loss 0.145453 Objective Loss 0.145453 LR 0.000250 Time 0.020194 -2022-12-06 11:33:23,014 - Epoch: [158][ 740/ 1200] Overall Loss 0.145216 Objective Loss 0.145216 LR 0.000250 Time 0.020177 -2022-12-06 11:33:23,205 - Epoch: [158][ 750/ 1200] Overall Loss 0.145315 Objective Loss 0.145315 LR 0.000250 Time 0.020162 -2022-12-06 11:33:23,395 - Epoch: [158][ 760/ 1200] Overall Loss 0.145153 Objective Loss 0.145153 LR 0.000250 Time 0.020146 -2022-12-06 11:33:23,585 - Epoch: [158][ 770/ 1200] Overall Loss 0.145238 Objective Loss 0.145238 LR 0.000250 Time 0.020130 -2022-12-06 11:33:23,775 - Epoch: [158][ 780/ 1200] Overall Loss 0.145359 Objective Loss 0.145359 LR 0.000250 Time 0.020115 -2022-12-06 11:33:23,966 - Epoch: [158][ 790/ 1200] Overall Loss 0.145777 Objective Loss 0.145777 LR 0.000250 Time 0.020102 -2022-12-06 11:33:24,157 - Epoch: [158][ 800/ 1200] Overall Loss 0.145713 Objective Loss 0.145713 LR 0.000250 Time 0.020089 -2022-12-06 11:33:24,348 - Epoch: [158][ 810/ 1200] Overall Loss 0.145827 Objective Loss 0.145827 LR 0.000250 Time 0.020075 -2022-12-06 11:33:24,538 - Epoch: [158][ 820/ 1200] Overall Loss 0.145920 Objective Loss 0.145920 LR 0.000250 Time 0.020062 -2022-12-06 11:33:24,729 - Epoch: [158][ 830/ 1200] Overall Loss 0.146031 Objective Loss 0.146031 LR 0.000250 Time 0.020049 -2022-12-06 11:33:24,919 - Epoch: [158][ 840/ 1200] Overall Loss 0.145944 Objective Loss 0.145944 LR 0.000250 Time 0.020037 -2022-12-06 11:33:25,110 - Epoch: [158][ 850/ 1200] Overall Loss 0.146037 Objective Loss 0.146037 LR 0.000250 Time 0.020025 -2022-12-06 11:33:25,300 - Epoch: [158][ 860/ 1200] Overall Loss 0.146136 Objective Loss 0.146136 LR 0.000250 Time 0.020012 -2022-12-06 11:33:25,490 - Epoch: [158][ 870/ 1200] Overall Loss 0.146159 Objective Loss 0.146159 LR 0.000250 Time 0.020001 -2022-12-06 11:33:25,680 - Epoch: [158][ 880/ 1200] Overall Loss 0.146137 Objective Loss 0.146137 LR 0.000250 Time 0.019989 -2022-12-06 11:33:25,871 - Epoch: [158][ 890/ 1200] Overall Loss 0.146414 Objective Loss 0.146414 LR 0.000250 Time 0.019978 -2022-12-06 11:33:26,062 - Epoch: [158][ 900/ 1200] Overall Loss 0.146744 Objective Loss 0.146744 LR 0.000250 Time 0.019967 -2022-12-06 11:33:26,253 - Epoch: [158][ 910/ 1200] Overall Loss 0.146694 Objective Loss 0.146694 LR 0.000250 Time 0.019957 -2022-12-06 11:33:26,444 - Epoch: [158][ 920/ 1200] Overall Loss 0.146693 Objective Loss 0.146693 LR 0.000250 Time 0.019947 -2022-12-06 11:33:26,635 - Epoch: [158][ 930/ 1200] Overall Loss 0.146883 Objective Loss 0.146883 LR 0.000250 Time 0.019937 -2022-12-06 11:33:26,825 - Epoch: [158][ 940/ 1200] Overall Loss 0.147019 Objective Loss 0.147019 LR 0.000250 Time 0.019927 -2022-12-06 11:33:27,016 - Epoch: [158][ 950/ 1200] Overall Loss 0.147047 Objective Loss 0.147047 LR 0.000250 Time 0.019917 -2022-12-06 11:33:27,206 - Epoch: [158][ 960/ 1200] Overall Loss 0.147127 Objective Loss 0.147127 LR 0.000250 Time 0.019908 -2022-12-06 11:33:27,398 - Epoch: [158][ 970/ 1200] Overall Loss 0.147156 Objective Loss 0.147156 LR 0.000250 Time 0.019899 -2022-12-06 11:33:27,589 - Epoch: [158][ 980/ 1200] Overall Loss 0.147164 Objective Loss 0.147164 LR 0.000250 Time 0.019891 -2022-12-06 11:33:27,780 - Epoch: [158][ 990/ 1200] Overall Loss 0.147273 Objective Loss 0.147273 LR 0.000250 Time 0.019882 -2022-12-06 11:33:27,971 - Epoch: [158][ 1000/ 1200] Overall Loss 0.147202 Objective Loss 0.147202 LR 0.000250 Time 0.019874 -2022-12-06 11:33:28,162 - Epoch: [158][ 1010/ 1200] Overall Loss 0.147187 Objective Loss 0.147187 LR 0.000250 Time 0.019866 -2022-12-06 11:33:28,352 - Epoch: [158][ 1020/ 1200] Overall Loss 0.147069 Objective Loss 0.147069 LR 0.000250 Time 0.019857 -2022-12-06 11:33:28,544 - Epoch: [158][ 1030/ 1200] Overall Loss 0.147312 Objective Loss 0.147312 LR 0.000250 Time 0.019850 -2022-12-06 11:33:28,734 - Epoch: [158][ 1040/ 1200] Overall Loss 0.147277 Objective Loss 0.147277 LR 0.000250 Time 0.019841 -2022-12-06 11:33:28,924 - Epoch: [158][ 1050/ 1200] Overall Loss 0.147400 Objective Loss 0.147400 LR 0.000250 Time 0.019833 -2022-12-06 11:33:29,115 - Epoch: [158][ 1060/ 1200] Overall Loss 0.147485 Objective Loss 0.147485 LR 0.000250 Time 0.019825 -2022-12-06 11:33:29,306 - Epoch: [158][ 1070/ 1200] Overall Loss 0.147463 Objective Loss 0.147463 LR 0.000250 Time 0.019818 -2022-12-06 11:33:29,496 - Epoch: [158][ 1080/ 1200] Overall Loss 0.147583 Objective Loss 0.147583 LR 0.000250 Time 0.019810 -2022-12-06 11:33:29,688 - Epoch: [158][ 1090/ 1200] Overall Loss 0.147681 Objective Loss 0.147681 LR 0.000250 Time 0.019803 -2022-12-06 11:33:29,878 - Epoch: [158][ 1100/ 1200] Overall Loss 0.147567 Objective Loss 0.147567 LR 0.000250 Time 0.019796 -2022-12-06 11:33:30,069 - Epoch: [158][ 1110/ 1200] Overall Loss 0.147753 Objective Loss 0.147753 LR 0.000250 Time 0.019789 -2022-12-06 11:33:30,259 - Epoch: [158][ 1120/ 1200] Overall Loss 0.147838 Objective Loss 0.147838 LR 0.000250 Time 0.019782 -2022-12-06 11:33:30,450 - Epoch: [158][ 1130/ 1200] Overall Loss 0.147818 Objective Loss 0.147818 LR 0.000250 Time 0.019775 -2022-12-06 11:33:30,640 - Epoch: [158][ 1140/ 1200] Overall Loss 0.147759 Objective Loss 0.147759 LR 0.000250 Time 0.019768 -2022-12-06 11:33:30,832 - Epoch: [158][ 1150/ 1200] Overall Loss 0.147811 Objective Loss 0.147811 LR 0.000250 Time 0.019762 -2022-12-06 11:33:31,022 - Epoch: [158][ 1160/ 1200] Overall Loss 0.147711 Objective Loss 0.147711 LR 0.000250 Time 0.019756 -2022-12-06 11:33:31,213 - Epoch: [158][ 1170/ 1200] Overall Loss 0.147725 Objective Loss 0.147725 LR 0.000250 Time 0.019750 -2022-12-06 11:33:31,403 - Epoch: [158][ 1180/ 1200] Overall Loss 0.147784 Objective Loss 0.147784 LR 0.000250 Time 0.019743 -2022-12-06 11:33:31,594 - Epoch: [158][ 1190/ 1200] Overall Loss 0.147902 Objective Loss 0.147902 LR 0.000250 Time 0.019736 -2022-12-06 11:33:31,823 - Epoch: [158][ 1200/ 1200] Overall Loss 0.148037 Objective Loss 0.148037 Top1 89.330544 Top5 99.163180 LR 0.000250 Time 0.019763 -2022-12-06 11:33:31,911 - --- validate (epoch=158)----------- -2022-12-06 11:33:31,912 - 34129 samples (256 per mini-batch) -2022-12-06 11:33:32,369 - Epoch: [158][ 10/ 134] Loss 0.213474 Top1 88.398438 Top5 98.671875 -2022-12-06 11:33:32,502 - Epoch: [158][ 20/ 134] Loss 0.226261 Top1 87.890625 Top5 98.750000 -2022-12-06 11:33:32,637 - Epoch: [158][ 30/ 134] Loss 0.226939 Top1 87.786458 Top5 98.723958 -2022-12-06 11:33:32,771 - Epoch: [158][ 40/ 134] Loss 0.221283 Top1 87.998047 Top5 98.730469 -2022-12-06 11:33:32,920 - Epoch: [158][ 50/ 134] Loss 0.226712 Top1 87.929688 Top5 98.679688 -2022-12-06 11:33:33,055 - Epoch: [158][ 60/ 134] Loss 0.228465 Top1 87.864583 Top5 98.593750 -2022-12-06 11:33:33,201 - Epoch: [158][ 70/ 134] Loss 0.229265 Top1 87.812500 Top5 98.549107 -2022-12-06 11:33:33,342 - Epoch: [158][ 80/ 134] Loss 0.227751 Top1 87.817383 Top5 98.515625 -2022-12-06 11:33:33,472 - Epoch: [158][ 90/ 134] Loss 0.230257 Top1 87.894965 Top5 98.519965 -2022-12-06 11:33:33,602 - Epoch: [158][ 100/ 134] Loss 0.234013 Top1 87.875000 Top5 98.523438 -2022-12-06 11:33:33,732 - Epoch: [158][ 110/ 134] Loss 0.232330 Top1 87.851562 Top5 98.536932 -2022-12-06 11:33:33,866 - Epoch: [158][ 120/ 134] Loss 0.234912 Top1 87.825521 Top5 98.551432 -2022-12-06 11:33:33,999 - Epoch: [158][ 130/ 134] Loss 0.235031 Top1 87.854567 Top5 98.548678 -2022-12-06 11:33:34,036 - Epoch: [158][ 134/ 134] Loss 0.235358 Top1 87.819743 Top5 98.564271 -2022-12-06 11:33:34,126 - ==> Top1: 87.820 Top5: 98.564 Loss: 0.235 - -2022-12-06 11:33:34,127 - ==> Confusion: -[[ 915 0 1 1 8 7 1 0 4 44 0 2 1 1 6 1 1 1 0 0 2] - [ 1 943 1 2 8 21 2 9 1 1 2 4 0 0 0 3 4 2 11 2 10] - [ 2 3 1010 11 4 2 17 7 0 4 7 3 0 1 3 1 3 3 4 5 13] - [ 3 0 15 953 0 2 1 0 0 1 7 0 5 3 10 0 1 3 10 0 6] - [ 10 4 1 0 963 2 1 2 0 5 1 2 1 2 9 4 6 2 2 0 3] - [ 0 8 0 3 7 987 1 19 2 4 0 9 7 8 3 2 2 1 0 5 1] - [ 2 2 13 0 2 2 1077 2 0 0 0 1 1 0 0 2 1 1 3 6 3] - [ 0 6 4 2 4 25 6 960 0 0 0 5 0 0 0 0 0 0 22 14 6] - [ 10 2 0 0 0 2 1 0 968 41 9 1 1 10 9 1 2 1 3 2 1] - [ 50 0 0 0 6 3 0 2 11 904 1 2 0 12 2 1 1 1 1 0 4] - [ 1 1 2 7 2 0 2 2 7 1 962 0 1 10 5 1 2 0 4 1 8] - [ 3 0 1 0 0 7 5 3 0 0 0 977 24 4 1 7 2 4 1 8 4] - [ 2 1 1 4 0 2 1 1 0 0 0 18 913 1 1 6 3 5 1 2 7] - [ 1 1 1 0 1 8 0 3 10 7 3 5 5 957 2 2 6 1 0 2 8] - [ 10 1 1 7 4 4 0 0 15 7 1 3 3 3 1057 0 1 1 7 1 4] - [ 1 0 1 0 1 0 3 0 2 1 0 7 9 1 0 999 4 7 0 2 5] - [ 3 0 0 0 3 1 1 0 0 1 0 3 3 1 1 8 1038 1 0 2 6] - [ 2 0 2 1 0 1 0 1 0 5 0 7 17 0 2 18 0 975 0 1 4] - [ 4 2 3 6 1 3 0 20 2 1 3 2 4 0 6 0 0 2 944 1 4] - [ 3 2 3 1 0 4 4 6 0 1 2 16 7 3 1 4 3 3 2 1011 4] - [ 113 193 153 101 112 147 74 119 74 82 142 80 301 244 146 118 147 68 166 191 10455]] - -2022-12-06 11:33:34,699 - ==> Best [Top1: 88.010 Top5: 98.500 Sparsity:0.00 Params: 5376 on epoch: 149] -2022-12-06 11:33:34,699 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:33:34,705 - - -2022-12-06 11:33:34,705 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:33:35,782 - Epoch: [159][ 10/ 1200] Overall Loss 0.130732 Objective Loss 0.130732 LR 0.000250 Time 0.107617 -2022-12-06 11:33:35,984 - Epoch: [159][ 20/ 1200] Overall Loss 0.131304 Objective Loss 0.131304 LR 0.000250 Time 0.063891 -2022-12-06 11:33:36,177 - Epoch: [159][ 30/ 1200] Overall Loss 0.139343 Objective Loss 0.139343 LR 0.000250 Time 0.048996 -2022-12-06 11:33:36,370 - Epoch: [159][ 40/ 1200] Overall Loss 0.142070 Objective Loss 0.142070 LR 0.000250 Time 0.041557 -2022-12-06 11:33:36,562 - Epoch: [159][ 50/ 1200] Overall Loss 0.144776 Objective Loss 0.144776 LR 0.000250 Time 0.037075 -2022-12-06 11:33:36,754 - Epoch: [159][ 60/ 1200] Overall Loss 0.145485 Objective Loss 0.145485 LR 0.000250 Time 0.034089 -2022-12-06 11:33:36,946 - Epoch: [159][ 70/ 1200] Overall Loss 0.146757 Objective Loss 0.146757 LR 0.000250 Time 0.031957 -2022-12-06 11:33:37,139 - Epoch: [159][ 80/ 1200] Overall Loss 0.145697 Objective Loss 0.145697 LR 0.000250 Time 0.030364 -2022-12-06 11:33:37,330 - Epoch: [159][ 90/ 1200] Overall Loss 0.146462 Objective Loss 0.146462 LR 0.000250 Time 0.029110 -2022-12-06 11:33:37,523 - Epoch: [159][ 100/ 1200] Overall Loss 0.146601 Objective Loss 0.146601 LR 0.000250 Time 0.028120 -2022-12-06 11:33:37,715 - Epoch: [159][ 110/ 1200] Overall Loss 0.146516 Objective Loss 0.146516 LR 0.000250 Time 0.027308 -2022-12-06 11:33:37,908 - Epoch: [159][ 120/ 1200] Overall Loss 0.146925 Objective Loss 0.146925 LR 0.000250 Time 0.026632 -2022-12-06 11:33:38,101 - Epoch: [159][ 130/ 1200] Overall Loss 0.146419 Objective Loss 0.146419 LR 0.000250 Time 0.026061 -2022-12-06 11:33:38,292 - Epoch: [159][ 140/ 1200] Overall Loss 0.146733 Objective Loss 0.146733 LR 0.000250 Time 0.025565 -2022-12-06 11:33:38,485 - Epoch: [159][ 150/ 1200] Overall Loss 0.146902 Objective Loss 0.146902 LR 0.000250 Time 0.025143 -2022-12-06 11:33:38,678 - Epoch: [159][ 160/ 1200] Overall Loss 0.146472 Objective Loss 0.146472 LR 0.000250 Time 0.024771 -2022-12-06 11:33:38,870 - Epoch: [159][ 170/ 1200] Overall Loss 0.146254 Objective Loss 0.146254 LR 0.000250 Time 0.024442 -2022-12-06 11:33:39,062 - Epoch: [159][ 180/ 1200] Overall Loss 0.147096 Objective Loss 0.147096 LR 0.000250 Time 0.024149 -2022-12-06 11:33:39,255 - Epoch: [159][ 190/ 1200] Overall Loss 0.147662 Objective Loss 0.147662 LR 0.000250 Time 0.023888 -2022-12-06 11:33:39,447 - Epoch: [159][ 200/ 1200] Overall Loss 0.148027 Objective Loss 0.148027 LR 0.000250 Time 0.023654 -2022-12-06 11:33:39,639 - Epoch: [159][ 210/ 1200] Overall Loss 0.147587 Objective Loss 0.147587 LR 0.000250 Time 0.023440 -2022-12-06 11:33:39,832 - Epoch: [159][ 220/ 1200] Overall Loss 0.147908 Objective Loss 0.147908 LR 0.000250 Time 0.023246 -2022-12-06 11:33:40,024 - Epoch: [159][ 230/ 1200] Overall Loss 0.148065 Objective Loss 0.148065 LR 0.000250 Time 0.023070 -2022-12-06 11:33:40,217 - Epoch: [159][ 240/ 1200] Overall Loss 0.147505 Objective Loss 0.147505 LR 0.000250 Time 0.022909 -2022-12-06 11:33:40,409 - Epoch: [159][ 250/ 1200] Overall Loss 0.147631 Objective Loss 0.147631 LR 0.000250 Time 0.022761 -2022-12-06 11:33:40,602 - Epoch: [159][ 260/ 1200] Overall Loss 0.149077 Objective Loss 0.149077 LR 0.000250 Time 0.022624 -2022-12-06 11:33:40,795 - Epoch: [159][ 270/ 1200] Overall Loss 0.148691 Objective Loss 0.148691 LR 0.000250 Time 0.022499 -2022-12-06 11:33:40,987 - Epoch: [159][ 280/ 1200] Overall Loss 0.148506 Objective Loss 0.148506 LR 0.000250 Time 0.022380 -2022-12-06 11:33:41,180 - Epoch: [159][ 290/ 1200] Overall Loss 0.148443 Objective Loss 0.148443 LR 0.000250 Time 0.022270 -2022-12-06 11:33:41,372 - Epoch: [159][ 300/ 1200] Overall Loss 0.149510 Objective Loss 0.149510 LR 0.000250 Time 0.022167 -2022-12-06 11:33:41,564 - Epoch: [159][ 310/ 1200] Overall Loss 0.149782 Objective Loss 0.149782 LR 0.000250 Time 0.022070 -2022-12-06 11:33:41,757 - Epoch: [159][ 320/ 1200] Overall Loss 0.149832 Objective Loss 0.149832 LR 0.000250 Time 0.021979 -2022-12-06 11:33:41,949 - Epoch: [159][ 330/ 1200] Overall Loss 0.149702 Objective Loss 0.149702 LR 0.000250 Time 0.021895 -2022-12-06 11:33:42,142 - Epoch: [159][ 340/ 1200] Overall Loss 0.149744 Objective Loss 0.149744 LR 0.000250 Time 0.021816 -2022-12-06 11:33:42,334 - Epoch: [159][ 350/ 1200] Overall Loss 0.149729 Objective Loss 0.149729 LR 0.000250 Time 0.021742 -2022-12-06 11:33:42,526 - Epoch: [159][ 360/ 1200] Overall Loss 0.149860 Objective Loss 0.149860 LR 0.000250 Time 0.021668 -2022-12-06 11:33:42,719 - Epoch: [159][ 370/ 1200] Overall Loss 0.149838 Objective Loss 0.149838 LR 0.000250 Time 0.021602 -2022-12-06 11:33:42,910 - Epoch: [159][ 380/ 1200] Overall Loss 0.149271 Objective Loss 0.149271 LR 0.000250 Time 0.021536 -2022-12-06 11:33:43,102 - Epoch: [159][ 390/ 1200] Overall Loss 0.149626 Objective Loss 0.149626 LR 0.000250 Time 0.021475 -2022-12-06 11:33:43,294 - Epoch: [159][ 400/ 1200] Overall Loss 0.149874 Objective Loss 0.149874 LR 0.000250 Time 0.021417 -2022-12-06 11:33:43,487 - Epoch: [159][ 410/ 1200] Overall Loss 0.149936 Objective Loss 0.149936 LR 0.000250 Time 0.021362 -2022-12-06 11:33:43,679 - Epoch: [159][ 420/ 1200] Overall Loss 0.149532 Objective Loss 0.149532 LR 0.000250 Time 0.021310 -2022-12-06 11:33:43,871 - Epoch: [159][ 430/ 1200] Overall Loss 0.149293 Objective Loss 0.149293 LR 0.000250 Time 0.021260 -2022-12-06 11:33:44,063 - Epoch: [159][ 440/ 1200] Overall Loss 0.149396 Objective Loss 0.149396 LR 0.000250 Time 0.021212 -2022-12-06 11:33:44,256 - Epoch: [159][ 450/ 1200] Overall Loss 0.149335 Objective Loss 0.149335 LR 0.000250 Time 0.021169 -2022-12-06 11:33:44,448 - Epoch: [159][ 460/ 1200] Overall Loss 0.149385 Objective Loss 0.149385 LR 0.000250 Time 0.021125 -2022-12-06 11:33:44,640 - Epoch: [159][ 470/ 1200] Overall Loss 0.149442 Objective Loss 0.149442 LR 0.000250 Time 0.021083 -2022-12-06 11:33:44,833 - Epoch: [159][ 480/ 1200] Overall Loss 0.149807 Objective Loss 0.149807 LR 0.000250 Time 0.021044 -2022-12-06 11:33:45,025 - Epoch: [159][ 490/ 1200] Overall Loss 0.149899 Objective Loss 0.149899 LR 0.000250 Time 0.021006 -2022-12-06 11:33:45,218 - Epoch: [159][ 500/ 1200] Overall Loss 0.149923 Objective Loss 0.149923 LR 0.000250 Time 0.020969 -2022-12-06 11:33:45,410 - Epoch: [159][ 510/ 1200] Overall Loss 0.149661 Objective Loss 0.149661 LR 0.000250 Time 0.020935 -2022-12-06 11:33:45,603 - Epoch: [159][ 520/ 1200] Overall Loss 0.149503 Objective Loss 0.149503 LR 0.000250 Time 0.020901 -2022-12-06 11:33:45,795 - Epoch: [159][ 530/ 1200] Overall Loss 0.149539 Objective Loss 0.149539 LR 0.000250 Time 0.020868 -2022-12-06 11:33:45,987 - Epoch: [159][ 540/ 1200] Overall Loss 0.149560 Objective Loss 0.149560 LR 0.000250 Time 0.020836 -2022-12-06 11:33:46,180 - Epoch: [159][ 550/ 1200] Overall Loss 0.149239 Objective Loss 0.149239 LR 0.000250 Time 0.020808 -2022-12-06 11:33:46,374 - Epoch: [159][ 560/ 1200] Overall Loss 0.149046 Objective Loss 0.149046 LR 0.000250 Time 0.020781 -2022-12-06 11:33:46,566 - Epoch: [159][ 570/ 1200] Overall Loss 0.148812 Objective Loss 0.148812 LR 0.000250 Time 0.020752 -2022-12-06 11:33:46,758 - Epoch: [159][ 580/ 1200] Overall Loss 0.149338 Objective Loss 0.149338 LR 0.000250 Time 0.020726 -2022-12-06 11:33:46,951 - Epoch: [159][ 590/ 1200] Overall Loss 0.149286 Objective Loss 0.149286 LR 0.000250 Time 0.020699 -2022-12-06 11:33:47,143 - Epoch: [159][ 600/ 1200] Overall Loss 0.149486 Objective Loss 0.149486 LR 0.000250 Time 0.020674 -2022-12-06 11:33:47,336 - Epoch: [159][ 610/ 1200] Overall Loss 0.149265 Objective Loss 0.149265 LR 0.000250 Time 0.020650 -2022-12-06 11:33:47,528 - Epoch: [159][ 620/ 1200] Overall Loss 0.149145 Objective Loss 0.149145 LR 0.000250 Time 0.020627 -2022-12-06 11:33:47,721 - Epoch: [159][ 630/ 1200] Overall Loss 0.149339 Objective Loss 0.149339 LR 0.000250 Time 0.020604 -2022-12-06 11:33:47,914 - Epoch: [159][ 640/ 1200] Overall Loss 0.149233 Objective Loss 0.149233 LR 0.000250 Time 0.020583 -2022-12-06 11:33:48,107 - Epoch: [159][ 650/ 1200] Overall Loss 0.149048 Objective Loss 0.149048 LR 0.000250 Time 0.020562 -2022-12-06 11:33:48,299 - Epoch: [159][ 660/ 1200] Overall Loss 0.148860 Objective Loss 0.148860 LR 0.000250 Time 0.020541 -2022-12-06 11:33:48,492 - Epoch: [159][ 670/ 1200] Overall Loss 0.148846 Objective Loss 0.148846 LR 0.000250 Time 0.020521 -2022-12-06 11:33:48,684 - Epoch: [159][ 680/ 1200] Overall Loss 0.148964 Objective Loss 0.148964 LR 0.000250 Time 0.020501 -2022-12-06 11:33:48,877 - Epoch: [159][ 690/ 1200] Overall Loss 0.149197 Objective Loss 0.149197 LR 0.000250 Time 0.020483 -2022-12-06 11:33:49,070 - Epoch: [159][ 700/ 1200] Overall Loss 0.149161 Objective Loss 0.149161 LR 0.000250 Time 0.020465 -2022-12-06 11:33:49,262 - Epoch: [159][ 710/ 1200] Overall Loss 0.149287 Objective Loss 0.149287 LR 0.000250 Time 0.020448 -2022-12-06 11:33:49,455 - Epoch: [159][ 720/ 1200] Overall Loss 0.149263 Objective Loss 0.149263 LR 0.000250 Time 0.020430 -2022-12-06 11:33:49,646 - Epoch: [159][ 730/ 1200] Overall Loss 0.149184 Objective Loss 0.149184 LR 0.000250 Time 0.020411 -2022-12-06 11:33:49,839 - Epoch: [159][ 740/ 1200] Overall Loss 0.148982 Objective Loss 0.148982 LR 0.000250 Time 0.020396 -2022-12-06 11:33:50,031 - Epoch: [159][ 750/ 1200] Overall Loss 0.148952 Objective Loss 0.148952 LR 0.000250 Time 0.020379 -2022-12-06 11:33:50,225 - Epoch: [159][ 760/ 1200] Overall Loss 0.148928 Objective Loss 0.148928 LR 0.000250 Time 0.020365 -2022-12-06 11:33:50,417 - Epoch: [159][ 770/ 1200] Overall Loss 0.149281 Objective Loss 0.149281 LR 0.000250 Time 0.020350 -2022-12-06 11:33:50,610 - Epoch: [159][ 780/ 1200] Overall Loss 0.149087 Objective Loss 0.149087 LR 0.000250 Time 0.020335 -2022-12-06 11:33:50,802 - Epoch: [159][ 790/ 1200] Overall Loss 0.148837 Objective Loss 0.148837 LR 0.000250 Time 0.020321 -2022-12-06 11:33:50,994 - Epoch: [159][ 800/ 1200] Overall Loss 0.148748 Objective Loss 0.148748 LR 0.000250 Time 0.020306 -2022-12-06 11:33:51,187 - Epoch: [159][ 810/ 1200] Overall Loss 0.148642 Objective Loss 0.148642 LR 0.000250 Time 0.020292 -2022-12-06 11:33:51,379 - Epoch: [159][ 820/ 1200] Overall Loss 0.148578 Objective Loss 0.148578 LR 0.000250 Time 0.020278 -2022-12-06 11:33:51,571 - Epoch: [159][ 830/ 1200] Overall Loss 0.148367 Objective Loss 0.148367 LR 0.000250 Time 0.020265 -2022-12-06 11:33:51,763 - Epoch: [159][ 840/ 1200] Overall Loss 0.148168 Objective Loss 0.148168 LR 0.000250 Time 0.020252 -2022-12-06 11:33:51,957 - Epoch: [159][ 850/ 1200] Overall Loss 0.148260 Objective Loss 0.148260 LR 0.000250 Time 0.020240 -2022-12-06 11:33:52,149 - Epoch: [159][ 860/ 1200] Overall Loss 0.148456 Objective Loss 0.148456 LR 0.000250 Time 0.020228 -2022-12-06 11:33:52,342 - Epoch: [159][ 870/ 1200] Overall Loss 0.148411 Objective Loss 0.148411 LR 0.000250 Time 0.020216 -2022-12-06 11:33:52,534 - Epoch: [159][ 880/ 1200] Overall Loss 0.148425 Objective Loss 0.148425 LR 0.000250 Time 0.020204 -2022-12-06 11:33:52,726 - Epoch: [159][ 890/ 1200] Overall Loss 0.148517 Objective Loss 0.148517 LR 0.000250 Time 0.020193 -2022-12-06 11:33:52,919 - Epoch: [159][ 900/ 1200] Overall Loss 0.148534 Objective Loss 0.148534 LR 0.000250 Time 0.020182 -2022-12-06 11:33:53,111 - Epoch: [159][ 910/ 1200] Overall Loss 0.148592 Objective Loss 0.148592 LR 0.000250 Time 0.020171 -2022-12-06 11:33:53,304 - Epoch: [159][ 920/ 1200] Overall Loss 0.148609 Objective Loss 0.148609 LR 0.000250 Time 0.020160 -2022-12-06 11:33:53,497 - Epoch: [159][ 930/ 1200] Overall Loss 0.148496 Objective Loss 0.148496 LR 0.000250 Time 0.020150 -2022-12-06 11:33:53,689 - Epoch: [159][ 940/ 1200] Overall Loss 0.148369 Objective Loss 0.148369 LR 0.000250 Time 0.020140 -2022-12-06 11:33:53,882 - Epoch: [159][ 950/ 1200] Overall Loss 0.148377 Objective Loss 0.148377 LR 0.000250 Time 0.020130 -2022-12-06 11:33:54,074 - Epoch: [159][ 960/ 1200] Overall Loss 0.148498 Objective Loss 0.148498 LR 0.000250 Time 0.020121 -2022-12-06 11:33:54,267 - Epoch: [159][ 970/ 1200] Overall Loss 0.148385 Objective Loss 0.148385 LR 0.000250 Time 0.020111 -2022-12-06 11:33:54,459 - Epoch: [159][ 980/ 1200] Overall Loss 0.148380 Objective Loss 0.148380 LR 0.000250 Time 0.020102 -2022-12-06 11:33:54,652 - Epoch: [159][ 990/ 1200] Overall Loss 0.148352 Objective Loss 0.148352 LR 0.000250 Time 0.020093 -2022-12-06 11:33:54,845 - Epoch: [159][ 1000/ 1200] Overall Loss 0.148431 Objective Loss 0.148431 LR 0.000250 Time 0.020084 -2022-12-06 11:33:55,037 - Epoch: [159][ 1010/ 1200] Overall Loss 0.148574 Objective Loss 0.148574 LR 0.000250 Time 0.020075 -2022-12-06 11:33:55,230 - Epoch: [159][ 1020/ 1200] Overall Loss 0.148478 Objective Loss 0.148478 LR 0.000250 Time 0.020066 -2022-12-06 11:33:55,422 - Epoch: [159][ 1030/ 1200] Overall Loss 0.148480 Objective Loss 0.148480 LR 0.000250 Time 0.020058 -2022-12-06 11:33:55,614 - Epoch: [159][ 1040/ 1200] Overall Loss 0.148544 Objective Loss 0.148544 LR 0.000250 Time 0.020049 -2022-12-06 11:33:55,806 - Epoch: [159][ 1050/ 1200] Overall Loss 0.148462 Objective Loss 0.148462 LR 0.000250 Time 0.020041 -2022-12-06 11:33:55,999 - Epoch: [159][ 1060/ 1200] Overall Loss 0.148310 Objective Loss 0.148310 LR 0.000250 Time 0.020033 -2022-12-06 11:33:56,192 - Epoch: [159][ 1070/ 1200] Overall Loss 0.148451 Objective Loss 0.148451 LR 0.000250 Time 0.020026 -2022-12-06 11:33:56,385 - Epoch: [159][ 1080/ 1200] Overall Loss 0.148537 Objective Loss 0.148537 LR 0.000250 Time 0.020018 -2022-12-06 11:33:56,578 - Epoch: [159][ 1090/ 1200] Overall Loss 0.148686 Objective Loss 0.148686 LR 0.000250 Time 0.020011 -2022-12-06 11:33:56,770 - Epoch: [159][ 1100/ 1200] Overall Loss 0.148742 Objective Loss 0.148742 LR 0.000250 Time 0.020004 -2022-12-06 11:33:56,963 - Epoch: [159][ 1110/ 1200] Overall Loss 0.148820 Objective Loss 0.148820 LR 0.000250 Time 0.019996 -2022-12-06 11:33:57,155 - Epoch: [159][ 1120/ 1200] Overall Loss 0.149014 Objective Loss 0.149014 LR 0.000250 Time 0.019989 -2022-12-06 11:33:57,347 - Epoch: [159][ 1130/ 1200] Overall Loss 0.149060 Objective Loss 0.149060 LR 0.000250 Time 0.019982 -2022-12-06 11:33:57,539 - Epoch: [159][ 1140/ 1200] Overall Loss 0.149235 Objective Loss 0.149235 LR 0.000250 Time 0.019974 -2022-12-06 11:33:57,732 - Epoch: [159][ 1150/ 1200] Overall Loss 0.149171 Objective Loss 0.149171 LR 0.000250 Time 0.019968 -2022-12-06 11:33:57,924 - Epoch: [159][ 1160/ 1200] Overall Loss 0.149045 Objective Loss 0.149045 LR 0.000250 Time 0.019961 -2022-12-06 11:33:58,116 - Epoch: [159][ 1170/ 1200] Overall Loss 0.149081 Objective Loss 0.149081 LR 0.000250 Time 0.019954 -2022-12-06 11:33:58,309 - Epoch: [159][ 1180/ 1200] Overall Loss 0.149072 Objective Loss 0.149072 LR 0.000250 Time 0.019948 -2022-12-06 11:33:58,502 - Epoch: [159][ 1190/ 1200] Overall Loss 0.149177 Objective Loss 0.149177 LR 0.000250 Time 0.019942 -2022-12-06 11:33:58,727 - Epoch: [159][ 1200/ 1200] Overall Loss 0.149211 Objective Loss 0.149211 Top1 88.284519 Top5 98.535565 LR 0.000250 Time 0.019962 -2022-12-06 11:33:58,815 - --- validate (epoch=159)----------- -2022-12-06 11:33:58,815 - 34129 samples (256 per mini-batch) -2022-12-06 11:33:59,270 - Epoch: [159][ 10/ 134] Loss 0.246329 Top1 87.539062 Top5 97.851562 -2022-12-06 11:33:59,407 - Epoch: [159][ 20/ 134] Loss 0.240446 Top1 87.968750 Top5 98.242188 -2022-12-06 11:33:59,542 - Epoch: [159][ 30/ 134] Loss 0.247683 Top1 88.216146 Top5 98.255208 -2022-12-06 11:33:59,678 - Epoch: [159][ 40/ 134] Loss 0.240384 Top1 88.535156 Top5 98.398438 -2022-12-06 11:33:59,812 - Epoch: [159][ 50/ 134] Loss 0.233788 Top1 88.601562 Top5 98.500000 -2022-12-06 11:33:59,947 - Epoch: [159][ 60/ 134] Loss 0.233765 Top1 88.509115 Top5 98.496094 -2022-12-06 11:34:00,082 - Epoch: [159][ 70/ 134] Loss 0.233947 Top1 88.498884 Top5 98.487723 -2022-12-06 11:34:00,216 - Epoch: [159][ 80/ 134] Loss 0.235848 Top1 88.505859 Top5 98.491211 -2022-12-06 11:34:00,350 - Epoch: [159][ 90/ 134] Loss 0.235642 Top1 88.489583 Top5 98.506944 -2022-12-06 11:34:00,485 - Epoch: [159][ 100/ 134] Loss 0.236356 Top1 88.417969 Top5 98.507812 -2022-12-06 11:34:00,619 - Epoch: [159][ 110/ 134] Loss 0.232750 Top1 88.398438 Top5 98.547585 -2022-12-06 11:34:00,754 - Epoch: [159][ 120/ 134] Loss 0.229620 Top1 88.421224 Top5 98.583984 -2022-12-06 11:34:00,888 - Epoch: [159][ 130/ 134] Loss 0.230427 Top1 88.335337 Top5 98.572716 -2022-12-06 11:34:00,928 - Epoch: [159][ 134/ 134] Loss 0.230552 Top1 88.367664 Top5 98.584781 -2022-12-06 11:34:01,022 - ==> Top1: 88.368 Top5: 98.585 Loss: 0.231 - -2022-12-06 11:34:01,023 - ==> Confusion: -[[ 909 0 0 2 8 6 0 1 4 49 0 2 1 4 4 1 0 0 0 0 5] - [ 2 949 1 2 9 18 1 11 2 1 2 1 0 1 0 1 6 2 8 3 7] - [ 5 1 1015 11 6 2 11 9 0 4 4 4 3 1 1 1 0 3 2 4 16] - [ 1 1 15 947 0 1 0 0 0 2 10 0 4 0 15 0 1 3 13 0 7] - [ 5 7 1 0 964 2 0 1 1 6 1 2 0 3 9 5 6 3 0 1 3] - [ 0 15 0 2 5 974 2 21 3 3 1 11 1 12 2 2 3 3 0 6 3] - [ 1 4 10 3 1 0 1076 3 0 0 3 0 0 2 0 3 1 1 2 7 1] - [ 1 9 4 2 3 21 4 967 0 1 1 4 1 1 0 0 0 0 16 12 7] - [ 6 1 0 0 0 2 1 0 988 37 10 1 3 7 5 0 1 0 1 1 0] - [ 37 0 0 0 3 5 0 2 16 914 2 2 0 9 3 1 0 0 1 0 6] - [ 0 1 5 1 0 0 1 3 6 1 977 0 1 9 3 1 0 0 1 2 7] - [ 3 1 1 0 1 11 3 3 0 0 0 980 19 2 0 6 4 6 0 9 2] - [ 0 1 0 2 2 2 1 0 0 1 0 23 909 2 2 3 1 8 0 3 9] - [ 0 1 1 0 1 8 0 3 11 11 2 3 4 960 1 0 3 0 0 2 12] - [ 9 2 1 10 3 4 0 0 18 5 0 2 1 3 1063 0 0 0 6 0 3] - [ 1 0 0 3 2 0 2 1 1 0 1 7 7 2 0 992 6 10 0 4 4] - [ 0 0 1 1 3 2 1 1 1 0 0 2 2 2 0 9 1033 0 2 2 10] - [ 2 0 1 0 1 1 0 2 0 3 0 6 13 2 0 12 0 989 0 1 3] - [ 3 2 2 7 0 3 1 22 2 1 5 2 3 0 8 0 0 2 939 2 4] - [ 1 4 1 2 0 4 6 6 0 0 1 11 6 5 0 4 2 3 1 1019 4] - [ 117 183 139 68 91 128 72 120 88 100 135 91 290 239 134 73 143 74 121 226 10594]] - -2022-12-06 11:34:01,602 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:34:01,603 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:34:01,610 - - -2022-12-06 11:34:01,610 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:34:02,537 - Epoch: [160][ 10/ 1200] Overall Loss 0.152444 Objective Loss 0.152444 LR 0.000250 Time 0.092653 -2022-12-06 11:34:02,742 - Epoch: [160][ 20/ 1200] Overall Loss 0.150493 Objective Loss 0.150493 LR 0.000250 Time 0.056522 -2022-12-06 11:34:02,938 - Epoch: [160][ 30/ 1200] Overall Loss 0.151720 Objective Loss 0.151720 LR 0.000250 Time 0.044198 -2022-12-06 11:34:03,136 - Epoch: [160][ 40/ 1200] Overall Loss 0.154015 Objective Loss 0.154015 LR 0.000250 Time 0.038093 -2022-12-06 11:34:03,331 - Epoch: [160][ 50/ 1200] Overall Loss 0.149769 Objective Loss 0.149769 LR 0.000250 Time 0.034364 -2022-12-06 11:34:03,529 - Epoch: [160][ 60/ 1200] Overall Loss 0.149476 Objective Loss 0.149476 LR 0.000250 Time 0.031934 -2022-12-06 11:34:03,724 - Epoch: [160][ 70/ 1200] Overall Loss 0.148707 Objective Loss 0.148707 LR 0.000250 Time 0.030139 -2022-12-06 11:34:03,921 - Epoch: [160][ 80/ 1200] Overall Loss 0.149630 Objective Loss 0.149630 LR 0.000250 Time 0.028838 -2022-12-06 11:34:04,116 - Epoch: [160][ 90/ 1200] Overall Loss 0.149386 Objective Loss 0.149386 LR 0.000250 Time 0.027790 -2022-12-06 11:34:04,314 - Epoch: [160][ 100/ 1200] Overall Loss 0.147566 Objective Loss 0.147566 LR 0.000250 Time 0.026985 -2022-12-06 11:34:04,509 - Epoch: [160][ 110/ 1200] Overall Loss 0.149068 Objective Loss 0.149068 LR 0.000250 Time 0.026296 -2022-12-06 11:34:04,706 - Epoch: [160][ 120/ 1200] Overall Loss 0.148927 Objective Loss 0.148927 LR 0.000250 Time 0.025744 -2022-12-06 11:34:04,900 - Epoch: [160][ 130/ 1200] Overall Loss 0.147966 Objective Loss 0.147966 LR 0.000250 Time 0.025256 -2022-12-06 11:34:05,099 - Epoch: [160][ 140/ 1200] Overall Loss 0.148118 Objective Loss 0.148118 LR 0.000250 Time 0.024863 -2022-12-06 11:34:05,294 - Epoch: [160][ 150/ 1200] Overall Loss 0.147310 Objective Loss 0.147310 LR 0.000250 Time 0.024502 -2022-12-06 11:34:05,492 - Epoch: [160][ 160/ 1200] Overall Loss 0.148193 Objective Loss 0.148193 LR 0.000250 Time 0.024206 -2022-12-06 11:34:05,688 - Epoch: [160][ 170/ 1200] Overall Loss 0.147844 Objective Loss 0.147844 LR 0.000250 Time 0.023933 -2022-12-06 11:34:05,886 - Epoch: [160][ 180/ 1200] Overall Loss 0.147261 Objective Loss 0.147261 LR 0.000250 Time 0.023701 -2022-12-06 11:34:06,081 - Epoch: [160][ 190/ 1200] Overall Loss 0.147383 Objective Loss 0.147383 LR 0.000250 Time 0.023478 -2022-12-06 11:34:06,280 - Epoch: [160][ 200/ 1200] Overall Loss 0.148155 Objective Loss 0.148155 LR 0.000250 Time 0.023295 -2022-12-06 11:34:06,475 - Epoch: [160][ 210/ 1200] Overall Loss 0.149146 Objective Loss 0.149146 LR 0.000250 Time 0.023112 -2022-12-06 11:34:06,673 - Epoch: [160][ 220/ 1200] Overall Loss 0.148263 Objective Loss 0.148263 LR 0.000250 Time 0.022958 -2022-12-06 11:34:06,867 - Epoch: [160][ 230/ 1200] Overall Loss 0.147969 Objective Loss 0.147969 LR 0.000250 Time 0.022803 -2022-12-06 11:34:07,065 - Epoch: [160][ 240/ 1200] Overall Loss 0.148105 Objective Loss 0.148105 LR 0.000250 Time 0.022676 -2022-12-06 11:34:07,260 - Epoch: [160][ 250/ 1200] Overall Loss 0.148020 Objective Loss 0.148020 LR 0.000250 Time 0.022545 -2022-12-06 11:34:07,457 - Epoch: [160][ 260/ 1200] Overall Loss 0.148217 Objective Loss 0.148217 LR 0.000250 Time 0.022435 -2022-12-06 11:34:07,652 - Epoch: [160][ 270/ 1200] Overall Loss 0.148617 Objective Loss 0.148617 LR 0.000250 Time 0.022325 -2022-12-06 11:34:07,851 - Epoch: [160][ 280/ 1200] Overall Loss 0.147908 Objective Loss 0.147908 LR 0.000250 Time 0.022233 -2022-12-06 11:34:08,045 - Epoch: [160][ 290/ 1200] Overall Loss 0.148115 Objective Loss 0.148115 LR 0.000250 Time 0.022136 -2022-12-06 11:34:08,244 - Epoch: [160][ 300/ 1200] Overall Loss 0.147984 Objective Loss 0.147984 LR 0.000250 Time 0.022058 -2022-12-06 11:34:08,439 - Epoch: [160][ 310/ 1200] Overall Loss 0.148027 Objective Loss 0.148027 LR 0.000250 Time 0.021974 -2022-12-06 11:34:08,636 - Epoch: [160][ 320/ 1200] Overall Loss 0.147790 Objective Loss 0.147790 LR 0.000250 Time 0.021903 -2022-12-06 11:34:08,831 - Epoch: [160][ 330/ 1200] Overall Loss 0.148028 Objective Loss 0.148028 LR 0.000250 Time 0.021828 -2022-12-06 11:34:09,028 - Epoch: [160][ 340/ 1200] Overall Loss 0.148009 Objective Loss 0.148009 LR 0.000250 Time 0.021764 -2022-12-06 11:34:09,225 - Epoch: [160][ 350/ 1200] Overall Loss 0.147665 Objective Loss 0.147665 LR 0.000250 Time 0.021701 -2022-12-06 11:34:09,423 - Epoch: [160][ 360/ 1200] Overall Loss 0.148202 Objective Loss 0.148202 LR 0.000250 Time 0.021647 -2022-12-06 11:34:09,617 - Epoch: [160][ 370/ 1200] Overall Loss 0.148498 Objective Loss 0.148498 LR 0.000250 Time 0.021587 -2022-12-06 11:34:09,815 - Epoch: [160][ 380/ 1200] Overall Loss 0.147963 Objective Loss 0.147963 LR 0.000250 Time 0.021539 -2022-12-06 11:34:10,011 - Epoch: [160][ 390/ 1200] Overall Loss 0.148122 Objective Loss 0.148122 LR 0.000250 Time 0.021485 -2022-12-06 11:34:10,209 - Epoch: [160][ 400/ 1200] Overall Loss 0.148259 Objective Loss 0.148259 LR 0.000250 Time 0.021444 -2022-12-06 11:34:10,406 - Epoch: [160][ 410/ 1200] Overall Loss 0.148485 Objective Loss 0.148485 LR 0.000250 Time 0.021398 -2022-12-06 11:34:10,605 - Epoch: [160][ 420/ 1200] Overall Loss 0.148570 Objective Loss 0.148570 LR 0.000250 Time 0.021361 -2022-12-06 11:34:10,800 - Epoch: [160][ 430/ 1200] Overall Loss 0.148260 Objective Loss 0.148260 LR 0.000250 Time 0.021317 -2022-12-06 11:34:10,997 - Epoch: [160][ 440/ 1200] Overall Loss 0.148147 Objective Loss 0.148147 LR 0.000250 Time 0.021280 -2022-12-06 11:34:11,194 - Epoch: [160][ 450/ 1200] Overall Loss 0.147799 Objective Loss 0.147799 LR 0.000250 Time 0.021243 -2022-12-06 11:34:11,392 - Epoch: [160][ 460/ 1200] Overall Loss 0.147833 Objective Loss 0.147833 LR 0.000250 Time 0.021210 -2022-12-06 11:34:11,587 - Epoch: [160][ 470/ 1200] Overall Loss 0.147848 Objective Loss 0.147848 LR 0.000250 Time 0.021174 -2022-12-06 11:34:11,786 - Epoch: [160][ 480/ 1200] Overall Loss 0.148438 Objective Loss 0.148438 LR 0.000250 Time 0.021144 -2022-12-06 11:34:11,981 - Epoch: [160][ 490/ 1200] Overall Loss 0.147975 Objective Loss 0.147975 LR 0.000250 Time 0.021110 -2022-12-06 11:34:12,179 - Epoch: [160][ 500/ 1200] Overall Loss 0.147712 Objective Loss 0.147712 LR 0.000250 Time 0.021083 -2022-12-06 11:34:12,375 - Epoch: [160][ 510/ 1200] Overall Loss 0.147638 Objective Loss 0.147638 LR 0.000250 Time 0.021053 -2022-12-06 11:34:12,573 - Epoch: [160][ 520/ 1200] Overall Loss 0.147445 Objective Loss 0.147445 LR 0.000250 Time 0.021027 -2022-12-06 11:34:12,768 - Epoch: [160][ 530/ 1200] Overall Loss 0.147507 Objective Loss 0.147507 LR 0.000250 Time 0.020997 -2022-12-06 11:34:12,965 - Epoch: [160][ 540/ 1200] Overall Loss 0.147382 Objective Loss 0.147382 LR 0.000250 Time 0.020973 -2022-12-06 11:34:13,160 - Epoch: [160][ 550/ 1200] Overall Loss 0.147187 Objective Loss 0.147187 LR 0.000250 Time 0.020945 -2022-12-06 11:34:13,358 - Epoch: [160][ 560/ 1200] Overall Loss 0.147548 Objective Loss 0.147548 LR 0.000250 Time 0.020923 -2022-12-06 11:34:13,553 - Epoch: [160][ 570/ 1200] Overall Loss 0.147642 Objective Loss 0.147642 LR 0.000250 Time 0.020897 -2022-12-06 11:34:13,751 - Epoch: [160][ 580/ 1200] Overall Loss 0.147553 Objective Loss 0.147553 LR 0.000250 Time 0.020877 -2022-12-06 11:34:13,945 - Epoch: [160][ 590/ 1200] Overall Loss 0.147621 Objective Loss 0.147621 LR 0.000250 Time 0.020852 -2022-12-06 11:34:14,143 - Epoch: [160][ 600/ 1200] Overall Loss 0.147796 Objective Loss 0.147796 LR 0.000250 Time 0.020833 -2022-12-06 11:34:14,338 - Epoch: [160][ 610/ 1200] Overall Loss 0.147771 Objective Loss 0.147771 LR 0.000250 Time 0.020811 -2022-12-06 11:34:14,536 - Epoch: [160][ 620/ 1200] Overall Loss 0.147785 Objective Loss 0.147785 LR 0.000250 Time 0.020794 -2022-12-06 11:34:14,731 - Epoch: [160][ 630/ 1200] Overall Loss 0.147741 Objective Loss 0.147741 LR 0.000250 Time 0.020772 -2022-12-06 11:34:14,929 - Epoch: [160][ 640/ 1200] Overall Loss 0.147577 Objective Loss 0.147577 LR 0.000250 Time 0.020755 -2022-12-06 11:34:15,123 - Epoch: [160][ 650/ 1200] Overall Loss 0.147447 Objective Loss 0.147447 LR 0.000250 Time 0.020735 -2022-12-06 11:34:15,321 - Epoch: [160][ 660/ 1200] Overall Loss 0.147408 Objective Loss 0.147408 LR 0.000250 Time 0.020720 -2022-12-06 11:34:15,516 - Epoch: [160][ 670/ 1200] Overall Loss 0.147325 Objective Loss 0.147325 LR 0.000250 Time 0.020701 -2022-12-06 11:34:15,714 - Epoch: [160][ 680/ 1200] Overall Loss 0.147579 Objective Loss 0.147579 LR 0.000250 Time 0.020686 -2022-12-06 11:34:15,909 - Epoch: [160][ 690/ 1200] Overall Loss 0.147734 Objective Loss 0.147734 LR 0.000250 Time 0.020668 -2022-12-06 11:34:16,107 - Epoch: [160][ 700/ 1200] Overall Loss 0.147860 Objective Loss 0.147860 LR 0.000250 Time 0.020656 -2022-12-06 11:34:16,303 - Epoch: [160][ 710/ 1200] Overall Loss 0.147759 Objective Loss 0.147759 LR 0.000250 Time 0.020639 -2022-12-06 11:34:16,500 - Epoch: [160][ 720/ 1200] Overall Loss 0.147593 Objective Loss 0.147593 LR 0.000250 Time 0.020626 -2022-12-06 11:34:16,695 - Epoch: [160][ 730/ 1200] Overall Loss 0.147932 Objective Loss 0.147932 LR 0.000250 Time 0.020609 -2022-12-06 11:34:16,893 - Epoch: [160][ 740/ 1200] Overall Loss 0.147812 Objective Loss 0.147812 LR 0.000250 Time 0.020597 -2022-12-06 11:34:17,089 - Epoch: [160][ 750/ 1200] Overall Loss 0.147795 Objective Loss 0.147795 LR 0.000250 Time 0.020584 -2022-12-06 11:34:17,288 - Epoch: [160][ 760/ 1200] Overall Loss 0.147697 Objective Loss 0.147697 LR 0.000250 Time 0.020574 -2022-12-06 11:34:17,483 - Epoch: [160][ 770/ 1200] Overall Loss 0.147735 Objective Loss 0.147735 LR 0.000250 Time 0.020559 -2022-12-06 11:34:17,680 - Epoch: [160][ 780/ 1200] Overall Loss 0.147606 Objective Loss 0.147606 LR 0.000250 Time 0.020548 -2022-12-06 11:34:17,875 - Epoch: [160][ 790/ 1200] Overall Loss 0.147651 Objective Loss 0.147651 LR 0.000250 Time 0.020534 -2022-12-06 11:34:18,074 - Epoch: [160][ 800/ 1200] Overall Loss 0.147765 Objective Loss 0.147765 LR 0.000250 Time 0.020524 -2022-12-06 11:34:18,270 - Epoch: [160][ 810/ 1200] Overall Loss 0.147744 Objective Loss 0.147744 LR 0.000250 Time 0.020513 -2022-12-06 11:34:18,470 - Epoch: [160][ 820/ 1200] Overall Loss 0.147784 Objective Loss 0.147784 LR 0.000250 Time 0.020506 -2022-12-06 11:34:18,667 - Epoch: [160][ 830/ 1200] Overall Loss 0.147705 Objective Loss 0.147705 LR 0.000250 Time 0.020495 -2022-12-06 11:34:18,868 - Epoch: [160][ 840/ 1200] Overall Loss 0.147732 Objective Loss 0.147732 LR 0.000250 Time 0.020490 -2022-12-06 11:34:19,065 - Epoch: [160][ 850/ 1200] Overall Loss 0.148036 Objective Loss 0.148036 LR 0.000250 Time 0.020480 -2022-12-06 11:34:19,265 - Epoch: [160][ 860/ 1200] Overall Loss 0.147959 Objective Loss 0.147959 LR 0.000250 Time 0.020474 -2022-12-06 11:34:19,462 - Epoch: [160][ 870/ 1200] Overall Loss 0.147921 Objective Loss 0.147921 LR 0.000250 Time 0.020464 -2022-12-06 11:34:19,662 - Epoch: [160][ 880/ 1200] Overall Loss 0.147870 Objective Loss 0.147870 LR 0.000250 Time 0.020459 -2022-12-06 11:34:19,859 - Epoch: [160][ 890/ 1200] Overall Loss 0.148011 Objective Loss 0.148011 LR 0.000250 Time 0.020450 -2022-12-06 11:34:20,059 - Epoch: [160][ 900/ 1200] Overall Loss 0.148061 Objective Loss 0.148061 LR 0.000250 Time 0.020444 -2022-12-06 11:34:20,256 - Epoch: [160][ 910/ 1200] Overall Loss 0.147785 Objective Loss 0.147785 LR 0.000250 Time 0.020435 -2022-12-06 11:34:20,456 - Epoch: [160][ 920/ 1200] Overall Loss 0.147804 Objective Loss 0.147804 LR 0.000250 Time 0.020430 -2022-12-06 11:34:20,654 - Epoch: [160][ 930/ 1200] Overall Loss 0.147864 Objective Loss 0.147864 LR 0.000250 Time 0.020422 -2022-12-06 11:34:20,854 - Epoch: [160][ 940/ 1200] Overall Loss 0.147782 Objective Loss 0.147782 LR 0.000250 Time 0.020417 -2022-12-06 11:34:21,051 - Epoch: [160][ 950/ 1200] Overall Loss 0.147972 Objective Loss 0.147972 LR 0.000250 Time 0.020409 -2022-12-06 11:34:21,252 - Epoch: [160][ 960/ 1200] Overall Loss 0.148001 Objective Loss 0.148001 LR 0.000250 Time 0.020405 -2022-12-06 11:34:21,449 - Epoch: [160][ 970/ 1200] Overall Loss 0.148077 Objective Loss 0.148077 LR 0.000250 Time 0.020397 -2022-12-06 11:34:21,649 - Epoch: [160][ 980/ 1200] Overall Loss 0.148159 Objective Loss 0.148159 LR 0.000250 Time 0.020393 -2022-12-06 11:34:21,845 - Epoch: [160][ 990/ 1200] Overall Loss 0.148270 Objective Loss 0.148270 LR 0.000250 Time 0.020385 -2022-12-06 11:34:22,045 - Epoch: [160][ 1000/ 1200] Overall Loss 0.148480 Objective Loss 0.148480 LR 0.000250 Time 0.020380 -2022-12-06 11:34:22,242 - Epoch: [160][ 1010/ 1200] Overall Loss 0.148442 Objective Loss 0.148442 LR 0.000250 Time 0.020372 -2022-12-06 11:34:22,441 - Epoch: [160][ 1020/ 1200] Overall Loss 0.148510 Objective Loss 0.148510 LR 0.000250 Time 0.020367 -2022-12-06 11:34:22,639 - Epoch: [160][ 1030/ 1200] Overall Loss 0.148483 Objective Loss 0.148483 LR 0.000250 Time 0.020361 -2022-12-06 11:34:22,839 - Epoch: [160][ 1040/ 1200] Overall Loss 0.148324 Objective Loss 0.148324 LR 0.000250 Time 0.020358 -2022-12-06 11:34:23,037 - Epoch: [160][ 1050/ 1200] Overall Loss 0.148370 Objective Loss 0.148370 LR 0.000250 Time 0.020351 -2022-12-06 11:34:23,237 - Epoch: [160][ 1060/ 1200] Overall Loss 0.148272 Objective Loss 0.148272 LR 0.000250 Time 0.020348 -2022-12-06 11:34:23,434 - Epoch: [160][ 1070/ 1200] Overall Loss 0.148124 Objective Loss 0.148124 LR 0.000250 Time 0.020341 -2022-12-06 11:34:23,634 - Epoch: [160][ 1080/ 1200] Overall Loss 0.148157 Objective Loss 0.148157 LR 0.000250 Time 0.020338 -2022-12-06 11:34:23,832 - Epoch: [160][ 1090/ 1200] Overall Loss 0.148136 Objective Loss 0.148136 LR 0.000250 Time 0.020332 -2022-12-06 11:34:24,032 - Epoch: [160][ 1100/ 1200] Overall Loss 0.147809 Objective Loss 0.147809 LR 0.000250 Time 0.020329 -2022-12-06 11:34:24,229 - Epoch: [160][ 1110/ 1200] Overall Loss 0.148026 Objective Loss 0.148026 LR 0.000250 Time 0.020322 -2022-12-06 11:34:24,428 - Epoch: [160][ 1120/ 1200] Overall Loss 0.147856 Objective Loss 0.147856 LR 0.000250 Time 0.020319 -2022-12-06 11:34:24,626 - Epoch: [160][ 1130/ 1200] Overall Loss 0.147744 Objective Loss 0.147744 LR 0.000250 Time 0.020313 -2022-12-06 11:34:24,826 - Epoch: [160][ 1140/ 1200] Overall Loss 0.147758 Objective Loss 0.147758 LR 0.000250 Time 0.020310 -2022-12-06 11:34:25,023 - Epoch: [160][ 1150/ 1200] Overall Loss 0.147920 Objective Loss 0.147920 LR 0.000250 Time 0.020304 -2022-12-06 11:34:25,223 - Epoch: [160][ 1160/ 1200] Overall Loss 0.147877 Objective Loss 0.147877 LR 0.000250 Time 0.020301 -2022-12-06 11:34:25,422 - Epoch: [160][ 1170/ 1200] Overall Loss 0.147927 Objective Loss 0.147927 LR 0.000250 Time 0.020297 -2022-12-06 11:34:25,621 - Epoch: [160][ 1180/ 1200] Overall Loss 0.147998 Objective Loss 0.147998 LR 0.000250 Time 0.020293 -2022-12-06 11:34:25,819 - Epoch: [160][ 1190/ 1200] Overall Loss 0.148038 Objective Loss 0.148038 LR 0.000250 Time 0.020288 -2022-12-06 11:34:26,049 - Epoch: [160][ 1200/ 1200] Overall Loss 0.148129 Objective Loss 0.148129 Top1 90.376569 Top5 98.117155 LR 0.000250 Time 0.020311 -2022-12-06 11:34:26,137 - --- validate (epoch=160)----------- -2022-12-06 11:34:26,137 - 34129 samples (256 per mini-batch) -2022-12-06 11:34:26,589 - Epoch: [160][ 10/ 134] Loss 0.245200 Top1 88.085938 Top5 98.398438 -2022-12-06 11:34:26,717 - Epoch: [160][ 20/ 134] Loss 0.237321 Top1 88.222656 Top5 98.750000 -2022-12-06 11:34:26,845 - Epoch: [160][ 30/ 134] Loss 0.243639 Top1 87.734375 Top5 98.619792 -2022-12-06 11:34:26,971 - Epoch: [160][ 40/ 134] Loss 0.238820 Top1 87.734375 Top5 98.642578 -2022-12-06 11:34:27,099 - Epoch: [160][ 50/ 134] Loss 0.231731 Top1 87.757812 Top5 98.664062 -2022-12-06 11:34:27,224 - Epoch: [160][ 60/ 134] Loss 0.236247 Top1 87.708333 Top5 98.632812 -2022-12-06 11:34:27,352 - Epoch: [160][ 70/ 134] Loss 0.232687 Top1 87.795759 Top5 98.577009 -2022-12-06 11:34:27,483 - Epoch: [160][ 80/ 134] Loss 0.231116 Top1 87.871094 Top5 98.549805 -2022-12-06 11:34:27,613 - Epoch: [160][ 90/ 134] Loss 0.233690 Top1 87.708333 Top5 98.550347 -2022-12-06 11:34:27,743 - Epoch: [160][ 100/ 134] Loss 0.231425 Top1 87.753906 Top5 98.546875 -2022-12-06 11:34:27,873 - Epoch: [160][ 110/ 134] Loss 0.232615 Top1 87.666903 Top5 98.544034 -2022-12-06 11:34:28,004 - Epoch: [160][ 120/ 134] Loss 0.232210 Top1 87.711589 Top5 98.593750 -2022-12-06 11:34:28,132 - Epoch: [160][ 130/ 134] Loss 0.231991 Top1 87.644231 Top5 98.599760 -2022-12-06 11:34:28,169 - Epoch: [160][ 134/ 134] Loss 0.230954 Top1 87.673240 Top5 98.596502 -2022-12-06 11:34:28,257 - ==> Top1: 87.673 Top5: 98.597 Loss: 0.231 - -2022-12-06 11:34:28,258 - ==> Confusion: -[[ 933 1 1 2 2 5 1 1 7 29 0 1 1 2 7 1 0 0 0 0 2] - [ 1 943 1 2 8 23 2 13 0 2 0 4 0 0 0 1 4 1 11 4 7] - [ 4 3 1014 11 6 2 14 11 1 4 4 5 2 0 1 1 2 2 3 4 9] - [ 2 4 18 940 0 2 0 1 0 0 9 0 4 3 13 0 2 2 13 0 7] - [ 8 3 1 1 963 1 1 3 0 5 1 2 1 2 7 6 5 3 0 2 5] - [ 0 11 1 2 4 988 2 14 3 1 0 12 2 16 1 2 1 0 0 6 3] - [ 0 2 10 1 1 2 1074 4 0 0 1 1 0 0 0 5 2 1 2 9 3] - [ 2 8 4 1 2 26 7 968 0 0 0 5 0 3 0 0 0 0 18 7 3] - [ 9 4 0 0 0 2 1 0 982 34 10 1 1 7 8 0 1 0 1 2 1] - [ 77 0 1 0 5 5 0 2 17 872 1 2 0 8 3 1 0 0 1 0 6] - [ 2 1 4 2 2 2 1 3 10 0 965 1 1 9 3 1 0 0 6 0 6] - [ 1 0 2 0 0 13 4 2 0 0 0 979 18 6 0 5 4 6 0 8 3] - [ 0 1 1 0 1 3 0 0 0 1 0 24 913 1 0 6 1 9 1 3 4] - [ 1 1 0 0 1 3 0 3 12 9 3 5 3 966 1 1 3 1 0 3 7] - [ 8 4 2 4 4 3 0 0 12 1 0 2 2 4 1072 0 0 1 5 2 4] - [ 1 0 1 2 2 0 2 0 1 0 0 6 5 2 0 1001 4 9 0 3 4] - [ 2 1 0 2 2 0 2 0 2 0 0 4 3 3 0 10 1032 0 1 4 4] - [ 3 0 1 1 0 1 0 1 0 4 0 7 21 1 2 15 0 977 0 0 2] - [ 3 3 1 4 1 1 0 23 1 1 2 4 4 0 8 0 0 1 944 2 5] - [ 2 2 0 0 1 3 7 7 0 1 2 12 6 5 1 2 5 5 1 1014 4] - [ 134 204 141 72 98 155 75 138 76 76 143 81 324 259 139 112 147 73 163 235 10381]] - -2022-12-06 11:34:28,927 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:34:28,927 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:34:28,933 - - -2022-12-06 11:34:28,933 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:34:29,882 - Epoch: [161][ 10/ 1200] Overall Loss 0.159175 Objective Loss 0.159175 LR 0.000250 Time 0.094794 -2022-12-06 11:34:30,091 - Epoch: [161][ 20/ 1200] Overall Loss 0.156900 Objective Loss 0.156900 LR 0.000250 Time 0.057832 -2022-12-06 11:34:30,294 - Epoch: [161][ 30/ 1200] Overall Loss 0.149770 Objective Loss 0.149770 LR 0.000250 Time 0.045317 -2022-12-06 11:34:30,502 - Epoch: [161][ 40/ 1200] Overall Loss 0.149451 Objective Loss 0.149451 LR 0.000250 Time 0.039163 -2022-12-06 11:34:30,705 - Epoch: [161][ 50/ 1200] Overall Loss 0.147095 Objective Loss 0.147095 LR 0.000250 Time 0.035370 -2022-12-06 11:34:30,912 - Epoch: [161][ 60/ 1200] Overall Loss 0.147969 Objective Loss 0.147969 LR 0.000250 Time 0.032912 -2022-12-06 11:34:31,115 - Epoch: [161][ 70/ 1200] Overall Loss 0.146690 Objective Loss 0.146690 LR 0.000250 Time 0.031109 -2022-12-06 11:34:31,322 - Epoch: [161][ 80/ 1200] Overall Loss 0.147153 Objective Loss 0.147153 LR 0.000250 Time 0.029793 -2022-12-06 11:34:31,525 - Epoch: [161][ 90/ 1200] Overall Loss 0.146585 Objective Loss 0.146585 LR 0.000250 Time 0.028732 -2022-12-06 11:34:31,731 - Epoch: [161][ 100/ 1200] Overall Loss 0.147480 Objective Loss 0.147480 LR 0.000250 Time 0.027919 -2022-12-06 11:34:31,934 - Epoch: [161][ 110/ 1200] Overall Loss 0.145593 Objective Loss 0.145593 LR 0.000250 Time 0.027216 -2022-12-06 11:34:32,141 - Epoch: [161][ 120/ 1200] Overall Loss 0.145405 Objective Loss 0.145405 LR 0.000250 Time 0.026672 -2022-12-06 11:34:32,346 - Epoch: [161][ 130/ 1200] Overall Loss 0.146066 Objective Loss 0.146066 LR 0.000250 Time 0.026188 -2022-12-06 11:34:32,552 - Epoch: [161][ 140/ 1200] Overall Loss 0.146070 Objective Loss 0.146070 LR 0.000250 Time 0.025789 -2022-12-06 11:34:32,756 - Epoch: [161][ 150/ 1200] Overall Loss 0.147067 Objective Loss 0.147067 LR 0.000250 Time 0.025422 -2022-12-06 11:34:32,963 - Epoch: [161][ 160/ 1200] Overall Loss 0.146633 Objective Loss 0.146633 LR 0.000250 Time 0.025126 -2022-12-06 11:34:33,166 - Epoch: [161][ 170/ 1200] Overall Loss 0.145402 Objective Loss 0.145402 LR 0.000250 Time 0.024837 -2022-12-06 11:34:33,373 - Epoch: [161][ 180/ 1200] Overall Loss 0.145006 Objective Loss 0.145006 LR 0.000250 Time 0.024602 -2022-12-06 11:34:33,576 - Epoch: [161][ 190/ 1200] Overall Loss 0.144214 Objective Loss 0.144214 LR 0.000250 Time 0.024373 -2022-12-06 11:34:33,784 - Epoch: [161][ 200/ 1200] Overall Loss 0.144044 Objective Loss 0.144044 LR 0.000250 Time 0.024191 -2022-12-06 11:34:33,987 - Epoch: [161][ 210/ 1200] Overall Loss 0.144338 Objective Loss 0.144338 LR 0.000250 Time 0.024002 -2022-12-06 11:34:34,194 - Epoch: [161][ 220/ 1200] Overall Loss 0.144304 Objective Loss 0.144304 LR 0.000250 Time 0.023849 -2022-12-06 11:34:34,397 - Epoch: [161][ 230/ 1200] Overall Loss 0.143528 Objective Loss 0.143528 LR 0.000250 Time 0.023693 -2022-12-06 11:34:34,604 - Epoch: [161][ 240/ 1200] Overall Loss 0.142976 Objective Loss 0.142976 LR 0.000250 Time 0.023564 -2022-12-06 11:34:34,807 - Epoch: [161][ 250/ 1200] Overall Loss 0.142713 Objective Loss 0.142713 LR 0.000250 Time 0.023432 -2022-12-06 11:34:35,013 - Epoch: [161][ 260/ 1200] Overall Loss 0.142569 Objective Loss 0.142569 LR 0.000250 Time 0.023322 -2022-12-06 11:34:35,217 - Epoch: [161][ 270/ 1200] Overall Loss 0.142403 Objective Loss 0.142403 LR 0.000250 Time 0.023210 -2022-12-06 11:34:35,424 - Epoch: [161][ 280/ 1200] Overall Loss 0.142050 Objective Loss 0.142050 LR 0.000250 Time 0.023119 -2022-12-06 11:34:35,628 - Epoch: [161][ 290/ 1200] Overall Loss 0.142807 Objective Loss 0.142807 LR 0.000250 Time 0.023024 -2022-12-06 11:34:35,834 - Epoch: [161][ 300/ 1200] Overall Loss 0.142912 Objective Loss 0.142912 LR 0.000250 Time 0.022939 -2022-12-06 11:34:36,036 - Epoch: [161][ 310/ 1200] Overall Loss 0.143349 Objective Loss 0.143349 LR 0.000250 Time 0.022851 -2022-12-06 11:34:36,242 - Epoch: [161][ 320/ 1200] Overall Loss 0.143531 Objective Loss 0.143531 LR 0.000250 Time 0.022777 -2022-12-06 11:34:36,445 - Epoch: [161][ 330/ 1200] Overall Loss 0.143733 Objective Loss 0.143733 LR 0.000250 Time 0.022702 -2022-12-06 11:34:36,652 - Epoch: [161][ 340/ 1200] Overall Loss 0.143859 Objective Loss 0.143859 LR 0.000250 Time 0.022640 -2022-12-06 11:34:36,854 - Epoch: [161][ 350/ 1200] Overall Loss 0.143663 Objective Loss 0.143663 LR 0.000250 Time 0.022570 -2022-12-06 11:34:37,061 - Epoch: [161][ 360/ 1200] Overall Loss 0.144080 Objective Loss 0.144080 LR 0.000250 Time 0.022515 -2022-12-06 11:34:37,264 - Epoch: [161][ 370/ 1200] Overall Loss 0.144342 Objective Loss 0.144342 LR 0.000250 Time 0.022454 -2022-12-06 11:34:37,470 - Epoch: [161][ 380/ 1200] Overall Loss 0.144303 Objective Loss 0.144303 LR 0.000250 Time 0.022404 -2022-12-06 11:34:37,673 - Epoch: [161][ 390/ 1200] Overall Loss 0.144337 Objective Loss 0.144337 LR 0.000250 Time 0.022347 -2022-12-06 11:34:37,880 - Epoch: [161][ 400/ 1200] Overall Loss 0.144161 Objective Loss 0.144161 LR 0.000250 Time 0.022304 -2022-12-06 11:34:38,082 - Epoch: [161][ 410/ 1200] Overall Loss 0.144660 Objective Loss 0.144660 LR 0.000250 Time 0.022251 -2022-12-06 11:34:38,288 - Epoch: [161][ 420/ 1200] Overall Loss 0.144736 Objective Loss 0.144736 LR 0.000250 Time 0.022211 -2022-12-06 11:34:38,490 - Epoch: [161][ 430/ 1200] Overall Loss 0.144628 Objective Loss 0.144628 LR 0.000250 Time 0.022163 -2022-12-06 11:34:38,696 - Epoch: [161][ 440/ 1200] Overall Loss 0.144475 Objective Loss 0.144475 LR 0.000250 Time 0.022127 -2022-12-06 11:34:38,899 - Epoch: [161][ 450/ 1200] Overall Loss 0.144350 Objective Loss 0.144350 LR 0.000250 Time 0.022084 -2022-12-06 11:34:39,106 - Epoch: [161][ 460/ 1200] Overall Loss 0.144674 Objective Loss 0.144674 LR 0.000250 Time 0.022052 -2022-12-06 11:34:39,308 - Epoch: [161][ 470/ 1200] Overall Loss 0.145137 Objective Loss 0.145137 LR 0.000250 Time 0.022013 -2022-12-06 11:34:39,515 - Epoch: [161][ 480/ 1200] Overall Loss 0.144899 Objective Loss 0.144899 LR 0.000250 Time 0.021983 -2022-12-06 11:34:39,718 - Epoch: [161][ 490/ 1200] Overall Loss 0.145322 Objective Loss 0.145322 LR 0.000250 Time 0.021948 -2022-12-06 11:34:39,925 - Epoch: [161][ 500/ 1200] Overall Loss 0.145468 Objective Loss 0.145468 LR 0.000250 Time 0.021921 -2022-12-06 11:34:40,128 - Epoch: [161][ 510/ 1200] Overall Loss 0.145456 Objective Loss 0.145456 LR 0.000250 Time 0.021888 -2022-12-06 11:34:40,334 - Epoch: [161][ 520/ 1200] Overall Loss 0.145725 Objective Loss 0.145725 LR 0.000250 Time 0.021862 -2022-12-06 11:34:40,537 - Epoch: [161][ 530/ 1200] Overall Loss 0.145756 Objective Loss 0.145756 LR 0.000250 Time 0.021832 -2022-12-06 11:34:40,744 - Epoch: [161][ 540/ 1200] Overall Loss 0.145594 Objective Loss 0.145594 LR 0.000250 Time 0.021809 -2022-12-06 11:34:40,946 - Epoch: [161][ 550/ 1200] Overall Loss 0.145497 Objective Loss 0.145497 LR 0.000250 Time 0.021780 -2022-12-06 11:34:41,153 - Epoch: [161][ 560/ 1200] Overall Loss 0.145640 Objective Loss 0.145640 LR 0.000250 Time 0.021760 -2022-12-06 11:34:41,355 - Epoch: [161][ 570/ 1200] Overall Loss 0.145885 Objective Loss 0.145885 LR 0.000250 Time 0.021731 -2022-12-06 11:34:41,562 - Epoch: [161][ 580/ 1200] Overall Loss 0.146113 Objective Loss 0.146113 LR 0.000250 Time 0.021712 -2022-12-06 11:34:41,765 - Epoch: [161][ 590/ 1200] Overall Loss 0.146079 Objective Loss 0.146079 LR 0.000250 Time 0.021687 -2022-12-06 11:34:41,971 - Epoch: [161][ 600/ 1200] Overall Loss 0.146057 Objective Loss 0.146057 LR 0.000250 Time 0.021668 -2022-12-06 11:34:42,173 - Epoch: [161][ 610/ 1200] Overall Loss 0.145926 Objective Loss 0.145926 LR 0.000250 Time 0.021643 -2022-12-06 11:34:42,380 - Epoch: [161][ 620/ 1200] Overall Loss 0.145923 Objective Loss 0.145923 LR 0.000250 Time 0.021626 -2022-12-06 11:34:42,582 - Epoch: [161][ 630/ 1200] Overall Loss 0.146260 Objective Loss 0.146260 LR 0.000250 Time 0.021603 -2022-12-06 11:34:42,789 - Epoch: [161][ 640/ 1200] Overall Loss 0.146344 Objective Loss 0.146344 LR 0.000250 Time 0.021587 -2022-12-06 11:34:42,990 - Epoch: [161][ 650/ 1200] Overall Loss 0.146398 Objective Loss 0.146398 LR 0.000250 Time 0.021564 -2022-12-06 11:34:43,197 - Epoch: [161][ 660/ 1200] Overall Loss 0.146449 Objective Loss 0.146449 LR 0.000250 Time 0.021550 -2022-12-06 11:34:43,399 - Epoch: [161][ 670/ 1200] Overall Loss 0.146262 Objective Loss 0.146262 LR 0.000250 Time 0.021529 -2022-12-06 11:34:43,607 - Epoch: [161][ 680/ 1200] Overall Loss 0.146058 Objective Loss 0.146058 LR 0.000250 Time 0.021516 -2022-12-06 11:34:43,809 - Epoch: [161][ 690/ 1200] Overall Loss 0.145854 Objective Loss 0.145854 LR 0.000250 Time 0.021496 -2022-12-06 11:34:44,016 - Epoch: [161][ 700/ 1200] Overall Loss 0.145827 Objective Loss 0.145827 LR 0.000250 Time 0.021485 -2022-12-06 11:34:44,219 - Epoch: [161][ 710/ 1200] Overall Loss 0.146210 Objective Loss 0.146210 LR 0.000250 Time 0.021468 -2022-12-06 11:34:44,425 - Epoch: [161][ 720/ 1200] Overall Loss 0.146187 Objective Loss 0.146187 LR 0.000250 Time 0.021455 -2022-12-06 11:34:44,629 - Epoch: [161][ 730/ 1200] Overall Loss 0.146257 Objective Loss 0.146257 LR 0.000250 Time 0.021439 -2022-12-06 11:34:44,836 - Epoch: [161][ 740/ 1200] Overall Loss 0.146440 Objective Loss 0.146440 LR 0.000250 Time 0.021428 -2022-12-06 11:34:45,036 - Epoch: [161][ 750/ 1200] Overall Loss 0.146474 Objective Loss 0.146474 LR 0.000250 Time 0.021409 -2022-12-06 11:34:45,243 - Epoch: [161][ 760/ 1200] Overall Loss 0.146377 Objective Loss 0.146377 LR 0.000250 Time 0.021398 -2022-12-06 11:34:45,446 - Epoch: [161][ 770/ 1200] Overall Loss 0.146342 Objective Loss 0.146342 LR 0.000250 Time 0.021383 -2022-12-06 11:34:45,652 - Epoch: [161][ 780/ 1200] Overall Loss 0.146359 Objective Loss 0.146359 LR 0.000250 Time 0.021371 -2022-12-06 11:34:45,854 - Epoch: [161][ 790/ 1200] Overall Loss 0.146456 Objective Loss 0.146456 LR 0.000250 Time 0.021357 -2022-12-06 11:34:46,060 - Epoch: [161][ 800/ 1200] Overall Loss 0.146493 Objective Loss 0.146493 LR 0.000250 Time 0.021346 -2022-12-06 11:34:46,263 - Epoch: [161][ 810/ 1200] Overall Loss 0.146628 Objective Loss 0.146628 LR 0.000250 Time 0.021332 -2022-12-06 11:34:46,470 - Epoch: [161][ 820/ 1200] Overall Loss 0.146676 Objective Loss 0.146676 LR 0.000250 Time 0.021324 -2022-12-06 11:34:46,673 - Epoch: [161][ 830/ 1200] Overall Loss 0.146943 Objective Loss 0.146943 LR 0.000250 Time 0.021310 -2022-12-06 11:34:46,879 - Epoch: [161][ 840/ 1200] Overall Loss 0.147043 Objective Loss 0.147043 LR 0.000250 Time 0.021302 -2022-12-06 11:34:47,081 - Epoch: [161][ 850/ 1200] Overall Loss 0.147000 Objective Loss 0.147000 LR 0.000250 Time 0.021288 -2022-12-06 11:34:47,288 - Epoch: [161][ 860/ 1200] Overall Loss 0.147308 Objective Loss 0.147308 LR 0.000250 Time 0.021280 -2022-12-06 11:34:47,491 - Epoch: [161][ 870/ 1200] Overall Loss 0.147415 Objective Loss 0.147415 LR 0.000250 Time 0.021268 -2022-12-06 11:34:47,697 - Epoch: [161][ 880/ 1200] Overall Loss 0.147473 Objective Loss 0.147473 LR 0.000250 Time 0.021261 -2022-12-06 11:34:47,901 - Epoch: [161][ 890/ 1200] Overall Loss 0.147314 Objective Loss 0.147314 LR 0.000250 Time 0.021250 -2022-12-06 11:34:48,108 - Epoch: [161][ 900/ 1200] Overall Loss 0.147463 Objective Loss 0.147463 LR 0.000250 Time 0.021243 -2022-12-06 11:34:48,312 - Epoch: [161][ 910/ 1200] Overall Loss 0.147397 Objective Loss 0.147397 LR 0.000250 Time 0.021233 -2022-12-06 11:34:48,518 - Epoch: [161][ 920/ 1200] Overall Loss 0.147444 Objective Loss 0.147444 LR 0.000250 Time 0.021226 -2022-12-06 11:34:48,721 - Epoch: [161][ 930/ 1200] Overall Loss 0.147295 Objective Loss 0.147295 LR 0.000250 Time 0.021215 -2022-12-06 11:34:48,928 - Epoch: [161][ 940/ 1200] Overall Loss 0.147475 Objective Loss 0.147475 LR 0.000250 Time 0.021209 -2022-12-06 11:34:49,132 - Epoch: [161][ 950/ 1200] Overall Loss 0.147650 Objective Loss 0.147650 LR 0.000250 Time 0.021199 -2022-12-06 11:34:49,339 - Epoch: [161][ 960/ 1200] Overall Loss 0.147619 Objective Loss 0.147619 LR 0.000250 Time 0.021194 -2022-12-06 11:34:49,542 - Epoch: [161][ 970/ 1200] Overall Loss 0.147614 Objective Loss 0.147614 LR 0.000250 Time 0.021184 -2022-12-06 11:34:49,750 - Epoch: [161][ 980/ 1200] Overall Loss 0.147647 Objective Loss 0.147647 LR 0.000250 Time 0.021179 -2022-12-06 11:34:49,952 - Epoch: [161][ 990/ 1200] Overall Loss 0.147659 Objective Loss 0.147659 LR 0.000250 Time 0.021169 -2022-12-06 11:34:50,159 - Epoch: [161][ 1000/ 1200] Overall Loss 0.147702 Objective Loss 0.147702 LR 0.000250 Time 0.021163 -2022-12-06 11:34:50,363 - Epoch: [161][ 1010/ 1200] Overall Loss 0.147930 Objective Loss 0.147930 LR 0.000250 Time 0.021155 -2022-12-06 11:34:50,571 - Epoch: [161][ 1020/ 1200] Overall Loss 0.147969 Objective Loss 0.147969 LR 0.000250 Time 0.021151 -2022-12-06 11:34:50,773 - Epoch: [161][ 1030/ 1200] Overall Loss 0.148105 Objective Loss 0.148105 LR 0.000250 Time 0.021142 -2022-12-06 11:34:50,980 - Epoch: [161][ 1040/ 1200] Overall Loss 0.148136 Objective Loss 0.148136 LR 0.000250 Time 0.021137 -2022-12-06 11:34:51,182 - Epoch: [161][ 1050/ 1200] Overall Loss 0.148129 Objective Loss 0.148129 LR 0.000250 Time 0.021127 -2022-12-06 11:34:51,388 - Epoch: [161][ 1060/ 1200] Overall Loss 0.148204 Objective Loss 0.148204 LR 0.000250 Time 0.021121 -2022-12-06 11:34:51,590 - Epoch: [161][ 1070/ 1200] Overall Loss 0.148325 Objective Loss 0.148325 LR 0.000250 Time 0.021112 -2022-12-06 11:34:51,797 - Epoch: [161][ 1080/ 1200] Overall Loss 0.148309 Objective Loss 0.148309 LR 0.000250 Time 0.021108 -2022-12-06 11:34:51,999 - Epoch: [161][ 1090/ 1200] Overall Loss 0.148464 Objective Loss 0.148464 LR 0.000250 Time 0.021099 -2022-12-06 11:34:52,205 - Epoch: [161][ 1100/ 1200] Overall Loss 0.148334 Objective Loss 0.148334 LR 0.000250 Time 0.021094 -2022-12-06 11:34:52,407 - Epoch: [161][ 1110/ 1200] Overall Loss 0.148456 Objective Loss 0.148456 LR 0.000250 Time 0.021085 -2022-12-06 11:34:52,614 - Epoch: [161][ 1120/ 1200] Overall Loss 0.148339 Objective Loss 0.148339 LR 0.000250 Time 0.021081 -2022-12-06 11:34:52,816 - Epoch: [161][ 1130/ 1200] Overall Loss 0.148289 Objective Loss 0.148289 LR 0.000250 Time 0.021073 -2022-12-06 11:34:53,023 - Epoch: [161][ 1140/ 1200] Overall Loss 0.148309 Objective Loss 0.148309 LR 0.000250 Time 0.021069 -2022-12-06 11:34:53,226 - Epoch: [161][ 1150/ 1200] Overall Loss 0.148307 Objective Loss 0.148307 LR 0.000250 Time 0.021062 -2022-12-06 11:34:53,432 - Epoch: [161][ 1160/ 1200] Overall Loss 0.148422 Objective Loss 0.148422 LR 0.000250 Time 0.021058 -2022-12-06 11:34:53,635 - Epoch: [161][ 1170/ 1200] Overall Loss 0.148512 Objective Loss 0.148512 LR 0.000250 Time 0.021050 -2022-12-06 11:34:53,841 - Epoch: [161][ 1180/ 1200] Overall Loss 0.148366 Objective Loss 0.148366 LR 0.000250 Time 0.021046 -2022-12-06 11:34:54,043 - Epoch: [161][ 1190/ 1200] Overall Loss 0.148335 Objective Loss 0.148335 LR 0.000250 Time 0.021038 -2022-12-06 11:34:54,278 - Epoch: [161][ 1200/ 1200] Overall Loss 0.148453 Objective Loss 0.148453 Top1 89.330544 Top5 99.372385 LR 0.000250 Time 0.021059 -2022-12-06 11:34:54,367 - --- validate (epoch=161)----------- -2022-12-06 11:34:54,367 - 34129 samples (256 per mini-batch) -2022-12-06 11:34:54,845 - Epoch: [161][ 10/ 134] Loss 0.230954 Top1 87.500000 Top5 98.710938 -2022-12-06 11:34:54,986 - Epoch: [161][ 20/ 134] Loss 0.245930 Top1 87.148438 Top5 98.554688 -2022-12-06 11:34:55,124 - Epoch: [161][ 30/ 134] Loss 0.240098 Top1 87.539062 Top5 98.515625 -2022-12-06 11:34:55,251 - Epoch: [161][ 40/ 134] Loss 0.233944 Top1 87.861328 Top5 98.505859 -2022-12-06 11:34:55,380 - Epoch: [161][ 50/ 134] Loss 0.235403 Top1 87.890625 Top5 98.484375 -2022-12-06 11:34:55,512 - Epoch: [161][ 60/ 134] Loss 0.228012 Top1 88.066406 Top5 98.502604 -2022-12-06 11:34:55,654 - Epoch: [161][ 70/ 134] Loss 0.223398 Top1 88.258929 Top5 98.554688 -2022-12-06 11:34:55,800 - Epoch: [161][ 80/ 134] Loss 0.222143 Top1 88.217773 Top5 98.574219 -2022-12-06 11:34:55,936 - Epoch: [161][ 90/ 134] Loss 0.223349 Top1 88.242188 Top5 98.524306 -2022-12-06 11:34:56,066 - Epoch: [161][ 100/ 134] Loss 0.224559 Top1 88.210938 Top5 98.546875 -2022-12-06 11:34:56,196 - Epoch: [161][ 110/ 134] Loss 0.222895 Top1 88.259943 Top5 98.568892 -2022-12-06 11:34:56,328 - Epoch: [161][ 120/ 134] Loss 0.226024 Top1 88.151042 Top5 98.564453 -2022-12-06 11:34:56,459 - Epoch: [161][ 130/ 134] Loss 0.225709 Top1 88.109976 Top5 98.566707 -2022-12-06 11:34:56,496 - Epoch: [161][ 134/ 134] Loss 0.228126 Top1 88.080518 Top5 98.549621 -2022-12-06 11:34:56,583 - ==> Top1: 88.081 Top5: 98.550 Loss: 0.228 - -2022-12-06 11:34:56,584 - ==> Confusion: -[[ 918 0 0 2 4 3 1 0 4 46 0 2 1 2 8 1 0 1 0 0 3] - [ 1 949 1 3 10 15 4 9 1 0 1 3 0 1 2 2 2 0 15 3 5] - [ 4 1 1021 9 3 3 13 8 0 2 5 6 2 0 1 2 1 3 2 2 15] - [ 2 2 15 945 1 1 1 1 0 2 11 0 5 1 13 0 0 2 10 1 7] - [ 9 3 1 0 966 0 1 2 1 6 2 1 0 2 11 2 5 2 1 1 4] - [ 0 18 0 3 9 978 2 16 3 5 0 9 2 8 2 2 3 1 1 4 3] - [ 2 2 7 2 0 2 1079 3 0 0 0 0 0 5 0 5 2 0 1 7 1] - [ 1 10 4 2 2 21 9 958 2 1 0 5 1 0 1 1 0 2 23 5 6] - [ 4 2 0 0 1 2 1 0 985 39 7 1 0 5 11 1 1 0 2 1 1] - [ 42 0 1 0 5 1 0 2 15 916 1 2 0 8 2 1 0 0 1 0 4] - [ 1 1 3 4 3 1 1 2 9 2 963 1 3 9 3 1 1 0 3 1 7] - [ 2 0 2 0 1 11 6 2 1 0 1 976 12 4 1 7 4 5 0 12 4] - [ 0 1 1 1 1 2 0 2 0 1 0 28 899 2 1 11 1 9 0 2 7] - [ 0 1 0 0 1 6 0 1 13 17 3 3 3 956 0 2 3 2 0 4 8] - [ 6 2 2 5 3 2 0 0 15 3 0 2 2 2 1074 0 1 1 6 0 4] - [ 1 0 1 0 1 0 3 0 2 0 0 8 3 3 0 1002 5 9 0 3 2] - [ 1 1 0 1 2 1 1 0 1 0 0 1 1 3 0 10 1040 2 0 2 5] - [ 3 0 1 0 0 1 0 0 0 2 1 6 10 2 0 15 1 991 0 1 2] - [ 2 4 5 6 1 1 0 15 1 2 2 2 3 1 9 2 0 2 944 1 5] - [ 2 4 2 2 0 3 8 3 0 1 2 14 5 4 0 3 4 2 1 1016 4] - [ 106 191 167 77 105 139 82 120 83 100 139 89 257 226 146 114 171 84 149 196 10485]] - -2022-12-06 11:34:57,263 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:34:57,263 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:34:57,269 - - -2022-12-06 11:34:57,269 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:34:58,208 - Epoch: [162][ 10/ 1200] Overall Loss 0.138849 Objective Loss 0.138849 LR 0.000250 Time 0.093877 -2022-12-06 11:34:58,403 - Epoch: [162][ 20/ 1200] Overall Loss 0.141188 Objective Loss 0.141188 LR 0.000250 Time 0.056657 -2022-12-06 11:34:58,595 - Epoch: [162][ 30/ 1200] Overall Loss 0.149972 Objective Loss 0.149972 LR 0.000250 Time 0.044145 -2022-12-06 11:34:58,787 - Epoch: [162][ 40/ 1200] Overall Loss 0.151718 Objective Loss 0.151718 LR 0.000250 Time 0.037897 -2022-12-06 11:34:58,979 - Epoch: [162][ 50/ 1200] Overall Loss 0.149332 Objective Loss 0.149332 LR 0.000250 Time 0.034139 -2022-12-06 11:34:59,169 - Epoch: [162][ 60/ 1200] Overall Loss 0.150315 Objective Loss 0.150315 LR 0.000250 Time 0.031610 -2022-12-06 11:34:59,360 - Epoch: [162][ 70/ 1200] Overall Loss 0.148984 Objective Loss 0.148984 LR 0.000250 Time 0.029817 -2022-12-06 11:34:59,552 - Epoch: [162][ 80/ 1200] Overall Loss 0.147599 Objective Loss 0.147599 LR 0.000250 Time 0.028484 -2022-12-06 11:34:59,744 - Epoch: [162][ 90/ 1200] Overall Loss 0.146903 Objective Loss 0.146903 LR 0.000250 Time 0.027440 -2022-12-06 11:34:59,936 - Epoch: [162][ 100/ 1200] Overall Loss 0.145402 Objective Loss 0.145402 LR 0.000250 Time 0.026611 -2022-12-06 11:35:00,127 - Epoch: [162][ 110/ 1200] Overall Loss 0.144674 Objective Loss 0.144674 LR 0.000250 Time 0.025926 -2022-12-06 11:35:00,319 - Epoch: [162][ 120/ 1200] Overall Loss 0.145257 Objective Loss 0.145257 LR 0.000250 Time 0.025358 -2022-12-06 11:35:00,510 - Epoch: [162][ 130/ 1200] Overall Loss 0.146910 Objective Loss 0.146910 LR 0.000250 Time 0.024878 -2022-12-06 11:35:00,702 - Epoch: [162][ 140/ 1200] Overall Loss 0.147142 Objective Loss 0.147142 LR 0.000250 Time 0.024464 -2022-12-06 11:35:00,893 - Epoch: [162][ 150/ 1200] Overall Loss 0.147666 Objective Loss 0.147666 LR 0.000250 Time 0.024108 -2022-12-06 11:35:01,086 - Epoch: [162][ 160/ 1200] Overall Loss 0.148725 Objective Loss 0.148725 LR 0.000250 Time 0.023802 -2022-12-06 11:35:01,277 - Epoch: [162][ 170/ 1200] Overall Loss 0.148361 Objective Loss 0.148361 LR 0.000250 Time 0.023522 -2022-12-06 11:35:01,470 - Epoch: [162][ 180/ 1200] Overall Loss 0.149327 Objective Loss 0.149327 LR 0.000250 Time 0.023281 -2022-12-06 11:35:01,661 - Epoch: [162][ 190/ 1200] Overall Loss 0.150848 Objective Loss 0.150848 LR 0.000250 Time 0.023059 -2022-12-06 11:35:01,853 - Epoch: [162][ 200/ 1200] Overall Loss 0.151287 Objective Loss 0.151287 LR 0.000250 Time 0.022866 -2022-12-06 11:35:02,045 - Epoch: [162][ 210/ 1200] Overall Loss 0.150909 Objective Loss 0.150909 LR 0.000250 Time 0.022687 -2022-12-06 11:35:02,238 - Epoch: [162][ 220/ 1200] Overall Loss 0.150516 Objective Loss 0.150516 LR 0.000250 Time 0.022529 -2022-12-06 11:35:02,428 - Epoch: [162][ 230/ 1200] Overall Loss 0.150624 Objective Loss 0.150624 LR 0.000250 Time 0.022376 -2022-12-06 11:35:02,619 - Epoch: [162][ 240/ 1200] Overall Loss 0.149562 Objective Loss 0.149562 LR 0.000250 Time 0.022237 -2022-12-06 11:35:02,811 - Epoch: [162][ 250/ 1200] Overall Loss 0.149282 Objective Loss 0.149282 LR 0.000250 Time 0.022112 -2022-12-06 11:35:03,002 - Epoch: [162][ 260/ 1200] Overall Loss 0.149304 Objective Loss 0.149304 LR 0.000250 Time 0.021995 -2022-12-06 11:35:03,194 - Epoch: [162][ 270/ 1200] Overall Loss 0.148822 Objective Loss 0.148822 LR 0.000250 Time 0.021888 -2022-12-06 11:35:03,385 - Epoch: [162][ 280/ 1200] Overall Loss 0.149143 Objective Loss 0.149143 LR 0.000250 Time 0.021788 -2022-12-06 11:35:03,576 - Epoch: [162][ 290/ 1200] Overall Loss 0.149412 Objective Loss 0.149412 LR 0.000250 Time 0.021694 -2022-12-06 11:35:03,767 - Epoch: [162][ 300/ 1200] Overall Loss 0.149261 Objective Loss 0.149261 LR 0.000250 Time 0.021605 -2022-12-06 11:35:03,961 - Epoch: [162][ 310/ 1200] Overall Loss 0.149064 Objective Loss 0.149064 LR 0.000250 Time 0.021532 -2022-12-06 11:35:04,153 - Epoch: [162][ 320/ 1200] Overall Loss 0.149308 Objective Loss 0.149308 LR 0.000250 Time 0.021456 -2022-12-06 11:35:04,344 - Epoch: [162][ 330/ 1200] Overall Loss 0.148886 Objective Loss 0.148886 LR 0.000250 Time 0.021384 -2022-12-06 11:35:04,536 - Epoch: [162][ 340/ 1200] Overall Loss 0.148801 Objective Loss 0.148801 LR 0.000250 Time 0.021318 -2022-12-06 11:35:04,728 - Epoch: [162][ 350/ 1200] Overall Loss 0.149458 Objective Loss 0.149458 LR 0.000250 Time 0.021255 -2022-12-06 11:35:04,919 - Epoch: [162][ 360/ 1200] Overall Loss 0.150134 Objective Loss 0.150134 LR 0.000250 Time 0.021195 -2022-12-06 11:35:05,111 - Epoch: [162][ 370/ 1200] Overall Loss 0.149958 Objective Loss 0.149958 LR 0.000250 Time 0.021139 -2022-12-06 11:35:05,304 - Epoch: [162][ 380/ 1200] Overall Loss 0.149714 Objective Loss 0.149714 LR 0.000250 Time 0.021088 -2022-12-06 11:35:05,495 - Epoch: [162][ 390/ 1200] Overall Loss 0.149545 Objective Loss 0.149545 LR 0.000250 Time 0.021038 -2022-12-06 11:35:05,688 - Epoch: [162][ 400/ 1200] Overall Loss 0.150016 Objective Loss 0.150016 LR 0.000250 Time 0.020991 -2022-12-06 11:35:05,879 - Epoch: [162][ 410/ 1200] Overall Loss 0.149451 Objective Loss 0.149451 LR 0.000250 Time 0.020945 -2022-12-06 11:35:06,071 - Epoch: [162][ 420/ 1200] Overall Loss 0.149376 Objective Loss 0.149376 LR 0.000250 Time 0.020901 -2022-12-06 11:35:06,262 - Epoch: [162][ 430/ 1200] Overall Loss 0.149460 Objective Loss 0.149460 LR 0.000250 Time 0.020859 -2022-12-06 11:35:06,453 - Epoch: [162][ 440/ 1200] Overall Loss 0.148765 Objective Loss 0.148765 LR 0.000250 Time 0.020817 -2022-12-06 11:35:06,644 - Epoch: [162][ 450/ 1200] Overall Loss 0.148242 Objective Loss 0.148242 LR 0.000250 Time 0.020778 -2022-12-06 11:35:06,837 - Epoch: [162][ 460/ 1200] Overall Loss 0.148199 Objective Loss 0.148199 LR 0.000250 Time 0.020744 -2022-12-06 11:35:07,028 - Epoch: [162][ 470/ 1200] Overall Loss 0.148114 Objective Loss 0.148114 LR 0.000250 Time 0.020708 -2022-12-06 11:35:07,220 - Epoch: [162][ 480/ 1200] Overall Loss 0.148167 Objective Loss 0.148167 LR 0.000250 Time 0.020676 -2022-12-06 11:35:07,412 - Epoch: [162][ 490/ 1200] Overall Loss 0.147994 Objective Loss 0.147994 LR 0.000250 Time 0.020644 -2022-12-06 11:35:07,604 - Epoch: [162][ 500/ 1200] Overall Loss 0.147890 Objective Loss 0.147890 LR 0.000250 Time 0.020615 -2022-12-06 11:35:07,796 - Epoch: [162][ 510/ 1200] Overall Loss 0.147461 Objective Loss 0.147461 LR 0.000250 Time 0.020586 -2022-12-06 11:35:07,988 - Epoch: [162][ 520/ 1200] Overall Loss 0.147605 Objective Loss 0.147605 LR 0.000250 Time 0.020558 -2022-12-06 11:35:08,180 - Epoch: [162][ 530/ 1200] Overall Loss 0.147675 Objective Loss 0.147675 LR 0.000250 Time 0.020531 -2022-12-06 11:35:08,372 - Epoch: [162][ 540/ 1200] Overall Loss 0.147511 Objective Loss 0.147511 LR 0.000250 Time 0.020506 -2022-12-06 11:35:08,563 - Epoch: [162][ 550/ 1200] Overall Loss 0.147965 Objective Loss 0.147965 LR 0.000250 Time 0.020480 -2022-12-06 11:35:08,755 - Epoch: [162][ 560/ 1200] Overall Loss 0.147890 Objective Loss 0.147890 LR 0.000250 Time 0.020454 -2022-12-06 11:35:08,947 - Epoch: [162][ 570/ 1200] Overall Loss 0.148126 Objective Loss 0.148126 LR 0.000250 Time 0.020431 -2022-12-06 11:35:09,138 - Epoch: [162][ 580/ 1200] Overall Loss 0.147675 Objective Loss 0.147675 LR 0.000250 Time 0.020409 -2022-12-06 11:35:09,331 - Epoch: [162][ 590/ 1200] Overall Loss 0.147325 Objective Loss 0.147325 LR 0.000250 Time 0.020388 -2022-12-06 11:35:09,523 - Epoch: [162][ 600/ 1200] Overall Loss 0.147453 Objective Loss 0.147453 LR 0.000250 Time 0.020367 -2022-12-06 11:35:09,714 - Epoch: [162][ 610/ 1200] Overall Loss 0.147354 Objective Loss 0.147354 LR 0.000250 Time 0.020346 -2022-12-06 11:35:09,906 - Epoch: [162][ 620/ 1200] Overall Loss 0.147335 Objective Loss 0.147335 LR 0.000250 Time 0.020327 -2022-12-06 11:35:10,098 - Epoch: [162][ 630/ 1200] Overall Loss 0.147154 Objective Loss 0.147154 LR 0.000250 Time 0.020308 -2022-12-06 11:35:10,290 - Epoch: [162][ 640/ 1200] Overall Loss 0.147313 Objective Loss 0.147313 LR 0.000250 Time 0.020289 -2022-12-06 11:35:10,482 - Epoch: [162][ 650/ 1200] Overall Loss 0.147536 Objective Loss 0.147536 LR 0.000250 Time 0.020273 -2022-12-06 11:35:10,675 - Epoch: [162][ 660/ 1200] Overall Loss 0.147402 Objective Loss 0.147402 LR 0.000250 Time 0.020256 -2022-12-06 11:35:10,866 - Epoch: [162][ 670/ 1200] Overall Loss 0.147190 Objective Loss 0.147190 LR 0.000250 Time 0.020239 -2022-12-06 11:35:11,058 - Epoch: [162][ 680/ 1200] Overall Loss 0.147218 Objective Loss 0.147218 LR 0.000250 Time 0.020222 -2022-12-06 11:35:11,250 - Epoch: [162][ 690/ 1200] Overall Loss 0.147040 Objective Loss 0.147040 LR 0.000250 Time 0.020206 -2022-12-06 11:35:11,441 - Epoch: [162][ 700/ 1200] Overall Loss 0.147055 Objective Loss 0.147055 LR 0.000250 Time 0.020191 -2022-12-06 11:35:11,633 - Epoch: [162][ 710/ 1200] Overall Loss 0.146956 Objective Loss 0.146956 LR 0.000250 Time 0.020175 -2022-12-06 11:35:11,825 - Epoch: [162][ 720/ 1200] Overall Loss 0.146779 Objective Loss 0.146779 LR 0.000250 Time 0.020161 -2022-12-06 11:35:12,016 - Epoch: [162][ 730/ 1200] Overall Loss 0.146780 Objective Loss 0.146780 LR 0.000250 Time 0.020146 -2022-12-06 11:35:12,208 - Epoch: [162][ 740/ 1200] Overall Loss 0.146595 Objective Loss 0.146595 LR 0.000250 Time 0.020132 -2022-12-06 11:35:12,401 - Epoch: [162][ 750/ 1200] Overall Loss 0.146523 Objective Loss 0.146523 LR 0.000250 Time 0.020120 -2022-12-06 11:35:12,595 - Epoch: [162][ 760/ 1200] Overall Loss 0.146594 Objective Loss 0.146594 LR 0.000250 Time 0.020109 -2022-12-06 11:35:12,789 - Epoch: [162][ 770/ 1200] Overall Loss 0.146419 Objective Loss 0.146419 LR 0.000250 Time 0.020100 -2022-12-06 11:35:12,982 - Epoch: [162][ 780/ 1200] Overall Loss 0.146437 Objective Loss 0.146437 LR 0.000250 Time 0.020089 -2022-12-06 11:35:13,177 - Epoch: [162][ 790/ 1200] Overall Loss 0.146596 Objective Loss 0.146596 LR 0.000250 Time 0.020081 -2022-12-06 11:35:13,371 - Epoch: [162][ 800/ 1200] Overall Loss 0.146431 Objective Loss 0.146431 LR 0.000250 Time 0.020072 -2022-12-06 11:35:13,565 - Epoch: [162][ 810/ 1200] Overall Loss 0.146485 Objective Loss 0.146485 LR 0.000250 Time 0.020063 -2022-12-06 11:35:13,760 - Epoch: [162][ 820/ 1200] Overall Loss 0.146386 Objective Loss 0.146386 LR 0.000250 Time 0.020055 -2022-12-06 11:35:13,954 - Epoch: [162][ 830/ 1200] Overall Loss 0.146452 Objective Loss 0.146452 LR 0.000250 Time 0.020046 -2022-12-06 11:35:14,148 - Epoch: [162][ 840/ 1200] Overall Loss 0.146556 Objective Loss 0.146556 LR 0.000250 Time 0.020038 -2022-12-06 11:35:14,342 - Epoch: [162][ 850/ 1200] Overall Loss 0.146331 Objective Loss 0.146331 LR 0.000250 Time 0.020030 -2022-12-06 11:35:14,536 - Epoch: [162][ 860/ 1200] Overall Loss 0.146254 Objective Loss 0.146254 LR 0.000250 Time 0.020022 -2022-12-06 11:35:14,728 - Epoch: [162][ 870/ 1200] Overall Loss 0.146246 Objective Loss 0.146246 LR 0.000250 Time 0.020011 -2022-12-06 11:35:14,919 - Epoch: [162][ 880/ 1200] Overall Loss 0.146324 Objective Loss 0.146324 LR 0.000250 Time 0.020001 -2022-12-06 11:35:15,110 - Epoch: [162][ 890/ 1200] Overall Loss 0.146343 Objective Loss 0.146343 LR 0.000250 Time 0.019990 -2022-12-06 11:35:15,302 - Epoch: [162][ 900/ 1200] Overall Loss 0.146279 Objective Loss 0.146279 LR 0.000250 Time 0.019981 -2022-12-06 11:35:15,494 - Epoch: [162][ 910/ 1200] Overall Loss 0.146425 Objective Loss 0.146425 LR 0.000250 Time 0.019972 -2022-12-06 11:35:15,687 - Epoch: [162][ 920/ 1200] Overall Loss 0.146441 Objective Loss 0.146441 LR 0.000250 Time 0.019963 -2022-12-06 11:35:15,879 - Epoch: [162][ 930/ 1200] Overall Loss 0.146284 Objective Loss 0.146284 LR 0.000250 Time 0.019955 -2022-12-06 11:35:16,071 - Epoch: [162][ 940/ 1200] Overall Loss 0.146206 Objective Loss 0.146206 LR 0.000250 Time 0.019946 -2022-12-06 11:35:16,263 - Epoch: [162][ 950/ 1200] Overall Loss 0.146088 Objective Loss 0.146088 LR 0.000250 Time 0.019938 -2022-12-06 11:35:16,455 - Epoch: [162][ 960/ 1200] Overall Loss 0.146107 Objective Loss 0.146107 LR 0.000250 Time 0.019929 -2022-12-06 11:35:16,647 - Epoch: [162][ 970/ 1200] Overall Loss 0.146201 Objective Loss 0.146201 LR 0.000250 Time 0.019922 -2022-12-06 11:35:16,839 - Epoch: [162][ 980/ 1200] Overall Loss 0.146585 Objective Loss 0.146585 LR 0.000250 Time 0.019914 -2022-12-06 11:35:17,032 - Epoch: [162][ 990/ 1200] Overall Loss 0.146529 Objective Loss 0.146529 LR 0.000250 Time 0.019907 -2022-12-06 11:35:17,224 - Epoch: [162][ 1000/ 1200] Overall Loss 0.146755 Objective Loss 0.146755 LR 0.000250 Time 0.019899 -2022-12-06 11:35:17,416 - Epoch: [162][ 1010/ 1200] Overall Loss 0.146783 Objective Loss 0.146783 LR 0.000250 Time 0.019892 -2022-12-06 11:35:17,609 - Epoch: [162][ 1020/ 1200] Overall Loss 0.146887 Objective Loss 0.146887 LR 0.000250 Time 0.019885 -2022-12-06 11:35:17,800 - Epoch: [162][ 1030/ 1200] Overall Loss 0.146863 Objective Loss 0.146863 LR 0.000250 Time 0.019877 -2022-12-06 11:35:17,993 - Epoch: [162][ 1040/ 1200] Overall Loss 0.147037 Objective Loss 0.147037 LR 0.000250 Time 0.019871 -2022-12-06 11:35:18,185 - Epoch: [162][ 1050/ 1200] Overall Loss 0.147051 Objective Loss 0.147051 LR 0.000250 Time 0.019864 -2022-12-06 11:35:18,377 - Epoch: [162][ 1060/ 1200] Overall Loss 0.146997 Objective Loss 0.146997 LR 0.000250 Time 0.019857 -2022-12-06 11:35:18,568 - Epoch: [162][ 1070/ 1200] Overall Loss 0.146863 Objective Loss 0.146863 LR 0.000250 Time 0.019850 -2022-12-06 11:35:18,759 - Epoch: [162][ 1080/ 1200] Overall Loss 0.146866 Objective Loss 0.146866 LR 0.000250 Time 0.019843 -2022-12-06 11:35:18,952 - Epoch: [162][ 1090/ 1200] Overall Loss 0.146778 Objective Loss 0.146778 LR 0.000250 Time 0.019836 -2022-12-06 11:35:19,144 - Epoch: [162][ 1100/ 1200] Overall Loss 0.146806 Objective Loss 0.146806 LR 0.000250 Time 0.019830 -2022-12-06 11:35:19,336 - Epoch: [162][ 1110/ 1200] Overall Loss 0.146952 Objective Loss 0.146952 LR 0.000250 Time 0.019824 -2022-12-06 11:35:19,527 - Epoch: [162][ 1120/ 1200] Overall Loss 0.147141 Objective Loss 0.147141 LR 0.000250 Time 0.019818 -2022-12-06 11:35:19,719 - Epoch: [162][ 1130/ 1200] Overall Loss 0.147239 Objective Loss 0.147239 LR 0.000250 Time 0.019811 -2022-12-06 11:35:19,910 - Epoch: [162][ 1140/ 1200] Overall Loss 0.147219 Objective Loss 0.147219 LR 0.000250 Time 0.019805 -2022-12-06 11:35:20,101 - Epoch: [162][ 1150/ 1200] Overall Loss 0.147306 Objective Loss 0.147306 LR 0.000250 Time 0.019798 -2022-12-06 11:35:20,294 - Epoch: [162][ 1160/ 1200] Overall Loss 0.147146 Objective Loss 0.147146 LR 0.000250 Time 0.019793 -2022-12-06 11:35:20,487 - Epoch: [162][ 1170/ 1200] Overall Loss 0.147148 Objective Loss 0.147148 LR 0.000250 Time 0.019789 -2022-12-06 11:35:20,682 - Epoch: [162][ 1180/ 1200] Overall Loss 0.147050 Objective Loss 0.147050 LR 0.000250 Time 0.019785 -2022-12-06 11:35:20,876 - Epoch: [162][ 1190/ 1200] Overall Loss 0.147010 Objective Loss 0.147010 LR 0.000250 Time 0.019782 -2022-12-06 11:35:21,101 - Epoch: [162][ 1200/ 1200] Overall Loss 0.147110 Objective Loss 0.147110 Top1 92.050209 Top5 99.581590 LR 0.000250 Time 0.019805 -2022-12-06 11:35:21,190 - --- validate (epoch=162)----------- -2022-12-06 11:35:21,190 - 34129 samples (256 per mini-batch) -2022-12-06 11:35:21,636 - Epoch: [162][ 10/ 134] Loss 0.232552 Top1 88.125000 Top5 98.554688 -2022-12-06 11:35:21,768 - Epoch: [162][ 20/ 134] Loss 0.238071 Top1 87.871094 Top5 98.554688 -2022-12-06 11:35:21,903 - Epoch: [162][ 30/ 134] Loss 0.239277 Top1 87.773438 Top5 98.502604 -2022-12-06 11:35:22,035 - Epoch: [162][ 40/ 134] Loss 0.239219 Top1 87.861328 Top5 98.447266 -2022-12-06 11:35:22,169 - Epoch: [162][ 50/ 134] Loss 0.237705 Top1 87.898438 Top5 98.500000 -2022-12-06 11:35:22,302 - Epoch: [162][ 60/ 134] Loss 0.237116 Top1 87.942708 Top5 98.522135 -2022-12-06 11:35:22,435 - Epoch: [162][ 70/ 134] Loss 0.237891 Top1 87.924107 Top5 98.498884 -2022-12-06 11:35:22,567 - Epoch: [162][ 80/ 134] Loss 0.235211 Top1 87.929688 Top5 98.530273 -2022-12-06 11:35:22,695 - Epoch: [162][ 90/ 134] Loss 0.234032 Top1 87.912326 Top5 98.502604 -2022-12-06 11:35:22,827 - Epoch: [162][ 100/ 134] Loss 0.233396 Top1 87.929688 Top5 98.531250 -2022-12-06 11:35:22,961 - Epoch: [162][ 110/ 134] Loss 0.233233 Top1 87.961648 Top5 98.519176 -2022-12-06 11:35:23,092 - Epoch: [162][ 120/ 134] Loss 0.231500 Top1 87.958984 Top5 98.528646 -2022-12-06 11:35:23,226 - Epoch: [162][ 130/ 134] Loss 0.232124 Top1 87.851562 Top5 98.512620 -2022-12-06 11:35:23,265 - Epoch: [162][ 134/ 134] Loss 0.233616 Top1 87.796302 Top5 98.502740 -2022-12-06 11:35:23,358 - ==> Top1: 87.796 Top5: 98.503 Loss: 0.234 - -2022-12-06 11:35:23,359 - ==> Confusion: -[[ 929 0 1 2 1 4 1 0 2 40 0 2 1 3 5 2 0 0 0 0 3] - [ 1 930 2 2 14 24 2 13 2 1 2 4 0 0 0 1 4 1 14 3 7] - [ 2 2 1016 17 4 3 18 8 1 3 7 2 1 0 1 3 0 1 2 4 8] - [ 0 1 15 950 1 1 0 2 0 1 9 1 4 2 12 0 1 2 13 0 5] - [ 11 3 1 0 964 1 1 1 0 5 1 0 2 2 9 6 4 2 1 1 5] - [ 0 7 0 2 5 999 2 14 2 4 1 9 3 9 2 2 1 1 0 2 4] - [ 1 0 5 2 0 3 1089 1 0 0 1 1 0 1 0 2 0 2 1 7 2] - [ 2 7 6 2 2 30 8 956 1 0 1 5 0 1 1 0 0 0 17 9 6] - [ 8 1 0 0 1 1 1 0 977 41 6 1 2 4 16 1 0 0 1 1 2] - [ 57 0 1 0 3 2 0 1 14 901 1 2 0 5 3 1 1 1 1 0 7] - [ 2 2 1 7 2 1 1 1 7 3 962 1 0 8 4 1 0 0 6 3 7] - [ 1 1 3 0 0 9 3 1 1 1 0 977 21 4 0 9 3 4 0 9 4] - [ 0 1 0 1 1 2 0 1 1 2 0 23 906 2 2 9 1 7 0 3 7] - [ 2 1 1 0 0 10 0 2 10 13 0 3 4 960 2 2 2 0 0 3 8] - [ 10 2 2 9 3 2 0 0 16 2 1 2 2 3 1066 0 0 1 6 0 3] - [ 1 0 1 0 0 0 4 0 1 1 1 6 9 1 0 1003 3 8 0 3 1] - [ 0 0 3 1 2 3 1 0 1 0 1 1 3 1 0 11 1035 1 0 2 6] - [ 5 0 1 0 0 1 0 0 0 3 0 5 12 1 3 17 0 985 0 1 2] - [ 3 4 3 9 1 2 0 21 1 1 2 2 5 0 9 0 0 2 940 0 3] - [ 2 4 3 1 0 6 5 4 0 1 1 10 8 4 0 2 2 1 1 1018 7] - [ 114 155 167 116 105 189 86 127 76 91 137 72 331 240 167 108 126 72 144 209 10394]] - -2022-12-06 11:35:23,931 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:35:23,931 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:35:23,937 - - -2022-12-06 11:35:23,937 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:35:24,989 - Epoch: [163][ 10/ 1200] Overall Loss 0.134352 Objective Loss 0.134352 LR 0.000250 Time 0.105059 -2022-12-06 11:35:25,194 - Epoch: [163][ 20/ 1200] Overall Loss 0.136908 Objective Loss 0.136908 LR 0.000250 Time 0.062763 -2022-12-06 11:35:25,390 - Epoch: [163][ 30/ 1200] Overall Loss 0.138556 Objective Loss 0.138556 LR 0.000250 Time 0.048345 -2022-12-06 11:35:25,588 - Epoch: [163][ 40/ 1200] Overall Loss 0.139710 Objective Loss 0.139710 LR 0.000250 Time 0.041199 -2022-12-06 11:35:25,783 - Epoch: [163][ 50/ 1200] Overall Loss 0.137482 Objective Loss 0.137482 LR 0.000250 Time 0.036857 -2022-12-06 11:35:25,982 - Epoch: [163][ 60/ 1200] Overall Loss 0.139855 Objective Loss 0.139855 LR 0.000250 Time 0.034012 -2022-12-06 11:35:26,176 - Epoch: [163][ 70/ 1200] Overall Loss 0.139049 Objective Loss 0.139049 LR 0.000250 Time 0.031930 -2022-12-06 11:35:26,376 - Epoch: [163][ 80/ 1200] Overall Loss 0.138483 Objective Loss 0.138483 LR 0.000250 Time 0.030423 -2022-12-06 11:35:26,571 - Epoch: [163][ 90/ 1200] Overall Loss 0.138524 Objective Loss 0.138524 LR 0.000250 Time 0.029201 -2022-12-06 11:35:26,769 - Epoch: [163][ 100/ 1200] Overall Loss 0.137939 Objective Loss 0.137939 LR 0.000250 Time 0.028263 -2022-12-06 11:35:26,964 - Epoch: [163][ 110/ 1200] Overall Loss 0.139049 Objective Loss 0.139049 LR 0.000250 Time 0.027461 -2022-12-06 11:35:27,163 - Epoch: [163][ 120/ 1200] Overall Loss 0.139359 Objective Loss 0.139359 LR 0.000250 Time 0.026822 -2022-12-06 11:35:27,357 - Epoch: [163][ 130/ 1200] Overall Loss 0.140077 Objective Loss 0.140077 LR 0.000250 Time 0.026252 -2022-12-06 11:35:27,556 - Epoch: [163][ 140/ 1200] Overall Loss 0.141618 Objective Loss 0.141618 LR 0.000250 Time 0.025791 -2022-12-06 11:35:27,751 - Epoch: [163][ 150/ 1200] Overall Loss 0.141833 Objective Loss 0.141833 LR 0.000250 Time 0.025366 -2022-12-06 11:35:27,948 - Epoch: [163][ 160/ 1200] Overall Loss 0.142973 Objective Loss 0.142973 LR 0.000250 Time 0.025013 -2022-12-06 11:35:28,143 - Epoch: [163][ 170/ 1200] Overall Loss 0.143105 Objective Loss 0.143105 LR 0.000250 Time 0.024681 -2022-12-06 11:35:28,342 - Epoch: [163][ 180/ 1200] Overall Loss 0.144209 Objective Loss 0.144209 LR 0.000250 Time 0.024415 -2022-12-06 11:35:28,537 - Epoch: [163][ 190/ 1200] Overall Loss 0.144654 Objective Loss 0.144654 LR 0.000250 Time 0.024150 -2022-12-06 11:35:28,730 - Epoch: [163][ 200/ 1200] Overall Loss 0.144987 Objective Loss 0.144987 LR 0.000250 Time 0.023909 -2022-12-06 11:35:28,923 - Epoch: [163][ 210/ 1200] Overall Loss 0.143951 Objective Loss 0.143951 LR 0.000250 Time 0.023685 -2022-12-06 11:35:29,116 - Epoch: [163][ 220/ 1200] Overall Loss 0.143721 Objective Loss 0.143721 LR 0.000250 Time 0.023482 -2022-12-06 11:35:29,308 - Epoch: [163][ 230/ 1200] Overall Loss 0.143167 Objective Loss 0.143167 LR 0.000250 Time 0.023296 -2022-12-06 11:35:29,501 - Epoch: [163][ 240/ 1200] Overall Loss 0.143190 Objective Loss 0.143190 LR 0.000250 Time 0.023127 -2022-12-06 11:35:29,692 - Epoch: [163][ 250/ 1200] Overall Loss 0.142940 Objective Loss 0.142940 LR 0.000250 Time 0.022965 -2022-12-06 11:35:29,885 - Epoch: [163][ 260/ 1200] Overall Loss 0.142332 Objective Loss 0.142332 LR 0.000250 Time 0.022820 -2022-12-06 11:35:30,076 - Epoch: [163][ 270/ 1200] Overall Loss 0.141383 Objective Loss 0.141383 LR 0.000250 Time 0.022682 -2022-12-06 11:35:30,269 - Epoch: [163][ 280/ 1200] Overall Loss 0.142489 Objective Loss 0.142489 LR 0.000250 Time 0.022558 -2022-12-06 11:35:30,462 - Epoch: [163][ 290/ 1200] Overall Loss 0.142670 Objective Loss 0.142670 LR 0.000250 Time 0.022443 -2022-12-06 11:35:30,654 - Epoch: [163][ 300/ 1200] Overall Loss 0.142980 Objective Loss 0.142980 LR 0.000250 Time 0.022334 -2022-12-06 11:35:30,846 - Epoch: [163][ 310/ 1200] Overall Loss 0.142684 Objective Loss 0.142684 LR 0.000250 Time 0.022230 -2022-12-06 11:35:31,038 - Epoch: [163][ 320/ 1200] Overall Loss 0.143451 Objective Loss 0.143451 LR 0.000250 Time 0.022134 -2022-12-06 11:35:31,229 - Epoch: [163][ 330/ 1200] Overall Loss 0.143607 Objective Loss 0.143607 LR 0.000250 Time 0.022042 -2022-12-06 11:35:31,422 - Epoch: [163][ 340/ 1200] Overall Loss 0.143632 Objective Loss 0.143632 LR 0.000250 Time 0.021959 -2022-12-06 11:35:31,615 - Epoch: [163][ 350/ 1200] Overall Loss 0.143722 Objective Loss 0.143722 LR 0.000250 Time 0.021880 -2022-12-06 11:35:31,808 - Epoch: [163][ 360/ 1200] Overall Loss 0.143466 Objective Loss 0.143466 LR 0.000250 Time 0.021808 -2022-12-06 11:35:32,001 - Epoch: [163][ 370/ 1200] Overall Loss 0.143403 Objective Loss 0.143403 LR 0.000250 Time 0.021739 -2022-12-06 11:35:32,194 - Epoch: [163][ 380/ 1200] Overall Loss 0.143154 Objective Loss 0.143154 LR 0.000250 Time 0.021672 -2022-12-06 11:35:32,386 - Epoch: [163][ 390/ 1200] Overall Loss 0.142962 Objective Loss 0.142962 LR 0.000250 Time 0.021608 -2022-12-06 11:35:32,579 - Epoch: [163][ 400/ 1200] Overall Loss 0.142963 Objective Loss 0.142963 LR 0.000250 Time 0.021550 -2022-12-06 11:35:32,772 - Epoch: [163][ 410/ 1200] Overall Loss 0.142887 Objective Loss 0.142887 LR 0.000250 Time 0.021493 -2022-12-06 11:35:32,965 - Epoch: [163][ 420/ 1200] Overall Loss 0.143199 Objective Loss 0.143199 LR 0.000250 Time 0.021440 -2022-12-06 11:35:33,158 - Epoch: [163][ 430/ 1200] Overall Loss 0.143052 Objective Loss 0.143052 LR 0.000250 Time 0.021388 -2022-12-06 11:35:33,350 - Epoch: [163][ 440/ 1200] Overall Loss 0.143185 Objective Loss 0.143185 LR 0.000250 Time 0.021338 -2022-12-06 11:35:33,543 - Epoch: [163][ 450/ 1200] Overall Loss 0.143640 Objective Loss 0.143640 LR 0.000250 Time 0.021292 -2022-12-06 11:35:33,736 - Epoch: [163][ 460/ 1200] Overall Loss 0.143563 Objective Loss 0.143563 LR 0.000250 Time 0.021246 -2022-12-06 11:35:33,928 - Epoch: [163][ 470/ 1200] Overall Loss 0.143898 Objective Loss 0.143898 LR 0.000250 Time 0.021203 -2022-12-06 11:35:34,122 - Epoch: [163][ 480/ 1200] Overall Loss 0.143661 Objective Loss 0.143661 LR 0.000250 Time 0.021163 -2022-12-06 11:35:34,315 - Epoch: [163][ 490/ 1200] Overall Loss 0.143623 Objective Loss 0.143623 LR 0.000250 Time 0.021124 -2022-12-06 11:35:34,508 - Epoch: [163][ 500/ 1200] Overall Loss 0.143509 Objective Loss 0.143509 LR 0.000250 Time 0.021087 -2022-12-06 11:35:34,700 - Epoch: [163][ 510/ 1200] Overall Loss 0.143215 Objective Loss 0.143215 LR 0.000250 Time 0.021049 -2022-12-06 11:35:34,894 - Epoch: [163][ 520/ 1200] Overall Loss 0.143340 Objective Loss 0.143340 LR 0.000250 Time 0.021016 -2022-12-06 11:35:35,088 - Epoch: [163][ 530/ 1200] Overall Loss 0.143201 Objective Loss 0.143201 LR 0.000250 Time 0.020983 -2022-12-06 11:35:35,282 - Epoch: [163][ 540/ 1200] Overall Loss 0.142850 Objective Loss 0.142850 LR 0.000250 Time 0.020953 -2022-12-06 11:35:35,474 - Epoch: [163][ 550/ 1200] Overall Loss 0.142588 Objective Loss 0.142588 LR 0.000250 Time 0.020922 -2022-12-06 11:35:35,667 - Epoch: [163][ 560/ 1200] Overall Loss 0.143186 Objective Loss 0.143186 LR 0.000250 Time 0.020891 -2022-12-06 11:35:35,861 - Epoch: [163][ 570/ 1200] Overall Loss 0.143194 Objective Loss 0.143194 LR 0.000250 Time 0.020864 -2022-12-06 11:35:36,054 - Epoch: [163][ 580/ 1200] Overall Loss 0.143244 Objective Loss 0.143244 LR 0.000250 Time 0.020836 -2022-12-06 11:35:36,247 - Epoch: [163][ 590/ 1200] Overall Loss 0.143376 Objective Loss 0.143376 LR 0.000250 Time 0.020809 -2022-12-06 11:35:36,440 - Epoch: [163][ 600/ 1200] Overall Loss 0.143159 Objective Loss 0.143159 LR 0.000250 Time 0.020783 -2022-12-06 11:35:36,633 - Epoch: [163][ 610/ 1200] Overall Loss 0.142951 Objective Loss 0.142951 LR 0.000250 Time 0.020757 -2022-12-06 11:35:36,825 - Epoch: [163][ 620/ 1200] Overall Loss 0.142706 Objective Loss 0.142706 LR 0.000250 Time 0.020732 -2022-12-06 11:35:37,018 - Epoch: [163][ 630/ 1200] Overall Loss 0.143134 Objective Loss 0.143134 LR 0.000250 Time 0.020708 -2022-12-06 11:35:37,210 - Epoch: [163][ 640/ 1200] Overall Loss 0.143450 Objective Loss 0.143450 LR 0.000250 Time 0.020684 -2022-12-06 11:35:37,403 - Epoch: [163][ 650/ 1200] Overall Loss 0.143471 Objective Loss 0.143471 LR 0.000250 Time 0.020661 -2022-12-06 11:35:37,596 - Epoch: [163][ 660/ 1200] Overall Loss 0.143348 Objective Loss 0.143348 LR 0.000250 Time 0.020640 -2022-12-06 11:35:37,789 - Epoch: [163][ 670/ 1200] Overall Loss 0.143569 Objective Loss 0.143569 LR 0.000250 Time 0.020619 -2022-12-06 11:35:37,981 - Epoch: [163][ 680/ 1200] Overall Loss 0.143581 Objective Loss 0.143581 LR 0.000250 Time 0.020598 -2022-12-06 11:35:38,173 - Epoch: [163][ 690/ 1200] Overall Loss 0.143399 Objective Loss 0.143399 LR 0.000250 Time 0.020576 -2022-12-06 11:35:38,365 - Epoch: [163][ 700/ 1200] Overall Loss 0.143646 Objective Loss 0.143646 LR 0.000250 Time 0.020556 -2022-12-06 11:35:38,557 - Epoch: [163][ 710/ 1200] Overall Loss 0.144000 Objective Loss 0.144000 LR 0.000250 Time 0.020536 -2022-12-06 11:35:38,750 - Epoch: [163][ 720/ 1200] Overall Loss 0.144184 Objective Loss 0.144184 LR 0.000250 Time 0.020518 -2022-12-06 11:35:38,942 - Epoch: [163][ 730/ 1200] Overall Loss 0.144060 Objective Loss 0.144060 LR 0.000250 Time 0.020500 -2022-12-06 11:35:39,135 - Epoch: [163][ 740/ 1200] Overall Loss 0.144275 Objective Loss 0.144275 LR 0.000250 Time 0.020482 -2022-12-06 11:35:39,327 - Epoch: [163][ 750/ 1200] Overall Loss 0.144519 Objective Loss 0.144519 LR 0.000250 Time 0.020465 -2022-12-06 11:35:39,520 - Epoch: [163][ 760/ 1200] Overall Loss 0.144806 Objective Loss 0.144806 LR 0.000250 Time 0.020448 -2022-12-06 11:35:39,713 - Epoch: [163][ 770/ 1200] Overall Loss 0.144685 Objective Loss 0.144685 LR 0.000250 Time 0.020432 -2022-12-06 11:35:39,905 - Epoch: [163][ 780/ 1200] Overall Loss 0.144717 Objective Loss 0.144717 LR 0.000250 Time 0.020416 -2022-12-06 11:35:40,097 - Epoch: [163][ 790/ 1200] Overall Loss 0.144738 Objective Loss 0.144738 LR 0.000250 Time 0.020401 -2022-12-06 11:35:40,290 - Epoch: [163][ 800/ 1200] Overall Loss 0.144610 Objective Loss 0.144610 LR 0.000250 Time 0.020386 -2022-12-06 11:35:40,481 - Epoch: [163][ 810/ 1200] Overall Loss 0.144774 Objective Loss 0.144774 LR 0.000250 Time 0.020369 -2022-12-06 11:35:40,674 - Epoch: [163][ 820/ 1200] Overall Loss 0.144835 Objective Loss 0.144835 LR 0.000250 Time 0.020356 -2022-12-06 11:35:40,867 - Epoch: [163][ 830/ 1200] Overall Loss 0.144973 Objective Loss 0.144973 LR 0.000250 Time 0.020342 -2022-12-06 11:35:41,060 - Epoch: [163][ 840/ 1200] Overall Loss 0.145067 Objective Loss 0.145067 LR 0.000250 Time 0.020329 -2022-12-06 11:35:41,253 - Epoch: [163][ 850/ 1200] Overall Loss 0.145079 Objective Loss 0.145079 LR 0.000250 Time 0.020316 -2022-12-06 11:35:41,446 - Epoch: [163][ 860/ 1200] Overall Loss 0.145166 Objective Loss 0.145166 LR 0.000250 Time 0.020304 -2022-12-06 11:35:41,639 - Epoch: [163][ 870/ 1200] Overall Loss 0.145235 Objective Loss 0.145235 LR 0.000250 Time 0.020292 -2022-12-06 11:35:41,831 - Epoch: [163][ 880/ 1200] Overall Loss 0.145259 Objective Loss 0.145259 LR 0.000250 Time 0.020279 -2022-12-06 11:35:42,023 - Epoch: [163][ 890/ 1200] Overall Loss 0.145297 Objective Loss 0.145297 LR 0.000250 Time 0.020267 -2022-12-06 11:35:42,217 - Epoch: [163][ 900/ 1200] Overall Loss 0.145382 Objective Loss 0.145382 LR 0.000250 Time 0.020256 -2022-12-06 11:35:42,409 - Epoch: [163][ 910/ 1200] Overall Loss 0.145511 Objective Loss 0.145511 LR 0.000250 Time 0.020244 -2022-12-06 11:35:42,602 - Epoch: [163][ 920/ 1200] Overall Loss 0.145487 Objective Loss 0.145487 LR 0.000250 Time 0.020233 -2022-12-06 11:35:42,795 - Epoch: [163][ 930/ 1200] Overall Loss 0.145551 Objective Loss 0.145551 LR 0.000250 Time 0.020223 -2022-12-06 11:35:42,988 - Epoch: [163][ 940/ 1200] Overall Loss 0.145470 Objective Loss 0.145470 LR 0.000250 Time 0.020212 -2022-12-06 11:35:43,182 - Epoch: [163][ 950/ 1200] Overall Loss 0.145701 Objective Loss 0.145701 LR 0.000250 Time 0.020202 -2022-12-06 11:35:43,375 - Epoch: [163][ 960/ 1200] Overall Loss 0.145817 Objective Loss 0.145817 LR 0.000250 Time 0.020193 -2022-12-06 11:35:43,568 - Epoch: [163][ 970/ 1200] Overall Loss 0.145784 Objective Loss 0.145784 LR 0.000250 Time 0.020182 -2022-12-06 11:35:43,760 - Epoch: [163][ 980/ 1200] Overall Loss 0.145863 Objective Loss 0.145863 LR 0.000250 Time 0.020173 -2022-12-06 11:35:43,953 - Epoch: [163][ 990/ 1200] Overall Loss 0.145869 Objective Loss 0.145869 LR 0.000250 Time 0.020163 -2022-12-06 11:35:44,145 - Epoch: [163][ 1000/ 1200] Overall Loss 0.145781 Objective Loss 0.145781 LR 0.000250 Time 0.020153 -2022-12-06 11:35:44,337 - Epoch: [163][ 1010/ 1200] Overall Loss 0.145849 Objective Loss 0.145849 LR 0.000250 Time 0.020143 -2022-12-06 11:35:44,530 - Epoch: [163][ 1020/ 1200] Overall Loss 0.146066 Objective Loss 0.146066 LR 0.000250 Time 0.020134 -2022-12-06 11:35:44,722 - Epoch: [163][ 1030/ 1200] Overall Loss 0.146039 Objective Loss 0.146039 LR 0.000250 Time 0.020125 -2022-12-06 11:35:44,915 - Epoch: [163][ 1040/ 1200] Overall Loss 0.145916 Objective Loss 0.145916 LR 0.000250 Time 0.020116 -2022-12-06 11:35:45,108 - Epoch: [163][ 1050/ 1200] Overall Loss 0.146017 Objective Loss 0.146017 LR 0.000250 Time 0.020107 -2022-12-06 11:35:45,300 - Epoch: [163][ 1060/ 1200] Overall Loss 0.145884 Objective Loss 0.145884 LR 0.000250 Time 0.020099 -2022-12-06 11:35:45,493 - Epoch: [163][ 1070/ 1200] Overall Loss 0.145888 Objective Loss 0.145888 LR 0.000250 Time 0.020091 -2022-12-06 11:35:45,685 - Epoch: [163][ 1080/ 1200] Overall Loss 0.145963 Objective Loss 0.145963 LR 0.000250 Time 0.020082 -2022-12-06 11:35:45,877 - Epoch: [163][ 1090/ 1200] Overall Loss 0.145903 Objective Loss 0.145903 LR 0.000250 Time 0.020073 -2022-12-06 11:35:46,069 - Epoch: [163][ 1100/ 1200] Overall Loss 0.145994 Objective Loss 0.145994 LR 0.000250 Time 0.020065 -2022-12-06 11:35:46,262 - Epoch: [163][ 1110/ 1200] Overall Loss 0.146108 Objective Loss 0.146108 LR 0.000250 Time 0.020057 -2022-12-06 11:35:46,454 - Epoch: [163][ 1120/ 1200] Overall Loss 0.146063 Objective Loss 0.146063 LR 0.000250 Time 0.020050 -2022-12-06 11:35:46,646 - Epoch: [163][ 1130/ 1200] Overall Loss 0.145878 Objective Loss 0.145878 LR 0.000250 Time 0.020041 -2022-12-06 11:35:46,839 - Epoch: [163][ 1140/ 1200] Overall Loss 0.146137 Objective Loss 0.146137 LR 0.000250 Time 0.020034 -2022-12-06 11:35:47,032 - Epoch: [163][ 1150/ 1200] Overall Loss 0.146157 Objective Loss 0.146157 LR 0.000250 Time 0.020027 -2022-12-06 11:35:47,224 - Epoch: [163][ 1160/ 1200] Overall Loss 0.146144 Objective Loss 0.146144 LR 0.000250 Time 0.020020 -2022-12-06 11:35:47,418 - Epoch: [163][ 1170/ 1200] Overall Loss 0.146264 Objective Loss 0.146264 LR 0.000250 Time 0.020014 -2022-12-06 11:35:47,611 - Epoch: [163][ 1180/ 1200] Overall Loss 0.146186 Objective Loss 0.146186 LR 0.000250 Time 0.020007 -2022-12-06 11:35:47,804 - Epoch: [163][ 1190/ 1200] Overall Loss 0.146139 Objective Loss 0.146139 LR 0.000250 Time 0.020001 -2022-12-06 11:35:48,026 - Epoch: [163][ 1200/ 1200] Overall Loss 0.146108 Objective Loss 0.146108 Top1 89.330544 Top5 99.790795 LR 0.000250 Time 0.020019 -2022-12-06 11:35:48,115 - --- validate (epoch=163)----------- -2022-12-06 11:35:48,115 - 34129 samples (256 per mini-batch) -2022-12-06 11:35:48,564 - Epoch: [163][ 10/ 134] Loss 0.209597 Top1 87.968750 Top5 98.437500 -2022-12-06 11:35:48,695 - Epoch: [163][ 20/ 134] Loss 0.219211 Top1 88.105469 Top5 98.730469 -2022-12-06 11:35:48,823 - Epoch: [163][ 30/ 134] Loss 0.215839 Top1 88.138021 Top5 98.697917 -2022-12-06 11:35:48,952 - Epoch: [163][ 40/ 134] Loss 0.222756 Top1 88.007812 Top5 98.613281 -2022-12-06 11:35:49,081 - Epoch: [163][ 50/ 134] Loss 0.225257 Top1 87.914062 Top5 98.539062 -2022-12-06 11:35:49,206 - Epoch: [163][ 60/ 134] Loss 0.221555 Top1 87.955729 Top5 98.574219 -2022-12-06 11:35:49,332 - Epoch: [163][ 70/ 134] Loss 0.224868 Top1 87.767857 Top5 98.487723 -2022-12-06 11:35:49,461 - Epoch: [163][ 80/ 134] Loss 0.226581 Top1 87.812500 Top5 98.510742 -2022-12-06 11:35:49,585 - Epoch: [163][ 90/ 134] Loss 0.227393 Top1 87.808160 Top5 98.511285 -2022-12-06 11:35:49,713 - Epoch: [163][ 100/ 134] Loss 0.230909 Top1 87.835938 Top5 98.511719 -2022-12-06 11:35:49,843 - Epoch: [163][ 110/ 134] Loss 0.231072 Top1 87.826705 Top5 98.529830 -2022-12-06 11:35:49,974 - Epoch: [163][ 120/ 134] Loss 0.232569 Top1 87.851562 Top5 98.548177 -2022-12-06 11:35:50,100 - Epoch: [163][ 130/ 134] Loss 0.233793 Top1 87.809495 Top5 98.506611 -2022-12-06 11:35:50,137 - Epoch: [163][ 134/ 134] Loss 0.234457 Top1 87.822673 Top5 98.499810 -2022-12-06 11:35:50,231 - ==> Top1: 87.823 Top5: 98.500 Loss: 0.234 - -2022-12-06 11:35:50,232 - ==> Confusion: -[[ 910 1 1 2 5 3 2 2 6 41 0 2 1 3 6 1 1 0 3 0 6] - [ 0 956 2 2 5 18 1 8 1 0 2 4 0 1 0 1 4 0 10 3 9] - [ 4 2 1019 13 3 2 10 6 0 3 9 4 0 2 2 1 1 2 3 3 14] - [ 1 1 14 954 0 3 1 0 1 1 10 0 5 0 10 0 0 1 10 0 8] - [ 9 6 2 0 964 1 1 2 1 6 1 3 1 1 7 3 5 2 1 0 4] - [ 0 23 0 3 3 979 2 18 2 2 1 8 5 8 2 1 5 0 1 5 1] - [ 1 3 5 2 0 1 1083 5 1 0 0 0 0 1 0 3 0 1 2 8 2] - [ 0 9 5 2 3 27 8 956 1 1 2 4 1 0 1 0 0 0 18 11 5] - [ 7 5 0 0 1 3 0 1 987 30 12 1 1 4 5 1 2 0 2 1 1] - [ 51 0 2 0 4 3 0 2 29 884 1 2 0 9 4 1 0 1 1 0 7] - [ 1 0 2 3 1 1 1 3 7 1 974 1 1 8 3 1 0 0 4 1 6] - [ 2 1 3 0 0 10 3 2 2 1 1 989 14 3 0 4 3 4 0 7 2] - [ 0 1 1 2 0 2 0 1 1 1 0 33 896 2 1 5 1 5 1 8 8] - [ 1 1 1 0 1 7 0 3 10 10 5 1 4 968 0 1 3 0 0 1 6] - [ 8 6 3 13 3 2 0 0 24 1 1 2 2 3 1044 0 1 1 13 0 3] - [ 0 0 1 2 1 0 6 0 1 0 0 9 3 3 0 998 4 8 0 4 3] - [ 1 2 1 1 1 1 1 1 0 0 0 3 0 3 0 9 1041 0 2 1 4] - [ 4 1 2 3 0 1 1 1 2 3 0 10 12 1 1 14 0 975 0 1 4] - [ 1 3 2 11 1 3 0 23 0 1 4 3 2 1 4 0 0 2 943 1 3] - [ 1 3 1 2 0 4 3 7 0 1 1 12 9 4 1 2 3 1 3 1019 3] - [ 83 228 163 97 82 146 76 125 84 71 151 99 299 256 111 88 189 69 153 224 10432]] - -2022-12-06 11:35:50,809 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:35:50,809 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:35:50,816 - - -2022-12-06 11:35:50,816 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:35:51,756 - Epoch: [164][ 10/ 1200] Overall Loss 0.116438 Objective Loss 0.116438 LR 0.000250 Time 0.093927 -2022-12-06 11:35:51,950 - Epoch: [164][ 20/ 1200] Overall Loss 0.126327 Objective Loss 0.126327 LR 0.000250 Time 0.056643 -2022-12-06 11:35:52,143 - Epoch: [164][ 30/ 1200] Overall Loss 0.137666 Objective Loss 0.137666 LR 0.000250 Time 0.044182 -2022-12-06 11:35:52,336 - Epoch: [164][ 40/ 1200] Overall Loss 0.138101 Objective Loss 0.138101 LR 0.000250 Time 0.037953 -2022-12-06 11:35:52,529 - Epoch: [164][ 50/ 1200] Overall Loss 0.139561 Objective Loss 0.139561 LR 0.000250 Time 0.034208 -2022-12-06 11:35:52,722 - Epoch: [164][ 60/ 1200] Overall Loss 0.140014 Objective Loss 0.140014 LR 0.000250 Time 0.031709 -2022-12-06 11:35:52,913 - Epoch: [164][ 70/ 1200] Overall Loss 0.140824 Objective Loss 0.140824 LR 0.000250 Time 0.029906 -2022-12-06 11:35:53,106 - Epoch: [164][ 80/ 1200] Overall Loss 0.142722 Objective Loss 0.142722 LR 0.000250 Time 0.028567 -2022-12-06 11:35:53,298 - Epoch: [164][ 90/ 1200] Overall Loss 0.145593 Objective Loss 0.145593 LR 0.000250 Time 0.027526 -2022-12-06 11:35:53,491 - Epoch: [164][ 100/ 1200] Overall Loss 0.146034 Objective Loss 0.146034 LR 0.000250 Time 0.026690 -2022-12-06 11:35:53,683 - Epoch: [164][ 110/ 1200] Overall Loss 0.144522 Objective Loss 0.144522 LR 0.000250 Time 0.026011 -2022-12-06 11:35:53,876 - Epoch: [164][ 120/ 1200] Overall Loss 0.143954 Objective Loss 0.143954 LR 0.000250 Time 0.025444 -2022-12-06 11:35:54,069 - Epoch: [164][ 130/ 1200] Overall Loss 0.142556 Objective Loss 0.142556 LR 0.000250 Time 0.024965 -2022-12-06 11:35:54,261 - Epoch: [164][ 140/ 1200] Overall Loss 0.143044 Objective Loss 0.143044 LR 0.000250 Time 0.024555 -2022-12-06 11:35:54,454 - Epoch: [164][ 150/ 1200] Overall Loss 0.142714 Objective Loss 0.142714 LR 0.000250 Time 0.024197 -2022-12-06 11:35:54,647 - Epoch: [164][ 160/ 1200] Overall Loss 0.141967 Objective Loss 0.141967 LR 0.000250 Time 0.023885 -2022-12-06 11:35:54,839 - Epoch: [164][ 170/ 1200] Overall Loss 0.141787 Objective Loss 0.141787 LR 0.000250 Time 0.023607 -2022-12-06 11:35:55,031 - Epoch: [164][ 180/ 1200] Overall Loss 0.141880 Objective Loss 0.141880 LR 0.000250 Time 0.023361 -2022-12-06 11:35:55,224 - Epoch: [164][ 190/ 1200] Overall Loss 0.140965 Objective Loss 0.140965 LR 0.000250 Time 0.023145 -2022-12-06 11:35:55,416 - Epoch: [164][ 200/ 1200] Overall Loss 0.141013 Objective Loss 0.141013 LR 0.000250 Time 0.022944 -2022-12-06 11:35:55,608 - Epoch: [164][ 210/ 1200] Overall Loss 0.141446 Objective Loss 0.141446 LR 0.000250 Time 0.022766 -2022-12-06 11:35:55,801 - Epoch: [164][ 220/ 1200] Overall Loss 0.140773 Objective Loss 0.140773 LR 0.000250 Time 0.022605 -2022-12-06 11:35:55,993 - Epoch: [164][ 230/ 1200] Overall Loss 0.140422 Objective Loss 0.140422 LR 0.000250 Time 0.022451 -2022-12-06 11:35:56,185 - Epoch: [164][ 240/ 1200] Overall Loss 0.141435 Objective Loss 0.141435 LR 0.000250 Time 0.022316 -2022-12-06 11:35:56,378 - Epoch: [164][ 250/ 1200] Overall Loss 0.141430 Objective Loss 0.141430 LR 0.000250 Time 0.022191 -2022-12-06 11:35:56,570 - Epoch: [164][ 260/ 1200] Overall Loss 0.142183 Objective Loss 0.142183 LR 0.000250 Time 0.022077 -2022-12-06 11:35:56,763 - Epoch: [164][ 270/ 1200] Overall Loss 0.142215 Objective Loss 0.142215 LR 0.000250 Time 0.021969 -2022-12-06 11:35:56,955 - Epoch: [164][ 280/ 1200] Overall Loss 0.143648 Objective Loss 0.143648 LR 0.000250 Time 0.021869 -2022-12-06 11:35:57,147 - Epoch: [164][ 290/ 1200] Overall Loss 0.143530 Objective Loss 0.143530 LR 0.000250 Time 0.021777 -2022-12-06 11:35:57,339 - Epoch: [164][ 300/ 1200] Overall Loss 0.143167 Objective Loss 0.143167 LR 0.000250 Time 0.021689 -2022-12-06 11:35:57,532 - Epoch: [164][ 310/ 1200] Overall Loss 0.143257 Objective Loss 0.143257 LR 0.000250 Time 0.021609 -2022-12-06 11:35:57,724 - Epoch: [164][ 320/ 1200] Overall Loss 0.142678 Objective Loss 0.142678 LR 0.000250 Time 0.021533 -2022-12-06 11:35:57,917 - Epoch: [164][ 330/ 1200] Overall Loss 0.143188 Objective Loss 0.143188 LR 0.000250 Time 0.021463 -2022-12-06 11:35:58,109 - Epoch: [164][ 340/ 1200] Overall Loss 0.142846 Objective Loss 0.142846 LR 0.000250 Time 0.021394 -2022-12-06 11:35:58,301 - Epoch: [164][ 350/ 1200] Overall Loss 0.143003 Objective Loss 0.143003 LR 0.000250 Time 0.021331 -2022-12-06 11:35:58,494 - Epoch: [164][ 360/ 1200] Overall Loss 0.143660 Objective Loss 0.143660 LR 0.000250 Time 0.021272 -2022-12-06 11:35:58,686 - Epoch: [164][ 370/ 1200] Overall Loss 0.143521 Objective Loss 0.143521 LR 0.000250 Time 0.021216 -2022-12-06 11:35:58,879 - Epoch: [164][ 380/ 1200] Overall Loss 0.143879 Objective Loss 0.143879 LR 0.000250 Time 0.021162 -2022-12-06 11:35:59,071 - Epoch: [164][ 390/ 1200] Overall Loss 0.143703 Objective Loss 0.143703 LR 0.000250 Time 0.021112 -2022-12-06 11:35:59,263 - Epoch: [164][ 400/ 1200] Overall Loss 0.143364 Objective Loss 0.143364 LR 0.000250 Time 0.021062 -2022-12-06 11:35:59,455 - Epoch: [164][ 410/ 1200] Overall Loss 0.143425 Objective Loss 0.143425 LR 0.000250 Time 0.021016 -2022-12-06 11:35:59,648 - Epoch: [164][ 420/ 1200] Overall Loss 0.142863 Objective Loss 0.142863 LR 0.000250 Time 0.020974 -2022-12-06 11:35:59,841 - Epoch: [164][ 430/ 1200] Overall Loss 0.143161 Objective Loss 0.143161 LR 0.000250 Time 0.020933 -2022-12-06 11:36:00,033 - Epoch: [164][ 440/ 1200] Overall Loss 0.142859 Objective Loss 0.142859 LR 0.000250 Time 0.020893 -2022-12-06 11:36:00,225 - Epoch: [164][ 450/ 1200] Overall Loss 0.143129 Objective Loss 0.143129 LR 0.000250 Time 0.020855 -2022-12-06 11:36:00,419 - Epoch: [164][ 460/ 1200] Overall Loss 0.143071 Objective Loss 0.143071 LR 0.000250 Time 0.020820 -2022-12-06 11:36:00,611 - Epoch: [164][ 470/ 1200] Overall Loss 0.142764 Objective Loss 0.142764 LR 0.000250 Time 0.020786 -2022-12-06 11:36:00,804 - Epoch: [164][ 480/ 1200] Overall Loss 0.143133 Objective Loss 0.143133 LR 0.000250 Time 0.020753 -2022-12-06 11:36:00,997 - Epoch: [164][ 490/ 1200] Overall Loss 0.142810 Objective Loss 0.142810 LR 0.000250 Time 0.020722 -2022-12-06 11:36:01,189 - Epoch: [164][ 500/ 1200] Overall Loss 0.143063 Objective Loss 0.143063 LR 0.000250 Time 0.020691 -2022-12-06 11:36:01,381 - Epoch: [164][ 510/ 1200] Overall Loss 0.143112 Objective Loss 0.143112 LR 0.000250 Time 0.020661 -2022-12-06 11:36:01,574 - Epoch: [164][ 520/ 1200] Overall Loss 0.143475 Objective Loss 0.143475 LR 0.000250 Time 0.020634 -2022-12-06 11:36:01,767 - Epoch: [164][ 530/ 1200] Overall Loss 0.143463 Objective Loss 0.143463 LR 0.000250 Time 0.020607 -2022-12-06 11:36:01,960 - Epoch: [164][ 540/ 1200] Overall Loss 0.143214 Objective Loss 0.143214 LR 0.000250 Time 0.020582 -2022-12-06 11:36:02,152 - Epoch: [164][ 550/ 1200] Overall Loss 0.142969 Objective Loss 0.142969 LR 0.000250 Time 0.020557 -2022-12-06 11:36:02,345 - Epoch: [164][ 560/ 1200] Overall Loss 0.142914 Objective Loss 0.142914 LR 0.000250 Time 0.020532 -2022-12-06 11:36:02,538 - Epoch: [164][ 570/ 1200] Overall Loss 0.143143 Objective Loss 0.143143 LR 0.000250 Time 0.020510 -2022-12-06 11:36:02,731 - Epoch: [164][ 580/ 1200] Overall Loss 0.143098 Objective Loss 0.143098 LR 0.000250 Time 0.020488 -2022-12-06 11:36:02,923 - Epoch: [164][ 590/ 1200] Overall Loss 0.143145 Objective Loss 0.143145 LR 0.000250 Time 0.020466 -2022-12-06 11:36:03,116 - Epoch: [164][ 600/ 1200] Overall Loss 0.143219 Objective Loss 0.143219 LR 0.000250 Time 0.020445 -2022-12-06 11:36:03,309 - Epoch: [164][ 610/ 1200] Overall Loss 0.143251 Objective Loss 0.143251 LR 0.000250 Time 0.020425 -2022-12-06 11:36:03,501 - Epoch: [164][ 620/ 1200] Overall Loss 0.143512 Objective Loss 0.143512 LR 0.000250 Time 0.020405 -2022-12-06 11:36:03,694 - Epoch: [164][ 630/ 1200] Overall Loss 0.143412 Objective Loss 0.143412 LR 0.000250 Time 0.020386 -2022-12-06 11:36:03,887 - Epoch: [164][ 640/ 1200] Overall Loss 0.143772 Objective Loss 0.143772 LR 0.000250 Time 0.020368 -2022-12-06 11:36:04,079 - Epoch: [164][ 650/ 1200] Overall Loss 0.143902 Objective Loss 0.143902 LR 0.000250 Time 0.020350 -2022-12-06 11:36:04,272 - Epoch: [164][ 660/ 1200] Overall Loss 0.144157 Objective Loss 0.144157 LR 0.000250 Time 0.020333 -2022-12-06 11:36:04,464 - Epoch: [164][ 670/ 1200] Overall Loss 0.144556 Objective Loss 0.144556 LR 0.000250 Time 0.020315 -2022-12-06 11:36:04,657 - Epoch: [164][ 680/ 1200] Overall Loss 0.144826 Objective Loss 0.144826 LR 0.000250 Time 0.020299 -2022-12-06 11:36:04,849 - Epoch: [164][ 690/ 1200] Overall Loss 0.144579 Objective Loss 0.144579 LR 0.000250 Time 0.020282 -2022-12-06 11:36:05,041 - Epoch: [164][ 700/ 1200] Overall Loss 0.144622 Objective Loss 0.144622 LR 0.000250 Time 0.020266 -2022-12-06 11:36:05,233 - Epoch: [164][ 710/ 1200] Overall Loss 0.144718 Objective Loss 0.144718 LR 0.000250 Time 0.020251 -2022-12-06 11:36:05,425 - Epoch: [164][ 720/ 1200] Overall Loss 0.144654 Objective Loss 0.144654 LR 0.000250 Time 0.020235 -2022-12-06 11:36:05,617 - Epoch: [164][ 730/ 1200] Overall Loss 0.144551 Objective Loss 0.144551 LR 0.000250 Time 0.020220 -2022-12-06 11:36:05,810 - Epoch: [164][ 740/ 1200] Overall Loss 0.144825 Objective Loss 0.144825 LR 0.000250 Time 0.020207 -2022-12-06 11:36:06,002 - Epoch: [164][ 750/ 1200] Overall Loss 0.144942 Objective Loss 0.144942 LR 0.000250 Time 0.020193 -2022-12-06 11:36:06,195 - Epoch: [164][ 760/ 1200] Overall Loss 0.144935 Objective Loss 0.144935 LR 0.000250 Time 0.020180 -2022-12-06 11:36:06,387 - Epoch: [164][ 770/ 1200] Overall Loss 0.144842 Objective Loss 0.144842 LR 0.000250 Time 0.020167 -2022-12-06 11:36:06,579 - Epoch: [164][ 780/ 1200] Overall Loss 0.145173 Objective Loss 0.145173 LR 0.000250 Time 0.020154 -2022-12-06 11:36:06,771 - Epoch: [164][ 790/ 1200] Overall Loss 0.145370 Objective Loss 0.145370 LR 0.000250 Time 0.020141 -2022-12-06 11:36:06,964 - Epoch: [164][ 800/ 1200] Overall Loss 0.145441 Objective Loss 0.145441 LR 0.000250 Time 0.020129 -2022-12-06 11:36:07,155 - Epoch: [164][ 810/ 1200] Overall Loss 0.145497 Objective Loss 0.145497 LR 0.000250 Time 0.020117 -2022-12-06 11:36:07,348 - Epoch: [164][ 820/ 1200] Overall Loss 0.145433 Objective Loss 0.145433 LR 0.000250 Time 0.020106 -2022-12-06 11:36:07,541 - Epoch: [164][ 830/ 1200] Overall Loss 0.145334 Objective Loss 0.145334 LR 0.000250 Time 0.020095 -2022-12-06 11:36:07,734 - Epoch: [164][ 840/ 1200] Overall Loss 0.145158 Objective Loss 0.145158 LR 0.000250 Time 0.020085 -2022-12-06 11:36:07,926 - Epoch: [164][ 850/ 1200] Overall Loss 0.145082 Objective Loss 0.145082 LR 0.000250 Time 0.020074 -2022-12-06 11:36:08,118 - Epoch: [164][ 860/ 1200] Overall Loss 0.145027 Objective Loss 0.145027 LR 0.000250 Time 0.020064 -2022-12-06 11:36:08,311 - Epoch: [164][ 870/ 1200] Overall Loss 0.145097 Objective Loss 0.145097 LR 0.000250 Time 0.020054 -2022-12-06 11:36:08,503 - Epoch: [164][ 880/ 1200] Overall Loss 0.145075 Objective Loss 0.145075 LR 0.000250 Time 0.020044 -2022-12-06 11:36:08,696 - Epoch: [164][ 890/ 1200] Overall Loss 0.145163 Objective Loss 0.145163 LR 0.000250 Time 0.020034 -2022-12-06 11:36:08,888 - Epoch: [164][ 900/ 1200] Overall Loss 0.145165 Objective Loss 0.145165 LR 0.000250 Time 0.020025 -2022-12-06 11:36:09,080 - Epoch: [164][ 910/ 1200] Overall Loss 0.145351 Objective Loss 0.145351 LR 0.000250 Time 0.020015 -2022-12-06 11:36:09,272 - Epoch: [164][ 920/ 1200] Overall Loss 0.145339 Objective Loss 0.145339 LR 0.000250 Time 0.020006 -2022-12-06 11:36:09,465 - Epoch: [164][ 930/ 1200] Overall Loss 0.145603 Objective Loss 0.145603 LR 0.000250 Time 0.019997 -2022-12-06 11:36:09,657 - Epoch: [164][ 940/ 1200] Overall Loss 0.145995 Objective Loss 0.145995 LR 0.000250 Time 0.019988 -2022-12-06 11:36:09,850 - Epoch: [164][ 950/ 1200] Overall Loss 0.145954 Objective Loss 0.145954 LR 0.000250 Time 0.019980 -2022-12-06 11:36:10,043 - Epoch: [164][ 960/ 1200] Overall Loss 0.145813 Objective Loss 0.145813 LR 0.000250 Time 0.019973 -2022-12-06 11:36:10,235 - Epoch: [164][ 970/ 1200] Overall Loss 0.145917 Objective Loss 0.145917 LR 0.000250 Time 0.019965 -2022-12-06 11:36:10,428 - Epoch: [164][ 980/ 1200] Overall Loss 0.146187 Objective Loss 0.146187 LR 0.000250 Time 0.019957 -2022-12-06 11:36:10,620 - Epoch: [164][ 990/ 1200] Overall Loss 0.145977 Objective Loss 0.145977 LR 0.000250 Time 0.019949 -2022-12-06 11:36:10,813 - Epoch: [164][ 1000/ 1200] Overall Loss 0.145984 Objective Loss 0.145984 LR 0.000250 Time 0.019942 -2022-12-06 11:36:11,004 - Epoch: [164][ 1010/ 1200] Overall Loss 0.146051 Objective Loss 0.146051 LR 0.000250 Time 0.019933 -2022-12-06 11:36:11,197 - Epoch: [164][ 1020/ 1200] Overall Loss 0.146152 Objective Loss 0.146152 LR 0.000250 Time 0.019926 -2022-12-06 11:36:11,390 - Epoch: [164][ 1030/ 1200] Overall Loss 0.146326 Objective Loss 0.146326 LR 0.000250 Time 0.019919 -2022-12-06 11:36:11,582 - Epoch: [164][ 1040/ 1200] Overall Loss 0.146310 Objective Loss 0.146310 LR 0.000250 Time 0.019912 -2022-12-06 11:36:11,775 - Epoch: [164][ 1050/ 1200] Overall Loss 0.146542 Objective Loss 0.146542 LR 0.000250 Time 0.019906 -2022-12-06 11:36:11,967 - Epoch: [164][ 1060/ 1200] Overall Loss 0.146384 Objective Loss 0.146384 LR 0.000250 Time 0.019899 -2022-12-06 11:36:12,160 - Epoch: [164][ 1070/ 1200] Overall Loss 0.146345 Objective Loss 0.146345 LR 0.000250 Time 0.019893 -2022-12-06 11:36:12,353 - Epoch: [164][ 1080/ 1200] Overall Loss 0.146212 Objective Loss 0.146212 LR 0.000250 Time 0.019886 -2022-12-06 11:36:12,545 - Epoch: [164][ 1090/ 1200] Overall Loss 0.146197 Objective Loss 0.146197 LR 0.000250 Time 0.019880 -2022-12-06 11:36:12,738 - Epoch: [164][ 1100/ 1200] Overall Loss 0.146269 Objective Loss 0.146269 LR 0.000250 Time 0.019874 -2022-12-06 11:36:12,931 - Epoch: [164][ 1110/ 1200] Overall Loss 0.146276 Objective Loss 0.146276 LR 0.000250 Time 0.019868 -2022-12-06 11:36:13,123 - Epoch: [164][ 1120/ 1200] Overall Loss 0.146121 Objective Loss 0.146121 LR 0.000250 Time 0.019862 -2022-12-06 11:36:13,316 - Epoch: [164][ 1130/ 1200] Overall Loss 0.145998 Objective Loss 0.145998 LR 0.000250 Time 0.019856 -2022-12-06 11:36:13,508 - Epoch: [164][ 1140/ 1200] Overall Loss 0.146066 Objective Loss 0.146066 LR 0.000250 Time 0.019851 -2022-12-06 11:36:13,701 - Epoch: [164][ 1150/ 1200] Overall Loss 0.146104 Objective Loss 0.146104 LR 0.000250 Time 0.019845 -2022-12-06 11:36:13,894 - Epoch: [164][ 1160/ 1200] Overall Loss 0.146229 Objective Loss 0.146229 LR 0.000250 Time 0.019839 -2022-12-06 11:36:14,085 - Epoch: [164][ 1170/ 1200] Overall Loss 0.146406 Objective Loss 0.146406 LR 0.000250 Time 0.019833 -2022-12-06 11:36:14,278 - Epoch: [164][ 1180/ 1200] Overall Loss 0.146384 Objective Loss 0.146384 LR 0.000250 Time 0.019828 -2022-12-06 11:36:14,471 - Epoch: [164][ 1190/ 1200] Overall Loss 0.146568 Objective Loss 0.146568 LR 0.000250 Time 0.019823 -2022-12-06 11:36:14,703 - Epoch: [164][ 1200/ 1200] Overall Loss 0.146546 Objective Loss 0.146546 Top1 89.958159 Top5 99.372385 LR 0.000250 Time 0.019851 -2022-12-06 11:36:14,792 - --- validate (epoch=164)----------- -2022-12-06 11:36:14,793 - 34129 samples (256 per mini-batch) -2022-12-06 11:36:15,231 - Epoch: [164][ 10/ 134] Loss 0.182005 Top1 88.359375 Top5 98.671875 -2022-12-06 11:36:15,357 - Epoch: [164][ 20/ 134] Loss 0.210199 Top1 87.871094 Top5 98.593750 -2022-12-06 11:36:15,486 - Epoch: [164][ 30/ 134] Loss 0.221531 Top1 87.721354 Top5 98.606771 -2022-12-06 11:36:15,616 - Epoch: [164][ 40/ 134] Loss 0.231871 Top1 87.597656 Top5 98.486328 -2022-12-06 11:36:15,748 - Epoch: [164][ 50/ 134] Loss 0.231226 Top1 87.578125 Top5 98.468750 -2022-12-06 11:36:15,880 - Epoch: [164][ 60/ 134] Loss 0.236325 Top1 87.545573 Top5 98.476562 -2022-12-06 11:36:16,011 - Epoch: [164][ 70/ 134] Loss 0.234339 Top1 87.444196 Top5 98.465402 -2022-12-06 11:36:16,144 - Epoch: [164][ 80/ 134] Loss 0.234210 Top1 87.402344 Top5 98.461914 -2022-12-06 11:36:16,276 - Epoch: [164][ 90/ 134] Loss 0.235846 Top1 87.382812 Top5 98.467882 -2022-12-06 11:36:16,406 - Epoch: [164][ 100/ 134] Loss 0.234758 Top1 87.386719 Top5 98.445312 -2022-12-06 11:36:16,537 - Epoch: [164][ 110/ 134] Loss 0.234115 Top1 87.443182 Top5 98.483665 -2022-12-06 11:36:16,667 - Epoch: [164][ 120/ 134] Loss 0.234369 Top1 87.347005 Top5 98.473307 -2022-12-06 11:36:16,800 - Epoch: [164][ 130/ 134] Loss 0.233618 Top1 87.388822 Top5 98.491587 -2022-12-06 11:36:16,839 - Epoch: [164][ 134/ 134] Loss 0.234525 Top1 87.421255 Top5 98.499810 -2022-12-06 11:36:16,927 - ==> Top1: 87.421 Top5: 98.500 Loss: 0.235 - -2022-12-06 11:36:16,928 - ==> Confusion: -[[ 931 1 2 4 2 5 0 0 8 30 0 1 1 2 5 2 0 1 0 0 1] - [ 1 941 3 2 8 23 2 12 1 1 1 3 1 1 0 1 3 0 10 2 11] - [ 4 4 1021 12 4 2 14 5 0 1 6 4 1 3 2 1 1 1 4 3 10] - [ 2 1 22 946 1 1 0 1 1 1 10 1 3 1 10 0 1 1 12 0 5] - [ 9 2 1 0 963 4 1 2 0 5 1 1 1 2 12 3 5 3 2 2 1] - [ 1 8 0 1 3 1000 2 17 2 2 0 12 2 8 2 2 2 1 0 2 2] - [ 1 4 4 2 0 1 1080 6 0 0 0 1 0 2 0 5 1 4 2 5 0] - [ 2 5 6 1 4 32 10 953 0 0 0 4 0 0 0 1 0 1 23 9 3] - [ 3 2 1 1 1 1 1 1 971 45 7 1 0 9 16 1 1 0 2 0 0] - [ 51 0 1 0 5 2 0 2 17 897 1 2 0 12 3 1 0 0 1 0 6] - [ 2 1 3 2 1 2 1 2 12 1 962 1 1 14 2 1 2 0 3 1 5] - [ 3 2 3 0 1 10 4 2 1 1 0 983 17 3 1 7 3 5 0 2 3] - [ 0 1 0 4 1 2 1 1 0 1 1 26 897 3 0 5 3 12 1 3 7] - [ 1 1 1 1 0 6 0 2 10 12 3 2 3 968 0 1 2 1 0 2 7] - [ 7 6 2 10 2 2 0 0 13 2 0 1 2 3 1069 0 0 1 6 1 3] - [ 0 0 0 2 4 1 3 0 0 0 2 7 4 0 0 1002 5 8 0 3 2] - [ 1 1 0 2 2 2 1 1 1 0 0 4 4 2 0 10 1031 0 0 2 8] - [ 2 0 1 1 1 0 0 1 1 3 0 9 8 2 0 11 1 991 0 0 4] - [ 2 5 3 3 1 3 0 21 1 1 7 2 2 1 5 0 0 2 945 2 2] - [ 2 3 2 2 0 3 7 4 1 2 2 12 6 6 1 3 3 2 2 1013 4] - [ 130 166 191 108 107 159 76 129 84 82 190 100 303 304 149 97 147 88 143 203 10270]] - -2022-12-06 11:36:17,588 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:36:17,588 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:36:17,594 - - -2022-12-06 11:36:17,594 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:36:18,527 - Epoch: [165][ 10/ 1200] Overall Loss 0.146849 Objective Loss 0.146849 LR 0.000250 Time 0.093236 -2022-12-06 11:36:18,727 - Epoch: [165][ 20/ 1200] Overall Loss 0.145148 Objective Loss 0.145148 LR 0.000250 Time 0.056594 -2022-12-06 11:36:18,919 - Epoch: [165][ 30/ 1200] Overall Loss 0.139546 Objective Loss 0.139546 LR 0.000250 Time 0.044120 -2022-12-06 11:36:19,110 - Epoch: [165][ 40/ 1200] Overall Loss 0.138069 Objective Loss 0.138069 LR 0.000250 Time 0.037834 -2022-12-06 11:36:19,301 - Epoch: [165][ 50/ 1200] Overall Loss 0.138385 Objective Loss 0.138385 LR 0.000250 Time 0.034084 -2022-12-06 11:36:19,492 - Epoch: [165][ 60/ 1200] Overall Loss 0.139052 Objective Loss 0.139052 LR 0.000250 Time 0.031572 -2022-12-06 11:36:19,683 - Epoch: [165][ 70/ 1200] Overall Loss 0.138799 Objective Loss 0.138799 LR 0.000250 Time 0.029784 -2022-12-06 11:36:19,874 - Epoch: [165][ 80/ 1200] Overall Loss 0.138620 Objective Loss 0.138620 LR 0.000250 Time 0.028438 -2022-12-06 11:36:20,064 - Epoch: [165][ 90/ 1200] Overall Loss 0.139007 Objective Loss 0.139007 LR 0.000250 Time 0.027386 -2022-12-06 11:36:20,255 - Epoch: [165][ 100/ 1200] Overall Loss 0.139096 Objective Loss 0.139096 LR 0.000250 Time 0.026551 -2022-12-06 11:36:20,445 - Epoch: [165][ 110/ 1200] Overall Loss 0.139022 Objective Loss 0.139022 LR 0.000250 Time 0.025865 -2022-12-06 11:36:20,636 - Epoch: [165][ 120/ 1200] Overall Loss 0.140812 Objective Loss 0.140812 LR 0.000250 Time 0.025292 -2022-12-06 11:36:20,827 - Epoch: [165][ 130/ 1200] Overall Loss 0.141313 Objective Loss 0.141313 LR 0.000250 Time 0.024811 -2022-12-06 11:36:21,018 - Epoch: [165][ 140/ 1200] Overall Loss 0.142316 Objective Loss 0.142316 LR 0.000250 Time 0.024402 -2022-12-06 11:36:21,209 - Epoch: [165][ 150/ 1200] Overall Loss 0.141598 Objective Loss 0.141598 LR 0.000250 Time 0.024045 -2022-12-06 11:36:21,399 - Epoch: [165][ 160/ 1200] Overall Loss 0.143089 Objective Loss 0.143089 LR 0.000250 Time 0.023729 -2022-12-06 11:36:21,590 - Epoch: [165][ 170/ 1200] Overall Loss 0.143407 Objective Loss 0.143407 LR 0.000250 Time 0.023452 -2022-12-06 11:36:21,782 - Epoch: [165][ 180/ 1200] Overall Loss 0.144177 Objective Loss 0.144177 LR 0.000250 Time 0.023210 -2022-12-06 11:36:21,972 - Epoch: [165][ 190/ 1200] Overall Loss 0.143781 Objective Loss 0.143781 LR 0.000250 Time 0.022989 -2022-12-06 11:36:22,163 - Epoch: [165][ 200/ 1200] Overall Loss 0.143647 Objective Loss 0.143647 LR 0.000250 Time 0.022789 -2022-12-06 11:36:22,353 - Epoch: [165][ 210/ 1200] Overall Loss 0.142553 Objective Loss 0.142553 LR 0.000250 Time 0.022608 -2022-12-06 11:36:22,544 - Epoch: [165][ 220/ 1200] Overall Loss 0.141127 Objective Loss 0.141127 LR 0.000250 Time 0.022445 -2022-12-06 11:36:22,736 - Epoch: [165][ 230/ 1200] Overall Loss 0.141872 Objective Loss 0.141872 LR 0.000250 Time 0.022301 -2022-12-06 11:36:22,927 - Epoch: [165][ 240/ 1200] Overall Loss 0.141284 Objective Loss 0.141284 LR 0.000250 Time 0.022164 -2022-12-06 11:36:23,117 - Epoch: [165][ 250/ 1200] Overall Loss 0.141657 Objective Loss 0.141657 LR 0.000250 Time 0.022038 -2022-12-06 11:36:23,308 - Epoch: [165][ 260/ 1200] Overall Loss 0.142175 Objective Loss 0.142175 LR 0.000250 Time 0.021921 -2022-12-06 11:36:23,500 - Epoch: [165][ 270/ 1200] Overall Loss 0.142029 Objective Loss 0.142029 LR 0.000250 Time 0.021818 -2022-12-06 11:36:23,690 - Epoch: [165][ 280/ 1200] Overall Loss 0.141784 Objective Loss 0.141784 LR 0.000250 Time 0.021717 -2022-12-06 11:36:23,881 - Epoch: [165][ 290/ 1200] Overall Loss 0.141973 Objective Loss 0.141973 LR 0.000250 Time 0.021624 -2022-12-06 11:36:24,071 - Epoch: [165][ 300/ 1200] Overall Loss 0.142148 Objective Loss 0.142148 LR 0.000250 Time 0.021536 -2022-12-06 11:36:24,263 - Epoch: [165][ 310/ 1200] Overall Loss 0.142313 Objective Loss 0.142313 LR 0.000250 Time 0.021457 -2022-12-06 11:36:24,454 - Epoch: [165][ 320/ 1200] Overall Loss 0.142230 Objective Loss 0.142230 LR 0.000250 Time 0.021383 -2022-12-06 11:36:24,645 - Epoch: [165][ 330/ 1200] Overall Loss 0.142439 Objective Loss 0.142439 LR 0.000250 Time 0.021312 -2022-12-06 11:36:24,836 - Epoch: [165][ 340/ 1200] Overall Loss 0.142408 Objective Loss 0.142408 LR 0.000250 Time 0.021245 -2022-12-06 11:36:25,027 - Epoch: [165][ 350/ 1200] Overall Loss 0.142805 Objective Loss 0.142805 LR 0.000250 Time 0.021183 -2022-12-06 11:36:25,218 - Epoch: [165][ 360/ 1200] Overall Loss 0.142755 Objective Loss 0.142755 LR 0.000250 Time 0.021122 -2022-12-06 11:36:25,408 - Epoch: [165][ 370/ 1200] Overall Loss 0.143483 Objective Loss 0.143483 LR 0.000250 Time 0.021065 -2022-12-06 11:36:25,599 - Epoch: [165][ 380/ 1200] Overall Loss 0.142930 Objective Loss 0.142930 LR 0.000250 Time 0.021011 -2022-12-06 11:36:25,790 - Epoch: [165][ 390/ 1200] Overall Loss 0.142695 Objective Loss 0.142695 LR 0.000250 Time 0.020961 -2022-12-06 11:36:25,981 - Epoch: [165][ 400/ 1200] Overall Loss 0.143456 Objective Loss 0.143456 LR 0.000250 Time 0.020913 -2022-12-06 11:36:26,172 - Epoch: [165][ 410/ 1200] Overall Loss 0.143623 Objective Loss 0.143623 LR 0.000250 Time 0.020867 -2022-12-06 11:36:26,363 - Epoch: [165][ 420/ 1200] Overall Loss 0.143434 Objective Loss 0.143434 LR 0.000250 Time 0.020823 -2022-12-06 11:36:26,554 - Epoch: [165][ 430/ 1200] Overall Loss 0.143265 Objective Loss 0.143265 LR 0.000250 Time 0.020782 -2022-12-06 11:36:26,745 - Epoch: [165][ 440/ 1200] Overall Loss 0.143365 Objective Loss 0.143365 LR 0.000250 Time 0.020743 -2022-12-06 11:36:26,937 - Epoch: [165][ 450/ 1200] Overall Loss 0.142948 Objective Loss 0.142948 LR 0.000250 Time 0.020707 -2022-12-06 11:36:27,127 - Epoch: [165][ 460/ 1200] Overall Loss 0.142941 Objective Loss 0.142941 LR 0.000250 Time 0.020669 -2022-12-06 11:36:27,318 - Epoch: [165][ 470/ 1200] Overall Loss 0.143079 Objective Loss 0.143079 LR 0.000250 Time 0.020634 -2022-12-06 11:36:27,508 - Epoch: [165][ 480/ 1200] Overall Loss 0.143076 Objective Loss 0.143076 LR 0.000250 Time 0.020599 -2022-12-06 11:36:27,699 - Epoch: [165][ 490/ 1200] Overall Loss 0.143165 Objective Loss 0.143165 LR 0.000250 Time 0.020568 -2022-12-06 11:36:27,890 - Epoch: [165][ 500/ 1200] Overall Loss 0.143081 Objective Loss 0.143081 LR 0.000250 Time 0.020537 -2022-12-06 11:36:28,081 - Epoch: [165][ 510/ 1200] Overall Loss 0.143125 Objective Loss 0.143125 LR 0.000250 Time 0.020509 -2022-12-06 11:36:28,272 - Epoch: [165][ 520/ 1200] Overall Loss 0.143465 Objective Loss 0.143465 LR 0.000250 Time 0.020479 -2022-12-06 11:36:28,463 - Epoch: [165][ 530/ 1200] Overall Loss 0.143604 Objective Loss 0.143604 LR 0.000250 Time 0.020452 -2022-12-06 11:36:28,654 - Epoch: [165][ 540/ 1200] Overall Loss 0.143595 Objective Loss 0.143595 LR 0.000250 Time 0.020426 -2022-12-06 11:36:28,844 - Epoch: [165][ 550/ 1200] Overall Loss 0.143781 Objective Loss 0.143781 LR 0.000250 Time 0.020399 -2022-12-06 11:36:29,034 - Epoch: [165][ 560/ 1200] Overall Loss 0.144348 Objective Loss 0.144348 LR 0.000250 Time 0.020374 -2022-12-06 11:36:29,226 - Epoch: [165][ 570/ 1200] Overall Loss 0.144376 Objective Loss 0.144376 LR 0.000250 Time 0.020352 -2022-12-06 11:36:29,416 - Epoch: [165][ 580/ 1200] Overall Loss 0.144624 Objective Loss 0.144624 LR 0.000250 Time 0.020328 -2022-12-06 11:36:29,607 - Epoch: [165][ 590/ 1200] Overall Loss 0.144552 Objective Loss 0.144552 LR 0.000250 Time 0.020306 -2022-12-06 11:36:29,798 - Epoch: [165][ 600/ 1200] Overall Loss 0.144856 Objective Loss 0.144856 LR 0.000250 Time 0.020285 -2022-12-06 11:36:29,989 - Epoch: [165][ 610/ 1200] Overall Loss 0.144743 Objective Loss 0.144743 LR 0.000250 Time 0.020264 -2022-12-06 11:36:30,180 - Epoch: [165][ 620/ 1200] Overall Loss 0.144720 Objective Loss 0.144720 LR 0.000250 Time 0.020245 -2022-12-06 11:36:30,371 - Epoch: [165][ 630/ 1200] Overall Loss 0.144786 Objective Loss 0.144786 LR 0.000250 Time 0.020226 -2022-12-06 11:36:30,562 - Epoch: [165][ 640/ 1200] Overall Loss 0.145265 Objective Loss 0.145265 LR 0.000250 Time 0.020207 -2022-12-06 11:36:30,752 - Epoch: [165][ 650/ 1200] Overall Loss 0.145241 Objective Loss 0.145241 LR 0.000250 Time 0.020188 -2022-12-06 11:36:30,943 - Epoch: [165][ 660/ 1200] Overall Loss 0.145046 Objective Loss 0.145046 LR 0.000250 Time 0.020171 -2022-12-06 11:36:31,134 - Epoch: [165][ 670/ 1200] Overall Loss 0.144937 Objective Loss 0.144937 LR 0.000250 Time 0.020154 -2022-12-06 11:36:31,324 - Epoch: [165][ 680/ 1200] Overall Loss 0.145231 Objective Loss 0.145231 LR 0.000250 Time 0.020136 -2022-12-06 11:36:31,515 - Epoch: [165][ 690/ 1200] Overall Loss 0.145287 Objective Loss 0.145287 LR 0.000250 Time 0.020120 -2022-12-06 11:36:31,706 - Epoch: [165][ 700/ 1200] Overall Loss 0.145207 Objective Loss 0.145207 LR 0.000250 Time 0.020105 -2022-12-06 11:36:31,897 - Epoch: [165][ 710/ 1200] Overall Loss 0.144897 Objective Loss 0.144897 LR 0.000250 Time 0.020090 -2022-12-06 11:36:32,087 - Epoch: [165][ 720/ 1200] Overall Loss 0.144645 Objective Loss 0.144645 LR 0.000250 Time 0.020075 -2022-12-06 11:36:32,278 - Epoch: [165][ 730/ 1200] Overall Loss 0.144527 Objective Loss 0.144527 LR 0.000250 Time 0.020061 -2022-12-06 11:36:32,468 - Epoch: [165][ 740/ 1200] Overall Loss 0.144188 Objective Loss 0.144188 LR 0.000250 Time 0.020046 -2022-12-06 11:36:32,659 - Epoch: [165][ 750/ 1200] Overall Loss 0.144407 Objective Loss 0.144407 LR 0.000250 Time 0.020032 -2022-12-06 11:36:32,850 - Epoch: [165][ 760/ 1200] Overall Loss 0.144270 Objective Loss 0.144270 LR 0.000250 Time 0.020019 -2022-12-06 11:36:33,045 - Epoch: [165][ 770/ 1200] Overall Loss 0.144097 Objective Loss 0.144097 LR 0.000250 Time 0.020011 -2022-12-06 11:36:33,235 - Epoch: [165][ 780/ 1200] Overall Loss 0.144209 Objective Loss 0.144209 LR 0.000250 Time 0.019998 -2022-12-06 11:36:33,426 - Epoch: [165][ 790/ 1200] Overall Loss 0.144333 Objective Loss 0.144333 LR 0.000250 Time 0.019986 -2022-12-06 11:36:33,616 - Epoch: [165][ 800/ 1200] Overall Loss 0.144119 Objective Loss 0.144119 LR 0.000250 Time 0.019973 -2022-12-06 11:36:33,806 - Epoch: [165][ 810/ 1200] Overall Loss 0.144387 Objective Loss 0.144387 LR 0.000250 Time 0.019960 -2022-12-06 11:36:33,997 - Epoch: [165][ 820/ 1200] Overall Loss 0.144576 Objective Loss 0.144576 LR 0.000250 Time 0.019949 -2022-12-06 11:36:34,188 - Epoch: [165][ 830/ 1200] Overall Loss 0.144465 Objective Loss 0.144465 LR 0.000250 Time 0.019938 -2022-12-06 11:36:34,379 - Epoch: [165][ 840/ 1200] Overall Loss 0.144596 Objective Loss 0.144596 LR 0.000250 Time 0.019927 -2022-12-06 11:36:34,569 - Epoch: [165][ 850/ 1200] Overall Loss 0.144719 Objective Loss 0.144719 LR 0.000250 Time 0.019916 -2022-12-06 11:36:34,760 - Epoch: [165][ 860/ 1200] Overall Loss 0.144674 Objective Loss 0.144674 LR 0.000250 Time 0.019905 -2022-12-06 11:36:34,950 - Epoch: [165][ 870/ 1200] Overall Loss 0.144530 Objective Loss 0.144530 LR 0.000250 Time 0.019894 -2022-12-06 11:36:35,140 - Epoch: [165][ 880/ 1200] Overall Loss 0.144712 Objective Loss 0.144712 LR 0.000250 Time 0.019883 -2022-12-06 11:36:35,330 - Epoch: [165][ 890/ 1200] Overall Loss 0.144646 Objective Loss 0.144646 LR 0.000250 Time 0.019873 -2022-12-06 11:36:35,521 - Epoch: [165][ 900/ 1200] Overall Loss 0.144658 Objective Loss 0.144658 LR 0.000250 Time 0.019864 -2022-12-06 11:36:35,712 - Epoch: [165][ 910/ 1200] Overall Loss 0.144768 Objective Loss 0.144768 LR 0.000250 Time 0.019855 -2022-12-06 11:36:35,903 - Epoch: [165][ 920/ 1200] Overall Loss 0.144655 Objective Loss 0.144655 LR 0.000250 Time 0.019846 -2022-12-06 11:36:36,092 - Epoch: [165][ 930/ 1200] Overall Loss 0.144612 Objective Loss 0.144612 LR 0.000250 Time 0.019836 -2022-12-06 11:36:36,283 - Epoch: [165][ 940/ 1200] Overall Loss 0.144384 Objective Loss 0.144384 LR 0.000250 Time 0.019827 -2022-12-06 11:36:36,474 - Epoch: [165][ 950/ 1200] Overall Loss 0.144363 Objective Loss 0.144363 LR 0.000250 Time 0.019819 -2022-12-06 11:36:36,664 - Epoch: [165][ 960/ 1200] Overall Loss 0.144331 Objective Loss 0.144331 LR 0.000250 Time 0.019810 -2022-12-06 11:36:36,857 - Epoch: [165][ 970/ 1200] Overall Loss 0.144193 Objective Loss 0.144193 LR 0.000250 Time 0.019805 -2022-12-06 11:36:37,051 - Epoch: [165][ 980/ 1200] Overall Loss 0.144354 Objective Loss 0.144354 LR 0.000250 Time 0.019800 -2022-12-06 11:36:37,244 - Epoch: [165][ 990/ 1200] Overall Loss 0.144538 Objective Loss 0.144538 LR 0.000250 Time 0.019794 -2022-12-06 11:36:37,438 - Epoch: [165][ 1000/ 1200] Overall Loss 0.144483 Objective Loss 0.144483 LR 0.000250 Time 0.019789 -2022-12-06 11:36:37,631 - Epoch: [165][ 1010/ 1200] Overall Loss 0.144430 Objective Loss 0.144430 LR 0.000250 Time 0.019784 -2022-12-06 11:36:37,825 - Epoch: [165][ 1020/ 1200] Overall Loss 0.144346 Objective Loss 0.144346 LR 0.000250 Time 0.019779 -2022-12-06 11:36:38,019 - Epoch: [165][ 1030/ 1200] Overall Loss 0.144553 Objective Loss 0.144553 LR 0.000250 Time 0.019775 -2022-12-06 11:36:38,213 - Epoch: [165][ 1040/ 1200] Overall Loss 0.144447 Objective Loss 0.144447 LR 0.000250 Time 0.019771 -2022-12-06 11:36:38,406 - Epoch: [165][ 1050/ 1200] Overall Loss 0.144481 Objective Loss 0.144481 LR 0.000250 Time 0.019766 -2022-12-06 11:36:38,601 - Epoch: [165][ 1060/ 1200] Overall Loss 0.144526 Objective Loss 0.144526 LR 0.000250 Time 0.019763 -2022-12-06 11:36:38,794 - Epoch: [165][ 1070/ 1200] Overall Loss 0.144551 Objective Loss 0.144551 LR 0.000250 Time 0.019759 -2022-12-06 11:36:38,988 - Epoch: [165][ 1080/ 1200] Overall Loss 0.144634 Objective Loss 0.144634 LR 0.000250 Time 0.019754 -2022-12-06 11:36:39,182 - Epoch: [165][ 1090/ 1200] Overall Loss 0.144573 Objective Loss 0.144573 LR 0.000250 Time 0.019750 -2022-12-06 11:36:39,375 - Epoch: [165][ 1100/ 1200] Overall Loss 0.144371 Objective Loss 0.144371 LR 0.000250 Time 0.019747 -2022-12-06 11:36:39,569 - Epoch: [165][ 1110/ 1200] Overall Loss 0.144390 Objective Loss 0.144390 LR 0.000250 Time 0.019743 -2022-12-06 11:36:39,762 - Epoch: [165][ 1120/ 1200] Overall Loss 0.144355 Objective Loss 0.144355 LR 0.000250 Time 0.019738 -2022-12-06 11:36:39,956 - Epoch: [165][ 1130/ 1200] Overall Loss 0.144605 Objective Loss 0.144605 LR 0.000250 Time 0.019735 -2022-12-06 11:36:40,150 - Epoch: [165][ 1140/ 1200] Overall Loss 0.144942 Objective Loss 0.144942 LR 0.000250 Time 0.019732 -2022-12-06 11:36:40,344 - Epoch: [165][ 1150/ 1200] Overall Loss 0.144996 Objective Loss 0.144996 LR 0.000250 Time 0.019728 -2022-12-06 11:36:40,537 - Epoch: [165][ 1160/ 1200] Overall Loss 0.144846 Objective Loss 0.144846 LR 0.000250 Time 0.019724 -2022-12-06 11:36:40,730 - Epoch: [165][ 1170/ 1200] Overall Loss 0.145144 Objective Loss 0.145144 LR 0.000250 Time 0.019720 -2022-12-06 11:36:40,924 - Epoch: [165][ 1180/ 1200] Overall Loss 0.145232 Objective Loss 0.145232 LR 0.000250 Time 0.019717 -2022-12-06 11:36:41,118 - Epoch: [165][ 1190/ 1200] Overall Loss 0.145314 Objective Loss 0.145314 LR 0.000250 Time 0.019713 -2022-12-06 11:36:41,341 - Epoch: [165][ 1200/ 1200] Overall Loss 0.145269 Objective Loss 0.145269 Top1 89.539749 Top5 98.953975 LR 0.000250 Time 0.019734 -2022-12-06 11:36:41,429 - --- validate (epoch=165)----------- -2022-12-06 11:36:41,430 - 34129 samples (256 per mini-batch) -2022-12-06 11:36:41,873 - Epoch: [165][ 10/ 134] Loss 0.238289 Top1 87.734375 Top5 98.593750 -2022-12-06 11:36:42,008 - Epoch: [165][ 20/ 134] Loss 0.246302 Top1 87.773438 Top5 98.300781 -2022-12-06 11:36:42,150 - Epoch: [165][ 30/ 134] Loss 0.232996 Top1 87.786458 Top5 98.437500 -2022-12-06 11:36:42,298 - Epoch: [165][ 40/ 134] Loss 0.230313 Top1 87.861328 Top5 98.476562 -2022-12-06 11:36:42,440 - Epoch: [165][ 50/ 134] Loss 0.227842 Top1 87.984375 Top5 98.492188 -2022-12-06 11:36:42,589 - Epoch: [165][ 60/ 134] Loss 0.228126 Top1 87.968750 Top5 98.574219 -2022-12-06 11:36:42,729 - Epoch: [165][ 70/ 134] Loss 0.230325 Top1 88.080357 Top5 98.582589 -2022-12-06 11:36:42,878 - Epoch: [165][ 80/ 134] Loss 0.228543 Top1 88.183594 Top5 98.588867 -2022-12-06 11:36:43,019 - Epoch: [165][ 90/ 134] Loss 0.232522 Top1 88.055556 Top5 98.593750 -2022-12-06 11:36:43,167 - Epoch: [165][ 100/ 134] Loss 0.230784 Top1 88.101562 Top5 98.582031 -2022-12-06 11:36:43,310 - Epoch: [165][ 110/ 134] Loss 0.229330 Top1 88.121449 Top5 98.597301 -2022-12-06 11:36:43,460 - Epoch: [165][ 120/ 134] Loss 0.230415 Top1 87.998047 Top5 98.613281 -2022-12-06 11:36:43,597 - Epoch: [165][ 130/ 134] Loss 0.231401 Top1 87.962740 Top5 98.584736 -2022-12-06 11:36:43,635 - Epoch: [165][ 134/ 134] Loss 0.229865 Top1 87.983826 Top5 98.578921 -2022-12-06 11:36:43,728 - ==> Top1: 87.984 Top5: 98.579 Loss: 0.230 - -2022-12-06 11:36:43,729 - ==> Confusion: -[[ 923 0 1 2 4 7 1 0 8 35 0 2 2 3 2 1 0 1 2 0 2] - [ 1 954 2 2 6 14 2 6 2 0 5 3 1 1 0 0 4 1 11 2 10] - [ 5 1 1010 13 3 2 23 9 0 2 4 3 2 1 2 2 1 3 4 3 10] - [ 1 1 15 948 1 3 1 0 1 1 8 0 3 2 11 0 0 1 15 0 8] - [ 9 3 1 0 968 1 1 1 0 6 1 2 0 2 9 4 4 3 1 1 3] - [ 1 18 0 2 4 978 3 19 3 3 1 13 2 10 1 1 1 1 0 4 4] - [ 1 3 4 1 0 0 1083 4 1 0 1 1 0 1 0 4 0 3 2 7 2] - [ 2 8 4 2 3 19 6 972 0 1 1 4 1 0 2 0 1 0 12 8 8] - [ 6 1 0 1 0 1 1 1 986 31 8 1 2 4 13 1 2 0 3 1 1] - [ 49 0 1 0 4 2 0 2 27 893 1 1 0 5 5 1 1 0 1 0 8] - [ 2 0 2 2 2 1 1 2 7 1 968 1 1 7 4 1 0 0 6 1 10] - [ 1 0 1 0 0 7 6 1 0 1 0 978 23 3 1 6 3 11 0 5 4] - [ 0 0 0 2 0 2 0 0 1 1 0 19 911 0 3 7 0 10 1 2 10] - [ 3 1 1 0 1 10 0 2 13 10 7 3 6 948 1 1 4 0 0 1 11] - [ 8 5 1 8 3 2 0 0 15 3 1 1 2 3 1070 0 0 0 4 0 4] - [ 0 0 0 2 2 0 4 0 1 1 0 5 12 1 0 994 5 10 0 2 4] - [ 0 4 1 1 2 0 1 0 1 1 0 3 3 2 0 10 1031 2 1 4 5] - [ 4 0 1 0 2 1 0 1 1 3 0 3 13 1 2 10 0 993 0 0 1] - [ 4 3 2 4 1 2 0 22 2 1 5 1 2 0 8 0 0 2 944 0 5] - [ 0 5 1 1 3 4 6 7 0 1 3 11 5 2 0 5 5 4 0 1010 7] - [ 126 221 138 96 98 144 74 136 83 83 151 84 283 233 139 83 164 90 153 181 10466]] - -2022-12-06 11:36:44,298 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:36:44,298 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:36:44,304 - - -2022-12-06 11:36:44,304 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:36:45,343 - Epoch: [166][ 10/ 1200] Overall Loss 0.134640 Objective Loss 0.134640 LR 0.000250 Time 0.103781 -2022-12-06 11:36:45,543 - Epoch: [166][ 20/ 1200] Overall Loss 0.142303 Objective Loss 0.142303 LR 0.000250 Time 0.061883 -2022-12-06 11:36:45,739 - Epoch: [166][ 30/ 1200] Overall Loss 0.134481 Objective Loss 0.134481 LR 0.000250 Time 0.047751 -2022-12-06 11:36:45,933 - Epoch: [166][ 40/ 1200] Overall Loss 0.142196 Objective Loss 0.142196 LR 0.000250 Time 0.040662 -2022-12-06 11:36:46,128 - Epoch: [166][ 50/ 1200] Overall Loss 0.142896 Objective Loss 0.142896 LR 0.000250 Time 0.036421 -2022-12-06 11:36:46,322 - Epoch: [166][ 60/ 1200] Overall Loss 0.142927 Objective Loss 0.142927 LR 0.000250 Time 0.033578 -2022-12-06 11:36:46,517 - Epoch: [166][ 70/ 1200] Overall Loss 0.142427 Objective Loss 0.142427 LR 0.000250 Time 0.031553 -2022-12-06 11:36:46,711 - Epoch: [166][ 80/ 1200] Overall Loss 0.142820 Objective Loss 0.142820 LR 0.000250 Time 0.030031 -2022-12-06 11:36:46,906 - Epoch: [166][ 90/ 1200] Overall Loss 0.145340 Objective Loss 0.145340 LR 0.000250 Time 0.028851 -2022-12-06 11:36:47,100 - Epoch: [166][ 100/ 1200] Overall Loss 0.145686 Objective Loss 0.145686 LR 0.000250 Time 0.027904 -2022-12-06 11:36:47,294 - Epoch: [166][ 110/ 1200] Overall Loss 0.142948 Objective Loss 0.142948 LR 0.000250 Time 0.027123 -2022-12-06 11:36:47,488 - Epoch: [166][ 120/ 1200] Overall Loss 0.142717 Objective Loss 0.142717 LR 0.000250 Time 0.026474 -2022-12-06 11:36:47,683 - Epoch: [166][ 130/ 1200] Overall Loss 0.143435 Objective Loss 0.143435 LR 0.000250 Time 0.025935 -2022-12-06 11:36:47,877 - Epoch: [166][ 140/ 1200] Overall Loss 0.143906 Objective Loss 0.143906 LR 0.000250 Time 0.025465 -2022-12-06 11:36:48,072 - Epoch: [166][ 150/ 1200] Overall Loss 0.145489 Objective Loss 0.145489 LR 0.000250 Time 0.025062 -2022-12-06 11:36:48,266 - Epoch: [166][ 160/ 1200] Overall Loss 0.145652 Objective Loss 0.145652 LR 0.000250 Time 0.024703 -2022-12-06 11:36:48,460 - Epoch: [166][ 170/ 1200] Overall Loss 0.145976 Objective Loss 0.145976 LR 0.000250 Time 0.024390 -2022-12-06 11:36:48,655 - Epoch: [166][ 180/ 1200] Overall Loss 0.145351 Objective Loss 0.145351 LR 0.000250 Time 0.024112 -2022-12-06 11:36:48,849 - Epoch: [166][ 190/ 1200] Overall Loss 0.144407 Objective Loss 0.144407 LR 0.000250 Time 0.023863 -2022-12-06 11:36:49,042 - Epoch: [166][ 200/ 1200] Overall Loss 0.145333 Objective Loss 0.145333 LR 0.000250 Time 0.023634 -2022-12-06 11:36:49,237 - Epoch: [166][ 210/ 1200] Overall Loss 0.145249 Objective Loss 0.145249 LR 0.000250 Time 0.023432 -2022-12-06 11:36:49,430 - Epoch: [166][ 220/ 1200] Overall Loss 0.145595 Objective Loss 0.145595 LR 0.000250 Time 0.023244 -2022-12-06 11:36:49,626 - Epoch: [166][ 230/ 1200] Overall Loss 0.145655 Objective Loss 0.145655 LR 0.000250 Time 0.023081 -2022-12-06 11:36:49,820 - Epoch: [166][ 240/ 1200] Overall Loss 0.145573 Objective Loss 0.145573 LR 0.000250 Time 0.022925 -2022-12-06 11:36:50,014 - Epoch: [166][ 250/ 1200] Overall Loss 0.145557 Objective Loss 0.145557 LR 0.000250 Time 0.022783 -2022-12-06 11:36:50,208 - Epoch: [166][ 260/ 1200] Overall Loss 0.144871 Objective Loss 0.144871 LR 0.000250 Time 0.022652 -2022-12-06 11:36:50,403 - Epoch: [166][ 270/ 1200] Overall Loss 0.144574 Objective Loss 0.144574 LR 0.000250 Time 0.022531 -2022-12-06 11:36:50,596 - Epoch: [166][ 280/ 1200] Overall Loss 0.144336 Objective Loss 0.144336 LR 0.000250 Time 0.022416 -2022-12-06 11:36:50,789 - Epoch: [166][ 290/ 1200] Overall Loss 0.144278 Objective Loss 0.144278 LR 0.000250 Time 0.022307 -2022-12-06 11:36:50,983 - Epoch: [166][ 300/ 1200] Overall Loss 0.144371 Objective Loss 0.144371 LR 0.000250 Time 0.022208 -2022-12-06 11:36:51,178 - Epoch: [166][ 310/ 1200] Overall Loss 0.144704 Objective Loss 0.144704 LR 0.000250 Time 0.022118 -2022-12-06 11:36:51,372 - Epoch: [166][ 320/ 1200] Overall Loss 0.144375 Objective Loss 0.144375 LR 0.000250 Time 0.022031 -2022-12-06 11:36:51,567 - Epoch: [166][ 330/ 1200] Overall Loss 0.144452 Objective Loss 0.144452 LR 0.000250 Time 0.021952 -2022-12-06 11:36:51,761 - Epoch: [166][ 340/ 1200] Overall Loss 0.144282 Objective Loss 0.144282 LR 0.000250 Time 0.021876 -2022-12-06 11:36:51,956 - Epoch: [166][ 350/ 1200] Overall Loss 0.144205 Objective Loss 0.144205 LR 0.000250 Time 0.021806 -2022-12-06 11:36:52,150 - Epoch: [166][ 360/ 1200] Overall Loss 0.143896 Objective Loss 0.143896 LR 0.000250 Time 0.021737 -2022-12-06 11:36:52,344 - Epoch: [166][ 370/ 1200] Overall Loss 0.144459 Objective Loss 0.144459 LR 0.000250 Time 0.021674 -2022-12-06 11:36:52,538 - Epoch: [166][ 380/ 1200] Overall Loss 0.144647 Objective Loss 0.144647 LR 0.000250 Time 0.021612 -2022-12-06 11:36:52,733 - Epoch: [166][ 390/ 1200] Overall Loss 0.144787 Objective Loss 0.144787 LR 0.000250 Time 0.021556 -2022-12-06 11:36:52,927 - Epoch: [166][ 400/ 1200] Overall Loss 0.144189 Objective Loss 0.144189 LR 0.000250 Time 0.021500 -2022-12-06 11:36:53,121 - Epoch: [166][ 410/ 1200] Overall Loss 0.144488 Objective Loss 0.144488 LR 0.000250 Time 0.021449 -2022-12-06 11:36:53,316 - Epoch: [166][ 420/ 1200] Overall Loss 0.144676 Objective Loss 0.144676 LR 0.000250 Time 0.021401 -2022-12-06 11:36:53,511 - Epoch: [166][ 430/ 1200] Overall Loss 0.144571 Objective Loss 0.144571 LR 0.000250 Time 0.021355 -2022-12-06 11:36:53,705 - Epoch: [166][ 440/ 1200] Overall Loss 0.144657 Objective Loss 0.144657 LR 0.000250 Time 0.021310 -2022-12-06 11:36:53,900 - Epoch: [166][ 450/ 1200] Overall Loss 0.145106 Objective Loss 0.145106 LR 0.000250 Time 0.021267 -2022-12-06 11:36:54,094 - Epoch: [166][ 460/ 1200] Overall Loss 0.145089 Objective Loss 0.145089 LR 0.000250 Time 0.021226 -2022-12-06 11:36:54,288 - Epoch: [166][ 470/ 1200] Overall Loss 0.145154 Objective Loss 0.145154 LR 0.000250 Time 0.021187 -2022-12-06 11:36:54,483 - Epoch: [166][ 480/ 1200] Overall Loss 0.144996 Objective Loss 0.144996 LR 0.000250 Time 0.021149 -2022-12-06 11:36:54,677 - Epoch: [166][ 490/ 1200] Overall Loss 0.145401 Objective Loss 0.145401 LR 0.000250 Time 0.021113 -2022-12-06 11:36:54,871 - Epoch: [166][ 500/ 1200] Overall Loss 0.145261 Objective Loss 0.145261 LR 0.000250 Time 0.021078 -2022-12-06 11:36:55,066 - Epoch: [166][ 510/ 1200] Overall Loss 0.145202 Objective Loss 0.145202 LR 0.000250 Time 0.021045 -2022-12-06 11:36:55,260 - Epoch: [166][ 520/ 1200] Overall Loss 0.145126 Objective Loss 0.145126 LR 0.000250 Time 0.021014 -2022-12-06 11:36:55,455 - Epoch: [166][ 530/ 1200] Overall Loss 0.145091 Objective Loss 0.145091 LR 0.000250 Time 0.020983 -2022-12-06 11:36:55,649 - Epoch: [166][ 540/ 1200] Overall Loss 0.145320 Objective Loss 0.145320 LR 0.000250 Time 0.020952 -2022-12-06 11:36:55,844 - Epoch: [166][ 550/ 1200] Overall Loss 0.145729 Objective Loss 0.145729 LR 0.000250 Time 0.020925 -2022-12-06 11:36:56,038 - Epoch: [166][ 560/ 1200] Overall Loss 0.145509 Objective Loss 0.145509 LR 0.000250 Time 0.020897 -2022-12-06 11:36:56,232 - Epoch: [166][ 570/ 1200] Overall Loss 0.145302 Objective Loss 0.145302 LR 0.000250 Time 0.020871 -2022-12-06 11:36:56,427 - Epoch: [166][ 580/ 1200] Overall Loss 0.145293 Objective Loss 0.145293 LR 0.000250 Time 0.020845 -2022-12-06 11:36:56,621 - Epoch: [166][ 590/ 1200] Overall Loss 0.144851 Objective Loss 0.144851 LR 0.000250 Time 0.020821 -2022-12-06 11:36:56,815 - Epoch: [166][ 600/ 1200] Overall Loss 0.144771 Objective Loss 0.144771 LR 0.000250 Time 0.020795 -2022-12-06 11:36:57,007 - Epoch: [166][ 610/ 1200] Overall Loss 0.144660 Objective Loss 0.144660 LR 0.000250 Time 0.020768 -2022-12-06 11:36:57,200 - Epoch: [166][ 620/ 1200] Overall Loss 0.144777 Objective Loss 0.144777 LR 0.000250 Time 0.020744 -2022-12-06 11:36:57,393 - Epoch: [166][ 630/ 1200] Overall Loss 0.144651 Objective Loss 0.144651 LR 0.000250 Time 0.020719 -2022-12-06 11:36:57,585 - Epoch: [166][ 640/ 1200] Overall Loss 0.144511 Objective Loss 0.144511 LR 0.000250 Time 0.020696 -2022-12-06 11:36:57,778 - Epoch: [166][ 650/ 1200] Overall Loss 0.144492 Objective Loss 0.144492 LR 0.000250 Time 0.020672 -2022-12-06 11:36:57,970 - Epoch: [166][ 660/ 1200] Overall Loss 0.144170 Objective Loss 0.144170 LR 0.000250 Time 0.020650 -2022-12-06 11:36:58,163 - Epoch: [166][ 670/ 1200] Overall Loss 0.144068 Objective Loss 0.144068 LR 0.000250 Time 0.020629 -2022-12-06 11:36:58,356 - Epoch: [166][ 680/ 1200] Overall Loss 0.144151 Objective Loss 0.144151 LR 0.000250 Time 0.020609 -2022-12-06 11:36:58,549 - Epoch: [166][ 690/ 1200] Overall Loss 0.143863 Objective Loss 0.143863 LR 0.000250 Time 0.020588 -2022-12-06 11:36:58,742 - Epoch: [166][ 700/ 1200] Overall Loss 0.143839 Objective Loss 0.143839 LR 0.000250 Time 0.020569 -2022-12-06 11:36:58,935 - Epoch: [166][ 710/ 1200] Overall Loss 0.143998 Objective Loss 0.143998 LR 0.000250 Time 0.020550 -2022-12-06 11:36:59,128 - Epoch: [166][ 720/ 1200] Overall Loss 0.143743 Objective Loss 0.143743 LR 0.000250 Time 0.020532 -2022-12-06 11:36:59,320 - Epoch: [166][ 730/ 1200] Overall Loss 0.143851 Objective Loss 0.143851 LR 0.000250 Time 0.020514 -2022-12-06 11:36:59,514 - Epoch: [166][ 740/ 1200] Overall Loss 0.143903 Objective Loss 0.143903 LR 0.000250 Time 0.020497 -2022-12-06 11:36:59,706 - Epoch: [166][ 750/ 1200] Overall Loss 0.144201 Objective Loss 0.144201 LR 0.000250 Time 0.020480 -2022-12-06 11:36:59,899 - Epoch: [166][ 760/ 1200] Overall Loss 0.144032 Objective Loss 0.144032 LR 0.000250 Time 0.020463 -2022-12-06 11:37:00,091 - Epoch: [166][ 770/ 1200] Overall Loss 0.144371 Objective Loss 0.144371 LR 0.000250 Time 0.020446 -2022-12-06 11:37:00,284 - Epoch: [166][ 780/ 1200] Overall Loss 0.144403 Objective Loss 0.144403 LR 0.000250 Time 0.020431 -2022-12-06 11:37:00,476 - Epoch: [166][ 790/ 1200] Overall Loss 0.144396 Objective Loss 0.144396 LR 0.000250 Time 0.020415 -2022-12-06 11:37:00,669 - Epoch: [166][ 800/ 1200] Overall Loss 0.144735 Objective Loss 0.144735 LR 0.000250 Time 0.020400 -2022-12-06 11:37:00,862 - Epoch: [166][ 810/ 1200] Overall Loss 0.145093 Objective Loss 0.145093 LR 0.000250 Time 0.020386 -2022-12-06 11:37:01,055 - Epoch: [166][ 820/ 1200] Overall Loss 0.145079 Objective Loss 0.145079 LR 0.000250 Time 0.020372 -2022-12-06 11:37:01,247 - Epoch: [166][ 830/ 1200] Overall Loss 0.145068 Objective Loss 0.145068 LR 0.000250 Time 0.020358 -2022-12-06 11:37:01,439 - Epoch: [166][ 840/ 1200] Overall Loss 0.144888 Objective Loss 0.144888 LR 0.000250 Time 0.020343 -2022-12-06 11:37:01,632 - Epoch: [166][ 850/ 1200] Overall Loss 0.145093 Objective Loss 0.145093 LR 0.000250 Time 0.020330 -2022-12-06 11:37:01,825 - Epoch: [166][ 860/ 1200] Overall Loss 0.145150 Objective Loss 0.145150 LR 0.000250 Time 0.020317 -2022-12-06 11:37:02,016 - Epoch: [166][ 870/ 1200] Overall Loss 0.144950 Objective Loss 0.144950 LR 0.000250 Time 0.020303 -2022-12-06 11:37:02,209 - Epoch: [166][ 880/ 1200] Overall Loss 0.144813 Objective Loss 0.144813 LR 0.000250 Time 0.020291 -2022-12-06 11:37:02,401 - Epoch: [166][ 890/ 1200] Overall Loss 0.144761 Objective Loss 0.144761 LR 0.000250 Time 0.020278 -2022-12-06 11:37:02,594 - Epoch: [166][ 900/ 1200] Overall Loss 0.144767 Objective Loss 0.144767 LR 0.000250 Time 0.020266 -2022-12-06 11:37:02,786 - Epoch: [166][ 910/ 1200] Overall Loss 0.144834 Objective Loss 0.144834 LR 0.000250 Time 0.020253 -2022-12-06 11:37:02,978 - Epoch: [166][ 920/ 1200] Overall Loss 0.144934 Objective Loss 0.144934 LR 0.000250 Time 0.020242 -2022-12-06 11:37:03,171 - Epoch: [166][ 930/ 1200] Overall Loss 0.145137 Objective Loss 0.145137 LR 0.000250 Time 0.020231 -2022-12-06 11:37:03,363 - Epoch: [166][ 940/ 1200] Overall Loss 0.144978 Objective Loss 0.144978 LR 0.000250 Time 0.020220 -2022-12-06 11:37:03,556 - Epoch: [166][ 950/ 1200] Overall Loss 0.144793 Objective Loss 0.144793 LR 0.000250 Time 0.020209 -2022-12-06 11:37:03,748 - Epoch: [166][ 960/ 1200] Overall Loss 0.144644 Objective Loss 0.144644 LR 0.000250 Time 0.020198 -2022-12-06 11:37:03,940 - Epoch: [166][ 970/ 1200] Overall Loss 0.144759 Objective Loss 0.144759 LR 0.000250 Time 0.020188 -2022-12-06 11:37:04,133 - Epoch: [166][ 980/ 1200] Overall Loss 0.145016 Objective Loss 0.145016 LR 0.000250 Time 0.020178 -2022-12-06 11:37:04,325 - Epoch: [166][ 990/ 1200] Overall Loss 0.145011 Objective Loss 0.145011 LR 0.000250 Time 0.020167 -2022-12-06 11:37:04,518 - Epoch: [166][ 1000/ 1200] Overall Loss 0.145119 Objective Loss 0.145119 LR 0.000250 Time 0.020158 -2022-12-06 11:37:04,710 - Epoch: [166][ 1010/ 1200] Overall Loss 0.145084 Objective Loss 0.145084 LR 0.000250 Time 0.020148 -2022-12-06 11:37:04,903 - Epoch: [166][ 1020/ 1200] Overall Loss 0.144973 Objective Loss 0.144973 LR 0.000250 Time 0.020139 -2022-12-06 11:37:05,096 - Epoch: [166][ 1030/ 1200] Overall Loss 0.145127 Objective Loss 0.145127 LR 0.000250 Time 0.020130 -2022-12-06 11:37:05,288 - Epoch: [166][ 1040/ 1200] Overall Loss 0.145244 Objective Loss 0.145244 LR 0.000250 Time 0.020121 -2022-12-06 11:37:05,481 - Epoch: [166][ 1050/ 1200] Overall Loss 0.145388 Objective Loss 0.145388 LR 0.000250 Time 0.020112 -2022-12-06 11:37:05,674 - Epoch: [166][ 1060/ 1200] Overall Loss 0.145530 Objective Loss 0.145530 LR 0.000250 Time 0.020104 -2022-12-06 11:37:05,867 - Epoch: [166][ 1070/ 1200] Overall Loss 0.145475 Objective Loss 0.145475 LR 0.000250 Time 0.020096 -2022-12-06 11:37:06,059 - Epoch: [166][ 1080/ 1200] Overall Loss 0.145521 Objective Loss 0.145521 LR 0.000250 Time 0.020088 -2022-12-06 11:37:06,252 - Epoch: [166][ 1090/ 1200] Overall Loss 0.145654 Objective Loss 0.145654 LR 0.000250 Time 0.020080 -2022-12-06 11:37:06,445 - Epoch: [166][ 1100/ 1200] Overall Loss 0.145783 Objective Loss 0.145783 LR 0.000250 Time 0.020072 -2022-12-06 11:37:06,638 - Epoch: [166][ 1110/ 1200] Overall Loss 0.145759 Objective Loss 0.145759 LR 0.000250 Time 0.020064 -2022-12-06 11:37:06,830 - Epoch: [166][ 1120/ 1200] Overall Loss 0.145767 Objective Loss 0.145767 LR 0.000250 Time 0.020056 -2022-12-06 11:37:07,023 - Epoch: [166][ 1130/ 1200] Overall Loss 0.146024 Objective Loss 0.146024 LR 0.000250 Time 0.020049 -2022-12-06 11:37:07,215 - Epoch: [166][ 1140/ 1200] Overall Loss 0.146085 Objective Loss 0.146085 LR 0.000250 Time 0.020041 -2022-12-06 11:37:07,408 - Epoch: [166][ 1150/ 1200] Overall Loss 0.145945 Objective Loss 0.145945 LR 0.000250 Time 0.020034 -2022-12-06 11:37:07,600 - Epoch: [166][ 1160/ 1200] Overall Loss 0.146151 Objective Loss 0.146151 LR 0.000250 Time 0.020027 -2022-12-06 11:37:07,792 - Epoch: [166][ 1170/ 1200] Overall Loss 0.146045 Objective Loss 0.146045 LR 0.000250 Time 0.020020 -2022-12-06 11:37:07,985 - Epoch: [166][ 1180/ 1200] Overall Loss 0.146076 Objective Loss 0.146076 LR 0.000250 Time 0.020013 -2022-12-06 11:37:08,178 - Epoch: [166][ 1190/ 1200] Overall Loss 0.146262 Objective Loss 0.146262 LR 0.000250 Time 0.020006 -2022-12-06 11:37:08,412 - Epoch: [166][ 1200/ 1200] Overall Loss 0.146247 Objective Loss 0.146247 Top1 89.748954 Top5 98.953975 LR 0.000250 Time 0.020034 -2022-12-06 11:37:08,502 - --- validate (epoch=166)----------- -2022-12-06 11:37:08,502 - 34129 samples (256 per mini-batch) -2022-12-06 11:37:08,942 - Epoch: [166][ 10/ 134] Loss 0.256535 Top1 87.226562 Top5 98.125000 -2022-12-06 11:37:09,076 - Epoch: [166][ 20/ 134] Loss 0.245083 Top1 87.675781 Top5 98.261719 -2022-12-06 11:37:09,208 - Epoch: [166][ 30/ 134] Loss 0.248840 Top1 87.552083 Top5 98.385417 -2022-12-06 11:37:09,341 - Epoch: [166][ 40/ 134] Loss 0.242670 Top1 87.773438 Top5 98.476562 -2022-12-06 11:37:09,473 - Epoch: [166][ 50/ 134] Loss 0.239391 Top1 87.875000 Top5 98.468750 -2022-12-06 11:37:09,606 - Epoch: [166][ 60/ 134] Loss 0.228670 Top1 88.170573 Top5 98.587240 -2022-12-06 11:37:09,738 - Epoch: [166][ 70/ 134] Loss 0.230916 Top1 88.119420 Top5 98.571429 -2022-12-06 11:37:09,870 - Epoch: [166][ 80/ 134] Loss 0.231113 Top1 88.081055 Top5 98.623047 -2022-12-06 11:37:10,002 - Epoch: [166][ 90/ 134] Loss 0.231021 Top1 88.255208 Top5 98.632812 -2022-12-06 11:37:10,131 - Epoch: [166][ 100/ 134] Loss 0.231466 Top1 88.207031 Top5 98.589844 -2022-12-06 11:37:10,263 - Epoch: [166][ 110/ 134] Loss 0.233242 Top1 88.220881 Top5 98.600852 -2022-12-06 11:37:10,396 - Epoch: [166][ 120/ 134] Loss 0.233370 Top1 88.118490 Top5 98.564453 -2022-12-06 11:37:10,528 - Epoch: [166][ 130/ 134] Loss 0.235348 Top1 88.091947 Top5 98.563702 -2022-12-06 11:37:10,567 - Epoch: [166][ 134/ 134] Loss 0.234636 Top1 88.106889 Top5 98.567201 -2022-12-06 11:37:10,654 - ==> Top1: 88.107 Top5: 98.567 Loss: 0.235 - -2022-12-06 11:37:10,655 - ==> Confusion: -[[ 916 0 2 2 6 5 0 0 5 37 0 1 2 3 9 3 1 0 1 0 3] - [ 1 946 2 2 12 24 2 8 2 0 1 5 0 1 0 2 3 1 7 2 6] - [ 3 6 1011 11 5 3 17 7 0 2 6 3 2 3 4 3 1 1 1 3 11] - [ 2 2 16 954 0 1 1 0 0 0 9 1 4 1 11 0 2 3 7 1 5] - [ 7 2 1 0 970 2 0 2 1 5 2 2 0 2 6 7 3 2 0 3 3] - [ 1 8 1 2 7 1004 1 10 1 3 1 9 1 12 1 0 1 1 0 4 1] - [ 1 3 2 1 0 1 1083 2 0 0 0 1 0 3 0 7 1 2 2 8 1] - [ 0 7 4 3 4 31 9 947 1 0 2 7 1 0 0 1 2 0 13 16 6] - [ 5 1 0 0 0 3 1 0 980 35 9 1 1 14 8 1 2 1 1 1 0] - [ 49 0 1 0 8 2 0 3 18 888 1 1 0 15 3 2 1 0 1 1 7] - [ 1 1 3 2 1 2 0 5 4 1 979 1 0 8 3 1 1 0 1 1 4] - [ 2 0 1 1 0 9 4 1 1 1 0 974 22 3 0 9 3 5 0 10 5] - [ 0 1 0 2 2 2 0 2 0 0 0 21 913 0 0 8 2 7 0 3 6] - [ 1 1 0 0 1 6 0 3 6 5 2 1 3 978 0 2 2 0 0 4 8] - [ 5 3 1 6 5 4 0 0 11 2 1 2 2 3 1067 0 0 1 4 1 12] - [ 1 0 1 0 1 1 3 1 1 0 2 6 5 1 0 1004 5 6 0 3 2] - [ 3 1 1 1 5 2 1 0 1 0 0 2 2 2 0 12 1030 0 0 5 4] - [ 2 0 2 0 2 1 0 0 0 4 0 7 17 1 1 21 0 975 0 0 3] - [ 2 5 5 6 3 3 0 20 2 1 5 2 3 0 6 0 0 2 940 1 2] - [ 0 5 0 2 1 3 3 3 0 0 2 13 4 5 0 4 3 3 0 1023 6] - [ 89 201 129 94 133 183 76 108 65 73 163 80 303 264 130 108 126 72 114 228 10487]] - -2022-12-06 11:37:11,220 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:37:11,221 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:37:11,226 - - -2022-12-06 11:37:11,227 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:37:12,148 - Epoch: [167][ 10/ 1200] Overall Loss 0.155034 Objective Loss 0.155034 LR 0.000250 Time 0.092081 -2022-12-06 11:37:12,343 - Epoch: [167][ 20/ 1200] Overall Loss 0.142813 Objective Loss 0.142813 LR 0.000250 Time 0.055766 -2022-12-06 11:37:12,534 - Epoch: [167][ 30/ 1200] Overall Loss 0.137526 Objective Loss 0.137526 LR 0.000250 Time 0.043514 -2022-12-06 11:37:12,724 - Epoch: [167][ 40/ 1200] Overall Loss 0.132754 Objective Loss 0.132754 LR 0.000250 Time 0.037386 -2022-12-06 11:37:12,914 - Epoch: [167][ 50/ 1200] Overall Loss 0.132098 Objective Loss 0.132098 LR 0.000250 Time 0.033693 -2022-12-06 11:37:13,104 - Epoch: [167][ 60/ 1200] Overall Loss 0.135188 Objective Loss 0.135188 LR 0.000250 Time 0.031233 -2022-12-06 11:37:13,294 - Epoch: [167][ 70/ 1200] Overall Loss 0.134869 Objective Loss 0.134869 LR 0.000250 Time 0.029479 -2022-12-06 11:37:13,484 - Epoch: [167][ 80/ 1200] Overall Loss 0.134913 Objective Loss 0.134913 LR 0.000250 Time 0.028160 -2022-12-06 11:37:13,675 - Epoch: [167][ 90/ 1200] Overall Loss 0.135225 Objective Loss 0.135225 LR 0.000250 Time 0.027144 -2022-12-06 11:37:13,865 - Epoch: [167][ 100/ 1200] Overall Loss 0.137398 Objective Loss 0.137398 LR 0.000250 Time 0.026332 -2022-12-06 11:37:14,055 - Epoch: [167][ 110/ 1200] Overall Loss 0.135469 Objective Loss 0.135469 LR 0.000250 Time 0.025657 -2022-12-06 11:37:14,245 - Epoch: [167][ 120/ 1200] Overall Loss 0.137037 Objective Loss 0.137037 LR 0.000250 Time 0.025098 -2022-12-06 11:37:14,436 - Epoch: [167][ 130/ 1200] Overall Loss 0.136588 Objective Loss 0.136588 LR 0.000250 Time 0.024632 -2022-12-06 11:37:14,626 - Epoch: [167][ 140/ 1200] Overall Loss 0.137116 Objective Loss 0.137116 LR 0.000250 Time 0.024224 -2022-12-06 11:37:14,815 - Epoch: [167][ 150/ 1200] Overall Loss 0.138665 Objective Loss 0.138665 LR 0.000250 Time 0.023868 -2022-12-06 11:37:15,006 - Epoch: [167][ 160/ 1200] Overall Loss 0.140331 Objective Loss 0.140331 LR 0.000250 Time 0.023563 -2022-12-06 11:37:15,196 - Epoch: [167][ 170/ 1200] Overall Loss 0.139913 Objective Loss 0.139913 LR 0.000250 Time 0.023295 -2022-12-06 11:37:15,386 - Epoch: [167][ 180/ 1200] Overall Loss 0.141014 Objective Loss 0.141014 LR 0.000250 Time 0.023054 -2022-12-06 11:37:15,578 - Epoch: [167][ 190/ 1200] Overall Loss 0.142030 Objective Loss 0.142030 LR 0.000250 Time 0.022843 -2022-12-06 11:37:15,768 - Epoch: [167][ 200/ 1200] Overall Loss 0.141276 Objective Loss 0.141276 LR 0.000250 Time 0.022649 -2022-12-06 11:37:15,958 - Epoch: [167][ 210/ 1200] Overall Loss 0.141634 Objective Loss 0.141634 LR 0.000250 Time 0.022472 -2022-12-06 11:37:16,147 - Epoch: [167][ 220/ 1200] Overall Loss 0.142600 Objective Loss 0.142600 LR 0.000250 Time 0.022309 -2022-12-06 11:37:16,337 - Epoch: [167][ 230/ 1200] Overall Loss 0.142099 Objective Loss 0.142099 LR 0.000250 Time 0.022165 -2022-12-06 11:37:16,528 - Epoch: [167][ 240/ 1200] Overall Loss 0.141237 Objective Loss 0.141237 LR 0.000250 Time 0.022033 -2022-12-06 11:37:16,718 - Epoch: [167][ 250/ 1200] Overall Loss 0.141293 Objective Loss 0.141293 LR 0.000250 Time 0.021910 -2022-12-06 11:37:16,913 - Epoch: [167][ 260/ 1200] Overall Loss 0.140712 Objective Loss 0.140712 LR 0.000250 Time 0.021814 -2022-12-06 11:37:17,113 - Epoch: [167][ 270/ 1200] Overall Loss 0.140588 Objective Loss 0.140588 LR 0.000250 Time 0.021745 -2022-12-06 11:37:17,320 - Epoch: [167][ 280/ 1200] Overall Loss 0.141064 Objective Loss 0.141064 LR 0.000250 Time 0.021705 -2022-12-06 11:37:17,521 - Epoch: [167][ 290/ 1200] Overall Loss 0.141792 Objective Loss 0.141792 LR 0.000250 Time 0.021648 -2022-12-06 11:37:17,727 - Epoch: [167][ 300/ 1200] Overall Loss 0.141615 Objective Loss 0.141615 LR 0.000250 Time 0.021613 -2022-12-06 11:37:17,928 - Epoch: [167][ 310/ 1200] Overall Loss 0.141357 Objective Loss 0.141357 LR 0.000250 Time 0.021562 -2022-12-06 11:37:18,134 - Epoch: [167][ 320/ 1200] Overall Loss 0.141376 Objective Loss 0.141376 LR 0.000250 Time 0.021528 -2022-12-06 11:37:18,335 - Epoch: [167][ 330/ 1200] Overall Loss 0.141725 Objective Loss 0.141725 LR 0.000250 Time 0.021483 -2022-12-06 11:37:18,541 - Epoch: [167][ 340/ 1200] Overall Loss 0.141858 Objective Loss 0.141858 LR 0.000250 Time 0.021456 -2022-12-06 11:37:18,742 - Epoch: [167][ 350/ 1200] Overall Loss 0.141461 Objective Loss 0.141461 LR 0.000250 Time 0.021417 -2022-12-06 11:37:18,948 - Epoch: [167][ 360/ 1200] Overall Loss 0.141713 Objective Loss 0.141713 LR 0.000250 Time 0.021392 -2022-12-06 11:37:19,149 - Epoch: [167][ 370/ 1200] Overall Loss 0.141994 Objective Loss 0.141994 LR 0.000250 Time 0.021354 -2022-12-06 11:37:19,354 - Epoch: [167][ 380/ 1200] Overall Loss 0.142162 Objective Loss 0.142162 LR 0.000250 Time 0.021332 -2022-12-06 11:37:19,555 - Epoch: [167][ 390/ 1200] Overall Loss 0.141885 Objective Loss 0.141885 LR 0.000250 Time 0.021299 -2022-12-06 11:37:19,761 - Epoch: [167][ 400/ 1200] Overall Loss 0.141892 Objective Loss 0.141892 LR 0.000250 Time 0.021278 -2022-12-06 11:37:19,961 - Epoch: [167][ 410/ 1200] Overall Loss 0.142633 Objective Loss 0.142633 LR 0.000250 Time 0.021247 -2022-12-06 11:37:20,166 - Epoch: [167][ 420/ 1200] Overall Loss 0.142141 Objective Loss 0.142141 LR 0.000250 Time 0.021227 -2022-12-06 11:37:20,368 - Epoch: [167][ 430/ 1200] Overall Loss 0.142410 Objective Loss 0.142410 LR 0.000250 Time 0.021202 -2022-12-06 11:37:20,573 - Epoch: [167][ 440/ 1200] Overall Loss 0.142350 Objective Loss 0.142350 LR 0.000250 Time 0.021185 -2022-12-06 11:37:20,774 - Epoch: [167][ 450/ 1200] Overall Loss 0.142606 Objective Loss 0.142606 LR 0.000250 Time 0.021160 -2022-12-06 11:37:20,979 - Epoch: [167][ 460/ 1200] Overall Loss 0.142422 Objective Loss 0.142422 LR 0.000250 Time 0.021144 -2022-12-06 11:37:21,180 - Epoch: [167][ 470/ 1200] Overall Loss 0.142295 Objective Loss 0.142295 LR 0.000250 Time 0.021120 -2022-12-06 11:37:21,386 - Epoch: [167][ 480/ 1200] Overall Loss 0.142073 Objective Loss 0.142073 LR 0.000250 Time 0.021108 -2022-12-06 11:37:21,587 - Epoch: [167][ 490/ 1200] Overall Loss 0.141999 Objective Loss 0.141999 LR 0.000250 Time 0.021086 -2022-12-06 11:37:21,792 - Epoch: [167][ 500/ 1200] Overall Loss 0.142264 Objective Loss 0.142264 LR 0.000250 Time 0.021074 -2022-12-06 11:37:21,994 - Epoch: [167][ 510/ 1200] Overall Loss 0.142305 Objective Loss 0.142305 LR 0.000250 Time 0.021055 -2022-12-06 11:37:22,199 - Epoch: [167][ 520/ 1200] Overall Loss 0.142044 Objective Loss 0.142044 LR 0.000250 Time 0.021043 -2022-12-06 11:37:22,400 - Epoch: [167][ 530/ 1200] Overall Loss 0.141887 Objective Loss 0.141887 LR 0.000250 Time 0.021025 -2022-12-06 11:37:22,606 - Epoch: [167][ 540/ 1200] Overall Loss 0.142102 Objective Loss 0.142102 LR 0.000250 Time 0.021015 -2022-12-06 11:37:22,806 - Epoch: [167][ 550/ 1200] Overall Loss 0.142132 Objective Loss 0.142132 LR 0.000250 Time 0.020997 -2022-12-06 11:37:23,013 - Epoch: [167][ 560/ 1200] Overall Loss 0.142267 Objective Loss 0.142267 LR 0.000250 Time 0.020989 -2022-12-06 11:37:23,213 - Epoch: [167][ 570/ 1200] Overall Loss 0.142078 Objective Loss 0.142078 LR 0.000250 Time 0.020972 -2022-12-06 11:37:23,419 - Epoch: [167][ 580/ 1200] Overall Loss 0.141969 Objective Loss 0.141969 LR 0.000250 Time 0.020964 -2022-12-06 11:37:23,620 - Epoch: [167][ 590/ 1200] Overall Loss 0.141817 Objective Loss 0.141817 LR 0.000250 Time 0.020948 -2022-12-06 11:37:23,826 - Epoch: [167][ 600/ 1200] Overall Loss 0.141600 Objective Loss 0.141600 LR 0.000250 Time 0.020942 -2022-12-06 11:37:24,027 - Epoch: [167][ 610/ 1200] Overall Loss 0.141688 Objective Loss 0.141688 LR 0.000250 Time 0.020927 -2022-12-06 11:37:24,232 - Epoch: [167][ 620/ 1200] Overall Loss 0.141934 Objective Loss 0.141934 LR 0.000250 Time 0.020920 -2022-12-06 11:37:24,433 - Epoch: [167][ 630/ 1200] Overall Loss 0.142144 Objective Loss 0.142144 LR 0.000250 Time 0.020906 -2022-12-06 11:37:24,639 - Epoch: [167][ 640/ 1200] Overall Loss 0.142693 Objective Loss 0.142693 LR 0.000250 Time 0.020900 -2022-12-06 11:37:24,841 - Epoch: [167][ 650/ 1200] Overall Loss 0.142661 Objective Loss 0.142661 LR 0.000250 Time 0.020888 -2022-12-06 11:37:25,046 - Epoch: [167][ 660/ 1200] Overall Loss 0.142423 Objective Loss 0.142423 LR 0.000250 Time 0.020882 -2022-12-06 11:37:25,248 - Epoch: [167][ 670/ 1200] Overall Loss 0.142411 Objective Loss 0.142411 LR 0.000250 Time 0.020870 -2022-12-06 11:37:25,454 - Epoch: [167][ 680/ 1200] Overall Loss 0.142506 Objective Loss 0.142506 LR 0.000250 Time 0.020866 -2022-12-06 11:37:25,655 - Epoch: [167][ 690/ 1200] Overall Loss 0.142353 Objective Loss 0.142353 LR 0.000250 Time 0.020855 -2022-12-06 11:37:25,862 - Epoch: [167][ 700/ 1200] Overall Loss 0.142358 Objective Loss 0.142358 LR 0.000250 Time 0.020851 -2022-12-06 11:37:26,063 - Epoch: [167][ 710/ 1200] Overall Loss 0.142167 Objective Loss 0.142167 LR 0.000250 Time 0.020840 -2022-12-06 11:37:26,269 - Epoch: [167][ 720/ 1200] Overall Loss 0.142147 Objective Loss 0.142147 LR 0.000250 Time 0.020836 -2022-12-06 11:37:26,471 - Epoch: [167][ 730/ 1200] Overall Loss 0.142020 Objective Loss 0.142020 LR 0.000250 Time 0.020826 -2022-12-06 11:37:26,677 - Epoch: [167][ 740/ 1200] Overall Loss 0.142149 Objective Loss 0.142149 LR 0.000250 Time 0.020822 -2022-12-06 11:37:26,878 - Epoch: [167][ 750/ 1200] Overall Loss 0.141914 Objective Loss 0.141914 LR 0.000250 Time 0.020812 -2022-12-06 11:37:27,084 - Epoch: [167][ 760/ 1200] Overall Loss 0.141772 Objective Loss 0.141772 LR 0.000250 Time 0.020809 -2022-12-06 11:37:27,285 - Epoch: [167][ 770/ 1200] Overall Loss 0.141967 Objective Loss 0.141967 LR 0.000250 Time 0.020799 -2022-12-06 11:37:27,491 - Epoch: [167][ 780/ 1200] Overall Loss 0.142051 Objective Loss 0.142051 LR 0.000250 Time 0.020795 -2022-12-06 11:37:27,692 - Epoch: [167][ 790/ 1200] Overall Loss 0.141960 Objective Loss 0.141960 LR 0.000250 Time 0.020786 -2022-12-06 11:37:27,898 - Epoch: [167][ 800/ 1200] Overall Loss 0.141859 Objective Loss 0.141859 LR 0.000250 Time 0.020783 -2022-12-06 11:37:28,100 - Epoch: [167][ 810/ 1200] Overall Loss 0.142060 Objective Loss 0.142060 LR 0.000250 Time 0.020774 -2022-12-06 11:37:28,306 - Epoch: [167][ 820/ 1200] Overall Loss 0.142113 Objective Loss 0.142113 LR 0.000250 Time 0.020771 -2022-12-06 11:37:28,507 - Epoch: [167][ 830/ 1200] Overall Loss 0.142215 Objective Loss 0.142215 LR 0.000250 Time 0.020763 -2022-12-06 11:37:28,713 - Epoch: [167][ 840/ 1200] Overall Loss 0.142109 Objective Loss 0.142109 LR 0.000250 Time 0.020760 -2022-12-06 11:37:28,914 - Epoch: [167][ 850/ 1200] Overall Loss 0.142029 Objective Loss 0.142029 LR 0.000250 Time 0.020751 -2022-12-06 11:37:29,119 - Epoch: [167][ 860/ 1200] Overall Loss 0.142309 Objective Loss 0.142309 LR 0.000250 Time 0.020749 -2022-12-06 11:37:29,320 - Epoch: [167][ 870/ 1200] Overall Loss 0.142394 Objective Loss 0.142394 LR 0.000250 Time 0.020741 -2022-12-06 11:37:29,526 - Epoch: [167][ 880/ 1200] Overall Loss 0.142396 Objective Loss 0.142396 LR 0.000250 Time 0.020738 -2022-12-06 11:37:29,728 - Epoch: [167][ 890/ 1200] Overall Loss 0.142447 Objective Loss 0.142447 LR 0.000250 Time 0.020731 -2022-12-06 11:37:29,934 - Epoch: [167][ 900/ 1200] Overall Loss 0.142543 Objective Loss 0.142543 LR 0.000250 Time 0.020729 -2022-12-06 11:37:30,134 - Epoch: [167][ 910/ 1200] Overall Loss 0.142523 Objective Loss 0.142523 LR 0.000250 Time 0.020721 -2022-12-06 11:37:30,340 - Epoch: [167][ 920/ 1200] Overall Loss 0.142647 Objective Loss 0.142647 LR 0.000250 Time 0.020719 -2022-12-06 11:37:30,541 - Epoch: [167][ 930/ 1200] Overall Loss 0.142730 Objective Loss 0.142730 LR 0.000250 Time 0.020711 -2022-12-06 11:37:30,747 - Epoch: [167][ 940/ 1200] Overall Loss 0.142906 Objective Loss 0.142906 LR 0.000250 Time 0.020710 -2022-12-06 11:37:30,948 - Epoch: [167][ 950/ 1200] Overall Loss 0.142911 Objective Loss 0.142911 LR 0.000250 Time 0.020703 -2022-12-06 11:37:31,154 - Epoch: [167][ 960/ 1200] Overall Loss 0.142906 Objective Loss 0.142906 LR 0.000250 Time 0.020701 -2022-12-06 11:37:31,356 - Epoch: [167][ 970/ 1200] Overall Loss 0.142900 Objective Loss 0.142900 LR 0.000250 Time 0.020695 -2022-12-06 11:37:31,562 - Epoch: [167][ 980/ 1200] Overall Loss 0.143044 Objective Loss 0.143044 LR 0.000250 Time 0.020694 -2022-12-06 11:37:31,763 - Epoch: [167][ 990/ 1200] Overall Loss 0.142971 Objective Loss 0.142971 LR 0.000250 Time 0.020687 -2022-12-06 11:37:31,969 - Epoch: [167][ 1000/ 1200] Overall Loss 0.142855 Objective Loss 0.142855 LR 0.000250 Time 0.020686 -2022-12-06 11:37:32,171 - Epoch: [167][ 1010/ 1200] Overall Loss 0.142874 Objective Loss 0.142874 LR 0.000250 Time 0.020680 -2022-12-06 11:37:32,377 - Epoch: [167][ 1020/ 1200] Overall Loss 0.142814 Objective Loss 0.142814 LR 0.000250 Time 0.020679 -2022-12-06 11:37:32,579 - Epoch: [167][ 1030/ 1200] Overall Loss 0.142902 Objective Loss 0.142902 LR 0.000250 Time 0.020674 -2022-12-06 11:37:32,785 - Epoch: [167][ 1040/ 1200] Overall Loss 0.142920 Objective Loss 0.142920 LR 0.000250 Time 0.020672 -2022-12-06 11:37:32,986 - Epoch: [167][ 1050/ 1200] Overall Loss 0.142958 Objective Loss 0.142958 LR 0.000250 Time 0.020666 -2022-12-06 11:37:33,191 - Epoch: [167][ 1060/ 1200] Overall Loss 0.143059 Objective Loss 0.143059 LR 0.000250 Time 0.020664 -2022-12-06 11:37:33,392 - Epoch: [167][ 1070/ 1200] Overall Loss 0.143419 Objective Loss 0.143419 LR 0.000250 Time 0.020658 -2022-12-06 11:37:33,597 - Epoch: [167][ 1080/ 1200] Overall Loss 0.143200 Objective Loss 0.143200 LR 0.000250 Time 0.020657 -2022-12-06 11:37:33,798 - Epoch: [167][ 1090/ 1200] Overall Loss 0.143054 Objective Loss 0.143054 LR 0.000250 Time 0.020651 -2022-12-06 11:37:34,004 - Epoch: [167][ 1100/ 1200] Overall Loss 0.143385 Objective Loss 0.143385 LR 0.000250 Time 0.020650 -2022-12-06 11:37:34,204 - Epoch: [167][ 1110/ 1200] Overall Loss 0.143307 Objective Loss 0.143307 LR 0.000250 Time 0.020644 -2022-12-06 11:37:34,411 - Epoch: [167][ 1120/ 1200] Overall Loss 0.143333 Objective Loss 0.143333 LR 0.000250 Time 0.020643 -2022-12-06 11:37:34,613 - Epoch: [167][ 1130/ 1200] Overall Loss 0.143345 Objective Loss 0.143345 LR 0.000250 Time 0.020640 -2022-12-06 11:37:34,825 - Epoch: [167][ 1140/ 1200] Overall Loss 0.143519 Objective Loss 0.143519 LR 0.000250 Time 0.020644 -2022-12-06 11:37:35,033 - Epoch: [167][ 1150/ 1200] Overall Loss 0.143672 Objective Loss 0.143672 LR 0.000250 Time 0.020645 -2022-12-06 11:37:35,245 - Epoch: [167][ 1160/ 1200] Overall Loss 0.143739 Objective Loss 0.143739 LR 0.000250 Time 0.020649 -2022-12-06 11:37:35,454 - Epoch: [167][ 1170/ 1200] Overall Loss 0.144009 Objective Loss 0.144009 LR 0.000250 Time 0.020650 -2022-12-06 11:37:35,665 - Epoch: [167][ 1180/ 1200] Overall Loss 0.144085 Objective Loss 0.144085 LR 0.000250 Time 0.020653 -2022-12-06 11:37:35,873 - Epoch: [167][ 1190/ 1200] Overall Loss 0.144189 Objective Loss 0.144189 LR 0.000250 Time 0.020655 -2022-12-06 11:37:36,103 - Epoch: [167][ 1200/ 1200] Overall Loss 0.144170 Objective Loss 0.144170 Top1 88.912134 Top5 99.372385 LR 0.000250 Time 0.020673 -2022-12-06 11:37:36,192 - --- validate (epoch=167)----------- -2022-12-06 11:37:36,192 - 34129 samples (256 per mini-batch) -2022-12-06 11:37:36,633 - Epoch: [167][ 10/ 134] Loss 0.243239 Top1 87.578125 Top5 98.710938 -2022-12-06 11:37:36,764 - Epoch: [167][ 20/ 134] Loss 0.242916 Top1 87.832031 Top5 98.671875 -2022-12-06 11:37:36,898 - Epoch: [167][ 30/ 134] Loss 0.237129 Top1 87.799479 Top5 98.606771 -2022-12-06 11:37:37,031 - Epoch: [167][ 40/ 134] Loss 0.235817 Top1 87.822266 Top5 98.544922 -2022-12-06 11:37:37,167 - Epoch: [167][ 50/ 134] Loss 0.241289 Top1 87.570312 Top5 98.515625 -2022-12-06 11:37:37,299 - Epoch: [167][ 60/ 134] Loss 0.240781 Top1 87.552083 Top5 98.476562 -2022-12-06 11:37:37,432 - Epoch: [167][ 70/ 134] Loss 0.236569 Top1 87.784598 Top5 98.537946 -2022-12-06 11:37:37,565 - Epoch: [167][ 80/ 134] Loss 0.234514 Top1 87.822266 Top5 98.549805 -2022-12-06 11:37:37,699 - Epoch: [167][ 90/ 134] Loss 0.231681 Top1 87.951389 Top5 98.580729 -2022-12-06 11:37:37,831 - Epoch: [167][ 100/ 134] Loss 0.231230 Top1 87.960938 Top5 98.554688 -2022-12-06 11:37:37,966 - Epoch: [167][ 110/ 134] Loss 0.229316 Top1 87.940341 Top5 98.583097 -2022-12-06 11:37:38,098 - Epoch: [167][ 120/ 134] Loss 0.231243 Top1 87.825521 Top5 98.561198 -2022-12-06 11:37:38,234 - Epoch: [167][ 130/ 134] Loss 0.229409 Top1 87.890625 Top5 98.587740 -2022-12-06 11:37:38,273 - Epoch: [167][ 134/ 134] Loss 0.228882 Top1 87.901784 Top5 98.590641 -2022-12-06 11:37:38,361 - ==> Top1: 87.902 Top5: 98.591 Loss: 0.229 - -2022-12-06 11:37:38,362 - ==> Confusion: -[[ 929 1 1 2 3 3 1 2 5 37 0 1 0 1 7 2 1 0 0 0 0] - [ 2 948 1 2 7 22 2 12 1 0 2 4 2 1 0 1 2 3 6 2 7] - [ 4 2 1016 13 4 1 14 8 0 1 7 5 2 4 2 1 0 2 2 4 11] - [ 1 1 11 957 1 1 0 1 0 0 9 0 4 1 9 1 1 3 13 0 6] - [ 12 5 2 0 955 2 1 2 0 4 2 3 0 2 11 5 6 2 1 2 3] - [ 1 16 0 2 5 987 1 19 3 4 1 13 2 5 1 1 1 0 1 4 2] - [ 1 1 5 3 1 1 1076 5 0 0 1 3 0 0 0 6 1 4 2 7 1] - [ 1 6 3 2 2 23 4 974 0 0 2 5 0 1 1 1 0 0 13 9 7] - [ 6 1 0 0 0 1 1 1 978 36 12 1 1 4 13 1 1 1 3 2 1] - [ 58 0 2 0 4 2 0 2 25 885 1 2 0 6 5 1 1 0 0 0 7] - [ 0 3 3 2 0 2 2 1 10 0 973 1 0 5 5 1 0 0 4 1 6] - [ 2 0 2 0 1 10 3 2 0 1 1 967 30 3 1 7 2 6 0 9 4] - [ 1 0 0 1 0 1 0 0 0 1 0 18 913 0 2 9 2 11 1 4 5] - [ 2 1 0 0 1 6 0 3 14 13 5 3 4 957 0 2 1 0 0 2 9] - [ 5 5 2 13 3 2 1 0 11 1 0 3 2 3 1066 0 0 1 8 0 4] - [ 1 0 1 0 2 0 1 1 2 0 0 6 6 3 0 998 4 11 0 3 4] - [ 3 2 1 2 1 0 0 0 2 0 0 5 3 1 0 8 1033 1 0 3 7] - [ 3 0 1 0 0 1 1 0 1 5 0 5 13 1 2 11 0 987 0 1 4] - [ 2 3 1 8 2 3 0 21 2 1 2 1 3 0 5 1 0 2 946 2 3] - [ 2 4 0 3 0 3 2 5 0 0 2 12 6 4 0 3 4 2 2 1022 4] - [ 96 210 152 93 94 145 65 150 79 75 166 80 314 237 134 112 174 87 142 193 10428]] - -2022-12-06 11:37:39,028 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:37:39,028 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:37:39,034 - - -2022-12-06 11:37:39,034 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:37:39,971 - Epoch: [168][ 10/ 1200] Overall Loss 0.152551 Objective Loss 0.152551 LR 0.000250 Time 0.093644 -2022-12-06 11:37:40,176 - Epoch: [168][ 20/ 1200] Overall Loss 0.139028 Objective Loss 0.139028 LR 0.000250 Time 0.057027 -2022-12-06 11:37:40,378 - Epoch: [168][ 30/ 1200] Overall Loss 0.139818 Objective Loss 0.139818 LR 0.000250 Time 0.044716 -2022-12-06 11:37:40,575 - Epoch: [168][ 40/ 1200] Overall Loss 0.141965 Objective Loss 0.141965 LR 0.000250 Time 0.038454 -2022-12-06 11:37:40,776 - Epoch: [168][ 50/ 1200] Overall Loss 0.139911 Objective Loss 0.139911 LR 0.000250 Time 0.034774 -2022-12-06 11:37:40,975 - Epoch: [168][ 60/ 1200] Overall Loss 0.137797 Objective Loss 0.137797 LR 0.000250 Time 0.032283 -2022-12-06 11:37:41,177 - Epoch: [168][ 70/ 1200] Overall Loss 0.141232 Objective Loss 0.141232 LR 0.000250 Time 0.030547 -2022-12-06 11:37:41,375 - Epoch: [168][ 80/ 1200] Overall Loss 0.141923 Objective Loss 0.141923 LR 0.000250 Time 0.029200 -2022-12-06 11:37:41,576 - Epoch: [168][ 90/ 1200] Overall Loss 0.141237 Objective Loss 0.141237 LR 0.000250 Time 0.028182 -2022-12-06 11:37:41,774 - Epoch: [168][ 100/ 1200] Overall Loss 0.140751 Objective Loss 0.140751 LR 0.000250 Time 0.027337 -2022-12-06 11:37:41,975 - Epoch: [168][ 110/ 1200] Overall Loss 0.140875 Objective Loss 0.140875 LR 0.000250 Time 0.026673 -2022-12-06 11:37:42,173 - Epoch: [168][ 120/ 1200] Overall Loss 0.139247 Objective Loss 0.139247 LR 0.000250 Time 0.026096 -2022-12-06 11:37:42,374 - Epoch: [168][ 130/ 1200] Overall Loss 0.140800 Objective Loss 0.140800 LR 0.000250 Time 0.025635 -2022-12-06 11:37:42,572 - Epoch: [168][ 140/ 1200] Overall Loss 0.140131 Objective Loss 0.140131 LR 0.000250 Time 0.025211 -2022-12-06 11:37:42,773 - Epoch: [168][ 150/ 1200] Overall Loss 0.139129 Objective Loss 0.139129 LR 0.000250 Time 0.024867 -2022-12-06 11:37:42,970 - Epoch: [168][ 160/ 1200] Overall Loss 0.138885 Objective Loss 0.138885 LR 0.000250 Time 0.024539 -2022-12-06 11:37:43,171 - Epoch: [168][ 170/ 1200] Overall Loss 0.138344 Objective Loss 0.138344 LR 0.000250 Time 0.024275 -2022-12-06 11:37:43,368 - Epoch: [168][ 180/ 1200] Overall Loss 0.139239 Objective Loss 0.139239 LR 0.000250 Time 0.024019 -2022-12-06 11:37:43,569 - Epoch: [168][ 190/ 1200] Overall Loss 0.138372 Objective Loss 0.138372 LR 0.000250 Time 0.023808 -2022-12-06 11:37:43,766 - Epoch: [168][ 200/ 1200] Overall Loss 0.138870 Objective Loss 0.138870 LR 0.000250 Time 0.023601 -2022-12-06 11:37:43,967 - Epoch: [168][ 210/ 1200] Overall Loss 0.138701 Objective Loss 0.138701 LR 0.000250 Time 0.023432 -2022-12-06 11:37:44,164 - Epoch: [168][ 220/ 1200] Overall Loss 0.139065 Objective Loss 0.139065 LR 0.000250 Time 0.023259 -2022-12-06 11:37:44,365 - Epoch: [168][ 230/ 1200] Overall Loss 0.138731 Objective Loss 0.138731 LR 0.000250 Time 0.023119 -2022-12-06 11:37:44,561 - Epoch: [168][ 240/ 1200] Overall Loss 0.138625 Objective Loss 0.138625 LR 0.000250 Time 0.022973 -2022-12-06 11:37:44,763 - Epoch: [168][ 250/ 1200] Overall Loss 0.137751 Objective Loss 0.137751 LR 0.000250 Time 0.022858 -2022-12-06 11:37:44,960 - Epoch: [168][ 260/ 1200] Overall Loss 0.137645 Objective Loss 0.137645 LR 0.000250 Time 0.022734 -2022-12-06 11:37:45,161 - Epoch: [168][ 270/ 1200] Overall Loss 0.137786 Objective Loss 0.137786 LR 0.000250 Time 0.022633 -2022-12-06 11:37:45,358 - Epoch: [168][ 280/ 1200] Overall Loss 0.138477 Objective Loss 0.138477 LR 0.000250 Time 0.022528 -2022-12-06 11:37:45,559 - Epoch: [168][ 290/ 1200] Overall Loss 0.138748 Objective Loss 0.138748 LR 0.000250 Time 0.022441 -2022-12-06 11:37:45,757 - Epoch: [168][ 300/ 1200] Overall Loss 0.139549 Objective Loss 0.139549 LR 0.000250 Time 0.022351 -2022-12-06 11:37:45,958 - Epoch: [168][ 310/ 1200] Overall Loss 0.140098 Objective Loss 0.140098 LR 0.000250 Time 0.022279 -2022-12-06 11:37:46,157 - Epoch: [168][ 320/ 1200] Overall Loss 0.140014 Objective Loss 0.140014 LR 0.000250 Time 0.022202 -2022-12-06 11:37:46,358 - Epoch: [168][ 330/ 1200] Overall Loss 0.140585 Objective Loss 0.140585 LR 0.000250 Time 0.022137 -2022-12-06 11:37:46,557 - Epoch: [168][ 340/ 1200] Overall Loss 0.140348 Objective Loss 0.140348 LR 0.000250 Time 0.022067 -2022-12-06 11:37:46,758 - Epoch: [168][ 350/ 1200] Overall Loss 0.140401 Objective Loss 0.140401 LR 0.000250 Time 0.022012 -2022-12-06 11:37:46,955 - Epoch: [168][ 360/ 1200] Overall Loss 0.140336 Objective Loss 0.140336 LR 0.000250 Time 0.021946 -2022-12-06 11:37:47,156 - Epoch: [168][ 370/ 1200] Overall Loss 0.140208 Objective Loss 0.140208 LR 0.000250 Time 0.021894 -2022-12-06 11:37:47,353 - Epoch: [168][ 380/ 1200] Overall Loss 0.140365 Objective Loss 0.140365 LR 0.000250 Time 0.021834 -2022-12-06 11:37:47,553 - Epoch: [168][ 390/ 1200] Overall Loss 0.140513 Objective Loss 0.140513 LR 0.000250 Time 0.021787 -2022-12-06 11:37:47,751 - Epoch: [168][ 400/ 1200] Overall Loss 0.140811 Objective Loss 0.140811 LR 0.000250 Time 0.021735 -2022-12-06 11:37:47,951 - Epoch: [168][ 410/ 1200] Overall Loss 0.141029 Objective Loss 0.141029 LR 0.000250 Time 0.021692 -2022-12-06 11:37:48,149 - Epoch: [168][ 420/ 1200] Overall Loss 0.140791 Objective Loss 0.140791 LR 0.000250 Time 0.021646 -2022-12-06 11:37:48,350 - Epoch: [168][ 430/ 1200] Overall Loss 0.141198 Objective Loss 0.141198 LR 0.000250 Time 0.021609 -2022-12-06 11:37:48,549 - Epoch: [168][ 440/ 1200] Overall Loss 0.141314 Objective Loss 0.141314 LR 0.000250 Time 0.021567 -2022-12-06 11:37:48,750 - Epoch: [168][ 450/ 1200] Overall Loss 0.141445 Objective Loss 0.141445 LR 0.000250 Time 0.021533 -2022-12-06 11:37:48,947 - Epoch: [168][ 460/ 1200] Overall Loss 0.141504 Objective Loss 0.141504 LR 0.000250 Time 0.021493 -2022-12-06 11:37:49,147 - Epoch: [168][ 470/ 1200] Overall Loss 0.141749 Objective Loss 0.141749 LR 0.000250 Time 0.021460 -2022-12-06 11:37:49,345 - Epoch: [168][ 480/ 1200] Overall Loss 0.141316 Objective Loss 0.141316 LR 0.000250 Time 0.021424 -2022-12-06 11:37:49,547 - Epoch: [168][ 490/ 1200] Overall Loss 0.141182 Objective Loss 0.141182 LR 0.000250 Time 0.021398 -2022-12-06 11:37:49,745 - Epoch: [168][ 500/ 1200] Overall Loss 0.141226 Objective Loss 0.141226 LR 0.000250 Time 0.021365 -2022-12-06 11:37:49,947 - Epoch: [168][ 510/ 1200] Overall Loss 0.141424 Objective Loss 0.141424 LR 0.000250 Time 0.021341 -2022-12-06 11:37:50,144 - Epoch: [168][ 520/ 1200] Overall Loss 0.141461 Objective Loss 0.141461 LR 0.000250 Time 0.021309 -2022-12-06 11:37:50,345 - Epoch: [168][ 530/ 1200] Overall Loss 0.141966 Objective Loss 0.141966 LR 0.000250 Time 0.021284 -2022-12-06 11:37:50,543 - Epoch: [168][ 540/ 1200] Overall Loss 0.142181 Objective Loss 0.142181 LR 0.000250 Time 0.021257 -2022-12-06 11:37:50,744 - Epoch: [168][ 550/ 1200] Overall Loss 0.142237 Objective Loss 0.142237 LR 0.000250 Time 0.021233 -2022-12-06 11:37:50,942 - Epoch: [168][ 560/ 1200] Overall Loss 0.141945 Objective Loss 0.141945 LR 0.000250 Time 0.021206 -2022-12-06 11:37:51,142 - Epoch: [168][ 570/ 1200] Overall Loss 0.141886 Objective Loss 0.141886 LR 0.000250 Time 0.021185 -2022-12-06 11:37:51,340 - Epoch: [168][ 580/ 1200] Overall Loss 0.142220 Objective Loss 0.142220 LR 0.000250 Time 0.021161 -2022-12-06 11:37:51,541 - Epoch: [168][ 590/ 1200] Overall Loss 0.142662 Objective Loss 0.142662 LR 0.000250 Time 0.021142 -2022-12-06 11:37:51,738 - Epoch: [168][ 600/ 1200] Overall Loss 0.142880 Objective Loss 0.142880 LR 0.000250 Time 0.021117 -2022-12-06 11:37:51,938 - Epoch: [168][ 610/ 1200] Overall Loss 0.142753 Objective Loss 0.142753 LR 0.000250 Time 0.021098 -2022-12-06 11:37:52,138 - Epoch: [168][ 620/ 1200] Overall Loss 0.143006 Objective Loss 0.143006 LR 0.000250 Time 0.021078 -2022-12-06 11:37:52,338 - Epoch: [168][ 630/ 1200] Overall Loss 0.143099 Objective Loss 0.143099 LR 0.000250 Time 0.021061 -2022-12-06 11:37:52,537 - Epoch: [168][ 640/ 1200] Overall Loss 0.143469 Objective Loss 0.143469 LR 0.000250 Time 0.021041 -2022-12-06 11:37:52,738 - Epoch: [168][ 650/ 1200] Overall Loss 0.143544 Objective Loss 0.143544 LR 0.000250 Time 0.021026 -2022-12-06 11:37:52,935 - Epoch: [168][ 660/ 1200] Overall Loss 0.143791 Objective Loss 0.143791 LR 0.000250 Time 0.021006 -2022-12-06 11:37:53,135 - Epoch: [168][ 670/ 1200] Overall Loss 0.143812 Objective Loss 0.143812 LR 0.000250 Time 0.020990 -2022-12-06 11:37:53,334 - Epoch: [168][ 680/ 1200] Overall Loss 0.143813 Objective Loss 0.143813 LR 0.000250 Time 0.020972 -2022-12-06 11:37:53,535 - Epoch: [168][ 690/ 1200] Overall Loss 0.143977 Objective Loss 0.143977 LR 0.000250 Time 0.020959 -2022-12-06 11:37:53,732 - Epoch: [168][ 700/ 1200] Overall Loss 0.143785 Objective Loss 0.143785 LR 0.000250 Time 0.020940 -2022-12-06 11:37:53,933 - Epoch: [168][ 710/ 1200] Overall Loss 0.144128 Objective Loss 0.144128 LR 0.000250 Time 0.020928 -2022-12-06 11:37:54,131 - Epoch: [168][ 720/ 1200] Overall Loss 0.144302 Objective Loss 0.144302 LR 0.000250 Time 0.020911 -2022-12-06 11:37:54,331 - Epoch: [168][ 730/ 1200] Overall Loss 0.144211 Objective Loss 0.144211 LR 0.000250 Time 0.020898 -2022-12-06 11:37:54,530 - Epoch: [168][ 740/ 1200] Overall Loss 0.144041 Objective Loss 0.144041 LR 0.000250 Time 0.020883 -2022-12-06 11:37:54,730 - Epoch: [168][ 750/ 1200] Overall Loss 0.144307 Objective Loss 0.144307 LR 0.000250 Time 0.020871 -2022-12-06 11:37:54,927 - Epoch: [168][ 760/ 1200] Overall Loss 0.144379 Objective Loss 0.144379 LR 0.000250 Time 0.020855 -2022-12-06 11:37:55,128 - Epoch: [168][ 770/ 1200] Overall Loss 0.144777 Objective Loss 0.144777 LR 0.000250 Time 0.020844 -2022-12-06 11:37:55,325 - Epoch: [168][ 780/ 1200] Overall Loss 0.144747 Objective Loss 0.144747 LR 0.000250 Time 0.020829 -2022-12-06 11:37:55,525 - Epoch: [168][ 790/ 1200] Overall Loss 0.144640 Objective Loss 0.144640 LR 0.000250 Time 0.020818 -2022-12-06 11:37:55,723 - Epoch: [168][ 800/ 1200] Overall Loss 0.144557 Objective Loss 0.144557 LR 0.000250 Time 0.020804 -2022-12-06 11:37:55,923 - Epoch: [168][ 810/ 1200] Overall Loss 0.144520 Objective Loss 0.144520 LR 0.000250 Time 0.020794 -2022-12-06 11:37:56,121 - Epoch: [168][ 820/ 1200] Overall Loss 0.144628 Objective Loss 0.144628 LR 0.000250 Time 0.020780 -2022-12-06 11:37:56,322 - Epoch: [168][ 830/ 1200] Overall Loss 0.144562 Objective Loss 0.144562 LR 0.000250 Time 0.020772 -2022-12-06 11:37:56,519 - Epoch: [168][ 840/ 1200] Overall Loss 0.144703 Objective Loss 0.144703 LR 0.000250 Time 0.020759 -2022-12-06 11:37:56,720 - Epoch: [168][ 850/ 1200] Overall Loss 0.145094 Objective Loss 0.145094 LR 0.000250 Time 0.020750 -2022-12-06 11:37:56,917 - Epoch: [168][ 860/ 1200] Overall Loss 0.145194 Objective Loss 0.145194 LR 0.000250 Time 0.020737 -2022-12-06 11:37:57,117 - Epoch: [168][ 870/ 1200] Overall Loss 0.145222 Objective Loss 0.145222 LR 0.000250 Time 0.020729 -2022-12-06 11:37:57,315 - Epoch: [168][ 880/ 1200] Overall Loss 0.145049 Objective Loss 0.145049 LR 0.000250 Time 0.020716 -2022-12-06 11:37:57,515 - Epoch: [168][ 890/ 1200] Overall Loss 0.145183 Objective Loss 0.145183 LR 0.000250 Time 0.020708 -2022-12-06 11:37:57,713 - Epoch: [168][ 900/ 1200] Overall Loss 0.145304 Objective Loss 0.145304 LR 0.000250 Time 0.020698 -2022-12-06 11:37:57,914 - Epoch: [168][ 910/ 1200] Overall Loss 0.145306 Objective Loss 0.145306 LR 0.000250 Time 0.020690 -2022-12-06 11:37:58,110 - Epoch: [168][ 920/ 1200] Overall Loss 0.145201 Objective Loss 0.145201 LR 0.000250 Time 0.020678 -2022-12-06 11:37:58,311 - Epoch: [168][ 930/ 1200] Overall Loss 0.145147 Objective Loss 0.145147 LR 0.000250 Time 0.020671 -2022-12-06 11:37:58,508 - Epoch: [168][ 940/ 1200] Overall Loss 0.145176 Objective Loss 0.145176 LR 0.000250 Time 0.020660 -2022-12-06 11:37:58,709 - Epoch: [168][ 950/ 1200] Overall Loss 0.145293 Objective Loss 0.145293 LR 0.000250 Time 0.020654 -2022-12-06 11:37:58,905 - Epoch: [168][ 960/ 1200] Overall Loss 0.145160 Objective Loss 0.145160 LR 0.000250 Time 0.020643 -2022-12-06 11:37:59,106 - Epoch: [168][ 970/ 1200] Overall Loss 0.145149 Objective Loss 0.145149 LR 0.000250 Time 0.020637 -2022-12-06 11:37:59,303 - Epoch: [168][ 980/ 1200] Overall Loss 0.145206 Objective Loss 0.145206 LR 0.000250 Time 0.020626 -2022-12-06 11:37:59,504 - Epoch: [168][ 990/ 1200] Overall Loss 0.145265 Objective Loss 0.145265 LR 0.000250 Time 0.020620 -2022-12-06 11:37:59,701 - Epoch: [168][ 1000/ 1200] Overall Loss 0.144979 Objective Loss 0.144979 LR 0.000250 Time 0.020610 -2022-12-06 11:37:59,902 - Epoch: [168][ 1010/ 1200] Overall Loss 0.145129 Objective Loss 0.145129 LR 0.000250 Time 0.020605 -2022-12-06 11:38:00,098 - Epoch: [168][ 1020/ 1200] Overall Loss 0.145108 Objective Loss 0.145108 LR 0.000250 Time 0.020595 -2022-12-06 11:38:00,299 - Epoch: [168][ 1030/ 1200] Overall Loss 0.145200 Objective Loss 0.145200 LR 0.000250 Time 0.020589 -2022-12-06 11:38:00,497 - Epoch: [168][ 1040/ 1200] Overall Loss 0.145356 Objective Loss 0.145356 LR 0.000250 Time 0.020580 -2022-12-06 11:38:00,698 - Epoch: [168][ 1050/ 1200] Overall Loss 0.145412 Objective Loss 0.145412 LR 0.000250 Time 0.020576 -2022-12-06 11:38:00,896 - Epoch: [168][ 1060/ 1200] Overall Loss 0.145401 Objective Loss 0.145401 LR 0.000250 Time 0.020568 -2022-12-06 11:38:01,096 - Epoch: [168][ 1070/ 1200] Overall Loss 0.145513 Objective Loss 0.145513 LR 0.000250 Time 0.020563 -2022-12-06 11:38:01,293 - Epoch: [168][ 1080/ 1200] Overall Loss 0.145716 Objective Loss 0.145716 LR 0.000250 Time 0.020554 -2022-12-06 11:38:01,494 - Epoch: [168][ 1090/ 1200] Overall Loss 0.145717 Objective Loss 0.145717 LR 0.000250 Time 0.020549 -2022-12-06 11:38:01,692 - Epoch: [168][ 1100/ 1200] Overall Loss 0.145703 Objective Loss 0.145703 LR 0.000250 Time 0.020541 -2022-12-06 11:38:01,892 - Epoch: [168][ 1110/ 1200] Overall Loss 0.145801 Objective Loss 0.145801 LR 0.000250 Time 0.020536 -2022-12-06 11:38:02,090 - Epoch: [168][ 1120/ 1200] Overall Loss 0.145829 Objective Loss 0.145829 LR 0.000250 Time 0.020529 -2022-12-06 11:38:02,292 - Epoch: [168][ 1130/ 1200] Overall Loss 0.146011 Objective Loss 0.146011 LR 0.000250 Time 0.020526 -2022-12-06 11:38:02,490 - Epoch: [168][ 1140/ 1200] Overall Loss 0.145994 Objective Loss 0.145994 LR 0.000250 Time 0.020519 -2022-12-06 11:38:02,691 - Epoch: [168][ 1150/ 1200] Overall Loss 0.145963 Objective Loss 0.145963 LR 0.000250 Time 0.020515 -2022-12-06 11:38:02,889 - Epoch: [168][ 1160/ 1200] Overall Loss 0.146199 Objective Loss 0.146199 LR 0.000250 Time 0.020508 -2022-12-06 11:38:03,090 - Epoch: [168][ 1170/ 1200] Overall Loss 0.146108 Objective Loss 0.146108 LR 0.000250 Time 0.020504 -2022-12-06 11:38:03,288 - Epoch: [168][ 1180/ 1200] Overall Loss 0.146084 Objective Loss 0.146084 LR 0.000250 Time 0.020498 -2022-12-06 11:38:03,490 - Epoch: [168][ 1190/ 1200] Overall Loss 0.145980 Objective Loss 0.145980 LR 0.000250 Time 0.020495 -2022-12-06 11:38:03,721 - Epoch: [168][ 1200/ 1200] Overall Loss 0.145988 Objective Loss 0.145988 Top1 90.376569 Top5 98.953975 LR 0.000250 Time 0.020516 -2022-12-06 11:38:03,809 - --- validate (epoch=168)----------- -2022-12-06 11:38:03,810 - 34129 samples (256 per mini-batch) -2022-12-06 11:38:04,254 - Epoch: [168][ 10/ 134] Loss 0.220371 Top1 87.968750 Top5 98.867188 -2022-12-06 11:38:04,388 - Epoch: [168][ 20/ 134] Loss 0.215258 Top1 88.085938 Top5 98.886719 -2022-12-06 11:38:04,518 - Epoch: [168][ 30/ 134] Loss 0.230360 Top1 87.851562 Top5 98.723958 -2022-12-06 11:38:04,654 - Epoch: [168][ 40/ 134] Loss 0.232970 Top1 87.919922 Top5 98.662109 -2022-12-06 11:38:04,795 - Epoch: [168][ 50/ 134] Loss 0.237358 Top1 88.031250 Top5 98.609375 -2022-12-06 11:38:04,935 - Epoch: [168][ 60/ 134] Loss 0.237476 Top1 88.053385 Top5 98.626302 -2022-12-06 11:38:05,069 - Epoch: [168][ 70/ 134] Loss 0.237647 Top1 88.208705 Top5 98.565848 -2022-12-06 11:38:05,204 - Epoch: [168][ 80/ 134] Loss 0.236822 Top1 88.266602 Top5 98.583984 -2022-12-06 11:38:05,337 - Epoch: [168][ 90/ 134] Loss 0.236224 Top1 88.307292 Top5 98.554688 -2022-12-06 11:38:05,474 - Epoch: [168][ 100/ 134] Loss 0.234733 Top1 88.347656 Top5 98.550781 -2022-12-06 11:38:05,606 - Epoch: [168][ 110/ 134] Loss 0.234503 Top1 88.277699 Top5 98.515625 -2022-12-06 11:38:05,743 - Epoch: [168][ 120/ 134] Loss 0.234958 Top1 88.222656 Top5 98.522135 -2022-12-06 11:38:05,876 - Epoch: [168][ 130/ 134] Loss 0.234682 Top1 88.248197 Top5 98.551683 -2022-12-06 11:38:05,913 - Epoch: [168][ 134/ 134] Loss 0.234886 Top1 88.270972 Top5 98.555481 -2022-12-06 11:38:06,007 - ==> Top1: 88.271 Top5: 98.555 Loss: 0.235 - -2022-12-06 11:38:06,008 - ==> Confusion: -[[ 926 1 2 0 2 5 0 0 6 36 0 1 0 2 7 1 0 0 4 0 3] - [ 0 935 1 3 6 21 2 12 2 0 3 4 1 1 0 0 3 1 16 4 12] - [ 5 2 1007 16 5 2 12 10 0 4 5 4 1 1 3 1 1 1 5 6 12] - [ 1 2 6 959 0 1 0 0 0 0 10 1 5 3 9 0 2 1 11 1 8] - [ 9 4 2 0 963 3 1 2 0 5 1 2 1 2 7 5 5 3 1 0 4] - [ 3 14 0 2 5 975 2 20 3 0 1 7 9 16 1 1 1 0 1 5 3] - [ 1 4 7 3 0 1 1073 4 0 1 1 2 1 0 0 3 2 4 1 8 2] - [ 1 5 2 3 3 19 6 960 0 0 3 3 2 0 1 0 0 0 26 11 9] - [ 4 4 0 1 1 2 1 1 974 38 9 2 0 6 13 1 1 0 2 1 3] - [ 58 0 1 0 5 2 0 3 22 882 1 1 0 12 4 1 1 0 1 0 7] - [ 0 1 2 1 0 0 1 3 7 2 974 1 2 13 3 0 0 0 2 0 7] - [ 2 0 2 0 0 8 3 2 3 0 1 987 19 4 1 3 1 6 0 5 4] - [ 0 0 0 1 2 2 0 1 0 0 0 22 916 0 1 6 2 6 0 3 7] - [ 1 1 0 0 0 5 0 3 9 9 1 2 4 973 0 2 4 0 0 1 8] - [ 6 5 0 11 3 3 0 0 14 1 1 3 2 4 1063 0 0 1 6 0 7] - [ 0 0 0 3 3 0 3 1 2 0 0 8 9 3 0 987 7 9 0 3 5] - [ 3 1 1 0 2 2 0 0 1 0 0 4 2 2 0 8 1035 1 0 2 8] - [ 3 0 1 2 0 1 1 0 0 4 1 5 18 1 1 9 0 985 0 0 4] - [ 2 4 2 5 2 2 0 15 2 2 2 2 2 0 8 1 0 0 952 1 4] - [ 3 3 1 2 0 4 2 6 0 1 2 10 7 9 0 3 3 1 2 1014 7] - [ 105 156 140 107 95 172 51 128 66 76 153 86 295 277 122 92 131 58 167 167 10582]] - -2022-12-06 11:38:06,678 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:38:06,679 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:38:06,684 - - -2022-12-06 11:38:06,685 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:38:07,619 - Epoch: [169][ 10/ 1200] Overall Loss 0.135753 Objective Loss 0.135753 LR 0.000250 Time 0.093345 -2022-12-06 11:38:07,822 - Epoch: [169][ 20/ 1200] Overall Loss 0.138439 Objective Loss 0.138439 LR 0.000250 Time 0.056815 -2022-12-06 11:38:08,014 - Epoch: [169][ 30/ 1200] Overall Loss 0.139487 Objective Loss 0.139487 LR 0.000250 Time 0.044254 -2022-12-06 11:38:08,206 - Epoch: [169][ 40/ 1200] Overall Loss 0.142081 Objective Loss 0.142081 LR 0.000250 Time 0.037980 -2022-12-06 11:38:08,398 - Epoch: [169][ 50/ 1200] Overall Loss 0.144440 Objective Loss 0.144440 LR 0.000250 Time 0.034202 -2022-12-06 11:38:08,589 - Epoch: [169][ 60/ 1200] Overall Loss 0.145040 Objective Loss 0.145040 LR 0.000250 Time 0.031681 -2022-12-06 11:38:08,780 - Epoch: [169][ 70/ 1200] Overall Loss 0.144665 Objective Loss 0.144665 LR 0.000250 Time 0.029881 -2022-12-06 11:38:08,971 - Epoch: [169][ 80/ 1200] Overall Loss 0.145777 Objective Loss 0.145777 LR 0.000250 Time 0.028528 -2022-12-06 11:38:09,163 - Epoch: [169][ 90/ 1200] Overall Loss 0.146071 Objective Loss 0.146071 LR 0.000250 Time 0.027482 -2022-12-06 11:38:09,354 - Epoch: [169][ 100/ 1200] Overall Loss 0.144158 Objective Loss 0.144158 LR 0.000250 Time 0.026638 -2022-12-06 11:38:09,545 - Epoch: [169][ 110/ 1200] Overall Loss 0.143014 Objective Loss 0.143014 LR 0.000250 Time 0.025947 -2022-12-06 11:38:09,736 - Epoch: [169][ 120/ 1200] Overall Loss 0.143651 Objective Loss 0.143651 LR 0.000250 Time 0.025374 -2022-12-06 11:38:09,927 - Epoch: [169][ 130/ 1200] Overall Loss 0.141724 Objective Loss 0.141724 LR 0.000250 Time 0.024888 -2022-12-06 11:38:10,119 - Epoch: [169][ 140/ 1200] Overall Loss 0.142118 Objective Loss 0.142118 LR 0.000250 Time 0.024476 -2022-12-06 11:38:10,310 - Epoch: [169][ 150/ 1200] Overall Loss 0.140583 Objective Loss 0.140583 LR 0.000250 Time 0.024110 -2022-12-06 11:38:10,500 - Epoch: [169][ 160/ 1200] Overall Loss 0.141981 Objective Loss 0.141981 LR 0.000250 Time 0.023792 -2022-12-06 11:38:10,691 - Epoch: [169][ 170/ 1200] Overall Loss 0.141642 Objective Loss 0.141642 LR 0.000250 Time 0.023512 -2022-12-06 11:38:10,882 - Epoch: [169][ 180/ 1200] Overall Loss 0.141222 Objective Loss 0.141222 LR 0.000250 Time 0.023262 -2022-12-06 11:38:11,073 - Epoch: [169][ 190/ 1200] Overall Loss 0.140988 Objective Loss 0.140988 LR 0.000250 Time 0.023040 -2022-12-06 11:38:11,264 - Epoch: [169][ 200/ 1200] Overall Loss 0.141530 Objective Loss 0.141530 LR 0.000250 Time 0.022841 -2022-12-06 11:38:11,455 - Epoch: [169][ 210/ 1200] Overall Loss 0.141331 Objective Loss 0.141331 LR 0.000250 Time 0.022661 -2022-12-06 11:38:11,646 - Epoch: [169][ 220/ 1200] Overall Loss 0.140767 Objective Loss 0.140767 LR 0.000250 Time 0.022498 -2022-12-06 11:38:11,837 - Epoch: [169][ 230/ 1200] Overall Loss 0.140631 Objective Loss 0.140631 LR 0.000250 Time 0.022347 -2022-12-06 11:38:12,029 - Epoch: [169][ 240/ 1200] Overall Loss 0.140699 Objective Loss 0.140699 LR 0.000250 Time 0.022210 -2022-12-06 11:38:12,220 - Epoch: [169][ 250/ 1200] Overall Loss 0.140547 Objective Loss 0.140547 LR 0.000250 Time 0.022084 -2022-12-06 11:38:12,411 - Epoch: [169][ 260/ 1200] Overall Loss 0.140480 Objective Loss 0.140480 LR 0.000250 Time 0.021967 -2022-12-06 11:38:12,602 - Epoch: [169][ 270/ 1200] Overall Loss 0.140725 Objective Loss 0.140725 LR 0.000250 Time 0.021861 -2022-12-06 11:38:12,794 - Epoch: [169][ 280/ 1200] Overall Loss 0.141035 Objective Loss 0.141035 LR 0.000250 Time 0.021761 -2022-12-06 11:38:12,985 - Epoch: [169][ 290/ 1200] Overall Loss 0.140531 Objective Loss 0.140531 LR 0.000250 Time 0.021668 -2022-12-06 11:38:13,176 - Epoch: [169][ 300/ 1200] Overall Loss 0.140770 Objective Loss 0.140770 LR 0.000250 Time 0.021582 -2022-12-06 11:38:13,367 - Epoch: [169][ 310/ 1200] Overall Loss 0.140727 Objective Loss 0.140727 LR 0.000250 Time 0.021500 -2022-12-06 11:38:13,559 - Epoch: [169][ 320/ 1200] Overall Loss 0.140434 Objective Loss 0.140434 LR 0.000250 Time 0.021425 -2022-12-06 11:38:13,750 - Epoch: [169][ 330/ 1200] Overall Loss 0.140380 Objective Loss 0.140380 LR 0.000250 Time 0.021353 -2022-12-06 11:38:13,941 - Epoch: [169][ 340/ 1200] Overall Loss 0.140203 Objective Loss 0.140203 LR 0.000250 Time 0.021286 -2022-12-06 11:38:14,132 - Epoch: [169][ 350/ 1200] Overall Loss 0.140172 Objective Loss 0.140172 LR 0.000250 Time 0.021221 -2022-12-06 11:38:14,323 - Epoch: [169][ 360/ 1200] Overall Loss 0.140177 Objective Loss 0.140177 LR 0.000250 Time 0.021161 -2022-12-06 11:38:14,513 - Epoch: [169][ 370/ 1200] Overall Loss 0.140855 Objective Loss 0.140855 LR 0.000250 Time 0.021102 -2022-12-06 11:38:14,705 - Epoch: [169][ 380/ 1200] Overall Loss 0.141554 Objective Loss 0.141554 LR 0.000250 Time 0.021050 -2022-12-06 11:38:14,895 - Epoch: [169][ 390/ 1200] Overall Loss 0.141795 Objective Loss 0.141795 LR 0.000250 Time 0.020998 -2022-12-06 11:38:15,086 - Epoch: [169][ 400/ 1200] Overall Loss 0.141489 Objective Loss 0.141489 LR 0.000250 Time 0.020948 -2022-12-06 11:38:15,277 - Epoch: [169][ 410/ 1200] Overall Loss 0.141375 Objective Loss 0.141375 LR 0.000250 Time 0.020902 -2022-12-06 11:38:15,468 - Epoch: [169][ 420/ 1200] Overall Loss 0.141898 Objective Loss 0.141898 LR 0.000250 Time 0.020858 -2022-12-06 11:38:15,660 - Epoch: [169][ 430/ 1200] Overall Loss 0.141826 Objective Loss 0.141826 LR 0.000250 Time 0.020816 -2022-12-06 11:38:15,851 - Epoch: [169][ 440/ 1200] Overall Loss 0.141725 Objective Loss 0.141725 LR 0.000250 Time 0.020776 -2022-12-06 11:38:16,042 - Epoch: [169][ 450/ 1200] Overall Loss 0.141397 Objective Loss 0.141397 LR 0.000250 Time 0.020739 -2022-12-06 11:38:16,233 - Epoch: [169][ 460/ 1200] Overall Loss 0.141526 Objective Loss 0.141526 LR 0.000250 Time 0.020702 -2022-12-06 11:38:16,425 - Epoch: [169][ 470/ 1200] Overall Loss 0.141424 Objective Loss 0.141424 LR 0.000250 Time 0.020669 -2022-12-06 11:38:16,616 - Epoch: [169][ 480/ 1200] Overall Loss 0.141658 Objective Loss 0.141658 LR 0.000250 Time 0.020635 -2022-12-06 11:38:16,807 - Epoch: [169][ 490/ 1200] Overall Loss 0.141626 Objective Loss 0.141626 LR 0.000250 Time 0.020603 -2022-12-06 11:38:16,999 - Epoch: [169][ 500/ 1200] Overall Loss 0.141826 Objective Loss 0.141826 LR 0.000250 Time 0.020572 -2022-12-06 11:38:17,190 - Epoch: [169][ 510/ 1200] Overall Loss 0.141957 Objective Loss 0.141957 LR 0.000250 Time 0.020543 -2022-12-06 11:38:17,381 - Epoch: [169][ 520/ 1200] Overall Loss 0.142484 Objective Loss 0.142484 LR 0.000250 Time 0.020515 -2022-12-06 11:38:17,572 - Epoch: [169][ 530/ 1200] Overall Loss 0.142518 Objective Loss 0.142518 LR 0.000250 Time 0.020487 -2022-12-06 11:38:17,763 - Epoch: [169][ 540/ 1200] Overall Loss 0.142140 Objective Loss 0.142140 LR 0.000250 Time 0.020460 -2022-12-06 11:38:17,954 - Epoch: [169][ 550/ 1200] Overall Loss 0.142259 Objective Loss 0.142259 LR 0.000250 Time 0.020434 -2022-12-06 11:38:18,145 - Epoch: [169][ 560/ 1200] Overall Loss 0.142377 Objective Loss 0.142377 LR 0.000250 Time 0.020409 -2022-12-06 11:38:18,337 - Epoch: [169][ 570/ 1200] Overall Loss 0.142575 Objective Loss 0.142575 LR 0.000250 Time 0.020386 -2022-12-06 11:38:18,528 - Epoch: [169][ 580/ 1200] Overall Loss 0.142573 Objective Loss 0.142573 LR 0.000250 Time 0.020365 -2022-12-06 11:38:18,719 - Epoch: [169][ 590/ 1200] Overall Loss 0.142820 Objective Loss 0.142820 LR 0.000250 Time 0.020342 -2022-12-06 11:38:18,910 - Epoch: [169][ 600/ 1200] Overall Loss 0.143031 Objective Loss 0.143031 LR 0.000250 Time 0.020320 -2022-12-06 11:38:19,100 - Epoch: [169][ 610/ 1200] Overall Loss 0.142794 Objective Loss 0.142794 LR 0.000250 Time 0.020298 -2022-12-06 11:38:19,291 - Epoch: [169][ 620/ 1200] Overall Loss 0.142545 Objective Loss 0.142545 LR 0.000250 Time 0.020278 -2022-12-06 11:38:19,482 - Epoch: [169][ 630/ 1200] Overall Loss 0.142415 Objective Loss 0.142415 LR 0.000250 Time 0.020258 -2022-12-06 11:38:19,674 - Epoch: [169][ 640/ 1200] Overall Loss 0.142332 Objective Loss 0.142332 LR 0.000250 Time 0.020239 -2022-12-06 11:38:19,864 - Epoch: [169][ 650/ 1200] Overall Loss 0.141839 Objective Loss 0.141839 LR 0.000250 Time 0.020220 -2022-12-06 11:38:20,055 - Epoch: [169][ 660/ 1200] Overall Loss 0.141975 Objective Loss 0.141975 LR 0.000250 Time 0.020203 -2022-12-06 11:38:20,247 - Epoch: [169][ 670/ 1200] Overall Loss 0.141860 Objective Loss 0.141860 LR 0.000250 Time 0.020186 -2022-12-06 11:38:20,437 - Epoch: [169][ 680/ 1200] Overall Loss 0.141845 Objective Loss 0.141845 LR 0.000250 Time 0.020169 -2022-12-06 11:38:20,628 - Epoch: [169][ 690/ 1200] Overall Loss 0.141996 Objective Loss 0.141996 LR 0.000250 Time 0.020152 -2022-12-06 11:38:20,819 - Epoch: [169][ 700/ 1200] Overall Loss 0.142208 Objective Loss 0.142208 LR 0.000250 Time 0.020137 -2022-12-06 11:38:21,011 - Epoch: [169][ 710/ 1200] Overall Loss 0.142247 Objective Loss 0.142247 LR 0.000250 Time 0.020122 -2022-12-06 11:38:21,202 - Epoch: [169][ 720/ 1200] Overall Loss 0.142234 Objective Loss 0.142234 LR 0.000250 Time 0.020107 -2022-12-06 11:38:21,394 - Epoch: [169][ 730/ 1200] Overall Loss 0.142184 Objective Loss 0.142184 LR 0.000250 Time 0.020093 -2022-12-06 11:38:21,586 - Epoch: [169][ 740/ 1200] Overall Loss 0.142231 Objective Loss 0.142231 LR 0.000250 Time 0.020081 -2022-12-06 11:38:21,776 - Epoch: [169][ 750/ 1200] Overall Loss 0.141993 Objective Loss 0.141993 LR 0.000250 Time 0.020066 -2022-12-06 11:38:21,967 - Epoch: [169][ 760/ 1200] Overall Loss 0.142049 Objective Loss 0.142049 LR 0.000250 Time 0.020053 -2022-12-06 11:38:22,158 - Epoch: [169][ 770/ 1200] Overall Loss 0.142332 Objective Loss 0.142332 LR 0.000250 Time 0.020040 -2022-12-06 11:38:22,350 - Epoch: [169][ 780/ 1200] Overall Loss 0.142450 Objective Loss 0.142450 LR 0.000250 Time 0.020027 -2022-12-06 11:38:22,541 - Epoch: [169][ 790/ 1200] Overall Loss 0.142604 Objective Loss 0.142604 LR 0.000250 Time 0.020015 -2022-12-06 11:38:22,731 - Epoch: [169][ 800/ 1200] Overall Loss 0.142566 Objective Loss 0.142566 LR 0.000250 Time 0.020003 -2022-12-06 11:38:22,923 - Epoch: [169][ 810/ 1200] Overall Loss 0.142885 Objective Loss 0.142885 LR 0.000250 Time 0.019991 -2022-12-06 11:38:23,114 - Epoch: [169][ 820/ 1200] Overall Loss 0.142951 Objective Loss 0.142951 LR 0.000250 Time 0.019980 -2022-12-06 11:38:23,305 - Epoch: [169][ 830/ 1200] Overall Loss 0.143044 Objective Loss 0.143044 LR 0.000250 Time 0.019969 -2022-12-06 11:38:23,496 - Epoch: [169][ 840/ 1200] Overall Loss 0.143055 Objective Loss 0.143055 LR 0.000250 Time 0.019958 -2022-12-06 11:38:23,688 - Epoch: [169][ 850/ 1200] Overall Loss 0.142964 Objective Loss 0.142964 LR 0.000250 Time 0.019948 -2022-12-06 11:38:23,879 - Epoch: [169][ 860/ 1200] Overall Loss 0.143039 Objective Loss 0.143039 LR 0.000250 Time 0.019937 -2022-12-06 11:38:24,070 - Epoch: [169][ 870/ 1200] Overall Loss 0.143013 Objective Loss 0.143013 LR 0.000250 Time 0.019928 -2022-12-06 11:38:24,262 - Epoch: [169][ 880/ 1200] Overall Loss 0.143136 Objective Loss 0.143136 LR 0.000250 Time 0.019918 -2022-12-06 11:38:24,452 - Epoch: [169][ 890/ 1200] Overall Loss 0.143246 Objective Loss 0.143246 LR 0.000250 Time 0.019908 -2022-12-06 11:38:24,643 - Epoch: [169][ 900/ 1200] Overall Loss 0.143506 Objective Loss 0.143506 LR 0.000250 Time 0.019898 -2022-12-06 11:38:24,834 - Epoch: [169][ 910/ 1200] Overall Loss 0.143645 Objective Loss 0.143645 LR 0.000250 Time 0.019889 -2022-12-06 11:38:25,025 - Epoch: [169][ 920/ 1200] Overall Loss 0.143767 Objective Loss 0.143767 LR 0.000250 Time 0.019879 -2022-12-06 11:38:25,216 - Epoch: [169][ 930/ 1200] Overall Loss 0.143709 Objective Loss 0.143709 LR 0.000250 Time 0.019870 -2022-12-06 11:38:25,407 - Epoch: [169][ 940/ 1200] Overall Loss 0.143743 Objective Loss 0.143743 LR 0.000250 Time 0.019862 -2022-12-06 11:38:25,598 - Epoch: [169][ 950/ 1200] Overall Loss 0.143822 Objective Loss 0.143822 LR 0.000250 Time 0.019853 -2022-12-06 11:38:25,790 - Epoch: [169][ 960/ 1200] Overall Loss 0.143807 Objective Loss 0.143807 LR 0.000250 Time 0.019846 -2022-12-06 11:38:25,981 - Epoch: [169][ 970/ 1200] Overall Loss 0.143863 Objective Loss 0.143863 LR 0.000250 Time 0.019837 -2022-12-06 11:38:26,172 - Epoch: [169][ 980/ 1200] Overall Loss 0.143954 Objective Loss 0.143954 LR 0.000250 Time 0.019830 -2022-12-06 11:38:26,364 - Epoch: [169][ 990/ 1200] Overall Loss 0.143901 Objective Loss 0.143901 LR 0.000250 Time 0.019822 -2022-12-06 11:38:26,555 - Epoch: [169][ 1000/ 1200] Overall Loss 0.144126 Objective Loss 0.144126 LR 0.000250 Time 0.019814 -2022-12-06 11:38:26,746 - Epoch: [169][ 1010/ 1200] Overall Loss 0.144245 Objective Loss 0.144245 LR 0.000250 Time 0.019807 -2022-12-06 11:38:26,937 - Epoch: [169][ 1020/ 1200] Overall Loss 0.144181 Objective Loss 0.144181 LR 0.000250 Time 0.019799 -2022-12-06 11:38:27,128 - Epoch: [169][ 1030/ 1200] Overall Loss 0.144035 Objective Loss 0.144035 LR 0.000250 Time 0.019792 -2022-12-06 11:38:27,319 - Epoch: [169][ 1040/ 1200] Overall Loss 0.144224 Objective Loss 0.144224 LR 0.000250 Time 0.019785 -2022-12-06 11:38:27,510 - Epoch: [169][ 1050/ 1200] Overall Loss 0.144348 Objective Loss 0.144348 LR 0.000250 Time 0.019778 -2022-12-06 11:38:27,701 - Epoch: [169][ 1060/ 1200] Overall Loss 0.144520 Objective Loss 0.144520 LR 0.000250 Time 0.019771 -2022-12-06 11:38:27,892 - Epoch: [169][ 1070/ 1200] Overall Loss 0.144560 Objective Loss 0.144560 LR 0.000250 Time 0.019764 -2022-12-06 11:38:28,083 - Epoch: [169][ 1080/ 1200] Overall Loss 0.144535 Objective Loss 0.144535 LR 0.000250 Time 0.019758 -2022-12-06 11:38:28,274 - Epoch: [169][ 1090/ 1200] Overall Loss 0.144688 Objective Loss 0.144688 LR 0.000250 Time 0.019751 -2022-12-06 11:38:28,466 - Epoch: [169][ 1100/ 1200] Overall Loss 0.144532 Objective Loss 0.144532 LR 0.000250 Time 0.019745 -2022-12-06 11:38:28,657 - Epoch: [169][ 1110/ 1200] Overall Loss 0.144454 Objective Loss 0.144454 LR 0.000250 Time 0.019739 -2022-12-06 11:38:28,848 - Epoch: [169][ 1120/ 1200] Overall Loss 0.144411 Objective Loss 0.144411 LR 0.000250 Time 0.019733 -2022-12-06 11:38:29,039 - Epoch: [169][ 1130/ 1200] Overall Loss 0.144506 Objective Loss 0.144506 LR 0.000250 Time 0.019727 -2022-12-06 11:38:29,230 - Epoch: [169][ 1140/ 1200] Overall Loss 0.144452 Objective Loss 0.144452 LR 0.000250 Time 0.019721 -2022-12-06 11:38:29,422 - Epoch: [169][ 1150/ 1200] Overall Loss 0.144552 Objective Loss 0.144552 LR 0.000250 Time 0.019715 -2022-12-06 11:38:29,613 - Epoch: [169][ 1160/ 1200] Overall Loss 0.144590 Objective Loss 0.144590 LR 0.000250 Time 0.019710 -2022-12-06 11:38:29,804 - Epoch: [169][ 1170/ 1200] Overall Loss 0.144464 Objective Loss 0.144464 LR 0.000250 Time 0.019704 -2022-12-06 11:38:29,995 - Epoch: [169][ 1180/ 1200] Overall Loss 0.144485 Objective Loss 0.144485 LR 0.000250 Time 0.019699 -2022-12-06 11:38:30,186 - Epoch: [169][ 1190/ 1200] Overall Loss 0.144641 Objective Loss 0.144641 LR 0.000250 Time 0.019693 -2022-12-06 11:38:30,407 - Epoch: [169][ 1200/ 1200] Overall Loss 0.144804 Objective Loss 0.144804 Top1 89.330544 Top5 99.581590 LR 0.000250 Time 0.019713 -2022-12-06 11:38:30,501 - --- validate (epoch=169)----------- -2022-12-06 11:38:30,502 - 34129 samples (256 per mini-batch) -2022-12-06 11:38:30,944 - Epoch: [169][ 10/ 134] Loss 0.258481 Top1 87.187500 Top5 98.554688 -2022-12-06 11:38:31,070 - Epoch: [169][ 20/ 134] Loss 0.236543 Top1 87.734375 Top5 98.789062 -2022-12-06 11:38:31,195 - Epoch: [169][ 30/ 134] Loss 0.237815 Top1 87.786458 Top5 98.606771 -2022-12-06 11:38:31,322 - Epoch: [169][ 40/ 134] Loss 0.237960 Top1 87.675781 Top5 98.505859 -2022-12-06 11:38:31,449 - Epoch: [169][ 50/ 134] Loss 0.230614 Top1 87.898438 Top5 98.570312 -2022-12-06 11:38:31,576 - Epoch: [169][ 60/ 134] Loss 0.233383 Top1 87.832031 Top5 98.567708 -2022-12-06 11:38:31,702 - Epoch: [169][ 70/ 134] Loss 0.232678 Top1 87.756696 Top5 98.565848 -2022-12-06 11:38:31,829 - Epoch: [169][ 80/ 134] Loss 0.234977 Top1 87.690430 Top5 98.520508 -2022-12-06 11:38:31,956 - Epoch: [169][ 90/ 134] Loss 0.238905 Top1 87.612847 Top5 98.493924 -2022-12-06 11:38:32,084 - Epoch: [169][ 100/ 134] Loss 0.236544 Top1 87.648438 Top5 98.542969 -2022-12-06 11:38:32,210 - Epoch: [169][ 110/ 134] Loss 0.233327 Top1 87.755682 Top5 98.558239 -2022-12-06 11:38:32,339 - Epoch: [169][ 120/ 134] Loss 0.234820 Top1 87.705078 Top5 98.535156 -2022-12-06 11:38:32,467 - Epoch: [169][ 130/ 134] Loss 0.233992 Top1 87.740385 Top5 98.554688 -2022-12-06 11:38:32,504 - Epoch: [169][ 134/ 134] Loss 0.233142 Top1 87.740631 Top5 98.561341 -2022-12-06 11:38:32,605 - ==> Top1: 87.741 Top5: 98.561 Loss: 0.233 - -2022-12-06 11:38:32,606 - ==> Confusion: -[[ 918 0 1 1 5 6 0 1 7 38 1 1 2 3 4 1 1 0 2 1 3] - [ 2 935 3 3 10 25 2 18 0 0 0 3 1 1 0 1 4 0 6 6 7] - [ 6 1 1013 10 7 2 20 13 0 2 4 2 1 2 1 3 2 2 1 3 8] - [ 1 1 11 956 1 4 1 0 0 1 9 0 7 1 9 0 2 2 8 0 6] - [ 7 3 1 0 968 2 1 1 1 4 1 5 0 3 6 4 5 2 0 2 4] - [ 0 9 0 2 8 993 2 14 1 3 1 9 3 14 0 1 1 1 0 4 3] - [ 1 3 5 1 0 1 1081 3 0 0 0 2 0 4 0 4 2 1 2 6 2] - [ 0 6 4 3 3 27 6 970 0 0 1 4 0 4 0 0 2 0 9 9 6] - [ 7 1 0 1 0 2 1 2 976 37 10 2 0 9 9 0 3 0 1 1 2] - [ 40 0 1 0 8 2 0 3 26 896 1 2 0 13 2 2 1 0 1 0 3] - [ 0 1 3 2 1 1 2 6 5 2 970 0 1 14 3 1 0 0 2 1 4] - [ 2 0 2 0 0 10 4 1 1 0 1 968 22 4 1 10 4 5 0 11 5] - [ 0 0 0 2 2 3 0 0 1 1 0 25 904 1 1 11 2 5 0 6 5] - [ 1 1 0 0 1 10 0 2 7 7 3 3 3 969 0 3 5 0 0 3 5] - [ 5 4 1 12 5 3 0 1 15 0 0 3 2 6 1057 1 1 1 6 0 7] - [ 0 0 0 1 1 0 2 1 1 0 1 5 6 1 0 999 6 12 0 4 3] - [ 1 1 1 1 2 3 1 0 0 1 0 1 2 3 0 8 1037 0 0 3 7] - [ 3 0 2 1 2 0 0 0 1 5 0 5 16 2 1 14 1 979 0 1 3] - [ 3 4 4 10 1 3 0 25 0 1 2 4 4 0 9 0 0 1 932 2 3] - [ 3 2 1 1 0 5 4 4 0 1 0 8 7 7 0 5 5 1 2 1018 6] - [ 99 185 162 96 117 189 81 137 66 77 144 78 279 301 106 119 184 73 108 228 10397]] - -2022-12-06 11:38:33,173 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:38:33,173 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:38:33,179 - - -2022-12-06 11:38:33,179 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:38:34,220 - Epoch: [170][ 10/ 1200] Overall Loss 0.128619 Objective Loss 0.128619 LR 0.000250 Time 0.104074 -2022-12-06 11:38:34,413 - Epoch: [170][ 20/ 1200] Overall Loss 0.127524 Objective Loss 0.127524 LR 0.000250 Time 0.061647 -2022-12-06 11:38:34,604 - Epoch: [170][ 30/ 1200] Overall Loss 0.127610 Objective Loss 0.127610 LR 0.000250 Time 0.047443 -2022-12-06 11:38:34,794 - Epoch: [170][ 40/ 1200] Overall Loss 0.133030 Objective Loss 0.133030 LR 0.000250 Time 0.040332 -2022-12-06 11:38:34,985 - Epoch: [170][ 50/ 1200] Overall Loss 0.132450 Objective Loss 0.132450 LR 0.000250 Time 0.036074 -2022-12-06 11:38:35,176 - Epoch: [170][ 60/ 1200] Overall Loss 0.136080 Objective Loss 0.136080 LR 0.000250 Time 0.033227 -2022-12-06 11:38:35,367 - Epoch: [170][ 70/ 1200] Overall Loss 0.136611 Objective Loss 0.136611 LR 0.000250 Time 0.031199 -2022-12-06 11:38:35,558 - Epoch: [170][ 80/ 1200] Overall Loss 0.136066 Objective Loss 0.136066 LR 0.000250 Time 0.029678 -2022-12-06 11:38:35,748 - Epoch: [170][ 90/ 1200] Overall Loss 0.136796 Objective Loss 0.136796 LR 0.000250 Time 0.028489 -2022-12-06 11:38:35,938 - Epoch: [170][ 100/ 1200] Overall Loss 0.138587 Objective Loss 0.138587 LR 0.000250 Time 0.027538 -2022-12-06 11:38:36,129 - Epoch: [170][ 110/ 1200] Overall Loss 0.140832 Objective Loss 0.140832 LR 0.000250 Time 0.026765 -2022-12-06 11:38:36,320 - Epoch: [170][ 120/ 1200] Overall Loss 0.139777 Objective Loss 0.139777 LR 0.000250 Time 0.026120 -2022-12-06 11:38:36,511 - Epoch: [170][ 130/ 1200] Overall Loss 0.140659 Objective Loss 0.140659 LR 0.000250 Time 0.025574 -2022-12-06 11:38:36,701 - Epoch: [170][ 140/ 1200] Overall Loss 0.140632 Objective Loss 0.140632 LR 0.000250 Time 0.025105 -2022-12-06 11:38:36,892 - Epoch: [170][ 150/ 1200] Overall Loss 0.140974 Objective Loss 0.140974 LR 0.000250 Time 0.024698 -2022-12-06 11:38:37,082 - Epoch: [170][ 160/ 1200] Overall Loss 0.141553 Objective Loss 0.141553 LR 0.000250 Time 0.024339 -2022-12-06 11:38:37,272 - Epoch: [170][ 170/ 1200] Overall Loss 0.141505 Objective Loss 0.141505 LR 0.000250 Time 0.024020 -2022-12-06 11:38:37,462 - Epoch: [170][ 180/ 1200] Overall Loss 0.141846 Objective Loss 0.141846 LR 0.000250 Time 0.023738 -2022-12-06 11:38:37,651 - Epoch: [170][ 190/ 1200] Overall Loss 0.142029 Objective Loss 0.142029 LR 0.000250 Time 0.023484 -2022-12-06 11:38:37,842 - Epoch: [170][ 200/ 1200] Overall Loss 0.141642 Objective Loss 0.141642 LR 0.000250 Time 0.023259 -2022-12-06 11:38:38,032 - Epoch: [170][ 210/ 1200] Overall Loss 0.143073 Objective Loss 0.143073 LR 0.000250 Time 0.023056 -2022-12-06 11:38:38,225 - Epoch: [170][ 220/ 1200] Overall Loss 0.143176 Objective Loss 0.143176 LR 0.000250 Time 0.022880 -2022-12-06 11:38:38,415 - Epoch: [170][ 230/ 1200] Overall Loss 0.143859 Objective Loss 0.143859 LR 0.000250 Time 0.022712 -2022-12-06 11:38:38,607 - Epoch: [170][ 240/ 1200] Overall Loss 0.144206 Objective Loss 0.144206 LR 0.000250 Time 0.022560 -2022-12-06 11:38:38,797 - Epoch: [170][ 250/ 1200] Overall Loss 0.143790 Objective Loss 0.143790 LR 0.000250 Time 0.022419 -2022-12-06 11:38:38,989 - Epoch: [170][ 260/ 1200] Overall Loss 0.143398 Objective Loss 0.143398 LR 0.000250 Time 0.022290 -2022-12-06 11:38:39,179 - Epoch: [170][ 270/ 1200] Overall Loss 0.142901 Objective Loss 0.142901 LR 0.000250 Time 0.022167 -2022-12-06 11:38:39,370 - Epoch: [170][ 280/ 1200] Overall Loss 0.142487 Objective Loss 0.142487 LR 0.000250 Time 0.022055 -2022-12-06 11:38:39,560 - Epoch: [170][ 290/ 1200] Overall Loss 0.142427 Objective Loss 0.142427 LR 0.000250 Time 0.021949 -2022-12-06 11:38:39,751 - Epoch: [170][ 300/ 1200] Overall Loss 0.142351 Objective Loss 0.142351 LR 0.000250 Time 0.021852 -2022-12-06 11:38:39,942 - Epoch: [170][ 310/ 1200] Overall Loss 0.142106 Objective Loss 0.142106 LR 0.000250 Time 0.021761 -2022-12-06 11:38:40,132 - Epoch: [170][ 320/ 1200] Overall Loss 0.142368 Objective Loss 0.142368 LR 0.000250 Time 0.021674 -2022-12-06 11:38:40,323 - Epoch: [170][ 330/ 1200] Overall Loss 0.142590 Objective Loss 0.142590 LR 0.000250 Time 0.021593 -2022-12-06 11:38:40,514 - Epoch: [170][ 340/ 1200] Overall Loss 0.142469 Objective Loss 0.142469 LR 0.000250 Time 0.021520 -2022-12-06 11:38:40,705 - Epoch: [170][ 350/ 1200] Overall Loss 0.142670 Objective Loss 0.142670 LR 0.000250 Time 0.021449 -2022-12-06 11:38:40,896 - Epoch: [170][ 360/ 1200] Overall Loss 0.142796 Objective Loss 0.142796 LR 0.000250 Time 0.021380 -2022-12-06 11:38:41,086 - Epoch: [170][ 370/ 1200] Overall Loss 0.143036 Objective Loss 0.143036 LR 0.000250 Time 0.021316 -2022-12-06 11:38:41,277 - Epoch: [170][ 380/ 1200] Overall Loss 0.143028 Objective Loss 0.143028 LR 0.000250 Time 0.021255 -2022-12-06 11:38:41,468 - Epoch: [170][ 390/ 1200] Overall Loss 0.143164 Objective Loss 0.143164 LR 0.000250 Time 0.021199 -2022-12-06 11:38:41,657 - Epoch: [170][ 400/ 1200] Overall Loss 0.143092 Objective Loss 0.143092 LR 0.000250 Time 0.021141 -2022-12-06 11:38:41,848 - Epoch: [170][ 410/ 1200] Overall Loss 0.142764 Objective Loss 0.142764 LR 0.000250 Time 0.021090 -2022-12-06 11:38:42,039 - Epoch: [170][ 420/ 1200] Overall Loss 0.143252 Objective Loss 0.143252 LR 0.000250 Time 0.021040 -2022-12-06 11:38:42,230 - Epoch: [170][ 430/ 1200] Overall Loss 0.143005 Objective Loss 0.143005 LR 0.000250 Time 0.020993 -2022-12-06 11:38:42,421 - Epoch: [170][ 440/ 1200] Overall Loss 0.142783 Objective Loss 0.142783 LR 0.000250 Time 0.020949 -2022-12-06 11:38:42,611 - Epoch: [170][ 450/ 1200] Overall Loss 0.142526 Objective Loss 0.142526 LR 0.000250 Time 0.020905 -2022-12-06 11:38:42,801 - Epoch: [170][ 460/ 1200] Overall Loss 0.142263 Objective Loss 0.142263 LR 0.000250 Time 0.020864 -2022-12-06 11:38:42,991 - Epoch: [170][ 470/ 1200] Overall Loss 0.142407 Objective Loss 0.142407 LR 0.000250 Time 0.020823 -2022-12-06 11:38:43,181 - Epoch: [170][ 480/ 1200] Overall Loss 0.142675 Objective Loss 0.142675 LR 0.000250 Time 0.020783 -2022-12-06 11:38:43,371 - Epoch: [170][ 490/ 1200] Overall Loss 0.142883 Objective Loss 0.142883 LR 0.000250 Time 0.020746 -2022-12-06 11:38:43,563 - Epoch: [170][ 500/ 1200] Overall Loss 0.142904 Objective Loss 0.142904 LR 0.000250 Time 0.020713 -2022-12-06 11:38:43,754 - Epoch: [170][ 510/ 1200] Overall Loss 0.142303 Objective Loss 0.142303 LR 0.000250 Time 0.020680 -2022-12-06 11:38:43,945 - Epoch: [170][ 520/ 1200] Overall Loss 0.142396 Objective Loss 0.142396 LR 0.000250 Time 0.020649 -2022-12-06 11:38:44,136 - Epoch: [170][ 530/ 1200] Overall Loss 0.142895 Objective Loss 0.142895 LR 0.000250 Time 0.020618 -2022-12-06 11:38:44,327 - Epoch: [170][ 540/ 1200] Overall Loss 0.142827 Objective Loss 0.142827 LR 0.000250 Time 0.020590 -2022-12-06 11:38:44,518 - Epoch: [170][ 550/ 1200] Overall Loss 0.143057 Objective Loss 0.143057 LR 0.000250 Time 0.020562 -2022-12-06 11:38:44,709 - Epoch: [170][ 560/ 1200] Overall Loss 0.142964 Objective Loss 0.142964 LR 0.000250 Time 0.020535 -2022-12-06 11:38:44,900 - Epoch: [170][ 570/ 1200] Overall Loss 0.142936 Objective Loss 0.142936 LR 0.000250 Time 0.020509 -2022-12-06 11:38:45,091 - Epoch: [170][ 580/ 1200] Overall Loss 0.143033 Objective Loss 0.143033 LR 0.000250 Time 0.020483 -2022-12-06 11:38:45,282 - Epoch: [170][ 590/ 1200] Overall Loss 0.143277 Objective Loss 0.143277 LR 0.000250 Time 0.020458 -2022-12-06 11:38:45,472 - Epoch: [170][ 600/ 1200] Overall Loss 0.143174 Objective Loss 0.143174 LR 0.000250 Time 0.020434 -2022-12-06 11:38:45,663 - Epoch: [170][ 610/ 1200] Overall Loss 0.143220 Objective Loss 0.143220 LR 0.000250 Time 0.020410 -2022-12-06 11:38:45,853 - Epoch: [170][ 620/ 1200] Overall Loss 0.142841 Objective Loss 0.142841 LR 0.000250 Time 0.020387 -2022-12-06 11:38:46,043 - Epoch: [170][ 630/ 1200] Overall Loss 0.142710 Objective Loss 0.142710 LR 0.000250 Time 0.020364 -2022-12-06 11:38:46,235 - Epoch: [170][ 640/ 1200] Overall Loss 0.142733 Objective Loss 0.142733 LR 0.000250 Time 0.020345 -2022-12-06 11:38:46,426 - Epoch: [170][ 650/ 1200] Overall Loss 0.142818 Objective Loss 0.142818 LR 0.000250 Time 0.020325 -2022-12-06 11:38:46,617 - Epoch: [170][ 660/ 1200] Overall Loss 0.142896 Objective Loss 0.142896 LR 0.000250 Time 0.020306 -2022-12-06 11:38:46,807 - Epoch: [170][ 670/ 1200] Overall Loss 0.142869 Objective Loss 0.142869 LR 0.000250 Time 0.020286 -2022-12-06 11:38:46,997 - Epoch: [170][ 680/ 1200] Overall Loss 0.142871 Objective Loss 0.142871 LR 0.000250 Time 0.020266 -2022-12-06 11:38:47,187 - Epoch: [170][ 690/ 1200] Overall Loss 0.142830 Objective Loss 0.142830 LR 0.000250 Time 0.020247 -2022-12-06 11:38:47,378 - Epoch: [170][ 700/ 1200] Overall Loss 0.142926 Objective Loss 0.142926 LR 0.000250 Time 0.020229 -2022-12-06 11:38:47,568 - Epoch: [170][ 710/ 1200] Overall Loss 0.142669 Objective Loss 0.142669 LR 0.000250 Time 0.020211 -2022-12-06 11:38:47,759 - Epoch: [170][ 720/ 1200] Overall Loss 0.142535 Objective Loss 0.142535 LR 0.000250 Time 0.020195 -2022-12-06 11:38:47,950 - Epoch: [170][ 730/ 1200] Overall Loss 0.142602 Objective Loss 0.142602 LR 0.000250 Time 0.020179 -2022-12-06 11:38:48,141 - Epoch: [170][ 740/ 1200] Overall Loss 0.142615 Objective Loss 0.142615 LR 0.000250 Time 0.020164 -2022-12-06 11:38:48,332 - Epoch: [170][ 750/ 1200] Overall Loss 0.142713 Objective Loss 0.142713 LR 0.000250 Time 0.020149 -2022-12-06 11:38:48,523 - Epoch: [170][ 760/ 1200] Overall Loss 0.142623 Objective Loss 0.142623 LR 0.000250 Time 0.020134 -2022-12-06 11:38:48,714 - Epoch: [170][ 770/ 1200] Overall Loss 0.143066 Objective Loss 0.143066 LR 0.000250 Time 0.020120 -2022-12-06 11:38:48,905 - Epoch: [170][ 780/ 1200] Overall Loss 0.143154 Objective Loss 0.143154 LR 0.000250 Time 0.020107 -2022-12-06 11:38:49,095 - Epoch: [170][ 790/ 1200] Overall Loss 0.142795 Objective Loss 0.142795 LR 0.000250 Time 0.020093 -2022-12-06 11:38:49,287 - Epoch: [170][ 800/ 1200] Overall Loss 0.142716 Objective Loss 0.142716 LR 0.000250 Time 0.020079 -2022-12-06 11:38:49,478 - Epoch: [170][ 810/ 1200] Overall Loss 0.142649 Objective Loss 0.142649 LR 0.000250 Time 0.020067 -2022-12-06 11:38:49,670 - Epoch: [170][ 820/ 1200] Overall Loss 0.142523 Objective Loss 0.142523 LR 0.000250 Time 0.020056 -2022-12-06 11:38:49,860 - Epoch: [170][ 830/ 1200] Overall Loss 0.142757 Objective Loss 0.142757 LR 0.000250 Time 0.020043 -2022-12-06 11:38:50,052 - Epoch: [170][ 840/ 1200] Overall Loss 0.142798 Objective Loss 0.142798 LR 0.000250 Time 0.020032 -2022-12-06 11:38:50,243 - Epoch: [170][ 850/ 1200] Overall Loss 0.142752 Objective Loss 0.142752 LR 0.000250 Time 0.020020 -2022-12-06 11:38:50,434 - Epoch: [170][ 860/ 1200] Overall Loss 0.142772 Objective Loss 0.142772 LR 0.000250 Time 0.020009 -2022-12-06 11:38:50,624 - Epoch: [170][ 870/ 1200] Overall Loss 0.142718 Objective Loss 0.142718 LR 0.000250 Time 0.019996 -2022-12-06 11:38:50,815 - Epoch: [170][ 880/ 1200] Overall Loss 0.142851 Objective Loss 0.142851 LR 0.000250 Time 0.019986 -2022-12-06 11:38:51,006 - Epoch: [170][ 890/ 1200] Overall Loss 0.143001 Objective Loss 0.143001 LR 0.000250 Time 0.019975 -2022-12-06 11:38:51,197 - Epoch: [170][ 900/ 1200] Overall Loss 0.143057 Objective Loss 0.143057 LR 0.000250 Time 0.019965 -2022-12-06 11:38:51,388 - Epoch: [170][ 910/ 1200] Overall Loss 0.143170 Objective Loss 0.143170 LR 0.000250 Time 0.019955 -2022-12-06 11:38:51,579 - Epoch: [170][ 920/ 1200] Overall Loss 0.142995 Objective Loss 0.142995 LR 0.000250 Time 0.019945 -2022-12-06 11:38:51,770 - Epoch: [170][ 930/ 1200] Overall Loss 0.143106 Objective Loss 0.143106 LR 0.000250 Time 0.019936 -2022-12-06 11:38:51,961 - Epoch: [170][ 940/ 1200] Overall Loss 0.143038 Objective Loss 0.143038 LR 0.000250 Time 0.019926 -2022-12-06 11:38:52,152 - Epoch: [170][ 950/ 1200] Overall Loss 0.143009 Objective Loss 0.143009 LR 0.000250 Time 0.019917 -2022-12-06 11:38:52,343 - Epoch: [170][ 960/ 1200] Overall Loss 0.143091 Objective Loss 0.143091 LR 0.000250 Time 0.019908 -2022-12-06 11:38:52,535 - Epoch: [170][ 970/ 1200] Overall Loss 0.143129 Objective Loss 0.143129 LR 0.000250 Time 0.019900 -2022-12-06 11:38:52,726 - Epoch: [170][ 980/ 1200] Overall Loss 0.143174 Objective Loss 0.143174 LR 0.000250 Time 0.019891 -2022-12-06 11:38:52,916 - Epoch: [170][ 990/ 1200] Overall Loss 0.143256 Objective Loss 0.143256 LR 0.000250 Time 0.019882 -2022-12-06 11:38:53,108 - Epoch: [170][ 1000/ 1200] Overall Loss 0.143421 Objective Loss 0.143421 LR 0.000250 Time 0.019874 -2022-12-06 11:38:53,299 - Epoch: [170][ 1010/ 1200] Overall Loss 0.143406 Objective Loss 0.143406 LR 0.000250 Time 0.019866 -2022-12-06 11:38:53,490 - Epoch: [170][ 1020/ 1200] Overall Loss 0.143330 Objective Loss 0.143330 LR 0.000250 Time 0.019858 -2022-12-06 11:38:53,681 - Epoch: [170][ 1030/ 1200] Overall Loss 0.143298 Objective Loss 0.143298 LR 0.000250 Time 0.019850 -2022-12-06 11:38:53,872 - Epoch: [170][ 1040/ 1200] Overall Loss 0.143169 Objective Loss 0.143169 LR 0.000250 Time 0.019842 -2022-12-06 11:38:54,063 - Epoch: [170][ 1050/ 1200] Overall Loss 0.143135 Objective Loss 0.143135 LR 0.000250 Time 0.019835 -2022-12-06 11:38:54,253 - Epoch: [170][ 1060/ 1200] Overall Loss 0.143226 Objective Loss 0.143226 LR 0.000250 Time 0.019827 -2022-12-06 11:38:54,444 - Epoch: [170][ 1070/ 1200] Overall Loss 0.143325 Objective Loss 0.143325 LR 0.000250 Time 0.019819 -2022-12-06 11:38:54,635 - Epoch: [170][ 1080/ 1200] Overall Loss 0.143370 Objective Loss 0.143370 LR 0.000250 Time 0.019812 -2022-12-06 11:38:54,826 - Epoch: [170][ 1090/ 1200] Overall Loss 0.143391 Objective Loss 0.143391 LR 0.000250 Time 0.019805 -2022-12-06 11:38:55,017 - Epoch: [170][ 1100/ 1200] Overall Loss 0.143490 Objective Loss 0.143490 LR 0.000250 Time 0.019798 -2022-12-06 11:38:55,208 - Epoch: [170][ 1110/ 1200] Overall Loss 0.143611 Objective Loss 0.143611 LR 0.000250 Time 0.019791 -2022-12-06 11:38:55,398 - Epoch: [170][ 1120/ 1200] Overall Loss 0.143792 Objective Loss 0.143792 LR 0.000250 Time 0.019784 -2022-12-06 11:38:55,589 - Epoch: [170][ 1130/ 1200] Overall Loss 0.144148 Objective Loss 0.144148 LR 0.000250 Time 0.019777 -2022-12-06 11:38:55,780 - Epoch: [170][ 1140/ 1200] Overall Loss 0.144201 Objective Loss 0.144201 LR 0.000250 Time 0.019771 -2022-12-06 11:38:55,971 - Epoch: [170][ 1150/ 1200] Overall Loss 0.144236 Objective Loss 0.144236 LR 0.000250 Time 0.019764 -2022-12-06 11:38:56,161 - Epoch: [170][ 1160/ 1200] Overall Loss 0.144473 Objective Loss 0.144473 LR 0.000250 Time 0.019757 -2022-12-06 11:38:56,352 - Epoch: [170][ 1170/ 1200] Overall Loss 0.144409 Objective Loss 0.144409 LR 0.000250 Time 0.019751 -2022-12-06 11:38:56,544 - Epoch: [170][ 1180/ 1200] Overall Loss 0.144437 Objective Loss 0.144437 LR 0.000250 Time 0.019746 -2022-12-06 11:38:56,735 - Epoch: [170][ 1190/ 1200] Overall Loss 0.144433 Objective Loss 0.144433 LR 0.000250 Time 0.019740 -2022-12-06 11:38:56,965 - Epoch: [170][ 1200/ 1200] Overall Loss 0.144503 Objective Loss 0.144503 Top1 89.330544 Top5 99.581590 LR 0.000250 Time 0.019767 -2022-12-06 11:38:57,060 - --- validate (epoch=170)----------- -2022-12-06 11:38:57,060 - 34129 samples (256 per mini-batch) -2022-12-06 11:38:57,507 - Epoch: [170][ 10/ 134] Loss 0.245263 Top1 87.929688 Top5 98.359375 -2022-12-06 11:38:57,640 - Epoch: [170][ 20/ 134] Loss 0.250349 Top1 87.343750 Top5 98.300781 -2022-12-06 11:38:57,767 - Epoch: [170][ 30/ 134] Loss 0.250924 Top1 87.565104 Top5 98.528646 -2022-12-06 11:38:57,898 - Epoch: [170][ 40/ 134] Loss 0.246380 Top1 87.666016 Top5 98.564453 -2022-12-06 11:38:58,028 - Epoch: [170][ 50/ 134] Loss 0.241354 Top1 87.843750 Top5 98.507812 -2022-12-06 11:38:58,157 - Epoch: [170][ 60/ 134] Loss 0.237670 Top1 88.111979 Top5 98.580729 -2022-12-06 11:38:58,286 - Epoch: [170][ 70/ 134] Loss 0.237772 Top1 88.007812 Top5 98.577009 -2022-12-06 11:38:58,426 - Epoch: [170][ 80/ 134] Loss 0.238697 Top1 88.007812 Top5 98.544922 -2022-12-06 11:38:58,559 - Epoch: [170][ 90/ 134] Loss 0.238668 Top1 88.012153 Top5 98.580729 -2022-12-06 11:38:58,684 - Epoch: [170][ 100/ 134] Loss 0.237809 Top1 88.093750 Top5 98.570312 -2022-12-06 11:38:58,813 - Epoch: [170][ 110/ 134] Loss 0.238366 Top1 88.121449 Top5 98.572443 -2022-12-06 11:38:58,959 - Epoch: [170][ 120/ 134] Loss 0.237698 Top1 88.115234 Top5 98.551432 -2022-12-06 11:38:59,095 - Epoch: [170][ 130/ 134] Loss 0.237775 Top1 88.082933 Top5 98.572716 -2022-12-06 11:38:59,131 - Epoch: [170][ 134/ 134] Loss 0.237522 Top1 88.065868 Top5 98.581851 -2022-12-06 11:38:59,219 - ==> Top1: 88.066 Top5: 98.582 Loss: 0.238 - -2022-12-06 11:38:59,220 - ==> Confusion: -[[ 920 0 1 3 4 7 1 0 3 36 0 2 1 3 5 1 1 1 4 0 3] - [ 2 941 1 2 8 22 1 13 0 0 2 4 0 1 0 2 4 2 9 3 10] - [ 5 3 1014 10 3 2 16 6 0 1 4 5 1 2 1 3 2 3 3 7 12] - [ 1 2 17 945 1 3 0 1 0 1 11 0 5 2 13 0 0 1 12 0 5] - [ 8 4 1 0 959 1 1 2 0 4 1 4 0 2 9 5 10 3 2 0 4] - [ 1 16 0 2 4 979 2 18 1 2 0 12 4 13 1 1 1 1 0 4 7] - [ 0 4 8 1 0 2 1075 3 1 0 0 1 1 2 0 2 3 4 2 7 2] - [ 0 5 4 3 2 29 5 959 0 0 0 6 3 1 1 0 1 0 16 12 7] - [ 6 2 0 0 0 2 0 0 982 31 8 2 1 12 13 1 2 0 1 1 0] - [ 61 1 1 0 6 4 0 3 23 870 1 2 0 14 4 2 0 2 0 0 7] - [ 3 1 3 4 1 1 2 2 8 1 963 1 1 14 2 1 0 0 4 1 6] - [ 1 1 2 0 1 8 2 1 0 2 0 994 16 2 0 5 4 7 0 4 1] - [ 0 1 1 0 0 3 0 0 0 1 0 28 913 0 2 6 1 5 0 2 6] - [ 0 1 0 0 2 5 0 3 6 7 2 5 3 972 1 3 5 0 0 1 7] - [ 8 3 2 4 4 2 0 0 11 0 0 4 1 3 1072 0 0 1 9 0 6] - [ 0 0 0 2 2 0 3 1 0 0 2 7 6 2 0 995 8 10 0 1 4] - [ 2 0 1 1 2 1 0 1 0 1 0 3 2 2 0 7 1035 2 0 5 7] - [ 3 0 0 3 0 0 1 0 1 4 0 7 12 3 1 15 0 983 0 0 3] - [ 4 4 3 6 1 3 0 18 1 1 4 4 3 1 6 0 0 2 938 2 7] - [ 2 5 4 1 1 6 2 4 0 1 3 18 9 3 0 3 5 3 0 1000 10] - [ 109 162 151 81 101 160 68 123 78 59 125 108 310 271 127 93 179 68 125 187 10541]] - -2022-12-06 11:38:59,786 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:38:59,786 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:38:59,792 - - -2022-12-06 11:38:59,792 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:39:00,718 - Epoch: [171][ 10/ 1200] Overall Loss 0.139751 Objective Loss 0.139751 LR 0.000250 Time 0.092522 -2022-12-06 11:39:00,911 - Epoch: [171][ 20/ 1200] Overall Loss 0.149780 Objective Loss 0.149780 LR 0.000250 Time 0.055845 -2022-12-06 11:39:01,101 - Epoch: [171][ 30/ 1200] Overall Loss 0.145250 Objective Loss 0.145250 LR 0.000250 Time 0.043550 -2022-12-06 11:39:01,292 - Epoch: [171][ 40/ 1200] Overall Loss 0.147205 Objective Loss 0.147205 LR 0.000250 Time 0.037418 -2022-12-06 11:39:01,481 - Epoch: [171][ 50/ 1200] Overall Loss 0.147515 Objective Loss 0.147515 LR 0.000250 Time 0.033713 -2022-12-06 11:39:01,672 - Epoch: [171][ 60/ 1200] Overall Loss 0.147904 Objective Loss 0.147904 LR 0.000250 Time 0.031261 -2022-12-06 11:39:01,862 - Epoch: [171][ 70/ 1200] Overall Loss 0.147768 Objective Loss 0.147768 LR 0.000250 Time 0.029514 -2022-12-06 11:39:02,052 - Epoch: [171][ 80/ 1200] Overall Loss 0.147223 Objective Loss 0.147223 LR 0.000250 Time 0.028193 -2022-12-06 11:39:02,242 - Epoch: [171][ 90/ 1200] Overall Loss 0.147781 Objective Loss 0.147781 LR 0.000250 Time 0.027164 -2022-12-06 11:39:02,432 - Epoch: [171][ 100/ 1200] Overall Loss 0.146554 Objective Loss 0.146554 LR 0.000250 Time 0.026343 -2022-12-06 11:39:02,623 - Epoch: [171][ 110/ 1200] Overall Loss 0.147261 Objective Loss 0.147261 LR 0.000250 Time 0.025673 -2022-12-06 11:39:02,813 - Epoch: [171][ 120/ 1200] Overall Loss 0.145501 Objective Loss 0.145501 LR 0.000250 Time 0.025114 -2022-12-06 11:39:03,003 - Epoch: [171][ 130/ 1200] Overall Loss 0.146398 Objective Loss 0.146398 LR 0.000250 Time 0.024639 -2022-12-06 11:39:03,192 - Epoch: [171][ 140/ 1200] Overall Loss 0.145880 Objective Loss 0.145880 LR 0.000250 Time 0.024230 -2022-12-06 11:39:03,383 - Epoch: [171][ 150/ 1200] Overall Loss 0.146480 Objective Loss 0.146480 LR 0.000250 Time 0.023879 -2022-12-06 11:39:03,572 - Epoch: [171][ 160/ 1200] Overall Loss 0.145241 Objective Loss 0.145241 LR 0.000250 Time 0.023569 -2022-12-06 11:39:03,762 - Epoch: [171][ 170/ 1200] Overall Loss 0.144598 Objective Loss 0.144598 LR 0.000250 Time 0.023295 -2022-12-06 11:39:03,953 - Epoch: [171][ 180/ 1200] Overall Loss 0.144275 Objective Loss 0.144275 LR 0.000250 Time 0.023056 -2022-12-06 11:39:04,143 - Epoch: [171][ 190/ 1200] Overall Loss 0.143968 Objective Loss 0.143968 LR 0.000250 Time 0.022840 -2022-12-06 11:39:04,333 - Epoch: [171][ 200/ 1200] Overall Loss 0.144261 Objective Loss 0.144261 LR 0.000250 Time 0.022648 -2022-12-06 11:39:04,523 - Epoch: [171][ 210/ 1200] Overall Loss 0.143478 Objective Loss 0.143478 LR 0.000250 Time 0.022470 -2022-12-06 11:39:04,713 - Epoch: [171][ 220/ 1200] Overall Loss 0.143909 Objective Loss 0.143909 LR 0.000250 Time 0.022311 -2022-12-06 11:39:04,903 - Epoch: [171][ 230/ 1200] Overall Loss 0.144267 Objective Loss 0.144267 LR 0.000250 Time 0.022166 -2022-12-06 11:39:05,093 - Epoch: [171][ 240/ 1200] Overall Loss 0.144503 Objective Loss 0.144503 LR 0.000250 Time 0.022030 -2022-12-06 11:39:05,282 - Epoch: [171][ 250/ 1200] Overall Loss 0.144517 Objective Loss 0.144517 LR 0.000250 Time 0.021903 -2022-12-06 11:39:05,472 - Epoch: [171][ 260/ 1200] Overall Loss 0.144314 Objective Loss 0.144314 LR 0.000250 Time 0.021790 -2022-12-06 11:39:05,663 - Epoch: [171][ 270/ 1200] Overall Loss 0.143498 Objective Loss 0.143498 LR 0.000250 Time 0.021686 -2022-12-06 11:39:05,852 - Epoch: [171][ 280/ 1200] Overall Loss 0.143968 Objective Loss 0.143968 LR 0.000250 Time 0.021586 -2022-12-06 11:39:06,042 - Epoch: [171][ 290/ 1200] Overall Loss 0.143268 Objective Loss 0.143268 LR 0.000250 Time 0.021494 -2022-12-06 11:39:06,232 - Epoch: [171][ 300/ 1200] Overall Loss 0.142703 Objective Loss 0.142703 LR 0.000250 Time 0.021409 -2022-12-06 11:39:06,422 - Epoch: [171][ 310/ 1200] Overall Loss 0.142622 Objective Loss 0.142622 LR 0.000250 Time 0.021329 -2022-12-06 11:39:06,612 - Epoch: [171][ 320/ 1200] Overall Loss 0.142424 Objective Loss 0.142424 LR 0.000250 Time 0.021256 -2022-12-06 11:39:06,802 - Epoch: [171][ 330/ 1200] Overall Loss 0.142042 Objective Loss 0.142042 LR 0.000250 Time 0.021186 -2022-12-06 11:39:06,992 - Epoch: [171][ 340/ 1200] Overall Loss 0.141519 Objective Loss 0.141519 LR 0.000250 Time 0.021120 -2022-12-06 11:39:07,183 - Epoch: [171][ 350/ 1200] Overall Loss 0.141566 Objective Loss 0.141566 LR 0.000250 Time 0.021060 -2022-12-06 11:39:07,373 - Epoch: [171][ 360/ 1200] Overall Loss 0.141632 Objective Loss 0.141632 LR 0.000250 Time 0.021003 -2022-12-06 11:39:07,563 - Epoch: [171][ 370/ 1200] Overall Loss 0.142103 Objective Loss 0.142103 LR 0.000250 Time 0.020946 -2022-12-06 11:39:07,753 - Epoch: [171][ 380/ 1200] Overall Loss 0.142193 Objective Loss 0.142193 LR 0.000250 Time 0.020894 -2022-12-06 11:39:07,942 - Epoch: [171][ 390/ 1200] Overall Loss 0.142201 Objective Loss 0.142201 LR 0.000250 Time 0.020842 -2022-12-06 11:39:08,132 - Epoch: [171][ 400/ 1200] Overall Loss 0.142385 Objective Loss 0.142385 LR 0.000250 Time 0.020795 -2022-12-06 11:39:08,322 - Epoch: [171][ 410/ 1200] Overall Loss 0.142212 Objective Loss 0.142212 LR 0.000250 Time 0.020749 -2022-12-06 11:39:08,512 - Epoch: [171][ 420/ 1200] Overall Loss 0.142351 Objective Loss 0.142351 LR 0.000250 Time 0.020706 -2022-12-06 11:39:08,701 - Epoch: [171][ 430/ 1200] Overall Loss 0.142034 Objective Loss 0.142034 LR 0.000250 Time 0.020663 -2022-12-06 11:39:08,892 - Epoch: [171][ 440/ 1200] Overall Loss 0.142254 Objective Loss 0.142254 LR 0.000250 Time 0.020625 -2022-12-06 11:39:09,081 - Epoch: [171][ 450/ 1200] Overall Loss 0.142145 Objective Loss 0.142145 LR 0.000250 Time 0.020586 -2022-12-06 11:39:09,272 - Epoch: [171][ 460/ 1200] Overall Loss 0.141990 Objective Loss 0.141990 LR 0.000250 Time 0.020552 -2022-12-06 11:39:09,462 - Epoch: [171][ 470/ 1200] Overall Loss 0.141530 Objective Loss 0.141530 LR 0.000250 Time 0.020519 -2022-12-06 11:39:09,653 - Epoch: [171][ 480/ 1200] Overall Loss 0.141673 Objective Loss 0.141673 LR 0.000250 Time 0.020487 -2022-12-06 11:39:09,843 - Epoch: [171][ 490/ 1200] Overall Loss 0.141885 Objective Loss 0.141885 LR 0.000250 Time 0.020455 -2022-12-06 11:39:10,033 - Epoch: [171][ 500/ 1200] Overall Loss 0.142050 Objective Loss 0.142050 LR 0.000250 Time 0.020425 -2022-12-06 11:39:10,222 - Epoch: [171][ 510/ 1200] Overall Loss 0.142186 Objective Loss 0.142186 LR 0.000250 Time 0.020396 -2022-12-06 11:39:10,412 - Epoch: [171][ 520/ 1200] Overall Loss 0.142377 Objective Loss 0.142377 LR 0.000250 Time 0.020368 -2022-12-06 11:39:10,602 - Epoch: [171][ 530/ 1200] Overall Loss 0.142183 Objective Loss 0.142183 LR 0.000250 Time 0.020341 -2022-12-06 11:39:10,792 - Epoch: [171][ 540/ 1200] Overall Loss 0.142554 Objective Loss 0.142554 LR 0.000250 Time 0.020315 -2022-12-06 11:39:10,982 - Epoch: [171][ 550/ 1200] Overall Loss 0.142431 Objective Loss 0.142431 LR 0.000250 Time 0.020289 -2022-12-06 11:39:11,172 - Epoch: [171][ 560/ 1200] Overall Loss 0.142182 Objective Loss 0.142182 LR 0.000250 Time 0.020265 -2022-12-06 11:39:11,362 - Epoch: [171][ 570/ 1200] Overall Loss 0.142171 Objective Loss 0.142171 LR 0.000250 Time 0.020242 -2022-12-06 11:39:11,552 - Epoch: [171][ 580/ 1200] Overall Loss 0.142314 Objective Loss 0.142314 LR 0.000250 Time 0.020220 -2022-12-06 11:39:11,742 - Epoch: [171][ 590/ 1200] Overall Loss 0.142292 Objective Loss 0.142292 LR 0.000250 Time 0.020199 -2022-12-06 11:39:11,932 - Epoch: [171][ 600/ 1200] Overall Loss 0.142138 Objective Loss 0.142138 LR 0.000250 Time 0.020178 -2022-12-06 11:39:12,123 - Epoch: [171][ 610/ 1200] Overall Loss 0.142293 Objective Loss 0.142293 LR 0.000250 Time 0.020159 -2022-12-06 11:39:12,314 - Epoch: [171][ 620/ 1200] Overall Loss 0.142231 Objective Loss 0.142231 LR 0.000250 Time 0.020141 -2022-12-06 11:39:12,503 - Epoch: [171][ 630/ 1200] Overall Loss 0.142028 Objective Loss 0.142028 LR 0.000250 Time 0.020121 -2022-12-06 11:39:12,694 - Epoch: [171][ 640/ 1200] Overall Loss 0.142454 Objective Loss 0.142454 LR 0.000250 Time 0.020103 -2022-12-06 11:39:12,882 - Epoch: [171][ 650/ 1200] Overall Loss 0.142543 Objective Loss 0.142543 LR 0.000250 Time 0.020083 -2022-12-06 11:39:13,072 - Epoch: [171][ 660/ 1200] Overall Loss 0.142474 Objective Loss 0.142474 LR 0.000250 Time 0.020066 -2022-12-06 11:39:13,262 - Epoch: [171][ 670/ 1200] Overall Loss 0.142451 Objective Loss 0.142451 LR 0.000250 Time 0.020049 -2022-12-06 11:39:13,452 - Epoch: [171][ 680/ 1200] Overall Loss 0.142338 Objective Loss 0.142338 LR 0.000250 Time 0.020033 -2022-12-06 11:39:13,642 - Epoch: [171][ 690/ 1200] Overall Loss 0.142182 Objective Loss 0.142182 LR 0.000250 Time 0.020017 -2022-12-06 11:39:13,833 - Epoch: [171][ 700/ 1200] Overall Loss 0.141923 Objective Loss 0.141923 LR 0.000250 Time 0.020003 -2022-12-06 11:39:14,024 - Epoch: [171][ 710/ 1200] Overall Loss 0.142016 Objective Loss 0.142016 LR 0.000250 Time 0.019989 -2022-12-06 11:39:14,214 - Epoch: [171][ 720/ 1200] Overall Loss 0.141881 Objective Loss 0.141881 LR 0.000250 Time 0.019975 -2022-12-06 11:39:14,402 - Epoch: [171][ 730/ 1200] Overall Loss 0.141784 Objective Loss 0.141784 LR 0.000250 Time 0.019959 -2022-12-06 11:39:14,592 - Epoch: [171][ 740/ 1200] Overall Loss 0.141672 Objective Loss 0.141672 LR 0.000250 Time 0.019945 -2022-12-06 11:39:14,782 - Epoch: [171][ 750/ 1200] Overall Loss 0.141678 Objective Loss 0.141678 LR 0.000250 Time 0.019931 -2022-12-06 11:39:14,973 - Epoch: [171][ 760/ 1200] Overall Loss 0.142003 Objective Loss 0.142003 LR 0.000250 Time 0.019919 -2022-12-06 11:39:15,162 - Epoch: [171][ 770/ 1200] Overall Loss 0.142402 Objective Loss 0.142402 LR 0.000250 Time 0.019906 -2022-12-06 11:39:15,353 - Epoch: [171][ 780/ 1200] Overall Loss 0.142478 Objective Loss 0.142478 LR 0.000250 Time 0.019894 -2022-12-06 11:39:15,543 - Epoch: [171][ 790/ 1200] Overall Loss 0.142638 Objective Loss 0.142638 LR 0.000250 Time 0.019883 -2022-12-06 11:39:15,733 - Epoch: [171][ 800/ 1200] Overall Loss 0.142541 Objective Loss 0.142541 LR 0.000250 Time 0.019871 -2022-12-06 11:39:15,923 - Epoch: [171][ 810/ 1200] Overall Loss 0.142588 Objective Loss 0.142588 LR 0.000250 Time 0.019859 -2022-12-06 11:39:16,112 - Epoch: [171][ 820/ 1200] Overall Loss 0.142642 Objective Loss 0.142642 LR 0.000250 Time 0.019847 -2022-12-06 11:39:16,303 - Epoch: [171][ 830/ 1200] Overall Loss 0.142851 Objective Loss 0.142851 LR 0.000250 Time 0.019837 -2022-12-06 11:39:16,495 - Epoch: [171][ 840/ 1200] Overall Loss 0.143196 Objective Loss 0.143196 LR 0.000250 Time 0.019830 -2022-12-06 11:39:16,687 - Epoch: [171][ 850/ 1200] Overall Loss 0.143290 Objective Loss 0.143290 LR 0.000250 Time 0.019821 -2022-12-06 11:39:16,880 - Epoch: [171][ 860/ 1200] Overall Loss 0.143206 Objective Loss 0.143206 LR 0.000250 Time 0.019814 -2022-12-06 11:39:17,072 - Epoch: [171][ 870/ 1200] Overall Loss 0.143249 Objective Loss 0.143249 LR 0.000250 Time 0.019807 -2022-12-06 11:39:17,264 - Epoch: [171][ 880/ 1200] Overall Loss 0.143031 Objective Loss 0.143031 LR 0.000250 Time 0.019799 -2022-12-06 11:39:17,456 - Epoch: [171][ 890/ 1200] Overall Loss 0.143121 Objective Loss 0.143121 LR 0.000250 Time 0.019792 -2022-12-06 11:39:17,648 - Epoch: [171][ 900/ 1200] Overall Loss 0.142830 Objective Loss 0.142830 LR 0.000250 Time 0.019785 -2022-12-06 11:39:17,840 - Epoch: [171][ 910/ 1200] Overall Loss 0.142633 Objective Loss 0.142633 LR 0.000250 Time 0.019777 -2022-12-06 11:39:18,032 - Epoch: [171][ 920/ 1200] Overall Loss 0.142658 Objective Loss 0.142658 LR 0.000250 Time 0.019771 -2022-12-06 11:39:18,224 - Epoch: [171][ 930/ 1200] Overall Loss 0.142745 Objective Loss 0.142745 LR 0.000250 Time 0.019764 -2022-12-06 11:39:18,416 - Epoch: [171][ 940/ 1200] Overall Loss 0.142747 Objective Loss 0.142747 LR 0.000250 Time 0.019757 -2022-12-06 11:39:18,608 - Epoch: [171][ 950/ 1200] Overall Loss 0.142642 Objective Loss 0.142642 LR 0.000250 Time 0.019751 -2022-12-06 11:39:18,800 - Epoch: [171][ 960/ 1200] Overall Loss 0.142470 Objective Loss 0.142470 LR 0.000250 Time 0.019744 -2022-12-06 11:39:18,991 - Epoch: [171][ 970/ 1200] Overall Loss 0.142703 Objective Loss 0.142703 LR 0.000250 Time 0.019738 -2022-12-06 11:39:19,184 - Epoch: [171][ 980/ 1200] Overall Loss 0.142750 Objective Loss 0.142750 LR 0.000250 Time 0.019732 -2022-12-06 11:39:19,376 - Epoch: [171][ 990/ 1200] Overall Loss 0.142618 Objective Loss 0.142618 LR 0.000250 Time 0.019726 -2022-12-06 11:39:19,566 - Epoch: [171][ 1000/ 1200] Overall Loss 0.142697 Objective Loss 0.142697 LR 0.000250 Time 0.019718 -2022-12-06 11:39:19,755 - Epoch: [171][ 1010/ 1200] Overall Loss 0.142630 Objective Loss 0.142630 LR 0.000250 Time 0.019710 -2022-12-06 11:39:19,946 - Epoch: [171][ 1020/ 1200] Overall Loss 0.142728 Objective Loss 0.142728 LR 0.000250 Time 0.019703 -2022-12-06 11:39:20,135 - Epoch: [171][ 1030/ 1200] Overall Loss 0.142556 Objective Loss 0.142556 LR 0.000250 Time 0.019695 -2022-12-06 11:39:20,326 - Epoch: [171][ 1040/ 1200] Overall Loss 0.142511 Objective Loss 0.142511 LR 0.000250 Time 0.019689 -2022-12-06 11:39:20,516 - Epoch: [171][ 1050/ 1200] Overall Loss 0.142649 Objective Loss 0.142649 LR 0.000250 Time 0.019682 -2022-12-06 11:39:20,706 - Epoch: [171][ 1060/ 1200] Overall Loss 0.142770 Objective Loss 0.142770 LR 0.000250 Time 0.019675 -2022-12-06 11:39:20,896 - Epoch: [171][ 1070/ 1200] Overall Loss 0.142738 Objective Loss 0.142738 LR 0.000250 Time 0.019668 -2022-12-06 11:39:21,087 - Epoch: [171][ 1080/ 1200] Overall Loss 0.142755 Objective Loss 0.142755 LR 0.000250 Time 0.019662 -2022-12-06 11:39:21,277 - Epoch: [171][ 1090/ 1200] Overall Loss 0.142736 Objective Loss 0.142736 LR 0.000250 Time 0.019655 -2022-12-06 11:39:21,467 - Epoch: [171][ 1100/ 1200] Overall Loss 0.142619 Objective Loss 0.142619 LR 0.000250 Time 0.019649 -2022-12-06 11:39:21,658 - Epoch: [171][ 1110/ 1200] Overall Loss 0.142537 Objective Loss 0.142537 LR 0.000250 Time 0.019644 -2022-12-06 11:39:21,848 - Epoch: [171][ 1120/ 1200] Overall Loss 0.142375 Objective Loss 0.142375 LR 0.000250 Time 0.019638 -2022-12-06 11:39:22,039 - Epoch: [171][ 1130/ 1200] Overall Loss 0.142450 Objective Loss 0.142450 LR 0.000250 Time 0.019632 -2022-12-06 11:39:22,229 - Epoch: [171][ 1140/ 1200] Overall Loss 0.142463 Objective Loss 0.142463 LR 0.000250 Time 0.019626 -2022-12-06 11:39:22,420 - Epoch: [171][ 1150/ 1200] Overall Loss 0.142432 Objective Loss 0.142432 LR 0.000250 Time 0.019621 -2022-12-06 11:39:22,610 - Epoch: [171][ 1160/ 1200] Overall Loss 0.142542 Objective Loss 0.142542 LR 0.000250 Time 0.019616 -2022-12-06 11:39:22,800 - Epoch: [171][ 1170/ 1200] Overall Loss 0.142605 Objective Loss 0.142605 LR 0.000250 Time 0.019610 -2022-12-06 11:39:22,990 - Epoch: [171][ 1180/ 1200] Overall Loss 0.142560 Objective Loss 0.142560 LR 0.000250 Time 0.019604 -2022-12-06 11:39:23,180 - Epoch: [171][ 1190/ 1200] Overall Loss 0.142842 Objective Loss 0.142842 LR 0.000250 Time 0.019599 -2022-12-06 11:39:23,411 - Epoch: [171][ 1200/ 1200] Overall Loss 0.143010 Objective Loss 0.143010 Top1 92.677824 Top5 99.581590 LR 0.000250 Time 0.019627 -2022-12-06 11:39:23,499 - --- validate (epoch=171)----------- -2022-12-06 11:39:23,500 - 34129 samples (256 per mini-batch) -2022-12-06 11:39:23,944 - Epoch: [171][ 10/ 134] Loss 0.225040 Top1 88.906250 Top5 98.710938 -2022-12-06 11:39:24,079 - Epoch: [171][ 20/ 134] Loss 0.218660 Top1 88.496094 Top5 98.632812 -2022-12-06 11:39:24,213 - Epoch: [171][ 30/ 134] Loss 0.228989 Top1 87.877604 Top5 98.567708 -2022-12-06 11:39:24,345 - Epoch: [171][ 40/ 134] Loss 0.225695 Top1 88.007812 Top5 98.593750 -2022-12-06 11:39:24,476 - Epoch: [171][ 50/ 134] Loss 0.227164 Top1 88.132812 Top5 98.609375 -2022-12-06 11:39:24,605 - Epoch: [171][ 60/ 134] Loss 0.226549 Top1 88.183594 Top5 98.652344 -2022-12-06 11:39:24,734 - Epoch: [171][ 70/ 134] Loss 0.225797 Top1 88.264509 Top5 98.638393 -2022-12-06 11:39:24,862 - Epoch: [171][ 80/ 134] Loss 0.225476 Top1 88.237305 Top5 98.662109 -2022-12-06 11:39:24,992 - Epoch: [171][ 90/ 134] Loss 0.226436 Top1 88.181424 Top5 98.658854 -2022-12-06 11:39:25,120 - Epoch: [171][ 100/ 134] Loss 0.229132 Top1 88.136719 Top5 98.621094 -2022-12-06 11:39:25,249 - Epoch: [171][ 110/ 134] Loss 0.231373 Top1 88.114347 Top5 98.618608 -2022-12-06 11:39:25,379 - Epoch: [171][ 120/ 134] Loss 0.232200 Top1 88.128255 Top5 98.610026 -2022-12-06 11:39:25,508 - Epoch: [171][ 130/ 134] Loss 0.231532 Top1 88.155048 Top5 98.617788 -2022-12-06 11:39:25,545 - Epoch: [171][ 134/ 134] Loss 0.231414 Top1 88.115679 Top5 98.605292 -2022-12-06 11:39:25,633 - ==> Top1: 88.116 Top5: 98.605 Loss: 0.231 - -2022-12-06 11:39:25,634 - ==> Confusion: -[[ 923 0 2 1 5 5 2 0 5 38 0 1 1 2 6 1 0 0 0 0 4] - [ 2 933 2 2 8 22 4 12 0 1 0 4 1 1 0 2 5 3 14 3 8] - [ 6 0 1017 16 3 1 16 4 0 2 2 5 2 4 2 3 2 1 3 4 10] - [ 3 3 12 959 0 1 1 0 0 0 7 1 4 0 12 0 2 1 5 1 8] - [ 12 2 2 0 959 1 0 1 0 6 1 4 0 2 9 5 6 3 0 0 7] - [ 3 7 0 3 3 990 1 17 3 1 0 14 2 12 1 1 0 1 0 5 5] - [ 2 1 8 1 0 2 1080 1 0 0 0 1 1 1 0 4 2 3 1 8 2] - [ 1 7 11 3 2 23 7 950 0 0 2 7 1 1 0 1 0 1 19 14 4] - [ 8 1 0 1 0 1 1 0 980 39 9 1 0 8 9 2 2 0 1 1 0] - [ 58 0 1 0 4 2 0 2 19 894 1 1 0 10 3 1 1 0 0 0 4] - [ 2 1 0 8 1 0 3 2 8 0 962 0 2 12 4 1 1 0 5 1 6] - [ 4 0 1 0 1 10 5 2 2 0 0 978 20 2 0 7 4 5 0 6 4] - [ 0 1 1 4 1 4 0 0 0 1 0 23 908 1 2 8 1 8 0 2 4] - [ 0 1 0 0 1 12 0 3 10 10 5 1 3 960 2 1 2 1 0 2 9] - [ 6 2 0 7 4 4 0 1 9 1 0 2 1 2 1076 0 1 2 6 0 6] - [ 1 0 0 2 2 0 3 1 0 0 2 5 4 1 0 1001 4 11 0 2 4] - [ 1 1 2 1 2 1 0 0 1 0 0 2 3 2 0 8 1038 0 0 3 7] - [ 3 1 1 4 2 0 0 0 1 2 0 7 11 3 2 8 0 988 0 0 3] - [ 2 2 2 10 1 3 0 20 3 1 5 3 1 0 10 2 0 2 937 0 4] - [ 4 2 1 2 1 4 6 2 0 1 2 10 6 6 0 3 4 2 0 1018 6] - [ 112 154 159 111 89 148 83 106 74 74 150 105 296 223 139 107 175 94 126 185 10516]] - -2022-12-06 11:39:26,304 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:39:26,304 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:39:26,310 - - -2022-12-06 11:39:26,310 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:39:27,261 - Epoch: [172][ 10/ 1200] Overall Loss 0.117954 Objective Loss 0.117954 LR 0.000250 Time 0.094971 -2022-12-06 11:39:27,464 - Epoch: [172][ 20/ 1200] Overall Loss 0.126219 Objective Loss 0.126219 LR 0.000250 Time 0.057618 -2022-12-06 11:39:27,656 - Epoch: [172][ 30/ 1200] Overall Loss 0.132191 Objective Loss 0.132191 LR 0.000250 Time 0.044802 -2022-12-06 11:39:27,848 - Epoch: [172][ 40/ 1200] Overall Loss 0.137591 Objective Loss 0.137591 LR 0.000250 Time 0.038394 -2022-12-06 11:39:28,040 - Epoch: [172][ 50/ 1200] Overall Loss 0.135728 Objective Loss 0.135728 LR 0.000250 Time 0.034540 -2022-12-06 11:39:28,232 - Epoch: [172][ 60/ 1200] Overall Loss 0.135594 Objective Loss 0.135594 LR 0.000250 Time 0.031963 -2022-12-06 11:39:28,423 - Epoch: [172][ 70/ 1200] Overall Loss 0.135637 Objective Loss 0.135637 LR 0.000250 Time 0.030122 -2022-12-06 11:39:28,615 - Epoch: [172][ 80/ 1200] Overall Loss 0.136671 Objective Loss 0.136671 LR 0.000250 Time 0.028753 -2022-12-06 11:39:28,806 - Epoch: [172][ 90/ 1200] Overall Loss 0.137730 Objective Loss 0.137730 LR 0.000250 Time 0.027679 -2022-12-06 11:39:28,999 - Epoch: [172][ 100/ 1200] Overall Loss 0.137663 Objective Loss 0.137663 LR 0.000250 Time 0.026826 -2022-12-06 11:39:29,190 - Epoch: [172][ 110/ 1200] Overall Loss 0.138026 Objective Loss 0.138026 LR 0.000250 Time 0.026126 -2022-12-06 11:39:29,382 - Epoch: [172][ 120/ 1200] Overall Loss 0.140972 Objective Loss 0.140972 LR 0.000250 Time 0.025543 -2022-12-06 11:39:29,574 - Epoch: [172][ 130/ 1200] Overall Loss 0.142448 Objective Loss 0.142448 LR 0.000250 Time 0.025046 -2022-12-06 11:39:29,765 - Epoch: [172][ 140/ 1200] Overall Loss 0.143146 Objective Loss 0.143146 LR 0.000250 Time 0.024622 -2022-12-06 11:39:29,957 - Epoch: [172][ 150/ 1200] Overall Loss 0.144151 Objective Loss 0.144151 LR 0.000250 Time 0.024253 -2022-12-06 11:39:30,149 - Epoch: [172][ 160/ 1200] Overall Loss 0.144340 Objective Loss 0.144340 LR 0.000250 Time 0.023933 -2022-12-06 11:39:30,340 - Epoch: [172][ 170/ 1200] Overall Loss 0.144406 Objective Loss 0.144406 LR 0.000250 Time 0.023648 -2022-12-06 11:39:30,532 - Epoch: [172][ 180/ 1200] Overall Loss 0.145368 Objective Loss 0.145368 LR 0.000250 Time 0.023397 -2022-12-06 11:39:30,723 - Epoch: [172][ 190/ 1200] Overall Loss 0.145055 Objective Loss 0.145055 LR 0.000250 Time 0.023169 -2022-12-06 11:39:30,915 - Epoch: [172][ 200/ 1200] Overall Loss 0.144677 Objective Loss 0.144677 LR 0.000250 Time 0.022966 -2022-12-06 11:39:31,106 - Epoch: [172][ 210/ 1200] Overall Loss 0.145408 Objective Loss 0.145408 LR 0.000250 Time 0.022779 -2022-12-06 11:39:31,297 - Epoch: [172][ 220/ 1200] Overall Loss 0.145248 Objective Loss 0.145248 LR 0.000250 Time 0.022612 -2022-12-06 11:39:31,488 - Epoch: [172][ 230/ 1200] Overall Loss 0.145162 Objective Loss 0.145162 LR 0.000250 Time 0.022458 -2022-12-06 11:39:31,680 - Epoch: [172][ 240/ 1200] Overall Loss 0.144968 Objective Loss 0.144968 LR 0.000250 Time 0.022319 -2022-12-06 11:39:31,872 - Epoch: [172][ 250/ 1200] Overall Loss 0.143920 Objective Loss 0.143920 LR 0.000250 Time 0.022189 -2022-12-06 11:39:32,063 - Epoch: [172][ 260/ 1200] Overall Loss 0.144123 Objective Loss 0.144123 LR 0.000250 Time 0.022070 -2022-12-06 11:39:32,255 - Epoch: [172][ 270/ 1200] Overall Loss 0.143901 Objective Loss 0.143901 LR 0.000250 Time 0.021961 -2022-12-06 11:39:32,446 - Epoch: [172][ 280/ 1200] Overall Loss 0.144187 Objective Loss 0.144187 LR 0.000250 Time 0.021859 -2022-12-06 11:39:32,637 - Epoch: [172][ 290/ 1200] Overall Loss 0.143957 Objective Loss 0.143957 LR 0.000250 Time 0.021762 -2022-12-06 11:39:32,829 - Epoch: [172][ 300/ 1200] Overall Loss 0.143890 Objective Loss 0.143890 LR 0.000250 Time 0.021673 -2022-12-06 11:39:33,020 - Epoch: [172][ 310/ 1200] Overall Loss 0.143332 Objective Loss 0.143332 LR 0.000250 Time 0.021589 -2022-12-06 11:39:33,212 - Epoch: [172][ 320/ 1200] Overall Loss 0.143150 Objective Loss 0.143150 LR 0.000250 Time 0.021512 -2022-12-06 11:39:33,403 - Epoch: [172][ 330/ 1200] Overall Loss 0.143076 Objective Loss 0.143076 LR 0.000250 Time 0.021438 -2022-12-06 11:39:33,595 - Epoch: [172][ 340/ 1200] Overall Loss 0.142780 Objective Loss 0.142780 LR 0.000250 Time 0.021369 -2022-12-06 11:39:33,786 - Epoch: [172][ 350/ 1200] Overall Loss 0.143173 Objective Loss 0.143173 LR 0.000250 Time 0.021303 -2022-12-06 11:39:33,978 - Epoch: [172][ 360/ 1200] Overall Loss 0.143151 Objective Loss 0.143151 LR 0.000250 Time 0.021243 -2022-12-06 11:39:34,169 - Epoch: [172][ 370/ 1200] Overall Loss 0.142582 Objective Loss 0.142582 LR 0.000250 Time 0.021185 -2022-12-06 11:39:34,361 - Epoch: [172][ 380/ 1200] Overall Loss 0.142553 Objective Loss 0.142553 LR 0.000250 Time 0.021130 -2022-12-06 11:39:34,552 - Epoch: [172][ 390/ 1200] Overall Loss 0.142218 Objective Loss 0.142218 LR 0.000250 Time 0.021077 -2022-12-06 11:39:34,744 - Epoch: [172][ 400/ 1200] Overall Loss 0.142261 Objective Loss 0.142261 LR 0.000250 Time 0.021029 -2022-12-06 11:39:34,935 - Epoch: [172][ 410/ 1200] Overall Loss 0.142320 Objective Loss 0.142320 LR 0.000250 Time 0.020981 -2022-12-06 11:39:35,127 - Epoch: [172][ 420/ 1200] Overall Loss 0.142170 Objective Loss 0.142170 LR 0.000250 Time 0.020937 -2022-12-06 11:39:35,319 - Epoch: [172][ 430/ 1200] Overall Loss 0.142602 Objective Loss 0.142602 LR 0.000250 Time 0.020894 -2022-12-06 11:39:35,510 - Epoch: [172][ 440/ 1200] Overall Loss 0.142386 Objective Loss 0.142386 LR 0.000250 Time 0.020854 -2022-12-06 11:39:35,702 - Epoch: [172][ 450/ 1200] Overall Loss 0.142754 Objective Loss 0.142754 LR 0.000250 Time 0.020815 -2022-12-06 11:39:35,894 - Epoch: [172][ 460/ 1200] Overall Loss 0.143229 Objective Loss 0.143229 LR 0.000250 Time 0.020778 -2022-12-06 11:39:36,085 - Epoch: [172][ 470/ 1200] Overall Loss 0.142961 Objective Loss 0.142961 LR 0.000250 Time 0.020743 -2022-12-06 11:39:36,277 - Epoch: [172][ 480/ 1200] Overall Loss 0.143085 Objective Loss 0.143085 LR 0.000250 Time 0.020709 -2022-12-06 11:39:36,469 - Epoch: [172][ 490/ 1200] Overall Loss 0.143156 Objective Loss 0.143156 LR 0.000250 Time 0.020676 -2022-12-06 11:39:36,660 - Epoch: [172][ 500/ 1200] Overall Loss 0.142985 Objective Loss 0.142985 LR 0.000250 Time 0.020644 -2022-12-06 11:39:36,851 - Epoch: [172][ 510/ 1200] Overall Loss 0.143109 Objective Loss 0.143109 LR 0.000250 Time 0.020613 -2022-12-06 11:39:37,043 - Epoch: [172][ 520/ 1200] Overall Loss 0.143045 Objective Loss 0.143045 LR 0.000250 Time 0.020584 -2022-12-06 11:39:37,235 - Epoch: [172][ 530/ 1200] Overall Loss 0.142881 Objective Loss 0.142881 LR 0.000250 Time 0.020557 -2022-12-06 11:39:37,426 - Epoch: [172][ 540/ 1200] Overall Loss 0.142755 Objective Loss 0.142755 LR 0.000250 Time 0.020530 -2022-12-06 11:39:37,619 - Epoch: [172][ 550/ 1200] Overall Loss 0.142872 Objective Loss 0.142872 LR 0.000250 Time 0.020505 -2022-12-06 11:39:37,810 - Epoch: [172][ 560/ 1200] Overall Loss 0.142993 Objective Loss 0.142993 LR 0.000250 Time 0.020479 -2022-12-06 11:39:38,000 - Epoch: [172][ 570/ 1200] Overall Loss 0.143059 Objective Loss 0.143059 LR 0.000250 Time 0.020454 -2022-12-06 11:39:38,192 - Epoch: [172][ 580/ 1200] Overall Loss 0.143002 Objective Loss 0.143002 LR 0.000250 Time 0.020431 -2022-12-06 11:39:38,383 - Epoch: [172][ 590/ 1200] Overall Loss 0.142850 Objective Loss 0.142850 LR 0.000250 Time 0.020408 -2022-12-06 11:39:38,575 - Epoch: [172][ 600/ 1200] Overall Loss 0.142660 Objective Loss 0.142660 LR 0.000250 Time 0.020385 -2022-12-06 11:39:38,767 - Epoch: [172][ 610/ 1200] Overall Loss 0.142487 Objective Loss 0.142487 LR 0.000250 Time 0.020365 -2022-12-06 11:39:38,959 - Epoch: [172][ 620/ 1200] Overall Loss 0.142711 Objective Loss 0.142711 LR 0.000250 Time 0.020346 -2022-12-06 11:39:39,151 - Epoch: [172][ 630/ 1200] Overall Loss 0.142552 Objective Loss 0.142552 LR 0.000250 Time 0.020328 -2022-12-06 11:39:39,344 - Epoch: [172][ 640/ 1200] Overall Loss 0.142550 Objective Loss 0.142550 LR 0.000250 Time 0.020309 -2022-12-06 11:39:39,535 - Epoch: [172][ 650/ 1200] Overall Loss 0.142557 Objective Loss 0.142557 LR 0.000250 Time 0.020291 -2022-12-06 11:39:39,728 - Epoch: [172][ 660/ 1200] Overall Loss 0.142592 Objective Loss 0.142592 LR 0.000250 Time 0.020274 -2022-12-06 11:39:39,919 - Epoch: [172][ 670/ 1200] Overall Loss 0.142628 Objective Loss 0.142628 LR 0.000250 Time 0.020256 -2022-12-06 11:39:40,110 - Epoch: [172][ 680/ 1200] Overall Loss 0.142474 Objective Loss 0.142474 LR 0.000250 Time 0.020239 -2022-12-06 11:39:40,303 - Epoch: [172][ 690/ 1200] Overall Loss 0.142719 Objective Loss 0.142719 LR 0.000250 Time 0.020223 -2022-12-06 11:39:40,495 - Epoch: [172][ 700/ 1200] Overall Loss 0.142791 Objective Loss 0.142791 LR 0.000250 Time 0.020208 -2022-12-06 11:39:40,687 - Epoch: [172][ 710/ 1200] Overall Loss 0.142648 Objective Loss 0.142648 LR 0.000250 Time 0.020193 -2022-12-06 11:39:40,878 - Epoch: [172][ 720/ 1200] Overall Loss 0.142769 Objective Loss 0.142769 LR 0.000250 Time 0.020178 -2022-12-06 11:39:41,071 - Epoch: [172][ 730/ 1200] Overall Loss 0.142754 Objective Loss 0.142754 LR 0.000250 Time 0.020164 -2022-12-06 11:39:41,262 - Epoch: [172][ 740/ 1200] Overall Loss 0.142866 Objective Loss 0.142866 LR 0.000250 Time 0.020150 -2022-12-06 11:39:41,454 - Epoch: [172][ 750/ 1200] Overall Loss 0.142965 Objective Loss 0.142965 LR 0.000250 Time 0.020136 -2022-12-06 11:39:41,646 - Epoch: [172][ 760/ 1200] Overall Loss 0.143373 Objective Loss 0.143373 LR 0.000250 Time 0.020123 -2022-12-06 11:39:41,837 - Epoch: [172][ 770/ 1200] Overall Loss 0.143579 Objective Loss 0.143579 LR 0.000250 Time 0.020109 -2022-12-06 11:39:42,029 - Epoch: [172][ 780/ 1200] Overall Loss 0.143515 Objective Loss 0.143515 LR 0.000250 Time 0.020097 -2022-12-06 11:39:42,221 - Epoch: [172][ 790/ 1200] Overall Loss 0.143546 Objective Loss 0.143546 LR 0.000250 Time 0.020085 -2022-12-06 11:39:42,412 - Epoch: [172][ 800/ 1200] Overall Loss 0.143473 Objective Loss 0.143473 LR 0.000250 Time 0.020072 -2022-12-06 11:39:42,603 - Epoch: [172][ 810/ 1200] Overall Loss 0.143526 Objective Loss 0.143526 LR 0.000250 Time 0.020060 -2022-12-06 11:39:42,795 - Epoch: [172][ 820/ 1200] Overall Loss 0.143397 Objective Loss 0.143397 LR 0.000250 Time 0.020048 -2022-12-06 11:39:42,987 - Epoch: [172][ 830/ 1200] Overall Loss 0.143214 Objective Loss 0.143214 LR 0.000250 Time 0.020037 -2022-12-06 11:39:43,179 - Epoch: [172][ 840/ 1200] Overall Loss 0.143316 Objective Loss 0.143316 LR 0.000250 Time 0.020026 -2022-12-06 11:39:43,371 - Epoch: [172][ 850/ 1200] Overall Loss 0.143310 Objective Loss 0.143310 LR 0.000250 Time 0.020015 -2022-12-06 11:39:43,563 - Epoch: [172][ 860/ 1200] Overall Loss 0.143327 Objective Loss 0.143327 LR 0.000250 Time 0.020006 -2022-12-06 11:39:43,754 - Epoch: [172][ 870/ 1200] Overall Loss 0.143415 Objective Loss 0.143415 LR 0.000250 Time 0.019995 -2022-12-06 11:39:43,946 - Epoch: [172][ 880/ 1200] Overall Loss 0.143518 Objective Loss 0.143518 LR 0.000250 Time 0.019986 -2022-12-06 11:39:44,138 - Epoch: [172][ 890/ 1200] Overall Loss 0.143520 Objective Loss 0.143520 LR 0.000250 Time 0.019976 -2022-12-06 11:39:44,330 - Epoch: [172][ 900/ 1200] Overall Loss 0.143680 Objective Loss 0.143680 LR 0.000250 Time 0.019967 -2022-12-06 11:39:44,522 - Epoch: [172][ 910/ 1200] Overall Loss 0.143539 Objective Loss 0.143539 LR 0.000250 Time 0.019958 -2022-12-06 11:39:44,714 - Epoch: [172][ 920/ 1200] Overall Loss 0.143501 Objective Loss 0.143501 LR 0.000250 Time 0.019949 -2022-12-06 11:39:44,906 - Epoch: [172][ 930/ 1200] Overall Loss 0.143390 Objective Loss 0.143390 LR 0.000250 Time 0.019940 -2022-12-06 11:39:45,099 - Epoch: [172][ 940/ 1200] Overall Loss 0.143729 Objective Loss 0.143729 LR 0.000250 Time 0.019932 -2022-12-06 11:39:45,291 - Epoch: [172][ 950/ 1200] Overall Loss 0.143928 Objective Loss 0.143928 LR 0.000250 Time 0.019924 -2022-12-06 11:39:45,483 - Epoch: [172][ 960/ 1200] Overall Loss 0.143794 Objective Loss 0.143794 LR 0.000250 Time 0.019916 -2022-12-06 11:39:45,674 - Epoch: [172][ 970/ 1200] Overall Loss 0.143732 Objective Loss 0.143732 LR 0.000250 Time 0.019908 -2022-12-06 11:39:45,866 - Epoch: [172][ 980/ 1200] Overall Loss 0.143621 Objective Loss 0.143621 LR 0.000250 Time 0.019899 -2022-12-06 11:39:46,058 - Epoch: [172][ 990/ 1200] Overall Loss 0.143735 Objective Loss 0.143735 LR 0.000250 Time 0.019891 -2022-12-06 11:39:46,250 - Epoch: [172][ 1000/ 1200] Overall Loss 0.143739 Objective Loss 0.143739 LR 0.000250 Time 0.019884 -2022-12-06 11:39:46,441 - Epoch: [172][ 1010/ 1200] Overall Loss 0.143816 Objective Loss 0.143816 LR 0.000250 Time 0.019876 -2022-12-06 11:39:46,633 - Epoch: [172][ 1020/ 1200] Overall Loss 0.143720 Objective Loss 0.143720 LR 0.000250 Time 0.019869 -2022-12-06 11:39:46,824 - Epoch: [172][ 1030/ 1200] Overall Loss 0.143689 Objective Loss 0.143689 LR 0.000250 Time 0.019861 -2022-12-06 11:39:47,016 - Epoch: [172][ 1040/ 1200] Overall Loss 0.143760 Objective Loss 0.143760 LR 0.000250 Time 0.019854 -2022-12-06 11:39:47,208 - Epoch: [172][ 1050/ 1200] Overall Loss 0.143891 Objective Loss 0.143891 LR 0.000250 Time 0.019847 -2022-12-06 11:39:47,400 - Epoch: [172][ 1060/ 1200] Overall Loss 0.143763 Objective Loss 0.143763 LR 0.000250 Time 0.019840 -2022-12-06 11:39:47,592 - Epoch: [172][ 1070/ 1200] Overall Loss 0.143708 Objective Loss 0.143708 LR 0.000250 Time 0.019834 -2022-12-06 11:39:47,783 - Epoch: [172][ 1080/ 1200] Overall Loss 0.143690 Objective Loss 0.143690 LR 0.000250 Time 0.019827 -2022-12-06 11:39:47,975 - Epoch: [172][ 1090/ 1200] Overall Loss 0.143881 Objective Loss 0.143881 LR 0.000250 Time 0.019820 -2022-12-06 11:39:48,167 - Epoch: [172][ 1100/ 1200] Overall Loss 0.143870 Objective Loss 0.143870 LR 0.000250 Time 0.019814 -2022-12-06 11:39:48,359 - Epoch: [172][ 1110/ 1200] Overall Loss 0.143735 Objective Loss 0.143735 LR 0.000250 Time 0.019808 -2022-12-06 11:39:48,551 - Epoch: [172][ 1120/ 1200] Overall Loss 0.143710 Objective Loss 0.143710 LR 0.000250 Time 0.019802 -2022-12-06 11:39:48,743 - Epoch: [172][ 1130/ 1200] Overall Loss 0.143826 Objective Loss 0.143826 LR 0.000250 Time 0.019796 -2022-12-06 11:39:48,935 - Epoch: [172][ 1140/ 1200] Overall Loss 0.143868 Objective Loss 0.143868 LR 0.000250 Time 0.019791 -2022-12-06 11:39:49,127 - Epoch: [172][ 1150/ 1200] Overall Loss 0.143777 Objective Loss 0.143777 LR 0.000250 Time 0.019786 -2022-12-06 11:39:49,319 - Epoch: [172][ 1160/ 1200] Overall Loss 0.143788 Objective Loss 0.143788 LR 0.000250 Time 0.019780 -2022-12-06 11:39:49,511 - Epoch: [172][ 1170/ 1200] Overall Loss 0.143814 Objective Loss 0.143814 LR 0.000250 Time 0.019774 -2022-12-06 11:39:49,703 - Epoch: [172][ 1180/ 1200] Overall Loss 0.143625 Objective Loss 0.143625 LR 0.000250 Time 0.019769 -2022-12-06 11:39:49,895 - Epoch: [172][ 1190/ 1200] Overall Loss 0.143676 Objective Loss 0.143676 LR 0.000250 Time 0.019764 -2022-12-06 11:39:50,128 - Epoch: [172][ 1200/ 1200] Overall Loss 0.143743 Objective Loss 0.143743 Top1 91.631799 Top5 99.163180 LR 0.000250 Time 0.019793 -2022-12-06 11:39:50,217 - --- validate (epoch=172)----------- -2022-12-06 11:39:50,217 - 34129 samples (256 per mini-batch) -2022-12-06 11:39:50,665 - Epoch: [172][ 10/ 134] Loss 0.219536 Top1 87.460938 Top5 98.593750 -2022-12-06 11:39:50,799 - Epoch: [172][ 20/ 134] Loss 0.240957 Top1 87.265625 Top5 98.300781 -2022-12-06 11:39:50,932 - Epoch: [172][ 30/ 134] Loss 0.232131 Top1 87.578125 Top5 98.450521 -2022-12-06 11:39:51,065 - Epoch: [172][ 40/ 134] Loss 0.224608 Top1 87.910156 Top5 98.574219 -2022-12-06 11:39:51,198 - Epoch: [172][ 50/ 134] Loss 0.228789 Top1 87.789062 Top5 98.578125 -2022-12-06 11:39:51,326 - Epoch: [172][ 60/ 134] Loss 0.231329 Top1 87.792969 Top5 98.574219 -2022-12-06 11:39:51,458 - Epoch: [172][ 70/ 134] Loss 0.232014 Top1 87.739955 Top5 98.549107 -2022-12-06 11:39:51,588 - Epoch: [172][ 80/ 134] Loss 0.233022 Top1 87.739258 Top5 98.544922 -2022-12-06 11:39:51,716 - Epoch: [172][ 90/ 134] Loss 0.234057 Top1 87.795139 Top5 98.567708 -2022-12-06 11:39:51,848 - Epoch: [172][ 100/ 134] Loss 0.235480 Top1 87.691406 Top5 98.531250 -2022-12-06 11:39:51,977 - Epoch: [172][ 110/ 134] Loss 0.235798 Top1 87.677557 Top5 98.529830 -2022-12-06 11:39:52,107 - Epoch: [172][ 120/ 134] Loss 0.235584 Top1 87.727865 Top5 98.515625 -2022-12-06 11:39:52,241 - Epoch: [172][ 130/ 134] Loss 0.236689 Top1 87.737380 Top5 98.515625 -2022-12-06 11:39:52,278 - Epoch: [172][ 134/ 134] Loss 0.236001 Top1 87.725981 Top5 98.534970 -2022-12-06 11:39:52,366 - ==> Top1: 87.726 Top5: 98.535 Loss: 0.236 - -2022-12-06 11:39:52,367 - ==> Confusion: -[[ 928 2 1 0 3 7 1 1 6 32 0 1 1 2 4 2 1 0 1 0 3] - [ 0 940 1 2 7 18 3 13 1 1 3 5 1 1 0 2 1 0 12 4 12] - [ 5 2 1009 16 6 2 16 8 1 1 9 4 1 1 1 3 1 0 4 3 10] - [ 2 2 10 954 1 3 0 1 0 0 8 0 8 2 10 0 0 0 12 1 6] - [ 11 4 1 0 960 1 1 2 1 4 1 1 1 1 13 6 4 3 0 2 3] - [ 0 9 0 3 4 993 1 16 4 1 1 7 4 11 1 1 4 0 0 5 4] - [ 0 3 10 2 0 1 1074 3 0 0 0 3 0 1 0 7 0 1 2 10 1] - [ 1 5 6 4 2 27 8 955 0 0 0 3 2 3 0 0 1 0 17 14 6] - [ 3 3 0 0 0 2 1 3 985 34 11 2 1 8 5 1 1 0 3 1 0] - [ 68 0 1 0 4 2 0 4 19 876 2 0 0 12 4 1 1 0 1 0 6] - [ 1 2 1 3 1 0 3 3 8 0 978 0 1 8 2 1 0 0 3 0 4] - [ 3 0 2 0 1 14 2 4 1 0 0 959 29 4 1 3 4 3 0 15 6] - [ 0 1 1 1 1 2 0 1 0 0 0 16 924 1 1 7 1 4 1 4 3] - [ 1 1 0 0 1 7 0 2 12 4 6 3 3 965 2 3 1 1 1 1 9] - [ 8 1 1 6 2 4 0 0 18 2 1 3 4 2 1061 0 0 1 10 1 5] - [ 0 0 0 1 2 0 2 1 0 0 1 5 10 2 0 997 4 7 0 5 6] - [ 2 0 1 1 3 0 1 0 1 0 0 2 4 2 0 12 1030 0 0 7 6] - [ 3 1 1 3 0 1 1 0 0 4 0 3 27 1 2 18 1 966 0 1 3] - [ 2 4 3 6 1 2 0 22 3 1 6 2 2 0 9 0 0 0 940 4 1] - [ 2 3 0 2 0 4 2 3 0 0 1 10 10 6 0 4 2 2 2 1023 4] - [ 100 186 141 95 101 172 76 131 66 84 163 74 351 245 125 100 132 60 164 239 10421]] - -2022-12-06 11:39:52,935 - ==> Best [Top1: 88.368 Top5: 98.585 Sparsity:0.00 Params: 5376 on epoch: 159] -2022-12-06 11:39:52,935 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:39:52,941 - - -2022-12-06 11:39:52,941 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:39:53,988 - Epoch: [173][ 10/ 1200] Overall Loss 0.162490 Objective Loss 0.162490 LR 0.000250 Time 0.104688 -2022-12-06 11:39:54,191 - Epoch: [173][ 20/ 1200] Overall Loss 0.155486 Objective Loss 0.155486 LR 0.000250 Time 0.062464 -2022-12-06 11:39:54,387 - Epoch: [173][ 30/ 1200] Overall Loss 0.148228 Objective Loss 0.148228 LR 0.000250 Time 0.048130 -2022-12-06 11:39:54,584 - Epoch: [173][ 40/ 1200] Overall Loss 0.140074 Objective Loss 0.140074 LR 0.000250 Time 0.041025 -2022-12-06 11:39:54,779 - Epoch: [173][ 50/ 1200] Overall Loss 0.139224 Objective Loss 0.139224 LR 0.000250 Time 0.036712 -2022-12-06 11:39:54,976 - Epoch: [173][ 60/ 1200] Overall Loss 0.137899 Objective Loss 0.137899 LR 0.000250 Time 0.033869 -2022-12-06 11:39:55,171 - Epoch: [173][ 70/ 1200] Overall Loss 0.135347 Objective Loss 0.135347 LR 0.000250 Time 0.031805 -2022-12-06 11:39:55,368 - Epoch: [173][ 80/ 1200] Overall Loss 0.134939 Objective Loss 0.134939 LR 0.000250 Time 0.030284 -2022-12-06 11:39:55,563 - Epoch: [173][ 90/ 1200] Overall Loss 0.134603 Objective Loss 0.134603 LR 0.000250 Time 0.029078 -2022-12-06 11:39:55,760 - Epoch: [173][ 100/ 1200] Overall Loss 0.135828 Objective Loss 0.135828 LR 0.000250 Time 0.028136 -2022-12-06 11:39:55,955 - Epoch: [173][ 110/ 1200] Overall Loss 0.137098 Objective Loss 0.137098 LR 0.000250 Time 0.027347 -2022-12-06 11:39:56,152 - Epoch: [173][ 120/ 1200] Overall Loss 0.135627 Objective Loss 0.135627 LR 0.000250 Time 0.026704 -2022-12-06 11:39:56,348 - Epoch: [173][ 130/ 1200] Overall Loss 0.136354 Objective Loss 0.136354 LR 0.000250 Time 0.026150 -2022-12-06 11:39:56,545 - Epoch: [173][ 140/ 1200] Overall Loss 0.136154 Objective Loss 0.136154 LR 0.000250 Time 0.025689 -2022-12-06 11:39:56,739 - Epoch: [173][ 150/ 1200] Overall Loss 0.134648 Objective Loss 0.134648 LR 0.000250 Time 0.025268 -2022-12-06 11:39:56,936 - Epoch: [173][ 160/ 1200] Overall Loss 0.134138 Objective Loss 0.134138 LR 0.000250 Time 0.024912 -2022-12-06 11:39:57,130 - Epoch: [173][ 170/ 1200] Overall Loss 0.133663 Objective Loss 0.133663 LR 0.000250 Time 0.024588 -2022-12-06 11:39:57,327 - Epoch: [173][ 180/ 1200] Overall Loss 0.133227 Objective Loss 0.133227 LR 0.000250 Time 0.024312 -2022-12-06 11:39:57,522 - Epoch: [173][ 190/ 1200] Overall Loss 0.133404 Objective Loss 0.133404 LR 0.000250 Time 0.024058 -2022-12-06 11:39:57,720 - Epoch: [173][ 200/ 1200] Overall Loss 0.134925 Objective Loss 0.134925 LR 0.000250 Time 0.023839 -2022-12-06 11:39:57,915 - Epoch: [173][ 210/ 1200] Overall Loss 0.135265 Objective Loss 0.135265 LR 0.000250 Time 0.023633 -2022-12-06 11:39:58,112 - Epoch: [173][ 220/ 1200] Overall Loss 0.135731 Objective Loss 0.135731 LR 0.000250 Time 0.023452 -2022-12-06 11:39:58,308 - Epoch: [173][ 230/ 1200] Overall Loss 0.136136 Objective Loss 0.136136 LR 0.000250 Time 0.023279 -2022-12-06 11:39:58,504 - Epoch: [173][ 240/ 1200] Overall Loss 0.136623 Objective Loss 0.136623 LR 0.000250 Time 0.023126 -2022-12-06 11:39:58,699 - Epoch: [173][ 250/ 1200] Overall Loss 0.136915 Objective Loss 0.136915 LR 0.000250 Time 0.022976 -2022-12-06 11:39:58,896 - Epoch: [173][ 260/ 1200] Overall Loss 0.137374 Objective Loss 0.137374 LR 0.000250 Time 0.022848 -2022-12-06 11:39:59,092 - Epoch: [173][ 270/ 1200] Overall Loss 0.136911 Objective Loss 0.136911 LR 0.000250 Time 0.022725 -2022-12-06 11:39:59,288 - Epoch: [173][ 280/ 1200] Overall Loss 0.136759 Objective Loss 0.136759 LR 0.000250 Time 0.022615 -2022-12-06 11:39:59,483 - Epoch: [173][ 290/ 1200] Overall Loss 0.137147 Objective Loss 0.137147 LR 0.000250 Time 0.022506 -2022-12-06 11:39:59,680 - Epoch: [173][ 300/ 1200] Overall Loss 0.137916 Objective Loss 0.137916 LR 0.000250 Time 0.022410 -2022-12-06 11:39:59,876 - Epoch: [173][ 310/ 1200] Overall Loss 0.138101 Objective Loss 0.138101 LR 0.000250 Time 0.022315 -2022-12-06 11:40:00,073 - Epoch: [173][ 320/ 1200] Overall Loss 0.138108 Objective Loss 0.138108 LR 0.000250 Time 0.022234 -2022-12-06 11:40:00,268 - Epoch: [173][ 330/ 1200] Overall Loss 0.138009 Objective Loss 0.138009 LR 0.000250 Time 0.022148 -2022-12-06 11:40:00,465 - Epoch: [173][ 340/ 1200] Overall Loss 0.138323 Objective Loss 0.138323 LR 0.000250 Time 0.022076 -2022-12-06 11:40:00,660 - Epoch: [173][ 350/ 1200] Overall Loss 0.138746 Objective Loss 0.138746 LR 0.000250 Time 0.022000 -2022-12-06 11:40:00,858 - Epoch: [173][ 360/ 1200] Overall Loss 0.138918 Objective Loss 0.138918 LR 0.000250 Time 0.021937 -2022-12-06 11:40:01,053 - Epoch: [173][ 370/ 1200] Overall Loss 0.138833 Objective Loss 0.138833 LR 0.000250 Time 0.021870 -2022-12-06 11:40:01,250 - Epoch: [173][ 380/ 1200] Overall Loss 0.138672 Objective Loss 0.138672 LR 0.000250 Time 0.021812 -2022-12-06 11:40:01,444 - Epoch: [173][ 390/ 1200] Overall Loss 0.138565 Objective Loss 0.138565 LR 0.000250 Time 0.021749 -2022-12-06 11:40:01,641 - Epoch: [173][ 400/ 1200] Overall Loss 0.138929 Objective Loss 0.138929 LR 0.000250 Time 0.021696 -2022-12-06 11:40:01,836 - Epoch: [173][ 410/ 1200] Overall Loss 0.139334 Objective Loss 0.139334 LR 0.000250 Time 0.021640 -2022-12-06 11:40:02,033 - Epoch: [173][ 420/ 1200] Overall Loss 0.139533 Objective Loss 0.139533 LR 0.000250 Time 0.021593 -2022-12-06 11:40:02,227 - Epoch: [173][ 430/ 1200] Overall Loss 0.139717 Objective Loss 0.139717 LR 0.000250 Time 0.021542 -2022-12-06 11:40:02,425 - Epoch: [173][ 440/ 1200] Overall Loss 0.140043 Objective Loss 0.140043 LR 0.000250 Time 0.021500 -2022-12-06 11:40:02,620 - Epoch: [173][ 450/ 1200] Overall Loss 0.140032 Objective Loss 0.140032 LR 0.000250 Time 0.021454 -2022-12-06 11:40:02,817 - Epoch: [173][ 460/ 1200] Overall Loss 0.140115 Objective Loss 0.140115 LR 0.000250 Time 0.021415 -2022-12-06 11:40:03,012 - Epoch: [173][ 470/ 1200] Overall Loss 0.140094 Objective Loss 0.140094 LR 0.000250 Time 0.021373 -2022-12-06 11:40:03,209 - Epoch: [173][ 480/ 1200] Overall Loss 0.139946 Objective Loss 0.139946 LR 0.000250 Time 0.021337 -2022-12-06 11:40:03,405 - Epoch: [173][ 490/ 1200] Overall Loss 0.140013 Objective Loss 0.140013 LR 0.000250 Time 0.021300 -2022-12-06 11:40:03,602 - Epoch: [173][ 500/ 1200] Overall Loss 0.139975 Objective Loss 0.139975 LR 0.000250 Time 0.021268 -2022-12-06 11:40:03,798 - Epoch: [173][ 510/ 1200] Overall Loss 0.140319 Objective Loss 0.140319 LR 0.000250 Time 0.021233 -2022-12-06 11:40:03,995 - Epoch: [173][ 520/ 1200] Overall Loss 0.140336 Objective Loss 0.140336 LR 0.000250 Time 0.021203 -2022-12-06 11:40:04,190 - Epoch: [173][ 530/ 1200] Overall Loss 0.140567 Objective Loss 0.140567 LR 0.000250 Time 0.021170 -2022-12-06 11:40:04,388 - Epoch: [173][ 540/ 1200] Overall Loss 0.140954 Objective Loss 0.140954 LR 0.000250 Time 0.021143 -2022-12-06 11:40:04,583 - Epoch: [173][ 550/ 1200] Overall Loss 0.141398 Objective Loss 0.141398 LR 0.000250 Time 0.021112 -2022-12-06 11:40:04,780 - Epoch: [173][ 560/ 1200] Overall Loss 0.141364 Objective Loss 0.141364 LR 0.000250 Time 0.021087 -2022-12-06 11:40:04,976 - Epoch: [173][ 570/ 1200] Overall Loss 0.141545 Objective Loss 0.141545 LR 0.000250 Time 0.021059 -2022-12-06 11:40:05,173 - Epoch: [173][ 580/ 1200] Overall Loss 0.141770 Objective Loss 0.141770 LR 0.000250 Time 0.021035 -2022-12-06 11:40:05,368 - Epoch: [173][ 590/ 1200] Overall Loss 0.141651 Objective Loss 0.141651 LR 0.000250 Time 0.021009 -2022-12-06 11:40:05,565 - Epoch: [173][ 600/ 1200] Overall Loss 0.141812 Objective Loss 0.141812 LR 0.000250 Time 0.020987 -2022-12-06 11:40:05,761 - Epoch: [173][ 610/ 1200] Overall Loss 0.142018 Objective Loss 0.142018 LR 0.000250 Time 0.020962 -2022-12-06 11:40:05,958 - Epoch: [173][ 620/ 1200] Overall Loss 0.141855 Objective Loss 0.141855 LR 0.000250 Time 0.020942 -2022-12-06 11:40:06,154 - Epoch: [173][ 630/ 1200] Overall Loss 0.142136 Objective Loss 0.142136 LR 0.000250 Time 0.020918 -2022-12-06 11:40:06,352 - Epoch: [173][ 640/ 1200] Overall Loss 0.141909 Objective Loss 0.141909 LR 0.000250 Time 0.020900 -2022-12-06 11:40:06,547 - Epoch: [173][ 650/ 1200] Overall Loss 0.141662 Objective Loss 0.141662 LR 0.000250 Time 0.020878 -2022-12-06 11:40:06,744 - Epoch: [173][ 660/ 1200] Overall Loss 0.141628 Objective Loss 0.141628 LR 0.000250 Time 0.020860 -2022-12-06 11:40:06,940 - Epoch: [173][ 670/ 1200] Overall Loss 0.141935 Objective Loss 0.141935 LR 0.000250 Time 0.020839 -2022-12-06 11:40:07,137 - Epoch: [173][ 680/ 1200] Overall Loss 0.141879 Objective Loss 0.141879 LR 0.000250 Time 0.020823 -2022-12-06 11:40:07,332 - Epoch: [173][ 690/ 1200] Overall Loss 0.141979 Objective Loss 0.141979 LR 0.000250 Time 0.020802 -2022-12-06 11:40:07,529 - Epoch: [173][ 700/ 1200] Overall Loss 0.141930 Objective Loss 0.141930 LR 0.000250 Time 0.020786 -2022-12-06 11:40:07,724 - Epoch: [173][ 710/ 1200] Overall Loss 0.142199 Objective Loss 0.142199 LR 0.000250 Time 0.020767 -2022-12-06 11:40:07,922 - Epoch: [173][ 720/ 1200] Overall Loss 0.142459 Objective Loss 0.142459 LR 0.000250 Time 0.020752 -2022-12-06 11:40:08,117 - Epoch: [173][ 730/ 1200] Overall Loss 0.142646 Objective Loss 0.142646 LR 0.000250 Time 0.020735 -2022-12-06 11:40:08,314 - Epoch: [173][ 740/ 1200] Overall Loss 0.142813 Objective Loss 0.142813 LR 0.000250 Time 0.020720 -2022-12-06 11:40:08,509 - Epoch: [173][ 750/ 1200] Overall Loss 0.142827 Objective Loss 0.142827 LR 0.000250 Time 0.020704 -2022-12-06 11:40:08,706 - Epoch: [173][ 760/ 1200] Overall Loss 0.142884 Objective Loss 0.142884 LR 0.000250 Time 0.020689 -2022-12-06 11:40:08,901 - Epoch: [173][ 770/ 1200] Overall Loss 0.143064 Objective Loss 0.143064 LR 0.000250 Time 0.020673 -2022-12-06 11:40:09,099 - Epoch: [173][ 780/ 1200] Overall Loss 0.142884 Objective Loss 0.142884 LR 0.000250 Time 0.020661 -2022-12-06 11:40:09,294 - Epoch: [173][ 790/ 1200] Overall Loss 0.142993 Objective Loss 0.142993 LR 0.000250 Time 0.020646 -2022-12-06 11:40:09,492 - Epoch: [173][ 800/ 1200] Overall Loss 0.142905 Objective Loss 0.142905 LR 0.000250 Time 0.020635 -2022-12-06 11:40:09,688 - Epoch: [173][ 810/ 1200] Overall Loss 0.142880 Objective Loss 0.142880 LR 0.000250 Time 0.020621 -2022-12-06 11:40:09,886 - Epoch: [173][ 820/ 1200] Overall Loss 0.142626 Objective Loss 0.142626 LR 0.000250 Time 0.020610 -2022-12-06 11:40:10,081 - Epoch: [173][ 830/ 1200] Overall Loss 0.142681 Objective Loss 0.142681 LR 0.000250 Time 0.020596 -2022-12-06 11:40:10,278 - Epoch: [173][ 840/ 1200] Overall Loss 0.142759 Objective Loss 0.142759 LR 0.000250 Time 0.020585 -2022-12-06 11:40:10,473 - Epoch: [173][ 850/ 1200] Overall Loss 0.142875 Objective Loss 0.142875 LR 0.000250 Time 0.020572 -2022-12-06 11:40:10,670 - Epoch: [173][ 860/ 1200] Overall Loss 0.142730 Objective Loss 0.142730 LR 0.000250 Time 0.020561 -2022-12-06 11:40:10,866 - Epoch: [173][ 870/ 1200] Overall Loss 0.142709 Objective Loss 0.142709 LR 0.000250 Time 0.020549 -2022-12-06 11:40:11,064 - Epoch: [173][ 880/ 1200] Overall Loss 0.142875 Objective Loss 0.142875 LR 0.000250 Time 0.020540 -2022-12-06 11:40:11,259 - Epoch: [173][ 890/ 1200] Overall Loss 0.143206 Objective Loss 0.143206 LR 0.000250 Time 0.020528 -2022-12-06 11:40:11,457 - Epoch: [173][ 900/ 1200] Overall Loss 0.143131 Objective Loss 0.143131 LR 0.000250 Time 0.020519 -2022-12-06 11:40:11,652 - Epoch: [173][ 910/ 1200] Overall Loss 0.143231 Objective Loss 0.143231 LR 0.000250 Time 0.020507 -2022-12-06 11:40:11,850 - Epoch: [173][ 920/ 1200] Overall Loss 0.143136 Objective Loss 0.143136 LR 0.000250 Time 0.020500 -2022-12-06 11:40:12,046 - Epoch: [173][ 930/ 1200] Overall Loss 0.143103 Objective Loss 0.143103 LR 0.000250 Time 0.020488 -2022-12-06 11:40:12,244 - Epoch: [173][ 940/ 1200] Overall Loss 0.143099 Objective Loss 0.143099 LR 0.000250 Time 0.020480 -2022-12-06 11:40:12,438 - Epoch: [173][ 950/ 1200] Overall Loss 0.143121 Objective Loss 0.143121 LR 0.000250 Time 0.020469 -2022-12-06 11:40:12,636 - Epoch: [173][ 960/ 1200] Overall Loss 0.143088 Objective Loss 0.143088 LR 0.000250 Time 0.020461 -2022-12-06 11:40:12,830 - Epoch: [173][ 970/ 1200] Overall Loss 0.143173 Objective Loss 0.143173 LR 0.000250 Time 0.020450 -2022-12-06 11:40:13,029 - Epoch: [173][ 980/ 1200] Overall Loss 0.143135 Objective Loss 0.143135 LR 0.000250 Time 0.020443 -2022-12-06 11:40:13,223 - Epoch: [173][ 990/ 1200] Overall Loss 0.143037 Objective Loss 0.143037 LR 0.000250 Time 0.020432 -2022-12-06 11:40:13,422 - Epoch: [173][ 1000/ 1200] Overall Loss 0.143041 Objective Loss 0.143041 LR 0.000250 Time 0.020426 -2022-12-06 11:40:13,616 - Epoch: [173][ 1010/ 1200] Overall Loss 0.143127 Objective Loss 0.143127 LR 0.000250 Time 0.020416 -2022-12-06 11:40:13,815 - Epoch: [173][ 1020/ 1200] Overall Loss 0.143127 Objective Loss 0.143127 LR 0.000250 Time 0.020410 -2022-12-06 11:40:14,010 - Epoch: [173][ 1030/ 1200] Overall Loss 0.143242 Objective Loss 0.143242 LR 0.000250 Time 0.020401 -2022-12-06 11:40:14,208 - Epoch: [173][ 1040/ 1200] Overall Loss 0.143172 Objective Loss 0.143172 LR 0.000250 Time 0.020394 -2022-12-06 11:40:14,403 - Epoch: [173][ 1050/ 1200] Overall Loss 0.143447 Objective Loss 0.143447 LR 0.000250 Time 0.020386 -2022-12-06 11:40:14,602 - Epoch: [173][ 1060/ 1200] Overall Loss 0.143370 Objective Loss 0.143370 LR 0.000250 Time 0.020381 -2022-12-06 11:40:14,797 - Epoch: [173][ 1070/ 1200] Overall Loss 0.143196 Objective Loss 0.143196 LR 0.000250 Time 0.020372 -2022-12-06 11:40:14,995 - Epoch: [173][ 1080/ 1200] Overall Loss 0.143411 Objective Loss 0.143411 LR 0.000250 Time 0.020366 -2022-12-06 11:40:15,190 - Epoch: [173][ 1090/ 1200] Overall Loss 0.143388 Objective Loss 0.143388 LR 0.000250 Time 0.020358 -2022-12-06 11:40:15,389 - Epoch: [173][ 1100/ 1200] Overall Loss 0.143408 Objective Loss 0.143408 LR 0.000250 Time 0.020353 -2022-12-06 11:40:15,583 - Epoch: [173][ 1110/ 1200] Overall Loss 0.143393 Objective Loss 0.143393 LR 0.000250 Time 0.020344 -2022-12-06 11:40:15,782 - Epoch: [173][ 1120/ 1200] Overall Loss 0.143458 Objective Loss 0.143458 LR 0.000250 Time 0.020340 -2022-12-06 11:40:15,976 - Epoch: [173][ 1130/ 1200] Overall Loss 0.143436 Objective Loss 0.143436 LR 0.000250 Time 0.020331 -2022-12-06 11:40:16,175 - Epoch: [173][ 1140/ 1200] Overall Loss 0.143615 Objective Loss 0.143615 LR 0.000250 Time 0.020326 -2022-12-06 11:40:16,370 - Epoch: [173][ 1150/ 1200] Overall Loss 0.143527 Objective Loss 0.143527 LR 0.000250 Time 0.020319 -2022-12-06 11:40:16,569 - Epoch: [173][ 1160/ 1200] Overall Loss 0.143511 Objective Loss 0.143511 LR 0.000250 Time 0.020315 -2022-12-06 11:40:16,764 - Epoch: [173][ 1170/ 1200] Overall Loss 0.143366 Objective Loss 0.143366 LR 0.000250 Time 0.020307 -2022-12-06 11:40:16,962 - Epoch: [173][ 1180/ 1200] Overall Loss 0.143445 Objective Loss 0.143445 LR 0.000250 Time 0.020303 -2022-12-06 11:40:17,158 - Epoch: [173][ 1190/ 1200] Overall Loss 0.143465 Objective Loss 0.143465 LR 0.000250 Time 0.020296 -2022-12-06 11:40:17,381 - Epoch: [173][ 1200/ 1200] Overall Loss 0.143396 Objective Loss 0.143396 Top1 90.794979 Top5 98.744770 LR 0.000250 Time 0.020313 -2022-12-06 11:40:17,470 - --- validate (epoch=173)----------- -2022-12-06 11:40:17,470 - 34129 samples (256 per mini-batch) -2022-12-06 11:40:17,924 - Epoch: [173][ 10/ 134] Loss 0.207994 Top1 88.085938 Top5 98.710938 -2022-12-06 11:40:18,070 - Epoch: [173][ 20/ 134] Loss 0.225335 Top1 88.085938 Top5 98.613281 -2022-12-06 11:40:18,209 - Epoch: [173][ 30/ 134] Loss 0.226176 Top1 88.216146 Top5 98.645833 -2022-12-06 11:40:18,354 - Epoch: [173][ 40/ 134] Loss 0.224752 Top1 88.388672 Top5 98.632812 -2022-12-06 11:40:18,492 - Epoch: [173][ 50/ 134] Loss 0.229666 Top1 88.289062 Top5 98.578125 -2022-12-06 11:40:18,637 - Epoch: [173][ 60/ 134] Loss 0.225731 Top1 88.548177 Top5 98.580729 -2022-12-06 11:40:18,771 - Epoch: [173][ 70/ 134] Loss 0.228539 Top1 88.582589 Top5 98.543527 -2022-12-06 11:40:18,904 - Epoch: [173][ 80/ 134] Loss 0.228420 Top1 88.452148 Top5 98.569336 -2022-12-06 11:40:19,036 - Epoch: [173][ 90/ 134] Loss 0.228944 Top1 88.480903 Top5 98.585069 -2022-12-06 11:40:19,171 - Epoch: [173][ 100/ 134] Loss 0.230348 Top1 88.488281 Top5 98.566406 -2022-12-06 11:40:19,302 - Epoch: [173][ 110/ 134] Loss 0.232115 Top1 88.423295 Top5 98.558239 -2022-12-06 11:40:19,442 - Epoch: [173][ 120/ 134] Loss 0.231879 Top1 88.476562 Top5 98.570964 -2022-12-06 11:40:19,573 - Epoch: [173][ 130/ 134] Loss 0.231135 Top1 88.446514 Top5 98.593750 -2022-12-06 11:40:19,610 - Epoch: [173][ 134/ 134] Loss 0.230907 Top1 88.426265 Top5 98.593571 -2022-12-06 11:40:19,698 - ==> Top1: 88.426 Top5: 98.594 Loss: 0.231 - -2022-12-06 11:40:19,699 - ==> Confusion: -[[ 922 0 1 1 4 8 0 0 7 38 0 1 1 2 2 1 2 1 1 0 4] - [ 2 949 1 1 11 14 3 15 1 1 3 3 0 0 0 2 3 0 6 3 9] - [ 5 4 1015 8 4 3 12 9 0 5 5 3 1 3 2 6 2 0 1 5 10] - [ 1 1 14 959 1 3 1 1 1 2 8 1 6 2 6 0 2 1 5 1 4] - [ 6 4 1 0 967 1 0 2 2 5 1 1 0 3 10 5 4 2 0 3 3] - [ 2 11 1 4 4 984 2 16 4 1 0 12 3 13 1 2 1 0 0 3 5] - [ 1 2 5 2 1 2 1077 3 0 1 0 1 0 3 0 5 0 2 1 8 4] - [ 0 8 6 4 2 29 4 962 0 1 1 6 0 2 1 0 0 1 6 14 7] - [ 7 2 0 0 1 2 0 1 976 38 9 1 1 7 10 1 2 1 1 2 2] - [ 50 0 0 0 6 3 0 2 20 894 1 3 0 13 2 1 1 0 0 0 5] - [ 2 1 2 7 2 0 2 2 9 1 960 0 1 11 2 1 0 0 6 2 8] - [ 1 0 2 0 0 7 4 1 0 0 0 977 27 6 0 5 2 5 0 9 5] - [ 1 1 2 0 1 2 0 0 0 0 0 18 921 0 1 6 1 7 0 2 6] - [ 2 1 1 0 0 6 0 0 7 10 2 5 3 976 1 1 0 1 0 1 6] - [ 5 4 1 10 3 1 0 1 11 3 0 2 3 3 1067 0 1 1 6 1 7] - [ 1 0 1 0 2 1 4 0 0 0 2 5 8 3 0 995 4 9 0 5 3] - [ 1 3 1 1 1 2 1 0 2 0 1 1 3 2 0 10 1029 1 0 5 8] - [ 4 0 1 1 1 0 1 1 0 2 0 4 14 2 2 15 1 984 0 1 2] - [ 3 4 4 9 2 4 1 23 3 1 3 3 1 0 8 0 0 0 934 1 4] - [ 1 2 1 2 0 5 3 3 1 1 1 8 8 7 0 5 2 2 1 1021 6] - [ 109 177 143 86 108 158 72 129 59 82 118 80 307 237 113 98 145 82 107 211 10605]] - -2022-12-06 11:40:20,264 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:40:20,264 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:40:20,271 - - -2022-12-06 11:40:20,271 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:40:21,202 - Epoch: [174][ 10/ 1200] Overall Loss 0.157594 Objective Loss 0.157594 LR 0.000250 Time 0.092976 -2022-12-06 11:40:21,404 - Epoch: [174][ 20/ 1200] Overall Loss 0.142230 Objective Loss 0.142230 LR 0.000250 Time 0.056593 -2022-12-06 11:40:21,596 - Epoch: [174][ 30/ 1200] Overall Loss 0.136768 Objective Loss 0.136768 LR 0.000250 Time 0.044099 -2022-12-06 11:40:21,788 - Epoch: [174][ 40/ 1200] Overall Loss 0.135496 Objective Loss 0.135496 LR 0.000250 Time 0.037857 -2022-12-06 11:40:21,979 - Epoch: [174][ 50/ 1200] Overall Loss 0.132797 Objective Loss 0.132797 LR 0.000250 Time 0.034094 -2022-12-06 11:40:22,169 - Epoch: [174][ 60/ 1200] Overall Loss 0.133704 Objective Loss 0.133704 LR 0.000250 Time 0.031579 -2022-12-06 11:40:22,360 - Epoch: [174][ 70/ 1200] Overall Loss 0.135134 Objective Loss 0.135134 LR 0.000250 Time 0.029788 -2022-12-06 11:40:22,551 - Epoch: [174][ 80/ 1200] Overall Loss 0.135409 Objective Loss 0.135409 LR 0.000250 Time 0.028436 -2022-12-06 11:40:22,741 - Epoch: [174][ 90/ 1200] Overall Loss 0.135413 Objective Loss 0.135413 LR 0.000250 Time 0.027391 -2022-12-06 11:40:22,932 - Epoch: [174][ 100/ 1200] Overall Loss 0.136651 Objective Loss 0.136651 LR 0.000250 Time 0.026556 -2022-12-06 11:40:23,123 - Epoch: [174][ 110/ 1200] Overall Loss 0.137729 Objective Loss 0.137729 LR 0.000250 Time 0.025871 -2022-12-06 11:40:23,314 - Epoch: [174][ 120/ 1200] Overall Loss 0.137419 Objective Loss 0.137419 LR 0.000250 Time 0.025298 -2022-12-06 11:40:23,504 - Epoch: [174][ 130/ 1200] Overall Loss 0.136918 Objective Loss 0.136918 LR 0.000250 Time 0.024814 -2022-12-06 11:40:23,695 - Epoch: [174][ 140/ 1200] Overall Loss 0.138425 Objective Loss 0.138425 LR 0.000250 Time 0.024399 -2022-12-06 11:40:23,886 - Epoch: [174][ 150/ 1200] Overall Loss 0.140253 Objective Loss 0.140253 LR 0.000250 Time 0.024042 -2022-12-06 11:40:24,076 - Epoch: [174][ 160/ 1200] Overall Loss 0.140828 Objective Loss 0.140828 LR 0.000250 Time 0.023723 -2022-12-06 11:40:24,267 - Epoch: [174][ 170/ 1200] Overall Loss 0.141288 Objective Loss 0.141288 LR 0.000250 Time 0.023447 -2022-12-06 11:40:24,457 - Epoch: [174][ 180/ 1200] Overall Loss 0.140790 Objective Loss 0.140790 LR 0.000250 Time 0.023198 -2022-12-06 11:40:24,647 - Epoch: [174][ 190/ 1200] Overall Loss 0.140948 Objective Loss 0.140948 LR 0.000250 Time 0.022975 -2022-12-06 11:40:24,837 - Epoch: [174][ 200/ 1200] Overall Loss 0.140233 Objective Loss 0.140233 LR 0.000250 Time 0.022773 -2022-12-06 11:40:25,027 - Epoch: [174][ 210/ 1200] Overall Loss 0.141432 Objective Loss 0.141432 LR 0.000250 Time 0.022591 -2022-12-06 11:40:25,217 - Epoch: [174][ 220/ 1200] Overall Loss 0.140972 Objective Loss 0.140972 LR 0.000250 Time 0.022427 -2022-12-06 11:40:25,408 - Epoch: [174][ 230/ 1200] Overall Loss 0.141014 Objective Loss 0.141014 LR 0.000250 Time 0.022278 -2022-12-06 11:40:25,599 - Epoch: [174][ 240/ 1200] Overall Loss 0.140387 Objective Loss 0.140387 LR 0.000250 Time 0.022142 -2022-12-06 11:40:25,789 - Epoch: [174][ 250/ 1200] Overall Loss 0.139477 Objective Loss 0.139477 LR 0.000250 Time 0.022018 -2022-12-06 11:40:25,980 - Epoch: [174][ 260/ 1200] Overall Loss 0.139535 Objective Loss 0.139535 LR 0.000250 Time 0.021902 -2022-12-06 11:40:26,171 - Epoch: [174][ 270/ 1200] Overall Loss 0.138923 Objective Loss 0.138923 LR 0.000250 Time 0.021796 -2022-12-06 11:40:26,361 - Epoch: [174][ 280/ 1200] Overall Loss 0.138608 Objective Loss 0.138608 LR 0.000250 Time 0.021695 -2022-12-06 11:40:26,552 - Epoch: [174][ 290/ 1200] Overall Loss 0.139142 Objective Loss 0.139142 LR 0.000250 Time 0.021603 -2022-12-06 11:40:26,743 - Epoch: [174][ 300/ 1200] Overall Loss 0.139160 Objective Loss 0.139160 LR 0.000250 Time 0.021516 -2022-12-06 11:40:26,934 - Epoch: [174][ 310/ 1200] Overall Loss 0.139906 Objective Loss 0.139906 LR 0.000250 Time 0.021437 -2022-12-06 11:40:27,124 - Epoch: [174][ 320/ 1200] Overall Loss 0.140127 Objective Loss 0.140127 LR 0.000250 Time 0.021361 -2022-12-06 11:40:27,315 - Epoch: [174][ 330/ 1200] Overall Loss 0.140590 Objective Loss 0.140590 LR 0.000250 Time 0.021290 -2022-12-06 11:40:27,505 - Epoch: [174][ 340/ 1200] Overall Loss 0.140480 Objective Loss 0.140480 LR 0.000250 Time 0.021222 -2022-12-06 11:40:27,696 - Epoch: [174][ 350/ 1200] Overall Loss 0.140685 Objective Loss 0.140685 LR 0.000250 Time 0.021159 -2022-12-06 11:40:27,887 - Epoch: [174][ 360/ 1200] Overall Loss 0.141107 Objective Loss 0.141107 LR 0.000250 Time 0.021099 -2022-12-06 11:40:28,078 - Epoch: [174][ 370/ 1200] Overall Loss 0.141460 Objective Loss 0.141460 LR 0.000250 Time 0.021044 -2022-12-06 11:40:28,268 - Epoch: [174][ 380/ 1200] Overall Loss 0.141806 Objective Loss 0.141806 LR 0.000250 Time 0.020990 -2022-12-06 11:40:28,459 - Epoch: [174][ 390/ 1200] Overall Loss 0.141944 Objective Loss 0.141944 LR 0.000250 Time 0.020939 -2022-12-06 11:40:28,649 - Epoch: [174][ 400/ 1200] Overall Loss 0.141975 Objective Loss 0.141975 LR 0.000250 Time 0.020890 -2022-12-06 11:40:28,839 - Epoch: [174][ 410/ 1200] Overall Loss 0.141426 Objective Loss 0.141426 LR 0.000250 Time 0.020843 -2022-12-06 11:40:29,031 - Epoch: [174][ 420/ 1200] Overall Loss 0.141255 Objective Loss 0.141255 LR 0.000250 Time 0.020802 -2022-12-06 11:40:29,222 - Epoch: [174][ 430/ 1200] Overall Loss 0.141405 Objective Loss 0.141405 LR 0.000250 Time 0.020762 -2022-12-06 11:40:29,412 - Epoch: [174][ 440/ 1200] Overall Loss 0.141411 Objective Loss 0.141411 LR 0.000250 Time 0.020721 -2022-12-06 11:40:29,603 - Epoch: [174][ 450/ 1200] Overall Loss 0.141361 Objective Loss 0.141361 LR 0.000250 Time 0.020683 -2022-12-06 11:40:29,794 - Epoch: [174][ 460/ 1200] Overall Loss 0.140828 Objective Loss 0.140828 LR 0.000250 Time 0.020646 -2022-12-06 11:40:29,984 - Epoch: [174][ 470/ 1200] Overall Loss 0.140423 Objective Loss 0.140423 LR 0.000250 Time 0.020612 -2022-12-06 11:40:30,175 - Epoch: [174][ 480/ 1200] Overall Loss 0.140080 Objective Loss 0.140080 LR 0.000250 Time 0.020579 -2022-12-06 11:40:30,366 - Epoch: [174][ 490/ 1200] Overall Loss 0.139845 Objective Loss 0.139845 LR 0.000250 Time 0.020548 -2022-12-06 11:40:30,557 - Epoch: [174][ 500/ 1200] Overall Loss 0.139810 Objective Loss 0.139810 LR 0.000250 Time 0.020518 -2022-12-06 11:40:30,748 - Epoch: [174][ 510/ 1200] Overall Loss 0.139495 Objective Loss 0.139495 LR 0.000250 Time 0.020488 -2022-12-06 11:40:30,939 - Epoch: [174][ 520/ 1200] Overall Loss 0.139314 Objective Loss 0.139314 LR 0.000250 Time 0.020459 -2022-12-06 11:40:31,129 - Epoch: [174][ 530/ 1200] Overall Loss 0.139380 Objective Loss 0.139380 LR 0.000250 Time 0.020432 -2022-12-06 11:40:31,320 - Epoch: [174][ 540/ 1200] Overall Loss 0.139175 Objective Loss 0.139175 LR 0.000250 Time 0.020406 -2022-12-06 11:40:31,510 - Epoch: [174][ 550/ 1200] Overall Loss 0.139173 Objective Loss 0.139173 LR 0.000250 Time 0.020380 -2022-12-06 11:40:31,701 - Epoch: [174][ 560/ 1200] Overall Loss 0.139777 Objective Loss 0.139777 LR 0.000250 Time 0.020356 -2022-12-06 11:40:31,892 - Epoch: [174][ 570/ 1200] Overall Loss 0.139509 Objective Loss 0.139509 LR 0.000250 Time 0.020332 -2022-12-06 11:40:32,082 - Epoch: [174][ 580/ 1200] Overall Loss 0.139706 Objective Loss 0.139706 LR 0.000250 Time 0.020309 -2022-12-06 11:40:32,273 - Epoch: [174][ 590/ 1200] Overall Loss 0.139478 Objective Loss 0.139478 LR 0.000250 Time 0.020288 -2022-12-06 11:40:32,464 - Epoch: [174][ 600/ 1200] Overall Loss 0.139341 Objective Loss 0.139341 LR 0.000250 Time 0.020266 -2022-12-06 11:40:32,655 - Epoch: [174][ 610/ 1200] Overall Loss 0.138941 Objective Loss 0.138941 LR 0.000250 Time 0.020247 -2022-12-06 11:40:32,846 - Epoch: [174][ 620/ 1200] Overall Loss 0.138928 Objective Loss 0.138928 LR 0.000250 Time 0.020227 -2022-12-06 11:40:33,038 - Epoch: [174][ 630/ 1200] Overall Loss 0.139192 Objective Loss 0.139192 LR 0.000250 Time 0.020210 -2022-12-06 11:40:33,230 - Epoch: [174][ 640/ 1200] Overall Loss 0.139388 Objective Loss 0.139388 LR 0.000250 Time 0.020194 -2022-12-06 11:40:33,424 - Epoch: [174][ 650/ 1200] Overall Loss 0.139583 Objective Loss 0.139583 LR 0.000250 Time 0.020180 -2022-12-06 11:40:33,616 - Epoch: [174][ 660/ 1200] Overall Loss 0.139785 Objective Loss 0.139785 LR 0.000250 Time 0.020165 -2022-12-06 11:40:33,809 - Epoch: [174][ 670/ 1200] Overall Loss 0.139506 Objective Loss 0.139506 LR 0.000250 Time 0.020151 -2022-12-06 11:40:34,001 - Epoch: [174][ 680/ 1200] Overall Loss 0.139554 Objective Loss 0.139554 LR 0.000250 Time 0.020137 -2022-12-06 11:40:34,195 - Epoch: [174][ 690/ 1200] Overall Loss 0.139653 Objective Loss 0.139653 LR 0.000250 Time 0.020125 -2022-12-06 11:40:34,387 - Epoch: [174][ 700/ 1200] Overall Loss 0.140056 Objective Loss 0.140056 LR 0.000250 Time 0.020111 -2022-12-06 11:40:34,581 - Epoch: [174][ 710/ 1200] Overall Loss 0.140236 Objective Loss 0.140236 LR 0.000250 Time 0.020100 -2022-12-06 11:40:34,773 - Epoch: [174][ 720/ 1200] Overall Loss 0.140215 Objective Loss 0.140215 LR 0.000250 Time 0.020087 -2022-12-06 11:40:34,967 - Epoch: [174][ 730/ 1200] Overall Loss 0.140400 Objective Loss 0.140400 LR 0.000250 Time 0.020077 -2022-12-06 11:40:35,160 - Epoch: [174][ 740/ 1200] Overall Loss 0.140609 Objective Loss 0.140609 LR 0.000250 Time 0.020066 -2022-12-06 11:40:35,353 - Epoch: [174][ 750/ 1200] Overall Loss 0.140812 Objective Loss 0.140812 LR 0.000250 Time 0.020055 -2022-12-06 11:40:35,546 - Epoch: [174][ 760/ 1200] Overall Loss 0.140873 Objective Loss 0.140873 LR 0.000250 Time 0.020044 -2022-12-06 11:40:35,740 - Epoch: [174][ 770/ 1200] Overall Loss 0.141011 Objective Loss 0.141011 LR 0.000250 Time 0.020034 -2022-12-06 11:40:35,933 - Epoch: [174][ 780/ 1200] Overall Loss 0.141024 Objective Loss 0.141024 LR 0.000250 Time 0.020025 -2022-12-06 11:40:36,127 - Epoch: [174][ 790/ 1200] Overall Loss 0.141051 Objective Loss 0.141051 LR 0.000250 Time 0.020016 -2022-12-06 11:40:36,320 - Epoch: [174][ 800/ 1200] Overall Loss 0.141209 Objective Loss 0.141209 LR 0.000250 Time 0.020007 -2022-12-06 11:40:36,514 - Epoch: [174][ 810/ 1200] Overall Loss 0.141172 Objective Loss 0.141172 LR 0.000250 Time 0.019998 -2022-12-06 11:40:36,705 - Epoch: [174][ 820/ 1200] Overall Loss 0.141113 Objective Loss 0.141113 LR 0.000250 Time 0.019987 -2022-12-06 11:40:36,899 - Epoch: [174][ 830/ 1200] Overall Loss 0.141047 Objective Loss 0.141047 LR 0.000250 Time 0.019979 -2022-12-06 11:40:37,092 - Epoch: [174][ 840/ 1200] Overall Loss 0.140978 Objective Loss 0.140978 LR 0.000250 Time 0.019970 -2022-12-06 11:40:37,286 - Epoch: [174][ 850/ 1200] Overall Loss 0.140776 Objective Loss 0.140776 LR 0.000250 Time 0.019962 -2022-12-06 11:40:37,478 - Epoch: [174][ 860/ 1200] Overall Loss 0.140819 Objective Loss 0.140819 LR 0.000250 Time 0.019953 -2022-12-06 11:40:37,672 - Epoch: [174][ 870/ 1200] Overall Loss 0.140989 Objective Loss 0.140989 LR 0.000250 Time 0.019946 -2022-12-06 11:40:37,865 - Epoch: [174][ 880/ 1200] Overall Loss 0.141013 Objective Loss 0.141013 LR 0.000250 Time 0.019938 -2022-12-06 11:40:38,058 - Epoch: [174][ 890/ 1200] Overall Loss 0.140800 Objective Loss 0.140800 LR 0.000250 Time 0.019931 -2022-12-06 11:40:38,251 - Epoch: [174][ 900/ 1200] Overall Loss 0.140998 Objective Loss 0.140998 LR 0.000250 Time 0.019922 -2022-12-06 11:40:38,444 - Epoch: [174][ 910/ 1200] Overall Loss 0.141098 Objective Loss 0.141098 LR 0.000250 Time 0.019915 -2022-12-06 11:40:38,636 - Epoch: [174][ 920/ 1200] Overall Loss 0.141074 Objective Loss 0.141074 LR 0.000250 Time 0.019907 -2022-12-06 11:40:38,830 - Epoch: [174][ 930/ 1200] Overall Loss 0.140996 Objective Loss 0.140996 LR 0.000250 Time 0.019901 -2022-12-06 11:40:39,023 - Epoch: [174][ 940/ 1200] Overall Loss 0.141116 Objective Loss 0.141116 LR 0.000250 Time 0.019894 -2022-12-06 11:40:39,217 - Epoch: [174][ 950/ 1200] Overall Loss 0.141168 Objective Loss 0.141168 LR 0.000250 Time 0.019888 -2022-12-06 11:40:39,410 - Epoch: [174][ 960/ 1200] Overall Loss 0.141336 Objective Loss 0.141336 LR 0.000250 Time 0.019881 -2022-12-06 11:40:39,604 - Epoch: [174][ 970/ 1200] Overall Loss 0.141255 Objective Loss 0.141255 LR 0.000250 Time 0.019876 -2022-12-06 11:40:39,797 - Epoch: [174][ 980/ 1200] Overall Loss 0.141421 Objective Loss 0.141421 LR 0.000250 Time 0.019869 -2022-12-06 11:40:39,990 - Epoch: [174][ 990/ 1200] Overall Loss 0.141522 Objective Loss 0.141522 LR 0.000250 Time 0.019864 -2022-12-06 11:40:40,183 - Epoch: [174][ 1000/ 1200] Overall Loss 0.141574 Objective Loss 0.141574 LR 0.000250 Time 0.019857 -2022-12-06 11:40:40,376 - Epoch: [174][ 1010/ 1200] Overall Loss 0.141575 Objective Loss 0.141575 LR 0.000250 Time 0.019851 -2022-12-06 11:40:40,568 - Epoch: [174][ 1020/ 1200] Overall Loss 0.141669 Objective Loss 0.141669 LR 0.000250 Time 0.019844 -2022-12-06 11:40:40,761 - Epoch: [174][ 1030/ 1200] Overall Loss 0.141587 Objective Loss 0.141587 LR 0.000250 Time 0.019838 -2022-12-06 11:40:40,954 - Epoch: [174][ 1040/ 1200] Overall Loss 0.141558 Objective Loss 0.141558 LR 0.000250 Time 0.019832 -2022-12-06 11:40:41,148 - Epoch: [174][ 1050/ 1200] Overall Loss 0.141450 Objective Loss 0.141450 LR 0.000250 Time 0.019828 -2022-12-06 11:40:41,341 - Epoch: [174][ 1060/ 1200] Overall Loss 0.141260 Objective Loss 0.141260 LR 0.000250 Time 0.019822 -2022-12-06 11:40:41,534 - Epoch: [174][ 1070/ 1200] Overall Loss 0.141291 Objective Loss 0.141291 LR 0.000250 Time 0.019817 -2022-12-06 11:40:41,727 - Epoch: [174][ 1080/ 1200] Overall Loss 0.141327 Objective Loss 0.141327 LR 0.000250 Time 0.019811 -2022-12-06 11:40:41,920 - Epoch: [174][ 1090/ 1200] Overall Loss 0.141448 Objective Loss 0.141448 LR 0.000250 Time 0.019806 -2022-12-06 11:40:42,113 - Epoch: [174][ 1100/ 1200] Overall Loss 0.141406 Objective Loss 0.141406 LR 0.000250 Time 0.019801 -2022-12-06 11:40:42,306 - Epoch: [174][ 1110/ 1200] Overall Loss 0.141661 Objective Loss 0.141661 LR 0.000250 Time 0.019797 -2022-12-06 11:40:42,499 - Epoch: [174][ 1120/ 1200] Overall Loss 0.141706 Objective Loss 0.141706 LR 0.000250 Time 0.019791 -2022-12-06 11:40:42,692 - Epoch: [174][ 1130/ 1200] Overall Loss 0.141758 Objective Loss 0.141758 LR 0.000250 Time 0.019786 -2022-12-06 11:40:42,885 - Epoch: [174][ 1140/ 1200] Overall Loss 0.141801 Objective Loss 0.141801 LR 0.000250 Time 0.019782 -2022-12-06 11:40:43,078 - Epoch: [174][ 1150/ 1200] Overall Loss 0.141585 Objective Loss 0.141585 LR 0.000250 Time 0.019777 -2022-12-06 11:40:43,270 - Epoch: [174][ 1160/ 1200] Overall Loss 0.141571 Objective Loss 0.141571 LR 0.000250 Time 0.019772 -2022-12-06 11:40:43,463 - Epoch: [174][ 1170/ 1200] Overall Loss 0.141422 Objective Loss 0.141422 LR 0.000250 Time 0.019767 -2022-12-06 11:40:43,655 - Epoch: [174][ 1180/ 1200] Overall Loss 0.141377 Objective Loss 0.141377 LR 0.000250 Time 0.019762 -2022-12-06 11:40:43,848 - Epoch: [174][ 1190/ 1200] Overall Loss 0.141479 Objective Loss 0.141479 LR 0.000250 Time 0.019758 -2022-12-06 11:40:44,078 - Epoch: [174][ 1200/ 1200] Overall Loss 0.141545 Objective Loss 0.141545 Top1 89.121339 Top5 98.953975 LR 0.000250 Time 0.019785 -2022-12-06 11:40:44,167 - --- validate (epoch=174)----------- -2022-12-06 11:40:44,168 - 34129 samples (256 per mini-batch) -2022-12-06 11:40:44,612 - Epoch: [174][ 10/ 134] Loss 0.223618 Top1 88.281250 Top5 98.437500 -2022-12-06 11:40:44,742 - Epoch: [174][ 20/ 134] Loss 0.231410 Top1 88.085938 Top5 98.457031 -2022-12-06 11:40:44,872 - Epoch: [174][ 30/ 134] Loss 0.222684 Top1 88.177083 Top5 98.463542 -2022-12-06 11:40:45,001 - Epoch: [174][ 40/ 134] Loss 0.225220 Top1 88.212891 Top5 98.554688 -2022-12-06 11:40:45,134 - Epoch: [174][ 50/ 134] Loss 0.229826 Top1 88.265625 Top5 98.585938 -2022-12-06 11:40:45,267 - Epoch: [174][ 60/ 134] Loss 0.229740 Top1 88.261719 Top5 98.600260 -2022-12-06 11:40:45,400 - Epoch: [174][ 70/ 134] Loss 0.234401 Top1 88.152902 Top5 98.616071 -2022-12-06 11:40:45,533 - Epoch: [174][ 80/ 134] Loss 0.233302 Top1 88.168945 Top5 98.618164 -2022-12-06 11:40:45,665 - Epoch: [174][ 90/ 134] Loss 0.232319 Top1 88.237847 Top5 98.637153 -2022-12-06 11:40:45,798 - Epoch: [174][ 100/ 134] Loss 0.235531 Top1 88.167969 Top5 98.609375 -2022-12-06 11:40:45,931 - Epoch: [174][ 110/ 134] Loss 0.233745 Top1 88.274148 Top5 98.654119 -2022-12-06 11:40:46,068 - Epoch: [174][ 120/ 134] Loss 0.233523 Top1 88.242188 Top5 98.619792 -2022-12-06 11:40:46,203 - Epoch: [174][ 130/ 134] Loss 0.232693 Top1 88.290264 Top5 98.599760 -2022-12-06 11:40:46,240 - Epoch: [174][ 134/ 134] Loss 0.232528 Top1 88.329573 Top5 98.605292 -2022-12-06 11:40:46,327 - ==> Top1: 88.330 Top5: 98.605 Loss: 0.233 - -2022-12-06 11:40:46,328 - ==> Confusion: -[[ 926 0 2 1 2 6 0 2 4 40 0 2 1 2 3 2 1 0 0 0 2] - [ 0 934 2 2 9 24 3 11 2 2 0 7 0 1 0 2 2 0 12 2 12] - [ 5 3 1014 14 3 2 20 7 0 1 5 4 3 3 2 1 0 0 3 3 10] - [ 4 3 15 953 0 3 0 0 0 1 8 0 3 2 12 0 1 1 9 1 4] - [ 14 5 1 0 958 1 1 1 1 6 1 2 0 2 7 4 5 2 1 1 7] - [ 2 5 0 4 3 996 2 13 3 2 1 12 1 15 1 1 1 2 0 1 4] - [ 1 1 4 3 1 1 1082 2 0 0 1 4 0 0 0 5 0 3 0 7 3] - [ 0 6 4 3 1 29 5 962 0 0 1 6 0 2 0 2 1 1 15 8 8] - [ 7 2 0 1 0 0 1 1 975 43 6 3 0 9 9 1 2 0 1 1 2] - [ 52 2 1 0 3 1 0 2 14 898 1 2 0 13 2 0 0 3 0 0 7] - [ 0 2 4 6 1 0 1 2 6 2 959 0 0 15 5 1 0 0 6 1 8] - [ 3 1 1 0 0 10 4 1 1 1 1 992 14 5 0 5 2 5 0 3 2] - [ 0 1 1 2 1 2 1 1 0 0 0 28 897 0 0 10 0 15 0 2 8] - [ 0 0 0 0 1 4 0 3 8 8 1 3 3 976 2 2 3 0 0 2 7] - [ 7 4 2 8 2 2 0 1 13 3 1 2 1 3 1070 0 0 1 5 1 4] - [ 1 0 1 0 2 0 2 0 2 0 0 8 4 1 0 995 6 17 0 2 2] - [ 2 3 0 2 1 1 1 0 2 1 0 2 3 3 0 10 1029 0 0 5 7] - [ 3 1 1 4 0 1 1 0 1 4 0 6 12 2 1 11 0 986 0 0 2] - [ 1 3 4 12 1 3 0 22 2 1 3 3 2 0 8 1 0 1 936 1 4] - [ 2 5 0 1 0 5 3 11 0 1 2 20 5 5 0 5 2 3 1 1002 7] - [ 123 184 142 98 79 163 69 155 64 69 126 102 249 247 132 110 137 96 113 167 10601]] - -2022-12-06 11:40:47,003 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:40:47,003 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:40:47,009 - - -2022-12-06 11:40:47,009 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:40:47,940 - Epoch: [175][ 10/ 1200] Overall Loss 0.144083 Objective Loss 0.144083 LR 0.000250 Time 0.093061 -2022-12-06 11:40:48,139 - Epoch: [175][ 20/ 1200] Overall Loss 0.137306 Objective Loss 0.137306 LR 0.000250 Time 0.056420 -2022-12-06 11:40:48,330 - Epoch: [175][ 30/ 1200] Overall Loss 0.139929 Objective Loss 0.139929 LR 0.000250 Time 0.043982 -2022-12-06 11:40:48,521 - Epoch: [175][ 40/ 1200] Overall Loss 0.140275 Objective Loss 0.140275 LR 0.000250 Time 0.037745 -2022-12-06 11:40:48,713 - Epoch: [175][ 50/ 1200] Overall Loss 0.134759 Objective Loss 0.134759 LR 0.000250 Time 0.034022 -2022-12-06 11:40:48,905 - Epoch: [175][ 60/ 1200] Overall Loss 0.136938 Objective Loss 0.136938 LR 0.000250 Time 0.031544 -2022-12-06 11:40:49,097 - Epoch: [175][ 70/ 1200] Overall Loss 0.135245 Objective Loss 0.135245 LR 0.000250 Time 0.029765 -2022-12-06 11:40:49,288 - Epoch: [175][ 80/ 1200] Overall Loss 0.138047 Objective Loss 0.138047 LR 0.000250 Time 0.028425 -2022-12-06 11:40:49,479 - Epoch: [175][ 90/ 1200] Overall Loss 0.137788 Objective Loss 0.137788 LR 0.000250 Time 0.027390 -2022-12-06 11:40:49,671 - Epoch: [175][ 100/ 1200] Overall Loss 0.136939 Objective Loss 0.136939 LR 0.000250 Time 0.026559 -2022-12-06 11:40:49,862 - Epoch: [175][ 110/ 1200] Overall Loss 0.137092 Objective Loss 0.137092 LR 0.000250 Time 0.025876 -2022-12-06 11:40:50,052 - Epoch: [175][ 120/ 1200] Overall Loss 0.136753 Objective Loss 0.136753 LR 0.000250 Time 0.025299 -2022-12-06 11:40:50,243 - Epoch: [175][ 130/ 1200] Overall Loss 0.136381 Objective Loss 0.136381 LR 0.000250 Time 0.024818 -2022-12-06 11:40:50,433 - Epoch: [175][ 140/ 1200] Overall Loss 0.137122 Objective Loss 0.137122 LR 0.000250 Time 0.024402 -2022-12-06 11:40:50,624 - Epoch: [175][ 150/ 1200] Overall Loss 0.136338 Objective Loss 0.136338 LR 0.000250 Time 0.024044 -2022-12-06 11:40:50,816 - Epoch: [175][ 160/ 1200] Overall Loss 0.136561 Objective Loss 0.136561 LR 0.000250 Time 0.023735 -2022-12-06 11:40:51,007 - Epoch: [175][ 170/ 1200] Overall Loss 0.136131 Objective Loss 0.136131 LR 0.000250 Time 0.023462 -2022-12-06 11:40:51,198 - Epoch: [175][ 180/ 1200] Overall Loss 0.136372 Objective Loss 0.136372 LR 0.000250 Time 0.023216 -2022-12-06 11:40:51,390 - Epoch: [175][ 190/ 1200] Overall Loss 0.137808 Objective Loss 0.137808 LR 0.000250 Time 0.022999 -2022-12-06 11:40:51,580 - Epoch: [175][ 200/ 1200] Overall Loss 0.138084 Objective Loss 0.138084 LR 0.000250 Time 0.022799 -2022-12-06 11:40:51,771 - Epoch: [175][ 210/ 1200] Overall Loss 0.138603 Objective Loss 0.138603 LR 0.000250 Time 0.022620 -2022-12-06 11:40:51,962 - Epoch: [175][ 220/ 1200] Overall Loss 0.139255 Objective Loss 0.139255 LR 0.000250 Time 0.022457 -2022-12-06 11:40:52,153 - Epoch: [175][ 230/ 1200] Overall Loss 0.138053 Objective Loss 0.138053 LR 0.000250 Time 0.022308 -2022-12-06 11:40:52,344 - Epoch: [175][ 240/ 1200] Overall Loss 0.138692 Objective Loss 0.138692 LR 0.000250 Time 0.022173 -2022-12-06 11:40:52,535 - Epoch: [175][ 250/ 1200] Overall Loss 0.138921 Objective Loss 0.138921 LR 0.000250 Time 0.022048 -2022-12-06 11:40:52,725 - Epoch: [175][ 260/ 1200] Overall Loss 0.139128 Objective Loss 0.139128 LR 0.000250 Time 0.021929 -2022-12-06 11:40:52,916 - Epoch: [175][ 270/ 1200] Overall Loss 0.139260 Objective Loss 0.139260 LR 0.000250 Time 0.021822 -2022-12-06 11:40:53,108 - Epoch: [175][ 280/ 1200] Overall Loss 0.138855 Objective Loss 0.138855 LR 0.000250 Time 0.021725 -2022-12-06 11:40:53,299 - Epoch: [175][ 290/ 1200] Overall Loss 0.139532 Objective Loss 0.139532 LR 0.000250 Time 0.021633 -2022-12-06 11:40:53,490 - Epoch: [175][ 300/ 1200] Overall Loss 0.138814 Objective Loss 0.138814 LR 0.000250 Time 0.021547 -2022-12-06 11:40:53,680 - Epoch: [175][ 310/ 1200] Overall Loss 0.138070 Objective Loss 0.138070 LR 0.000250 Time 0.021463 -2022-12-06 11:40:53,871 - Epoch: [175][ 320/ 1200] Overall Loss 0.138622 Objective Loss 0.138622 LR 0.000250 Time 0.021387 -2022-12-06 11:40:54,063 - Epoch: [175][ 330/ 1200] Overall Loss 0.138395 Objective Loss 0.138395 LR 0.000250 Time 0.021318 -2022-12-06 11:40:54,253 - Epoch: [175][ 340/ 1200] Overall Loss 0.139319 Objective Loss 0.139319 LR 0.000250 Time 0.021250 -2022-12-06 11:40:54,444 - Epoch: [175][ 350/ 1200] Overall Loss 0.139438 Objective Loss 0.139438 LR 0.000250 Time 0.021187 -2022-12-06 11:40:54,636 - Epoch: [175][ 360/ 1200] Overall Loss 0.139158 Objective Loss 0.139158 LR 0.000250 Time 0.021129 -2022-12-06 11:40:54,826 - Epoch: [175][ 370/ 1200] Overall Loss 0.139056 Objective Loss 0.139056 LR 0.000250 Time 0.021072 -2022-12-06 11:40:55,016 - Epoch: [175][ 380/ 1200] Overall Loss 0.139068 Objective Loss 0.139068 LR 0.000250 Time 0.021016 -2022-12-06 11:40:55,207 - Epoch: [175][ 390/ 1200] Overall Loss 0.139164 Objective Loss 0.139164 LR 0.000250 Time 0.020965 -2022-12-06 11:40:55,399 - Epoch: [175][ 400/ 1200] Overall Loss 0.139305 Objective Loss 0.139305 LR 0.000250 Time 0.020918 -2022-12-06 11:40:55,590 - Epoch: [175][ 410/ 1200] Overall Loss 0.139288 Objective Loss 0.139288 LR 0.000250 Time 0.020874 -2022-12-06 11:40:55,782 - Epoch: [175][ 420/ 1200] Overall Loss 0.139935 Objective Loss 0.139935 LR 0.000250 Time 0.020832 -2022-12-06 11:40:55,973 - Epoch: [175][ 430/ 1200] Overall Loss 0.140035 Objective Loss 0.140035 LR 0.000250 Time 0.020791 -2022-12-06 11:40:56,164 - Epoch: [175][ 440/ 1200] Overall Loss 0.139988 Objective Loss 0.139988 LR 0.000250 Time 0.020751 -2022-12-06 11:40:56,355 - Epoch: [175][ 450/ 1200] Overall Loss 0.140008 Objective Loss 0.140008 LR 0.000250 Time 0.020714 -2022-12-06 11:40:56,546 - Epoch: [175][ 460/ 1200] Overall Loss 0.139683 Objective Loss 0.139683 LR 0.000250 Time 0.020677 -2022-12-06 11:40:56,738 - Epoch: [175][ 470/ 1200] Overall Loss 0.140183 Objective Loss 0.140183 LR 0.000250 Time 0.020644 -2022-12-06 11:40:56,929 - Epoch: [175][ 480/ 1200] Overall Loss 0.140355 Objective Loss 0.140355 LR 0.000250 Time 0.020611 -2022-12-06 11:40:57,121 - Epoch: [175][ 490/ 1200] Overall Loss 0.140364 Objective Loss 0.140364 LR 0.000250 Time 0.020581 -2022-12-06 11:40:57,312 - Epoch: [175][ 500/ 1200] Overall Loss 0.140016 Objective Loss 0.140016 LR 0.000250 Time 0.020551 -2022-12-06 11:40:57,504 - Epoch: [175][ 510/ 1200] Overall Loss 0.140110 Objective Loss 0.140110 LR 0.000250 Time 0.020522 -2022-12-06 11:40:57,695 - Epoch: [175][ 520/ 1200] Overall Loss 0.139799 Objective Loss 0.139799 LR 0.000250 Time 0.020494 -2022-12-06 11:40:57,886 - Epoch: [175][ 530/ 1200] Overall Loss 0.139750 Objective Loss 0.139750 LR 0.000250 Time 0.020467 -2022-12-06 11:40:58,078 - Epoch: [175][ 540/ 1200] Overall Loss 0.139761 Objective Loss 0.139761 LR 0.000250 Time 0.020442 -2022-12-06 11:40:58,270 - Epoch: [175][ 550/ 1200] Overall Loss 0.139813 Objective Loss 0.139813 LR 0.000250 Time 0.020418 -2022-12-06 11:40:58,461 - Epoch: [175][ 560/ 1200] Overall Loss 0.140042 Objective Loss 0.140042 LR 0.000250 Time 0.020393 -2022-12-06 11:40:58,652 - Epoch: [175][ 570/ 1200] Overall Loss 0.140328 Objective Loss 0.140328 LR 0.000250 Time 0.020371 -2022-12-06 11:40:58,842 - Epoch: [175][ 580/ 1200] Overall Loss 0.140460 Objective Loss 0.140460 LR 0.000250 Time 0.020347 -2022-12-06 11:40:59,034 - Epoch: [175][ 590/ 1200] Overall Loss 0.140546 Objective Loss 0.140546 LR 0.000250 Time 0.020326 -2022-12-06 11:40:59,224 - Epoch: [175][ 600/ 1200] Overall Loss 0.140582 Objective Loss 0.140582 LR 0.000250 Time 0.020302 -2022-12-06 11:40:59,415 - Epoch: [175][ 610/ 1200] Overall Loss 0.140379 Objective Loss 0.140379 LR 0.000250 Time 0.020282 -2022-12-06 11:40:59,606 - Epoch: [175][ 620/ 1200] Overall Loss 0.140532 Objective Loss 0.140532 LR 0.000250 Time 0.020261 -2022-12-06 11:40:59,797 - Epoch: [175][ 630/ 1200] Overall Loss 0.140566 Objective Loss 0.140566 LR 0.000250 Time 0.020243 -2022-12-06 11:40:59,988 - Epoch: [175][ 640/ 1200] Overall Loss 0.140425 Objective Loss 0.140425 LR 0.000250 Time 0.020224 -2022-12-06 11:41:00,179 - Epoch: [175][ 650/ 1200] Overall Loss 0.140548 Objective Loss 0.140548 LR 0.000250 Time 0.020206 -2022-12-06 11:41:00,370 - Epoch: [175][ 660/ 1200] Overall Loss 0.140514 Objective Loss 0.140514 LR 0.000250 Time 0.020188 -2022-12-06 11:41:00,562 - Epoch: [175][ 670/ 1200] Overall Loss 0.140875 Objective Loss 0.140875 LR 0.000250 Time 0.020173 -2022-12-06 11:41:00,752 - Epoch: [175][ 680/ 1200] Overall Loss 0.140790 Objective Loss 0.140790 LR 0.000250 Time 0.020154 -2022-12-06 11:41:00,944 - Epoch: [175][ 690/ 1200] Overall Loss 0.141027 Objective Loss 0.141027 LR 0.000250 Time 0.020139 -2022-12-06 11:41:01,134 - Epoch: [175][ 700/ 1200] Overall Loss 0.141113 Objective Loss 0.141113 LR 0.000250 Time 0.020122 -2022-12-06 11:41:01,325 - Epoch: [175][ 710/ 1200] Overall Loss 0.140979 Objective Loss 0.140979 LR 0.000250 Time 0.020107 -2022-12-06 11:41:01,515 - Epoch: [175][ 720/ 1200] Overall Loss 0.140789 Objective Loss 0.140789 LR 0.000250 Time 0.020091 -2022-12-06 11:41:01,706 - Epoch: [175][ 730/ 1200] Overall Loss 0.140637 Objective Loss 0.140637 LR 0.000250 Time 0.020076 -2022-12-06 11:41:01,897 - Epoch: [175][ 740/ 1200] Overall Loss 0.140529 Objective Loss 0.140529 LR 0.000250 Time 0.020063 -2022-12-06 11:41:02,088 - Epoch: [175][ 750/ 1200] Overall Loss 0.140807 Objective Loss 0.140807 LR 0.000250 Time 0.020050 -2022-12-06 11:41:02,279 - Epoch: [175][ 760/ 1200] Overall Loss 0.140702 Objective Loss 0.140702 LR 0.000250 Time 0.020036 -2022-12-06 11:41:02,471 - Epoch: [175][ 770/ 1200] Overall Loss 0.140861 Objective Loss 0.140861 LR 0.000250 Time 0.020024 -2022-12-06 11:41:02,662 - Epoch: [175][ 780/ 1200] Overall Loss 0.140770 Objective Loss 0.140770 LR 0.000250 Time 0.020011 -2022-12-06 11:41:02,854 - Epoch: [175][ 790/ 1200] Overall Loss 0.140786 Objective Loss 0.140786 LR 0.000250 Time 0.020000 -2022-12-06 11:41:03,045 - Epoch: [175][ 800/ 1200] Overall Loss 0.140636 Objective Loss 0.140636 LR 0.000250 Time 0.019988 -2022-12-06 11:41:03,235 - Epoch: [175][ 810/ 1200] Overall Loss 0.140769 Objective Loss 0.140769 LR 0.000250 Time 0.019976 -2022-12-06 11:41:03,425 - Epoch: [175][ 820/ 1200] Overall Loss 0.140777 Objective Loss 0.140777 LR 0.000250 Time 0.019963 -2022-12-06 11:41:03,616 - Epoch: [175][ 830/ 1200] Overall Loss 0.140627 Objective Loss 0.140627 LR 0.000250 Time 0.019952 -2022-12-06 11:41:03,806 - Epoch: [175][ 840/ 1200] Overall Loss 0.140425 Objective Loss 0.140425 LR 0.000250 Time 0.019940 -2022-12-06 11:41:03,996 - Epoch: [175][ 850/ 1200] Overall Loss 0.140433 Objective Loss 0.140433 LR 0.000250 Time 0.019929 -2022-12-06 11:41:04,186 - Epoch: [175][ 860/ 1200] Overall Loss 0.140668 Objective Loss 0.140668 LR 0.000250 Time 0.019917 -2022-12-06 11:41:04,378 - Epoch: [175][ 870/ 1200] Overall Loss 0.140532 Objective Loss 0.140532 LR 0.000250 Time 0.019908 -2022-12-06 11:41:04,569 - Epoch: [175][ 880/ 1200] Overall Loss 0.140630 Objective Loss 0.140630 LR 0.000250 Time 0.019898 -2022-12-06 11:41:04,760 - Epoch: [175][ 890/ 1200] Overall Loss 0.140853 Objective Loss 0.140853 LR 0.000250 Time 0.019888 -2022-12-06 11:41:04,951 - Epoch: [175][ 900/ 1200] Overall Loss 0.141031 Objective Loss 0.141031 LR 0.000250 Time 0.019879 -2022-12-06 11:41:05,142 - Epoch: [175][ 910/ 1200] Overall Loss 0.141313 Objective Loss 0.141313 LR 0.000250 Time 0.019870 -2022-12-06 11:41:05,333 - Epoch: [175][ 920/ 1200] Overall Loss 0.141376 Objective Loss 0.141376 LR 0.000250 Time 0.019861 -2022-12-06 11:41:05,525 - Epoch: [175][ 930/ 1200] Overall Loss 0.141355 Objective Loss 0.141355 LR 0.000250 Time 0.019853 -2022-12-06 11:41:05,715 - Epoch: [175][ 940/ 1200] Overall Loss 0.141592 Objective Loss 0.141592 LR 0.000250 Time 0.019844 -2022-12-06 11:41:05,906 - Epoch: [175][ 950/ 1200] Overall Loss 0.141689 Objective Loss 0.141689 LR 0.000250 Time 0.019835 -2022-12-06 11:41:06,098 - Epoch: [175][ 960/ 1200] Overall Loss 0.141727 Objective Loss 0.141727 LR 0.000250 Time 0.019827 -2022-12-06 11:41:06,289 - Epoch: [175][ 970/ 1200] Overall Loss 0.141764 Objective Loss 0.141764 LR 0.000250 Time 0.019819 -2022-12-06 11:41:06,480 - Epoch: [175][ 980/ 1200] Overall Loss 0.141467 Objective Loss 0.141467 LR 0.000250 Time 0.019812 -2022-12-06 11:41:06,672 - Epoch: [175][ 990/ 1200] Overall Loss 0.141399 Objective Loss 0.141399 LR 0.000250 Time 0.019805 -2022-12-06 11:41:06,863 - Epoch: [175][ 1000/ 1200] Overall Loss 0.141311 Objective Loss 0.141311 LR 0.000250 Time 0.019797 -2022-12-06 11:41:07,054 - Epoch: [175][ 1010/ 1200] Overall Loss 0.141195 Objective Loss 0.141195 LR 0.000250 Time 0.019790 -2022-12-06 11:41:07,245 - Epoch: [175][ 1020/ 1200] Overall Loss 0.141182 Objective Loss 0.141182 LR 0.000250 Time 0.019782 -2022-12-06 11:41:07,436 - Epoch: [175][ 1030/ 1200] Overall Loss 0.141256 Objective Loss 0.141256 LR 0.000250 Time 0.019775 -2022-12-06 11:41:07,627 - Epoch: [175][ 1040/ 1200] Overall Loss 0.141141 Objective Loss 0.141141 LR 0.000250 Time 0.019768 -2022-12-06 11:41:07,819 - Epoch: [175][ 1050/ 1200] Overall Loss 0.141221 Objective Loss 0.141221 LR 0.000250 Time 0.019762 -2022-12-06 11:41:08,009 - Epoch: [175][ 1060/ 1200] Overall Loss 0.141144 Objective Loss 0.141144 LR 0.000250 Time 0.019755 -2022-12-06 11:41:08,200 - Epoch: [175][ 1070/ 1200] Overall Loss 0.141105 Objective Loss 0.141105 LR 0.000250 Time 0.019748 -2022-12-06 11:41:08,391 - Epoch: [175][ 1080/ 1200] Overall Loss 0.141068 Objective Loss 0.141068 LR 0.000250 Time 0.019741 -2022-12-06 11:41:08,583 - Epoch: [175][ 1090/ 1200] Overall Loss 0.141113 Objective Loss 0.141113 LR 0.000250 Time 0.019736 -2022-12-06 11:41:08,773 - Epoch: [175][ 1100/ 1200] Overall Loss 0.141160 Objective Loss 0.141160 LR 0.000250 Time 0.019728 -2022-12-06 11:41:08,964 - Epoch: [175][ 1110/ 1200] Overall Loss 0.140911 Objective Loss 0.140911 LR 0.000250 Time 0.019723 -2022-12-06 11:41:09,155 - Epoch: [175][ 1120/ 1200] Overall Loss 0.140881 Objective Loss 0.140881 LR 0.000250 Time 0.019716 -2022-12-06 11:41:09,347 - Epoch: [175][ 1130/ 1200] Overall Loss 0.140811 Objective Loss 0.140811 LR 0.000250 Time 0.019711 -2022-12-06 11:41:09,537 - Epoch: [175][ 1140/ 1200] Overall Loss 0.140932 Objective Loss 0.140932 LR 0.000250 Time 0.019704 -2022-12-06 11:41:09,728 - Epoch: [175][ 1150/ 1200] Overall Loss 0.141058 Objective Loss 0.141058 LR 0.000250 Time 0.019699 -2022-12-06 11:41:09,919 - Epoch: [175][ 1160/ 1200] Overall Loss 0.141090 Objective Loss 0.141090 LR 0.000250 Time 0.019693 -2022-12-06 11:41:10,111 - Epoch: [175][ 1170/ 1200] Overall Loss 0.140877 Objective Loss 0.140877 LR 0.000250 Time 0.019688 -2022-12-06 11:41:10,302 - Epoch: [175][ 1180/ 1200] Overall Loss 0.140925 Objective Loss 0.140925 LR 0.000250 Time 0.019683 -2022-12-06 11:41:10,493 - Epoch: [175][ 1190/ 1200] Overall Loss 0.141012 Objective Loss 0.141012 LR 0.000250 Time 0.019678 -2022-12-06 11:41:10,721 - Epoch: [175][ 1200/ 1200] Overall Loss 0.141008 Objective Loss 0.141008 Top1 91.631799 Top5 98.535565 LR 0.000250 Time 0.019703 -2022-12-06 11:41:10,810 - --- validate (epoch=175)----------- -2022-12-06 11:41:10,811 - 34129 samples (256 per mini-batch) -2022-12-06 11:41:11,258 - Epoch: [175][ 10/ 134] Loss 0.223933 Top1 87.851562 Top5 98.750000 -2022-12-06 11:41:11,385 - Epoch: [175][ 20/ 134] Loss 0.223593 Top1 88.320312 Top5 98.691406 -2022-12-06 11:41:11,528 - Epoch: [175][ 30/ 134] Loss 0.229936 Top1 87.864583 Top5 98.710938 -2022-12-06 11:41:11,674 - Epoch: [175][ 40/ 134] Loss 0.246268 Top1 87.636719 Top5 98.691406 -2022-12-06 11:41:11,807 - Epoch: [175][ 50/ 134] Loss 0.252422 Top1 87.531250 Top5 98.703125 -2022-12-06 11:41:11,939 - Epoch: [175][ 60/ 134] Loss 0.252908 Top1 87.545573 Top5 98.619792 -2022-12-06 11:41:12,068 - Epoch: [175][ 70/ 134] Loss 0.249384 Top1 87.712054 Top5 98.632812 -2022-12-06 11:41:12,199 - Epoch: [175][ 80/ 134] Loss 0.244589 Top1 87.812500 Top5 98.623047 -2022-12-06 11:41:12,330 - Epoch: [175][ 90/ 134] Loss 0.244068 Top1 87.773438 Top5 98.632812 -2022-12-06 11:41:12,460 - Epoch: [175][ 100/ 134] Loss 0.242351 Top1 87.718750 Top5 98.574219 -2022-12-06 11:41:12,591 - Epoch: [175][ 110/ 134] Loss 0.237774 Top1 87.862216 Top5 98.600852 -2022-12-06 11:41:12,724 - Epoch: [175][ 120/ 134] Loss 0.235314 Top1 87.864583 Top5 98.636068 -2022-12-06 11:41:12,857 - Epoch: [175][ 130/ 134] Loss 0.235604 Top1 87.764423 Top5 98.629808 -2022-12-06 11:41:12,895 - Epoch: [175][ 134/ 134] Loss 0.233394 Top1 87.810953 Top5 98.640452 -2022-12-06 11:41:12,990 - ==> Top1: 87.811 Top5: 98.640 Loss: 0.233 - -2022-12-06 11:41:12,991 - ==> Confusion: -[[ 926 0 2 2 2 9 0 1 3 36 0 1 0 2 3 2 3 1 1 0 2] - [ 3 936 2 2 7 26 3 14 1 0 3 4 0 1 1 2 1 0 9 4 8] - [ 6 3 1019 12 5 3 16 5 0 1 4 4 1 1 2 2 1 3 2 4 9] - [ 3 3 13 960 0 5 0 0 0 0 7 0 2 0 7 0 0 3 8 1 8] - [ 7 2 2 0 963 2 0 2 0 5 1 2 1 2 10 5 7 3 0 1 5] - [ 2 4 0 3 3 1012 2 14 1 0 1 7 2 5 1 1 2 1 1 5 2] - [ 2 2 6 2 2 2 1079 1 0 1 1 1 0 1 0 5 0 1 2 7 3] - [ 2 6 3 2 2 42 8 950 0 0 0 6 1 0 1 0 2 0 13 10 6] - [ 7 2 0 0 1 1 1 2 982 37 8 1 1 3 11 1 3 0 2 1 0] - [ 73 0 0 0 7 4 0 4 23 865 1 1 0 11 4 0 0 1 0 0 7] - [ 1 1 2 6 1 3 3 3 8 3 966 0 1 9 4 1 0 0 1 1 5] - [ 2 2 1 0 1 13 2 1 0 0 0 978 22 2 0 6 4 6 0 7 4] - [ 1 1 0 2 1 2 1 0 0 0 0 26 912 1 0 5 3 6 0 3 5] - [ 1 0 1 0 1 15 0 0 12 6 2 3 4 962 0 1 2 1 0 3 9] - [ 7 2 1 14 2 4 0 0 14 2 1 3 1 3 1065 0 0 1 4 1 5] - [ 0 0 1 3 3 0 2 1 0 0 0 6 5 2 0 985 11 16 0 4 4] - [ 0 3 1 1 2 2 2 0 0 0 1 0 3 2 0 8 1036 1 0 5 5] - [ 3 0 2 3 1 1 0 0 0 1 0 7 13 1 1 13 0 987 0 1 2] - [ 3 4 1 8 1 3 0 20 1 1 4 4 3 0 5 0 0 0 944 1 5] - [ 2 4 1 2 0 6 5 3 0 2 1 10 6 5 0 3 5 2 0 1017 6] - [ 124 171 151 121 107 214 67 121 66 71 150 76 324 225 139 81 164 79 127 224 10424]] - -2022-12-06 11:41:13,650 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:41:13,651 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:41:13,656 - - -2022-12-06 11:41:13,657 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:41:14,593 - Epoch: [176][ 10/ 1200] Overall Loss 0.150301 Objective Loss 0.150301 LR 0.000250 Time 0.093544 -2022-12-06 11:41:14,800 - Epoch: [176][ 20/ 1200] Overall Loss 0.143579 Objective Loss 0.143579 LR 0.000250 Time 0.057110 -2022-12-06 11:41:14,997 - Epoch: [176][ 30/ 1200] Overall Loss 0.143465 Objective Loss 0.143465 LR 0.000250 Time 0.044624 -2022-12-06 11:41:15,196 - Epoch: [176][ 40/ 1200] Overall Loss 0.142537 Objective Loss 0.142537 LR 0.000250 Time 0.038438 -2022-12-06 11:41:15,392 - Epoch: [176][ 50/ 1200] Overall Loss 0.138541 Objective Loss 0.138541 LR 0.000250 Time 0.034660 -2022-12-06 11:41:15,591 - Epoch: [176][ 60/ 1200] Overall Loss 0.136754 Objective Loss 0.136754 LR 0.000250 Time 0.032190 -2022-12-06 11:41:15,787 - Epoch: [176][ 70/ 1200] Overall Loss 0.139371 Objective Loss 0.139371 LR 0.000250 Time 0.030384 -2022-12-06 11:41:15,986 - Epoch: [176][ 80/ 1200] Overall Loss 0.141885 Objective Loss 0.141885 LR 0.000250 Time 0.029064 -2022-12-06 11:41:16,183 - Epoch: [176][ 90/ 1200] Overall Loss 0.142641 Objective Loss 0.142641 LR 0.000250 Time 0.028014 -2022-12-06 11:41:16,382 - Epoch: [176][ 100/ 1200] Overall Loss 0.141835 Objective Loss 0.141835 LR 0.000250 Time 0.027200 -2022-12-06 11:41:16,579 - Epoch: [176][ 110/ 1200] Overall Loss 0.140791 Objective Loss 0.140791 LR 0.000250 Time 0.026509 -2022-12-06 11:41:16,777 - Epoch: [176][ 120/ 1200] Overall Loss 0.138750 Objective Loss 0.138750 LR 0.000250 Time 0.025951 -2022-12-06 11:41:16,974 - Epoch: [176][ 130/ 1200] Overall Loss 0.138626 Objective Loss 0.138626 LR 0.000250 Time 0.025463 -2022-12-06 11:41:17,174 - Epoch: [176][ 140/ 1200] Overall Loss 0.138772 Objective Loss 0.138772 LR 0.000250 Time 0.025066 -2022-12-06 11:41:17,370 - Epoch: [176][ 150/ 1200] Overall Loss 0.139213 Objective Loss 0.139213 LR 0.000250 Time 0.024702 -2022-12-06 11:41:17,570 - Epoch: [176][ 160/ 1200] Overall Loss 0.139961 Objective Loss 0.139961 LR 0.000250 Time 0.024401 -2022-12-06 11:41:17,766 - Epoch: [176][ 170/ 1200] Overall Loss 0.138862 Objective Loss 0.138862 LR 0.000250 Time 0.024117 -2022-12-06 11:41:17,965 - Epoch: [176][ 180/ 1200] Overall Loss 0.139529 Objective Loss 0.139529 LR 0.000250 Time 0.023880 -2022-12-06 11:41:18,162 - Epoch: [176][ 190/ 1200] Overall Loss 0.138887 Objective Loss 0.138887 LR 0.000250 Time 0.023656 -2022-12-06 11:41:18,361 - Epoch: [176][ 200/ 1200] Overall Loss 0.139068 Objective Loss 0.139068 LR 0.000250 Time 0.023466 -2022-12-06 11:41:18,556 - Epoch: [176][ 210/ 1200] Overall Loss 0.139034 Objective Loss 0.139034 LR 0.000250 Time 0.023275 -2022-12-06 11:41:18,749 - Epoch: [176][ 220/ 1200] Overall Loss 0.138740 Objective Loss 0.138740 LR 0.000250 Time 0.023091 -2022-12-06 11:41:18,940 - Epoch: [176][ 230/ 1200] Overall Loss 0.138649 Objective Loss 0.138649 LR 0.000250 Time 0.022918 -2022-12-06 11:41:19,133 - Epoch: [176][ 240/ 1200] Overall Loss 0.139239 Objective Loss 0.139239 LR 0.000250 Time 0.022763 -2022-12-06 11:41:19,324 - Epoch: [176][ 250/ 1200] Overall Loss 0.139252 Objective Loss 0.139252 LR 0.000250 Time 0.022617 -2022-12-06 11:41:19,518 - Epoch: [176][ 260/ 1200] Overall Loss 0.139122 Objective Loss 0.139122 LR 0.000250 Time 0.022489 -2022-12-06 11:41:19,710 - Epoch: [176][ 270/ 1200] Overall Loss 0.139202 Objective Loss 0.139202 LR 0.000250 Time 0.022366 -2022-12-06 11:41:19,903 - Epoch: [176][ 280/ 1200] Overall Loss 0.139402 Objective Loss 0.139402 LR 0.000250 Time 0.022252 -2022-12-06 11:41:20,095 - Epoch: [176][ 290/ 1200] Overall Loss 0.139840 Objective Loss 0.139840 LR 0.000250 Time 0.022146 -2022-12-06 11:41:20,288 - Epoch: [176][ 300/ 1200] Overall Loss 0.139362 Objective Loss 0.139362 LR 0.000250 Time 0.022050 -2022-12-06 11:41:20,481 - Epoch: [176][ 310/ 1200] Overall Loss 0.139589 Objective Loss 0.139589 LR 0.000250 Time 0.021958 -2022-12-06 11:41:20,673 - Epoch: [176][ 320/ 1200] Overall Loss 0.139074 Objective Loss 0.139074 LR 0.000250 Time 0.021871 -2022-12-06 11:41:20,865 - Epoch: [176][ 330/ 1200] Overall Loss 0.138547 Objective Loss 0.138547 LR 0.000250 Time 0.021789 -2022-12-06 11:41:21,058 - Epoch: [176][ 340/ 1200] Overall Loss 0.138375 Objective Loss 0.138375 LR 0.000250 Time 0.021713 -2022-12-06 11:41:21,250 - Epoch: [176][ 350/ 1200] Overall Loss 0.138141 Objective Loss 0.138141 LR 0.000250 Time 0.021639 -2022-12-06 11:41:21,441 - Epoch: [176][ 360/ 1200] Overall Loss 0.138327 Objective Loss 0.138327 LR 0.000250 Time 0.021569 -2022-12-06 11:41:21,633 - Epoch: [176][ 370/ 1200] Overall Loss 0.138619 Objective Loss 0.138619 LR 0.000250 Time 0.021503 -2022-12-06 11:41:21,825 - Epoch: [176][ 380/ 1200] Overall Loss 0.138630 Objective Loss 0.138630 LR 0.000250 Time 0.021441 -2022-12-06 11:41:22,017 - Epoch: [176][ 390/ 1200] Overall Loss 0.138949 Objective Loss 0.138949 LR 0.000250 Time 0.021381 -2022-12-06 11:41:22,209 - Epoch: [176][ 400/ 1200] Overall Loss 0.138979 Objective Loss 0.138979 LR 0.000250 Time 0.021327 -2022-12-06 11:41:22,401 - Epoch: [176][ 410/ 1200] Overall Loss 0.138913 Objective Loss 0.138913 LR 0.000250 Time 0.021273 -2022-12-06 11:41:22,594 - Epoch: [176][ 420/ 1200] Overall Loss 0.138643 Objective Loss 0.138643 LR 0.000250 Time 0.021223 -2022-12-06 11:41:22,786 - Epoch: [176][ 430/ 1200] Overall Loss 0.138215 Objective Loss 0.138215 LR 0.000250 Time 0.021176 -2022-12-06 11:41:22,978 - Epoch: [176][ 440/ 1200] Overall Loss 0.138658 Objective Loss 0.138658 LR 0.000250 Time 0.021130 -2022-12-06 11:41:23,170 - Epoch: [176][ 450/ 1200] Overall Loss 0.138376 Objective Loss 0.138376 LR 0.000250 Time 0.021086 -2022-12-06 11:41:23,362 - Epoch: [176][ 460/ 1200] Overall Loss 0.138100 Objective Loss 0.138100 LR 0.000250 Time 0.021044 -2022-12-06 11:41:23,554 - Epoch: [176][ 470/ 1200] Overall Loss 0.138253 Objective Loss 0.138253 LR 0.000250 Time 0.021004 -2022-12-06 11:41:23,747 - Epoch: [176][ 480/ 1200] Overall Loss 0.138685 Objective Loss 0.138685 LR 0.000250 Time 0.020966 -2022-12-06 11:41:23,939 - Epoch: [176][ 490/ 1200] Overall Loss 0.138966 Objective Loss 0.138966 LR 0.000250 Time 0.020930 -2022-12-06 11:41:24,132 - Epoch: [176][ 500/ 1200] Overall Loss 0.138758 Objective Loss 0.138758 LR 0.000250 Time 0.020895 -2022-12-06 11:41:24,324 - Epoch: [176][ 510/ 1200] Overall Loss 0.138708 Objective Loss 0.138708 LR 0.000250 Time 0.020861 -2022-12-06 11:41:24,517 - Epoch: [176][ 520/ 1200] Overall Loss 0.138778 Objective Loss 0.138778 LR 0.000250 Time 0.020829 -2022-12-06 11:41:24,708 - Epoch: [176][ 530/ 1200] Overall Loss 0.138757 Objective Loss 0.138757 LR 0.000250 Time 0.020797 -2022-12-06 11:41:24,902 - Epoch: [176][ 540/ 1200] Overall Loss 0.138846 Objective Loss 0.138846 LR 0.000250 Time 0.020769 -2022-12-06 11:41:25,095 - Epoch: [176][ 550/ 1200] Overall Loss 0.138791 Objective Loss 0.138791 LR 0.000250 Time 0.020741 -2022-12-06 11:41:25,287 - Epoch: [176][ 560/ 1200] Overall Loss 0.138692 Objective Loss 0.138692 LR 0.000250 Time 0.020712 -2022-12-06 11:41:25,479 - Epoch: [176][ 570/ 1200] Overall Loss 0.138837 Objective Loss 0.138837 LR 0.000250 Time 0.020685 -2022-12-06 11:41:25,672 - Epoch: [176][ 580/ 1200] Overall Loss 0.138901 Objective Loss 0.138901 LR 0.000250 Time 0.020661 -2022-12-06 11:41:25,865 - Epoch: [176][ 590/ 1200] Overall Loss 0.138769 Objective Loss 0.138769 LR 0.000250 Time 0.020636 -2022-12-06 11:41:26,057 - Epoch: [176][ 600/ 1200] Overall Loss 0.138355 Objective Loss 0.138355 LR 0.000250 Time 0.020612 -2022-12-06 11:41:26,250 - Epoch: [176][ 610/ 1200] Overall Loss 0.138192 Objective Loss 0.138192 LR 0.000250 Time 0.020589 -2022-12-06 11:41:26,442 - Epoch: [176][ 620/ 1200] Overall Loss 0.138531 Objective Loss 0.138531 LR 0.000250 Time 0.020567 -2022-12-06 11:41:26,635 - Epoch: [176][ 630/ 1200] Overall Loss 0.138655 Objective Loss 0.138655 LR 0.000250 Time 0.020545 -2022-12-06 11:41:26,827 - Epoch: [176][ 640/ 1200] Overall Loss 0.138705 Objective Loss 0.138705 LR 0.000250 Time 0.020524 -2022-12-06 11:41:27,020 - Epoch: [176][ 650/ 1200] Overall Loss 0.139027 Objective Loss 0.139027 LR 0.000250 Time 0.020504 -2022-12-06 11:41:27,214 - Epoch: [176][ 660/ 1200] Overall Loss 0.139134 Objective Loss 0.139134 LR 0.000250 Time 0.020486 -2022-12-06 11:41:27,407 - Epoch: [176][ 670/ 1200] Overall Loss 0.139429 Objective Loss 0.139429 LR 0.000250 Time 0.020467 -2022-12-06 11:41:27,599 - Epoch: [176][ 680/ 1200] Overall Loss 0.139375 Objective Loss 0.139375 LR 0.000250 Time 0.020448 -2022-12-06 11:41:27,792 - Epoch: [176][ 690/ 1200] Overall Loss 0.139300 Objective Loss 0.139300 LR 0.000250 Time 0.020430 -2022-12-06 11:41:27,985 - Epoch: [176][ 700/ 1200] Overall Loss 0.139231 Objective Loss 0.139231 LR 0.000250 Time 0.020413 -2022-12-06 11:41:28,177 - Epoch: [176][ 710/ 1200] Overall Loss 0.139200 Objective Loss 0.139200 LR 0.000250 Time 0.020396 -2022-12-06 11:41:28,370 - Epoch: [176][ 720/ 1200] Overall Loss 0.139244 Objective Loss 0.139244 LR 0.000250 Time 0.020379 -2022-12-06 11:41:28,563 - Epoch: [176][ 730/ 1200] Overall Loss 0.139265 Objective Loss 0.139265 LR 0.000250 Time 0.020364 -2022-12-06 11:41:28,755 - Epoch: [176][ 740/ 1200] Overall Loss 0.139130 Objective Loss 0.139130 LR 0.000250 Time 0.020348 -2022-12-06 11:41:28,948 - Epoch: [176][ 750/ 1200] Overall Loss 0.139119 Objective Loss 0.139119 LR 0.000250 Time 0.020333 -2022-12-06 11:41:29,140 - Epoch: [176][ 760/ 1200] Overall Loss 0.139093 Objective Loss 0.139093 LR 0.000250 Time 0.020318 -2022-12-06 11:41:29,333 - Epoch: [176][ 770/ 1200] Overall Loss 0.139321 Objective Loss 0.139321 LR 0.000250 Time 0.020303 -2022-12-06 11:41:29,526 - Epoch: [176][ 780/ 1200] Overall Loss 0.139220 Objective Loss 0.139220 LR 0.000250 Time 0.020290 -2022-12-06 11:41:29,718 - Epoch: [176][ 790/ 1200] Overall Loss 0.139438 Objective Loss 0.139438 LR 0.000250 Time 0.020275 -2022-12-06 11:41:29,911 - Epoch: [176][ 800/ 1200] Overall Loss 0.139529 Objective Loss 0.139529 LR 0.000250 Time 0.020262 -2022-12-06 11:41:30,103 - Epoch: [176][ 810/ 1200] Overall Loss 0.139629 Objective Loss 0.139629 LR 0.000250 Time 0.020249 -2022-12-06 11:41:30,297 - Epoch: [176][ 820/ 1200] Overall Loss 0.139669 Objective Loss 0.139669 LR 0.000250 Time 0.020237 -2022-12-06 11:41:30,489 - Epoch: [176][ 830/ 1200] Overall Loss 0.139935 Objective Loss 0.139935 LR 0.000250 Time 0.020224 -2022-12-06 11:41:30,682 - Epoch: [176][ 840/ 1200] Overall Loss 0.139834 Objective Loss 0.139834 LR 0.000250 Time 0.020212 -2022-12-06 11:41:30,874 - Epoch: [176][ 850/ 1200] Overall Loss 0.139810 Objective Loss 0.139810 LR 0.000250 Time 0.020201 -2022-12-06 11:41:31,067 - Epoch: [176][ 860/ 1200] Overall Loss 0.139569 Objective Loss 0.139569 LR 0.000250 Time 0.020189 -2022-12-06 11:41:31,259 - Epoch: [176][ 870/ 1200] Overall Loss 0.139756 Objective Loss 0.139756 LR 0.000250 Time 0.020177 -2022-12-06 11:41:31,451 - Epoch: [176][ 880/ 1200] Overall Loss 0.139733 Objective Loss 0.139733 LR 0.000250 Time 0.020166 -2022-12-06 11:41:31,644 - Epoch: [176][ 890/ 1200] Overall Loss 0.139921 Objective Loss 0.139921 LR 0.000250 Time 0.020155 -2022-12-06 11:41:31,836 - Epoch: [176][ 900/ 1200] Overall Loss 0.140073 Objective Loss 0.140073 LR 0.000250 Time 0.020144 -2022-12-06 11:41:32,028 - Epoch: [176][ 910/ 1200] Overall Loss 0.140263 Objective Loss 0.140263 LR 0.000250 Time 0.020133 -2022-12-06 11:41:32,221 - Epoch: [176][ 920/ 1200] Overall Loss 0.140638 Objective Loss 0.140638 LR 0.000250 Time 0.020123 -2022-12-06 11:41:32,412 - Epoch: [176][ 930/ 1200] Overall Loss 0.140623 Objective Loss 0.140623 LR 0.000250 Time 0.020112 -2022-12-06 11:41:32,606 - Epoch: [176][ 940/ 1200] Overall Loss 0.140715 Objective Loss 0.140715 LR 0.000250 Time 0.020103 -2022-12-06 11:41:32,798 - Epoch: [176][ 950/ 1200] Overall Loss 0.140685 Objective Loss 0.140685 LR 0.000250 Time 0.020093 -2022-12-06 11:41:32,991 - Epoch: [176][ 960/ 1200] Overall Loss 0.140715 Objective Loss 0.140715 LR 0.000250 Time 0.020084 -2022-12-06 11:41:33,183 - Epoch: [176][ 970/ 1200] Overall Loss 0.140761 Objective Loss 0.140761 LR 0.000250 Time 0.020075 -2022-12-06 11:41:33,375 - Epoch: [176][ 980/ 1200] Overall Loss 0.140778 Objective Loss 0.140778 LR 0.000250 Time 0.020066 -2022-12-06 11:41:33,568 - Epoch: [176][ 990/ 1200] Overall Loss 0.140796 Objective Loss 0.140796 LR 0.000250 Time 0.020057 -2022-12-06 11:41:33,761 - Epoch: [176][ 1000/ 1200] Overall Loss 0.140492 Objective Loss 0.140492 LR 0.000250 Time 0.020049 -2022-12-06 11:41:33,954 - Epoch: [176][ 1010/ 1200] Overall Loss 0.140383 Objective Loss 0.140383 LR 0.000250 Time 0.020041 -2022-12-06 11:41:34,146 - Epoch: [176][ 1020/ 1200] Overall Loss 0.140357 Objective Loss 0.140357 LR 0.000250 Time 0.020033 -2022-12-06 11:41:34,339 - Epoch: [176][ 1030/ 1200] Overall Loss 0.140346 Objective Loss 0.140346 LR 0.000250 Time 0.020025 -2022-12-06 11:41:34,531 - Epoch: [176][ 1040/ 1200] Overall Loss 0.140280 Objective Loss 0.140280 LR 0.000250 Time 0.020017 -2022-12-06 11:41:34,724 - Epoch: [176][ 1050/ 1200] Overall Loss 0.140199 Objective Loss 0.140199 LR 0.000250 Time 0.020008 -2022-12-06 11:41:34,917 - Epoch: [176][ 1060/ 1200] Overall Loss 0.140313 Objective Loss 0.140313 LR 0.000250 Time 0.020001 -2022-12-06 11:41:35,109 - Epoch: [176][ 1070/ 1200] Overall Loss 0.140517 Objective Loss 0.140517 LR 0.000250 Time 0.019993 -2022-12-06 11:41:35,301 - Epoch: [176][ 1080/ 1200] Overall Loss 0.140500 Objective Loss 0.140500 LR 0.000250 Time 0.019986 -2022-12-06 11:41:35,493 - Epoch: [176][ 1090/ 1200] Overall Loss 0.140572 Objective Loss 0.140572 LR 0.000250 Time 0.019978 -2022-12-06 11:41:35,686 - Epoch: [176][ 1100/ 1200] Overall Loss 0.140615 Objective Loss 0.140615 LR 0.000250 Time 0.019971 -2022-12-06 11:41:35,878 - Epoch: [176][ 1110/ 1200] Overall Loss 0.140856 Objective Loss 0.140856 LR 0.000250 Time 0.019964 -2022-12-06 11:41:36,071 - Epoch: [176][ 1120/ 1200] Overall Loss 0.140809 Objective Loss 0.140809 LR 0.000250 Time 0.019957 -2022-12-06 11:41:36,264 - Epoch: [176][ 1130/ 1200] Overall Loss 0.140706 Objective Loss 0.140706 LR 0.000250 Time 0.019951 -2022-12-06 11:41:36,456 - Epoch: [176][ 1140/ 1200] Overall Loss 0.140624 Objective Loss 0.140624 LR 0.000250 Time 0.019944 -2022-12-06 11:41:36,648 - Epoch: [176][ 1150/ 1200] Overall Loss 0.140781 Objective Loss 0.140781 LR 0.000250 Time 0.019937 -2022-12-06 11:41:36,841 - Epoch: [176][ 1160/ 1200] Overall Loss 0.140730 Objective Loss 0.140730 LR 0.000250 Time 0.019931 -2022-12-06 11:41:37,034 - Epoch: [176][ 1170/ 1200] Overall Loss 0.140690 Objective Loss 0.140690 LR 0.000250 Time 0.019925 -2022-12-06 11:41:37,227 - Epoch: [176][ 1180/ 1200] Overall Loss 0.140658 Objective Loss 0.140658 LR 0.000250 Time 0.019919 -2022-12-06 11:41:37,420 - Epoch: [176][ 1190/ 1200] Overall Loss 0.140503 Objective Loss 0.140503 LR 0.000250 Time 0.019913 -2022-12-06 11:41:37,653 - Epoch: [176][ 1200/ 1200] Overall Loss 0.140233 Objective Loss 0.140233 Top1 90.585774 Top5 99.790795 LR 0.000250 Time 0.019942 -2022-12-06 11:41:37,756 - --- validate (epoch=176)----------- -2022-12-06 11:41:37,756 - 34129 samples (256 per mini-batch) -2022-12-06 11:41:38,202 - Epoch: [176][ 10/ 134] Loss 0.247895 Top1 87.968750 Top5 98.007812 -2022-12-06 11:41:38,333 - Epoch: [176][ 20/ 134] Loss 0.231583 Top1 88.281250 Top5 98.300781 -2022-12-06 11:41:38,463 - Epoch: [176][ 30/ 134] Loss 0.228150 Top1 88.255208 Top5 98.515625 -2022-12-06 11:41:38,597 - Epoch: [176][ 40/ 134] Loss 0.226803 Top1 88.515625 Top5 98.554688 -2022-12-06 11:41:38,734 - Epoch: [176][ 50/ 134] Loss 0.227871 Top1 88.554688 Top5 98.585938 -2022-12-06 11:41:38,869 - Epoch: [176][ 60/ 134] Loss 0.231163 Top1 88.313802 Top5 98.580729 -2022-12-06 11:41:39,003 - Epoch: [176][ 70/ 134] Loss 0.231808 Top1 88.320312 Top5 98.565848 -2022-12-06 11:41:39,136 - Epoch: [176][ 80/ 134] Loss 0.233461 Top1 88.369141 Top5 98.583984 -2022-12-06 11:41:39,267 - Epoch: [176][ 90/ 134] Loss 0.236996 Top1 88.307292 Top5 98.641493 -2022-12-06 11:41:39,401 - Epoch: [176][ 100/ 134] Loss 0.238797 Top1 88.269531 Top5 98.617188 -2022-12-06 11:41:39,533 - Epoch: [176][ 110/ 134] Loss 0.236218 Top1 88.299006 Top5 98.604403 -2022-12-06 11:41:39,665 - Epoch: [176][ 120/ 134] Loss 0.236730 Top1 88.333333 Top5 98.597005 -2022-12-06 11:41:39,798 - Epoch: [176][ 130/ 134] Loss 0.235073 Top1 88.335337 Top5 98.593750 -2022-12-06 11:41:39,838 - Epoch: [176][ 134/ 134] Loss 0.235640 Top1 88.294412 Top5 98.593571 -2022-12-06 11:41:39,925 - ==> Top1: 88.294 Top5: 98.594 Loss: 0.236 - -2022-12-06 11:41:39,926 - ==> Confusion: -[[ 911 0 1 3 4 9 0 3 4 41 0 1 1 2 5 2 1 0 2 0 6] - [ 2 948 1 2 10 20 1 10 0 1 3 3 0 1 0 2 4 0 8 3 8] - [ 4 1 1018 12 5 4 11 8 0 2 6 4 3 1 1 2 1 1 2 5 12] - [ 2 1 16 960 1 3 0 0 1 1 8 1 1 1 7 1 0 1 9 0 6] - [ 8 5 1 0 962 3 1 2 3 6 1 3 0 3 5 4 5 1 1 1 5] - [ 1 12 1 2 6 1001 2 14 3 1 1 9 2 9 1 0 0 2 0 0 2] - [ 1 2 12 4 1 3 1075 3 0 0 0 1 1 1 0 4 0 0 2 7 1] - [ 1 10 3 3 1 34 5 953 0 1 1 4 2 1 1 0 1 0 13 12 8] - [ 4 4 0 0 0 2 1 1 975 39 8 1 1 7 15 2 1 0 2 1 0] - [ 57 0 1 0 7 6 0 3 21 887 2 2 0 6 3 0 0 1 1 0 4] - [ 1 1 3 6 1 3 2 1 10 3 965 0 0 7 4 1 0 0 2 3 6] - [ 1 3 2 0 1 9 2 4 0 0 0 984 21 2 0 4 5 6 0 4 3] - [ 0 1 1 0 1 2 0 0 0 1 0 20 915 0 3 7 1 5 0 4 8] - [ 2 0 0 0 0 7 0 2 13 10 4 6 3 961 0 2 2 0 1 3 7] - [ 6 2 1 13 2 4 0 0 10 4 0 2 1 3 1066 1 0 1 4 1 9] - [ 0 0 1 0 4 1 5 1 0 0 1 11 6 2 0 988 6 9 1 3 4] - [ 0 0 1 1 2 2 1 0 1 0 1 2 2 1 0 10 1035 2 0 6 5] - [ 2 0 2 3 2 2 1 1 0 5 1 9 19 1 3 13 0 966 0 1 5] - [ 3 4 5 14 0 4 0 25 1 1 4 4 1 0 6 0 0 1 929 2 4] - [ 1 4 1 3 0 6 3 6 0 0 1 22 8 5 0 1 4 1 1 1009 4] - [ 111 197 145 111 91 179 62 126 68 75 131 85 277 224 115 91 142 64 117 192 10623]] - -2022-12-06 11:41:40,507 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:41:40,508 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:41:40,514 - - -2022-12-06 11:41:40,514 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:41:41,559 - Epoch: [177][ 10/ 1200] Overall Loss 0.128578 Objective Loss 0.128578 LR 0.000250 Time 0.104440 -2022-12-06 11:41:41,757 - Epoch: [177][ 20/ 1200] Overall Loss 0.152292 Objective Loss 0.152292 LR 0.000250 Time 0.062109 -2022-12-06 11:41:41,957 - Epoch: [177][ 30/ 1200] Overall Loss 0.150731 Objective Loss 0.150731 LR 0.000250 Time 0.048062 -2022-12-06 11:41:42,154 - Epoch: [177][ 40/ 1200] Overall Loss 0.149321 Objective Loss 0.149321 LR 0.000250 Time 0.040949 -2022-12-06 11:41:42,353 - Epoch: [177][ 50/ 1200] Overall Loss 0.152113 Objective Loss 0.152113 LR 0.000250 Time 0.036731 -2022-12-06 11:41:42,549 - Epoch: [177][ 60/ 1200] Overall Loss 0.151818 Objective Loss 0.151818 LR 0.000250 Time 0.033855 -2022-12-06 11:41:42,748 - Epoch: [177][ 70/ 1200] Overall Loss 0.150726 Objective Loss 0.150726 LR 0.000250 Time 0.031866 -2022-12-06 11:41:42,944 - Epoch: [177][ 80/ 1200] Overall Loss 0.148015 Objective Loss 0.148015 LR 0.000250 Time 0.030315 -2022-12-06 11:41:43,143 - Epoch: [177][ 90/ 1200] Overall Loss 0.145757 Objective Loss 0.145757 LR 0.000250 Time 0.029154 -2022-12-06 11:41:43,339 - Epoch: [177][ 100/ 1200] Overall Loss 0.145241 Objective Loss 0.145241 LR 0.000250 Time 0.028192 -2022-12-06 11:41:43,538 - Epoch: [177][ 110/ 1200] Overall Loss 0.146687 Objective Loss 0.146687 LR 0.000250 Time 0.027434 -2022-12-06 11:41:43,734 - Epoch: [177][ 120/ 1200] Overall Loss 0.146087 Objective Loss 0.146087 LR 0.000250 Time 0.026777 -2022-12-06 11:41:43,933 - Epoch: [177][ 130/ 1200] Overall Loss 0.146108 Objective Loss 0.146108 LR 0.000250 Time 0.026247 -2022-12-06 11:41:44,129 - Epoch: [177][ 140/ 1200] Overall Loss 0.145137 Objective Loss 0.145137 LR 0.000250 Time 0.025766 -2022-12-06 11:41:44,328 - Epoch: [177][ 150/ 1200] Overall Loss 0.144282 Objective Loss 0.144282 LR 0.000250 Time 0.025373 -2022-12-06 11:41:44,523 - Epoch: [177][ 160/ 1200] Overall Loss 0.144967 Objective Loss 0.144967 LR 0.000250 Time 0.025003 -2022-12-06 11:41:44,722 - Epoch: [177][ 170/ 1200] Overall Loss 0.145240 Objective Loss 0.145240 LR 0.000250 Time 0.024700 -2022-12-06 11:41:44,918 - Epoch: [177][ 180/ 1200] Overall Loss 0.143720 Objective Loss 0.143720 LR 0.000250 Time 0.024410 -2022-12-06 11:41:45,117 - Epoch: [177][ 190/ 1200] Overall Loss 0.143168 Objective Loss 0.143168 LR 0.000250 Time 0.024172 -2022-12-06 11:41:45,313 - Epoch: [177][ 200/ 1200] Overall Loss 0.143257 Objective Loss 0.143257 LR 0.000250 Time 0.023938 -2022-12-06 11:41:45,512 - Epoch: [177][ 210/ 1200] Overall Loss 0.142872 Objective Loss 0.142872 LR 0.000250 Time 0.023744 -2022-12-06 11:41:45,708 - Epoch: [177][ 220/ 1200] Overall Loss 0.142504 Objective Loss 0.142504 LR 0.000250 Time 0.023553 -2022-12-06 11:41:45,907 - Epoch: [177][ 230/ 1200] Overall Loss 0.142158 Objective Loss 0.142158 LR 0.000250 Time 0.023392 -2022-12-06 11:41:46,102 - Epoch: [177][ 240/ 1200] Overall Loss 0.142419 Objective Loss 0.142419 LR 0.000250 Time 0.023229 -2022-12-06 11:41:46,301 - Epoch: [177][ 250/ 1200] Overall Loss 0.142983 Objective Loss 0.142983 LR 0.000250 Time 0.023094 -2022-12-06 11:41:46,497 - Epoch: [177][ 260/ 1200] Overall Loss 0.142713 Objective Loss 0.142713 LR 0.000250 Time 0.022956 -2022-12-06 11:41:46,696 - Epoch: [177][ 270/ 1200] Overall Loss 0.142931 Objective Loss 0.142931 LR 0.000250 Time 0.022839 -2022-12-06 11:41:46,891 - Epoch: [177][ 280/ 1200] Overall Loss 0.142832 Objective Loss 0.142832 LR 0.000250 Time 0.022718 -2022-12-06 11:41:47,089 - Epoch: [177][ 290/ 1200] Overall Loss 0.142688 Objective Loss 0.142688 LR 0.000250 Time 0.022617 -2022-12-06 11:41:47,284 - Epoch: [177][ 300/ 1200] Overall Loss 0.142554 Objective Loss 0.142554 LR 0.000250 Time 0.022512 -2022-12-06 11:41:47,483 - Epoch: [177][ 310/ 1200] Overall Loss 0.143159 Objective Loss 0.143159 LR 0.000250 Time 0.022426 -2022-12-06 11:41:47,679 - Epoch: [177][ 320/ 1200] Overall Loss 0.142650 Objective Loss 0.142650 LR 0.000250 Time 0.022335 -2022-12-06 11:41:47,878 - Epoch: [177][ 330/ 1200] Overall Loss 0.142640 Objective Loss 0.142640 LR 0.000250 Time 0.022258 -2022-12-06 11:41:48,073 - Epoch: [177][ 340/ 1200] Overall Loss 0.143326 Objective Loss 0.143326 LR 0.000250 Time 0.022178 -2022-12-06 11:41:48,272 - Epoch: [177][ 350/ 1200] Overall Loss 0.143660 Objective Loss 0.143660 LR 0.000250 Time 0.022110 -2022-12-06 11:41:48,468 - Epoch: [177][ 360/ 1200] Overall Loss 0.143215 Objective Loss 0.143215 LR 0.000250 Time 0.022038 -2022-12-06 11:41:48,667 - Epoch: [177][ 370/ 1200] Overall Loss 0.143004 Objective Loss 0.143004 LR 0.000250 Time 0.021979 -2022-12-06 11:41:48,862 - Epoch: [177][ 380/ 1200] Overall Loss 0.142536 Objective Loss 0.142536 LR 0.000250 Time 0.021913 -2022-12-06 11:41:49,061 - Epoch: [177][ 390/ 1200] Overall Loss 0.142570 Objective Loss 0.142570 LR 0.000250 Time 0.021860 -2022-12-06 11:41:49,257 - Epoch: [177][ 400/ 1200] Overall Loss 0.142681 Objective Loss 0.142681 LR 0.000250 Time 0.021802 -2022-12-06 11:41:49,456 - Epoch: [177][ 410/ 1200] Overall Loss 0.142971 Objective Loss 0.142971 LR 0.000250 Time 0.021753 -2022-12-06 11:41:49,652 - Epoch: [177][ 420/ 1200] Overall Loss 0.143393 Objective Loss 0.143393 LR 0.000250 Time 0.021701 -2022-12-06 11:41:49,850 - Epoch: [177][ 430/ 1200] Overall Loss 0.142974 Objective Loss 0.142974 LR 0.000250 Time 0.021657 -2022-12-06 11:41:50,046 - Epoch: [177][ 440/ 1200] Overall Loss 0.142730 Objective Loss 0.142730 LR 0.000250 Time 0.021607 -2022-12-06 11:41:50,245 - Epoch: [177][ 450/ 1200] Overall Loss 0.142612 Objective Loss 0.142612 LR 0.000250 Time 0.021568 -2022-12-06 11:41:50,440 - Epoch: [177][ 460/ 1200] Overall Loss 0.142340 Objective Loss 0.142340 LR 0.000250 Time 0.021523 -2022-12-06 11:41:50,639 - Epoch: [177][ 470/ 1200] Overall Loss 0.142290 Objective Loss 0.142290 LR 0.000250 Time 0.021487 -2022-12-06 11:41:50,834 - Epoch: [177][ 480/ 1200] Overall Loss 0.142329 Objective Loss 0.142329 LR 0.000250 Time 0.021445 -2022-12-06 11:41:51,033 - Epoch: [177][ 490/ 1200] Overall Loss 0.142139 Objective Loss 0.142139 LR 0.000250 Time 0.021413 -2022-12-06 11:41:51,229 - Epoch: [177][ 500/ 1200] Overall Loss 0.141947 Objective Loss 0.141947 LR 0.000250 Time 0.021374 -2022-12-06 11:41:51,428 - Epoch: [177][ 510/ 1200] Overall Loss 0.142063 Objective Loss 0.142063 LR 0.000250 Time 0.021344 -2022-12-06 11:41:51,624 - Epoch: [177][ 520/ 1200] Overall Loss 0.141909 Objective Loss 0.141909 LR 0.000250 Time 0.021310 -2022-12-06 11:41:51,823 - Epoch: [177][ 530/ 1200] Overall Loss 0.141807 Objective Loss 0.141807 LR 0.000250 Time 0.021282 -2022-12-06 11:41:52,019 - Epoch: [177][ 540/ 1200] Overall Loss 0.142007 Objective Loss 0.142007 LR 0.000250 Time 0.021250 -2022-12-06 11:41:52,218 - Epoch: [177][ 550/ 1200] Overall Loss 0.141988 Objective Loss 0.141988 LR 0.000250 Time 0.021224 -2022-12-06 11:41:52,414 - Epoch: [177][ 560/ 1200] Overall Loss 0.141847 Objective Loss 0.141847 LR 0.000250 Time 0.021194 -2022-12-06 11:41:52,613 - Epoch: [177][ 570/ 1200] Overall Loss 0.141930 Objective Loss 0.141930 LR 0.000250 Time 0.021170 -2022-12-06 11:41:52,808 - Epoch: [177][ 580/ 1200] Overall Loss 0.141927 Objective Loss 0.141927 LR 0.000250 Time 0.021141 -2022-12-06 11:41:53,007 - Epoch: [177][ 590/ 1200] Overall Loss 0.141795 Objective Loss 0.141795 LR 0.000250 Time 0.021119 -2022-12-06 11:41:53,204 - Epoch: [177][ 600/ 1200] Overall Loss 0.141909 Objective Loss 0.141909 LR 0.000250 Time 0.021094 -2022-12-06 11:41:53,403 - Epoch: [177][ 610/ 1200] Overall Loss 0.141733 Objective Loss 0.141733 LR 0.000250 Time 0.021074 -2022-12-06 11:41:53,599 - Epoch: [177][ 620/ 1200] Overall Loss 0.141317 Objective Loss 0.141317 LR 0.000250 Time 0.021050 -2022-12-06 11:41:53,798 - Epoch: [177][ 630/ 1200] Overall Loss 0.141826 Objective Loss 0.141826 LR 0.000250 Time 0.021031 -2022-12-06 11:41:53,994 - Epoch: [177][ 640/ 1200] Overall Loss 0.141571 Objective Loss 0.141571 LR 0.000250 Time 0.021007 -2022-12-06 11:41:54,193 - Epoch: [177][ 650/ 1200] Overall Loss 0.141968 Objective Loss 0.141968 LR 0.000250 Time 0.020989 -2022-12-06 11:41:54,388 - Epoch: [177][ 660/ 1200] Overall Loss 0.142041 Objective Loss 0.142041 LR 0.000250 Time 0.020966 -2022-12-06 11:41:54,587 - Epoch: [177][ 670/ 1200] Overall Loss 0.142097 Objective Loss 0.142097 LR 0.000250 Time 0.020949 -2022-12-06 11:41:54,782 - Epoch: [177][ 680/ 1200] Overall Loss 0.141913 Objective Loss 0.141913 LR 0.000250 Time 0.020927 -2022-12-06 11:41:54,981 - Epoch: [177][ 690/ 1200] Overall Loss 0.141735 Objective Loss 0.141735 LR 0.000250 Time 0.020911 -2022-12-06 11:41:55,177 - Epoch: [177][ 700/ 1200] Overall Loss 0.141592 Objective Loss 0.141592 LR 0.000250 Time 0.020891 -2022-12-06 11:41:55,376 - Epoch: [177][ 710/ 1200] Overall Loss 0.141469 Objective Loss 0.141469 LR 0.000250 Time 0.020877 -2022-12-06 11:41:55,571 - Epoch: [177][ 720/ 1200] Overall Loss 0.141237 Objective Loss 0.141237 LR 0.000250 Time 0.020857 -2022-12-06 11:41:55,770 - Epoch: [177][ 730/ 1200] Overall Loss 0.141024 Objective Loss 0.141024 LR 0.000250 Time 0.020843 -2022-12-06 11:41:55,965 - Epoch: [177][ 740/ 1200] Overall Loss 0.141118 Objective Loss 0.141118 LR 0.000250 Time 0.020824 -2022-12-06 11:41:56,164 - Epoch: [177][ 750/ 1200] Overall Loss 0.141154 Objective Loss 0.141154 LR 0.000250 Time 0.020811 -2022-12-06 11:41:56,359 - Epoch: [177][ 760/ 1200] Overall Loss 0.141193 Objective Loss 0.141193 LR 0.000250 Time 0.020793 -2022-12-06 11:41:56,558 - Epoch: [177][ 770/ 1200] Overall Loss 0.141039 Objective Loss 0.141039 LR 0.000250 Time 0.020781 -2022-12-06 11:41:56,753 - Epoch: [177][ 780/ 1200] Overall Loss 0.141333 Objective Loss 0.141333 LR 0.000250 Time 0.020764 -2022-12-06 11:41:56,952 - Epoch: [177][ 790/ 1200] Overall Loss 0.141290 Objective Loss 0.141290 LR 0.000250 Time 0.020752 -2022-12-06 11:41:57,148 - Epoch: [177][ 800/ 1200] Overall Loss 0.141383 Objective Loss 0.141383 LR 0.000250 Time 0.020737 -2022-12-06 11:41:57,347 - Epoch: [177][ 810/ 1200] Overall Loss 0.141463 Objective Loss 0.141463 LR 0.000250 Time 0.020726 -2022-12-06 11:41:57,543 - Epoch: [177][ 820/ 1200] Overall Loss 0.141522 Objective Loss 0.141522 LR 0.000250 Time 0.020711 -2022-12-06 11:41:57,742 - Epoch: [177][ 830/ 1200] Overall Loss 0.141764 Objective Loss 0.141764 LR 0.000250 Time 0.020701 -2022-12-06 11:41:57,937 - Epoch: [177][ 840/ 1200] Overall Loss 0.141725 Objective Loss 0.141725 LR 0.000250 Time 0.020686 -2022-12-06 11:41:58,136 - Epoch: [177][ 850/ 1200] Overall Loss 0.141562 Objective Loss 0.141562 LR 0.000250 Time 0.020676 -2022-12-06 11:41:58,332 - Epoch: [177][ 860/ 1200] Overall Loss 0.141671 Objective Loss 0.141671 LR 0.000250 Time 0.020663 -2022-12-06 11:41:58,530 - Epoch: [177][ 870/ 1200] Overall Loss 0.141529 Objective Loss 0.141529 LR 0.000250 Time 0.020653 -2022-12-06 11:41:58,726 - Epoch: [177][ 880/ 1200] Overall Loss 0.141667 Objective Loss 0.141667 LR 0.000250 Time 0.020640 -2022-12-06 11:41:58,925 - Epoch: [177][ 890/ 1200] Overall Loss 0.141700 Objective Loss 0.141700 LR 0.000250 Time 0.020631 -2022-12-06 11:41:59,122 - Epoch: [177][ 900/ 1200] Overall Loss 0.141867 Objective Loss 0.141867 LR 0.000250 Time 0.020620 -2022-12-06 11:41:59,321 - Epoch: [177][ 910/ 1200] Overall Loss 0.141929 Objective Loss 0.141929 LR 0.000250 Time 0.020611 -2022-12-06 11:41:59,517 - Epoch: [177][ 920/ 1200] Overall Loss 0.141768 Objective Loss 0.141768 LR 0.000250 Time 0.020600 -2022-12-06 11:41:59,717 - Epoch: [177][ 930/ 1200] Overall Loss 0.141885 Objective Loss 0.141885 LR 0.000250 Time 0.020593 -2022-12-06 11:41:59,916 - Epoch: [177][ 940/ 1200] Overall Loss 0.142050 Objective Loss 0.142050 LR 0.000250 Time 0.020585 -2022-12-06 11:42:00,117 - Epoch: [177][ 950/ 1200] Overall Loss 0.142321 Objective Loss 0.142321 LR 0.000250 Time 0.020579 -2022-12-06 11:42:00,315 - Epoch: [177][ 960/ 1200] Overall Loss 0.142469 Objective Loss 0.142469 LR 0.000250 Time 0.020571 -2022-12-06 11:42:00,517 - Epoch: [177][ 970/ 1200] Overall Loss 0.142349 Objective Loss 0.142349 LR 0.000250 Time 0.020566 -2022-12-06 11:42:00,716 - Epoch: [177][ 980/ 1200] Overall Loss 0.142393 Objective Loss 0.142393 LR 0.000250 Time 0.020559 -2022-12-06 11:42:00,917 - Epoch: [177][ 990/ 1200] Overall Loss 0.142456 Objective Loss 0.142456 LR 0.000250 Time 0.020553 -2022-12-06 11:42:01,115 - Epoch: [177][ 1000/ 1200] Overall Loss 0.142494 Objective Loss 0.142494 LR 0.000250 Time 0.020546 -2022-12-06 11:42:01,317 - Epoch: [177][ 1010/ 1200] Overall Loss 0.142479 Objective Loss 0.142479 LR 0.000250 Time 0.020541 -2022-12-06 11:42:01,515 - Epoch: [177][ 1020/ 1200] Overall Loss 0.142723 Objective Loss 0.142723 LR 0.000250 Time 0.020533 -2022-12-06 11:42:01,716 - Epoch: [177][ 1030/ 1200] Overall Loss 0.142571 Objective Loss 0.142571 LR 0.000250 Time 0.020529 -2022-12-06 11:42:01,914 - Epoch: [177][ 1040/ 1200] Overall Loss 0.142317 Objective Loss 0.142317 LR 0.000250 Time 0.020521 -2022-12-06 11:42:02,115 - Epoch: [177][ 1050/ 1200] Overall Loss 0.142483 Objective Loss 0.142483 LR 0.000250 Time 0.020517 -2022-12-06 11:42:02,313 - Epoch: [177][ 1060/ 1200] Overall Loss 0.142596 Objective Loss 0.142596 LR 0.000250 Time 0.020510 -2022-12-06 11:42:02,514 - Epoch: [177][ 1070/ 1200] Overall Loss 0.142516 Objective Loss 0.142516 LR 0.000250 Time 0.020505 -2022-12-06 11:42:02,712 - Epoch: [177][ 1080/ 1200] Overall Loss 0.142456 Objective Loss 0.142456 LR 0.000250 Time 0.020498 -2022-12-06 11:42:02,913 - Epoch: [177][ 1090/ 1200] Overall Loss 0.142514 Objective Loss 0.142514 LR 0.000250 Time 0.020495 -2022-12-06 11:42:03,112 - Epoch: [177][ 1100/ 1200] Overall Loss 0.142474 Objective Loss 0.142474 LR 0.000250 Time 0.020489 -2022-12-06 11:42:03,314 - Epoch: [177][ 1110/ 1200] Overall Loss 0.142340 Objective Loss 0.142340 LR 0.000250 Time 0.020485 -2022-12-06 11:42:03,512 - Epoch: [177][ 1120/ 1200] Overall Loss 0.142336 Objective Loss 0.142336 LR 0.000250 Time 0.020478 -2022-12-06 11:42:03,713 - Epoch: [177][ 1130/ 1200] Overall Loss 0.142302 Objective Loss 0.142302 LR 0.000250 Time 0.020475 -2022-12-06 11:42:03,912 - Epoch: [177][ 1140/ 1200] Overall Loss 0.142280 Objective Loss 0.142280 LR 0.000250 Time 0.020469 -2022-12-06 11:42:04,113 - Epoch: [177][ 1150/ 1200] Overall Loss 0.142347 Objective Loss 0.142347 LR 0.000250 Time 0.020465 -2022-12-06 11:42:04,311 - Epoch: [177][ 1160/ 1200] Overall Loss 0.142473 Objective Loss 0.142473 LR 0.000250 Time 0.020460 -2022-12-06 11:42:04,512 - Epoch: [177][ 1170/ 1200] Overall Loss 0.142351 Objective Loss 0.142351 LR 0.000250 Time 0.020456 -2022-12-06 11:42:04,710 - Epoch: [177][ 1180/ 1200] Overall Loss 0.142261 Objective Loss 0.142261 LR 0.000250 Time 0.020450 -2022-12-06 11:42:04,911 - Epoch: [177][ 1190/ 1200] Overall Loss 0.142238 Objective Loss 0.142238 LR 0.000250 Time 0.020446 -2022-12-06 11:42:05,144 - Epoch: [177][ 1200/ 1200] Overall Loss 0.142314 Objective Loss 0.142314 Top1 91.213389 Top5 98.535565 LR 0.000250 Time 0.020470 -2022-12-06 11:42:05,232 - --- validate (epoch=177)----------- -2022-12-06 11:42:05,232 - 34129 samples (256 per mini-batch) -2022-12-06 11:42:05,685 - Epoch: [177][ 10/ 134] Loss 0.217252 Top1 88.125000 Top5 98.437500 -2022-12-06 11:42:05,820 - Epoch: [177][ 20/ 134] Loss 0.224316 Top1 88.359375 Top5 98.417969 -2022-12-06 11:42:05,953 - Epoch: [177][ 30/ 134] Loss 0.225596 Top1 88.463542 Top5 98.528646 -2022-12-06 11:42:06,088 - Epoch: [177][ 40/ 134] Loss 0.225223 Top1 88.515625 Top5 98.662109 -2022-12-06 11:42:06,220 - Epoch: [177][ 50/ 134] Loss 0.230134 Top1 88.468750 Top5 98.625000 -2022-12-06 11:42:06,348 - Epoch: [177][ 60/ 134] Loss 0.226892 Top1 88.404948 Top5 98.704427 -2022-12-06 11:42:06,475 - Epoch: [177][ 70/ 134] Loss 0.223852 Top1 88.487723 Top5 98.643973 -2022-12-06 11:42:06,602 - Epoch: [177][ 80/ 134] Loss 0.234751 Top1 88.178711 Top5 98.559570 -2022-12-06 11:42:06,728 - Epoch: [177][ 90/ 134] Loss 0.234810 Top1 88.177083 Top5 98.576389 -2022-12-06 11:42:06,856 - Epoch: [177][ 100/ 134] Loss 0.235004 Top1 88.265625 Top5 98.570312 -2022-12-06 11:42:06,985 - Epoch: [177][ 110/ 134] Loss 0.236649 Top1 88.252841 Top5 98.561790 -2022-12-06 11:42:07,114 - Epoch: [177][ 120/ 134] Loss 0.234572 Top1 88.310547 Top5 98.544922 -2022-12-06 11:42:07,243 - Epoch: [177][ 130/ 134] Loss 0.234582 Top1 88.209135 Top5 98.551683 -2022-12-06 11:42:07,280 - Epoch: [177][ 134/ 134] Loss 0.233728 Top1 88.218231 Top5 98.552551 -2022-12-06 11:42:07,367 - ==> Top1: 88.218 Top5: 98.553 Loss: 0.234 - -2022-12-06 11:42:07,368 - ==> Confusion: -[[ 929 1 2 0 2 5 0 2 5 38 0 1 1 2 2 2 0 0 3 0 1] - [ 1 943 0 2 6 23 1 14 2 0 2 3 1 1 0 1 3 1 14 3 6] - [ 6 3 1020 14 4 3 9 8 0 1 3 3 2 1 2 1 1 2 4 5 11] - [ 2 2 15 957 0 1 1 0 1 1 7 1 4 1 7 0 1 2 11 0 6] - [ 7 6 1 0 954 4 1 2 1 7 1 3 1 2 7 4 9 3 1 0 6] - [ 1 10 0 2 2 997 2 17 1 1 1 10 4 14 1 2 0 0 0 2 2] - [ 2 2 17 2 1 2 1069 3 0 0 1 2 1 1 0 4 0 1 1 5 4] - [ 0 9 3 2 2 20 5 973 2 0 2 6 1 0 0 0 1 0 14 9 5] - [ 6 3 0 0 0 1 1 2 987 33 8 1 2 3 8 0 2 0 3 1 3] - [ 66 1 1 0 3 1 0 1 25 878 2 1 0 8 4 1 0 2 1 0 6] - [ 0 3 5 7 1 2 0 3 9 0 954 2 1 9 5 1 0 0 10 1 6] - [ 4 0 2 0 0 8 2 1 1 0 1 991 20 3 0 3 2 4 0 5 4] - [ 0 1 0 2 0 1 0 0 0 0 1 19 920 3 1 5 1 9 0 2 4] - [ 2 1 0 0 2 8 0 2 13 7 4 6 4 957 0 2 1 2 0 2 10] - [ 5 4 1 11 1 3 0 0 16 0 0 2 4 3 1060 0 0 1 10 1 8] - [ 1 0 2 0 1 0 4 1 1 0 1 6 4 3 0 991 5 15 1 1 6] - [ 0 2 0 2 1 1 1 0 2 1 1 2 4 1 0 6 1037 1 1 4 5] - [ 3 0 1 1 1 1 0 0 0 1 0 6 17 0 1 10 2 989 0 0 3] - [ 2 2 4 6 2 2 0 18 2 1 4 3 2 0 5 0 0 0 950 2 3] - [ 1 3 0 2 0 5 5 5 0 1 3 14 8 6 0 4 3 2 1 1011 6] - [ 116 199 136 102 77 169 63 153 81 59 130 82 319 243 110 86 116 84 175 189 10537]] - -2022-12-06 11:42:07,940 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:42:07,941 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:42:07,947 - - -2022-12-06 11:42:07,947 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:42:08,880 - Epoch: [178][ 10/ 1200] Overall Loss 0.132134 Objective Loss 0.132134 LR 0.000250 Time 0.093282 -2022-12-06 11:42:09,079 - Epoch: [178][ 20/ 1200] Overall Loss 0.134530 Objective Loss 0.134530 LR 0.000250 Time 0.056559 -2022-12-06 11:42:09,278 - Epoch: [178][ 30/ 1200] Overall Loss 0.130233 Objective Loss 0.130233 LR 0.000250 Time 0.044303 -2022-12-06 11:42:09,474 - Epoch: [178][ 40/ 1200] Overall Loss 0.127991 Objective Loss 0.127991 LR 0.000250 Time 0.038126 -2022-12-06 11:42:09,673 - Epoch: [178][ 50/ 1200] Overall Loss 0.131032 Objective Loss 0.131032 LR 0.000250 Time 0.034463 -2022-12-06 11:42:09,869 - Epoch: [178][ 60/ 1200] Overall Loss 0.133435 Objective Loss 0.133435 LR 0.000250 Time 0.031973 -2022-12-06 11:42:10,067 - Epoch: [178][ 70/ 1200] Overall Loss 0.134106 Objective Loss 0.134106 LR 0.000250 Time 0.030224 -2022-12-06 11:42:10,263 - Epoch: [178][ 80/ 1200] Overall Loss 0.135922 Objective Loss 0.135922 LR 0.000250 Time 0.028890 -2022-12-06 11:42:10,461 - Epoch: [178][ 90/ 1200] Overall Loss 0.134989 Objective Loss 0.134989 LR 0.000250 Time 0.027874 -2022-12-06 11:42:10,656 - Epoch: [178][ 100/ 1200] Overall Loss 0.135780 Objective Loss 0.135780 LR 0.000250 Time 0.027036 -2022-12-06 11:42:10,854 - Epoch: [178][ 110/ 1200] Overall Loss 0.136450 Objective Loss 0.136450 LR 0.000250 Time 0.026373 -2022-12-06 11:42:11,050 - Epoch: [178][ 120/ 1200] Overall Loss 0.136277 Objective Loss 0.136277 LR 0.000250 Time 0.025804 -2022-12-06 11:42:11,248 - Epoch: [178][ 130/ 1200] Overall Loss 0.138060 Objective Loss 0.138060 LR 0.000250 Time 0.025337 -2022-12-06 11:42:11,444 - Epoch: [178][ 140/ 1200] Overall Loss 0.138526 Objective Loss 0.138526 LR 0.000250 Time 0.024921 -2022-12-06 11:42:11,641 - Epoch: [178][ 150/ 1200] Overall Loss 0.138006 Objective Loss 0.138006 LR 0.000250 Time 0.024574 -2022-12-06 11:42:11,837 - Epoch: [178][ 160/ 1200] Overall Loss 0.137390 Objective Loss 0.137390 LR 0.000250 Time 0.024258 -2022-12-06 11:42:12,035 - Epoch: [178][ 170/ 1200] Overall Loss 0.136916 Objective Loss 0.136916 LR 0.000250 Time 0.023990 -2022-12-06 11:42:12,230 - Epoch: [178][ 180/ 1200] Overall Loss 0.136981 Objective Loss 0.136981 LR 0.000250 Time 0.023742 -2022-12-06 11:42:12,429 - Epoch: [178][ 190/ 1200] Overall Loss 0.136495 Objective Loss 0.136495 LR 0.000250 Time 0.023536 -2022-12-06 11:42:12,625 - Epoch: [178][ 200/ 1200] Overall Loss 0.137031 Objective Loss 0.137031 LR 0.000250 Time 0.023336 -2022-12-06 11:42:12,824 - Epoch: [178][ 210/ 1200] Overall Loss 0.137134 Objective Loss 0.137134 LR 0.000250 Time 0.023168 -2022-12-06 11:42:13,019 - Epoch: [178][ 220/ 1200] Overall Loss 0.137762 Objective Loss 0.137762 LR 0.000250 Time 0.023002 -2022-12-06 11:42:13,218 - Epoch: [178][ 230/ 1200] Overall Loss 0.137265 Objective Loss 0.137265 LR 0.000250 Time 0.022862 -2022-12-06 11:42:13,414 - Epoch: [178][ 240/ 1200] Overall Loss 0.137495 Objective Loss 0.137495 LR 0.000250 Time 0.022725 -2022-12-06 11:42:13,612 - Epoch: [178][ 250/ 1200] Overall Loss 0.137742 Objective Loss 0.137742 LR 0.000250 Time 0.022606 -2022-12-06 11:42:13,808 - Epoch: [178][ 260/ 1200] Overall Loss 0.138317 Objective Loss 0.138317 LR 0.000250 Time 0.022487 -2022-12-06 11:42:14,007 - Epoch: [178][ 270/ 1200] Overall Loss 0.138443 Objective Loss 0.138443 LR 0.000250 Time 0.022388 -2022-12-06 11:42:14,202 - Epoch: [178][ 280/ 1200] Overall Loss 0.138360 Objective Loss 0.138360 LR 0.000250 Time 0.022285 -2022-12-06 11:42:14,400 - Epoch: [178][ 290/ 1200] Overall Loss 0.137921 Objective Loss 0.137921 LR 0.000250 Time 0.022196 -2022-12-06 11:42:14,597 - Epoch: [178][ 300/ 1200] Overall Loss 0.137315 Objective Loss 0.137315 LR 0.000250 Time 0.022110 -2022-12-06 11:42:14,795 - Epoch: [178][ 310/ 1200] Overall Loss 0.137637 Objective Loss 0.137637 LR 0.000250 Time 0.022035 -2022-12-06 11:42:14,991 - Epoch: [178][ 320/ 1200] Overall Loss 0.137780 Objective Loss 0.137780 LR 0.000250 Time 0.021956 -2022-12-06 11:42:15,189 - Epoch: [178][ 330/ 1200] Overall Loss 0.138173 Objective Loss 0.138173 LR 0.000250 Time 0.021889 -2022-12-06 11:42:15,385 - Epoch: [178][ 340/ 1200] Overall Loss 0.137972 Objective Loss 0.137972 LR 0.000250 Time 0.021820 -2022-12-06 11:42:15,583 - Epoch: [178][ 350/ 1200] Overall Loss 0.137299 Objective Loss 0.137299 LR 0.000250 Time 0.021761 -2022-12-06 11:42:15,778 - Epoch: [178][ 360/ 1200] Overall Loss 0.137575 Objective Loss 0.137575 LR 0.000250 Time 0.021699 -2022-12-06 11:42:15,977 - Epoch: [178][ 370/ 1200] Overall Loss 0.138328 Objective Loss 0.138328 LR 0.000250 Time 0.021648 -2022-12-06 11:42:16,174 - Epoch: [178][ 380/ 1200] Overall Loss 0.138386 Objective Loss 0.138386 LR 0.000250 Time 0.021594 -2022-12-06 11:42:16,372 - Epoch: [178][ 390/ 1200] Overall Loss 0.138670 Objective Loss 0.138670 LR 0.000250 Time 0.021547 -2022-12-06 11:42:16,568 - Epoch: [178][ 400/ 1200] Overall Loss 0.138497 Objective Loss 0.138497 LR 0.000250 Time 0.021497 -2022-12-06 11:42:16,766 - Epoch: [178][ 410/ 1200] Overall Loss 0.138954 Objective Loss 0.138954 LR 0.000250 Time 0.021455 -2022-12-06 11:42:16,962 - Epoch: [178][ 420/ 1200] Overall Loss 0.139039 Objective Loss 0.139039 LR 0.000250 Time 0.021410 -2022-12-06 11:42:17,160 - Epoch: [178][ 430/ 1200] Overall Loss 0.139087 Objective Loss 0.139087 LR 0.000250 Time 0.021371 -2022-12-06 11:42:17,356 - Epoch: [178][ 440/ 1200] Overall Loss 0.139678 Objective Loss 0.139678 LR 0.000250 Time 0.021328 -2022-12-06 11:42:17,554 - Epoch: [178][ 450/ 1200] Overall Loss 0.139684 Objective Loss 0.139684 LR 0.000250 Time 0.021294 -2022-12-06 11:42:17,750 - Epoch: [178][ 460/ 1200] Overall Loss 0.139563 Objective Loss 0.139563 LR 0.000250 Time 0.021257 -2022-12-06 11:42:17,949 - Epoch: [178][ 470/ 1200] Overall Loss 0.139166 Objective Loss 0.139166 LR 0.000250 Time 0.021225 -2022-12-06 11:42:18,144 - Epoch: [178][ 480/ 1200] Overall Loss 0.138916 Objective Loss 0.138916 LR 0.000250 Time 0.021189 -2022-12-06 11:42:18,343 - Epoch: [178][ 490/ 1200] Overall Loss 0.138681 Objective Loss 0.138681 LR 0.000250 Time 0.021161 -2022-12-06 11:42:18,539 - Epoch: [178][ 500/ 1200] Overall Loss 0.139145 Objective Loss 0.139145 LR 0.000250 Time 0.021128 -2022-12-06 11:42:18,739 - Epoch: [178][ 510/ 1200] Overall Loss 0.139544 Objective Loss 0.139544 LR 0.000250 Time 0.021105 -2022-12-06 11:42:18,934 - Epoch: [178][ 520/ 1200] Overall Loss 0.139205 Objective Loss 0.139205 LR 0.000250 Time 0.021074 -2022-12-06 11:42:19,132 - Epoch: [178][ 530/ 1200] Overall Loss 0.139474 Objective Loss 0.139474 LR 0.000250 Time 0.021049 -2022-12-06 11:42:19,328 - Epoch: [178][ 540/ 1200] Overall Loss 0.139308 Objective Loss 0.139308 LR 0.000250 Time 0.021021 -2022-12-06 11:42:19,526 - Epoch: [178][ 550/ 1200] Overall Loss 0.139368 Objective Loss 0.139368 LR 0.000250 Time 0.020998 -2022-12-06 11:42:19,722 - Epoch: [178][ 560/ 1200] Overall Loss 0.139491 Objective Loss 0.139491 LR 0.000250 Time 0.020971 -2022-12-06 11:42:19,920 - Epoch: [178][ 570/ 1200] Overall Loss 0.139642 Objective Loss 0.139642 LR 0.000250 Time 0.020950 -2022-12-06 11:42:20,116 - Epoch: [178][ 580/ 1200] Overall Loss 0.139733 Objective Loss 0.139733 LR 0.000250 Time 0.020925 -2022-12-06 11:42:20,313 - Epoch: [178][ 590/ 1200] Overall Loss 0.139775 Objective Loss 0.139775 LR 0.000250 Time 0.020905 -2022-12-06 11:42:20,511 - Epoch: [178][ 600/ 1200] Overall Loss 0.139606 Objective Loss 0.139606 LR 0.000250 Time 0.020884 -2022-12-06 11:42:20,709 - Epoch: [178][ 610/ 1200] Overall Loss 0.139578 Objective Loss 0.139578 LR 0.000250 Time 0.020867 -2022-12-06 11:42:20,906 - Epoch: [178][ 620/ 1200] Overall Loss 0.139669 Objective Loss 0.139669 LR 0.000250 Time 0.020846 -2022-12-06 11:42:21,104 - Epoch: [178][ 630/ 1200] Overall Loss 0.139821 Objective Loss 0.139821 LR 0.000250 Time 0.020829 -2022-12-06 11:42:21,300 - Epoch: [178][ 640/ 1200] Overall Loss 0.139815 Objective Loss 0.139815 LR 0.000250 Time 0.020808 -2022-12-06 11:42:21,499 - Epoch: [178][ 650/ 1200] Overall Loss 0.139794 Objective Loss 0.139794 LR 0.000250 Time 0.020793 -2022-12-06 11:42:21,695 - Epoch: [178][ 660/ 1200] Overall Loss 0.140020 Objective Loss 0.140020 LR 0.000250 Time 0.020775 -2022-12-06 11:42:21,892 - Epoch: [178][ 670/ 1200] Overall Loss 0.139959 Objective Loss 0.139959 LR 0.000250 Time 0.020759 -2022-12-06 11:42:22,089 - Epoch: [178][ 680/ 1200] Overall Loss 0.139845 Objective Loss 0.139845 LR 0.000250 Time 0.020741 -2022-12-06 11:42:22,287 - Epoch: [178][ 690/ 1200] Overall Loss 0.139716 Objective Loss 0.139716 LR 0.000250 Time 0.020727 -2022-12-06 11:42:22,482 - Epoch: [178][ 700/ 1200] Overall Loss 0.139833 Objective Loss 0.139833 LR 0.000250 Time 0.020710 -2022-12-06 11:42:22,681 - Epoch: [178][ 710/ 1200] Overall Loss 0.139948 Objective Loss 0.139948 LR 0.000250 Time 0.020696 -2022-12-06 11:42:22,877 - Epoch: [178][ 720/ 1200] Overall Loss 0.139650 Objective Loss 0.139650 LR 0.000250 Time 0.020681 -2022-12-06 11:42:23,076 - Epoch: [178][ 730/ 1200] Overall Loss 0.139746 Objective Loss 0.139746 LR 0.000250 Time 0.020669 -2022-12-06 11:42:23,272 - Epoch: [178][ 740/ 1200] Overall Loss 0.139894 Objective Loss 0.139894 LR 0.000250 Time 0.020655 -2022-12-06 11:42:23,471 - Epoch: [178][ 750/ 1200] Overall Loss 0.140091 Objective Loss 0.140091 LR 0.000250 Time 0.020643 -2022-12-06 11:42:23,667 - Epoch: [178][ 760/ 1200] Overall Loss 0.139898 Objective Loss 0.139898 LR 0.000250 Time 0.020629 -2022-12-06 11:42:23,865 - Epoch: [178][ 770/ 1200] Overall Loss 0.139929 Objective Loss 0.139929 LR 0.000250 Time 0.020618 -2022-12-06 11:42:24,062 - Epoch: [178][ 780/ 1200] Overall Loss 0.139945 Objective Loss 0.139945 LR 0.000250 Time 0.020605 -2022-12-06 11:42:24,260 - Epoch: [178][ 790/ 1200] Overall Loss 0.140025 Objective Loss 0.140025 LR 0.000250 Time 0.020594 -2022-12-06 11:42:24,456 - Epoch: [178][ 800/ 1200] Overall Loss 0.139915 Objective Loss 0.139915 LR 0.000250 Time 0.020581 -2022-12-06 11:42:24,654 - Epoch: [178][ 810/ 1200] Overall Loss 0.139909 Objective Loss 0.139909 LR 0.000250 Time 0.020570 -2022-12-06 11:42:24,850 - Epoch: [178][ 820/ 1200] Overall Loss 0.139920 Objective Loss 0.139920 LR 0.000250 Time 0.020558 -2022-12-06 11:42:25,048 - Epoch: [178][ 830/ 1200] Overall Loss 0.140399 Objective Loss 0.140399 LR 0.000250 Time 0.020548 -2022-12-06 11:42:25,244 - Epoch: [178][ 840/ 1200] Overall Loss 0.140378 Objective Loss 0.140378 LR 0.000250 Time 0.020536 -2022-12-06 11:42:25,442 - Epoch: [178][ 850/ 1200] Overall Loss 0.140231 Objective Loss 0.140231 LR 0.000250 Time 0.020527 -2022-12-06 11:42:25,637 - Epoch: [178][ 860/ 1200] Overall Loss 0.140342 Objective Loss 0.140342 LR 0.000250 Time 0.020515 -2022-12-06 11:42:25,836 - Epoch: [178][ 870/ 1200] Overall Loss 0.140340 Objective Loss 0.140340 LR 0.000250 Time 0.020506 -2022-12-06 11:42:26,031 - Epoch: [178][ 880/ 1200] Overall Loss 0.140332 Objective Loss 0.140332 LR 0.000250 Time 0.020495 -2022-12-06 11:42:26,230 - Epoch: [178][ 890/ 1200] Overall Loss 0.140369 Objective Loss 0.140369 LR 0.000250 Time 0.020488 -2022-12-06 11:42:26,426 - Epoch: [178][ 900/ 1200] Overall Loss 0.140195 Objective Loss 0.140195 LR 0.000250 Time 0.020477 -2022-12-06 11:42:26,624 - Epoch: [178][ 910/ 1200] Overall Loss 0.140287 Objective Loss 0.140287 LR 0.000250 Time 0.020469 -2022-12-06 11:42:26,820 - Epoch: [178][ 920/ 1200] Overall Loss 0.140427 Objective Loss 0.140427 LR 0.000250 Time 0.020459 -2022-12-06 11:42:27,019 - Epoch: [178][ 930/ 1200] Overall Loss 0.140514 Objective Loss 0.140514 LR 0.000250 Time 0.020452 -2022-12-06 11:42:27,215 - Epoch: [178][ 940/ 1200] Overall Loss 0.140600 Objective Loss 0.140600 LR 0.000250 Time 0.020443 -2022-12-06 11:42:27,413 - Epoch: [178][ 950/ 1200] Overall Loss 0.140786 Objective Loss 0.140786 LR 0.000250 Time 0.020436 -2022-12-06 11:42:27,610 - Epoch: [178][ 960/ 1200] Overall Loss 0.140843 Objective Loss 0.140843 LR 0.000250 Time 0.020427 -2022-12-06 11:42:27,808 - Epoch: [178][ 970/ 1200] Overall Loss 0.141086 Objective Loss 0.141086 LR 0.000250 Time 0.020420 -2022-12-06 11:42:28,004 - Epoch: [178][ 980/ 1200] Overall Loss 0.141229 Objective Loss 0.141229 LR 0.000250 Time 0.020411 -2022-12-06 11:42:28,202 - Epoch: [178][ 990/ 1200] Overall Loss 0.141235 Objective Loss 0.141235 LR 0.000250 Time 0.020405 -2022-12-06 11:42:28,399 - Epoch: [178][ 1000/ 1200] Overall Loss 0.141209 Objective Loss 0.141209 LR 0.000250 Time 0.020397 -2022-12-06 11:42:28,598 - Epoch: [178][ 1010/ 1200] Overall Loss 0.141047 Objective Loss 0.141047 LR 0.000250 Time 0.020392 -2022-12-06 11:42:28,794 - Epoch: [178][ 1020/ 1200] Overall Loss 0.141022 Objective Loss 0.141022 LR 0.000250 Time 0.020383 -2022-12-06 11:42:28,992 - Epoch: [178][ 1030/ 1200] Overall Loss 0.141166 Objective Loss 0.141166 LR 0.000250 Time 0.020377 -2022-12-06 11:42:29,187 - Epoch: [178][ 1040/ 1200] Overall Loss 0.141423 Objective Loss 0.141423 LR 0.000250 Time 0.020368 -2022-12-06 11:42:29,386 - Epoch: [178][ 1050/ 1200] Overall Loss 0.141593 Objective Loss 0.141593 LR 0.000250 Time 0.020363 -2022-12-06 11:42:29,582 - Epoch: [178][ 1060/ 1200] Overall Loss 0.141540 Objective Loss 0.141540 LR 0.000250 Time 0.020355 -2022-12-06 11:42:29,780 - Epoch: [178][ 1070/ 1200] Overall Loss 0.141545 Objective Loss 0.141545 LR 0.000250 Time 0.020350 -2022-12-06 11:42:29,976 - Epoch: [178][ 1080/ 1200] Overall Loss 0.141456 Objective Loss 0.141456 LR 0.000250 Time 0.020342 -2022-12-06 11:42:30,174 - Epoch: [178][ 1090/ 1200] Overall Loss 0.141451 Objective Loss 0.141451 LR 0.000250 Time 0.020337 -2022-12-06 11:42:30,370 - Epoch: [178][ 1100/ 1200] Overall Loss 0.141454 Objective Loss 0.141454 LR 0.000250 Time 0.020329 -2022-12-06 11:42:30,568 - Epoch: [178][ 1110/ 1200] Overall Loss 0.141305 Objective Loss 0.141305 LR 0.000250 Time 0.020324 -2022-12-06 11:42:30,764 - Epoch: [178][ 1120/ 1200] Overall Loss 0.141193 Objective Loss 0.141193 LR 0.000250 Time 0.020317 -2022-12-06 11:42:30,962 - Epoch: [178][ 1130/ 1200] Overall Loss 0.141166 Objective Loss 0.141166 LR 0.000250 Time 0.020312 -2022-12-06 11:42:31,158 - Epoch: [178][ 1140/ 1200] Overall Loss 0.141087 Objective Loss 0.141087 LR 0.000250 Time 0.020305 -2022-12-06 11:42:31,356 - Epoch: [178][ 1150/ 1200] Overall Loss 0.141024 Objective Loss 0.141024 LR 0.000250 Time 0.020300 -2022-12-06 11:42:31,552 - Epoch: [178][ 1160/ 1200] Overall Loss 0.141111 Objective Loss 0.141111 LR 0.000250 Time 0.020294 -2022-12-06 11:42:31,751 - Epoch: [178][ 1170/ 1200] Overall Loss 0.141022 Objective Loss 0.141022 LR 0.000250 Time 0.020290 -2022-12-06 11:42:31,946 - Epoch: [178][ 1180/ 1200] Overall Loss 0.140992 Objective Loss 0.140992 LR 0.000250 Time 0.020283 -2022-12-06 11:42:32,144 - Epoch: [178][ 1190/ 1200] Overall Loss 0.141059 Objective Loss 0.141059 LR 0.000250 Time 0.020279 -2022-12-06 11:42:32,375 - Epoch: [178][ 1200/ 1200] Overall Loss 0.141183 Objective Loss 0.141183 Top1 90.167364 Top5 98.535565 LR 0.000250 Time 0.020302 -2022-12-06 11:42:32,464 - --- validate (epoch=178)----------- -2022-12-06 11:42:32,464 - 34129 samples (256 per mini-batch) -2022-12-06 11:42:32,908 - Epoch: [178][ 10/ 134] Loss 0.220687 Top1 88.320312 Top5 98.984375 -2022-12-06 11:42:33,036 - Epoch: [178][ 20/ 134] Loss 0.228762 Top1 88.027344 Top5 98.671875 -2022-12-06 11:42:33,163 - Epoch: [178][ 30/ 134] Loss 0.226724 Top1 88.098958 Top5 98.815104 -2022-12-06 11:42:33,294 - Epoch: [178][ 40/ 134] Loss 0.227288 Top1 88.251953 Top5 98.798828 -2022-12-06 11:42:33,421 - Epoch: [178][ 50/ 134] Loss 0.228113 Top1 88.101562 Top5 98.734375 -2022-12-06 11:42:33,551 - Epoch: [178][ 60/ 134] Loss 0.230629 Top1 88.138021 Top5 98.723958 -2022-12-06 11:42:33,677 - Epoch: [178][ 70/ 134] Loss 0.233004 Top1 88.147321 Top5 98.727679 -2022-12-06 11:42:33,804 - Epoch: [178][ 80/ 134] Loss 0.231209 Top1 88.198242 Top5 98.745117 -2022-12-06 11:42:33,934 - Epoch: [178][ 90/ 134] Loss 0.233797 Top1 88.177083 Top5 98.741319 -2022-12-06 11:42:34,062 - Epoch: [178][ 100/ 134] Loss 0.237120 Top1 88.160156 Top5 98.734375 -2022-12-06 11:42:34,193 - Epoch: [178][ 110/ 134] Loss 0.236513 Top1 88.096591 Top5 98.753551 -2022-12-06 11:42:34,325 - Epoch: [178][ 120/ 134] Loss 0.235565 Top1 88.125000 Top5 98.763021 -2022-12-06 11:42:34,459 - Epoch: [178][ 130/ 134] Loss 0.237507 Top1 88.052885 Top5 98.731971 -2022-12-06 11:42:34,499 - Epoch: [178][ 134/ 134] Loss 0.236608 Top1 88.071728 Top5 98.722494 -2022-12-06 11:42:34,602 - ==> Top1: 88.072 Top5: 98.722 Loss: 0.237 - -2022-12-06 11:42:34,603 - ==> Confusion: -[[ 923 1 2 2 4 6 0 0 7 37 0 1 1 3 4 1 0 0 1 0 3] - [ 3 941 3 2 6 20 0 11 2 0 3 4 0 0 0 0 5 1 12 3 11] - [ 5 4 1021 9 3 2 11 4 0 2 5 6 3 4 1 2 0 1 2 4 14] - [ 2 0 20 941 1 2 1 0 0 1 12 1 1 3 13 0 0 3 9 1 9] - [ 9 6 0 0 961 1 1 2 1 4 1 3 1 4 8 6 4 3 0 0 5] - [ 2 14 0 4 5 980 1 15 2 1 2 16 5 11 1 1 3 0 0 0 6] - [ 1 3 8 2 1 3 1074 3 0 0 2 1 0 1 0 6 0 2 0 6 5] - [ 3 5 2 3 3 29 9 954 0 0 3 6 1 3 0 0 2 0 19 7 5] - [ 4 0 0 0 0 3 1 0 1004 28 8 1 1 4 4 1 1 1 1 1 1] - [ 57 2 1 0 4 2 0 2 30 878 1 2 0 9 2 1 1 2 0 0 7] - [ 1 1 1 3 0 1 2 2 11 0 976 1 0 10 1 1 0 0 5 0 3] - [ 2 0 2 1 1 7 4 2 1 0 0 980 27 2 0 4 4 6 0 3 5] - [ 0 0 0 1 0 2 0 0 0 1 0 27 910 2 1 5 1 8 1 3 7] - [ 0 0 1 0 0 12 0 3 14 10 4 6 4 952 0 1 5 1 0 1 9] - [ 5 5 3 9 2 3 0 0 21 1 1 1 3 3 1060 0 1 1 5 0 6] - [ 1 0 1 0 3 0 2 1 1 0 0 9 8 3 0 986 5 17 0 1 5] - [ 1 2 1 2 2 2 0 0 0 0 0 1 4 1 1 10 1033 1 1 3 7] - [ 4 1 2 2 2 1 2 1 1 2 0 5 16 0 0 11 1 982 0 1 2] - [ 4 3 4 5 0 3 0 19 2 1 6 4 3 0 8 0 0 3 939 1 3] - [ 2 4 2 0 0 7 6 3 0 1 2 22 9 4 0 2 6 4 0 998 8] - [ 107 190 137 79 72 160 69 127 89 57 160 86 298 238 132 99 185 85 131 166 10559]] - -2022-12-06 11:42:35,264 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:42:35,264 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:42:35,270 - - -2022-12-06 11:42:35,270 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:42:36,213 - Epoch: [179][ 10/ 1200] Overall Loss 0.141819 Objective Loss 0.141819 LR 0.000250 Time 0.094169 -2022-12-06 11:42:36,418 - Epoch: [179][ 20/ 1200] Overall Loss 0.133544 Objective Loss 0.133544 LR 0.000250 Time 0.057337 -2022-12-06 11:42:36,618 - Epoch: [179][ 30/ 1200] Overall Loss 0.139748 Objective Loss 0.139748 LR 0.000250 Time 0.044858 -2022-12-06 11:42:36,815 - Epoch: [179][ 40/ 1200] Overall Loss 0.137044 Objective Loss 0.137044 LR 0.000250 Time 0.038546 -2022-12-06 11:42:37,015 - Epoch: [179][ 50/ 1200] Overall Loss 0.135479 Objective Loss 0.135479 LR 0.000250 Time 0.034828 -2022-12-06 11:42:37,210 - Epoch: [179][ 60/ 1200] Overall Loss 0.132554 Objective Loss 0.132554 LR 0.000250 Time 0.032269 -2022-12-06 11:42:37,410 - Epoch: [179][ 70/ 1200] Overall Loss 0.132504 Objective Loss 0.132504 LR 0.000250 Time 0.030504 -2022-12-06 11:42:37,606 - Epoch: [179][ 80/ 1200] Overall Loss 0.133474 Objective Loss 0.133474 LR 0.000250 Time 0.029137 -2022-12-06 11:42:37,806 - Epoch: [179][ 90/ 1200] Overall Loss 0.136831 Objective Loss 0.136831 LR 0.000250 Time 0.028116 -2022-12-06 11:42:38,003 - Epoch: [179][ 100/ 1200] Overall Loss 0.136332 Objective Loss 0.136332 LR 0.000250 Time 0.027270 -2022-12-06 11:42:38,202 - Epoch: [179][ 110/ 1200] Overall Loss 0.137286 Objective Loss 0.137286 LR 0.000250 Time 0.026595 -2022-12-06 11:42:38,398 - Epoch: [179][ 120/ 1200] Overall Loss 0.139212 Objective Loss 0.139212 LR 0.000250 Time 0.026007 -2022-12-06 11:42:38,598 - Epoch: [179][ 130/ 1200] Overall Loss 0.139147 Objective Loss 0.139147 LR 0.000250 Time 0.025539 -2022-12-06 11:42:38,794 - Epoch: [179][ 140/ 1200] Overall Loss 0.139117 Objective Loss 0.139117 LR 0.000250 Time 0.025113 -2022-12-06 11:42:38,993 - Epoch: [179][ 150/ 1200] Overall Loss 0.138597 Objective Loss 0.138597 LR 0.000250 Time 0.024763 -2022-12-06 11:42:39,190 - Epoch: [179][ 160/ 1200] Overall Loss 0.137851 Objective Loss 0.137851 LR 0.000250 Time 0.024439 -2022-12-06 11:42:39,389 - Epoch: [179][ 170/ 1200] Overall Loss 0.137000 Objective Loss 0.137000 LR 0.000250 Time 0.024169 -2022-12-06 11:42:39,586 - Epoch: [179][ 180/ 1200] Overall Loss 0.136748 Objective Loss 0.136748 LR 0.000250 Time 0.023920 -2022-12-06 11:42:39,785 - Epoch: [179][ 190/ 1200] Overall Loss 0.137040 Objective Loss 0.137040 LR 0.000250 Time 0.023705 -2022-12-06 11:42:39,981 - Epoch: [179][ 200/ 1200] Overall Loss 0.136541 Objective Loss 0.136541 LR 0.000250 Time 0.023499 -2022-12-06 11:42:40,181 - Epoch: [179][ 210/ 1200] Overall Loss 0.137257 Objective Loss 0.137257 LR 0.000250 Time 0.023326 -2022-12-06 11:42:40,377 - Epoch: [179][ 220/ 1200] Overall Loss 0.137760 Objective Loss 0.137760 LR 0.000250 Time 0.023155 -2022-12-06 11:42:40,577 - Epoch: [179][ 230/ 1200] Overall Loss 0.137335 Objective Loss 0.137335 LR 0.000250 Time 0.023015 -2022-12-06 11:42:40,772 - Epoch: [179][ 240/ 1200] Overall Loss 0.138298 Objective Loss 0.138298 LR 0.000250 Time 0.022868 -2022-12-06 11:42:40,971 - Epoch: [179][ 250/ 1200] Overall Loss 0.138818 Objective Loss 0.138818 LR 0.000250 Time 0.022747 -2022-12-06 11:42:41,167 - Epoch: [179][ 260/ 1200] Overall Loss 0.138975 Objective Loss 0.138975 LR 0.000250 Time 0.022624 -2022-12-06 11:42:41,366 - Epoch: [179][ 270/ 1200] Overall Loss 0.139190 Objective Loss 0.139190 LR 0.000250 Time 0.022520 -2022-12-06 11:42:41,562 - Epoch: [179][ 280/ 1200] Overall Loss 0.139036 Objective Loss 0.139036 LR 0.000250 Time 0.022415 -2022-12-06 11:42:41,762 - Epoch: [179][ 290/ 1200] Overall Loss 0.138617 Objective Loss 0.138617 LR 0.000250 Time 0.022328 -2022-12-06 11:42:41,957 - Epoch: [179][ 300/ 1200] Overall Loss 0.138522 Objective Loss 0.138522 LR 0.000250 Time 0.022235 -2022-12-06 11:42:42,156 - Epoch: [179][ 310/ 1200] Overall Loss 0.137925 Objective Loss 0.137925 LR 0.000250 Time 0.022158 -2022-12-06 11:42:42,353 - Epoch: [179][ 320/ 1200] Overall Loss 0.137802 Objective Loss 0.137802 LR 0.000250 Time 0.022078 -2022-12-06 11:42:42,553 - Epoch: [179][ 330/ 1200] Overall Loss 0.137729 Objective Loss 0.137729 LR 0.000250 Time 0.022012 -2022-12-06 11:42:42,749 - Epoch: [179][ 340/ 1200] Overall Loss 0.137655 Objective Loss 0.137655 LR 0.000250 Time 0.021940 -2022-12-06 11:42:42,948 - Epoch: [179][ 350/ 1200] Overall Loss 0.137645 Objective Loss 0.137645 LR 0.000250 Time 0.021880 -2022-12-06 11:42:43,144 - Epoch: [179][ 360/ 1200] Overall Loss 0.137534 Objective Loss 0.137534 LR 0.000250 Time 0.021815 -2022-12-06 11:42:43,343 - Epoch: [179][ 370/ 1200] Overall Loss 0.137260 Objective Loss 0.137260 LR 0.000250 Time 0.021763 -2022-12-06 11:42:43,540 - Epoch: [179][ 380/ 1200] Overall Loss 0.137769 Objective Loss 0.137769 LR 0.000250 Time 0.021706 -2022-12-06 11:42:43,739 - Epoch: [179][ 390/ 1200] Overall Loss 0.137628 Objective Loss 0.137628 LR 0.000250 Time 0.021660 -2022-12-06 11:42:43,936 - Epoch: [179][ 400/ 1200] Overall Loss 0.136992 Objective Loss 0.136992 LR 0.000250 Time 0.021609 -2022-12-06 11:42:44,135 - Epoch: [179][ 410/ 1200] Overall Loss 0.137341 Objective Loss 0.137341 LR 0.000250 Time 0.021566 -2022-12-06 11:42:44,331 - Epoch: [179][ 420/ 1200] Overall Loss 0.137040 Objective Loss 0.137040 LR 0.000250 Time 0.021518 -2022-12-06 11:42:44,530 - Epoch: [179][ 430/ 1200] Overall Loss 0.137041 Objective Loss 0.137041 LR 0.000250 Time 0.021480 -2022-12-06 11:42:44,726 - Epoch: [179][ 440/ 1200] Overall Loss 0.137048 Objective Loss 0.137048 LR 0.000250 Time 0.021434 -2022-12-06 11:42:44,925 - Epoch: [179][ 450/ 1200] Overall Loss 0.136919 Objective Loss 0.136919 LR 0.000250 Time 0.021401 -2022-12-06 11:42:45,122 - Epoch: [179][ 460/ 1200] Overall Loss 0.136790 Objective Loss 0.136790 LR 0.000250 Time 0.021361 -2022-12-06 11:42:45,320 - Epoch: [179][ 470/ 1200] Overall Loss 0.136765 Objective Loss 0.136765 LR 0.000250 Time 0.021328 -2022-12-06 11:42:45,517 - Epoch: [179][ 480/ 1200] Overall Loss 0.136707 Objective Loss 0.136707 LR 0.000250 Time 0.021293 -2022-12-06 11:42:45,717 - Epoch: [179][ 490/ 1200] Overall Loss 0.136440 Objective Loss 0.136440 LR 0.000250 Time 0.021264 -2022-12-06 11:42:45,913 - Epoch: [179][ 500/ 1200] Overall Loss 0.136632 Objective Loss 0.136632 LR 0.000250 Time 0.021231 -2022-12-06 11:42:46,113 - Epoch: [179][ 510/ 1200] Overall Loss 0.136530 Objective Loss 0.136530 LR 0.000250 Time 0.021205 -2022-12-06 11:42:46,310 - Epoch: [179][ 520/ 1200] Overall Loss 0.136814 Objective Loss 0.136814 LR 0.000250 Time 0.021174 -2022-12-06 11:42:46,509 - Epoch: [179][ 530/ 1200] Overall Loss 0.137503 Objective Loss 0.137503 LR 0.000250 Time 0.021151 -2022-12-06 11:42:46,708 - Epoch: [179][ 540/ 1200] Overall Loss 0.137366 Objective Loss 0.137366 LR 0.000250 Time 0.021125 -2022-12-06 11:42:46,903 - Epoch: [179][ 550/ 1200] Overall Loss 0.137294 Objective Loss 0.137294 LR 0.000250 Time 0.021095 -2022-12-06 11:42:47,102 - Epoch: [179][ 560/ 1200] Overall Loss 0.137438 Objective Loss 0.137438 LR 0.000250 Time 0.021073 -2022-12-06 11:42:47,296 - Epoch: [179][ 570/ 1200] Overall Loss 0.137323 Objective Loss 0.137323 LR 0.000250 Time 0.021043 -2022-12-06 11:42:47,495 - Epoch: [179][ 580/ 1200] Overall Loss 0.137533 Objective Loss 0.137533 LR 0.000250 Time 0.021022 -2022-12-06 11:42:47,691 - Epoch: [179][ 590/ 1200] Overall Loss 0.137794 Objective Loss 0.137794 LR 0.000250 Time 0.020998 -2022-12-06 11:42:47,890 - Epoch: [179][ 600/ 1200] Overall Loss 0.137833 Objective Loss 0.137833 LR 0.000250 Time 0.020977 -2022-12-06 11:42:48,085 - Epoch: [179][ 610/ 1200] Overall Loss 0.138044 Objective Loss 0.138044 LR 0.000250 Time 0.020954 -2022-12-06 11:42:48,285 - Epoch: [179][ 620/ 1200] Overall Loss 0.137758 Objective Loss 0.137758 LR 0.000250 Time 0.020936 -2022-12-06 11:42:48,480 - Epoch: [179][ 630/ 1200] Overall Loss 0.137817 Objective Loss 0.137817 LR 0.000250 Time 0.020914 -2022-12-06 11:42:48,679 - Epoch: [179][ 640/ 1200] Overall Loss 0.138010 Objective Loss 0.138010 LR 0.000250 Time 0.020896 -2022-12-06 11:42:48,875 - Epoch: [179][ 650/ 1200] Overall Loss 0.138178 Objective Loss 0.138178 LR 0.000250 Time 0.020875 -2022-12-06 11:42:49,074 - Epoch: [179][ 660/ 1200] Overall Loss 0.138510 Objective Loss 0.138510 LR 0.000250 Time 0.020859 -2022-12-06 11:42:49,268 - Epoch: [179][ 670/ 1200] Overall Loss 0.138408 Objective Loss 0.138408 LR 0.000250 Time 0.020838 -2022-12-06 11:42:49,467 - Epoch: [179][ 680/ 1200] Overall Loss 0.138611 Objective Loss 0.138611 LR 0.000250 Time 0.020823 -2022-12-06 11:42:49,663 - Epoch: [179][ 690/ 1200] Overall Loss 0.138651 Objective Loss 0.138651 LR 0.000250 Time 0.020804 -2022-12-06 11:42:49,861 - Epoch: [179][ 700/ 1200] Overall Loss 0.138737 Objective Loss 0.138737 LR 0.000250 Time 0.020789 -2022-12-06 11:42:50,056 - Epoch: [179][ 710/ 1200] Overall Loss 0.138950 Objective Loss 0.138950 LR 0.000250 Time 0.020770 -2022-12-06 11:42:50,254 - Epoch: [179][ 720/ 1200] Overall Loss 0.138970 Objective Loss 0.138970 LR 0.000250 Time 0.020756 -2022-12-06 11:42:50,449 - Epoch: [179][ 730/ 1200] Overall Loss 0.139079 Objective Loss 0.139079 LR 0.000250 Time 0.020738 -2022-12-06 11:42:50,648 - Epoch: [179][ 740/ 1200] Overall Loss 0.138816 Objective Loss 0.138816 LR 0.000250 Time 0.020726 -2022-12-06 11:42:50,843 - Epoch: [179][ 750/ 1200] Overall Loss 0.138843 Objective Loss 0.138843 LR 0.000250 Time 0.020709 -2022-12-06 11:42:51,041 - Epoch: [179][ 760/ 1200] Overall Loss 0.139037 Objective Loss 0.139037 LR 0.000250 Time 0.020697 -2022-12-06 11:42:51,236 - Epoch: [179][ 770/ 1200] Overall Loss 0.139138 Objective Loss 0.139138 LR 0.000250 Time 0.020680 -2022-12-06 11:42:51,435 - Epoch: [179][ 780/ 1200] Overall Loss 0.138929 Objective Loss 0.138929 LR 0.000250 Time 0.020670 -2022-12-06 11:42:51,631 - Epoch: [179][ 790/ 1200] Overall Loss 0.139123 Objective Loss 0.139123 LR 0.000250 Time 0.020655 -2022-12-06 11:42:51,829 - Epoch: [179][ 800/ 1200] Overall Loss 0.139152 Objective Loss 0.139152 LR 0.000250 Time 0.020644 -2022-12-06 11:42:52,024 - Epoch: [179][ 810/ 1200] Overall Loss 0.139071 Objective Loss 0.139071 LR 0.000250 Time 0.020629 -2022-12-06 11:42:52,222 - Epoch: [179][ 820/ 1200] Overall Loss 0.138934 Objective Loss 0.138934 LR 0.000250 Time 0.020618 -2022-12-06 11:42:52,417 - Epoch: [179][ 830/ 1200] Overall Loss 0.138885 Objective Loss 0.138885 LR 0.000250 Time 0.020604 -2022-12-06 11:42:52,616 - Epoch: [179][ 840/ 1200] Overall Loss 0.139014 Objective Loss 0.139014 LR 0.000250 Time 0.020595 -2022-12-06 11:42:52,811 - Epoch: [179][ 850/ 1200] Overall Loss 0.138972 Objective Loss 0.138972 LR 0.000250 Time 0.020581 -2022-12-06 11:42:53,009 - Epoch: [179][ 860/ 1200] Overall Loss 0.138930 Objective Loss 0.138930 LR 0.000250 Time 0.020572 -2022-12-06 11:42:53,204 - Epoch: [179][ 870/ 1200] Overall Loss 0.138947 Objective Loss 0.138947 LR 0.000250 Time 0.020559 -2022-12-06 11:42:53,402 - Epoch: [179][ 880/ 1200] Overall Loss 0.138915 Objective Loss 0.138915 LR 0.000250 Time 0.020550 -2022-12-06 11:42:53,597 - Epoch: [179][ 890/ 1200] Overall Loss 0.138829 Objective Loss 0.138829 LR 0.000250 Time 0.020537 -2022-12-06 11:42:53,796 - Epoch: [179][ 900/ 1200] Overall Loss 0.138942 Objective Loss 0.138942 LR 0.000250 Time 0.020530 -2022-12-06 11:42:53,992 - Epoch: [179][ 910/ 1200] Overall Loss 0.138944 Objective Loss 0.138944 LR 0.000250 Time 0.020518 -2022-12-06 11:42:54,191 - Epoch: [179][ 920/ 1200] Overall Loss 0.138887 Objective Loss 0.138887 LR 0.000250 Time 0.020511 -2022-12-06 11:42:54,385 - Epoch: [179][ 930/ 1200] Overall Loss 0.139154 Objective Loss 0.139154 LR 0.000250 Time 0.020499 -2022-12-06 11:42:54,583 - Epoch: [179][ 940/ 1200] Overall Loss 0.139030 Objective Loss 0.139030 LR 0.000250 Time 0.020491 -2022-12-06 11:42:54,779 - Epoch: [179][ 950/ 1200] Overall Loss 0.139087 Objective Loss 0.139087 LR 0.000250 Time 0.020481 -2022-12-06 11:42:54,978 - Epoch: [179][ 960/ 1200] Overall Loss 0.139238 Objective Loss 0.139238 LR 0.000250 Time 0.020474 -2022-12-06 11:42:55,173 - Epoch: [179][ 970/ 1200] Overall Loss 0.139298 Objective Loss 0.139298 LR 0.000250 Time 0.020463 -2022-12-06 11:42:55,371 - Epoch: [179][ 980/ 1200] Overall Loss 0.139465 Objective Loss 0.139465 LR 0.000250 Time 0.020457 -2022-12-06 11:42:55,567 - Epoch: [179][ 990/ 1200] Overall Loss 0.139341 Objective Loss 0.139341 LR 0.000250 Time 0.020447 -2022-12-06 11:42:55,766 - Epoch: [179][ 1000/ 1200] Overall Loss 0.139390 Objective Loss 0.139390 LR 0.000250 Time 0.020441 -2022-12-06 11:42:55,960 - Epoch: [179][ 1010/ 1200] Overall Loss 0.139331 Objective Loss 0.139331 LR 0.000250 Time 0.020430 -2022-12-06 11:42:56,159 - Epoch: [179][ 1020/ 1200] Overall Loss 0.139369 Objective Loss 0.139369 LR 0.000250 Time 0.020424 -2022-12-06 11:42:56,354 - Epoch: [179][ 1030/ 1200] Overall Loss 0.139369 Objective Loss 0.139369 LR 0.000250 Time 0.020415 -2022-12-06 11:42:56,553 - Epoch: [179][ 1040/ 1200] Overall Loss 0.139435 Objective Loss 0.139435 LR 0.000250 Time 0.020409 -2022-12-06 11:42:56,747 - Epoch: [179][ 1050/ 1200] Overall Loss 0.139387 Objective Loss 0.139387 LR 0.000250 Time 0.020399 -2022-12-06 11:42:56,946 - Epoch: [179][ 1060/ 1200] Overall Loss 0.139380 Objective Loss 0.139380 LR 0.000250 Time 0.020394 -2022-12-06 11:42:57,140 - Epoch: [179][ 1070/ 1200] Overall Loss 0.139599 Objective Loss 0.139599 LR 0.000250 Time 0.020385 -2022-12-06 11:42:57,339 - Epoch: [179][ 1080/ 1200] Overall Loss 0.139667 Objective Loss 0.139667 LR 0.000250 Time 0.020379 -2022-12-06 11:42:57,534 - Epoch: [179][ 1090/ 1200] Overall Loss 0.139702 Objective Loss 0.139702 LR 0.000250 Time 0.020370 -2022-12-06 11:42:57,732 - Epoch: [179][ 1100/ 1200] Overall Loss 0.139631 Objective Loss 0.139631 LR 0.000250 Time 0.020365 -2022-12-06 11:42:57,927 - Epoch: [179][ 1110/ 1200] Overall Loss 0.139697 Objective Loss 0.139697 LR 0.000250 Time 0.020357 -2022-12-06 11:42:58,124 - Epoch: [179][ 1120/ 1200] Overall Loss 0.139862 Objective Loss 0.139862 LR 0.000250 Time 0.020351 -2022-12-06 11:42:58,319 - Epoch: [179][ 1130/ 1200] Overall Loss 0.139863 Objective Loss 0.139863 LR 0.000250 Time 0.020343 -2022-12-06 11:42:58,518 - Epoch: [179][ 1140/ 1200] Overall Loss 0.139913 Objective Loss 0.139913 LR 0.000250 Time 0.020338 -2022-12-06 11:42:58,713 - Epoch: [179][ 1150/ 1200] Overall Loss 0.139903 Objective Loss 0.139903 LR 0.000250 Time 0.020331 -2022-12-06 11:42:58,912 - Epoch: [179][ 1160/ 1200] Overall Loss 0.139795 Objective Loss 0.139795 LR 0.000250 Time 0.020326 -2022-12-06 11:42:59,107 - Epoch: [179][ 1170/ 1200] Overall Loss 0.139973 Objective Loss 0.139973 LR 0.000250 Time 0.020318 -2022-12-06 11:42:59,305 - Epoch: [179][ 1180/ 1200] Overall Loss 0.139988 Objective Loss 0.139988 LR 0.000250 Time 0.020314 -2022-12-06 11:42:59,500 - Epoch: [179][ 1190/ 1200] Overall Loss 0.140005 Objective Loss 0.140005 LR 0.000250 Time 0.020307 -2022-12-06 11:42:59,723 - Epoch: [179][ 1200/ 1200] Overall Loss 0.140163 Objective Loss 0.140163 Top1 92.259414 Top5 98.953975 LR 0.000250 Time 0.020323 -2022-12-06 11:42:59,813 - --- validate (epoch=179)----------- -2022-12-06 11:42:59,813 - 34129 samples (256 per mini-batch) -2022-12-06 11:43:00,266 - Epoch: [179][ 10/ 134] Loss 0.210851 Top1 88.476562 Top5 98.359375 -2022-12-06 11:43:00,414 - Epoch: [179][ 20/ 134] Loss 0.208615 Top1 88.515625 Top5 98.515625 -2022-12-06 11:43:00,553 - Epoch: [179][ 30/ 134] Loss 0.228293 Top1 87.955729 Top5 98.411458 -2022-12-06 11:43:00,699 - Epoch: [179][ 40/ 134] Loss 0.228850 Top1 88.007812 Top5 98.554688 -2022-12-06 11:43:00,839 - Epoch: [179][ 50/ 134] Loss 0.237325 Top1 87.875000 Top5 98.570312 -2022-12-06 11:43:00,985 - Epoch: [179][ 60/ 134] Loss 0.236272 Top1 87.968750 Top5 98.574219 -2022-12-06 11:43:01,123 - Epoch: [179][ 70/ 134] Loss 0.235115 Top1 88.108259 Top5 98.588170 -2022-12-06 11:43:01,269 - Epoch: [179][ 80/ 134] Loss 0.235794 Top1 88.095703 Top5 98.632812 -2022-12-06 11:43:01,407 - Epoch: [179][ 90/ 134] Loss 0.233223 Top1 88.029514 Top5 98.637153 -2022-12-06 11:43:01,552 - Epoch: [179][ 100/ 134] Loss 0.235816 Top1 87.960938 Top5 98.601562 -2022-12-06 11:43:01,692 - Epoch: [179][ 110/ 134] Loss 0.236495 Top1 87.997159 Top5 98.568892 -2022-12-06 11:43:01,828 - Epoch: [179][ 120/ 134] Loss 0.237268 Top1 87.952474 Top5 98.557943 -2022-12-06 11:43:01,961 - Epoch: [179][ 130/ 134] Loss 0.235435 Top1 87.962740 Top5 98.560697 -2022-12-06 11:43:01,999 - Epoch: [179][ 134/ 134] Loss 0.234339 Top1 87.992616 Top5 98.567201 -2022-12-06 11:43:02,087 - ==> Top1: 87.993 Top5: 98.567 Loss: 0.234 - -2022-12-06 11:43:02,087 - ==> Confusion: -[[ 904 2 0 1 8 8 0 1 5 51 0 2 1 2 4 1 3 0 2 0 1] - [ 1 954 1 1 6 19 2 9 2 1 4 4 1 2 0 1 3 1 4 4 7] - [ 3 4 1009 13 3 3 14 13 1 3 5 4 1 4 3 1 0 1 5 4 9] - [ 0 5 13 953 0 3 1 0 0 1 8 0 4 1 9 0 1 2 10 1 8] - [ 3 5 3 0 965 2 0 2 1 6 1 1 1 1 10 5 5 3 1 2 3] - [ 1 11 0 2 5 991 2 14 2 3 1 11 2 9 2 1 2 2 0 4 4] - [ 1 2 6 2 0 3 1076 4 0 0 1 3 0 1 0 3 0 5 0 9 2] - [ 1 12 3 2 1 26 7 958 0 1 1 5 1 1 0 0 2 2 13 13 5] - [ 2 5 0 0 0 2 1 1 978 42 9 1 3 6 8 0 2 1 1 1 1] - [ 40 2 1 0 3 3 0 1 18 908 1 1 0 12 3 1 1 2 1 0 3] - [ 0 3 3 4 2 2 0 4 9 1 968 0 0 7 3 1 0 0 3 1 8] - [ 3 2 2 0 1 7 1 4 1 0 0 979 23 1 0 7 2 4 0 9 5] - [ 0 1 1 3 0 4 1 0 0 1 0 25 899 2 2 7 1 12 0 3 7] - [ 1 1 0 0 0 10 0 1 9 10 6 2 4 959 0 1 4 2 1 2 10] - [ 2 4 0 6 2 3 0 0 10 5 1 3 2 3 1074 1 1 1 6 0 6] - [ 0 0 0 2 4 0 2 1 0 1 3 8 4 1 0 999 4 10 0 0 4] - [ 3 1 1 0 3 0 1 0 2 0 1 1 2 2 0 8 1037 0 0 4 6] - [ 3 0 1 1 1 1 0 1 0 4 0 9 11 0 2 14 0 985 0 0 3] - [ 3 3 5 6 2 4 0 21 1 1 2 4 3 0 11 0 0 0 936 1 5] - [ 2 4 1 0 0 6 6 7 1 1 2 14 6 5 0 5 2 2 1 1009 6] - [ 93 211 117 87 94 160 86 138 93 77 141 89 278 247 137 116 175 86 141 175 10485]] - -2022-12-06 11:43:02,654 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:43:02,654 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:43:02,660 - - -2022-12-06 11:43:02,660 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:43:03,702 - Epoch: [180][ 10/ 1200] Overall Loss 0.146293 Objective Loss 0.146293 LR 0.000125 Time 0.104171 -2022-12-06 11:43:03,913 - Epoch: [180][ 20/ 1200] Overall Loss 0.142156 Objective Loss 0.142156 LR 0.000125 Time 0.062582 -2022-12-06 11:43:04,111 - Epoch: [180][ 30/ 1200] Overall Loss 0.141449 Objective Loss 0.141449 LR 0.000125 Time 0.048305 -2022-12-06 11:43:04,311 - Epoch: [180][ 40/ 1200] Overall Loss 0.138965 Objective Loss 0.138965 LR 0.000125 Time 0.041208 -2022-12-06 11:43:04,508 - Epoch: [180][ 50/ 1200] Overall Loss 0.140528 Objective Loss 0.140528 LR 0.000125 Time 0.036903 -2022-12-06 11:43:04,708 - Epoch: [180][ 60/ 1200] Overall Loss 0.140728 Objective Loss 0.140728 LR 0.000125 Time 0.034072 -2022-12-06 11:43:04,905 - Epoch: [180][ 70/ 1200] Overall Loss 0.139332 Objective Loss 0.139332 LR 0.000125 Time 0.032016 -2022-12-06 11:43:05,105 - Epoch: [180][ 80/ 1200] Overall Loss 0.141074 Objective Loss 0.141074 LR 0.000125 Time 0.030507 -2022-12-06 11:43:05,302 - Epoch: [180][ 90/ 1200] Overall Loss 0.139785 Objective Loss 0.139785 LR 0.000125 Time 0.029296 -2022-12-06 11:43:05,502 - Epoch: [180][ 100/ 1200] Overall Loss 0.141847 Objective Loss 0.141847 LR 0.000125 Time 0.028364 -2022-12-06 11:43:05,699 - Epoch: [180][ 110/ 1200] Overall Loss 0.140361 Objective Loss 0.140361 LR 0.000125 Time 0.027568 -2022-12-06 11:43:05,899 - Epoch: [180][ 120/ 1200] Overall Loss 0.138793 Objective Loss 0.138793 LR 0.000125 Time 0.026935 -2022-12-06 11:43:06,096 - Epoch: [180][ 130/ 1200] Overall Loss 0.136101 Objective Loss 0.136101 LR 0.000125 Time 0.026373 -2022-12-06 11:43:06,296 - Epoch: [180][ 140/ 1200] Overall Loss 0.136046 Objective Loss 0.136046 LR 0.000125 Time 0.025913 -2022-12-06 11:43:06,492 - Epoch: [180][ 150/ 1200] Overall Loss 0.136282 Objective Loss 0.136282 LR 0.000125 Time 0.025493 -2022-12-06 11:43:06,693 - Epoch: [180][ 160/ 1200] Overall Loss 0.134645 Objective Loss 0.134645 LR 0.000125 Time 0.025153 -2022-12-06 11:43:06,891 - Epoch: [180][ 170/ 1200] Overall Loss 0.134339 Objective Loss 0.134339 LR 0.000125 Time 0.024831 -2022-12-06 11:43:07,091 - Epoch: [180][ 180/ 1200] Overall Loss 0.134480 Objective Loss 0.134480 LR 0.000125 Time 0.024561 -2022-12-06 11:43:07,288 - Epoch: [180][ 190/ 1200] Overall Loss 0.134826 Objective Loss 0.134826 LR 0.000125 Time 0.024304 -2022-12-06 11:43:07,488 - Epoch: [180][ 200/ 1200] Overall Loss 0.134049 Objective Loss 0.134049 LR 0.000125 Time 0.024087 -2022-12-06 11:43:07,686 - Epoch: [180][ 210/ 1200] Overall Loss 0.134328 Objective Loss 0.134328 LR 0.000125 Time 0.023880 -2022-12-06 11:43:07,887 - Epoch: [180][ 220/ 1200] Overall Loss 0.133794 Objective Loss 0.133794 LR 0.000125 Time 0.023704 -2022-12-06 11:43:08,084 - Epoch: [180][ 230/ 1200] Overall Loss 0.132817 Objective Loss 0.132817 LR 0.000125 Time 0.023529 -2022-12-06 11:43:08,284 - Epoch: [180][ 240/ 1200] Overall Loss 0.131861 Objective Loss 0.131861 LR 0.000125 Time 0.023379 -2022-12-06 11:43:08,481 - Epoch: [180][ 250/ 1200] Overall Loss 0.131715 Objective Loss 0.131715 LR 0.000125 Time 0.023229 -2022-12-06 11:43:08,681 - Epoch: [180][ 260/ 1200] Overall Loss 0.132237 Objective Loss 0.132237 LR 0.000125 Time 0.023102 -2022-12-06 11:43:08,877 - Epoch: [180][ 270/ 1200] Overall Loss 0.132244 Objective Loss 0.132244 LR 0.000125 Time 0.022971 -2022-12-06 11:43:09,078 - Epoch: [180][ 280/ 1200] Overall Loss 0.132421 Objective Loss 0.132421 LR 0.000125 Time 0.022867 -2022-12-06 11:43:09,275 - Epoch: [180][ 290/ 1200] Overall Loss 0.132138 Objective Loss 0.132138 LR 0.000125 Time 0.022754 -2022-12-06 11:43:09,475 - Epoch: [180][ 300/ 1200] Overall Loss 0.132835 Objective Loss 0.132835 LR 0.000125 Time 0.022660 -2022-12-06 11:43:09,672 - Epoch: [180][ 310/ 1200] Overall Loss 0.133132 Objective Loss 0.133132 LR 0.000125 Time 0.022564 -2022-12-06 11:43:09,873 - Epoch: [180][ 320/ 1200] Overall Loss 0.133205 Objective Loss 0.133205 LR 0.000125 Time 0.022486 -2022-12-06 11:43:10,071 - Epoch: [180][ 330/ 1200] Overall Loss 0.133112 Objective Loss 0.133112 LR 0.000125 Time 0.022402 -2022-12-06 11:43:10,271 - Epoch: [180][ 340/ 1200] Overall Loss 0.132929 Objective Loss 0.132929 LR 0.000125 Time 0.022331 -2022-12-06 11:43:10,468 - Epoch: [180][ 350/ 1200] Overall Loss 0.133301 Objective Loss 0.133301 LR 0.000125 Time 0.022254 -2022-12-06 11:43:10,667 - Epoch: [180][ 360/ 1200] Overall Loss 0.133477 Objective Loss 0.133477 LR 0.000125 Time 0.022187 -2022-12-06 11:43:10,864 - Epoch: [180][ 370/ 1200] Overall Loss 0.133577 Objective Loss 0.133577 LR 0.000125 Time 0.022118 -2022-12-06 11:43:11,064 - Epoch: [180][ 380/ 1200] Overall Loss 0.133496 Objective Loss 0.133496 LR 0.000125 Time 0.022061 -2022-12-06 11:43:11,261 - Epoch: [180][ 390/ 1200] Overall Loss 0.133272 Objective Loss 0.133272 LR 0.000125 Time 0.021998 -2022-12-06 11:43:11,461 - Epoch: [180][ 400/ 1200] Overall Loss 0.133634 Objective Loss 0.133634 LR 0.000125 Time 0.021947 -2022-12-06 11:43:11,658 - Epoch: [180][ 410/ 1200] Overall Loss 0.133486 Objective Loss 0.133486 LR 0.000125 Time 0.021892 -2022-12-06 11:43:11,859 - Epoch: [180][ 420/ 1200] Overall Loss 0.133329 Objective Loss 0.133329 LR 0.000125 Time 0.021848 -2022-12-06 11:43:12,056 - Epoch: [180][ 430/ 1200] Overall Loss 0.133285 Objective Loss 0.133285 LR 0.000125 Time 0.021797 -2022-12-06 11:43:12,257 - Epoch: [180][ 440/ 1200] Overall Loss 0.133243 Objective Loss 0.133243 LR 0.000125 Time 0.021756 -2022-12-06 11:43:12,453 - Epoch: [180][ 450/ 1200] Overall Loss 0.133473 Objective Loss 0.133473 LR 0.000125 Time 0.021708 -2022-12-06 11:43:12,654 - Epoch: [180][ 460/ 1200] Overall Loss 0.133332 Objective Loss 0.133332 LR 0.000125 Time 0.021671 -2022-12-06 11:43:12,851 - Epoch: [180][ 470/ 1200] Overall Loss 0.133313 Objective Loss 0.133313 LR 0.000125 Time 0.021629 -2022-12-06 11:43:13,051 - Epoch: [180][ 480/ 1200] Overall Loss 0.133092 Objective Loss 0.133092 LR 0.000125 Time 0.021592 -2022-12-06 11:43:13,248 - Epoch: [180][ 490/ 1200] Overall Loss 0.132877 Objective Loss 0.132877 LR 0.000125 Time 0.021554 -2022-12-06 11:43:13,449 - Epoch: [180][ 500/ 1200] Overall Loss 0.132960 Objective Loss 0.132960 LR 0.000125 Time 0.021522 -2022-12-06 11:43:13,646 - Epoch: [180][ 510/ 1200] Overall Loss 0.133077 Objective Loss 0.133077 LR 0.000125 Time 0.021487 -2022-12-06 11:43:13,846 - Epoch: [180][ 520/ 1200] Overall Loss 0.133079 Objective Loss 0.133079 LR 0.000125 Time 0.021457 -2022-12-06 11:43:14,045 - Epoch: [180][ 530/ 1200] Overall Loss 0.133110 Objective Loss 0.133110 LR 0.000125 Time 0.021425 -2022-12-06 11:43:14,244 - Epoch: [180][ 540/ 1200] Overall Loss 0.133248 Objective Loss 0.133248 LR 0.000125 Time 0.021397 -2022-12-06 11:43:14,442 - Epoch: [180][ 550/ 1200] Overall Loss 0.133365 Objective Loss 0.133365 LR 0.000125 Time 0.021366 -2022-12-06 11:43:14,642 - Epoch: [180][ 560/ 1200] Overall Loss 0.133259 Objective Loss 0.133259 LR 0.000125 Time 0.021341 -2022-12-06 11:43:14,839 - Epoch: [180][ 570/ 1200] Overall Loss 0.133460 Objective Loss 0.133460 LR 0.000125 Time 0.021311 -2022-12-06 11:43:15,040 - Epoch: [180][ 580/ 1200] Overall Loss 0.133343 Objective Loss 0.133343 LR 0.000125 Time 0.021290 -2022-12-06 11:43:15,238 - Epoch: [180][ 590/ 1200] Overall Loss 0.133174 Objective Loss 0.133174 LR 0.000125 Time 0.021263 -2022-12-06 11:43:15,438 - Epoch: [180][ 600/ 1200] Overall Loss 0.133010 Objective Loss 0.133010 LR 0.000125 Time 0.021242 -2022-12-06 11:43:15,634 - Epoch: [180][ 610/ 1200] Overall Loss 0.133022 Objective Loss 0.133022 LR 0.000125 Time 0.021214 -2022-12-06 11:43:15,835 - Epoch: [180][ 620/ 1200] Overall Loss 0.133164 Objective Loss 0.133164 LR 0.000125 Time 0.021194 -2022-12-06 11:43:16,032 - Epoch: [180][ 630/ 1200] Overall Loss 0.133159 Objective Loss 0.133159 LR 0.000125 Time 0.021171 -2022-12-06 11:43:16,232 - Epoch: [180][ 640/ 1200] Overall Loss 0.133073 Objective Loss 0.133073 LR 0.000125 Time 0.021152 -2022-12-06 11:43:16,430 - Epoch: [180][ 650/ 1200] Overall Loss 0.132786 Objective Loss 0.132786 LR 0.000125 Time 0.021130 -2022-12-06 11:43:16,630 - Epoch: [180][ 660/ 1200] Overall Loss 0.132962 Objective Loss 0.132962 LR 0.000125 Time 0.021112 -2022-12-06 11:43:16,827 - Epoch: [180][ 670/ 1200] Overall Loss 0.132901 Objective Loss 0.132901 LR 0.000125 Time 0.021090 -2022-12-06 11:43:17,027 - Epoch: [180][ 680/ 1200] Overall Loss 0.132925 Objective Loss 0.132925 LR 0.000125 Time 0.021073 -2022-12-06 11:43:17,224 - Epoch: [180][ 690/ 1200] Overall Loss 0.132908 Objective Loss 0.132908 LR 0.000125 Time 0.021052 -2022-12-06 11:43:17,424 - Epoch: [180][ 700/ 1200] Overall Loss 0.132804 Objective Loss 0.132804 LR 0.000125 Time 0.021037 -2022-12-06 11:43:17,622 - Epoch: [180][ 710/ 1200] Overall Loss 0.132719 Objective Loss 0.132719 LR 0.000125 Time 0.021019 -2022-12-06 11:43:17,823 - Epoch: [180][ 720/ 1200] Overall Loss 0.132659 Objective Loss 0.132659 LR 0.000125 Time 0.021004 -2022-12-06 11:43:18,019 - Epoch: [180][ 730/ 1200] Overall Loss 0.132644 Objective Loss 0.132644 LR 0.000125 Time 0.020984 -2022-12-06 11:43:18,219 - Epoch: [180][ 740/ 1200] Overall Loss 0.132581 Objective Loss 0.132581 LR 0.000125 Time 0.020971 -2022-12-06 11:43:18,416 - Epoch: [180][ 750/ 1200] Overall Loss 0.132694 Objective Loss 0.132694 LR 0.000125 Time 0.020953 -2022-12-06 11:43:18,615 - Epoch: [180][ 760/ 1200] Overall Loss 0.132677 Objective Loss 0.132677 LR 0.000125 Time 0.020939 -2022-12-06 11:43:18,812 - Epoch: [180][ 770/ 1200] Overall Loss 0.132713 Objective Loss 0.132713 LR 0.000125 Time 0.020922 -2022-12-06 11:43:19,012 - Epoch: [180][ 780/ 1200] Overall Loss 0.133072 Objective Loss 0.133072 LR 0.000125 Time 0.020909 -2022-12-06 11:43:19,209 - Epoch: [180][ 790/ 1200] Overall Loss 0.133140 Objective Loss 0.133140 LR 0.000125 Time 0.020893 -2022-12-06 11:43:19,409 - Epoch: [180][ 800/ 1200] Overall Loss 0.133318 Objective Loss 0.133318 LR 0.000125 Time 0.020882 -2022-12-06 11:43:19,606 - Epoch: [180][ 810/ 1200] Overall Loss 0.133265 Objective Loss 0.133265 LR 0.000125 Time 0.020866 -2022-12-06 11:43:19,806 - Epoch: [180][ 820/ 1200] Overall Loss 0.133224 Objective Loss 0.133224 LR 0.000125 Time 0.020855 -2022-12-06 11:43:20,004 - Epoch: [180][ 830/ 1200] Overall Loss 0.133162 Objective Loss 0.133162 LR 0.000125 Time 0.020841 -2022-12-06 11:43:20,204 - Epoch: [180][ 840/ 1200] Overall Loss 0.133294 Objective Loss 0.133294 LR 0.000125 Time 0.020831 -2022-12-06 11:43:20,401 - Epoch: [180][ 850/ 1200] Overall Loss 0.133469 Objective Loss 0.133469 LR 0.000125 Time 0.020816 -2022-12-06 11:43:20,601 - Epoch: [180][ 860/ 1200] Overall Loss 0.133228 Objective Loss 0.133228 LR 0.000125 Time 0.020807 -2022-12-06 11:43:20,798 - Epoch: [180][ 870/ 1200] Overall Loss 0.133255 Objective Loss 0.133255 LR 0.000125 Time 0.020793 -2022-12-06 11:43:20,998 - Epoch: [180][ 880/ 1200] Overall Loss 0.133198 Objective Loss 0.133198 LR 0.000125 Time 0.020784 -2022-12-06 11:43:21,196 - Epoch: [180][ 890/ 1200] Overall Loss 0.133174 Objective Loss 0.133174 LR 0.000125 Time 0.020772 -2022-12-06 11:43:21,395 - Epoch: [180][ 900/ 1200] Overall Loss 0.133245 Objective Loss 0.133245 LR 0.000125 Time 0.020762 -2022-12-06 11:43:21,592 - Epoch: [180][ 910/ 1200] Overall Loss 0.133066 Objective Loss 0.133066 LR 0.000125 Time 0.020749 -2022-12-06 11:43:21,791 - Epoch: [180][ 920/ 1200] Overall Loss 0.132933 Objective Loss 0.132933 LR 0.000125 Time 0.020739 -2022-12-06 11:43:21,988 - Epoch: [180][ 930/ 1200] Overall Loss 0.133072 Objective Loss 0.133072 LR 0.000125 Time 0.020728 -2022-12-06 11:43:22,189 - Epoch: [180][ 940/ 1200] Overall Loss 0.132868 Objective Loss 0.132868 LR 0.000125 Time 0.020721 -2022-12-06 11:43:22,386 - Epoch: [180][ 950/ 1200] Overall Loss 0.132810 Objective Loss 0.132810 LR 0.000125 Time 0.020710 -2022-12-06 11:43:22,587 - Epoch: [180][ 960/ 1200] Overall Loss 0.132953 Objective Loss 0.132953 LR 0.000125 Time 0.020703 -2022-12-06 11:43:22,785 - Epoch: [180][ 970/ 1200] Overall Loss 0.132828 Objective Loss 0.132828 LR 0.000125 Time 0.020692 -2022-12-06 11:43:22,985 - Epoch: [180][ 980/ 1200] Overall Loss 0.132735 Objective Loss 0.132735 LR 0.000125 Time 0.020685 -2022-12-06 11:43:23,182 - Epoch: [180][ 990/ 1200] Overall Loss 0.133026 Objective Loss 0.133026 LR 0.000125 Time 0.020674 -2022-12-06 11:43:23,382 - Epoch: [180][ 1000/ 1200] Overall Loss 0.133110 Objective Loss 0.133110 LR 0.000125 Time 0.020667 -2022-12-06 11:43:23,579 - Epoch: [180][ 1010/ 1200] Overall Loss 0.133021 Objective Loss 0.133021 LR 0.000125 Time 0.020657 -2022-12-06 11:43:23,779 - Epoch: [180][ 1020/ 1200] Overall Loss 0.133020 Objective Loss 0.133020 LR 0.000125 Time 0.020650 -2022-12-06 11:43:23,976 - Epoch: [180][ 1030/ 1200] Overall Loss 0.132946 Objective Loss 0.132946 LR 0.000125 Time 0.020640 -2022-12-06 11:43:24,175 - Epoch: [180][ 1040/ 1200] Overall Loss 0.132872 Objective Loss 0.132872 LR 0.000125 Time 0.020633 -2022-12-06 11:43:24,372 - Epoch: [180][ 1050/ 1200] Overall Loss 0.132804 Objective Loss 0.132804 LR 0.000125 Time 0.020623 -2022-12-06 11:43:24,572 - Epoch: [180][ 1060/ 1200] Overall Loss 0.132696 Objective Loss 0.132696 LR 0.000125 Time 0.020617 -2022-12-06 11:43:24,769 - Epoch: [180][ 1070/ 1200] Overall Loss 0.132748 Objective Loss 0.132748 LR 0.000125 Time 0.020607 -2022-12-06 11:43:24,969 - Epoch: [180][ 1080/ 1200] Overall Loss 0.132723 Objective Loss 0.132723 LR 0.000125 Time 0.020601 -2022-12-06 11:43:25,165 - Epoch: [180][ 1090/ 1200] Overall Loss 0.132850 Objective Loss 0.132850 LR 0.000125 Time 0.020592 -2022-12-06 11:43:25,366 - Epoch: [180][ 1100/ 1200] Overall Loss 0.132833 Objective Loss 0.132833 LR 0.000125 Time 0.020587 -2022-12-06 11:43:25,562 - Epoch: [180][ 1110/ 1200] Overall Loss 0.133052 Objective Loss 0.133052 LR 0.000125 Time 0.020578 -2022-12-06 11:43:25,762 - Epoch: [180][ 1120/ 1200] Overall Loss 0.133091 Objective Loss 0.133091 LR 0.000125 Time 0.020572 -2022-12-06 11:43:25,960 - Epoch: [180][ 1130/ 1200] Overall Loss 0.133071 Objective Loss 0.133071 LR 0.000125 Time 0.020564 -2022-12-06 11:43:26,160 - Epoch: [180][ 1140/ 1200] Overall Loss 0.133066 Objective Loss 0.133066 LR 0.000125 Time 0.020559 -2022-12-06 11:43:26,358 - Epoch: [180][ 1150/ 1200] Overall Loss 0.133143 Objective Loss 0.133143 LR 0.000125 Time 0.020552 -2022-12-06 11:43:26,560 - Epoch: [180][ 1160/ 1200] Overall Loss 0.133072 Objective Loss 0.133072 LR 0.000125 Time 0.020548 -2022-12-06 11:43:26,757 - Epoch: [180][ 1170/ 1200] Overall Loss 0.132879 Objective Loss 0.132879 LR 0.000125 Time 0.020541 -2022-12-06 11:43:26,958 - Epoch: [180][ 1180/ 1200] Overall Loss 0.133007 Objective Loss 0.133007 LR 0.000125 Time 0.020536 -2022-12-06 11:43:27,156 - Epoch: [180][ 1190/ 1200] Overall Loss 0.132820 Objective Loss 0.132820 LR 0.000125 Time 0.020529 -2022-12-06 11:43:27,380 - Epoch: [180][ 1200/ 1200] Overall Loss 0.132629 Objective Loss 0.132629 Top1 91.631799 Top5 99.163180 LR 0.000125 Time 0.020545 -2022-12-06 11:43:27,469 - --- validate (epoch=180)----------- -2022-12-06 11:43:27,469 - 34129 samples (256 per mini-batch) -2022-12-06 11:43:27,934 - Epoch: [180][ 10/ 134] Loss 0.222765 Top1 87.851562 Top5 98.828125 -2022-12-06 11:43:28,080 - Epoch: [180][ 20/ 134] Loss 0.223486 Top1 87.597656 Top5 98.945312 -2022-12-06 11:43:28,215 - Epoch: [180][ 30/ 134] Loss 0.223122 Top1 87.851562 Top5 98.828125 -2022-12-06 11:43:28,347 - Epoch: [180][ 40/ 134] Loss 0.224498 Top1 87.988281 Top5 98.769531 -2022-12-06 11:43:28,477 - Epoch: [180][ 50/ 134] Loss 0.223248 Top1 87.929688 Top5 98.695312 -2022-12-06 11:43:28,608 - Epoch: [180][ 60/ 134] Loss 0.219374 Top1 88.007812 Top5 98.723958 -2022-12-06 11:43:28,737 - Epoch: [180][ 70/ 134] Loss 0.220938 Top1 88.002232 Top5 98.722098 -2022-12-06 11:43:28,870 - Epoch: [180][ 80/ 134] Loss 0.221844 Top1 87.822266 Top5 98.720703 -2022-12-06 11:43:29,010 - Epoch: [180][ 90/ 134] Loss 0.220027 Top1 87.942708 Top5 98.728299 -2022-12-06 11:43:29,151 - Epoch: [180][ 100/ 134] Loss 0.222284 Top1 87.957031 Top5 98.687500 -2022-12-06 11:43:29,300 - Epoch: [180][ 110/ 134] Loss 0.226351 Top1 87.936790 Top5 98.657670 -2022-12-06 11:43:29,442 - Epoch: [180][ 120/ 134] Loss 0.224673 Top1 87.952474 Top5 98.681641 -2022-12-06 11:43:29,586 - Epoch: [180][ 130/ 134] Loss 0.225720 Top1 87.956731 Top5 98.704928 -2022-12-06 11:43:29,624 - Epoch: [180][ 134/ 134] Loss 0.226830 Top1 87.945735 Top5 98.690263 -2022-12-06 11:43:29,711 - ==> Top1: 87.946 Top5: 98.690 Loss: 0.227 - -2022-12-06 11:43:29,712 - ==> Confusion: -[[ 924 1 1 1 3 6 1 0 3 40 0 1 1 2 4 2 3 0 1 0 2] - [ 1 939 1 2 9 24 2 12 3 1 2 6 0 0 1 0 2 1 11 4 6] - [ 5 3 1022 9 3 2 11 10 0 3 4 6 4 2 2 1 0 0 3 3 10] - [ 2 1 18 948 0 3 0 0 0 1 10 0 5 2 10 1 1 2 10 0 6] - [ 6 3 1 0 969 1 1 2 1 6 1 1 1 2 5 7 5 3 1 0 4] - [ 0 10 1 2 7 978 2 17 2 3 1 15 2 14 2 1 2 1 0 6 3] - [ 2 0 8 3 1 0 1081 4 0 0 0 2 0 1 0 4 0 3 0 7 2] - [ 1 9 5 2 2 24 9 954 0 0 0 6 0 1 1 2 1 1 14 14 8] - [ 5 2 0 0 0 1 1 0 991 39 8 1 1 5 6 0 1 0 1 1 1] - [ 52 1 0 0 8 4 0 3 19 899 1 1 0 8 2 2 0 0 0 0 1] - [ 1 1 3 2 1 1 2 3 8 1 966 0 0 12 5 1 0 0 4 0 8] - [ 1 1 1 0 0 7 4 1 2 1 0 988 22 2 0 6 3 4 0 4 4] - [ 0 0 2 1 1 2 0 0 0 0 0 25 917 1 1 5 1 5 1 3 4] - [ 2 1 2 0 1 5 0 1 10 8 3 9 3 961 1 2 4 0 0 2 8] - [ 7 3 1 9 4 2 0 0 14 3 0 1 1 3 1066 0 0 1 5 2 8] - [ 1 0 0 1 2 0 2 0 1 0 1 6 5 2 0 1001 3 12 0 3 3] - [ 2 2 1 1 2 1 0 0 1 0 0 0 3 2 0 13 1032 1 1 5 5] - [ 2 0 1 1 0 1 0 0 0 4 0 9 17 1 2 12 0 983 0 0 3] - [ 2 4 3 4 1 4 0 20 3 1 2 3 3 0 4 0 0 1 948 2 3] - [ 1 2 1 2 0 4 5 2 0 1 2 12 7 6 0 5 2 0 1 1021 6] - [ 120 189 151 86 103 146 73 134 81 88 147 89 297 251 120 122 162 82 144 214 10427]] - -2022-12-06 11:43:30,290 - ==> Best [Top1: 88.426 Top5: 98.594 Sparsity:0.00 Params: 5376 on epoch: 173] -2022-12-06 11:43:30,291 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:43:30,297 - - -2022-12-06 11:43:30,297 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:43:31,228 - Epoch: [181][ 10/ 1200] Overall Loss 0.130580 Objective Loss 0.130580 LR 0.000125 Time 0.093060 -2022-12-06 11:43:31,427 - Epoch: [181][ 20/ 1200] Overall Loss 0.126214 Objective Loss 0.126214 LR 0.000125 Time 0.056428 -2022-12-06 11:43:31,618 - Epoch: [181][ 30/ 1200] Overall Loss 0.133870 Objective Loss 0.133870 LR 0.000125 Time 0.043991 -2022-12-06 11:43:31,811 - Epoch: [181][ 40/ 1200] Overall Loss 0.133387 Objective Loss 0.133387 LR 0.000125 Time 0.037787 -2022-12-06 11:43:32,002 - Epoch: [181][ 50/ 1200] Overall Loss 0.134156 Objective Loss 0.134156 LR 0.000125 Time 0.034049 -2022-12-06 11:43:32,194 - Epoch: [181][ 60/ 1200] Overall Loss 0.131546 Objective Loss 0.131546 LR 0.000125 Time 0.031560 -2022-12-06 11:43:32,386 - Epoch: [181][ 70/ 1200] Overall Loss 0.132128 Objective Loss 0.132128 LR 0.000125 Time 0.029791 -2022-12-06 11:43:32,578 - Epoch: [181][ 80/ 1200] Overall Loss 0.129879 Objective Loss 0.129879 LR 0.000125 Time 0.028461 -2022-12-06 11:43:32,770 - Epoch: [181][ 90/ 1200] Overall Loss 0.129843 Objective Loss 0.129843 LR 0.000125 Time 0.027424 -2022-12-06 11:43:32,963 - Epoch: [181][ 100/ 1200] Overall Loss 0.128438 Objective Loss 0.128438 LR 0.000125 Time 0.026602 -2022-12-06 11:43:33,155 - Epoch: [181][ 110/ 1200] Overall Loss 0.129470 Objective Loss 0.129470 LR 0.000125 Time 0.025930 -2022-12-06 11:43:33,347 - Epoch: [181][ 120/ 1200] Overall Loss 0.129886 Objective Loss 0.129886 LR 0.000125 Time 0.025360 -2022-12-06 11:43:33,539 - Epoch: [181][ 130/ 1200] Overall Loss 0.130899 Objective Loss 0.130899 LR 0.000125 Time 0.024883 -2022-12-06 11:43:33,730 - Epoch: [181][ 140/ 1200] Overall Loss 0.131345 Objective Loss 0.131345 LR 0.000125 Time 0.024469 -2022-12-06 11:43:33,922 - Epoch: [181][ 150/ 1200] Overall Loss 0.131044 Objective Loss 0.131044 LR 0.000125 Time 0.024116 -2022-12-06 11:43:34,115 - Epoch: [181][ 160/ 1200] Overall Loss 0.129797 Objective Loss 0.129797 LR 0.000125 Time 0.023807 -2022-12-06 11:43:34,306 - Epoch: [181][ 170/ 1200] Overall Loss 0.129904 Objective Loss 0.129904 LR 0.000125 Time 0.023529 -2022-12-06 11:43:34,499 - Epoch: [181][ 180/ 1200] Overall Loss 0.130128 Objective Loss 0.130128 LR 0.000125 Time 0.023288 -2022-12-06 11:43:34,691 - Epoch: [181][ 190/ 1200] Overall Loss 0.130298 Objective Loss 0.130298 LR 0.000125 Time 0.023070 -2022-12-06 11:43:34,883 - Epoch: [181][ 200/ 1200] Overall Loss 0.130805 Objective Loss 0.130805 LR 0.000125 Time 0.022874 -2022-12-06 11:43:35,074 - Epoch: [181][ 210/ 1200] Overall Loss 0.129803 Objective Loss 0.129803 LR 0.000125 Time 0.022696 -2022-12-06 11:43:35,267 - Epoch: [181][ 220/ 1200] Overall Loss 0.129705 Objective Loss 0.129705 LR 0.000125 Time 0.022535 -2022-12-06 11:43:35,458 - Epoch: [181][ 230/ 1200] Overall Loss 0.129934 Objective Loss 0.129934 LR 0.000125 Time 0.022386 -2022-12-06 11:43:35,650 - Epoch: [181][ 240/ 1200] Overall Loss 0.130592 Objective Loss 0.130592 LR 0.000125 Time 0.022251 -2022-12-06 11:43:35,841 - Epoch: [181][ 250/ 1200] Overall Loss 0.130835 Objective Loss 0.130835 LR 0.000125 Time 0.022123 -2022-12-06 11:43:36,033 - Epoch: [181][ 260/ 1200] Overall Loss 0.131081 Objective Loss 0.131081 LR 0.000125 Time 0.022007 -2022-12-06 11:43:36,225 - Epoch: [181][ 270/ 1200] Overall Loss 0.130927 Objective Loss 0.130927 LR 0.000125 Time 0.021901 -2022-12-06 11:43:36,416 - Epoch: [181][ 280/ 1200] Overall Loss 0.130513 Objective Loss 0.130513 LR 0.000125 Time 0.021800 -2022-12-06 11:43:36,608 - Epoch: [181][ 290/ 1200] Overall Loss 0.130694 Objective Loss 0.130694 LR 0.000125 Time 0.021708 -2022-12-06 11:43:36,800 - Epoch: [181][ 300/ 1200] Overall Loss 0.131232 Objective Loss 0.131232 LR 0.000125 Time 0.021623 -2022-12-06 11:43:36,992 - Epoch: [181][ 310/ 1200] Overall Loss 0.131516 Objective Loss 0.131516 LR 0.000125 Time 0.021541 -2022-12-06 11:43:37,184 - Epoch: [181][ 320/ 1200] Overall Loss 0.131588 Objective Loss 0.131588 LR 0.000125 Time 0.021467 -2022-12-06 11:43:37,376 - Epoch: [181][ 330/ 1200] Overall Loss 0.131200 Objective Loss 0.131200 LR 0.000125 Time 0.021397 -2022-12-06 11:43:37,568 - Epoch: [181][ 340/ 1200] Overall Loss 0.131375 Objective Loss 0.131375 LR 0.000125 Time 0.021331 -2022-12-06 11:43:37,760 - Epoch: [181][ 350/ 1200] Overall Loss 0.130526 Objective Loss 0.130526 LR 0.000125 Time 0.021268 -2022-12-06 11:43:37,952 - Epoch: [181][ 360/ 1200] Overall Loss 0.131171 Objective Loss 0.131171 LR 0.000125 Time 0.021208 -2022-12-06 11:43:38,143 - Epoch: [181][ 370/ 1200] Overall Loss 0.131329 Objective Loss 0.131329 LR 0.000125 Time 0.021152 -2022-12-06 11:43:38,335 - Epoch: [181][ 380/ 1200] Overall Loss 0.131025 Objective Loss 0.131025 LR 0.000125 Time 0.021098 -2022-12-06 11:43:38,527 - Epoch: [181][ 390/ 1200] Overall Loss 0.130301 Objective Loss 0.130301 LR 0.000125 Time 0.021047 -2022-12-06 11:43:38,718 - Epoch: [181][ 400/ 1200] Overall Loss 0.129797 Objective Loss 0.129797 LR 0.000125 Time 0.020998 -2022-12-06 11:43:38,910 - Epoch: [181][ 410/ 1200] Overall Loss 0.129691 Objective Loss 0.129691 LR 0.000125 Time 0.020953 -2022-12-06 11:43:39,102 - Epoch: [181][ 420/ 1200] Overall Loss 0.129631 Objective Loss 0.129631 LR 0.000125 Time 0.020910 -2022-12-06 11:43:39,294 - Epoch: [181][ 430/ 1200] Overall Loss 0.129702 Objective Loss 0.129702 LR 0.000125 Time 0.020869 -2022-12-06 11:43:39,486 - Epoch: [181][ 440/ 1200] Overall Loss 0.129446 Objective Loss 0.129446 LR 0.000125 Time 0.020830 -2022-12-06 11:43:39,678 - Epoch: [181][ 450/ 1200] Overall Loss 0.129524 Objective Loss 0.129524 LR 0.000125 Time 0.020793 -2022-12-06 11:43:39,871 - Epoch: [181][ 460/ 1200] Overall Loss 0.129603 Objective Loss 0.129603 LR 0.000125 Time 0.020758 -2022-12-06 11:43:40,062 - Epoch: [181][ 470/ 1200] Overall Loss 0.129786 Objective Loss 0.129786 LR 0.000125 Time 0.020722 -2022-12-06 11:43:40,254 - Epoch: [181][ 480/ 1200] Overall Loss 0.130036 Objective Loss 0.130036 LR 0.000125 Time 0.020688 -2022-12-06 11:43:40,446 - Epoch: [181][ 490/ 1200] Overall Loss 0.130051 Objective Loss 0.130051 LR 0.000125 Time 0.020657 -2022-12-06 11:43:40,638 - Epoch: [181][ 500/ 1200] Overall Loss 0.130375 Objective Loss 0.130375 LR 0.000125 Time 0.020627 -2022-12-06 11:43:40,830 - Epoch: [181][ 510/ 1200] Overall Loss 0.130237 Objective Loss 0.130237 LR 0.000125 Time 0.020598 -2022-12-06 11:43:41,022 - Epoch: [181][ 520/ 1200] Overall Loss 0.129979 Objective Loss 0.129979 LR 0.000125 Time 0.020570 -2022-12-06 11:43:41,214 - Epoch: [181][ 530/ 1200] Overall Loss 0.129931 Objective Loss 0.129931 LR 0.000125 Time 0.020543 -2022-12-06 11:43:41,406 - Epoch: [181][ 540/ 1200] Overall Loss 0.129882 Objective Loss 0.129882 LR 0.000125 Time 0.020517 -2022-12-06 11:43:41,597 - Epoch: [181][ 550/ 1200] Overall Loss 0.129994 Objective Loss 0.129994 LR 0.000125 Time 0.020491 -2022-12-06 11:43:41,788 - Epoch: [181][ 560/ 1200] Overall Loss 0.129858 Objective Loss 0.129858 LR 0.000125 Time 0.020466 -2022-12-06 11:43:41,980 - Epoch: [181][ 570/ 1200] Overall Loss 0.130127 Objective Loss 0.130127 LR 0.000125 Time 0.020442 -2022-12-06 11:43:42,172 - Epoch: [181][ 580/ 1200] Overall Loss 0.130261 Objective Loss 0.130261 LR 0.000125 Time 0.020419 -2022-12-06 11:43:42,363 - Epoch: [181][ 590/ 1200] Overall Loss 0.130306 Objective Loss 0.130306 LR 0.000125 Time 0.020397 -2022-12-06 11:43:42,555 - Epoch: [181][ 600/ 1200] Overall Loss 0.130364 Objective Loss 0.130364 LR 0.000125 Time 0.020375 -2022-12-06 11:43:42,747 - Epoch: [181][ 610/ 1200] Overall Loss 0.130499 Objective Loss 0.130499 LR 0.000125 Time 0.020355 -2022-12-06 11:43:42,938 - Epoch: [181][ 620/ 1200] Overall Loss 0.130669 Objective Loss 0.130669 LR 0.000125 Time 0.020334 -2022-12-06 11:43:43,130 - Epoch: [181][ 630/ 1200] Overall Loss 0.130807 Objective Loss 0.130807 LR 0.000125 Time 0.020315 -2022-12-06 11:43:43,321 - Epoch: [181][ 640/ 1200] Overall Loss 0.130413 Objective Loss 0.130413 LR 0.000125 Time 0.020295 -2022-12-06 11:43:43,512 - Epoch: [181][ 650/ 1200] Overall Loss 0.130669 Objective Loss 0.130669 LR 0.000125 Time 0.020276 -2022-12-06 11:43:43,704 - Epoch: [181][ 660/ 1200] Overall Loss 0.130870 Objective Loss 0.130870 LR 0.000125 Time 0.020259 -2022-12-06 11:43:43,895 - Epoch: [181][ 670/ 1200] Overall Loss 0.130744 Objective Loss 0.130744 LR 0.000125 Time 0.020242 -2022-12-06 11:43:44,087 - Epoch: [181][ 680/ 1200] Overall Loss 0.130717 Objective Loss 0.130717 LR 0.000125 Time 0.020225 -2022-12-06 11:43:44,278 - Epoch: [181][ 690/ 1200] Overall Loss 0.130917 Objective Loss 0.130917 LR 0.000125 Time 0.020208 -2022-12-06 11:43:44,470 - Epoch: [181][ 700/ 1200] Overall Loss 0.131371 Objective Loss 0.131371 LR 0.000125 Time 0.020192 -2022-12-06 11:43:44,662 - Epoch: [181][ 710/ 1200] Overall Loss 0.131197 Objective Loss 0.131197 LR 0.000125 Time 0.020177 -2022-12-06 11:43:44,853 - Epoch: [181][ 720/ 1200] Overall Loss 0.131118 Objective Loss 0.131118 LR 0.000125 Time 0.020162 -2022-12-06 11:43:45,045 - Epoch: [181][ 730/ 1200] Overall Loss 0.131060 Objective Loss 0.131060 LR 0.000125 Time 0.020148 -2022-12-06 11:43:45,236 - Epoch: [181][ 740/ 1200] Overall Loss 0.131176 Objective Loss 0.131176 LR 0.000125 Time 0.020133 -2022-12-06 11:43:45,427 - Epoch: [181][ 750/ 1200] Overall Loss 0.131340 Objective Loss 0.131340 LR 0.000125 Time 0.020119 -2022-12-06 11:43:45,619 - Epoch: [181][ 760/ 1200] Overall Loss 0.131095 Objective Loss 0.131095 LR 0.000125 Time 0.020106 -2022-12-06 11:43:45,811 - Epoch: [181][ 770/ 1200] Overall Loss 0.131369 Objective Loss 0.131369 LR 0.000125 Time 0.020093 -2022-12-06 11:43:46,002 - Epoch: [181][ 780/ 1200] Overall Loss 0.131532 Objective Loss 0.131532 LR 0.000125 Time 0.020080 -2022-12-06 11:43:46,194 - Epoch: [181][ 790/ 1200] Overall Loss 0.131502 Objective Loss 0.131502 LR 0.000125 Time 0.020068 -2022-12-06 11:43:46,386 - Epoch: [181][ 800/ 1200] Overall Loss 0.131554 Objective Loss 0.131554 LR 0.000125 Time 0.020056 -2022-12-06 11:43:46,577 - Epoch: [181][ 810/ 1200] Overall Loss 0.131475 Objective Loss 0.131475 LR 0.000125 Time 0.020043 -2022-12-06 11:43:46,768 - Epoch: [181][ 820/ 1200] Overall Loss 0.131453 Objective Loss 0.131453 LR 0.000125 Time 0.020032 -2022-12-06 11:43:46,960 - Epoch: [181][ 830/ 1200] Overall Loss 0.131348 Objective Loss 0.131348 LR 0.000125 Time 0.020021 -2022-12-06 11:43:47,152 - Epoch: [181][ 840/ 1200] Overall Loss 0.131426 Objective Loss 0.131426 LR 0.000125 Time 0.020010 -2022-12-06 11:43:47,343 - Epoch: [181][ 850/ 1200] Overall Loss 0.131435 Objective Loss 0.131435 LR 0.000125 Time 0.019999 -2022-12-06 11:43:47,535 - Epoch: [181][ 860/ 1200] Overall Loss 0.131735 Objective Loss 0.131735 LR 0.000125 Time 0.019989 -2022-12-06 11:43:47,727 - Epoch: [181][ 870/ 1200] Overall Loss 0.131861 Objective Loss 0.131861 LR 0.000125 Time 0.019980 -2022-12-06 11:43:47,919 - Epoch: [181][ 880/ 1200] Overall Loss 0.131873 Objective Loss 0.131873 LR 0.000125 Time 0.019970 -2022-12-06 11:43:48,111 - Epoch: [181][ 890/ 1200] Overall Loss 0.131750 Objective Loss 0.131750 LR 0.000125 Time 0.019960 -2022-12-06 11:43:48,302 - Epoch: [181][ 900/ 1200] Overall Loss 0.131810 Objective Loss 0.131810 LR 0.000125 Time 0.019951 -2022-12-06 11:43:48,493 - Epoch: [181][ 910/ 1200] Overall Loss 0.131940 Objective Loss 0.131940 LR 0.000125 Time 0.019941 -2022-12-06 11:43:48,685 - Epoch: [181][ 920/ 1200] Overall Loss 0.131791 Objective Loss 0.131791 LR 0.000125 Time 0.019932 -2022-12-06 11:43:48,876 - Epoch: [181][ 930/ 1200] Overall Loss 0.131741 Objective Loss 0.131741 LR 0.000125 Time 0.019923 -2022-12-06 11:43:49,068 - Epoch: [181][ 940/ 1200] Overall Loss 0.131961 Objective Loss 0.131961 LR 0.000125 Time 0.019914 -2022-12-06 11:43:49,259 - Epoch: [181][ 950/ 1200] Overall Loss 0.131966 Objective Loss 0.131966 LR 0.000125 Time 0.019905 -2022-12-06 11:43:49,451 - Epoch: [181][ 960/ 1200] Overall Loss 0.131843 Objective Loss 0.131843 LR 0.000125 Time 0.019897 -2022-12-06 11:43:49,642 - Epoch: [181][ 970/ 1200] Overall Loss 0.131858 Objective Loss 0.131858 LR 0.000125 Time 0.019889 -2022-12-06 11:43:49,834 - Epoch: [181][ 980/ 1200] Overall Loss 0.131909 Objective Loss 0.131909 LR 0.000125 Time 0.019881 -2022-12-06 11:43:50,026 - Epoch: [181][ 990/ 1200] Overall Loss 0.131747 Objective Loss 0.131747 LR 0.000125 Time 0.019874 -2022-12-06 11:43:50,218 - Epoch: [181][ 1000/ 1200] Overall Loss 0.131608 Objective Loss 0.131608 LR 0.000125 Time 0.019866 -2022-12-06 11:43:50,410 - Epoch: [181][ 1010/ 1200] Overall Loss 0.131733 Objective Loss 0.131733 LR 0.000125 Time 0.019859 -2022-12-06 11:43:50,602 - Epoch: [181][ 1020/ 1200] Overall Loss 0.131787 Objective Loss 0.131787 LR 0.000125 Time 0.019852 -2022-12-06 11:43:50,794 - Epoch: [181][ 1030/ 1200] Overall Loss 0.131791 Objective Loss 0.131791 LR 0.000125 Time 0.019845 -2022-12-06 11:43:50,986 - Epoch: [181][ 1040/ 1200] Overall Loss 0.131971 Objective Loss 0.131971 LR 0.000125 Time 0.019839 -2022-12-06 11:43:51,179 - Epoch: [181][ 1050/ 1200] Overall Loss 0.131692 Objective Loss 0.131692 LR 0.000125 Time 0.019832 -2022-12-06 11:43:51,371 - Epoch: [181][ 1060/ 1200] Overall Loss 0.131629 Objective Loss 0.131629 LR 0.000125 Time 0.019826 -2022-12-06 11:43:51,564 - Epoch: [181][ 1070/ 1200] Overall Loss 0.131522 Objective Loss 0.131522 LR 0.000125 Time 0.019821 -2022-12-06 11:43:51,757 - Epoch: [181][ 1080/ 1200] Overall Loss 0.131422 Objective Loss 0.131422 LR 0.000125 Time 0.019815 -2022-12-06 11:43:51,949 - Epoch: [181][ 1090/ 1200] Overall Loss 0.131431 Objective Loss 0.131431 LR 0.000125 Time 0.019809 -2022-12-06 11:43:52,142 - Epoch: [181][ 1100/ 1200] Overall Loss 0.131529 Objective Loss 0.131529 LR 0.000125 Time 0.019804 -2022-12-06 11:43:52,334 - Epoch: [181][ 1110/ 1200] Overall Loss 0.131555 Objective Loss 0.131555 LR 0.000125 Time 0.019799 -2022-12-06 11:43:52,527 - Epoch: [181][ 1120/ 1200] Overall Loss 0.131513 Objective Loss 0.131513 LR 0.000125 Time 0.019793 -2022-12-06 11:43:52,720 - Epoch: [181][ 1130/ 1200] Overall Loss 0.131424 Objective Loss 0.131424 LR 0.000125 Time 0.019788 -2022-12-06 11:43:52,913 - Epoch: [181][ 1140/ 1200] Overall Loss 0.131436 Objective Loss 0.131436 LR 0.000125 Time 0.019783 -2022-12-06 11:43:53,105 - Epoch: [181][ 1150/ 1200] Overall Loss 0.131475 Objective Loss 0.131475 LR 0.000125 Time 0.019779 -2022-12-06 11:43:53,298 - Epoch: [181][ 1160/ 1200] Overall Loss 0.131504 Objective Loss 0.131504 LR 0.000125 Time 0.019773 -2022-12-06 11:43:53,490 - Epoch: [181][ 1170/ 1200] Overall Loss 0.131495 Objective Loss 0.131495 LR 0.000125 Time 0.019768 -2022-12-06 11:43:53,683 - Epoch: [181][ 1180/ 1200] Overall Loss 0.131509 Objective Loss 0.131509 LR 0.000125 Time 0.019764 -2022-12-06 11:43:53,875 - Epoch: [181][ 1190/ 1200] Overall Loss 0.131422 Objective Loss 0.131422 LR 0.000125 Time 0.019759 -2022-12-06 11:43:54,110 - Epoch: [181][ 1200/ 1200] Overall Loss 0.131398 Objective Loss 0.131398 Top1 91.004184 Top5 100.000000 LR 0.000125 Time 0.019790 -2022-12-06 11:43:54,199 - --- validate (epoch=181)----------- -2022-12-06 11:43:54,199 - 34129 samples (256 per mini-batch) -2022-12-06 11:43:54,650 - Epoch: [181][ 10/ 134] Loss 0.235794 Top1 88.164062 Top5 98.671875 -2022-12-06 11:43:54,778 - Epoch: [181][ 20/ 134] Loss 0.232156 Top1 88.710938 Top5 98.984375 -2022-12-06 11:43:54,910 - Epoch: [181][ 30/ 134] Loss 0.225064 Top1 88.984375 Top5 98.789062 -2022-12-06 11:43:55,038 - Epoch: [181][ 40/ 134] Loss 0.220894 Top1 88.945312 Top5 98.710938 -2022-12-06 11:43:55,165 - Epoch: [181][ 50/ 134] Loss 0.229596 Top1 88.757812 Top5 98.625000 -2022-12-06 11:43:55,286 - Epoch: [181][ 60/ 134] Loss 0.228304 Top1 88.841146 Top5 98.600260 -2022-12-06 11:43:55,410 - Epoch: [181][ 70/ 134] Loss 0.227543 Top1 88.928571 Top5 98.610491 -2022-12-06 11:43:55,532 - Epoch: [181][ 80/ 134] Loss 0.230733 Top1 88.798828 Top5 98.623047 -2022-12-06 11:43:55,655 - Epoch: [181][ 90/ 134] Loss 0.227205 Top1 88.862847 Top5 98.606771 -2022-12-06 11:43:55,798 - Epoch: [181][ 100/ 134] Loss 0.224062 Top1 88.792969 Top5 98.636719 -2022-12-06 11:43:55,939 - Epoch: [181][ 110/ 134] Loss 0.225393 Top1 88.689631 Top5 98.636364 -2022-12-06 11:43:56,078 - Epoch: [181][ 120/ 134] Loss 0.224303 Top1 88.743490 Top5 98.649089 -2022-12-06 11:43:56,214 - Epoch: [181][ 130/ 134] Loss 0.224916 Top1 88.689904 Top5 98.653846 -2022-12-06 11:43:56,250 - Epoch: [181][ 134/ 134] Loss 0.224302 Top1 88.704621 Top5 98.666823 -2022-12-06 11:43:56,340 - ==> Top1: 88.705 Top5: 98.667 Loss: 0.224 - -2022-12-06 11:43:56,341 - ==> Confusion: -[[ 913 0 0 2 5 7 1 2 5 48 0 1 1 2 3 1 0 1 1 0 3] - [ 1 955 1 2 6 14 2 8 2 2 4 4 2 0 0 1 3 1 9 3 7] - [ 4 2 1019 10 4 2 13 9 0 4 3 2 2 2 3 1 1 0 4 4 14] - [ 4 2 16 949 1 3 1 0 0 1 9 0 4 3 9 0 1 1 10 0 6] - [ 6 4 3 0 965 1 0 2 1 6 1 3 1 1 8 3 6 2 0 0 7] - [ 1 11 1 3 5 992 2 16 2 3 1 11 1 11 1 1 1 1 0 4 1] - [ 1 2 11 2 2 3 1078 3 0 0 0 2 0 0 0 3 0 1 2 7 1] - [ 1 5 8 2 1 26 7 967 0 0 0 5 0 1 1 0 0 0 17 8 5] - [ 4 1 0 1 0 2 1 1 982 43 5 1 2 4 11 1 1 0 1 1 2] - [ 45 0 0 0 4 1 0 3 22 905 1 0 0 9 3 0 0 1 1 0 6] - [ 0 2 4 6 1 1 0 2 9 1 965 1 0 10 3 1 0 0 4 1 8] - [ 1 0 1 0 1 7 3 2 1 1 0 993 15 4 0 4 6 4 0 5 3] - [ 1 0 2 1 1 2 0 0 0 0 0 29 905 0 0 6 1 10 2 3 6] - [ 1 0 2 0 0 7 0 1 8 16 2 4 3 965 1 1 2 0 1 1 8] - [ 5 4 2 7 3 2 0 0 12 1 0 2 1 3 1076 0 0 1 5 1 5] - [ 0 0 0 2 2 0 3 0 2 1 0 9 4 2 0 995 4 9 0 3 7] - [ 1 2 2 1 2 0 0 1 1 0 0 1 3 1 0 11 1030 1 0 6 9] - [ 4 0 1 1 0 1 1 0 0 3 1 8 16 0 1 10 0 986 0 0 3] - [ 3 3 3 6 0 2 0 18 1 1 5 4 2 0 8 0 0 3 944 1 4] - [ 1 3 1 3 0 4 4 6 0 2 2 16 8 6 1 4 3 1 1 1008 6] - [ 117 182 150 79 84 143 68 125 65 83 118 97 289 227 127 82 115 82 128 186 10679]] - -2022-12-06 11:43:57,002 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:43:57,002 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:43:57,009 - - -2022-12-06 11:43:57,009 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:43:57,943 - Epoch: [182][ 10/ 1200] Overall Loss 0.125597 Objective Loss 0.125597 LR 0.000125 Time 0.093330 -2022-12-06 11:43:58,139 - Epoch: [182][ 20/ 1200] Overall Loss 0.125725 Objective Loss 0.125725 LR 0.000125 Time 0.056419 -2022-12-06 11:43:58,330 - Epoch: [182][ 30/ 1200] Overall Loss 0.129638 Objective Loss 0.129638 LR 0.000125 Time 0.043958 -2022-12-06 11:43:58,521 - Epoch: [182][ 40/ 1200] Overall Loss 0.130294 Objective Loss 0.130294 LR 0.000125 Time 0.037737 -2022-12-06 11:43:58,712 - Epoch: [182][ 50/ 1200] Overall Loss 0.126295 Objective Loss 0.126295 LR 0.000125 Time 0.033989 -2022-12-06 11:43:58,903 - Epoch: [182][ 60/ 1200] Overall Loss 0.129597 Objective Loss 0.129597 LR 0.000125 Time 0.031497 -2022-12-06 11:43:59,093 - Epoch: [182][ 70/ 1200] Overall Loss 0.130077 Objective Loss 0.130077 LR 0.000125 Time 0.029709 -2022-12-06 11:43:59,284 - Epoch: [182][ 80/ 1200] Overall Loss 0.132987 Objective Loss 0.132987 LR 0.000125 Time 0.028373 -2022-12-06 11:43:59,474 - Epoch: [182][ 90/ 1200] Overall Loss 0.130005 Objective Loss 0.130005 LR 0.000125 Time 0.027334 -2022-12-06 11:43:59,665 - Epoch: [182][ 100/ 1200] Overall Loss 0.130310 Objective Loss 0.130310 LR 0.000125 Time 0.026499 -2022-12-06 11:43:59,855 - Epoch: [182][ 110/ 1200] Overall Loss 0.130919 Objective Loss 0.130919 LR 0.000125 Time 0.025812 -2022-12-06 11:44:00,046 - Epoch: [182][ 120/ 1200] Overall Loss 0.129814 Objective Loss 0.129814 LR 0.000125 Time 0.025246 -2022-12-06 11:44:00,235 - Epoch: [182][ 130/ 1200] Overall Loss 0.130019 Objective Loss 0.130019 LR 0.000125 Time 0.024760 -2022-12-06 11:44:00,426 - Epoch: [182][ 140/ 1200] Overall Loss 0.129266 Objective Loss 0.129266 LR 0.000125 Time 0.024348 -2022-12-06 11:44:00,616 - Epoch: [182][ 150/ 1200] Overall Loss 0.129000 Objective Loss 0.129000 LR 0.000125 Time 0.023988 -2022-12-06 11:44:00,806 - Epoch: [182][ 160/ 1200] Overall Loss 0.129384 Objective Loss 0.129384 LR 0.000125 Time 0.023676 -2022-12-06 11:44:00,996 - Epoch: [182][ 170/ 1200] Overall Loss 0.130270 Objective Loss 0.130270 LR 0.000125 Time 0.023398 -2022-12-06 11:44:01,187 - Epoch: [182][ 180/ 1200] Overall Loss 0.130502 Objective Loss 0.130502 LR 0.000125 Time 0.023152 -2022-12-06 11:44:01,377 - Epoch: [182][ 190/ 1200] Overall Loss 0.130884 Objective Loss 0.130884 LR 0.000125 Time 0.022931 -2022-12-06 11:44:01,567 - Epoch: [182][ 200/ 1200] Overall Loss 0.131075 Objective Loss 0.131075 LR 0.000125 Time 0.022735 -2022-12-06 11:44:01,757 - Epoch: [182][ 210/ 1200] Overall Loss 0.131040 Objective Loss 0.131040 LR 0.000125 Time 0.022555 -2022-12-06 11:44:01,947 - Epoch: [182][ 220/ 1200] Overall Loss 0.130716 Objective Loss 0.130716 LR 0.000125 Time 0.022388 -2022-12-06 11:44:02,137 - Epoch: [182][ 230/ 1200] Overall Loss 0.131178 Objective Loss 0.131178 LR 0.000125 Time 0.022239 -2022-12-06 11:44:02,327 - Epoch: [182][ 240/ 1200] Overall Loss 0.131238 Objective Loss 0.131238 LR 0.000125 Time 0.022104 -2022-12-06 11:44:02,518 - Epoch: [182][ 250/ 1200] Overall Loss 0.132037 Objective Loss 0.132037 LR 0.000125 Time 0.021979 -2022-12-06 11:44:02,708 - Epoch: [182][ 260/ 1200] Overall Loss 0.131188 Objective Loss 0.131188 LR 0.000125 Time 0.021865 -2022-12-06 11:44:02,899 - Epoch: [182][ 270/ 1200] Overall Loss 0.130656 Objective Loss 0.130656 LR 0.000125 Time 0.021757 -2022-12-06 11:44:03,089 - Epoch: [182][ 280/ 1200] Overall Loss 0.130640 Objective Loss 0.130640 LR 0.000125 Time 0.021658 -2022-12-06 11:44:03,279 - Epoch: [182][ 290/ 1200] Overall Loss 0.131015 Objective Loss 0.131015 LR 0.000125 Time 0.021566 -2022-12-06 11:44:03,470 - Epoch: [182][ 300/ 1200] Overall Loss 0.130653 Objective Loss 0.130653 LR 0.000125 Time 0.021480 -2022-12-06 11:44:03,659 - Epoch: [182][ 310/ 1200] Overall Loss 0.131187 Objective Loss 0.131187 LR 0.000125 Time 0.021397 -2022-12-06 11:44:03,850 - Epoch: [182][ 320/ 1200] Overall Loss 0.131334 Objective Loss 0.131334 LR 0.000125 Time 0.021322 -2022-12-06 11:44:04,040 - Epoch: [182][ 330/ 1200] Overall Loss 0.130873 Objective Loss 0.130873 LR 0.000125 Time 0.021251 -2022-12-06 11:44:04,230 - Epoch: [182][ 340/ 1200] Overall Loss 0.130732 Objective Loss 0.130732 LR 0.000125 Time 0.021183 -2022-12-06 11:44:04,421 - Epoch: [182][ 350/ 1200] Overall Loss 0.130145 Objective Loss 0.130145 LR 0.000125 Time 0.021122 -2022-12-06 11:44:04,612 - Epoch: [182][ 360/ 1200] Overall Loss 0.129969 Objective Loss 0.129969 LR 0.000125 Time 0.021063 -2022-12-06 11:44:04,802 - Epoch: [182][ 370/ 1200] Overall Loss 0.129837 Objective Loss 0.129837 LR 0.000125 Time 0.021006 -2022-12-06 11:44:04,993 - Epoch: [182][ 380/ 1200] Overall Loss 0.129834 Objective Loss 0.129834 LR 0.000125 Time 0.020956 -2022-12-06 11:44:05,183 - Epoch: [182][ 390/ 1200] Overall Loss 0.129912 Objective Loss 0.129912 LR 0.000125 Time 0.020904 -2022-12-06 11:44:05,374 - Epoch: [182][ 400/ 1200] Overall Loss 0.130102 Objective Loss 0.130102 LR 0.000125 Time 0.020856 -2022-12-06 11:44:05,564 - Epoch: [182][ 410/ 1200] Overall Loss 0.129643 Objective Loss 0.129643 LR 0.000125 Time 0.020810 -2022-12-06 11:44:05,755 - Epoch: [182][ 420/ 1200] Overall Loss 0.129369 Objective Loss 0.129369 LR 0.000125 Time 0.020768 -2022-12-06 11:44:05,945 - Epoch: [182][ 430/ 1200] Overall Loss 0.129571 Objective Loss 0.129571 LR 0.000125 Time 0.020726 -2022-12-06 11:44:06,135 - Epoch: [182][ 440/ 1200] Overall Loss 0.129055 Objective Loss 0.129055 LR 0.000125 Time 0.020686 -2022-12-06 11:44:06,326 - Epoch: [182][ 450/ 1200] Overall Loss 0.128857 Objective Loss 0.128857 LR 0.000125 Time 0.020650 -2022-12-06 11:44:06,517 - Epoch: [182][ 460/ 1200] Overall Loss 0.128669 Objective Loss 0.128669 LR 0.000125 Time 0.020614 -2022-12-06 11:44:06,707 - Epoch: [182][ 470/ 1200] Overall Loss 0.128348 Objective Loss 0.128348 LR 0.000125 Time 0.020578 -2022-12-06 11:44:06,897 - Epoch: [182][ 480/ 1200] Overall Loss 0.128561 Objective Loss 0.128561 LR 0.000125 Time 0.020545 -2022-12-06 11:44:07,088 - Epoch: [182][ 490/ 1200] Overall Loss 0.128341 Objective Loss 0.128341 LR 0.000125 Time 0.020513 -2022-12-06 11:44:07,278 - Epoch: [182][ 500/ 1200] Overall Loss 0.128421 Objective Loss 0.128421 LR 0.000125 Time 0.020483 -2022-12-06 11:44:07,469 - Epoch: [182][ 510/ 1200] Overall Loss 0.128668 Objective Loss 0.128668 LR 0.000125 Time 0.020455 -2022-12-06 11:44:07,660 - Epoch: [182][ 520/ 1200] Overall Loss 0.128704 Objective Loss 0.128704 LR 0.000125 Time 0.020428 -2022-12-06 11:44:07,851 - Epoch: [182][ 530/ 1200] Overall Loss 0.128986 Objective Loss 0.128986 LR 0.000125 Time 0.020402 -2022-12-06 11:44:08,042 - Epoch: [182][ 540/ 1200] Overall Loss 0.128958 Objective Loss 0.128958 LR 0.000125 Time 0.020376 -2022-12-06 11:44:08,232 - Epoch: [182][ 550/ 1200] Overall Loss 0.129143 Objective Loss 0.129143 LR 0.000125 Time 0.020350 -2022-12-06 11:44:08,423 - Epoch: [182][ 560/ 1200] Overall Loss 0.129491 Objective Loss 0.129491 LR 0.000125 Time 0.020326 -2022-12-06 11:44:08,613 - Epoch: [182][ 570/ 1200] Overall Loss 0.129640 Objective Loss 0.129640 LR 0.000125 Time 0.020304 -2022-12-06 11:44:08,805 - Epoch: [182][ 580/ 1200] Overall Loss 0.129881 Objective Loss 0.129881 LR 0.000125 Time 0.020282 -2022-12-06 11:44:08,995 - Epoch: [182][ 590/ 1200] Overall Loss 0.130042 Objective Loss 0.130042 LR 0.000125 Time 0.020261 -2022-12-06 11:44:09,186 - Epoch: [182][ 600/ 1200] Overall Loss 0.130403 Objective Loss 0.130403 LR 0.000125 Time 0.020240 -2022-12-06 11:44:09,376 - Epoch: [182][ 610/ 1200] Overall Loss 0.130294 Objective Loss 0.130294 LR 0.000125 Time 0.020219 -2022-12-06 11:44:09,566 - Epoch: [182][ 620/ 1200] Overall Loss 0.130150 Objective Loss 0.130150 LR 0.000125 Time 0.020199 -2022-12-06 11:44:09,758 - Epoch: [182][ 630/ 1200] Overall Loss 0.130162 Objective Loss 0.130162 LR 0.000125 Time 0.020181 -2022-12-06 11:44:09,949 - Epoch: [182][ 640/ 1200] Overall Loss 0.130290 Objective Loss 0.130290 LR 0.000125 Time 0.020163 -2022-12-06 11:44:10,139 - Epoch: [182][ 650/ 1200] Overall Loss 0.130254 Objective Loss 0.130254 LR 0.000125 Time 0.020145 -2022-12-06 11:44:10,330 - Epoch: [182][ 660/ 1200] Overall Loss 0.130097 Objective Loss 0.130097 LR 0.000125 Time 0.020128 -2022-12-06 11:44:10,520 - Epoch: [182][ 670/ 1200] Overall Loss 0.130432 Objective Loss 0.130432 LR 0.000125 Time 0.020111 -2022-12-06 11:44:10,711 - Epoch: [182][ 680/ 1200] Overall Loss 0.130487 Objective Loss 0.130487 LR 0.000125 Time 0.020096 -2022-12-06 11:44:10,902 - Epoch: [182][ 690/ 1200] Overall Loss 0.130521 Objective Loss 0.130521 LR 0.000125 Time 0.020080 -2022-12-06 11:44:11,092 - Epoch: [182][ 700/ 1200] Overall Loss 0.130337 Objective Loss 0.130337 LR 0.000125 Time 0.020064 -2022-12-06 11:44:11,283 - Epoch: [182][ 710/ 1200] Overall Loss 0.130400 Objective Loss 0.130400 LR 0.000125 Time 0.020049 -2022-12-06 11:44:11,474 - Epoch: [182][ 720/ 1200] Overall Loss 0.130651 Objective Loss 0.130651 LR 0.000125 Time 0.020035 -2022-12-06 11:44:11,664 - Epoch: [182][ 730/ 1200] Overall Loss 0.130914 Objective Loss 0.130914 LR 0.000125 Time 0.020020 -2022-12-06 11:44:11,854 - Epoch: [182][ 740/ 1200] Overall Loss 0.130703 Objective Loss 0.130703 LR 0.000125 Time 0.020007 -2022-12-06 11:44:12,045 - Epoch: [182][ 750/ 1200] Overall Loss 0.130697 Objective Loss 0.130697 LR 0.000125 Time 0.019994 -2022-12-06 11:44:12,236 - Epoch: [182][ 760/ 1200] Overall Loss 0.130623 Objective Loss 0.130623 LR 0.000125 Time 0.019980 -2022-12-06 11:44:12,427 - Epoch: [182][ 770/ 1200] Overall Loss 0.130553 Objective Loss 0.130553 LR 0.000125 Time 0.019968 -2022-12-06 11:44:12,618 - Epoch: [182][ 780/ 1200] Overall Loss 0.130483 Objective Loss 0.130483 LR 0.000125 Time 0.019956 -2022-12-06 11:44:12,808 - Epoch: [182][ 790/ 1200] Overall Loss 0.130595 Objective Loss 0.130595 LR 0.000125 Time 0.019945 -2022-12-06 11:44:13,000 - Epoch: [182][ 800/ 1200] Overall Loss 0.130475 Objective Loss 0.130475 LR 0.000125 Time 0.019934 -2022-12-06 11:44:13,191 - Epoch: [182][ 810/ 1200] Overall Loss 0.130159 Objective Loss 0.130159 LR 0.000125 Time 0.019923 -2022-12-06 11:44:13,382 - Epoch: [182][ 820/ 1200] Overall Loss 0.130112 Objective Loss 0.130112 LR 0.000125 Time 0.019913 -2022-12-06 11:44:13,574 - Epoch: [182][ 830/ 1200] Overall Loss 0.129995 Objective Loss 0.129995 LR 0.000125 Time 0.019903 -2022-12-06 11:44:13,765 - Epoch: [182][ 840/ 1200] Overall Loss 0.130167 Objective Loss 0.130167 LR 0.000125 Time 0.019893 -2022-12-06 11:44:13,957 - Epoch: [182][ 850/ 1200] Overall Loss 0.130209 Objective Loss 0.130209 LR 0.000125 Time 0.019883 -2022-12-06 11:44:14,148 - Epoch: [182][ 860/ 1200] Overall Loss 0.130332 Objective Loss 0.130332 LR 0.000125 Time 0.019874 -2022-12-06 11:44:14,340 - Epoch: [182][ 870/ 1200] Overall Loss 0.130359 Objective Loss 0.130359 LR 0.000125 Time 0.019865 -2022-12-06 11:44:14,531 - Epoch: [182][ 880/ 1200] Overall Loss 0.130440 Objective Loss 0.130440 LR 0.000125 Time 0.019857 -2022-12-06 11:44:14,723 - Epoch: [182][ 890/ 1200] Overall Loss 0.130426 Objective Loss 0.130426 LR 0.000125 Time 0.019848 -2022-12-06 11:44:14,914 - Epoch: [182][ 900/ 1200] Overall Loss 0.130536 Objective Loss 0.130536 LR 0.000125 Time 0.019840 -2022-12-06 11:44:15,106 - Epoch: [182][ 910/ 1200] Overall Loss 0.130591 Objective Loss 0.130591 LR 0.000125 Time 0.019832 -2022-12-06 11:44:15,298 - Epoch: [182][ 920/ 1200] Overall Loss 0.131097 Objective Loss 0.131097 LR 0.000125 Time 0.019824 -2022-12-06 11:44:15,490 - Epoch: [182][ 930/ 1200] Overall Loss 0.131093 Objective Loss 0.131093 LR 0.000125 Time 0.019817 -2022-12-06 11:44:15,682 - Epoch: [182][ 940/ 1200] Overall Loss 0.131184 Objective Loss 0.131184 LR 0.000125 Time 0.019810 -2022-12-06 11:44:15,873 - Epoch: [182][ 950/ 1200] Overall Loss 0.131202 Objective Loss 0.131202 LR 0.000125 Time 0.019802 -2022-12-06 11:44:16,065 - Epoch: [182][ 960/ 1200] Overall Loss 0.131287 Objective Loss 0.131287 LR 0.000125 Time 0.019795 -2022-12-06 11:44:16,256 - Epoch: [182][ 970/ 1200] Overall Loss 0.131184 Objective Loss 0.131184 LR 0.000125 Time 0.019788 -2022-12-06 11:44:16,449 - Epoch: [182][ 980/ 1200] Overall Loss 0.131032 Objective Loss 0.131032 LR 0.000125 Time 0.019782 -2022-12-06 11:44:16,640 - Epoch: [182][ 990/ 1200] Overall Loss 0.131051 Objective Loss 0.131051 LR 0.000125 Time 0.019775 -2022-12-06 11:44:16,832 - Epoch: [182][ 1000/ 1200] Overall Loss 0.131092 Objective Loss 0.131092 LR 0.000125 Time 0.019768 -2022-12-06 11:44:17,023 - Epoch: [182][ 1010/ 1200] Overall Loss 0.131175 Objective Loss 0.131175 LR 0.000125 Time 0.019761 -2022-12-06 11:44:17,215 - Epoch: [182][ 1020/ 1200] Overall Loss 0.130942 Objective Loss 0.130942 LR 0.000125 Time 0.019755 -2022-12-06 11:44:17,406 - Epoch: [182][ 1030/ 1200] Overall Loss 0.130879 Objective Loss 0.130879 LR 0.000125 Time 0.019748 -2022-12-06 11:44:17,597 - Epoch: [182][ 1040/ 1200] Overall Loss 0.130785 Objective Loss 0.130785 LR 0.000125 Time 0.019741 -2022-12-06 11:44:17,790 - Epoch: [182][ 1050/ 1200] Overall Loss 0.130867 Objective Loss 0.130867 LR 0.000125 Time 0.019736 -2022-12-06 11:44:17,981 - Epoch: [182][ 1060/ 1200] Overall Loss 0.130788 Objective Loss 0.130788 LR 0.000125 Time 0.019730 -2022-12-06 11:44:18,173 - Epoch: [182][ 1070/ 1200] Overall Loss 0.130784 Objective Loss 0.130784 LR 0.000125 Time 0.019724 -2022-12-06 11:44:18,364 - Epoch: [182][ 1080/ 1200] Overall Loss 0.130944 Objective Loss 0.130944 LR 0.000125 Time 0.019719 -2022-12-06 11:44:18,556 - Epoch: [182][ 1090/ 1200] Overall Loss 0.130785 Objective Loss 0.130785 LR 0.000125 Time 0.019713 -2022-12-06 11:44:18,747 - Epoch: [182][ 1100/ 1200] Overall Loss 0.130685 Objective Loss 0.130685 LR 0.000125 Time 0.019707 -2022-12-06 11:44:18,938 - Epoch: [182][ 1110/ 1200] Overall Loss 0.130578 Objective Loss 0.130578 LR 0.000125 Time 0.019701 -2022-12-06 11:44:19,130 - Epoch: [182][ 1120/ 1200] Overall Loss 0.130581 Objective Loss 0.130581 LR 0.000125 Time 0.019696 -2022-12-06 11:44:19,322 - Epoch: [182][ 1130/ 1200] Overall Loss 0.130594 Objective Loss 0.130594 LR 0.000125 Time 0.019691 -2022-12-06 11:44:19,514 - Epoch: [182][ 1140/ 1200] Overall Loss 0.130713 Objective Loss 0.130713 LR 0.000125 Time 0.019686 -2022-12-06 11:44:19,706 - Epoch: [182][ 1150/ 1200] Overall Loss 0.130593 Objective Loss 0.130593 LR 0.000125 Time 0.019681 -2022-12-06 11:44:19,898 - Epoch: [182][ 1160/ 1200] Overall Loss 0.130639 Objective Loss 0.130639 LR 0.000125 Time 0.019677 -2022-12-06 11:44:20,089 - Epoch: [182][ 1170/ 1200] Overall Loss 0.130695 Objective Loss 0.130695 LR 0.000125 Time 0.019672 -2022-12-06 11:44:20,280 - Epoch: [182][ 1180/ 1200] Overall Loss 0.130754 Objective Loss 0.130754 LR 0.000125 Time 0.019667 -2022-12-06 11:44:20,472 - Epoch: [182][ 1190/ 1200] Overall Loss 0.130709 Objective Loss 0.130709 LR 0.000125 Time 0.019662 -2022-12-06 11:44:20,706 - Epoch: [182][ 1200/ 1200] Overall Loss 0.130469 Objective Loss 0.130469 Top1 89.958159 Top5 98.953975 LR 0.000125 Time 0.019693 -2022-12-06 11:44:20,795 - --- validate (epoch=182)----------- -2022-12-06 11:44:20,795 - 34129 samples (256 per mini-batch) -2022-12-06 11:44:21,243 - Epoch: [182][ 10/ 134] Loss 0.239659 Top1 87.343750 Top5 98.515625 -2022-12-06 11:44:21,375 - Epoch: [182][ 20/ 134] Loss 0.229524 Top1 87.285156 Top5 98.554688 -2022-12-06 11:44:21,507 - Epoch: [182][ 30/ 134] Loss 0.226062 Top1 87.734375 Top5 98.554688 -2022-12-06 11:44:21,638 - Epoch: [182][ 40/ 134] Loss 0.229366 Top1 87.792969 Top5 98.544922 -2022-12-06 11:44:21,770 - Epoch: [182][ 50/ 134] Loss 0.231010 Top1 87.843750 Top5 98.539062 -2022-12-06 11:44:21,902 - Epoch: [182][ 60/ 134] Loss 0.225982 Top1 87.832031 Top5 98.619792 -2022-12-06 11:44:22,052 - Epoch: [182][ 70/ 134] Loss 0.222931 Top1 88.058036 Top5 98.632812 -2022-12-06 11:44:22,182 - Epoch: [182][ 80/ 134] Loss 0.225627 Top1 87.988281 Top5 98.627930 -2022-12-06 11:44:22,309 - Epoch: [182][ 90/ 134] Loss 0.226006 Top1 88.059896 Top5 98.650174 -2022-12-06 11:44:22,448 - Epoch: [182][ 100/ 134] Loss 0.227897 Top1 88.058594 Top5 98.656250 -2022-12-06 11:44:22,581 - Epoch: [182][ 110/ 134] Loss 0.228329 Top1 88.117898 Top5 98.654119 -2022-12-06 11:44:22,711 - Epoch: [182][ 120/ 134] Loss 0.228275 Top1 88.144531 Top5 98.658854 -2022-12-06 11:44:22,841 - Epoch: [182][ 130/ 134] Loss 0.227970 Top1 88.167067 Top5 98.650841 -2022-12-06 11:44:22,878 - Epoch: [182][ 134/ 134] Loss 0.229427 Top1 88.139119 Top5 98.643382 -2022-12-06 11:44:22,966 - ==> Top1: 88.139 Top5: 98.643 Loss: 0.229 - -2022-12-06 11:44:22,967 - ==> Confusion: -[[ 918 0 0 2 4 5 1 1 6 43 0 1 1 2 6 1 2 0 1 0 2] - [ 2 941 1 2 9 22 1 13 3 1 1 4 1 1 0 1 3 0 10 4 7] - [ 4 3 1024 11 3 3 13 10 0 1 4 4 1 2 2 2 2 1 1 2 10] - [ 4 1 13 955 1 3 0 0 0 1 9 0 4 1 7 0 1 1 11 1 7] - [ 10 3 1 0 967 3 1 2 1 6 1 4 0 2 8 1 6 2 0 0 2] - [ 2 12 1 4 5 985 2 13 2 2 1 11 4 14 1 1 0 0 2 3 4] - [ 1 1 9 0 0 3 1077 2 0 1 0 2 0 1 0 5 1 2 2 8 3] - [ 1 5 3 2 1 30 9 957 0 1 0 6 0 0 1 0 1 0 18 13 6] - [ 3 2 0 1 0 3 1 1 992 37 8 1 1 3 6 0 2 0 1 1 1] - [ 50 0 1 0 7 3 0 2 20 895 1 1 0 10 2 0 0 1 1 0 7] - [ 0 3 3 4 0 0 1 2 10 3 965 0 1 9 2 1 0 0 3 2 10] - [ 2 1 1 0 2 8 4 2 2 0 0 981 23 2 0 4 3 4 0 10 2] - [ 1 1 2 2 0 3 0 0 1 0 0 21 919 1 0 6 2 2 0 3 5] - [ 1 0 0 0 2 7 0 1 9 16 2 7 2 961 0 1 4 0 1 2 7] - [ 8 1 2 9 4 4 0 0 14 0 0 3 1 3 1067 0 1 1 8 0 4] - [ 1 0 1 1 2 0 3 1 0 0 1 8 7 2 0 996 7 9 0 2 2] - [ 2 0 1 1 3 1 0 0 0 0 1 2 3 3 0 6 1036 0 1 6 6] - [ 3 1 2 2 2 1 1 0 0 4 0 8 16 1 3 13 0 973 0 3 3] - [ 3 3 4 7 1 4 0 21 2 1 3 4 2 0 5 0 0 3 939 1 5] - [ 1 4 0 2 0 5 4 4 0 1 1 12 3 6 1 4 2 1 1 1023 5] - [ 115 170 148 99 107 163 68 118 76 80 140 88 305 224 131 91 162 67 146 219 10509]] - -2022-12-06 11:44:23,635 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:44:23,635 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:44:23,641 - - -2022-12-06 11:44:23,641 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:44:24,583 - Epoch: [183][ 10/ 1200] Overall Loss 0.131825 Objective Loss 0.131825 LR 0.000125 Time 0.094064 -2022-12-06 11:44:24,782 - Epoch: [183][ 20/ 1200] Overall Loss 0.137998 Objective Loss 0.137998 LR 0.000125 Time 0.056965 -2022-12-06 11:44:24,974 - Epoch: [183][ 30/ 1200] Overall Loss 0.138922 Objective Loss 0.138922 LR 0.000125 Time 0.044370 -2022-12-06 11:44:25,166 - Epoch: [183][ 40/ 1200] Overall Loss 0.134811 Objective Loss 0.134811 LR 0.000125 Time 0.038064 -2022-12-06 11:44:25,358 - Epoch: [183][ 50/ 1200] Overall Loss 0.133806 Objective Loss 0.133806 LR 0.000125 Time 0.034276 -2022-12-06 11:44:25,550 - Epoch: [183][ 60/ 1200] Overall Loss 0.131810 Objective Loss 0.131810 LR 0.000125 Time 0.031747 -2022-12-06 11:44:25,741 - Epoch: [183][ 70/ 1200] Overall Loss 0.132011 Objective Loss 0.132011 LR 0.000125 Time 0.029943 -2022-12-06 11:44:25,933 - Epoch: [183][ 80/ 1200] Overall Loss 0.132182 Objective Loss 0.132182 LR 0.000125 Time 0.028584 -2022-12-06 11:44:26,124 - Epoch: [183][ 90/ 1200] Overall Loss 0.130637 Objective Loss 0.130637 LR 0.000125 Time 0.027530 -2022-12-06 11:44:26,316 - Epoch: [183][ 100/ 1200] Overall Loss 0.130289 Objective Loss 0.130289 LR 0.000125 Time 0.026689 -2022-12-06 11:44:26,508 - Epoch: [183][ 110/ 1200] Overall Loss 0.131551 Objective Loss 0.131551 LR 0.000125 Time 0.026003 -2022-12-06 11:44:26,699 - Epoch: [183][ 120/ 1200] Overall Loss 0.130836 Objective Loss 0.130836 LR 0.000125 Time 0.025427 -2022-12-06 11:44:26,891 - Epoch: [183][ 130/ 1200] Overall Loss 0.129907 Objective Loss 0.129907 LR 0.000125 Time 0.024940 -2022-12-06 11:44:27,082 - Epoch: [183][ 140/ 1200] Overall Loss 0.130524 Objective Loss 0.130524 LR 0.000125 Time 0.024522 -2022-12-06 11:44:27,274 - Epoch: [183][ 150/ 1200] Overall Loss 0.130035 Objective Loss 0.130035 LR 0.000125 Time 0.024162 -2022-12-06 11:44:27,465 - Epoch: [183][ 160/ 1200] Overall Loss 0.130447 Objective Loss 0.130447 LR 0.000125 Time 0.023843 -2022-12-06 11:44:27,656 - Epoch: [183][ 170/ 1200] Overall Loss 0.130248 Objective Loss 0.130248 LR 0.000125 Time 0.023561 -2022-12-06 11:44:27,848 - Epoch: [183][ 180/ 1200] Overall Loss 0.128940 Objective Loss 0.128940 LR 0.000125 Time 0.023313 -2022-12-06 11:44:28,039 - Epoch: [183][ 190/ 1200] Overall Loss 0.128862 Objective Loss 0.128862 LR 0.000125 Time 0.023088 -2022-12-06 11:44:28,230 - Epoch: [183][ 200/ 1200] Overall Loss 0.128290 Objective Loss 0.128290 LR 0.000125 Time 0.022888 -2022-12-06 11:44:28,422 - Epoch: [183][ 210/ 1200] Overall Loss 0.128946 Objective Loss 0.128946 LR 0.000125 Time 0.022709 -2022-12-06 11:44:28,614 - Epoch: [183][ 220/ 1200] Overall Loss 0.129452 Objective Loss 0.129452 LR 0.000125 Time 0.022547 -2022-12-06 11:44:28,805 - Epoch: [183][ 230/ 1200] Overall Loss 0.129320 Objective Loss 0.129320 LR 0.000125 Time 0.022397 -2022-12-06 11:44:28,996 - Epoch: [183][ 240/ 1200] Overall Loss 0.129290 Objective Loss 0.129290 LR 0.000125 Time 0.022256 -2022-12-06 11:44:29,187 - Epoch: [183][ 250/ 1200] Overall Loss 0.129578 Objective Loss 0.129578 LR 0.000125 Time 0.022127 -2022-12-06 11:44:29,378 - Epoch: [183][ 260/ 1200] Overall Loss 0.129671 Objective Loss 0.129671 LR 0.000125 Time 0.022010 -2022-12-06 11:44:29,570 - Epoch: [183][ 270/ 1200] Overall Loss 0.129249 Objective Loss 0.129249 LR 0.000125 Time 0.021901 -2022-12-06 11:44:29,760 - Epoch: [183][ 280/ 1200] Overall Loss 0.129325 Objective Loss 0.129325 LR 0.000125 Time 0.021797 -2022-12-06 11:44:29,952 - Epoch: [183][ 290/ 1200] Overall Loss 0.129937 Objective Loss 0.129937 LR 0.000125 Time 0.021704 -2022-12-06 11:44:30,143 - Epoch: [183][ 300/ 1200] Overall Loss 0.130233 Objective Loss 0.130233 LR 0.000125 Time 0.021619 -2022-12-06 11:44:30,335 - Epoch: [183][ 310/ 1200] Overall Loss 0.129816 Objective Loss 0.129816 LR 0.000125 Time 0.021536 -2022-12-06 11:44:30,527 - Epoch: [183][ 320/ 1200] Overall Loss 0.129909 Objective Loss 0.129909 LR 0.000125 Time 0.021461 -2022-12-06 11:44:30,718 - Epoch: [183][ 330/ 1200] Overall Loss 0.130614 Objective Loss 0.130614 LR 0.000125 Time 0.021389 -2022-12-06 11:44:30,910 - Epoch: [183][ 340/ 1200] Overall Loss 0.130370 Objective Loss 0.130370 LR 0.000125 Time 0.021322 -2022-12-06 11:44:31,102 - Epoch: [183][ 350/ 1200] Overall Loss 0.130152 Objective Loss 0.130152 LR 0.000125 Time 0.021261 -2022-12-06 11:44:31,294 - Epoch: [183][ 360/ 1200] Overall Loss 0.130489 Objective Loss 0.130489 LR 0.000125 Time 0.021201 -2022-12-06 11:44:31,485 - Epoch: [183][ 370/ 1200] Overall Loss 0.130194 Objective Loss 0.130194 LR 0.000125 Time 0.021144 -2022-12-06 11:44:31,676 - Epoch: [183][ 380/ 1200] Overall Loss 0.130217 Objective Loss 0.130217 LR 0.000125 Time 0.021089 -2022-12-06 11:44:31,868 - Epoch: [183][ 390/ 1200] Overall Loss 0.130358 Objective Loss 0.130358 LR 0.000125 Time 0.021038 -2022-12-06 11:44:32,059 - Epoch: [183][ 400/ 1200] Overall Loss 0.130406 Objective Loss 0.130406 LR 0.000125 Time 0.020989 -2022-12-06 11:44:32,250 - Epoch: [183][ 410/ 1200] Overall Loss 0.130891 Objective Loss 0.130891 LR 0.000125 Time 0.020942 -2022-12-06 11:44:32,442 - Epoch: [183][ 420/ 1200] Overall Loss 0.130793 Objective Loss 0.130793 LR 0.000125 Time 0.020898 -2022-12-06 11:44:32,634 - Epoch: [183][ 430/ 1200] Overall Loss 0.130690 Objective Loss 0.130690 LR 0.000125 Time 0.020858 -2022-12-06 11:44:32,826 - Epoch: [183][ 440/ 1200] Overall Loss 0.130702 Objective Loss 0.130702 LR 0.000125 Time 0.020818 -2022-12-06 11:44:33,018 - Epoch: [183][ 450/ 1200] Overall Loss 0.130633 Objective Loss 0.130633 LR 0.000125 Time 0.020781 -2022-12-06 11:44:33,208 - Epoch: [183][ 460/ 1200] Overall Loss 0.130641 Objective Loss 0.130641 LR 0.000125 Time 0.020743 -2022-12-06 11:44:33,400 - Epoch: [183][ 470/ 1200] Overall Loss 0.130390 Objective Loss 0.130390 LR 0.000125 Time 0.020708 -2022-12-06 11:44:33,591 - Epoch: [183][ 480/ 1200] Overall Loss 0.130618 Objective Loss 0.130618 LR 0.000125 Time 0.020674 -2022-12-06 11:44:33,783 - Epoch: [183][ 490/ 1200] Overall Loss 0.130518 Objective Loss 0.130518 LR 0.000125 Time 0.020643 -2022-12-06 11:44:33,975 - Epoch: [183][ 500/ 1200] Overall Loss 0.130517 Objective Loss 0.130517 LR 0.000125 Time 0.020613 -2022-12-06 11:44:34,167 - Epoch: [183][ 510/ 1200] Overall Loss 0.130416 Objective Loss 0.130416 LR 0.000125 Time 0.020583 -2022-12-06 11:44:34,358 - Epoch: [183][ 520/ 1200] Overall Loss 0.130216 Objective Loss 0.130216 LR 0.000125 Time 0.020554 -2022-12-06 11:44:34,550 - Epoch: [183][ 530/ 1200] Overall Loss 0.129812 Objective Loss 0.129812 LR 0.000125 Time 0.020527 -2022-12-06 11:44:34,742 - Epoch: [183][ 540/ 1200] Overall Loss 0.129942 Objective Loss 0.129942 LR 0.000125 Time 0.020501 -2022-12-06 11:44:34,933 - Epoch: [183][ 550/ 1200] Overall Loss 0.129687 Objective Loss 0.129687 LR 0.000125 Time 0.020475 -2022-12-06 11:44:35,125 - Epoch: [183][ 560/ 1200] Overall Loss 0.129655 Objective Loss 0.129655 LR 0.000125 Time 0.020451 -2022-12-06 11:44:35,316 - Epoch: [183][ 570/ 1200] Overall Loss 0.129414 Objective Loss 0.129414 LR 0.000125 Time 0.020427 -2022-12-06 11:44:35,508 - Epoch: [183][ 580/ 1200] Overall Loss 0.129543 Objective Loss 0.129543 LR 0.000125 Time 0.020404 -2022-12-06 11:44:35,699 - Epoch: [183][ 590/ 1200] Overall Loss 0.129600 Objective Loss 0.129600 LR 0.000125 Time 0.020382 -2022-12-06 11:44:35,890 - Epoch: [183][ 600/ 1200] Overall Loss 0.129450 Objective Loss 0.129450 LR 0.000125 Time 0.020360 -2022-12-06 11:44:36,082 - Epoch: [183][ 610/ 1200] Overall Loss 0.129581 Objective Loss 0.129581 LR 0.000125 Time 0.020339 -2022-12-06 11:44:36,273 - Epoch: [183][ 620/ 1200] Overall Loss 0.129568 Objective Loss 0.129568 LR 0.000125 Time 0.020319 -2022-12-06 11:44:36,465 - Epoch: [183][ 630/ 1200] Overall Loss 0.129318 Objective Loss 0.129318 LR 0.000125 Time 0.020300 -2022-12-06 11:44:36,657 - Epoch: [183][ 640/ 1200] Overall Loss 0.129410 Objective Loss 0.129410 LR 0.000125 Time 0.020281 -2022-12-06 11:44:36,848 - Epoch: [183][ 650/ 1200] Overall Loss 0.129433 Objective Loss 0.129433 LR 0.000125 Time 0.020263 -2022-12-06 11:44:37,039 - Epoch: [183][ 660/ 1200] Overall Loss 0.129265 Objective Loss 0.129265 LR 0.000125 Time 0.020245 -2022-12-06 11:44:37,230 - Epoch: [183][ 670/ 1200] Overall Loss 0.129479 Objective Loss 0.129479 LR 0.000125 Time 0.020227 -2022-12-06 11:44:37,422 - Epoch: [183][ 680/ 1200] Overall Loss 0.129452 Objective Loss 0.129452 LR 0.000125 Time 0.020210 -2022-12-06 11:44:37,613 - Epoch: [183][ 690/ 1200] Overall Loss 0.129455 Objective Loss 0.129455 LR 0.000125 Time 0.020193 -2022-12-06 11:44:37,803 - Epoch: [183][ 700/ 1200] Overall Loss 0.129513 Objective Loss 0.129513 LR 0.000125 Time 0.020176 -2022-12-06 11:44:37,993 - Epoch: [183][ 710/ 1200] Overall Loss 0.129506 Objective Loss 0.129506 LR 0.000125 Time 0.020159 -2022-12-06 11:44:38,185 - Epoch: [183][ 720/ 1200] Overall Loss 0.129476 Objective Loss 0.129476 LR 0.000125 Time 0.020144 -2022-12-06 11:44:38,376 - Epoch: [183][ 730/ 1200] Overall Loss 0.129712 Objective Loss 0.129712 LR 0.000125 Time 0.020129 -2022-12-06 11:44:38,566 - Epoch: [183][ 740/ 1200] Overall Loss 0.129651 Objective Loss 0.129651 LR 0.000125 Time 0.020114 -2022-12-06 11:44:38,757 - Epoch: [183][ 750/ 1200] Overall Loss 0.129724 Objective Loss 0.129724 LR 0.000125 Time 0.020099 -2022-12-06 11:44:38,948 - Epoch: [183][ 760/ 1200] Overall Loss 0.129609 Objective Loss 0.129609 LR 0.000125 Time 0.020085 -2022-12-06 11:44:39,138 - Epoch: [183][ 770/ 1200] Overall Loss 0.129713 Objective Loss 0.129713 LR 0.000125 Time 0.020071 -2022-12-06 11:44:39,329 - Epoch: [183][ 780/ 1200] Overall Loss 0.129805 Objective Loss 0.129805 LR 0.000125 Time 0.020058 -2022-12-06 11:44:39,521 - Epoch: [183][ 790/ 1200] Overall Loss 0.129673 Objective Loss 0.129673 LR 0.000125 Time 0.020046 -2022-12-06 11:44:39,712 - Epoch: [183][ 800/ 1200] Overall Loss 0.129635 Objective Loss 0.129635 LR 0.000125 Time 0.020033 -2022-12-06 11:44:39,903 - Epoch: [183][ 810/ 1200] Overall Loss 0.129674 Objective Loss 0.129674 LR 0.000125 Time 0.020021 -2022-12-06 11:44:40,095 - Epoch: [183][ 820/ 1200] Overall Loss 0.129564 Objective Loss 0.129564 LR 0.000125 Time 0.020010 -2022-12-06 11:44:40,286 - Epoch: [183][ 830/ 1200] Overall Loss 0.129541 Objective Loss 0.129541 LR 0.000125 Time 0.019999 -2022-12-06 11:44:40,476 - Epoch: [183][ 840/ 1200] Overall Loss 0.129539 Objective Loss 0.129539 LR 0.000125 Time 0.019987 -2022-12-06 11:44:40,667 - Epoch: [183][ 850/ 1200] Overall Loss 0.129548 Objective Loss 0.129548 LR 0.000125 Time 0.019976 -2022-12-06 11:44:40,857 - Epoch: [183][ 860/ 1200] Overall Loss 0.129545 Objective Loss 0.129545 LR 0.000125 Time 0.019964 -2022-12-06 11:44:41,048 - Epoch: [183][ 870/ 1200] Overall Loss 0.129418 Objective Loss 0.129418 LR 0.000125 Time 0.019953 -2022-12-06 11:44:41,239 - Epoch: [183][ 880/ 1200] Overall Loss 0.129620 Objective Loss 0.129620 LR 0.000125 Time 0.019942 -2022-12-06 11:44:41,430 - Epoch: [183][ 890/ 1200] Overall Loss 0.129607 Objective Loss 0.129607 LR 0.000125 Time 0.019932 -2022-12-06 11:44:41,620 - Epoch: [183][ 900/ 1200] Overall Loss 0.129641 Objective Loss 0.129641 LR 0.000125 Time 0.019922 -2022-12-06 11:44:41,811 - Epoch: [183][ 910/ 1200] Overall Loss 0.129621 Objective Loss 0.129621 LR 0.000125 Time 0.019911 -2022-12-06 11:44:42,002 - Epoch: [183][ 920/ 1200] Overall Loss 0.129647 Objective Loss 0.129647 LR 0.000125 Time 0.019902 -2022-12-06 11:44:42,193 - Epoch: [183][ 930/ 1200] Overall Loss 0.129739 Objective Loss 0.129739 LR 0.000125 Time 0.019893 -2022-12-06 11:44:42,384 - Epoch: [183][ 940/ 1200] Overall Loss 0.129676 Objective Loss 0.129676 LR 0.000125 Time 0.019884 -2022-12-06 11:44:42,575 - Epoch: [183][ 950/ 1200] Overall Loss 0.129799 Objective Loss 0.129799 LR 0.000125 Time 0.019875 -2022-12-06 11:44:42,765 - Epoch: [183][ 960/ 1200] Overall Loss 0.129613 Objective Loss 0.129613 LR 0.000125 Time 0.019866 -2022-12-06 11:44:42,956 - Epoch: [183][ 970/ 1200] Overall Loss 0.129936 Objective Loss 0.129936 LR 0.000125 Time 0.019857 -2022-12-06 11:44:43,147 - Epoch: [183][ 980/ 1200] Overall Loss 0.129889 Objective Loss 0.129889 LR 0.000125 Time 0.019849 -2022-12-06 11:44:43,338 - Epoch: [183][ 990/ 1200] Overall Loss 0.129740 Objective Loss 0.129740 LR 0.000125 Time 0.019841 -2022-12-06 11:44:43,529 - Epoch: [183][ 1000/ 1200] Overall Loss 0.129707 Objective Loss 0.129707 LR 0.000125 Time 0.019833 -2022-12-06 11:44:43,720 - Epoch: [183][ 1010/ 1200] Overall Loss 0.129688 Objective Loss 0.129688 LR 0.000125 Time 0.019825 -2022-12-06 11:44:43,912 - Epoch: [183][ 1020/ 1200] Overall Loss 0.129887 Objective Loss 0.129887 LR 0.000125 Time 0.019818 -2022-12-06 11:44:44,102 - Epoch: [183][ 1030/ 1200] Overall Loss 0.129909 Objective Loss 0.129909 LR 0.000125 Time 0.019810 -2022-12-06 11:44:44,294 - Epoch: [183][ 1040/ 1200] Overall Loss 0.129831 Objective Loss 0.129831 LR 0.000125 Time 0.019803 -2022-12-06 11:44:44,485 - Epoch: [183][ 1050/ 1200] Overall Loss 0.129731 Objective Loss 0.129731 LR 0.000125 Time 0.019796 -2022-12-06 11:44:44,676 - Epoch: [183][ 1060/ 1200] Overall Loss 0.129858 Objective Loss 0.129858 LR 0.000125 Time 0.019789 -2022-12-06 11:44:44,867 - Epoch: [183][ 1070/ 1200] Overall Loss 0.129865 Objective Loss 0.129865 LR 0.000125 Time 0.019782 -2022-12-06 11:44:45,058 - Epoch: [183][ 1080/ 1200] Overall Loss 0.129804 Objective Loss 0.129804 LR 0.000125 Time 0.019775 -2022-12-06 11:44:45,249 - Epoch: [183][ 1090/ 1200] Overall Loss 0.129557 Objective Loss 0.129557 LR 0.000125 Time 0.019769 -2022-12-06 11:44:45,439 - Epoch: [183][ 1100/ 1200] Overall Loss 0.129528 Objective Loss 0.129528 LR 0.000125 Time 0.019762 -2022-12-06 11:44:45,630 - Epoch: [183][ 1110/ 1200] Overall Loss 0.129375 Objective Loss 0.129375 LR 0.000125 Time 0.019755 -2022-12-06 11:44:45,821 - Epoch: [183][ 1120/ 1200] Overall Loss 0.129401 Objective Loss 0.129401 LR 0.000125 Time 0.019749 -2022-12-06 11:44:46,012 - Epoch: [183][ 1130/ 1200] Overall Loss 0.129322 Objective Loss 0.129322 LR 0.000125 Time 0.019742 -2022-12-06 11:44:46,203 - Epoch: [183][ 1140/ 1200] Overall Loss 0.129442 Objective Loss 0.129442 LR 0.000125 Time 0.019736 -2022-12-06 11:44:46,394 - Epoch: [183][ 1150/ 1200] Overall Loss 0.129464 Objective Loss 0.129464 LR 0.000125 Time 0.019730 -2022-12-06 11:44:46,584 - Epoch: [183][ 1160/ 1200] Overall Loss 0.129348 Objective Loss 0.129348 LR 0.000125 Time 0.019723 -2022-12-06 11:44:46,775 - Epoch: [183][ 1170/ 1200] Overall Loss 0.129331 Objective Loss 0.129331 LR 0.000125 Time 0.019717 -2022-12-06 11:44:46,964 - Epoch: [183][ 1180/ 1200] Overall Loss 0.129530 Objective Loss 0.129530 LR 0.000125 Time 0.019710 -2022-12-06 11:44:47,156 - Epoch: [183][ 1190/ 1200] Overall Loss 0.129625 Objective Loss 0.129625 LR 0.000125 Time 0.019705 -2022-12-06 11:44:47,388 - Epoch: [183][ 1200/ 1200] Overall Loss 0.129839 Objective Loss 0.129839 Top1 91.004184 Top5 99.163180 LR 0.000125 Time 0.019734 -2022-12-06 11:44:47,477 - --- validate (epoch=183)----------- -2022-12-06 11:44:47,478 - 34129 samples (256 per mini-batch) -2022-12-06 11:44:47,939 - Epoch: [183][ 10/ 134] Loss 0.242352 Top1 88.515625 Top5 98.984375 -2022-12-06 11:44:48,071 - Epoch: [183][ 20/ 134] Loss 0.230633 Top1 88.906250 Top5 98.847656 -2022-12-06 11:44:48,201 - Epoch: [183][ 30/ 134] Loss 0.231645 Top1 88.802083 Top5 98.763021 -2022-12-06 11:44:48,338 - Epoch: [183][ 40/ 134] Loss 0.235541 Top1 88.535156 Top5 98.681641 -2022-12-06 11:44:48,480 - Epoch: [183][ 50/ 134] Loss 0.224662 Top1 88.617188 Top5 98.765625 -2022-12-06 11:44:48,624 - Epoch: [183][ 60/ 134] Loss 0.229839 Top1 88.580729 Top5 98.691406 -2022-12-06 11:44:48,767 - Epoch: [183][ 70/ 134] Loss 0.231244 Top1 88.549107 Top5 98.632812 -2022-12-06 11:44:48,906 - Epoch: [183][ 80/ 134] Loss 0.226394 Top1 88.505859 Top5 98.627930 -2022-12-06 11:44:49,037 - Epoch: [183][ 90/ 134] Loss 0.228213 Top1 88.550347 Top5 98.628472 -2022-12-06 11:44:49,170 - Epoch: [183][ 100/ 134] Loss 0.228237 Top1 88.460938 Top5 98.648438 -2022-12-06 11:44:49,301 - Epoch: [183][ 110/ 134] Loss 0.229282 Top1 88.330966 Top5 98.615057 -2022-12-06 11:44:49,432 - Epoch: [183][ 120/ 134] Loss 0.225873 Top1 88.450521 Top5 98.658854 -2022-12-06 11:44:49,563 - Epoch: [183][ 130/ 134] Loss 0.225765 Top1 88.341346 Top5 98.650841 -2022-12-06 11:44:49,600 - Epoch: [183][ 134/ 134] Loss 0.226524 Top1 88.361804 Top5 98.652173 -2022-12-06 11:44:49,691 - ==> Top1: 88.362 Top5: 98.652 Loss: 0.227 - -2022-12-06 11:44:49,691 - ==> Confusion: -[[ 931 0 1 1 1 8 0 1 4 35 0 1 1 2 4 2 0 1 2 0 1] - [ 2 942 0 2 7 23 3 11 1 1 3 5 0 1 0 0 2 2 10 3 9] - [ 2 3 1022 11 2 2 14 10 0 3 3 4 3 2 2 2 2 1 2 2 11] - [ 5 1 14 956 1 2 0 0 0 0 8 1 3 3 9 0 1 1 8 0 7] - [ 7 4 1 0 960 3 1 2 1 6 1 5 0 2 10 4 4 3 1 1 4] - [ 0 10 0 5 3 992 3 15 3 1 1 11 3 11 1 1 0 0 0 4 5] - [ 1 1 5 2 0 4 1083 2 0 1 0 2 0 1 0 2 0 1 2 7 4] - [ 0 6 5 2 0 23 10 971 0 0 2 6 0 2 0 0 0 0 15 7 5] - [ 5 1 0 0 0 1 0 3 982 37 10 1 2 7 9 0 2 1 1 1 1] - [ 63 1 0 0 2 3 0 2 21 883 1 0 0 10 5 1 0 1 1 0 7] - [ 0 2 3 6 1 1 2 3 7 1 971 0 0 7 2 1 0 0 6 0 6] - [ 2 1 1 0 1 12 4 3 1 0 0 986 22 1 0 3 2 4 0 4 4] - [ 0 1 1 2 0 3 0 2 0 1 0 22 912 2 0 7 0 9 0 3 4] - [ 1 1 1 0 0 8 0 3 8 11 4 5 3 962 0 2 2 1 0 1 10] - [ 5 4 3 10 3 3 0 0 13 1 0 4 1 3 1066 0 0 1 6 1 6] - [ 1 0 0 1 2 1 3 1 0 1 1 7 5 0 0 993 8 12 0 3 4] - [ 3 2 0 2 3 1 0 1 1 0 0 2 1 3 0 8 1029 0 1 5 10] - [ 4 1 1 1 0 0 0 0 0 3 0 6 13 2 2 12 0 988 0 0 3] - [ 4 5 5 6 0 4 0 19 1 0 2 2 3 1 7 0 0 3 940 2 4] - [ 1 2 0 3 0 5 4 7 0 1 2 12 4 5 0 2 2 2 1 1019 8] - [ 109 199 160 99 89 159 72 139 73 73 157 88 291 218 142 90 113 78 142 166 10569]] - -2022-12-06 11:44:50,259 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:44:50,259 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:44:50,265 - - -2022-12-06 11:44:50,265 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:44:51,322 - Epoch: [184][ 10/ 1200] Overall Loss 0.124201 Objective Loss 0.124201 LR 0.000125 Time 0.105544 -2022-12-06 11:44:51,536 - Epoch: [184][ 20/ 1200] Overall Loss 0.126530 Objective Loss 0.126530 LR 0.000125 Time 0.063444 -2022-12-06 11:44:51,749 - Epoch: [184][ 30/ 1200] Overall Loss 0.129730 Objective Loss 0.129730 LR 0.000125 Time 0.049380 -2022-12-06 11:44:51,960 - Epoch: [184][ 40/ 1200] Overall Loss 0.129143 Objective Loss 0.129143 LR 0.000125 Time 0.042313 -2022-12-06 11:44:52,173 - Epoch: [184][ 50/ 1200] Overall Loss 0.126786 Objective Loss 0.126786 LR 0.000125 Time 0.038091 -2022-12-06 11:44:52,383 - Epoch: [184][ 60/ 1200] Overall Loss 0.125232 Objective Loss 0.125232 LR 0.000125 Time 0.035237 -2022-12-06 11:44:52,596 - Epoch: [184][ 70/ 1200] Overall Loss 0.126990 Objective Loss 0.126990 LR 0.000125 Time 0.033241 -2022-12-06 11:44:52,807 - Epoch: [184][ 80/ 1200] Overall Loss 0.125220 Objective Loss 0.125220 LR 0.000125 Time 0.031711 -2022-12-06 11:44:53,019 - Epoch: [184][ 90/ 1200] Overall Loss 0.125163 Objective Loss 0.125163 LR 0.000125 Time 0.030537 -2022-12-06 11:44:53,229 - Epoch: [184][ 100/ 1200] Overall Loss 0.123689 Objective Loss 0.123689 LR 0.000125 Time 0.029579 -2022-12-06 11:44:53,441 - Epoch: [184][ 110/ 1200] Overall Loss 0.122996 Objective Loss 0.122996 LR 0.000125 Time 0.028810 -2022-12-06 11:44:53,651 - Epoch: [184][ 120/ 1200] Overall Loss 0.123505 Objective Loss 0.123505 LR 0.000125 Time 0.028153 -2022-12-06 11:44:53,863 - Epoch: [184][ 130/ 1200] Overall Loss 0.123690 Objective Loss 0.123690 LR 0.000125 Time 0.027616 -2022-12-06 11:44:54,074 - Epoch: [184][ 140/ 1200] Overall Loss 0.124717 Objective Loss 0.124717 LR 0.000125 Time 0.027147 -2022-12-06 11:44:54,286 - Epoch: [184][ 150/ 1200] Overall Loss 0.125127 Objective Loss 0.125127 LR 0.000125 Time 0.026747 -2022-12-06 11:44:54,497 - Epoch: [184][ 160/ 1200] Overall Loss 0.125008 Objective Loss 0.125008 LR 0.000125 Time 0.026392 -2022-12-06 11:44:54,709 - Epoch: [184][ 170/ 1200] Overall Loss 0.124988 Objective Loss 0.124988 LR 0.000125 Time 0.026084 -2022-12-06 11:44:54,919 - Epoch: [184][ 180/ 1200] Overall Loss 0.125294 Objective Loss 0.125294 LR 0.000125 Time 0.025799 -2022-12-06 11:44:55,133 - Epoch: [184][ 190/ 1200] Overall Loss 0.125228 Objective Loss 0.125228 LR 0.000125 Time 0.025560 -2022-12-06 11:44:55,343 - Epoch: [184][ 200/ 1200] Overall Loss 0.126344 Objective Loss 0.126344 LR 0.000125 Time 0.025330 -2022-12-06 11:44:55,556 - Epoch: [184][ 210/ 1200] Overall Loss 0.126136 Objective Loss 0.126136 LR 0.000125 Time 0.025134 -2022-12-06 11:44:55,766 - Epoch: [184][ 220/ 1200] Overall Loss 0.126838 Objective Loss 0.126838 LR 0.000125 Time 0.024946 -2022-12-06 11:44:55,978 - Epoch: [184][ 230/ 1200] Overall Loss 0.127163 Objective Loss 0.127163 LR 0.000125 Time 0.024779 -2022-12-06 11:44:56,189 - Epoch: [184][ 240/ 1200] Overall Loss 0.127281 Objective Loss 0.127281 LR 0.000125 Time 0.024624 -2022-12-06 11:44:56,401 - Epoch: [184][ 250/ 1200] Overall Loss 0.127051 Objective Loss 0.127051 LR 0.000125 Time 0.024486 -2022-12-06 11:44:56,611 - Epoch: [184][ 260/ 1200] Overall Loss 0.127320 Objective Loss 0.127320 LR 0.000125 Time 0.024351 -2022-12-06 11:44:56,824 - Epoch: [184][ 270/ 1200] Overall Loss 0.127055 Objective Loss 0.127055 LR 0.000125 Time 0.024235 -2022-12-06 11:44:57,035 - Epoch: [184][ 280/ 1200] Overall Loss 0.127911 Objective Loss 0.127911 LR 0.000125 Time 0.024118 -2022-12-06 11:44:57,248 - Epoch: [184][ 290/ 1200] Overall Loss 0.127985 Objective Loss 0.127985 LR 0.000125 Time 0.024020 -2022-12-06 11:44:57,458 - Epoch: [184][ 300/ 1200] Overall Loss 0.127690 Objective Loss 0.127690 LR 0.000125 Time 0.023920 -2022-12-06 11:44:57,671 - Epoch: [184][ 310/ 1200] Overall Loss 0.127950 Objective Loss 0.127950 LR 0.000125 Time 0.023831 -2022-12-06 11:44:57,881 - Epoch: [184][ 320/ 1200] Overall Loss 0.127923 Objective Loss 0.127923 LR 0.000125 Time 0.023742 -2022-12-06 11:44:58,094 - Epoch: [184][ 330/ 1200] Overall Loss 0.127787 Objective Loss 0.127787 LR 0.000125 Time 0.023665 -2022-12-06 11:44:58,304 - Epoch: [184][ 340/ 1200] Overall Loss 0.127732 Objective Loss 0.127732 LR 0.000125 Time 0.023586 -2022-12-06 11:44:58,517 - Epoch: [184][ 350/ 1200] Overall Loss 0.127903 Objective Loss 0.127903 LR 0.000125 Time 0.023520 -2022-12-06 11:44:58,727 - Epoch: [184][ 360/ 1200] Overall Loss 0.127582 Objective Loss 0.127582 LR 0.000125 Time 0.023448 -2022-12-06 11:44:58,940 - Epoch: [184][ 370/ 1200] Overall Loss 0.127289 Objective Loss 0.127289 LR 0.000125 Time 0.023387 -2022-12-06 11:44:59,150 - Epoch: [184][ 380/ 1200] Overall Loss 0.127592 Objective Loss 0.127592 LR 0.000125 Time 0.023324 -2022-12-06 11:44:59,362 - Epoch: [184][ 390/ 1200] Overall Loss 0.127231 Objective Loss 0.127231 LR 0.000125 Time 0.023269 -2022-12-06 11:44:59,573 - Epoch: [184][ 400/ 1200] Overall Loss 0.127963 Objective Loss 0.127963 LR 0.000125 Time 0.023213 -2022-12-06 11:44:59,786 - Epoch: [184][ 410/ 1200] Overall Loss 0.127997 Objective Loss 0.127997 LR 0.000125 Time 0.023164 -2022-12-06 11:44:59,996 - Epoch: [184][ 420/ 1200] Overall Loss 0.128574 Objective Loss 0.128574 LR 0.000125 Time 0.023112 -2022-12-06 11:45:00,208 - Epoch: [184][ 430/ 1200] Overall Loss 0.128588 Objective Loss 0.128588 LR 0.000125 Time 0.023066 -2022-12-06 11:45:00,418 - Epoch: [184][ 440/ 1200] Overall Loss 0.128535 Objective Loss 0.128535 LR 0.000125 Time 0.023018 -2022-12-06 11:45:00,631 - Epoch: [184][ 450/ 1200] Overall Loss 0.128199 Objective Loss 0.128199 LR 0.000125 Time 0.022978 -2022-12-06 11:45:00,842 - Epoch: [184][ 460/ 1200] Overall Loss 0.128106 Objective Loss 0.128106 LR 0.000125 Time 0.022935 -2022-12-06 11:45:01,054 - Epoch: [184][ 470/ 1200] Overall Loss 0.128366 Objective Loss 0.128366 LR 0.000125 Time 0.022898 -2022-12-06 11:45:01,265 - Epoch: [184][ 480/ 1200] Overall Loss 0.128774 Objective Loss 0.128774 LR 0.000125 Time 0.022859 -2022-12-06 11:45:01,478 - Epoch: [184][ 490/ 1200] Overall Loss 0.128964 Objective Loss 0.128964 LR 0.000125 Time 0.022826 -2022-12-06 11:45:01,689 - Epoch: [184][ 500/ 1200] Overall Loss 0.128971 Objective Loss 0.128971 LR 0.000125 Time 0.022790 -2022-12-06 11:45:01,901 - Epoch: [184][ 510/ 1200] Overall Loss 0.129170 Objective Loss 0.129170 LR 0.000125 Time 0.022759 -2022-12-06 11:45:02,112 - Epoch: [184][ 520/ 1200] Overall Loss 0.128926 Objective Loss 0.128926 LR 0.000125 Time 0.022725 -2022-12-06 11:45:02,325 - Epoch: [184][ 530/ 1200] Overall Loss 0.129139 Objective Loss 0.129139 LR 0.000125 Time 0.022697 -2022-12-06 11:45:02,536 - Epoch: [184][ 540/ 1200] Overall Loss 0.129137 Objective Loss 0.129137 LR 0.000125 Time 0.022667 -2022-12-06 11:45:02,749 - Epoch: [184][ 550/ 1200] Overall Loss 0.129125 Objective Loss 0.129125 LR 0.000125 Time 0.022641 -2022-12-06 11:45:02,959 - Epoch: [184][ 560/ 1200] Overall Loss 0.129234 Objective Loss 0.129234 LR 0.000125 Time 0.022611 -2022-12-06 11:45:03,172 - Epoch: [184][ 570/ 1200] Overall Loss 0.129157 Objective Loss 0.129157 LR 0.000125 Time 0.022586 -2022-12-06 11:45:03,383 - Epoch: [184][ 580/ 1200] Overall Loss 0.129286 Objective Loss 0.129286 LR 0.000125 Time 0.022560 -2022-12-06 11:45:03,596 - Epoch: [184][ 590/ 1200] Overall Loss 0.128974 Objective Loss 0.128974 LR 0.000125 Time 0.022538 -2022-12-06 11:45:03,807 - Epoch: [184][ 600/ 1200] Overall Loss 0.128730 Objective Loss 0.128730 LR 0.000125 Time 0.022512 -2022-12-06 11:45:04,019 - Epoch: [184][ 610/ 1200] Overall Loss 0.128441 Objective Loss 0.128441 LR 0.000125 Time 0.022490 -2022-12-06 11:45:04,229 - Epoch: [184][ 620/ 1200] Overall Loss 0.128292 Objective Loss 0.128292 LR 0.000125 Time 0.022465 -2022-12-06 11:45:04,441 - Epoch: [184][ 630/ 1200] Overall Loss 0.128166 Objective Loss 0.128166 LR 0.000125 Time 0.022444 -2022-12-06 11:45:04,641 - Epoch: [184][ 640/ 1200] Overall Loss 0.128210 Objective Loss 0.128210 LR 0.000125 Time 0.022406 -2022-12-06 11:45:04,838 - Epoch: [184][ 650/ 1200] Overall Loss 0.128072 Objective Loss 0.128072 LR 0.000125 Time 0.022363 -2022-12-06 11:45:05,038 - Epoch: [184][ 660/ 1200] Overall Loss 0.128029 Objective Loss 0.128029 LR 0.000125 Time 0.022327 -2022-12-06 11:45:05,236 - Epoch: [184][ 670/ 1200] Overall Loss 0.128099 Objective Loss 0.128099 LR 0.000125 Time 0.022287 -2022-12-06 11:45:05,436 - Epoch: [184][ 680/ 1200] Overall Loss 0.128145 Objective Loss 0.128145 LR 0.000125 Time 0.022253 -2022-12-06 11:45:05,633 - Epoch: [184][ 690/ 1200] Overall Loss 0.128405 Objective Loss 0.128405 LR 0.000125 Time 0.022216 -2022-12-06 11:45:05,833 - Epoch: [184][ 700/ 1200] Overall Loss 0.128318 Objective Loss 0.128318 LR 0.000125 Time 0.022184 -2022-12-06 11:45:06,030 - Epoch: [184][ 710/ 1200] Overall Loss 0.128208 Objective Loss 0.128208 LR 0.000125 Time 0.022148 -2022-12-06 11:45:06,230 - Epoch: [184][ 720/ 1200] Overall Loss 0.128328 Objective Loss 0.128328 LR 0.000125 Time 0.022117 -2022-12-06 11:45:06,427 - Epoch: [184][ 730/ 1200] Overall Loss 0.128417 Objective Loss 0.128417 LR 0.000125 Time 0.022083 -2022-12-06 11:45:06,627 - Epoch: [184][ 740/ 1200] Overall Loss 0.128258 Objective Loss 0.128258 LR 0.000125 Time 0.022054 -2022-12-06 11:45:06,824 - Epoch: [184][ 750/ 1200] Overall Loss 0.128337 Objective Loss 0.128337 LR 0.000125 Time 0.022022 -2022-12-06 11:45:07,024 - Epoch: [184][ 760/ 1200] Overall Loss 0.128369 Objective Loss 0.128369 LR 0.000125 Time 0.021995 -2022-12-06 11:45:07,221 - Epoch: [184][ 770/ 1200] Overall Loss 0.128389 Objective Loss 0.128389 LR 0.000125 Time 0.021965 -2022-12-06 11:45:07,422 - Epoch: [184][ 780/ 1200] Overall Loss 0.128333 Objective Loss 0.128333 LR 0.000125 Time 0.021939 -2022-12-06 11:45:07,619 - Epoch: [184][ 790/ 1200] Overall Loss 0.128409 Objective Loss 0.128409 LR 0.000125 Time 0.021911 -2022-12-06 11:45:07,819 - Epoch: [184][ 800/ 1200] Overall Loss 0.128398 Objective Loss 0.128398 LR 0.000125 Time 0.021886 -2022-12-06 11:45:08,016 - Epoch: [184][ 810/ 1200] Overall Loss 0.128476 Objective Loss 0.128476 LR 0.000125 Time 0.021858 -2022-12-06 11:45:08,216 - Epoch: [184][ 820/ 1200] Overall Loss 0.128464 Objective Loss 0.128464 LR 0.000125 Time 0.021834 -2022-12-06 11:45:08,413 - Epoch: [184][ 830/ 1200] Overall Loss 0.128683 Objective Loss 0.128683 LR 0.000125 Time 0.021808 -2022-12-06 11:45:08,613 - Epoch: [184][ 840/ 1200] Overall Loss 0.128655 Objective Loss 0.128655 LR 0.000125 Time 0.021786 -2022-12-06 11:45:08,810 - Epoch: [184][ 850/ 1200] Overall Loss 0.128719 Objective Loss 0.128719 LR 0.000125 Time 0.021761 -2022-12-06 11:45:09,010 - Epoch: [184][ 860/ 1200] Overall Loss 0.128820 Objective Loss 0.128820 LR 0.000125 Time 0.021740 -2022-12-06 11:45:09,207 - Epoch: [184][ 870/ 1200] Overall Loss 0.128696 Objective Loss 0.128696 LR 0.000125 Time 0.021716 -2022-12-06 11:45:09,407 - Epoch: [184][ 880/ 1200] Overall Loss 0.128727 Objective Loss 0.128727 LR 0.000125 Time 0.021696 -2022-12-06 11:45:09,605 - Epoch: [184][ 890/ 1200] Overall Loss 0.129004 Objective Loss 0.129004 LR 0.000125 Time 0.021673 -2022-12-06 11:45:09,805 - Epoch: [184][ 900/ 1200] Overall Loss 0.128793 Objective Loss 0.128793 LR 0.000125 Time 0.021655 -2022-12-06 11:45:10,002 - Epoch: [184][ 910/ 1200] Overall Loss 0.128855 Objective Loss 0.128855 LR 0.000125 Time 0.021633 -2022-12-06 11:45:10,202 - Epoch: [184][ 920/ 1200] Overall Loss 0.128808 Objective Loss 0.128808 LR 0.000125 Time 0.021614 -2022-12-06 11:45:10,399 - Epoch: [184][ 930/ 1200] Overall Loss 0.128890 Objective Loss 0.128890 LR 0.000125 Time 0.021593 -2022-12-06 11:45:10,599 - Epoch: [184][ 940/ 1200] Overall Loss 0.128793 Objective Loss 0.128793 LR 0.000125 Time 0.021576 -2022-12-06 11:45:10,796 - Epoch: [184][ 950/ 1200] Overall Loss 0.128811 Objective Loss 0.128811 LR 0.000125 Time 0.021556 -2022-12-06 11:45:10,997 - Epoch: [184][ 960/ 1200] Overall Loss 0.128832 Objective Loss 0.128832 LR 0.000125 Time 0.021540 -2022-12-06 11:45:11,195 - Epoch: [184][ 970/ 1200] Overall Loss 0.128740 Objective Loss 0.128740 LR 0.000125 Time 0.021521 -2022-12-06 11:45:11,395 - Epoch: [184][ 980/ 1200] Overall Loss 0.128792 Objective Loss 0.128792 LR 0.000125 Time 0.021505 -2022-12-06 11:45:11,593 - Epoch: [184][ 990/ 1200] Overall Loss 0.128905 Objective Loss 0.128905 LR 0.000125 Time 0.021487 -2022-12-06 11:45:11,792 - Epoch: [184][ 1000/ 1200] Overall Loss 0.129150 Objective Loss 0.129150 LR 0.000125 Time 0.021471 -2022-12-06 11:45:11,990 - Epoch: [184][ 1010/ 1200] Overall Loss 0.129047 Objective Loss 0.129047 LR 0.000125 Time 0.021453 -2022-12-06 11:45:12,190 - Epoch: [184][ 1020/ 1200] Overall Loss 0.129077 Objective Loss 0.129077 LR 0.000125 Time 0.021438 -2022-12-06 11:45:12,386 - Epoch: [184][ 1030/ 1200] Overall Loss 0.129077 Objective Loss 0.129077 LR 0.000125 Time 0.021421 -2022-12-06 11:45:12,586 - Epoch: [184][ 1040/ 1200] Overall Loss 0.129058 Objective Loss 0.129058 LR 0.000125 Time 0.021406 -2022-12-06 11:45:12,783 - Epoch: [184][ 1050/ 1200] Overall Loss 0.129149 Objective Loss 0.129149 LR 0.000125 Time 0.021390 -2022-12-06 11:45:12,983 - Epoch: [184][ 1060/ 1200] Overall Loss 0.129190 Objective Loss 0.129190 LR 0.000125 Time 0.021376 -2022-12-06 11:45:13,180 - Epoch: [184][ 1070/ 1200] Overall Loss 0.129153 Objective Loss 0.129153 LR 0.000125 Time 0.021359 -2022-12-06 11:45:13,379 - Epoch: [184][ 1080/ 1200] Overall Loss 0.129127 Objective Loss 0.129127 LR 0.000125 Time 0.021346 -2022-12-06 11:45:13,576 - Epoch: [184][ 1090/ 1200] Overall Loss 0.129264 Objective Loss 0.129264 LR 0.000125 Time 0.021330 -2022-12-06 11:45:13,776 - Epoch: [184][ 1100/ 1200] Overall Loss 0.129200 Objective Loss 0.129200 LR 0.000125 Time 0.021317 -2022-12-06 11:45:13,973 - Epoch: [184][ 1110/ 1200] Overall Loss 0.129382 Objective Loss 0.129382 LR 0.000125 Time 0.021302 -2022-12-06 11:45:14,173 - Epoch: [184][ 1120/ 1200] Overall Loss 0.129437 Objective Loss 0.129437 LR 0.000125 Time 0.021290 -2022-12-06 11:45:14,370 - Epoch: [184][ 1130/ 1200] Overall Loss 0.129551 Objective Loss 0.129551 LR 0.000125 Time 0.021276 -2022-12-06 11:45:14,575 - Epoch: [184][ 1140/ 1200] Overall Loss 0.129634 Objective Loss 0.129634 LR 0.000125 Time 0.021268 -2022-12-06 11:45:14,781 - Epoch: [184][ 1150/ 1200] Overall Loss 0.129583 Objective Loss 0.129583 LR 0.000125 Time 0.021262 -2022-12-06 11:45:14,994 - Epoch: [184][ 1160/ 1200] Overall Loss 0.129463 Objective Loss 0.129463 LR 0.000125 Time 0.021262 -2022-12-06 11:45:15,200 - Epoch: [184][ 1170/ 1200] Overall Loss 0.129355 Objective Loss 0.129355 LR 0.000125 Time 0.021256 -2022-12-06 11:45:15,414 - Epoch: [184][ 1180/ 1200] Overall Loss 0.129388 Objective Loss 0.129388 LR 0.000125 Time 0.021256 -2022-12-06 11:45:15,620 - Epoch: [184][ 1190/ 1200] Overall Loss 0.129368 Objective Loss 0.129368 LR 0.000125 Time 0.021250 -2022-12-06 11:45:15,848 - Epoch: [184][ 1200/ 1200] Overall Loss 0.129388 Objective Loss 0.129388 Top1 89.539749 Top5 99.372385 LR 0.000125 Time 0.021263 -2022-12-06 11:45:15,937 - --- validate (epoch=184)----------- -2022-12-06 11:45:15,937 - 34129 samples (256 per mini-batch) -2022-12-06 11:45:16,384 - Epoch: [184][ 10/ 134] Loss 0.236219 Top1 87.773438 Top5 98.789062 -2022-12-06 11:45:16,516 - Epoch: [184][ 20/ 134] Loss 0.232315 Top1 88.007812 Top5 98.750000 -2022-12-06 11:45:16,649 - Epoch: [184][ 30/ 134] Loss 0.230354 Top1 88.085938 Top5 98.750000 -2022-12-06 11:45:16,781 - Epoch: [184][ 40/ 134] Loss 0.230014 Top1 88.193359 Top5 98.740234 -2022-12-06 11:45:16,912 - Epoch: [184][ 50/ 134] Loss 0.231026 Top1 88.218750 Top5 98.773438 -2022-12-06 11:45:17,046 - Epoch: [184][ 60/ 134] Loss 0.238527 Top1 88.098958 Top5 98.691406 -2022-12-06 11:45:17,177 - Epoch: [184][ 70/ 134] Loss 0.232771 Top1 88.258929 Top5 98.694196 -2022-12-06 11:45:17,309 - Epoch: [184][ 80/ 134] Loss 0.229835 Top1 88.432617 Top5 98.696289 -2022-12-06 11:45:17,442 - Epoch: [184][ 90/ 134] Loss 0.226649 Top1 88.420139 Top5 98.645833 -2022-12-06 11:45:17,573 - Epoch: [184][ 100/ 134] Loss 0.229226 Top1 88.402344 Top5 98.636719 -2022-12-06 11:45:17,706 - Epoch: [184][ 110/ 134] Loss 0.227692 Top1 88.487216 Top5 98.650568 -2022-12-06 11:45:17,839 - Epoch: [184][ 120/ 134] Loss 0.228316 Top1 88.476562 Top5 98.645833 -2022-12-06 11:45:17,973 - Epoch: [184][ 130/ 134] Loss 0.228225 Top1 88.527644 Top5 98.644832 -2022-12-06 11:45:18,011 - Epoch: [184][ 134/ 134] Loss 0.227519 Top1 88.555188 Top5 98.655103 -2022-12-06 11:45:18,100 - ==> Top1: 88.555 Top5: 98.655 Loss: 0.228 - -2022-12-06 11:45:18,101 - ==> Confusion: -[[ 917 1 1 1 4 7 0 0 5 44 0 2 1 2 4 1 0 0 3 0 3] - [ 1 944 2 2 9 20 4 11 4 1 0 4 0 0 0 0 3 2 9 1 10] - [ 4 3 1016 12 2 2 12 8 0 5 6 3 2 1 2 2 1 1 3 5 13] - [ 4 3 12 957 0 2 1 0 0 0 10 1 5 1 8 0 1 0 8 1 6] - [ 5 7 1 0 958 3 1 2 1 7 2 4 0 1 11 4 4 3 1 1 4] - [ 0 10 0 4 3 992 1 14 1 1 1 12 3 17 3 1 0 0 0 1 5] - [ 0 2 10 1 0 1 1074 3 0 0 1 2 0 1 0 4 0 5 2 8 4] - [ 0 8 3 1 2 26 4 969 0 1 1 6 0 1 1 0 1 0 15 10 5] - [ 3 1 0 0 0 1 1 1 992 35 7 1 2 4 9 1 1 0 1 1 3] - [ 41 2 0 0 5 3 0 2 21 907 1 0 0 10 3 0 1 1 1 0 3] - [ 0 2 3 3 1 0 0 2 11 2 967 0 0 7 4 1 0 0 7 2 7] - [ 2 0 2 0 1 15 2 2 1 1 0 990 15 1 0 6 3 3 0 3 4] - [ 0 0 1 1 1 4 0 0 0 1 0 22 909 2 2 7 1 8 0 3 7] - [ 1 1 0 0 0 6 0 3 14 14 2 4 3 958 1 1 3 1 0 1 10] - [ 6 4 0 8 2 4 0 0 14 2 0 3 1 4 1069 0 0 1 5 0 7] - [ 0 0 0 1 4 1 2 0 2 1 0 8 3 1 0 997 6 11 0 3 3] - [ 2 2 0 1 2 1 1 0 2 0 0 2 2 1 0 7 1032 1 0 7 9] - [ 3 0 2 1 0 1 0 0 0 4 0 10 9 2 1 13 0 986 0 1 3] - [ 4 3 3 7 0 3 0 25 1 1 4 2 2 0 7 1 0 2 939 1 3] - [ 1 3 1 2 0 7 3 5 0 1 1 12 4 6 0 5 4 1 2 1017 5] - [ 109 199 136 72 91 156 69 129 78 88 141 88 250 239 132 88 133 73 126 198 10631]] - -2022-12-06 11:45:18,685 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:45:18,685 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:45:18,691 - - -2022-12-06 11:45:18,691 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:45:19,641 - Epoch: [185][ 10/ 1200] Overall Loss 0.112329 Objective Loss 0.112329 LR 0.000125 Time 0.094894 -2022-12-06 11:45:19,839 - Epoch: [185][ 20/ 1200] Overall Loss 0.119719 Objective Loss 0.119719 LR 0.000125 Time 0.057342 -2022-12-06 11:45:20,040 - Epoch: [185][ 30/ 1200] Overall Loss 0.122368 Objective Loss 0.122368 LR 0.000125 Time 0.044902 -2022-12-06 11:45:20,238 - Epoch: [185][ 40/ 1200] Overall Loss 0.120779 Objective Loss 0.120779 LR 0.000125 Time 0.038607 -2022-12-06 11:45:20,437 - Epoch: [185][ 50/ 1200] Overall Loss 0.120723 Objective Loss 0.120723 LR 0.000125 Time 0.034856 -2022-12-06 11:45:20,633 - Epoch: [185][ 60/ 1200] Overall Loss 0.125487 Objective Loss 0.125487 LR 0.000125 Time 0.032310 -2022-12-06 11:45:20,833 - Epoch: [185][ 70/ 1200] Overall Loss 0.126945 Objective Loss 0.126945 LR 0.000125 Time 0.030538 -2022-12-06 11:45:21,030 - Epoch: [185][ 80/ 1200] Overall Loss 0.125871 Objective Loss 0.125871 LR 0.000125 Time 0.029176 -2022-12-06 11:45:21,230 - Epoch: [185][ 90/ 1200] Overall Loss 0.127139 Objective Loss 0.127139 LR 0.000125 Time 0.028150 -2022-12-06 11:45:21,427 - Epoch: [185][ 100/ 1200] Overall Loss 0.127637 Objective Loss 0.127637 LR 0.000125 Time 0.027299 -2022-12-06 11:45:21,627 - Epoch: [185][ 110/ 1200] Overall Loss 0.129379 Objective Loss 0.129379 LR 0.000125 Time 0.026635 -2022-12-06 11:45:21,824 - Epoch: [185][ 120/ 1200] Overall Loss 0.128285 Objective Loss 0.128285 LR 0.000125 Time 0.026048 -2022-12-06 11:45:22,024 - Epoch: [185][ 130/ 1200] Overall Loss 0.128058 Objective Loss 0.128058 LR 0.000125 Time 0.025582 -2022-12-06 11:45:22,222 - Epoch: [185][ 140/ 1200] Overall Loss 0.127666 Objective Loss 0.127666 LR 0.000125 Time 0.025161 -2022-12-06 11:45:22,422 - Epoch: [185][ 150/ 1200] Overall Loss 0.128347 Objective Loss 0.128347 LR 0.000125 Time 0.024813 -2022-12-06 11:45:22,619 - Epoch: [185][ 160/ 1200] Overall Loss 0.129641 Objective Loss 0.129641 LR 0.000125 Time 0.024492 -2022-12-06 11:45:22,820 - Epoch: [185][ 170/ 1200] Overall Loss 0.130889 Objective Loss 0.130889 LR 0.000125 Time 0.024230 -2022-12-06 11:45:23,016 - Epoch: [185][ 180/ 1200] Overall Loss 0.131043 Objective Loss 0.131043 LR 0.000125 Time 0.023974 -2022-12-06 11:45:23,217 - Epoch: [185][ 190/ 1200] Overall Loss 0.130218 Objective Loss 0.130218 LR 0.000125 Time 0.023763 -2022-12-06 11:45:23,414 - Epoch: [185][ 200/ 1200] Overall Loss 0.130752 Objective Loss 0.130752 LR 0.000125 Time 0.023557 -2022-12-06 11:45:23,614 - Epoch: [185][ 210/ 1200] Overall Loss 0.130894 Objective Loss 0.130894 LR 0.000125 Time 0.023388 -2022-12-06 11:45:23,811 - Epoch: [185][ 220/ 1200] Overall Loss 0.130685 Objective Loss 0.130685 LR 0.000125 Time 0.023217 -2022-12-06 11:45:24,012 - Epoch: [185][ 230/ 1200] Overall Loss 0.130384 Objective Loss 0.130384 LR 0.000125 Time 0.023077 -2022-12-06 11:45:24,209 - Epoch: [185][ 240/ 1200] Overall Loss 0.131042 Objective Loss 0.131042 LR 0.000125 Time 0.022934 -2022-12-06 11:45:24,408 - Epoch: [185][ 250/ 1200] Overall Loss 0.131027 Objective Loss 0.131027 LR 0.000125 Time 0.022814 -2022-12-06 11:45:24,606 - Epoch: [185][ 260/ 1200] Overall Loss 0.130596 Objective Loss 0.130596 LR 0.000125 Time 0.022692 -2022-12-06 11:45:24,806 - Epoch: [185][ 270/ 1200] Overall Loss 0.130813 Objective Loss 0.130813 LR 0.000125 Time 0.022592 -2022-12-06 11:45:25,003 - Epoch: [185][ 280/ 1200] Overall Loss 0.130147 Objective Loss 0.130147 LR 0.000125 Time 0.022486 -2022-12-06 11:45:25,203 - Epoch: [185][ 290/ 1200] Overall Loss 0.130535 Objective Loss 0.130535 LR 0.000125 Time 0.022400 -2022-12-06 11:45:25,400 - Epoch: [185][ 300/ 1200] Overall Loss 0.130963 Objective Loss 0.130963 LR 0.000125 Time 0.022308 -2022-12-06 11:45:25,601 - Epoch: [185][ 310/ 1200] Overall Loss 0.130855 Objective Loss 0.130855 LR 0.000125 Time 0.022233 -2022-12-06 11:45:25,798 - Epoch: [185][ 320/ 1200] Overall Loss 0.130672 Objective Loss 0.130672 LR 0.000125 Time 0.022153 -2022-12-06 11:45:25,998 - Epoch: [185][ 330/ 1200] Overall Loss 0.130798 Objective Loss 0.130798 LR 0.000125 Time 0.022086 -2022-12-06 11:45:26,195 - Epoch: [185][ 340/ 1200] Overall Loss 0.131114 Objective Loss 0.131114 LR 0.000125 Time 0.022014 -2022-12-06 11:45:26,394 - Epoch: [185][ 350/ 1200] Overall Loss 0.131170 Objective Loss 0.131170 LR 0.000125 Time 0.021954 -2022-12-06 11:45:26,591 - Epoch: [185][ 360/ 1200] Overall Loss 0.131036 Objective Loss 0.131036 LR 0.000125 Time 0.021890 -2022-12-06 11:45:26,792 - Epoch: [185][ 370/ 1200] Overall Loss 0.131269 Objective Loss 0.131269 LR 0.000125 Time 0.021838 -2022-12-06 11:45:26,988 - Epoch: [185][ 380/ 1200] Overall Loss 0.131226 Objective Loss 0.131226 LR 0.000125 Time 0.021779 -2022-12-06 11:45:27,188 - Epoch: [185][ 390/ 1200] Overall Loss 0.131067 Objective Loss 0.131067 LR 0.000125 Time 0.021732 -2022-12-06 11:45:27,385 - Epoch: [185][ 400/ 1200] Overall Loss 0.130933 Objective Loss 0.130933 LR 0.000125 Time 0.021679 -2022-12-06 11:45:27,585 - Epoch: [185][ 410/ 1200] Overall Loss 0.130650 Objective Loss 0.130650 LR 0.000125 Time 0.021637 -2022-12-06 11:45:27,782 - Epoch: [185][ 420/ 1200] Overall Loss 0.130572 Objective Loss 0.130572 LR 0.000125 Time 0.021590 -2022-12-06 11:45:27,982 - Epoch: [185][ 430/ 1200] Overall Loss 0.130455 Objective Loss 0.130455 LR 0.000125 Time 0.021551 -2022-12-06 11:45:28,179 - Epoch: [185][ 440/ 1200] Overall Loss 0.130733 Objective Loss 0.130733 LR 0.000125 Time 0.021507 -2022-12-06 11:45:28,378 - Epoch: [185][ 450/ 1200] Overall Loss 0.131049 Objective Loss 0.131049 LR 0.000125 Time 0.021472 -2022-12-06 11:45:28,576 - Epoch: [185][ 460/ 1200] Overall Loss 0.130681 Objective Loss 0.130681 LR 0.000125 Time 0.021434 -2022-12-06 11:45:28,777 - Epoch: [185][ 470/ 1200] Overall Loss 0.130323 Objective Loss 0.130323 LR 0.000125 Time 0.021403 -2022-12-06 11:45:28,975 - Epoch: [185][ 480/ 1200] Overall Loss 0.129991 Objective Loss 0.129991 LR 0.000125 Time 0.021369 -2022-12-06 11:45:29,176 - Epoch: [185][ 490/ 1200] Overall Loss 0.130042 Objective Loss 0.130042 LR 0.000125 Time 0.021342 -2022-12-06 11:45:29,373 - Epoch: [185][ 500/ 1200] Overall Loss 0.130121 Objective Loss 0.130121 LR 0.000125 Time 0.021308 -2022-12-06 11:45:29,573 - Epoch: [185][ 510/ 1200] Overall Loss 0.129982 Objective Loss 0.129982 LR 0.000125 Time 0.021283 -2022-12-06 11:45:29,770 - Epoch: [185][ 520/ 1200] Overall Loss 0.130279 Objective Loss 0.130279 LR 0.000125 Time 0.021251 -2022-12-06 11:45:29,970 - Epoch: [185][ 530/ 1200] Overall Loss 0.130475 Objective Loss 0.130475 LR 0.000125 Time 0.021227 -2022-12-06 11:45:30,168 - Epoch: [185][ 540/ 1200] Overall Loss 0.130131 Objective Loss 0.130131 LR 0.000125 Time 0.021198 -2022-12-06 11:45:30,367 - Epoch: [185][ 550/ 1200] Overall Loss 0.129965 Objective Loss 0.129965 LR 0.000125 Time 0.021174 -2022-12-06 11:45:30,564 - Epoch: [185][ 560/ 1200] Overall Loss 0.129873 Objective Loss 0.129873 LR 0.000125 Time 0.021146 -2022-12-06 11:45:30,764 - Epoch: [185][ 570/ 1200] Overall Loss 0.129758 Objective Loss 0.129758 LR 0.000125 Time 0.021126 -2022-12-06 11:45:30,961 - Epoch: [185][ 580/ 1200] Overall Loss 0.129834 Objective Loss 0.129834 LR 0.000125 Time 0.021100 -2022-12-06 11:45:31,161 - Epoch: [185][ 590/ 1200] Overall Loss 0.129622 Objective Loss 0.129622 LR 0.000125 Time 0.021081 -2022-12-06 11:45:31,358 - Epoch: [185][ 600/ 1200] Overall Loss 0.129910 Objective Loss 0.129910 LR 0.000125 Time 0.021057 -2022-12-06 11:45:31,558 - Epoch: [185][ 610/ 1200] Overall Loss 0.130013 Objective Loss 0.130013 LR 0.000125 Time 0.021038 -2022-12-06 11:45:31,755 - Epoch: [185][ 620/ 1200] Overall Loss 0.130004 Objective Loss 0.130004 LR 0.000125 Time 0.021016 -2022-12-06 11:45:31,956 - Epoch: [185][ 630/ 1200] Overall Loss 0.129720 Objective Loss 0.129720 LR 0.000125 Time 0.021000 -2022-12-06 11:45:32,153 - Epoch: [185][ 640/ 1200] Overall Loss 0.129508 Objective Loss 0.129508 LR 0.000125 Time 0.020978 -2022-12-06 11:45:32,353 - Epoch: [185][ 650/ 1200] Overall Loss 0.129660 Objective Loss 0.129660 LR 0.000125 Time 0.020963 -2022-12-06 11:45:32,550 - Epoch: [185][ 660/ 1200] Overall Loss 0.129545 Objective Loss 0.129545 LR 0.000125 Time 0.020943 -2022-12-06 11:45:32,750 - Epoch: [185][ 670/ 1200] Overall Loss 0.129659 Objective Loss 0.129659 LR 0.000125 Time 0.020928 -2022-12-06 11:45:32,947 - Epoch: [185][ 680/ 1200] Overall Loss 0.129754 Objective Loss 0.129754 LR 0.000125 Time 0.020909 -2022-12-06 11:45:33,146 - Epoch: [185][ 690/ 1200] Overall Loss 0.129876 Objective Loss 0.129876 LR 0.000125 Time 0.020894 -2022-12-06 11:45:33,343 - Epoch: [185][ 700/ 1200] Overall Loss 0.130083 Objective Loss 0.130083 LR 0.000125 Time 0.020876 -2022-12-06 11:45:33,542 - Epoch: [185][ 710/ 1200] Overall Loss 0.130166 Objective Loss 0.130166 LR 0.000125 Time 0.020862 -2022-12-06 11:45:33,739 - Epoch: [185][ 720/ 1200] Overall Loss 0.130377 Objective Loss 0.130377 LR 0.000125 Time 0.020845 -2022-12-06 11:45:33,939 - Epoch: [185][ 730/ 1200] Overall Loss 0.130364 Objective Loss 0.130364 LR 0.000125 Time 0.020833 -2022-12-06 11:45:34,136 - Epoch: [185][ 740/ 1200] Overall Loss 0.130455 Objective Loss 0.130455 LR 0.000125 Time 0.020817 -2022-12-06 11:45:34,336 - Epoch: [185][ 750/ 1200] Overall Loss 0.130334 Objective Loss 0.130334 LR 0.000125 Time 0.020805 -2022-12-06 11:45:34,534 - Epoch: [185][ 760/ 1200] Overall Loss 0.130500 Objective Loss 0.130500 LR 0.000125 Time 0.020790 -2022-12-06 11:45:34,734 - Epoch: [185][ 770/ 1200] Overall Loss 0.130464 Objective Loss 0.130464 LR 0.000125 Time 0.020780 -2022-12-06 11:45:34,932 - Epoch: [185][ 780/ 1200] Overall Loss 0.130487 Objective Loss 0.130487 LR 0.000125 Time 0.020766 -2022-12-06 11:45:35,132 - Epoch: [185][ 790/ 1200] Overall Loss 0.130575 Objective Loss 0.130575 LR 0.000125 Time 0.020756 -2022-12-06 11:45:35,329 - Epoch: [185][ 800/ 1200] Overall Loss 0.130484 Objective Loss 0.130484 LR 0.000125 Time 0.020742 -2022-12-06 11:45:35,529 - Epoch: [185][ 810/ 1200] Overall Loss 0.130296 Objective Loss 0.130296 LR 0.000125 Time 0.020732 -2022-12-06 11:45:35,726 - Epoch: [185][ 820/ 1200] Overall Loss 0.130419 Objective Loss 0.130419 LR 0.000125 Time 0.020719 -2022-12-06 11:45:35,925 - Epoch: [185][ 830/ 1200] Overall Loss 0.130336 Objective Loss 0.130336 LR 0.000125 Time 0.020709 -2022-12-06 11:45:36,122 - Epoch: [185][ 840/ 1200] Overall Loss 0.130240 Objective Loss 0.130240 LR 0.000125 Time 0.020695 -2022-12-06 11:45:36,321 - Epoch: [185][ 850/ 1200] Overall Loss 0.130342 Objective Loss 0.130342 LR 0.000125 Time 0.020686 -2022-12-06 11:45:36,518 - Epoch: [185][ 860/ 1200] Overall Loss 0.130428 Objective Loss 0.130428 LR 0.000125 Time 0.020674 -2022-12-06 11:45:36,718 - Epoch: [185][ 870/ 1200] Overall Loss 0.130465 Objective Loss 0.130465 LR 0.000125 Time 0.020665 -2022-12-06 11:45:36,915 - Epoch: [185][ 880/ 1200] Overall Loss 0.130360 Objective Loss 0.130360 LR 0.000125 Time 0.020654 -2022-12-06 11:45:37,116 - Epoch: [185][ 890/ 1200] Overall Loss 0.130398 Objective Loss 0.130398 LR 0.000125 Time 0.020647 -2022-12-06 11:45:37,313 - Epoch: [185][ 900/ 1200] Overall Loss 0.130310 Objective Loss 0.130310 LR 0.000125 Time 0.020636 -2022-12-06 11:45:37,512 - Epoch: [185][ 910/ 1200] Overall Loss 0.130397 Objective Loss 0.130397 LR 0.000125 Time 0.020628 -2022-12-06 11:45:37,709 - Epoch: [185][ 920/ 1200] Overall Loss 0.130492 Objective Loss 0.130492 LR 0.000125 Time 0.020617 -2022-12-06 11:45:37,910 - Epoch: [185][ 930/ 1200] Overall Loss 0.130520 Objective Loss 0.130520 LR 0.000125 Time 0.020610 -2022-12-06 11:45:38,107 - Epoch: [185][ 940/ 1200] Overall Loss 0.130515 Objective Loss 0.130515 LR 0.000125 Time 0.020600 -2022-12-06 11:45:38,308 - Epoch: [185][ 950/ 1200] Overall Loss 0.130641 Objective Loss 0.130641 LR 0.000125 Time 0.020594 -2022-12-06 11:45:38,504 - Epoch: [185][ 960/ 1200] Overall Loss 0.130778 Objective Loss 0.130778 LR 0.000125 Time 0.020584 -2022-12-06 11:45:38,705 - Epoch: [185][ 970/ 1200] Overall Loss 0.130603 Objective Loss 0.130603 LR 0.000125 Time 0.020578 -2022-12-06 11:45:38,902 - Epoch: [185][ 980/ 1200] Overall Loss 0.130566 Objective Loss 0.130566 LR 0.000125 Time 0.020568 -2022-12-06 11:45:39,102 - Epoch: [185][ 990/ 1200] Overall Loss 0.130401 Objective Loss 0.130401 LR 0.000125 Time 0.020562 -2022-12-06 11:45:39,299 - Epoch: [185][ 1000/ 1200] Overall Loss 0.130448 Objective Loss 0.130448 LR 0.000125 Time 0.020553 -2022-12-06 11:45:39,500 - Epoch: [185][ 1010/ 1200] Overall Loss 0.130405 Objective Loss 0.130405 LR 0.000125 Time 0.020548 -2022-12-06 11:45:39,697 - Epoch: [185][ 1020/ 1200] Overall Loss 0.130415 Objective Loss 0.130415 LR 0.000125 Time 0.020539 -2022-12-06 11:45:39,897 - Epoch: [185][ 1030/ 1200] Overall Loss 0.130269 Objective Loss 0.130269 LR 0.000125 Time 0.020533 -2022-12-06 11:45:40,094 - Epoch: [185][ 1040/ 1200] Overall Loss 0.130360 Objective Loss 0.130360 LR 0.000125 Time 0.020525 -2022-12-06 11:45:40,294 - Epoch: [185][ 1050/ 1200] Overall Loss 0.130323 Objective Loss 0.130323 LR 0.000125 Time 0.020519 -2022-12-06 11:45:40,490 - Epoch: [185][ 1060/ 1200] Overall Loss 0.130305 Objective Loss 0.130305 LR 0.000125 Time 0.020510 -2022-12-06 11:45:40,691 - Epoch: [185][ 1070/ 1200] Overall Loss 0.130330 Objective Loss 0.130330 LR 0.000125 Time 0.020506 -2022-12-06 11:45:40,888 - Epoch: [185][ 1080/ 1200] Overall Loss 0.130290 Objective Loss 0.130290 LR 0.000125 Time 0.020498 -2022-12-06 11:45:41,088 - Epoch: [185][ 1090/ 1200] Overall Loss 0.130383 Objective Loss 0.130383 LR 0.000125 Time 0.020492 -2022-12-06 11:45:41,284 - Epoch: [185][ 1100/ 1200] Overall Loss 0.130420 Objective Loss 0.130420 LR 0.000125 Time 0.020484 -2022-12-06 11:45:41,485 - Epoch: [185][ 1110/ 1200] Overall Loss 0.130201 Objective Loss 0.130201 LR 0.000125 Time 0.020480 -2022-12-06 11:45:41,682 - Epoch: [185][ 1120/ 1200] Overall Loss 0.130285 Objective Loss 0.130285 LR 0.000125 Time 0.020472 -2022-12-06 11:45:41,882 - Epoch: [185][ 1130/ 1200] Overall Loss 0.130352 Objective Loss 0.130352 LR 0.000125 Time 0.020468 -2022-12-06 11:45:42,079 - Epoch: [185][ 1140/ 1200] Overall Loss 0.130456 Objective Loss 0.130456 LR 0.000125 Time 0.020461 -2022-12-06 11:45:42,279 - Epoch: [185][ 1150/ 1200] Overall Loss 0.130387 Objective Loss 0.130387 LR 0.000125 Time 0.020456 -2022-12-06 11:45:42,476 - Epoch: [185][ 1160/ 1200] Overall Loss 0.130357 Objective Loss 0.130357 LR 0.000125 Time 0.020449 -2022-12-06 11:45:42,676 - Epoch: [185][ 1170/ 1200] Overall Loss 0.130238 Objective Loss 0.130238 LR 0.000125 Time 0.020445 -2022-12-06 11:45:42,873 - Epoch: [185][ 1180/ 1200] Overall Loss 0.130096 Objective Loss 0.130096 LR 0.000125 Time 0.020438 -2022-12-06 11:45:43,073 - Epoch: [185][ 1190/ 1200] Overall Loss 0.129991 Objective Loss 0.129991 LR 0.000125 Time 0.020434 -2022-12-06 11:45:43,305 - Epoch: [185][ 1200/ 1200] Overall Loss 0.129950 Objective Loss 0.129950 Top1 89.539749 Top5 98.326360 LR 0.000125 Time 0.020456 -2022-12-06 11:45:43,393 - --- validate (epoch=185)----------- -2022-12-06 11:45:43,393 - 34129 samples (256 per mini-batch) -2022-12-06 11:45:43,836 - Epoch: [185][ 10/ 134] Loss 0.198810 Top1 88.671875 Top5 98.789062 -2022-12-06 11:45:43,967 - Epoch: [185][ 20/ 134] Loss 0.223046 Top1 88.242188 Top5 98.671875 -2022-12-06 11:45:44,097 - Epoch: [185][ 30/ 134] Loss 0.227905 Top1 88.346354 Top5 98.593750 -2022-12-06 11:45:44,227 - Epoch: [185][ 40/ 134] Loss 0.223145 Top1 88.222656 Top5 98.544922 -2022-12-06 11:45:44,357 - Epoch: [185][ 50/ 134] Loss 0.220335 Top1 88.296875 Top5 98.632812 -2022-12-06 11:45:44,486 - Epoch: [185][ 60/ 134] Loss 0.219792 Top1 88.287760 Top5 98.626302 -2022-12-06 11:45:44,617 - Epoch: [185][ 70/ 134] Loss 0.219279 Top1 88.292411 Top5 98.621652 -2022-12-06 11:45:44,746 - Epoch: [185][ 80/ 134] Loss 0.221895 Top1 88.247070 Top5 98.608398 -2022-12-06 11:45:44,876 - Epoch: [185][ 90/ 134] Loss 0.223456 Top1 88.237847 Top5 98.606771 -2022-12-06 11:45:45,005 - Epoch: [185][ 100/ 134] Loss 0.225625 Top1 88.277344 Top5 98.582031 -2022-12-06 11:45:45,136 - Epoch: [185][ 110/ 134] Loss 0.225068 Top1 88.359375 Top5 98.611506 -2022-12-06 11:45:45,265 - Epoch: [185][ 120/ 134] Loss 0.228488 Top1 88.310547 Top5 98.590495 -2022-12-06 11:45:45,397 - Epoch: [185][ 130/ 134] Loss 0.229082 Top1 88.311298 Top5 98.575721 -2022-12-06 11:45:45,435 - Epoch: [185][ 134/ 134] Loss 0.230068 Top1 88.355944 Top5 98.596502 -2022-12-06 11:45:45,522 - ==> Top1: 88.356 Top5: 98.597 Loss: 0.230 - -2022-12-06 11:45:45,523 - ==> Confusion: -[[ 925 1 1 2 10 6 0 0 3 35 0 2 1 2 3 1 1 1 1 0 1] - [ 0 938 2 2 13 22 2 14 2 0 2 4 1 0 0 1 3 1 7 5 8] - [ 4 2 1013 11 5 3 18 7 0 2 4 3 2 3 3 4 2 2 3 3 9] - [ 4 0 14 957 1 4 1 0 0 1 8 0 3 1 7 0 2 0 10 1 6] - [ 4 4 1 0 967 2 1 2 1 4 2 2 0 2 8 5 7 3 1 1 3] - [ 0 9 0 5 6 987 2 14 2 2 1 12 4 12 1 1 0 0 0 5 6] - [ 2 1 8 3 0 1 1083 2 0 0 0 1 0 3 0 3 0 2 2 6 1] - [ 0 6 5 2 2 27 13 958 0 0 1 8 0 1 0 1 0 1 13 10 6] - [ 4 1 0 1 0 0 1 2 986 35 6 1 2 8 9 1 2 0 1 1 3] - [ 55 2 2 0 5 1 0 4 20 886 1 0 0 12 4 1 0 1 1 0 6] - [ 0 2 4 4 1 0 2 3 11 1 958 0 0 12 4 1 0 0 8 0 8] - [ 1 0 2 0 1 7 5 1 1 0 0 988 18 3 1 5 2 5 0 7 4] - [ 0 1 1 1 0 2 1 1 1 1 0 21 915 1 2 7 1 5 0 3 5] - [ 2 1 0 0 2 5 0 2 9 7 2 3 4 974 0 1 1 0 1 2 7] - [ 6 2 2 10 3 4 0 1 9 1 0 2 1 4 1070 0 1 1 4 1 8] - [ 1 0 0 1 3 0 3 1 0 1 2 8 5 1 0 1000 3 11 0 2 1] - [ 1 0 0 2 3 0 0 0 1 0 0 1 2 1 0 14 1037 0 0 4 6] - [ 4 0 1 2 2 1 1 1 0 4 0 7 15 2 1 14 0 979 0 0 2] - [ 3 2 4 7 0 3 0 23 0 1 4 2 1 0 9 1 0 2 940 2 4] - [ 3 3 0 1 0 5 4 5 0 1 2 11 5 6 0 4 3 2 1 1018 6] - [ 103 180 147 93 110 144 92 118 69 68 124 89 294 233 131 111 146 72 120 208 10574]] - -2022-12-06 11:45:46,201 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:45:46,201 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:45:46,207 - - -2022-12-06 11:45:46,207 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:45:47,151 - Epoch: [186][ 10/ 1200] Overall Loss 0.115475 Objective Loss 0.115475 LR 0.000125 Time 0.094313 -2022-12-06 11:45:47,348 - Epoch: [186][ 20/ 1200] Overall Loss 0.125048 Objective Loss 0.125048 LR 0.000125 Time 0.056989 -2022-12-06 11:45:47,541 - Epoch: [186][ 30/ 1200] Overall Loss 0.125892 Objective Loss 0.125892 LR 0.000125 Time 0.044399 -2022-12-06 11:45:47,733 - Epoch: [186][ 40/ 1200] Overall Loss 0.123254 Objective Loss 0.123254 LR 0.000125 Time 0.038079 -2022-12-06 11:45:47,925 - Epoch: [186][ 50/ 1200] Overall Loss 0.123818 Objective Loss 0.123818 LR 0.000125 Time 0.034290 -2022-12-06 11:45:48,116 - Epoch: [186][ 60/ 1200] Overall Loss 0.125405 Objective Loss 0.125405 LR 0.000125 Time 0.031750 -2022-12-06 11:45:48,308 - Epoch: [186][ 70/ 1200] Overall Loss 0.129987 Objective Loss 0.129987 LR 0.000125 Time 0.029954 -2022-12-06 11:45:48,500 - Epoch: [186][ 80/ 1200] Overall Loss 0.129279 Objective Loss 0.129279 LR 0.000125 Time 0.028599 -2022-12-06 11:45:48,691 - Epoch: [186][ 90/ 1200] Overall Loss 0.127196 Objective Loss 0.127196 LR 0.000125 Time 0.027543 -2022-12-06 11:45:48,883 - Epoch: [186][ 100/ 1200] Overall Loss 0.127799 Objective Loss 0.127799 LR 0.000125 Time 0.026696 -2022-12-06 11:45:49,074 - Epoch: [186][ 110/ 1200] Overall Loss 0.126289 Objective Loss 0.126289 LR 0.000125 Time 0.026001 -2022-12-06 11:45:49,264 - Epoch: [186][ 120/ 1200] Overall Loss 0.127823 Objective Loss 0.127823 LR 0.000125 Time 0.025418 -2022-12-06 11:45:49,456 - Epoch: [186][ 130/ 1200] Overall Loss 0.127800 Objective Loss 0.127800 LR 0.000125 Time 0.024932 -2022-12-06 11:45:49,646 - Epoch: [186][ 140/ 1200] Overall Loss 0.129035 Objective Loss 0.129035 LR 0.000125 Time 0.024510 -2022-12-06 11:45:49,837 - Epoch: [186][ 150/ 1200] Overall Loss 0.128605 Objective Loss 0.128605 LR 0.000125 Time 0.024145 -2022-12-06 11:45:50,030 - Epoch: [186][ 160/ 1200] Overall Loss 0.128549 Objective Loss 0.128549 LR 0.000125 Time 0.023834 -2022-12-06 11:45:50,221 - Epoch: [186][ 170/ 1200] Overall Loss 0.129057 Objective Loss 0.129057 LR 0.000125 Time 0.023555 -2022-12-06 11:45:50,413 - Epoch: [186][ 180/ 1200] Overall Loss 0.128183 Objective Loss 0.128183 LR 0.000125 Time 0.023310 -2022-12-06 11:45:50,604 - Epoch: [186][ 190/ 1200] Overall Loss 0.128375 Objective Loss 0.128375 LR 0.000125 Time 0.023087 -2022-12-06 11:45:50,795 - Epoch: [186][ 200/ 1200] Overall Loss 0.127259 Objective Loss 0.127259 LR 0.000125 Time 0.022885 -2022-12-06 11:45:50,988 - Epoch: [186][ 210/ 1200] Overall Loss 0.128259 Objective Loss 0.128259 LR 0.000125 Time 0.022708 -2022-12-06 11:45:51,179 - Epoch: [186][ 220/ 1200] Overall Loss 0.127771 Objective Loss 0.127771 LR 0.000125 Time 0.022544 -2022-12-06 11:45:51,371 - Epoch: [186][ 230/ 1200] Overall Loss 0.127429 Objective Loss 0.127429 LR 0.000125 Time 0.022395 -2022-12-06 11:45:51,562 - Epoch: [186][ 240/ 1200] Overall Loss 0.126715 Objective Loss 0.126715 LR 0.000125 Time 0.022256 -2022-12-06 11:45:51,753 - Epoch: [186][ 250/ 1200] Overall Loss 0.126333 Objective Loss 0.126333 LR 0.000125 Time 0.022129 -2022-12-06 11:45:51,945 - Epoch: [186][ 260/ 1200] Overall Loss 0.127373 Objective Loss 0.127373 LR 0.000125 Time 0.022013 -2022-12-06 11:45:52,137 - Epoch: [186][ 270/ 1200] Overall Loss 0.128590 Objective Loss 0.128590 LR 0.000125 Time 0.021906 -2022-12-06 11:45:52,328 - Epoch: [186][ 280/ 1200] Overall Loss 0.128599 Objective Loss 0.128599 LR 0.000125 Time 0.021803 -2022-12-06 11:45:52,520 - Epoch: [186][ 290/ 1200] Overall Loss 0.129414 Objective Loss 0.129414 LR 0.000125 Time 0.021712 -2022-12-06 11:45:52,711 - Epoch: [186][ 300/ 1200] Overall Loss 0.129561 Objective Loss 0.129561 LR 0.000125 Time 0.021625 -2022-12-06 11:45:52,903 - Epoch: [186][ 310/ 1200] Overall Loss 0.129693 Objective Loss 0.129693 LR 0.000125 Time 0.021543 -2022-12-06 11:45:53,094 - Epoch: [186][ 320/ 1200] Overall Loss 0.129253 Objective Loss 0.129253 LR 0.000125 Time 0.021464 -2022-12-06 11:45:53,288 - Epoch: [186][ 330/ 1200] Overall Loss 0.128604 Objective Loss 0.128604 LR 0.000125 Time 0.021401 -2022-12-06 11:45:53,484 - Epoch: [186][ 340/ 1200] Overall Loss 0.128268 Objective Loss 0.128268 LR 0.000125 Time 0.021346 -2022-12-06 11:45:53,679 - Epoch: [186][ 350/ 1200] Overall Loss 0.128158 Objective Loss 0.128158 LR 0.000125 Time 0.021292 -2022-12-06 11:45:53,876 - Epoch: [186][ 360/ 1200] Overall Loss 0.128183 Objective Loss 0.128183 LR 0.000125 Time 0.021246 -2022-12-06 11:45:54,070 - Epoch: [186][ 370/ 1200] Overall Loss 0.128203 Objective Loss 0.128203 LR 0.000125 Time 0.021196 -2022-12-06 11:45:54,266 - Epoch: [186][ 380/ 1200] Overall Loss 0.127900 Objective Loss 0.127900 LR 0.000125 Time 0.021153 -2022-12-06 11:45:54,461 - Epoch: [186][ 390/ 1200] Overall Loss 0.127854 Objective Loss 0.127854 LR 0.000125 Time 0.021109 -2022-12-06 11:45:54,658 - Epoch: [186][ 400/ 1200] Overall Loss 0.127305 Objective Loss 0.127305 LR 0.000125 Time 0.021071 -2022-12-06 11:45:54,852 - Epoch: [186][ 410/ 1200] Overall Loss 0.127108 Objective Loss 0.127108 LR 0.000125 Time 0.021030 -2022-12-06 11:45:55,049 - Epoch: [186][ 420/ 1200] Overall Loss 0.127555 Objective Loss 0.127555 LR 0.000125 Time 0.020997 -2022-12-06 11:45:55,244 - Epoch: [186][ 430/ 1200] Overall Loss 0.127308 Objective Loss 0.127308 LR 0.000125 Time 0.020960 -2022-12-06 11:45:55,440 - Epoch: [186][ 440/ 1200] Overall Loss 0.127177 Objective Loss 0.127177 LR 0.000125 Time 0.020928 -2022-12-06 11:45:55,636 - Epoch: [186][ 450/ 1200] Overall Loss 0.127311 Objective Loss 0.127311 LR 0.000125 Time 0.020897 -2022-12-06 11:45:55,832 - Epoch: [186][ 460/ 1200] Overall Loss 0.127248 Objective Loss 0.127248 LR 0.000125 Time 0.020868 -2022-12-06 11:45:56,026 - Epoch: [186][ 470/ 1200] Overall Loss 0.127312 Objective Loss 0.127312 LR 0.000125 Time 0.020837 -2022-12-06 11:45:56,223 - Epoch: [186][ 480/ 1200] Overall Loss 0.127280 Objective Loss 0.127280 LR 0.000125 Time 0.020811 -2022-12-06 11:45:56,418 - Epoch: [186][ 490/ 1200] Overall Loss 0.127054 Objective Loss 0.127054 LR 0.000125 Time 0.020783 -2022-12-06 11:45:56,615 - Epoch: [186][ 500/ 1200] Overall Loss 0.126777 Objective Loss 0.126777 LR 0.000125 Time 0.020760 -2022-12-06 11:45:56,809 - Epoch: [186][ 510/ 1200] Overall Loss 0.126695 Objective Loss 0.126695 LR 0.000125 Time 0.020733 -2022-12-06 11:45:57,005 - Epoch: [186][ 520/ 1200] Overall Loss 0.126888 Objective Loss 0.126888 LR 0.000125 Time 0.020710 -2022-12-06 11:45:57,199 - Epoch: [186][ 530/ 1200] Overall Loss 0.127113 Objective Loss 0.127113 LR 0.000125 Time 0.020685 -2022-12-06 11:45:57,395 - Epoch: [186][ 540/ 1200] Overall Loss 0.126966 Objective Loss 0.126966 LR 0.000125 Time 0.020664 -2022-12-06 11:45:57,590 - Epoch: [186][ 550/ 1200] Overall Loss 0.127071 Objective Loss 0.127071 LR 0.000125 Time 0.020641 -2022-12-06 11:45:57,786 - Epoch: [186][ 560/ 1200] Overall Loss 0.127121 Objective Loss 0.127121 LR 0.000125 Time 0.020622 -2022-12-06 11:45:57,981 - Epoch: [186][ 570/ 1200] Overall Loss 0.127438 Objective Loss 0.127438 LR 0.000125 Time 0.020600 -2022-12-06 11:45:58,177 - Epoch: [186][ 580/ 1200] Overall Loss 0.127926 Objective Loss 0.127926 LR 0.000125 Time 0.020583 -2022-12-06 11:45:58,372 - Epoch: [186][ 590/ 1200] Overall Loss 0.128057 Objective Loss 0.128057 LR 0.000125 Time 0.020563 -2022-12-06 11:45:58,568 - Epoch: [186][ 600/ 1200] Overall Loss 0.128316 Objective Loss 0.128316 LR 0.000125 Time 0.020546 -2022-12-06 11:45:58,762 - Epoch: [186][ 610/ 1200] Overall Loss 0.128081 Objective Loss 0.128081 LR 0.000125 Time 0.020527 -2022-12-06 11:45:58,959 - Epoch: [186][ 620/ 1200] Overall Loss 0.127983 Objective Loss 0.127983 LR 0.000125 Time 0.020512 -2022-12-06 11:45:59,154 - Epoch: [186][ 630/ 1200] Overall Loss 0.127740 Objective Loss 0.127740 LR 0.000125 Time 0.020496 -2022-12-06 11:45:59,351 - Epoch: [186][ 640/ 1200] Overall Loss 0.127962 Objective Loss 0.127962 LR 0.000125 Time 0.020481 -2022-12-06 11:45:59,545 - Epoch: [186][ 650/ 1200] Overall Loss 0.127615 Objective Loss 0.127615 LR 0.000125 Time 0.020465 -2022-12-06 11:45:59,741 - Epoch: [186][ 660/ 1200] Overall Loss 0.127633 Objective Loss 0.127633 LR 0.000125 Time 0.020451 -2022-12-06 11:45:59,938 - Epoch: [186][ 670/ 1200] Overall Loss 0.127588 Objective Loss 0.127588 LR 0.000125 Time 0.020439 -2022-12-06 11:46:00,136 - Epoch: [186][ 680/ 1200] Overall Loss 0.127512 Objective Loss 0.127512 LR 0.000125 Time 0.020429 -2022-12-06 11:46:00,333 - Epoch: [186][ 690/ 1200] Overall Loss 0.127488 Objective Loss 0.127488 LR 0.000125 Time 0.020417 -2022-12-06 11:46:00,531 - Epoch: [186][ 700/ 1200] Overall Loss 0.127392 Objective Loss 0.127392 LR 0.000125 Time 0.020408 -2022-12-06 11:46:00,727 - Epoch: [186][ 710/ 1200] Overall Loss 0.127381 Objective Loss 0.127381 LR 0.000125 Time 0.020395 -2022-12-06 11:46:00,925 - Epoch: [186][ 720/ 1200] Overall Loss 0.127239 Objective Loss 0.127239 LR 0.000125 Time 0.020386 -2022-12-06 11:46:01,120 - Epoch: [186][ 730/ 1200] Overall Loss 0.127163 Objective Loss 0.127163 LR 0.000125 Time 0.020373 -2022-12-06 11:46:01,318 - Epoch: [186][ 740/ 1200] Overall Loss 0.127227 Objective Loss 0.127227 LR 0.000125 Time 0.020365 -2022-12-06 11:46:01,513 - Epoch: [186][ 750/ 1200] Overall Loss 0.127427 Objective Loss 0.127427 LR 0.000125 Time 0.020353 -2022-12-06 11:46:01,712 - Epoch: [186][ 760/ 1200] Overall Loss 0.127231 Objective Loss 0.127231 LR 0.000125 Time 0.020346 -2022-12-06 11:46:01,908 - Epoch: [186][ 770/ 1200] Overall Loss 0.127192 Objective Loss 0.127192 LR 0.000125 Time 0.020335 -2022-12-06 11:46:02,107 - Epoch: [186][ 780/ 1200] Overall Loss 0.127199 Objective Loss 0.127199 LR 0.000125 Time 0.020329 -2022-12-06 11:46:02,302 - Epoch: [186][ 790/ 1200] Overall Loss 0.127209 Objective Loss 0.127209 LR 0.000125 Time 0.020318 -2022-12-06 11:46:02,499 - Epoch: [186][ 800/ 1200] Overall Loss 0.127254 Objective Loss 0.127254 LR 0.000125 Time 0.020310 -2022-12-06 11:46:02,695 - Epoch: [186][ 810/ 1200] Overall Loss 0.127370 Objective Loss 0.127370 LR 0.000125 Time 0.020300 -2022-12-06 11:46:02,893 - Epoch: [186][ 820/ 1200] Overall Loss 0.127445 Objective Loss 0.127445 LR 0.000125 Time 0.020294 -2022-12-06 11:46:03,089 - Epoch: [186][ 830/ 1200] Overall Loss 0.127350 Objective Loss 0.127350 LR 0.000125 Time 0.020285 -2022-12-06 11:46:03,288 - Epoch: [186][ 840/ 1200] Overall Loss 0.127375 Objective Loss 0.127375 LR 0.000125 Time 0.020279 -2022-12-06 11:46:03,483 - Epoch: [186][ 850/ 1200] Overall Loss 0.127469 Objective Loss 0.127469 LR 0.000125 Time 0.020270 -2022-12-06 11:46:03,681 - Epoch: [186][ 860/ 1200] Overall Loss 0.127561 Objective Loss 0.127561 LR 0.000125 Time 0.020264 -2022-12-06 11:46:03,877 - Epoch: [186][ 870/ 1200] Overall Loss 0.127532 Objective Loss 0.127532 LR 0.000125 Time 0.020255 -2022-12-06 11:46:04,075 - Epoch: [186][ 880/ 1200] Overall Loss 0.127424 Objective Loss 0.127424 LR 0.000125 Time 0.020250 -2022-12-06 11:46:04,271 - Epoch: [186][ 890/ 1200] Overall Loss 0.127468 Objective Loss 0.127468 LR 0.000125 Time 0.020241 -2022-12-06 11:46:04,468 - Epoch: [186][ 900/ 1200] Overall Loss 0.127509 Objective Loss 0.127509 LR 0.000125 Time 0.020235 -2022-12-06 11:46:04,664 - Epoch: [186][ 910/ 1200] Overall Loss 0.127428 Objective Loss 0.127428 LR 0.000125 Time 0.020227 -2022-12-06 11:46:04,861 - Epoch: [186][ 920/ 1200] Overall Loss 0.127337 Objective Loss 0.127337 LR 0.000125 Time 0.020221 -2022-12-06 11:46:05,057 - Epoch: [186][ 930/ 1200] Overall Loss 0.127269 Objective Loss 0.127269 LR 0.000125 Time 0.020213 -2022-12-06 11:46:05,255 - Epoch: [186][ 940/ 1200] Overall Loss 0.127225 Objective Loss 0.127225 LR 0.000125 Time 0.020209 -2022-12-06 11:46:05,450 - Epoch: [186][ 950/ 1200] Overall Loss 0.127295 Objective Loss 0.127295 LR 0.000125 Time 0.020201 -2022-12-06 11:46:05,648 - Epoch: [186][ 960/ 1200] Overall Loss 0.127121 Objective Loss 0.127121 LR 0.000125 Time 0.020196 -2022-12-06 11:46:05,843 - Epoch: [186][ 970/ 1200] Overall Loss 0.127096 Objective Loss 0.127096 LR 0.000125 Time 0.020188 -2022-12-06 11:46:06,042 - Epoch: [186][ 980/ 1200] Overall Loss 0.127135 Objective Loss 0.127135 LR 0.000125 Time 0.020185 -2022-12-06 11:46:06,237 - Epoch: [186][ 990/ 1200] Overall Loss 0.127060 Objective Loss 0.127060 LR 0.000125 Time 0.020177 -2022-12-06 11:46:06,436 - Epoch: [186][ 1000/ 1200] Overall Loss 0.127367 Objective Loss 0.127367 LR 0.000125 Time 0.020173 -2022-12-06 11:46:06,631 - Epoch: [186][ 1010/ 1200] Overall Loss 0.127387 Objective Loss 0.127387 LR 0.000125 Time 0.020166 -2022-12-06 11:46:06,829 - Epoch: [186][ 1020/ 1200] Overall Loss 0.127583 Objective Loss 0.127583 LR 0.000125 Time 0.020162 -2022-12-06 11:46:07,024 - Epoch: [186][ 1030/ 1200] Overall Loss 0.127857 Objective Loss 0.127857 LR 0.000125 Time 0.020155 -2022-12-06 11:46:07,222 - Epoch: [186][ 1040/ 1200] Overall Loss 0.127921 Objective Loss 0.127921 LR 0.000125 Time 0.020152 -2022-12-06 11:46:07,418 - Epoch: [186][ 1050/ 1200] Overall Loss 0.128041 Objective Loss 0.128041 LR 0.000125 Time 0.020146 -2022-12-06 11:46:07,616 - Epoch: [186][ 1060/ 1200] Overall Loss 0.128140 Objective Loss 0.128140 LR 0.000125 Time 0.020142 -2022-12-06 11:46:07,812 - Epoch: [186][ 1070/ 1200] Overall Loss 0.128204 Objective Loss 0.128204 LR 0.000125 Time 0.020136 -2022-12-06 11:46:08,010 - Epoch: [186][ 1080/ 1200] Overall Loss 0.128128 Objective Loss 0.128128 LR 0.000125 Time 0.020133 -2022-12-06 11:46:08,207 - Epoch: [186][ 1090/ 1200] Overall Loss 0.128207 Objective Loss 0.128207 LR 0.000125 Time 0.020128 -2022-12-06 11:46:08,406 - Epoch: [186][ 1100/ 1200] Overall Loss 0.128027 Objective Loss 0.128027 LR 0.000125 Time 0.020125 -2022-12-06 11:46:08,601 - Epoch: [186][ 1110/ 1200] Overall Loss 0.128039 Objective Loss 0.128039 LR 0.000125 Time 0.020120 -2022-12-06 11:46:08,800 - Epoch: [186][ 1120/ 1200] Overall Loss 0.128146 Objective Loss 0.128146 LR 0.000125 Time 0.020117 -2022-12-06 11:46:08,995 - Epoch: [186][ 1130/ 1200] Overall Loss 0.128378 Objective Loss 0.128378 LR 0.000125 Time 0.020111 -2022-12-06 11:46:09,195 - Epoch: [186][ 1140/ 1200] Overall Loss 0.128240 Objective Loss 0.128240 LR 0.000125 Time 0.020110 -2022-12-06 11:46:09,391 - Epoch: [186][ 1150/ 1200] Overall Loss 0.128292 Objective Loss 0.128292 LR 0.000125 Time 0.020105 -2022-12-06 11:46:09,589 - Epoch: [186][ 1160/ 1200] Overall Loss 0.128330 Objective Loss 0.128330 LR 0.000125 Time 0.020101 -2022-12-06 11:46:09,784 - Epoch: [186][ 1170/ 1200] Overall Loss 0.128253 Objective Loss 0.128253 LR 0.000125 Time 0.020096 -2022-12-06 11:46:09,984 - Epoch: [186][ 1180/ 1200] Overall Loss 0.128272 Objective Loss 0.128272 LR 0.000125 Time 0.020094 -2022-12-06 11:46:10,179 - Epoch: [186][ 1190/ 1200] Overall Loss 0.128394 Objective Loss 0.128394 LR 0.000125 Time 0.020090 -2022-12-06 11:46:10,409 - Epoch: [186][ 1200/ 1200] Overall Loss 0.128590 Objective Loss 0.128590 Top1 89.121339 Top5 98.535565 LR 0.000125 Time 0.020113 -2022-12-06 11:46:10,497 - --- validate (epoch=186)----------- -2022-12-06 11:46:10,497 - 34129 samples (256 per mini-batch) -2022-12-06 11:46:10,944 - Epoch: [186][ 10/ 134] Loss 0.240425 Top1 87.890625 Top5 98.710938 -2022-12-06 11:46:11,078 - Epoch: [186][ 20/ 134] Loss 0.224607 Top1 88.457031 Top5 98.730469 -2022-12-06 11:46:11,210 - Epoch: [186][ 30/ 134] Loss 0.220042 Top1 89.023438 Top5 98.802083 -2022-12-06 11:46:11,341 - Epoch: [186][ 40/ 134] Loss 0.225605 Top1 88.681641 Top5 98.681641 -2022-12-06 11:46:11,473 - Epoch: [186][ 50/ 134] Loss 0.226757 Top1 88.546875 Top5 98.695312 -2022-12-06 11:46:11,605 - Epoch: [186][ 60/ 134] Loss 0.220721 Top1 88.730469 Top5 98.704427 -2022-12-06 11:46:11,737 - Epoch: [186][ 70/ 134] Loss 0.219733 Top1 88.604911 Top5 98.666295 -2022-12-06 11:46:11,869 - Epoch: [186][ 80/ 134] Loss 0.224058 Top1 88.442383 Top5 98.666992 -2022-12-06 11:46:12,001 - Epoch: [186][ 90/ 134] Loss 0.225564 Top1 88.381076 Top5 98.641493 -2022-12-06 11:46:12,133 - Epoch: [186][ 100/ 134] Loss 0.227468 Top1 88.375000 Top5 98.632812 -2022-12-06 11:46:12,266 - Epoch: [186][ 110/ 134] Loss 0.225498 Top1 88.419744 Top5 98.671875 -2022-12-06 11:46:12,400 - Epoch: [186][ 120/ 134] Loss 0.224653 Top1 88.515625 Top5 98.684896 -2022-12-06 11:46:12,535 - Epoch: [186][ 130/ 134] Loss 0.226519 Top1 88.377404 Top5 98.677885 -2022-12-06 11:46:12,573 - Epoch: [186][ 134/ 134] Loss 0.229414 Top1 88.379384 Top5 98.666823 -2022-12-06 11:46:12,663 - ==> Top1: 88.379 Top5: 98.667 Loss: 0.229 - -2022-12-06 11:46:12,664 - ==> Confusion: -[[ 909 1 3 1 9 7 0 1 2 44 0 2 1 2 9 1 1 0 2 0 1] - [ 1 941 2 1 9 22 3 12 1 2 1 3 0 0 0 1 3 2 9 4 10] - [ 3 4 1014 15 5 3 14 9 0 4 5 2 1 1 3 1 1 1 2 4 11] - [ 3 2 13 957 0 2 2 0 1 0 10 0 4 2 8 0 0 1 9 0 6] - [ 6 4 1 0 971 3 1 2 1 5 2 2 0 2 7 3 5 3 0 0 2] - [ 0 7 1 2 5 999 1 14 3 2 1 9 2 11 2 1 1 1 0 2 5] - [ 2 2 6 1 0 2 1081 2 0 0 1 1 0 1 0 4 0 2 2 8 3] - [ 1 7 5 2 2 29 7 964 0 0 1 4 1 1 0 0 1 0 15 8 6] - [ 4 2 0 0 0 2 0 1 981 37 11 1 0 9 8 2 2 0 2 1 1] - [ 44 0 1 0 5 3 0 4 19 901 1 0 0 13 3 0 1 1 0 0 5] - [ 0 2 5 1 1 2 0 2 8 1 964 0 1 10 5 1 0 0 8 0 8] - [ 1 0 3 0 1 13 2 1 0 1 0 987 17 3 0 4 3 5 0 7 3] - [ 0 0 1 0 1 2 1 1 1 0 0 22 919 0 0 6 1 6 1 2 5] - [ 1 1 0 0 2 6 0 1 11 5 4 3 4 970 0 1 3 0 1 2 8] - [ 4 3 3 9 4 5 0 0 12 3 0 2 1 6 1064 0 1 1 6 1 5] - [ 0 0 0 2 4 0 3 1 2 0 0 7 7 3 0 990 6 13 0 4 1] - [ 1 4 1 2 2 0 1 0 0 0 0 3 3 2 1 5 1034 0 1 6 6] - [ 3 0 2 1 0 1 0 0 0 3 0 7 18 2 1 12 0 983 0 0 3] - [ 2 5 5 4 1 3 0 21 2 1 3 4 1 0 6 0 0 2 941 1 6] - [ 0 3 0 3 2 5 4 3 0 0 3 10 6 6 0 2 5 0 2 1021 5] - [ 87 179 159 83 109 167 69 126 77 78 134 89 282 240 123 87 134 65 142 229 10567]] - -2022-12-06 11:46:13,242 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:46:13,242 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:46:13,248 - - -2022-12-06 11:46:13,248 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:46:14,295 - Epoch: [187][ 10/ 1200] Overall Loss 0.120882 Objective Loss 0.120882 LR 0.000125 Time 0.104592 -2022-12-06 11:46:14,497 - Epoch: [187][ 20/ 1200] Overall Loss 0.118623 Objective Loss 0.118623 LR 0.000125 Time 0.062363 -2022-12-06 11:46:14,689 - Epoch: [187][ 30/ 1200] Overall Loss 0.126911 Objective Loss 0.126911 LR 0.000125 Time 0.047963 -2022-12-06 11:46:14,881 - Epoch: [187][ 40/ 1200] Overall Loss 0.125921 Objective Loss 0.125921 LR 0.000125 Time 0.040751 -2022-12-06 11:46:15,072 - Epoch: [187][ 50/ 1200] Overall Loss 0.125790 Objective Loss 0.125790 LR 0.000125 Time 0.036424 -2022-12-06 11:46:15,264 - Epoch: [187][ 60/ 1200] Overall Loss 0.124557 Objective Loss 0.124557 LR 0.000125 Time 0.033546 -2022-12-06 11:46:15,456 - Epoch: [187][ 70/ 1200] Overall Loss 0.123106 Objective Loss 0.123106 LR 0.000125 Time 0.031480 -2022-12-06 11:46:15,647 - Epoch: [187][ 80/ 1200] Overall Loss 0.121502 Objective Loss 0.121502 LR 0.000125 Time 0.029929 -2022-12-06 11:46:15,839 - Epoch: [187][ 90/ 1200] Overall Loss 0.119815 Objective Loss 0.119815 LR 0.000125 Time 0.028735 -2022-12-06 11:46:16,031 - Epoch: [187][ 100/ 1200] Overall Loss 0.119879 Objective Loss 0.119879 LR 0.000125 Time 0.027772 -2022-12-06 11:46:16,223 - Epoch: [187][ 110/ 1200] Overall Loss 0.120863 Objective Loss 0.120863 LR 0.000125 Time 0.026991 -2022-12-06 11:46:16,416 - Epoch: [187][ 120/ 1200] Overall Loss 0.120978 Objective Loss 0.120978 LR 0.000125 Time 0.026342 -2022-12-06 11:46:16,608 - Epoch: [187][ 130/ 1200] Overall Loss 0.120708 Objective Loss 0.120708 LR 0.000125 Time 0.025790 -2022-12-06 11:46:16,800 - Epoch: [187][ 140/ 1200] Overall Loss 0.122374 Objective Loss 0.122374 LR 0.000125 Time 0.025315 -2022-12-06 11:46:16,991 - Epoch: [187][ 150/ 1200] Overall Loss 0.123345 Objective Loss 0.123345 LR 0.000125 Time 0.024898 -2022-12-06 11:46:17,183 - Epoch: [187][ 160/ 1200] Overall Loss 0.124268 Objective Loss 0.124268 LR 0.000125 Time 0.024538 -2022-12-06 11:46:17,375 - Epoch: [187][ 170/ 1200] Overall Loss 0.124231 Objective Loss 0.124231 LR 0.000125 Time 0.024221 -2022-12-06 11:46:17,568 - Epoch: [187][ 180/ 1200] Overall Loss 0.124060 Objective Loss 0.124060 LR 0.000125 Time 0.023940 -2022-12-06 11:46:17,759 - Epoch: [187][ 190/ 1200] Overall Loss 0.124382 Objective Loss 0.124382 LR 0.000125 Time 0.023687 -2022-12-06 11:46:17,951 - Epoch: [187][ 200/ 1200] Overall Loss 0.124520 Objective Loss 0.124520 LR 0.000125 Time 0.023458 -2022-12-06 11:46:18,142 - Epoch: [187][ 210/ 1200] Overall Loss 0.125188 Objective Loss 0.125188 LR 0.000125 Time 0.023249 -2022-12-06 11:46:18,334 - Epoch: [187][ 220/ 1200] Overall Loss 0.124980 Objective Loss 0.124980 LR 0.000125 Time 0.023061 -2022-12-06 11:46:18,526 - Epoch: [187][ 230/ 1200] Overall Loss 0.125239 Objective Loss 0.125239 LR 0.000125 Time 0.022890 -2022-12-06 11:46:18,718 - Epoch: [187][ 240/ 1200] Overall Loss 0.126328 Objective Loss 0.126328 LR 0.000125 Time 0.022734 -2022-12-06 11:46:18,909 - Epoch: [187][ 250/ 1200] Overall Loss 0.126860 Objective Loss 0.126860 LR 0.000125 Time 0.022589 -2022-12-06 11:46:19,101 - Epoch: [187][ 260/ 1200] Overall Loss 0.126997 Objective Loss 0.126997 LR 0.000125 Time 0.022456 -2022-12-06 11:46:19,292 - Epoch: [187][ 270/ 1200] Overall Loss 0.127308 Objective Loss 0.127308 LR 0.000125 Time 0.022330 -2022-12-06 11:46:19,484 - Epoch: [187][ 280/ 1200] Overall Loss 0.127339 Objective Loss 0.127339 LR 0.000125 Time 0.022216 -2022-12-06 11:46:19,675 - Epoch: [187][ 290/ 1200] Overall Loss 0.127761 Objective Loss 0.127761 LR 0.000125 Time 0.022107 -2022-12-06 11:46:19,867 - Epoch: [187][ 300/ 1200] Overall Loss 0.127675 Objective Loss 0.127675 LR 0.000125 Time 0.022008 -2022-12-06 11:46:20,059 - Epoch: [187][ 310/ 1200] Overall Loss 0.127384 Objective Loss 0.127384 LR 0.000125 Time 0.021916 -2022-12-06 11:46:20,251 - Epoch: [187][ 320/ 1200] Overall Loss 0.128206 Objective Loss 0.128206 LR 0.000125 Time 0.021830 -2022-12-06 11:46:20,443 - Epoch: [187][ 330/ 1200] Overall Loss 0.128534 Objective Loss 0.128534 LR 0.000125 Time 0.021746 -2022-12-06 11:46:20,636 - Epoch: [187][ 340/ 1200] Overall Loss 0.129279 Objective Loss 0.129279 LR 0.000125 Time 0.021673 -2022-12-06 11:46:20,830 - Epoch: [187][ 350/ 1200] Overall Loss 0.129095 Objective Loss 0.129095 LR 0.000125 Time 0.021607 -2022-12-06 11:46:21,024 - Epoch: [187][ 360/ 1200] Overall Loss 0.129519 Objective Loss 0.129519 LR 0.000125 Time 0.021545 -2022-12-06 11:46:21,218 - Epoch: [187][ 370/ 1200] Overall Loss 0.129395 Objective Loss 0.129395 LR 0.000125 Time 0.021486 -2022-12-06 11:46:21,413 - Epoch: [187][ 380/ 1200] Overall Loss 0.129004 Objective Loss 0.129004 LR 0.000125 Time 0.021430 -2022-12-06 11:46:21,607 - Epoch: [187][ 390/ 1200] Overall Loss 0.129016 Objective Loss 0.129016 LR 0.000125 Time 0.021378 -2022-12-06 11:46:21,802 - Epoch: [187][ 400/ 1200] Overall Loss 0.128983 Objective Loss 0.128983 LR 0.000125 Time 0.021328 -2022-12-06 11:46:21,996 - Epoch: [187][ 410/ 1200] Overall Loss 0.129186 Objective Loss 0.129186 LR 0.000125 Time 0.021281 -2022-12-06 11:46:22,190 - Epoch: [187][ 420/ 1200] Overall Loss 0.128803 Objective Loss 0.128803 LR 0.000125 Time 0.021235 -2022-12-06 11:46:22,384 - Epoch: [187][ 430/ 1200] Overall Loss 0.128692 Objective Loss 0.128692 LR 0.000125 Time 0.021191 -2022-12-06 11:46:22,578 - Epoch: [187][ 440/ 1200] Overall Loss 0.128781 Objective Loss 0.128781 LR 0.000125 Time 0.021150 -2022-12-06 11:46:22,773 - Epoch: [187][ 450/ 1200] Overall Loss 0.129095 Objective Loss 0.129095 LR 0.000125 Time 0.021111 -2022-12-06 11:46:22,967 - Epoch: [187][ 460/ 1200] Overall Loss 0.129123 Objective Loss 0.129123 LR 0.000125 Time 0.021072 -2022-12-06 11:46:23,161 - Epoch: [187][ 470/ 1200] Overall Loss 0.129339 Objective Loss 0.129339 LR 0.000125 Time 0.021037 -2022-12-06 11:46:23,355 - Epoch: [187][ 480/ 1200] Overall Loss 0.129522 Objective Loss 0.129522 LR 0.000125 Time 0.021001 -2022-12-06 11:46:23,550 - Epoch: [187][ 490/ 1200] Overall Loss 0.129399 Objective Loss 0.129399 LR 0.000125 Time 0.020969 -2022-12-06 11:46:23,744 - Epoch: [187][ 500/ 1200] Overall Loss 0.129409 Objective Loss 0.129409 LR 0.000125 Time 0.020937 -2022-12-06 11:46:23,939 - Epoch: [187][ 510/ 1200] Overall Loss 0.129266 Objective Loss 0.129266 LR 0.000125 Time 0.020907 -2022-12-06 11:46:24,132 - Epoch: [187][ 520/ 1200] Overall Loss 0.129130 Objective Loss 0.129130 LR 0.000125 Time 0.020876 -2022-12-06 11:46:24,327 - Epoch: [187][ 530/ 1200] Overall Loss 0.129004 Objective Loss 0.129004 LR 0.000125 Time 0.020849 -2022-12-06 11:46:24,522 - Epoch: [187][ 540/ 1200] Overall Loss 0.129437 Objective Loss 0.129437 LR 0.000125 Time 0.020822 -2022-12-06 11:46:24,716 - Epoch: [187][ 550/ 1200] Overall Loss 0.129164 Objective Loss 0.129164 LR 0.000125 Time 0.020796 -2022-12-06 11:46:24,910 - Epoch: [187][ 560/ 1200] Overall Loss 0.129011 Objective Loss 0.129011 LR 0.000125 Time 0.020770 -2022-12-06 11:46:25,105 - Epoch: [187][ 570/ 1200] Overall Loss 0.128686 Objective Loss 0.128686 LR 0.000125 Time 0.020746 -2022-12-06 11:46:25,299 - Epoch: [187][ 580/ 1200] Overall Loss 0.128421 Objective Loss 0.128421 LR 0.000125 Time 0.020721 -2022-12-06 11:46:25,492 - Epoch: [187][ 590/ 1200] Overall Loss 0.128641 Objective Loss 0.128641 LR 0.000125 Time 0.020698 -2022-12-06 11:46:25,687 - Epoch: [187][ 600/ 1200] Overall Loss 0.128569 Objective Loss 0.128569 LR 0.000125 Time 0.020676 -2022-12-06 11:46:25,882 - Epoch: [187][ 610/ 1200] Overall Loss 0.128624 Objective Loss 0.128624 LR 0.000125 Time 0.020656 -2022-12-06 11:46:26,076 - Epoch: [187][ 620/ 1200] Overall Loss 0.128764 Objective Loss 0.128764 LR 0.000125 Time 0.020635 -2022-12-06 11:46:26,270 - Epoch: [187][ 630/ 1200] Overall Loss 0.128589 Objective Loss 0.128589 LR 0.000125 Time 0.020615 -2022-12-06 11:46:26,464 - Epoch: [187][ 640/ 1200] Overall Loss 0.128716 Objective Loss 0.128716 LR 0.000125 Time 0.020595 -2022-12-06 11:46:26,659 - Epoch: [187][ 650/ 1200] Overall Loss 0.128573 Objective Loss 0.128573 LR 0.000125 Time 0.020577 -2022-12-06 11:46:26,854 - Epoch: [187][ 660/ 1200] Overall Loss 0.128463 Objective Loss 0.128463 LR 0.000125 Time 0.020559 -2022-12-06 11:46:27,048 - Epoch: [187][ 670/ 1200] Overall Loss 0.128098 Objective Loss 0.128098 LR 0.000125 Time 0.020542 -2022-12-06 11:46:27,242 - Epoch: [187][ 680/ 1200] Overall Loss 0.127871 Objective Loss 0.127871 LR 0.000125 Time 0.020525 -2022-12-06 11:46:27,437 - Epoch: [187][ 690/ 1200] Overall Loss 0.127971 Objective Loss 0.127971 LR 0.000125 Time 0.020508 -2022-12-06 11:46:27,631 - Epoch: [187][ 700/ 1200] Overall Loss 0.127867 Objective Loss 0.127867 LR 0.000125 Time 0.020492 -2022-12-06 11:46:27,826 - Epoch: [187][ 710/ 1200] Overall Loss 0.127748 Objective Loss 0.127748 LR 0.000125 Time 0.020477 -2022-12-06 11:46:28,018 - Epoch: [187][ 720/ 1200] Overall Loss 0.127765 Objective Loss 0.127765 LR 0.000125 Time 0.020459 -2022-12-06 11:46:28,211 - Epoch: [187][ 730/ 1200] Overall Loss 0.127590 Objective Loss 0.127590 LR 0.000125 Time 0.020441 -2022-12-06 11:46:28,402 - Epoch: [187][ 740/ 1200] Overall Loss 0.127565 Objective Loss 0.127565 LR 0.000125 Time 0.020423 -2022-12-06 11:46:28,594 - Epoch: [187][ 750/ 1200] Overall Loss 0.127612 Objective Loss 0.127612 LR 0.000125 Time 0.020406 -2022-12-06 11:46:28,786 - Epoch: [187][ 760/ 1200] Overall Loss 0.127493 Objective Loss 0.127493 LR 0.000125 Time 0.020389 -2022-12-06 11:46:28,977 - Epoch: [187][ 770/ 1200] Overall Loss 0.127388 Objective Loss 0.127388 LR 0.000125 Time 0.020372 -2022-12-06 11:46:29,169 - Epoch: [187][ 780/ 1200] Overall Loss 0.127206 Objective Loss 0.127206 LR 0.000125 Time 0.020356 -2022-12-06 11:46:29,360 - Epoch: [187][ 790/ 1200] Overall Loss 0.127122 Objective Loss 0.127122 LR 0.000125 Time 0.020340 -2022-12-06 11:46:29,552 - Epoch: [187][ 800/ 1200] Overall Loss 0.127140 Objective Loss 0.127140 LR 0.000125 Time 0.020325 -2022-12-06 11:46:29,743 - Epoch: [187][ 810/ 1200] Overall Loss 0.127090 Objective Loss 0.127090 LR 0.000125 Time 0.020309 -2022-12-06 11:46:29,936 - Epoch: [187][ 820/ 1200] Overall Loss 0.127221 Objective Loss 0.127221 LR 0.000125 Time 0.020295 -2022-12-06 11:46:30,128 - Epoch: [187][ 830/ 1200] Overall Loss 0.127090 Objective Loss 0.127090 LR 0.000125 Time 0.020282 -2022-12-06 11:46:30,319 - Epoch: [187][ 840/ 1200] Overall Loss 0.127193 Objective Loss 0.127193 LR 0.000125 Time 0.020268 -2022-12-06 11:46:30,512 - Epoch: [187][ 850/ 1200] Overall Loss 0.127211 Objective Loss 0.127211 LR 0.000125 Time 0.020255 -2022-12-06 11:46:30,703 - Epoch: [187][ 860/ 1200] Overall Loss 0.127274 Objective Loss 0.127274 LR 0.000125 Time 0.020241 -2022-12-06 11:46:30,895 - Epoch: [187][ 870/ 1200] Overall Loss 0.127530 Objective Loss 0.127530 LR 0.000125 Time 0.020228 -2022-12-06 11:46:31,086 - Epoch: [187][ 880/ 1200] Overall Loss 0.127706 Objective Loss 0.127706 LR 0.000125 Time 0.020215 -2022-12-06 11:46:31,277 - Epoch: [187][ 890/ 1200] Overall Loss 0.127893 Objective Loss 0.127893 LR 0.000125 Time 0.020202 -2022-12-06 11:46:31,468 - Epoch: [187][ 900/ 1200] Overall Loss 0.127855 Objective Loss 0.127855 LR 0.000125 Time 0.020189 -2022-12-06 11:46:31,660 - Epoch: [187][ 910/ 1200] Overall Loss 0.128043 Objective Loss 0.128043 LR 0.000125 Time 0.020178 -2022-12-06 11:46:31,852 - Epoch: [187][ 920/ 1200] Overall Loss 0.127881 Objective Loss 0.127881 LR 0.000125 Time 0.020167 -2022-12-06 11:46:32,044 - Epoch: [187][ 930/ 1200] Overall Loss 0.127650 Objective Loss 0.127650 LR 0.000125 Time 0.020155 -2022-12-06 11:46:32,235 - Epoch: [187][ 940/ 1200] Overall Loss 0.127574 Objective Loss 0.127574 LR 0.000125 Time 0.020144 -2022-12-06 11:46:32,427 - Epoch: [187][ 950/ 1200] Overall Loss 0.127424 Objective Loss 0.127424 LR 0.000125 Time 0.020133 -2022-12-06 11:46:32,618 - Epoch: [187][ 960/ 1200] Overall Loss 0.127294 Objective Loss 0.127294 LR 0.000125 Time 0.020122 -2022-12-06 11:46:32,810 - Epoch: [187][ 970/ 1200] Overall Loss 0.127041 Objective Loss 0.127041 LR 0.000125 Time 0.020112 -2022-12-06 11:46:33,002 - Epoch: [187][ 980/ 1200] Overall Loss 0.126976 Objective Loss 0.126976 LR 0.000125 Time 0.020102 -2022-12-06 11:46:33,193 - Epoch: [187][ 990/ 1200] Overall Loss 0.126964 Objective Loss 0.126964 LR 0.000125 Time 0.020091 -2022-12-06 11:46:33,384 - Epoch: [187][ 1000/ 1200] Overall Loss 0.127012 Objective Loss 0.127012 LR 0.000125 Time 0.020081 -2022-12-06 11:46:33,576 - Epoch: [187][ 1010/ 1200] Overall Loss 0.127061 Objective Loss 0.127061 LR 0.000125 Time 0.020071 -2022-12-06 11:46:33,767 - Epoch: [187][ 1020/ 1200] Overall Loss 0.126939 Objective Loss 0.126939 LR 0.000125 Time 0.020062 -2022-12-06 11:46:33,959 - Epoch: [187][ 1030/ 1200] Overall Loss 0.127071 Objective Loss 0.127071 LR 0.000125 Time 0.020052 -2022-12-06 11:46:34,152 - Epoch: [187][ 1040/ 1200] Overall Loss 0.127084 Objective Loss 0.127084 LR 0.000125 Time 0.020044 -2022-12-06 11:46:34,344 - Epoch: [187][ 1050/ 1200] Overall Loss 0.127115 Objective Loss 0.127115 LR 0.000125 Time 0.020036 -2022-12-06 11:46:34,536 - Epoch: [187][ 1060/ 1200] Overall Loss 0.127244 Objective Loss 0.127244 LR 0.000125 Time 0.020028 -2022-12-06 11:46:34,728 - Epoch: [187][ 1070/ 1200] Overall Loss 0.127279 Objective Loss 0.127279 LR 0.000125 Time 0.020019 -2022-12-06 11:46:34,919 - Epoch: [187][ 1080/ 1200] Overall Loss 0.127346 Objective Loss 0.127346 LR 0.000125 Time 0.020011 -2022-12-06 11:46:35,111 - Epoch: [187][ 1090/ 1200] Overall Loss 0.127246 Objective Loss 0.127246 LR 0.000125 Time 0.020002 -2022-12-06 11:46:35,302 - Epoch: [187][ 1100/ 1200] Overall Loss 0.127103 Objective Loss 0.127103 LR 0.000125 Time 0.019994 -2022-12-06 11:46:35,494 - Epoch: [187][ 1110/ 1200] Overall Loss 0.127175 Objective Loss 0.127175 LR 0.000125 Time 0.019986 -2022-12-06 11:46:35,685 - Epoch: [187][ 1120/ 1200] Overall Loss 0.127107 Objective Loss 0.127107 LR 0.000125 Time 0.019978 -2022-12-06 11:46:35,876 - Epoch: [187][ 1130/ 1200] Overall Loss 0.127188 Objective Loss 0.127188 LR 0.000125 Time 0.019970 -2022-12-06 11:46:36,068 - Epoch: [187][ 1140/ 1200] Overall Loss 0.127239 Objective Loss 0.127239 LR 0.000125 Time 0.019962 -2022-12-06 11:46:36,260 - Epoch: [187][ 1150/ 1200] Overall Loss 0.127166 Objective Loss 0.127166 LR 0.000125 Time 0.019955 -2022-12-06 11:46:36,452 - Epoch: [187][ 1160/ 1200] Overall Loss 0.127162 Objective Loss 0.127162 LR 0.000125 Time 0.019948 -2022-12-06 11:46:36,644 - Epoch: [187][ 1170/ 1200] Overall Loss 0.127114 Objective Loss 0.127114 LR 0.000125 Time 0.019941 -2022-12-06 11:46:36,836 - Epoch: [187][ 1180/ 1200] Overall Loss 0.126959 Objective Loss 0.126959 LR 0.000125 Time 0.019934 -2022-12-06 11:46:37,028 - Epoch: [187][ 1190/ 1200] Overall Loss 0.127061 Objective Loss 0.127061 LR 0.000125 Time 0.019928 -2022-12-06 11:46:37,249 - Epoch: [187][ 1200/ 1200] Overall Loss 0.127265 Objective Loss 0.127265 Top1 91.631799 Top5 98.953975 LR 0.000125 Time 0.019946 -2022-12-06 11:46:37,338 - --- validate (epoch=187)----------- -2022-12-06 11:46:37,338 - 34129 samples (256 per mini-batch) -2022-12-06 11:46:37,779 - Epoch: [187][ 10/ 134] Loss 0.224068 Top1 88.476562 Top5 98.789062 -2022-12-06 11:46:37,911 - Epoch: [187][ 20/ 134] Loss 0.220861 Top1 88.261719 Top5 98.632812 -2022-12-06 11:46:38,044 - Epoch: [187][ 30/ 134] Loss 0.223232 Top1 88.203125 Top5 98.736979 -2022-12-06 11:46:38,175 - Epoch: [187][ 40/ 134] Loss 0.223855 Top1 88.095703 Top5 98.632812 -2022-12-06 11:46:38,305 - Epoch: [187][ 50/ 134] Loss 0.222266 Top1 88.109375 Top5 98.640625 -2022-12-06 11:46:38,436 - Epoch: [187][ 60/ 134] Loss 0.219307 Top1 88.359375 Top5 98.678385 -2022-12-06 11:46:38,566 - Epoch: [187][ 70/ 134] Loss 0.220155 Top1 88.404018 Top5 98.716518 -2022-12-06 11:46:38,699 - Epoch: [187][ 80/ 134] Loss 0.222860 Top1 88.320312 Top5 98.696289 -2022-12-06 11:46:38,829 - Epoch: [187][ 90/ 134] Loss 0.225865 Top1 88.320312 Top5 98.645833 -2022-12-06 11:46:38,958 - Epoch: [187][ 100/ 134] Loss 0.223036 Top1 88.363281 Top5 98.660156 -2022-12-06 11:46:39,090 - Epoch: [187][ 110/ 134] Loss 0.223645 Top1 88.330966 Top5 98.647017 -2022-12-06 11:46:39,221 - Epoch: [187][ 120/ 134] Loss 0.226729 Top1 88.271484 Top5 98.655599 -2022-12-06 11:46:39,353 - Epoch: [187][ 130/ 134] Loss 0.226108 Top1 88.299279 Top5 98.638822 -2022-12-06 11:46:39,392 - Epoch: [187][ 134/ 134] Loss 0.225258 Top1 88.297342 Top5 98.643382 -2022-12-06 11:46:39,480 - ==> Top1: 88.297 Top5: 98.643 Loss: 0.225 - -2022-12-06 11:46:39,480 - ==> Confusion: -[[ 912 1 1 0 3 6 0 1 8 48 0 0 1 4 6 1 1 0 2 0 1] - [ 1 953 1 2 8 18 2 12 3 1 3 4 0 1 0 0 1 1 6 2 8] - [ 3 3 1010 13 3 2 18 8 2 4 6 4 1 2 1 2 1 1 4 4 11] - [ 5 0 15 950 1 2 1 1 1 1 10 0 4 0 9 0 2 1 10 1 6] - [ 6 6 1 0 971 2 1 2 1 6 2 1 0 2 3 3 6 3 1 0 3] - [ 0 7 1 3 7 988 3 16 3 1 1 13 2 14 1 1 0 1 0 4 3] - [ 1 1 5 1 1 2 1084 3 0 0 0 2 0 2 0 3 0 1 1 7 4] - [ 1 6 4 2 1 18 11 968 0 1 0 7 0 2 0 0 1 0 16 9 7] - [ 4 1 0 0 0 2 0 2 993 35 5 2 1 8 7 0 1 0 1 1 1] - [ 40 1 2 1 6 1 0 3 22 905 1 0 0 11 2 1 0 0 0 0 5] - [ 0 2 3 2 0 0 2 1 7 2 974 0 0 9 4 1 0 0 5 1 6] - [ 1 2 2 0 0 7 4 1 2 0 0 994 18 2 0 3 3 5 0 5 2] - [ 0 0 1 1 0 2 0 0 0 1 0 26 908 1 0 9 1 7 2 3 7] - [ 1 1 0 0 0 4 0 2 14 9 1 7 3 972 0 1 2 0 0 1 5] - [ 8 3 3 8 4 3 0 1 13 1 1 2 1 6 1065 0 0 1 5 1 4] - [ 0 0 0 3 2 0 3 1 1 1 1 8 4 2 0 995 3 13 0 4 2] - [ 0 2 1 2 1 0 1 0 1 0 0 4 2 1 1 11 1033 1 1 6 4] - [ 3 0 2 0 1 1 0 0 0 2 0 9 12 2 2 10 0 988 0 1 3] - [ 2 2 4 6 1 2 0 22 2 1 5 3 3 0 6 1 0 2 940 1 5] - [ 2 4 0 2 1 4 4 7 0 0 2 13 4 9 0 3 1 2 1 1016 5] - [ 113 177 139 72 98 152 75 127 99 93 155 94 290 268 122 96 134 75 134 201 10512]] - -2022-12-06 11:46:40,045 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:46:40,045 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:46:40,051 - - -2022-12-06 11:46:40,051 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:46:40,981 - Epoch: [188][ 10/ 1200] Overall Loss 0.137403 Objective Loss 0.137403 LR 0.000125 Time 0.092913 -2022-12-06 11:46:41,182 - Epoch: [188][ 20/ 1200] Overall Loss 0.129067 Objective Loss 0.129067 LR 0.000125 Time 0.056476 -2022-12-06 11:46:41,373 - Epoch: [188][ 30/ 1200] Overall Loss 0.127082 Objective Loss 0.127082 LR 0.000125 Time 0.044007 -2022-12-06 11:46:41,565 - Epoch: [188][ 40/ 1200] Overall Loss 0.125035 Objective Loss 0.125035 LR 0.000125 Time 0.037788 -2022-12-06 11:46:41,756 - Epoch: [188][ 50/ 1200] Overall Loss 0.130550 Objective Loss 0.130550 LR 0.000125 Time 0.034038 -2022-12-06 11:46:41,947 - Epoch: [188][ 60/ 1200] Overall Loss 0.129252 Objective Loss 0.129252 LR 0.000125 Time 0.031542 -2022-12-06 11:46:42,139 - Epoch: [188][ 70/ 1200] Overall Loss 0.131330 Objective Loss 0.131330 LR 0.000125 Time 0.029768 -2022-12-06 11:46:42,331 - Epoch: [188][ 80/ 1200] Overall Loss 0.131616 Objective Loss 0.131616 LR 0.000125 Time 0.028445 -2022-12-06 11:46:42,523 - Epoch: [188][ 90/ 1200] Overall Loss 0.130460 Objective Loss 0.130460 LR 0.000125 Time 0.027402 -2022-12-06 11:46:42,714 - Epoch: [188][ 100/ 1200] Overall Loss 0.131174 Objective Loss 0.131174 LR 0.000125 Time 0.026569 -2022-12-06 11:46:42,904 - Epoch: [188][ 110/ 1200] Overall Loss 0.130139 Objective Loss 0.130139 LR 0.000125 Time 0.025882 -2022-12-06 11:46:43,096 - Epoch: [188][ 120/ 1200] Overall Loss 0.129682 Objective Loss 0.129682 LR 0.000125 Time 0.025316 -2022-12-06 11:46:43,287 - Epoch: [188][ 130/ 1200] Overall Loss 0.128668 Objective Loss 0.128668 LR 0.000125 Time 0.024836 -2022-12-06 11:46:43,479 - Epoch: [188][ 140/ 1200] Overall Loss 0.128984 Objective Loss 0.128984 LR 0.000125 Time 0.024427 -2022-12-06 11:46:43,670 - Epoch: [188][ 150/ 1200] Overall Loss 0.128432 Objective Loss 0.128432 LR 0.000125 Time 0.024070 -2022-12-06 11:46:43,862 - Epoch: [188][ 160/ 1200] Overall Loss 0.129245 Objective Loss 0.129245 LR 0.000125 Time 0.023759 -2022-12-06 11:46:44,053 - Epoch: [188][ 170/ 1200] Overall Loss 0.128033 Objective Loss 0.128033 LR 0.000125 Time 0.023483 -2022-12-06 11:46:44,244 - Epoch: [188][ 180/ 1200] Overall Loss 0.127007 Objective Loss 0.127007 LR 0.000125 Time 0.023239 -2022-12-06 11:46:44,436 - Epoch: [188][ 190/ 1200] Overall Loss 0.127249 Objective Loss 0.127249 LR 0.000125 Time 0.023020 -2022-12-06 11:46:44,627 - Epoch: [188][ 200/ 1200] Overall Loss 0.126886 Objective Loss 0.126886 LR 0.000125 Time 0.022822 -2022-12-06 11:46:44,818 - Epoch: [188][ 210/ 1200] Overall Loss 0.125930 Objective Loss 0.125930 LR 0.000125 Time 0.022642 -2022-12-06 11:46:45,009 - Epoch: [188][ 220/ 1200] Overall Loss 0.125536 Objective Loss 0.125536 LR 0.000125 Time 0.022481 -2022-12-06 11:46:45,200 - Epoch: [188][ 230/ 1200] Overall Loss 0.126137 Objective Loss 0.126137 LR 0.000125 Time 0.022331 -2022-12-06 11:46:45,391 - Epoch: [188][ 240/ 1200] Overall Loss 0.125644 Objective Loss 0.125644 LR 0.000125 Time 0.022194 -2022-12-06 11:46:45,582 - Epoch: [188][ 250/ 1200] Overall Loss 0.125972 Objective Loss 0.125972 LR 0.000125 Time 0.022068 -2022-12-06 11:46:45,773 - Epoch: [188][ 260/ 1200] Overall Loss 0.126205 Objective Loss 0.126205 LR 0.000125 Time 0.021950 -2022-12-06 11:46:45,963 - Epoch: [188][ 270/ 1200] Overall Loss 0.125673 Objective Loss 0.125673 LR 0.000125 Time 0.021840 -2022-12-06 11:46:46,154 - Epoch: [188][ 280/ 1200] Overall Loss 0.126017 Objective Loss 0.126017 LR 0.000125 Time 0.021740 -2022-12-06 11:46:46,345 - Epoch: [188][ 290/ 1200] Overall Loss 0.125900 Objective Loss 0.125900 LR 0.000125 Time 0.021648 -2022-12-06 11:46:46,537 - Epoch: [188][ 300/ 1200] Overall Loss 0.125920 Objective Loss 0.125920 LR 0.000125 Time 0.021564 -2022-12-06 11:46:46,728 - Epoch: [188][ 310/ 1200] Overall Loss 0.125695 Objective Loss 0.125695 LR 0.000125 Time 0.021482 -2022-12-06 11:46:46,920 - Epoch: [188][ 320/ 1200] Overall Loss 0.125791 Objective Loss 0.125791 LR 0.000125 Time 0.021408 -2022-12-06 11:46:47,111 - Epoch: [188][ 330/ 1200] Overall Loss 0.125637 Objective Loss 0.125637 LR 0.000125 Time 0.021336 -2022-12-06 11:46:47,302 - Epoch: [188][ 340/ 1200] Overall Loss 0.126157 Objective Loss 0.126157 LR 0.000125 Time 0.021269 -2022-12-06 11:46:47,493 - Epoch: [188][ 350/ 1200] Overall Loss 0.126501 Objective Loss 0.126501 LR 0.000125 Time 0.021205 -2022-12-06 11:46:47,684 - Epoch: [188][ 360/ 1200] Overall Loss 0.126609 Objective Loss 0.126609 LR 0.000125 Time 0.021147 -2022-12-06 11:46:47,875 - Epoch: [188][ 370/ 1200] Overall Loss 0.126828 Objective Loss 0.126828 LR 0.000125 Time 0.021090 -2022-12-06 11:46:48,067 - Epoch: [188][ 380/ 1200] Overall Loss 0.126806 Objective Loss 0.126806 LR 0.000125 Time 0.021037 -2022-12-06 11:46:48,258 - Epoch: [188][ 390/ 1200] Overall Loss 0.126931 Objective Loss 0.126931 LR 0.000125 Time 0.020987 -2022-12-06 11:46:48,450 - Epoch: [188][ 400/ 1200] Overall Loss 0.126962 Objective Loss 0.126962 LR 0.000125 Time 0.020941 -2022-12-06 11:46:48,641 - Epoch: [188][ 410/ 1200] Overall Loss 0.127131 Objective Loss 0.127131 LR 0.000125 Time 0.020896 -2022-12-06 11:46:48,832 - Epoch: [188][ 420/ 1200] Overall Loss 0.127098 Objective Loss 0.127098 LR 0.000125 Time 0.020852 -2022-12-06 11:46:49,023 - Epoch: [188][ 430/ 1200] Overall Loss 0.127430 Objective Loss 0.127430 LR 0.000125 Time 0.020808 -2022-12-06 11:46:49,214 - Epoch: [188][ 440/ 1200] Overall Loss 0.127694 Objective Loss 0.127694 LR 0.000125 Time 0.020769 -2022-12-06 11:46:49,405 - Epoch: [188][ 450/ 1200] Overall Loss 0.127678 Objective Loss 0.127678 LR 0.000125 Time 0.020731 -2022-12-06 11:46:49,597 - Epoch: [188][ 460/ 1200] Overall Loss 0.127651 Objective Loss 0.127651 LR 0.000125 Time 0.020696 -2022-12-06 11:46:49,788 - Epoch: [188][ 470/ 1200] Overall Loss 0.127923 Objective Loss 0.127923 LR 0.000125 Time 0.020660 -2022-12-06 11:46:49,979 - Epoch: [188][ 480/ 1200] Overall Loss 0.127835 Objective Loss 0.127835 LR 0.000125 Time 0.020626 -2022-12-06 11:46:50,169 - Epoch: [188][ 490/ 1200] Overall Loss 0.127984 Objective Loss 0.127984 LR 0.000125 Time 0.020594 -2022-12-06 11:46:50,361 - Epoch: [188][ 500/ 1200] Overall Loss 0.127772 Objective Loss 0.127772 LR 0.000125 Time 0.020564 -2022-12-06 11:46:50,552 - Epoch: [188][ 510/ 1200] Overall Loss 0.127411 Objective Loss 0.127411 LR 0.000125 Time 0.020534 -2022-12-06 11:46:50,743 - Epoch: [188][ 520/ 1200] Overall Loss 0.127089 Objective Loss 0.127089 LR 0.000125 Time 0.020506 -2022-12-06 11:46:50,936 - Epoch: [188][ 530/ 1200] Overall Loss 0.127534 Objective Loss 0.127534 LR 0.000125 Time 0.020481 -2022-12-06 11:46:51,127 - Epoch: [188][ 540/ 1200] Overall Loss 0.127658 Objective Loss 0.127658 LR 0.000125 Time 0.020455 -2022-12-06 11:46:51,319 - Epoch: [188][ 550/ 1200] Overall Loss 0.127372 Objective Loss 0.127372 LR 0.000125 Time 0.020430 -2022-12-06 11:46:51,510 - Epoch: [188][ 560/ 1200] Overall Loss 0.127494 Objective Loss 0.127494 LR 0.000125 Time 0.020406 -2022-12-06 11:46:51,701 - Epoch: [188][ 570/ 1200] Overall Loss 0.127468 Objective Loss 0.127468 LR 0.000125 Time 0.020382 -2022-12-06 11:46:51,893 - Epoch: [188][ 580/ 1200] Overall Loss 0.127450 Objective Loss 0.127450 LR 0.000125 Time 0.020361 -2022-12-06 11:46:52,084 - Epoch: [188][ 590/ 1200] Overall Loss 0.127614 Objective Loss 0.127614 LR 0.000125 Time 0.020339 -2022-12-06 11:46:52,275 - Epoch: [188][ 600/ 1200] Overall Loss 0.127863 Objective Loss 0.127863 LR 0.000125 Time 0.020318 -2022-12-06 11:46:52,466 - Epoch: [188][ 610/ 1200] Overall Loss 0.128005 Objective Loss 0.128005 LR 0.000125 Time 0.020296 -2022-12-06 11:46:52,657 - Epoch: [188][ 620/ 1200] Overall Loss 0.127941 Objective Loss 0.127941 LR 0.000125 Time 0.020276 -2022-12-06 11:46:52,848 - Epoch: [188][ 630/ 1200] Overall Loss 0.128069 Objective Loss 0.128069 LR 0.000125 Time 0.020256 -2022-12-06 11:46:53,039 - Epoch: [188][ 640/ 1200] Overall Loss 0.128264 Objective Loss 0.128264 LR 0.000125 Time 0.020238 -2022-12-06 11:46:53,232 - Epoch: [188][ 650/ 1200] Overall Loss 0.128256 Objective Loss 0.128256 LR 0.000125 Time 0.020222 -2022-12-06 11:46:53,423 - Epoch: [188][ 660/ 1200] Overall Loss 0.128154 Objective Loss 0.128154 LR 0.000125 Time 0.020205 -2022-12-06 11:46:53,614 - Epoch: [188][ 670/ 1200] Overall Loss 0.128241 Objective Loss 0.128241 LR 0.000125 Time 0.020187 -2022-12-06 11:46:53,806 - Epoch: [188][ 680/ 1200] Overall Loss 0.128030 Objective Loss 0.128030 LR 0.000125 Time 0.020172 -2022-12-06 11:46:53,997 - Epoch: [188][ 690/ 1200] Overall Loss 0.128181 Objective Loss 0.128181 LR 0.000125 Time 0.020156 -2022-12-06 11:46:54,189 - Epoch: [188][ 700/ 1200] Overall Loss 0.128197 Objective Loss 0.128197 LR 0.000125 Time 0.020141 -2022-12-06 11:46:54,380 - Epoch: [188][ 710/ 1200] Overall Loss 0.128220 Objective Loss 0.128220 LR 0.000125 Time 0.020126 -2022-12-06 11:46:54,571 - Epoch: [188][ 720/ 1200] Overall Loss 0.128137 Objective Loss 0.128137 LR 0.000125 Time 0.020111 -2022-12-06 11:46:54,763 - Epoch: [188][ 730/ 1200] Overall Loss 0.127886 Objective Loss 0.127886 LR 0.000125 Time 0.020097 -2022-12-06 11:46:54,955 - Epoch: [188][ 740/ 1200] Overall Loss 0.127970 Objective Loss 0.127970 LR 0.000125 Time 0.020084 -2022-12-06 11:46:55,146 - Epoch: [188][ 750/ 1200] Overall Loss 0.128053 Objective Loss 0.128053 LR 0.000125 Time 0.020070 -2022-12-06 11:46:55,338 - Epoch: [188][ 760/ 1200] Overall Loss 0.128317 Objective Loss 0.128317 LR 0.000125 Time 0.020058 -2022-12-06 11:46:55,529 - Epoch: [188][ 770/ 1200] Overall Loss 0.128184 Objective Loss 0.128184 LR 0.000125 Time 0.020045 -2022-12-06 11:46:55,720 - Epoch: [188][ 780/ 1200] Overall Loss 0.128212 Objective Loss 0.128212 LR 0.000125 Time 0.020033 -2022-12-06 11:46:55,912 - Epoch: [188][ 790/ 1200] Overall Loss 0.128039 Objective Loss 0.128039 LR 0.000125 Time 0.020021 -2022-12-06 11:46:56,103 - Epoch: [188][ 800/ 1200] Overall Loss 0.128148 Objective Loss 0.128148 LR 0.000125 Time 0.020009 -2022-12-06 11:46:56,294 - Epoch: [188][ 810/ 1200] Overall Loss 0.128138 Objective Loss 0.128138 LR 0.000125 Time 0.019997 -2022-12-06 11:46:56,486 - Epoch: [188][ 820/ 1200] Overall Loss 0.128339 Objective Loss 0.128339 LR 0.000125 Time 0.019986 -2022-12-06 11:46:56,677 - Epoch: [188][ 830/ 1200] Overall Loss 0.128353 Objective Loss 0.128353 LR 0.000125 Time 0.019975 -2022-12-06 11:46:56,868 - Epoch: [188][ 840/ 1200] Overall Loss 0.128291 Objective Loss 0.128291 LR 0.000125 Time 0.019964 -2022-12-06 11:46:57,059 - Epoch: [188][ 850/ 1200] Overall Loss 0.128248 Objective Loss 0.128248 LR 0.000125 Time 0.019954 -2022-12-06 11:46:57,251 - Epoch: [188][ 860/ 1200] Overall Loss 0.128089 Objective Loss 0.128089 LR 0.000125 Time 0.019944 -2022-12-06 11:46:57,442 - Epoch: [188][ 870/ 1200] Overall Loss 0.128085 Objective Loss 0.128085 LR 0.000125 Time 0.019933 -2022-12-06 11:46:57,634 - Epoch: [188][ 880/ 1200] Overall Loss 0.128287 Objective Loss 0.128287 LR 0.000125 Time 0.019924 -2022-12-06 11:46:57,825 - Epoch: [188][ 890/ 1200] Overall Loss 0.128202 Objective Loss 0.128202 LR 0.000125 Time 0.019915 -2022-12-06 11:46:58,017 - Epoch: [188][ 900/ 1200] Overall Loss 0.128186 Objective Loss 0.128186 LR 0.000125 Time 0.019906 -2022-12-06 11:46:58,208 - Epoch: [188][ 910/ 1200] Overall Loss 0.128402 Objective Loss 0.128402 LR 0.000125 Time 0.019897 -2022-12-06 11:46:58,400 - Epoch: [188][ 920/ 1200] Overall Loss 0.128531 Objective Loss 0.128531 LR 0.000125 Time 0.019888 -2022-12-06 11:46:58,590 - Epoch: [188][ 930/ 1200] Overall Loss 0.128655 Objective Loss 0.128655 LR 0.000125 Time 0.019879 -2022-12-06 11:46:58,781 - Epoch: [188][ 940/ 1200] Overall Loss 0.128671 Objective Loss 0.128671 LR 0.000125 Time 0.019870 -2022-12-06 11:46:58,973 - Epoch: [188][ 950/ 1200] Overall Loss 0.128646 Objective Loss 0.128646 LR 0.000125 Time 0.019862 -2022-12-06 11:46:59,166 - Epoch: [188][ 960/ 1200] Overall Loss 0.128626 Objective Loss 0.128626 LR 0.000125 Time 0.019855 -2022-12-06 11:46:59,359 - Epoch: [188][ 970/ 1200] Overall Loss 0.128688 Objective Loss 0.128688 LR 0.000125 Time 0.019849 -2022-12-06 11:46:59,552 - Epoch: [188][ 980/ 1200] Overall Loss 0.128825 Objective Loss 0.128825 LR 0.000125 Time 0.019843 -2022-12-06 11:46:59,745 - Epoch: [188][ 990/ 1200] Overall Loss 0.128848 Objective Loss 0.128848 LR 0.000125 Time 0.019837 -2022-12-06 11:46:59,938 - Epoch: [188][ 1000/ 1200] Overall Loss 0.128637 Objective Loss 0.128637 LR 0.000125 Time 0.019831 -2022-12-06 11:47:00,130 - Epoch: [188][ 1010/ 1200] Overall Loss 0.128722 Objective Loss 0.128722 LR 0.000125 Time 0.019825 -2022-12-06 11:47:00,323 - Epoch: [188][ 1020/ 1200] Overall Loss 0.128635 Objective Loss 0.128635 LR 0.000125 Time 0.019819 -2022-12-06 11:47:00,517 - Epoch: [188][ 1030/ 1200] Overall Loss 0.128836 Objective Loss 0.128836 LR 0.000125 Time 0.019814 -2022-12-06 11:47:00,709 - Epoch: [188][ 1040/ 1200] Overall Loss 0.128801 Objective Loss 0.128801 LR 0.000125 Time 0.019808 -2022-12-06 11:47:00,903 - Epoch: [188][ 1050/ 1200] Overall Loss 0.128689 Objective Loss 0.128689 LR 0.000125 Time 0.019803 -2022-12-06 11:47:01,095 - Epoch: [188][ 1060/ 1200] Overall Loss 0.128524 Objective Loss 0.128524 LR 0.000125 Time 0.019797 -2022-12-06 11:47:01,288 - Epoch: [188][ 1070/ 1200] Overall Loss 0.128470 Objective Loss 0.128470 LR 0.000125 Time 0.019792 -2022-12-06 11:47:01,480 - Epoch: [188][ 1080/ 1200] Overall Loss 0.128399 Objective Loss 0.128399 LR 0.000125 Time 0.019786 -2022-12-06 11:47:01,674 - Epoch: [188][ 1090/ 1200] Overall Loss 0.128360 Objective Loss 0.128360 LR 0.000125 Time 0.019782 -2022-12-06 11:47:01,867 - Epoch: [188][ 1100/ 1200] Overall Loss 0.128373 Objective Loss 0.128373 LR 0.000125 Time 0.019777 -2022-12-06 11:47:02,059 - Epoch: [188][ 1110/ 1200] Overall Loss 0.128329 Objective Loss 0.128329 LR 0.000125 Time 0.019771 -2022-12-06 11:47:02,252 - Epoch: [188][ 1120/ 1200] Overall Loss 0.128498 Objective Loss 0.128498 LR 0.000125 Time 0.019766 -2022-12-06 11:47:02,445 - Epoch: [188][ 1130/ 1200] Overall Loss 0.128603 Objective Loss 0.128603 LR 0.000125 Time 0.019762 -2022-12-06 11:47:02,637 - Epoch: [188][ 1140/ 1200] Overall Loss 0.128473 Objective Loss 0.128473 LR 0.000125 Time 0.019756 -2022-12-06 11:47:02,830 - Epoch: [188][ 1150/ 1200] Overall Loss 0.128491 Objective Loss 0.128491 LR 0.000125 Time 0.019752 -2022-12-06 11:47:03,022 - Epoch: [188][ 1160/ 1200] Overall Loss 0.128417 Objective Loss 0.128417 LR 0.000125 Time 0.019747 -2022-12-06 11:47:03,215 - Epoch: [188][ 1170/ 1200] Overall Loss 0.128420 Objective Loss 0.128420 LR 0.000125 Time 0.019743 -2022-12-06 11:47:03,408 - Epoch: [188][ 1180/ 1200] Overall Loss 0.128549 Objective Loss 0.128549 LR 0.000125 Time 0.019738 -2022-12-06 11:47:03,601 - Epoch: [188][ 1190/ 1200] Overall Loss 0.128583 Objective Loss 0.128583 LR 0.000125 Time 0.019734 -2022-12-06 11:47:03,827 - Epoch: [188][ 1200/ 1200] Overall Loss 0.128566 Objective Loss 0.128566 Top1 91.422594 Top5 99.372385 LR 0.000125 Time 0.019758 -2022-12-06 11:47:03,916 - --- validate (epoch=188)----------- -2022-12-06 11:47:03,916 - 34129 samples (256 per mini-batch) -2022-12-06 11:47:04,360 - Epoch: [188][ 10/ 134] Loss 0.218064 Top1 88.789062 Top5 99.062500 -2022-12-06 11:47:04,500 - Epoch: [188][ 20/ 134] Loss 0.218517 Top1 88.613281 Top5 98.984375 -2022-12-06 11:47:04,627 - Epoch: [188][ 30/ 134] Loss 0.226896 Top1 88.567708 Top5 98.828125 -2022-12-06 11:47:04,756 - Epoch: [188][ 40/ 134] Loss 0.241622 Top1 88.242188 Top5 98.515625 -2022-12-06 11:47:04,886 - Epoch: [188][ 50/ 134] Loss 0.240768 Top1 88.375000 Top5 98.546875 -2022-12-06 11:47:05,018 - Epoch: [188][ 60/ 134] Loss 0.237642 Top1 88.411458 Top5 98.587240 -2022-12-06 11:47:05,150 - Epoch: [188][ 70/ 134] Loss 0.241530 Top1 88.420759 Top5 98.582589 -2022-12-06 11:47:05,282 - Epoch: [188][ 80/ 134] Loss 0.238520 Top1 88.486328 Top5 98.549805 -2022-12-06 11:47:05,412 - Epoch: [188][ 90/ 134] Loss 0.233301 Top1 88.519965 Top5 98.546007 -2022-12-06 11:47:05,545 - Epoch: [188][ 100/ 134] Loss 0.231544 Top1 88.453125 Top5 98.546875 -2022-12-06 11:47:05,677 - Epoch: [188][ 110/ 134] Loss 0.230802 Top1 88.476562 Top5 98.607955 -2022-12-06 11:47:05,809 - Epoch: [188][ 120/ 134] Loss 0.231876 Top1 88.450521 Top5 98.645833 -2022-12-06 11:47:05,943 - Epoch: [188][ 130/ 134] Loss 0.229662 Top1 88.530649 Top5 98.659856 -2022-12-06 11:47:05,983 - Epoch: [188][ 134/ 134] Loss 0.231035 Top1 88.511237 Top5 98.649243 -2022-12-06 11:47:06,071 - ==> Top1: 88.511 Top5: 98.649 Loss: 0.231 - -2022-12-06 11:47:06,071 - ==> Confusion: -[[ 930 1 1 1 2 5 0 1 5 35 0 1 1 4 4 1 2 0 1 0 1] - [ 1 946 1 3 9 20 2 14 0 0 1 4 1 1 0 1 4 1 7 2 9] - [ 4 4 1011 17 3 2 13 8 0 1 5 2 2 2 3 3 1 2 1 4 15] - [ 2 0 8 960 1 3 0 1 0 1 8 0 5 0 9 0 3 2 8 1 8] - [ 9 4 2 0 963 4 0 1 1 5 1 2 0 3 5 6 7 3 1 0 3] - [ 1 10 0 3 7 993 2 14 2 2 1 8 2 14 1 1 0 1 0 4 3] - [ 1 2 12 2 0 3 1077 2 0 0 0 1 0 2 0 3 0 1 1 8 3] - [ 1 5 5 2 1 30 8 964 0 0 0 4 1 1 1 1 2 0 11 8 9] - [ 6 1 0 0 0 1 1 2 980 35 11 1 1 8 12 1 1 0 1 1 1] - [ 61 0 1 0 7 3 0 3 17 886 1 1 1 9 3 1 1 0 0 1 5] - [ 1 2 2 4 0 0 1 1 12 1 970 1 0 7 4 1 0 0 4 2 6] - [ 1 1 1 0 1 11 3 2 1 1 0 963 31 4 0 9 3 5 0 11 3] - [ 1 0 0 2 0 2 1 0 0 1 0 18 917 0 2 10 1 6 1 2 5] - [ 2 1 2 0 2 7 0 3 13 13 4 3 4 957 0 0 4 0 0 1 7] - [ 7 5 3 8 2 4 0 0 12 3 1 2 1 3 1070 0 0 1 3 1 4] - [ 0 0 0 2 2 1 2 1 0 0 1 3 1 2 0 1004 4 11 0 3 6] - [ 1 0 1 2 1 0 0 0 1 0 1 0 4 3 0 11 1038 0 0 2 7] - [ 3 0 1 1 1 1 2 0 0 4 0 5 16 2 1 17 0 978 0 0 4] - [ 2 3 3 8 1 3 0 21 2 1 3 3 4 0 7 0 0 1 942 1 3] - [ 0 3 1 2 1 5 4 3 0 0 2 10 5 6 0 6 6 1 1 1014 10] - [ 116 172 118 88 101 144 73 124 78 78 143 72 287 212 114 113 167 71 131 183 10641]] - -2022-12-06 11:47:06,735 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:47:06,735 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:47:06,766 - - -2022-12-06 11:47:06,766 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:47:07,692 - Epoch: [189][ 10/ 1200] Overall Loss 0.133164 Objective Loss 0.133164 LR 0.000125 Time 0.092546 -2022-12-06 11:47:07,892 - Epoch: [189][ 20/ 1200] Overall Loss 0.136493 Objective Loss 0.136493 LR 0.000125 Time 0.056235 -2022-12-06 11:47:08,085 - Epoch: [189][ 30/ 1200] Overall Loss 0.137685 Objective Loss 0.137685 LR 0.000125 Time 0.043900 -2022-12-06 11:47:08,279 - Epoch: [189][ 40/ 1200] Overall Loss 0.128560 Objective Loss 0.128560 LR 0.000125 Time 0.037750 -2022-12-06 11:47:08,470 - Epoch: [189][ 50/ 1200] Overall Loss 0.128097 Objective Loss 0.128097 LR 0.000125 Time 0.034025 -2022-12-06 11:47:08,663 - Epoch: [189][ 60/ 1200] Overall Loss 0.129369 Objective Loss 0.129369 LR 0.000125 Time 0.031558 -2022-12-06 11:47:08,854 - Epoch: [189][ 70/ 1200] Overall Loss 0.126963 Objective Loss 0.126963 LR 0.000125 Time 0.029776 -2022-12-06 11:47:09,047 - Epoch: [189][ 80/ 1200] Overall Loss 0.129364 Objective Loss 0.129364 LR 0.000125 Time 0.028452 -2022-12-06 11:47:09,238 - Epoch: [189][ 90/ 1200] Overall Loss 0.128776 Objective Loss 0.128776 LR 0.000125 Time 0.027412 -2022-12-06 11:47:09,431 - Epoch: [189][ 100/ 1200] Overall Loss 0.127722 Objective Loss 0.127722 LR 0.000125 Time 0.026589 -2022-12-06 11:47:09,623 - Epoch: [189][ 110/ 1200] Overall Loss 0.126298 Objective Loss 0.126298 LR 0.000125 Time 0.025918 -2022-12-06 11:47:09,815 - Epoch: [189][ 120/ 1200] Overall Loss 0.125345 Objective Loss 0.125345 LR 0.000125 Time 0.025355 -2022-12-06 11:47:10,007 - Epoch: [189][ 130/ 1200] Overall Loss 0.124780 Objective Loss 0.124780 LR 0.000125 Time 0.024874 -2022-12-06 11:47:10,199 - Epoch: [189][ 140/ 1200] Overall Loss 0.124572 Objective Loss 0.124572 LR 0.000125 Time 0.024464 -2022-12-06 11:47:10,390 - Epoch: [189][ 150/ 1200] Overall Loss 0.122449 Objective Loss 0.122449 LR 0.000125 Time 0.024107 -2022-12-06 11:47:10,583 - Epoch: [189][ 160/ 1200] Overall Loss 0.122846 Objective Loss 0.122846 LR 0.000125 Time 0.023797 -2022-12-06 11:47:10,774 - Epoch: [189][ 170/ 1200] Overall Loss 0.122577 Objective Loss 0.122577 LR 0.000125 Time 0.023522 -2022-12-06 11:47:10,967 - Epoch: [189][ 180/ 1200] Overall Loss 0.123326 Objective Loss 0.123326 LR 0.000125 Time 0.023283 -2022-12-06 11:47:11,159 - Epoch: [189][ 190/ 1200] Overall Loss 0.123299 Objective Loss 0.123299 LR 0.000125 Time 0.023063 -2022-12-06 11:47:11,351 - Epoch: [189][ 200/ 1200] Overall Loss 0.123733 Objective Loss 0.123733 LR 0.000125 Time 0.022868 -2022-12-06 11:47:11,542 - Epoch: [189][ 210/ 1200] Overall Loss 0.124919 Objective Loss 0.124919 LR 0.000125 Time 0.022690 -2022-12-06 11:47:11,735 - Epoch: [189][ 220/ 1200] Overall Loss 0.124726 Objective Loss 0.124726 LR 0.000125 Time 0.022530 -2022-12-06 11:47:11,927 - Epoch: [189][ 230/ 1200] Overall Loss 0.125543 Objective Loss 0.125543 LR 0.000125 Time 0.022384 -2022-12-06 11:47:12,119 - Epoch: [189][ 240/ 1200] Overall Loss 0.126338 Objective Loss 0.126338 LR 0.000125 Time 0.022250 -2022-12-06 11:47:12,311 - Epoch: [189][ 250/ 1200] Overall Loss 0.125722 Objective Loss 0.125722 LR 0.000125 Time 0.022125 -2022-12-06 11:47:12,503 - Epoch: [189][ 260/ 1200] Overall Loss 0.125609 Objective Loss 0.125609 LR 0.000125 Time 0.022010 -2022-12-06 11:47:12,695 - Epoch: [189][ 270/ 1200] Overall Loss 0.125362 Objective Loss 0.125362 LR 0.000125 Time 0.021903 -2022-12-06 11:47:12,887 - Epoch: [189][ 280/ 1200] Overall Loss 0.125775 Objective Loss 0.125775 LR 0.000125 Time 0.021806 -2022-12-06 11:47:13,079 - Epoch: [189][ 290/ 1200] Overall Loss 0.125546 Objective Loss 0.125546 LR 0.000125 Time 0.021714 -2022-12-06 11:47:13,271 - Epoch: [189][ 300/ 1200] Overall Loss 0.125479 Objective Loss 0.125479 LR 0.000125 Time 0.021629 -2022-12-06 11:47:13,463 - Epoch: [189][ 310/ 1200] Overall Loss 0.125275 Objective Loss 0.125275 LR 0.000125 Time 0.021547 -2022-12-06 11:47:13,655 - Epoch: [189][ 320/ 1200] Overall Loss 0.124958 Objective Loss 0.124958 LR 0.000125 Time 0.021474 -2022-12-06 11:47:13,847 - Epoch: [189][ 330/ 1200] Overall Loss 0.125116 Objective Loss 0.125116 LR 0.000125 Time 0.021403 -2022-12-06 11:47:14,040 - Epoch: [189][ 340/ 1200] Overall Loss 0.125371 Objective Loss 0.125371 LR 0.000125 Time 0.021338 -2022-12-06 11:47:14,232 - Epoch: [189][ 350/ 1200] Overall Loss 0.125564 Objective Loss 0.125564 LR 0.000125 Time 0.021276 -2022-12-06 11:47:14,424 - Epoch: [189][ 360/ 1200] Overall Loss 0.126086 Objective Loss 0.126086 LR 0.000125 Time 0.021218 -2022-12-06 11:47:14,616 - Epoch: [189][ 370/ 1200] Overall Loss 0.126255 Objective Loss 0.126255 LR 0.000125 Time 0.021161 -2022-12-06 11:47:14,809 - Epoch: [189][ 380/ 1200] Overall Loss 0.126607 Objective Loss 0.126607 LR 0.000125 Time 0.021109 -2022-12-06 11:47:15,000 - Epoch: [189][ 390/ 1200] Overall Loss 0.126336 Objective Loss 0.126336 LR 0.000125 Time 0.021057 -2022-12-06 11:47:15,192 - Epoch: [189][ 400/ 1200] Overall Loss 0.126503 Objective Loss 0.126503 LR 0.000125 Time 0.021010 -2022-12-06 11:47:15,384 - Epoch: [189][ 410/ 1200] Overall Loss 0.125978 Objective Loss 0.125978 LR 0.000125 Time 0.020964 -2022-12-06 11:47:15,578 - Epoch: [189][ 420/ 1200] Overall Loss 0.126055 Objective Loss 0.126055 LR 0.000125 Time 0.020924 -2022-12-06 11:47:15,769 - Epoch: [189][ 430/ 1200] Overall Loss 0.125770 Objective Loss 0.125770 LR 0.000125 Time 0.020882 -2022-12-06 11:47:15,962 - Epoch: [189][ 440/ 1200] Overall Loss 0.125451 Objective Loss 0.125451 LR 0.000125 Time 0.020845 -2022-12-06 11:47:16,154 - Epoch: [189][ 450/ 1200] Overall Loss 0.125652 Objective Loss 0.125652 LR 0.000125 Time 0.020807 -2022-12-06 11:47:16,347 - Epoch: [189][ 460/ 1200] Overall Loss 0.125580 Objective Loss 0.125580 LR 0.000125 Time 0.020773 -2022-12-06 11:47:16,539 - Epoch: [189][ 470/ 1200] Overall Loss 0.125359 Objective Loss 0.125359 LR 0.000125 Time 0.020738 -2022-12-06 11:47:16,731 - Epoch: [189][ 480/ 1200] Overall Loss 0.125307 Objective Loss 0.125307 LR 0.000125 Time 0.020705 -2022-12-06 11:47:16,923 - Epoch: [189][ 490/ 1200] Overall Loss 0.125275 Objective Loss 0.125275 LR 0.000125 Time 0.020673 -2022-12-06 11:47:17,115 - Epoch: [189][ 500/ 1200] Overall Loss 0.125401 Objective Loss 0.125401 LR 0.000125 Time 0.020643 -2022-12-06 11:47:17,308 - Epoch: [189][ 510/ 1200] Overall Loss 0.125341 Objective Loss 0.125341 LR 0.000125 Time 0.020615 -2022-12-06 11:47:17,501 - Epoch: [189][ 520/ 1200] Overall Loss 0.125463 Objective Loss 0.125463 LR 0.000125 Time 0.020588 -2022-12-06 11:47:17,692 - Epoch: [189][ 530/ 1200] Overall Loss 0.125264 Objective Loss 0.125264 LR 0.000125 Time 0.020560 -2022-12-06 11:47:17,884 - Epoch: [189][ 540/ 1200] Overall Loss 0.125520 Objective Loss 0.125520 LR 0.000125 Time 0.020534 -2022-12-06 11:47:18,076 - Epoch: [189][ 550/ 1200] Overall Loss 0.125445 Objective Loss 0.125445 LR 0.000125 Time 0.020509 -2022-12-06 11:47:18,270 - Epoch: [189][ 560/ 1200] Overall Loss 0.125565 Objective Loss 0.125565 LR 0.000125 Time 0.020487 -2022-12-06 11:47:18,462 - Epoch: [189][ 570/ 1200] Overall Loss 0.125887 Objective Loss 0.125887 LR 0.000125 Time 0.020464 -2022-12-06 11:47:18,655 - Epoch: [189][ 580/ 1200] Overall Loss 0.125727 Objective Loss 0.125727 LR 0.000125 Time 0.020443 -2022-12-06 11:47:18,848 - Epoch: [189][ 590/ 1200] Overall Loss 0.125679 Objective Loss 0.125679 LR 0.000125 Time 0.020422 -2022-12-06 11:47:19,040 - Epoch: [189][ 600/ 1200] Overall Loss 0.125902 Objective Loss 0.125902 LR 0.000125 Time 0.020402 -2022-12-06 11:47:19,232 - Epoch: [189][ 610/ 1200] Overall Loss 0.125990 Objective Loss 0.125990 LR 0.000125 Time 0.020381 -2022-12-06 11:47:19,424 - Epoch: [189][ 620/ 1200] Overall Loss 0.126099 Objective Loss 0.126099 LR 0.000125 Time 0.020362 -2022-12-06 11:47:19,616 - Epoch: [189][ 630/ 1200] Overall Loss 0.126306 Objective Loss 0.126306 LR 0.000125 Time 0.020341 -2022-12-06 11:47:19,809 - Epoch: [189][ 640/ 1200] Overall Loss 0.126337 Objective Loss 0.126337 LR 0.000125 Time 0.020324 -2022-12-06 11:47:20,000 - Epoch: [189][ 650/ 1200] Overall Loss 0.126242 Objective Loss 0.126242 LR 0.000125 Time 0.020305 -2022-12-06 11:47:20,193 - Epoch: [189][ 660/ 1200] Overall Loss 0.126339 Objective Loss 0.126339 LR 0.000125 Time 0.020288 -2022-12-06 11:47:20,384 - Epoch: [189][ 670/ 1200] Overall Loss 0.126255 Objective Loss 0.126255 LR 0.000125 Time 0.020271 -2022-12-06 11:47:20,577 - Epoch: [189][ 680/ 1200] Overall Loss 0.126040 Objective Loss 0.126040 LR 0.000125 Time 0.020255 -2022-12-06 11:47:20,769 - Epoch: [189][ 690/ 1200] Overall Loss 0.126056 Objective Loss 0.126056 LR 0.000125 Time 0.020239 -2022-12-06 11:47:20,961 - Epoch: [189][ 700/ 1200] Overall Loss 0.126021 Objective Loss 0.126021 LR 0.000125 Time 0.020224 -2022-12-06 11:47:21,153 - Epoch: [189][ 710/ 1200] Overall Loss 0.126310 Objective Loss 0.126310 LR 0.000125 Time 0.020209 -2022-12-06 11:47:21,344 - Epoch: [189][ 720/ 1200] Overall Loss 0.126336 Objective Loss 0.126336 LR 0.000125 Time 0.020193 -2022-12-06 11:47:21,534 - Epoch: [189][ 730/ 1200] Overall Loss 0.126134 Objective Loss 0.126134 LR 0.000125 Time 0.020175 -2022-12-06 11:47:21,724 - Epoch: [189][ 740/ 1200] Overall Loss 0.126203 Objective Loss 0.126203 LR 0.000125 Time 0.020158 -2022-12-06 11:47:21,913 - Epoch: [189][ 750/ 1200] Overall Loss 0.126227 Objective Loss 0.126227 LR 0.000125 Time 0.020141 -2022-12-06 11:47:22,103 - Epoch: [189][ 760/ 1200] Overall Loss 0.126225 Objective Loss 0.126225 LR 0.000125 Time 0.020125 -2022-12-06 11:47:22,292 - Epoch: [189][ 770/ 1200] Overall Loss 0.126283 Objective Loss 0.126283 LR 0.000125 Time 0.020109 -2022-12-06 11:47:22,482 - Epoch: [189][ 780/ 1200] Overall Loss 0.126206 Objective Loss 0.126206 LR 0.000125 Time 0.020094 -2022-12-06 11:47:22,672 - Epoch: [189][ 790/ 1200] Overall Loss 0.126274 Objective Loss 0.126274 LR 0.000125 Time 0.020079 -2022-12-06 11:47:22,861 - Epoch: [189][ 800/ 1200] Overall Loss 0.126269 Objective Loss 0.126269 LR 0.000125 Time 0.020064 -2022-12-06 11:47:23,051 - Epoch: [189][ 810/ 1200] Overall Loss 0.126284 Objective Loss 0.126284 LR 0.000125 Time 0.020050 -2022-12-06 11:47:23,241 - Epoch: [189][ 820/ 1200] Overall Loss 0.126301 Objective Loss 0.126301 LR 0.000125 Time 0.020037 -2022-12-06 11:47:23,431 - Epoch: [189][ 830/ 1200] Overall Loss 0.126055 Objective Loss 0.126055 LR 0.000125 Time 0.020023 -2022-12-06 11:47:23,621 - Epoch: [189][ 840/ 1200] Overall Loss 0.125993 Objective Loss 0.125993 LR 0.000125 Time 0.020011 -2022-12-06 11:47:23,812 - Epoch: [189][ 850/ 1200] Overall Loss 0.125917 Objective Loss 0.125917 LR 0.000125 Time 0.019999 -2022-12-06 11:47:24,002 - Epoch: [189][ 860/ 1200] Overall Loss 0.125966 Objective Loss 0.125966 LR 0.000125 Time 0.019987 -2022-12-06 11:47:24,192 - Epoch: [189][ 870/ 1200] Overall Loss 0.126048 Objective Loss 0.126048 LR 0.000125 Time 0.019975 -2022-12-06 11:47:24,382 - Epoch: [189][ 880/ 1200] Overall Loss 0.125983 Objective Loss 0.125983 LR 0.000125 Time 0.019963 -2022-12-06 11:47:24,572 - Epoch: [189][ 890/ 1200] Overall Loss 0.126221 Objective Loss 0.126221 LR 0.000125 Time 0.019952 -2022-12-06 11:47:24,762 - Epoch: [189][ 900/ 1200] Overall Loss 0.126275 Objective Loss 0.126275 LR 0.000125 Time 0.019941 -2022-12-06 11:47:24,952 - Epoch: [189][ 910/ 1200] Overall Loss 0.126290 Objective Loss 0.126290 LR 0.000125 Time 0.019930 -2022-12-06 11:47:25,142 - Epoch: [189][ 920/ 1200] Overall Loss 0.126293 Objective Loss 0.126293 LR 0.000125 Time 0.019919 -2022-12-06 11:47:25,332 - Epoch: [189][ 930/ 1200] Overall Loss 0.126388 Objective Loss 0.126388 LR 0.000125 Time 0.019909 -2022-12-06 11:47:25,523 - Epoch: [189][ 940/ 1200] Overall Loss 0.126550 Objective Loss 0.126550 LR 0.000125 Time 0.019899 -2022-12-06 11:47:25,713 - Epoch: [189][ 950/ 1200] Overall Loss 0.126361 Objective Loss 0.126361 LR 0.000125 Time 0.019890 -2022-12-06 11:47:25,903 - Epoch: [189][ 960/ 1200] Overall Loss 0.126553 Objective Loss 0.126553 LR 0.000125 Time 0.019879 -2022-12-06 11:47:26,093 - Epoch: [189][ 970/ 1200] Overall Loss 0.126471 Objective Loss 0.126471 LR 0.000125 Time 0.019871 -2022-12-06 11:47:26,284 - Epoch: [189][ 980/ 1200] Overall Loss 0.126429 Objective Loss 0.126429 LR 0.000125 Time 0.019861 -2022-12-06 11:47:26,473 - Epoch: [189][ 990/ 1200] Overall Loss 0.126393 Objective Loss 0.126393 LR 0.000125 Time 0.019852 -2022-12-06 11:47:26,663 - Epoch: [189][ 1000/ 1200] Overall Loss 0.126417 Objective Loss 0.126417 LR 0.000125 Time 0.019842 -2022-12-06 11:47:26,854 - Epoch: [189][ 1010/ 1200] Overall Loss 0.126294 Objective Loss 0.126294 LR 0.000125 Time 0.019834 -2022-12-06 11:47:27,044 - Epoch: [189][ 1020/ 1200] Overall Loss 0.126375 Objective Loss 0.126375 LR 0.000125 Time 0.019826 -2022-12-06 11:47:27,234 - Epoch: [189][ 1030/ 1200] Overall Loss 0.126296 Objective Loss 0.126296 LR 0.000125 Time 0.019817 -2022-12-06 11:47:27,424 - Epoch: [189][ 1040/ 1200] Overall Loss 0.126326 Objective Loss 0.126326 LR 0.000125 Time 0.019808 -2022-12-06 11:47:27,614 - Epoch: [189][ 1050/ 1200] Overall Loss 0.126509 Objective Loss 0.126509 LR 0.000125 Time 0.019800 -2022-12-06 11:47:27,804 - Epoch: [189][ 1060/ 1200] Overall Loss 0.126475 Objective Loss 0.126475 LR 0.000125 Time 0.019792 -2022-12-06 11:47:27,994 - Epoch: [189][ 1070/ 1200] Overall Loss 0.126572 Objective Loss 0.126572 LR 0.000125 Time 0.019784 -2022-12-06 11:47:28,184 - Epoch: [189][ 1080/ 1200] Overall Loss 0.126571 Objective Loss 0.126571 LR 0.000125 Time 0.019776 -2022-12-06 11:47:28,374 - Epoch: [189][ 1090/ 1200] Overall Loss 0.126632 Objective Loss 0.126632 LR 0.000125 Time 0.019769 -2022-12-06 11:47:28,564 - Epoch: [189][ 1100/ 1200] Overall Loss 0.126633 Objective Loss 0.126633 LR 0.000125 Time 0.019762 -2022-12-06 11:47:28,754 - Epoch: [189][ 1110/ 1200] Overall Loss 0.126669 Objective Loss 0.126669 LR 0.000125 Time 0.019754 -2022-12-06 11:47:28,944 - Epoch: [189][ 1120/ 1200] Overall Loss 0.126677 Objective Loss 0.126677 LR 0.000125 Time 0.019747 -2022-12-06 11:47:29,134 - Epoch: [189][ 1130/ 1200] Overall Loss 0.126684 Objective Loss 0.126684 LR 0.000125 Time 0.019740 -2022-12-06 11:47:29,324 - Epoch: [189][ 1140/ 1200] Overall Loss 0.126821 Objective Loss 0.126821 LR 0.000125 Time 0.019733 -2022-12-06 11:47:29,514 - Epoch: [189][ 1150/ 1200] Overall Loss 0.126988 Objective Loss 0.126988 LR 0.000125 Time 0.019726 -2022-12-06 11:47:29,704 - Epoch: [189][ 1160/ 1200] Overall Loss 0.127209 Objective Loss 0.127209 LR 0.000125 Time 0.019719 -2022-12-06 11:47:29,894 - Epoch: [189][ 1170/ 1200] Overall Loss 0.127264 Objective Loss 0.127264 LR 0.000125 Time 0.019712 -2022-12-06 11:47:30,084 - Epoch: [189][ 1180/ 1200] Overall Loss 0.127468 Objective Loss 0.127468 LR 0.000125 Time 0.019706 -2022-12-06 11:47:30,274 - Epoch: [189][ 1190/ 1200] Overall Loss 0.127394 Objective Loss 0.127394 LR 0.000125 Time 0.019700 -2022-12-06 11:47:30,503 - Epoch: [189][ 1200/ 1200] Overall Loss 0.127411 Objective Loss 0.127411 Top1 91.213389 Top5 98.744770 LR 0.000125 Time 0.019726 -2022-12-06 11:47:30,596 - --- validate (epoch=189)----------- -2022-12-06 11:47:30,596 - 34129 samples (256 per mini-batch) -2022-12-06 11:47:31,039 - Epoch: [189][ 10/ 134] Loss 0.264511 Top1 88.046875 Top5 98.476562 -2022-12-06 11:47:31,172 - Epoch: [189][ 20/ 134] Loss 0.237700 Top1 88.476562 Top5 98.476562 -2022-12-06 11:47:31,301 - Epoch: [189][ 30/ 134] Loss 0.239772 Top1 88.411458 Top5 98.502604 -2022-12-06 11:47:31,430 - Epoch: [189][ 40/ 134] Loss 0.234753 Top1 88.632812 Top5 98.564453 -2022-12-06 11:47:31,560 - Epoch: [189][ 50/ 134] Loss 0.237139 Top1 88.375000 Top5 98.578125 -2022-12-06 11:47:31,689 - Epoch: [189][ 60/ 134] Loss 0.236015 Top1 88.496094 Top5 98.606771 -2022-12-06 11:47:31,821 - Epoch: [189][ 70/ 134] Loss 0.237090 Top1 88.420759 Top5 98.577009 -2022-12-06 11:47:31,954 - Epoch: [189][ 80/ 134] Loss 0.235018 Top1 88.491211 Top5 98.608398 -2022-12-06 11:47:32,086 - Epoch: [189][ 90/ 134] Loss 0.232744 Top1 88.554688 Top5 98.606771 -2022-12-06 11:47:32,218 - Epoch: [189][ 100/ 134] Loss 0.234480 Top1 88.417969 Top5 98.601562 -2022-12-06 11:47:32,351 - Epoch: [189][ 110/ 134] Loss 0.234107 Top1 88.409091 Top5 98.622159 -2022-12-06 11:47:32,484 - Epoch: [189][ 120/ 134] Loss 0.231495 Top1 88.483073 Top5 98.642578 -2022-12-06 11:47:32,618 - Epoch: [189][ 130/ 134] Loss 0.228485 Top1 88.473558 Top5 98.671875 -2022-12-06 11:47:32,657 - Epoch: [189][ 134/ 134] Loss 0.227593 Top1 88.473146 Top5 98.678543 -2022-12-06 11:47:32,745 - ==> Top1: 88.473 Top5: 98.679 Loss: 0.228 - -2022-12-06 11:47:32,746 - ==> Confusion: -[[ 931 1 2 3 2 5 0 0 4 38 0 1 1 1 3 1 0 0 0 0 3] - [ 1 943 2 2 8 21 2 14 2 0 2 4 0 1 0 1 2 2 8 3 9] - [ 4 2 1021 13 4 3 12 8 0 5 3 4 2 1 2 1 1 2 2 5 8] - [ 2 0 13 960 0 3 1 0 0 3 6 0 3 2 8 0 1 1 9 0 8] - [ 8 4 3 0 962 3 1 1 1 7 1 3 0 3 9 3 5 2 0 0 4] - [ 0 7 0 4 4 994 2 15 2 2 1 13 3 11 1 1 0 0 1 4 4] - [ 2 0 8 3 1 2 1081 2 0 0 0 1 0 2 0 4 0 2 1 8 1] - [ 3 5 3 2 1 32 8 967 0 1 1 4 0 1 1 1 1 0 11 7 5] - [ 4 1 0 0 0 2 1 0 987 42 4 1 1 5 11 1 2 0 0 1 1] - [ 57 1 0 0 4 3 0 1 18 898 1 1 0 7 2 0 0 1 1 0 6] - [ 1 3 4 6 1 0 1 3 10 1 953 0 0 12 6 1 0 0 8 0 9] - [ 1 0 2 0 1 7 4 2 1 0 0 993 15 4 0 3 3 5 0 4 6] - [ 0 1 1 4 0 2 0 1 0 1 0 27 898 1 0 8 1 12 1 4 7] - [ 2 1 1 0 1 7 0 2 12 13 0 4 2 968 0 1 1 0 0 1 7] - [ 7 2 2 10 3 5 0 0 14 5 0 1 1 5 1065 0 1 1 4 0 4] - [ 0 0 0 2 3 0 1 1 1 1 1 6 2 3 0 997 7 10 1 2 5] - [ 1 0 1 2 2 2 1 1 1 0 0 1 2 3 0 11 1033 0 0 5 6] - [ 4 0 1 1 0 1 0 0 0 5 0 5 9 1 1 14 0 990 0 0 4] - [ 1 2 5 7 0 3 0 25 1 1 4 2 1 2 6 1 0 2 939 2 4] - [ 2 3 2 2 1 5 2 5 0 1 1 13 5 7 0 6 2 2 0 1015 6] - [ 130 188 150 103 95 172 79 135 80 87 100 82 255 229 126 105 118 77 121 197 10597]] - -2022-12-06 11:47:33,410 - ==> Best [Top1: 88.705 Top5: 98.667 Sparsity:0.00 Params: 5376 on epoch: 181] -2022-12-06 11:47:33,410 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:47:33,416 - - -2022-12-06 11:47:33,416 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:47:34,350 - Epoch: [190][ 10/ 1200] Overall Loss 0.122961 Objective Loss 0.122961 LR 0.000063 Time 0.093297 -2022-12-06 11:47:34,550 - Epoch: [190][ 20/ 1200] Overall Loss 0.118758 Objective Loss 0.118758 LR 0.000063 Time 0.056650 -2022-12-06 11:47:34,743 - Epoch: [190][ 30/ 1200] Overall Loss 0.120215 Objective Loss 0.120215 LR 0.000063 Time 0.044178 -2022-12-06 11:47:34,936 - Epoch: [190][ 40/ 1200] Overall Loss 0.123431 Objective Loss 0.123431 LR 0.000063 Time 0.037942 -2022-12-06 11:47:35,128 - Epoch: [190][ 50/ 1200] Overall Loss 0.124090 Objective Loss 0.124090 LR 0.000063 Time 0.034182 -2022-12-06 11:47:35,320 - Epoch: [190][ 60/ 1200] Overall Loss 0.123236 Objective Loss 0.123236 LR 0.000063 Time 0.031677 -2022-12-06 11:47:35,511 - Epoch: [190][ 70/ 1200] Overall Loss 0.122509 Objective Loss 0.122509 LR 0.000063 Time 0.029875 -2022-12-06 11:47:35,703 - Epoch: [190][ 80/ 1200] Overall Loss 0.122012 Objective Loss 0.122012 LR 0.000063 Time 0.028532 -2022-12-06 11:47:35,895 - Epoch: [190][ 90/ 1200] Overall Loss 0.122266 Objective Loss 0.122266 LR 0.000063 Time 0.027483 -2022-12-06 11:47:36,087 - Epoch: [190][ 100/ 1200] Overall Loss 0.122044 Objective Loss 0.122044 LR 0.000063 Time 0.026658 -2022-12-06 11:47:36,279 - Epoch: [190][ 110/ 1200] Overall Loss 0.121364 Objective Loss 0.121364 LR 0.000063 Time 0.025974 -2022-12-06 11:47:36,471 - Epoch: [190][ 120/ 1200] Overall Loss 0.122594 Objective Loss 0.122594 LR 0.000063 Time 0.025400 -2022-12-06 11:47:36,663 - Epoch: [190][ 130/ 1200] Overall Loss 0.122421 Objective Loss 0.122421 LR 0.000063 Time 0.024916 -2022-12-06 11:47:36,855 - Epoch: [190][ 140/ 1200] Overall Loss 0.124045 Objective Loss 0.124045 LR 0.000063 Time 0.024508 -2022-12-06 11:47:37,046 - Epoch: [190][ 150/ 1200] Overall Loss 0.123646 Objective Loss 0.123646 LR 0.000063 Time 0.024144 -2022-12-06 11:47:37,239 - Epoch: [190][ 160/ 1200] Overall Loss 0.122464 Objective Loss 0.122464 LR 0.000063 Time 0.023836 -2022-12-06 11:47:37,430 - Epoch: [190][ 170/ 1200] Overall Loss 0.123024 Objective Loss 0.123024 LR 0.000063 Time 0.023554 -2022-12-06 11:47:37,623 - Epoch: [190][ 180/ 1200] Overall Loss 0.123464 Objective Loss 0.123464 LR 0.000063 Time 0.023314 -2022-12-06 11:47:37,814 - Epoch: [190][ 190/ 1200] Overall Loss 0.123905 Objective Loss 0.123905 LR 0.000063 Time 0.023092 -2022-12-06 11:47:38,007 - Epoch: [190][ 200/ 1200] Overall Loss 0.123602 Objective Loss 0.123602 LR 0.000063 Time 0.022900 -2022-12-06 11:47:38,199 - Epoch: [190][ 210/ 1200] Overall Loss 0.122398 Objective Loss 0.122398 LR 0.000063 Time 0.022720 -2022-12-06 11:47:38,391 - Epoch: [190][ 220/ 1200] Overall Loss 0.121873 Objective Loss 0.121873 LR 0.000063 Time 0.022560 -2022-12-06 11:47:38,584 - Epoch: [190][ 230/ 1200] Overall Loss 0.121479 Objective Loss 0.121479 LR 0.000063 Time 0.022412 -2022-12-06 11:47:38,776 - Epoch: [190][ 240/ 1200] Overall Loss 0.121374 Objective Loss 0.121374 LR 0.000063 Time 0.022278 -2022-12-06 11:47:38,969 - Epoch: [190][ 250/ 1200] Overall Loss 0.120986 Objective Loss 0.120986 LR 0.000063 Time 0.022155 -2022-12-06 11:47:39,161 - Epoch: [190][ 260/ 1200] Overall Loss 0.121654 Objective Loss 0.121654 LR 0.000063 Time 0.022041 -2022-12-06 11:47:39,353 - Epoch: [190][ 270/ 1200] Overall Loss 0.121511 Objective Loss 0.121511 LR 0.000063 Time 0.021934 -2022-12-06 11:47:39,546 - Epoch: [190][ 280/ 1200] Overall Loss 0.121725 Objective Loss 0.121725 LR 0.000063 Time 0.021836 -2022-12-06 11:47:39,737 - Epoch: [190][ 290/ 1200] Overall Loss 0.121589 Objective Loss 0.121589 LR 0.000063 Time 0.021741 -2022-12-06 11:47:39,930 - Epoch: [190][ 300/ 1200] Overall Loss 0.121468 Objective Loss 0.121468 LR 0.000063 Time 0.021658 -2022-12-06 11:47:40,122 - Epoch: [190][ 310/ 1200] Overall Loss 0.121206 Objective Loss 0.121206 LR 0.000063 Time 0.021575 -2022-12-06 11:47:40,314 - Epoch: [190][ 320/ 1200] Overall Loss 0.121782 Objective Loss 0.121782 LR 0.000063 Time 0.021501 -2022-12-06 11:47:40,506 - Epoch: [190][ 330/ 1200] Overall Loss 0.122282 Objective Loss 0.122282 LR 0.000063 Time 0.021429 -2022-12-06 11:47:40,698 - Epoch: [190][ 340/ 1200] Overall Loss 0.122668 Objective Loss 0.122668 LR 0.000063 Time 0.021362 -2022-12-06 11:47:40,890 - Epoch: [190][ 350/ 1200] Overall Loss 0.122547 Objective Loss 0.122547 LR 0.000063 Time 0.021298 -2022-12-06 11:47:41,082 - Epoch: [190][ 360/ 1200] Overall Loss 0.122779 Objective Loss 0.122779 LR 0.000063 Time 0.021240 -2022-12-06 11:47:41,274 - Epoch: [190][ 370/ 1200] Overall Loss 0.123131 Objective Loss 0.123131 LR 0.000063 Time 0.021183 -2022-12-06 11:47:41,467 - Epoch: [190][ 380/ 1200] Overall Loss 0.123100 Objective Loss 0.123100 LR 0.000063 Time 0.021131 -2022-12-06 11:47:41,659 - Epoch: [190][ 390/ 1200] Overall Loss 0.123458 Objective Loss 0.123458 LR 0.000063 Time 0.021081 -2022-12-06 11:47:41,853 - Epoch: [190][ 400/ 1200] Overall Loss 0.123107 Objective Loss 0.123107 LR 0.000063 Time 0.021036 -2022-12-06 11:47:42,045 - Epoch: [190][ 410/ 1200] Overall Loss 0.123158 Objective Loss 0.123158 LR 0.000063 Time 0.020990 -2022-12-06 11:47:42,237 - Epoch: [190][ 420/ 1200] Overall Loss 0.123065 Objective Loss 0.123065 LR 0.000063 Time 0.020948 -2022-12-06 11:47:42,429 - Epoch: [190][ 430/ 1200] Overall Loss 0.123118 Objective Loss 0.123118 LR 0.000063 Time 0.020904 -2022-12-06 11:47:42,621 - Epoch: [190][ 440/ 1200] Overall Loss 0.123208 Objective Loss 0.123208 LR 0.000063 Time 0.020864 -2022-12-06 11:47:42,813 - Epoch: [190][ 450/ 1200] Overall Loss 0.123386 Objective Loss 0.123386 LR 0.000063 Time 0.020826 -2022-12-06 11:47:43,005 - Epoch: [190][ 460/ 1200] Overall Loss 0.123273 Objective Loss 0.123273 LR 0.000063 Time 0.020791 -2022-12-06 11:47:43,197 - Epoch: [190][ 470/ 1200] Overall Loss 0.123099 Objective Loss 0.123099 LR 0.000063 Time 0.020756 -2022-12-06 11:47:43,390 - Epoch: [190][ 480/ 1200] Overall Loss 0.122768 Objective Loss 0.122768 LR 0.000063 Time 0.020724 -2022-12-06 11:47:43,581 - Epoch: [190][ 490/ 1200] Overall Loss 0.122340 Objective Loss 0.122340 LR 0.000063 Time 0.020691 -2022-12-06 11:47:43,774 - Epoch: [190][ 500/ 1200] Overall Loss 0.122365 Objective Loss 0.122365 LR 0.000063 Time 0.020661 -2022-12-06 11:47:43,966 - Epoch: [190][ 510/ 1200] Overall Loss 0.122711 Objective Loss 0.122711 LR 0.000063 Time 0.020631 -2022-12-06 11:47:44,158 - Epoch: [190][ 520/ 1200] Overall Loss 0.122707 Objective Loss 0.122707 LR 0.000063 Time 0.020603 -2022-12-06 11:47:44,350 - Epoch: [190][ 530/ 1200] Overall Loss 0.122682 Objective Loss 0.122682 LR 0.000063 Time 0.020576 -2022-12-06 11:47:44,543 - Epoch: [190][ 540/ 1200] Overall Loss 0.122715 Objective Loss 0.122715 LR 0.000063 Time 0.020551 -2022-12-06 11:47:44,735 - Epoch: [190][ 550/ 1200] Overall Loss 0.122632 Objective Loss 0.122632 LR 0.000063 Time 0.020525 -2022-12-06 11:47:44,928 - Epoch: [190][ 560/ 1200] Overall Loss 0.122603 Objective Loss 0.122603 LR 0.000063 Time 0.020501 -2022-12-06 11:47:45,120 - Epoch: [190][ 570/ 1200] Overall Loss 0.122663 Objective Loss 0.122663 LR 0.000063 Time 0.020478 -2022-12-06 11:47:45,312 - Epoch: [190][ 580/ 1200] Overall Loss 0.122561 Objective Loss 0.122561 LR 0.000063 Time 0.020455 -2022-12-06 11:47:45,504 - Epoch: [190][ 590/ 1200] Overall Loss 0.122613 Objective Loss 0.122613 LR 0.000063 Time 0.020433 -2022-12-06 11:47:45,696 - Epoch: [190][ 600/ 1200] Overall Loss 0.122655 Objective Loss 0.122655 LR 0.000063 Time 0.020412 -2022-12-06 11:47:45,888 - Epoch: [190][ 610/ 1200] Overall Loss 0.122721 Objective Loss 0.122721 LR 0.000063 Time 0.020391 -2022-12-06 11:47:46,080 - Epoch: [190][ 620/ 1200] Overall Loss 0.122741 Objective Loss 0.122741 LR 0.000063 Time 0.020371 -2022-12-06 11:47:46,272 - Epoch: [190][ 630/ 1200] Overall Loss 0.122550 Objective Loss 0.122550 LR 0.000063 Time 0.020352 -2022-12-06 11:47:46,465 - Epoch: [190][ 640/ 1200] Overall Loss 0.122357 Objective Loss 0.122357 LR 0.000063 Time 0.020334 -2022-12-06 11:47:46,657 - Epoch: [190][ 650/ 1200] Overall Loss 0.122642 Objective Loss 0.122642 LR 0.000063 Time 0.020316 -2022-12-06 11:47:46,849 - Epoch: [190][ 660/ 1200] Overall Loss 0.122547 Objective Loss 0.122547 LR 0.000063 Time 0.020299 -2022-12-06 11:47:47,041 - Epoch: [190][ 670/ 1200] Overall Loss 0.122805 Objective Loss 0.122805 LR 0.000063 Time 0.020282 -2022-12-06 11:47:47,234 - Epoch: [190][ 680/ 1200] Overall Loss 0.122656 Objective Loss 0.122656 LR 0.000063 Time 0.020266 -2022-12-06 11:47:47,426 - Epoch: [190][ 690/ 1200] Overall Loss 0.122747 Objective Loss 0.122747 LR 0.000063 Time 0.020249 -2022-12-06 11:47:47,618 - Epoch: [190][ 700/ 1200] Overall Loss 0.122876 Objective Loss 0.122876 LR 0.000063 Time 0.020234 -2022-12-06 11:47:47,810 - Epoch: [190][ 710/ 1200] Overall Loss 0.122969 Objective Loss 0.122969 LR 0.000063 Time 0.020219 -2022-12-06 11:47:48,003 - Epoch: [190][ 720/ 1200] Overall Loss 0.123070 Objective Loss 0.123070 LR 0.000063 Time 0.020205 -2022-12-06 11:47:48,195 - Epoch: [190][ 730/ 1200] Overall Loss 0.123233 Objective Loss 0.123233 LR 0.000063 Time 0.020191 -2022-12-06 11:47:48,388 - Epoch: [190][ 740/ 1200] Overall Loss 0.123232 Objective Loss 0.123232 LR 0.000063 Time 0.020177 -2022-12-06 11:47:48,580 - Epoch: [190][ 750/ 1200] Overall Loss 0.123383 Objective Loss 0.123383 LR 0.000063 Time 0.020163 -2022-12-06 11:47:48,772 - Epoch: [190][ 760/ 1200] Overall Loss 0.123329 Objective Loss 0.123329 LR 0.000063 Time 0.020150 -2022-12-06 11:47:48,964 - Epoch: [190][ 770/ 1200] Overall Loss 0.123512 Objective Loss 0.123512 LR 0.000063 Time 0.020137 -2022-12-06 11:47:49,156 - Epoch: [190][ 780/ 1200] Overall Loss 0.123375 Objective Loss 0.123375 LR 0.000063 Time 0.020125 -2022-12-06 11:47:49,348 - Epoch: [190][ 790/ 1200] Overall Loss 0.123398 Objective Loss 0.123398 LR 0.000063 Time 0.020112 -2022-12-06 11:47:49,540 - Epoch: [190][ 800/ 1200] Overall Loss 0.123268 Objective Loss 0.123268 LR 0.000063 Time 0.020100 -2022-12-06 11:47:49,732 - Epoch: [190][ 810/ 1200] Overall Loss 0.123524 Objective Loss 0.123524 LR 0.000063 Time 0.020089 -2022-12-06 11:47:49,925 - Epoch: [190][ 820/ 1200] Overall Loss 0.123597 Objective Loss 0.123597 LR 0.000063 Time 0.020078 -2022-12-06 11:47:50,117 - Epoch: [190][ 830/ 1200] Overall Loss 0.123713 Objective Loss 0.123713 LR 0.000063 Time 0.020067 -2022-12-06 11:47:50,310 - Epoch: [190][ 840/ 1200] Overall Loss 0.124042 Objective Loss 0.124042 LR 0.000063 Time 0.020057 -2022-12-06 11:47:50,502 - Epoch: [190][ 850/ 1200] Overall Loss 0.123997 Objective Loss 0.123997 LR 0.000063 Time 0.020046 -2022-12-06 11:47:50,694 - Epoch: [190][ 860/ 1200] Overall Loss 0.123898 Objective Loss 0.123898 LR 0.000063 Time 0.020036 -2022-12-06 11:47:50,886 - Epoch: [190][ 870/ 1200] Overall Loss 0.123636 Objective Loss 0.123636 LR 0.000063 Time 0.020026 -2022-12-06 11:47:51,079 - Epoch: [190][ 880/ 1200] Overall Loss 0.123622 Objective Loss 0.123622 LR 0.000063 Time 0.020017 -2022-12-06 11:47:51,270 - Epoch: [190][ 890/ 1200] Overall Loss 0.123597 Objective Loss 0.123597 LR 0.000063 Time 0.020006 -2022-12-06 11:47:51,462 - Epoch: [190][ 900/ 1200] Overall Loss 0.123829 Objective Loss 0.123829 LR 0.000063 Time 0.019997 -2022-12-06 11:47:51,654 - Epoch: [190][ 910/ 1200] Overall Loss 0.123750 Objective Loss 0.123750 LR 0.000063 Time 0.019987 -2022-12-06 11:47:51,847 - Epoch: [190][ 920/ 1200] Overall Loss 0.123731 Objective Loss 0.123731 LR 0.000063 Time 0.019979 -2022-12-06 11:47:52,039 - Epoch: [190][ 930/ 1200] Overall Loss 0.124027 Objective Loss 0.124027 LR 0.000063 Time 0.019970 -2022-12-06 11:47:52,232 - Epoch: [190][ 940/ 1200] Overall Loss 0.123872 Objective Loss 0.123872 LR 0.000063 Time 0.019962 -2022-12-06 11:47:52,424 - Epoch: [190][ 950/ 1200] Overall Loss 0.123873 Objective Loss 0.123873 LR 0.000063 Time 0.019954 -2022-12-06 11:47:52,617 - Epoch: [190][ 960/ 1200] Overall Loss 0.123989 Objective Loss 0.123989 LR 0.000063 Time 0.019946 -2022-12-06 11:47:52,809 - Epoch: [190][ 970/ 1200] Overall Loss 0.123895 Objective Loss 0.123895 LR 0.000063 Time 0.019938 -2022-12-06 11:47:53,002 - Epoch: [190][ 980/ 1200] Overall Loss 0.123842 Objective Loss 0.123842 LR 0.000063 Time 0.019931 -2022-12-06 11:47:53,193 - Epoch: [190][ 990/ 1200] Overall Loss 0.123859 Objective Loss 0.123859 LR 0.000063 Time 0.019922 -2022-12-06 11:47:53,385 - Epoch: [190][ 1000/ 1200] Overall Loss 0.123671 Objective Loss 0.123671 LR 0.000063 Time 0.019914 -2022-12-06 11:47:53,575 - Epoch: [190][ 1010/ 1200] Overall Loss 0.123625 Objective Loss 0.123625 LR 0.000063 Time 0.019904 -2022-12-06 11:47:53,765 - Epoch: [190][ 1020/ 1200] Overall Loss 0.123475 Objective Loss 0.123475 LR 0.000063 Time 0.019895 -2022-12-06 11:47:53,955 - Epoch: [190][ 1030/ 1200] Overall Loss 0.123418 Objective Loss 0.123418 LR 0.000063 Time 0.019886 -2022-12-06 11:47:54,146 - Epoch: [190][ 1040/ 1200] Overall Loss 0.123351 Objective Loss 0.123351 LR 0.000063 Time 0.019878 -2022-12-06 11:47:54,336 - Epoch: [190][ 1050/ 1200] Overall Loss 0.123303 Objective Loss 0.123303 LR 0.000063 Time 0.019869 -2022-12-06 11:47:54,526 - Epoch: [190][ 1060/ 1200] Overall Loss 0.123359 Objective Loss 0.123359 LR 0.000063 Time 0.019860 -2022-12-06 11:47:54,716 - Epoch: [190][ 1070/ 1200] Overall Loss 0.123488 Objective Loss 0.123488 LR 0.000063 Time 0.019852 -2022-12-06 11:47:54,906 - Epoch: [190][ 1080/ 1200] Overall Loss 0.123422 Objective Loss 0.123422 LR 0.000063 Time 0.019843 -2022-12-06 11:47:55,096 - Epoch: [190][ 1090/ 1200] Overall Loss 0.123547 Objective Loss 0.123547 LR 0.000063 Time 0.019835 -2022-12-06 11:47:55,286 - Epoch: [190][ 1100/ 1200] Overall Loss 0.123761 Objective Loss 0.123761 LR 0.000063 Time 0.019827 -2022-12-06 11:47:55,476 - Epoch: [190][ 1110/ 1200] Overall Loss 0.123674 Objective Loss 0.123674 LR 0.000063 Time 0.019819 -2022-12-06 11:47:55,666 - Epoch: [190][ 1120/ 1200] Overall Loss 0.123553 Objective Loss 0.123553 LR 0.000063 Time 0.019812 -2022-12-06 11:47:55,857 - Epoch: [190][ 1130/ 1200] Overall Loss 0.123451 Objective Loss 0.123451 LR 0.000063 Time 0.019805 -2022-12-06 11:47:56,046 - Epoch: [190][ 1140/ 1200] Overall Loss 0.123487 Objective Loss 0.123487 LR 0.000063 Time 0.019797 -2022-12-06 11:47:56,236 - Epoch: [190][ 1150/ 1200] Overall Loss 0.123580 Objective Loss 0.123580 LR 0.000063 Time 0.019789 -2022-12-06 11:47:56,426 - Epoch: [190][ 1160/ 1200] Overall Loss 0.123635 Objective Loss 0.123635 LR 0.000063 Time 0.019781 -2022-12-06 11:47:56,616 - Epoch: [190][ 1170/ 1200] Overall Loss 0.123671 Objective Loss 0.123671 LR 0.000063 Time 0.019774 -2022-12-06 11:47:56,806 - Epoch: [190][ 1180/ 1200] Overall Loss 0.123535 Objective Loss 0.123535 LR 0.000063 Time 0.019767 -2022-12-06 11:47:56,996 - Epoch: [190][ 1190/ 1200] Overall Loss 0.123558 Objective Loss 0.123558 LR 0.000063 Time 0.019761 -2022-12-06 11:47:57,227 - Epoch: [190][ 1200/ 1200] Overall Loss 0.123550 Objective Loss 0.123550 Top1 91.841004 Top5 99.163180 LR 0.000063 Time 0.019788 -2022-12-06 11:47:57,316 - --- validate (epoch=190)----------- -2022-12-06 11:47:57,316 - 34129 samples (256 per mini-batch) -2022-12-06 11:47:57,779 - Epoch: [190][ 10/ 134] Loss 0.198975 Top1 89.023438 Top5 98.750000 -2022-12-06 11:47:57,914 - Epoch: [190][ 20/ 134] Loss 0.184806 Top1 89.160156 Top5 98.847656 -2022-12-06 11:47:58,048 - Epoch: [190][ 30/ 134] Loss 0.209066 Top1 88.750000 Top5 98.763021 -2022-12-06 11:47:58,180 - Epoch: [190][ 40/ 134] Loss 0.204063 Top1 88.828125 Top5 98.750000 -2022-12-06 11:47:58,316 - Epoch: [190][ 50/ 134] Loss 0.210116 Top1 88.968750 Top5 98.687500 -2022-12-06 11:47:58,460 - Epoch: [190][ 60/ 134] Loss 0.211315 Top1 89.062500 Top5 98.710938 -2022-12-06 11:47:58,605 - Epoch: [190][ 70/ 134] Loss 0.212567 Top1 88.978795 Top5 98.705357 -2022-12-06 11:47:58,746 - Epoch: [190][ 80/ 134] Loss 0.216642 Top1 88.818359 Top5 98.715820 -2022-12-06 11:47:58,893 - Epoch: [190][ 90/ 134] Loss 0.219966 Top1 88.806424 Top5 98.676215 -2022-12-06 11:47:59,034 - Epoch: [190][ 100/ 134] Loss 0.218874 Top1 88.785156 Top5 98.710938 -2022-12-06 11:47:59,180 - Epoch: [190][ 110/ 134] Loss 0.219725 Top1 88.774858 Top5 98.721591 -2022-12-06 11:47:59,322 - Epoch: [190][ 120/ 134] Loss 0.220952 Top1 88.766276 Top5 98.697917 -2022-12-06 11:47:59,458 - Epoch: [190][ 130/ 134] Loss 0.221767 Top1 88.756010 Top5 98.704928 -2022-12-06 11:47:59,495 - Epoch: [190][ 134/ 134] Loss 0.221657 Top1 88.754432 Top5 98.699054 -2022-12-06 11:47:59,582 - ==> Top1: 88.754 Top5: 98.699 Loss: 0.222 - -2022-12-06 11:47:59,583 - ==> Confusion: -[[ 930 1 1 3 3 5 1 0 4 34 0 1 1 1 6 1 1 0 1 0 2] - [ 1 948 2 2 7 17 4 8 3 1 1 5 0 1 0 1 5 1 5 4 11] - [ 3 3 1021 11 2 2 13 7 0 3 4 3 2 1 2 2 2 1 3 6 12] - [ 3 0 17 955 1 1 0 0 1 2 9 0 3 1 7 0 1 1 9 0 9] - [ 10 4 1 0 968 2 1 1 1 5 2 2 0 2 6 2 5 3 1 0 4] - [ 0 13 0 4 6 982 1 14 2 1 1 14 2 18 1 2 2 0 0 3 3] - [ 2 1 11 2 0 0 1076 2 1 0 0 3 0 1 0 3 0 3 1 8 4] - [ 1 6 3 2 2 24 7 963 0 1 0 7 0 1 1 1 1 0 14 12 8] - [ 5 2 0 1 0 1 1 0 986 37 8 1 2 4 11 0 1 1 1 1 1] - [ 56 1 1 0 6 2 0 1 24 885 1 0 0 10 4 0 0 2 1 0 7] - [ 0 2 3 2 0 0 1 2 8 1 968 0 0 12 5 1 0 0 4 2 8] - [ 1 0 1 0 0 6 4 1 2 0 0 986 18 7 0 7 2 4 0 5 7] - [ 0 1 2 0 1 2 1 1 0 0 0 26 910 1 0 7 1 7 0 2 7] - [ 1 1 0 0 1 6 0 0 9 10 3 2 2 972 1 3 2 0 1 1 8] - [ 5 3 3 5 2 3 0 0 11 2 1 4 2 3 1075 0 0 1 5 1 4] - [ 0 0 0 2 3 0 1 1 0 1 2 5 1 2 0 1004 6 8 0 4 3] - [ 1 0 2 3 2 0 0 0 1 0 0 2 2 1 0 9 1037 0 0 5 7] - [ 2 0 1 1 2 1 0 0 0 2 0 6 17 2 1 14 0 984 0 0 3] - [ 2 2 6 5 0 3 0 21 2 1 2 4 3 1 8 0 0 2 942 0 4] - [ 2 2 0 1 1 4 3 3 0 1 2 13 7 6 0 3 5 1 1 1018 7] - [ 114 172 142 94 105 125 66 115 79 81 137 85 264 227 128 108 146 72 116 172 10678]] - -2022-12-06 11:48:00,147 - ==> Best [Top1: 88.754 Top5: 98.699 Sparsity:0.00 Params: 5376 on epoch: 190] -2022-12-06 11:48:00,147 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:48:00,154 - - -2022-12-06 11:48:00,154 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:48:01,196 - Epoch: [191][ 10/ 1200] Overall Loss 0.130999 Objective Loss 0.130999 LR 0.000063 Time 0.104147 -2022-12-06 11:48:01,402 - Epoch: [191][ 20/ 1200] Overall Loss 0.127307 Objective Loss 0.127307 LR 0.000063 Time 0.062314 -2022-12-06 11:48:01,600 - Epoch: [191][ 30/ 1200] Overall Loss 0.130482 Objective Loss 0.130482 LR 0.000063 Time 0.048135 -2022-12-06 11:48:01,796 - Epoch: [191][ 40/ 1200] Overall Loss 0.131216 Objective Loss 0.131216 LR 0.000063 Time 0.040974 -2022-12-06 11:48:01,993 - Epoch: [191][ 50/ 1200] Overall Loss 0.130208 Objective Loss 0.130208 LR 0.000063 Time 0.036716 -2022-12-06 11:48:02,188 - Epoch: [191][ 60/ 1200] Overall Loss 0.128893 Objective Loss 0.128893 LR 0.000063 Time 0.033837 -2022-12-06 11:48:02,385 - Epoch: [191][ 70/ 1200] Overall Loss 0.125033 Objective Loss 0.125033 LR 0.000063 Time 0.031812 -2022-12-06 11:48:02,581 - Epoch: [191][ 80/ 1200] Overall Loss 0.124029 Objective Loss 0.124029 LR 0.000063 Time 0.030271 -2022-12-06 11:48:02,778 - Epoch: [191][ 90/ 1200] Overall Loss 0.124219 Objective Loss 0.124219 LR 0.000063 Time 0.029094 -2022-12-06 11:48:02,973 - Epoch: [191][ 100/ 1200] Overall Loss 0.122396 Objective Loss 0.122396 LR 0.000063 Time 0.028133 -2022-12-06 11:48:03,170 - Epoch: [191][ 110/ 1200] Overall Loss 0.122394 Objective Loss 0.122394 LR 0.000063 Time 0.027362 -2022-12-06 11:48:03,366 - Epoch: [191][ 120/ 1200] Overall Loss 0.123248 Objective Loss 0.123248 LR 0.000063 Time 0.026706 -2022-12-06 11:48:03,563 - Epoch: [191][ 130/ 1200] Overall Loss 0.122740 Objective Loss 0.122740 LR 0.000063 Time 0.026164 -2022-12-06 11:48:03,760 - Epoch: [191][ 140/ 1200] Overall Loss 0.124052 Objective Loss 0.124052 LR 0.000063 Time 0.025696 -2022-12-06 11:48:03,957 - Epoch: [191][ 150/ 1200] Overall Loss 0.124413 Objective Loss 0.124413 LR 0.000063 Time 0.025296 -2022-12-06 11:48:04,153 - Epoch: [191][ 160/ 1200] Overall Loss 0.124167 Objective Loss 0.124167 LR 0.000063 Time 0.024935 -2022-12-06 11:48:04,350 - Epoch: [191][ 170/ 1200] Overall Loss 0.123716 Objective Loss 0.123716 LR 0.000063 Time 0.024626 -2022-12-06 11:48:04,546 - Epoch: [191][ 180/ 1200] Overall Loss 0.123102 Objective Loss 0.123102 LR 0.000063 Time 0.024342 -2022-12-06 11:48:04,744 - Epoch: [191][ 190/ 1200] Overall Loss 0.123118 Objective Loss 0.123118 LR 0.000063 Time 0.024099 -2022-12-06 11:48:04,940 - Epoch: [191][ 200/ 1200] Overall Loss 0.123375 Objective Loss 0.123375 LR 0.000063 Time 0.023870 -2022-12-06 11:48:05,137 - Epoch: [191][ 210/ 1200] Overall Loss 0.122793 Objective Loss 0.122793 LR 0.000063 Time 0.023673 -2022-12-06 11:48:05,332 - Epoch: [191][ 220/ 1200] Overall Loss 0.122555 Objective Loss 0.122555 LR 0.000063 Time 0.023480 -2022-12-06 11:48:05,530 - Epoch: [191][ 230/ 1200] Overall Loss 0.122415 Objective Loss 0.122415 LR 0.000063 Time 0.023318 -2022-12-06 11:48:05,726 - Epoch: [191][ 240/ 1200] Overall Loss 0.121476 Objective Loss 0.121476 LR 0.000063 Time 0.023161 -2022-12-06 11:48:05,924 - Epoch: [191][ 250/ 1200] Overall Loss 0.121715 Objective Loss 0.121715 LR 0.000063 Time 0.023022 -2022-12-06 11:48:06,120 - Epoch: [191][ 260/ 1200] Overall Loss 0.121291 Objective Loss 0.121291 LR 0.000063 Time 0.022888 -2022-12-06 11:48:06,318 - Epoch: [191][ 270/ 1200] Overall Loss 0.121159 Objective Loss 0.121159 LR 0.000063 Time 0.022771 -2022-12-06 11:48:06,514 - Epoch: [191][ 280/ 1200] Overall Loss 0.121580 Objective Loss 0.121580 LR 0.000063 Time 0.022657 -2022-12-06 11:48:06,712 - Epoch: [191][ 290/ 1200] Overall Loss 0.121311 Objective Loss 0.121311 LR 0.000063 Time 0.022556 -2022-12-06 11:48:06,908 - Epoch: [191][ 300/ 1200] Overall Loss 0.121448 Objective Loss 0.121448 LR 0.000063 Time 0.022456 -2022-12-06 11:48:07,105 - Epoch: [191][ 310/ 1200] Overall Loss 0.121437 Objective Loss 0.121437 LR 0.000063 Time 0.022366 -2022-12-06 11:48:07,301 - Epoch: [191][ 320/ 1200] Overall Loss 0.121288 Objective Loss 0.121288 LR 0.000063 Time 0.022277 -2022-12-06 11:48:07,498 - Epoch: [191][ 330/ 1200] Overall Loss 0.121095 Objective Loss 0.121095 LR 0.000063 Time 0.022199 -2022-12-06 11:48:07,694 - Epoch: [191][ 340/ 1200] Overall Loss 0.121041 Objective Loss 0.121041 LR 0.000063 Time 0.022120 -2022-12-06 11:48:07,892 - Epoch: [191][ 350/ 1200] Overall Loss 0.120896 Objective Loss 0.120896 LR 0.000063 Time 0.022051 -2022-12-06 11:48:08,087 - Epoch: [191][ 360/ 1200] Overall Loss 0.120998 Objective Loss 0.120998 LR 0.000063 Time 0.021981 -2022-12-06 11:48:08,285 - Epoch: [191][ 370/ 1200] Overall Loss 0.121023 Objective Loss 0.121023 LR 0.000063 Time 0.021920 -2022-12-06 11:48:08,481 - Epoch: [191][ 380/ 1200] Overall Loss 0.120784 Objective Loss 0.120784 LR 0.000063 Time 0.021856 -2022-12-06 11:48:08,678 - Epoch: [191][ 390/ 1200] Overall Loss 0.120787 Objective Loss 0.120787 LR 0.000063 Time 0.021800 -2022-12-06 11:48:08,874 - Epoch: [191][ 400/ 1200] Overall Loss 0.120709 Objective Loss 0.120709 LR 0.000063 Time 0.021743 -2022-12-06 11:48:09,072 - Epoch: [191][ 410/ 1200] Overall Loss 0.120745 Objective Loss 0.120745 LR 0.000063 Time 0.021694 -2022-12-06 11:48:09,268 - Epoch: [191][ 420/ 1200] Overall Loss 0.120783 Objective Loss 0.120783 LR 0.000063 Time 0.021642 -2022-12-06 11:48:09,465 - Epoch: [191][ 430/ 1200] Overall Loss 0.120908 Objective Loss 0.120908 LR 0.000063 Time 0.021596 -2022-12-06 11:48:09,660 - Epoch: [191][ 440/ 1200] Overall Loss 0.120900 Objective Loss 0.120900 LR 0.000063 Time 0.021548 -2022-12-06 11:48:09,858 - Epoch: [191][ 450/ 1200] Overall Loss 0.121288 Objective Loss 0.121288 LR 0.000063 Time 0.021508 -2022-12-06 11:48:10,054 - Epoch: [191][ 460/ 1200] Overall Loss 0.121170 Objective Loss 0.121170 LR 0.000063 Time 0.021465 -2022-12-06 11:48:10,252 - Epoch: [191][ 470/ 1200] Overall Loss 0.121474 Objective Loss 0.121474 LR 0.000063 Time 0.021428 -2022-12-06 11:48:10,448 - Epoch: [191][ 480/ 1200] Overall Loss 0.121422 Objective Loss 0.121422 LR 0.000063 Time 0.021389 -2022-12-06 11:48:10,645 - Epoch: [191][ 490/ 1200] Overall Loss 0.121109 Objective Loss 0.121109 LR 0.000063 Time 0.021354 -2022-12-06 11:48:10,841 - Epoch: [191][ 500/ 1200] Overall Loss 0.121299 Objective Loss 0.121299 LR 0.000063 Time 0.021317 -2022-12-06 11:48:11,038 - Epoch: [191][ 510/ 1200] Overall Loss 0.121576 Objective Loss 0.121576 LR 0.000063 Time 0.021285 -2022-12-06 11:48:11,234 - Epoch: [191][ 520/ 1200] Overall Loss 0.121703 Objective Loss 0.121703 LR 0.000063 Time 0.021251 -2022-12-06 11:48:11,432 - Epoch: [191][ 530/ 1200] Overall Loss 0.121639 Objective Loss 0.121639 LR 0.000063 Time 0.021222 -2022-12-06 11:48:11,628 - Epoch: [191][ 540/ 1200] Overall Loss 0.121979 Objective Loss 0.121979 LR 0.000063 Time 0.021191 -2022-12-06 11:48:11,826 - Epoch: [191][ 550/ 1200] Overall Loss 0.121885 Objective Loss 0.121885 LR 0.000063 Time 0.021165 -2022-12-06 11:48:12,022 - Epoch: [191][ 560/ 1200] Overall Loss 0.121850 Objective Loss 0.121850 LR 0.000063 Time 0.021137 -2022-12-06 11:48:12,220 - Epoch: [191][ 570/ 1200] Overall Loss 0.121638 Objective Loss 0.121638 LR 0.000063 Time 0.021112 -2022-12-06 11:48:12,416 - Epoch: [191][ 580/ 1200] Overall Loss 0.121693 Objective Loss 0.121693 LR 0.000063 Time 0.021085 -2022-12-06 11:48:12,613 - Epoch: [191][ 590/ 1200] Overall Loss 0.121508 Objective Loss 0.121508 LR 0.000063 Time 0.021061 -2022-12-06 11:48:12,809 - Epoch: [191][ 600/ 1200] Overall Loss 0.121548 Objective Loss 0.121548 LR 0.000063 Time 0.021035 -2022-12-06 11:48:13,007 - Epoch: [191][ 610/ 1200] Overall Loss 0.121675 Objective Loss 0.121675 LR 0.000063 Time 0.021013 -2022-12-06 11:48:13,202 - Epoch: [191][ 620/ 1200] Overall Loss 0.121646 Objective Loss 0.121646 LR 0.000063 Time 0.020989 -2022-12-06 11:48:13,400 - Epoch: [191][ 630/ 1200] Overall Loss 0.121625 Objective Loss 0.121625 LR 0.000063 Time 0.020969 -2022-12-06 11:48:13,596 - Epoch: [191][ 640/ 1200] Overall Loss 0.121724 Objective Loss 0.121724 LR 0.000063 Time 0.020946 -2022-12-06 11:48:13,794 - Epoch: [191][ 650/ 1200] Overall Loss 0.121708 Objective Loss 0.121708 LR 0.000063 Time 0.020928 -2022-12-06 11:48:13,990 - Epoch: [191][ 660/ 1200] Overall Loss 0.121572 Objective Loss 0.121572 LR 0.000063 Time 0.020907 -2022-12-06 11:48:14,188 - Epoch: [191][ 670/ 1200] Overall Loss 0.121671 Objective Loss 0.121671 LR 0.000063 Time 0.020889 -2022-12-06 11:48:14,384 - Epoch: [191][ 680/ 1200] Overall Loss 0.121676 Objective Loss 0.121676 LR 0.000063 Time 0.020869 -2022-12-06 11:48:14,581 - Epoch: [191][ 690/ 1200] Overall Loss 0.121499 Objective Loss 0.121499 LR 0.000063 Time 0.020853 -2022-12-06 11:48:14,776 - Epoch: [191][ 700/ 1200] Overall Loss 0.121670 Objective Loss 0.121670 LR 0.000063 Time 0.020833 -2022-12-06 11:48:14,975 - Epoch: [191][ 710/ 1200] Overall Loss 0.121736 Objective Loss 0.121736 LR 0.000063 Time 0.020817 -2022-12-06 11:48:15,171 - Epoch: [191][ 720/ 1200] Overall Loss 0.121686 Objective Loss 0.121686 LR 0.000063 Time 0.020800 -2022-12-06 11:48:15,368 - Epoch: [191][ 730/ 1200] Overall Loss 0.121531 Objective Loss 0.121531 LR 0.000063 Time 0.020784 -2022-12-06 11:48:15,563 - Epoch: [191][ 740/ 1200] Overall Loss 0.121518 Objective Loss 0.121518 LR 0.000063 Time 0.020766 -2022-12-06 11:48:15,760 - Epoch: [191][ 750/ 1200] Overall Loss 0.121552 Objective Loss 0.121552 LR 0.000063 Time 0.020752 -2022-12-06 11:48:15,956 - Epoch: [191][ 760/ 1200] Overall Loss 0.121304 Objective Loss 0.121304 LR 0.000063 Time 0.020736 -2022-12-06 11:48:16,154 - Epoch: [191][ 770/ 1200] Overall Loss 0.121125 Objective Loss 0.121125 LR 0.000063 Time 0.020723 -2022-12-06 11:48:16,350 - Epoch: [191][ 780/ 1200] Overall Loss 0.121139 Objective Loss 0.121139 LR 0.000063 Time 0.020708 -2022-12-06 11:48:16,548 - Epoch: [191][ 790/ 1200] Overall Loss 0.121109 Objective Loss 0.121109 LR 0.000063 Time 0.020695 -2022-12-06 11:48:16,743 - Epoch: [191][ 800/ 1200] Overall Loss 0.120986 Objective Loss 0.120986 LR 0.000063 Time 0.020679 -2022-12-06 11:48:16,940 - Epoch: [191][ 810/ 1200] Overall Loss 0.120969 Objective Loss 0.120969 LR 0.000063 Time 0.020667 -2022-12-06 11:48:17,136 - Epoch: [191][ 820/ 1200] Overall Loss 0.121138 Objective Loss 0.121138 LR 0.000063 Time 0.020653 -2022-12-06 11:48:17,334 - Epoch: [191][ 830/ 1200] Overall Loss 0.121446 Objective Loss 0.121446 LR 0.000063 Time 0.020642 -2022-12-06 11:48:17,529 - Epoch: [191][ 840/ 1200] Overall Loss 0.121479 Objective Loss 0.121479 LR 0.000063 Time 0.020628 -2022-12-06 11:48:17,726 - Epoch: [191][ 850/ 1200] Overall Loss 0.121666 Objective Loss 0.121666 LR 0.000063 Time 0.020617 -2022-12-06 11:48:17,922 - Epoch: [191][ 860/ 1200] Overall Loss 0.121763 Objective Loss 0.121763 LR 0.000063 Time 0.020604 -2022-12-06 11:48:18,119 - Epoch: [191][ 870/ 1200] Overall Loss 0.121847 Objective Loss 0.121847 LR 0.000063 Time 0.020593 -2022-12-06 11:48:18,315 - Epoch: [191][ 880/ 1200] Overall Loss 0.121958 Objective Loss 0.121958 LR 0.000063 Time 0.020581 -2022-12-06 11:48:18,512 - Epoch: [191][ 890/ 1200] Overall Loss 0.121908 Objective Loss 0.121908 LR 0.000063 Time 0.020570 -2022-12-06 11:48:18,707 - Epoch: [191][ 900/ 1200] Overall Loss 0.121760 Objective Loss 0.121760 LR 0.000063 Time 0.020558 -2022-12-06 11:48:18,904 - Epoch: [191][ 910/ 1200] Overall Loss 0.121858 Objective Loss 0.121858 LR 0.000063 Time 0.020548 -2022-12-06 11:48:19,099 - Epoch: [191][ 920/ 1200] Overall Loss 0.121988 Objective Loss 0.121988 LR 0.000063 Time 0.020536 -2022-12-06 11:48:19,297 - Epoch: [191][ 930/ 1200] Overall Loss 0.122020 Objective Loss 0.122020 LR 0.000063 Time 0.020527 -2022-12-06 11:48:19,493 - Epoch: [191][ 940/ 1200] Overall Loss 0.122074 Objective Loss 0.122074 LR 0.000063 Time 0.020517 -2022-12-06 11:48:19,690 - Epoch: [191][ 950/ 1200] Overall Loss 0.122184 Objective Loss 0.122184 LR 0.000063 Time 0.020508 -2022-12-06 11:48:19,885 - Epoch: [191][ 960/ 1200] Overall Loss 0.122352 Objective Loss 0.122352 LR 0.000063 Time 0.020497 -2022-12-06 11:48:20,082 - Epoch: [191][ 970/ 1200] Overall Loss 0.122354 Objective Loss 0.122354 LR 0.000063 Time 0.020488 -2022-12-06 11:48:20,277 - Epoch: [191][ 980/ 1200] Overall Loss 0.122258 Objective Loss 0.122258 LR 0.000063 Time 0.020477 -2022-12-06 11:48:20,474 - Epoch: [191][ 990/ 1200] Overall Loss 0.122327 Objective Loss 0.122327 LR 0.000063 Time 0.020469 -2022-12-06 11:48:20,670 - Epoch: [191][ 1000/ 1200] Overall Loss 0.122488 Objective Loss 0.122488 LR 0.000063 Time 0.020460 -2022-12-06 11:48:20,868 - Epoch: [191][ 1010/ 1200] Overall Loss 0.122540 Objective Loss 0.122540 LR 0.000063 Time 0.020452 -2022-12-06 11:48:21,063 - Epoch: [191][ 1020/ 1200] Overall Loss 0.122616 Objective Loss 0.122616 LR 0.000063 Time 0.020443 -2022-12-06 11:48:21,259 - Epoch: [191][ 1030/ 1200] Overall Loss 0.122626 Objective Loss 0.122626 LR 0.000063 Time 0.020434 -2022-12-06 11:48:21,455 - Epoch: [191][ 1040/ 1200] Overall Loss 0.122663 Objective Loss 0.122663 LR 0.000063 Time 0.020425 -2022-12-06 11:48:21,652 - Epoch: [191][ 1050/ 1200] Overall Loss 0.122557 Objective Loss 0.122557 LR 0.000063 Time 0.020418 -2022-12-06 11:48:21,847 - Epoch: [191][ 1060/ 1200] Overall Loss 0.122562 Objective Loss 0.122562 LR 0.000063 Time 0.020409 -2022-12-06 11:48:22,045 - Epoch: [191][ 1070/ 1200] Overall Loss 0.122533 Objective Loss 0.122533 LR 0.000063 Time 0.020403 -2022-12-06 11:48:22,240 - Epoch: [191][ 1080/ 1200] Overall Loss 0.122483 Objective Loss 0.122483 LR 0.000063 Time 0.020394 -2022-12-06 11:48:22,437 - Epoch: [191][ 1090/ 1200] Overall Loss 0.122346 Objective Loss 0.122346 LR 0.000063 Time 0.020387 -2022-12-06 11:48:22,633 - Epoch: [191][ 1100/ 1200] Overall Loss 0.122284 Objective Loss 0.122284 LR 0.000063 Time 0.020379 -2022-12-06 11:48:22,830 - Epoch: [191][ 1110/ 1200] Overall Loss 0.122206 Objective Loss 0.122206 LR 0.000063 Time 0.020373 -2022-12-06 11:48:23,026 - Epoch: [191][ 1120/ 1200] Overall Loss 0.122207 Objective Loss 0.122207 LR 0.000063 Time 0.020365 -2022-12-06 11:48:23,222 - Epoch: [191][ 1130/ 1200] Overall Loss 0.122264 Objective Loss 0.122264 LR 0.000063 Time 0.020358 -2022-12-06 11:48:23,418 - Epoch: [191][ 1140/ 1200] Overall Loss 0.122382 Objective Loss 0.122382 LR 0.000063 Time 0.020351 -2022-12-06 11:48:23,616 - Epoch: [191][ 1150/ 1200] Overall Loss 0.122390 Objective Loss 0.122390 LR 0.000063 Time 0.020345 -2022-12-06 11:48:23,811 - Epoch: [191][ 1160/ 1200] Overall Loss 0.122251 Objective Loss 0.122251 LR 0.000063 Time 0.020338 -2022-12-06 11:48:24,008 - Epoch: [191][ 1170/ 1200] Overall Loss 0.122256 Objective Loss 0.122256 LR 0.000063 Time 0.020331 -2022-12-06 11:48:24,203 - Epoch: [191][ 1180/ 1200] Overall Loss 0.122240 Objective Loss 0.122240 LR 0.000063 Time 0.020324 -2022-12-06 11:48:24,400 - Epoch: [191][ 1190/ 1200] Overall Loss 0.122211 Objective Loss 0.122211 LR 0.000063 Time 0.020319 -2022-12-06 11:48:24,621 - Epoch: [191][ 1200/ 1200] Overall Loss 0.122351 Objective Loss 0.122351 Top1 91.841004 Top5 99.372385 LR 0.000063 Time 0.020333 -2022-12-06 11:48:24,709 - --- validate (epoch=191)----------- -2022-12-06 11:48:24,709 - 34129 samples (256 per mini-batch) -2022-12-06 11:48:25,163 - Epoch: [191][ 10/ 134] Loss 0.212663 Top1 89.414062 Top5 98.906250 -2022-12-06 11:48:25,291 - Epoch: [191][ 20/ 134] Loss 0.224739 Top1 88.613281 Top5 98.613281 -2022-12-06 11:48:25,422 - Epoch: [191][ 30/ 134] Loss 0.229490 Top1 88.580729 Top5 98.619792 -2022-12-06 11:48:25,550 - Epoch: [191][ 40/ 134] Loss 0.232093 Top1 88.242188 Top5 98.681641 -2022-12-06 11:48:25,680 - Epoch: [191][ 50/ 134] Loss 0.229572 Top1 88.359375 Top5 98.750000 -2022-12-06 11:48:25,811 - Epoch: [191][ 60/ 134] Loss 0.227868 Top1 88.430990 Top5 98.756510 -2022-12-06 11:48:25,939 - Epoch: [191][ 70/ 134] Loss 0.229261 Top1 88.498884 Top5 98.755580 -2022-12-06 11:48:26,066 - Epoch: [191][ 80/ 134] Loss 0.226130 Top1 88.657227 Top5 98.759766 -2022-12-06 11:48:26,197 - Epoch: [191][ 90/ 134] Loss 0.223786 Top1 88.667535 Top5 98.736979 -2022-12-06 11:48:26,328 - Epoch: [191][ 100/ 134] Loss 0.221205 Top1 88.683594 Top5 98.714844 -2022-12-06 11:48:26,458 - Epoch: [191][ 110/ 134] Loss 0.223951 Top1 88.607955 Top5 98.675426 -2022-12-06 11:48:26,589 - Epoch: [191][ 120/ 134] Loss 0.223747 Top1 88.600260 Top5 98.697917 -2022-12-06 11:48:26,723 - Epoch: [191][ 130/ 134] Loss 0.225089 Top1 88.599760 Top5 98.677885 -2022-12-06 11:48:26,761 - Epoch: [191][ 134/ 134] Loss 0.223053 Top1 88.651880 Top5 98.684403 -2022-12-06 11:48:26,850 - ==> Top1: 88.652 Top5: 98.684 Loss: 0.223 - -2022-12-06 11:48:26,851 - ==> Confusion: -[[ 937 1 1 1 1 5 1 0 5 32 0 1 1 2 2 2 1 0 2 0 1] - [ 1 947 2 2 9 18 2 10 3 2 2 4 0 0 0 0 5 2 7 2 9] - [ 2 4 1016 13 3 1 15 8 0 3 3 2 4 1 2 2 2 1 3 5 13] - [ 2 1 12 961 1 2 0 0 1 1 9 0 3 1 9 0 1 1 9 0 6] - [ 9 2 2 0 965 2 1 1 1 5 1 4 0 1 7 4 5 3 0 0 7] - [ 0 8 0 4 7 988 3 13 2 2 1 13 4 13 2 1 1 0 1 1 5] - [ 2 1 8 3 0 0 1082 1 0 0 0 2 0 1 0 4 0 2 2 7 3] - [ 2 8 7 2 2 27 10 950 0 1 0 5 0 2 1 1 1 0 14 10 11] - [ 7 1 0 1 0 1 1 0 988 38 5 1 2 4 9 0 1 0 3 1 1] - [ 54 1 0 0 7 2 0 0 19 896 1 1 0 12 1 1 0 1 0 0 5] - [ 1 2 4 3 0 3 1 1 7 2 966 0 0 10 6 1 0 0 3 1 8] - [ 2 0 3 1 1 9 4 1 1 0 0 984 19 2 1 5 2 6 0 7 3] - [ 0 1 1 3 0 2 2 1 0 0 0 23 910 0 1 8 0 7 1 3 6] - [ 2 0 0 0 1 7 0 1 12 7 2 5 4 967 0 2 2 0 1 1 9] - [ 6 5 3 9 3 3 0 0 13 2 0 3 1 3 1067 0 1 1 5 1 4] - [ 1 0 0 1 3 0 2 1 1 1 1 5 4 2 0 1001 4 10 0 2 4] - [ 3 0 1 1 2 1 1 0 1 0 0 1 2 1 0 10 1039 0 0 3 6] - [ 2 0 1 1 0 0 1 0 0 5 0 7 14 2 3 13 0 983 0 1 3] - [ 3 2 3 6 0 4 0 16 2 1 4 3 3 0 7 1 0 1 947 0 5] - [ 1 3 1 2 0 3 8 4 0 1 2 13 7 5 1 5 2 2 0 1014 6] - [ 118 179 130 97 103 154 74 111 74 71 141 77 279 217 133 102 144 75 125 178 10644]] - -2022-12-06 11:48:27,433 - ==> Best [Top1: 88.754 Top5: 98.699 Sparsity:0.00 Params: 5376 on epoch: 190] -2022-12-06 11:48:27,433 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:48:27,440 - - -2022-12-06 11:48:27,440 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:48:28,385 - Epoch: [192][ 10/ 1200] Overall Loss 0.130604 Objective Loss 0.130604 LR 0.000063 Time 0.094479 -2022-12-06 11:48:28,583 - Epoch: [192][ 20/ 1200] Overall Loss 0.134703 Objective Loss 0.134703 LR 0.000063 Time 0.057066 -2022-12-06 11:48:28,782 - Epoch: [192][ 30/ 1200] Overall Loss 0.132221 Objective Loss 0.132221 LR 0.000063 Time 0.044677 -2022-12-06 11:48:28,977 - Epoch: [192][ 40/ 1200] Overall Loss 0.128280 Objective Loss 0.128280 LR 0.000063 Time 0.038375 -2022-12-06 11:48:29,176 - Epoch: [192][ 50/ 1200] Overall Loss 0.126156 Objective Loss 0.126156 LR 0.000063 Time 0.034672 -2022-12-06 11:48:29,372 - Epoch: [192][ 60/ 1200] Overall Loss 0.125185 Objective Loss 0.125185 LR 0.000063 Time 0.032152 -2022-12-06 11:48:29,571 - Epoch: [192][ 70/ 1200] Overall Loss 0.124275 Objective Loss 0.124275 LR 0.000063 Time 0.030382 -2022-12-06 11:48:29,766 - Epoch: [192][ 80/ 1200] Overall Loss 0.123467 Objective Loss 0.123467 LR 0.000063 Time 0.029024 -2022-12-06 11:48:29,965 - Epoch: [192][ 90/ 1200] Overall Loss 0.123022 Objective Loss 0.123022 LR 0.000063 Time 0.028002 -2022-12-06 11:48:30,161 - Epoch: [192][ 100/ 1200] Overall Loss 0.123740 Objective Loss 0.123740 LR 0.000063 Time 0.027154 -2022-12-06 11:48:30,359 - Epoch: [192][ 110/ 1200] Overall Loss 0.124117 Objective Loss 0.124117 LR 0.000063 Time 0.026481 -2022-12-06 11:48:30,554 - Epoch: [192][ 120/ 1200] Overall Loss 0.124056 Objective Loss 0.124056 LR 0.000063 Time 0.025897 -2022-12-06 11:48:30,753 - Epoch: [192][ 130/ 1200] Overall Loss 0.125425 Objective Loss 0.125425 LR 0.000063 Time 0.025433 -2022-12-06 11:48:30,949 - Epoch: [192][ 140/ 1200] Overall Loss 0.124814 Objective Loss 0.124814 LR 0.000063 Time 0.025009 -2022-12-06 11:48:31,147 - Epoch: [192][ 150/ 1200] Overall Loss 0.124114 Objective Loss 0.124114 LR 0.000063 Time 0.024660 -2022-12-06 11:48:31,342 - Epoch: [192][ 160/ 1200] Overall Loss 0.123728 Objective Loss 0.123728 LR 0.000063 Time 0.024334 -2022-12-06 11:48:31,541 - Epoch: [192][ 170/ 1200] Overall Loss 0.122640 Objective Loss 0.122640 LR 0.000063 Time 0.024071 -2022-12-06 11:48:31,736 - Epoch: [192][ 180/ 1200] Overall Loss 0.122673 Objective Loss 0.122673 LR 0.000063 Time 0.023814 -2022-12-06 11:48:31,935 - Epoch: [192][ 190/ 1200] Overall Loss 0.123387 Objective Loss 0.123387 LR 0.000063 Time 0.023605 -2022-12-06 11:48:32,131 - Epoch: [192][ 200/ 1200] Overall Loss 0.122741 Objective Loss 0.122741 LR 0.000063 Time 0.023400 -2022-12-06 11:48:32,330 - Epoch: [192][ 210/ 1200] Overall Loss 0.122173 Objective Loss 0.122173 LR 0.000063 Time 0.023230 -2022-12-06 11:48:32,525 - Epoch: [192][ 220/ 1200] Overall Loss 0.121806 Objective Loss 0.121806 LR 0.000063 Time 0.023059 -2022-12-06 11:48:32,723 - Epoch: [192][ 230/ 1200] Overall Loss 0.122189 Objective Loss 0.122189 LR 0.000063 Time 0.022917 -2022-12-06 11:48:32,919 - Epoch: [192][ 240/ 1200] Overall Loss 0.121613 Objective Loss 0.121613 LR 0.000063 Time 0.022775 -2022-12-06 11:48:33,117 - Epoch: [192][ 250/ 1200] Overall Loss 0.122079 Objective Loss 0.122079 LR 0.000063 Time 0.022655 -2022-12-06 11:48:33,312 - Epoch: [192][ 260/ 1200] Overall Loss 0.121369 Objective Loss 0.121369 LR 0.000063 Time 0.022531 -2022-12-06 11:48:33,511 - Epoch: [192][ 270/ 1200] Overall Loss 0.121517 Objective Loss 0.121517 LR 0.000063 Time 0.022430 -2022-12-06 11:48:33,706 - Epoch: [192][ 280/ 1200] Overall Loss 0.121420 Objective Loss 0.121420 LR 0.000063 Time 0.022325 -2022-12-06 11:48:33,905 - Epoch: [192][ 290/ 1200] Overall Loss 0.121952 Objective Loss 0.121952 LR 0.000063 Time 0.022239 -2022-12-06 11:48:34,100 - Epoch: [192][ 300/ 1200] Overall Loss 0.121774 Objective Loss 0.121774 LR 0.000063 Time 0.022147 -2022-12-06 11:48:34,299 - Epoch: [192][ 310/ 1200] Overall Loss 0.121889 Objective Loss 0.121889 LR 0.000063 Time 0.022071 -2022-12-06 11:48:34,495 - Epoch: [192][ 320/ 1200] Overall Loss 0.121182 Objective Loss 0.121182 LR 0.000063 Time 0.021991 -2022-12-06 11:48:34,694 - Epoch: [192][ 330/ 1200] Overall Loss 0.120689 Objective Loss 0.120689 LR 0.000063 Time 0.021925 -2022-12-06 11:48:34,888 - Epoch: [192][ 340/ 1200] Overall Loss 0.120443 Objective Loss 0.120443 LR 0.000063 Time 0.021852 -2022-12-06 11:48:35,087 - Epoch: [192][ 350/ 1200] Overall Loss 0.120724 Objective Loss 0.120724 LR 0.000063 Time 0.021794 -2022-12-06 11:48:35,283 - Epoch: [192][ 360/ 1200] Overall Loss 0.120986 Objective Loss 0.120986 LR 0.000063 Time 0.021731 -2022-12-06 11:48:35,482 - Epoch: [192][ 370/ 1200] Overall Loss 0.120624 Objective Loss 0.120624 LR 0.000063 Time 0.021679 -2022-12-06 11:48:35,677 - Epoch: [192][ 380/ 1200] Overall Loss 0.121055 Objective Loss 0.121055 LR 0.000063 Time 0.021621 -2022-12-06 11:48:35,876 - Epoch: [192][ 390/ 1200] Overall Loss 0.120981 Objective Loss 0.120981 LR 0.000063 Time 0.021576 -2022-12-06 11:48:36,072 - Epoch: [192][ 400/ 1200] Overall Loss 0.121113 Objective Loss 0.121113 LR 0.000063 Time 0.021525 -2022-12-06 11:48:36,271 - Epoch: [192][ 410/ 1200] Overall Loss 0.120683 Objective Loss 0.120683 LR 0.000063 Time 0.021483 -2022-12-06 11:48:36,466 - Epoch: [192][ 420/ 1200] Overall Loss 0.120922 Objective Loss 0.120922 LR 0.000063 Time 0.021436 -2022-12-06 11:48:36,665 - Epoch: [192][ 430/ 1200] Overall Loss 0.120771 Objective Loss 0.120771 LR 0.000063 Time 0.021400 -2022-12-06 11:48:36,861 - Epoch: [192][ 440/ 1200] Overall Loss 0.120640 Objective Loss 0.120640 LR 0.000063 Time 0.021355 -2022-12-06 11:48:37,059 - Epoch: [192][ 450/ 1200] Overall Loss 0.120705 Objective Loss 0.120705 LR 0.000063 Time 0.021321 -2022-12-06 11:48:37,254 - Epoch: [192][ 460/ 1200] Overall Loss 0.120824 Objective Loss 0.120824 LR 0.000063 Time 0.021280 -2022-12-06 11:48:37,453 - Epoch: [192][ 470/ 1200] Overall Loss 0.120780 Objective Loss 0.120780 LR 0.000063 Time 0.021250 -2022-12-06 11:48:37,649 - Epoch: [192][ 480/ 1200] Overall Loss 0.120887 Objective Loss 0.120887 LR 0.000063 Time 0.021214 -2022-12-06 11:48:37,847 - Epoch: [192][ 490/ 1200] Overall Loss 0.120827 Objective Loss 0.120827 LR 0.000063 Time 0.021185 -2022-12-06 11:48:38,043 - Epoch: [192][ 500/ 1200] Overall Loss 0.121261 Objective Loss 0.121261 LR 0.000063 Time 0.021150 -2022-12-06 11:48:38,242 - Epoch: [192][ 510/ 1200] Overall Loss 0.120901 Objective Loss 0.120901 LR 0.000063 Time 0.021125 -2022-12-06 11:48:38,437 - Epoch: [192][ 520/ 1200] Overall Loss 0.120805 Objective Loss 0.120805 LR 0.000063 Time 0.021092 -2022-12-06 11:48:38,635 - Epoch: [192][ 530/ 1200] Overall Loss 0.120785 Objective Loss 0.120785 LR 0.000063 Time 0.021068 -2022-12-06 11:48:38,830 - Epoch: [192][ 540/ 1200] Overall Loss 0.120753 Objective Loss 0.120753 LR 0.000063 Time 0.021038 -2022-12-06 11:48:39,029 - Epoch: [192][ 550/ 1200] Overall Loss 0.120612 Objective Loss 0.120612 LR 0.000063 Time 0.021015 -2022-12-06 11:48:39,224 - Epoch: [192][ 560/ 1200] Overall Loss 0.120739 Objective Loss 0.120739 LR 0.000063 Time 0.020988 -2022-12-06 11:48:39,423 - Epoch: [192][ 570/ 1200] Overall Loss 0.120691 Objective Loss 0.120691 LR 0.000063 Time 0.020968 -2022-12-06 11:48:39,618 - Epoch: [192][ 580/ 1200] Overall Loss 0.120838 Objective Loss 0.120838 LR 0.000063 Time 0.020942 -2022-12-06 11:48:39,816 - Epoch: [192][ 590/ 1200] Overall Loss 0.120994 Objective Loss 0.120994 LR 0.000063 Time 0.020922 -2022-12-06 11:48:40,012 - Epoch: [192][ 600/ 1200] Overall Loss 0.120802 Objective Loss 0.120802 LR 0.000063 Time 0.020898 -2022-12-06 11:48:40,211 - Epoch: [192][ 610/ 1200] Overall Loss 0.120533 Objective Loss 0.120533 LR 0.000063 Time 0.020881 -2022-12-06 11:48:40,407 - Epoch: [192][ 620/ 1200] Overall Loss 0.120638 Objective Loss 0.120638 LR 0.000063 Time 0.020859 -2022-12-06 11:48:40,606 - Epoch: [192][ 630/ 1200] Overall Loss 0.120561 Objective Loss 0.120561 LR 0.000063 Time 0.020843 -2022-12-06 11:48:40,801 - Epoch: [192][ 640/ 1200] Overall Loss 0.120692 Objective Loss 0.120692 LR 0.000063 Time 0.020822 -2022-12-06 11:48:41,000 - Epoch: [192][ 650/ 1200] Overall Loss 0.120991 Objective Loss 0.120991 LR 0.000063 Time 0.020807 -2022-12-06 11:48:41,195 - Epoch: [192][ 660/ 1200] Overall Loss 0.121059 Objective Loss 0.121059 LR 0.000063 Time 0.020786 -2022-12-06 11:48:41,394 - Epoch: [192][ 670/ 1200] Overall Loss 0.121101 Objective Loss 0.121101 LR 0.000063 Time 0.020771 -2022-12-06 11:48:41,589 - Epoch: [192][ 680/ 1200] Overall Loss 0.121295 Objective Loss 0.121295 LR 0.000063 Time 0.020752 -2022-12-06 11:48:41,787 - Epoch: [192][ 690/ 1200] Overall Loss 0.121111 Objective Loss 0.121111 LR 0.000063 Time 0.020738 -2022-12-06 11:48:41,982 - Epoch: [192][ 700/ 1200] Overall Loss 0.121127 Objective Loss 0.121127 LR 0.000063 Time 0.020719 -2022-12-06 11:48:42,181 - Epoch: [192][ 710/ 1200] Overall Loss 0.120968 Objective Loss 0.120968 LR 0.000063 Time 0.020706 -2022-12-06 11:48:42,376 - Epoch: [192][ 720/ 1200] Overall Loss 0.121041 Objective Loss 0.121041 LR 0.000063 Time 0.020689 -2022-12-06 11:48:42,575 - Epoch: [192][ 730/ 1200] Overall Loss 0.121127 Objective Loss 0.121127 LR 0.000063 Time 0.020678 -2022-12-06 11:48:42,771 - Epoch: [192][ 740/ 1200] Overall Loss 0.121032 Objective Loss 0.121032 LR 0.000063 Time 0.020663 -2022-12-06 11:48:42,970 - Epoch: [192][ 750/ 1200] Overall Loss 0.121205 Objective Loss 0.121205 LR 0.000063 Time 0.020652 -2022-12-06 11:48:43,166 - Epoch: [192][ 760/ 1200] Overall Loss 0.121315 Objective Loss 0.121315 LR 0.000063 Time 0.020637 -2022-12-06 11:48:43,365 - Epoch: [192][ 770/ 1200] Overall Loss 0.121175 Objective Loss 0.121175 LR 0.000063 Time 0.020626 -2022-12-06 11:48:43,560 - Epoch: [192][ 780/ 1200] Overall Loss 0.121115 Objective Loss 0.121115 LR 0.000063 Time 0.020612 -2022-12-06 11:48:43,759 - Epoch: [192][ 790/ 1200] Overall Loss 0.120870 Objective Loss 0.120870 LR 0.000063 Time 0.020602 -2022-12-06 11:48:43,955 - Epoch: [192][ 800/ 1200] Overall Loss 0.120752 Objective Loss 0.120752 LR 0.000063 Time 0.020589 -2022-12-06 11:48:44,154 - Epoch: [192][ 810/ 1200] Overall Loss 0.120743 Objective Loss 0.120743 LR 0.000063 Time 0.020580 -2022-12-06 11:48:44,350 - Epoch: [192][ 820/ 1200] Overall Loss 0.120828 Objective Loss 0.120828 LR 0.000063 Time 0.020567 -2022-12-06 11:48:44,548 - Epoch: [192][ 830/ 1200] Overall Loss 0.120788 Objective Loss 0.120788 LR 0.000063 Time 0.020558 -2022-12-06 11:48:44,744 - Epoch: [192][ 840/ 1200] Overall Loss 0.120866 Objective Loss 0.120866 LR 0.000063 Time 0.020545 -2022-12-06 11:48:44,943 - Epoch: [192][ 850/ 1200] Overall Loss 0.120894 Objective Loss 0.120894 LR 0.000063 Time 0.020537 -2022-12-06 11:48:45,139 - Epoch: [192][ 860/ 1200] Overall Loss 0.121084 Objective Loss 0.121084 LR 0.000063 Time 0.020525 -2022-12-06 11:48:45,338 - Epoch: [192][ 870/ 1200] Overall Loss 0.120900 Objective Loss 0.120900 LR 0.000063 Time 0.020517 -2022-12-06 11:48:45,533 - Epoch: [192][ 880/ 1200] Overall Loss 0.120906 Objective Loss 0.120906 LR 0.000063 Time 0.020505 -2022-12-06 11:48:45,732 - Epoch: [192][ 890/ 1200] Overall Loss 0.120909 Objective Loss 0.120909 LR 0.000063 Time 0.020497 -2022-12-06 11:48:45,927 - Epoch: [192][ 900/ 1200] Overall Loss 0.121143 Objective Loss 0.121143 LR 0.000063 Time 0.020487 -2022-12-06 11:48:46,126 - Epoch: [192][ 910/ 1200] Overall Loss 0.121175 Objective Loss 0.121175 LR 0.000063 Time 0.020479 -2022-12-06 11:48:46,322 - Epoch: [192][ 920/ 1200] Overall Loss 0.121141 Objective Loss 0.121141 LR 0.000063 Time 0.020469 -2022-12-06 11:48:46,521 - Epoch: [192][ 930/ 1200] Overall Loss 0.121108 Objective Loss 0.121108 LR 0.000063 Time 0.020462 -2022-12-06 11:48:46,716 - Epoch: [192][ 940/ 1200] Overall Loss 0.121367 Objective Loss 0.121367 LR 0.000063 Time 0.020451 -2022-12-06 11:48:46,915 - Epoch: [192][ 950/ 1200] Overall Loss 0.121391 Objective Loss 0.121391 LR 0.000063 Time 0.020445 -2022-12-06 11:48:47,111 - Epoch: [192][ 960/ 1200] Overall Loss 0.121202 Objective Loss 0.121202 LR 0.000063 Time 0.020435 -2022-12-06 11:48:47,310 - Epoch: [192][ 970/ 1200] Overall Loss 0.121140 Objective Loss 0.121140 LR 0.000063 Time 0.020429 -2022-12-06 11:48:47,506 - Epoch: [192][ 980/ 1200] Overall Loss 0.121177 Objective Loss 0.121177 LR 0.000063 Time 0.020421 -2022-12-06 11:48:47,705 - Epoch: [192][ 990/ 1200] Overall Loss 0.121177 Objective Loss 0.121177 LR 0.000063 Time 0.020415 -2022-12-06 11:48:47,901 - Epoch: [192][ 1000/ 1200] Overall Loss 0.121203 Objective Loss 0.121203 LR 0.000063 Time 0.020406 -2022-12-06 11:48:48,100 - Epoch: [192][ 1010/ 1200] Overall Loss 0.121254 Objective Loss 0.121254 LR 0.000063 Time 0.020400 -2022-12-06 11:48:48,295 - Epoch: [192][ 1020/ 1200] Overall Loss 0.121367 Objective Loss 0.121367 LR 0.000063 Time 0.020391 -2022-12-06 11:48:48,494 - Epoch: [192][ 1030/ 1200] Overall Loss 0.121608 Objective Loss 0.121608 LR 0.000063 Time 0.020386 -2022-12-06 11:48:48,690 - Epoch: [192][ 1040/ 1200] Overall Loss 0.121613 Objective Loss 0.121613 LR 0.000063 Time 0.020377 -2022-12-06 11:48:48,888 - Epoch: [192][ 1050/ 1200] Overall Loss 0.121686 Objective Loss 0.121686 LR 0.000063 Time 0.020372 -2022-12-06 11:48:49,084 - Epoch: [192][ 1060/ 1200] Overall Loss 0.121488 Objective Loss 0.121488 LR 0.000063 Time 0.020364 -2022-12-06 11:48:49,282 - Epoch: [192][ 1070/ 1200] Overall Loss 0.121550 Objective Loss 0.121550 LR 0.000063 Time 0.020358 -2022-12-06 11:48:49,477 - Epoch: [192][ 1080/ 1200] Overall Loss 0.121554 Objective Loss 0.121554 LR 0.000063 Time 0.020350 -2022-12-06 11:48:49,676 - Epoch: [192][ 1090/ 1200] Overall Loss 0.121430 Objective Loss 0.121430 LR 0.000063 Time 0.020345 -2022-12-06 11:48:49,871 - Epoch: [192][ 1100/ 1200] Overall Loss 0.121417 Objective Loss 0.121417 LR 0.000063 Time 0.020337 -2022-12-06 11:48:50,070 - Epoch: [192][ 1110/ 1200] Overall Loss 0.121286 Objective Loss 0.121286 LR 0.000063 Time 0.020333 -2022-12-06 11:48:50,266 - Epoch: [192][ 1120/ 1200] Overall Loss 0.121419 Objective Loss 0.121419 LR 0.000063 Time 0.020325 -2022-12-06 11:48:50,464 - Epoch: [192][ 1130/ 1200] Overall Loss 0.121402 Objective Loss 0.121402 LR 0.000063 Time 0.020320 -2022-12-06 11:48:50,659 - Epoch: [192][ 1140/ 1200] Overall Loss 0.121438 Objective Loss 0.121438 LR 0.000063 Time 0.020313 -2022-12-06 11:48:50,858 - Epoch: [192][ 1150/ 1200] Overall Loss 0.121532 Objective Loss 0.121532 LR 0.000063 Time 0.020308 -2022-12-06 11:48:51,053 - Epoch: [192][ 1160/ 1200] Overall Loss 0.121544 Objective Loss 0.121544 LR 0.000063 Time 0.020301 -2022-12-06 11:48:51,252 - Epoch: [192][ 1170/ 1200] Overall Loss 0.121483 Objective Loss 0.121483 LR 0.000063 Time 0.020298 -2022-12-06 11:48:51,447 - Epoch: [192][ 1180/ 1200] Overall Loss 0.121364 Objective Loss 0.121364 LR 0.000063 Time 0.020290 -2022-12-06 11:48:51,646 - Epoch: [192][ 1190/ 1200] Overall Loss 0.121502 Objective Loss 0.121502 LR 0.000063 Time 0.020286 -2022-12-06 11:48:51,869 - Epoch: [192][ 1200/ 1200] Overall Loss 0.121564 Objective Loss 0.121564 Top1 91.213389 Top5 99.163180 LR 0.000063 Time 0.020302 -2022-12-06 11:48:51,957 - --- validate (epoch=192)----------- -2022-12-06 11:48:51,957 - 34129 samples (256 per mini-batch) -2022-12-06 11:48:52,426 - Epoch: [192][ 10/ 134] Loss 0.238102 Top1 88.320312 Top5 98.828125 -2022-12-06 11:48:52,558 - Epoch: [192][ 20/ 134] Loss 0.218367 Top1 88.730469 Top5 98.789062 -2022-12-06 11:48:52,689 - Epoch: [192][ 30/ 134] Loss 0.223541 Top1 88.906250 Top5 98.750000 -2022-12-06 11:48:52,815 - Epoch: [192][ 40/ 134] Loss 0.219952 Top1 88.964844 Top5 98.720703 -2022-12-06 11:48:52,940 - Epoch: [192][ 50/ 134] Loss 0.217359 Top1 89.015625 Top5 98.710938 -2022-12-06 11:48:53,066 - Epoch: [192][ 60/ 134] Loss 0.222553 Top1 88.893229 Top5 98.723958 -2022-12-06 11:48:53,191 - Epoch: [192][ 70/ 134] Loss 0.227587 Top1 88.816964 Top5 98.699777 -2022-12-06 11:48:53,317 - Epoch: [192][ 80/ 134] Loss 0.231631 Top1 88.715820 Top5 98.652344 -2022-12-06 11:48:53,441 - Epoch: [192][ 90/ 134] Loss 0.226763 Top1 88.867188 Top5 98.671875 -2022-12-06 11:48:53,575 - Epoch: [192][ 100/ 134] Loss 0.225203 Top1 88.941406 Top5 98.703125 -2022-12-06 11:48:53,722 - Epoch: [192][ 110/ 134] Loss 0.223983 Top1 89.005682 Top5 98.700284 -2022-12-06 11:48:53,863 - Epoch: [192][ 120/ 134] Loss 0.224072 Top1 89.020182 Top5 98.697917 -2022-12-06 11:48:54,004 - Epoch: [192][ 130/ 134] Loss 0.223448 Top1 89.050481 Top5 98.668870 -2022-12-06 11:48:54,041 - Epoch: [192][ 134/ 134] Loss 0.221651 Top1 89.085528 Top5 98.669753 -2022-12-06 11:48:54,128 - ==> Top1: 89.086 Top5: 98.670 Loss: 0.222 - -2022-12-06 11:48:54,129 - ==> Confusion: -[[ 928 1 1 1 2 5 1 0 5 39 0 1 2 1 2 1 1 1 2 0 2] - [ 1 952 1 3 6 14 3 11 0 1 1 5 1 1 0 1 3 0 8 3 12] - [ 2 1 1015 11 3 2 14 8 0 4 4 4 1 1 2 4 3 1 3 4 16] - [ 3 0 14 957 1 1 1 0 0 1 8 0 4 2 7 0 1 1 11 0 8] - [ 8 7 1 1 961 3 1 1 1 7 1 3 0 1 6 3 6 3 0 0 6] - [ 0 13 0 3 5 976 2 16 2 4 0 13 3 15 1 2 3 0 2 3 6] - [ 2 1 7 2 0 1 1084 2 1 1 0 1 1 0 0 3 0 1 1 8 2] - [ 0 5 3 3 1 24 9 973 0 0 0 4 1 1 0 0 1 0 13 8 8] - [ 6 3 0 0 0 1 2 2 983 38 8 1 1 6 8 0 1 0 2 1 1] - [ 47 1 0 0 5 1 0 2 16 908 1 1 0 11 1 2 0 0 0 0 5] - [ 1 3 4 3 1 0 1 4 6 1 971 0 0 9 4 1 0 0 4 0 6] - [ 3 0 2 0 1 8 6 2 2 0 0 986 15 2 0 6 2 5 0 7 4] - [ 1 1 0 2 0 3 0 1 1 1 0 23 908 1 0 7 1 8 0 3 8] - [ 2 1 0 0 0 7 0 3 10 7 2 3 3 970 0 4 3 0 0 2 6] - [ 8 5 3 8 3 3 0 0 10 1 0 4 2 4 1067 0 1 1 5 0 5] - [ 1 0 1 0 2 0 2 0 0 0 2 6 4 2 0 1000 5 11 1 4 2] - [ 2 2 0 1 1 0 2 0 0 0 0 1 2 1 0 13 1036 1 0 3 7] - [ 3 0 1 1 1 1 0 0 0 3 0 6 13 1 1 15 0 987 0 0 3] - [ 1 3 4 5 0 4 0 20 3 1 2 2 3 0 4 1 0 2 946 1 6] - [ 1 2 0 2 1 5 8 4 0 2 1 14 5 4 0 6 5 1 1 1011 7] - [ 102 180 133 81 91 128 84 123 61 76 137 87 244 224 107 92 118 73 129 173 10783]] - -2022-12-06 11:48:54,800 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:48:54,801 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:48:54,807 - - -2022-12-06 11:48:54,807 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:48:55,739 - Epoch: [193][ 10/ 1200] Overall Loss 0.134299 Objective Loss 0.134299 LR 0.000063 Time 0.093076 -2022-12-06 11:48:55,941 - Epoch: [193][ 20/ 1200] Overall Loss 0.128488 Objective Loss 0.128488 LR 0.000063 Time 0.056613 -2022-12-06 11:48:56,132 - Epoch: [193][ 30/ 1200] Overall Loss 0.126306 Objective Loss 0.126306 LR 0.000063 Time 0.044100 -2022-12-06 11:48:56,323 - Epoch: [193][ 40/ 1200] Overall Loss 0.125019 Objective Loss 0.125019 LR 0.000063 Time 0.037834 -2022-12-06 11:48:56,513 - Epoch: [193][ 50/ 1200] Overall Loss 0.121308 Objective Loss 0.121308 LR 0.000063 Time 0.034056 -2022-12-06 11:48:56,704 - Epoch: [193][ 60/ 1200] Overall Loss 0.123085 Objective Loss 0.123085 LR 0.000063 Time 0.031547 -2022-12-06 11:48:56,894 - Epoch: [193][ 70/ 1200] Overall Loss 0.120132 Objective Loss 0.120132 LR 0.000063 Time 0.029756 -2022-12-06 11:48:57,084 - Epoch: [193][ 80/ 1200] Overall Loss 0.120469 Objective Loss 0.120469 LR 0.000063 Time 0.028403 -2022-12-06 11:48:57,275 - Epoch: [193][ 90/ 1200] Overall Loss 0.122031 Objective Loss 0.122031 LR 0.000063 Time 0.027359 -2022-12-06 11:48:57,465 - Epoch: [193][ 100/ 1200] Overall Loss 0.121798 Objective Loss 0.121798 LR 0.000063 Time 0.026517 -2022-12-06 11:48:57,655 - Epoch: [193][ 110/ 1200] Overall Loss 0.120513 Objective Loss 0.120513 LR 0.000063 Time 0.025832 -2022-12-06 11:48:57,845 - Epoch: [193][ 120/ 1200] Overall Loss 0.120855 Objective Loss 0.120855 LR 0.000063 Time 0.025254 -2022-12-06 11:48:58,035 - Epoch: [193][ 130/ 1200] Overall Loss 0.119512 Objective Loss 0.119512 LR 0.000063 Time 0.024768 -2022-12-06 11:48:58,225 - Epoch: [193][ 140/ 1200] Overall Loss 0.119959 Objective Loss 0.119959 LR 0.000063 Time 0.024353 -2022-12-06 11:48:58,415 - Epoch: [193][ 150/ 1200] Overall Loss 0.120270 Objective Loss 0.120270 LR 0.000063 Time 0.023993 -2022-12-06 11:48:58,605 - Epoch: [193][ 160/ 1200] Overall Loss 0.119594 Objective Loss 0.119594 LR 0.000063 Time 0.023681 -2022-12-06 11:48:58,795 - Epoch: [193][ 170/ 1200] Overall Loss 0.119794 Objective Loss 0.119794 LR 0.000063 Time 0.023402 -2022-12-06 11:48:58,986 - Epoch: [193][ 180/ 1200] Overall Loss 0.120213 Objective Loss 0.120213 LR 0.000063 Time 0.023157 -2022-12-06 11:48:59,176 - Epoch: [193][ 190/ 1200] Overall Loss 0.120134 Objective Loss 0.120134 LR 0.000063 Time 0.022936 -2022-12-06 11:48:59,366 - Epoch: [193][ 200/ 1200] Overall Loss 0.120768 Objective Loss 0.120768 LR 0.000063 Time 0.022736 -2022-12-06 11:48:59,556 - Epoch: [193][ 210/ 1200] Overall Loss 0.120652 Objective Loss 0.120652 LR 0.000063 Time 0.022559 -2022-12-06 11:48:59,747 - Epoch: [193][ 220/ 1200] Overall Loss 0.120777 Objective Loss 0.120777 LR 0.000063 Time 0.022397 -2022-12-06 11:48:59,938 - Epoch: [193][ 230/ 1200] Overall Loss 0.120547 Objective Loss 0.120547 LR 0.000063 Time 0.022250 -2022-12-06 11:49:00,128 - Epoch: [193][ 240/ 1200] Overall Loss 0.120362 Objective Loss 0.120362 LR 0.000063 Time 0.022114 -2022-12-06 11:49:00,318 - Epoch: [193][ 250/ 1200] Overall Loss 0.120278 Objective Loss 0.120278 LR 0.000063 Time 0.021987 -2022-12-06 11:49:00,509 - Epoch: [193][ 260/ 1200] Overall Loss 0.120987 Objective Loss 0.120987 LR 0.000063 Time 0.021872 -2022-12-06 11:49:00,699 - Epoch: [193][ 270/ 1200] Overall Loss 0.120794 Objective Loss 0.120794 LR 0.000063 Time 0.021767 -2022-12-06 11:49:00,890 - Epoch: [193][ 280/ 1200] Overall Loss 0.120449 Objective Loss 0.120449 LR 0.000063 Time 0.021667 -2022-12-06 11:49:01,080 - Epoch: [193][ 290/ 1200] Overall Loss 0.120077 Objective Loss 0.120077 LR 0.000063 Time 0.021574 -2022-12-06 11:49:01,270 - Epoch: [193][ 300/ 1200] Overall Loss 0.119508 Objective Loss 0.119508 LR 0.000063 Time 0.021487 -2022-12-06 11:49:01,460 - Epoch: [193][ 310/ 1200] Overall Loss 0.119424 Objective Loss 0.119424 LR 0.000063 Time 0.021405 -2022-12-06 11:49:01,650 - Epoch: [193][ 320/ 1200] Overall Loss 0.119837 Objective Loss 0.119837 LR 0.000063 Time 0.021329 -2022-12-06 11:49:01,841 - Epoch: [193][ 330/ 1200] Overall Loss 0.119936 Objective Loss 0.119936 LR 0.000063 Time 0.021260 -2022-12-06 11:49:02,032 - Epoch: [193][ 340/ 1200] Overall Loss 0.119101 Objective Loss 0.119101 LR 0.000063 Time 0.021193 -2022-12-06 11:49:02,222 - Epoch: [193][ 350/ 1200] Overall Loss 0.119075 Objective Loss 0.119075 LR 0.000063 Time 0.021130 -2022-12-06 11:49:02,413 - Epoch: [193][ 360/ 1200] Overall Loss 0.119323 Objective Loss 0.119323 LR 0.000063 Time 0.021071 -2022-12-06 11:49:02,603 - Epoch: [193][ 370/ 1200] Overall Loss 0.118996 Objective Loss 0.118996 LR 0.000063 Time 0.021015 -2022-12-06 11:49:02,794 - Epoch: [193][ 380/ 1200] Overall Loss 0.118656 Objective Loss 0.118656 LR 0.000063 Time 0.020962 -2022-12-06 11:49:02,984 - Epoch: [193][ 390/ 1200] Overall Loss 0.118731 Objective Loss 0.118731 LR 0.000063 Time 0.020911 -2022-12-06 11:49:03,174 - Epoch: [193][ 400/ 1200] Overall Loss 0.118680 Objective Loss 0.118680 LR 0.000063 Time 0.020862 -2022-12-06 11:49:03,365 - Epoch: [193][ 410/ 1200] Overall Loss 0.118772 Objective Loss 0.118772 LR 0.000063 Time 0.020818 -2022-12-06 11:49:03,555 - Epoch: [193][ 420/ 1200] Overall Loss 0.118873 Objective Loss 0.118873 LR 0.000063 Time 0.020773 -2022-12-06 11:49:03,745 - Epoch: [193][ 430/ 1200] Overall Loss 0.119072 Objective Loss 0.119072 LR 0.000063 Time 0.020731 -2022-12-06 11:49:03,936 - Epoch: [193][ 440/ 1200] Overall Loss 0.119359 Objective Loss 0.119359 LR 0.000063 Time 0.020691 -2022-12-06 11:49:04,126 - Epoch: [193][ 450/ 1200] Overall Loss 0.119618 Objective Loss 0.119618 LR 0.000063 Time 0.020653 -2022-12-06 11:49:04,316 - Epoch: [193][ 460/ 1200] Overall Loss 0.119678 Objective Loss 0.119678 LR 0.000063 Time 0.020617 -2022-12-06 11:49:04,507 - Epoch: [193][ 470/ 1200] Overall Loss 0.119621 Objective Loss 0.119621 LR 0.000063 Time 0.020582 -2022-12-06 11:49:04,697 - Epoch: [193][ 480/ 1200] Overall Loss 0.119704 Objective Loss 0.119704 LR 0.000063 Time 0.020549 -2022-12-06 11:49:04,887 - Epoch: [193][ 490/ 1200] Overall Loss 0.119747 Objective Loss 0.119747 LR 0.000063 Time 0.020517 -2022-12-06 11:49:05,078 - Epoch: [193][ 500/ 1200] Overall Loss 0.119755 Objective Loss 0.119755 LR 0.000063 Time 0.020485 -2022-12-06 11:49:05,270 - Epoch: [193][ 510/ 1200] Overall Loss 0.119844 Objective Loss 0.119844 LR 0.000063 Time 0.020460 -2022-12-06 11:49:05,462 - Epoch: [193][ 520/ 1200] Overall Loss 0.119923 Objective Loss 0.119923 LR 0.000063 Time 0.020435 -2022-12-06 11:49:05,655 - Epoch: [193][ 530/ 1200] Overall Loss 0.119788 Objective Loss 0.119788 LR 0.000063 Time 0.020412 -2022-12-06 11:49:05,846 - Epoch: [193][ 540/ 1200] Overall Loss 0.120155 Objective Loss 0.120155 LR 0.000063 Time 0.020388 -2022-12-06 11:49:06,039 - Epoch: [193][ 550/ 1200] Overall Loss 0.120096 Objective Loss 0.120096 LR 0.000063 Time 0.020366 -2022-12-06 11:49:06,231 - Epoch: [193][ 560/ 1200] Overall Loss 0.119916 Objective Loss 0.119916 LR 0.000063 Time 0.020345 -2022-12-06 11:49:06,424 - Epoch: [193][ 570/ 1200] Overall Loss 0.119864 Objective Loss 0.119864 LR 0.000063 Time 0.020326 -2022-12-06 11:49:06,616 - Epoch: [193][ 580/ 1200] Overall Loss 0.119721 Objective Loss 0.119721 LR 0.000063 Time 0.020305 -2022-12-06 11:49:06,808 - Epoch: [193][ 590/ 1200] Overall Loss 0.119574 Objective Loss 0.119574 LR 0.000063 Time 0.020286 -2022-12-06 11:49:07,000 - Epoch: [193][ 600/ 1200] Overall Loss 0.119559 Objective Loss 0.119559 LR 0.000063 Time 0.020266 -2022-12-06 11:49:07,192 - Epoch: [193][ 610/ 1200] Overall Loss 0.119911 Objective Loss 0.119911 LR 0.000063 Time 0.020249 -2022-12-06 11:49:07,384 - Epoch: [193][ 620/ 1200] Overall Loss 0.120073 Objective Loss 0.120073 LR 0.000063 Time 0.020230 -2022-12-06 11:49:07,576 - Epoch: [193][ 630/ 1200] Overall Loss 0.120062 Objective Loss 0.120062 LR 0.000063 Time 0.020213 -2022-12-06 11:49:07,768 - Epoch: [193][ 640/ 1200] Overall Loss 0.120027 Objective Loss 0.120027 LR 0.000063 Time 0.020196 -2022-12-06 11:49:07,961 - Epoch: [193][ 650/ 1200] Overall Loss 0.119942 Objective Loss 0.119942 LR 0.000063 Time 0.020181 -2022-12-06 11:49:08,152 - Epoch: [193][ 660/ 1200] Overall Loss 0.119964 Objective Loss 0.119964 LR 0.000063 Time 0.020165 -2022-12-06 11:49:08,345 - Epoch: [193][ 670/ 1200] Overall Loss 0.120048 Objective Loss 0.120048 LR 0.000063 Time 0.020151 -2022-12-06 11:49:08,536 - Epoch: [193][ 680/ 1200] Overall Loss 0.119801 Objective Loss 0.119801 LR 0.000063 Time 0.020135 -2022-12-06 11:49:08,728 - Epoch: [193][ 690/ 1200] Overall Loss 0.119631 Objective Loss 0.119631 LR 0.000063 Time 0.020121 -2022-12-06 11:49:08,920 - Epoch: [193][ 700/ 1200] Overall Loss 0.119930 Objective Loss 0.119930 LR 0.000063 Time 0.020106 -2022-12-06 11:49:09,113 - Epoch: [193][ 710/ 1200] Overall Loss 0.119918 Objective Loss 0.119918 LR 0.000063 Time 0.020094 -2022-12-06 11:49:09,304 - Epoch: [193][ 720/ 1200] Overall Loss 0.120140 Objective Loss 0.120140 LR 0.000063 Time 0.020080 -2022-12-06 11:49:09,497 - Epoch: [193][ 730/ 1200] Overall Loss 0.120221 Objective Loss 0.120221 LR 0.000063 Time 0.020068 -2022-12-06 11:49:09,689 - Epoch: [193][ 740/ 1200] Overall Loss 0.120200 Objective Loss 0.120200 LR 0.000063 Time 0.020055 -2022-12-06 11:49:09,881 - Epoch: [193][ 750/ 1200] Overall Loss 0.120244 Objective Loss 0.120244 LR 0.000063 Time 0.020044 -2022-12-06 11:49:10,074 - Epoch: [193][ 760/ 1200] Overall Loss 0.120421 Objective Loss 0.120421 LR 0.000063 Time 0.020033 -2022-12-06 11:49:10,266 - Epoch: [193][ 770/ 1200] Overall Loss 0.120492 Objective Loss 0.120492 LR 0.000063 Time 0.020022 -2022-12-06 11:49:10,458 - Epoch: [193][ 780/ 1200] Overall Loss 0.120292 Objective Loss 0.120292 LR 0.000063 Time 0.020011 -2022-12-06 11:49:10,651 - Epoch: [193][ 790/ 1200] Overall Loss 0.120337 Objective Loss 0.120337 LR 0.000063 Time 0.020001 -2022-12-06 11:49:10,843 - Epoch: [193][ 800/ 1200] Overall Loss 0.120312 Objective Loss 0.120312 LR 0.000063 Time 0.019989 -2022-12-06 11:49:11,035 - Epoch: [193][ 810/ 1200] Overall Loss 0.120251 Objective Loss 0.120251 LR 0.000063 Time 0.019980 -2022-12-06 11:49:11,227 - Epoch: [193][ 820/ 1200] Overall Loss 0.120150 Objective Loss 0.120150 LR 0.000063 Time 0.019969 -2022-12-06 11:49:11,419 - Epoch: [193][ 830/ 1200] Overall Loss 0.120445 Objective Loss 0.120445 LR 0.000063 Time 0.019959 -2022-12-06 11:49:11,610 - Epoch: [193][ 840/ 1200] Overall Loss 0.120720 Objective Loss 0.120720 LR 0.000063 Time 0.019949 -2022-12-06 11:49:11,803 - Epoch: [193][ 850/ 1200] Overall Loss 0.120736 Objective Loss 0.120736 LR 0.000063 Time 0.019940 -2022-12-06 11:49:11,995 - Epoch: [193][ 860/ 1200] Overall Loss 0.120829 Objective Loss 0.120829 LR 0.000063 Time 0.019931 -2022-12-06 11:49:12,188 - Epoch: [193][ 870/ 1200] Overall Loss 0.120850 Objective Loss 0.120850 LR 0.000063 Time 0.019923 -2022-12-06 11:49:12,379 - Epoch: [193][ 880/ 1200] Overall Loss 0.120769 Objective Loss 0.120769 LR 0.000063 Time 0.019913 -2022-12-06 11:49:12,572 - Epoch: [193][ 890/ 1200] Overall Loss 0.120605 Objective Loss 0.120605 LR 0.000063 Time 0.019906 -2022-12-06 11:49:12,764 - Epoch: [193][ 900/ 1200] Overall Loss 0.120656 Objective Loss 0.120656 LR 0.000063 Time 0.019897 -2022-12-06 11:49:12,956 - Epoch: [193][ 910/ 1200] Overall Loss 0.120577 Objective Loss 0.120577 LR 0.000063 Time 0.019889 -2022-12-06 11:49:13,148 - Epoch: [193][ 920/ 1200] Overall Loss 0.120677 Objective Loss 0.120677 LR 0.000063 Time 0.019881 -2022-12-06 11:49:13,341 - Epoch: [193][ 930/ 1200] Overall Loss 0.120683 Objective Loss 0.120683 LR 0.000063 Time 0.019874 -2022-12-06 11:49:13,533 - Epoch: [193][ 940/ 1200] Overall Loss 0.120732 Objective Loss 0.120732 LR 0.000063 Time 0.019866 -2022-12-06 11:49:13,726 - Epoch: [193][ 950/ 1200] Overall Loss 0.120685 Objective Loss 0.120685 LR 0.000063 Time 0.019860 -2022-12-06 11:49:13,918 - Epoch: [193][ 960/ 1200] Overall Loss 0.120734 Objective Loss 0.120734 LR 0.000063 Time 0.019852 -2022-12-06 11:49:14,111 - Epoch: [193][ 970/ 1200] Overall Loss 0.120776 Objective Loss 0.120776 LR 0.000063 Time 0.019846 -2022-12-06 11:49:14,303 - Epoch: [193][ 980/ 1200] Overall Loss 0.121094 Objective Loss 0.121094 LR 0.000063 Time 0.019839 -2022-12-06 11:49:14,496 - Epoch: [193][ 990/ 1200] Overall Loss 0.121091 Objective Loss 0.121091 LR 0.000063 Time 0.019833 -2022-12-06 11:49:14,687 - Epoch: [193][ 1000/ 1200] Overall Loss 0.121232 Objective Loss 0.121232 LR 0.000063 Time 0.019825 -2022-12-06 11:49:14,881 - Epoch: [193][ 1010/ 1200] Overall Loss 0.121387 Objective Loss 0.121387 LR 0.000063 Time 0.019820 -2022-12-06 11:49:15,072 - Epoch: [193][ 1020/ 1200] Overall Loss 0.121545 Objective Loss 0.121545 LR 0.000063 Time 0.019813 -2022-12-06 11:49:15,265 - Epoch: [193][ 1030/ 1200] Overall Loss 0.121598 Objective Loss 0.121598 LR 0.000063 Time 0.019807 -2022-12-06 11:49:15,458 - Epoch: [193][ 1040/ 1200] Overall Loss 0.121658 Objective Loss 0.121658 LR 0.000063 Time 0.019801 -2022-12-06 11:49:15,651 - Epoch: [193][ 1050/ 1200] Overall Loss 0.121647 Objective Loss 0.121647 LR 0.000063 Time 0.019796 -2022-12-06 11:49:15,843 - Epoch: [193][ 1060/ 1200] Overall Loss 0.121654 Objective Loss 0.121654 LR 0.000063 Time 0.019790 -2022-12-06 11:49:16,035 - Epoch: [193][ 1070/ 1200] Overall Loss 0.121669 Objective Loss 0.121669 LR 0.000063 Time 0.019784 -2022-12-06 11:49:16,227 - Epoch: [193][ 1080/ 1200] Overall Loss 0.121725 Objective Loss 0.121725 LR 0.000063 Time 0.019778 -2022-12-06 11:49:16,420 - Epoch: [193][ 1090/ 1200] Overall Loss 0.121709 Objective Loss 0.121709 LR 0.000063 Time 0.019773 -2022-12-06 11:49:16,611 - Epoch: [193][ 1100/ 1200] Overall Loss 0.121693 Objective Loss 0.121693 LR 0.000063 Time 0.019767 -2022-12-06 11:49:16,804 - Epoch: [193][ 1110/ 1200] Overall Loss 0.121713 Objective Loss 0.121713 LR 0.000063 Time 0.019762 -2022-12-06 11:49:16,996 - Epoch: [193][ 1120/ 1200] Overall Loss 0.121816 Objective Loss 0.121816 LR 0.000063 Time 0.019756 -2022-12-06 11:49:17,189 - Epoch: [193][ 1130/ 1200] Overall Loss 0.121819 Objective Loss 0.121819 LR 0.000063 Time 0.019752 -2022-12-06 11:49:17,380 - Epoch: [193][ 1140/ 1200] Overall Loss 0.121798 Objective Loss 0.121798 LR 0.000063 Time 0.019746 -2022-12-06 11:49:17,573 - Epoch: [193][ 1150/ 1200] Overall Loss 0.121831 Objective Loss 0.121831 LR 0.000063 Time 0.019742 -2022-12-06 11:49:17,765 - Epoch: [193][ 1160/ 1200] Overall Loss 0.121884 Objective Loss 0.121884 LR 0.000063 Time 0.019737 -2022-12-06 11:49:17,957 - Epoch: [193][ 1170/ 1200] Overall Loss 0.121855 Objective Loss 0.121855 LR 0.000063 Time 0.019731 -2022-12-06 11:49:18,149 - Epoch: [193][ 1180/ 1200] Overall Loss 0.121846 Objective Loss 0.121846 LR 0.000063 Time 0.019727 -2022-12-06 11:49:18,342 - Epoch: [193][ 1190/ 1200] Overall Loss 0.121868 Objective Loss 0.121868 LR 0.000063 Time 0.019722 -2022-12-06 11:49:18,571 - Epoch: [193][ 1200/ 1200] Overall Loss 0.121903 Objective Loss 0.121903 Top1 89.748954 Top5 98.326360 LR 0.000063 Time 0.019749 -2022-12-06 11:49:18,659 - --- validate (epoch=193)----------- -2022-12-06 11:49:18,660 - 34129 samples (256 per mini-batch) -2022-12-06 11:49:19,109 - Epoch: [193][ 10/ 134] Loss 0.223948 Top1 88.007812 Top5 98.476562 -2022-12-06 11:49:19,257 - Epoch: [193][ 20/ 134] Loss 0.220521 Top1 88.535156 Top5 98.671875 -2022-12-06 11:49:19,397 - Epoch: [193][ 30/ 134] Loss 0.221142 Top1 88.502604 Top5 98.723958 -2022-12-06 11:49:19,544 - Epoch: [193][ 40/ 134] Loss 0.221166 Top1 88.564453 Top5 98.632812 -2022-12-06 11:49:19,683 - Epoch: [193][ 50/ 134] Loss 0.220278 Top1 88.578125 Top5 98.562500 -2022-12-06 11:49:19,830 - Epoch: [193][ 60/ 134] Loss 0.219531 Top1 88.528646 Top5 98.632812 -2022-12-06 11:49:19,968 - Epoch: [193][ 70/ 134] Loss 0.220252 Top1 88.510045 Top5 98.671875 -2022-12-06 11:49:20,114 - Epoch: [193][ 80/ 134] Loss 0.222967 Top1 88.491211 Top5 98.676758 -2022-12-06 11:49:20,253 - Epoch: [193][ 90/ 134] Loss 0.222892 Top1 88.515625 Top5 98.689236 -2022-12-06 11:49:20,399 - Epoch: [193][ 100/ 134] Loss 0.222023 Top1 88.566406 Top5 98.714844 -2022-12-06 11:49:20,538 - Epoch: [193][ 110/ 134] Loss 0.222573 Top1 88.526278 Top5 98.710938 -2022-12-06 11:49:20,685 - Epoch: [193][ 120/ 134] Loss 0.223758 Top1 88.502604 Top5 98.701172 -2022-12-06 11:49:20,819 - Epoch: [193][ 130/ 134] Loss 0.223566 Top1 88.566707 Top5 98.683894 -2022-12-06 11:49:20,856 - Epoch: [193][ 134/ 134] Loss 0.221961 Top1 88.587418 Top5 98.687333 -2022-12-06 11:49:20,944 - ==> Top1: 88.587 Top5: 98.687 Loss: 0.222 - -2022-12-06 11:49:20,945 - ==> Confusion: -[[ 930 1 1 2 1 6 1 0 3 36 0 1 1 2 4 2 2 0 1 0 2] - [ 3 948 2 2 6 15 4 12 4 0 3 3 0 0 0 1 4 0 8 4 8] - [ 1 4 1022 12 3 1 13 8 0 4 3 4 1 1 2 1 2 1 3 5 12] - [ 1 1 16 953 1 1 0 2 1 2 8 1 4 2 6 0 0 1 13 0 7] - [ 7 4 2 0 963 2 1 2 1 9 2 3 0 2 5 4 7 2 0 0 4] - [ 0 16 0 4 5 987 1 14 4 3 1 11 4 9 2 2 2 0 0 2 2] - [ 0 1 10 2 0 2 1081 2 0 0 0 4 0 2 0 2 0 0 2 8 2] - [ 0 7 2 2 1 23 8 972 0 0 2 4 1 2 0 0 1 0 15 7 7] - [ 1 2 0 1 0 3 1 1 988 39 11 1 1 4 7 0 1 0 2 1 0] - [ 42 2 0 0 3 2 0 0 18 911 1 1 0 10 1 2 0 0 1 0 7] - [ 1 2 3 2 0 0 1 2 7 2 973 0 0 7 5 1 0 0 5 3 5] - [ 2 1 2 0 1 7 5 3 2 0 0 987 18 2 0 5 3 5 0 6 2] - [ 0 1 1 0 0 3 0 1 0 2 0 27 912 1 0 5 2 4 0 3 7] - [ 2 1 0 0 1 7 0 2 13 9 3 5 3 964 1 2 3 0 0 1 6] - [ 7 5 3 7 2 3 0 1 15 2 0 3 1 4 1063 0 0 1 8 1 4] - [ 0 0 1 1 3 0 2 0 1 0 1 5 4 1 0 999 5 11 0 5 4] - [ 1 0 1 2 3 1 1 0 2 0 1 1 2 1 0 7 1039 1 1 3 5] - [ 3 0 1 1 1 1 0 1 0 2 0 7 16 2 1 14 0 981 0 0 5] - [ 1 3 5 4 0 3 0 20 3 1 3 4 3 0 5 0 0 1 947 2 3] - [ 2 3 2 1 1 4 4 5 0 1 1 14 5 3 1 2 6 1 0 1018 6] - [ 99 182 145 79 95 132 80 133 74 90 158 97 268 225 124 98 176 58 130 190 10593]] - -2022-12-06 11:49:21,508 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:49:21,508 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:49:21,514 - - -2022-12-06 11:49:21,514 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:49:22,556 - Epoch: [194][ 10/ 1200] Overall Loss 0.132476 Objective Loss 0.132476 LR 0.000063 Time 0.104113 -2022-12-06 11:49:22,761 - Epoch: [194][ 20/ 1200] Overall Loss 0.123126 Objective Loss 0.123126 LR 0.000063 Time 0.062287 -2022-12-06 11:49:22,962 - Epoch: [194][ 30/ 1200] Overall Loss 0.120048 Objective Loss 0.120048 LR 0.000063 Time 0.048201 -2022-12-06 11:49:23,166 - Epoch: [194][ 40/ 1200] Overall Loss 0.119520 Objective Loss 0.119520 LR 0.000063 Time 0.041250 -2022-12-06 11:49:23,367 - Epoch: [194][ 50/ 1200] Overall Loss 0.124412 Objective Loss 0.124412 LR 0.000063 Time 0.037002 -2022-12-06 11:49:23,572 - Epoch: [194][ 60/ 1200] Overall Loss 0.125536 Objective Loss 0.125536 LR 0.000063 Time 0.034243 -2022-12-06 11:49:23,772 - Epoch: [194][ 70/ 1200] Overall Loss 0.127156 Objective Loss 0.127156 LR 0.000063 Time 0.032203 -2022-12-06 11:49:23,977 - Epoch: [194][ 80/ 1200] Overall Loss 0.125342 Objective Loss 0.125342 LR 0.000063 Time 0.030733 -2022-12-06 11:49:24,178 - Epoch: [194][ 90/ 1200] Overall Loss 0.123289 Objective Loss 0.123289 LR 0.000063 Time 0.029543 -2022-12-06 11:49:24,383 - Epoch: [194][ 100/ 1200] Overall Loss 0.122096 Objective Loss 0.122096 LR 0.000063 Time 0.028632 -2022-12-06 11:49:24,580 - Epoch: [194][ 110/ 1200] Overall Loss 0.122399 Objective Loss 0.122399 LR 0.000063 Time 0.027809 -2022-12-06 11:49:24,771 - Epoch: [194][ 120/ 1200] Overall Loss 0.121663 Objective Loss 0.121663 LR 0.000063 Time 0.027083 -2022-12-06 11:49:24,963 - Epoch: [194][ 130/ 1200] Overall Loss 0.122020 Objective Loss 0.122020 LR 0.000063 Time 0.026469 -2022-12-06 11:49:25,155 - Epoch: [194][ 140/ 1200] Overall Loss 0.121766 Objective Loss 0.121766 LR 0.000063 Time 0.025945 -2022-12-06 11:49:25,345 - Epoch: [194][ 150/ 1200] Overall Loss 0.122031 Objective Loss 0.122031 LR 0.000063 Time 0.025483 -2022-12-06 11:49:25,536 - Epoch: [194][ 160/ 1200] Overall Loss 0.121199 Objective Loss 0.121199 LR 0.000063 Time 0.025077 -2022-12-06 11:49:25,727 - Epoch: [194][ 170/ 1200] Overall Loss 0.121365 Objective Loss 0.121365 LR 0.000063 Time 0.024724 -2022-12-06 11:49:25,918 - Epoch: [194][ 180/ 1200] Overall Loss 0.120383 Objective Loss 0.120383 LR 0.000063 Time 0.024408 -2022-12-06 11:49:26,109 - Epoch: [194][ 190/ 1200] Overall Loss 0.120426 Objective Loss 0.120426 LR 0.000063 Time 0.024128 -2022-12-06 11:49:26,301 - Epoch: [194][ 200/ 1200] Overall Loss 0.121492 Objective Loss 0.121492 LR 0.000063 Time 0.023876 -2022-12-06 11:49:26,491 - Epoch: [194][ 210/ 1200] Overall Loss 0.122213 Objective Loss 0.122213 LR 0.000063 Time 0.023642 -2022-12-06 11:49:26,683 - Epoch: [194][ 220/ 1200] Overall Loss 0.121854 Objective Loss 0.121854 LR 0.000063 Time 0.023438 -2022-12-06 11:49:26,875 - Epoch: [194][ 230/ 1200] Overall Loss 0.121050 Objective Loss 0.121050 LR 0.000063 Time 0.023251 -2022-12-06 11:49:27,066 - Epoch: [194][ 240/ 1200] Overall Loss 0.121290 Objective Loss 0.121290 LR 0.000063 Time 0.023076 -2022-12-06 11:49:27,257 - Epoch: [194][ 250/ 1200] Overall Loss 0.121051 Objective Loss 0.121051 LR 0.000063 Time 0.022917 -2022-12-06 11:49:27,449 - Epoch: [194][ 260/ 1200] Overall Loss 0.120977 Objective Loss 0.120977 LR 0.000063 Time 0.022771 -2022-12-06 11:49:27,640 - Epoch: [194][ 270/ 1200] Overall Loss 0.121126 Objective Loss 0.121126 LR 0.000063 Time 0.022634 -2022-12-06 11:49:27,832 - Epoch: [194][ 280/ 1200] Overall Loss 0.120863 Objective Loss 0.120863 LR 0.000063 Time 0.022507 -2022-12-06 11:49:28,022 - Epoch: [194][ 290/ 1200] Overall Loss 0.120598 Objective Loss 0.120598 LR 0.000063 Time 0.022386 -2022-12-06 11:49:28,214 - Epoch: [194][ 300/ 1200] Overall Loss 0.120707 Objective Loss 0.120707 LR 0.000063 Time 0.022277 -2022-12-06 11:49:28,405 - Epoch: [194][ 310/ 1200] Overall Loss 0.120682 Objective Loss 0.120682 LR 0.000063 Time 0.022174 -2022-12-06 11:49:28,597 - Epoch: [194][ 320/ 1200] Overall Loss 0.120399 Objective Loss 0.120399 LR 0.000063 Time 0.022077 -2022-12-06 11:49:28,788 - Epoch: [194][ 330/ 1200] Overall Loss 0.120036 Objective Loss 0.120036 LR 0.000063 Time 0.021985 -2022-12-06 11:49:28,979 - Epoch: [194][ 340/ 1200] Overall Loss 0.120293 Objective Loss 0.120293 LR 0.000063 Time 0.021900 -2022-12-06 11:49:29,170 - Epoch: [194][ 350/ 1200] Overall Loss 0.120090 Objective Loss 0.120090 LR 0.000063 Time 0.021819 -2022-12-06 11:49:29,361 - Epoch: [194][ 360/ 1200] Overall Loss 0.120975 Objective Loss 0.120975 LR 0.000063 Time 0.021742 -2022-12-06 11:49:29,553 - Epoch: [194][ 370/ 1200] Overall Loss 0.121054 Objective Loss 0.121054 LR 0.000063 Time 0.021670 -2022-12-06 11:49:29,744 - Epoch: [194][ 380/ 1200] Overall Loss 0.121264 Objective Loss 0.121264 LR 0.000063 Time 0.021603 -2022-12-06 11:49:29,935 - Epoch: [194][ 390/ 1200] Overall Loss 0.121252 Objective Loss 0.121252 LR 0.000063 Time 0.021537 -2022-12-06 11:49:30,126 - Epoch: [194][ 400/ 1200] Overall Loss 0.121337 Objective Loss 0.121337 LR 0.000063 Time 0.021475 -2022-12-06 11:49:30,318 - Epoch: [194][ 410/ 1200] Overall Loss 0.120773 Objective Loss 0.120773 LR 0.000063 Time 0.021416 -2022-12-06 11:49:30,509 - Epoch: [194][ 420/ 1200] Overall Loss 0.120634 Objective Loss 0.120634 LR 0.000063 Time 0.021360 -2022-12-06 11:49:30,699 - Epoch: [194][ 430/ 1200] Overall Loss 0.120557 Objective Loss 0.120557 LR 0.000063 Time 0.021306 -2022-12-06 11:49:30,891 - Epoch: [194][ 440/ 1200] Overall Loss 0.120421 Objective Loss 0.120421 LR 0.000063 Time 0.021256 -2022-12-06 11:49:31,082 - Epoch: [194][ 450/ 1200] Overall Loss 0.120426 Objective Loss 0.120426 LR 0.000063 Time 0.021207 -2022-12-06 11:49:31,274 - Epoch: [194][ 460/ 1200] Overall Loss 0.120600 Objective Loss 0.120600 LR 0.000063 Time 0.021161 -2022-12-06 11:49:31,465 - Epoch: [194][ 470/ 1200] Overall Loss 0.120513 Objective Loss 0.120513 LR 0.000063 Time 0.021116 -2022-12-06 11:49:31,656 - Epoch: [194][ 480/ 1200] Overall Loss 0.120559 Objective Loss 0.120559 LR 0.000063 Time 0.021075 -2022-12-06 11:49:31,848 - Epoch: [194][ 490/ 1200] Overall Loss 0.120850 Objective Loss 0.120850 LR 0.000063 Time 0.021034 -2022-12-06 11:49:32,039 - Epoch: [194][ 500/ 1200] Overall Loss 0.120472 Objective Loss 0.120472 LR 0.000063 Time 0.020994 -2022-12-06 11:49:32,230 - Epoch: [194][ 510/ 1200] Overall Loss 0.120376 Objective Loss 0.120376 LR 0.000063 Time 0.020956 -2022-12-06 11:49:32,422 - Epoch: [194][ 520/ 1200] Overall Loss 0.120191 Objective Loss 0.120191 LR 0.000063 Time 0.020921 -2022-12-06 11:49:32,613 - Epoch: [194][ 530/ 1200] Overall Loss 0.120202 Objective Loss 0.120202 LR 0.000063 Time 0.020887 -2022-12-06 11:49:32,805 - Epoch: [194][ 540/ 1200] Overall Loss 0.120236 Objective Loss 0.120236 LR 0.000063 Time 0.020854 -2022-12-06 11:49:32,996 - Epoch: [194][ 550/ 1200] Overall Loss 0.120414 Objective Loss 0.120414 LR 0.000063 Time 0.020821 -2022-12-06 11:49:33,188 - Epoch: [194][ 560/ 1200] Overall Loss 0.120364 Objective Loss 0.120364 LR 0.000063 Time 0.020791 -2022-12-06 11:49:33,379 - Epoch: [194][ 570/ 1200] Overall Loss 0.120493 Objective Loss 0.120493 LR 0.000063 Time 0.020760 -2022-12-06 11:49:33,571 - Epoch: [194][ 580/ 1200] Overall Loss 0.120681 Objective Loss 0.120681 LR 0.000063 Time 0.020732 -2022-12-06 11:49:33,762 - Epoch: [194][ 590/ 1200] Overall Loss 0.120969 Objective Loss 0.120969 LR 0.000063 Time 0.020704 -2022-12-06 11:49:33,953 - Epoch: [194][ 600/ 1200] Overall Loss 0.120604 Objective Loss 0.120604 LR 0.000063 Time 0.020676 -2022-12-06 11:49:34,145 - Epoch: [194][ 610/ 1200] Overall Loss 0.120755 Objective Loss 0.120755 LR 0.000063 Time 0.020651 -2022-12-06 11:49:34,336 - Epoch: [194][ 620/ 1200] Overall Loss 0.120794 Objective Loss 0.120794 LR 0.000063 Time 0.020625 -2022-12-06 11:49:34,527 - Epoch: [194][ 630/ 1200] Overall Loss 0.120983 Objective Loss 0.120983 LR 0.000063 Time 0.020600 -2022-12-06 11:49:34,718 - Epoch: [194][ 640/ 1200] Overall Loss 0.121004 Objective Loss 0.121004 LR 0.000063 Time 0.020577 -2022-12-06 11:49:34,909 - Epoch: [194][ 650/ 1200] Overall Loss 0.121030 Objective Loss 0.121030 LR 0.000063 Time 0.020553 -2022-12-06 11:49:35,101 - Epoch: [194][ 660/ 1200] Overall Loss 0.121080 Objective Loss 0.121080 LR 0.000063 Time 0.020531 -2022-12-06 11:49:35,292 - Epoch: [194][ 670/ 1200] Overall Loss 0.121007 Objective Loss 0.121007 LR 0.000063 Time 0.020509 -2022-12-06 11:49:35,484 - Epoch: [194][ 680/ 1200] Overall Loss 0.121075 Objective Loss 0.121075 LR 0.000063 Time 0.020489 -2022-12-06 11:49:35,675 - Epoch: [194][ 690/ 1200] Overall Loss 0.121358 Objective Loss 0.121358 LR 0.000063 Time 0.020468 -2022-12-06 11:49:35,866 - Epoch: [194][ 700/ 1200] Overall Loss 0.121366 Objective Loss 0.121366 LR 0.000063 Time 0.020448 -2022-12-06 11:49:36,058 - Epoch: [194][ 710/ 1200] Overall Loss 0.121319 Objective Loss 0.121319 LR 0.000063 Time 0.020429 -2022-12-06 11:49:36,248 - Epoch: [194][ 720/ 1200] Overall Loss 0.121413 Objective Loss 0.121413 LR 0.000063 Time 0.020410 -2022-12-06 11:49:36,439 - Epoch: [194][ 730/ 1200] Overall Loss 0.121509 Objective Loss 0.121509 LR 0.000063 Time 0.020390 -2022-12-06 11:49:36,629 - Epoch: [194][ 740/ 1200] Overall Loss 0.121472 Objective Loss 0.121472 LR 0.000063 Time 0.020371 -2022-12-06 11:49:36,819 - Epoch: [194][ 750/ 1200] Overall Loss 0.121484 Objective Loss 0.121484 LR 0.000063 Time 0.020352 -2022-12-06 11:49:37,010 - Epoch: [194][ 760/ 1200] Overall Loss 0.121438 Objective Loss 0.121438 LR 0.000063 Time 0.020335 -2022-12-06 11:49:37,201 - Epoch: [194][ 770/ 1200] Overall Loss 0.121201 Objective Loss 0.121201 LR 0.000063 Time 0.020318 -2022-12-06 11:49:37,393 - Epoch: [194][ 780/ 1200] Overall Loss 0.121165 Objective Loss 0.121165 LR 0.000063 Time 0.020303 -2022-12-06 11:49:37,585 - Epoch: [194][ 790/ 1200] Overall Loss 0.121004 Objective Loss 0.121004 LR 0.000063 Time 0.020288 -2022-12-06 11:49:37,776 - Epoch: [194][ 800/ 1200] Overall Loss 0.121298 Objective Loss 0.121298 LR 0.000063 Time 0.020272 -2022-12-06 11:49:37,966 - Epoch: [194][ 810/ 1200] Overall Loss 0.121116 Objective Loss 0.121116 LR 0.000063 Time 0.020256 -2022-12-06 11:49:38,157 - Epoch: [194][ 820/ 1200] Overall Loss 0.121129 Objective Loss 0.121129 LR 0.000063 Time 0.020241 -2022-12-06 11:49:38,348 - Epoch: [194][ 830/ 1200] Overall Loss 0.121143 Objective Loss 0.121143 LR 0.000063 Time 0.020227 -2022-12-06 11:49:38,539 - Epoch: [194][ 840/ 1200] Overall Loss 0.121263 Objective Loss 0.121263 LR 0.000063 Time 0.020213 -2022-12-06 11:49:38,730 - Epoch: [194][ 850/ 1200] Overall Loss 0.121373 Objective Loss 0.121373 LR 0.000063 Time 0.020199 -2022-12-06 11:49:38,921 - Epoch: [194][ 860/ 1200] Overall Loss 0.121609 Objective Loss 0.121609 LR 0.000063 Time 0.020186 -2022-12-06 11:49:39,111 - Epoch: [194][ 870/ 1200] Overall Loss 0.121575 Objective Loss 0.121575 LR 0.000063 Time 0.020171 -2022-12-06 11:49:39,302 - Epoch: [194][ 880/ 1200] Overall Loss 0.121447 Objective Loss 0.121447 LR 0.000063 Time 0.020159 -2022-12-06 11:49:39,492 - Epoch: [194][ 890/ 1200] Overall Loss 0.121572 Objective Loss 0.121572 LR 0.000063 Time 0.020146 -2022-12-06 11:49:39,683 - Epoch: [194][ 900/ 1200] Overall Loss 0.121479 Objective Loss 0.121479 LR 0.000063 Time 0.020133 -2022-12-06 11:49:39,874 - Epoch: [194][ 910/ 1200] Overall Loss 0.121551 Objective Loss 0.121551 LR 0.000063 Time 0.020121 -2022-12-06 11:49:40,065 - Epoch: [194][ 920/ 1200] Overall Loss 0.121460 Objective Loss 0.121460 LR 0.000063 Time 0.020109 -2022-12-06 11:49:40,255 - Epoch: [194][ 930/ 1200] Overall Loss 0.121476 Objective Loss 0.121476 LR 0.000063 Time 0.020097 -2022-12-06 11:49:40,447 - Epoch: [194][ 940/ 1200] Overall Loss 0.121407 Objective Loss 0.121407 LR 0.000063 Time 0.020086 -2022-12-06 11:49:40,637 - Epoch: [194][ 950/ 1200] Overall Loss 0.121396 Objective Loss 0.121396 LR 0.000063 Time 0.020075 -2022-12-06 11:49:40,828 - Epoch: [194][ 960/ 1200] Overall Loss 0.121436 Objective Loss 0.121436 LR 0.000063 Time 0.020064 -2022-12-06 11:49:41,019 - Epoch: [194][ 970/ 1200] Overall Loss 0.121374 Objective Loss 0.121374 LR 0.000063 Time 0.020053 -2022-12-06 11:49:41,210 - Epoch: [194][ 980/ 1200] Overall Loss 0.121361 Objective Loss 0.121361 LR 0.000063 Time 0.020043 -2022-12-06 11:49:41,400 - Epoch: [194][ 990/ 1200] Overall Loss 0.121301 Objective Loss 0.121301 LR 0.000063 Time 0.020032 -2022-12-06 11:49:41,591 - Epoch: [194][ 1000/ 1200] Overall Loss 0.121252 Objective Loss 0.121252 LR 0.000063 Time 0.020022 -2022-12-06 11:49:41,782 - Epoch: [194][ 1010/ 1200] Overall Loss 0.121168 Objective Loss 0.121168 LR 0.000063 Time 0.020012 -2022-12-06 11:49:41,973 - Epoch: [194][ 1020/ 1200] Overall Loss 0.121293 Objective Loss 0.121293 LR 0.000063 Time 0.020003 -2022-12-06 11:49:42,163 - Epoch: [194][ 1030/ 1200] Overall Loss 0.121268 Objective Loss 0.121268 LR 0.000063 Time 0.019993 -2022-12-06 11:49:42,355 - Epoch: [194][ 1040/ 1200] Overall Loss 0.121276 Objective Loss 0.121276 LR 0.000063 Time 0.019985 -2022-12-06 11:49:42,546 - Epoch: [194][ 1050/ 1200] Overall Loss 0.121331 Objective Loss 0.121331 LR 0.000063 Time 0.019975 -2022-12-06 11:49:42,736 - Epoch: [194][ 1060/ 1200] Overall Loss 0.121365 Objective Loss 0.121365 LR 0.000063 Time 0.019966 -2022-12-06 11:49:42,927 - Epoch: [194][ 1070/ 1200] Overall Loss 0.121345 Objective Loss 0.121345 LR 0.000063 Time 0.019957 -2022-12-06 11:49:43,118 - Epoch: [194][ 1080/ 1200] Overall Loss 0.121146 Objective Loss 0.121146 LR 0.000063 Time 0.019949 -2022-12-06 11:49:43,309 - Epoch: [194][ 1090/ 1200] Overall Loss 0.121198 Objective Loss 0.121198 LR 0.000063 Time 0.019940 -2022-12-06 11:49:43,499 - Epoch: [194][ 1100/ 1200] Overall Loss 0.121145 Objective Loss 0.121145 LR 0.000063 Time 0.019932 -2022-12-06 11:49:43,690 - Epoch: [194][ 1110/ 1200] Overall Loss 0.121303 Objective Loss 0.121303 LR 0.000063 Time 0.019924 -2022-12-06 11:49:43,882 - Epoch: [194][ 1120/ 1200] Overall Loss 0.121321 Objective Loss 0.121321 LR 0.000063 Time 0.019917 -2022-12-06 11:49:44,072 - Epoch: [194][ 1130/ 1200] Overall Loss 0.121295 Objective Loss 0.121295 LR 0.000063 Time 0.019908 -2022-12-06 11:49:44,264 - Epoch: [194][ 1140/ 1200] Overall Loss 0.121129 Objective Loss 0.121129 LR 0.000063 Time 0.019901 -2022-12-06 11:49:44,455 - Epoch: [194][ 1150/ 1200] Overall Loss 0.121055 Objective Loss 0.121055 LR 0.000063 Time 0.019894 -2022-12-06 11:49:44,646 - Epoch: [194][ 1160/ 1200] Overall Loss 0.121125 Objective Loss 0.121125 LR 0.000063 Time 0.019887 -2022-12-06 11:49:44,837 - Epoch: [194][ 1170/ 1200] Overall Loss 0.121020 Objective Loss 0.121020 LR 0.000063 Time 0.019880 -2022-12-06 11:49:45,027 - Epoch: [194][ 1180/ 1200] Overall Loss 0.120965 Objective Loss 0.120965 LR 0.000063 Time 0.019872 -2022-12-06 11:49:45,218 - Epoch: [194][ 1190/ 1200] Overall Loss 0.120985 Objective Loss 0.120985 LR 0.000063 Time 0.019865 -2022-12-06 11:49:45,448 - Epoch: [194][ 1200/ 1200] Overall Loss 0.120928 Objective Loss 0.120928 Top1 92.050209 Top5 99.372385 LR 0.000063 Time 0.019891 -2022-12-06 11:49:45,537 - --- validate (epoch=194)----------- -2022-12-06 11:49:45,537 - 34129 samples (256 per mini-batch) -2022-12-06 11:49:45,982 - Epoch: [194][ 10/ 134] Loss 0.210002 Top1 88.828125 Top5 98.671875 -2022-12-06 11:49:46,109 - Epoch: [194][ 20/ 134] Loss 0.205754 Top1 88.671875 Top5 98.593750 -2022-12-06 11:49:46,237 - Epoch: [194][ 30/ 134] Loss 0.227539 Top1 88.580729 Top5 98.619792 -2022-12-06 11:49:46,366 - Epoch: [194][ 40/ 134] Loss 0.212846 Top1 88.730469 Top5 98.671875 -2022-12-06 11:49:46,496 - Epoch: [194][ 50/ 134] Loss 0.210947 Top1 88.929688 Top5 98.679688 -2022-12-06 11:49:46,621 - Epoch: [194][ 60/ 134] Loss 0.212779 Top1 88.925781 Top5 98.671875 -2022-12-06 11:49:46,747 - Epoch: [194][ 70/ 134] Loss 0.216510 Top1 88.839286 Top5 98.705357 -2022-12-06 11:49:46,876 - Epoch: [194][ 80/ 134] Loss 0.218485 Top1 88.764648 Top5 98.691406 -2022-12-06 11:49:47,008 - Epoch: [194][ 90/ 134] Loss 0.220192 Top1 88.697917 Top5 98.676215 -2022-12-06 11:49:47,141 - Epoch: [194][ 100/ 134] Loss 0.220438 Top1 88.722656 Top5 98.667969 -2022-12-06 11:49:47,273 - Epoch: [194][ 110/ 134] Loss 0.223399 Top1 88.682528 Top5 98.643466 -2022-12-06 11:49:47,408 - Epoch: [194][ 120/ 134] Loss 0.221965 Top1 88.736979 Top5 98.645833 -2022-12-06 11:49:47,542 - Epoch: [194][ 130/ 134] Loss 0.223358 Top1 88.689904 Top5 98.635817 -2022-12-06 11:49:47,579 - Epoch: [194][ 134/ 134] Loss 0.221525 Top1 88.687040 Top5 98.646313 -2022-12-06 11:49:47,666 - ==> Top1: 88.687 Top5: 98.646 Loss: 0.222 - -2022-12-06 11:49:47,667 - ==> Confusion: -[[ 930 1 0 2 5 5 1 1 3 36 0 1 1 1 4 1 2 0 1 0 1] - [ 1 943 1 2 10 19 3 9 4 1 2 4 0 0 0 1 5 1 10 3 8] - [ 2 4 1013 12 6 2 13 10 0 3 6 3 1 2 1 2 1 1 2 6 13] - [ 3 1 11 960 1 2 0 0 1 0 8 0 3 1 8 0 2 2 11 0 6] - [ 8 3 1 0 970 1 1 2 2 4 1 3 0 2 7 3 5 2 0 0 5] - [ 1 10 0 3 5 989 2 14 3 3 0 15 3 11 2 1 2 0 1 2 2] - [ 2 0 4 4 0 2 1082 2 1 1 0 1 0 0 0 4 0 2 2 9 2] - [ 2 7 3 3 1 27 11 965 0 0 1 4 0 2 0 0 1 0 13 6 8] - [ 6 1 0 0 0 1 1 1 981 37 11 2 0 6 10 1 1 1 2 1 1] - [ 49 1 0 0 6 1 0 1 18 899 2 1 0 11 2 1 0 0 1 0 8] - [ 1 2 2 2 1 0 2 1 8 1 968 0 1 12 5 1 0 0 3 3 6] - [ 3 0 2 0 1 5 4 1 2 0 0 987 17 4 1 6 3 5 0 7 3] - [ 0 1 1 0 2 2 0 1 0 1 0 23 915 0 1 8 1 6 0 3 4] - [ 2 1 0 0 1 5 0 1 7 7 4 5 3 975 1 2 2 0 0 1 6] - [ 7 3 2 9 4 3 0 1 10 2 0 2 1 3 1072 0 1 1 4 0 5] - [ 1 0 0 1 2 0 2 0 1 0 1 7 3 3 0 1000 5 10 0 4 3] - [ 0 0 0 2 2 0 1 0 2 0 0 0 3 2 0 13 1037 0 0 3 7] - [ 4 0 1 1 2 1 0 0 0 4 0 3 12 2 1 13 0 988 0 1 3] - [ 1 3 4 9 0 2 0 21 3 1 5 3 3 0 6 0 0 1 942 0 4] - [ 2 3 2 0 2 4 3 3 0 2 3 15 5 6 0 4 2 1 0 1018 5] - [ 105 169 137 86 103 149 74 118 66 66 147 89 278 232 131 105 145 78 130 186 10632]] - -2022-12-06 11:49:48,246 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:49:48,246 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:49:48,252 - - -2022-12-06 11:49:48,252 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:49:49,189 - Epoch: [195][ 10/ 1200] Overall Loss 0.141591 Objective Loss 0.141591 LR 0.000031 Time 0.093683 -2022-12-06 11:49:49,388 - Epoch: [195][ 20/ 1200] Overall Loss 0.130157 Objective Loss 0.130157 LR 0.000031 Time 0.056736 -2022-12-06 11:49:49,586 - Epoch: [195][ 30/ 1200] Overall Loss 0.122060 Objective Loss 0.122060 LR 0.000031 Time 0.044410 -2022-12-06 11:49:49,781 - Epoch: [195][ 40/ 1200] Overall Loss 0.123974 Objective Loss 0.123974 LR 0.000031 Time 0.038172 -2022-12-06 11:49:49,979 - Epoch: [195][ 50/ 1200] Overall Loss 0.123831 Objective Loss 0.123831 LR 0.000031 Time 0.034494 -2022-12-06 11:49:50,175 - Epoch: [195][ 60/ 1200] Overall Loss 0.122388 Objective Loss 0.122388 LR 0.000031 Time 0.031999 -2022-12-06 11:49:50,375 - Epoch: [195][ 70/ 1200] Overall Loss 0.123061 Objective Loss 0.123061 LR 0.000031 Time 0.030266 -2022-12-06 11:49:50,570 - Epoch: [195][ 80/ 1200] Overall Loss 0.120805 Objective Loss 0.120805 LR 0.000031 Time 0.028923 -2022-12-06 11:49:50,770 - Epoch: [195][ 90/ 1200] Overall Loss 0.121793 Objective Loss 0.121793 LR 0.000031 Time 0.027926 -2022-12-06 11:49:50,966 - Epoch: [195][ 100/ 1200] Overall Loss 0.121317 Objective Loss 0.121317 LR 0.000031 Time 0.027083 -2022-12-06 11:49:51,165 - Epoch: [195][ 110/ 1200] Overall Loss 0.122108 Objective Loss 0.122108 LR 0.000031 Time 0.026429 -2022-12-06 11:49:51,361 - Epoch: [195][ 120/ 1200] Overall Loss 0.122367 Objective Loss 0.122367 LR 0.000031 Time 0.025855 -2022-12-06 11:49:51,561 - Epoch: [195][ 130/ 1200] Overall Loss 0.122935 Objective Loss 0.122935 LR 0.000031 Time 0.025395 -2022-12-06 11:49:51,756 - Epoch: [195][ 140/ 1200] Overall Loss 0.124010 Objective Loss 0.124010 LR 0.000031 Time 0.024970 -2022-12-06 11:49:51,955 - Epoch: [195][ 150/ 1200] Overall Loss 0.124888 Objective Loss 0.124888 LR 0.000031 Time 0.024629 -2022-12-06 11:49:52,151 - Epoch: [195][ 160/ 1200] Overall Loss 0.124361 Objective Loss 0.124361 LR 0.000031 Time 0.024314 -2022-12-06 11:49:52,350 - Epoch: [195][ 170/ 1200] Overall Loss 0.123808 Objective Loss 0.123808 LR 0.000031 Time 0.024053 -2022-12-06 11:49:52,546 - Epoch: [195][ 180/ 1200] Overall Loss 0.123414 Objective Loss 0.123414 LR 0.000031 Time 0.023802 -2022-12-06 11:49:52,746 - Epoch: [195][ 190/ 1200] Overall Loss 0.122176 Objective Loss 0.122176 LR 0.000031 Time 0.023597 -2022-12-06 11:49:52,942 - Epoch: [195][ 200/ 1200] Overall Loss 0.122231 Objective Loss 0.122231 LR 0.000031 Time 0.023395 -2022-12-06 11:49:53,140 - Epoch: [195][ 210/ 1200] Overall Loss 0.121909 Objective Loss 0.121909 LR 0.000031 Time 0.023222 -2022-12-06 11:49:53,336 - Epoch: [195][ 220/ 1200] Overall Loss 0.121455 Objective Loss 0.121455 LR 0.000031 Time 0.023051 -2022-12-06 11:49:53,534 - Epoch: [195][ 230/ 1200] Overall Loss 0.121012 Objective Loss 0.121012 LR 0.000031 Time 0.022911 -2022-12-06 11:49:53,730 - Epoch: [195][ 240/ 1200] Overall Loss 0.121649 Objective Loss 0.121649 LR 0.000031 Time 0.022770 -2022-12-06 11:49:53,928 - Epoch: [195][ 250/ 1200] Overall Loss 0.121579 Objective Loss 0.121579 LR 0.000031 Time 0.022650 -2022-12-06 11:49:54,125 - Epoch: [195][ 260/ 1200] Overall Loss 0.121574 Objective Loss 0.121574 LR 0.000031 Time 0.022534 -2022-12-06 11:49:54,323 - Epoch: [195][ 270/ 1200] Overall Loss 0.121295 Objective Loss 0.121295 LR 0.000031 Time 0.022429 -2022-12-06 11:49:54,518 - Epoch: [195][ 280/ 1200] Overall Loss 0.121614 Objective Loss 0.121614 LR 0.000031 Time 0.022323 -2022-12-06 11:49:54,716 - Epoch: [195][ 290/ 1200] Overall Loss 0.121383 Objective Loss 0.121383 LR 0.000031 Time 0.022235 -2022-12-06 11:49:54,911 - Epoch: [195][ 300/ 1200] Overall Loss 0.120709 Objective Loss 0.120709 LR 0.000031 Time 0.022143 -2022-12-06 11:49:55,111 - Epoch: [195][ 310/ 1200] Overall Loss 0.120816 Objective Loss 0.120816 LR 0.000031 Time 0.022069 -2022-12-06 11:49:55,306 - Epoch: [195][ 320/ 1200] Overall Loss 0.120873 Objective Loss 0.120873 LR 0.000031 Time 0.021988 -2022-12-06 11:49:55,505 - Epoch: [195][ 330/ 1200] Overall Loss 0.120382 Objective Loss 0.120382 LR 0.000031 Time 0.021923 -2022-12-06 11:49:55,700 - Epoch: [195][ 340/ 1200] Overall Loss 0.120839 Objective Loss 0.120839 LR 0.000031 Time 0.021852 -2022-12-06 11:49:55,900 - Epoch: [195][ 350/ 1200] Overall Loss 0.121299 Objective Loss 0.121299 LR 0.000031 Time 0.021795 -2022-12-06 11:49:56,095 - Epoch: [195][ 360/ 1200] Overall Loss 0.121473 Objective Loss 0.121473 LR 0.000031 Time 0.021731 -2022-12-06 11:49:56,294 - Epoch: [195][ 370/ 1200] Overall Loss 0.120810 Objective Loss 0.120810 LR 0.000031 Time 0.021680 -2022-12-06 11:49:56,489 - Epoch: [195][ 380/ 1200] Overall Loss 0.120822 Objective Loss 0.120822 LR 0.000031 Time 0.021622 -2022-12-06 11:49:56,688 - Epoch: [195][ 390/ 1200] Overall Loss 0.120819 Objective Loss 0.120819 LR 0.000031 Time 0.021575 -2022-12-06 11:49:56,884 - Epoch: [195][ 400/ 1200] Overall Loss 0.120961 Objective Loss 0.120961 LR 0.000031 Time 0.021524 -2022-12-06 11:49:57,084 - Epoch: [195][ 410/ 1200] Overall Loss 0.121214 Objective Loss 0.121214 LR 0.000031 Time 0.021485 -2022-12-06 11:49:57,280 - Epoch: [195][ 420/ 1200] Overall Loss 0.120905 Objective Loss 0.120905 LR 0.000031 Time 0.021439 -2022-12-06 11:49:57,479 - Epoch: [195][ 430/ 1200] Overall Loss 0.121298 Objective Loss 0.121298 LR 0.000031 Time 0.021402 -2022-12-06 11:49:57,674 - Epoch: [195][ 440/ 1200] Overall Loss 0.121239 Objective Loss 0.121239 LR 0.000031 Time 0.021359 -2022-12-06 11:49:57,873 - Epoch: [195][ 450/ 1200] Overall Loss 0.121228 Objective Loss 0.121228 LR 0.000031 Time 0.021325 -2022-12-06 11:49:58,069 - Epoch: [195][ 460/ 1200] Overall Loss 0.121018 Objective Loss 0.121018 LR 0.000031 Time 0.021286 -2022-12-06 11:49:58,268 - Epoch: [195][ 470/ 1200] Overall Loss 0.120858 Objective Loss 0.120858 LR 0.000031 Time 0.021257 -2022-12-06 11:49:58,464 - Epoch: [195][ 480/ 1200] Overall Loss 0.120669 Objective Loss 0.120669 LR 0.000031 Time 0.021219 -2022-12-06 11:49:58,663 - Epoch: [195][ 490/ 1200] Overall Loss 0.120806 Objective Loss 0.120806 LR 0.000031 Time 0.021192 -2022-12-06 11:49:58,858 - Epoch: [195][ 500/ 1200] Overall Loss 0.120704 Objective Loss 0.120704 LR 0.000031 Time 0.021157 -2022-12-06 11:49:59,057 - Epoch: [195][ 510/ 1200] Overall Loss 0.120708 Objective Loss 0.120708 LR 0.000031 Time 0.021132 -2022-12-06 11:49:59,254 - Epoch: [195][ 520/ 1200] Overall Loss 0.120629 Objective Loss 0.120629 LR 0.000031 Time 0.021103 -2022-12-06 11:49:59,454 - Epoch: [195][ 530/ 1200] Overall Loss 0.120576 Objective Loss 0.120576 LR 0.000031 Time 0.021081 -2022-12-06 11:49:59,650 - Epoch: [195][ 540/ 1200] Overall Loss 0.120609 Objective Loss 0.120609 LR 0.000031 Time 0.021052 -2022-12-06 11:49:59,849 - Epoch: [195][ 550/ 1200] Overall Loss 0.120549 Objective Loss 0.120549 LR 0.000031 Time 0.021030 -2022-12-06 11:50:00,045 - Epoch: [195][ 560/ 1200] Overall Loss 0.120487 Objective Loss 0.120487 LR 0.000031 Time 0.021004 -2022-12-06 11:50:00,245 - Epoch: [195][ 570/ 1200] Overall Loss 0.120589 Objective Loss 0.120589 LR 0.000031 Time 0.020985 -2022-12-06 11:50:00,441 - Epoch: [195][ 580/ 1200] Overall Loss 0.120634 Objective Loss 0.120634 LR 0.000031 Time 0.020960 -2022-12-06 11:50:00,640 - Epoch: [195][ 590/ 1200] Overall Loss 0.120568 Objective Loss 0.120568 LR 0.000031 Time 0.020942 -2022-12-06 11:50:00,835 - Epoch: [195][ 600/ 1200] Overall Loss 0.120399 Objective Loss 0.120399 LR 0.000031 Time 0.020918 -2022-12-06 11:50:01,035 - Epoch: [195][ 610/ 1200] Overall Loss 0.120127 Objective Loss 0.120127 LR 0.000031 Time 0.020901 -2022-12-06 11:50:01,231 - Epoch: [195][ 620/ 1200] Overall Loss 0.120130 Objective Loss 0.120130 LR 0.000031 Time 0.020879 -2022-12-06 11:50:01,431 - Epoch: [195][ 630/ 1200] Overall Loss 0.120499 Objective Loss 0.120499 LR 0.000031 Time 0.020863 -2022-12-06 11:50:01,627 - Epoch: [195][ 640/ 1200] Overall Loss 0.120398 Objective Loss 0.120398 LR 0.000031 Time 0.020843 -2022-12-06 11:50:01,826 - Epoch: [195][ 650/ 1200] Overall Loss 0.120263 Objective Loss 0.120263 LR 0.000031 Time 0.020828 -2022-12-06 11:50:02,022 - Epoch: [195][ 660/ 1200] Overall Loss 0.120288 Objective Loss 0.120288 LR 0.000031 Time 0.020809 -2022-12-06 11:50:02,222 - Epoch: [195][ 670/ 1200] Overall Loss 0.120279 Objective Loss 0.120279 LR 0.000031 Time 0.020795 -2022-12-06 11:50:02,417 - Epoch: [195][ 680/ 1200] Overall Loss 0.120278 Objective Loss 0.120278 LR 0.000031 Time 0.020777 -2022-12-06 11:50:02,616 - Epoch: [195][ 690/ 1200] Overall Loss 0.120194 Objective Loss 0.120194 LR 0.000031 Time 0.020763 -2022-12-06 11:50:02,812 - Epoch: [195][ 700/ 1200] Overall Loss 0.120333 Objective Loss 0.120333 LR 0.000031 Time 0.020745 -2022-12-06 11:50:03,011 - Epoch: [195][ 710/ 1200] Overall Loss 0.120353 Objective Loss 0.120353 LR 0.000031 Time 0.020733 -2022-12-06 11:50:03,207 - Epoch: [195][ 720/ 1200] Overall Loss 0.120418 Objective Loss 0.120418 LR 0.000031 Time 0.020715 -2022-12-06 11:50:03,406 - Epoch: [195][ 730/ 1200] Overall Loss 0.120249 Objective Loss 0.120249 LR 0.000031 Time 0.020704 -2022-12-06 11:50:03,602 - Epoch: [195][ 740/ 1200] Overall Loss 0.120406 Objective Loss 0.120406 LR 0.000031 Time 0.020688 -2022-12-06 11:50:03,801 - Epoch: [195][ 750/ 1200] Overall Loss 0.120502 Objective Loss 0.120502 LR 0.000031 Time 0.020677 -2022-12-06 11:50:03,997 - Epoch: [195][ 760/ 1200] Overall Loss 0.120405 Objective Loss 0.120405 LR 0.000031 Time 0.020662 -2022-12-06 11:50:04,197 - Epoch: [195][ 770/ 1200] Overall Loss 0.120231 Objective Loss 0.120231 LR 0.000031 Time 0.020652 -2022-12-06 11:50:04,393 - Epoch: [195][ 780/ 1200] Overall Loss 0.120276 Objective Loss 0.120276 LR 0.000031 Time 0.020638 -2022-12-06 11:50:04,591 - Epoch: [195][ 790/ 1200] Overall Loss 0.120176 Objective Loss 0.120176 LR 0.000031 Time 0.020628 -2022-12-06 11:50:04,789 - Epoch: [195][ 800/ 1200] Overall Loss 0.120260 Objective Loss 0.120260 LR 0.000031 Time 0.020617 -2022-12-06 11:50:04,991 - Epoch: [195][ 810/ 1200] Overall Loss 0.120216 Objective Loss 0.120216 LR 0.000031 Time 0.020610 -2022-12-06 11:50:05,190 - Epoch: [195][ 820/ 1200] Overall Loss 0.120110 Objective Loss 0.120110 LR 0.000031 Time 0.020601 -2022-12-06 11:50:05,393 - Epoch: [195][ 830/ 1200] Overall Loss 0.120168 Objective Loss 0.120168 LR 0.000031 Time 0.020596 -2022-12-06 11:50:05,592 - Epoch: [195][ 840/ 1200] Overall Loss 0.120143 Objective Loss 0.120143 LR 0.000031 Time 0.020587 -2022-12-06 11:50:05,794 - Epoch: [195][ 850/ 1200] Overall Loss 0.120090 Objective Loss 0.120090 LR 0.000031 Time 0.020582 -2022-12-06 11:50:05,992 - Epoch: [195][ 860/ 1200] Overall Loss 0.120133 Objective Loss 0.120133 LR 0.000031 Time 0.020573 -2022-12-06 11:50:06,194 - Epoch: [195][ 870/ 1200] Overall Loss 0.120089 Objective Loss 0.120089 LR 0.000031 Time 0.020568 -2022-12-06 11:50:06,392 - Epoch: [195][ 880/ 1200] Overall Loss 0.120055 Objective Loss 0.120055 LR 0.000031 Time 0.020559 -2022-12-06 11:50:06,593 - Epoch: [195][ 890/ 1200] Overall Loss 0.120139 Objective Loss 0.120139 LR 0.000031 Time 0.020553 -2022-12-06 11:50:06,792 - Epoch: [195][ 900/ 1200] Overall Loss 0.120188 Objective Loss 0.120188 LR 0.000031 Time 0.020544 -2022-12-06 11:50:06,994 - Epoch: [195][ 910/ 1200] Overall Loss 0.120151 Objective Loss 0.120151 LR 0.000031 Time 0.020540 -2022-12-06 11:50:07,192 - Epoch: [195][ 920/ 1200] Overall Loss 0.120204 Objective Loss 0.120204 LR 0.000031 Time 0.020531 -2022-12-06 11:50:07,394 - Epoch: [195][ 930/ 1200] Overall Loss 0.120192 Objective Loss 0.120192 LR 0.000031 Time 0.020527 -2022-12-06 11:50:07,592 - Epoch: [195][ 940/ 1200] Overall Loss 0.120312 Objective Loss 0.120312 LR 0.000031 Time 0.020519 -2022-12-06 11:50:07,793 - Epoch: [195][ 950/ 1200] Overall Loss 0.120331 Objective Loss 0.120331 LR 0.000031 Time 0.020514 -2022-12-06 11:50:07,992 - Epoch: [195][ 960/ 1200] Overall Loss 0.120130 Objective Loss 0.120130 LR 0.000031 Time 0.020507 -2022-12-06 11:50:08,194 - Epoch: [195][ 970/ 1200] Overall Loss 0.120060 Objective Loss 0.120060 LR 0.000031 Time 0.020503 -2022-12-06 11:50:08,392 - Epoch: [195][ 980/ 1200] Overall Loss 0.120212 Objective Loss 0.120212 LR 0.000031 Time 0.020495 -2022-12-06 11:50:08,594 - Epoch: [195][ 990/ 1200] Overall Loss 0.120293 Objective Loss 0.120293 LR 0.000031 Time 0.020492 -2022-12-06 11:50:08,792 - Epoch: [195][ 1000/ 1200] Overall Loss 0.120185 Objective Loss 0.120185 LR 0.000031 Time 0.020485 -2022-12-06 11:50:08,994 - Epoch: [195][ 1010/ 1200] Overall Loss 0.120180 Objective Loss 0.120180 LR 0.000031 Time 0.020481 -2022-12-06 11:50:09,192 - Epoch: [195][ 1020/ 1200] Overall Loss 0.120101 Objective Loss 0.120101 LR 0.000031 Time 0.020474 -2022-12-06 11:50:09,394 - Epoch: [195][ 1030/ 1200] Overall Loss 0.120001 Objective Loss 0.120001 LR 0.000031 Time 0.020471 -2022-12-06 11:50:09,593 - Epoch: [195][ 1040/ 1200] Overall Loss 0.120106 Objective Loss 0.120106 LR 0.000031 Time 0.020464 -2022-12-06 11:50:09,794 - Epoch: [195][ 1050/ 1200] Overall Loss 0.119952 Objective Loss 0.119952 LR 0.000031 Time 0.020461 -2022-12-06 11:50:09,993 - Epoch: [195][ 1060/ 1200] Overall Loss 0.119879 Objective Loss 0.119879 LR 0.000031 Time 0.020455 -2022-12-06 11:50:10,195 - Epoch: [195][ 1070/ 1200] Overall Loss 0.119954 Objective Loss 0.119954 LR 0.000031 Time 0.020451 -2022-12-06 11:50:10,393 - Epoch: [195][ 1080/ 1200] Overall Loss 0.120097 Objective Loss 0.120097 LR 0.000031 Time 0.020445 -2022-12-06 11:50:10,594 - Epoch: [195][ 1090/ 1200] Overall Loss 0.120001 Objective Loss 0.120001 LR 0.000031 Time 0.020442 -2022-12-06 11:50:10,793 - Epoch: [195][ 1100/ 1200] Overall Loss 0.120015 Objective Loss 0.120015 LR 0.000031 Time 0.020436 -2022-12-06 11:50:10,994 - Epoch: [195][ 1110/ 1200] Overall Loss 0.120062 Objective Loss 0.120062 LR 0.000031 Time 0.020433 -2022-12-06 11:50:11,193 - Epoch: [195][ 1120/ 1200] Overall Loss 0.120036 Objective Loss 0.120036 LR 0.000031 Time 0.020427 -2022-12-06 11:50:11,394 - Epoch: [195][ 1130/ 1200] Overall Loss 0.120035 Objective Loss 0.120035 LR 0.000031 Time 0.020424 -2022-12-06 11:50:11,593 - Epoch: [195][ 1140/ 1200] Overall Loss 0.120140 Objective Loss 0.120140 LR 0.000031 Time 0.020419 -2022-12-06 11:50:11,795 - Epoch: [195][ 1150/ 1200] Overall Loss 0.119972 Objective Loss 0.119972 LR 0.000031 Time 0.020416 -2022-12-06 11:50:11,993 - Epoch: [195][ 1160/ 1200] Overall Loss 0.119865 Objective Loss 0.119865 LR 0.000031 Time 0.020411 -2022-12-06 11:50:12,194 - Epoch: [195][ 1170/ 1200] Overall Loss 0.119903 Objective Loss 0.119903 LR 0.000031 Time 0.020408 -2022-12-06 11:50:12,392 - Epoch: [195][ 1180/ 1200] Overall Loss 0.119764 Objective Loss 0.119764 LR 0.000031 Time 0.020402 -2022-12-06 11:50:12,594 - Epoch: [195][ 1190/ 1200] Overall Loss 0.119665 Objective Loss 0.119665 LR 0.000031 Time 0.020400 -2022-12-06 11:50:12,823 - Epoch: [195][ 1200/ 1200] Overall Loss 0.119668 Objective Loss 0.119668 Top1 90.794979 Top5 98.326360 LR 0.000031 Time 0.020420 -2022-12-06 11:50:12,912 - --- validate (epoch=195)----------- -2022-12-06 11:50:12,912 - 34129 samples (256 per mini-batch) -2022-12-06 11:50:13,354 - Epoch: [195][ 10/ 134] Loss 0.210272 Top1 89.179688 Top5 98.789062 -2022-12-06 11:50:13,485 - Epoch: [195][ 20/ 134] Loss 0.219396 Top1 89.062500 Top5 98.554688 -2022-12-06 11:50:13,618 - Epoch: [195][ 30/ 134] Loss 0.220279 Top1 89.231771 Top5 98.684896 -2022-12-06 11:50:13,747 - Epoch: [195][ 40/ 134] Loss 0.224414 Top1 89.072266 Top5 98.632812 -2022-12-06 11:50:13,879 - Epoch: [195][ 50/ 134] Loss 0.221755 Top1 89.078125 Top5 98.625000 -2022-12-06 11:50:14,011 - Epoch: [195][ 60/ 134] Loss 0.223938 Top1 89.029948 Top5 98.645833 -2022-12-06 11:50:14,142 - Epoch: [195][ 70/ 134] Loss 0.223630 Top1 88.978795 Top5 98.655134 -2022-12-06 11:50:14,273 - Epoch: [195][ 80/ 134] Loss 0.223479 Top1 88.901367 Top5 98.627930 -2022-12-06 11:50:14,404 - Epoch: [195][ 90/ 134] Loss 0.220146 Top1 88.980035 Top5 98.663194 -2022-12-06 11:50:14,536 - Epoch: [195][ 100/ 134] Loss 0.223853 Top1 88.886719 Top5 98.617188 -2022-12-06 11:50:14,667 - Epoch: [195][ 110/ 134] Loss 0.223781 Top1 88.867188 Top5 98.622159 -2022-12-06 11:50:14,799 - Epoch: [195][ 120/ 134] Loss 0.220066 Top1 88.994141 Top5 98.675130 -2022-12-06 11:50:14,930 - Epoch: [195][ 130/ 134] Loss 0.221444 Top1 89.002404 Top5 98.677885 -2022-12-06 11:50:14,968 - Epoch: [195][ 134/ 134] Loss 0.222334 Top1 88.936095 Top5 98.684403 -2022-12-06 11:50:15,056 - ==> Top1: 88.936 Top5: 98.684 Loss: 0.222 - -2022-12-06 11:50:15,056 - ==> Confusion: -[[ 929 1 0 3 3 6 1 0 4 37 0 1 1 1 4 1 2 0 1 0 1] - [ 1 949 4 2 7 14 2 11 2 1 3 3 0 0 1 1 3 0 10 3 10] - [ 4 2 1022 13 4 2 11 10 0 2 4 5 1 2 1 1 0 1 2 6 10] - [ 5 1 13 957 1 2 1 0 0 1 7 0 3 1 8 0 0 1 11 0 8] - [ 7 5 1 0 964 1 1 2 1 6 1 5 0 2 8 4 4 3 0 0 5] - [ 1 12 1 5 5 983 2 15 4 3 0 15 2 12 2 1 1 1 0 1 3] - [ 1 2 10 3 0 0 1078 3 0 0 0 2 0 1 0 4 0 3 1 7 3] - [ 1 7 3 3 2 21 9 975 0 0 1 4 0 1 0 0 1 0 13 8 5] - [ 9 1 0 1 0 1 1 1 984 35 8 2 1 4 10 1 2 0 2 1 0] - [ 46 1 0 0 2 1 0 1 21 907 1 1 0 9 3 0 0 1 1 1 5] - [ 0 2 4 3 1 0 1 2 9 1 968 0 0 9 5 1 0 0 4 2 7] - [ 2 1 2 0 1 10 3 1 1 0 1 985 20 3 0 5 3 4 0 6 3] - [ 0 1 2 2 0 2 1 1 1 0 1 24 910 1 1 7 1 6 0 2 6] - [ 2 1 0 0 2 7 0 1 11 11 3 5 3 964 1 2 2 0 0 2 6] - [ 6 5 2 8 2 3 0 1 10 1 0 3 2 3 1072 0 0 1 5 2 4] - [ 1 0 1 0 2 0 1 0 1 1 1 7 3 2 0 998 4 12 0 5 4] - [ 2 0 1 2 2 1 0 0 1 0 0 1 3 1 0 9 1036 1 1 5 6] - [ 3 0 1 1 1 1 0 0 0 3 0 6 15 2 1 11 0 987 0 0 4] - [ 1 3 4 3 0 3 0 21 3 1 3 4 2 1 6 0 0 1 947 1 4] - [ 1 3 1 1 1 4 4 7 0 1 2 12 7 5 0 3 3 2 1 1015 7] - [ 104 193 147 78 95 133 70 128 80 66 139 82 262 220 137 78 117 79 127 172 10719]] - -2022-12-06 11:50:15,713 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:50:15,713 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:50:15,719 - - -2022-12-06 11:50:15,719 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:50:16,674 - Epoch: [196][ 10/ 1200] Overall Loss 0.119217 Objective Loss 0.119217 LR 0.000031 Time 0.095455 -2022-12-06 11:50:16,876 - Epoch: [196][ 20/ 1200] Overall Loss 0.116718 Objective Loss 0.116718 LR 0.000031 Time 0.057759 -2022-12-06 11:50:17,077 - Epoch: [196][ 30/ 1200] Overall Loss 0.119404 Objective Loss 0.119404 LR 0.000031 Time 0.045209 -2022-12-06 11:50:17,275 - Epoch: [196][ 40/ 1200] Overall Loss 0.118367 Objective Loss 0.118367 LR 0.000031 Time 0.038841 -2022-12-06 11:50:17,476 - Epoch: [196][ 50/ 1200] Overall Loss 0.116977 Objective Loss 0.116977 LR 0.000031 Time 0.035081 -2022-12-06 11:50:17,674 - Epoch: [196][ 60/ 1200] Overall Loss 0.117180 Objective Loss 0.117180 LR 0.000031 Time 0.032525 -2022-12-06 11:50:17,875 - Epoch: [196][ 70/ 1200] Overall Loss 0.117259 Objective Loss 0.117259 LR 0.000031 Time 0.030741 -2022-12-06 11:50:18,073 - Epoch: [196][ 80/ 1200] Overall Loss 0.116830 Objective Loss 0.116830 LR 0.000031 Time 0.029366 -2022-12-06 11:50:18,274 - Epoch: [196][ 90/ 1200] Overall Loss 0.119268 Objective Loss 0.119268 LR 0.000031 Time 0.028327 -2022-12-06 11:50:18,471 - Epoch: [196][ 100/ 1200] Overall Loss 0.118592 Objective Loss 0.118592 LR 0.000031 Time 0.027463 -2022-12-06 11:50:18,671 - Epoch: [196][ 110/ 1200] Overall Loss 0.118722 Objective Loss 0.118722 LR 0.000031 Time 0.026782 -2022-12-06 11:50:18,869 - Epoch: [196][ 120/ 1200] Overall Loss 0.117663 Objective Loss 0.117663 LR 0.000031 Time 0.026192 -2022-12-06 11:50:19,070 - Epoch: [196][ 130/ 1200] Overall Loss 0.117439 Objective Loss 0.117439 LR 0.000031 Time 0.025717 -2022-12-06 11:50:19,267 - Epoch: [196][ 140/ 1200] Overall Loss 0.117386 Objective Loss 0.117386 LR 0.000031 Time 0.025287 -2022-12-06 11:50:19,468 - Epoch: [196][ 150/ 1200] Overall Loss 0.117674 Objective Loss 0.117674 LR 0.000031 Time 0.024935 -2022-12-06 11:50:19,666 - Epoch: [196][ 160/ 1200] Overall Loss 0.118282 Objective Loss 0.118282 LR 0.000031 Time 0.024608 -2022-12-06 11:50:19,867 - Epoch: [196][ 170/ 1200] Overall Loss 0.117914 Objective Loss 0.117914 LR 0.000031 Time 0.024340 -2022-12-06 11:50:20,064 - Epoch: [196][ 180/ 1200] Overall Loss 0.117440 Objective Loss 0.117440 LR 0.000031 Time 0.024084 -2022-12-06 11:50:20,266 - Epoch: [196][ 190/ 1200] Overall Loss 0.116960 Objective Loss 0.116960 LR 0.000031 Time 0.023876 -2022-12-06 11:50:20,467 - Epoch: [196][ 200/ 1200] Overall Loss 0.116941 Objective Loss 0.116941 LR 0.000031 Time 0.023682 -2022-12-06 11:50:20,670 - Epoch: [196][ 210/ 1200] Overall Loss 0.116118 Objective Loss 0.116118 LR 0.000031 Time 0.023518 -2022-12-06 11:50:20,870 - Epoch: [196][ 220/ 1200] Overall Loss 0.115967 Objective Loss 0.115967 LR 0.000031 Time 0.023359 -2022-12-06 11:50:21,073 - Epoch: [196][ 230/ 1200] Overall Loss 0.115699 Objective Loss 0.115699 LR 0.000031 Time 0.023222 -2022-12-06 11:50:21,273 - Epoch: [196][ 240/ 1200] Overall Loss 0.116019 Objective Loss 0.116019 LR 0.000031 Time 0.023087 -2022-12-06 11:50:21,477 - Epoch: [196][ 250/ 1200] Overall Loss 0.115743 Objective Loss 0.115743 LR 0.000031 Time 0.022974 -2022-12-06 11:50:21,677 - Epoch: [196][ 260/ 1200] Overall Loss 0.115483 Objective Loss 0.115483 LR 0.000031 Time 0.022859 -2022-12-06 11:50:21,880 - Epoch: [196][ 270/ 1200] Overall Loss 0.115455 Objective Loss 0.115455 LR 0.000031 Time 0.022762 -2022-12-06 11:50:22,081 - Epoch: [196][ 280/ 1200] Overall Loss 0.115115 Objective Loss 0.115115 LR 0.000031 Time 0.022664 -2022-12-06 11:50:22,284 - Epoch: [196][ 290/ 1200] Overall Loss 0.114724 Objective Loss 0.114724 LR 0.000031 Time 0.022581 -2022-12-06 11:50:22,484 - Epoch: [196][ 300/ 1200] Overall Loss 0.115008 Objective Loss 0.115008 LR 0.000031 Time 0.022494 -2022-12-06 11:50:22,687 - Epoch: [196][ 310/ 1200] Overall Loss 0.114595 Objective Loss 0.114595 LR 0.000031 Time 0.022421 -2022-12-06 11:50:22,887 - Epoch: [196][ 320/ 1200] Overall Loss 0.115301 Objective Loss 0.115301 LR 0.000031 Time 0.022343 -2022-12-06 11:50:23,090 - Epoch: [196][ 330/ 1200] Overall Loss 0.115162 Objective Loss 0.115162 LR 0.000031 Time 0.022279 -2022-12-06 11:50:23,290 - Epoch: [196][ 340/ 1200] Overall Loss 0.114830 Objective Loss 0.114830 LR 0.000031 Time 0.022210 -2022-12-06 11:50:23,493 - Epoch: [196][ 350/ 1200] Overall Loss 0.114782 Objective Loss 0.114782 LR 0.000031 Time 0.022155 -2022-12-06 11:50:23,693 - Epoch: [196][ 360/ 1200] Overall Loss 0.114974 Objective Loss 0.114974 LR 0.000031 Time 0.022094 -2022-12-06 11:50:23,896 - Epoch: [196][ 370/ 1200] Overall Loss 0.115149 Objective Loss 0.115149 LR 0.000031 Time 0.022044 -2022-12-06 11:50:24,096 - Epoch: [196][ 380/ 1200] Overall Loss 0.115023 Objective Loss 0.115023 LR 0.000031 Time 0.021989 -2022-12-06 11:50:24,299 - Epoch: [196][ 390/ 1200] Overall Loss 0.115225 Objective Loss 0.115225 LR 0.000031 Time 0.021945 -2022-12-06 11:50:24,499 - Epoch: [196][ 400/ 1200] Overall Loss 0.115557 Objective Loss 0.115557 LR 0.000031 Time 0.021894 -2022-12-06 11:50:24,702 - Epoch: [196][ 410/ 1200] Overall Loss 0.116110 Objective Loss 0.116110 LR 0.000031 Time 0.021854 -2022-12-06 11:50:24,903 - Epoch: [196][ 420/ 1200] Overall Loss 0.116246 Objective Loss 0.116246 LR 0.000031 Time 0.021811 -2022-12-06 11:50:25,106 - Epoch: [196][ 430/ 1200] Overall Loss 0.115990 Objective Loss 0.115990 LR 0.000031 Time 0.021774 -2022-12-06 11:50:25,307 - Epoch: [196][ 440/ 1200] Overall Loss 0.116071 Objective Loss 0.116071 LR 0.000031 Time 0.021735 -2022-12-06 11:50:25,510 - Epoch: [196][ 450/ 1200] Overall Loss 0.116494 Objective Loss 0.116494 LR 0.000031 Time 0.021701 -2022-12-06 11:50:25,711 - Epoch: [196][ 460/ 1200] Overall Loss 0.116272 Objective Loss 0.116272 LR 0.000031 Time 0.021664 -2022-12-06 11:50:25,913 - Epoch: [196][ 470/ 1200] Overall Loss 0.116356 Objective Loss 0.116356 LR 0.000031 Time 0.021634 -2022-12-06 11:50:26,113 - Epoch: [196][ 480/ 1200] Overall Loss 0.116189 Objective Loss 0.116189 LR 0.000031 Time 0.021598 -2022-12-06 11:50:26,316 - Epoch: [196][ 490/ 1200] Overall Loss 0.116022 Objective Loss 0.116022 LR 0.000031 Time 0.021571 -2022-12-06 11:50:26,517 - Epoch: [196][ 500/ 1200] Overall Loss 0.116383 Objective Loss 0.116383 LR 0.000031 Time 0.021539 -2022-12-06 11:50:26,719 - Epoch: [196][ 510/ 1200] Overall Loss 0.116325 Objective Loss 0.116325 LR 0.000031 Time 0.021513 -2022-12-06 11:50:26,920 - Epoch: [196][ 520/ 1200] Overall Loss 0.116137 Objective Loss 0.116137 LR 0.000031 Time 0.021483 -2022-12-06 11:50:27,122 - Epoch: [196][ 530/ 1200] Overall Loss 0.116110 Objective Loss 0.116110 LR 0.000031 Time 0.021459 -2022-12-06 11:50:27,320 - Epoch: [196][ 540/ 1200] Overall Loss 0.116117 Objective Loss 0.116117 LR 0.000031 Time 0.021426 -2022-12-06 11:50:27,521 - Epoch: [196][ 550/ 1200] Overall Loss 0.116335 Objective Loss 0.116335 LR 0.000031 Time 0.021401 -2022-12-06 11:50:27,718 - Epoch: [196][ 560/ 1200] Overall Loss 0.116438 Objective Loss 0.116438 LR 0.000031 Time 0.021371 -2022-12-06 11:50:27,919 - Epoch: [196][ 570/ 1200] Overall Loss 0.116381 Objective Loss 0.116381 LR 0.000031 Time 0.021347 -2022-12-06 11:50:28,117 - Epoch: [196][ 580/ 1200] Overall Loss 0.116237 Objective Loss 0.116237 LR 0.000031 Time 0.021319 -2022-12-06 11:50:28,317 - Epoch: [196][ 590/ 1200] Overall Loss 0.116252 Objective Loss 0.116252 LR 0.000031 Time 0.021297 -2022-12-06 11:50:28,515 - Epoch: [196][ 600/ 1200] Overall Loss 0.116093 Objective Loss 0.116093 LR 0.000031 Time 0.021270 -2022-12-06 11:50:28,716 - Epoch: [196][ 610/ 1200] Overall Loss 0.116514 Objective Loss 0.116514 LR 0.000031 Time 0.021250 -2022-12-06 11:50:28,914 - Epoch: [196][ 620/ 1200] Overall Loss 0.116585 Objective Loss 0.116585 LR 0.000031 Time 0.021225 -2022-12-06 11:50:29,114 - Epoch: [196][ 630/ 1200] Overall Loss 0.116565 Objective Loss 0.116565 LR 0.000031 Time 0.021205 -2022-12-06 11:50:29,311 - Epoch: [196][ 640/ 1200] Overall Loss 0.116633 Objective Loss 0.116633 LR 0.000031 Time 0.021181 -2022-12-06 11:50:29,511 - Epoch: [196][ 650/ 1200] Overall Loss 0.116800 Objective Loss 0.116800 LR 0.000031 Time 0.021162 -2022-12-06 11:50:29,709 - Epoch: [196][ 660/ 1200] Overall Loss 0.117000 Objective Loss 0.117000 LR 0.000031 Time 0.021141 -2022-12-06 11:50:29,911 - Epoch: [196][ 670/ 1200] Overall Loss 0.116972 Objective Loss 0.116972 LR 0.000031 Time 0.021125 -2022-12-06 11:50:30,108 - Epoch: [196][ 680/ 1200] Overall Loss 0.117217 Objective Loss 0.117217 LR 0.000031 Time 0.021105 -2022-12-06 11:50:30,308 - Epoch: [196][ 690/ 1200] Overall Loss 0.117132 Objective Loss 0.117132 LR 0.000031 Time 0.021087 -2022-12-06 11:50:30,506 - Epoch: [196][ 700/ 1200] Overall Loss 0.116876 Objective Loss 0.116876 LR 0.000031 Time 0.021068 -2022-12-06 11:50:30,707 - Epoch: [196][ 710/ 1200] Overall Loss 0.116825 Objective Loss 0.116825 LR 0.000031 Time 0.021053 -2022-12-06 11:50:30,904 - Epoch: [196][ 720/ 1200] Overall Loss 0.116795 Objective Loss 0.116795 LR 0.000031 Time 0.021034 -2022-12-06 11:50:31,105 - Epoch: [196][ 730/ 1200] Overall Loss 0.116971 Objective Loss 0.116971 LR 0.000031 Time 0.021020 -2022-12-06 11:50:31,303 - Epoch: [196][ 740/ 1200] Overall Loss 0.117044 Objective Loss 0.117044 LR 0.000031 Time 0.021003 -2022-12-06 11:50:31,503 - Epoch: [196][ 750/ 1200] Overall Loss 0.117016 Objective Loss 0.117016 LR 0.000031 Time 0.020989 -2022-12-06 11:50:31,700 - Epoch: [196][ 760/ 1200] Overall Loss 0.117001 Objective Loss 0.117001 LR 0.000031 Time 0.020972 -2022-12-06 11:50:31,901 - Epoch: [196][ 770/ 1200] Overall Loss 0.117030 Objective Loss 0.117030 LR 0.000031 Time 0.020959 -2022-12-06 11:50:32,099 - Epoch: [196][ 780/ 1200] Overall Loss 0.116981 Objective Loss 0.116981 LR 0.000031 Time 0.020943 -2022-12-06 11:50:32,299 - Epoch: [196][ 790/ 1200] Overall Loss 0.117087 Objective Loss 0.117087 LR 0.000031 Time 0.020931 -2022-12-06 11:50:32,498 - Epoch: [196][ 800/ 1200] Overall Loss 0.117177 Objective Loss 0.117177 LR 0.000031 Time 0.020917 -2022-12-06 11:50:32,698 - Epoch: [196][ 810/ 1200] Overall Loss 0.117205 Objective Loss 0.117205 LR 0.000031 Time 0.020905 -2022-12-06 11:50:32,895 - Epoch: [196][ 820/ 1200] Overall Loss 0.117368 Objective Loss 0.117368 LR 0.000031 Time 0.020890 -2022-12-06 11:50:33,096 - Epoch: [196][ 830/ 1200] Overall Loss 0.117551 Objective Loss 0.117551 LR 0.000031 Time 0.020880 -2022-12-06 11:50:33,294 - Epoch: [196][ 840/ 1200] Overall Loss 0.117695 Objective Loss 0.117695 LR 0.000031 Time 0.020866 -2022-12-06 11:50:33,496 - Epoch: [196][ 850/ 1200] Overall Loss 0.117807 Objective Loss 0.117807 LR 0.000031 Time 0.020857 -2022-12-06 11:50:33,694 - Epoch: [196][ 860/ 1200] Overall Loss 0.118051 Objective Loss 0.118051 LR 0.000031 Time 0.020845 -2022-12-06 11:50:33,895 - Epoch: [196][ 870/ 1200] Overall Loss 0.118134 Objective Loss 0.118134 LR 0.000031 Time 0.020835 -2022-12-06 11:50:34,093 - Epoch: [196][ 880/ 1200] Overall Loss 0.118185 Objective Loss 0.118185 LR 0.000031 Time 0.020823 -2022-12-06 11:50:34,293 - Epoch: [196][ 890/ 1200] Overall Loss 0.118161 Objective Loss 0.118161 LR 0.000031 Time 0.020813 -2022-12-06 11:50:34,491 - Epoch: [196][ 900/ 1200] Overall Loss 0.117973 Objective Loss 0.117973 LR 0.000031 Time 0.020802 -2022-12-06 11:50:34,692 - Epoch: [196][ 910/ 1200] Overall Loss 0.117979 Objective Loss 0.117979 LR 0.000031 Time 0.020793 -2022-12-06 11:50:34,890 - Epoch: [196][ 920/ 1200] Overall Loss 0.118186 Objective Loss 0.118186 LR 0.000031 Time 0.020781 -2022-12-06 11:50:35,091 - Epoch: [196][ 930/ 1200] Overall Loss 0.118355 Objective Loss 0.118355 LR 0.000031 Time 0.020774 -2022-12-06 11:50:35,290 - Epoch: [196][ 940/ 1200] Overall Loss 0.118424 Objective Loss 0.118424 LR 0.000031 Time 0.020764 -2022-12-06 11:50:35,491 - Epoch: [196][ 950/ 1200] Overall Loss 0.118629 Objective Loss 0.118629 LR 0.000031 Time 0.020756 -2022-12-06 11:50:35,690 - Epoch: [196][ 960/ 1200] Overall Loss 0.118550 Objective Loss 0.118550 LR 0.000031 Time 0.020746 -2022-12-06 11:50:35,891 - Epoch: [196][ 970/ 1200] Overall Loss 0.118551 Objective Loss 0.118551 LR 0.000031 Time 0.020740 -2022-12-06 11:50:36,089 - Epoch: [196][ 980/ 1200] Overall Loss 0.118439 Objective Loss 0.118439 LR 0.000031 Time 0.020729 -2022-12-06 11:50:36,290 - Epoch: [196][ 990/ 1200] Overall Loss 0.118318 Objective Loss 0.118318 LR 0.000031 Time 0.020722 -2022-12-06 11:50:36,489 - Epoch: [196][ 1000/ 1200] Overall Loss 0.118278 Objective Loss 0.118278 LR 0.000031 Time 0.020713 -2022-12-06 11:50:36,689 - Epoch: [196][ 1010/ 1200] Overall Loss 0.118360 Objective Loss 0.118360 LR 0.000031 Time 0.020706 -2022-12-06 11:50:36,887 - Epoch: [196][ 1020/ 1200] Overall Loss 0.118474 Objective Loss 0.118474 LR 0.000031 Time 0.020696 -2022-12-06 11:50:37,088 - Epoch: [196][ 1030/ 1200] Overall Loss 0.118553 Objective Loss 0.118553 LR 0.000031 Time 0.020690 -2022-12-06 11:50:37,286 - Epoch: [196][ 1040/ 1200] Overall Loss 0.118610 Objective Loss 0.118610 LR 0.000031 Time 0.020681 -2022-12-06 11:50:37,486 - Epoch: [196][ 1050/ 1200] Overall Loss 0.118529 Objective Loss 0.118529 LR 0.000031 Time 0.020674 -2022-12-06 11:50:37,684 - Epoch: [196][ 1060/ 1200] Overall Loss 0.118484 Objective Loss 0.118484 LR 0.000031 Time 0.020665 -2022-12-06 11:50:37,884 - Epoch: [196][ 1070/ 1200] Overall Loss 0.118413 Objective Loss 0.118413 LR 0.000031 Time 0.020659 -2022-12-06 11:50:38,082 - Epoch: [196][ 1080/ 1200] Overall Loss 0.118365 Objective Loss 0.118365 LR 0.000031 Time 0.020650 -2022-12-06 11:50:38,284 - Epoch: [196][ 1090/ 1200] Overall Loss 0.118295 Objective Loss 0.118295 LR 0.000031 Time 0.020645 -2022-12-06 11:50:38,482 - Epoch: [196][ 1100/ 1200] Overall Loss 0.118249 Objective Loss 0.118249 LR 0.000031 Time 0.020637 -2022-12-06 11:50:38,683 - Epoch: [196][ 1110/ 1200] Overall Loss 0.118310 Objective Loss 0.118310 LR 0.000031 Time 0.020632 -2022-12-06 11:50:38,881 - Epoch: [196][ 1120/ 1200] Overall Loss 0.118344 Objective Loss 0.118344 LR 0.000031 Time 0.020624 -2022-12-06 11:50:39,081 - Epoch: [196][ 1130/ 1200] Overall Loss 0.118314 Objective Loss 0.118314 LR 0.000031 Time 0.020618 -2022-12-06 11:50:39,279 - Epoch: [196][ 1140/ 1200] Overall Loss 0.118300 Objective Loss 0.118300 LR 0.000031 Time 0.020610 -2022-12-06 11:50:39,480 - Epoch: [196][ 1150/ 1200] Overall Loss 0.118235 Objective Loss 0.118235 LR 0.000031 Time 0.020605 -2022-12-06 11:50:39,678 - Epoch: [196][ 1160/ 1200] Overall Loss 0.118259 Objective Loss 0.118259 LR 0.000031 Time 0.020598 -2022-12-06 11:50:39,879 - Epoch: [196][ 1170/ 1200] Overall Loss 0.118210 Objective Loss 0.118210 LR 0.000031 Time 0.020593 -2022-12-06 11:50:40,076 - Epoch: [196][ 1180/ 1200] Overall Loss 0.118235 Objective Loss 0.118235 LR 0.000031 Time 0.020585 -2022-12-06 11:50:40,276 - Epoch: [196][ 1190/ 1200] Overall Loss 0.118072 Objective Loss 0.118072 LR 0.000031 Time 0.020580 -2022-12-06 11:50:40,510 - Epoch: [196][ 1200/ 1200] Overall Loss 0.118058 Objective Loss 0.118058 Top1 92.259414 Top5 99.163180 LR 0.000031 Time 0.020603 -2022-12-06 11:50:40,599 - --- validate (epoch=196)----------- -2022-12-06 11:50:40,599 - 34129 samples (256 per mini-batch) -2022-12-06 11:50:41,055 - Epoch: [196][ 10/ 134] Loss 0.223114 Top1 88.906250 Top5 98.671875 -2022-12-06 11:50:41,184 - Epoch: [196][ 20/ 134] Loss 0.223628 Top1 88.828125 Top5 98.496094 -2022-12-06 11:50:41,312 - Epoch: [196][ 30/ 134] Loss 0.218270 Top1 88.763021 Top5 98.541667 -2022-12-06 11:50:41,439 - Epoch: [196][ 40/ 134] Loss 0.216653 Top1 89.160156 Top5 98.662109 -2022-12-06 11:50:41,567 - Epoch: [196][ 50/ 134] Loss 0.212144 Top1 89.132812 Top5 98.726562 -2022-12-06 11:50:41,704 - Epoch: [196][ 60/ 134] Loss 0.216457 Top1 89.095052 Top5 98.769531 -2022-12-06 11:50:41,837 - Epoch: [196][ 70/ 134] Loss 0.217603 Top1 89.107143 Top5 98.783482 -2022-12-06 11:50:41,976 - Epoch: [196][ 80/ 134] Loss 0.220936 Top1 89.052734 Top5 98.740234 -2022-12-06 11:50:42,109 - Epoch: [196][ 90/ 134] Loss 0.220961 Top1 89.105903 Top5 98.758681 -2022-12-06 11:50:42,247 - Epoch: [196][ 100/ 134] Loss 0.222111 Top1 88.960938 Top5 98.769531 -2022-12-06 11:50:42,380 - Epoch: [196][ 110/ 134] Loss 0.222153 Top1 88.977273 Top5 98.781960 -2022-12-06 11:50:42,519 - Epoch: [196][ 120/ 134] Loss 0.224161 Top1 88.974609 Top5 98.772786 -2022-12-06 11:50:42,652 - Epoch: [196][ 130/ 134] Loss 0.225035 Top1 88.957332 Top5 98.765024 -2022-12-06 11:50:42,689 - Epoch: [196][ 134/ 134] Loss 0.223186 Top1 88.977116 Top5 98.745935 -2022-12-06 11:50:42,777 - ==> Top1: 88.977 Top5: 98.746 Loss: 0.223 - -2022-12-06 11:50:42,778 - ==> Confusion: -[[ 918 1 1 2 4 7 0 1 5 43 0 1 1 2 4 1 2 0 2 0 1] - [ 1 947 2 2 6 20 1 10 2 2 1 3 2 1 0 1 2 0 11 3 10] - [ 2 4 1022 14 3 2 9 10 0 4 3 3 1 1 2 2 1 1 4 4 11] - [ 5 0 12 961 1 2 0 0 0 1 8 0 2 1 6 0 0 2 11 0 8] - [ 8 5 1 0 959 1 1 1 1 7 2 4 0 2 9 5 5 3 1 0 5] - [ 1 11 0 5 4 985 2 16 4 3 1 10 3 15 1 1 1 0 2 2 2] - [ 1 2 6 4 0 1 1082 2 1 1 0 1 0 1 0 3 0 1 2 7 3] - [ 2 8 2 3 0 25 8 963 0 0 2 5 0 1 2 0 1 0 17 10 5] - [ 5 2 0 0 0 1 1 1 985 34 9 2 0 7 12 1 1 0 1 1 1] - [ 42 0 3 0 4 1 0 3 22 904 1 1 0 8 3 0 0 2 1 0 6] - [ 1 2 4 2 1 0 0 2 9 1 968 0 0 10 4 1 0 0 4 2 8] - [ 3 0 1 0 1 7 3 1 2 1 0 982 23 2 0 7 2 5 0 5 6] - [ 1 1 1 2 0 2 0 1 1 1 0 24 915 0 0 5 1 5 0 3 6] - [ 2 1 0 0 1 7 0 1 11 8 5 5 3 966 1 2 2 0 0 1 7] - [ 6 5 3 9 2 3 0 0 12 1 0 2 1 4 1069 0 0 1 6 2 4] - [ 1 0 0 3 3 0 3 1 0 0 2 5 2 2 0 1001 5 9 0 3 3] - [ 0 0 1 1 2 1 1 0 2 0 0 1 3 1 0 10 1036 0 0 5 8] - [ 3 0 1 2 2 1 0 0 0 2 0 3 18 2 4 9 0 986 0 0 3] - [ 1 2 5 6 0 3 0 18 4 1 4 3 2 0 6 0 0 2 946 1 4] - [ 1 4 1 2 2 4 5 6 0 2 2 12 2 7 0 4 5 1 0 1015 5] - [ 90 172 140 97 92 144 67 118 73 69 131 72 278 221 128 79 109 66 142 183 10755]] - -2022-12-06 11:50:43,449 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:50:43,449 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:50:43,455 - - -2022-12-06 11:50:43,455 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:50:44,407 - Epoch: [197][ 10/ 1200] Overall Loss 0.115472 Objective Loss 0.115472 LR 0.000031 Time 0.095103 -2022-12-06 11:50:44,619 - Epoch: [197][ 20/ 1200] Overall Loss 0.121610 Objective Loss 0.121610 LR 0.000031 Time 0.058147 -2022-12-06 11:50:44,829 - Epoch: [197][ 30/ 1200] Overall Loss 0.116917 Objective Loss 0.116917 LR 0.000031 Time 0.045737 -2022-12-06 11:50:45,040 - Epoch: [197][ 40/ 1200] Overall Loss 0.117840 Objective Loss 0.117840 LR 0.000031 Time 0.039574 -2022-12-06 11:50:45,250 - Epoch: [197][ 50/ 1200] Overall Loss 0.119884 Objective Loss 0.119884 LR 0.000031 Time 0.035837 -2022-12-06 11:50:45,461 - Epoch: [197][ 60/ 1200] Overall Loss 0.116069 Objective Loss 0.116069 LR 0.000031 Time 0.033369 -2022-12-06 11:50:45,671 - Epoch: [197][ 70/ 1200] Overall Loss 0.115252 Objective Loss 0.115252 LR 0.000031 Time 0.031589 -2022-12-06 11:50:45,881 - Epoch: [197][ 80/ 1200] Overall Loss 0.114859 Objective Loss 0.114859 LR 0.000031 Time 0.030267 -2022-12-06 11:50:46,091 - Epoch: [197][ 90/ 1200] Overall Loss 0.113791 Objective Loss 0.113791 LR 0.000031 Time 0.029225 -2022-12-06 11:50:46,301 - Epoch: [197][ 100/ 1200] Overall Loss 0.113427 Objective Loss 0.113427 LR 0.000031 Time 0.028404 -2022-12-06 11:50:46,511 - Epoch: [197][ 110/ 1200] Overall Loss 0.115008 Objective Loss 0.115008 LR 0.000031 Time 0.027720 -2022-12-06 11:50:46,721 - Epoch: [197][ 120/ 1200] Overall Loss 0.115478 Objective Loss 0.115478 LR 0.000031 Time 0.027152 -2022-12-06 11:50:46,925 - Epoch: [197][ 130/ 1200] Overall Loss 0.116038 Objective Loss 0.116038 LR 0.000031 Time 0.026629 -2022-12-06 11:50:47,129 - Epoch: [197][ 140/ 1200] Overall Loss 0.116169 Objective Loss 0.116169 LR 0.000031 Time 0.026181 -2022-12-06 11:50:47,332 - Epoch: [197][ 150/ 1200] Overall Loss 0.117021 Objective Loss 0.117021 LR 0.000031 Time 0.025788 -2022-12-06 11:50:47,537 - Epoch: [197][ 160/ 1200] Overall Loss 0.117714 Objective Loss 0.117714 LR 0.000031 Time 0.025451 -2022-12-06 11:50:47,740 - Epoch: [197][ 170/ 1200] Overall Loss 0.118229 Objective Loss 0.118229 LR 0.000031 Time 0.025150 -2022-12-06 11:50:47,945 - Epoch: [197][ 180/ 1200] Overall Loss 0.118736 Objective Loss 0.118736 LR 0.000031 Time 0.024884 -2022-12-06 11:50:48,149 - Epoch: [197][ 190/ 1200] Overall Loss 0.118284 Objective Loss 0.118284 LR 0.000031 Time 0.024646 -2022-12-06 11:50:48,353 - Epoch: [197][ 200/ 1200] Overall Loss 0.117367 Objective Loss 0.117367 LR 0.000031 Time 0.024432 -2022-12-06 11:50:48,556 - Epoch: [197][ 210/ 1200] Overall Loss 0.118002 Objective Loss 0.118002 LR 0.000031 Time 0.024235 -2022-12-06 11:50:48,760 - Epoch: [197][ 220/ 1200] Overall Loss 0.118061 Objective Loss 0.118061 LR 0.000031 Time 0.024058 -2022-12-06 11:50:48,964 - Epoch: [197][ 230/ 1200] Overall Loss 0.117864 Objective Loss 0.117864 LR 0.000031 Time 0.023893 -2022-12-06 11:50:49,168 - Epoch: [197][ 240/ 1200] Overall Loss 0.117707 Objective Loss 0.117707 LR 0.000031 Time 0.023746 -2022-12-06 11:50:49,371 - Epoch: [197][ 250/ 1200] Overall Loss 0.118180 Objective Loss 0.118180 LR 0.000031 Time 0.023608 -2022-12-06 11:50:49,575 - Epoch: [197][ 260/ 1200] Overall Loss 0.117873 Objective Loss 0.117873 LR 0.000031 Time 0.023483 -2022-12-06 11:50:49,779 - Epoch: [197][ 270/ 1200] Overall Loss 0.118276 Objective Loss 0.118276 LR 0.000031 Time 0.023366 -2022-12-06 11:50:49,983 - Epoch: [197][ 280/ 1200] Overall Loss 0.117909 Objective Loss 0.117909 LR 0.000031 Time 0.023258 -2022-12-06 11:50:50,188 - Epoch: [197][ 290/ 1200] Overall Loss 0.118307 Objective Loss 0.118307 LR 0.000031 Time 0.023159 -2022-12-06 11:50:50,392 - Epoch: [197][ 300/ 1200] Overall Loss 0.118186 Objective Loss 0.118186 LR 0.000031 Time 0.023065 -2022-12-06 11:50:50,595 - Epoch: [197][ 310/ 1200] Overall Loss 0.118533 Objective Loss 0.118533 LR 0.000031 Time 0.022975 -2022-12-06 11:50:50,799 - Epoch: [197][ 320/ 1200] Overall Loss 0.118947 Objective Loss 0.118947 LR 0.000031 Time 0.022894 -2022-12-06 11:50:51,003 - Epoch: [197][ 330/ 1200] Overall Loss 0.118768 Objective Loss 0.118768 LR 0.000031 Time 0.022816 -2022-12-06 11:50:51,207 - Epoch: [197][ 340/ 1200] Overall Loss 0.118171 Objective Loss 0.118171 LR 0.000031 Time 0.022745 -2022-12-06 11:50:51,411 - Epoch: [197][ 350/ 1200] Overall Loss 0.117777 Objective Loss 0.117777 LR 0.000031 Time 0.022676 -2022-12-06 11:50:51,615 - Epoch: [197][ 360/ 1200] Overall Loss 0.117664 Objective Loss 0.117664 LR 0.000031 Time 0.022610 -2022-12-06 11:50:51,819 - Epoch: [197][ 370/ 1200] Overall Loss 0.117972 Objective Loss 0.117972 LR 0.000031 Time 0.022548 -2022-12-06 11:50:52,023 - Epoch: [197][ 380/ 1200] Overall Loss 0.117991 Objective Loss 0.117991 LR 0.000031 Time 0.022490 -2022-12-06 11:50:52,226 - Epoch: [197][ 390/ 1200] Overall Loss 0.118457 Objective Loss 0.118457 LR 0.000031 Time 0.022434 -2022-12-06 11:50:52,431 - Epoch: [197][ 400/ 1200] Overall Loss 0.118586 Objective Loss 0.118586 LR 0.000031 Time 0.022383 -2022-12-06 11:50:52,634 - Epoch: [197][ 410/ 1200] Overall Loss 0.118278 Objective Loss 0.118278 LR 0.000031 Time 0.022332 -2022-12-06 11:50:52,837 - Epoch: [197][ 420/ 1200] Overall Loss 0.118422 Objective Loss 0.118422 LR 0.000031 Time 0.022283 -2022-12-06 11:50:53,041 - Epoch: [197][ 430/ 1200] Overall Loss 0.118144 Objective Loss 0.118144 LR 0.000031 Time 0.022237 -2022-12-06 11:50:53,245 - Epoch: [197][ 440/ 1200] Overall Loss 0.118223 Objective Loss 0.118223 LR 0.000031 Time 0.022194 -2022-12-06 11:50:53,448 - Epoch: [197][ 450/ 1200] Overall Loss 0.118622 Objective Loss 0.118622 LR 0.000031 Time 0.022152 -2022-12-06 11:50:53,652 - Epoch: [197][ 460/ 1200] Overall Loss 0.118499 Objective Loss 0.118499 LR 0.000031 Time 0.022112 -2022-12-06 11:50:53,855 - Epoch: [197][ 470/ 1200] Overall Loss 0.118725 Objective Loss 0.118725 LR 0.000031 Time 0.022072 -2022-12-06 11:50:54,059 - Epoch: [197][ 480/ 1200] Overall Loss 0.119044 Objective Loss 0.119044 LR 0.000031 Time 0.022035 -2022-12-06 11:50:54,261 - Epoch: [197][ 490/ 1200] Overall Loss 0.119205 Objective Loss 0.119205 LR 0.000031 Time 0.021998 -2022-12-06 11:50:54,465 - Epoch: [197][ 500/ 1200] Overall Loss 0.119022 Objective Loss 0.119022 LR 0.000031 Time 0.021965 -2022-12-06 11:50:54,669 - Epoch: [197][ 510/ 1200] Overall Loss 0.119139 Objective Loss 0.119139 LR 0.000031 Time 0.021932 -2022-12-06 11:50:54,873 - Epoch: [197][ 520/ 1200] Overall Loss 0.119388 Objective Loss 0.119388 LR 0.000031 Time 0.021902 -2022-12-06 11:50:55,077 - Epoch: [197][ 530/ 1200] Overall Loss 0.119229 Objective Loss 0.119229 LR 0.000031 Time 0.021873 -2022-12-06 11:50:55,281 - Epoch: [197][ 540/ 1200] Overall Loss 0.119196 Objective Loss 0.119196 LR 0.000031 Time 0.021845 -2022-12-06 11:50:55,485 - Epoch: [197][ 550/ 1200] Overall Loss 0.118843 Objective Loss 0.118843 LR 0.000031 Time 0.021817 -2022-12-06 11:50:55,689 - Epoch: [197][ 560/ 1200] Overall Loss 0.119054 Objective Loss 0.119054 LR 0.000031 Time 0.021792 -2022-12-06 11:50:55,893 - Epoch: [197][ 570/ 1200] Overall Loss 0.118842 Objective Loss 0.118842 LR 0.000031 Time 0.021766 -2022-12-06 11:50:56,097 - Epoch: [197][ 580/ 1200] Overall Loss 0.119047 Objective Loss 0.119047 LR 0.000031 Time 0.021741 -2022-12-06 11:50:56,301 - Epoch: [197][ 590/ 1200] Overall Loss 0.118935 Objective Loss 0.118935 LR 0.000031 Time 0.021718 -2022-12-06 11:50:56,506 - Epoch: [197][ 600/ 1200] Overall Loss 0.119282 Objective Loss 0.119282 LR 0.000031 Time 0.021696 -2022-12-06 11:50:56,709 - Epoch: [197][ 610/ 1200] Overall Loss 0.119391 Objective Loss 0.119391 LR 0.000031 Time 0.021673 -2022-12-06 11:50:56,914 - Epoch: [197][ 620/ 1200] Overall Loss 0.119386 Objective Loss 0.119386 LR 0.000031 Time 0.021652 -2022-12-06 11:50:57,116 - Epoch: [197][ 630/ 1200] Overall Loss 0.119303 Objective Loss 0.119303 LR 0.000031 Time 0.021629 -2022-12-06 11:50:57,321 - Epoch: [197][ 640/ 1200] Overall Loss 0.119270 Objective Loss 0.119270 LR 0.000031 Time 0.021610 -2022-12-06 11:50:57,525 - Epoch: [197][ 650/ 1200] Overall Loss 0.119217 Objective Loss 0.119217 LR 0.000031 Time 0.021590 -2022-12-06 11:50:57,729 - Epoch: [197][ 660/ 1200] Overall Loss 0.118724 Objective Loss 0.118724 LR 0.000031 Time 0.021572 -2022-12-06 11:50:57,932 - Epoch: [197][ 670/ 1200] Overall Loss 0.118739 Objective Loss 0.118739 LR 0.000031 Time 0.021553 -2022-12-06 11:50:58,136 - Epoch: [197][ 680/ 1200] Overall Loss 0.118733 Objective Loss 0.118733 LR 0.000031 Time 0.021535 -2022-12-06 11:50:58,340 - Epoch: [197][ 690/ 1200] Overall Loss 0.118677 Objective Loss 0.118677 LR 0.000031 Time 0.021517 -2022-12-06 11:50:58,544 - Epoch: [197][ 700/ 1200] Overall Loss 0.118733 Objective Loss 0.118733 LR 0.000031 Time 0.021501 -2022-12-06 11:50:58,747 - Epoch: [197][ 710/ 1200] Overall Loss 0.118634 Objective Loss 0.118634 LR 0.000031 Time 0.021483 -2022-12-06 11:50:58,951 - Epoch: [197][ 720/ 1200] Overall Loss 0.118695 Objective Loss 0.118695 LR 0.000031 Time 0.021467 -2022-12-06 11:50:59,155 - Epoch: [197][ 730/ 1200] Overall Loss 0.118650 Objective Loss 0.118650 LR 0.000031 Time 0.021451 -2022-12-06 11:50:59,359 - Epoch: [197][ 740/ 1200] Overall Loss 0.118522 Objective Loss 0.118522 LR 0.000031 Time 0.021436 -2022-12-06 11:50:59,562 - Epoch: [197][ 750/ 1200] Overall Loss 0.118471 Objective Loss 0.118471 LR 0.000031 Time 0.021421 -2022-12-06 11:50:59,766 - Epoch: [197][ 760/ 1200] Overall Loss 0.118560 Objective Loss 0.118560 LR 0.000031 Time 0.021407 -2022-12-06 11:50:59,970 - Epoch: [197][ 770/ 1200] Overall Loss 0.118362 Objective Loss 0.118362 LR 0.000031 Time 0.021393 -2022-12-06 11:51:00,174 - Epoch: [197][ 780/ 1200] Overall Loss 0.118497 Objective Loss 0.118497 LR 0.000031 Time 0.021379 -2022-12-06 11:51:00,378 - Epoch: [197][ 790/ 1200] Overall Loss 0.118394 Objective Loss 0.118394 LR 0.000031 Time 0.021367 -2022-12-06 11:51:00,582 - Epoch: [197][ 800/ 1200] Overall Loss 0.118549 Objective Loss 0.118549 LR 0.000031 Time 0.021354 -2022-12-06 11:51:00,786 - Epoch: [197][ 810/ 1200] Overall Loss 0.118813 Objective Loss 0.118813 LR 0.000031 Time 0.021341 -2022-12-06 11:51:00,990 - Epoch: [197][ 820/ 1200] Overall Loss 0.118717 Objective Loss 0.118717 LR 0.000031 Time 0.021328 -2022-12-06 11:51:01,193 - Epoch: [197][ 830/ 1200] Overall Loss 0.118576 Objective Loss 0.118576 LR 0.000031 Time 0.021317 -2022-12-06 11:51:01,397 - Epoch: [197][ 840/ 1200] Overall Loss 0.118491 Objective Loss 0.118491 LR 0.000031 Time 0.021305 -2022-12-06 11:51:01,601 - Epoch: [197][ 850/ 1200] Overall Loss 0.118257 Objective Loss 0.118257 LR 0.000031 Time 0.021293 -2022-12-06 11:51:01,804 - Epoch: [197][ 860/ 1200] Overall Loss 0.118004 Objective Loss 0.118004 LR 0.000031 Time 0.021281 -2022-12-06 11:51:02,008 - Epoch: [197][ 870/ 1200] Overall Loss 0.118114 Objective Loss 0.118114 LR 0.000031 Time 0.021270 -2022-12-06 11:51:02,212 - Epoch: [197][ 880/ 1200] Overall Loss 0.118070 Objective Loss 0.118070 LR 0.000031 Time 0.021259 -2022-12-06 11:51:02,415 - Epoch: [197][ 890/ 1200] Overall Loss 0.118094 Objective Loss 0.118094 LR 0.000031 Time 0.021249 -2022-12-06 11:51:02,613 - Epoch: [197][ 900/ 1200] Overall Loss 0.118036 Objective Loss 0.118036 LR 0.000031 Time 0.021231 -2022-12-06 11:51:02,812 - Epoch: [197][ 910/ 1200] Overall Loss 0.118006 Objective Loss 0.118006 LR 0.000031 Time 0.021216 -2022-12-06 11:51:03,010 - Epoch: [197][ 920/ 1200] Overall Loss 0.118020 Objective Loss 0.118020 LR 0.000031 Time 0.021200 -2022-12-06 11:51:03,209 - Epoch: [197][ 930/ 1200] Overall Loss 0.117924 Objective Loss 0.117924 LR 0.000031 Time 0.021185 -2022-12-06 11:51:03,406 - Epoch: [197][ 940/ 1200] Overall Loss 0.117893 Objective Loss 0.117893 LR 0.000031 Time 0.021170 -2022-12-06 11:51:03,605 - Epoch: [197][ 950/ 1200] Overall Loss 0.117861 Objective Loss 0.117861 LR 0.000031 Time 0.021155 -2022-12-06 11:51:03,803 - Epoch: [197][ 960/ 1200] Overall Loss 0.118042 Objective Loss 0.118042 LR 0.000031 Time 0.021141 -2022-12-06 11:51:04,002 - Epoch: [197][ 970/ 1200] Overall Loss 0.118066 Objective Loss 0.118066 LR 0.000031 Time 0.021127 -2022-12-06 11:51:04,200 - Epoch: [197][ 980/ 1200] Overall Loss 0.117963 Objective Loss 0.117963 LR 0.000031 Time 0.021113 -2022-12-06 11:51:04,399 - Epoch: [197][ 990/ 1200] Overall Loss 0.117895 Objective Loss 0.117895 LR 0.000031 Time 0.021101 -2022-12-06 11:51:04,597 - Epoch: [197][ 1000/ 1200] Overall Loss 0.117879 Objective Loss 0.117879 LR 0.000031 Time 0.021087 -2022-12-06 11:51:04,797 - Epoch: [197][ 1010/ 1200] Overall Loss 0.117845 Objective Loss 0.117845 LR 0.000031 Time 0.021076 -2022-12-06 11:51:04,995 - Epoch: [197][ 1020/ 1200] Overall Loss 0.117832 Objective Loss 0.117832 LR 0.000031 Time 0.021063 -2022-12-06 11:51:05,194 - Epoch: [197][ 1030/ 1200] Overall Loss 0.117970 Objective Loss 0.117970 LR 0.000031 Time 0.021051 -2022-12-06 11:51:05,393 - Epoch: [197][ 1040/ 1200] Overall Loss 0.117944 Objective Loss 0.117944 LR 0.000031 Time 0.021039 -2022-12-06 11:51:05,592 - Epoch: [197][ 1050/ 1200] Overall Loss 0.117844 Objective Loss 0.117844 LR 0.000031 Time 0.021028 -2022-12-06 11:51:05,791 - Epoch: [197][ 1060/ 1200] Overall Loss 0.117831 Objective Loss 0.117831 LR 0.000031 Time 0.021016 -2022-12-06 11:51:05,990 - Epoch: [197][ 1070/ 1200] Overall Loss 0.117839 Objective Loss 0.117839 LR 0.000031 Time 0.021006 -2022-12-06 11:51:06,188 - Epoch: [197][ 1080/ 1200] Overall Loss 0.117886 Objective Loss 0.117886 LR 0.000031 Time 0.020994 -2022-12-06 11:51:06,388 - Epoch: [197][ 1090/ 1200] Overall Loss 0.117703 Objective Loss 0.117703 LR 0.000031 Time 0.020984 -2022-12-06 11:51:06,586 - Epoch: [197][ 1100/ 1200] Overall Loss 0.117735 Objective Loss 0.117735 LR 0.000031 Time 0.020973 -2022-12-06 11:51:06,785 - Epoch: [197][ 1110/ 1200] Overall Loss 0.117624 Objective Loss 0.117624 LR 0.000031 Time 0.020963 -2022-12-06 11:51:06,984 - Epoch: [197][ 1120/ 1200] Overall Loss 0.117596 Objective Loss 0.117596 LR 0.000031 Time 0.020953 -2022-12-06 11:51:07,183 - Epoch: [197][ 1130/ 1200] Overall Loss 0.117645 Objective Loss 0.117645 LR 0.000031 Time 0.020944 -2022-12-06 11:51:07,382 - Epoch: [197][ 1140/ 1200] Overall Loss 0.117637 Objective Loss 0.117637 LR 0.000031 Time 0.020933 -2022-12-06 11:51:07,581 - Epoch: [197][ 1150/ 1200] Overall Loss 0.117677 Objective Loss 0.117677 LR 0.000031 Time 0.020924 -2022-12-06 11:51:07,779 - Epoch: [197][ 1160/ 1200] Overall Loss 0.117713 Objective Loss 0.117713 LR 0.000031 Time 0.020914 -2022-12-06 11:51:07,978 - Epoch: [197][ 1170/ 1200] Overall Loss 0.117839 Objective Loss 0.117839 LR 0.000031 Time 0.020905 -2022-12-06 11:51:08,176 - Epoch: [197][ 1180/ 1200] Overall Loss 0.117814 Objective Loss 0.117814 LR 0.000031 Time 0.020895 -2022-12-06 11:51:08,376 - Epoch: [197][ 1190/ 1200] Overall Loss 0.117909 Objective Loss 0.117909 LR 0.000031 Time 0.020887 -2022-12-06 11:51:08,605 - Epoch: [197][ 1200/ 1200] Overall Loss 0.117853 Objective Loss 0.117853 Top1 91.841004 Top5 99.163180 LR 0.000031 Time 0.020903 -2022-12-06 11:51:08,693 - --- validate (epoch=197)----------- -2022-12-06 11:51:08,694 - 34129 samples (256 per mini-batch) -2022-12-06 11:51:09,142 - Epoch: [197][ 10/ 134] Loss 0.213052 Top1 89.101562 Top5 98.867188 -2022-12-06 11:51:09,275 - Epoch: [197][ 20/ 134] Loss 0.221568 Top1 88.808594 Top5 98.730469 -2022-12-06 11:51:09,408 - Epoch: [197][ 30/ 134] Loss 0.231992 Top1 88.619792 Top5 98.684896 -2022-12-06 11:51:09,541 - Epoch: [197][ 40/ 134] Loss 0.225642 Top1 88.837891 Top5 98.681641 -2022-12-06 11:51:09,671 - Epoch: [197][ 50/ 134] Loss 0.219989 Top1 88.937500 Top5 98.750000 -2022-12-06 11:51:09,803 - Epoch: [197][ 60/ 134] Loss 0.222538 Top1 88.873698 Top5 98.730469 -2022-12-06 11:51:09,933 - Epoch: [197][ 70/ 134] Loss 0.220643 Top1 88.895089 Top5 98.761161 -2022-12-06 11:51:10,065 - Epoch: [197][ 80/ 134] Loss 0.222050 Top1 88.872070 Top5 98.774414 -2022-12-06 11:51:10,198 - Epoch: [197][ 90/ 134] Loss 0.223594 Top1 88.893229 Top5 98.784722 -2022-12-06 11:51:10,325 - Epoch: [197][ 100/ 134] Loss 0.222346 Top1 88.921875 Top5 98.750000 -2022-12-06 11:51:10,455 - Epoch: [197][ 110/ 134] Loss 0.223170 Top1 88.817472 Top5 98.739347 -2022-12-06 11:51:10,587 - Epoch: [197][ 120/ 134] Loss 0.221288 Top1 88.837891 Top5 98.733724 -2022-12-06 11:51:10,720 - Epoch: [197][ 130/ 134] Loss 0.221878 Top1 88.765024 Top5 98.725962 -2022-12-06 11:51:10,759 - Epoch: [197][ 134/ 134] Loss 0.220901 Top1 88.763222 Top5 98.716634 -2022-12-06 11:51:10,853 - ==> Top1: 88.763 Top5: 98.717 Loss: 0.221 - -2022-12-06 11:51:10,854 - ==> Confusion: -[[ 917 1 1 2 3 6 1 0 6 42 0 1 0 3 6 1 2 0 2 0 2] - [ 0 943 4 2 6 21 2 11 2 3 3 4 0 1 0 0 2 1 11 3 8] - [ 4 2 1024 11 4 1 14 12 0 2 4 2 1 2 2 1 1 1 3 2 10] - [ 5 1 12 963 1 1 1 0 0 1 7 0 3 2 5 0 1 1 10 0 6] - [ 7 2 2 0 964 2 1 2 1 7 2 3 0 2 8 5 5 3 0 0 4] - [ 1 6 1 4 4 986 2 16 3 2 0 10 3 17 2 1 1 0 0 7 3] - [ 2 1 7 3 0 2 1082 3 0 0 0 1 0 1 0 3 0 2 1 7 3] - [ 1 7 5 2 1 26 9 965 0 0 1 5 1 1 0 0 1 0 15 10 4] - [ 3 1 0 0 1 1 2 1 985 34 10 1 1 6 9 1 2 0 1 1 4] - [ 42 1 1 0 4 1 0 2 22 903 2 1 0 13 3 0 0 1 0 0 5] - [ 1 3 3 2 1 0 0 2 6 1 969 0 0 10 5 1 0 0 6 2 7] - [ 1 0 3 0 1 7 3 2 0 0 0 987 19 4 0 7 2 4 0 8 3] - [ 2 1 1 2 0 1 0 1 0 1 0 26 910 1 1 6 1 6 1 3 5] - [ 2 1 0 0 1 6 0 2 8 9 5 4 3 969 0 2 3 0 0 1 7] - [ 6 5 3 12 2 4 0 0 11 2 0 3 1 3 1067 0 0 1 4 2 4] - [ 1 0 0 3 3 0 3 0 0 0 1 8 2 3 0 1000 4 9 1 3 2] - [ 1 0 1 2 1 0 0 0 1 0 0 0 2 2 0 11 1035 2 1 5 8] - [ 3 0 1 1 2 1 0 0 0 4 0 4 14 2 1 15 0 985 0 0 3] - [ 1 2 4 6 0 3 0 23 4 1 3 1 3 0 5 1 0 2 943 0 6] - [ 1 2 1 1 1 4 5 4 0 1 3 10 4 6 0 4 5 2 0 1019 7] - [ 106 164 150 99 96 145 79 128 71 71 131 82 274 231 116 91 136 73 126 185 10672]] - -2022-12-06 11:51:11,436 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:51:11,436 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:51:11,442 - - -2022-12-06 11:51:11,442 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:51:12,490 - Epoch: [198][ 10/ 1200] Overall Loss 0.115583 Objective Loss 0.115583 LR 0.000031 Time 0.104755 -2022-12-06 11:51:12,684 - Epoch: [198][ 20/ 1200] Overall Loss 0.118249 Objective Loss 0.118249 LR 0.000031 Time 0.062032 -2022-12-06 11:51:12,876 - Epoch: [198][ 30/ 1200] Overall Loss 0.123164 Objective Loss 0.123164 LR 0.000031 Time 0.047748 -2022-12-06 11:51:13,069 - Epoch: [198][ 40/ 1200] Overall Loss 0.123762 Objective Loss 0.123762 LR 0.000031 Time 0.040621 -2022-12-06 11:51:13,263 - Epoch: [198][ 50/ 1200] Overall Loss 0.120236 Objective Loss 0.120236 LR 0.000031 Time 0.036359 -2022-12-06 11:51:13,455 - Epoch: [198][ 60/ 1200] Overall Loss 0.120233 Objective Loss 0.120233 LR 0.000031 Time 0.033504 -2022-12-06 11:51:13,649 - Epoch: [198][ 70/ 1200] Overall Loss 0.118395 Objective Loss 0.118395 LR 0.000031 Time 0.031472 -2022-12-06 11:51:13,842 - Epoch: [198][ 80/ 1200] Overall Loss 0.119796 Objective Loss 0.119796 LR 0.000031 Time 0.029939 -2022-12-06 11:51:14,035 - Epoch: [198][ 90/ 1200] Overall Loss 0.118203 Objective Loss 0.118203 LR 0.000031 Time 0.028753 -2022-12-06 11:51:14,227 - Epoch: [198][ 100/ 1200] Overall Loss 0.116934 Objective Loss 0.116934 LR 0.000031 Time 0.027795 -2022-12-06 11:51:14,420 - Epoch: [198][ 110/ 1200] Overall Loss 0.119302 Objective Loss 0.119302 LR 0.000031 Time 0.027018 -2022-12-06 11:51:14,611 - Epoch: [198][ 120/ 1200] Overall Loss 0.118945 Objective Loss 0.118945 LR 0.000031 Time 0.026355 -2022-12-06 11:51:14,804 - Epoch: [198][ 130/ 1200] Overall Loss 0.118130 Objective Loss 0.118130 LR 0.000031 Time 0.025810 -2022-12-06 11:51:14,997 - Epoch: [198][ 140/ 1200] Overall Loss 0.118722 Objective Loss 0.118722 LR 0.000031 Time 0.025340 -2022-12-06 11:51:15,190 - Epoch: [198][ 150/ 1200] Overall Loss 0.119215 Objective Loss 0.119215 LR 0.000031 Time 0.024929 -2022-12-06 11:51:15,382 - Epoch: [198][ 160/ 1200] Overall Loss 0.119327 Objective Loss 0.119327 LR 0.000031 Time 0.024568 -2022-12-06 11:51:15,574 - Epoch: [198][ 170/ 1200] Overall Loss 0.118515 Objective Loss 0.118515 LR 0.000031 Time 0.024252 -2022-12-06 11:51:15,767 - Epoch: [198][ 180/ 1200] Overall Loss 0.118787 Objective Loss 0.118787 LR 0.000031 Time 0.023972 -2022-12-06 11:51:15,960 - Epoch: [198][ 190/ 1200] Overall Loss 0.118771 Objective Loss 0.118771 LR 0.000031 Time 0.023723 -2022-12-06 11:51:16,153 - Epoch: [198][ 200/ 1200] Overall Loss 0.118644 Objective Loss 0.118644 LR 0.000031 Time 0.023499 -2022-12-06 11:51:16,346 - Epoch: [198][ 210/ 1200] Overall Loss 0.118777 Objective Loss 0.118777 LR 0.000031 Time 0.023297 -2022-12-06 11:51:16,538 - Epoch: [198][ 220/ 1200] Overall Loss 0.118295 Objective Loss 0.118295 LR 0.000031 Time 0.023110 -2022-12-06 11:51:16,732 - Epoch: [198][ 230/ 1200] Overall Loss 0.117913 Objective Loss 0.117913 LR 0.000031 Time 0.022943 -2022-12-06 11:51:16,924 - Epoch: [198][ 240/ 1200] Overall Loss 0.117505 Objective Loss 0.117505 LR 0.000031 Time 0.022787 -2022-12-06 11:51:17,117 - Epoch: [198][ 250/ 1200] Overall Loss 0.117686 Objective Loss 0.117686 LR 0.000031 Time 0.022645 -2022-12-06 11:51:17,309 - Epoch: [198][ 260/ 1200] Overall Loss 0.117551 Objective Loss 0.117551 LR 0.000031 Time 0.022511 -2022-12-06 11:51:17,502 - Epoch: [198][ 270/ 1200] Overall Loss 0.117140 Objective Loss 0.117140 LR 0.000031 Time 0.022390 -2022-12-06 11:51:17,695 - Epoch: [198][ 280/ 1200] Overall Loss 0.117456 Objective Loss 0.117456 LR 0.000031 Time 0.022277 -2022-12-06 11:51:17,888 - Epoch: [198][ 290/ 1200] Overall Loss 0.117324 Objective Loss 0.117324 LR 0.000031 Time 0.022173 -2022-12-06 11:51:18,082 - Epoch: [198][ 300/ 1200] Overall Loss 0.117381 Objective Loss 0.117381 LR 0.000031 Time 0.022076 -2022-12-06 11:51:18,274 - Epoch: [198][ 310/ 1200] Overall Loss 0.117315 Objective Loss 0.117315 LR 0.000031 Time 0.021984 -2022-12-06 11:51:18,466 - Epoch: [198][ 320/ 1200] Overall Loss 0.117354 Objective Loss 0.117354 LR 0.000031 Time 0.021896 -2022-12-06 11:51:18,659 - Epoch: [198][ 330/ 1200] Overall Loss 0.118188 Objective Loss 0.118188 LR 0.000031 Time 0.021815 -2022-12-06 11:51:18,851 - Epoch: [198][ 340/ 1200] Overall Loss 0.118390 Objective Loss 0.118390 LR 0.000031 Time 0.021736 -2022-12-06 11:51:19,044 - Epoch: [198][ 350/ 1200] Overall Loss 0.118314 Objective Loss 0.118314 LR 0.000031 Time 0.021665 -2022-12-06 11:51:19,236 - Epoch: [198][ 360/ 1200] Overall Loss 0.118342 Objective Loss 0.118342 LR 0.000031 Time 0.021595 -2022-12-06 11:51:19,429 - Epoch: [198][ 370/ 1200] Overall Loss 0.118061 Objective Loss 0.118061 LR 0.000031 Time 0.021532 -2022-12-06 11:51:19,621 - Epoch: [198][ 380/ 1200] Overall Loss 0.117993 Objective Loss 0.117993 LR 0.000031 Time 0.021469 -2022-12-06 11:51:19,814 - Epoch: [198][ 390/ 1200] Overall Loss 0.117705 Objective Loss 0.117705 LR 0.000031 Time 0.021410 -2022-12-06 11:51:20,006 - Epoch: [198][ 400/ 1200] Overall Loss 0.117575 Objective Loss 0.117575 LR 0.000031 Time 0.021354 -2022-12-06 11:51:20,199 - Epoch: [198][ 410/ 1200] Overall Loss 0.117868 Objective Loss 0.117868 LR 0.000031 Time 0.021303 -2022-12-06 11:51:20,391 - Epoch: [198][ 420/ 1200] Overall Loss 0.117551 Objective Loss 0.117551 LR 0.000031 Time 0.021251 -2022-12-06 11:51:20,584 - Epoch: [198][ 430/ 1200] Overall Loss 0.117632 Objective Loss 0.117632 LR 0.000031 Time 0.021205 -2022-12-06 11:51:20,776 - Epoch: [198][ 440/ 1200] Overall Loss 0.117367 Objective Loss 0.117367 LR 0.000031 Time 0.021159 -2022-12-06 11:51:20,969 - Epoch: [198][ 450/ 1200] Overall Loss 0.117367 Objective Loss 0.117367 LR 0.000031 Time 0.021115 -2022-12-06 11:51:21,161 - Epoch: [198][ 460/ 1200] Overall Loss 0.117455 Objective Loss 0.117455 LR 0.000031 Time 0.021073 -2022-12-06 11:51:21,355 - Epoch: [198][ 470/ 1200] Overall Loss 0.117455 Objective Loss 0.117455 LR 0.000031 Time 0.021035 -2022-12-06 11:51:21,547 - Epoch: [198][ 480/ 1200] Overall Loss 0.117438 Objective Loss 0.117438 LR 0.000031 Time 0.020997 -2022-12-06 11:51:21,740 - Epoch: [198][ 490/ 1200] Overall Loss 0.117216 Objective Loss 0.117216 LR 0.000031 Time 0.020962 -2022-12-06 11:51:21,933 - Epoch: [198][ 500/ 1200] Overall Loss 0.117164 Objective Loss 0.117164 LR 0.000031 Time 0.020927 -2022-12-06 11:51:22,126 - Epoch: [198][ 510/ 1200] Overall Loss 0.117134 Objective Loss 0.117134 LR 0.000031 Time 0.020894 -2022-12-06 11:51:22,319 - Epoch: [198][ 520/ 1200] Overall Loss 0.117129 Objective Loss 0.117129 LR 0.000031 Time 0.020862 -2022-12-06 11:51:22,512 - Epoch: [198][ 530/ 1200] Overall Loss 0.116922 Objective Loss 0.116922 LR 0.000031 Time 0.020830 -2022-12-06 11:51:22,704 - Epoch: [198][ 540/ 1200] Overall Loss 0.116921 Objective Loss 0.116921 LR 0.000031 Time 0.020799 -2022-12-06 11:51:22,896 - Epoch: [198][ 550/ 1200] Overall Loss 0.117202 Objective Loss 0.117202 LR 0.000031 Time 0.020770 -2022-12-06 11:51:23,089 - Epoch: [198][ 560/ 1200] Overall Loss 0.117325 Objective Loss 0.117325 LR 0.000031 Time 0.020742 -2022-12-06 11:51:23,282 - Epoch: [198][ 570/ 1200] Overall Loss 0.117275 Objective Loss 0.117275 LR 0.000031 Time 0.020716 -2022-12-06 11:51:23,474 - Epoch: [198][ 580/ 1200] Overall Loss 0.117505 Objective Loss 0.117505 LR 0.000031 Time 0.020688 -2022-12-06 11:51:23,667 - Epoch: [198][ 590/ 1200] Overall Loss 0.117449 Objective Loss 0.117449 LR 0.000031 Time 0.020664 -2022-12-06 11:51:23,859 - Epoch: [198][ 600/ 1200] Overall Loss 0.117469 Objective Loss 0.117469 LR 0.000031 Time 0.020639 -2022-12-06 11:51:24,052 - Epoch: [198][ 610/ 1200] Overall Loss 0.117232 Objective Loss 0.117232 LR 0.000031 Time 0.020616 -2022-12-06 11:51:24,244 - Epoch: [198][ 620/ 1200] Overall Loss 0.117435 Objective Loss 0.117435 LR 0.000031 Time 0.020593 -2022-12-06 11:51:24,437 - Epoch: [198][ 630/ 1200] Overall Loss 0.117332 Objective Loss 0.117332 LR 0.000031 Time 0.020572 -2022-12-06 11:51:24,630 - Epoch: [198][ 640/ 1200] Overall Loss 0.117407 Objective Loss 0.117407 LR 0.000031 Time 0.020550 -2022-12-06 11:51:24,823 - Epoch: [198][ 650/ 1200] Overall Loss 0.117223 Objective Loss 0.117223 LR 0.000031 Time 0.020531 -2022-12-06 11:51:25,015 - Epoch: [198][ 660/ 1200] Overall Loss 0.117208 Objective Loss 0.117208 LR 0.000031 Time 0.020510 -2022-12-06 11:51:25,209 - Epoch: [198][ 670/ 1200] Overall Loss 0.117182 Objective Loss 0.117182 LR 0.000031 Time 0.020492 -2022-12-06 11:51:25,401 - Epoch: [198][ 680/ 1200] Overall Loss 0.117763 Objective Loss 0.117763 LR 0.000031 Time 0.020472 -2022-12-06 11:51:25,594 - Epoch: [198][ 690/ 1200] Overall Loss 0.117912 Objective Loss 0.117912 LR 0.000031 Time 0.020455 -2022-12-06 11:51:25,787 - Epoch: [198][ 700/ 1200] Overall Loss 0.117793 Objective Loss 0.117793 LR 0.000031 Time 0.020437 -2022-12-06 11:51:25,979 - Epoch: [198][ 710/ 1200] Overall Loss 0.117728 Objective Loss 0.117728 LR 0.000031 Time 0.020420 -2022-12-06 11:51:26,172 - Epoch: [198][ 720/ 1200] Overall Loss 0.117759 Objective Loss 0.117759 LR 0.000031 Time 0.020403 -2022-12-06 11:51:26,365 - Epoch: [198][ 730/ 1200] Overall Loss 0.117643 Objective Loss 0.117643 LR 0.000031 Time 0.020386 -2022-12-06 11:51:26,557 - Epoch: [198][ 740/ 1200] Overall Loss 0.117798 Objective Loss 0.117798 LR 0.000031 Time 0.020370 -2022-12-06 11:51:26,750 - Epoch: [198][ 750/ 1200] Overall Loss 0.117833 Objective Loss 0.117833 LR 0.000031 Time 0.020355 -2022-12-06 11:51:26,942 - Epoch: [198][ 760/ 1200] Overall Loss 0.117846 Objective Loss 0.117846 LR 0.000031 Time 0.020339 -2022-12-06 11:51:27,135 - Epoch: [198][ 770/ 1200] Overall Loss 0.117946 Objective Loss 0.117946 LR 0.000031 Time 0.020325 -2022-12-06 11:51:27,327 - Epoch: [198][ 780/ 1200] Overall Loss 0.117958 Objective Loss 0.117958 LR 0.000031 Time 0.020310 -2022-12-06 11:51:27,520 - Epoch: [198][ 790/ 1200] Overall Loss 0.118027 Objective Loss 0.118027 LR 0.000031 Time 0.020297 -2022-12-06 11:51:27,713 - Epoch: [198][ 800/ 1200] Overall Loss 0.117919 Objective Loss 0.117919 LR 0.000031 Time 0.020284 -2022-12-06 11:51:27,906 - Epoch: [198][ 810/ 1200] Overall Loss 0.117894 Objective Loss 0.117894 LR 0.000031 Time 0.020271 -2022-12-06 11:51:28,099 - Epoch: [198][ 820/ 1200] Overall Loss 0.117700 Objective Loss 0.117700 LR 0.000031 Time 0.020258 -2022-12-06 11:51:28,292 - Epoch: [198][ 830/ 1200] Overall Loss 0.117802 Objective Loss 0.117802 LR 0.000031 Time 0.020246 -2022-12-06 11:51:28,485 - Epoch: [198][ 840/ 1200] Overall Loss 0.117698 Objective Loss 0.117698 LR 0.000031 Time 0.020234 -2022-12-06 11:51:28,678 - Epoch: [198][ 850/ 1200] Overall Loss 0.117497 Objective Loss 0.117497 LR 0.000031 Time 0.020223 -2022-12-06 11:51:28,871 - Epoch: [198][ 860/ 1200] Overall Loss 0.117500 Objective Loss 0.117500 LR 0.000031 Time 0.020210 -2022-12-06 11:51:29,063 - Epoch: [198][ 870/ 1200] Overall Loss 0.117488 Objective Loss 0.117488 LR 0.000031 Time 0.020198 -2022-12-06 11:51:29,256 - Epoch: [198][ 880/ 1200] Overall Loss 0.117540 Objective Loss 0.117540 LR 0.000031 Time 0.020188 -2022-12-06 11:51:29,449 - Epoch: [198][ 890/ 1200] Overall Loss 0.117515 Objective Loss 0.117515 LR 0.000031 Time 0.020177 -2022-12-06 11:51:29,642 - Epoch: [198][ 900/ 1200] Overall Loss 0.117664 Objective Loss 0.117664 LR 0.000031 Time 0.020167 -2022-12-06 11:51:29,835 - Epoch: [198][ 910/ 1200] Overall Loss 0.117589 Objective Loss 0.117589 LR 0.000031 Time 0.020157 -2022-12-06 11:51:30,028 - Epoch: [198][ 920/ 1200] Overall Loss 0.117604 Objective Loss 0.117604 LR 0.000031 Time 0.020147 -2022-12-06 11:51:30,222 - Epoch: [198][ 930/ 1200] Overall Loss 0.117715 Objective Loss 0.117715 LR 0.000031 Time 0.020138 -2022-12-06 11:51:30,415 - Epoch: [198][ 940/ 1200] Overall Loss 0.117652 Objective Loss 0.117652 LR 0.000031 Time 0.020128 -2022-12-06 11:51:30,608 - Epoch: [198][ 950/ 1200] Overall Loss 0.117649 Objective Loss 0.117649 LR 0.000031 Time 0.020120 -2022-12-06 11:51:30,802 - Epoch: [198][ 960/ 1200] Overall Loss 0.117621 Objective Loss 0.117621 LR 0.000031 Time 0.020111 -2022-12-06 11:51:30,995 - Epoch: [198][ 970/ 1200] Overall Loss 0.117620 Objective Loss 0.117620 LR 0.000031 Time 0.020103 -2022-12-06 11:51:31,188 - Epoch: [198][ 980/ 1200] Overall Loss 0.117608 Objective Loss 0.117608 LR 0.000031 Time 0.020094 -2022-12-06 11:51:31,382 - Epoch: [198][ 990/ 1200] Overall Loss 0.117582 Objective Loss 0.117582 LR 0.000031 Time 0.020086 -2022-12-06 11:51:31,575 - Epoch: [198][ 1000/ 1200] Overall Loss 0.117724 Objective Loss 0.117724 LR 0.000031 Time 0.020078 -2022-12-06 11:51:31,768 - Epoch: [198][ 1010/ 1200] Overall Loss 0.117739 Objective Loss 0.117739 LR 0.000031 Time 0.020069 -2022-12-06 11:51:31,962 - Epoch: [198][ 1020/ 1200] Overall Loss 0.117841 Objective Loss 0.117841 LR 0.000031 Time 0.020062 -2022-12-06 11:51:32,155 - Epoch: [198][ 1030/ 1200] Overall Loss 0.117852 Objective Loss 0.117852 LR 0.000031 Time 0.020055 -2022-12-06 11:51:32,348 - Epoch: [198][ 1040/ 1200] Overall Loss 0.117761 Objective Loss 0.117761 LR 0.000031 Time 0.020047 -2022-12-06 11:51:32,542 - Epoch: [198][ 1050/ 1200] Overall Loss 0.117880 Objective Loss 0.117880 LR 0.000031 Time 0.020040 -2022-12-06 11:51:32,734 - Epoch: [198][ 1060/ 1200] Overall Loss 0.117886 Objective Loss 0.117886 LR 0.000031 Time 0.020032 -2022-12-06 11:51:32,927 - Epoch: [198][ 1070/ 1200] Overall Loss 0.117780 Objective Loss 0.117780 LR 0.000031 Time 0.020024 -2022-12-06 11:51:33,119 - Epoch: [198][ 1080/ 1200] Overall Loss 0.117720 Objective Loss 0.117720 LR 0.000031 Time 0.020016 -2022-12-06 11:51:33,311 - Epoch: [198][ 1090/ 1200] Overall Loss 0.117801 Objective Loss 0.117801 LR 0.000031 Time 0.020008 -2022-12-06 11:51:33,502 - Epoch: [198][ 1100/ 1200] Overall Loss 0.117805 Objective Loss 0.117805 LR 0.000031 Time 0.019999 -2022-12-06 11:51:33,694 - Epoch: [198][ 1110/ 1200] Overall Loss 0.117855 Objective Loss 0.117855 LR 0.000031 Time 0.019991 -2022-12-06 11:51:33,885 - Epoch: [198][ 1120/ 1200] Overall Loss 0.117752 Objective Loss 0.117752 LR 0.000031 Time 0.019984 -2022-12-06 11:51:34,077 - Epoch: [198][ 1130/ 1200] Overall Loss 0.117765 Objective Loss 0.117765 LR 0.000031 Time 0.019976 -2022-12-06 11:51:34,269 - Epoch: [198][ 1140/ 1200] Overall Loss 0.117700 Objective Loss 0.117700 LR 0.000031 Time 0.019968 -2022-12-06 11:51:34,461 - Epoch: [198][ 1150/ 1200] Overall Loss 0.117792 Objective Loss 0.117792 LR 0.000031 Time 0.019961 -2022-12-06 11:51:34,652 - Epoch: [198][ 1160/ 1200] Overall Loss 0.117733 Objective Loss 0.117733 LR 0.000031 Time 0.019954 -2022-12-06 11:51:34,844 - Epoch: [198][ 1170/ 1200] Overall Loss 0.117824 Objective Loss 0.117824 LR 0.000031 Time 0.019947 -2022-12-06 11:51:35,035 - Epoch: [198][ 1180/ 1200] Overall Loss 0.117826 Objective Loss 0.117826 LR 0.000031 Time 0.019939 -2022-12-06 11:51:35,227 - Epoch: [198][ 1190/ 1200] Overall Loss 0.117859 Objective Loss 0.117859 LR 0.000031 Time 0.019933 -2022-12-06 11:51:35,452 - Epoch: [198][ 1200/ 1200] Overall Loss 0.117902 Objective Loss 0.117902 Top1 89.539749 Top5 99.372385 LR 0.000031 Time 0.019953 -2022-12-06 11:51:35,540 - --- validate (epoch=198)----------- -2022-12-06 11:51:35,540 - 34129 samples (256 per mini-batch) -2022-12-06 11:51:35,989 - Epoch: [198][ 10/ 134] Loss 0.185815 Top1 88.984375 Top5 98.906250 -2022-12-06 11:51:36,119 - Epoch: [198][ 20/ 134] Loss 0.202751 Top1 88.476562 Top5 98.652344 -2022-12-06 11:51:36,246 - Epoch: [198][ 30/ 134] Loss 0.210629 Top1 88.606771 Top5 98.632812 -2022-12-06 11:51:36,372 - Epoch: [198][ 40/ 134] Loss 0.208311 Top1 88.818359 Top5 98.691406 -2022-12-06 11:51:36,498 - Epoch: [198][ 50/ 134] Loss 0.209040 Top1 88.718750 Top5 98.695312 -2022-12-06 11:51:36,623 - Epoch: [198][ 60/ 134] Loss 0.211565 Top1 88.678385 Top5 98.684896 -2022-12-06 11:51:36,749 - Epoch: [198][ 70/ 134] Loss 0.212962 Top1 88.621652 Top5 98.660714 -2022-12-06 11:51:36,875 - Epoch: [198][ 80/ 134] Loss 0.215403 Top1 88.632812 Top5 98.676758 -2022-12-06 11:51:36,999 - Epoch: [198][ 90/ 134] Loss 0.217470 Top1 88.580729 Top5 98.676215 -2022-12-06 11:51:37,124 - Epoch: [198][ 100/ 134] Loss 0.218605 Top1 88.550781 Top5 98.675781 -2022-12-06 11:51:37,248 - Epoch: [198][ 110/ 134] Loss 0.218740 Top1 88.611506 Top5 98.678977 -2022-12-06 11:51:37,374 - Epoch: [198][ 120/ 134] Loss 0.220534 Top1 88.554688 Top5 98.668620 -2022-12-06 11:51:37,504 - Epoch: [198][ 130/ 134] Loss 0.221554 Top1 88.524639 Top5 98.683894 -2022-12-06 11:51:37,540 - Epoch: [198][ 134/ 134] Loss 0.222554 Top1 88.520027 Top5 98.672683 -2022-12-06 11:51:37,628 - ==> Top1: 88.520 Top5: 98.673 Loss: 0.223 - -2022-12-06 11:51:37,628 - ==> Confusion: -[[ 928 1 2 3 4 6 0 0 5 32 0 1 0 2 7 1 2 0 1 0 1] - [ 1 944 0 2 6 19 4 12 1 2 1 5 0 1 0 1 2 2 11 4 9] - [ 4 2 1022 12 4 3 14 10 0 3 3 2 2 1 2 2 0 2 3 3 9] - [ 5 1 13 957 1 3 0 0 0 0 10 0 2 0 10 0 0 1 12 0 5] - [ 8 3 1 0 963 2 1 2 1 6 1 3 0 2 10 5 5 3 1 0 3] - [ 0 10 0 4 5 992 2 13 1 2 0 11 4 11 2 1 2 1 0 2 6] - [ 1 1 6 5 0 2 1082 2 0 0 0 1 0 1 0 3 0 3 1 8 2] - [ 2 7 4 2 0 25 11 956 0 1 1 8 0 1 0 1 1 0 22 8 4] - [ 5 3 0 1 0 2 1 1 986 36 11 1 1 4 8 1 2 0 0 1 0] - [ 53 1 0 0 3 2 0 2 23 892 1 1 0 12 3 1 0 1 1 1 4] - [ 0 2 3 3 1 1 1 2 7 2 969 0 0 10 3 1 0 0 6 2 6] - [ 3 1 1 0 1 7 3 2 1 0 1 982 22 3 0 4 3 5 0 6 6] - [ 1 1 0 2 0 2 0 1 1 1 0 25 912 0 0 7 1 7 0 2 6] - [ 2 1 1 0 1 7 0 1 9 8 4 3 4 971 0 1 3 0 0 1 6] - [ 6 6 2 12 2 3 0 0 12 0 0 3 1 3 1069 0 0 1 5 1 4] - [ 1 0 1 2 2 0 2 0 1 0 1 5 2 3 0 999 5 11 0 5 3] - [ 1 0 0 2 1 0 1 0 1 0 0 1 3 2 0 12 1032 1 0 6 9] - [ 3 0 1 3 1 1 0 0 0 5 0 6 15 2 1 11 0 985 0 0 2] - [ 1 2 4 6 0 2 0 20 2 1 4 4 3 1 6 0 0 3 944 0 5] - [ 1 4 1 1 1 4 5 3 0 1 2 15 4 6 0 5 5 1 2 1013 6] - [ 111 182 145 107 98 154 69 116 73 63 147 87 270 236 133 94 128 71 136 195 10611]] - -2022-12-06 11:51:38,198 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:51:38,198 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:51:38,204 - - -2022-12-06 11:51:38,204 - Training epoch: 307166 samples (256 per mini-batch) -2022-12-06 11:51:39,134 - Epoch: [199][ 10/ 1200] Overall Loss 0.113605 Objective Loss 0.113605 LR 0.000031 Time 0.092899 -2022-12-06 11:51:39,328 - Epoch: [199][ 20/ 1200] Overall Loss 0.111090 Objective Loss 0.111090 LR 0.000031 Time 0.056139 -2022-12-06 11:51:39,520 - Epoch: [199][ 30/ 1200] Overall Loss 0.107720 Objective Loss 0.107720 LR 0.000031 Time 0.043784 -2022-12-06 11:51:39,711 - Epoch: [199][ 40/ 1200] Overall Loss 0.113254 Objective Loss 0.113254 LR 0.000031 Time 0.037597 -2022-12-06 11:51:39,901 - Epoch: [199][ 50/ 1200] Overall Loss 0.116971 Objective Loss 0.116971 LR 0.000031 Time 0.033869 -2022-12-06 11:51:40,091 - Epoch: [199][ 60/ 1200] Overall Loss 0.115854 Objective Loss 0.115854 LR 0.000031 Time 0.031381 -2022-12-06 11:51:40,280 - Epoch: [199][ 70/ 1200] Overall Loss 0.117077 Objective Loss 0.117077 LR 0.000031 Time 0.029595 -2022-12-06 11:51:40,470 - Epoch: [199][ 80/ 1200] Overall Loss 0.116888 Objective Loss 0.116888 LR 0.000031 Time 0.028267 -2022-12-06 11:51:40,660 - Epoch: [199][ 90/ 1200] Overall Loss 0.117166 Objective Loss 0.117166 LR 0.000031 Time 0.027233 -2022-12-06 11:51:40,851 - Epoch: [199][ 100/ 1200] Overall Loss 0.116294 Objective Loss 0.116294 LR 0.000031 Time 0.026406 -2022-12-06 11:51:41,041 - Epoch: [199][ 110/ 1200] Overall Loss 0.115311 Objective Loss 0.115311 LR 0.000031 Time 0.025728 -2022-12-06 11:51:41,230 - Epoch: [199][ 120/ 1200] Overall Loss 0.115650 Objective Loss 0.115650 LR 0.000031 Time 0.025160 -2022-12-06 11:51:41,421 - Epoch: [199][ 130/ 1200] Overall Loss 0.116370 Objective Loss 0.116370 LR 0.000031 Time 0.024690 -2022-12-06 11:51:41,611 - Epoch: [199][ 140/ 1200] Overall Loss 0.117509 Objective Loss 0.117509 LR 0.000031 Time 0.024278 -2022-12-06 11:51:41,801 - Epoch: [199][ 150/ 1200] Overall Loss 0.117733 Objective Loss 0.117733 LR 0.000031 Time 0.023922 -2022-12-06 11:51:41,991 - Epoch: [199][ 160/ 1200] Overall Loss 0.118668 Objective Loss 0.118668 LR 0.000031 Time 0.023610 -2022-12-06 11:51:42,180 - Epoch: [199][ 170/ 1200] Overall Loss 0.117944 Objective Loss 0.117944 LR 0.000031 Time 0.023332 -2022-12-06 11:51:42,369 - Epoch: [199][ 180/ 1200] Overall Loss 0.117365 Objective Loss 0.117365 LR 0.000031 Time 0.023084 -2022-12-06 11:51:42,559 - Epoch: [199][ 190/ 1200] Overall Loss 0.117854 Objective Loss 0.117854 LR 0.000031 Time 0.022867 -2022-12-06 11:51:42,750 - Epoch: [199][ 200/ 1200] Overall Loss 0.117723 Objective Loss 0.117723 LR 0.000031 Time 0.022671 -2022-12-06 11:51:42,939 - Epoch: [199][ 210/ 1200] Overall Loss 0.117791 Objective Loss 0.117791 LR 0.000031 Time 0.022492 -2022-12-06 11:51:43,129 - Epoch: [199][ 220/ 1200] Overall Loss 0.117418 Objective Loss 0.117418 LR 0.000031 Time 0.022330 -2022-12-06 11:51:43,319 - Epoch: [199][ 230/ 1200] Overall Loss 0.117751 Objective Loss 0.117751 LR 0.000031 Time 0.022184 -2022-12-06 11:51:43,509 - Epoch: [199][ 240/ 1200] Overall Loss 0.117775 Objective Loss 0.117775 LR 0.000031 Time 0.022049 -2022-12-06 11:51:43,699 - Epoch: [199][ 250/ 1200] Overall Loss 0.118116 Objective Loss 0.118116 LR 0.000031 Time 0.021926 -2022-12-06 11:51:43,889 - Epoch: [199][ 260/ 1200] Overall Loss 0.118100 Objective Loss 0.118100 LR 0.000031 Time 0.021811 -2022-12-06 11:51:44,079 - Epoch: [199][ 270/ 1200] Overall Loss 0.118399 Objective Loss 0.118399 LR 0.000031 Time 0.021704 -2022-12-06 11:51:44,270 - Epoch: [199][ 280/ 1200] Overall Loss 0.118410 Objective Loss 0.118410 LR 0.000031 Time 0.021607 -2022-12-06 11:51:44,459 - Epoch: [199][ 290/ 1200] Overall Loss 0.118460 Objective Loss 0.118460 LR 0.000031 Time 0.021515 -2022-12-06 11:51:44,651 - Epoch: [199][ 300/ 1200] Overall Loss 0.118584 Objective Loss 0.118584 LR 0.000031 Time 0.021433 -2022-12-06 11:51:44,841 - Epoch: [199][ 310/ 1200] Overall Loss 0.118567 Objective Loss 0.118567 LR 0.000031 Time 0.021355 -2022-12-06 11:51:45,031 - Epoch: [199][ 320/ 1200] Overall Loss 0.118780 Objective Loss 0.118780 LR 0.000031 Time 0.021278 -2022-12-06 11:51:45,222 - Epoch: [199][ 330/ 1200] Overall Loss 0.118704 Objective Loss 0.118704 LR 0.000031 Time 0.021210 -2022-12-06 11:51:45,412 - Epoch: [199][ 340/ 1200] Overall Loss 0.118542 Objective Loss 0.118542 LR 0.000031 Time 0.021144 -2022-12-06 11:51:45,602 - Epoch: [199][ 350/ 1200] Overall Loss 0.119072 Objective Loss 0.119072 LR 0.000031 Time 0.021082 -2022-12-06 11:51:45,792 - Epoch: [199][ 360/ 1200] Overall Loss 0.118497 Objective Loss 0.118497 LR 0.000031 Time 0.021023 -2022-12-06 11:51:45,983 - Epoch: [199][ 370/ 1200] Overall Loss 0.118535 Objective Loss 0.118535 LR 0.000031 Time 0.020969 -2022-12-06 11:51:46,173 - Epoch: [199][ 380/ 1200] Overall Loss 0.118722 Objective Loss 0.118722 LR 0.000031 Time 0.020916 -2022-12-06 11:51:46,364 - Epoch: [199][ 390/ 1200] Overall Loss 0.118677 Objective Loss 0.118677 LR 0.000031 Time 0.020869 -2022-12-06 11:51:46,555 - Epoch: [199][ 400/ 1200] Overall Loss 0.118982 Objective Loss 0.118982 LR 0.000031 Time 0.020823 -2022-12-06 11:51:46,745 - Epoch: [199][ 410/ 1200] Overall Loss 0.118790 Objective Loss 0.118790 LR 0.000031 Time 0.020777 -2022-12-06 11:51:46,936 - Epoch: [199][ 420/ 1200] Overall Loss 0.118966 Objective Loss 0.118966 LR 0.000031 Time 0.020734 -2022-12-06 11:51:47,126 - Epoch: [199][ 430/ 1200] Overall Loss 0.119306 Objective Loss 0.119306 LR 0.000031 Time 0.020693 -2022-12-06 11:51:47,316 - Epoch: [199][ 440/ 1200] Overall Loss 0.119344 Objective Loss 0.119344 LR 0.000031 Time 0.020653 -2022-12-06 11:51:47,506 - Epoch: [199][ 450/ 1200] Overall Loss 0.119122 Objective Loss 0.119122 LR 0.000031 Time 0.020615 -2022-12-06 11:51:47,696 - Epoch: [199][ 460/ 1200] Overall Loss 0.118716 Objective Loss 0.118716 LR 0.000031 Time 0.020579 -2022-12-06 11:51:47,886 - Epoch: [199][ 470/ 1200] Overall Loss 0.118875 Objective Loss 0.118875 LR 0.000031 Time 0.020544 -2022-12-06 11:51:48,077 - Epoch: [199][ 480/ 1200] Overall Loss 0.118926 Objective Loss 0.118926 LR 0.000031 Time 0.020513 -2022-12-06 11:51:48,267 - Epoch: [199][ 490/ 1200] Overall Loss 0.118904 Objective Loss 0.118904 LR 0.000031 Time 0.020481 -2022-12-06 11:51:48,458 - Epoch: [199][ 500/ 1200] Overall Loss 0.119299 Objective Loss 0.119299 LR 0.000031 Time 0.020453 -2022-12-06 11:51:48,648 - Epoch: [199][ 510/ 1200] Overall Loss 0.119161 Objective Loss 0.119161 LR 0.000031 Time 0.020424 -2022-12-06 11:51:48,838 - Epoch: [199][ 520/ 1200] Overall Loss 0.119092 Objective Loss 0.119092 LR 0.000031 Time 0.020396 -2022-12-06 11:51:49,029 - Epoch: [199][ 530/ 1200] Overall Loss 0.119239 Objective Loss 0.119239 LR 0.000031 Time 0.020369 -2022-12-06 11:51:49,219 - Epoch: [199][ 540/ 1200] Overall Loss 0.119263 Objective Loss 0.119263 LR 0.000031 Time 0.020343 -2022-12-06 11:51:49,410 - Epoch: [199][ 550/ 1200] Overall Loss 0.119755 Objective Loss 0.119755 LR 0.000031 Time 0.020319 -2022-12-06 11:51:49,601 - Epoch: [199][ 560/ 1200] Overall Loss 0.119677 Objective Loss 0.119677 LR 0.000031 Time 0.020296 -2022-12-06 11:51:49,791 - Epoch: [199][ 570/ 1200] Overall Loss 0.119580 Objective Loss 0.119580 LR 0.000031 Time 0.020274 -2022-12-06 11:51:49,982 - Epoch: [199][ 580/ 1200] Overall Loss 0.119563 Objective Loss 0.119563 LR 0.000031 Time 0.020251 -2022-12-06 11:51:50,172 - Epoch: [199][ 590/ 1200] Overall Loss 0.119732 Objective Loss 0.119732 LR 0.000031 Time 0.020230 -2022-12-06 11:51:50,362 - Epoch: [199][ 600/ 1200] Overall Loss 0.119646 Objective Loss 0.119646 LR 0.000031 Time 0.020209 -2022-12-06 11:51:50,553 - Epoch: [199][ 610/ 1200] Overall Loss 0.119901 Objective Loss 0.119901 LR 0.000031 Time 0.020190 -2022-12-06 11:51:50,744 - Epoch: [199][ 620/ 1200] Overall Loss 0.119943 Objective Loss 0.119943 LR 0.000031 Time 0.020170 -2022-12-06 11:51:50,935 - Epoch: [199][ 630/ 1200] Overall Loss 0.119931 Objective Loss 0.119931 LR 0.000031 Time 0.020152 -2022-12-06 11:51:51,125 - Epoch: [199][ 640/ 1200] Overall Loss 0.119920 Objective Loss 0.119920 LR 0.000031 Time 0.020134 -2022-12-06 11:51:51,315 - Epoch: [199][ 650/ 1200] Overall Loss 0.119750 Objective Loss 0.119750 LR 0.000031 Time 0.020116 -2022-12-06 11:51:51,506 - Epoch: [199][ 660/ 1200] Overall Loss 0.119930 Objective Loss 0.119930 LR 0.000031 Time 0.020099 -2022-12-06 11:51:51,696 - Epoch: [199][ 670/ 1200] Overall Loss 0.120106 Objective Loss 0.120106 LR 0.000031 Time 0.020082 -2022-12-06 11:51:51,886 - Epoch: [199][ 680/ 1200] Overall Loss 0.120110 Objective Loss 0.120110 LR 0.000031 Time 0.020066 -2022-12-06 11:51:52,077 - Epoch: [199][ 690/ 1200] Overall Loss 0.120397 Objective Loss 0.120397 LR 0.000031 Time 0.020050 -2022-12-06 11:51:52,267 - Epoch: [199][ 700/ 1200] Overall Loss 0.120509 Objective Loss 0.120509 LR 0.000031 Time 0.020035 -2022-12-06 11:51:52,457 - Epoch: [199][ 710/ 1200] Overall Loss 0.120613 Objective Loss 0.120613 LR 0.000031 Time 0.020020 -2022-12-06 11:51:52,648 - Epoch: [199][ 720/ 1200] Overall Loss 0.120643 Objective Loss 0.120643 LR 0.000031 Time 0.020006 -2022-12-06 11:51:52,838 - Epoch: [199][ 730/ 1200] Overall Loss 0.120467 Objective Loss 0.120467 LR 0.000031 Time 0.019992 -2022-12-06 11:51:53,029 - Epoch: [199][ 740/ 1200] Overall Loss 0.120474 Objective Loss 0.120474 LR 0.000031 Time 0.019979 -2022-12-06 11:51:53,219 - Epoch: [199][ 750/ 1200] Overall Loss 0.120495 Objective Loss 0.120495 LR 0.000031 Time 0.019966 -2022-12-06 11:51:53,410 - Epoch: [199][ 760/ 1200] Overall Loss 0.120317 Objective Loss 0.120317 LR 0.000031 Time 0.019953 -2022-12-06 11:51:53,600 - Epoch: [199][ 770/ 1200] Overall Loss 0.120137 Objective Loss 0.120137 LR 0.000031 Time 0.019940 -2022-12-06 11:51:53,791 - Epoch: [199][ 780/ 1200] Overall Loss 0.120128 Objective Loss 0.120128 LR 0.000031 Time 0.019928 -2022-12-06 11:51:53,981 - Epoch: [199][ 790/ 1200] Overall Loss 0.120236 Objective Loss 0.120236 LR 0.000031 Time 0.019916 -2022-12-06 11:51:54,172 - Epoch: [199][ 800/ 1200] Overall Loss 0.120039 Objective Loss 0.120039 LR 0.000031 Time 0.019905 -2022-12-06 11:51:54,362 - Epoch: [199][ 810/ 1200] Overall Loss 0.119853 Objective Loss 0.119853 LR 0.000031 Time 0.019893 -2022-12-06 11:51:54,552 - Epoch: [199][ 820/ 1200] Overall Loss 0.119778 Objective Loss 0.119778 LR 0.000031 Time 0.019882 -2022-12-06 11:51:54,743 - Epoch: [199][ 830/ 1200] Overall Loss 0.119848 Objective Loss 0.119848 LR 0.000031 Time 0.019871 -2022-12-06 11:51:54,933 - Epoch: [199][ 840/ 1200] Overall Loss 0.119921 Objective Loss 0.119921 LR 0.000031 Time 0.019861 -2022-12-06 11:51:55,124 - Epoch: [199][ 850/ 1200] Overall Loss 0.119931 Objective Loss 0.119931 LR 0.000031 Time 0.019850 -2022-12-06 11:51:55,314 - Epoch: [199][ 860/ 1200] Overall Loss 0.119894 Objective Loss 0.119894 LR 0.000031 Time 0.019840 -2022-12-06 11:51:55,503 - Epoch: [199][ 870/ 1200] Overall Loss 0.119809 Objective Loss 0.119809 LR 0.000031 Time 0.019830 -2022-12-06 11:51:55,695 - Epoch: [199][ 880/ 1200] Overall Loss 0.119805 Objective Loss 0.119805 LR 0.000031 Time 0.019821 -2022-12-06 11:51:55,885 - Epoch: [199][ 890/ 1200] Overall Loss 0.119673 Objective Loss 0.119673 LR 0.000031 Time 0.019811 -2022-12-06 11:51:56,076 - Epoch: [199][ 900/ 1200] Overall Loss 0.119623 Objective Loss 0.119623 LR 0.000031 Time 0.019803 -2022-12-06 11:51:56,266 - Epoch: [199][ 910/ 1200] Overall Loss 0.119520 Objective Loss 0.119520 LR 0.000031 Time 0.019794 -2022-12-06 11:51:56,457 - Epoch: [199][ 920/ 1200] Overall Loss 0.119339 Objective Loss 0.119339 LR 0.000031 Time 0.019785 -2022-12-06 11:51:56,648 - Epoch: [199][ 930/ 1200] Overall Loss 0.119391 Objective Loss 0.119391 LR 0.000031 Time 0.019777 -2022-12-06 11:51:56,839 - Epoch: [199][ 940/ 1200] Overall Loss 0.119355 Objective Loss 0.119355 LR 0.000031 Time 0.019769 -2022-12-06 11:51:57,028 - Epoch: [199][ 950/ 1200] Overall Loss 0.119340 Objective Loss 0.119340 LR 0.000031 Time 0.019760 -2022-12-06 11:51:57,219 - Epoch: [199][ 960/ 1200] Overall Loss 0.119357 Objective Loss 0.119357 LR 0.000031 Time 0.019752 -2022-12-06 11:51:57,409 - Epoch: [199][ 970/ 1200] Overall Loss 0.119428 Objective Loss 0.119428 LR 0.000031 Time 0.019744 -2022-12-06 11:51:57,599 - Epoch: [199][ 980/ 1200] Overall Loss 0.119451 Objective Loss 0.119451 LR 0.000031 Time 0.019736 -2022-12-06 11:51:57,790 - Epoch: [199][ 990/ 1200] Overall Loss 0.119554 Objective Loss 0.119554 LR 0.000031 Time 0.019729 -2022-12-06 11:51:57,980 - Epoch: [199][ 1000/ 1200] Overall Loss 0.119355 Objective Loss 0.119355 LR 0.000031 Time 0.019722 -2022-12-06 11:51:58,170 - Epoch: [199][ 1010/ 1200] Overall Loss 0.119433 Objective Loss 0.119433 LR 0.000031 Time 0.019714 -2022-12-06 11:51:58,360 - Epoch: [199][ 1020/ 1200] Overall Loss 0.119384 Objective Loss 0.119384 LR 0.000031 Time 0.019706 -2022-12-06 11:51:58,550 - Epoch: [199][ 1030/ 1200] Overall Loss 0.119321 Objective Loss 0.119321 LR 0.000031 Time 0.019699 -2022-12-06 11:51:58,741 - Epoch: [199][ 1040/ 1200] Overall Loss 0.119186 Objective Loss 0.119186 LR 0.000031 Time 0.019692 -2022-12-06 11:51:58,931 - Epoch: [199][ 1050/ 1200] Overall Loss 0.118966 Objective Loss 0.118966 LR 0.000031 Time 0.019685 -2022-12-06 11:51:59,122 - Epoch: [199][ 1060/ 1200] Overall Loss 0.118938 Objective Loss 0.118938 LR 0.000031 Time 0.019679 -2022-12-06 11:51:59,312 - Epoch: [199][ 1070/ 1200] Overall Loss 0.119021 Objective Loss 0.119021 LR 0.000031 Time 0.019672 -2022-12-06 11:51:59,502 - Epoch: [199][ 1080/ 1200] Overall Loss 0.119105 Objective Loss 0.119105 LR 0.000031 Time 0.019666 -2022-12-06 11:51:59,692 - Epoch: [199][ 1090/ 1200] Overall Loss 0.119129 Objective Loss 0.119129 LR 0.000031 Time 0.019659 -2022-12-06 11:51:59,884 - Epoch: [199][ 1100/ 1200] Overall Loss 0.119165 Objective Loss 0.119165 LR 0.000031 Time 0.019654 -2022-12-06 11:52:00,074 - Epoch: [199][ 1110/ 1200] Overall Loss 0.119067 Objective Loss 0.119067 LR 0.000031 Time 0.019648 -2022-12-06 11:52:00,265 - Epoch: [199][ 1120/ 1200] Overall Loss 0.119004 Objective Loss 0.119004 LR 0.000031 Time 0.019642 -2022-12-06 11:52:00,455 - Epoch: [199][ 1130/ 1200] Overall Loss 0.119008 Objective Loss 0.119008 LR 0.000031 Time 0.019637 -2022-12-06 11:52:00,646 - Epoch: [199][ 1140/ 1200] Overall Loss 0.118949 Objective Loss 0.118949 LR 0.000031 Time 0.019631 -2022-12-06 11:52:00,836 - Epoch: [199][ 1150/ 1200] Overall Loss 0.119093 Objective Loss 0.119093 LR 0.000031 Time 0.019625 -2022-12-06 11:52:01,027 - Epoch: [199][ 1160/ 1200] Overall Loss 0.119128 Objective Loss 0.119128 LR 0.000031 Time 0.019620 -2022-12-06 11:52:01,217 - Epoch: [199][ 1170/ 1200] Overall Loss 0.119134 Objective Loss 0.119134 LR 0.000031 Time 0.019614 -2022-12-06 11:52:01,407 - Epoch: [199][ 1180/ 1200] Overall Loss 0.119237 Objective Loss 0.119237 LR 0.000031 Time 0.019609 -2022-12-06 11:52:01,597 - Epoch: [199][ 1190/ 1200] Overall Loss 0.119338 Objective Loss 0.119338 LR 0.000031 Time 0.019603 -2022-12-06 11:52:01,820 - Epoch: [199][ 1200/ 1200] Overall Loss 0.119171 Objective Loss 0.119171 Top1 94.142259 Top5 99.581590 LR 0.000031 Time 0.019625 -2022-12-06 11:52:01,908 - --- validate (epoch=199)----------- -2022-12-06 11:52:01,908 - 34129 samples (256 per mini-batch) -2022-12-06 11:52:02,355 - Epoch: [199][ 10/ 134] Loss 0.205084 Top1 89.296875 Top5 98.437500 -2022-12-06 11:52:02,484 - Epoch: [199][ 20/ 134] Loss 0.208329 Top1 89.238281 Top5 98.691406 -2022-12-06 11:52:02,616 - Epoch: [199][ 30/ 134] Loss 0.206838 Top1 89.257812 Top5 98.723958 -2022-12-06 11:52:02,742 - Epoch: [199][ 40/ 134] Loss 0.213025 Top1 89.091797 Top5 98.671875 -2022-12-06 11:52:02,868 - Epoch: [199][ 50/ 134] Loss 0.211692 Top1 89.132812 Top5 98.671875 -2022-12-06 11:52:02,996 - Epoch: [199][ 60/ 134] Loss 0.213703 Top1 89.095052 Top5 98.658854 -2022-12-06 11:52:03,122 - Epoch: [199][ 70/ 134] Loss 0.214060 Top1 89.073661 Top5 98.671875 -2022-12-06 11:52:03,250 - Epoch: [199][ 80/ 134] Loss 0.211955 Top1 89.199219 Top5 98.710938 -2022-12-06 11:52:03,376 - Epoch: [199][ 90/ 134] Loss 0.212933 Top1 89.149306 Top5 98.732639 -2022-12-06 11:52:03,504 - Epoch: [199][ 100/ 134] Loss 0.214134 Top1 89.136719 Top5 98.699219 -2022-12-06 11:52:03,630 - Epoch: [199][ 110/ 134] Loss 0.216518 Top1 89.058949 Top5 98.671875 -2022-12-06 11:52:03,757 - Epoch: [199][ 120/ 134] Loss 0.217798 Top1 89.026693 Top5 98.701172 -2022-12-06 11:52:03,885 - Epoch: [199][ 130/ 134] Loss 0.219549 Top1 88.933293 Top5 98.677885 -2022-12-06 11:52:03,921 - Epoch: [199][ 134/ 134] Loss 0.219794 Top1 88.930235 Top5 98.681473 -2022-12-06 11:52:04,009 - ==> Top1: 88.930 Top5: 98.681 Loss: 0.220 - -2022-12-06 11:52:04,010 - ==> Confusion: -[[ 929 1 1 3 4 5 1 0 4 35 0 1 1 2 3 1 2 0 2 0 1] - [ 1 948 1 2 6 18 2 13 2 2 1 3 1 0 0 1 3 2 7 4 10] - [ 4 2 1014 14 3 2 16 9 0 4 3 3 1 2 1 1 0 1 3 6 14] - [ 4 1 11 959 1 2 1 0 0 1 8 0 3 2 8 0 1 1 11 0 6] - [ 9 4 1 0 960 1 1 2 1 7 1 4 0 2 10 5 5 3 0 0 4] - [ 2 9 0 4 4 994 2 16 3 3 0 9 3 11 3 1 0 0 0 3 2] - [ 1 0 7 1 0 1 1084 3 0 1 0 2 0 1 0 2 0 2 2 7 4] - [ 1 5 3 3 1 24 11 967 0 0 1 6 0 1 0 0 1 0 13 9 8] - [ 6 0 0 0 1 1 1 2 993 35 8 2 0 5 4 0 1 1 1 1 2] - [ 53 0 0 0 4 1 0 3 23 892 1 1 0 11 3 1 0 0 1 0 7] - [ 1 2 3 3 0 1 1 2 9 1 966 0 0 9 4 1 0 0 4 2 10] - [ 3 0 1 0 1 8 3 3 1 0 1 975 23 4 0 8 2 5 0 8 5] - [ 1 1 1 2 0 2 0 0 1 1 0 19 915 0 1 8 1 8 0 2 6] - [ 1 1 0 0 1 7 0 3 11 10 4 2 3 966 1 3 2 0 0 1 7] - [ 8 4 1 11 3 3 0 0 11 2 0 3 1 4 1070 0 0 1 5 0 3] - [ 0 0 0 2 3 0 3 0 0 0 2 6 4 1 0 1002 2 9 0 4 5] - [ 2 0 1 2 1 0 0 0 1 0 0 1 3 1 0 15 1031 2 0 4 8] - [ 3 0 1 1 2 1 0 0 0 2 0 3 14 2 1 14 0 989 0 0 3] - [ 2 2 3 7 0 2 0 20 2 1 3 3 3 1 5 0 0 2 947 0 5] - [ 1 4 1 1 3 4 5 4 0 1 2 11 4 6 0 4 4 1 0 1018 6] - [ 118 169 139 88 87 149 71 120 72 79 127 81 275 214 127 96 115 79 116 179 10725]] - -2022-12-06 11:52:04,688 - ==> Best [Top1: 89.086 Top5: 98.670 Sparsity:0.00 Params: 5376 on epoch: 192] -2022-12-06 11:52:04,688 - Saving checkpoint to: logs/2022.12.06-093055/qat_checkpoint.pth.tar -2022-12-06 11:52:04,694 - --- test --------------------- -2022-12-06 11:52:04,694 - 38058 samples (256 per mini-batch) -2022-12-06 11:52:05,141 - Test: [ 10/ 149] Loss 0.271637 Top1 88.242188 Top5 98.593750 -2022-12-06 11:52:05,276 - Test: [ 20/ 149] Loss 0.286768 Top1 87.929688 Top5 98.398438 -2022-12-06 11:52:05,407 - Test: [ 30/ 149] Loss 0.290988 Top1 87.656250 Top5 98.281250 -2022-12-06 11:52:05,543 - Test: [ 40/ 149] Loss 0.305289 Top1 87.392578 Top5 98.154297 -2022-12-06 11:52:05,680 - Test: [ 50/ 149] Loss 0.292040 Top1 87.656250 Top5 98.265625 -2022-12-06 11:52:05,815 - Test: [ 60/ 149] Loss 0.294143 Top1 87.480469 Top5 98.235677 -2022-12-06 11:52:05,950 - Test: [ 70/ 149] Loss 0.293202 Top1 87.494420 Top5 98.281250 -2022-12-06 11:52:06,087 - Test: [ 80/ 149] Loss 0.292344 Top1 87.485352 Top5 98.291016 -2022-12-06 11:52:06,224 - Test: [ 90/ 149] Loss 0.292362 Top1 87.500000 Top5 98.315972 -2022-12-06 11:52:06,363 - Test: [ 100/ 149] Loss 0.293850 Top1 87.429688 Top5 98.312500 -2022-12-06 11:52:06,502 - Test: [ 110/ 149] Loss 0.297461 Top1 87.425426 Top5 98.309659 -2022-12-06 11:52:06,639 - Test: [ 120/ 149] Loss 0.300190 Top1 87.389323 Top5 98.284505 -2022-12-06 11:52:06,771 - Test: [ 130/ 149] Loss 0.298059 Top1 87.451923 Top5 98.299279 -2022-12-06 11:52:06,901 - Test: [ 140/ 149] Loss 0.298582 Top1 87.441406 Top5 98.297991 -2022-12-06 11:52:07,007 - Test: [ 149/ 149] Loss 0.299148 Top1 87.448106 Top5 98.286825 -2022-12-06 11:52:07,102 - ==> Top1: 87.448 Top5: 98.287 Loss: 0.299 - -2022-12-06 11:52:07,103 - ==> Confusion: -[[ 965 2 3 0 8 4 0 0 10 60 2 1 1 6 5 3 3 5 1 0 4] - [ 2 1036 1 0 8 29 1 8 7 2 0 6 2 1 1 5 3 3 8 2 6] - [ 11 2 1034 11 0 1 28 12 0 3 5 3 5 1 3 8 0 4 2 3 7] - [ 0 0 10 1113 1 5 0 1 2 3 11 0 15 3 12 3 6 10 10 0 7] - [ 11 9 5 0 1101 11 0 1 1 12 2 8 1 1 3 6 2 3 0 0 8] - [ 5 19 1 2 0 1038 2 24 1 5 8 9 1 7 0 1 1 1 2 8 2] - [ 0 5 11 1 0 1 1226 1 0 0 0 4 1 2 0 1 0 3 1 14 1] - [ 2 14 8 0 1 32 5 1111 0 0 0 6 0 1 0 1 0 2 20 7 5] - [ 10 4 0 2 0 2 0 1 1012 32 15 7 3 13 17 2 0 3 1 0 7] - [ 60 0 3 0 2 4 5 5 33 998 2 0 0 16 5 3 1 2 0 1 3] - [ 1 1 3 6 2 1 1 3 10 0 1092 0 2 12 10 0 4 0 4 1 11] - [ 8 0 1 0 2 20 1 3 1 1 2 1072 25 6 0 3 1 8 1 12 3] - [ 3 1 2 1 0 2 1 0 1 0 2 22 1005 2 5 8 2 31 1 3 15] - [ 1 0 2 4 3 8 0 1 11 11 14 10 2 998 3 2 3 3 1 2 10] - [ 11 9 2 11 7 1 0 1 10 12 6 2 6 5 1110 0 4 5 2 2 9] - [ 1 0 0 0 3 1 3 0 1 0 0 14 5 3 2 1073 4 17 1 1 2] - [ 4 6 2 1 6 7 1 1 4 0 1 5 4 2 2 19 1121 3 0 3 23] - [ 0 0 1 4 0 0 0 1 0 1 0 5 18 2 1 15 0 1073 0 3 1] - [ 1 1 9 7 0 1 2 32 2 0 5 2 0 1 7 0 0 0 1094 0 6] - [ 2 3 1 1 3 5 7 13 0 2 2 17 5 3 0 6 3 0 1 1145 8] - [ 113 167 161 91 73 207 63 134 119 92 181 95 325 273 139 101 168 84 156 193 11858]] - -2022-12-06 11:52:07,226 - -2022-12-06 11:52:07,226 - Log file for this run: /data/ml/afshin/ai/kws20-enhancement/ai8x-training/logs/2022.12.06-093055/2022.12.06-093055.log +2023-10-02 19:52:28,941 - Log file for this run: /home/alicangok/Projects/AI8X/train_clean/logs/2023.10.02-195228/2023.10.02-195228.log +2023-10-02 19:52:31,089 - Optimizer Type: +2023-10-02 19:52:31,090 - Optimizer Args: {'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0.0, 'amsgrad': False} +2023-10-02 20:29:09,721 - Dataset sizes: + training=316395 + validation=29943 + test=33015 +2023-10-02 20:29:09,721 - Reading compression schedule from: policies/schedule_kws20.yaml +2023-10-02 20:29:09,725 - + +2023-10-02 20:29:09,725 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:29:11,015 - Epoch: [0][ 10/ 1236] Overall Loss 3.044487 Objective Loss 3.044487 LR 0.001000 Time 0.128923 +2023-10-02 20:29:11,175 - Epoch: [0][ 20/ 1236] Overall Loss 3.043869 Objective Loss 3.043869 LR 0.001000 Time 0.072466 +2023-10-02 20:29:11,335 - Epoch: [0][ 30/ 1236] Overall Loss 3.041371 Objective Loss 3.041371 LR 0.001000 Time 0.053635 +2023-10-02 20:29:11,496 - Epoch: [0][ 40/ 1236] Overall Loss 3.036939 Objective Loss 3.036939 LR 0.001000 Time 0.044234 +2023-10-02 20:29:11,656 - Epoch: [0][ 50/ 1236] Overall Loss 3.028362 Objective Loss 3.028362 LR 0.001000 Time 0.038572 +2023-10-02 20:29:11,816 - Epoch: [0][ 60/ 1236] Overall Loss 3.013532 Objective Loss 3.013532 LR 0.001000 Time 0.034814 +2023-10-02 20:29:11,976 - Epoch: [0][ 70/ 1236] Overall Loss 3.001722 Objective Loss 3.001722 LR 0.001000 Time 0.032119 +2023-10-02 20:29:12,137 - Epoch: [0][ 80/ 1236] Overall Loss 2.987417 Objective Loss 2.987417 LR 0.001000 Time 0.030110 +2023-10-02 20:29:12,297 - Epoch: [0][ 90/ 1236] Overall Loss 2.965891 Objective Loss 2.965891 LR 0.001000 Time 0.028537 +2023-10-02 20:29:12,457 - Epoch: [0][ 100/ 1236] Overall Loss 2.939472 Objective Loss 2.939472 LR 0.001000 Time 0.027284 +2023-10-02 20:29:12,617 - Epoch: [0][ 110/ 1236] Overall Loss 2.911176 Objective Loss 2.911176 LR 0.001000 Time 0.026252 +2023-10-02 20:29:12,777 - Epoch: [0][ 120/ 1236] Overall Loss 2.886002 Objective Loss 2.886002 LR 0.001000 Time 0.025399 +2023-10-02 20:29:12,937 - Epoch: [0][ 130/ 1236] Overall Loss 2.859854 Objective Loss 2.859854 LR 0.001000 Time 0.024670 +2023-10-02 20:29:13,097 - Epoch: [0][ 140/ 1236] Overall Loss 2.831811 Objective Loss 2.831811 LR 0.001000 Time 0.024052 +2023-10-02 20:29:13,256 - Epoch: [0][ 150/ 1236] Overall Loss 2.808278 Objective Loss 2.808278 LR 0.001000 Time 0.023508 +2023-10-02 20:29:13,417 - Epoch: [0][ 160/ 1236] Overall Loss 2.786254 Objective Loss 2.786254 LR 0.001000 Time 0.023042 +2023-10-02 20:29:13,577 - Epoch: [0][ 170/ 1236] Overall Loss 2.763223 Objective Loss 2.763223 LR 0.001000 Time 0.022623 +2023-10-02 20:29:13,737 - Epoch: [0][ 180/ 1236] Overall Loss 2.739347 Objective Loss 2.739347 LR 0.001000 Time 0.022256 +2023-10-02 20:29:13,898 - Epoch: [0][ 190/ 1236] Overall Loss 2.719697 Objective Loss 2.719697 LR 0.001000 Time 0.021928 +2023-10-02 20:29:14,059 - Epoch: [0][ 200/ 1236] Overall Loss 2.700654 Objective Loss 2.700654 LR 0.001000 Time 0.021637 +2023-10-02 20:29:14,220 - Epoch: [0][ 210/ 1236] Overall Loss 2.680273 Objective Loss 2.680273 LR 0.001000 Time 0.021369 +2023-10-02 20:29:14,381 - Epoch: [0][ 220/ 1236] Overall Loss 2.658086 Objective Loss 2.658086 LR 0.001000 Time 0.021130 +2023-10-02 20:29:14,541 - Epoch: [0][ 230/ 1236] Overall Loss 2.635074 Objective Loss 2.635074 LR 0.001000 Time 0.020908 +2023-10-02 20:29:14,703 - Epoch: [0][ 240/ 1236] Overall Loss 2.613678 Objective Loss 2.613678 LR 0.001000 Time 0.020707 +2023-10-02 20:29:14,863 - Epoch: [0][ 250/ 1236] Overall Loss 2.594949 Objective Loss 2.594949 LR 0.001000 Time 0.020519 +2023-10-02 20:29:15,024 - Epoch: [0][ 260/ 1236] Overall Loss 2.576883 Objective Loss 2.576883 LR 0.001000 Time 0.020349 +2023-10-02 20:29:15,185 - Epoch: [0][ 270/ 1236] Overall Loss 2.558841 Objective Loss 2.558841 LR 0.001000 Time 0.020190 +2023-10-02 20:29:15,344 - Epoch: [0][ 280/ 1236] Overall Loss 2.539332 Objective Loss 2.539332 LR 0.001000 Time 0.020036 +2023-10-02 20:29:15,506 - Epoch: [0][ 290/ 1236] Overall Loss 2.520477 Objective Loss 2.520477 LR 0.001000 Time 0.019903 +2023-10-02 20:29:15,665 - Epoch: [0][ 300/ 1236] Overall Loss 2.504157 Objective Loss 2.504157 LR 0.001000 Time 0.019768 +2023-10-02 20:29:15,827 - Epoch: [0][ 310/ 1236] Overall Loss 2.487104 Objective Loss 2.487104 LR 0.001000 Time 0.019650 +2023-10-02 20:29:15,986 - Epoch: [0][ 320/ 1236] Overall Loss 2.470560 Objective Loss 2.470560 LR 0.001000 Time 0.019532 +2023-10-02 20:29:16,147 - Epoch: [0][ 330/ 1236] Overall Loss 2.453117 Objective Loss 2.453117 LR 0.001000 Time 0.019427 +2023-10-02 20:29:16,306 - Epoch: [0][ 340/ 1236] Overall Loss 2.435903 Objective Loss 2.435903 LR 0.001000 Time 0.019324 +2023-10-02 20:29:16,467 - Epoch: [0][ 350/ 1236] Overall Loss 2.418946 Objective Loss 2.418946 LR 0.001000 Time 0.019230 +2023-10-02 20:29:16,626 - Epoch: [0][ 360/ 1236] Overall Loss 2.403944 Objective Loss 2.403944 LR 0.001000 Time 0.019137 +2023-10-02 20:29:16,787 - Epoch: [0][ 370/ 1236] Overall Loss 2.388543 Objective Loss 2.388543 LR 0.001000 Time 0.019054 +2023-10-02 20:29:16,946 - Epoch: [0][ 380/ 1236] Overall Loss 2.373802 Objective Loss 2.373802 LR 0.001000 Time 0.018970 +2023-10-02 20:29:17,107 - Epoch: [0][ 390/ 1236] Overall Loss 2.360220 Objective Loss 2.360220 LR 0.001000 Time 0.018895 +2023-10-02 20:29:17,266 - Epoch: [0][ 400/ 1236] Overall Loss 2.345238 Objective Loss 2.345238 LR 0.001000 Time 0.018820 +2023-10-02 20:29:17,427 - Epoch: [0][ 410/ 1236] Overall Loss 2.331127 Objective Loss 2.331127 LR 0.001000 Time 0.018754 +2023-10-02 20:29:17,587 - Epoch: [0][ 420/ 1236] Overall Loss 2.317458 Objective Loss 2.317458 LR 0.001000 Time 0.018686 +2023-10-02 20:29:17,748 - Epoch: [0][ 430/ 1236] Overall Loss 2.303499 Objective Loss 2.303499 LR 0.001000 Time 0.018626 +2023-10-02 20:29:17,908 - Epoch: [0][ 440/ 1236] Overall Loss 2.290037 Objective Loss 2.290037 LR 0.001000 Time 0.018565 +2023-10-02 20:29:18,069 - Epoch: [0][ 450/ 1236] Overall Loss 2.276128 Objective Loss 2.276128 LR 0.001000 Time 0.018511 +2023-10-02 20:29:18,229 - Epoch: [0][ 460/ 1236] Overall Loss 2.263184 Objective Loss 2.263184 LR 0.001000 Time 0.018455 +2023-10-02 20:29:18,390 - Epoch: [0][ 470/ 1236] Overall Loss 2.250054 Objective Loss 2.250054 LR 0.001000 Time 0.018405 +2023-10-02 20:29:18,549 - Epoch: [0][ 480/ 1236] Overall Loss 2.237896 Objective Loss 2.237896 LR 0.001000 Time 0.018353 +2023-10-02 20:29:18,711 - Epoch: [0][ 490/ 1236] Overall Loss 2.226133 Objective Loss 2.226133 LR 0.001000 Time 0.018308 +2023-10-02 20:29:18,872 - Epoch: [0][ 500/ 1236] Overall Loss 2.215156 Objective Loss 2.215156 LR 0.001000 Time 0.018262 +2023-10-02 20:29:19,034 - Epoch: [0][ 510/ 1236] Overall Loss 2.204911 Objective Loss 2.204911 LR 0.001000 Time 0.018222 +2023-10-02 20:29:19,194 - Epoch: [0][ 520/ 1236] Overall Loss 2.192753 Objective Loss 2.192753 LR 0.001000 Time 0.018178 +2023-10-02 20:29:19,356 - Epoch: [0][ 530/ 1236] Overall Loss 2.180165 Objective Loss 2.180165 LR 0.001000 Time 0.018139 +2023-10-02 20:29:19,515 - Epoch: [0][ 540/ 1236] Overall Loss 2.168805 Objective Loss 2.168805 LR 0.001000 Time 0.018099 +2023-10-02 20:29:19,677 - Epoch: [0][ 550/ 1236] Overall Loss 2.156828 Objective Loss 2.156828 LR 0.001000 Time 0.018064 +2023-10-02 20:29:19,837 - Epoch: [0][ 560/ 1236] Overall Loss 2.145467 Objective Loss 2.145467 LR 0.001000 Time 0.018026 +2023-10-02 20:29:19,999 - Epoch: [0][ 570/ 1236] Overall Loss 2.134713 Objective Loss 2.134713 LR 0.001000 Time 0.017993 +2023-10-02 20:29:20,159 - Epoch: [0][ 580/ 1236] Overall Loss 2.122831 Objective Loss 2.122831 LR 0.001000 Time 0.017958 +2023-10-02 20:29:20,321 - Epoch: [0][ 590/ 1236] Overall Loss 2.111729 Objective Loss 2.111729 LR 0.001000 Time 0.017928 +2023-10-02 20:29:20,481 - Epoch: [0][ 600/ 1236] Overall Loss 2.101706 Objective Loss 2.101706 LR 0.001000 Time 0.017895 +2023-10-02 20:29:20,642 - Epoch: [0][ 610/ 1236] Overall Loss 2.091479 Objective Loss 2.091479 LR 0.001000 Time 0.017866 +2023-10-02 20:29:20,803 - Epoch: [0][ 620/ 1236] Overall Loss 2.081530 Objective Loss 2.081530 LR 0.001000 Time 0.017836 +2023-10-02 20:29:20,965 - Epoch: [0][ 630/ 1236] Overall Loss 2.071963 Objective Loss 2.071963 LR 0.001000 Time 0.017809 +2023-10-02 20:29:21,125 - Epoch: [0][ 640/ 1236] Overall Loss 2.061962 Objective Loss 2.061962 LR 0.001000 Time 0.017782 +2023-10-02 20:29:21,287 - Epoch: [0][ 650/ 1236] Overall Loss 2.051943 Objective Loss 2.051943 LR 0.001000 Time 0.017757 +2023-10-02 20:29:21,448 - Epoch: [0][ 660/ 1236] Overall Loss 2.042663 Objective Loss 2.042663 LR 0.001000 Time 0.017731 +2023-10-02 20:29:21,609 - Epoch: [0][ 670/ 1236] Overall Loss 2.032590 Objective Loss 2.032590 LR 0.001000 Time 0.017707 +2023-10-02 20:29:21,770 - Epoch: [0][ 680/ 1236] Overall Loss 2.023060 Objective Loss 2.023060 LR 0.001000 Time 0.017682 +2023-10-02 20:29:21,932 - Epoch: [0][ 690/ 1236] Overall Loss 2.014065 Objective Loss 2.014065 LR 0.001000 Time 0.017660 +2023-10-02 20:29:22,093 - Epoch: [0][ 700/ 1236] Overall Loss 2.005264 Objective Loss 2.005264 LR 0.001000 Time 0.017637 +2023-10-02 20:29:22,254 - Epoch: [0][ 710/ 1236] Overall Loss 1.996290 Objective Loss 1.996290 LR 0.001000 Time 0.017616 +2023-10-02 20:29:22,415 - Epoch: [0][ 720/ 1236] Overall Loss 1.987387 Objective Loss 1.987387 LR 0.001000 Time 0.017593 +2023-10-02 20:29:22,576 - Epoch: [0][ 730/ 1236] Overall Loss 1.978740 Objective Loss 1.978740 LR 0.001000 Time 0.017574 +2023-10-02 20:29:22,737 - Epoch: [0][ 740/ 1236] Overall Loss 1.970352 Objective Loss 1.970352 LR 0.001000 Time 0.017553 +2023-10-02 20:29:22,899 - Epoch: [0][ 750/ 1236] Overall Loss 1.961302 Objective Loss 1.961302 LR 0.001000 Time 0.017534 +2023-10-02 20:29:23,060 - Epoch: [0][ 760/ 1236] Overall Loss 1.952575 Objective Loss 1.952575 LR 0.001000 Time 0.017515 +2023-10-02 20:29:23,222 - Epoch: [0][ 770/ 1236] Overall Loss 1.943852 Objective Loss 1.943852 LR 0.001000 Time 0.017497 +2023-10-02 20:29:23,383 - Epoch: [0][ 780/ 1236] Overall Loss 1.934989 Objective Loss 1.934989 LR 0.001000 Time 0.017479 +2023-10-02 20:29:23,545 - Epoch: [0][ 790/ 1236] Overall Loss 1.927068 Objective Loss 1.927068 LR 0.001000 Time 0.017462 +2023-10-02 20:29:23,706 - Epoch: [0][ 800/ 1236] Overall Loss 1.919172 Objective Loss 1.919172 LR 0.001000 Time 0.017445 +2023-10-02 20:29:23,867 - Epoch: [0][ 810/ 1236] Overall Loss 1.911605 Objective Loss 1.911605 LR 0.001000 Time 0.017429 +2023-10-02 20:29:24,028 - Epoch: [0][ 820/ 1236] Overall Loss 1.903098 Objective Loss 1.903098 LR 0.001000 Time 0.017411 +2023-10-02 20:29:24,190 - Epoch: [0][ 830/ 1236] Overall Loss 1.895773 Objective Loss 1.895773 LR 0.001000 Time 0.017397 +2023-10-02 20:29:24,351 - Epoch: [0][ 840/ 1236] Overall Loss 1.888093 Objective Loss 1.888093 LR 0.001000 Time 0.017381 +2023-10-02 20:29:24,513 - Epoch: [0][ 850/ 1236] Overall Loss 1.880607 Objective Loss 1.880607 LR 0.001000 Time 0.017366 +2023-10-02 20:29:24,674 - Epoch: [0][ 860/ 1236] Overall Loss 1.873310 Objective Loss 1.873310 LR 0.001000 Time 0.017351 +2023-10-02 20:29:24,835 - Epoch: [0][ 870/ 1236] Overall Loss 1.865662 Objective Loss 1.865662 LR 0.001000 Time 0.017337 +2023-10-02 20:29:24,997 - Epoch: [0][ 880/ 1236] Overall Loss 1.858511 Objective Loss 1.858511 LR 0.001000 Time 0.017323 +2023-10-02 20:29:25,159 - Epoch: [0][ 890/ 1236] Overall Loss 1.851172 Objective Loss 1.851172 LR 0.001000 Time 0.017310 +2023-10-02 20:29:25,320 - Epoch: [0][ 900/ 1236] Overall Loss 1.843839 Objective Loss 1.843839 LR 0.001000 Time 0.017296 +2023-10-02 20:29:25,481 - Epoch: [0][ 910/ 1236] Overall Loss 1.837367 Objective Loss 1.837367 LR 0.001000 Time 0.017284 +2023-10-02 20:29:25,642 - Epoch: [0][ 920/ 1236] Overall Loss 1.830195 Objective Loss 1.830195 LR 0.001000 Time 0.017271 +2023-10-02 20:29:25,804 - Epoch: [0][ 930/ 1236] Overall Loss 1.823499 Objective Loss 1.823499 LR 0.001000 Time 0.017259 +2023-10-02 20:29:25,965 - Epoch: [0][ 940/ 1236] Overall Loss 1.816869 Objective Loss 1.816869 LR 0.001000 Time 0.017246 +2023-10-02 20:29:26,128 - Epoch: [0][ 950/ 1236] Overall Loss 1.810617 Objective Loss 1.810617 LR 0.001000 Time 0.017235 +2023-10-02 20:29:26,289 - Epoch: [0][ 960/ 1236] Overall Loss 1.803959 Objective Loss 1.803959 LR 0.001000 Time 0.017223 +2023-10-02 20:29:26,451 - Epoch: [0][ 970/ 1236] Overall Loss 1.797753 Objective Loss 1.797753 LR 0.001000 Time 0.017212 +2023-10-02 20:29:26,612 - Epoch: [0][ 980/ 1236] Overall Loss 1.790909 Objective Loss 1.790909 LR 0.001000 Time 0.017201 +2023-10-02 20:29:26,774 - Epoch: [0][ 990/ 1236] Overall Loss 1.784527 Objective Loss 1.784527 LR 0.001000 Time 0.017190 +2023-10-02 20:29:26,935 - Epoch: [0][ 1000/ 1236] Overall Loss 1.777918 Objective Loss 1.777918 LR 0.001000 Time 0.017179 +2023-10-02 20:29:27,097 - Epoch: [0][ 1010/ 1236] Overall Loss 1.771322 Objective Loss 1.771322 LR 0.001000 Time 0.017169 +2023-10-02 20:29:27,258 - Epoch: [0][ 1020/ 1236] Overall Loss 1.765629 Objective Loss 1.765629 LR 0.001000 Time 0.017159 +2023-10-02 20:29:27,420 - Epoch: [0][ 1030/ 1236] Overall Loss 1.759648 Objective Loss 1.759648 LR 0.001000 Time 0.017149 +2023-10-02 20:29:27,581 - Epoch: [0][ 1040/ 1236] Overall Loss 1.753800 Objective Loss 1.753800 LR 0.001000 Time 0.017138 +2023-10-02 20:29:27,743 - Epoch: [0][ 1050/ 1236] Overall Loss 1.747705 Objective Loss 1.747705 LR 0.001000 Time 0.017129 +2023-10-02 20:29:27,904 - Epoch: [0][ 1060/ 1236] Overall Loss 1.742214 Objective Loss 1.742214 LR 0.001000 Time 0.017119 +2023-10-02 20:29:28,066 - Epoch: [0][ 1070/ 1236] Overall Loss 1.736095 Objective Loss 1.736095 LR 0.001000 Time 0.017110 +2023-10-02 20:29:28,227 - Epoch: [0][ 1080/ 1236] Overall Loss 1.730729 Objective Loss 1.730729 LR 0.001000 Time 0.017100 +2023-10-02 20:29:28,389 - Epoch: [0][ 1090/ 1236] Overall Loss 1.725180 Objective Loss 1.725180 LR 0.001000 Time 0.017092 +2023-10-02 20:29:28,550 - Epoch: [0][ 1100/ 1236] Overall Loss 1.719527 Objective Loss 1.719527 LR 0.001000 Time 0.017082 +2023-10-02 20:29:28,711 - Epoch: [0][ 1110/ 1236] Overall Loss 1.713755 Objective Loss 1.713755 LR 0.001000 Time 0.017074 +2023-10-02 20:29:28,872 - Epoch: [0][ 1120/ 1236] Overall Loss 1.708270 Objective Loss 1.708270 LR 0.001000 Time 0.017065 +2023-10-02 20:29:29,034 - Epoch: [0][ 1130/ 1236] Overall Loss 1.702935 Objective Loss 1.702935 LR 0.001000 Time 0.017057 +2023-10-02 20:29:29,196 - Epoch: [0][ 1140/ 1236] Overall Loss 1.697737 Objective Loss 1.697737 LR 0.001000 Time 0.017049 +2023-10-02 20:29:29,357 - Epoch: [0][ 1150/ 1236] Overall Loss 1.692514 Objective Loss 1.692514 LR 0.001000 Time 0.017041 +2023-10-02 20:29:29,518 - Epoch: [0][ 1160/ 1236] Overall Loss 1.687165 Objective Loss 1.687165 LR 0.001000 Time 0.017033 +2023-10-02 20:29:29,680 - Epoch: [0][ 1170/ 1236] Overall Loss 1.682086 Objective Loss 1.682086 LR 0.001000 Time 0.017025 +2023-10-02 20:29:29,842 - Epoch: [0][ 1180/ 1236] Overall Loss 1.676742 Objective Loss 1.676742 LR 0.001000 Time 0.017017 +2023-10-02 20:29:30,004 - Epoch: [0][ 1190/ 1236] Overall Loss 1.671557 Objective Loss 1.671557 LR 0.001000 Time 0.017010 +2023-10-02 20:29:30,165 - Epoch: [0][ 1200/ 1236] Overall Loss 1.666229 Objective Loss 1.666229 LR 0.001000 Time 0.017002 +2023-10-02 20:29:30,327 - Epoch: [0][ 1210/ 1236] Overall Loss 1.661537 Objective Loss 1.661537 LR 0.001000 Time 0.016996 +2023-10-02 20:29:30,488 - Epoch: [0][ 1220/ 1236] Overall Loss 1.656808 Objective Loss 1.656808 LR 0.001000 Time 0.016988 +2023-10-02 20:29:30,690 - Epoch: [0][ 1230/ 1236] Overall Loss 1.651959 Objective Loss 1.651959 LR 0.001000 Time 0.017014 +2023-10-02 20:29:30,783 - Epoch: [0][ 1236/ 1236] Overall Loss 1.649043 Objective Loss 1.649043 Top1 57.230143 Top5 88.594705 LR 0.001000 Time 0.017006 +2023-10-02 20:29:30,918 - --- validate (epoch=0)----------- +2023-10-02 20:29:30,918 - 29943 samples (256 per mini-batch) +2023-10-02 20:29:31,319 - Epoch: [0][ 10/ 117] Loss 0.976131 Top1 57.382812 Top5 89.648438 +2023-10-02 20:29:31,426 - Epoch: [0][ 20/ 117] Loss 0.986808 Top1 56.777344 Top5 89.902344 +2023-10-02 20:29:31,528 - Epoch: [0][ 30/ 117] Loss 0.984363 Top1 56.210938 Top5 89.960938 +2023-10-02 20:29:31,632 - Epoch: [0][ 40/ 117] Loss 0.979813 Top1 56.337891 Top5 89.951172 +2023-10-02 20:29:31,734 - Epoch: [0][ 50/ 117] Loss 0.972696 Top1 56.476562 Top5 90.078125 +2023-10-02 20:29:31,839 - Epoch: [0][ 60/ 117] Loss 0.975039 Top1 56.393229 Top5 90.169271 +2023-10-02 20:29:31,941 - Epoch: [0][ 70/ 117] Loss 0.971929 Top1 56.428571 Top5 90.089286 +2023-10-02 20:29:32,043 - Epoch: [0][ 80/ 117] Loss 0.973728 Top1 56.420898 Top5 90.073242 +2023-10-02 20:29:32,146 - Epoch: [0][ 90/ 117] Loss 0.973328 Top1 56.358507 Top5 90.073785 +2023-10-02 20:29:32,250 - Epoch: [0][ 100/ 117] Loss 0.972897 Top1 56.273438 Top5 90.042969 +2023-10-02 20:29:32,358 - Epoch: [0][ 110/ 117] Loss 0.973993 Top1 56.189631 Top5 90.035511 +2023-10-02 20:29:32,417 - Epoch: [0][ 117/ 117] Loss 0.973671 Top1 56.183415 Top5 89.970945 +2023-10-02 20:29:32,544 - ==> Top1: 56.183 Top5: 89.971 Loss: 0.974 + +2023-10-02 20:29:32,545 - ==> Confusion: +[[ 816 0 6 0 14 4 0 1 10 109 2 2 1 37 5 2 7 20 2 0 12] + [ 1 707 12 2 46 61 1 25 21 0 31 3 2 29 38 1 49 0 84 3 15] + [ 27 1 583 11 14 20 184 77 0 12 12 18 3 30 5 10 3 1 6 17 22] + [ 3 2 21 715 5 47 17 8 4 0 38 2 15 17 69 5 20 23 58 0 20] + [ 38 4 4 1 878 9 0 0 0 11 1 0 0 19 19 12 29 1 7 1 16] + [ 6 39 25 5 21 682 3 69 5 3 6 11 9 141 15 4 23 1 16 14 18] + [ 1 5 112 4 4 15 944 43 0 0 5 5 0 0 1 18 0 0 7 16 11] + [ 4 5 57 8 8 65 8 831 3 1 14 5 3 12 4 0 0 0 153 22 15] + [ 60 6 1 0 1 3 0 0 819 57 4 0 3 32 70 0 12 7 10 0 4] + [ 315 1 1 1 20 1 1 1 30 638 0 0 1 59 16 1 1 6 5 0 21] + [ 9 42 49 13 17 65 5 40 23 5 526 1 0 6 8 0 15 1 209 1 18] + [ 1 0 2 0 1 27 2 5 0 0 0 792 70 41 0 7 9 13 4 54 7] + [ 1 0 0 4 0 11 2 1 4 0 0 228 658 21 4 1 33 89 4 2 5] + [ 13 1 5 0 5 52 1 6 4 24 0 33 2 914 11 4 18 9 1 5 11] + [ 19 21 0 11 48 5 0 0 49 21 2 0 3 6 867 1 16 8 2 0 22] + [ 2 0 2 0 20 1 5 0 0 0 0 89 6 2 0 906 7 67 9 1 17] + [ 5 13 0 5 43 31 0 1 10 1 1 27 2 17 15 26 936 6 1 0 21] + [ 3 0 0 2 0 2 2 0 6 1 0 76 150 5 2 9 4 757 3 1 15] + [ 4 7 6 31 2 12 0 63 12 0 20 0 2 1 29 0 9 0 848 5 17] + [ 0 0 9 0 0 8 16 52 0 0 0 70 2 34 0 9 2 1 9 928 12] + [ 356 361 152 116 320 693 40 250 154 306 117 483 341 577 527 195 362 278 620 579 1078]] + +2023-10-02 20:29:32,546 - ==> Best [Top1: 56.183 Top5: 89.971 Sparsity:0.00 Params: 169472 on epoch: 0] +2023-10-02 20:29:32,546 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:29:32,553 - + +2023-10-02 20:29:32,553 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:29:33,459 - Epoch: [1][ 10/ 1236] Overall Loss 0.998323 Objective Loss 0.998323 LR 0.001000 Time 0.090553 +2023-10-02 20:29:33,619 - Epoch: [1][ 20/ 1236] Overall Loss 1.009624 Objective Loss 1.009624 LR 0.001000 Time 0.053272 +2023-10-02 20:29:33,779 - Epoch: [1][ 30/ 1236] Overall Loss 1.010244 Objective Loss 1.010244 LR 0.001000 Time 0.040819 +2023-10-02 20:29:33,939 - Epoch: [1][ 40/ 1236] Overall Loss 1.014783 Objective Loss 1.014783 LR 0.001000 Time 0.034607 +2023-10-02 20:29:34,098 - Epoch: [1][ 50/ 1236] Overall Loss 1.010928 Objective Loss 1.010928 LR 0.001000 Time 0.030872 +2023-10-02 20:29:34,262 - Epoch: [1][ 60/ 1236] Overall Loss 1.015953 Objective Loss 1.015953 LR 0.001000 Time 0.028452 +2023-10-02 20:29:34,423 - Epoch: [1][ 70/ 1236] Overall Loss 1.005329 Objective Loss 1.005329 LR 0.001000 Time 0.026674 +2023-10-02 20:29:34,583 - Epoch: [1][ 80/ 1236] Overall Loss 1.003348 Objective Loss 1.003348 LR 0.001000 Time 0.025346 +2023-10-02 20:29:34,741 - Epoch: [1][ 90/ 1236] Overall Loss 1.001600 Objective Loss 1.001600 LR 0.001000 Time 0.024276 +2023-10-02 20:29:34,900 - Epoch: [1][ 100/ 1236] Overall Loss 1.001433 Objective Loss 1.001433 LR 0.001000 Time 0.023440 +2023-10-02 20:29:35,059 - Epoch: [1][ 110/ 1236] Overall Loss 1.002290 Objective Loss 1.002290 LR 0.001000 Time 0.022746 +2023-10-02 20:29:35,218 - Epoch: [1][ 120/ 1236] Overall Loss 0.999363 Objective Loss 0.999363 LR 0.001000 Time 0.022179 +2023-10-02 20:29:35,377 - Epoch: [1][ 130/ 1236] Overall Loss 0.998120 Objective Loss 0.998120 LR 0.001000 Time 0.021689 +2023-10-02 20:29:35,537 - Epoch: [1][ 140/ 1236] Overall Loss 0.995912 Objective Loss 0.995912 LR 0.001000 Time 0.021279 +2023-10-02 20:29:35,695 - Epoch: [1][ 150/ 1236] Overall Loss 0.992902 Objective Loss 0.992902 LR 0.001000 Time 0.020913 +2023-10-02 20:29:35,855 - Epoch: [1][ 160/ 1236] Overall Loss 0.992915 Objective Loss 0.992915 LR 0.001000 Time 0.020602 +2023-10-02 20:29:36,013 - Epoch: [1][ 170/ 1236] Overall Loss 0.990588 Objective Loss 0.990588 LR 0.001000 Time 0.020319 +2023-10-02 20:29:36,180 - Epoch: [1][ 180/ 1236] Overall Loss 0.989282 Objective Loss 0.989282 LR 0.001000 Time 0.020120 +2023-10-02 20:29:36,349 - Epoch: [1][ 190/ 1236] Overall Loss 0.988336 Objective Loss 0.988336 LR 0.001000 Time 0.019947 +2023-10-02 20:29:36,522 - Epoch: [1][ 200/ 1236] Overall Loss 0.987864 Objective Loss 0.987864 LR 0.001000 Time 0.019812 +2023-10-02 20:29:36,688 - Epoch: [1][ 210/ 1236] Overall Loss 0.984803 Objective Loss 0.984803 LR 0.001000 Time 0.019656 +2023-10-02 20:29:36,861 - Epoch: [1][ 220/ 1236] Overall Loss 0.982538 Objective Loss 0.982538 LR 0.001000 Time 0.019550 +2023-10-02 20:29:37,031 - Epoch: [1][ 230/ 1236] Overall Loss 0.980301 Objective Loss 0.980301 LR 0.001000 Time 0.019434 +2023-10-02 20:29:37,204 - Epoch: [1][ 240/ 1236] Overall Loss 0.978029 Objective Loss 0.978029 LR 0.001000 Time 0.019345 +2023-10-02 20:29:37,371 - Epoch: [1][ 250/ 1236] Overall Loss 0.978619 Objective Loss 0.978619 LR 0.001000 Time 0.019237 +2023-10-02 20:29:37,545 - Epoch: [1][ 260/ 1236] Overall Loss 0.977719 Objective Loss 0.977719 LR 0.001000 Time 0.019165 +2023-10-02 20:29:37,708 - Epoch: [1][ 270/ 1236] Overall Loss 0.977773 Objective Loss 0.977773 LR 0.001000 Time 0.019059 +2023-10-02 20:29:37,873 - Epoch: [1][ 280/ 1236] Overall Loss 0.976844 Objective Loss 0.976844 LR 0.001000 Time 0.018964 +2023-10-02 20:29:38,035 - Epoch: [1][ 290/ 1236] Overall Loss 0.973578 Objective Loss 0.973578 LR 0.001000 Time 0.018869 +2023-10-02 20:29:38,199 - Epoch: [1][ 300/ 1236] Overall Loss 0.969515 Objective Loss 0.969515 LR 0.001000 Time 0.018786 +2023-10-02 20:29:38,361 - Epoch: [1][ 310/ 1236] Overall Loss 0.967092 Objective Loss 0.967092 LR 0.001000 Time 0.018703 +2023-10-02 20:29:38,526 - Epoch: [1][ 320/ 1236] Overall Loss 0.964108 Objective Loss 0.964108 LR 0.001000 Time 0.018632 +2023-10-02 20:29:38,690 - Epoch: [1][ 330/ 1236] Overall Loss 0.962569 Objective Loss 0.962569 LR 0.001000 Time 0.018562 +2023-10-02 20:29:38,855 - Epoch: [1][ 340/ 1236] Overall Loss 0.961649 Objective Loss 0.961649 LR 0.001000 Time 0.018500 +2023-10-02 20:29:39,017 - Epoch: [1][ 350/ 1236] Overall Loss 0.958429 Objective Loss 0.958429 LR 0.001000 Time 0.018434 +2023-10-02 20:29:39,182 - Epoch: [1][ 360/ 1236] Overall Loss 0.958079 Objective Loss 0.958079 LR 0.001000 Time 0.018379 +2023-10-02 20:29:39,346 - Epoch: [1][ 370/ 1236] Overall Loss 0.957927 Objective Loss 0.957927 LR 0.001000 Time 0.018325 +2023-10-02 20:29:39,511 - Epoch: [1][ 380/ 1236] Overall Loss 0.955733 Objective Loss 0.955733 LR 0.001000 Time 0.018275 +2023-10-02 20:29:39,675 - Epoch: [1][ 390/ 1236] Overall Loss 0.954561 Objective Loss 0.954561 LR 0.001000 Time 0.018226 +2023-10-02 20:29:39,838 - Epoch: [1][ 400/ 1236] Overall Loss 0.952868 Objective Loss 0.952868 LR 0.001000 Time 0.018178 +2023-10-02 20:29:39,999 - Epoch: [1][ 410/ 1236] Overall Loss 0.951984 Objective Loss 0.951984 LR 0.001000 Time 0.018127 +2023-10-02 20:29:40,162 - Epoch: [1][ 420/ 1236] Overall Loss 0.951376 Objective Loss 0.951376 LR 0.001000 Time 0.018081 +2023-10-02 20:29:40,323 - Epoch: [1][ 430/ 1236] Overall Loss 0.948843 Objective Loss 0.948843 LR 0.001000 Time 0.018035 +2023-10-02 20:29:40,486 - Epoch: [1][ 440/ 1236] Overall Loss 0.946861 Objective Loss 0.946861 LR 0.001000 Time 0.017993 +2023-10-02 20:29:40,648 - Epoch: [1][ 450/ 1236] Overall Loss 0.945360 Objective Loss 0.945360 LR 0.001000 Time 0.017953 +2023-10-02 20:29:40,811 - Epoch: [1][ 460/ 1236] Overall Loss 0.944568 Objective Loss 0.944568 LR 0.001000 Time 0.017917 +2023-10-02 20:29:40,974 - Epoch: [1][ 470/ 1236] Overall Loss 0.943626 Objective Loss 0.943626 LR 0.001000 Time 0.017881 +2023-10-02 20:29:41,137 - Epoch: [1][ 480/ 1236] Overall Loss 0.942766 Objective Loss 0.942766 LR 0.001000 Time 0.017848 +2023-10-02 20:29:41,299 - Epoch: [1][ 490/ 1236] Overall Loss 0.940870 Objective Loss 0.940870 LR 0.001000 Time 0.017814 +2023-10-02 20:29:41,463 - Epoch: [1][ 500/ 1236] Overall Loss 0.940469 Objective Loss 0.940469 LR 0.001000 Time 0.017785 +2023-10-02 20:29:41,626 - Epoch: [1][ 510/ 1236] Overall Loss 0.940434 Objective Loss 0.940434 LR 0.001000 Time 0.017754 +2023-10-02 20:29:41,790 - Epoch: [1][ 520/ 1236] Overall Loss 0.939145 Objective Loss 0.939145 LR 0.001000 Time 0.017729 +2023-10-02 20:29:41,953 - Epoch: [1][ 530/ 1236] Overall Loss 0.938434 Objective Loss 0.938434 LR 0.001000 Time 0.017700 +2023-10-02 20:29:42,118 - Epoch: [1][ 540/ 1236] Overall Loss 0.937306 Objective Loss 0.937306 LR 0.001000 Time 0.017677 +2023-10-02 20:29:42,282 - Epoch: [1][ 550/ 1236] Overall Loss 0.936309 Objective Loss 0.936309 LR 0.001000 Time 0.017653 +2023-10-02 20:29:42,447 - Epoch: [1][ 560/ 1236] Overall Loss 0.934290 Objective Loss 0.934290 LR 0.001000 Time 0.017632 +2023-10-02 20:29:42,610 - Epoch: [1][ 570/ 1236] Overall Loss 0.931697 Objective Loss 0.931697 LR 0.001000 Time 0.017608 +2023-10-02 20:29:42,775 - Epoch: [1][ 580/ 1236] Overall Loss 0.929879 Objective Loss 0.929879 LR 0.001000 Time 0.017588 +2023-10-02 20:29:42,937 - Epoch: [1][ 590/ 1236] Overall Loss 0.927496 Objective Loss 0.927496 LR 0.001000 Time 0.017565 +2023-10-02 20:29:43,103 - Epoch: [1][ 600/ 1236] Overall Loss 0.926155 Objective Loss 0.926155 LR 0.001000 Time 0.017547 +2023-10-02 20:29:43,265 - Epoch: [1][ 610/ 1236] Overall Loss 0.925046 Objective Loss 0.925046 LR 0.001000 Time 0.017526 +2023-10-02 20:29:43,430 - Epoch: [1][ 620/ 1236] Overall Loss 0.924173 Objective Loss 0.924173 LR 0.001000 Time 0.017508 +2023-10-02 20:29:43,593 - Epoch: [1][ 630/ 1236] Overall Loss 0.923398 Objective Loss 0.923398 LR 0.001000 Time 0.017488 +2023-10-02 20:29:43,758 - Epoch: [1][ 640/ 1236] Overall Loss 0.922094 Objective Loss 0.922094 LR 0.001000 Time 0.017473 +2023-10-02 20:29:43,921 - Epoch: [1][ 650/ 1236] Overall Loss 0.920523 Objective Loss 0.920523 LR 0.001000 Time 0.017454 +2023-10-02 20:29:44,087 - Epoch: [1][ 660/ 1236] Overall Loss 0.919240 Objective Loss 0.919240 LR 0.001000 Time 0.017439 +2023-10-02 20:29:44,250 - Epoch: [1][ 670/ 1236] Overall Loss 0.917085 Objective Loss 0.917085 LR 0.001000 Time 0.017422 +2023-10-02 20:29:44,415 - Epoch: [1][ 680/ 1236] Overall Loss 0.916313 Objective Loss 0.916313 LR 0.001000 Time 0.017408 +2023-10-02 20:29:44,578 - Epoch: [1][ 690/ 1236] Overall Loss 0.914568 Objective Loss 0.914568 LR 0.001000 Time 0.017392 +2023-10-02 20:29:44,744 - Epoch: [1][ 700/ 1236] Overall Loss 0.913081 Objective Loss 0.913081 LR 0.001000 Time 0.017379 +2023-10-02 20:29:44,907 - Epoch: [1][ 710/ 1236] Overall Loss 0.912046 Objective Loss 0.912046 LR 0.001000 Time 0.017364 +2023-10-02 20:29:45,072 - Epoch: [1][ 720/ 1236] Overall Loss 0.911261 Objective Loss 0.911261 LR 0.001000 Time 0.017351 +2023-10-02 20:29:45,235 - Epoch: [1][ 730/ 1236] Overall Loss 0.909942 Objective Loss 0.909942 LR 0.001000 Time 0.017337 +2023-10-02 20:29:45,401 - Epoch: [1][ 740/ 1236] Overall Loss 0.908960 Objective Loss 0.908960 LR 0.001000 Time 0.017326 +2023-10-02 20:29:45,564 - Epoch: [1][ 750/ 1236] Overall Loss 0.907651 Objective Loss 0.907651 LR 0.001000 Time 0.017312 +2023-10-02 20:29:45,729 - Epoch: [1][ 760/ 1236] Overall Loss 0.905870 Objective Loss 0.905870 LR 0.001000 Time 0.017301 +2023-10-02 20:29:45,893 - Epoch: [1][ 770/ 1236] Overall Loss 0.904375 Objective Loss 0.904375 LR 0.001000 Time 0.017288 +2023-10-02 20:29:46,058 - Epoch: [1][ 780/ 1236] Overall Loss 0.903252 Objective Loss 0.903252 LR 0.001000 Time 0.017278 +2023-10-02 20:29:46,221 - Epoch: [1][ 790/ 1236] Overall Loss 0.902163 Objective Loss 0.902163 LR 0.001000 Time 0.017266 +2023-10-02 20:29:46,387 - Epoch: [1][ 800/ 1236] Overall Loss 0.900815 Objective Loss 0.900815 LR 0.001000 Time 0.017256 +2023-10-02 20:29:46,550 - Epoch: [1][ 810/ 1236] Overall Loss 0.899665 Objective Loss 0.899665 LR 0.001000 Time 0.017244 +2023-10-02 20:29:46,715 - Epoch: [1][ 820/ 1236] Overall Loss 0.898200 Objective Loss 0.898200 LR 0.001000 Time 0.017235 +2023-10-02 20:29:46,878 - Epoch: [1][ 830/ 1236] Overall Loss 0.897290 Objective Loss 0.897290 LR 0.001000 Time 0.017223 +2023-10-02 20:29:47,043 - Epoch: [1][ 840/ 1236] Overall Loss 0.896358 Objective Loss 0.896358 LR 0.001000 Time 0.017215 +2023-10-02 20:29:47,206 - Epoch: [1][ 850/ 1236] Overall Loss 0.895259 Objective Loss 0.895259 LR 0.001000 Time 0.017204 +2023-10-02 20:29:47,371 - Epoch: [1][ 860/ 1236] Overall Loss 0.894120 Objective Loss 0.894120 LR 0.001000 Time 0.017195 +2023-10-02 20:29:47,535 - Epoch: [1][ 870/ 1236] Overall Loss 0.892806 Objective Loss 0.892806 LR 0.001000 Time 0.017184 +2023-10-02 20:29:47,700 - Epoch: [1][ 880/ 1236] Overall Loss 0.891544 Objective Loss 0.891544 LR 0.001000 Time 0.017177 +2023-10-02 20:29:47,863 - Epoch: [1][ 890/ 1236] Overall Loss 0.890354 Objective Loss 0.890354 LR 0.001000 Time 0.017167 +2023-10-02 20:29:48,029 - Epoch: [1][ 900/ 1236] Overall Loss 0.888979 Objective Loss 0.888979 LR 0.001000 Time 0.017159 +2023-10-02 20:29:48,192 - Epoch: [1][ 910/ 1236] Overall Loss 0.887750 Objective Loss 0.887750 LR 0.001000 Time 0.017150 +2023-10-02 20:29:48,357 - Epoch: [1][ 920/ 1236] Overall Loss 0.887074 Objective Loss 0.887074 LR 0.001000 Time 0.017143 +2023-10-02 20:29:48,521 - Epoch: [1][ 930/ 1236] Overall Loss 0.886646 Objective Loss 0.886646 LR 0.001000 Time 0.017133 +2023-10-02 20:29:48,686 - Epoch: [1][ 940/ 1236] Overall Loss 0.886203 Objective Loss 0.886203 LR 0.001000 Time 0.017127 +2023-10-02 20:29:48,849 - Epoch: [1][ 950/ 1236] Overall Loss 0.885434 Objective Loss 0.885434 LR 0.001000 Time 0.017118 +2023-10-02 20:29:49,015 - Epoch: [1][ 960/ 1236] Overall Loss 0.884454 Objective Loss 0.884454 LR 0.001000 Time 0.017112 +2023-10-02 20:29:49,178 - Epoch: [1][ 970/ 1236] Overall Loss 0.883103 Objective Loss 0.883103 LR 0.001000 Time 0.017103 +2023-10-02 20:29:49,344 - Epoch: [1][ 980/ 1236] Overall Loss 0.882000 Objective Loss 0.882000 LR 0.001000 Time 0.017097 +2023-10-02 20:29:49,507 - Epoch: [1][ 990/ 1236] Overall Loss 0.880359 Objective Loss 0.880359 LR 0.001000 Time 0.017089 +2023-10-02 20:29:49,672 - Epoch: [1][ 1000/ 1236] Overall Loss 0.879519 Objective Loss 0.879519 LR 0.001000 Time 0.017083 +2023-10-02 20:29:49,835 - Epoch: [1][ 1010/ 1236] Overall Loss 0.878280 Objective Loss 0.878280 LR 0.001000 Time 0.017075 +2023-10-02 20:29:50,001 - Epoch: [1][ 1020/ 1236] Overall Loss 0.877079 Objective Loss 0.877079 LR 0.001000 Time 0.017070 +2023-10-02 20:29:50,164 - Epoch: [1][ 1030/ 1236] Overall Loss 0.875989 Objective Loss 0.875989 LR 0.001000 Time 0.017062 +2023-10-02 20:29:50,330 - Epoch: [1][ 1040/ 1236] Overall Loss 0.874856 Objective Loss 0.874856 LR 0.001000 Time 0.017057 +2023-10-02 20:29:50,493 - Epoch: [1][ 1050/ 1236] Overall Loss 0.874043 Objective Loss 0.874043 LR 0.001000 Time 0.017050 +2023-10-02 20:29:50,659 - Epoch: [1][ 1060/ 1236] Overall Loss 0.872660 Objective Loss 0.872660 LR 0.001000 Time 0.017045 +2023-10-02 20:29:50,822 - Epoch: [1][ 1070/ 1236] Overall Loss 0.871642 Objective Loss 0.871642 LR 0.001000 Time 0.017038 +2023-10-02 20:29:50,988 - Epoch: [1][ 1080/ 1236] Overall Loss 0.870820 Objective Loss 0.870820 LR 0.001000 Time 0.017033 +2023-10-02 20:29:51,152 - Epoch: [1][ 1090/ 1236] Overall Loss 0.869430 Objective Loss 0.869430 LR 0.001000 Time 0.017027 +2023-10-02 20:29:51,318 - Epoch: [1][ 1100/ 1236] Overall Loss 0.868111 Objective Loss 0.868111 LR 0.001000 Time 0.017023 +2023-10-02 20:29:51,481 - Epoch: [1][ 1110/ 1236] Overall Loss 0.867214 Objective Loss 0.867214 LR 0.001000 Time 0.017016 +2023-10-02 20:29:51,647 - Epoch: [1][ 1120/ 1236] Overall Loss 0.865910 Objective Loss 0.865910 LR 0.001000 Time 0.017012 +2023-10-02 20:29:51,810 - Epoch: [1][ 1130/ 1236] Overall Loss 0.865116 Objective Loss 0.865116 LR 0.001000 Time 0.017006 +2023-10-02 20:29:51,976 - Epoch: [1][ 1140/ 1236] Overall Loss 0.864525 Objective Loss 0.864525 LR 0.001000 Time 0.017001 +2023-10-02 20:29:52,140 - Epoch: [1][ 1150/ 1236] Overall Loss 0.862988 Objective Loss 0.862988 LR 0.001000 Time 0.016996 +2023-10-02 20:29:52,306 - Epoch: [1][ 1160/ 1236] Overall Loss 0.861717 Objective Loss 0.861717 LR 0.001000 Time 0.016992 +2023-10-02 20:29:52,470 - Epoch: [1][ 1170/ 1236] Overall Loss 0.860609 Objective Loss 0.860609 LR 0.001000 Time 0.016987 +2023-10-02 20:29:52,636 - Epoch: [1][ 1180/ 1236] Overall Loss 0.859502 Objective Loss 0.859502 LR 0.001000 Time 0.016983 +2023-10-02 20:29:52,800 - Epoch: [1][ 1190/ 1236] Overall Loss 0.858554 Objective Loss 0.858554 LR 0.001000 Time 0.016978 +2023-10-02 20:29:52,963 - Epoch: [1][ 1200/ 1236] Overall Loss 0.857685 Objective Loss 0.857685 LR 0.001000 Time 0.016973 +2023-10-02 20:29:53,127 - Epoch: [1][ 1210/ 1236] Overall Loss 0.856650 Objective Loss 0.856650 LR 0.001000 Time 0.016967 +2023-10-02 20:29:53,291 - Epoch: [1][ 1220/ 1236] Overall Loss 0.855628 Objective Loss 0.855628 LR 0.001000 Time 0.016962 +2023-10-02 20:29:53,501 - Epoch: [1][ 1230/ 1236] Overall Loss 0.854895 Objective Loss 0.854895 LR 0.001000 Time 0.016995 +2023-10-02 20:29:53,596 - Epoch: [1][ 1236/ 1236] Overall Loss 0.854470 Objective Loss 0.854470 Top1 64.358452 Top5 94.501018 LR 0.001000 Time 0.016989 +2023-10-02 20:29:53,733 - --- validate (epoch=1)----------- +2023-10-02 20:29:53,733 - 29943 samples (256 per mini-batch) +2023-10-02 20:29:54,164 - Epoch: [1][ 10/ 117] Loss 0.689302 Top1 64.921875 Top5 93.554688 +2023-10-02 20:29:54,269 - Epoch: [1][ 20/ 117] Loss 0.684404 Top1 64.648438 Top5 93.691406 +2023-10-02 20:29:54,371 - Epoch: [1][ 30/ 117] Loss 0.677350 Top1 64.895833 Top5 93.671875 +2023-10-02 20:29:54,473 - Epoch: [1][ 40/ 117] Loss 0.687991 Top1 64.619141 Top5 93.476562 +2023-10-02 20:29:54,575 - Epoch: [1][ 50/ 117] Loss 0.689858 Top1 64.617188 Top5 93.484375 +2023-10-02 20:29:54,676 - Epoch: [1][ 60/ 117] Loss 0.688566 Top1 64.759115 Top5 93.704427 +2023-10-02 20:29:54,779 - Epoch: [1][ 70/ 117] Loss 0.686510 Top1 64.743304 Top5 93.677455 +2023-10-02 20:29:54,883 - Epoch: [1][ 80/ 117] Loss 0.692886 Top1 64.501953 Top5 93.735352 +2023-10-02 20:29:54,993 - Epoch: [1][ 90/ 117] Loss 0.694719 Top1 64.331597 Top5 93.671875 +2023-10-02 20:29:55,098 - Epoch: [1][ 100/ 117] Loss 0.692814 Top1 64.390625 Top5 93.792969 +2023-10-02 20:29:55,218 - Epoch: [1][ 110/ 117] Loss 0.693490 Top1 64.385653 Top5 93.746449 +2023-10-02 20:29:55,276 - Epoch: [1][ 117/ 117] Loss 0.692469 Top1 64.399025 Top5 93.731423 +2023-10-02 20:29:55,409 - ==> Top1: 64.399 Top5: 93.731 Loss: 0.692 + +2023-10-02 20:29:55,409 - ==> Confusion: +[[ 746 1 10 2 17 4 0 2 24 174 3 2 1 21 8 0 10 6 2 1 16] + [ 1 938 5 1 7 42 1 37 13 0 29 1 0 2 9 0 6 0 33 1 5] + [ 17 2 849 14 6 2 29 33 0 5 33 2 2 8 2 6 1 1 24 4 16] + [ 1 3 13 848 0 10 1 6 5 0 90 0 2 7 46 3 4 9 27 0 14] + [ 21 28 11 0 885 22 0 0 7 20 2 1 0 18 8 3 15 1 2 0 6] + [ 2 118 6 7 2 708 3 107 9 2 33 8 4 62 2 0 3 3 6 24 7] + [ 0 9 131 1 2 4 946 54 0 0 18 1 1 1 0 3 1 0 4 8 7] + [ 1 8 23 1 6 27 5 983 6 0 18 3 0 4 1 0 0 0 108 10 14] + [ 23 7 2 1 0 1 0 2 946 36 12 0 0 24 17 0 6 1 7 0 4] + [ 130 3 3 2 14 3 1 1 80 776 1 0 0 76 7 0 0 3 1 1 17] + [ 6 6 9 19 4 10 0 11 29 3 895 0 0 9 3 0 2 0 41 1 5] + [ 0 2 4 1 1 53 2 3 2 0 1 738 132 18 0 3 12 10 1 46 6] + [ 0 4 7 14 0 16 2 3 9 0 2 76 849 5 5 11 18 30 6 6 5] + [ 2 0 2 3 10 67 1 6 25 8 12 8 1 950 1 3 7 1 0 6 6] + [ 8 13 2 17 17 6 0 0 84 8 7 0 4 3 889 0 5 1 10 0 27] + [ 0 2 14 5 2 15 6 3 2 2 0 39 23 2 0 948 23 22 1 9 16] + [ 1 41 6 1 30 12 1 0 8 0 8 22 1 8 3 6 997 0 4 1 11] + [ 0 0 3 8 0 4 0 0 16 2 0 10 181 5 4 13 6 772 2 0 12] + [ 0 14 6 16 1 0 0 46 12 0 25 0 2 0 22 0 1 0 910 2 11] + [ 0 5 2 0 1 8 16 53 1 0 4 26 6 6 0 3 0 0 5 1009 7] + [ 233 433 216 134 177 475 58 249 255 288 474 189 496 618 196 175 330 60 444 704 1701]] + +2023-10-02 20:29:55,411 - ==> Best [Top1: 64.399 Top5: 93.731 Sparsity:0.00 Params: 169472 on epoch: 1] +2023-10-02 20:29:55,411 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:29:55,424 - + +2023-10-02 20:29:55,425 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:29:56,343 - Epoch: [2][ 10/ 1236] Overall Loss 0.755000 Objective Loss 0.755000 LR 0.001000 Time 0.091734 +2023-10-02 20:29:56,508 - Epoch: [2][ 20/ 1236] Overall Loss 0.744512 Objective Loss 0.744512 LR 0.001000 Time 0.054148 +2023-10-02 20:29:56,673 - Epoch: [2][ 30/ 1236] Overall Loss 0.730218 Objective Loss 0.730218 LR 0.001000 Time 0.041559 +2023-10-02 20:29:56,838 - Epoch: [2][ 40/ 1236] Overall Loss 0.733793 Objective Loss 0.733793 LR 0.001000 Time 0.035308 +2023-10-02 20:29:57,002 - Epoch: [2][ 50/ 1236] Overall Loss 0.730884 Objective Loss 0.730884 LR 0.001000 Time 0.031520 +2023-10-02 20:29:57,171 - Epoch: [2][ 60/ 1236] Overall Loss 0.726676 Objective Loss 0.726676 LR 0.001000 Time 0.029072 +2023-10-02 20:29:57,336 - Epoch: [2][ 70/ 1236] Overall Loss 0.727207 Objective Loss 0.727207 LR 0.001000 Time 0.027263 +2023-10-02 20:29:57,501 - Epoch: [2][ 80/ 1236] Overall Loss 0.724659 Objective Loss 0.724659 LR 0.001000 Time 0.025920 +2023-10-02 20:29:57,664 - Epoch: [2][ 90/ 1236] Overall Loss 0.721913 Objective Loss 0.721913 LR 0.001000 Time 0.024842 +2023-10-02 20:29:57,827 - Epoch: [2][ 100/ 1236] Overall Loss 0.716290 Objective Loss 0.716290 LR 0.001000 Time 0.023989 +2023-10-02 20:29:57,990 - Epoch: [2][ 110/ 1236] Overall Loss 0.717953 Objective Loss 0.717953 LR 0.001000 Time 0.023283 +2023-10-02 20:29:58,153 - Epoch: [2][ 120/ 1236] Overall Loss 0.715200 Objective Loss 0.715200 LR 0.001000 Time 0.022703 +2023-10-02 20:29:58,316 - Epoch: [2][ 130/ 1236] Overall Loss 0.708578 Objective Loss 0.708578 LR 0.001000 Time 0.022205 +2023-10-02 20:29:58,480 - Epoch: [2][ 140/ 1236] Overall Loss 0.705892 Objective Loss 0.705892 LR 0.001000 Time 0.021787 +2023-10-02 20:29:58,642 - Epoch: [2][ 150/ 1236] Overall Loss 0.704598 Objective Loss 0.704598 LR 0.001000 Time 0.021415 +2023-10-02 20:29:58,805 - Epoch: [2][ 160/ 1236] Overall Loss 0.703644 Objective Loss 0.703644 LR 0.001000 Time 0.021095 +2023-10-02 20:29:58,968 - Epoch: [2][ 170/ 1236] Overall Loss 0.700980 Objective Loss 0.700980 LR 0.001000 Time 0.020808 +2023-10-02 20:29:59,131 - Epoch: [2][ 180/ 1236] Overall Loss 0.699255 Objective Loss 0.699255 LR 0.001000 Time 0.020559 +2023-10-02 20:29:59,294 - Epoch: [2][ 190/ 1236] Overall Loss 0.698369 Objective Loss 0.698369 LR 0.001000 Time 0.020331 +2023-10-02 20:29:59,458 - Epoch: [2][ 200/ 1236] Overall Loss 0.696867 Objective Loss 0.696867 LR 0.001000 Time 0.020132 +2023-10-02 20:29:59,620 - Epoch: [2][ 210/ 1236] Overall Loss 0.696520 Objective Loss 0.696520 LR 0.001000 Time 0.019948 +2023-10-02 20:29:59,784 - Epoch: [2][ 220/ 1236] Overall Loss 0.694559 Objective Loss 0.694559 LR 0.001000 Time 0.019784 +2023-10-02 20:29:59,947 - Epoch: [2][ 230/ 1236] Overall Loss 0.692248 Objective Loss 0.692248 LR 0.001000 Time 0.019631 +2023-10-02 20:30:00,111 - Epoch: [2][ 240/ 1236] Overall Loss 0.691741 Objective Loss 0.691741 LR 0.001000 Time 0.019493 +2023-10-02 20:30:00,274 - Epoch: [2][ 250/ 1236] Overall Loss 0.692532 Objective Loss 0.692532 LR 0.001000 Time 0.019364 +2023-10-02 20:30:00,438 - Epoch: [2][ 260/ 1236] Overall Loss 0.694142 Objective Loss 0.694142 LR 0.001000 Time 0.019248 +2023-10-02 20:30:00,600 - Epoch: [2][ 270/ 1236] Overall Loss 0.694042 Objective Loss 0.694042 LR 0.001000 Time 0.019138 +2023-10-02 20:30:00,765 - Epoch: [2][ 280/ 1236] Overall Loss 0.694564 Objective Loss 0.694564 LR 0.001000 Time 0.019039 +2023-10-02 20:30:00,928 - Epoch: [2][ 290/ 1236] Overall Loss 0.696759 Objective Loss 0.696759 LR 0.001000 Time 0.018944 +2023-10-02 20:30:01,093 - Epoch: [2][ 300/ 1236] Overall Loss 0.697194 Objective Loss 0.697194 LR 0.001000 Time 0.018863 +2023-10-02 20:30:01,258 - Epoch: [2][ 310/ 1236] Overall Loss 0.697910 Objective Loss 0.697910 LR 0.001000 Time 0.018784 +2023-10-02 20:30:01,423 - Epoch: [2][ 320/ 1236] Overall Loss 0.697223 Objective Loss 0.697223 LR 0.001000 Time 0.018712 +2023-10-02 20:30:01,588 - Epoch: [2][ 330/ 1236] Overall Loss 0.696053 Objective Loss 0.696053 LR 0.001000 Time 0.018646 +2023-10-02 20:30:01,755 - Epoch: [2][ 340/ 1236] Overall Loss 0.694463 Objective Loss 0.694463 LR 0.001000 Time 0.018586 +2023-10-02 20:30:01,921 - Epoch: [2][ 350/ 1236] Overall Loss 0.692324 Objective Loss 0.692324 LR 0.001000 Time 0.018528 +2023-10-02 20:30:02,087 - Epoch: [2][ 360/ 1236] Overall Loss 0.692113 Objective Loss 0.692113 LR 0.001000 Time 0.018473 +2023-10-02 20:30:02,252 - Epoch: [2][ 370/ 1236] Overall Loss 0.692043 Objective Loss 0.692043 LR 0.001000 Time 0.018421 +2023-10-02 20:30:02,419 - Epoch: [2][ 380/ 1236] Overall Loss 0.691962 Objective Loss 0.691962 LR 0.001000 Time 0.018374 +2023-10-02 20:30:02,585 - Epoch: [2][ 390/ 1236] Overall Loss 0.691204 Objective Loss 0.691204 LR 0.001000 Time 0.018328 +2023-10-02 20:30:02,752 - Epoch: [2][ 400/ 1236] Overall Loss 0.690941 Objective Loss 0.690941 LR 0.001000 Time 0.018286 +2023-10-02 20:30:02,918 - Epoch: [2][ 410/ 1236] Overall Loss 0.691925 Objective Loss 0.691925 LR 0.001000 Time 0.018245 +2023-10-02 20:30:03,084 - Epoch: [2][ 420/ 1236] Overall Loss 0.691885 Objective Loss 0.691885 LR 0.001000 Time 0.018206 +2023-10-02 20:30:03,249 - Epoch: [2][ 430/ 1236] Overall Loss 0.691583 Objective Loss 0.691583 LR 0.001000 Time 0.018165 +2023-10-02 20:30:03,414 - Epoch: [2][ 440/ 1236] Overall Loss 0.691763 Objective Loss 0.691763 LR 0.001000 Time 0.018126 +2023-10-02 20:30:03,577 - Epoch: [2][ 450/ 1236] Overall Loss 0.691088 Objective Loss 0.691088 LR 0.001000 Time 0.018086 +2023-10-02 20:30:03,742 - Epoch: [2][ 460/ 1236] Overall Loss 0.690559 Objective Loss 0.690559 LR 0.001000 Time 0.018050 +2023-10-02 20:30:03,906 - Epoch: [2][ 470/ 1236] Overall Loss 0.689313 Objective Loss 0.689313 LR 0.001000 Time 0.018014 +2023-10-02 20:30:04,071 - Epoch: [2][ 480/ 1236] Overall Loss 0.689272 Objective Loss 0.689272 LR 0.001000 Time 0.017981 +2023-10-02 20:30:04,235 - Epoch: [2][ 490/ 1236] Overall Loss 0.689137 Objective Loss 0.689137 LR 0.001000 Time 0.017950 +2023-10-02 20:30:04,400 - Epoch: [2][ 500/ 1236] Overall Loss 0.689423 Objective Loss 0.689423 LR 0.001000 Time 0.017919 +2023-10-02 20:30:04,563 - Epoch: [2][ 510/ 1236] Overall Loss 0.688735 Objective Loss 0.688735 LR 0.001000 Time 0.017888 +2023-10-02 20:30:04,729 - Epoch: [2][ 520/ 1236] Overall Loss 0.688703 Objective Loss 0.688703 LR 0.001000 Time 0.017861 +2023-10-02 20:30:04,892 - Epoch: [2][ 530/ 1236] Overall Loss 0.688392 Objective Loss 0.688392 LR 0.001000 Time 0.017831 +2023-10-02 20:30:05,057 - Epoch: [2][ 540/ 1236] Overall Loss 0.688573 Objective Loss 0.688573 LR 0.001000 Time 0.017806 +2023-10-02 20:30:05,221 - Epoch: [2][ 550/ 1236] Overall Loss 0.688042 Objective Loss 0.688042 LR 0.001000 Time 0.017780 +2023-10-02 20:30:05,387 - Epoch: [2][ 560/ 1236] Overall Loss 0.687897 Objective Loss 0.687897 LR 0.001000 Time 0.017758 +2023-10-02 20:30:05,550 - Epoch: [2][ 570/ 1236] Overall Loss 0.687545 Objective Loss 0.687545 LR 0.001000 Time 0.017733 +2023-10-02 20:30:05,716 - Epoch: [2][ 580/ 1236] Overall Loss 0.686814 Objective Loss 0.686814 LR 0.001000 Time 0.017712 +2023-10-02 20:30:05,879 - Epoch: [2][ 590/ 1236] Overall Loss 0.685797 Objective Loss 0.685797 LR 0.001000 Time 0.017687 +2023-10-02 20:30:06,044 - Epoch: [2][ 600/ 1236] Overall Loss 0.685894 Objective Loss 0.685894 LR 0.001000 Time 0.017667 +2023-10-02 20:30:06,207 - Epoch: [2][ 610/ 1236] Overall Loss 0.686546 Objective Loss 0.686546 LR 0.001000 Time 0.017645 +2023-10-02 20:30:06,373 - Epoch: [2][ 620/ 1236] Overall Loss 0.686431 Objective Loss 0.686431 LR 0.001000 Time 0.017627 +2023-10-02 20:30:06,537 - Epoch: [2][ 630/ 1236] Overall Loss 0.686557 Objective Loss 0.686557 LR 0.001000 Time 0.017608 +2023-10-02 20:30:06,703 - Epoch: [2][ 640/ 1236] Overall Loss 0.685323 Objective Loss 0.685323 LR 0.001000 Time 0.017591 +2023-10-02 20:30:06,867 - Epoch: [2][ 650/ 1236] Overall Loss 0.684975 Objective Loss 0.684975 LR 0.001000 Time 0.017571 +2023-10-02 20:30:07,032 - Epoch: [2][ 660/ 1236] Overall Loss 0.684663 Objective Loss 0.684663 LR 0.001000 Time 0.017555 +2023-10-02 20:30:07,195 - Epoch: [2][ 670/ 1236] Overall Loss 0.684469 Objective Loss 0.684469 LR 0.001000 Time 0.017536 +2023-10-02 20:30:07,360 - Epoch: [2][ 680/ 1236] Overall Loss 0.684276 Objective Loss 0.684276 LR 0.001000 Time 0.017520 +2023-10-02 20:30:07,523 - Epoch: [2][ 690/ 1236] Overall Loss 0.683442 Objective Loss 0.683442 LR 0.001000 Time 0.017503 +2023-10-02 20:30:07,689 - Epoch: [2][ 700/ 1236] Overall Loss 0.683001 Objective Loss 0.683001 LR 0.001000 Time 0.017489 +2023-10-02 20:30:07,852 - Epoch: [2][ 710/ 1236] Overall Loss 0.683341 Objective Loss 0.683341 LR 0.001000 Time 0.017472 +2023-10-02 20:30:08,018 - Epoch: [2][ 720/ 1236] Overall Loss 0.682428 Objective Loss 0.682428 LR 0.001000 Time 0.017459 +2023-10-02 20:30:08,182 - Epoch: [2][ 730/ 1236] Overall Loss 0.682483 Objective Loss 0.682483 LR 0.001000 Time 0.017444 +2023-10-02 20:30:08,347 - Epoch: [2][ 740/ 1236] Overall Loss 0.681875 Objective Loss 0.681875 LR 0.001000 Time 0.017431 +2023-10-02 20:30:08,510 - Epoch: [2][ 750/ 1236] Overall Loss 0.680998 Objective Loss 0.680998 LR 0.001000 Time 0.017416 +2023-10-02 20:30:08,676 - Epoch: [2][ 760/ 1236] Overall Loss 0.680256 Objective Loss 0.680256 LR 0.001000 Time 0.017404 +2023-10-02 20:30:08,839 - Epoch: [2][ 770/ 1236] Overall Loss 0.680332 Objective Loss 0.680332 LR 0.001000 Time 0.017390 +2023-10-02 20:30:09,005 - Epoch: [2][ 780/ 1236] Overall Loss 0.679827 Objective Loss 0.679827 LR 0.001000 Time 0.017379 +2023-10-02 20:30:09,169 - Epoch: [2][ 790/ 1236] Overall Loss 0.679389 Objective Loss 0.679389 LR 0.001000 Time 0.017366 +2023-10-02 20:30:09,334 - Epoch: [2][ 800/ 1236] Overall Loss 0.679052 Objective Loss 0.679052 LR 0.001000 Time 0.017355 +2023-10-02 20:30:09,497 - Epoch: [2][ 810/ 1236] Overall Loss 0.678259 Objective Loss 0.678259 LR 0.001000 Time 0.017342 +2023-10-02 20:30:09,663 - Epoch: [2][ 820/ 1236] Overall Loss 0.677989 Objective Loss 0.677989 LR 0.001000 Time 0.017332 +2023-10-02 20:30:09,826 - Epoch: [2][ 830/ 1236] Overall Loss 0.677400 Objective Loss 0.677400 LR 0.001000 Time 0.017320 +2023-10-02 20:30:09,992 - Epoch: [2][ 840/ 1236] Overall Loss 0.677184 Objective Loss 0.677184 LR 0.001000 Time 0.017311 +2023-10-02 20:30:10,156 - Epoch: [2][ 850/ 1236] Overall Loss 0.676893 Objective Loss 0.676893 LR 0.001000 Time 0.017299 +2023-10-02 20:30:10,321 - Epoch: [2][ 860/ 1236] Overall Loss 0.676708 Objective Loss 0.676708 LR 0.001000 Time 0.017290 +2023-10-02 20:30:10,484 - Epoch: [2][ 870/ 1236] Overall Loss 0.676416 Objective Loss 0.676416 LR 0.001000 Time 0.017279 +2023-10-02 20:30:10,649 - Epoch: [2][ 880/ 1236] Overall Loss 0.676124 Objective Loss 0.676124 LR 0.001000 Time 0.017269 +2023-10-02 20:30:10,813 - Epoch: [2][ 890/ 1236] Overall Loss 0.675564 Objective Loss 0.675564 LR 0.001000 Time 0.017259 +2023-10-02 20:30:10,978 - Epoch: [2][ 900/ 1236] Overall Loss 0.674681 Objective Loss 0.674681 LR 0.001000 Time 0.017250 +2023-10-02 20:30:11,141 - Epoch: [2][ 910/ 1236] Overall Loss 0.674578 Objective Loss 0.674578 LR 0.001000 Time 0.017240 +2023-10-02 20:30:11,307 - Epoch: [2][ 920/ 1236] Overall Loss 0.674212 Objective Loss 0.674212 LR 0.001000 Time 0.017232 +2023-10-02 20:30:11,471 - Epoch: [2][ 930/ 1236] Overall Loss 0.674658 Objective Loss 0.674658 LR 0.001000 Time 0.017222 +2023-10-02 20:30:11,636 - Epoch: [2][ 940/ 1236] Overall Loss 0.674517 Objective Loss 0.674517 LR 0.001000 Time 0.017215 +2023-10-02 20:30:11,799 - Epoch: [2][ 950/ 1236] Overall Loss 0.674011 Objective Loss 0.674011 LR 0.001000 Time 0.017205 +2023-10-02 20:30:11,965 - Epoch: [2][ 960/ 1236] Overall Loss 0.673857 Objective Loss 0.673857 LR 0.001000 Time 0.017198 +2023-10-02 20:30:12,128 - Epoch: [2][ 970/ 1236] Overall Loss 0.673178 Objective Loss 0.673178 LR 0.001000 Time 0.017189 +2023-10-02 20:30:12,294 - Epoch: [2][ 980/ 1236] Overall Loss 0.672060 Objective Loss 0.672060 LR 0.001000 Time 0.017182 +2023-10-02 20:30:12,457 - Epoch: [2][ 990/ 1236] Overall Loss 0.671550 Objective Loss 0.671550 LR 0.001000 Time 0.017173 +2023-10-02 20:30:12,623 - Epoch: [2][ 1000/ 1236] Overall Loss 0.671079 Objective Loss 0.671079 LR 0.001000 Time 0.017167 +2023-10-02 20:30:12,787 - Epoch: [2][ 1010/ 1236] Overall Loss 0.670086 Objective Loss 0.670086 LR 0.001000 Time 0.017158 +2023-10-02 20:30:12,951 - Epoch: [2][ 1020/ 1236] Overall Loss 0.669528 Objective Loss 0.669528 LR 0.001000 Time 0.017151 +2023-10-02 20:30:13,114 - Epoch: [2][ 1030/ 1236] Overall Loss 0.669019 Objective Loss 0.669019 LR 0.001000 Time 0.017143 +2023-10-02 20:30:13,279 - Epoch: [2][ 1040/ 1236] Overall Loss 0.668252 Objective Loss 0.668252 LR 0.001000 Time 0.017136 +2023-10-02 20:30:13,443 - Epoch: [2][ 1050/ 1236] Overall Loss 0.667599 Objective Loss 0.667599 LR 0.001000 Time 0.017129 +2023-10-02 20:30:13,609 - Epoch: [2][ 1060/ 1236] Overall Loss 0.666845 Objective Loss 0.666845 LR 0.001000 Time 0.017124 +2023-10-02 20:30:13,774 - Epoch: [2][ 1070/ 1236] Overall Loss 0.666601 Objective Loss 0.666601 LR 0.001000 Time 0.017117 +2023-10-02 20:30:13,939 - Epoch: [2][ 1080/ 1236] Overall Loss 0.666480 Objective Loss 0.666480 LR 0.001000 Time 0.017111 +2023-10-02 20:30:14,103 - Epoch: [2][ 1090/ 1236] Overall Loss 0.666288 Objective Loss 0.666288 LR 0.001000 Time 0.017105 +2023-10-02 20:30:14,269 - Epoch: [2][ 1100/ 1236] Overall Loss 0.666112 Objective Loss 0.666112 LR 0.001000 Time 0.017100 +2023-10-02 20:30:14,433 - Epoch: [2][ 1110/ 1236] Overall Loss 0.665456 Objective Loss 0.665456 LR 0.001000 Time 0.017093 +2023-10-02 20:30:14,598 - Epoch: [2][ 1120/ 1236] Overall Loss 0.664779 Objective Loss 0.664779 LR 0.001000 Time 0.017088 +2023-10-02 20:30:14,762 - Epoch: [2][ 1130/ 1236] Overall Loss 0.664128 Objective Loss 0.664128 LR 0.001000 Time 0.017081 +2023-10-02 20:30:14,928 - Epoch: [2][ 1140/ 1236] Overall Loss 0.663810 Objective Loss 0.663810 LR 0.001000 Time 0.017077 +2023-10-02 20:30:15,092 - Epoch: [2][ 1150/ 1236] Overall Loss 0.663143 Objective Loss 0.663143 LR 0.001000 Time 0.017071 +2023-10-02 20:30:15,257 - Epoch: [2][ 1160/ 1236] Overall Loss 0.662840 Objective Loss 0.662840 LR 0.001000 Time 0.017065 +2023-10-02 20:30:15,420 - Epoch: [2][ 1170/ 1236] Overall Loss 0.662333 Objective Loss 0.662333 LR 0.001000 Time 0.017059 +2023-10-02 20:30:15,585 - Epoch: [2][ 1180/ 1236] Overall Loss 0.661949 Objective Loss 0.661949 LR 0.001000 Time 0.017054 +2023-10-02 20:30:15,749 - Epoch: [2][ 1190/ 1236] Overall Loss 0.661942 Objective Loss 0.661942 LR 0.001000 Time 0.017047 +2023-10-02 20:30:15,914 - Epoch: [2][ 1200/ 1236] Overall Loss 0.661406 Objective Loss 0.661406 LR 0.001000 Time 0.017043 +2023-10-02 20:30:16,077 - Epoch: [2][ 1210/ 1236] Overall Loss 0.661323 Objective Loss 0.661323 LR 0.001000 Time 0.017037 +2023-10-02 20:30:16,243 - Epoch: [2][ 1220/ 1236] Overall Loss 0.660656 Objective Loss 0.660656 LR 0.001000 Time 0.017033 +2023-10-02 20:30:16,455 - Epoch: [2][ 1230/ 1236] Overall Loss 0.660044 Objective Loss 0.660044 LR 0.001000 Time 0.017066 +2023-10-02 20:30:16,549 - Epoch: [2][ 1236/ 1236] Overall Loss 0.659847 Objective Loss 0.659847 Top1 73.319756 Top5 95.112016 LR 0.001000 Time 0.017060 +2023-10-02 20:30:16,685 - --- validate (epoch=2)----------- +2023-10-02 20:30:16,685 - 29943 samples (256 per mini-batch) +2023-10-02 20:30:17,105 - Epoch: [2][ 10/ 117] Loss 0.516382 Top1 71.328125 Top5 95.664062 +2023-10-02 20:30:17,208 - Epoch: [2][ 20/ 117] Loss 0.532524 Top1 70.312500 Top5 95.527344 +2023-10-02 20:30:17,313 - Epoch: [2][ 30/ 117] Loss 0.536903 Top1 70.143229 Top5 95.572917 +2023-10-02 20:30:17,422 - Epoch: [2][ 40/ 117] Loss 0.536868 Top1 70.107422 Top5 95.683594 +2023-10-02 20:30:17,534 - Epoch: [2][ 50/ 117] Loss 0.540914 Top1 69.789062 Top5 95.617188 +2023-10-02 20:30:17,644 - Epoch: [2][ 60/ 117] Loss 0.546444 Top1 69.667969 Top5 95.540365 +2023-10-02 20:30:17,756 - Epoch: [2][ 70/ 117] Loss 0.543505 Top1 69.492188 Top5 95.524554 +2023-10-02 20:30:17,867 - Epoch: [2][ 80/ 117] Loss 0.550265 Top1 69.355469 Top5 95.502930 +2023-10-02 20:30:17,980 - Epoch: [2][ 90/ 117] Loss 0.551394 Top1 69.236111 Top5 95.477431 +2023-10-02 20:30:18,091 - Epoch: [2][ 100/ 117] Loss 0.553273 Top1 69.234375 Top5 95.492188 +2023-10-02 20:30:18,214 - Epoch: [2][ 110/ 117] Loss 0.555733 Top1 69.080256 Top5 95.511364 +2023-10-02 20:30:18,273 - Epoch: [2][ 117/ 117] Loss 0.556937 Top1 69.007781 Top5 95.501453 +2023-10-02 20:30:18,389 - ==> Top1: 69.008 Top5: 95.501 Loss: 0.557 + +2023-10-02 20:30:18,389 - ==> Confusion: +[[ 852 1 1 1 13 3 0 2 8 117 1 6 2 10 8 4 7 6 1 0 7] + [ 1 1002 3 1 7 27 3 29 5 0 8 3 0 0 6 2 5 1 24 0 4] + [ 16 0 829 12 11 1 66 29 1 5 6 9 5 17 7 8 1 1 14 3 15] + [ 3 3 12 857 0 5 2 4 8 0 44 0 9 7 72 7 3 13 31 0 9] + [ 23 19 4 0 915 15 1 3 2 17 0 1 0 5 10 6 21 0 0 0 8] + [ 2 139 9 3 4 808 3 31 0 3 3 25 4 36 10 2 7 1 4 12 10] + [ 1 12 51 0 4 6 1043 39 0 0 2 4 1 0 0 8 0 0 1 14 5] + [ 3 26 27 1 10 35 5 977 1 1 4 13 0 1 7 2 0 0 81 15 9] + [ 23 3 0 0 0 1 1 3 894 66 15 3 6 30 33 2 4 1 2 0 2] + [ 185 2 1 0 19 2 1 4 28 792 0 2 0 54 8 0 1 6 0 0 14] + [ 5 11 19 14 3 15 2 7 25 4 866 0 0 19 14 0 6 0 28 6 9] + [ 0 3 0 0 0 20 1 2 0 0 0 902 37 10 2 6 4 24 0 21 3] + [ 3 0 1 2 0 9 1 1 0 0 1 159 811 1 7 11 8 47 2 3 1] + [ 1 1 4 0 9 36 0 3 6 11 8 27 5 982 5 4 5 3 0 5 4] + [ 13 2 0 8 19 2 0 0 31 6 0 0 10 4 982 2 3 4 2 0 13] + [ 0 1 7 2 3 5 3 0 1 0 0 43 9 0 0 998 22 24 0 2 14] + [ 0 30 3 0 12 13 1 2 0 1 0 13 2 2 4 11 1053 0 0 0 14] + [ 0 1 1 3 0 0 1 1 0 0 0 21 47 0 4 12 0 945 0 0 2] + [ 1 12 5 10 1 6 0 35 12 0 8 2 6 0 28 0 3 0 926 3 10] + [ 0 13 1 0 1 5 11 38 0 0 0 27 7 9 2 3 4 1 3 1024 3] + [ 262 573 125 76 207 374 47 174 116 224 196 343 491 518 449 166 240 108 428 583 2205]] + +2023-10-02 20:30:18,391 - ==> Best [Top1: 69.008 Top5: 95.501 Sparsity:0.00 Params: 169472 on epoch: 2] +2023-10-02 20:30:18,391 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:30:18,404 - + +2023-10-02 20:30:18,404 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:30:19,449 - Epoch: [3][ 10/ 1236] Overall Loss 0.557656 Objective Loss 0.557656 LR 0.001000 Time 0.104440 +2023-10-02 20:30:19,613 - Epoch: [3][ 20/ 1236] Overall Loss 0.568612 Objective Loss 0.568612 LR 0.001000 Time 0.060387 +2023-10-02 20:30:19,776 - Epoch: [3][ 30/ 1236] Overall Loss 0.577605 Objective Loss 0.577605 LR 0.001000 Time 0.045672 +2023-10-02 20:30:19,940 - Epoch: [3][ 40/ 1236] Overall Loss 0.580304 Objective Loss 0.580304 LR 0.001000 Time 0.038353 +2023-10-02 20:30:20,103 - Epoch: [3][ 50/ 1236] Overall Loss 0.579573 Objective Loss 0.579573 LR 0.001000 Time 0.033930 +2023-10-02 20:30:20,267 - Epoch: [3][ 60/ 1236] Overall Loss 0.578036 Objective Loss 0.578036 LR 0.001000 Time 0.031010 +2023-10-02 20:30:20,430 - Epoch: [3][ 70/ 1236] Overall Loss 0.577420 Objective Loss 0.577420 LR 0.001000 Time 0.028911 +2023-10-02 20:30:20,594 - Epoch: [3][ 80/ 1236] Overall Loss 0.577634 Objective Loss 0.577634 LR 0.001000 Time 0.027342 +2023-10-02 20:30:20,756 - Epoch: [3][ 90/ 1236] Overall Loss 0.577411 Objective Loss 0.577411 LR 0.001000 Time 0.026102 +2023-10-02 20:30:20,920 - Epoch: [3][ 100/ 1236] Overall Loss 0.581600 Objective Loss 0.581600 LR 0.001000 Time 0.025128 +2023-10-02 20:30:21,083 - Epoch: [3][ 110/ 1236] Overall Loss 0.582976 Objective Loss 0.582976 LR 0.001000 Time 0.024320 +2023-10-02 20:30:21,247 - Epoch: [3][ 120/ 1236] Overall Loss 0.585291 Objective Loss 0.585291 LR 0.001000 Time 0.023658 +2023-10-02 20:30:21,409 - Epoch: [3][ 130/ 1236] Overall Loss 0.584878 Objective Loss 0.584878 LR 0.001000 Time 0.023083 +2023-10-02 20:30:21,574 - Epoch: [3][ 140/ 1236] Overall Loss 0.583106 Objective Loss 0.583106 LR 0.001000 Time 0.022606 +2023-10-02 20:30:21,736 - Epoch: [3][ 150/ 1236] Overall Loss 0.581220 Objective Loss 0.581220 LR 0.001000 Time 0.022180 +2023-10-02 20:30:21,900 - Epoch: [3][ 160/ 1236] Overall Loss 0.582098 Objective Loss 0.582098 LR 0.001000 Time 0.021815 +2023-10-02 20:30:22,062 - Epoch: [3][ 170/ 1236] Overall Loss 0.582996 Objective Loss 0.582996 LR 0.001000 Time 0.021484 +2023-10-02 20:30:22,225 - Epoch: [3][ 180/ 1236] Overall Loss 0.581050 Objective Loss 0.581050 LR 0.001000 Time 0.021196 +2023-10-02 20:30:22,387 - Epoch: [3][ 190/ 1236] Overall Loss 0.579492 Objective Loss 0.579492 LR 0.001000 Time 0.020929 +2023-10-02 20:30:22,551 - Epoch: [3][ 200/ 1236] Overall Loss 0.582104 Objective Loss 0.582104 LR 0.001000 Time 0.020700 +2023-10-02 20:30:22,713 - Epoch: [3][ 210/ 1236] Overall Loss 0.582859 Objective Loss 0.582859 LR 0.001000 Time 0.020486 +2023-10-02 20:30:22,877 - Epoch: [3][ 220/ 1236] Overall Loss 0.583672 Objective Loss 0.583672 LR 0.001000 Time 0.020300 +2023-10-02 20:30:23,039 - Epoch: [3][ 230/ 1236] Overall Loss 0.585421 Objective Loss 0.585421 LR 0.001000 Time 0.020120 +2023-10-02 20:30:23,203 - Epoch: [3][ 240/ 1236] Overall Loss 0.584570 Objective Loss 0.584570 LR 0.001000 Time 0.019964 +2023-10-02 20:30:23,366 - Epoch: [3][ 250/ 1236] Overall Loss 0.584740 Objective Loss 0.584740 LR 0.001000 Time 0.019813 +2023-10-02 20:30:23,529 - Epoch: [3][ 260/ 1236] Overall Loss 0.583934 Objective Loss 0.583934 LR 0.001000 Time 0.019680 +2023-10-02 20:30:23,692 - Epoch: [3][ 270/ 1236] Overall Loss 0.582895 Objective Loss 0.582895 LR 0.001000 Time 0.019552 +2023-10-02 20:30:23,856 - Epoch: [3][ 280/ 1236] Overall Loss 0.583757 Objective Loss 0.583757 LR 0.001000 Time 0.019440 +2023-10-02 20:30:24,021 - Epoch: [3][ 290/ 1236] Overall Loss 0.582533 Objective Loss 0.582533 LR 0.001000 Time 0.019335 +2023-10-02 20:30:24,185 - Epoch: [3][ 300/ 1236] Overall Loss 0.582667 Objective Loss 0.582667 LR 0.001000 Time 0.019239 +2023-10-02 20:30:24,350 - Epoch: [3][ 310/ 1236] Overall Loss 0.581325 Objective Loss 0.581325 LR 0.001000 Time 0.019147 +2023-10-02 20:30:24,515 - Epoch: [3][ 320/ 1236] Overall Loss 0.580916 Objective Loss 0.580916 LR 0.001000 Time 0.019063 +2023-10-02 20:30:24,679 - Epoch: [3][ 330/ 1236] Overall Loss 0.579948 Objective Loss 0.579948 LR 0.001000 Time 0.018981 +2023-10-02 20:30:24,844 - Epoch: [3][ 340/ 1236] Overall Loss 0.579453 Objective Loss 0.579453 LR 0.001000 Time 0.018908 +2023-10-02 20:30:25,008 - Epoch: [3][ 350/ 1236] Overall Loss 0.579416 Objective Loss 0.579416 LR 0.001000 Time 0.018836 +2023-10-02 20:30:25,173 - Epoch: [3][ 360/ 1236] Overall Loss 0.580179 Objective Loss 0.580179 LR 0.001000 Time 0.018771 +2023-10-02 20:30:25,338 - Epoch: [3][ 370/ 1236] Overall Loss 0.581370 Objective Loss 0.581370 LR 0.001000 Time 0.018708 +2023-10-02 20:30:25,504 - Epoch: [3][ 380/ 1236] Overall Loss 0.581001 Objective Loss 0.581001 LR 0.001000 Time 0.018651 +2023-10-02 20:30:25,669 - Epoch: [3][ 390/ 1236] Overall Loss 0.581023 Objective Loss 0.581023 LR 0.001000 Time 0.018595 +2023-10-02 20:30:25,834 - Epoch: [3][ 400/ 1236] Overall Loss 0.581332 Objective Loss 0.581332 LR 0.001000 Time 0.018543 +2023-10-02 20:30:25,999 - Epoch: [3][ 410/ 1236] Overall Loss 0.580146 Objective Loss 0.580146 LR 0.001000 Time 0.018491 +2023-10-02 20:30:26,164 - Epoch: [3][ 420/ 1236] Overall Loss 0.579618 Objective Loss 0.579618 LR 0.001000 Time 0.018445 +2023-10-02 20:30:26,329 - Epoch: [3][ 430/ 1236] Overall Loss 0.580847 Objective Loss 0.580847 LR 0.001000 Time 0.018398 +2023-10-02 20:30:26,494 - Epoch: [3][ 440/ 1236] Overall Loss 0.580751 Objective Loss 0.580751 LR 0.001000 Time 0.018355 +2023-10-02 20:30:26,659 - Epoch: [3][ 450/ 1236] Overall Loss 0.579789 Objective Loss 0.579789 LR 0.001000 Time 0.018312 +2023-10-02 20:30:26,824 - Epoch: [3][ 460/ 1236] Overall Loss 0.578769 Objective Loss 0.578769 LR 0.001000 Time 0.018272 +2023-10-02 20:30:26,988 - Epoch: [3][ 470/ 1236] Overall Loss 0.579890 Objective Loss 0.579890 LR 0.001000 Time 0.018231 +2023-10-02 20:30:27,153 - Epoch: [3][ 480/ 1236] Overall Loss 0.579275 Objective Loss 0.579275 LR 0.001000 Time 0.018195 +2023-10-02 20:30:27,318 - Epoch: [3][ 490/ 1236] Overall Loss 0.579700 Objective Loss 0.579700 LR 0.001000 Time 0.018161 +2023-10-02 20:30:27,484 - Epoch: [3][ 500/ 1236] Overall Loss 0.580243 Objective Loss 0.580243 LR 0.001000 Time 0.018128 +2023-10-02 20:30:27,648 - Epoch: [3][ 510/ 1236] Overall Loss 0.580733 Objective Loss 0.580733 LR 0.001000 Time 0.018093 +2023-10-02 20:30:27,813 - Epoch: [3][ 520/ 1236] Overall Loss 0.580461 Objective Loss 0.580461 LR 0.001000 Time 0.018063 +2023-10-02 20:30:27,978 - Epoch: [3][ 530/ 1236] Overall Loss 0.578854 Objective Loss 0.578854 LR 0.001000 Time 0.018032 +2023-10-02 20:30:28,143 - Epoch: [3][ 540/ 1236] Overall Loss 0.578506 Objective Loss 0.578506 LR 0.001000 Time 0.018003 +2023-10-02 20:30:28,307 - Epoch: [3][ 550/ 1236] Overall Loss 0.578814 Objective Loss 0.578814 LR 0.001000 Time 0.017975 +2023-10-02 20:30:28,473 - Epoch: [3][ 560/ 1236] Overall Loss 0.579209 Objective Loss 0.579209 LR 0.001000 Time 0.017948 +2023-10-02 20:30:28,637 - Epoch: [3][ 570/ 1236] Overall Loss 0.579372 Objective Loss 0.579372 LR 0.001000 Time 0.017921 +2023-10-02 20:30:28,814 - Epoch: [3][ 580/ 1236] Overall Loss 0.579728 Objective Loss 0.579728 LR 0.001000 Time 0.017917 +2023-10-02 20:30:28,986 - Epoch: [3][ 590/ 1236] Overall Loss 0.579240 Objective Loss 0.579240 LR 0.001000 Time 0.017903 +2023-10-02 20:30:29,162 - Epoch: [3][ 600/ 1236] Overall Loss 0.579701 Objective Loss 0.579701 LR 0.001000 Time 0.017898 +2023-10-02 20:30:29,333 - Epoch: [3][ 610/ 1236] Overall Loss 0.579178 Objective Loss 0.579178 LR 0.001000 Time 0.017884 +2023-10-02 20:30:29,508 - Epoch: [3][ 620/ 1236] Overall Loss 0.578480 Objective Loss 0.578480 LR 0.001000 Time 0.017878 +2023-10-02 20:30:29,672 - Epoch: [3][ 630/ 1236] Overall Loss 0.578482 Objective Loss 0.578482 LR 0.001000 Time 0.017853 +2023-10-02 20:30:29,837 - Epoch: [3][ 640/ 1236] Overall Loss 0.578063 Objective Loss 0.578063 LR 0.001000 Time 0.017832 +2023-10-02 20:30:30,001 - Epoch: [3][ 650/ 1236] Overall Loss 0.578480 Objective Loss 0.578480 LR 0.001000 Time 0.017810 +2023-10-02 20:30:30,166 - Epoch: [3][ 660/ 1236] Overall Loss 0.578271 Objective Loss 0.578271 LR 0.001000 Time 0.017790 +2023-10-02 20:30:30,331 - Epoch: [3][ 670/ 1236] Overall Loss 0.578216 Objective Loss 0.578216 LR 0.001000 Time 0.017770 +2023-10-02 20:30:30,496 - Epoch: [3][ 680/ 1236] Overall Loss 0.577491 Objective Loss 0.577491 LR 0.001000 Time 0.017751 +2023-10-02 20:30:30,661 - Epoch: [3][ 690/ 1236] Overall Loss 0.577221 Objective Loss 0.577221 LR 0.001000 Time 0.017732 +2023-10-02 20:30:30,826 - Epoch: [3][ 700/ 1236] Overall Loss 0.576813 Objective Loss 0.576813 LR 0.001000 Time 0.017715 +2023-10-02 20:30:30,991 - Epoch: [3][ 710/ 1236] Overall Loss 0.577366 Objective Loss 0.577366 LR 0.001000 Time 0.017696 +2023-10-02 20:30:31,157 - Epoch: [3][ 720/ 1236] Overall Loss 0.577003 Objective Loss 0.577003 LR 0.001000 Time 0.017680 +2023-10-02 20:30:31,321 - Epoch: [3][ 730/ 1236] Overall Loss 0.576046 Objective Loss 0.576046 LR 0.001000 Time 0.017663 +2023-10-02 20:30:31,486 - Epoch: [3][ 740/ 1236] Overall Loss 0.576131 Objective Loss 0.576131 LR 0.001000 Time 0.017647 +2023-10-02 20:30:31,651 - Epoch: [3][ 750/ 1236] Overall Loss 0.575844 Objective Loss 0.575844 LR 0.001000 Time 0.017631 +2023-10-02 20:30:31,819 - Epoch: [3][ 760/ 1236] Overall Loss 0.574971 Objective Loss 0.574971 LR 0.001000 Time 0.017620 +2023-10-02 20:30:31,983 - Epoch: [3][ 770/ 1236] Overall Loss 0.574505 Objective Loss 0.574505 LR 0.001000 Time 0.017604 +2023-10-02 20:30:32,159 - Epoch: [3][ 780/ 1236] Overall Loss 0.574642 Objective Loss 0.574642 LR 0.001000 Time 0.017603 +2023-10-02 20:30:32,330 - Epoch: [3][ 790/ 1236] Overall Loss 0.574370 Objective Loss 0.574370 LR 0.001000 Time 0.017596 +2023-10-02 20:30:32,506 - Epoch: [3][ 800/ 1236] Overall Loss 0.573910 Objective Loss 0.573910 LR 0.001000 Time 0.017595 +2023-10-02 20:30:32,676 - Epoch: [3][ 810/ 1236] Overall Loss 0.574124 Objective Loss 0.574124 LR 0.001000 Time 0.017588 +2023-10-02 20:30:32,842 - Epoch: [3][ 820/ 1236] Overall Loss 0.573459 Objective Loss 0.573459 LR 0.001000 Time 0.017575 +2023-10-02 20:30:33,006 - Epoch: [3][ 830/ 1236] Overall Loss 0.573314 Objective Loss 0.573314 LR 0.001000 Time 0.017561 +2023-10-02 20:30:33,172 - Epoch: [3][ 840/ 1236] Overall Loss 0.572895 Objective Loss 0.572895 LR 0.001000 Time 0.017549 +2023-10-02 20:30:33,336 - Epoch: [3][ 850/ 1236] Overall Loss 0.572833 Objective Loss 0.572833 LR 0.001000 Time 0.017536 +2023-10-02 20:30:33,503 - Epoch: [3][ 860/ 1236] Overall Loss 0.572572 Objective Loss 0.572572 LR 0.001000 Time 0.017525 +2023-10-02 20:30:33,667 - Epoch: [3][ 870/ 1236] Overall Loss 0.572885 Objective Loss 0.572885 LR 0.001000 Time 0.017512 +2023-10-02 20:30:33,833 - Epoch: [3][ 880/ 1236] Overall Loss 0.572542 Objective Loss 0.572542 LR 0.001000 Time 0.017502 +2023-10-02 20:30:33,997 - Epoch: [3][ 890/ 1236] Overall Loss 0.572649 Objective Loss 0.572649 LR 0.001000 Time 0.017489 +2023-10-02 20:30:34,163 - Epoch: [3][ 900/ 1236] Overall Loss 0.572118 Objective Loss 0.572118 LR 0.001000 Time 0.017479 +2023-10-02 20:30:34,328 - Epoch: [3][ 910/ 1236] Overall Loss 0.571873 Objective Loss 0.571873 LR 0.001000 Time 0.017467 +2023-10-02 20:30:34,493 - Epoch: [3][ 920/ 1236] Overall Loss 0.571749 Objective Loss 0.571749 LR 0.001000 Time 0.017457 +2023-10-02 20:30:34,657 - Epoch: [3][ 930/ 1236] Overall Loss 0.571178 Objective Loss 0.571178 LR 0.001000 Time 0.017445 +2023-10-02 20:30:34,834 - Epoch: [3][ 940/ 1236] Overall Loss 0.571004 Objective Loss 0.571004 LR 0.001000 Time 0.017447 +2023-10-02 20:30:35,005 - Epoch: [3][ 950/ 1236] Overall Loss 0.570504 Objective Loss 0.570504 LR 0.001000 Time 0.017444 +2023-10-02 20:30:35,182 - Epoch: [3][ 960/ 1236] Overall Loss 0.570348 Objective Loss 0.570348 LR 0.001000 Time 0.017446 +2023-10-02 20:30:35,353 - Epoch: [3][ 970/ 1236] Overall Loss 0.570323 Objective Loss 0.570323 LR 0.001000 Time 0.017441 +2023-10-02 20:30:35,529 - Epoch: [3][ 980/ 1236] Overall Loss 0.570463 Objective Loss 0.570463 LR 0.001000 Time 0.017443 +2023-10-02 20:30:35,700 - Epoch: [3][ 990/ 1236] Overall Loss 0.570006 Objective Loss 0.570006 LR 0.001000 Time 0.017439 +2023-10-02 20:30:35,876 - Epoch: [3][ 1000/ 1236] Overall Loss 0.569787 Objective Loss 0.569787 LR 0.001000 Time 0.017441 +2023-10-02 20:30:36,043 - Epoch: [3][ 1010/ 1236] Overall Loss 0.569820 Objective Loss 0.569820 LR 0.001000 Time 0.017433 +2023-10-02 20:30:36,219 - Epoch: [3][ 1020/ 1236] Overall Loss 0.569517 Objective Loss 0.569517 LR 0.001000 Time 0.017435 +2023-10-02 20:30:36,390 - Epoch: [3][ 1030/ 1236] Overall Loss 0.569536 Objective Loss 0.569536 LR 0.001000 Time 0.017431 +2023-10-02 20:30:36,566 - Epoch: [3][ 1040/ 1236] Overall Loss 0.568888 Objective Loss 0.568888 LR 0.001000 Time 0.017432 +2023-10-02 20:30:36,736 - Epoch: [3][ 1050/ 1236] Overall Loss 0.568921 Objective Loss 0.568921 LR 0.001000 Time 0.017428 +2023-10-02 20:30:36,906 - Epoch: [3][ 1060/ 1236] Overall Loss 0.568507 Objective Loss 0.568507 LR 0.001000 Time 0.017423 +2023-10-02 20:30:37,070 - Epoch: [3][ 1070/ 1236] Overall Loss 0.568305 Objective Loss 0.568305 LR 0.001000 Time 0.017414 +2023-10-02 20:30:37,246 - Epoch: [3][ 1080/ 1236] Overall Loss 0.568090 Objective Loss 0.568090 LR 0.001000 Time 0.017416 +2023-10-02 20:30:37,417 - Epoch: [3][ 1090/ 1236] Overall Loss 0.568333 Objective Loss 0.568333 LR 0.001000 Time 0.017412 +2023-10-02 20:30:37,593 - Epoch: [3][ 1100/ 1236] Overall Loss 0.567812 Objective Loss 0.567812 LR 0.001000 Time 0.017413 +2023-10-02 20:30:37,763 - Epoch: [3][ 1110/ 1236] Overall Loss 0.567390 Objective Loss 0.567390 LR 0.001000 Time 0.017410 +2023-10-02 20:30:37,940 - Epoch: [3][ 1120/ 1236] Overall Loss 0.567141 Objective Loss 0.567141 LR 0.001000 Time 0.017411 +2023-10-02 20:30:38,106 - Epoch: [3][ 1130/ 1236] Overall Loss 0.567183 Objective Loss 0.567183 LR 0.001000 Time 0.017404 +2023-10-02 20:30:38,272 - Epoch: [3][ 1140/ 1236] Overall Loss 0.566997 Objective Loss 0.566997 LR 0.001000 Time 0.017397 +2023-10-02 20:30:38,436 - Epoch: [3][ 1150/ 1236] Overall Loss 0.566347 Objective Loss 0.566347 LR 0.001000 Time 0.017388 +2023-10-02 20:30:38,602 - Epoch: [3][ 1160/ 1236] Overall Loss 0.565859 Objective Loss 0.565859 LR 0.001000 Time 0.017381 +2023-10-02 20:30:38,766 - Epoch: [3][ 1170/ 1236] Overall Loss 0.565342 Objective Loss 0.565342 LR 0.001000 Time 0.017373 +2023-10-02 20:30:38,932 - Epoch: [3][ 1180/ 1236] Overall Loss 0.565048 Objective Loss 0.565048 LR 0.001000 Time 0.017366 +2023-10-02 20:30:39,095 - Epoch: [3][ 1190/ 1236] Overall Loss 0.564751 Objective Loss 0.564751 LR 0.001000 Time 0.017357 +2023-10-02 20:30:39,261 - Epoch: [3][ 1200/ 1236] Overall Loss 0.564677 Objective Loss 0.564677 LR 0.001000 Time 0.017350 +2023-10-02 20:30:39,426 - Epoch: [3][ 1210/ 1236] Overall Loss 0.564337 Objective Loss 0.564337 LR 0.001000 Time 0.017342 +2023-10-02 20:30:39,591 - Epoch: [3][ 1220/ 1236] Overall Loss 0.564246 Objective Loss 0.564246 LR 0.001000 Time 0.017335 +2023-10-02 20:30:39,804 - Epoch: [3][ 1230/ 1236] Overall Loss 0.563804 Objective Loss 0.563804 LR 0.001000 Time 0.017367 +2023-10-02 20:30:39,900 - Epoch: [3][ 1236/ 1236] Overall Loss 0.563381 Objective Loss 0.563381 Top1 76.171079 Top5 96.537678 LR 0.001000 Time 0.017360 +2023-10-02 20:30:40,024 - --- validate (epoch=3)----------- +2023-10-02 20:30:40,024 - 29943 samples (256 per mini-batch) +2023-10-02 20:30:40,467 - Epoch: [3][ 10/ 117] Loss 0.520867 Top1 71.718750 Top5 96.484375 +2023-10-02 20:30:40,571 - Epoch: [3][ 20/ 117] Loss 0.497321 Top1 71.718750 Top5 96.562500 +2023-10-02 20:30:40,678 - Epoch: [3][ 30/ 117] Loss 0.489066 Top1 72.552083 Top5 96.718750 +2023-10-02 20:30:40,782 - Epoch: [3][ 40/ 117] Loss 0.485015 Top1 72.451172 Top5 96.787109 +2023-10-02 20:30:40,886 - Epoch: [3][ 50/ 117] Loss 0.480986 Top1 72.859375 Top5 96.851562 +2023-10-02 20:30:40,989 - Epoch: [3][ 60/ 117] Loss 0.481588 Top1 73.092448 Top5 96.842448 +2023-10-02 20:30:41,094 - Epoch: [3][ 70/ 117] Loss 0.481567 Top1 73.180804 Top5 96.852679 +2023-10-02 20:30:41,196 - Epoch: [3][ 80/ 117] Loss 0.480558 Top1 73.325195 Top5 96.767578 +2023-10-02 20:30:41,299 - Epoch: [3][ 90/ 117] Loss 0.479242 Top1 73.372396 Top5 96.788194 +2023-10-02 20:30:41,402 - Epoch: [3][ 100/ 117] Loss 0.477858 Top1 73.277344 Top5 96.785156 +2023-10-02 20:30:41,518 - Epoch: [3][ 110/ 117] Loss 0.478229 Top1 73.199574 Top5 96.835938 +2023-10-02 20:30:41,577 - Epoch: [3][ 117/ 117] Loss 0.480634 Top1 73.098888 Top5 96.797248 +2023-10-02 20:30:41,696 - ==> Top1: 73.099 Top5: 96.797 Loss: 0.481 + +2023-10-02 20:30:41,697 - ==> Confusion: +[[ 883 1 3 2 13 3 0 1 14 83 1 6 0 2 7 3 6 1 2 0 19] + [ 1 986 2 2 9 43 3 24 5 0 5 3 0 0 7 3 11 0 24 2 1] + [ 10 1 912 9 8 1 38 16 0 0 4 7 2 10 2 5 2 0 12 3 14] + [ 5 3 25 883 0 4 1 1 3 0 32 0 5 7 58 4 3 7 31 1 16] + [ 22 13 2 0 944 11 0 1 0 8 1 1 0 5 12 5 15 0 1 3 6] + [ 2 52 8 1 2 913 2 40 2 2 1 16 3 29 7 2 6 0 2 16 10] + [ 1 7 40 1 2 0 1075 16 0 0 6 3 0 0 0 7 2 0 2 17 12] + [ 1 19 28 0 8 39 4 1016 0 1 7 8 0 0 1 2 2 0 62 15 5] + [ 18 8 0 0 1 1 0 1 944 32 17 2 0 20 29 2 4 1 2 3 4] + [ 151 4 1 0 32 3 1 1 62 787 0 0 0 41 8 1 1 2 1 4 19] + [ 7 6 7 16 4 10 4 6 19 3 908 0 0 13 6 0 6 0 21 9 8] + [ 0 2 2 0 0 18 2 0 0 1 0 924 15 9 0 6 4 13 0 36 3] + [ 2 0 5 3 2 3 1 2 1 0 0 156 832 1 6 15 5 20 5 9 0] + [ 3 0 0 1 8 27 1 4 10 12 9 8 1 1008 5 3 4 0 0 7 8] + [ 7 3 1 4 13 6 0 0 28 1 5 0 7 2 999 0 3 1 3 0 18] + [ 0 1 3 1 3 5 4 0 0 0 0 22 3 0 2 1031 20 7 0 12 20] + [ 0 18 4 0 13 12 0 1 1 0 0 10 0 1 4 8 1069 0 0 8 12] + [ 1 0 1 3 0 0 5 0 2 0 0 28 56 1 5 19 2 897 1 2 15] + [ 1 10 11 15 0 4 0 32 4 0 7 0 4 0 18 0 6 0 942 6 8] + [ 0 7 4 0 2 5 14 29 0 0 2 19 2 4 0 1 3 0 2 1056 2] + [ 215 373 180 60 167 405 45 195 134 109 159 302 377 404 272 141 385 38 391 674 2879]] + +2023-10-02 20:30:41,698 - ==> Best [Top1: 73.099 Top5: 96.797 Sparsity:0.00 Params: 169472 on epoch: 3] +2023-10-02 20:30:41,698 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:30:41,712 - + +2023-10-02 20:30:41,712 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:30:42,684 - Epoch: [4][ 10/ 1236] Overall Loss 0.533025 Objective Loss 0.533025 LR 0.001000 Time 0.097174 +2023-10-02 20:30:42,848 - Epoch: [4][ 20/ 1236] Overall Loss 0.521617 Objective Loss 0.521617 LR 0.001000 Time 0.056747 +2023-10-02 20:30:43,012 - Epoch: [4][ 30/ 1236] Overall Loss 0.518795 Objective Loss 0.518795 LR 0.001000 Time 0.043272 +2023-10-02 20:30:43,176 - Epoch: [4][ 40/ 1236] Overall Loss 0.514134 Objective Loss 0.514134 LR 0.001000 Time 0.036556 +2023-10-02 20:30:43,339 - Epoch: [4][ 50/ 1236] Overall Loss 0.507405 Objective Loss 0.507405 LR 0.001000 Time 0.032507 +2023-10-02 20:30:43,504 - Epoch: [4][ 60/ 1236] Overall Loss 0.504185 Objective Loss 0.504185 LR 0.001000 Time 0.029827 +2023-10-02 20:30:43,667 - Epoch: [4][ 70/ 1236] Overall Loss 0.505534 Objective Loss 0.505534 LR 0.001000 Time 0.027894 +2023-10-02 20:30:43,831 - Epoch: [4][ 80/ 1236] Overall Loss 0.508115 Objective Loss 0.508115 LR 0.001000 Time 0.026449 +2023-10-02 20:30:43,994 - Epoch: [4][ 90/ 1236] Overall Loss 0.509941 Objective Loss 0.509941 LR 0.001000 Time 0.025319 +2023-10-02 20:30:44,158 - Epoch: [4][ 100/ 1236] Overall Loss 0.506431 Objective Loss 0.506431 LR 0.001000 Time 0.024422 +2023-10-02 20:30:44,321 - Epoch: [4][ 110/ 1236] Overall Loss 0.504059 Objective Loss 0.504059 LR 0.001000 Time 0.023686 +2023-10-02 20:30:44,485 - Epoch: [4][ 120/ 1236] Overall Loss 0.504287 Objective Loss 0.504287 LR 0.001000 Time 0.023076 +2023-10-02 20:30:44,649 - Epoch: [4][ 130/ 1236] Overall Loss 0.505156 Objective Loss 0.505156 LR 0.001000 Time 0.022557 +2023-10-02 20:30:44,813 - Epoch: [4][ 140/ 1236] Overall Loss 0.506606 Objective Loss 0.506606 LR 0.001000 Time 0.022114 +2023-10-02 20:30:44,976 - Epoch: [4][ 150/ 1236] Overall Loss 0.505742 Objective Loss 0.505742 LR 0.001000 Time 0.021725 +2023-10-02 20:30:45,140 - Epoch: [4][ 160/ 1236] Overall Loss 0.504448 Objective Loss 0.504448 LR 0.001000 Time 0.021388 +2023-10-02 20:30:45,303 - Epoch: [4][ 170/ 1236] Overall Loss 0.502875 Objective Loss 0.502875 LR 0.001000 Time 0.021088 +2023-10-02 20:30:45,467 - Epoch: [4][ 180/ 1236] Overall Loss 0.505855 Objective Loss 0.505855 LR 0.001000 Time 0.020826 +2023-10-02 20:30:45,630 - Epoch: [4][ 190/ 1236] Overall Loss 0.506052 Objective Loss 0.506052 LR 0.001000 Time 0.020590 +2023-10-02 20:30:45,795 - Epoch: [4][ 200/ 1236] Overall Loss 0.507845 Objective Loss 0.507845 LR 0.001000 Time 0.020380 +2023-10-02 20:30:45,958 - Epoch: [4][ 210/ 1236] Overall Loss 0.508831 Objective Loss 0.508831 LR 0.001000 Time 0.020186 +2023-10-02 20:30:46,122 - Epoch: [4][ 220/ 1236] Overall Loss 0.508647 Objective Loss 0.508647 LR 0.001000 Time 0.020013 +2023-10-02 20:30:46,286 - Epoch: [4][ 230/ 1236] Overall Loss 0.509145 Objective Loss 0.509145 LR 0.001000 Time 0.019852 +2023-10-02 20:30:46,450 - Epoch: [4][ 240/ 1236] Overall Loss 0.509879 Objective Loss 0.509879 LR 0.001000 Time 0.019707 +2023-10-02 20:30:46,613 - Epoch: [4][ 250/ 1236] Overall Loss 0.508810 Objective Loss 0.508810 LR 0.001000 Time 0.019573 +2023-10-02 20:30:46,778 - Epoch: [4][ 260/ 1236] Overall Loss 0.507584 Objective Loss 0.507584 LR 0.001000 Time 0.019451 +2023-10-02 20:30:46,945 - Epoch: [4][ 270/ 1236] Overall Loss 0.506355 Objective Loss 0.506355 LR 0.001000 Time 0.019335 +2023-10-02 20:30:47,109 - Epoch: [4][ 280/ 1236] Overall Loss 0.506022 Objective Loss 0.506022 LR 0.001000 Time 0.019229 +2023-10-02 20:30:47,273 - Epoch: [4][ 290/ 1236] Overall Loss 0.505602 Objective Loss 0.505602 LR 0.001000 Time 0.019129 +2023-10-02 20:30:47,438 - Epoch: [4][ 300/ 1236] Overall Loss 0.504391 Objective Loss 0.504391 LR 0.001000 Time 0.019042 +2023-10-02 20:30:47,604 - Epoch: [4][ 310/ 1236] Overall Loss 0.503439 Objective Loss 0.503439 LR 0.001000 Time 0.018961 +2023-10-02 20:30:47,769 - Epoch: [4][ 320/ 1236] Overall Loss 0.503593 Objective Loss 0.503593 LR 0.001000 Time 0.018884 +2023-10-02 20:30:47,934 - Epoch: [4][ 330/ 1236] Overall Loss 0.504068 Objective Loss 0.504068 LR 0.001000 Time 0.018811 +2023-10-02 20:30:48,101 - Epoch: [4][ 340/ 1236] Overall Loss 0.503283 Objective Loss 0.503283 LR 0.001000 Time 0.018747 +2023-10-02 20:30:48,266 - Epoch: [4][ 350/ 1236] Overall Loss 0.503447 Objective Loss 0.503447 LR 0.001000 Time 0.018683 +2023-10-02 20:30:48,434 - Epoch: [4][ 360/ 1236] Overall Loss 0.503981 Objective Loss 0.503981 LR 0.001000 Time 0.018628 +2023-10-02 20:30:48,600 - Epoch: [4][ 370/ 1236] Overall Loss 0.503707 Objective Loss 0.503707 LR 0.001000 Time 0.018573 +2023-10-02 20:30:48,767 - Epoch: [4][ 380/ 1236] Overall Loss 0.502822 Objective Loss 0.502822 LR 0.001000 Time 0.018524 +2023-10-02 20:30:48,933 - Epoch: [4][ 390/ 1236] Overall Loss 0.501256 Objective Loss 0.501256 LR 0.001000 Time 0.018474 +2023-10-02 20:30:49,101 - Epoch: [4][ 400/ 1236] Overall Loss 0.501576 Objective Loss 0.501576 LR 0.001000 Time 0.018429 +2023-10-02 20:30:49,266 - Epoch: [4][ 410/ 1236] Overall Loss 0.501278 Objective Loss 0.501278 LR 0.001000 Time 0.018383 +2023-10-02 20:30:49,434 - Epoch: [4][ 420/ 1236] Overall Loss 0.502353 Objective Loss 0.502353 LR 0.001000 Time 0.018343 +2023-10-02 20:30:49,599 - Epoch: [4][ 430/ 1236] Overall Loss 0.503557 Objective Loss 0.503557 LR 0.001000 Time 0.018301 +2023-10-02 20:30:49,766 - Epoch: [4][ 440/ 1236] Overall Loss 0.503633 Objective Loss 0.503633 LR 0.001000 Time 0.018264 +2023-10-02 20:30:49,931 - Epoch: [4][ 450/ 1236] Overall Loss 0.503598 Objective Loss 0.503598 LR 0.001000 Time 0.018224 +2023-10-02 20:30:50,095 - Epoch: [4][ 460/ 1236] Overall Loss 0.504705 Objective Loss 0.504705 LR 0.001000 Time 0.018183 +2023-10-02 20:30:50,259 - Epoch: [4][ 470/ 1236] Overall Loss 0.504660 Objective Loss 0.504660 LR 0.001000 Time 0.018144 +2023-10-02 20:30:50,423 - Epoch: [4][ 480/ 1236] Overall Loss 0.505411 Objective Loss 0.505411 LR 0.001000 Time 0.018107 +2023-10-02 20:30:50,587 - Epoch: [4][ 490/ 1236] Overall Loss 0.506130 Objective Loss 0.506130 LR 0.001000 Time 0.018071 +2023-10-02 20:30:50,751 - Epoch: [4][ 500/ 1236] Overall Loss 0.505228 Objective Loss 0.505228 LR 0.001000 Time 0.018038 +2023-10-02 20:30:50,915 - Epoch: [4][ 510/ 1236] Overall Loss 0.504821 Objective Loss 0.504821 LR 0.001000 Time 0.018004 +2023-10-02 20:30:51,079 - Epoch: [4][ 520/ 1236] Overall Loss 0.504592 Objective Loss 0.504592 LR 0.001000 Time 0.017974 +2023-10-02 20:30:51,243 - Epoch: [4][ 530/ 1236] Overall Loss 0.504712 Objective Loss 0.504712 LR 0.001000 Time 0.017943 +2023-10-02 20:30:51,409 - Epoch: [4][ 540/ 1236] Overall Loss 0.504844 Objective Loss 0.504844 LR 0.001000 Time 0.017918 +2023-10-02 20:30:51,573 - Epoch: [4][ 550/ 1236] Overall Loss 0.504454 Objective Loss 0.504454 LR 0.001000 Time 0.017889 +2023-10-02 20:30:51,738 - Epoch: [4][ 560/ 1236] Overall Loss 0.504003 Objective Loss 0.504003 LR 0.001000 Time 0.017864 +2023-10-02 20:30:51,902 - Epoch: [4][ 570/ 1236] Overall Loss 0.504057 Objective Loss 0.504057 LR 0.001000 Time 0.017838 +2023-10-02 20:30:52,066 - Epoch: [4][ 580/ 1236] Overall Loss 0.504084 Objective Loss 0.504084 LR 0.001000 Time 0.017813 +2023-10-02 20:30:52,230 - Epoch: [4][ 590/ 1236] Overall Loss 0.503936 Objective Loss 0.503936 LR 0.001000 Time 0.017788 +2023-10-02 20:30:52,395 - Epoch: [4][ 600/ 1236] Overall Loss 0.504309 Objective Loss 0.504309 LR 0.001000 Time 0.017766 +2023-10-02 20:30:52,559 - Epoch: [4][ 610/ 1236] Overall Loss 0.504408 Objective Loss 0.504408 LR 0.001000 Time 0.017743 +2023-10-02 20:30:52,724 - Epoch: [4][ 620/ 1236] Overall Loss 0.504088 Objective Loss 0.504088 LR 0.001000 Time 0.017722 +2023-10-02 20:30:52,888 - Epoch: [4][ 630/ 1236] Overall Loss 0.503292 Objective Loss 0.503292 LR 0.001000 Time 0.017701 +2023-10-02 20:30:53,053 - Epoch: [4][ 640/ 1236] Overall Loss 0.502597 Objective Loss 0.502597 LR 0.001000 Time 0.017682 +2023-10-02 20:30:53,217 - Epoch: [4][ 650/ 1236] Overall Loss 0.502001 Objective Loss 0.502001 LR 0.001000 Time 0.017662 +2023-10-02 20:30:53,381 - Epoch: [4][ 660/ 1236] Overall Loss 0.502017 Objective Loss 0.502017 LR 0.001000 Time 0.017643 +2023-10-02 20:30:53,545 - Epoch: [4][ 670/ 1236] Overall Loss 0.501629 Objective Loss 0.501629 LR 0.001000 Time 0.017624 +2023-10-02 20:30:53,710 - Epoch: [4][ 680/ 1236] Overall Loss 0.501356 Objective Loss 0.501356 LR 0.001000 Time 0.017607 +2023-10-02 20:30:53,874 - Epoch: [4][ 690/ 1236] Overall Loss 0.500753 Objective Loss 0.500753 LR 0.001000 Time 0.017589 +2023-10-02 20:30:54,039 - Epoch: [4][ 700/ 1236] Overall Loss 0.500889 Objective Loss 0.500889 LR 0.001000 Time 0.017573 +2023-10-02 20:30:54,203 - Epoch: [4][ 710/ 1236] Overall Loss 0.501109 Objective Loss 0.501109 LR 0.001000 Time 0.017556 +2023-10-02 20:30:54,368 - Epoch: [4][ 720/ 1236] Overall Loss 0.500793 Objective Loss 0.500793 LR 0.001000 Time 0.017541 +2023-10-02 20:30:54,538 - Epoch: [4][ 730/ 1236] Overall Loss 0.500557 Objective Loss 0.500557 LR 0.001000 Time 0.017533 +2023-10-02 20:30:54,707 - Epoch: [4][ 740/ 1236] Overall Loss 0.500942 Objective Loss 0.500942 LR 0.001000 Time 0.017523 +2023-10-02 20:30:54,876 - Epoch: [4][ 750/ 1236] Overall Loss 0.500853 Objective Loss 0.500853 LR 0.001000 Time 0.017515 +2023-10-02 20:30:55,044 - Epoch: [4][ 760/ 1236] Overall Loss 0.500552 Objective Loss 0.500552 LR 0.001000 Time 0.017505 +2023-10-02 20:30:55,214 - Epoch: [4][ 770/ 1236] Overall Loss 0.500909 Objective Loss 0.500909 LR 0.001000 Time 0.017498 +2023-10-02 20:30:55,382 - Epoch: [4][ 780/ 1236] Overall Loss 0.501065 Objective Loss 0.501065 LR 0.001000 Time 0.017489 +2023-10-02 20:30:55,552 - Epoch: [4][ 790/ 1236] Overall Loss 0.501046 Objective Loss 0.501046 LR 0.001000 Time 0.017482 +2023-10-02 20:30:55,721 - Epoch: [4][ 800/ 1236] Overall Loss 0.501119 Objective Loss 0.501119 LR 0.001000 Time 0.017474 +2023-10-02 20:30:55,890 - Epoch: [4][ 810/ 1236] Overall Loss 0.500462 Objective Loss 0.500462 LR 0.001000 Time 0.017467 +2023-10-02 20:30:56,059 - Epoch: [4][ 820/ 1236] Overall Loss 0.500330 Objective Loss 0.500330 LR 0.001000 Time 0.017460 +2023-10-02 20:30:56,229 - Epoch: [4][ 830/ 1236] Overall Loss 0.500205 Objective Loss 0.500205 LR 0.001000 Time 0.017453 +2023-10-02 20:30:56,397 - Epoch: [4][ 840/ 1236] Overall Loss 0.500149 Objective Loss 0.500149 LR 0.001000 Time 0.017446 +2023-10-02 20:30:56,566 - Epoch: [4][ 850/ 1236] Overall Loss 0.500000 Objective Loss 0.500000 LR 0.001000 Time 0.017439 +2023-10-02 20:30:56,735 - Epoch: [4][ 860/ 1236] Overall Loss 0.500018 Objective Loss 0.500018 LR 0.001000 Time 0.017432 +2023-10-02 20:30:56,905 - Epoch: [4][ 870/ 1236] Overall Loss 0.499496 Objective Loss 0.499496 LR 0.001000 Time 0.017427 +2023-10-02 20:30:57,074 - Epoch: [4][ 880/ 1236] Overall Loss 0.499624 Objective Loss 0.499624 LR 0.001000 Time 0.017420 +2023-10-02 20:30:57,243 - Epoch: [4][ 890/ 1236] Overall Loss 0.499493 Objective Loss 0.499493 LR 0.001000 Time 0.017414 +2023-10-02 20:30:57,411 - Epoch: [4][ 900/ 1236] Overall Loss 0.499296 Objective Loss 0.499296 LR 0.001000 Time 0.017407 +2023-10-02 20:30:57,581 - Epoch: [4][ 910/ 1236] Overall Loss 0.499004 Objective Loss 0.499004 LR 0.001000 Time 0.017402 +2023-10-02 20:30:57,749 - Epoch: [4][ 920/ 1236] Overall Loss 0.498708 Objective Loss 0.498708 LR 0.001000 Time 0.017396 +2023-10-02 20:30:57,919 - Epoch: [4][ 930/ 1236] Overall Loss 0.498467 Objective Loss 0.498467 LR 0.001000 Time 0.017391 +2023-10-02 20:30:58,088 - Epoch: [4][ 940/ 1236] Overall Loss 0.498533 Objective Loss 0.498533 LR 0.001000 Time 0.017385 +2023-10-02 20:30:58,257 - Epoch: [4][ 950/ 1236] Overall Loss 0.498126 Objective Loss 0.498126 LR 0.001000 Time 0.017380 +2023-10-02 20:30:58,426 - Epoch: [4][ 960/ 1236] Overall Loss 0.497859 Objective Loss 0.497859 LR 0.001000 Time 0.017374 +2023-10-02 20:30:58,595 - Epoch: [4][ 970/ 1236] Overall Loss 0.497733 Objective Loss 0.497733 LR 0.001000 Time 0.017370 +2023-10-02 20:30:58,764 - Epoch: [4][ 980/ 1236] Overall Loss 0.498006 Objective Loss 0.498006 LR 0.001000 Time 0.017364 +2023-10-02 20:30:58,933 - Epoch: [4][ 990/ 1236] Overall Loss 0.497856 Objective Loss 0.497856 LR 0.001000 Time 0.017360 +2023-10-02 20:30:59,102 - Epoch: [4][ 1000/ 1236] Overall Loss 0.497525 Objective Loss 0.497525 LR 0.001000 Time 0.017354 +2023-10-02 20:30:59,271 - Epoch: [4][ 1010/ 1236] Overall Loss 0.497262 Objective Loss 0.497262 LR 0.001000 Time 0.017350 +2023-10-02 20:30:59,441 - Epoch: [4][ 1020/ 1236] Overall Loss 0.496791 Objective Loss 0.496791 LR 0.001000 Time 0.017345 +2023-10-02 20:30:59,610 - Epoch: [4][ 1030/ 1236] Overall Loss 0.496663 Objective Loss 0.496663 LR 0.001000 Time 0.017341 +2023-10-02 20:30:59,779 - Epoch: [4][ 1040/ 1236] Overall Loss 0.497005 Objective Loss 0.497005 LR 0.001000 Time 0.017336 +2023-10-02 20:30:59,948 - Epoch: [4][ 1050/ 1236] Overall Loss 0.497126 Objective Loss 0.497126 LR 0.001000 Time 0.017332 +2023-10-02 20:31:00,116 - Epoch: [4][ 1060/ 1236] Overall Loss 0.497108 Objective Loss 0.497108 LR 0.001000 Time 0.017327 +2023-10-02 20:31:00,286 - Epoch: [4][ 1070/ 1236] Overall Loss 0.496984 Objective Loss 0.496984 LR 0.001000 Time 0.017324 +2023-10-02 20:31:00,455 - Epoch: [4][ 1080/ 1236] Overall Loss 0.496658 Objective Loss 0.496658 LR 0.001000 Time 0.017319 +2023-10-02 20:31:00,625 - Epoch: [4][ 1090/ 1236] Overall Loss 0.496681 Objective Loss 0.496681 LR 0.001000 Time 0.017316 +2023-10-02 20:31:00,793 - Epoch: [4][ 1100/ 1236] Overall Loss 0.496363 Objective Loss 0.496363 LR 0.001000 Time 0.017311 +2023-10-02 20:31:00,963 - Epoch: [4][ 1110/ 1236] Overall Loss 0.496240 Objective Loss 0.496240 LR 0.001000 Time 0.017308 +2023-10-02 20:31:01,132 - Epoch: [4][ 1120/ 1236] Overall Loss 0.496070 Objective Loss 0.496070 LR 0.001000 Time 0.017304 +2023-10-02 20:31:01,302 - Epoch: [4][ 1130/ 1236] Overall Loss 0.495728 Objective Loss 0.495728 LR 0.001000 Time 0.017301 +2023-10-02 20:31:01,470 - Epoch: [4][ 1140/ 1236] Overall Loss 0.495215 Objective Loss 0.495215 LR 0.001000 Time 0.017297 +2023-10-02 20:31:01,639 - Epoch: [4][ 1150/ 1236] Overall Loss 0.494885 Objective Loss 0.494885 LR 0.001000 Time 0.017293 +2023-10-02 20:31:01,808 - Epoch: [4][ 1160/ 1236] Overall Loss 0.494911 Objective Loss 0.494911 LR 0.001000 Time 0.017289 +2023-10-02 20:31:01,978 - Epoch: [4][ 1170/ 1236] Overall Loss 0.494903 Objective Loss 0.494903 LR 0.001000 Time 0.017286 +2023-10-02 20:31:02,146 - Epoch: [4][ 1180/ 1236] Overall Loss 0.494789 Objective Loss 0.494789 LR 0.001000 Time 0.017282 +2023-10-02 20:31:02,315 - Epoch: [4][ 1190/ 1236] Overall Loss 0.494960 Objective Loss 0.494960 LR 0.001000 Time 0.017279 +2023-10-02 20:31:02,484 - Epoch: [4][ 1200/ 1236] Overall Loss 0.495057 Objective Loss 0.495057 LR 0.001000 Time 0.017275 +2023-10-02 20:31:02,653 - Epoch: [4][ 1210/ 1236] Overall Loss 0.495257 Objective Loss 0.495257 LR 0.001000 Time 0.017272 +2023-10-02 20:31:02,822 - Epoch: [4][ 1220/ 1236] Overall Loss 0.495289 Objective Loss 0.495289 LR 0.001000 Time 0.017268 +2023-10-02 20:31:03,040 - Epoch: [4][ 1230/ 1236] Overall Loss 0.495217 Objective Loss 0.495217 LR 0.001000 Time 0.017305 +2023-10-02 20:31:03,137 - Epoch: [4][ 1236/ 1236] Overall Loss 0.495273 Objective Loss 0.495273 Top1 74.745418 Top5 95.315682 LR 0.001000 Time 0.017299 +2023-10-02 20:31:03,261 - --- validate (epoch=4)----------- +2023-10-02 20:31:03,261 - 29943 samples (256 per mini-batch) +2023-10-02 20:31:03,700 - Epoch: [4][ 10/ 117] Loss 0.461145 Top1 74.062500 Top5 96.875000 +2023-10-02 20:31:03,803 - Epoch: [4][ 20/ 117] Loss 0.460336 Top1 73.945312 Top5 96.972656 +2023-10-02 20:31:03,906 - Epoch: [4][ 30/ 117] Loss 0.468151 Top1 73.841146 Top5 96.875000 +2023-10-02 20:31:04,009 - Epoch: [4][ 40/ 117] Loss 0.460057 Top1 74.169922 Top5 96.855469 +2023-10-02 20:31:04,121 - Epoch: [4][ 50/ 117] Loss 0.449816 Top1 74.210938 Top5 96.937500 +2023-10-02 20:31:04,231 - Epoch: [4][ 60/ 117] Loss 0.448917 Top1 74.309896 Top5 96.920573 +2023-10-02 20:31:04,341 - Epoch: [4][ 70/ 117] Loss 0.448560 Top1 74.291295 Top5 96.897321 +2023-10-02 20:31:04,448 - Epoch: [4][ 80/ 117] Loss 0.447940 Top1 74.360352 Top5 96.860352 +2023-10-02 20:31:04,559 - Epoch: [4][ 90/ 117] Loss 0.447153 Top1 74.388021 Top5 96.870660 +2023-10-02 20:31:04,669 - Epoch: [4][ 100/ 117] Loss 0.451035 Top1 74.343750 Top5 96.886719 +2023-10-02 20:31:04,791 - Epoch: [4][ 110/ 117] Loss 0.448988 Top1 74.350142 Top5 96.899858 +2023-10-02 20:31:04,850 - Epoch: [4][ 117/ 117] Loss 0.448532 Top1 74.294493 Top5 96.890759 +2023-10-02 20:31:04,981 - ==> Top1: 74.294 Top5: 96.891 Loss: 0.449 + +2023-10-02 20:31:04,982 - ==> Confusion: +[[ 905 3 0 1 6 2 0 0 19 82 2 1 1 7 0 1 6 3 0 0 11] + [ 1 1047 0 2 4 16 2 24 5 0 6 4 0 1 0 3 3 0 5 4 4] + [ 22 3 866 5 7 1 62 23 1 3 12 1 4 8 2 5 2 4 8 2 15] + [ 3 4 17 879 0 3 5 2 7 0 53 0 4 5 50 2 3 14 23 1 14] + [ 26 19 1 0 912 15 0 0 3 16 0 0 0 8 5 8 27 0 0 4 6] + [ 2 119 0 1 1 840 2 36 6 2 11 14 4 32 10 3 5 1 0 21 6] + [ 2 6 24 1 1 0 1097 15 0 0 3 0 1 0 0 8 0 0 1 25 7] + [ 7 32 13 0 2 28 5 1041 5 0 9 9 1 0 1 1 2 0 42 15 5] + [ 24 4 0 0 0 1 0 2 965 43 19 2 0 10 13 1 4 1 0 0 0] + [ 112 3 0 0 3 0 2 0 46 891 0 1 0 43 3 1 1 0 0 4 9] + [ 8 9 4 5 1 3 3 6 23 1 959 0 0 10 2 2 4 0 6 2 5] + [ 1 5 0 0 0 18 2 0 2 2 1 846 83 14 0 6 4 25 1 17 8] + [ 1 6 0 3 0 5 0 2 7 0 1 47 928 0 4 8 9 40 1 4 2] + [ 2 1 0 0 2 11 1 1 19 15 12 7 1 1025 5 2 3 0 0 5 7] + [ 17 8 1 5 18 2 0 0 57 8 7 0 2 6 946 0 4 7 2 0 11] + [ 1 1 1 2 3 2 2 0 2 1 0 10 14 0 0 1035 20 25 0 7 8] + [ 1 22 2 0 3 9 1 0 2 0 3 7 3 1 2 14 1076 1 0 5 9] + [ 0 0 0 2 0 0 1 0 2 0 0 2 28 0 1 8 1 991 0 2 0] + [ 3 22 6 12 1 1 0 30 17 0 16 0 4 0 11 0 4 1 929 2 9] + [ 1 5 2 0 1 3 7 20 1 0 4 11 9 3 0 7 4 2 2 1065 5] + [ 283 509 112 69 110 310 50 172 231 199 284 157 537 441 190 138 308 104 221 477 3003]] + +2023-10-02 20:31:04,983 - ==> Best [Top1: 74.294 Top5: 96.891 Sparsity:0.00 Params: 169472 on epoch: 4] +2023-10-02 20:31:04,983 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:31:04,997 - + +2023-10-02 20:31:04,997 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:31:05,973 - Epoch: [5][ 10/ 1236] Overall Loss 0.456980 Objective Loss 0.456980 LR 0.001000 Time 0.097581 +2023-10-02 20:31:06,136 - Epoch: [5][ 20/ 1236] Overall Loss 0.459248 Objective Loss 0.459248 LR 0.001000 Time 0.056921 +2023-10-02 20:31:06,299 - Epoch: [5][ 30/ 1236] Overall Loss 0.456136 Objective Loss 0.456136 LR 0.001000 Time 0.043370 +2023-10-02 20:31:06,464 - Epoch: [5][ 40/ 1236] Overall Loss 0.458568 Objective Loss 0.458568 LR 0.001000 Time 0.036636 +2023-10-02 20:31:06,626 - Epoch: [5][ 50/ 1236] Overall Loss 0.453504 Objective Loss 0.453504 LR 0.001000 Time 0.032556 +2023-10-02 20:31:06,790 - Epoch: [5][ 60/ 1236] Overall Loss 0.455895 Objective Loss 0.455895 LR 0.001000 Time 0.029854 +2023-10-02 20:31:06,953 - Epoch: [5][ 70/ 1236] Overall Loss 0.452304 Objective Loss 0.452304 LR 0.001000 Time 0.027908 +2023-10-02 20:31:07,116 - Epoch: [5][ 80/ 1236] Overall Loss 0.449383 Objective Loss 0.449383 LR 0.001000 Time 0.026459 +2023-10-02 20:31:07,279 - Epoch: [5][ 90/ 1236] Overall Loss 0.448043 Objective Loss 0.448043 LR 0.001000 Time 0.025322 +2023-10-02 20:31:07,443 - Epoch: [5][ 100/ 1236] Overall Loss 0.452293 Objective Loss 0.452293 LR 0.001000 Time 0.024425 +2023-10-02 20:31:07,607 - Epoch: [5][ 110/ 1236] Overall Loss 0.453253 Objective Loss 0.453253 LR 0.001000 Time 0.023696 +2023-10-02 20:31:07,771 - Epoch: [5][ 120/ 1236] Overall Loss 0.452727 Objective Loss 0.452727 LR 0.001000 Time 0.023083 +2023-10-02 20:31:07,934 - Epoch: [5][ 130/ 1236] Overall Loss 0.454852 Objective Loss 0.454852 LR 0.001000 Time 0.022560 +2023-10-02 20:31:08,098 - Epoch: [5][ 140/ 1236] Overall Loss 0.455544 Objective Loss 0.455544 LR 0.001000 Time 0.022116 +2023-10-02 20:31:08,261 - Epoch: [5][ 150/ 1236] Overall Loss 0.456988 Objective Loss 0.456988 LR 0.001000 Time 0.021727 +2023-10-02 20:31:08,424 - Epoch: [5][ 160/ 1236] Overall Loss 0.453837 Objective Loss 0.453837 LR 0.001000 Time 0.021389 +2023-10-02 20:31:08,587 - Epoch: [5][ 170/ 1236] Overall Loss 0.454825 Objective Loss 0.454825 LR 0.001000 Time 0.021086 +2023-10-02 20:31:08,750 - Epoch: [5][ 180/ 1236] Overall Loss 0.455503 Objective Loss 0.455503 LR 0.001000 Time 0.020819 +2023-10-02 20:31:08,913 - Epoch: [5][ 190/ 1236] Overall Loss 0.454979 Objective Loss 0.454979 LR 0.001000 Time 0.020579 +2023-10-02 20:31:09,077 - Epoch: [5][ 200/ 1236] Overall Loss 0.452487 Objective Loss 0.452487 LR 0.001000 Time 0.020366 +2023-10-02 20:31:09,240 - Epoch: [5][ 210/ 1236] Overall Loss 0.452873 Objective Loss 0.452873 LR 0.001000 Time 0.020171 +2023-10-02 20:31:09,403 - Epoch: [5][ 220/ 1236] Overall Loss 0.452211 Objective Loss 0.452211 LR 0.001000 Time 0.019997 +2023-10-02 20:31:09,566 - Epoch: [5][ 230/ 1236] Overall Loss 0.451077 Objective Loss 0.451077 LR 0.001000 Time 0.019834 +2023-10-02 20:31:09,730 - Epoch: [5][ 240/ 1236] Overall Loss 0.449667 Objective Loss 0.449667 LR 0.001000 Time 0.019690 +2023-10-02 20:31:09,893 - Epoch: [5][ 250/ 1236] Overall Loss 0.450552 Objective Loss 0.450552 LR 0.001000 Time 0.019554 +2023-10-02 20:31:10,057 - Epoch: [5][ 260/ 1236] Overall Loss 0.451159 Objective Loss 0.451159 LR 0.001000 Time 0.019431 +2023-10-02 20:31:10,220 - Epoch: [5][ 270/ 1236] Overall Loss 0.451800 Objective Loss 0.451800 LR 0.001000 Time 0.019313 +2023-10-02 20:31:10,384 - Epoch: [5][ 280/ 1236] Overall Loss 0.452084 Objective Loss 0.452084 LR 0.001000 Time 0.019207 +2023-10-02 20:31:10,547 - Epoch: [5][ 290/ 1236] Overall Loss 0.452976 Objective Loss 0.452976 LR 0.001000 Time 0.019107 +2023-10-02 20:31:10,712 - Epoch: [5][ 300/ 1236] Overall Loss 0.454686 Objective Loss 0.454686 LR 0.001000 Time 0.019018 +2023-10-02 20:31:10,877 - Epoch: [5][ 310/ 1236] Overall Loss 0.455148 Objective Loss 0.455148 LR 0.001000 Time 0.018935 +2023-10-02 20:31:11,040 - Epoch: [5][ 320/ 1236] Overall Loss 0.454894 Objective Loss 0.454894 LR 0.001000 Time 0.018853 +2023-10-02 20:31:11,203 - Epoch: [5][ 330/ 1236] Overall Loss 0.454946 Objective Loss 0.454946 LR 0.001000 Time 0.018775 +2023-10-02 20:31:11,367 - Epoch: [5][ 340/ 1236] Overall Loss 0.455006 Objective Loss 0.455006 LR 0.001000 Time 0.018703 +2023-10-02 20:31:11,530 - Epoch: [5][ 350/ 1236] Overall Loss 0.454760 Objective Loss 0.454760 LR 0.001000 Time 0.018633 +2023-10-02 20:31:11,693 - Epoch: [5][ 360/ 1236] Overall Loss 0.455535 Objective Loss 0.455535 LR 0.001000 Time 0.018569 +2023-10-02 20:31:11,856 - Epoch: [5][ 370/ 1236] Overall Loss 0.454618 Objective Loss 0.454618 LR 0.001000 Time 0.018507 +2023-10-02 20:31:12,020 - Epoch: [5][ 380/ 1236] Overall Loss 0.454725 Objective Loss 0.454725 LR 0.001000 Time 0.018450 +2023-10-02 20:31:12,184 - Epoch: [5][ 390/ 1236] Overall Loss 0.455115 Objective Loss 0.455115 LR 0.001000 Time 0.018397 +2023-10-02 20:31:12,349 - Epoch: [5][ 400/ 1236] Overall Loss 0.455137 Objective Loss 0.455137 LR 0.001000 Time 0.018350 +2023-10-02 20:31:12,514 - Epoch: [5][ 410/ 1236] Overall Loss 0.455355 Objective Loss 0.455355 LR 0.001000 Time 0.018302 +2023-10-02 20:31:12,679 - Epoch: [5][ 420/ 1236] Overall Loss 0.454765 Objective Loss 0.454765 LR 0.001000 Time 0.018260 +2023-10-02 20:31:12,844 - Epoch: [5][ 430/ 1236] Overall Loss 0.455250 Objective Loss 0.455250 LR 0.001000 Time 0.018217 +2023-10-02 20:31:13,009 - Epoch: [5][ 440/ 1236] Overall Loss 0.456320 Objective Loss 0.456320 LR 0.001000 Time 0.018178 +2023-10-02 20:31:13,173 - Epoch: [5][ 450/ 1236] Overall Loss 0.456548 Objective Loss 0.456548 LR 0.001000 Time 0.018138 +2023-10-02 20:31:13,338 - Epoch: [5][ 460/ 1236] Overall Loss 0.456022 Objective Loss 0.456022 LR 0.001000 Time 0.018102 +2023-10-02 20:31:13,501 - Epoch: [5][ 470/ 1236] Overall Loss 0.456877 Objective Loss 0.456877 LR 0.001000 Time 0.018063 +2023-10-02 20:31:13,666 - Epoch: [5][ 480/ 1236] Overall Loss 0.456484 Objective Loss 0.456484 LR 0.001000 Time 0.018028 +2023-10-02 20:31:13,828 - Epoch: [5][ 490/ 1236] Overall Loss 0.456338 Objective Loss 0.456338 LR 0.001000 Time 0.017992 +2023-10-02 20:31:13,993 - Epoch: [5][ 500/ 1236] Overall Loss 0.456624 Objective Loss 0.456624 LR 0.001000 Time 0.017960 +2023-10-02 20:31:14,156 - Epoch: [5][ 510/ 1236] Overall Loss 0.456224 Objective Loss 0.456224 LR 0.001000 Time 0.017927 +2023-10-02 20:31:14,320 - Epoch: [5][ 520/ 1236] Overall Loss 0.457032 Objective Loss 0.457032 LR 0.001000 Time 0.017898 +2023-10-02 20:31:14,484 - Epoch: [5][ 530/ 1236] Overall Loss 0.456901 Objective Loss 0.456901 LR 0.001000 Time 0.017868 +2023-10-02 20:31:14,648 - Epoch: [5][ 540/ 1236] Overall Loss 0.457249 Objective Loss 0.457249 LR 0.001000 Time 0.017841 +2023-10-02 20:31:14,811 - Epoch: [5][ 550/ 1236] Overall Loss 0.457357 Objective Loss 0.457357 LR 0.001000 Time 0.017813 +2023-10-02 20:31:14,976 - Epoch: [5][ 560/ 1236] Overall Loss 0.457393 Objective Loss 0.457393 LR 0.001000 Time 0.017788 +2023-10-02 20:31:15,139 - Epoch: [5][ 570/ 1236] Overall Loss 0.456912 Objective Loss 0.456912 LR 0.001000 Time 0.017761 +2023-10-02 20:31:15,303 - Epoch: [5][ 580/ 1236] Overall Loss 0.456417 Objective Loss 0.456417 LR 0.001000 Time 0.017738 +2023-10-02 20:31:15,467 - Epoch: [5][ 590/ 1236] Overall Loss 0.456817 Objective Loss 0.456817 LR 0.001000 Time 0.017714 +2023-10-02 20:31:15,631 - Epoch: [5][ 600/ 1236] Overall Loss 0.456638 Objective Loss 0.456638 LR 0.001000 Time 0.017692 +2023-10-02 20:31:15,794 - Epoch: [5][ 610/ 1236] Overall Loss 0.456628 Objective Loss 0.456628 LR 0.001000 Time 0.017669 +2023-10-02 20:31:15,958 - Epoch: [5][ 620/ 1236] Overall Loss 0.456098 Objective Loss 0.456098 LR 0.001000 Time 0.017648 +2023-10-02 20:31:16,121 - Epoch: [5][ 630/ 1236] Overall Loss 0.455895 Objective Loss 0.455895 LR 0.001000 Time 0.017626 +2023-10-02 20:31:16,286 - Epoch: [5][ 640/ 1236] Overall Loss 0.455368 Objective Loss 0.455368 LR 0.001000 Time 0.017607 +2023-10-02 20:31:16,449 - Epoch: [5][ 650/ 1236] Overall Loss 0.455085 Objective Loss 0.455085 LR 0.001000 Time 0.017587 +2023-10-02 20:31:16,613 - Epoch: [5][ 660/ 1236] Overall Loss 0.455945 Objective Loss 0.455945 LR 0.001000 Time 0.017569 +2023-10-02 20:31:16,777 - Epoch: [5][ 670/ 1236] Overall Loss 0.456226 Objective Loss 0.456226 LR 0.001000 Time 0.017550 +2023-10-02 20:31:16,941 - Epoch: [5][ 680/ 1236] Overall Loss 0.456726 Objective Loss 0.456726 LR 0.001000 Time 0.017534 +2023-10-02 20:31:17,105 - Epoch: [5][ 690/ 1236] Overall Loss 0.456383 Objective Loss 0.456383 LR 0.001000 Time 0.017516 +2023-10-02 20:31:17,269 - Epoch: [5][ 700/ 1236] Overall Loss 0.456502 Objective Loss 0.456502 LR 0.001000 Time 0.017500 +2023-10-02 20:31:17,432 - Epoch: [5][ 710/ 1236] Overall Loss 0.456847 Objective Loss 0.456847 LR 0.001000 Time 0.017483 +2023-10-02 20:31:17,597 - Epoch: [5][ 720/ 1236] Overall Loss 0.456116 Objective Loss 0.456116 LR 0.001000 Time 0.017469 +2023-10-02 20:31:17,761 - Epoch: [5][ 730/ 1236] Overall Loss 0.456392 Objective Loss 0.456392 LR 0.001000 Time 0.017453 +2023-10-02 20:31:17,926 - Epoch: [5][ 740/ 1236] Overall Loss 0.456326 Objective Loss 0.456326 LR 0.001000 Time 0.017440 +2023-10-02 20:31:18,089 - Epoch: [5][ 750/ 1236] Overall Loss 0.455614 Objective Loss 0.455614 LR 0.001000 Time 0.017424 +2023-10-02 20:31:18,253 - Epoch: [5][ 760/ 1236] Overall Loss 0.455406 Objective Loss 0.455406 LR 0.001000 Time 0.017411 +2023-10-02 20:31:18,416 - Epoch: [5][ 770/ 1236] Overall Loss 0.455608 Objective Loss 0.455608 LR 0.001000 Time 0.017396 +2023-10-02 20:31:18,581 - Epoch: [5][ 780/ 1236] Overall Loss 0.455400 Objective Loss 0.455400 LR 0.001000 Time 0.017384 +2023-10-02 20:31:18,744 - Epoch: [5][ 790/ 1236] Overall Loss 0.455681 Objective Loss 0.455681 LR 0.001000 Time 0.017370 +2023-10-02 20:31:18,909 - Epoch: [5][ 800/ 1236] Overall Loss 0.455781 Objective Loss 0.455781 LR 0.001000 Time 0.017359 +2023-10-02 20:31:19,072 - Epoch: [5][ 810/ 1236] Overall Loss 0.455421 Objective Loss 0.455421 LR 0.001000 Time 0.017346 +2023-10-02 20:31:19,237 - Epoch: [5][ 820/ 1236] Overall Loss 0.454894 Objective Loss 0.454894 LR 0.001000 Time 0.017335 +2023-10-02 20:31:19,401 - Epoch: [5][ 830/ 1236] Overall Loss 0.454726 Objective Loss 0.454726 LR 0.001000 Time 0.017322 +2023-10-02 20:31:19,565 - Epoch: [5][ 840/ 1236] Overall Loss 0.455152 Objective Loss 0.455152 LR 0.001000 Time 0.017312 +2023-10-02 20:31:19,729 - Epoch: [5][ 850/ 1236] Overall Loss 0.454736 Objective Loss 0.454736 LR 0.001000 Time 0.017300 +2023-10-02 20:31:19,894 - Epoch: [5][ 860/ 1236] Overall Loss 0.454555 Objective Loss 0.454555 LR 0.001000 Time 0.017290 +2023-10-02 20:31:20,057 - Epoch: [5][ 870/ 1236] Overall Loss 0.454417 Objective Loss 0.454417 LR 0.001000 Time 0.017279 +2023-10-02 20:31:20,222 - Epoch: [5][ 880/ 1236] Overall Loss 0.454161 Objective Loss 0.454161 LR 0.001000 Time 0.017270 +2023-10-02 20:31:20,385 - Epoch: [5][ 890/ 1236] Overall Loss 0.454290 Objective Loss 0.454290 LR 0.001000 Time 0.017259 +2023-10-02 20:31:20,550 - Epoch: [5][ 900/ 1236] Overall Loss 0.454288 Objective Loss 0.454288 LR 0.001000 Time 0.017250 +2023-10-02 20:31:20,713 - Epoch: [5][ 910/ 1236] Overall Loss 0.454600 Objective Loss 0.454600 LR 0.001000 Time 0.017239 +2023-10-02 20:31:20,878 - Epoch: [5][ 920/ 1236] Overall Loss 0.454443 Objective Loss 0.454443 LR 0.001000 Time 0.017230 +2023-10-02 20:31:21,041 - Epoch: [5][ 930/ 1236] Overall Loss 0.454886 Objective Loss 0.454886 LR 0.001000 Time 0.017221 +2023-10-02 20:31:21,206 - Epoch: [5][ 940/ 1236] Overall Loss 0.454996 Objective Loss 0.454996 LR 0.001000 Time 0.017212 +2023-10-02 20:31:21,369 - Epoch: [5][ 950/ 1236] Overall Loss 0.455322 Objective Loss 0.455322 LR 0.001000 Time 0.017203 +2023-10-02 20:31:21,534 - Epoch: [5][ 960/ 1236] Overall Loss 0.455452 Objective Loss 0.455452 LR 0.001000 Time 0.017195 +2023-10-02 20:31:21,698 - Epoch: [5][ 970/ 1236] Overall Loss 0.455486 Objective Loss 0.455486 LR 0.001000 Time 0.017186 +2023-10-02 20:31:21,863 - Epoch: [5][ 980/ 1236] Overall Loss 0.455100 Objective Loss 0.455100 LR 0.001000 Time 0.017179 +2023-10-02 20:31:22,027 - Epoch: [5][ 990/ 1236] Overall Loss 0.455390 Objective Loss 0.455390 LR 0.001000 Time 0.017170 +2023-10-02 20:31:22,192 - Epoch: [5][ 1000/ 1236] Overall Loss 0.455669 Objective Loss 0.455669 LR 0.001000 Time 0.017163 +2023-10-02 20:31:22,355 - Epoch: [5][ 1010/ 1236] Overall Loss 0.455645 Objective Loss 0.455645 LR 0.001000 Time 0.017155 +2023-10-02 20:31:22,520 - Epoch: [5][ 1020/ 1236] Overall Loss 0.455679 Objective Loss 0.455679 LR 0.001000 Time 0.017148 +2023-10-02 20:31:22,684 - Epoch: [5][ 1030/ 1236] Overall Loss 0.455677 Objective Loss 0.455677 LR 0.001000 Time 0.017141 +2023-10-02 20:31:22,849 - Epoch: [5][ 1040/ 1236] Overall Loss 0.455665 Objective Loss 0.455665 LR 0.001000 Time 0.017134 +2023-10-02 20:31:23,013 - Epoch: [5][ 1050/ 1236] Overall Loss 0.455858 Objective Loss 0.455858 LR 0.001000 Time 0.017126 +2023-10-02 20:31:23,177 - Epoch: [5][ 1060/ 1236] Overall Loss 0.455708 Objective Loss 0.455708 LR 0.001000 Time 0.017120 +2023-10-02 20:31:23,341 - Epoch: [5][ 1070/ 1236] Overall Loss 0.455594 Objective Loss 0.455594 LR 0.001000 Time 0.017112 +2023-10-02 20:31:23,506 - Epoch: [5][ 1080/ 1236] Overall Loss 0.455912 Objective Loss 0.455912 LR 0.001000 Time 0.017106 +2023-10-02 20:31:23,670 - Epoch: [5][ 1090/ 1236] Overall Loss 0.455865 Objective Loss 0.455865 LR 0.001000 Time 0.017100 +2023-10-02 20:31:23,835 - Epoch: [5][ 1100/ 1236] Overall Loss 0.455618 Objective Loss 0.455618 LR 0.001000 Time 0.017094 +2023-10-02 20:31:23,999 - Epoch: [5][ 1110/ 1236] Overall Loss 0.455541 Objective Loss 0.455541 LR 0.001000 Time 0.017087 +2023-10-02 20:31:24,164 - Epoch: [5][ 1120/ 1236] Overall Loss 0.455130 Objective Loss 0.455130 LR 0.001000 Time 0.017082 +2023-10-02 20:31:24,328 - Epoch: [5][ 1130/ 1236] Overall Loss 0.454694 Objective Loss 0.454694 LR 0.001000 Time 0.017076 +2023-10-02 20:31:24,493 - Epoch: [5][ 1140/ 1236] Overall Loss 0.454509 Objective Loss 0.454509 LR 0.001000 Time 0.017071 +2023-10-02 20:31:24,658 - Epoch: [5][ 1150/ 1236] Overall Loss 0.454880 Objective Loss 0.454880 LR 0.001000 Time 0.017065 +2023-10-02 20:31:24,823 - Epoch: [5][ 1160/ 1236] Overall Loss 0.454642 Objective Loss 0.454642 LR 0.001000 Time 0.017060 +2023-10-02 20:31:24,986 - Epoch: [5][ 1170/ 1236] Overall Loss 0.454563 Objective Loss 0.454563 LR 0.001000 Time 0.017054 +2023-10-02 20:31:25,151 - Epoch: [5][ 1180/ 1236] Overall Loss 0.454207 Objective Loss 0.454207 LR 0.001000 Time 0.017049 +2023-10-02 20:31:25,315 - Epoch: [5][ 1190/ 1236] Overall Loss 0.454457 Objective Loss 0.454457 LR 0.001000 Time 0.017043 +2023-10-02 20:31:25,480 - Epoch: [5][ 1200/ 1236] Overall Loss 0.454459 Objective Loss 0.454459 LR 0.001000 Time 0.017037 +2023-10-02 20:31:25,643 - Epoch: [5][ 1210/ 1236] Overall Loss 0.454421 Objective Loss 0.454421 LR 0.001000 Time 0.017031 +2023-10-02 20:31:25,808 - Epoch: [5][ 1220/ 1236] Overall Loss 0.454046 Objective Loss 0.454046 LR 0.001000 Time 0.017027 +2023-10-02 20:31:26,023 - Epoch: [5][ 1230/ 1236] Overall Loss 0.453575 Objective Loss 0.453575 LR 0.001000 Time 0.017063 +2023-10-02 20:31:26,118 - Epoch: [5][ 1236/ 1236] Overall Loss 0.453425 Objective Loss 0.453425 Top1 80.448065 Top5 96.537678 LR 0.001000 Time 0.017057 +2023-10-02 20:31:26,260 - --- validate (epoch=5)----------- +2023-10-02 20:31:26,261 - 29943 samples (256 per mini-batch) +2023-10-02 20:31:26,715 - Epoch: [5][ 10/ 117] Loss 0.419440 Top1 76.718750 Top5 97.109375 +2023-10-02 20:31:26,817 - Epoch: [5][ 20/ 117] Loss 0.407351 Top1 77.753906 Top5 97.226562 +2023-10-02 20:31:26,915 - Epoch: [5][ 30/ 117] Loss 0.397409 Top1 77.526042 Top5 97.213542 +2023-10-02 20:31:27,017 - Epoch: [5][ 40/ 117] Loss 0.405055 Top1 77.685547 Top5 97.197266 +2023-10-02 20:31:27,115 - Epoch: [5][ 50/ 117] Loss 0.404882 Top1 77.656250 Top5 97.328125 +2023-10-02 20:31:27,215 - Epoch: [5][ 60/ 117] Loss 0.405307 Top1 77.688802 Top5 97.441406 +2023-10-02 20:31:27,314 - Epoch: [5][ 70/ 117] Loss 0.405488 Top1 77.656250 Top5 97.477679 +2023-10-02 20:31:27,415 - Epoch: [5][ 80/ 117] Loss 0.404470 Top1 77.578125 Top5 97.529297 +2023-10-02 20:31:27,513 - Epoch: [5][ 90/ 117] Loss 0.409306 Top1 77.439236 Top5 97.500000 +2023-10-02 20:31:27,615 - Epoch: [5][ 100/ 117] Loss 0.411163 Top1 77.453125 Top5 97.515625 +2023-10-02 20:31:27,729 - Epoch: [5][ 110/ 117] Loss 0.410370 Top1 77.510653 Top5 97.531960 +2023-10-02 20:31:27,788 - Epoch: [5][ 117/ 117] Loss 0.409093 Top1 77.660889 Top5 97.531977 +2023-10-02 20:31:27,928 - ==> Top1: 77.661 Top5: 97.532 Loss: 0.409 + +2023-10-02 20:31:27,928 - ==> Confusion: +[[ 878 1 2 0 12 2 0 0 7 116 0 1 0 4 3 1 4 1 1 0 17] + [ 0 1043 1 3 6 26 2 17 3 0 3 4 0 0 2 4 1 0 9 1 6] + [ 16 1 912 13 5 2 29 14 0 5 3 5 6 3 6 3 0 6 9 2 16] + [ 1 2 18 876 1 6 1 0 4 1 26 0 12 4 72 5 2 7 36 0 15] + [ 22 11 2 0 964 6 1 0 1 17 0 3 0 2 6 2 6 0 0 4 3] + [ 2 101 2 0 11 901 1 28 3 3 7 10 6 14 7 1 5 3 0 4 7] + [ 0 5 41 0 1 2 1090 12 0 0 3 2 1 2 0 9 0 2 1 10 10] + [ 6 18 19 0 6 32 3 1006 0 0 4 15 1 1 0 4 0 1 80 15 7] + [ 23 5 0 1 1 0 0 2 950 66 6 4 5 6 14 3 0 1 2 0 0] + [ 95 0 0 0 9 0 1 1 26 947 0 2 1 14 4 0 0 4 0 3 12] + [ 6 2 4 10 5 2 3 8 29 1 926 2 0 22 4 5 1 1 14 0 8] + [ 0 0 0 0 0 19 2 0 0 0 0 916 38 8 0 2 1 21 0 15 13] + [ 0 1 2 3 1 3 0 1 0 0 0 67 933 0 6 10 3 26 2 3 7] + [ 2 0 1 0 3 22 1 3 15 21 2 8 3 1020 3 1 2 0 0 2 10] + [ 10 2 0 5 21 0 0 0 39 18 3 0 5 1 968 0 0 4 5 0 20] + [ 1 1 5 0 3 0 0 0 0 1 0 17 13 1 0 1032 16 24 0 7 13] + [ 1 18 2 1 17 8 1 0 1 0 0 5 0 3 4 9 1072 1 0 5 13] + [ 0 0 0 1 0 0 2 0 2 0 0 10 20 0 0 5 1 990 0 0 7] + [ 2 13 5 14 2 1 1 21 6 0 6 0 6 0 14 1 1 1 962 2 10] + [ 0 5 3 1 2 8 13 22 0 0 0 18 3 2 1 2 2 1 3 1059 7] + [ 196 385 111 55 216 248 51 101 170 182 149 220 459 324 203 106 159 106 279 376 3809]] + +2023-10-02 20:31:27,930 - ==> Best [Top1: 77.661 Top5: 97.532 Sparsity:0.00 Params: 169472 on epoch: 5] +2023-10-02 20:31:27,930 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:31:27,943 - + +2023-10-02 20:31:27,943 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:31:28,918 - Epoch: [6][ 10/ 1236] Overall Loss 0.359992 Objective Loss 0.359992 LR 0.001000 Time 0.097368 +2023-10-02 20:31:29,083 - Epoch: [6][ 20/ 1236] Overall Loss 0.398223 Objective Loss 0.398223 LR 0.001000 Time 0.056924 +2023-10-02 20:31:29,247 - Epoch: [6][ 30/ 1236] Overall Loss 0.393535 Objective Loss 0.393535 LR 0.001000 Time 0.043409 +2023-10-02 20:31:29,412 - Epoch: [6][ 40/ 1236] Overall Loss 0.405931 Objective Loss 0.405931 LR 0.001000 Time 0.036672 +2023-10-02 20:31:29,575 - Epoch: [6][ 50/ 1236] Overall Loss 0.410769 Objective Loss 0.410769 LR 0.001000 Time 0.032597 +2023-10-02 20:31:29,740 - Epoch: [6][ 60/ 1236] Overall Loss 0.411893 Objective Loss 0.411893 LR 0.001000 Time 0.029916 +2023-10-02 20:31:29,904 - Epoch: [6][ 70/ 1236] Overall Loss 0.410483 Objective Loss 0.410483 LR 0.001000 Time 0.027968 +2023-10-02 20:31:30,068 - Epoch: [6][ 80/ 1236] Overall Loss 0.416205 Objective Loss 0.416205 LR 0.001000 Time 0.026529 +2023-10-02 20:31:30,232 - Epoch: [6][ 90/ 1236] Overall Loss 0.412095 Objective Loss 0.412095 LR 0.001000 Time 0.025390 +2023-10-02 20:31:30,397 - Epoch: [6][ 100/ 1236] Overall Loss 0.415011 Objective Loss 0.415011 LR 0.001000 Time 0.024499 +2023-10-02 20:31:30,560 - Epoch: [6][ 110/ 1236] Overall Loss 0.414499 Objective Loss 0.414499 LR 0.001000 Time 0.023751 +2023-10-02 20:31:30,725 - Epoch: [6][ 120/ 1236] Overall Loss 0.415052 Objective Loss 0.415052 LR 0.001000 Time 0.023145 +2023-10-02 20:31:30,888 - Epoch: [6][ 130/ 1236] Overall Loss 0.415690 Objective Loss 0.415690 LR 0.001000 Time 0.022616 +2023-10-02 20:31:31,053 - Epoch: [6][ 140/ 1236] Overall Loss 0.416916 Objective Loss 0.416916 LR 0.001000 Time 0.022182 +2023-10-02 20:31:31,217 - Epoch: [6][ 150/ 1236] Overall Loss 0.419072 Objective Loss 0.419072 LR 0.001000 Time 0.021789 +2023-10-02 20:31:31,382 - Epoch: [6][ 160/ 1236] Overall Loss 0.417076 Objective Loss 0.417076 LR 0.001000 Time 0.021457 +2023-10-02 20:31:31,545 - Epoch: [6][ 170/ 1236] Overall Loss 0.417255 Objective Loss 0.417255 LR 0.001000 Time 0.021152 +2023-10-02 20:31:31,709 - Epoch: [6][ 180/ 1236] Overall Loss 0.417487 Objective Loss 0.417487 LR 0.001000 Time 0.020890 +2023-10-02 20:31:31,872 - Epoch: [6][ 190/ 1236] Overall Loss 0.419137 Objective Loss 0.419137 LR 0.001000 Time 0.020647 +2023-10-02 20:31:32,038 - Epoch: [6][ 200/ 1236] Overall Loss 0.419802 Objective Loss 0.419802 LR 0.001000 Time 0.020438 +2023-10-02 20:31:32,201 - Epoch: [6][ 210/ 1236] Overall Loss 0.422477 Objective Loss 0.422477 LR 0.001000 Time 0.020240 +2023-10-02 20:31:32,366 - Epoch: [6][ 220/ 1236] Overall Loss 0.423181 Objective Loss 0.423181 LR 0.001000 Time 0.020070 +2023-10-02 20:31:32,529 - Epoch: [6][ 230/ 1236] Overall Loss 0.423081 Objective Loss 0.423081 LR 0.001000 Time 0.019905 +2023-10-02 20:31:32,694 - Epoch: [6][ 240/ 1236] Overall Loss 0.422366 Objective Loss 0.422366 LR 0.001000 Time 0.019763 +2023-10-02 20:31:32,858 - Epoch: [6][ 250/ 1236] Overall Loss 0.422007 Objective Loss 0.422007 LR 0.001000 Time 0.019625 +2023-10-02 20:31:33,023 - Epoch: [6][ 260/ 1236] Overall Loss 0.421881 Objective Loss 0.421881 LR 0.001000 Time 0.019505 +2023-10-02 20:31:33,186 - Epoch: [6][ 270/ 1236] Overall Loss 0.421925 Objective Loss 0.421925 LR 0.001000 Time 0.019385 +2023-10-02 20:31:33,352 - Epoch: [6][ 280/ 1236] Overall Loss 0.423384 Objective Loss 0.423384 LR 0.001000 Time 0.019283 +2023-10-02 20:31:33,516 - Epoch: [6][ 290/ 1236] Overall Loss 0.424025 Objective Loss 0.424025 LR 0.001000 Time 0.019185 +2023-10-02 20:31:33,682 - Epoch: [6][ 300/ 1236] Overall Loss 0.426006 Objective Loss 0.426006 LR 0.001000 Time 0.019096 +2023-10-02 20:31:33,847 - Epoch: [6][ 310/ 1236] Overall Loss 0.426373 Objective Loss 0.426373 LR 0.001000 Time 0.019012 +2023-10-02 20:31:34,013 - Epoch: [6][ 320/ 1236] Overall Loss 0.427228 Objective Loss 0.427228 LR 0.001000 Time 0.018935 +2023-10-02 20:31:34,176 - Epoch: [6][ 330/ 1236] Overall Loss 0.426828 Objective Loss 0.426828 LR 0.001000 Time 0.018854 +2023-10-02 20:31:34,340 - Epoch: [6][ 340/ 1236] Overall Loss 0.426788 Objective Loss 0.426788 LR 0.001000 Time 0.018783 +2023-10-02 20:31:34,503 - Epoch: [6][ 350/ 1236] Overall Loss 0.426242 Objective Loss 0.426242 LR 0.001000 Time 0.018709 +2023-10-02 20:31:34,668 - Epoch: [6][ 360/ 1236] Overall Loss 0.425287 Objective Loss 0.425287 LR 0.001000 Time 0.018649 +2023-10-02 20:31:34,835 - Epoch: [6][ 370/ 1236] Overall Loss 0.425375 Objective Loss 0.425375 LR 0.001000 Time 0.018595 +2023-10-02 20:31:35,002 - Epoch: [6][ 380/ 1236] Overall Loss 0.425880 Objective Loss 0.425880 LR 0.001000 Time 0.018544 +2023-10-02 20:31:35,170 - Epoch: [6][ 390/ 1236] Overall Loss 0.425729 Objective Loss 0.425729 LR 0.001000 Time 0.018497 +2023-10-02 20:31:35,335 - Epoch: [6][ 400/ 1236] Overall Loss 0.425087 Objective Loss 0.425087 LR 0.001000 Time 0.018447 +2023-10-02 20:31:35,499 - Epoch: [6][ 410/ 1236] Overall Loss 0.425438 Objective Loss 0.425438 LR 0.001000 Time 0.018397 +2023-10-02 20:31:35,664 - Epoch: [6][ 420/ 1236] Overall Loss 0.424839 Objective Loss 0.424839 LR 0.001000 Time 0.018350 +2023-10-02 20:31:35,828 - Epoch: [6][ 430/ 1236] Overall Loss 0.423723 Objective Loss 0.423723 LR 0.001000 Time 0.018304 +2023-10-02 20:31:35,992 - Epoch: [6][ 440/ 1236] Overall Loss 0.423541 Objective Loss 0.423541 LR 0.001000 Time 0.018261 +2023-10-02 20:31:36,156 - Epoch: [6][ 450/ 1236] Overall Loss 0.423112 Objective Loss 0.423112 LR 0.001000 Time 0.018218 +2023-10-02 20:31:36,320 - Epoch: [6][ 460/ 1236] Overall Loss 0.422720 Objective Loss 0.422720 LR 0.001000 Time 0.018177 +2023-10-02 20:31:36,483 - Epoch: [6][ 470/ 1236] Overall Loss 0.422184 Objective Loss 0.422184 LR 0.001000 Time 0.018138 +2023-10-02 20:31:36,648 - Epoch: [6][ 480/ 1236] Overall Loss 0.421660 Objective Loss 0.421660 LR 0.001000 Time 0.018102 +2023-10-02 20:31:36,811 - Epoch: [6][ 490/ 1236] Overall Loss 0.421573 Objective Loss 0.421573 LR 0.001000 Time 0.018066 +2023-10-02 20:31:36,976 - Epoch: [6][ 500/ 1236] Overall Loss 0.422163 Objective Loss 0.422163 LR 0.001000 Time 0.018033 +2023-10-02 20:31:37,139 - Epoch: [6][ 510/ 1236] Overall Loss 0.422597 Objective Loss 0.422597 LR 0.001000 Time 0.018000 +2023-10-02 20:31:37,304 - Epoch: [6][ 520/ 1236] Overall Loss 0.422324 Objective Loss 0.422324 LR 0.001000 Time 0.017969 +2023-10-02 20:31:37,467 - Epoch: [6][ 530/ 1236] Overall Loss 0.422060 Objective Loss 0.422060 LR 0.001000 Time 0.017938 +2023-10-02 20:31:37,632 - Epoch: [6][ 540/ 1236] Overall Loss 0.422010 Objective Loss 0.422010 LR 0.001000 Time 0.017909 +2023-10-02 20:31:37,795 - Epoch: [6][ 550/ 1236] Overall Loss 0.421790 Objective Loss 0.421790 LR 0.001000 Time 0.017881 +2023-10-02 20:31:37,959 - Epoch: [6][ 560/ 1236] Overall Loss 0.421798 Objective Loss 0.421798 LR 0.001000 Time 0.017854 +2023-10-02 20:31:38,123 - Epoch: [6][ 570/ 1236] Overall Loss 0.421906 Objective Loss 0.421906 LR 0.001000 Time 0.017828 +2023-10-02 20:31:38,287 - Epoch: [6][ 580/ 1236] Overall Loss 0.422412 Objective Loss 0.422412 LR 0.001000 Time 0.017803 +2023-10-02 20:31:38,451 - Epoch: [6][ 590/ 1236] Overall Loss 0.423448 Objective Loss 0.423448 LR 0.001000 Time 0.017778 +2023-10-02 20:31:38,615 - Epoch: [6][ 600/ 1236] Overall Loss 0.423633 Objective Loss 0.423633 LR 0.001000 Time 0.017755 +2023-10-02 20:31:38,779 - Epoch: [6][ 610/ 1236] Overall Loss 0.424536 Objective Loss 0.424536 LR 0.001000 Time 0.017731 +2023-10-02 20:31:38,943 - Epoch: [6][ 620/ 1236] Overall Loss 0.424321 Objective Loss 0.424321 LR 0.001000 Time 0.017709 +2023-10-02 20:31:39,107 - Epoch: [6][ 630/ 1236] Overall Loss 0.424194 Objective Loss 0.424194 LR 0.001000 Time 0.017688 +2023-10-02 20:31:39,271 - Epoch: [6][ 640/ 1236] Overall Loss 0.424449 Objective Loss 0.424449 LR 0.001000 Time 0.017668 +2023-10-02 20:31:39,435 - Epoch: [6][ 650/ 1236] Overall Loss 0.424127 Objective Loss 0.424127 LR 0.001000 Time 0.017648 +2023-10-02 20:31:39,600 - Epoch: [6][ 660/ 1236] Overall Loss 0.424587 Objective Loss 0.424587 LR 0.001000 Time 0.017630 +2023-10-02 20:31:39,764 - Epoch: [6][ 670/ 1236] Overall Loss 0.424867 Objective Loss 0.424867 LR 0.001000 Time 0.017611 +2023-10-02 20:31:39,928 - Epoch: [6][ 680/ 1236] Overall Loss 0.425209 Objective Loss 0.425209 LR 0.001000 Time 0.017593 +2023-10-02 20:31:40,092 - Epoch: [6][ 690/ 1236] Overall Loss 0.424812 Objective Loss 0.424812 LR 0.001000 Time 0.017575 +2023-10-02 20:31:40,257 - Epoch: [6][ 700/ 1236] Overall Loss 0.424714 Objective Loss 0.424714 LR 0.001000 Time 0.017559 +2023-10-02 20:31:40,421 - Epoch: [6][ 710/ 1236] Overall Loss 0.424759 Objective Loss 0.424759 LR 0.001000 Time 0.017542 +2023-10-02 20:31:40,585 - Epoch: [6][ 720/ 1236] Overall Loss 0.424317 Objective Loss 0.424317 LR 0.001000 Time 0.017526 +2023-10-02 20:31:40,748 - Epoch: [6][ 730/ 1236] Overall Loss 0.424667 Objective Loss 0.424667 LR 0.001000 Time 0.017509 +2023-10-02 20:31:40,912 - Epoch: [6][ 740/ 1236] Overall Loss 0.424654 Objective Loss 0.424654 LR 0.001000 Time 0.017494 +2023-10-02 20:31:41,076 - Epoch: [6][ 750/ 1236] Overall Loss 0.424873 Objective Loss 0.424873 LR 0.001000 Time 0.017479 +2023-10-02 20:31:41,240 - Epoch: [6][ 760/ 1236] Overall Loss 0.424290 Objective Loss 0.424290 LR 0.001000 Time 0.017464 +2023-10-02 20:31:41,404 - Epoch: [6][ 770/ 1236] Overall Loss 0.424257 Objective Loss 0.424257 LR 0.001000 Time 0.017450 +2023-10-02 20:31:41,568 - Epoch: [6][ 780/ 1236] Overall Loss 0.424217 Objective Loss 0.424217 LR 0.001000 Time 0.017436 +2023-10-02 20:31:41,732 - Epoch: [6][ 790/ 1236] Overall Loss 0.424858 Objective Loss 0.424858 LR 0.001000 Time 0.017422 +2023-10-02 20:31:41,896 - Epoch: [6][ 800/ 1236] Overall Loss 0.425006 Objective Loss 0.425006 LR 0.001000 Time 0.017409 +2023-10-02 20:31:42,059 - Epoch: [6][ 810/ 1236] Overall Loss 0.424740 Objective Loss 0.424740 LR 0.001000 Time 0.017395 +2023-10-02 20:31:42,223 - Epoch: [6][ 820/ 1236] Overall Loss 0.424813 Objective Loss 0.424813 LR 0.001000 Time 0.017383 +2023-10-02 20:31:42,387 - Epoch: [6][ 830/ 1236] Overall Loss 0.424735 Objective Loss 0.424735 LR 0.001000 Time 0.017371 +2023-10-02 20:31:42,551 - Epoch: [6][ 840/ 1236] Overall Loss 0.424985 Objective Loss 0.424985 LR 0.001000 Time 0.017359 +2023-10-02 20:31:42,715 - Epoch: [6][ 850/ 1236] Overall Loss 0.424793 Objective Loss 0.424793 LR 0.001000 Time 0.017347 +2023-10-02 20:31:42,880 - Epoch: [6][ 860/ 1236] Overall Loss 0.424365 Objective Loss 0.424365 LR 0.001000 Time 0.017336 +2023-10-02 20:31:43,044 - Epoch: [6][ 870/ 1236] Overall Loss 0.424335 Objective Loss 0.424335 LR 0.001000 Time 0.017325 +2023-10-02 20:31:43,208 - Epoch: [6][ 880/ 1236] Overall Loss 0.423850 Objective Loss 0.423850 LR 0.001000 Time 0.017314 +2023-10-02 20:31:43,371 - Epoch: [6][ 890/ 1236] Overall Loss 0.423818 Objective Loss 0.423818 LR 0.001000 Time 0.017303 +2023-10-02 20:31:43,535 - Epoch: [6][ 900/ 1236] Overall Loss 0.423668 Objective Loss 0.423668 LR 0.001000 Time 0.017293 +2023-10-02 20:31:43,699 - Epoch: [6][ 910/ 1236] Overall Loss 0.423623 Objective Loss 0.423623 LR 0.001000 Time 0.017283 +2023-10-02 20:31:43,863 - Epoch: [6][ 920/ 1236] Overall Loss 0.423475 Objective Loss 0.423475 LR 0.001000 Time 0.017273 +2023-10-02 20:31:44,027 - Epoch: [6][ 930/ 1236] Overall Loss 0.423169 Objective Loss 0.423169 LR 0.001000 Time 0.017263 +2023-10-02 20:31:44,191 - Epoch: [6][ 940/ 1236] Overall Loss 0.422907 Objective Loss 0.422907 LR 0.001000 Time 0.017253 +2023-10-02 20:31:44,355 - Epoch: [6][ 950/ 1236] Overall Loss 0.423130 Objective Loss 0.423130 LR 0.001000 Time 0.017244 +2023-10-02 20:31:44,519 - Epoch: [6][ 960/ 1236] Overall Loss 0.422853 Objective Loss 0.422853 LR 0.001000 Time 0.017235 +2023-10-02 20:31:44,683 - Epoch: [6][ 970/ 1236] Overall Loss 0.422740 Objective Loss 0.422740 LR 0.001000 Time 0.017226 +2023-10-02 20:31:44,847 - Epoch: [6][ 980/ 1236] Overall Loss 0.422300 Objective Loss 0.422300 LR 0.001000 Time 0.017217 +2023-10-02 20:31:45,011 - Epoch: [6][ 990/ 1236] Overall Loss 0.422311 Objective Loss 0.422311 LR 0.001000 Time 0.017208 +2023-10-02 20:31:45,175 - Epoch: [6][ 1000/ 1236] Overall Loss 0.422407 Objective Loss 0.422407 LR 0.001000 Time 0.017200 +2023-10-02 20:31:45,338 - Epoch: [6][ 1010/ 1236] Overall Loss 0.421863 Objective Loss 0.421863 LR 0.001000 Time 0.017191 +2023-10-02 20:31:45,503 - Epoch: [6][ 1020/ 1236] Overall Loss 0.421769 Objective Loss 0.421769 LR 0.001000 Time 0.017184 +2023-10-02 20:31:45,666 - Epoch: [6][ 1030/ 1236] Overall Loss 0.421550 Objective Loss 0.421550 LR 0.001000 Time 0.017175 +2023-10-02 20:31:45,830 - Epoch: [6][ 1040/ 1236] Overall Loss 0.421543 Objective Loss 0.421543 LR 0.001000 Time 0.017167 +2023-10-02 20:31:45,994 - Epoch: [6][ 1050/ 1236] Overall Loss 0.421499 Objective Loss 0.421499 LR 0.001000 Time 0.017159 +2023-10-02 20:31:46,158 - Epoch: [6][ 1060/ 1236] Overall Loss 0.421277 Objective Loss 0.421277 LR 0.001000 Time 0.017152 +2023-10-02 20:31:46,322 - Epoch: [6][ 1070/ 1236] Overall Loss 0.421273 Objective Loss 0.421273 LR 0.001000 Time 0.017145 +2023-10-02 20:31:46,487 - Epoch: [6][ 1080/ 1236] Overall Loss 0.421317 Objective Loss 0.421317 LR 0.001000 Time 0.017138 +2023-10-02 20:31:46,651 - Epoch: [6][ 1090/ 1236] Overall Loss 0.421279 Objective Loss 0.421279 LR 0.001000 Time 0.017131 +2023-10-02 20:31:46,815 - Epoch: [6][ 1100/ 1236] Overall Loss 0.421079 Objective Loss 0.421079 LR 0.001000 Time 0.017124 +2023-10-02 20:31:46,979 - Epoch: [6][ 1110/ 1236] Overall Loss 0.420782 Objective Loss 0.420782 LR 0.001000 Time 0.017118 +2023-10-02 20:31:47,143 - Epoch: [6][ 1120/ 1236] Overall Loss 0.420384 Objective Loss 0.420384 LR 0.001000 Time 0.017111 +2023-10-02 20:31:47,307 - Epoch: [6][ 1130/ 1236] Overall Loss 0.420272 Objective Loss 0.420272 LR 0.001000 Time 0.017104 +2023-10-02 20:31:47,471 - Epoch: [6][ 1140/ 1236] Overall Loss 0.420175 Objective Loss 0.420175 LR 0.001000 Time 0.017098 +2023-10-02 20:31:47,635 - Epoch: [6][ 1150/ 1236] Overall Loss 0.420046 Objective Loss 0.420046 LR 0.001000 Time 0.017092 +2023-10-02 20:31:47,800 - Epoch: [6][ 1160/ 1236] Overall Loss 0.419758 Objective Loss 0.419758 LR 0.001000 Time 0.017086 +2023-10-02 20:31:47,963 - Epoch: [6][ 1170/ 1236] Overall Loss 0.419930 Objective Loss 0.419930 LR 0.001000 Time 0.017080 +2023-10-02 20:31:48,127 - Epoch: [6][ 1180/ 1236] Overall Loss 0.419734 Objective Loss 0.419734 LR 0.001000 Time 0.017074 +2023-10-02 20:31:48,291 - Epoch: [6][ 1190/ 1236] Overall Loss 0.419614 Objective Loss 0.419614 LR 0.001000 Time 0.017067 +2023-10-02 20:31:48,455 - Epoch: [6][ 1200/ 1236] Overall Loss 0.419558 Objective Loss 0.419558 LR 0.001000 Time 0.017062 +2023-10-02 20:31:48,620 - Epoch: [6][ 1210/ 1236] Overall Loss 0.419682 Objective Loss 0.419682 LR 0.001000 Time 0.017056 +2023-10-02 20:31:48,784 - Epoch: [6][ 1220/ 1236] Overall Loss 0.419752 Objective Loss 0.419752 LR 0.001000 Time 0.017051 +2023-10-02 20:31:48,999 - Epoch: [6][ 1230/ 1236] Overall Loss 0.419658 Objective Loss 0.419658 LR 0.001000 Time 0.017087 +2023-10-02 20:31:49,094 - Epoch: [6][ 1236/ 1236] Overall Loss 0.419703 Objective Loss 0.419703 Top1 78.004073 Top5 98.370672 LR 0.001000 Time 0.017081 +2023-10-02 20:31:49,222 - --- validate (epoch=6)----------- +2023-10-02 20:31:49,222 - 29943 samples (256 per mini-batch) +2023-10-02 20:31:49,663 - Epoch: [6][ 10/ 117] Loss 0.426203 Top1 78.281250 Top5 96.835938 +2023-10-02 20:31:49,765 - Epoch: [6][ 20/ 117] Loss 0.420866 Top1 78.242188 Top5 97.226562 +2023-10-02 20:31:49,866 - Epoch: [6][ 30/ 117] Loss 0.412213 Top1 78.346354 Top5 97.343750 +2023-10-02 20:31:49,966 - Epoch: [6][ 40/ 117] Loss 0.411394 Top1 78.515625 Top5 97.460938 +2023-10-02 20:31:50,066 - Epoch: [6][ 50/ 117] Loss 0.412221 Top1 78.367188 Top5 97.406250 +2023-10-02 20:31:50,169 - Epoch: [6][ 60/ 117] Loss 0.414724 Top1 78.522135 Top5 97.447917 +2023-10-02 20:31:50,270 - Epoch: [6][ 70/ 117] Loss 0.414176 Top1 78.565848 Top5 97.472098 +2023-10-02 20:31:50,371 - Epoch: [6][ 80/ 117] Loss 0.414243 Top1 78.549805 Top5 97.475586 +2023-10-02 20:31:50,472 - Epoch: [6][ 90/ 117] Loss 0.414782 Top1 78.441840 Top5 97.439236 +2023-10-02 20:31:50,572 - Epoch: [6][ 100/ 117] Loss 0.414878 Top1 78.511719 Top5 97.421875 +2023-10-02 20:31:50,681 - Epoch: [6][ 110/ 117] Loss 0.408198 Top1 78.632812 Top5 97.489347 +2023-10-02 20:31:50,740 - Epoch: [6][ 117/ 117] Loss 0.406561 Top1 78.719567 Top5 97.505260 +2023-10-02 20:31:50,854 - ==> Top1: 78.720 Top5: 97.505 Loss: 0.407 + +2023-10-02 20:31:50,855 - ==> Confusion: +[[ 932 0 5 0 5 3 0 0 6 53 2 0 1 2 7 2 6 2 0 0 24] + [ 1 991 3 3 3 73 3 21 6 0 4 0 0 0 2 5 5 0 4 3 4] + [ 6 2 878 31 1 3 65 20 0 0 1 0 5 4 1 6 0 4 6 6 17] + [ 3 1 7 987 0 8 2 0 3 0 12 0 9 4 19 3 1 6 10 4 10] + [ 28 15 3 1 915 20 1 1 1 3 0 0 0 4 8 7 31 0 0 5 7] + [ 2 20 1 8 3 974 5 20 2 3 1 4 9 18 6 0 6 3 2 16 13] + [ 0 4 22 0 1 2 1122 13 0 0 1 0 0 0 0 6 0 1 0 10 9] + [ 4 18 21 3 4 33 6 1026 0 0 3 2 3 0 1 1 0 3 45 36 9] + [ 23 2 0 1 1 1 1 1 968 37 9 0 0 8 20 0 5 6 3 3 0] + [ 171 0 0 1 2 2 1 0 38 831 1 0 0 31 10 0 1 3 2 5 20] + [ 6 1 2 28 0 2 7 6 15 0 933 0 2 10 7 2 4 1 5 10 12] + [ 0 0 0 0 0 25 2 0 0 0 0 874 49 3 0 7 2 30 0 29 14] + [ 0 1 2 6 0 3 2 2 0 0 0 34 933 0 0 13 2 44 0 6 20] + [ 2 0 3 3 2 27 1 0 17 11 7 9 3 984 8 8 3 4 0 12 15] + [ 14 1 0 45 11 2 0 0 31 1 5 0 3 2 961 1 3 5 0 0 16] + [ 3 1 1 4 2 2 3 0 0 0 0 8 12 0 0 1048 9 21 1 13 6] + [ 1 9 1 1 2 10 1 0 2 0 0 3 0 0 2 16 1090 1 0 10 12] + [ 0 0 1 6 0 0 1 0 0 0 0 4 13 0 0 7 1 1001 0 2 2] + [ 3 7 8 32 0 1 3 35 6 0 1 0 7 0 14 1 0 0 933 2 15] + [ 0 0 2 2 1 7 10 9 0 0 1 14 7 0 0 3 1 1 1 1090 3] + [ 190 231 107 155 65 353 61 125 143 81 132 134 445 237 198 111 258 121 159 499 4100]] + +2023-10-02 20:31:50,856 - ==> Best [Top1: 78.720 Top5: 97.505 Sparsity:0.00 Params: 169472 on epoch: 6] +2023-10-02 20:31:50,856 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:31:50,870 - + +2023-10-02 20:31:50,870 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:31:51,945 - Epoch: [7][ 10/ 1236] Overall Loss 0.376847 Objective Loss 0.376847 LR 0.001000 Time 0.107500 +2023-10-02 20:31:52,110 - Epoch: [7][ 20/ 1236] Overall Loss 0.381580 Objective Loss 0.381580 LR 0.001000 Time 0.061988 +2023-10-02 20:31:52,275 - Epoch: [7][ 30/ 1236] Overall Loss 0.374206 Objective Loss 0.374206 LR 0.001000 Time 0.046804 +2023-10-02 20:31:52,440 - Epoch: [7][ 40/ 1236] Overall Loss 0.385426 Objective Loss 0.385426 LR 0.001000 Time 0.039225 +2023-10-02 20:31:52,604 - Epoch: [7][ 50/ 1236] Overall Loss 0.381507 Objective Loss 0.381507 LR 0.001000 Time 0.034655 +2023-10-02 20:31:52,769 - Epoch: [7][ 60/ 1236] Overall Loss 0.386329 Objective Loss 0.386329 LR 0.001000 Time 0.031623 +2023-10-02 20:31:52,933 - Epoch: [7][ 70/ 1236] Overall Loss 0.382800 Objective Loss 0.382800 LR 0.001000 Time 0.029445 +2023-10-02 20:31:53,098 - Epoch: [7][ 80/ 1236] Overall Loss 0.383944 Objective Loss 0.383944 LR 0.001000 Time 0.027822 +2023-10-02 20:31:53,263 - Epoch: [7][ 90/ 1236] Overall Loss 0.387631 Objective Loss 0.387631 LR 0.001000 Time 0.026553 +2023-10-02 20:31:53,428 - Epoch: [7][ 100/ 1236] Overall Loss 0.387902 Objective Loss 0.387902 LR 0.001000 Time 0.025547 +2023-10-02 20:31:53,592 - Epoch: [7][ 110/ 1236] Overall Loss 0.386901 Objective Loss 0.386901 LR 0.001000 Time 0.024716 +2023-10-02 20:31:53,757 - Epoch: [7][ 120/ 1236] Overall Loss 0.386678 Objective Loss 0.386678 LR 0.001000 Time 0.024030 +2023-10-02 20:31:53,922 - Epoch: [7][ 130/ 1236] Overall Loss 0.387011 Objective Loss 0.387011 LR 0.001000 Time 0.023446 +2023-10-02 20:31:54,088 - Epoch: [7][ 140/ 1236] Overall Loss 0.385205 Objective Loss 0.385205 LR 0.001000 Time 0.022952 +2023-10-02 20:31:54,252 - Epoch: [7][ 150/ 1236] Overall Loss 0.385126 Objective Loss 0.385126 LR 0.001000 Time 0.022518 +2023-10-02 20:31:54,418 - Epoch: [7][ 160/ 1236] Overall Loss 0.387319 Objective Loss 0.387319 LR 0.001000 Time 0.022142 +2023-10-02 20:31:54,582 - Epoch: [7][ 170/ 1236] Overall Loss 0.387251 Objective Loss 0.387251 LR 0.001000 Time 0.021806 +2023-10-02 20:31:54,748 - Epoch: [7][ 180/ 1236] Overall Loss 0.387231 Objective Loss 0.387231 LR 0.001000 Time 0.021510 +2023-10-02 20:31:54,912 - Epoch: [7][ 190/ 1236] Overall Loss 0.389329 Objective Loss 0.389329 LR 0.001000 Time 0.021244 +2023-10-02 20:31:55,078 - Epoch: [7][ 200/ 1236] Overall Loss 0.388938 Objective Loss 0.388938 LR 0.001000 Time 0.021008 +2023-10-02 20:31:55,243 - Epoch: [7][ 210/ 1236] Overall Loss 0.388816 Objective Loss 0.388816 LR 0.001000 Time 0.020790 +2023-10-02 20:31:55,408 - Epoch: [7][ 220/ 1236] Overall Loss 0.389918 Objective Loss 0.389918 LR 0.001000 Time 0.020595 +2023-10-02 20:31:55,573 - Epoch: [7][ 230/ 1236] Overall Loss 0.390645 Objective Loss 0.390645 LR 0.001000 Time 0.020416 +2023-10-02 20:31:55,738 - Epoch: [7][ 240/ 1236] Overall Loss 0.389344 Objective Loss 0.389344 LR 0.001000 Time 0.020254 +2023-10-02 20:31:55,903 - Epoch: [7][ 250/ 1236] Overall Loss 0.390180 Objective Loss 0.390180 LR 0.001000 Time 0.020102 +2023-10-02 20:31:56,069 - Epoch: [7][ 260/ 1236] Overall Loss 0.390363 Objective Loss 0.390363 LR 0.001000 Time 0.019964 +2023-10-02 20:31:56,233 - Epoch: [7][ 270/ 1236] Overall Loss 0.389964 Objective Loss 0.389964 LR 0.001000 Time 0.019834 +2023-10-02 20:31:56,399 - Epoch: [7][ 280/ 1236] Overall Loss 0.389776 Objective Loss 0.389776 LR 0.001000 Time 0.019715 +2023-10-02 20:31:56,563 - Epoch: [7][ 290/ 1236] Overall Loss 0.391510 Objective Loss 0.391510 LR 0.001000 Time 0.019601 +2023-10-02 20:31:56,729 - Epoch: [7][ 300/ 1236] Overall Loss 0.390229 Objective Loss 0.390229 LR 0.001000 Time 0.019498 +2023-10-02 20:31:56,894 - Epoch: [7][ 310/ 1236] Overall Loss 0.390425 Objective Loss 0.390425 LR 0.001000 Time 0.019401 +2023-10-02 20:31:57,061 - Epoch: [7][ 320/ 1236] Overall Loss 0.389686 Objective Loss 0.389686 LR 0.001000 Time 0.019315 +2023-10-02 20:31:57,228 - Epoch: [7][ 330/ 1236] Overall Loss 0.390060 Objective Loss 0.390060 LR 0.001000 Time 0.019236 +2023-10-02 20:31:57,394 - Epoch: [7][ 340/ 1236] Overall Loss 0.390131 Objective Loss 0.390131 LR 0.001000 Time 0.019158 +2023-10-02 20:31:57,561 - Epoch: [7][ 350/ 1236] Overall Loss 0.389553 Objective Loss 0.389553 LR 0.001000 Time 0.019086 +2023-10-02 20:31:57,728 - Epoch: [7][ 360/ 1236] Overall Loss 0.388788 Objective Loss 0.388788 LR 0.001000 Time 0.019019 +2023-10-02 20:31:57,893 - Epoch: [7][ 370/ 1236] Overall Loss 0.388882 Objective Loss 0.388882 LR 0.001000 Time 0.018951 +2023-10-02 20:31:58,059 - Epoch: [7][ 380/ 1236] Overall Loss 0.389394 Objective Loss 0.389394 LR 0.001000 Time 0.018888 +2023-10-02 20:31:58,224 - Epoch: [7][ 390/ 1236] Overall Loss 0.388050 Objective Loss 0.388050 LR 0.001000 Time 0.018824 +2023-10-02 20:31:58,389 - Epoch: [7][ 400/ 1236] Overall Loss 0.388163 Objective Loss 0.388163 LR 0.001000 Time 0.018767 +2023-10-02 20:31:58,554 - Epoch: [7][ 410/ 1236] Overall Loss 0.388309 Objective Loss 0.388309 LR 0.001000 Time 0.018710 +2023-10-02 20:31:58,719 - Epoch: [7][ 420/ 1236] Overall Loss 0.388732 Objective Loss 0.388732 LR 0.001000 Time 0.018657 +2023-10-02 20:31:58,884 - Epoch: [7][ 430/ 1236] Overall Loss 0.389124 Objective Loss 0.389124 LR 0.001000 Time 0.018605 +2023-10-02 20:31:59,049 - Epoch: [7][ 440/ 1236] Overall Loss 0.389575 Objective Loss 0.389575 LR 0.001000 Time 0.018558 +2023-10-02 20:31:59,213 - Epoch: [7][ 450/ 1236] Overall Loss 0.390326 Objective Loss 0.390326 LR 0.001000 Time 0.018510 +2023-10-02 20:31:59,379 - Epoch: [7][ 460/ 1236] Overall Loss 0.390359 Objective Loss 0.390359 LR 0.001000 Time 0.018467 +2023-10-02 20:31:59,545 - Epoch: [7][ 470/ 1236] Overall Loss 0.389817 Objective Loss 0.389817 LR 0.001000 Time 0.018425 +2023-10-02 20:31:59,710 - Epoch: [7][ 480/ 1236] Overall Loss 0.390605 Objective Loss 0.390605 LR 0.001000 Time 0.018385 +2023-10-02 20:31:59,875 - Epoch: [7][ 490/ 1236] Overall Loss 0.390418 Objective Loss 0.390418 LR 0.001000 Time 0.018346 +2023-10-02 20:32:00,040 - Epoch: [7][ 500/ 1236] Overall Loss 0.390955 Objective Loss 0.390955 LR 0.001000 Time 0.018309 +2023-10-02 20:32:00,205 - Epoch: [7][ 510/ 1236] Overall Loss 0.391008 Objective Loss 0.391008 LR 0.001000 Time 0.018273 +2023-10-02 20:32:00,371 - Epoch: [7][ 520/ 1236] Overall Loss 0.390294 Objective Loss 0.390294 LR 0.001000 Time 0.018240 +2023-10-02 20:32:00,536 - Epoch: [7][ 530/ 1236] Overall Loss 0.391837 Objective Loss 0.391837 LR 0.001000 Time 0.018206 +2023-10-02 20:32:00,701 - Epoch: [7][ 540/ 1236] Overall Loss 0.391135 Objective Loss 0.391135 LR 0.001000 Time 0.018174 +2023-10-02 20:32:00,866 - Epoch: [7][ 550/ 1236] Overall Loss 0.391462 Objective Loss 0.391462 LR 0.001000 Time 0.018144 +2023-10-02 20:32:01,032 - Epoch: [7][ 560/ 1236] Overall Loss 0.391530 Objective Loss 0.391530 LR 0.001000 Time 0.018116 +2023-10-02 20:32:01,198 - Epoch: [7][ 570/ 1236] Overall Loss 0.391786 Objective Loss 0.391786 LR 0.001000 Time 0.018088 +2023-10-02 20:32:01,363 - Epoch: [7][ 580/ 1236] Overall Loss 0.391723 Objective Loss 0.391723 LR 0.001000 Time 0.018061 +2023-10-02 20:32:01,529 - Epoch: [7][ 590/ 1236] Overall Loss 0.392236 Objective Loss 0.392236 LR 0.001000 Time 0.018034 +2023-10-02 20:32:01,694 - Epoch: [7][ 600/ 1236] Overall Loss 0.392224 Objective Loss 0.392224 LR 0.001000 Time 0.018010 +2023-10-02 20:32:01,860 - Epoch: [7][ 610/ 1236] Overall Loss 0.392266 Objective Loss 0.392266 LR 0.001000 Time 0.017985 +2023-10-02 20:32:02,026 - Epoch: [7][ 620/ 1236] Overall Loss 0.392722 Objective Loss 0.392722 LR 0.001000 Time 0.017962 +2023-10-02 20:32:02,191 - Epoch: [7][ 630/ 1236] Overall Loss 0.392507 Objective Loss 0.392507 LR 0.001000 Time 0.017939 +2023-10-02 20:32:02,357 - Epoch: [7][ 640/ 1236] Overall Loss 0.392482 Objective Loss 0.392482 LR 0.001000 Time 0.017917 +2023-10-02 20:32:02,522 - Epoch: [7][ 650/ 1236] Overall Loss 0.392361 Objective Loss 0.392361 LR 0.001000 Time 0.017894 +2023-10-02 20:32:02,687 - Epoch: [7][ 660/ 1236] Overall Loss 0.392358 Objective Loss 0.392358 LR 0.001000 Time 0.017873 +2023-10-02 20:32:02,852 - Epoch: [7][ 670/ 1236] Overall Loss 0.392790 Objective Loss 0.392790 LR 0.001000 Time 0.017853 +2023-10-02 20:32:03,018 - Epoch: [7][ 680/ 1236] Overall Loss 0.392544 Objective Loss 0.392544 LR 0.001000 Time 0.017834 +2023-10-02 20:32:03,184 - Epoch: [7][ 690/ 1236] Overall Loss 0.392109 Objective Loss 0.392109 LR 0.001000 Time 0.017815 +2023-10-02 20:32:03,350 - Epoch: [7][ 700/ 1236] Overall Loss 0.392035 Objective Loss 0.392035 LR 0.001000 Time 0.017798 +2023-10-02 20:32:03,516 - Epoch: [7][ 710/ 1236] Overall Loss 0.392001 Objective Loss 0.392001 LR 0.001000 Time 0.017780 +2023-10-02 20:32:03,682 - Epoch: [7][ 720/ 1236] Overall Loss 0.392349 Objective Loss 0.392349 LR 0.001000 Time 0.017764 +2023-10-02 20:32:03,848 - Epoch: [7][ 730/ 1236] Overall Loss 0.392369 Objective Loss 0.392369 LR 0.001000 Time 0.017747 +2023-10-02 20:32:04,013 - Epoch: [7][ 740/ 1236] Overall Loss 0.392309 Objective Loss 0.392309 LR 0.001000 Time 0.017730 +2023-10-02 20:32:04,179 - Epoch: [7][ 750/ 1236] Overall Loss 0.392696 Objective Loss 0.392696 LR 0.001000 Time 0.017714 +2023-10-02 20:32:04,345 - Epoch: [7][ 760/ 1236] Overall Loss 0.392697 Objective Loss 0.392697 LR 0.001000 Time 0.017699 +2023-10-02 20:32:04,510 - Epoch: [7][ 770/ 1236] Overall Loss 0.393046 Objective Loss 0.393046 LR 0.001000 Time 0.017684 +2023-10-02 20:32:04,676 - Epoch: [7][ 780/ 1236] Overall Loss 0.393293 Objective Loss 0.393293 LR 0.001000 Time 0.017669 +2023-10-02 20:32:04,841 - Epoch: [7][ 790/ 1236] Overall Loss 0.393435 Objective Loss 0.393435 LR 0.001000 Time 0.017653 +2023-10-02 20:32:05,006 - Epoch: [7][ 800/ 1236] Overall Loss 0.393432 Objective Loss 0.393432 LR 0.001000 Time 0.017639 +2023-10-02 20:32:05,171 - Epoch: [7][ 810/ 1236] Overall Loss 0.393558 Objective Loss 0.393558 LR 0.001000 Time 0.017625 +2023-10-02 20:32:05,337 - Epoch: [7][ 820/ 1236] Overall Loss 0.393927 Objective Loss 0.393927 LR 0.001000 Time 0.017612 +2023-10-02 20:32:05,503 - Epoch: [7][ 830/ 1236] Overall Loss 0.394426 Objective Loss 0.394426 LR 0.001000 Time 0.017599 +2023-10-02 20:32:05,669 - Epoch: [7][ 840/ 1236] Overall Loss 0.394495 Objective Loss 0.394495 LR 0.001000 Time 0.017587 +2023-10-02 20:32:05,834 - Epoch: [7][ 850/ 1236] Overall Loss 0.394164 Objective Loss 0.394164 LR 0.001000 Time 0.017574 +2023-10-02 20:32:06,000 - Epoch: [7][ 860/ 1236] Overall Loss 0.394277 Objective Loss 0.394277 LR 0.001000 Time 0.017562 +2023-10-02 20:32:06,166 - Epoch: [7][ 870/ 1236] Overall Loss 0.394653 Objective Loss 0.394653 LR 0.001000 Time 0.017550 +2023-10-02 20:32:06,332 - Epoch: [7][ 880/ 1236] Overall Loss 0.394963 Objective Loss 0.394963 LR 0.001000 Time 0.017540 +2023-10-02 20:32:06,498 - Epoch: [7][ 890/ 1236] Overall Loss 0.394823 Objective Loss 0.394823 LR 0.001000 Time 0.017528 +2023-10-02 20:32:06,664 - Epoch: [7][ 900/ 1236] Overall Loss 0.394961 Objective Loss 0.394961 LR 0.001000 Time 0.017518 +2023-10-02 20:32:06,830 - Epoch: [7][ 910/ 1236] Overall Loss 0.394877 Objective Loss 0.394877 LR 0.001000 Time 0.017507 +2023-10-02 20:32:06,996 - Epoch: [7][ 920/ 1236] Overall Loss 0.394759 Objective Loss 0.394759 LR 0.001000 Time 0.017497 +2023-10-02 20:32:07,162 - Epoch: [7][ 930/ 1236] Overall Loss 0.394946 Objective Loss 0.394946 LR 0.001000 Time 0.017487 +2023-10-02 20:32:07,328 - Epoch: [7][ 940/ 1236] Overall Loss 0.395060 Objective Loss 0.395060 LR 0.001000 Time 0.017477 +2023-10-02 20:32:07,493 - Epoch: [7][ 950/ 1236] Overall Loss 0.395252 Objective Loss 0.395252 LR 0.001000 Time 0.017467 +2023-10-02 20:32:07,659 - Epoch: [7][ 960/ 1236] Overall Loss 0.395441 Objective Loss 0.395441 LR 0.001000 Time 0.017458 +2023-10-02 20:32:07,825 - Epoch: [7][ 970/ 1236] Overall Loss 0.395109 Objective Loss 0.395109 LR 0.001000 Time 0.017449 +2023-10-02 20:32:07,991 - Epoch: [7][ 980/ 1236] Overall Loss 0.395168 Objective Loss 0.395168 LR 0.001000 Time 0.017440 +2023-10-02 20:32:08,157 - Epoch: [7][ 990/ 1236] Overall Loss 0.394985 Objective Loss 0.394985 LR 0.001000 Time 0.017431 +2023-10-02 20:32:08,324 - Epoch: [7][ 1000/ 1236] Overall Loss 0.394673 Objective Loss 0.394673 LR 0.001000 Time 0.017423 +2023-10-02 20:32:08,490 - Epoch: [7][ 1010/ 1236] Overall Loss 0.394915 Objective Loss 0.394915 LR 0.001000 Time 0.017414 +2023-10-02 20:32:08,656 - Epoch: [7][ 1020/ 1236] Overall Loss 0.395421 Objective Loss 0.395421 LR 0.001000 Time 0.017406 +2023-10-02 20:32:08,822 - Epoch: [7][ 1030/ 1236] Overall Loss 0.395629 Objective Loss 0.395629 LR 0.001000 Time 0.017398 +2023-10-02 20:32:08,988 - Epoch: [7][ 1040/ 1236] Overall Loss 0.395914 Objective Loss 0.395914 LR 0.001000 Time 0.017390 +2023-10-02 20:32:09,155 - Epoch: [7][ 1050/ 1236] Overall Loss 0.396328 Objective Loss 0.396328 LR 0.001000 Time 0.017383 +2023-10-02 20:32:09,321 - Epoch: [7][ 1060/ 1236] Overall Loss 0.396622 Objective Loss 0.396622 LR 0.001000 Time 0.017376 +2023-10-02 20:32:09,487 - Epoch: [7][ 1070/ 1236] Overall Loss 0.396724 Objective Loss 0.396724 LR 0.001000 Time 0.017368 +2023-10-02 20:32:09,654 - Epoch: [7][ 1080/ 1236] Overall Loss 0.396873 Objective Loss 0.396873 LR 0.001000 Time 0.017361 +2023-10-02 20:32:09,819 - Epoch: [7][ 1090/ 1236] Overall Loss 0.396898 Objective Loss 0.396898 LR 0.001000 Time 0.017354 +2023-10-02 20:32:09,985 - Epoch: [7][ 1100/ 1236] Overall Loss 0.397019 Objective Loss 0.397019 LR 0.001000 Time 0.017346 +2023-10-02 20:32:10,151 - Epoch: [7][ 1110/ 1236] Overall Loss 0.396714 Objective Loss 0.396714 LR 0.001000 Time 0.017339 +2023-10-02 20:32:10,317 - Epoch: [7][ 1120/ 1236] Overall Loss 0.397156 Objective Loss 0.397156 LR 0.001000 Time 0.017332 +2023-10-02 20:32:10,483 - Epoch: [7][ 1130/ 1236] Overall Loss 0.397119 Objective Loss 0.397119 LR 0.001000 Time 0.017326 +2023-10-02 20:32:10,650 - Epoch: [7][ 1140/ 1236] Overall Loss 0.397242 Objective Loss 0.397242 LR 0.001000 Time 0.017320 +2023-10-02 20:32:10,816 - Epoch: [7][ 1150/ 1236] Overall Loss 0.397295 Objective Loss 0.397295 LR 0.001000 Time 0.017313 +2023-10-02 20:32:10,983 - Epoch: [7][ 1160/ 1236] Overall Loss 0.397455 Objective Loss 0.397455 LR 0.001000 Time 0.017307 +2023-10-02 20:32:11,149 - Epoch: [7][ 1170/ 1236] Overall Loss 0.397432 Objective Loss 0.397432 LR 0.001000 Time 0.017301 +2023-10-02 20:32:11,316 - Epoch: [7][ 1180/ 1236] Overall Loss 0.397634 Objective Loss 0.397634 LR 0.001000 Time 0.017296 +2023-10-02 20:32:11,482 - Epoch: [7][ 1190/ 1236] Overall Loss 0.397555 Objective Loss 0.397555 LR 0.001000 Time 0.017290 +2023-10-02 20:32:11,649 - Epoch: [7][ 1200/ 1236] Overall Loss 0.397379 Objective Loss 0.397379 LR 0.001000 Time 0.017284 +2023-10-02 20:32:11,815 - Epoch: [7][ 1210/ 1236] Overall Loss 0.397387 Objective Loss 0.397387 LR 0.001000 Time 0.017279 +2023-10-02 20:32:11,982 - Epoch: [7][ 1220/ 1236] Overall Loss 0.397525 Objective Loss 0.397525 LR 0.001000 Time 0.017274 +2023-10-02 20:32:12,198 - Epoch: [7][ 1230/ 1236] Overall Loss 0.397654 Objective Loss 0.397654 LR 0.001000 Time 0.017309 +2023-10-02 20:32:12,293 - Epoch: [7][ 1236/ 1236] Overall Loss 0.397875 Objective Loss 0.397875 Top1 78.411405 Top5 97.556008 LR 0.001000 Time 0.017302 +2023-10-02 20:32:12,430 - --- validate (epoch=7)----------- +2023-10-02 20:32:12,430 - 29943 samples (256 per mini-batch) +2023-10-02 20:32:12,870 - Epoch: [7][ 10/ 117] Loss 0.393176 Top1 77.226562 Top5 97.734375 +2023-10-02 20:32:12,981 - Epoch: [7][ 20/ 117] Loss 0.373721 Top1 77.558594 Top5 97.578125 +2023-10-02 20:32:13,090 - Epoch: [7][ 30/ 117] Loss 0.376481 Top1 77.630208 Top5 97.552083 +2023-10-02 20:32:13,199 - Epoch: [7][ 40/ 117] Loss 0.371076 Top1 77.539062 Top5 97.441406 +2023-10-02 20:32:13,307 - Epoch: [7][ 50/ 117] Loss 0.380264 Top1 77.453125 Top5 97.359375 +2023-10-02 20:32:13,416 - Epoch: [7][ 60/ 117] Loss 0.377469 Top1 77.415365 Top5 97.356771 +2023-10-02 20:32:13,522 - Epoch: [7][ 70/ 117] Loss 0.379291 Top1 77.254464 Top5 97.293527 +2023-10-02 20:32:13,630 - Epoch: [7][ 80/ 117] Loss 0.373884 Top1 77.368164 Top5 97.290039 +2023-10-02 20:32:13,736 - Epoch: [7][ 90/ 117] Loss 0.375949 Top1 77.447917 Top5 97.309028 +2023-10-02 20:32:13,845 - Epoch: [7][ 100/ 117] Loss 0.377397 Top1 77.386719 Top5 97.312500 +2023-10-02 20:32:13,966 - Epoch: [7][ 110/ 117] Loss 0.379044 Top1 77.468040 Top5 97.297585 +2023-10-02 20:32:14,025 - Epoch: [7][ 117/ 117] Loss 0.380469 Top1 77.510603 Top5 97.308219 +2023-10-02 20:32:14,118 - ==> Top1: 77.511 Top5: 97.308 Loss: 0.380 + +2023-10-02 20:32:14,119 - ==> Confusion: +[[ 939 2 5 2 3 2 0 0 2 59 2 1 1 2 4 1 8 2 0 0 15] + [ 1 1056 1 1 2 19 2 18 5 0 0 0 0 0 0 3 7 0 10 3 3] + [ 9 1 927 15 3 0 39 24 0 2 2 2 0 2 0 3 1 3 10 6 7] + [ 3 2 21 991 0 3 1 2 3 0 6 0 5 1 13 0 2 4 20 2 10] + [ 33 13 3 0 942 7 0 0 0 7 0 0 0 0 7 4 23 0 1 7 3] + [ 4 75 4 6 5 903 2 34 3 4 5 3 8 16 7 1 8 2 4 14 8] + [ 0 5 30 1 2 0 1101 19 0 0 2 0 1 0 0 10 1 0 2 13 4] + [ 3 19 24 1 4 25 8 1020 2 0 3 4 0 0 1 0 3 0 69 26 6] + [ 22 2 1 0 2 0 0 1 936 60 11 0 1 9 29 1 3 4 3 4 0] + [ 167 0 0 0 6 1 1 1 17 886 0 0 0 11 11 0 1 0 0 6 11] + [ 10 3 8 20 2 1 4 2 11 0 951 0 0 9 8 0 1 0 8 6 9] + [ 2 0 3 0 0 21 1 2 0 2 0 865 57 5 0 5 5 27 0 31 9] + [ 0 4 2 7 2 4 0 2 0 0 0 34 930 0 2 16 8 39 1 11 6] + [ 2 1 7 0 12 14 0 2 10 26 7 4 2 994 9 0 9 1 0 13 6] + [ 12 4 0 38 18 1 0 0 22 4 3 0 4 0 964 0 4 5 6 0 16] + [ 0 1 4 1 4 0 2 0 0 1 0 8 6 0 0 1055 24 14 1 10 3] + [ 0 20 1 0 7 2 0 0 3 0 1 3 0 0 4 13 1094 0 1 4 8] + [ 1 0 0 3 0 0 1 0 1 0 0 2 12 0 0 4 1 1007 0 2 4] + [ 2 8 6 23 0 1 0 18 6 0 5 0 5 0 6 0 2 0 979 1 6] + [ 0 1 4 1 1 2 8 9 0 0 2 10 4 0 0 4 7 0 2 1096 1] + [ 231 418 200 156 156 214 50 125 110 130 139 114 477 281 230 107 316 119 267 492 3573]] + +2023-10-02 20:32:14,120 - ==> Best [Top1: 78.720 Top5: 97.505 Sparsity:0.00 Params: 169472 on epoch: 6] +2023-10-02 20:32:14,120 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:32:14,126 - + +2023-10-02 20:32:14,126 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:32:15,108 - Epoch: [8][ 10/ 1236] Overall Loss 0.413261 Objective Loss 0.413261 LR 0.001000 Time 0.098111 +2023-10-02 20:32:15,273 - Epoch: [8][ 20/ 1236] Overall Loss 0.403671 Objective Loss 0.403671 LR 0.001000 Time 0.057290 +2023-10-02 20:32:15,436 - Epoch: [8][ 30/ 1236] Overall Loss 0.395845 Objective Loss 0.395845 LR 0.001000 Time 0.043615 +2023-10-02 20:32:15,600 - Epoch: [8][ 40/ 1236] Overall Loss 0.396802 Objective Loss 0.396802 LR 0.001000 Time 0.036809 +2023-10-02 20:32:15,763 - Epoch: [8][ 50/ 1236] Overall Loss 0.392963 Objective Loss 0.392963 LR 0.001000 Time 0.032696 +2023-10-02 20:32:15,927 - Epoch: [8][ 60/ 1236] Overall Loss 0.393314 Objective Loss 0.393314 LR 0.001000 Time 0.029973 +2023-10-02 20:32:16,089 - Epoch: [8][ 70/ 1236] Overall Loss 0.391339 Objective Loss 0.391339 LR 0.001000 Time 0.028005 +2023-10-02 20:32:16,253 - Epoch: [8][ 80/ 1236] Overall Loss 0.386598 Objective Loss 0.386598 LR 0.001000 Time 0.026551 +2023-10-02 20:32:16,416 - Epoch: [8][ 90/ 1236] Overall Loss 0.381514 Objective Loss 0.381514 LR 0.001000 Time 0.025407 +2023-10-02 20:32:16,580 - Epoch: [8][ 100/ 1236] Overall Loss 0.379200 Objective Loss 0.379200 LR 0.001000 Time 0.024500 +2023-10-02 20:32:16,743 - Epoch: [8][ 110/ 1236] Overall Loss 0.377777 Objective Loss 0.377777 LR 0.001000 Time 0.023750 +2023-10-02 20:32:16,906 - Epoch: [8][ 120/ 1236] Overall Loss 0.379261 Objective Loss 0.379261 LR 0.001000 Time 0.023129 +2023-10-02 20:32:17,068 - Epoch: [8][ 130/ 1236] Overall Loss 0.379481 Objective Loss 0.379481 LR 0.001000 Time 0.022595 +2023-10-02 20:32:17,232 - Epoch: [8][ 140/ 1236] Overall Loss 0.379145 Objective Loss 0.379145 LR 0.001000 Time 0.022150 +2023-10-02 20:32:17,394 - Epoch: [8][ 150/ 1236] Overall Loss 0.377805 Objective Loss 0.377805 LR 0.001000 Time 0.021753 +2023-10-02 20:32:17,558 - Epoch: [8][ 160/ 1236] Overall Loss 0.378282 Objective Loss 0.378282 LR 0.001000 Time 0.021414 +2023-10-02 20:32:17,720 - Epoch: [8][ 170/ 1236] Overall Loss 0.380625 Objective Loss 0.380625 LR 0.001000 Time 0.021108 +2023-10-02 20:32:17,884 - Epoch: [8][ 180/ 1236] Overall Loss 0.378236 Objective Loss 0.378236 LR 0.001000 Time 0.020842 +2023-10-02 20:32:18,046 - Epoch: [8][ 190/ 1236] Overall Loss 0.378099 Objective Loss 0.378099 LR 0.001000 Time 0.020599 +2023-10-02 20:32:18,210 - Epoch: [8][ 200/ 1236] Overall Loss 0.377938 Objective Loss 0.377938 LR 0.001000 Time 0.020386 +2023-10-02 20:32:18,372 - Epoch: [8][ 210/ 1236] Overall Loss 0.376126 Objective Loss 0.376126 LR 0.001000 Time 0.020187 +2023-10-02 20:32:18,536 - Epoch: [8][ 220/ 1236] Overall Loss 0.375435 Objective Loss 0.375435 LR 0.001000 Time 0.020012 +2023-10-02 20:32:18,698 - Epoch: [8][ 230/ 1236] Overall Loss 0.374740 Objective Loss 0.374740 LR 0.001000 Time 0.019844 +2023-10-02 20:32:18,862 - Epoch: [8][ 240/ 1236] Overall Loss 0.373406 Objective Loss 0.373406 LR 0.001000 Time 0.019700 +2023-10-02 20:32:19,025 - Epoch: [8][ 250/ 1236] Overall Loss 0.371659 Objective Loss 0.371659 LR 0.001000 Time 0.019560 +2023-10-02 20:32:19,188 - Epoch: [8][ 260/ 1236] Overall Loss 0.371450 Objective Loss 0.371450 LR 0.001000 Time 0.019436 +2023-10-02 20:32:19,351 - Epoch: [8][ 270/ 1236] Overall Loss 0.372385 Objective Loss 0.372385 LR 0.001000 Time 0.019316 +2023-10-02 20:32:19,514 - Epoch: [8][ 280/ 1236] Overall Loss 0.372507 Objective Loss 0.372507 LR 0.001000 Time 0.019210 +2023-10-02 20:32:19,677 - Epoch: [8][ 290/ 1236] Overall Loss 0.373855 Objective Loss 0.373855 LR 0.001000 Time 0.019107 +2023-10-02 20:32:19,840 - Epoch: [8][ 300/ 1236] Overall Loss 0.373214 Objective Loss 0.373214 LR 0.001000 Time 0.019014 +2023-10-02 20:32:20,002 - Epoch: [8][ 310/ 1236] Overall Loss 0.373156 Objective Loss 0.373156 LR 0.001000 Time 0.018923 +2023-10-02 20:32:20,166 - Epoch: [8][ 320/ 1236] Overall Loss 0.374194 Objective Loss 0.374194 LR 0.001000 Time 0.018842 +2023-10-02 20:32:20,329 - Epoch: [8][ 330/ 1236] Overall Loss 0.374709 Objective Loss 0.374709 LR 0.001000 Time 0.018762 +2023-10-02 20:32:20,495 - Epoch: [8][ 340/ 1236] Overall Loss 0.374464 Objective Loss 0.374464 LR 0.001000 Time 0.018700 +2023-10-02 20:32:20,664 - Epoch: [8][ 350/ 1236] Overall Loss 0.375731 Objective Loss 0.375731 LR 0.001000 Time 0.018646 +2023-10-02 20:32:20,830 - Epoch: [8][ 360/ 1236] Overall Loss 0.376311 Objective Loss 0.376311 LR 0.001000 Time 0.018590 +2023-10-02 20:32:20,997 - Epoch: [8][ 370/ 1236] Overall Loss 0.375548 Objective Loss 0.375548 LR 0.001000 Time 0.018539 +2023-10-02 20:32:21,164 - Epoch: [8][ 380/ 1236] Overall Loss 0.374894 Objective Loss 0.374894 LR 0.001000 Time 0.018488 +2023-10-02 20:32:21,332 - Epoch: [8][ 390/ 1236] Overall Loss 0.374490 Objective Loss 0.374490 LR 0.001000 Time 0.018443 +2023-10-02 20:32:21,498 - Epoch: [8][ 400/ 1236] Overall Loss 0.374572 Objective Loss 0.374572 LR 0.001000 Time 0.018396 +2023-10-02 20:32:21,665 - Epoch: [8][ 410/ 1236] Overall Loss 0.375334 Objective Loss 0.375334 LR 0.001000 Time 0.018355 +2023-10-02 20:32:21,831 - Epoch: [8][ 420/ 1236] Overall Loss 0.375623 Objective Loss 0.375623 LR 0.001000 Time 0.018314 +2023-10-02 20:32:21,999 - Epoch: [8][ 430/ 1236] Overall Loss 0.375983 Objective Loss 0.375983 LR 0.001000 Time 0.018276 +2023-10-02 20:32:22,165 - Epoch: [8][ 440/ 1236] Overall Loss 0.375913 Objective Loss 0.375913 LR 0.001000 Time 0.018239 +2023-10-02 20:32:22,333 - Epoch: [8][ 450/ 1236] Overall Loss 0.376987 Objective Loss 0.376987 LR 0.001000 Time 0.018206 +2023-10-02 20:32:22,501 - Epoch: [8][ 460/ 1236] Overall Loss 0.377025 Objective Loss 0.377025 LR 0.001000 Time 0.018174 +2023-10-02 20:32:22,669 - Epoch: [8][ 470/ 1236] Overall Loss 0.377028 Objective Loss 0.377028 LR 0.001000 Time 0.018143 +2023-10-02 20:32:22,836 - Epoch: [8][ 480/ 1236] Overall Loss 0.376414 Objective Loss 0.376414 LR 0.001000 Time 0.018113 +2023-10-02 20:32:23,004 - Epoch: [8][ 490/ 1236] Overall Loss 0.376409 Objective Loss 0.376409 LR 0.001000 Time 0.018086 +2023-10-02 20:32:23,171 - Epoch: [8][ 500/ 1236] Overall Loss 0.375551 Objective Loss 0.375551 LR 0.001000 Time 0.018058 +2023-10-02 20:32:23,339 - Epoch: [8][ 510/ 1236] Overall Loss 0.375689 Objective Loss 0.375689 LR 0.001000 Time 0.018032 +2023-10-02 20:32:23,506 - Epoch: [8][ 520/ 1236] Overall Loss 0.375303 Objective Loss 0.375303 LR 0.001000 Time 0.018005 +2023-10-02 20:32:23,674 - Epoch: [8][ 530/ 1236] Overall Loss 0.374743 Objective Loss 0.374743 LR 0.001000 Time 0.017982 +2023-10-02 20:32:23,842 - Epoch: [8][ 540/ 1236] Overall Loss 0.375082 Objective Loss 0.375082 LR 0.001000 Time 0.017960 +2023-10-02 20:32:24,010 - Epoch: [8][ 550/ 1236] Overall Loss 0.375329 Objective Loss 0.375329 LR 0.001000 Time 0.017938 +2023-10-02 20:32:24,177 - Epoch: [8][ 560/ 1236] Overall Loss 0.375264 Objective Loss 0.375264 LR 0.001000 Time 0.017916 +2023-10-02 20:32:24,345 - Epoch: [8][ 570/ 1236] Overall Loss 0.375067 Objective Loss 0.375067 LR 0.001000 Time 0.017895 +2023-10-02 20:32:24,512 - Epoch: [8][ 580/ 1236] Overall Loss 0.375101 Objective Loss 0.375101 LR 0.001000 Time 0.017874 +2023-10-02 20:32:24,680 - Epoch: [8][ 590/ 1236] Overall Loss 0.375461 Objective Loss 0.375461 LR 0.001000 Time 0.017855 +2023-10-02 20:32:24,847 - Epoch: [8][ 600/ 1236] Overall Loss 0.375335 Objective Loss 0.375335 LR 0.001000 Time 0.017835 +2023-10-02 20:32:25,015 - Epoch: [8][ 610/ 1236] Overall Loss 0.375792 Objective Loss 0.375792 LR 0.001000 Time 0.017818 +2023-10-02 20:32:25,181 - Epoch: [8][ 620/ 1236] Overall Loss 0.376315 Objective Loss 0.376315 LR 0.001000 Time 0.017799 +2023-10-02 20:32:25,349 - Epoch: [8][ 630/ 1236] Overall Loss 0.375923 Objective Loss 0.375923 LR 0.001000 Time 0.017782 +2023-10-02 20:32:25,516 - Epoch: [8][ 640/ 1236] Overall Loss 0.375393 Objective Loss 0.375393 LR 0.001000 Time 0.017765 +2023-10-02 20:32:25,684 - Epoch: [8][ 650/ 1236] Overall Loss 0.375446 Objective Loss 0.375446 LR 0.001000 Time 0.017749 +2023-10-02 20:32:25,849 - Epoch: [8][ 660/ 1236] Overall Loss 0.375819 Objective Loss 0.375819 LR 0.001000 Time 0.017729 +2023-10-02 20:32:26,011 - Epoch: [8][ 670/ 1236] Overall Loss 0.375601 Objective Loss 0.375601 LR 0.001000 Time 0.017706 +2023-10-02 20:32:26,174 - Epoch: [8][ 680/ 1236] Overall Loss 0.375676 Objective Loss 0.375676 LR 0.001000 Time 0.017685 +2023-10-02 20:32:26,337 - Epoch: [8][ 690/ 1236] Overall Loss 0.375591 Objective Loss 0.375591 LR 0.001000 Time 0.017664 +2023-10-02 20:32:26,500 - Epoch: [8][ 700/ 1236] Overall Loss 0.375736 Objective Loss 0.375736 LR 0.001000 Time 0.017644 +2023-10-02 20:32:26,662 - Epoch: [8][ 710/ 1236] Overall Loss 0.375531 Objective Loss 0.375531 LR 0.001000 Time 0.017624 +2023-10-02 20:32:26,826 - Epoch: [8][ 720/ 1236] Overall Loss 0.375336 Objective Loss 0.375336 LR 0.001000 Time 0.017606 +2023-10-02 20:32:26,988 - Epoch: [8][ 730/ 1236] Overall Loss 0.375519 Objective Loss 0.375519 LR 0.001000 Time 0.017587 +2023-10-02 20:32:27,152 - Epoch: [8][ 740/ 1236] Overall Loss 0.375560 Objective Loss 0.375560 LR 0.001000 Time 0.017570 +2023-10-02 20:32:27,314 - Epoch: [8][ 750/ 1236] Overall Loss 0.375644 Objective Loss 0.375644 LR 0.001000 Time 0.017552 +2023-10-02 20:32:27,477 - Epoch: [8][ 760/ 1236] Overall Loss 0.375719 Objective Loss 0.375719 LR 0.001000 Time 0.017535 +2023-10-02 20:32:27,640 - Epoch: [8][ 770/ 1236] Overall Loss 0.375866 Objective Loss 0.375866 LR 0.001000 Time 0.017518 +2023-10-02 20:32:27,803 - Epoch: [8][ 780/ 1236] Overall Loss 0.375803 Objective Loss 0.375803 LR 0.001000 Time 0.017502 +2023-10-02 20:32:27,965 - Epoch: [8][ 790/ 1236] Overall Loss 0.376080 Objective Loss 0.376080 LR 0.001000 Time 0.017486 +2023-10-02 20:32:28,128 - Epoch: [8][ 800/ 1236] Overall Loss 0.375431 Objective Loss 0.375431 LR 0.001000 Time 0.017470 +2023-10-02 20:32:28,291 - Epoch: [8][ 810/ 1236] Overall Loss 0.375494 Objective Loss 0.375494 LR 0.001000 Time 0.017455 +2023-10-02 20:32:28,454 - Epoch: [8][ 820/ 1236] Overall Loss 0.375434 Objective Loss 0.375434 LR 0.001000 Time 0.017441 +2023-10-02 20:32:28,616 - Epoch: [8][ 830/ 1236] Overall Loss 0.375220 Objective Loss 0.375220 LR 0.001000 Time 0.017426 +2023-10-02 20:32:28,780 - Epoch: [8][ 840/ 1236] Overall Loss 0.375031 Objective Loss 0.375031 LR 0.001000 Time 0.017413 +2023-10-02 20:32:28,942 - Epoch: [8][ 850/ 1236] Overall Loss 0.374798 Objective Loss 0.374798 LR 0.001000 Time 0.017398 +2023-10-02 20:32:29,106 - Epoch: [8][ 860/ 1236] Overall Loss 0.374200 Objective Loss 0.374200 LR 0.001000 Time 0.017386 +2023-10-02 20:32:29,268 - Epoch: [8][ 870/ 1236] Overall Loss 0.374539 Objective Loss 0.374539 LR 0.001000 Time 0.017373 +2023-10-02 20:32:29,432 - Epoch: [8][ 880/ 1236] Overall Loss 0.374552 Objective Loss 0.374552 LR 0.001000 Time 0.017361 +2023-10-02 20:32:29,594 - Epoch: [8][ 890/ 1236] Overall Loss 0.374576 Objective Loss 0.374576 LR 0.001000 Time 0.017348 +2023-10-02 20:32:29,758 - Epoch: [8][ 900/ 1236] Overall Loss 0.374739 Objective Loss 0.374739 LR 0.001000 Time 0.017336 +2023-10-02 20:32:29,921 - Epoch: [8][ 910/ 1236] Overall Loss 0.374695 Objective Loss 0.374695 LR 0.001000 Time 0.017325 +2023-10-02 20:32:30,084 - Epoch: [8][ 920/ 1236] Overall Loss 0.374736 Objective Loss 0.374736 LR 0.001000 Time 0.017314 +2023-10-02 20:32:30,246 - Epoch: [8][ 930/ 1236] Overall Loss 0.374556 Objective Loss 0.374556 LR 0.001000 Time 0.017301 +2023-10-02 20:32:30,410 - Epoch: [8][ 940/ 1236] Overall Loss 0.374658 Objective Loss 0.374658 LR 0.001000 Time 0.017291 +2023-10-02 20:32:30,572 - Epoch: [8][ 950/ 1236] Overall Loss 0.374771 Objective Loss 0.374771 LR 0.001000 Time 0.017280 +2023-10-02 20:32:30,736 - Epoch: [8][ 960/ 1236] Overall Loss 0.374472 Objective Loss 0.374472 LR 0.001000 Time 0.017270 +2023-10-02 20:32:30,899 - Epoch: [8][ 970/ 1236] Overall Loss 0.374432 Objective Loss 0.374432 LR 0.001000 Time 0.017260 +2023-10-02 20:32:31,064 - Epoch: [8][ 980/ 1236] Overall Loss 0.374337 Objective Loss 0.374337 LR 0.001000 Time 0.017251 +2023-10-02 20:32:31,227 - Epoch: [8][ 990/ 1236] Overall Loss 0.374104 Objective Loss 0.374104 LR 0.001000 Time 0.017242 +2023-10-02 20:32:31,391 - Epoch: [8][ 1000/ 1236] Overall Loss 0.374109 Objective Loss 0.374109 LR 0.001000 Time 0.017233 +2023-10-02 20:32:31,553 - Epoch: [8][ 1010/ 1236] Overall Loss 0.374099 Objective Loss 0.374099 LR 0.001000 Time 0.017222 +2023-10-02 20:32:31,717 - Epoch: [8][ 1020/ 1236] Overall Loss 0.374041 Objective Loss 0.374041 LR 0.001000 Time 0.017214 +2023-10-02 20:32:31,879 - Epoch: [8][ 1030/ 1236] Overall Loss 0.374414 Objective Loss 0.374414 LR 0.001000 Time 0.017204 +2023-10-02 20:32:32,043 - Epoch: [8][ 1040/ 1236] Overall Loss 0.374554 Objective Loss 0.374554 LR 0.001000 Time 0.017196 +2023-10-02 20:32:32,207 - Epoch: [8][ 1050/ 1236] Overall Loss 0.374695 Objective Loss 0.374695 LR 0.001000 Time 0.017188 +2023-10-02 20:32:32,371 - Epoch: [8][ 1060/ 1236] Overall Loss 0.374639 Objective Loss 0.374639 LR 0.001000 Time 0.017180 +2023-10-02 20:32:32,533 - Epoch: [8][ 1070/ 1236] Overall Loss 0.374827 Objective Loss 0.374827 LR 0.001000 Time 0.017171 +2023-10-02 20:32:32,697 - Epoch: [8][ 1080/ 1236] Overall Loss 0.374968 Objective Loss 0.374968 LR 0.001000 Time 0.017163 +2023-10-02 20:32:32,860 - Epoch: [8][ 1090/ 1236] Overall Loss 0.374965 Objective Loss 0.374965 LR 0.001000 Time 0.017155 +2023-10-02 20:32:33,023 - Epoch: [8][ 1100/ 1236] Overall Loss 0.374916 Objective Loss 0.374916 LR 0.001000 Time 0.017147 +2023-10-02 20:32:33,186 - Epoch: [8][ 1110/ 1236] Overall Loss 0.375510 Objective Loss 0.375510 LR 0.001000 Time 0.017139 +2023-10-02 20:32:33,349 - Epoch: [8][ 1120/ 1236] Overall Loss 0.375408 Objective Loss 0.375408 LR 0.001000 Time 0.017131 +2023-10-02 20:32:33,512 - Epoch: [8][ 1130/ 1236] Overall Loss 0.375539 Objective Loss 0.375539 LR 0.001000 Time 0.017124 +2023-10-02 20:32:33,676 - Epoch: [8][ 1140/ 1236] Overall Loss 0.375333 Objective Loss 0.375333 LR 0.001000 Time 0.017117 +2023-10-02 20:32:33,839 - Epoch: [8][ 1150/ 1236] Overall Loss 0.375063 Objective Loss 0.375063 LR 0.001000 Time 0.017109 +2023-10-02 20:32:34,002 - Epoch: [8][ 1160/ 1236] Overall Loss 0.375109 Objective Loss 0.375109 LR 0.001000 Time 0.017103 +2023-10-02 20:32:34,165 - Epoch: [8][ 1170/ 1236] Overall Loss 0.375035 Objective Loss 0.375035 LR 0.001000 Time 0.017095 +2023-10-02 20:32:34,329 - Epoch: [8][ 1180/ 1236] Overall Loss 0.375347 Objective Loss 0.375347 LR 0.001000 Time 0.017089 +2023-10-02 20:32:34,491 - Epoch: [8][ 1190/ 1236] Overall Loss 0.375235 Objective Loss 0.375235 LR 0.001000 Time 0.017082 +2023-10-02 20:32:34,655 - Epoch: [8][ 1200/ 1236] Overall Loss 0.375250 Objective Loss 0.375250 LR 0.001000 Time 0.017075 +2023-10-02 20:32:34,817 - Epoch: [8][ 1210/ 1236] Overall Loss 0.375435 Objective Loss 0.375435 LR 0.001000 Time 0.017068 +2023-10-02 20:32:34,981 - Epoch: [8][ 1220/ 1236] Overall Loss 0.375715 Objective Loss 0.375715 LR 0.001000 Time 0.017062 +2023-10-02 20:32:35,194 - Epoch: [8][ 1230/ 1236] Overall Loss 0.375978 Objective Loss 0.375978 LR 0.001000 Time 0.017097 +2023-10-02 20:32:35,289 - Epoch: [8][ 1236/ 1236] Overall Loss 0.376142 Objective Loss 0.376142 Top1 78.818737 Top5 96.945010 LR 0.001000 Time 0.017091 +2023-10-02 20:32:35,430 - --- validate (epoch=8)----------- +2023-10-02 20:32:35,430 - 29943 samples (256 per mini-batch) +2023-10-02 20:32:35,868 - Epoch: [8][ 10/ 117] Loss 0.370028 Top1 78.750000 Top5 98.085938 +2023-10-02 20:32:35,972 - Epoch: [8][ 20/ 117] Loss 0.377311 Top1 78.730469 Top5 97.851562 +2023-10-02 20:32:36,074 - Epoch: [8][ 30/ 117] Loss 0.375381 Top1 78.281250 Top5 97.591146 +2023-10-02 20:32:36,176 - Epoch: [8][ 40/ 117] Loss 0.379607 Top1 78.134766 Top5 97.519531 +2023-10-02 20:32:36,277 - Epoch: [8][ 50/ 117] Loss 0.383377 Top1 78.179688 Top5 97.515625 +2023-10-02 20:32:36,379 - Epoch: [8][ 60/ 117] Loss 0.384674 Top1 78.307292 Top5 97.552083 +2023-10-02 20:32:36,480 - Epoch: [8][ 70/ 117] Loss 0.378233 Top1 78.493304 Top5 97.522321 +2023-10-02 20:32:36,582 - Epoch: [8][ 80/ 117] Loss 0.375202 Top1 78.740234 Top5 97.553711 +2023-10-02 20:32:36,683 - Epoch: [8][ 90/ 117] Loss 0.377909 Top1 78.715278 Top5 97.513021 +2023-10-02 20:32:36,787 - Epoch: [8][ 100/ 117] Loss 0.376880 Top1 78.730469 Top5 97.539062 +2023-10-02 20:32:36,902 - Epoch: [8][ 110/ 117] Loss 0.380212 Top1 78.675426 Top5 97.542614 +2023-10-02 20:32:36,961 - Epoch: [8][ 117/ 117] Loss 0.380072 Top1 78.666132 Top5 97.511939 +2023-10-02 20:32:37,054 - ==> Top1: 78.666 Top5: 97.512 Loss: 0.380 + +2023-10-02 20:32:37,054 - ==> Confusion: +[[ 952 1 4 1 4 3 0 0 9 38 1 0 0 5 3 2 2 2 1 0 22] + [ 1 1002 2 2 5 53 3 18 5 0 7 2 0 1 3 2 1 0 19 1 4] + [ 10 0 945 12 2 2 32 13 0 0 2 4 5 3 0 3 0 1 10 2 10] + [ 1 0 14 987 0 7 0 2 6 0 5 0 7 2 25 2 0 4 17 0 10] + [ 35 8 5 0 956 13 0 1 1 2 0 0 2 3 12 3 1 0 2 2 4] + [ 1 22 1 8 1 993 1 32 1 3 3 6 3 12 5 1 3 0 8 5 7] + [ 0 4 51 1 3 0 1093 13 0 0 6 0 1 0 0 9 0 0 1 4 5] + [ 4 18 20 0 3 27 6 1042 1 0 3 3 1 0 1 0 0 1 77 6 5] + [ 25 6 1 0 0 5 0 2 941 31 19 0 3 14 25 4 0 5 4 1 3] + [ 182 1 5 1 11 6 1 1 41 813 4 0 1 23 8 1 0 2 1 3 14] + [ 5 2 9 20 0 1 2 9 14 0 946 2 0 9 9 1 1 0 14 1 8] + [ 0 0 0 0 0 21 2 1 0 0 0 923 41 6 1 3 0 18 1 9 9] + [ 0 0 1 5 0 3 0 4 0 0 1 67 928 2 5 10 0 27 3 7 5] + [ 0 0 3 2 4 25 0 0 4 5 9 12 1 1037 6 0 0 0 0 1 10] + [ 18 0 2 27 10 3 0 0 25 1 4 0 5 0 980 0 0 8 5 0 13] + [ 0 0 7 0 3 0 0 3 0 0 0 12 6 1 0 1064 8 15 1 6 8] + [ 0 10 3 0 11 17 1 1 3 0 1 11 3 2 6 13 1050 1 1 4 23] + [ 0 1 0 2 0 0 3 1 0 0 0 8 28 0 1 9 0 976 0 1 8] + [ 0 2 6 20 0 1 0 25 6 0 4 0 4 0 12 0 1 0 981 0 6] + [ 0 1 3 2 3 1 15 25 0 0 1 24 2 2 0 5 3 0 5 1050 10] + [ 188 245 263 172 132 401 46 140 95 72 156 205 457 328 259 117 60 72 343 258 3896]] + +2023-10-02 20:32:37,056 - ==> Best [Top1: 78.720 Top5: 97.505 Sparsity:0.00 Params: 169472 on epoch: 6] +2023-10-02 20:32:37,056 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:32:37,062 - + +2023-10-02 20:32:37,062 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:32:38,047 - Epoch: [9][ 10/ 1236] Overall Loss 0.365136 Objective Loss 0.365136 LR 0.001000 Time 0.098481 +2023-10-02 20:32:38,215 - Epoch: [9][ 20/ 1236] Overall Loss 0.366859 Objective Loss 0.366859 LR 0.001000 Time 0.057625 +2023-10-02 20:32:38,380 - Epoch: [9][ 30/ 1236] Overall Loss 0.351138 Objective Loss 0.351138 LR 0.001000 Time 0.043915 +2023-10-02 20:32:38,545 - Epoch: [9][ 40/ 1236] Overall Loss 0.343084 Objective Loss 0.343084 LR 0.001000 Time 0.037052 +2023-10-02 20:32:38,710 - Epoch: [9][ 50/ 1236] Overall Loss 0.345223 Objective Loss 0.345223 LR 0.001000 Time 0.032925 +2023-10-02 20:32:38,875 - Epoch: [9][ 60/ 1236] Overall Loss 0.345754 Objective Loss 0.345754 LR 0.001000 Time 0.030192 +2023-10-02 20:32:39,040 - Epoch: [9][ 70/ 1236] Overall Loss 0.347083 Objective Loss 0.347083 LR 0.001000 Time 0.028229 +2023-10-02 20:32:39,205 - Epoch: [9][ 80/ 1236] Overall Loss 0.349764 Objective Loss 0.349764 LR 0.001000 Time 0.026762 +2023-10-02 20:32:39,370 - Epoch: [9][ 90/ 1236] Overall Loss 0.353021 Objective Loss 0.353021 LR 0.001000 Time 0.025614 +2023-10-02 20:32:39,535 - Epoch: [9][ 100/ 1236] Overall Loss 0.356843 Objective Loss 0.356843 LR 0.001000 Time 0.024703 +2023-10-02 20:32:39,700 - Epoch: [9][ 110/ 1236] Overall Loss 0.356749 Objective Loss 0.356749 LR 0.001000 Time 0.023952 +2023-10-02 20:32:39,866 - Epoch: [9][ 120/ 1236] Overall Loss 0.356909 Objective Loss 0.356909 LR 0.001000 Time 0.023332 +2023-10-02 20:32:40,030 - Epoch: [9][ 130/ 1236] Overall Loss 0.358265 Objective Loss 0.358265 LR 0.001000 Time 0.022800 +2023-10-02 20:32:40,195 - Epoch: [9][ 140/ 1236] Overall Loss 0.356772 Objective Loss 0.356772 LR 0.001000 Time 0.022351 +2023-10-02 20:32:40,360 - Epoch: [9][ 150/ 1236] Overall Loss 0.359110 Objective Loss 0.359110 LR 0.001000 Time 0.021957 +2023-10-02 20:32:40,525 - Epoch: [9][ 160/ 1236] Overall Loss 0.359763 Objective Loss 0.359763 LR 0.001000 Time 0.021616 +2023-10-02 20:32:40,690 - Epoch: [9][ 170/ 1236] Overall Loss 0.359693 Objective Loss 0.359693 LR 0.001000 Time 0.021312 +2023-10-02 20:32:40,856 - Epoch: [9][ 180/ 1236] Overall Loss 0.357785 Objective Loss 0.357785 LR 0.001000 Time 0.021044 +2023-10-02 20:32:41,021 - Epoch: [9][ 190/ 1236] Overall Loss 0.358259 Objective Loss 0.358259 LR 0.001000 Time 0.020806 +2023-10-02 20:32:41,187 - Epoch: [9][ 200/ 1236] Overall Loss 0.357867 Objective Loss 0.357867 LR 0.001000 Time 0.020594 +2023-10-02 20:32:41,352 - Epoch: [9][ 210/ 1236] Overall Loss 0.358330 Objective Loss 0.358330 LR 0.001000 Time 0.020400 +2023-10-02 20:32:41,518 - Epoch: [9][ 220/ 1236] Overall Loss 0.358688 Objective Loss 0.358688 LR 0.001000 Time 0.020224 +2023-10-02 20:32:41,683 - Epoch: [9][ 230/ 1236] Overall Loss 0.356813 Objective Loss 0.356813 LR 0.001000 Time 0.020062 +2023-10-02 20:32:41,849 - Epoch: [9][ 240/ 1236] Overall Loss 0.355700 Objective Loss 0.355700 LR 0.001000 Time 0.019917 +2023-10-02 20:32:42,015 - Epoch: [9][ 250/ 1236] Overall Loss 0.355466 Objective Loss 0.355466 LR 0.001000 Time 0.019781 +2023-10-02 20:32:42,180 - Epoch: [9][ 260/ 1236] Overall Loss 0.354245 Objective Loss 0.354245 LR 0.001000 Time 0.019655 +2023-10-02 20:32:42,345 - Epoch: [9][ 270/ 1236] Overall Loss 0.354936 Objective Loss 0.354936 LR 0.001000 Time 0.019537 +2023-10-02 20:32:42,511 - Epoch: [9][ 280/ 1236] Overall Loss 0.354743 Objective Loss 0.354743 LR 0.001000 Time 0.019429 +2023-10-02 20:32:42,675 - Epoch: [9][ 290/ 1236] Overall Loss 0.354843 Objective Loss 0.354843 LR 0.001000 Time 0.019326 +2023-10-02 20:32:42,840 - Epoch: [9][ 300/ 1236] Overall Loss 0.356006 Objective Loss 0.356006 LR 0.001000 Time 0.019231 +2023-10-02 20:32:43,004 - Epoch: [9][ 310/ 1236] Overall Loss 0.356546 Objective Loss 0.356546 LR 0.001000 Time 0.019138 +2023-10-02 20:32:43,170 - Epoch: [9][ 320/ 1236] Overall Loss 0.356790 Objective Loss 0.356790 LR 0.001000 Time 0.019058 +2023-10-02 20:32:43,335 - Epoch: [9][ 330/ 1236] Overall Loss 0.357864 Objective Loss 0.357864 LR 0.001000 Time 0.018979 +2023-10-02 20:32:43,501 - Epoch: [9][ 340/ 1236] Overall Loss 0.358833 Objective Loss 0.358833 LR 0.001000 Time 0.018907 +2023-10-02 20:32:43,665 - Epoch: [9][ 350/ 1236] Overall Loss 0.359136 Objective Loss 0.359136 LR 0.001000 Time 0.018836 +2023-10-02 20:32:43,831 - Epoch: [9][ 360/ 1236] Overall Loss 0.359491 Objective Loss 0.359491 LR 0.001000 Time 0.018772 +2023-10-02 20:32:43,996 - Epoch: [9][ 370/ 1236] Overall Loss 0.359548 Objective Loss 0.359548 LR 0.001000 Time 0.018710 +2023-10-02 20:32:44,161 - Epoch: [9][ 380/ 1236] Overall Loss 0.360193 Objective Loss 0.360193 LR 0.001000 Time 0.018651 +2023-10-02 20:32:44,325 - Epoch: [9][ 390/ 1236] Overall Loss 0.359930 Objective Loss 0.359930 LR 0.001000 Time 0.018592 +2023-10-02 20:32:44,489 - Epoch: [9][ 400/ 1236] Overall Loss 0.359671 Objective Loss 0.359671 LR 0.001000 Time 0.018538 +2023-10-02 20:32:44,654 - Epoch: [9][ 410/ 1236] Overall Loss 0.360740 Objective Loss 0.360740 LR 0.001000 Time 0.018486 +2023-10-02 20:32:44,818 - Epoch: [9][ 420/ 1236] Overall Loss 0.360800 Objective Loss 0.360800 LR 0.001000 Time 0.018437 +2023-10-02 20:32:44,982 - Epoch: [9][ 430/ 1236] Overall Loss 0.360986 Objective Loss 0.360986 LR 0.001000 Time 0.018388 +2023-10-02 20:32:45,146 - Epoch: [9][ 440/ 1236] Overall Loss 0.360681 Objective Loss 0.360681 LR 0.001000 Time 0.018343 +2023-10-02 20:32:45,312 - Epoch: [9][ 450/ 1236] Overall Loss 0.360739 Objective Loss 0.360739 LR 0.001000 Time 0.018302 +2023-10-02 20:32:45,477 - Epoch: [9][ 460/ 1236] Overall Loss 0.359768 Objective Loss 0.359768 LR 0.001000 Time 0.018263 +2023-10-02 20:32:45,642 - Epoch: [9][ 470/ 1236] Overall Loss 0.359832 Objective Loss 0.359832 LR 0.001000 Time 0.018226 +2023-10-02 20:32:45,807 - Epoch: [9][ 480/ 1236] Overall Loss 0.360381 Objective Loss 0.360381 LR 0.001000 Time 0.018189 +2023-10-02 20:32:45,972 - Epoch: [9][ 490/ 1236] Overall Loss 0.360168 Objective Loss 0.360168 LR 0.001000 Time 0.018154 +2023-10-02 20:32:46,138 - Epoch: [9][ 500/ 1236] Overall Loss 0.360575 Objective Loss 0.360575 LR 0.001000 Time 0.018121 +2023-10-02 20:32:46,302 - Epoch: [9][ 510/ 1236] Overall Loss 0.360234 Objective Loss 0.360234 LR 0.001000 Time 0.018088 +2023-10-02 20:32:46,468 - Epoch: [9][ 520/ 1236] Overall Loss 0.359919 Objective Loss 0.359919 LR 0.001000 Time 0.018058 +2023-10-02 20:32:46,632 - Epoch: [9][ 530/ 1236] Overall Loss 0.359803 Objective Loss 0.359803 LR 0.001000 Time 0.018027 +2023-10-02 20:32:46,798 - Epoch: [9][ 540/ 1236] Overall Loss 0.359951 Objective Loss 0.359951 LR 0.001000 Time 0.017999 +2023-10-02 20:32:46,962 - Epoch: [9][ 550/ 1236] Overall Loss 0.359935 Objective Loss 0.359935 LR 0.001000 Time 0.017970 +2023-10-02 20:32:47,128 - Epoch: [9][ 560/ 1236] Overall Loss 0.360205 Objective Loss 0.360205 LR 0.001000 Time 0.017945 +2023-10-02 20:32:47,293 - Epoch: [9][ 570/ 1236] Overall Loss 0.359660 Objective Loss 0.359660 LR 0.001000 Time 0.017919 +2023-10-02 20:32:47,459 - Epoch: [9][ 580/ 1236] Overall Loss 0.359592 Objective Loss 0.359592 LR 0.001000 Time 0.017895 +2023-10-02 20:32:47,623 - Epoch: [9][ 590/ 1236] Overall Loss 0.359902 Objective Loss 0.359902 LR 0.001000 Time 0.017870 +2023-10-02 20:32:47,789 - Epoch: [9][ 600/ 1236] Overall Loss 0.359895 Objective Loss 0.359895 LR 0.001000 Time 0.017847 +2023-10-02 20:32:47,953 - Epoch: [9][ 610/ 1236] Overall Loss 0.359863 Objective Loss 0.359863 LR 0.001000 Time 0.017824 +2023-10-02 20:32:48,119 - Epoch: [9][ 620/ 1236] Overall Loss 0.359786 Objective Loss 0.359786 LR 0.001000 Time 0.017803 +2023-10-02 20:32:48,283 - Epoch: [9][ 630/ 1236] Overall Loss 0.360345 Objective Loss 0.360345 LR 0.001000 Time 0.017781 +2023-10-02 20:32:48,449 - Epoch: [9][ 640/ 1236] Overall Loss 0.360080 Objective Loss 0.360080 LR 0.001000 Time 0.017761 +2023-10-02 20:32:48,613 - Epoch: [9][ 650/ 1236] Overall Loss 0.360073 Objective Loss 0.360073 LR 0.001000 Time 0.017741 +2023-10-02 20:32:48,779 - Epoch: [9][ 660/ 1236] Overall Loss 0.359837 Objective Loss 0.359837 LR 0.001000 Time 0.017722 +2023-10-02 20:32:48,944 - Epoch: [9][ 670/ 1236] Overall Loss 0.359882 Objective Loss 0.359882 LR 0.001000 Time 0.017704 +2023-10-02 20:32:49,109 - Epoch: [9][ 680/ 1236] Overall Loss 0.359631 Objective Loss 0.359631 LR 0.001000 Time 0.017686 +2023-10-02 20:32:49,274 - Epoch: [9][ 690/ 1236] Overall Loss 0.359649 Objective Loss 0.359649 LR 0.001000 Time 0.017668 +2023-10-02 20:32:49,440 - Epoch: [9][ 700/ 1236] Overall Loss 0.359800 Objective Loss 0.359800 LR 0.001000 Time 0.017652 +2023-10-02 20:32:49,605 - Epoch: [9][ 710/ 1236] Overall Loss 0.359842 Objective Loss 0.359842 LR 0.001000 Time 0.017635 +2023-10-02 20:32:49,771 - Epoch: [9][ 720/ 1236] Overall Loss 0.359647 Objective Loss 0.359647 LR 0.001000 Time 0.017621 +2023-10-02 20:32:49,936 - Epoch: [9][ 730/ 1236] Overall Loss 0.360036 Objective Loss 0.360036 LR 0.001000 Time 0.017605 +2023-10-02 20:32:50,101 - Epoch: [9][ 740/ 1236] Overall Loss 0.360095 Objective Loss 0.360095 LR 0.001000 Time 0.017590 +2023-10-02 20:32:50,266 - Epoch: [9][ 750/ 1236] Overall Loss 0.359888 Objective Loss 0.359888 LR 0.001000 Time 0.017575 +2023-10-02 20:32:50,432 - Epoch: [9][ 760/ 1236] Overall Loss 0.359404 Objective Loss 0.359404 LR 0.001000 Time 0.017561 +2023-10-02 20:32:50,597 - Epoch: [9][ 770/ 1236] Overall Loss 0.359359 Objective Loss 0.359359 LR 0.001000 Time 0.017547 +2023-10-02 20:32:50,762 - Epoch: [9][ 780/ 1236] Overall Loss 0.359102 Objective Loss 0.359102 LR 0.001000 Time 0.017534 +2023-10-02 20:32:50,928 - Epoch: [9][ 790/ 1236] Overall Loss 0.358913 Objective Loss 0.358913 LR 0.001000 Time 0.017521 +2023-10-02 20:32:51,093 - Epoch: [9][ 800/ 1236] Overall Loss 0.359393 Objective Loss 0.359393 LR 0.001000 Time 0.017509 +2023-10-02 20:32:51,258 - Epoch: [9][ 810/ 1236] Overall Loss 0.359328 Objective Loss 0.359328 LR 0.001000 Time 0.017496 +2023-10-02 20:32:51,426 - Epoch: [9][ 820/ 1236] Overall Loss 0.359394 Objective Loss 0.359394 LR 0.001000 Time 0.017487 +2023-10-02 20:32:51,591 - Epoch: [9][ 830/ 1236] Overall Loss 0.359384 Objective Loss 0.359384 LR 0.001000 Time 0.017474 +2023-10-02 20:32:51,757 - Epoch: [9][ 840/ 1236] Overall Loss 0.359491 Objective Loss 0.359491 LR 0.001000 Time 0.017464 +2023-10-02 20:32:51,922 - Epoch: [9][ 850/ 1236] Overall Loss 0.359219 Objective Loss 0.359219 LR 0.001000 Time 0.017452 +2023-10-02 20:32:52,088 - Epoch: [9][ 860/ 1236] Overall Loss 0.359281 Objective Loss 0.359281 LR 0.001000 Time 0.017441 +2023-10-02 20:32:52,253 - Epoch: [9][ 870/ 1236] Overall Loss 0.359300 Objective Loss 0.359300 LR 0.001000 Time 0.017431 +2023-10-02 20:32:52,420 - Epoch: [9][ 880/ 1236] Overall Loss 0.359374 Objective Loss 0.359374 LR 0.001000 Time 0.017422 +2023-10-02 20:32:52,586 - Epoch: [9][ 890/ 1236] Overall Loss 0.359189 Objective Loss 0.359189 LR 0.001000 Time 0.017412 +2023-10-02 20:32:52,753 - Epoch: [9][ 900/ 1236] Overall Loss 0.359062 Objective Loss 0.359062 LR 0.001000 Time 0.017404 +2023-10-02 20:32:52,919 - Epoch: [9][ 910/ 1236] Overall Loss 0.359297 Objective Loss 0.359297 LR 0.001000 Time 0.017395 +2023-10-02 20:32:53,085 - Epoch: [9][ 920/ 1236] Overall Loss 0.359225 Objective Loss 0.359225 LR 0.001000 Time 0.017386 +2023-10-02 20:32:53,251 - Epoch: [9][ 930/ 1236] Overall Loss 0.359347 Objective Loss 0.359347 LR 0.001000 Time 0.017377 +2023-10-02 20:32:53,418 - Epoch: [9][ 940/ 1236] Overall Loss 0.359341 Objective Loss 0.359341 LR 0.001000 Time 0.017369 +2023-10-02 20:32:53,583 - Epoch: [9][ 950/ 1236] Overall Loss 0.359148 Objective Loss 0.359148 LR 0.001000 Time 0.017360 +2023-10-02 20:32:53,749 - Epoch: [9][ 960/ 1236] Overall Loss 0.359200 Objective Loss 0.359200 LR 0.001000 Time 0.017352 +2023-10-02 20:32:53,915 - Epoch: [9][ 970/ 1236] Overall Loss 0.359101 Objective Loss 0.359101 LR 0.001000 Time 0.017344 +2023-10-02 20:32:54,081 - Epoch: [9][ 980/ 1236] Overall Loss 0.358976 Objective Loss 0.358976 LR 0.001000 Time 0.017337 +2023-10-02 20:32:54,247 - Epoch: [9][ 990/ 1236] Overall Loss 0.359002 Objective Loss 0.359002 LR 0.001000 Time 0.017328 +2023-10-02 20:32:54,413 - Epoch: [9][ 1000/ 1236] Overall Loss 0.358729 Objective Loss 0.358729 LR 0.001000 Time 0.017321 +2023-10-02 20:32:54,579 - Epoch: [9][ 1010/ 1236] Overall Loss 0.358737 Objective Loss 0.358737 LR 0.001000 Time 0.017313 +2023-10-02 20:32:54,745 - Epoch: [9][ 1020/ 1236] Overall Loss 0.358722 Objective Loss 0.358722 LR 0.001000 Time 0.017306 +2023-10-02 20:32:54,911 - Epoch: [9][ 1030/ 1236] Overall Loss 0.359143 Objective Loss 0.359143 LR 0.001000 Time 0.017299 +2023-10-02 20:32:55,078 - Epoch: [9][ 1040/ 1236] Overall Loss 0.359373 Objective Loss 0.359373 LR 0.001000 Time 0.017292 +2023-10-02 20:32:55,244 - Epoch: [9][ 1050/ 1236] Overall Loss 0.359176 Objective Loss 0.359176 LR 0.001000 Time 0.017286 +2023-10-02 20:32:55,410 - Epoch: [9][ 1060/ 1236] Overall Loss 0.359287 Objective Loss 0.359287 LR 0.001000 Time 0.017279 +2023-10-02 20:32:55,576 - Epoch: [9][ 1070/ 1236] Overall Loss 0.359275 Objective Loss 0.359275 LR 0.001000 Time 0.017273 +2023-10-02 20:32:55,743 - Epoch: [9][ 1080/ 1236] Overall Loss 0.359099 Objective Loss 0.359099 LR 0.001000 Time 0.017267 +2023-10-02 20:32:55,910 - Epoch: [9][ 1090/ 1236] Overall Loss 0.358827 Objective Loss 0.358827 LR 0.001000 Time 0.017261 +2023-10-02 20:32:56,076 - Epoch: [9][ 1100/ 1236] Overall Loss 0.359070 Objective Loss 0.359070 LR 0.001000 Time 0.017256 +2023-10-02 20:32:56,242 - Epoch: [9][ 1110/ 1236] Overall Loss 0.359305 Objective Loss 0.359305 LR 0.001000 Time 0.017249 +2023-10-02 20:32:56,409 - Epoch: [9][ 1120/ 1236] Overall Loss 0.359151 Objective Loss 0.359151 LR 0.001000 Time 0.017244 +2023-10-02 20:32:56,574 - Epoch: [9][ 1130/ 1236] Overall Loss 0.359165 Objective Loss 0.359165 LR 0.001000 Time 0.017237 +2023-10-02 20:32:56,741 - Epoch: [9][ 1140/ 1236] Overall Loss 0.359442 Objective Loss 0.359442 LR 0.001000 Time 0.017232 +2023-10-02 20:32:56,909 - Epoch: [9][ 1150/ 1236] Overall Loss 0.359515 Objective Loss 0.359515 LR 0.001000 Time 0.017228 +2023-10-02 20:32:57,076 - Epoch: [9][ 1160/ 1236] Overall Loss 0.359408 Objective Loss 0.359408 LR 0.001000 Time 0.017223 +2023-10-02 20:32:57,244 - Epoch: [9][ 1170/ 1236] Overall Loss 0.359404 Objective Loss 0.359404 LR 0.001000 Time 0.017219 +2023-10-02 20:32:57,411 - Epoch: [9][ 1180/ 1236] Overall Loss 0.359651 Objective Loss 0.359651 LR 0.001000 Time 0.017214 +2023-10-02 20:32:57,579 - Epoch: [9][ 1190/ 1236] Overall Loss 0.359862 Objective Loss 0.359862 LR 0.001000 Time 0.017211 +2023-10-02 20:32:57,747 - Epoch: [9][ 1200/ 1236] Overall Loss 0.360173 Objective Loss 0.360173 LR 0.001000 Time 0.017207 +2023-10-02 20:32:57,915 - Epoch: [9][ 1210/ 1236] Overall Loss 0.360311 Objective Loss 0.360311 LR 0.001000 Time 0.017203 +2023-10-02 20:32:58,082 - Epoch: [9][ 1220/ 1236] Overall Loss 0.360616 Objective Loss 0.360616 LR 0.001000 Time 0.017199 +2023-10-02 20:32:58,300 - Epoch: [9][ 1230/ 1236] Overall Loss 0.360790 Objective Loss 0.360790 LR 0.001000 Time 0.017236 +2023-10-02 20:32:58,396 - Epoch: [9][ 1236/ 1236] Overall Loss 0.360793 Objective Loss 0.360793 Top1 82.892057 Top5 97.556008 LR 0.001000 Time 0.017229 +2023-10-02 20:32:58,519 - --- validate (epoch=9)----------- +2023-10-02 20:32:58,519 - 29943 samples (256 per mini-batch) +2023-10-02 20:32:58,960 - Epoch: [9][ 10/ 117] Loss 0.368703 Top1 80.507812 Top5 97.382812 +2023-10-02 20:32:59,061 - Epoch: [9][ 20/ 117] Loss 0.368147 Top1 79.687500 Top5 97.500000 +2023-10-02 20:32:59,162 - Epoch: [9][ 30/ 117] Loss 0.357485 Top1 80.182292 Top5 97.760417 +2023-10-02 20:32:59,263 - Epoch: [9][ 40/ 117] Loss 0.357279 Top1 80.185547 Top5 97.744141 +2023-10-02 20:32:59,365 - Epoch: [9][ 50/ 117] Loss 0.350992 Top1 80.210938 Top5 97.843750 +2023-10-02 20:32:59,466 - Epoch: [9][ 60/ 117] Loss 0.354469 Top1 80.240885 Top5 97.825521 +2023-10-02 20:32:59,569 - Epoch: [9][ 70/ 117] Loss 0.353134 Top1 80.435268 Top5 97.823661 +2023-10-02 20:32:59,673 - Epoch: [9][ 80/ 117] Loss 0.353948 Top1 80.415039 Top5 97.792969 +2023-10-02 20:32:59,773 - Epoch: [9][ 90/ 117] Loss 0.351387 Top1 80.412326 Top5 97.738715 +2023-10-02 20:32:59,877 - Epoch: [9][ 100/ 117] Loss 0.352205 Top1 80.406250 Top5 97.718750 +2023-10-02 20:32:59,988 - Epoch: [9][ 110/ 117] Loss 0.351250 Top1 80.433239 Top5 97.759233 +2023-10-02 20:33:00,047 - Epoch: [9][ 117/ 117] Loss 0.348918 Top1 80.456200 Top5 97.792472 +2023-10-02 20:33:00,178 - ==> Top1: 80.456 Top5: 97.792 Loss: 0.349 + +2023-10-02 20:33:00,179 - ==> Confusion: +[[ 928 0 3 0 9 1 1 0 2 76 3 1 2 1 3 3 4 0 0 1 12] + [ 1 1044 1 0 15 18 7 17 2 1 7 0 0 0 3 4 2 0 1 3 5] + [ 6 0 916 3 4 1 69 18 0 4 2 2 6 3 0 4 0 1 4 5 8] + [ 4 3 28 932 0 3 9 2 3 1 16 0 18 1 31 3 2 6 15 1 11] + [ 18 8 0 0 980 5 0 1 0 10 0 0 0 0 9 4 7 0 0 5 3] + [ 3 53 5 3 11 946 4 15 3 5 3 5 3 23 6 3 6 1 3 7 8] + [ 0 1 16 0 0 1 1145 7 0 0 3 0 3 0 0 3 0 0 0 10 2] + [ 5 33 13 0 9 40 3 1050 0 1 2 4 1 0 0 3 1 1 30 15 7] + [ 20 2 0 0 3 1 0 3 924 66 11 0 0 19 21 1 3 1 3 6 5] + [ 95 0 1 0 6 0 1 0 26 939 0 0 0 24 6 0 1 0 0 6 14] + [ 5 2 8 11 0 0 1 3 9 1 983 0 1 8 4 0 3 1 2 5 6] + [ 0 0 3 0 1 20 2 1 0 2 0 914 35 4 0 3 2 14 0 27 7] + [ 0 1 3 4 1 1 1 1 2 1 0 43 956 0 4 10 3 20 4 6 7] + [ 1 0 2 0 6 6 2 0 8 15 11 5 3 1040 2 0 0 1 0 2 15] + [ 8 0 1 14 14 0 0 0 16 8 6 0 2 1 1005 0 1 3 1 1 20] + [ 2 1 1 0 3 0 2 0 0 0 0 13 5 2 0 1064 9 9 1 11 11] + [ 0 23 0 0 8 4 6 1 1 0 1 4 1 1 4 18 1071 0 1 5 12] + [ 0 0 2 1 0 0 3 0 0 0 0 8 15 0 0 12 1 988 0 3 5] + [ 1 12 7 17 1 1 1 35 8 0 9 0 5 0 14 2 2 0 942 0 11] + [ 0 2 2 1 2 3 14 13 0 0 3 8 7 0 0 4 4 0 1 1082 6] + [ 204 304 158 49 168 205 112 115 91 116 225 149 483 271 199 139 115 71 170 319 4242]] + +2023-10-02 20:33:00,180 - ==> Best [Top1: 80.456 Top5: 97.792 Sparsity:0.00 Params: 169472 on epoch: 9] +2023-10-02 20:33:00,180 - Saving checkpoint to: logs/2023.10.02-195228/checkpoint.pth.tar +2023-10-02 20:33:00,194 - + +2023-10-02 20:33:00,194 - Initiating quantization aware training (QAT)... +2023-10-02 20:33:00,206 - + +2023-10-02 20:33:00,206 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:33:01,214 - Epoch: [10][ 10/ 1236] Overall Loss 0.361530 Objective Loss 0.361530 LR 0.001000 Time 0.100821 +2023-10-02 20:33:01,425 - Epoch: [10][ 20/ 1236] Overall Loss 0.341313 Objective Loss 0.341313 LR 0.001000 Time 0.060948 +2023-10-02 20:33:01,635 - Epoch: [10][ 30/ 1236] Overall Loss 0.342815 Objective Loss 0.342815 LR 0.001000 Time 0.047615 +2023-10-02 20:33:01,847 - Epoch: [10][ 40/ 1236] Overall Loss 0.353541 Objective Loss 0.353541 LR 0.001000 Time 0.040983 +2023-10-02 20:33:02,056 - Epoch: [10][ 50/ 1236] Overall Loss 0.365221 Objective Loss 0.365221 LR 0.001000 Time 0.036950 +2023-10-02 20:33:02,267 - Epoch: [10][ 60/ 1236] Overall Loss 0.366251 Objective Loss 0.366251 LR 0.001000 Time 0.034303 +2023-10-02 20:33:02,477 - Epoch: [10][ 70/ 1236] Overall Loss 0.364868 Objective Loss 0.364868 LR 0.001000 Time 0.032372 +2023-10-02 20:33:02,687 - Epoch: [10][ 80/ 1236] Overall Loss 0.364388 Objective Loss 0.364388 LR 0.001000 Time 0.030944 +2023-10-02 20:33:02,896 - Epoch: [10][ 90/ 1236] Overall Loss 0.365296 Objective Loss 0.365296 LR 0.001000 Time 0.029824 +2023-10-02 20:33:03,106 - Epoch: [10][ 100/ 1236] Overall Loss 0.362133 Objective Loss 0.362133 LR 0.001000 Time 0.028943 +2023-10-02 20:33:03,316 - Epoch: [10][ 110/ 1236] Overall Loss 0.365629 Objective Loss 0.365629 LR 0.001000 Time 0.028212 +2023-10-02 20:33:03,526 - Epoch: [10][ 120/ 1236] Overall Loss 0.366205 Objective Loss 0.366205 LR 0.001000 Time 0.027607 +2023-10-02 20:33:03,735 - Epoch: [10][ 130/ 1236] Overall Loss 0.364943 Objective Loss 0.364943 LR 0.001000 Time 0.027088 +2023-10-02 20:33:03,945 - Epoch: [10][ 140/ 1236] Overall Loss 0.365581 Objective Loss 0.365581 LR 0.001000 Time 0.026653 +2023-10-02 20:33:04,155 - Epoch: [10][ 150/ 1236] Overall Loss 0.364886 Objective Loss 0.364886 LR 0.001000 Time 0.026271 +2023-10-02 20:33:04,366 - Epoch: [10][ 160/ 1236] Overall Loss 0.365849 Objective Loss 0.365849 LR 0.001000 Time 0.025948 +2023-10-02 20:33:04,577 - Epoch: [10][ 170/ 1236] Overall Loss 0.365424 Objective Loss 0.365424 LR 0.001000 Time 0.025654 +2023-10-02 20:33:04,789 - Epoch: [10][ 180/ 1236] Overall Loss 0.365639 Objective Loss 0.365639 LR 0.001000 Time 0.025403 +2023-10-02 20:33:05,000 - Epoch: [10][ 190/ 1236] Overall Loss 0.365441 Objective Loss 0.365441 LR 0.001000 Time 0.025169 +2023-10-02 20:33:05,212 - Epoch: [10][ 200/ 1236] Overall Loss 0.366055 Objective Loss 0.366055 LR 0.001000 Time 0.024968 +2023-10-02 20:33:05,423 - Epoch: [10][ 210/ 1236] Overall Loss 0.365337 Objective Loss 0.365337 LR 0.001000 Time 0.024777 +2023-10-02 20:33:05,635 - Epoch: [10][ 220/ 1236] Overall Loss 0.366665 Objective Loss 0.366665 LR 0.001000 Time 0.024613 +2023-10-02 20:33:05,845 - Epoch: [10][ 230/ 1236] Overall Loss 0.365149 Objective Loss 0.365149 LR 0.001000 Time 0.024452 +2023-10-02 20:33:06,057 - Epoch: [10][ 240/ 1236] Overall Loss 0.365598 Objective Loss 0.365598 LR 0.001000 Time 0.024314 +2023-10-02 20:33:06,268 - Epoch: [10][ 250/ 1236] Overall Loss 0.365337 Objective Loss 0.365337 LR 0.001000 Time 0.024178 +2023-10-02 20:33:06,480 - Epoch: [10][ 260/ 1236] Overall Loss 0.364886 Objective Loss 0.364886 LR 0.001000 Time 0.024061 +2023-10-02 20:33:06,690 - Epoch: [10][ 270/ 1236] Overall Loss 0.364299 Objective Loss 0.364299 LR 0.001000 Time 0.023945 +2023-10-02 20:33:06,902 - Epoch: [10][ 280/ 1236] Overall Loss 0.365772 Objective Loss 0.365772 LR 0.001000 Time 0.023845 +2023-10-02 20:33:07,113 - Epoch: [10][ 290/ 1236] Overall Loss 0.365124 Objective Loss 0.365124 LR 0.001000 Time 0.023745 +2023-10-02 20:33:07,326 - Epoch: [10][ 300/ 1236] Overall Loss 0.365367 Objective Loss 0.365367 LR 0.001000 Time 0.023660 +2023-10-02 20:33:07,537 - Epoch: [10][ 310/ 1236] Overall Loss 0.365172 Objective Loss 0.365172 LR 0.001000 Time 0.023573 +2023-10-02 20:33:07,750 - Epoch: [10][ 320/ 1236] Overall Loss 0.365461 Objective Loss 0.365461 LR 0.001000 Time 0.023503 +2023-10-02 20:33:07,961 - Epoch: [10][ 330/ 1236] Overall Loss 0.365152 Objective Loss 0.365152 LR 0.001000 Time 0.023427 +2023-10-02 20:33:08,172 - Epoch: [10][ 340/ 1236] Overall Loss 0.365020 Objective Loss 0.365020 LR 0.001000 Time 0.023359 +2023-10-02 20:33:08,381 - Epoch: [10][ 350/ 1236] Overall Loss 0.364450 Objective Loss 0.364450 LR 0.001000 Time 0.023287 +2023-10-02 20:33:08,591 - Epoch: [10][ 360/ 1236] Overall Loss 0.366142 Objective Loss 0.366142 LR 0.001000 Time 0.023223 +2023-10-02 20:33:08,801 - Epoch: [10][ 370/ 1236] Overall Loss 0.366244 Objective Loss 0.366244 LR 0.001000 Time 0.023160 +2023-10-02 20:33:09,011 - Epoch: [10][ 380/ 1236] Overall Loss 0.365923 Objective Loss 0.365923 LR 0.001000 Time 0.023104 +2023-10-02 20:33:09,221 - Epoch: [10][ 390/ 1236] Overall Loss 0.366244 Objective Loss 0.366244 LR 0.001000 Time 0.023048 +2023-10-02 20:33:09,431 - Epoch: [10][ 400/ 1236] Overall Loss 0.366538 Objective Loss 0.366538 LR 0.001000 Time 0.022997 +2023-10-02 20:33:09,641 - Epoch: [10][ 410/ 1236] Overall Loss 0.367402 Objective Loss 0.367402 LR 0.001000 Time 0.022946 +2023-10-02 20:33:09,851 - Epoch: [10][ 420/ 1236] Overall Loss 0.367622 Objective Loss 0.367622 LR 0.001000 Time 0.022899 +2023-10-02 20:33:10,061 - Epoch: [10][ 430/ 1236] Overall Loss 0.367582 Objective Loss 0.367582 LR 0.001000 Time 0.022853 +2023-10-02 20:33:10,271 - Epoch: [10][ 440/ 1236] Overall Loss 0.368058 Objective Loss 0.368058 LR 0.001000 Time 0.022811 +2023-10-02 20:33:10,481 - Epoch: [10][ 450/ 1236] Overall Loss 0.368946 Objective Loss 0.368946 LR 0.001000 Time 0.022769 +2023-10-02 20:33:10,691 - Epoch: [10][ 460/ 1236] Overall Loss 0.368865 Objective Loss 0.368865 LR 0.001000 Time 0.022731 +2023-10-02 20:33:10,901 - Epoch: [10][ 470/ 1236] Overall Loss 0.369531 Objective Loss 0.369531 LR 0.001000 Time 0.022692 +2023-10-02 20:33:11,111 - Epoch: [10][ 480/ 1236] Overall Loss 0.369673 Objective Loss 0.369673 LR 0.001000 Time 0.022657 +2023-10-02 20:33:11,321 - Epoch: [10][ 490/ 1236] Overall Loss 0.369669 Objective Loss 0.369669 LR 0.001000 Time 0.022621 +2023-10-02 20:33:11,531 - Epoch: [10][ 500/ 1236] Overall Loss 0.369155 Objective Loss 0.369155 LR 0.001000 Time 0.022589 +2023-10-02 20:33:11,741 - Epoch: [10][ 510/ 1236] Overall Loss 0.369058 Objective Loss 0.369058 LR 0.001000 Time 0.022556 +2023-10-02 20:33:11,951 - Epoch: [10][ 520/ 1236] Overall Loss 0.368927 Objective Loss 0.368927 LR 0.001000 Time 0.022526 +2023-10-02 20:33:12,161 - Epoch: [10][ 530/ 1236] Overall Loss 0.369387 Objective Loss 0.369387 LR 0.001000 Time 0.022496 +2023-10-02 20:33:12,371 - Epoch: [10][ 540/ 1236] Overall Loss 0.369502 Objective Loss 0.369502 LR 0.001000 Time 0.022468 +2023-10-02 20:33:12,581 - Epoch: [10][ 550/ 1236] Overall Loss 0.369371 Objective Loss 0.369371 LR 0.001000 Time 0.022440 +2023-10-02 20:33:12,791 - Epoch: [10][ 560/ 1236] Overall Loss 0.369310 Objective Loss 0.369310 LR 0.001000 Time 0.022414 +2023-10-02 20:33:13,001 - Epoch: [10][ 570/ 1236] Overall Loss 0.369256 Objective Loss 0.369256 LR 0.001000 Time 0.022388 +2023-10-02 20:33:13,211 - Epoch: [10][ 580/ 1236] Overall Loss 0.369549 Objective Loss 0.369549 LR 0.001000 Time 0.022364 +2023-10-02 20:33:13,421 - Epoch: [10][ 590/ 1236] Overall Loss 0.368946 Objective Loss 0.368946 LR 0.001000 Time 0.022340 +2023-10-02 20:33:13,631 - Epoch: [10][ 600/ 1236] Overall Loss 0.368836 Objective Loss 0.368836 LR 0.001000 Time 0.022317 +2023-10-02 20:33:13,841 - Epoch: [10][ 610/ 1236] Overall Loss 0.368701 Objective Loss 0.368701 LR 0.001000 Time 0.022294 +2023-10-02 20:33:14,051 - Epoch: [10][ 620/ 1236] Overall Loss 0.368205 Objective Loss 0.368205 LR 0.001000 Time 0.022274 +2023-10-02 20:33:14,261 - Epoch: [10][ 630/ 1236] Overall Loss 0.368031 Objective Loss 0.368031 LR 0.001000 Time 0.022252 +2023-10-02 20:33:14,471 - Epoch: [10][ 640/ 1236] Overall Loss 0.368153 Objective Loss 0.368153 LR 0.001000 Time 0.022233 +2023-10-02 20:33:14,681 - Epoch: [10][ 650/ 1236] Overall Loss 0.368320 Objective Loss 0.368320 LR 0.001000 Time 0.022212 +2023-10-02 20:33:14,891 - Epoch: [10][ 660/ 1236] Overall Loss 0.368495 Objective Loss 0.368495 LR 0.001000 Time 0.022194 +2023-10-02 20:33:15,101 - Epoch: [10][ 670/ 1236] Overall Loss 0.368656 Objective Loss 0.368656 LR 0.001000 Time 0.022175 +2023-10-02 20:33:15,311 - Epoch: [10][ 680/ 1236] Overall Loss 0.368609 Objective Loss 0.368609 LR 0.001000 Time 0.022158 +2023-10-02 20:33:15,521 - Epoch: [10][ 690/ 1236] Overall Loss 0.368762 Objective Loss 0.368762 LR 0.001000 Time 0.022140 +2023-10-02 20:33:15,731 - Epoch: [10][ 700/ 1236] Overall Loss 0.368879 Objective Loss 0.368879 LR 0.001000 Time 0.022123 +2023-10-02 20:33:15,941 - Epoch: [10][ 710/ 1236] Overall Loss 0.368628 Objective Loss 0.368628 LR 0.001000 Time 0.022106 +2023-10-02 20:33:16,151 - Epoch: [10][ 720/ 1236] Overall Loss 0.368761 Objective Loss 0.368761 LR 0.001000 Time 0.022091 +2023-10-02 20:33:16,361 - Epoch: [10][ 730/ 1236] Overall Loss 0.368894 Objective Loss 0.368894 LR 0.001000 Time 0.022075 +2023-10-02 20:33:16,571 - Epoch: [10][ 740/ 1236] Overall Loss 0.368849 Objective Loss 0.368849 LR 0.001000 Time 0.022060 +2023-10-02 20:33:16,781 - Epoch: [10][ 750/ 1236] Overall Loss 0.368900 Objective Loss 0.368900 LR 0.001000 Time 0.022045 +2023-10-02 20:33:16,991 - Epoch: [10][ 760/ 1236] Overall Loss 0.368595 Objective Loss 0.368595 LR 0.001000 Time 0.022031 +2023-10-02 20:33:17,201 - Epoch: [10][ 770/ 1236] Overall Loss 0.369150 Objective Loss 0.369150 LR 0.001000 Time 0.022017 +2023-10-02 20:33:17,411 - Epoch: [10][ 780/ 1236] Overall Loss 0.369411 Objective Loss 0.369411 LR 0.001000 Time 0.022004 +2023-10-02 20:33:17,621 - Epoch: [10][ 790/ 1236] Overall Loss 0.369279 Objective Loss 0.369279 LR 0.001000 Time 0.021990 +2023-10-02 20:33:17,831 - Epoch: [10][ 800/ 1236] Overall Loss 0.369507 Objective Loss 0.369507 LR 0.001000 Time 0.021978 +2023-10-02 20:33:18,041 - Epoch: [10][ 810/ 1236] Overall Loss 0.369075 Objective Loss 0.369075 LR 0.001000 Time 0.021965 +2023-10-02 20:33:18,251 - Epoch: [10][ 820/ 1236] Overall Loss 0.369316 Objective Loss 0.369316 LR 0.001000 Time 0.021953 +2023-10-02 20:33:18,461 - Epoch: [10][ 830/ 1236] Overall Loss 0.368807 Objective Loss 0.368807 LR 0.001000 Time 0.021941 +2023-10-02 20:33:18,671 - Epoch: [10][ 840/ 1236] Overall Loss 0.368638 Objective Loss 0.368638 LR 0.001000 Time 0.021929 +2023-10-02 20:33:18,881 - Epoch: [10][ 850/ 1236] Overall Loss 0.368352 Objective Loss 0.368352 LR 0.001000 Time 0.021917 +2023-10-02 20:33:19,091 - Epoch: [10][ 860/ 1236] Overall Loss 0.368819 Objective Loss 0.368819 LR 0.001000 Time 0.021907 +2023-10-02 20:33:19,301 - Epoch: [10][ 870/ 1236] Overall Loss 0.369061 Objective Loss 0.369061 LR 0.001000 Time 0.021895 +2023-10-02 20:33:19,511 - Epoch: [10][ 880/ 1236] Overall Loss 0.368463 Objective Loss 0.368463 LR 0.001000 Time 0.021885 +2023-10-02 20:33:19,721 - Epoch: [10][ 890/ 1236] Overall Loss 0.368933 Objective Loss 0.368933 LR 0.001000 Time 0.021874 +2023-10-02 20:33:19,931 - Epoch: [10][ 900/ 1236] Overall Loss 0.368688 Objective Loss 0.368688 LR 0.001000 Time 0.021865 +2023-10-02 20:33:20,141 - Epoch: [10][ 910/ 1236] Overall Loss 0.368437 Objective Loss 0.368437 LR 0.001000 Time 0.021854 +2023-10-02 20:33:20,351 - Epoch: [10][ 920/ 1236] Overall Loss 0.368423 Objective Loss 0.368423 LR 0.001000 Time 0.021844 +2023-10-02 20:33:20,560 - Epoch: [10][ 930/ 1236] Overall Loss 0.368269 Objective Loss 0.368269 LR 0.001000 Time 0.021834 +2023-10-02 20:33:20,771 - Epoch: [10][ 940/ 1236] Overall Loss 0.368179 Objective Loss 0.368179 LR 0.001000 Time 0.021825 +2023-10-02 20:33:20,980 - Epoch: [10][ 950/ 1236] Overall Loss 0.367827 Objective Loss 0.367827 LR 0.001000 Time 0.021815 +2023-10-02 20:33:21,191 - Epoch: [10][ 960/ 1236] Overall Loss 0.368009 Objective Loss 0.368009 LR 0.001000 Time 0.021807 +2023-10-02 20:33:21,400 - Epoch: [10][ 970/ 1236] Overall Loss 0.367992 Objective Loss 0.367992 LR 0.001000 Time 0.021798 +2023-10-02 20:33:21,610 - Epoch: [10][ 980/ 1236] Overall Loss 0.368012 Objective Loss 0.368012 LR 0.001000 Time 0.021790 +2023-10-02 20:33:21,820 - Epoch: [10][ 990/ 1236] Overall Loss 0.367857 Objective Loss 0.367857 LR 0.001000 Time 0.021781 +2023-10-02 20:33:22,030 - Epoch: [10][ 1000/ 1236] Overall Loss 0.367979 Objective Loss 0.367979 LR 0.001000 Time 0.021773 +2023-10-02 20:33:22,240 - Epoch: [10][ 1010/ 1236] Overall Loss 0.368042 Objective Loss 0.368042 LR 0.001000 Time 0.021764 +2023-10-02 20:33:22,450 - Epoch: [10][ 1020/ 1236] Overall Loss 0.368059 Objective Loss 0.368059 LR 0.001000 Time 0.021757 +2023-10-02 20:33:22,660 - Epoch: [10][ 1030/ 1236] Overall Loss 0.368360 Objective Loss 0.368360 LR 0.001000 Time 0.021749 +2023-10-02 20:33:22,871 - Epoch: [10][ 1040/ 1236] Overall Loss 0.368209 Objective Loss 0.368209 LR 0.001000 Time 0.021742 +2023-10-02 20:33:23,080 - Epoch: [10][ 1050/ 1236] Overall Loss 0.368285 Objective Loss 0.368285 LR 0.001000 Time 0.021734 +2023-10-02 20:33:23,290 - Epoch: [10][ 1060/ 1236] Overall Loss 0.367903 Objective Loss 0.367903 LR 0.001000 Time 0.021727 +2023-10-02 20:33:23,500 - Epoch: [10][ 1070/ 1236] Overall Loss 0.367826 Objective Loss 0.367826 LR 0.001000 Time 0.021719 +2023-10-02 20:33:23,710 - Epoch: [10][ 1080/ 1236] Overall Loss 0.368399 Objective Loss 0.368399 LR 0.001000 Time 0.021713 +2023-10-02 20:33:23,920 - Epoch: [10][ 1090/ 1236] Overall Loss 0.368308 Objective Loss 0.368308 LR 0.001000 Time 0.021705 +2023-10-02 20:33:24,131 - Epoch: [10][ 1100/ 1236] Overall Loss 0.368406 Objective Loss 0.368406 LR 0.001000 Time 0.021699 +2023-10-02 20:33:24,340 - Epoch: [10][ 1110/ 1236] Overall Loss 0.368265 Objective Loss 0.368265 LR 0.001000 Time 0.021692 +2023-10-02 20:33:24,550 - Epoch: [10][ 1120/ 1236] Overall Loss 0.368033 Objective Loss 0.368033 LR 0.001000 Time 0.021685 +2023-10-02 20:33:24,758 - Epoch: [10][ 1130/ 1236] Overall Loss 0.367890 Objective Loss 0.367890 LR 0.001000 Time 0.021677 +2023-10-02 20:33:24,968 - Epoch: [10][ 1140/ 1236] Overall Loss 0.367733 Objective Loss 0.367733 LR 0.001000 Time 0.021671 +2023-10-02 20:33:25,176 - Epoch: [10][ 1150/ 1236] Overall Loss 0.367689 Objective Loss 0.367689 LR 0.001000 Time 0.021663 +2023-10-02 20:33:25,386 - Epoch: [10][ 1160/ 1236] Overall Loss 0.367826 Objective Loss 0.367826 LR 0.001000 Time 0.021657 +2023-10-02 20:33:25,595 - Epoch: [10][ 1170/ 1236] Overall Loss 0.368029 Objective Loss 0.368029 LR 0.001000 Time 0.021650 +2023-10-02 20:33:25,805 - Epoch: [10][ 1180/ 1236] Overall Loss 0.367913 Objective Loss 0.367913 LR 0.001000 Time 0.021644 +2023-10-02 20:33:26,013 - Epoch: [10][ 1190/ 1236] Overall Loss 0.367759 Objective Loss 0.367759 LR 0.001000 Time 0.021637 +2023-10-02 20:33:26,224 - Epoch: [10][ 1200/ 1236] Overall Loss 0.367711 Objective Loss 0.367711 LR 0.001000 Time 0.021632 +2023-10-02 20:33:26,432 - Epoch: [10][ 1210/ 1236] Overall Loss 0.367608 Objective Loss 0.367608 LR 0.001000 Time 0.021625 +2023-10-02 20:33:26,642 - Epoch: [10][ 1220/ 1236] Overall Loss 0.367264 Objective Loss 0.367264 LR 0.001000 Time 0.021619 +2023-10-02 20:33:26,906 - Epoch: [10][ 1230/ 1236] Overall Loss 0.367216 Objective Loss 0.367216 LR 0.001000 Time 0.021658 +2023-10-02 20:33:27,029 - Epoch: [10][ 1236/ 1236] Overall Loss 0.367147 Objective Loss 0.367147 Top1 81.670061 Top5 97.963340 LR 0.001000 Time 0.021652 +2023-10-02 20:33:27,174 - --- validate (epoch=10)----------- +2023-10-02 20:33:27,175 - 29943 samples (256 per mini-batch) +2023-10-02 20:33:27,650 - Epoch: [10][ 10/ 117] Loss 0.393310 Top1 78.515625 Top5 97.304688 +2023-10-02 20:33:27,801 - Epoch: [10][ 20/ 117] Loss 0.389408 Top1 78.281250 Top5 97.128906 +2023-10-02 20:33:27,952 - Epoch: [10][ 30/ 117] Loss 0.382660 Top1 78.619792 Top5 97.252604 +2023-10-02 20:33:28,102 - Epoch: [10][ 40/ 117] Loss 0.378430 Top1 78.603516 Top5 97.216797 +2023-10-02 20:33:28,252 - Epoch: [10][ 50/ 117] Loss 0.376696 Top1 78.789062 Top5 97.296875 +2023-10-02 20:33:28,403 - Epoch: [10][ 60/ 117] Loss 0.371494 Top1 78.925781 Top5 97.285156 +2023-10-02 20:33:28,553 - Epoch: [10][ 70/ 117] Loss 0.370364 Top1 78.922991 Top5 97.315848 +2023-10-02 20:33:28,703 - Epoch: [10][ 80/ 117] Loss 0.368947 Top1 78.930664 Top5 97.314453 +2023-10-02 20:33:28,853 - Epoch: [10][ 90/ 117] Loss 0.369392 Top1 78.875868 Top5 97.317708 +2023-10-02 20:33:29,003 - Epoch: [10][ 100/ 117] Loss 0.365491 Top1 79.023438 Top5 97.292969 +2023-10-02 20:33:29,169 - Epoch: [10][ 110/ 117] Loss 0.366007 Top1 78.980824 Top5 97.276278 +2023-10-02 20:33:29,259 - Epoch: [10][ 117/ 117] Loss 0.365191 Top1 79.050195 Top5 97.318238 +2023-10-02 20:33:29,385 - ==> Top1: 79.050 Top5: 97.318 Loss: 0.365 + +2023-10-02 20:33:29,386 - ==> Confusion: +[[ 873 2 3 0 6 1 0 0 12 124 2 1 0 3 4 1 3 3 0 0 12] + [ 3 1042 2 1 10 13 2 20 4 2 7 1 1 0 2 2 1 0 13 1 4] + [ 14 0 935 4 4 1 39 14 0 5 5 2 6 1 4 3 1 4 4 1 9] + [ 2 4 23 914 0 2 2 1 15 3 19 0 7 3 64 1 2 10 9 0 8] + [ 23 6 1 0 954 5 0 1 4 26 0 0 1 1 15 2 5 1 0 2 3] + [ 5 65 6 1 4 925 0 24 2 17 13 4 5 25 4 0 1 1 5 3 6] + [ 1 2 27 0 2 1 1127 7 1 0 9 1 1 2 0 1 0 1 1 5 2] + [ 4 24 13 0 4 21 6 1076 3 1 7 6 1 1 2 2 1 2 33 5 6] + [ 18 5 0 0 2 1 0 2 981 44 7 0 0 5 15 2 0 6 1 0 0] + [ 54 1 0 0 3 0 2 0 43 988 0 3 0 9 8 0 0 1 0 1 6] + [ 5 4 8 13 1 0 1 2 31 2 963 0 0 7 5 1 2 1 2 0 5] + [ 0 1 0 0 0 18 1 3 1 6 0 888 48 6 0 3 2 34 0 15 9] + [ 1 5 4 2 1 0 0 0 3 2 0 30 945 5 12 7 3 38 2 2 6] + [ 1 0 2 0 3 2 2 0 33 25 8 2 0 1026 5 0 2 0 0 3 5] + [ 5 0 0 8 4 0 0 0 39 9 2 0 0 0 1018 0 0 6 2 0 8] + [ 1 0 3 2 8 0 0 0 0 0 0 9 4 3 0 1044 17 33 0 5 5] + [ 2 29 4 0 10 5 1 1 1 0 0 2 0 0 4 9 1081 1 0 4 7] + [ 1 0 1 0 0 0 1 0 0 0 0 2 14 0 2 5 0 1007 0 2 3] + [ 0 9 11 13 1 1 0 26 8 0 6 0 4 0 13 1 2 1 963 0 9] + [ 0 11 4 2 3 3 14 15 0 1 2 19 2 5 0 3 6 0 1 1058 3] + [ 262 344 161 62 127 147 57 122 239 235 256 131 483 315 244 65 222 148 171 252 3862]] + +2023-10-02 20:33:29,387 - ==> Best [Top1: 79.050 Top5: 97.318 Sparsity:0.00 Params: 169472 on epoch: 10] +2023-10-02 20:33:29,387 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:33:29,394 - + +2023-10-02 20:33:29,394 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:33:30,535 - Epoch: [11][ 10/ 1236] Overall Loss 0.355770 Objective Loss 0.355770 LR 0.001000 Time 0.114029 +2023-10-02 20:33:30,747 - Epoch: [11][ 20/ 1236] Overall Loss 0.374588 Objective Loss 0.374588 LR 0.001000 Time 0.067594 +2023-10-02 20:33:30,955 - Epoch: [11][ 30/ 1236] Overall Loss 0.362149 Objective Loss 0.362149 LR 0.001000 Time 0.051977 +2023-10-02 20:33:31,167 - Epoch: [11][ 40/ 1236] Overall Loss 0.362338 Objective Loss 0.362338 LR 0.001000 Time 0.044270 +2023-10-02 20:33:31,374 - Epoch: [11][ 50/ 1236] Overall Loss 0.361361 Objective Loss 0.361361 LR 0.001000 Time 0.039560 +2023-10-02 20:33:31,586 - Epoch: [11][ 60/ 1236] Overall Loss 0.358252 Objective Loss 0.358252 LR 0.001000 Time 0.036491 +2023-10-02 20:33:31,794 - Epoch: [11][ 70/ 1236] Overall Loss 0.357035 Objective Loss 0.357035 LR 0.001000 Time 0.034238 +2023-10-02 20:33:32,006 - Epoch: [11][ 80/ 1236] Overall Loss 0.352501 Objective Loss 0.352501 LR 0.001000 Time 0.032605 +2023-10-02 20:33:32,213 - Epoch: [11][ 90/ 1236] Overall Loss 0.352751 Objective Loss 0.352751 LR 0.001000 Time 0.031283 +2023-10-02 20:33:32,425 - Epoch: [11][ 100/ 1236] Overall Loss 0.352235 Objective Loss 0.352235 LR 0.001000 Time 0.030274 +2023-10-02 20:33:32,632 - Epoch: [11][ 110/ 1236] Overall Loss 0.350826 Objective Loss 0.350826 LR 0.001000 Time 0.029401 +2023-10-02 20:33:32,842 - Epoch: [11][ 120/ 1236] Overall Loss 0.350503 Objective Loss 0.350503 LR 0.001000 Time 0.028696 +2023-10-02 20:33:33,050 - Epoch: [11][ 130/ 1236] Overall Loss 0.348485 Objective Loss 0.348485 LR 0.001000 Time 0.028077 +2023-10-02 20:33:33,260 - Epoch: [11][ 140/ 1236] Overall Loss 0.346341 Objective Loss 0.346341 LR 0.001000 Time 0.027568 +2023-10-02 20:33:33,468 - Epoch: [11][ 150/ 1236] Overall Loss 0.344942 Objective Loss 0.344942 LR 0.001000 Time 0.027118 +2023-10-02 20:33:33,681 - Epoch: [11][ 160/ 1236] Overall Loss 0.345103 Objective Loss 0.345103 LR 0.001000 Time 0.026752 +2023-10-02 20:33:33,890 - Epoch: [11][ 170/ 1236] Overall Loss 0.344286 Objective Loss 0.344286 LR 0.001000 Time 0.026406 +2023-10-02 20:33:34,101 - Epoch: [11][ 180/ 1236] Overall Loss 0.342035 Objective Loss 0.342035 LR 0.001000 Time 0.026108 +2023-10-02 20:33:34,311 - Epoch: [11][ 190/ 1236] Overall Loss 0.342876 Objective Loss 0.342876 LR 0.001000 Time 0.025837 +2023-10-02 20:33:34,522 - Epoch: [11][ 200/ 1236] Overall Loss 0.342804 Objective Loss 0.342804 LR 0.001000 Time 0.025602 +2023-10-02 20:33:34,733 - Epoch: [11][ 210/ 1236] Overall Loss 0.340792 Objective Loss 0.340792 LR 0.001000 Time 0.025384 +2023-10-02 20:33:34,945 - Epoch: [11][ 220/ 1236] Overall Loss 0.341886 Objective Loss 0.341886 LR 0.001000 Time 0.025194 +2023-10-02 20:33:35,155 - Epoch: [11][ 230/ 1236] Overall Loss 0.341973 Objective Loss 0.341973 LR 0.001000 Time 0.025007 +2023-10-02 20:33:35,367 - Epoch: [11][ 240/ 1236] Overall Loss 0.341022 Objective Loss 0.341022 LR 0.001000 Time 0.024851 +2023-10-02 20:33:35,576 - Epoch: [11][ 250/ 1236] Overall Loss 0.344009 Objective Loss 0.344009 LR 0.001000 Time 0.024692 +2023-10-02 20:33:35,788 - Epoch: [11][ 260/ 1236] Overall Loss 0.349697 Objective Loss 0.349697 LR 0.001000 Time 0.024556 +2023-10-02 20:33:35,998 - Epoch: [11][ 270/ 1236] Overall Loss 0.352731 Objective Loss 0.352731 LR 0.001000 Time 0.024420 +2023-10-02 20:33:36,210 - Epoch: [11][ 280/ 1236] Overall Loss 0.356459 Objective Loss 0.356459 LR 0.001000 Time 0.024304 +2023-10-02 20:33:36,420 - Epoch: [11][ 290/ 1236] Overall Loss 0.358166 Objective Loss 0.358166 LR 0.001000 Time 0.024189 +2023-10-02 20:33:36,632 - Epoch: [11][ 300/ 1236] Overall Loss 0.360010 Objective Loss 0.360010 LR 0.001000 Time 0.024089 +2023-10-02 20:33:36,840 - Epoch: [11][ 310/ 1236] Overall Loss 0.361356 Objective Loss 0.361356 LR 0.001000 Time 0.023982 +2023-10-02 20:33:37,052 - Epoch: [11][ 320/ 1236] Overall Loss 0.361952 Objective Loss 0.361952 LR 0.001000 Time 0.023895 +2023-10-02 20:33:37,260 - Epoch: [11][ 330/ 1236] Overall Loss 0.362849 Objective Loss 0.362849 LR 0.001000 Time 0.023800 +2023-10-02 20:33:37,472 - Epoch: [11][ 340/ 1236] Overall Loss 0.362981 Objective Loss 0.362981 LR 0.001000 Time 0.023722 +2023-10-02 20:33:37,680 - Epoch: [11][ 350/ 1236] Overall Loss 0.363169 Objective Loss 0.363169 LR 0.001000 Time 0.023638 +2023-10-02 20:33:37,892 - Epoch: [11][ 360/ 1236] Overall Loss 0.362941 Objective Loss 0.362941 LR 0.001000 Time 0.023571 +2023-10-02 20:33:38,100 - Epoch: [11][ 370/ 1236] Overall Loss 0.362978 Objective Loss 0.362978 LR 0.001000 Time 0.023495 +2023-10-02 20:33:38,312 - Epoch: [11][ 380/ 1236] Overall Loss 0.363227 Objective Loss 0.363227 LR 0.001000 Time 0.023433 +2023-10-02 20:33:38,520 - Epoch: [11][ 390/ 1236] Overall Loss 0.362769 Objective Loss 0.362769 LR 0.001000 Time 0.023365 +2023-10-02 20:33:38,732 - Epoch: [11][ 400/ 1236] Overall Loss 0.362670 Objective Loss 0.362670 LR 0.001000 Time 0.023310 +2023-10-02 20:33:38,940 - Epoch: [11][ 410/ 1236] Overall Loss 0.362034 Objective Loss 0.362034 LR 0.001000 Time 0.023247 +2023-10-02 20:33:39,152 - Epoch: [11][ 420/ 1236] Overall Loss 0.362931 Objective Loss 0.362931 LR 0.001000 Time 0.023199 +2023-10-02 20:33:39,360 - Epoch: [11][ 430/ 1236] Overall Loss 0.362478 Objective Loss 0.362478 LR 0.001000 Time 0.023142 +2023-10-02 20:33:39,573 - Epoch: [11][ 440/ 1236] Overall Loss 0.362033 Objective Loss 0.362033 LR 0.001000 Time 0.023098 +2023-10-02 20:33:39,781 - Epoch: [11][ 450/ 1236] Overall Loss 0.362331 Objective Loss 0.362331 LR 0.001000 Time 0.023046 +2023-10-02 20:33:39,993 - Epoch: [11][ 460/ 1236] Overall Loss 0.362310 Objective Loss 0.362310 LR 0.001000 Time 0.023005 +2023-10-02 20:33:40,200 - Epoch: [11][ 470/ 1236] Overall Loss 0.362921 Objective Loss 0.362921 LR 0.001000 Time 0.022957 +2023-10-02 20:33:40,413 - Epoch: [11][ 480/ 1236] Overall Loss 0.363079 Objective Loss 0.363079 LR 0.001000 Time 0.022921 +2023-10-02 20:33:40,621 - Epoch: [11][ 490/ 1236] Overall Loss 0.362769 Objective Loss 0.362769 LR 0.001000 Time 0.022878 +2023-10-02 20:33:40,833 - Epoch: [11][ 500/ 1236] Overall Loss 0.363654 Objective Loss 0.363654 LR 0.001000 Time 0.022844 +2023-10-02 20:33:41,041 - Epoch: [11][ 510/ 1236] Overall Loss 0.363847 Objective Loss 0.363847 LR 0.001000 Time 0.022803 +2023-10-02 20:33:41,254 - Epoch: [11][ 520/ 1236] Overall Loss 0.363936 Objective Loss 0.363936 LR 0.001000 Time 0.022772 +2023-10-02 20:33:41,461 - Epoch: [11][ 530/ 1236] Overall Loss 0.363842 Objective Loss 0.363842 LR 0.001000 Time 0.022734 +2023-10-02 20:33:41,674 - Epoch: [11][ 540/ 1236] Overall Loss 0.363824 Objective Loss 0.363824 LR 0.001000 Time 0.022705 +2023-10-02 20:33:41,881 - Epoch: [11][ 550/ 1236] Overall Loss 0.363726 Objective Loss 0.363726 LR 0.001000 Time 0.022670 +2023-10-02 20:33:42,094 - Epoch: [11][ 560/ 1236] Overall Loss 0.364132 Objective Loss 0.364132 LR 0.001000 Time 0.022644 +2023-10-02 20:33:42,301 - Epoch: [11][ 570/ 1236] Overall Loss 0.363893 Objective Loss 0.363893 LR 0.001000 Time 0.022610 +2023-10-02 20:33:42,512 - Epoch: [11][ 580/ 1236] Overall Loss 0.363696 Objective Loss 0.363696 LR 0.001000 Time 0.022584 +2023-10-02 20:33:42,722 - Epoch: [11][ 590/ 1236] Overall Loss 0.363256 Objective Loss 0.363256 LR 0.001000 Time 0.022555 +2023-10-02 20:33:42,934 - Epoch: [11][ 600/ 1236] Overall Loss 0.363315 Objective Loss 0.363315 LR 0.001000 Time 0.022533 +2023-10-02 20:33:43,142 - Epoch: [11][ 610/ 1236] Overall Loss 0.362535 Objective Loss 0.362535 LR 0.001000 Time 0.022504 +2023-10-02 20:33:43,355 - Epoch: [11][ 620/ 1236] Overall Loss 0.362129 Objective Loss 0.362129 LR 0.001000 Time 0.022483 +2023-10-02 20:33:43,562 - Epoch: [11][ 630/ 1236] Overall Loss 0.361748 Objective Loss 0.361748 LR 0.001000 Time 0.022455 +2023-10-02 20:33:43,774 - Epoch: [11][ 640/ 1236] Overall Loss 0.361103 Objective Loss 0.361103 LR 0.001000 Time 0.022435 +2023-10-02 20:33:43,982 - Epoch: [11][ 650/ 1236] Overall Loss 0.361085 Objective Loss 0.361085 LR 0.001000 Time 0.022409 +2023-10-02 20:33:44,194 - Epoch: [11][ 660/ 1236] Overall Loss 0.360681 Objective Loss 0.360681 LR 0.001000 Time 0.022391 +2023-10-02 20:33:44,402 - Epoch: [11][ 670/ 1236] Overall Loss 0.360411 Objective Loss 0.360411 LR 0.001000 Time 0.022366 +2023-10-02 20:33:44,615 - Epoch: [11][ 680/ 1236] Overall Loss 0.359823 Objective Loss 0.359823 LR 0.001000 Time 0.022349 +2023-10-02 20:33:44,822 - Epoch: [11][ 690/ 1236] Overall Loss 0.359641 Objective Loss 0.359641 LR 0.001000 Time 0.022326 +2023-10-02 20:33:45,035 - Epoch: [11][ 700/ 1236] Overall Loss 0.359462 Objective Loss 0.359462 LR 0.001000 Time 0.022310 +2023-10-02 20:33:45,242 - Epoch: [11][ 710/ 1236] Overall Loss 0.359707 Objective Loss 0.359707 LR 0.001000 Time 0.022288 +2023-10-02 20:33:45,454 - Epoch: [11][ 720/ 1236] Overall Loss 0.359290 Objective Loss 0.359290 LR 0.001000 Time 0.022272 +2023-10-02 20:33:45,662 - Epoch: [11][ 730/ 1236] Overall Loss 0.358829 Objective Loss 0.358829 LR 0.001000 Time 0.022251 +2023-10-02 20:33:45,875 - Epoch: [11][ 740/ 1236] Overall Loss 0.358757 Objective Loss 0.358757 LR 0.001000 Time 0.022237 +2023-10-02 20:33:46,082 - Epoch: [11][ 750/ 1236] Overall Loss 0.358546 Objective Loss 0.358546 LR 0.001000 Time 0.022217 +2023-10-02 20:33:46,294 - Epoch: [11][ 760/ 1236] Overall Loss 0.358557 Objective Loss 0.358557 LR 0.001000 Time 0.022203 +2023-10-02 20:33:46,503 - Epoch: [11][ 770/ 1236] Overall Loss 0.358026 Objective Loss 0.358026 LR 0.001000 Time 0.022185 +2023-10-02 20:33:46,716 - Epoch: [11][ 780/ 1236] Overall Loss 0.358139 Objective Loss 0.358139 LR 0.001000 Time 0.022172 +2023-10-02 20:33:46,924 - Epoch: [11][ 790/ 1236] Overall Loss 0.358222 Objective Loss 0.358222 LR 0.001000 Time 0.022154 +2023-10-02 20:33:47,135 - Epoch: [11][ 800/ 1236] Overall Loss 0.357996 Objective Loss 0.357996 LR 0.001000 Time 0.022141 +2023-10-02 20:33:47,344 - Epoch: [11][ 810/ 1236] Overall Loss 0.358070 Objective Loss 0.358070 LR 0.001000 Time 0.022124 +2023-10-02 20:33:47,555 - Epoch: [11][ 820/ 1236] Overall Loss 0.358105 Objective Loss 0.358105 LR 0.001000 Time 0.022111 +2023-10-02 20:33:47,764 - Epoch: [11][ 830/ 1236] Overall Loss 0.357832 Objective Loss 0.357832 LR 0.001000 Time 0.022096 +2023-10-02 20:33:47,976 - Epoch: [11][ 840/ 1236] Overall Loss 0.358051 Objective Loss 0.358051 LR 0.001000 Time 0.022085 +2023-10-02 20:33:48,184 - Epoch: [11][ 850/ 1236] Overall Loss 0.358061 Objective Loss 0.358061 LR 0.001000 Time 0.022070 +2023-10-02 20:33:48,397 - Epoch: [11][ 860/ 1236] Overall Loss 0.358346 Objective Loss 0.358346 LR 0.001000 Time 0.022060 +2023-10-02 20:33:48,605 - Epoch: [11][ 870/ 1236] Overall Loss 0.358264 Objective Loss 0.358264 LR 0.001000 Time 0.022045 +2023-10-02 20:33:48,817 - Epoch: [11][ 880/ 1236] Overall Loss 0.357873 Objective Loss 0.357873 LR 0.001000 Time 0.022035 +2023-10-02 20:33:49,025 - Epoch: [11][ 890/ 1236] Overall Loss 0.357701 Objective Loss 0.357701 LR 0.001000 Time 0.022021 +2023-10-02 20:33:49,236 - Epoch: [11][ 900/ 1236] Overall Loss 0.357306 Objective Loss 0.357306 LR 0.001000 Time 0.022010 +2023-10-02 20:33:49,445 - Epoch: [11][ 910/ 1236] Overall Loss 0.357463 Objective Loss 0.357463 LR 0.001000 Time 0.021997 +2023-10-02 20:33:49,658 - Epoch: [11][ 920/ 1236] Overall Loss 0.357494 Objective Loss 0.357494 LR 0.001000 Time 0.021988 +2023-10-02 20:33:49,866 - Epoch: [11][ 930/ 1236] Overall Loss 0.357580 Objective Loss 0.357580 LR 0.001000 Time 0.021975 +2023-10-02 20:33:50,077 - Epoch: [11][ 940/ 1236] Overall Loss 0.357562 Objective Loss 0.357562 LR 0.001000 Time 0.021966 +2023-10-02 20:33:50,286 - Epoch: [11][ 950/ 1236] Overall Loss 0.357084 Objective Loss 0.357084 LR 0.001000 Time 0.021953 +2023-10-02 20:33:50,499 - Epoch: [11][ 960/ 1236] Overall Loss 0.357067 Objective Loss 0.357067 LR 0.001000 Time 0.021946 +2023-10-02 20:33:50,706 - Epoch: [11][ 970/ 1236] Overall Loss 0.357061 Objective Loss 0.357061 LR 0.001000 Time 0.021933 +2023-10-02 20:33:50,919 - Epoch: [11][ 980/ 1236] Overall Loss 0.356923 Objective Loss 0.356923 LR 0.001000 Time 0.021926 +2023-10-02 20:33:51,127 - Epoch: [11][ 990/ 1236] Overall Loss 0.356718 Objective Loss 0.356718 LR 0.001000 Time 0.021914 +2023-10-02 20:33:51,339 - Epoch: [11][ 1000/ 1236] Overall Loss 0.356732 Objective Loss 0.356732 LR 0.001000 Time 0.021907 +2023-10-02 20:33:51,547 - Epoch: [11][ 1010/ 1236] Overall Loss 0.356747 Objective Loss 0.356747 LR 0.001000 Time 0.021896 +2023-10-02 20:33:51,759 - Epoch: [11][ 1020/ 1236] Overall Loss 0.357078 Objective Loss 0.357078 LR 0.001000 Time 0.021888 +2023-10-02 20:33:51,968 - Epoch: [11][ 1030/ 1236] Overall Loss 0.356956 Objective Loss 0.356956 LR 0.001000 Time 0.021877 +2023-10-02 20:33:52,181 - Epoch: [11][ 1040/ 1236] Overall Loss 0.357132 Objective Loss 0.357132 LR 0.001000 Time 0.021871 +2023-10-02 20:33:52,388 - Epoch: [11][ 1050/ 1236] Overall Loss 0.356989 Objective Loss 0.356989 LR 0.001000 Time 0.021860 +2023-10-02 20:33:52,601 - Epoch: [11][ 1060/ 1236] Overall Loss 0.357095 Objective Loss 0.357095 LR 0.001000 Time 0.021854 +2023-10-02 20:33:52,810 - Epoch: [11][ 1070/ 1236] Overall Loss 0.357039 Objective Loss 0.357039 LR 0.001000 Time 0.021846 +2023-10-02 20:33:53,017 - Epoch: [11][ 1080/ 1236] Overall Loss 0.356939 Objective Loss 0.356939 LR 0.001000 Time 0.021834 +2023-10-02 20:33:53,226 - Epoch: [11][ 1090/ 1236] Overall Loss 0.356804 Objective Loss 0.356804 LR 0.001000 Time 0.021826 +2023-10-02 20:33:53,434 - Epoch: [11][ 1100/ 1236] Overall Loss 0.356522 Objective Loss 0.356522 LR 0.001000 Time 0.021815 +2023-10-02 20:33:53,643 - Epoch: [11][ 1110/ 1236] Overall Loss 0.356710 Objective Loss 0.356710 LR 0.001000 Time 0.021807 +2023-10-02 20:33:53,850 - Epoch: [11][ 1120/ 1236] Overall Loss 0.356837 Objective Loss 0.356837 LR 0.001000 Time 0.021797 +2023-10-02 20:33:54,060 - Epoch: [11][ 1130/ 1236] Overall Loss 0.356583 Objective Loss 0.356583 LR 0.001000 Time 0.021789 +2023-10-02 20:33:54,267 - Epoch: [11][ 1140/ 1236] Overall Loss 0.356553 Objective Loss 0.356553 LR 0.001000 Time 0.021779 +2023-10-02 20:33:54,477 - Epoch: [11][ 1150/ 1236] Overall Loss 0.356357 Objective Loss 0.356357 LR 0.001000 Time 0.021772 +2023-10-02 20:33:54,683 - Epoch: [11][ 1160/ 1236] Overall Loss 0.356080 Objective Loss 0.356080 LR 0.001000 Time 0.021762 +2023-10-02 20:33:54,893 - Epoch: [11][ 1170/ 1236] Overall Loss 0.355951 Objective Loss 0.355951 LR 0.001000 Time 0.021755 +2023-10-02 20:33:55,100 - Epoch: [11][ 1180/ 1236] Overall Loss 0.356319 Objective Loss 0.356319 LR 0.001000 Time 0.021746 +2023-10-02 20:33:55,310 - Epoch: [11][ 1190/ 1236] Overall Loss 0.356646 Objective Loss 0.356646 LR 0.001000 Time 0.021740 +2023-10-02 20:33:55,517 - Epoch: [11][ 1200/ 1236] Overall Loss 0.356499 Objective Loss 0.356499 LR 0.001000 Time 0.021730 +2023-10-02 20:33:55,725 - Epoch: [11][ 1210/ 1236] Overall Loss 0.356766 Objective Loss 0.356766 LR 0.001000 Time 0.021723 +2023-10-02 20:33:55,933 - Epoch: [11][ 1220/ 1236] Overall Loss 0.356578 Objective Loss 0.356578 LR 0.001000 Time 0.021714 +2023-10-02 20:33:56,194 - Epoch: [11][ 1230/ 1236] Overall Loss 0.356572 Objective Loss 0.356572 LR 0.001000 Time 0.021749 +2023-10-02 20:33:56,315 - Epoch: [11][ 1236/ 1236] Overall Loss 0.356679 Objective Loss 0.356679 Top1 85.132383 Top5 96.537678 LR 0.001000 Time 0.021741 +2023-10-02 20:33:56,448 - --- validate (epoch=11)----------- +2023-10-02 20:33:56,448 - 29943 samples (256 per mini-batch) +2023-10-02 20:33:56,931 - Epoch: [11][ 10/ 117] Loss 0.354232 Top1 80.351562 Top5 98.085938 +2023-10-02 20:33:57,081 - Epoch: [11][ 20/ 117] Loss 0.360625 Top1 80.859375 Top5 97.832031 +2023-10-02 20:33:57,229 - Epoch: [11][ 30/ 117] Loss 0.375589 Top1 80.091146 Top5 97.708333 +2023-10-02 20:33:57,380 - Epoch: [11][ 40/ 117] Loss 0.371489 Top1 80.224609 Top5 97.812500 +2023-10-02 20:33:57,527 - Epoch: [11][ 50/ 117] Loss 0.362110 Top1 80.421875 Top5 97.820312 +2023-10-02 20:33:57,677 - Epoch: [11][ 60/ 117] Loss 0.365909 Top1 80.449219 Top5 97.812500 +2023-10-02 20:33:57,825 - Epoch: [11][ 70/ 117] Loss 0.366237 Top1 80.446429 Top5 97.806920 +2023-10-02 20:33:57,975 - Epoch: [11][ 80/ 117] Loss 0.367885 Top1 80.332031 Top5 97.827148 +2023-10-02 20:33:58,125 - Epoch: [11][ 90/ 117] Loss 0.364913 Top1 80.390625 Top5 97.877604 +2023-10-02 20:33:58,276 - Epoch: [11][ 100/ 117] Loss 0.366300 Top1 80.281250 Top5 97.917969 +2023-10-02 20:33:58,433 - Epoch: [11][ 110/ 117] Loss 0.364682 Top1 80.273438 Top5 97.855114 +2023-10-02 20:33:58,521 - Epoch: [11][ 117/ 117] Loss 0.363265 Top1 80.239121 Top5 97.865945 +2023-10-02 20:33:58,660 - ==> Top1: 80.239 Top5: 97.866 Loss: 0.363 + +2023-10-02 20:33:58,661 - ==> Confusion: +[[ 965 0 3 1 7 5 0 0 1 38 0 1 0 6 2 4 5 0 0 0 12] + [ 3 996 3 1 5 69 2 17 3 0 1 0 1 0 1 3 5 0 13 4 4] + [ 9 0 921 8 4 4 59 10 1 0 0 1 8 5 0 5 0 0 7 3 11] + [ 3 1 27 968 0 7 5 0 1 0 4 0 10 4 24 3 2 4 9 0 17] + [ 26 10 3 1 947 14 0 0 1 1 0 0 2 7 12 5 14 0 1 1 5] + [ 4 19 2 2 1 1018 1 9 0 2 2 4 4 25 7 2 4 0 2 2 6] + [ 0 2 19 1 1 3 1141 8 0 0 0 1 2 1 0 2 1 0 1 5 3] + [ 10 20 24 1 4 51 4 1030 0 0 1 4 7 0 0 3 0 0 49 4 6] + [ 42 7 1 0 1 5 0 1 856 51 15 0 7 37 51 3 3 4 2 2 1] + [ 177 1 0 0 6 7 1 0 12 839 0 0 1 41 13 2 1 1 0 3 14] + [ 6 3 10 20 1 4 6 3 7 1 949 0 3 16 6 3 2 0 3 0 10] + [ 0 1 1 0 0 24 1 2 0 0 0 922 27 14 1 6 2 16 0 15 3] + [ 1 0 4 2 0 7 0 2 1 0 0 45 954 3 5 13 3 8 0 6 14] + [ 0 0 3 0 6 18 0 0 1 6 5 3 0 1062 2 1 1 0 0 3 8] + [ 10 2 2 21 6 2 0 0 8 1 4 0 5 2 1019 1 3 1 1 0 13] + [ 2 1 4 3 5 2 2 0 0 0 0 10 8 1 0 1067 15 7 0 5 2] + [ 3 15 0 0 6 12 2 0 2 0 1 4 0 1 5 14 1081 0 1 2 12] + [ 0 1 2 5 0 0 2 0 0 0 0 17 38 0 1 19 1 945 0 3 4] + [ 1 9 8 24 1 4 1 19 1 0 2 0 2 0 20 3 2 0 962 1 8] + [ 0 5 4 2 2 13 11 11 0 0 1 13 3 0 0 8 5 0 1 1071 2] + [ 235 258 169 59 103 425 58 85 37 60 136 158 453 406 203 123 144 39 173 268 4313]] + +2023-10-02 20:33:58,662 - ==> Best [Top1: 80.239 Top5: 97.866 Sparsity:0.00 Params: 169472 on epoch: 11] +2023-10-02 20:33:58,662 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:33:58,675 - + +2023-10-02 20:33:58,675 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:33:59,668 - Epoch: [12][ 10/ 1236] Overall Loss 0.325942 Objective Loss 0.325942 LR 0.001000 Time 0.099163 +2023-10-02 20:33:59,874 - Epoch: [12][ 20/ 1236] Overall Loss 0.335389 Objective Loss 0.335389 LR 0.001000 Time 0.059875 +2023-10-02 20:34:00,084 - Epoch: [12][ 30/ 1236] Overall Loss 0.346114 Objective Loss 0.346114 LR 0.001000 Time 0.046911 +2023-10-02 20:34:00,294 - Epoch: [12][ 40/ 1236] Overall Loss 0.339447 Objective Loss 0.339447 LR 0.001000 Time 0.040414 +2023-10-02 20:34:00,503 - Epoch: [12][ 50/ 1236] Overall Loss 0.337739 Objective Loss 0.337739 LR 0.001000 Time 0.036522 +2023-10-02 20:34:00,713 - Epoch: [12][ 60/ 1236] Overall Loss 0.347436 Objective Loss 0.347436 LR 0.001000 Time 0.033923 +2023-10-02 20:34:00,923 - Epoch: [12][ 70/ 1236] Overall Loss 0.346090 Objective Loss 0.346090 LR 0.001000 Time 0.032075 +2023-10-02 20:34:01,132 - Epoch: [12][ 80/ 1236] Overall Loss 0.347724 Objective Loss 0.347724 LR 0.001000 Time 0.030674 +2023-10-02 20:34:01,340 - Epoch: [12][ 90/ 1236] Overall Loss 0.350798 Objective Loss 0.350798 LR 0.001000 Time 0.029564 +2023-10-02 20:34:01,549 - Epoch: [12][ 100/ 1236] Overall Loss 0.347681 Objective Loss 0.347681 LR 0.001000 Time 0.028693 +2023-10-02 20:34:01,755 - Epoch: [12][ 110/ 1236] Overall Loss 0.344607 Objective Loss 0.344607 LR 0.001000 Time 0.027955 +2023-10-02 20:34:01,964 - Epoch: [12][ 120/ 1236] Overall Loss 0.343763 Objective Loss 0.343763 LR 0.001000 Time 0.027364 +2023-10-02 20:34:02,170 - Epoch: [12][ 130/ 1236] Overall Loss 0.340836 Objective Loss 0.340836 LR 0.001000 Time 0.026844 +2023-10-02 20:34:02,378 - Epoch: [12][ 140/ 1236] Overall Loss 0.341394 Objective Loss 0.341394 LR 0.001000 Time 0.026408 +2023-10-02 20:34:02,585 - Epoch: [12][ 150/ 1236] Overall Loss 0.341270 Objective Loss 0.341270 LR 0.001000 Time 0.026015 +2023-10-02 20:34:02,793 - Epoch: [12][ 160/ 1236] Overall Loss 0.340217 Objective Loss 0.340217 LR 0.001000 Time 0.025686 +2023-10-02 20:34:02,999 - Epoch: [12][ 170/ 1236] Overall Loss 0.341651 Objective Loss 0.341651 LR 0.001000 Time 0.025382 +2023-10-02 20:34:03,207 - Epoch: [12][ 180/ 1236] Overall Loss 0.344018 Objective Loss 0.344018 LR 0.001000 Time 0.025125 +2023-10-02 20:34:03,414 - Epoch: [12][ 190/ 1236] Overall Loss 0.343908 Objective Loss 0.343908 LR 0.001000 Time 0.024884 +2023-10-02 20:34:03,621 - Epoch: [12][ 200/ 1236] Overall Loss 0.344647 Objective Loss 0.344647 LR 0.001000 Time 0.024675 +2023-10-02 20:34:03,827 - Epoch: [12][ 210/ 1236] Overall Loss 0.344553 Objective Loss 0.344553 LR 0.001000 Time 0.024470 +2023-10-02 20:34:04,034 - Epoch: [12][ 220/ 1236] Overall Loss 0.342387 Objective Loss 0.342387 LR 0.001000 Time 0.024300 +2023-10-02 20:34:04,241 - Epoch: [12][ 230/ 1236] Overall Loss 0.341988 Objective Loss 0.341988 LR 0.001000 Time 0.024137 +2023-10-02 20:34:04,449 - Epoch: [12][ 240/ 1236] Overall Loss 0.342844 Objective Loss 0.342844 LR 0.001000 Time 0.023996 +2023-10-02 20:34:04,656 - Epoch: [12][ 250/ 1236] Overall Loss 0.343663 Objective Loss 0.343663 LR 0.001000 Time 0.023857 +2023-10-02 20:34:04,863 - Epoch: [12][ 260/ 1236] Overall Loss 0.342830 Objective Loss 0.342830 LR 0.001000 Time 0.023736 +2023-10-02 20:34:05,070 - Epoch: [12][ 270/ 1236] Overall Loss 0.342485 Objective Loss 0.342485 LR 0.001000 Time 0.023617 +2023-10-02 20:34:05,278 - Epoch: [12][ 280/ 1236] Overall Loss 0.342985 Objective Loss 0.342985 LR 0.001000 Time 0.023517 +2023-10-02 20:34:05,485 - Epoch: [12][ 290/ 1236] Overall Loss 0.342322 Objective Loss 0.342322 LR 0.001000 Time 0.023414 +2023-10-02 20:34:05,692 - Epoch: [12][ 300/ 1236] Overall Loss 0.341660 Objective Loss 0.341660 LR 0.001000 Time 0.023324 +2023-10-02 20:34:05,901 - Epoch: [12][ 310/ 1236] Overall Loss 0.341546 Objective Loss 0.341546 LR 0.001000 Time 0.023240 +2023-10-02 20:34:06,108 - Epoch: [12][ 320/ 1236] Overall Loss 0.341773 Objective Loss 0.341773 LR 0.001000 Time 0.023158 +2023-10-02 20:34:06,320 - Epoch: [12][ 330/ 1236] Overall Loss 0.341581 Objective Loss 0.341581 LR 0.001000 Time 0.023098 +2023-10-02 20:34:06,528 - Epoch: [12][ 340/ 1236] Overall Loss 0.342524 Objective Loss 0.342524 LR 0.001000 Time 0.023031 +2023-10-02 20:34:06,741 - Epoch: [12][ 350/ 1236] Overall Loss 0.342226 Objective Loss 0.342226 LR 0.001000 Time 0.022979 +2023-10-02 20:34:06,949 - Epoch: [12][ 360/ 1236] Overall Loss 0.342180 Objective Loss 0.342180 LR 0.001000 Time 0.022918 +2023-10-02 20:34:07,162 - Epoch: [12][ 370/ 1236] Overall Loss 0.341351 Objective Loss 0.341351 LR 0.001000 Time 0.022872 +2023-10-02 20:34:07,370 - Epoch: [12][ 380/ 1236] Overall Loss 0.341335 Objective Loss 0.341335 LR 0.001000 Time 0.022818 +2023-10-02 20:34:07,583 - Epoch: [12][ 390/ 1236] Overall Loss 0.341604 Objective Loss 0.341604 LR 0.001000 Time 0.022777 +2023-10-02 20:34:07,792 - Epoch: [12][ 400/ 1236] Overall Loss 0.341578 Objective Loss 0.341578 LR 0.001000 Time 0.022729 +2023-10-02 20:34:08,004 - Epoch: [12][ 410/ 1236] Overall Loss 0.340985 Objective Loss 0.340985 LR 0.001000 Time 0.022691 +2023-10-02 20:34:08,213 - Epoch: [12][ 420/ 1236] Overall Loss 0.341675 Objective Loss 0.341675 LR 0.001000 Time 0.022647 +2023-10-02 20:34:08,426 - Epoch: [12][ 430/ 1236] Overall Loss 0.342235 Objective Loss 0.342235 LR 0.001000 Time 0.022614 +2023-10-02 20:34:08,634 - Epoch: [12][ 440/ 1236] Overall Loss 0.342901 Objective Loss 0.342901 LR 0.001000 Time 0.022573 +2023-10-02 20:34:08,847 - Epoch: [12][ 450/ 1236] Overall Loss 0.343101 Objective Loss 0.343101 LR 0.001000 Time 0.022542 +2023-10-02 20:34:09,055 - Epoch: [12][ 460/ 1236] Overall Loss 0.343575 Objective Loss 0.343575 LR 0.001000 Time 0.022505 +2023-10-02 20:34:09,268 - Epoch: [12][ 470/ 1236] Overall Loss 0.342999 Objective Loss 0.342999 LR 0.001000 Time 0.022477 +2023-10-02 20:34:09,476 - Epoch: [12][ 480/ 1236] Overall Loss 0.342671 Objective Loss 0.342671 LR 0.001000 Time 0.022443 +2023-10-02 20:34:09,689 - Epoch: [12][ 490/ 1236] Overall Loss 0.341577 Objective Loss 0.341577 LR 0.001000 Time 0.022418 +2023-10-02 20:34:09,898 - Epoch: [12][ 500/ 1236] Overall Loss 0.341373 Objective Loss 0.341373 LR 0.001000 Time 0.022386 +2023-10-02 20:34:10,110 - Epoch: [12][ 510/ 1236] Overall Loss 0.341119 Objective Loss 0.341119 LR 0.001000 Time 0.022363 +2023-10-02 20:34:10,319 - Epoch: [12][ 520/ 1236] Overall Loss 0.340771 Objective Loss 0.340771 LR 0.001000 Time 0.022333 +2023-10-02 20:34:10,531 - Epoch: [12][ 530/ 1236] Overall Loss 0.341497 Objective Loss 0.341497 LR 0.001000 Time 0.022312 +2023-10-02 20:34:10,740 - Epoch: [12][ 540/ 1236] Overall Loss 0.341032 Objective Loss 0.341032 LR 0.001000 Time 0.022285 +2023-10-02 20:34:10,953 - Epoch: [12][ 550/ 1236] Overall Loss 0.340627 Objective Loss 0.340627 LR 0.001000 Time 0.022265 +2023-10-02 20:34:11,161 - Epoch: [12][ 560/ 1236] Overall Loss 0.340765 Objective Loss 0.340765 LR 0.001000 Time 0.022240 +2023-10-02 20:34:11,374 - Epoch: [12][ 570/ 1236] Overall Loss 0.341190 Objective Loss 0.341190 LR 0.001000 Time 0.022222 +2023-10-02 20:34:11,583 - Epoch: [12][ 580/ 1236] Overall Loss 0.341011 Objective Loss 0.341011 LR 0.001000 Time 0.022197 +2023-10-02 20:34:11,795 - Epoch: [12][ 590/ 1236] Overall Loss 0.340784 Objective Loss 0.340784 LR 0.001000 Time 0.022181 +2023-10-02 20:34:12,004 - Epoch: [12][ 600/ 1236] Overall Loss 0.340862 Objective Loss 0.340862 LR 0.001000 Time 0.022159 +2023-10-02 20:34:12,216 - Epoch: [12][ 610/ 1236] Overall Loss 0.341144 Objective Loss 0.341144 LR 0.001000 Time 0.022143 +2023-10-02 20:34:12,425 - Epoch: [12][ 620/ 1236] Overall Loss 0.340908 Objective Loss 0.340908 LR 0.001000 Time 0.022122 +2023-10-02 20:34:12,638 - Epoch: [12][ 630/ 1236] Overall Loss 0.341365 Objective Loss 0.341365 LR 0.001000 Time 0.022108 +2023-10-02 20:34:12,847 - Epoch: [12][ 640/ 1236] Overall Loss 0.341749 Objective Loss 0.341749 LR 0.001000 Time 0.022088 +2023-10-02 20:34:13,060 - Epoch: [12][ 650/ 1236] Overall Loss 0.342193 Objective Loss 0.342193 LR 0.001000 Time 0.022075 +2023-10-02 20:34:13,269 - Epoch: [12][ 660/ 1236] Overall Loss 0.341430 Objective Loss 0.341430 LR 0.001000 Time 0.022056 +2023-10-02 20:34:13,482 - Epoch: [12][ 670/ 1236] Overall Loss 0.341231 Objective Loss 0.341231 LR 0.001000 Time 0.022044 +2023-10-02 20:34:13,691 - Epoch: [12][ 680/ 1236] Overall Loss 0.341318 Objective Loss 0.341318 LR 0.001000 Time 0.022027 +2023-10-02 20:34:13,903 - Epoch: [12][ 690/ 1236] Overall Loss 0.340690 Objective Loss 0.340690 LR 0.001000 Time 0.022015 +2023-10-02 20:34:14,112 - Epoch: [12][ 700/ 1236] Overall Loss 0.340904 Objective Loss 0.340904 LR 0.001000 Time 0.021999 +2023-10-02 20:34:14,325 - Epoch: [12][ 710/ 1236] Overall Loss 0.341067 Objective Loss 0.341067 LR 0.001000 Time 0.021988 +2023-10-02 20:34:14,534 - Epoch: [12][ 720/ 1236] Overall Loss 0.341348 Objective Loss 0.341348 LR 0.001000 Time 0.021972 +2023-10-02 20:34:14,746 - Epoch: [12][ 730/ 1236] Overall Loss 0.341900 Objective Loss 0.341900 LR 0.001000 Time 0.021961 +2023-10-02 20:34:14,955 - Epoch: [12][ 740/ 1236] Overall Loss 0.341809 Objective Loss 0.341809 LR 0.001000 Time 0.021946 +2023-10-02 20:34:15,168 - Epoch: [12][ 750/ 1236] Overall Loss 0.342301 Objective Loss 0.342301 LR 0.001000 Time 0.021937 +2023-10-02 20:34:15,377 - Epoch: [12][ 760/ 1236] Overall Loss 0.342707 Objective Loss 0.342707 LR 0.001000 Time 0.021923 +2023-10-02 20:34:15,590 - Epoch: [12][ 770/ 1236] Overall Loss 0.342411 Objective Loss 0.342411 LR 0.001000 Time 0.021914 +2023-10-02 20:34:15,798 - Epoch: [12][ 780/ 1236] Overall Loss 0.342251 Objective Loss 0.342251 LR 0.001000 Time 0.021900 +2023-10-02 20:34:16,015 - Epoch: [12][ 790/ 1236] Overall Loss 0.342526 Objective Loss 0.342526 LR 0.001000 Time 0.021890 +2023-10-02 20:34:16,224 - Epoch: [12][ 800/ 1236] Overall Loss 0.341905 Objective Loss 0.341905 LR 0.001000 Time 0.021877 +2023-10-02 20:34:16,437 - Epoch: [12][ 810/ 1236] Overall Loss 0.341555 Objective Loss 0.341555 LR 0.001000 Time 0.021869 +2023-10-02 20:34:16,645 - Epoch: [12][ 820/ 1236] Overall Loss 0.341675 Objective Loss 0.341675 LR 0.001000 Time 0.021857 +2023-10-02 20:34:16,858 - Epoch: [12][ 830/ 1236] Overall Loss 0.341806 Objective Loss 0.341806 LR 0.001000 Time 0.021849 +2023-10-02 20:34:17,067 - Epoch: [12][ 840/ 1236] Overall Loss 0.341986 Objective Loss 0.341986 LR 0.001000 Time 0.021838 +2023-10-02 20:34:17,280 - Epoch: [12][ 850/ 1236] Overall Loss 0.342115 Objective Loss 0.342115 LR 0.001000 Time 0.021830 +2023-10-02 20:34:17,489 - Epoch: [12][ 860/ 1236] Overall Loss 0.341578 Objective Loss 0.341578 LR 0.001000 Time 0.021819 +2023-10-02 20:34:17,702 - Epoch: [12][ 870/ 1236] Overall Loss 0.342075 Objective Loss 0.342075 LR 0.001000 Time 0.021812 +2023-10-02 20:34:17,910 - Epoch: [12][ 880/ 1236] Overall Loss 0.342345 Objective Loss 0.342345 LR 0.001000 Time 0.021801 +2023-10-02 20:34:18,123 - Epoch: [12][ 890/ 1236] Overall Loss 0.342363 Objective Loss 0.342363 LR 0.001000 Time 0.021795 +2023-10-02 20:34:18,334 - Epoch: [12][ 900/ 1236] Overall Loss 0.342417 Objective Loss 0.342417 LR 0.001000 Time 0.021787 +2023-10-02 20:34:18,548 - Epoch: [12][ 910/ 1236] Overall Loss 0.342374 Objective Loss 0.342374 LR 0.001000 Time 0.021781 +2023-10-02 20:34:18,768 - Epoch: [12][ 920/ 1236] Overall Loss 0.342384 Objective Loss 0.342384 LR 0.001000 Time 0.021783 +2023-10-02 20:34:18,983 - Epoch: [12][ 930/ 1236] Overall Loss 0.342350 Objective Loss 0.342350 LR 0.001000 Time 0.021780 +2023-10-02 20:34:19,203 - Epoch: [12][ 940/ 1236] Overall Loss 0.342049 Objective Loss 0.342049 LR 0.001000 Time 0.021782 +2023-10-02 20:34:19,419 - Epoch: [12][ 950/ 1236] Overall Loss 0.342171 Objective Loss 0.342171 LR 0.001000 Time 0.021780 +2023-10-02 20:34:19,639 - Epoch: [12][ 960/ 1236] Overall Loss 0.342248 Objective Loss 0.342248 LR 0.001000 Time 0.021782 +2023-10-02 20:34:19,855 - Epoch: [12][ 970/ 1236] Overall Loss 0.342019 Objective Loss 0.342019 LR 0.001000 Time 0.021780 +2023-10-02 20:34:20,069 - Epoch: [12][ 980/ 1236] Overall Loss 0.341937 Objective Loss 0.341937 LR 0.001000 Time 0.021775 +2023-10-02 20:34:20,274 - Epoch: [12][ 990/ 1236] Overall Loss 0.341749 Objective Loss 0.341749 LR 0.001000 Time 0.021762 +2023-10-02 20:34:20,482 - Epoch: [12][ 1000/ 1236] Overall Loss 0.341683 Objective Loss 0.341683 LR 0.001000 Time 0.021752 +2023-10-02 20:34:20,689 - Epoch: [12][ 1010/ 1236] Overall Loss 0.341660 Objective Loss 0.341660 LR 0.001000 Time 0.021740 +2023-10-02 20:34:20,898 - Epoch: [12][ 1020/ 1236] Overall Loss 0.341683 Objective Loss 0.341683 LR 0.001000 Time 0.021731 +2023-10-02 20:34:21,103 - Epoch: [12][ 1030/ 1236] Overall Loss 0.341940 Objective Loss 0.341940 LR 0.001000 Time 0.021719 +2023-10-02 20:34:21,312 - Epoch: [12][ 1040/ 1236] Overall Loss 0.342065 Objective Loss 0.342065 LR 0.001000 Time 0.021711 +2023-10-02 20:34:21,517 - Epoch: [12][ 1050/ 1236] Overall Loss 0.341783 Objective Loss 0.341783 LR 0.001000 Time 0.021699 +2023-10-02 20:34:21,726 - Epoch: [12][ 1060/ 1236] Overall Loss 0.341690 Objective Loss 0.341690 LR 0.001000 Time 0.021692 +2023-10-02 20:34:21,934 - Epoch: [12][ 1070/ 1236] Overall Loss 0.341802 Objective Loss 0.341802 LR 0.001000 Time 0.021683 +2023-10-02 20:34:22,144 - Epoch: [12][ 1080/ 1236] Overall Loss 0.341949 Objective Loss 0.341949 LR 0.001000 Time 0.021676 +2023-10-02 20:34:22,349 - Epoch: [12][ 1090/ 1236] Overall Loss 0.341821 Objective Loss 0.341821 LR 0.001000 Time 0.021665 +2023-10-02 20:34:22,556 - Epoch: [12][ 1100/ 1236] Overall Loss 0.341636 Objective Loss 0.341636 LR 0.001000 Time 0.021656 +2023-10-02 20:34:22,762 - Epoch: [12][ 1110/ 1236] Overall Loss 0.341612 Objective Loss 0.341612 LR 0.001000 Time 0.021646 +2023-10-02 20:34:22,970 - Epoch: [12][ 1120/ 1236] Overall Loss 0.341513 Objective Loss 0.341513 LR 0.001000 Time 0.021638 +2023-10-02 20:34:23,175 - Epoch: [12][ 1130/ 1236] Overall Loss 0.341356 Objective Loss 0.341356 LR 0.001000 Time 0.021628 +2023-10-02 20:34:23,382 - Epoch: [12][ 1140/ 1236] Overall Loss 0.341389 Objective Loss 0.341389 LR 0.001000 Time 0.021619 +2023-10-02 20:34:23,587 - Epoch: [12][ 1150/ 1236] Overall Loss 0.341376 Objective Loss 0.341376 LR 0.001000 Time 0.021610 +2023-10-02 20:34:23,794 - Epoch: [12][ 1160/ 1236] Overall Loss 0.341344 Objective Loss 0.341344 LR 0.001000 Time 0.021601 +2023-10-02 20:34:23,999 - Epoch: [12][ 1170/ 1236] Overall Loss 0.341312 Objective Loss 0.341312 LR 0.001000 Time 0.021592 +2023-10-02 20:34:24,206 - Epoch: [12][ 1180/ 1236] Overall Loss 0.341341 Objective Loss 0.341341 LR 0.001000 Time 0.021583 +2023-10-02 20:34:24,412 - Epoch: [12][ 1190/ 1236] Overall Loss 0.341206 Objective Loss 0.341206 LR 0.001000 Time 0.021575 +2023-10-02 20:34:24,619 - Epoch: [12][ 1200/ 1236] Overall Loss 0.341023 Objective Loss 0.341023 LR 0.001000 Time 0.021567 +2023-10-02 20:34:24,825 - Epoch: [12][ 1210/ 1236] Overall Loss 0.341090 Objective Loss 0.341090 LR 0.001000 Time 0.021559 +2023-10-02 20:34:25,031 - Epoch: [12][ 1220/ 1236] Overall Loss 0.341055 Objective Loss 0.341055 LR 0.001000 Time 0.021551 +2023-10-02 20:34:25,292 - Epoch: [12][ 1230/ 1236] Overall Loss 0.341373 Objective Loss 0.341373 LR 0.001000 Time 0.021588 +2023-10-02 20:34:25,415 - Epoch: [12][ 1236/ 1236] Overall Loss 0.341401 Objective Loss 0.341401 Top1 81.262729 Top5 98.574338 LR 0.001000 Time 0.021582 +2023-10-02 20:34:25,541 - --- validate (epoch=12)----------- +2023-10-02 20:34:25,541 - 29943 samples (256 per mini-batch) +2023-10-02 20:34:26,019 - Epoch: [12][ 10/ 117] Loss 0.377895 Top1 79.492188 Top5 97.890625 +2023-10-02 20:34:26,170 - Epoch: [12][ 20/ 117] Loss 0.369470 Top1 80.136719 Top5 97.871094 +2023-10-02 20:34:26,322 - Epoch: [12][ 30/ 117] Loss 0.357875 Top1 80.169271 Top5 97.838542 +2023-10-02 20:34:26,471 - Epoch: [12][ 40/ 117] Loss 0.352711 Top1 80.097656 Top5 97.714844 +2023-10-02 20:34:26,622 - Epoch: [12][ 50/ 117] Loss 0.348865 Top1 80.078125 Top5 97.671875 +2023-10-02 20:34:26,772 - Epoch: [12][ 60/ 117] Loss 0.344686 Top1 80.292969 Top5 97.682292 +2023-10-02 20:34:26,924 - Epoch: [12][ 70/ 117] Loss 0.347657 Top1 80.273438 Top5 97.667411 +2023-10-02 20:34:27,073 - Epoch: [12][ 80/ 117] Loss 0.347775 Top1 80.356445 Top5 97.666016 +2023-10-02 20:34:27,223 - Epoch: [12][ 90/ 117] Loss 0.349057 Top1 80.412326 Top5 97.638889 +2023-10-02 20:34:27,372 - Epoch: [12][ 100/ 117] Loss 0.345721 Top1 80.457031 Top5 97.601562 +2023-10-02 20:34:27,531 - Epoch: [12][ 110/ 117] Loss 0.346911 Top1 80.458097 Top5 97.610085 +2023-10-02 20:34:27,621 - Epoch: [12][ 117/ 117] Loss 0.347451 Top1 80.399426 Top5 97.618809 +2023-10-02 20:34:27,760 - ==> Top1: 80.399 Top5: 97.619 Loss: 0.347 + +2023-10-02 20:34:27,760 - ==> Confusion: +[[ 926 0 7 1 1 3 0 0 9 79 2 0 1 1 1 1 5 1 0 0 12] + [ 1 986 4 1 1 50 3 28 5 0 10 0 0 0 2 4 3 0 19 7 7] + [ 9 0 927 10 3 1 51 16 0 1 4 2 2 1 0 3 1 2 11 1 11] + [ 5 1 21 970 0 3 0 3 10 1 9 0 6 3 21 0 0 3 22 0 11] + [ 34 7 5 0 941 7 0 6 2 11 0 0 1 3 7 6 9 0 3 4 4] + [ 2 21 3 2 2 977 0 33 0 8 8 4 8 16 6 0 2 0 7 11 6] + [ 0 3 25 1 0 3 1110 11 0 0 3 1 0 1 0 5 0 0 2 21 5] + [ 4 8 15 0 2 26 0 1069 3 0 4 7 2 0 0 0 0 2 51 17 8] + [ 21 0 2 0 1 2 0 2 995 32 7 1 2 8 9 0 0 4 3 0 0] + [ 93 1 1 0 5 0 1 1 40 934 0 2 0 20 9 0 0 1 0 5 6] + [ 5 1 6 8 1 0 1 5 27 0 972 0 0 4 5 2 0 0 8 2 6] + [ 0 0 2 1 0 16 0 2 0 0 0 899 52 5 0 3 3 18 0 24 10] + [ 0 1 6 3 0 3 0 2 1 0 0 37 962 0 3 4 1 26 1 12 6] + [ 1 0 1 0 1 10 2 1 17 22 7 7 4 1026 2 0 0 0 0 8 10] + [ 11 1 2 15 4 0 0 0 35 10 6 0 3 2 986 0 0 3 9 0 14] + [ 1 1 5 1 2 0 3 2 0 0 0 16 7 1 0 1042 12 22 0 14 5] + [ 1 13 2 1 3 8 2 1 2 0 2 9 3 0 5 13 1070 0 0 14 12] + [ 0 0 0 5 0 0 2 1 0 0 0 3 16 0 3 7 0 992 0 3 6] + [ 1 3 7 12 0 1 1 15 9 0 8 0 0 0 11 0 0 1 986 5 8] + [ 0 0 1 1 1 6 10 12 0 0 2 5 0 1 0 1 1 1 4 1102 4] + [ 203 191 159 79 95 255 51 140 189 158 202 115 432 366 158 66 131 86 217 410 4202]] + +2023-10-02 20:34:27,762 - ==> Best [Top1: 80.399 Top5: 97.619 Sparsity:0.00 Params: 169472 on epoch: 12] +2023-10-02 20:34:27,762 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:34:27,775 - + +2023-10-02 20:34:27,776 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:34:28,799 - Epoch: [13][ 10/ 1236] Overall Loss 0.300415 Objective Loss 0.300415 LR 0.001000 Time 0.102284 +2023-10-02 20:34:29,007 - Epoch: [13][ 20/ 1236] Overall Loss 0.317811 Objective Loss 0.317811 LR 0.001000 Time 0.061521 +2023-10-02 20:34:29,214 - Epoch: [13][ 30/ 1236] Overall Loss 0.313549 Objective Loss 0.313549 LR 0.001000 Time 0.047900 +2023-10-02 20:34:29,422 - Epoch: [13][ 40/ 1236] Overall Loss 0.325850 Objective Loss 0.325850 LR 0.001000 Time 0.041112 +2023-10-02 20:34:29,628 - Epoch: [13][ 50/ 1236] Overall Loss 0.326195 Objective Loss 0.326195 LR 0.001000 Time 0.036986 +2023-10-02 20:34:29,836 - Epoch: [13][ 60/ 1236] Overall Loss 0.320370 Objective Loss 0.320370 LR 0.001000 Time 0.034282 +2023-10-02 20:34:30,043 - Epoch: [13][ 70/ 1236] Overall Loss 0.322188 Objective Loss 0.322188 LR 0.001000 Time 0.032314 +2023-10-02 20:34:30,251 - Epoch: [13][ 80/ 1236] Overall Loss 0.319008 Objective Loss 0.319008 LR 0.001000 Time 0.030881 +2023-10-02 20:34:30,457 - Epoch: [13][ 90/ 1236] Overall Loss 0.322749 Objective Loss 0.322749 LR 0.001000 Time 0.029728 +2023-10-02 20:34:30,666 - Epoch: [13][ 100/ 1236] Overall Loss 0.322734 Objective Loss 0.322734 LR 0.001000 Time 0.028841 +2023-10-02 20:34:30,871 - Epoch: [13][ 110/ 1236] Overall Loss 0.323258 Objective Loss 0.323258 LR 0.001000 Time 0.028084 +2023-10-02 20:34:31,080 - Epoch: [13][ 120/ 1236] Overall Loss 0.325366 Objective Loss 0.325366 LR 0.001000 Time 0.027483 +2023-10-02 20:34:31,286 - Epoch: [13][ 130/ 1236] Overall Loss 0.324533 Objective Loss 0.324533 LR 0.001000 Time 0.026947 +2023-10-02 20:34:31,495 - Epoch: [13][ 140/ 1236] Overall Loss 0.323847 Objective Loss 0.323847 LR 0.001000 Time 0.026515 +2023-10-02 20:34:31,700 - Epoch: [13][ 150/ 1236] Overall Loss 0.322679 Objective Loss 0.322679 LR 0.001000 Time 0.026115 +2023-10-02 20:34:31,909 - Epoch: [13][ 160/ 1236] Overall Loss 0.321838 Objective Loss 0.321838 LR 0.001000 Time 0.025788 +2023-10-02 20:34:32,114 - Epoch: [13][ 170/ 1236] Overall Loss 0.323840 Objective Loss 0.323840 LR 0.001000 Time 0.025476 +2023-10-02 20:34:32,324 - Epoch: [13][ 180/ 1236] Overall Loss 0.325705 Objective Loss 0.325705 LR 0.001000 Time 0.025222 +2023-10-02 20:34:32,529 - Epoch: [13][ 190/ 1236] Overall Loss 0.328207 Objective Loss 0.328207 LR 0.001000 Time 0.024973 +2023-10-02 20:34:32,737 - Epoch: [13][ 200/ 1236] Overall Loss 0.330661 Objective Loss 0.330661 LR 0.001000 Time 0.024763 +2023-10-02 20:34:32,943 - Epoch: [13][ 210/ 1236] Overall Loss 0.331100 Objective Loss 0.331100 LR 0.001000 Time 0.024559 +2023-10-02 20:34:33,152 - Epoch: [13][ 220/ 1236] Overall Loss 0.331696 Objective Loss 0.331696 LR 0.001000 Time 0.024390 +2023-10-02 20:34:33,359 - Epoch: [13][ 230/ 1236] Overall Loss 0.331790 Objective Loss 0.331790 LR 0.001000 Time 0.024230 +2023-10-02 20:34:33,568 - Epoch: [13][ 240/ 1236] Overall Loss 0.332742 Objective Loss 0.332742 LR 0.001000 Time 0.024090 +2023-10-02 20:34:33,775 - Epoch: [13][ 250/ 1236] Overall Loss 0.332050 Objective Loss 0.332050 LR 0.001000 Time 0.023954 +2023-10-02 20:34:33,985 - Epoch: [13][ 260/ 1236] Overall Loss 0.331491 Objective Loss 0.331491 LR 0.001000 Time 0.023835 +2023-10-02 20:34:34,192 - Epoch: [13][ 270/ 1236] Overall Loss 0.332404 Objective Loss 0.332404 LR 0.001000 Time 0.023718 +2023-10-02 20:34:34,401 - Epoch: [13][ 280/ 1236] Overall Loss 0.331691 Objective Loss 0.331691 LR 0.001000 Time 0.023619 +2023-10-02 20:34:34,608 - Epoch: [13][ 290/ 1236] Overall Loss 0.332780 Objective Loss 0.332780 LR 0.001000 Time 0.023517 +2023-10-02 20:34:34,817 - Epoch: [13][ 300/ 1236] Overall Loss 0.332719 Objective Loss 0.332719 LR 0.001000 Time 0.023429 +2023-10-02 20:34:35,025 - Epoch: [13][ 310/ 1236] Overall Loss 0.332676 Objective Loss 0.332676 LR 0.001000 Time 0.023341 +2023-10-02 20:34:35,232 - Epoch: [13][ 320/ 1236] Overall Loss 0.332596 Objective Loss 0.332596 LR 0.001000 Time 0.023258 +2023-10-02 20:34:35,441 - Epoch: [13][ 330/ 1236] Overall Loss 0.331998 Objective Loss 0.331998 LR 0.001000 Time 0.023186 +2023-10-02 20:34:35,651 - Epoch: [13][ 340/ 1236] Overall Loss 0.332200 Objective Loss 0.332200 LR 0.001000 Time 0.023121 +2023-10-02 20:34:35,858 - Epoch: [13][ 350/ 1236] Overall Loss 0.331693 Objective Loss 0.331693 LR 0.001000 Time 0.023051 +2023-10-02 20:34:36,068 - Epoch: [13][ 360/ 1236] Overall Loss 0.330915 Objective Loss 0.330915 LR 0.001000 Time 0.022993 +2023-10-02 20:34:36,275 - Epoch: [13][ 370/ 1236] Overall Loss 0.330925 Objective Loss 0.330925 LR 0.001000 Time 0.022931 +2023-10-02 20:34:36,485 - Epoch: [13][ 380/ 1236] Overall Loss 0.330639 Objective Loss 0.330639 LR 0.001000 Time 0.022879 +2023-10-02 20:34:36,692 - Epoch: [13][ 390/ 1236] Overall Loss 0.329640 Objective Loss 0.329640 LR 0.001000 Time 0.022822 +2023-10-02 20:34:36,902 - Epoch: [13][ 400/ 1236] Overall Loss 0.328906 Objective Loss 0.328906 LR 0.001000 Time 0.022776 +2023-10-02 20:34:37,110 - Epoch: [13][ 410/ 1236] Overall Loss 0.328131 Objective Loss 0.328131 LR 0.001000 Time 0.022725 +2023-10-02 20:34:37,320 - Epoch: [13][ 420/ 1236] Overall Loss 0.327700 Objective Loss 0.327700 LR 0.001000 Time 0.022684 +2023-10-02 20:34:37,527 - Epoch: [13][ 430/ 1236] Overall Loss 0.326922 Objective Loss 0.326922 LR 0.001000 Time 0.022637 +2023-10-02 20:34:37,741 - Epoch: [13][ 440/ 1236] Overall Loss 0.326692 Objective Loss 0.326692 LR 0.001000 Time 0.022607 +2023-10-02 20:34:37,947 - Epoch: [13][ 450/ 1236] Overall Loss 0.327912 Objective Loss 0.327912 LR 0.001000 Time 0.022564 +2023-10-02 20:34:38,157 - Epoch: [13][ 460/ 1236] Overall Loss 0.327291 Objective Loss 0.327291 LR 0.001000 Time 0.022528 +2023-10-02 20:34:38,364 - Epoch: [13][ 470/ 1236] Overall Loss 0.327441 Objective Loss 0.327441 LR 0.001000 Time 0.022489 +2023-10-02 20:34:38,573 - Epoch: [13][ 480/ 1236] Overall Loss 0.326234 Objective Loss 0.326234 LR 0.001000 Time 0.022456 +2023-10-02 20:34:38,780 - Epoch: [13][ 490/ 1236] Overall Loss 0.325861 Objective Loss 0.325861 LR 0.001000 Time 0.022419 +2023-10-02 20:34:38,990 - Epoch: [13][ 500/ 1236] Overall Loss 0.326049 Objective Loss 0.326049 LR 0.001000 Time 0.022390 +2023-10-02 20:34:39,197 - Epoch: [13][ 510/ 1236] Overall Loss 0.325994 Objective Loss 0.325994 LR 0.001000 Time 0.022355 +2023-10-02 20:34:39,407 - Epoch: [13][ 520/ 1236] Overall Loss 0.326280 Objective Loss 0.326280 LR 0.001000 Time 0.022329 +2023-10-02 20:34:39,614 - Epoch: [13][ 530/ 1236] Overall Loss 0.326530 Objective Loss 0.326530 LR 0.001000 Time 0.022297 +2023-10-02 20:34:39,824 - Epoch: [13][ 540/ 1236] Overall Loss 0.326516 Objective Loss 0.326516 LR 0.001000 Time 0.022272 +2023-10-02 20:34:40,030 - Epoch: [13][ 550/ 1236] Overall Loss 0.325893 Objective Loss 0.325893 LR 0.001000 Time 0.022243 +2023-10-02 20:34:40,240 - Epoch: [13][ 560/ 1236] Overall Loss 0.326213 Objective Loss 0.326213 LR 0.001000 Time 0.022219 +2023-10-02 20:34:40,447 - Epoch: [13][ 570/ 1236] Overall Loss 0.326092 Objective Loss 0.326092 LR 0.001000 Time 0.022192 +2023-10-02 20:34:40,657 - Epoch: [13][ 580/ 1236] Overall Loss 0.326609 Objective Loss 0.326609 LR 0.001000 Time 0.022170 +2023-10-02 20:34:40,864 - Epoch: [13][ 590/ 1236] Overall Loss 0.327456 Objective Loss 0.327456 LR 0.001000 Time 0.022145 +2023-10-02 20:34:41,074 - Epoch: [13][ 600/ 1236] Overall Loss 0.327237 Objective Loss 0.327237 LR 0.001000 Time 0.022126 +2023-10-02 20:34:41,282 - Epoch: [13][ 610/ 1236] Overall Loss 0.327839 Objective Loss 0.327839 LR 0.001000 Time 0.022103 +2023-10-02 20:34:41,492 - Epoch: [13][ 620/ 1236] Overall Loss 0.327774 Objective Loss 0.327774 LR 0.001000 Time 0.022086 +2023-10-02 20:34:41,700 - Epoch: [13][ 630/ 1236] Overall Loss 0.327975 Objective Loss 0.327975 LR 0.001000 Time 0.022065 +2023-10-02 20:34:41,911 - Epoch: [13][ 640/ 1236] Overall Loss 0.328231 Objective Loss 0.328231 LR 0.001000 Time 0.022048 +2023-10-02 20:34:42,118 - Epoch: [13][ 650/ 1236] Overall Loss 0.327906 Objective Loss 0.327906 LR 0.001000 Time 0.022027 +2023-10-02 20:34:42,329 - Epoch: [13][ 660/ 1236] Overall Loss 0.328400 Objective Loss 0.328400 LR 0.001000 Time 0.022012 +2023-10-02 20:34:42,536 - Epoch: [13][ 670/ 1236] Overall Loss 0.328572 Objective Loss 0.328572 LR 0.001000 Time 0.021993 +2023-10-02 20:34:42,747 - Epoch: [13][ 680/ 1236] Overall Loss 0.328511 Objective Loss 0.328511 LR 0.001000 Time 0.021978 +2023-10-02 20:34:42,954 - Epoch: [13][ 690/ 1236] Overall Loss 0.328369 Objective Loss 0.328369 LR 0.001000 Time 0.021960 +2023-10-02 20:34:43,164 - Epoch: [13][ 700/ 1236] Overall Loss 0.327652 Objective Loss 0.327652 LR 0.001000 Time 0.021947 +2023-10-02 20:34:43,372 - Epoch: [13][ 710/ 1236] Overall Loss 0.327802 Objective Loss 0.327802 LR 0.001000 Time 0.021929 +2023-10-02 20:34:43,582 - Epoch: [13][ 720/ 1236] Overall Loss 0.327535 Objective Loss 0.327535 LR 0.001000 Time 0.021916 +2023-10-02 20:34:43,790 - Epoch: [13][ 730/ 1236] Overall Loss 0.327294 Objective Loss 0.327294 LR 0.001000 Time 0.021900 +2023-10-02 20:34:44,000 - Epoch: [13][ 740/ 1236] Overall Loss 0.327514 Objective Loss 0.327514 LR 0.001000 Time 0.021888 +2023-10-02 20:34:44,208 - Epoch: [13][ 750/ 1236] Overall Loss 0.327511 Objective Loss 0.327511 LR 0.001000 Time 0.021873 +2023-10-02 20:34:44,418 - Epoch: [13][ 760/ 1236] Overall Loss 0.327544 Objective Loss 0.327544 LR 0.001000 Time 0.021861 +2023-10-02 20:34:44,626 - Epoch: [13][ 770/ 1236] Overall Loss 0.327497 Objective Loss 0.327497 LR 0.001000 Time 0.021846 +2023-10-02 20:34:44,836 - Epoch: [13][ 780/ 1236] Overall Loss 0.327876 Objective Loss 0.327876 LR 0.001000 Time 0.021836 +2023-10-02 20:34:45,044 - Epoch: [13][ 790/ 1236] Overall Loss 0.327747 Objective Loss 0.327747 LR 0.001000 Time 0.021822 +2023-10-02 20:34:45,254 - Epoch: [13][ 800/ 1236] Overall Loss 0.327644 Objective Loss 0.327644 LR 0.001000 Time 0.021812 +2023-10-02 20:34:45,462 - Epoch: [13][ 810/ 1236] Overall Loss 0.327441 Objective Loss 0.327441 LR 0.001000 Time 0.021798 +2023-10-02 20:34:45,672 - Epoch: [13][ 820/ 1236] Overall Loss 0.328163 Objective Loss 0.328163 LR 0.001000 Time 0.021788 +2023-10-02 20:34:45,880 - Epoch: [13][ 830/ 1236] Overall Loss 0.328176 Objective Loss 0.328176 LR 0.001000 Time 0.021775 +2023-10-02 20:34:46,090 - Epoch: [13][ 840/ 1236] Overall Loss 0.328137 Objective Loss 0.328137 LR 0.001000 Time 0.021766 +2023-10-02 20:34:46,297 - Epoch: [13][ 850/ 1236] Overall Loss 0.328084 Objective Loss 0.328084 LR 0.001000 Time 0.021754 +2023-10-02 20:34:46,508 - Epoch: [13][ 860/ 1236] Overall Loss 0.327964 Objective Loss 0.327964 LR 0.001000 Time 0.021745 +2023-10-02 20:34:46,715 - Epoch: [13][ 870/ 1236] Overall Loss 0.328354 Objective Loss 0.328354 LR 0.001000 Time 0.021733 +2023-10-02 20:34:46,925 - Epoch: [13][ 880/ 1236] Overall Loss 0.328429 Objective Loss 0.328429 LR 0.001000 Time 0.021725 +2023-10-02 20:34:47,133 - Epoch: [13][ 890/ 1236] Overall Loss 0.328475 Objective Loss 0.328475 LR 0.001000 Time 0.021713 +2023-10-02 20:34:47,343 - Epoch: [13][ 900/ 1236] Overall Loss 0.328349 Objective Loss 0.328349 LR 0.001000 Time 0.021705 +2023-10-02 20:34:47,551 - Epoch: [13][ 910/ 1236] Overall Loss 0.328194 Objective Loss 0.328194 LR 0.001000 Time 0.021695 +2023-10-02 20:34:47,761 - Epoch: [13][ 920/ 1236] Overall Loss 0.328325 Objective Loss 0.328325 LR 0.001000 Time 0.021687 +2023-10-02 20:34:47,969 - Epoch: [13][ 930/ 1236] Overall Loss 0.328262 Objective Loss 0.328262 LR 0.001000 Time 0.021677 +2023-10-02 20:34:48,179 - Epoch: [13][ 940/ 1236] Overall Loss 0.328375 Objective Loss 0.328375 LR 0.001000 Time 0.021669 +2023-10-02 20:34:48,386 - Epoch: [13][ 950/ 1236] Overall Loss 0.328853 Objective Loss 0.328853 LR 0.001000 Time 0.021659 +2023-10-02 20:34:48,596 - Epoch: [13][ 960/ 1236] Overall Loss 0.328658 Objective Loss 0.328658 LR 0.001000 Time 0.021652 +2023-10-02 20:34:48,804 - Epoch: [13][ 970/ 1236] Overall Loss 0.328689 Objective Loss 0.328689 LR 0.001000 Time 0.021643 +2023-10-02 20:34:49,014 - Epoch: [13][ 980/ 1236] Overall Loss 0.328505 Objective Loss 0.328505 LR 0.001000 Time 0.021636 +2023-10-02 20:34:49,221 - Epoch: [13][ 990/ 1236] Overall Loss 0.328629 Objective Loss 0.328629 LR 0.001000 Time 0.021626 +2023-10-02 20:34:49,431 - Epoch: [13][ 1000/ 1236] Overall Loss 0.328844 Objective Loss 0.328844 LR 0.001000 Time 0.021620 +2023-10-02 20:34:49,639 - Epoch: [13][ 1010/ 1236] Overall Loss 0.328953 Objective Loss 0.328953 LR 0.001000 Time 0.021611 +2023-10-02 20:34:49,849 - Epoch: [13][ 1020/ 1236] Overall Loss 0.328967 Objective Loss 0.328967 LR 0.001000 Time 0.021604 +2023-10-02 20:34:50,056 - Epoch: [13][ 1030/ 1236] Overall Loss 0.328741 Objective Loss 0.328741 LR 0.001000 Time 0.021596 +2023-10-02 20:34:50,266 - Epoch: [13][ 1040/ 1236] Overall Loss 0.328805 Objective Loss 0.328805 LR 0.001000 Time 0.021590 +2023-10-02 20:34:50,474 - Epoch: [13][ 1050/ 1236] Overall Loss 0.328913 Objective Loss 0.328913 LR 0.001000 Time 0.021581 +2023-10-02 20:34:50,684 - Epoch: [13][ 1060/ 1236] Overall Loss 0.328677 Objective Loss 0.328677 LR 0.001000 Time 0.021576 +2023-10-02 20:34:50,891 - Epoch: [13][ 1070/ 1236] Overall Loss 0.328476 Objective Loss 0.328476 LR 0.001000 Time 0.021568 +2023-10-02 20:34:51,101 - Epoch: [13][ 1080/ 1236] Overall Loss 0.328640 Objective Loss 0.328640 LR 0.001000 Time 0.021562 +2023-10-02 20:34:51,309 - Epoch: [13][ 1090/ 1236] Overall Loss 0.328529 Objective Loss 0.328529 LR 0.001000 Time 0.021555 +2023-10-02 20:34:51,520 - Epoch: [13][ 1100/ 1236] Overall Loss 0.328919 Objective Loss 0.328919 LR 0.001000 Time 0.021550 +2023-10-02 20:34:51,728 - Epoch: [13][ 1110/ 1236] Overall Loss 0.329208 Objective Loss 0.329208 LR 0.001000 Time 0.021543 +2023-10-02 20:34:51,940 - Epoch: [13][ 1120/ 1236] Overall Loss 0.329220 Objective Loss 0.329220 LR 0.001000 Time 0.021539 +2023-10-02 20:34:52,148 - Epoch: [13][ 1130/ 1236] Overall Loss 0.329380 Objective Loss 0.329380 LR 0.001000 Time 0.021533 +2023-10-02 20:34:52,359 - Epoch: [13][ 1140/ 1236] Overall Loss 0.329822 Objective Loss 0.329822 LR 0.001000 Time 0.021529 +2023-10-02 20:34:52,567 - Epoch: [13][ 1150/ 1236] Overall Loss 0.329873 Objective Loss 0.329873 LR 0.001000 Time 0.021522 +2023-10-02 20:34:52,778 - Epoch: [13][ 1160/ 1236] Overall Loss 0.330110 Objective Loss 0.330110 LR 0.001000 Time 0.021519 +2023-10-02 20:34:52,986 - Epoch: [13][ 1170/ 1236] Overall Loss 0.330015 Objective Loss 0.330015 LR 0.001000 Time 0.021512 +2023-10-02 20:34:53,197 - Epoch: [13][ 1180/ 1236] Overall Loss 0.329725 Objective Loss 0.329725 LR 0.001000 Time 0.021508 +2023-10-02 20:34:53,405 - Epoch: [13][ 1190/ 1236] Overall Loss 0.329796 Objective Loss 0.329796 LR 0.001000 Time 0.021502 +2023-10-02 20:34:53,617 - Epoch: [13][ 1200/ 1236] Overall Loss 0.329648 Objective Loss 0.329648 LR 0.001000 Time 0.021499 +2023-10-02 20:34:53,824 - Epoch: [13][ 1210/ 1236] Overall Loss 0.329709 Objective Loss 0.329709 LR 0.001000 Time 0.021493 +2023-10-02 20:34:54,036 - Epoch: [13][ 1220/ 1236] Overall Loss 0.330039 Objective Loss 0.330039 LR 0.001000 Time 0.021490 +2023-10-02 20:34:54,298 - Epoch: [13][ 1230/ 1236] Overall Loss 0.330154 Objective Loss 0.330154 LR 0.001000 Time 0.021528 +2023-10-02 20:34:54,421 - Epoch: [13][ 1236/ 1236] Overall Loss 0.330080 Objective Loss 0.330080 Top1 84.317719 Top5 97.148676 LR 0.001000 Time 0.021522 +2023-10-02 20:34:54,556 - --- validate (epoch=13)----------- +2023-10-02 20:34:54,556 - 29943 samples (256 per mini-batch) +2023-10-02 20:34:55,046 - Epoch: [13][ 10/ 117] Loss 0.360513 Top1 80.937500 Top5 97.304688 +2023-10-02 20:34:55,193 - Epoch: [13][ 20/ 117] Loss 0.352335 Top1 81.074219 Top5 97.343750 +2023-10-02 20:34:55,339 - Epoch: [13][ 30/ 117] Loss 0.346915 Top1 81.054688 Top5 97.708333 +2023-10-02 20:34:55,486 - Epoch: [13][ 40/ 117] Loss 0.347576 Top1 81.083984 Top5 97.675781 +2023-10-02 20:34:55,631 - Epoch: [13][ 50/ 117] Loss 0.354303 Top1 80.921875 Top5 97.640625 +2023-10-02 20:34:55,776 - Epoch: [13][ 60/ 117] Loss 0.347239 Top1 81.100260 Top5 97.688802 +2023-10-02 20:34:55,920 - Epoch: [13][ 70/ 117] Loss 0.348174 Top1 81.015625 Top5 97.667411 +2023-10-02 20:34:56,065 - Epoch: [13][ 80/ 117] Loss 0.344435 Top1 81.074219 Top5 97.719727 +2023-10-02 20:34:56,217 - Epoch: [13][ 90/ 117] Loss 0.345271 Top1 80.933160 Top5 97.717014 +2023-10-02 20:34:56,372 - Epoch: [13][ 100/ 117] Loss 0.346588 Top1 80.902344 Top5 97.738281 +2023-10-02 20:34:56,534 - Epoch: [13][ 110/ 117] Loss 0.349322 Top1 80.823864 Top5 97.702415 +2023-10-02 20:34:56,623 - Epoch: [13][ 117/ 117] Loss 0.347227 Top1 80.826904 Top5 97.708980 +2023-10-02 20:34:56,766 - ==> Top1: 80.827 Top5: 97.709 Loss: 0.347 + +2023-10-02 20:34:56,767 - ==> Confusion: +[[ 905 1 2 0 6 2 0 0 9 88 1 3 1 5 4 3 6 3 0 0 11] + [ 2 997 2 0 5 51 4 26 9 0 8 1 1 0 2 1 0 0 12 3 7] + [ 11 0 882 20 3 3 75 11 1 3 6 1 10 0 0 3 2 1 7 6 11] + [ 1 2 14 931 0 6 4 0 12 2 16 0 8 5 55 2 0 6 15 0 10] + [ 22 11 0 0 944 12 1 1 1 17 0 1 1 4 16 2 5 1 2 3 6] + [ 0 20 1 2 1 981 1 19 8 2 6 8 2 33 11 0 2 0 3 6 10] + [ 0 1 17 0 0 0 1136 8 0 0 9 2 3 2 0 1 0 1 1 7 3] + [ 4 13 13 0 2 20 6 1061 3 0 8 6 6 3 4 0 0 0 51 10 8] + [ 20 1 1 0 0 0 0 0 982 41 2 0 3 9 20 1 0 2 6 0 1] + [ 91 0 0 0 2 0 1 0 30 953 0 0 2 24 3 0 1 1 0 3 8] + [ 1 4 3 14 0 2 1 3 37 0 948 3 1 15 10 0 0 0 5 0 6] + [ 0 0 2 0 1 14 1 2 0 1 0 965 12 8 0 1 2 15 0 6 5] + [ 0 3 2 2 0 3 2 1 1 0 0 51 957 2 6 7 4 17 2 4 4] + [ 0 0 1 0 1 10 0 0 25 17 9 5 2 1030 5 0 0 2 0 3 9] + [ 8 0 2 7 9 0 0 0 30 12 2 0 2 4 1008 0 0 1 4 0 12] + [ 0 0 2 4 5 0 5 0 0 0 0 13 7 0 0 1062 16 15 3 0 2] + [ 1 14 0 0 7 10 3 0 2 0 1 7 0 0 6 11 1082 0 0 1 16] + [ 0 0 0 1 0 1 2 0 1 0 0 7 12 0 4 9 0 999 1 0 1] + [ 1 6 3 13 1 0 1 14 14 0 9 0 2 0 18 1 0 0 976 0 9] + [ 0 1 0 1 4 8 7 13 0 0 3 21 9 4 1 3 5 0 2 1064 6] + [ 176 166 110 79 101 288 68 96 166 118 204 215 402 337 255 74 146 79 221 265 4339]] + +2023-10-02 20:34:56,768 - ==> Best [Top1: 80.827 Top5: 97.709 Sparsity:0.00 Params: 169472 on epoch: 13] +2023-10-02 20:34:56,768 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:34:56,782 - + +2023-10-02 20:34:56,782 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:34:57,784 - Epoch: [14][ 10/ 1236] Overall Loss 0.316541 Objective Loss 0.316541 LR 0.001000 Time 0.100177 +2023-10-02 20:34:57,994 - Epoch: [14][ 20/ 1236] Overall Loss 0.326926 Objective Loss 0.326926 LR 0.001000 Time 0.060531 +2023-10-02 20:34:58,201 - Epoch: [14][ 30/ 1236] Overall Loss 0.324688 Objective Loss 0.324688 LR 0.001000 Time 0.047228 +2023-10-02 20:34:58,411 - Epoch: [14][ 40/ 1236] Overall Loss 0.319012 Objective Loss 0.319012 LR 0.001000 Time 0.040662 +2023-10-02 20:34:58,622 - Epoch: [14][ 50/ 1236] Overall Loss 0.321918 Objective Loss 0.321918 LR 0.001000 Time 0.036736 +2023-10-02 20:34:58,836 - Epoch: [14][ 60/ 1236] Overall Loss 0.324693 Objective Loss 0.324693 LR 0.001000 Time 0.034171 +2023-10-02 20:34:59,046 - Epoch: [14][ 70/ 1236] Overall Loss 0.324027 Objective Loss 0.324027 LR 0.001000 Time 0.032284 +2023-10-02 20:34:59,259 - Epoch: [14][ 80/ 1236] Overall Loss 0.319484 Objective Loss 0.319484 LR 0.001000 Time 0.030904 +2023-10-02 20:34:59,470 - Epoch: [14][ 90/ 1236] Overall Loss 0.313069 Objective Loss 0.313069 LR 0.001000 Time 0.029807 +2023-10-02 20:34:59,687 - Epoch: [14][ 100/ 1236] Overall Loss 0.315146 Objective Loss 0.315146 LR 0.001000 Time 0.028992 +2023-10-02 20:34:59,897 - Epoch: [14][ 110/ 1236] Overall Loss 0.316712 Objective Loss 0.316712 LR 0.001000 Time 0.028268 +2023-10-02 20:35:00,111 - Epoch: [14][ 120/ 1236] Overall Loss 0.319932 Objective Loss 0.319932 LR 0.001000 Time 0.027692 +2023-10-02 20:35:00,322 - Epoch: [14][ 130/ 1236] Overall Loss 0.320797 Objective Loss 0.320797 LR 0.001000 Time 0.027181 +2023-10-02 20:35:00,535 - Epoch: [14][ 140/ 1236] Overall Loss 0.321488 Objective Loss 0.321488 LR 0.001000 Time 0.026755 +2023-10-02 20:35:00,746 - Epoch: [14][ 150/ 1236] Overall Loss 0.321387 Objective Loss 0.321387 LR 0.001000 Time 0.026377 +2023-10-02 20:35:00,958 - Epoch: [14][ 160/ 1236] Overall Loss 0.318508 Objective Loss 0.318508 LR 0.001000 Time 0.026052 +2023-10-02 20:35:01,165 - Epoch: [14][ 170/ 1236] Overall Loss 0.319318 Objective Loss 0.319318 LR 0.001000 Time 0.025734 +2023-10-02 20:35:01,377 - Epoch: [14][ 180/ 1236] Overall Loss 0.320103 Objective Loss 0.320103 LR 0.001000 Time 0.025481 +2023-10-02 20:35:01,586 - Epoch: [14][ 190/ 1236] Overall Loss 0.318507 Objective Loss 0.318507 LR 0.001000 Time 0.025239 +2023-10-02 20:35:01,796 - Epoch: [14][ 200/ 1236] Overall Loss 0.317531 Objective Loss 0.317531 LR 0.001000 Time 0.025024 +2023-10-02 20:35:02,005 - Epoch: [14][ 210/ 1236] Overall Loss 0.317495 Objective Loss 0.317495 LR 0.001000 Time 0.024826 +2023-10-02 20:35:02,218 - Epoch: [14][ 220/ 1236] Overall Loss 0.318050 Objective Loss 0.318050 LR 0.001000 Time 0.024661 +2023-10-02 20:35:02,427 - Epoch: [14][ 230/ 1236] Overall Loss 0.317502 Objective Loss 0.317502 LR 0.001000 Time 0.024497 +2023-10-02 20:35:02,639 - Epoch: [14][ 240/ 1236] Overall Loss 0.316478 Objective Loss 0.316478 LR 0.001000 Time 0.024359 +2023-10-02 20:35:02,848 - Epoch: [14][ 250/ 1236] Overall Loss 0.316110 Objective Loss 0.316110 LR 0.001000 Time 0.024220 +2023-10-02 20:35:03,060 - Epoch: [14][ 260/ 1236] Overall Loss 0.315280 Objective Loss 0.315280 LR 0.001000 Time 0.024101 +2023-10-02 20:35:03,269 - Epoch: [14][ 270/ 1236] Overall Loss 0.314408 Objective Loss 0.314408 LR 0.001000 Time 0.023981 +2023-10-02 20:35:03,481 - Epoch: [14][ 280/ 1236] Overall Loss 0.314277 Objective Loss 0.314277 LR 0.001000 Time 0.023880 +2023-10-02 20:35:03,689 - Epoch: [14][ 290/ 1236] Overall Loss 0.313635 Objective Loss 0.313635 LR 0.001000 Time 0.023773 +2023-10-02 20:35:03,901 - Epoch: [14][ 300/ 1236] Overall Loss 0.313935 Objective Loss 0.313935 LR 0.001000 Time 0.023685 +2023-10-02 20:35:04,110 - Epoch: [14][ 310/ 1236] Overall Loss 0.313897 Objective Loss 0.313897 LR 0.001000 Time 0.023595 +2023-10-02 20:35:04,321 - Epoch: [14][ 320/ 1236] Overall Loss 0.313103 Objective Loss 0.313103 LR 0.001000 Time 0.023516 +2023-10-02 20:35:04,530 - Epoch: [14][ 330/ 1236] Overall Loss 0.313059 Objective Loss 0.313059 LR 0.001000 Time 0.023437 +2023-10-02 20:35:04,742 - Epoch: [14][ 340/ 1236] Overall Loss 0.312492 Objective Loss 0.312492 LR 0.001000 Time 0.023368 +2023-10-02 20:35:04,952 - Epoch: [14][ 350/ 1236] Overall Loss 0.313829 Objective Loss 0.313829 LR 0.001000 Time 0.023301 +2023-10-02 20:35:05,166 - Epoch: [14][ 360/ 1236] Overall Loss 0.314612 Objective Loss 0.314612 LR 0.001000 Time 0.023245 +2023-10-02 20:35:05,376 - Epoch: [14][ 370/ 1236] Overall Loss 0.315154 Objective Loss 0.315154 LR 0.001000 Time 0.023184 +2023-10-02 20:35:05,591 - Epoch: [14][ 380/ 1236] Overall Loss 0.315238 Objective Loss 0.315238 LR 0.001000 Time 0.023138 +2023-10-02 20:35:05,801 - Epoch: [14][ 390/ 1236] Overall Loss 0.314849 Objective Loss 0.314849 LR 0.001000 Time 0.023082 +2023-10-02 20:35:06,014 - Epoch: [14][ 400/ 1236] Overall Loss 0.314928 Objective Loss 0.314928 LR 0.001000 Time 0.023038 +2023-10-02 20:35:06,225 - Epoch: [14][ 410/ 1236] Overall Loss 0.316295 Objective Loss 0.316295 LR 0.001000 Time 0.022987 +2023-10-02 20:35:06,438 - Epoch: [14][ 420/ 1236] Overall Loss 0.317439 Objective Loss 0.317439 LR 0.001000 Time 0.022947 +2023-10-02 20:35:06,648 - Epoch: [14][ 430/ 1236] Overall Loss 0.317023 Objective Loss 0.317023 LR 0.001000 Time 0.022901 +2023-10-02 20:35:06,862 - Epoch: [14][ 440/ 1236] Overall Loss 0.317185 Objective Loss 0.317185 LR 0.001000 Time 0.022865 +2023-10-02 20:35:07,072 - Epoch: [14][ 450/ 1236] Overall Loss 0.317550 Objective Loss 0.317550 LR 0.001000 Time 0.022822 +2023-10-02 20:35:07,284 - Epoch: [14][ 460/ 1236] Overall Loss 0.318195 Objective Loss 0.318195 LR 0.001000 Time 0.022787 +2023-10-02 20:35:07,494 - Epoch: [14][ 470/ 1236] Overall Loss 0.318351 Objective Loss 0.318351 LR 0.001000 Time 0.022750 +2023-10-02 20:35:07,708 - Epoch: [14][ 480/ 1236] Overall Loss 0.317735 Objective Loss 0.317735 LR 0.001000 Time 0.022719 +2023-10-02 20:35:07,918 - Epoch: [14][ 490/ 1236] Overall Loss 0.317303 Objective Loss 0.317303 LR 0.001000 Time 0.022684 +2023-10-02 20:35:08,131 - Epoch: [14][ 500/ 1236] Overall Loss 0.317206 Objective Loss 0.317206 LR 0.001000 Time 0.022656 +2023-10-02 20:35:08,342 - Epoch: [14][ 510/ 1236] Overall Loss 0.316864 Objective Loss 0.316864 LR 0.001000 Time 0.022624 +2023-10-02 20:35:08,556 - Epoch: [14][ 520/ 1236] Overall Loss 0.317460 Objective Loss 0.317460 LR 0.001000 Time 0.022599 +2023-10-02 20:35:08,766 - Epoch: [14][ 530/ 1236] Overall Loss 0.317501 Objective Loss 0.317501 LR 0.001000 Time 0.022569 +2023-10-02 20:35:08,980 - Epoch: [14][ 540/ 1236] Overall Loss 0.318551 Objective Loss 0.318551 LR 0.001000 Time 0.022546 +2023-10-02 20:35:09,190 - Epoch: [14][ 550/ 1236] Overall Loss 0.319157 Objective Loss 0.319157 LR 0.001000 Time 0.022517 +2023-10-02 20:35:09,403 - Epoch: [14][ 560/ 1236] Overall Loss 0.318038 Objective Loss 0.318038 LR 0.001000 Time 0.022495 +2023-10-02 20:35:09,613 - Epoch: [14][ 570/ 1236] Overall Loss 0.317960 Objective Loss 0.317960 LR 0.001000 Time 0.022468 +2023-10-02 20:35:09,826 - Epoch: [14][ 580/ 1236] Overall Loss 0.318497 Objective Loss 0.318497 LR 0.001000 Time 0.022447 +2023-10-02 20:35:10,037 - Epoch: [14][ 590/ 1236] Overall Loss 0.319216 Objective Loss 0.319216 LR 0.001000 Time 0.022423 +2023-10-02 20:35:10,250 - Epoch: [14][ 600/ 1236] Overall Loss 0.319216 Objective Loss 0.319216 LR 0.001000 Time 0.022404 +2023-10-02 20:35:10,460 - Epoch: [14][ 610/ 1236] Overall Loss 0.319614 Objective Loss 0.319614 LR 0.001000 Time 0.022381 +2023-10-02 20:35:10,674 - Epoch: [14][ 620/ 1236] Overall Loss 0.319400 Objective Loss 0.319400 LR 0.001000 Time 0.022363 +2023-10-02 20:35:10,884 - Epoch: [14][ 630/ 1236] Overall Loss 0.319509 Objective Loss 0.319509 LR 0.001000 Time 0.022341 +2023-10-02 20:35:11,097 - Epoch: [14][ 640/ 1236] Overall Loss 0.319534 Objective Loss 0.319534 LR 0.001000 Time 0.022325 +2023-10-02 20:35:11,308 - Epoch: [14][ 650/ 1236] Overall Loss 0.320153 Objective Loss 0.320153 LR 0.001000 Time 0.022304 +2023-10-02 20:35:11,521 - Epoch: [14][ 660/ 1236] Overall Loss 0.320475 Objective Loss 0.320475 LR 0.001000 Time 0.022289 +2023-10-02 20:35:11,732 - Epoch: [14][ 670/ 1236] Overall Loss 0.320732 Objective Loss 0.320732 LR 0.001000 Time 0.022270 +2023-10-02 20:35:11,945 - Epoch: [14][ 680/ 1236] Overall Loss 0.320980 Objective Loss 0.320980 LR 0.001000 Time 0.022256 +2023-10-02 20:35:12,155 - Epoch: [14][ 690/ 1236] Overall Loss 0.320891 Objective Loss 0.320891 LR 0.001000 Time 0.022237 +2023-10-02 20:35:12,369 - Epoch: [14][ 700/ 1236] Overall Loss 0.320660 Objective Loss 0.320660 LR 0.001000 Time 0.022224 +2023-10-02 20:35:12,579 - Epoch: [14][ 710/ 1236] Overall Loss 0.320637 Objective Loss 0.320637 LR 0.001000 Time 0.022206 +2023-10-02 20:35:12,792 - Epoch: [14][ 720/ 1236] Overall Loss 0.320685 Objective Loss 0.320685 LR 0.001000 Time 0.022193 +2023-10-02 20:35:13,002 - Epoch: [14][ 730/ 1236] Overall Loss 0.320922 Objective Loss 0.320922 LR 0.001000 Time 0.022177 +2023-10-02 20:35:13,216 - Epoch: [14][ 740/ 1236] Overall Loss 0.321161 Objective Loss 0.321161 LR 0.001000 Time 0.022165 +2023-10-02 20:35:13,426 - Epoch: [14][ 750/ 1236] Overall Loss 0.321589 Objective Loss 0.321589 LR 0.001000 Time 0.022149 +2023-10-02 20:35:13,640 - Epoch: [14][ 760/ 1236] Overall Loss 0.322113 Objective Loss 0.322113 LR 0.001000 Time 0.022138 +2023-10-02 20:35:13,850 - Epoch: [14][ 770/ 1236] Overall Loss 0.322213 Objective Loss 0.322213 LR 0.001000 Time 0.022123 +2023-10-02 20:35:14,063 - Epoch: [14][ 780/ 1236] Overall Loss 0.322192 Objective Loss 0.322192 LR 0.001000 Time 0.022112 +2023-10-02 20:35:14,273 - Epoch: [14][ 790/ 1236] Overall Loss 0.322695 Objective Loss 0.322695 LR 0.001000 Time 0.022098 +2023-10-02 20:35:14,487 - Epoch: [14][ 800/ 1236] Overall Loss 0.322503 Objective Loss 0.322503 LR 0.001000 Time 0.022088 +2023-10-02 20:35:14,697 - Epoch: [14][ 810/ 1236] Overall Loss 0.322274 Objective Loss 0.322274 LR 0.001000 Time 0.022074 +2023-10-02 20:35:14,911 - Epoch: [14][ 820/ 1236] Overall Loss 0.321733 Objective Loss 0.321733 LR 0.001000 Time 0.022065 +2023-10-02 20:35:15,121 - Epoch: [14][ 830/ 1236] Overall Loss 0.321579 Objective Loss 0.321579 LR 0.001000 Time 0.022053 +2023-10-02 20:35:15,335 - Epoch: [14][ 840/ 1236] Overall Loss 0.321157 Objective Loss 0.321157 LR 0.001000 Time 0.022044 +2023-10-02 20:35:15,545 - Epoch: [14][ 850/ 1236] Overall Loss 0.321208 Objective Loss 0.321208 LR 0.001000 Time 0.022031 +2023-10-02 20:35:15,759 - Epoch: [14][ 860/ 1236] Overall Loss 0.321282 Objective Loss 0.321282 LR 0.001000 Time 0.022023 +2023-10-02 20:35:15,969 - Epoch: [14][ 870/ 1236] Overall Loss 0.320962 Objective Loss 0.320962 LR 0.001000 Time 0.022011 +2023-10-02 20:35:16,183 - Epoch: [14][ 880/ 1236] Overall Loss 0.320333 Objective Loss 0.320333 LR 0.001000 Time 0.022003 +2023-10-02 20:35:16,393 - Epoch: [14][ 890/ 1236] Overall Loss 0.320401 Objective Loss 0.320401 LR 0.001000 Time 0.021991 +2023-10-02 20:35:16,606 - Epoch: [14][ 900/ 1236] Overall Loss 0.320106 Objective Loss 0.320106 LR 0.001000 Time 0.021984 +2023-10-02 20:35:16,816 - Epoch: [14][ 910/ 1236] Overall Loss 0.320193 Objective Loss 0.320193 LR 0.001000 Time 0.021972 +2023-10-02 20:35:17,030 - Epoch: [14][ 920/ 1236] Overall Loss 0.320140 Objective Loss 0.320140 LR 0.001000 Time 0.021965 +2023-10-02 20:35:17,240 - Epoch: [14][ 930/ 1236] Overall Loss 0.320366 Objective Loss 0.320366 LR 0.001000 Time 0.021955 +2023-10-02 20:35:17,454 - Epoch: [14][ 940/ 1236] Overall Loss 0.320182 Objective Loss 0.320182 LR 0.001000 Time 0.021948 +2023-10-02 20:35:17,664 - Epoch: [14][ 950/ 1236] Overall Loss 0.320180 Objective Loss 0.320180 LR 0.001000 Time 0.021938 +2023-10-02 20:35:17,878 - Epoch: [14][ 960/ 1236] Overall Loss 0.319975 Objective Loss 0.319975 LR 0.001000 Time 0.021932 +2023-10-02 20:35:18,088 - Epoch: [14][ 970/ 1236] Overall Loss 0.319692 Objective Loss 0.319692 LR 0.001000 Time 0.021922 +2023-10-02 20:35:18,302 - Epoch: [14][ 980/ 1236] Overall Loss 0.319748 Objective Loss 0.319748 LR 0.001000 Time 0.021916 +2023-10-02 20:35:18,512 - Epoch: [14][ 990/ 1236] Overall Loss 0.319669 Objective Loss 0.319669 LR 0.001000 Time 0.021906 +2023-10-02 20:35:18,725 - Epoch: [14][ 1000/ 1236] Overall Loss 0.319505 Objective Loss 0.319505 LR 0.001000 Time 0.021900 +2023-10-02 20:35:18,935 - Epoch: [14][ 1010/ 1236] Overall Loss 0.319236 Objective Loss 0.319236 LR 0.001000 Time 0.021891 +2023-10-02 20:35:19,149 - Epoch: [14][ 1020/ 1236] Overall Loss 0.319057 Objective Loss 0.319057 LR 0.001000 Time 0.021885 +2023-10-02 20:35:19,359 - Epoch: [14][ 1030/ 1236] Overall Loss 0.319457 Objective Loss 0.319457 LR 0.001000 Time 0.021876 +2023-10-02 20:35:19,573 - Epoch: [14][ 1040/ 1236] Overall Loss 0.319460 Objective Loss 0.319460 LR 0.001000 Time 0.021871 +2023-10-02 20:35:19,783 - Epoch: [14][ 1050/ 1236] Overall Loss 0.319976 Objective Loss 0.319976 LR 0.001000 Time 0.021862 +2023-10-02 20:35:19,996 - Epoch: [14][ 1060/ 1236] Overall Loss 0.320034 Objective Loss 0.320034 LR 0.001000 Time 0.021857 +2023-10-02 20:35:20,206 - Epoch: [14][ 1070/ 1236] Overall Loss 0.320128 Objective Loss 0.320128 LR 0.001000 Time 0.021849 +2023-10-02 20:35:20,420 - Epoch: [14][ 1080/ 1236] Overall Loss 0.320578 Objective Loss 0.320578 LR 0.001000 Time 0.021844 +2023-10-02 20:35:20,630 - Epoch: [14][ 1090/ 1236] Overall Loss 0.320641 Objective Loss 0.320641 LR 0.001000 Time 0.021836 +2023-10-02 20:35:20,843 - Epoch: [14][ 1100/ 1236] Overall Loss 0.321189 Objective Loss 0.321189 LR 0.001000 Time 0.021831 +2023-10-02 20:35:21,053 - Epoch: [14][ 1110/ 1236] Overall Loss 0.321364 Objective Loss 0.321364 LR 0.001000 Time 0.021823 +2023-10-02 20:35:21,267 - Epoch: [14][ 1120/ 1236] Overall Loss 0.321420 Objective Loss 0.321420 LR 0.001000 Time 0.021819 +2023-10-02 20:35:21,477 - Epoch: [14][ 1130/ 1236] Overall Loss 0.321527 Objective Loss 0.321527 LR 0.001000 Time 0.021811 +2023-10-02 20:35:21,691 - Epoch: [14][ 1140/ 1236] Overall Loss 0.321833 Objective Loss 0.321833 LR 0.001000 Time 0.021807 +2023-10-02 20:35:21,901 - Epoch: [14][ 1150/ 1236] Overall Loss 0.321979 Objective Loss 0.321979 LR 0.001000 Time 0.021800 +2023-10-02 20:35:22,115 - Epoch: [14][ 1160/ 1236] Overall Loss 0.321951 Objective Loss 0.321951 LR 0.001000 Time 0.021796 +2023-10-02 20:35:22,325 - Epoch: [14][ 1170/ 1236] Overall Loss 0.322207 Objective Loss 0.322207 LR 0.001000 Time 0.021789 +2023-10-02 20:35:22,539 - Epoch: [14][ 1180/ 1236] Overall Loss 0.322404 Objective Loss 0.322404 LR 0.001000 Time 0.021785 +2023-10-02 20:35:22,749 - Epoch: [14][ 1190/ 1236] Overall Loss 0.322424 Objective Loss 0.322424 LR 0.001000 Time 0.021778 +2023-10-02 20:35:22,963 - Epoch: [14][ 1200/ 1236] Overall Loss 0.322373 Objective Loss 0.322373 LR 0.001000 Time 0.021774 +2023-10-02 20:35:23,173 - Epoch: [14][ 1210/ 1236] Overall Loss 0.322271 Objective Loss 0.322271 LR 0.001000 Time 0.021768 +2023-10-02 20:35:23,387 - Epoch: [14][ 1220/ 1236] Overall Loss 0.322187 Objective Loss 0.322187 LR 0.001000 Time 0.021764 +2023-10-02 20:35:23,651 - Epoch: [14][ 1230/ 1236] Overall Loss 0.322188 Objective Loss 0.322188 LR 0.001000 Time 0.021802 +2023-10-02 20:35:23,773 - Epoch: [14][ 1236/ 1236] Overall Loss 0.322341 Objective Loss 0.322341 Top1 86.761711 Top5 98.574338 LR 0.001000 Time 0.021795 +2023-10-02 20:35:23,903 - --- validate (epoch=14)----------- +2023-10-02 20:35:23,903 - 29943 samples (256 per mini-batch) +2023-10-02 20:35:24,372 - Epoch: [14][ 10/ 117] Loss 0.328927 Top1 81.835938 Top5 97.734375 +2023-10-02 20:35:24,522 - Epoch: [14][ 20/ 117] Loss 0.335041 Top1 81.953125 Top5 97.890625 +2023-10-02 20:35:24,670 - Epoch: [14][ 30/ 117] Loss 0.346794 Top1 81.979167 Top5 97.968750 +2023-10-02 20:35:24,820 - Epoch: [14][ 40/ 117] Loss 0.344475 Top1 82.089844 Top5 98.085938 +2023-10-02 20:35:24,967 - Epoch: [14][ 50/ 117] Loss 0.338239 Top1 82.320312 Top5 98.070312 +2023-10-02 20:35:25,117 - Epoch: [14][ 60/ 117] Loss 0.333690 Top1 82.447917 Top5 98.125000 +2023-10-02 20:35:25,265 - Epoch: [14][ 70/ 117] Loss 0.329518 Top1 82.656250 Top5 98.152902 +2023-10-02 20:35:25,414 - Epoch: [14][ 80/ 117] Loss 0.330048 Top1 82.700195 Top5 98.154297 +2023-10-02 20:35:25,562 - Epoch: [14][ 90/ 117] Loss 0.334567 Top1 82.656250 Top5 98.111979 +2023-10-02 20:35:25,712 - Epoch: [14][ 100/ 117] Loss 0.335294 Top1 82.628906 Top5 98.078125 +2023-10-02 20:35:25,867 - Epoch: [14][ 110/ 117] Loss 0.339931 Top1 82.649148 Top5 98.039773 +2023-10-02 20:35:25,957 - Epoch: [14][ 117/ 117] Loss 0.340913 Top1 82.590255 Top5 98.016231 +2023-10-02 20:35:26,050 - ==> Top1: 82.590 Top5: 98.016 Loss: 0.341 + +2023-10-02 20:35:26,051 - ==> Confusion: +[[ 845 7 6 0 14 2 0 0 5 131 1 1 0 3 2 3 6 0 1 0 23] + [ 0 1046 3 1 6 24 9 16 1 1 2 1 1 0 0 2 3 0 7 4 4] + [ 0 0 937 8 3 1 53 12 0 0 0 1 8 2 0 4 0 2 6 7 12] + [ 2 3 32 931 0 1 9 1 4 0 7 0 13 8 29 2 2 6 16 1 22] + [ 11 6 3 0 981 9 0 0 0 8 0 0 4 3 5 7 8 0 0 1 4] + [ 2 38 0 2 2 986 2 19 1 1 1 9 4 11 7 1 1 0 9 8 12] + [ 0 1 20 0 0 0 1147 5 0 0 0 0 1 0 0 5 0 0 1 7 4] + [ 2 27 21 0 3 26 8 1046 1 0 2 6 9 0 1 2 1 0 47 9 7] + [ 11 6 0 2 1 1 0 0 952 41 10 3 4 17 27 2 3 1 6 0 2] + [ 52 0 0 1 7 2 1 0 36 961 0 0 5 29 6 2 1 1 0 2 13] + [ 1 2 5 12 2 1 11 6 21 0 934 5 3 21 5 0 1 1 11 1 10] + [ 0 0 4 0 0 14 0 0 0 0 0 948 24 11 0 3 1 12 0 12 6] + [ 0 1 2 1 1 2 2 0 0 0 1 54 956 4 3 11 2 7 0 9 12] + [ 2 0 1 0 5 10 1 0 6 7 5 4 1 1051 4 1 0 1 0 6 14] + [ 7 3 2 8 9 0 0 0 14 3 2 0 3 4 1004 0 3 3 9 1 26] + [ 1 0 5 0 5 0 1 0 0 0 0 11 9 1 0 1078 6 1 0 7 9] + [ 0 19 1 0 2 6 2 0 1 0 0 6 0 0 3 15 1079 1 0 5 21] + [ 0 0 1 1 0 0 4 0 1 0 0 21 30 0 0 25 0 942 1 1 11] + [ 1 7 6 9 1 0 0 18 5 0 2 0 4 1 12 1 1 0 982 2 16] + [ 0 0 4 0 1 5 13 10 0 0 1 9 5 0 0 3 3 0 1 1093 4] + [ 89 280 139 51 114 184 82 91 78 96 176 187 343 323 131 94 115 32 175 294 4831]] + +2023-10-02 20:35:26,052 - ==> Best [Top1: 82.590 Top5: 98.016 Sparsity:0.00 Params: 169472 on epoch: 14] +2023-10-02 20:35:26,052 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:35:26,066 - + +2023-10-02 20:35:26,066 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:35:27,196 - Epoch: [15][ 10/ 1236] Overall Loss 0.294808 Objective Loss 0.294808 LR 0.001000 Time 0.112940 +2023-10-02 20:35:27,404 - Epoch: [15][ 20/ 1236] Overall Loss 0.295697 Objective Loss 0.295697 LR 0.001000 Time 0.066824 +2023-10-02 20:35:27,610 - Epoch: [15][ 30/ 1236] Overall Loss 0.301942 Objective Loss 0.301942 LR 0.001000 Time 0.051382 +2023-10-02 20:35:27,819 - Epoch: [15][ 40/ 1236] Overall Loss 0.294456 Objective Loss 0.294456 LR 0.001000 Time 0.043755 +2023-10-02 20:35:28,024 - Epoch: [15][ 50/ 1236] Overall Loss 0.298650 Objective Loss 0.298650 LR 0.001000 Time 0.039103 +2023-10-02 20:35:28,233 - Epoch: [15][ 60/ 1236] Overall Loss 0.305414 Objective Loss 0.305414 LR 0.001000 Time 0.036060 +2023-10-02 20:35:28,438 - Epoch: [15][ 70/ 1236] Overall Loss 0.306780 Objective Loss 0.306780 LR 0.001000 Time 0.033837 +2023-10-02 20:35:28,647 - Epoch: [15][ 80/ 1236] Overall Loss 0.306346 Objective Loss 0.306346 LR 0.001000 Time 0.032215 +2023-10-02 20:35:28,853 - Epoch: [15][ 90/ 1236] Overall Loss 0.308965 Objective Loss 0.308965 LR 0.001000 Time 0.030913 +2023-10-02 20:35:29,061 - Epoch: [15][ 100/ 1236] Overall Loss 0.308268 Objective Loss 0.308268 LR 0.001000 Time 0.029907 +2023-10-02 20:35:29,267 - Epoch: [15][ 110/ 1236] Overall Loss 0.309846 Objective Loss 0.309846 LR 0.001000 Time 0.029051 +2023-10-02 20:35:29,476 - Epoch: [15][ 120/ 1236] Overall Loss 0.307799 Objective Loss 0.307799 LR 0.001000 Time 0.028369 +2023-10-02 20:35:29,681 - Epoch: [15][ 130/ 1236] Overall Loss 0.308841 Objective Loss 0.308841 LR 0.001000 Time 0.027766 +2023-10-02 20:35:29,890 - Epoch: [15][ 140/ 1236] Overall Loss 0.310095 Objective Loss 0.310095 LR 0.001000 Time 0.027272 +2023-10-02 20:35:30,095 - Epoch: [15][ 150/ 1236] Overall Loss 0.311729 Objective Loss 0.311729 LR 0.001000 Time 0.026820 +2023-10-02 20:35:30,304 - Epoch: [15][ 160/ 1236] Overall Loss 0.311600 Objective Loss 0.311600 LR 0.001000 Time 0.026448 +2023-10-02 20:35:30,509 - Epoch: [15][ 170/ 1236] Overall Loss 0.313469 Objective Loss 0.313469 LR 0.001000 Time 0.026098 +2023-10-02 20:35:30,717 - Epoch: [15][ 180/ 1236] Overall Loss 0.313001 Objective Loss 0.313001 LR 0.001000 Time 0.025797 +2023-10-02 20:35:30,923 - Epoch: [15][ 190/ 1236] Overall Loss 0.313190 Objective Loss 0.313190 LR 0.001000 Time 0.025523 +2023-10-02 20:35:31,130 - Epoch: [15][ 200/ 1236] Overall Loss 0.311528 Objective Loss 0.311528 LR 0.001000 Time 0.025282 +2023-10-02 20:35:31,335 - Epoch: [15][ 210/ 1236] Overall Loss 0.309923 Objective Loss 0.309923 LR 0.001000 Time 0.025053 +2023-10-02 20:35:31,544 - Epoch: [15][ 220/ 1236] Overall Loss 0.309682 Objective Loss 0.309682 LR 0.001000 Time 0.024863 +2023-10-02 20:35:31,749 - Epoch: [15][ 230/ 1236] Overall Loss 0.309444 Objective Loss 0.309444 LR 0.001000 Time 0.024672 +2023-10-02 20:35:31,958 - Epoch: [15][ 240/ 1236] Overall Loss 0.310083 Objective Loss 0.310083 LR 0.001000 Time 0.024514 +2023-10-02 20:35:32,163 - Epoch: [15][ 250/ 1236] Overall Loss 0.312284 Objective Loss 0.312284 LR 0.001000 Time 0.024353 +2023-10-02 20:35:32,372 - Epoch: [15][ 260/ 1236] Overall Loss 0.311984 Objective Loss 0.311984 LR 0.001000 Time 0.024219 +2023-10-02 20:35:32,578 - Epoch: [15][ 270/ 1236] Overall Loss 0.312105 Objective Loss 0.312105 LR 0.001000 Time 0.024081 +2023-10-02 20:35:32,787 - Epoch: [15][ 280/ 1236] Overall Loss 0.312164 Objective Loss 0.312164 LR 0.001000 Time 0.023966 +2023-10-02 20:35:32,992 - Epoch: [15][ 290/ 1236] Overall Loss 0.312526 Objective Loss 0.312526 LR 0.001000 Time 0.023846 +2023-10-02 20:35:33,201 - Epoch: [15][ 300/ 1236] Overall Loss 0.312283 Objective Loss 0.312283 LR 0.001000 Time 0.023748 +2023-10-02 20:35:33,406 - Epoch: [15][ 310/ 1236] Overall Loss 0.311715 Objective Loss 0.311715 LR 0.001000 Time 0.023643 +2023-10-02 20:35:33,615 - Epoch: [15][ 320/ 1236] Overall Loss 0.311690 Objective Loss 0.311690 LR 0.001000 Time 0.023556 +2023-10-02 20:35:33,821 - Epoch: [15][ 330/ 1236] Overall Loss 0.311739 Objective Loss 0.311739 LR 0.001000 Time 0.023464 +2023-10-02 20:35:34,030 - Epoch: [15][ 340/ 1236] Overall Loss 0.312501 Objective Loss 0.312501 LR 0.001000 Time 0.023388 +2023-10-02 20:35:34,235 - Epoch: [15][ 350/ 1236] Overall Loss 0.312777 Objective Loss 0.312777 LR 0.001000 Time 0.023307 +2023-10-02 20:35:34,445 - Epoch: [15][ 360/ 1236] Overall Loss 0.312827 Objective Loss 0.312827 LR 0.001000 Time 0.023240 +2023-10-02 20:35:34,650 - Epoch: [15][ 370/ 1236] Overall Loss 0.312439 Objective Loss 0.312439 LR 0.001000 Time 0.023165 +2023-10-02 20:35:34,859 - Epoch: [15][ 380/ 1236] Overall Loss 0.312053 Objective Loss 0.312053 LR 0.001000 Time 0.023105 +2023-10-02 20:35:35,064 - Epoch: [15][ 390/ 1236] Overall Loss 0.311305 Objective Loss 0.311305 LR 0.001000 Time 0.023038 +2023-10-02 20:35:35,273 - Epoch: [15][ 400/ 1236] Overall Loss 0.310910 Objective Loss 0.310910 LR 0.001000 Time 0.022984 +2023-10-02 20:35:35,479 - Epoch: [15][ 410/ 1236] Overall Loss 0.310941 Objective Loss 0.310941 LR 0.001000 Time 0.022924 +2023-10-02 20:35:35,688 - Epoch: [15][ 420/ 1236] Overall Loss 0.311257 Objective Loss 0.311257 LR 0.001000 Time 0.022875 +2023-10-02 20:35:35,893 - Epoch: [15][ 430/ 1236] Overall Loss 0.311669 Objective Loss 0.311669 LR 0.001000 Time 0.022819 +2023-10-02 20:35:36,102 - Epoch: [15][ 440/ 1236] Overall Loss 0.312106 Objective Loss 0.312106 LR 0.001000 Time 0.022775 +2023-10-02 20:35:36,307 - Epoch: [15][ 450/ 1236] Overall Loss 0.312006 Objective Loss 0.312006 LR 0.001000 Time 0.022725 +2023-10-02 20:35:36,516 - Epoch: [15][ 460/ 1236] Overall Loss 0.312027 Objective Loss 0.312027 LR 0.001000 Time 0.022684 +2023-10-02 20:35:36,721 - Epoch: [15][ 470/ 1236] Overall Loss 0.312728 Objective Loss 0.312728 LR 0.001000 Time 0.022637 +2023-10-02 20:35:36,930 - Epoch: [15][ 480/ 1236] Overall Loss 0.311818 Objective Loss 0.311818 LR 0.001000 Time 0.022600 +2023-10-02 20:35:37,136 - Epoch: [15][ 490/ 1236] Overall Loss 0.311696 Objective Loss 0.311696 LR 0.001000 Time 0.022558 +2023-10-02 20:35:37,345 - Epoch: [15][ 500/ 1236] Overall Loss 0.311255 Objective Loss 0.311255 LR 0.001000 Time 0.022524 +2023-10-02 20:35:37,550 - Epoch: [15][ 510/ 1236] Overall Loss 0.311612 Objective Loss 0.311612 LR 0.001000 Time 0.022485 +2023-10-02 20:35:37,759 - Epoch: [15][ 520/ 1236] Overall Loss 0.311578 Objective Loss 0.311578 LR 0.001000 Time 0.022454 +2023-10-02 20:35:37,965 - Epoch: [15][ 530/ 1236] Overall Loss 0.312925 Objective Loss 0.312925 LR 0.001000 Time 0.022418 +2023-10-02 20:35:38,176 - Epoch: [15][ 540/ 1236] Overall Loss 0.313556 Objective Loss 0.313556 LR 0.001000 Time 0.022392 +2023-10-02 20:35:38,381 - Epoch: [15][ 550/ 1236] Overall Loss 0.314183 Objective Loss 0.314183 LR 0.001000 Time 0.022358 +2023-10-02 20:35:38,590 - Epoch: [15][ 560/ 1236] Overall Loss 0.314075 Objective Loss 0.314075 LR 0.001000 Time 0.022332 +2023-10-02 20:35:38,796 - Epoch: [15][ 570/ 1236] Overall Loss 0.314397 Objective Loss 0.314397 LR 0.001000 Time 0.022300 +2023-10-02 20:35:39,005 - Epoch: [15][ 580/ 1236] Overall Loss 0.314540 Objective Loss 0.314540 LR 0.001000 Time 0.022277 +2023-10-02 20:35:39,211 - Epoch: [15][ 590/ 1236] Overall Loss 0.314396 Objective Loss 0.314396 LR 0.001000 Time 0.022247 +2023-10-02 20:35:39,420 - Epoch: [15][ 600/ 1236] Overall Loss 0.314722 Objective Loss 0.314722 LR 0.001000 Time 0.022225 +2023-10-02 20:35:39,626 - Epoch: [15][ 610/ 1236] Overall Loss 0.314914 Objective Loss 0.314914 LR 0.001000 Time 0.022197 +2023-10-02 20:35:39,835 - Epoch: [15][ 620/ 1236] Overall Loss 0.315156 Objective Loss 0.315156 LR 0.001000 Time 0.022176 +2023-10-02 20:35:40,041 - Epoch: [15][ 630/ 1236] Overall Loss 0.315713 Objective Loss 0.315713 LR 0.001000 Time 0.022150 +2023-10-02 20:35:40,250 - Epoch: [15][ 640/ 1236] Overall Loss 0.315776 Objective Loss 0.315776 LR 0.001000 Time 0.022130 +2023-10-02 20:35:40,456 - Epoch: [15][ 650/ 1236] Overall Loss 0.315932 Objective Loss 0.315932 LR 0.001000 Time 0.022105 +2023-10-02 20:35:40,665 - Epoch: [15][ 660/ 1236] Overall Loss 0.315862 Objective Loss 0.315862 LR 0.001000 Time 0.022088 +2023-10-02 20:35:40,871 - Epoch: [15][ 670/ 1236] Overall Loss 0.315810 Objective Loss 0.315810 LR 0.001000 Time 0.022064 +2023-10-02 20:35:41,080 - Epoch: [15][ 680/ 1236] Overall Loss 0.315730 Objective Loss 0.315730 LR 0.001000 Time 0.022047 +2023-10-02 20:35:41,286 - Epoch: [15][ 690/ 1236] Overall Loss 0.316191 Objective Loss 0.316191 LR 0.001000 Time 0.022025 +2023-10-02 20:35:41,495 - Epoch: [15][ 700/ 1236] Overall Loss 0.316512 Objective Loss 0.316512 LR 0.001000 Time 0.022009 +2023-10-02 20:35:41,701 - Epoch: [15][ 710/ 1236] Overall Loss 0.316262 Objective Loss 0.316262 LR 0.001000 Time 0.021988 +2023-10-02 20:35:41,910 - Epoch: [15][ 720/ 1236] Overall Loss 0.316289 Objective Loss 0.316289 LR 0.001000 Time 0.021973 +2023-10-02 20:35:42,116 - Epoch: [15][ 730/ 1236] Overall Loss 0.316930 Objective Loss 0.316930 LR 0.001000 Time 0.021954 +2023-10-02 20:35:42,325 - Epoch: [15][ 740/ 1236] Overall Loss 0.316892 Objective Loss 0.316892 LR 0.001000 Time 0.021940 +2023-10-02 20:35:42,531 - Epoch: [15][ 750/ 1236] Overall Loss 0.316803 Objective Loss 0.316803 LR 0.001000 Time 0.021921 +2023-10-02 20:35:42,740 - Epoch: [15][ 760/ 1236] Overall Loss 0.317045 Objective Loss 0.317045 LR 0.001000 Time 0.021908 +2023-10-02 20:35:42,946 - Epoch: [15][ 770/ 1236] Overall Loss 0.316786 Objective Loss 0.316786 LR 0.001000 Time 0.021890 +2023-10-02 20:35:43,155 - Epoch: [15][ 780/ 1236] Overall Loss 0.317048 Objective Loss 0.317048 LR 0.001000 Time 0.021877 +2023-10-02 20:35:43,360 - Epoch: [15][ 790/ 1236] Overall Loss 0.317116 Objective Loss 0.317116 LR 0.001000 Time 0.021859 +2023-10-02 20:35:43,569 - Epoch: [15][ 800/ 1236] Overall Loss 0.317686 Objective Loss 0.317686 LR 0.001000 Time 0.021847 +2023-10-02 20:35:43,775 - Epoch: [15][ 810/ 1236] Overall Loss 0.317751 Objective Loss 0.317751 LR 0.001000 Time 0.021831 +2023-10-02 20:35:43,984 - Epoch: [15][ 820/ 1236] Overall Loss 0.317506 Objective Loss 0.317506 LR 0.001000 Time 0.021820 +2023-10-02 20:35:44,190 - Epoch: [15][ 830/ 1236] Overall Loss 0.317492 Objective Loss 0.317492 LR 0.001000 Time 0.021804 +2023-10-02 20:35:44,399 - Epoch: [15][ 840/ 1236] Overall Loss 0.317762 Objective Loss 0.317762 LR 0.001000 Time 0.021793 +2023-10-02 20:35:44,605 - Epoch: [15][ 850/ 1236] Overall Loss 0.317600 Objective Loss 0.317600 LR 0.001000 Time 0.021779 +2023-10-02 20:35:44,814 - Epoch: [15][ 860/ 1236] Overall Loss 0.317805 Objective Loss 0.317805 LR 0.001000 Time 0.021769 +2023-10-02 20:35:45,020 - Epoch: [15][ 870/ 1236] Overall Loss 0.317864 Objective Loss 0.317864 LR 0.001000 Time 0.021754 +2023-10-02 20:35:45,229 - Epoch: [15][ 880/ 1236] Overall Loss 0.317852 Objective Loss 0.317852 LR 0.001000 Time 0.021745 +2023-10-02 20:35:45,435 - Epoch: [15][ 890/ 1236] Overall Loss 0.317871 Objective Loss 0.317871 LR 0.001000 Time 0.021731 +2023-10-02 20:35:45,644 - Epoch: [15][ 900/ 1236] Overall Loss 0.317998 Objective Loss 0.317998 LR 0.001000 Time 0.021722 +2023-10-02 20:35:45,850 - Epoch: [15][ 910/ 1236] Overall Loss 0.318280 Objective Loss 0.318280 LR 0.001000 Time 0.021709 +2023-10-02 20:35:46,059 - Epoch: [15][ 920/ 1236] Overall Loss 0.318042 Objective Loss 0.318042 LR 0.001000 Time 0.021700 +2023-10-02 20:35:46,265 - Epoch: [15][ 930/ 1236] Overall Loss 0.317975 Objective Loss 0.317975 LR 0.001000 Time 0.021687 +2023-10-02 20:35:46,474 - Epoch: [15][ 940/ 1236] Overall Loss 0.317790 Objective Loss 0.317790 LR 0.001000 Time 0.021679 +2023-10-02 20:35:46,680 - Epoch: [15][ 950/ 1236] Overall Loss 0.317564 Objective Loss 0.317564 LR 0.001000 Time 0.021667 +2023-10-02 20:35:46,889 - Epoch: [15][ 960/ 1236] Overall Loss 0.317580 Objective Loss 0.317580 LR 0.001000 Time 0.021660 +2023-10-02 20:35:47,095 - Epoch: [15][ 970/ 1236] Overall Loss 0.317548 Objective Loss 0.317548 LR 0.001000 Time 0.021648 +2023-10-02 20:35:47,304 - Epoch: [15][ 980/ 1236] Overall Loss 0.317312 Objective Loss 0.317312 LR 0.001000 Time 0.021640 +2023-10-02 20:35:47,510 - Epoch: [15][ 990/ 1236] Overall Loss 0.317450 Objective Loss 0.317450 LR 0.001000 Time 0.021629 +2023-10-02 20:35:47,719 - Epoch: [15][ 1000/ 1236] Overall Loss 0.318026 Objective Loss 0.318026 LR 0.001000 Time 0.021622 +2023-10-02 20:35:47,925 - Epoch: [15][ 1010/ 1236] Overall Loss 0.318127 Objective Loss 0.318127 LR 0.001000 Time 0.021611 +2023-10-02 20:35:48,134 - Epoch: [15][ 1020/ 1236] Overall Loss 0.317967 Objective Loss 0.317967 LR 0.001000 Time 0.021604 +2023-10-02 20:35:48,340 - Epoch: [15][ 1030/ 1236] Overall Loss 0.318125 Objective Loss 0.318125 LR 0.001000 Time 0.021594 +2023-10-02 20:35:48,549 - Epoch: [15][ 1040/ 1236] Overall Loss 0.317749 Objective Loss 0.317749 LR 0.001000 Time 0.021587 +2023-10-02 20:35:48,755 - Epoch: [15][ 1050/ 1236] Overall Loss 0.317869 Objective Loss 0.317869 LR 0.001000 Time 0.021577 +2023-10-02 20:35:48,964 - Epoch: [15][ 1060/ 1236] Overall Loss 0.317645 Objective Loss 0.317645 LR 0.001000 Time 0.021571 +2023-10-02 20:35:49,170 - Epoch: [15][ 1070/ 1236] Overall Loss 0.317555 Objective Loss 0.317555 LR 0.001000 Time 0.021561 +2023-10-02 20:35:49,378 - Epoch: [15][ 1080/ 1236] Overall Loss 0.317494 Objective Loss 0.317494 LR 0.001000 Time 0.021554 +2023-10-02 20:35:49,585 - Epoch: [15][ 1090/ 1236] Overall Loss 0.317294 Objective Loss 0.317294 LR 0.001000 Time 0.021545 +2023-10-02 20:35:49,794 - Epoch: [15][ 1100/ 1236] Overall Loss 0.317316 Objective Loss 0.317316 LR 0.001000 Time 0.021539 +2023-10-02 20:35:50,000 - Epoch: [15][ 1110/ 1236] Overall Loss 0.317199 Objective Loss 0.317199 LR 0.001000 Time 0.021530 +2023-10-02 20:35:50,210 - Epoch: [15][ 1120/ 1236] Overall Loss 0.317220 Objective Loss 0.317220 LR 0.001000 Time 0.021524 +2023-10-02 20:35:50,415 - Epoch: [15][ 1130/ 1236] Overall Loss 0.317063 Objective Loss 0.317063 LR 0.001000 Time 0.021515 +2023-10-02 20:35:50,625 - Epoch: [15][ 1140/ 1236] Overall Loss 0.317302 Objective Loss 0.317302 LR 0.001000 Time 0.021510 +2023-10-02 20:35:50,830 - Epoch: [15][ 1150/ 1236] Overall Loss 0.317280 Objective Loss 0.317280 LR 0.001000 Time 0.021502 +2023-10-02 20:35:51,039 - Epoch: [15][ 1160/ 1236] Overall Loss 0.317436 Objective Loss 0.317436 LR 0.001000 Time 0.021497 +2023-10-02 20:35:51,245 - Epoch: [15][ 1170/ 1236] Overall Loss 0.317700 Objective Loss 0.317700 LR 0.001000 Time 0.021488 +2023-10-02 20:35:51,455 - Epoch: [15][ 1180/ 1236] Overall Loss 0.317752 Objective Loss 0.317752 LR 0.001000 Time 0.021484 +2023-10-02 20:35:51,660 - Epoch: [15][ 1190/ 1236] Overall Loss 0.317757 Objective Loss 0.317757 LR 0.001000 Time 0.021475 +2023-10-02 20:35:51,868 - Epoch: [15][ 1200/ 1236] Overall Loss 0.317902 Objective Loss 0.317902 LR 0.001000 Time 0.021470 +2023-10-02 20:35:52,075 - Epoch: [15][ 1210/ 1236] Overall Loss 0.317935 Objective Loss 0.317935 LR 0.001000 Time 0.021462 +2023-10-02 20:35:52,285 - Epoch: [15][ 1220/ 1236] Overall Loss 0.318140 Objective Loss 0.318140 LR 0.001000 Time 0.021458 +2023-10-02 20:35:52,546 - Epoch: [15][ 1230/ 1236] Overall Loss 0.318187 Objective Loss 0.318187 LR 0.001000 Time 0.021495 +2023-10-02 20:35:52,668 - Epoch: [15][ 1236/ 1236] Overall Loss 0.317946 Objective Loss 0.317946 Top1 85.743381 Top5 99.185336 LR 0.001000 Time 0.021490 +2023-10-02 20:35:52,775 - --- validate (epoch=15)----------- +2023-10-02 20:35:52,775 - 29943 samples (256 per mini-batch) +2023-10-02 20:35:53,263 - Epoch: [15][ 10/ 117] Loss 0.321541 Top1 83.242188 Top5 97.929688 +2023-10-02 20:35:53,418 - Epoch: [15][ 20/ 117] Loss 0.316068 Top1 82.285156 Top5 98.203125 +2023-10-02 20:35:53,568 - Epoch: [15][ 30/ 117] Loss 0.325168 Top1 81.953125 Top5 98.372396 +2023-10-02 20:35:53,724 - Epoch: [15][ 40/ 117] Loss 0.325420 Top1 82.216797 Top5 98.281250 +2023-10-02 20:35:53,874 - Epoch: [15][ 50/ 117] Loss 0.326025 Top1 82.375000 Top5 98.195312 +2023-10-02 20:35:54,026 - Epoch: [15][ 60/ 117] Loss 0.330765 Top1 82.356771 Top5 98.229167 +2023-10-02 20:35:54,175 - Epoch: [15][ 70/ 117] Loss 0.332641 Top1 82.299107 Top5 98.203125 +2023-10-02 20:35:54,324 - Epoch: [15][ 80/ 117] Loss 0.338566 Top1 82.197266 Top5 98.120117 +2023-10-02 20:35:54,472 - Epoch: [15][ 90/ 117] Loss 0.337160 Top1 82.322049 Top5 98.081597 +2023-10-02 20:35:54,622 - Epoch: [15][ 100/ 117] Loss 0.335855 Top1 82.398438 Top5 98.097656 +2023-10-02 20:35:54,777 - Epoch: [15][ 110/ 117] Loss 0.337320 Top1 82.301136 Top5 98.064631 +2023-10-02 20:35:54,866 - Epoch: [15][ 117/ 117] Loss 0.336461 Top1 82.406572 Top5 98.056307 +2023-10-02 20:35:54,976 - ==> Top1: 82.407 Top5: 98.056 Loss: 0.336 + +2023-10-02 20:35:54,977 - ==> Confusion: +[[ 874 1 2 0 3 3 0 1 15 111 1 2 0 4 1 5 4 3 1 0 19] + [ 0 1031 2 0 5 35 1 18 5 2 5 0 2 0 0 2 3 0 7 3 10] + [ 4 0 927 24 0 2 40 14 0 3 2 2 8 4 0 4 0 2 6 5 9] + [ 1 2 18 986 0 2 4 2 18 1 7 0 6 5 12 0 1 7 5 0 12] + [ 20 11 4 0 931 13 0 1 0 23 0 0 3 8 13 5 9 0 0 4 5] + [ 1 41 1 2 2 999 1 21 7 3 4 1 4 8 2 1 2 0 2 4 10] + [ 0 3 20 0 0 0 1131 13 0 0 3 1 1 1 0 4 1 0 1 7 5] + [ 5 23 17 1 0 19 3 1082 1 0 3 7 6 0 0 1 0 0 31 12 7] + [ 13 2 1 1 0 0 0 1 1000 38 5 1 7 11 3 0 0 1 1 0 4] + [ 65 1 0 0 3 1 0 0 43 963 0 0 1 23 7 0 0 2 0 5 5] + [ 3 1 6 10 1 0 1 5 17 2 961 3 0 21 5 1 0 1 4 2 9] + [ 0 5 0 0 0 14 0 3 0 1 0 933 25 11 0 2 2 18 0 11 10] + [ 0 3 2 1 1 1 0 1 1 0 0 47 962 7 2 6 3 18 1 9 3] + [ 0 0 2 0 1 15 1 0 15 10 5 7 1 1047 3 1 0 1 0 2 8] + [ 8 2 2 29 5 0 0 0 68 10 5 0 5 8 933 0 1 7 4 1 13] + [ 0 1 2 3 3 0 4 0 0 0 0 9 13 0 0 1059 10 13 0 13 4] + [ 2 12 2 2 6 5 3 0 1 0 0 5 1 1 5 9 1086 1 0 8 12] + [ 0 0 0 0 0 0 2 0 0 0 0 5 20 0 0 5 1 1000 0 3 2] + [ 1 6 3 25 0 0 1 34 10 0 7 0 4 0 14 0 0 0 952 2 9] + [ 0 3 3 1 1 5 6 13 0 1 0 11 6 4 0 0 3 0 0 1092 3] + [ 119 232 123 113 68 227 56 118 180 129 184 132 441 353 110 64 93 74 128 235 4726]] + +2023-10-02 20:35:54,978 - ==> Best [Top1: 82.590 Top5: 98.016 Sparsity:0.00 Params: 169472 on epoch: 14] +2023-10-02 20:35:54,978 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:35:54,984 - + +2023-10-02 20:35:54,984 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:35:56,019 - Epoch: [16][ 10/ 1236] Overall Loss 0.316848 Objective Loss 0.316848 LR 0.001000 Time 0.103386 +2023-10-02 20:35:56,226 - Epoch: [16][ 20/ 1236] Overall Loss 0.303217 Objective Loss 0.303217 LR 0.001000 Time 0.062028 +2023-10-02 20:35:56,431 - Epoch: [16][ 30/ 1236] Overall Loss 0.306849 Objective Loss 0.306849 LR 0.001000 Time 0.048193 +2023-10-02 20:35:56,639 - Epoch: [16][ 40/ 1236] Overall Loss 0.308321 Objective Loss 0.308321 LR 0.001000 Time 0.041333 +2023-10-02 20:35:56,843 - Epoch: [16][ 50/ 1236] Overall Loss 0.305958 Objective Loss 0.305958 LR 0.001000 Time 0.037147 +2023-10-02 20:35:57,051 - Epoch: [16][ 60/ 1236] Overall Loss 0.298141 Objective Loss 0.298141 LR 0.001000 Time 0.034416 +2023-10-02 20:35:57,256 - Epoch: [16][ 70/ 1236] Overall Loss 0.300149 Objective Loss 0.300149 LR 0.001000 Time 0.032414 +2023-10-02 20:35:57,464 - Epoch: [16][ 80/ 1236] Overall Loss 0.296405 Objective Loss 0.296405 LR 0.001000 Time 0.030957 +2023-10-02 20:35:57,668 - Epoch: [16][ 90/ 1236] Overall Loss 0.298275 Objective Loss 0.298275 LR 0.001000 Time 0.029783 +2023-10-02 20:35:57,876 - Epoch: [16][ 100/ 1236] Overall Loss 0.297602 Objective Loss 0.297602 LR 0.001000 Time 0.028885 +2023-10-02 20:35:58,080 - Epoch: [16][ 110/ 1236] Overall Loss 0.293563 Objective Loss 0.293563 LR 0.001000 Time 0.028110 +2023-10-02 20:35:58,288 - Epoch: [16][ 120/ 1236] Overall Loss 0.294009 Objective Loss 0.294009 LR 0.001000 Time 0.027499 +2023-10-02 20:35:58,492 - Epoch: [16][ 130/ 1236] Overall Loss 0.294447 Objective Loss 0.294447 LR 0.001000 Time 0.026954 +2023-10-02 20:35:58,700 - Epoch: [16][ 140/ 1236] Overall Loss 0.292792 Objective Loss 0.292792 LR 0.001000 Time 0.026511 +2023-10-02 20:35:58,904 - Epoch: [16][ 150/ 1236] Overall Loss 0.293404 Objective Loss 0.293404 LR 0.001000 Time 0.026102 +2023-10-02 20:35:59,112 - Epoch: [16][ 160/ 1236] Overall Loss 0.295191 Objective Loss 0.295191 LR 0.001000 Time 0.025769 +2023-10-02 20:35:59,317 - Epoch: [16][ 170/ 1236] Overall Loss 0.296996 Objective Loss 0.296996 LR 0.001000 Time 0.025455 +2023-10-02 20:35:59,525 - Epoch: [16][ 180/ 1236] Overall Loss 0.295685 Objective Loss 0.295685 LR 0.001000 Time 0.025194 +2023-10-02 20:35:59,729 - Epoch: [16][ 190/ 1236] Overall Loss 0.296124 Objective Loss 0.296124 LR 0.001000 Time 0.024941 +2023-10-02 20:35:59,937 - Epoch: [16][ 200/ 1236] Overall Loss 0.297483 Objective Loss 0.297483 LR 0.001000 Time 0.024733 +2023-10-02 20:36:00,141 - Epoch: [16][ 210/ 1236] Overall Loss 0.296827 Objective Loss 0.296827 LR 0.001000 Time 0.024526 +2023-10-02 20:36:00,348 - Epoch: [16][ 220/ 1236] Overall Loss 0.294599 Objective Loss 0.294599 LR 0.001000 Time 0.024350 +2023-10-02 20:36:00,553 - Epoch: [16][ 230/ 1236] Overall Loss 0.294730 Objective Loss 0.294730 LR 0.001000 Time 0.024178 +2023-10-02 20:36:00,762 - Epoch: [16][ 240/ 1236] Overall Loss 0.295501 Objective Loss 0.295501 LR 0.001000 Time 0.024038 +2023-10-02 20:36:00,966 - Epoch: [16][ 250/ 1236] Overall Loss 0.295388 Objective Loss 0.295388 LR 0.001000 Time 0.023892 +2023-10-02 20:36:01,173 - Epoch: [16][ 260/ 1236] Overall Loss 0.295130 Objective Loss 0.295130 LR 0.001000 Time 0.023769 +2023-10-02 20:36:01,378 - Epoch: [16][ 270/ 1236] Overall Loss 0.293715 Objective Loss 0.293715 LR 0.001000 Time 0.023643 +2023-10-02 20:36:01,586 - Epoch: [16][ 280/ 1236] Overall Loss 0.293895 Objective Loss 0.293895 LR 0.001000 Time 0.023541 +2023-10-02 20:36:01,791 - Epoch: [16][ 290/ 1236] Overall Loss 0.293346 Objective Loss 0.293346 LR 0.001000 Time 0.023433 +2023-10-02 20:36:01,999 - Epoch: [16][ 300/ 1236] Overall Loss 0.293047 Objective Loss 0.293047 LR 0.001000 Time 0.023345 +2023-10-02 20:36:02,203 - Epoch: [16][ 310/ 1236] Overall Loss 0.292589 Objective Loss 0.292589 LR 0.001000 Time 0.023250 +2023-10-02 20:36:02,411 - Epoch: [16][ 320/ 1236] Overall Loss 0.293345 Objective Loss 0.293345 LR 0.001000 Time 0.023170 +2023-10-02 20:36:02,616 - Epoch: [16][ 330/ 1236] Overall Loss 0.293160 Objective Loss 0.293160 LR 0.001000 Time 0.023086 +2023-10-02 20:36:02,824 - Epoch: [16][ 340/ 1236] Overall Loss 0.293579 Objective Loss 0.293579 LR 0.001000 Time 0.023019 +2023-10-02 20:36:03,029 - Epoch: [16][ 350/ 1236] Overall Loss 0.292841 Objective Loss 0.292841 LR 0.001000 Time 0.022946 +2023-10-02 20:36:03,237 - Epoch: [16][ 360/ 1236] Overall Loss 0.293449 Objective Loss 0.293449 LR 0.001000 Time 0.022885 +2023-10-02 20:36:03,442 - Epoch: [16][ 370/ 1236] Overall Loss 0.293619 Objective Loss 0.293619 LR 0.001000 Time 0.022819 +2023-10-02 20:36:03,649 - Epoch: [16][ 380/ 1236] Overall Loss 0.294485 Objective Loss 0.294485 LR 0.001000 Time 0.022762 +2023-10-02 20:36:03,855 - Epoch: [16][ 390/ 1236] Overall Loss 0.295534 Objective Loss 0.295534 LR 0.001000 Time 0.022703 +2023-10-02 20:36:04,062 - Epoch: [16][ 400/ 1236] Overall Loss 0.295435 Objective Loss 0.295435 LR 0.001000 Time 0.022652 +2023-10-02 20:36:04,267 - Epoch: [16][ 410/ 1236] Overall Loss 0.296094 Objective Loss 0.296094 LR 0.001000 Time 0.022600 +2023-10-02 20:36:04,474 - Epoch: [16][ 420/ 1236] Overall Loss 0.296577 Objective Loss 0.296577 LR 0.001000 Time 0.022555 +2023-10-02 20:36:04,680 - Epoch: [16][ 430/ 1236] Overall Loss 0.297338 Objective Loss 0.297338 LR 0.001000 Time 0.022508 +2023-10-02 20:36:04,887 - Epoch: [16][ 440/ 1236] Overall Loss 0.297448 Objective Loss 0.297448 LR 0.001000 Time 0.022466 +2023-10-02 20:36:05,093 - Epoch: [16][ 450/ 1236] Overall Loss 0.298003 Objective Loss 0.298003 LR 0.001000 Time 0.022423 +2023-10-02 20:36:05,299 - Epoch: [16][ 460/ 1236] Overall Loss 0.297946 Objective Loss 0.297946 LR 0.001000 Time 0.022384 +2023-10-02 20:36:05,506 - Epoch: [16][ 470/ 1236] Overall Loss 0.298251 Objective Loss 0.298251 LR 0.001000 Time 0.022346 +2023-10-02 20:36:05,714 - Epoch: [16][ 480/ 1236] Overall Loss 0.298513 Objective Loss 0.298513 LR 0.001000 Time 0.022315 +2023-10-02 20:36:05,922 - Epoch: [16][ 490/ 1236] Overall Loss 0.298792 Objective Loss 0.298792 LR 0.001000 Time 0.022282 +2023-10-02 20:36:06,131 - Epoch: [16][ 500/ 1236] Overall Loss 0.299281 Objective Loss 0.299281 LR 0.001000 Time 0.022253 +2023-10-02 20:36:06,338 - Epoch: [16][ 510/ 1236] Overall Loss 0.299509 Objective Loss 0.299509 LR 0.001000 Time 0.022223 +2023-10-02 20:36:06,547 - Epoch: [16][ 520/ 1236] Overall Loss 0.299427 Objective Loss 0.299427 LR 0.001000 Time 0.022197 +2023-10-02 20:36:06,755 - Epoch: [16][ 530/ 1236] Overall Loss 0.300211 Objective Loss 0.300211 LR 0.001000 Time 0.022169 +2023-10-02 20:36:06,963 - Epoch: [16][ 540/ 1236] Overall Loss 0.299817 Objective Loss 0.299817 LR 0.001000 Time 0.022144 +2023-10-02 20:36:07,170 - Epoch: [16][ 550/ 1236] Overall Loss 0.300996 Objective Loss 0.300996 LR 0.001000 Time 0.022118 +2023-10-02 20:36:07,379 - Epoch: [16][ 560/ 1236] Overall Loss 0.301946 Objective Loss 0.301946 LR 0.001000 Time 0.022095 +2023-10-02 20:36:07,587 - Epoch: [16][ 570/ 1236] Overall Loss 0.301911 Objective Loss 0.301911 LR 0.001000 Time 0.022071 +2023-10-02 20:36:07,796 - Epoch: [16][ 580/ 1236] Overall Loss 0.301915 Objective Loss 0.301915 LR 0.001000 Time 0.022050 +2023-10-02 20:36:08,003 - Epoch: [16][ 590/ 1236] Overall Loss 0.302107 Objective Loss 0.302107 LR 0.001000 Time 0.022028 +2023-10-02 20:36:08,212 - Epoch: [16][ 600/ 1236] Overall Loss 0.302712 Objective Loss 0.302712 LR 0.001000 Time 0.022008 +2023-10-02 20:36:08,420 - Epoch: [16][ 610/ 1236] Overall Loss 0.302615 Objective Loss 0.302615 LR 0.001000 Time 0.021987 +2023-10-02 20:36:08,628 - Epoch: [16][ 620/ 1236] Overall Loss 0.302523 Objective Loss 0.302523 LR 0.001000 Time 0.021968 +2023-10-02 20:36:08,836 - Epoch: [16][ 630/ 1236] Overall Loss 0.302916 Objective Loss 0.302916 LR 0.001000 Time 0.021949 +2023-10-02 20:36:09,044 - Epoch: [16][ 640/ 1236] Overall Loss 0.302648 Objective Loss 0.302648 LR 0.001000 Time 0.021931 +2023-10-02 20:36:09,252 - Epoch: [16][ 650/ 1236] Overall Loss 0.303430 Objective Loss 0.303430 LR 0.001000 Time 0.021913 +2023-10-02 20:36:09,461 - Epoch: [16][ 660/ 1236] Overall Loss 0.303362 Objective Loss 0.303362 LR 0.001000 Time 0.021897 +2023-10-02 20:36:09,668 - Epoch: [16][ 670/ 1236] Overall Loss 0.303858 Objective Loss 0.303858 LR 0.001000 Time 0.021879 +2023-10-02 20:36:09,877 - Epoch: [16][ 680/ 1236] Overall Loss 0.303922 Objective Loss 0.303922 LR 0.001000 Time 0.021864 +2023-10-02 20:36:10,083 - Epoch: [16][ 690/ 1236] Overall Loss 0.303954 Objective Loss 0.303954 LR 0.001000 Time 0.021845 +2023-10-02 20:36:10,292 - Epoch: [16][ 700/ 1236] Overall Loss 0.303872 Objective Loss 0.303872 LR 0.001000 Time 0.021831 +2023-10-02 20:36:10,499 - Epoch: [16][ 710/ 1236] Overall Loss 0.304078 Objective Loss 0.304078 LR 0.001000 Time 0.021815 +2023-10-02 20:36:10,708 - Epoch: [16][ 720/ 1236] Overall Loss 0.303850 Objective Loss 0.303850 LR 0.001000 Time 0.021801 +2023-10-02 20:36:10,915 - Epoch: [16][ 730/ 1236] Overall Loss 0.303084 Objective Loss 0.303084 LR 0.001000 Time 0.021786 +2023-10-02 20:36:11,124 - Epoch: [16][ 740/ 1236] Overall Loss 0.303148 Objective Loss 0.303148 LR 0.001000 Time 0.021774 +2023-10-02 20:36:11,331 - Epoch: [16][ 750/ 1236] Overall Loss 0.303226 Objective Loss 0.303226 LR 0.001000 Time 0.021760 +2023-10-02 20:36:11,540 - Epoch: [16][ 760/ 1236] Overall Loss 0.303077 Objective Loss 0.303077 LR 0.001000 Time 0.021748 +2023-10-02 20:36:11,748 - Epoch: [16][ 770/ 1236] Overall Loss 0.303217 Objective Loss 0.303217 LR 0.001000 Time 0.021735 +2023-10-02 20:36:11,956 - Epoch: [16][ 780/ 1236] Overall Loss 0.303319 Objective Loss 0.303319 LR 0.001000 Time 0.021723 +2023-10-02 20:36:12,164 - Epoch: [16][ 790/ 1236] Overall Loss 0.303000 Objective Loss 0.303000 LR 0.001000 Time 0.021710 +2023-10-02 20:36:12,373 - Epoch: [16][ 800/ 1236] Overall Loss 0.303030 Objective Loss 0.303030 LR 0.001000 Time 0.021700 +2023-10-02 20:36:12,580 - Epoch: [16][ 810/ 1236] Overall Loss 0.303295 Objective Loss 0.303295 LR 0.001000 Time 0.021687 +2023-10-02 20:36:12,789 - Epoch: [16][ 820/ 1236] Overall Loss 0.303599 Objective Loss 0.303599 LR 0.001000 Time 0.021677 +2023-10-02 20:36:12,997 - Epoch: [16][ 830/ 1236] Overall Loss 0.303835 Objective Loss 0.303835 LR 0.001000 Time 0.021666 +2023-10-02 20:36:13,205 - Epoch: [16][ 840/ 1236] Overall Loss 0.304315 Objective Loss 0.304315 LR 0.001000 Time 0.021656 +2023-10-02 20:36:13,413 - Epoch: [16][ 850/ 1236] Overall Loss 0.304581 Objective Loss 0.304581 LR 0.001000 Time 0.021645 +2023-10-02 20:36:13,622 - Epoch: [16][ 860/ 1236] Overall Loss 0.304638 Objective Loss 0.304638 LR 0.001000 Time 0.021636 +2023-10-02 20:36:13,829 - Epoch: [16][ 870/ 1236] Overall Loss 0.304575 Objective Loss 0.304575 LR 0.001000 Time 0.021625 +2023-10-02 20:36:14,038 - Epoch: [16][ 880/ 1236] Overall Loss 0.304845 Objective Loss 0.304845 LR 0.001000 Time 0.021617 +2023-10-02 20:36:14,246 - Epoch: [16][ 890/ 1236] Overall Loss 0.305469 Objective Loss 0.305469 LR 0.001000 Time 0.021607 +2023-10-02 20:36:14,455 - Epoch: [16][ 900/ 1236] Overall Loss 0.305441 Objective Loss 0.305441 LR 0.001000 Time 0.021599 +2023-10-02 20:36:14,663 - Epoch: [16][ 910/ 1236] Overall Loss 0.305448 Objective Loss 0.305448 LR 0.001000 Time 0.021589 +2023-10-02 20:36:14,871 - Epoch: [16][ 920/ 1236] Overall Loss 0.305738 Objective Loss 0.305738 LR 0.001000 Time 0.021581 +2023-10-02 20:36:15,079 - Epoch: [16][ 930/ 1236] Overall Loss 0.306084 Objective Loss 0.306084 LR 0.001000 Time 0.021572 +2023-10-02 20:36:15,288 - Epoch: [16][ 940/ 1236] Overall Loss 0.306129 Objective Loss 0.306129 LR 0.001000 Time 0.021564 +2023-10-02 20:36:15,495 - Epoch: [16][ 950/ 1236] Overall Loss 0.306067 Objective Loss 0.306067 LR 0.001000 Time 0.021555 +2023-10-02 20:36:15,704 - Epoch: [16][ 960/ 1236] Overall Loss 0.305785 Objective Loss 0.305785 LR 0.001000 Time 0.021548 +2023-10-02 20:36:15,911 - Epoch: [16][ 970/ 1236] Overall Loss 0.305939 Objective Loss 0.305939 LR 0.001000 Time 0.021539 +2023-10-02 20:36:16,120 - Epoch: [16][ 980/ 1236] Overall Loss 0.306053 Objective Loss 0.306053 LR 0.001000 Time 0.021532 +2023-10-02 20:36:16,327 - Epoch: [16][ 990/ 1236] Overall Loss 0.306086 Objective Loss 0.306086 LR 0.001000 Time 0.021524 +2023-10-02 20:36:16,536 - Epoch: [16][ 1000/ 1236] Overall Loss 0.306082 Objective Loss 0.306082 LR 0.001000 Time 0.021517 +2023-10-02 20:36:16,744 - Epoch: [16][ 1010/ 1236] Overall Loss 0.305934 Objective Loss 0.305934 LR 0.001000 Time 0.021509 +2023-10-02 20:36:16,953 - Epoch: [16][ 1020/ 1236] Overall Loss 0.306012 Objective Loss 0.306012 LR 0.001000 Time 0.021503 +2023-10-02 20:36:17,160 - Epoch: [16][ 1030/ 1236] Overall Loss 0.306210 Objective Loss 0.306210 LR 0.001000 Time 0.021495 +2023-10-02 20:36:17,369 - Epoch: [16][ 1040/ 1236] Overall Loss 0.306163 Objective Loss 0.306163 LR 0.001000 Time 0.021489 +2023-10-02 20:36:17,577 - Epoch: [16][ 1050/ 1236] Overall Loss 0.306267 Objective Loss 0.306267 LR 0.001000 Time 0.021482 +2023-10-02 20:36:17,786 - Epoch: [16][ 1060/ 1236] Overall Loss 0.306358 Objective Loss 0.306358 LR 0.001000 Time 0.021476 +2023-10-02 20:36:17,993 - Epoch: [16][ 1070/ 1236] Overall Loss 0.306252 Objective Loss 0.306252 LR 0.001000 Time 0.021469 +2023-10-02 20:36:18,202 - Epoch: [16][ 1080/ 1236] Overall Loss 0.306413 Objective Loss 0.306413 LR 0.001000 Time 0.021463 +2023-10-02 20:36:18,409 - Epoch: [16][ 1090/ 1236] Overall Loss 0.306162 Objective Loss 0.306162 LR 0.001000 Time 0.021456 +2023-10-02 20:36:18,618 - Epoch: [16][ 1100/ 1236] Overall Loss 0.306296 Objective Loss 0.306296 LR 0.001000 Time 0.021451 +2023-10-02 20:36:18,826 - Epoch: [16][ 1110/ 1236] Overall Loss 0.306470 Objective Loss 0.306470 LR 0.001000 Time 0.021444 +2023-10-02 20:36:19,034 - Epoch: [16][ 1120/ 1236] Overall Loss 0.306539 Objective Loss 0.306539 LR 0.001000 Time 0.021439 +2023-10-02 20:36:19,242 - Epoch: [16][ 1130/ 1236] Overall Loss 0.306693 Objective Loss 0.306693 LR 0.001000 Time 0.021433 +2023-10-02 20:36:19,450 - Epoch: [16][ 1140/ 1236] Overall Loss 0.307160 Objective Loss 0.307160 LR 0.001000 Time 0.021427 +2023-10-02 20:36:19,658 - Epoch: [16][ 1150/ 1236] Overall Loss 0.307416 Objective Loss 0.307416 LR 0.001000 Time 0.021421 +2023-10-02 20:36:19,867 - Epoch: [16][ 1160/ 1236] Overall Loss 0.307533 Objective Loss 0.307533 LR 0.001000 Time 0.021416 +2023-10-02 20:36:20,075 - Epoch: [16][ 1170/ 1236] Overall Loss 0.307791 Objective Loss 0.307791 LR 0.001000 Time 0.021411 +2023-10-02 20:36:20,285 - Epoch: [16][ 1180/ 1236] Overall Loss 0.307771 Objective Loss 0.307771 LR 0.001000 Time 0.021407 +2023-10-02 20:36:20,491 - Epoch: [16][ 1190/ 1236] Overall Loss 0.307490 Objective Loss 0.307490 LR 0.001000 Time 0.021400 +2023-10-02 20:36:20,700 - Epoch: [16][ 1200/ 1236] Overall Loss 0.307682 Objective Loss 0.307682 LR 0.001000 Time 0.021396 +2023-10-02 20:36:20,908 - Epoch: [16][ 1210/ 1236] Overall Loss 0.307821 Objective Loss 0.307821 LR 0.001000 Time 0.021390 +2023-10-02 20:36:21,116 - Epoch: [16][ 1220/ 1236] Overall Loss 0.307873 Objective Loss 0.307873 LR 0.001000 Time 0.021386 +2023-10-02 20:36:21,378 - Epoch: [16][ 1230/ 1236] Overall Loss 0.307783 Objective Loss 0.307783 LR 0.001000 Time 0.021424 +2023-10-02 20:36:21,501 - Epoch: [16][ 1236/ 1236] Overall Loss 0.307695 Objective Loss 0.307695 Top1 82.484725 Top5 98.167006 LR 0.001000 Time 0.021419 +2023-10-02 20:36:21,627 - --- validate (epoch=16)----------- +2023-10-02 20:36:21,627 - 29943 samples (256 per mini-batch) +2023-10-02 20:36:22,124 - Epoch: [16][ 10/ 117] Loss 0.327147 Top1 83.242188 Top5 97.734375 +2023-10-02 20:36:22,276 - Epoch: [16][ 20/ 117] Loss 0.351478 Top1 81.933594 Top5 97.832031 +2023-10-02 20:36:22,426 - Epoch: [16][ 30/ 117] Loss 0.360239 Top1 81.822917 Top5 97.721354 +2023-10-02 20:36:22,577 - Epoch: [16][ 40/ 117] Loss 0.356497 Top1 81.835938 Top5 97.744141 +2023-10-02 20:36:22,728 - Epoch: [16][ 50/ 117] Loss 0.358886 Top1 81.828125 Top5 97.703125 +2023-10-02 20:36:22,878 - Epoch: [16][ 60/ 117] Loss 0.354243 Top1 81.855469 Top5 97.832031 +2023-10-02 20:36:23,029 - Epoch: [16][ 70/ 117] Loss 0.348563 Top1 81.819196 Top5 97.901786 +2023-10-02 20:36:23,179 - Epoch: [16][ 80/ 117] Loss 0.344023 Top1 81.992188 Top5 97.915039 +2023-10-02 20:36:23,330 - Epoch: [16][ 90/ 117] Loss 0.343499 Top1 81.992188 Top5 97.873264 +2023-10-02 20:36:23,480 - Epoch: [16][ 100/ 117] Loss 0.346207 Top1 81.921875 Top5 97.894531 +2023-10-02 20:36:42,476 - Epoch: [16][ 110/ 117] Loss 0.346009 Top1 81.821733 Top5 97.876420 +2023-10-02 20:36:42,568 - Epoch: [16][ 117/ 117] Loss 0.345233 Top1 81.825468 Top5 97.865945 +2023-10-02 20:36:42,716 - ==> Top1: 81.825 Top5: 97.866 Loss: 0.345 + +2023-10-02 20:36:42,716 - ==> Confusion: +[[ 925 1 2 1 5 3 0 0 11 69 2 1 0 1 2 3 7 2 2 0 13] + [ 0 1067 2 1 4 10 2 11 4 1 0 0 0 0 0 6 7 0 6 2 8] + [ 8 0 941 13 4 1 41 5 0 0 2 1 4 1 0 3 1 0 15 7 9] + [ 2 3 19 956 0 2 7 1 12 0 4 0 1 0 38 2 2 1 21 1 17] + [ 28 7 1 0 937 7 1 2 2 13 1 1 1 2 19 8 15 0 1 1 3] + [ 2 89 3 2 4 910 1 28 3 4 3 5 3 9 7 1 7 0 11 12 12] + [ 0 3 21 1 0 0 1143 4 0 0 3 0 1 1 0 3 0 0 3 5 3] + [ 4 46 16 0 0 16 3 1017 2 0 6 4 1 0 1 2 3 0 68 19 10] + [ 21 2 0 0 0 0 0 0 1001 26 7 0 1 4 10 3 2 0 6 4 2] + [ 99 2 1 0 3 0 1 0 58 922 1 2 0 9 9 1 1 0 1 3 6] + [ 1 5 2 12 2 0 5 4 25 2 953 1 1 6 5 1 3 2 10 5 8] + [ 1 0 3 0 0 11 0 4 0 4 0 916 27 3 1 2 13 15 0 25 10] + [ 0 0 1 2 0 0 0 1 2 0 2 31 958 1 6 19 5 16 2 10 12] + [ 2 0 1 0 2 8 2 1 32 21 9 6 4 995 6 5 5 1 0 8 11] + [ 9 0 0 11 4 0 0 0 45 4 4 0 3 0 997 1 2 4 6 0 11] + [ 0 0 2 3 5 1 5 0 0 0 0 6 6 0 0 1072 15 7 3 6 3] + [ 2 15 3 0 2 3 0 0 2 0 0 3 0 0 4 10 1106 0 0 4 7] + [ 0 0 1 0 0 0 2 0 3 1 1 4 15 1 2 10 1 990 1 1 5] + [ 1 6 5 14 0 0 2 10 5 0 2 0 0 0 12 1 0 1 997 3 9] + [ 0 0 2 1 1 2 11 14 0 0 3 1 3 0 1 4 14 1 2 1086 6] + [ 159 263 120 100 110 110 73 62 187 81 148 119 392 229 214 89 283 67 207 280 4612]] + +2023-10-02 20:36:42,718 - ==> Best [Top1: 82.590 Top5: 98.016 Sparsity:0.00 Params: 169472 on epoch: 14] +2023-10-02 20:36:42,718 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:36:42,724 - + +2023-10-02 20:36:42,724 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:36:43,880 - Epoch: [17][ 10/ 1236] Overall Loss 0.296165 Objective Loss 0.296165 LR 0.001000 Time 0.115553 +2023-10-02 20:36:44,088 - Epoch: [17][ 20/ 1236] Overall Loss 0.303367 Objective Loss 0.303367 LR 0.001000 Time 0.068148 +2023-10-02 20:36:44,295 - Epoch: [17][ 30/ 1236] Overall Loss 0.302629 Objective Loss 0.302629 LR 0.001000 Time 0.052313 +2023-10-02 20:36:44,503 - Epoch: [17][ 40/ 1236] Overall Loss 0.299779 Objective Loss 0.299779 LR 0.001000 Time 0.044423 +2023-10-02 20:36:44,709 - Epoch: [17][ 50/ 1236] Overall Loss 0.303387 Objective Loss 0.303387 LR 0.001000 Time 0.039662 +2023-10-02 20:36:44,919 - Epoch: [17][ 60/ 1236] Overall Loss 0.299789 Objective Loss 0.299789 LR 0.001000 Time 0.036542 +2023-10-02 20:36:45,124 - Epoch: [17][ 70/ 1236] Overall Loss 0.301262 Objective Loss 0.301262 LR 0.001000 Time 0.034253 +2023-10-02 20:36:45,333 - Epoch: [17][ 80/ 1236] Overall Loss 0.303725 Objective Loss 0.303725 LR 0.001000 Time 0.032584 +2023-10-02 20:36:45,539 - Epoch: [17][ 90/ 1236] Overall Loss 0.300918 Objective Loss 0.300918 LR 0.001000 Time 0.031241 +2023-10-02 20:36:45,748 - Epoch: [17][ 100/ 1236] Overall Loss 0.299241 Objective Loss 0.299241 LR 0.001000 Time 0.030208 +2023-10-02 20:36:45,954 - Epoch: [17][ 110/ 1236] Overall Loss 0.299417 Objective Loss 0.299417 LR 0.001000 Time 0.029326 +2023-10-02 20:36:46,163 - Epoch: [17][ 120/ 1236] Overall Loss 0.296661 Objective Loss 0.296661 LR 0.001000 Time 0.028624 +2023-10-02 20:36:46,368 - Epoch: [17][ 130/ 1236] Overall Loss 0.295324 Objective Loss 0.295324 LR 0.001000 Time 0.027998 +2023-10-02 20:36:46,576 - Epoch: [17][ 140/ 1236] Overall Loss 0.294978 Objective Loss 0.294978 LR 0.001000 Time 0.027479 +2023-10-02 20:36:46,782 - Epoch: [17][ 150/ 1236] Overall Loss 0.294894 Objective Loss 0.294894 LR 0.001000 Time 0.027023 +2023-10-02 20:36:47,001 - Epoch: [17][ 160/ 1236] Overall Loss 0.294237 Objective Loss 0.294237 LR 0.001000 Time 0.026700 +2023-10-02 20:36:47,215 - Epoch: [17][ 170/ 1236] Overall Loss 0.294166 Objective Loss 0.294166 LR 0.001000 Time 0.026383 +2023-10-02 20:36:47,434 - Epoch: [17][ 180/ 1236] Overall Loss 0.293889 Objective Loss 0.293889 LR 0.001000 Time 0.026131 +2023-10-02 20:36:47,647 - Epoch: [17][ 190/ 1236] Overall Loss 0.294707 Objective Loss 0.294707 LR 0.001000 Time 0.025878 +2023-10-02 20:36:47,866 - Epoch: [17][ 200/ 1236] Overall Loss 0.295502 Objective Loss 0.295502 LR 0.001000 Time 0.025677 +2023-10-02 20:36:48,080 - Epoch: [17][ 210/ 1236] Overall Loss 0.295512 Objective Loss 0.295512 LR 0.001000 Time 0.025470 +2023-10-02 20:36:48,298 - Epoch: [17][ 220/ 1236] Overall Loss 0.294552 Objective Loss 0.294552 LR 0.001000 Time 0.025305 +2023-10-02 20:36:48,512 - Epoch: [17][ 230/ 1236] Overall Loss 0.295129 Objective Loss 0.295129 LR 0.001000 Time 0.025132 +2023-10-02 20:36:48,731 - Epoch: [17][ 240/ 1236] Overall Loss 0.295571 Objective Loss 0.295571 LR 0.001000 Time 0.024995 +2023-10-02 20:36:48,945 - Epoch: [17][ 250/ 1236] Overall Loss 0.295647 Objective Loss 0.295647 LR 0.001000 Time 0.024850 +2023-10-02 20:36:49,164 - Epoch: [17][ 260/ 1236] Overall Loss 0.296030 Objective Loss 0.296030 LR 0.001000 Time 0.024736 +2023-10-02 20:36:49,378 - Epoch: [17][ 270/ 1236] Overall Loss 0.295800 Objective Loss 0.295800 LR 0.001000 Time 0.024610 +2023-10-02 20:36:49,597 - Epoch: [17][ 280/ 1236] Overall Loss 0.296572 Objective Loss 0.296572 LR 0.001000 Time 0.024512 +2023-10-02 20:36:49,811 - Epoch: [17][ 290/ 1236] Overall Loss 0.297623 Objective Loss 0.297623 LR 0.001000 Time 0.024403 +2023-10-02 20:36:50,018 - Epoch: [17][ 300/ 1236] Overall Loss 0.296991 Objective Loss 0.296991 LR 0.001000 Time 0.024280 +2023-10-02 20:36:50,225 - Epoch: [17][ 310/ 1236] Overall Loss 0.296364 Objective Loss 0.296364 LR 0.001000 Time 0.024163 +2023-10-02 20:36:50,434 - Epoch: [17][ 320/ 1236] Overall Loss 0.296102 Objective Loss 0.296102 LR 0.001000 Time 0.024059 +2023-10-02 20:36:50,641 - Epoch: [17][ 330/ 1236] Overall Loss 0.296617 Objective Loss 0.296617 LR 0.001000 Time 0.023957 +2023-10-02 20:36:50,850 - Epoch: [17][ 340/ 1236] Overall Loss 0.297121 Objective Loss 0.297121 LR 0.001000 Time 0.023866 +2023-10-02 20:36:51,057 - Epoch: [17][ 350/ 1236] Overall Loss 0.297396 Objective Loss 0.297396 LR 0.001000 Time 0.023775 +2023-10-02 20:36:51,266 - Epoch: [17][ 360/ 1236] Overall Loss 0.297063 Objective Loss 0.297063 LR 0.001000 Time 0.023693 +2023-10-02 20:36:51,473 - Epoch: [17][ 370/ 1236] Overall Loss 0.297414 Objective Loss 0.297414 LR 0.001000 Time 0.023609 +2023-10-02 20:36:51,681 - Epoch: [17][ 380/ 1236] Overall Loss 0.297227 Objective Loss 0.297227 LR 0.001000 Time 0.023535 +2023-10-02 20:36:51,888 - Epoch: [17][ 390/ 1236] Overall Loss 0.297545 Objective Loss 0.297545 LR 0.001000 Time 0.023462 +2023-10-02 20:36:52,097 - Epoch: [17][ 400/ 1236] Overall Loss 0.297264 Objective Loss 0.297264 LR 0.001000 Time 0.023396 +2023-10-02 20:36:52,304 - Epoch: [17][ 410/ 1236] Overall Loss 0.297004 Objective Loss 0.297004 LR 0.001000 Time 0.023330 +2023-10-02 20:36:52,513 - Epoch: [17][ 420/ 1236] Overall Loss 0.297270 Objective Loss 0.297270 LR 0.001000 Time 0.023270 +2023-10-02 20:36:52,720 - Epoch: [17][ 430/ 1236] Overall Loss 0.297685 Objective Loss 0.297685 LR 0.001000 Time 0.023210 +2023-10-02 20:36:52,929 - Epoch: [17][ 440/ 1236] Overall Loss 0.296977 Objective Loss 0.296977 LR 0.001000 Time 0.023157 +2023-10-02 20:36:53,136 - Epoch: [17][ 450/ 1236] Overall Loss 0.296693 Objective Loss 0.296693 LR 0.001000 Time 0.023102 +2023-10-02 20:36:53,344 - Epoch: [17][ 460/ 1236] Overall Loss 0.296988 Objective Loss 0.296988 LR 0.001000 Time 0.023052 +2023-10-02 20:36:53,551 - Epoch: [17][ 470/ 1236] Overall Loss 0.297165 Objective Loss 0.297165 LR 0.001000 Time 0.023001 +2023-10-02 20:36:53,760 - Epoch: [17][ 480/ 1236] Overall Loss 0.296896 Objective Loss 0.296896 LR 0.001000 Time 0.022956 +2023-10-02 20:36:53,967 - Epoch: [17][ 490/ 1236] Overall Loss 0.296462 Objective Loss 0.296462 LR 0.001000 Time 0.022910 +2023-10-02 20:36:54,176 - Epoch: [17][ 500/ 1236] Overall Loss 0.296002 Objective Loss 0.296002 LR 0.001000 Time 0.022868 +2023-10-02 20:36:54,383 - Epoch: [17][ 510/ 1236] Overall Loss 0.295898 Objective Loss 0.295898 LR 0.001000 Time 0.022826 +2023-10-02 20:36:54,591 - Epoch: [17][ 520/ 1236] Overall Loss 0.296690 Objective Loss 0.296690 LR 0.001000 Time 0.022787 +2023-10-02 20:36:54,799 - Epoch: [17][ 530/ 1236] Overall Loss 0.296534 Objective Loss 0.296534 LR 0.001000 Time 0.022747 +2023-10-02 20:36:55,007 - Epoch: [17][ 540/ 1236] Overall Loss 0.296607 Objective Loss 0.296607 LR 0.001000 Time 0.022712 +2023-10-02 20:36:55,215 - Epoch: [17][ 550/ 1236] Overall Loss 0.296206 Objective Loss 0.296206 LR 0.001000 Time 0.022675 +2023-10-02 20:36:55,423 - Epoch: [17][ 560/ 1236] Overall Loss 0.296997 Objective Loss 0.296997 LR 0.001000 Time 0.022643 +2023-10-02 20:36:55,631 - Epoch: [17][ 570/ 1236] Overall Loss 0.296702 Objective Loss 0.296702 LR 0.001000 Time 0.022609 +2023-10-02 20:36:55,839 - Epoch: [17][ 580/ 1236] Overall Loss 0.296610 Objective Loss 0.296610 LR 0.001000 Time 0.022578 +2023-10-02 20:36:56,047 - Epoch: [17][ 590/ 1236] Overall Loss 0.296954 Objective Loss 0.296954 LR 0.001000 Time 0.022546 +2023-10-02 20:36:56,255 - Epoch: [17][ 600/ 1236] Overall Loss 0.297205 Objective Loss 0.297205 LR 0.001000 Time 0.022517 +2023-10-02 20:36:56,463 - Epoch: [17][ 610/ 1236] Overall Loss 0.297276 Objective Loss 0.297276 LR 0.001000 Time 0.022488 +2023-10-02 20:36:56,672 - Epoch: [17][ 620/ 1236] Overall Loss 0.297153 Objective Loss 0.297153 LR 0.001000 Time 0.022462 +2023-10-02 20:36:56,879 - Epoch: [17][ 630/ 1236] Overall Loss 0.297184 Objective Loss 0.297184 LR 0.001000 Time 0.022434 +2023-10-02 20:36:57,088 - Epoch: [17][ 640/ 1236] Overall Loss 0.297275 Objective Loss 0.297275 LR 0.001000 Time 0.022409 +2023-10-02 20:36:57,295 - Epoch: [17][ 650/ 1236] Overall Loss 0.297834 Objective Loss 0.297834 LR 0.001000 Time 0.022383 +2023-10-02 20:36:57,504 - Epoch: [17][ 660/ 1236] Overall Loss 0.298157 Objective Loss 0.298157 LR 0.001000 Time 0.022359 +2023-10-02 20:36:57,711 - Epoch: [17][ 670/ 1236] Overall Loss 0.298702 Objective Loss 0.298702 LR 0.001000 Time 0.022334 +2023-10-02 20:36:57,920 - Epoch: [17][ 680/ 1236] Overall Loss 0.298907 Objective Loss 0.298907 LR 0.001000 Time 0.022313 +2023-10-02 20:36:58,127 - Epoch: [17][ 690/ 1236] Overall Loss 0.299378 Objective Loss 0.299378 LR 0.001000 Time 0.022289 +2023-10-02 20:36:58,335 - Epoch: [17][ 700/ 1236] Overall Loss 0.299865 Objective Loss 0.299865 LR 0.001000 Time 0.022268 +2023-10-02 20:36:58,543 - Epoch: [17][ 710/ 1236] Overall Loss 0.300073 Objective Loss 0.300073 LR 0.001000 Time 0.022246 +2023-10-02 20:36:58,751 - Epoch: [17][ 720/ 1236] Overall Loss 0.300617 Objective Loss 0.300617 LR 0.001000 Time 0.022226 +2023-10-02 20:36:58,960 - Epoch: [17][ 730/ 1236] Overall Loss 0.300830 Objective Loss 0.300830 LR 0.001000 Time 0.022206 +2023-10-02 20:36:59,168 - Epoch: [17][ 740/ 1236] Overall Loss 0.300894 Objective Loss 0.300894 LR 0.001000 Time 0.022188 +2023-10-02 20:36:59,376 - Epoch: [17][ 750/ 1236] Overall Loss 0.301062 Objective Loss 0.301062 LR 0.001000 Time 0.022168 +2023-10-02 20:36:59,584 - Epoch: [17][ 760/ 1236] Overall Loss 0.301300 Objective Loss 0.301300 LR 0.001000 Time 0.022151 +2023-10-02 20:36:59,792 - Epoch: [17][ 770/ 1236] Overall Loss 0.301103 Objective Loss 0.301103 LR 0.001000 Time 0.022131 +2023-10-02 20:37:00,000 - Epoch: [17][ 780/ 1236] Overall Loss 0.301120 Objective Loss 0.301120 LR 0.001000 Time 0.022114 +2023-10-02 20:37:00,208 - Epoch: [17][ 790/ 1236] Overall Loss 0.301378 Objective Loss 0.301378 LR 0.001000 Time 0.022096 +2023-10-02 20:37:00,416 - Epoch: [17][ 800/ 1236] Overall Loss 0.301620 Objective Loss 0.301620 LR 0.001000 Time 0.022080 +2023-10-02 20:37:00,624 - Epoch: [17][ 810/ 1236] Overall Loss 0.301547 Objective Loss 0.301547 LR 0.001000 Time 0.022063 +2023-10-02 20:37:00,832 - Epoch: [17][ 820/ 1236] Overall Loss 0.301722 Objective Loss 0.301722 LR 0.001000 Time 0.022048 +2023-10-02 20:37:01,040 - Epoch: [17][ 830/ 1236] Overall Loss 0.301355 Objective Loss 0.301355 LR 0.001000 Time 0.022032 +2023-10-02 20:37:01,248 - Epoch: [17][ 840/ 1236] Overall Loss 0.300942 Objective Loss 0.300942 LR 0.001000 Time 0.022018 +2023-10-02 20:37:01,456 - Epoch: [17][ 850/ 1236] Overall Loss 0.300749 Objective Loss 0.300749 LR 0.001000 Time 0.022002 +2023-10-02 20:37:01,664 - Epoch: [17][ 860/ 1236] Overall Loss 0.301141 Objective Loss 0.301141 LR 0.001000 Time 0.021989 +2023-10-02 20:37:01,872 - Epoch: [17][ 870/ 1236] Overall Loss 0.301194 Objective Loss 0.301194 LR 0.001000 Time 0.021974 +2023-10-02 20:37:02,080 - Epoch: [17][ 880/ 1236] Overall Loss 0.301217 Objective Loss 0.301217 LR 0.001000 Time 0.021961 +2023-10-02 20:37:02,288 - Epoch: [17][ 890/ 1236] Overall Loss 0.301481 Objective Loss 0.301481 LR 0.001000 Time 0.021947 +2023-10-02 20:37:02,497 - Epoch: [17][ 900/ 1236] Overall Loss 0.301493 Objective Loss 0.301493 LR 0.001000 Time 0.021935 +2023-10-02 20:37:02,704 - Epoch: [17][ 910/ 1236] Overall Loss 0.301546 Objective Loss 0.301546 LR 0.001000 Time 0.021921 +2023-10-02 20:37:02,913 - Epoch: [17][ 920/ 1236] Overall Loss 0.301298 Objective Loss 0.301298 LR 0.001000 Time 0.021910 +2023-10-02 20:37:03,120 - Epoch: [17][ 930/ 1236] Overall Loss 0.301514 Objective Loss 0.301514 LR 0.001000 Time 0.021897 +2023-10-02 20:37:03,329 - Epoch: [17][ 940/ 1236] Overall Loss 0.301715 Objective Loss 0.301715 LR 0.001000 Time 0.021886 +2023-10-02 20:37:03,537 - Epoch: [17][ 950/ 1236] Overall Loss 0.301623 Objective Loss 0.301623 LR 0.001000 Time 0.021874 +2023-10-02 20:37:03,745 - Epoch: [17][ 960/ 1236] Overall Loss 0.301674 Objective Loss 0.301674 LR 0.001000 Time 0.021863 +2023-10-02 20:37:03,953 - Epoch: [17][ 970/ 1236] Overall Loss 0.301647 Objective Loss 0.301647 LR 0.001000 Time 0.021849 +2023-10-02 20:37:04,161 - Epoch: [17][ 980/ 1236] Overall Loss 0.301771 Objective Loss 0.301771 LR 0.001000 Time 0.021839 +2023-10-02 20:37:04,368 - Epoch: [17][ 990/ 1236] Overall Loss 0.301955 Objective Loss 0.301955 LR 0.001000 Time 0.021827 +2023-10-02 20:37:04,577 - Epoch: [17][ 1000/ 1236] Overall Loss 0.301797 Objective Loss 0.301797 LR 0.001000 Time 0.021817 +2023-10-02 20:37:04,784 - Epoch: [17][ 1010/ 1236] Overall Loss 0.301725 Objective Loss 0.301725 LR 0.001000 Time 0.021806 +2023-10-02 20:37:04,993 - Epoch: [17][ 1020/ 1236] Overall Loss 0.301428 Objective Loss 0.301428 LR 0.001000 Time 0.021796 +2023-10-02 20:37:05,200 - Epoch: [17][ 1030/ 1236] Overall Loss 0.301453 Objective Loss 0.301453 LR 0.001000 Time 0.021786 +2023-10-02 20:37:05,409 - Epoch: [17][ 1040/ 1236] Overall Loss 0.302003 Objective Loss 0.302003 LR 0.001000 Time 0.021777 +2023-10-02 20:37:05,616 - Epoch: [17][ 1050/ 1236] Overall Loss 0.301919 Objective Loss 0.301919 LR 0.001000 Time 0.021766 +2023-10-02 20:37:05,824 - Epoch: [17][ 1060/ 1236] Overall Loss 0.302228 Objective Loss 0.302228 LR 0.001000 Time 0.021757 +2023-10-02 20:37:06,031 - Epoch: [17][ 1070/ 1236] Overall Loss 0.302324 Objective Loss 0.302324 LR 0.001000 Time 0.021747 +2023-10-02 20:37:06,240 - Epoch: [17][ 1080/ 1236] Overall Loss 0.302706 Objective Loss 0.302706 LR 0.001000 Time 0.021739 +2023-10-02 20:37:06,448 - Epoch: [17][ 1090/ 1236] Overall Loss 0.302779 Objective Loss 0.302779 LR 0.001000 Time 0.021729 +2023-10-02 20:37:06,656 - Epoch: [17][ 1100/ 1236] Overall Loss 0.302869 Objective Loss 0.302869 LR 0.001000 Time 0.021721 +2023-10-02 20:37:06,863 - Epoch: [17][ 1110/ 1236] Overall Loss 0.302808 Objective Loss 0.302808 LR 0.001000 Time 0.021712 +2023-10-02 20:37:07,072 - Epoch: [17][ 1120/ 1236] Overall Loss 0.302805 Objective Loss 0.302805 LR 0.001000 Time 0.021704 +2023-10-02 20:37:07,279 - Epoch: [17][ 1130/ 1236] Overall Loss 0.303004 Objective Loss 0.303004 LR 0.001000 Time 0.021695 +2023-10-02 20:37:07,488 - Epoch: [17][ 1140/ 1236] Overall Loss 0.302955 Objective Loss 0.302955 LR 0.001000 Time 0.021688 +2023-10-02 20:37:07,695 - Epoch: [17][ 1150/ 1236] Overall Loss 0.303150 Objective Loss 0.303150 LR 0.001000 Time 0.021679 +2023-10-02 20:37:07,904 - Epoch: [17][ 1160/ 1236] Overall Loss 0.303051 Objective Loss 0.303051 LR 0.001000 Time 0.021671 +2023-10-02 20:37:08,111 - Epoch: [17][ 1170/ 1236] Overall Loss 0.303187 Objective Loss 0.303187 LR 0.001000 Time 0.021663 +2023-10-02 20:37:08,319 - Epoch: [17][ 1180/ 1236] Overall Loss 0.302900 Objective Loss 0.302900 LR 0.001000 Time 0.021656 +2023-10-02 20:37:08,527 - Epoch: [17][ 1190/ 1236] Overall Loss 0.303137 Objective Loss 0.303137 LR 0.001000 Time 0.021648 +2023-10-02 20:37:08,736 - Epoch: [17][ 1200/ 1236] Overall Loss 0.303062 Objective Loss 0.303062 LR 0.001000 Time 0.021641 +2023-10-02 20:37:08,943 - Epoch: [17][ 1210/ 1236] Overall Loss 0.302879 Objective Loss 0.302879 LR 0.001000 Time 0.021634 +2023-10-02 20:37:09,152 - Epoch: [17][ 1220/ 1236] Overall Loss 0.303078 Objective Loss 0.303078 LR 0.001000 Time 0.021628 +2023-10-02 20:37:09,414 - Epoch: [17][ 1230/ 1236] Overall Loss 0.303194 Objective Loss 0.303194 LR 0.001000 Time 0.021664 +2023-10-02 20:37:09,537 - Epoch: [17][ 1236/ 1236] Overall Loss 0.303336 Objective Loss 0.303336 Top1 84.114053 Top5 97.963340 LR 0.001000 Time 0.021658 +2023-10-02 20:37:09,665 - --- validate (epoch=17)----------- +2023-10-02 20:37:09,665 - 29943 samples (256 per mini-batch) +2023-10-02 20:37:10,143 - Epoch: [17][ 10/ 117] Loss 0.338277 Top1 82.617188 Top5 97.890625 +2023-10-02 20:37:10,294 - Epoch: [17][ 20/ 117] Loss 0.323206 Top1 82.890625 Top5 97.929688 +2023-10-02 20:37:10,445 - Epoch: [17][ 30/ 117] Loss 0.322957 Top1 83.203125 Top5 97.877604 +2023-10-02 20:37:10,595 - Epoch: [17][ 40/ 117] Loss 0.318081 Top1 83.183594 Top5 97.978516 +2023-10-02 20:37:10,745 - Epoch: [17][ 50/ 117] Loss 0.326411 Top1 82.976562 Top5 97.914062 +2023-10-02 20:37:10,895 - Epoch: [17][ 60/ 117] Loss 0.328762 Top1 82.884115 Top5 97.923177 +2023-10-02 20:37:11,045 - Epoch: [17][ 70/ 117] Loss 0.325544 Top1 83.069196 Top5 97.968750 +2023-10-02 20:37:11,195 - Epoch: [17][ 80/ 117] Loss 0.328379 Top1 82.802734 Top5 97.958984 +2023-10-02 20:37:11,345 - Epoch: [17][ 90/ 117] Loss 0.330567 Top1 82.699653 Top5 97.977431 +2023-10-02 20:37:11,496 - Epoch: [17][ 100/ 117] Loss 0.334087 Top1 82.535156 Top5 97.921875 +2023-10-02 20:37:11,654 - Epoch: [17][ 110/ 117] Loss 0.332529 Top1 82.588778 Top5 97.926136 +2023-10-02 20:37:11,744 - Epoch: [17][ 117/ 117] Loss 0.331097 Top1 82.590255 Top5 97.939418 +2023-10-02 20:37:11,870 - ==> Top1: 82.590 Top5: 97.939 Loss: 0.331 + +2023-10-02 20:37:11,871 - ==> Confusion: +[[ 940 3 5 0 7 3 0 0 6 59 2 0 2 4 6 4 1 2 0 0 6] + [ 2 1066 0 1 6 21 1 11 4 1 1 0 1 0 0 3 0 0 9 0 4] + [ 3 0 963 20 4 1 20 7 0 1 6 3 6 1 0 2 1 2 7 1 8] + [ 4 2 17 995 0 4 1 2 8 0 4 0 3 0 26 0 1 1 6 0 15] + [ 30 9 3 0 946 5 0 0 2 12 1 1 3 3 14 5 6 0 1 3 6] + [ 1 43 2 2 5 976 1 22 2 3 4 11 1 15 4 1 2 1 6 0 14] + [ 0 2 35 0 0 2 1121 11 0 0 3 2 2 0 0 2 0 1 1 5 4] + [ 2 27 18 1 1 25 0 1064 2 0 3 6 5 1 1 0 0 0 49 5 8] + [ 21 2 0 1 1 2 0 0 980 39 9 1 4 5 11 1 1 4 4 1 2] + [ 142 2 3 0 3 1 0 0 46 868 1 0 0 26 11 1 0 2 0 4 9] + [ 4 3 3 17 0 2 1 4 13 1 972 3 0 6 5 1 1 3 5 1 8] + [ 1 1 3 0 2 13 0 2 0 0 0 947 17 4 0 0 4 21 0 9 11] + [ 1 4 4 1 0 2 1 1 2 0 1 49 943 3 7 10 1 17 5 2 14] + [ 0 0 2 0 4 11 1 0 8 8 18 8 2 1038 4 0 2 1 1 3 8] + [ 8 1 3 22 2 0 0 0 29 2 5 0 0 3 1010 0 0 3 4 0 9] + [ 1 0 4 2 3 1 5 0 0 1 0 10 6 1 0 1048 15 21 0 8 8] + [ 1 19 1 0 3 5 1 0 2 0 0 5 0 0 3 9 1088 2 0 7 15] + [ 0 0 0 2 0 1 2 1 0 0 1 2 20 0 1 5 3 996 0 0 4] + [ 1 9 9 23 0 0 0 15 8 0 5 0 0 0 12 1 1 0 972 0 12] + [ 0 7 8 3 3 9 10 19 2 0 4 8 5 0 0 5 4 0 2 1056 7] + [ 211 250 142 113 101 163 51 80 128 110 270 160 352 278 167 51 114 77 154 192 4741]] + +2023-10-02 20:37:11,872 - ==> Best [Top1: 82.590 Top5: 98.016 Sparsity:0.00 Params: 169472 on epoch: 14] +2023-10-02 20:37:11,872 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:37:11,878 - + +2023-10-02 20:37:11,878 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:37:12,915 - Epoch: [18][ 10/ 1236] Overall Loss 0.290554 Objective Loss 0.290554 LR 0.001000 Time 0.103620 +2023-10-02 20:37:13,126 - Epoch: [18][ 20/ 1236] Overall Loss 0.300097 Objective Loss 0.300097 LR 0.001000 Time 0.062309 +2023-10-02 20:37:13,332 - Epoch: [18][ 30/ 1236] Overall Loss 0.289872 Objective Loss 0.289872 LR 0.001000 Time 0.048380 +2023-10-02 20:37:13,542 - Epoch: [18][ 40/ 1236] Overall Loss 0.288264 Objective Loss 0.288264 LR 0.001000 Time 0.041525 +2023-10-02 20:37:13,750 - Epoch: [18][ 50/ 1236] Overall Loss 0.289488 Objective Loss 0.289488 LR 0.001000 Time 0.037374 +2023-10-02 20:37:13,960 - Epoch: [18][ 60/ 1236] Overall Loss 0.289336 Objective Loss 0.289336 LR 0.001000 Time 0.034632 +2023-10-02 20:37:14,168 - Epoch: [18][ 70/ 1236] Overall Loss 0.289623 Objective Loss 0.289623 LR 0.001000 Time 0.032652 +2023-10-02 20:37:14,377 - Epoch: [18][ 80/ 1236] Overall Loss 0.289497 Objective Loss 0.289497 LR 0.001000 Time 0.031188 +2023-10-02 20:37:14,585 - Epoch: [18][ 90/ 1236] Overall Loss 0.293028 Objective Loss 0.293028 LR 0.001000 Time 0.030030 +2023-10-02 20:37:14,795 - Epoch: [18][ 100/ 1236] Overall Loss 0.294628 Objective Loss 0.294628 LR 0.001000 Time 0.029123 +2023-10-02 20:37:15,004 - Epoch: [18][ 110/ 1236] Overall Loss 0.294641 Objective Loss 0.294641 LR 0.001000 Time 0.028355 +2023-10-02 20:37:15,213 - Epoch: [18][ 120/ 1236] Overall Loss 0.294513 Objective Loss 0.294513 LR 0.001000 Time 0.027737 +2023-10-02 20:37:15,421 - Epoch: [18][ 130/ 1236] Overall Loss 0.291749 Objective Loss 0.291749 LR 0.001000 Time 0.027188 +2023-10-02 20:37:15,631 - Epoch: [18][ 140/ 1236] Overall Loss 0.290030 Objective Loss 0.290030 LR 0.001000 Time 0.026745 +2023-10-02 20:37:15,839 - Epoch: [18][ 150/ 1236] Overall Loss 0.290275 Objective Loss 0.290275 LR 0.001000 Time 0.026340 +2023-10-02 20:37:16,049 - Epoch: [18][ 160/ 1236] Overall Loss 0.290381 Objective Loss 0.290381 LR 0.001000 Time 0.026005 +2023-10-02 20:37:16,257 - Epoch: [18][ 170/ 1236] Overall Loss 0.289642 Objective Loss 0.289642 LR 0.001000 Time 0.025688 +2023-10-02 20:37:16,467 - Epoch: [18][ 180/ 1236] Overall Loss 0.290353 Objective Loss 0.290353 LR 0.001000 Time 0.025426 +2023-10-02 20:37:16,675 - Epoch: [18][ 190/ 1236] Overall Loss 0.289043 Objective Loss 0.289043 LR 0.001000 Time 0.025173 +2023-10-02 20:37:16,885 - Epoch: [18][ 200/ 1236] Overall Loss 0.287609 Objective Loss 0.287609 LR 0.001000 Time 0.024963 +2023-10-02 20:37:17,093 - Epoch: [18][ 210/ 1236] Overall Loss 0.288384 Objective Loss 0.288384 LR 0.001000 Time 0.024757 +2023-10-02 20:37:17,303 - Epoch: [18][ 220/ 1236] Overall Loss 0.289496 Objective Loss 0.289496 LR 0.001000 Time 0.024586 +2023-10-02 20:37:17,511 - Epoch: [18][ 230/ 1236] Overall Loss 0.289147 Objective Loss 0.289147 LR 0.001000 Time 0.024415 +2023-10-02 20:37:17,721 - Epoch: [18][ 240/ 1236] Overall Loss 0.288127 Objective Loss 0.288127 LR 0.001000 Time 0.024271 +2023-10-02 20:37:17,929 - Epoch: [18][ 250/ 1236] Overall Loss 0.289267 Objective Loss 0.289267 LR 0.001000 Time 0.024125 +2023-10-02 20:37:18,138 - Epoch: [18][ 260/ 1236] Overall Loss 0.289471 Objective Loss 0.289471 LR 0.001000 Time 0.024003 +2023-10-02 20:37:18,346 - Epoch: [18][ 270/ 1236] Overall Loss 0.288607 Objective Loss 0.288607 LR 0.001000 Time 0.023878 +2023-10-02 20:37:18,556 - Epoch: [18][ 280/ 1236] Overall Loss 0.288425 Objective Loss 0.288425 LR 0.001000 Time 0.023773 +2023-10-02 20:37:18,764 - Epoch: [18][ 290/ 1236] Overall Loss 0.287723 Objective Loss 0.287723 LR 0.001000 Time 0.023664 +2023-10-02 20:37:18,974 - Epoch: [18][ 300/ 1236] Overall Loss 0.287956 Objective Loss 0.287956 LR 0.001000 Time 0.023574 +2023-10-02 20:37:19,182 - Epoch: [18][ 310/ 1236] Overall Loss 0.287220 Objective Loss 0.287220 LR 0.001000 Time 0.023481 +2023-10-02 20:37:19,392 - Epoch: [18][ 320/ 1236] Overall Loss 0.288142 Objective Loss 0.288142 LR 0.001000 Time 0.023402 +2023-10-02 20:37:19,600 - Epoch: [18][ 330/ 1236] Overall Loss 0.288848 Objective Loss 0.288848 LR 0.001000 Time 0.023324 +2023-10-02 20:37:19,810 - Epoch: [18][ 340/ 1236] Overall Loss 0.289364 Objective Loss 0.289364 LR 0.001000 Time 0.023254 +2023-10-02 20:37:20,018 - Epoch: [18][ 350/ 1236] Overall Loss 0.290419 Objective Loss 0.290419 LR 0.001000 Time 0.023179 +2023-10-02 20:37:20,228 - Epoch: [18][ 360/ 1236] Overall Loss 0.290720 Objective Loss 0.290720 LR 0.001000 Time 0.023118 +2023-10-02 20:37:20,436 - Epoch: [18][ 370/ 1236] Overall Loss 0.290543 Objective Loss 0.290543 LR 0.001000 Time 0.023056 +2023-10-02 20:37:20,646 - Epoch: [18][ 380/ 1236] Overall Loss 0.290671 Objective Loss 0.290671 LR 0.001000 Time 0.023001 +2023-10-02 20:37:20,855 - Epoch: [18][ 390/ 1236] Overall Loss 0.290450 Objective Loss 0.290450 LR 0.001000 Time 0.022944 +2023-10-02 20:37:21,065 - Epoch: [18][ 400/ 1236] Overall Loss 0.290501 Objective Loss 0.290501 LR 0.001000 Time 0.022895 +2023-10-02 20:37:21,272 - Epoch: [18][ 410/ 1236] Overall Loss 0.291366 Objective Loss 0.291366 LR 0.001000 Time 0.022840 +2023-10-02 20:37:21,480 - Epoch: [18][ 420/ 1236] Overall Loss 0.292015 Objective Loss 0.292015 LR 0.001000 Time 0.022793 +2023-10-02 20:37:21,686 - Epoch: [18][ 430/ 1236] Overall Loss 0.291990 Objective Loss 0.291990 LR 0.001000 Time 0.022740 +2023-10-02 20:37:21,895 - Epoch: [18][ 440/ 1236] Overall Loss 0.292523 Objective Loss 0.292523 LR 0.001000 Time 0.022697 +2023-10-02 20:37:22,100 - Epoch: [18][ 450/ 1236] Overall Loss 0.293264 Objective Loss 0.293264 LR 0.001000 Time 0.022649 +2023-10-02 20:37:22,308 - Epoch: [18][ 460/ 1236] Overall Loss 0.292417 Objective Loss 0.292417 LR 0.001000 Time 0.022607 +2023-10-02 20:37:22,515 - Epoch: [18][ 470/ 1236] Overall Loss 0.292700 Objective Loss 0.292700 LR 0.001000 Time 0.022566 +2023-10-02 20:37:22,722 - Epoch: [18][ 480/ 1236] Overall Loss 0.292711 Objective Loss 0.292711 LR 0.001000 Time 0.022527 +2023-10-02 20:37:22,929 - Epoch: [18][ 490/ 1236] Overall Loss 0.293237 Objective Loss 0.293237 LR 0.001000 Time 0.022489 +2023-10-02 20:37:23,138 - Epoch: [18][ 500/ 1236] Overall Loss 0.293366 Objective Loss 0.293366 LR 0.001000 Time 0.022456 +2023-10-02 20:37:23,343 - Epoch: [18][ 510/ 1236] Overall Loss 0.293142 Objective Loss 0.293142 LR 0.001000 Time 0.022418 +2023-10-02 20:37:23,551 - Epoch: [18][ 520/ 1236] Overall Loss 0.292439 Objective Loss 0.292439 LR 0.001000 Time 0.022386 +2023-10-02 20:37:23,758 - Epoch: [18][ 530/ 1236] Overall Loss 0.292381 Objective Loss 0.292381 LR 0.001000 Time 0.022353 +2023-10-02 20:37:23,967 - Epoch: [18][ 540/ 1236] Overall Loss 0.292374 Objective Loss 0.292374 LR 0.001000 Time 0.022325 +2023-10-02 20:37:24,171 - Epoch: [18][ 550/ 1236] Overall Loss 0.292687 Objective Loss 0.292687 LR 0.001000 Time 0.022291 +2023-10-02 20:37:24,379 - Epoch: [18][ 560/ 1236] Overall Loss 0.292128 Objective Loss 0.292128 LR 0.001000 Time 0.022264 +2023-10-02 20:37:24,586 - Epoch: [18][ 570/ 1236] Overall Loss 0.292318 Objective Loss 0.292318 LR 0.001000 Time 0.022233 +2023-10-02 20:37:24,795 - Epoch: [18][ 580/ 1236] Overall Loss 0.292916 Objective Loss 0.292916 LR 0.001000 Time 0.022209 +2023-10-02 20:37:25,000 - Epoch: [18][ 590/ 1236] Overall Loss 0.293151 Objective Loss 0.293151 LR 0.001000 Time 0.022181 +2023-10-02 20:37:25,210 - Epoch: [18][ 600/ 1236] Overall Loss 0.293357 Objective Loss 0.293357 LR 0.001000 Time 0.022159 +2023-10-02 20:37:25,415 - Epoch: [18][ 610/ 1236] Overall Loss 0.293161 Objective Loss 0.293161 LR 0.001000 Time 0.022133 +2023-10-02 20:37:25,623 - Epoch: [18][ 620/ 1236] Overall Loss 0.292914 Objective Loss 0.292914 LR 0.001000 Time 0.022110 +2023-10-02 20:37:25,830 - Epoch: [18][ 630/ 1236] Overall Loss 0.292847 Objective Loss 0.292847 LR 0.001000 Time 0.022085 +2023-10-02 20:37:26,038 - Epoch: [18][ 640/ 1236] Overall Loss 0.293089 Objective Loss 0.293089 LR 0.001000 Time 0.022065 +2023-10-02 20:37:26,244 - Epoch: [18][ 650/ 1236] Overall Loss 0.293423 Objective Loss 0.293423 LR 0.001000 Time 0.022042 +2023-10-02 20:37:26,453 - Epoch: [18][ 660/ 1236] Overall Loss 0.293668 Objective Loss 0.293668 LR 0.001000 Time 0.022024 +2023-10-02 20:37:26,658 - Epoch: [18][ 670/ 1236] Overall Loss 0.293625 Objective Loss 0.293625 LR 0.001000 Time 0.022002 +2023-10-02 20:37:26,866 - Epoch: [18][ 680/ 1236] Overall Loss 0.293561 Objective Loss 0.293561 LR 0.001000 Time 0.021983 +2023-10-02 20:37:27,073 - Epoch: [18][ 690/ 1236] Overall Loss 0.293505 Objective Loss 0.293505 LR 0.001000 Time 0.021964 +2023-10-02 20:37:27,280 - Epoch: [18][ 700/ 1236] Overall Loss 0.293242 Objective Loss 0.293242 LR 0.001000 Time 0.021946 +2023-10-02 20:37:27,487 - Epoch: [18][ 710/ 1236] Overall Loss 0.293110 Objective Loss 0.293110 LR 0.001000 Time 0.021928 +2023-10-02 20:37:27,695 - Epoch: [18][ 720/ 1236] Overall Loss 0.293682 Objective Loss 0.293682 LR 0.001000 Time 0.021911 +2023-10-02 20:37:27,902 - Epoch: [18][ 730/ 1236] Overall Loss 0.293530 Objective Loss 0.293530 LR 0.001000 Time 0.021894 +2023-10-02 20:37:28,111 - Epoch: [18][ 740/ 1236] Overall Loss 0.293731 Objective Loss 0.293731 LR 0.001000 Time 0.021880 +2023-10-02 20:37:28,316 - Epoch: [18][ 750/ 1236] Overall Loss 0.294213 Objective Loss 0.294213 LR 0.001000 Time 0.021862 +2023-10-02 20:37:28,524 - Epoch: [18][ 760/ 1236] Overall Loss 0.294492 Objective Loss 0.294492 LR 0.001000 Time 0.021847 +2023-10-02 20:37:28,731 - Epoch: [18][ 770/ 1236] Overall Loss 0.294241 Objective Loss 0.294241 LR 0.001000 Time 0.021832 +2023-10-02 20:37:28,939 - Epoch: [18][ 780/ 1236] Overall Loss 0.294353 Objective Loss 0.294353 LR 0.001000 Time 0.021819 +2023-10-02 20:37:29,145 - Epoch: [18][ 790/ 1236] Overall Loss 0.294569 Objective Loss 0.294569 LR 0.001000 Time 0.021803 +2023-10-02 20:37:29,354 - Epoch: [18][ 800/ 1236] Overall Loss 0.294498 Objective Loss 0.294498 LR 0.001000 Time 0.021791 +2023-10-02 20:37:29,560 - Epoch: [18][ 810/ 1236] Overall Loss 0.294380 Objective Loss 0.294380 LR 0.001000 Time 0.021776 +2023-10-02 20:37:29,769 - Epoch: [18][ 820/ 1236] Overall Loss 0.294298 Objective Loss 0.294298 LR 0.001000 Time 0.021765 +2023-10-02 20:37:29,974 - Epoch: [18][ 830/ 1236] Overall Loss 0.294487 Objective Loss 0.294487 LR 0.001000 Time 0.021750 +2023-10-02 20:37:30,183 - Epoch: [18][ 840/ 1236] Overall Loss 0.294533 Objective Loss 0.294533 LR 0.001000 Time 0.021739 +2023-10-02 20:37:30,389 - Epoch: [18][ 850/ 1236] Overall Loss 0.294218 Objective Loss 0.294218 LR 0.001000 Time 0.021725 +2023-10-02 20:37:30,598 - Epoch: [18][ 860/ 1236] Overall Loss 0.293943 Objective Loss 0.293943 LR 0.001000 Time 0.021715 +2023-10-02 20:37:30,803 - Epoch: [18][ 870/ 1236] Overall Loss 0.294145 Objective Loss 0.294145 LR 0.001000 Time 0.021701 +2023-10-02 20:37:31,011 - Epoch: [18][ 880/ 1236] Overall Loss 0.294535 Objective Loss 0.294535 LR 0.001000 Time 0.021690 +2023-10-02 20:37:31,218 - Epoch: [18][ 890/ 1236] Overall Loss 0.294538 Objective Loss 0.294538 LR 0.001000 Time 0.021679 +2023-10-02 20:37:31,426 - Epoch: [18][ 900/ 1236] Overall Loss 0.294657 Objective Loss 0.294657 LR 0.001000 Time 0.021668 +2023-10-02 20:37:31,632 - Epoch: [18][ 910/ 1236] Overall Loss 0.294853 Objective Loss 0.294853 LR 0.001000 Time 0.021656 +2023-10-02 20:37:31,840 - Epoch: [18][ 920/ 1236] Overall Loss 0.294440 Objective Loss 0.294440 LR 0.001000 Time 0.021646 +2023-10-02 20:37:32,047 - Epoch: [18][ 930/ 1236] Overall Loss 0.294363 Objective Loss 0.294363 LR 0.001000 Time 0.021635 +2023-10-02 20:37:32,263 - Epoch: [18][ 940/ 1236] Overall Loss 0.294205 Objective Loss 0.294205 LR 0.001000 Time 0.021634 +2023-10-02 20:37:32,469 - Epoch: [18][ 950/ 1236] Overall Loss 0.294243 Objective Loss 0.294243 LR 0.001000 Time 0.021624 +2023-10-02 20:37:32,677 - Epoch: [18][ 960/ 1236] Overall Loss 0.294487 Objective Loss 0.294487 LR 0.001000 Time 0.021614 +2023-10-02 20:37:32,883 - Epoch: [18][ 970/ 1236] Overall Loss 0.294735 Objective Loss 0.294735 LR 0.001000 Time 0.021604 +2023-10-02 20:37:33,091 - Epoch: [18][ 980/ 1236] Overall Loss 0.294718 Objective Loss 0.294718 LR 0.001000 Time 0.021595 +2023-10-02 20:37:33,298 - Epoch: [18][ 990/ 1236] Overall Loss 0.294635 Objective Loss 0.294635 LR 0.001000 Time 0.021586 +2023-10-02 20:37:33,505 - Epoch: [18][ 1000/ 1236] Overall Loss 0.294905 Objective Loss 0.294905 LR 0.001000 Time 0.021577 +2023-10-02 20:37:33,712 - Epoch: [18][ 1010/ 1236] Overall Loss 0.295246 Objective Loss 0.295246 LR 0.001000 Time 0.021568 +2023-10-02 20:37:33,920 - Epoch: [18][ 1020/ 1236] Overall Loss 0.295161 Objective Loss 0.295161 LR 0.001000 Time 0.021560 +2023-10-02 20:37:34,127 - Epoch: [18][ 1030/ 1236] Overall Loss 0.295193 Objective Loss 0.295193 LR 0.001000 Time 0.021551 +2023-10-02 20:37:34,334 - Epoch: [18][ 1040/ 1236] Overall Loss 0.295485 Objective Loss 0.295485 LR 0.001000 Time 0.021543 +2023-10-02 20:37:34,541 - Epoch: [18][ 1050/ 1236] Overall Loss 0.295542 Objective Loss 0.295542 LR 0.001000 Time 0.021535 +2023-10-02 20:37:34,749 - Epoch: [18][ 1060/ 1236] Overall Loss 0.296042 Objective Loss 0.296042 LR 0.001000 Time 0.021527 +2023-10-02 20:37:34,956 - Epoch: [18][ 1070/ 1236] Overall Loss 0.296176 Objective Loss 0.296176 LR 0.001000 Time 0.021519 +2023-10-02 20:37:35,163 - Epoch: [18][ 1080/ 1236] Overall Loss 0.296045 Objective Loss 0.296045 LR 0.001000 Time 0.021512 +2023-10-02 20:37:35,371 - Epoch: [18][ 1090/ 1236] Overall Loss 0.295845 Objective Loss 0.295845 LR 0.001000 Time 0.021504 +2023-10-02 20:37:35,578 - Epoch: [18][ 1100/ 1236] Overall Loss 0.295968 Objective Loss 0.295968 LR 0.001000 Time 0.021497 +2023-10-02 20:37:35,785 - Epoch: [18][ 1110/ 1236] Overall Loss 0.295864 Objective Loss 0.295864 LR 0.001000 Time 0.021489 +2023-10-02 20:37:35,993 - Epoch: [18][ 1120/ 1236] Overall Loss 0.296097 Objective Loss 0.296097 LR 0.001000 Time 0.021483 +2023-10-02 20:37:36,200 - Epoch: [18][ 1130/ 1236] Overall Loss 0.295837 Objective Loss 0.295837 LR 0.001000 Time 0.021475 +2023-10-02 20:37:36,407 - Epoch: [18][ 1140/ 1236] Overall Loss 0.295693 Objective Loss 0.295693 LR 0.001000 Time 0.021469 +2023-10-02 20:37:36,614 - Epoch: [18][ 1150/ 1236] Overall Loss 0.295968 Objective Loss 0.295968 LR 0.001000 Time 0.021462 +2023-10-02 20:37:36,822 - Epoch: [18][ 1160/ 1236] Overall Loss 0.296257 Objective Loss 0.296257 LR 0.001000 Time 0.021455 +2023-10-02 20:37:37,028 - Epoch: [18][ 1170/ 1236] Overall Loss 0.296330 Objective Loss 0.296330 LR 0.001000 Time 0.021448 +2023-10-02 20:37:37,236 - Epoch: [18][ 1180/ 1236] Overall Loss 0.296446 Objective Loss 0.296446 LR 0.001000 Time 0.021442 +2023-10-02 20:37:37,443 - Epoch: [18][ 1190/ 1236] Overall Loss 0.296611 Objective Loss 0.296611 LR 0.001000 Time 0.021436 +2023-10-02 20:37:37,651 - Epoch: [18][ 1200/ 1236] Overall Loss 0.296809 Objective Loss 0.296809 LR 0.001000 Time 0.021430 +2023-10-02 20:37:37,858 - Epoch: [18][ 1210/ 1236] Overall Loss 0.297219 Objective Loss 0.297219 LR 0.001000 Time 0.021424 +2023-10-02 20:37:38,068 - Epoch: [18][ 1220/ 1236] Overall Loss 0.297219 Objective Loss 0.297219 LR 0.001000 Time 0.021420 +2023-10-02 20:37:38,332 - Epoch: [18][ 1230/ 1236] Overall Loss 0.297229 Objective Loss 0.297229 LR 0.001000 Time 0.021460 +2023-10-02 20:37:38,455 - Epoch: [18][ 1236/ 1236] Overall Loss 0.297373 Objective Loss 0.297373 Top1 80.448065 Top5 96.945010 LR 0.001000 Time 0.021455 +2023-10-02 20:37:38,607 - --- validate (epoch=18)----------- +2023-10-02 20:37:38,607 - 29943 samples (256 per mini-batch) +2023-10-02 20:37:39,105 - Epoch: [18][ 10/ 117] Loss 0.322898 Top1 81.953125 Top5 97.851562 +2023-10-02 20:37:39,252 - Epoch: [18][ 20/ 117] Loss 0.331321 Top1 81.250000 Top5 97.929688 +2023-10-02 20:37:39,396 - Epoch: [18][ 30/ 117] Loss 0.334251 Top1 80.768229 Top5 97.799479 +2023-10-02 20:37:39,541 - Epoch: [18][ 40/ 117] Loss 0.332839 Top1 80.703125 Top5 97.773438 +2023-10-02 20:37:39,686 - Epoch: [18][ 50/ 117] Loss 0.335742 Top1 80.664062 Top5 97.648438 +2023-10-02 20:37:39,831 - Epoch: [18][ 60/ 117] Loss 0.333016 Top1 80.722656 Top5 97.623698 +2023-10-02 20:37:39,976 - Epoch: [18][ 70/ 117] Loss 0.331849 Top1 80.786830 Top5 97.633929 +2023-10-02 20:37:40,122 - Epoch: [18][ 80/ 117] Loss 0.333428 Top1 80.737305 Top5 97.646484 +2023-10-02 20:37:40,269 - Epoch: [18][ 90/ 117] Loss 0.331946 Top1 80.785590 Top5 97.682292 +2023-10-02 20:37:40,414 - Epoch: [18][ 100/ 117] Loss 0.329020 Top1 80.921875 Top5 97.667969 +2023-10-02 20:37:40,567 - Epoch: [18][ 110/ 117] Loss 0.331345 Top1 80.873580 Top5 97.645597 +2023-10-02 20:37:40,657 - Epoch: [18][ 117/ 117] Loss 0.330701 Top1 80.930434 Top5 97.642187 +2023-10-02 20:37:40,795 - ==> Top1: 80.930 Top5: 97.642 Loss: 0.331 + +2023-10-02 20:37:40,796 - ==> Confusion: +[[ 911 0 7 1 10 6 0 0 3 74 1 1 1 4 5 5 4 2 0 0 15] + [ 0 1035 2 0 10 37 0 22 2 1 1 0 0 0 2 5 1 0 9 1 3] + [ 6 1 953 20 4 3 31 6 0 1 1 3 6 0 0 2 1 0 5 2 11] + [ 1 1 16 972 0 7 2 2 4 0 1 0 6 4 33 3 1 4 19 0 13] + [ 24 5 1 0 980 2 0 0 0 6 2 0 2 2 7 5 8 0 0 2 4] + [ 0 31 1 2 3 999 2 24 2 4 5 1 1 14 7 1 5 1 8 0 5] + [ 0 0 33 0 0 0 1122 4 0 0 3 3 0 0 1 10 0 1 1 9 4] + [ 3 23 25 0 5 34 3 1052 0 1 3 4 2 0 2 1 0 0 42 13 5] + [ 22 7 2 0 2 5 0 2 946 49 10 0 2 6 23 2 4 2 5 0 0] + [ 124 0 0 0 13 4 0 0 19 930 1 0 0 7 8 1 1 0 0 5 6] + [ 3 2 6 18 3 2 1 2 15 1 952 6 1 9 9 1 2 0 12 0 8] + [ 0 2 2 0 4 26 0 4 0 2 0 937 30 3 0 2 1 14 0 2 6] + [ 0 5 2 0 0 5 1 1 1 0 3 47 958 1 5 8 1 15 4 6 5] + [ 1 0 3 0 9 22 0 0 13 28 7 3 2 1004 11 1 3 1 1 5 5] + [ 6 1 3 9 5 1 0 0 18 4 0 1 3 0 1035 0 0 3 7 0 5] + [ 0 1 2 1 6 0 0 0 1 1 0 11 6 2 0 1071 12 10 1 4 5] + [ 0 9 1 0 3 10 0 0 1 0 0 6 0 0 4 13 1098 0 2 4 10] + [ 0 0 1 1 0 0 3 0 0 0 0 2 22 0 2 5 1 996 4 0 1] + [ 4 7 6 16 1 1 0 16 6 0 3 0 0 0 11 2 3 0 985 0 7] + [ 0 0 2 0 2 7 11 9 0 0 2 10 2 4 0 8 5 2 2 1080 6] + [ 141 248 167 97 158 278 54 115 121 126 121 199 389 295 263 75 265 72 244 260 4217]] + +2023-10-02 20:37:40,797 - ==> Best [Top1: 82.590 Top5: 98.016 Sparsity:0.00 Params: 169472 on epoch: 14] +2023-10-02 20:37:40,797 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:37:40,804 - + +2023-10-02 20:37:40,804 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:37:41,841 - Epoch: [19][ 10/ 1236] Overall Loss 0.304617 Objective Loss 0.304617 LR 0.001000 Time 0.103697 +2023-10-02 20:37:42,052 - Epoch: [19][ 20/ 1236] Overall Loss 0.295770 Objective Loss 0.295770 LR 0.001000 Time 0.062382 +2023-10-02 20:37:42,261 - Epoch: [19][ 30/ 1236] Overall Loss 0.295603 Objective Loss 0.295603 LR 0.001000 Time 0.048500 +2023-10-02 20:37:42,472 - Epoch: [19][ 40/ 1236] Overall Loss 0.298046 Objective Loss 0.298046 LR 0.001000 Time 0.041625 +2023-10-02 20:37:42,679 - Epoch: [19][ 50/ 1236] Overall Loss 0.292960 Objective Loss 0.292960 LR 0.001000 Time 0.037428 +2023-10-02 20:37:42,890 - Epoch: [19][ 60/ 1236] Overall Loss 0.297368 Objective Loss 0.297368 LR 0.001000 Time 0.034691 +2023-10-02 20:37:43,098 - Epoch: [19][ 70/ 1236] Overall Loss 0.294083 Objective Loss 0.294083 LR 0.001000 Time 0.032686 +2023-10-02 20:37:43,309 - Epoch: [19][ 80/ 1236] Overall Loss 0.292908 Objective Loss 0.292908 LR 0.001000 Time 0.031239 +2023-10-02 20:37:43,516 - Epoch: [19][ 90/ 1236] Overall Loss 0.293163 Objective Loss 0.293163 LR 0.001000 Time 0.030066 +2023-10-02 20:37:43,727 - Epoch: [19][ 100/ 1236] Overall Loss 0.292432 Objective Loss 0.292432 LR 0.001000 Time 0.029159 +2023-10-02 20:37:43,935 - Epoch: [19][ 110/ 1236] Overall Loss 0.293385 Objective Loss 0.293385 LR 0.001000 Time 0.028387 +2023-10-02 20:37:44,145 - Epoch: [19][ 120/ 1236] Overall Loss 0.291842 Objective Loss 0.291842 LR 0.001000 Time 0.027771 +2023-10-02 20:37:44,353 - Epoch: [19][ 130/ 1236] Overall Loss 0.291538 Objective Loss 0.291538 LR 0.001000 Time 0.027223 +2023-10-02 20:37:44,563 - Epoch: [19][ 140/ 1236] Overall Loss 0.290844 Objective Loss 0.290844 LR 0.001000 Time 0.026778 +2023-10-02 20:37:44,772 - Epoch: [19][ 150/ 1236] Overall Loss 0.290794 Objective Loss 0.290794 LR 0.001000 Time 0.026373 +2023-10-02 20:37:44,982 - Epoch: [19][ 160/ 1236] Overall Loss 0.289848 Objective Loss 0.289848 LR 0.001000 Time 0.026037 +2023-10-02 20:37:45,190 - Epoch: [19][ 170/ 1236] Overall Loss 0.290245 Objective Loss 0.290245 LR 0.001000 Time 0.025721 +2023-10-02 20:37:45,401 - Epoch: [19][ 180/ 1236] Overall Loss 0.288603 Objective Loss 0.288603 LR 0.001000 Time 0.025460 +2023-10-02 20:37:45,609 - Epoch: [19][ 190/ 1236] Overall Loss 0.289234 Objective Loss 0.289234 LR 0.001000 Time 0.025208 +2023-10-02 20:37:45,820 - Epoch: [19][ 200/ 1236] Overall Loss 0.288713 Objective Loss 0.288713 LR 0.001000 Time 0.025003 +2023-10-02 20:37:46,027 - Epoch: [19][ 210/ 1236] Overall Loss 0.290082 Objective Loss 0.290082 LR 0.001000 Time 0.024797 +2023-10-02 20:37:46,238 - Epoch: [19][ 220/ 1236] Overall Loss 0.289906 Objective Loss 0.289906 LR 0.001000 Time 0.024625 +2023-10-02 20:37:46,446 - Epoch: [19][ 230/ 1236] Overall Loss 0.290707 Objective Loss 0.290707 LR 0.001000 Time 0.024453 +2023-10-02 20:37:46,657 - Epoch: [19][ 240/ 1236] Overall Loss 0.290078 Objective Loss 0.290078 LR 0.001000 Time 0.024314 +2023-10-02 20:37:46,864 - Epoch: [19][ 250/ 1236] Overall Loss 0.289872 Objective Loss 0.289872 LR 0.001000 Time 0.024169 +2023-10-02 20:37:47,073 - Epoch: [19][ 260/ 1236] Overall Loss 0.289075 Objective Loss 0.289075 LR 0.001000 Time 0.024041 +2023-10-02 20:37:47,281 - Epoch: [19][ 270/ 1236] Overall Loss 0.288764 Objective Loss 0.288764 LR 0.001000 Time 0.023915 +2023-10-02 20:37:47,492 - Epoch: [19][ 280/ 1236] Overall Loss 0.288617 Objective Loss 0.288617 LR 0.001000 Time 0.023812 +2023-10-02 20:37:47,700 - Epoch: [19][ 290/ 1236] Overall Loss 0.288692 Objective Loss 0.288692 LR 0.001000 Time 0.023703 +2023-10-02 20:37:47,910 - Epoch: [19][ 300/ 1236] Overall Loss 0.288892 Objective Loss 0.288892 LR 0.001000 Time 0.023614 +2023-10-02 20:37:48,119 - Epoch: [19][ 310/ 1236] Overall Loss 0.288193 Objective Loss 0.288193 LR 0.001000 Time 0.023519 +2023-10-02 20:37:48,329 - Epoch: [19][ 320/ 1236] Overall Loss 0.287919 Objective Loss 0.287919 LR 0.001000 Time 0.023441 +2023-10-02 20:37:48,537 - Epoch: [19][ 330/ 1236] Overall Loss 0.288479 Objective Loss 0.288479 LR 0.001000 Time 0.023357 +2023-10-02 20:37:48,748 - Epoch: [19][ 340/ 1236] Overall Loss 0.289525 Objective Loss 0.289525 LR 0.001000 Time 0.023288 +2023-10-02 20:37:48,956 - Epoch: [19][ 350/ 1236] Overall Loss 0.289135 Objective Loss 0.289135 LR 0.001000 Time 0.023214 +2023-10-02 20:37:49,166 - Epoch: [19][ 360/ 1236] Overall Loss 0.289737 Objective Loss 0.289737 LR 0.001000 Time 0.023152 +2023-10-02 20:37:49,375 - Epoch: [19][ 370/ 1236] Overall Loss 0.289893 Objective Loss 0.289893 LR 0.001000 Time 0.023087 +2023-10-02 20:37:49,585 - Epoch: [19][ 380/ 1236] Overall Loss 0.290070 Objective Loss 0.290070 LR 0.001000 Time 0.023032 +2023-10-02 20:37:49,794 - Epoch: [19][ 390/ 1236] Overall Loss 0.290142 Objective Loss 0.290142 LR 0.001000 Time 0.022972 +2023-10-02 20:37:50,005 - Epoch: [19][ 400/ 1236] Overall Loss 0.290051 Objective Loss 0.290051 LR 0.001000 Time 0.022923 +2023-10-02 20:37:50,213 - Epoch: [19][ 410/ 1236] Overall Loss 0.289969 Objective Loss 0.289969 LR 0.001000 Time 0.022870 +2023-10-02 20:37:50,423 - Epoch: [19][ 420/ 1236] Overall Loss 0.289889 Objective Loss 0.289889 LR 0.001000 Time 0.022825 +2023-10-02 20:37:50,632 - Epoch: [19][ 430/ 1236] Overall Loss 0.289751 Objective Loss 0.289751 LR 0.001000 Time 0.022775 +2023-10-02 20:37:50,843 - Epoch: [19][ 440/ 1236] Overall Loss 0.289191 Objective Loss 0.289191 LR 0.001000 Time 0.022736 +2023-10-02 20:37:51,051 - Epoch: [19][ 450/ 1236] Overall Loss 0.289856 Objective Loss 0.289856 LR 0.001000 Time 0.022691 +2023-10-02 20:37:51,262 - Epoch: [19][ 460/ 1236] Overall Loss 0.290261 Objective Loss 0.290261 LR 0.001000 Time 0.022655 +2023-10-02 20:37:51,471 - Epoch: [19][ 470/ 1236] Overall Loss 0.289834 Objective Loss 0.289834 LR 0.001000 Time 0.022613 +2023-10-02 20:37:51,681 - Epoch: [19][ 480/ 1236] Overall Loss 0.290109 Objective Loss 0.290109 LR 0.001000 Time 0.022580 +2023-10-02 20:37:51,890 - Epoch: [19][ 490/ 1236] Overall Loss 0.289580 Objective Loss 0.289580 LR 0.001000 Time 0.022542 +2023-10-02 20:37:52,100 - Epoch: [19][ 500/ 1236] Overall Loss 0.290318 Objective Loss 0.290318 LR 0.001000 Time 0.022511 +2023-10-02 20:37:52,309 - Epoch: [19][ 510/ 1236] Overall Loss 0.290511 Objective Loss 0.290511 LR 0.001000 Time 0.022476 +2023-10-02 20:37:52,520 - Epoch: [19][ 520/ 1236] Overall Loss 0.291536 Objective Loss 0.291536 LR 0.001000 Time 0.022449 +2023-10-02 20:37:52,728 - Epoch: [19][ 530/ 1236] Overall Loss 0.291839 Objective Loss 0.291839 LR 0.001000 Time 0.022416 +2023-10-02 20:37:52,938 - Epoch: [19][ 540/ 1236] Overall Loss 0.291919 Objective Loss 0.291919 LR 0.001000 Time 0.022389 +2023-10-02 20:37:53,147 - Epoch: [19][ 550/ 1236] Overall Loss 0.292408 Objective Loss 0.292408 LR 0.001000 Time 0.022359 +2023-10-02 20:37:53,358 - Epoch: [19][ 560/ 1236] Overall Loss 0.292439 Objective Loss 0.292439 LR 0.001000 Time 0.022335 +2023-10-02 20:37:53,566 - Epoch: [19][ 570/ 1236] Overall Loss 0.292129 Objective Loss 0.292129 LR 0.001000 Time 0.022307 +2023-10-02 20:37:53,777 - Epoch: [19][ 580/ 1236] Overall Loss 0.292307 Objective Loss 0.292307 LR 0.001000 Time 0.022284 +2023-10-02 20:37:53,985 - Epoch: [19][ 590/ 1236] Overall Loss 0.292059 Objective Loss 0.292059 LR 0.001000 Time 0.022257 +2023-10-02 20:37:54,196 - Epoch: [19][ 600/ 1236] Overall Loss 0.291907 Objective Loss 0.291907 LR 0.001000 Time 0.022237 +2023-10-02 20:37:54,404 - Epoch: [19][ 610/ 1236] Overall Loss 0.291887 Objective Loss 0.291887 LR 0.001000 Time 0.022211 +2023-10-02 20:37:54,615 - Epoch: [19][ 620/ 1236] Overall Loss 0.291880 Objective Loss 0.291880 LR 0.001000 Time 0.022192 +2023-10-02 20:37:54,823 - Epoch: [19][ 630/ 1236] Overall Loss 0.291770 Objective Loss 0.291770 LR 0.001000 Time 0.022168 +2023-10-02 20:37:55,034 - Epoch: [19][ 640/ 1236] Overall Loss 0.291951 Objective Loss 0.291951 LR 0.001000 Time 0.022150 +2023-10-02 20:37:55,242 - Epoch: [19][ 650/ 1236] Overall Loss 0.292070 Objective Loss 0.292070 LR 0.001000 Time 0.022128 +2023-10-02 20:37:55,453 - Epoch: [19][ 660/ 1236] Overall Loss 0.292518 Objective Loss 0.292518 LR 0.001000 Time 0.022112 +2023-10-02 20:37:55,662 - Epoch: [19][ 670/ 1236] Overall Loss 0.292665 Objective Loss 0.292665 LR 0.001000 Time 0.022092 +2023-10-02 20:37:55,872 - Epoch: [19][ 680/ 1236] Overall Loss 0.292720 Objective Loss 0.292720 LR 0.001000 Time 0.022077 +2023-10-02 20:37:56,081 - Epoch: [19][ 690/ 1236] Overall Loss 0.293027 Objective Loss 0.293027 LR 0.001000 Time 0.022059 +2023-10-02 20:37:56,292 - Epoch: [19][ 700/ 1236] Overall Loss 0.293046 Objective Loss 0.293046 LR 0.001000 Time 0.022044 +2023-10-02 20:37:56,500 - Epoch: [19][ 710/ 1236] Overall Loss 0.293185 Objective Loss 0.293185 LR 0.001000 Time 0.022026 +2023-10-02 20:37:56,711 - Epoch: [19][ 720/ 1236] Overall Loss 0.293617 Objective Loss 0.293617 LR 0.001000 Time 0.022011 +2023-10-02 20:37:56,920 - Epoch: [19][ 730/ 1236] Overall Loss 0.293477 Objective Loss 0.293477 LR 0.001000 Time 0.021994 +2023-10-02 20:37:57,130 - Epoch: [19][ 740/ 1236] Overall Loss 0.293698 Objective Loss 0.293698 LR 0.001000 Time 0.021980 +2023-10-02 20:37:57,339 - Epoch: [19][ 750/ 1236] Overall Loss 0.293868 Objective Loss 0.293868 LR 0.001000 Time 0.021963 +2023-10-02 20:37:57,549 - Epoch: [19][ 760/ 1236] Overall Loss 0.293759 Objective Loss 0.293759 LR 0.001000 Time 0.021951 +2023-10-02 20:37:57,758 - Epoch: [19][ 770/ 1236] Overall Loss 0.293493 Objective Loss 0.293493 LR 0.001000 Time 0.021935 +2023-10-02 20:37:57,969 - Epoch: [19][ 780/ 1236] Overall Loss 0.294084 Objective Loss 0.294084 LR 0.001000 Time 0.021923 +2023-10-02 20:37:58,177 - Epoch: [19][ 790/ 1236] Overall Loss 0.294144 Objective Loss 0.294144 LR 0.001000 Time 0.021908 +2023-10-02 20:37:58,388 - Epoch: [19][ 800/ 1236] Overall Loss 0.294357 Objective Loss 0.294357 LR 0.001000 Time 0.021897 +2023-10-02 20:37:58,597 - Epoch: [19][ 810/ 1236] Overall Loss 0.294589 Objective Loss 0.294589 LR 0.001000 Time 0.021883 +2023-10-02 20:37:58,807 - Epoch: [19][ 820/ 1236] Overall Loss 0.294472 Objective Loss 0.294472 LR 0.001000 Time 0.021872 +2023-10-02 20:37:59,016 - Epoch: [19][ 830/ 1236] Overall Loss 0.294664 Objective Loss 0.294664 LR 0.001000 Time 0.021860 +2023-10-02 20:37:59,226 - Epoch: [19][ 840/ 1236] Overall Loss 0.294746 Objective Loss 0.294746 LR 0.001000 Time 0.021850 +2023-10-02 20:37:59,435 - Epoch: [19][ 850/ 1236] Overall Loss 0.294486 Objective Loss 0.294486 LR 0.001000 Time 0.021836 +2023-10-02 20:37:59,646 - Epoch: [19][ 860/ 1236] Overall Loss 0.294642 Objective Loss 0.294642 LR 0.001000 Time 0.021827 +2023-10-02 20:37:59,854 - Epoch: [19][ 870/ 1236] Overall Loss 0.294565 Objective Loss 0.294565 LR 0.001000 Time 0.021813 +2023-10-02 20:38:00,064 - Epoch: [19][ 880/ 1236] Overall Loss 0.294550 Objective Loss 0.294550 LR 0.001000 Time 0.021805 +2023-10-02 20:38:00,273 - Epoch: [19][ 890/ 1236] Overall Loss 0.294067 Objective Loss 0.294067 LR 0.001000 Time 0.021794 +2023-10-02 20:38:00,484 - Epoch: [19][ 900/ 1236] Overall Loss 0.293864 Objective Loss 0.293864 LR 0.001000 Time 0.021785 +2023-10-02 20:38:00,692 - Epoch: [19][ 910/ 1236] Overall Loss 0.293671 Objective Loss 0.293671 LR 0.001000 Time 0.021773 +2023-10-02 20:38:00,903 - Epoch: [19][ 920/ 1236] Overall Loss 0.293889 Objective Loss 0.293889 LR 0.001000 Time 0.021765 +2023-10-02 20:38:01,112 - Epoch: [19][ 930/ 1236] Overall Loss 0.293947 Objective Loss 0.293947 LR 0.001000 Time 0.021754 +2023-10-02 20:38:01,322 - Epoch: [19][ 940/ 1236] Overall Loss 0.293773 Objective Loss 0.293773 LR 0.001000 Time 0.021746 +2023-10-02 20:38:01,532 - Epoch: [19][ 950/ 1236] Overall Loss 0.293753 Objective Loss 0.293753 LR 0.001000 Time 0.021737 +2023-10-02 20:38:01,742 - Epoch: [19][ 960/ 1236] Overall Loss 0.293627 Objective Loss 0.293627 LR 0.001000 Time 0.021730 +2023-10-02 20:38:01,951 - Epoch: [19][ 970/ 1236] Overall Loss 0.293823 Objective Loss 0.293823 LR 0.001000 Time 0.021721 +2023-10-02 20:38:02,162 - Epoch: [19][ 980/ 1236] Overall Loss 0.293621 Objective Loss 0.293621 LR 0.001000 Time 0.021714 +2023-10-02 20:38:02,371 - Epoch: [19][ 990/ 1236] Overall Loss 0.293520 Objective Loss 0.293520 LR 0.001000 Time 0.021704 +2023-10-02 20:38:02,581 - Epoch: [19][ 1000/ 1236] Overall Loss 0.293570 Objective Loss 0.293570 LR 0.001000 Time 0.021697 +2023-10-02 20:38:02,790 - Epoch: [19][ 1010/ 1236] Overall Loss 0.293797 Objective Loss 0.293797 LR 0.001000 Time 0.021688 +2023-10-02 20:38:03,001 - Epoch: [19][ 1020/ 1236] Overall Loss 0.293667 Objective Loss 0.293667 LR 0.001000 Time 0.021681 +2023-10-02 20:38:03,210 - Epoch: [19][ 1030/ 1236] Overall Loss 0.293842 Objective Loss 0.293842 LR 0.001000 Time 0.021672 +2023-10-02 20:38:03,420 - Epoch: [19][ 1040/ 1236] Overall Loss 0.293418 Objective Loss 0.293418 LR 0.001000 Time 0.021666 +2023-10-02 20:38:03,629 - Epoch: [19][ 1050/ 1236] Overall Loss 0.293122 Objective Loss 0.293122 LR 0.001000 Time 0.021658 +2023-10-02 20:38:03,840 - Epoch: [19][ 1060/ 1236] Overall Loss 0.293222 Objective Loss 0.293222 LR 0.001000 Time 0.021652 +2023-10-02 20:38:04,049 - Epoch: [19][ 1070/ 1236] Overall Loss 0.293357 Objective Loss 0.293357 LR 0.001000 Time 0.021645 +2023-10-02 20:38:04,259 - Epoch: [19][ 1080/ 1236] Overall Loss 0.293586 Objective Loss 0.293586 LR 0.001000 Time 0.021639 +2023-10-02 20:38:04,468 - Epoch: [19][ 1090/ 1236] Overall Loss 0.293757 Objective Loss 0.293757 LR 0.001000 Time 0.021631 +2023-10-02 20:38:04,679 - Epoch: [19][ 1100/ 1236] Overall Loss 0.293623 Objective Loss 0.293623 LR 0.001000 Time 0.021625 +2023-10-02 20:38:04,887 - Epoch: [19][ 1110/ 1236] Overall Loss 0.294049 Objective Loss 0.294049 LR 0.001000 Time 0.021617 +2023-10-02 20:38:05,099 - Epoch: [19][ 1120/ 1236] Overall Loss 0.294398 Objective Loss 0.294398 LR 0.001000 Time 0.021612 +2023-10-02 20:38:05,308 - Epoch: [19][ 1130/ 1236] Overall Loss 0.294543 Objective Loss 0.294543 LR 0.001000 Time 0.021605 +2023-10-02 20:38:05,519 - Epoch: [19][ 1140/ 1236] Overall Loss 0.294534 Objective Loss 0.294534 LR 0.001000 Time 0.021600 +2023-10-02 20:38:05,727 - Epoch: [19][ 1150/ 1236] Overall Loss 0.294730 Objective Loss 0.294730 LR 0.001000 Time 0.021593 +2023-10-02 20:38:05,938 - Epoch: [19][ 1160/ 1236] Overall Loss 0.294933 Objective Loss 0.294933 LR 0.001000 Time 0.021588 +2023-10-02 20:38:06,147 - Epoch: [19][ 1170/ 1236] Overall Loss 0.294915 Objective Loss 0.294915 LR 0.001000 Time 0.021581 +2023-10-02 20:38:06,357 - Epoch: [19][ 1180/ 1236] Overall Loss 0.294652 Objective Loss 0.294652 LR 0.001000 Time 0.021576 +2023-10-02 20:38:06,566 - Epoch: [19][ 1190/ 1236] Overall Loss 0.294837 Objective Loss 0.294837 LR 0.001000 Time 0.021570 +2023-10-02 20:38:06,777 - Epoch: [19][ 1200/ 1236] Overall Loss 0.294912 Objective Loss 0.294912 LR 0.001000 Time 0.021565 +2023-10-02 20:38:06,985 - Epoch: [19][ 1210/ 1236] Overall Loss 0.294734 Objective Loss 0.294734 LR 0.001000 Time 0.021558 +2023-10-02 20:38:07,196 - Epoch: [19][ 1220/ 1236] Overall Loss 0.295121 Objective Loss 0.295121 LR 0.001000 Time 0.021554 +2023-10-02 20:38:07,459 - Epoch: [19][ 1230/ 1236] Overall Loss 0.294976 Objective Loss 0.294976 LR 0.001000 Time 0.021592 +2023-10-02 20:38:07,582 - Epoch: [19][ 1236/ 1236] Overall Loss 0.295018 Objective Loss 0.295018 Top1 83.095723 Top5 97.963340 LR 0.001000 Time 0.021587 +2023-10-02 20:38:07,707 - --- validate (epoch=19)----------- +2023-10-02 20:38:07,708 - 29943 samples (256 per mini-batch) +2023-10-02 20:38:08,189 - Epoch: [19][ 10/ 117] Loss 0.361192 Top1 81.601562 Top5 97.734375 +2023-10-02 20:38:08,343 - Epoch: [19][ 20/ 117] Loss 0.330714 Top1 82.402344 Top5 98.085938 +2023-10-02 20:38:08,496 - Epoch: [19][ 30/ 117] Loss 0.347283 Top1 82.239583 Top5 98.046875 +2023-10-02 20:38:08,649 - Epoch: [19][ 40/ 117] Loss 0.348395 Top1 82.451172 Top5 98.056641 +2023-10-02 20:38:08,802 - Epoch: [19][ 50/ 117] Loss 0.347886 Top1 82.507812 Top5 98.031250 +2023-10-02 20:38:08,954 - Epoch: [19][ 60/ 117] Loss 0.347044 Top1 82.610677 Top5 98.033854 +2023-10-02 20:38:09,107 - Epoch: [19][ 70/ 117] Loss 0.344722 Top1 82.823661 Top5 98.041295 +2023-10-02 20:38:09,260 - Epoch: [19][ 80/ 117] Loss 0.346875 Top1 82.788086 Top5 98.041992 +2023-10-02 20:38:09,410 - Epoch: [19][ 90/ 117] Loss 0.344762 Top1 82.916667 Top5 98.059896 +2023-10-02 20:38:09,561 - Epoch: [19][ 100/ 117] Loss 0.341758 Top1 82.988281 Top5 98.074219 +2023-10-02 20:38:09,718 - Epoch: [19][ 110/ 117] Loss 0.342466 Top1 82.950994 Top5 98.032670 +2023-10-02 20:38:09,808 - Epoch: [19][ 117/ 117] Loss 0.344352 Top1 82.947600 Top5 97.986174 +2023-10-02 20:38:09,904 - ==> Top1: 82.948 Top5: 97.986 Loss: 0.344 + +2023-10-02 20:38:09,905 - ==> Confusion: +[[ 960 0 3 1 6 1 0 0 6 46 1 2 1 1 2 3 3 2 0 0 12] + [ 0 1041 2 0 11 24 3 11 5 1 7 0 0 0 2 7 1 0 7 3 6] + [ 9 1 947 17 3 0 33 3 1 0 0 0 9 1 0 3 0 1 9 6 13] + [ 3 1 15 947 2 2 3 1 14 1 2 0 13 3 41 1 3 3 12 1 21] + [ 29 2 2 1 969 6 0 0 2 2 1 0 0 3 8 5 9 3 0 3 5] + [ 0 44 3 0 5 980 2 20 4 7 1 6 7 10 3 1 5 3 2 3 10] + [ 0 2 28 0 0 0 1130 2 0 0 1 1 1 1 0 8 1 1 1 9 5] + [ 5 23 26 1 7 39 9 1031 1 1 1 4 4 0 2 1 0 1 41 13 8] + [ 24 3 1 0 1 0 0 0 991 35 4 0 2 6 9 1 2 3 4 2 1] + [ 176 0 0 0 6 0 1 0 31 866 1 0 3 15 6 2 0 0 0 2 10] + [ 3 3 7 17 2 0 4 4 26 0 950 1 2 6 7 1 0 3 7 0 10] + [ 1 0 5 0 2 9 1 0 0 4 0 950 22 6 0 5 1 15 0 6 8] + [ 0 3 4 2 0 1 2 0 1 0 0 37 975 0 1 14 1 13 0 2 12] + [ 1 0 5 0 1 10 0 3 30 12 10 5 3 1007 10 0 1 4 0 2 15] + [ 16 0 3 15 5 0 0 0 35 4 0 0 1 1 1001 0 0 2 5 0 13] + [ 0 1 2 0 3 0 1 0 0 0 0 10 9 0 0 1078 9 9 0 4 8] + [ 2 20 0 0 2 8 1 0 2 0 0 7 1 0 2 16 1077 1 0 6 16] + [ 0 0 0 2 0 0 3 0 0 0 0 4 19 0 0 7 2 996 2 2 1] + [ 3 8 10 17 1 0 3 15 8 0 6 0 3 0 14 0 1 0 966 3 10] + [ 0 2 4 0 1 4 8 4 0 0 3 18 8 0 1 5 0 2 4 1076 12] + [ 212 162 151 63 142 175 58 71 143 84 143 192 352 256 144 107 129 86 132 204 4899]] + +2023-10-02 20:38:09,906 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:38:09,906 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:38:09,920 - + +2023-10-02 20:38:09,920 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:38:10,946 - Epoch: [20][ 10/ 1236] Overall Loss 0.292965 Objective Loss 0.292965 LR 0.001000 Time 0.102559 +2023-10-02 20:38:11,166 - Epoch: [20][ 20/ 1236] Overall Loss 0.274602 Objective Loss 0.274602 LR 0.001000 Time 0.062228 +2023-10-02 20:38:11,379 - Epoch: [20][ 30/ 1236] Overall Loss 0.276347 Objective Loss 0.276347 LR 0.001000 Time 0.048591 +2023-10-02 20:38:11,598 - Epoch: [20][ 40/ 1236] Overall Loss 0.274370 Objective Loss 0.274370 LR 0.001000 Time 0.041910 +2023-10-02 20:38:11,809 - Epoch: [20][ 50/ 1236] Overall Loss 0.272217 Objective Loss 0.272217 LR 0.001000 Time 0.037734 +2023-10-02 20:38:12,018 - Epoch: [20][ 60/ 1236] Overall Loss 0.276719 Objective Loss 0.276719 LR 0.001000 Time 0.034930 +2023-10-02 20:38:12,224 - Epoch: [20][ 70/ 1236] Overall Loss 0.275669 Objective Loss 0.275669 LR 0.001000 Time 0.032874 +2023-10-02 20:38:12,433 - Epoch: [20][ 80/ 1236] Overall Loss 0.274861 Objective Loss 0.274861 LR 0.001000 Time 0.031379 +2023-10-02 20:38:12,639 - Epoch: [20][ 90/ 1236] Overall Loss 0.276442 Objective Loss 0.276442 LR 0.001000 Time 0.030174 +2023-10-02 20:38:12,847 - Epoch: [20][ 100/ 1236] Overall Loss 0.279650 Objective Loss 0.279650 LR 0.001000 Time 0.029235 +2023-10-02 20:38:13,054 - Epoch: [20][ 110/ 1236] Overall Loss 0.278422 Objective Loss 0.278422 LR 0.001000 Time 0.028458 +2023-10-02 20:38:13,263 - Epoch: [20][ 120/ 1236] Overall Loss 0.279323 Objective Loss 0.279323 LR 0.001000 Time 0.027822 +2023-10-02 20:38:13,470 - Epoch: [20][ 130/ 1236] Overall Loss 0.276787 Objective Loss 0.276787 LR 0.001000 Time 0.027260 +2023-10-02 20:38:13,678 - Epoch: [20][ 140/ 1236] Overall Loss 0.277040 Objective Loss 0.277040 LR 0.001000 Time 0.026798 +2023-10-02 20:38:13,884 - Epoch: [20][ 150/ 1236] Overall Loss 0.276867 Objective Loss 0.276867 LR 0.001000 Time 0.026382 +2023-10-02 20:38:14,093 - Epoch: [20][ 160/ 1236] Overall Loss 0.277311 Objective Loss 0.277311 LR 0.001000 Time 0.026041 +2023-10-02 20:38:14,299 - Epoch: [20][ 170/ 1236] Overall Loss 0.278941 Objective Loss 0.278941 LR 0.001000 Time 0.025715 +2023-10-02 20:38:14,508 - Epoch: [20][ 180/ 1236] Overall Loss 0.278781 Objective Loss 0.278781 LR 0.001000 Time 0.025450 +2023-10-02 20:38:14,714 - Epoch: [20][ 190/ 1236] Overall Loss 0.278977 Objective Loss 0.278977 LR 0.001000 Time 0.025189 +2023-10-02 20:38:14,923 - Epoch: [20][ 200/ 1236] Overall Loss 0.279165 Objective Loss 0.279165 LR 0.001000 Time 0.024976 +2023-10-02 20:38:15,129 - Epoch: [20][ 210/ 1236] Overall Loss 0.279588 Objective Loss 0.279588 LR 0.001000 Time 0.024764 +2023-10-02 20:38:15,338 - Epoch: [20][ 220/ 1236] Overall Loss 0.280025 Objective Loss 0.280025 LR 0.001000 Time 0.024590 +2023-10-02 20:38:15,544 - Epoch: [20][ 230/ 1236] Overall Loss 0.281463 Objective Loss 0.281463 LR 0.001000 Time 0.024413 +2023-10-02 20:38:15,752 - Epoch: [20][ 240/ 1236] Overall Loss 0.281172 Objective Loss 0.281172 LR 0.001000 Time 0.024263 +2023-10-02 20:38:15,959 - Epoch: [20][ 250/ 1236] Overall Loss 0.281722 Objective Loss 0.281722 LR 0.001000 Time 0.024113 +2023-10-02 20:38:16,169 - Epoch: [20][ 260/ 1236] Overall Loss 0.283569 Objective Loss 0.283569 LR 0.001000 Time 0.023993 +2023-10-02 20:38:16,375 - Epoch: [20][ 270/ 1236] Overall Loss 0.283710 Objective Loss 0.283710 LR 0.001000 Time 0.023865 +2023-10-02 20:38:16,583 - Epoch: [20][ 280/ 1236] Overall Loss 0.285051 Objective Loss 0.285051 LR 0.001000 Time 0.023756 +2023-10-02 20:38:16,790 - Epoch: [20][ 290/ 1236] Overall Loss 0.285244 Objective Loss 0.285244 LR 0.001000 Time 0.023646 +2023-10-02 20:38:17,000 - Epoch: [20][ 300/ 1236] Overall Loss 0.284495 Objective Loss 0.284495 LR 0.001000 Time 0.023556 +2023-10-02 20:38:17,206 - Epoch: [20][ 310/ 1236] Overall Loss 0.285996 Objective Loss 0.285996 LR 0.001000 Time 0.023459 +2023-10-02 20:38:17,414 - Epoch: [20][ 320/ 1236] Overall Loss 0.286577 Objective Loss 0.286577 LR 0.001000 Time 0.023376 +2023-10-02 20:38:17,622 - Epoch: [20][ 330/ 1236] Overall Loss 0.286638 Objective Loss 0.286638 LR 0.001000 Time 0.023292 +2023-10-02 20:38:17,831 - Epoch: [20][ 340/ 1236] Overall Loss 0.285710 Objective Loss 0.285710 LR 0.001000 Time 0.023223 +2023-10-02 20:38:18,037 - Epoch: [20][ 350/ 1236] Overall Loss 0.285936 Objective Loss 0.285936 LR 0.001000 Time 0.023146 +2023-10-02 20:38:18,247 - Epoch: [20][ 360/ 1236] Overall Loss 0.286138 Objective Loss 0.286138 LR 0.001000 Time 0.023085 +2023-10-02 20:38:18,453 - Epoch: [20][ 370/ 1236] Overall Loss 0.286659 Objective Loss 0.286659 LR 0.001000 Time 0.023017 +2023-10-02 20:38:18,661 - Epoch: [20][ 380/ 1236] Overall Loss 0.287243 Objective Loss 0.287243 LR 0.001000 Time 0.022959 +2023-10-02 20:38:18,869 - Epoch: [20][ 390/ 1236] Overall Loss 0.286833 Objective Loss 0.286833 LR 0.001000 Time 0.022898 +2023-10-02 20:38:19,077 - Epoch: [20][ 400/ 1236] Overall Loss 0.286041 Objective Loss 0.286041 LR 0.001000 Time 0.022846 +2023-10-02 20:38:19,284 - Epoch: [20][ 410/ 1236] Overall Loss 0.286166 Objective Loss 0.286166 LR 0.001000 Time 0.022790 +2023-10-02 20:38:19,492 - Epoch: [20][ 420/ 1236] Overall Loss 0.285396 Objective Loss 0.285396 LR 0.001000 Time 0.022742 +2023-10-02 20:38:19,700 - Epoch: [20][ 430/ 1236] Overall Loss 0.285559 Objective Loss 0.285559 LR 0.001000 Time 0.022695 +2023-10-02 20:38:19,908 - Epoch: [20][ 440/ 1236] Overall Loss 0.285158 Objective Loss 0.285158 LR 0.001000 Time 0.022652 +2023-10-02 20:38:20,115 - Epoch: [20][ 450/ 1236] Overall Loss 0.285498 Objective Loss 0.285498 LR 0.001000 Time 0.022608 +2023-10-02 20:38:20,324 - Epoch: [20][ 460/ 1236] Overall Loss 0.285899 Objective Loss 0.285899 LR 0.001000 Time 0.022570 +2023-10-02 20:38:20,531 - Epoch: [20][ 470/ 1236] Overall Loss 0.285977 Objective Loss 0.285977 LR 0.001000 Time 0.022530 +2023-10-02 20:38:20,739 - Epoch: [20][ 480/ 1236] Overall Loss 0.285358 Objective Loss 0.285358 LR 0.001000 Time 0.022493 +2023-10-02 20:38:20,947 - Epoch: [20][ 490/ 1236] Overall Loss 0.284988 Objective Loss 0.284988 LR 0.001000 Time 0.022457 +2023-10-02 20:38:21,155 - Epoch: [20][ 500/ 1236] Overall Loss 0.285025 Objective Loss 0.285025 LR 0.001000 Time 0.022424 +2023-10-02 20:38:21,362 - Epoch: [20][ 510/ 1236] Overall Loss 0.284970 Objective Loss 0.284970 LR 0.001000 Time 0.022390 +2023-10-02 20:38:21,571 - Epoch: [20][ 520/ 1236] Overall Loss 0.285838 Objective Loss 0.285838 LR 0.001000 Time 0.022360 +2023-10-02 20:38:21,778 - Epoch: [20][ 530/ 1236] Overall Loss 0.285729 Objective Loss 0.285729 LR 0.001000 Time 0.022329 +2023-10-02 20:38:21,986 - Epoch: [20][ 540/ 1236] Overall Loss 0.285893 Objective Loss 0.285893 LR 0.001000 Time 0.022300 +2023-10-02 20:38:22,194 - Epoch: [20][ 550/ 1236] Overall Loss 0.286038 Objective Loss 0.286038 LR 0.001000 Time 0.022271 +2023-10-02 20:38:22,402 - Epoch: [20][ 560/ 1236] Overall Loss 0.286175 Objective Loss 0.286175 LR 0.001000 Time 0.022245 +2023-10-02 20:38:22,609 - Epoch: [20][ 570/ 1236] Overall Loss 0.286410 Objective Loss 0.286410 LR 0.001000 Time 0.022218 +2023-10-02 20:38:22,818 - Epoch: [20][ 580/ 1236] Overall Loss 0.286253 Objective Loss 0.286253 LR 0.001000 Time 0.022194 +2023-10-02 20:38:23,025 - Epoch: [20][ 590/ 1236] Overall Loss 0.286748 Objective Loss 0.286748 LR 0.001000 Time 0.022168 +2023-10-02 20:38:23,233 - Epoch: [20][ 600/ 1236] Overall Loss 0.286497 Objective Loss 0.286497 LR 0.001000 Time 0.022145 +2023-10-02 20:38:23,441 - Epoch: [20][ 610/ 1236] Overall Loss 0.286691 Objective Loss 0.286691 LR 0.001000 Time 0.022122 +2023-10-02 20:38:23,649 - Epoch: [20][ 620/ 1236] Overall Loss 0.286286 Objective Loss 0.286286 LR 0.001000 Time 0.022101 +2023-10-02 20:38:23,857 - Epoch: [20][ 630/ 1236] Overall Loss 0.286220 Objective Loss 0.286220 LR 0.001000 Time 0.022079 +2023-10-02 20:38:24,065 - Epoch: [20][ 640/ 1236] Overall Loss 0.286444 Objective Loss 0.286444 LR 0.001000 Time 0.022059 +2023-10-02 20:38:24,273 - Epoch: [20][ 650/ 1236] Overall Loss 0.286209 Objective Loss 0.286209 LR 0.001000 Time 0.022039 +2023-10-02 20:38:24,481 - Epoch: [20][ 660/ 1236] Overall Loss 0.286350 Objective Loss 0.286350 LR 0.001000 Time 0.022020 +2023-10-02 20:38:24,689 - Epoch: [20][ 670/ 1236] Overall Loss 0.286043 Objective Loss 0.286043 LR 0.001000 Time 0.022001 +2023-10-02 20:38:24,897 - Epoch: [20][ 680/ 1236] Overall Loss 0.286273 Objective Loss 0.286273 LR 0.001000 Time 0.021983 +2023-10-02 20:38:25,105 - Epoch: [20][ 690/ 1236] Overall Loss 0.286897 Objective Loss 0.286897 LR 0.001000 Time 0.021965 +2023-10-02 20:38:25,313 - Epoch: [20][ 700/ 1236] Overall Loss 0.286796 Objective Loss 0.286796 LR 0.001000 Time 0.021948 +2023-10-02 20:38:25,521 - Epoch: [20][ 710/ 1236] Overall Loss 0.286808 Objective Loss 0.286808 LR 0.001000 Time 0.021931 +2023-10-02 20:38:25,729 - Epoch: [20][ 720/ 1236] Overall Loss 0.286486 Objective Loss 0.286486 LR 0.001000 Time 0.021916 +2023-10-02 20:38:25,937 - Epoch: [20][ 730/ 1236] Overall Loss 0.286432 Objective Loss 0.286432 LR 0.001000 Time 0.021900 +2023-10-02 20:38:26,145 - Epoch: [20][ 740/ 1236] Overall Loss 0.286757 Objective Loss 0.286757 LR 0.001000 Time 0.021885 +2023-10-02 20:38:26,353 - Epoch: [20][ 750/ 1236] Overall Loss 0.287183 Objective Loss 0.287183 LR 0.001000 Time 0.021870 +2023-10-02 20:38:26,561 - Epoch: [20][ 760/ 1236] Overall Loss 0.287593 Objective Loss 0.287593 LR 0.001000 Time 0.021855 +2023-10-02 20:38:26,769 - Epoch: [20][ 770/ 1236] Overall Loss 0.287885 Objective Loss 0.287885 LR 0.001000 Time 0.021839 +2023-10-02 20:38:26,977 - Epoch: [20][ 780/ 1236] Overall Loss 0.287817 Objective Loss 0.287817 LR 0.001000 Time 0.021826 +2023-10-02 20:38:27,185 - Epoch: [20][ 790/ 1236] Overall Loss 0.287822 Objective Loss 0.287822 LR 0.001000 Time 0.021811 +2023-10-02 20:38:27,393 - Epoch: [20][ 800/ 1236] Overall Loss 0.287621 Objective Loss 0.287621 LR 0.001000 Time 0.021798 +2023-10-02 20:38:27,601 - Epoch: [20][ 810/ 1236] Overall Loss 0.287649 Objective Loss 0.287649 LR 0.001000 Time 0.021785 +2023-10-02 20:38:27,809 - Epoch: [20][ 820/ 1236] Overall Loss 0.287870 Objective Loss 0.287870 LR 0.001000 Time 0.021773 +2023-10-02 20:38:28,017 - Epoch: [20][ 830/ 1236] Overall Loss 0.287861 Objective Loss 0.287861 LR 0.001000 Time 0.021760 +2023-10-02 20:38:28,226 - Epoch: [20][ 840/ 1236] Overall Loss 0.287547 Objective Loss 0.287547 LR 0.001000 Time 0.021749 +2023-10-02 20:38:28,433 - Epoch: [20][ 850/ 1236] Overall Loss 0.287642 Objective Loss 0.287642 LR 0.001000 Time 0.021737 +2023-10-02 20:38:28,642 - Epoch: [20][ 860/ 1236] Overall Loss 0.287621 Objective Loss 0.287621 LR 0.001000 Time 0.021726 +2023-10-02 20:38:28,849 - Epoch: [20][ 870/ 1236] Overall Loss 0.287527 Objective Loss 0.287527 LR 0.001000 Time 0.021715 +2023-10-02 20:38:29,057 - Epoch: [20][ 880/ 1236] Overall Loss 0.287487 Objective Loss 0.287487 LR 0.001000 Time 0.021704 +2023-10-02 20:38:29,265 - Epoch: [20][ 890/ 1236] Overall Loss 0.287735 Objective Loss 0.287735 LR 0.001000 Time 0.021693 +2023-10-02 20:38:29,474 - Epoch: [20][ 900/ 1236] Overall Loss 0.287781 Objective Loss 0.287781 LR 0.001000 Time 0.021683 +2023-10-02 20:38:29,681 - Epoch: [20][ 910/ 1236] Overall Loss 0.288053 Objective Loss 0.288053 LR 0.001000 Time 0.021673 +2023-10-02 20:38:29,890 - Epoch: [20][ 920/ 1236] Overall Loss 0.287857 Objective Loss 0.287857 LR 0.001000 Time 0.021664 +2023-10-02 20:38:30,097 - Epoch: [20][ 930/ 1236] Overall Loss 0.287927 Objective Loss 0.287927 LR 0.001000 Time 0.021653 +2023-10-02 20:38:30,306 - Epoch: [20][ 940/ 1236] Overall Loss 0.288161 Objective Loss 0.288161 LR 0.001000 Time 0.021645 +2023-10-02 20:38:30,513 - Epoch: [20][ 950/ 1236] Overall Loss 0.288258 Objective Loss 0.288258 LR 0.001000 Time 0.021635 +2023-10-02 20:38:30,721 - Epoch: [20][ 960/ 1236] Overall Loss 0.288474 Objective Loss 0.288474 LR 0.001000 Time 0.021626 +2023-10-02 20:38:30,929 - Epoch: [20][ 970/ 1236] Overall Loss 0.288604 Objective Loss 0.288604 LR 0.001000 Time 0.021617 +2023-10-02 20:38:31,138 - Epoch: [20][ 980/ 1236] Overall Loss 0.288880 Objective Loss 0.288880 LR 0.001000 Time 0.021609 +2023-10-02 20:38:31,345 - Epoch: [20][ 990/ 1236] Overall Loss 0.289136 Objective Loss 0.289136 LR 0.001000 Time 0.021600 +2023-10-02 20:38:31,554 - Epoch: [20][ 1000/ 1236] Overall Loss 0.289513 Objective Loss 0.289513 LR 0.001000 Time 0.021592 +2023-10-02 20:38:31,762 - Epoch: [20][ 1010/ 1236] Overall Loss 0.289661 Objective Loss 0.289661 LR 0.001000 Time 0.021584 +2023-10-02 20:38:31,971 - Epoch: [20][ 1020/ 1236] Overall Loss 0.289770 Objective Loss 0.289770 LR 0.001000 Time 0.021577 +2023-10-02 20:38:32,178 - Epoch: [20][ 1030/ 1236] Overall Loss 0.289955 Objective Loss 0.289955 LR 0.001000 Time 0.021569 +2023-10-02 20:38:32,387 - Epoch: [20][ 1040/ 1236] Overall Loss 0.289757 Objective Loss 0.289757 LR 0.001000 Time 0.021562 +2023-10-02 20:38:32,594 - Epoch: [20][ 1050/ 1236] Overall Loss 0.289793 Objective Loss 0.289793 LR 0.001000 Time 0.021553 +2023-10-02 20:38:32,803 - Epoch: [20][ 1060/ 1236] Overall Loss 0.289888 Objective Loss 0.289888 LR 0.001000 Time 0.021547 +2023-10-02 20:38:33,010 - Epoch: [20][ 1070/ 1236] Overall Loss 0.289997 Objective Loss 0.289997 LR 0.001000 Time 0.021539 +2023-10-02 20:38:33,219 - Epoch: [20][ 1080/ 1236] Overall Loss 0.289707 Objective Loss 0.289707 LR 0.001000 Time 0.021532 +2023-10-02 20:38:33,427 - Epoch: [20][ 1090/ 1236] Overall Loss 0.289680 Objective Loss 0.289680 LR 0.001000 Time 0.021525 +2023-10-02 20:38:33,635 - Epoch: [20][ 1100/ 1236] Overall Loss 0.289725 Objective Loss 0.289725 LR 0.001000 Time 0.021518 +2023-10-02 20:38:33,843 - Epoch: [20][ 1110/ 1236] Overall Loss 0.289738 Objective Loss 0.289738 LR 0.001000 Time 0.021511 +2023-10-02 20:38:34,052 - Epoch: [20][ 1120/ 1236] Overall Loss 0.289759 Objective Loss 0.289759 LR 0.001000 Time 0.021505 +2023-10-02 20:38:34,259 - Epoch: [20][ 1130/ 1236] Overall Loss 0.289737 Objective Loss 0.289737 LR 0.001000 Time 0.021498 +2023-10-02 20:38:34,468 - Epoch: [20][ 1140/ 1236] Overall Loss 0.289864 Objective Loss 0.289864 LR 0.001000 Time 0.021492 +2023-10-02 20:38:34,675 - Epoch: [20][ 1150/ 1236] Overall Loss 0.290127 Objective Loss 0.290127 LR 0.001000 Time 0.021484 +2023-10-02 20:38:34,884 - Epoch: [20][ 1160/ 1236] Overall Loss 0.289917 Objective Loss 0.289917 LR 0.001000 Time 0.021479 +2023-10-02 20:38:35,091 - Epoch: [20][ 1170/ 1236] Overall Loss 0.289786 Objective Loss 0.289786 LR 0.001000 Time 0.021472 +2023-10-02 20:38:35,300 - Epoch: [20][ 1180/ 1236] Overall Loss 0.289937 Objective Loss 0.289937 LR 0.001000 Time 0.021467 +2023-10-02 20:38:35,507 - Epoch: [20][ 1190/ 1236] Overall Loss 0.289949 Objective Loss 0.289949 LR 0.001000 Time 0.021461 +2023-10-02 20:38:35,716 - Epoch: [20][ 1200/ 1236] Overall Loss 0.289927 Objective Loss 0.289927 LR 0.001000 Time 0.021455 +2023-10-02 20:38:35,923 - Epoch: [20][ 1210/ 1236] Overall Loss 0.289987 Objective Loss 0.289987 LR 0.001000 Time 0.021449 +2023-10-02 20:38:36,132 - Epoch: [20][ 1220/ 1236] Overall Loss 0.290026 Objective Loss 0.290026 LR 0.001000 Time 0.021444 +2023-10-02 20:38:36,394 - Epoch: [20][ 1230/ 1236] Overall Loss 0.290145 Objective Loss 0.290145 LR 0.001000 Time 0.021483 +2023-10-02 20:38:36,517 - Epoch: [20][ 1236/ 1236] Overall Loss 0.290139 Objective Loss 0.290139 Top1 86.761711 Top5 98.574338 LR 0.001000 Time 0.021478 +2023-10-02 20:38:36,662 - --- validate (epoch=20)----------- +2023-10-02 20:38:36,662 - 29943 samples (256 per mini-batch) +2023-10-02 20:38:37,162 - Epoch: [20][ 10/ 117] Loss 0.330156 Top1 83.632812 Top5 98.398438 +2023-10-02 20:38:37,314 - Epoch: [20][ 20/ 117] Loss 0.330728 Top1 83.300781 Top5 98.261719 +2023-10-02 20:38:37,465 - Epoch: [20][ 30/ 117] Loss 0.330717 Top1 83.111979 Top5 98.281250 +2023-10-02 20:38:37,614 - Epoch: [20][ 40/ 117] Loss 0.328627 Top1 83.173828 Top5 98.222656 +2023-10-02 20:38:37,766 - Epoch: [20][ 50/ 117] Loss 0.329061 Top1 83.187500 Top5 98.257812 +2023-10-02 20:38:37,916 - Epoch: [20][ 60/ 117] Loss 0.330070 Top1 83.125000 Top5 98.287760 +2023-10-02 20:38:38,067 - Epoch: [20][ 70/ 117] Loss 0.327687 Top1 83.063616 Top5 98.270089 +2023-10-02 20:38:38,217 - Epoch: [20][ 80/ 117] Loss 0.328782 Top1 82.963867 Top5 98.251953 +2023-10-02 20:38:38,369 - Epoch: [20][ 90/ 117] Loss 0.328556 Top1 82.873264 Top5 98.272569 +2023-10-02 20:38:38,520 - Epoch: [20][ 100/ 117] Loss 0.331903 Top1 82.667969 Top5 98.238281 +2023-10-02 20:38:38,677 - Epoch: [20][ 110/ 117] Loss 0.331387 Top1 82.688210 Top5 98.227983 +2023-10-02 20:38:38,767 - Epoch: [20][ 117/ 117] Loss 0.331964 Top1 82.690445 Top5 98.199913 +2023-10-02 20:38:38,913 - ==> Top1: 82.690 Top5: 98.200 Loss: 0.332 + +2023-10-02 20:38:38,914 - ==> Confusion: +[[ 907 0 9 1 13 4 0 0 3 85 0 1 0 3 3 0 3 0 0 0 18] + [ 0 1034 3 1 1 41 1 19 4 1 1 0 1 0 2 3 4 0 5 2 8] + [ 6 0 950 17 3 1 28 12 1 1 1 0 7 3 0 4 0 1 9 2 10] + [ 1 2 8 977 2 6 2 1 9 1 3 0 7 4 25 1 1 2 24 0 13] + [ 23 5 0 0 956 11 0 0 1 13 0 0 0 3 9 5 16 0 0 3 5] + [ 3 38 2 3 3 1007 1 21 3 7 4 3 1 5 5 0 1 0 1 0 8] + [ 0 7 18 0 0 1 1120 10 0 0 0 0 0 1 0 11 1 3 1 11 7] + [ 1 24 16 0 2 31 5 1061 1 1 1 5 4 4 1 2 1 1 41 10 6] + [ 17 3 1 0 2 1 0 0 964 65 3 1 4 4 16 1 1 0 3 1 2] + [ 93 1 1 0 7 4 0 0 13 966 0 0 2 12 8 1 1 0 0 4 6] + [ 3 2 5 16 2 5 0 2 30 2 954 1 1 4 6 1 1 2 6 0 10] + [ 0 1 1 1 0 20 0 2 0 3 0 931 35 5 0 5 1 17 0 6 7] + [ 1 0 4 3 1 5 0 0 2 2 1 36 966 1 4 5 0 20 2 3 12] + [ 0 0 1 0 3 34 3 0 18 31 9 4 3 985 6 1 3 1 0 2 15] + [ 8 2 1 15 6 2 0 0 33 5 0 0 3 2 1002 0 0 3 6 0 13] + [ 0 0 2 2 4 0 1 0 0 3 0 7 6 1 0 1059 19 20 1 4 5] + [ 1 19 0 0 5 8 0 1 1 0 1 2 1 0 1 10 1092 0 1 5 13] + [ 0 1 1 5 0 1 3 0 1 0 0 5 19 0 0 2 0 998 0 0 2] + [ 0 8 3 14 1 1 0 39 3 0 0 0 1 0 13 0 1 0 972 4 8] + [ 0 1 3 1 1 6 9 6 0 0 2 13 3 2 0 4 4 1 1 1089 6] + [ 134 217 120 105 116 294 39 111 117 120 115 134 443 185 189 76 130 82 187 221 4770]] + +2023-10-02 20:38:38,915 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:38:38,915 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:38:38,921 - + +2023-10-02 20:38:38,922 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:38:40,051 - Epoch: [21][ 10/ 1236] Overall Loss 0.282059 Objective Loss 0.282059 LR 0.001000 Time 0.112850 +2023-10-02 20:38:40,258 - Epoch: [21][ 20/ 1236] Overall Loss 0.288274 Objective Loss 0.288274 LR 0.001000 Time 0.066760 +2023-10-02 20:38:40,464 - Epoch: [21][ 30/ 1236] Overall Loss 0.283312 Objective Loss 0.283312 LR 0.001000 Time 0.051370 +2023-10-02 20:38:40,673 - Epoch: [21][ 40/ 1236] Overall Loss 0.282348 Objective Loss 0.282348 LR 0.001000 Time 0.043733 +2023-10-02 20:38:40,877 - Epoch: [21][ 50/ 1236] Overall Loss 0.277429 Objective Loss 0.277429 LR 0.001000 Time 0.039075 +2023-10-02 20:38:41,086 - Epoch: [21][ 60/ 1236] Overall Loss 0.273706 Objective Loss 0.273706 LR 0.001000 Time 0.036035 +2023-10-02 20:38:41,291 - Epoch: [21][ 70/ 1236] Overall Loss 0.272889 Objective Loss 0.272889 LR 0.001000 Time 0.033812 +2023-10-02 20:38:41,498 - Epoch: [21][ 80/ 1236] Overall Loss 0.272974 Objective Loss 0.272974 LR 0.001000 Time 0.032171 +2023-10-02 20:38:41,704 - Epoch: [21][ 90/ 1236] Overall Loss 0.270948 Objective Loss 0.270948 LR 0.001000 Time 0.030870 +2023-10-02 20:38:41,912 - Epoch: [21][ 100/ 1236] Overall Loss 0.272144 Objective Loss 0.272144 LR 0.001000 Time 0.029854 +2023-10-02 20:38:42,118 - Epoch: [21][ 110/ 1236] Overall Loss 0.274103 Objective Loss 0.274103 LR 0.001000 Time 0.029001 +2023-10-02 20:38:42,327 - Epoch: [21][ 120/ 1236] Overall Loss 0.275279 Objective Loss 0.275279 LR 0.001000 Time 0.028321 +2023-10-02 20:38:42,532 - Epoch: [21][ 130/ 1236] Overall Loss 0.278150 Objective Loss 0.278150 LR 0.001000 Time 0.027723 +2023-10-02 20:38:42,739 - Epoch: [21][ 140/ 1236] Overall Loss 0.278456 Objective Loss 0.278456 LR 0.001000 Time 0.027216 +2023-10-02 20:38:42,947 - Epoch: [21][ 150/ 1236] Overall Loss 0.279572 Objective Loss 0.279572 LR 0.001000 Time 0.026784 +2023-10-02 20:38:43,158 - Epoch: [21][ 160/ 1236] Overall Loss 0.280283 Objective Loss 0.280283 LR 0.001000 Time 0.026422 +2023-10-02 20:38:43,370 - Epoch: [21][ 170/ 1236] Overall Loss 0.280495 Objective Loss 0.280495 LR 0.001000 Time 0.026105 +2023-10-02 20:38:43,582 - Epoch: [21][ 180/ 1236] Overall Loss 0.280875 Objective Loss 0.280875 LR 0.001000 Time 0.025824 +2023-10-02 20:38:43,794 - Epoch: [21][ 190/ 1236] Overall Loss 0.282989 Objective Loss 0.282989 LR 0.001000 Time 0.025571 +2023-10-02 20:38:44,006 - Epoch: [21][ 200/ 1236] Overall Loss 0.284359 Objective Loss 0.284359 LR 0.001000 Time 0.025345 +2023-10-02 20:38:44,218 - Epoch: [21][ 210/ 1236] Overall Loss 0.283967 Objective Loss 0.283967 LR 0.001000 Time 0.025139 +2023-10-02 20:38:44,430 - Epoch: [21][ 220/ 1236] Overall Loss 0.283514 Objective Loss 0.283514 LR 0.001000 Time 0.024955 +2023-10-02 20:38:44,642 - Epoch: [21][ 230/ 1236] Overall Loss 0.283257 Objective Loss 0.283257 LR 0.001000 Time 0.024784 +2023-10-02 20:38:44,856 - Epoch: [21][ 240/ 1236] Overall Loss 0.283963 Objective Loss 0.283963 LR 0.001000 Time 0.024633 +2023-10-02 20:38:45,066 - Epoch: [21][ 250/ 1236] Overall Loss 0.283439 Objective Loss 0.283439 LR 0.001000 Time 0.024488 +2023-10-02 20:38:45,279 - Epoch: [21][ 260/ 1236] Overall Loss 0.283856 Objective Loss 0.283856 LR 0.001000 Time 0.024360 +2023-10-02 20:38:45,490 - Epoch: [21][ 270/ 1236] Overall Loss 0.283277 Objective Loss 0.283277 LR 0.001000 Time 0.024236 +2023-10-02 20:38:45,703 - Epoch: [21][ 280/ 1236] Overall Loss 0.283371 Objective Loss 0.283371 LR 0.001000 Time 0.024125 +2023-10-02 20:38:45,914 - Epoch: [21][ 290/ 1236] Overall Loss 0.283765 Objective Loss 0.283765 LR 0.001000 Time 0.024018 +2023-10-02 20:38:46,125 - Epoch: [21][ 300/ 1236] Overall Loss 0.282715 Objective Loss 0.282715 LR 0.001000 Time 0.023915 +2023-10-02 20:38:46,335 - Epoch: [21][ 310/ 1236] Overall Loss 0.282031 Objective Loss 0.282031 LR 0.001000 Time 0.023822 +2023-10-02 20:38:46,543 - Epoch: [21][ 320/ 1236] Overall Loss 0.283040 Objective Loss 0.283040 LR 0.001000 Time 0.023726 +2023-10-02 20:38:46,753 - Epoch: [21][ 330/ 1236] Overall Loss 0.282237 Objective Loss 0.282237 LR 0.001000 Time 0.023637 +2023-10-02 20:38:46,962 - Epoch: [21][ 340/ 1236] Overall Loss 0.281394 Objective Loss 0.281394 LR 0.001000 Time 0.023552 +2023-10-02 20:38:47,173 - Epoch: [21][ 350/ 1236] Overall Loss 0.282631 Objective Loss 0.282631 LR 0.001000 Time 0.023481 +2023-10-02 20:38:47,380 - Epoch: [21][ 360/ 1236] Overall Loss 0.282110 Objective Loss 0.282110 LR 0.001000 Time 0.023403 +2023-10-02 20:38:47,592 - Epoch: [21][ 370/ 1236] Overall Loss 0.282734 Objective Loss 0.282734 LR 0.001000 Time 0.023343 +2023-10-02 20:38:47,800 - Epoch: [21][ 380/ 1236] Overall Loss 0.282551 Objective Loss 0.282551 LR 0.001000 Time 0.023276 +2023-10-02 20:38:48,011 - Epoch: [21][ 390/ 1236] Overall Loss 0.283197 Objective Loss 0.283197 LR 0.001000 Time 0.023219 +2023-10-02 20:38:48,219 - Epoch: [21][ 400/ 1236] Overall Loss 0.283212 Objective Loss 0.283212 LR 0.001000 Time 0.023157 +2023-10-02 20:38:48,429 - Epoch: [21][ 410/ 1236] Overall Loss 0.283585 Objective Loss 0.283585 LR 0.001000 Time 0.023099 +2023-10-02 20:38:48,638 - Epoch: [21][ 420/ 1236] Overall Loss 0.283123 Objective Loss 0.283123 LR 0.001000 Time 0.023043 +2023-10-02 20:38:48,847 - Epoch: [21][ 430/ 1236] Overall Loss 0.282790 Objective Loss 0.282790 LR 0.001000 Time 0.022993 +2023-10-02 20:38:49,056 - Epoch: [21][ 440/ 1236] Overall Loss 0.282866 Objective Loss 0.282866 LR 0.001000 Time 0.022941 +2023-10-02 20:38:49,266 - Epoch: [21][ 450/ 1236] Overall Loss 0.282520 Objective Loss 0.282520 LR 0.001000 Time 0.022897 +2023-10-02 20:38:49,474 - Epoch: [21][ 460/ 1236] Overall Loss 0.283038 Objective Loss 0.283038 LR 0.001000 Time 0.022849 +2023-10-02 20:38:49,684 - Epoch: [21][ 470/ 1236] Overall Loss 0.283204 Objective Loss 0.283204 LR 0.001000 Time 0.022808 +2023-10-02 20:38:49,893 - Epoch: [21][ 480/ 1236] Overall Loss 0.282651 Objective Loss 0.282651 LR 0.001000 Time 0.022765 +2023-10-02 20:38:50,102 - Epoch: [21][ 490/ 1236] Overall Loss 0.283147 Objective Loss 0.283147 LR 0.001000 Time 0.022725 +2023-10-02 20:38:50,311 - Epoch: [21][ 500/ 1236] Overall Loss 0.282612 Objective Loss 0.282612 LR 0.001000 Time 0.022684 +2023-10-02 20:38:50,522 - Epoch: [21][ 510/ 1236] Overall Loss 0.282393 Objective Loss 0.282393 LR 0.001000 Time 0.022652 +2023-10-02 20:38:50,730 - Epoch: [21][ 520/ 1236] Overall Loss 0.281928 Objective Loss 0.281928 LR 0.001000 Time 0.022616 +2023-10-02 20:38:50,941 - Epoch: [21][ 530/ 1236] Overall Loss 0.282805 Objective Loss 0.282805 LR 0.001000 Time 0.022587 +2023-10-02 20:38:51,148 - Epoch: [21][ 540/ 1236] Overall Loss 0.283023 Objective Loss 0.283023 LR 0.001000 Time 0.022552 +2023-10-02 20:38:51,359 - Epoch: [21][ 550/ 1236] Overall Loss 0.283019 Objective Loss 0.283019 LR 0.001000 Time 0.022524 +2023-10-02 20:38:51,567 - Epoch: [21][ 560/ 1236] Overall Loss 0.283774 Objective Loss 0.283774 LR 0.001000 Time 0.022492 +2023-10-02 20:38:51,778 - Epoch: [21][ 570/ 1236] Overall Loss 0.283628 Objective Loss 0.283628 LR 0.001000 Time 0.022467 +2023-10-02 20:38:51,985 - Epoch: [21][ 580/ 1236] Overall Loss 0.283179 Objective Loss 0.283179 LR 0.001000 Time 0.022436 +2023-10-02 20:38:52,195 - Epoch: [21][ 590/ 1236] Overall Loss 0.283487 Objective Loss 0.283487 LR 0.001000 Time 0.022411 +2023-10-02 20:38:52,404 - Epoch: [21][ 600/ 1236] Overall Loss 0.283589 Objective Loss 0.283589 LR 0.001000 Time 0.022383 +2023-10-02 20:38:52,614 - Epoch: [21][ 610/ 1236] Overall Loss 0.283890 Objective Loss 0.283890 LR 0.001000 Time 0.022359 +2023-10-02 20:38:52,822 - Epoch: [21][ 620/ 1236] Overall Loss 0.283730 Objective Loss 0.283730 LR 0.001000 Time 0.022333 +2023-10-02 20:38:53,032 - Epoch: [21][ 630/ 1236] Overall Loss 0.283755 Objective Loss 0.283755 LR 0.001000 Time 0.022310 +2023-10-02 20:38:53,241 - Epoch: [21][ 640/ 1236] Overall Loss 0.283778 Objective Loss 0.283778 LR 0.001000 Time 0.022285 +2023-10-02 20:38:53,451 - Epoch: [21][ 650/ 1236] Overall Loss 0.283293 Objective Loss 0.283293 LR 0.001000 Time 0.022266 +2023-10-02 20:38:53,659 - Epoch: [21][ 660/ 1236] Overall Loss 0.282987 Objective Loss 0.282987 LR 0.001000 Time 0.022243 +2023-10-02 20:38:53,872 - Epoch: [21][ 670/ 1236] Overall Loss 0.283134 Objective Loss 0.283134 LR 0.001000 Time 0.022227 +2023-10-02 20:38:54,079 - Epoch: [21][ 680/ 1236] Overall Loss 0.283502 Objective Loss 0.283502 LR 0.001000 Time 0.022203 +2023-10-02 20:38:54,291 - Epoch: [21][ 690/ 1236] Overall Loss 0.283839 Objective Loss 0.283839 LR 0.001000 Time 0.022189 +2023-10-02 20:38:54,500 - Epoch: [21][ 700/ 1236] Overall Loss 0.283925 Objective Loss 0.283925 LR 0.001000 Time 0.022169 +2023-10-02 20:38:54,712 - Epoch: [21][ 710/ 1236] Overall Loss 0.283920 Objective Loss 0.283920 LR 0.001000 Time 0.022155 +2023-10-02 20:38:54,920 - Epoch: [21][ 720/ 1236] Overall Loss 0.284222 Objective Loss 0.284222 LR 0.001000 Time 0.022135 +2023-10-02 20:38:55,132 - Epoch: [21][ 730/ 1236] Overall Loss 0.284549 Objective Loss 0.284549 LR 0.001000 Time 0.022123 +2023-10-02 20:38:55,339 - Epoch: [21][ 740/ 1236] Overall Loss 0.284703 Objective Loss 0.284703 LR 0.001000 Time 0.022103 +2023-10-02 20:38:55,552 - Epoch: [21][ 750/ 1236] Overall Loss 0.284482 Objective Loss 0.284482 LR 0.001000 Time 0.022090 +2023-10-02 20:38:55,759 - Epoch: [21][ 760/ 1236] Overall Loss 0.284548 Objective Loss 0.284548 LR 0.001000 Time 0.022072 +2023-10-02 20:38:55,971 - Epoch: [21][ 770/ 1236] Overall Loss 0.284538 Objective Loss 0.284538 LR 0.001000 Time 0.022061 +2023-10-02 20:38:56,179 - Epoch: [21][ 780/ 1236] Overall Loss 0.284661 Objective Loss 0.284661 LR 0.001000 Time 0.022044 +2023-10-02 20:38:56,391 - Epoch: [21][ 790/ 1236] Overall Loss 0.284476 Objective Loss 0.284476 LR 0.001000 Time 0.022032 +2023-10-02 20:38:56,599 - Epoch: [21][ 800/ 1236] Overall Loss 0.284474 Objective Loss 0.284474 LR 0.001000 Time 0.022016 +2023-10-02 20:38:56,811 - Epoch: [21][ 810/ 1236] Overall Loss 0.284621 Objective Loss 0.284621 LR 0.001000 Time 0.022006 +2023-10-02 20:38:57,019 - Epoch: [21][ 820/ 1236] Overall Loss 0.284741 Objective Loss 0.284741 LR 0.001000 Time 0.021991 +2023-10-02 20:38:57,231 - Epoch: [21][ 830/ 1236] Overall Loss 0.284827 Objective Loss 0.284827 LR 0.001000 Time 0.021981 +2023-10-02 20:38:57,441 - Epoch: [21][ 840/ 1236] Overall Loss 0.285263 Objective Loss 0.285263 LR 0.001000 Time 0.021966 +2023-10-02 20:38:57,653 - Epoch: [21][ 850/ 1236] Overall Loss 0.285686 Objective Loss 0.285686 LR 0.001000 Time 0.021957 +2023-10-02 20:38:57,861 - Epoch: [21][ 860/ 1236] Overall Loss 0.285880 Objective Loss 0.285880 LR 0.001000 Time 0.021943 +2023-10-02 20:38:58,073 - Epoch: [21][ 870/ 1236] Overall Loss 0.286154 Objective Loss 0.286154 LR 0.001000 Time 0.021935 +2023-10-02 20:38:58,282 - Epoch: [21][ 880/ 1236] Overall Loss 0.285968 Objective Loss 0.285968 LR 0.001000 Time 0.021921 +2023-10-02 20:38:58,494 - Epoch: [21][ 890/ 1236] Overall Loss 0.286360 Objective Loss 0.286360 LR 0.001000 Time 0.021913 +2023-10-02 20:38:58,702 - Epoch: [21][ 900/ 1236] Overall Loss 0.286577 Objective Loss 0.286577 LR 0.001000 Time 0.021900 +2023-10-02 20:38:58,914 - Epoch: [21][ 910/ 1236] Overall Loss 0.286779 Objective Loss 0.286779 LR 0.001000 Time 0.021892 +2023-10-02 20:38:59,122 - Epoch: [21][ 920/ 1236] Overall Loss 0.286519 Objective Loss 0.286519 LR 0.001000 Time 0.021879 +2023-10-02 20:38:59,334 - Epoch: [21][ 930/ 1236] Overall Loss 0.286791 Objective Loss 0.286791 LR 0.001000 Time 0.021872 +2023-10-02 20:38:59,542 - Epoch: [21][ 940/ 1236] Overall Loss 0.287198 Objective Loss 0.287198 LR 0.001000 Time 0.021860 +2023-10-02 20:38:59,754 - Epoch: [21][ 950/ 1236] Overall Loss 0.287599 Objective Loss 0.287599 LR 0.001000 Time 0.021853 +2023-10-02 20:38:59,962 - Epoch: [21][ 960/ 1236] Overall Loss 0.287813 Objective Loss 0.287813 LR 0.001000 Time 0.021841 +2023-10-02 20:39:00,173 - Epoch: [21][ 970/ 1236] Overall Loss 0.287871 Objective Loss 0.287871 LR 0.001000 Time 0.021833 +2023-10-02 20:39:00,379 - Epoch: [21][ 980/ 1236] Overall Loss 0.288341 Objective Loss 0.288341 LR 0.001000 Time 0.021821 +2023-10-02 20:39:00,589 - Epoch: [21][ 990/ 1236] Overall Loss 0.288576 Objective Loss 0.288576 LR 0.001000 Time 0.021811 +2023-10-02 20:39:00,796 - Epoch: [21][ 1000/ 1236] Overall Loss 0.288641 Objective Loss 0.288641 LR 0.001000 Time 0.021800 +2023-10-02 20:39:01,008 - Epoch: [21][ 1010/ 1236] Overall Loss 0.288577 Objective Loss 0.288577 LR 0.001000 Time 0.021793 +2023-10-02 20:39:01,214 - Epoch: [21][ 1020/ 1236] Overall Loss 0.288764 Objective Loss 0.288764 LR 0.001000 Time 0.021781 +2023-10-02 20:39:01,424 - Epoch: [21][ 1030/ 1236] Overall Loss 0.288771 Objective Loss 0.288771 LR 0.001000 Time 0.021773 +2023-10-02 20:39:01,631 - Epoch: [21][ 1040/ 1236] Overall Loss 0.288667 Objective Loss 0.288667 LR 0.001000 Time 0.021763 +2023-10-02 20:39:01,842 - Epoch: [21][ 1050/ 1236] Overall Loss 0.288976 Objective Loss 0.288976 LR 0.001000 Time 0.021756 +2023-10-02 20:39:02,048 - Epoch: [21][ 1060/ 1236] Overall Loss 0.288938 Objective Loss 0.288938 LR 0.001000 Time 0.021745 +2023-10-02 20:39:02,259 - Epoch: [21][ 1070/ 1236] Overall Loss 0.289380 Objective Loss 0.289380 LR 0.001000 Time 0.021739 +2023-10-02 20:39:02,466 - Epoch: [21][ 1080/ 1236] Overall Loss 0.289066 Objective Loss 0.289066 LR 0.001000 Time 0.021728 +2023-10-02 20:39:02,677 - Epoch: [21][ 1090/ 1236] Overall Loss 0.289137 Objective Loss 0.289137 LR 0.001000 Time 0.021722 +2023-10-02 20:39:02,883 - Epoch: [21][ 1100/ 1236] Overall Loss 0.289403 Objective Loss 0.289403 LR 0.001000 Time 0.021711 +2023-10-02 20:39:03,094 - Epoch: [21][ 1110/ 1236] Overall Loss 0.289368 Objective Loss 0.289368 LR 0.001000 Time 0.021706 +2023-10-02 20:39:03,300 - Epoch: [21][ 1120/ 1236] Overall Loss 0.289513 Objective Loss 0.289513 LR 0.001000 Time 0.021696 +2023-10-02 20:39:03,510 - Epoch: [21][ 1130/ 1236] Overall Loss 0.289736 Objective Loss 0.289736 LR 0.001000 Time 0.021689 +2023-10-02 20:39:03,718 - Epoch: [21][ 1140/ 1236] Overall Loss 0.289659 Objective Loss 0.289659 LR 0.001000 Time 0.021681 +2023-10-02 20:39:03,928 - Epoch: [21][ 1150/ 1236] Overall Loss 0.289916 Objective Loss 0.289916 LR 0.001000 Time 0.021674 +2023-10-02 20:39:04,136 - Epoch: [21][ 1160/ 1236] Overall Loss 0.289993 Objective Loss 0.289993 LR 0.001000 Time 0.021666 +2023-10-02 20:39:04,347 - Epoch: [21][ 1170/ 1236] Overall Loss 0.290109 Objective Loss 0.290109 LR 0.001000 Time 0.021661 +2023-10-02 20:39:04,553 - Epoch: [21][ 1180/ 1236] Overall Loss 0.290254 Objective Loss 0.290254 LR 0.001000 Time 0.021652 +2023-10-02 20:39:04,764 - Epoch: [21][ 1190/ 1236] Overall Loss 0.290236 Objective Loss 0.290236 LR 0.001000 Time 0.021647 +2023-10-02 20:39:04,970 - Epoch: [21][ 1200/ 1236] Overall Loss 0.290286 Objective Loss 0.290286 LR 0.001000 Time 0.021638 +2023-10-02 20:39:05,180 - Epoch: [21][ 1210/ 1236] Overall Loss 0.290346 Objective Loss 0.290346 LR 0.001000 Time 0.021632 +2023-10-02 20:39:05,387 - Epoch: [21][ 1220/ 1236] Overall Loss 0.290337 Objective Loss 0.290337 LR 0.001000 Time 0.021624 +2023-10-02 20:39:05,646 - Epoch: [21][ 1230/ 1236] Overall Loss 0.290596 Objective Loss 0.290596 LR 0.001000 Time 0.021659 +2023-10-02 20:39:05,767 - Epoch: [21][ 1236/ 1236] Overall Loss 0.290684 Objective Loss 0.290684 Top1 84.928717 Top5 97.963340 LR 0.001000 Time 0.021651 +2023-10-02 20:39:05,890 - --- validate (epoch=21)----------- +2023-10-02 20:39:05,890 - 29943 samples (256 per mini-batch) +2023-10-02 20:39:06,367 - Epoch: [21][ 10/ 117] Loss 0.370108 Top1 81.250000 Top5 97.265625 +2023-10-02 20:39:06,516 - Epoch: [21][ 20/ 117] Loss 0.384052 Top1 80.468750 Top5 97.597656 +2023-10-02 20:39:06,665 - Epoch: [21][ 30/ 117] Loss 0.355985 Top1 81.197917 Top5 97.643229 +2023-10-02 20:39:06,814 - Epoch: [21][ 40/ 117] Loss 0.351144 Top1 81.552734 Top5 97.724609 +2023-10-02 20:39:06,963 - Epoch: [21][ 50/ 117] Loss 0.346184 Top1 81.648438 Top5 97.750000 +2023-10-02 20:39:07,112 - Epoch: [21][ 60/ 117] Loss 0.347801 Top1 81.529948 Top5 97.819010 +2023-10-02 20:39:07,261 - Epoch: [21][ 70/ 117] Loss 0.345292 Top1 81.523438 Top5 97.868304 +2023-10-02 20:39:07,409 - Epoch: [21][ 80/ 117] Loss 0.347528 Top1 81.489258 Top5 97.753906 +2023-10-02 20:39:07,558 - Epoch: [21][ 90/ 117] Loss 0.346432 Top1 81.467014 Top5 97.795139 +2023-10-02 20:39:07,706 - Epoch: [21][ 100/ 117] Loss 0.348617 Top1 81.574219 Top5 97.753906 +2023-10-02 20:39:07,862 - Epoch: [21][ 110/ 117] Loss 0.344566 Top1 81.704545 Top5 97.776989 +2023-10-02 20:39:07,950 - Epoch: [21][ 117/ 117] Loss 0.344215 Top1 81.678523 Top5 97.792472 +2023-10-02 20:39:08,093 - ==> Top1: 81.679 Top5: 97.792 Loss: 0.344 + +2023-10-02 20:39:08,093 - ==> Confusion: +[[ 899 0 6 0 10 2 0 0 1 93 1 4 3 1 7 2 6 3 0 0 12] + [ 0 1051 4 0 8 26 1 10 3 0 0 0 1 0 2 3 7 0 3 7 5] + [ 9 0 968 13 4 0 22 3 0 0 3 1 8 1 0 6 1 1 5 4 7] + [ 1 1 27 933 2 4 3 1 8 0 4 0 12 2 46 7 4 0 10 2 22] + [ 27 3 4 0 959 6 0 1 1 8 0 3 2 2 13 3 9 1 0 3 5] + [ 1 38 1 0 3 976 3 11 2 4 0 20 5 10 8 2 7 0 0 12 13] + [ 0 1 32 0 0 1 1128 3 0 0 0 3 1 0 0 2 0 0 1 12 7] + [ 3 22 26 0 5 44 3 1021 0 1 2 7 2 2 3 2 1 0 38 26 10] + [ 16 3 0 0 0 2 0 0 943 54 4 4 5 22 24 2 3 1 0 3 3] + [ 94 1 1 0 8 4 0 0 11 948 0 2 1 24 5 0 2 0 0 7 11] + [ 2 3 7 13 3 2 8 2 20 1 937 5 3 19 9 2 1 0 7 3 6] + [ 0 0 2 0 0 7 0 1 0 0 0 952 33 7 0 4 2 10 0 13 4] + [ 1 0 4 0 1 3 0 0 0 0 0 55 971 3 2 11 2 8 0 3 4] + [ 1 0 1 1 2 8 2 0 2 6 5 10 2 1060 4 1 1 1 0 7 5] + [ 12 1 1 6 3 0 0 0 15 2 1 0 2 3 1040 0 2 1 5 0 7] + [ 0 0 1 0 5 0 2 0 0 2 0 12 10 0 0 1062 20 7 0 8 5] + [ 0 12 2 0 3 3 1 0 2 1 0 9 0 1 2 7 1100 1 1 8 8] + [ 0 0 1 3 1 1 4 0 0 0 0 12 37 1 4 7 1 963 1 2 0] + [ 2 8 8 17 1 0 0 34 6 1 4 2 5 0 33 3 3 0 926 3 12] + [ 0 0 3 0 1 3 10 4 0 0 1 9 4 0 0 0 2 0 0 1111 4] + [ 128 164 177 64 122 222 50 67 88 97 122 166 507 340 211 82 262 47 126 354 4509]] + +2023-10-02 20:39:08,095 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:39:08,095 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:39:08,101 - + +2023-10-02 20:39:08,101 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:39:09,107 - Epoch: [22][ 10/ 1236] Overall Loss 0.282554 Objective Loss 0.282554 LR 0.001000 Time 0.100521 +2023-10-02 20:39:09,312 - Epoch: [22][ 20/ 1236] Overall Loss 0.285841 Objective Loss 0.285841 LR 0.001000 Time 0.060527 +2023-10-02 20:39:09,517 - Epoch: [22][ 30/ 1236] Overall Loss 0.285633 Objective Loss 0.285633 LR 0.001000 Time 0.047174 +2023-10-02 20:39:09,725 - Epoch: [22][ 40/ 1236] Overall Loss 0.290102 Objective Loss 0.290102 LR 0.001000 Time 0.040555 +2023-10-02 20:39:09,929 - Epoch: [22][ 50/ 1236] Overall Loss 0.287230 Objective Loss 0.287230 LR 0.001000 Time 0.036523 +2023-10-02 20:39:10,135 - Epoch: [22][ 60/ 1236] Overall Loss 0.284720 Objective Loss 0.284720 LR 0.001000 Time 0.033864 +2023-10-02 20:39:10,339 - Epoch: [22][ 70/ 1236] Overall Loss 0.280625 Objective Loss 0.280625 LR 0.001000 Time 0.031938 +2023-10-02 20:39:10,547 - Epoch: [22][ 80/ 1236] Overall Loss 0.278611 Objective Loss 0.278611 LR 0.001000 Time 0.030542 +2023-10-02 20:39:10,750 - Epoch: [22][ 90/ 1236] Overall Loss 0.274021 Objective Loss 0.274021 LR 0.001000 Time 0.029402 +2023-10-02 20:39:10,957 - Epoch: [22][ 100/ 1236] Overall Loss 0.272386 Objective Loss 0.272386 LR 0.001000 Time 0.028530 +2023-10-02 20:39:11,161 - Epoch: [22][ 110/ 1236] Overall Loss 0.271477 Objective Loss 0.271477 LR 0.001000 Time 0.027786 +2023-10-02 20:39:11,367 - Epoch: [22][ 120/ 1236] Overall Loss 0.269411 Objective Loss 0.269411 LR 0.001000 Time 0.027188 +2023-10-02 20:39:11,571 - Epoch: [22][ 130/ 1236] Overall Loss 0.271982 Objective Loss 0.271982 LR 0.001000 Time 0.026659 +2023-10-02 20:39:11,777 - Epoch: [22][ 140/ 1236] Overall Loss 0.272179 Objective Loss 0.272179 LR 0.001000 Time 0.026226 +2023-10-02 20:39:11,980 - Epoch: [22][ 150/ 1236] Overall Loss 0.273326 Objective Loss 0.273326 LR 0.001000 Time 0.025829 +2023-10-02 20:39:12,187 - Epoch: [22][ 160/ 1236] Overall Loss 0.274283 Objective Loss 0.274283 LR 0.001000 Time 0.025507 +2023-10-02 20:39:12,390 - Epoch: [22][ 170/ 1236] Overall Loss 0.273773 Objective Loss 0.273773 LR 0.001000 Time 0.025199 +2023-10-02 20:39:12,595 - Epoch: [22][ 180/ 1236] Overall Loss 0.273781 Objective Loss 0.273781 LR 0.001000 Time 0.024939 +2023-10-02 20:39:12,799 - Epoch: [22][ 190/ 1236] Overall Loss 0.273123 Objective Loss 0.273123 LR 0.001000 Time 0.024698 +2023-10-02 20:39:13,006 - Epoch: [22][ 200/ 1236] Overall Loss 0.273888 Objective Loss 0.273888 LR 0.001000 Time 0.024498 +2023-10-02 20:39:13,210 - Epoch: [22][ 210/ 1236] Overall Loss 0.273521 Objective Loss 0.273521 LR 0.001000 Time 0.024298 +2023-10-02 20:39:13,417 - Epoch: [22][ 220/ 1236] Overall Loss 0.275124 Objective Loss 0.275124 LR 0.001000 Time 0.024134 +2023-10-02 20:39:13,620 - Epoch: [22][ 230/ 1236] Overall Loss 0.274987 Objective Loss 0.274987 LR 0.001000 Time 0.023966 +2023-10-02 20:39:13,827 - Epoch: [22][ 240/ 1236] Overall Loss 0.274499 Objective Loss 0.274499 LR 0.001000 Time 0.023829 +2023-10-02 20:39:14,030 - Epoch: [22][ 250/ 1236] Overall Loss 0.273215 Objective Loss 0.273215 LR 0.001000 Time 0.023687 +2023-10-02 20:39:14,236 - Epoch: [22][ 260/ 1236] Overall Loss 0.272127 Objective Loss 0.272127 LR 0.001000 Time 0.023566 +2023-10-02 20:39:14,440 - Epoch: [22][ 270/ 1236] Overall Loss 0.273358 Objective Loss 0.273358 LR 0.001000 Time 0.023449 +2023-10-02 20:39:14,646 - Epoch: [22][ 280/ 1236] Overall Loss 0.273261 Objective Loss 0.273261 LR 0.001000 Time 0.023344 +2023-10-02 20:39:14,853 - Epoch: [22][ 290/ 1236] Overall Loss 0.274397 Objective Loss 0.274397 LR 0.001000 Time 0.023254 +2023-10-02 20:39:15,060 - Epoch: [22][ 300/ 1236] Overall Loss 0.273858 Objective Loss 0.273858 LR 0.001000 Time 0.023166 +2023-10-02 20:39:15,267 - Epoch: [22][ 310/ 1236] Overall Loss 0.274185 Objective Loss 0.274185 LR 0.001000 Time 0.023088 +2023-10-02 20:39:15,472 - Epoch: [22][ 320/ 1236] Overall Loss 0.275291 Objective Loss 0.275291 LR 0.001000 Time 0.023005 +2023-10-02 20:39:15,681 - Epoch: [22][ 330/ 1236] Overall Loss 0.276102 Objective Loss 0.276102 LR 0.001000 Time 0.022940 +2023-10-02 20:39:15,890 - Epoch: [22][ 340/ 1236] Overall Loss 0.277019 Objective Loss 0.277019 LR 0.001000 Time 0.022880 +2023-10-02 20:39:16,100 - Epoch: [22][ 350/ 1236] Overall Loss 0.277160 Objective Loss 0.277160 LR 0.001000 Time 0.022824 +2023-10-02 20:39:16,308 - Epoch: [22][ 360/ 1236] Overall Loss 0.277424 Objective Loss 0.277424 LR 0.001000 Time 0.022769 +2023-10-02 20:39:16,518 - Epoch: [22][ 370/ 1236] Overall Loss 0.278091 Objective Loss 0.278091 LR 0.001000 Time 0.022718 +2023-10-02 20:39:16,727 - Epoch: [22][ 380/ 1236] Overall Loss 0.279300 Objective Loss 0.279300 LR 0.001000 Time 0.022670 +2023-10-02 20:39:16,935 - Epoch: [22][ 390/ 1236] Overall Loss 0.279840 Objective Loss 0.279840 LR 0.001000 Time 0.022622 +2023-10-02 20:39:17,143 - Epoch: [22][ 400/ 1236] Overall Loss 0.280212 Objective Loss 0.280212 LR 0.001000 Time 0.022575 +2023-10-02 20:39:17,351 - Epoch: [22][ 410/ 1236] Overall Loss 0.280945 Objective Loss 0.280945 LR 0.001000 Time 0.022532 +2023-10-02 20:39:17,559 - Epoch: [22][ 420/ 1236] Overall Loss 0.281225 Objective Loss 0.281225 LR 0.001000 Time 0.022489 +2023-10-02 20:39:17,767 - Epoch: [22][ 430/ 1236] Overall Loss 0.281536 Objective Loss 0.281536 LR 0.001000 Time 0.022450 +2023-10-02 20:39:17,975 - Epoch: [22][ 440/ 1236] Overall Loss 0.281641 Objective Loss 0.281641 LR 0.001000 Time 0.022411 +2023-10-02 20:39:18,184 - Epoch: [22][ 450/ 1236] Overall Loss 0.282096 Objective Loss 0.282096 LR 0.001000 Time 0.022377 +2023-10-02 20:39:18,392 - Epoch: [22][ 460/ 1236] Overall Loss 0.282658 Objective Loss 0.282658 LR 0.001000 Time 0.022341 +2023-10-02 20:39:18,601 - Epoch: [22][ 470/ 1236] Overall Loss 0.282912 Objective Loss 0.282912 LR 0.001000 Time 0.022310 +2023-10-02 20:39:18,808 - Epoch: [22][ 480/ 1236] Overall Loss 0.283314 Objective Loss 0.283314 LR 0.001000 Time 0.022277 +2023-10-02 20:39:19,017 - Epoch: [22][ 490/ 1236] Overall Loss 0.283231 Objective Loss 0.283231 LR 0.001000 Time 0.022247 +2023-10-02 20:39:19,225 - Epoch: [22][ 500/ 1236] Overall Loss 0.282722 Objective Loss 0.282722 LR 0.001000 Time 0.022217 +2023-10-02 20:39:19,434 - Epoch: [22][ 510/ 1236] Overall Loss 0.282471 Objective Loss 0.282471 LR 0.001000 Time 0.022191 +2023-10-02 20:39:19,641 - Epoch: [22][ 520/ 1236] Overall Loss 0.282147 Objective Loss 0.282147 LR 0.001000 Time 0.022163 +2023-10-02 20:39:19,850 - Epoch: [22][ 530/ 1236] Overall Loss 0.282148 Objective Loss 0.282148 LR 0.001000 Time 0.022138 +2023-10-02 20:39:20,058 - Epoch: [22][ 540/ 1236] Overall Loss 0.281394 Objective Loss 0.281394 LR 0.001000 Time 0.022112 +2023-10-02 20:39:20,267 - Epoch: [22][ 550/ 1236] Overall Loss 0.281337 Objective Loss 0.281337 LR 0.001000 Time 0.022090 +2023-10-02 20:39:20,475 - Epoch: [22][ 560/ 1236] Overall Loss 0.281872 Objective Loss 0.281872 LR 0.001000 Time 0.022066 +2023-10-02 20:39:20,684 - Epoch: [22][ 570/ 1236] Overall Loss 0.281889 Objective Loss 0.281889 LR 0.001000 Time 0.022045 +2023-10-02 20:39:20,891 - Epoch: [22][ 580/ 1236] Overall Loss 0.281759 Objective Loss 0.281759 LR 0.001000 Time 0.022022 +2023-10-02 20:39:21,100 - Epoch: [22][ 590/ 1236] Overall Loss 0.281854 Objective Loss 0.281854 LR 0.001000 Time 0.022003 +2023-10-02 20:39:21,308 - Epoch: [22][ 600/ 1236] Overall Loss 0.282040 Objective Loss 0.282040 LR 0.001000 Time 0.021982 +2023-10-02 20:39:21,517 - Epoch: [22][ 610/ 1236] Overall Loss 0.281283 Objective Loss 0.281283 LR 0.001000 Time 0.021964 +2023-10-02 20:39:21,725 - Epoch: [22][ 620/ 1236] Overall Loss 0.281227 Objective Loss 0.281227 LR 0.001000 Time 0.021944 +2023-10-02 20:39:21,934 - Epoch: [22][ 630/ 1236] Overall Loss 0.281012 Objective Loss 0.281012 LR 0.001000 Time 0.021928 +2023-10-02 20:39:22,142 - Epoch: [22][ 640/ 1236] Overall Loss 0.281068 Objective Loss 0.281068 LR 0.001000 Time 0.021909 +2023-10-02 20:39:22,350 - Epoch: [22][ 650/ 1236] Overall Loss 0.281325 Objective Loss 0.281325 LR 0.001000 Time 0.021892 +2023-10-02 20:39:22,558 - Epoch: [22][ 660/ 1236] Overall Loss 0.281619 Objective Loss 0.281619 LR 0.001000 Time 0.021875 +2023-10-02 20:39:22,767 - Epoch: [22][ 670/ 1236] Overall Loss 0.281976 Objective Loss 0.281976 LR 0.001000 Time 0.021860 +2023-10-02 20:39:22,975 - Epoch: [22][ 680/ 1236] Overall Loss 0.282258 Objective Loss 0.282258 LR 0.001000 Time 0.021844 +2023-10-02 20:39:23,184 - Epoch: [22][ 690/ 1236] Overall Loss 0.282114 Objective Loss 0.282114 LR 0.001000 Time 0.021830 +2023-10-02 20:39:23,392 - Epoch: [22][ 700/ 1236] Overall Loss 0.282422 Objective Loss 0.282422 LR 0.001000 Time 0.021814 +2023-10-02 20:39:23,600 - Epoch: [22][ 710/ 1236] Overall Loss 0.283044 Objective Loss 0.283044 LR 0.001000 Time 0.021800 +2023-10-02 20:39:23,808 - Epoch: [22][ 720/ 1236] Overall Loss 0.283296 Objective Loss 0.283296 LR 0.001000 Time 0.021786 +2023-10-02 20:39:24,017 - Epoch: [22][ 730/ 1236] Overall Loss 0.283545 Objective Loss 0.283545 LR 0.001000 Time 0.021773 +2023-10-02 20:39:24,224 - Epoch: [22][ 740/ 1236] Overall Loss 0.283652 Objective Loss 0.283652 LR 0.001000 Time 0.021758 +2023-10-02 20:39:24,434 - Epoch: [22][ 750/ 1236] Overall Loss 0.284117 Objective Loss 0.284117 LR 0.001000 Time 0.021747 +2023-10-02 20:39:24,641 - Epoch: [22][ 760/ 1236] Overall Loss 0.283877 Objective Loss 0.283877 LR 0.001000 Time 0.021733 +2023-10-02 20:39:24,850 - Epoch: [22][ 770/ 1236] Overall Loss 0.284055 Objective Loss 0.284055 LR 0.001000 Time 0.021722 +2023-10-02 20:39:25,057 - Epoch: [22][ 780/ 1236] Overall Loss 0.284026 Objective Loss 0.284026 LR 0.001000 Time 0.021709 +2023-10-02 20:39:25,266 - Epoch: [22][ 790/ 1236] Overall Loss 0.283852 Objective Loss 0.283852 LR 0.001000 Time 0.021698 +2023-10-02 20:39:25,474 - Epoch: [22][ 800/ 1236] Overall Loss 0.284144 Objective Loss 0.284144 LR 0.001000 Time 0.021686 +2023-10-02 20:39:25,683 - Epoch: [22][ 810/ 1236] Overall Loss 0.284437 Objective Loss 0.284437 LR 0.001000 Time 0.021676 +2023-10-02 20:39:25,891 - Epoch: [22][ 820/ 1236] Overall Loss 0.284625 Objective Loss 0.284625 LR 0.001000 Time 0.021665 +2023-10-02 20:39:26,099 - Epoch: [22][ 830/ 1236] Overall Loss 0.284651 Objective Loss 0.284651 LR 0.001000 Time 0.021655 +2023-10-02 20:39:26,307 - Epoch: [22][ 840/ 1236] Overall Loss 0.284510 Objective Loss 0.284510 LR 0.001000 Time 0.021644 +2023-10-02 20:39:26,516 - Epoch: [22][ 850/ 1236] Overall Loss 0.284342 Objective Loss 0.284342 LR 0.001000 Time 0.021635 +2023-10-02 20:39:26,724 - Epoch: [22][ 860/ 1236] Overall Loss 0.284427 Objective Loss 0.284427 LR 0.001000 Time 0.021624 +2023-10-02 20:39:26,933 - Epoch: [22][ 870/ 1236] Overall Loss 0.284201 Objective Loss 0.284201 LR 0.001000 Time 0.021615 +2023-10-02 20:39:27,141 - Epoch: [22][ 880/ 1236] Overall Loss 0.284244 Objective Loss 0.284244 LR 0.001000 Time 0.021606 +2023-10-02 20:39:27,349 - Epoch: [22][ 890/ 1236] Overall Loss 0.284585 Objective Loss 0.284585 LR 0.001000 Time 0.021597 +2023-10-02 20:39:27,558 - Epoch: [22][ 900/ 1236] Overall Loss 0.284900 Objective Loss 0.284900 LR 0.001000 Time 0.021589 +2023-10-02 20:39:27,766 - Epoch: [22][ 910/ 1236] Overall Loss 0.285124 Objective Loss 0.285124 LR 0.001000 Time 0.021579 +2023-10-02 20:39:27,973 - Epoch: [22][ 920/ 1236] Overall Loss 0.285280 Objective Loss 0.285280 LR 0.001000 Time 0.021570 +2023-10-02 20:39:28,182 - Epoch: [22][ 930/ 1236] Overall Loss 0.285558 Objective Loss 0.285558 LR 0.001000 Time 0.021563 +2023-10-02 20:39:28,390 - Epoch: [22][ 940/ 1236] Overall Loss 0.285806 Objective Loss 0.285806 LR 0.001000 Time 0.021554 +2023-10-02 20:39:28,599 - Epoch: [22][ 950/ 1236] Overall Loss 0.285797 Objective Loss 0.285797 LR 0.001000 Time 0.021547 +2023-10-02 20:39:28,807 - Epoch: [22][ 960/ 1236] Overall Loss 0.286113 Objective Loss 0.286113 LR 0.001000 Time 0.021538 +2023-10-02 20:39:29,015 - Epoch: [22][ 970/ 1236] Overall Loss 0.286276 Objective Loss 0.286276 LR 0.001000 Time 0.021531 +2023-10-02 20:39:29,223 - Epoch: [22][ 980/ 1236] Overall Loss 0.286122 Objective Loss 0.286122 LR 0.001000 Time 0.021523 +2023-10-02 20:39:29,432 - Epoch: [22][ 990/ 1236] Overall Loss 0.286103 Objective Loss 0.286103 LR 0.001000 Time 0.021516 +2023-10-02 20:39:29,640 - Epoch: [22][ 1000/ 1236] Overall Loss 0.286345 Objective Loss 0.286345 LR 0.001000 Time 0.021509 +2023-10-02 20:39:29,849 - Epoch: [22][ 1010/ 1236] Overall Loss 0.286613 Objective Loss 0.286613 LR 0.001000 Time 0.021503 +2023-10-02 20:39:30,057 - Epoch: [22][ 1020/ 1236] Overall Loss 0.286821 Objective Loss 0.286821 LR 0.001000 Time 0.021496 +2023-10-02 20:39:30,266 - Epoch: [22][ 1030/ 1236] Overall Loss 0.287026 Objective Loss 0.287026 LR 0.001000 Time 0.021489 +2023-10-02 20:39:30,474 - Epoch: [22][ 1040/ 1236] Overall Loss 0.286853 Objective Loss 0.286853 LR 0.001000 Time 0.021483 +2023-10-02 20:39:30,683 - Epoch: [22][ 1050/ 1236] Overall Loss 0.286961 Objective Loss 0.286961 LR 0.001000 Time 0.021477 +2023-10-02 20:39:30,891 - Epoch: [22][ 1060/ 1236] Overall Loss 0.287255 Objective Loss 0.287255 LR 0.001000 Time 0.021470 +2023-10-02 20:39:31,099 - Epoch: [22][ 1070/ 1236] Overall Loss 0.287127 Objective Loss 0.287127 LR 0.001000 Time 0.021464 +2023-10-02 20:39:31,308 - Epoch: [22][ 1080/ 1236] Overall Loss 0.286903 Objective Loss 0.286903 LR 0.001000 Time 0.021458 +2023-10-02 20:39:31,517 - Epoch: [22][ 1090/ 1236] Overall Loss 0.286982 Objective Loss 0.286982 LR 0.001000 Time 0.021453 +2023-10-02 20:39:31,725 - Epoch: [22][ 1100/ 1236] Overall Loss 0.287267 Objective Loss 0.287267 LR 0.001000 Time 0.021446 +2023-10-02 20:39:31,935 - Epoch: [22][ 1110/ 1236] Overall Loss 0.287223 Objective Loss 0.287223 LR 0.001000 Time 0.021442 +2023-10-02 20:39:32,143 - Epoch: [22][ 1120/ 1236] Overall Loss 0.287039 Objective Loss 0.287039 LR 0.001000 Time 0.021436 +2023-10-02 20:39:32,352 - Epoch: [22][ 1130/ 1236] Overall Loss 0.287255 Objective Loss 0.287255 LR 0.001000 Time 0.021431 +2023-10-02 20:39:32,561 - Epoch: [22][ 1140/ 1236] Overall Loss 0.287053 Objective Loss 0.287053 LR 0.001000 Time 0.021426 +2023-10-02 20:39:32,770 - Epoch: [22][ 1150/ 1236] Overall Loss 0.287159 Objective Loss 0.287159 LR 0.001000 Time 0.021421 +2023-10-02 20:39:32,979 - Epoch: [22][ 1160/ 1236] Overall Loss 0.287018 Objective Loss 0.287018 LR 0.001000 Time 0.021416 +2023-10-02 20:39:33,188 - Epoch: [22][ 1170/ 1236] Overall Loss 0.287110 Objective Loss 0.287110 LR 0.001000 Time 0.021412 +2023-10-02 20:39:33,396 - Epoch: [22][ 1180/ 1236] Overall Loss 0.287106 Objective Loss 0.287106 LR 0.001000 Time 0.021407 +2023-10-02 20:39:33,605 - Epoch: [22][ 1190/ 1236] Overall Loss 0.287037 Objective Loss 0.287037 LR 0.001000 Time 0.021402 +2023-10-02 20:39:33,814 - Epoch: [22][ 1200/ 1236] Overall Loss 0.286972 Objective Loss 0.286972 LR 0.001000 Time 0.021397 +2023-10-02 20:39:34,023 - Epoch: [22][ 1210/ 1236] Overall Loss 0.287086 Objective Loss 0.287086 LR 0.001000 Time 0.021393 +2023-10-02 20:39:34,232 - Epoch: [22][ 1220/ 1236] Overall Loss 0.287008 Objective Loss 0.287008 LR 0.001000 Time 0.021389 +2023-10-02 20:39:34,491 - Epoch: [22][ 1230/ 1236] Overall Loss 0.287328 Objective Loss 0.287328 LR 0.001000 Time 0.021425 +2023-10-02 20:39:34,612 - Epoch: [22][ 1236/ 1236] Overall Loss 0.287340 Objective Loss 0.287340 Top1 84.114053 Top5 97.352342 LR 0.001000 Time 0.021419 +2023-10-02 20:39:34,747 - --- validate (epoch=22)----------- +2023-10-02 20:39:34,747 - 29943 samples (256 per mini-batch) +2023-10-02 20:39:35,233 - Epoch: [22][ 10/ 117] Loss 0.374025 Top1 81.132812 Top5 97.773438 +2023-10-02 20:39:35,382 - Epoch: [22][ 20/ 117] Loss 0.358134 Top1 81.953125 Top5 97.500000 +2023-10-02 20:39:35,529 - Epoch: [22][ 30/ 117] Loss 0.362646 Top1 81.809896 Top5 97.656250 +2023-10-02 20:39:35,677 - Epoch: [22][ 40/ 117] Loss 0.357897 Top1 81.777344 Top5 97.607422 +2023-10-02 20:39:35,823 - Epoch: [22][ 50/ 117] Loss 0.364775 Top1 81.445312 Top5 97.570312 +2023-10-02 20:39:35,969 - Epoch: [22][ 60/ 117] Loss 0.362856 Top1 81.406250 Top5 97.597656 +2023-10-02 20:39:36,115 - Epoch: [22][ 70/ 117] Loss 0.358270 Top1 81.612723 Top5 97.617188 +2023-10-02 20:39:36,262 - Epoch: [22][ 80/ 117] Loss 0.359800 Top1 81.660156 Top5 97.558594 +2023-10-02 20:39:36,408 - Epoch: [22][ 90/ 117] Loss 0.358129 Top1 81.636285 Top5 97.547743 +2023-10-02 20:39:36,554 - Epoch: [22][ 100/ 117] Loss 0.355500 Top1 81.593750 Top5 97.605469 +2023-10-02 20:39:36,709 - Epoch: [22][ 110/ 117] Loss 0.357833 Top1 81.495028 Top5 97.638494 +2023-10-02 20:39:36,798 - Epoch: [22][ 117/ 117] Loss 0.358870 Top1 81.474802 Top5 97.635507 +2023-10-02 20:39:36,927 - ==> Top1: 81.475 Top5: 97.636 Loss: 0.359 + +2023-10-02 20:39:36,928 - ==> Confusion: +[[ 967 1 4 0 3 1 0 0 4 43 0 0 0 3 2 3 4 2 0 0 13] + [ 1 1004 4 1 16 25 1 18 5 0 1 0 0 0 5 6 14 0 11 3 16] + [ 14 0 973 12 1 0 20 1 0 1 3 0 5 7 0 3 1 3 0 1 11] + [ 1 0 18 978 1 0 4 2 9 0 4 0 2 9 29 1 3 2 7 0 19] + [ 49 4 5 0 943 2 0 0 0 2 0 0 2 5 12 5 15 2 0 1 3] + [ 11 30 4 8 2 953 0 12 5 11 5 6 5 28 5 1 6 1 2 8 13] + [ 1 2 47 1 0 0 1109 1 0 0 2 1 1 2 0 5 0 1 1 9 8] + [ 12 18 37 1 11 39 7 980 3 1 8 10 3 3 3 1 2 1 47 21 10] + [ 23 1 1 0 1 0 0 0 979 36 10 1 1 10 11 3 3 8 1 0 0] + [ 185 0 1 0 7 0 0 0 27 866 1 1 2 12 4 0 3 0 0 3 7] + [ 7 1 13 10 2 1 1 1 30 0 945 3 0 24 3 2 0 2 2 0 6] + [ 2 1 4 0 3 12 0 2 0 1 0 894 49 19 0 6 3 24 1 6 8] + [ 2 0 3 3 3 1 0 0 1 0 2 25 958 5 7 9 1 30 2 5 11] + [ 3 0 1 0 2 2 0 0 7 11 7 1 0 1064 6 0 3 2 0 3 7] + [ 18 0 4 20 3 0 0 0 27 4 1 0 1 2 1006 0 2 2 3 0 8] + [ 0 0 2 4 5 0 0 0 1 1 0 2 8 3 0 1056 18 24 1 3 6] + [ 3 9 1 2 3 1 1 0 2 0 0 4 1 1 3 8 1108 0 0 2 12] + [ 0 0 1 5 0 0 2 0 0 1 1 0 17 2 3 3 0 999 0 1 3] + [ 5 3 11 35 0 1 1 9 13 0 11 0 0 0 16 0 2 1 950 0 10] + [ 0 1 4 2 1 2 7 4 1 0 3 16 6 7 0 5 7 1 3 1079 3] + [ 275 122 212 86 123 112 42 62 124 139 204 121 400 388 188 65 222 75 118 242 4585]] + +2023-10-02 20:39:36,929 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:39:36,929 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:39:36,935 - + +2023-10-02 20:39:36,935 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:39:37,926 - Epoch: [23][ 10/ 1236] Overall Loss 0.260148 Objective Loss 0.260148 LR 0.001000 Time 0.099012 +2023-10-02 20:39:38,132 - Epoch: [23][ 20/ 1236] Overall Loss 0.257514 Objective Loss 0.257514 LR 0.001000 Time 0.059799 +2023-10-02 20:39:38,338 - Epoch: [23][ 30/ 1236] Overall Loss 0.273796 Objective Loss 0.273796 LR 0.001000 Time 0.046716 +2023-10-02 20:39:38,545 - Epoch: [23][ 40/ 1236] Overall Loss 0.279170 Objective Loss 0.279170 LR 0.001000 Time 0.040215 +2023-10-02 20:39:38,750 - Epoch: [23][ 50/ 1236] Overall Loss 0.282984 Objective Loss 0.282984 LR 0.001000 Time 0.036258 +2023-10-02 20:39:38,957 - Epoch: [23][ 60/ 1236] Overall Loss 0.284196 Objective Loss 0.284196 LR 0.001000 Time 0.033670 +2023-10-02 20:39:39,162 - Epoch: [23][ 70/ 1236] Overall Loss 0.283173 Objective Loss 0.283173 LR 0.001000 Time 0.031776 +2023-10-02 20:39:39,369 - Epoch: [23][ 80/ 1236] Overall Loss 0.284924 Objective Loss 0.284924 LR 0.001000 Time 0.030393 +2023-10-02 20:39:39,574 - Epoch: [23][ 90/ 1236] Overall Loss 0.281523 Objective Loss 0.281523 LR 0.001000 Time 0.029286 +2023-10-02 20:39:39,781 - Epoch: [23][ 100/ 1236] Overall Loss 0.280143 Objective Loss 0.280143 LR 0.001000 Time 0.028427 +2023-10-02 20:39:39,985 - Epoch: [23][ 110/ 1236] Overall Loss 0.276863 Objective Loss 0.276863 LR 0.001000 Time 0.027697 +2023-10-02 20:39:40,193 - Epoch: [23][ 120/ 1236] Overall Loss 0.278565 Objective Loss 0.278565 LR 0.001000 Time 0.027115 +2023-10-02 20:39:40,397 - Epoch: [23][ 130/ 1236] Overall Loss 0.280182 Objective Loss 0.280182 LR 0.001000 Time 0.026598 +2023-10-02 20:39:40,605 - Epoch: [23][ 140/ 1236] Overall Loss 0.279038 Objective Loss 0.279038 LR 0.001000 Time 0.026180 +2023-10-02 20:39:40,809 - Epoch: [23][ 150/ 1236] Overall Loss 0.278770 Objective Loss 0.278770 LR 0.001000 Time 0.025795 +2023-10-02 20:39:41,017 - Epoch: [23][ 160/ 1236] Overall Loss 0.278520 Objective Loss 0.278520 LR 0.001000 Time 0.025478 +2023-10-02 20:39:41,221 - Epoch: [23][ 170/ 1236] Overall Loss 0.278619 Objective Loss 0.278619 LR 0.001000 Time 0.025181 +2023-10-02 20:39:41,429 - Epoch: [23][ 180/ 1236] Overall Loss 0.280221 Objective Loss 0.280221 LR 0.001000 Time 0.024933 +2023-10-02 20:39:41,633 - Epoch: [23][ 190/ 1236] Overall Loss 0.278169 Objective Loss 0.278169 LR 0.001000 Time 0.024695 +2023-10-02 20:39:41,841 - Epoch: [23][ 200/ 1236] Overall Loss 0.277336 Objective Loss 0.277336 LR 0.001000 Time 0.024497 +2023-10-02 20:39:42,045 - Epoch: [23][ 210/ 1236] Overall Loss 0.278365 Objective Loss 0.278365 LR 0.001000 Time 0.024302 +2023-10-02 20:39:42,252 - Epoch: [23][ 220/ 1236] Overall Loss 0.278364 Objective Loss 0.278364 LR 0.001000 Time 0.024139 +2023-10-02 20:39:42,457 - Epoch: [23][ 230/ 1236] Overall Loss 0.278110 Objective Loss 0.278110 LR 0.001000 Time 0.023977 +2023-10-02 20:39:42,665 - Epoch: [23][ 240/ 1236] Overall Loss 0.278535 Objective Loss 0.278535 LR 0.001000 Time 0.023843 +2023-10-02 20:39:42,869 - Epoch: [23][ 250/ 1236] Overall Loss 0.279040 Objective Loss 0.279040 LR 0.001000 Time 0.023705 +2023-10-02 20:39:43,077 - Epoch: [23][ 260/ 1236] Overall Loss 0.279583 Objective Loss 0.279583 LR 0.001000 Time 0.023591 +2023-10-02 20:39:43,281 - Epoch: [23][ 270/ 1236] Overall Loss 0.279969 Objective Loss 0.279969 LR 0.001000 Time 0.023473 +2023-10-02 20:39:43,489 - Epoch: [23][ 280/ 1236] Overall Loss 0.281205 Objective Loss 0.281205 LR 0.001000 Time 0.023376 +2023-10-02 20:39:43,693 - Epoch: [23][ 290/ 1236] Overall Loss 0.279623 Objective Loss 0.279623 LR 0.001000 Time 0.023273 +2023-10-02 20:39:43,901 - Epoch: [23][ 300/ 1236] Overall Loss 0.280042 Objective Loss 0.280042 LR 0.001000 Time 0.023188 +2023-10-02 20:39:44,105 - Epoch: [23][ 310/ 1236] Overall Loss 0.279514 Objective Loss 0.279514 LR 0.001000 Time 0.023099 +2023-10-02 20:39:44,313 - Epoch: [23][ 320/ 1236] Overall Loss 0.280072 Objective Loss 0.280072 LR 0.001000 Time 0.023025 +2023-10-02 20:39:44,517 - Epoch: [23][ 330/ 1236] Overall Loss 0.279805 Objective Loss 0.279805 LR 0.001000 Time 0.022947 +2023-10-02 20:39:44,725 - Epoch: [23][ 340/ 1236] Overall Loss 0.279858 Objective Loss 0.279858 LR 0.001000 Time 0.022882 +2023-10-02 20:39:44,930 - Epoch: [23][ 350/ 1236] Overall Loss 0.279918 Objective Loss 0.279918 LR 0.001000 Time 0.022812 +2023-10-02 20:39:45,138 - Epoch: [23][ 360/ 1236] Overall Loss 0.280393 Objective Loss 0.280393 LR 0.001000 Time 0.022755 +2023-10-02 20:39:45,343 - Epoch: [23][ 370/ 1236] Overall Loss 0.280316 Objective Loss 0.280316 LR 0.001000 Time 0.022693 +2023-10-02 20:39:45,549 - Epoch: [23][ 380/ 1236] Overall Loss 0.281218 Objective Loss 0.281218 LR 0.001000 Time 0.022638 +2023-10-02 20:39:45,755 - Epoch: [23][ 390/ 1236] Overall Loss 0.280823 Objective Loss 0.280823 LR 0.001000 Time 0.022586 +2023-10-02 20:39:45,962 - Epoch: [23][ 400/ 1236] Overall Loss 0.281117 Objective Loss 0.281117 LR 0.001000 Time 0.022537 +2023-10-02 20:39:46,168 - Epoch: [23][ 410/ 1236] Overall Loss 0.280544 Objective Loss 0.280544 LR 0.001000 Time 0.022489 +2023-10-02 20:39:46,374 - Epoch: [23][ 420/ 1236] Overall Loss 0.280086 Objective Loss 0.280086 LR 0.001000 Time 0.022444 +2023-10-02 20:39:46,580 - Epoch: [23][ 430/ 1236] Overall Loss 0.279776 Objective Loss 0.279776 LR 0.001000 Time 0.022401 +2023-10-02 20:39:46,787 - Epoch: [23][ 440/ 1236] Overall Loss 0.279680 Objective Loss 0.279680 LR 0.001000 Time 0.022361 +2023-10-02 20:39:46,993 - Epoch: [23][ 450/ 1236] Overall Loss 0.279656 Objective Loss 0.279656 LR 0.001000 Time 0.022321 +2023-10-02 20:39:47,200 - Epoch: [23][ 460/ 1236] Overall Loss 0.279390 Objective Loss 0.279390 LR 0.001000 Time 0.022284 +2023-10-02 20:39:47,406 - Epoch: [23][ 470/ 1236] Overall Loss 0.278907 Objective Loss 0.278907 LR 0.001000 Time 0.022248 +2023-10-02 20:39:47,613 - Epoch: [23][ 480/ 1236] Overall Loss 0.279106 Objective Loss 0.279106 LR 0.001000 Time 0.022217 +2023-10-02 20:39:47,818 - Epoch: [23][ 490/ 1236] Overall Loss 0.278924 Objective Loss 0.278924 LR 0.001000 Time 0.022181 +2023-10-02 20:39:48,025 - Epoch: [23][ 500/ 1236] Overall Loss 0.278379 Objective Loss 0.278379 LR 0.001000 Time 0.022150 +2023-10-02 20:39:48,231 - Epoch: [23][ 510/ 1236] Overall Loss 0.279253 Objective Loss 0.279253 LR 0.001000 Time 0.022119 +2023-10-02 20:39:48,437 - Epoch: [23][ 520/ 1236] Overall Loss 0.278873 Objective Loss 0.278873 LR 0.001000 Time 0.022090 +2023-10-02 20:39:48,644 - Epoch: [23][ 530/ 1236] Overall Loss 0.279477 Objective Loss 0.279477 LR 0.001000 Time 0.022062 +2023-10-02 20:39:48,850 - Epoch: [23][ 540/ 1236] Overall Loss 0.280453 Objective Loss 0.280453 LR 0.001000 Time 0.022036 +2023-10-02 20:39:49,057 - Epoch: [23][ 550/ 1236] Overall Loss 0.280796 Objective Loss 0.280796 LR 0.001000 Time 0.022010 +2023-10-02 20:39:49,263 - Epoch: [23][ 560/ 1236] Overall Loss 0.280974 Objective Loss 0.280974 LR 0.001000 Time 0.021985 +2023-10-02 20:39:49,469 - Epoch: [23][ 570/ 1236] Overall Loss 0.281215 Objective Loss 0.281215 LR 0.001000 Time 0.021960 +2023-10-02 20:39:49,676 - Epoch: [23][ 580/ 1236] Overall Loss 0.281172 Objective Loss 0.281172 LR 0.001000 Time 0.021938 +2023-10-02 20:39:49,882 - Epoch: [23][ 590/ 1236] Overall Loss 0.281014 Objective Loss 0.281014 LR 0.001000 Time 0.021915 +2023-10-02 20:39:50,089 - Epoch: [23][ 600/ 1236] Overall Loss 0.280966 Objective Loss 0.280966 LR 0.001000 Time 0.021894 +2023-10-02 20:39:50,295 - Epoch: [23][ 610/ 1236] Overall Loss 0.281422 Objective Loss 0.281422 LR 0.001000 Time 0.021872 +2023-10-02 20:39:50,502 - Epoch: [23][ 620/ 1236] Overall Loss 0.281584 Objective Loss 0.281584 LR 0.001000 Time 0.021852 +2023-10-02 20:39:50,709 - Epoch: [23][ 630/ 1236] Overall Loss 0.281614 Objective Loss 0.281614 LR 0.001000 Time 0.021833 +2023-10-02 20:39:50,917 - Epoch: [23][ 640/ 1236] Overall Loss 0.282253 Objective Loss 0.282253 LR 0.001000 Time 0.021817 +2023-10-02 20:39:51,120 - Epoch: [23][ 650/ 1236] Overall Loss 0.282324 Objective Loss 0.282324 LR 0.001000 Time 0.021794 +2023-10-02 20:39:51,327 - Epoch: [23][ 660/ 1236] Overall Loss 0.282233 Objective Loss 0.282233 LR 0.001000 Time 0.021776 +2023-10-02 20:39:51,533 - Epoch: [23][ 670/ 1236] Overall Loss 0.282181 Objective Loss 0.282181 LR 0.001000 Time 0.021758 +2023-10-02 20:39:51,741 - Epoch: [23][ 680/ 1236] Overall Loss 0.282597 Objective Loss 0.282597 LR 0.001000 Time 0.021744 +2023-10-02 20:39:51,946 - Epoch: [23][ 690/ 1236] Overall Loss 0.282534 Objective Loss 0.282534 LR 0.001000 Time 0.021725 +2023-10-02 20:39:52,153 - Epoch: [23][ 700/ 1236] Overall Loss 0.282677 Objective Loss 0.282677 LR 0.001000 Time 0.021710 +2023-10-02 20:39:52,359 - Epoch: [23][ 710/ 1236] Overall Loss 0.282249 Objective Loss 0.282249 LR 0.001000 Time 0.021694 +2023-10-02 20:39:52,565 - Epoch: [23][ 720/ 1236] Overall Loss 0.282474 Objective Loss 0.282474 LR 0.001000 Time 0.021679 +2023-10-02 20:39:52,772 - Epoch: [23][ 730/ 1236] Overall Loss 0.282567 Objective Loss 0.282567 LR 0.001000 Time 0.021664 +2023-10-02 20:39:52,978 - Epoch: [23][ 740/ 1236] Overall Loss 0.282869 Objective Loss 0.282869 LR 0.001000 Time 0.021651 +2023-10-02 20:39:53,184 - Epoch: [23][ 750/ 1236] Overall Loss 0.283222 Objective Loss 0.283222 LR 0.001000 Time 0.021636 +2023-10-02 20:39:53,391 - Epoch: [23][ 760/ 1236] Overall Loss 0.282789 Objective Loss 0.282789 LR 0.001000 Time 0.021623 +2023-10-02 20:39:53,597 - Epoch: [23][ 770/ 1236] Overall Loss 0.283410 Objective Loss 0.283410 LR 0.001000 Time 0.021610 +2023-10-02 20:39:53,804 - Epoch: [23][ 780/ 1236] Overall Loss 0.283431 Objective Loss 0.283431 LR 0.001000 Time 0.021598 +2023-10-02 20:39:54,010 - Epoch: [23][ 790/ 1236] Overall Loss 0.283656 Objective Loss 0.283656 LR 0.001000 Time 0.021585 +2023-10-02 20:39:54,217 - Epoch: [23][ 800/ 1236] Overall Loss 0.283816 Objective Loss 0.283816 LR 0.001000 Time 0.021573 +2023-10-02 20:39:54,423 - Epoch: [23][ 810/ 1236] Overall Loss 0.283972 Objective Loss 0.283972 LR 0.001000 Time 0.021561 +2023-10-02 20:39:54,631 - Epoch: [23][ 820/ 1236] Overall Loss 0.283676 Objective Loss 0.283676 LR 0.001000 Time 0.021551 +2023-10-02 20:39:54,836 - Epoch: [23][ 830/ 1236] Overall Loss 0.284264 Objective Loss 0.284264 LR 0.001000 Time 0.021538 +2023-10-02 20:39:55,044 - Epoch: [23][ 840/ 1236] Overall Loss 0.284441 Objective Loss 0.284441 LR 0.001000 Time 0.021529 +2023-10-02 20:39:55,249 - Epoch: [23][ 850/ 1236] Overall Loss 0.284367 Objective Loss 0.284367 LR 0.001000 Time 0.021517 +2023-10-02 20:39:55,456 - Epoch: [23][ 860/ 1236] Overall Loss 0.284605 Objective Loss 0.284605 LR 0.001000 Time 0.021506 +2023-10-02 20:39:55,662 - Epoch: [23][ 870/ 1236] Overall Loss 0.284775 Objective Loss 0.284775 LR 0.001000 Time 0.021496 +2023-10-02 20:39:55,868 - Epoch: [23][ 880/ 1236] Overall Loss 0.284959 Objective Loss 0.284959 LR 0.001000 Time 0.021486 +2023-10-02 20:39:56,074 - Epoch: [23][ 890/ 1236] Overall Loss 0.284707 Objective Loss 0.284707 LR 0.001000 Time 0.021476 +2023-10-02 20:39:56,282 - Epoch: [23][ 900/ 1236] Overall Loss 0.284262 Objective Loss 0.284262 LR 0.001000 Time 0.021467 +2023-10-02 20:39:56,487 - Epoch: [23][ 910/ 1236] Overall Loss 0.284104 Objective Loss 0.284104 LR 0.001000 Time 0.021456 +2023-10-02 20:39:56,695 - Epoch: [23][ 920/ 1236] Overall Loss 0.283998 Objective Loss 0.283998 LR 0.001000 Time 0.021449 +2023-10-02 20:39:56,900 - Epoch: [23][ 930/ 1236] Overall Loss 0.284164 Objective Loss 0.284164 LR 0.001000 Time 0.021438 +2023-10-02 20:39:57,107 - Epoch: [23][ 940/ 1236] Overall Loss 0.283968 Objective Loss 0.283968 LR 0.001000 Time 0.021430 +2023-10-02 20:39:57,313 - Epoch: [23][ 950/ 1236] Overall Loss 0.283998 Objective Loss 0.283998 LR 0.001000 Time 0.021420 +2023-10-02 20:39:57,520 - Epoch: [23][ 960/ 1236] Overall Loss 0.284115 Objective Loss 0.284115 LR 0.001000 Time 0.021412 +2023-10-02 20:39:57,726 - Epoch: [23][ 970/ 1236] Overall Loss 0.284253 Objective Loss 0.284253 LR 0.001000 Time 0.021403 +2023-10-02 20:39:57,932 - Epoch: [23][ 980/ 1236] Overall Loss 0.284105 Objective Loss 0.284105 LR 0.001000 Time 0.021395 +2023-10-02 20:39:58,139 - Epoch: [23][ 990/ 1236] Overall Loss 0.284063 Objective Loss 0.284063 LR 0.001000 Time 0.021387 +2023-10-02 20:39:58,345 - Epoch: [23][ 1000/ 1236] Overall Loss 0.284036 Objective Loss 0.284036 LR 0.001000 Time 0.021380 +2023-10-02 20:39:58,551 - Epoch: [23][ 1010/ 1236] Overall Loss 0.284081 Objective Loss 0.284081 LR 0.001000 Time 0.021372 +2023-10-02 20:39:58,758 - Epoch: [23][ 1020/ 1236] Overall Loss 0.283969 Objective Loss 0.283969 LR 0.001000 Time 0.021365 +2023-10-02 20:39:58,964 - Epoch: [23][ 1030/ 1236] Overall Loss 0.284391 Objective Loss 0.284391 LR 0.001000 Time 0.021357 +2023-10-02 20:39:59,171 - Epoch: [23][ 1040/ 1236] Overall Loss 0.284090 Objective Loss 0.284090 LR 0.001000 Time 0.021350 +2023-10-02 20:39:59,377 - Epoch: [23][ 1050/ 1236] Overall Loss 0.284226 Objective Loss 0.284226 LR 0.001000 Time 0.021343 +2023-10-02 20:39:59,584 - Epoch: [23][ 1060/ 1236] Overall Loss 0.283970 Objective Loss 0.283970 LR 0.001000 Time 0.021336 +2023-10-02 20:39:59,790 - Epoch: [23][ 1070/ 1236] Overall Loss 0.283930 Objective Loss 0.283930 LR 0.001000 Time 0.021329 +2023-10-02 20:39:59,997 - Epoch: [23][ 1080/ 1236] Overall Loss 0.283950 Objective Loss 0.283950 LR 0.001000 Time 0.021323 +2023-10-02 20:40:00,203 - Epoch: [23][ 1090/ 1236] Overall Loss 0.283987 Objective Loss 0.283987 LR 0.001000 Time 0.021316 +2023-10-02 20:40:00,410 - Epoch: [23][ 1100/ 1236] Overall Loss 0.283924 Objective Loss 0.283924 LR 0.001000 Time 0.021310 +2023-10-02 20:40:00,616 - Epoch: [23][ 1110/ 1236] Overall Loss 0.283920 Objective Loss 0.283920 LR 0.001000 Time 0.021304 +2023-10-02 20:40:00,823 - Epoch: [23][ 1120/ 1236] Overall Loss 0.283811 Objective Loss 0.283811 LR 0.001000 Time 0.021298 +2023-10-02 20:40:01,029 - Epoch: [23][ 1130/ 1236] Overall Loss 0.283818 Objective Loss 0.283818 LR 0.001000 Time 0.021292 +2023-10-02 20:40:01,236 - Epoch: [23][ 1140/ 1236] Overall Loss 0.284017 Objective Loss 0.284017 LR 0.001000 Time 0.021286 +2023-10-02 20:40:01,442 - Epoch: [23][ 1150/ 1236] Overall Loss 0.283890 Objective Loss 0.283890 LR 0.001000 Time 0.021280 +2023-10-02 20:40:01,649 - Epoch: [23][ 1160/ 1236] Overall Loss 0.284111 Objective Loss 0.284111 LR 0.001000 Time 0.021274 +2023-10-02 20:40:01,855 - Epoch: [23][ 1170/ 1236] Overall Loss 0.283974 Objective Loss 0.283974 LR 0.001000 Time 0.021269 +2023-10-02 20:40:02,062 - Epoch: [23][ 1180/ 1236] Overall Loss 0.283823 Objective Loss 0.283823 LR 0.001000 Time 0.021263 +2023-10-02 20:40:02,268 - Epoch: [23][ 1190/ 1236] Overall Loss 0.283814 Objective Loss 0.283814 LR 0.001000 Time 0.021258 +2023-10-02 20:40:02,475 - Epoch: [23][ 1200/ 1236] Overall Loss 0.283811 Objective Loss 0.283811 LR 0.001000 Time 0.021253 +2023-10-02 20:40:02,681 - Epoch: [23][ 1210/ 1236] Overall Loss 0.283823 Objective Loss 0.283823 LR 0.001000 Time 0.021247 +2023-10-02 20:40:02,889 - Epoch: [23][ 1220/ 1236] Overall Loss 0.283541 Objective Loss 0.283541 LR 0.001000 Time 0.021243 +2023-10-02 20:40:03,146 - Epoch: [23][ 1230/ 1236] Overall Loss 0.283695 Objective Loss 0.283695 LR 0.001000 Time 0.021280 +2023-10-02 20:40:03,268 - Epoch: [23][ 1236/ 1236] Overall Loss 0.283792 Objective Loss 0.283792 Top1 82.484725 Top5 96.945010 LR 0.001000 Time 0.021275 +2023-10-02 20:40:03,389 - --- validate (epoch=23)----------- +2023-10-02 20:40:03,389 - 29943 samples (256 per mini-batch) +2023-10-02 20:40:03,870 - Epoch: [23][ 10/ 117] Loss 0.348734 Top1 81.835938 Top5 97.539062 +2023-10-02 20:40:04,020 - Epoch: [23][ 20/ 117] Loss 0.344549 Top1 81.621094 Top5 97.441406 +2023-10-02 20:40:04,169 - Epoch: [23][ 30/ 117] Loss 0.334682 Top1 81.692708 Top5 97.591146 +2023-10-02 20:40:04,319 - Epoch: [23][ 40/ 117] Loss 0.352337 Top1 81.542969 Top5 97.607422 +2023-10-02 20:40:04,468 - Epoch: [23][ 50/ 117] Loss 0.348806 Top1 81.664062 Top5 97.640625 +2023-10-02 20:40:04,619 - Epoch: [23][ 60/ 117] Loss 0.352411 Top1 81.438802 Top5 97.649740 +2023-10-02 20:40:04,770 - Epoch: [23][ 70/ 117] Loss 0.349556 Top1 81.556920 Top5 97.712054 +2023-10-02 20:40:04,920 - Epoch: [23][ 80/ 117] Loss 0.346600 Top1 81.606445 Top5 97.724609 +2023-10-02 20:40:05,069 - Epoch: [23][ 90/ 117] Loss 0.346543 Top1 81.697049 Top5 97.703993 +2023-10-02 20:40:05,218 - Epoch: [23][ 100/ 117] Loss 0.347993 Top1 81.625000 Top5 97.714844 +2023-10-02 20:40:05,374 - Epoch: [23][ 110/ 117] Loss 0.344996 Top1 81.740057 Top5 97.766335 +2023-10-02 20:40:05,463 - Epoch: [23][ 117/ 117] Loss 0.344016 Top1 81.805430 Top5 97.782453 +2023-10-02 20:40:05,603 - ==> Top1: 81.805 Top5: 97.782 Loss: 0.344 + +2023-10-02 20:40:05,604 - ==> Confusion: +[[ 963 0 3 1 4 3 0 0 7 38 1 2 1 3 1 0 2 4 0 0 17] + [ 0 1047 3 0 15 10 2 21 2 2 3 1 0 0 3 3 2 0 13 0 4] + [ 8 0 983 13 3 0 19 3 0 0 3 1 5 1 0 1 0 3 7 4 2] + [ 4 3 19 987 1 1 2 1 5 1 4 0 1 1 22 2 0 7 15 1 12] + [ 38 7 4 0 959 3 0 0 2 2 0 0 0 1 14 4 10 0 0 2 4] + [ 5 72 3 1 1 934 0 42 5 1 5 6 5 6 6 2 2 1 5 4 10] + [ 0 2 47 1 1 0 1107 4 0 0 1 1 1 1 0 7 0 1 0 12 5] + [ 7 20 31 1 3 13 2 1039 3 0 7 5 2 1 5 1 1 2 61 7 7] + [ 26 1 1 0 2 0 0 1 996 27 6 0 1 6 12 1 0 4 1 0 4] + [ 189 1 0 0 8 1 1 1 47 836 0 0 0 12 6 2 1 2 1 3 8] + [ 5 5 11 12 1 0 1 1 30 0 954 2 0 9 5 1 0 3 6 0 7] + [ 1 4 2 0 2 16 0 6 0 1 0 927 24 5 1 4 5 21 0 10 6] + [ 1 3 4 4 3 3 0 0 5 0 1 41 936 0 3 10 1 31 0 6 16] + [ 2 1 3 0 4 11 1 1 36 10 14 4 1 993 11 1 2 2 0 8 14] + [ 14 1 1 19 2 0 0 0 35 3 3 0 1 1 1003 0 0 1 8 0 9] + [ 2 0 4 3 5 2 2 0 1 0 0 4 5 2 0 1057 14 21 2 6 4] + [ 0 18 2 0 8 2 0 0 1 0 0 3 1 0 3 11 1097 1 0 2 12] + [ 0 1 2 1 0 0 2 0 0 0 0 1 12 0 1 4 2 1010 0 1 1] + [ 4 3 8 22 0 0 0 11 13 0 4 0 0 0 9 0 1 1 983 0 9] + [ 0 3 5 2 1 2 14 13 1 0 2 6 5 0 2 5 7 2 0 1075 7] + [ 219 252 207 117 145 101 48 74 196 68 207 125 344 235 217 77 131 103 199 231 4609]] + +2023-10-02 20:40:05,605 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:40:05,605 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:40:05,611 - + +2023-10-02 20:40:05,611 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:40:06,613 - Epoch: [24][ 10/ 1236] Overall Loss 0.261094 Objective Loss 0.261094 LR 0.001000 Time 0.100111 +2023-10-02 20:40:06,819 - Epoch: [24][ 20/ 1236] Overall Loss 0.266274 Objective Loss 0.266274 LR 0.001000 Time 0.060356 +2023-10-02 20:40:07,025 - Epoch: [24][ 30/ 1236] Overall Loss 0.281352 Objective Loss 0.281352 LR 0.001000 Time 0.047080 +2023-10-02 20:40:07,232 - Epoch: [24][ 40/ 1236] Overall Loss 0.278725 Objective Loss 0.278725 LR 0.001000 Time 0.040478 +2023-10-02 20:40:07,436 - Epoch: [24][ 50/ 1236] Overall Loss 0.277556 Objective Loss 0.277556 LR 0.001000 Time 0.036458 +2023-10-02 20:40:07,644 - Epoch: [24][ 60/ 1236] Overall Loss 0.276639 Objective Loss 0.276639 LR 0.001000 Time 0.033838 +2023-10-02 20:40:07,848 - Epoch: [24][ 70/ 1236] Overall Loss 0.275742 Objective Loss 0.275742 LR 0.001000 Time 0.031917 +2023-10-02 20:40:08,055 - Epoch: [24][ 80/ 1236] Overall Loss 0.274661 Objective Loss 0.274661 LR 0.001000 Time 0.030515 +2023-10-02 20:40:08,259 - Epoch: [24][ 90/ 1236] Overall Loss 0.270587 Objective Loss 0.270587 LR 0.001000 Time 0.029391 +2023-10-02 20:40:08,467 - Epoch: [24][ 100/ 1236] Overall Loss 0.269228 Objective Loss 0.269228 LR 0.001000 Time 0.028525 +2023-10-02 20:40:08,671 - Epoch: [24][ 110/ 1236] Overall Loss 0.268726 Objective Loss 0.268726 LR 0.001000 Time 0.027786 +2023-10-02 20:40:08,879 - Epoch: [24][ 120/ 1236] Overall Loss 0.268773 Objective Loss 0.268773 LR 0.001000 Time 0.027197 +2023-10-02 20:40:09,083 - Epoch: [24][ 130/ 1236] Overall Loss 0.271197 Objective Loss 0.271197 LR 0.001000 Time 0.026674 +2023-10-02 20:40:09,290 - Epoch: [24][ 140/ 1236] Overall Loss 0.270371 Objective Loss 0.270371 LR 0.001000 Time 0.026250 +2023-10-02 20:40:09,495 - Epoch: [24][ 150/ 1236] Overall Loss 0.269160 Objective Loss 0.269160 LR 0.001000 Time 0.025860 +2023-10-02 20:40:09,702 - Epoch: [24][ 160/ 1236] Overall Loss 0.269982 Objective Loss 0.269982 LR 0.001000 Time 0.025540 +2023-10-02 20:40:09,907 - Epoch: [24][ 170/ 1236] Overall Loss 0.270239 Objective Loss 0.270239 LR 0.001000 Time 0.025241 +2023-10-02 20:40:10,115 - Epoch: [24][ 180/ 1236] Overall Loss 0.271591 Objective Loss 0.271591 LR 0.001000 Time 0.024989 +2023-10-02 20:40:10,319 - Epoch: [24][ 190/ 1236] Overall Loss 0.272238 Objective Loss 0.272238 LR 0.001000 Time 0.024747 +2023-10-02 20:40:10,527 - Epoch: [24][ 200/ 1236] Overall Loss 0.271656 Objective Loss 0.271656 LR 0.001000 Time 0.024547 +2023-10-02 20:40:10,731 - Epoch: [24][ 210/ 1236] Overall Loss 0.274602 Objective Loss 0.274602 LR 0.001000 Time 0.024350 +2023-10-02 20:40:10,938 - Epoch: [24][ 220/ 1236] Overall Loss 0.275104 Objective Loss 0.275104 LR 0.001000 Time 0.024185 +2023-10-02 20:40:11,143 - Epoch: [24][ 230/ 1236] Overall Loss 0.275671 Objective Loss 0.275671 LR 0.001000 Time 0.024021 +2023-10-02 20:40:11,349 - Epoch: [24][ 240/ 1236] Overall Loss 0.276539 Objective Loss 0.276539 LR 0.001000 Time 0.023877 +2023-10-02 20:40:11,553 - Epoch: [24][ 250/ 1236] Overall Loss 0.276199 Objective Loss 0.276199 LR 0.001000 Time 0.023738 +2023-10-02 20:40:11,761 - Epoch: [24][ 260/ 1236] Overall Loss 0.276092 Objective Loss 0.276092 LR 0.001000 Time 0.023623 +2023-10-02 20:40:11,965 - Epoch: [24][ 270/ 1236] Overall Loss 0.276260 Objective Loss 0.276260 LR 0.001000 Time 0.023503 +2023-10-02 20:40:12,173 - Epoch: [24][ 280/ 1236] Overall Loss 0.277031 Objective Loss 0.277031 LR 0.001000 Time 0.023405 +2023-10-02 20:40:12,377 - Epoch: [24][ 290/ 1236] Overall Loss 0.275845 Objective Loss 0.275845 LR 0.001000 Time 0.023300 +2023-10-02 20:40:12,583 - Epoch: [24][ 300/ 1236] Overall Loss 0.276179 Objective Loss 0.276179 LR 0.001000 Time 0.023211 +2023-10-02 20:40:12,788 - Epoch: [24][ 310/ 1236] Overall Loss 0.276880 Objective Loss 0.276880 LR 0.001000 Time 0.023121 +2023-10-02 20:40:12,996 - Epoch: [24][ 320/ 1236] Overall Loss 0.276339 Objective Loss 0.276339 LR 0.001000 Time 0.023047 +2023-10-02 20:40:13,200 - Epoch: [24][ 330/ 1236] Overall Loss 0.276505 Objective Loss 0.276505 LR 0.001000 Time 0.022966 +2023-10-02 20:40:13,408 - Epoch: [24][ 340/ 1236] Overall Loss 0.276170 Objective Loss 0.276170 LR 0.001000 Time 0.022901 +2023-10-02 20:40:13,612 - Epoch: [24][ 350/ 1236] Overall Loss 0.276596 Objective Loss 0.276596 LR 0.001000 Time 0.022830 +2023-10-02 20:40:13,820 - Epoch: [24][ 360/ 1236] Overall Loss 0.277066 Objective Loss 0.277066 LR 0.001000 Time 0.022773 +2023-10-02 20:40:14,024 - Epoch: [24][ 370/ 1236] Overall Loss 0.277075 Objective Loss 0.277075 LR 0.001000 Time 0.022709 +2023-10-02 20:40:14,232 - Epoch: [24][ 380/ 1236] Overall Loss 0.276310 Objective Loss 0.276310 LR 0.001000 Time 0.022657 +2023-10-02 20:40:14,436 - Epoch: [24][ 390/ 1236] Overall Loss 0.276063 Objective Loss 0.276063 LR 0.001000 Time 0.022599 +2023-10-02 20:40:14,644 - Epoch: [24][ 400/ 1236] Overall Loss 0.276063 Objective Loss 0.276063 LR 0.001000 Time 0.022553 +2023-10-02 20:40:14,849 - Epoch: [24][ 410/ 1236] Overall Loss 0.276141 Objective Loss 0.276141 LR 0.001000 Time 0.022502 +2023-10-02 20:40:15,057 - Epoch: [24][ 420/ 1236] Overall Loss 0.276647 Objective Loss 0.276647 LR 0.001000 Time 0.022460 +2023-10-02 20:40:15,261 - Epoch: [24][ 430/ 1236] Overall Loss 0.276508 Objective Loss 0.276508 LR 0.001000 Time 0.022412 +2023-10-02 20:40:15,469 - Epoch: [24][ 440/ 1236] Overall Loss 0.276637 Objective Loss 0.276637 LR 0.001000 Time 0.022375 +2023-10-02 20:40:15,673 - Epoch: [24][ 450/ 1236] Overall Loss 0.276276 Objective Loss 0.276276 LR 0.001000 Time 0.022331 +2023-10-02 20:40:15,881 - Epoch: [24][ 460/ 1236] Overall Loss 0.276257 Objective Loss 0.276257 LR 0.001000 Time 0.022297 +2023-10-02 20:40:16,086 - Epoch: [24][ 470/ 1236] Overall Loss 0.276116 Objective Loss 0.276116 LR 0.001000 Time 0.022257 +2023-10-02 20:40:16,293 - Epoch: [24][ 480/ 1236] Overall Loss 0.276083 Objective Loss 0.276083 LR 0.001000 Time 0.022225 +2023-10-02 20:40:16,498 - Epoch: [24][ 490/ 1236] Overall Loss 0.275549 Objective Loss 0.275549 LR 0.001000 Time 0.022189 +2023-10-02 20:40:16,706 - Epoch: [24][ 500/ 1236] Overall Loss 0.275948 Objective Loss 0.275948 LR 0.001000 Time 0.022160 +2023-10-02 20:40:16,911 - Epoch: [24][ 510/ 1236] Overall Loss 0.275900 Objective Loss 0.275900 LR 0.001000 Time 0.022127 +2023-10-02 20:40:17,119 - Epoch: [24][ 520/ 1236] Overall Loss 0.275798 Objective Loss 0.275798 LR 0.001000 Time 0.022101 +2023-10-02 20:40:17,323 - Epoch: [24][ 530/ 1236] Overall Loss 0.276198 Objective Loss 0.276198 LR 0.001000 Time 0.022069 +2023-10-02 20:40:17,530 - Epoch: [24][ 540/ 1236] Overall Loss 0.276198 Objective Loss 0.276198 LR 0.001000 Time 0.022043 +2023-10-02 20:40:17,735 - Epoch: [24][ 550/ 1236] Overall Loss 0.276312 Objective Loss 0.276312 LR 0.001000 Time 0.022013 +2023-10-02 20:40:17,943 - Epoch: [24][ 560/ 1236] Overall Loss 0.276840 Objective Loss 0.276840 LR 0.001000 Time 0.021991 +2023-10-02 20:40:18,147 - Epoch: [24][ 570/ 1236] Overall Loss 0.276723 Objective Loss 0.276723 LR 0.001000 Time 0.021964 +2023-10-02 20:40:18,355 - Epoch: [24][ 580/ 1236] Overall Loss 0.276495 Objective Loss 0.276495 LR 0.001000 Time 0.021943 +2023-10-02 20:40:18,559 - Epoch: [24][ 590/ 1236] Overall Loss 0.277096 Objective Loss 0.277096 LR 0.001000 Time 0.021917 +2023-10-02 20:40:18,766 - Epoch: [24][ 600/ 1236] Overall Loss 0.277491 Objective Loss 0.277491 LR 0.001000 Time 0.021896 +2023-10-02 20:40:18,972 - Epoch: [24][ 610/ 1236] Overall Loss 0.277266 Objective Loss 0.277266 LR 0.001000 Time 0.021874 +2023-10-02 20:40:19,179 - Epoch: [24][ 620/ 1236] Overall Loss 0.277538 Objective Loss 0.277538 LR 0.001000 Time 0.021854 +2023-10-02 20:40:19,384 - Epoch: [24][ 630/ 1236] Overall Loss 0.277564 Objective Loss 0.277564 LR 0.001000 Time 0.021833 +2023-10-02 20:40:19,592 - Epoch: [24][ 640/ 1236] Overall Loss 0.277177 Objective Loss 0.277177 LR 0.001000 Time 0.021816 +2023-10-02 20:40:19,797 - Epoch: [24][ 650/ 1236] Overall Loss 0.277558 Objective Loss 0.277558 LR 0.001000 Time 0.021795 +2023-10-02 20:40:20,005 - Epoch: [24][ 660/ 1236] Overall Loss 0.277859 Objective Loss 0.277859 LR 0.001000 Time 0.021780 +2023-10-02 20:40:20,209 - Epoch: [24][ 670/ 1236] Overall Loss 0.277922 Objective Loss 0.277922 LR 0.001000 Time 0.021759 +2023-10-02 20:40:20,417 - Epoch: [24][ 680/ 1236] Overall Loss 0.278163 Objective Loss 0.278163 LR 0.001000 Time 0.021745 +2023-10-02 20:40:20,622 - Epoch: [24][ 690/ 1236] Overall Loss 0.278235 Objective Loss 0.278235 LR 0.001000 Time 0.021726 +2023-10-02 20:40:20,830 - Epoch: [24][ 700/ 1236] Overall Loss 0.278260 Objective Loss 0.278260 LR 0.001000 Time 0.021712 +2023-10-02 20:40:21,034 - Epoch: [24][ 710/ 1236] Overall Loss 0.278710 Objective Loss 0.278710 LR 0.001000 Time 0.021694 +2023-10-02 20:40:21,241 - Epoch: [24][ 720/ 1236] Overall Loss 0.278602 Objective Loss 0.278602 LR 0.001000 Time 0.021679 +2023-10-02 20:40:21,447 - Epoch: [24][ 730/ 1236] Overall Loss 0.278808 Objective Loss 0.278808 LR 0.001000 Time 0.021662 +2023-10-02 20:40:21,655 - Epoch: [24][ 740/ 1236] Overall Loss 0.279589 Objective Loss 0.279589 LR 0.001000 Time 0.021650 +2023-10-02 20:40:21,859 - Epoch: [24][ 750/ 1236] Overall Loss 0.279846 Objective Loss 0.279846 LR 0.001000 Time 0.021634 +2023-10-02 20:40:22,067 - Epoch: [24][ 760/ 1236] Overall Loss 0.279963 Objective Loss 0.279963 LR 0.001000 Time 0.021623 +2023-10-02 20:40:22,272 - Epoch: [24][ 770/ 1236] Overall Loss 0.279602 Objective Loss 0.279602 LR 0.001000 Time 0.021607 +2023-10-02 20:40:22,479 - Epoch: [24][ 780/ 1236] Overall Loss 0.279762 Objective Loss 0.279762 LR 0.001000 Time 0.021594 +2023-10-02 20:40:22,684 - Epoch: [24][ 790/ 1236] Overall Loss 0.279857 Objective Loss 0.279857 LR 0.001000 Time 0.021581 +2023-10-02 20:40:22,892 - Epoch: [24][ 800/ 1236] Overall Loss 0.280265 Objective Loss 0.280265 LR 0.001000 Time 0.021571 +2023-10-02 20:40:23,097 - Epoch: [24][ 810/ 1236] Overall Loss 0.280714 Objective Loss 0.280714 LR 0.001000 Time 0.021557 +2023-10-02 20:40:23,305 - Epoch: [24][ 820/ 1236] Overall Loss 0.280830 Objective Loss 0.280830 LR 0.001000 Time 0.021548 +2023-10-02 20:40:23,509 - Epoch: [24][ 830/ 1236] Overall Loss 0.281064 Objective Loss 0.281064 LR 0.001000 Time 0.021534 +2023-10-02 20:40:23,718 - Epoch: [24][ 840/ 1236] Overall Loss 0.280814 Objective Loss 0.280814 LR 0.001000 Time 0.021525 +2023-10-02 20:40:23,922 - Epoch: [24][ 850/ 1236] Overall Loss 0.280863 Objective Loss 0.280863 LR 0.001000 Time 0.021512 +2023-10-02 20:40:24,130 - Epoch: [24][ 860/ 1236] Overall Loss 0.280666 Objective Loss 0.280666 LR 0.001000 Time 0.021504 +2023-10-02 20:40:24,335 - Epoch: [24][ 870/ 1236] Overall Loss 0.280625 Objective Loss 0.280625 LR 0.001000 Time 0.021491 +2023-10-02 20:40:24,543 - Epoch: [24][ 880/ 1236] Overall Loss 0.280598 Objective Loss 0.280598 LR 0.001000 Time 0.021483 +2023-10-02 20:40:24,748 - Epoch: [24][ 890/ 1236] Overall Loss 0.280803 Objective Loss 0.280803 LR 0.001000 Time 0.021472 +2023-10-02 20:40:24,954 - Epoch: [24][ 900/ 1236] Overall Loss 0.280993 Objective Loss 0.280993 LR 0.001000 Time 0.021462 +2023-10-02 20:40:25,160 - Epoch: [24][ 910/ 1236] Overall Loss 0.281230 Objective Loss 0.281230 LR 0.001000 Time 0.021452 +2023-10-02 20:40:25,368 - Epoch: [24][ 920/ 1236] Overall Loss 0.281031 Objective Loss 0.281031 LR 0.001000 Time 0.021445 +2023-10-02 20:40:25,572 - Epoch: [24][ 930/ 1236] Overall Loss 0.281439 Objective Loss 0.281439 LR 0.001000 Time 0.021434 +2023-10-02 20:40:25,781 - Epoch: [24][ 940/ 1236] Overall Loss 0.281500 Objective Loss 0.281500 LR 0.001000 Time 0.021427 +2023-10-02 20:40:25,985 - Epoch: [24][ 950/ 1236] Overall Loss 0.281758 Objective Loss 0.281758 LR 0.001000 Time 0.021417 +2023-10-02 20:40:26,194 - Epoch: [24][ 960/ 1236] Overall Loss 0.282094 Objective Loss 0.282094 LR 0.001000 Time 0.021410 +2023-10-02 20:40:26,398 - Epoch: [24][ 970/ 1236] Overall Loss 0.281870 Objective Loss 0.281870 LR 0.001000 Time 0.021400 +2023-10-02 20:40:26,606 - Epoch: [24][ 980/ 1236] Overall Loss 0.282115 Objective Loss 0.282115 LR 0.001000 Time 0.021394 +2023-10-02 20:40:26,811 - Epoch: [24][ 990/ 1236] Overall Loss 0.282112 Objective Loss 0.282112 LR 0.001000 Time 0.021384 +2023-10-02 20:40:27,019 - Epoch: [24][ 1000/ 1236] Overall Loss 0.281879 Objective Loss 0.281879 LR 0.001000 Time 0.021378 +2023-10-02 20:40:27,224 - Epoch: [24][ 1010/ 1236] Overall Loss 0.282116 Objective Loss 0.282116 LR 0.001000 Time 0.021369 +2023-10-02 20:40:27,431 - Epoch: [24][ 1020/ 1236] Overall Loss 0.282106 Objective Loss 0.282106 LR 0.001000 Time 0.021363 +2023-10-02 20:40:27,636 - Epoch: [24][ 1030/ 1236] Overall Loss 0.282122 Objective Loss 0.282122 LR 0.001000 Time 0.021353 +2023-10-02 20:40:27,844 - Epoch: [24][ 1040/ 1236] Overall Loss 0.282366 Objective Loss 0.282366 LR 0.001000 Time 0.021348 +2023-10-02 20:40:28,048 - Epoch: [24][ 1050/ 1236] Overall Loss 0.282370 Objective Loss 0.282370 LR 0.001000 Time 0.021339 +2023-10-02 20:40:28,255 - Epoch: [24][ 1060/ 1236] Overall Loss 0.282757 Objective Loss 0.282757 LR 0.001000 Time 0.021332 +2023-10-02 20:40:28,461 - Epoch: [24][ 1070/ 1236] Overall Loss 0.282805 Objective Loss 0.282805 LR 0.001000 Time 0.021324 +2023-10-02 20:40:28,669 - Epoch: [24][ 1080/ 1236] Overall Loss 0.283436 Objective Loss 0.283436 LR 0.001000 Time 0.021319 +2023-10-02 20:40:28,873 - Epoch: [24][ 1090/ 1236] Overall Loss 0.283521 Objective Loss 0.283521 LR 0.001000 Time 0.021311 +2023-10-02 20:40:29,081 - Epoch: [24][ 1100/ 1236] Overall Loss 0.283700 Objective Loss 0.283700 LR 0.001000 Time 0.021305 +2023-10-02 20:40:29,285 - Epoch: [24][ 1110/ 1236] Overall Loss 0.283984 Objective Loss 0.283984 LR 0.001000 Time 0.021297 +2023-10-02 20:40:29,493 - Epoch: [24][ 1120/ 1236] Overall Loss 0.283912 Objective Loss 0.283912 LR 0.001000 Time 0.021293 +2023-10-02 20:40:29,698 - Epoch: [24][ 1130/ 1236] Overall Loss 0.284045 Objective Loss 0.284045 LR 0.001000 Time 0.021285 +2023-10-02 20:40:29,905 - Epoch: [24][ 1140/ 1236] Overall Loss 0.284144 Objective Loss 0.284144 LR 0.001000 Time 0.021279 +2023-10-02 20:40:30,110 - Epoch: [24][ 1150/ 1236] Overall Loss 0.284216 Objective Loss 0.284216 LR 0.001000 Time 0.021272 +2023-10-02 20:40:30,318 - Epoch: [24][ 1160/ 1236] Overall Loss 0.284170 Objective Loss 0.284170 LR 0.001000 Time 0.021268 +2023-10-02 20:40:30,523 - Epoch: [24][ 1170/ 1236] Overall Loss 0.284222 Objective Loss 0.284222 LR 0.001000 Time 0.021260 +2023-10-02 20:40:30,731 - Epoch: [24][ 1180/ 1236] Overall Loss 0.284505 Objective Loss 0.284505 LR 0.001000 Time 0.021256 +2023-10-02 20:40:30,936 - Epoch: [24][ 1190/ 1236] Overall Loss 0.284666 Objective Loss 0.284666 LR 0.001000 Time 0.021249 +2023-10-02 20:40:31,142 - Epoch: [24][ 1200/ 1236] Overall Loss 0.284799 Objective Loss 0.284799 LR 0.001000 Time 0.021244 +2023-10-02 20:40:31,348 - Epoch: [24][ 1210/ 1236] Overall Loss 0.284681 Objective Loss 0.284681 LR 0.001000 Time 0.021239 +2023-10-02 20:40:31,556 - Epoch: [24][ 1220/ 1236] Overall Loss 0.284731 Objective Loss 0.284731 LR 0.001000 Time 0.021235 +2023-10-02 20:40:31,814 - Epoch: [24][ 1230/ 1236] Overall Loss 0.284882 Objective Loss 0.284882 LR 0.001000 Time 0.021271 +2023-10-02 20:40:31,936 - Epoch: [24][ 1236/ 1236] Overall Loss 0.284913 Objective Loss 0.284913 Top1 86.354379 Top5 97.352342 LR 0.001000 Time 0.021267 +2023-10-02 20:40:32,091 - --- validate (epoch=24)----------- +2023-10-02 20:40:32,091 - 29943 samples (256 per mini-batch) +2023-10-02 20:40:32,687 - Epoch: [24][ 10/ 117] Loss 0.349555 Top1 81.601562 Top5 98.007812 +2023-10-02 20:40:32,836 - Epoch: [24][ 20/ 117] Loss 0.342996 Top1 82.500000 Top5 97.773438 +2023-10-02 20:40:32,985 - Epoch: [24][ 30/ 117] Loss 0.323360 Top1 83.125000 Top5 98.098958 +2023-10-02 20:40:33,133 - Epoch: [24][ 40/ 117] Loss 0.324608 Top1 82.958984 Top5 97.988281 +2023-10-02 20:40:33,282 - Epoch: [24][ 50/ 117] Loss 0.326640 Top1 83.039062 Top5 98.015625 +2023-10-02 20:40:33,431 - Epoch: [24][ 60/ 117] Loss 0.324743 Top1 83.092448 Top5 98.040365 +2023-10-02 20:40:33,580 - Epoch: [24][ 70/ 117] Loss 0.331651 Top1 82.873884 Top5 98.052455 +2023-10-02 20:40:33,730 - Epoch: [24][ 80/ 117] Loss 0.330415 Top1 82.861328 Top5 98.032227 +2023-10-02 20:40:33,879 - Epoch: [24][ 90/ 117] Loss 0.329271 Top1 82.907986 Top5 97.986111 +2023-10-02 20:40:34,028 - Epoch: [24][ 100/ 117] Loss 0.327884 Top1 82.937500 Top5 98.015625 +2023-10-02 20:40:34,184 - Epoch: [24][ 110/ 117] Loss 0.328909 Top1 82.869318 Top5 98.050426 +2023-10-02 20:40:34,273 - Epoch: [24][ 117/ 117] Loss 0.327635 Top1 82.914204 Top5 98.032929 +2023-10-02 20:40:34,417 - ==> Top1: 82.914 Top5: 98.033 Loss: 0.328 + +2023-10-02 20:40:34,418 - ==> Confusion: +[[ 929 0 2 0 8 5 0 0 4 69 1 1 0 4 4 2 4 1 0 0 16] + [ 0 1048 1 1 6 32 1 13 1 0 1 0 1 0 1 4 5 0 5 1 10] + [ 4 0 970 13 2 1 26 10 0 0 0 1 6 6 0 3 1 0 4 4 5] + [ 0 3 25 980 1 6 3 2 4 1 3 0 3 4 28 1 1 3 6 0 15] + [ 25 4 3 0 952 9 0 1 1 7 0 0 1 2 16 6 15 0 0 2 6] + [ 3 41 2 0 5 990 3 23 2 3 1 3 2 18 4 1 7 0 1 0 7] + [ 1 3 30 0 0 3 1128 5 0 0 2 0 0 1 0 4 1 0 0 7 6] + [ 4 25 23 0 3 28 1 1064 0 0 4 6 0 7 2 5 0 2 29 6 9] + [ 22 3 0 0 1 1 0 1 967 26 9 2 1 13 31 3 3 0 3 0 3] + [ 125 0 0 0 11 0 1 0 38 897 0 0 0 24 12 1 2 0 1 2 5] + [ 2 1 18 10 3 0 3 4 20 0 949 5 1 20 3 2 1 0 4 0 7] + [ 0 3 2 0 1 17 1 6 0 0 0 944 20 12 0 5 0 13 0 4 7] + [ 0 3 6 0 2 7 0 3 0 0 1 54 948 7 3 13 0 13 0 2 6] + [ 1 0 3 0 1 5 0 0 10 10 4 2 0 1064 5 1 0 2 0 2 9] + [ 6 3 2 22 6 0 0 0 18 1 2 0 4 3 1011 0 2 2 4 0 15] + [ 0 0 4 2 5 1 6 0 0 0 0 8 10 2 0 1060 17 10 1 6 2] + [ 2 14 0 0 5 6 1 1 0 0 0 1 1 2 3 6 1106 0 0 2 11] + [ 0 0 3 0 0 0 2 1 0 0 0 4 27 5 2 15 0 978 0 0 1] + [ 1 12 12 29 0 1 3 41 7 1 2 0 1 1 10 3 2 0 932 0 10] + [ 0 4 4 2 1 11 11 17 0 0 4 13 4 3 2 2 9 2 2 1053 8] + [ 129 159 185 83 99 216 56 112 101 88 155 130 394 351 175 72 170 62 100 211 4857]] + +2023-10-02 20:40:34,419 - ==> Best [Top1: 82.948 Top5: 97.986 Sparsity:0.00 Params: 169472 on epoch: 19] +2023-10-02 20:40:34,419 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:40:34,425 - + +2023-10-02 20:40:34,425 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:40:35,437 - Epoch: [25][ 10/ 1236] Overall Loss 0.261089 Objective Loss 0.261089 LR 0.001000 Time 0.101162 +2023-10-02 20:40:35,644 - Epoch: [25][ 20/ 1236] Overall Loss 0.269800 Objective Loss 0.269800 LR 0.001000 Time 0.060892 +2023-10-02 20:40:35,850 - Epoch: [25][ 30/ 1236] Overall Loss 0.275057 Objective Loss 0.275057 LR 0.001000 Time 0.047455 +2023-10-02 20:40:36,057 - Epoch: [25][ 40/ 1236] Overall Loss 0.271730 Objective Loss 0.271730 LR 0.001000 Time 0.040775 +2023-10-02 20:40:36,262 - Epoch: [25][ 50/ 1236] Overall Loss 0.273315 Objective Loss 0.273315 LR 0.001000 Time 0.036708 +2023-10-02 20:40:36,468 - Epoch: [25][ 60/ 1236] Overall Loss 0.271829 Objective Loss 0.271829 LR 0.001000 Time 0.034021 +2023-10-02 20:40:36,673 - Epoch: [25][ 70/ 1236] Overall Loss 0.269014 Objective Loss 0.269014 LR 0.001000 Time 0.032079 +2023-10-02 20:40:36,881 - Epoch: [25][ 80/ 1236] Overall Loss 0.273645 Objective Loss 0.273645 LR 0.001000 Time 0.030663 +2023-10-02 20:40:37,085 - Epoch: [25][ 90/ 1236] Overall Loss 0.272404 Objective Loss 0.272404 LR 0.001000 Time 0.029527 +2023-10-02 20:40:37,293 - Epoch: [25][ 100/ 1236] Overall Loss 0.270475 Objective Loss 0.270475 LR 0.001000 Time 0.028647 +2023-10-02 20:40:37,497 - Epoch: [25][ 110/ 1236] Overall Loss 0.269873 Objective Loss 0.269873 LR 0.001000 Time 0.027899 +2023-10-02 20:40:37,705 - Epoch: [25][ 120/ 1236] Overall Loss 0.272297 Objective Loss 0.272297 LR 0.001000 Time 0.027304 +2023-10-02 20:40:37,910 - Epoch: [25][ 130/ 1236] Overall Loss 0.271605 Objective Loss 0.271605 LR 0.001000 Time 0.026775 +2023-10-02 20:40:38,118 - Epoch: [25][ 140/ 1236] Overall Loss 0.266953 Objective Loss 0.266953 LR 0.001000 Time 0.026346 +2023-10-02 20:40:38,322 - Epoch: [25][ 150/ 1236] Overall Loss 0.267154 Objective Loss 0.267154 LR 0.001000 Time 0.025948 +2023-10-02 20:40:38,529 - Epoch: [25][ 160/ 1236] Overall Loss 0.267791 Objective Loss 0.267791 LR 0.001000 Time 0.025622 +2023-10-02 20:40:38,734 - Epoch: [25][ 170/ 1236] Overall Loss 0.267192 Objective Loss 0.267192 LR 0.001000 Time 0.025316 +2023-10-02 20:40:38,942 - Epoch: [25][ 180/ 1236] Overall Loss 0.266163 Objective Loss 0.266163 LR 0.001000 Time 0.025063 +2023-10-02 20:40:39,146 - Epoch: [25][ 190/ 1236] Overall Loss 0.264963 Objective Loss 0.264963 LR 0.001000 Time 0.024817 +2023-10-02 20:40:39,354 - Epoch: [25][ 200/ 1236] Overall Loss 0.264930 Objective Loss 0.264930 LR 0.001000 Time 0.024613 +2023-10-02 20:40:39,558 - Epoch: [25][ 210/ 1236] Overall Loss 0.265620 Objective Loss 0.265620 LR 0.001000 Time 0.024412 +2023-10-02 20:40:39,766 - Epoch: [25][ 220/ 1236] Overall Loss 0.265860 Objective Loss 0.265860 LR 0.001000 Time 0.024247 +2023-10-02 20:40:39,971 - Epoch: [25][ 230/ 1236] Overall Loss 0.266742 Objective Loss 0.266742 LR 0.001000 Time 0.024081 +2023-10-02 20:40:40,178 - Epoch: [25][ 240/ 1236] Overall Loss 0.268546 Objective Loss 0.268546 LR 0.001000 Time 0.023943 +2023-10-02 20:40:40,383 - Epoch: [25][ 250/ 1236] Overall Loss 0.269306 Objective Loss 0.269306 LR 0.001000 Time 0.023802 +2023-10-02 20:40:40,591 - Epoch: [25][ 260/ 1236] Overall Loss 0.269081 Objective Loss 0.269081 LR 0.001000 Time 0.023685 +2023-10-02 20:40:40,795 - Epoch: [25][ 270/ 1236] Overall Loss 0.269713 Objective Loss 0.269713 LR 0.001000 Time 0.023565 +2023-10-02 20:40:41,002 - Epoch: [25][ 280/ 1236] Overall Loss 0.269011 Objective Loss 0.269011 LR 0.001000 Time 0.023459 +2023-10-02 20:40:41,208 - Epoch: [25][ 290/ 1236] Overall Loss 0.269069 Objective Loss 0.269069 LR 0.001000 Time 0.023355 +2023-10-02 20:40:41,415 - Epoch: [25][ 300/ 1236] Overall Loss 0.269689 Objective Loss 0.269689 LR 0.001000 Time 0.023267 +2023-10-02 20:40:41,620 - Epoch: [25][ 310/ 1236] Overall Loss 0.270628 Objective Loss 0.270628 LR 0.001000 Time 0.023177 +2023-10-02 20:40:41,826 - Epoch: [25][ 320/ 1236] Overall Loss 0.269110 Objective Loss 0.269110 LR 0.001000 Time 0.023097 +2023-10-02 20:40:42,033 - Epoch: [25][ 330/ 1236] Overall Loss 0.268242 Objective Loss 0.268242 LR 0.001000 Time 0.023017 +2023-10-02 20:40:42,239 - Epoch: [25][ 340/ 1236] Overall Loss 0.267663 Objective Loss 0.267663 LR 0.001000 Time 0.022947 +2023-10-02 20:40:42,445 - Epoch: [25][ 350/ 1236] Overall Loss 0.267732 Objective Loss 0.267732 LR 0.001000 Time 0.022879 +2023-10-02 20:40:42,652 - Epoch: [25][ 360/ 1236] Overall Loss 0.267357 Objective Loss 0.267357 LR 0.001000 Time 0.022817 +2023-10-02 20:40:42,859 - Epoch: [25][ 370/ 1236] Overall Loss 0.266963 Objective Loss 0.266963 LR 0.001000 Time 0.022758 +2023-10-02 20:40:43,065 - Epoch: [25][ 380/ 1236] Overall Loss 0.267188 Objective Loss 0.267188 LR 0.001000 Time 0.022701 +2023-10-02 20:40:43,271 - Epoch: [25][ 390/ 1236] Overall Loss 0.267276 Objective Loss 0.267276 LR 0.001000 Time 0.022647 +2023-10-02 20:40:43,478 - Epoch: [25][ 400/ 1236] Overall Loss 0.267675 Objective Loss 0.267675 LR 0.001000 Time 0.022597 +2023-10-02 20:40:43,684 - Epoch: [25][ 410/ 1236] Overall Loss 0.268093 Objective Loss 0.268093 LR 0.001000 Time 0.022548 +2023-10-02 20:40:43,891 - Epoch: [25][ 420/ 1236] Overall Loss 0.269002 Objective Loss 0.269002 LR 0.001000 Time 0.022502 +2023-10-02 20:40:44,097 - Epoch: [25][ 430/ 1236] Overall Loss 0.268994 Objective Loss 0.268994 LR 0.001000 Time 0.022458 +2023-10-02 20:40:44,304 - Epoch: [25][ 440/ 1236] Overall Loss 0.268825 Objective Loss 0.268825 LR 0.001000 Time 0.022417 +2023-10-02 20:40:44,510 - Epoch: [25][ 450/ 1236] Overall Loss 0.269016 Objective Loss 0.269016 LR 0.001000 Time 0.022376 +2023-10-02 20:40:44,716 - Epoch: [25][ 460/ 1236] Overall Loss 0.269344 Objective Loss 0.269344 LR 0.001000 Time 0.022338 +2023-10-02 20:40:44,923 - Epoch: [25][ 470/ 1236] Overall Loss 0.269335 Objective Loss 0.269335 LR 0.001000 Time 0.022301 +2023-10-02 20:40:45,129 - Epoch: [25][ 480/ 1236] Overall Loss 0.269784 Objective Loss 0.269784 LR 0.001000 Time 0.022267 +2023-10-02 20:40:45,336 - Epoch: [25][ 490/ 1236] Overall Loss 0.270192 Objective Loss 0.270192 LR 0.001000 Time 0.022232 +2023-10-02 20:40:45,542 - Epoch: [25][ 500/ 1236] Overall Loss 0.270157 Objective Loss 0.270157 LR 0.001000 Time 0.022200 +2023-10-02 20:40:45,749 - Epoch: [25][ 510/ 1236] Overall Loss 0.270163 Objective Loss 0.270163 LR 0.001000 Time 0.022169 +2023-10-02 20:40:45,955 - Epoch: [25][ 520/ 1236] Overall Loss 0.270622 Objective Loss 0.270622 LR 0.001000 Time 0.022140 +2023-10-02 20:40:46,161 - Epoch: [25][ 530/ 1236] Overall Loss 0.270930 Objective Loss 0.270930 LR 0.001000 Time 0.022109 +2023-10-02 20:40:46,367 - Epoch: [25][ 540/ 1236] Overall Loss 0.271007 Objective Loss 0.271007 LR 0.001000 Time 0.022081 +2023-10-02 20:40:46,574 - Epoch: [25][ 550/ 1236] Overall Loss 0.270699 Objective Loss 0.270699 LR 0.001000 Time 0.022055 +2023-10-02 20:40:46,780 - Epoch: [25][ 560/ 1236] Overall Loss 0.270851 Objective Loss 0.270851 LR 0.001000 Time 0.022029 +2023-10-02 20:40:46,986 - Epoch: [25][ 570/ 1236] Overall Loss 0.271456 Objective Loss 0.271456 LR 0.001000 Time 0.022004 +2023-10-02 20:40:47,193 - Epoch: [25][ 580/ 1236] Overall Loss 0.271731 Objective Loss 0.271731 LR 0.001000 Time 0.021980 +2023-10-02 20:40:47,399 - Epoch: [25][ 590/ 1236] Overall Loss 0.271811 Objective Loss 0.271811 LR 0.001000 Time 0.021957 +2023-10-02 20:40:47,606 - Epoch: [25][ 600/ 1236] Overall Loss 0.271716 Objective Loss 0.271716 LR 0.001000 Time 0.021935 +2023-10-02 20:40:47,812 - Epoch: [25][ 610/ 1236] Overall Loss 0.271544 Objective Loss 0.271544 LR 0.001000 Time 0.021913 +2023-10-02 20:40:48,019 - Epoch: [25][ 620/ 1236] Overall Loss 0.271265 Objective Loss 0.271265 LR 0.001000 Time 0.021893 +2023-10-02 20:40:48,225 - Epoch: [25][ 630/ 1236] Overall Loss 0.271473 Objective Loss 0.271473 LR 0.001000 Time 0.021872 +2023-10-02 20:40:48,432 - Epoch: [25][ 640/ 1236] Overall Loss 0.271454 Objective Loss 0.271454 LR 0.001000 Time 0.021853 +2023-10-02 20:40:48,638 - Epoch: [25][ 650/ 1236] Overall Loss 0.271464 Objective Loss 0.271464 LR 0.001000 Time 0.021833 +2023-10-02 20:40:48,845 - Epoch: [25][ 660/ 1236] Overall Loss 0.271874 Objective Loss 0.271874 LR 0.001000 Time 0.021815 +2023-10-02 20:40:49,051 - Epoch: [25][ 670/ 1236] Overall Loss 0.272415 Objective Loss 0.272415 LR 0.001000 Time 0.021797 +2023-10-02 20:40:49,258 - Epoch: [25][ 680/ 1236] Overall Loss 0.272845 Objective Loss 0.272845 LR 0.001000 Time 0.021780 +2023-10-02 20:40:49,464 - Epoch: [25][ 690/ 1236] Overall Loss 0.272503 Objective Loss 0.272503 LR 0.001000 Time 0.021763 +2023-10-02 20:40:49,671 - Epoch: [25][ 700/ 1236] Overall Loss 0.272439 Objective Loss 0.272439 LR 0.001000 Time 0.021747 +2023-10-02 20:40:49,877 - Epoch: [25][ 710/ 1236] Overall Loss 0.272655 Objective Loss 0.272655 LR 0.001000 Time 0.021731 +2023-10-02 20:40:50,084 - Epoch: [25][ 720/ 1236] Overall Loss 0.273036 Objective Loss 0.273036 LR 0.001000 Time 0.021716 +2023-10-02 20:40:50,290 - Epoch: [25][ 730/ 1236] Overall Loss 0.273238 Objective Loss 0.273238 LR 0.001000 Time 0.021700 +2023-10-02 20:40:50,497 - Epoch: [25][ 740/ 1236] Overall Loss 0.273508 Objective Loss 0.273508 LR 0.001000 Time 0.021686 +2023-10-02 20:40:50,703 - Epoch: [25][ 750/ 1236] Overall Loss 0.273571 Objective Loss 0.273571 LR 0.001000 Time 0.021671 +2023-10-02 20:40:50,910 - Epoch: [25][ 760/ 1236] Overall Loss 0.273612 Objective Loss 0.273612 LR 0.001000 Time 0.021658 +2023-10-02 20:40:51,116 - Epoch: [25][ 770/ 1236] Overall Loss 0.273635 Objective Loss 0.273635 LR 0.001000 Time 0.021644 +2023-10-02 20:40:51,323 - Epoch: [25][ 780/ 1236] Overall Loss 0.273299 Objective Loss 0.273299 LR 0.001000 Time 0.021632 +2023-10-02 20:40:51,529 - Epoch: [25][ 790/ 1236] Overall Loss 0.273705 Objective Loss 0.273705 LR 0.001000 Time 0.021619 +2023-10-02 20:40:51,736 - Epoch: [25][ 800/ 1236] Overall Loss 0.274142 Objective Loss 0.274142 LR 0.001000 Time 0.021607 +2023-10-02 20:40:51,942 - Epoch: [25][ 810/ 1236] Overall Loss 0.274568 Objective Loss 0.274568 LR 0.001000 Time 0.021594 +2023-10-02 20:40:52,154 - Epoch: [25][ 820/ 1236] Overall Loss 0.274449 Objective Loss 0.274449 LR 0.001000 Time 0.021589 +2023-10-02 20:40:52,361 - Epoch: [25][ 830/ 1236] Overall Loss 0.274655 Objective Loss 0.274655 LR 0.001000 Time 0.021578 +2023-10-02 20:40:52,572 - Epoch: [25][ 840/ 1236] Overall Loss 0.274889 Objective Loss 0.274889 LR 0.001000 Time 0.021572 +2023-10-02 20:40:52,780 - Epoch: [25][ 850/ 1236] Overall Loss 0.275154 Objective Loss 0.275154 LR 0.001000 Time 0.021562 +2023-10-02 20:40:52,991 - Epoch: [25][ 860/ 1236] Overall Loss 0.275813 Objective Loss 0.275813 LR 0.001000 Time 0.021556 +2023-10-02 20:40:53,200 - Epoch: [25][ 870/ 1236] Overall Loss 0.275927 Objective Loss 0.275927 LR 0.001000 Time 0.021548 +2023-10-02 20:40:53,411 - Epoch: [25][ 880/ 1236] Overall Loss 0.275921 Objective Loss 0.275921 LR 0.001000 Time 0.021543 +2023-10-02 20:40:53,620 - Epoch: [25][ 890/ 1236] Overall Loss 0.276561 Objective Loss 0.276561 LR 0.001000 Time 0.021535 +2023-10-02 20:40:53,831 - Epoch: [25][ 900/ 1236] Overall Loss 0.276911 Objective Loss 0.276911 LR 0.001000 Time 0.021530 +2023-10-02 20:40:54,039 - Epoch: [25][ 910/ 1236] Overall Loss 0.276779 Objective Loss 0.276779 LR 0.001000 Time 0.021521 +2023-10-02 20:40:54,250 - Epoch: [25][ 920/ 1236] Overall Loss 0.276797 Objective Loss 0.276797 LR 0.001000 Time 0.021517 +2023-10-02 20:40:54,459 - Epoch: [25][ 930/ 1236] Overall Loss 0.277059 Objective Loss 0.277059 LR 0.001000 Time 0.021509 +2023-10-02 20:40:54,670 - Epoch: [25][ 940/ 1236] Overall Loss 0.277277 Objective Loss 0.277277 LR 0.001000 Time 0.021505 +2023-10-02 20:40:54,879 - Epoch: [25][ 950/ 1236] Overall Loss 0.277398 Objective Loss 0.277398 LR 0.001000 Time 0.021497 +2023-10-02 20:40:55,090 - Epoch: [25][ 960/ 1236] Overall Loss 0.277406 Objective Loss 0.277406 LR 0.001000 Time 0.021493 +2023-10-02 20:40:55,298 - Epoch: [25][ 970/ 1236] Overall Loss 0.277807 Objective Loss 0.277807 LR 0.001000 Time 0.021486 +2023-10-02 20:40:55,509 - Epoch: [25][ 980/ 1236] Overall Loss 0.277884 Objective Loss 0.277884 LR 0.001000 Time 0.021482 +2023-10-02 20:40:55,718 - Epoch: [25][ 990/ 1236] Overall Loss 0.277834 Objective Loss 0.277834 LR 0.001000 Time 0.021475 +2023-10-02 20:40:55,928 - Epoch: [25][ 1000/ 1236] Overall Loss 0.278068 Objective Loss 0.278068 LR 0.001000 Time 0.021470 +2023-10-02 20:40:56,137 - Epoch: [25][ 1010/ 1236] Overall Loss 0.278084 Objective Loss 0.278084 LR 0.001000 Time 0.021463 +2023-10-02 20:40:56,346 - Epoch: [25][ 1020/ 1236] Overall Loss 0.278237 Objective Loss 0.278237 LR 0.001000 Time 0.021458 +2023-10-02 20:40:56,555 - Epoch: [25][ 1030/ 1236] Overall Loss 0.278598 Objective Loss 0.278598 LR 0.001000 Time 0.021452 +2023-10-02 20:40:56,766 - Epoch: [25][ 1040/ 1236] Overall Loss 0.279056 Objective Loss 0.279056 LR 0.001000 Time 0.021448 +2023-10-02 20:40:56,974 - Epoch: [25][ 1050/ 1236] Overall Loss 0.279199 Objective Loss 0.279199 LR 0.001000 Time 0.021442 +2023-10-02 20:40:57,185 - Epoch: [25][ 1060/ 1236] Overall Loss 0.279042 Objective Loss 0.279042 LR 0.001000 Time 0.021438 +2023-10-02 20:40:57,394 - Epoch: [25][ 1070/ 1236] Overall Loss 0.278905 Objective Loss 0.278905 LR 0.001000 Time 0.021433 +2023-10-02 20:40:57,605 - Epoch: [25][ 1080/ 1236] Overall Loss 0.278911 Objective Loss 0.278911 LR 0.001000 Time 0.021429 +2023-10-02 20:40:57,813 - Epoch: [25][ 1090/ 1236] Overall Loss 0.278766 Objective Loss 0.278766 LR 0.001000 Time 0.021423 +2023-10-02 20:40:58,024 - Epoch: [25][ 1100/ 1236] Overall Loss 0.279062 Objective Loss 0.279062 LR 0.001000 Time 0.021419 +2023-10-02 20:40:58,233 - Epoch: [25][ 1110/ 1236] Overall Loss 0.278955 Objective Loss 0.278955 LR 0.001000 Time 0.021414 +2023-10-02 20:40:58,443 - Epoch: [25][ 1120/ 1236] Overall Loss 0.278956 Objective Loss 0.278956 LR 0.001000 Time 0.021410 +2023-10-02 20:40:58,652 - Epoch: [25][ 1130/ 1236] Overall Loss 0.279039 Objective Loss 0.279039 LR 0.001000 Time 0.021405 +2023-10-02 20:40:58,864 - Epoch: [25][ 1140/ 1236] Overall Loss 0.278903 Objective Loss 0.278903 LR 0.001000 Time 0.021403 +2023-10-02 20:40:59,073 - Epoch: [25][ 1150/ 1236] Overall Loss 0.278819 Objective Loss 0.278819 LR 0.001000 Time 0.021397 +2023-10-02 20:40:59,284 - Epoch: [25][ 1160/ 1236] Overall Loss 0.278748 Objective Loss 0.278748 LR 0.001000 Time 0.021394 +2023-10-02 20:40:59,493 - Epoch: [25][ 1170/ 1236] Overall Loss 0.278965 Objective Loss 0.278965 LR 0.001000 Time 0.021389 +2023-10-02 20:40:59,704 - Epoch: [25][ 1180/ 1236] Overall Loss 0.279044 Objective Loss 0.279044 LR 0.001000 Time 0.021386 +2023-10-02 20:40:59,913 - Epoch: [25][ 1190/ 1236] Overall Loss 0.279153 Objective Loss 0.279153 LR 0.001000 Time 0.021382 +2023-10-02 20:41:00,123 - Epoch: [25][ 1200/ 1236] Overall Loss 0.279217 Objective Loss 0.279217 LR 0.001000 Time 0.021379 +2023-10-02 20:41:00,332 - Epoch: [25][ 1210/ 1236] Overall Loss 0.279101 Objective Loss 0.279101 LR 0.001000 Time 0.021374 +2023-10-02 20:41:00,542 - Epoch: [25][ 1220/ 1236] Overall Loss 0.278908 Objective Loss 0.278908 LR 0.001000 Time 0.021371 +2023-10-02 20:41:00,803 - Epoch: [25][ 1230/ 1236] Overall Loss 0.279061 Objective Loss 0.279061 LR 0.001000 Time 0.021409 +2023-10-02 20:41:00,925 - Epoch: [25][ 1236/ 1236] Overall Loss 0.279159 Objective Loss 0.279159 Top1 86.354379 Top5 97.963340 LR 0.001000 Time 0.021404 +2023-10-02 20:41:01,059 - --- validate (epoch=25)----------- +2023-10-02 20:41:01,059 - 29943 samples (256 per mini-batch) +2023-10-02 20:41:01,547 - Epoch: [25][ 10/ 117] Loss 0.355878 Top1 83.515625 Top5 98.046875 +2023-10-02 20:41:01,698 - Epoch: [25][ 20/ 117] Loss 0.358870 Top1 83.085938 Top5 98.144531 +2023-10-02 20:41:01,852 - Epoch: [25][ 30/ 117] Loss 0.334088 Top1 83.697917 Top5 98.151042 +2023-10-02 20:41:02,008 - Epoch: [25][ 40/ 117] Loss 0.334420 Top1 83.437500 Top5 98.222656 +2023-10-02 20:41:02,163 - Epoch: [25][ 50/ 117] Loss 0.332628 Top1 83.398438 Top5 98.312500 +2023-10-02 20:41:02,318 - Epoch: [25][ 60/ 117] Loss 0.336140 Top1 83.307292 Top5 98.255208 +2023-10-02 20:41:02,471 - Epoch: [25][ 70/ 117] Loss 0.335004 Top1 83.191964 Top5 98.225446 +2023-10-02 20:41:02,626 - Epoch: [25][ 80/ 117] Loss 0.332419 Top1 83.227539 Top5 98.247070 +2023-10-02 20:41:02,781 - Epoch: [25][ 90/ 117] Loss 0.332052 Top1 83.194444 Top5 98.224826 +2023-10-02 20:41:02,936 - Epoch: [25][ 100/ 117] Loss 0.332797 Top1 83.195312 Top5 98.222656 +2023-10-02 20:41:03,096 - Epoch: [25][ 110/ 117] Loss 0.334097 Top1 83.160511 Top5 98.199574 +2023-10-02 20:41:03,185 - Epoch: [25][ 117/ 117] Loss 0.333950 Top1 83.121264 Top5 98.176535 +2023-10-02 20:41:03,321 - ==> Top1: 83.121 Top5: 98.177 Loss: 0.334 + +2023-10-02 20:41:03,322 - ==> Confusion: +[[ 901 2 2 0 4 2 0 0 6 103 0 1 0 4 4 3 1 3 0 0 14] + [ 0 1050 1 0 6 24 2 19 3 2 1 0 0 1 2 5 2 0 7 0 6] + [ 4 0 949 17 4 0 19 11 0 5 5 2 3 4 1 2 2 1 5 10 12] + [ 1 3 15 937 0 6 4 3 9 2 5 2 6 9 49 1 1 4 12 1 19] + [ 27 4 2 0 959 7 0 1 1 20 0 0 1 4 14 3 2 0 0 2 3] + [ 3 40 0 0 4 988 2 18 6 9 0 8 2 7 2 1 3 0 4 6 13] + [ 0 2 34 1 2 1 1106 11 0 0 3 3 2 1 0 5 0 0 0 11 9] + [ 3 19 12 0 2 23 0 1087 1 1 2 7 1 2 1 1 0 0 40 9 7] + [ 14 0 0 0 1 1 0 0 993 43 5 2 1 8 13 2 0 4 0 1 1] + [ 80 1 0 0 2 0 0 0 39 963 1 0 1 17 2 2 1 1 0 3 6] + [ 2 2 4 13 2 2 0 7 43 1 920 3 1 26 4 2 0 3 8 1 9] + [ 1 3 1 0 2 13 0 1 0 3 0 944 32 7 0 2 0 14 0 4 8] + [ 0 0 1 1 1 2 1 1 1 0 1 65 940 5 3 9 1 12 3 4 17] + [ 2 0 1 0 3 5 0 0 12 18 1 9 0 1051 4 0 0 1 0 1 11] + [ 6 3 3 7 3 0 0 0 39 8 0 0 2 6 1010 0 0 2 5 0 7] + [ 0 1 1 1 5 0 1 0 0 2 0 7 4 1 0 1076 13 10 2 4 6] + [ 0 15 2 0 12 4 0 0 3 0 0 7 1 5 4 11 1069 0 1 6 21] + [ 0 0 1 1 0 0 1 0 1 2 1 5 25 3 3 10 2 978 1 1 3] + [ 2 9 8 8 1 0 0 24 8 0 3 0 0 0 20 2 1 0 969 1 12] + [ 0 5 4 0 1 6 2 27 1 0 1 13 3 2 0 3 4 0 2 1066 12] + [ 143 187 121 57 94 168 36 132 195 125 127 150 331 327 192 70 93 49 171 204 4933]] + +2023-10-02 20:41:03,323 - ==> Best [Top1: 83.121 Top5: 98.177 Sparsity:0.00 Params: 169472 on epoch: 25] +2023-10-02 20:41:03,323 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:41:03,336 - + +2023-10-02 20:41:03,337 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:41:04,330 - Epoch: [26][ 10/ 1236] Overall Loss 0.267357 Objective Loss 0.267357 LR 0.001000 Time 0.099335 +2023-10-02 20:41:04,536 - Epoch: [26][ 20/ 1236] Overall Loss 0.263256 Objective Loss 0.263256 LR 0.001000 Time 0.059937 +2023-10-02 20:41:04,742 - Epoch: [26][ 30/ 1236] Overall Loss 0.270939 Objective Loss 0.270939 LR 0.001000 Time 0.046761 +2023-10-02 20:41:04,949 - Epoch: [26][ 40/ 1236] Overall Loss 0.269998 Objective Loss 0.269998 LR 0.001000 Time 0.040248 +2023-10-02 20:41:05,154 - Epoch: [26][ 50/ 1236] Overall Loss 0.265776 Objective Loss 0.265776 LR 0.001000 Time 0.036283 +2023-10-02 20:41:05,361 - Epoch: [26][ 60/ 1236] Overall Loss 0.265787 Objective Loss 0.265787 LR 0.001000 Time 0.033684 +2023-10-02 20:41:05,565 - Epoch: [26][ 70/ 1236] Overall Loss 0.260693 Objective Loss 0.260693 LR 0.001000 Time 0.031789 +2023-10-02 20:41:05,772 - Epoch: [26][ 80/ 1236] Overall Loss 0.257663 Objective Loss 0.257663 LR 0.001000 Time 0.030399 +2023-10-02 20:41:05,977 - Epoch: [26][ 90/ 1236] Overall Loss 0.255549 Objective Loss 0.255549 LR 0.001000 Time 0.029291 +2023-10-02 20:41:06,184 - Epoch: [26][ 100/ 1236] Overall Loss 0.255355 Objective Loss 0.255355 LR 0.001000 Time 0.028435 +2023-10-02 20:41:06,389 - Epoch: [26][ 110/ 1236] Overall Loss 0.255597 Objective Loss 0.255597 LR 0.001000 Time 0.027703 +2023-10-02 20:41:06,596 - Epoch: [26][ 120/ 1236] Overall Loss 0.256427 Objective Loss 0.256427 LR 0.001000 Time 0.027122 +2023-10-02 20:41:06,801 - Epoch: [26][ 130/ 1236] Overall Loss 0.256030 Objective Loss 0.256030 LR 0.001000 Time 0.026607 +2023-10-02 20:41:07,008 - Epoch: [26][ 140/ 1236] Overall Loss 0.256222 Objective Loss 0.256222 LR 0.001000 Time 0.026186 +2023-10-02 20:41:07,213 - Epoch: [26][ 150/ 1236] Overall Loss 0.256886 Objective Loss 0.256886 LR 0.001000 Time 0.025802 +2023-10-02 20:41:07,421 - Epoch: [26][ 160/ 1236] Overall Loss 0.257736 Objective Loss 0.257736 LR 0.001000 Time 0.025487 +2023-10-02 20:41:07,625 - Epoch: [26][ 170/ 1236] Overall Loss 0.257385 Objective Loss 0.257385 LR 0.001000 Time 0.025187 +2023-10-02 20:41:07,833 - Epoch: [26][ 180/ 1236] Overall Loss 0.258322 Objective Loss 0.258322 LR 0.001000 Time 0.024941 +2023-10-02 20:41:08,037 - Epoch: [26][ 190/ 1236] Overall Loss 0.258482 Objective Loss 0.258482 LR 0.001000 Time 0.024701 +2023-10-02 20:41:08,244 - Epoch: [26][ 200/ 1236] Overall Loss 0.258539 Objective Loss 0.258539 LR 0.001000 Time 0.024503 +2023-10-02 20:41:08,448 - Epoch: [26][ 210/ 1236] Overall Loss 0.258044 Objective Loss 0.258044 LR 0.001000 Time 0.024306 +2023-10-02 20:41:08,656 - Epoch: [26][ 220/ 1236] Overall Loss 0.258369 Objective Loss 0.258369 LR 0.001000 Time 0.024142 +2023-10-02 20:41:08,860 - Epoch: [26][ 230/ 1236] Overall Loss 0.259122 Objective Loss 0.259122 LR 0.001000 Time 0.023979 +2023-10-02 20:41:09,068 - Epoch: [26][ 240/ 1236] Overall Loss 0.260658 Objective Loss 0.260658 LR 0.001000 Time 0.023844 +2023-10-02 20:41:09,272 - Epoch: [26][ 250/ 1236] Overall Loss 0.260670 Objective Loss 0.260670 LR 0.001000 Time 0.023706 +2023-10-02 20:41:09,479 - Epoch: [26][ 260/ 1236] Overall Loss 0.260390 Objective Loss 0.260390 LR 0.001000 Time 0.023592 +2023-10-02 20:41:09,684 - Epoch: [26][ 270/ 1236] Overall Loss 0.261408 Objective Loss 0.261408 LR 0.001000 Time 0.023474 +2023-10-02 20:41:09,892 - Epoch: [26][ 280/ 1236] Overall Loss 0.261878 Objective Loss 0.261878 LR 0.001000 Time 0.023378 +2023-10-02 20:41:10,096 - Epoch: [26][ 290/ 1236] Overall Loss 0.261928 Objective Loss 0.261928 LR 0.001000 Time 0.023275 +2023-10-02 20:41:10,304 - Epoch: [26][ 300/ 1236] Overall Loss 0.262062 Objective Loss 0.262062 LR 0.001000 Time 0.023191 +2023-10-02 20:41:10,508 - Epoch: [26][ 310/ 1236] Overall Loss 0.262316 Objective Loss 0.262316 LR 0.001000 Time 0.023101 +2023-10-02 20:41:10,716 - Epoch: [26][ 320/ 1236] Overall Loss 0.262024 Objective Loss 0.262024 LR 0.001000 Time 0.023028 +2023-10-02 20:41:10,921 - Epoch: [26][ 330/ 1236] Overall Loss 0.262240 Objective Loss 0.262240 LR 0.001000 Time 0.022949 +2023-10-02 20:41:11,129 - Epoch: [26][ 340/ 1236] Overall Loss 0.263372 Objective Loss 0.263372 LR 0.001000 Time 0.022885 +2023-10-02 20:41:11,333 - Epoch: [26][ 350/ 1236] Overall Loss 0.264815 Objective Loss 0.264815 LR 0.001000 Time 0.022815 +2023-10-02 20:41:11,540 - Epoch: [26][ 360/ 1236] Overall Loss 0.265335 Objective Loss 0.265335 LR 0.001000 Time 0.022756 +2023-10-02 20:41:11,745 - Epoch: [26][ 370/ 1236] Overall Loss 0.265812 Objective Loss 0.265812 LR 0.001000 Time 0.022693 +2023-10-02 20:41:11,953 - Epoch: [26][ 380/ 1236] Overall Loss 0.266266 Objective Loss 0.266266 LR 0.001000 Time 0.022641 +2023-10-02 20:41:12,158 - Epoch: [26][ 390/ 1236] Overall Loss 0.267000 Objective Loss 0.267000 LR 0.001000 Time 0.022586 +2023-10-02 20:41:12,365 - Epoch: [26][ 400/ 1236] Overall Loss 0.267494 Objective Loss 0.267494 LR 0.001000 Time 0.022539 +2023-10-02 20:41:12,570 - Epoch: [26][ 410/ 1236] Overall Loss 0.267434 Objective Loss 0.267434 LR 0.001000 Time 0.022488 +2023-10-02 20:41:12,777 - Epoch: [26][ 420/ 1236] Overall Loss 0.267863 Objective Loss 0.267863 LR 0.001000 Time 0.022446 +2023-10-02 20:41:12,983 - Epoch: [26][ 430/ 1236] Overall Loss 0.268405 Objective Loss 0.268405 LR 0.001000 Time 0.022401 +2023-10-02 20:41:13,191 - Epoch: [26][ 440/ 1236] Overall Loss 0.268536 Objective Loss 0.268536 LR 0.001000 Time 0.022364 +2023-10-02 20:41:13,395 - Epoch: [26][ 450/ 1236] Overall Loss 0.269436 Objective Loss 0.269436 LR 0.001000 Time 0.022321 +2023-10-02 20:41:13,603 - Epoch: [26][ 460/ 1236] Overall Loss 0.270638 Objective Loss 0.270638 LR 0.001000 Time 0.022286 +2023-10-02 20:41:13,808 - Epoch: [26][ 470/ 1236] Overall Loss 0.270389 Objective Loss 0.270389 LR 0.001000 Time 0.022248 +2023-10-02 20:41:14,015 - Epoch: [26][ 480/ 1236] Overall Loss 0.270364 Objective Loss 0.270364 LR 0.001000 Time 0.022216 +2023-10-02 20:41:14,220 - Epoch: [26][ 490/ 1236] Overall Loss 0.270487 Objective Loss 0.270487 LR 0.001000 Time 0.022180 +2023-10-02 20:41:14,428 - Epoch: [26][ 500/ 1236] Overall Loss 0.270586 Objective Loss 0.270586 LR 0.001000 Time 0.022151 +2023-10-02 20:41:14,633 - Epoch: [26][ 510/ 1236] Overall Loss 0.270791 Objective Loss 0.270791 LR 0.001000 Time 0.022118 +2023-10-02 20:41:14,840 - Epoch: [26][ 520/ 1236] Overall Loss 0.271041 Objective Loss 0.271041 LR 0.001000 Time 0.022091 +2023-10-02 20:41:15,045 - Epoch: [26][ 530/ 1236] Overall Loss 0.271126 Objective Loss 0.271126 LR 0.001000 Time 0.022061 +2023-10-02 20:41:15,253 - Epoch: [26][ 540/ 1236] Overall Loss 0.271484 Objective Loss 0.271484 LR 0.001000 Time 0.022036 +2023-10-02 20:41:15,457 - Epoch: [26][ 550/ 1236] Overall Loss 0.271747 Objective Loss 0.271747 LR 0.001000 Time 0.022007 +2023-10-02 20:41:15,665 - Epoch: [26][ 560/ 1236] Overall Loss 0.271867 Objective Loss 0.271867 LR 0.001000 Time 0.021984 +2023-10-02 20:41:15,870 - Epoch: [26][ 570/ 1236] Overall Loss 0.271489 Objective Loss 0.271489 LR 0.001000 Time 0.021958 +2023-10-02 20:41:16,078 - Epoch: [26][ 580/ 1236] Overall Loss 0.272076 Objective Loss 0.272076 LR 0.001000 Time 0.021937 +2023-10-02 20:41:16,282 - Epoch: [26][ 590/ 1236] Overall Loss 0.272643 Objective Loss 0.272643 LR 0.001000 Time 0.021912 +2023-10-02 20:41:16,490 - Epoch: [26][ 600/ 1236] Overall Loss 0.273235 Objective Loss 0.273235 LR 0.001000 Time 0.021892 +2023-10-02 20:41:16,695 - Epoch: [26][ 610/ 1236] Overall Loss 0.273480 Objective Loss 0.273480 LR 0.001000 Time 0.021869 +2023-10-02 20:41:16,903 - Epoch: [26][ 620/ 1236] Overall Loss 0.274057 Objective Loss 0.274057 LR 0.001000 Time 0.021850 +2023-10-02 20:41:17,108 - Epoch: [26][ 630/ 1236] Overall Loss 0.274201 Objective Loss 0.274201 LR 0.001000 Time 0.021828 +2023-10-02 20:41:17,315 - Epoch: [26][ 640/ 1236] Overall Loss 0.275001 Objective Loss 0.275001 LR 0.001000 Time 0.021811 +2023-10-02 20:41:17,520 - Epoch: [26][ 650/ 1236] Overall Loss 0.275405 Objective Loss 0.275405 LR 0.001000 Time 0.021790 +2023-10-02 20:41:17,728 - Epoch: [26][ 660/ 1236] Overall Loss 0.275056 Objective Loss 0.275056 LR 0.001000 Time 0.021775 +2023-10-02 20:41:17,933 - Epoch: [26][ 670/ 1236] Overall Loss 0.275248 Objective Loss 0.275248 LR 0.001000 Time 0.021755 +2023-10-02 20:41:18,141 - Epoch: [26][ 680/ 1236] Overall Loss 0.275229 Objective Loss 0.275229 LR 0.001000 Time 0.021740 +2023-10-02 20:41:18,345 - Epoch: [26][ 690/ 1236] Overall Loss 0.275907 Objective Loss 0.275907 LR 0.001000 Time 0.021722 +2023-10-02 20:41:18,553 - Epoch: [26][ 700/ 1236] Overall Loss 0.275839 Objective Loss 0.275839 LR 0.001000 Time 0.021707 +2023-10-02 20:41:18,758 - Epoch: [26][ 710/ 1236] Overall Loss 0.276423 Objective Loss 0.276423 LR 0.001000 Time 0.021690 +2023-10-02 20:41:18,966 - Epoch: [26][ 720/ 1236] Overall Loss 0.276618 Objective Loss 0.276618 LR 0.001000 Time 0.021677 +2023-10-02 20:41:19,171 - Epoch: [26][ 730/ 1236] Overall Loss 0.276666 Objective Loss 0.276666 LR 0.001000 Time 0.021661 +2023-10-02 20:41:19,378 - Epoch: [26][ 740/ 1236] Overall Loss 0.276576 Objective Loss 0.276576 LR 0.001000 Time 0.021648 +2023-10-02 20:41:19,583 - Epoch: [26][ 750/ 1236] Overall Loss 0.276769 Objective Loss 0.276769 LR 0.001000 Time 0.021632 +2023-10-02 20:41:19,791 - Epoch: [26][ 760/ 1236] Overall Loss 0.276656 Objective Loss 0.276656 LR 0.001000 Time 0.021620 +2023-10-02 20:41:19,996 - Epoch: [26][ 770/ 1236] Overall Loss 0.276392 Objective Loss 0.276392 LR 0.001000 Time 0.021605 +2023-10-02 20:41:20,202 - Epoch: [26][ 780/ 1236] Overall Loss 0.276240 Objective Loss 0.276240 LR 0.001000 Time 0.021592 +2023-10-02 20:41:20,408 - Epoch: [26][ 790/ 1236] Overall Loss 0.276245 Objective Loss 0.276245 LR 0.001000 Time 0.021578 +2023-10-02 20:41:20,614 - Epoch: [26][ 800/ 1236] Overall Loss 0.276297 Objective Loss 0.276297 LR 0.001000 Time 0.021566 +2023-10-02 20:41:20,820 - Epoch: [26][ 810/ 1236] Overall Loss 0.276300 Objective Loss 0.276300 LR 0.001000 Time 0.021552 +2023-10-02 20:41:21,027 - Epoch: [26][ 820/ 1236] Overall Loss 0.276349 Objective Loss 0.276349 LR 0.001000 Time 0.021541 +2023-10-02 20:41:21,233 - Epoch: [26][ 830/ 1236] Overall Loss 0.276003 Objective Loss 0.276003 LR 0.001000 Time 0.021528 +2023-10-02 20:41:21,441 - Epoch: [26][ 840/ 1236] Overall Loss 0.276013 Objective Loss 0.276013 LR 0.001000 Time 0.021519 +2023-10-02 20:41:21,646 - Epoch: [26][ 850/ 1236] Overall Loss 0.276086 Objective Loss 0.276086 LR 0.001000 Time 0.021506 +2023-10-02 20:41:21,852 - Epoch: [26][ 860/ 1236] Overall Loss 0.275974 Objective Loss 0.275974 LR 0.001000 Time 0.021496 +2023-10-02 20:41:22,058 - Epoch: [26][ 870/ 1236] Overall Loss 0.276280 Objective Loss 0.276280 LR 0.001000 Time 0.021484 +2023-10-02 20:41:22,265 - Epoch: [26][ 880/ 1236] Overall Loss 0.276167 Objective Loss 0.276167 LR 0.001000 Time 0.021474 +2023-10-02 20:41:22,471 - Epoch: [26][ 890/ 1236] Overall Loss 0.275928 Objective Loss 0.275928 LR 0.001000 Time 0.021463 +2023-10-02 20:41:22,678 - Epoch: [26][ 900/ 1236] Overall Loss 0.275689 Objective Loss 0.275689 LR 0.001000 Time 0.021454 +2023-10-02 20:41:22,883 - Epoch: [26][ 910/ 1236] Overall Loss 0.275433 Objective Loss 0.275433 LR 0.001000 Time 0.021443 +2023-10-02 20:41:23,090 - Epoch: [26][ 920/ 1236] Overall Loss 0.275388 Objective Loss 0.275388 LR 0.001000 Time 0.021435 +2023-10-02 20:41:23,294 - Epoch: [26][ 930/ 1236] Overall Loss 0.275474 Objective Loss 0.275474 LR 0.001000 Time 0.021422 +2023-10-02 20:41:23,502 - Epoch: [26][ 940/ 1236] Overall Loss 0.275394 Objective Loss 0.275394 LR 0.001000 Time 0.021416 +2023-10-02 20:41:23,709 - Epoch: [26][ 950/ 1236] Overall Loss 0.275521 Objective Loss 0.275521 LR 0.001000 Time 0.021406 +2023-10-02 20:41:23,916 - Epoch: [26][ 960/ 1236] Overall Loss 0.276024 Objective Loss 0.276024 LR 0.001000 Time 0.021399 +2023-10-02 20:41:24,123 - Epoch: [26][ 970/ 1236] Overall Loss 0.275904 Objective Loss 0.275904 LR 0.001000 Time 0.021390 +2023-10-02 20:41:24,331 - Epoch: [26][ 980/ 1236] Overall Loss 0.276187 Objective Loss 0.276187 LR 0.001000 Time 0.021384 +2023-10-02 20:41:24,537 - Epoch: [26][ 990/ 1236] Overall Loss 0.276409 Objective Loss 0.276409 LR 0.001000 Time 0.021374 +2023-10-02 20:41:24,745 - Epoch: [26][ 1000/ 1236] Overall Loss 0.276379 Objective Loss 0.276379 LR 0.001000 Time 0.021368 +2023-10-02 20:41:24,952 - Epoch: [26][ 1010/ 1236] Overall Loss 0.276625 Objective Loss 0.276625 LR 0.001000 Time 0.021360 +2023-10-02 20:41:25,160 - Epoch: [26][ 1020/ 1236] Overall Loss 0.277014 Objective Loss 0.277014 LR 0.001000 Time 0.021354 +2023-10-02 20:41:25,367 - Epoch: [26][ 1030/ 1236] Overall Loss 0.277002 Objective Loss 0.277002 LR 0.001000 Time 0.021346 +2023-10-02 20:41:25,575 - Epoch: [26][ 1040/ 1236] Overall Loss 0.276963 Objective Loss 0.276963 LR 0.001000 Time 0.021340 +2023-10-02 20:41:25,782 - Epoch: [26][ 1050/ 1236] Overall Loss 0.277150 Objective Loss 0.277150 LR 0.001000 Time 0.021333 +2023-10-02 20:41:25,990 - Epoch: [26][ 1060/ 1236] Overall Loss 0.277066 Objective Loss 0.277066 LR 0.001000 Time 0.021328 +2023-10-02 20:41:26,197 - Epoch: [26][ 1070/ 1236] Overall Loss 0.276946 Objective Loss 0.276946 LR 0.001000 Time 0.021320 +2023-10-02 20:41:26,405 - Epoch: [26][ 1080/ 1236] Overall Loss 0.276927 Objective Loss 0.276927 LR 0.001000 Time 0.021315 +2023-10-02 20:41:26,612 - Epoch: [26][ 1090/ 1236] Overall Loss 0.277012 Objective Loss 0.277012 LR 0.001000 Time 0.021308 +2023-10-02 20:41:26,820 - Epoch: [26][ 1100/ 1236] Overall Loss 0.277117 Objective Loss 0.277117 LR 0.001000 Time 0.021303 +2023-10-02 20:41:27,027 - Epoch: [26][ 1110/ 1236] Overall Loss 0.277017 Objective Loss 0.277017 LR 0.001000 Time 0.021297 +2023-10-02 20:41:27,235 - Epoch: [26][ 1120/ 1236] Overall Loss 0.276504 Objective Loss 0.276504 LR 0.001000 Time 0.021292 +2023-10-02 20:41:27,442 - Epoch: [26][ 1130/ 1236] Overall Loss 0.276287 Objective Loss 0.276287 LR 0.001000 Time 0.021285 +2023-10-02 20:41:27,650 - Epoch: [26][ 1140/ 1236] Overall Loss 0.276162 Objective Loss 0.276162 LR 0.001000 Time 0.021281 +2023-10-02 20:41:27,855 - Epoch: [26][ 1150/ 1236] Overall Loss 0.276353 Objective Loss 0.276353 LR 0.001000 Time 0.021272 +2023-10-02 20:41:28,064 - Epoch: [26][ 1160/ 1236] Overall Loss 0.276268 Objective Loss 0.276268 LR 0.001000 Time 0.021270 +2023-10-02 20:41:28,270 - Epoch: [26][ 1170/ 1236] Overall Loss 0.276068 Objective Loss 0.276068 LR 0.001000 Time 0.021264 +2023-10-02 20:41:28,479 - Epoch: [26][ 1180/ 1236] Overall Loss 0.276257 Objective Loss 0.276257 LR 0.001000 Time 0.021260 +2023-10-02 20:41:28,685 - Epoch: [26][ 1190/ 1236] Overall Loss 0.276205 Objective Loss 0.276205 LR 0.001000 Time 0.021254 +2023-10-02 20:41:28,893 - Epoch: [26][ 1200/ 1236] Overall Loss 0.275995 Objective Loss 0.275995 LR 0.001000 Time 0.021250 +2023-10-02 20:41:29,100 - Epoch: [26][ 1210/ 1236] Overall Loss 0.275898 Objective Loss 0.275898 LR 0.001000 Time 0.021245 +2023-10-02 20:41:29,309 - Epoch: [26][ 1220/ 1236] Overall Loss 0.275736 Objective Loss 0.275736 LR 0.001000 Time 0.021241 +2023-10-02 20:41:29,567 - Epoch: [26][ 1230/ 1236] Overall Loss 0.275798 Objective Loss 0.275798 LR 0.001000 Time 0.021277 +2023-10-02 20:41:29,689 - Epoch: [26][ 1236/ 1236] Overall Loss 0.275794 Objective Loss 0.275794 Top1 86.354379 Top5 98.370672 LR 0.001000 Time 0.021272 +2023-10-02 20:41:29,832 - --- validate (epoch=26)----------- +2023-10-02 20:41:29,833 - 29943 samples (256 per mini-batch) +2023-10-02 20:41:30,437 - Epoch: [26][ 10/ 117] Loss 0.325697 Top1 82.656250 Top5 98.203125 +2023-10-02 20:41:30,586 - Epoch: [26][ 20/ 117] Loss 0.328461 Top1 83.222656 Top5 98.164062 +2023-10-02 20:41:30,736 - Epoch: [26][ 30/ 117] Loss 0.321422 Top1 83.111979 Top5 98.268229 +2023-10-02 20:41:30,884 - Epoch: [26][ 40/ 117] Loss 0.327662 Top1 82.910156 Top5 98.203125 +2023-10-02 20:41:31,033 - Epoch: [26][ 50/ 117] Loss 0.322328 Top1 82.945312 Top5 98.125000 +2023-10-02 20:41:31,181 - Epoch: [26][ 60/ 117] Loss 0.327983 Top1 82.630208 Top5 98.085938 +2023-10-02 20:41:31,328 - Epoch: [26][ 70/ 117] Loss 0.328146 Top1 82.784598 Top5 98.046875 +2023-10-02 20:41:31,476 - Epoch: [26][ 80/ 117] Loss 0.324032 Top1 82.856445 Top5 98.032227 +2023-10-02 20:41:31,623 - Epoch: [26][ 90/ 117] Loss 0.323363 Top1 82.855903 Top5 98.020833 +2023-10-02 20:41:31,771 - Epoch: [26][ 100/ 117] Loss 0.322346 Top1 82.812500 Top5 98.082031 +2023-10-02 20:41:31,925 - Epoch: [26][ 110/ 117] Loss 0.326069 Top1 82.674006 Top5 98.061080 +2023-10-02 20:41:32,014 - Epoch: [26][ 117/ 117] Loss 0.328182 Top1 82.620312 Top5 98.032929 +2023-10-02 20:41:32,151 - ==> Top1: 82.620 Top5: 98.033 Loss: 0.328 + +2023-10-02 20:41:32,152 - ==> Confusion: +[[ 937 0 2 1 1 6 0 0 15 64 2 0 0 1 2 0 9 1 0 0 9] + [ 1 1053 0 1 2 27 0 22 2 2 1 1 1 1 0 1 2 0 6 2 6] + [ 10 0 951 14 2 0 28 14 1 2 4 0 5 3 0 5 0 0 7 4 6] + [ 2 3 21 990 1 6 4 2 10 0 3 0 1 2 20 0 2 0 5 2 15] + [ 34 12 5 0 932 10 0 0 3 14 1 0 0 3 9 3 19 0 0 3 2] + [ 5 39 2 0 1 982 0 38 4 5 1 4 3 10 3 1 3 0 2 4 9] + [ 1 7 21 0 0 0 1123 11 0 0 4 1 0 3 0 1 1 2 0 11 5] + [ 4 14 15 0 0 26 4 1090 3 0 5 6 1 4 0 1 1 0 24 12 8] + [ 23 2 0 0 1 0 0 1 990 33 7 2 0 9 11 1 0 2 0 4 3] + [ 115 0 0 0 0 0 1 1 53 913 2 0 1 17 5 0 1 0 0 4 6] + [ 3 2 10 10 2 2 1 8 22 2 953 2 1 10 3 2 1 2 7 4 6] + [ 0 1 2 0 0 17 1 6 0 2 0 936 26 8 0 2 3 18 0 9 4] + [ 1 3 3 4 1 5 2 2 0 1 4 40 953 2 4 8 4 22 0 4 5] + [ 3 0 3 0 3 10 1 3 25 12 6 4 0 1029 4 0 2 1 0 3 10] + [ 14 4 1 16 2 0 0 0 43 2 2 0 4 2 983 0 2 2 7 0 17] + [ 0 0 2 4 6 1 5 0 2 0 0 7 8 3 0 1045 23 19 1 4 4] + [ 0 14 0 0 3 8 0 2 2 0 0 4 0 1 2 5 1108 0 0 2 10] + [ 2 0 0 4 0 2 4 0 3 0 0 6 15 2 0 7 2 988 0 0 3] + [ 1 10 9 24 0 0 0 42 9 2 6 0 0 2 8 0 3 0 943 1 8] + [ 0 2 2 0 1 3 10 16 1 0 2 9 5 2 0 2 11 1 0 1082 3] + [ 211 178 150 77 70 261 29 145 152 84 168 117 405 275 117 52 222 66 144 224 4758]] + +2023-10-02 20:41:32,153 - ==> Best [Top1: 83.121 Top5: 98.177 Sparsity:0.00 Params: 169472 on epoch: 25] +2023-10-02 20:41:32,153 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:41:32,159 - + +2023-10-02 20:41:32,159 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:41:33,185 - Epoch: [27][ 10/ 1236] Overall Loss 0.275361 Objective Loss 0.275361 LR 0.001000 Time 0.102539 +2023-10-02 20:41:33,393 - Epoch: [27][ 20/ 1236] Overall Loss 0.270961 Objective Loss 0.270961 LR 0.001000 Time 0.061629 +2023-10-02 20:41:33,599 - Epoch: [27][ 30/ 1236] Overall Loss 0.271645 Objective Loss 0.271645 LR 0.001000 Time 0.047908 +2023-10-02 20:41:33,807 - Epoch: [27][ 40/ 1236] Overall Loss 0.267852 Objective Loss 0.267852 LR 0.001000 Time 0.041132 +2023-10-02 20:41:34,012 - Epoch: [27][ 50/ 1236] Overall Loss 0.272177 Objective Loss 0.272177 LR 0.001000 Time 0.036998 +2023-10-02 20:41:34,220 - Epoch: [27][ 60/ 1236] Overall Loss 0.273812 Objective Loss 0.273812 LR 0.001000 Time 0.034291 +2023-10-02 20:41:34,425 - Epoch: [27][ 70/ 1236] Overall Loss 0.269601 Objective Loss 0.269601 LR 0.001000 Time 0.032312 +2023-10-02 20:41:34,633 - Epoch: [27][ 80/ 1236] Overall Loss 0.270107 Objective Loss 0.270107 LR 0.001000 Time 0.030875 +2023-10-02 20:41:34,838 - Epoch: [27][ 90/ 1236] Overall Loss 0.267606 Objective Loss 0.267606 LR 0.001000 Time 0.029712 +2023-10-02 20:41:35,046 - Epoch: [27][ 100/ 1236] Overall Loss 0.265350 Objective Loss 0.265350 LR 0.001000 Time 0.028820 +2023-10-02 20:41:35,251 - Epoch: [27][ 110/ 1236] Overall Loss 0.265191 Objective Loss 0.265191 LR 0.001000 Time 0.028061 +2023-10-02 20:41:35,459 - Epoch: [27][ 120/ 1236] Overall Loss 0.264995 Objective Loss 0.264995 LR 0.001000 Time 0.027451 +2023-10-02 20:41:35,663 - Epoch: [27][ 130/ 1236] Overall Loss 0.265180 Objective Loss 0.265180 LR 0.001000 Time 0.026910 +2023-10-02 20:41:35,870 - Epoch: [27][ 140/ 1236] Overall Loss 0.266020 Objective Loss 0.266020 LR 0.001000 Time 0.026468 +2023-10-02 20:41:36,075 - Epoch: [27][ 150/ 1236] Overall Loss 0.264715 Objective Loss 0.264715 LR 0.001000 Time 0.026066 +2023-10-02 20:41:36,283 - Epoch: [27][ 160/ 1236] Overall Loss 0.265123 Objective Loss 0.265123 LR 0.001000 Time 0.025733 +2023-10-02 20:41:36,487 - Epoch: [27][ 170/ 1236] Overall Loss 0.265769 Objective Loss 0.265769 LR 0.001000 Time 0.025418 +2023-10-02 20:41:36,694 - Epoch: [27][ 180/ 1236] Overall Loss 0.265764 Objective Loss 0.265764 LR 0.001000 Time 0.025155 +2023-10-02 20:41:36,899 - Epoch: [27][ 190/ 1236] Overall Loss 0.265061 Objective Loss 0.265061 LR 0.001000 Time 0.024908 +2023-10-02 20:41:37,106 - Epoch: [27][ 200/ 1236] Overall Loss 0.264133 Objective Loss 0.264133 LR 0.001000 Time 0.024699 +2023-10-02 20:41:37,311 - Epoch: [27][ 210/ 1236] Overall Loss 0.264617 Objective Loss 0.264617 LR 0.001000 Time 0.024496 +2023-10-02 20:41:37,518 - Epoch: [27][ 220/ 1236] Overall Loss 0.264618 Objective Loss 0.264618 LR 0.001000 Time 0.024324 +2023-10-02 20:41:37,723 - Epoch: [27][ 230/ 1236] Overall Loss 0.263780 Objective Loss 0.263780 LR 0.001000 Time 0.024155 +2023-10-02 20:41:37,930 - Epoch: [27][ 240/ 1236] Overall Loss 0.263529 Objective Loss 0.263529 LR 0.001000 Time 0.024011 +2023-10-02 20:41:38,135 - Epoch: [27][ 250/ 1236] Overall Loss 0.262490 Objective Loss 0.262490 LR 0.001000 Time 0.023868 +2023-10-02 20:41:38,342 - Epoch: [27][ 260/ 1236] Overall Loss 0.262706 Objective Loss 0.262706 LR 0.001000 Time 0.023746 +2023-10-02 20:41:38,547 - Epoch: [27][ 270/ 1236] Overall Loss 0.263260 Objective Loss 0.263260 LR 0.001000 Time 0.023623 +2023-10-02 20:41:38,754 - Epoch: [27][ 280/ 1236] Overall Loss 0.264898 Objective Loss 0.264898 LR 0.001000 Time 0.023519 +2023-10-02 20:41:38,959 - Epoch: [27][ 290/ 1236] Overall Loss 0.265308 Objective Loss 0.265308 LR 0.001000 Time 0.023413 +2023-10-02 20:41:39,167 - Epoch: [27][ 300/ 1236] Overall Loss 0.264061 Objective Loss 0.264061 LR 0.001000 Time 0.023324 +2023-10-02 20:41:39,372 - Epoch: [27][ 310/ 1236] Overall Loss 0.264771 Objective Loss 0.264771 LR 0.001000 Time 0.023231 +2023-10-02 20:41:39,578 - Epoch: [27][ 320/ 1236] Overall Loss 0.265313 Objective Loss 0.265313 LR 0.001000 Time 0.023151 +2023-10-02 20:41:39,783 - Epoch: [27][ 330/ 1236] Overall Loss 0.265157 Objective Loss 0.265157 LR 0.001000 Time 0.023068 +2023-10-02 20:41:39,991 - Epoch: [27][ 340/ 1236] Overall Loss 0.264854 Objective Loss 0.264854 LR 0.001000 Time 0.023001 +2023-10-02 20:41:40,196 - Epoch: [27][ 350/ 1236] Overall Loss 0.265587 Objective Loss 0.265587 LR 0.001000 Time 0.022929 +2023-10-02 20:41:40,404 - Epoch: [27][ 360/ 1236] Overall Loss 0.265076 Objective Loss 0.265076 LR 0.001000 Time 0.022868 +2023-10-02 20:41:40,609 - Epoch: [27][ 370/ 1236] Overall Loss 0.265367 Objective Loss 0.265367 LR 0.001000 Time 0.022805 +2023-10-02 20:41:40,817 - Epoch: [27][ 380/ 1236] Overall Loss 0.264274 Objective Loss 0.264274 LR 0.001000 Time 0.022751 +2023-10-02 20:41:41,023 - Epoch: [27][ 390/ 1236] Overall Loss 0.264112 Objective Loss 0.264112 LR 0.001000 Time 0.022694 +2023-10-02 20:41:41,231 - Epoch: [27][ 400/ 1236] Overall Loss 0.264447 Objective Loss 0.264447 LR 0.001000 Time 0.022647 +2023-10-02 20:41:41,437 - Epoch: [27][ 410/ 1236] Overall Loss 0.264941 Objective Loss 0.264941 LR 0.001000 Time 0.022595 +2023-10-02 20:41:41,645 - Epoch: [27][ 420/ 1236] Overall Loss 0.265125 Objective Loss 0.265125 LR 0.001000 Time 0.022551 +2023-10-02 20:41:41,850 - Epoch: [27][ 430/ 1236] Overall Loss 0.265159 Objective Loss 0.265159 LR 0.001000 Time 0.022504 +2023-10-02 20:41:42,058 - Epoch: [27][ 440/ 1236] Overall Loss 0.265240 Objective Loss 0.265240 LR 0.001000 Time 0.022465 +2023-10-02 20:41:42,264 - Epoch: [27][ 450/ 1236] Overall Loss 0.265115 Objective Loss 0.265115 LR 0.001000 Time 0.022422 +2023-10-02 20:41:42,472 - Epoch: [27][ 460/ 1236] Overall Loss 0.264985 Objective Loss 0.264985 LR 0.001000 Time 0.022386 +2023-10-02 20:41:42,677 - Epoch: [27][ 470/ 1236] Overall Loss 0.265736 Objective Loss 0.265736 LR 0.001000 Time 0.022346 +2023-10-02 20:41:42,886 - Epoch: [27][ 480/ 1236] Overall Loss 0.266484 Objective Loss 0.266484 LR 0.001000 Time 0.022315 +2023-10-02 20:41:43,091 - Epoch: [27][ 490/ 1236] Overall Loss 0.266894 Objective Loss 0.266894 LR 0.001000 Time 0.022278 +2023-10-02 20:41:43,300 - Epoch: [27][ 500/ 1236] Overall Loss 0.266902 Objective Loss 0.266902 LR 0.001000 Time 0.022248 +2023-10-02 20:41:43,505 - Epoch: [27][ 510/ 1236] Overall Loss 0.267521 Objective Loss 0.267521 LR 0.001000 Time 0.022215 +2023-10-02 20:41:43,713 - Epoch: [27][ 520/ 1236] Overall Loss 0.267876 Objective Loss 0.267876 LR 0.001000 Time 0.022187 +2023-10-02 20:41:43,919 - Epoch: [27][ 530/ 1236] Overall Loss 0.267667 Objective Loss 0.267667 LR 0.001000 Time 0.022156 +2023-10-02 20:41:44,127 - Epoch: [27][ 540/ 1236] Overall Loss 0.268082 Objective Loss 0.268082 LR 0.001000 Time 0.022131 +2023-10-02 20:41:44,333 - Epoch: [27][ 550/ 1236] Overall Loss 0.268293 Objective Loss 0.268293 LR 0.001000 Time 0.022102 +2023-10-02 20:41:44,540 - Epoch: [27][ 560/ 1236] Overall Loss 0.268697 Objective Loss 0.268697 LR 0.001000 Time 0.022076 +2023-10-02 20:41:44,747 - Epoch: [27][ 570/ 1236] Overall Loss 0.268565 Objective Loss 0.268565 LR 0.001000 Time 0.022049 +2023-10-02 20:41:44,955 - Epoch: [27][ 580/ 1236] Overall Loss 0.268777 Objective Loss 0.268777 LR 0.001000 Time 0.022027 +2023-10-02 20:41:45,160 - Epoch: [27][ 590/ 1236] Overall Loss 0.269172 Objective Loss 0.269172 LR 0.001000 Time 0.022002 +2023-10-02 20:41:45,369 - Epoch: [27][ 600/ 1236] Overall Loss 0.268910 Objective Loss 0.268910 LR 0.001000 Time 0.021982 +2023-10-02 20:41:45,574 - Epoch: [27][ 610/ 1236] Overall Loss 0.268990 Objective Loss 0.268990 LR 0.001000 Time 0.021958 +2023-10-02 20:41:45,783 - Epoch: [27][ 620/ 1236] Overall Loss 0.268967 Objective Loss 0.268967 LR 0.001000 Time 0.021939 +2023-10-02 20:41:45,989 - Epoch: [27][ 630/ 1236] Overall Loss 0.268857 Objective Loss 0.268857 LR 0.001000 Time 0.021917 +2023-10-02 20:41:46,197 - Epoch: [27][ 640/ 1236] Overall Loss 0.268920 Objective Loss 0.268920 LR 0.001000 Time 0.021901 +2023-10-02 20:41:46,403 - Epoch: [27][ 650/ 1236] Overall Loss 0.268825 Objective Loss 0.268825 LR 0.001000 Time 0.021879 +2023-10-02 20:41:46,611 - Epoch: [27][ 660/ 1236] Overall Loss 0.268811 Objective Loss 0.268811 LR 0.001000 Time 0.021863 +2023-10-02 20:41:46,817 - Epoch: [27][ 670/ 1236] Overall Loss 0.268915 Objective Loss 0.268915 LR 0.001000 Time 0.021843 +2023-10-02 20:41:47,025 - Epoch: [27][ 680/ 1236] Overall Loss 0.269062 Objective Loss 0.269062 LR 0.001000 Time 0.021828 +2023-10-02 20:41:47,231 - Epoch: [27][ 690/ 1236] Overall Loss 0.268943 Objective Loss 0.268943 LR 0.001000 Time 0.021809 +2023-10-02 20:41:47,439 - Epoch: [27][ 700/ 1236] Overall Loss 0.268896 Objective Loss 0.268896 LR 0.001000 Time 0.021795 +2023-10-02 20:41:47,644 - Epoch: [27][ 710/ 1236] Overall Loss 0.268856 Objective Loss 0.268856 LR 0.001000 Time 0.021777 +2023-10-02 20:41:47,853 - Epoch: [27][ 720/ 1236] Overall Loss 0.268531 Objective Loss 0.268531 LR 0.001000 Time 0.021763 +2023-10-02 20:41:48,058 - Epoch: [27][ 730/ 1236] Overall Loss 0.268710 Objective Loss 0.268710 LR 0.001000 Time 0.021746 +2023-10-02 20:41:48,267 - Epoch: [27][ 740/ 1236] Overall Loss 0.268570 Objective Loss 0.268570 LR 0.001000 Time 0.021734 +2023-10-02 20:41:48,472 - Epoch: [27][ 750/ 1236] Overall Loss 0.268349 Objective Loss 0.268349 LR 0.001000 Time 0.021718 +2023-10-02 20:41:48,680 - Epoch: [27][ 760/ 1236] Overall Loss 0.268546 Objective Loss 0.268546 LR 0.001000 Time 0.021705 +2023-10-02 20:41:48,886 - Epoch: [27][ 770/ 1236] Overall Loss 0.268479 Objective Loss 0.268479 LR 0.001000 Time 0.021690 +2023-10-02 20:41:49,094 - Epoch: [27][ 780/ 1236] Overall Loss 0.268504 Objective Loss 0.268504 LR 0.001000 Time 0.021678 +2023-10-02 20:41:49,300 - Epoch: [27][ 790/ 1236] Overall Loss 0.268664 Objective Loss 0.268664 LR 0.001000 Time 0.021664 +2023-10-02 20:41:49,508 - Epoch: [27][ 800/ 1236] Overall Loss 0.268958 Objective Loss 0.268958 LR 0.001000 Time 0.021653 +2023-10-02 20:41:49,714 - Epoch: [27][ 810/ 1236] Overall Loss 0.269221 Objective Loss 0.269221 LR 0.001000 Time 0.021639 +2023-10-02 20:41:49,922 - Epoch: [27][ 820/ 1236] Overall Loss 0.269328 Objective Loss 0.269328 LR 0.001000 Time 0.021629 +2023-10-02 20:41:50,128 - Epoch: [27][ 830/ 1236] Overall Loss 0.269590 Objective Loss 0.269590 LR 0.001000 Time 0.021616 +2023-10-02 20:41:50,336 - Epoch: [27][ 840/ 1236] Overall Loss 0.269872 Objective Loss 0.269872 LR 0.001000 Time 0.021606 +2023-10-02 20:41:50,542 - Epoch: [27][ 850/ 1236] Overall Loss 0.270125 Objective Loss 0.270125 LR 0.001000 Time 0.021593 +2023-10-02 20:41:50,750 - Epoch: [27][ 860/ 1236] Overall Loss 0.270200 Objective Loss 0.270200 LR 0.001000 Time 0.021584 +2023-10-02 20:41:50,955 - Epoch: [27][ 870/ 1236] Overall Loss 0.270375 Objective Loss 0.270375 LR 0.001000 Time 0.021572 +2023-10-02 20:41:51,164 - Epoch: [27][ 880/ 1236] Overall Loss 0.270815 Objective Loss 0.270815 LR 0.001000 Time 0.021564 +2023-10-02 20:41:51,369 - Epoch: [27][ 890/ 1236] Overall Loss 0.270937 Objective Loss 0.270937 LR 0.001000 Time 0.021551 +2023-10-02 20:41:51,578 - Epoch: [27][ 900/ 1236] Overall Loss 0.271325 Objective Loss 0.271325 LR 0.001000 Time 0.021544 +2023-10-02 20:41:51,783 - Epoch: [27][ 910/ 1236] Overall Loss 0.271602 Objective Loss 0.271602 LR 0.001000 Time 0.021532 +2023-10-02 20:41:51,992 - Epoch: [27][ 920/ 1236] Overall Loss 0.271616 Objective Loss 0.271616 LR 0.001000 Time 0.021525 +2023-10-02 20:41:52,197 - Epoch: [27][ 930/ 1236] Overall Loss 0.271784 Objective Loss 0.271784 LR 0.001000 Time 0.021513 +2023-10-02 20:41:52,406 - Epoch: [27][ 940/ 1236] Overall Loss 0.272147 Objective Loss 0.272147 LR 0.001000 Time 0.021506 +2023-10-02 20:41:52,611 - Epoch: [27][ 950/ 1236] Overall Loss 0.272033 Objective Loss 0.272033 LR 0.001000 Time 0.021496 +2023-10-02 20:41:52,819 - Epoch: [27][ 960/ 1236] Overall Loss 0.272099 Objective Loss 0.272099 LR 0.001000 Time 0.021488 +2023-10-02 20:41:53,025 - Epoch: [27][ 970/ 1236] Overall Loss 0.272033 Objective Loss 0.272033 LR 0.001000 Time 0.021478 +2023-10-02 20:41:53,233 - Epoch: [27][ 980/ 1236] Overall Loss 0.271939 Objective Loss 0.271939 LR 0.001000 Time 0.021471 +2023-10-02 20:41:53,438 - Epoch: [27][ 990/ 1236] Overall Loss 0.271812 Objective Loss 0.271812 LR 0.001000 Time 0.021461 +2023-10-02 20:41:53,646 - Epoch: [27][ 1000/ 1236] Overall Loss 0.272087 Objective Loss 0.272087 LR 0.001000 Time 0.021455 +2023-10-02 20:41:53,852 - Epoch: [27][ 1010/ 1236] Overall Loss 0.272058 Objective Loss 0.272058 LR 0.001000 Time 0.021445 +2023-10-02 20:41:54,060 - Epoch: [27][ 1020/ 1236] Overall Loss 0.272287 Objective Loss 0.272287 LR 0.001000 Time 0.021439 +2023-10-02 20:41:54,265 - Epoch: [27][ 1030/ 1236] Overall Loss 0.272398 Objective Loss 0.272398 LR 0.001000 Time 0.021430 +2023-10-02 20:41:54,474 - Epoch: [27][ 1040/ 1236] Overall Loss 0.272495 Objective Loss 0.272495 LR 0.001000 Time 0.021424 +2023-10-02 20:41:54,679 - Epoch: [27][ 1050/ 1236] Overall Loss 0.272660 Objective Loss 0.272660 LR 0.001000 Time 0.021415 +2023-10-02 20:41:54,888 - Epoch: [27][ 1060/ 1236] Overall Loss 0.272508 Objective Loss 0.272508 LR 0.001000 Time 0.021410 +2023-10-02 20:41:55,093 - Epoch: [27][ 1070/ 1236] Overall Loss 0.272654 Objective Loss 0.272654 LR 0.001000 Time 0.021401 +2023-10-02 20:41:55,302 - Epoch: [27][ 1080/ 1236] Overall Loss 0.272521 Objective Loss 0.272521 LR 0.001000 Time 0.021396 +2023-10-02 20:41:55,507 - Epoch: [27][ 1090/ 1236] Overall Loss 0.272504 Objective Loss 0.272504 LR 0.001000 Time 0.021388 +2023-10-02 20:41:55,716 - Epoch: [27][ 1100/ 1236] Overall Loss 0.272778 Objective Loss 0.272778 LR 0.001000 Time 0.021383 +2023-10-02 20:41:55,921 - Epoch: [27][ 1110/ 1236] Overall Loss 0.273150 Objective Loss 0.273150 LR 0.001000 Time 0.021375 +2023-10-02 20:41:56,130 - Epoch: [27][ 1120/ 1236] Overall Loss 0.273191 Objective Loss 0.273191 LR 0.001000 Time 0.021370 +2023-10-02 20:41:56,335 - Epoch: [27][ 1130/ 1236] Overall Loss 0.273357 Objective Loss 0.273357 LR 0.001000 Time 0.021362 +2023-10-02 20:41:56,544 - Epoch: [27][ 1140/ 1236] Overall Loss 0.273460 Objective Loss 0.273460 LR 0.001000 Time 0.021358 +2023-10-02 20:41:56,749 - Epoch: [27][ 1150/ 1236] Overall Loss 0.273668 Objective Loss 0.273668 LR 0.001000 Time 0.021350 +2023-10-02 20:41:56,958 - Epoch: [27][ 1160/ 1236] Overall Loss 0.273949 Objective Loss 0.273949 LR 0.001000 Time 0.021346 +2023-10-02 20:41:57,164 - Epoch: [27][ 1170/ 1236] Overall Loss 0.273745 Objective Loss 0.273745 LR 0.001000 Time 0.021339 +2023-10-02 20:41:57,373 - Epoch: [27][ 1180/ 1236] Overall Loss 0.273767 Objective Loss 0.273767 LR 0.001000 Time 0.021335 +2023-10-02 20:41:57,578 - Epoch: [27][ 1190/ 1236] Overall Loss 0.273911 Objective Loss 0.273911 LR 0.001000 Time 0.021328 +2023-10-02 20:41:57,787 - Epoch: [27][ 1200/ 1236] Overall Loss 0.273963 Objective Loss 0.273963 LR 0.001000 Time 0.021324 +2023-10-02 20:41:57,992 - Epoch: [27][ 1210/ 1236] Overall Loss 0.273900 Objective Loss 0.273900 LR 0.001000 Time 0.021317 +2023-10-02 20:41:58,201 - Epoch: [27][ 1220/ 1236] Overall Loss 0.273764 Objective Loss 0.273764 LR 0.001000 Time 0.021313 +2023-10-02 20:41:58,459 - Epoch: [27][ 1230/ 1236] Overall Loss 0.273383 Objective Loss 0.273383 LR 0.001000 Time 0.021350 +2023-10-02 20:41:58,581 - Epoch: [27][ 1236/ 1236] Overall Loss 0.273208 Objective Loss 0.273208 Top1 82.688391 Top5 97.759674 LR 0.001000 Time 0.021345 +2023-10-02 20:41:58,704 - --- validate (epoch=27)----------- +2023-10-02 20:41:58,704 - 29943 samples (256 per mini-batch) +2023-10-02 20:41:59,182 - Epoch: [27][ 10/ 117] Loss 0.297796 Top1 84.609375 Top5 98.515625 +2023-10-02 20:41:59,336 - Epoch: [27][ 20/ 117] Loss 0.304349 Top1 84.218750 Top5 98.359375 +2023-10-02 20:41:59,490 - Epoch: [27][ 30/ 117] Loss 0.313495 Top1 83.763021 Top5 98.216146 +2023-10-02 20:41:59,642 - Epoch: [27][ 40/ 117] Loss 0.321215 Top1 83.837891 Top5 98.291016 +2023-10-02 20:41:59,796 - Epoch: [27][ 50/ 117] Loss 0.330063 Top1 83.562500 Top5 98.265625 +2023-10-02 20:41:59,949 - Epoch: [27][ 60/ 117] Loss 0.329673 Top1 83.522135 Top5 98.287760 +2023-10-02 20:42:00,103 - Epoch: [27][ 70/ 117] Loss 0.328858 Top1 83.437500 Top5 98.297991 +2023-10-02 20:42:00,255 - Epoch: [27][ 80/ 117] Loss 0.328806 Top1 83.447266 Top5 98.281250 +2023-10-02 20:42:00,404 - Epoch: [27][ 90/ 117] Loss 0.327208 Top1 83.480903 Top5 98.255208 +2023-10-02 20:42:00,555 - Epoch: [27][ 100/ 117] Loss 0.326878 Top1 83.488281 Top5 98.261719 +2023-10-02 20:42:00,713 - Epoch: [27][ 110/ 117] Loss 0.325301 Top1 83.444602 Top5 98.259943 +2023-10-02 20:42:00,802 - Epoch: [27][ 117/ 117] Loss 0.322305 Top1 83.491968 Top5 98.283405 +2023-10-02 20:42:00,918 - ==> Top1: 83.492 Top5: 98.283 Loss: 0.322 + +2023-10-02 20:42:00,919 - ==> Confusion: +[[ 898 0 3 0 7 3 0 0 6 102 2 1 0 2 5 0 3 1 0 0 17] + [ 0 1045 4 3 6 32 1 7 1 0 0 0 0 1 3 2 8 0 8 4 6] + [ 6 0 955 22 3 0 10 15 0 2 2 4 6 1 0 3 0 1 11 5 10] + [ 1 2 19 982 1 0 1 4 5 1 4 0 4 3 32 0 2 0 12 0 16] + [ 21 5 2 0 947 9 1 1 0 17 0 2 2 3 16 2 17 0 2 2 1] + [ 2 34 1 4 4 981 1 28 1 4 0 9 1 13 4 2 4 0 5 4 14] + [ 0 7 42 1 0 2 1092 5 0 0 7 2 0 1 0 2 2 1 4 13 10] + [ 2 26 22 0 5 21 1 1058 1 2 2 6 0 4 2 0 1 0 49 8 8] + [ 14 2 0 0 1 0 0 1 980 35 8 2 1 12 22 1 0 1 5 1 3] + [ 59 1 1 0 1 0 0 0 40 976 3 0 0 19 6 1 3 0 0 2 7] + [ 5 0 10 20 1 2 1 5 17 1 941 3 2 17 3 1 1 3 14 0 6] + [ 1 1 2 0 0 13 0 7 0 1 0 963 10 7 0 3 2 15 0 4 6] + [ 0 0 1 1 1 2 1 3 2 1 0 63 940 2 0 7 5 13 8 7 11] + [ 1 0 4 0 2 11 0 1 14 12 4 8 0 1035 6 1 3 1 0 4 12] + [ 10 0 5 10 2 2 0 0 24 5 1 0 3 1 1010 0 2 1 17 1 7] + [ 0 0 4 1 6 0 0 0 0 3 0 9 7 1 0 1074 8 12 1 3 5] + [ 1 8 2 0 5 4 0 0 1 0 0 6 1 0 4 9 1102 0 2 2 14] + [ 0 0 2 7 0 0 1 0 1 0 1 10 17 2 2 10 2 973 3 2 5] + [ 2 6 5 17 0 2 0 20 7 0 3 0 0 0 6 0 0 0 988 0 12] + [ 0 2 4 2 1 7 3 14 1 0 2 12 4 1 0 0 3 0 2 1090 4] + [ 124 175 157 83 85 170 33 114 119 122 145 165 338 290 141 54 159 47 200 214 4970]] + +2023-10-02 20:42:00,920 - ==> Best [Top1: 83.492 Top5: 98.283 Sparsity:0.00 Params: 169472 on epoch: 27] +2023-10-02 20:42:00,920 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:42:00,927 - + +2023-10-02 20:42:00,927 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:42:01,923 - Epoch: [28][ 10/ 1236] Overall Loss 0.226898 Objective Loss 0.226898 LR 0.001000 Time 0.099559 +2023-10-02 20:42:02,130 - Epoch: [28][ 20/ 1236] Overall Loss 0.254091 Objective Loss 0.254091 LR 0.001000 Time 0.060105 +2023-10-02 20:42:02,336 - Epoch: [28][ 30/ 1236] Overall Loss 0.248111 Objective Loss 0.248111 LR 0.001000 Time 0.046929 +2023-10-02 20:42:02,544 - Epoch: [28][ 40/ 1236] Overall Loss 0.250484 Objective Loss 0.250484 LR 0.001000 Time 0.040370 +2023-10-02 20:42:02,748 - Epoch: [28][ 50/ 1236] Overall Loss 0.255414 Objective Loss 0.255414 LR 0.001000 Time 0.036388 +2023-10-02 20:42:02,956 - Epoch: [28][ 60/ 1236] Overall Loss 0.258673 Objective Loss 0.258673 LR 0.001000 Time 0.033780 +2023-10-02 20:42:03,161 - Epoch: [28][ 70/ 1236] Overall Loss 0.259909 Objective Loss 0.259909 LR 0.001000 Time 0.031872 +2023-10-02 20:42:03,368 - Epoch: [28][ 80/ 1236] Overall Loss 0.261977 Objective Loss 0.261977 LR 0.001000 Time 0.030479 +2023-10-02 20:42:03,573 - Epoch: [28][ 90/ 1236] Overall Loss 0.262871 Objective Loss 0.262871 LR 0.001000 Time 0.029365 +2023-10-02 20:42:03,780 - Epoch: [28][ 100/ 1236] Overall Loss 0.264046 Objective Loss 0.264046 LR 0.001000 Time 0.028498 +2023-10-02 20:42:03,984 - Epoch: [28][ 110/ 1236] Overall Loss 0.263512 Objective Loss 0.263512 LR 0.001000 Time 0.027762 +2023-10-02 20:42:04,192 - Epoch: [28][ 120/ 1236] Overall Loss 0.264855 Objective Loss 0.264855 LR 0.001000 Time 0.027176 +2023-10-02 20:42:04,397 - Epoch: [28][ 130/ 1236] Overall Loss 0.266749 Objective Loss 0.266749 LR 0.001000 Time 0.026656 +2023-10-02 20:42:04,604 - Epoch: [28][ 140/ 1236] Overall Loss 0.266363 Objective Loss 0.266363 LR 0.001000 Time 0.026234 +2023-10-02 20:42:04,809 - Epoch: [28][ 150/ 1236] Overall Loss 0.265620 Objective Loss 0.265620 LR 0.001000 Time 0.025848 +2023-10-02 20:42:05,016 - Epoch: [28][ 160/ 1236] Overall Loss 0.264746 Objective Loss 0.264746 LR 0.001000 Time 0.025527 +2023-10-02 20:42:05,221 - Epoch: [28][ 170/ 1236] Overall Loss 0.264390 Objective Loss 0.264390 LR 0.001000 Time 0.025225 +2023-10-02 20:42:05,429 - Epoch: [28][ 180/ 1236] Overall Loss 0.264703 Objective Loss 0.264703 LR 0.001000 Time 0.024978 +2023-10-02 20:42:05,633 - Epoch: [28][ 190/ 1236] Overall Loss 0.264834 Objective Loss 0.264834 LR 0.001000 Time 0.024739 +2023-10-02 20:42:05,841 - Epoch: [28][ 200/ 1236] Overall Loss 0.263544 Objective Loss 0.263544 LR 0.001000 Time 0.024540 +2023-10-02 20:42:06,046 - Epoch: [28][ 210/ 1236] Overall Loss 0.264861 Objective Loss 0.264861 LR 0.001000 Time 0.024344 +2023-10-02 20:42:06,254 - Epoch: [28][ 220/ 1236] Overall Loss 0.265268 Objective Loss 0.265268 LR 0.001000 Time 0.024182 +2023-10-02 20:42:06,459 - Epoch: [28][ 230/ 1236] Overall Loss 0.265057 Objective Loss 0.265057 LR 0.001000 Time 0.024019 +2023-10-02 20:42:06,666 - Epoch: [28][ 240/ 1236] Overall Loss 0.265965 Objective Loss 0.265965 LR 0.001000 Time 0.023883 +2023-10-02 20:42:06,871 - Epoch: [28][ 250/ 1236] Overall Loss 0.266305 Objective Loss 0.266305 LR 0.001000 Time 0.023747 +2023-10-02 20:42:07,080 - Epoch: [28][ 260/ 1236] Overall Loss 0.265441 Objective Loss 0.265441 LR 0.001000 Time 0.023635 +2023-10-02 20:42:07,286 - Epoch: [28][ 270/ 1236] Overall Loss 0.265311 Objective Loss 0.265311 LR 0.001000 Time 0.023519 +2023-10-02 20:42:07,496 - Epoch: [28][ 280/ 1236] Overall Loss 0.265014 Objective Loss 0.265014 LR 0.001000 Time 0.023425 +2023-10-02 20:42:07,701 - Epoch: [28][ 290/ 1236] Overall Loss 0.264714 Objective Loss 0.264714 LR 0.001000 Time 0.023324 +2023-10-02 20:42:07,910 - Epoch: [28][ 300/ 1236] Overall Loss 0.264255 Objective Loss 0.264255 LR 0.001000 Time 0.023243 +2023-10-02 20:42:08,116 - Epoch: [28][ 310/ 1236] Overall Loss 0.263279 Objective Loss 0.263279 LR 0.001000 Time 0.023157 +2023-10-02 20:42:08,324 - Epoch: [28][ 320/ 1236] Overall Loss 0.263214 Objective Loss 0.263214 LR 0.001000 Time 0.023083 +2023-10-02 20:42:08,531 - Epoch: [28][ 330/ 1236] Overall Loss 0.263579 Objective Loss 0.263579 LR 0.001000 Time 0.023005 +2023-10-02 20:42:08,740 - Epoch: [28][ 340/ 1236] Overall Loss 0.264478 Objective Loss 0.264478 LR 0.001000 Time 0.022944 +2023-10-02 20:42:08,946 - Epoch: [28][ 350/ 1236] Overall Loss 0.264362 Objective Loss 0.264362 LR 0.001000 Time 0.022875 +2023-10-02 20:42:09,155 - Epoch: [28][ 360/ 1236] Overall Loss 0.266256 Objective Loss 0.266256 LR 0.001000 Time 0.022820 +2023-10-02 20:42:09,361 - Epoch: [28][ 370/ 1236] Overall Loss 0.267061 Objective Loss 0.267061 LR 0.001000 Time 0.022759 +2023-10-02 20:42:09,570 - Epoch: [28][ 380/ 1236] Overall Loss 0.267662 Objective Loss 0.267662 LR 0.001000 Time 0.022710 +2023-10-02 20:42:09,776 - Epoch: [28][ 390/ 1236] Overall Loss 0.267926 Objective Loss 0.267926 LR 0.001000 Time 0.022655 +2023-10-02 20:42:09,987 - Epoch: [28][ 400/ 1236] Overall Loss 0.268990 Objective Loss 0.268990 LR 0.001000 Time 0.022614 +2023-10-02 20:42:10,193 - Epoch: [28][ 410/ 1236] Overall Loss 0.268213 Objective Loss 0.268213 LR 0.001000 Time 0.022564 +2023-10-02 20:42:10,402 - Epoch: [28][ 420/ 1236] Overall Loss 0.267843 Objective Loss 0.267843 LR 0.001000 Time 0.022525 +2023-10-02 20:42:10,608 - Epoch: [28][ 430/ 1236] Overall Loss 0.267470 Objective Loss 0.267470 LR 0.001000 Time 0.022479 +2023-10-02 20:42:10,816 - Epoch: [28][ 440/ 1236] Overall Loss 0.268305 Objective Loss 0.268305 LR 0.001000 Time 0.022441 +2023-10-02 20:42:11,023 - Epoch: [28][ 450/ 1236] Overall Loss 0.268206 Objective Loss 0.268206 LR 0.001000 Time 0.022401 +2023-10-02 20:42:11,233 - Epoch: [28][ 460/ 1236] Overall Loss 0.268286 Objective Loss 0.268286 LR 0.001000 Time 0.022369 +2023-10-02 20:42:11,438 - Epoch: [28][ 470/ 1236] Overall Loss 0.268312 Objective Loss 0.268312 LR 0.001000 Time 0.022330 +2023-10-02 20:42:11,648 - Epoch: [28][ 480/ 1236] Overall Loss 0.268677 Objective Loss 0.268677 LR 0.001000 Time 0.022300 +2023-10-02 20:42:11,852 - Epoch: [28][ 490/ 1236] Overall Loss 0.269062 Objective Loss 0.269062 LR 0.001000 Time 0.022262 +2023-10-02 20:42:12,061 - Epoch: [28][ 500/ 1236] Overall Loss 0.269121 Objective Loss 0.269121 LR 0.001000 Time 0.022233 +2023-10-02 20:42:12,268 - Epoch: [28][ 510/ 1236] Overall Loss 0.269061 Objective Loss 0.269061 LR 0.001000 Time 0.022202 +2023-10-02 20:42:12,477 - Epoch: [28][ 520/ 1236] Overall Loss 0.269155 Objective Loss 0.269155 LR 0.001000 Time 0.022178 +2023-10-02 20:42:12,683 - Epoch: [28][ 530/ 1236] Overall Loss 0.268991 Objective Loss 0.268991 LR 0.001000 Time 0.022147 +2023-10-02 20:42:12,893 - Epoch: [28][ 540/ 1236] Overall Loss 0.268966 Objective Loss 0.268966 LR 0.001000 Time 0.022125 +2023-10-02 20:42:13,098 - Epoch: [28][ 550/ 1236] Overall Loss 0.269020 Objective Loss 0.269020 LR 0.001000 Time 0.022096 +2023-10-02 20:42:13,308 - Epoch: [28][ 560/ 1236] Overall Loss 0.269499 Objective Loss 0.269499 LR 0.001000 Time 0.022076 +2023-10-02 20:42:13,514 - Epoch: [28][ 570/ 1236] Overall Loss 0.270118 Objective Loss 0.270118 LR 0.001000 Time 0.022049 +2023-10-02 20:42:13,722 - Epoch: [28][ 580/ 1236] Overall Loss 0.271005 Objective Loss 0.271005 LR 0.001000 Time 0.022028 +2023-10-02 20:42:13,929 - Epoch: [28][ 590/ 1236] Overall Loss 0.271034 Objective Loss 0.271034 LR 0.001000 Time 0.022005 +2023-10-02 20:42:14,139 - Epoch: [28][ 600/ 1236] Overall Loss 0.271250 Objective Loss 0.271250 LR 0.001000 Time 0.021987 +2023-10-02 20:42:14,345 - Epoch: [28][ 610/ 1236] Overall Loss 0.271365 Objective Loss 0.271365 LR 0.001000 Time 0.021963 +2023-10-02 20:42:14,553 - Epoch: [28][ 620/ 1236] Overall Loss 0.272053 Objective Loss 0.272053 LR 0.001000 Time 0.021944 +2023-10-02 20:42:14,760 - Epoch: [28][ 630/ 1236] Overall Loss 0.271727 Objective Loss 0.271727 LR 0.001000 Time 0.021925 +2023-10-02 20:42:14,970 - Epoch: [28][ 640/ 1236] Overall Loss 0.271078 Objective Loss 0.271078 LR 0.001000 Time 0.021910 +2023-10-02 20:42:15,176 - Epoch: [28][ 650/ 1236] Overall Loss 0.270865 Objective Loss 0.270865 LR 0.001000 Time 0.021889 +2023-10-02 20:42:15,385 - Epoch: [28][ 660/ 1236] Overall Loss 0.271365 Objective Loss 0.271365 LR 0.001000 Time 0.021873 +2023-10-02 20:42:15,592 - Epoch: [28][ 670/ 1236] Overall Loss 0.271320 Objective Loss 0.271320 LR 0.001000 Time 0.021853 +2023-10-02 20:42:15,800 - Epoch: [28][ 680/ 1236] Overall Loss 0.271678 Objective Loss 0.271678 LR 0.001000 Time 0.021837 +2023-10-02 20:42:16,007 - Epoch: [28][ 690/ 1236] Overall Loss 0.272229 Objective Loss 0.272229 LR 0.001000 Time 0.021821 +2023-10-02 20:42:16,217 - Epoch: [28][ 700/ 1236] Overall Loss 0.273057 Objective Loss 0.273057 LR 0.001000 Time 0.021808 +2023-10-02 20:42:16,423 - Epoch: [28][ 710/ 1236] Overall Loss 0.273477 Objective Loss 0.273477 LR 0.001000 Time 0.021791 +2023-10-02 20:42:16,632 - Epoch: [28][ 720/ 1236] Overall Loss 0.273321 Objective Loss 0.273321 LR 0.001000 Time 0.021779 +2023-10-02 20:42:16,838 - Epoch: [28][ 730/ 1236] Overall Loss 0.273066 Objective Loss 0.273066 LR 0.001000 Time 0.021762 +2023-10-02 20:42:17,046 - Epoch: [28][ 740/ 1236] Overall Loss 0.273396 Objective Loss 0.273396 LR 0.001000 Time 0.021749 +2023-10-02 20:42:17,254 - Epoch: [28][ 750/ 1236] Overall Loss 0.273253 Objective Loss 0.273253 LR 0.001000 Time 0.021733 +2023-10-02 20:42:17,462 - Epoch: [28][ 760/ 1236] Overall Loss 0.273294 Objective Loss 0.273294 LR 0.001000 Time 0.021721 +2023-10-02 20:42:17,669 - Epoch: [28][ 770/ 1236] Overall Loss 0.272935 Objective Loss 0.272935 LR 0.001000 Time 0.021707 +2023-10-02 20:42:17,879 - Epoch: [28][ 780/ 1236] Overall Loss 0.273295 Objective Loss 0.273295 LR 0.001000 Time 0.021698 +2023-10-02 20:42:18,085 - Epoch: [28][ 790/ 1236] Overall Loss 0.273454 Objective Loss 0.273454 LR 0.001000 Time 0.021683 +2023-10-02 20:42:18,294 - Epoch: [28][ 800/ 1236] Overall Loss 0.273513 Objective Loss 0.273513 LR 0.001000 Time 0.021673 +2023-10-02 20:42:18,501 - Epoch: [28][ 810/ 1236] Overall Loss 0.273452 Objective Loss 0.273452 LR 0.001000 Time 0.021661 +2023-10-02 20:42:18,709 - Epoch: [28][ 820/ 1236] Overall Loss 0.273127 Objective Loss 0.273127 LR 0.001000 Time 0.021650 +2023-10-02 20:42:18,916 - Epoch: [28][ 830/ 1236] Overall Loss 0.273695 Objective Loss 0.273695 LR 0.001000 Time 0.021638 +2023-10-02 20:42:19,125 - Epoch: [28][ 840/ 1236] Overall Loss 0.273464 Objective Loss 0.273464 LR 0.001000 Time 0.021630 +2023-10-02 20:42:19,331 - Epoch: [28][ 850/ 1236] Overall Loss 0.273702 Objective Loss 0.273702 LR 0.001000 Time 0.021617 +2023-10-02 20:42:19,540 - Epoch: [28][ 860/ 1236] Overall Loss 0.273910 Objective Loss 0.273910 LR 0.001000 Time 0.021608 +2023-10-02 20:42:19,747 - Epoch: [28][ 870/ 1236] Overall Loss 0.274162 Objective Loss 0.274162 LR 0.001000 Time 0.021597 +2023-10-02 20:42:19,955 - Epoch: [28][ 880/ 1236] Overall Loss 0.274042 Objective Loss 0.274042 LR 0.001000 Time 0.021588 +2023-10-02 20:42:20,162 - Epoch: [28][ 890/ 1236] Overall Loss 0.274117 Objective Loss 0.274117 LR 0.001000 Time 0.021578 +2023-10-02 20:42:20,371 - Epoch: [28][ 900/ 1236] Overall Loss 0.273814 Objective Loss 0.273814 LR 0.001000 Time 0.021569 +2023-10-02 20:42:20,578 - Epoch: [28][ 910/ 1236] Overall Loss 0.273880 Objective Loss 0.273880 LR 0.001000 Time 0.021559 +2023-10-02 20:42:20,787 - Epoch: [28][ 920/ 1236] Overall Loss 0.274202 Objective Loss 0.274202 LR 0.001000 Time 0.021551 +2023-10-02 20:42:20,994 - Epoch: [28][ 930/ 1236] Overall Loss 0.274235 Objective Loss 0.274235 LR 0.001000 Time 0.021542 +2023-10-02 20:42:21,203 - Epoch: [28][ 940/ 1236] Overall Loss 0.274461 Objective Loss 0.274461 LR 0.001000 Time 0.021534 +2023-10-02 20:42:21,410 - Epoch: [28][ 950/ 1236] Overall Loss 0.274632 Objective Loss 0.274632 LR 0.001000 Time 0.021525 +2023-10-02 20:42:21,619 - Epoch: [28][ 960/ 1236] Overall Loss 0.274800 Objective Loss 0.274800 LR 0.001000 Time 0.021518 +2023-10-02 20:42:21,826 - Epoch: [28][ 970/ 1236] Overall Loss 0.275164 Objective Loss 0.275164 LR 0.001000 Time 0.021510 +2023-10-02 20:42:22,035 - Epoch: [28][ 980/ 1236] Overall Loss 0.275296 Objective Loss 0.275296 LR 0.001000 Time 0.021503 +2023-10-02 20:42:22,240 - Epoch: [28][ 990/ 1236] Overall Loss 0.275258 Objective Loss 0.275258 LR 0.001000 Time 0.021493 +2023-10-02 20:42:22,449 - Epoch: [28][ 1000/ 1236] Overall Loss 0.275139 Objective Loss 0.275139 LR 0.001000 Time 0.021486 +2023-10-02 20:42:22,656 - Epoch: [28][ 1010/ 1236] Overall Loss 0.275419 Objective Loss 0.275419 LR 0.001000 Time 0.021479 +2023-10-02 20:42:22,865 - Epoch: [28][ 1020/ 1236] Overall Loss 0.275539 Objective Loss 0.275539 LR 0.001000 Time 0.021472 +2023-10-02 20:42:23,072 - Epoch: [28][ 1030/ 1236] Overall Loss 0.275584 Objective Loss 0.275584 LR 0.001000 Time 0.021465 +2023-10-02 20:42:23,281 - Epoch: [28][ 1040/ 1236] Overall Loss 0.275636 Objective Loss 0.275636 LR 0.001000 Time 0.021459 +2023-10-02 20:42:23,488 - Epoch: [28][ 1050/ 1236] Overall Loss 0.275707 Objective Loss 0.275707 LR 0.001000 Time 0.021451 +2023-10-02 20:42:23,696 - Epoch: [28][ 1060/ 1236] Overall Loss 0.275686 Objective Loss 0.275686 LR 0.001000 Time 0.021445 +2023-10-02 20:42:23,904 - Epoch: [28][ 1070/ 1236] Overall Loss 0.275724 Objective Loss 0.275724 LR 0.001000 Time 0.021438 +2023-10-02 20:42:24,112 - Epoch: [28][ 1080/ 1236] Overall Loss 0.275876 Objective Loss 0.275876 LR 0.001000 Time 0.021432 +2023-10-02 20:42:24,319 - Epoch: [28][ 1090/ 1236] Overall Loss 0.275749 Objective Loss 0.275749 LR 0.001000 Time 0.021426 +2023-10-02 20:42:24,528 - Epoch: [28][ 1100/ 1236] Overall Loss 0.275384 Objective Loss 0.275384 LR 0.001000 Time 0.021420 +2023-10-02 20:42:24,735 - Epoch: [28][ 1110/ 1236] Overall Loss 0.275596 Objective Loss 0.275596 LR 0.001000 Time 0.021414 +2023-10-02 20:42:24,943 - Epoch: [28][ 1120/ 1236] Overall Loss 0.275480 Objective Loss 0.275480 LR 0.001000 Time 0.021408 +2023-10-02 20:42:25,151 - Epoch: [28][ 1130/ 1236] Overall Loss 0.275461 Objective Loss 0.275461 LR 0.001000 Time 0.021402 +2023-10-02 20:42:25,359 - Epoch: [28][ 1140/ 1236] Overall Loss 0.275177 Objective Loss 0.275177 LR 0.001000 Time 0.021397 +2023-10-02 20:42:25,567 - Epoch: [28][ 1150/ 1236] Overall Loss 0.275083 Objective Loss 0.275083 LR 0.001000 Time 0.021391 +2023-10-02 20:42:25,775 - Epoch: [28][ 1160/ 1236] Overall Loss 0.274955 Objective Loss 0.274955 LR 0.001000 Time 0.021386 +2023-10-02 20:42:25,982 - Epoch: [28][ 1170/ 1236] Overall Loss 0.274877 Objective Loss 0.274877 LR 0.001000 Time 0.021380 +2023-10-02 20:42:26,191 - Epoch: [28][ 1180/ 1236] Overall Loss 0.274957 Objective Loss 0.274957 LR 0.001000 Time 0.021375 +2023-10-02 20:42:26,398 - Epoch: [28][ 1190/ 1236] Overall Loss 0.274897 Objective Loss 0.274897 LR 0.001000 Time 0.021369 +2023-10-02 20:42:26,607 - Epoch: [28][ 1200/ 1236] Overall Loss 0.274908 Objective Loss 0.274908 LR 0.001000 Time 0.021364 +2023-10-02 20:42:26,814 - Epoch: [28][ 1210/ 1236] Overall Loss 0.274802 Objective Loss 0.274802 LR 0.001000 Time 0.021359 +2023-10-02 20:42:27,023 - Epoch: [28][ 1220/ 1236] Overall Loss 0.274845 Objective Loss 0.274845 LR 0.001000 Time 0.021354 +2023-10-02 20:42:27,282 - Epoch: [28][ 1230/ 1236] Overall Loss 0.274838 Objective Loss 0.274838 LR 0.001000 Time 0.021390 +2023-10-02 20:42:27,403 - Epoch: [28][ 1236/ 1236] Overall Loss 0.274798 Objective Loss 0.274798 Top1 85.947047 Top5 98.778004 LR 0.001000 Time 0.021385 +2023-10-02 20:42:27,539 - --- validate (epoch=28)----------- +2023-10-02 20:42:27,539 - 29943 samples (256 per mini-batch) +2023-10-02 20:42:28,008 - Epoch: [28][ 10/ 117] Loss 0.342596 Top1 83.945312 Top5 97.851562 +2023-10-02 20:42:28,159 - Epoch: [28][ 20/ 117] Loss 0.333856 Top1 83.574219 Top5 98.007812 +2023-10-02 20:42:28,310 - Epoch: [28][ 30/ 117] Loss 0.345308 Top1 83.502604 Top5 98.059896 +2023-10-02 20:42:28,461 - Epoch: [28][ 40/ 117] Loss 0.352465 Top1 83.613281 Top5 97.919922 +2023-10-02 20:42:28,613 - Epoch: [28][ 50/ 117] Loss 0.340226 Top1 83.726562 Top5 98.015625 +2023-10-02 20:42:28,764 - Epoch: [28][ 60/ 117] Loss 0.347048 Top1 83.404948 Top5 97.981771 +2023-10-02 20:42:28,914 - Epoch: [28][ 70/ 117] Loss 0.344539 Top1 83.270089 Top5 98.030134 +2023-10-02 20:42:29,067 - Epoch: [28][ 80/ 117] Loss 0.347039 Top1 83.222656 Top5 98.051758 +2023-10-02 20:42:29,228 - Epoch: [28][ 90/ 117] Loss 0.342657 Top1 83.289931 Top5 98.090278 +2023-10-02 20:42:29,388 - Epoch: [28][ 100/ 117] Loss 0.339243 Top1 83.375000 Top5 98.148438 +2023-10-02 20:42:29,553 - Epoch: [28][ 110/ 117] Loss 0.336993 Top1 83.409091 Top5 98.135653 +2023-10-02 20:42:29,641 - Epoch: [28][ 117/ 117] Loss 0.336299 Top1 83.435194 Top5 98.136459 +2023-10-02 20:42:29,775 - ==> Top1: 83.435 Top5: 98.136 Loss: 0.336 + +2023-10-02 20:42:29,776 - ==> Confusion: +[[ 905 0 4 0 6 5 0 0 5 93 2 1 2 1 7 2 2 2 0 0 13] + [ 0 1058 3 1 8 19 1 10 2 2 0 3 1 0 3 4 0 0 8 1 7] + [ 5 1 961 7 4 1 28 9 1 3 0 1 5 5 2 4 0 1 3 4 11] + [ 1 4 25 924 2 5 3 2 2 0 4 0 9 6 56 3 1 8 15 0 19] + [ 23 1 1 0 976 2 0 0 0 9 1 0 1 2 16 5 6 1 0 3 3] + [ 3 53 4 0 2 981 0 15 1 9 1 5 2 17 6 2 0 0 2 3 10] + [ 0 4 24 0 0 1 1119 5 0 0 2 2 2 2 0 9 1 0 1 8 11] + [ 3 21 24 0 7 24 1 1046 1 0 2 8 5 5 4 2 0 1 50 7 7] + [ 21 4 0 1 1 1 0 0 939 48 12 1 0 10 36 4 1 4 3 0 3] + [ 99 1 0 0 3 1 1 0 18 956 1 1 0 12 12 1 1 0 0 3 9] + [ 2 4 8 10 3 1 5 3 15 2 945 2 2 15 5 1 0 5 17 0 8] + [ 0 0 3 0 0 16 0 1 0 3 0 941 31 2 0 4 2 21 0 4 7] + [ 1 2 3 1 0 1 1 0 0 0 1 49 952 2 6 11 1 25 0 3 9] + [ 1 0 1 0 4 6 2 0 10 22 5 4 3 1030 10 6 0 1 0 1 13] + [ 5 1 3 8 1 0 0 0 8 4 0 0 3 4 1044 0 0 6 5 0 9] + [ 0 1 1 1 5 0 2 0 0 0 0 8 8 0 0 1070 10 15 2 6 5] + [ 1 22 0 0 9 4 0 0 1 1 0 6 1 2 4 15 1081 0 0 3 11] + [ 0 0 0 0 0 0 2 0 1 0 0 4 21 1 2 5 0 1001 0 1 0] + [ 3 10 8 9 0 0 2 12 2 1 0 0 2 0 18 3 1 1 983 1 12] + [ 0 1 2 3 0 7 10 11 0 0 4 15 6 2 0 5 5 1 1 1067 12] + [ 128 249 151 37 134 151 50 69 106 99 123 143 357 293 236 90 65 80 204 136 5004]] + +2023-10-02 20:42:29,777 - ==> Best [Top1: 83.492 Top5: 98.283 Sparsity:0.00 Params: 169472 on epoch: 27] +2023-10-02 20:42:29,777 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:42:29,783 - + +2023-10-02 20:42:29,783 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:42:30,885 - Epoch: [29][ 10/ 1236] Overall Loss 0.246822 Objective Loss 0.246822 LR 0.001000 Time 0.110124 +2023-10-02 20:42:31,094 - Epoch: [29][ 20/ 1236] Overall Loss 0.250899 Objective Loss 0.250899 LR 0.001000 Time 0.065472 +2023-10-02 20:42:31,302 - Epoch: [29][ 30/ 1236] Overall Loss 0.251306 Objective Loss 0.251306 LR 0.001000 Time 0.050572 +2023-10-02 20:42:31,511 - Epoch: [29][ 40/ 1236] Overall Loss 0.252661 Objective Loss 0.252661 LR 0.001000 Time 0.043157 +2023-10-02 20:42:31,718 - Epoch: [29][ 50/ 1236] Overall Loss 0.258060 Objective Loss 0.258060 LR 0.001000 Time 0.038651 +2023-10-02 20:42:31,928 - Epoch: [29][ 60/ 1236] Overall Loss 0.252371 Objective Loss 0.252371 LR 0.001000 Time 0.035701 +2023-10-02 20:42:32,135 - Epoch: [29][ 70/ 1236] Overall Loss 0.251319 Objective Loss 0.251319 LR 0.001000 Time 0.033552 +2023-10-02 20:42:32,345 - Epoch: [29][ 80/ 1236] Overall Loss 0.254163 Objective Loss 0.254163 LR 0.001000 Time 0.031980 +2023-10-02 20:42:32,552 - Epoch: [29][ 90/ 1236] Overall Loss 0.258781 Objective Loss 0.258781 LR 0.001000 Time 0.030721 +2023-10-02 20:42:32,761 - Epoch: [29][ 100/ 1236] Overall Loss 0.258514 Objective Loss 0.258514 LR 0.001000 Time 0.029736 +2023-10-02 20:42:32,973 - Epoch: [29][ 110/ 1236] Overall Loss 0.260200 Objective Loss 0.260200 LR 0.001000 Time 0.028955 +2023-10-02 20:42:33,186 - Epoch: [29][ 120/ 1236] Overall Loss 0.261964 Objective Loss 0.261964 LR 0.001000 Time 0.028300 +2023-10-02 20:42:33,397 - Epoch: [29][ 130/ 1236] Overall Loss 0.263137 Objective Loss 0.263137 LR 0.001000 Time 0.027741 +2023-10-02 20:42:33,609 - Epoch: [29][ 140/ 1236] Overall Loss 0.264301 Objective Loss 0.264301 LR 0.001000 Time 0.027269 +2023-10-02 20:42:33,820 - Epoch: [29][ 150/ 1236] Overall Loss 0.262166 Objective Loss 0.262166 LR 0.001000 Time 0.026850 +2023-10-02 20:42:34,032 - Epoch: [29][ 160/ 1236] Overall Loss 0.261875 Objective Loss 0.261875 LR 0.001000 Time 0.026487 +2023-10-02 20:42:34,244 - Epoch: [29][ 170/ 1236] Overall Loss 0.261539 Objective Loss 0.261539 LR 0.001000 Time 0.026168 +2023-10-02 20:42:34,457 - Epoch: [29][ 180/ 1236] Overall Loss 0.265048 Objective Loss 0.265048 LR 0.001000 Time 0.025889 +2023-10-02 20:42:34,668 - Epoch: [29][ 190/ 1236] Overall Loss 0.266667 Objective Loss 0.266667 LR 0.001000 Time 0.025633 +2023-10-02 20:42:34,881 - Epoch: [29][ 200/ 1236] Overall Loss 0.267291 Objective Loss 0.267291 LR 0.001000 Time 0.025409 +2023-10-02 20:42:35,091 - Epoch: [29][ 210/ 1236] Overall Loss 0.268043 Objective Loss 0.268043 LR 0.001000 Time 0.025198 +2023-10-02 20:42:35,304 - Epoch: [29][ 220/ 1236] Overall Loss 0.267658 Objective Loss 0.267658 LR 0.001000 Time 0.025010 +2023-10-02 20:42:35,513 - Epoch: [29][ 230/ 1236] Overall Loss 0.268520 Objective Loss 0.268520 LR 0.001000 Time 0.024831 +2023-10-02 20:42:35,725 - Epoch: [29][ 240/ 1236] Overall Loss 0.267674 Objective Loss 0.267674 LR 0.001000 Time 0.024671 +2023-10-02 20:42:35,934 - Epoch: [29][ 250/ 1236] Overall Loss 0.267403 Objective Loss 0.267403 LR 0.001000 Time 0.024519 +2023-10-02 20:42:36,145 - Epoch: [29][ 260/ 1236] Overall Loss 0.267808 Objective Loss 0.267808 LR 0.001000 Time 0.024384 +2023-10-02 20:42:36,355 - Epoch: [29][ 270/ 1236] Overall Loss 0.268552 Objective Loss 0.268552 LR 0.001000 Time 0.024254 +2023-10-02 20:42:36,567 - Epoch: [29][ 280/ 1236] Overall Loss 0.268935 Objective Loss 0.268935 LR 0.001000 Time 0.024138 +2023-10-02 20:42:36,776 - Epoch: [29][ 290/ 1236] Overall Loss 0.269354 Objective Loss 0.269354 LR 0.001000 Time 0.024026 +2023-10-02 20:42:36,986 - Epoch: [29][ 300/ 1236] Overall Loss 0.269299 Objective Loss 0.269299 LR 0.001000 Time 0.023919 +2023-10-02 20:42:37,192 - Epoch: [29][ 310/ 1236] Overall Loss 0.269208 Objective Loss 0.269208 LR 0.001000 Time 0.023812 +2023-10-02 20:42:37,402 - Epoch: [29][ 320/ 1236] Overall Loss 0.269425 Objective Loss 0.269425 LR 0.001000 Time 0.023724 +2023-10-02 20:42:37,608 - Epoch: [29][ 330/ 1236] Overall Loss 0.269569 Objective Loss 0.269569 LR 0.001000 Time 0.023629 +2023-10-02 20:42:37,819 - Epoch: [29][ 340/ 1236] Overall Loss 0.269736 Objective Loss 0.269736 LR 0.001000 Time 0.023552 +2023-10-02 20:42:38,026 - Epoch: [29][ 350/ 1236] Overall Loss 0.269332 Objective Loss 0.269332 LR 0.001000 Time 0.023470 +2023-10-02 20:42:38,236 - Epoch: [29][ 360/ 1236] Overall Loss 0.269507 Objective Loss 0.269507 LR 0.001000 Time 0.023401 +2023-10-02 20:42:38,442 - Epoch: [29][ 370/ 1236] Overall Loss 0.269057 Objective Loss 0.269057 LR 0.001000 Time 0.023325 +2023-10-02 20:42:38,651 - Epoch: [29][ 380/ 1236] Overall Loss 0.269507 Objective Loss 0.269507 LR 0.001000 Time 0.023260 +2023-10-02 20:42:38,857 - Epoch: [29][ 390/ 1236] Overall Loss 0.268968 Objective Loss 0.268968 LR 0.001000 Time 0.023191 +2023-10-02 20:42:39,064 - Epoch: [29][ 400/ 1236] Overall Loss 0.268526 Objective Loss 0.268526 LR 0.001000 Time 0.023129 +2023-10-02 20:42:39,268 - Epoch: [29][ 410/ 1236] Overall Loss 0.268723 Objective Loss 0.268723 LR 0.001000 Time 0.023061 +2023-10-02 20:42:39,475 - Epoch: [29][ 420/ 1236] Overall Loss 0.269176 Objective Loss 0.269176 LR 0.001000 Time 0.023004 +2023-10-02 20:42:39,679 - Epoch: [29][ 430/ 1236] Overall Loss 0.269217 Objective Loss 0.269217 LR 0.001000 Time 0.022943 +2023-10-02 20:42:39,885 - Epoch: [29][ 440/ 1236] Overall Loss 0.269576 Objective Loss 0.269576 LR 0.001000 Time 0.022889 +2023-10-02 20:42:40,090 - Epoch: [29][ 450/ 1236] Overall Loss 0.269656 Objective Loss 0.269656 LR 0.001000 Time 0.022835 +2023-10-02 20:42:40,297 - Epoch: [29][ 460/ 1236] Overall Loss 0.269976 Objective Loss 0.269976 LR 0.001000 Time 0.022788 +2023-10-02 20:42:40,501 - Epoch: [29][ 470/ 1236] Overall Loss 0.270233 Objective Loss 0.270233 LR 0.001000 Time 0.022737 +2023-10-02 20:42:40,707 - Epoch: [29][ 480/ 1236] Overall Loss 0.270047 Objective Loss 0.270047 LR 0.001000 Time 0.022690 +2023-10-02 20:42:40,912 - Epoch: [29][ 490/ 1236] Overall Loss 0.269951 Objective Loss 0.269951 LR 0.001000 Time 0.022645 +2023-10-02 20:42:41,117 - Epoch: [29][ 500/ 1236] Overall Loss 0.270217 Objective Loss 0.270217 LR 0.001000 Time 0.022603 +2023-10-02 20:42:41,323 - Epoch: [29][ 510/ 1236] Overall Loss 0.270302 Objective Loss 0.270302 LR 0.001000 Time 0.022562 +2023-10-02 20:42:41,528 - Epoch: [29][ 520/ 1236] Overall Loss 0.270694 Objective Loss 0.270694 LR 0.001000 Time 0.022523 +2023-10-02 20:42:41,733 - Epoch: [29][ 530/ 1236] Overall Loss 0.270460 Objective Loss 0.270460 LR 0.001000 Time 0.022484 +2023-10-02 20:42:41,939 - Epoch: [29][ 540/ 1236] Overall Loss 0.270810 Objective Loss 0.270810 LR 0.001000 Time 0.022449 +2023-10-02 20:42:42,144 - Epoch: [29][ 550/ 1236] Overall Loss 0.270294 Objective Loss 0.270294 LR 0.001000 Time 0.022413 +2023-10-02 20:42:42,350 - Epoch: [29][ 560/ 1236] Overall Loss 0.270322 Objective Loss 0.270322 LR 0.001000 Time 0.022380 +2023-10-02 20:42:42,555 - Epoch: [29][ 570/ 1236] Overall Loss 0.270323 Objective Loss 0.270323 LR 0.001000 Time 0.022347 +2023-10-02 20:42:42,761 - Epoch: [29][ 580/ 1236] Overall Loss 0.270428 Objective Loss 0.270428 LR 0.001000 Time 0.022315 +2023-10-02 20:42:42,966 - Epoch: [29][ 590/ 1236] Overall Loss 0.271011 Objective Loss 0.271011 LR 0.001000 Time 0.022284 +2023-10-02 20:42:43,173 - Epoch: [29][ 600/ 1236] Overall Loss 0.270962 Objective Loss 0.270962 LR 0.001000 Time 0.022258 +2023-10-02 20:42:43,377 - Epoch: [29][ 610/ 1236] Overall Loss 0.270760 Objective Loss 0.270760 LR 0.001000 Time 0.022227 +2023-10-02 20:42:43,583 - Epoch: [29][ 620/ 1236] Overall Loss 0.270631 Objective Loss 0.270631 LR 0.001000 Time 0.022200 +2023-10-02 20:42:43,788 - Epoch: [29][ 630/ 1236] Overall Loss 0.270906 Objective Loss 0.270906 LR 0.001000 Time 0.022173 +2023-10-02 20:42:43,994 - Epoch: [29][ 640/ 1236] Overall Loss 0.270950 Objective Loss 0.270950 LR 0.001000 Time 0.022147 +2023-10-02 20:42:44,199 - Epoch: [29][ 650/ 1236] Overall Loss 0.270934 Objective Loss 0.270934 LR 0.001000 Time 0.022121 +2023-10-02 20:42:44,405 - Epoch: [29][ 660/ 1236] Overall Loss 0.271074 Objective Loss 0.271074 LR 0.001000 Time 0.022098 +2023-10-02 20:42:44,610 - Epoch: [29][ 670/ 1236] Overall Loss 0.271003 Objective Loss 0.271003 LR 0.001000 Time 0.022073 +2023-10-02 20:42:44,815 - Epoch: [29][ 680/ 1236] Overall Loss 0.271203 Objective Loss 0.271203 LR 0.001000 Time 0.022051 +2023-10-02 20:42:45,021 - Epoch: [29][ 690/ 1236] Overall Loss 0.271541 Objective Loss 0.271541 LR 0.001000 Time 0.022026 +2023-10-02 20:42:45,228 - Epoch: [29][ 700/ 1236] Overall Loss 0.272048 Objective Loss 0.272048 LR 0.001000 Time 0.022007 +2023-10-02 20:42:45,432 - Epoch: [29][ 710/ 1236] Overall Loss 0.272000 Objective Loss 0.272000 LR 0.001000 Time 0.021984 +2023-10-02 20:42:45,639 - Epoch: [29][ 720/ 1236] Overall Loss 0.272084 Objective Loss 0.272084 LR 0.001000 Time 0.021966 +2023-10-02 20:42:45,843 - Epoch: [29][ 730/ 1236] Overall Loss 0.272017 Objective Loss 0.272017 LR 0.001000 Time 0.021944 +2023-10-02 20:42:46,048 - Epoch: [29][ 740/ 1236] Overall Loss 0.272203 Objective Loss 0.272203 LR 0.001000 Time 0.021925 +2023-10-02 20:42:46,253 - Epoch: [29][ 750/ 1236] Overall Loss 0.272315 Objective Loss 0.272315 LR 0.001000 Time 0.021906 +2023-10-02 20:42:46,460 - Epoch: [29][ 760/ 1236] Overall Loss 0.272523 Objective Loss 0.272523 LR 0.001000 Time 0.021889 +2023-10-02 20:42:46,664 - Epoch: [29][ 770/ 1236] Overall Loss 0.272661 Objective Loss 0.272661 LR 0.001000 Time 0.021869 +2023-10-02 20:42:46,871 - Epoch: [29][ 780/ 1236] Overall Loss 0.272509 Objective Loss 0.272509 LR 0.001000 Time 0.021854 +2023-10-02 20:42:47,075 - Epoch: [29][ 790/ 1236] Overall Loss 0.272390 Objective Loss 0.272390 LR 0.001000 Time 0.021835 +2023-10-02 20:42:47,281 - Epoch: [29][ 800/ 1236] Overall Loss 0.271972 Objective Loss 0.271972 LR 0.001000 Time 0.021819 +2023-10-02 20:42:47,486 - Epoch: [29][ 810/ 1236] Overall Loss 0.272008 Objective Loss 0.272008 LR 0.001000 Time 0.021803 +2023-10-02 20:42:47,691 - Epoch: [29][ 820/ 1236] Overall Loss 0.272151 Objective Loss 0.272151 LR 0.001000 Time 0.021787 +2023-10-02 20:42:47,897 - Epoch: [29][ 830/ 1236] Overall Loss 0.271931 Objective Loss 0.271931 LR 0.001000 Time 0.021772 +2023-10-02 20:42:48,104 - Epoch: [29][ 840/ 1236] Overall Loss 0.271902 Objective Loss 0.271902 LR 0.001000 Time 0.021759 +2023-10-02 20:42:48,308 - Epoch: [29][ 850/ 1236] Overall Loss 0.271836 Objective Loss 0.271836 LR 0.001000 Time 0.021742 +2023-10-02 20:42:48,514 - Epoch: [29][ 860/ 1236] Overall Loss 0.272105 Objective Loss 0.272105 LR 0.001000 Time 0.021728 +2023-10-02 20:42:48,719 - Epoch: [29][ 870/ 1236] Overall Loss 0.272275 Objective Loss 0.272275 LR 0.001000 Time 0.021714 +2023-10-02 20:42:48,926 - Epoch: [29][ 880/ 1236] Overall Loss 0.271742 Objective Loss 0.271742 LR 0.001000 Time 0.021702 +2023-10-02 20:42:49,130 - Epoch: [29][ 890/ 1236] Overall Loss 0.271717 Objective Loss 0.271717 LR 0.001000 Time 0.021687 +2023-10-02 20:42:49,337 - Epoch: [29][ 900/ 1236] Overall Loss 0.271416 Objective Loss 0.271416 LR 0.001000 Time 0.021676 +2023-10-02 20:42:49,541 - Epoch: [29][ 910/ 1236] Overall Loss 0.271557 Objective Loss 0.271557 LR 0.001000 Time 0.021662 +2023-10-02 20:42:49,746 - Epoch: [29][ 920/ 1236] Overall Loss 0.271971 Objective Loss 0.271971 LR 0.001000 Time 0.021649 +2023-10-02 20:42:49,951 - Epoch: [29][ 930/ 1236] Overall Loss 0.272163 Objective Loss 0.272163 LR 0.001000 Time 0.021637 +2023-10-02 20:42:50,157 - Epoch: [29][ 940/ 1236] Overall Loss 0.272297 Objective Loss 0.272297 LR 0.001000 Time 0.021625 +2023-10-02 20:42:50,362 - Epoch: [29][ 950/ 1236] Overall Loss 0.272288 Objective Loss 0.272288 LR 0.001000 Time 0.021613 +2023-10-02 20:42:50,570 - Epoch: [29][ 960/ 1236] Overall Loss 0.272425 Objective Loss 0.272425 LR 0.001000 Time 0.021604 +2023-10-02 20:42:50,774 - Epoch: [29][ 970/ 1236] Overall Loss 0.272403 Objective Loss 0.272403 LR 0.001000 Time 0.021591 +2023-10-02 20:42:50,979 - Epoch: [29][ 980/ 1236] Overall Loss 0.272318 Objective Loss 0.272318 LR 0.001000 Time 0.021580 +2023-10-02 20:42:51,184 - Epoch: [29][ 990/ 1236] Overall Loss 0.272523 Objective Loss 0.272523 LR 0.001000 Time 0.021568 +2023-10-02 20:42:51,390 - Epoch: [29][ 1000/ 1236] Overall Loss 0.272170 Objective Loss 0.272170 LR 0.001000 Time 0.021558 +2023-10-02 20:42:51,595 - Epoch: [29][ 1010/ 1236] Overall Loss 0.272025 Objective Loss 0.272025 LR 0.001000 Time 0.021547 +2023-10-02 20:42:51,801 - Epoch: [29][ 1020/ 1236] Overall Loss 0.272054 Objective Loss 0.272054 LR 0.001000 Time 0.021537 +2023-10-02 20:42:52,006 - Epoch: [29][ 1030/ 1236] Overall Loss 0.271747 Objective Loss 0.271747 LR 0.001000 Time 0.021527 +2023-10-02 20:42:52,212 - Epoch: [29][ 1040/ 1236] Overall Loss 0.271615 Objective Loss 0.271615 LR 0.001000 Time 0.021518 +2023-10-02 20:42:52,417 - Epoch: [29][ 1050/ 1236] Overall Loss 0.271815 Objective Loss 0.271815 LR 0.001000 Time 0.021508 +2023-10-02 20:42:52,623 - Epoch: [29][ 1060/ 1236] Overall Loss 0.272075 Objective Loss 0.272075 LR 0.001000 Time 0.021499 +2023-10-02 20:42:52,828 - Epoch: [29][ 1070/ 1236] Overall Loss 0.271899 Objective Loss 0.271899 LR 0.001000 Time 0.021490 +2023-10-02 20:42:53,035 - Epoch: [29][ 1080/ 1236] Overall Loss 0.271989 Objective Loss 0.271989 LR 0.001000 Time 0.021482 +2023-10-02 20:42:53,239 - Epoch: [29][ 1090/ 1236] Overall Loss 0.271885 Objective Loss 0.271885 LR 0.001000 Time 0.021472 +2023-10-02 20:42:53,446 - Epoch: [29][ 1100/ 1236] Overall Loss 0.271878 Objective Loss 0.271878 LR 0.001000 Time 0.021465 +2023-10-02 20:42:53,650 - Epoch: [29][ 1110/ 1236] Overall Loss 0.272036 Objective Loss 0.272036 LR 0.001000 Time 0.021455 +2023-10-02 20:42:53,857 - Epoch: [29][ 1120/ 1236] Overall Loss 0.272188 Objective Loss 0.272188 LR 0.001000 Time 0.021448 +2023-10-02 20:42:54,061 - Epoch: [29][ 1130/ 1236] Overall Loss 0.272397 Objective Loss 0.272397 LR 0.001000 Time 0.021438 +2023-10-02 20:42:54,268 - Epoch: [29][ 1140/ 1236] Overall Loss 0.272821 Objective Loss 0.272821 LR 0.001000 Time 0.021431 +2023-10-02 20:42:54,472 - Epoch: [29][ 1150/ 1236] Overall Loss 0.273296 Objective Loss 0.273296 LR 0.001000 Time 0.021422 +2023-10-02 20:42:54,678 - Epoch: [29][ 1160/ 1236] Overall Loss 0.272992 Objective Loss 0.272992 LR 0.001000 Time 0.021415 +2023-10-02 20:42:54,883 - Epoch: [29][ 1170/ 1236] Overall Loss 0.273432 Objective Loss 0.273432 LR 0.001000 Time 0.021407 +2023-10-02 20:42:55,091 - Epoch: [29][ 1180/ 1236] Overall Loss 0.273536 Objective Loss 0.273536 LR 0.001000 Time 0.021401 +2023-10-02 20:42:55,295 - Epoch: [29][ 1190/ 1236] Overall Loss 0.273817 Objective Loss 0.273817 LR 0.001000 Time 0.021392 +2023-10-02 20:42:55,502 - Epoch: [29][ 1200/ 1236] Overall Loss 0.273695 Objective Loss 0.273695 LR 0.001000 Time 0.021386 +2023-10-02 20:42:55,706 - Epoch: [29][ 1210/ 1236] Overall Loss 0.273757 Objective Loss 0.273757 LR 0.001000 Time 0.021378 +2023-10-02 20:42:55,912 - Epoch: [29][ 1220/ 1236] Overall Loss 0.273989 Objective Loss 0.273989 LR 0.001000 Time 0.021371 +2023-10-02 20:42:56,169 - Epoch: [29][ 1230/ 1236] Overall Loss 0.274006 Objective Loss 0.274006 LR 0.001000 Time 0.021407 +2023-10-02 20:42:56,289 - Epoch: [29][ 1236/ 1236] Overall Loss 0.274107 Objective Loss 0.274107 Top1 85.743381 Top5 97.759674 LR 0.001000 Time 0.021400 +2023-10-02 20:42:56,410 - --- validate (epoch=29)----------- +2023-10-02 20:42:56,410 - 29943 samples (256 per mini-batch) +2023-10-02 20:42:56,868 - Epoch: [29][ 10/ 117] Loss 0.326659 Top1 82.851562 Top5 98.476562 +2023-10-02 20:42:57,019 - Epoch: [29][ 20/ 117] Loss 0.324747 Top1 83.398438 Top5 98.398438 +2023-10-02 20:42:57,167 - Epoch: [29][ 30/ 117] Loss 0.312444 Top1 83.671875 Top5 98.203125 +2023-10-02 20:42:57,318 - Epoch: [29][ 40/ 117] Loss 0.321636 Top1 83.349609 Top5 98.134766 +2023-10-02 20:42:57,468 - Epoch: [29][ 50/ 117] Loss 0.322021 Top1 83.390625 Top5 98.078125 +2023-10-02 20:42:57,618 - Epoch: [29][ 60/ 117] Loss 0.326476 Top1 83.255208 Top5 98.014323 +2023-10-02 20:42:57,767 - Epoch: [29][ 70/ 117] Loss 0.322151 Top1 83.247768 Top5 98.030134 +2023-10-02 20:42:57,918 - Epoch: [29][ 80/ 117] Loss 0.322714 Top1 83.300781 Top5 98.037109 +2023-10-02 20:42:58,067 - Epoch: [29][ 90/ 117] Loss 0.323393 Top1 83.333333 Top5 98.051215 +2023-10-02 20:42:58,218 - Epoch: [29][ 100/ 117] Loss 0.323760 Top1 83.265625 Top5 98.089844 +2023-10-02 20:42:58,373 - Epoch: [29][ 110/ 117] Loss 0.324873 Top1 83.398438 Top5 98.100142 +2023-10-02 20:42:58,461 - Epoch: [29][ 117/ 117] Loss 0.324606 Top1 83.435194 Top5 98.096383 +2023-10-02 20:42:58,607 - ==> Top1: 83.435 Top5: 98.096 Loss: 0.325 + +2023-10-02 20:42:58,608 - ==> Confusion: +[[ 933 0 5 0 11 5 0 0 4 63 2 1 0 0 4 1 6 0 0 0 15] + [ 1 1072 2 1 3 14 2 12 3 1 1 0 0 0 0 4 3 1 6 2 3] + [ 6 2 974 9 0 0 21 8 0 0 0 5 0 3 0 8 0 1 10 1 8] + [ 0 2 15 989 1 0 0 2 3 0 4 0 4 4 27 1 0 2 17 0 18] + [ 24 8 3 0 960 7 0 0 1 14 0 2 0 3 8 3 10 0 0 1 6] + [ 1 69 0 2 3 938 3 29 2 4 4 10 3 14 5 2 4 0 5 4 14] + [ 0 4 41 0 0 0 1101 7 0 0 3 2 1 3 0 12 0 0 2 6 9] + [ 3 25 27 0 0 20 6 1055 0 0 1 11 2 5 2 5 0 1 40 6 9] + [ 19 2 0 0 2 1 0 1 955 49 15 2 0 15 17 4 4 0 2 0 1] + [ 86 1 1 1 6 1 1 0 20 954 1 0 0 22 8 2 1 1 1 2 10] + [ 1 2 11 12 3 0 1 5 12 1 955 5 1 18 3 2 1 3 10 0 7] + [ 0 0 2 0 0 12 0 5 0 0 0 947 36 3 0 4 1 16 0 4 5] + [ 0 2 4 5 0 1 1 4 0 0 2 40 958 0 1 12 1 21 1 4 11] + [ 1 0 2 0 1 9 1 2 10 16 6 6 1 1041 4 0 0 2 0 1 16] + [ 10 4 1 12 5 0 0 0 28 3 3 1 4 4 1004 0 2 0 12 0 8] + [ 0 1 3 1 6 0 0 0 0 0 0 5 6 1 1 1067 16 13 3 4 7] + [ 2 11 2 0 5 7 0 1 1 0 0 13 0 1 2 8 1090 1 0 5 12] + [ 0 0 2 6 0 0 1 0 1 0 0 3 27 0 2 10 2 976 1 1 6] + [ 1 9 11 8 0 0 0 16 8 0 5 0 2 0 10 1 1 0 979 1 16] + [ 0 1 9 1 0 4 11 15 1 1 1 22 4 2 0 3 5 0 2 1060 10] + [ 132 292 163 87 96 115 23 105 116 98 168 165 318 276 126 89 142 64 158 197 4975]] + +2023-10-02 20:42:58,609 - ==> Best [Top1: 83.492 Top5: 98.283 Sparsity:0.00 Params: 169472 on epoch: 27] +2023-10-02 20:42:58,609 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:42:58,615 - + +2023-10-02 20:42:58,615 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:42:59,609 - Epoch: [30][ 10/ 1236] Overall Loss 0.255917 Objective Loss 0.255917 LR 0.001000 Time 0.099316 +2023-10-02 20:42:59,816 - Epoch: [30][ 20/ 1236] Overall Loss 0.260315 Objective Loss 0.260315 LR 0.001000 Time 0.059966 +2023-10-02 20:43:00,021 - Epoch: [30][ 30/ 1236] Overall Loss 0.260369 Objective Loss 0.260369 LR 0.001000 Time 0.046792 +2023-10-02 20:43:00,230 - Epoch: [30][ 40/ 1236] Overall Loss 0.259566 Objective Loss 0.259566 LR 0.001000 Time 0.040287 +2023-10-02 20:43:00,435 - Epoch: [30][ 50/ 1236] Overall Loss 0.263047 Objective Loss 0.263047 LR 0.001000 Time 0.036327 +2023-10-02 20:43:00,643 - Epoch: [30][ 60/ 1236] Overall Loss 0.264413 Objective Loss 0.264413 LR 0.001000 Time 0.033735 +2023-10-02 20:43:00,848 - Epoch: [30][ 70/ 1236] Overall Loss 0.263293 Objective Loss 0.263293 LR 0.001000 Time 0.031841 +2023-10-02 20:43:01,056 - Epoch: [30][ 80/ 1236] Overall Loss 0.266372 Objective Loss 0.266372 LR 0.001000 Time 0.030460 +2023-10-02 20:43:01,261 - Epoch: [30][ 90/ 1236] Overall Loss 0.267533 Objective Loss 0.267533 LR 0.001000 Time 0.029352 +2023-10-02 20:43:01,469 - Epoch: [30][ 100/ 1236] Overall Loss 0.265481 Objective Loss 0.265481 LR 0.001000 Time 0.028496 +2023-10-02 20:43:01,674 - Epoch: [30][ 110/ 1236] Overall Loss 0.263851 Objective Loss 0.263851 LR 0.001000 Time 0.027763 +2023-10-02 20:43:01,882 - Epoch: [30][ 120/ 1236] Overall Loss 0.263345 Objective Loss 0.263345 LR 0.001000 Time 0.027178 +2023-10-02 20:43:02,087 - Epoch: [30][ 130/ 1236] Overall Loss 0.263051 Objective Loss 0.263051 LR 0.001000 Time 0.026663 +2023-10-02 20:43:02,295 - Epoch: [30][ 140/ 1236] Overall Loss 0.263190 Objective Loss 0.263190 LR 0.001000 Time 0.026245 +2023-10-02 20:43:02,500 - Epoch: [30][ 150/ 1236] Overall Loss 0.263482 Objective Loss 0.263482 LR 0.001000 Time 0.025858 +2023-10-02 20:43:02,708 - Epoch: [30][ 160/ 1236] Overall Loss 0.265433 Objective Loss 0.265433 LR 0.001000 Time 0.025538 +2023-10-02 20:43:02,912 - Epoch: [30][ 170/ 1236] Overall Loss 0.266541 Objective Loss 0.266541 LR 0.001000 Time 0.025237 +2023-10-02 20:43:03,120 - Epoch: [30][ 180/ 1236] Overall Loss 0.267424 Objective Loss 0.267424 LR 0.001000 Time 0.024988 +2023-10-02 20:43:03,325 - Epoch: [30][ 190/ 1236] Overall Loss 0.268018 Objective Loss 0.268018 LR 0.001000 Time 0.024748 +2023-10-02 20:43:03,532 - Epoch: [30][ 200/ 1236] Overall Loss 0.265887 Objective Loss 0.265887 LR 0.001000 Time 0.024548 +2023-10-02 20:43:03,737 - Epoch: [30][ 210/ 1236] Overall Loss 0.266515 Objective Loss 0.266515 LR 0.001000 Time 0.024352 +2023-10-02 20:43:03,945 - Epoch: [30][ 220/ 1236] Overall Loss 0.267110 Objective Loss 0.267110 LR 0.001000 Time 0.024188 +2023-10-02 20:43:04,149 - Epoch: [30][ 230/ 1236] Overall Loss 0.266351 Objective Loss 0.266351 LR 0.001000 Time 0.024024 +2023-10-02 20:43:04,357 - Epoch: [30][ 240/ 1236] Overall Loss 0.264829 Objective Loss 0.264829 LR 0.001000 Time 0.023887 +2023-10-02 20:43:04,561 - Epoch: [30][ 250/ 1236] Overall Loss 0.264185 Objective Loss 0.264185 LR 0.001000 Time 0.023748 +2023-10-02 20:43:04,769 - Epoch: [30][ 260/ 1236] Overall Loss 0.266171 Objective Loss 0.266171 LR 0.001000 Time 0.023633 +2023-10-02 20:43:04,974 - Epoch: [30][ 270/ 1236] Overall Loss 0.266145 Objective Loss 0.266145 LR 0.001000 Time 0.023515 +2023-10-02 20:43:05,181 - Epoch: [30][ 280/ 1236] Overall Loss 0.266553 Objective Loss 0.266553 LR 0.001000 Time 0.023415 +2023-10-02 20:43:05,386 - Epoch: [30][ 290/ 1236] Overall Loss 0.266036 Objective Loss 0.266036 LR 0.001000 Time 0.023313 +2023-10-02 20:43:05,594 - Epoch: [30][ 300/ 1236] Overall Loss 0.266290 Objective Loss 0.266290 LR 0.001000 Time 0.023228 +2023-10-02 20:43:05,799 - Epoch: [30][ 310/ 1236] Overall Loss 0.265509 Objective Loss 0.265509 LR 0.001000 Time 0.023138 +2023-10-02 20:43:06,007 - Epoch: [30][ 320/ 1236] Overall Loss 0.265144 Objective Loss 0.265144 LR 0.001000 Time 0.023065 +2023-10-02 20:43:06,212 - Epoch: [30][ 330/ 1236] Overall Loss 0.265387 Objective Loss 0.265387 LR 0.001000 Time 0.022986 +2023-10-02 20:43:06,420 - Epoch: [30][ 340/ 1236] Overall Loss 0.265124 Objective Loss 0.265124 LR 0.001000 Time 0.022920 +2023-10-02 20:43:06,624 - Epoch: [30][ 350/ 1236] Overall Loss 0.264898 Objective Loss 0.264898 LR 0.001000 Time 0.022848 +2023-10-02 20:43:06,832 - Epoch: [30][ 360/ 1236] Overall Loss 0.265671 Objective Loss 0.265671 LR 0.001000 Time 0.022791 +2023-10-02 20:43:07,037 - Epoch: [30][ 370/ 1236] Overall Loss 0.264953 Objective Loss 0.264953 LR 0.001000 Time 0.022728 +2023-10-02 20:43:07,245 - Epoch: [30][ 380/ 1236] Overall Loss 0.264380 Objective Loss 0.264380 LR 0.001000 Time 0.022677 +2023-10-02 20:43:07,450 - Epoch: [30][ 390/ 1236] Overall Loss 0.264039 Objective Loss 0.264039 LR 0.001000 Time 0.022621 +2023-10-02 20:43:07,659 - Epoch: [30][ 400/ 1236] Overall Loss 0.264243 Objective Loss 0.264243 LR 0.001000 Time 0.022576 +2023-10-02 20:43:07,864 - Epoch: [30][ 410/ 1236] Overall Loss 0.264296 Objective Loss 0.264296 LR 0.001000 Time 0.022525 +2023-10-02 20:43:08,073 - Epoch: [30][ 420/ 1236] Overall Loss 0.264445 Objective Loss 0.264445 LR 0.001000 Time 0.022485 +2023-10-02 20:43:08,278 - Epoch: [30][ 430/ 1236] Overall Loss 0.264149 Objective Loss 0.264149 LR 0.001000 Time 0.022438 +2023-10-02 20:43:08,486 - Epoch: [30][ 440/ 1236] Overall Loss 0.263591 Objective Loss 0.263591 LR 0.001000 Time 0.022401 +2023-10-02 20:43:08,691 - Epoch: [30][ 450/ 1236] Overall Loss 0.263394 Objective Loss 0.263394 LR 0.001000 Time 0.022359 +2023-10-02 20:43:08,900 - Epoch: [30][ 460/ 1236] Overall Loss 0.262600 Objective Loss 0.262600 LR 0.001000 Time 0.022325 +2023-10-02 20:43:09,105 - Epoch: [30][ 470/ 1236] Overall Loss 0.262665 Objective Loss 0.262665 LR 0.001000 Time 0.022286 +2023-10-02 20:43:09,314 - Epoch: [30][ 480/ 1236] Overall Loss 0.262483 Objective Loss 0.262483 LR 0.001000 Time 0.022257 +2023-10-02 20:43:09,520 - Epoch: [30][ 490/ 1236] Overall Loss 0.261869 Objective Loss 0.261869 LR 0.001000 Time 0.022221 +2023-10-02 20:43:09,727 - Epoch: [30][ 500/ 1236] Overall Loss 0.262024 Objective Loss 0.262024 LR 0.001000 Time 0.022191 +2023-10-02 20:43:09,934 - Epoch: [30][ 510/ 1236] Overall Loss 0.261530 Objective Loss 0.261530 LR 0.001000 Time 0.022159 +2023-10-02 20:43:10,143 - Epoch: [30][ 520/ 1236] Overall Loss 0.261533 Objective Loss 0.261533 LR 0.001000 Time 0.022134 +2023-10-02 20:43:10,349 - Epoch: [30][ 530/ 1236] Overall Loss 0.261195 Objective Loss 0.261195 LR 0.001000 Time 0.022104 +2023-10-02 20:43:10,556 - Epoch: [30][ 540/ 1236] Overall Loss 0.261455 Objective Loss 0.261455 LR 0.001000 Time 0.022078 +2023-10-02 20:43:10,763 - Epoch: [30][ 550/ 1236] Overall Loss 0.261630 Objective Loss 0.261630 LR 0.001000 Time 0.022051 +2023-10-02 20:43:10,971 - Epoch: [30][ 560/ 1236] Overall Loss 0.261974 Objective Loss 0.261974 LR 0.001000 Time 0.022027 +2023-10-02 20:43:11,178 - Epoch: [30][ 570/ 1236] Overall Loss 0.262408 Objective Loss 0.262408 LR 0.001000 Time 0.022001 +2023-10-02 20:43:11,386 - Epoch: [30][ 580/ 1236] Overall Loss 0.262932 Objective Loss 0.262932 LR 0.001000 Time 0.021980 +2023-10-02 20:43:11,592 - Epoch: [30][ 590/ 1236] Overall Loss 0.263556 Objective Loss 0.263556 LR 0.001000 Time 0.021955 +2023-10-02 20:43:11,800 - Epoch: [30][ 600/ 1236] Overall Loss 0.263494 Objective Loss 0.263494 LR 0.001000 Time 0.021935 +2023-10-02 20:43:12,007 - Epoch: [30][ 610/ 1236] Overall Loss 0.264424 Objective Loss 0.264424 LR 0.001000 Time 0.021912 +2023-10-02 20:43:12,216 - Epoch: [30][ 620/ 1236] Overall Loss 0.264663 Objective Loss 0.264663 LR 0.001000 Time 0.021894 +2023-10-02 20:43:12,421 - Epoch: [30][ 630/ 1236] Overall Loss 0.264908 Objective Loss 0.264908 LR 0.001000 Time 0.021873 +2023-10-02 20:43:12,630 - Epoch: [30][ 640/ 1236] Overall Loss 0.265404 Objective Loss 0.265404 LR 0.001000 Time 0.021857 +2023-10-02 20:43:12,836 - Epoch: [30][ 650/ 1236] Overall Loss 0.265083 Objective Loss 0.265083 LR 0.001000 Time 0.021837 +2023-10-02 20:43:13,043 - Epoch: [30][ 660/ 1236] Overall Loss 0.264976 Objective Loss 0.264976 LR 0.001000 Time 0.021819 +2023-10-02 20:43:13,250 - Epoch: [30][ 670/ 1236] Overall Loss 0.264497 Objective Loss 0.264497 LR 0.001000 Time 0.021800 +2023-10-02 20:43:13,458 - Epoch: [30][ 680/ 1236] Overall Loss 0.264540 Objective Loss 0.264540 LR 0.001000 Time 0.021785 +2023-10-02 20:43:13,665 - Epoch: [30][ 690/ 1236] Overall Loss 0.264351 Objective Loss 0.264351 LR 0.001000 Time 0.021766 +2023-10-02 20:43:13,873 - Epoch: [30][ 700/ 1236] Overall Loss 0.264312 Objective Loss 0.264312 LR 0.001000 Time 0.021753 +2023-10-02 20:43:14,079 - Epoch: [30][ 710/ 1236] Overall Loss 0.264214 Objective Loss 0.264214 LR 0.001000 Time 0.021736 +2023-10-02 20:43:14,286 - Epoch: [30][ 720/ 1236] Overall Loss 0.263604 Objective Loss 0.263604 LR 0.001000 Time 0.021722 +2023-10-02 20:43:14,493 - Epoch: [30][ 730/ 1236] Overall Loss 0.263875 Objective Loss 0.263875 LR 0.001000 Time 0.021706 +2023-10-02 20:43:14,701 - Epoch: [30][ 740/ 1236] Overall Loss 0.263830 Objective Loss 0.263830 LR 0.001000 Time 0.021692 +2023-10-02 20:43:14,908 - Epoch: [30][ 750/ 1236] Overall Loss 0.263955 Objective Loss 0.263955 LR 0.001000 Time 0.021677 +2023-10-02 20:43:15,115 - Epoch: [30][ 760/ 1236] Overall Loss 0.264271 Objective Loss 0.264271 LR 0.001000 Time 0.021664 +2023-10-02 20:43:15,322 - Epoch: [30][ 770/ 1236] Overall Loss 0.263775 Objective Loss 0.263775 LR 0.001000 Time 0.021650 +2023-10-02 20:43:15,529 - Epoch: [30][ 780/ 1236] Overall Loss 0.263410 Objective Loss 0.263410 LR 0.001000 Time 0.021638 +2023-10-02 20:43:15,737 - Epoch: [30][ 790/ 1236] Overall Loss 0.263621 Objective Loss 0.263621 LR 0.001000 Time 0.021624 +2023-10-02 20:43:15,944 - Epoch: [30][ 800/ 1236] Overall Loss 0.263359 Objective Loss 0.263359 LR 0.001000 Time 0.021613 +2023-10-02 20:43:16,152 - Epoch: [30][ 810/ 1236] Overall Loss 0.263368 Objective Loss 0.263368 LR 0.001000 Time 0.021600 +2023-10-02 20:43:16,359 - Epoch: [30][ 820/ 1236] Overall Loss 0.263450 Objective Loss 0.263450 LR 0.001000 Time 0.021589 +2023-10-02 20:43:16,566 - Epoch: [30][ 830/ 1236] Overall Loss 0.263823 Objective Loss 0.263823 LR 0.001000 Time 0.021577 +2023-10-02 20:43:16,773 - Epoch: [30][ 840/ 1236] Overall Loss 0.264224 Objective Loss 0.264224 LR 0.001000 Time 0.021566 +2023-10-02 20:43:16,981 - Epoch: [30][ 850/ 1236] Overall Loss 0.264326 Objective Loss 0.264326 LR 0.001000 Time 0.021555 +2023-10-02 20:43:17,188 - Epoch: [30][ 860/ 1236] Overall Loss 0.264357 Objective Loss 0.264357 LR 0.001000 Time 0.021545 +2023-10-02 20:43:17,395 - Epoch: [30][ 870/ 1236] Overall Loss 0.264408 Objective Loss 0.264408 LR 0.001000 Time 0.021533 +2023-10-02 20:43:17,602 - Epoch: [30][ 880/ 1236] Overall Loss 0.264168 Objective Loss 0.264168 LR 0.001000 Time 0.021524 +2023-10-02 20:43:17,809 - Epoch: [30][ 890/ 1236] Overall Loss 0.264057 Objective Loss 0.264057 LR 0.001000 Time 0.021513 +2023-10-02 20:43:18,017 - Epoch: [30][ 900/ 1236] Overall Loss 0.264128 Objective Loss 0.264128 LR 0.001000 Time 0.021504 +2023-10-02 20:43:18,224 - Epoch: [30][ 910/ 1236] Overall Loss 0.264247 Objective Loss 0.264247 LR 0.001000 Time 0.021494 +2023-10-02 20:43:18,431 - Epoch: [30][ 920/ 1236] Overall Loss 0.264264 Objective Loss 0.264264 LR 0.001000 Time 0.021485 +2023-10-02 20:43:18,638 - Epoch: [30][ 930/ 1236] Overall Loss 0.264565 Objective Loss 0.264565 LR 0.001000 Time 0.021475 +2023-10-02 20:43:18,846 - Epoch: [30][ 940/ 1236] Overall Loss 0.264682 Objective Loss 0.264682 LR 0.001000 Time 0.021467 +2023-10-02 20:43:19,053 - Epoch: [30][ 950/ 1236] Overall Loss 0.264545 Objective Loss 0.264545 LR 0.001000 Time 0.021457 +2023-10-02 20:43:19,260 - Epoch: [30][ 960/ 1236] Overall Loss 0.264816 Objective Loss 0.264816 LR 0.001000 Time 0.021450 +2023-10-02 20:43:19,468 - Epoch: [30][ 970/ 1236] Overall Loss 0.264957 Objective Loss 0.264957 LR 0.001000 Time 0.021441 +2023-10-02 20:43:19,675 - Epoch: [30][ 980/ 1236] Overall Loss 0.264741 Objective Loss 0.264741 LR 0.001000 Time 0.021433 +2023-10-02 20:43:19,882 - Epoch: [30][ 990/ 1236] Overall Loss 0.264698 Objective Loss 0.264698 LR 0.001000 Time 0.021424 +2023-10-02 20:43:20,089 - Epoch: [30][ 1000/ 1236] Overall Loss 0.264571 Objective Loss 0.264571 LR 0.001000 Time 0.021417 +2023-10-02 20:43:20,296 - Epoch: [30][ 1010/ 1236] Overall Loss 0.264741 Objective Loss 0.264741 LR 0.001000 Time 0.021409 +2023-10-02 20:43:20,505 - Epoch: [30][ 1020/ 1236] Overall Loss 0.264889 Objective Loss 0.264889 LR 0.001000 Time 0.021403 +2023-10-02 20:43:20,711 - Epoch: [30][ 1030/ 1236] Overall Loss 0.265132 Objective Loss 0.265132 LR 0.001000 Time 0.021395 +2023-10-02 20:43:20,919 - Epoch: [30][ 1040/ 1236] Overall Loss 0.265389 Objective Loss 0.265389 LR 0.001000 Time 0.021389 +2023-10-02 20:43:21,125 - Epoch: [30][ 1050/ 1236] Overall Loss 0.265367 Objective Loss 0.265367 LR 0.001000 Time 0.021382 +2023-10-02 20:43:21,333 - Epoch: [30][ 1060/ 1236] Overall Loss 0.265329 Objective Loss 0.265329 LR 0.001000 Time 0.021375 +2023-10-02 20:43:21,540 - Epoch: [30][ 1070/ 1236] Overall Loss 0.265570 Objective Loss 0.265570 LR 0.001000 Time 0.021368 +2023-10-02 20:43:21,748 - Epoch: [30][ 1080/ 1236] Overall Loss 0.265583 Objective Loss 0.265583 LR 0.001000 Time 0.021363 +2023-10-02 20:43:21,954 - Epoch: [30][ 1090/ 1236] Overall Loss 0.265541 Objective Loss 0.265541 LR 0.001000 Time 0.021355 +2023-10-02 20:43:22,162 - Epoch: [30][ 1100/ 1236] Overall Loss 0.265634 Objective Loss 0.265634 LR 0.001000 Time 0.021349 +2023-10-02 20:43:22,369 - Epoch: [30][ 1110/ 1236] Overall Loss 0.265878 Objective Loss 0.265878 LR 0.001000 Time 0.021343 +2023-10-02 20:43:22,577 - Epoch: [30][ 1120/ 1236] Overall Loss 0.265955 Objective Loss 0.265955 LR 0.001000 Time 0.021337 +2023-10-02 20:43:22,784 - Epoch: [30][ 1130/ 1236] Overall Loss 0.265917 Objective Loss 0.265917 LR 0.001000 Time 0.021330 +2023-10-02 20:43:22,991 - Epoch: [30][ 1140/ 1236] Overall Loss 0.265984 Objective Loss 0.265984 LR 0.001000 Time 0.021325 +2023-10-02 20:43:23,198 - Epoch: [30][ 1150/ 1236] Overall Loss 0.266371 Objective Loss 0.266371 LR 0.001000 Time 0.021318 +2023-10-02 20:43:23,406 - Epoch: [30][ 1160/ 1236] Overall Loss 0.266510 Objective Loss 0.266510 LR 0.001000 Time 0.021313 +2023-10-02 20:43:23,613 - Epoch: [30][ 1170/ 1236] Overall Loss 0.266526 Objective Loss 0.266526 LR 0.001000 Time 0.021306 +2023-10-02 20:43:23,821 - Epoch: [30][ 1180/ 1236] Overall Loss 0.266751 Objective Loss 0.266751 LR 0.001000 Time 0.021302 +2023-10-02 20:43:24,028 - Epoch: [30][ 1190/ 1236] Overall Loss 0.267055 Objective Loss 0.267055 LR 0.001000 Time 0.021296 +2023-10-02 20:43:24,235 - Epoch: [30][ 1200/ 1236] Overall Loss 0.267034 Objective Loss 0.267034 LR 0.001000 Time 0.021292 +2023-10-02 20:43:24,442 - Epoch: [30][ 1210/ 1236] Overall Loss 0.266927 Objective Loss 0.266927 LR 0.001000 Time 0.021286 +2023-10-02 20:43:24,649 - Epoch: [30][ 1220/ 1236] Overall Loss 0.266716 Objective Loss 0.266716 LR 0.001000 Time 0.021281 +2023-10-02 20:43:24,910 - Epoch: [30][ 1230/ 1236] Overall Loss 0.266761 Objective Loss 0.266761 LR 0.001000 Time 0.021319 +2023-10-02 20:43:25,032 - Epoch: [30][ 1236/ 1236] Overall Loss 0.266675 Objective Loss 0.266675 Top1 85.132383 Top5 97.963340 LR 0.001000 Time 0.021313 +2023-10-02 20:43:25,182 - --- validate (epoch=30)----------- +2023-10-02 20:43:25,182 - 29943 samples (256 per mini-batch) +2023-10-02 20:43:25,665 - Epoch: [30][ 10/ 117] Loss 0.345639 Top1 81.562500 Top5 97.968750 +2023-10-02 20:43:25,816 - Epoch: [30][ 20/ 117] Loss 0.348209 Top1 81.933594 Top5 97.890625 +2023-10-02 20:43:25,968 - Epoch: [30][ 30/ 117] Loss 0.331150 Top1 82.382812 Top5 97.981771 +2023-10-02 20:43:26,117 - Epoch: [30][ 40/ 117] Loss 0.324520 Top1 82.480469 Top5 98.007812 +2023-10-02 20:43:26,268 - Epoch: [30][ 50/ 117] Loss 0.327784 Top1 82.390625 Top5 97.976562 +2023-10-02 20:43:26,416 - Epoch: [30][ 60/ 117] Loss 0.327008 Top1 82.350260 Top5 97.923177 +2023-10-02 20:43:26,566 - Epoch: [30][ 70/ 117] Loss 0.326509 Top1 82.405134 Top5 97.963170 +2023-10-02 20:43:26,716 - Epoch: [30][ 80/ 117] Loss 0.325668 Top1 82.412109 Top5 98.032227 +2023-10-02 20:43:26,865 - Epoch: [30][ 90/ 117] Loss 0.324183 Top1 82.382812 Top5 98.020833 +2023-10-02 20:43:27,013 - Epoch: [30][ 100/ 117] Loss 0.328212 Top1 82.316406 Top5 97.996094 +2023-10-02 20:43:27,170 - Epoch: [30][ 110/ 117] Loss 0.327295 Top1 82.365057 Top5 98.046875 +2023-10-02 20:43:27,258 - Epoch: [30][ 117/ 117] Loss 0.325049 Top1 82.386534 Top5 98.019571 +2023-10-02 20:43:27,399 - ==> Top1: 82.387 Top5: 98.020 Loss: 0.325 + +2023-10-02 20:43:27,400 - ==> Confusion: +[[ 935 1 2 0 6 3 0 0 11 66 2 1 0 1 2 1 4 3 0 0 12] + [ 0 1035 0 1 1 44 2 27 2 1 0 1 0 1 0 5 4 0 3 1 3] + [ 9 0 968 8 2 0 17 9 0 2 7 2 3 4 0 6 2 2 4 6 5] + [ 2 1 13 972 0 5 1 2 2 1 14 1 6 5 32 0 2 5 10 1 14] + [ 27 14 2 0 933 9 0 2 1 9 1 1 0 5 9 7 22 0 0 3 5] + [ 3 34 2 2 0 966 0 29 1 5 4 16 2 30 3 0 3 1 3 3 9] + [ 1 7 33 1 0 0 1111 9 0 0 4 2 0 0 1 10 1 0 1 5 5] + [ 3 17 23 0 1 31 5 1077 0 1 6 10 2 5 0 0 0 1 24 7 5] + [ 17 2 0 1 0 3 0 0 1000 30 6 1 0 4 12 3 3 3 2 2 0] + [ 92 0 0 0 3 0 0 1 63 921 1 0 0 17 5 1 2 1 0 5 7] + [ 1 3 8 10 0 2 0 2 18 1 963 4 1 17 3 1 0 2 5 0 12] + [ 0 0 1 0 0 12 0 2 0 0 0 948 23 11 0 4 5 19 0 7 3] + [ 1 0 2 2 0 3 1 2 0 0 2 40 948 2 4 16 5 30 1 3 6] + [ 3 0 1 0 1 3 0 0 22 14 11 5 0 1036 4 2 3 1 0 1 12] + [ 11 1 3 13 1 1 0 0 35 3 4 0 4 2 1000 0 2 4 7 0 10] + [ 0 1 0 0 2 1 1 0 0 0 1 8 1 0 0 1087 10 11 2 6 3] + [ 1 17 0 1 1 9 0 0 3 0 0 10 1 2 3 9 1097 0 0 3 4] + [ 1 0 0 3 0 0 0 1 0 0 2 4 12 0 1 8 2 1001 1 2 0] + [ 2 7 8 16 0 1 0 36 4 1 4 0 1 0 12 0 0 0 964 1 11] + [ 0 2 3 0 0 5 5 14 0 0 2 17 2 7 0 7 7 0 0 1076 5] + [ 148 216 113 71 69 222 31 123 163 76 192 189 402 303 135 101 262 101 132 225 4631]] + +2023-10-02 20:43:27,401 - ==> Best [Top1: 83.492 Top5: 98.283 Sparsity:0.00 Params: 169472 on epoch: 27] +2023-10-02 20:43:27,401 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:43:27,407 - + +2023-10-02 20:43:27,407 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:43:28,530 - Epoch: [31][ 10/ 1236] Overall Loss 0.265759 Objective Loss 0.265759 LR 0.001000 Time 0.112222 +2023-10-02 20:43:28,736 - Epoch: [31][ 20/ 1236] Overall Loss 0.257560 Objective Loss 0.257560 LR 0.001000 Time 0.066413 +2023-10-02 20:43:28,942 - Epoch: [31][ 30/ 1236] Overall Loss 0.255595 Objective Loss 0.255595 LR 0.001000 Time 0.051123 +2023-10-02 20:43:29,149 - Epoch: [31][ 40/ 1236] Overall Loss 0.253211 Objective Loss 0.253211 LR 0.001000 Time 0.043495 +2023-10-02 20:43:29,353 - Epoch: [31][ 50/ 1236] Overall Loss 0.257711 Objective Loss 0.257711 LR 0.001000 Time 0.038883 +2023-10-02 20:43:29,560 - Epoch: [31][ 60/ 1236] Overall Loss 0.263619 Objective Loss 0.263619 LR 0.001000 Time 0.035846 +2023-10-02 20:43:29,764 - Epoch: [31][ 70/ 1236] Overall Loss 0.260133 Objective Loss 0.260133 LR 0.001000 Time 0.033641 +2023-10-02 20:43:29,971 - Epoch: [31][ 80/ 1236] Overall Loss 0.259993 Objective Loss 0.259993 LR 0.001000 Time 0.032010 +2023-10-02 20:43:30,175 - Epoch: [31][ 90/ 1236] Overall Loss 0.258985 Objective Loss 0.258985 LR 0.001000 Time 0.030724 +2023-10-02 20:43:30,381 - Epoch: [31][ 100/ 1236] Overall Loss 0.257064 Objective Loss 0.257064 LR 0.001000 Time 0.029710 +2023-10-02 20:43:30,586 - Epoch: [31][ 110/ 1236] Overall Loss 0.256946 Objective Loss 0.256946 LR 0.001000 Time 0.028868 +2023-10-02 20:43:30,794 - Epoch: [31][ 120/ 1236] Overall Loss 0.259049 Objective Loss 0.259049 LR 0.001000 Time 0.028190 +2023-10-02 20:43:30,998 - Epoch: [31][ 130/ 1236] Overall Loss 0.260470 Objective Loss 0.260470 LR 0.001000 Time 0.027592 +2023-10-02 20:43:31,206 - Epoch: [31][ 140/ 1236] Overall Loss 0.259023 Objective Loss 0.259023 LR 0.001000 Time 0.027102 +2023-10-02 20:43:31,410 - Epoch: [31][ 150/ 1236] Overall Loss 0.257149 Objective Loss 0.257149 LR 0.001000 Time 0.026656 +2023-10-02 20:43:31,618 - Epoch: [31][ 160/ 1236] Overall Loss 0.257614 Objective Loss 0.257614 LR 0.001000 Time 0.026285 +2023-10-02 20:43:31,822 - Epoch: [31][ 170/ 1236] Overall Loss 0.257762 Objective Loss 0.257762 LR 0.001000 Time 0.025939 +2023-10-02 20:43:32,030 - Epoch: [31][ 180/ 1236] Overall Loss 0.259682 Objective Loss 0.259682 LR 0.001000 Time 0.025650 +2023-10-02 20:43:32,234 - Epoch: [31][ 190/ 1236] Overall Loss 0.261130 Objective Loss 0.261130 LR 0.001000 Time 0.025375 +2023-10-02 20:43:32,442 - Epoch: [31][ 200/ 1236] Overall Loss 0.260858 Objective Loss 0.260858 LR 0.001000 Time 0.025144 +2023-10-02 20:43:32,646 - Epoch: [31][ 210/ 1236] Overall Loss 0.261496 Objective Loss 0.261496 LR 0.001000 Time 0.024918 +2023-10-02 20:43:32,853 - Epoch: [31][ 220/ 1236] Overall Loss 0.261459 Objective Loss 0.261459 LR 0.001000 Time 0.024722 +2023-10-02 20:43:33,058 - Epoch: [31][ 230/ 1236] Overall Loss 0.260532 Objective Loss 0.260532 LR 0.001000 Time 0.024535 +2023-10-02 20:43:33,266 - Epoch: [31][ 240/ 1236] Overall Loss 0.262053 Objective Loss 0.262053 LR 0.001000 Time 0.024377 +2023-10-02 20:43:33,470 - Epoch: [31][ 250/ 1236] Overall Loss 0.260433 Objective Loss 0.260433 LR 0.001000 Time 0.024219 +2023-10-02 20:43:33,678 - Epoch: [31][ 260/ 1236] Overall Loss 0.259953 Objective Loss 0.259953 LR 0.001000 Time 0.024085 +2023-10-02 20:43:33,883 - Epoch: [31][ 270/ 1236] Overall Loss 0.259942 Objective Loss 0.259942 LR 0.001000 Time 0.023950 +2023-10-02 20:43:34,086 - Epoch: [31][ 280/ 1236] Overall Loss 0.259988 Objective Loss 0.259988 LR 0.001000 Time 0.023821 +2023-10-02 20:43:34,292 - Epoch: [31][ 290/ 1236] Overall Loss 0.261342 Objective Loss 0.261342 LR 0.001000 Time 0.023707 +2023-10-02 20:43:34,500 - Epoch: [31][ 300/ 1236] Overall Loss 0.262205 Objective Loss 0.262205 LR 0.001000 Time 0.023609 +2023-10-02 20:43:34,704 - Epoch: [31][ 310/ 1236] Overall Loss 0.262578 Objective Loss 0.262578 LR 0.001000 Time 0.023507 +2023-10-02 20:43:34,912 - Epoch: [31][ 320/ 1236] Overall Loss 0.262232 Objective Loss 0.262232 LR 0.001000 Time 0.023419 +2023-10-02 20:43:35,117 - Epoch: [31][ 330/ 1236] Overall Loss 0.262276 Objective Loss 0.262276 LR 0.001000 Time 0.023329 +2023-10-02 20:43:35,324 - Epoch: [31][ 340/ 1236] Overall Loss 0.262072 Objective Loss 0.262072 LR 0.001000 Time 0.023253 +2023-10-02 20:43:35,529 - Epoch: [31][ 350/ 1236] Overall Loss 0.261399 Objective Loss 0.261399 LR 0.001000 Time 0.023174 +2023-10-02 20:43:35,734 - Epoch: [31][ 360/ 1236] Overall Loss 0.261286 Objective Loss 0.261286 LR 0.001000 Time 0.023099 +2023-10-02 20:43:35,940 - Epoch: [31][ 370/ 1236] Overall Loss 0.261340 Objective Loss 0.261340 LR 0.001000 Time 0.023030 +2023-10-02 20:43:36,147 - Epoch: [31][ 380/ 1236] Overall Loss 0.261929 Objective Loss 0.261929 LR 0.001000 Time 0.022966 +2023-10-02 20:43:36,353 - Epoch: [31][ 390/ 1236] Overall Loss 0.262421 Objective Loss 0.262421 LR 0.001000 Time 0.022902 +2023-10-02 20:43:36,560 - Epoch: [31][ 400/ 1236] Overall Loss 0.262206 Objective Loss 0.262206 LR 0.001000 Time 0.022847 +2023-10-02 20:43:36,765 - Epoch: [31][ 410/ 1236] Overall Loss 0.262229 Objective Loss 0.262229 LR 0.001000 Time 0.022790 +2023-10-02 20:43:36,970 - Epoch: [31][ 420/ 1236] Overall Loss 0.262265 Objective Loss 0.262265 LR 0.001000 Time 0.022734 +2023-10-02 20:43:37,176 - Epoch: [31][ 430/ 1236] Overall Loss 0.262083 Objective Loss 0.262083 LR 0.001000 Time 0.022683 +2023-10-02 20:43:37,383 - Epoch: [31][ 440/ 1236] Overall Loss 0.262775 Objective Loss 0.262775 LR 0.001000 Time 0.022639 +2023-10-02 20:43:37,589 - Epoch: [31][ 450/ 1236] Overall Loss 0.262721 Objective Loss 0.262721 LR 0.001000 Time 0.022591 +2023-10-02 20:43:37,795 - Epoch: [31][ 460/ 1236] Overall Loss 0.262945 Objective Loss 0.262945 LR 0.001000 Time 0.022547 +2023-10-02 20:43:38,001 - Epoch: [31][ 470/ 1236] Overall Loss 0.263262 Objective Loss 0.263262 LR 0.001000 Time 0.022503 +2023-10-02 20:43:38,208 - Epoch: [31][ 480/ 1236] Overall Loss 0.263303 Objective Loss 0.263303 LR 0.001000 Time 0.022466 +2023-10-02 20:43:38,413 - Epoch: [31][ 490/ 1236] Overall Loss 0.263552 Objective Loss 0.263552 LR 0.001000 Time 0.022425 +2023-10-02 20:43:38,621 - Epoch: [31][ 500/ 1236] Overall Loss 0.263824 Objective Loss 0.263824 LR 0.001000 Time 0.022392 +2023-10-02 20:43:38,826 - Epoch: [31][ 510/ 1236] Overall Loss 0.264309 Objective Loss 0.264309 LR 0.001000 Time 0.022355 +2023-10-02 20:43:39,034 - Epoch: [31][ 520/ 1236] Overall Loss 0.264376 Objective Loss 0.264376 LR 0.001000 Time 0.022324 +2023-10-02 20:43:39,239 - Epoch: [31][ 530/ 1236] Overall Loss 0.264229 Objective Loss 0.264229 LR 0.001000 Time 0.022289 +2023-10-02 20:43:39,447 - Epoch: [31][ 540/ 1236] Overall Loss 0.264151 Objective Loss 0.264151 LR 0.001000 Time 0.022261 +2023-10-02 20:43:39,652 - Epoch: [31][ 550/ 1236] Overall Loss 0.264718 Objective Loss 0.264718 LR 0.001000 Time 0.022228 +2023-10-02 20:43:39,860 - Epoch: [31][ 560/ 1236] Overall Loss 0.265186 Objective Loss 0.265186 LR 0.001000 Time 0.022202 +2023-10-02 20:43:40,065 - Epoch: [31][ 570/ 1236] Overall Loss 0.265076 Objective Loss 0.265076 LR 0.001000 Time 0.022172 +2023-10-02 20:43:40,273 - Epoch: [31][ 580/ 1236] Overall Loss 0.264720 Objective Loss 0.264720 LR 0.001000 Time 0.022147 +2023-10-02 20:43:40,478 - Epoch: [31][ 590/ 1236] Overall Loss 0.264656 Objective Loss 0.264656 LR 0.001000 Time 0.022119 +2023-10-02 20:43:40,686 - Epoch: [31][ 600/ 1236] Overall Loss 0.264182 Objective Loss 0.264182 LR 0.001000 Time 0.022096 +2023-10-02 20:43:40,891 - Epoch: [31][ 610/ 1236] Overall Loss 0.264740 Objective Loss 0.264740 LR 0.001000 Time 0.022069 +2023-10-02 20:43:41,098 - Epoch: [31][ 620/ 1236] Overall Loss 0.264395 Objective Loss 0.264395 LR 0.001000 Time 0.022048 +2023-10-02 20:43:41,304 - Epoch: [31][ 630/ 1236] Overall Loss 0.264425 Objective Loss 0.264425 LR 0.001000 Time 0.022023 +2023-10-02 20:43:41,510 - Epoch: [31][ 640/ 1236] Overall Loss 0.264353 Objective Loss 0.264353 LR 0.001000 Time 0.022001 +2023-10-02 20:43:41,715 - Epoch: [31][ 650/ 1236] Overall Loss 0.264751 Objective Loss 0.264751 LR 0.001000 Time 0.021977 +2023-10-02 20:43:41,923 - Epoch: [31][ 660/ 1236] Overall Loss 0.264438 Objective Loss 0.264438 LR 0.001000 Time 0.021959 +2023-10-02 20:43:42,127 - Epoch: [31][ 670/ 1236] Overall Loss 0.264332 Objective Loss 0.264332 LR 0.001000 Time 0.021936 +2023-10-02 20:43:42,334 - Epoch: [31][ 680/ 1236] Overall Loss 0.264453 Objective Loss 0.264453 LR 0.001000 Time 0.021917 +2023-10-02 20:43:42,540 - Epoch: [31][ 690/ 1236] Overall Loss 0.264401 Objective Loss 0.264401 LR 0.001000 Time 0.021896 +2023-10-02 20:43:42,746 - Epoch: [31][ 700/ 1236] Overall Loss 0.264758 Objective Loss 0.264758 LR 0.001000 Time 0.021877 +2023-10-02 20:43:42,951 - Epoch: [31][ 710/ 1236] Overall Loss 0.264686 Objective Loss 0.264686 LR 0.001000 Time 0.021858 +2023-10-02 20:43:43,158 - Epoch: [31][ 720/ 1236] Overall Loss 0.265158 Objective Loss 0.265158 LR 0.001000 Time 0.021840 +2023-10-02 20:43:43,364 - Epoch: [31][ 730/ 1236] Overall Loss 0.265348 Objective Loss 0.265348 LR 0.001000 Time 0.021821 +2023-10-02 20:43:43,572 - Epoch: [31][ 740/ 1236] Overall Loss 0.265537 Objective Loss 0.265537 LR 0.001000 Time 0.021807 +2023-10-02 20:43:43,777 - Epoch: [31][ 750/ 1236] Overall Loss 0.265535 Objective Loss 0.265535 LR 0.001000 Time 0.021790 +2023-10-02 20:43:43,984 - Epoch: [31][ 760/ 1236] Overall Loss 0.265390 Objective Loss 0.265390 LR 0.001000 Time 0.021776 +2023-10-02 20:43:44,189 - Epoch: [31][ 770/ 1236] Overall Loss 0.265775 Objective Loss 0.265775 LR 0.001000 Time 0.021759 +2023-10-02 20:43:44,396 - Epoch: [31][ 780/ 1236] Overall Loss 0.265752 Objective Loss 0.265752 LR 0.001000 Time 0.021744 +2023-10-02 20:43:44,601 - Epoch: [31][ 790/ 1236] Overall Loss 0.265916 Objective Loss 0.265916 LR 0.001000 Time 0.021729 +2023-10-02 20:43:44,808 - Epoch: [31][ 800/ 1236] Overall Loss 0.266056 Objective Loss 0.266056 LR 0.001000 Time 0.021715 +2023-10-02 20:43:45,014 - Epoch: [31][ 810/ 1236] Overall Loss 0.265971 Objective Loss 0.265971 LR 0.001000 Time 0.021700 +2023-10-02 20:43:45,221 - Epoch: [31][ 820/ 1236] Overall Loss 0.265746 Objective Loss 0.265746 LR 0.001000 Time 0.021687 +2023-10-02 20:43:45,427 - Epoch: [31][ 830/ 1236] Overall Loss 0.266006 Objective Loss 0.266006 LR 0.001000 Time 0.021672 +2023-10-02 20:43:45,635 - Epoch: [31][ 840/ 1236] Overall Loss 0.266456 Objective Loss 0.266456 LR 0.001000 Time 0.021662 +2023-10-02 20:43:45,839 - Epoch: [31][ 850/ 1236] Overall Loss 0.267173 Objective Loss 0.267173 LR 0.001000 Time 0.021646 +2023-10-02 20:43:46,047 - Epoch: [31][ 860/ 1236] Overall Loss 0.267286 Objective Loss 0.267286 LR 0.001000 Time 0.021635 +2023-10-02 20:43:46,252 - Epoch: [31][ 870/ 1236] Overall Loss 0.267663 Objective Loss 0.267663 LR 0.001000 Time 0.021622 +2023-10-02 20:43:46,460 - Epoch: [31][ 880/ 1236] Overall Loss 0.267564 Objective Loss 0.267564 LR 0.001000 Time 0.021612 +2023-10-02 20:43:46,665 - Epoch: [31][ 890/ 1236] Overall Loss 0.267412 Objective Loss 0.267412 LR 0.001000 Time 0.021600 +2023-10-02 20:43:46,873 - Epoch: [31][ 900/ 1236] Overall Loss 0.267497 Objective Loss 0.267497 LR 0.001000 Time 0.021590 +2023-10-02 20:43:47,077 - Epoch: [31][ 910/ 1236] Overall Loss 0.267634 Objective Loss 0.267634 LR 0.001000 Time 0.021578 +2023-10-02 20:43:47,285 - Epoch: [31][ 920/ 1236] Overall Loss 0.267673 Objective Loss 0.267673 LR 0.001000 Time 0.021569 +2023-10-02 20:43:47,491 - Epoch: [31][ 930/ 1236] Overall Loss 0.267901 Objective Loss 0.267901 LR 0.001000 Time 0.021557 +2023-10-02 20:43:47,697 - Epoch: [31][ 940/ 1236] Overall Loss 0.268574 Objective Loss 0.268574 LR 0.001000 Time 0.021547 +2023-10-02 20:43:47,903 - Epoch: [31][ 950/ 1236] Overall Loss 0.268733 Objective Loss 0.268733 LR 0.001000 Time 0.021536 +2023-10-02 20:43:48,110 - Epoch: [31][ 960/ 1236] Overall Loss 0.268762 Objective Loss 0.268762 LR 0.001000 Time 0.021526 +2023-10-02 20:43:48,315 - Epoch: [31][ 970/ 1236] Overall Loss 0.268758 Objective Loss 0.268758 LR 0.001000 Time 0.021516 +2023-10-02 20:43:48,522 - Epoch: [31][ 980/ 1236] Overall Loss 0.268907 Objective Loss 0.268907 LR 0.001000 Time 0.021508 +2023-10-02 20:43:48,727 - Epoch: [31][ 990/ 1236] Overall Loss 0.268974 Objective Loss 0.268974 LR 0.001000 Time 0.021497 +2023-10-02 20:43:48,934 - Epoch: [31][ 1000/ 1236] Overall Loss 0.269027 Objective Loss 0.269027 LR 0.001000 Time 0.021488 +2023-10-02 20:43:49,140 - Epoch: [31][ 1010/ 1236] Overall Loss 0.268799 Objective Loss 0.268799 LR 0.001000 Time 0.021478 +2023-10-02 20:43:49,348 - Epoch: [31][ 1020/ 1236] Overall Loss 0.269006 Objective Loss 0.269006 LR 0.001000 Time 0.021471 +2023-10-02 20:43:49,553 - Epoch: [31][ 1030/ 1236] Overall Loss 0.269034 Objective Loss 0.269034 LR 0.001000 Time 0.021462 +2023-10-02 20:43:49,761 - Epoch: [31][ 1040/ 1236] Overall Loss 0.269101 Objective Loss 0.269101 LR 0.001000 Time 0.021455 +2023-10-02 20:43:49,966 - Epoch: [31][ 1050/ 1236] Overall Loss 0.268836 Objective Loss 0.268836 LR 0.001000 Time 0.021445 +2023-10-02 20:43:50,171 - Epoch: [31][ 1060/ 1236] Overall Loss 0.269150 Objective Loss 0.269150 LR 0.001000 Time 0.021437 +2023-10-02 20:43:50,377 - Epoch: [31][ 1070/ 1236] Overall Loss 0.269244 Objective Loss 0.269244 LR 0.001000 Time 0.021429 +2023-10-02 20:43:50,582 - Epoch: [31][ 1080/ 1236] Overall Loss 0.269222 Objective Loss 0.269222 LR 0.001000 Time 0.021420 +2023-10-02 20:43:50,789 - Epoch: [31][ 1090/ 1236] Overall Loss 0.269449 Objective Loss 0.269449 LR 0.001000 Time 0.021411 +2023-10-02 20:43:50,997 - Epoch: [31][ 1100/ 1236] Overall Loss 0.269768 Objective Loss 0.269768 LR 0.001000 Time 0.021406 +2023-10-02 20:43:51,202 - Epoch: [31][ 1110/ 1236] Overall Loss 0.269661 Objective Loss 0.269661 LR 0.001000 Time 0.021397 +2023-10-02 20:43:51,410 - Epoch: [31][ 1120/ 1236] Overall Loss 0.269656 Objective Loss 0.269656 LR 0.001000 Time 0.021392 +2023-10-02 20:43:51,615 - Epoch: [31][ 1130/ 1236] Overall Loss 0.270082 Objective Loss 0.270082 LR 0.001000 Time 0.021384 +2023-10-02 20:43:51,822 - Epoch: [31][ 1140/ 1236] Overall Loss 0.270133 Objective Loss 0.270133 LR 0.001000 Time 0.021377 +2023-10-02 20:43:52,028 - Epoch: [31][ 1150/ 1236] Overall Loss 0.270182 Objective Loss 0.270182 LR 0.001000 Time 0.021370 +2023-10-02 20:43:52,235 - Epoch: [31][ 1160/ 1236] Overall Loss 0.270195 Objective Loss 0.270195 LR 0.001000 Time 0.021363 +2023-10-02 20:43:52,441 - Epoch: [31][ 1170/ 1236] Overall Loss 0.270040 Objective Loss 0.270040 LR 0.001000 Time 0.021356 +2023-10-02 20:43:52,649 - Epoch: [31][ 1180/ 1236] Overall Loss 0.269932 Objective Loss 0.269932 LR 0.001000 Time 0.021351 +2023-10-02 20:43:52,854 - Epoch: [31][ 1190/ 1236] Overall Loss 0.269653 Objective Loss 0.269653 LR 0.001000 Time 0.021343 +2023-10-02 20:43:53,062 - Epoch: [31][ 1200/ 1236] Overall Loss 0.269494 Objective Loss 0.269494 LR 0.001000 Time 0.021338 +2023-10-02 20:43:53,267 - Epoch: [31][ 1210/ 1236] Overall Loss 0.269465 Objective Loss 0.269465 LR 0.001000 Time 0.021331 +2023-10-02 20:43:53,475 - Epoch: [31][ 1220/ 1236] Overall Loss 0.269284 Objective Loss 0.269284 LR 0.001000 Time 0.021327 +2023-10-02 20:43:53,733 - Epoch: [31][ 1230/ 1236] Overall Loss 0.269362 Objective Loss 0.269362 LR 0.001000 Time 0.021363 +2023-10-02 20:43:53,854 - Epoch: [31][ 1236/ 1236] Overall Loss 0.269308 Objective Loss 0.269308 Top1 85.132383 Top5 98.167006 LR 0.001000 Time 0.021357 +2023-10-02 20:43:54,012 - --- validate (epoch=31)----------- +2023-10-02 20:43:54,013 - 29943 samples (256 per mini-batch) +2023-10-02 20:43:54,499 - Epoch: [31][ 10/ 117] Loss 0.296345 Top1 83.554688 Top5 98.320312 +2023-10-02 20:43:54,659 - Epoch: [31][ 20/ 117] Loss 0.299195 Top1 84.003906 Top5 98.515625 +2023-10-02 20:43:54,815 - Epoch: [31][ 30/ 117] Loss 0.308149 Top1 84.062500 Top5 98.359375 +2023-10-02 20:43:54,975 - Epoch: [31][ 40/ 117] Loss 0.314248 Top1 83.974609 Top5 98.369141 +2023-10-02 20:43:55,131 - Epoch: [31][ 50/ 117] Loss 0.313053 Top1 83.984375 Top5 98.367188 +2023-10-02 20:43:55,290 - Epoch: [31][ 60/ 117] Loss 0.317179 Top1 83.841146 Top5 98.333333 +2023-10-02 20:43:55,446 - Epoch: [31][ 70/ 117] Loss 0.322821 Top1 83.710938 Top5 98.359375 +2023-10-02 20:43:55,606 - Epoch: [31][ 80/ 117] Loss 0.324586 Top1 83.632812 Top5 98.388672 +2023-10-02 20:43:55,763 - Epoch: [31][ 90/ 117] Loss 0.327156 Top1 83.519965 Top5 98.368056 +2023-10-02 20:43:55,921 - Epoch: [31][ 100/ 117] Loss 0.323160 Top1 83.527344 Top5 98.359375 +2023-10-02 20:43:56,085 - Epoch: [31][ 110/ 117] Loss 0.321896 Top1 83.615057 Top5 98.352273 +2023-10-02 20:43:56,174 - Epoch: [31][ 117/ 117] Loss 0.322861 Top1 83.575460 Top5 98.353538 +2023-10-02 20:43:56,307 - ==> Top1: 83.575 Top5: 98.354 Loss: 0.323 + +2023-10-02 20:43:56,308 - ==> Confusion: +[[ 933 0 5 0 6 2 0 0 7 64 0 1 2 6 7 1 0 4 0 0 12] + [ 1 1048 2 1 7 23 1 14 4 1 2 1 1 0 2 5 5 0 8 0 5] + [ 5 1 947 14 1 0 25 10 0 1 5 2 3 2 1 11 1 1 11 6 9] + [ 5 1 12 978 0 3 2 0 4 0 5 0 4 4 39 1 1 3 12 1 14] + [ 34 8 3 0 945 6 0 0 0 7 3 2 0 4 13 5 13 0 0 1 6] + [ 3 43 0 1 2 997 2 20 1 5 2 8 2 10 4 1 1 0 4 1 9] + [ 0 7 35 0 0 2 1107 4 0 0 4 0 1 1 0 7 0 1 6 6 10] + [ 4 27 22 2 4 30 2 1046 1 3 1 4 0 5 2 1 0 0 48 11 5] + [ 24 1 1 0 2 3 0 1 975 38 6 2 1 13 10 4 2 2 1 1 2] + [ 117 0 1 1 5 1 0 0 29 905 1 0 0 34 6 1 2 1 0 4 11] + [ 2 4 9 14 2 1 2 1 10 2 963 2 0 18 4 2 0 3 8 0 6] + [ 0 2 5 0 0 13 0 5 0 0 0 961 16 5 0 2 1 16 0 4 5] + [ 1 2 2 3 2 1 0 2 1 0 2 54 955 0 3 9 0 10 2 6 13] + [ 1 0 1 0 4 13 1 2 2 9 5 5 1 1053 6 1 2 1 0 2 10] + [ 16 2 3 18 3 0 0 0 22 5 1 0 4 2 1006 0 2 3 9 0 5] + [ 0 0 3 2 1 0 1 0 0 1 0 10 6 1 0 1064 17 10 2 8 8] + [ 0 14 0 0 2 4 0 1 2 1 0 8 1 1 2 10 1097 1 0 2 15] + [ 0 0 2 2 0 0 0 0 2 0 1 6 27 2 0 9 0 983 1 2 1] + [ 0 6 8 20 0 0 0 16 9 0 1 0 1 0 12 1 1 0 982 0 11] + [ 0 1 2 5 1 8 9 11 0 0 0 20 1 0 0 3 6 0 4 1073 8] + [ 138 192 177 102 79 182 25 86 117 68 145 164 374 299 173 80 82 83 156 176 5007]] + +2023-10-02 20:43:56,309 - ==> Best [Top1: 83.575 Top5: 98.354 Sparsity:0.00 Params: 169472 on epoch: 31] +2023-10-02 20:43:56,309 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:43:56,317 - + +2023-10-02 20:43:56,317 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:43:57,330 - Epoch: [32][ 10/ 1236] Overall Loss 0.247052 Objective Loss 0.247052 LR 0.001000 Time 0.101230 +2023-10-02 20:43:57,536 - Epoch: [32][ 20/ 1236] Overall Loss 0.255475 Objective Loss 0.255475 LR 0.001000 Time 0.060921 +2023-10-02 20:43:57,742 - Epoch: [32][ 30/ 1236] Overall Loss 0.256613 Objective Loss 0.256613 LR 0.001000 Time 0.047451 +2023-10-02 20:43:57,949 - Epoch: [32][ 40/ 1236] Overall Loss 0.259478 Objective Loss 0.259478 LR 0.001000 Time 0.040776 +2023-10-02 20:43:58,153 - Epoch: [32][ 50/ 1236] Overall Loss 0.257560 Objective Loss 0.257560 LR 0.001000 Time 0.036699 +2023-10-02 20:43:58,361 - Epoch: [32][ 60/ 1236] Overall Loss 0.252230 Objective Loss 0.252230 LR 0.001000 Time 0.034037 +2023-10-02 20:43:58,565 - Epoch: [32][ 70/ 1236] Overall Loss 0.251799 Objective Loss 0.251799 LR 0.001000 Time 0.032090 +2023-10-02 20:43:58,773 - Epoch: [32][ 80/ 1236] Overall Loss 0.249941 Objective Loss 0.249941 LR 0.001000 Time 0.030667 +2023-10-02 20:43:58,977 - Epoch: [32][ 90/ 1236] Overall Loss 0.251252 Objective Loss 0.251252 LR 0.001000 Time 0.029527 +2023-10-02 20:43:59,185 - Epoch: [32][ 100/ 1236] Overall Loss 0.250340 Objective Loss 0.250340 LR 0.001000 Time 0.028650 +2023-10-02 20:43:59,389 - Epoch: [32][ 110/ 1236] Overall Loss 0.251803 Objective Loss 0.251803 LR 0.001000 Time 0.027901 +2023-10-02 20:43:59,597 - Epoch: [32][ 120/ 1236] Overall Loss 0.254945 Objective Loss 0.254945 LR 0.001000 Time 0.027306 +2023-10-02 20:43:59,802 - Epoch: [32][ 130/ 1236] Overall Loss 0.254740 Objective Loss 0.254740 LR 0.001000 Time 0.026775 +2023-10-02 20:44:00,009 - Epoch: [32][ 140/ 1236] Overall Loss 0.254075 Objective Loss 0.254075 LR 0.001000 Time 0.026344 +2023-10-02 20:44:00,214 - Epoch: [32][ 150/ 1236] Overall Loss 0.253152 Objective Loss 0.253152 LR 0.001000 Time 0.025948 +2023-10-02 20:44:00,421 - Epoch: [32][ 160/ 1236] Overall Loss 0.251959 Objective Loss 0.251959 LR 0.001000 Time 0.025623 +2023-10-02 20:44:00,625 - Epoch: [32][ 170/ 1236] Overall Loss 0.252442 Objective Loss 0.252442 LR 0.001000 Time 0.025312 +2023-10-02 20:44:00,831 - Epoch: [32][ 180/ 1236] Overall Loss 0.254413 Objective Loss 0.254413 LR 0.001000 Time 0.025046 +2023-10-02 20:44:01,036 - Epoch: [32][ 190/ 1236] Overall Loss 0.253872 Objective Loss 0.253872 LR 0.001000 Time 0.024801 +2023-10-02 20:44:01,243 - Epoch: [32][ 200/ 1236] Overall Loss 0.253721 Objective Loss 0.253721 LR 0.001000 Time 0.024595 +2023-10-02 20:44:01,449 - Epoch: [32][ 210/ 1236] Overall Loss 0.254162 Objective Loss 0.254162 LR 0.001000 Time 0.024405 +2023-10-02 20:44:01,658 - Epoch: [32][ 220/ 1236] Overall Loss 0.254131 Objective Loss 0.254131 LR 0.001000 Time 0.024242 +2023-10-02 20:44:01,867 - Epoch: [32][ 230/ 1236] Overall Loss 0.255022 Objective Loss 0.255022 LR 0.001000 Time 0.024097 +2023-10-02 20:44:02,076 - Epoch: [32][ 240/ 1236] Overall Loss 0.254603 Objective Loss 0.254603 LR 0.001000 Time 0.023957 +2023-10-02 20:44:02,287 - Epoch: [32][ 250/ 1236] Overall Loss 0.254704 Objective Loss 0.254704 LR 0.001000 Time 0.023841 +2023-10-02 20:44:02,494 - Epoch: [32][ 260/ 1236] Overall Loss 0.255567 Objective Loss 0.255567 LR 0.001000 Time 0.023721 +2023-10-02 20:44:02,704 - Epoch: [32][ 270/ 1236] Overall Loss 0.255524 Objective Loss 0.255524 LR 0.001000 Time 0.023617 +2023-10-02 20:44:02,912 - Epoch: [32][ 280/ 1236] Overall Loss 0.256419 Objective Loss 0.256419 LR 0.001000 Time 0.023514 +2023-10-02 20:44:03,122 - Epoch: [32][ 290/ 1236] Overall Loss 0.256816 Objective Loss 0.256816 LR 0.001000 Time 0.023424 +2023-10-02 20:44:03,331 - Epoch: [32][ 300/ 1236] Overall Loss 0.257988 Objective Loss 0.257988 LR 0.001000 Time 0.023335 +2023-10-02 20:44:03,540 - Epoch: [32][ 310/ 1236] Overall Loss 0.258350 Objective Loss 0.258350 LR 0.001000 Time 0.023257 +2023-10-02 20:44:03,749 - Epoch: [32][ 320/ 1236] Overall Loss 0.258456 Objective Loss 0.258456 LR 0.001000 Time 0.023178 +2023-10-02 20:44:03,959 - Epoch: [32][ 330/ 1236] Overall Loss 0.259976 Objective Loss 0.259976 LR 0.001000 Time 0.023110 +2023-10-02 20:44:04,166 - Epoch: [32][ 340/ 1236] Overall Loss 0.261277 Objective Loss 0.261277 LR 0.001000 Time 0.023036 +2023-10-02 20:44:04,376 - Epoch: [32][ 350/ 1236] Overall Loss 0.261110 Objective Loss 0.261110 LR 0.001000 Time 0.022976 +2023-10-02 20:44:04,586 - Epoch: [32][ 360/ 1236] Overall Loss 0.261044 Objective Loss 0.261044 LR 0.001000 Time 0.022917 +2023-10-02 20:44:04,798 - Epoch: [32][ 370/ 1236] Overall Loss 0.262057 Objective Loss 0.262057 LR 0.001000 Time 0.022870 +2023-10-02 20:44:05,006 - Epoch: [32][ 380/ 1236] Overall Loss 0.261311 Objective Loss 0.261311 LR 0.001000 Time 0.022814 +2023-10-02 20:44:05,218 - Epoch: [32][ 390/ 1236] Overall Loss 0.261112 Objective Loss 0.261112 LR 0.001000 Time 0.022772 +2023-10-02 20:44:05,427 - Epoch: [32][ 400/ 1236] Overall Loss 0.260329 Objective Loss 0.260329 LR 0.001000 Time 0.022725 +2023-10-02 20:44:05,638 - Epoch: [32][ 410/ 1236] Overall Loss 0.260330 Objective Loss 0.260330 LR 0.001000 Time 0.022684 +2023-10-02 20:44:05,847 - Epoch: [32][ 420/ 1236] Overall Loss 0.260301 Objective Loss 0.260301 LR 0.001000 Time 0.022639 +2023-10-02 20:44:06,058 - Epoch: [32][ 430/ 1236] Overall Loss 0.260064 Objective Loss 0.260064 LR 0.001000 Time 0.022601 +2023-10-02 20:44:06,267 - Epoch: [32][ 440/ 1236] Overall Loss 0.259543 Objective Loss 0.259543 LR 0.001000 Time 0.022560 +2023-10-02 20:44:06,477 - Epoch: [32][ 450/ 1236] Overall Loss 0.260002 Objective Loss 0.260002 LR 0.001000 Time 0.022525 +2023-10-02 20:44:06,687 - Epoch: [32][ 460/ 1236] Overall Loss 0.260384 Objective Loss 0.260384 LR 0.001000 Time 0.022488 +2023-10-02 20:44:06,899 - Epoch: [32][ 470/ 1236] Overall Loss 0.260551 Objective Loss 0.260551 LR 0.001000 Time 0.022460 +2023-10-02 20:44:07,107 - Epoch: [32][ 480/ 1236] Overall Loss 0.260901 Objective Loss 0.260901 LR 0.001000 Time 0.022426 +2023-10-02 20:44:07,319 - Epoch: [32][ 490/ 1236] Overall Loss 0.260659 Objective Loss 0.260659 LR 0.001000 Time 0.022400 +2023-10-02 20:44:07,528 - Epoch: [32][ 500/ 1236] Overall Loss 0.260499 Objective Loss 0.260499 LR 0.001000 Time 0.022368 +2023-10-02 20:44:07,740 - Epoch: [32][ 510/ 1236] Overall Loss 0.261002 Objective Loss 0.261002 LR 0.001000 Time 0.022345 +2023-10-02 20:44:07,949 - Epoch: [32][ 520/ 1236] Overall Loss 0.260925 Objective Loss 0.260925 LR 0.001000 Time 0.022318 +2023-10-02 20:44:08,160 - Epoch: [32][ 530/ 1236] Overall Loss 0.261099 Objective Loss 0.261099 LR 0.001000 Time 0.022294 +2023-10-02 20:44:08,367 - Epoch: [32][ 540/ 1236] Overall Loss 0.261673 Objective Loss 0.261673 LR 0.001000 Time 0.022264 +2023-10-02 20:44:08,578 - Epoch: [32][ 550/ 1236] Overall Loss 0.261962 Objective Loss 0.261962 LR 0.001000 Time 0.022241 +2023-10-02 20:44:08,785 - Epoch: [32][ 560/ 1236] Overall Loss 0.261991 Objective Loss 0.261991 LR 0.001000 Time 0.022213 +2023-10-02 20:44:08,996 - Epoch: [32][ 570/ 1236] Overall Loss 0.261993 Objective Loss 0.261993 LR 0.001000 Time 0.022193 +2023-10-02 20:44:09,203 - Epoch: [32][ 580/ 1236] Overall Loss 0.262243 Objective Loss 0.262243 LR 0.001000 Time 0.022167 +2023-10-02 20:44:09,412 - Epoch: [32][ 590/ 1236] Overall Loss 0.262475 Objective Loss 0.262475 LR 0.001000 Time 0.022145 +2023-10-02 20:44:09,620 - Epoch: [32][ 600/ 1236] Overall Loss 0.262316 Objective Loss 0.262316 LR 0.001000 Time 0.022120 +2023-10-02 20:44:09,831 - Epoch: [32][ 610/ 1236] Overall Loss 0.261822 Objective Loss 0.261822 LR 0.001000 Time 0.022102 +2023-10-02 20:44:10,042 - Epoch: [32][ 620/ 1236] Overall Loss 0.262510 Objective Loss 0.262510 LR 0.001000 Time 0.022083 +2023-10-02 20:44:10,254 - Epoch: [32][ 630/ 1236] Overall Loss 0.262541 Objective Loss 0.262541 LR 0.001000 Time 0.022070 +2023-10-02 20:44:10,465 - Epoch: [32][ 640/ 1236] Overall Loss 0.262839 Objective Loss 0.262839 LR 0.001000 Time 0.022052 +2023-10-02 20:44:10,678 - Epoch: [32][ 650/ 1236] Overall Loss 0.263138 Objective Loss 0.263138 LR 0.001000 Time 0.022039 +2023-10-02 20:44:10,888 - Epoch: [32][ 660/ 1236] Overall Loss 0.263185 Objective Loss 0.263185 LR 0.001000 Time 0.022022 +2023-10-02 20:44:11,103 - Epoch: [32][ 670/ 1236] Overall Loss 0.263037 Objective Loss 0.263037 LR 0.001000 Time 0.022010 +2023-10-02 20:44:11,315 - Epoch: [32][ 680/ 1236] Overall Loss 0.263026 Objective Loss 0.263026 LR 0.001000 Time 0.021995 +2023-10-02 20:44:11,526 - Epoch: [32][ 690/ 1236] Overall Loss 0.262810 Objective Loss 0.262810 LR 0.001000 Time 0.021982 +2023-10-02 20:44:11,736 - Epoch: [32][ 700/ 1236] Overall Loss 0.263472 Objective Loss 0.263472 LR 0.001000 Time 0.021968 +2023-10-02 20:44:11,948 - Epoch: [32][ 710/ 1236] Overall Loss 0.263479 Objective Loss 0.263479 LR 0.001000 Time 0.021955 +2023-10-02 20:44:12,158 - Epoch: [32][ 720/ 1236] Overall Loss 0.263801 Objective Loss 0.263801 LR 0.001000 Time 0.021942 +2023-10-02 20:44:12,369 - Epoch: [32][ 730/ 1236] Overall Loss 0.263949 Objective Loss 0.263949 LR 0.001000 Time 0.021930 +2023-10-02 20:44:12,579 - Epoch: [32][ 740/ 1236] Overall Loss 0.264011 Objective Loss 0.264011 LR 0.001000 Time 0.021918 +2023-10-02 20:44:12,790 - Epoch: [32][ 750/ 1236] Overall Loss 0.264002 Objective Loss 0.264002 LR 0.001000 Time 0.021906 +2023-10-02 20:44:13,001 - Epoch: [32][ 760/ 1236] Overall Loss 0.264346 Objective Loss 0.264346 LR 0.001000 Time 0.021895 +2023-10-02 20:44:13,212 - Epoch: [32][ 770/ 1236] Overall Loss 0.264671 Objective Loss 0.264671 LR 0.001000 Time 0.021884 +2023-10-02 20:44:13,422 - Epoch: [32][ 780/ 1236] Overall Loss 0.265211 Objective Loss 0.265211 LR 0.001000 Time 0.021873 +2023-10-02 20:44:13,633 - Epoch: [32][ 790/ 1236] Overall Loss 0.265174 Objective Loss 0.265174 LR 0.001000 Time 0.021863 +2023-10-02 20:44:13,843 - Epoch: [32][ 800/ 1236] Overall Loss 0.265513 Objective Loss 0.265513 LR 0.001000 Time 0.021852 +2023-10-02 20:44:14,055 - Epoch: [32][ 810/ 1236] Overall Loss 0.265539 Objective Loss 0.265539 LR 0.001000 Time 0.021843 +2023-10-02 20:44:14,265 - Epoch: [32][ 820/ 1236] Overall Loss 0.265233 Objective Loss 0.265233 LR 0.001000 Time 0.021832 +2023-10-02 20:44:14,476 - Epoch: [32][ 830/ 1236] Overall Loss 0.265163 Objective Loss 0.265163 LR 0.001000 Time 0.021823 +2023-10-02 20:44:14,686 - Epoch: [32][ 840/ 1236] Overall Loss 0.265105 Objective Loss 0.265105 LR 0.001000 Time 0.021814 +2023-10-02 20:44:14,898 - Epoch: [32][ 850/ 1236] Overall Loss 0.265014 Objective Loss 0.265014 LR 0.001000 Time 0.021805 +2023-10-02 20:44:15,108 - Epoch: [32][ 860/ 1236] Overall Loss 0.264934 Objective Loss 0.264934 LR 0.001000 Time 0.021796 +2023-10-02 20:44:15,319 - Epoch: [32][ 870/ 1236] Overall Loss 0.264981 Objective Loss 0.264981 LR 0.001000 Time 0.021788 +2023-10-02 20:44:15,530 - Epoch: [32][ 880/ 1236] Overall Loss 0.265128 Objective Loss 0.265128 LR 0.001000 Time 0.021779 +2023-10-02 20:44:15,741 - Epoch: [32][ 890/ 1236] Overall Loss 0.265487 Objective Loss 0.265487 LR 0.001000 Time 0.021772 +2023-10-02 20:44:15,951 - Epoch: [32][ 900/ 1236] Overall Loss 0.265679 Objective Loss 0.265679 LR 0.001000 Time 0.021763 +2023-10-02 20:44:16,162 - Epoch: [32][ 910/ 1236] Overall Loss 0.265875 Objective Loss 0.265875 LR 0.001000 Time 0.021756 +2023-10-02 20:44:16,373 - Epoch: [32][ 920/ 1236] Overall Loss 0.265936 Objective Loss 0.265936 LR 0.001000 Time 0.021748 +2023-10-02 20:44:16,584 - Epoch: [32][ 930/ 1236] Overall Loss 0.265680 Objective Loss 0.265680 LR 0.001000 Time 0.021741 +2023-10-02 20:44:16,794 - Epoch: [32][ 940/ 1236] Overall Loss 0.265681 Objective Loss 0.265681 LR 0.001000 Time 0.021733 +2023-10-02 20:44:17,006 - Epoch: [32][ 950/ 1236] Overall Loss 0.265555 Objective Loss 0.265555 LR 0.001000 Time 0.021726 +2023-10-02 20:44:17,216 - Epoch: [32][ 960/ 1236] Overall Loss 0.265679 Objective Loss 0.265679 LR 0.001000 Time 0.021719 +2023-10-02 20:44:17,426 - Epoch: [32][ 970/ 1236] Overall Loss 0.265626 Objective Loss 0.265626 LR 0.001000 Time 0.021711 +2023-10-02 20:44:17,634 - Epoch: [32][ 980/ 1236] Overall Loss 0.265671 Objective Loss 0.265671 LR 0.001000 Time 0.021702 +2023-10-02 20:44:17,843 - Epoch: [32][ 990/ 1236] Overall Loss 0.266141 Objective Loss 0.266141 LR 0.001000 Time 0.021693 +2023-10-02 20:44:18,052 - Epoch: [32][ 1000/ 1236] Overall Loss 0.266323 Objective Loss 0.266323 LR 0.001000 Time 0.021685 +2023-10-02 20:44:18,259 - Epoch: [32][ 1010/ 1236] Overall Loss 0.266329 Objective Loss 0.266329 LR 0.001000 Time 0.021674 +2023-10-02 20:44:18,470 - Epoch: [32][ 1020/ 1236] Overall Loss 0.266377 Objective Loss 0.266377 LR 0.001000 Time 0.021667 +2023-10-02 20:44:18,676 - Epoch: [32][ 1030/ 1236] Overall Loss 0.266592 Objective Loss 0.266592 LR 0.001000 Time 0.021657 +2023-10-02 20:44:18,884 - Epoch: [32][ 1040/ 1236] Overall Loss 0.266494 Objective Loss 0.266494 LR 0.001000 Time 0.021649 +2023-10-02 20:44:19,092 - Epoch: [32][ 1050/ 1236] Overall Loss 0.266623 Objective Loss 0.266623 LR 0.001000 Time 0.021639 +2023-10-02 20:44:19,302 - Epoch: [32][ 1060/ 1236] Overall Loss 0.266626 Objective Loss 0.266626 LR 0.001000 Time 0.021633 +2023-10-02 20:44:19,508 - Epoch: [32][ 1070/ 1236] Overall Loss 0.266529 Objective Loss 0.266529 LR 0.001000 Time 0.021623 +2023-10-02 20:44:19,717 - Epoch: [32][ 1080/ 1236] Overall Loss 0.266614 Objective Loss 0.266614 LR 0.001000 Time 0.021616 +2023-10-02 20:44:19,924 - Epoch: [32][ 1090/ 1236] Overall Loss 0.266687 Objective Loss 0.266687 LR 0.001000 Time 0.021608 +2023-10-02 20:44:20,133 - Epoch: [32][ 1100/ 1236] Overall Loss 0.266740 Objective Loss 0.266740 LR 0.001000 Time 0.021601 +2023-10-02 20:44:20,341 - Epoch: [32][ 1110/ 1236] Overall Loss 0.266946 Objective Loss 0.266946 LR 0.001000 Time 0.021593 +2023-10-02 20:44:20,549 - Epoch: [32][ 1120/ 1236] Overall Loss 0.266879 Objective Loss 0.266879 LR 0.001000 Time 0.021586 +2023-10-02 20:44:20,757 - Epoch: [32][ 1130/ 1236] Overall Loss 0.266883 Objective Loss 0.266883 LR 0.001000 Time 0.021579 +2023-10-02 20:44:20,966 - Epoch: [32][ 1140/ 1236] Overall Loss 0.267029 Objective Loss 0.267029 LR 0.001000 Time 0.021573 +2023-10-02 20:44:21,173 - Epoch: [32][ 1150/ 1236] Overall Loss 0.267376 Objective Loss 0.267376 LR 0.001000 Time 0.021564 +2023-10-02 20:44:21,384 - Epoch: [32][ 1160/ 1236] Overall Loss 0.267483 Objective Loss 0.267483 LR 0.001000 Time 0.021559 +2023-10-02 20:44:21,590 - Epoch: [32][ 1170/ 1236] Overall Loss 0.267794 Objective Loss 0.267794 LR 0.001000 Time 0.021551 +2023-10-02 20:44:21,800 - Epoch: [32][ 1180/ 1236] Overall Loss 0.268152 Objective Loss 0.268152 LR 0.001000 Time 0.021546 +2023-10-02 20:44:22,007 - Epoch: [32][ 1190/ 1236] Overall Loss 0.268108 Objective Loss 0.268108 LR 0.001000 Time 0.021539 +2023-10-02 20:44:22,217 - Epoch: [32][ 1200/ 1236] Overall Loss 0.268404 Objective Loss 0.268404 LR 0.001000 Time 0.021534 +2023-10-02 20:44:22,423 - Epoch: [32][ 1210/ 1236] Overall Loss 0.268532 Objective Loss 0.268532 LR 0.001000 Time 0.021527 +2023-10-02 20:44:22,634 - Epoch: [32][ 1220/ 1236] Overall Loss 0.268834 Objective Loss 0.268834 LR 0.001000 Time 0.021522 +2023-10-02 20:44:22,893 - Epoch: [32][ 1230/ 1236] Overall Loss 0.268717 Objective Loss 0.268717 LR 0.001000 Time 0.021558 +2023-10-02 20:44:23,015 - Epoch: [32][ 1236/ 1236] Overall Loss 0.268784 Objective Loss 0.268784 Top1 86.354379 Top5 98.167006 LR 0.001000 Time 0.021552 +2023-10-02 20:44:23,162 - --- validate (epoch=32)----------- +2023-10-02 20:44:23,163 - 29943 samples (256 per mini-batch) +2023-10-02 20:44:23,648 - Epoch: [32][ 10/ 117] Loss 0.342186 Top1 84.101562 Top5 98.281250 +2023-10-02 20:44:23,803 - Epoch: [32][ 20/ 117] Loss 0.349008 Top1 84.082031 Top5 98.085938 +2023-10-02 20:44:23,959 - Epoch: [32][ 30/ 117] Loss 0.334860 Top1 84.283854 Top5 98.125000 +2023-10-02 20:44:24,121 - Epoch: [32][ 40/ 117] Loss 0.333430 Top1 84.316406 Top5 98.115234 +2023-10-02 20:44:24,276 - Epoch: [32][ 50/ 117] Loss 0.339286 Top1 83.937500 Top5 98.054688 +2023-10-02 20:44:24,437 - Epoch: [32][ 60/ 117] Loss 0.333942 Top1 84.055990 Top5 98.151042 +2023-10-02 20:44:24,592 - Epoch: [32][ 70/ 117] Loss 0.331440 Top1 84.308036 Top5 98.203125 +2023-10-02 20:44:24,753 - Epoch: [32][ 80/ 117] Loss 0.331664 Top1 84.204102 Top5 98.149414 +2023-10-02 20:44:24,909 - Epoch: [32][ 90/ 117] Loss 0.334459 Top1 84.171007 Top5 98.138021 +2023-10-02 20:44:25,070 - Epoch: [32][ 100/ 117] Loss 0.332247 Top1 84.085938 Top5 98.175781 +2023-10-02 20:44:25,232 - Epoch: [32][ 110/ 117] Loss 0.330379 Top1 84.190341 Top5 98.185369 +2023-10-02 20:44:25,321 - Epoch: [32][ 117/ 117] Loss 0.331146 Top1 84.159904 Top5 98.173196 +2023-10-02 20:44:25,435 - ==> Top1: 84.160 Top5: 98.173 Loss: 0.331 + +2023-10-02 20:44:25,436 - ==> Confusion: +[[ 920 2 7 1 16 2 0 0 4 61 1 0 1 2 6 3 5 1 0 0 18] + [ 1 1060 2 0 13 13 1 16 1 0 1 1 1 0 0 4 0 0 5 1 11] + [ 5 0 960 5 7 0 30 15 0 2 1 2 3 3 0 2 1 1 9 3 7] + [ 2 2 29 962 1 1 5 2 3 2 4 0 5 4 25 2 0 2 17 1 20] + [ 22 5 1 0 984 2 1 0 0 5 0 2 0 4 6 4 5 0 0 4 5] + [ 2 57 1 1 9 957 1 18 3 5 3 4 9 13 4 1 2 0 5 4 17] + [ 0 4 29 0 0 1 1119 6 0 0 4 1 1 0 0 4 1 1 3 8 9] + [ 5 20 17 1 7 22 3 1072 1 1 7 7 1 3 2 1 0 0 27 9 12] + [ 24 2 1 0 1 5 0 1 968 40 9 3 1 9 19 2 2 0 1 0 1] + [ 123 0 2 0 18 2 0 0 26 909 0 0 0 16 7 1 1 0 0 4 10] + [ 4 4 12 5 2 1 3 3 12 1 957 4 0 12 7 0 2 2 11 3 8] + [ 1 0 4 0 1 6 0 1 0 4 0 912 63 10 0 6 2 12 0 8 5] + [ 0 0 6 3 4 2 5 0 1 0 0 28 981 1 4 11 2 6 2 3 9] + [ 3 0 2 0 3 7 1 1 12 20 7 5 1 1024 5 3 1 1 0 4 19] + [ 9 2 4 17 12 0 0 0 22 6 1 0 3 2 1004 0 0 1 4 1 13] + [ 0 0 1 0 8 0 2 0 0 0 0 5 8 2 0 1076 14 5 2 6 5] + [ 0 17 0 0 7 2 2 2 0 0 0 4 1 1 3 18 1091 0 0 5 8] + [ 0 0 0 7 1 0 3 0 1 0 1 3 25 1 2 7 1 980 0 0 6] + [ 4 7 4 13 1 0 0 37 8 0 1 0 0 0 10 1 0 0 966 1 15] + [ 0 1 1 1 0 3 10 8 0 0 3 9 6 2 0 9 2 0 0 1090 7] + [ 154 167 145 56 186 98 51 83 98 64 142 110 410 240 138 90 102 36 135 192 5208]] + +2023-10-02 20:44:25,437 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:44:25,437 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:44:25,450 - + +2023-10-02 20:44:25,450 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:44:26,464 - Epoch: [33][ 10/ 1236] Overall Loss 0.232928 Objective Loss 0.232928 LR 0.001000 Time 0.101321 +2023-10-02 20:44:26,674 - Epoch: [33][ 20/ 1236] Overall Loss 0.227844 Objective Loss 0.227844 LR 0.001000 Time 0.061147 +2023-10-02 20:44:26,883 - Epoch: [33][ 30/ 1236] Overall Loss 0.237461 Objective Loss 0.237461 LR 0.001000 Time 0.047661 +2023-10-02 20:44:27,094 - Epoch: [33][ 40/ 1236] Overall Loss 0.246002 Objective Loss 0.246002 LR 0.001000 Time 0.041030 +2023-10-02 20:44:27,301 - Epoch: [33][ 50/ 1236] Overall Loss 0.245803 Objective Loss 0.245803 LR 0.001000 Time 0.036953 +2023-10-02 20:44:27,512 - Epoch: [33][ 60/ 1236] Overall Loss 0.249139 Objective Loss 0.249139 LR 0.001000 Time 0.034312 +2023-10-02 20:44:27,719 - Epoch: [33][ 70/ 1236] Overall Loss 0.255102 Objective Loss 0.255102 LR 0.001000 Time 0.032362 +2023-10-02 20:44:27,930 - Epoch: [33][ 80/ 1236] Overall Loss 0.254928 Objective Loss 0.254928 LR 0.001000 Time 0.030953 +2023-10-02 20:44:28,138 - Epoch: [33][ 90/ 1236] Overall Loss 0.255570 Objective Loss 0.255570 LR 0.001000 Time 0.029812 +2023-10-02 20:44:28,348 - Epoch: [33][ 100/ 1236] Overall Loss 0.258365 Objective Loss 0.258365 LR 0.001000 Time 0.028938 +2023-10-02 20:44:28,555 - Epoch: [33][ 110/ 1236] Overall Loss 0.258707 Objective Loss 0.258707 LR 0.001000 Time 0.028185 +2023-10-02 20:44:28,767 - Epoch: [33][ 120/ 1236] Overall Loss 0.260758 Objective Loss 0.260758 LR 0.001000 Time 0.027597 +2023-10-02 20:44:28,974 - Epoch: [33][ 130/ 1236] Overall Loss 0.261077 Objective Loss 0.261077 LR 0.001000 Time 0.027065 +2023-10-02 20:44:29,186 - Epoch: [33][ 140/ 1236] Overall Loss 0.261486 Objective Loss 0.261486 LR 0.001000 Time 0.026643 +2023-10-02 20:44:29,392 - Epoch: [33][ 150/ 1236] Overall Loss 0.260801 Objective Loss 0.260801 LR 0.001000 Time 0.026240 +2023-10-02 20:44:29,601 - Epoch: [33][ 160/ 1236] Overall Loss 0.262791 Objective Loss 0.262791 LR 0.001000 Time 0.025905 +2023-10-02 20:44:29,809 - Epoch: [33][ 170/ 1236] Overall Loss 0.262450 Objective Loss 0.262450 LR 0.001000 Time 0.025595 +2023-10-02 20:44:30,016 - Epoch: [33][ 180/ 1236] Overall Loss 0.263233 Objective Loss 0.263233 LR 0.001000 Time 0.025324 +2023-10-02 20:44:30,224 - Epoch: [33][ 190/ 1236] Overall Loss 0.263315 Objective Loss 0.263315 LR 0.001000 Time 0.025077 +2023-10-02 20:44:30,431 - Epoch: [33][ 200/ 1236] Overall Loss 0.264063 Objective Loss 0.264063 LR 0.001000 Time 0.024858 +2023-10-02 20:44:30,641 - Epoch: [33][ 210/ 1236] Overall Loss 0.263290 Objective Loss 0.263290 LR 0.001000 Time 0.024669 +2023-10-02 20:44:30,848 - Epoch: [33][ 220/ 1236] Overall Loss 0.264969 Objective Loss 0.264969 LR 0.001000 Time 0.024488 +2023-10-02 20:44:31,057 - Epoch: [33][ 230/ 1236] Overall Loss 0.264954 Objective Loss 0.264954 LR 0.001000 Time 0.024331 +2023-10-02 20:44:31,264 - Epoch: [33][ 240/ 1236] Overall Loss 0.264620 Objective Loss 0.264620 LR 0.001000 Time 0.024180 +2023-10-02 20:44:31,473 - Epoch: [33][ 250/ 1236] Overall Loss 0.264931 Objective Loss 0.264931 LR 0.001000 Time 0.024048 +2023-10-02 20:44:31,681 - Epoch: [33][ 260/ 1236] Overall Loss 0.265507 Objective Loss 0.265507 LR 0.001000 Time 0.023918 +2023-10-02 20:44:31,890 - Epoch: [33][ 270/ 1236] Overall Loss 0.265272 Objective Loss 0.265272 LR 0.001000 Time 0.023806 +2023-10-02 20:44:32,097 - Epoch: [33][ 280/ 1236] Overall Loss 0.264980 Objective Loss 0.264980 LR 0.001000 Time 0.023695 +2023-10-02 20:44:32,306 - Epoch: [33][ 290/ 1236] Overall Loss 0.265203 Objective Loss 0.265203 LR 0.001000 Time 0.023598 +2023-10-02 20:44:32,513 - Epoch: [33][ 300/ 1236] Overall Loss 0.264666 Objective Loss 0.264666 LR 0.001000 Time 0.023499 +2023-10-02 20:44:32,732 - Epoch: [33][ 310/ 1236] Overall Loss 0.264892 Objective Loss 0.264892 LR 0.001000 Time 0.023448 +2023-10-02 20:44:32,942 - Epoch: [33][ 320/ 1236] Overall Loss 0.264969 Objective Loss 0.264969 LR 0.001000 Time 0.023370 +2023-10-02 20:44:33,151 - Epoch: [33][ 330/ 1236] Overall Loss 0.265208 Objective Loss 0.265208 LR 0.001000 Time 0.023295 +2023-10-02 20:44:33,359 - Epoch: [33][ 340/ 1236] Overall Loss 0.264893 Objective Loss 0.264893 LR 0.001000 Time 0.023219 +2023-10-02 20:44:33,568 - Epoch: [33][ 350/ 1236] Overall Loss 0.265128 Objective Loss 0.265128 LR 0.001000 Time 0.023153 +2023-10-02 20:44:33,778 - Epoch: [33][ 360/ 1236] Overall Loss 0.264434 Objective Loss 0.264434 LR 0.001000 Time 0.023092 +2023-10-02 20:44:33,988 - Epoch: [33][ 370/ 1236] Overall Loss 0.263923 Objective Loss 0.263923 LR 0.001000 Time 0.023035 +2023-10-02 20:44:34,198 - Epoch: [33][ 380/ 1236] Overall Loss 0.264065 Objective Loss 0.264065 LR 0.001000 Time 0.022980 +2023-10-02 20:44:34,408 - Epoch: [33][ 390/ 1236] Overall Loss 0.265398 Objective Loss 0.265398 LR 0.001000 Time 0.022925 +2023-10-02 20:44:34,618 - Epoch: [33][ 400/ 1236] Overall Loss 0.265724 Objective Loss 0.265724 LR 0.001000 Time 0.022876 +2023-10-02 20:44:34,828 - Epoch: [33][ 410/ 1236] Overall Loss 0.265305 Objective Loss 0.265305 LR 0.001000 Time 0.022826 +2023-10-02 20:44:35,038 - Epoch: [33][ 420/ 1236] Overall Loss 0.265707 Objective Loss 0.265707 LR 0.001000 Time 0.022781 +2023-10-02 20:44:35,248 - Epoch: [33][ 430/ 1236] Overall Loss 0.265498 Objective Loss 0.265498 LR 0.001000 Time 0.022736 +2023-10-02 20:44:35,458 - Epoch: [33][ 440/ 1236] Overall Loss 0.265943 Objective Loss 0.265943 LR 0.001000 Time 0.022695 +2023-10-02 20:44:35,669 - Epoch: [33][ 450/ 1236] Overall Loss 0.265303 Objective Loss 0.265303 LR 0.001000 Time 0.022658 +2023-10-02 20:44:35,879 - Epoch: [33][ 460/ 1236] Overall Loss 0.265909 Objective Loss 0.265909 LR 0.001000 Time 0.022622 +2023-10-02 20:44:36,090 - Epoch: [33][ 470/ 1236] Overall Loss 0.266034 Objective Loss 0.266034 LR 0.001000 Time 0.022589 +2023-10-02 20:44:36,301 - Epoch: [33][ 480/ 1236] Overall Loss 0.266237 Objective Loss 0.266237 LR 0.001000 Time 0.022556 +2023-10-02 20:44:36,512 - Epoch: [33][ 490/ 1236] Overall Loss 0.266829 Objective Loss 0.266829 LR 0.001000 Time 0.022523 +2023-10-02 20:44:36,723 - Epoch: [33][ 500/ 1236] Overall Loss 0.266601 Objective Loss 0.266601 LR 0.001000 Time 0.022493 +2023-10-02 20:44:36,933 - Epoch: [33][ 510/ 1236] Overall Loss 0.267082 Objective Loss 0.267082 LR 0.001000 Time 0.022462 +2023-10-02 20:44:37,144 - Epoch: [33][ 520/ 1236] Overall Loss 0.266816 Objective Loss 0.266816 LR 0.001000 Time 0.022433 +2023-10-02 20:44:37,354 - Epoch: [33][ 530/ 1236] Overall Loss 0.266593 Objective Loss 0.266593 LR 0.001000 Time 0.022407 +2023-10-02 20:44:37,565 - Epoch: [33][ 540/ 1236] Overall Loss 0.265905 Objective Loss 0.265905 LR 0.001000 Time 0.022381 +2023-10-02 20:44:37,776 - Epoch: [33][ 550/ 1236] Overall Loss 0.265734 Objective Loss 0.265734 LR 0.001000 Time 0.022354 +2023-10-02 20:44:37,986 - Epoch: [33][ 560/ 1236] Overall Loss 0.265656 Objective Loss 0.265656 LR 0.001000 Time 0.022330 +2023-10-02 20:44:38,196 - Epoch: [33][ 570/ 1236] Overall Loss 0.265880 Objective Loss 0.265880 LR 0.001000 Time 0.022304 +2023-10-02 20:44:38,407 - Epoch: [33][ 580/ 1236] Overall Loss 0.265811 Objective Loss 0.265811 LR 0.001000 Time 0.022282 +2023-10-02 20:44:38,618 - Epoch: [33][ 590/ 1236] Overall Loss 0.265683 Objective Loss 0.265683 LR 0.001000 Time 0.022260 +2023-10-02 20:44:38,828 - Epoch: [33][ 600/ 1236] Overall Loss 0.265858 Objective Loss 0.265858 LR 0.001000 Time 0.022239 +2023-10-02 20:44:39,039 - Epoch: [33][ 610/ 1236] Overall Loss 0.266340 Objective Loss 0.266340 LR 0.001000 Time 0.022220 +2023-10-02 20:44:39,249 - Epoch: [33][ 620/ 1236] Overall Loss 0.266118 Objective Loss 0.266118 LR 0.001000 Time 0.022200 +2023-10-02 20:44:39,460 - Epoch: [33][ 630/ 1236] Overall Loss 0.266277 Objective Loss 0.266277 LR 0.001000 Time 0.022181 +2023-10-02 20:44:39,671 - Epoch: [33][ 640/ 1236] Overall Loss 0.266186 Objective Loss 0.266186 LR 0.001000 Time 0.022163 +2023-10-02 20:44:39,882 - Epoch: [33][ 650/ 1236] Overall Loss 0.266221 Objective Loss 0.266221 LR 0.001000 Time 0.022146 +2023-10-02 20:44:40,092 - Epoch: [33][ 660/ 1236] Overall Loss 0.266069 Objective Loss 0.266069 LR 0.001000 Time 0.022128 +2023-10-02 20:44:40,303 - Epoch: [33][ 670/ 1236] Overall Loss 0.265909 Objective Loss 0.265909 LR 0.001000 Time 0.022113 +2023-10-02 20:44:40,513 - Epoch: [33][ 680/ 1236] Overall Loss 0.265695 Objective Loss 0.265695 LR 0.001000 Time 0.022096 +2023-10-02 20:44:40,724 - Epoch: [33][ 690/ 1236] Overall Loss 0.265784 Objective Loss 0.265784 LR 0.001000 Time 0.022079 +2023-10-02 20:44:40,935 - Epoch: [33][ 700/ 1236] Overall Loss 0.265923 Objective Loss 0.265923 LR 0.001000 Time 0.022063 +2023-10-02 20:44:41,146 - Epoch: [33][ 710/ 1236] Overall Loss 0.265626 Objective Loss 0.265626 LR 0.001000 Time 0.022049 +2023-10-02 20:44:41,356 - Epoch: [33][ 720/ 1236] Overall Loss 0.265600 Objective Loss 0.265600 LR 0.001000 Time 0.022034 +2023-10-02 20:44:41,567 - Epoch: [33][ 730/ 1236] Overall Loss 0.265773 Objective Loss 0.265773 LR 0.001000 Time 0.022021 +2023-10-02 20:44:41,777 - Epoch: [33][ 740/ 1236] Overall Loss 0.266214 Objective Loss 0.266214 LR 0.001000 Time 0.022006 +2023-10-02 20:44:41,988 - Epoch: [33][ 750/ 1236] Overall Loss 0.265974 Objective Loss 0.265974 LR 0.001000 Time 0.021994 +2023-10-02 20:44:42,198 - Epoch: [33][ 760/ 1236] Overall Loss 0.265898 Objective Loss 0.265898 LR 0.001000 Time 0.021980 +2023-10-02 20:44:42,410 - Epoch: [33][ 770/ 1236] Overall Loss 0.265830 Objective Loss 0.265830 LR 0.001000 Time 0.021967 +2023-10-02 20:44:42,620 - Epoch: [33][ 780/ 1236] Overall Loss 0.265540 Objective Loss 0.265540 LR 0.001000 Time 0.021954 +2023-10-02 20:44:42,831 - Epoch: [33][ 790/ 1236] Overall Loss 0.265544 Objective Loss 0.265544 LR 0.001000 Time 0.021943 +2023-10-02 20:44:43,041 - Epoch: [33][ 800/ 1236] Overall Loss 0.265253 Objective Loss 0.265253 LR 0.001000 Time 0.021931 +2023-10-02 20:44:43,252 - Epoch: [33][ 810/ 1236] Overall Loss 0.265332 Objective Loss 0.265332 LR 0.001000 Time 0.021919 +2023-10-02 20:44:43,462 - Epoch: [33][ 820/ 1236] Overall Loss 0.265026 Objective Loss 0.265026 LR 0.001000 Time 0.021906 +2023-10-02 20:44:43,673 - Epoch: [33][ 830/ 1236] Overall Loss 0.264676 Objective Loss 0.264676 LR 0.001000 Time 0.021896 +2023-10-02 20:44:43,883 - Epoch: [33][ 840/ 1236] Overall Loss 0.264616 Objective Loss 0.264616 LR 0.001000 Time 0.021885 +2023-10-02 20:44:44,094 - Epoch: [33][ 850/ 1236] Overall Loss 0.264533 Objective Loss 0.264533 LR 0.001000 Time 0.021875 +2023-10-02 20:44:44,304 - Epoch: [33][ 860/ 1236] Overall Loss 0.264660 Objective Loss 0.264660 LR 0.001000 Time 0.021865 +2023-10-02 20:44:44,516 - Epoch: [33][ 870/ 1236] Overall Loss 0.264536 Objective Loss 0.264536 LR 0.001000 Time 0.021856 +2023-10-02 20:44:44,726 - Epoch: [33][ 880/ 1236] Overall Loss 0.264500 Objective Loss 0.264500 LR 0.001000 Time 0.021846 +2023-10-02 20:44:44,937 - Epoch: [33][ 890/ 1236] Overall Loss 0.264461 Objective Loss 0.264461 LR 0.001000 Time 0.021837 +2023-10-02 20:44:45,147 - Epoch: [33][ 900/ 1236] Overall Loss 0.264712 Objective Loss 0.264712 LR 0.001000 Time 0.021828 +2023-10-02 20:44:45,358 - Epoch: [33][ 910/ 1236] Overall Loss 0.264625 Objective Loss 0.264625 LR 0.001000 Time 0.021819 +2023-10-02 20:44:45,568 - Epoch: [33][ 920/ 1236] Overall Loss 0.264783 Objective Loss 0.264783 LR 0.001000 Time 0.021810 +2023-10-02 20:44:45,780 - Epoch: [33][ 930/ 1236] Overall Loss 0.265068 Objective Loss 0.265068 LR 0.001000 Time 0.021802 +2023-10-02 20:44:45,990 - Epoch: [33][ 940/ 1236] Overall Loss 0.264955 Objective Loss 0.264955 LR 0.001000 Time 0.021793 +2023-10-02 20:44:46,202 - Epoch: [33][ 950/ 1236] Overall Loss 0.265345 Objective Loss 0.265345 LR 0.001000 Time 0.021786 +2023-10-02 20:44:46,412 - Epoch: [33][ 960/ 1236] Overall Loss 0.265397 Objective Loss 0.265397 LR 0.001000 Time 0.021778 +2023-10-02 20:44:46,623 - Epoch: [33][ 970/ 1236] Overall Loss 0.265618 Objective Loss 0.265618 LR 0.001000 Time 0.021770 +2023-10-02 20:44:46,833 - Epoch: [33][ 980/ 1236] Overall Loss 0.265669 Objective Loss 0.265669 LR 0.001000 Time 0.021762 +2023-10-02 20:44:47,044 - Epoch: [33][ 990/ 1236] Overall Loss 0.265715 Objective Loss 0.265715 LR 0.001000 Time 0.021755 +2023-10-02 20:44:47,254 - Epoch: [33][ 1000/ 1236] Overall Loss 0.265994 Objective Loss 0.265994 LR 0.001000 Time 0.021747 +2023-10-02 20:44:47,465 - Epoch: [33][ 1010/ 1236] Overall Loss 0.265941 Objective Loss 0.265941 LR 0.001000 Time 0.021740 +2023-10-02 20:44:47,675 - Epoch: [33][ 1020/ 1236] Overall Loss 0.265522 Objective Loss 0.265522 LR 0.001000 Time 0.021733 +2023-10-02 20:44:47,886 - Epoch: [33][ 1030/ 1236] Overall Loss 0.265241 Objective Loss 0.265241 LR 0.001000 Time 0.021726 +2023-10-02 20:44:48,096 - Epoch: [33][ 1040/ 1236] Overall Loss 0.265504 Objective Loss 0.265504 LR 0.001000 Time 0.021719 +2023-10-02 20:44:48,308 - Epoch: [33][ 1050/ 1236] Overall Loss 0.265747 Objective Loss 0.265747 LR 0.001000 Time 0.021713 +2023-10-02 20:44:48,517 - Epoch: [33][ 1060/ 1236] Overall Loss 0.266088 Objective Loss 0.266088 LR 0.001000 Time 0.021705 +2023-10-02 20:44:48,728 - Epoch: [33][ 1070/ 1236] Overall Loss 0.266343 Objective Loss 0.266343 LR 0.001000 Time 0.021699 +2023-10-02 20:44:48,938 - Epoch: [33][ 1080/ 1236] Overall Loss 0.266562 Objective Loss 0.266562 LR 0.001000 Time 0.021692 +2023-10-02 20:44:49,150 - Epoch: [33][ 1090/ 1236] Overall Loss 0.266671 Objective Loss 0.266671 LR 0.001000 Time 0.021687 +2023-10-02 20:44:49,359 - Epoch: [33][ 1100/ 1236] Overall Loss 0.266932 Objective Loss 0.266932 LR 0.001000 Time 0.021680 +2023-10-02 20:44:49,570 - Epoch: [33][ 1110/ 1236] Overall Loss 0.267014 Objective Loss 0.267014 LR 0.001000 Time 0.021674 +2023-10-02 20:44:49,780 - Epoch: [33][ 1120/ 1236] Overall Loss 0.267058 Objective Loss 0.267058 LR 0.001000 Time 0.021668 +2023-10-02 20:44:49,991 - Epoch: [33][ 1130/ 1236] Overall Loss 0.267193 Objective Loss 0.267193 LR 0.001000 Time 0.021662 +2023-10-02 20:44:50,201 - Epoch: [33][ 1140/ 1236] Overall Loss 0.267309 Objective Loss 0.267309 LR 0.001000 Time 0.021656 +2023-10-02 20:44:50,412 - Epoch: [33][ 1150/ 1236] Overall Loss 0.267031 Objective Loss 0.267031 LR 0.001000 Time 0.021651 +2023-10-02 20:44:50,622 - Epoch: [33][ 1160/ 1236] Overall Loss 0.267136 Objective Loss 0.267136 LR 0.001000 Time 0.021645 +2023-10-02 20:44:50,833 - Epoch: [33][ 1170/ 1236] Overall Loss 0.267082 Objective Loss 0.267082 LR 0.001000 Time 0.021640 +2023-10-02 20:44:51,043 - Epoch: [33][ 1180/ 1236] Overall Loss 0.267617 Objective Loss 0.267617 LR 0.001000 Time 0.021634 +2023-10-02 20:44:51,254 - Epoch: [33][ 1190/ 1236] Overall Loss 0.267672 Objective Loss 0.267672 LR 0.001000 Time 0.021628 +2023-10-02 20:44:51,464 - Epoch: [33][ 1200/ 1236] Overall Loss 0.267848 Objective Loss 0.267848 LR 0.001000 Time 0.021623 +2023-10-02 20:44:51,675 - Epoch: [33][ 1210/ 1236] Overall Loss 0.268029 Objective Loss 0.268029 LR 0.001000 Time 0.021618 +2023-10-02 20:44:51,885 - Epoch: [33][ 1220/ 1236] Overall Loss 0.268223 Objective Loss 0.268223 LR 0.001000 Time 0.021613 +2023-10-02 20:44:52,149 - Epoch: [33][ 1230/ 1236] Overall Loss 0.268353 Objective Loss 0.268353 LR 0.001000 Time 0.021652 +2023-10-02 20:44:52,272 - Epoch: [33][ 1236/ 1236] Overall Loss 0.268675 Objective Loss 0.268675 Top1 84.725051 Top5 96.741344 LR 0.001000 Time 0.021646 +2023-10-02 20:44:52,394 - --- validate (epoch=33)----------- +2023-10-02 20:44:52,394 - 29943 samples (256 per mini-batch) +2023-10-02 20:44:52,863 - Epoch: [33][ 10/ 117] Loss 0.328059 Top1 83.242188 Top5 98.359375 +2023-10-02 20:44:53,014 - Epoch: [33][ 20/ 117] Loss 0.331249 Top1 83.046875 Top5 98.027344 +2023-10-02 20:44:53,165 - Epoch: [33][ 30/ 117] Loss 0.328929 Top1 83.229167 Top5 98.046875 +2023-10-02 20:44:53,316 - Epoch: [33][ 40/ 117] Loss 0.325521 Top1 83.427734 Top5 98.076172 +2023-10-02 20:44:53,467 - Epoch: [33][ 50/ 117] Loss 0.337300 Top1 82.976562 Top5 97.992188 +2023-10-02 20:44:53,618 - Epoch: [33][ 60/ 117] Loss 0.332144 Top1 83.040365 Top5 97.955729 +2023-10-02 20:44:53,768 - Epoch: [33][ 70/ 117] Loss 0.332125 Top1 82.918527 Top5 97.912946 +2023-10-02 20:44:53,920 - Epoch: [33][ 80/ 117] Loss 0.331402 Top1 82.958984 Top5 97.910156 +2023-10-02 20:44:54,072 - Epoch: [33][ 90/ 117] Loss 0.332603 Top1 82.981771 Top5 97.968750 +2023-10-02 20:44:54,224 - Epoch: [33][ 100/ 117] Loss 0.333255 Top1 83.042969 Top5 98.003906 +2023-10-02 20:44:54,382 - Epoch: [33][ 110/ 117] Loss 0.331899 Top1 83.029119 Top5 98.014915 +2023-10-02 20:44:54,472 - Epoch: [33][ 117/ 117] Loss 0.330572 Top1 83.107905 Top5 98.052967 +2023-10-02 20:44:54,607 - ==> Top1: 83.108 Top5: 98.053 Loss: 0.331 + +2023-10-02 20:44:54,608 - ==> Confusion: +[[ 886 2 6 0 6 3 0 0 2 108 2 1 2 2 7 2 1 3 0 0 17] + [ 1 1038 1 1 6 18 1 27 5 3 3 0 0 0 2 5 3 0 8 3 6] + [ 1 0 970 9 2 0 20 8 0 4 2 1 7 4 0 6 0 1 9 5 7] + [ 1 2 20 968 0 3 1 3 9 1 9 1 4 9 30 1 2 2 7 0 16] + [ 20 4 3 0 961 2 1 0 1 14 1 0 0 6 14 4 12 0 0 1 6] + [ 1 45 1 1 2 971 0 30 4 6 3 8 0 15 7 1 6 0 2 4 9] + [ 0 3 40 0 0 0 1114 3 0 0 4 2 2 0 0 6 0 0 2 8 7] + [ 2 9 21 1 6 18 3 1070 3 4 3 8 6 3 4 1 1 0 38 8 9] + [ 16 0 0 1 2 2 0 0 978 49 11 1 2 9 13 1 1 1 2 0 0] + [ 65 1 0 0 4 1 1 0 36 967 1 1 0 26 5 1 1 0 0 1 8] + [ 1 0 8 11 0 0 5 4 20 1 953 3 1 19 5 0 0 1 12 1 8] + [ 1 0 3 0 0 13 1 2 0 3 0 934 40 6 0 4 3 17 0 5 3] + [ 0 1 2 0 0 5 1 2 1 3 0 52 956 1 0 13 2 20 0 3 6] + [ 0 0 1 0 3 11 1 0 13 15 4 6 1 1046 2 0 0 3 0 3 10] + [ 9 1 6 13 5 0 0 0 21 5 1 0 4 4 1014 1 1 4 3 0 9] + [ 0 0 1 2 4 0 0 0 0 1 1 7 10 1 0 1069 17 7 2 8 4] + [ 0 9 3 0 4 3 2 0 1 0 0 7 3 4 4 14 1088 2 1 5 11] + [ 2 0 0 4 0 0 2 0 1 1 0 8 23 1 2 9 2 980 1 1 1] + [ 3 6 9 14 0 2 0 26 10 0 1 2 4 0 18 0 0 1 959 0 13] + [ 0 0 2 1 1 7 6 3 0 1 1 16 7 3 1 2 4 3 1 1089 4] + [ 110 166 167 55 78 166 30 107 150 141 194 160 355 333 163 61 122 74 142 257 4874]] + +2023-10-02 20:44:54,609 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:44:54,609 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:44:54,616 - + +2023-10-02 20:44:54,616 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:44:55,619 - Epoch: [34][ 10/ 1236] Overall Loss 0.256775 Objective Loss 0.256775 LR 0.001000 Time 0.100331 +2023-10-02 20:44:55,830 - Epoch: [34][ 20/ 1236] Overall Loss 0.265402 Objective Loss 0.265402 LR 0.001000 Time 0.060672 +2023-10-02 20:44:56,038 - Epoch: [34][ 30/ 1236] Overall Loss 0.256243 Objective Loss 0.256243 LR 0.001000 Time 0.047329 +2023-10-02 20:44:56,248 - Epoch: [34][ 40/ 1236] Overall Loss 0.250131 Objective Loss 0.250131 LR 0.001000 Time 0.040753 +2023-10-02 20:44:56,455 - Epoch: [34][ 50/ 1236] Overall Loss 0.248368 Objective Loss 0.248368 LR 0.001000 Time 0.036730 +2023-10-02 20:44:56,666 - Epoch: [34][ 60/ 1236] Overall Loss 0.255527 Objective Loss 0.255527 LR 0.001000 Time 0.034113 +2023-10-02 20:44:56,872 - Epoch: [34][ 70/ 1236] Overall Loss 0.258864 Objective Loss 0.258864 LR 0.001000 Time 0.032189 +2023-10-02 20:44:57,081 - Epoch: [34][ 80/ 1236] Overall Loss 0.257493 Objective Loss 0.257493 LR 0.001000 Time 0.030773 +2023-10-02 20:44:57,289 - Epoch: [34][ 90/ 1236] Overall Loss 0.258104 Objective Loss 0.258104 LR 0.001000 Time 0.029646 +2023-10-02 20:44:57,500 - Epoch: [34][ 100/ 1236] Overall Loss 0.262091 Objective Loss 0.262091 LR 0.001000 Time 0.028786 +2023-10-02 20:44:57,706 - Epoch: [34][ 110/ 1236] Overall Loss 0.258847 Objective Loss 0.258847 LR 0.001000 Time 0.028039 +2023-10-02 20:44:57,914 - Epoch: [34][ 120/ 1236] Overall Loss 0.259942 Objective Loss 0.259942 LR 0.001000 Time 0.027440 +2023-10-02 20:44:58,122 - Epoch: [34][ 130/ 1236] Overall Loss 0.259322 Objective Loss 0.259322 LR 0.001000 Time 0.026922 +2023-10-02 20:44:58,331 - Epoch: [34][ 140/ 1236] Overall Loss 0.259155 Objective Loss 0.259155 LR 0.001000 Time 0.026489 +2023-10-02 20:44:58,539 - Epoch: [34][ 150/ 1236] Overall Loss 0.259352 Objective Loss 0.259352 LR 0.001000 Time 0.026099 +2023-10-02 20:44:58,747 - Epoch: [34][ 160/ 1236] Overall Loss 0.259301 Objective Loss 0.259301 LR 0.001000 Time 0.025770 +2023-10-02 20:44:58,954 - Epoch: [34][ 170/ 1236] Overall Loss 0.260433 Objective Loss 0.260433 LR 0.001000 Time 0.025464 +2023-10-02 20:44:59,163 - Epoch: [34][ 180/ 1236] Overall Loss 0.260996 Objective Loss 0.260996 LR 0.001000 Time 0.025208 +2023-10-02 20:44:59,370 - Epoch: [34][ 190/ 1236] Overall Loss 0.259915 Objective Loss 0.259915 LR 0.001000 Time 0.024963 +2023-10-02 20:44:59,579 - Epoch: [34][ 200/ 1236] Overall Loss 0.260836 Objective Loss 0.260836 LR 0.001000 Time 0.024759 +2023-10-02 20:44:59,787 - Epoch: [34][ 210/ 1236] Overall Loss 0.262734 Objective Loss 0.262734 LR 0.001000 Time 0.024560 +2023-10-02 20:44:59,995 - Epoch: [34][ 220/ 1236] Overall Loss 0.262671 Objective Loss 0.262671 LR 0.001000 Time 0.024391 +2023-10-02 20:45:00,203 - Epoch: [34][ 230/ 1236] Overall Loss 0.263076 Objective Loss 0.263076 LR 0.001000 Time 0.024230 +2023-10-02 20:45:00,411 - Epoch: [34][ 240/ 1236] Overall Loss 0.263760 Objective Loss 0.263760 LR 0.001000 Time 0.024090 +2023-10-02 20:45:00,619 - Epoch: [34][ 250/ 1236] Overall Loss 0.264644 Objective Loss 0.264644 LR 0.001000 Time 0.023950 +2023-10-02 20:45:00,828 - Epoch: [34][ 260/ 1236] Overall Loss 0.264616 Objective Loss 0.264616 LR 0.001000 Time 0.023831 +2023-10-02 20:45:01,035 - Epoch: [34][ 270/ 1236] Overall Loss 0.264439 Objective Loss 0.264439 LR 0.001000 Time 0.023710 +2023-10-02 20:45:01,244 - Epoch: [34][ 280/ 1236] Overall Loss 0.264457 Objective Loss 0.264457 LR 0.001000 Time 0.023609 +2023-10-02 20:45:01,451 - Epoch: [34][ 290/ 1236] Overall Loss 0.265116 Objective Loss 0.265116 LR 0.001000 Time 0.023504 +2023-10-02 20:45:01,661 - Epoch: [34][ 300/ 1236] Overall Loss 0.265369 Objective Loss 0.265369 LR 0.001000 Time 0.023421 +2023-10-02 20:45:01,867 - Epoch: [34][ 310/ 1236] Overall Loss 0.264653 Objective Loss 0.264653 LR 0.001000 Time 0.023329 +2023-10-02 20:45:02,077 - Epoch: [34][ 320/ 1236] Overall Loss 0.264192 Objective Loss 0.264192 LR 0.001000 Time 0.023256 +2023-10-02 20:45:02,284 - Epoch: [34][ 330/ 1236] Overall Loss 0.265485 Objective Loss 0.265485 LR 0.001000 Time 0.023175 +2023-10-02 20:45:02,493 - Epoch: [34][ 340/ 1236] Overall Loss 0.264905 Objective Loss 0.264905 LR 0.001000 Time 0.023108 +2023-10-02 20:45:02,700 - Epoch: [34][ 350/ 1236] Overall Loss 0.265362 Objective Loss 0.265362 LR 0.001000 Time 0.023037 +2023-10-02 20:45:02,911 - Epoch: [34][ 360/ 1236] Overall Loss 0.266157 Objective Loss 0.266157 LR 0.001000 Time 0.022982 +2023-10-02 20:45:03,118 - Epoch: [34][ 370/ 1236] Overall Loss 0.266447 Objective Loss 0.266447 LR 0.001000 Time 0.022919 +2023-10-02 20:45:03,329 - Epoch: [34][ 380/ 1236] Overall Loss 0.265811 Objective Loss 0.265811 LR 0.001000 Time 0.022870 +2023-10-02 20:45:03,536 - Epoch: [34][ 390/ 1236] Overall Loss 0.265028 Objective Loss 0.265028 LR 0.001000 Time 0.022813 +2023-10-02 20:45:03,747 - Epoch: [34][ 400/ 1236] Overall Loss 0.264523 Objective Loss 0.264523 LR 0.001000 Time 0.022770 +2023-10-02 20:45:03,954 - Epoch: [34][ 410/ 1236] Overall Loss 0.265489 Objective Loss 0.265489 LR 0.001000 Time 0.022718 +2023-10-02 20:45:04,165 - Epoch: [34][ 420/ 1236] Overall Loss 0.264660 Objective Loss 0.264660 LR 0.001000 Time 0.022679 +2023-10-02 20:45:04,372 - Epoch: [34][ 430/ 1236] Overall Loss 0.265169 Objective Loss 0.265169 LR 0.001000 Time 0.022632 +2023-10-02 20:45:04,582 - Epoch: [34][ 440/ 1236] Overall Loss 0.265458 Objective Loss 0.265458 LR 0.001000 Time 0.022596 +2023-10-02 20:45:04,790 - Epoch: [34][ 450/ 1236] Overall Loss 0.265561 Objective Loss 0.265561 LR 0.001000 Time 0.022554 +2023-10-02 20:45:05,000 - Epoch: [34][ 460/ 1236] Overall Loss 0.265429 Objective Loss 0.265429 LR 0.001000 Time 0.022521 +2023-10-02 20:45:05,207 - Epoch: [34][ 470/ 1236] Overall Loss 0.264976 Objective Loss 0.264976 LR 0.001000 Time 0.022481 +2023-10-02 20:45:05,418 - Epoch: [34][ 480/ 1236] Overall Loss 0.265038 Objective Loss 0.265038 LR 0.001000 Time 0.022452 +2023-10-02 20:45:05,625 - Epoch: [34][ 490/ 1236] Overall Loss 0.265163 Objective Loss 0.265163 LR 0.001000 Time 0.022416 +2023-10-02 20:45:05,836 - Epoch: [34][ 500/ 1236] Overall Loss 0.265683 Objective Loss 0.265683 LR 0.001000 Time 0.022389 +2023-10-02 20:45:06,043 - Epoch: [34][ 510/ 1236] Overall Loss 0.265511 Objective Loss 0.265511 LR 0.001000 Time 0.022355 +2023-10-02 20:45:06,254 - Epoch: [34][ 520/ 1236] Overall Loss 0.266564 Objective Loss 0.266564 LR 0.001000 Time 0.022331 +2023-10-02 20:45:06,461 - Epoch: [34][ 530/ 1236] Overall Loss 0.266847 Objective Loss 0.266847 LR 0.001000 Time 0.022299 +2023-10-02 20:45:06,672 - Epoch: [34][ 540/ 1236] Overall Loss 0.266716 Objective Loss 0.266716 LR 0.001000 Time 0.022277 +2023-10-02 20:45:06,879 - Epoch: [34][ 550/ 1236] Overall Loss 0.266685 Objective Loss 0.266685 LR 0.001000 Time 0.022247 +2023-10-02 20:45:07,090 - Epoch: [34][ 560/ 1236] Overall Loss 0.266566 Objective Loss 0.266566 LR 0.001000 Time 0.022226 +2023-10-02 20:45:07,297 - Epoch: [34][ 570/ 1236] Overall Loss 0.266494 Objective Loss 0.266494 LR 0.001000 Time 0.022199 +2023-10-02 20:45:07,508 - Epoch: [34][ 580/ 1236] Overall Loss 0.267163 Objective Loss 0.267163 LR 0.001000 Time 0.022180 +2023-10-02 20:45:07,715 - Epoch: [34][ 590/ 1236] Overall Loss 0.266875 Objective Loss 0.266875 LR 0.001000 Time 0.022154 +2023-10-02 20:45:07,926 - Epoch: [34][ 600/ 1236] Overall Loss 0.267112 Objective Loss 0.267112 LR 0.001000 Time 0.022136 +2023-10-02 20:45:08,133 - Epoch: [34][ 610/ 1236] Overall Loss 0.267407 Objective Loss 0.267407 LR 0.001000 Time 0.022112 +2023-10-02 20:45:08,344 - Epoch: [34][ 620/ 1236] Overall Loss 0.267930 Objective Loss 0.267930 LR 0.001000 Time 0.022095 +2023-10-02 20:45:08,551 - Epoch: [34][ 630/ 1236] Overall Loss 0.267903 Objective Loss 0.267903 LR 0.001000 Time 0.022073 +2023-10-02 20:45:08,762 - Epoch: [34][ 640/ 1236] Overall Loss 0.267685 Objective Loss 0.267685 LR 0.001000 Time 0.022057 +2023-10-02 20:45:08,969 - Epoch: [34][ 650/ 1236] Overall Loss 0.267591 Objective Loss 0.267591 LR 0.001000 Time 0.022036 +2023-10-02 20:45:09,180 - Epoch: [34][ 660/ 1236] Overall Loss 0.268263 Objective Loss 0.268263 LR 0.001000 Time 0.022022 +2023-10-02 20:45:09,387 - Epoch: [34][ 670/ 1236] Overall Loss 0.268561 Objective Loss 0.268561 LR 0.001000 Time 0.022001 +2023-10-02 20:45:09,597 - Epoch: [34][ 680/ 1236] Overall Loss 0.268508 Objective Loss 0.268508 LR 0.001000 Time 0.021986 +2023-10-02 20:45:09,805 - Epoch: [34][ 690/ 1236] Overall Loss 0.268856 Objective Loss 0.268856 LR 0.001000 Time 0.021967 +2023-10-02 20:45:10,016 - Epoch: [34][ 700/ 1236] Overall Loss 0.268874 Objective Loss 0.268874 LR 0.001000 Time 0.021954 +2023-10-02 20:45:10,223 - Epoch: [34][ 710/ 1236] Overall Loss 0.268937 Objective Loss 0.268937 LR 0.001000 Time 0.021936 +2023-10-02 20:45:10,434 - Epoch: [34][ 720/ 1236] Overall Loss 0.269005 Objective Loss 0.269005 LR 0.001000 Time 0.021924 +2023-10-02 20:45:10,641 - Epoch: [34][ 730/ 1236] Overall Loss 0.268579 Objective Loss 0.268579 LR 0.001000 Time 0.021906 +2023-10-02 20:45:10,852 - Epoch: [34][ 740/ 1236] Overall Loss 0.268699 Objective Loss 0.268699 LR 0.001000 Time 0.021895 +2023-10-02 20:45:11,059 - Epoch: [34][ 750/ 1236] Overall Loss 0.268724 Objective Loss 0.268724 LR 0.001000 Time 0.021879 +2023-10-02 20:45:11,270 - Epoch: [34][ 760/ 1236] Overall Loss 0.268670 Objective Loss 0.268670 LR 0.001000 Time 0.021868 +2023-10-02 20:45:11,477 - Epoch: [34][ 770/ 1236] Overall Loss 0.269010 Objective Loss 0.269010 LR 0.001000 Time 0.021853 +2023-10-02 20:45:11,688 - Epoch: [34][ 780/ 1236] Overall Loss 0.268974 Objective Loss 0.268974 LR 0.001000 Time 0.021843 +2023-10-02 20:45:11,896 - Epoch: [34][ 790/ 1236] Overall Loss 0.269286 Objective Loss 0.269286 LR 0.001000 Time 0.021830 +2023-10-02 20:45:12,108 - Epoch: [34][ 800/ 1236] Overall Loss 0.269492 Objective Loss 0.269492 LR 0.001000 Time 0.021821 +2023-10-02 20:45:12,315 - Epoch: [34][ 810/ 1236] Overall Loss 0.269457 Objective Loss 0.269457 LR 0.001000 Time 0.021807 +2023-10-02 20:45:12,525 - Epoch: [34][ 820/ 1236] Overall Loss 0.270100 Objective Loss 0.270100 LR 0.001000 Time 0.021797 +2023-10-02 20:45:12,733 - Epoch: [34][ 830/ 1236] Overall Loss 0.270426 Objective Loss 0.270426 LR 0.001000 Time 0.021783 +2023-10-02 20:45:12,945 - Epoch: [34][ 840/ 1236] Overall Loss 0.270838 Objective Loss 0.270838 LR 0.001000 Time 0.021775 +2023-10-02 20:45:13,152 - Epoch: [34][ 850/ 1236] Overall Loss 0.270791 Objective Loss 0.270791 LR 0.001000 Time 0.021763 +2023-10-02 20:45:13,363 - Epoch: [34][ 860/ 1236] Overall Loss 0.270790 Objective Loss 0.270790 LR 0.001000 Time 0.021755 +2023-10-02 20:45:13,570 - Epoch: [34][ 870/ 1236] Overall Loss 0.271117 Objective Loss 0.271117 LR 0.001000 Time 0.021742 +2023-10-02 20:45:13,784 - Epoch: [34][ 880/ 1236] Overall Loss 0.271211 Objective Loss 0.271211 LR 0.001000 Time 0.021738 +2023-10-02 20:45:13,999 - Epoch: [34][ 890/ 1236] Overall Loss 0.271274 Objective Loss 0.271274 LR 0.001000 Time 0.021735 +2023-10-02 20:45:14,220 - Epoch: [34][ 900/ 1236] Overall Loss 0.271323 Objective Loss 0.271323 LR 0.001000 Time 0.021739 +2023-10-02 20:45:14,428 - Epoch: [34][ 910/ 1236] Overall Loss 0.271451 Objective Loss 0.271451 LR 0.001000 Time 0.021728 +2023-10-02 20:45:14,636 - Epoch: [34][ 920/ 1236] Overall Loss 0.271716 Objective Loss 0.271716 LR 0.001000 Time 0.021718 +2023-10-02 20:45:14,845 - Epoch: [34][ 930/ 1236] Overall Loss 0.272191 Objective Loss 0.272191 LR 0.001000 Time 0.021707 +2023-10-02 20:45:15,054 - Epoch: [34][ 940/ 1236] Overall Loss 0.272251 Objective Loss 0.272251 LR 0.001000 Time 0.021698 +2023-10-02 20:45:15,261 - Epoch: [34][ 950/ 1236] Overall Loss 0.272472 Objective Loss 0.272472 LR 0.001000 Time 0.021686 +2023-10-02 20:45:15,470 - Epoch: [34][ 960/ 1236] Overall Loss 0.272799 Objective Loss 0.272799 LR 0.001000 Time 0.021678 +2023-10-02 20:45:15,678 - Epoch: [34][ 970/ 1236] Overall Loss 0.272981 Objective Loss 0.272981 LR 0.001000 Time 0.021667 +2023-10-02 20:45:15,887 - Epoch: [34][ 980/ 1236] Overall Loss 0.273058 Objective Loss 0.273058 LR 0.001000 Time 0.021659 +2023-10-02 20:45:16,095 - Epoch: [34][ 990/ 1236] Overall Loss 0.272827 Objective Loss 0.272827 LR 0.001000 Time 0.021648 +2023-10-02 20:45:16,304 - Epoch: [34][ 1000/ 1236] Overall Loss 0.272755 Objective Loss 0.272755 LR 0.001000 Time 0.021640 +2023-10-02 20:45:16,512 - Epoch: [34][ 1010/ 1236] Overall Loss 0.273173 Objective Loss 0.273173 LR 0.001000 Time 0.021630 +2023-10-02 20:45:16,721 - Epoch: [34][ 1020/ 1236] Overall Loss 0.273434 Objective Loss 0.273434 LR 0.001000 Time 0.021623 +2023-10-02 20:45:16,928 - Epoch: [34][ 1030/ 1236] Overall Loss 0.273271 Objective Loss 0.273271 LR 0.001000 Time 0.021613 +2023-10-02 20:45:17,137 - Epoch: [34][ 1040/ 1236] Overall Loss 0.273306 Objective Loss 0.273306 LR 0.001000 Time 0.021606 +2023-10-02 20:45:17,345 - Epoch: [34][ 1050/ 1236] Overall Loss 0.273345 Objective Loss 0.273345 LR 0.001000 Time 0.021596 +2023-10-02 20:45:17,554 - Epoch: [34][ 1060/ 1236] Overall Loss 0.273482 Objective Loss 0.273482 LR 0.001000 Time 0.021589 +2023-10-02 20:45:17,761 - Epoch: [34][ 1070/ 1236] Overall Loss 0.273567 Objective Loss 0.273567 LR 0.001000 Time 0.021580 +2023-10-02 20:45:17,970 - Epoch: [34][ 1080/ 1236] Overall Loss 0.273748 Objective Loss 0.273748 LR 0.001000 Time 0.021573 +2023-10-02 20:45:18,178 - Epoch: [34][ 1090/ 1236] Overall Loss 0.273832 Objective Loss 0.273832 LR 0.001000 Time 0.021565 +2023-10-02 20:45:18,387 - Epoch: [34][ 1100/ 1236] Overall Loss 0.273724 Objective Loss 0.273724 LR 0.001000 Time 0.021558 +2023-10-02 20:45:18,595 - Epoch: [34][ 1110/ 1236] Overall Loss 0.273661 Objective Loss 0.273661 LR 0.001000 Time 0.021550 +2023-10-02 20:45:18,804 - Epoch: [34][ 1120/ 1236] Overall Loss 0.273322 Objective Loss 0.273322 LR 0.001000 Time 0.021544 +2023-10-02 20:45:19,012 - Epoch: [34][ 1130/ 1236] Overall Loss 0.273277 Objective Loss 0.273277 LR 0.001000 Time 0.021536 +2023-10-02 20:45:19,221 - Epoch: [34][ 1140/ 1236] Overall Loss 0.273326 Objective Loss 0.273326 LR 0.001000 Time 0.021530 +2023-10-02 20:45:19,428 - Epoch: [34][ 1150/ 1236] Overall Loss 0.273259 Objective Loss 0.273259 LR 0.001000 Time 0.021522 +2023-10-02 20:45:19,637 - Epoch: [34][ 1160/ 1236] Overall Loss 0.273157 Objective Loss 0.273157 LR 0.001000 Time 0.021516 +2023-10-02 20:45:19,845 - Epoch: [34][ 1170/ 1236] Overall Loss 0.272855 Objective Loss 0.272855 LR 0.001000 Time 0.021509 +2023-10-02 20:45:20,054 - Epoch: [34][ 1180/ 1236] Overall Loss 0.272683 Objective Loss 0.272683 LR 0.001000 Time 0.021503 +2023-10-02 20:45:20,262 - Epoch: [34][ 1190/ 1236] Overall Loss 0.272595 Objective Loss 0.272595 LR 0.001000 Time 0.021496 +2023-10-02 20:45:20,471 - Epoch: [34][ 1200/ 1236] Overall Loss 0.272592 Objective Loss 0.272592 LR 0.001000 Time 0.021491 +2023-10-02 20:45:20,679 - Epoch: [34][ 1210/ 1236] Overall Loss 0.272613 Objective Loss 0.272613 LR 0.001000 Time 0.021484 +2023-10-02 20:45:20,888 - Epoch: [34][ 1220/ 1236] Overall Loss 0.272394 Objective Loss 0.272394 LR 0.001000 Time 0.021479 +2023-10-02 20:45:21,149 - Epoch: [34][ 1230/ 1236] Overall Loss 0.272239 Objective Loss 0.272239 LR 0.001000 Time 0.021515 +2023-10-02 20:45:21,272 - Epoch: [34][ 1236/ 1236] Overall Loss 0.272190 Objective Loss 0.272190 Top1 86.354379 Top5 99.389002 LR 0.001000 Time 0.021510 +2023-10-02 20:45:21,411 - --- validate (epoch=34)----------- +2023-10-02 20:45:21,411 - 29943 samples (256 per mini-batch) +2023-10-02 20:45:21,896 - Epoch: [34][ 10/ 117] Loss 0.325640 Top1 83.359375 Top5 98.203125 +2023-10-02 20:45:22,049 - Epoch: [34][ 20/ 117] Loss 0.307293 Top1 83.554688 Top5 98.339844 +2023-10-02 20:45:22,199 - Epoch: [34][ 30/ 117] Loss 0.304123 Top1 83.697917 Top5 98.346354 +2023-10-02 20:45:22,351 - Epoch: [34][ 40/ 117] Loss 0.311082 Top1 83.935547 Top5 98.330078 +2023-10-02 20:45:22,504 - Epoch: [34][ 50/ 117] Loss 0.315541 Top1 83.812500 Top5 98.335938 +2023-10-02 20:45:22,655 - Epoch: [34][ 60/ 117] Loss 0.319316 Top1 83.652344 Top5 98.255208 +2023-10-02 20:45:22,804 - Epoch: [34][ 70/ 117] Loss 0.329066 Top1 83.415179 Top5 98.219866 +2023-10-02 20:45:22,955 - Epoch: [34][ 80/ 117] Loss 0.328451 Top1 83.417969 Top5 98.242188 +2023-10-02 20:45:23,104 - Epoch: [34][ 90/ 117] Loss 0.332927 Top1 83.164062 Top5 98.224826 +2023-10-02 20:45:23,255 - Epoch: [34][ 100/ 117] Loss 0.334782 Top1 83.136719 Top5 98.148438 +2023-10-02 20:45:23,413 - Epoch: [34][ 110/ 117] Loss 0.334937 Top1 83.139205 Top5 98.132102 +2023-10-02 20:45:23,502 - Epoch: [34][ 117/ 117] Loss 0.334046 Top1 83.154661 Top5 98.156497 +2023-10-02 20:45:23,641 - ==> Top1: 83.155 Top5: 98.156 Loss: 0.334 + +2023-10-02 20:45:23,642 - ==> Confusion: +[[ 906 3 5 0 13 4 1 3 2 73 2 1 0 2 6 1 5 4 0 0 19] + [ 0 1065 1 1 7 15 3 18 2 1 1 1 0 0 1 3 1 0 5 1 5] + [ 4 0 977 7 2 1 32 4 0 1 1 1 1 2 0 2 0 1 7 6 7] + [ 1 6 27 990 0 3 3 0 5 3 3 0 1 4 21 0 0 1 9 1 11] + [ 15 8 3 0 966 7 0 1 0 13 1 0 1 4 8 3 11 0 1 3 5] + [ 4 56 2 3 2 973 1 35 0 3 1 2 1 6 8 1 3 0 4 1 10] + [ 0 3 33 0 0 0 1129 4 0 0 3 2 0 1 0 2 1 1 2 6 4] + [ 3 19 23 1 2 18 8 1072 0 1 0 4 0 7 3 1 0 0 40 8 8] + [ 18 3 0 0 2 3 0 2 934 61 15 1 3 8 18 0 1 5 11 1 3] + [ 112 1 3 1 12 1 0 0 20 938 0 0 1 13 7 0 1 0 0 5 4] + [ 5 3 17 15 2 2 6 4 15 1 938 3 0 19 4 0 2 3 4 2 8] + [ 0 2 3 0 0 23 0 2 0 1 0 943 19 5 0 3 3 17 0 9 5] + [ 0 2 5 5 1 8 1 2 1 1 2 56 915 4 2 12 3 25 3 7 13] + [ 2 0 3 1 3 23 1 1 12 19 6 3 0 1016 11 0 1 1 1 2 13] + [ 13 0 5 13 5 0 0 0 16 6 4 1 3 2 1010 0 2 0 14 0 7] + [ 0 1 6 2 5 1 4 0 0 2 0 7 5 1 0 1054 25 7 1 7 6] + [ 0 13 2 0 2 6 2 0 0 0 0 4 2 0 3 5 1110 0 0 4 8] + [ 0 0 2 4 0 0 3 1 0 1 0 5 13 1 7 6 3 983 0 4 5] + [ 1 3 10 16 1 0 0 23 2 0 2 0 0 0 13 0 0 0 989 0 8] + [ 0 4 1 1 1 3 21 15 0 1 0 15 3 0 1 2 10 1 1 1060 12] + [ 115 259 218 92 107 164 58 123 100 105 148 143 267 278 168 39 140 55 162 233 4931]] + +2023-10-02 20:45:23,643 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:45:23,644 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:45:23,650 - + +2023-10-02 20:45:23,650 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:45:24,785 - Epoch: [35][ 10/ 1236] Overall Loss 0.277928 Objective Loss 0.277928 LR 0.001000 Time 0.113463 +2023-10-02 20:45:24,993 - Epoch: [35][ 20/ 1236] Overall Loss 0.255429 Objective Loss 0.255429 LR 0.001000 Time 0.067099 +2023-10-02 20:45:25,199 - Epoch: [35][ 30/ 1236] Overall Loss 0.251014 Objective Loss 0.251014 LR 0.001000 Time 0.051612 +2023-10-02 20:45:25,407 - Epoch: [35][ 40/ 1236] Overall Loss 0.249006 Objective Loss 0.249006 LR 0.001000 Time 0.043898 +2023-10-02 20:45:25,614 - Epoch: [35][ 50/ 1236] Overall Loss 0.254641 Objective Loss 0.254641 LR 0.001000 Time 0.039214 +2023-10-02 20:45:25,823 - Epoch: [35][ 60/ 1236] Overall Loss 0.258493 Objective Loss 0.258493 LR 0.001000 Time 0.036160 +2023-10-02 20:45:26,028 - Epoch: [35][ 70/ 1236] Overall Loss 0.259160 Objective Loss 0.259160 LR 0.001000 Time 0.033925 +2023-10-02 20:45:26,238 - Epoch: [35][ 80/ 1236] Overall Loss 0.260312 Objective Loss 0.260312 LR 0.001000 Time 0.032298 +2023-10-02 20:45:26,443 - Epoch: [35][ 90/ 1236] Overall Loss 0.259896 Objective Loss 0.259896 LR 0.001000 Time 0.030988 +2023-10-02 20:45:26,652 - Epoch: [35][ 100/ 1236] Overall Loss 0.260063 Objective Loss 0.260063 LR 0.001000 Time 0.029980 +2023-10-02 20:45:26,857 - Epoch: [35][ 110/ 1236] Overall Loss 0.260883 Objective Loss 0.260883 LR 0.001000 Time 0.029116 +2023-10-02 20:45:27,067 - Epoch: [35][ 120/ 1236] Overall Loss 0.259286 Objective Loss 0.259286 LR 0.001000 Time 0.028431 +2023-10-02 20:45:27,272 - Epoch: [35][ 130/ 1236] Overall Loss 0.260625 Objective Loss 0.260625 LR 0.001000 Time 0.027823 +2023-10-02 20:45:27,482 - Epoch: [35][ 140/ 1236] Overall Loss 0.259833 Objective Loss 0.259833 LR 0.001000 Time 0.027329 +2023-10-02 20:45:27,687 - Epoch: [35][ 150/ 1236] Overall Loss 0.261195 Objective Loss 0.261195 LR 0.001000 Time 0.026875 +2023-10-02 20:45:27,896 - Epoch: [35][ 160/ 1236] Overall Loss 0.261809 Objective Loss 0.261809 LR 0.001000 Time 0.026501 +2023-10-02 20:45:28,102 - Epoch: [35][ 170/ 1236] Overall Loss 0.262478 Objective Loss 0.262478 LR 0.001000 Time 0.026150 +2023-10-02 20:45:28,311 - Epoch: [35][ 180/ 1236] Overall Loss 0.263673 Objective Loss 0.263673 LR 0.001000 Time 0.025855 +2023-10-02 20:45:28,516 - Epoch: [35][ 190/ 1236] Overall Loss 0.263356 Objective Loss 0.263356 LR 0.001000 Time 0.025573 +2023-10-02 20:45:28,725 - Epoch: [35][ 200/ 1236] Overall Loss 0.263589 Objective Loss 0.263589 LR 0.001000 Time 0.025337 +2023-10-02 20:45:28,930 - Epoch: [35][ 210/ 1236] Overall Loss 0.264527 Objective Loss 0.264527 LR 0.001000 Time 0.025107 +2023-10-02 20:45:29,139 - Epoch: [35][ 220/ 1236] Overall Loss 0.263554 Objective Loss 0.263554 LR 0.001000 Time 0.024916 +2023-10-02 20:45:29,345 - Epoch: [35][ 230/ 1236] Overall Loss 0.263710 Objective Loss 0.263710 LR 0.001000 Time 0.024724 +2023-10-02 20:45:29,553 - Epoch: [35][ 240/ 1236] Overall Loss 0.264317 Objective Loss 0.264317 LR 0.001000 Time 0.024561 +2023-10-02 20:45:29,758 - Epoch: [35][ 250/ 1236] Overall Loss 0.263080 Objective Loss 0.263080 LR 0.001000 Time 0.024399 +2023-10-02 20:45:29,968 - Epoch: [35][ 260/ 1236] Overall Loss 0.263163 Objective Loss 0.263163 LR 0.001000 Time 0.024264 +2023-10-02 20:45:30,173 - Epoch: [35][ 270/ 1236] Overall Loss 0.262783 Objective Loss 0.262783 LR 0.001000 Time 0.024124 +2023-10-02 20:45:30,382 - Epoch: [35][ 280/ 1236] Overall Loss 0.262956 Objective Loss 0.262956 LR 0.001000 Time 0.024010 +2023-10-02 20:45:30,587 - Epoch: [35][ 290/ 1236] Overall Loss 0.263305 Objective Loss 0.263305 LR 0.001000 Time 0.023885 +2023-10-02 20:45:30,795 - Epoch: [35][ 300/ 1236] Overall Loss 0.262494 Objective Loss 0.262494 LR 0.001000 Time 0.023781 +2023-10-02 20:45:31,004 - Epoch: [35][ 310/ 1236] Overall Loss 0.262483 Objective Loss 0.262483 LR 0.001000 Time 0.023687 +2023-10-02 20:45:31,210 - Epoch: [35][ 320/ 1236] Overall Loss 0.261935 Objective Loss 0.261935 LR 0.001000 Time 0.023590 +2023-10-02 20:45:31,419 - Epoch: [35][ 330/ 1236] Overall Loss 0.261541 Objective Loss 0.261541 LR 0.001000 Time 0.023509 +2023-10-02 20:45:31,625 - Epoch: [35][ 340/ 1236] Overall Loss 0.261253 Objective Loss 0.261253 LR 0.001000 Time 0.023423 +2023-10-02 20:45:31,835 - Epoch: [35][ 350/ 1236] Overall Loss 0.261082 Objective Loss 0.261082 LR 0.001000 Time 0.023351 +2023-10-02 20:45:32,041 - Epoch: [35][ 360/ 1236] Overall Loss 0.260676 Objective Loss 0.260676 LR 0.001000 Time 0.023275 +2023-10-02 20:45:32,251 - Epoch: [35][ 370/ 1236] Overall Loss 0.261303 Objective Loss 0.261303 LR 0.001000 Time 0.023213 +2023-10-02 20:45:32,461 - Epoch: [35][ 380/ 1236] Overall Loss 0.260820 Objective Loss 0.260820 LR 0.001000 Time 0.023154 +2023-10-02 20:45:32,671 - Epoch: [35][ 390/ 1236] Overall Loss 0.261130 Objective Loss 0.261130 LR 0.001000 Time 0.023095 +2023-10-02 20:45:32,881 - Epoch: [35][ 400/ 1236] Overall Loss 0.260391 Objective Loss 0.260391 LR 0.001000 Time 0.023041 +2023-10-02 20:45:33,092 - Epoch: [35][ 410/ 1236] Overall Loss 0.260538 Objective Loss 0.260538 LR 0.001000 Time 0.022989 +2023-10-02 20:45:33,302 - Epoch: [35][ 420/ 1236] Overall Loss 0.260669 Objective Loss 0.260669 LR 0.001000 Time 0.022941 +2023-10-02 20:45:33,513 - Epoch: [35][ 430/ 1236] Overall Loss 0.260161 Objective Loss 0.260161 LR 0.001000 Time 0.022896 +2023-10-02 20:45:33,724 - Epoch: [35][ 440/ 1236] Overall Loss 0.260877 Objective Loss 0.260877 LR 0.001000 Time 0.022855 +2023-10-02 20:45:33,934 - Epoch: [35][ 450/ 1236] Overall Loss 0.261434 Objective Loss 0.261434 LR 0.001000 Time 0.022814 +2023-10-02 20:45:34,144 - Epoch: [35][ 460/ 1236] Overall Loss 0.262067 Objective Loss 0.262067 LR 0.001000 Time 0.022773 +2023-10-02 20:45:34,356 - Epoch: [35][ 470/ 1236] Overall Loss 0.261788 Objective Loss 0.261788 LR 0.001000 Time 0.022738 +2023-10-02 20:45:34,566 - Epoch: [35][ 480/ 1236] Overall Loss 0.261479 Objective Loss 0.261479 LR 0.001000 Time 0.022701 +2023-10-02 20:45:34,776 - Epoch: [35][ 490/ 1236] Overall Loss 0.261981 Objective Loss 0.261981 LR 0.001000 Time 0.022666 +2023-10-02 20:45:34,987 - Epoch: [35][ 500/ 1236] Overall Loss 0.262022 Objective Loss 0.262022 LR 0.001000 Time 0.022632 +2023-10-02 20:45:35,197 - Epoch: [35][ 510/ 1236] Overall Loss 0.262984 Objective Loss 0.262984 LR 0.001000 Time 0.022601 +2023-10-02 20:45:35,407 - Epoch: [35][ 520/ 1236] Overall Loss 0.263116 Objective Loss 0.263116 LR 0.001000 Time 0.022569 +2023-10-02 20:45:35,618 - Epoch: [35][ 530/ 1236] Overall Loss 0.263082 Objective Loss 0.263082 LR 0.001000 Time 0.022541 +2023-10-02 20:45:35,828 - Epoch: [35][ 540/ 1236] Overall Loss 0.262711 Objective Loss 0.262711 LR 0.001000 Time 0.022512 +2023-10-02 20:45:36,040 - Epoch: [35][ 550/ 1236] Overall Loss 0.262587 Objective Loss 0.262587 LR 0.001000 Time 0.022487 +2023-10-02 20:45:36,250 - Epoch: [35][ 560/ 1236] Overall Loss 0.262668 Objective Loss 0.262668 LR 0.001000 Time 0.022459 +2023-10-02 20:45:36,461 - Epoch: [35][ 570/ 1236] Overall Loss 0.262595 Objective Loss 0.262595 LR 0.001000 Time 0.022433 +2023-10-02 20:45:36,671 - Epoch: [35][ 580/ 1236] Overall Loss 0.263023 Objective Loss 0.263023 LR 0.001000 Time 0.022409 +2023-10-02 20:45:36,881 - Epoch: [35][ 590/ 1236] Overall Loss 0.263140 Objective Loss 0.263140 LR 0.001000 Time 0.022385 +2023-10-02 20:45:37,092 - Epoch: [35][ 600/ 1236] Overall Loss 0.263313 Objective Loss 0.263313 LR 0.001000 Time 0.022361 +2023-10-02 20:45:37,302 - Epoch: [35][ 610/ 1236] Overall Loss 0.262937 Objective Loss 0.262937 LR 0.001000 Time 0.022339 +2023-10-02 20:45:37,512 - Epoch: [35][ 620/ 1236] Overall Loss 0.263073 Objective Loss 0.263073 LR 0.001000 Time 0.022317 +2023-10-02 20:45:37,723 - Epoch: [35][ 630/ 1236] Overall Loss 0.263215 Objective Loss 0.263215 LR 0.001000 Time 0.022297 +2023-10-02 20:45:37,937 - Epoch: [35][ 640/ 1236] Overall Loss 0.263243 Objective Loss 0.263243 LR 0.001000 Time 0.022283 +2023-10-02 20:45:38,159 - Epoch: [35][ 650/ 1236] Overall Loss 0.263064 Objective Loss 0.263064 LR 0.001000 Time 0.022278 +2023-10-02 20:45:38,375 - Epoch: [35][ 660/ 1236] Overall Loss 0.263233 Objective Loss 0.263233 LR 0.001000 Time 0.022268 +2023-10-02 20:45:38,597 - Epoch: [35][ 670/ 1236] Overall Loss 0.263582 Objective Loss 0.263582 LR 0.001000 Time 0.022265 +2023-10-02 20:45:38,813 - Epoch: [35][ 680/ 1236] Overall Loss 0.263924 Objective Loss 0.263924 LR 0.001000 Time 0.022255 +2023-10-02 20:45:39,034 - Epoch: [35][ 690/ 1236] Overall Loss 0.264352 Objective Loss 0.264352 LR 0.001000 Time 0.022252 +2023-10-02 20:45:39,250 - Epoch: [35][ 700/ 1236] Overall Loss 0.264488 Objective Loss 0.264488 LR 0.001000 Time 0.022242 +2023-10-02 20:45:39,472 - Epoch: [35][ 710/ 1236] Overall Loss 0.264619 Objective Loss 0.264619 LR 0.001000 Time 0.022241 +2023-10-02 20:45:39,685 - Epoch: [35][ 720/ 1236] Overall Loss 0.265092 Objective Loss 0.265092 LR 0.001000 Time 0.022227 +2023-10-02 20:45:39,907 - Epoch: [35][ 730/ 1236] Overall Loss 0.265392 Objective Loss 0.265392 LR 0.001000 Time 0.022226 +2023-10-02 20:45:40,123 - Epoch: [35][ 740/ 1236] Overall Loss 0.265338 Objective Loss 0.265338 LR 0.001000 Time 0.022217 +2023-10-02 20:45:40,345 - Epoch: [35][ 750/ 1236] Overall Loss 0.265396 Objective Loss 0.265396 LR 0.001000 Time 0.022216 +2023-10-02 20:45:40,561 - Epoch: [35][ 760/ 1236] Overall Loss 0.265432 Objective Loss 0.265432 LR 0.001000 Time 0.022207 +2023-10-02 20:45:40,782 - Epoch: [35][ 770/ 1236] Overall Loss 0.265703 Objective Loss 0.265703 LR 0.001000 Time 0.022206 +2023-10-02 20:45:40,999 - Epoch: [35][ 780/ 1236] Overall Loss 0.265827 Objective Loss 0.265827 LR 0.001000 Time 0.022198 +2023-10-02 20:45:41,221 - Epoch: [35][ 790/ 1236] Overall Loss 0.265568 Objective Loss 0.265568 LR 0.001000 Time 0.022197 +2023-10-02 20:45:41,437 - Epoch: [35][ 800/ 1236] Overall Loss 0.265755 Objective Loss 0.265755 LR 0.001000 Time 0.022189 +2023-10-02 20:45:41,658 - Epoch: [35][ 810/ 1236] Overall Loss 0.265993 Objective Loss 0.265993 LR 0.001000 Time 0.022188 +2023-10-02 20:45:41,874 - Epoch: [35][ 820/ 1236] Overall Loss 0.266023 Objective Loss 0.266023 LR 0.001000 Time 0.022180 +2023-10-02 20:45:42,096 - Epoch: [35][ 830/ 1236] Overall Loss 0.266241 Objective Loss 0.266241 LR 0.001000 Time 0.022179 +2023-10-02 20:45:42,312 - Epoch: [35][ 840/ 1236] Overall Loss 0.265977 Objective Loss 0.265977 LR 0.001000 Time 0.022172 +2023-10-02 20:45:42,534 - Epoch: [35][ 850/ 1236] Overall Loss 0.265993 Objective Loss 0.265993 LR 0.001000 Time 0.022172 +2023-10-02 20:45:42,751 - Epoch: [35][ 860/ 1236] Overall Loss 0.265957 Objective Loss 0.265957 LR 0.001000 Time 0.022165 +2023-10-02 20:45:42,972 - Epoch: [35][ 870/ 1236] Overall Loss 0.266199 Objective Loss 0.266199 LR 0.001000 Time 0.022165 +2023-10-02 20:45:43,188 - Epoch: [35][ 880/ 1236] Overall Loss 0.266449 Objective Loss 0.266449 LR 0.001000 Time 0.022158 +2023-10-02 20:45:43,410 - Epoch: [35][ 890/ 1236] Overall Loss 0.266345 Objective Loss 0.266345 LR 0.001000 Time 0.022157 +2023-10-02 20:45:43,626 - Epoch: [35][ 900/ 1236] Overall Loss 0.266469 Objective Loss 0.266469 LR 0.001000 Time 0.022150 +2023-10-02 20:45:43,844 - Epoch: [35][ 910/ 1236] Overall Loss 0.266394 Objective Loss 0.266394 LR 0.001000 Time 0.022146 +2023-10-02 20:45:44,054 - Epoch: [35][ 920/ 1236] Overall Loss 0.266683 Objective Loss 0.266683 LR 0.001000 Time 0.022133 +2023-10-02 20:45:44,265 - Epoch: [35][ 930/ 1236] Overall Loss 0.266500 Objective Loss 0.266500 LR 0.001000 Time 0.022122 +2023-10-02 20:45:44,475 - Epoch: [35][ 940/ 1236] Overall Loss 0.266827 Objective Loss 0.266827 LR 0.001000 Time 0.022109 +2023-10-02 20:45:44,686 - Epoch: [35][ 950/ 1236] Overall Loss 0.266610 Objective Loss 0.266610 LR 0.001000 Time 0.022098 +2023-10-02 20:45:44,896 - Epoch: [35][ 960/ 1236] Overall Loss 0.266740 Objective Loss 0.266740 LR 0.001000 Time 0.022086 +2023-10-02 20:45:45,113 - Epoch: [35][ 970/ 1236] Overall Loss 0.266868 Objective Loss 0.266868 LR 0.001000 Time 0.022082 +2023-10-02 20:45:45,330 - Epoch: [35][ 980/ 1236] Overall Loss 0.266675 Objective Loss 0.266675 LR 0.001000 Time 0.022077 +2023-10-02 20:45:45,552 - Epoch: [35][ 990/ 1236] Overall Loss 0.266639 Objective Loss 0.266639 LR 0.001000 Time 0.022078 +2023-10-02 20:45:45,769 - Epoch: [35][ 1000/ 1236] Overall Loss 0.266323 Objective Loss 0.266323 LR 0.001000 Time 0.022074 +2023-10-02 20:45:45,992 - Epoch: [35][ 1010/ 1236] Overall Loss 0.266340 Objective Loss 0.266340 LR 0.001000 Time 0.022075 +2023-10-02 20:45:46,208 - Epoch: [35][ 1020/ 1236] Overall Loss 0.266373 Objective Loss 0.266373 LR 0.001000 Time 0.022071 +2023-10-02 20:45:46,431 - Epoch: [35][ 1030/ 1236] Overall Loss 0.266315 Objective Loss 0.266315 LR 0.001000 Time 0.022072 +2023-10-02 20:45:46,647 - Epoch: [35][ 1040/ 1236] Overall Loss 0.266279 Objective Loss 0.266279 LR 0.001000 Time 0.022068 +2023-10-02 20:45:46,869 - Epoch: [35][ 1050/ 1236] Overall Loss 0.266241 Objective Loss 0.266241 LR 0.001000 Time 0.022068 +2023-10-02 20:45:47,086 - Epoch: [35][ 1060/ 1236] Overall Loss 0.266424 Objective Loss 0.266424 LR 0.001000 Time 0.022064 +2023-10-02 20:45:47,309 - Epoch: [35][ 1070/ 1236] Overall Loss 0.266573 Objective Loss 0.266573 LR 0.001000 Time 0.022065 +2023-10-02 20:45:47,516 - Epoch: [35][ 1080/ 1236] Overall Loss 0.266753 Objective Loss 0.266753 LR 0.001000 Time 0.022053 +2023-10-02 20:45:47,727 - Epoch: [35][ 1090/ 1236] Overall Loss 0.267099 Objective Loss 0.267099 LR 0.001000 Time 0.022044 +2023-10-02 20:45:47,933 - Epoch: [35][ 1100/ 1236] Overall Loss 0.267357 Objective Loss 0.267357 LR 0.001000 Time 0.022031 +2023-10-02 20:45:48,144 - Epoch: [35][ 1110/ 1236] Overall Loss 0.267506 Objective Loss 0.267506 LR 0.001000 Time 0.022022 +2023-10-02 20:45:48,350 - Epoch: [35][ 1120/ 1236] Overall Loss 0.267655 Objective Loss 0.267655 LR 0.001000 Time 0.022009 +2023-10-02 20:45:48,561 - Epoch: [35][ 1130/ 1236] Overall Loss 0.267948 Objective Loss 0.267948 LR 0.001000 Time 0.022000 +2023-10-02 20:45:48,767 - Epoch: [35][ 1140/ 1236] Overall Loss 0.268166 Objective Loss 0.268166 LR 0.001000 Time 0.021988 +2023-10-02 20:45:48,978 - Epoch: [35][ 1150/ 1236] Overall Loss 0.268316 Objective Loss 0.268316 LR 0.001000 Time 0.021980 +2023-10-02 20:45:49,184 - Epoch: [35][ 1160/ 1236] Overall Loss 0.268363 Objective Loss 0.268363 LR 0.001000 Time 0.021968 +2023-10-02 20:45:49,395 - Epoch: [35][ 1170/ 1236] Overall Loss 0.268287 Objective Loss 0.268287 LR 0.001000 Time 0.021960 +2023-10-02 20:45:49,602 - Epoch: [35][ 1180/ 1236] Overall Loss 0.268418 Objective Loss 0.268418 LR 0.001000 Time 0.021949 +2023-10-02 20:45:49,813 - Epoch: [35][ 1190/ 1236] Overall Loss 0.268504 Objective Loss 0.268504 LR 0.001000 Time 0.021942 +2023-10-02 20:45:50,020 - Epoch: [35][ 1200/ 1236] Overall Loss 0.268648 Objective Loss 0.268648 LR 0.001000 Time 0.021931 +2023-10-02 20:45:50,231 - Epoch: [35][ 1210/ 1236] Overall Loss 0.268708 Objective Loss 0.268708 LR 0.001000 Time 0.021924 +2023-10-02 20:45:50,437 - Epoch: [35][ 1220/ 1236] Overall Loss 0.268667 Objective Loss 0.268667 LR 0.001000 Time 0.021913 +2023-10-02 20:45:50,699 - Epoch: [35][ 1230/ 1236] Overall Loss 0.268543 Objective Loss 0.268543 LR 0.001000 Time 0.021947 +2023-10-02 20:45:50,821 - Epoch: [35][ 1236/ 1236] Overall Loss 0.268400 Objective Loss 0.268400 Top1 89.409369 Top5 98.574338 LR 0.001000 Time 0.021939 +2023-10-02 20:45:50,951 - --- validate (epoch=35)----------- +2023-10-02 20:45:50,952 - 29943 samples (256 per mini-batch) +2023-10-02 20:45:51,425 - Epoch: [35][ 10/ 117] Loss 0.345326 Top1 83.593750 Top5 98.242188 +2023-10-02 20:45:51,586 - Epoch: [35][ 20/ 117] Loss 0.333778 Top1 84.160156 Top5 98.203125 +2023-10-02 20:45:51,745 - Epoch: [35][ 30/ 117] Loss 0.324490 Top1 84.270833 Top5 98.385417 +2023-10-02 20:45:51,905 - Epoch: [35][ 40/ 117] Loss 0.331708 Top1 84.189453 Top5 98.320312 +2023-10-02 20:45:52,063 - Epoch: [35][ 50/ 117] Loss 0.330983 Top1 84.335938 Top5 98.445312 +2023-10-02 20:45:52,216 - Epoch: [35][ 60/ 117] Loss 0.333367 Top1 84.121094 Top5 98.398438 +2023-10-02 20:45:52,363 - Epoch: [35][ 70/ 117] Loss 0.331580 Top1 84.246652 Top5 98.437500 +2023-10-02 20:45:52,509 - Epoch: [35][ 80/ 117] Loss 0.335679 Top1 84.199219 Top5 98.413086 +2023-10-02 20:45:52,654 - Epoch: [35][ 90/ 117] Loss 0.336260 Top1 84.062500 Top5 98.372396 +2023-10-02 20:45:52,799 - Epoch: [35][ 100/ 117] Loss 0.340001 Top1 83.917969 Top5 98.347656 +2023-10-02 20:45:52,953 - Epoch: [35][ 110/ 117] Loss 0.342190 Top1 83.941761 Top5 98.345170 +2023-10-02 20:45:53,042 - Epoch: [35][ 117/ 117] Loss 0.339590 Top1 83.982901 Top5 98.316802 +2023-10-02 20:45:53,153 - ==> Top1: 83.983 Top5: 98.317 Loss: 0.340 + +2023-10-02 20:45:53,154 - ==> Confusion: +[[ 905 0 3 0 8 1 0 1 6 88 1 0 2 3 5 2 5 4 1 0 15] + [ 1 1038 0 1 14 25 1 21 5 1 3 0 0 0 0 3 1 2 9 2 4] + [ 6 0 936 14 2 0 25 14 0 5 6 1 5 1 3 6 0 2 14 7 9] + [ 4 3 13 975 0 0 1 1 2 3 4 1 7 3 38 2 0 6 9 0 17] + [ 27 5 1 0 958 2 0 2 2 21 2 0 2 3 10 2 8 1 0 3 1] + [ 2 39 2 1 1 941 1 23 3 14 8 7 8 19 8 0 2 0 5 3 29] + [ 0 5 26 0 0 2 1118 6 0 0 6 2 1 0 1 4 1 0 3 7 9] + [ 6 17 16 3 6 18 2 1062 0 3 8 5 3 6 2 1 1 1 41 5 12] + [ 16 0 0 0 2 0 0 0 971 46 8 1 5 7 25 3 2 1 1 0 1] + [ 99 1 0 0 7 0 0 0 35 950 0 0 0 9 7 0 1 0 0 4 6] + [ 3 0 11 6 2 0 1 3 16 3 961 2 1 13 7 0 1 3 9 0 11] + [ 0 0 2 0 2 7 0 4 0 3 0 925 43 11 1 6 0 19 0 4 8] + [ 0 0 2 2 0 1 4 0 1 2 4 38 929 4 5 7 4 41 0 9 15] + [ 1 0 1 0 1 2 0 2 18 27 5 2 0 1033 10 0 0 2 1 0 14] + [ 10 1 1 11 1 0 0 0 15 5 2 0 3 2 1030 0 2 8 3 0 7] + [ 0 0 1 0 4 0 4 0 0 1 0 9 7 0 0 1064 17 11 2 8 6] + [ 0 10 3 0 7 7 1 2 1 0 0 7 1 4 3 14 1080 2 1 3 15] + [ 1 0 0 0 0 0 1 0 1 1 1 0 14 0 0 8 0 1006 0 2 3] + [ 1 3 9 12 1 0 0 20 5 1 3 0 1 1 17 0 0 2 975 0 17] + [ 0 0 5 2 3 4 15 11 0 1 1 8 1 4 3 3 7 1 3 1071 9] + [ 160 157 96 61 132 101 40 94 133 108 157 120 292 279 165 67 99 81 160 184 5219]] + +2023-10-02 20:45:53,155 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:45:53,155 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:45:53,161 - + +2023-10-02 20:45:53,161 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:45:54,190 - Epoch: [36][ 10/ 1236] Overall Loss 0.279520 Objective Loss 0.279520 LR 0.001000 Time 0.102880 +2023-10-02 20:45:54,399 - Epoch: [36][ 20/ 1236] Overall Loss 0.267815 Objective Loss 0.267815 LR 0.001000 Time 0.061878 +2023-10-02 20:45:54,608 - Epoch: [36][ 30/ 1236] Overall Loss 0.266561 Objective Loss 0.266561 LR 0.001000 Time 0.048181 +2023-10-02 20:45:54,817 - Epoch: [36][ 40/ 1236] Overall Loss 0.272089 Objective Loss 0.272089 LR 0.001000 Time 0.041365 +2023-10-02 20:45:55,025 - Epoch: [36][ 50/ 1236] Overall Loss 0.271896 Objective Loss 0.271896 LR 0.001000 Time 0.037252 +2023-10-02 20:45:55,236 - Epoch: [36][ 60/ 1236] Overall Loss 0.272779 Objective Loss 0.272779 LR 0.001000 Time 0.034548 +2023-10-02 20:45:55,443 - Epoch: [36][ 70/ 1236] Overall Loss 0.269909 Objective Loss 0.269909 LR 0.001000 Time 0.032565 +2023-10-02 20:45:55,654 - Epoch: [36][ 80/ 1236] Overall Loss 0.268984 Objective Loss 0.268984 LR 0.001000 Time 0.031126 +2023-10-02 20:45:55,861 - Epoch: [36][ 90/ 1236] Overall Loss 0.268003 Objective Loss 0.268003 LR 0.001000 Time 0.029963 +2023-10-02 20:45:56,072 - Epoch: [36][ 100/ 1236] Overall Loss 0.268553 Objective Loss 0.268553 LR 0.001000 Time 0.029076 +2023-10-02 20:45:56,279 - Epoch: [36][ 110/ 1236] Overall Loss 0.266037 Objective Loss 0.266037 LR 0.001000 Time 0.028311 +2023-10-02 20:45:56,490 - Epoch: [36][ 120/ 1236] Overall Loss 0.263738 Objective Loss 0.263738 LR 0.001000 Time 0.027706 +2023-10-02 20:45:56,697 - Epoch: [36][ 130/ 1236] Overall Loss 0.265656 Objective Loss 0.265656 LR 0.001000 Time 0.027166 +2023-10-02 20:45:56,910 - Epoch: [36][ 140/ 1236] Overall Loss 0.264147 Objective Loss 0.264147 LR 0.001000 Time 0.026743 +2023-10-02 20:45:57,121 - Epoch: [36][ 150/ 1236] Overall Loss 0.265308 Objective Loss 0.265308 LR 0.001000 Time 0.026363 +2023-10-02 20:45:57,332 - Epoch: [36][ 160/ 1236] Overall Loss 0.264726 Objective Loss 0.264726 LR 0.001000 Time 0.026036 +2023-10-02 20:45:57,543 - Epoch: [36][ 170/ 1236] Overall Loss 0.263877 Objective Loss 0.263877 LR 0.001000 Time 0.025742 +2023-10-02 20:45:57,755 - Epoch: [36][ 180/ 1236] Overall Loss 0.264143 Objective Loss 0.264143 LR 0.001000 Time 0.025486 +2023-10-02 20:45:57,966 - Epoch: [36][ 190/ 1236] Overall Loss 0.263002 Objective Loss 0.263002 LR 0.001000 Time 0.025253 +2023-10-02 20:45:58,178 - Epoch: [36][ 200/ 1236] Overall Loss 0.262780 Objective Loss 0.262780 LR 0.001000 Time 0.025050 +2023-10-02 20:45:58,389 - Epoch: [36][ 210/ 1236] Overall Loss 0.262058 Objective Loss 0.262058 LR 0.001000 Time 0.024859 +2023-10-02 20:45:58,601 - Epoch: [36][ 220/ 1236] Overall Loss 0.261813 Objective Loss 0.261813 LR 0.001000 Time 0.024690 +2023-10-02 20:45:58,812 - Epoch: [36][ 230/ 1236] Overall Loss 0.261747 Objective Loss 0.261747 LR 0.001000 Time 0.024533 +2023-10-02 20:45:59,025 - Epoch: [36][ 240/ 1236] Overall Loss 0.261108 Objective Loss 0.261108 LR 0.001000 Time 0.024394 +2023-10-02 20:45:59,235 - Epoch: [36][ 250/ 1236] Overall Loss 0.259640 Objective Loss 0.259640 LR 0.001000 Time 0.024256 +2023-10-02 20:45:59,448 - Epoch: [36][ 260/ 1236] Overall Loss 0.259300 Objective Loss 0.259300 LR 0.001000 Time 0.024137 +2023-10-02 20:45:59,658 - Epoch: [36][ 270/ 1236] Overall Loss 0.258849 Objective Loss 0.258849 LR 0.001000 Time 0.024019 +2023-10-02 20:45:59,870 - Epoch: [36][ 280/ 1236] Overall Loss 0.259076 Objective Loss 0.259076 LR 0.001000 Time 0.023915 +2023-10-02 20:46:00,082 - Epoch: [36][ 290/ 1236] Overall Loss 0.259487 Objective Loss 0.259487 LR 0.001000 Time 0.023817 +2023-10-02 20:46:00,294 - Epoch: [36][ 300/ 1236] Overall Loss 0.258110 Objective Loss 0.258110 LR 0.001000 Time 0.023729 +2023-10-02 20:46:00,504 - Epoch: [36][ 310/ 1236] Overall Loss 0.257982 Objective Loss 0.257982 LR 0.001000 Time 0.023639 +2023-10-02 20:46:00,717 - Epoch: [36][ 320/ 1236] Overall Loss 0.257530 Objective Loss 0.257530 LR 0.001000 Time 0.023562 +2023-10-02 20:46:00,928 - Epoch: [36][ 330/ 1236] Overall Loss 0.257741 Objective Loss 0.257741 LR 0.001000 Time 0.023487 +2023-10-02 20:46:01,141 - Epoch: [36][ 340/ 1236] Overall Loss 0.256911 Objective Loss 0.256911 LR 0.001000 Time 0.023420 +2023-10-02 20:46:01,348 - Epoch: [36][ 350/ 1236] Overall Loss 0.257109 Objective Loss 0.257109 LR 0.001000 Time 0.023341 +2023-10-02 20:46:01,559 - Epoch: [36][ 360/ 1236] Overall Loss 0.256423 Objective Loss 0.256423 LR 0.001000 Time 0.023278 +2023-10-02 20:46:01,766 - Epoch: [36][ 370/ 1236] Overall Loss 0.256605 Objective Loss 0.256605 LR 0.001000 Time 0.023209 +2023-10-02 20:46:01,977 - Epoch: [36][ 380/ 1236] Overall Loss 0.256124 Objective Loss 0.256124 LR 0.001000 Time 0.023153 +2023-10-02 20:46:02,185 - Epoch: [36][ 390/ 1236] Overall Loss 0.256436 Objective Loss 0.256436 LR 0.001000 Time 0.023090 +2023-10-02 20:46:02,396 - Epoch: [36][ 400/ 1236] Overall Loss 0.255706 Objective Loss 0.255706 LR 0.001000 Time 0.023040 +2023-10-02 20:46:02,603 - Epoch: [36][ 410/ 1236] Overall Loss 0.256021 Objective Loss 0.256021 LR 0.001000 Time 0.022983 +2023-10-02 20:46:02,814 - Epoch: [36][ 420/ 1236] Overall Loss 0.256044 Objective Loss 0.256044 LR 0.001000 Time 0.022938 +2023-10-02 20:46:03,022 - Epoch: [36][ 430/ 1236] Overall Loss 0.256043 Objective Loss 0.256043 LR 0.001000 Time 0.022886 +2023-10-02 20:46:03,233 - Epoch: [36][ 440/ 1236] Overall Loss 0.255995 Objective Loss 0.255995 LR 0.001000 Time 0.022845 +2023-10-02 20:46:03,440 - Epoch: [36][ 450/ 1236] Overall Loss 0.256854 Objective Loss 0.256854 LR 0.001000 Time 0.022797 +2023-10-02 20:46:03,651 - Epoch: [36][ 460/ 1236] Overall Loss 0.256879 Objective Loss 0.256879 LR 0.001000 Time 0.022759 +2023-10-02 20:46:03,858 - Epoch: [36][ 470/ 1236] Overall Loss 0.257455 Objective Loss 0.257455 LR 0.001000 Time 0.022715 +2023-10-02 20:46:04,069 - Epoch: [36][ 480/ 1236] Overall Loss 0.257324 Objective Loss 0.257324 LR 0.001000 Time 0.022682 +2023-10-02 20:46:04,277 - Epoch: [36][ 490/ 1236] Overall Loss 0.256785 Objective Loss 0.256785 LR 0.001000 Time 0.022641 +2023-10-02 20:46:04,488 - Epoch: [36][ 500/ 1236] Overall Loss 0.257032 Objective Loss 0.257032 LR 0.001000 Time 0.022610 +2023-10-02 20:46:04,695 - Epoch: [36][ 510/ 1236] Overall Loss 0.257010 Objective Loss 0.257010 LR 0.001000 Time 0.022573 +2023-10-02 20:46:04,907 - Epoch: [36][ 520/ 1236] Overall Loss 0.257217 Objective Loss 0.257217 LR 0.001000 Time 0.022544 +2023-10-02 20:46:05,114 - Epoch: [36][ 530/ 1236] Overall Loss 0.256756 Objective Loss 0.256756 LR 0.001000 Time 0.022510 +2023-10-02 20:46:05,325 - Epoch: [36][ 540/ 1236] Overall Loss 0.256947 Objective Loss 0.256947 LR 0.001000 Time 0.022483 +2023-10-02 20:46:05,533 - Epoch: [36][ 550/ 1236] Overall Loss 0.256954 Objective Loss 0.256954 LR 0.001000 Time 0.022451 +2023-10-02 20:46:05,744 - Epoch: [36][ 560/ 1236] Overall Loss 0.256563 Objective Loss 0.256563 LR 0.001000 Time 0.022427 +2023-10-02 20:46:05,951 - Epoch: [36][ 570/ 1236] Overall Loss 0.256576 Objective Loss 0.256576 LR 0.001000 Time 0.022397 +2023-10-02 20:46:06,163 - Epoch: [36][ 580/ 1236] Overall Loss 0.256458 Objective Loss 0.256458 LR 0.001000 Time 0.022375 +2023-10-02 20:46:06,370 - Epoch: [36][ 590/ 1236] Overall Loss 0.257092 Objective Loss 0.257092 LR 0.001000 Time 0.022347 +2023-10-02 20:46:06,582 - Epoch: [36][ 600/ 1236] Overall Loss 0.257035 Objective Loss 0.257035 LR 0.001000 Time 0.022326 +2023-10-02 20:46:06,789 - Epoch: [36][ 610/ 1236] Overall Loss 0.256719 Objective Loss 0.256719 LR 0.001000 Time 0.022300 +2023-10-02 20:46:07,000 - Epoch: [36][ 620/ 1236] Overall Loss 0.256630 Objective Loss 0.256630 LR 0.001000 Time 0.022280 +2023-10-02 20:46:07,207 - Epoch: [36][ 630/ 1236] Overall Loss 0.257162 Objective Loss 0.257162 LR 0.001000 Time 0.022255 +2023-10-02 20:46:07,420 - Epoch: [36][ 640/ 1236] Overall Loss 0.256646 Objective Loss 0.256646 LR 0.001000 Time 0.022239 +2023-10-02 20:46:07,629 - Epoch: [36][ 650/ 1236] Overall Loss 0.256890 Objective Loss 0.256890 LR 0.001000 Time 0.022218 +2023-10-02 20:46:07,841 - Epoch: [36][ 660/ 1236] Overall Loss 0.257450 Objective Loss 0.257450 LR 0.001000 Time 0.022202 +2023-10-02 20:46:08,051 - Epoch: [36][ 670/ 1236] Overall Loss 0.257981 Objective Loss 0.257981 LR 0.001000 Time 0.022184 +2023-10-02 20:46:08,265 - Epoch: [36][ 680/ 1236] Overall Loss 0.258285 Objective Loss 0.258285 LR 0.001000 Time 0.022172 +2023-10-02 20:46:08,475 - Epoch: [36][ 690/ 1236] Overall Loss 0.258286 Objective Loss 0.258286 LR 0.001000 Time 0.022154 +2023-10-02 20:46:08,690 - Epoch: [36][ 700/ 1236] Overall Loss 0.258384 Objective Loss 0.258384 LR 0.001000 Time 0.022143 +2023-10-02 20:46:08,906 - Epoch: [36][ 710/ 1236] Overall Loss 0.258398 Objective Loss 0.258398 LR 0.001000 Time 0.022134 +2023-10-02 20:46:09,120 - Epoch: [36][ 720/ 1236] Overall Loss 0.258318 Objective Loss 0.258318 LR 0.001000 Time 0.022123 +2023-10-02 20:46:09,335 - Epoch: [36][ 730/ 1236] Overall Loss 0.258562 Objective Loss 0.258562 LR 0.001000 Time 0.022115 +2023-10-02 20:46:09,550 - Epoch: [36][ 740/ 1236] Overall Loss 0.258791 Objective Loss 0.258791 LR 0.001000 Time 0.022105 +2023-10-02 20:46:09,763 - Epoch: [36][ 750/ 1236] Overall Loss 0.259092 Objective Loss 0.259092 LR 0.001000 Time 0.022095 +2023-10-02 20:46:09,979 - Epoch: [36][ 760/ 1236] Overall Loss 0.259035 Objective Loss 0.259035 LR 0.001000 Time 0.022087 +2023-10-02 20:46:10,194 - Epoch: [36][ 770/ 1236] Overall Loss 0.258975 Objective Loss 0.258975 LR 0.001000 Time 0.022080 +2023-10-02 20:46:10,409 - Epoch: [36][ 780/ 1236] Overall Loss 0.259143 Objective Loss 0.259143 LR 0.001000 Time 0.022071 +2023-10-02 20:46:10,623 - Epoch: [36][ 790/ 1236] Overall Loss 0.259170 Objective Loss 0.259170 LR 0.001000 Time 0.022063 +2023-10-02 20:46:10,839 - Epoch: [36][ 800/ 1236] Overall Loss 0.259404 Objective Loss 0.259404 LR 0.001000 Time 0.022054 +2023-10-02 20:46:11,053 - Epoch: [36][ 810/ 1236] Overall Loss 0.259753 Objective Loss 0.259753 LR 0.001000 Time 0.022046 +2023-10-02 20:46:11,265 - Epoch: [36][ 820/ 1236] Overall Loss 0.259633 Objective Loss 0.259633 LR 0.001000 Time 0.022036 +2023-10-02 20:46:11,478 - Epoch: [36][ 830/ 1236] Overall Loss 0.259670 Objective Loss 0.259670 LR 0.001000 Time 0.022026 +2023-10-02 20:46:11,690 - Epoch: [36][ 840/ 1236] Overall Loss 0.259579 Objective Loss 0.259579 LR 0.001000 Time 0.022016 +2023-10-02 20:46:11,903 - Epoch: [36][ 850/ 1236] Overall Loss 0.259546 Objective Loss 0.259546 LR 0.001000 Time 0.022005 +2023-10-02 20:46:12,115 - Epoch: [36][ 860/ 1236] Overall Loss 0.259638 Objective Loss 0.259638 LR 0.001000 Time 0.021996 +2023-10-02 20:46:12,328 - Epoch: [36][ 870/ 1236] Overall Loss 0.259883 Objective Loss 0.259883 LR 0.001000 Time 0.021987 +2023-10-02 20:46:12,540 - Epoch: [36][ 880/ 1236] Overall Loss 0.259820 Objective Loss 0.259820 LR 0.001000 Time 0.021979 +2023-10-02 20:46:12,753 - Epoch: [36][ 890/ 1236] Overall Loss 0.259834 Objective Loss 0.259834 LR 0.001000 Time 0.021969 +2023-10-02 20:46:12,965 - Epoch: [36][ 900/ 1236] Overall Loss 0.259651 Objective Loss 0.259651 LR 0.001000 Time 0.021960 +2023-10-02 20:46:13,178 - Epoch: [36][ 910/ 1236] Overall Loss 0.259918 Objective Loss 0.259918 LR 0.001000 Time 0.021952 +2023-10-02 20:46:13,390 - Epoch: [36][ 920/ 1236] Overall Loss 0.260148 Objective Loss 0.260148 LR 0.001000 Time 0.021944 +2023-10-02 20:46:13,603 - Epoch: [36][ 930/ 1236] Overall Loss 0.260501 Objective Loss 0.260501 LR 0.001000 Time 0.021936 +2023-10-02 20:46:13,815 - Epoch: [36][ 940/ 1236] Overall Loss 0.260780 Objective Loss 0.260780 LR 0.001000 Time 0.021929 +2023-10-02 20:46:14,028 - Epoch: [36][ 950/ 1236] Overall Loss 0.260880 Objective Loss 0.260880 LR 0.001000 Time 0.021921 +2023-10-02 20:46:14,240 - Epoch: [36][ 960/ 1236] Overall Loss 0.260757 Objective Loss 0.260757 LR 0.001000 Time 0.021914 +2023-10-02 20:46:14,453 - Epoch: [36][ 970/ 1236] Overall Loss 0.260814 Objective Loss 0.260814 LR 0.001000 Time 0.021907 +2023-10-02 20:46:14,665 - Epoch: [36][ 980/ 1236] Overall Loss 0.260970 Objective Loss 0.260970 LR 0.001000 Time 0.021900 +2023-10-02 20:46:14,878 - Epoch: [36][ 990/ 1236] Overall Loss 0.260830 Objective Loss 0.260830 LR 0.001000 Time 0.021892 +2023-10-02 20:46:15,090 - Epoch: [36][ 1000/ 1236] Overall Loss 0.261216 Objective Loss 0.261216 LR 0.001000 Time 0.021885 +2023-10-02 20:46:15,303 - Epoch: [36][ 1010/ 1236] Overall Loss 0.261181 Objective Loss 0.261181 LR 0.001000 Time 0.021878 +2023-10-02 20:46:15,515 - Epoch: [36][ 1020/ 1236] Overall Loss 0.261244 Objective Loss 0.261244 LR 0.001000 Time 0.021872 +2023-10-02 20:46:15,728 - Epoch: [36][ 1030/ 1236] Overall Loss 0.261461 Objective Loss 0.261461 LR 0.001000 Time 0.021866 +2023-10-02 20:46:15,940 - Epoch: [36][ 1040/ 1236] Overall Loss 0.261148 Objective Loss 0.261148 LR 0.001000 Time 0.021859 +2023-10-02 20:46:16,153 - Epoch: [36][ 1050/ 1236] Overall Loss 0.261188 Objective Loss 0.261188 LR 0.001000 Time 0.021852 +2023-10-02 20:46:16,365 - Epoch: [36][ 1060/ 1236] Overall Loss 0.261238 Objective Loss 0.261238 LR 0.001000 Time 0.021846 +2023-10-02 20:46:16,578 - Epoch: [36][ 1070/ 1236] Overall Loss 0.261056 Objective Loss 0.261056 LR 0.001000 Time 0.021840 +2023-10-02 20:46:16,790 - Epoch: [36][ 1080/ 1236] Overall Loss 0.260997 Objective Loss 0.260997 LR 0.001000 Time 0.021835 +2023-10-02 20:46:17,003 - Epoch: [36][ 1090/ 1236] Overall Loss 0.261218 Objective Loss 0.261218 LR 0.001000 Time 0.021829 +2023-10-02 20:46:17,215 - Epoch: [36][ 1100/ 1236] Overall Loss 0.261228 Objective Loss 0.261228 LR 0.001000 Time 0.021824 +2023-10-02 20:46:17,426 - Epoch: [36][ 1110/ 1236] Overall Loss 0.261600 Objective Loss 0.261600 LR 0.001000 Time 0.021816 +2023-10-02 20:46:17,638 - Epoch: [36][ 1120/ 1236] Overall Loss 0.261627 Objective Loss 0.261627 LR 0.001000 Time 0.021810 +2023-10-02 20:46:17,849 - Epoch: [36][ 1130/ 1236] Overall Loss 0.261521 Objective Loss 0.261521 LR 0.001000 Time 0.021803 +2023-10-02 20:46:18,061 - Epoch: [36][ 1140/ 1236] Overall Loss 0.261538 Objective Loss 0.261538 LR 0.001000 Time 0.021797 +2023-10-02 20:46:18,273 - Epoch: [36][ 1150/ 1236] Overall Loss 0.261722 Objective Loss 0.261722 LR 0.001000 Time 0.021791 +2023-10-02 20:46:18,485 - Epoch: [36][ 1160/ 1236] Overall Loss 0.261662 Objective Loss 0.261662 LR 0.001000 Time 0.021784 +2023-10-02 20:46:18,697 - Epoch: [36][ 1170/ 1236] Overall Loss 0.261944 Objective Loss 0.261944 LR 0.001000 Time 0.021779 +2023-10-02 20:46:18,909 - Epoch: [36][ 1180/ 1236] Overall Loss 0.262160 Objective Loss 0.262160 LR 0.001000 Time 0.021774 +2023-10-02 20:46:19,121 - Epoch: [36][ 1190/ 1236] Overall Loss 0.262079 Objective Loss 0.262079 LR 0.001000 Time 0.021768 +2023-10-02 20:46:19,333 - Epoch: [36][ 1200/ 1236] Overall Loss 0.262050 Objective Loss 0.262050 LR 0.001000 Time 0.021764 +2023-10-02 20:46:19,545 - Epoch: [36][ 1210/ 1236] Overall Loss 0.262161 Objective Loss 0.262161 LR 0.001000 Time 0.021758 +2023-10-02 20:46:19,757 - Epoch: [36][ 1220/ 1236] Overall Loss 0.262282 Objective Loss 0.262282 LR 0.001000 Time 0.021753 +2023-10-02 20:46:20,023 - Epoch: [36][ 1230/ 1236] Overall Loss 0.262479 Objective Loss 0.262479 LR 0.001000 Time 0.021791 +2023-10-02 20:46:20,144 - Epoch: [36][ 1236/ 1236] Overall Loss 0.262650 Objective Loss 0.262650 Top1 85.539715 Top5 97.352342 LR 0.001000 Time 0.021784 +2023-10-02 20:46:20,269 - --- validate (epoch=36)----------- +2023-10-02 20:46:20,269 - 29943 samples (256 per mini-batch) +2023-10-02 20:46:20,752 - Epoch: [36][ 10/ 117] Loss 0.311412 Top1 84.804688 Top5 98.867188 +2023-10-02 20:46:20,903 - Epoch: [36][ 20/ 117] Loss 0.317148 Top1 84.238281 Top5 98.476562 +2023-10-02 20:46:21,054 - Epoch: [36][ 30/ 117] Loss 0.336251 Top1 83.723958 Top5 98.333333 +2023-10-02 20:46:21,204 - Epoch: [36][ 40/ 117] Loss 0.334723 Top1 83.828125 Top5 98.261719 +2023-10-02 20:46:21,357 - Epoch: [36][ 50/ 117] Loss 0.344283 Top1 83.492188 Top5 98.242188 +2023-10-02 20:46:21,508 - Epoch: [36][ 60/ 117] Loss 0.340340 Top1 83.658854 Top5 98.248698 +2023-10-02 20:46:21,660 - Epoch: [36][ 70/ 117] Loss 0.334926 Top1 83.816964 Top5 98.292411 +2023-10-02 20:46:21,811 - Epoch: [36][ 80/ 117] Loss 0.334857 Top1 83.833008 Top5 98.266602 +2023-10-02 20:46:21,964 - Epoch: [36][ 90/ 117] Loss 0.332909 Top1 83.893229 Top5 98.263889 +2023-10-02 20:46:22,115 - Epoch: [36][ 100/ 117] Loss 0.333900 Top1 83.894531 Top5 98.250000 +2023-10-02 20:46:22,273 - Epoch: [36][ 110/ 117] Loss 0.329881 Top1 84.005682 Top5 98.249290 +2023-10-02 20:46:22,361 - Epoch: [36][ 117/ 117] Loss 0.330060 Top1 84.059713 Top5 98.286745 +2023-10-02 20:46:22,500 - ==> Top1: 84.060 Top5: 98.287 Loss: 0.330 + +2023-10-02 20:46:22,500 - ==> Confusion: +[[ 937 0 10 0 13 1 1 1 10 47 0 0 3 2 2 0 6 3 0 0 14] + [ 0 1054 2 1 3 20 1 14 0 2 2 1 1 0 0 4 4 1 11 0 10] + [ 2 0 981 13 1 0 12 3 0 0 4 1 10 3 0 3 2 2 6 2 11] + [ 3 5 19 975 0 2 2 0 12 0 4 0 1 3 17 1 0 3 22 1 19] + [ 24 12 4 0 953 7 0 0 0 7 3 2 2 3 11 2 11 1 0 2 6] + [ 4 46 3 2 2 971 0 30 1 4 1 14 2 1 10 0 3 0 4 4 14] + [ 0 2 68 0 0 0 1088 2 0 0 0 6 1 0 1 3 0 0 1 5 14] + [ 5 17 37 0 3 20 5 1027 0 2 2 7 2 4 2 2 1 0 66 7 9] + [ 16 3 0 1 1 3 0 0 999 21 12 4 1 9 13 1 2 1 1 1 0] + [ 128 1 0 0 14 3 1 0 34 890 4 0 0 16 8 4 2 0 0 2 12] + [ 2 1 15 8 1 0 0 3 9 0 973 2 3 8 5 1 0 2 10 1 9] + [ 0 1 1 0 1 5 0 2 0 2 0 956 29 6 0 5 1 14 0 3 9] + [ 1 2 3 2 2 3 0 2 1 0 0 47 960 2 0 10 2 10 0 2 19] + [ 2 0 2 1 5 13 1 0 23 14 14 8 2 998 7 1 3 1 0 3 21] + [ 6 1 6 10 4 0 0 0 21 2 1 0 5 1 1031 0 2 0 6 0 5] + [ 0 0 2 1 5 0 1 0 0 0 0 9 11 0 0 1071 19 9 0 3 3] + [ 1 12 2 1 2 0 0 0 1 0 0 5 4 0 3 8 1103 1 1 1 16] + [ 0 0 1 2 0 0 1 0 0 0 1 7 28 0 8 8 1 972 3 1 5] + [ 1 4 9 15 1 0 0 16 1 0 2 2 2 0 9 0 1 0 992 0 13] + [ 0 1 4 1 1 2 4 9 0 1 1 16 4 1 0 6 15 1 3 1068 14] + [ 151 240 193 84 71 117 21 69 121 66 185 144 356 192 137 57 130 54 192 154 5171]] + +2023-10-02 20:46:22,501 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:46:22,502 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:46:22,507 - + +2023-10-02 20:46:22,508 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:46:23,511 - Epoch: [37][ 10/ 1236] Overall Loss 0.247425 Objective Loss 0.247425 LR 0.001000 Time 0.100279 +2023-10-02 20:46:23,718 - Epoch: [37][ 20/ 1236] Overall Loss 0.247309 Objective Loss 0.247309 LR 0.001000 Time 0.060494 +2023-10-02 20:46:23,926 - Epoch: [37][ 30/ 1236] Overall Loss 0.241749 Objective Loss 0.241749 LR 0.001000 Time 0.047199 +2023-10-02 20:46:24,135 - Epoch: [37][ 40/ 1236] Overall Loss 0.241904 Objective Loss 0.241904 LR 0.001000 Time 0.040609 +2023-10-02 20:46:24,341 - Epoch: [37][ 50/ 1236] Overall Loss 0.246032 Objective Loss 0.246032 LR 0.001000 Time 0.036617 +2023-10-02 20:46:24,550 - Epoch: [37][ 60/ 1236] Overall Loss 0.245130 Objective Loss 0.245130 LR 0.001000 Time 0.033993 +2023-10-02 20:46:24,757 - Epoch: [37][ 70/ 1236] Overall Loss 0.245622 Objective Loss 0.245622 LR 0.001000 Time 0.032082 +2023-10-02 20:46:24,965 - Epoch: [37][ 80/ 1236] Overall Loss 0.246634 Objective Loss 0.246634 LR 0.001000 Time 0.030677 +2023-10-02 20:46:25,172 - Epoch: [37][ 90/ 1236] Overall Loss 0.244586 Objective Loss 0.244586 LR 0.001000 Time 0.029565 +2023-10-02 20:46:25,381 - Epoch: [37][ 100/ 1236] Overall Loss 0.241474 Objective Loss 0.241474 LR 0.001000 Time 0.028696 +2023-10-02 20:46:25,588 - Epoch: [37][ 110/ 1236] Overall Loss 0.242925 Objective Loss 0.242925 LR 0.001000 Time 0.027963 +2023-10-02 20:46:25,797 - Epoch: [37][ 120/ 1236] Overall Loss 0.244001 Objective Loss 0.244001 LR 0.001000 Time 0.027371 +2023-10-02 20:46:26,003 - Epoch: [37][ 130/ 1236] Overall Loss 0.244287 Objective Loss 0.244287 LR 0.001000 Time 0.026851 +2023-10-02 20:46:26,212 - Epoch: [37][ 140/ 1236] Overall Loss 0.245854 Objective Loss 0.245854 LR 0.001000 Time 0.026424 +2023-10-02 20:46:26,419 - Epoch: [37][ 150/ 1236] Overall Loss 0.247493 Objective Loss 0.247493 LR 0.001000 Time 0.026038 +2023-10-02 20:46:26,628 - Epoch: [37][ 160/ 1236] Overall Loss 0.250061 Objective Loss 0.250061 LR 0.001000 Time 0.025713 +2023-10-02 20:46:26,834 - Epoch: [37][ 170/ 1236] Overall Loss 0.252081 Objective Loss 0.252081 LR 0.001000 Time 0.025414 +2023-10-02 20:46:27,043 - Epoch: [37][ 180/ 1236] Overall Loss 0.252481 Objective Loss 0.252481 LR 0.001000 Time 0.025161 +2023-10-02 20:46:27,250 - Epoch: [37][ 190/ 1236] Overall Loss 0.253357 Objective Loss 0.253357 LR 0.001000 Time 0.024923 +2023-10-02 20:46:27,459 - Epoch: [37][ 200/ 1236] Overall Loss 0.253736 Objective Loss 0.253736 LR 0.001000 Time 0.024722 +2023-10-02 20:46:27,665 - Epoch: [37][ 210/ 1236] Overall Loss 0.253150 Objective Loss 0.253150 LR 0.001000 Time 0.024524 +2023-10-02 20:46:27,873 - Epoch: [37][ 220/ 1236] Overall Loss 0.254498 Objective Loss 0.254498 LR 0.001000 Time 0.024355 +2023-10-02 20:46:28,080 - Epoch: [37][ 230/ 1236] Overall Loss 0.255521 Objective Loss 0.255521 LR 0.001000 Time 0.024192 +2023-10-02 20:46:28,288 - Epoch: [37][ 240/ 1236] Overall Loss 0.255520 Objective Loss 0.255520 LR 0.001000 Time 0.024051 +2023-10-02 20:46:28,493 - Epoch: [37][ 250/ 1236] Overall Loss 0.255432 Objective Loss 0.255432 LR 0.001000 Time 0.023910 +2023-10-02 20:46:28,702 - Epoch: [37][ 260/ 1236] Overall Loss 0.256056 Objective Loss 0.256056 LR 0.001000 Time 0.023792 +2023-10-02 20:46:28,907 - Epoch: [37][ 270/ 1236] Overall Loss 0.256139 Objective Loss 0.256139 LR 0.001000 Time 0.023669 +2023-10-02 20:46:29,116 - Epoch: [37][ 280/ 1236] Overall Loss 0.254972 Objective Loss 0.254972 LR 0.001000 Time 0.023569 +2023-10-02 20:46:29,321 - Epoch: [37][ 290/ 1236] Overall Loss 0.255095 Objective Loss 0.255095 LR 0.001000 Time 0.023463 +2023-10-02 20:46:29,530 - Epoch: [37][ 300/ 1236] Overall Loss 0.254011 Objective Loss 0.254011 LR 0.001000 Time 0.023376 +2023-10-02 20:46:29,735 - Epoch: [37][ 310/ 1236] Overall Loss 0.253803 Objective Loss 0.253803 LR 0.001000 Time 0.023282 +2023-10-02 20:46:29,944 - Epoch: [37][ 320/ 1236] Overall Loss 0.253738 Objective Loss 0.253738 LR 0.001000 Time 0.023206 +2023-10-02 20:46:30,149 - Epoch: [37][ 330/ 1236] Overall Loss 0.254610 Objective Loss 0.254610 LR 0.001000 Time 0.023124 +2023-10-02 20:46:30,358 - Epoch: [37][ 340/ 1236] Overall Loss 0.255960 Objective Loss 0.255960 LR 0.001000 Time 0.023057 +2023-10-02 20:46:30,563 - Epoch: [37][ 350/ 1236] Overall Loss 0.256277 Objective Loss 0.256277 LR 0.001000 Time 0.022984 +2023-10-02 20:46:30,772 - Epoch: [37][ 360/ 1236] Overall Loss 0.256803 Objective Loss 0.256803 LR 0.001000 Time 0.022926 +2023-10-02 20:46:30,977 - Epoch: [37][ 370/ 1236] Overall Loss 0.257203 Objective Loss 0.257203 LR 0.001000 Time 0.022860 +2023-10-02 20:46:31,185 - Epoch: [37][ 380/ 1236] Overall Loss 0.257158 Objective Loss 0.257158 LR 0.001000 Time 0.022805 +2023-10-02 20:46:31,392 - Epoch: [37][ 390/ 1236] Overall Loss 0.257632 Objective Loss 0.257632 LR 0.001000 Time 0.022746 +2023-10-02 20:46:31,601 - Epoch: [37][ 400/ 1236] Overall Loss 0.257918 Objective Loss 0.257918 LR 0.001000 Time 0.022699 +2023-10-02 20:46:31,806 - Epoch: [37][ 410/ 1236] Overall Loss 0.257871 Objective Loss 0.257871 LR 0.001000 Time 0.022646 +2023-10-02 20:46:32,014 - Epoch: [37][ 420/ 1236] Overall Loss 0.257632 Objective Loss 0.257632 LR 0.001000 Time 0.022601 +2023-10-02 20:46:32,221 - Epoch: [37][ 430/ 1236] Overall Loss 0.256916 Objective Loss 0.256916 LR 0.001000 Time 0.022555 +2023-10-02 20:46:32,428 - Epoch: [37][ 440/ 1236] Overall Loss 0.256515 Objective Loss 0.256515 LR 0.001000 Time 0.022514 +2023-10-02 20:46:32,635 - Epoch: [37][ 450/ 1236] Overall Loss 0.256334 Objective Loss 0.256334 LR 0.001000 Time 0.022472 +2023-10-02 20:46:32,843 - Epoch: [37][ 460/ 1236] Overall Loss 0.256512 Objective Loss 0.256512 LR 0.001000 Time 0.022434 +2023-10-02 20:46:33,049 - Epoch: [37][ 470/ 1236] Overall Loss 0.256687 Objective Loss 0.256687 LR 0.001000 Time 0.022396 +2023-10-02 20:46:33,257 - Epoch: [37][ 480/ 1236] Overall Loss 0.256823 Objective Loss 0.256823 LR 0.001000 Time 0.022362 +2023-10-02 20:46:33,464 - Epoch: [37][ 490/ 1236] Overall Loss 0.256020 Objective Loss 0.256020 LR 0.001000 Time 0.022324 +2023-10-02 20:46:33,673 - Epoch: [37][ 500/ 1236] Overall Loss 0.255926 Objective Loss 0.255926 LR 0.001000 Time 0.022295 +2023-10-02 20:46:33,878 - Epoch: [37][ 510/ 1236] Overall Loss 0.256121 Objective Loss 0.256121 LR 0.001000 Time 0.022260 +2023-10-02 20:46:34,087 - Epoch: [37][ 520/ 1236] Overall Loss 0.256408 Objective Loss 0.256408 LR 0.001000 Time 0.022233 +2023-10-02 20:46:34,293 - Epoch: [37][ 530/ 1236] Overall Loss 0.255849 Objective Loss 0.255849 LR 0.001000 Time 0.022201 +2023-10-02 20:46:34,500 - Epoch: [37][ 540/ 1236] Overall Loss 0.255942 Objective Loss 0.255942 LR 0.001000 Time 0.022174 +2023-10-02 20:46:34,707 - Epoch: [37][ 550/ 1236] Overall Loss 0.255941 Objective Loss 0.255941 LR 0.001000 Time 0.022145 +2023-10-02 20:46:34,914 - Epoch: [37][ 560/ 1236] Overall Loss 0.256005 Objective Loss 0.256005 LR 0.001000 Time 0.022120 +2023-10-02 20:46:35,121 - Epoch: [37][ 570/ 1236] Overall Loss 0.256018 Objective Loss 0.256018 LR 0.001000 Time 0.022094 +2023-10-02 20:46:35,330 - Epoch: [37][ 580/ 1236] Overall Loss 0.255962 Objective Loss 0.255962 LR 0.001000 Time 0.022073 +2023-10-02 20:46:35,536 - Epoch: [37][ 590/ 1236] Overall Loss 0.256674 Objective Loss 0.256674 LR 0.001000 Time 0.022047 +2023-10-02 20:46:35,743 - Epoch: [37][ 600/ 1236] Overall Loss 0.257192 Objective Loss 0.257192 LR 0.001000 Time 0.022025 +2023-10-02 20:46:35,950 - Epoch: [37][ 610/ 1236] Overall Loss 0.257679 Objective Loss 0.257679 LR 0.001000 Time 0.022003 +2023-10-02 20:46:36,158 - Epoch: [37][ 620/ 1236] Overall Loss 0.257349 Objective Loss 0.257349 LR 0.001000 Time 0.021982 +2023-10-02 20:46:36,364 - Epoch: [37][ 630/ 1236] Overall Loss 0.257618 Objective Loss 0.257618 LR 0.001000 Time 0.021959 +2023-10-02 20:46:36,573 - Epoch: [37][ 640/ 1236] Overall Loss 0.257667 Objective Loss 0.257667 LR 0.001000 Time 0.021942 +2023-10-02 20:46:36,779 - Epoch: [37][ 650/ 1236] Overall Loss 0.257962 Objective Loss 0.257962 LR 0.001000 Time 0.021920 +2023-10-02 20:46:36,988 - Epoch: [37][ 660/ 1236] Overall Loss 0.258140 Objective Loss 0.258140 LR 0.001000 Time 0.021904 +2023-10-02 20:46:37,193 - Epoch: [37][ 670/ 1236] Overall Loss 0.258606 Objective Loss 0.258606 LR 0.001000 Time 0.021883 +2023-10-02 20:46:37,402 - Epoch: [37][ 680/ 1236] Overall Loss 0.258402 Objective Loss 0.258402 LR 0.001000 Time 0.021868 +2023-10-02 20:46:37,608 - Epoch: [37][ 690/ 1236] Overall Loss 0.258809 Objective Loss 0.258809 LR 0.001000 Time 0.021849 +2023-10-02 20:46:37,815 - Epoch: [37][ 700/ 1236] Overall Loss 0.258802 Objective Loss 0.258802 LR 0.001000 Time 0.021833 +2023-10-02 20:46:38,022 - Epoch: [37][ 710/ 1236] Overall Loss 0.259183 Objective Loss 0.259183 LR 0.001000 Time 0.021815 +2023-10-02 20:46:38,231 - Epoch: [37][ 720/ 1236] Overall Loss 0.259364 Objective Loss 0.259364 LR 0.001000 Time 0.021802 +2023-10-02 20:46:38,437 - Epoch: [37][ 730/ 1236] Overall Loss 0.259693 Objective Loss 0.259693 LR 0.001000 Time 0.021784 +2023-10-02 20:46:38,644 - Epoch: [37][ 740/ 1236] Overall Loss 0.260002 Objective Loss 0.260002 LR 0.001000 Time 0.021770 +2023-10-02 20:46:38,851 - Epoch: [37][ 750/ 1236] Overall Loss 0.260141 Objective Loss 0.260141 LR 0.001000 Time 0.021753 +2023-10-02 20:46:39,060 - Epoch: [37][ 760/ 1236] Overall Loss 0.260144 Objective Loss 0.260144 LR 0.001000 Time 0.021742 +2023-10-02 20:46:39,265 - Epoch: [37][ 770/ 1236] Overall Loss 0.260458 Objective Loss 0.260458 LR 0.001000 Time 0.021726 +2023-10-02 20:46:39,474 - Epoch: [37][ 780/ 1236] Overall Loss 0.260440 Objective Loss 0.260440 LR 0.001000 Time 0.021715 +2023-10-02 20:46:39,680 - Epoch: [37][ 790/ 1236] Overall Loss 0.260518 Objective Loss 0.260518 LR 0.001000 Time 0.021700 +2023-10-02 20:46:39,887 - Epoch: [37][ 800/ 1236] Overall Loss 0.260487 Objective Loss 0.260487 LR 0.001000 Time 0.021687 +2023-10-02 20:46:40,094 - Epoch: [37][ 810/ 1236] Overall Loss 0.260959 Objective Loss 0.260959 LR 0.001000 Time 0.021673 +2023-10-02 20:46:40,302 - Epoch: [37][ 820/ 1236] Overall Loss 0.261054 Objective Loss 0.261054 LR 0.001000 Time 0.021663 +2023-10-02 20:46:40,508 - Epoch: [37][ 830/ 1236] Overall Loss 0.261092 Objective Loss 0.261092 LR 0.001000 Time 0.021649 +2023-10-02 20:46:40,716 - Epoch: [37][ 840/ 1236] Overall Loss 0.261013 Objective Loss 0.261013 LR 0.001000 Time 0.021639 +2023-10-02 20:46:40,922 - Epoch: [37][ 850/ 1236] Overall Loss 0.261353 Objective Loss 0.261353 LR 0.001000 Time 0.021626 +2023-10-02 20:46:41,131 - Epoch: [37][ 860/ 1236] Overall Loss 0.261421 Objective Loss 0.261421 LR 0.001000 Time 0.021617 +2023-10-02 20:46:41,336 - Epoch: [37][ 870/ 1236] Overall Loss 0.261234 Objective Loss 0.261234 LR 0.001000 Time 0.021605 +2023-10-02 20:46:41,544 - Epoch: [37][ 880/ 1236] Overall Loss 0.261405 Objective Loss 0.261405 LR 0.001000 Time 0.021595 +2023-10-02 20:46:41,751 - Epoch: [37][ 890/ 1236] Overall Loss 0.261609 Objective Loss 0.261609 LR 0.001000 Time 0.021583 +2023-10-02 20:46:41,960 - Epoch: [37][ 900/ 1236] Overall Loss 0.261830 Objective Loss 0.261830 LR 0.001000 Time 0.021575 +2023-10-02 20:46:42,166 - Epoch: [37][ 910/ 1236] Overall Loss 0.261593 Objective Loss 0.261593 LR 0.001000 Time 0.021564 +2023-10-02 20:46:42,374 - Epoch: [37][ 920/ 1236] Overall Loss 0.262010 Objective Loss 0.262010 LR 0.001000 Time 0.021556 +2023-10-02 20:46:42,580 - Epoch: [37][ 930/ 1236] Overall Loss 0.262203 Objective Loss 0.262203 LR 0.001000 Time 0.021545 +2023-10-02 20:46:42,787 - Epoch: [37][ 940/ 1236] Overall Loss 0.262482 Objective Loss 0.262482 LR 0.001000 Time 0.021536 +2023-10-02 20:46:42,994 - Epoch: [37][ 950/ 1236] Overall Loss 0.262539 Objective Loss 0.262539 LR 0.001000 Time 0.021525 +2023-10-02 20:46:43,203 - Epoch: [37][ 960/ 1236] Overall Loss 0.262785 Objective Loss 0.262785 LR 0.001000 Time 0.021519 +2023-10-02 20:46:43,409 - Epoch: [37][ 970/ 1236] Overall Loss 0.262984 Objective Loss 0.262984 LR 0.001000 Time 0.021509 +2023-10-02 20:46:43,617 - Epoch: [37][ 980/ 1236] Overall Loss 0.263425 Objective Loss 0.263425 LR 0.001000 Time 0.021502 +2023-10-02 20:46:43,823 - Epoch: [37][ 990/ 1236] Overall Loss 0.264209 Objective Loss 0.264209 LR 0.001000 Time 0.021492 +2023-10-02 20:46:44,032 - Epoch: [37][ 1000/ 1236] Overall Loss 0.264356 Objective Loss 0.264356 LR 0.001000 Time 0.021486 +2023-10-02 20:46:44,238 - Epoch: [37][ 1010/ 1236] Overall Loss 0.264366 Objective Loss 0.264366 LR 0.001000 Time 0.021477 +2023-10-02 20:46:44,447 - Epoch: [37][ 1020/ 1236] Overall Loss 0.264624 Objective Loss 0.264624 LR 0.001000 Time 0.021471 +2023-10-02 20:46:44,652 - Epoch: [37][ 1030/ 1236] Overall Loss 0.265007 Objective Loss 0.265007 LR 0.001000 Time 0.021462 +2023-10-02 20:46:44,860 - Epoch: [37][ 1040/ 1236] Overall Loss 0.265075 Objective Loss 0.265075 LR 0.001000 Time 0.021455 +2023-10-02 20:46:45,067 - Epoch: [37][ 1050/ 1236] Overall Loss 0.265190 Objective Loss 0.265190 LR 0.001000 Time 0.021446 +2023-10-02 20:46:45,276 - Epoch: [37][ 1060/ 1236] Overall Loss 0.265176 Objective Loss 0.265176 LR 0.001000 Time 0.021441 +2023-10-02 20:46:45,482 - Epoch: [37][ 1070/ 1236] Overall Loss 0.265154 Objective Loss 0.265154 LR 0.001000 Time 0.021432 +2023-10-02 20:46:45,690 - Epoch: [37][ 1080/ 1236] Overall Loss 0.265097 Objective Loss 0.265097 LR 0.001000 Time 0.021426 +2023-10-02 20:46:45,896 - Epoch: [37][ 1090/ 1236] Overall Loss 0.265027 Objective Loss 0.265027 LR 0.001000 Time 0.021418 +2023-10-02 20:46:46,104 - Epoch: [37][ 1100/ 1236] Overall Loss 0.265103 Objective Loss 0.265103 LR 0.001000 Time 0.021411 +2023-10-02 20:46:46,311 - Epoch: [37][ 1110/ 1236] Overall Loss 0.265224 Objective Loss 0.265224 LR 0.001000 Time 0.021404 +2023-10-02 20:46:46,519 - Epoch: [37][ 1120/ 1236] Overall Loss 0.265311 Objective Loss 0.265311 LR 0.001000 Time 0.021398 +2023-10-02 20:46:46,726 - Epoch: [37][ 1130/ 1236] Overall Loss 0.265784 Objective Loss 0.265784 LR 0.001000 Time 0.021390 +2023-10-02 20:46:46,933 - Epoch: [37][ 1140/ 1236] Overall Loss 0.265734 Objective Loss 0.265734 LR 0.001000 Time 0.021385 +2023-10-02 20:46:47,140 - Epoch: [37][ 1150/ 1236] Overall Loss 0.265569 Objective Loss 0.265569 LR 0.001000 Time 0.021377 +2023-10-02 20:46:47,348 - Epoch: [37][ 1160/ 1236] Overall Loss 0.265542 Objective Loss 0.265542 LR 0.001000 Time 0.021372 +2023-10-02 20:46:47,555 - Epoch: [37][ 1170/ 1236] Overall Loss 0.265556 Objective Loss 0.265556 LR 0.001000 Time 0.021365 +2023-10-02 20:46:47,764 - Epoch: [37][ 1180/ 1236] Overall Loss 0.265646 Objective Loss 0.265646 LR 0.001000 Time 0.021361 +2023-10-02 20:46:47,970 - Epoch: [37][ 1190/ 1236] Overall Loss 0.265714 Objective Loss 0.265714 LR 0.001000 Time 0.021354 +2023-10-02 20:46:48,179 - Epoch: [37][ 1200/ 1236] Overall Loss 0.265597 Objective Loss 0.265597 LR 0.001000 Time 0.021350 +2023-10-02 20:46:48,385 - Epoch: [37][ 1210/ 1236] Overall Loss 0.265908 Objective Loss 0.265908 LR 0.001000 Time 0.021343 +2023-10-02 20:46:48,592 - Epoch: [37][ 1220/ 1236] Overall Loss 0.265881 Objective Loss 0.265881 LR 0.001000 Time 0.021338 +2023-10-02 20:46:48,852 - Epoch: [37][ 1230/ 1236] Overall Loss 0.265801 Objective Loss 0.265801 LR 0.001000 Time 0.021375 +2023-10-02 20:46:48,974 - Epoch: [37][ 1236/ 1236] Overall Loss 0.265954 Objective Loss 0.265954 Top1 85.743381 Top5 98.574338 LR 0.001000 Time 0.021370 +2023-10-02 20:46:49,108 - --- validate (epoch=37)----------- +2023-10-02 20:46:49,109 - 29943 samples (256 per mini-batch) +2023-10-02 20:46:49,579 - Epoch: [37][ 10/ 117] Loss 0.339739 Top1 84.140625 Top5 98.125000 +2023-10-02 20:46:49,726 - Epoch: [37][ 20/ 117] Loss 0.353602 Top1 83.183594 Top5 97.890625 +2023-10-02 20:46:49,871 - Epoch: [37][ 30/ 117] Loss 0.339142 Top1 83.085938 Top5 97.968750 +2023-10-02 20:46:50,016 - Epoch: [37][ 40/ 117] Loss 0.326502 Top1 83.173828 Top5 98.066406 +2023-10-02 20:46:50,160 - Epoch: [37][ 50/ 117] Loss 0.321020 Top1 83.398438 Top5 98.078125 +2023-10-02 20:46:50,305 - Epoch: [37][ 60/ 117] Loss 0.324587 Top1 83.307292 Top5 98.092448 +2023-10-02 20:46:50,451 - Epoch: [37][ 70/ 117] Loss 0.321796 Top1 83.364955 Top5 98.158482 +2023-10-02 20:46:50,596 - Epoch: [37][ 80/ 117] Loss 0.322668 Top1 83.393555 Top5 98.149414 +2023-10-02 20:46:50,742 - Epoch: [37][ 90/ 117] Loss 0.322885 Top1 83.342014 Top5 98.203125 +2023-10-02 20:46:50,889 - Epoch: [37][ 100/ 117] Loss 0.323221 Top1 83.222656 Top5 98.187500 +2023-10-02 20:46:51,042 - Epoch: [37][ 110/ 117] Loss 0.327358 Top1 83.160511 Top5 98.139205 +2023-10-02 20:46:51,130 - Epoch: [37][ 117/ 117] Loss 0.326631 Top1 83.147981 Top5 98.143139 +2023-10-02 20:46:51,234 - ==> Top1: 83.148 Top5: 98.143 Loss: 0.327 + +2023-10-02 20:46:51,235 - ==> Confusion: +[[ 934 0 1 1 13 1 1 0 0 71 2 0 1 1 4 3 4 1 1 0 11] + [ 0 1041 1 0 4 33 5 17 2 0 4 0 0 0 1 4 0 0 8 3 8] + [ 4 0 950 16 1 0 34 8 0 1 4 1 4 2 1 2 2 1 11 9 5] + [ 3 7 8 993 0 0 2 0 3 0 4 0 2 4 25 3 2 2 15 0 16] + [ 27 3 2 0 967 10 0 0 0 10 2 0 0 5 8 2 5 1 0 2 6] + [ 6 34 1 7 1 989 4 12 4 4 4 2 3 17 2 1 2 1 8 5 9] + [ 0 3 33 0 0 1 1129 4 0 0 2 0 0 0 1 2 0 1 3 8 4] + [ 3 20 25 0 5 37 2 1032 1 2 4 3 0 5 1 2 1 1 63 5 6] + [ 26 3 3 0 2 1 0 1 944 49 11 0 2 14 20 0 3 0 6 1 3] + [ 103 1 1 0 12 1 0 0 20 944 3 0 0 16 6 2 0 0 0 5 5] + [ 3 0 10 10 3 1 8 0 11 0 966 2 0 11 3 1 1 3 13 2 5] + [ 0 2 4 0 3 12 0 6 0 2 0 938 17 7 0 7 1 22 1 7 6] + [ 0 2 2 4 1 2 4 1 0 0 2 56 920 1 5 11 4 26 10 5 12] + [ 2 0 2 0 5 12 1 0 11 13 10 9 0 1030 5 0 1 2 0 5 11] + [ 13 2 3 19 6 0 0 0 21 8 4 0 0 3 997 0 1 8 6 0 10] + [ 0 0 1 3 6 0 4 0 0 0 0 7 3 1 0 1069 13 13 3 7 4] + [ 0 10 2 0 9 5 1 0 1 0 0 5 3 1 3 17 1088 2 0 5 9] + [ 0 1 0 4 0 0 1 0 1 0 1 3 20 0 2 5 0 996 0 2 2] + [ 2 8 5 17 1 0 1 9 7 0 3 0 1 0 5 1 0 0 1000 0 8] + [ 0 2 3 1 2 1 11 6 0 1 4 12 2 0 1 5 3 2 3 1086 7] + [ 159 187 135 97 83 222 72 63 131 108 182 111 312 278 153 79 109 75 226 239 4884]] + +2023-10-02 20:46:51,236 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:46:51,236 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:46:51,242 - + +2023-10-02 20:46:51,242 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:46:52,362 - Epoch: [38][ 10/ 1236] Overall Loss 0.272967 Objective Loss 0.272967 LR 0.001000 Time 0.111968 +2023-10-02 20:46:52,570 - Epoch: [38][ 20/ 1236] Overall Loss 0.266909 Objective Loss 0.266909 LR 0.001000 Time 0.066371 +2023-10-02 20:46:52,777 - Epoch: [38][ 30/ 1236] Overall Loss 0.258645 Objective Loss 0.258645 LR 0.001000 Time 0.051105 +2023-10-02 20:46:52,986 - Epoch: [38][ 40/ 1236] Overall Loss 0.255425 Objective Loss 0.255425 LR 0.001000 Time 0.043542 +2023-10-02 20:46:53,192 - Epoch: [38][ 50/ 1236] Overall Loss 0.252420 Objective Loss 0.252420 LR 0.001000 Time 0.038940 +2023-10-02 20:46:53,400 - Epoch: [38][ 60/ 1236] Overall Loss 0.253520 Objective Loss 0.253520 LR 0.001000 Time 0.035924 +2023-10-02 20:46:53,606 - Epoch: [38][ 70/ 1236] Overall Loss 0.250622 Objective Loss 0.250622 LR 0.001000 Time 0.033726 +2023-10-02 20:46:53,815 - Epoch: [38][ 80/ 1236] Overall Loss 0.245390 Objective Loss 0.245390 LR 0.001000 Time 0.032115 +2023-10-02 20:46:54,020 - Epoch: [38][ 90/ 1236] Overall Loss 0.246236 Objective Loss 0.246236 LR 0.001000 Time 0.030828 +2023-10-02 20:46:54,229 - Epoch: [38][ 100/ 1236] Overall Loss 0.249822 Objective Loss 0.249822 LR 0.001000 Time 0.029832 +2023-10-02 20:46:54,435 - Epoch: [38][ 110/ 1236] Overall Loss 0.248728 Objective Loss 0.248728 LR 0.001000 Time 0.028983 +2023-10-02 20:46:54,643 - Epoch: [38][ 120/ 1236] Overall Loss 0.250356 Objective Loss 0.250356 LR 0.001000 Time 0.028305 +2023-10-02 20:46:54,849 - Epoch: [38][ 130/ 1236] Overall Loss 0.251288 Objective Loss 0.251288 LR 0.001000 Time 0.027706 +2023-10-02 20:46:55,057 - Epoch: [38][ 140/ 1236] Overall Loss 0.251464 Objective Loss 0.251464 LR 0.001000 Time 0.027216 +2023-10-02 20:46:55,263 - Epoch: [38][ 150/ 1236] Overall Loss 0.253238 Objective Loss 0.253238 LR 0.001000 Time 0.026772 +2023-10-02 20:46:55,472 - Epoch: [38][ 160/ 1236] Overall Loss 0.254557 Objective Loss 0.254557 LR 0.001000 Time 0.026403 +2023-10-02 20:46:55,677 - Epoch: [38][ 170/ 1236] Overall Loss 0.255973 Objective Loss 0.255973 LR 0.001000 Time 0.026056 +2023-10-02 20:46:55,888 - Epoch: [38][ 180/ 1236] Overall Loss 0.255154 Objective Loss 0.255154 LR 0.001000 Time 0.025777 +2023-10-02 20:46:56,095 - Epoch: [38][ 190/ 1236] Overall Loss 0.254546 Objective Loss 0.254546 LR 0.001000 Time 0.025507 +2023-10-02 20:46:56,305 - Epoch: [38][ 200/ 1236] Overall Loss 0.253825 Objective Loss 0.253825 LR 0.001000 Time 0.025279 +2023-10-02 20:46:56,510 - Epoch: [38][ 210/ 1236] Overall Loss 0.253592 Objective Loss 0.253592 LR 0.001000 Time 0.025052 +2023-10-02 20:46:56,717 - Epoch: [38][ 220/ 1236] Overall Loss 0.254931 Objective Loss 0.254931 LR 0.001000 Time 0.024854 +2023-10-02 20:46:56,923 - Epoch: [38][ 230/ 1236] Overall Loss 0.255068 Objective Loss 0.255068 LR 0.001000 Time 0.024662 +2023-10-02 20:46:57,131 - Epoch: [38][ 240/ 1236] Overall Loss 0.254680 Objective Loss 0.254680 LR 0.001000 Time 0.024501 +2023-10-02 20:46:57,337 - Epoch: [38][ 250/ 1236] Overall Loss 0.253157 Objective Loss 0.253157 LR 0.001000 Time 0.024341 +2023-10-02 20:46:57,545 - Epoch: [38][ 260/ 1236] Overall Loss 0.254457 Objective Loss 0.254457 LR 0.001000 Time 0.024206 +2023-10-02 20:46:57,750 - Epoch: [38][ 270/ 1236] Overall Loss 0.253453 Objective Loss 0.253453 LR 0.001000 Time 0.024068 +2023-10-02 20:46:57,959 - Epoch: [38][ 280/ 1236] Overall Loss 0.253664 Objective Loss 0.253664 LR 0.001000 Time 0.023953 +2023-10-02 20:46:58,165 - Epoch: [38][ 290/ 1236] Overall Loss 0.253348 Objective Loss 0.253348 LR 0.001000 Time 0.023835 +2023-10-02 20:46:58,372 - Epoch: [38][ 300/ 1236] Overall Loss 0.253142 Objective Loss 0.253142 LR 0.001000 Time 0.023732 +2023-10-02 20:46:58,579 - Epoch: [38][ 310/ 1236] Overall Loss 0.253066 Objective Loss 0.253066 LR 0.001000 Time 0.023630 +2023-10-02 20:46:58,788 - Epoch: [38][ 320/ 1236] Overall Loss 0.252817 Objective Loss 0.252817 LR 0.001000 Time 0.023544 +2023-10-02 20:46:58,994 - Epoch: [38][ 330/ 1236] Overall Loss 0.252705 Objective Loss 0.252705 LR 0.001000 Time 0.023453 +2023-10-02 20:46:59,202 - Epoch: [38][ 340/ 1236] Overall Loss 0.253430 Objective Loss 0.253430 LR 0.001000 Time 0.023372 +2023-10-02 20:46:59,408 - Epoch: [38][ 350/ 1236] Overall Loss 0.253344 Objective Loss 0.253344 LR 0.001000 Time 0.023291 +2023-10-02 20:46:59,616 - Epoch: [38][ 360/ 1236] Overall Loss 0.253232 Objective Loss 0.253232 LR 0.001000 Time 0.023221 +2023-10-02 20:46:59,824 - Epoch: [38][ 370/ 1236] Overall Loss 0.253766 Objective Loss 0.253766 LR 0.001000 Time 0.023150 +2023-10-02 20:47:00,033 - Epoch: [38][ 380/ 1236] Overall Loss 0.253058 Objective Loss 0.253058 LR 0.001000 Time 0.023092 +2023-10-02 20:47:00,240 - Epoch: [38][ 390/ 1236] Overall Loss 0.253541 Objective Loss 0.253541 LR 0.001000 Time 0.023029 +2023-10-02 20:47:00,450 - Epoch: [38][ 400/ 1236] Overall Loss 0.253917 Objective Loss 0.253917 LR 0.001000 Time 0.022977 +2023-10-02 20:47:00,656 - Epoch: [38][ 410/ 1236] Overall Loss 0.254320 Objective Loss 0.254320 LR 0.001000 Time 0.022918 +2023-10-02 20:47:00,865 - Epoch: [38][ 420/ 1236] Overall Loss 0.254791 Objective Loss 0.254791 LR 0.001000 Time 0.022870 +2023-10-02 20:47:01,071 - Epoch: [38][ 430/ 1236] Overall Loss 0.254848 Objective Loss 0.254848 LR 0.001000 Time 0.022816 +2023-10-02 20:47:01,281 - Epoch: [38][ 440/ 1236] Overall Loss 0.255201 Objective Loss 0.255201 LR 0.001000 Time 0.022774 +2023-10-02 20:47:01,487 - Epoch: [38][ 450/ 1236] Overall Loss 0.255704 Objective Loss 0.255704 LR 0.001000 Time 0.022725 +2023-10-02 20:47:01,696 - Epoch: [38][ 460/ 1236] Overall Loss 0.256002 Objective Loss 0.256002 LR 0.001000 Time 0.022686 +2023-10-02 20:47:01,903 - Epoch: [38][ 470/ 1236] Overall Loss 0.256636 Objective Loss 0.256636 LR 0.001000 Time 0.022642 +2023-10-02 20:47:02,111 - Epoch: [38][ 480/ 1236] Overall Loss 0.256926 Objective Loss 0.256926 LR 0.001000 Time 0.022603 +2023-10-02 20:47:02,319 - Epoch: [38][ 490/ 1236] Overall Loss 0.257051 Objective Loss 0.257051 LR 0.001000 Time 0.022563 +2023-10-02 20:47:02,528 - Epoch: [38][ 500/ 1236] Overall Loss 0.257552 Objective Loss 0.257552 LR 0.001000 Time 0.022530 +2023-10-02 20:47:02,735 - Epoch: [38][ 510/ 1236] Overall Loss 0.257461 Objective Loss 0.257461 LR 0.001000 Time 0.022493 +2023-10-02 20:47:02,944 - Epoch: [38][ 520/ 1236] Overall Loss 0.258256 Objective Loss 0.258256 LR 0.001000 Time 0.022463 +2023-10-02 20:47:03,151 - Epoch: [38][ 530/ 1236] Overall Loss 0.258168 Objective Loss 0.258168 LR 0.001000 Time 0.022428 +2023-10-02 20:47:03,361 - Epoch: [38][ 540/ 1236] Overall Loss 0.258099 Objective Loss 0.258099 LR 0.001000 Time 0.022401 +2023-10-02 20:47:03,567 - Epoch: [38][ 550/ 1236] Overall Loss 0.258246 Objective Loss 0.258246 LR 0.001000 Time 0.022369 +2023-10-02 20:47:03,776 - Epoch: [38][ 560/ 1236] Overall Loss 0.258721 Objective Loss 0.258721 LR 0.001000 Time 0.022341 +2023-10-02 20:47:03,983 - Epoch: [38][ 570/ 1236] Overall Loss 0.259248 Objective Loss 0.259248 LR 0.001000 Time 0.022311 +2023-10-02 20:47:04,193 - Epoch: [38][ 580/ 1236] Overall Loss 0.259532 Objective Loss 0.259532 LR 0.001000 Time 0.022288 +2023-10-02 20:47:04,400 - Epoch: [38][ 590/ 1236] Overall Loss 0.259633 Objective Loss 0.259633 LR 0.001000 Time 0.022260 +2023-10-02 20:47:04,610 - Epoch: [38][ 600/ 1236] Overall Loss 0.259543 Objective Loss 0.259543 LR 0.001000 Time 0.022238 +2023-10-02 20:47:04,816 - Epoch: [38][ 610/ 1236] Overall Loss 0.259421 Objective Loss 0.259421 LR 0.001000 Time 0.022211 +2023-10-02 20:47:05,026 - Epoch: [38][ 620/ 1236] Overall Loss 0.259325 Objective Loss 0.259325 LR 0.001000 Time 0.022191 +2023-10-02 20:47:05,232 - Epoch: [38][ 630/ 1236] Overall Loss 0.258880 Objective Loss 0.258880 LR 0.001000 Time 0.022166 +2023-10-02 20:47:05,441 - Epoch: [38][ 640/ 1236] Overall Loss 0.258919 Objective Loss 0.258919 LR 0.001000 Time 0.022145 +2023-10-02 20:47:05,648 - Epoch: [38][ 650/ 1236] Overall Loss 0.259222 Objective Loss 0.259222 LR 0.001000 Time 0.022121 +2023-10-02 20:47:05,858 - Epoch: [38][ 660/ 1236] Overall Loss 0.259185 Objective Loss 0.259185 LR 0.001000 Time 0.022104 +2023-10-02 20:47:06,064 - Epoch: [38][ 670/ 1236] Overall Loss 0.258850 Objective Loss 0.258850 LR 0.001000 Time 0.022081 +2023-10-02 20:47:06,273 - Epoch: [38][ 680/ 1236] Overall Loss 0.259056 Objective Loss 0.259056 LR 0.001000 Time 0.022062 +2023-10-02 20:47:06,480 - Epoch: [38][ 690/ 1236] Overall Loss 0.259169 Objective Loss 0.259169 LR 0.001000 Time 0.022041 +2023-10-02 20:47:06,689 - Epoch: [38][ 700/ 1236] Overall Loss 0.259580 Objective Loss 0.259580 LR 0.001000 Time 0.022024 +2023-10-02 20:47:06,896 - Epoch: [38][ 710/ 1236] Overall Loss 0.259411 Objective Loss 0.259411 LR 0.001000 Time 0.022004 +2023-10-02 20:47:07,105 - Epoch: [38][ 720/ 1236] Overall Loss 0.259429 Objective Loss 0.259429 LR 0.001000 Time 0.021987 +2023-10-02 20:47:07,312 - Epoch: [38][ 730/ 1236] Overall Loss 0.259344 Objective Loss 0.259344 LR 0.001000 Time 0.021969 +2023-10-02 20:47:07,521 - Epoch: [38][ 740/ 1236] Overall Loss 0.259664 Objective Loss 0.259664 LR 0.001000 Time 0.021953 +2023-10-02 20:47:07,728 - Epoch: [38][ 750/ 1236] Overall Loss 0.259955 Objective Loss 0.259955 LR 0.001000 Time 0.021934 +2023-10-02 20:47:07,938 - Epoch: [38][ 760/ 1236] Overall Loss 0.260146 Objective Loss 0.260146 LR 0.001000 Time 0.021921 +2023-10-02 20:47:08,144 - Epoch: [38][ 770/ 1236] Overall Loss 0.260018 Objective Loss 0.260018 LR 0.001000 Time 0.021904 +2023-10-02 20:47:08,354 - Epoch: [38][ 780/ 1236] Overall Loss 0.260376 Objective Loss 0.260376 LR 0.001000 Time 0.021892 +2023-10-02 20:47:08,560 - Epoch: [38][ 790/ 1236] Overall Loss 0.260430 Objective Loss 0.260430 LR 0.001000 Time 0.021876 +2023-10-02 20:47:08,769 - Epoch: [38][ 800/ 1236] Overall Loss 0.260077 Objective Loss 0.260077 LR 0.001000 Time 0.021864 +2023-10-02 20:47:08,976 - Epoch: [38][ 810/ 1236] Overall Loss 0.260131 Objective Loss 0.260131 LR 0.001000 Time 0.021848 +2023-10-02 20:47:09,185 - Epoch: [38][ 820/ 1236] Overall Loss 0.260293 Objective Loss 0.260293 LR 0.001000 Time 0.021836 +2023-10-02 20:47:09,392 - Epoch: [38][ 830/ 1236] Overall Loss 0.260359 Objective Loss 0.260359 LR 0.001000 Time 0.021822 +2023-10-02 20:47:09,602 - Epoch: [38][ 840/ 1236] Overall Loss 0.260232 Objective Loss 0.260232 LR 0.001000 Time 0.021811 +2023-10-02 20:47:09,809 - Epoch: [38][ 850/ 1236] Overall Loss 0.260482 Objective Loss 0.260482 LR 0.001000 Time 0.021797 +2023-10-02 20:47:10,018 - Epoch: [38][ 860/ 1236] Overall Loss 0.260834 Objective Loss 0.260834 LR 0.001000 Time 0.021787 +2023-10-02 20:47:10,225 - Epoch: [38][ 870/ 1236] Overall Loss 0.261121 Objective Loss 0.261121 LR 0.001000 Time 0.021774 +2023-10-02 20:47:10,434 - Epoch: [38][ 880/ 1236] Overall Loss 0.261216 Objective Loss 0.261216 LR 0.001000 Time 0.021764 +2023-10-02 20:47:10,642 - Epoch: [38][ 890/ 1236] Overall Loss 0.261596 Objective Loss 0.261596 LR 0.001000 Time 0.021751 +2023-10-02 20:47:10,851 - Epoch: [38][ 900/ 1236] Overall Loss 0.261837 Objective Loss 0.261837 LR 0.001000 Time 0.021742 +2023-10-02 20:47:11,058 - Epoch: [38][ 910/ 1236] Overall Loss 0.261771 Objective Loss 0.261771 LR 0.001000 Time 0.021730 +2023-10-02 20:47:11,266 - Epoch: [38][ 920/ 1236] Overall Loss 0.261795 Objective Loss 0.261795 LR 0.001000 Time 0.021719 +2023-10-02 20:47:11,474 - Epoch: [38][ 930/ 1236] Overall Loss 0.262032 Objective Loss 0.262032 LR 0.001000 Time 0.021708 +2023-10-02 20:47:11,683 - Epoch: [38][ 940/ 1236] Overall Loss 0.262069 Objective Loss 0.262069 LR 0.001000 Time 0.021699 +2023-10-02 20:47:11,889 - Epoch: [38][ 950/ 1236] Overall Loss 0.262229 Objective Loss 0.262229 LR 0.001000 Time 0.021688 +2023-10-02 20:47:12,099 - Epoch: [38][ 960/ 1236] Overall Loss 0.262748 Objective Loss 0.262748 LR 0.001000 Time 0.021680 +2023-10-02 20:47:12,306 - Epoch: [38][ 970/ 1236] Overall Loss 0.262776 Objective Loss 0.262776 LR 0.001000 Time 0.021669 +2023-10-02 20:47:12,514 - Epoch: [38][ 980/ 1236] Overall Loss 0.263014 Objective Loss 0.263014 LR 0.001000 Time 0.021660 +2023-10-02 20:47:12,721 - Epoch: [38][ 990/ 1236] Overall Loss 0.263018 Objective Loss 0.263018 LR 0.001000 Time 0.021649 +2023-10-02 20:47:12,930 - Epoch: [38][ 1000/ 1236] Overall Loss 0.263391 Objective Loss 0.263391 LR 0.001000 Time 0.021641 +2023-10-02 20:47:13,137 - Epoch: [38][ 1010/ 1236] Overall Loss 0.263375 Objective Loss 0.263375 LR 0.001000 Time 0.021631 +2023-10-02 20:47:13,346 - Epoch: [38][ 1020/ 1236] Overall Loss 0.263305 Objective Loss 0.263305 LR 0.001000 Time 0.021623 +2023-10-02 20:47:13,553 - Epoch: [38][ 1030/ 1236] Overall Loss 0.263322 Objective Loss 0.263322 LR 0.001000 Time 0.021613 +2023-10-02 20:47:13,763 - Epoch: [38][ 1040/ 1236] Overall Loss 0.263631 Objective Loss 0.263631 LR 0.001000 Time 0.021607 +2023-10-02 20:47:13,969 - Epoch: [38][ 1050/ 1236] Overall Loss 0.263698 Objective Loss 0.263698 LR 0.001000 Time 0.021597 +2023-10-02 20:47:14,179 - Epoch: [38][ 1060/ 1236] Overall Loss 0.263589 Objective Loss 0.263589 LR 0.001000 Time 0.021591 +2023-10-02 20:47:14,385 - Epoch: [38][ 1070/ 1236] Overall Loss 0.263952 Objective Loss 0.263952 LR 0.001000 Time 0.021582 +2023-10-02 20:47:14,595 - Epoch: [38][ 1080/ 1236] Overall Loss 0.264265 Objective Loss 0.264265 LR 0.001000 Time 0.021575 +2023-10-02 20:47:14,801 - Epoch: [38][ 1090/ 1236] Overall Loss 0.264174 Objective Loss 0.264174 LR 0.001000 Time 0.021567 +2023-10-02 20:47:15,011 - Epoch: [38][ 1100/ 1236] Overall Loss 0.264777 Objective Loss 0.264777 LR 0.001000 Time 0.021561 +2023-10-02 20:47:15,217 - Epoch: [38][ 1110/ 1236] Overall Loss 0.264891 Objective Loss 0.264891 LR 0.001000 Time 0.021552 +2023-10-02 20:47:15,426 - Epoch: [38][ 1120/ 1236] Overall Loss 0.264771 Objective Loss 0.264771 LR 0.001000 Time 0.021546 +2023-10-02 20:47:15,633 - Epoch: [38][ 1130/ 1236] Overall Loss 0.264885 Objective Loss 0.264885 LR 0.001000 Time 0.021538 +2023-10-02 20:47:15,842 - Epoch: [38][ 1140/ 1236] Overall Loss 0.264872 Objective Loss 0.264872 LR 0.001000 Time 0.021531 +2023-10-02 20:47:16,049 - Epoch: [38][ 1150/ 1236] Overall Loss 0.264904 Objective Loss 0.264904 LR 0.001000 Time 0.021523 +2023-10-02 20:47:16,258 - Epoch: [38][ 1160/ 1236] Overall Loss 0.265406 Objective Loss 0.265406 LR 0.001000 Time 0.021517 +2023-10-02 20:47:16,465 - Epoch: [38][ 1170/ 1236] Overall Loss 0.265282 Objective Loss 0.265282 LR 0.001000 Time 0.021510 +2023-10-02 20:47:16,674 - Epoch: [38][ 1180/ 1236] Overall Loss 0.265128 Objective Loss 0.265128 LR 0.001000 Time 0.021504 +2023-10-02 20:47:16,882 - Epoch: [38][ 1190/ 1236] Overall Loss 0.265419 Objective Loss 0.265419 LR 0.001000 Time 0.021496 +2023-10-02 20:47:17,090 - Epoch: [38][ 1200/ 1236] Overall Loss 0.265541 Objective Loss 0.265541 LR 0.001000 Time 0.021490 +2023-10-02 20:47:17,298 - Epoch: [38][ 1210/ 1236] Overall Loss 0.265604 Objective Loss 0.265604 LR 0.001000 Time 0.021483 +2023-10-02 20:47:17,508 - Epoch: [38][ 1220/ 1236] Overall Loss 0.265414 Objective Loss 0.265414 LR 0.001000 Time 0.021479 +2023-10-02 20:47:17,768 - Epoch: [38][ 1230/ 1236] Overall Loss 0.265768 Objective Loss 0.265768 LR 0.001000 Time 0.021516 +2023-10-02 20:47:17,891 - Epoch: [38][ 1236/ 1236] Overall Loss 0.265883 Objective Loss 0.265883 Top1 85.947047 Top5 97.963340 LR 0.001000 Time 0.021511 +2023-10-02 20:47:18,021 - --- validate (epoch=38)----------- +2023-10-02 20:47:18,021 - 29943 samples (256 per mini-batch) +2023-10-02 20:47:18,503 - Epoch: [38][ 10/ 117] Loss 0.337668 Top1 82.617188 Top5 97.812500 +2023-10-02 20:47:18,657 - Epoch: [38][ 20/ 117] Loss 0.336302 Top1 83.046875 Top5 97.851562 +2023-10-02 20:47:18,807 - Epoch: [38][ 30/ 117] Loss 0.333724 Top1 83.072917 Top5 97.890625 +2023-10-02 20:47:18,959 - Epoch: [38][ 40/ 117] Loss 0.335833 Top1 82.968750 Top5 97.890625 +2023-10-02 20:47:19,109 - Epoch: [38][ 50/ 117] Loss 0.329022 Top1 82.945312 Top5 97.945312 +2023-10-02 20:47:19,261 - Epoch: [38][ 60/ 117] Loss 0.333490 Top1 82.994792 Top5 97.884115 +2023-10-02 20:47:19,410 - Epoch: [38][ 70/ 117] Loss 0.330655 Top1 82.918527 Top5 98.024554 +2023-10-02 20:47:19,560 - Epoch: [38][ 80/ 117] Loss 0.333722 Top1 82.822266 Top5 97.993164 +2023-10-02 20:47:19,708 - Epoch: [38][ 90/ 117] Loss 0.334901 Top1 82.738715 Top5 97.921007 +2023-10-02 20:47:19,859 - Epoch: [38][ 100/ 117] Loss 0.330535 Top1 82.742188 Top5 97.917969 +2023-10-02 20:47:20,016 - Epoch: [38][ 110/ 117] Loss 0.327842 Top1 82.642045 Top5 97.972301 +2023-10-02 20:47:20,105 - Epoch: [38][ 117/ 117] Loss 0.328542 Top1 82.663728 Top5 97.939418 +2023-10-02 20:47:20,242 - ==> Top1: 82.664 Top5: 97.939 Loss: 0.329 + +2023-10-02 20:47:20,242 - ==> Confusion: +[[ 922 2 12 0 10 2 0 1 7 57 3 2 1 1 14 3 4 1 0 0 8] + [ 0 1031 1 0 1 44 3 21 3 1 1 1 2 0 0 4 1 0 9 3 5] + [ 1 0 943 17 1 0 45 6 0 1 6 0 6 1 0 5 2 1 7 8 6] + [ 1 4 11 1006 0 2 1 3 4 0 4 1 2 6 15 1 1 1 11 1 14] + [ 20 10 3 1 953 9 1 1 0 9 3 1 2 3 12 2 8 0 0 5 7] + [ 0 36 0 2 2 993 2 16 1 4 3 10 6 12 8 1 1 1 2 6 10] + [ 0 0 24 0 0 0 1137 3 0 0 5 4 0 0 1 4 0 0 2 7 4] + [ 1 18 22 4 2 32 7 1046 1 2 4 7 2 3 3 0 2 1 43 11 7] + [ 17 5 0 0 3 4 0 0 955 45 12 2 2 9 19 0 4 3 7 1 1] + [ 110 3 0 1 12 1 1 0 36 896 0 0 0 20 24 1 1 0 0 4 9] + [ 1 6 10 9 2 2 5 6 14 0 968 2 1 8 5 1 0 3 2 3 5] + [ 1 0 0 0 1 16 0 3 0 1 2 937 26 7 0 1 1 13 2 19 5] + [ 0 1 1 5 1 10 1 2 1 0 6 52 942 3 1 3 6 13 1 7 12] + [ 0 0 0 0 2 13 1 1 12 13 9 6 0 1036 8 3 1 1 0 1 12] + [ 6 2 5 26 4 0 0 0 12 1 3 0 2 2 1015 0 1 1 7 0 14] + [ 0 0 0 2 2 1 2 0 0 2 0 12 4 1 0 1070 18 7 2 6 5] + [ 0 15 0 0 3 9 1 0 1 0 0 2 0 1 4 9 1099 1 0 8 8] + [ 0 2 0 15 0 0 3 0 1 1 0 6 28 0 5 5 1 965 0 2 4] + [ 1 9 7 26 1 1 0 19 4 0 1 1 0 0 3 0 1 0 985 2 7] + [ 0 1 2 3 0 5 10 5 0 0 2 10 2 0 0 1 4 1 0 1103 3] + [ 112 270 156 140 60 278 69 114 94 58 192 157 289 274 186 69 163 61 147 266 4750]] + +2023-10-02 20:47:20,244 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:47:20,244 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:47:20,250 - + +2023-10-02 20:47:20,250 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:47:21,268 - Epoch: [39][ 10/ 1236] Overall Loss 0.278624 Objective Loss 0.278624 LR 0.001000 Time 0.101805 +2023-10-02 20:47:21,476 - Epoch: [39][ 20/ 1236] Overall Loss 0.267102 Objective Loss 0.267102 LR 0.001000 Time 0.061292 +2023-10-02 20:47:21,685 - Epoch: [39][ 30/ 1236] Overall Loss 0.260951 Objective Loss 0.260951 LR 0.001000 Time 0.047764 +2023-10-02 20:47:21,896 - Epoch: [39][ 40/ 1236] Overall Loss 0.261118 Objective Loss 0.261118 LR 0.001000 Time 0.041080 +2023-10-02 20:47:22,102 - Epoch: [39][ 50/ 1236] Overall Loss 0.264760 Objective Loss 0.264760 LR 0.001000 Time 0.036988 +2023-10-02 20:47:22,313 - Epoch: [39][ 60/ 1236] Overall Loss 0.261787 Objective Loss 0.261787 LR 0.001000 Time 0.034329 +2023-10-02 20:47:22,520 - Epoch: [39][ 70/ 1236] Overall Loss 0.264311 Objective Loss 0.264311 LR 0.001000 Time 0.032383 +2023-10-02 20:47:22,732 - Epoch: [39][ 80/ 1236] Overall Loss 0.263368 Objective Loss 0.263368 LR 0.001000 Time 0.030977 +2023-10-02 20:47:22,938 - Epoch: [39][ 90/ 1236] Overall Loss 0.266805 Objective Loss 0.266805 LR 0.001000 Time 0.029826 +2023-10-02 20:47:23,145 - Epoch: [39][ 100/ 1236] Overall Loss 0.264112 Objective Loss 0.264112 LR 0.001000 Time 0.028907 +2023-10-02 20:47:23,351 - Epoch: [39][ 110/ 1236] Overall Loss 0.264390 Objective Loss 0.264390 LR 0.001000 Time 0.028144 +2023-10-02 20:47:23,561 - Epoch: [39][ 120/ 1236] Overall Loss 0.262731 Objective Loss 0.262731 LR 0.001000 Time 0.027542 +2023-10-02 20:47:23,771 - Epoch: [39][ 130/ 1236] Overall Loss 0.260846 Objective Loss 0.260846 LR 0.001000 Time 0.027030 +2023-10-02 20:47:23,981 - Epoch: [39][ 140/ 1236] Overall Loss 0.259253 Objective Loss 0.259253 LR 0.001000 Time 0.026599 +2023-10-02 20:47:24,189 - Epoch: [39][ 150/ 1236] Overall Loss 0.258164 Objective Loss 0.258164 LR 0.001000 Time 0.026210 +2023-10-02 20:47:24,400 - Epoch: [39][ 160/ 1236] Overall Loss 0.256885 Objective Loss 0.256885 LR 0.001000 Time 0.025890 +2023-10-02 20:47:24,608 - Epoch: [39][ 170/ 1236] Overall Loss 0.257563 Objective Loss 0.257563 LR 0.001000 Time 0.025586 +2023-10-02 20:47:24,818 - Epoch: [39][ 180/ 1236] Overall Loss 0.258492 Objective Loss 0.258492 LR 0.001000 Time 0.025323 +2023-10-02 20:47:25,027 - Epoch: [39][ 190/ 1236] Overall Loss 0.265539 Objective Loss 0.265539 LR 0.001000 Time 0.025084 +2023-10-02 20:47:25,237 - Epoch: [39][ 200/ 1236] Overall Loss 0.269781 Objective Loss 0.269781 LR 0.001000 Time 0.024877 +2023-10-02 20:47:25,446 - Epoch: [39][ 210/ 1236] Overall Loss 0.272082 Objective Loss 0.272082 LR 0.001000 Time 0.024682 +2023-10-02 20:47:25,656 - Epoch: [39][ 220/ 1236] Overall Loss 0.274951 Objective Loss 0.274951 LR 0.001000 Time 0.024514 +2023-10-02 20:47:25,865 - Epoch: [39][ 230/ 1236] Overall Loss 0.276785 Objective Loss 0.276785 LR 0.001000 Time 0.024350 +2023-10-02 20:47:26,075 - Epoch: [39][ 240/ 1236] Overall Loss 0.279077 Objective Loss 0.279077 LR 0.001000 Time 0.024209 +2023-10-02 20:47:26,284 - Epoch: [39][ 250/ 1236] Overall Loss 0.280416 Objective Loss 0.280416 LR 0.001000 Time 0.024071 +2023-10-02 20:47:26,494 - Epoch: [39][ 260/ 1236] Overall Loss 0.280465 Objective Loss 0.280465 LR 0.001000 Time 0.023952 +2023-10-02 20:47:26,703 - Epoch: [39][ 270/ 1236] Overall Loss 0.281246 Objective Loss 0.281246 LR 0.001000 Time 0.023832 +2023-10-02 20:47:26,913 - Epoch: [39][ 280/ 1236] Overall Loss 0.281543 Objective Loss 0.281543 LR 0.001000 Time 0.023730 +2023-10-02 20:47:27,122 - Epoch: [39][ 290/ 1236] Overall Loss 0.281776 Objective Loss 0.281776 LR 0.001000 Time 0.023628 +2023-10-02 20:47:27,332 - Epoch: [39][ 300/ 1236] Overall Loss 0.282462 Objective Loss 0.282462 LR 0.001000 Time 0.023540 +2023-10-02 20:47:27,541 - Epoch: [39][ 310/ 1236] Overall Loss 0.282826 Objective Loss 0.282826 LR 0.001000 Time 0.023449 +2023-10-02 20:47:27,751 - Epoch: [39][ 320/ 1236] Overall Loss 0.283395 Objective Loss 0.283395 LR 0.001000 Time 0.023368 +2023-10-02 20:47:27,960 - Epoch: [39][ 330/ 1236] Overall Loss 0.283868 Objective Loss 0.283868 LR 0.001000 Time 0.023289 +2023-10-02 20:47:28,171 - Epoch: [39][ 340/ 1236] Overall Loss 0.284086 Objective Loss 0.284086 LR 0.001000 Time 0.023225 +2023-10-02 20:47:28,378 - Epoch: [39][ 350/ 1236] Overall Loss 0.284424 Objective Loss 0.284424 LR 0.001000 Time 0.023151 +2023-10-02 20:47:28,587 - Epoch: [39][ 360/ 1236] Overall Loss 0.284271 Objective Loss 0.284271 LR 0.001000 Time 0.023087 +2023-10-02 20:47:28,795 - Epoch: [39][ 370/ 1236] Overall Loss 0.284774 Objective Loss 0.284774 LR 0.001000 Time 0.023023 +2023-10-02 20:47:29,008 - Epoch: [39][ 380/ 1236] Overall Loss 0.284878 Objective Loss 0.284878 LR 0.001000 Time 0.022976 +2023-10-02 20:47:29,220 - Epoch: [39][ 390/ 1236] Overall Loss 0.284946 Objective Loss 0.284946 LR 0.001000 Time 0.022929 +2023-10-02 20:47:29,433 - Epoch: [39][ 400/ 1236] Overall Loss 0.285039 Objective Loss 0.285039 LR 0.001000 Time 0.022888 +2023-10-02 20:47:29,645 - Epoch: [39][ 410/ 1236] Overall Loss 0.285023 Objective Loss 0.285023 LR 0.001000 Time 0.022846 +2023-10-02 20:47:29,859 - Epoch: [39][ 420/ 1236] Overall Loss 0.284611 Objective Loss 0.284611 LR 0.001000 Time 0.022810 +2023-10-02 20:47:30,071 - Epoch: [39][ 430/ 1236] Overall Loss 0.284521 Objective Loss 0.284521 LR 0.001000 Time 0.022772 +2023-10-02 20:47:30,284 - Epoch: [39][ 440/ 1236] Overall Loss 0.284315 Objective Loss 0.284315 LR 0.001000 Time 0.022738 +2023-10-02 20:47:30,497 - Epoch: [39][ 450/ 1236] Overall Loss 0.284621 Objective Loss 0.284621 LR 0.001000 Time 0.022704 +2023-10-02 20:47:30,710 - Epoch: [39][ 460/ 1236] Overall Loss 0.284799 Objective Loss 0.284799 LR 0.001000 Time 0.022673 +2023-10-02 20:47:30,922 - Epoch: [39][ 470/ 1236] Overall Loss 0.284578 Objective Loss 0.284578 LR 0.001000 Time 0.022641 +2023-10-02 20:47:31,135 - Epoch: [39][ 480/ 1236] Overall Loss 0.284053 Objective Loss 0.284053 LR 0.001000 Time 0.022613 +2023-10-02 20:47:31,347 - Epoch: [39][ 490/ 1236] Overall Loss 0.283866 Objective Loss 0.283866 LR 0.001000 Time 0.022583 +2023-10-02 20:47:31,560 - Epoch: [39][ 500/ 1236] Overall Loss 0.283432 Objective Loss 0.283432 LR 0.001000 Time 0.022557 +2023-10-02 20:47:31,772 - Epoch: [39][ 510/ 1236] Overall Loss 0.283316 Objective Loss 0.283316 LR 0.001000 Time 0.022530 +2023-10-02 20:47:31,986 - Epoch: [39][ 520/ 1236] Overall Loss 0.283294 Objective Loss 0.283294 LR 0.001000 Time 0.022506 +2023-10-02 20:47:32,198 - Epoch: [39][ 530/ 1236] Overall Loss 0.283465 Objective Loss 0.283465 LR 0.001000 Time 0.022482 +2023-10-02 20:47:32,411 - Epoch: [39][ 540/ 1236] Overall Loss 0.283860 Objective Loss 0.283860 LR 0.001000 Time 0.022460 +2023-10-02 20:47:32,623 - Epoch: [39][ 550/ 1236] Overall Loss 0.283647 Objective Loss 0.283647 LR 0.001000 Time 0.022436 +2023-10-02 20:47:32,836 - Epoch: [39][ 560/ 1236] Overall Loss 0.283404 Objective Loss 0.283404 LR 0.001000 Time 0.022415 +2023-10-02 20:47:33,048 - Epoch: [39][ 570/ 1236] Overall Loss 0.283194 Objective Loss 0.283194 LR 0.001000 Time 0.022393 +2023-10-02 20:47:33,261 - Epoch: [39][ 580/ 1236] Overall Loss 0.282914 Objective Loss 0.282914 LR 0.001000 Time 0.022374 +2023-10-02 20:47:33,473 - Epoch: [39][ 590/ 1236] Overall Loss 0.282621 Objective Loss 0.282621 LR 0.001000 Time 0.022353 +2023-10-02 20:47:33,687 - Epoch: [39][ 600/ 1236] Overall Loss 0.282848 Objective Loss 0.282848 LR 0.001000 Time 0.022335 +2023-10-02 20:47:33,898 - Epoch: [39][ 610/ 1236] Overall Loss 0.283056 Objective Loss 0.283056 LR 0.001000 Time 0.022316 +2023-10-02 20:47:34,114 - Epoch: [39][ 620/ 1236] Overall Loss 0.283058 Objective Loss 0.283058 LR 0.001000 Time 0.022302 +2023-10-02 20:47:34,326 - Epoch: [39][ 630/ 1236] Overall Loss 0.282783 Objective Loss 0.282783 LR 0.001000 Time 0.022284 +2023-10-02 20:47:34,539 - Epoch: [39][ 640/ 1236] Overall Loss 0.282614 Objective Loss 0.282614 LR 0.001000 Time 0.022269 +2023-10-02 20:47:34,751 - Epoch: [39][ 650/ 1236] Overall Loss 0.282306 Objective Loss 0.282306 LR 0.001000 Time 0.022252 +2023-10-02 20:47:34,965 - Epoch: [39][ 660/ 1236] Overall Loss 0.282072 Objective Loss 0.282072 LR 0.001000 Time 0.022238 +2023-10-02 20:47:35,177 - Epoch: [39][ 670/ 1236] Overall Loss 0.282007 Objective Loss 0.282007 LR 0.001000 Time 0.022222 +2023-10-02 20:47:35,390 - Epoch: [39][ 680/ 1236] Overall Loss 0.281826 Objective Loss 0.281826 LR 0.001000 Time 0.022209 +2023-10-02 20:47:35,602 - Epoch: [39][ 690/ 1236] Overall Loss 0.281295 Objective Loss 0.281295 LR 0.001000 Time 0.022193 +2023-10-02 20:47:35,815 - Epoch: [39][ 700/ 1236] Overall Loss 0.281143 Objective Loss 0.281143 LR 0.001000 Time 0.022180 +2023-10-02 20:47:36,028 - Epoch: [39][ 710/ 1236] Overall Loss 0.280727 Objective Loss 0.280727 LR 0.001000 Time 0.022166 +2023-10-02 20:47:36,241 - Epoch: [39][ 720/ 1236] Overall Loss 0.280585 Objective Loss 0.280585 LR 0.001000 Time 0.022154 +2023-10-02 20:47:36,453 - Epoch: [39][ 730/ 1236] Overall Loss 0.280423 Objective Loss 0.280423 LR 0.001000 Time 0.022140 +2023-10-02 20:47:36,666 - Epoch: [39][ 740/ 1236] Overall Loss 0.280811 Objective Loss 0.280811 LR 0.001000 Time 0.022129 +2023-10-02 20:47:36,878 - Epoch: [39][ 750/ 1236] Overall Loss 0.280878 Objective Loss 0.280878 LR 0.001000 Time 0.022116 +2023-10-02 20:47:37,092 - Epoch: [39][ 760/ 1236] Overall Loss 0.280654 Objective Loss 0.280654 LR 0.001000 Time 0.022105 +2023-10-02 20:47:37,303 - Epoch: [39][ 770/ 1236] Overall Loss 0.280499 Objective Loss 0.280499 LR 0.001000 Time 0.022093 +2023-10-02 20:47:37,517 - Epoch: [39][ 780/ 1236] Overall Loss 0.280301 Objective Loss 0.280301 LR 0.001000 Time 0.022083 +2023-10-02 20:47:37,729 - Epoch: [39][ 790/ 1236] Overall Loss 0.280188 Objective Loss 0.280188 LR 0.001000 Time 0.022072 +2023-10-02 20:47:37,942 - Epoch: [39][ 800/ 1236] Overall Loss 0.279939 Objective Loss 0.279939 LR 0.001000 Time 0.022062 +2023-10-02 20:47:38,155 - Epoch: [39][ 810/ 1236] Overall Loss 0.280032 Objective Loss 0.280032 LR 0.001000 Time 0.022052 +2023-10-02 20:47:38,368 - Epoch: [39][ 820/ 1236] Overall Loss 0.279877 Objective Loss 0.279877 LR 0.001000 Time 0.022042 +2023-10-02 20:47:38,581 - Epoch: [39][ 830/ 1236] Overall Loss 0.279645 Objective Loss 0.279645 LR 0.001000 Time 0.022032 +2023-10-02 20:47:38,794 - Epoch: [39][ 840/ 1236] Overall Loss 0.279533 Objective Loss 0.279533 LR 0.001000 Time 0.022023 +2023-10-02 20:47:39,006 - Epoch: [39][ 850/ 1236] Overall Loss 0.279315 Objective Loss 0.279315 LR 0.001000 Time 0.022014 +2023-10-02 20:47:39,220 - Epoch: [39][ 860/ 1236] Overall Loss 0.279346 Objective Loss 0.279346 LR 0.001000 Time 0.022005 +2023-10-02 20:47:39,433 - Epoch: [39][ 870/ 1236] Overall Loss 0.279253 Objective Loss 0.279253 LR 0.001000 Time 0.021997 +2023-10-02 20:47:39,648 - Epoch: [39][ 880/ 1236] Overall Loss 0.279097 Objective Loss 0.279097 LR 0.001000 Time 0.021991 +2023-10-02 20:47:39,870 - Epoch: [39][ 890/ 1236] Overall Loss 0.279121 Objective Loss 0.279121 LR 0.001000 Time 0.021993 +2023-10-02 20:47:40,097 - Epoch: [39][ 900/ 1236] Overall Loss 0.278962 Objective Loss 0.278962 LR 0.001000 Time 0.022000 +2023-10-02 20:47:40,320 - Epoch: [39][ 910/ 1236] Overall Loss 0.278984 Objective Loss 0.278984 LR 0.001000 Time 0.022003 +2023-10-02 20:47:40,547 - Epoch: [39][ 920/ 1236] Overall Loss 0.278776 Objective Loss 0.278776 LR 0.001000 Time 0.022010 +2023-10-02 20:47:40,767 - Epoch: [39][ 930/ 1236] Overall Loss 0.278846 Objective Loss 0.278846 LR 0.001000 Time 0.022010 +2023-10-02 20:47:40,993 - Epoch: [39][ 940/ 1236] Overall Loss 0.278979 Objective Loss 0.278979 LR 0.001000 Time 0.022015 +2023-10-02 20:47:41,212 - Epoch: [39][ 950/ 1236] Overall Loss 0.279009 Objective Loss 0.279009 LR 0.001000 Time 0.022014 +2023-10-02 20:47:41,438 - Epoch: [39][ 960/ 1236] Overall Loss 0.278641 Objective Loss 0.278641 LR 0.001000 Time 0.022019 +2023-10-02 20:47:41,658 - Epoch: [39][ 970/ 1236] Overall Loss 0.278416 Objective Loss 0.278416 LR 0.001000 Time 0.022019 +2023-10-02 20:47:41,883 - Epoch: [39][ 980/ 1236] Overall Loss 0.278272 Objective Loss 0.278272 LR 0.001000 Time 0.022023 +2023-10-02 20:47:42,103 - Epoch: [39][ 990/ 1236] Overall Loss 0.278161 Objective Loss 0.278161 LR 0.001000 Time 0.022022 +2023-10-02 20:47:42,328 - Epoch: [39][ 1000/ 1236] Overall Loss 0.278207 Objective Loss 0.278207 LR 0.001000 Time 0.022026 +2023-10-02 20:47:42,540 - Epoch: [39][ 1010/ 1236] Overall Loss 0.278286 Objective Loss 0.278286 LR 0.001000 Time 0.022018 +2023-10-02 20:47:42,754 - Epoch: [39][ 1020/ 1236] Overall Loss 0.278184 Objective Loss 0.278184 LR 0.001000 Time 0.022011 +2023-10-02 20:47:42,966 - Epoch: [39][ 1030/ 1236] Overall Loss 0.277886 Objective Loss 0.277886 LR 0.001000 Time 0.022003 +2023-10-02 20:47:43,180 - Epoch: [39][ 1040/ 1236] Overall Loss 0.277955 Objective Loss 0.277955 LR 0.001000 Time 0.021997 +2023-10-02 20:47:43,392 - Epoch: [39][ 1050/ 1236] Overall Loss 0.277864 Objective Loss 0.277864 LR 0.001000 Time 0.021989 +2023-10-02 20:47:43,606 - Epoch: [39][ 1060/ 1236] Overall Loss 0.277937 Objective Loss 0.277937 LR 0.001000 Time 0.021983 +2023-10-02 20:47:43,818 - Epoch: [39][ 1070/ 1236] Overall Loss 0.277732 Objective Loss 0.277732 LR 0.001000 Time 0.021976 +2023-10-02 20:47:44,031 - Epoch: [39][ 1080/ 1236] Overall Loss 0.277650 Objective Loss 0.277650 LR 0.001000 Time 0.021970 +2023-10-02 20:47:44,244 - Epoch: [39][ 1090/ 1236] Overall Loss 0.277455 Objective Loss 0.277455 LR 0.001000 Time 0.021962 +2023-10-02 20:47:44,458 - Epoch: [39][ 1100/ 1236] Overall Loss 0.277346 Objective Loss 0.277346 LR 0.001000 Time 0.021957 +2023-10-02 20:47:44,670 - Epoch: [39][ 1110/ 1236] Overall Loss 0.277179 Objective Loss 0.277179 LR 0.001000 Time 0.021950 +2023-10-02 20:47:44,883 - Epoch: [39][ 1120/ 1236] Overall Loss 0.276763 Objective Loss 0.276763 LR 0.001000 Time 0.021944 +2023-10-02 20:47:45,096 - Epoch: [39][ 1130/ 1236] Overall Loss 0.276540 Objective Loss 0.276540 LR 0.001000 Time 0.021938 +2023-10-02 20:47:45,309 - Epoch: [39][ 1140/ 1236] Overall Loss 0.276220 Objective Loss 0.276220 LR 0.001000 Time 0.021932 +2023-10-02 20:47:45,521 - Epoch: [39][ 1150/ 1236] Overall Loss 0.276541 Objective Loss 0.276541 LR 0.001000 Time 0.021926 +2023-10-02 20:47:45,735 - Epoch: [39][ 1160/ 1236] Overall Loss 0.276349 Objective Loss 0.276349 LR 0.001000 Time 0.021921 +2023-10-02 20:47:45,948 - Epoch: [39][ 1170/ 1236] Overall Loss 0.276280 Objective Loss 0.276280 LR 0.001000 Time 0.021915 +2023-10-02 20:47:46,161 - Epoch: [39][ 1180/ 1236] Overall Loss 0.276067 Objective Loss 0.276067 LR 0.001000 Time 0.021910 +2023-10-02 20:47:46,373 - Epoch: [39][ 1190/ 1236] Overall Loss 0.276374 Objective Loss 0.276374 LR 0.001000 Time 0.021904 +2023-10-02 20:47:46,587 - Epoch: [39][ 1200/ 1236] Overall Loss 0.276336 Objective Loss 0.276336 LR 0.001000 Time 0.021899 +2023-10-02 20:47:46,801 - Epoch: [39][ 1210/ 1236] Overall Loss 0.276417 Objective Loss 0.276417 LR 0.001000 Time 0.021894 +2023-10-02 20:47:47,016 - Epoch: [39][ 1220/ 1236] Overall Loss 0.276272 Objective Loss 0.276272 LR 0.001000 Time 0.021890 +2023-10-02 20:47:47,282 - Epoch: [39][ 1230/ 1236] Overall Loss 0.276270 Objective Loss 0.276270 LR 0.001000 Time 0.021929 +2023-10-02 20:47:47,405 - Epoch: [39][ 1236/ 1236] Overall Loss 0.276254 Objective Loss 0.276254 Top1 84.521385 Top5 98.574338 LR 0.001000 Time 0.021922 +2023-10-02 20:47:47,549 - --- validate (epoch=39)----------- +2023-10-02 20:47:47,550 - 29943 samples (256 per mini-batch) +2023-10-02 20:47:48,039 - Epoch: [39][ 10/ 117] Loss 0.290228 Top1 84.218750 Top5 98.554688 +2023-10-02 20:47:48,201 - Epoch: [39][ 20/ 117] Loss 0.300858 Top1 84.179688 Top5 98.691406 +2023-10-02 20:47:48,363 - Epoch: [39][ 30/ 117] Loss 0.299663 Top1 84.205729 Top5 98.502604 +2023-10-02 20:47:48,525 - Epoch: [39][ 40/ 117] Loss 0.302759 Top1 84.042969 Top5 98.427734 +2023-10-02 20:47:48,686 - Epoch: [39][ 50/ 117] Loss 0.305374 Top1 83.960938 Top5 98.367188 +2023-10-02 20:47:48,844 - Epoch: [39][ 60/ 117] Loss 0.306892 Top1 83.860677 Top5 98.333333 +2023-10-02 20:47:49,000 - Epoch: [39][ 70/ 117] Loss 0.306490 Top1 84.040179 Top5 98.320312 +2023-10-02 20:47:49,150 - Epoch: [39][ 80/ 117] Loss 0.307716 Top1 83.984375 Top5 98.330078 +2023-10-02 20:47:49,300 - Epoch: [39][ 90/ 117] Loss 0.305137 Top1 83.940972 Top5 98.355035 +2023-10-02 20:47:49,451 - Epoch: [39][ 100/ 117] Loss 0.305554 Top1 83.976562 Top5 98.347656 +2023-10-02 20:47:49,608 - Epoch: [39][ 110/ 117] Loss 0.307196 Top1 83.973722 Top5 98.352273 +2023-10-02 20:47:49,699 - Epoch: [39][ 117/ 117] Loss 0.305995 Top1 83.956183 Top5 98.380256 +2023-10-02 20:47:49,837 - ==> Top1: 83.956 Top5: 98.380 Loss: 0.306 + +2023-10-02 20:47:49,837 - ==> Confusion: +[[ 958 3 2 0 10 4 0 0 5 45 1 0 0 2 1 2 0 1 0 0 16] + [ 1 1058 0 3 3 31 1 16 2 0 0 1 1 0 0 7 1 0 4 1 1] + [ 6 0 971 11 5 0 19 7 0 1 8 2 5 2 0 2 0 1 2 6 8] + [ 5 3 15 1008 0 1 1 1 4 0 1 0 3 2 19 2 0 0 10 1 13] + [ 32 5 2 0 966 10 0 1 0 8 0 3 1 2 8 3 5 1 1 1 1] + [ 4 33 1 2 0 1009 0 17 2 3 2 4 2 14 3 0 1 0 5 4 10] + [ 2 3 25 0 0 2 1129 5 0 0 2 3 2 0 0 2 0 1 3 8 4] + [ 3 14 19 0 2 35 5 1058 1 3 2 4 10 2 1 1 2 0 37 15 4] + [ 24 1 0 0 1 3 0 0 958 59 8 2 6 9 13 1 0 2 1 0 1] + [ 140 1 1 1 6 0 0 0 11 924 0 0 0 21 3 0 1 0 1 3 6] + [ 8 3 13 13 1 0 2 2 14 1 959 2 3 15 0 0 0 1 6 2 8] + [ 1 0 2 0 0 16 0 2 0 0 0 956 16 8 0 3 1 14 0 8 8] + [ 2 2 4 2 1 4 2 1 0 0 0 44 962 1 1 8 1 18 0 5 10] + [ 1 0 0 0 4 15 0 0 10 13 6 6 1 1044 4 1 0 1 0 1 12] + [ 16 3 3 27 6 2 0 0 17 5 1 0 5 4 995 0 1 2 5 0 9] + [ 0 0 3 1 5 0 0 0 0 0 0 11 5 2 0 1068 14 13 2 4 6] + [ 1 15 3 0 11 5 0 0 1 0 0 5 1 3 5 10 1080 1 0 5 15] + [ 1 0 1 5 0 0 1 0 1 0 0 5 21 0 1 7 0 989 0 3 3] + [ 4 7 6 22 1 0 0 17 4 0 5 0 4 0 11 0 0 0 977 2 8] + [ 0 2 3 1 1 5 6 8 0 0 1 14 3 4 0 1 3 0 2 1094 4] + [ 234 177 143 114 109 236 45 90 98 92 167 133 316 277 121 58 68 67 131 253 4976]] + +2023-10-02 20:47:49,839 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:47:49,839 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:47:49,845 - + +2023-10-02 20:47:49,845 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:47:50,982 - Epoch: [40][ 10/ 1236] Overall Loss 0.237735 Objective Loss 0.237735 LR 0.001000 Time 0.113681 +2023-10-02 20:47:51,193 - Epoch: [40][ 20/ 1236] Overall Loss 0.243155 Objective Loss 0.243155 LR 0.001000 Time 0.067371 +2023-10-02 20:47:51,403 - Epoch: [40][ 30/ 1236] Overall Loss 0.246887 Objective Loss 0.246887 LR 0.001000 Time 0.051843 +2023-10-02 20:47:51,613 - Epoch: [40][ 40/ 1236] Overall Loss 0.247347 Objective Loss 0.247347 LR 0.001000 Time 0.044137 +2023-10-02 20:47:51,822 - Epoch: [40][ 50/ 1236] Overall Loss 0.243266 Objective Loss 0.243266 LR 0.001000 Time 0.039459 +2023-10-02 20:47:52,034 - Epoch: [40][ 60/ 1236] Overall Loss 0.245125 Objective Loss 0.245125 LR 0.001000 Time 0.036412 +2023-10-02 20:47:52,241 - Epoch: [40][ 70/ 1236] Overall Loss 0.242796 Objective Loss 0.242796 LR 0.001000 Time 0.034168 +2023-10-02 20:47:52,456 - Epoch: [40][ 80/ 1236] Overall Loss 0.247200 Objective Loss 0.247200 LR 0.001000 Time 0.032570 +2023-10-02 20:47:52,671 - Epoch: [40][ 90/ 1236] Overall Loss 0.251074 Objective Loss 0.251074 LR 0.001000 Time 0.031344 +2023-10-02 20:47:52,893 - Epoch: [40][ 100/ 1236] Overall Loss 0.251544 Objective Loss 0.251544 LR 0.001000 Time 0.030421 +2023-10-02 20:47:53,108 - Epoch: [40][ 110/ 1236] Overall Loss 0.253163 Objective Loss 0.253163 LR 0.001000 Time 0.029613 +2023-10-02 20:47:53,329 - Epoch: [40][ 120/ 1236] Overall Loss 0.250712 Objective Loss 0.250712 LR 0.001000 Time 0.028983 +2023-10-02 20:47:53,544 - Epoch: [40][ 130/ 1236] Overall Loss 0.251491 Objective Loss 0.251491 LR 0.001000 Time 0.028404 +2023-10-02 20:47:53,765 - Epoch: [40][ 140/ 1236] Overall Loss 0.253511 Objective Loss 0.253511 LR 0.001000 Time 0.027951 +2023-10-02 20:47:53,980 - Epoch: [40][ 150/ 1236] Overall Loss 0.252875 Objective Loss 0.252875 LR 0.001000 Time 0.027518 +2023-10-02 20:47:54,201 - Epoch: [40][ 160/ 1236] Overall Loss 0.254376 Objective Loss 0.254376 LR 0.001000 Time 0.027177 +2023-10-02 20:47:54,417 - Epoch: [40][ 170/ 1236] Overall Loss 0.253380 Objective Loss 0.253380 LR 0.001000 Time 0.026843 +2023-10-02 20:47:54,626 - Epoch: [40][ 180/ 1236] Overall Loss 0.253827 Objective Loss 0.253827 LR 0.001000 Time 0.026515 +2023-10-02 20:47:54,835 - Epoch: [40][ 190/ 1236] Overall Loss 0.253356 Objective Loss 0.253356 LR 0.001000 Time 0.026210 +2023-10-02 20:47:55,044 - Epoch: [40][ 200/ 1236] Overall Loss 0.253232 Objective Loss 0.253232 LR 0.001000 Time 0.025946 +2023-10-02 20:47:55,255 - Epoch: [40][ 210/ 1236] Overall Loss 0.253269 Objective Loss 0.253269 LR 0.001000 Time 0.025704 +2023-10-02 20:47:55,465 - Epoch: [40][ 220/ 1236] Overall Loss 0.253704 Objective Loss 0.253704 LR 0.001000 Time 0.025492 +2023-10-02 20:47:55,674 - Epoch: [40][ 230/ 1236] Overall Loss 0.255503 Objective Loss 0.255503 LR 0.001000 Time 0.025287 +2023-10-02 20:47:55,884 - Epoch: [40][ 240/ 1236] Overall Loss 0.255406 Objective Loss 0.255406 LR 0.001000 Time 0.025106 +2023-10-02 20:47:56,093 - Epoch: [40][ 250/ 1236] Overall Loss 0.254983 Objective Loss 0.254983 LR 0.001000 Time 0.024932 +2023-10-02 20:47:56,303 - Epoch: [40][ 260/ 1236] Overall Loss 0.254617 Objective Loss 0.254617 LR 0.001000 Time 0.024778 +2023-10-02 20:47:56,512 - Epoch: [40][ 270/ 1236] Overall Loss 0.254712 Objective Loss 0.254712 LR 0.001000 Time 0.024629 +2023-10-02 20:47:56,721 - Epoch: [40][ 280/ 1236] Overall Loss 0.255454 Objective Loss 0.255454 LR 0.001000 Time 0.024497 +2023-10-02 20:47:56,930 - Epoch: [40][ 290/ 1236] Overall Loss 0.255647 Objective Loss 0.255647 LR 0.001000 Time 0.024366 +2023-10-02 20:47:57,140 - Epoch: [40][ 300/ 1236] Overall Loss 0.255949 Objective Loss 0.255949 LR 0.001000 Time 0.024252 +2023-10-02 20:47:57,348 - Epoch: [40][ 310/ 1236] Overall Loss 0.256638 Objective Loss 0.256638 LR 0.001000 Time 0.024138 +2023-10-02 20:47:57,558 - Epoch: [40][ 320/ 1236] Overall Loss 0.257001 Objective Loss 0.257001 LR 0.001000 Time 0.024039 +2023-10-02 20:47:57,767 - Epoch: [40][ 330/ 1236] Overall Loss 0.256811 Objective Loss 0.256811 LR 0.001000 Time 0.023940 +2023-10-02 20:47:57,977 - Epoch: [40][ 340/ 1236] Overall Loss 0.256287 Objective Loss 0.256287 LR 0.001000 Time 0.023852 +2023-10-02 20:47:58,187 - Epoch: [40][ 350/ 1236] Overall Loss 0.256256 Objective Loss 0.256256 LR 0.001000 Time 0.023764 +2023-10-02 20:47:58,397 - Epoch: [40][ 360/ 1236] Overall Loss 0.256850 Objective Loss 0.256850 LR 0.001000 Time 0.023687 +2023-10-02 20:47:58,607 - Epoch: [40][ 370/ 1236] Overall Loss 0.257129 Objective Loss 0.257129 LR 0.001000 Time 0.023611 +2023-10-02 20:47:58,817 - Epoch: [40][ 380/ 1236] Overall Loss 0.257316 Objective Loss 0.257316 LR 0.001000 Time 0.023542 +2023-10-02 20:47:59,027 - Epoch: [40][ 390/ 1236] Overall Loss 0.257114 Objective Loss 0.257114 LR 0.001000 Time 0.023473 +2023-10-02 20:47:59,236 - Epoch: [40][ 400/ 1236] Overall Loss 0.257888 Objective Loss 0.257888 LR 0.001000 Time 0.023408 +2023-10-02 20:47:59,447 - Epoch: [40][ 410/ 1236] Overall Loss 0.258273 Objective Loss 0.258273 LR 0.001000 Time 0.023349 +2023-10-02 20:47:59,657 - Epoch: [40][ 420/ 1236] Overall Loss 0.258331 Objective Loss 0.258331 LR 0.001000 Time 0.023294 +2023-10-02 20:47:59,867 - Epoch: [40][ 430/ 1236] Overall Loss 0.258386 Objective Loss 0.258386 LR 0.001000 Time 0.023237 +2023-10-02 20:48:00,076 - Epoch: [40][ 440/ 1236] Overall Loss 0.258430 Objective Loss 0.258430 LR 0.001000 Time 0.023184 +2023-10-02 20:48:00,288 - Epoch: [40][ 450/ 1236] Overall Loss 0.258458 Objective Loss 0.258458 LR 0.001000 Time 0.023138 +2023-10-02 20:48:00,501 - Epoch: [40][ 460/ 1236] Overall Loss 0.258715 Objective Loss 0.258715 LR 0.001000 Time 0.023098 +2023-10-02 20:48:00,715 - Epoch: [40][ 470/ 1236] Overall Loss 0.259025 Objective Loss 0.259025 LR 0.001000 Time 0.023061 +2023-10-02 20:48:00,928 - Epoch: [40][ 480/ 1236] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.001000 Time 0.023025 +2023-10-02 20:48:01,142 - Epoch: [40][ 490/ 1236] Overall Loss 0.258514 Objective Loss 0.258514 LR 0.001000 Time 0.022991 +2023-10-02 20:48:01,356 - Epoch: [40][ 500/ 1236] Overall Loss 0.258336 Objective Loss 0.258336 LR 0.001000 Time 0.022957 +2023-10-02 20:48:01,570 - Epoch: [40][ 510/ 1236] Overall Loss 0.258434 Objective Loss 0.258434 LR 0.001000 Time 0.022923 +2023-10-02 20:48:01,783 - Epoch: [40][ 520/ 1236] Overall Loss 0.258361 Objective Loss 0.258361 LR 0.001000 Time 0.022893 +2023-10-02 20:48:01,997 - Epoch: [40][ 530/ 1236] Overall Loss 0.258524 Objective Loss 0.258524 LR 0.001000 Time 0.022864 +2023-10-02 20:48:02,210 - Epoch: [40][ 540/ 1236] Overall Loss 0.258549 Objective Loss 0.258549 LR 0.001000 Time 0.022835 +2023-10-02 20:48:02,424 - Epoch: [40][ 550/ 1236] Overall Loss 0.258508 Objective Loss 0.258508 LR 0.001000 Time 0.022806 +2023-10-02 20:48:02,638 - Epoch: [40][ 560/ 1236] Overall Loss 0.259243 Objective Loss 0.259243 LR 0.001000 Time 0.022779 +2023-10-02 20:48:02,852 - Epoch: [40][ 570/ 1236] Overall Loss 0.259555 Objective Loss 0.259555 LR 0.001000 Time 0.022754 +2023-10-02 20:48:03,065 - Epoch: [40][ 580/ 1236] Overall Loss 0.259012 Objective Loss 0.259012 LR 0.001000 Time 0.022729 +2023-10-02 20:48:03,279 - Epoch: [40][ 590/ 1236] Overall Loss 0.259119 Objective Loss 0.259119 LR 0.001000 Time 0.022706 +2023-10-02 20:48:03,492 - Epoch: [40][ 600/ 1236] Overall Loss 0.259419 Objective Loss 0.259419 LR 0.001000 Time 0.022682 +2023-10-02 20:48:03,706 - Epoch: [40][ 610/ 1236] Overall Loss 0.259759 Objective Loss 0.259759 LR 0.001000 Time 0.022660 +2023-10-02 20:48:03,919 - Epoch: [40][ 620/ 1236] Overall Loss 0.259848 Objective Loss 0.259848 LR 0.001000 Time 0.022638 +2023-10-02 20:48:04,133 - Epoch: [40][ 630/ 1236] Overall Loss 0.259904 Objective Loss 0.259904 LR 0.001000 Time 0.022618 +2023-10-02 20:48:04,346 - Epoch: [40][ 640/ 1236] Overall Loss 0.259754 Objective Loss 0.259754 LR 0.001000 Time 0.022598 +2023-10-02 20:48:04,560 - Epoch: [40][ 650/ 1236] Overall Loss 0.260311 Objective Loss 0.260311 LR 0.001000 Time 0.022579 +2023-10-02 20:48:04,774 - Epoch: [40][ 660/ 1236] Overall Loss 0.260277 Objective Loss 0.260277 LR 0.001000 Time 0.022560 +2023-10-02 20:48:04,987 - Epoch: [40][ 670/ 1236] Overall Loss 0.260254 Objective Loss 0.260254 LR 0.001000 Time 0.022540 +2023-10-02 20:48:05,201 - Epoch: [40][ 680/ 1236] Overall Loss 0.260372 Objective Loss 0.260372 LR 0.001000 Time 0.022522 +2023-10-02 20:48:05,415 - Epoch: [40][ 690/ 1236] Overall Loss 0.260748 Objective Loss 0.260748 LR 0.001000 Time 0.022503 +2023-10-02 20:48:05,628 - Epoch: [40][ 700/ 1236] Overall Loss 0.260890 Objective Loss 0.260890 LR 0.001000 Time 0.022486 +2023-10-02 20:48:05,842 - Epoch: [40][ 710/ 1236] Overall Loss 0.261004 Objective Loss 0.261004 LR 0.001000 Time 0.022471 +2023-10-02 20:48:06,056 - Epoch: [40][ 720/ 1236] Overall Loss 0.260992 Objective Loss 0.260992 LR 0.001000 Time 0.022454 +2023-10-02 20:48:06,270 - Epoch: [40][ 730/ 1236] Overall Loss 0.261165 Objective Loss 0.261165 LR 0.001000 Time 0.022440 +2023-10-02 20:48:06,483 - Epoch: [40][ 740/ 1236] Overall Loss 0.261457 Objective Loss 0.261457 LR 0.001000 Time 0.022425 +2023-10-02 20:48:06,697 - Epoch: [40][ 750/ 1236] Overall Loss 0.261268 Objective Loss 0.261268 LR 0.001000 Time 0.022410 +2023-10-02 20:48:06,910 - Epoch: [40][ 760/ 1236] Overall Loss 0.261251 Objective Loss 0.261251 LR 0.001000 Time 0.022396 +2023-10-02 20:48:07,124 - Epoch: [40][ 770/ 1236] Overall Loss 0.261065 Objective Loss 0.261065 LR 0.001000 Time 0.022383 +2023-10-02 20:48:07,338 - Epoch: [40][ 780/ 1236] Overall Loss 0.261228 Objective Loss 0.261228 LR 0.001000 Time 0.022369 +2023-10-02 20:48:07,552 - Epoch: [40][ 790/ 1236] Overall Loss 0.261015 Objective Loss 0.261015 LR 0.001000 Time 0.022356 +2023-10-02 20:48:07,765 - Epoch: [40][ 800/ 1236] Overall Loss 0.261050 Objective Loss 0.261050 LR 0.001000 Time 0.022343 +2023-10-02 20:48:07,979 - Epoch: [40][ 810/ 1236] Overall Loss 0.260871 Objective Loss 0.260871 LR 0.001000 Time 0.022329 +2023-10-02 20:48:08,192 - Epoch: [40][ 820/ 1236] Overall Loss 0.260780 Objective Loss 0.260780 LR 0.001000 Time 0.022317 +2023-10-02 20:48:08,406 - Epoch: [40][ 830/ 1236] Overall Loss 0.260895 Objective Loss 0.260895 LR 0.001000 Time 0.022305 +2023-10-02 20:48:08,620 - Epoch: [40][ 840/ 1236] Overall Loss 0.260827 Objective Loss 0.260827 LR 0.001000 Time 0.022293 +2023-10-02 20:48:08,833 - Epoch: [40][ 850/ 1236] Overall Loss 0.260945 Objective Loss 0.260945 LR 0.001000 Time 0.022282 +2023-10-02 20:48:09,047 - Epoch: [40][ 860/ 1236] Overall Loss 0.260670 Objective Loss 0.260670 LR 0.001000 Time 0.022271 +2023-10-02 20:48:09,261 - Epoch: [40][ 870/ 1236] Overall Loss 0.260785 Objective Loss 0.260785 LR 0.001000 Time 0.022261 +2023-10-02 20:48:09,475 - Epoch: [40][ 880/ 1236] Overall Loss 0.260819 Objective Loss 0.260819 LR 0.001000 Time 0.022250 +2023-10-02 20:48:09,689 - Epoch: [40][ 890/ 1236] Overall Loss 0.260999 Objective Loss 0.260999 LR 0.001000 Time 0.022240 +2023-10-02 20:48:09,902 - Epoch: [40][ 900/ 1236] Overall Loss 0.260772 Objective Loss 0.260772 LR 0.001000 Time 0.022230 +2023-10-02 20:48:10,114 - Epoch: [40][ 910/ 1236] Overall Loss 0.261065 Objective Loss 0.261065 LR 0.001000 Time 0.022218 +2023-10-02 20:48:10,325 - Epoch: [40][ 920/ 1236] Overall Loss 0.261134 Objective Loss 0.261134 LR 0.001000 Time 0.022206 +2023-10-02 20:48:10,536 - Epoch: [40][ 930/ 1236] Overall Loss 0.261110 Objective Loss 0.261110 LR 0.001000 Time 0.022194 +2023-10-02 20:48:10,748 - Epoch: [40][ 940/ 1236] Overall Loss 0.261177 Objective Loss 0.261177 LR 0.001000 Time 0.022183 +2023-10-02 20:48:10,959 - Epoch: [40][ 950/ 1236] Overall Loss 0.261085 Objective Loss 0.261085 LR 0.001000 Time 0.022171 +2023-10-02 20:48:11,171 - Epoch: [40][ 960/ 1236] Overall Loss 0.261357 Objective Loss 0.261357 LR 0.001000 Time 0.022160 +2023-10-02 20:48:11,382 - Epoch: [40][ 970/ 1236] Overall Loss 0.261324 Objective Loss 0.261324 LR 0.001000 Time 0.022149 +2023-10-02 20:48:11,594 - Epoch: [40][ 980/ 1236] Overall Loss 0.261676 Objective Loss 0.261676 LR 0.001000 Time 0.022139 +2023-10-02 20:48:11,805 - Epoch: [40][ 990/ 1236] Overall Loss 0.262042 Objective Loss 0.262042 LR 0.001000 Time 0.022129 +2023-10-02 20:48:12,017 - Epoch: [40][ 1000/ 1236] Overall Loss 0.261948 Objective Loss 0.261948 LR 0.001000 Time 0.022118 +2023-10-02 20:48:12,227 - Epoch: [40][ 1010/ 1236] Overall Loss 0.261963 Objective Loss 0.261963 LR 0.001000 Time 0.022108 +2023-10-02 20:48:12,439 - Epoch: [40][ 1020/ 1236] Overall Loss 0.262094 Objective Loss 0.262094 LR 0.001000 Time 0.022098 +2023-10-02 20:48:12,650 - Epoch: [40][ 1030/ 1236] Overall Loss 0.262176 Objective Loss 0.262176 LR 0.001000 Time 0.022089 +2023-10-02 20:48:12,862 - Epoch: [40][ 1040/ 1236] Overall Loss 0.262204 Objective Loss 0.262204 LR 0.001000 Time 0.022079 +2023-10-02 20:48:13,073 - Epoch: [40][ 1050/ 1236] Overall Loss 0.262212 Objective Loss 0.262212 LR 0.001000 Time 0.022070 +2023-10-02 20:48:13,284 - Epoch: [40][ 1060/ 1236] Overall Loss 0.262046 Objective Loss 0.262046 LR 0.001000 Time 0.022060 +2023-10-02 20:48:13,495 - Epoch: [40][ 1070/ 1236] Overall Loss 0.262133 Objective Loss 0.262133 LR 0.001000 Time 0.022050 +2023-10-02 20:48:13,707 - Epoch: [40][ 1080/ 1236] Overall Loss 0.262121 Objective Loss 0.262121 LR 0.001000 Time 0.022042 +2023-10-02 20:48:13,918 - Epoch: [40][ 1090/ 1236] Overall Loss 0.262001 Objective Loss 0.262001 LR 0.001000 Time 0.022033 +2023-10-02 20:48:14,129 - Epoch: [40][ 1100/ 1236] Overall Loss 0.261984 Objective Loss 0.261984 LR 0.001000 Time 0.022024 +2023-10-02 20:48:14,340 - Epoch: [40][ 1110/ 1236] Overall Loss 0.261842 Objective Loss 0.261842 LR 0.001000 Time 0.022016 +2023-10-02 20:48:14,552 - Epoch: [40][ 1120/ 1236] Overall Loss 0.261744 Objective Loss 0.261744 LR 0.001000 Time 0.022008 +2023-10-02 20:48:14,763 - Epoch: [40][ 1130/ 1236] Overall Loss 0.261754 Objective Loss 0.261754 LR 0.001000 Time 0.022000 +2023-10-02 20:48:14,975 - Epoch: [40][ 1140/ 1236] Overall Loss 0.261954 Objective Loss 0.261954 LR 0.001000 Time 0.021993 +2023-10-02 20:48:15,186 - Epoch: [40][ 1150/ 1236] Overall Loss 0.262063 Objective Loss 0.262063 LR 0.001000 Time 0.021985 +2023-10-02 20:48:15,398 - Epoch: [40][ 1160/ 1236] Overall Loss 0.262298 Objective Loss 0.262298 LR 0.001000 Time 0.021977 +2023-10-02 20:48:15,609 - Epoch: [40][ 1170/ 1236] Overall Loss 0.262433 Objective Loss 0.262433 LR 0.001000 Time 0.021970 +2023-10-02 20:48:15,820 - Epoch: [40][ 1180/ 1236] Overall Loss 0.262722 Objective Loss 0.262722 LR 0.001000 Time 0.021962 +2023-10-02 20:48:16,031 - Epoch: [40][ 1190/ 1236] Overall Loss 0.262837 Objective Loss 0.262837 LR 0.001000 Time 0.021955 +2023-10-02 20:48:16,243 - Epoch: [40][ 1200/ 1236] Overall Loss 0.263117 Objective Loss 0.263117 LR 0.001000 Time 0.021948 +2023-10-02 20:48:16,454 - Epoch: [40][ 1210/ 1236] Overall Loss 0.263135 Objective Loss 0.263135 LR 0.001000 Time 0.021941 +2023-10-02 20:48:16,666 - Epoch: [40][ 1220/ 1236] Overall Loss 0.263393 Objective Loss 0.263393 LR 0.001000 Time 0.021934 +2023-10-02 20:48:16,931 - Epoch: [40][ 1230/ 1236] Overall Loss 0.263412 Objective Loss 0.263412 LR 0.001000 Time 0.021971 +2023-10-02 20:48:17,054 - Epoch: [40][ 1236/ 1236] Overall Loss 0.263594 Objective Loss 0.263594 Top1 85.132383 Top5 97.352342 LR 0.001000 Time 0.021964 +2023-10-02 20:48:17,199 - --- validate (epoch=40)----------- +2023-10-02 20:48:17,199 - 29943 samples (256 per mini-batch) +2023-10-02 20:48:17,693 - Epoch: [40][ 10/ 117] Loss 0.299763 Top1 83.554688 Top5 98.164062 +2023-10-02 20:48:17,844 - Epoch: [40][ 20/ 117] Loss 0.306003 Top1 83.593750 Top5 98.164062 +2023-10-02 20:48:18,003 - Epoch: [40][ 30/ 117] Loss 0.318857 Top1 83.164062 Top5 98.098958 +2023-10-02 20:48:18,165 - Epoch: [40][ 40/ 117] Loss 0.315121 Top1 83.232422 Top5 98.251953 +2023-10-02 20:48:18,325 - Epoch: [40][ 50/ 117] Loss 0.320457 Top1 83.101562 Top5 98.195312 +2023-10-02 20:48:18,487 - Epoch: [40][ 60/ 117] Loss 0.319461 Top1 83.138021 Top5 98.157552 +2023-10-02 20:48:18,647 - Epoch: [40][ 70/ 117] Loss 0.321539 Top1 82.974330 Top5 98.097098 +2023-10-02 20:48:18,807 - Epoch: [40][ 80/ 117] Loss 0.321649 Top1 82.998047 Top5 98.105469 +2023-10-02 20:48:18,962 - Epoch: [40][ 90/ 117] Loss 0.322659 Top1 82.947049 Top5 98.081597 +2023-10-02 20:48:19,113 - Epoch: [40][ 100/ 117] Loss 0.322973 Top1 83.031250 Top5 98.101562 +2023-10-02 20:48:19,271 - Epoch: [40][ 110/ 117] Loss 0.324733 Top1 82.954545 Top5 98.121449 +2023-10-02 20:48:19,361 - Epoch: [40][ 117/ 117] Loss 0.322202 Top1 82.950940 Top5 98.106402 +2023-10-02 20:48:19,519 - ==> Top1: 82.951 Top5: 98.106 Loss: 0.322 + +2023-10-02 20:48:19,520 - ==> Confusion: +[[ 947 2 4 1 12 3 0 0 4 42 1 0 2 3 7 1 4 1 0 0 16] + [ 1 1047 1 0 3 16 2 21 4 0 1 1 2 0 1 4 2 0 17 0 8] + [ 5 0 958 12 3 0 30 10 0 0 1 0 11 2 2 4 1 1 4 5 7] + [ 1 1 10 949 1 1 3 5 6 1 3 0 18 3 32 1 0 6 28 1 19] + [ 31 12 1 0 960 4 0 1 0 2 0 2 1 3 12 2 10 1 1 3 4] + [ 1 52 1 1 2 939 2 39 10 7 3 6 5 13 7 0 1 0 8 8 11] + [ 0 6 19 0 1 1 1129 6 0 0 3 1 1 0 0 8 0 1 4 6 5] + [ 4 13 10 0 3 13 7 1073 1 0 3 7 11 0 1 2 0 1 54 7 8] + [ 19 1 2 0 1 1 0 2 984 32 10 1 7 5 17 0 3 1 1 1 1] + [ 149 0 0 0 12 2 1 0 50 868 0 1 2 17 6 0 0 1 0 2 8] + [ 2 2 13 5 2 0 4 3 22 1 955 2 3 7 4 0 0 3 17 0 8] + [ 0 1 2 0 3 10 0 3 0 0 0 938 44 5 0 4 1 14 0 5 5] + [ 1 1 0 1 0 1 2 3 0 0 2 39 984 0 3 4 1 7 3 6 10] + [ 3 0 3 0 3 9 0 1 24 9 6 10 3 1029 4 0 3 1 0 4 7] + [ 9 2 3 10 5 0 0 0 22 1 2 0 3 3 1023 0 1 2 4 0 11] + [ 0 0 2 0 4 1 1 0 0 0 0 10 8 0 1 1071 11 11 3 5 6] + [ 1 18 2 0 4 3 2 0 2 0 0 2 1 1 4 5 1097 0 1 5 13] + [ 0 0 0 3 0 0 2 0 1 0 0 3 30 0 0 3 0 990 0 1 5] + [ 3 3 9 8 0 0 0 17 4 0 2 0 1 0 9 0 1 0 1001 3 7] + [ 0 1 3 0 1 4 20 11 0 0 1 12 4 1 1 4 7 2 2 1071 7] + [ 157 228 144 59 81 144 61 119 156 71 151 109 467 248 171 52 149 65 219 229 4825]] + +2023-10-02 20:48:19,521 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:48:19,521 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:48:19,527 - + +2023-10-02 20:48:19,527 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:48:20,566 - Epoch: [41][ 10/ 1236] Overall Loss 0.269228 Objective Loss 0.269228 LR 0.001000 Time 0.103882 +2023-10-02 20:48:20,777 - Epoch: [41][ 20/ 1236] Overall Loss 0.259641 Objective Loss 0.259641 LR 0.001000 Time 0.062445 +2023-10-02 20:48:20,987 - Epoch: [41][ 30/ 1236] Overall Loss 0.264499 Objective Loss 0.264499 LR 0.001000 Time 0.048632 +2023-10-02 20:48:21,199 - Epoch: [41][ 40/ 1236] Overall Loss 0.256350 Objective Loss 0.256350 LR 0.001000 Time 0.041752 +2023-10-02 20:48:21,409 - Epoch: [41][ 50/ 1236] Overall Loss 0.260372 Objective Loss 0.260372 LR 0.001000 Time 0.037574 +2023-10-02 20:48:21,619 - Epoch: [41][ 60/ 1236] Overall Loss 0.258914 Objective Loss 0.258914 LR 0.001000 Time 0.034803 +2023-10-02 20:48:21,828 - Epoch: [41][ 70/ 1236] Overall Loss 0.261061 Objective Loss 0.261061 LR 0.001000 Time 0.032826 +2023-10-02 20:48:22,039 - Epoch: [41][ 80/ 1236] Overall Loss 0.263097 Objective Loss 0.263097 LR 0.001000 Time 0.031353 +2023-10-02 20:48:22,248 - Epoch: [41][ 90/ 1236] Overall Loss 0.267658 Objective Loss 0.267658 LR 0.001000 Time 0.030172 +2023-10-02 20:48:22,459 - Epoch: [41][ 100/ 1236] Overall Loss 0.268308 Objective Loss 0.268308 LR 0.001000 Time 0.029265 +2023-10-02 20:48:22,670 - Epoch: [41][ 110/ 1236] Overall Loss 0.268929 Objective Loss 0.268929 LR 0.001000 Time 0.028505 +2023-10-02 20:48:22,880 - Epoch: [41][ 120/ 1236] Overall Loss 0.269170 Objective Loss 0.269170 LR 0.001000 Time 0.027876 +2023-10-02 20:48:23,088 - Epoch: [41][ 130/ 1236] Overall Loss 0.266774 Objective Loss 0.266774 LR 0.001000 Time 0.027325 +2023-10-02 20:48:23,299 - Epoch: [41][ 140/ 1236] Overall Loss 0.265690 Objective Loss 0.265690 LR 0.001000 Time 0.026880 +2023-10-02 20:48:23,509 - Epoch: [41][ 150/ 1236] Overall Loss 0.267353 Objective Loss 0.267353 LR 0.001000 Time 0.026474 +2023-10-02 20:48:23,718 - Epoch: [41][ 160/ 1236] Overall Loss 0.267305 Objective Loss 0.267305 LR 0.001000 Time 0.026126 +2023-10-02 20:48:23,927 - Epoch: [41][ 170/ 1236] Overall Loss 0.268409 Objective Loss 0.268409 LR 0.001000 Time 0.025807 +2023-10-02 20:48:24,137 - Epoch: [41][ 180/ 1236] Overall Loss 0.268908 Objective Loss 0.268908 LR 0.001000 Time 0.025538 +2023-10-02 20:48:24,345 - Epoch: [41][ 190/ 1236] Overall Loss 0.267580 Objective Loss 0.267580 LR 0.001000 Time 0.025283 +2023-10-02 20:48:24,555 - Epoch: [41][ 200/ 1236] Overall Loss 0.267151 Objective Loss 0.267151 LR 0.001000 Time 0.025067 +2023-10-02 20:48:24,763 - Epoch: [41][ 210/ 1236] Overall Loss 0.266768 Objective Loss 0.266768 LR 0.001000 Time 0.024858 +2023-10-02 20:48:24,974 - Epoch: [41][ 220/ 1236] Overall Loss 0.266030 Objective Loss 0.266030 LR 0.001000 Time 0.024683 +2023-10-02 20:48:25,182 - Epoch: [41][ 230/ 1236] Overall Loss 0.266861 Objective Loss 0.266861 LR 0.001000 Time 0.024509 +2023-10-02 20:48:25,392 - Epoch: [41][ 240/ 1236] Overall Loss 0.266623 Objective Loss 0.266623 LR 0.001000 Time 0.024361 +2023-10-02 20:48:25,600 - Epoch: [41][ 250/ 1236] Overall Loss 0.266623 Objective Loss 0.266623 LR 0.001000 Time 0.024213 +2023-10-02 20:48:25,810 - Epoch: [41][ 260/ 1236] Overall Loss 0.266350 Objective Loss 0.266350 LR 0.001000 Time 0.024087 +2023-10-02 20:48:26,018 - Epoch: [41][ 270/ 1236] Overall Loss 0.266244 Objective Loss 0.266244 LR 0.001000 Time 0.023962 +2023-10-02 20:48:26,228 - Epoch: [41][ 280/ 1236] Overall Loss 0.265038 Objective Loss 0.265038 LR 0.001000 Time 0.023854 +2023-10-02 20:48:26,436 - Epoch: [41][ 290/ 1236] Overall Loss 0.264070 Objective Loss 0.264070 LR 0.001000 Time 0.023744 +2023-10-02 20:48:26,646 - Epoch: [41][ 300/ 1236] Overall Loss 0.263728 Objective Loss 0.263728 LR 0.001000 Time 0.023651 +2023-10-02 20:48:26,854 - Epoch: [41][ 310/ 1236] Overall Loss 0.263994 Objective Loss 0.263994 LR 0.001000 Time 0.023556 +2023-10-02 20:48:27,065 - Epoch: [41][ 320/ 1236] Overall Loss 0.264167 Objective Loss 0.264167 LR 0.001000 Time 0.023477 +2023-10-02 20:48:27,275 - Epoch: [41][ 330/ 1236] Overall Loss 0.264050 Objective Loss 0.264050 LR 0.001000 Time 0.023398 +2023-10-02 20:48:27,485 - Epoch: [41][ 340/ 1236] Overall Loss 0.264252 Objective Loss 0.264252 LR 0.001000 Time 0.023326 +2023-10-02 20:48:27,697 - Epoch: [41][ 350/ 1236] Overall Loss 0.263805 Objective Loss 0.263805 LR 0.001000 Time 0.023264 +2023-10-02 20:48:27,908 - Epoch: [41][ 360/ 1236] Overall Loss 0.263365 Objective Loss 0.263365 LR 0.001000 Time 0.023204 +2023-10-02 20:48:28,120 - Epoch: [41][ 370/ 1236] Overall Loss 0.263820 Objective Loss 0.263820 LR 0.001000 Time 0.023150 +2023-10-02 20:48:28,331 - Epoch: [41][ 380/ 1236] Overall Loss 0.263823 Objective Loss 0.263823 LR 0.001000 Time 0.023095 +2023-10-02 20:48:28,541 - Epoch: [41][ 390/ 1236] Overall Loss 0.263480 Objective Loss 0.263480 LR 0.001000 Time 0.023037 +2023-10-02 20:48:28,753 - Epoch: [41][ 400/ 1236] Overall Loss 0.263338 Objective Loss 0.263338 LR 0.001000 Time 0.022990 +2023-10-02 20:48:28,965 - Epoch: [41][ 410/ 1236] Overall Loss 0.263436 Objective Loss 0.263436 LR 0.001000 Time 0.022942 +2023-10-02 20:48:29,176 - Epoch: [41][ 420/ 1236] Overall Loss 0.264487 Objective Loss 0.264487 LR 0.001000 Time 0.022897 +2023-10-02 20:48:29,386 - Epoch: [41][ 430/ 1236] Overall Loss 0.264540 Objective Loss 0.264540 LR 0.001000 Time 0.022850 +2023-10-02 20:48:29,597 - Epoch: [41][ 440/ 1236] Overall Loss 0.264103 Objective Loss 0.264103 LR 0.001000 Time 0.022809 +2023-10-02 20:48:29,806 - Epoch: [41][ 450/ 1236] Overall Loss 0.264087 Objective Loss 0.264087 LR 0.001000 Time 0.022765 +2023-10-02 20:48:30,017 - Epoch: [41][ 460/ 1236] Overall Loss 0.263469 Objective Loss 0.263469 LR 0.001000 Time 0.022728 +2023-10-02 20:48:30,227 - Epoch: [41][ 470/ 1236] Overall Loss 0.262967 Objective Loss 0.262967 LR 0.001000 Time 0.022687 +2023-10-02 20:48:30,438 - Epoch: [41][ 480/ 1236] Overall Loss 0.262861 Objective Loss 0.262861 LR 0.001000 Time 0.022654 +2023-10-02 20:48:30,648 - Epoch: [41][ 490/ 1236] Overall Loss 0.263108 Objective Loss 0.263108 LR 0.001000 Time 0.022617 +2023-10-02 20:48:30,858 - Epoch: [41][ 500/ 1236] Overall Loss 0.263200 Objective Loss 0.263200 LR 0.001000 Time 0.022585 +2023-10-02 20:48:31,069 - Epoch: [41][ 510/ 1236] Overall Loss 0.263032 Objective Loss 0.263032 LR 0.001000 Time 0.022551 +2023-10-02 20:48:31,279 - Epoch: [41][ 520/ 1236] Overall Loss 0.262563 Objective Loss 0.262563 LR 0.001000 Time 0.022523 +2023-10-02 20:48:31,489 - Epoch: [41][ 530/ 1236] Overall Loss 0.262354 Objective Loss 0.262354 LR 0.001000 Time 0.022491 +2023-10-02 20:48:31,700 - Epoch: [41][ 540/ 1236] Overall Loss 0.262080 Objective Loss 0.262080 LR 0.001000 Time 0.022465 +2023-10-02 20:48:31,910 - Epoch: [41][ 550/ 1236] Overall Loss 0.262290 Objective Loss 0.262290 LR 0.001000 Time 0.022435 +2023-10-02 20:48:32,121 - Epoch: [41][ 560/ 1236] Overall Loss 0.262247 Objective Loss 0.262247 LR 0.001000 Time 0.022411 +2023-10-02 20:48:32,331 - Epoch: [41][ 570/ 1236] Overall Loss 0.262308 Objective Loss 0.262308 LR 0.001000 Time 0.022383 +2023-10-02 20:48:32,542 - Epoch: [41][ 580/ 1236] Overall Loss 0.262072 Objective Loss 0.262072 LR 0.001000 Time 0.022361 +2023-10-02 20:48:32,752 - Epoch: [41][ 590/ 1236] Overall Loss 0.261748 Objective Loss 0.261748 LR 0.001000 Time 0.022335 +2023-10-02 20:48:32,963 - Epoch: [41][ 600/ 1236] Overall Loss 0.261500 Objective Loss 0.261500 LR 0.001000 Time 0.022314 +2023-10-02 20:48:33,172 - Epoch: [41][ 610/ 1236] Overall Loss 0.261434 Objective Loss 0.261434 LR 0.001000 Time 0.022289 +2023-10-02 20:48:33,383 - Epoch: [41][ 620/ 1236] Overall Loss 0.261757 Objective Loss 0.261757 LR 0.001000 Time 0.022269 +2023-10-02 20:48:33,593 - Epoch: [41][ 630/ 1236] Overall Loss 0.261918 Objective Loss 0.261918 LR 0.001000 Time 0.022246 +2023-10-02 20:48:33,803 - Epoch: [41][ 640/ 1236] Overall Loss 0.262424 Objective Loss 0.262424 LR 0.001000 Time 0.022227 +2023-10-02 20:48:34,013 - Epoch: [41][ 650/ 1236] Overall Loss 0.262998 Objective Loss 0.262998 LR 0.001000 Time 0.022205 +2023-10-02 20:48:34,224 - Epoch: [41][ 660/ 1236] Overall Loss 0.263101 Objective Loss 0.263101 LR 0.001000 Time 0.022188 +2023-10-02 20:48:34,434 - Epoch: [41][ 670/ 1236] Overall Loss 0.263398 Objective Loss 0.263398 LR 0.001000 Time 0.022168 +2023-10-02 20:48:34,645 - Epoch: [41][ 680/ 1236] Overall Loss 0.263539 Objective Loss 0.263539 LR 0.001000 Time 0.022151 +2023-10-02 20:48:34,854 - Epoch: [41][ 690/ 1236] Overall Loss 0.263408 Objective Loss 0.263408 LR 0.001000 Time 0.022132 +2023-10-02 20:48:35,065 - Epoch: [41][ 700/ 1236] Overall Loss 0.263724 Objective Loss 0.263724 LR 0.001000 Time 0.022117 +2023-10-02 20:48:35,275 - Epoch: [41][ 710/ 1236] Overall Loss 0.263724 Objective Loss 0.263724 LR 0.001000 Time 0.022099 +2023-10-02 20:48:35,486 - Epoch: [41][ 720/ 1236] Overall Loss 0.263639 Objective Loss 0.263639 LR 0.001000 Time 0.022085 +2023-10-02 20:48:35,696 - Epoch: [41][ 730/ 1236] Overall Loss 0.263583 Objective Loss 0.263583 LR 0.001000 Time 0.022067 +2023-10-02 20:48:35,907 - Epoch: [41][ 740/ 1236] Overall Loss 0.263472 Objective Loss 0.263472 LR 0.001000 Time 0.022054 +2023-10-02 20:48:36,117 - Epoch: [41][ 750/ 1236] Overall Loss 0.263371 Objective Loss 0.263371 LR 0.001000 Time 0.022038 +2023-10-02 20:48:36,328 - Epoch: [41][ 760/ 1236] Overall Loss 0.263406 Objective Loss 0.263406 LR 0.001000 Time 0.022025 +2023-10-02 20:48:36,537 - Epoch: [41][ 770/ 1236] Overall Loss 0.263679 Objective Loss 0.263679 LR 0.001000 Time 0.022009 +2023-10-02 20:48:36,748 - Epoch: [41][ 780/ 1236] Overall Loss 0.264001 Objective Loss 0.264001 LR 0.001000 Time 0.021997 +2023-10-02 20:48:36,958 - Epoch: [41][ 790/ 1236] Overall Loss 0.264087 Objective Loss 0.264087 LR 0.001000 Time 0.021983 +2023-10-02 20:48:37,169 - Epoch: [41][ 800/ 1236] Overall Loss 0.264382 Objective Loss 0.264382 LR 0.001000 Time 0.021971 +2023-10-02 20:48:37,380 - Epoch: [41][ 810/ 1236] Overall Loss 0.264561 Objective Loss 0.264561 LR 0.001000 Time 0.021958 +2023-10-02 20:48:37,591 - Epoch: [41][ 820/ 1236] Overall Loss 0.264836 Objective Loss 0.264836 LR 0.001000 Time 0.021947 +2023-10-02 20:48:37,801 - Epoch: [41][ 830/ 1236] Overall Loss 0.265129 Objective Loss 0.265129 LR 0.001000 Time 0.021934 +2023-10-02 20:48:38,012 - Epoch: [41][ 840/ 1236] Overall Loss 0.265558 Objective Loss 0.265558 LR 0.001000 Time 0.021924 +2023-10-02 20:48:38,222 - Epoch: [41][ 850/ 1236] Overall Loss 0.265924 Objective Loss 0.265924 LR 0.001000 Time 0.021911 +2023-10-02 20:48:38,433 - Epoch: [41][ 860/ 1236] Overall Loss 0.266159 Objective Loss 0.266159 LR 0.001000 Time 0.021901 +2023-10-02 20:48:38,643 - Epoch: [41][ 870/ 1236] Overall Loss 0.266126 Objective Loss 0.266126 LR 0.001000 Time 0.021889 +2023-10-02 20:48:38,853 - Epoch: [41][ 880/ 1236] Overall Loss 0.266350 Objective Loss 0.266350 LR 0.001000 Time 0.021880 +2023-10-02 20:48:39,063 - Epoch: [41][ 890/ 1236] Overall Loss 0.266327 Objective Loss 0.266327 LR 0.001000 Time 0.021868 +2023-10-02 20:48:39,274 - Epoch: [41][ 900/ 1236] Overall Loss 0.266920 Objective Loss 0.266920 LR 0.001000 Time 0.021859 +2023-10-02 20:48:39,484 - Epoch: [41][ 910/ 1236] Overall Loss 0.266878 Objective Loss 0.266878 LR 0.001000 Time 0.021848 +2023-10-02 20:48:39,695 - Epoch: [41][ 920/ 1236] Overall Loss 0.267041 Objective Loss 0.267041 LR 0.001000 Time 0.021839 +2023-10-02 20:48:39,905 - Epoch: [41][ 930/ 1236] Overall Loss 0.267184 Objective Loss 0.267184 LR 0.001000 Time 0.021829 +2023-10-02 20:48:40,116 - Epoch: [41][ 940/ 1236] Overall Loss 0.267057 Objective Loss 0.267057 LR 0.001000 Time 0.021821 +2023-10-02 20:48:40,326 - Epoch: [41][ 950/ 1236] Overall Loss 0.267099 Objective Loss 0.267099 LR 0.001000 Time 0.021811 +2023-10-02 20:48:40,537 - Epoch: [41][ 960/ 1236] Overall Loss 0.267044 Objective Loss 0.267044 LR 0.001000 Time 0.021803 +2023-10-02 20:48:40,747 - Epoch: [41][ 970/ 1236] Overall Loss 0.267344 Objective Loss 0.267344 LR 0.001000 Time 0.021793 +2023-10-02 20:48:40,958 - Epoch: [41][ 980/ 1236] Overall Loss 0.267299 Objective Loss 0.267299 LR 0.001000 Time 0.021785 +2023-10-02 20:48:41,168 - Epoch: [41][ 990/ 1236] Overall Loss 0.267322 Objective Loss 0.267322 LR 0.001000 Time 0.021776 +2023-10-02 20:48:41,379 - Epoch: [41][ 1000/ 1236] Overall Loss 0.267477 Objective Loss 0.267477 LR 0.001000 Time 0.021769 +2023-10-02 20:48:41,589 - Epoch: [41][ 1010/ 1236] Overall Loss 0.267480 Objective Loss 0.267480 LR 0.001000 Time 0.021760 +2023-10-02 20:48:41,800 - Epoch: [41][ 1020/ 1236] Overall Loss 0.267413 Objective Loss 0.267413 LR 0.001000 Time 0.021753 +2023-10-02 20:48:42,012 - Epoch: [41][ 1030/ 1236] Overall Loss 0.267485 Objective Loss 0.267485 LR 0.001000 Time 0.021746 +2023-10-02 20:48:42,223 - Epoch: [41][ 1040/ 1236] Overall Loss 0.267439 Objective Loss 0.267439 LR 0.001000 Time 0.021740 +2023-10-02 20:48:42,433 - Epoch: [41][ 1050/ 1236] Overall Loss 0.267532 Objective Loss 0.267532 LR 0.001000 Time 0.021731 +2023-10-02 20:48:42,644 - Epoch: [41][ 1060/ 1236] Overall Loss 0.267781 Objective Loss 0.267781 LR 0.001000 Time 0.021725 +2023-10-02 20:48:42,854 - Epoch: [41][ 1070/ 1236] Overall Loss 0.268138 Objective Loss 0.268138 LR 0.001000 Time 0.021717 +2023-10-02 20:48:43,065 - Epoch: [41][ 1080/ 1236] Overall Loss 0.268326 Objective Loss 0.268326 LR 0.001000 Time 0.021711 +2023-10-02 20:48:43,275 - Epoch: [41][ 1090/ 1236] Overall Loss 0.268644 Objective Loss 0.268644 LR 0.001000 Time 0.021703 +2023-10-02 20:48:43,485 - Epoch: [41][ 1100/ 1236] Overall Loss 0.268821 Objective Loss 0.268821 LR 0.001000 Time 0.021697 +2023-10-02 20:48:43,695 - Epoch: [41][ 1110/ 1236] Overall Loss 0.268806 Objective Loss 0.268806 LR 0.001000 Time 0.021689 +2023-10-02 20:48:43,906 - Epoch: [41][ 1120/ 1236] Overall Loss 0.268816 Objective Loss 0.268816 LR 0.001000 Time 0.021684 +2023-10-02 20:48:44,116 - Epoch: [41][ 1130/ 1236] Overall Loss 0.269098 Objective Loss 0.269098 LR 0.001000 Time 0.021676 +2023-10-02 20:48:44,327 - Epoch: [41][ 1140/ 1236] Overall Loss 0.269281 Objective Loss 0.269281 LR 0.001000 Time 0.021671 +2023-10-02 20:48:44,537 - Epoch: [41][ 1150/ 1236] Overall Loss 0.269505 Objective Loss 0.269505 LR 0.001000 Time 0.021664 +2023-10-02 20:48:44,748 - Epoch: [41][ 1160/ 1236] Overall Loss 0.269635 Objective Loss 0.269635 LR 0.001000 Time 0.021659 +2023-10-02 20:48:44,959 - Epoch: [41][ 1170/ 1236] Overall Loss 0.269904 Objective Loss 0.269904 LR 0.001000 Time 0.021652 +2023-10-02 20:48:45,170 - Epoch: [41][ 1180/ 1236] Overall Loss 0.269792 Objective Loss 0.269792 LR 0.001000 Time 0.021647 +2023-10-02 20:48:45,380 - Epoch: [41][ 1190/ 1236] Overall Loss 0.269934 Objective Loss 0.269934 LR 0.001000 Time 0.021641 +2023-10-02 20:48:45,591 - Epoch: [41][ 1200/ 1236] Overall Loss 0.270021 Objective Loss 0.270021 LR 0.001000 Time 0.021636 +2023-10-02 20:48:45,802 - Epoch: [41][ 1210/ 1236] Overall Loss 0.270169 Objective Loss 0.270169 LR 0.001000 Time 0.021630 +2023-10-02 20:48:46,013 - Epoch: [41][ 1220/ 1236] Overall Loss 0.270305 Objective Loss 0.270305 LR 0.001000 Time 0.021626 +2023-10-02 20:48:46,278 - Epoch: [41][ 1230/ 1236] Overall Loss 0.270113 Objective Loss 0.270113 LR 0.001000 Time 0.021664 +2023-10-02 20:48:46,401 - Epoch: [41][ 1236/ 1236] Overall Loss 0.270173 Objective Loss 0.270173 Top1 86.761711 Top5 98.778004 LR 0.001000 Time 0.021658 +2023-10-02 20:48:46,523 - --- validate (epoch=41)----------- +2023-10-02 20:48:46,524 - 29943 samples (256 per mini-batch) +2023-10-02 20:48:47,007 - Epoch: [41][ 10/ 117] Loss 0.324044 Top1 82.304688 Top5 98.203125 +2023-10-02 20:48:47,161 - Epoch: [41][ 20/ 117] Loss 0.319540 Top1 83.281250 Top5 98.144531 +2023-10-02 20:48:47,313 - Epoch: [41][ 30/ 117] Loss 0.307940 Top1 83.580729 Top5 98.177083 +2023-10-02 20:48:47,466 - Epoch: [41][ 40/ 117] Loss 0.309000 Top1 83.525391 Top5 98.212891 +2023-10-02 20:48:47,617 - Epoch: [41][ 50/ 117] Loss 0.310941 Top1 83.507812 Top5 98.140625 +2023-10-02 20:48:47,769 - Epoch: [41][ 60/ 117] Loss 0.312501 Top1 83.385417 Top5 98.164062 +2023-10-02 20:48:47,920 - Epoch: [41][ 70/ 117] Loss 0.312108 Top1 83.337054 Top5 98.191964 +2023-10-02 20:48:48,073 - Epoch: [41][ 80/ 117] Loss 0.315703 Top1 83.378906 Top5 98.149414 +2023-10-02 20:48:48,225 - Epoch: [41][ 90/ 117] Loss 0.316079 Top1 83.515625 Top5 98.116319 +2023-10-02 20:48:48,384 - Epoch: [41][ 100/ 117] Loss 0.317599 Top1 83.457031 Top5 98.097656 +2023-10-02 20:48:48,549 - Epoch: [41][ 110/ 117] Loss 0.317465 Top1 83.348722 Top5 98.093040 +2023-10-02 20:48:48,639 - Epoch: [41][ 117/ 117] Loss 0.317192 Top1 83.284908 Top5 98.106402 +2023-10-02 20:48:48,778 - ==> Top1: 83.285 Top5: 98.106 Loss: 0.317 + +2023-10-02 20:48:48,779 - ==> Confusion: +[[ 940 2 7 1 7 3 0 1 5 53 2 1 0 1 8 3 2 3 1 0 10] + [ 2 1067 2 1 5 20 0 16 2 0 1 1 0 0 1 4 1 0 5 2 1] + [ 3 0 950 24 1 1 25 7 0 1 2 4 4 3 0 6 0 2 13 3 7] + [ 0 1 2 1023 0 3 2 0 2 2 1 0 1 1 15 1 1 3 15 1 15] + [ 25 11 2 0 962 6 0 2 0 8 0 3 0 4 9 5 7 1 0 2 3] + [ 3 54 1 6 1 971 1 18 2 4 3 9 4 15 4 1 2 0 6 2 9] + [ 1 7 34 1 0 0 1123 4 0 0 4 2 0 0 1 4 0 2 2 4 2] + [ 2 39 14 1 4 29 4 1029 0 3 3 3 1 4 2 1 0 0 59 14 6] + [ 24 2 0 1 2 0 0 1 959 39 8 2 2 15 24 1 1 1 5 2 0] + [ 130 0 1 0 5 2 1 0 29 911 1 0 0 14 11 0 1 0 0 5 8] + [ 3 2 11 17 1 1 2 1 17 0 959 4 0 9 4 0 0 1 12 1 8] + [ 0 1 1 0 1 16 0 3 0 1 0 936 39 7 0 3 0 16 0 7 4] + [ 3 1 2 7 1 3 4 1 1 0 1 27 959 3 0 10 1 25 4 7 8] + [ 3 0 3 0 5 12 0 0 10 17 10 8 0 1027 8 2 1 2 0 0 11] + [ 10 1 6 23 4 0 0 0 22 1 1 0 2 2 1013 0 1 2 9 0 4] + [ 0 2 4 5 6 1 1 0 0 0 0 9 4 1 1 1064 11 13 2 4 6] + [ 0 16 1 3 8 3 0 0 1 0 0 6 0 2 2 12 1086 1 1 4 15] + [ 0 0 2 6 1 0 2 0 0 0 0 5 19 1 3 5 0 991 1 0 2] + [ 1 7 5 19 1 0 1 14 4 2 1 0 1 0 5 0 0 0 998 1 8] + [ 0 1 1 2 1 5 13 7 0 0 1 18 1 1 1 5 4 1 3 1082 5] + [ 194 220 157 156 85 169 43 73 125 78 170 149 337 250 167 65 116 66 186 211 4888]] + +2023-10-02 20:48:48,780 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:48:48,780 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:48:48,786 - + +2023-10-02 20:48:48,786 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:48:49,818 - Epoch: [42][ 10/ 1236] Overall Loss 0.258973 Objective Loss 0.258973 LR 0.001000 Time 0.103125 +2023-10-02 20:48:50,027 - Epoch: [42][ 20/ 1236] Overall Loss 0.250245 Objective Loss 0.250245 LR 0.001000 Time 0.062002 +2023-10-02 20:48:50,236 - Epoch: [42][ 30/ 1236] Overall Loss 0.261321 Objective Loss 0.261321 LR 0.001000 Time 0.048278 +2023-10-02 20:48:50,445 - Epoch: [42][ 40/ 1236] Overall Loss 0.261978 Objective Loss 0.261978 LR 0.001000 Time 0.041434 +2023-10-02 20:48:50,658 - Epoch: [42][ 50/ 1236] Overall Loss 0.268560 Objective Loss 0.268560 LR 0.001000 Time 0.037365 +2023-10-02 20:48:50,869 - Epoch: [42][ 60/ 1236] Overall Loss 0.265237 Objective Loss 0.265237 LR 0.001000 Time 0.034650 +2023-10-02 20:48:51,080 - Epoch: [42][ 70/ 1236] Overall Loss 0.267624 Objective Loss 0.267624 LR 0.001000 Time 0.032701 +2023-10-02 20:48:51,291 - Epoch: [42][ 80/ 1236] Overall Loss 0.266044 Objective Loss 0.266044 LR 0.001000 Time 0.031249 +2023-10-02 20:48:51,503 - Epoch: [42][ 90/ 1236] Overall Loss 0.267964 Objective Loss 0.267964 LR 0.001000 Time 0.030110 +2023-10-02 20:48:51,712 - Epoch: [42][ 100/ 1236] Overall Loss 0.268090 Objective Loss 0.268090 LR 0.001000 Time 0.029190 +2023-10-02 20:48:51,921 - Epoch: [42][ 110/ 1236] Overall Loss 0.265258 Objective Loss 0.265258 LR 0.001000 Time 0.028419 +2023-10-02 20:48:52,130 - Epoch: [42][ 120/ 1236] Overall Loss 0.265783 Objective Loss 0.265783 LR 0.001000 Time 0.027795 +2023-10-02 20:48:52,339 - Epoch: [42][ 130/ 1236] Overall Loss 0.264560 Objective Loss 0.264560 LR 0.001000 Time 0.027250 +2023-10-02 20:48:52,549 - Epoch: [42][ 140/ 1236] Overall Loss 0.264446 Objective Loss 0.264446 LR 0.001000 Time 0.026800 +2023-10-02 20:48:52,760 - Epoch: [42][ 150/ 1236] Overall Loss 0.263233 Objective Loss 0.263233 LR 0.001000 Time 0.026413 +2023-10-02 20:48:52,974 - Epoch: [42][ 160/ 1236] Overall Loss 0.262577 Objective Loss 0.262577 LR 0.001000 Time 0.026098 +2023-10-02 20:48:53,185 - Epoch: [42][ 170/ 1236] Overall Loss 0.261264 Objective Loss 0.261264 LR 0.001000 Time 0.025803 +2023-10-02 20:48:53,399 - Epoch: [42][ 180/ 1236] Overall Loss 0.261907 Objective Loss 0.261907 LR 0.001000 Time 0.025556 +2023-10-02 20:48:53,610 - Epoch: [42][ 190/ 1236] Overall Loss 0.262139 Objective Loss 0.262139 LR 0.001000 Time 0.025321 +2023-10-02 20:48:53,824 - Epoch: [42][ 200/ 1236] Overall Loss 0.263112 Objective Loss 0.263112 LR 0.001000 Time 0.025123 +2023-10-02 20:48:54,035 - Epoch: [42][ 210/ 1236] Overall Loss 0.261934 Objective Loss 0.261934 LR 0.001000 Time 0.024930 +2023-10-02 20:48:54,249 - Epoch: [42][ 220/ 1236] Overall Loss 0.260209 Objective Loss 0.260209 LR 0.001000 Time 0.024768 +2023-10-02 20:48:54,460 - Epoch: [42][ 230/ 1236] Overall Loss 0.259325 Objective Loss 0.259325 LR 0.001000 Time 0.024607 +2023-10-02 20:48:54,674 - Epoch: [42][ 240/ 1236] Overall Loss 0.258766 Objective Loss 0.258766 LR 0.001000 Time 0.024472 +2023-10-02 20:48:54,885 - Epoch: [42][ 250/ 1236] Overall Loss 0.259426 Objective Loss 0.259426 LR 0.001000 Time 0.024337 +2023-10-02 20:48:55,099 - Epoch: [42][ 260/ 1236] Overall Loss 0.259944 Objective Loss 0.259944 LR 0.001000 Time 0.024222 +2023-10-02 20:48:55,310 - Epoch: [42][ 270/ 1236] Overall Loss 0.259453 Objective Loss 0.259453 LR 0.001000 Time 0.024106 +2023-10-02 20:48:55,524 - Epoch: [42][ 280/ 1236] Overall Loss 0.259815 Objective Loss 0.259815 LR 0.001000 Time 0.024008 +2023-10-02 20:48:55,735 - Epoch: [42][ 290/ 1236] Overall Loss 0.259934 Objective Loss 0.259934 LR 0.001000 Time 0.023907 +2023-10-02 20:48:55,949 - Epoch: [42][ 300/ 1236] Overall Loss 0.260121 Objective Loss 0.260121 LR 0.001000 Time 0.023822 +2023-10-02 20:48:56,164 - Epoch: [42][ 310/ 1236] Overall Loss 0.259522 Objective Loss 0.259522 LR 0.001000 Time 0.023734 +2023-10-02 20:48:56,377 - Epoch: [42][ 320/ 1236] Overall Loss 0.259144 Objective Loss 0.259144 LR 0.001000 Time 0.023658 +2023-10-02 20:48:56,587 - Epoch: [42][ 330/ 1236] Overall Loss 0.258901 Objective Loss 0.258901 LR 0.001000 Time 0.023575 +2023-10-02 20:48:56,800 - Epoch: [42][ 340/ 1236] Overall Loss 0.258829 Objective Loss 0.258829 LR 0.001000 Time 0.023507 +2023-10-02 20:48:57,009 - Epoch: [42][ 350/ 1236] Overall Loss 0.259345 Objective Loss 0.259345 LR 0.001000 Time 0.023432 +2023-10-02 20:48:57,222 - Epoch: [42][ 360/ 1236] Overall Loss 0.260686 Objective Loss 0.260686 LR 0.001000 Time 0.023371 +2023-10-02 20:48:57,431 - Epoch: [42][ 370/ 1236] Overall Loss 0.260061 Objective Loss 0.260061 LR 0.001000 Time 0.023305 +2023-10-02 20:48:57,647 - Epoch: [42][ 380/ 1236] Overall Loss 0.259689 Objective Loss 0.259689 LR 0.001000 Time 0.023258 +2023-10-02 20:48:57,864 - Epoch: [42][ 390/ 1236] Overall Loss 0.259911 Objective Loss 0.259911 LR 0.001000 Time 0.023219 +2023-10-02 20:48:58,087 - Epoch: [42][ 400/ 1236] Overall Loss 0.260427 Objective Loss 0.260427 LR 0.001000 Time 0.023195 +2023-10-02 20:48:58,297 - Epoch: [42][ 410/ 1236] Overall Loss 0.260420 Objective Loss 0.260420 LR 0.001000 Time 0.023139 +2023-10-02 20:48:58,508 - Epoch: [42][ 420/ 1236] Overall Loss 0.260845 Objective Loss 0.260845 LR 0.001000 Time 0.023091 +2023-10-02 20:48:58,719 - Epoch: [42][ 430/ 1236] Overall Loss 0.260394 Objective Loss 0.260394 LR 0.001000 Time 0.023041 +2023-10-02 20:48:58,931 - Epoch: [42][ 440/ 1236] Overall Loss 0.260107 Objective Loss 0.260107 LR 0.001000 Time 0.022998 +2023-10-02 20:48:59,141 - Epoch: [42][ 450/ 1236] Overall Loss 0.260329 Objective Loss 0.260329 LR 0.001000 Time 0.022952 +2023-10-02 20:48:59,353 - Epoch: [42][ 460/ 1236] Overall Loss 0.260211 Objective Loss 0.260211 LR 0.001000 Time 0.022912 +2023-10-02 20:48:59,563 - Epoch: [42][ 470/ 1236] Overall Loss 0.260374 Objective Loss 0.260374 LR 0.001000 Time 0.022868 +2023-10-02 20:48:59,776 - Epoch: [42][ 480/ 1236] Overall Loss 0.260650 Objective Loss 0.260650 LR 0.001000 Time 0.022836 +2023-10-02 20:48:59,986 - Epoch: [42][ 490/ 1236] Overall Loss 0.260485 Objective Loss 0.260485 LR 0.001000 Time 0.022796 +2023-10-02 20:49:00,197 - Epoch: [42][ 500/ 1236] Overall Loss 0.260445 Objective Loss 0.260445 LR 0.001000 Time 0.022763 +2023-10-02 20:49:00,408 - Epoch: [42][ 510/ 1236] Overall Loss 0.260008 Objective Loss 0.260008 LR 0.001000 Time 0.022727 +2023-10-02 20:49:00,616 - Epoch: [42][ 520/ 1236] Overall Loss 0.259781 Objective Loss 0.259781 LR 0.001000 Time 0.022689 +2023-10-02 20:49:00,825 - Epoch: [42][ 530/ 1236] Overall Loss 0.259256 Objective Loss 0.259256 LR 0.001000 Time 0.022652 +2023-10-02 20:49:01,034 - Epoch: [42][ 540/ 1236] Overall Loss 0.259358 Objective Loss 0.259358 LR 0.001000 Time 0.022619 +2023-10-02 20:49:01,243 - Epoch: [42][ 550/ 1236] Overall Loss 0.259777 Objective Loss 0.259777 LR 0.001000 Time 0.022587 +2023-10-02 20:49:01,452 - Epoch: [42][ 560/ 1236] Overall Loss 0.260216 Objective Loss 0.260216 LR 0.001000 Time 0.022556 +2023-10-02 20:49:01,660 - Epoch: [42][ 570/ 1236] Overall Loss 0.260461 Objective Loss 0.260461 LR 0.001000 Time 0.022526 +2023-10-02 20:49:01,870 - Epoch: [42][ 580/ 1236] Overall Loss 0.260351 Objective Loss 0.260351 LR 0.001000 Time 0.022498 +2023-10-02 20:49:02,079 - Epoch: [42][ 590/ 1236] Overall Loss 0.260420 Objective Loss 0.260420 LR 0.001000 Time 0.022470 +2023-10-02 20:49:02,288 - Epoch: [42][ 600/ 1236] Overall Loss 0.260650 Objective Loss 0.260650 LR 0.001000 Time 0.022444 +2023-10-02 20:49:02,496 - Epoch: [42][ 610/ 1236] Overall Loss 0.260819 Objective Loss 0.260819 LR 0.001000 Time 0.022415 +2023-10-02 20:49:02,706 - Epoch: [42][ 620/ 1236] Overall Loss 0.260627 Objective Loss 0.260627 LR 0.001000 Time 0.022391 +2023-10-02 20:49:02,915 - Epoch: [42][ 630/ 1236] Overall Loss 0.260329 Objective Loss 0.260329 LR 0.001000 Time 0.022365 +2023-10-02 20:49:03,123 - Epoch: [42][ 640/ 1236] Overall Loss 0.260249 Objective Loss 0.260249 LR 0.001000 Time 0.022340 +2023-10-02 20:49:03,331 - Epoch: [42][ 650/ 1236] Overall Loss 0.259981 Objective Loss 0.259981 LR 0.001000 Time 0.022316 +2023-10-02 20:49:03,541 - Epoch: [42][ 660/ 1236] Overall Loss 0.259707 Objective Loss 0.259707 LR 0.001000 Time 0.022296 +2023-10-02 20:49:03,749 - Epoch: [42][ 670/ 1236] Overall Loss 0.259565 Objective Loss 0.259565 LR 0.001000 Time 0.022273 +2023-10-02 20:49:03,959 - Epoch: [42][ 680/ 1236] Overall Loss 0.259653 Objective Loss 0.259653 LR 0.001000 Time 0.022253 +2023-10-02 20:49:04,167 - Epoch: [42][ 690/ 1236] Overall Loss 0.259347 Objective Loss 0.259347 LR 0.001000 Time 0.022232 +2023-10-02 20:49:04,377 - Epoch: [42][ 700/ 1236] Overall Loss 0.259450 Objective Loss 0.259450 LR 0.001000 Time 0.022214 +2023-10-02 20:49:04,585 - Epoch: [42][ 710/ 1236] Overall Loss 0.259522 Objective Loss 0.259522 LR 0.001000 Time 0.022192 +2023-10-02 20:49:04,795 - Epoch: [42][ 720/ 1236] Overall Loss 0.259765 Objective Loss 0.259765 LR 0.001000 Time 0.022174 +2023-10-02 20:49:05,002 - Epoch: [42][ 730/ 1236] Overall Loss 0.259768 Objective Loss 0.259768 LR 0.001000 Time 0.022152 +2023-10-02 20:49:05,212 - Epoch: [42][ 740/ 1236] Overall Loss 0.259699 Objective Loss 0.259699 LR 0.001000 Time 0.022136 +2023-10-02 20:49:05,421 - Epoch: [42][ 750/ 1236] Overall Loss 0.259962 Objective Loss 0.259962 LR 0.001000 Time 0.022118 +2023-10-02 20:49:05,631 - Epoch: [42][ 760/ 1236] Overall Loss 0.259679 Objective Loss 0.259679 LR 0.001000 Time 0.022102 +2023-10-02 20:49:05,840 - Epoch: [42][ 770/ 1236] Overall Loss 0.259826 Objective Loss 0.259826 LR 0.001000 Time 0.022085 +2023-10-02 20:49:06,049 - Epoch: [42][ 780/ 1236] Overall Loss 0.260081 Objective Loss 0.260081 LR 0.001000 Time 0.022069 +2023-10-02 20:49:06,258 - Epoch: [42][ 790/ 1236] Overall Loss 0.260136 Objective Loss 0.260136 LR 0.001000 Time 0.022052 +2023-10-02 20:49:06,468 - Epoch: [42][ 800/ 1236] Overall Loss 0.260000 Objective Loss 0.260000 LR 0.001000 Time 0.022039 +2023-10-02 20:49:06,676 - Epoch: [42][ 810/ 1236] Overall Loss 0.259939 Objective Loss 0.259939 LR 0.001000 Time 0.022024 +2023-10-02 20:49:06,885 - Epoch: [42][ 820/ 1236] Overall Loss 0.260208 Objective Loss 0.260208 LR 0.001000 Time 0.022009 +2023-10-02 20:49:07,094 - Epoch: [42][ 830/ 1236] Overall Loss 0.260474 Objective Loss 0.260474 LR 0.001000 Time 0.021995 +2023-10-02 20:49:07,301 - Epoch: [42][ 840/ 1236] Overall Loss 0.260647 Objective Loss 0.260647 LR 0.001000 Time 0.021980 +2023-10-02 20:49:07,510 - Epoch: [42][ 850/ 1236] Overall Loss 0.260563 Objective Loss 0.260563 LR 0.001000 Time 0.021967 +2023-10-02 20:49:07,718 - Epoch: [42][ 860/ 1236] Overall Loss 0.260682 Objective Loss 0.260682 LR 0.001000 Time 0.021953 +2023-10-02 20:49:07,927 - Epoch: [42][ 870/ 1236] Overall Loss 0.260535 Objective Loss 0.260535 LR 0.001000 Time 0.021940 +2023-10-02 20:49:08,136 - Epoch: [42][ 880/ 1236] Overall Loss 0.260317 Objective Loss 0.260317 LR 0.001000 Time 0.021928 +2023-10-02 20:49:08,345 - Epoch: [42][ 890/ 1236] Overall Loss 0.260555 Objective Loss 0.260555 LR 0.001000 Time 0.021914 +2023-10-02 20:49:08,554 - Epoch: [42][ 900/ 1236] Overall Loss 0.260643 Objective Loss 0.260643 LR 0.001000 Time 0.021903 +2023-10-02 20:49:08,763 - Epoch: [42][ 910/ 1236] Overall Loss 0.260908 Objective Loss 0.260908 LR 0.001000 Time 0.021890 +2023-10-02 20:49:08,972 - Epoch: [42][ 920/ 1236] Overall Loss 0.260940 Objective Loss 0.260940 LR 0.001000 Time 0.021880 +2023-10-02 20:49:09,181 - Epoch: [42][ 930/ 1236] Overall Loss 0.261218 Objective Loss 0.261218 LR 0.001000 Time 0.021867 +2023-10-02 20:49:09,390 - Epoch: [42][ 940/ 1236] Overall Loss 0.261323 Objective Loss 0.261323 LR 0.001000 Time 0.021857 +2023-10-02 20:49:09,599 - Epoch: [42][ 950/ 1236] Overall Loss 0.261302 Objective Loss 0.261302 LR 0.001000 Time 0.021845 +2023-10-02 20:49:09,808 - Epoch: [42][ 960/ 1236] Overall Loss 0.261371 Objective Loss 0.261371 LR 0.001000 Time 0.021835 +2023-10-02 20:49:10,017 - Epoch: [42][ 970/ 1236] Overall Loss 0.261439 Objective Loss 0.261439 LR 0.001000 Time 0.021824 +2023-10-02 20:49:10,226 - Epoch: [42][ 980/ 1236] Overall Loss 0.261412 Objective Loss 0.261412 LR 0.001000 Time 0.021814 +2023-10-02 20:49:10,435 - Epoch: [42][ 990/ 1236] Overall Loss 0.261286 Objective Loss 0.261286 LR 0.001000 Time 0.021804 +2023-10-02 20:49:10,645 - Epoch: [42][ 1000/ 1236] Overall Loss 0.261365 Objective Loss 0.261365 LR 0.001000 Time 0.021795 +2023-10-02 20:49:10,853 - Epoch: [42][ 1010/ 1236] Overall Loss 0.261631 Objective Loss 0.261631 LR 0.001000 Time 0.021786 +2023-10-02 20:49:11,062 - Epoch: [42][ 1020/ 1236] Overall Loss 0.261411 Objective Loss 0.261411 LR 0.001000 Time 0.021776 +2023-10-02 20:49:11,269 - Epoch: [42][ 1030/ 1236] Overall Loss 0.261578 Objective Loss 0.261578 LR 0.001000 Time 0.021766 +2023-10-02 20:49:11,477 - Epoch: [42][ 1040/ 1236] Overall Loss 0.261852 Objective Loss 0.261852 LR 0.001000 Time 0.021756 +2023-10-02 20:49:11,684 - Epoch: [42][ 1050/ 1236] Overall Loss 0.261994 Objective Loss 0.261994 LR 0.001000 Time 0.021745 +2023-10-02 20:49:11,892 - Epoch: [42][ 1060/ 1236] Overall Loss 0.262163 Objective Loss 0.262163 LR 0.001000 Time 0.021735 +2023-10-02 20:49:12,099 - Epoch: [42][ 1070/ 1236] Overall Loss 0.262054 Objective Loss 0.262054 LR 0.001000 Time 0.021726 +2023-10-02 20:49:12,306 - Epoch: [42][ 1080/ 1236] Overall Loss 0.262154 Objective Loss 0.262154 LR 0.001000 Time 0.021716 +2023-10-02 20:49:12,511 - Epoch: [42][ 1090/ 1236] Overall Loss 0.262256 Objective Loss 0.262256 LR 0.001000 Time 0.021704 +2023-10-02 20:49:12,719 - Epoch: [42][ 1100/ 1236] Overall Loss 0.262092 Objective Loss 0.262092 LR 0.001000 Time 0.021696 +2023-10-02 20:49:12,927 - Epoch: [42][ 1110/ 1236] Overall Loss 0.262104 Objective Loss 0.262104 LR 0.001000 Time 0.021687 +2023-10-02 20:49:13,135 - Epoch: [42][ 1120/ 1236] Overall Loss 0.262125 Objective Loss 0.262125 LR 0.001000 Time 0.021679 +2023-10-02 20:49:13,341 - Epoch: [42][ 1130/ 1236] Overall Loss 0.262025 Objective Loss 0.262025 LR 0.001000 Time 0.021669 +2023-10-02 20:49:13,550 - Epoch: [42][ 1140/ 1236] Overall Loss 0.262448 Objective Loss 0.262448 LR 0.001000 Time 0.021662 +2023-10-02 20:49:13,755 - Epoch: [42][ 1150/ 1236] Overall Loss 0.262505 Objective Loss 0.262505 LR 0.001000 Time 0.021652 +2023-10-02 20:49:13,963 - Epoch: [42][ 1160/ 1236] Overall Loss 0.262646 Objective Loss 0.262646 LR 0.001000 Time 0.021644 +2023-10-02 20:49:14,170 - Epoch: [42][ 1170/ 1236] Overall Loss 0.262552 Objective Loss 0.262552 LR 0.001000 Time 0.021636 +2023-10-02 20:49:14,379 - Epoch: [42][ 1180/ 1236] Overall Loss 0.262426 Objective Loss 0.262426 LR 0.001000 Time 0.021629 +2023-10-02 20:49:14,586 - Epoch: [42][ 1190/ 1236] Overall Loss 0.262274 Objective Loss 0.262274 LR 0.001000 Time 0.021620 +2023-10-02 20:49:14,794 - Epoch: [42][ 1200/ 1236] Overall Loss 0.262218 Objective Loss 0.262218 LR 0.001000 Time 0.021613 +2023-10-02 20:49:15,001 - Epoch: [42][ 1210/ 1236] Overall Loss 0.262282 Objective Loss 0.262282 LR 0.001000 Time 0.021605 +2023-10-02 20:49:15,209 - Epoch: [42][ 1220/ 1236] Overall Loss 0.261970 Objective Loss 0.261970 LR 0.001000 Time 0.021598 +2023-10-02 20:49:15,469 - Epoch: [42][ 1230/ 1236] Overall Loss 0.261799 Objective Loss 0.261799 LR 0.001000 Time 0.021634 +2023-10-02 20:49:15,591 - Epoch: [42][ 1236/ 1236] Overall Loss 0.261774 Objective Loss 0.261774 Top1 86.150713 Top5 98.370672 LR 0.001000 Time 0.021627 +2023-10-02 20:49:15,716 - --- validate (epoch=42)----------- +2023-10-02 20:49:15,716 - 29943 samples (256 per mini-batch) +2023-10-02 20:49:16,182 - Epoch: [42][ 10/ 117] Loss 0.362495 Top1 82.734375 Top5 98.046875 +2023-10-02 20:49:16,334 - Epoch: [42][ 20/ 117] Loss 0.333032 Top1 83.320312 Top5 98.242188 +2023-10-02 20:49:16,485 - Epoch: [42][ 30/ 117] Loss 0.328988 Top1 83.111979 Top5 98.072917 +2023-10-02 20:49:16,638 - Epoch: [42][ 40/ 117] Loss 0.332122 Top1 83.027344 Top5 98.115234 +2023-10-02 20:49:16,788 - Epoch: [42][ 50/ 117] Loss 0.330370 Top1 83.140625 Top5 98.125000 +2023-10-02 20:49:16,940 - Epoch: [42][ 60/ 117] Loss 0.327359 Top1 83.235677 Top5 98.203125 +2023-10-02 20:49:17,089 - Epoch: [42][ 70/ 117] Loss 0.329117 Top1 83.281250 Top5 98.180804 +2023-10-02 20:49:17,241 - Epoch: [42][ 80/ 117] Loss 0.324912 Top1 83.525391 Top5 98.227539 +2023-10-02 20:49:17,398 - Epoch: [42][ 90/ 117] Loss 0.324755 Top1 83.563368 Top5 98.229167 +2023-10-02 20:49:17,555 - Epoch: [42][ 100/ 117] Loss 0.324276 Top1 83.562500 Top5 98.273438 +2023-10-02 20:49:17,720 - Epoch: [42][ 110/ 117] Loss 0.321660 Top1 83.618608 Top5 98.270597 +2023-10-02 20:49:17,810 - Epoch: [42][ 117/ 117] Loss 0.322415 Top1 83.622216 Top5 98.290084 +2023-10-02 20:49:17,941 - ==> Top1: 83.622 Top5: 98.290 Loss: 0.322 + +2023-10-02 20:49:17,942 - ==> Confusion: +[[ 904 0 3 0 10 6 0 0 10 83 2 0 0 2 10 0 2 1 1 0 16] + [ 0 1050 0 1 4 29 1 18 3 2 1 0 1 0 1 2 1 0 13 0 4] + [ 7 0 946 11 1 0 24 11 0 4 4 0 8 1 4 5 0 1 11 2 16] + [ 2 3 14 975 1 5 3 1 4 1 2 1 4 3 36 1 1 4 15 1 12] + [ 23 9 2 0 946 14 0 0 1 17 2 1 3 4 10 3 6 0 0 2 7] + [ 3 30 1 3 1 1010 0 20 3 5 0 2 0 18 5 0 0 0 0 4 11] + [ 0 5 34 0 0 1 1118 7 0 0 4 2 0 1 1 6 0 0 1 5 6] + [ 4 13 16 2 2 24 6 1070 2 3 5 3 3 8 1 0 0 0 41 7 8] + [ 14 2 0 0 2 4 2 2 970 54 6 0 6 11 14 0 0 1 1 0 0] + [ 83 0 0 0 3 3 0 0 34 959 1 0 0 23 5 0 0 0 0 3 5] + [ 2 2 7 13 1 4 4 0 19 3 950 2 1 15 7 1 0 0 10 1 11] + [ 0 1 0 0 1 19 0 3 0 0 0 931 32 12 0 2 1 13 0 11 9] + [ 0 0 0 4 1 11 1 1 1 1 2 36 967 1 6 7 2 13 2 3 9] + [ 1 0 2 0 1 8 0 0 6 13 5 8 1 1058 6 1 0 0 0 2 7] + [ 6 0 0 12 3 0 0 0 25 4 0 0 3 1 1031 0 2 2 6 0 6] + [ 0 0 1 0 8 0 2 1 0 0 0 8 15 1 1 1047 15 12 1 8 14] + [ 1 21 1 1 8 9 1 1 2 0 1 4 2 4 3 6 1069 1 0 9 17] + [ 0 0 3 2 0 1 3 0 4 1 1 6 29 0 2 4 1 974 0 0 7] + [ 3 9 4 11 2 0 0 22 6 0 1 0 2 0 14 1 0 0 982 1 10] + [ 0 2 5 0 1 11 9 11 0 1 1 8 4 1 1 1 7 1 3 1074 11] + [ 141 199 113 81 84 239 47 100 132 109 118 102 340 348 157 38 92 48 177 232 5008]] + +2023-10-02 20:49:17,943 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:49:17,943 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:49:17,949 - + +2023-10-02 20:49:17,949 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:49:18,964 - Epoch: [43][ 10/ 1236] Overall Loss 0.233867 Objective Loss 0.233867 LR 0.001000 Time 0.101403 +2023-10-02 20:49:19,174 - Epoch: [43][ 20/ 1236] Overall Loss 0.231038 Objective Loss 0.231038 LR 0.001000 Time 0.061200 +2023-10-02 20:49:19,383 - Epoch: [43][ 30/ 1236] Overall Loss 0.238371 Objective Loss 0.238371 LR 0.001000 Time 0.047749 +2023-10-02 20:49:19,593 - Epoch: [43][ 40/ 1236] Overall Loss 0.241194 Objective Loss 0.241194 LR 0.001000 Time 0.041055 +2023-10-02 20:49:19,799 - Epoch: [43][ 50/ 1236] Overall Loss 0.241664 Objective Loss 0.241664 LR 0.001000 Time 0.036936 +2023-10-02 20:49:20,010 - Epoch: [43][ 60/ 1236] Overall Loss 0.242554 Objective Loss 0.242554 LR 0.001000 Time 0.034281 +2023-10-02 20:49:20,218 - Epoch: [43][ 70/ 1236] Overall Loss 0.243780 Objective Loss 0.243780 LR 0.001000 Time 0.032341 +2023-10-02 20:49:20,428 - Epoch: [43][ 80/ 1236] Overall Loss 0.241353 Objective Loss 0.241353 LR 0.001000 Time 0.030921 +2023-10-02 20:49:20,637 - Epoch: [43][ 90/ 1236] Overall Loss 0.245457 Objective Loss 0.245457 LR 0.001000 Time 0.029783 +2023-10-02 20:49:20,847 - Epoch: [43][ 100/ 1236] Overall Loss 0.245421 Objective Loss 0.245421 LR 0.001000 Time 0.028907 +2023-10-02 20:49:21,055 - Epoch: [43][ 110/ 1236] Overall Loss 0.246661 Objective Loss 0.246661 LR 0.001000 Time 0.028158 +2023-10-02 20:49:21,265 - Epoch: [43][ 120/ 1236] Overall Loss 0.249272 Objective Loss 0.249272 LR 0.001000 Time 0.027560 +2023-10-02 20:49:21,474 - Epoch: [43][ 130/ 1236] Overall Loss 0.247399 Objective Loss 0.247399 LR 0.001000 Time 0.027035 +2023-10-02 20:49:21,684 - Epoch: [43][ 140/ 1236] Overall Loss 0.248428 Objective Loss 0.248428 LR 0.001000 Time 0.026601 +2023-10-02 20:49:21,892 - Epoch: [43][ 150/ 1236] Overall Loss 0.247665 Objective Loss 0.247665 LR 0.001000 Time 0.026206 +2023-10-02 20:49:22,102 - Epoch: [43][ 160/ 1236] Overall Loss 0.248239 Objective Loss 0.248239 LR 0.001000 Time 0.025879 +2023-10-02 20:49:22,310 - Epoch: [43][ 170/ 1236] Overall Loss 0.248293 Objective Loss 0.248293 LR 0.001000 Time 0.025571 +2023-10-02 20:49:22,518 - Epoch: [43][ 180/ 1236] Overall Loss 0.248456 Objective Loss 0.248456 LR 0.001000 Time 0.025304 +2023-10-02 20:49:22,726 - Epoch: [43][ 190/ 1236] Overall Loss 0.247891 Objective Loss 0.247891 LR 0.001000 Time 0.025057 +2023-10-02 20:49:22,938 - Epoch: [43][ 200/ 1236] Overall Loss 0.248354 Objective Loss 0.248354 LR 0.001000 Time 0.024861 +2023-10-02 20:49:23,145 - Epoch: [43][ 210/ 1236] Overall Loss 0.248289 Objective Loss 0.248289 LR 0.001000 Time 0.024662 +2023-10-02 20:49:23,355 - Epoch: [43][ 220/ 1236] Overall Loss 0.249175 Objective Loss 0.249175 LR 0.001000 Time 0.024495 +2023-10-02 20:49:23,563 - Epoch: [43][ 230/ 1236] Overall Loss 0.249564 Objective Loss 0.249564 LR 0.001000 Time 0.024328 +2023-10-02 20:49:23,773 - Epoch: [43][ 240/ 1236] Overall Loss 0.250569 Objective Loss 0.250569 LR 0.001000 Time 0.024187 +2023-10-02 20:49:23,981 - Epoch: [43][ 250/ 1236] Overall Loss 0.249811 Objective Loss 0.249811 LR 0.001000 Time 0.024048 +2023-10-02 20:49:24,191 - Epoch: [43][ 260/ 1236] Overall Loss 0.250287 Objective Loss 0.250287 LR 0.001000 Time 0.023930 +2023-10-02 20:49:24,400 - Epoch: [43][ 270/ 1236] Overall Loss 0.250077 Objective Loss 0.250077 LR 0.001000 Time 0.023809 +2023-10-02 20:49:24,608 - Epoch: [43][ 280/ 1236] Overall Loss 0.250153 Objective Loss 0.250153 LR 0.001000 Time 0.023704 +2023-10-02 20:49:24,816 - Epoch: [43][ 290/ 1236] Overall Loss 0.250395 Objective Loss 0.250395 LR 0.001000 Time 0.023603 +2023-10-02 20:49:25,027 - Epoch: [43][ 300/ 1236] Overall Loss 0.250754 Objective Loss 0.250754 LR 0.001000 Time 0.023516 +2023-10-02 20:49:25,235 - Epoch: [43][ 310/ 1236] Overall Loss 0.251252 Objective Loss 0.251252 LR 0.001000 Time 0.023423 +2023-10-02 20:49:25,445 - Epoch: [43][ 320/ 1236] Overall Loss 0.251593 Objective Loss 0.251593 LR 0.001000 Time 0.023347 +2023-10-02 20:49:25,653 - Epoch: [43][ 330/ 1236] Overall Loss 0.251153 Objective Loss 0.251153 LR 0.001000 Time 0.023265 +2023-10-02 20:49:25,863 - Epoch: [43][ 340/ 1236] Overall Loss 0.251065 Objective Loss 0.251065 LR 0.001000 Time 0.023199 +2023-10-02 20:49:26,072 - Epoch: [43][ 350/ 1236] Overall Loss 0.251127 Objective Loss 0.251127 LR 0.001000 Time 0.023131 +2023-10-02 20:49:26,282 - Epoch: [43][ 360/ 1236] Overall Loss 0.251441 Objective Loss 0.251441 LR 0.001000 Time 0.023072 +2023-10-02 20:49:26,490 - Epoch: [43][ 370/ 1236] Overall Loss 0.251497 Objective Loss 0.251497 LR 0.001000 Time 0.023007 +2023-10-02 20:49:26,701 - Epoch: [43][ 380/ 1236] Overall Loss 0.251764 Objective Loss 0.251764 LR 0.001000 Time 0.022955 +2023-10-02 20:49:26,909 - Epoch: [43][ 390/ 1236] Overall Loss 0.251239 Objective Loss 0.251239 LR 0.001000 Time 0.022897 +2023-10-02 20:49:27,120 - Epoch: [43][ 400/ 1236] Overall Loss 0.251180 Objective Loss 0.251180 LR 0.001000 Time 0.022850 +2023-10-02 20:49:27,328 - Epoch: [43][ 410/ 1236] Overall Loss 0.251123 Objective Loss 0.251123 LR 0.001000 Time 0.022798 +2023-10-02 20:49:27,539 - Epoch: [43][ 420/ 1236] Overall Loss 0.251296 Objective Loss 0.251296 LR 0.001000 Time 0.022755 +2023-10-02 20:49:27,747 - Epoch: [43][ 430/ 1236] Overall Loss 0.251249 Objective Loss 0.251249 LR 0.001000 Time 0.022708 +2023-10-02 20:49:27,957 - Epoch: [43][ 440/ 1236] Overall Loss 0.251832 Objective Loss 0.251832 LR 0.001000 Time 0.022668 +2023-10-02 20:49:28,166 - Epoch: [43][ 450/ 1236] Overall Loss 0.252027 Objective Loss 0.252027 LR 0.001000 Time 0.022625 +2023-10-02 20:49:28,377 - Epoch: [43][ 460/ 1236] Overall Loss 0.252835 Objective Loss 0.252835 LR 0.001000 Time 0.022590 +2023-10-02 20:49:28,585 - Epoch: [43][ 470/ 1236] Overall Loss 0.253140 Objective Loss 0.253140 LR 0.001000 Time 0.022550 +2023-10-02 20:49:28,795 - Epoch: [43][ 480/ 1236] Overall Loss 0.253878 Objective Loss 0.253878 LR 0.001000 Time 0.022517 +2023-10-02 20:49:29,004 - Epoch: [43][ 490/ 1236] Overall Loss 0.254679 Objective Loss 0.254679 LR 0.001000 Time 0.022481 +2023-10-02 20:49:29,215 - Epoch: [43][ 500/ 1236] Overall Loss 0.254894 Objective Loss 0.254894 LR 0.001000 Time 0.022453 +2023-10-02 20:49:29,424 - Epoch: [43][ 510/ 1236] Overall Loss 0.255126 Objective Loss 0.255126 LR 0.001000 Time 0.022419 +2023-10-02 20:49:29,635 - Epoch: [43][ 520/ 1236] Overall Loss 0.255070 Objective Loss 0.255070 LR 0.001000 Time 0.022392 +2023-10-02 20:49:29,844 - Epoch: [43][ 530/ 1236] Overall Loss 0.255002 Objective Loss 0.255002 LR 0.001000 Time 0.022363 +2023-10-02 20:49:30,054 - Epoch: [43][ 540/ 1236] Overall Loss 0.254971 Objective Loss 0.254971 LR 0.001000 Time 0.022339 +2023-10-02 20:49:30,261 - Epoch: [43][ 550/ 1236] Overall Loss 0.255113 Objective Loss 0.255113 LR 0.001000 Time 0.022307 +2023-10-02 20:49:30,472 - Epoch: [43][ 560/ 1236] Overall Loss 0.255297 Objective Loss 0.255297 LR 0.001000 Time 0.022284 +2023-10-02 20:49:30,681 - Epoch: [43][ 570/ 1236] Overall Loss 0.255396 Objective Loss 0.255396 LR 0.001000 Time 0.022257 +2023-10-02 20:49:30,890 - Epoch: [43][ 580/ 1236] Overall Loss 0.255761 Objective Loss 0.255761 LR 0.001000 Time 0.022233 +2023-10-02 20:49:31,099 - Epoch: [43][ 590/ 1236] Overall Loss 0.255951 Objective Loss 0.255951 LR 0.001000 Time 0.022208 +2023-10-02 20:49:31,310 - Epoch: [43][ 600/ 1236] Overall Loss 0.256157 Objective Loss 0.256157 LR 0.001000 Time 0.022188 +2023-10-02 20:49:31,518 - Epoch: [43][ 610/ 1236] Overall Loss 0.256314 Objective Loss 0.256314 LR 0.001000 Time 0.022164 +2023-10-02 20:49:31,727 - Epoch: [43][ 620/ 1236] Overall Loss 0.256411 Objective Loss 0.256411 LR 0.001000 Time 0.022143 +2023-10-02 20:49:31,934 - Epoch: [43][ 630/ 1236] Overall Loss 0.256629 Objective Loss 0.256629 LR 0.001000 Time 0.022119 +2023-10-02 20:49:32,144 - Epoch: [43][ 640/ 1236] Overall Loss 0.256575 Objective Loss 0.256575 LR 0.001000 Time 0.022101 +2023-10-02 20:49:32,352 - Epoch: [43][ 650/ 1236] Overall Loss 0.256378 Objective Loss 0.256378 LR 0.001000 Time 0.022080 +2023-10-02 20:49:32,562 - Epoch: [43][ 660/ 1236] Overall Loss 0.256768 Objective Loss 0.256768 LR 0.001000 Time 0.022064 +2023-10-02 20:49:32,769 - Epoch: [43][ 670/ 1236] Overall Loss 0.257264 Objective Loss 0.257264 LR 0.001000 Time 0.022042 +2023-10-02 20:49:32,980 - Epoch: [43][ 680/ 1236] Overall Loss 0.257274 Objective Loss 0.257274 LR 0.001000 Time 0.022028 +2023-10-02 20:49:33,187 - Epoch: [43][ 690/ 1236] Overall Loss 0.257146 Objective Loss 0.257146 LR 0.001000 Time 0.022006 +2023-10-02 20:49:33,398 - Epoch: [43][ 700/ 1236] Overall Loss 0.257067 Objective Loss 0.257067 LR 0.001000 Time 0.021992 +2023-10-02 20:49:33,607 - Epoch: [43][ 710/ 1236] Overall Loss 0.257217 Objective Loss 0.257217 LR 0.001000 Time 0.021974 +2023-10-02 20:49:33,815 - Epoch: [43][ 720/ 1236] Overall Loss 0.257090 Objective Loss 0.257090 LR 0.001000 Time 0.021958 +2023-10-02 20:49:34,028 - Epoch: [43][ 730/ 1236] Overall Loss 0.257245 Objective Loss 0.257245 LR 0.001000 Time 0.021946 +2023-10-02 20:49:34,240 - Epoch: [43][ 740/ 1236] Overall Loss 0.257960 Objective Loss 0.257960 LR 0.001000 Time 0.021936 +2023-10-02 20:49:34,451 - Epoch: [43][ 750/ 1236] Overall Loss 0.258321 Objective Loss 0.258321 LR 0.001000 Time 0.021922 +2023-10-02 20:49:34,662 - Epoch: [43][ 760/ 1236] Overall Loss 0.258380 Objective Loss 0.258380 LR 0.001000 Time 0.021911 +2023-10-02 20:49:34,873 - Epoch: [43][ 770/ 1236] Overall Loss 0.258102 Objective Loss 0.258102 LR 0.001000 Time 0.021901 +2023-10-02 20:49:35,086 - Epoch: [43][ 780/ 1236] Overall Loss 0.258234 Objective Loss 0.258234 LR 0.001000 Time 0.021892 +2023-10-02 20:49:35,299 - Epoch: [43][ 790/ 1236] Overall Loss 0.258106 Objective Loss 0.258106 LR 0.001000 Time 0.021884 +2023-10-02 20:49:35,511 - Epoch: [43][ 800/ 1236] Overall Loss 0.258218 Objective Loss 0.258218 LR 0.001000 Time 0.021875 +2023-10-02 20:49:35,724 - Epoch: [43][ 810/ 1236] Overall Loss 0.258525 Objective Loss 0.258525 LR 0.001000 Time 0.021866 +2023-10-02 20:49:35,935 - Epoch: [43][ 820/ 1236] Overall Loss 0.258696 Objective Loss 0.258696 LR 0.001000 Time 0.021856 +2023-10-02 20:49:36,148 - Epoch: [43][ 830/ 1236] Overall Loss 0.258613 Objective Loss 0.258613 LR 0.001000 Time 0.021849 +2023-10-02 20:49:36,360 - Epoch: [43][ 840/ 1236] Overall Loss 0.258742 Objective Loss 0.258742 LR 0.001000 Time 0.021841 +2023-10-02 20:49:36,573 - Epoch: [43][ 850/ 1236] Overall Loss 0.258784 Objective Loss 0.258784 LR 0.001000 Time 0.021834 +2023-10-02 20:49:36,784 - Epoch: [43][ 860/ 1236] Overall Loss 0.258617 Objective Loss 0.258617 LR 0.001000 Time 0.021825 +2023-10-02 20:49:36,997 - Epoch: [43][ 870/ 1236] Overall Loss 0.258511 Objective Loss 0.258511 LR 0.001000 Time 0.021817 +2023-10-02 20:49:37,209 - Epoch: [43][ 880/ 1236] Overall Loss 0.258610 Objective Loss 0.258610 LR 0.001000 Time 0.021810 +2023-10-02 20:49:37,422 - Epoch: [43][ 890/ 1236] Overall Loss 0.258648 Objective Loss 0.258648 LR 0.001000 Time 0.021802 +2023-10-02 20:49:37,634 - Epoch: [43][ 900/ 1236] Overall Loss 0.258771 Objective Loss 0.258771 LR 0.001000 Time 0.021795 +2023-10-02 20:49:37,846 - Epoch: [43][ 910/ 1236] Overall Loss 0.258733 Objective Loss 0.258733 LR 0.001000 Time 0.021788 +2023-10-02 20:49:38,059 - Epoch: [43][ 920/ 1236] Overall Loss 0.258703 Objective Loss 0.258703 LR 0.001000 Time 0.021783 +2023-10-02 20:49:38,272 - Epoch: [43][ 930/ 1236] Overall Loss 0.258546 Objective Loss 0.258546 LR 0.001000 Time 0.021776 +2023-10-02 20:49:38,485 - Epoch: [43][ 940/ 1236] Overall Loss 0.258462 Objective Loss 0.258462 LR 0.001000 Time 0.021770 +2023-10-02 20:49:38,698 - Epoch: [43][ 950/ 1236] Overall Loss 0.258730 Objective Loss 0.258730 LR 0.001000 Time 0.021764 +2023-10-02 20:49:38,911 - Epoch: [43][ 960/ 1236] Overall Loss 0.258589 Objective Loss 0.258589 LR 0.001000 Time 0.021760 +2023-10-02 20:49:39,123 - Epoch: [43][ 970/ 1236] Overall Loss 0.258372 Objective Loss 0.258372 LR 0.001000 Time 0.021753 +2023-10-02 20:49:39,336 - Epoch: [43][ 980/ 1236] Overall Loss 0.258693 Objective Loss 0.258693 LR 0.001000 Time 0.021749 +2023-10-02 20:49:39,548 - Epoch: [43][ 990/ 1236] Overall Loss 0.258605 Objective Loss 0.258605 LR 0.001000 Time 0.021742 +2023-10-02 20:49:39,761 - Epoch: [43][ 1000/ 1236] Overall Loss 0.258709 Objective Loss 0.258709 LR 0.001000 Time 0.021737 +2023-10-02 20:49:39,974 - Epoch: [43][ 1010/ 1236] Overall Loss 0.258805 Objective Loss 0.258805 LR 0.001000 Time 0.021732 +2023-10-02 20:49:40,185 - Epoch: [43][ 1020/ 1236] Overall Loss 0.258842 Objective Loss 0.258842 LR 0.001000 Time 0.021726 +2023-10-02 20:49:40,397 - Epoch: [43][ 1030/ 1236] Overall Loss 0.258820 Objective Loss 0.258820 LR 0.001000 Time 0.021720 +2023-10-02 20:49:40,610 - Epoch: [43][ 1040/ 1236] Overall Loss 0.258767 Objective Loss 0.258767 LR 0.001000 Time 0.021715 +2023-10-02 20:49:40,821 - Epoch: [43][ 1050/ 1236] Overall Loss 0.258758 Objective Loss 0.258758 LR 0.001000 Time 0.021709 +2023-10-02 20:49:41,034 - Epoch: [43][ 1060/ 1236] Overall Loss 0.258708 Objective Loss 0.258708 LR 0.001000 Time 0.021703 +2023-10-02 20:49:41,247 - Epoch: [43][ 1070/ 1236] Overall Loss 0.258841 Objective Loss 0.258841 LR 0.001000 Time 0.021699 +2023-10-02 20:49:41,458 - Epoch: [43][ 1080/ 1236] Overall Loss 0.259012 Objective Loss 0.259012 LR 0.001000 Time 0.021693 +2023-10-02 20:49:41,670 - Epoch: [43][ 1090/ 1236] Overall Loss 0.259027 Objective Loss 0.259027 LR 0.001000 Time 0.021688 +2023-10-02 20:49:41,882 - Epoch: [43][ 1100/ 1236] Overall Loss 0.259116 Objective Loss 0.259116 LR 0.001000 Time 0.021684 +2023-10-02 20:49:42,094 - Epoch: [43][ 1110/ 1236] Overall Loss 0.259217 Objective Loss 0.259217 LR 0.001000 Time 0.021679 +2023-10-02 20:49:42,305 - Epoch: [43][ 1120/ 1236] Overall Loss 0.259110 Objective Loss 0.259110 LR 0.001000 Time 0.021674 +2023-10-02 20:49:42,518 - Epoch: [43][ 1130/ 1236] Overall Loss 0.259165 Objective Loss 0.259165 LR 0.001000 Time 0.021670 +2023-10-02 20:49:42,729 - Epoch: [43][ 1140/ 1236] Overall Loss 0.259167 Objective Loss 0.259167 LR 0.001000 Time 0.021665 +2023-10-02 20:49:42,942 - Epoch: [43][ 1150/ 1236] Overall Loss 0.259276 Objective Loss 0.259276 LR 0.001000 Time 0.021661 +2023-10-02 20:49:43,154 - Epoch: [43][ 1160/ 1236] Overall Loss 0.259270 Objective Loss 0.259270 LR 0.001000 Time 0.021657 +2023-10-02 20:49:43,367 - Epoch: [43][ 1170/ 1236] Overall Loss 0.259336 Objective Loss 0.259336 LR 0.001000 Time 0.021652 +2023-10-02 20:49:43,577 - Epoch: [43][ 1180/ 1236] Overall Loss 0.259199 Objective Loss 0.259199 LR 0.001000 Time 0.021646 +2023-10-02 20:49:43,788 - Epoch: [43][ 1190/ 1236] Overall Loss 0.259265 Objective Loss 0.259265 LR 0.001000 Time 0.021642 +2023-10-02 20:49:44,000 - Epoch: [43][ 1200/ 1236] Overall Loss 0.259456 Objective Loss 0.259456 LR 0.001000 Time 0.021637 +2023-10-02 20:49:44,211 - Epoch: [43][ 1210/ 1236] Overall Loss 0.259518 Objective Loss 0.259518 LR 0.001000 Time 0.021632 +2023-10-02 20:49:44,425 - Epoch: [43][ 1220/ 1236] Overall Loss 0.259678 Objective Loss 0.259678 LR 0.001000 Time 0.021629 +2023-10-02 20:49:44,689 - Epoch: [43][ 1230/ 1236] Overall Loss 0.259809 Objective Loss 0.259809 LR 0.001000 Time 0.021668 +2023-10-02 20:49:44,812 - Epoch: [43][ 1236/ 1236] Overall Loss 0.259816 Objective Loss 0.259816 Top1 85.947047 Top5 97.352342 LR 0.001000 Time 0.021662 +2023-10-02 20:49:44,950 - --- validate (epoch=43)----------- +2023-10-02 20:49:44,950 - 29943 samples (256 per mini-batch) +2023-10-02 20:49:45,435 - Epoch: [43][ 10/ 117] Loss 0.322454 Top1 83.554688 Top5 98.164062 +2023-10-02 20:49:45,582 - Epoch: [43][ 20/ 117] Loss 0.300853 Top1 84.218750 Top5 98.222656 +2023-10-02 20:49:45,728 - Epoch: [43][ 30/ 117] Loss 0.315109 Top1 84.127604 Top5 98.033854 +2023-10-02 20:49:45,874 - Epoch: [43][ 40/ 117] Loss 0.314283 Top1 84.023438 Top5 98.027344 +2023-10-02 20:49:46,021 - Epoch: [43][ 50/ 117] Loss 0.312339 Top1 84.054688 Top5 98.078125 +2023-10-02 20:49:46,168 - Epoch: [43][ 60/ 117] Loss 0.321859 Top1 83.880208 Top5 98.033854 +2023-10-02 20:49:46,314 - Epoch: [43][ 70/ 117] Loss 0.324998 Top1 83.828125 Top5 98.007812 +2023-10-02 20:49:46,461 - Epoch: [43][ 80/ 117] Loss 0.324016 Top1 83.867188 Top5 98.007812 +2023-10-02 20:49:46,608 - Epoch: [43][ 90/ 117] Loss 0.321821 Top1 83.953993 Top5 98.051215 +2023-10-02 20:49:46,755 - Epoch: [43][ 100/ 117] Loss 0.319191 Top1 83.988281 Top5 98.062500 +2023-10-02 20:49:46,910 - Epoch: [43][ 110/ 117] Loss 0.317986 Top1 84.073153 Top5 98.078835 +2023-10-02 20:49:47,000 - Epoch: [43][ 117/ 117] Loss 0.317521 Top1 84.053034 Top5 98.083024 +2023-10-02 20:49:47,142 - ==> Top1: 84.053 Top5: 98.083 Loss: 0.318 + +2023-10-02 20:49:47,142 - ==> Confusion: +[[ 925 1 2 0 12 4 0 0 8 61 2 3 1 3 4 0 4 1 0 1 18] + [ 0 1056 0 2 5 21 0 16 1 0 3 1 1 0 1 4 3 0 7 2 8] + [ 5 0 917 29 2 0 47 8 0 0 5 1 9 5 0 5 0 1 11 0 11] + [ 4 3 3 988 0 5 1 2 3 0 10 1 7 4 19 1 4 4 17 0 13] + [ 28 4 0 1 965 5 0 0 0 7 0 1 1 4 12 7 8 0 0 2 5] + [ 3 53 0 1 1 966 2 24 5 4 8 3 6 9 5 1 4 0 3 7 11] + [ 0 3 15 0 0 0 1138 5 0 0 8 0 0 0 0 3 0 0 3 10 6] + [ 5 24 13 0 2 21 3 1049 1 1 5 9 4 8 1 0 0 0 55 8 9] + [ 18 1 0 1 0 3 0 0 989 29 16 2 6 9 6 0 1 1 4 3 0] + [ 94 3 1 0 4 2 0 0 45 929 1 0 1 24 5 1 1 1 0 1 6] + [ 1 2 5 6 1 1 4 1 8 0 982 3 0 13 7 0 0 3 6 2 8] + [ 0 0 3 0 0 8 0 6 0 0 1 936 36 10 0 4 3 14 0 9 5] + [ 0 1 3 6 0 2 2 1 0 0 2 25 977 3 4 5 4 10 2 3 18] + [ 0 0 1 0 2 9 0 0 17 5 16 5 1 1037 4 1 3 1 0 1 16] + [ 12 0 1 17 3 1 0 0 35 1 4 0 4 5 998 0 3 1 3 0 13] + [ 0 0 2 0 4 0 1 0 1 0 0 8 12 1 0 1072 17 6 1 3 6] + [ 0 14 1 0 5 6 1 0 1 0 0 4 1 1 5 6 1098 1 2 3 12] + [ 0 1 2 4 0 0 3 0 1 0 1 5 25 2 3 8 0 974 2 1 6] + [ 2 5 1 17 1 0 0 12 7 0 5 3 4 0 12 1 0 0 987 0 11] + [ 0 3 2 2 0 2 7 7 0 0 1 7 7 3 2 2 3 0 3 1095 6] + [ 143 195 104 84 102 148 58 88 163 83 237 84 310 280 130 70 138 59 145 194 5090]] + +2023-10-02 20:49:47,144 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:49:47,144 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:49:47,150 - + +2023-10-02 20:49:47,150 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:49:48,282 - Epoch: [44][ 10/ 1236] Overall Loss 0.269739 Objective Loss 0.269739 LR 0.001000 Time 0.113183 +2023-10-02 20:49:48,490 - Epoch: [44][ 20/ 1236] Overall Loss 0.247689 Objective Loss 0.247689 LR 0.001000 Time 0.066982 +2023-10-02 20:49:48,697 - Epoch: [44][ 30/ 1236] Overall Loss 0.245720 Objective Loss 0.245720 LR 0.001000 Time 0.051500 +2023-10-02 20:49:48,905 - Epoch: [44][ 40/ 1236] Overall Loss 0.248479 Objective Loss 0.248479 LR 0.001000 Time 0.043813 +2023-10-02 20:49:49,112 - Epoch: [44][ 50/ 1236] Overall Loss 0.248733 Objective Loss 0.248733 LR 0.001000 Time 0.039156 +2023-10-02 20:49:49,320 - Epoch: [44][ 60/ 1236] Overall Loss 0.251625 Objective Loss 0.251625 LR 0.001000 Time 0.036088 +2023-10-02 20:49:49,526 - Epoch: [44][ 70/ 1236] Overall Loss 0.247483 Objective Loss 0.247483 LR 0.001000 Time 0.033863 +2023-10-02 20:49:49,734 - Epoch: [44][ 80/ 1236] Overall Loss 0.246884 Objective Loss 0.246884 LR 0.001000 Time 0.032226 +2023-10-02 20:49:49,941 - Epoch: [44][ 90/ 1236] Overall Loss 0.246701 Objective Loss 0.246701 LR 0.001000 Time 0.030926 +2023-10-02 20:49:50,149 - Epoch: [44][ 100/ 1236] Overall Loss 0.248030 Objective Loss 0.248030 LR 0.001000 Time 0.029913 +2023-10-02 20:49:50,356 - Epoch: [44][ 110/ 1236] Overall Loss 0.250556 Objective Loss 0.250556 LR 0.001000 Time 0.029070 +2023-10-02 20:49:50,564 - Epoch: [44][ 120/ 1236] Overall Loss 0.250627 Objective Loss 0.250627 LR 0.001000 Time 0.028379 +2023-10-02 20:49:50,771 - Epoch: [44][ 130/ 1236] Overall Loss 0.252624 Objective Loss 0.252624 LR 0.001000 Time 0.027783 +2023-10-02 20:49:50,978 - Epoch: [44][ 140/ 1236] Overall Loss 0.251874 Objective Loss 0.251874 LR 0.001000 Time 0.027279 +2023-10-02 20:49:51,186 - Epoch: [44][ 150/ 1236] Overall Loss 0.251134 Objective Loss 0.251134 LR 0.001000 Time 0.026841 +2023-10-02 20:49:51,392 - Epoch: [44][ 160/ 1236] Overall Loss 0.251675 Objective Loss 0.251675 LR 0.001000 Time 0.026453 +2023-10-02 20:49:51,600 - Epoch: [44][ 170/ 1236] Overall Loss 0.252994 Objective Loss 0.252994 LR 0.001000 Time 0.026116 +2023-10-02 20:49:51,807 - Epoch: [44][ 180/ 1236] Overall Loss 0.252585 Objective Loss 0.252585 LR 0.001000 Time 0.025813 +2023-10-02 20:49:52,015 - Epoch: [44][ 190/ 1236] Overall Loss 0.252697 Objective Loss 0.252697 LR 0.001000 Time 0.025547 +2023-10-02 20:49:52,222 - Epoch: [44][ 200/ 1236] Overall Loss 0.250961 Objective Loss 0.250961 LR 0.001000 Time 0.025304 +2023-10-02 20:49:52,429 - Epoch: [44][ 210/ 1236] Overall Loss 0.251986 Objective Loss 0.251986 LR 0.001000 Time 0.025086 +2023-10-02 20:49:52,636 - Epoch: [44][ 220/ 1236] Overall Loss 0.253053 Objective Loss 0.253053 LR 0.001000 Time 0.024884 +2023-10-02 20:49:52,844 - Epoch: [44][ 230/ 1236] Overall Loss 0.253063 Objective Loss 0.253063 LR 0.001000 Time 0.024702 +2023-10-02 20:49:53,050 - Epoch: [44][ 240/ 1236] Overall Loss 0.253584 Objective Loss 0.253584 LR 0.001000 Time 0.024534 +2023-10-02 20:49:53,258 - Epoch: [44][ 250/ 1236] Overall Loss 0.253388 Objective Loss 0.253388 LR 0.001000 Time 0.024383 +2023-10-02 20:49:53,465 - Epoch: [44][ 260/ 1236] Overall Loss 0.253881 Objective Loss 0.253881 LR 0.001000 Time 0.024239 +2023-10-02 20:49:53,673 - Epoch: [44][ 270/ 1236] Overall Loss 0.253439 Objective Loss 0.253439 LR 0.001000 Time 0.024109 +2023-10-02 20:49:53,880 - Epoch: [44][ 280/ 1236] Overall Loss 0.253703 Objective Loss 0.253703 LR 0.001000 Time 0.023986 +2023-10-02 20:49:54,087 - Epoch: [44][ 290/ 1236] Overall Loss 0.254630 Objective Loss 0.254630 LR 0.001000 Time 0.023875 +2023-10-02 20:49:54,294 - Epoch: [44][ 300/ 1236] Overall Loss 0.255500 Objective Loss 0.255500 LR 0.001000 Time 0.023767 +2023-10-02 20:49:54,502 - Epoch: [44][ 310/ 1236] Overall Loss 0.255564 Objective Loss 0.255564 LR 0.001000 Time 0.023671 +2023-10-02 20:49:54,709 - Epoch: [44][ 320/ 1236] Overall Loss 0.255964 Objective Loss 0.255964 LR 0.001000 Time 0.023576 +2023-10-02 20:49:54,917 - Epoch: [44][ 330/ 1236] Overall Loss 0.255021 Objective Loss 0.255021 LR 0.001000 Time 0.023489 +2023-10-02 20:49:55,123 - Epoch: [44][ 340/ 1236] Overall Loss 0.255476 Objective Loss 0.255476 LR 0.001000 Time 0.023406 +2023-10-02 20:49:55,332 - Epoch: [44][ 350/ 1236] Overall Loss 0.255237 Objective Loss 0.255237 LR 0.001000 Time 0.023331 +2023-10-02 20:49:55,538 - Epoch: [44][ 360/ 1236] Overall Loss 0.254503 Objective Loss 0.254503 LR 0.001000 Time 0.023257 +2023-10-02 20:49:55,747 - Epoch: [44][ 370/ 1236] Overall Loss 0.254957 Objective Loss 0.254957 LR 0.001000 Time 0.023190 +2023-10-02 20:49:55,956 - Epoch: [44][ 380/ 1236] Overall Loss 0.254226 Objective Loss 0.254226 LR 0.001000 Time 0.023129 +2023-10-02 20:49:56,163 - Epoch: [44][ 390/ 1236] Overall Loss 0.254515 Objective Loss 0.254515 LR 0.001000 Time 0.023066 +2023-10-02 20:49:56,371 - Epoch: [44][ 400/ 1236] Overall Loss 0.254044 Objective Loss 0.254044 LR 0.001000 Time 0.023008 +2023-10-02 20:49:56,579 - Epoch: [44][ 410/ 1236] Overall Loss 0.254648 Objective Loss 0.254648 LR 0.001000 Time 0.022951 +2023-10-02 20:49:56,787 - Epoch: [44][ 420/ 1236] Overall Loss 0.255104 Objective Loss 0.255104 LR 0.001000 Time 0.022898 +2023-10-02 20:49:56,995 - Epoch: [44][ 430/ 1236] Overall Loss 0.254080 Objective Loss 0.254080 LR 0.001000 Time 0.022847 +2023-10-02 20:49:57,204 - Epoch: [44][ 440/ 1236] Overall Loss 0.253588 Objective Loss 0.253588 LR 0.001000 Time 0.022801 +2023-10-02 20:49:57,412 - Epoch: [44][ 450/ 1236] Overall Loss 0.254209 Objective Loss 0.254209 LR 0.001000 Time 0.022754 +2023-10-02 20:49:57,620 - Epoch: [44][ 460/ 1236] Overall Loss 0.254055 Objective Loss 0.254055 LR 0.001000 Time 0.022710 +2023-10-02 20:49:57,828 - Epoch: [44][ 470/ 1236] Overall Loss 0.254454 Objective Loss 0.254454 LR 0.001000 Time 0.022666 +2023-10-02 20:49:58,037 - Epoch: [44][ 480/ 1236] Overall Loss 0.255244 Objective Loss 0.255244 LR 0.001000 Time 0.022627 +2023-10-02 20:49:58,245 - Epoch: [44][ 490/ 1236] Overall Loss 0.255020 Objective Loss 0.255020 LR 0.001000 Time 0.022588 +2023-10-02 20:49:58,453 - Epoch: [44][ 500/ 1236] Overall Loss 0.254842 Objective Loss 0.254842 LR 0.001000 Time 0.022552 +2023-10-02 20:49:58,662 - Epoch: [44][ 510/ 1236] Overall Loss 0.255598 Objective Loss 0.255598 LR 0.001000 Time 0.022515 +2023-10-02 20:49:58,870 - Epoch: [44][ 520/ 1236] Overall Loss 0.255363 Objective Loss 0.255363 LR 0.001000 Time 0.022481 +2023-10-02 20:49:59,078 - Epoch: [44][ 530/ 1236] Overall Loss 0.255706 Objective Loss 0.255706 LR 0.001000 Time 0.022447 +2023-10-02 20:49:59,287 - Epoch: [44][ 540/ 1236] Overall Loss 0.255899 Objective Loss 0.255899 LR 0.001000 Time 0.022417 +2023-10-02 20:49:59,495 - Epoch: [44][ 550/ 1236] Overall Loss 0.256405 Objective Loss 0.256405 LR 0.001000 Time 0.022385 +2023-10-02 20:49:59,703 - Epoch: [44][ 560/ 1236] Overall Loss 0.255842 Objective Loss 0.255842 LR 0.001000 Time 0.022356 +2023-10-02 20:49:59,911 - Epoch: [44][ 570/ 1236] Overall Loss 0.256416 Objective Loss 0.256416 LR 0.001000 Time 0.022327 +2023-10-02 20:50:00,120 - Epoch: [44][ 580/ 1236] Overall Loss 0.256250 Objective Loss 0.256250 LR 0.001000 Time 0.022300 +2023-10-02 20:50:00,328 - Epoch: [44][ 590/ 1236] Overall Loss 0.256702 Objective Loss 0.256702 LR 0.001000 Time 0.022273 +2023-10-02 20:50:00,536 - Epoch: [44][ 600/ 1236] Overall Loss 0.256891 Objective Loss 0.256891 LR 0.001000 Time 0.022248 +2023-10-02 20:50:00,744 - Epoch: [44][ 610/ 1236] Overall Loss 0.256803 Objective Loss 0.256803 LR 0.001000 Time 0.022222 +2023-10-02 20:50:00,954 - Epoch: [44][ 620/ 1236] Overall Loss 0.256824 Objective Loss 0.256824 LR 0.001000 Time 0.022201 +2023-10-02 20:50:01,161 - Epoch: [44][ 630/ 1236] Overall Loss 0.256903 Objective Loss 0.256903 LR 0.001000 Time 0.022176 +2023-10-02 20:50:01,369 - Epoch: [44][ 640/ 1236] Overall Loss 0.257156 Objective Loss 0.257156 LR 0.001000 Time 0.022154 +2023-10-02 20:50:01,577 - Epoch: [44][ 650/ 1236] Overall Loss 0.257338 Objective Loss 0.257338 LR 0.001000 Time 0.022131 +2023-10-02 20:50:01,785 - Epoch: [44][ 660/ 1236] Overall Loss 0.257516 Objective Loss 0.257516 LR 0.001000 Time 0.022111 +2023-10-02 20:50:01,994 - Epoch: [44][ 670/ 1236] Overall Loss 0.257225 Objective Loss 0.257225 LR 0.001000 Time 0.022090 +2023-10-02 20:50:02,202 - Epoch: [44][ 680/ 1236] Overall Loss 0.257465 Objective Loss 0.257465 LR 0.001000 Time 0.022070 +2023-10-02 20:50:02,410 - Epoch: [44][ 690/ 1236] Overall Loss 0.257524 Objective Loss 0.257524 LR 0.001000 Time 0.022050 +2023-10-02 20:50:02,618 - Epoch: [44][ 700/ 1236] Overall Loss 0.257625 Objective Loss 0.257625 LR 0.001000 Time 0.022031 +2023-10-02 20:50:02,826 - Epoch: [44][ 710/ 1236] Overall Loss 0.257491 Objective Loss 0.257491 LR 0.001000 Time 0.022012 +2023-10-02 20:50:03,034 - Epoch: [44][ 720/ 1236] Overall Loss 0.257495 Objective Loss 0.257495 LR 0.001000 Time 0.021995 +2023-10-02 20:50:03,243 - Epoch: [44][ 730/ 1236] Overall Loss 0.257606 Objective Loss 0.257606 LR 0.001000 Time 0.021977 +2023-10-02 20:50:03,451 - Epoch: [44][ 740/ 1236] Overall Loss 0.257751 Objective Loss 0.257751 LR 0.001000 Time 0.021960 +2023-10-02 20:50:03,659 - Epoch: [44][ 750/ 1236] Overall Loss 0.258068 Objective Loss 0.258068 LR 0.001000 Time 0.021943 +2023-10-02 20:50:03,868 - Epoch: [44][ 760/ 1236] Overall Loss 0.258519 Objective Loss 0.258519 LR 0.001000 Time 0.021928 +2023-10-02 20:50:04,075 - Epoch: [44][ 770/ 1236] Overall Loss 0.258485 Objective Loss 0.258485 LR 0.001000 Time 0.021913 +2023-10-02 20:50:04,281 - Epoch: [44][ 780/ 1236] Overall Loss 0.258575 Objective Loss 0.258575 LR 0.001000 Time 0.021896 +2023-10-02 20:50:04,489 - Epoch: [44][ 790/ 1236] Overall Loss 0.258679 Objective Loss 0.258679 LR 0.001000 Time 0.021880 +2023-10-02 20:50:04,695 - Epoch: [44][ 800/ 1236] Overall Loss 0.258707 Objective Loss 0.258707 LR 0.001000 Time 0.021864 +2023-10-02 20:50:04,902 - Epoch: [44][ 810/ 1236] Overall Loss 0.259333 Objective Loss 0.259333 LR 0.001000 Time 0.021850 +2023-10-02 20:50:05,108 - Epoch: [44][ 820/ 1236] Overall Loss 0.259671 Objective Loss 0.259671 LR 0.001000 Time 0.021835 +2023-10-02 20:50:05,316 - Epoch: [44][ 830/ 1236] Overall Loss 0.259432 Objective Loss 0.259432 LR 0.001000 Time 0.021821 +2023-10-02 20:50:05,522 - Epoch: [44][ 840/ 1236] Overall Loss 0.259908 Objective Loss 0.259908 LR 0.001000 Time 0.021806 +2023-10-02 20:50:05,730 - Epoch: [44][ 850/ 1236] Overall Loss 0.260115 Objective Loss 0.260115 LR 0.001000 Time 0.021794 +2023-10-02 20:50:05,936 - Epoch: [44][ 860/ 1236] Overall Loss 0.259964 Objective Loss 0.259964 LR 0.001000 Time 0.021779 +2023-10-02 20:50:06,143 - Epoch: [44][ 870/ 1236] Overall Loss 0.259922 Objective Loss 0.259922 LR 0.001000 Time 0.021767 +2023-10-02 20:50:06,350 - Epoch: [44][ 880/ 1236] Overall Loss 0.260002 Objective Loss 0.260002 LR 0.001000 Time 0.021754 +2023-10-02 20:50:06,557 - Epoch: [44][ 890/ 1236] Overall Loss 0.259833 Objective Loss 0.259833 LR 0.001000 Time 0.021743 +2023-10-02 20:50:06,763 - Epoch: [44][ 900/ 1236] Overall Loss 0.259708 Objective Loss 0.259708 LR 0.001000 Time 0.021729 +2023-10-02 20:50:06,971 - Epoch: [44][ 910/ 1236] Overall Loss 0.259376 Objective Loss 0.259376 LR 0.001000 Time 0.021719 +2023-10-02 20:50:07,177 - Epoch: [44][ 920/ 1236] Overall Loss 0.259487 Objective Loss 0.259487 LR 0.001000 Time 0.021706 +2023-10-02 20:50:07,384 - Epoch: [44][ 930/ 1236] Overall Loss 0.259397 Objective Loss 0.259397 LR 0.001000 Time 0.021696 +2023-10-02 20:50:07,590 - Epoch: [44][ 940/ 1236] Overall Loss 0.259413 Objective Loss 0.259413 LR 0.001000 Time 0.021683 +2023-10-02 20:50:07,798 - Epoch: [44][ 950/ 1236] Overall Loss 0.259493 Objective Loss 0.259493 LR 0.001000 Time 0.021673 +2023-10-02 20:50:08,004 - Epoch: [44][ 960/ 1236] Overall Loss 0.259594 Objective Loss 0.259594 LR 0.001000 Time 0.021662 +2023-10-02 20:50:08,212 - Epoch: [44][ 970/ 1236] Overall Loss 0.259448 Objective Loss 0.259448 LR 0.001000 Time 0.021652 +2023-10-02 20:50:08,418 - Epoch: [44][ 980/ 1236] Overall Loss 0.259118 Objective Loss 0.259118 LR 0.001000 Time 0.021641 +2023-10-02 20:50:08,625 - Epoch: [44][ 990/ 1236] Overall Loss 0.258976 Objective Loss 0.258976 LR 0.001000 Time 0.021632 +2023-10-02 20:50:08,831 - Epoch: [44][ 1000/ 1236] Overall Loss 0.259015 Objective Loss 0.259015 LR 0.001000 Time 0.021621 +2023-10-02 20:50:09,039 - Epoch: [44][ 1010/ 1236] Overall Loss 0.259106 Objective Loss 0.259106 LR 0.001000 Time 0.021612 +2023-10-02 20:50:09,245 - Epoch: [44][ 1020/ 1236] Overall Loss 0.259223 Objective Loss 0.259223 LR 0.001000 Time 0.021602 +2023-10-02 20:50:09,452 - Epoch: [44][ 1030/ 1236] Overall Loss 0.259426 Objective Loss 0.259426 LR 0.001000 Time 0.021594 +2023-10-02 20:50:09,658 - Epoch: [44][ 1040/ 1236] Overall Loss 0.259370 Objective Loss 0.259370 LR 0.001000 Time 0.021584 +2023-10-02 20:50:09,866 - Epoch: [44][ 1050/ 1236] Overall Loss 0.259059 Objective Loss 0.259059 LR 0.001000 Time 0.021576 +2023-10-02 20:50:10,072 - Epoch: [44][ 1060/ 1236] Overall Loss 0.258932 Objective Loss 0.258932 LR 0.001000 Time 0.021566 +2023-10-02 20:50:10,280 - Epoch: [44][ 1070/ 1236] Overall Loss 0.258812 Objective Loss 0.258812 LR 0.001000 Time 0.021559 +2023-10-02 20:50:10,487 - Epoch: [44][ 1080/ 1236] Overall Loss 0.258557 Objective Loss 0.258557 LR 0.001000 Time 0.021550 +2023-10-02 20:50:10,694 - Epoch: [44][ 1090/ 1236] Overall Loss 0.258824 Objective Loss 0.258824 LR 0.001000 Time 0.021543 +2023-10-02 20:50:10,900 - Epoch: [44][ 1100/ 1236] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.001000 Time 0.021534 +2023-10-02 20:50:11,108 - Epoch: [44][ 1110/ 1236] Overall Loss 0.258892 Objective Loss 0.258892 LR 0.001000 Time 0.021527 +2023-10-02 20:50:11,314 - Epoch: [44][ 1120/ 1236] Overall Loss 0.258960 Objective Loss 0.258960 LR 0.001000 Time 0.021518 +2023-10-02 20:50:11,521 - Epoch: [44][ 1130/ 1236] Overall Loss 0.259103 Objective Loss 0.259103 LR 0.001000 Time 0.021511 +2023-10-02 20:50:11,727 - Epoch: [44][ 1140/ 1236] Overall Loss 0.259073 Objective Loss 0.259073 LR 0.001000 Time 0.021503 +2023-10-02 20:50:11,935 - Epoch: [44][ 1150/ 1236] Overall Loss 0.259182 Objective Loss 0.259182 LR 0.001000 Time 0.021496 +2023-10-02 20:50:12,141 - Epoch: [44][ 1160/ 1236] Overall Loss 0.259270 Objective Loss 0.259270 LR 0.001000 Time 0.021488 +2023-10-02 20:50:12,349 - Epoch: [44][ 1170/ 1236] Overall Loss 0.259474 Objective Loss 0.259474 LR 0.001000 Time 0.021482 +2023-10-02 20:50:12,555 - Epoch: [44][ 1180/ 1236] Overall Loss 0.259713 Objective Loss 0.259713 LR 0.001000 Time 0.021474 +2023-10-02 20:50:12,762 - Epoch: [44][ 1190/ 1236] Overall Loss 0.259907 Objective Loss 0.259907 LR 0.001000 Time 0.021468 +2023-10-02 20:50:12,968 - Epoch: [44][ 1200/ 1236] Overall Loss 0.260038 Objective Loss 0.260038 LR 0.001000 Time 0.021460 +2023-10-02 20:50:13,176 - Epoch: [44][ 1210/ 1236] Overall Loss 0.260226 Objective Loss 0.260226 LR 0.001000 Time 0.021454 +2023-10-02 20:50:13,382 - Epoch: [44][ 1220/ 1236] Overall Loss 0.260152 Objective Loss 0.260152 LR 0.001000 Time 0.021447 +2023-10-02 20:50:13,640 - Epoch: [44][ 1230/ 1236] Overall Loss 0.260352 Objective Loss 0.260352 LR 0.001000 Time 0.021482 +2023-10-02 20:50:13,761 - Epoch: [44][ 1236/ 1236] Overall Loss 0.260325 Objective Loss 0.260325 Top1 86.761711 Top5 98.574338 LR 0.001000 Time 0.021476 +2023-10-02 20:50:13,898 - --- validate (epoch=44)----------- +2023-10-02 20:50:13,899 - 29943 samples (256 per mini-batch) +2023-10-02 20:50:14,384 - Epoch: [44][ 10/ 117] Loss 0.295180 Top1 84.765625 Top5 98.046875 +2023-10-02 20:50:14,535 - Epoch: [44][ 20/ 117] Loss 0.308784 Top1 84.609375 Top5 98.066406 +2023-10-02 20:50:14,685 - Epoch: [44][ 30/ 117] Loss 0.304891 Top1 84.283854 Top5 97.994792 +2023-10-02 20:50:14,835 - Epoch: [44][ 40/ 117] Loss 0.299715 Top1 84.316406 Top5 98.125000 +2023-10-02 20:50:14,986 - Epoch: [44][ 50/ 117] Loss 0.297527 Top1 84.382812 Top5 98.226562 +2023-10-02 20:50:15,135 - Epoch: [44][ 60/ 117] Loss 0.299428 Top1 84.322917 Top5 98.196615 +2023-10-02 20:50:15,283 - Epoch: [44][ 70/ 117] Loss 0.299277 Top1 84.324777 Top5 98.225446 +2023-10-02 20:50:15,432 - Epoch: [44][ 80/ 117] Loss 0.305385 Top1 84.121094 Top5 98.203125 +2023-10-02 20:50:15,581 - Epoch: [44][ 90/ 117] Loss 0.307897 Top1 84.027778 Top5 98.198785 +2023-10-02 20:50:15,730 - Epoch: [44][ 100/ 117] Loss 0.308749 Top1 83.996094 Top5 98.234375 +2023-10-02 20:50:15,887 - Epoch: [44][ 110/ 117] Loss 0.310717 Top1 83.970170 Top5 98.199574 +2023-10-02 20:50:15,975 - Epoch: [44][ 117/ 117] Loss 0.310659 Top1 83.979561 Top5 98.229970 +2023-10-02 20:50:16,092 - ==> Top1: 83.980 Top5: 98.230 Loss: 0.311 + +2023-10-02 20:50:16,093 - ==> Confusion: +[[ 942 3 4 0 10 5 0 1 4 52 1 2 1 2 4 0 4 1 0 0 14] + [ 0 1079 2 0 4 9 0 24 0 1 0 0 1 0 0 2 1 0 5 1 2] + [ 1 1 943 13 3 0 44 14 0 1 1 1 6 2 0 4 3 1 8 8 2] + [ 1 4 20 991 2 3 5 5 0 0 5 0 3 3 13 1 1 3 11 1 17] + [ 19 9 2 0 975 5 0 1 0 8 2 0 2 2 6 4 10 0 0 3 2] + [ 4 50 1 0 6 979 2 33 1 3 0 6 2 6 1 1 3 0 3 5 10] + [ 0 6 18 0 0 0 1143 3 0 0 2 1 1 0 0 4 0 0 1 5 7] + [ 1 24 17 0 2 28 2 1069 0 0 2 3 4 2 0 2 0 0 33 22 7] + [ 19 5 0 0 0 3 0 4 982 30 8 1 3 13 11 0 6 0 4 0 0] + [ 137 1 3 0 9 3 1 1 25 902 2 0 0 18 2 1 3 0 0 4 7] + [ 4 6 9 13 2 2 8 5 17 0 949 1 4 14 2 1 1 1 6 2 6] + [ 1 2 2 0 2 16 0 4 0 0 0 945 22 5 0 2 1 15 0 14 4] + [ 0 4 3 5 0 1 2 3 0 0 0 46 954 3 4 9 4 9 1 12 8] + [ 2 1 1 0 4 14 2 1 5 13 7 2 0 1043 5 0 3 1 0 4 11] + [ 10 3 7 27 7 0 0 0 21 5 3 0 4 4 990 0 2 2 10 0 6] + [ 0 3 6 0 3 1 0 0 0 0 0 9 5 0 0 1069 13 11 1 8 5] + [ 0 21 1 0 4 1 0 0 0 0 0 5 0 1 3 9 1102 2 1 6 5] + [ 0 0 1 3 0 0 5 0 0 0 0 8 23 1 3 6 1 980 2 2 3] + [ 1 10 7 10 1 0 1 24 3 0 1 0 0 0 10 0 0 0 987 1 12] + [ 0 6 0 0 1 4 9 4 0 0 0 9 3 0 0 2 7 0 1 1104 2] + [ 156 299 153 70 110 156 66 122 84 61 135 141 314 303 97 47 119 57 133 264 5018]] + +2023-10-02 20:50:16,094 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:50:16,094 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:50:16,100 - + +2023-10-02 20:50:16,100 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:50:17,105 - Epoch: [45][ 10/ 1236] Overall Loss 0.246970 Objective Loss 0.246970 LR 0.001000 Time 0.100462 +2023-10-02 20:50:17,313 - Epoch: [45][ 20/ 1236] Overall Loss 0.247767 Objective Loss 0.247767 LR 0.001000 Time 0.060600 +2023-10-02 20:50:17,519 - Epoch: [45][ 30/ 1236] Overall Loss 0.235817 Objective Loss 0.235817 LR 0.001000 Time 0.047257 +2023-10-02 20:50:17,728 - Epoch: [45][ 40/ 1236] Overall Loss 0.235952 Objective Loss 0.235952 LR 0.001000 Time 0.040654 +2023-10-02 20:50:17,932 - Epoch: [45][ 50/ 1236] Overall Loss 0.236128 Objective Loss 0.236128 LR 0.001000 Time 0.036608 +2023-10-02 20:50:18,141 - Epoch: [45][ 60/ 1236] Overall Loss 0.240804 Objective Loss 0.240804 LR 0.001000 Time 0.033983 +2023-10-02 20:50:18,346 - Epoch: [45][ 70/ 1236] Overall Loss 0.244927 Objective Loss 0.244927 LR 0.001000 Time 0.032047 +2023-10-02 20:50:18,554 - Epoch: [45][ 80/ 1236] Overall Loss 0.244094 Objective Loss 0.244094 LR 0.001000 Time 0.030643 +2023-10-02 20:50:18,759 - Epoch: [45][ 90/ 1236] Overall Loss 0.246671 Objective Loss 0.246671 LR 0.001000 Time 0.029510 +2023-10-02 20:50:18,968 - Epoch: [45][ 100/ 1236] Overall Loss 0.245638 Objective Loss 0.245638 LR 0.001000 Time 0.028645 +2023-10-02 20:50:19,172 - Epoch: [45][ 110/ 1236] Overall Loss 0.244894 Objective Loss 0.244894 LR 0.001000 Time 0.027898 +2023-10-02 20:50:19,381 - Epoch: [45][ 120/ 1236] Overall Loss 0.247408 Objective Loss 0.247408 LR 0.001000 Time 0.027310 +2023-10-02 20:50:19,585 - Epoch: [45][ 130/ 1236] Overall Loss 0.248096 Objective Loss 0.248096 LR 0.001000 Time 0.026778 +2023-10-02 20:50:19,794 - Epoch: [45][ 140/ 1236] Overall Loss 0.249899 Objective Loss 0.249899 LR 0.001000 Time 0.026355 +2023-10-02 20:50:19,999 - Epoch: [45][ 150/ 1236] Overall Loss 0.248385 Objective Loss 0.248385 LR 0.001000 Time 0.025960 +2023-10-02 20:50:20,208 - Epoch: [45][ 160/ 1236] Overall Loss 0.249101 Objective Loss 0.249101 LR 0.001000 Time 0.025642 +2023-10-02 20:50:20,412 - Epoch: [45][ 170/ 1236] Overall Loss 0.247831 Objective Loss 0.247831 LR 0.001000 Time 0.025336 +2023-10-02 20:50:20,621 - Epoch: [45][ 180/ 1236] Overall Loss 0.248249 Objective Loss 0.248249 LR 0.001000 Time 0.025088 +2023-10-02 20:50:20,826 - Epoch: [45][ 190/ 1236] Overall Loss 0.248449 Objective Loss 0.248449 LR 0.001000 Time 0.024842 +2023-10-02 20:50:21,035 - Epoch: [45][ 200/ 1236] Overall Loss 0.248440 Objective Loss 0.248440 LR 0.001000 Time 0.024644 +2023-10-02 20:50:21,240 - Epoch: [45][ 210/ 1236] Overall Loss 0.247874 Objective Loss 0.247874 LR 0.001000 Time 0.024445 +2023-10-02 20:50:21,449 - Epoch: [45][ 220/ 1236] Overall Loss 0.247117 Objective Loss 0.247117 LR 0.001000 Time 0.024283 +2023-10-02 20:50:21,654 - Epoch: [45][ 230/ 1236] Overall Loss 0.248058 Objective Loss 0.248058 LR 0.001000 Time 0.024116 +2023-10-02 20:50:21,863 - Epoch: [45][ 240/ 1236] Overall Loss 0.248982 Objective Loss 0.248982 LR 0.001000 Time 0.023982 +2023-10-02 20:50:22,068 - Epoch: [45][ 250/ 1236] Overall Loss 0.251520 Objective Loss 0.251520 LR 0.001000 Time 0.023842 +2023-10-02 20:50:22,276 - Epoch: [45][ 260/ 1236] Overall Loss 0.252891 Objective Loss 0.252891 LR 0.001000 Time 0.023724 +2023-10-02 20:50:22,482 - Epoch: [45][ 270/ 1236] Overall Loss 0.253731 Objective Loss 0.253731 LR 0.001000 Time 0.023602 +2023-10-02 20:50:22,691 - Epoch: [45][ 280/ 1236] Overall Loss 0.253122 Objective Loss 0.253122 LR 0.001000 Time 0.023506 +2023-10-02 20:50:22,896 - Epoch: [45][ 290/ 1236] Overall Loss 0.252394 Objective Loss 0.252394 LR 0.001000 Time 0.023400 +2023-10-02 20:50:23,105 - Epoch: [45][ 300/ 1236] Overall Loss 0.252316 Objective Loss 0.252316 LR 0.001000 Time 0.023316 +2023-10-02 20:50:23,310 - Epoch: [45][ 310/ 1236] Overall Loss 0.252515 Objective Loss 0.252515 LR 0.001000 Time 0.023224 +2023-10-02 20:50:23,519 - Epoch: [45][ 320/ 1236] Overall Loss 0.252745 Objective Loss 0.252745 LR 0.001000 Time 0.023152 +2023-10-02 20:50:23,723 - Epoch: [45][ 330/ 1236] Overall Loss 0.252744 Objective Loss 0.252744 LR 0.001000 Time 0.023067 +2023-10-02 20:50:23,932 - Epoch: [45][ 340/ 1236] Overall Loss 0.253389 Objective Loss 0.253389 LR 0.001000 Time 0.023003 +2023-10-02 20:50:24,137 - Epoch: [45][ 350/ 1236] Overall Loss 0.253813 Objective Loss 0.253813 LR 0.001000 Time 0.022930 +2023-10-02 20:50:24,347 - Epoch: [45][ 360/ 1236] Overall Loss 0.253491 Objective Loss 0.253491 LR 0.001000 Time 0.022874 +2023-10-02 20:50:24,551 - Epoch: [45][ 370/ 1236] Overall Loss 0.253488 Objective Loss 0.253488 LR 0.001000 Time 0.022808 +2023-10-02 20:50:24,760 - Epoch: [45][ 380/ 1236] Overall Loss 0.254032 Objective Loss 0.254032 LR 0.001000 Time 0.022758 +2023-10-02 20:50:24,965 - Epoch: [45][ 390/ 1236] Overall Loss 0.253613 Objective Loss 0.253613 LR 0.001000 Time 0.022699 +2023-10-02 20:50:25,179 - Epoch: [45][ 400/ 1236] Overall Loss 0.253803 Objective Loss 0.253803 LR 0.001000 Time 0.022666 +2023-10-02 20:50:25,391 - Epoch: [45][ 410/ 1236] Overall Loss 0.253393 Objective Loss 0.253393 LR 0.001000 Time 0.022629 +2023-10-02 20:50:25,606 - Epoch: [45][ 420/ 1236] Overall Loss 0.253679 Objective Loss 0.253679 LR 0.001000 Time 0.022601 +2023-10-02 20:50:25,810 - Epoch: [45][ 430/ 1236] Overall Loss 0.253964 Objective Loss 0.253964 LR 0.001000 Time 0.022550 +2023-10-02 20:50:26,017 - Epoch: [45][ 440/ 1236] Overall Loss 0.253968 Objective Loss 0.253968 LR 0.001000 Time 0.022506 +2023-10-02 20:50:26,223 - Epoch: [45][ 450/ 1236] Overall Loss 0.254371 Objective Loss 0.254371 LR 0.001000 Time 0.022463 +2023-10-02 20:50:26,430 - Epoch: [45][ 460/ 1236] Overall Loss 0.254560 Objective Loss 0.254560 LR 0.001000 Time 0.022424 +2023-10-02 20:50:26,635 - Epoch: [45][ 470/ 1236] Overall Loss 0.254898 Objective Loss 0.254898 LR 0.001000 Time 0.022384 +2023-10-02 20:50:26,843 - Epoch: [45][ 480/ 1236] Overall Loss 0.254799 Objective Loss 0.254799 LR 0.001000 Time 0.022348 +2023-10-02 20:50:27,048 - Epoch: [45][ 490/ 1236] Overall Loss 0.254443 Objective Loss 0.254443 LR 0.001000 Time 0.022311 +2023-10-02 20:50:27,256 - Epoch: [45][ 500/ 1236] Overall Loss 0.254383 Objective Loss 0.254383 LR 0.001000 Time 0.022281 +2023-10-02 20:50:27,461 - Epoch: [45][ 510/ 1236] Overall Loss 0.254485 Objective Loss 0.254485 LR 0.001000 Time 0.022245 +2023-10-02 20:50:27,668 - Epoch: [45][ 520/ 1236] Overall Loss 0.254354 Objective Loss 0.254354 LR 0.001000 Time 0.022214 +2023-10-02 20:50:27,874 - Epoch: [45][ 530/ 1236] Overall Loss 0.254503 Objective Loss 0.254503 LR 0.001000 Time 0.022182 +2023-10-02 20:50:28,082 - Epoch: [45][ 540/ 1236] Overall Loss 0.254741 Objective Loss 0.254741 LR 0.001000 Time 0.022156 +2023-10-02 20:50:28,286 - Epoch: [45][ 550/ 1236] Overall Loss 0.255545 Objective Loss 0.255545 LR 0.001000 Time 0.022125 +2023-10-02 20:50:28,494 - Epoch: [45][ 560/ 1236] Overall Loss 0.256117 Objective Loss 0.256117 LR 0.001000 Time 0.022101 +2023-10-02 20:50:28,699 - Epoch: [45][ 570/ 1236] Overall Loss 0.256351 Objective Loss 0.256351 LR 0.001000 Time 0.022071 +2023-10-02 20:50:28,907 - Epoch: [45][ 580/ 1236] Overall Loss 0.256131 Objective Loss 0.256131 LR 0.001000 Time 0.022050 +2023-10-02 20:50:29,112 - Epoch: [45][ 590/ 1236] Overall Loss 0.255902 Objective Loss 0.255902 LR 0.001000 Time 0.022023 +2023-10-02 20:50:29,319 - Epoch: [45][ 600/ 1236] Overall Loss 0.255962 Objective Loss 0.255962 LR 0.001000 Time 0.022000 +2023-10-02 20:50:29,525 - Epoch: [45][ 610/ 1236] Overall Loss 0.255836 Objective Loss 0.255836 LR 0.001000 Time 0.021976 +2023-10-02 20:50:29,732 - Epoch: [45][ 620/ 1236] Overall Loss 0.255634 Objective Loss 0.255634 LR 0.001000 Time 0.021955 +2023-10-02 20:50:29,937 - Epoch: [45][ 630/ 1236] Overall Loss 0.255549 Objective Loss 0.255549 LR 0.001000 Time 0.021933 +2023-10-02 20:50:30,144 - Epoch: [45][ 640/ 1236] Overall Loss 0.255796 Objective Loss 0.255796 LR 0.001000 Time 0.021913 +2023-10-02 20:50:30,350 - Epoch: [45][ 650/ 1236] Overall Loss 0.255615 Objective Loss 0.255615 LR 0.001000 Time 0.021892 +2023-10-02 20:50:30,557 - Epoch: [45][ 660/ 1236] Overall Loss 0.255318 Objective Loss 0.255318 LR 0.001000 Time 0.021873 +2023-10-02 20:50:30,763 - Epoch: [45][ 670/ 1236] Overall Loss 0.255605 Objective Loss 0.255605 LR 0.001000 Time 0.021853 +2023-10-02 20:50:30,970 - Epoch: [45][ 680/ 1236] Overall Loss 0.255591 Objective Loss 0.255591 LR 0.001000 Time 0.021836 +2023-10-02 20:50:31,176 - Epoch: [45][ 690/ 1236] Overall Loss 0.255780 Objective Loss 0.255780 LR 0.001000 Time 0.021817 +2023-10-02 20:50:31,383 - Epoch: [45][ 700/ 1236] Overall Loss 0.256430 Objective Loss 0.256430 LR 0.001000 Time 0.021801 +2023-10-02 20:50:31,588 - Epoch: [45][ 710/ 1236] Overall Loss 0.256567 Objective Loss 0.256567 LR 0.001000 Time 0.021782 +2023-10-02 20:50:31,795 - Epoch: [45][ 720/ 1236] Overall Loss 0.257210 Objective Loss 0.257210 LR 0.001000 Time 0.021766 +2023-10-02 20:50:32,001 - Epoch: [45][ 730/ 1236] Overall Loss 0.257266 Objective Loss 0.257266 LR 0.001000 Time 0.021749 +2023-10-02 20:50:32,208 - Epoch: [45][ 740/ 1236] Overall Loss 0.257233 Objective Loss 0.257233 LR 0.001000 Time 0.021734 +2023-10-02 20:50:32,413 - Epoch: [45][ 750/ 1236] Overall Loss 0.257246 Objective Loss 0.257246 LR 0.001000 Time 0.021718 +2023-10-02 20:50:32,620 - Epoch: [45][ 760/ 1236] Overall Loss 0.257718 Objective Loss 0.257718 LR 0.001000 Time 0.021704 +2023-10-02 20:50:32,826 - Epoch: [45][ 770/ 1236] Overall Loss 0.257750 Objective Loss 0.257750 LR 0.001000 Time 0.021689 +2023-10-02 20:50:33,033 - Epoch: [45][ 780/ 1236] Overall Loss 0.257899 Objective Loss 0.257899 LR 0.001000 Time 0.021676 +2023-10-02 20:50:33,238 - Epoch: [45][ 790/ 1236] Overall Loss 0.257991 Objective Loss 0.257991 LR 0.001000 Time 0.021661 +2023-10-02 20:50:33,445 - Epoch: [45][ 800/ 1236] Overall Loss 0.258127 Objective Loss 0.258127 LR 0.001000 Time 0.021649 +2023-10-02 20:50:33,651 - Epoch: [45][ 810/ 1236] Overall Loss 0.258355 Objective Loss 0.258355 LR 0.001000 Time 0.021635 +2023-10-02 20:50:33,861 - Epoch: [45][ 820/ 1236] Overall Loss 0.258347 Objective Loss 0.258347 LR 0.001000 Time 0.021627 +2023-10-02 20:50:34,072 - Epoch: [45][ 830/ 1236] Overall Loss 0.258476 Objective Loss 0.258476 LR 0.001000 Time 0.021620 +2023-10-02 20:50:34,284 - Epoch: [45][ 840/ 1236] Overall Loss 0.258433 Objective Loss 0.258433 LR 0.001000 Time 0.021615 +2023-10-02 20:50:34,495 - Epoch: [45][ 850/ 1236] Overall Loss 0.258368 Objective Loss 0.258368 LR 0.001000 Time 0.021608 +2023-10-02 20:50:34,707 - Epoch: [45][ 860/ 1236] Overall Loss 0.258273 Objective Loss 0.258273 LR 0.001000 Time 0.021603 +2023-10-02 20:50:34,918 - Epoch: [45][ 870/ 1236] Overall Loss 0.258155 Objective Loss 0.258155 LR 0.001000 Time 0.021597 +2023-10-02 20:50:35,130 - Epoch: [45][ 880/ 1236] Overall Loss 0.258365 Objective Loss 0.258365 LR 0.001000 Time 0.021592 +2023-10-02 20:50:35,342 - Epoch: [45][ 890/ 1236] Overall Loss 0.258060 Objective Loss 0.258060 LR 0.001000 Time 0.021587 +2023-10-02 20:50:35,552 - Epoch: [45][ 900/ 1236] Overall Loss 0.257977 Objective Loss 0.257977 LR 0.001000 Time 0.021581 +2023-10-02 20:50:35,763 - Epoch: [45][ 910/ 1236] Overall Loss 0.257737 Objective Loss 0.257737 LR 0.001000 Time 0.021575 +2023-10-02 20:50:35,975 - Epoch: [45][ 920/ 1236] Overall Loss 0.257738 Objective Loss 0.257738 LR 0.001000 Time 0.021570 +2023-10-02 20:50:36,186 - Epoch: [45][ 930/ 1236] Overall Loss 0.257430 Objective Loss 0.257430 LR 0.001000 Time 0.021565 +2023-10-02 20:50:36,398 - Epoch: [45][ 940/ 1236] Overall Loss 0.257718 Objective Loss 0.257718 LR 0.001000 Time 0.021559 +2023-10-02 20:50:36,610 - Epoch: [45][ 950/ 1236] Overall Loss 0.257483 Objective Loss 0.257483 LR 0.001000 Time 0.021555 +2023-10-02 20:50:36,821 - Epoch: [45][ 960/ 1236] Overall Loss 0.257474 Objective Loss 0.257474 LR 0.001000 Time 0.021549 +2023-10-02 20:50:37,032 - Epoch: [45][ 970/ 1236] Overall Loss 0.257247 Objective Loss 0.257247 LR 0.001000 Time 0.021544 +2023-10-02 20:50:37,243 - Epoch: [45][ 980/ 1236] Overall Loss 0.257338 Objective Loss 0.257338 LR 0.001000 Time 0.021540 +2023-10-02 20:50:37,454 - Epoch: [45][ 990/ 1236] Overall Loss 0.257593 Objective Loss 0.257593 LR 0.001000 Time 0.021535 +2023-10-02 20:50:37,664 - Epoch: [45][ 1000/ 1236] Overall Loss 0.257846 Objective Loss 0.257846 LR 0.001000 Time 0.021529 +2023-10-02 20:50:37,876 - Epoch: [45][ 1010/ 1236] Overall Loss 0.257753 Objective Loss 0.257753 LR 0.001000 Time 0.021524 +2023-10-02 20:50:38,087 - Epoch: [45][ 1020/ 1236] Overall Loss 0.257539 Objective Loss 0.257539 LR 0.001000 Time 0.021520 +2023-10-02 20:50:38,298 - Epoch: [45][ 1030/ 1236] Overall Loss 0.257463 Objective Loss 0.257463 LR 0.001000 Time 0.021516 +2023-10-02 20:50:38,509 - Epoch: [45][ 1040/ 1236] Overall Loss 0.257713 Objective Loss 0.257713 LR 0.001000 Time 0.021511 +2023-10-02 20:50:38,721 - Epoch: [45][ 1050/ 1236] Overall Loss 0.257924 Objective Loss 0.257924 LR 0.001000 Time 0.021508 +2023-10-02 20:50:38,932 - Epoch: [45][ 1060/ 1236] Overall Loss 0.257954 Objective Loss 0.257954 LR 0.001000 Time 0.021504 +2023-10-02 20:50:39,144 - Epoch: [45][ 1070/ 1236] Overall Loss 0.258123 Objective Loss 0.258123 LR 0.001000 Time 0.021500 +2023-10-02 20:50:39,355 - Epoch: [45][ 1080/ 1236] Overall Loss 0.258432 Objective Loss 0.258432 LR 0.001000 Time 0.021496 +2023-10-02 20:50:39,566 - Epoch: [45][ 1090/ 1236] Overall Loss 0.258650 Objective Loss 0.258650 LR 0.001000 Time 0.021493 +2023-10-02 20:50:39,776 - Epoch: [45][ 1100/ 1236] Overall Loss 0.258520 Objective Loss 0.258520 LR 0.001000 Time 0.021488 +2023-10-02 20:50:39,987 - Epoch: [45][ 1110/ 1236] Overall Loss 0.258484 Objective Loss 0.258484 LR 0.001000 Time 0.021484 +2023-10-02 20:50:40,198 - Epoch: [45][ 1120/ 1236] Overall Loss 0.258389 Objective Loss 0.258389 LR 0.001000 Time 0.021480 +2023-10-02 20:50:40,409 - Epoch: [45][ 1130/ 1236] Overall Loss 0.258306 Objective Loss 0.258306 LR 0.001000 Time 0.021477 +2023-10-02 20:50:40,620 - Epoch: [45][ 1140/ 1236] Overall Loss 0.258430 Objective Loss 0.258430 LR 0.001000 Time 0.021473 +2023-10-02 20:50:40,832 - Epoch: [45][ 1150/ 1236] Overall Loss 0.258281 Objective Loss 0.258281 LR 0.001000 Time 0.021470 +2023-10-02 20:50:41,046 - Epoch: [45][ 1160/ 1236] Overall Loss 0.258539 Objective Loss 0.258539 LR 0.001000 Time 0.021469 +2023-10-02 20:50:41,263 - Epoch: [45][ 1170/ 1236] Overall Loss 0.258322 Objective Loss 0.258322 LR 0.001000 Time 0.021471 +2023-10-02 20:50:41,484 - Epoch: [45][ 1180/ 1236] Overall Loss 0.258591 Objective Loss 0.258591 LR 0.001000 Time 0.021476 +2023-10-02 20:50:41,702 - Epoch: [45][ 1190/ 1236] Overall Loss 0.258855 Objective Loss 0.258855 LR 0.001000 Time 0.021478 +2023-10-02 20:50:41,911 - Epoch: [45][ 1200/ 1236] Overall Loss 0.258996 Objective Loss 0.258996 LR 0.001000 Time 0.021473 +2023-10-02 20:50:42,123 - Epoch: [45][ 1210/ 1236] Overall Loss 0.258937 Objective Loss 0.258937 LR 0.001000 Time 0.021471 +2023-10-02 20:50:42,336 - Epoch: [45][ 1220/ 1236] Overall Loss 0.259122 Objective Loss 0.259122 LR 0.001000 Time 0.021469 +2023-10-02 20:50:42,597 - Epoch: [45][ 1230/ 1236] Overall Loss 0.259375 Objective Loss 0.259375 LR 0.001000 Time 0.021506 +2023-10-02 20:50:42,718 - Epoch: [45][ 1236/ 1236] Overall Loss 0.259447 Objective Loss 0.259447 Top1 84.928717 Top5 97.352342 LR 0.001000 Time 0.021500 +2023-10-02 20:50:42,858 - --- validate (epoch=45)----------- +2023-10-02 20:50:42,859 - 29943 samples (256 per mini-batch) +2023-10-02 20:50:43,337 - Epoch: [45][ 10/ 117] Loss 0.324454 Top1 83.281250 Top5 98.164062 +2023-10-02 20:50:43,489 - Epoch: [45][ 20/ 117] Loss 0.300440 Top1 84.179688 Top5 98.183594 +2023-10-02 20:50:43,641 - Epoch: [45][ 30/ 117] Loss 0.311848 Top1 83.841146 Top5 98.033854 +2023-10-02 20:50:43,793 - Epoch: [45][ 40/ 117] Loss 0.312984 Top1 83.642578 Top5 98.046875 +2023-10-02 20:50:43,946 - Epoch: [45][ 50/ 117] Loss 0.319217 Top1 83.632812 Top5 98.000000 +2023-10-02 20:50:44,098 - Epoch: [45][ 60/ 117] Loss 0.319616 Top1 83.561198 Top5 97.988281 +2023-10-02 20:50:44,249 - Epoch: [45][ 70/ 117] Loss 0.323918 Top1 83.426339 Top5 97.968750 +2023-10-02 20:50:44,401 - Epoch: [45][ 80/ 117] Loss 0.322681 Top1 83.583984 Top5 97.963867 +2023-10-02 20:50:44,554 - Epoch: [45][ 90/ 117] Loss 0.320816 Top1 83.606771 Top5 97.955729 +2023-10-02 20:50:44,705 - Epoch: [45][ 100/ 117] Loss 0.320792 Top1 83.609375 Top5 97.953125 +2023-10-02 20:50:44,862 - Epoch: [45][ 110/ 117] Loss 0.318739 Top1 83.647017 Top5 97.982955 +2023-10-02 20:50:44,951 - Epoch: [45][ 117/ 117] Loss 0.319623 Top1 83.682330 Top5 98.009551 +2023-10-02 20:50:45,073 - ==> Top1: 83.682 Top5: 98.010 Loss: 0.320 + +2023-10-02 20:50:45,073 - ==> Confusion: +[[ 925 5 0 0 10 4 0 2 10 57 2 2 1 3 6 4 1 2 1 0 15] + [ 0 1060 1 1 4 25 0 11 1 2 2 0 0 0 1 5 0 1 11 0 6] + [ 4 1 936 21 3 0 28 8 0 1 3 2 10 1 3 9 0 2 11 6 7] + [ 2 3 5 979 1 3 4 3 6 0 6 0 10 4 23 4 0 5 13 0 18] + [ 20 11 2 1 960 7 0 0 1 11 0 0 1 3 9 5 13 2 2 2 0] + [ 1 48 1 0 1 1004 0 9 1 8 1 6 1 5 6 0 2 0 6 5 11] + [ 0 5 24 2 1 2 1114 2 0 0 6 3 2 0 0 8 0 1 3 9 9] + [ 3 33 14 0 4 43 5 993 0 4 3 12 3 6 1 2 0 0 68 17 7] + [ 17 5 0 0 2 6 0 0 980 32 8 1 4 9 12 1 3 1 5 3 0] + [ 93 1 0 0 8 1 0 0 26 957 0 0 0 11 7 3 0 0 1 4 7] + [ 1 2 10 9 1 4 1 0 12 0 976 1 2 11 5 0 0 4 6 2 6] + [ 0 0 1 0 0 7 0 2 0 0 0 970 14 12 0 6 3 13 0 6 1] + [ 0 5 1 4 1 3 0 0 0 0 2 66 942 5 0 11 1 13 2 7 5] + [ 1 0 0 1 1 10 0 0 10 15 10 7 2 1040 7 1 3 2 0 1 8] + [ 10 0 2 16 4 1 0 0 25 3 2 2 5 2 1012 0 0 0 9 0 8] + [ 0 0 3 1 4 0 1 0 0 0 0 7 5 1 1 1082 13 9 3 3 1] + [ 0 15 1 0 6 6 1 0 1 0 0 7 1 1 4 12 1089 0 1 5 11] + [ 0 0 0 3 0 0 1 0 0 1 0 11 16 2 0 9 1 985 3 5 1] + [ 1 8 3 17 1 0 1 10 5 0 3 2 3 0 10 0 0 0 994 1 9] + [ 0 4 2 2 1 6 6 1 0 0 1 16 4 3 1 8 9 1 2 1080 5] + [ 119 237 93 86 103 194 35 50 125 84 162 188 326 272 139 108 118 65 203 219 4979]] + +2023-10-02 20:50:45,075 - ==> Best [Top1: 84.160 Top5: 98.173 Sparsity:0.00 Params: 169472 on epoch: 32] +2023-10-02 20:50:45,075 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:50:45,081 - + +2023-10-02 20:50:45,081 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:50:46,073 - Epoch: [46][ 10/ 1236] Overall Loss 0.247343 Objective Loss 0.247343 LR 0.001000 Time 0.099230 +2023-10-02 20:50:46,280 - Epoch: [46][ 20/ 1236] Overall Loss 0.259715 Objective Loss 0.259715 LR 0.001000 Time 0.059920 +2023-10-02 20:50:46,485 - Epoch: [46][ 30/ 1236] Overall Loss 0.258355 Objective Loss 0.258355 LR 0.001000 Time 0.046778 +2023-10-02 20:50:46,691 - Epoch: [46][ 40/ 1236] Overall Loss 0.258809 Objective Loss 0.258809 LR 0.001000 Time 0.040222 +2023-10-02 20:50:46,896 - Epoch: [46][ 50/ 1236] Overall Loss 0.258205 Objective Loss 0.258205 LR 0.001000 Time 0.036260 +2023-10-02 20:50:47,101 - Epoch: [46][ 60/ 1236] Overall Loss 0.251037 Objective Loss 0.251037 LR 0.001000 Time 0.033638 +2023-10-02 20:50:47,306 - Epoch: [46][ 70/ 1236] Overall Loss 0.249018 Objective Loss 0.249018 LR 0.001000 Time 0.031759 +2023-10-02 20:50:47,514 - Epoch: [46][ 80/ 1236] Overall Loss 0.245343 Objective Loss 0.245343 LR 0.001000 Time 0.030383 +2023-10-02 20:50:47,718 - Epoch: [46][ 90/ 1236] Overall Loss 0.243876 Objective Loss 0.243876 LR 0.001000 Time 0.029271 +2023-10-02 20:50:47,926 - Epoch: [46][ 100/ 1236] Overall Loss 0.244763 Objective Loss 0.244763 LR 0.001000 Time 0.028417 +2023-10-02 20:50:48,130 - Epoch: [46][ 110/ 1236] Overall Loss 0.241402 Objective Loss 0.241402 LR 0.001000 Time 0.027689 +2023-10-02 20:50:48,336 - Epoch: [46][ 120/ 1236] Overall Loss 0.243177 Objective Loss 0.243177 LR 0.001000 Time 0.027099 +2023-10-02 20:50:48,542 - Epoch: [46][ 130/ 1236] Overall Loss 0.242664 Objective Loss 0.242664 LR 0.001000 Time 0.026592 +2023-10-02 20:50:48,749 - Epoch: [46][ 140/ 1236] Overall Loss 0.244485 Objective Loss 0.244485 LR 0.001000 Time 0.026169 +2023-10-02 20:50:48,953 - Epoch: [46][ 150/ 1236] Overall Loss 0.244966 Objective Loss 0.244966 LR 0.001000 Time 0.025783 +2023-10-02 20:50:49,159 - Epoch: [46][ 160/ 1236] Overall Loss 0.243815 Objective Loss 0.243815 LR 0.001000 Time 0.025459 +2023-10-02 20:50:49,364 - Epoch: [46][ 170/ 1236] Overall Loss 0.243512 Objective Loss 0.243512 LR 0.001000 Time 0.025156 +2023-10-02 20:50:49,571 - Epoch: [46][ 180/ 1236] Overall Loss 0.243565 Objective Loss 0.243565 LR 0.001000 Time 0.024910 +2023-10-02 20:50:49,775 - Epoch: [46][ 190/ 1236] Overall Loss 0.243952 Objective Loss 0.243952 LR 0.001000 Time 0.024667 +2023-10-02 20:50:49,982 - Epoch: [46][ 200/ 1236] Overall Loss 0.244926 Objective Loss 0.244926 LR 0.001000 Time 0.024469 +2023-10-02 20:50:50,186 - Epoch: [46][ 210/ 1236] Overall Loss 0.246056 Objective Loss 0.246056 LR 0.001000 Time 0.024274 +2023-10-02 20:50:50,392 - Epoch: [46][ 220/ 1236] Overall Loss 0.245474 Objective Loss 0.245474 LR 0.001000 Time 0.024108 +2023-10-02 20:50:50,597 - Epoch: [46][ 230/ 1236] Overall Loss 0.245976 Objective Loss 0.245976 LR 0.001000 Time 0.023943 +2023-10-02 20:50:50,803 - Epoch: [46][ 240/ 1236] Overall Loss 0.246648 Objective Loss 0.246648 LR 0.001000 Time 0.023804 +2023-10-02 20:50:51,008 - Epoch: [46][ 250/ 1236] Overall Loss 0.246810 Objective Loss 0.246810 LR 0.001000 Time 0.023666 +2023-10-02 20:50:51,216 - Epoch: [46][ 260/ 1236] Overall Loss 0.246120 Objective Loss 0.246120 LR 0.001000 Time 0.023553 +2023-10-02 20:50:51,420 - Epoch: [46][ 270/ 1236] Overall Loss 0.245984 Objective Loss 0.245984 LR 0.001000 Time 0.023434 +2023-10-02 20:50:51,626 - Epoch: [46][ 280/ 1236] Overall Loss 0.246204 Objective Loss 0.246204 LR 0.001000 Time 0.023333 +2023-10-02 20:50:51,831 - Epoch: [46][ 290/ 1236] Overall Loss 0.247518 Objective Loss 0.247518 LR 0.001000 Time 0.023229 +2023-10-02 20:50:52,038 - Epoch: [46][ 300/ 1236] Overall Loss 0.247766 Objective Loss 0.247766 LR 0.001000 Time 0.023145 +2023-10-02 20:50:52,242 - Epoch: [46][ 310/ 1236] Overall Loss 0.247705 Objective Loss 0.247705 LR 0.001000 Time 0.023054 +2023-10-02 20:50:52,450 - Epoch: [46][ 320/ 1236] Overall Loss 0.249052 Objective Loss 0.249052 LR 0.001000 Time 0.022982 +2023-10-02 20:50:52,653 - Epoch: [46][ 330/ 1236] Overall Loss 0.250219 Objective Loss 0.250219 LR 0.001000 Time 0.022903 +2023-10-02 20:50:52,861 - Epoch: [46][ 340/ 1236] Overall Loss 0.250394 Objective Loss 0.250394 LR 0.001000 Time 0.022839 +2023-10-02 20:50:53,065 - Epoch: [46][ 350/ 1236] Overall Loss 0.250762 Objective Loss 0.250762 LR 0.001000 Time 0.022768 +2023-10-02 20:50:53,273 - Epoch: [46][ 360/ 1236] Overall Loss 0.251249 Objective Loss 0.251249 LR 0.001000 Time 0.022713 +2023-10-02 20:50:53,477 - Epoch: [46][ 370/ 1236] Overall Loss 0.251303 Objective Loss 0.251303 LR 0.001000 Time 0.022649 +2023-10-02 20:50:53,683 - Epoch: [46][ 380/ 1236] Overall Loss 0.250441 Objective Loss 0.250441 LR 0.001000 Time 0.022596 +2023-10-02 20:50:53,888 - Epoch: [46][ 390/ 1236] Overall Loss 0.250765 Objective Loss 0.250765 LR 0.001000 Time 0.022542 +2023-10-02 20:50:54,095 - Epoch: [46][ 400/ 1236] Overall Loss 0.251133 Objective Loss 0.251133 LR 0.001000 Time 0.022494 +2023-10-02 20:50:54,300 - Epoch: [46][ 410/ 1236] Overall Loss 0.251046 Objective Loss 0.251046 LR 0.001000 Time 0.022445 +2023-10-02 20:50:54,508 - Epoch: [46][ 420/ 1236] Overall Loss 0.251326 Objective Loss 0.251326 LR 0.001000 Time 0.022405 +2023-10-02 20:50:54,712 - Epoch: [46][ 430/ 1236] Overall Loss 0.251894 Objective Loss 0.251894 LR 0.001000 Time 0.022357 +2023-10-02 20:50:54,920 - Epoch: [46][ 440/ 1236] Overall Loss 0.251674 Objective Loss 0.251674 LR 0.001000 Time 0.022321 +2023-10-02 20:50:55,124 - Epoch: [46][ 450/ 1236] Overall Loss 0.252422 Objective Loss 0.252422 LR 0.001000 Time 0.022277 +2023-10-02 20:50:55,332 - Epoch: [46][ 460/ 1236] Overall Loss 0.252745 Objective Loss 0.252745 LR 0.001000 Time 0.022244 +2023-10-02 20:50:55,536 - Epoch: [46][ 470/ 1236] Overall Loss 0.253110 Objective Loss 0.253110 LR 0.001000 Time 0.022205 +2023-10-02 20:50:55,744 - Epoch: [46][ 480/ 1236] Overall Loss 0.252541 Objective Loss 0.252541 LR 0.001000 Time 0.022175 +2023-10-02 20:50:55,948 - Epoch: [46][ 490/ 1236] Overall Loss 0.252648 Objective Loss 0.252648 LR 0.001000 Time 0.022139 +2023-10-02 20:50:56,155 - Epoch: [46][ 500/ 1236] Overall Loss 0.251880 Objective Loss 0.251880 LR 0.001000 Time 0.022109 +2023-10-02 20:50:56,360 - Epoch: [46][ 510/ 1236] Overall Loss 0.252190 Objective Loss 0.252190 LR 0.001000 Time 0.022078 +2023-10-02 20:50:56,567 - Epoch: [46][ 520/ 1236] Overall Loss 0.252224 Objective Loss 0.252224 LR 0.001000 Time 0.022050 +2023-10-02 20:50:56,772 - Epoch: [46][ 530/ 1236] Overall Loss 0.252433 Objective Loss 0.252433 LR 0.001000 Time 0.022021 +2023-10-02 20:50:56,979 - Epoch: [46][ 540/ 1236] Overall Loss 0.253720 Objective Loss 0.253720 LR 0.001000 Time 0.021995 +2023-10-02 20:50:57,184 - Epoch: [46][ 550/ 1236] Overall Loss 0.253798 Objective Loss 0.253798 LR 0.001000 Time 0.021968 +2023-10-02 20:50:57,391 - Epoch: [46][ 560/ 1236] Overall Loss 0.253721 Objective Loss 0.253721 LR 0.001000 Time 0.021944 +2023-10-02 20:50:57,596 - Epoch: [46][ 570/ 1236] Overall Loss 0.253564 Objective Loss 0.253564 LR 0.001000 Time 0.021916 +2023-10-02 20:50:57,802 - Epoch: [46][ 580/ 1236] Overall Loss 0.253309 Objective Loss 0.253309 LR 0.001000 Time 0.021894 +2023-10-02 20:50:58,008 - Epoch: [46][ 590/ 1236] Overall Loss 0.253251 Objective Loss 0.253251 LR 0.001000 Time 0.021870 +2023-10-02 20:50:58,216 - Epoch: [46][ 600/ 1236] Overall Loss 0.253156 Objective Loss 0.253156 LR 0.001000 Time 0.021852 +2023-10-02 20:50:58,420 - Epoch: [46][ 610/ 1236] Overall Loss 0.253025 Objective Loss 0.253025 LR 0.001000 Time 0.021828 +2023-10-02 20:50:58,628 - Epoch: [46][ 620/ 1236] Overall Loss 0.253177 Objective Loss 0.253177 LR 0.001000 Time 0.021811 +2023-10-02 20:50:58,832 - Epoch: [46][ 630/ 1236] Overall Loss 0.253298 Objective Loss 0.253298 LR 0.001000 Time 0.021788 +2023-10-02 20:50:59,039 - Epoch: [46][ 640/ 1236] Overall Loss 0.253048 Objective Loss 0.253048 LR 0.001000 Time 0.021772 +2023-10-02 20:50:59,244 - Epoch: [46][ 650/ 1236] Overall Loss 0.252927 Objective Loss 0.252927 LR 0.001000 Time 0.021750 +2023-10-02 20:50:59,450 - Epoch: [46][ 660/ 1236] Overall Loss 0.252982 Objective Loss 0.252982 LR 0.001000 Time 0.021733 +2023-10-02 20:50:59,655 - Epoch: [46][ 670/ 1236] Overall Loss 0.252667 Objective Loss 0.252667 LR 0.001000 Time 0.021715 +2023-10-02 20:50:59,862 - Epoch: [46][ 680/ 1236] Overall Loss 0.252445 Objective Loss 0.252445 LR 0.001000 Time 0.021698 +2023-10-02 20:51:00,067 - Epoch: [46][ 690/ 1236] Overall Loss 0.252566 Objective Loss 0.252566 LR 0.001000 Time 0.021681 +2023-10-02 20:51:00,275 - Epoch: [46][ 700/ 1236] Overall Loss 0.252202 Objective Loss 0.252202 LR 0.001000 Time 0.021668 +2023-10-02 20:51:00,479 - Epoch: [46][ 710/ 1236] Overall Loss 0.252809 Objective Loss 0.252809 LR 0.001000 Time 0.021649 +2023-10-02 20:51:00,688 - Epoch: [46][ 720/ 1236] Overall Loss 0.253095 Objective Loss 0.253095 LR 0.001000 Time 0.021638 +2023-10-02 20:51:00,892 - Epoch: [46][ 730/ 1236] Overall Loss 0.253074 Objective Loss 0.253074 LR 0.001000 Time 0.021621 +2023-10-02 20:51:01,098 - Epoch: [46][ 740/ 1236] Overall Loss 0.253489 Objective Loss 0.253489 LR 0.001000 Time 0.021608 +2023-10-02 20:51:01,304 - Epoch: [46][ 750/ 1236] Overall Loss 0.253484 Objective Loss 0.253484 LR 0.001000 Time 0.021593 +2023-10-02 20:51:01,510 - Epoch: [46][ 760/ 1236] Overall Loss 0.253727 Objective Loss 0.253727 LR 0.001000 Time 0.021581 +2023-10-02 20:51:01,716 - Epoch: [46][ 770/ 1236] Overall Loss 0.253971 Objective Loss 0.253971 LR 0.001000 Time 0.021567 +2023-10-02 20:51:01,922 - Epoch: [46][ 780/ 1236] Overall Loss 0.254110 Objective Loss 0.254110 LR 0.001000 Time 0.021555 +2023-10-02 20:51:02,127 - Epoch: [46][ 790/ 1236] Overall Loss 0.254527 Objective Loss 0.254527 LR 0.001000 Time 0.021541 +2023-10-02 20:51:02,334 - Epoch: [46][ 800/ 1236] Overall Loss 0.254697 Objective Loss 0.254697 LR 0.001000 Time 0.021530 +2023-10-02 20:51:02,539 - Epoch: [46][ 810/ 1236] Overall Loss 0.254884 Objective Loss 0.254884 LR 0.001000 Time 0.021515 +2023-10-02 20:51:02,747 - Epoch: [46][ 820/ 1236] Overall Loss 0.255319 Objective Loss 0.255319 LR 0.001000 Time 0.021506 +2023-10-02 20:51:02,951 - Epoch: [46][ 830/ 1236] Overall Loss 0.255602 Objective Loss 0.255602 LR 0.001000 Time 0.021493 +2023-10-02 20:51:03,159 - Epoch: [46][ 840/ 1236] Overall Loss 0.255535 Objective Loss 0.255535 LR 0.001000 Time 0.021484 +2023-10-02 20:51:03,363 - Epoch: [46][ 850/ 1236] Overall Loss 0.255376 Objective Loss 0.255376 LR 0.001000 Time 0.021470 +2023-10-02 20:51:03,569 - Epoch: [46][ 860/ 1236] Overall Loss 0.255144 Objective Loss 0.255144 LR 0.001000 Time 0.021461 +2023-10-02 20:51:03,775 - Epoch: [46][ 870/ 1236] Overall Loss 0.255284 Objective Loss 0.255284 LR 0.001000 Time 0.021450 +2023-10-02 20:51:03,981 - Epoch: [46][ 880/ 1236] Overall Loss 0.255305 Objective Loss 0.255305 LR 0.001000 Time 0.021440 +2023-10-02 20:51:04,186 - Epoch: [46][ 890/ 1236] Overall Loss 0.255263 Objective Loss 0.255263 LR 0.001000 Time 0.021430 +2023-10-02 20:51:04,394 - Epoch: [46][ 900/ 1236] Overall Loss 0.255375 Objective Loss 0.255375 LR 0.001000 Time 0.021422 +2023-10-02 20:51:04,598 - Epoch: [46][ 910/ 1236] Overall Loss 0.255708 Objective Loss 0.255708 LR 0.001000 Time 0.021411 +2023-10-02 20:51:04,806 - Epoch: [46][ 920/ 1236] Overall Loss 0.255743 Objective Loss 0.255743 LR 0.001000 Time 0.021403 +2023-10-02 20:51:05,010 - Epoch: [46][ 930/ 1236] Overall Loss 0.255872 Objective Loss 0.255872 LR 0.001000 Time 0.021393 +2023-10-02 20:51:05,218 - Epoch: [46][ 940/ 1236] Overall Loss 0.256358 Objective Loss 0.256358 LR 0.001000 Time 0.021386 +2023-10-02 20:51:05,422 - Epoch: [46][ 950/ 1236] Overall Loss 0.256241 Objective Loss 0.256241 LR 0.001000 Time 0.021375 +2023-10-02 20:51:05,629 - Epoch: [46][ 960/ 1236] Overall Loss 0.256584 Objective Loss 0.256584 LR 0.001000 Time 0.021368 +2023-10-02 20:51:05,835 - Epoch: [46][ 970/ 1236] Overall Loss 0.256667 Objective Loss 0.256667 LR 0.001000 Time 0.021358 +2023-10-02 20:51:06,041 - Epoch: [46][ 980/ 1236] Overall Loss 0.256791 Objective Loss 0.256791 LR 0.001000 Time 0.021350 +2023-10-02 20:51:06,247 - Epoch: [46][ 990/ 1236] Overall Loss 0.256877 Objective Loss 0.256877 LR 0.001000 Time 0.021342 +2023-10-02 20:51:06,453 - Epoch: [46][ 1000/ 1236] Overall Loss 0.256944 Objective Loss 0.256944 LR 0.001000 Time 0.021335 +2023-10-02 20:51:06,659 - Epoch: [46][ 1010/ 1236] Overall Loss 0.256842 Objective Loss 0.256842 LR 0.001000 Time 0.021326 +2023-10-02 20:51:06,866 - Epoch: [46][ 1020/ 1236] Overall Loss 0.256690 Objective Loss 0.256690 LR 0.001000 Time 0.021319 +2023-10-02 20:51:07,071 - Epoch: [46][ 1030/ 1236] Overall Loss 0.256659 Objective Loss 0.256659 LR 0.001000 Time 0.021311 +2023-10-02 20:51:07,277 - Epoch: [46][ 1040/ 1236] Overall Loss 0.256784 Objective Loss 0.256784 LR 0.001000 Time 0.021304 +2023-10-02 20:51:07,483 - Epoch: [46][ 1050/ 1236] Overall Loss 0.257027 Objective Loss 0.257027 LR 0.001000 Time 0.021297 +2023-10-02 20:51:07,690 - Epoch: [46][ 1060/ 1236] Overall Loss 0.256952 Objective Loss 0.256952 LR 0.001000 Time 0.021291 +2023-10-02 20:51:07,895 - Epoch: [46][ 1070/ 1236] Overall Loss 0.256950 Objective Loss 0.256950 LR 0.001000 Time 0.021284 +2023-10-02 20:51:08,103 - Epoch: [46][ 1080/ 1236] Overall Loss 0.257085 Objective Loss 0.257085 LR 0.001000 Time 0.021279 +2023-10-02 20:51:08,310 - Epoch: [46][ 1090/ 1236] Overall Loss 0.257068 Objective Loss 0.257068 LR 0.001000 Time 0.021273 +2023-10-02 20:51:08,520 - Epoch: [46][ 1100/ 1236] Overall Loss 0.257045 Objective Loss 0.257045 LR 0.001000 Time 0.021270 +2023-10-02 20:51:08,730 - Epoch: [46][ 1110/ 1236] Overall Loss 0.257577 Objective Loss 0.257577 LR 0.001000 Time 0.021266 +2023-10-02 20:51:08,940 - Epoch: [46][ 1120/ 1236] Overall Loss 0.257732 Objective Loss 0.257732 LR 0.001000 Time 0.021263 +2023-10-02 20:51:09,150 - Epoch: [46][ 1130/ 1236] Overall Loss 0.257866 Objective Loss 0.257866 LR 0.001000 Time 0.021260 +2023-10-02 20:51:09,360 - Epoch: [46][ 1140/ 1236] Overall Loss 0.257985 Objective Loss 0.257985 LR 0.001000 Time 0.021258 +2023-10-02 20:51:09,570 - Epoch: [46][ 1150/ 1236] Overall Loss 0.258124 Objective Loss 0.258124 LR 0.001000 Time 0.021255 +2023-10-02 20:51:09,780 - Epoch: [46][ 1160/ 1236] Overall Loss 0.258431 Objective Loss 0.258431 LR 0.001000 Time 0.021253 +2023-10-02 20:51:09,990 - Epoch: [46][ 1170/ 1236] Overall Loss 0.258428 Objective Loss 0.258428 LR 0.001000 Time 0.021250 +2023-10-02 20:51:10,200 - Epoch: [46][ 1180/ 1236] Overall Loss 0.258766 Objective Loss 0.258766 LR 0.001000 Time 0.021248 +2023-10-02 20:51:10,410 - Epoch: [46][ 1190/ 1236] Overall Loss 0.259075 Objective Loss 0.259075 LR 0.001000 Time 0.021246 +2023-10-02 20:51:10,620 - Epoch: [46][ 1200/ 1236] Overall Loss 0.259054 Objective Loss 0.259054 LR 0.001000 Time 0.021243 +2023-10-02 20:51:10,830 - Epoch: [46][ 1210/ 1236] Overall Loss 0.258906 Objective Loss 0.258906 LR 0.001000 Time 0.021241 +2023-10-02 20:51:11,040 - Epoch: [46][ 1220/ 1236] Overall Loss 0.258924 Objective Loss 0.258924 LR 0.001000 Time 0.021239 +2023-10-02 20:51:11,301 - Epoch: [46][ 1230/ 1236] Overall Loss 0.258948 Objective Loss 0.258948 LR 0.001000 Time 0.021277 +2023-10-02 20:51:11,422 - Epoch: [46][ 1236/ 1236] Overall Loss 0.258721 Objective Loss 0.258721 Top1 87.780041 Top5 97.759674 LR 0.001000 Time 0.021272 +2023-10-02 20:51:11,561 - --- validate (epoch=46)----------- +2023-10-02 20:51:11,561 - 29943 samples (256 per mini-batch) +2023-10-02 20:51:12,050 - Epoch: [46][ 10/ 117] Loss 0.307922 Top1 84.257812 Top5 98.320312 +2023-10-02 20:51:12,200 - Epoch: [46][ 20/ 117] Loss 0.335863 Top1 84.023438 Top5 98.320312 +2023-10-02 20:51:12,350 - Epoch: [46][ 30/ 117] Loss 0.338614 Top1 84.062500 Top5 98.307292 +2023-10-02 20:51:12,499 - Epoch: [46][ 40/ 117] Loss 0.329891 Top1 84.072266 Top5 98.378906 +2023-10-02 20:51:12,648 - Epoch: [46][ 50/ 117] Loss 0.330071 Top1 84.218750 Top5 98.289062 +2023-10-02 20:51:12,796 - Epoch: [46][ 60/ 117] Loss 0.328509 Top1 84.303385 Top5 98.274740 +2023-10-02 20:51:12,945 - Epoch: [46][ 70/ 117] Loss 0.328725 Top1 84.414062 Top5 98.364955 +2023-10-02 20:51:13,094 - Epoch: [46][ 80/ 117] Loss 0.323588 Top1 84.462891 Top5 98.354492 +2023-10-02 20:51:13,242 - Epoch: [46][ 90/ 117] Loss 0.323136 Top1 84.288194 Top5 98.320312 +2023-10-02 20:51:13,391 - Epoch: [46][ 100/ 117] Loss 0.324291 Top1 84.320312 Top5 98.281250 +2023-10-02 20:51:13,546 - Epoch: [46][ 110/ 117] Loss 0.324144 Top1 84.350142 Top5 98.291903 +2023-10-02 20:51:13,634 - Epoch: [46][ 117/ 117] Loss 0.323321 Top1 84.290151 Top5 98.313462 +2023-10-02 20:51:13,774 - ==> Top1: 84.290 Top5: 98.313 Loss: 0.323 + +2023-10-02 20:51:13,775 - ==> Confusion: +[[ 935 1 0 0 5 3 0 1 4 65 2 0 0 4 10 1 2 3 0 0 14] + [ 0 1036 3 1 10 22 1 19 3 1 3 1 1 0 2 3 3 1 12 0 9] + [ 3 1 951 10 1 2 32 11 0 1 3 0 7 0 2 10 0 1 11 1 9] + [ 2 2 13 972 3 4 3 2 5 0 6 0 6 3 26 3 1 7 14 0 17] + [ 20 6 2 0 968 6 0 0 1 16 1 0 2 5 8 4 7 0 0 1 3] + [ 2 40 0 1 2 987 2 25 2 5 3 4 4 6 9 0 1 0 5 4 14] + [ 0 3 30 0 0 0 1122 6 0 0 6 2 1 0 0 9 0 1 1 4 6] + [ 3 21 20 0 10 28 4 1040 0 2 10 4 2 2 2 0 0 0 56 7 7] + [ 20 2 0 0 1 6 0 1 967 38 6 1 5 10 22 1 2 3 3 0 1] + [ 91 0 0 0 9 3 0 0 23 950 2 2 0 16 6 1 2 0 1 4 9] + [ 2 1 10 13 3 1 1 1 12 0 976 3 0 6 5 0 0 3 6 0 10] + [ 0 0 4 0 1 15 0 9 0 1 0 912 55 8 0 3 1 16 0 3 7] + [ 0 2 9 3 1 3 1 2 1 2 2 31 974 1 3 8 1 7 2 2 13] + [ 2 0 0 1 7 11 1 2 10 12 9 5 2 1034 7 1 1 1 0 2 11] + [ 11 1 3 10 3 1 0 0 19 2 3 0 4 3 1017 0 0 1 10 0 13] + [ 0 0 3 0 5 2 2 0 0 0 0 5 6 1 0 1073 12 11 3 4 7] + [ 1 10 1 0 11 7 0 0 0 0 0 4 3 1 4 15 1075 1 2 4 22] + [ 1 0 0 1 0 0 0 0 0 0 2 4 26 1 4 6 0 984 2 2 5] + [ 2 1 4 9 1 0 0 15 4 0 2 0 2 0 14 0 0 0 1002 1 11] + [ 0 4 7 2 1 6 10 23 0 1 3 8 9 2 2 3 4 0 1 1050 16] + [ 160 136 129 77 112 184 44 86 104 83 174 103 382 242 159 90 62 57 173 134 5214]] + +2023-10-02 20:51:13,776 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:51:13,776 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:51:13,790 - + +2023-10-02 20:51:13,790 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:51:14,931 - Epoch: [47][ 10/ 1236] Overall Loss 0.240251 Objective Loss 0.240251 LR 0.001000 Time 0.114084 +2023-10-02 20:51:15,137 - Epoch: [47][ 20/ 1236] Overall Loss 0.239540 Objective Loss 0.239540 LR 0.001000 Time 0.067323 +2023-10-02 20:51:15,342 - Epoch: [47][ 30/ 1236] Overall Loss 0.244203 Objective Loss 0.244203 LR 0.001000 Time 0.051713 +2023-10-02 20:51:15,548 - Epoch: [47][ 40/ 1236] Overall Loss 0.239786 Objective Loss 0.239786 LR 0.001000 Time 0.043917 +2023-10-02 20:51:15,753 - Epoch: [47][ 50/ 1236] Overall Loss 0.244959 Objective Loss 0.244959 LR 0.001000 Time 0.039223 +2023-10-02 20:51:15,959 - Epoch: [47][ 60/ 1236] Overall Loss 0.247925 Objective Loss 0.247925 LR 0.001000 Time 0.036126 +2023-10-02 20:51:16,171 - Epoch: [47][ 70/ 1236] Overall Loss 0.245685 Objective Loss 0.245685 LR 0.001000 Time 0.033985 +2023-10-02 20:51:16,386 - Epoch: [47][ 80/ 1236] Overall Loss 0.249618 Objective Loss 0.249618 LR 0.001000 Time 0.032405 +2023-10-02 20:51:16,601 - Epoch: [47][ 90/ 1236] Overall Loss 0.250365 Objective Loss 0.250365 LR 0.001000 Time 0.031167 +2023-10-02 20:51:16,817 - Epoch: [47][ 100/ 1236] Overall Loss 0.245719 Objective Loss 0.245719 LR 0.001000 Time 0.030191 +2023-10-02 20:51:17,031 - Epoch: [47][ 110/ 1236] Overall Loss 0.245960 Objective Loss 0.245960 LR 0.001000 Time 0.029382 +2023-10-02 20:51:17,244 - Epoch: [47][ 120/ 1236] Overall Loss 0.245268 Objective Loss 0.245268 LR 0.001000 Time 0.028702 +2023-10-02 20:51:17,455 - Epoch: [47][ 130/ 1236] Overall Loss 0.244456 Objective Loss 0.244456 LR 0.001000 Time 0.028112 +2023-10-02 20:51:17,666 - Epoch: [47][ 140/ 1236] Overall Loss 0.244769 Objective Loss 0.244769 LR 0.001000 Time 0.027613 +2023-10-02 20:51:17,878 - Epoch: [47][ 150/ 1236] Overall Loss 0.244249 Objective Loss 0.244249 LR 0.001000 Time 0.027182 +2023-10-02 20:51:18,089 - Epoch: [47][ 160/ 1236] Overall Loss 0.245053 Objective Loss 0.245053 LR 0.001000 Time 0.026802 +2023-10-02 20:51:18,300 - Epoch: [47][ 170/ 1236] Overall Loss 0.246147 Objective Loss 0.246147 LR 0.001000 Time 0.026465 +2023-10-02 20:51:18,512 - Epoch: [47][ 180/ 1236] Overall Loss 0.246110 Objective Loss 0.246110 LR 0.001000 Time 0.026170 +2023-10-02 20:51:18,723 - Epoch: [47][ 190/ 1236] Overall Loss 0.248996 Objective Loss 0.248996 LR 0.001000 Time 0.025899 +2023-10-02 20:51:18,936 - Epoch: [47][ 200/ 1236] Overall Loss 0.247498 Objective Loss 0.247498 LR 0.001000 Time 0.025666 +2023-10-02 20:51:19,147 - Epoch: [47][ 210/ 1236] Overall Loss 0.248648 Objective Loss 0.248648 LR 0.001000 Time 0.025446 +2023-10-02 20:51:19,359 - Epoch: [47][ 220/ 1236] Overall Loss 0.249131 Objective Loss 0.249131 LR 0.001000 Time 0.025253 +2023-10-02 20:51:19,570 - Epoch: [47][ 230/ 1236] Overall Loss 0.249956 Objective Loss 0.249956 LR 0.001000 Time 0.025069 +2023-10-02 20:51:19,781 - Epoch: [47][ 240/ 1236] Overall Loss 0.250978 Objective Loss 0.250978 LR 0.001000 Time 0.024905 +2023-10-02 20:51:19,992 - Epoch: [47][ 250/ 1236] Overall Loss 0.250918 Objective Loss 0.250918 LR 0.001000 Time 0.024750 +2023-10-02 20:51:20,205 - Epoch: [47][ 260/ 1236] Overall Loss 0.252997 Objective Loss 0.252997 LR 0.001000 Time 0.024614 +2023-10-02 20:51:20,416 - Epoch: [47][ 270/ 1236] Overall Loss 0.253103 Objective Loss 0.253103 LR 0.001000 Time 0.024482 +2023-10-02 20:51:20,628 - Epoch: [47][ 280/ 1236] Overall Loss 0.254427 Objective Loss 0.254427 LR 0.001000 Time 0.024365 +2023-10-02 20:51:20,840 - Epoch: [47][ 290/ 1236] Overall Loss 0.254950 Objective Loss 0.254950 LR 0.001000 Time 0.024253 +2023-10-02 20:51:21,051 - Epoch: [47][ 300/ 1236] Overall Loss 0.254369 Objective Loss 0.254369 LR 0.001000 Time 0.024149 +2023-10-02 20:51:21,264 - Epoch: [47][ 310/ 1236] Overall Loss 0.254013 Objective Loss 0.254013 LR 0.001000 Time 0.024053 +2023-10-02 20:51:21,473 - Epoch: [47][ 320/ 1236] Overall Loss 0.255397 Objective Loss 0.255397 LR 0.001000 Time 0.023957 +2023-10-02 20:51:21,687 - Epoch: [47][ 330/ 1236] Overall Loss 0.254557 Objective Loss 0.254557 LR 0.001000 Time 0.023873 +2023-10-02 20:51:21,897 - Epoch: [47][ 340/ 1236] Overall Loss 0.254340 Objective Loss 0.254340 LR 0.001000 Time 0.023789 +2023-10-02 20:51:22,110 - Epoch: [47][ 350/ 1236] Overall Loss 0.253968 Objective Loss 0.253968 LR 0.001000 Time 0.023713 +2023-10-02 20:51:22,321 - Epoch: [47][ 360/ 1236] Overall Loss 0.253578 Objective Loss 0.253578 LR 0.001000 Time 0.023638 +2023-10-02 20:51:22,532 - Epoch: [47][ 370/ 1236] Overall Loss 0.253983 Objective Loss 0.253983 LR 0.001000 Time 0.023570 +2023-10-02 20:51:22,745 - Epoch: [47][ 380/ 1236] Overall Loss 0.254464 Objective Loss 0.254464 LR 0.001000 Time 0.023509 +2023-10-02 20:51:22,957 - Epoch: [47][ 390/ 1236] Overall Loss 0.254173 Objective Loss 0.254173 LR 0.001000 Time 0.023448 +2023-10-02 20:51:23,168 - Epoch: [47][ 400/ 1236] Overall Loss 0.255146 Objective Loss 0.255146 LR 0.001000 Time 0.023388 +2023-10-02 20:51:23,379 - Epoch: [47][ 410/ 1236] Overall Loss 0.254726 Objective Loss 0.254726 LR 0.001000 Time 0.023332 +2023-10-02 20:51:23,592 - Epoch: [47][ 420/ 1236] Overall Loss 0.254987 Objective Loss 0.254987 LR 0.001000 Time 0.023282 +2023-10-02 20:51:23,804 - Epoch: [47][ 430/ 1236] Overall Loss 0.255045 Objective Loss 0.255045 LR 0.001000 Time 0.023234 +2023-10-02 20:51:24,016 - Epoch: [47][ 440/ 1236] Overall Loss 0.253944 Objective Loss 0.253944 LR 0.001000 Time 0.023187 +2023-10-02 20:51:24,227 - Epoch: [47][ 450/ 1236] Overall Loss 0.253430 Objective Loss 0.253430 LR 0.001000 Time 0.023139 +2023-10-02 20:51:24,438 - Epoch: [47][ 460/ 1236] Overall Loss 0.253610 Objective Loss 0.253610 LR 0.001000 Time 0.023092 +2023-10-02 20:51:24,650 - Epoch: [47][ 470/ 1236] Overall Loss 0.253311 Objective Loss 0.253311 LR 0.001000 Time 0.023050 +2023-10-02 20:51:24,860 - Epoch: [47][ 480/ 1236] Overall Loss 0.252857 Objective Loss 0.252857 LR 0.001000 Time 0.023005 +2023-10-02 20:51:25,071 - Epoch: [47][ 490/ 1236] Overall Loss 0.253051 Objective Loss 0.253051 LR 0.001000 Time 0.022965 +2023-10-02 20:51:25,282 - Epoch: [47][ 500/ 1236] Overall Loss 0.253708 Objective Loss 0.253708 LR 0.001000 Time 0.022925 +2023-10-02 20:51:25,494 - Epoch: [47][ 510/ 1236] Overall Loss 0.254046 Objective Loss 0.254046 LR 0.001000 Time 0.022889 +2023-10-02 20:51:25,702 - Epoch: [47][ 520/ 1236] Overall Loss 0.253962 Objective Loss 0.253962 LR 0.001000 Time 0.022850 +2023-10-02 20:51:25,914 - Epoch: [47][ 530/ 1236] Overall Loss 0.253857 Objective Loss 0.253857 LR 0.001000 Time 0.022817 +2023-10-02 20:51:26,123 - Epoch: [47][ 540/ 1236] Overall Loss 0.253585 Objective Loss 0.253585 LR 0.001000 Time 0.022781 +2023-10-02 20:51:26,334 - Epoch: [47][ 550/ 1236] Overall Loss 0.253684 Objective Loss 0.253684 LR 0.001000 Time 0.022748 +2023-10-02 20:51:26,552 - Epoch: [47][ 560/ 1236] Overall Loss 0.253673 Objective Loss 0.253673 LR 0.001000 Time 0.022731 +2023-10-02 20:51:26,783 - Epoch: [47][ 570/ 1236] Overall Loss 0.253936 Objective Loss 0.253936 LR 0.001000 Time 0.022736 +2023-10-02 20:51:27,014 - Epoch: [47][ 580/ 1236] Overall Loss 0.253877 Objective Loss 0.253877 LR 0.001000 Time 0.022741 +2023-10-02 20:51:27,239 - Epoch: [47][ 590/ 1236] Overall Loss 0.254197 Objective Loss 0.254197 LR 0.001000 Time 0.022734 +2023-10-02 20:51:27,467 - Epoch: [47][ 600/ 1236] Overall Loss 0.254404 Objective Loss 0.254404 LR 0.001000 Time 0.022735 +2023-10-02 20:51:27,691 - Epoch: [47][ 610/ 1236] Overall Loss 0.254881 Objective Loss 0.254881 LR 0.001000 Time 0.022727 +2023-10-02 20:51:27,916 - Epoch: [47][ 620/ 1236] Overall Loss 0.254656 Objective Loss 0.254656 LR 0.001000 Time 0.022722 +2023-10-02 20:51:28,128 - Epoch: [47][ 630/ 1236] Overall Loss 0.254610 Objective Loss 0.254610 LR 0.001000 Time 0.022699 +2023-10-02 20:51:28,342 - Epoch: [47][ 640/ 1236] Overall Loss 0.254581 Objective Loss 0.254581 LR 0.001000 Time 0.022677 +2023-10-02 20:51:28,554 - Epoch: [47][ 650/ 1236] Overall Loss 0.254867 Objective Loss 0.254867 LR 0.001000 Time 0.022654 +2023-10-02 20:51:28,767 - Epoch: [47][ 660/ 1236] Overall Loss 0.255194 Objective Loss 0.255194 LR 0.001000 Time 0.022632 +2023-10-02 20:51:28,980 - Epoch: [47][ 670/ 1236] Overall Loss 0.255235 Objective Loss 0.255235 LR 0.001000 Time 0.022611 +2023-10-02 20:51:29,192 - Epoch: [47][ 680/ 1236] Overall Loss 0.255259 Objective Loss 0.255259 LR 0.001000 Time 0.022589 +2023-10-02 20:51:29,402 - Epoch: [47][ 690/ 1236] Overall Loss 0.255518 Objective Loss 0.255518 LR 0.001000 Time 0.022566 +2023-10-02 20:51:29,615 - Epoch: [47][ 700/ 1236] Overall Loss 0.255390 Objective Loss 0.255390 LR 0.001000 Time 0.022547 +2023-10-02 20:51:29,828 - Epoch: [47][ 710/ 1236] Overall Loss 0.255420 Objective Loss 0.255420 LR 0.001000 Time 0.022528 +2023-10-02 20:51:30,040 - Epoch: [47][ 720/ 1236] Overall Loss 0.255355 Objective Loss 0.255355 LR 0.001000 Time 0.022509 +2023-10-02 20:51:30,251 - Epoch: [47][ 730/ 1236] Overall Loss 0.255240 Objective Loss 0.255240 LR 0.001000 Time 0.022487 +2023-10-02 20:51:30,463 - Epoch: [47][ 740/ 1236] Overall Loss 0.255557 Objective Loss 0.255557 LR 0.001000 Time 0.022469 +2023-10-02 20:51:30,674 - Epoch: [47][ 750/ 1236] Overall Loss 0.256049 Objective Loss 0.256049 LR 0.001000 Time 0.022451 +2023-10-02 20:51:30,885 - Epoch: [47][ 760/ 1236] Overall Loss 0.256165 Objective Loss 0.256165 LR 0.001000 Time 0.022433 +2023-10-02 20:51:31,096 - Epoch: [47][ 770/ 1236] Overall Loss 0.256362 Objective Loss 0.256362 LR 0.001000 Time 0.022413 +2023-10-02 20:51:31,308 - Epoch: [47][ 780/ 1236] Overall Loss 0.256715 Objective Loss 0.256715 LR 0.001000 Time 0.022397 +2023-10-02 20:51:31,518 - Epoch: [47][ 790/ 1236] Overall Loss 0.257034 Objective Loss 0.257034 LR 0.001000 Time 0.022379 +2023-10-02 20:51:31,730 - Epoch: [47][ 800/ 1236] Overall Loss 0.256481 Objective Loss 0.256481 LR 0.001000 Time 0.022364 +2023-10-02 20:51:31,941 - Epoch: [47][ 810/ 1236] Overall Loss 0.256658 Objective Loss 0.256658 LR 0.001000 Time 0.022348 +2023-10-02 20:51:32,153 - Epoch: [47][ 820/ 1236] Overall Loss 0.256511 Objective Loss 0.256511 LR 0.001000 Time 0.022333 +2023-10-02 20:51:32,364 - Epoch: [47][ 830/ 1236] Overall Loss 0.256359 Objective Loss 0.256359 LR 0.001000 Time 0.022318 +2023-10-02 20:51:32,577 - Epoch: [47][ 840/ 1236] Overall Loss 0.256094 Objective Loss 0.256094 LR 0.001000 Time 0.022304 +2023-10-02 20:51:32,788 - Epoch: [47][ 850/ 1236] Overall Loss 0.255898 Objective Loss 0.255898 LR 0.001000 Time 0.022290 +2023-10-02 20:51:33,001 - Epoch: [47][ 860/ 1236] Overall Loss 0.255801 Objective Loss 0.255801 LR 0.001000 Time 0.022277 +2023-10-02 20:51:33,212 - Epoch: [47][ 870/ 1236] Overall Loss 0.255917 Objective Loss 0.255917 LR 0.001000 Time 0.022264 +2023-10-02 20:51:33,425 - Epoch: [47][ 880/ 1236] Overall Loss 0.255942 Objective Loss 0.255942 LR 0.001000 Time 0.022252 +2023-10-02 20:51:33,636 - Epoch: [47][ 890/ 1236] Overall Loss 0.256037 Objective Loss 0.256037 LR 0.001000 Time 0.022239 +2023-10-02 20:51:33,849 - Epoch: [47][ 900/ 1236] Overall Loss 0.256126 Objective Loss 0.256126 LR 0.001000 Time 0.022227 +2023-10-02 20:51:34,060 - Epoch: [47][ 910/ 1236] Overall Loss 0.256236 Objective Loss 0.256236 LR 0.001000 Time 0.022215 +2023-10-02 20:51:34,273 - Epoch: [47][ 920/ 1236] Overall Loss 0.256226 Objective Loss 0.256226 LR 0.001000 Time 0.022204 +2023-10-02 20:51:34,484 - Epoch: [47][ 930/ 1236] Overall Loss 0.256333 Objective Loss 0.256333 LR 0.001000 Time 0.022193 +2023-10-02 20:51:34,697 - Epoch: [47][ 940/ 1236] Overall Loss 0.256370 Objective Loss 0.256370 LR 0.001000 Time 0.022182 +2023-10-02 20:51:34,908 - Epoch: [47][ 950/ 1236] Overall Loss 0.256520 Objective Loss 0.256520 LR 0.001000 Time 0.022170 +2023-10-02 20:51:35,121 - Epoch: [47][ 960/ 1236] Overall Loss 0.256717 Objective Loss 0.256717 LR 0.001000 Time 0.022160 +2023-10-02 20:51:35,333 - Epoch: [47][ 970/ 1236] Overall Loss 0.256871 Objective Loss 0.256871 LR 0.001000 Time 0.022149 +2023-10-02 20:51:35,546 - Epoch: [47][ 980/ 1236] Overall Loss 0.256891 Objective Loss 0.256891 LR 0.001000 Time 0.022140 +2023-10-02 20:51:35,756 - Epoch: [47][ 990/ 1236] Overall Loss 0.256948 Objective Loss 0.256948 LR 0.001000 Time 0.022129 +2023-10-02 20:51:35,969 - Epoch: [47][ 1000/ 1236] Overall Loss 0.256845 Objective Loss 0.256845 LR 0.001000 Time 0.022120 +2023-10-02 20:51:36,181 - Epoch: [47][ 1010/ 1236] Overall Loss 0.256539 Objective Loss 0.256539 LR 0.001000 Time 0.022110 +2023-10-02 20:51:36,394 - Epoch: [47][ 1020/ 1236] Overall Loss 0.256523 Objective Loss 0.256523 LR 0.001000 Time 0.022102 +2023-10-02 20:51:36,605 - Epoch: [47][ 1030/ 1236] Overall Loss 0.256799 Objective Loss 0.256799 LR 0.001000 Time 0.022092 +2023-10-02 20:51:36,819 - Epoch: [47][ 1040/ 1236] Overall Loss 0.257039 Objective Loss 0.257039 LR 0.001000 Time 0.022085 +2023-10-02 20:51:37,033 - Epoch: [47][ 1050/ 1236] Overall Loss 0.257227 Objective Loss 0.257227 LR 0.001000 Time 0.022078 +2023-10-02 20:51:37,247 - Epoch: [47][ 1060/ 1236] Overall Loss 0.257182 Objective Loss 0.257182 LR 0.001000 Time 0.022072 +2023-10-02 20:51:37,461 - Epoch: [47][ 1070/ 1236] Overall Loss 0.257262 Objective Loss 0.257262 LR 0.001000 Time 0.022065 +2023-10-02 20:51:37,676 - Epoch: [47][ 1080/ 1236] Overall Loss 0.257114 Objective Loss 0.257114 LR 0.001000 Time 0.022059 +2023-10-02 20:51:37,889 - Epoch: [47][ 1090/ 1236] Overall Loss 0.257100 Objective Loss 0.257100 LR 0.001000 Time 0.022052 +2023-10-02 20:51:38,104 - Epoch: [47][ 1100/ 1236] Overall Loss 0.257186 Objective Loss 0.257186 LR 0.001000 Time 0.022046 +2023-10-02 20:51:38,317 - Epoch: [47][ 1110/ 1236] Overall Loss 0.257200 Objective Loss 0.257200 LR 0.001000 Time 0.022039 +2023-10-02 20:51:38,531 - Epoch: [47][ 1120/ 1236] Overall Loss 0.257562 Objective Loss 0.257562 LR 0.001000 Time 0.022033 +2023-10-02 20:51:38,745 - Epoch: [47][ 1130/ 1236] Overall Loss 0.257408 Objective Loss 0.257408 LR 0.001000 Time 0.022027 +2023-10-02 20:51:38,960 - Epoch: [47][ 1140/ 1236] Overall Loss 0.257600 Objective Loss 0.257600 LR 0.001000 Time 0.022022 +2023-10-02 20:51:39,173 - Epoch: [47][ 1150/ 1236] Overall Loss 0.257395 Objective Loss 0.257395 LR 0.001000 Time 0.022016 +2023-10-02 20:51:39,388 - Epoch: [47][ 1160/ 1236] Overall Loss 0.257245 Objective Loss 0.257245 LR 0.001000 Time 0.022010 +2023-10-02 20:51:39,602 - Epoch: [47][ 1170/ 1236] Overall Loss 0.257389 Objective Loss 0.257389 LR 0.001000 Time 0.022005 +2023-10-02 20:51:39,816 - Epoch: [47][ 1180/ 1236] Overall Loss 0.257674 Objective Loss 0.257674 LR 0.001000 Time 0.022000 +2023-10-02 20:51:40,030 - Epoch: [47][ 1190/ 1236] Overall Loss 0.257627 Objective Loss 0.257627 LR 0.001000 Time 0.021994 +2023-10-02 20:51:40,244 - Epoch: [47][ 1200/ 1236] Overall Loss 0.257534 Objective Loss 0.257534 LR 0.001000 Time 0.021989 +2023-10-02 20:51:40,458 - Epoch: [47][ 1210/ 1236] Overall Loss 0.257589 Objective Loss 0.257589 LR 0.001000 Time 0.021984 +2023-10-02 20:51:40,672 - Epoch: [47][ 1220/ 1236] Overall Loss 0.257817 Objective Loss 0.257817 LR 0.001000 Time 0.021978 +2023-10-02 20:51:40,938 - Epoch: [47][ 1230/ 1236] Overall Loss 0.257988 Objective Loss 0.257988 LR 0.001000 Time 0.022016 +2023-10-02 20:51:41,059 - Epoch: [47][ 1236/ 1236] Overall Loss 0.258065 Objective Loss 0.258065 Top1 84.725051 Top5 98.370672 LR 0.001000 Time 0.022007 +2023-10-02 20:51:41,159 - --- validate (epoch=47)----------- +2023-10-02 20:51:41,159 - 29943 samples (256 per mini-batch) +2023-10-02 20:51:41,633 - Epoch: [47][ 10/ 117] Loss 0.318775 Top1 82.539062 Top5 98.046875 +2023-10-02 20:51:41,781 - Epoch: [47][ 20/ 117] Loss 0.300027 Top1 83.261719 Top5 98.105469 +2023-10-02 20:51:41,928 - Epoch: [47][ 30/ 117] Loss 0.311644 Top1 83.567708 Top5 98.111979 +2023-10-02 20:51:42,075 - Epoch: [47][ 40/ 117] Loss 0.307447 Top1 83.632812 Top5 98.232422 +2023-10-02 20:51:42,223 - Epoch: [47][ 50/ 117] Loss 0.312549 Top1 83.632812 Top5 98.148438 +2023-10-02 20:51:42,370 - Epoch: [47][ 60/ 117] Loss 0.320999 Top1 83.359375 Top5 98.085938 +2023-10-02 20:51:42,518 - Epoch: [47][ 70/ 117] Loss 0.326646 Top1 83.415179 Top5 98.058036 +2023-10-02 20:51:42,666 - Epoch: [47][ 80/ 117] Loss 0.327543 Top1 83.417969 Top5 98.037109 +2023-10-02 20:51:42,813 - Epoch: [47][ 90/ 117] Loss 0.326359 Top1 83.424479 Top5 98.025174 +2023-10-02 20:51:42,959 - Epoch: [47][ 100/ 117] Loss 0.325186 Top1 83.433594 Top5 98.011719 +2023-10-02 20:51:43,112 - Epoch: [47][ 110/ 117] Loss 0.322769 Top1 83.522727 Top5 98.029119 +2023-10-02 20:51:43,200 - Epoch: [47][ 117/ 117] Loss 0.323749 Top1 83.485289 Top5 98.042948 +2023-10-02 20:51:43,342 - ==> Top1: 83.485 Top5: 98.043 Loss: 0.324 + +2023-10-02 20:51:43,342 - ==> Confusion: +[[ 917 2 2 2 10 2 0 2 8 59 5 2 2 4 9 0 4 1 0 0 19] + [ 0 1072 0 2 5 18 1 12 2 0 2 1 2 0 1 2 1 0 4 3 3] + [ 5 0 932 17 4 2 25 9 0 0 6 1 7 3 5 4 1 0 17 4 14] + [ 0 5 10 983 2 3 1 1 2 0 9 0 7 3 26 2 2 3 12 3 15] + [ 22 7 2 1 957 7 0 2 1 7 2 0 3 4 17 4 8 0 0 0 6] + [ 2 59 0 0 2 970 1 27 0 4 2 11 3 7 5 0 6 0 3 5 9] + [ 1 5 34 0 0 1 1108 6 1 0 6 2 1 1 0 4 0 1 1 13 6] + [ 2 33 8 0 3 23 3 1042 2 1 7 6 7 1 1 3 0 0 56 12 8] + [ 20 3 0 0 0 3 0 1 987 21 13 0 5 13 13 1 2 4 2 0 1] + [ 96 3 0 2 4 3 0 1 48 908 4 0 0 26 12 1 0 1 0 2 8] + [ 2 0 1 9 0 0 2 1 11 0 984 2 1 7 6 0 1 5 11 3 7] + [ 0 0 2 0 0 3 0 5 0 1 0 946 25 9 0 3 2 21 0 10 8] + [ 0 1 4 2 0 1 0 1 1 0 1 49 961 1 0 10 2 15 5 6 8] + [ 0 0 3 0 1 10 0 0 20 9 7 10 1 1029 8 2 4 1 0 2 12] + [ 5 1 0 15 3 0 0 0 26 0 3 0 3 2 1026 0 2 2 5 0 8] + [ 1 0 1 2 4 1 2 0 0 0 0 6 7 0 0 1065 19 14 1 6 5] + [ 0 23 0 0 3 2 1 0 1 0 1 6 0 2 4 11 1091 1 0 5 10] + [ 0 0 1 4 0 0 0 0 1 0 0 4 22 1 2 4 2 993 2 1 1] + [ 0 8 4 11 2 0 1 9 4 0 2 1 5 0 15 0 0 0 996 0 10] + [ 0 3 1 0 0 3 6 8 0 0 3 7 4 0 1 4 9 1 2 1097 3] + [ 97 256 82 98 74 148 27 82 143 48 207 108 410 237 224 65 186 78 182 219 4934]] + +2023-10-02 20:51:43,344 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:51:43,344 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:51:43,350 - + +2023-10-02 20:51:43,350 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:51:44,358 - Epoch: [48][ 10/ 1236] Overall Loss 0.217819 Objective Loss 0.217819 LR 0.001000 Time 0.100744 +2023-10-02 20:51:44,566 - Epoch: [48][ 20/ 1236] Overall Loss 0.214623 Objective Loss 0.214623 LR 0.001000 Time 0.060756 +2023-10-02 20:51:44,773 - Epoch: [48][ 30/ 1236] Overall Loss 0.220421 Objective Loss 0.220421 LR 0.001000 Time 0.047352 +2023-10-02 20:51:44,982 - Epoch: [48][ 40/ 1236] Overall Loss 0.232314 Objective Loss 0.232314 LR 0.001000 Time 0.040738 +2023-10-02 20:51:45,187 - Epoch: [48][ 50/ 1236] Overall Loss 0.231076 Objective Loss 0.231076 LR 0.001000 Time 0.036692 +2023-10-02 20:51:45,397 - Epoch: [48][ 60/ 1236] Overall Loss 0.228848 Objective Loss 0.228848 LR 0.001000 Time 0.034059 +2023-10-02 20:51:45,602 - Epoch: [48][ 70/ 1236] Overall Loss 0.235234 Objective Loss 0.235234 LR 0.001000 Time 0.032124 +2023-10-02 20:51:45,811 - Epoch: [48][ 80/ 1236] Overall Loss 0.233620 Objective Loss 0.233620 LR 0.001000 Time 0.030718 +2023-10-02 20:51:46,017 - Epoch: [48][ 90/ 1236] Overall Loss 0.236153 Objective Loss 0.236153 LR 0.001000 Time 0.029587 +2023-10-02 20:51:46,227 - Epoch: [48][ 100/ 1236] Overall Loss 0.239118 Objective Loss 0.239118 LR 0.001000 Time 0.028734 +2023-10-02 20:51:46,437 - Epoch: [48][ 110/ 1236] Overall Loss 0.238603 Objective Loss 0.238603 LR 0.001000 Time 0.028020 +2023-10-02 20:51:46,647 - Epoch: [48][ 120/ 1236] Overall Loss 0.237214 Objective Loss 0.237214 LR 0.001000 Time 0.027432 +2023-10-02 20:51:46,852 - Epoch: [48][ 130/ 1236] Overall Loss 0.239128 Objective Loss 0.239128 LR 0.001000 Time 0.026901 +2023-10-02 20:51:47,059 - Epoch: [48][ 140/ 1236] Overall Loss 0.242050 Objective Loss 0.242050 LR 0.001000 Time 0.026454 +2023-10-02 20:51:47,268 - Epoch: [48][ 150/ 1236] Overall Loss 0.240846 Objective Loss 0.240846 LR 0.001000 Time 0.026079 +2023-10-02 20:51:47,475 - Epoch: [48][ 160/ 1236] Overall Loss 0.239537 Objective Loss 0.239537 LR 0.001000 Time 0.025742 +2023-10-02 20:51:47,680 - Epoch: [48][ 170/ 1236] Overall Loss 0.239169 Objective Loss 0.239169 LR 0.001000 Time 0.025435 +2023-10-02 20:51:47,893 - Epoch: [48][ 180/ 1236] Overall Loss 0.238836 Objective Loss 0.238836 LR 0.001000 Time 0.025203 +2023-10-02 20:51:48,099 - Epoch: [48][ 190/ 1236] Overall Loss 0.239414 Objective Loss 0.239414 LR 0.001000 Time 0.024958 +2023-10-02 20:51:48,306 - Epoch: [48][ 200/ 1236] Overall Loss 0.240356 Objective Loss 0.240356 LR 0.001000 Time 0.024744 +2023-10-02 20:51:48,512 - Epoch: [48][ 210/ 1236] Overall Loss 0.241025 Objective Loss 0.241025 LR 0.001000 Time 0.024543 +2023-10-02 20:51:48,719 - Epoch: [48][ 220/ 1236] Overall Loss 0.242820 Objective Loss 0.242820 LR 0.001000 Time 0.024366 +2023-10-02 20:51:48,924 - Epoch: [48][ 230/ 1236] Overall Loss 0.242625 Objective Loss 0.242625 LR 0.001000 Time 0.024198 +2023-10-02 20:51:49,132 - Epoch: [48][ 240/ 1236] Overall Loss 0.243429 Objective Loss 0.243429 LR 0.001000 Time 0.024052 +2023-10-02 20:51:49,337 - Epoch: [48][ 250/ 1236] Overall Loss 0.244898 Objective Loss 0.244898 LR 0.001000 Time 0.023910 +2023-10-02 20:51:49,544 - Epoch: [48][ 260/ 1236] Overall Loss 0.244797 Objective Loss 0.244797 LR 0.001000 Time 0.023785 +2023-10-02 20:51:49,753 - Epoch: [48][ 270/ 1236] Overall Loss 0.245403 Objective Loss 0.245403 LR 0.001000 Time 0.023678 +2023-10-02 20:51:49,971 - Epoch: [48][ 280/ 1236] Overall Loss 0.245508 Objective Loss 0.245508 LR 0.001000 Time 0.023610 +2023-10-02 20:51:50,183 - Epoch: [48][ 290/ 1236] Overall Loss 0.245371 Objective Loss 0.245371 LR 0.001000 Time 0.023524 +2023-10-02 20:51:50,400 - Epoch: [48][ 300/ 1236] Overall Loss 0.245038 Objective Loss 0.245038 LR 0.001000 Time 0.023462 +2023-10-02 20:51:50,612 - Epoch: [48][ 310/ 1236] Overall Loss 0.245081 Objective Loss 0.245081 LR 0.001000 Time 0.023388 +2023-10-02 20:51:50,829 - Epoch: [48][ 320/ 1236] Overall Loss 0.245394 Objective Loss 0.245394 LR 0.001000 Time 0.023334 +2023-10-02 20:51:51,041 - Epoch: [48][ 330/ 1236] Overall Loss 0.246063 Objective Loss 0.246063 LR 0.001000 Time 0.023271 +2023-10-02 20:51:51,262 - Epoch: [48][ 340/ 1236] Overall Loss 0.246070 Objective Loss 0.246070 LR 0.001000 Time 0.023235 +2023-10-02 20:51:51,477 - Epoch: [48][ 350/ 1236] Overall Loss 0.247173 Objective Loss 0.247173 LR 0.001000 Time 0.023184 +2023-10-02 20:51:51,695 - Epoch: [48][ 360/ 1236] Overall Loss 0.247112 Objective Loss 0.247112 LR 0.001000 Time 0.023145 +2023-10-02 20:51:51,911 - Epoch: [48][ 370/ 1236] Overall Loss 0.247254 Objective Loss 0.247254 LR 0.001000 Time 0.023101 +2023-10-02 20:51:52,130 - Epoch: [48][ 380/ 1236] Overall Loss 0.247551 Objective Loss 0.247551 LR 0.001000 Time 0.023069 +2023-10-02 20:51:52,346 - Epoch: [48][ 390/ 1236] Overall Loss 0.247092 Objective Loss 0.247092 LR 0.001000 Time 0.023030 +2023-10-02 20:51:52,566 - Epoch: [48][ 400/ 1236] Overall Loss 0.247331 Objective Loss 0.247331 LR 0.001000 Time 0.023004 +2023-10-02 20:51:52,782 - Epoch: [48][ 410/ 1236] Overall Loss 0.248017 Objective Loss 0.248017 LR 0.001000 Time 0.022968 +2023-10-02 20:51:53,001 - Epoch: [48][ 420/ 1236] Overall Loss 0.247832 Objective Loss 0.247832 LR 0.001000 Time 0.022943 +2023-10-02 20:51:53,217 - Epoch: [48][ 430/ 1236] Overall Loss 0.248082 Objective Loss 0.248082 LR 0.001000 Time 0.022909 +2023-10-02 20:51:53,436 - Epoch: [48][ 440/ 1236] Overall Loss 0.248511 Objective Loss 0.248511 LR 0.001000 Time 0.022887 +2023-10-02 20:51:53,652 - Epoch: [48][ 450/ 1236] Overall Loss 0.248758 Objective Loss 0.248758 LR 0.001000 Time 0.022857 +2023-10-02 20:51:53,868 - Epoch: [48][ 460/ 1236] Overall Loss 0.248527 Objective Loss 0.248527 LR 0.001000 Time 0.022829 +2023-10-02 20:51:54,076 - Epoch: [48][ 470/ 1236] Overall Loss 0.247920 Objective Loss 0.247920 LR 0.001000 Time 0.022786 +2023-10-02 20:51:54,286 - Epoch: [48][ 480/ 1236] Overall Loss 0.247768 Objective Loss 0.247768 LR 0.001000 Time 0.022748 +2023-10-02 20:51:54,495 - Epoch: [48][ 490/ 1236] Overall Loss 0.248022 Objective Loss 0.248022 LR 0.001000 Time 0.022709 +2023-10-02 20:51:54,705 - Epoch: [48][ 500/ 1236] Overall Loss 0.247964 Objective Loss 0.247964 LR 0.001000 Time 0.022674 +2023-10-02 20:51:54,914 - Epoch: [48][ 510/ 1236] Overall Loss 0.248302 Objective Loss 0.248302 LR 0.001000 Time 0.022638 +2023-10-02 20:51:55,123 - Epoch: [48][ 520/ 1236] Overall Loss 0.248724 Objective Loss 0.248724 LR 0.001000 Time 0.022605 +2023-10-02 20:51:55,332 - Epoch: [48][ 530/ 1236] Overall Loss 0.248857 Objective Loss 0.248857 LR 0.001000 Time 0.022572 +2023-10-02 20:51:55,542 - Epoch: [48][ 540/ 1236] Overall Loss 0.248805 Objective Loss 0.248805 LR 0.001000 Time 0.022541 +2023-10-02 20:51:55,751 - Epoch: [48][ 550/ 1236] Overall Loss 0.249157 Objective Loss 0.249157 LR 0.001000 Time 0.022511 +2023-10-02 20:51:55,960 - Epoch: [48][ 560/ 1236] Overall Loss 0.249556 Objective Loss 0.249556 LR 0.001000 Time 0.022483 +2023-10-02 20:51:56,169 - Epoch: [48][ 570/ 1236] Overall Loss 0.249596 Objective Loss 0.249596 LR 0.001000 Time 0.022454 +2023-10-02 20:51:56,379 - Epoch: [48][ 580/ 1236] Overall Loss 0.249867 Objective Loss 0.249867 LR 0.001000 Time 0.022428 +2023-10-02 20:51:56,588 - Epoch: [48][ 590/ 1236] Overall Loss 0.250258 Objective Loss 0.250258 LR 0.001000 Time 0.022401 +2023-10-02 20:51:56,797 - Epoch: [48][ 600/ 1236] Overall Loss 0.250073 Objective Loss 0.250073 LR 0.001000 Time 0.022376 +2023-10-02 20:51:57,006 - Epoch: [48][ 610/ 1236] Overall Loss 0.250511 Objective Loss 0.250511 LR 0.001000 Time 0.022351 +2023-10-02 20:51:57,216 - Epoch: [48][ 620/ 1236] Overall Loss 0.250556 Objective Loss 0.250556 LR 0.001000 Time 0.022329 +2023-10-02 20:51:57,425 - Epoch: [48][ 630/ 1236] Overall Loss 0.250836 Objective Loss 0.250836 LR 0.001000 Time 0.022305 +2023-10-02 20:51:57,636 - Epoch: [48][ 640/ 1236] Overall Loss 0.251253 Objective Loss 0.251253 LR 0.001000 Time 0.022286 +2023-10-02 20:51:57,845 - Epoch: [48][ 650/ 1236] Overall Loss 0.250851 Objective Loss 0.250851 LR 0.001000 Time 0.022264 +2023-10-02 20:51:58,054 - Epoch: [48][ 660/ 1236] Overall Loss 0.250923 Objective Loss 0.250923 LR 0.001000 Time 0.022244 +2023-10-02 20:51:58,263 - Epoch: [48][ 670/ 1236] Overall Loss 0.250914 Objective Loss 0.250914 LR 0.001000 Time 0.022223 +2023-10-02 20:51:58,473 - Epoch: [48][ 680/ 1236] Overall Loss 0.251092 Objective Loss 0.251092 LR 0.001000 Time 0.022204 +2023-10-02 20:51:58,681 - Epoch: [48][ 690/ 1236] Overall Loss 0.251188 Objective Loss 0.251188 LR 0.001000 Time 0.022184 +2023-10-02 20:51:58,891 - Epoch: [48][ 700/ 1236] Overall Loss 0.251602 Objective Loss 0.251602 LR 0.001000 Time 0.022166 +2023-10-02 20:51:59,100 - Epoch: [48][ 710/ 1236] Overall Loss 0.251695 Objective Loss 0.251695 LR 0.001000 Time 0.022147 +2023-10-02 20:51:59,310 - Epoch: [48][ 720/ 1236] Overall Loss 0.251575 Objective Loss 0.251575 LR 0.001000 Time 0.022131 +2023-10-02 20:51:59,518 - Epoch: [48][ 730/ 1236] Overall Loss 0.251342 Objective Loss 0.251342 LR 0.001000 Time 0.022113 +2023-10-02 20:51:59,728 - Epoch: [48][ 740/ 1236] Overall Loss 0.251350 Objective Loss 0.251350 LR 0.001000 Time 0.022097 +2023-10-02 20:51:59,937 - Epoch: [48][ 750/ 1236] Overall Loss 0.251354 Objective Loss 0.251354 LR 0.001000 Time 0.022081 +2023-10-02 20:52:00,147 - Epoch: [48][ 760/ 1236] Overall Loss 0.251306 Objective Loss 0.251306 LR 0.001000 Time 0.022066 +2023-10-02 20:52:00,356 - Epoch: [48][ 770/ 1236] Overall Loss 0.251228 Objective Loss 0.251228 LR 0.001000 Time 0.022050 +2023-10-02 20:52:00,567 - Epoch: [48][ 780/ 1236] Overall Loss 0.251127 Objective Loss 0.251127 LR 0.001000 Time 0.022037 +2023-10-02 20:52:00,775 - Epoch: [48][ 790/ 1236] Overall Loss 0.251295 Objective Loss 0.251295 LR 0.001000 Time 0.022022 +2023-10-02 20:52:00,985 - Epoch: [48][ 800/ 1236] Overall Loss 0.251258 Objective Loss 0.251258 LR 0.001000 Time 0.022008 +2023-10-02 20:52:01,194 - Epoch: [48][ 810/ 1236] Overall Loss 0.251103 Objective Loss 0.251103 LR 0.001000 Time 0.021994 +2023-10-02 20:52:01,404 - Epoch: [48][ 820/ 1236] Overall Loss 0.251240 Objective Loss 0.251240 LR 0.001000 Time 0.021981 +2023-10-02 20:52:01,612 - Epoch: [48][ 830/ 1236] Overall Loss 0.251449 Objective Loss 0.251449 LR 0.001000 Time 0.021968 +2023-10-02 20:52:01,822 - Epoch: [48][ 840/ 1236] Overall Loss 0.251056 Objective Loss 0.251056 LR 0.001000 Time 0.021956 +2023-10-02 20:52:02,031 - Epoch: [48][ 850/ 1236] Overall Loss 0.251121 Objective Loss 0.251121 LR 0.001000 Time 0.021943 +2023-10-02 20:52:02,241 - Epoch: [48][ 860/ 1236] Overall Loss 0.251033 Objective Loss 0.251033 LR 0.001000 Time 0.021931 +2023-10-02 20:52:02,450 - Epoch: [48][ 870/ 1236] Overall Loss 0.251403 Objective Loss 0.251403 LR 0.001000 Time 0.021919 +2023-10-02 20:52:02,662 - Epoch: [48][ 880/ 1236] Overall Loss 0.251395 Objective Loss 0.251395 LR 0.001000 Time 0.021911 +2023-10-02 20:52:02,872 - Epoch: [48][ 890/ 1236] Overall Loss 0.251495 Objective Loss 0.251495 LR 0.001000 Time 0.021900 +2023-10-02 20:52:03,082 - Epoch: [48][ 900/ 1236] Overall Loss 0.251388 Objective Loss 0.251388 LR 0.001000 Time 0.021889 +2023-10-02 20:52:03,290 - Epoch: [48][ 910/ 1236] Overall Loss 0.251424 Objective Loss 0.251424 LR 0.001000 Time 0.021877 +2023-10-02 20:52:03,500 - Epoch: [48][ 920/ 1236] Overall Loss 0.251730 Objective Loss 0.251730 LR 0.001000 Time 0.021867 +2023-10-02 20:52:03,709 - Epoch: [48][ 930/ 1236] Overall Loss 0.251819 Objective Loss 0.251819 LR 0.001000 Time 0.021856 +2023-10-02 20:52:03,919 - Epoch: [48][ 940/ 1236] Overall Loss 0.251719 Objective Loss 0.251719 LR 0.001000 Time 0.021847 +2023-10-02 20:52:04,128 - Epoch: [48][ 950/ 1236] Overall Loss 0.251958 Objective Loss 0.251958 LR 0.001000 Time 0.021836 +2023-10-02 20:52:04,338 - Epoch: [48][ 960/ 1236] Overall Loss 0.251961 Objective Loss 0.251961 LR 0.001000 Time 0.021828 +2023-10-02 20:52:04,547 - Epoch: [48][ 970/ 1236] Overall Loss 0.251988 Objective Loss 0.251988 LR 0.001000 Time 0.021817 +2023-10-02 20:52:04,757 - Epoch: [48][ 980/ 1236] Overall Loss 0.252217 Objective Loss 0.252217 LR 0.001000 Time 0.021809 +2023-10-02 20:52:04,966 - Epoch: [48][ 990/ 1236] Overall Loss 0.252393 Objective Loss 0.252393 LR 0.001000 Time 0.021799 +2023-10-02 20:52:05,176 - Epoch: [48][ 1000/ 1236] Overall Loss 0.252768 Objective Loss 0.252768 LR 0.001000 Time 0.021791 +2023-10-02 20:52:05,385 - Epoch: [48][ 1010/ 1236] Overall Loss 0.253157 Objective Loss 0.253157 LR 0.001000 Time 0.021782 +2023-10-02 20:52:05,595 - Epoch: [48][ 1020/ 1236] Overall Loss 0.253078 Objective Loss 0.253078 LR 0.001000 Time 0.021774 +2023-10-02 20:52:05,803 - Epoch: [48][ 1030/ 1236] Overall Loss 0.253161 Objective Loss 0.253161 LR 0.001000 Time 0.021765 +2023-10-02 20:52:06,013 - Epoch: [48][ 1040/ 1236] Overall Loss 0.253247 Objective Loss 0.253247 LR 0.001000 Time 0.021757 +2023-10-02 20:52:06,222 - Epoch: [48][ 1050/ 1236] Overall Loss 0.253346 Objective Loss 0.253346 LR 0.001000 Time 0.021749 +2023-10-02 20:52:06,433 - Epoch: [48][ 1060/ 1236] Overall Loss 0.253350 Objective Loss 0.253350 LR 0.001000 Time 0.021741 +2023-10-02 20:52:06,641 - Epoch: [48][ 1070/ 1236] Overall Loss 0.253346 Objective Loss 0.253346 LR 0.001000 Time 0.021733 +2023-10-02 20:52:06,851 - Epoch: [48][ 1080/ 1236] Overall Loss 0.253089 Objective Loss 0.253089 LR 0.001000 Time 0.021726 +2023-10-02 20:52:07,060 - Epoch: [48][ 1090/ 1236] Overall Loss 0.253071 Objective Loss 0.253071 LR 0.001000 Time 0.021718 +2023-10-02 20:52:07,270 - Epoch: [48][ 1100/ 1236] Overall Loss 0.253087 Objective Loss 0.253087 LR 0.001000 Time 0.021711 +2023-10-02 20:52:07,479 - Epoch: [48][ 1110/ 1236] Overall Loss 0.253323 Objective Loss 0.253323 LR 0.001000 Time 0.021703 +2023-10-02 20:52:07,689 - Epoch: [48][ 1120/ 1236] Overall Loss 0.253339 Objective Loss 0.253339 LR 0.001000 Time 0.021697 +2023-10-02 20:52:07,898 - Epoch: [48][ 1130/ 1236] Overall Loss 0.253599 Objective Loss 0.253599 LR 0.001000 Time 0.021689 +2023-10-02 20:52:08,109 - Epoch: [48][ 1140/ 1236] Overall Loss 0.253775 Objective Loss 0.253775 LR 0.001000 Time 0.021684 +2023-10-02 20:52:08,319 - Epoch: [48][ 1150/ 1236] Overall Loss 0.253658 Objective Loss 0.253658 LR 0.001000 Time 0.021677 +2023-10-02 20:52:08,529 - Epoch: [48][ 1160/ 1236] Overall Loss 0.253874 Objective Loss 0.253874 LR 0.001000 Time 0.021672 +2023-10-02 20:52:08,740 - Epoch: [48][ 1170/ 1236] Overall Loss 0.253889 Objective Loss 0.253889 LR 0.001000 Time 0.021666 +2023-10-02 20:52:08,951 - Epoch: [48][ 1180/ 1236] Overall Loss 0.253891 Objective Loss 0.253891 LR 0.001000 Time 0.021661 +2023-10-02 20:52:09,161 - Epoch: [48][ 1190/ 1236] Overall Loss 0.254234 Objective Loss 0.254234 LR 0.001000 Time 0.021655 +2023-10-02 20:52:09,372 - Epoch: [48][ 1200/ 1236] Overall Loss 0.254168 Objective Loss 0.254168 LR 0.001000 Time 0.021650 +2023-10-02 20:52:09,583 - Epoch: [48][ 1210/ 1236] Overall Loss 0.254144 Objective Loss 0.254144 LR 0.001000 Time 0.021645 +2023-10-02 20:52:09,795 - Epoch: [48][ 1220/ 1236] Overall Loss 0.254232 Objective Loss 0.254232 LR 0.001000 Time 0.021641 +2023-10-02 20:52:10,059 - Epoch: [48][ 1230/ 1236] Overall Loss 0.254625 Objective Loss 0.254625 LR 0.001000 Time 0.021680 +2023-10-02 20:52:10,182 - Epoch: [48][ 1236/ 1236] Overall Loss 0.254805 Objective Loss 0.254805 Top1 86.761711 Top5 97.556008 LR 0.001000 Time 0.021674 +2023-10-02 20:52:10,315 - --- validate (epoch=48)----------- +2023-10-02 20:52:10,315 - 29943 samples (256 per mini-batch) +2023-10-02 20:52:10,797 - Epoch: [48][ 10/ 117] Loss 0.304934 Top1 83.984375 Top5 98.515625 +2023-10-02 20:52:10,949 - Epoch: [48][ 20/ 117] Loss 0.322889 Top1 83.339844 Top5 98.281250 +2023-10-02 20:52:11,099 - Epoch: [48][ 30/ 117] Loss 0.312212 Top1 83.789062 Top5 98.307292 +2023-10-02 20:52:11,248 - Epoch: [48][ 40/ 117] Loss 0.314443 Top1 83.955078 Top5 98.261719 +2023-10-02 20:52:11,398 - Epoch: [48][ 50/ 117] Loss 0.311366 Top1 83.976562 Top5 98.203125 +2023-10-02 20:52:11,548 - Epoch: [48][ 60/ 117] Loss 0.310933 Top1 84.023438 Top5 98.170573 +2023-10-02 20:52:11,697 - Epoch: [48][ 70/ 117] Loss 0.311194 Top1 84.017857 Top5 98.164062 +2023-10-02 20:52:11,848 - Epoch: [48][ 80/ 117] Loss 0.312970 Top1 83.867188 Top5 98.178711 +2023-10-02 20:52:11,997 - Epoch: [48][ 90/ 117] Loss 0.316515 Top1 83.789062 Top5 98.177083 +2023-10-02 20:52:12,148 - Epoch: [48][ 100/ 117] Loss 0.319022 Top1 83.824219 Top5 98.132812 +2023-10-02 20:52:12,305 - Epoch: [48][ 110/ 117] Loss 0.315278 Top1 83.884943 Top5 98.135653 +2023-10-02 20:52:12,393 - Epoch: [48][ 117/ 117] Loss 0.317440 Top1 83.869352 Top5 98.126440 +2023-10-02 20:52:12,524 - ==> Top1: 83.869 Top5: 98.126 Loss: 0.317 + +2023-10-02 20:52:12,524 - ==> Confusion: +[[ 898 0 2 0 9 4 0 0 6 92 3 0 0 3 6 2 3 1 0 0 21] + [ 0 1052 3 0 10 16 1 21 1 1 5 0 0 0 1 2 2 0 10 0 6] + [ 5 0 977 9 5 0 20 7 1 3 5 1 5 2 0 3 0 1 4 1 7] + [ 1 3 18 983 1 3 0 3 4 1 13 0 2 3 25 1 1 3 7 1 16] + [ 21 4 1 1 981 3 0 0 0 11 1 0 0 6 7 4 5 0 1 0 4] + [ 2 57 3 0 9 953 0 30 0 8 5 2 4 13 6 1 1 0 4 4 14] + [ 0 3 32 0 0 1 1132 2 0 0 7 0 1 0 1 3 0 1 1 3 4] + [ 1 16 19 0 4 24 4 1049 1 6 8 5 1 7 2 0 2 0 46 12 11] + [ 17 2 0 0 3 1 0 0 945 65 14 1 0 18 15 0 0 2 3 1 2] + [ 65 0 0 3 6 0 0 0 15 993 1 0 0 17 7 0 2 0 0 2 8] + [ 0 3 6 8 2 0 4 4 10 1 978 2 0 12 5 1 0 2 6 1 8] + [ 1 1 1 0 4 17 0 7 0 4 0 901 39 12 0 7 4 17 0 13 7] + [ 0 0 6 2 0 4 2 2 0 3 1 33 975 3 2 8 0 14 0 4 9] + [ 3 0 1 0 5 4 2 1 9 19 8 7 0 1044 4 1 1 1 0 1 8] + [ 10 0 2 21 8 0 0 0 19 6 5 0 3 4 1005 0 0 1 4 0 13] + [ 0 0 1 1 6 3 0 0 0 1 0 5 9 0 0 1073 17 7 2 4 5] + [ 1 19 3 0 17 4 2 1 1 0 0 5 1 3 3 6 1075 0 1 5 14] + [ 0 0 1 3 0 0 1 0 2 3 2 6 31 3 3 7 2 962 2 1 9] + [ 3 4 9 21 0 1 0 11 6 0 8 0 1 1 12 0 0 0 978 1 12] + [ 0 4 4 1 2 2 18 9 0 2 2 11 2 4 0 1 7 0 0 1076 7] + [ 140 148 178 99 167 142 46 80 88 102 183 116 396 271 129 55 119 57 138 168 5083]] + +2023-10-02 20:52:12,526 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:52:12,526 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:52:12,532 - + +2023-10-02 20:52:12,532 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:52:13,659 - Epoch: [49][ 10/ 1236] Overall Loss 0.228186 Objective Loss 0.228186 LR 0.001000 Time 0.112676 +2023-10-02 20:52:13,871 - Epoch: [49][ 20/ 1236] Overall Loss 0.239329 Objective Loss 0.239329 LR 0.001000 Time 0.066890 +2023-10-02 20:52:14,079 - Epoch: [49][ 30/ 1236] Overall Loss 0.231923 Objective Loss 0.231923 LR 0.001000 Time 0.051501 +2023-10-02 20:52:14,290 - Epoch: [49][ 40/ 1236] Overall Loss 0.235533 Objective Loss 0.235533 LR 0.001000 Time 0.043897 +2023-10-02 20:52:14,497 - Epoch: [49][ 50/ 1236] Overall Loss 0.241700 Objective Loss 0.241700 LR 0.001000 Time 0.039253 +2023-10-02 20:52:14,708 - Epoch: [49][ 60/ 1236] Overall Loss 0.244050 Objective Loss 0.244050 LR 0.001000 Time 0.036219 +2023-10-02 20:52:14,915 - Epoch: [49][ 70/ 1236] Overall Loss 0.242135 Objective Loss 0.242135 LR 0.001000 Time 0.033999 +2023-10-02 20:52:15,126 - Epoch: [49][ 80/ 1236] Overall Loss 0.241101 Objective Loss 0.241101 LR 0.001000 Time 0.032381 +2023-10-02 20:52:15,334 - Epoch: [49][ 90/ 1236] Overall Loss 0.242402 Objective Loss 0.242402 LR 0.001000 Time 0.031091 +2023-10-02 20:52:15,547 - Epoch: [49][ 100/ 1236] Overall Loss 0.244124 Objective Loss 0.244124 LR 0.001000 Time 0.030107 +2023-10-02 20:52:15,756 - Epoch: [49][ 110/ 1236] Overall Loss 0.242961 Objective Loss 0.242961 LR 0.001000 Time 0.029264 +2023-10-02 20:52:15,967 - Epoch: [49][ 120/ 1236] Overall Loss 0.242979 Objective Loss 0.242979 LR 0.001000 Time 0.028581 +2023-10-02 20:52:16,177 - Epoch: [49][ 130/ 1236] Overall Loss 0.242130 Objective Loss 0.242130 LR 0.001000 Time 0.027995 +2023-10-02 20:52:16,389 - Epoch: [49][ 140/ 1236] Overall Loss 0.243187 Objective Loss 0.243187 LR 0.001000 Time 0.027508 +2023-10-02 20:52:16,598 - Epoch: [49][ 150/ 1236] Overall Loss 0.245791 Objective Loss 0.245791 LR 0.001000 Time 0.027070 +2023-10-02 20:52:16,818 - Epoch: [49][ 160/ 1236] Overall Loss 0.247070 Objective Loss 0.247070 LR 0.001000 Time 0.026747 +2023-10-02 20:52:17,027 - Epoch: [49][ 170/ 1236] Overall Loss 0.247016 Objective Loss 0.247016 LR 0.001000 Time 0.026403 +2023-10-02 20:52:17,242 - Epoch: [49][ 180/ 1236] Overall Loss 0.248452 Objective Loss 0.248452 LR 0.001000 Time 0.026128 +2023-10-02 20:52:17,450 - Epoch: [49][ 190/ 1236] Overall Loss 0.249475 Objective Loss 0.249475 LR 0.001000 Time 0.025846 +2023-10-02 20:52:17,664 - Epoch: [49][ 200/ 1236] Overall Loss 0.250575 Objective Loss 0.250575 LR 0.001000 Time 0.025624 +2023-10-02 20:52:17,873 - Epoch: [49][ 210/ 1236] Overall Loss 0.250937 Objective Loss 0.250937 LR 0.001000 Time 0.025397 +2023-10-02 20:52:18,088 - Epoch: [49][ 220/ 1236] Overall Loss 0.251291 Objective Loss 0.251291 LR 0.001000 Time 0.025215 +2023-10-02 20:52:18,299 - Epoch: [49][ 230/ 1236] Overall Loss 0.250758 Objective Loss 0.250758 LR 0.001000 Time 0.025037 +2023-10-02 20:52:18,516 - Epoch: [49][ 240/ 1236] Overall Loss 0.251425 Objective Loss 0.251425 LR 0.001000 Time 0.024894 +2023-10-02 20:52:18,725 - Epoch: [49][ 250/ 1236] Overall Loss 0.253206 Objective Loss 0.253206 LR 0.001000 Time 0.024733 +2023-10-02 20:52:18,939 - Epoch: [49][ 260/ 1236] Overall Loss 0.253040 Objective Loss 0.253040 LR 0.001000 Time 0.024605 +2023-10-02 20:52:19,150 - Epoch: [49][ 270/ 1236] Overall Loss 0.252276 Objective Loss 0.252276 LR 0.001000 Time 0.024474 +2023-10-02 20:52:19,365 - Epoch: [49][ 280/ 1236] Overall Loss 0.251049 Objective Loss 0.251049 LR 0.001000 Time 0.024366 +2023-10-02 20:52:19,576 - Epoch: [49][ 290/ 1236] Overall Loss 0.250055 Objective Loss 0.250055 LR 0.001000 Time 0.024252 +2023-10-02 20:52:19,790 - Epoch: [49][ 300/ 1236] Overall Loss 0.248997 Objective Loss 0.248997 LR 0.001000 Time 0.024156 +2023-10-02 20:52:20,001 - Epoch: [49][ 310/ 1236] Overall Loss 0.249558 Objective Loss 0.249558 LR 0.001000 Time 0.024056 +2023-10-02 20:52:20,218 - Epoch: [49][ 320/ 1236] Overall Loss 0.249815 Objective Loss 0.249815 LR 0.001000 Time 0.023979 +2023-10-02 20:52:20,429 - Epoch: [49][ 330/ 1236] Overall Loss 0.250348 Objective Loss 0.250348 LR 0.001000 Time 0.023891 +2023-10-02 20:52:20,645 - Epoch: [49][ 340/ 1236] Overall Loss 0.249988 Objective Loss 0.249988 LR 0.001000 Time 0.023824 +2023-10-02 20:52:20,857 - Epoch: [49][ 350/ 1236] Overall Loss 0.249357 Objective Loss 0.249357 LR 0.001000 Time 0.023744 +2023-10-02 20:52:21,074 - Epoch: [49][ 360/ 1236] Overall Loss 0.248789 Objective Loss 0.248789 LR 0.001000 Time 0.023687 +2023-10-02 20:52:21,284 - Epoch: [49][ 370/ 1236] Overall Loss 0.250129 Objective Loss 0.250129 LR 0.001000 Time 0.023613 +2023-10-02 20:52:21,498 - Epoch: [49][ 380/ 1236] Overall Loss 0.249557 Objective Loss 0.249557 LR 0.001000 Time 0.023553 +2023-10-02 20:52:21,708 - Epoch: [49][ 390/ 1236] Overall Loss 0.249727 Objective Loss 0.249727 LR 0.001000 Time 0.023487 +2023-10-02 20:52:21,922 - Epoch: [49][ 400/ 1236] Overall Loss 0.249341 Objective Loss 0.249341 LR 0.001000 Time 0.023432 +2023-10-02 20:52:22,131 - Epoch: [49][ 410/ 1236] Overall Loss 0.249431 Objective Loss 0.249431 LR 0.001000 Time 0.023371 +2023-10-02 20:52:22,346 - Epoch: [49][ 420/ 1236] Overall Loss 0.249797 Objective Loss 0.249797 LR 0.001000 Time 0.023325 +2023-10-02 20:52:22,557 - Epoch: [49][ 430/ 1236] Overall Loss 0.250723 Objective Loss 0.250723 LR 0.001000 Time 0.023271 +2023-10-02 20:52:22,771 - Epoch: [49][ 440/ 1236] Overall Loss 0.250969 Objective Loss 0.250969 LR 0.001000 Time 0.023228 +2023-10-02 20:52:22,981 - Epoch: [49][ 450/ 1236] Overall Loss 0.251296 Objective Loss 0.251296 LR 0.001000 Time 0.023177 +2023-10-02 20:52:23,194 - Epoch: [49][ 460/ 1236] Overall Loss 0.251090 Objective Loss 0.251090 LR 0.001000 Time 0.023137 +2023-10-02 20:52:23,404 - Epoch: [49][ 470/ 1236] Overall Loss 0.251094 Objective Loss 0.251094 LR 0.001000 Time 0.023090 +2023-10-02 20:52:23,617 - Epoch: [49][ 480/ 1236] Overall Loss 0.251008 Objective Loss 0.251008 LR 0.001000 Time 0.023053 +2023-10-02 20:52:23,827 - Epoch: [49][ 490/ 1236] Overall Loss 0.251346 Objective Loss 0.251346 LR 0.001000 Time 0.023010 +2023-10-02 20:52:24,040 - Epoch: [49][ 500/ 1236] Overall Loss 0.251938 Objective Loss 0.251938 LR 0.001000 Time 0.022975 +2023-10-02 20:52:24,251 - Epoch: [49][ 510/ 1236] Overall Loss 0.252229 Objective Loss 0.252229 LR 0.001000 Time 0.022936 +2023-10-02 20:52:24,464 - Epoch: [49][ 520/ 1236] Overall Loss 0.251992 Objective Loss 0.251992 LR 0.001000 Time 0.022904 +2023-10-02 20:52:24,674 - Epoch: [49][ 530/ 1236] Overall Loss 0.252709 Objective Loss 0.252709 LR 0.001000 Time 0.022868 +2023-10-02 20:52:24,887 - Epoch: [49][ 540/ 1236] Overall Loss 0.252784 Objective Loss 0.252784 LR 0.001000 Time 0.022839 +2023-10-02 20:52:25,098 - Epoch: [49][ 550/ 1236] Overall Loss 0.252808 Objective Loss 0.252808 LR 0.001000 Time 0.022805 +2023-10-02 20:52:25,312 - Epoch: [49][ 560/ 1236] Overall Loss 0.253131 Objective Loss 0.253131 LR 0.001000 Time 0.022781 +2023-10-02 20:52:25,523 - Epoch: [49][ 570/ 1236] Overall Loss 0.253474 Objective Loss 0.253474 LR 0.001000 Time 0.022750 +2023-10-02 20:52:25,736 - Epoch: [49][ 580/ 1236] Overall Loss 0.253560 Objective Loss 0.253560 LR 0.001000 Time 0.022724 +2023-10-02 20:52:25,945 - Epoch: [49][ 590/ 1236] Overall Loss 0.253814 Objective Loss 0.253814 LR 0.001000 Time 0.022691 +2023-10-02 20:52:26,157 - Epoch: [49][ 600/ 1236] Overall Loss 0.253957 Objective Loss 0.253957 LR 0.001000 Time 0.022665 +2023-10-02 20:52:26,366 - Epoch: [49][ 610/ 1236] Overall Loss 0.253587 Objective Loss 0.253587 LR 0.001000 Time 0.022636 +2023-10-02 20:52:26,579 - Epoch: [49][ 620/ 1236] Overall Loss 0.254164 Objective Loss 0.254164 LR 0.001000 Time 0.022613 +2023-10-02 20:52:26,787 - Epoch: [49][ 630/ 1236] Overall Loss 0.254222 Objective Loss 0.254222 LR 0.001000 Time 0.022584 +2023-10-02 20:52:26,998 - Epoch: [49][ 640/ 1236] Overall Loss 0.254202 Objective Loss 0.254202 LR 0.001000 Time 0.022561 +2023-10-02 20:52:27,207 - Epoch: [49][ 650/ 1236] Overall Loss 0.254726 Objective Loss 0.254726 LR 0.001000 Time 0.022533 +2023-10-02 20:52:27,419 - Epoch: [49][ 660/ 1236] Overall Loss 0.254170 Objective Loss 0.254170 LR 0.001000 Time 0.022512 +2023-10-02 20:52:27,633 - Epoch: [49][ 670/ 1236] Overall Loss 0.254488 Objective Loss 0.254488 LR 0.001000 Time 0.022492 +2023-10-02 20:52:27,848 - Epoch: [49][ 680/ 1236] Overall Loss 0.254538 Objective Loss 0.254538 LR 0.001000 Time 0.022477 +2023-10-02 20:52:28,068 - Epoch: [49][ 690/ 1236] Overall Loss 0.254229 Objective Loss 0.254229 LR 0.001000 Time 0.022469 +2023-10-02 20:52:28,284 - Epoch: [49][ 700/ 1236] Overall Loss 0.253753 Objective Loss 0.253753 LR 0.001000 Time 0.022457 +2023-10-02 20:52:28,504 - Epoch: [49][ 710/ 1236] Overall Loss 0.253467 Objective Loss 0.253467 LR 0.001000 Time 0.022449 +2023-10-02 20:52:28,719 - Epoch: [49][ 720/ 1236] Overall Loss 0.253361 Objective Loss 0.253361 LR 0.001000 Time 0.022436 +2023-10-02 20:52:28,939 - Epoch: [49][ 730/ 1236] Overall Loss 0.253550 Objective Loss 0.253550 LR 0.001000 Time 0.022429 +2023-10-02 20:52:29,154 - Epoch: [49][ 740/ 1236] Overall Loss 0.253803 Objective Loss 0.253803 LR 0.001000 Time 0.022417 +2023-10-02 20:52:29,371 - Epoch: [49][ 750/ 1236] Overall Loss 0.253631 Objective Loss 0.253631 LR 0.001000 Time 0.022406 +2023-10-02 20:52:29,586 - Epoch: [49][ 760/ 1236] Overall Loss 0.253648 Objective Loss 0.253648 LR 0.001000 Time 0.022395 +2023-10-02 20:52:29,812 - Epoch: [49][ 770/ 1236] Overall Loss 0.253758 Objective Loss 0.253758 LR 0.001000 Time 0.022396 +2023-10-02 20:52:30,023 - Epoch: [49][ 780/ 1236] Overall Loss 0.253840 Objective Loss 0.253840 LR 0.001000 Time 0.022379 +2023-10-02 20:52:30,236 - Epoch: [49][ 790/ 1236] Overall Loss 0.253790 Objective Loss 0.253790 LR 0.001000 Time 0.022366 +2023-10-02 20:52:30,447 - Epoch: [49][ 800/ 1236] Overall Loss 0.253765 Objective Loss 0.253765 LR 0.001000 Time 0.022350 +2023-10-02 20:52:30,660 - Epoch: [49][ 810/ 1236] Overall Loss 0.253828 Objective Loss 0.253828 LR 0.001000 Time 0.022336 +2023-10-02 20:52:30,869 - Epoch: [49][ 820/ 1236] Overall Loss 0.254256 Objective Loss 0.254256 LR 0.001000 Time 0.022318 +2023-10-02 20:52:31,079 - Epoch: [49][ 830/ 1236] Overall Loss 0.254192 Objective Loss 0.254192 LR 0.001000 Time 0.022301 +2023-10-02 20:52:31,291 - Epoch: [49][ 840/ 1236] Overall Loss 0.253912 Objective Loss 0.253912 LR 0.001000 Time 0.022288 +2023-10-02 20:52:31,506 - Epoch: [49][ 850/ 1236] Overall Loss 0.253832 Objective Loss 0.253832 LR 0.001000 Time 0.022278 +2023-10-02 20:52:31,726 - Epoch: [49][ 860/ 1236] Overall Loss 0.253971 Objective Loss 0.253971 LR 0.001000 Time 0.022275 +2023-10-02 20:52:31,949 - Epoch: [49][ 870/ 1236] Overall Loss 0.253783 Objective Loss 0.253783 LR 0.001000 Time 0.022274 +2023-10-02 20:52:32,167 - Epoch: [49][ 880/ 1236] Overall Loss 0.253809 Objective Loss 0.253809 LR 0.001000 Time 0.022269 +2023-10-02 20:52:32,387 - Epoch: [49][ 890/ 1236] Overall Loss 0.254027 Objective Loss 0.254027 LR 0.001000 Time 0.022265 +2023-10-02 20:52:32,603 - Epoch: [49][ 900/ 1236] Overall Loss 0.254181 Objective Loss 0.254181 LR 0.001000 Time 0.022256 +2023-10-02 20:52:32,822 - Epoch: [49][ 910/ 1236] Overall Loss 0.254218 Objective Loss 0.254218 LR 0.001000 Time 0.022252 +2023-10-02 20:52:33,036 - Epoch: [49][ 920/ 1236] Overall Loss 0.254349 Objective Loss 0.254349 LR 0.001000 Time 0.022241 +2023-10-02 20:52:33,254 - Epoch: [49][ 930/ 1236] Overall Loss 0.254478 Objective Loss 0.254478 LR 0.001000 Time 0.022236 +2023-10-02 20:52:33,469 - Epoch: [49][ 940/ 1236] Overall Loss 0.254897 Objective Loss 0.254897 LR 0.001000 Time 0.022228 +2023-10-02 20:52:33,687 - Epoch: [49][ 950/ 1236] Overall Loss 0.255063 Objective Loss 0.255063 LR 0.001000 Time 0.022223 +2023-10-02 20:52:33,902 - Epoch: [49][ 960/ 1236] Overall Loss 0.255168 Objective Loss 0.255168 LR 0.001000 Time 0.022215 +2023-10-02 20:52:34,120 - Epoch: [49][ 970/ 1236] Overall Loss 0.255079 Objective Loss 0.255079 LR 0.001000 Time 0.022211 +2023-10-02 20:52:34,335 - Epoch: [49][ 980/ 1236] Overall Loss 0.254859 Objective Loss 0.254859 LR 0.001000 Time 0.022203 +2023-10-02 20:52:34,554 - Epoch: [49][ 990/ 1236] Overall Loss 0.254713 Objective Loss 0.254713 LR 0.001000 Time 0.022199 +2023-10-02 20:52:34,768 - Epoch: [49][ 1000/ 1236] Overall Loss 0.254603 Objective Loss 0.254603 LR 0.001000 Time 0.022191 +2023-10-02 20:52:34,986 - Epoch: [49][ 1010/ 1236] Overall Loss 0.254514 Objective Loss 0.254514 LR 0.001000 Time 0.022187 +2023-10-02 20:52:35,196 - Epoch: [49][ 1020/ 1236] Overall Loss 0.254664 Objective Loss 0.254664 LR 0.001000 Time 0.022175 +2023-10-02 20:52:35,404 - Epoch: [49][ 1030/ 1236] Overall Loss 0.254687 Objective Loss 0.254687 LR 0.001000 Time 0.022161 +2023-10-02 20:52:35,612 - Epoch: [49][ 1040/ 1236] Overall Loss 0.254874 Objective Loss 0.254874 LR 0.001000 Time 0.022148 +2023-10-02 20:52:35,820 - Epoch: [49][ 1050/ 1236] Overall Loss 0.255035 Objective Loss 0.255035 LR 0.001000 Time 0.022134 +2023-10-02 20:52:36,027 - Epoch: [49][ 1060/ 1236] Overall Loss 0.255026 Objective Loss 0.255026 LR 0.001000 Time 0.022121 +2023-10-02 20:52:36,235 - Epoch: [49][ 1070/ 1236] Overall Loss 0.255222 Objective Loss 0.255222 LR 0.001000 Time 0.022108 +2023-10-02 20:52:36,442 - Epoch: [49][ 1080/ 1236] Overall Loss 0.255266 Objective Loss 0.255266 LR 0.001000 Time 0.022095 +2023-10-02 20:52:36,651 - Epoch: [49][ 1090/ 1236] Overall Loss 0.255446 Objective Loss 0.255446 LR 0.001000 Time 0.022083 +2023-10-02 20:52:36,858 - Epoch: [49][ 1100/ 1236] Overall Loss 0.255580 Objective Loss 0.255580 LR 0.001000 Time 0.022070 +2023-10-02 20:52:37,067 - Epoch: [49][ 1110/ 1236] Overall Loss 0.255604 Objective Loss 0.255604 LR 0.001000 Time 0.022059 +2023-10-02 20:52:37,273 - Epoch: [49][ 1120/ 1236] Overall Loss 0.255873 Objective Loss 0.255873 LR 0.001000 Time 0.022047 +2023-10-02 20:52:37,481 - Epoch: [49][ 1130/ 1236] Overall Loss 0.255979 Objective Loss 0.255979 LR 0.001000 Time 0.022035 +2023-10-02 20:52:37,689 - Epoch: [49][ 1140/ 1236] Overall Loss 0.255762 Objective Loss 0.255762 LR 0.001000 Time 0.022024 +2023-10-02 20:52:37,897 - Epoch: [49][ 1150/ 1236] Overall Loss 0.256102 Objective Loss 0.256102 LR 0.001000 Time 0.022013 +2023-10-02 20:52:38,104 - Epoch: [49][ 1160/ 1236] Overall Loss 0.256211 Objective Loss 0.256211 LR 0.001000 Time 0.022001 +2023-10-02 20:52:38,312 - Epoch: [49][ 1170/ 1236] Overall Loss 0.256172 Objective Loss 0.256172 LR 0.001000 Time 0.021990 +2023-10-02 20:52:38,519 - Epoch: [49][ 1180/ 1236] Overall Loss 0.256283 Objective Loss 0.256283 LR 0.001000 Time 0.021979 +2023-10-02 20:52:38,728 - Epoch: [49][ 1190/ 1236] Overall Loss 0.256197 Objective Loss 0.256197 LR 0.001000 Time 0.021970 +2023-10-02 20:52:38,935 - Epoch: [49][ 1200/ 1236] Overall Loss 0.256086 Objective Loss 0.256086 LR 0.001000 Time 0.021959 +2023-10-02 20:52:39,143 - Epoch: [49][ 1210/ 1236] Overall Loss 0.256142 Objective Loss 0.256142 LR 0.001000 Time 0.021949 +2023-10-02 20:52:39,358 - Epoch: [49][ 1220/ 1236] Overall Loss 0.256103 Objective Loss 0.256103 LR 0.001000 Time 0.021945 +2023-10-02 20:52:39,624 - Epoch: [49][ 1230/ 1236] Overall Loss 0.256139 Objective Loss 0.256139 LR 0.001000 Time 0.021983 +2023-10-02 20:52:39,746 - Epoch: [49][ 1236/ 1236] Overall Loss 0.256190 Objective Loss 0.256190 Top1 85.132383 Top5 97.148676 LR 0.001000 Time 0.021975 +2023-10-02 20:52:39,881 - --- validate (epoch=49)----------- +2023-10-02 20:52:39,882 - 29943 samples (256 per mini-batch) +2023-10-02 20:52:40,359 - Epoch: [49][ 10/ 117] Loss 0.323624 Top1 83.437500 Top5 97.929688 +2023-10-02 20:52:40,512 - Epoch: [49][ 20/ 117] Loss 0.317291 Top1 83.808594 Top5 98.046875 +2023-10-02 20:52:40,664 - Epoch: [49][ 30/ 117] Loss 0.321697 Top1 83.932292 Top5 98.033854 +2023-10-02 20:52:40,814 - Epoch: [49][ 40/ 117] Loss 0.315581 Top1 83.925781 Top5 98.193359 +2023-10-02 20:52:40,963 - Epoch: [49][ 50/ 117] Loss 0.316377 Top1 83.937500 Top5 98.203125 +2023-10-02 20:52:41,113 - Epoch: [49][ 60/ 117] Loss 0.322239 Top1 83.736979 Top5 98.164062 +2023-10-02 20:52:41,262 - Epoch: [49][ 70/ 117] Loss 0.325011 Top1 83.822545 Top5 98.169643 +2023-10-02 20:52:41,413 - Epoch: [49][ 80/ 117] Loss 0.324332 Top1 83.837891 Top5 98.193359 +2023-10-02 20:52:41,564 - Epoch: [49][ 90/ 117] Loss 0.324815 Top1 83.945312 Top5 98.242188 +2023-10-02 20:52:41,716 - Epoch: [49][ 100/ 117] Loss 0.323185 Top1 83.902344 Top5 98.214844 +2023-10-02 20:52:41,874 - Epoch: [49][ 110/ 117] Loss 0.323135 Top1 83.849432 Top5 98.164062 +2023-10-02 20:52:41,963 - Epoch: [49][ 117/ 117] Loss 0.321160 Top1 83.842634 Top5 98.169856 +2023-10-02 20:52:42,080 - ==> Top1: 83.843 Top5: 98.170 Loss: 0.321 + +2023-10-02 20:52:42,081 - ==> Confusion: +[[ 948 0 7 0 10 3 0 1 0 52 2 0 0 2 0 3 6 3 0 0 13] + [ 0 1037 1 2 11 32 0 24 2 1 2 0 1 0 2 4 1 0 8 0 3] + [ 9 0 964 18 3 0 22 8 0 0 2 0 5 1 1 3 0 0 10 4 6] + [ 1 1 16 992 2 1 3 3 4 1 2 0 3 4 27 0 2 5 8 0 14] + [ 30 3 0 3 967 7 0 0 1 8 1 2 1 2 8 6 3 0 2 2 4] + [ 4 38 3 2 4 996 0 20 0 5 2 4 2 8 3 1 1 0 7 4 12] + [ 0 2 32 1 0 0 1131 2 0 0 2 3 0 0 0 2 0 1 2 6 7] + [ 3 11 21 2 1 17 5 1077 2 1 3 4 4 4 3 0 1 0 41 10 8] + [ 28 2 0 0 1 2 1 0 944 58 10 2 2 11 15 3 2 1 5 0 2] + [ 107 0 0 0 7 1 0 0 10 970 4 1 0 9 3 1 1 0 0 1 4] + [ 5 0 18 16 0 2 2 5 13 4 954 3 1 9 2 2 1 1 6 3 6] + [ 0 1 3 0 2 19 0 5 0 0 1 939 22 8 0 5 1 15 0 10 4] + [ 3 0 3 9 1 1 4 0 2 5 1 42 927 3 3 12 4 28 0 6 14] + [ 3 0 1 1 6 13 0 2 11 19 4 4 0 1033 6 1 1 1 0 3 10] + [ 18 0 4 22 5 0 0 0 26 2 2 0 3 1 1001 0 2 2 7 0 6] + [ 0 2 3 0 4 0 2 0 0 0 0 5 4 0 0 1073 18 13 1 6 3] + [ 2 7 0 0 6 11 1 2 1 1 0 5 0 2 3 8 1092 0 0 8 12] + [ 0 0 2 4 0 1 1 1 0 1 2 7 20 4 5 7 1 976 1 1 4] + [ 1 7 4 15 1 0 0 21 7 1 1 1 0 0 8 1 0 0 989 1 10] + [ 0 4 5 2 0 7 7 13 0 1 0 11 1 3 2 3 5 1 2 1082 3] + [ 185 132 178 103 127 195 57 106 95 120 133 110 288 240 155 65 124 56 166 257 5013]] + +2023-10-02 20:52:42,082 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:52:42,082 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:52:42,088 - + +2023-10-02 20:52:42,088 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:52:43,115 - Epoch: [50][ 10/ 1236] Overall Loss 0.219099 Objective Loss 0.219099 LR 0.001000 Time 0.102586 +2023-10-02 20:52:43,324 - Epoch: [50][ 20/ 1236] Overall Loss 0.227311 Objective Loss 0.227311 LR 0.001000 Time 0.061741 +2023-10-02 20:52:43,532 - Epoch: [50][ 30/ 1236] Overall Loss 0.229514 Objective Loss 0.229514 LR 0.001000 Time 0.048093 +2023-10-02 20:52:43,743 - Epoch: [50][ 40/ 1236] Overall Loss 0.237156 Objective Loss 0.237156 LR 0.001000 Time 0.041336 +2023-10-02 20:52:43,950 - Epoch: [50][ 50/ 1236] Overall Loss 0.237962 Objective Loss 0.237962 LR 0.001000 Time 0.037203 +2023-10-02 20:52:44,162 - Epoch: [50][ 60/ 1236] Overall Loss 0.241697 Objective Loss 0.241697 LR 0.001000 Time 0.034516 +2023-10-02 20:52:44,368 - Epoch: [50][ 70/ 1236] Overall Loss 0.241148 Objective Loss 0.241148 LR 0.001000 Time 0.032538 +2023-10-02 20:52:44,580 - Epoch: [50][ 80/ 1236] Overall Loss 0.239508 Objective Loss 0.239508 LR 0.001000 Time 0.031105 +2023-10-02 20:52:44,787 - Epoch: [50][ 90/ 1236] Overall Loss 0.238958 Objective Loss 0.238958 LR 0.001000 Time 0.029947 +2023-10-02 20:52:44,998 - Epoch: [50][ 100/ 1236] Overall Loss 0.241040 Objective Loss 0.241040 LR 0.001000 Time 0.029059 +2023-10-02 20:52:45,205 - Epoch: [50][ 110/ 1236] Overall Loss 0.241884 Objective Loss 0.241884 LR 0.001000 Time 0.028296 +2023-10-02 20:52:45,416 - Epoch: [50][ 120/ 1236] Overall Loss 0.243566 Objective Loss 0.243566 LR 0.001000 Time 0.027693 +2023-10-02 20:52:45,625 - Epoch: [50][ 130/ 1236] Overall Loss 0.242845 Objective Loss 0.242845 LR 0.001000 Time 0.027169 +2023-10-02 20:52:45,833 - Epoch: [50][ 140/ 1236] Overall Loss 0.244377 Objective Loss 0.244377 LR 0.001000 Time 0.026714 +2023-10-02 20:52:46,041 - Epoch: [50][ 150/ 1236] Overall Loss 0.244814 Objective Loss 0.244814 LR 0.001000 Time 0.026319 +2023-10-02 20:52:46,252 - Epoch: [50][ 160/ 1236] Overall Loss 0.245954 Objective Loss 0.245954 LR 0.001000 Time 0.025992 +2023-10-02 20:52:46,462 - Epoch: [50][ 170/ 1236] Overall Loss 0.247335 Objective Loss 0.247335 LR 0.001000 Time 0.025688 +2023-10-02 20:52:46,674 - Epoch: [50][ 180/ 1236] Overall Loss 0.246631 Objective Loss 0.246631 LR 0.001000 Time 0.025434 +2023-10-02 20:52:46,884 - Epoch: [50][ 190/ 1236] Overall Loss 0.246609 Objective Loss 0.246609 LR 0.001000 Time 0.025202 +2023-10-02 20:52:47,096 - Epoch: [50][ 200/ 1236] Overall Loss 0.245878 Objective Loss 0.245878 LR 0.001000 Time 0.024997 +2023-10-02 20:52:47,305 - Epoch: [50][ 210/ 1236] Overall Loss 0.246848 Objective Loss 0.246848 LR 0.001000 Time 0.024798 +2023-10-02 20:52:47,517 - Epoch: [50][ 220/ 1236] Overall Loss 0.247451 Objective Loss 0.247451 LR 0.001000 Time 0.024630 +2023-10-02 20:52:47,727 - Epoch: [50][ 230/ 1236] Overall Loss 0.247447 Objective Loss 0.247447 LR 0.001000 Time 0.024465 +2023-10-02 20:52:47,937 - Epoch: [50][ 240/ 1236] Overall Loss 0.247729 Objective Loss 0.247729 LR 0.001000 Time 0.024322 +2023-10-02 20:52:48,147 - Epoch: [50][ 250/ 1236] Overall Loss 0.248141 Objective Loss 0.248141 LR 0.001000 Time 0.024181 +2023-10-02 20:52:48,358 - Epoch: [50][ 260/ 1236] Overall Loss 0.247537 Objective Loss 0.247537 LR 0.001000 Time 0.024061 +2023-10-02 20:52:48,568 - Epoch: [50][ 270/ 1236] Overall Loss 0.248129 Objective Loss 0.248129 LR 0.001000 Time 0.023942 +2023-10-02 20:52:48,780 - Epoch: [50][ 280/ 1236] Overall Loss 0.248807 Objective Loss 0.248807 LR 0.001000 Time 0.023841 +2023-10-02 20:52:48,990 - Epoch: [50][ 290/ 1236] Overall Loss 0.248739 Objective Loss 0.248739 LR 0.001000 Time 0.023738 +2023-10-02 20:52:49,201 - Epoch: [50][ 300/ 1236] Overall Loss 0.249904 Objective Loss 0.249904 LR 0.001000 Time 0.023649 +2023-10-02 20:52:49,411 - Epoch: [50][ 310/ 1236] Overall Loss 0.250104 Objective Loss 0.250104 LR 0.001000 Time 0.023563 +2023-10-02 20:52:49,623 - Epoch: [50][ 320/ 1236] Overall Loss 0.249895 Objective Loss 0.249895 LR 0.001000 Time 0.023487 +2023-10-02 20:52:49,833 - Epoch: [50][ 330/ 1236] Overall Loss 0.251140 Objective Loss 0.251140 LR 0.001000 Time 0.023407 +2023-10-02 20:52:50,044 - Epoch: [50][ 340/ 1236] Overall Loss 0.251614 Objective Loss 0.251614 LR 0.001000 Time 0.023339 +2023-10-02 20:52:50,251 - Epoch: [50][ 350/ 1236] Overall Loss 0.252829 Objective Loss 0.252829 LR 0.001000 Time 0.023264 +2023-10-02 20:52:50,462 - Epoch: [50][ 360/ 1236] Overall Loss 0.252779 Objective Loss 0.252779 LR 0.001000 Time 0.023201 +2023-10-02 20:52:50,671 - Epoch: [50][ 370/ 1236] Overall Loss 0.253317 Objective Loss 0.253317 LR 0.001000 Time 0.023138 +2023-10-02 20:52:50,881 - Epoch: [50][ 380/ 1236] Overall Loss 0.254064 Objective Loss 0.254064 LR 0.001000 Time 0.023082 +2023-10-02 20:52:51,090 - Epoch: [50][ 390/ 1236] Overall Loss 0.254649 Objective Loss 0.254649 LR 0.001000 Time 0.023023 +2023-10-02 20:52:51,301 - Epoch: [50][ 400/ 1236] Overall Loss 0.254244 Objective Loss 0.254244 LR 0.001000 Time 0.022973 +2023-10-02 20:52:51,510 - Epoch: [50][ 410/ 1236] Overall Loss 0.254375 Objective Loss 0.254375 LR 0.001000 Time 0.022922 +2023-10-02 20:52:51,720 - Epoch: [50][ 420/ 1236] Overall Loss 0.255453 Objective Loss 0.255453 LR 0.001000 Time 0.022876 +2023-10-02 20:52:51,929 - Epoch: [50][ 430/ 1236] Overall Loss 0.255641 Objective Loss 0.255641 LR 0.001000 Time 0.022826 +2023-10-02 20:52:52,140 - Epoch: [50][ 440/ 1236] Overall Loss 0.255462 Objective Loss 0.255462 LR 0.001000 Time 0.022786 +2023-10-02 20:52:52,349 - Epoch: [50][ 450/ 1236] Overall Loss 0.255562 Objective Loss 0.255562 LR 0.001000 Time 0.022744 +2023-10-02 20:52:52,560 - Epoch: [50][ 460/ 1236] Overall Loss 0.254657 Objective Loss 0.254657 LR 0.001000 Time 0.022706 +2023-10-02 20:52:52,768 - Epoch: [50][ 470/ 1236] Overall Loss 0.254576 Objective Loss 0.254576 LR 0.001000 Time 0.022664 +2023-10-02 20:52:52,977 - Epoch: [50][ 480/ 1236] Overall Loss 0.254096 Objective Loss 0.254096 LR 0.001000 Time 0.022626 +2023-10-02 20:52:53,185 - Epoch: [50][ 490/ 1236] Overall Loss 0.254208 Objective Loss 0.254208 LR 0.001000 Time 0.022587 +2023-10-02 20:52:53,392 - Epoch: [50][ 500/ 1236] Overall Loss 0.254157 Objective Loss 0.254157 LR 0.001000 Time 0.022550 +2023-10-02 20:52:53,600 - Epoch: [50][ 510/ 1236] Overall Loss 0.253669 Objective Loss 0.253669 LR 0.001000 Time 0.022514 +2023-10-02 20:52:53,808 - Epoch: [50][ 520/ 1236] Overall Loss 0.254117 Objective Loss 0.254117 LR 0.001000 Time 0.022481 +2023-10-02 20:52:54,016 - Epoch: [50][ 530/ 1236] Overall Loss 0.253928 Objective Loss 0.253928 LR 0.001000 Time 0.022448 +2023-10-02 20:52:54,223 - Epoch: [50][ 540/ 1236] Overall Loss 0.254102 Objective Loss 0.254102 LR 0.001000 Time 0.022416 +2023-10-02 20:52:54,431 - Epoch: [50][ 550/ 1236] Overall Loss 0.254466 Objective Loss 0.254466 LR 0.001000 Time 0.022385 +2023-10-02 20:52:54,639 - Epoch: [50][ 560/ 1236] Overall Loss 0.254770 Objective Loss 0.254770 LR 0.001000 Time 0.022356 +2023-10-02 20:52:54,846 - Epoch: [50][ 570/ 1236] Overall Loss 0.254913 Objective Loss 0.254913 LR 0.001000 Time 0.022324 +2023-10-02 20:52:55,053 - Epoch: [50][ 580/ 1236] Overall Loss 0.254293 Objective Loss 0.254293 LR 0.001000 Time 0.022297 +2023-10-02 20:52:55,261 - Epoch: [50][ 590/ 1236] Overall Loss 0.253987 Objective Loss 0.253987 LR 0.001000 Time 0.022270 +2023-10-02 20:52:55,468 - Epoch: [50][ 600/ 1236] Overall Loss 0.253863 Objective Loss 0.253863 LR 0.001000 Time 0.022244 +2023-10-02 20:52:55,676 - Epoch: [50][ 610/ 1236] Overall Loss 0.253699 Objective Loss 0.253699 LR 0.001000 Time 0.022219 +2023-10-02 20:52:55,883 - Epoch: [50][ 620/ 1236] Overall Loss 0.253546 Objective Loss 0.253546 LR 0.001000 Time 0.022195 +2023-10-02 20:52:56,091 - Epoch: [50][ 630/ 1236] Overall Loss 0.253343 Objective Loss 0.253343 LR 0.001000 Time 0.022171 +2023-10-02 20:52:56,299 - Epoch: [50][ 640/ 1236] Overall Loss 0.253462 Objective Loss 0.253462 LR 0.001000 Time 0.022149 +2023-10-02 20:52:56,506 - Epoch: [50][ 650/ 1236] Overall Loss 0.253399 Objective Loss 0.253399 LR 0.001000 Time 0.022125 +2023-10-02 20:52:56,714 - Epoch: [50][ 660/ 1236] Overall Loss 0.253123 Objective Loss 0.253123 LR 0.001000 Time 0.022105 +2023-10-02 20:52:56,922 - Epoch: [50][ 670/ 1236] Overall Loss 0.253468 Objective Loss 0.253468 LR 0.001000 Time 0.022084 +2023-10-02 20:52:57,129 - Epoch: [50][ 680/ 1236] Overall Loss 0.253519 Objective Loss 0.253519 LR 0.001000 Time 0.022064 +2023-10-02 20:52:57,337 - Epoch: [50][ 690/ 1236] Overall Loss 0.253231 Objective Loss 0.253231 LR 0.001000 Time 0.022044 +2023-10-02 20:52:57,545 - Epoch: [50][ 700/ 1236] Overall Loss 0.253216 Objective Loss 0.253216 LR 0.001000 Time 0.022026 +2023-10-02 20:52:57,756 - Epoch: [50][ 710/ 1236] Overall Loss 0.253817 Objective Loss 0.253817 LR 0.001000 Time 0.022011 +2023-10-02 20:52:57,964 - Epoch: [50][ 720/ 1236] Overall Loss 0.254085 Objective Loss 0.254085 LR 0.001000 Time 0.021993 +2023-10-02 20:52:58,171 - Epoch: [50][ 730/ 1236] Overall Loss 0.254260 Objective Loss 0.254260 LR 0.001000 Time 0.021976 +2023-10-02 20:52:58,380 - Epoch: [50][ 740/ 1236] Overall Loss 0.254278 Objective Loss 0.254278 LR 0.001000 Time 0.021960 +2023-10-02 20:52:58,587 - Epoch: [50][ 750/ 1236] Overall Loss 0.254108 Objective Loss 0.254108 LR 0.001000 Time 0.021943 +2023-10-02 20:52:58,795 - Epoch: [50][ 760/ 1236] Overall Loss 0.254446 Objective Loss 0.254446 LR 0.001000 Time 0.021928 +2023-10-02 20:52:59,003 - Epoch: [50][ 770/ 1236] Overall Loss 0.254701 Objective Loss 0.254701 LR 0.001000 Time 0.021913 +2023-10-02 20:52:59,211 - Epoch: [50][ 780/ 1236] Overall Loss 0.254524 Objective Loss 0.254524 LR 0.001000 Time 0.021897 +2023-10-02 20:52:59,418 - Epoch: [50][ 790/ 1236] Overall Loss 0.254594 Objective Loss 0.254594 LR 0.001000 Time 0.021883 +2023-10-02 20:52:59,626 - Epoch: [50][ 800/ 1236] Overall Loss 0.254460 Objective Loss 0.254460 LR 0.001000 Time 0.021869 +2023-10-02 20:52:59,834 - Epoch: [50][ 810/ 1236] Overall Loss 0.254810 Objective Loss 0.254810 LR 0.001000 Time 0.021854 +2023-10-02 20:53:00,042 - Epoch: [50][ 820/ 1236] Overall Loss 0.254811 Objective Loss 0.254811 LR 0.001000 Time 0.021841 +2023-10-02 20:53:00,250 - Epoch: [50][ 830/ 1236] Overall Loss 0.255001 Objective Loss 0.255001 LR 0.001000 Time 0.021828 +2023-10-02 20:53:00,458 - Epoch: [50][ 840/ 1236] Overall Loss 0.255431 Objective Loss 0.255431 LR 0.001000 Time 0.021815 +2023-10-02 20:53:00,666 - Epoch: [50][ 850/ 1236] Overall Loss 0.255756 Objective Loss 0.255756 LR 0.001000 Time 0.021802 +2023-10-02 20:53:00,874 - Epoch: [50][ 860/ 1236] Overall Loss 0.256016 Objective Loss 0.256016 LR 0.001000 Time 0.021790 +2023-10-02 20:53:01,083 - Epoch: [50][ 870/ 1236] Overall Loss 0.256000 Objective Loss 0.256000 LR 0.001000 Time 0.021780 +2023-10-02 20:53:01,293 - Epoch: [50][ 880/ 1236] Overall Loss 0.256457 Objective Loss 0.256457 LR 0.001000 Time 0.021770 +2023-10-02 20:53:01,501 - Epoch: [50][ 890/ 1236] Overall Loss 0.256964 Objective Loss 0.256964 LR 0.001000 Time 0.021758 +2023-10-02 20:53:01,711 - Epoch: [50][ 900/ 1236] Overall Loss 0.256825 Objective Loss 0.256825 LR 0.001000 Time 0.021749 +2023-10-02 20:53:01,919 - Epoch: [50][ 910/ 1236] Overall Loss 0.257154 Objective Loss 0.257154 LR 0.001000 Time 0.021737 +2023-10-02 20:53:02,129 - Epoch: [50][ 920/ 1236] Overall Loss 0.257294 Objective Loss 0.257294 LR 0.001000 Time 0.021728 +2023-10-02 20:53:02,337 - Epoch: [50][ 930/ 1236] Overall Loss 0.257152 Objective Loss 0.257152 LR 0.001000 Time 0.021717 +2023-10-02 20:53:02,546 - Epoch: [50][ 940/ 1236] Overall Loss 0.257191 Objective Loss 0.257191 LR 0.001000 Time 0.021708 +2023-10-02 20:53:02,755 - Epoch: [50][ 950/ 1236] Overall Loss 0.257266 Objective Loss 0.257266 LR 0.001000 Time 0.021699 +2023-10-02 20:53:02,964 - Epoch: [50][ 960/ 1236] Overall Loss 0.256990 Objective Loss 0.256990 LR 0.001000 Time 0.021691 +2023-10-02 20:53:03,173 - Epoch: [50][ 970/ 1236] Overall Loss 0.256976 Objective Loss 0.256976 LR 0.001000 Time 0.021680 +2023-10-02 20:53:03,382 - Epoch: [50][ 980/ 1236] Overall Loss 0.257183 Objective Loss 0.257183 LR 0.001000 Time 0.021672 +2023-10-02 20:53:03,591 - Epoch: [50][ 990/ 1236] Overall Loss 0.257470 Objective Loss 0.257470 LR 0.001000 Time 0.021664 +2023-10-02 20:53:03,800 - Epoch: [50][ 1000/ 1236] Overall Loss 0.257645 Objective Loss 0.257645 LR 0.001000 Time 0.021656 +2023-10-02 20:53:04,009 - Epoch: [50][ 1010/ 1236] Overall Loss 0.257574 Objective Loss 0.257574 LR 0.001000 Time 0.021648 +2023-10-02 20:53:04,218 - Epoch: [50][ 1020/ 1236] Overall Loss 0.257605 Objective Loss 0.257605 LR 0.001000 Time 0.021641 +2023-10-02 20:53:04,427 - Epoch: [50][ 1030/ 1236] Overall Loss 0.257665 Objective Loss 0.257665 LR 0.001000 Time 0.021633 +2023-10-02 20:53:04,636 - Epoch: [50][ 1040/ 1236] Overall Loss 0.258015 Objective Loss 0.258015 LR 0.001000 Time 0.021626 +2023-10-02 20:53:04,845 - Epoch: [50][ 1050/ 1236] Overall Loss 0.258055 Objective Loss 0.258055 LR 0.001000 Time 0.021617 +2023-10-02 20:53:05,054 - Epoch: [50][ 1060/ 1236] Overall Loss 0.258439 Objective Loss 0.258439 LR 0.001000 Time 0.021611 +2023-10-02 20:53:05,263 - Epoch: [50][ 1070/ 1236] Overall Loss 0.258677 Objective Loss 0.258677 LR 0.001000 Time 0.021603 +2023-10-02 20:53:05,472 - Epoch: [50][ 1080/ 1236] Overall Loss 0.258648 Objective Loss 0.258648 LR 0.001000 Time 0.021597 +2023-10-02 20:53:05,681 - Epoch: [50][ 1090/ 1236] Overall Loss 0.258620 Objective Loss 0.258620 LR 0.001000 Time 0.021590 +2023-10-02 20:53:05,891 - Epoch: [50][ 1100/ 1236] Overall Loss 0.258952 Objective Loss 0.258952 LR 0.001000 Time 0.021584 +2023-10-02 20:53:06,099 - Epoch: [50][ 1110/ 1236] Overall Loss 0.258920 Objective Loss 0.258920 LR 0.001000 Time 0.021577 +2023-10-02 20:53:06,309 - Epoch: [50][ 1120/ 1236] Overall Loss 0.258800 Objective Loss 0.258800 LR 0.001000 Time 0.021571 +2023-10-02 20:53:06,518 - Epoch: [50][ 1130/ 1236] Overall Loss 0.258916 Objective Loss 0.258916 LR 0.001000 Time 0.021565 +2023-10-02 20:53:06,727 - Epoch: [50][ 1140/ 1236] Overall Loss 0.259067 Objective Loss 0.259067 LR 0.001000 Time 0.021559 +2023-10-02 20:53:06,936 - Epoch: [50][ 1150/ 1236] Overall Loss 0.259139 Objective Loss 0.259139 LR 0.001000 Time 0.021552 +2023-10-02 20:53:07,146 - Epoch: [50][ 1160/ 1236] Overall Loss 0.259311 Objective Loss 0.259311 LR 0.001000 Time 0.021547 +2023-10-02 20:53:07,355 - Epoch: [50][ 1170/ 1236] Overall Loss 0.259740 Objective Loss 0.259740 LR 0.001000 Time 0.021541 +2023-10-02 20:53:07,564 - Epoch: [50][ 1180/ 1236] Overall Loss 0.259712 Objective Loss 0.259712 LR 0.001000 Time 0.021536 +2023-10-02 20:53:07,773 - Epoch: [50][ 1190/ 1236] Overall Loss 0.259614 Objective Loss 0.259614 LR 0.001000 Time 0.021530 +2023-10-02 20:53:07,983 - Epoch: [50][ 1200/ 1236] Overall Loss 0.259750 Objective Loss 0.259750 LR 0.001000 Time 0.021525 +2023-10-02 20:53:08,192 - Epoch: [50][ 1210/ 1236] Overall Loss 0.259586 Objective Loss 0.259586 LR 0.001000 Time 0.021519 +2023-10-02 20:53:08,401 - Epoch: [50][ 1220/ 1236] Overall Loss 0.259707 Objective Loss 0.259707 LR 0.001000 Time 0.021515 +2023-10-02 20:53:08,663 - Epoch: [50][ 1230/ 1236] Overall Loss 0.259534 Objective Loss 0.259534 LR 0.001000 Time 0.021552 +2023-10-02 20:53:08,785 - Epoch: [50][ 1236/ 1236] Overall Loss 0.259594 Objective Loss 0.259594 Top1 84.114053 Top5 98.370672 LR 0.001000 Time 0.021546 +2023-10-02 20:53:08,917 - --- validate (epoch=50)----------- +2023-10-02 20:53:08,918 - 29943 samples (256 per mini-batch) +2023-10-02 20:53:09,401 - Epoch: [50][ 10/ 117] Loss 0.351419 Top1 83.476562 Top5 98.828125 +2023-10-02 20:53:09,554 - Epoch: [50][ 20/ 117] Loss 0.338946 Top1 83.222656 Top5 98.632812 +2023-10-02 20:53:09,707 - Epoch: [50][ 30/ 117] Loss 0.336142 Top1 83.528646 Top5 98.606771 +2023-10-02 20:53:09,859 - Epoch: [50][ 40/ 117] Loss 0.335142 Top1 83.574219 Top5 98.408203 +2023-10-02 20:53:10,010 - Epoch: [50][ 50/ 117] Loss 0.331838 Top1 83.726562 Top5 98.359375 +2023-10-02 20:53:10,161 - Epoch: [50][ 60/ 117] Loss 0.332358 Top1 83.906250 Top5 98.378906 +2023-10-02 20:53:10,314 - Epoch: [50][ 70/ 117] Loss 0.331610 Top1 83.850446 Top5 98.348214 +2023-10-02 20:53:10,468 - Epoch: [50][ 80/ 117] Loss 0.329377 Top1 83.955078 Top5 98.349609 +2023-10-02 20:53:10,624 - Epoch: [50][ 90/ 117] Loss 0.326893 Top1 84.001736 Top5 98.315972 +2023-10-02 20:53:10,780 - Epoch: [50][ 100/ 117] Loss 0.323386 Top1 84.066406 Top5 98.343750 +2023-10-02 20:53:10,942 - Epoch: [50][ 110/ 117] Loss 0.318702 Top1 84.101562 Top5 98.345170 +2023-10-02 20:53:11,031 - Epoch: [50][ 117/ 117] Loss 0.318055 Top1 84.129847 Top5 98.346859 +2023-10-02 20:53:11,147 - ==> Top1: 84.130 Top5: 98.347 Loss: 0.318 + +2023-10-02 20:53:11,147 - ==> Confusion: +[[ 961 0 3 0 7 4 0 0 11 34 1 0 0 4 2 1 6 0 0 0 16] + [ 0 1069 0 0 3 18 2 10 2 2 1 0 0 0 0 3 9 0 5 2 5] + [ 6 2 980 8 1 0 21 5 0 0 4 1 3 2 1 2 1 1 8 5 5] + [ 4 2 22 972 1 4 4 1 4 0 3 0 7 4 21 3 2 7 12 1 15] + [ 28 9 1 0 949 7 0 2 3 9 0 1 1 5 5 2 20 0 0 2 6] + [ 5 56 1 0 3 965 0 20 4 2 1 4 4 20 5 1 3 0 4 4 14] + [ 1 3 30 0 0 0 1126 3 0 0 6 0 0 0 0 6 0 0 1 10 5] + [ 2 42 25 0 2 30 6 1015 2 2 7 4 2 4 1 2 1 0 47 13 11] + [ 22 1 1 0 1 1 0 0 987 34 7 1 1 8 13 3 3 2 3 0 1] + [ 155 0 0 1 7 2 1 0 39 873 2 0 0 13 11 1 2 0 0 4 8] + [ 2 3 11 14 3 3 2 4 23 2 932 2 0 19 6 1 3 1 11 2 9] + [ 0 5 4 0 0 9 1 1 0 0 0 927 32 14 1 6 4 16 0 8 7] + [ 2 1 5 1 1 1 2 2 1 1 2 41 954 4 1 10 3 19 3 5 9] + [ 3 0 5 0 2 5 0 0 8 8 8 3 0 1063 4 0 1 0 0 2 7] + [ 14 2 2 13 4 0 0 0 30 2 2 1 5 4 1005 0 2 1 3 0 11] + [ 0 1 5 0 3 0 3 0 0 1 0 7 8 1 0 1059 21 11 2 5 7] + [ 0 11 1 0 5 5 0 0 2 0 0 2 0 2 2 8 1110 0 0 5 8] + [ 0 0 2 1 0 0 1 0 1 1 1 1 14 0 2 6 0 1000 0 4 4] + [ 1 7 6 20 2 0 0 13 5 0 5 0 0 0 10 1 0 0 989 1 8] + [ 0 2 5 2 0 2 14 7 0 1 1 9 1 2 0 2 4 0 0 1092 8] + [ 155 217 155 62 75 121 51 83 106 60 130 90 358 303 131 59 172 58 149 207 5163]] + +2023-10-02 20:53:11,149 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:53:11,149 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:53:11,155 - + +2023-10-02 20:53:11,155 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:53:12,162 - Epoch: [51][ 10/ 1236] Overall Loss 0.245486 Objective Loss 0.245486 LR 0.001000 Time 0.100629 +2023-10-02 20:53:12,371 - Epoch: [51][ 20/ 1236] Overall Loss 0.241369 Objective Loss 0.241369 LR 0.001000 Time 0.060729 +2023-10-02 20:53:12,578 - Epoch: [51][ 30/ 1236] Overall Loss 0.248036 Objective Loss 0.248036 LR 0.001000 Time 0.047355 +2023-10-02 20:53:12,788 - Epoch: [51][ 40/ 1236] Overall Loss 0.250873 Objective Loss 0.250873 LR 0.001000 Time 0.040757 +2023-10-02 20:53:12,994 - Epoch: [51][ 50/ 1236] Overall Loss 0.243991 Objective Loss 0.243991 LR 0.001000 Time 0.036726 +2023-10-02 20:53:13,204 - Epoch: [51][ 60/ 1236] Overall Loss 0.242806 Objective Loss 0.242806 LR 0.001000 Time 0.034099 +2023-10-02 20:53:13,411 - Epoch: [51][ 70/ 1236] Overall Loss 0.239894 Objective Loss 0.239894 LR 0.001000 Time 0.032171 +2023-10-02 20:53:13,621 - Epoch: [51][ 80/ 1236] Overall Loss 0.240289 Objective Loss 0.240289 LR 0.001000 Time 0.030768 +2023-10-02 20:53:13,827 - Epoch: [51][ 90/ 1236] Overall Loss 0.243845 Objective Loss 0.243845 LR 0.001000 Time 0.029639 +2023-10-02 20:53:14,037 - Epoch: [51][ 100/ 1236] Overall Loss 0.243597 Objective Loss 0.243597 LR 0.001000 Time 0.028770 +2023-10-02 20:53:14,243 - Epoch: [51][ 110/ 1236] Overall Loss 0.244266 Objective Loss 0.244266 LR 0.001000 Time 0.028026 +2023-10-02 20:53:14,453 - Epoch: [51][ 120/ 1236] Overall Loss 0.243897 Objective Loss 0.243897 LR 0.001000 Time 0.027436 +2023-10-02 20:53:14,660 - Epoch: [51][ 130/ 1236] Overall Loss 0.244673 Objective Loss 0.244673 LR 0.001000 Time 0.026920 +2023-10-02 20:53:14,872 - Epoch: [51][ 140/ 1236] Overall Loss 0.247386 Objective Loss 0.247386 LR 0.001000 Time 0.026497 +2023-10-02 20:53:15,085 - Epoch: [51][ 150/ 1236] Overall Loss 0.245744 Objective Loss 0.245744 LR 0.001000 Time 0.026139 +2023-10-02 20:53:15,295 - Epoch: [51][ 160/ 1236] Overall Loss 0.246498 Objective Loss 0.246498 LR 0.001000 Time 0.025818 +2023-10-02 20:53:15,501 - Epoch: [51][ 170/ 1236] Overall Loss 0.243954 Objective Loss 0.243954 LR 0.001000 Time 0.025508 +2023-10-02 20:53:15,710 - Epoch: [51][ 180/ 1236] Overall Loss 0.244267 Objective Loss 0.244267 LR 0.001000 Time 0.025250 +2023-10-02 20:53:15,916 - Epoch: [51][ 190/ 1236] Overall Loss 0.244311 Objective Loss 0.244311 LR 0.001000 Time 0.025003 +2023-10-02 20:53:16,124 - Epoch: [51][ 200/ 1236] Overall Loss 0.246450 Objective Loss 0.246450 LR 0.001000 Time 0.024792 +2023-10-02 20:53:16,330 - Epoch: [51][ 210/ 1236] Overall Loss 0.246414 Objective Loss 0.246414 LR 0.001000 Time 0.024590 +2023-10-02 20:53:16,539 - Epoch: [51][ 220/ 1236] Overall Loss 0.247023 Objective Loss 0.247023 LR 0.001000 Time 0.024416 +2023-10-02 20:53:16,746 - Epoch: [51][ 230/ 1236] Overall Loss 0.247452 Objective Loss 0.247452 LR 0.001000 Time 0.024248 +2023-10-02 20:53:16,954 - Epoch: [51][ 240/ 1236] Overall Loss 0.247685 Objective Loss 0.247685 LR 0.001000 Time 0.024103 +2023-10-02 20:53:17,161 - Epoch: [51][ 250/ 1236] Overall Loss 0.248530 Objective Loss 0.248530 LR 0.001000 Time 0.023962 +2023-10-02 20:53:17,369 - Epoch: [51][ 260/ 1236] Overall Loss 0.249403 Objective Loss 0.249403 LR 0.001000 Time 0.023839 +2023-10-02 20:53:17,576 - Epoch: [51][ 270/ 1236] Overall Loss 0.249550 Objective Loss 0.249550 LR 0.001000 Time 0.023717 +2023-10-02 20:53:17,784 - Epoch: [51][ 280/ 1236] Overall Loss 0.249139 Objective Loss 0.249139 LR 0.001000 Time 0.023612 +2023-10-02 20:53:17,991 - Epoch: [51][ 290/ 1236] Overall Loss 0.250662 Objective Loss 0.250662 LR 0.001000 Time 0.023507 +2023-10-02 20:53:18,199 - Epoch: [51][ 300/ 1236] Overall Loss 0.250719 Objective Loss 0.250719 LR 0.001000 Time 0.023415 +2023-10-02 20:53:18,406 - Epoch: [51][ 310/ 1236] Overall Loss 0.250552 Objective Loss 0.250552 LR 0.001000 Time 0.023323 +2023-10-02 20:53:18,614 - Epoch: [51][ 320/ 1236] Overall Loss 0.249791 Objective Loss 0.249791 LR 0.001000 Time 0.023244 +2023-10-02 20:53:18,822 - Epoch: [51][ 330/ 1236] Overall Loss 0.249033 Objective Loss 0.249033 LR 0.001000 Time 0.023166 +2023-10-02 20:53:19,033 - Epoch: [51][ 340/ 1236] Overall Loss 0.249562 Objective Loss 0.249562 LR 0.001000 Time 0.023103 +2023-10-02 20:53:19,240 - Epoch: [51][ 350/ 1236] Overall Loss 0.249601 Objective Loss 0.249601 LR 0.001000 Time 0.023034 +2023-10-02 20:53:19,450 - Epoch: [51][ 360/ 1236] Overall Loss 0.248582 Objective Loss 0.248582 LR 0.001000 Time 0.022978 +2023-10-02 20:53:19,658 - Epoch: [51][ 370/ 1236] Overall Loss 0.248419 Objective Loss 0.248419 LR 0.001000 Time 0.022916 +2023-10-02 20:53:19,868 - Epoch: [51][ 380/ 1236] Overall Loss 0.248896 Objective Loss 0.248896 LR 0.001000 Time 0.022867 +2023-10-02 20:53:20,075 - Epoch: [51][ 390/ 1236] Overall Loss 0.248473 Objective Loss 0.248473 LR 0.001000 Time 0.022811 +2023-10-02 20:53:20,286 - Epoch: [51][ 400/ 1236] Overall Loss 0.248395 Objective Loss 0.248395 LR 0.001000 Time 0.022766 +2023-10-02 20:53:20,493 - Epoch: [51][ 410/ 1236] Overall Loss 0.249190 Objective Loss 0.249190 LR 0.001000 Time 0.022715 +2023-10-02 20:53:20,704 - Epoch: [51][ 420/ 1236] Overall Loss 0.250210 Objective Loss 0.250210 LR 0.001000 Time 0.022675 +2023-10-02 20:53:20,911 - Epoch: [51][ 430/ 1236] Overall Loss 0.251330 Objective Loss 0.251330 LR 0.001000 Time 0.022629 +2023-10-02 20:53:21,121 - Epoch: [51][ 440/ 1236] Overall Loss 0.251706 Objective Loss 0.251706 LR 0.001000 Time 0.022593 +2023-10-02 20:53:21,329 - Epoch: [51][ 450/ 1236] Overall Loss 0.251676 Objective Loss 0.251676 LR 0.001000 Time 0.022550 +2023-10-02 20:53:21,539 - Epoch: [51][ 460/ 1236] Overall Loss 0.251961 Objective Loss 0.251961 LR 0.001000 Time 0.022517 +2023-10-02 20:53:21,747 - Epoch: [51][ 470/ 1236] Overall Loss 0.251825 Objective Loss 0.251825 LR 0.001000 Time 0.022479 +2023-10-02 20:53:21,957 - Epoch: [51][ 480/ 1236] Overall Loss 0.252331 Objective Loss 0.252331 LR 0.001000 Time 0.022448 +2023-10-02 20:53:22,164 - Epoch: [51][ 490/ 1236] Overall Loss 0.252062 Objective Loss 0.252062 LR 0.001000 Time 0.022412 +2023-10-02 20:53:22,375 - Epoch: [51][ 500/ 1236] Overall Loss 0.252893 Objective Loss 0.252893 LR 0.001000 Time 0.022385 +2023-10-02 20:53:22,582 - Epoch: [51][ 510/ 1236] Overall Loss 0.252826 Objective Loss 0.252826 LR 0.001000 Time 0.022351 +2023-10-02 20:53:22,793 - Epoch: [51][ 520/ 1236] Overall Loss 0.252752 Objective Loss 0.252752 LR 0.001000 Time 0.022326 +2023-10-02 20:53:23,000 - Epoch: [51][ 530/ 1236] Overall Loss 0.252655 Objective Loss 0.252655 LR 0.001000 Time 0.022295 +2023-10-02 20:53:23,210 - Epoch: [51][ 540/ 1236] Overall Loss 0.252972 Objective Loss 0.252972 LR 0.001000 Time 0.022272 +2023-10-02 20:53:23,418 - Epoch: [51][ 550/ 1236] Overall Loss 0.253119 Objective Loss 0.253119 LR 0.001000 Time 0.022243 +2023-10-02 20:53:23,628 - Epoch: [51][ 560/ 1236] Overall Loss 0.252878 Objective Loss 0.252878 LR 0.001000 Time 0.022221 +2023-10-02 20:53:23,836 - Epoch: [51][ 570/ 1236] Overall Loss 0.252730 Objective Loss 0.252730 LR 0.001000 Time 0.022194 +2023-10-02 20:53:24,046 - Epoch: [51][ 580/ 1236] Overall Loss 0.252846 Objective Loss 0.252846 LR 0.001000 Time 0.022174 +2023-10-02 20:53:24,253 - Epoch: [51][ 590/ 1236] Overall Loss 0.253436 Objective Loss 0.253436 LR 0.001000 Time 0.022149 +2023-10-02 20:53:24,464 - Epoch: [51][ 600/ 1236] Overall Loss 0.253293 Objective Loss 0.253293 LR 0.001000 Time 0.022131 +2023-10-02 20:53:24,671 - Epoch: [51][ 610/ 1236] Overall Loss 0.252702 Objective Loss 0.252702 LR 0.001000 Time 0.022107 +2023-10-02 20:53:24,882 - Epoch: [51][ 620/ 1236] Overall Loss 0.252555 Objective Loss 0.252555 LR 0.001000 Time 0.022090 +2023-10-02 20:53:25,089 - Epoch: [51][ 630/ 1236] Overall Loss 0.252885 Objective Loss 0.252885 LR 0.001000 Time 0.022068 +2023-10-02 20:53:25,299 - Epoch: [51][ 640/ 1236] Overall Loss 0.253050 Objective Loss 0.253050 LR 0.001000 Time 0.022050 +2023-10-02 20:53:25,508 - Epoch: [51][ 650/ 1236] Overall Loss 0.253227 Objective Loss 0.253227 LR 0.001000 Time 0.022030 +2023-10-02 20:53:25,718 - Epoch: [51][ 660/ 1236] Overall Loss 0.253503 Objective Loss 0.253503 LR 0.001000 Time 0.022014 +2023-10-02 20:53:25,927 - Epoch: [51][ 670/ 1236] Overall Loss 0.254142 Objective Loss 0.254142 LR 0.001000 Time 0.021994 +2023-10-02 20:53:26,136 - Epoch: [51][ 680/ 1236] Overall Loss 0.253828 Objective Loss 0.253828 LR 0.001000 Time 0.021979 +2023-10-02 20:53:26,345 - Epoch: [51][ 690/ 1236] Overall Loss 0.253761 Objective Loss 0.253761 LR 0.001000 Time 0.021961 +2023-10-02 20:53:26,555 - Epoch: [51][ 700/ 1236] Overall Loss 0.253819 Objective Loss 0.253819 LR 0.001000 Time 0.021946 +2023-10-02 20:53:26,764 - Epoch: [51][ 710/ 1236] Overall Loss 0.253588 Objective Loss 0.253588 LR 0.001000 Time 0.021929 +2023-10-02 20:53:26,973 - Epoch: [51][ 720/ 1236] Overall Loss 0.253652 Objective Loss 0.253652 LR 0.001000 Time 0.021915 +2023-10-02 20:53:27,182 - Epoch: [51][ 730/ 1236] Overall Loss 0.253554 Objective Loss 0.253554 LR 0.001000 Time 0.021899 +2023-10-02 20:53:27,392 - Epoch: [51][ 740/ 1236] Overall Loss 0.253758 Objective Loss 0.253758 LR 0.001000 Time 0.021886 +2023-10-02 20:53:27,601 - Epoch: [51][ 750/ 1236] Overall Loss 0.253623 Objective Loss 0.253623 LR 0.001000 Time 0.021870 +2023-10-02 20:53:27,811 - Epoch: [51][ 760/ 1236] Overall Loss 0.253765 Objective Loss 0.253765 LR 0.001000 Time 0.021858 +2023-10-02 20:53:28,023 - Epoch: [51][ 770/ 1236] Overall Loss 0.254376 Objective Loss 0.254376 LR 0.001000 Time 0.021848 +2023-10-02 20:53:28,235 - Epoch: [51][ 780/ 1236] Overall Loss 0.255135 Objective Loss 0.255135 LR 0.001000 Time 0.021839 +2023-10-02 20:53:28,447 - Epoch: [51][ 790/ 1236] Overall Loss 0.255495 Objective Loss 0.255495 LR 0.001000 Time 0.021829 +2023-10-02 20:53:28,658 - Epoch: [51][ 800/ 1236] Overall Loss 0.255740 Objective Loss 0.255740 LR 0.001000 Time 0.021819 +2023-10-02 20:53:28,867 - Epoch: [51][ 810/ 1236] Overall Loss 0.255747 Objective Loss 0.255747 LR 0.001000 Time 0.021808 +2023-10-02 20:53:29,077 - Epoch: [51][ 820/ 1236] Overall Loss 0.255824 Objective Loss 0.255824 LR 0.001000 Time 0.021797 +2023-10-02 20:53:29,286 - Epoch: [51][ 830/ 1236] Overall Loss 0.255922 Objective Loss 0.255922 LR 0.001000 Time 0.021786 +2023-10-02 20:53:29,496 - Epoch: [51][ 840/ 1236] Overall Loss 0.256125 Objective Loss 0.256125 LR 0.001000 Time 0.021776 +2023-10-02 20:53:29,704 - Epoch: [51][ 850/ 1236] Overall Loss 0.256669 Objective Loss 0.256669 LR 0.001000 Time 0.021765 +2023-10-02 20:53:29,915 - Epoch: [51][ 860/ 1236] Overall Loss 0.256547 Objective Loss 0.256547 LR 0.001000 Time 0.021756 +2023-10-02 20:53:30,124 - Epoch: [51][ 870/ 1236] Overall Loss 0.256485 Objective Loss 0.256485 LR 0.001000 Time 0.021746 +2023-10-02 20:53:30,335 - Epoch: [51][ 880/ 1236] Overall Loss 0.256696 Objective Loss 0.256696 LR 0.001000 Time 0.021738 +2023-10-02 20:53:30,543 - Epoch: [51][ 890/ 1236] Overall Loss 0.256775 Objective Loss 0.256775 LR 0.001000 Time 0.021728 +2023-10-02 20:53:30,754 - Epoch: [51][ 900/ 1236] Overall Loss 0.256846 Objective Loss 0.256846 LR 0.001000 Time 0.021720 +2023-10-02 20:53:30,963 - Epoch: [51][ 910/ 1236] Overall Loss 0.256661 Objective Loss 0.256661 LR 0.001000 Time 0.021710 +2023-10-02 20:53:31,174 - Epoch: [51][ 920/ 1236] Overall Loss 0.256669 Objective Loss 0.256669 LR 0.001000 Time 0.021703 +2023-10-02 20:53:31,382 - Epoch: [51][ 930/ 1236] Overall Loss 0.256810 Objective Loss 0.256810 LR 0.001000 Time 0.021694 +2023-10-02 20:53:31,594 - Epoch: [51][ 940/ 1236] Overall Loss 0.257225 Objective Loss 0.257225 LR 0.001000 Time 0.021688 +2023-10-02 20:53:31,802 - Epoch: [51][ 950/ 1236] Overall Loss 0.257108 Objective Loss 0.257108 LR 0.001000 Time 0.021679 +2023-10-02 20:53:32,014 - Epoch: [51][ 960/ 1236] Overall Loss 0.257080 Objective Loss 0.257080 LR 0.001000 Time 0.021672 +2023-10-02 20:53:32,223 - Epoch: [51][ 970/ 1236] Overall Loss 0.257368 Objective Loss 0.257368 LR 0.001000 Time 0.021664 +2023-10-02 20:53:32,433 - Epoch: [51][ 980/ 1236] Overall Loss 0.257555 Objective Loss 0.257555 LR 0.001000 Time 0.021658 +2023-10-02 20:53:32,642 - Epoch: [51][ 990/ 1236] Overall Loss 0.257528 Objective Loss 0.257528 LR 0.001000 Time 0.021650 +2023-10-02 20:53:32,853 - Epoch: [51][ 1000/ 1236] Overall Loss 0.257548 Objective Loss 0.257548 LR 0.001000 Time 0.021644 +2023-10-02 20:53:33,061 - Epoch: [51][ 1010/ 1236] Overall Loss 0.257272 Objective Loss 0.257272 LR 0.001000 Time 0.021635 +2023-10-02 20:53:33,272 - Epoch: [51][ 1020/ 1236] Overall Loss 0.257353 Objective Loss 0.257353 LR 0.001000 Time 0.021629 +2023-10-02 20:53:33,481 - Epoch: [51][ 1030/ 1236] Overall Loss 0.257394 Objective Loss 0.257394 LR 0.001000 Time 0.021622 +2023-10-02 20:53:33,692 - Epoch: [51][ 1040/ 1236] Overall Loss 0.257041 Objective Loss 0.257041 LR 0.001000 Time 0.021616 +2023-10-02 20:53:33,901 - Epoch: [51][ 1050/ 1236] Overall Loss 0.256767 Objective Loss 0.256767 LR 0.001000 Time 0.021609 +2023-10-02 20:53:34,111 - Epoch: [51][ 1060/ 1236] Overall Loss 0.256641 Objective Loss 0.256641 LR 0.001000 Time 0.021603 +2023-10-02 20:53:34,320 - Epoch: [51][ 1070/ 1236] Overall Loss 0.256463 Objective Loss 0.256463 LR 0.001000 Time 0.021596 +2023-10-02 20:53:34,531 - Epoch: [51][ 1080/ 1236] Overall Loss 0.256435 Objective Loss 0.256435 LR 0.001000 Time 0.021591 +2023-10-02 20:53:34,740 - Epoch: [51][ 1090/ 1236] Overall Loss 0.256516 Objective Loss 0.256516 LR 0.001000 Time 0.021585 +2023-10-02 20:53:34,951 - Epoch: [51][ 1100/ 1236] Overall Loss 0.256366 Objective Loss 0.256366 LR 0.001000 Time 0.021580 +2023-10-02 20:53:35,160 - Epoch: [51][ 1110/ 1236] Overall Loss 0.256464 Objective Loss 0.256464 LR 0.001000 Time 0.021574 +2023-10-02 20:53:35,371 - Epoch: [51][ 1120/ 1236] Overall Loss 0.256451 Objective Loss 0.256451 LR 0.001000 Time 0.021569 +2023-10-02 20:53:35,580 - Epoch: [51][ 1130/ 1236] Overall Loss 0.256639 Objective Loss 0.256639 LR 0.001000 Time 0.021562 +2023-10-02 20:53:35,790 - Epoch: [51][ 1140/ 1236] Overall Loss 0.256731 Objective Loss 0.256731 LR 0.001000 Time 0.021558 +2023-10-02 20:53:35,999 - Epoch: [51][ 1150/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.021552 +2023-10-02 20:53:36,210 - Epoch: [51][ 1160/ 1236] Overall Loss 0.256845 Objective Loss 0.256845 LR 0.001000 Time 0.021547 +2023-10-02 20:53:36,419 - Epoch: [51][ 1170/ 1236] Overall Loss 0.256986 Objective Loss 0.256986 LR 0.001000 Time 0.021541 +2023-10-02 20:53:36,630 - Epoch: [51][ 1180/ 1236] Overall Loss 0.256895 Objective Loss 0.256895 LR 0.001000 Time 0.021537 +2023-10-02 20:53:36,839 - Epoch: [51][ 1190/ 1236] Overall Loss 0.256770 Objective Loss 0.256770 LR 0.001000 Time 0.021532 +2023-10-02 20:53:37,050 - Epoch: [51][ 1200/ 1236] Overall Loss 0.256941 Objective Loss 0.256941 LR 0.001000 Time 0.021528 +2023-10-02 20:53:37,259 - Epoch: [51][ 1210/ 1236] Overall Loss 0.257325 Objective Loss 0.257325 LR 0.001000 Time 0.021522 +2023-10-02 20:53:37,470 - Epoch: [51][ 1220/ 1236] Overall Loss 0.257315 Objective Loss 0.257315 LR 0.001000 Time 0.021519 +2023-10-02 20:53:37,733 - Epoch: [51][ 1230/ 1236] Overall Loss 0.257534 Objective Loss 0.257534 LR 0.001000 Time 0.021557 +2023-10-02 20:53:37,856 - Epoch: [51][ 1236/ 1236] Overall Loss 0.257488 Objective Loss 0.257488 Top1 87.576375 Top5 97.759674 LR 0.001000 Time 0.021552 +2023-10-02 20:53:37,996 - --- validate (epoch=51)----------- +2023-10-02 20:53:37,996 - 29943 samples (256 per mini-batch) +2023-10-02 20:53:38,504 - Epoch: [51][ 10/ 117] Loss 0.346206 Top1 83.242188 Top5 98.437500 +2023-10-02 20:53:38,666 - Epoch: [51][ 20/ 117] Loss 0.337934 Top1 83.027344 Top5 98.457031 +2023-10-02 20:53:38,825 - Epoch: [51][ 30/ 117] Loss 0.341324 Top1 82.981771 Top5 98.294271 +2023-10-02 20:53:38,987 - Epoch: [51][ 40/ 117] Loss 0.343753 Top1 83.027344 Top5 98.369141 +2023-10-02 20:53:39,145 - Epoch: [51][ 50/ 117] Loss 0.336453 Top1 83.218750 Top5 98.351562 +2023-10-02 20:53:39,307 - Epoch: [51][ 60/ 117] Loss 0.333243 Top1 83.417969 Top5 98.365885 +2023-10-02 20:53:39,467 - Epoch: [51][ 70/ 117] Loss 0.326360 Top1 83.526786 Top5 98.392857 +2023-10-02 20:53:39,636 - Epoch: [51][ 80/ 117] Loss 0.324611 Top1 83.618164 Top5 98.374023 +2023-10-02 20:53:39,804 - Epoch: [51][ 90/ 117] Loss 0.324146 Top1 83.641493 Top5 98.372396 +2023-10-02 20:53:39,973 - Epoch: [51][ 100/ 117] Loss 0.323397 Top1 83.660156 Top5 98.316406 +2023-10-02 20:53:40,146 - Epoch: [51][ 110/ 117] Loss 0.323103 Top1 83.643466 Top5 98.291903 +2023-10-02 20:53:40,236 - Epoch: [51][ 117/ 117] Loss 0.321397 Top1 83.685669 Top5 98.306783 +2023-10-02 20:53:40,377 - ==> Top1: 83.686 Top5: 98.307 Loss: 0.321 + +2023-10-02 20:53:40,378 - ==> Confusion: +[[ 951 0 2 0 7 8 0 1 5 42 0 2 2 4 7 0 1 2 0 0 16] + [ 0 1046 1 2 2 40 2 26 1 0 0 0 0 0 0 1 1 1 2 2 4] + [ 6 1 941 14 3 2 26 14 0 0 2 3 7 1 5 5 0 1 12 7 6] + [ 0 4 8 983 0 9 5 2 4 1 3 0 7 3 25 0 0 3 15 2 15] + [ 31 8 2 1 953 13 0 2 2 8 0 0 0 5 7 6 7 0 1 2 2] + [ 4 35 1 0 1 998 1 27 1 6 1 11 3 7 3 0 0 0 4 5 8] + [ 1 4 21 0 0 1 1133 8 0 0 3 3 0 0 1 2 0 1 0 7 6] + [ 2 13 13 0 0 28 6 1087 1 0 1 7 6 5 1 0 0 1 33 10 4] + [ 22 4 0 0 0 3 0 1 974 35 5 3 4 8 15 3 2 2 4 3 1] + [ 119 3 0 0 7 5 0 0 25 917 0 1 0 20 4 2 1 0 0 2 13] + [ 2 2 8 8 2 1 1 4 26 1 953 4 1 12 5 0 0 1 8 2 12] + [ 0 0 0 0 0 12 0 4 0 0 0 976 23 4 0 2 0 12 0 1 1] + [ 0 1 4 1 0 7 3 2 0 0 1 53 962 2 2 6 1 9 1 6 7] + [ 0 0 1 0 2 15 0 1 14 9 8 8 1 1039 6 1 0 1 0 3 10] + [ 15 1 2 17 9 4 0 0 27 9 4 0 4 2 992 0 1 1 3 0 10] + [ 0 0 2 0 4 0 0 0 0 0 0 12 10 0 0 1064 17 11 1 5 8] + [ 1 17 1 0 7 9 0 2 3 0 0 6 2 2 2 5 1085 0 0 6 13] + [ 1 0 0 0 0 1 1 0 0 0 1 17 30 2 4 8 1 967 0 1 4] + [ 1 4 6 15 0 0 1 44 4 2 1 0 1 0 12 0 0 0 963 1 13] + [ 0 1 4 1 0 4 4 18 0 0 1 18 5 2 1 3 5 0 1 1075 9] + [ 158 239 86 65 88 233 53 149 92 72 116 196 385 280 133 59 96 46 154 206 4999]] + +2023-10-02 20:53:40,379 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:53:40,379 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:53:40,386 - + +2023-10-02 20:53:40,386 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:53:41,413 - Epoch: [52][ 10/ 1236] Overall Loss 0.202937 Objective Loss 0.202937 LR 0.001000 Time 0.102709 +2023-10-02 20:53:41,624 - Epoch: [52][ 20/ 1236] Overall Loss 0.210040 Objective Loss 0.210040 LR 0.001000 Time 0.061880 +2023-10-02 20:53:41,834 - Epoch: [52][ 30/ 1236] Overall Loss 0.229591 Objective Loss 0.229591 LR 0.001000 Time 0.048241 +2023-10-02 20:53:42,046 - Epoch: [52][ 40/ 1236] Overall Loss 0.239097 Objective Loss 0.239097 LR 0.001000 Time 0.041468 +2023-10-02 20:53:42,255 - Epoch: [52][ 50/ 1236] Overall Loss 0.239770 Objective Loss 0.239770 LR 0.001000 Time 0.037341 +2023-10-02 20:53:42,467 - Epoch: [52][ 60/ 1236] Overall Loss 0.243927 Objective Loss 0.243927 LR 0.001000 Time 0.034648 +2023-10-02 20:53:42,675 - Epoch: [52][ 70/ 1236] Overall Loss 0.243429 Objective Loss 0.243429 LR 0.001000 Time 0.032674 +2023-10-02 20:53:42,887 - Epoch: [52][ 80/ 1236] Overall Loss 0.238501 Objective Loss 0.238501 LR 0.001000 Time 0.031236 +2023-10-02 20:53:43,096 - Epoch: [52][ 90/ 1236] Overall Loss 0.238014 Objective Loss 0.238014 LR 0.001000 Time 0.030080 +2023-10-02 20:53:43,307 - Epoch: [52][ 100/ 1236] Overall Loss 0.236915 Objective Loss 0.236915 LR 0.001000 Time 0.029177 +2023-10-02 20:53:43,516 - Epoch: [52][ 110/ 1236] Overall Loss 0.235667 Objective Loss 0.235667 LR 0.001000 Time 0.028416 +2023-10-02 20:53:43,728 - Epoch: [52][ 120/ 1236] Overall Loss 0.236883 Objective Loss 0.236883 LR 0.001000 Time 0.027813 +2023-10-02 20:53:43,937 - Epoch: [52][ 130/ 1236] Overall Loss 0.238663 Objective Loss 0.238663 LR 0.001000 Time 0.027279 +2023-10-02 20:53:44,150 - Epoch: [52][ 140/ 1236] Overall Loss 0.239003 Objective Loss 0.239003 LR 0.001000 Time 0.026848 +2023-10-02 20:53:44,359 - Epoch: [52][ 150/ 1236] Overall Loss 0.239867 Objective Loss 0.239867 LR 0.001000 Time 0.026448 +2023-10-02 20:53:44,571 - Epoch: [52][ 160/ 1236] Overall Loss 0.241768 Objective Loss 0.241768 LR 0.001000 Time 0.026121 +2023-10-02 20:53:44,780 - Epoch: [52][ 170/ 1236] Overall Loss 0.244058 Objective Loss 0.244058 LR 0.001000 Time 0.025810 +2023-10-02 20:53:44,993 - Epoch: [52][ 180/ 1236] Overall Loss 0.243079 Objective Loss 0.243079 LR 0.001000 Time 0.025551 +2023-10-02 20:53:45,202 - Epoch: [52][ 190/ 1236] Overall Loss 0.242736 Objective Loss 0.242736 LR 0.001000 Time 0.025304 +2023-10-02 20:53:45,414 - Epoch: [52][ 200/ 1236] Overall Loss 0.244564 Objective Loss 0.244564 LR 0.001000 Time 0.025093 +2023-10-02 20:53:45,623 - Epoch: [52][ 210/ 1236] Overall Loss 0.245211 Objective Loss 0.245211 LR 0.001000 Time 0.024892 +2023-10-02 20:53:45,836 - Epoch: [52][ 220/ 1236] Overall Loss 0.246173 Objective Loss 0.246173 LR 0.001000 Time 0.024721 +2023-10-02 20:53:46,045 - Epoch: [52][ 230/ 1236] Overall Loss 0.245642 Objective Loss 0.245642 LR 0.001000 Time 0.024554 +2023-10-02 20:53:46,257 - Epoch: [52][ 240/ 1236] Overall Loss 0.244776 Objective Loss 0.244776 LR 0.001000 Time 0.024410 +2023-10-02 20:53:46,466 - Epoch: [52][ 250/ 1236] Overall Loss 0.245885 Objective Loss 0.245885 LR 0.001000 Time 0.024268 +2023-10-02 20:53:46,679 - Epoch: [52][ 260/ 1236] Overall Loss 0.245637 Objective Loss 0.245637 LR 0.001000 Time 0.024152 +2023-10-02 20:53:46,888 - Epoch: [52][ 270/ 1236] Overall Loss 0.245862 Objective Loss 0.245862 LR 0.001000 Time 0.024031 +2023-10-02 20:53:47,101 - Epoch: [52][ 280/ 1236] Overall Loss 0.245561 Objective Loss 0.245561 LR 0.001000 Time 0.023931 +2023-10-02 20:53:47,310 - Epoch: [52][ 290/ 1236] Overall Loss 0.246352 Objective Loss 0.246352 LR 0.001000 Time 0.023826 +2023-10-02 20:53:47,522 - Epoch: [52][ 300/ 1236] Overall Loss 0.246205 Objective Loss 0.246205 LR 0.001000 Time 0.023735 +2023-10-02 20:53:47,731 - Epoch: [52][ 310/ 1236] Overall Loss 0.246146 Objective Loss 0.246146 LR 0.001000 Time 0.023641 +2023-10-02 20:53:47,942 - Epoch: [52][ 320/ 1236] Overall Loss 0.245999 Objective Loss 0.245999 LR 0.001000 Time 0.023563 +2023-10-02 20:53:48,151 - Epoch: [52][ 330/ 1236] Overall Loss 0.246193 Objective Loss 0.246193 LR 0.001000 Time 0.023481 +2023-10-02 20:53:48,362 - Epoch: [52][ 340/ 1236] Overall Loss 0.246625 Objective Loss 0.246625 LR 0.001000 Time 0.023409 +2023-10-02 20:53:48,572 - Epoch: [52][ 350/ 1236] Overall Loss 0.246870 Objective Loss 0.246870 LR 0.001000 Time 0.023336 +2023-10-02 20:53:48,784 - Epoch: [52][ 360/ 1236] Overall Loss 0.247098 Objective Loss 0.247098 LR 0.001000 Time 0.023276 +2023-10-02 20:53:48,993 - Epoch: [52][ 370/ 1236] Overall Loss 0.247179 Objective Loss 0.247179 LR 0.001000 Time 0.023211 +2023-10-02 20:53:49,204 - Epoch: [52][ 380/ 1236] Overall Loss 0.247454 Objective Loss 0.247454 LR 0.001000 Time 0.023154 +2023-10-02 20:53:49,414 - Epoch: [52][ 390/ 1236] Overall Loss 0.247890 Objective Loss 0.247890 LR 0.001000 Time 0.023098 +2023-10-02 20:53:49,626 - Epoch: [52][ 400/ 1236] Overall Loss 0.247342 Objective Loss 0.247342 LR 0.001000 Time 0.023050 +2023-10-02 20:53:49,835 - Epoch: [52][ 410/ 1236] Overall Loss 0.247379 Objective Loss 0.247379 LR 0.001000 Time 0.022997 +2023-10-02 20:53:50,047 - Epoch: [52][ 420/ 1236] Overall Loss 0.246800 Objective Loss 0.246800 LR 0.001000 Time 0.022954 +2023-10-02 20:53:50,255 - Epoch: [52][ 430/ 1236] Overall Loss 0.246604 Objective Loss 0.246604 LR 0.001000 Time 0.022904 +2023-10-02 20:53:50,466 - Epoch: [52][ 440/ 1236] Overall Loss 0.246558 Objective Loss 0.246558 LR 0.001000 Time 0.022862 +2023-10-02 20:53:50,676 - Epoch: [52][ 450/ 1236] Overall Loss 0.246775 Objective Loss 0.246775 LR 0.001000 Time 0.022817 +2023-10-02 20:53:50,888 - Epoch: [52][ 460/ 1236] Overall Loss 0.247340 Objective Loss 0.247340 LR 0.001000 Time 0.022782 +2023-10-02 20:53:51,097 - Epoch: [52][ 470/ 1236] Overall Loss 0.247113 Objective Loss 0.247113 LR 0.001000 Time 0.022741 +2023-10-02 20:53:51,310 - Epoch: [52][ 480/ 1236] Overall Loss 0.247353 Objective Loss 0.247353 LR 0.001000 Time 0.022709 +2023-10-02 20:53:51,519 - Epoch: [52][ 490/ 1236] Overall Loss 0.247539 Objective Loss 0.247539 LR 0.001000 Time 0.022672 +2023-10-02 20:53:51,729 - Epoch: [52][ 500/ 1236] Overall Loss 0.247231 Objective Loss 0.247231 LR 0.001000 Time 0.022639 +2023-10-02 20:53:51,940 - Epoch: [52][ 510/ 1236] Overall Loss 0.247351 Objective Loss 0.247351 LR 0.001000 Time 0.022607 +2023-10-02 20:53:52,152 - Epoch: [52][ 520/ 1236] Overall Loss 0.247289 Objective Loss 0.247289 LR 0.001000 Time 0.022580 +2023-10-02 20:53:52,361 - Epoch: [52][ 530/ 1236] Overall Loss 0.247488 Objective Loss 0.247488 LR 0.001000 Time 0.022548 +2023-10-02 20:53:52,572 - Epoch: [52][ 540/ 1236] Overall Loss 0.246951 Objective Loss 0.246951 LR 0.001000 Time 0.022520 +2023-10-02 20:53:52,782 - Epoch: [52][ 550/ 1236] Overall Loss 0.246769 Objective Loss 0.246769 LR 0.001000 Time 0.022492 +2023-10-02 20:53:52,993 - Epoch: [52][ 560/ 1236] Overall Loss 0.246501 Objective Loss 0.246501 LR 0.001000 Time 0.022466 +2023-10-02 20:53:53,203 - Epoch: [52][ 570/ 1236] Overall Loss 0.246554 Objective Loss 0.246554 LR 0.001000 Time 0.022440 +2023-10-02 20:53:53,415 - Epoch: [52][ 580/ 1236] Overall Loss 0.246224 Objective Loss 0.246224 LR 0.001000 Time 0.022419 +2023-10-02 20:53:53,624 - Epoch: [52][ 590/ 1236] Overall Loss 0.245988 Objective Loss 0.245988 LR 0.001000 Time 0.022392 +2023-10-02 20:53:53,836 - Epoch: [52][ 600/ 1236] Overall Loss 0.246687 Objective Loss 0.246687 LR 0.001000 Time 0.022373 +2023-10-02 20:53:54,045 - Epoch: [52][ 610/ 1236] Overall Loss 0.246446 Objective Loss 0.246446 LR 0.001000 Time 0.022348 +2023-10-02 20:53:54,258 - Epoch: [52][ 620/ 1236] Overall Loss 0.246567 Objective Loss 0.246567 LR 0.001000 Time 0.022330 +2023-10-02 20:53:54,467 - Epoch: [52][ 630/ 1236] Overall Loss 0.246710 Objective Loss 0.246710 LR 0.001000 Time 0.022307 +2023-10-02 20:53:54,679 - Epoch: [52][ 640/ 1236] Overall Loss 0.247110 Objective Loss 0.247110 LR 0.001000 Time 0.022289 +2023-10-02 20:53:54,889 - Epoch: [52][ 650/ 1236] Overall Loss 0.247414 Objective Loss 0.247414 LR 0.001000 Time 0.022269 +2023-10-02 20:53:55,098 - Epoch: [52][ 660/ 1236] Overall Loss 0.247724 Objective Loss 0.247724 LR 0.001000 Time 0.022248 +2023-10-02 20:53:55,308 - Epoch: [52][ 670/ 1236] Overall Loss 0.247621 Objective Loss 0.247621 LR 0.001000 Time 0.022229 +2023-10-02 20:53:55,519 - Epoch: [52][ 680/ 1236] Overall Loss 0.247765 Objective Loss 0.247765 LR 0.001000 Time 0.022211 +2023-10-02 20:53:55,729 - Epoch: [52][ 690/ 1236] Overall Loss 0.247739 Objective Loss 0.247739 LR 0.001000 Time 0.022193 +2023-10-02 20:53:55,939 - Epoch: [52][ 700/ 1236] Overall Loss 0.247835 Objective Loss 0.247835 LR 0.001000 Time 0.022176 +2023-10-02 20:53:56,149 - Epoch: [52][ 710/ 1236] Overall Loss 0.247880 Objective Loss 0.247880 LR 0.001000 Time 0.022157 +2023-10-02 20:53:56,360 - Epoch: [52][ 720/ 1236] Overall Loss 0.247883 Objective Loss 0.247883 LR 0.001000 Time 0.022142 +2023-10-02 20:53:56,570 - Epoch: [52][ 730/ 1236] Overall Loss 0.248018 Objective Loss 0.248018 LR 0.001000 Time 0.022124 +2023-10-02 20:53:56,780 - Epoch: [52][ 740/ 1236] Overall Loss 0.248258 Objective Loss 0.248258 LR 0.001000 Time 0.022110 +2023-10-02 20:53:56,990 - Epoch: [52][ 750/ 1236] Overall Loss 0.249080 Objective Loss 0.249080 LR 0.001000 Time 0.022093 +2023-10-02 20:53:57,201 - Epoch: [52][ 760/ 1236] Overall Loss 0.248966 Objective Loss 0.248966 LR 0.001000 Time 0.022079 +2023-10-02 20:53:57,411 - Epoch: [52][ 770/ 1236] Overall Loss 0.249014 Objective Loss 0.249014 LR 0.001000 Time 0.022062 +2023-10-02 20:53:57,621 - Epoch: [52][ 780/ 1236] Overall Loss 0.249073 Objective Loss 0.249073 LR 0.001000 Time 0.022049 +2023-10-02 20:53:57,831 - Epoch: [52][ 790/ 1236] Overall Loss 0.249639 Objective Loss 0.249639 LR 0.001000 Time 0.022034 +2023-10-02 20:53:58,042 - Epoch: [52][ 800/ 1236] Overall Loss 0.249695 Objective Loss 0.249695 LR 0.001000 Time 0.022021 +2023-10-02 20:53:58,252 - Epoch: [52][ 810/ 1236] Overall Loss 0.249470 Objective Loss 0.249470 LR 0.001000 Time 0.022007 +2023-10-02 20:53:58,462 - Epoch: [52][ 820/ 1236] Overall Loss 0.249305 Objective Loss 0.249305 LR 0.001000 Time 0.021995 +2023-10-02 20:53:58,672 - Epoch: [52][ 830/ 1236] Overall Loss 0.249647 Objective Loss 0.249647 LR 0.001000 Time 0.021981 +2023-10-02 20:53:58,884 - Epoch: [52][ 840/ 1236] Overall Loss 0.249554 Objective Loss 0.249554 LR 0.001000 Time 0.021972 +2023-10-02 20:53:59,093 - Epoch: [52][ 850/ 1236] Overall Loss 0.249985 Objective Loss 0.249985 LR 0.001000 Time 0.021958 +2023-10-02 20:53:59,304 - Epoch: [52][ 860/ 1236] Overall Loss 0.250447 Objective Loss 0.250447 LR 0.001000 Time 0.021948 +2023-10-02 20:53:59,513 - Epoch: [52][ 870/ 1236] Overall Loss 0.250576 Objective Loss 0.250576 LR 0.001000 Time 0.021934 +2023-10-02 20:53:59,724 - Epoch: [52][ 880/ 1236] Overall Loss 0.250860 Objective Loss 0.250860 LR 0.001000 Time 0.021924 +2023-10-02 20:53:59,934 - Epoch: [52][ 890/ 1236] Overall Loss 0.250941 Objective Loss 0.250941 LR 0.001000 Time 0.021912 +2023-10-02 20:54:00,145 - Epoch: [52][ 900/ 1236] Overall Loss 0.250982 Objective Loss 0.250982 LR 0.001000 Time 0.021903 +2023-10-02 20:54:00,355 - Epoch: [52][ 910/ 1236] Overall Loss 0.251510 Objective Loss 0.251510 LR 0.001000 Time 0.021891 +2023-10-02 20:54:00,566 - Epoch: [52][ 920/ 1236] Overall Loss 0.251770 Objective Loss 0.251770 LR 0.001000 Time 0.021882 +2023-10-02 20:54:00,776 - Epoch: [52][ 930/ 1236] Overall Loss 0.251928 Objective Loss 0.251928 LR 0.001000 Time 0.021871 +2023-10-02 20:54:00,987 - Epoch: [52][ 940/ 1236] Overall Loss 0.252251 Objective Loss 0.252251 LR 0.001000 Time 0.021862 +2023-10-02 20:54:01,198 - Epoch: [52][ 950/ 1236] Overall Loss 0.252542 Objective Loss 0.252542 LR 0.001000 Time 0.021852 +2023-10-02 20:54:01,409 - Epoch: [52][ 960/ 1236] Overall Loss 0.252754 Objective Loss 0.252754 LR 0.001000 Time 0.021845 +2023-10-02 20:54:01,620 - Epoch: [52][ 970/ 1236] Overall Loss 0.252818 Objective Loss 0.252818 LR 0.001000 Time 0.021836 +2023-10-02 20:54:01,832 - Epoch: [52][ 980/ 1236] Overall Loss 0.252672 Objective Loss 0.252672 LR 0.001000 Time 0.021828 +2023-10-02 20:54:02,042 - Epoch: [52][ 990/ 1236] Overall Loss 0.252831 Objective Loss 0.252831 LR 0.001000 Time 0.021819 +2023-10-02 20:54:02,254 - Epoch: [52][ 1000/ 1236] Overall Loss 0.253052 Objective Loss 0.253052 LR 0.001000 Time 0.021812 +2023-10-02 20:54:02,465 - Epoch: [52][ 1010/ 1236] Overall Loss 0.253128 Objective Loss 0.253128 LR 0.001000 Time 0.021803 +2023-10-02 20:54:02,676 - Epoch: [52][ 1020/ 1236] Overall Loss 0.253393 Objective Loss 0.253393 LR 0.001000 Time 0.021797 +2023-10-02 20:54:02,887 - Epoch: [52][ 1030/ 1236] Overall Loss 0.253677 Objective Loss 0.253677 LR 0.001000 Time 0.021788 +2023-10-02 20:54:03,099 - Epoch: [52][ 1040/ 1236] Overall Loss 0.253899 Objective Loss 0.253899 LR 0.001000 Time 0.021782 +2023-10-02 20:54:03,310 - Epoch: [52][ 1050/ 1236] Overall Loss 0.253814 Objective Loss 0.253814 LR 0.001000 Time 0.021774 +2023-10-02 20:54:03,521 - Epoch: [52][ 1060/ 1236] Overall Loss 0.254075 Objective Loss 0.254075 LR 0.001000 Time 0.021768 +2023-10-02 20:54:03,732 - Epoch: [52][ 1070/ 1236] Overall Loss 0.254215 Objective Loss 0.254215 LR 0.001000 Time 0.021760 +2023-10-02 20:54:03,944 - Epoch: [52][ 1080/ 1236] Overall Loss 0.254325 Objective Loss 0.254325 LR 0.001000 Time 0.021754 +2023-10-02 20:54:04,155 - Epoch: [52][ 1090/ 1236] Overall Loss 0.254962 Objective Loss 0.254962 LR 0.001000 Time 0.021747 +2023-10-02 20:54:04,366 - Epoch: [52][ 1100/ 1236] Overall Loss 0.255242 Objective Loss 0.255242 LR 0.001000 Time 0.021741 +2023-10-02 20:54:04,577 - Epoch: [52][ 1110/ 1236] Overall Loss 0.255498 Objective Loss 0.255498 LR 0.001000 Time 0.021734 +2023-10-02 20:54:04,789 - Epoch: [52][ 1120/ 1236] Overall Loss 0.255668 Objective Loss 0.255668 LR 0.001000 Time 0.021729 +2023-10-02 20:54:05,000 - Epoch: [52][ 1130/ 1236] Overall Loss 0.255746 Objective Loss 0.255746 LR 0.001000 Time 0.021721 +2023-10-02 20:54:05,211 - Epoch: [52][ 1140/ 1236] Overall Loss 0.255493 Objective Loss 0.255493 LR 0.001000 Time 0.021716 +2023-10-02 20:54:05,422 - Epoch: [52][ 1150/ 1236] Overall Loss 0.255559 Objective Loss 0.255559 LR 0.001000 Time 0.021710 +2023-10-02 20:54:05,634 - Epoch: [52][ 1160/ 1236] Overall Loss 0.255584 Objective Loss 0.255584 LR 0.001000 Time 0.021705 +2023-10-02 20:54:05,845 - Epoch: [52][ 1170/ 1236] Overall Loss 0.255462 Objective Loss 0.255462 LR 0.001000 Time 0.021698 +2023-10-02 20:54:06,056 - Epoch: [52][ 1180/ 1236] Overall Loss 0.255443 Objective Loss 0.255443 LR 0.001000 Time 0.021693 +2023-10-02 20:54:06,267 - Epoch: [52][ 1190/ 1236] Overall Loss 0.255766 Objective Loss 0.255766 LR 0.001000 Time 0.021687 +2023-10-02 20:54:06,479 - Epoch: [52][ 1200/ 1236] Overall Loss 0.255909 Objective Loss 0.255909 LR 0.001000 Time 0.021682 +2023-10-02 20:54:06,690 - Epoch: [52][ 1210/ 1236] Overall Loss 0.256056 Objective Loss 0.256056 LR 0.001000 Time 0.021676 +2023-10-02 20:54:06,901 - Epoch: [52][ 1220/ 1236] Overall Loss 0.256155 Objective Loss 0.256155 LR 0.001000 Time 0.021672 +2023-10-02 20:54:07,166 - Epoch: [52][ 1230/ 1236] Overall Loss 0.256657 Objective Loss 0.256657 LR 0.001000 Time 0.021710 +2023-10-02 20:54:07,290 - Epoch: [52][ 1236/ 1236] Overall Loss 0.256598 Objective Loss 0.256598 Top1 84.928717 Top5 98.370672 LR 0.001000 Time 0.021704 +2023-10-02 20:54:07,417 - --- validate (epoch=52)----------- +2023-10-02 20:54:07,418 - 29943 samples (256 per mini-batch) +2023-10-02 20:54:07,919 - Epoch: [52][ 10/ 117] Loss 0.287684 Top1 83.671875 Top5 97.890625 +2023-10-02 20:54:08,081 - Epoch: [52][ 20/ 117] Loss 0.302383 Top1 83.652344 Top5 97.753906 +2023-10-02 20:54:08,243 - Epoch: [52][ 30/ 117] Loss 0.308586 Top1 83.776042 Top5 97.786458 +2023-10-02 20:54:08,406 - Epoch: [52][ 40/ 117] Loss 0.314397 Top1 83.896484 Top5 97.783203 +2023-10-02 20:54:08,567 - Epoch: [52][ 50/ 117] Loss 0.313150 Top1 83.859375 Top5 97.890625 +2023-10-02 20:54:08,730 - Epoch: [52][ 60/ 117] Loss 0.315027 Top1 83.730469 Top5 97.936198 +2023-10-02 20:54:08,882 - Epoch: [52][ 70/ 117] Loss 0.314739 Top1 83.716518 Top5 97.901786 +2023-10-02 20:54:09,035 - Epoch: [52][ 80/ 117] Loss 0.318860 Top1 83.608398 Top5 97.949219 +2023-10-02 20:54:09,189 - Epoch: [52][ 90/ 117] Loss 0.316708 Top1 83.628472 Top5 97.894965 +2023-10-02 20:54:09,343 - Epoch: [52][ 100/ 117] Loss 0.316896 Top1 83.527344 Top5 97.898438 +2023-10-02 20:54:09,503 - Epoch: [52][ 110/ 117] Loss 0.317198 Top1 83.689631 Top5 97.926136 +2023-10-02 20:54:09,593 - Epoch: [52][ 117/ 117] Loss 0.316494 Top1 83.689009 Top5 97.926060 +2023-10-02 20:54:09,717 - ==> Top1: 83.689 Top5: 97.926 Loss: 0.316 + +2023-10-02 20:54:09,717 - ==> Confusion: +[[ 931 0 8 3 6 2 0 0 7 64 2 0 1 3 7 0 1 1 1 0 13] + [ 0 1054 2 1 5 21 2 19 4 2 3 0 0 0 2 4 1 0 6 1 4] + [ 3 0 981 13 2 0 17 5 0 0 1 0 7 3 2 8 1 1 6 4 2] + [ 0 2 10 1004 2 2 1 4 4 0 1 0 2 2 29 1 2 1 12 0 10] + [ 22 9 2 1 962 5 0 2 0 15 2 0 1 4 8 3 7 0 2 3 2] + [ 2 37 1 4 2 970 0 20 3 9 6 12 4 13 4 0 2 0 7 3 17] + [ 0 0 42 0 0 1 1114 5 0 0 7 2 2 1 0 4 0 2 2 5 4] + [ 1 19 40 2 2 27 4 1032 0 4 12 1 3 9 2 1 1 0 50 3 5] + [ 16 4 1 1 1 1 0 0 963 48 20 1 1 13 10 1 4 3 1 0 0] + [ 98 1 3 0 7 3 1 0 28 934 2 1 1 24 7 0 1 1 0 2 5] + [ 3 2 9 9 0 0 1 1 9 1 985 3 1 10 5 0 0 2 4 3 5] + [ 1 1 5 0 2 11 1 3 0 0 0 946 23 16 0 1 2 18 0 4 1] + [ 0 1 7 4 0 0 2 2 1 0 4 34 956 8 2 6 2 26 6 2 5] + [ 1 0 3 0 2 5 0 0 12 10 7 2 0 1061 7 0 0 1 0 0 8] + [ 11 2 5 15 2 0 0 0 18 1 4 0 3 3 1020 0 0 1 9 0 7] + [ 0 0 5 1 5 0 8 0 0 0 0 5 10 2 1 1054 17 15 3 4 4] + [ 1 18 2 0 7 5 0 1 1 0 1 6 4 3 4 7 1090 0 2 2 7] + [ 0 0 1 1 0 1 3 0 1 1 0 6 13 2 5 6 2 995 0 1 0] + [ 1 4 9 15 0 0 1 13 5 0 9 1 0 0 15 0 1 0 985 0 9] + [ 0 4 7 1 0 4 11 13 0 0 2 15 6 4 1 5 9 2 0 1066 2] + [ 140 168 208 122 104 106 55 83 120 83 209 111 353 294 156 65 102 75 196 199 4956]] + +2023-10-02 20:54:09,719 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:54:09,719 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:54:09,725 - + +2023-10-02 20:54:09,725 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:54:10,867 - Epoch: [53][ 10/ 1236] Overall Loss 0.253399 Objective Loss 0.253399 LR 0.001000 Time 0.114150 +2023-10-02 20:54:11,079 - Epoch: [53][ 20/ 1236] Overall Loss 0.256220 Objective Loss 0.256220 LR 0.001000 Time 0.067653 +2023-10-02 20:54:11,292 - Epoch: [53][ 30/ 1236] Overall Loss 0.254979 Objective Loss 0.254979 LR 0.001000 Time 0.052190 +2023-10-02 20:54:11,504 - Epoch: [53][ 40/ 1236] Overall Loss 0.258291 Objective Loss 0.258291 LR 0.001000 Time 0.044436 +2023-10-02 20:54:11,717 - Epoch: [53][ 50/ 1236] Overall Loss 0.259640 Objective Loss 0.259640 LR 0.001000 Time 0.039765 +2023-10-02 20:54:11,928 - Epoch: [53][ 60/ 1236] Overall Loss 0.257642 Objective Loss 0.257642 LR 0.001000 Time 0.036661 +2023-10-02 20:54:12,140 - Epoch: [53][ 70/ 1236] Overall Loss 0.257867 Objective Loss 0.257867 LR 0.001000 Time 0.034433 +2023-10-02 20:54:12,351 - Epoch: [53][ 80/ 1236] Overall Loss 0.258143 Objective Loss 0.258143 LR 0.001000 Time 0.032762 +2023-10-02 20:54:12,560 - Epoch: [53][ 90/ 1236] Overall Loss 0.254702 Objective Loss 0.254702 LR 0.001000 Time 0.031423 +2023-10-02 20:54:12,769 - Epoch: [53][ 100/ 1236] Overall Loss 0.254450 Objective Loss 0.254450 LR 0.001000 Time 0.030367 +2023-10-02 20:54:12,980 - Epoch: [53][ 110/ 1236] Overall Loss 0.253973 Objective Loss 0.253973 LR 0.001000 Time 0.029508 +2023-10-02 20:54:13,195 - Epoch: [53][ 120/ 1236] Overall Loss 0.253649 Objective Loss 0.253649 LR 0.001000 Time 0.028827 +2023-10-02 20:54:13,409 - Epoch: [53][ 130/ 1236] Overall Loss 0.252805 Objective Loss 0.252805 LR 0.001000 Time 0.028248 +2023-10-02 20:54:13,625 - Epoch: [53][ 140/ 1236] Overall Loss 0.250932 Objective Loss 0.250932 LR 0.001000 Time 0.027760 +2023-10-02 20:54:13,839 - Epoch: [53][ 150/ 1236] Overall Loss 0.252192 Objective Loss 0.252192 LR 0.001000 Time 0.027325 +2023-10-02 20:54:14,055 - Epoch: [53][ 160/ 1236] Overall Loss 0.252088 Objective Loss 0.252088 LR 0.001000 Time 0.026963 +2023-10-02 20:54:14,269 - Epoch: [53][ 170/ 1236] Overall Loss 0.250383 Objective Loss 0.250383 LR 0.001000 Time 0.026631 +2023-10-02 20:54:14,485 - Epoch: [53][ 180/ 1236] Overall Loss 0.249334 Objective Loss 0.249334 LR 0.001000 Time 0.026350 +2023-10-02 20:54:14,699 - Epoch: [53][ 190/ 1236] Overall Loss 0.249128 Objective Loss 0.249128 LR 0.001000 Time 0.026087 +2023-10-02 20:54:14,915 - Epoch: [53][ 200/ 1236] Overall Loss 0.249115 Objective Loss 0.249115 LR 0.001000 Time 0.025861 +2023-10-02 20:54:15,129 - Epoch: [53][ 210/ 1236] Overall Loss 0.249012 Objective Loss 0.249012 LR 0.001000 Time 0.025646 +2023-10-02 20:54:15,345 - Epoch: [53][ 220/ 1236] Overall Loss 0.248920 Objective Loss 0.248920 LR 0.001000 Time 0.025460 +2023-10-02 20:54:15,559 - Epoch: [53][ 230/ 1236] Overall Loss 0.247683 Objective Loss 0.247683 LR 0.001000 Time 0.025282 +2023-10-02 20:54:15,775 - Epoch: [53][ 240/ 1236] Overall Loss 0.248027 Objective Loss 0.248027 LR 0.001000 Time 0.025127 +2023-10-02 20:54:15,989 - Epoch: [53][ 250/ 1236] Overall Loss 0.248587 Objective Loss 0.248587 LR 0.001000 Time 0.024975 +2023-10-02 20:54:16,205 - Epoch: [53][ 260/ 1236] Overall Loss 0.249051 Objective Loss 0.249051 LR 0.001000 Time 0.024843 +2023-10-02 20:54:16,419 - Epoch: [53][ 270/ 1236] Overall Loss 0.248850 Objective Loss 0.248850 LR 0.001000 Time 0.024715 +2023-10-02 20:54:16,635 - Epoch: [53][ 280/ 1236] Overall Loss 0.248657 Objective Loss 0.248657 LR 0.001000 Time 0.024601 +2023-10-02 20:54:16,849 - Epoch: [53][ 290/ 1236] Overall Loss 0.249018 Objective Loss 0.249018 LR 0.001000 Time 0.024489 +2023-10-02 20:54:17,064 - Epoch: [53][ 300/ 1236] Overall Loss 0.248265 Objective Loss 0.248265 LR 0.001000 Time 0.024389 +2023-10-02 20:54:17,275 - Epoch: [53][ 310/ 1236] Overall Loss 0.248569 Objective Loss 0.248569 LR 0.001000 Time 0.024281 +2023-10-02 20:54:17,486 - Epoch: [53][ 320/ 1236] Overall Loss 0.248976 Objective Loss 0.248976 LR 0.001000 Time 0.024181 +2023-10-02 20:54:17,697 - Epoch: [53][ 330/ 1236] Overall Loss 0.248979 Objective Loss 0.248979 LR 0.001000 Time 0.024085 +2023-10-02 20:54:17,908 - Epoch: [53][ 340/ 1236] Overall Loss 0.248369 Objective Loss 0.248369 LR 0.001000 Time 0.023997 +2023-10-02 20:54:18,119 - Epoch: [53][ 350/ 1236] Overall Loss 0.247574 Objective Loss 0.247574 LR 0.001000 Time 0.023912 +2023-10-02 20:54:18,330 - Epoch: [53][ 360/ 1236] Overall Loss 0.248173 Objective Loss 0.248173 LR 0.001000 Time 0.023834 +2023-10-02 20:54:18,541 - Epoch: [53][ 370/ 1236] Overall Loss 0.249342 Objective Loss 0.249342 LR 0.001000 Time 0.023758 +2023-10-02 20:54:18,752 - Epoch: [53][ 380/ 1236] Overall Loss 0.249762 Objective Loss 0.249762 LR 0.001000 Time 0.023688 +2023-10-02 20:54:18,963 - Epoch: [53][ 390/ 1236] Overall Loss 0.250040 Objective Loss 0.250040 LR 0.001000 Time 0.023620 +2023-10-02 20:54:19,174 - Epoch: [53][ 400/ 1236] Overall Loss 0.250230 Objective Loss 0.250230 LR 0.001000 Time 0.023556 +2023-10-02 20:54:19,385 - Epoch: [53][ 410/ 1236] Overall Loss 0.250063 Objective Loss 0.250063 LR 0.001000 Time 0.023494 +2023-10-02 20:54:19,596 - Epoch: [53][ 420/ 1236] Overall Loss 0.250895 Objective Loss 0.250895 LR 0.001000 Time 0.023437 +2023-10-02 20:54:19,808 - Epoch: [53][ 430/ 1236] Overall Loss 0.251119 Objective Loss 0.251119 LR 0.001000 Time 0.023384 +2023-10-02 20:54:20,018 - Epoch: [53][ 440/ 1236] Overall Loss 0.251578 Objective Loss 0.251578 LR 0.001000 Time 0.023330 +2023-10-02 20:54:20,229 - Epoch: [53][ 450/ 1236] Overall Loss 0.252090 Objective Loss 0.252090 LR 0.001000 Time 0.023278 +2023-10-02 20:54:20,440 - Epoch: [53][ 460/ 1236] Overall Loss 0.252238 Objective Loss 0.252238 LR 0.001000 Time 0.023230 +2023-10-02 20:54:20,651 - Epoch: [53][ 470/ 1236] Overall Loss 0.251949 Objective Loss 0.251949 LR 0.001000 Time 0.023183 +2023-10-02 20:54:20,861 - Epoch: [53][ 480/ 1236] Overall Loss 0.252301 Objective Loss 0.252301 LR 0.001000 Time 0.023137 +2023-10-02 20:54:21,071 - Epoch: [53][ 490/ 1236] Overall Loss 0.252388 Objective Loss 0.252388 LR 0.001000 Time 0.023094 +2023-10-02 20:54:21,283 - Epoch: [53][ 500/ 1236] Overall Loss 0.252689 Objective Loss 0.252689 LR 0.001000 Time 0.023054 +2023-10-02 20:54:21,493 - Epoch: [53][ 510/ 1236] Overall Loss 0.253060 Objective Loss 0.253060 LR 0.001000 Time 0.023014 +2023-10-02 20:54:21,703 - Epoch: [53][ 520/ 1236] Overall Loss 0.252821 Objective Loss 0.252821 LR 0.001000 Time 0.022974 +2023-10-02 20:54:21,914 - Epoch: [53][ 530/ 1236] Overall Loss 0.252463 Objective Loss 0.252463 LR 0.001000 Time 0.022937 +2023-10-02 20:54:22,125 - Epoch: [53][ 540/ 1236] Overall Loss 0.252309 Objective Loss 0.252309 LR 0.001000 Time 0.022903 +2023-10-02 20:54:22,336 - Epoch: [53][ 550/ 1236] Overall Loss 0.252319 Objective Loss 0.252319 LR 0.001000 Time 0.022869 +2023-10-02 20:54:22,547 - Epoch: [53][ 560/ 1236] Overall Loss 0.252455 Objective Loss 0.252455 LR 0.001000 Time 0.022837 +2023-10-02 20:54:22,757 - Epoch: [53][ 570/ 1236] Overall Loss 0.252474 Objective Loss 0.252474 LR 0.001000 Time 0.022805 +2023-10-02 20:54:22,969 - Epoch: [53][ 580/ 1236] Overall Loss 0.252226 Objective Loss 0.252226 LR 0.001000 Time 0.022775 +2023-10-02 20:54:23,179 - Epoch: [53][ 590/ 1236] Overall Loss 0.252208 Objective Loss 0.252208 LR 0.001000 Time 0.022746 +2023-10-02 20:54:23,390 - Epoch: [53][ 600/ 1236] Overall Loss 0.252248 Objective Loss 0.252248 LR 0.001000 Time 0.022718 +2023-10-02 20:54:23,600 - Epoch: [53][ 610/ 1236] Overall Loss 0.252049 Objective Loss 0.252049 LR 0.001000 Time 0.022688 +2023-10-02 20:54:23,811 - Epoch: [53][ 620/ 1236] Overall Loss 0.252054 Objective Loss 0.252054 LR 0.001000 Time 0.022662 +2023-10-02 20:54:24,021 - Epoch: [53][ 630/ 1236] Overall Loss 0.252044 Objective Loss 0.252044 LR 0.001000 Time 0.022635 +2023-10-02 20:54:24,232 - Epoch: [53][ 640/ 1236] Overall Loss 0.252135 Objective Loss 0.252135 LR 0.001000 Time 0.022611 +2023-10-02 20:54:24,443 - Epoch: [53][ 650/ 1236] Overall Loss 0.251964 Objective Loss 0.251964 LR 0.001000 Time 0.022587 +2023-10-02 20:54:24,654 - Epoch: [53][ 660/ 1236] Overall Loss 0.252022 Objective Loss 0.252022 LR 0.001000 Time 0.022564 +2023-10-02 20:54:24,865 - Epoch: [53][ 670/ 1236] Overall Loss 0.252339 Objective Loss 0.252339 LR 0.001000 Time 0.022541 +2023-10-02 20:54:25,076 - Epoch: [53][ 680/ 1236] Overall Loss 0.252327 Objective Loss 0.252327 LR 0.001000 Time 0.022519 +2023-10-02 20:54:25,287 - Epoch: [53][ 690/ 1236] Overall Loss 0.252206 Objective Loss 0.252206 LR 0.001000 Time 0.022497 +2023-10-02 20:54:25,498 - Epoch: [53][ 700/ 1236] Overall Loss 0.252534 Objective Loss 0.252534 LR 0.001000 Time 0.022477 +2023-10-02 20:54:25,708 - Epoch: [53][ 710/ 1236] Overall Loss 0.252481 Objective Loss 0.252481 LR 0.001000 Time 0.022456 +2023-10-02 20:54:25,922 - Epoch: [53][ 720/ 1236] Overall Loss 0.253040 Objective Loss 0.253040 LR 0.001000 Time 0.022441 +2023-10-02 20:54:26,131 - Epoch: [53][ 730/ 1236] Overall Loss 0.252934 Objective Loss 0.252934 LR 0.001000 Time 0.022420 +2023-10-02 20:54:26,341 - Epoch: [53][ 740/ 1236] Overall Loss 0.252439 Objective Loss 0.252439 LR 0.001000 Time 0.022400 +2023-10-02 20:54:26,551 - Epoch: [53][ 750/ 1236] Overall Loss 0.252835 Objective Loss 0.252835 LR 0.001000 Time 0.022381 +2023-10-02 20:54:26,762 - Epoch: [53][ 760/ 1236] Overall Loss 0.252862 Objective Loss 0.252862 LR 0.001000 Time 0.022363 +2023-10-02 20:54:26,971 - Epoch: [53][ 770/ 1236] Overall Loss 0.252783 Objective Loss 0.252783 LR 0.001000 Time 0.022344 +2023-10-02 20:54:27,182 - Epoch: [53][ 780/ 1236] Overall Loss 0.252776 Objective Loss 0.252776 LR 0.001000 Time 0.022327 +2023-10-02 20:54:27,391 - Epoch: [53][ 790/ 1236] Overall Loss 0.252563 Objective Loss 0.252563 LR 0.001000 Time 0.022309 +2023-10-02 20:54:27,602 - Epoch: [53][ 800/ 1236] Overall Loss 0.252631 Objective Loss 0.252631 LR 0.001000 Time 0.022293 +2023-10-02 20:54:27,811 - Epoch: [53][ 810/ 1236] Overall Loss 0.253113 Objective Loss 0.253113 LR 0.001000 Time 0.022276 +2023-10-02 20:54:28,022 - Epoch: [53][ 820/ 1236] Overall Loss 0.253102 Objective Loss 0.253102 LR 0.001000 Time 0.022261 +2023-10-02 20:54:28,231 - Epoch: [53][ 830/ 1236] Overall Loss 0.253082 Objective Loss 0.253082 LR 0.001000 Time 0.022244 +2023-10-02 20:54:28,442 - Epoch: [53][ 840/ 1236] Overall Loss 0.253307 Objective Loss 0.253307 LR 0.001000 Time 0.022230 +2023-10-02 20:54:28,651 - Epoch: [53][ 850/ 1236] Overall Loss 0.252923 Objective Loss 0.252923 LR 0.001000 Time 0.022214 +2023-10-02 20:54:28,862 - Epoch: [53][ 860/ 1236] Overall Loss 0.252948 Objective Loss 0.252948 LR 0.001000 Time 0.022201 +2023-10-02 20:54:29,073 - Epoch: [53][ 870/ 1236] Overall Loss 0.253108 Objective Loss 0.253108 LR 0.001000 Time 0.022187 +2023-10-02 20:54:29,285 - Epoch: [53][ 880/ 1236] Overall Loss 0.252870 Objective Loss 0.252870 LR 0.001000 Time 0.022176 +2023-10-02 20:54:29,497 - Epoch: [53][ 890/ 1236] Overall Loss 0.253029 Objective Loss 0.253029 LR 0.001000 Time 0.022164 +2023-10-02 20:54:29,710 - Epoch: [53][ 900/ 1236] Overall Loss 0.252861 Objective Loss 0.252861 LR 0.001000 Time 0.022155 +2023-10-02 20:54:29,922 - Epoch: [53][ 910/ 1236] Overall Loss 0.252741 Objective Loss 0.252741 LR 0.001000 Time 0.022143 +2023-10-02 20:54:30,133 - Epoch: [53][ 920/ 1236] Overall Loss 0.252696 Objective Loss 0.252696 LR 0.001000 Time 0.022132 +2023-10-02 20:54:30,345 - Epoch: [53][ 930/ 1236] Overall Loss 0.252550 Objective Loss 0.252550 LR 0.001000 Time 0.022121 +2023-10-02 20:54:30,558 - Epoch: [53][ 940/ 1236] Overall Loss 0.252793 Objective Loss 0.252793 LR 0.001000 Time 0.022112 +2023-10-02 20:54:30,769 - Epoch: [53][ 950/ 1236] Overall Loss 0.252820 Objective Loss 0.252820 LR 0.001000 Time 0.022102 +2023-10-02 20:54:30,982 - Epoch: [53][ 960/ 1236] Overall Loss 0.252864 Objective Loss 0.252864 LR 0.001000 Time 0.022093 +2023-10-02 20:54:31,194 - Epoch: [53][ 970/ 1236] Overall Loss 0.252875 Objective Loss 0.252875 LR 0.001000 Time 0.022083 +2023-10-02 20:54:31,407 - Epoch: [53][ 980/ 1236] Overall Loss 0.252941 Objective Loss 0.252941 LR 0.001000 Time 0.022075 +2023-10-02 20:54:31,619 - Epoch: [53][ 990/ 1236] Overall Loss 0.253126 Objective Loss 0.253126 LR 0.001000 Time 0.022065 +2023-10-02 20:54:31,832 - Epoch: [53][ 1000/ 1236] Overall Loss 0.253221 Objective Loss 0.253221 LR 0.001000 Time 0.022057 +2023-10-02 20:54:32,043 - Epoch: [53][ 1010/ 1236] Overall Loss 0.253690 Objective Loss 0.253690 LR 0.001000 Time 0.022048 +2023-10-02 20:54:32,256 - Epoch: [53][ 1020/ 1236] Overall Loss 0.253698 Objective Loss 0.253698 LR 0.001000 Time 0.022040 +2023-10-02 20:54:32,468 - Epoch: [53][ 1030/ 1236] Overall Loss 0.253836 Objective Loss 0.253836 LR 0.001000 Time 0.022031 +2023-10-02 20:54:32,681 - Epoch: [53][ 1040/ 1236] Overall Loss 0.253743 Objective Loss 0.253743 LR 0.001000 Time 0.022024 +2023-10-02 20:54:32,892 - Epoch: [53][ 1050/ 1236] Overall Loss 0.253735 Objective Loss 0.253735 LR 0.001000 Time 0.022015 +2023-10-02 20:54:33,105 - Epoch: [53][ 1060/ 1236] Overall Loss 0.254269 Objective Loss 0.254269 LR 0.001000 Time 0.022008 +2023-10-02 20:54:33,317 - Epoch: [53][ 1070/ 1236] Overall Loss 0.254324 Objective Loss 0.254324 LR 0.001000 Time 0.022000 +2023-10-02 20:54:33,530 - Epoch: [53][ 1080/ 1236] Overall Loss 0.254421 Objective Loss 0.254421 LR 0.001000 Time 0.021993 +2023-10-02 20:54:33,741 - Epoch: [53][ 1090/ 1236] Overall Loss 0.254394 Objective Loss 0.254394 LR 0.001000 Time 0.021985 +2023-10-02 20:54:33,954 - Epoch: [53][ 1100/ 1236] Overall Loss 0.254093 Objective Loss 0.254093 LR 0.001000 Time 0.021978 +2023-10-02 20:54:34,165 - Epoch: [53][ 1110/ 1236] Overall Loss 0.254094 Objective Loss 0.254094 LR 0.001000 Time 0.021971 +2023-10-02 20:54:34,378 - Epoch: [53][ 1120/ 1236] Overall Loss 0.254444 Objective Loss 0.254444 LR 0.001000 Time 0.021964 +2023-10-02 20:54:34,590 - Epoch: [53][ 1130/ 1236] Overall Loss 0.254442 Objective Loss 0.254442 LR 0.001000 Time 0.021957 +2023-10-02 20:54:34,803 - Epoch: [53][ 1140/ 1236] Overall Loss 0.254384 Objective Loss 0.254384 LR 0.001000 Time 0.021951 +2023-10-02 20:54:35,015 - Epoch: [53][ 1150/ 1236] Overall Loss 0.254526 Objective Loss 0.254526 LR 0.001000 Time 0.021944 +2023-10-02 20:54:35,227 - Epoch: [53][ 1160/ 1236] Overall Loss 0.254524 Objective Loss 0.254524 LR 0.001000 Time 0.021938 +2023-10-02 20:54:35,439 - Epoch: [53][ 1170/ 1236] Overall Loss 0.254605 Objective Loss 0.254605 LR 0.001000 Time 0.021931 +2023-10-02 20:54:35,652 - Epoch: [53][ 1180/ 1236] Overall Loss 0.254658 Objective Loss 0.254658 LR 0.001000 Time 0.021925 +2023-10-02 20:54:35,863 - Epoch: [53][ 1190/ 1236] Overall Loss 0.254942 Objective Loss 0.254942 LR 0.001000 Time 0.021918 +2023-10-02 20:54:36,076 - Epoch: [53][ 1200/ 1236] Overall Loss 0.255048 Objective Loss 0.255048 LR 0.001000 Time 0.021913 +2023-10-02 20:54:36,288 - Epoch: [53][ 1210/ 1236] Overall Loss 0.254931 Objective Loss 0.254931 LR 0.001000 Time 0.021906 +2023-10-02 20:54:36,501 - Epoch: [53][ 1220/ 1236] Overall Loss 0.255024 Objective Loss 0.255024 LR 0.001000 Time 0.021901 +2023-10-02 20:54:36,766 - Epoch: [53][ 1230/ 1236] Overall Loss 0.255219 Objective Loss 0.255219 LR 0.001000 Time 0.021938 +2023-10-02 20:54:36,889 - Epoch: [53][ 1236/ 1236] Overall Loss 0.255138 Objective Loss 0.255138 Top1 84.317719 Top5 98.370672 LR 0.001000 Time 0.021931 +2023-10-02 20:54:37,016 - --- validate (epoch=53)----------- +2023-10-02 20:54:37,016 - 29943 samples (256 per mini-batch) +2023-10-02 20:54:37,517 - Epoch: [53][ 10/ 117] Loss 0.375751 Top1 82.500000 Top5 97.734375 +2023-10-02 20:54:37,670 - Epoch: [53][ 20/ 117] Loss 0.330027 Top1 83.945312 Top5 98.183594 +2023-10-02 20:54:37,822 - Epoch: [53][ 30/ 117] Loss 0.336670 Top1 84.088542 Top5 98.255208 +2023-10-02 20:54:37,975 - Epoch: [53][ 40/ 117] Loss 0.331810 Top1 84.326172 Top5 98.183594 +2023-10-02 20:54:38,127 - Epoch: [53][ 50/ 117] Loss 0.325350 Top1 84.484375 Top5 98.296875 +2023-10-02 20:54:38,279 - Epoch: [53][ 60/ 117] Loss 0.323618 Top1 84.329427 Top5 98.333333 +2023-10-02 20:54:38,433 - Epoch: [53][ 70/ 117] Loss 0.328328 Top1 84.101562 Top5 98.275670 +2023-10-02 20:54:38,585 - Epoch: [53][ 80/ 117] Loss 0.330585 Top1 83.994141 Top5 98.325195 +2023-10-02 20:54:38,737 - Epoch: [53][ 90/ 117] Loss 0.330480 Top1 84.084201 Top5 98.346354 +2023-10-02 20:54:38,889 - Epoch: [53][ 100/ 117] Loss 0.328450 Top1 84.109375 Top5 98.339844 +2023-10-02 20:54:39,049 - Epoch: [53][ 110/ 117] Loss 0.327330 Top1 84.080256 Top5 98.327415 +2023-10-02 20:54:39,138 - Epoch: [53][ 117/ 117] Loss 0.328318 Top1 84.059713 Top5 98.306783 +2023-10-02 20:54:39,274 - ==> Top1: 84.060 Top5: 98.307 Loss: 0.328 + +2023-10-02 20:54:39,275 - ==> Confusion: +[[ 937 2 2 0 13 7 0 0 8 53 2 0 2 1 6 2 5 1 0 0 9] + [ 1 1035 2 0 3 44 2 21 1 1 2 1 2 1 0 5 0 0 5 0 5] + [ 4 0 946 9 1 0 30 15 1 1 2 1 10 5 1 7 0 2 12 3 6] + [ 0 3 18 954 0 7 2 1 8 0 5 0 17 4 24 1 1 3 23 0 18] + [ 21 5 1 0 954 11 0 0 0 14 1 0 1 5 7 3 10 1 1 3 12] + [ 3 19 1 2 3 1010 1 17 0 7 1 6 4 16 4 0 3 0 2 1 16] + [ 1 7 21 0 0 1 1126 7 1 0 5 1 1 0 0 7 0 1 0 5 7] + [ 1 16 14 1 5 37 9 1051 2 1 6 6 5 3 2 1 0 0 34 11 13] + [ 18 3 2 3 0 2 0 0 976 37 9 2 5 7 14 1 4 1 2 1 2] + [ 88 0 0 0 2 2 1 0 32 945 2 0 1 24 5 0 1 0 0 5 11] + [ 2 2 13 9 1 2 2 5 24 3 937 3 0 25 3 0 0 1 7 2 12] + [ 0 0 3 0 0 9 0 4 0 1 1 917 53 6 0 6 5 18 0 6 6] + [ 0 0 2 4 0 0 3 1 1 0 2 33 981 9 2 5 3 11 4 3 4] + [ 2 1 1 0 1 8 0 0 8 9 2 7 2 1052 6 0 1 2 0 1 16] + [ 11 0 2 7 3 1 0 0 27 4 1 0 5 4 1014 0 2 4 4 0 12] + [ 0 0 3 2 3 1 2 0 0 0 0 7 10 0 0 1059 20 12 0 10 5] + [ 1 15 2 0 4 7 1 1 1 0 0 5 4 3 3 5 1089 0 0 7 13] + [ 1 0 1 1 0 0 1 0 0 1 0 4 31 1 1 5 0 988 0 0 3] + [ 3 5 2 14 0 0 0 35 6 0 3 1 2 0 16 0 0 0 965 1 15] + [ 0 6 4 1 1 9 7 14 0 0 2 16 7 2 1 3 10 1 1 1055 12] + [ 143 154 109 64 74 246 32 124 106 69 114 138 358 271 124 61 109 73 168 189 5179]] + +2023-10-02 20:54:39,276 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:54:39,276 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:54:39,282 - + +2023-10-02 20:54:39,282 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:54:40,320 - Epoch: [54][ 10/ 1236] Overall Loss 0.269036 Objective Loss 0.269036 LR 0.001000 Time 0.103697 +2023-10-02 20:54:40,533 - Epoch: [54][ 20/ 1236] Overall Loss 0.250837 Objective Loss 0.250837 LR 0.001000 Time 0.062482 +2023-10-02 20:54:40,744 - Epoch: [54][ 30/ 1236] Overall Loss 0.257284 Objective Loss 0.257284 LR 0.001000 Time 0.048631 +2023-10-02 20:54:40,957 - Epoch: [54][ 40/ 1236] Overall Loss 0.253388 Objective Loss 0.253388 LR 0.001000 Time 0.041791 +2023-10-02 20:54:41,165 - Epoch: [54][ 50/ 1236] Overall Loss 0.260349 Objective Loss 0.260349 LR 0.001000 Time 0.037601 +2023-10-02 20:54:41,377 - Epoch: [54][ 60/ 1236] Overall Loss 0.260967 Objective Loss 0.260967 LR 0.001000 Time 0.034859 +2023-10-02 20:54:41,587 - Epoch: [54][ 70/ 1236] Overall Loss 0.259169 Objective Loss 0.259169 LR 0.001000 Time 0.032856 +2023-10-02 20:54:41,799 - Epoch: [54][ 80/ 1236] Overall Loss 0.258917 Objective Loss 0.258917 LR 0.001000 Time 0.031394 +2023-10-02 20:54:42,009 - Epoch: [54][ 90/ 1236] Overall Loss 0.255892 Objective Loss 0.255892 LR 0.001000 Time 0.030223 +2023-10-02 20:54:42,219 - Epoch: [54][ 100/ 1236] Overall Loss 0.255738 Objective Loss 0.255738 LR 0.001000 Time 0.029299 +2023-10-02 20:54:42,428 - Epoch: [54][ 110/ 1236] Overall Loss 0.257442 Objective Loss 0.257442 LR 0.001000 Time 0.028516 +2023-10-02 20:54:42,638 - Epoch: [54][ 120/ 1236] Overall Loss 0.254759 Objective Loss 0.254759 LR 0.001000 Time 0.027889 +2023-10-02 20:54:42,846 - Epoch: [54][ 130/ 1236] Overall Loss 0.253912 Objective Loss 0.253912 LR 0.001000 Time 0.027334 +2023-10-02 20:54:43,057 - Epoch: [54][ 140/ 1236] Overall Loss 0.252369 Objective Loss 0.252369 LR 0.001000 Time 0.026889 +2023-10-02 20:54:43,265 - Epoch: [54][ 150/ 1236] Overall Loss 0.253543 Objective Loss 0.253543 LR 0.001000 Time 0.026476 +2023-10-02 20:54:43,476 - Epoch: [54][ 160/ 1236] Overall Loss 0.252163 Objective Loss 0.252163 LR 0.001000 Time 0.026141 +2023-10-02 20:54:43,683 - Epoch: [54][ 170/ 1236] Overall Loss 0.251862 Objective Loss 0.251862 LR 0.001000 Time 0.025818 +2023-10-02 20:54:43,893 - Epoch: [54][ 180/ 1236] Overall Loss 0.252150 Objective Loss 0.252150 LR 0.001000 Time 0.025551 +2023-10-02 20:54:44,101 - Epoch: [54][ 190/ 1236] Overall Loss 0.251469 Objective Loss 0.251469 LR 0.001000 Time 0.025293 +2023-10-02 20:54:44,313 - Epoch: [54][ 200/ 1236] Overall Loss 0.251409 Objective Loss 0.251409 LR 0.001000 Time 0.025085 +2023-10-02 20:54:44,520 - Epoch: [54][ 210/ 1236] Overall Loss 0.251098 Objective Loss 0.251098 LR 0.001000 Time 0.024874 +2023-10-02 20:54:44,730 - Epoch: [54][ 220/ 1236] Overall Loss 0.249876 Objective Loss 0.249876 LR 0.001000 Time 0.024698 +2023-10-02 20:54:44,938 - Epoch: [54][ 230/ 1236] Overall Loss 0.248144 Objective Loss 0.248144 LR 0.001000 Time 0.024522 +2023-10-02 20:54:45,148 - Epoch: [54][ 240/ 1236] Overall Loss 0.248366 Objective Loss 0.248366 LR 0.001000 Time 0.024375 +2023-10-02 20:54:45,356 - Epoch: [54][ 250/ 1236] Overall Loss 0.249232 Objective Loss 0.249232 LR 0.001000 Time 0.024226 +2023-10-02 20:54:45,567 - Epoch: [54][ 260/ 1236] Overall Loss 0.248957 Objective Loss 0.248957 LR 0.001000 Time 0.024102 +2023-10-02 20:54:45,775 - Epoch: [54][ 270/ 1236] Overall Loss 0.248972 Objective Loss 0.248972 LR 0.001000 Time 0.023974 +2023-10-02 20:54:45,985 - Epoch: [54][ 280/ 1236] Overall Loss 0.249419 Objective Loss 0.249419 LR 0.001000 Time 0.023868 +2023-10-02 20:54:46,193 - Epoch: [54][ 290/ 1236] Overall Loss 0.249810 Objective Loss 0.249810 LR 0.001000 Time 0.023759 +2023-10-02 20:54:46,405 - Epoch: [54][ 300/ 1236] Overall Loss 0.249954 Objective Loss 0.249954 LR 0.001000 Time 0.023672 +2023-10-02 20:54:46,615 - Epoch: [54][ 310/ 1236] Overall Loss 0.249577 Objective Loss 0.249577 LR 0.001000 Time 0.023581 +2023-10-02 20:54:46,827 - Epoch: [54][ 320/ 1236] Overall Loss 0.249247 Objective Loss 0.249247 LR 0.001000 Time 0.023506 +2023-10-02 20:54:47,038 - Epoch: [54][ 330/ 1236] Overall Loss 0.248963 Objective Loss 0.248963 LR 0.001000 Time 0.023427 +2023-10-02 20:54:47,251 - Epoch: [54][ 340/ 1236] Overall Loss 0.249048 Objective Loss 0.249048 LR 0.001000 Time 0.023363 +2023-10-02 20:54:47,462 - Epoch: [54][ 350/ 1236] Overall Loss 0.249157 Objective Loss 0.249157 LR 0.001000 Time 0.023294 +2023-10-02 20:54:47,674 - Epoch: [54][ 360/ 1236] Overall Loss 0.249048 Objective Loss 0.249048 LR 0.001000 Time 0.023236 +2023-10-02 20:54:47,885 - Epoch: [54][ 370/ 1236] Overall Loss 0.248392 Objective Loss 0.248392 LR 0.001000 Time 0.023174 +2023-10-02 20:54:48,098 - Epoch: [54][ 380/ 1236] Overall Loss 0.248290 Objective Loss 0.248290 LR 0.001000 Time 0.023123 +2023-10-02 20:54:48,309 - Epoch: [54][ 390/ 1236] Overall Loss 0.248959 Objective Loss 0.248959 LR 0.001000 Time 0.023067 +2023-10-02 20:54:48,521 - Epoch: [54][ 400/ 1236] Overall Loss 0.248995 Objective Loss 0.248995 LR 0.001000 Time 0.023021 +2023-10-02 20:54:48,733 - Epoch: [54][ 410/ 1236] Overall Loss 0.249338 Objective Loss 0.249338 LR 0.001000 Time 0.022971 +2023-10-02 20:54:48,945 - Epoch: [54][ 420/ 1236] Overall Loss 0.249882 Objective Loss 0.249882 LR 0.001000 Time 0.022930 +2023-10-02 20:54:49,156 - Epoch: [54][ 430/ 1236] Overall Loss 0.249831 Objective Loss 0.249831 LR 0.001000 Time 0.022883 +2023-10-02 20:54:49,368 - Epoch: [54][ 440/ 1236] Overall Loss 0.249983 Objective Loss 0.249983 LR 0.001000 Time 0.022845 +2023-10-02 20:54:49,579 - Epoch: [54][ 450/ 1236] Overall Loss 0.249170 Objective Loss 0.249170 LR 0.001000 Time 0.022802 +2023-10-02 20:54:49,792 - Epoch: [54][ 460/ 1236] Overall Loss 0.249042 Objective Loss 0.249042 LR 0.001000 Time 0.022768 +2023-10-02 20:54:50,002 - Epoch: [54][ 470/ 1236] Overall Loss 0.249638 Objective Loss 0.249638 LR 0.001000 Time 0.022728 +2023-10-02 20:54:50,215 - Epoch: [54][ 480/ 1236] Overall Loss 0.249689 Objective Loss 0.249689 LR 0.001000 Time 0.022697 +2023-10-02 20:54:50,424 - Epoch: [54][ 490/ 1236] Overall Loss 0.248467 Objective Loss 0.248467 LR 0.001000 Time 0.022658 +2023-10-02 20:54:50,635 - Epoch: [54][ 500/ 1236] Overall Loss 0.248554 Objective Loss 0.248554 LR 0.001000 Time 0.022626 +2023-10-02 20:54:50,843 - Epoch: [54][ 510/ 1236] Overall Loss 0.248327 Objective Loss 0.248327 LR 0.001000 Time 0.022589 +2023-10-02 20:54:51,053 - Epoch: [54][ 520/ 1236] Overall Loss 0.248456 Objective Loss 0.248456 LR 0.001000 Time 0.022558 +2023-10-02 20:54:51,260 - Epoch: [54][ 530/ 1236] Overall Loss 0.248433 Objective Loss 0.248433 LR 0.001000 Time 0.022522 +2023-10-02 20:54:51,469 - Epoch: [54][ 540/ 1236] Overall Loss 0.248287 Objective Loss 0.248287 LR 0.001000 Time 0.022492 +2023-10-02 20:54:51,677 - Epoch: [54][ 550/ 1236] Overall Loss 0.248570 Objective Loss 0.248570 LR 0.001000 Time 0.022459 +2023-10-02 20:54:51,888 - Epoch: [54][ 560/ 1236] Overall Loss 0.249005 Objective Loss 0.249005 LR 0.001000 Time 0.022433 +2023-10-02 20:54:52,094 - Epoch: [54][ 570/ 1236] Overall Loss 0.248846 Objective Loss 0.248846 LR 0.001000 Time 0.022401 +2023-10-02 20:54:52,305 - Epoch: [54][ 580/ 1236] Overall Loss 0.249192 Objective Loss 0.249192 LR 0.001000 Time 0.022378 +2023-10-02 20:54:52,512 - Epoch: [54][ 590/ 1236] Overall Loss 0.249675 Objective Loss 0.249675 LR 0.001000 Time 0.022349 +2023-10-02 20:54:52,722 - Epoch: [54][ 600/ 1236] Overall Loss 0.249514 Objective Loss 0.249514 LR 0.001000 Time 0.022326 +2023-10-02 20:54:52,929 - Epoch: [54][ 610/ 1236] Overall Loss 0.249189 Objective Loss 0.249189 LR 0.001000 Time 0.022299 +2023-10-02 20:54:53,140 - Epoch: [54][ 620/ 1236] Overall Loss 0.249507 Objective Loss 0.249507 LR 0.001000 Time 0.022278 +2023-10-02 20:54:53,346 - Epoch: [54][ 630/ 1236] Overall Loss 0.249814 Objective Loss 0.249814 LR 0.001000 Time 0.022252 +2023-10-02 20:54:53,557 - Epoch: [54][ 640/ 1236] Overall Loss 0.250226 Objective Loss 0.250226 LR 0.001000 Time 0.022234 +2023-10-02 20:54:53,764 - Epoch: [54][ 650/ 1236] Overall Loss 0.250108 Objective Loss 0.250108 LR 0.001000 Time 0.022209 +2023-10-02 20:54:53,974 - Epoch: [54][ 660/ 1236] Overall Loss 0.250140 Objective Loss 0.250140 LR 0.001000 Time 0.022190 +2023-10-02 20:54:54,182 - Epoch: [54][ 670/ 1236] Overall Loss 0.250314 Objective Loss 0.250314 LR 0.001000 Time 0.022168 +2023-10-02 20:54:54,392 - Epoch: [54][ 680/ 1236] Overall Loss 0.250260 Objective Loss 0.250260 LR 0.001000 Time 0.022150 +2023-10-02 20:54:54,601 - Epoch: [54][ 690/ 1236] Overall Loss 0.250620 Objective Loss 0.250620 LR 0.001000 Time 0.022129 +2023-10-02 20:54:54,812 - Epoch: [54][ 700/ 1236] Overall Loss 0.251014 Objective Loss 0.251014 LR 0.001000 Time 0.022114 +2023-10-02 20:54:55,019 - Epoch: [54][ 710/ 1236] Overall Loss 0.251307 Objective Loss 0.251307 LR 0.001000 Time 0.022094 +2023-10-02 20:54:55,230 - Epoch: [54][ 720/ 1236] Overall Loss 0.251498 Objective Loss 0.251498 LR 0.001000 Time 0.022079 +2023-10-02 20:54:55,437 - Epoch: [54][ 730/ 1236] Overall Loss 0.251277 Objective Loss 0.251277 LR 0.001000 Time 0.022060 +2023-10-02 20:54:55,647 - Epoch: [54][ 740/ 1236] Overall Loss 0.251519 Objective Loss 0.251519 LR 0.001000 Time 0.022046 +2023-10-02 20:54:55,854 - Epoch: [54][ 750/ 1236] Overall Loss 0.251876 Objective Loss 0.251876 LR 0.001000 Time 0.022027 +2023-10-02 20:54:56,065 - Epoch: [54][ 760/ 1236] Overall Loss 0.251665 Objective Loss 0.251665 LR 0.001000 Time 0.022014 +2023-10-02 20:54:56,272 - Epoch: [54][ 770/ 1236] Overall Loss 0.251757 Objective Loss 0.251757 LR 0.001000 Time 0.021997 +2023-10-02 20:54:56,481 - Epoch: [54][ 780/ 1236] Overall Loss 0.251528 Objective Loss 0.251528 LR 0.001000 Time 0.021983 +2023-10-02 20:54:56,689 - Epoch: [54][ 790/ 1236] Overall Loss 0.251766 Objective Loss 0.251766 LR 0.001000 Time 0.021966 +2023-10-02 20:54:56,898 - Epoch: [54][ 800/ 1236] Overall Loss 0.252165 Objective Loss 0.252165 LR 0.001000 Time 0.021952 +2023-10-02 20:54:57,106 - Epoch: [54][ 810/ 1236] Overall Loss 0.252518 Objective Loss 0.252518 LR 0.001000 Time 0.021936 +2023-10-02 20:54:57,317 - Epoch: [54][ 820/ 1236] Overall Loss 0.252225 Objective Loss 0.252225 LR 0.001000 Time 0.021925 +2023-10-02 20:54:57,524 - Epoch: [54][ 830/ 1236] Overall Loss 0.252310 Objective Loss 0.252310 LR 0.001000 Time 0.021910 +2023-10-02 20:54:57,735 - Epoch: [54][ 840/ 1236] Overall Loss 0.252110 Objective Loss 0.252110 LR 0.001000 Time 0.021900 +2023-10-02 20:54:57,942 - Epoch: [54][ 850/ 1236] Overall Loss 0.252585 Objective Loss 0.252585 LR 0.001000 Time 0.021885 +2023-10-02 20:54:58,152 - Epoch: [54][ 860/ 1236] Overall Loss 0.252512 Objective Loss 0.252512 LR 0.001000 Time 0.021875 +2023-10-02 20:54:58,359 - Epoch: [54][ 870/ 1236] Overall Loss 0.252551 Objective Loss 0.252551 LR 0.001000 Time 0.021861 +2023-10-02 20:54:58,570 - Epoch: [54][ 880/ 1236] Overall Loss 0.252557 Objective Loss 0.252557 LR 0.001000 Time 0.021852 +2023-10-02 20:54:58,777 - Epoch: [54][ 890/ 1236] Overall Loss 0.252750 Objective Loss 0.252750 LR 0.001000 Time 0.021839 +2023-10-02 20:54:58,986 - Epoch: [54][ 900/ 1236] Overall Loss 0.252642 Objective Loss 0.252642 LR 0.001000 Time 0.021829 +2023-10-02 20:54:59,195 - Epoch: [54][ 910/ 1236] Overall Loss 0.252721 Objective Loss 0.252721 LR 0.001000 Time 0.021816 +2023-10-02 20:54:59,404 - Epoch: [54][ 920/ 1236] Overall Loss 0.252829 Objective Loss 0.252829 LR 0.001000 Time 0.021806 +2023-10-02 20:54:59,613 - Epoch: [54][ 930/ 1236] Overall Loss 0.253069 Objective Loss 0.253069 LR 0.001000 Time 0.021794 +2023-10-02 20:54:59,822 - Epoch: [54][ 940/ 1236] Overall Loss 0.253030 Objective Loss 0.253030 LR 0.001000 Time 0.021785 +2023-10-02 20:55:00,030 - Epoch: [54][ 950/ 1236] Overall Loss 0.253159 Objective Loss 0.253159 LR 0.001000 Time 0.021773 +2023-10-02 20:55:00,241 - Epoch: [54][ 960/ 1236] Overall Loss 0.252773 Objective Loss 0.252773 LR 0.001000 Time 0.021766 +2023-10-02 20:55:00,448 - Epoch: [54][ 970/ 1236] Overall Loss 0.252785 Objective Loss 0.252785 LR 0.001000 Time 0.021754 +2023-10-02 20:55:00,659 - Epoch: [54][ 980/ 1236] Overall Loss 0.252626 Objective Loss 0.252626 LR 0.001000 Time 0.021747 +2023-10-02 20:55:00,866 - Epoch: [54][ 990/ 1236] Overall Loss 0.252430 Objective Loss 0.252430 LR 0.001000 Time 0.021736 +2023-10-02 20:55:01,076 - Epoch: [54][ 1000/ 1236] Overall Loss 0.252307 Objective Loss 0.252307 LR 0.001000 Time 0.021729 +2023-10-02 20:55:01,283 - Epoch: [54][ 1010/ 1236] Overall Loss 0.252330 Objective Loss 0.252330 LR 0.001000 Time 0.021718 +2023-10-02 20:55:01,494 - Epoch: [54][ 1020/ 1236] Overall Loss 0.252435 Objective Loss 0.252435 LR 0.001000 Time 0.021711 +2023-10-02 20:55:01,701 - Epoch: [54][ 1030/ 1236] Overall Loss 0.252603 Objective Loss 0.252603 LR 0.001000 Time 0.021701 +2023-10-02 20:55:01,911 - Epoch: [54][ 1040/ 1236] Overall Loss 0.252922 Objective Loss 0.252922 LR 0.001000 Time 0.021695 +2023-10-02 20:55:02,118 - Epoch: [54][ 1050/ 1236] Overall Loss 0.252808 Objective Loss 0.252808 LR 0.001000 Time 0.021685 +2023-10-02 20:55:02,329 - Epoch: [54][ 1060/ 1236] Overall Loss 0.252898 Objective Loss 0.252898 LR 0.001000 Time 0.021679 +2023-10-02 20:55:02,536 - Epoch: [54][ 1070/ 1236] Overall Loss 0.253142 Objective Loss 0.253142 LR 0.001000 Time 0.021669 +2023-10-02 20:55:02,746 - Epoch: [54][ 1080/ 1236] Overall Loss 0.253039 Objective Loss 0.253039 LR 0.001000 Time 0.021663 +2023-10-02 20:55:02,953 - Epoch: [54][ 1090/ 1236] Overall Loss 0.253076 Objective Loss 0.253076 LR 0.001000 Time 0.021654 +2023-10-02 20:55:03,164 - Epoch: [54][ 1100/ 1236] Overall Loss 0.253111 Objective Loss 0.253111 LR 0.001000 Time 0.021648 +2023-10-02 20:55:03,371 - Epoch: [54][ 1110/ 1236] Overall Loss 0.253328 Objective Loss 0.253328 LR 0.001000 Time 0.021640 +2023-10-02 20:55:03,581 - Epoch: [54][ 1120/ 1236] Overall Loss 0.253534 Objective Loss 0.253534 LR 0.001000 Time 0.021634 +2023-10-02 20:55:03,788 - Epoch: [54][ 1130/ 1236] Overall Loss 0.253754 Objective Loss 0.253754 LR 0.001000 Time 0.021626 +2023-10-02 20:55:03,998 - Epoch: [54][ 1140/ 1236] Overall Loss 0.253914 Objective Loss 0.253914 LR 0.001000 Time 0.021619 +2023-10-02 20:55:04,206 - Epoch: [54][ 1150/ 1236] Overall Loss 0.253947 Objective Loss 0.253947 LR 0.001000 Time 0.021611 +2023-10-02 20:55:04,417 - Epoch: [54][ 1160/ 1236] Overall Loss 0.254165 Objective Loss 0.254165 LR 0.001000 Time 0.021606 +2023-10-02 20:55:04,624 - Epoch: [54][ 1170/ 1236] Overall Loss 0.254420 Objective Loss 0.254420 LR 0.001000 Time 0.021598 +2023-10-02 20:55:04,834 - Epoch: [54][ 1180/ 1236] Overall Loss 0.254337 Objective Loss 0.254337 LR 0.001000 Time 0.021593 +2023-10-02 20:55:05,041 - Epoch: [54][ 1190/ 1236] Overall Loss 0.254441 Objective Loss 0.254441 LR 0.001000 Time 0.021585 +2023-10-02 20:55:05,250 - Epoch: [54][ 1200/ 1236] Overall Loss 0.254562 Objective Loss 0.254562 LR 0.001000 Time 0.021580 +2023-10-02 20:55:05,459 - Epoch: [54][ 1210/ 1236] Overall Loss 0.254478 Objective Loss 0.254478 LR 0.001000 Time 0.021572 +2023-10-02 20:55:05,669 - Epoch: [54][ 1220/ 1236] Overall Loss 0.254384 Objective Loss 0.254384 LR 0.001000 Time 0.021568 +2023-10-02 20:55:05,931 - Epoch: [54][ 1230/ 1236] Overall Loss 0.254445 Objective Loss 0.254445 LR 0.001000 Time 0.021605 +2023-10-02 20:55:06,053 - Epoch: [54][ 1236/ 1236] Overall Loss 0.254368 Objective Loss 0.254368 Top1 85.539715 Top5 98.574338 LR 0.001000 Time 0.021599 +2023-10-02 20:55:06,190 - --- validate (epoch=54)----------- +2023-10-02 20:55:06,190 - 29943 samples (256 per mini-batch) +2023-10-02 20:55:06,673 - Epoch: [54][ 10/ 117] Loss 0.322417 Top1 84.609375 Top5 98.203125 +2023-10-02 20:55:06,820 - Epoch: [54][ 20/ 117] Loss 0.322706 Top1 84.511719 Top5 98.105469 +2023-10-02 20:55:06,966 - Epoch: [54][ 30/ 117] Loss 0.330710 Top1 83.776042 Top5 98.059896 +2023-10-02 20:55:07,113 - Epoch: [54][ 40/ 117] Loss 0.331797 Top1 83.671875 Top5 97.988281 +2023-10-02 20:55:07,260 - Epoch: [54][ 50/ 117] Loss 0.327701 Top1 83.648438 Top5 98.015625 +2023-10-02 20:55:07,408 - Epoch: [54][ 60/ 117] Loss 0.326274 Top1 83.671875 Top5 98.040365 +2023-10-02 20:55:07,555 - Epoch: [54][ 70/ 117] Loss 0.325413 Top1 83.632812 Top5 98.024554 +2023-10-02 20:55:07,702 - Epoch: [54][ 80/ 117] Loss 0.320666 Top1 83.681641 Top5 98.081055 +2023-10-02 20:55:07,850 - Epoch: [54][ 90/ 117] Loss 0.319341 Top1 83.658854 Top5 98.107639 +2023-10-02 20:55:08,001 - Epoch: [54][ 100/ 117] Loss 0.322126 Top1 83.667969 Top5 98.113281 +2023-10-02 20:55:08,161 - Epoch: [54][ 110/ 117] Loss 0.325276 Top1 83.547585 Top5 98.064631 +2023-10-02 20:55:08,250 - Epoch: [54][ 117/ 117] Loss 0.324022 Top1 83.588819 Top5 98.069666 +2023-10-02 20:55:08,375 - ==> Top1: 83.589 Top5: 98.070 Loss: 0.324 + +2023-10-02 20:55:08,376 - ==> Confusion: +[[ 936 0 7 0 5 4 0 0 6 69 1 2 0 3 3 0 1 1 1 0 11] + [ 1 1033 1 0 5 32 2 24 3 1 2 1 0 1 2 5 5 0 9 1 3] + [ 12 0 967 10 0 0 23 6 0 3 2 1 2 4 1 9 0 1 4 6 5] + [ 2 2 8 991 0 5 1 4 3 1 8 0 5 3 30 1 1 3 4 2 15] + [ 27 6 2 0 956 11 0 0 0 10 0 0 0 3 7 6 15 0 0 3 4] + [ 4 24 1 2 4 1008 2 16 2 4 3 9 0 10 4 0 4 1 3 6 9] + [ 0 3 32 0 0 0 1119 4 0 0 5 1 0 0 0 12 0 0 1 7 7] + [ 3 7 23 2 2 43 5 1049 1 2 8 9 2 2 3 2 0 0 34 14 7] + [ 18 2 1 0 3 3 1 2 955 53 12 1 1 7 17 1 3 3 0 1 5] + [ 101 2 1 0 3 2 0 0 20 951 0 0 0 15 8 0 1 0 1 5 9] + [ 4 2 11 10 1 1 5 2 18 3 961 1 0 14 4 1 2 3 2 0 8] + [ 2 1 3 0 0 14 0 1 0 0 0 957 13 8 0 3 5 15 0 10 3] + [ 0 2 3 3 1 1 1 1 1 1 0 72 914 1 3 9 6 27 2 10 10] + [ 2 0 2 0 3 19 0 0 11 15 4 4 0 1033 4 1 1 1 0 3 16] + [ 19 0 4 15 3 0 0 0 12 8 2 0 3 2 1022 0 1 2 4 0 4] + [ 0 0 2 0 6 0 2 0 0 1 0 6 6 1 0 1073 16 12 0 5 4] + [ 0 14 4 0 4 7 0 0 0 2 0 2 1 1 2 10 1097 1 1 5 10] + [ 1 0 2 3 0 0 0 0 0 0 1 7 24 1 4 13 2 976 0 2 2] + [ 2 10 10 20 1 0 0 28 6 1 6 2 0 0 17 0 0 0 951 2 12] + [ 0 3 3 2 2 2 8 6 0 0 2 10 1 4 1 5 2 1 0 1098 2] + [ 186 157 184 112 81 238 36 111 87 95 174 141 321 269 159 69 109 51 130 213 4982]] + +2023-10-02 20:55:08,377 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:55:08,377 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:55:08,383 - + +2023-10-02 20:55:08,383 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:55:09,381 - Epoch: [55][ 10/ 1236] Overall Loss 0.231699 Objective Loss 0.231699 LR 0.001000 Time 0.099760 +2023-10-02 20:55:09,589 - Epoch: [55][ 20/ 1236] Overall Loss 0.225752 Objective Loss 0.225752 LR 0.001000 Time 0.060238 +2023-10-02 20:55:09,797 - Epoch: [55][ 30/ 1236] Overall Loss 0.230936 Objective Loss 0.230936 LR 0.001000 Time 0.047074 +2023-10-02 20:55:10,005 - Epoch: [55][ 40/ 1236] Overall Loss 0.238352 Objective Loss 0.238352 LR 0.001000 Time 0.040502 +2023-10-02 20:55:10,214 - Epoch: [55][ 50/ 1236] Overall Loss 0.239834 Objective Loss 0.239834 LR 0.001000 Time 0.036569 +2023-10-02 20:55:10,422 - Epoch: [55][ 60/ 1236] Overall Loss 0.242917 Objective Loss 0.242917 LR 0.001000 Time 0.033945 +2023-10-02 20:55:10,630 - Epoch: [55][ 70/ 1236] Overall Loss 0.241463 Objective Loss 0.241463 LR 0.001000 Time 0.032040 +2023-10-02 20:55:10,839 - Epoch: [55][ 80/ 1236] Overall Loss 0.240340 Objective Loss 0.240340 LR 0.001000 Time 0.030639 +2023-10-02 20:55:11,047 - Epoch: [55][ 90/ 1236] Overall Loss 0.241203 Objective Loss 0.241203 LR 0.001000 Time 0.029543 +2023-10-02 20:55:11,255 - Epoch: [55][ 100/ 1236] Overall Loss 0.239651 Objective Loss 0.239651 LR 0.001000 Time 0.028669 +2023-10-02 20:55:11,463 - Epoch: [55][ 110/ 1236] Overall Loss 0.237153 Objective Loss 0.237153 LR 0.001000 Time 0.027938 +2023-10-02 20:55:11,672 - Epoch: [55][ 120/ 1236] Overall Loss 0.237300 Objective Loss 0.237300 LR 0.001000 Time 0.027349 +2023-10-02 20:55:11,881 - Epoch: [55][ 130/ 1236] Overall Loss 0.235460 Objective Loss 0.235460 LR 0.001000 Time 0.026837 +2023-10-02 20:55:12,093 - Epoch: [55][ 140/ 1236] Overall Loss 0.237224 Objective Loss 0.237224 LR 0.001000 Time 0.026432 +2023-10-02 20:55:12,300 - Epoch: [55][ 150/ 1236] Overall Loss 0.238351 Objective Loss 0.238351 LR 0.001000 Time 0.026048 +2023-10-02 20:55:12,509 - Epoch: [55][ 160/ 1236] Overall Loss 0.238445 Objective Loss 0.238445 LR 0.001000 Time 0.025727 +2023-10-02 20:55:12,715 - Epoch: [55][ 170/ 1236] Overall Loss 0.239849 Objective Loss 0.239849 LR 0.001000 Time 0.025417 +2023-10-02 20:55:12,923 - Epoch: [55][ 180/ 1236] Overall Loss 0.243171 Objective Loss 0.243171 LR 0.001000 Time 0.025158 +2023-10-02 20:55:13,129 - Epoch: [55][ 190/ 1236] Overall Loss 0.245478 Objective Loss 0.245478 LR 0.001000 Time 0.024910 +2023-10-02 20:55:13,337 - Epoch: [55][ 200/ 1236] Overall Loss 0.244015 Objective Loss 0.244015 LR 0.001000 Time 0.024704 +2023-10-02 20:55:13,543 - Epoch: [55][ 210/ 1236] Overall Loss 0.244987 Objective Loss 0.244987 LR 0.001000 Time 0.024502 +2023-10-02 20:55:13,751 - Epoch: [55][ 220/ 1236] Overall Loss 0.246482 Objective Loss 0.246482 LR 0.001000 Time 0.024332 +2023-10-02 20:55:13,958 - Epoch: [55][ 230/ 1236] Overall Loss 0.246720 Objective Loss 0.246720 LR 0.001000 Time 0.024164 +2023-10-02 20:55:14,165 - Epoch: [55][ 240/ 1236] Overall Loss 0.247705 Objective Loss 0.247705 LR 0.001000 Time 0.024021 +2023-10-02 20:55:14,371 - Epoch: [55][ 250/ 1236] Overall Loss 0.246859 Objective Loss 0.246859 LR 0.001000 Time 0.023880 +2023-10-02 20:55:14,579 - Epoch: [55][ 260/ 1236] Overall Loss 0.248084 Objective Loss 0.248084 LR 0.001000 Time 0.023759 +2023-10-02 20:55:14,785 - Epoch: [55][ 270/ 1236] Overall Loss 0.249684 Objective Loss 0.249684 LR 0.001000 Time 0.023638 +2023-10-02 20:55:14,993 - Epoch: [55][ 280/ 1236] Overall Loss 0.249780 Objective Loss 0.249780 LR 0.001000 Time 0.023534 +2023-10-02 20:55:15,199 - Epoch: [55][ 290/ 1236] Overall Loss 0.250175 Objective Loss 0.250175 LR 0.001000 Time 0.023429 +2023-10-02 20:55:15,407 - Epoch: [55][ 300/ 1236] Overall Loss 0.250242 Objective Loss 0.250242 LR 0.001000 Time 0.023340 +2023-10-02 20:55:15,613 - Epoch: [55][ 310/ 1236] Overall Loss 0.250578 Objective Loss 0.250578 LR 0.001000 Time 0.023247 +2023-10-02 20:55:15,823 - Epoch: [55][ 320/ 1236] Overall Loss 0.250505 Objective Loss 0.250505 LR 0.001000 Time 0.023173 +2023-10-02 20:55:16,029 - Epoch: [55][ 330/ 1236] Overall Loss 0.250509 Objective Loss 0.250509 LR 0.001000 Time 0.023094 +2023-10-02 20:55:16,238 - Epoch: [55][ 340/ 1236] Overall Loss 0.249972 Objective Loss 0.249972 LR 0.001000 Time 0.023030 +2023-10-02 20:55:16,446 - Epoch: [55][ 350/ 1236] Overall Loss 0.250416 Objective Loss 0.250416 LR 0.001000 Time 0.022964 +2023-10-02 20:55:16,655 - Epoch: [55][ 360/ 1236] Overall Loss 0.250335 Objective Loss 0.250335 LR 0.001000 Time 0.022906 +2023-10-02 20:55:16,862 - Epoch: [55][ 370/ 1236] Overall Loss 0.250338 Objective Loss 0.250338 LR 0.001000 Time 0.022846 +2023-10-02 20:55:17,072 - Epoch: [55][ 380/ 1236] Overall Loss 0.249773 Objective Loss 0.249773 LR 0.001000 Time 0.022795 +2023-10-02 20:55:17,279 - Epoch: [55][ 390/ 1236] Overall Loss 0.249320 Objective Loss 0.249320 LR 0.001000 Time 0.022741 +2023-10-02 20:55:17,489 - Epoch: [55][ 400/ 1236] Overall Loss 0.249755 Objective Loss 0.249755 LR 0.001000 Time 0.022696 +2023-10-02 20:55:17,696 - Epoch: [55][ 410/ 1236] Overall Loss 0.249680 Objective Loss 0.249680 LR 0.001000 Time 0.022647 +2023-10-02 20:55:17,906 - Epoch: [55][ 420/ 1236] Overall Loss 0.249618 Objective Loss 0.249618 LR 0.001000 Time 0.022605 +2023-10-02 20:55:18,113 - Epoch: [55][ 430/ 1236] Overall Loss 0.249098 Objective Loss 0.249098 LR 0.001000 Time 0.022561 +2023-10-02 20:55:18,322 - Epoch: [55][ 440/ 1236] Overall Loss 0.249585 Objective Loss 0.249585 LR 0.001000 Time 0.022522 +2023-10-02 20:55:18,530 - Epoch: [55][ 450/ 1236] Overall Loss 0.249776 Objective Loss 0.249776 LR 0.001000 Time 0.022482 +2023-10-02 20:55:18,738 - Epoch: [55][ 460/ 1236] Overall Loss 0.250328 Objective Loss 0.250328 LR 0.001000 Time 0.022445 +2023-10-02 20:55:18,945 - Epoch: [55][ 470/ 1236] Overall Loss 0.250663 Objective Loss 0.250663 LR 0.001000 Time 0.022404 +2023-10-02 20:55:19,153 - Epoch: [55][ 480/ 1236] Overall Loss 0.250811 Objective Loss 0.250811 LR 0.001000 Time 0.022370 +2023-10-02 20:55:19,360 - Epoch: [55][ 490/ 1236] Overall Loss 0.251630 Objective Loss 0.251630 LR 0.001000 Time 0.022332 +2023-10-02 20:55:19,569 - Epoch: [55][ 500/ 1236] Overall Loss 0.251645 Objective Loss 0.251645 LR 0.001000 Time 0.022304 +2023-10-02 20:55:19,777 - Epoch: [55][ 510/ 1236] Overall Loss 0.252474 Objective Loss 0.252474 LR 0.001000 Time 0.022272 +2023-10-02 20:55:19,986 - Epoch: [55][ 520/ 1236] Overall Loss 0.252352 Objective Loss 0.252352 LR 0.001000 Time 0.022246 +2023-10-02 20:55:20,194 - Epoch: [55][ 530/ 1236] Overall Loss 0.252712 Objective Loss 0.252712 LR 0.001000 Time 0.022217 +2023-10-02 20:55:20,403 - Epoch: [55][ 540/ 1236] Overall Loss 0.252512 Objective Loss 0.252512 LR 0.001000 Time 0.022193 +2023-10-02 20:55:20,611 - Epoch: [55][ 550/ 1236] Overall Loss 0.252688 Objective Loss 0.252688 LR 0.001000 Time 0.022166 +2023-10-02 20:55:20,820 - Epoch: [55][ 560/ 1236] Overall Loss 0.252234 Objective Loss 0.252234 LR 0.001000 Time 0.022143 +2023-10-02 20:55:21,027 - Epoch: [55][ 570/ 1236] Overall Loss 0.251885 Objective Loss 0.251885 LR 0.001000 Time 0.022118 +2023-10-02 20:55:21,237 - Epoch: [55][ 580/ 1236] Overall Loss 0.251723 Objective Loss 0.251723 LR 0.001000 Time 0.022097 +2023-10-02 20:55:21,444 - Epoch: [55][ 590/ 1236] Overall Loss 0.251696 Objective Loss 0.251696 LR 0.001000 Time 0.022073 +2023-10-02 20:55:21,653 - Epoch: [55][ 600/ 1236] Overall Loss 0.251830 Objective Loss 0.251830 LR 0.001000 Time 0.022053 +2023-10-02 20:55:21,861 - Epoch: [55][ 610/ 1236] Overall Loss 0.251139 Objective Loss 0.251139 LR 0.001000 Time 0.022031 +2023-10-02 20:55:22,070 - Epoch: [55][ 620/ 1236] Overall Loss 0.251416 Objective Loss 0.251416 LR 0.001000 Time 0.022012 +2023-10-02 20:55:22,278 - Epoch: [55][ 630/ 1236] Overall Loss 0.251382 Objective Loss 0.251382 LR 0.001000 Time 0.021992 +2023-10-02 20:55:22,487 - Epoch: [55][ 640/ 1236] Overall Loss 0.251507 Objective Loss 0.251507 LR 0.001000 Time 0.021975 +2023-10-02 20:55:22,694 - Epoch: [55][ 650/ 1236] Overall Loss 0.251810 Objective Loss 0.251810 LR 0.001000 Time 0.021955 +2023-10-02 20:55:22,904 - Epoch: [55][ 660/ 1236] Overall Loss 0.251323 Objective Loss 0.251323 LR 0.001000 Time 0.021939 +2023-10-02 20:55:23,111 - Epoch: [55][ 670/ 1236] Overall Loss 0.251240 Objective Loss 0.251240 LR 0.001000 Time 0.021920 +2023-10-02 20:55:23,321 - Epoch: [55][ 680/ 1236] Overall Loss 0.251294 Objective Loss 0.251294 LR 0.001000 Time 0.021906 +2023-10-02 20:55:23,528 - Epoch: [55][ 690/ 1236] Overall Loss 0.251059 Objective Loss 0.251059 LR 0.001000 Time 0.021888 +2023-10-02 20:55:23,738 - Epoch: [55][ 700/ 1236] Overall Loss 0.251348 Objective Loss 0.251348 LR 0.001000 Time 0.021874 +2023-10-02 20:55:23,945 - Epoch: [55][ 710/ 1236] Overall Loss 0.251255 Objective Loss 0.251255 LR 0.001000 Time 0.021858 +2023-10-02 20:55:24,155 - Epoch: [55][ 720/ 1236] Overall Loss 0.251715 Objective Loss 0.251715 LR 0.001000 Time 0.021845 +2023-10-02 20:55:24,362 - Epoch: [55][ 730/ 1236] Overall Loss 0.251810 Objective Loss 0.251810 LR 0.001000 Time 0.021829 +2023-10-02 20:55:24,572 - Epoch: [55][ 740/ 1236] Overall Loss 0.251733 Objective Loss 0.251733 LR 0.001000 Time 0.021817 +2023-10-02 20:55:24,779 - Epoch: [55][ 750/ 1236] Overall Loss 0.251769 Objective Loss 0.251769 LR 0.001000 Time 0.021802 +2023-10-02 20:55:24,988 - Epoch: [55][ 760/ 1236] Overall Loss 0.251668 Objective Loss 0.251668 LR 0.001000 Time 0.021789 +2023-10-02 20:55:25,196 - Epoch: [55][ 770/ 1236] Overall Loss 0.251700 Objective Loss 0.251700 LR 0.001000 Time 0.021776 +2023-10-02 20:55:25,406 - Epoch: [55][ 780/ 1236] Overall Loss 0.251308 Objective Loss 0.251308 LR 0.001000 Time 0.021766 +2023-10-02 20:55:25,614 - Epoch: [55][ 790/ 1236] Overall Loss 0.251256 Objective Loss 0.251256 LR 0.001000 Time 0.021752 +2023-10-02 20:55:25,823 - Epoch: [55][ 800/ 1236] Overall Loss 0.251499 Objective Loss 0.251499 LR 0.001000 Time 0.021742 +2023-10-02 20:55:26,030 - Epoch: [55][ 810/ 1236] Overall Loss 0.251381 Objective Loss 0.251381 LR 0.001000 Time 0.021728 +2023-10-02 20:55:26,240 - Epoch: [55][ 820/ 1236] Overall Loss 0.251575 Objective Loss 0.251575 LR 0.001000 Time 0.021719 +2023-10-02 20:55:26,447 - Epoch: [55][ 830/ 1236] Overall Loss 0.252039 Objective Loss 0.252039 LR 0.001000 Time 0.021707 +2023-10-02 20:55:26,657 - Epoch: [55][ 840/ 1236] Overall Loss 0.252428 Objective Loss 0.252428 LR 0.001000 Time 0.021697 +2023-10-02 20:55:26,864 - Epoch: [55][ 850/ 1236] Overall Loss 0.252553 Objective Loss 0.252553 LR 0.001000 Time 0.021686 +2023-10-02 20:55:27,073 - Epoch: [55][ 860/ 1236] Overall Loss 0.252525 Objective Loss 0.252525 LR 0.001000 Time 0.021676 +2023-10-02 20:55:27,281 - Epoch: [55][ 870/ 1236] Overall Loss 0.252712 Objective Loss 0.252712 LR 0.001000 Time 0.021666 +2023-10-02 20:55:27,490 - Epoch: [55][ 880/ 1236] Overall Loss 0.252868 Objective Loss 0.252868 LR 0.001000 Time 0.021657 +2023-10-02 20:55:27,698 - Epoch: [55][ 890/ 1236] Overall Loss 0.252776 Objective Loss 0.252776 LR 0.001000 Time 0.021646 +2023-10-02 20:55:27,907 - Epoch: [55][ 900/ 1236] Overall Loss 0.252810 Objective Loss 0.252810 LR 0.001000 Time 0.021638 +2023-10-02 20:55:28,115 - Epoch: [55][ 910/ 1236] Overall Loss 0.252847 Objective Loss 0.252847 LR 0.001000 Time 0.021628 +2023-10-02 20:55:28,325 - Epoch: [55][ 920/ 1236] Overall Loss 0.253075 Objective Loss 0.253075 LR 0.001000 Time 0.021621 +2023-10-02 20:55:28,532 - Epoch: [55][ 930/ 1236] Overall Loss 0.253352 Objective Loss 0.253352 LR 0.001000 Time 0.021611 +2023-10-02 20:55:28,741 - Epoch: [55][ 940/ 1236] Overall Loss 0.253503 Objective Loss 0.253503 LR 0.001000 Time 0.021603 +2023-10-02 20:55:28,949 - Epoch: [55][ 950/ 1236] Overall Loss 0.253525 Objective Loss 0.253525 LR 0.001000 Time 0.021594 +2023-10-02 20:55:29,158 - Epoch: [55][ 960/ 1236] Overall Loss 0.253837 Objective Loss 0.253837 LR 0.001000 Time 0.021587 +2023-10-02 20:55:29,366 - Epoch: [55][ 970/ 1236] Overall Loss 0.254169 Objective Loss 0.254169 LR 0.001000 Time 0.021578 +2023-10-02 20:55:29,576 - Epoch: [55][ 980/ 1236] Overall Loss 0.254277 Objective Loss 0.254277 LR 0.001000 Time 0.021571 +2023-10-02 20:55:29,783 - Epoch: [55][ 990/ 1236] Overall Loss 0.254197 Objective Loss 0.254197 LR 0.001000 Time 0.021563 +2023-10-02 20:55:29,992 - Epoch: [55][ 1000/ 1236] Overall Loss 0.254230 Objective Loss 0.254230 LR 0.001000 Time 0.021556 +2023-10-02 20:55:30,200 - Epoch: [55][ 1010/ 1236] Overall Loss 0.254524 Objective Loss 0.254524 LR 0.001000 Time 0.021548 +2023-10-02 20:55:30,409 - Epoch: [55][ 1020/ 1236] Overall Loss 0.254313 Objective Loss 0.254313 LR 0.001000 Time 0.021541 +2023-10-02 20:55:30,617 - Epoch: [55][ 1030/ 1236] Overall Loss 0.254259 Objective Loss 0.254259 LR 0.001000 Time 0.021533 +2023-10-02 20:55:30,826 - Epoch: [55][ 1040/ 1236] Overall Loss 0.254304 Objective Loss 0.254304 LR 0.001000 Time 0.021527 +2023-10-02 20:55:31,033 - Epoch: [55][ 1050/ 1236] Overall Loss 0.254648 Objective Loss 0.254648 LR 0.001000 Time 0.021519 +2023-10-02 20:55:31,242 - Epoch: [55][ 1060/ 1236] Overall Loss 0.254976 Objective Loss 0.254976 LR 0.001000 Time 0.021513 +2023-10-02 20:55:31,450 - Epoch: [55][ 1070/ 1236] Overall Loss 0.255025 Objective Loss 0.255025 LR 0.001000 Time 0.021506 +2023-10-02 20:55:31,660 - Epoch: [55][ 1080/ 1236] Overall Loss 0.255073 Objective Loss 0.255073 LR 0.001000 Time 0.021501 +2023-10-02 20:55:31,867 - Epoch: [55][ 1090/ 1236] Overall Loss 0.254922 Objective Loss 0.254922 LR 0.001000 Time 0.021494 +2023-10-02 20:55:32,077 - Epoch: [55][ 1100/ 1236] Overall Loss 0.254621 Objective Loss 0.254621 LR 0.001000 Time 0.021488 +2023-10-02 20:55:32,284 - Epoch: [55][ 1110/ 1236] Overall Loss 0.254314 Objective Loss 0.254314 LR 0.001000 Time 0.021481 +2023-10-02 20:55:32,494 - Epoch: [55][ 1120/ 1236] Overall Loss 0.254384 Objective Loss 0.254384 LR 0.001000 Time 0.021477 +2023-10-02 20:55:32,701 - Epoch: [55][ 1130/ 1236] Overall Loss 0.254552 Objective Loss 0.254552 LR 0.001000 Time 0.021470 +2023-10-02 20:55:32,911 - Epoch: [55][ 1140/ 1236] Overall Loss 0.254698 Objective Loss 0.254698 LR 0.001000 Time 0.021465 +2023-10-02 20:55:33,118 - Epoch: [55][ 1150/ 1236] Overall Loss 0.254976 Objective Loss 0.254976 LR 0.001000 Time 0.021458 +2023-10-02 20:55:33,328 - Epoch: [55][ 1160/ 1236] Overall Loss 0.255178 Objective Loss 0.255178 LR 0.001000 Time 0.021454 +2023-10-02 20:55:33,536 - Epoch: [55][ 1170/ 1236] Overall Loss 0.255257 Objective Loss 0.255257 LR 0.001000 Time 0.021448 +2023-10-02 20:55:33,745 - Epoch: [55][ 1180/ 1236] Overall Loss 0.255373 Objective Loss 0.255373 LR 0.001000 Time 0.021443 +2023-10-02 20:55:33,952 - Epoch: [55][ 1190/ 1236] Overall Loss 0.255369 Objective Loss 0.255369 LR 0.001000 Time 0.021437 +2023-10-02 20:55:34,162 - Epoch: [55][ 1200/ 1236] Overall Loss 0.255557 Objective Loss 0.255557 LR 0.001000 Time 0.021432 +2023-10-02 20:55:34,369 - Epoch: [55][ 1210/ 1236] Overall Loss 0.255769 Objective Loss 0.255769 LR 0.001000 Time 0.021427 +2023-10-02 20:55:34,579 - Epoch: [55][ 1220/ 1236] Overall Loss 0.255697 Objective Loss 0.255697 LR 0.001000 Time 0.021423 +2023-10-02 20:55:34,842 - Epoch: [55][ 1230/ 1236] Overall Loss 0.255579 Objective Loss 0.255579 LR 0.001000 Time 0.021461 +2023-10-02 20:55:34,965 - Epoch: [55][ 1236/ 1236] Overall Loss 0.255483 Objective Loss 0.255483 Top1 86.150713 Top5 99.185336 LR 0.001000 Time 0.021457 +2023-10-02 20:55:35,084 - --- validate (epoch=55)----------- +2023-10-02 20:55:35,085 - 29943 samples (256 per mini-batch) +2023-10-02 20:55:35,585 - Epoch: [55][ 10/ 117] Loss 0.357457 Top1 84.179688 Top5 98.476562 +2023-10-02 20:55:35,738 - Epoch: [55][ 20/ 117] Loss 0.346356 Top1 84.296875 Top5 98.613281 +2023-10-02 20:55:35,893 - Epoch: [55][ 30/ 117] Loss 0.340561 Top1 84.401042 Top5 98.515625 +2023-10-02 20:55:36,048 - Epoch: [55][ 40/ 117] Loss 0.333244 Top1 84.189453 Top5 98.515625 +2023-10-02 20:55:36,202 - Epoch: [55][ 50/ 117] Loss 0.326756 Top1 84.414062 Top5 98.453125 +2023-10-02 20:55:36,356 - Epoch: [55][ 60/ 117] Loss 0.324821 Top1 84.303385 Top5 98.411458 +2023-10-02 20:55:36,509 - Epoch: [55][ 70/ 117] Loss 0.321415 Top1 84.358259 Top5 98.420759 +2023-10-02 20:55:36,662 - Epoch: [55][ 80/ 117] Loss 0.327264 Top1 84.438477 Top5 98.359375 +2023-10-02 20:55:36,816 - Epoch: [55][ 90/ 117] Loss 0.329390 Top1 84.292535 Top5 98.320312 +2023-10-02 20:55:36,979 - Epoch: [55][ 100/ 117] Loss 0.330985 Top1 84.148438 Top5 98.265625 +2023-10-02 20:55:37,146 - Epoch: [55][ 110/ 117] Loss 0.330196 Top1 84.215199 Top5 98.267045 +2023-10-02 20:55:37,236 - Epoch: [55][ 117/ 117] Loss 0.331132 Top1 84.133186 Top5 98.260027 +2023-10-02 20:55:37,329 - ==> Top1: 84.133 Top5: 98.260 Loss: 0.331 + +2023-10-02 20:55:37,330 - ==> Confusion: +[[ 929 1 4 0 5 1 0 0 2 72 1 1 2 4 4 2 2 2 0 0 18] + [ 0 1040 1 0 5 38 2 16 4 1 0 1 0 0 3 4 1 0 6 3 6] + [ 7 0 955 7 4 2 24 10 0 1 1 1 8 2 5 7 0 1 7 4 10] + [ 2 3 13 960 0 2 3 3 6 2 4 0 9 6 42 3 2 1 7 2 19] + [ 35 7 2 1 950 4 0 0 0 15 0 1 0 3 9 6 8 0 0 4 5] + [ 3 25 1 0 3 1003 0 21 1 7 1 3 5 9 8 3 0 0 4 5 14] + [ 0 3 30 0 0 0 1120 8 0 0 5 1 0 0 0 10 0 0 1 6 7] + [ 3 19 24 0 4 43 6 1051 1 1 4 0 3 5 2 0 1 0 27 13 11] + [ 17 3 2 0 2 1 0 1 962 38 4 1 3 19 28 2 2 1 1 0 2] + [ 114 1 0 0 5 0 1 0 30 924 2 0 0 19 10 0 1 0 0 5 7] + [ 4 3 10 5 1 3 4 5 18 2 950 2 1 18 4 1 0 1 3 1 17] + [ 0 2 2 0 1 12 0 1 0 0 0 930 37 10 0 8 1 14 0 8 9] + [ 0 1 5 3 1 2 1 1 2 2 1 30 958 2 4 11 2 19 2 7 14] + [ 2 1 1 0 1 8 0 0 9 8 5 4 2 1059 4 1 0 2 0 3 9] + [ 13 1 2 12 2 0 0 0 15 6 1 0 3 3 1026 0 2 6 5 0 4] + [ 0 0 2 1 3 0 1 0 0 0 0 1 7 0 0 1083 10 10 3 8 5] + [ 1 13 2 0 12 5 0 0 1 1 0 5 3 1 6 12 1074 0 0 9 16] + [ 0 0 1 3 0 0 5 0 2 0 0 3 20 1 2 14 0 979 0 1 7] + [ 4 6 8 11 1 2 1 32 5 0 2 1 1 0 28 0 0 0 947 0 19] + [ 0 5 3 0 0 6 7 11 0 0 1 6 3 2 1 10 4 0 2 1083 8] + [ 157 151 132 60 101 154 36 102 113 87 138 84 350 318 174 80 76 54 107 222 5209]] + +2023-10-02 20:55:37,331 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:55:37,331 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:55:37,337 - + +2023-10-02 20:55:37,337 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:55:38,451 - Epoch: [56][ 10/ 1236] Overall Loss 0.236135 Objective Loss 0.236135 LR 0.001000 Time 0.111384 +2023-10-02 20:55:38,660 - Epoch: [56][ 20/ 1236] Overall Loss 0.231275 Objective Loss 0.231275 LR 0.001000 Time 0.066081 +2023-10-02 20:55:38,867 - Epoch: [56][ 30/ 1236] Overall Loss 0.243336 Objective Loss 0.243336 LR 0.001000 Time 0.050969 +2023-10-02 20:55:39,076 - Epoch: [56][ 40/ 1236] Overall Loss 0.242700 Objective Loss 0.242700 LR 0.001000 Time 0.043427 +2023-10-02 20:55:39,283 - Epoch: [56][ 50/ 1236] Overall Loss 0.239171 Objective Loss 0.239171 LR 0.001000 Time 0.038866 +2023-10-02 20:55:39,492 - Epoch: [56][ 60/ 1236] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.001000 Time 0.035861 +2023-10-02 20:55:39,699 - Epoch: [56][ 70/ 1236] Overall Loss 0.241405 Objective Loss 0.241405 LR 0.001000 Time 0.033676 +2023-10-02 20:55:39,907 - Epoch: [56][ 80/ 1236] Overall Loss 0.243426 Objective Loss 0.243426 LR 0.001000 Time 0.032066 +2023-10-02 20:55:40,113 - Epoch: [56][ 90/ 1236] Overall Loss 0.247988 Objective Loss 0.247988 LR 0.001000 Time 0.030775 +2023-10-02 20:55:40,322 - Epoch: [56][ 100/ 1236] Overall Loss 0.245204 Objective Loss 0.245204 LR 0.001000 Time 0.029783 +2023-10-02 20:55:40,529 - Epoch: [56][ 110/ 1236] Overall Loss 0.244175 Objective Loss 0.244175 LR 0.001000 Time 0.028945 +2023-10-02 20:55:40,738 - Epoch: [56][ 120/ 1236] Overall Loss 0.246480 Objective Loss 0.246480 LR 0.001000 Time 0.028270 +2023-10-02 20:55:40,945 - Epoch: [56][ 130/ 1236] Overall Loss 0.246388 Objective Loss 0.246388 LR 0.001000 Time 0.027687 +2023-10-02 20:55:41,154 - Epoch: [56][ 140/ 1236] Overall Loss 0.244485 Objective Loss 0.244485 LR 0.001000 Time 0.027196 +2023-10-02 20:55:41,361 - Epoch: [56][ 150/ 1236] Overall Loss 0.243062 Objective Loss 0.243062 LR 0.001000 Time 0.026757 +2023-10-02 20:55:41,571 - Epoch: [56][ 160/ 1236] Overall Loss 0.243622 Objective Loss 0.243622 LR 0.001000 Time 0.026393 +2023-10-02 20:55:41,777 - Epoch: [56][ 170/ 1236] Overall Loss 0.244498 Objective Loss 0.244498 LR 0.001000 Time 0.026050 +2023-10-02 20:55:41,987 - Epoch: [56][ 180/ 1236] Overall Loss 0.243339 Objective Loss 0.243339 LR 0.001000 Time 0.025765 +2023-10-02 20:55:42,192 - Epoch: [56][ 190/ 1236] Overall Loss 0.243306 Objective Loss 0.243306 LR 0.001000 Time 0.025491 +2023-10-02 20:55:42,403 - Epoch: [56][ 200/ 1236] Overall Loss 0.243046 Objective Loss 0.243046 LR 0.001000 Time 0.025266 +2023-10-02 20:55:42,608 - Epoch: [56][ 210/ 1236] Overall Loss 0.244789 Objective Loss 0.244789 LR 0.001000 Time 0.025042 +2023-10-02 20:55:42,819 - Epoch: [56][ 220/ 1236] Overall Loss 0.245626 Objective Loss 0.245626 LR 0.001000 Time 0.024860 +2023-10-02 20:55:43,027 - Epoch: [56][ 230/ 1236] Overall Loss 0.246124 Objective Loss 0.246124 LR 0.001000 Time 0.024674 +2023-10-02 20:55:43,237 - Epoch: [56][ 240/ 1236] Overall Loss 0.245744 Objective Loss 0.245744 LR 0.001000 Time 0.024521 +2023-10-02 20:55:43,443 - Epoch: [56][ 250/ 1236] Overall Loss 0.245447 Objective Loss 0.245447 LR 0.001000 Time 0.024365 +2023-10-02 20:55:43,653 - Epoch: [56][ 260/ 1236] Overall Loss 0.246359 Objective Loss 0.246359 LR 0.001000 Time 0.024231 +2023-10-02 20:55:43,860 - Epoch: [56][ 270/ 1236] Overall Loss 0.247378 Objective Loss 0.247378 LR 0.001000 Time 0.024096 +2023-10-02 20:55:44,070 - Epoch: [56][ 280/ 1236] Overall Loss 0.248446 Objective Loss 0.248446 LR 0.001000 Time 0.023984 +2023-10-02 20:55:44,276 - Epoch: [56][ 290/ 1236] Overall Loss 0.248086 Objective Loss 0.248086 LR 0.001000 Time 0.023867 +2023-10-02 20:55:44,486 - Epoch: [56][ 300/ 1236] Overall Loss 0.248115 Objective Loss 0.248115 LR 0.001000 Time 0.023770 +2023-10-02 20:55:44,692 - Epoch: [56][ 310/ 1236] Overall Loss 0.246886 Objective Loss 0.246886 LR 0.001000 Time 0.023667 +2023-10-02 20:55:44,902 - Epoch: [56][ 320/ 1236] Overall Loss 0.246804 Objective Loss 0.246804 LR 0.001000 Time 0.023583 +2023-10-02 20:55:45,108 - Epoch: [56][ 330/ 1236] Overall Loss 0.247684 Objective Loss 0.247684 LR 0.001000 Time 0.023492 +2023-10-02 20:55:45,319 - Epoch: [56][ 340/ 1236] Overall Loss 0.247306 Objective Loss 0.247306 LR 0.001000 Time 0.023419 +2023-10-02 20:55:45,525 - Epoch: [56][ 350/ 1236] Overall Loss 0.247877 Objective Loss 0.247877 LR 0.001000 Time 0.023338 +2023-10-02 20:55:45,734 - Epoch: [56][ 360/ 1236] Overall Loss 0.247736 Objective Loss 0.247736 LR 0.001000 Time 0.023269 +2023-10-02 20:55:45,941 - Epoch: [56][ 370/ 1236] Overall Loss 0.247696 Objective Loss 0.247696 LR 0.001000 Time 0.023197 +2023-10-02 20:55:46,150 - Epoch: [56][ 380/ 1236] Overall Loss 0.247756 Objective Loss 0.247756 LR 0.001000 Time 0.023135 +2023-10-02 20:55:46,357 - Epoch: [56][ 390/ 1236] Overall Loss 0.247759 Objective Loss 0.247759 LR 0.001000 Time 0.023070 +2023-10-02 20:55:46,567 - Epoch: [56][ 400/ 1236] Overall Loss 0.247973 Objective Loss 0.247973 LR 0.001000 Time 0.023017 +2023-10-02 20:55:46,773 - Epoch: [56][ 410/ 1236] Overall Loss 0.248018 Objective Loss 0.248018 LR 0.001000 Time 0.022958 +2023-10-02 20:55:46,984 - Epoch: [56][ 420/ 1236] Overall Loss 0.248289 Objective Loss 0.248289 LR 0.001000 Time 0.022911 +2023-10-02 20:55:47,192 - Epoch: [56][ 430/ 1236] Overall Loss 0.247715 Objective Loss 0.247715 LR 0.001000 Time 0.022863 +2023-10-02 20:55:47,402 - Epoch: [56][ 440/ 1236] Overall Loss 0.247570 Objective Loss 0.247570 LR 0.001000 Time 0.022819 +2023-10-02 20:55:47,614 - Epoch: [56][ 450/ 1236] Overall Loss 0.248179 Objective Loss 0.248179 LR 0.001000 Time 0.022779 +2023-10-02 20:55:47,827 - Epoch: [56][ 460/ 1236] Overall Loss 0.248174 Objective Loss 0.248174 LR 0.001000 Time 0.022747 +2023-10-02 20:55:48,040 - Epoch: [56][ 470/ 1236] Overall Loss 0.248336 Objective Loss 0.248336 LR 0.001000 Time 0.022715 +2023-10-02 20:55:48,252 - Epoch: [56][ 480/ 1236] Overall Loss 0.248547 Objective Loss 0.248547 LR 0.001000 Time 0.022683 +2023-10-02 20:55:48,462 - Epoch: [56][ 490/ 1236] Overall Loss 0.248715 Objective Loss 0.248715 LR 0.001000 Time 0.022648 +2023-10-02 20:55:48,674 - Epoch: [56][ 500/ 1236] Overall Loss 0.248259 Objective Loss 0.248259 LR 0.001000 Time 0.022619 +2023-10-02 20:55:48,883 - Epoch: [56][ 510/ 1236] Overall Loss 0.248731 Objective Loss 0.248731 LR 0.001000 Time 0.022583 +2023-10-02 20:55:49,094 - Epoch: [56][ 520/ 1236] Overall Loss 0.248577 Objective Loss 0.248577 LR 0.001000 Time 0.022555 +2023-10-02 20:55:49,303 - Epoch: [56][ 530/ 1236] Overall Loss 0.248755 Objective Loss 0.248755 LR 0.001000 Time 0.022523 +2023-10-02 20:55:49,515 - Epoch: [56][ 540/ 1236] Overall Loss 0.248589 Objective Loss 0.248589 LR 0.001000 Time 0.022497 +2023-10-02 20:55:49,727 - Epoch: [56][ 550/ 1236] Overall Loss 0.249019 Objective Loss 0.249019 LR 0.001000 Time 0.022472 +2023-10-02 20:55:49,945 - Epoch: [56][ 560/ 1236] Overall Loss 0.248773 Objective Loss 0.248773 LR 0.001000 Time 0.022461 +2023-10-02 20:55:50,151 - Epoch: [56][ 570/ 1236] Overall Loss 0.248502 Objective Loss 0.248502 LR 0.001000 Time 0.022425 +2023-10-02 20:55:50,360 - Epoch: [56][ 580/ 1236] Overall Loss 0.249168 Objective Loss 0.249168 LR 0.001000 Time 0.022398 +2023-10-02 20:55:50,567 - Epoch: [56][ 590/ 1236] Overall Loss 0.249074 Objective Loss 0.249074 LR 0.001000 Time 0.022369 +2023-10-02 20:55:50,777 - Epoch: [56][ 600/ 1236] Overall Loss 0.249067 Objective Loss 0.249067 LR 0.001000 Time 0.022345 +2023-10-02 20:55:50,984 - Epoch: [56][ 610/ 1236] Overall Loss 0.249115 Objective Loss 0.249115 LR 0.001000 Time 0.022318 +2023-10-02 20:55:51,193 - Epoch: [56][ 620/ 1236] Overall Loss 0.248983 Objective Loss 0.248983 LR 0.001000 Time 0.022295 +2023-10-02 20:55:51,400 - Epoch: [56][ 630/ 1236] Overall Loss 0.249224 Objective Loss 0.249224 LR 0.001000 Time 0.022269 +2023-10-02 20:55:51,610 - Epoch: [56][ 640/ 1236] Overall Loss 0.249138 Objective Loss 0.249138 LR 0.001000 Time 0.022249 +2023-10-02 20:55:51,817 - Epoch: [56][ 650/ 1236] Overall Loss 0.249549 Objective Loss 0.249549 LR 0.001000 Time 0.022225 +2023-10-02 20:55:52,026 - Epoch: [56][ 660/ 1236] Overall Loss 0.249682 Objective Loss 0.249682 LR 0.001000 Time 0.022205 +2023-10-02 20:55:52,233 - Epoch: [56][ 670/ 1236] Overall Loss 0.249446 Objective Loss 0.249446 LR 0.001000 Time 0.022181 +2023-10-02 20:55:52,442 - Epoch: [56][ 680/ 1236] Overall Loss 0.249684 Objective Loss 0.249684 LR 0.001000 Time 0.022161 +2023-10-02 20:55:52,648 - Epoch: [56][ 690/ 1236] Overall Loss 0.249632 Objective Loss 0.249632 LR 0.001000 Time 0.022140 +2023-10-02 20:55:52,858 - Epoch: [56][ 700/ 1236] Overall Loss 0.250181 Objective Loss 0.250181 LR 0.001000 Time 0.022123 +2023-10-02 20:55:53,066 - Epoch: [56][ 710/ 1236] Overall Loss 0.250502 Objective Loss 0.250502 LR 0.001000 Time 0.022103 +2023-10-02 20:55:53,275 - Epoch: [56][ 720/ 1236] Overall Loss 0.250612 Objective Loss 0.250612 LR 0.001000 Time 0.022086 +2023-10-02 20:55:53,481 - Epoch: [56][ 730/ 1236] Overall Loss 0.250378 Objective Loss 0.250378 LR 0.001000 Time 0.022066 +2023-10-02 20:55:53,689 - Epoch: [56][ 740/ 1236] Overall Loss 0.250471 Objective Loss 0.250471 LR 0.001000 Time 0.022048 +2023-10-02 20:55:53,894 - Epoch: [56][ 750/ 1236] Overall Loss 0.250406 Objective Loss 0.250406 LR 0.001000 Time 0.022028 +2023-10-02 20:55:54,102 - Epoch: [56][ 760/ 1236] Overall Loss 0.250563 Objective Loss 0.250563 LR 0.001000 Time 0.022010 +2023-10-02 20:55:54,307 - Epoch: [56][ 770/ 1236] Overall Loss 0.250827 Objective Loss 0.250827 LR 0.001000 Time 0.021991 +2023-10-02 20:55:54,514 - Epoch: [56][ 780/ 1236] Overall Loss 0.251038 Objective Loss 0.251038 LR 0.001000 Time 0.021973 +2023-10-02 20:55:54,719 - Epoch: [56][ 790/ 1236] Overall Loss 0.251468 Objective Loss 0.251468 LR 0.001000 Time 0.021955 +2023-10-02 20:55:54,926 - Epoch: [56][ 800/ 1236] Overall Loss 0.251226 Objective Loss 0.251226 LR 0.001000 Time 0.021939 +2023-10-02 20:55:55,132 - Epoch: [56][ 810/ 1236] Overall Loss 0.251483 Objective Loss 0.251483 LR 0.001000 Time 0.021922 +2023-10-02 20:55:55,339 - Epoch: [56][ 820/ 1236] Overall Loss 0.251204 Objective Loss 0.251204 LR 0.001000 Time 0.021906 +2023-10-02 20:55:55,545 - Epoch: [56][ 830/ 1236] Overall Loss 0.250844 Objective Loss 0.250844 LR 0.001000 Time 0.021891 +2023-10-02 20:55:55,754 - Epoch: [56][ 840/ 1236] Overall Loss 0.250855 Objective Loss 0.250855 LR 0.001000 Time 0.021879 +2023-10-02 20:55:55,961 - Epoch: [56][ 850/ 1236] Overall Loss 0.250875 Objective Loss 0.250875 LR 0.001000 Time 0.021865 +2023-10-02 20:55:56,170 - Epoch: [56][ 860/ 1236] Overall Loss 0.251275 Objective Loss 0.251275 LR 0.001000 Time 0.021853 +2023-10-02 20:55:56,377 - Epoch: [56][ 870/ 1236] Overall Loss 0.251467 Objective Loss 0.251467 LR 0.001000 Time 0.021840 +2023-10-02 20:55:56,587 - Epoch: [56][ 880/ 1236] Overall Loss 0.251296 Objective Loss 0.251296 LR 0.001000 Time 0.021829 +2023-10-02 20:55:56,794 - Epoch: [56][ 890/ 1236] Overall Loss 0.251550 Objective Loss 0.251550 LR 0.001000 Time 0.021816 +2023-10-02 20:55:57,003 - Epoch: [56][ 900/ 1236] Overall Loss 0.251816 Objective Loss 0.251816 LR 0.001000 Time 0.021807 +2023-10-02 20:55:57,210 - Epoch: [56][ 910/ 1236] Overall Loss 0.251884 Objective Loss 0.251884 LR 0.001000 Time 0.021794 +2023-10-02 20:55:57,419 - Epoch: [56][ 920/ 1236] Overall Loss 0.252111 Objective Loss 0.252111 LR 0.001000 Time 0.021784 +2023-10-02 20:55:57,627 - Epoch: [56][ 930/ 1236] Overall Loss 0.252035 Objective Loss 0.252035 LR 0.001000 Time 0.021773 +2023-10-02 20:55:57,836 - Epoch: [56][ 940/ 1236] Overall Loss 0.252046 Objective Loss 0.252046 LR 0.001000 Time 0.021764 +2023-10-02 20:55:58,044 - Epoch: [56][ 950/ 1236] Overall Loss 0.252070 Objective Loss 0.252070 LR 0.001000 Time 0.021753 +2023-10-02 20:55:58,253 - Epoch: [56][ 960/ 1236] Overall Loss 0.252310 Objective Loss 0.252310 LR 0.001000 Time 0.021744 +2023-10-02 20:55:58,461 - Epoch: [56][ 970/ 1236] Overall Loss 0.252260 Objective Loss 0.252260 LR 0.001000 Time 0.021734 +2023-10-02 20:55:58,670 - Epoch: [56][ 980/ 1236] Overall Loss 0.252693 Objective Loss 0.252693 LR 0.001000 Time 0.021725 +2023-10-02 20:55:58,877 - Epoch: [56][ 990/ 1236] Overall Loss 0.253056 Objective Loss 0.253056 LR 0.001000 Time 0.021714 +2023-10-02 20:55:59,084 - Epoch: [56][ 1000/ 1236] Overall Loss 0.253450 Objective Loss 0.253450 LR 0.001000 Time 0.021704 +2023-10-02 20:55:59,290 - Epoch: [56][ 1010/ 1236] Overall Loss 0.253782 Objective Loss 0.253782 LR 0.001000 Time 0.021693 +2023-10-02 20:55:59,497 - Epoch: [56][ 1020/ 1236] Overall Loss 0.254012 Objective Loss 0.254012 LR 0.001000 Time 0.021683 +2023-10-02 20:55:59,712 - Epoch: [56][ 1030/ 1236] Overall Loss 0.254410 Objective Loss 0.254410 LR 0.001000 Time 0.021680 +2023-10-02 20:55:59,926 - Epoch: [56][ 1040/ 1236] Overall Loss 0.254321 Objective Loss 0.254321 LR 0.001000 Time 0.021678 +2023-10-02 20:56:00,135 - Epoch: [56][ 1050/ 1236] Overall Loss 0.254174 Objective Loss 0.254174 LR 0.001000 Time 0.021670 +2023-10-02 20:56:00,345 - Epoch: [56][ 1060/ 1236] Overall Loss 0.254114 Objective Loss 0.254114 LR 0.001000 Time 0.021664 +2023-10-02 20:56:00,554 - Epoch: [56][ 1070/ 1236] Overall Loss 0.254086 Objective Loss 0.254086 LR 0.001000 Time 0.021656 +2023-10-02 20:56:00,765 - Epoch: [56][ 1080/ 1236] Overall Loss 0.254485 Objective Loss 0.254485 LR 0.001000 Time 0.021650 +2023-10-02 20:56:00,973 - Epoch: [56][ 1090/ 1236] Overall Loss 0.254620 Objective Loss 0.254620 LR 0.001000 Time 0.021643 +2023-10-02 20:56:01,184 - Epoch: [56][ 1100/ 1236] Overall Loss 0.254763 Objective Loss 0.254763 LR 0.001000 Time 0.021637 +2023-10-02 20:56:01,393 - Epoch: [56][ 1110/ 1236] Overall Loss 0.254784 Objective Loss 0.254784 LR 0.001000 Time 0.021630 +2023-10-02 20:56:01,603 - Epoch: [56][ 1120/ 1236] Overall Loss 0.254753 Objective Loss 0.254753 LR 0.001000 Time 0.021624 +2023-10-02 20:56:01,812 - Epoch: [56][ 1130/ 1236] Overall Loss 0.254888 Objective Loss 0.254888 LR 0.001000 Time 0.021618 +2023-10-02 20:56:02,023 - Epoch: [56][ 1140/ 1236] Overall Loss 0.254997 Objective Loss 0.254997 LR 0.001000 Time 0.021613 +2023-10-02 20:56:02,234 - Epoch: [56][ 1150/ 1236] Overall Loss 0.255246 Objective Loss 0.255246 LR 0.001000 Time 0.021608 +2023-10-02 20:56:02,444 - Epoch: [56][ 1160/ 1236] Overall Loss 0.255472 Objective Loss 0.255472 LR 0.001000 Time 0.021602 +2023-10-02 20:56:02,653 - Epoch: [56][ 1170/ 1236] Overall Loss 0.255602 Objective Loss 0.255602 LR 0.001000 Time 0.021596 +2023-10-02 20:56:02,863 - Epoch: [56][ 1180/ 1236] Overall Loss 0.255863 Objective Loss 0.255863 LR 0.001000 Time 0.021591 +2023-10-02 20:56:03,073 - Epoch: [56][ 1190/ 1236] Overall Loss 0.255850 Objective Loss 0.255850 LR 0.001000 Time 0.021585 +2023-10-02 20:56:03,283 - Epoch: [56][ 1200/ 1236] Overall Loss 0.255901 Objective Loss 0.255901 LR 0.001000 Time 0.021580 +2023-10-02 20:56:03,492 - Epoch: [56][ 1210/ 1236] Overall Loss 0.255903 Objective Loss 0.255903 LR 0.001000 Time 0.021574 +2023-10-02 20:56:03,705 - Epoch: [56][ 1220/ 1236] Overall Loss 0.255971 Objective Loss 0.255971 LR 0.001000 Time 0.021572 +2023-10-02 20:56:03,970 - Epoch: [56][ 1230/ 1236] Overall Loss 0.256073 Objective Loss 0.256073 LR 0.001000 Time 0.021612 +2023-10-02 20:56:04,093 - Epoch: [56][ 1236/ 1236] Overall Loss 0.256073 Objective Loss 0.256073 Top1 88.391039 Top5 98.370672 LR 0.001000 Time 0.021606 +2023-10-02 20:56:04,238 - --- validate (epoch=56)----------- +2023-10-02 20:56:04,238 - 29943 samples (256 per mini-batch) +2023-10-02 20:56:04,713 - Epoch: [56][ 10/ 117] Loss 0.344279 Top1 83.710938 Top5 98.359375 +2023-10-02 20:56:04,868 - Epoch: [56][ 20/ 117] Loss 0.340779 Top1 83.613281 Top5 98.339844 +2023-10-02 20:56:05,019 - Epoch: [56][ 30/ 117] Loss 0.326106 Top1 84.101562 Top5 98.463542 +2023-10-02 20:56:05,170 - Epoch: [56][ 40/ 117] Loss 0.325295 Top1 84.091797 Top5 98.486328 +2023-10-02 20:56:05,321 - Epoch: [56][ 50/ 117] Loss 0.331440 Top1 83.843750 Top5 98.390625 +2023-10-02 20:56:05,472 - Epoch: [56][ 60/ 117] Loss 0.335337 Top1 83.867188 Top5 98.385417 +2023-10-02 20:56:05,623 - Epoch: [56][ 70/ 117] Loss 0.330812 Top1 83.967634 Top5 98.392857 +2023-10-02 20:56:05,774 - Epoch: [56][ 80/ 117] Loss 0.330627 Top1 83.925781 Top5 98.344727 +2023-10-02 20:56:05,924 - Epoch: [56][ 90/ 117] Loss 0.328840 Top1 84.071181 Top5 98.355035 +2023-10-02 20:56:06,075 - Epoch: [56][ 100/ 117] Loss 0.329997 Top1 84.097656 Top5 98.382812 +2023-10-02 20:56:06,233 - Epoch: [56][ 110/ 117] Loss 0.333562 Top1 83.881392 Top5 98.348722 +2023-10-02 20:56:06,322 - Epoch: [56][ 117/ 117] Loss 0.330665 Top1 84.056374 Top5 98.383595 +2023-10-02 20:56:06,456 - ==> Top1: 84.056 Top5: 98.384 Loss: 0.331 + +2023-10-02 20:56:06,457 - ==> Confusion: +[[ 935 1 7 0 12 3 0 0 7 59 4 0 0 3 3 1 3 0 2 0 10] + [ 0 1051 1 1 3 22 3 17 0 0 2 1 1 0 2 3 2 0 17 2 3] + [ 5 0 927 17 3 0 38 7 0 1 10 2 8 3 1 3 0 1 14 2 14] + [ 1 3 7 981 0 5 1 4 4 0 7 0 10 3 28 5 1 2 10 0 17] + [ 18 9 1 2 968 7 0 0 2 10 1 0 0 4 7 7 8 0 0 1 5] + [ 3 47 0 3 1 972 2 18 2 3 7 5 5 10 8 1 3 0 9 4 13] + [ 0 6 22 0 0 0 1129 4 0 1 5 2 0 0 1 6 0 2 2 7 4] + [ 2 18 18 2 4 24 9 1042 0 4 9 7 5 4 2 0 2 1 45 12 8] + [ 17 2 0 1 2 1 0 0 986 31 10 1 2 13 13 1 2 2 2 1 2] + [ 107 0 1 0 8 1 0 0 38 906 1 1 0 32 8 2 2 0 0 3 9] + [ 4 0 12 4 2 1 4 3 7 2 973 0 1 16 5 2 1 2 5 3 6] + [ 0 2 1 0 1 10 0 1 0 0 0 917 55 10 0 3 1 16 0 8 10] + [ 1 1 3 2 1 2 1 1 1 0 0 34 957 2 2 12 1 20 5 7 15] + [ 1 0 0 0 3 11 2 0 14 6 9 7 1 1042 4 0 1 1 0 3 14] + [ 6 1 6 18 4 0 0 0 19 1 1 0 4 3 1013 0 2 4 12 0 7] + [ 0 0 1 1 4 0 4 0 0 1 0 5 11 1 0 1072 10 7 1 8 8] + [ 1 13 0 0 6 7 3 0 1 0 0 6 2 2 3 14 1080 1 3 4 15] + [ 0 0 0 2 0 0 5 0 1 0 0 1 23 1 4 6 0 989 0 1 5] + [ 1 5 2 15 1 0 0 15 7 0 4 0 3 0 10 0 1 0 992 1 11] + [ 0 3 0 3 1 4 12 12 0 1 2 12 5 2 0 4 6 0 1 1072 12] + [ 127 222 97 103 79 146 57 108 105 72 220 91 305 299 125 81 110 79 163 151 5165]] + +2023-10-02 20:56:06,458 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:56:06,458 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:56:06,465 - + +2023-10-02 20:56:06,465 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:56:07,509 - Epoch: [57][ 10/ 1236] Overall Loss 0.245523 Objective Loss 0.245523 LR 0.001000 Time 0.104335 +2023-10-02 20:56:07,718 - Epoch: [57][ 20/ 1236] Overall Loss 0.241015 Objective Loss 0.241015 LR 0.001000 Time 0.062636 +2023-10-02 20:56:07,928 - Epoch: [57][ 30/ 1236] Overall Loss 0.244096 Objective Loss 0.244096 LR 0.001000 Time 0.048745 +2023-10-02 20:56:08,139 - Epoch: [57][ 40/ 1236] Overall Loss 0.241597 Objective Loss 0.241597 LR 0.001000 Time 0.041810 +2023-10-02 20:56:08,348 - Epoch: [57][ 50/ 1236] Overall Loss 0.246325 Objective Loss 0.246325 LR 0.001000 Time 0.037609 +2023-10-02 20:56:08,558 - Epoch: [57][ 60/ 1236] Overall Loss 0.249382 Objective Loss 0.249382 LR 0.001000 Time 0.034841 +2023-10-02 20:56:08,768 - Epoch: [57][ 70/ 1236] Overall Loss 0.248792 Objective Loss 0.248792 LR 0.001000 Time 0.032839 +2023-10-02 20:56:08,980 - Epoch: [57][ 80/ 1236] Overall Loss 0.250557 Objective Loss 0.250557 LR 0.001000 Time 0.031374 +2023-10-02 20:56:09,188 - Epoch: [57][ 90/ 1236] Overall Loss 0.248939 Objective Loss 0.248939 LR 0.001000 Time 0.030204 +2023-10-02 20:56:09,399 - Epoch: [57][ 100/ 1236] Overall Loss 0.249491 Objective Loss 0.249491 LR 0.001000 Time 0.029290 +2023-10-02 20:56:09,607 - Epoch: [57][ 110/ 1236] Overall Loss 0.248552 Objective Loss 0.248552 LR 0.001000 Time 0.028515 +2023-10-02 20:56:09,817 - Epoch: [57][ 120/ 1236] Overall Loss 0.247410 Objective Loss 0.247410 LR 0.001000 Time 0.027885 +2023-10-02 20:56:10,030 - Epoch: [57][ 130/ 1236] Overall Loss 0.249859 Objective Loss 0.249859 LR 0.001000 Time 0.027369 +2023-10-02 20:56:10,241 - Epoch: [57][ 140/ 1236] Overall Loss 0.249011 Objective Loss 0.249011 LR 0.001000 Time 0.026913 +2023-10-02 20:56:10,451 - Epoch: [57][ 150/ 1236] Overall Loss 0.249704 Objective Loss 0.249704 LR 0.001000 Time 0.026509 +2023-10-02 20:56:10,661 - Epoch: [57][ 160/ 1236] Overall Loss 0.249711 Objective Loss 0.249711 LR 0.001000 Time 0.026162 +2023-10-02 20:56:10,870 - Epoch: [57][ 170/ 1236] Overall Loss 0.251006 Objective Loss 0.251006 LR 0.001000 Time 0.025845 +2023-10-02 20:56:11,080 - Epoch: [57][ 180/ 1236] Overall Loss 0.251389 Objective Loss 0.251389 LR 0.001000 Time 0.025573 +2023-10-02 20:56:11,289 - Epoch: [57][ 190/ 1236] Overall Loss 0.250639 Objective Loss 0.250639 LR 0.001000 Time 0.025323 +2023-10-02 20:56:11,498 - Epoch: [57][ 200/ 1236] Overall Loss 0.249398 Objective Loss 0.249398 LR 0.001000 Time 0.025100 +2023-10-02 20:56:11,707 - Epoch: [57][ 210/ 1236] Overall Loss 0.249974 Objective Loss 0.249974 LR 0.001000 Time 0.024891 +2023-10-02 20:56:11,916 - Epoch: [57][ 220/ 1236] Overall Loss 0.250081 Objective Loss 0.250081 LR 0.001000 Time 0.024708 +2023-10-02 20:56:12,130 - Epoch: [57][ 230/ 1236] Overall Loss 0.248340 Objective Loss 0.248340 LR 0.001000 Time 0.024559 +2023-10-02 20:56:12,341 - Epoch: [57][ 240/ 1236] Overall Loss 0.247849 Objective Loss 0.247849 LR 0.001000 Time 0.024415 +2023-10-02 20:56:12,550 - Epoch: [57][ 250/ 1236] Overall Loss 0.249216 Objective Loss 0.249216 LR 0.001000 Time 0.024270 +2023-10-02 20:56:12,759 - Epoch: [57][ 260/ 1236] Overall Loss 0.248575 Objective Loss 0.248575 LR 0.001000 Time 0.024141 +2023-10-02 20:56:12,968 - Epoch: [57][ 270/ 1236] Overall Loss 0.248275 Objective Loss 0.248275 LR 0.001000 Time 0.024014 +2023-10-02 20:56:13,177 - Epoch: [57][ 280/ 1236] Overall Loss 0.248351 Objective Loss 0.248351 LR 0.001000 Time 0.023902 +2023-10-02 20:56:13,386 - Epoch: [57][ 290/ 1236] Overall Loss 0.249545 Objective Loss 0.249545 LR 0.001000 Time 0.023793 +2023-10-02 20:56:13,597 - Epoch: [57][ 300/ 1236] Overall Loss 0.250130 Objective Loss 0.250130 LR 0.001000 Time 0.023704 +2023-10-02 20:56:13,805 - Epoch: [57][ 310/ 1236] Overall Loss 0.250636 Objective Loss 0.250636 LR 0.001000 Time 0.023604 +2023-10-02 20:56:14,015 - Epoch: [57][ 320/ 1236] Overall Loss 0.250984 Objective Loss 0.250984 LR 0.001000 Time 0.023522 +2023-10-02 20:56:14,223 - Epoch: [57][ 330/ 1236] Overall Loss 0.251039 Objective Loss 0.251039 LR 0.001000 Time 0.023435 +2023-10-02 20:56:14,433 - Epoch: [57][ 340/ 1236] Overall Loss 0.251891 Objective Loss 0.251891 LR 0.001000 Time 0.023363 +2023-10-02 20:56:14,642 - Epoch: [57][ 350/ 1236] Overall Loss 0.252104 Objective Loss 0.252104 LR 0.001000 Time 0.023286 +2023-10-02 20:56:14,851 - Epoch: [57][ 360/ 1236] Overall Loss 0.251596 Objective Loss 0.251596 LR 0.001000 Time 0.023221 +2023-10-02 20:56:15,061 - Epoch: [57][ 370/ 1236] Overall Loss 0.251843 Objective Loss 0.251843 LR 0.001000 Time 0.023154 +2023-10-02 20:56:15,271 - Epoch: [57][ 380/ 1236] Overall Loss 0.252179 Objective Loss 0.252179 LR 0.001000 Time 0.023097 +2023-10-02 20:56:15,479 - Epoch: [57][ 390/ 1236] Overall Loss 0.252498 Objective Loss 0.252498 LR 0.001000 Time 0.023035 +2023-10-02 20:56:15,689 - Epoch: [57][ 400/ 1236] Overall Loss 0.253186 Objective Loss 0.253186 LR 0.001000 Time 0.022983 +2023-10-02 20:56:15,897 - Epoch: [57][ 410/ 1236] Overall Loss 0.252850 Objective Loss 0.252850 LR 0.001000 Time 0.022927 +2023-10-02 20:56:16,107 - Epoch: [57][ 420/ 1236] Overall Loss 0.253467 Objective Loss 0.253467 LR 0.001000 Time 0.022881 +2023-10-02 20:56:16,316 - Epoch: [57][ 430/ 1236] Overall Loss 0.253047 Objective Loss 0.253047 LR 0.001000 Time 0.022829 +2023-10-02 20:56:16,526 - Epoch: [57][ 440/ 1236] Overall Loss 0.253416 Objective Loss 0.253416 LR 0.001000 Time 0.022789 +2023-10-02 20:56:16,736 - Epoch: [57][ 450/ 1236] Overall Loss 0.253006 Objective Loss 0.253006 LR 0.001000 Time 0.022744 +2023-10-02 20:56:16,946 - Epoch: [57][ 460/ 1236] Overall Loss 0.252623 Objective Loss 0.252623 LR 0.001000 Time 0.022705 +2023-10-02 20:56:17,154 - Epoch: [57][ 470/ 1236] Overall Loss 0.252378 Objective Loss 0.252378 LR 0.001000 Time 0.022662 +2023-10-02 20:56:17,364 - Epoch: [57][ 480/ 1236] Overall Loss 0.252923 Objective Loss 0.252923 LR 0.001000 Time 0.022627 +2023-10-02 20:56:17,573 - Epoch: [57][ 490/ 1236] Overall Loss 0.253787 Objective Loss 0.253787 LR 0.001000 Time 0.022588 +2023-10-02 20:56:17,783 - Epoch: [57][ 500/ 1236] Overall Loss 0.254763 Objective Loss 0.254763 LR 0.001000 Time 0.022556 +2023-10-02 20:56:17,992 - Epoch: [57][ 510/ 1236] Overall Loss 0.254612 Objective Loss 0.254612 LR 0.001000 Time 0.022520 +2023-10-02 20:56:18,202 - Epoch: [57][ 520/ 1236] Overall Loss 0.254439 Objective Loss 0.254439 LR 0.001000 Time 0.022490 +2023-10-02 20:56:18,411 - Epoch: [57][ 530/ 1236] Overall Loss 0.254575 Objective Loss 0.254575 LR 0.001000 Time 0.022459 +2023-10-02 20:56:18,621 - Epoch: [57][ 540/ 1236] Overall Loss 0.253800 Objective Loss 0.253800 LR 0.001000 Time 0.022432 +2023-10-02 20:56:18,830 - Epoch: [57][ 550/ 1236] Overall Loss 0.253949 Objective Loss 0.253949 LR 0.001000 Time 0.022403 +2023-10-02 20:56:19,041 - Epoch: [57][ 560/ 1236] Overall Loss 0.254153 Objective Loss 0.254153 LR 0.001000 Time 0.022380 +2023-10-02 20:56:19,250 - Epoch: [57][ 570/ 1236] Overall Loss 0.254407 Objective Loss 0.254407 LR 0.001000 Time 0.022353 +2023-10-02 20:56:19,460 - Epoch: [57][ 580/ 1236] Overall Loss 0.254868 Objective Loss 0.254868 LR 0.001000 Time 0.022330 +2023-10-02 20:56:19,669 - Epoch: [57][ 590/ 1236] Overall Loss 0.254798 Objective Loss 0.254798 LR 0.001000 Time 0.022305 +2023-10-02 20:56:19,879 - Epoch: [57][ 600/ 1236] Overall Loss 0.254921 Objective Loss 0.254921 LR 0.001000 Time 0.022283 +2023-10-02 20:56:20,088 - Epoch: [57][ 610/ 1236] Overall Loss 0.255268 Objective Loss 0.255268 LR 0.001000 Time 0.022258 +2023-10-02 20:56:20,298 - Epoch: [57][ 620/ 1236] Overall Loss 0.255331 Objective Loss 0.255331 LR 0.001000 Time 0.022237 +2023-10-02 20:56:20,507 - Epoch: [57][ 630/ 1236] Overall Loss 0.255813 Objective Loss 0.255813 LR 0.001000 Time 0.022213 +2023-10-02 20:56:20,717 - Epoch: [57][ 640/ 1236] Overall Loss 0.255934 Objective Loss 0.255934 LR 0.001000 Time 0.022194 +2023-10-02 20:56:20,926 - Epoch: [57][ 650/ 1236] Overall Loss 0.255986 Objective Loss 0.255986 LR 0.001000 Time 0.022171 +2023-10-02 20:56:21,136 - Epoch: [57][ 660/ 1236] Overall Loss 0.255570 Objective Loss 0.255570 LR 0.001000 Time 0.022152 +2023-10-02 20:56:21,344 - Epoch: [57][ 670/ 1236] Overall Loss 0.255407 Objective Loss 0.255407 LR 0.001000 Time 0.022131 +2023-10-02 20:56:21,555 - Epoch: [57][ 680/ 1236] Overall Loss 0.255468 Objective Loss 0.255468 LR 0.001000 Time 0.022114 +2023-10-02 20:56:21,764 - Epoch: [57][ 690/ 1236] Overall Loss 0.255583 Objective Loss 0.255583 LR 0.001000 Time 0.022094 +2023-10-02 20:56:21,974 - Epoch: [57][ 700/ 1236] Overall Loss 0.255523 Objective Loss 0.255523 LR 0.001000 Time 0.022078 +2023-10-02 20:56:22,182 - Epoch: [57][ 710/ 1236] Overall Loss 0.255239 Objective Loss 0.255239 LR 0.001000 Time 0.022059 +2023-10-02 20:56:22,393 - Epoch: [57][ 720/ 1236] Overall Loss 0.255302 Objective Loss 0.255302 LR 0.001000 Time 0.022044 +2023-10-02 20:56:22,602 - Epoch: [57][ 730/ 1236] Overall Loss 0.255549 Objective Loss 0.255549 LR 0.001000 Time 0.022026 +2023-10-02 20:56:22,812 - Epoch: [57][ 740/ 1236] Overall Loss 0.255794 Objective Loss 0.255794 LR 0.001000 Time 0.022013 +2023-10-02 20:56:23,021 - Epoch: [57][ 750/ 1236] Overall Loss 0.256061 Objective Loss 0.256061 LR 0.001000 Time 0.021996 +2023-10-02 20:56:23,231 - Epoch: [57][ 760/ 1236] Overall Loss 0.255985 Objective Loss 0.255985 LR 0.001000 Time 0.021982 +2023-10-02 20:56:23,440 - Epoch: [57][ 770/ 1236] Overall Loss 0.256342 Objective Loss 0.256342 LR 0.001000 Time 0.021966 +2023-10-02 20:56:23,650 - Epoch: [57][ 780/ 1236] Overall Loss 0.256649 Objective Loss 0.256649 LR 0.001000 Time 0.021953 +2023-10-02 20:56:23,859 - Epoch: [57][ 790/ 1236] Overall Loss 0.256741 Objective Loss 0.256741 LR 0.001000 Time 0.021938 +2023-10-02 20:56:24,069 - Epoch: [57][ 800/ 1236] Overall Loss 0.256553 Objective Loss 0.256553 LR 0.001000 Time 0.021926 +2023-10-02 20:56:24,279 - Epoch: [57][ 810/ 1236] Overall Loss 0.256222 Objective Loss 0.256222 LR 0.001000 Time 0.021912 +2023-10-02 20:56:24,489 - Epoch: [57][ 820/ 1236] Overall Loss 0.256373 Objective Loss 0.256373 LR 0.001000 Time 0.021901 +2023-10-02 20:56:24,698 - Epoch: [57][ 830/ 1236] Overall Loss 0.256347 Objective Loss 0.256347 LR 0.001000 Time 0.021887 +2023-10-02 20:56:24,908 - Epoch: [57][ 840/ 1236] Overall Loss 0.256433 Objective Loss 0.256433 LR 0.001000 Time 0.021876 +2023-10-02 20:56:25,116 - Epoch: [57][ 850/ 1236] Overall Loss 0.256084 Objective Loss 0.256084 LR 0.001000 Time 0.021862 +2023-10-02 20:56:25,326 - Epoch: [57][ 860/ 1236] Overall Loss 0.256046 Objective Loss 0.256046 LR 0.001000 Time 0.021852 +2023-10-02 20:56:25,535 - Epoch: [57][ 870/ 1236] Overall Loss 0.256180 Objective Loss 0.256180 LR 0.001000 Time 0.021839 +2023-10-02 20:56:25,745 - Epoch: [57][ 880/ 1236] Overall Loss 0.256488 Objective Loss 0.256488 LR 0.001000 Time 0.021829 +2023-10-02 20:56:25,954 - Epoch: [57][ 890/ 1236] Overall Loss 0.256436 Objective Loss 0.256436 LR 0.001000 Time 0.021817 +2023-10-02 20:56:26,165 - Epoch: [57][ 900/ 1236] Overall Loss 0.256454 Objective Loss 0.256454 LR 0.001000 Time 0.021808 +2023-10-02 20:56:26,374 - Epoch: [57][ 910/ 1236] Overall Loss 0.256279 Objective Loss 0.256279 LR 0.001000 Time 0.021796 +2023-10-02 20:56:26,584 - Epoch: [57][ 920/ 1236] Overall Loss 0.256302 Objective Loss 0.256302 LR 0.001000 Time 0.021787 +2023-10-02 20:56:26,792 - Epoch: [57][ 930/ 1236] Overall Loss 0.256278 Objective Loss 0.256278 LR 0.001000 Time 0.021776 +2023-10-02 20:56:27,002 - Epoch: [57][ 940/ 1236] Overall Loss 0.256419 Objective Loss 0.256419 LR 0.001000 Time 0.021767 +2023-10-02 20:56:27,211 - Epoch: [57][ 950/ 1236] Overall Loss 0.256403 Objective Loss 0.256403 LR 0.001000 Time 0.021758 +2023-10-02 20:56:27,421 - Epoch: [57][ 960/ 1236] Overall Loss 0.256465 Objective Loss 0.256465 LR 0.001000 Time 0.021749 +2023-10-02 20:56:27,630 - Epoch: [57][ 970/ 1236] Overall Loss 0.256523 Objective Loss 0.256523 LR 0.001000 Time 0.021739 +2023-10-02 20:56:27,840 - Epoch: [57][ 980/ 1236] Overall Loss 0.256559 Objective Loss 0.256559 LR 0.001000 Time 0.021731 +2023-10-02 20:56:28,049 - Epoch: [57][ 990/ 1236] Overall Loss 0.256569 Objective Loss 0.256569 LR 0.001000 Time 0.021721 +2023-10-02 20:56:28,259 - Epoch: [57][ 1000/ 1236] Overall Loss 0.256336 Objective Loss 0.256336 LR 0.001000 Time 0.021713 +2023-10-02 20:56:28,468 - Epoch: [57][ 1010/ 1236] Overall Loss 0.256268 Objective Loss 0.256268 LR 0.001000 Time 0.021703 +2023-10-02 20:56:28,678 - Epoch: [57][ 1020/ 1236] Overall Loss 0.256205 Objective Loss 0.256205 LR 0.001000 Time 0.021696 +2023-10-02 20:56:28,887 - Epoch: [57][ 1030/ 1236] Overall Loss 0.256239 Objective Loss 0.256239 LR 0.001000 Time 0.021687 +2023-10-02 20:56:29,097 - Epoch: [57][ 1040/ 1236] Overall Loss 0.256290 Objective Loss 0.256290 LR 0.001000 Time 0.021680 +2023-10-02 20:56:29,305 - Epoch: [57][ 1050/ 1236] Overall Loss 0.256531 Objective Loss 0.256531 LR 0.001000 Time 0.021671 +2023-10-02 20:56:29,515 - Epoch: [57][ 1060/ 1236] Overall Loss 0.256535 Objective Loss 0.256535 LR 0.001000 Time 0.021664 +2023-10-02 20:56:29,724 - Epoch: [57][ 1070/ 1236] Overall Loss 0.256590 Objective Loss 0.256590 LR 0.001000 Time 0.021656 +2023-10-02 20:56:29,934 - Epoch: [57][ 1080/ 1236] Overall Loss 0.256548 Objective Loss 0.256548 LR 0.001000 Time 0.021649 +2023-10-02 20:56:30,143 - Epoch: [57][ 1090/ 1236] Overall Loss 0.256521 Objective Loss 0.256521 LR 0.001000 Time 0.021641 +2023-10-02 20:56:30,353 - Epoch: [57][ 1100/ 1236] Overall Loss 0.256582 Objective Loss 0.256582 LR 0.001000 Time 0.021635 +2023-10-02 20:56:30,562 - Epoch: [57][ 1110/ 1236] Overall Loss 0.256646 Objective Loss 0.256646 LR 0.001000 Time 0.021627 +2023-10-02 20:56:30,772 - Epoch: [57][ 1120/ 1236] Overall Loss 0.256591 Objective Loss 0.256591 LR 0.001000 Time 0.021621 +2023-10-02 20:56:30,981 - Epoch: [57][ 1130/ 1236] Overall Loss 0.256542 Objective Loss 0.256542 LR 0.001000 Time 0.021613 +2023-10-02 20:56:31,191 - Epoch: [57][ 1140/ 1236] Overall Loss 0.256462 Objective Loss 0.256462 LR 0.001000 Time 0.021608 +2023-10-02 20:56:31,400 - Epoch: [57][ 1150/ 1236] Overall Loss 0.256582 Objective Loss 0.256582 LR 0.001000 Time 0.021600 +2023-10-02 20:56:31,610 - Epoch: [57][ 1160/ 1236] Overall Loss 0.256649 Objective Loss 0.256649 LR 0.001000 Time 0.021594 +2023-10-02 20:56:31,819 - Epoch: [57][ 1170/ 1236] Overall Loss 0.256906 Objective Loss 0.256906 LR 0.001000 Time 0.021587 +2023-10-02 20:56:32,029 - Epoch: [57][ 1180/ 1236] Overall Loss 0.256927 Objective Loss 0.256927 LR 0.001000 Time 0.021582 +2023-10-02 20:56:32,238 - Epoch: [57][ 1190/ 1236] Overall Loss 0.257104 Objective Loss 0.257104 LR 0.001000 Time 0.021574 +2023-10-02 20:56:32,448 - Epoch: [57][ 1200/ 1236] Overall Loss 0.257204 Objective Loss 0.257204 LR 0.001000 Time 0.021569 +2023-10-02 20:56:32,656 - Epoch: [57][ 1210/ 1236] Overall Loss 0.257252 Objective Loss 0.257252 LR 0.001000 Time 0.021562 +2023-10-02 20:56:32,867 - Epoch: [57][ 1220/ 1236] Overall Loss 0.257222 Objective Loss 0.257222 LR 0.001000 Time 0.021558 +2023-10-02 20:56:33,130 - Epoch: [57][ 1230/ 1236] Overall Loss 0.257307 Objective Loss 0.257307 LR 0.001000 Time 0.021595 +2023-10-02 20:56:33,253 - Epoch: [57][ 1236/ 1236] Overall Loss 0.257375 Objective Loss 0.257375 Top1 87.780041 Top5 98.778004 LR 0.001000 Time 0.021590 +2023-10-02 20:56:33,394 - --- validate (epoch=57)----------- +2023-10-02 20:56:33,394 - 29943 samples (256 per mini-batch) +2023-10-02 20:56:33,894 - Epoch: [57][ 10/ 117] Loss 0.298464 Top1 84.960938 Top5 98.476562 +2023-10-02 20:56:34,051 - Epoch: [57][ 20/ 117] Loss 0.305434 Top1 84.277344 Top5 98.496094 +2023-10-02 20:56:34,207 - Epoch: [57][ 30/ 117] Loss 0.316179 Top1 84.114583 Top5 98.424479 +2023-10-02 20:56:34,363 - Epoch: [57][ 40/ 117] Loss 0.309556 Top1 84.296875 Top5 98.457031 +2023-10-02 20:56:34,519 - Epoch: [57][ 50/ 117] Loss 0.318455 Top1 84.164062 Top5 98.257812 +2023-10-02 20:56:34,675 - Epoch: [57][ 60/ 117] Loss 0.323351 Top1 83.893229 Top5 98.216146 +2023-10-02 20:56:34,831 - Epoch: [57][ 70/ 117] Loss 0.328192 Top1 83.978795 Top5 98.197545 +2023-10-02 20:56:34,987 - Epoch: [57][ 80/ 117] Loss 0.324836 Top1 84.086914 Top5 98.251953 +2023-10-02 20:56:35,145 - Epoch: [57][ 90/ 117] Loss 0.326711 Top1 84.153646 Top5 98.220486 +2023-10-02 20:56:35,301 - Epoch: [57][ 100/ 117] Loss 0.329200 Top1 84.097656 Top5 98.218750 +2023-10-02 20:56:35,464 - Epoch: [57][ 110/ 117] Loss 0.326789 Top1 84.147727 Top5 98.252841 +2023-10-02 20:56:35,554 - Epoch: [57][ 117/ 117] Loss 0.330217 Top1 84.133186 Top5 98.243329 +2023-10-02 20:56:35,679 - ==> Top1: 84.133 Top5: 98.243 Loss: 0.330 + +2023-10-02 20:56:35,679 - ==> Confusion: +[[ 926 1 1 2 5 1 0 0 8 72 2 0 1 2 5 3 5 1 0 0 15] + [ 0 1061 2 4 2 17 0 20 2 0 1 1 0 2 0 5 1 0 7 2 4] + [ 3 1 983 10 2 1 16 4 0 1 2 1 6 1 1 4 1 2 5 4 8] + [ 1 4 11 975 0 3 1 2 2 0 8 0 1 3 24 4 2 4 21 1 22] + [ 26 9 1 0 949 4 0 0 0 8 2 0 1 6 8 4 22 1 0 1 8] + [ 2 69 1 4 1 915 2 38 7 5 2 8 4 23 7 3 2 0 6 2 15] + [ 0 6 31 0 0 0 1117 5 0 0 7 1 0 0 0 9 0 1 1 6 7] + [ 4 21 24 1 2 22 1 1050 1 4 6 10 1 3 0 0 3 0 45 9 11] + [ 20 2 1 1 3 3 0 0 975 32 12 2 1 9 12 2 6 4 0 0 4] + [ 78 0 1 0 3 1 0 0 29 963 3 0 0 18 5 1 1 1 0 4 11] + [ 3 2 9 6 0 1 1 4 9 1 977 2 1 8 3 4 4 0 6 1 11] + [ 1 1 5 0 0 10 0 6 0 3 1 914 34 8 0 5 5 27 0 9 6] + [ 1 0 4 2 2 2 1 1 0 0 3 29 949 3 2 13 2 33 4 7 10] + [ 1 0 1 0 2 4 1 1 11 12 4 4 0 1048 7 2 1 3 0 0 17] + [ 11 0 5 19 0 0 0 0 23 4 3 0 3 4 1001 0 1 4 6 1 16] + [ 1 0 2 1 7 0 2 0 0 0 0 4 9 1 0 1061 12 18 3 5 8] + [ 0 14 2 0 4 5 0 1 3 0 0 2 1 3 1 14 1092 0 3 6 10] + [ 0 0 3 1 0 0 1 0 3 0 0 0 14 0 2 7 2 1002 0 0 3] + [ 1 6 9 10 0 0 0 24 4 0 10 1 0 0 12 0 1 0 976 1 13] + [ 0 3 5 1 0 8 14 20 0 0 0 11 3 0 0 5 7 1 2 1063 9] + [ 136 179 151 65 88 109 39 95 111 82 192 94 325 255 117 66 188 81 140 197 5195]] + +2023-10-02 20:56:35,681 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:56:35,681 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:56:35,687 - + +2023-10-02 20:56:35,687 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:56:36,824 - Epoch: [58][ 10/ 1236] Overall Loss 0.244496 Objective Loss 0.244496 LR 0.001000 Time 0.113650 +2023-10-02 20:56:37,036 - Epoch: [58][ 20/ 1236] Overall Loss 0.266012 Objective Loss 0.266012 LR 0.001000 Time 0.067390 +2023-10-02 20:56:37,246 - Epoch: [58][ 30/ 1236] Overall Loss 0.273644 Objective Loss 0.273644 LR 0.001000 Time 0.051908 +2023-10-02 20:56:37,457 - Epoch: [58][ 40/ 1236] Overall Loss 0.271897 Objective Loss 0.271897 LR 0.001000 Time 0.044216 +2023-10-02 20:56:37,669 - Epoch: [58][ 50/ 1236] Overall Loss 0.267441 Objective Loss 0.267441 LR 0.001000 Time 0.039586 +2023-10-02 20:56:37,881 - Epoch: [58][ 60/ 1236] Overall Loss 0.269443 Objective Loss 0.269443 LR 0.001000 Time 0.036527 +2023-10-02 20:56:38,091 - Epoch: [58][ 70/ 1236] Overall Loss 0.266973 Objective Loss 0.266973 LR 0.001000 Time 0.034300 +2023-10-02 20:56:38,304 - Epoch: [58][ 80/ 1236] Overall Loss 0.263369 Objective Loss 0.263369 LR 0.001000 Time 0.032665 +2023-10-02 20:56:38,515 - Epoch: [58][ 90/ 1236] Overall Loss 0.263751 Objective Loss 0.263751 LR 0.001000 Time 0.031377 +2023-10-02 20:56:38,728 - Epoch: [58][ 100/ 1236] Overall Loss 0.261015 Objective Loss 0.261015 LR 0.001000 Time 0.030359 +2023-10-02 20:56:38,941 - Epoch: [58][ 110/ 1236] Overall Loss 0.259441 Objective Loss 0.259441 LR 0.001000 Time 0.029531 +2023-10-02 20:56:39,157 - Epoch: [58][ 120/ 1236] Overall Loss 0.257438 Objective Loss 0.257438 LR 0.001000 Time 0.028861 +2023-10-02 20:56:39,371 - Epoch: [58][ 130/ 1236] Overall Loss 0.257472 Objective Loss 0.257472 LR 0.001000 Time 0.028288 +2023-10-02 20:56:39,587 - Epoch: [58][ 140/ 1236] Overall Loss 0.257241 Objective Loss 0.257241 LR 0.001000 Time 0.027796 +2023-10-02 20:56:39,802 - Epoch: [58][ 150/ 1236] Overall Loss 0.259083 Objective Loss 0.259083 LR 0.001000 Time 0.027364 +2023-10-02 20:56:40,016 - Epoch: [58][ 160/ 1236] Overall Loss 0.259202 Objective Loss 0.259202 LR 0.001000 Time 0.026981 +2023-10-02 20:56:40,227 - Epoch: [58][ 170/ 1236] Overall Loss 0.260934 Objective Loss 0.260934 LR 0.001000 Time 0.026634 +2023-10-02 20:56:40,439 - Epoch: [58][ 180/ 1236] Overall Loss 0.260250 Objective Loss 0.260250 LR 0.001000 Time 0.026328 +2023-10-02 20:56:40,651 - Epoch: [58][ 190/ 1236] Overall Loss 0.259775 Objective Loss 0.259775 LR 0.001000 Time 0.026053 +2023-10-02 20:56:40,862 - Epoch: [58][ 200/ 1236] Overall Loss 0.258899 Objective Loss 0.258899 LR 0.001000 Time 0.025797 +2023-10-02 20:56:41,069 - Epoch: [58][ 210/ 1236] Overall Loss 0.258801 Objective Loss 0.258801 LR 0.001000 Time 0.025553 +2023-10-02 20:56:41,275 - Epoch: [58][ 220/ 1236] Overall Loss 0.258152 Objective Loss 0.258152 LR 0.001000 Time 0.025325 +2023-10-02 20:56:41,483 - Epoch: [58][ 230/ 1236] Overall Loss 0.259269 Objective Loss 0.259269 LR 0.001000 Time 0.025125 +2023-10-02 20:56:41,688 - Epoch: [58][ 240/ 1236] Overall Loss 0.258265 Objective Loss 0.258265 LR 0.001000 Time 0.024933 +2023-10-02 20:56:41,894 - Epoch: [58][ 250/ 1236] Overall Loss 0.258321 Objective Loss 0.258321 LR 0.001000 Time 0.024759 +2023-10-02 20:56:42,101 - Epoch: [58][ 260/ 1236] Overall Loss 0.257987 Objective Loss 0.257987 LR 0.001000 Time 0.024600 +2023-10-02 20:56:42,309 - Epoch: [58][ 270/ 1236] Overall Loss 0.257547 Objective Loss 0.257547 LR 0.001000 Time 0.024461 +2023-10-02 20:56:42,515 - Epoch: [58][ 280/ 1236] Overall Loss 0.258207 Objective Loss 0.258207 LR 0.001000 Time 0.024320 +2023-10-02 20:56:42,722 - Epoch: [58][ 290/ 1236] Overall Loss 0.258768 Objective Loss 0.258768 LR 0.001000 Time 0.024195 +2023-10-02 20:56:42,928 - Epoch: [58][ 300/ 1236] Overall Loss 0.258788 Objective Loss 0.258788 LR 0.001000 Time 0.024073 +2023-10-02 20:56:43,136 - Epoch: [58][ 310/ 1236] Overall Loss 0.258819 Objective Loss 0.258819 LR 0.001000 Time 0.023966 +2023-10-02 20:56:43,343 - Epoch: [58][ 320/ 1236] Overall Loss 0.258522 Objective Loss 0.258522 LR 0.001000 Time 0.023863 +2023-10-02 20:56:43,550 - Epoch: [58][ 330/ 1236] Overall Loss 0.258814 Objective Loss 0.258814 LR 0.001000 Time 0.023766 +2023-10-02 20:56:43,756 - Epoch: [58][ 340/ 1236] Overall Loss 0.258873 Objective Loss 0.258873 LR 0.001000 Time 0.023673 +2023-10-02 20:56:43,963 - Epoch: [58][ 350/ 1236] Overall Loss 0.259499 Objective Loss 0.259499 LR 0.001000 Time 0.023588 +2023-10-02 20:56:44,171 - Epoch: [58][ 360/ 1236] Overall Loss 0.260105 Objective Loss 0.260105 LR 0.001000 Time 0.023509 +2023-10-02 20:56:44,392 - Epoch: [58][ 370/ 1236] Overall Loss 0.260912 Objective Loss 0.260912 LR 0.001000 Time 0.023469 +2023-10-02 20:56:44,608 - Epoch: [58][ 380/ 1236] Overall Loss 0.261003 Objective Loss 0.261003 LR 0.001000 Time 0.023419 +2023-10-02 20:56:44,827 - Epoch: [58][ 390/ 1236] Overall Loss 0.261750 Objective Loss 0.261750 LR 0.001000 Time 0.023380 +2023-10-02 20:56:45,044 - Epoch: [58][ 400/ 1236] Overall Loss 0.261762 Objective Loss 0.261762 LR 0.001000 Time 0.023335 +2023-10-02 20:56:45,263 - Epoch: [58][ 410/ 1236] Overall Loss 0.262010 Objective Loss 0.262010 LR 0.001000 Time 0.023301 +2023-10-02 20:56:45,481 - Epoch: [58][ 420/ 1236] Overall Loss 0.261657 Objective Loss 0.261657 LR 0.001000 Time 0.023263 +2023-10-02 20:56:45,701 - Epoch: [58][ 430/ 1236] Overall Loss 0.262613 Objective Loss 0.262613 LR 0.001000 Time 0.023233 +2023-10-02 20:56:45,917 - Epoch: [58][ 440/ 1236] Overall Loss 0.262885 Objective Loss 0.262885 LR 0.001000 Time 0.023195 +2023-10-02 20:56:46,137 - Epoch: [58][ 450/ 1236] Overall Loss 0.263498 Objective Loss 0.263498 LR 0.001000 Time 0.023167 +2023-10-02 20:56:46,352 - Epoch: [58][ 460/ 1236] Overall Loss 0.263522 Objective Loss 0.263522 LR 0.001000 Time 0.023131 +2023-10-02 20:56:46,571 - Epoch: [58][ 470/ 1236] Overall Loss 0.264290 Objective Loss 0.264290 LR 0.001000 Time 0.023105 +2023-10-02 20:56:46,787 - Epoch: [58][ 480/ 1236] Overall Loss 0.264518 Objective Loss 0.264518 LR 0.001000 Time 0.023072 +2023-10-02 20:56:47,007 - Epoch: [58][ 490/ 1236] Overall Loss 0.263681 Objective Loss 0.263681 LR 0.001000 Time 0.023048 +2023-10-02 20:56:47,224 - Epoch: [58][ 500/ 1236] Overall Loss 0.263620 Objective Loss 0.263620 LR 0.001000 Time 0.023021 +2023-10-02 20:56:47,443 - Epoch: [58][ 510/ 1236] Overall Loss 0.264441 Objective Loss 0.264441 LR 0.001000 Time 0.022996 +2023-10-02 20:56:47,659 - Epoch: [58][ 520/ 1236] Overall Loss 0.264317 Objective Loss 0.264317 LR 0.001000 Time 0.022967 +2023-10-02 20:56:47,878 - Epoch: [58][ 530/ 1236] Overall Loss 0.264541 Objective Loss 0.264541 LR 0.001000 Time 0.022947 +2023-10-02 20:56:48,095 - Epoch: [58][ 540/ 1236] Overall Loss 0.265051 Objective Loss 0.265051 LR 0.001000 Time 0.022922 +2023-10-02 20:56:48,314 - Epoch: [58][ 550/ 1236] Overall Loss 0.265044 Objective Loss 0.265044 LR 0.001000 Time 0.022902 +2023-10-02 20:56:48,530 - Epoch: [58][ 560/ 1236] Overall Loss 0.265049 Objective Loss 0.265049 LR 0.001000 Time 0.022877 +2023-10-02 20:56:48,749 - Epoch: [58][ 570/ 1236] Overall Loss 0.265193 Objective Loss 0.265193 LR 0.001000 Time 0.022859 +2023-10-02 20:56:48,965 - Epoch: [58][ 580/ 1236] Overall Loss 0.265399 Objective Loss 0.265399 LR 0.001000 Time 0.022837 +2023-10-02 20:56:49,186 - Epoch: [58][ 590/ 1236] Overall Loss 0.265185 Objective Loss 0.265185 LR 0.001000 Time 0.022824 +2023-10-02 20:56:49,402 - Epoch: [58][ 600/ 1236] Overall Loss 0.265164 Objective Loss 0.265164 LR 0.001000 Time 0.022803 +2023-10-02 20:56:49,621 - Epoch: [58][ 610/ 1236] Overall Loss 0.265102 Objective Loss 0.265102 LR 0.001000 Time 0.022788 +2023-10-02 20:56:49,837 - Epoch: [58][ 620/ 1236] Overall Loss 0.264874 Objective Loss 0.264874 LR 0.001000 Time 0.022767 +2023-10-02 20:56:50,056 - Epoch: [58][ 630/ 1236] Overall Loss 0.265376 Objective Loss 0.265376 LR 0.001000 Time 0.022753 +2023-10-02 20:56:50,272 - Epoch: [58][ 640/ 1236] Overall Loss 0.265024 Objective Loss 0.265024 LR 0.001000 Time 0.022734 +2023-10-02 20:56:50,491 - Epoch: [58][ 650/ 1236] Overall Loss 0.265393 Objective Loss 0.265393 LR 0.001000 Time 0.022721 +2023-10-02 20:56:50,704 - Epoch: [58][ 660/ 1236] Overall Loss 0.265018 Objective Loss 0.265018 LR 0.001000 Time 0.022699 +2023-10-02 20:56:50,914 - Epoch: [58][ 670/ 1236] Overall Loss 0.264888 Objective Loss 0.264888 LR 0.001000 Time 0.022673 +2023-10-02 20:56:51,124 - Epoch: [58][ 680/ 1236] Overall Loss 0.265059 Objective Loss 0.265059 LR 0.001000 Time 0.022647 +2023-10-02 20:56:51,333 - Epoch: [58][ 690/ 1236] Overall Loss 0.264903 Objective Loss 0.264903 LR 0.001000 Time 0.022622 +2023-10-02 20:56:51,542 - Epoch: [58][ 700/ 1236] Overall Loss 0.264561 Objective Loss 0.264561 LR 0.001000 Time 0.022597 +2023-10-02 20:56:51,751 - Epoch: [58][ 710/ 1236] Overall Loss 0.264717 Objective Loss 0.264717 LR 0.001000 Time 0.022572 +2023-10-02 20:56:51,961 - Epoch: [58][ 720/ 1236] Overall Loss 0.264699 Objective Loss 0.264699 LR 0.001000 Time 0.022549 +2023-10-02 20:56:52,170 - Epoch: [58][ 730/ 1236] Overall Loss 0.265234 Objective Loss 0.265234 LR 0.001000 Time 0.022527 +2023-10-02 20:56:52,379 - Epoch: [58][ 740/ 1236] Overall Loss 0.264810 Objective Loss 0.264810 LR 0.001000 Time 0.022505 +2023-10-02 20:56:52,589 - Epoch: [58][ 750/ 1236] Overall Loss 0.264716 Objective Loss 0.264716 LR 0.001000 Time 0.022484 +2023-10-02 20:56:52,798 - Epoch: [58][ 760/ 1236] Overall Loss 0.264923 Objective Loss 0.264923 LR 0.001000 Time 0.022463 +2023-10-02 20:56:53,008 - Epoch: [58][ 770/ 1236] Overall Loss 0.265279 Objective Loss 0.265279 LR 0.001000 Time 0.022443 +2023-10-02 20:56:53,217 - Epoch: [58][ 780/ 1236] Overall Loss 0.265399 Objective Loss 0.265399 LR 0.001000 Time 0.022423 +2023-10-02 20:56:53,431 - Epoch: [58][ 790/ 1236] Overall Loss 0.265144 Objective Loss 0.265144 LR 0.001000 Time 0.022402 +2023-10-02 20:56:53,640 - Epoch: [58][ 800/ 1236] Overall Loss 0.265002 Objective Loss 0.265002 LR 0.001000 Time 0.022383 +2023-10-02 20:56:53,850 - Epoch: [58][ 810/ 1236] Overall Loss 0.265168 Objective Loss 0.265168 LR 0.001000 Time 0.022365 +2023-10-02 20:56:54,059 - Epoch: [58][ 820/ 1236] Overall Loss 0.265328 Objective Loss 0.265328 LR 0.001000 Time 0.022347 +2023-10-02 20:56:54,269 - Epoch: [58][ 830/ 1236] Overall Loss 0.265516 Objective Loss 0.265516 LR 0.001000 Time 0.022329 +2023-10-02 20:56:54,478 - Epoch: [58][ 840/ 1236] Overall Loss 0.265452 Objective Loss 0.265452 LR 0.001000 Time 0.022313 +2023-10-02 20:56:54,687 - Epoch: [58][ 850/ 1236] Overall Loss 0.265455 Objective Loss 0.265455 LR 0.001000 Time 0.022296 +2023-10-02 20:56:54,897 - Epoch: [58][ 860/ 1236] Overall Loss 0.265545 Objective Loss 0.265545 LR 0.001000 Time 0.022280 +2023-10-02 20:56:55,107 - Epoch: [58][ 870/ 1236] Overall Loss 0.265507 Objective Loss 0.265507 LR 0.001000 Time 0.022264 +2023-10-02 20:56:55,316 - Epoch: [58][ 880/ 1236] Overall Loss 0.265270 Objective Loss 0.265270 LR 0.001000 Time 0.022249 +2023-10-02 20:56:55,526 - Epoch: [58][ 890/ 1236] Overall Loss 0.265143 Objective Loss 0.265143 LR 0.001000 Time 0.022234 +2023-10-02 20:56:55,735 - Epoch: [58][ 900/ 1236] Overall Loss 0.265009 Objective Loss 0.265009 LR 0.001000 Time 0.022220 +2023-10-02 20:56:55,945 - Epoch: [58][ 910/ 1236] Overall Loss 0.265153 Objective Loss 0.265153 LR 0.001000 Time 0.022206 +2023-10-02 20:56:56,156 - Epoch: [58][ 920/ 1236] Overall Loss 0.265089 Objective Loss 0.265089 LR 0.001000 Time 0.022193 +2023-10-02 20:56:56,366 - Epoch: [58][ 930/ 1236] Overall Loss 0.264944 Objective Loss 0.264944 LR 0.001000 Time 0.022180 +2023-10-02 20:56:56,576 - Epoch: [58][ 940/ 1236] Overall Loss 0.264745 Objective Loss 0.264745 LR 0.001000 Time 0.022167 +2023-10-02 20:56:56,786 - Epoch: [58][ 950/ 1236] Overall Loss 0.264484 Objective Loss 0.264484 LR 0.001000 Time 0.022154 +2023-10-02 20:56:56,996 - Epoch: [58][ 960/ 1236] Overall Loss 0.264367 Objective Loss 0.264367 LR 0.001000 Time 0.022142 +2023-10-02 20:56:57,206 - Epoch: [58][ 970/ 1236] Overall Loss 0.264188 Objective Loss 0.264188 LR 0.001000 Time 0.022130 +2023-10-02 20:56:57,415 - Epoch: [58][ 980/ 1236] Overall Loss 0.264133 Objective Loss 0.264133 LR 0.001000 Time 0.022117 +2023-10-02 20:56:57,625 - Epoch: [58][ 990/ 1236] Overall Loss 0.264212 Objective Loss 0.264212 LR 0.001000 Time 0.022106 +2023-10-02 20:56:57,835 - Epoch: [58][ 1000/ 1236] Overall Loss 0.264358 Objective Loss 0.264358 LR 0.001000 Time 0.022094 +2023-10-02 20:56:58,045 - Epoch: [58][ 1010/ 1236] Overall Loss 0.264275 Objective Loss 0.264275 LR 0.001000 Time 0.022082 +2023-10-02 20:56:58,254 - Epoch: [58][ 1020/ 1236] Overall Loss 0.264211 Objective Loss 0.264211 LR 0.001000 Time 0.022071 +2023-10-02 20:56:58,464 - Epoch: [58][ 1030/ 1236] Overall Loss 0.264381 Objective Loss 0.264381 LR 0.001000 Time 0.022060 +2023-10-02 20:56:58,673 - Epoch: [58][ 1040/ 1236] Overall Loss 0.264205 Objective Loss 0.264205 LR 0.001000 Time 0.022049 +2023-10-02 20:56:58,883 - Epoch: [58][ 1050/ 1236] Overall Loss 0.264614 Objective Loss 0.264614 LR 0.001000 Time 0.022038 +2023-10-02 20:56:59,093 - Epoch: [58][ 1060/ 1236] Overall Loss 0.264784 Objective Loss 0.264784 LR 0.001000 Time 0.022028 +2023-10-02 20:56:59,303 - Epoch: [58][ 1070/ 1236] Overall Loss 0.264588 Objective Loss 0.264588 LR 0.001000 Time 0.022018 +2023-10-02 20:56:59,513 - Epoch: [58][ 1080/ 1236] Overall Loss 0.264418 Objective Loss 0.264418 LR 0.001000 Time 0.022008 +2023-10-02 20:56:59,725 - Epoch: [58][ 1090/ 1236] Overall Loss 0.264214 Objective Loss 0.264214 LR 0.001000 Time 0.022000 +2023-10-02 20:56:59,934 - Epoch: [58][ 1100/ 1236] Overall Loss 0.264276 Objective Loss 0.264276 LR 0.001000 Time 0.021991 +2023-10-02 20:57:00,144 - Epoch: [58][ 1110/ 1236] Overall Loss 0.264454 Objective Loss 0.264454 LR 0.001000 Time 0.021981 +2023-10-02 20:57:00,353 - Epoch: [58][ 1120/ 1236] Overall Loss 0.264689 Objective Loss 0.264689 LR 0.001000 Time 0.021971 +2023-10-02 20:57:00,563 - Epoch: [58][ 1130/ 1236] Overall Loss 0.264770 Objective Loss 0.264770 LR 0.001000 Time 0.021962 +2023-10-02 20:57:00,772 - Epoch: [58][ 1140/ 1236] Overall Loss 0.265045 Objective Loss 0.265045 LR 0.001000 Time 0.021953 +2023-10-02 20:57:00,983 - Epoch: [58][ 1150/ 1236] Overall Loss 0.264909 Objective Loss 0.264909 LR 0.001000 Time 0.021944 +2023-10-02 20:57:01,193 - Epoch: [58][ 1160/ 1236] Overall Loss 0.264570 Objective Loss 0.264570 LR 0.001000 Time 0.021936 +2023-10-02 20:57:01,403 - Epoch: [58][ 1170/ 1236] Overall Loss 0.264674 Objective Loss 0.264674 LR 0.001000 Time 0.021928 +2023-10-02 20:57:01,613 - Epoch: [58][ 1180/ 1236] Overall Loss 0.264667 Objective Loss 0.264667 LR 0.001000 Time 0.021920 +2023-10-02 20:57:01,823 - Epoch: [58][ 1190/ 1236] Overall Loss 0.264462 Objective Loss 0.264462 LR 0.001000 Time 0.021912 +2023-10-02 20:57:02,033 - Epoch: [58][ 1200/ 1236] Overall Loss 0.264448 Objective Loss 0.264448 LR 0.001000 Time 0.021904 +2023-10-02 20:57:02,243 - Epoch: [58][ 1210/ 1236] Overall Loss 0.264389 Objective Loss 0.264389 LR 0.001000 Time 0.021896 +2023-10-02 20:57:02,454 - Epoch: [58][ 1220/ 1236] Overall Loss 0.264680 Objective Loss 0.264680 LR 0.001000 Time 0.021889 +2023-10-02 20:57:02,715 - Epoch: [58][ 1230/ 1236] Overall Loss 0.264675 Objective Loss 0.264675 LR 0.001000 Time 0.021923 +2023-10-02 20:57:02,836 - Epoch: [58][ 1236/ 1236] Overall Loss 0.264618 Objective Loss 0.264618 Top1 83.706721 Top5 97.556008 LR 0.001000 Time 0.021914 +2023-10-02 20:57:02,952 - --- validate (epoch=58)----------- +2023-10-02 20:57:02,952 - 29943 samples (256 per mini-batch) +2023-10-02 20:57:03,435 - Epoch: [58][ 10/ 117] Loss 0.325082 Top1 83.828125 Top5 98.046875 +2023-10-02 20:57:03,591 - Epoch: [58][ 20/ 117] Loss 0.314937 Top1 84.394531 Top5 98.203125 +2023-10-02 20:57:03,744 - Epoch: [58][ 30/ 117] Loss 0.323241 Top1 84.518229 Top5 98.281250 +2023-10-02 20:57:03,899 - Epoch: [58][ 40/ 117] Loss 0.321505 Top1 84.541016 Top5 98.291016 +2023-10-02 20:57:04,051 - Epoch: [58][ 50/ 117] Loss 0.325069 Top1 84.421875 Top5 98.289062 +2023-10-02 20:57:04,205 - Epoch: [58][ 60/ 117] Loss 0.331535 Top1 84.316406 Top5 98.242188 +2023-10-02 20:57:04,361 - Epoch: [58][ 70/ 117] Loss 0.330737 Top1 84.335938 Top5 98.303571 +2023-10-02 20:57:04,517 - Epoch: [58][ 80/ 117] Loss 0.332042 Top1 84.370117 Top5 98.364258 +2023-10-02 20:57:04,671 - Epoch: [58][ 90/ 117] Loss 0.332774 Top1 84.279514 Top5 98.350694 +2023-10-02 20:57:04,824 - Epoch: [58][ 100/ 117] Loss 0.332489 Top1 84.246094 Top5 98.367188 +2023-10-02 20:57:04,983 - Epoch: [58][ 110/ 117] Loss 0.332474 Top1 84.169034 Top5 98.334517 +2023-10-02 20:57:05,073 - Epoch: [58][ 117/ 117] Loss 0.331511 Top1 84.163243 Top5 98.346859 +2023-10-02 20:57:05,221 - ==> Top1: 84.163 Top5: 98.347 Loss: 0.332 + +2023-10-02 20:57:05,222 - ==> Confusion: +[[ 897 3 4 0 14 3 0 0 4 82 1 5 2 3 4 0 4 2 1 0 21] + [ 0 1035 1 0 8 35 1 21 2 1 6 0 1 0 2 4 1 0 6 1 6] + [ 1 0 944 14 2 0 29 9 0 3 3 3 11 3 0 3 1 0 10 6 14] + [ 1 2 12 972 1 2 4 4 3 0 13 0 6 4 25 3 1 5 8 0 23] + [ 17 6 1 0 953 10 0 1 0 14 0 0 1 3 13 5 17 0 0 5 4] + [ 3 26 0 1 0 1014 1 18 2 5 2 4 4 11 4 1 1 0 3 1 15] + [ 0 4 22 0 0 2 1123 6 0 0 10 2 0 0 0 5 0 0 0 11 6] + [ 2 18 14 0 6 32 10 1041 0 3 8 7 1 4 2 1 0 0 41 19 9] + [ 16 7 0 1 0 1 1 0 958 43 12 2 4 6 27 2 2 1 2 1 3] + [ 83 1 0 0 6 1 0 0 21 962 1 2 0 19 7 1 1 0 1 4 9] + [ 3 3 6 9 1 2 1 2 12 0 965 3 2 16 8 0 0 3 7 2 8] + [ 0 0 1 0 0 19 0 4 0 1 0 908 55 7 0 2 3 16 0 12 7] + [ 0 0 3 4 2 4 1 1 0 0 3 32 966 6 1 8 4 17 3 7 6] + [ 1 0 1 0 1 13 0 1 14 14 5 2 0 1039 4 1 0 2 0 0 21] + [ 6 1 4 19 6 0 0 0 15 4 2 0 4 6 1016 0 0 2 7 0 9] + [ 0 0 2 2 4 2 2 0 0 1 1 8 7 0 0 1065 13 16 1 6 4] + [ 1 17 4 1 5 9 0 0 0 0 0 4 1 1 4 12 1079 1 1 10 11] + [ 0 0 0 2 0 0 4 0 0 1 2 5 23 1 3 6 2 981 0 1 7] + [ 1 3 6 12 0 0 1 23 3 1 11 1 3 0 6 0 1 0 982 2 12] + [ 0 2 3 0 0 7 3 12 0 0 1 12 6 0 0 5 8 0 2 1089 2] + [ 134 155 105 86 91 207 45 85 106 81 199 119 368 204 125 58 107 56 144 218 5212]] + +2023-10-02 20:57:05,223 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:57:05,223 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:57:05,229 - + +2023-10-02 20:57:05,229 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:57:06,272 - Epoch: [59][ 10/ 1236] Overall Loss 0.226025 Objective Loss 0.226025 LR 0.001000 Time 0.104183 +2023-10-02 20:57:06,483 - Epoch: [59][ 20/ 1236] Overall Loss 0.248539 Objective Loss 0.248539 LR 0.001000 Time 0.062620 +2023-10-02 20:57:06,695 - Epoch: [59][ 30/ 1236] Overall Loss 0.250022 Objective Loss 0.250022 LR 0.001000 Time 0.048814 +2023-10-02 20:57:06,906 - Epoch: [59][ 40/ 1236] Overall Loss 0.253959 Objective Loss 0.253959 LR 0.001000 Time 0.041860 +2023-10-02 20:57:07,114 - Epoch: [59][ 50/ 1236] Overall Loss 0.253345 Objective Loss 0.253345 LR 0.001000 Time 0.037627 +2023-10-02 20:57:07,324 - Epoch: [59][ 60/ 1236] Overall Loss 0.256934 Objective Loss 0.256934 LR 0.001000 Time 0.034843 +2023-10-02 20:57:07,532 - Epoch: [59][ 70/ 1236] Overall Loss 0.254552 Objective Loss 0.254552 LR 0.001000 Time 0.032822 +2023-10-02 20:57:07,742 - Epoch: [59][ 80/ 1236] Overall Loss 0.252551 Objective Loss 0.252551 LR 0.001000 Time 0.031337 +2023-10-02 20:57:07,953 - Epoch: [59][ 90/ 1236] Overall Loss 0.253587 Objective Loss 0.253587 LR 0.001000 Time 0.030183 +2023-10-02 20:57:08,173 - Epoch: [59][ 100/ 1236] Overall Loss 0.253532 Objective Loss 0.253532 LR 0.001000 Time 0.029352 +2023-10-02 20:57:08,392 - Epoch: [59][ 110/ 1236] Overall Loss 0.252589 Objective Loss 0.252589 LR 0.001000 Time 0.028672 +2023-10-02 20:57:08,616 - Epoch: [59][ 120/ 1236] Overall Loss 0.251290 Objective Loss 0.251290 LR 0.001000 Time 0.028133 +2023-10-02 20:57:08,835 - Epoch: [59][ 130/ 1236] Overall Loss 0.251520 Objective Loss 0.251520 LR 0.001000 Time 0.027651 +2023-10-02 20:57:09,057 - Epoch: [59][ 140/ 1236] Overall Loss 0.252454 Objective Loss 0.252454 LR 0.001000 Time 0.027253 +2023-10-02 20:57:09,274 - Epoch: [59][ 150/ 1236] Overall Loss 0.252797 Objective Loss 0.252797 LR 0.001000 Time 0.026878 +2023-10-02 20:57:09,496 - Epoch: [59][ 160/ 1236] Overall Loss 0.253248 Objective Loss 0.253248 LR 0.001000 Time 0.026584 +2023-10-02 20:57:09,713 - Epoch: [59][ 170/ 1236] Overall Loss 0.253599 Objective Loss 0.253599 LR 0.001000 Time 0.026294 +2023-10-02 20:57:09,935 - Epoch: [59][ 180/ 1236] Overall Loss 0.254609 Objective Loss 0.254609 LR 0.001000 Time 0.026065 +2023-10-02 20:57:10,152 - Epoch: [59][ 190/ 1236] Overall Loss 0.254914 Objective Loss 0.254914 LR 0.001000 Time 0.025832 +2023-10-02 20:57:10,374 - Epoch: [59][ 200/ 1236] Overall Loss 0.255562 Objective Loss 0.255562 LR 0.001000 Time 0.025649 +2023-10-02 20:57:10,591 - Epoch: [59][ 210/ 1236] Overall Loss 0.255127 Objective Loss 0.255127 LR 0.001000 Time 0.025462 +2023-10-02 20:57:10,813 - Epoch: [59][ 220/ 1236] Overall Loss 0.254383 Objective Loss 0.254383 LR 0.001000 Time 0.025311 +2023-10-02 20:57:11,023 - Epoch: [59][ 230/ 1236] Overall Loss 0.255154 Objective Loss 0.255154 LR 0.001000 Time 0.025120 +2023-10-02 20:57:11,235 - Epoch: [59][ 240/ 1236] Overall Loss 0.254963 Objective Loss 0.254963 LR 0.001000 Time 0.024957 +2023-10-02 20:57:11,444 - Epoch: [59][ 250/ 1236] Overall Loss 0.254467 Objective Loss 0.254467 LR 0.001000 Time 0.024795 +2023-10-02 20:57:11,657 - Epoch: [59][ 260/ 1236] Overall Loss 0.255313 Objective Loss 0.255313 LR 0.001000 Time 0.024658 +2023-10-02 20:57:11,866 - Epoch: [59][ 270/ 1236] Overall Loss 0.254800 Objective Loss 0.254800 LR 0.001000 Time 0.024517 +2023-10-02 20:57:12,078 - Epoch: [59][ 280/ 1236] Overall Loss 0.256142 Objective Loss 0.256142 LR 0.001000 Time 0.024399 +2023-10-02 20:57:12,287 - Epoch: [59][ 290/ 1236] Overall Loss 0.256284 Objective Loss 0.256284 LR 0.001000 Time 0.024276 +2023-10-02 20:57:12,504 - Epoch: [59][ 300/ 1236] Overall Loss 0.256323 Objective Loss 0.256323 LR 0.001000 Time 0.024190 +2023-10-02 20:57:12,720 - Epoch: [59][ 310/ 1236] Overall Loss 0.256349 Objective Loss 0.256349 LR 0.001000 Time 0.024104 +2023-10-02 20:57:12,941 - Epoch: [59][ 320/ 1236] Overall Loss 0.255991 Objective Loss 0.255991 LR 0.001000 Time 0.024039 +2023-10-02 20:57:13,157 - Epoch: [59][ 330/ 1236] Overall Loss 0.257101 Objective Loss 0.257101 LR 0.001000 Time 0.023965 +2023-10-02 20:57:13,379 - Epoch: [59][ 340/ 1236] Overall Loss 0.257614 Objective Loss 0.257614 LR 0.001000 Time 0.023910 +2023-10-02 20:57:13,596 - Epoch: [59][ 350/ 1236] Overall Loss 0.258307 Objective Loss 0.258307 LR 0.001000 Time 0.023848 +2023-10-02 20:57:13,818 - Epoch: [59][ 360/ 1236] Overall Loss 0.258207 Objective Loss 0.258207 LR 0.001000 Time 0.023800 +2023-10-02 20:57:14,034 - Epoch: [59][ 370/ 1236] Overall Loss 0.257641 Objective Loss 0.257641 LR 0.001000 Time 0.023741 +2023-10-02 20:57:14,257 - Epoch: [59][ 380/ 1236] Overall Loss 0.257179 Objective Loss 0.257179 LR 0.001000 Time 0.023700 +2023-10-02 20:57:14,474 - Epoch: [59][ 390/ 1236] Overall Loss 0.257093 Objective Loss 0.257093 LR 0.001000 Time 0.023647 +2023-10-02 20:57:14,696 - Epoch: [59][ 400/ 1236] Overall Loss 0.257229 Objective Loss 0.257229 LR 0.001000 Time 0.023610 +2023-10-02 20:57:14,912 - Epoch: [59][ 410/ 1236] Overall Loss 0.257325 Objective Loss 0.257325 LR 0.001000 Time 0.023562 +2023-10-02 20:57:15,134 - Epoch: [59][ 420/ 1236] Overall Loss 0.257230 Objective Loss 0.257230 LR 0.001000 Time 0.023529 +2023-10-02 20:57:15,351 - Epoch: [59][ 430/ 1236] Overall Loss 0.256652 Objective Loss 0.256652 LR 0.001000 Time 0.023486 +2023-10-02 20:57:15,574 - Epoch: [59][ 440/ 1236] Overall Loss 0.256360 Objective Loss 0.256360 LR 0.001000 Time 0.023456 +2023-10-02 20:57:15,790 - Epoch: [59][ 450/ 1236] Overall Loss 0.256250 Objective Loss 0.256250 LR 0.001000 Time 0.023416 +2023-10-02 20:57:16,012 - Epoch: [59][ 460/ 1236] Overall Loss 0.255592 Objective Loss 0.255592 LR 0.001000 Time 0.023388 +2023-10-02 20:57:16,229 - Epoch: [59][ 470/ 1236] Overall Loss 0.255740 Objective Loss 0.255740 LR 0.001000 Time 0.023352 +2023-10-02 20:57:16,452 - Epoch: [59][ 480/ 1236] Overall Loss 0.255799 Objective Loss 0.255799 LR 0.001000 Time 0.023329 +2023-10-02 20:57:16,669 - Epoch: [59][ 490/ 1236] Overall Loss 0.255965 Objective Loss 0.255965 LR 0.001000 Time 0.023293 +2023-10-02 20:57:16,890 - Epoch: [59][ 500/ 1236] Overall Loss 0.255835 Objective Loss 0.255835 LR 0.001000 Time 0.023270 +2023-10-02 20:57:17,099 - Epoch: [59][ 510/ 1236] Overall Loss 0.255733 Objective Loss 0.255733 LR 0.001000 Time 0.023219 +2023-10-02 20:57:17,308 - Epoch: [59][ 520/ 1236] Overall Loss 0.255285 Objective Loss 0.255285 LR 0.001000 Time 0.023174 +2023-10-02 20:57:17,517 - Epoch: [59][ 530/ 1236] Overall Loss 0.255414 Objective Loss 0.255414 LR 0.001000 Time 0.023130 +2023-10-02 20:57:17,726 - Epoch: [59][ 540/ 1236] Overall Loss 0.255600 Objective Loss 0.255600 LR 0.001000 Time 0.023089 +2023-10-02 20:57:17,935 - Epoch: [59][ 550/ 1236] Overall Loss 0.255900 Objective Loss 0.255900 LR 0.001000 Time 0.023047 +2023-10-02 20:57:18,144 - Epoch: [59][ 560/ 1236] Overall Loss 0.255996 Objective Loss 0.255996 LR 0.001000 Time 0.023009 +2023-10-02 20:57:18,353 - Epoch: [59][ 570/ 1236] Overall Loss 0.256349 Objective Loss 0.256349 LR 0.001000 Time 0.022968 +2023-10-02 20:57:18,562 - Epoch: [59][ 580/ 1236] Overall Loss 0.256484 Objective Loss 0.256484 LR 0.001000 Time 0.022932 +2023-10-02 20:57:18,771 - Epoch: [59][ 590/ 1236] Overall Loss 0.256620 Objective Loss 0.256620 LR 0.001000 Time 0.022896 +2023-10-02 20:57:18,980 - Epoch: [59][ 600/ 1236] Overall Loss 0.257077 Objective Loss 0.257077 LR 0.001000 Time 0.022863 +2023-10-02 20:57:19,189 - Epoch: [59][ 610/ 1236] Overall Loss 0.256928 Objective Loss 0.256928 LR 0.001000 Time 0.022830 +2023-10-02 20:57:19,401 - Epoch: [59][ 620/ 1236] Overall Loss 0.256822 Objective Loss 0.256822 LR 0.001000 Time 0.022803 +2023-10-02 20:57:19,610 - Epoch: [59][ 630/ 1236] Overall Loss 0.256409 Objective Loss 0.256409 LR 0.001000 Time 0.022773 +2023-10-02 20:57:19,821 - Epoch: [59][ 640/ 1236] Overall Loss 0.256648 Objective Loss 0.256648 LR 0.001000 Time 0.022745 +2023-10-02 20:57:20,030 - Epoch: [59][ 650/ 1236] Overall Loss 0.256239 Objective Loss 0.256239 LR 0.001000 Time 0.022716 +2023-10-02 20:57:20,240 - Epoch: [59][ 660/ 1236] Overall Loss 0.256282 Objective Loss 0.256282 LR 0.001000 Time 0.022689 +2023-10-02 20:57:20,450 - Epoch: [59][ 670/ 1236] Overall Loss 0.255998 Objective Loss 0.255998 LR 0.001000 Time 0.022663 +2023-10-02 20:57:20,659 - Epoch: [59][ 680/ 1236] Overall Loss 0.256077 Objective Loss 0.256077 LR 0.001000 Time 0.022637 +2023-10-02 20:57:20,869 - Epoch: [59][ 690/ 1236] Overall Loss 0.256013 Objective Loss 0.256013 LR 0.001000 Time 0.022612 +2023-10-02 20:57:21,079 - Epoch: [59][ 700/ 1236] Overall Loss 0.255695 Objective Loss 0.255695 LR 0.001000 Time 0.022589 +2023-10-02 20:57:21,288 - Epoch: [59][ 710/ 1236] Overall Loss 0.255937 Objective Loss 0.255937 LR 0.001000 Time 0.022564 +2023-10-02 20:57:21,498 - Epoch: [59][ 720/ 1236] Overall Loss 0.256597 Objective Loss 0.256597 LR 0.001000 Time 0.022542 +2023-10-02 20:57:21,708 - Epoch: [59][ 730/ 1236] Overall Loss 0.256626 Objective Loss 0.256626 LR 0.001000 Time 0.022520 +2023-10-02 20:57:21,918 - Epoch: [59][ 740/ 1236] Overall Loss 0.256495 Objective Loss 0.256495 LR 0.001000 Time 0.022499 +2023-10-02 20:57:22,127 - Epoch: [59][ 750/ 1236] Overall Loss 0.256660 Objective Loss 0.256660 LR 0.001000 Time 0.022477 +2023-10-02 20:57:22,337 - Epoch: [59][ 760/ 1236] Overall Loss 0.256330 Objective Loss 0.256330 LR 0.001000 Time 0.022458 +2023-10-02 20:57:22,547 - Epoch: [59][ 770/ 1236] Overall Loss 0.256390 Objective Loss 0.256390 LR 0.001000 Time 0.022438 +2023-10-02 20:57:22,757 - Epoch: [59][ 780/ 1236] Overall Loss 0.256804 Objective Loss 0.256804 LR 0.001000 Time 0.022419 +2023-10-02 20:57:22,972 - Epoch: [59][ 790/ 1236] Overall Loss 0.257040 Objective Loss 0.257040 LR 0.001000 Time 0.022406 +2023-10-02 20:57:23,183 - Epoch: [59][ 800/ 1236] Overall Loss 0.257205 Objective Loss 0.257205 LR 0.001000 Time 0.022389 +2023-10-02 20:57:23,392 - Epoch: [59][ 810/ 1236] Overall Loss 0.256945 Objective Loss 0.256945 LR 0.001000 Time 0.022370 +2023-10-02 20:57:23,602 - Epoch: [59][ 820/ 1236] Overall Loss 0.257138 Objective Loss 0.257138 LR 0.001000 Time 0.022353 +2023-10-02 20:57:23,811 - Epoch: [59][ 830/ 1236] Overall Loss 0.257459 Objective Loss 0.257459 LR 0.001000 Time 0.022335 +2023-10-02 20:57:24,021 - Epoch: [59][ 840/ 1236] Overall Loss 0.257897 Objective Loss 0.257897 LR 0.001000 Time 0.022319 +2023-10-02 20:57:24,230 - Epoch: [59][ 850/ 1236] Overall Loss 0.257748 Objective Loss 0.257748 LR 0.001000 Time 0.022302 +2023-10-02 20:57:24,441 - Epoch: [59][ 860/ 1236] Overall Loss 0.257908 Objective Loss 0.257908 LR 0.001000 Time 0.022287 +2023-10-02 20:57:24,649 - Epoch: [59][ 870/ 1236] Overall Loss 0.258268 Objective Loss 0.258268 LR 0.001000 Time 0.022270 +2023-10-02 20:57:24,860 - Epoch: [59][ 880/ 1236] Overall Loss 0.258303 Objective Loss 0.258303 LR 0.001000 Time 0.022255 +2023-10-02 20:57:25,069 - Epoch: [59][ 890/ 1236] Overall Loss 0.258233 Objective Loss 0.258233 LR 0.001000 Time 0.022240 +2023-10-02 20:57:25,279 - Epoch: [59][ 900/ 1236] Overall Loss 0.258189 Objective Loss 0.258189 LR 0.001000 Time 0.022226 +2023-10-02 20:57:25,488 - Epoch: [59][ 910/ 1236] Overall Loss 0.258305 Objective Loss 0.258305 LR 0.001000 Time 0.022211 +2023-10-02 20:57:25,698 - Epoch: [59][ 920/ 1236] Overall Loss 0.258449 Objective Loss 0.258449 LR 0.001000 Time 0.022197 +2023-10-02 20:57:25,907 - Epoch: [59][ 930/ 1236] Overall Loss 0.258501 Objective Loss 0.258501 LR 0.001000 Time 0.022183 +2023-10-02 20:57:26,117 - Epoch: [59][ 940/ 1236] Overall Loss 0.258199 Objective Loss 0.258199 LR 0.001000 Time 0.022170 +2023-10-02 20:57:26,326 - Epoch: [59][ 950/ 1236] Overall Loss 0.258234 Objective Loss 0.258234 LR 0.001000 Time 0.022156 +2023-10-02 20:57:26,537 - Epoch: [59][ 960/ 1236] Overall Loss 0.257873 Objective Loss 0.257873 LR 0.001000 Time 0.022144 +2023-10-02 20:57:26,747 - Epoch: [59][ 970/ 1236] Overall Loss 0.257741 Objective Loss 0.257741 LR 0.001000 Time 0.022133 +2023-10-02 20:57:26,964 - Epoch: [59][ 980/ 1236] Overall Loss 0.257563 Objective Loss 0.257563 LR 0.001000 Time 0.022127 +2023-10-02 20:57:27,175 - Epoch: [59][ 990/ 1236] Overall Loss 0.257716 Objective Loss 0.257716 LR 0.001000 Time 0.022115 +2023-10-02 20:57:27,387 - Epoch: [59][ 1000/ 1236] Overall Loss 0.257754 Objective Loss 0.257754 LR 0.001000 Time 0.022106 +2023-10-02 20:57:27,598 - Epoch: [59][ 1010/ 1236] Overall Loss 0.257716 Objective Loss 0.257716 LR 0.001000 Time 0.022095 +2023-10-02 20:57:27,810 - Epoch: [59][ 1020/ 1236] Overall Loss 0.257526 Objective Loss 0.257526 LR 0.001000 Time 0.022087 +2023-10-02 20:57:28,021 - Epoch: [59][ 1030/ 1236] Overall Loss 0.257557 Objective Loss 0.257557 LR 0.001000 Time 0.022076 +2023-10-02 20:57:28,233 - Epoch: [59][ 1040/ 1236] Overall Loss 0.257475 Objective Loss 0.257475 LR 0.001000 Time 0.022067 +2023-10-02 20:57:28,443 - Epoch: [59][ 1050/ 1236] Overall Loss 0.257401 Objective Loss 0.257401 LR 0.001000 Time 0.022057 +2023-10-02 20:57:28,655 - Epoch: [59][ 1060/ 1236] Overall Loss 0.257551 Objective Loss 0.257551 LR 0.001000 Time 0.022048 +2023-10-02 20:57:28,866 - Epoch: [59][ 1070/ 1236] Overall Loss 0.257663 Objective Loss 0.257663 LR 0.001000 Time 0.022039 +2023-10-02 20:57:29,079 - Epoch: [59][ 1080/ 1236] Overall Loss 0.257646 Objective Loss 0.257646 LR 0.001000 Time 0.022031 +2023-10-02 20:57:29,293 - Epoch: [59][ 1090/ 1236] Overall Loss 0.258006 Objective Loss 0.258006 LR 0.001000 Time 0.022025 +2023-10-02 20:57:29,506 - Epoch: [59][ 1100/ 1236] Overall Loss 0.257944 Objective Loss 0.257944 LR 0.001000 Time 0.022018 +2023-10-02 20:57:29,717 - Epoch: [59][ 1110/ 1236] Overall Loss 0.257800 Objective Loss 0.257800 LR 0.001000 Time 0.022010 +2023-10-02 20:57:29,930 - Epoch: [59][ 1120/ 1236] Overall Loss 0.257644 Objective Loss 0.257644 LR 0.001000 Time 0.022003 +2023-10-02 20:57:30,140 - Epoch: [59][ 1130/ 1236] Overall Loss 0.257794 Objective Loss 0.257794 LR 0.001000 Time 0.021994 +2023-10-02 20:57:30,352 - Epoch: [59][ 1140/ 1236] Overall Loss 0.258023 Objective Loss 0.258023 LR 0.001000 Time 0.021987 +2023-10-02 20:57:30,563 - Epoch: [59][ 1150/ 1236] Overall Loss 0.258251 Objective Loss 0.258251 LR 0.001000 Time 0.021978 +2023-10-02 20:57:30,775 - Epoch: [59][ 1160/ 1236] Overall Loss 0.258627 Objective Loss 0.258627 LR 0.001000 Time 0.021972 +2023-10-02 20:57:30,987 - Epoch: [59][ 1170/ 1236] Overall Loss 0.258799 Objective Loss 0.258799 LR 0.001000 Time 0.021964 +2023-10-02 20:57:31,199 - Epoch: [59][ 1180/ 1236] Overall Loss 0.258887 Objective Loss 0.258887 LR 0.001000 Time 0.021957 +2023-10-02 20:57:31,414 - Epoch: [59][ 1190/ 1236] Overall Loss 0.258967 Objective Loss 0.258967 LR 0.001000 Time 0.021953 +2023-10-02 20:57:31,627 - Epoch: [59][ 1200/ 1236] Overall Loss 0.258935 Objective Loss 0.258935 LR 0.001000 Time 0.021948 +2023-10-02 20:57:31,841 - Epoch: [59][ 1210/ 1236] Overall Loss 0.258969 Objective Loss 0.258969 LR 0.001000 Time 0.021942 +2023-10-02 20:57:32,053 - Epoch: [59][ 1220/ 1236] Overall Loss 0.258753 Objective Loss 0.258753 LR 0.001000 Time 0.021936 +2023-10-02 20:57:32,322 - Epoch: [59][ 1230/ 1236] Overall Loss 0.258608 Objective Loss 0.258608 LR 0.001000 Time 0.021976 +2023-10-02 20:57:32,445 - Epoch: [59][ 1236/ 1236] Overall Loss 0.258703 Objective Loss 0.258703 Top1 85.743381 Top5 98.167006 LR 0.001000 Time 0.021969 +2023-10-02 20:57:32,583 - --- validate (epoch=59)----------- +2023-10-02 20:57:32,583 - 29943 samples (256 per mini-batch) +2023-10-02 20:57:33,075 - Epoch: [59][ 10/ 117] Loss 0.379843 Top1 82.578125 Top5 97.734375 +2023-10-02 20:57:33,227 - Epoch: [59][ 20/ 117] Loss 0.352745 Top1 83.300781 Top5 97.773438 +2023-10-02 20:57:33,379 - Epoch: [59][ 30/ 117] Loss 0.352065 Top1 83.763021 Top5 97.877604 +2023-10-02 20:57:33,535 - Epoch: [59][ 40/ 117] Loss 0.345462 Top1 83.828125 Top5 97.998047 +2023-10-02 20:57:33,693 - Epoch: [59][ 50/ 117] Loss 0.341891 Top1 83.828125 Top5 98.031250 +2023-10-02 20:57:33,856 - Epoch: [59][ 60/ 117] Loss 0.339938 Top1 83.743490 Top5 98.066406 +2023-10-02 20:57:34,015 - Epoch: [59][ 70/ 117] Loss 0.336477 Top1 83.856027 Top5 98.091518 +2023-10-02 20:57:34,167 - Epoch: [59][ 80/ 117] Loss 0.337133 Top1 83.842773 Top5 98.120117 +2023-10-02 20:57:34,317 - Epoch: [59][ 90/ 117] Loss 0.332843 Top1 83.984375 Top5 98.146701 +2023-10-02 20:57:34,469 - Epoch: [59][ 100/ 117] Loss 0.333420 Top1 83.945312 Top5 98.167969 +2023-10-02 20:57:34,626 - Epoch: [59][ 110/ 117] Loss 0.333980 Top1 83.980824 Top5 98.167614 +2023-10-02 20:57:34,718 - Epoch: [59][ 117/ 117] Loss 0.334492 Top1 83.952844 Top5 98.166516 +2023-10-02 20:57:34,859 - ==> Top1: 83.953 Top5: 98.167 Loss: 0.334 + +2023-10-02 20:57:34,860 - ==> Confusion: +[[ 914 0 3 0 14 3 0 0 6 70 1 2 1 4 8 0 4 2 1 0 17] + [ 1 1034 1 0 4 29 2 21 3 0 1 1 1 0 1 3 3 0 16 3 7] + [ 3 0 961 5 2 0 26 8 0 2 2 3 8 1 1 2 3 2 11 3 13] + [ 1 4 8 982 2 3 2 4 0 0 5 0 9 9 26 1 1 2 15 0 15] + [ 28 6 1 0 951 5 0 0 0 16 0 2 0 5 7 4 17 0 0 5 3] + [ 3 44 0 2 4 967 2 19 4 8 2 5 5 16 6 2 3 0 4 6 14] + [ 1 2 27 0 0 1 1121 8 0 0 4 2 1 0 0 5 0 1 2 6 10] + [ 6 17 19 0 3 31 9 1023 3 3 1 12 2 2 0 1 1 2 59 13 11] + [ 17 3 1 0 0 1 0 2 972 43 4 3 1 4 25 0 4 5 2 0 2] + [ 89 1 0 0 4 0 1 1 29 947 1 0 0 19 10 1 1 0 0 5 10] + [ 4 3 10 7 1 1 6 5 24 1 918 2 2 34 10 0 1 3 13 2 6] + [ 0 0 2 0 0 13 0 3 0 2 0 937 39 9 0 2 1 15 0 4 8] + [ 0 1 2 3 1 4 2 2 3 0 2 33 973 3 3 4 2 15 2 4 9] + [ 0 0 3 1 1 4 0 0 17 19 2 5 1 1048 6 0 1 0 0 1 10] + [ 9 0 5 22 8 0 1 0 18 3 0 0 4 4 1012 0 0 5 2 0 8] + [ 0 0 1 3 4 3 1 0 0 0 0 9 6 3 1 1051 13 18 0 9 12] + [ 0 12 1 0 3 4 0 0 1 0 0 8 5 2 2 8 1091 0 2 5 17] + [ 0 0 1 4 0 0 1 0 0 1 0 0 19 1 4 3 0 998 0 2 4] + [ 4 4 5 24 0 1 1 9 6 0 3 2 0 0 18 0 0 0 979 1 11] + [ 0 2 2 0 0 7 5 4 0 1 1 17 7 3 0 2 5 1 2 1082 11] + [ 125 164 116 85 79 173 37 81 132 72 153 116 343 316 176 46 119 62 162 171 5177]] + +2023-10-02 20:57:34,861 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:57:34,861 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:57:34,867 - + +2023-10-02 20:57:34,867 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:57:35,875 - Epoch: [60][ 10/ 1236] Overall Loss 0.246125 Objective Loss 0.246125 LR 0.001000 Time 0.100764 +2023-10-02 20:57:36,084 - Epoch: [60][ 20/ 1236] Overall Loss 0.245142 Objective Loss 0.245142 LR 0.001000 Time 0.060813 +2023-10-02 20:57:36,292 - Epoch: [60][ 30/ 1236] Overall Loss 0.243237 Objective Loss 0.243237 LR 0.001000 Time 0.047419 +2023-10-02 20:57:36,502 - Epoch: [60][ 40/ 1236] Overall Loss 0.239424 Objective Loss 0.239424 LR 0.001000 Time 0.040802 +2023-10-02 20:57:36,708 - Epoch: [60][ 50/ 1236] Overall Loss 0.238520 Objective Loss 0.238520 LR 0.001000 Time 0.036766 +2023-10-02 20:57:36,922 - Epoch: [60][ 60/ 1236] Overall Loss 0.239608 Objective Loss 0.239608 LR 0.001000 Time 0.034184 +2023-10-02 20:57:37,130 - Epoch: [60][ 70/ 1236] Overall Loss 0.242470 Objective Loss 0.242470 LR 0.001000 Time 0.032280 +2023-10-02 20:57:37,340 - Epoch: [60][ 80/ 1236] Overall Loss 0.246320 Objective Loss 0.246320 LR 0.001000 Time 0.030856 +2023-10-02 20:57:37,548 - Epoch: [60][ 90/ 1236] Overall Loss 0.249690 Objective Loss 0.249690 LR 0.001000 Time 0.029740 +2023-10-02 20:57:37,760 - Epoch: [60][ 100/ 1236] Overall Loss 0.254464 Objective Loss 0.254464 LR 0.001000 Time 0.028884 +2023-10-02 20:57:37,969 - Epoch: [60][ 110/ 1236] Overall Loss 0.253688 Objective Loss 0.253688 LR 0.001000 Time 0.028155 +2023-10-02 20:57:38,178 - Epoch: [60][ 120/ 1236] Overall Loss 0.252748 Objective Loss 0.252748 LR 0.001000 Time 0.027549 +2023-10-02 20:57:38,386 - Epoch: [60][ 130/ 1236] Overall Loss 0.255052 Objective Loss 0.255052 LR 0.001000 Time 0.027027 +2023-10-02 20:57:38,594 - Epoch: [60][ 140/ 1236] Overall Loss 0.253825 Objective Loss 0.253825 LR 0.001000 Time 0.026579 +2023-10-02 20:57:38,803 - Epoch: [60][ 150/ 1236] Overall Loss 0.253480 Objective Loss 0.253480 LR 0.001000 Time 0.026195 +2023-10-02 20:57:39,013 - Epoch: [60][ 160/ 1236] Overall Loss 0.251929 Objective Loss 0.251929 LR 0.001000 Time 0.025869 +2023-10-02 20:57:39,222 - Epoch: [60][ 170/ 1236] Overall Loss 0.251264 Objective Loss 0.251264 LR 0.001000 Time 0.025570 +2023-10-02 20:57:39,432 - Epoch: [60][ 180/ 1236] Overall Loss 0.250311 Objective Loss 0.250311 LR 0.001000 Time 0.025315 +2023-10-02 20:57:39,647 - Epoch: [60][ 190/ 1236] Overall Loss 0.250207 Objective Loss 0.250207 LR 0.001000 Time 0.025102 +2023-10-02 20:57:39,860 - Epoch: [60][ 200/ 1236] Overall Loss 0.249775 Objective Loss 0.249775 LR 0.001000 Time 0.024913 +2023-10-02 20:57:40,068 - Epoch: [60][ 210/ 1236] Overall Loss 0.249091 Objective Loss 0.249091 LR 0.001000 Time 0.024716 +2023-10-02 20:57:40,278 - Epoch: [60][ 220/ 1236] Overall Loss 0.248566 Objective Loss 0.248566 LR 0.001000 Time 0.024547 +2023-10-02 20:57:40,487 - Epoch: [60][ 230/ 1236] Overall Loss 0.250550 Objective Loss 0.250550 LR 0.001000 Time 0.024381 +2023-10-02 20:57:40,697 - Epoch: [60][ 240/ 1236] Overall Loss 0.249619 Objective Loss 0.249619 LR 0.001000 Time 0.024239 +2023-10-02 20:57:40,903 - Epoch: [60][ 250/ 1236] Overall Loss 0.250336 Objective Loss 0.250336 LR 0.001000 Time 0.024086 +2023-10-02 20:57:41,110 - Epoch: [60][ 260/ 1236] Overall Loss 0.249856 Objective Loss 0.249856 LR 0.001000 Time 0.023953 +2023-10-02 20:57:41,316 - Epoch: [60][ 270/ 1236] Overall Loss 0.250047 Objective Loss 0.250047 LR 0.001000 Time 0.023827 +2023-10-02 20:57:41,522 - Epoch: [60][ 280/ 1236] Overall Loss 0.250127 Objective Loss 0.250127 LR 0.001000 Time 0.023713 +2023-10-02 20:57:41,728 - Epoch: [60][ 290/ 1236] Overall Loss 0.250556 Objective Loss 0.250556 LR 0.001000 Time 0.023602 +2023-10-02 20:57:41,936 - Epoch: [60][ 300/ 1236] Overall Loss 0.250962 Objective Loss 0.250962 LR 0.001000 Time 0.023510 +2023-10-02 20:57:42,142 - Epoch: [60][ 310/ 1236] Overall Loss 0.250650 Objective Loss 0.250650 LR 0.001000 Time 0.023415 +2023-10-02 20:57:42,349 - Epoch: [60][ 320/ 1236] Overall Loss 0.250156 Objective Loss 0.250156 LR 0.001000 Time 0.023329 +2023-10-02 20:57:42,555 - Epoch: [60][ 330/ 1236] Overall Loss 0.249419 Objective Loss 0.249419 LR 0.001000 Time 0.023239 +2023-10-02 20:57:42,762 - Epoch: [60][ 340/ 1236] Overall Loss 0.250160 Objective Loss 0.250160 LR 0.001000 Time 0.023163 +2023-10-02 20:57:42,967 - Epoch: [60][ 350/ 1236] Overall Loss 0.249523 Objective Loss 0.249523 LR 0.001000 Time 0.023084 +2023-10-02 20:57:43,174 - Epoch: [60][ 360/ 1236] Overall Loss 0.249970 Objective Loss 0.249970 LR 0.001000 Time 0.023017 +2023-10-02 20:57:43,384 - Epoch: [60][ 370/ 1236] Overall Loss 0.250961 Objective Loss 0.250961 LR 0.001000 Time 0.022962 +2023-10-02 20:57:43,593 - Epoch: [60][ 380/ 1236] Overall Loss 0.250830 Objective Loss 0.250830 LR 0.001000 Time 0.022906 +2023-10-02 20:57:43,801 - Epoch: [60][ 390/ 1236] Overall Loss 0.250980 Objective Loss 0.250980 LR 0.001000 Time 0.022850 +2023-10-02 20:57:44,010 - Epoch: [60][ 400/ 1236] Overall Loss 0.250139 Objective Loss 0.250139 LR 0.001000 Time 0.022800 +2023-10-02 20:57:44,216 - Epoch: [60][ 410/ 1236] Overall Loss 0.250829 Objective Loss 0.250829 LR 0.001000 Time 0.022747 +2023-10-02 20:57:44,425 - Epoch: [60][ 420/ 1236] Overall Loss 0.251019 Objective Loss 0.251019 LR 0.001000 Time 0.022702 +2023-10-02 20:57:44,632 - Epoch: [60][ 430/ 1236] Overall Loss 0.251047 Objective Loss 0.251047 LR 0.001000 Time 0.022654 +2023-10-02 20:57:44,841 - Epoch: [60][ 440/ 1236] Overall Loss 0.250948 Objective Loss 0.250948 LR 0.001000 Time 0.022614 +2023-10-02 20:57:45,048 - Epoch: [60][ 450/ 1236] Overall Loss 0.250530 Objective Loss 0.250530 LR 0.001000 Time 0.022570 +2023-10-02 20:57:45,256 - Epoch: [60][ 460/ 1236] Overall Loss 0.250338 Objective Loss 0.250338 LR 0.001000 Time 0.022532 +2023-10-02 20:57:45,463 - Epoch: [60][ 470/ 1236] Overall Loss 0.250140 Objective Loss 0.250140 LR 0.001000 Time 0.022492 +2023-10-02 20:57:45,673 - Epoch: [60][ 480/ 1236] Overall Loss 0.250357 Objective Loss 0.250357 LR 0.001000 Time 0.022460 +2023-10-02 20:57:45,880 - Epoch: [60][ 490/ 1236] Overall Loss 0.250464 Objective Loss 0.250464 LR 0.001000 Time 0.022424 +2023-10-02 20:57:46,089 - Epoch: [60][ 500/ 1236] Overall Loss 0.249917 Objective Loss 0.249917 LR 0.001000 Time 0.022392 +2023-10-02 20:57:46,295 - Epoch: [60][ 510/ 1236] Overall Loss 0.250105 Objective Loss 0.250105 LR 0.001000 Time 0.022357 +2023-10-02 20:57:46,504 - Epoch: [60][ 520/ 1236] Overall Loss 0.249799 Objective Loss 0.249799 LR 0.001000 Time 0.022328 +2023-10-02 20:57:46,711 - Epoch: [60][ 530/ 1236] Overall Loss 0.249658 Objective Loss 0.249658 LR 0.001000 Time 0.022296 +2023-10-02 20:57:46,920 - Epoch: [60][ 540/ 1236] Overall Loss 0.249997 Objective Loss 0.249997 LR 0.001000 Time 0.022270 +2023-10-02 20:57:47,128 - Epoch: [60][ 550/ 1236] Overall Loss 0.250084 Objective Loss 0.250084 LR 0.001000 Time 0.022242 +2023-10-02 20:57:47,340 - Epoch: [60][ 560/ 1236] Overall Loss 0.250086 Objective Loss 0.250086 LR 0.001000 Time 0.022224 +2023-10-02 20:57:47,552 - Epoch: [60][ 570/ 1236] Overall Loss 0.250252 Objective Loss 0.250252 LR 0.001000 Time 0.022202 +2023-10-02 20:57:47,765 - Epoch: [60][ 580/ 1236] Overall Loss 0.250717 Objective Loss 0.250717 LR 0.001000 Time 0.022185 +2023-10-02 20:57:47,973 - Epoch: [60][ 590/ 1236] Overall Loss 0.250675 Objective Loss 0.250675 LR 0.001000 Time 0.022162 +2023-10-02 20:57:48,185 - Epoch: [60][ 600/ 1236] Overall Loss 0.250663 Objective Loss 0.250663 LR 0.001000 Time 0.022145 +2023-10-02 20:57:48,394 - Epoch: [60][ 610/ 1236] Overall Loss 0.250728 Objective Loss 0.250728 LR 0.001000 Time 0.022124 +2023-10-02 20:57:48,606 - Epoch: [60][ 620/ 1236] Overall Loss 0.250794 Objective Loss 0.250794 LR 0.001000 Time 0.022108 +2023-10-02 20:57:48,814 - Epoch: [60][ 630/ 1236] Overall Loss 0.250665 Objective Loss 0.250665 LR 0.001000 Time 0.022088 +2023-10-02 20:57:49,027 - Epoch: [60][ 640/ 1236] Overall Loss 0.250167 Objective Loss 0.250167 LR 0.001000 Time 0.022074 +2023-10-02 20:57:49,237 - Epoch: [60][ 650/ 1236] Overall Loss 0.250320 Objective Loss 0.250320 LR 0.001000 Time 0.022058 +2023-10-02 20:57:49,450 - Epoch: [60][ 660/ 1236] Overall Loss 0.250470 Objective Loss 0.250470 LR 0.001000 Time 0.022045 +2023-10-02 20:57:49,657 - Epoch: [60][ 670/ 1236] Overall Loss 0.250385 Objective Loss 0.250385 LR 0.001000 Time 0.022025 +2023-10-02 20:57:49,870 - Epoch: [60][ 680/ 1236] Overall Loss 0.250692 Objective Loss 0.250692 LR 0.001000 Time 0.022013 +2023-10-02 20:57:50,079 - Epoch: [60][ 690/ 1236] Overall Loss 0.250886 Objective Loss 0.250886 LR 0.001000 Time 0.021996 +2023-10-02 20:57:50,289 - Epoch: [60][ 700/ 1236] Overall Loss 0.250970 Objective Loss 0.250970 LR 0.001000 Time 0.021982 +2023-10-02 20:57:50,498 - Epoch: [60][ 710/ 1236] Overall Loss 0.251202 Objective Loss 0.251202 LR 0.001000 Time 0.021967 +2023-10-02 20:57:50,710 - Epoch: [60][ 720/ 1236] Overall Loss 0.251101 Objective Loss 0.251101 LR 0.001000 Time 0.021956 +2023-10-02 20:57:50,919 - Epoch: [60][ 730/ 1236] Overall Loss 0.251060 Objective Loss 0.251060 LR 0.001000 Time 0.021939 +2023-10-02 20:57:51,132 - Epoch: [60][ 740/ 1236] Overall Loss 0.251168 Objective Loss 0.251168 LR 0.001000 Time 0.021930 +2023-10-02 20:57:51,340 - Epoch: [60][ 750/ 1236] Overall Loss 0.251189 Objective Loss 0.251189 LR 0.001000 Time 0.021915 +2023-10-02 20:57:51,553 - Epoch: [60][ 760/ 1236] Overall Loss 0.250984 Objective Loss 0.250984 LR 0.001000 Time 0.021906 +2023-10-02 20:57:51,762 - Epoch: [60][ 770/ 1236] Overall Loss 0.251162 Objective Loss 0.251162 LR 0.001000 Time 0.021893 +2023-10-02 20:57:51,974 - Epoch: [60][ 780/ 1236] Overall Loss 0.250927 Objective Loss 0.250927 LR 0.001000 Time 0.021883 +2023-10-02 20:57:52,184 - Epoch: [60][ 790/ 1236] Overall Loss 0.251239 Objective Loss 0.251239 LR 0.001000 Time 0.021871 +2023-10-02 20:57:52,396 - Epoch: [60][ 800/ 1236] Overall Loss 0.251027 Objective Loss 0.251027 LR 0.001000 Time 0.021862 +2023-10-02 20:57:52,605 - Epoch: [60][ 810/ 1236] Overall Loss 0.251225 Objective Loss 0.251225 LR 0.001000 Time 0.021850 +2023-10-02 20:57:52,816 - Epoch: [60][ 820/ 1236] Overall Loss 0.251405 Objective Loss 0.251405 LR 0.001000 Time 0.021841 +2023-10-02 20:57:53,024 - Epoch: [60][ 830/ 1236] Overall Loss 0.251675 Objective Loss 0.251675 LR 0.001000 Time 0.021828 +2023-10-02 20:57:53,238 - Epoch: [60][ 840/ 1236] Overall Loss 0.251831 Objective Loss 0.251831 LR 0.001000 Time 0.021822 +2023-10-02 20:57:53,449 - Epoch: [60][ 850/ 1236] Overall Loss 0.251965 Objective Loss 0.251965 LR 0.001000 Time 0.021813 +2023-10-02 20:57:53,661 - Epoch: [60][ 860/ 1236] Overall Loss 0.251912 Objective Loss 0.251912 LR 0.001000 Time 0.021805 +2023-10-02 20:57:53,868 - Epoch: [60][ 870/ 1236] Overall Loss 0.251778 Objective Loss 0.251778 LR 0.001000 Time 0.021793 +2023-10-02 20:57:54,079 - Epoch: [60][ 880/ 1236] Overall Loss 0.252175 Objective Loss 0.252175 LR 0.001000 Time 0.021784 +2023-10-02 20:57:54,287 - Epoch: [60][ 890/ 1236] Overall Loss 0.252563 Objective Loss 0.252563 LR 0.001000 Time 0.021773 +2023-10-02 20:57:54,499 - Epoch: [60][ 900/ 1236] Overall Loss 0.252901 Objective Loss 0.252901 LR 0.001000 Time 0.021766 +2023-10-02 20:57:54,708 - Epoch: [60][ 910/ 1236] Overall Loss 0.252886 Objective Loss 0.252886 LR 0.001000 Time 0.021756 +2023-10-02 20:57:54,919 - Epoch: [60][ 920/ 1236] Overall Loss 0.252827 Objective Loss 0.252827 LR 0.001000 Time 0.021748 +2023-10-02 20:57:55,127 - Epoch: [60][ 930/ 1236] Overall Loss 0.253150 Objective Loss 0.253150 LR 0.001000 Time 0.021738 +2023-10-02 20:57:55,340 - Epoch: [60][ 940/ 1236] Overall Loss 0.253139 Objective Loss 0.253139 LR 0.001000 Time 0.021732 +2023-10-02 20:57:55,547 - Epoch: [60][ 950/ 1236] Overall Loss 0.252780 Objective Loss 0.252780 LR 0.001000 Time 0.021721 +2023-10-02 20:57:55,764 - Epoch: [60][ 960/ 1236] Overall Loss 0.252663 Objective Loss 0.252663 LR 0.001000 Time 0.021721 +2023-10-02 20:57:55,980 - Epoch: [60][ 970/ 1236] Overall Loss 0.252558 Objective Loss 0.252558 LR 0.001000 Time 0.021719 +2023-10-02 20:57:56,197 - Epoch: [60][ 980/ 1236] Overall Loss 0.252452 Objective Loss 0.252452 LR 0.001000 Time 0.021718 +2023-10-02 20:57:56,412 - Epoch: [60][ 990/ 1236] Overall Loss 0.252310 Objective Loss 0.252310 LR 0.001000 Time 0.021714 +2023-10-02 20:57:56,629 - Epoch: [60][ 1000/ 1236] Overall Loss 0.252416 Objective Loss 0.252416 LR 0.001000 Time 0.021712 +2023-10-02 20:57:56,844 - Epoch: [60][ 1010/ 1236] Overall Loss 0.252478 Objective Loss 0.252478 LR 0.001000 Time 0.021709 +2023-10-02 20:57:57,061 - Epoch: [60][ 1020/ 1236] Overall Loss 0.252341 Objective Loss 0.252341 LR 0.001000 Time 0.021709 +2023-10-02 20:57:57,277 - Epoch: [60][ 1030/ 1236] Overall Loss 0.252220 Objective Loss 0.252220 LR 0.001000 Time 0.021706 +2023-10-02 20:57:57,493 - Epoch: [60][ 1040/ 1236] Overall Loss 0.252813 Objective Loss 0.252813 LR 0.001000 Time 0.021704 +2023-10-02 20:57:57,709 - Epoch: [60][ 1050/ 1236] Overall Loss 0.252772 Objective Loss 0.252772 LR 0.001000 Time 0.021702 +2023-10-02 20:57:57,927 - Epoch: [60][ 1060/ 1236] Overall Loss 0.253028 Objective Loss 0.253028 LR 0.001000 Time 0.021703 +2023-10-02 20:57:58,142 - Epoch: [60][ 1070/ 1236] Overall Loss 0.253188 Objective Loss 0.253188 LR 0.001000 Time 0.021699 +2023-10-02 20:57:58,358 - Epoch: [60][ 1080/ 1236] Overall Loss 0.253344 Objective Loss 0.253344 LR 0.001000 Time 0.021697 +2023-10-02 20:57:58,573 - Epoch: [60][ 1090/ 1236] Overall Loss 0.253557 Objective Loss 0.253557 LR 0.001000 Time 0.021693 +2023-10-02 20:57:58,789 - Epoch: [60][ 1100/ 1236] Overall Loss 0.253733 Objective Loss 0.253733 LR 0.001000 Time 0.021691 +2023-10-02 20:57:59,004 - Epoch: [60][ 1110/ 1236] Overall Loss 0.253989 Objective Loss 0.253989 LR 0.001000 Time 0.021687 +2023-10-02 20:57:59,220 - Epoch: [60][ 1120/ 1236] Overall Loss 0.253842 Objective Loss 0.253842 LR 0.001000 Time 0.021686 +2023-10-02 20:57:59,434 - Epoch: [60][ 1130/ 1236] Overall Loss 0.254248 Objective Loss 0.254248 LR 0.001000 Time 0.021682 +2023-10-02 20:57:59,651 - Epoch: [60][ 1140/ 1236] Overall Loss 0.254173 Objective Loss 0.254173 LR 0.001000 Time 0.021682 +2023-10-02 20:57:59,866 - Epoch: [60][ 1150/ 1236] Overall Loss 0.254667 Objective Loss 0.254667 LR 0.001000 Time 0.021679 +2023-10-02 20:58:00,084 - Epoch: [60][ 1160/ 1236] Overall Loss 0.254796 Objective Loss 0.254796 LR 0.001000 Time 0.021678 +2023-10-02 20:58:00,299 - Epoch: [60][ 1170/ 1236] Overall Loss 0.255116 Objective Loss 0.255116 LR 0.001000 Time 0.021675 +2023-10-02 20:58:00,514 - Epoch: [60][ 1180/ 1236] Overall Loss 0.255228 Objective Loss 0.255228 LR 0.001000 Time 0.021673 +2023-10-02 20:58:00,729 - Epoch: [60][ 1190/ 1236] Overall Loss 0.255249 Objective Loss 0.255249 LR 0.001000 Time 0.021670 +2023-10-02 20:58:00,944 - Epoch: [60][ 1200/ 1236] Overall Loss 0.255167 Objective Loss 0.255167 LR 0.001000 Time 0.021667 +2023-10-02 20:58:01,159 - Epoch: [60][ 1210/ 1236] Overall Loss 0.255137 Objective Loss 0.255137 LR 0.001000 Time 0.021664 +2023-10-02 20:58:01,375 - Epoch: [60][ 1220/ 1236] Overall Loss 0.255292 Objective Loss 0.255292 LR 0.001000 Time 0.021663 +2023-10-02 20:58:01,643 - Epoch: [60][ 1230/ 1236] Overall Loss 0.255429 Objective Loss 0.255429 LR 0.001000 Time 0.021703 +2023-10-02 20:58:01,765 - Epoch: [60][ 1236/ 1236] Overall Loss 0.255412 Objective Loss 0.255412 Top1 84.725051 Top5 99.185336 LR 0.001000 Time 0.021696 +2023-10-02 20:58:01,906 - --- validate (epoch=60)----------- +2023-10-02 20:58:01,907 - 29943 samples (256 per mini-batch) +2023-10-02 20:58:02,401 - Epoch: [60][ 10/ 117] Loss 0.367151 Top1 81.796875 Top5 97.812500 +2023-10-02 20:58:02,556 - Epoch: [60][ 20/ 117] Loss 0.365376 Top1 81.855469 Top5 97.910156 +2023-10-02 20:58:02,710 - Epoch: [60][ 30/ 117] Loss 0.339539 Top1 82.630208 Top5 98.085938 +2023-10-02 20:58:02,867 - Epoch: [60][ 40/ 117] Loss 0.347994 Top1 82.441406 Top5 98.085938 +2023-10-02 20:58:03,025 - Epoch: [60][ 50/ 117] Loss 0.342312 Top1 82.429688 Top5 98.125000 +2023-10-02 20:58:03,184 - Epoch: [60][ 60/ 117] Loss 0.340266 Top1 82.421875 Top5 98.144531 +2023-10-02 20:58:03,341 - Epoch: [60][ 70/ 117] Loss 0.333092 Top1 82.572545 Top5 98.141741 +2023-10-02 20:58:03,500 - Epoch: [60][ 80/ 117] Loss 0.335100 Top1 82.490234 Top5 98.100586 +2023-10-02 20:58:03,659 - Epoch: [60][ 90/ 117] Loss 0.334097 Top1 82.573785 Top5 98.116319 +2023-10-02 20:58:03,819 - Epoch: [60][ 100/ 117] Loss 0.334534 Top1 82.519531 Top5 98.074219 +2023-10-02 20:58:03,987 - Epoch: [60][ 110/ 117] Loss 0.331846 Top1 82.546165 Top5 98.046875 +2023-10-02 20:58:04,077 - Epoch: [60][ 117/ 117] Loss 0.332542 Top1 82.490064 Top5 98.009551 +2023-10-02 20:58:04,176 - ==> Top1: 82.490 Top5: 98.010 Loss: 0.333 + +2023-10-02 20:58:04,176 - ==> Confusion: +[[ 927 2 2 0 6 4 0 0 7 70 2 1 1 2 6 1 3 2 1 0 13] + [ 0 1064 0 1 3 20 0 27 0 2 1 1 1 1 0 2 0 0 6 1 1] + [ 4 0 956 10 5 0 14 11 0 2 2 2 10 3 1 3 1 2 13 4 13] + [ 1 4 7 967 0 3 2 2 8 0 6 0 3 7 20 1 0 4 36 0 18] + [ 32 10 1 0 950 7 0 0 0 11 1 4 1 3 7 3 16 2 0 1 1] + [ 2 67 1 2 2 957 2 20 4 3 5 10 6 15 4 1 2 3 3 2 5] + [ 0 8 36 0 0 0 1095 14 0 0 6 4 1 1 1 5 0 1 1 9 9] + [ 1 27 18 1 5 22 2 1048 2 5 5 7 6 2 1 0 3 3 47 9 4] + [ 18 3 0 0 2 1 0 2 985 36 8 2 0 12 8 2 2 3 2 0 3] + [ 87 2 0 0 4 2 0 0 44 937 2 0 1 20 10 0 0 0 0 2 8] + [ 1 5 14 4 0 2 2 3 19 1 958 1 0 16 3 0 0 3 8 3 10] + [ 1 1 0 0 0 10 0 1 0 0 0 950 34 11 0 2 1 15 0 7 2] + [ 0 2 1 1 0 2 2 2 3 1 2 46 954 3 1 6 5 19 3 5 10] + [ 1 0 0 0 4 6 0 0 21 10 5 5 1 1044 6 0 0 1 0 1 14] + [ 14 2 5 17 2 0 0 0 24 5 2 0 4 6 990 0 2 3 18 0 7] + [ 0 1 2 1 4 0 1 0 0 1 0 9 11 0 0 1056 13 17 2 9 7] + [ 1 17 1 0 6 8 0 0 0 1 0 7 2 2 3 9 1097 0 1 3 3] + [ 0 1 0 11 0 0 2 0 0 0 0 5 27 1 4 3 2 979 0 1 2] + [ 3 11 5 9 0 0 0 9 3 1 2 0 0 0 8 1 1 0 1004 0 11] + [ 0 7 4 0 0 7 5 17 0 1 0 14 5 5 0 2 10 1 3 1071 0] + [ 162 296 112 78 85 179 26 105 156 87 199 164 372 329 135 45 147 76 207 234 4711]] + +2023-10-02 20:58:04,177 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:58:04,178 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:58:04,184 - + +2023-10-02 20:58:04,184 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:58:05,209 - Epoch: [61][ 10/ 1236] Overall Loss 0.248665 Objective Loss 0.248665 LR 0.001000 Time 0.102456 +2023-10-02 20:58:05,419 - Epoch: [61][ 20/ 1236] Overall Loss 0.259311 Objective Loss 0.259311 LR 0.001000 Time 0.061712 +2023-10-02 20:58:05,628 - Epoch: [61][ 30/ 1236] Overall Loss 0.257365 Objective Loss 0.257365 LR 0.001000 Time 0.048047 +2023-10-02 20:58:05,837 - Epoch: [61][ 40/ 1236] Overall Loss 0.253091 Objective Loss 0.253091 LR 0.001000 Time 0.041258 +2023-10-02 20:58:06,045 - Epoch: [61][ 50/ 1236] Overall Loss 0.249372 Objective Loss 0.249372 LR 0.001000 Time 0.037143 +2023-10-02 20:58:06,255 - Epoch: [61][ 60/ 1236] Overall Loss 0.246690 Objective Loss 0.246690 LR 0.001000 Time 0.034440 +2023-10-02 20:58:06,464 - Epoch: [61][ 70/ 1236] Overall Loss 0.245116 Objective Loss 0.245116 LR 0.001000 Time 0.032479 +2023-10-02 20:58:06,674 - Epoch: [61][ 80/ 1236] Overall Loss 0.243421 Objective Loss 0.243421 LR 0.001000 Time 0.031044 +2023-10-02 20:58:06,885 - Epoch: [61][ 90/ 1236] Overall Loss 0.243897 Objective Loss 0.243897 LR 0.001000 Time 0.029926 +2023-10-02 20:58:07,095 - Epoch: [61][ 100/ 1236] Overall Loss 0.245484 Objective Loss 0.245484 LR 0.001000 Time 0.029030 +2023-10-02 20:58:07,304 - Epoch: [61][ 110/ 1236] Overall Loss 0.243743 Objective Loss 0.243743 LR 0.001000 Time 0.028272 +2023-10-02 20:58:07,518 - Epoch: [61][ 120/ 1236] Overall Loss 0.244481 Objective Loss 0.244481 LR 0.001000 Time 0.027697 +2023-10-02 20:58:07,726 - Epoch: [61][ 130/ 1236] Overall Loss 0.245014 Objective Loss 0.245014 LR 0.001000 Time 0.027169 +2023-10-02 20:58:07,936 - Epoch: [61][ 140/ 1236] Overall Loss 0.246053 Objective Loss 0.246053 LR 0.001000 Time 0.026721 +2023-10-02 20:58:08,144 - Epoch: [61][ 150/ 1236] Overall Loss 0.245380 Objective Loss 0.245380 LR 0.001000 Time 0.026327 +2023-10-02 20:58:08,353 - Epoch: [61][ 160/ 1236] Overall Loss 0.244571 Objective Loss 0.244571 LR 0.001000 Time 0.025988 +2023-10-02 20:58:08,565 - Epoch: [61][ 170/ 1236] Overall Loss 0.244992 Objective Loss 0.244992 LR 0.001000 Time 0.025705 +2023-10-02 20:58:08,776 - Epoch: [61][ 180/ 1236] Overall Loss 0.245781 Objective Loss 0.245781 LR 0.001000 Time 0.025443 +2023-10-02 20:58:08,984 - Epoch: [61][ 190/ 1236] Overall Loss 0.246042 Objective Loss 0.246042 LR 0.001000 Time 0.025197 +2023-10-02 20:58:09,194 - Epoch: [61][ 200/ 1236] Overall Loss 0.245269 Objective Loss 0.245269 LR 0.001000 Time 0.024986 +2023-10-02 20:58:09,403 - Epoch: [61][ 210/ 1236] Overall Loss 0.245511 Objective Loss 0.245511 LR 0.001000 Time 0.024790 +2023-10-02 20:58:09,612 - Epoch: [61][ 220/ 1236] Overall Loss 0.245454 Objective Loss 0.245454 LR 0.001000 Time 0.024614 +2023-10-02 20:58:09,821 - Epoch: [61][ 230/ 1236] Overall Loss 0.246141 Objective Loss 0.246141 LR 0.001000 Time 0.024448 +2023-10-02 20:58:10,033 - Epoch: [61][ 240/ 1236] Overall Loss 0.247163 Objective Loss 0.247163 LR 0.001000 Time 0.024312 +2023-10-02 20:58:10,245 - Epoch: [61][ 250/ 1236] Overall Loss 0.246702 Objective Loss 0.246702 LR 0.001000 Time 0.024186 +2023-10-02 20:58:10,456 - Epoch: [61][ 260/ 1236] Overall Loss 0.246709 Objective Loss 0.246709 LR 0.001000 Time 0.024065 +2023-10-02 20:58:10,664 - Epoch: [61][ 270/ 1236] Overall Loss 0.247097 Objective Loss 0.247097 LR 0.001000 Time 0.023945 +2023-10-02 20:58:10,874 - Epoch: [61][ 280/ 1236] Overall Loss 0.247489 Objective Loss 0.247489 LR 0.001000 Time 0.023839 +2023-10-02 20:58:11,083 - Epoch: [61][ 290/ 1236] Overall Loss 0.248409 Objective Loss 0.248409 LR 0.001000 Time 0.023734 +2023-10-02 20:58:11,293 - Epoch: [61][ 300/ 1236] Overall Loss 0.249367 Objective Loss 0.249367 LR 0.001000 Time 0.023643 +2023-10-02 20:58:11,502 - Epoch: [61][ 310/ 1236] Overall Loss 0.249425 Objective Loss 0.249425 LR 0.001000 Time 0.023553 +2023-10-02 20:58:11,713 - Epoch: [61][ 320/ 1236] Overall Loss 0.249206 Objective Loss 0.249206 LR 0.001000 Time 0.023475 +2023-10-02 20:58:11,924 - Epoch: [61][ 330/ 1236] Overall Loss 0.249859 Objective Loss 0.249859 LR 0.001000 Time 0.023399 +2023-10-02 20:58:12,134 - Epoch: [61][ 340/ 1236] Overall Loss 0.249459 Objective Loss 0.249459 LR 0.001000 Time 0.023329 +2023-10-02 20:58:12,344 - Epoch: [61][ 350/ 1236] Overall Loss 0.249341 Objective Loss 0.249341 LR 0.001000 Time 0.023257 +2023-10-02 20:58:12,557 - Epoch: [61][ 360/ 1236] Overall Loss 0.248787 Objective Loss 0.248787 LR 0.001000 Time 0.023200 +2023-10-02 20:58:12,766 - Epoch: [61][ 370/ 1236] Overall Loss 0.248817 Objective Loss 0.248817 LR 0.001000 Time 0.023138 +2023-10-02 20:58:12,982 - Epoch: [61][ 380/ 1236] Overall Loss 0.248923 Objective Loss 0.248923 LR 0.001000 Time 0.023098 +2023-10-02 20:58:13,192 - Epoch: [61][ 390/ 1236] Overall Loss 0.249049 Objective Loss 0.249049 LR 0.001000 Time 0.023042 +2023-10-02 20:58:13,403 - Epoch: [61][ 400/ 1236] Overall Loss 0.249665 Objective Loss 0.249665 LR 0.001000 Time 0.022991 +2023-10-02 20:58:13,612 - Epoch: [61][ 410/ 1236] Overall Loss 0.249998 Objective Loss 0.249998 LR 0.001000 Time 0.022938 +2023-10-02 20:58:13,823 - Epoch: [61][ 420/ 1236] Overall Loss 0.250508 Objective Loss 0.250508 LR 0.001000 Time 0.022892 +2023-10-02 20:58:14,033 - Epoch: [61][ 430/ 1236] Overall Loss 0.250831 Objective Loss 0.250831 LR 0.001000 Time 0.022847 +2023-10-02 20:58:14,243 - Epoch: [61][ 440/ 1236] Overall Loss 0.250218 Objective Loss 0.250218 LR 0.001000 Time 0.022805 +2023-10-02 20:58:14,453 - Epoch: [61][ 450/ 1236] Overall Loss 0.250328 Objective Loss 0.250328 LR 0.001000 Time 0.022763 +2023-10-02 20:58:14,664 - Epoch: [61][ 460/ 1236] Overall Loss 0.251147 Objective Loss 0.251147 LR 0.001000 Time 0.022726 +2023-10-02 20:58:14,874 - Epoch: [61][ 470/ 1236] Overall Loss 0.251152 Objective Loss 0.251152 LR 0.001000 Time 0.022689 +2023-10-02 20:58:15,090 - Epoch: [61][ 480/ 1236] Overall Loss 0.250960 Objective Loss 0.250960 LR 0.001000 Time 0.022665 +2023-10-02 20:58:15,299 - Epoch: [61][ 490/ 1236] Overall Loss 0.251379 Objective Loss 0.251379 LR 0.001000 Time 0.022629 +2023-10-02 20:58:15,510 - Epoch: [61][ 500/ 1236] Overall Loss 0.251337 Objective Loss 0.251337 LR 0.001000 Time 0.022597 +2023-10-02 20:58:15,719 - Epoch: [61][ 510/ 1236] Overall Loss 0.250731 Objective Loss 0.250731 LR 0.001000 Time 0.022562 +2023-10-02 20:58:15,931 - Epoch: [61][ 520/ 1236] Overall Loss 0.250841 Objective Loss 0.250841 LR 0.001000 Time 0.022534 +2023-10-02 20:58:16,140 - Epoch: [61][ 530/ 1236] Overall Loss 0.250368 Objective Loss 0.250368 LR 0.001000 Time 0.022501 +2023-10-02 20:58:16,357 - Epoch: [61][ 540/ 1236] Overall Loss 0.251162 Objective Loss 0.251162 LR 0.001000 Time 0.022486 +2023-10-02 20:58:16,568 - Epoch: [61][ 550/ 1236] Overall Loss 0.251270 Objective Loss 0.251270 LR 0.001000 Time 0.022457 +2023-10-02 20:58:16,779 - Epoch: [61][ 560/ 1236] Overall Loss 0.251415 Objective Loss 0.251415 LR 0.001000 Time 0.022432 +2023-10-02 20:58:16,988 - Epoch: [61][ 570/ 1236] Overall Loss 0.251231 Objective Loss 0.251231 LR 0.001000 Time 0.022405 +2023-10-02 20:58:17,199 - Epoch: [61][ 580/ 1236] Overall Loss 0.251184 Objective Loss 0.251184 LR 0.001000 Time 0.022381 +2023-10-02 20:58:17,409 - Epoch: [61][ 590/ 1236] Overall Loss 0.251502 Objective Loss 0.251502 LR 0.001000 Time 0.022357 +2023-10-02 20:58:17,621 - Epoch: [61][ 600/ 1236] Overall Loss 0.251169 Objective Loss 0.251169 LR 0.001000 Time 0.022337 +2023-10-02 20:58:17,830 - Epoch: [61][ 610/ 1236] Overall Loss 0.250800 Objective Loss 0.250800 LR 0.001000 Time 0.022312 +2023-10-02 20:58:18,044 - Epoch: [61][ 620/ 1236] Overall Loss 0.251230 Objective Loss 0.251230 LR 0.001000 Time 0.022297 +2023-10-02 20:58:18,261 - Epoch: [61][ 630/ 1236] Overall Loss 0.251217 Objective Loss 0.251217 LR 0.001000 Time 0.022287 +2023-10-02 20:58:18,476 - Epoch: [61][ 640/ 1236] Overall Loss 0.250927 Objective Loss 0.250927 LR 0.001000 Time 0.022273 +2023-10-02 20:58:18,690 - Epoch: [61][ 650/ 1236] Overall Loss 0.251392 Objective Loss 0.251392 LR 0.001000 Time 0.022257 +2023-10-02 20:58:18,905 - Epoch: [61][ 660/ 1236] Overall Loss 0.251753 Objective Loss 0.251753 LR 0.001000 Time 0.022244 +2023-10-02 20:58:19,118 - Epoch: [61][ 670/ 1236] Overall Loss 0.252069 Objective Loss 0.252069 LR 0.001000 Time 0.022229 +2023-10-02 20:58:19,332 - Epoch: [61][ 680/ 1236] Overall Loss 0.252325 Objective Loss 0.252325 LR 0.001000 Time 0.022216 +2023-10-02 20:58:19,546 - Epoch: [61][ 690/ 1236] Overall Loss 0.252842 Objective Loss 0.252842 LR 0.001000 Time 0.022202 +2023-10-02 20:58:19,767 - Epoch: [61][ 700/ 1236] Overall Loss 0.252927 Objective Loss 0.252927 LR 0.001000 Time 0.022200 +2023-10-02 20:58:19,989 - Epoch: [61][ 710/ 1236] Overall Loss 0.253459 Objective Loss 0.253459 LR 0.001000 Time 0.022198 +2023-10-02 20:58:20,213 - Epoch: [61][ 720/ 1236] Overall Loss 0.253667 Objective Loss 0.253667 LR 0.001000 Time 0.022201 +2023-10-02 20:58:20,434 - Epoch: [61][ 730/ 1236] Overall Loss 0.253288 Objective Loss 0.253288 LR 0.001000 Time 0.022199 +2023-10-02 20:58:20,649 - Epoch: [61][ 740/ 1236] Overall Loss 0.253668 Objective Loss 0.253668 LR 0.001000 Time 0.022189 +2023-10-02 20:58:20,855 - Epoch: [61][ 750/ 1236] Overall Loss 0.253933 Objective Loss 0.253933 LR 0.001000 Time 0.022168 +2023-10-02 20:58:21,061 - Epoch: [61][ 760/ 1236] Overall Loss 0.254466 Objective Loss 0.254466 LR 0.001000 Time 0.022147 +2023-10-02 20:58:21,266 - Epoch: [61][ 770/ 1236] Overall Loss 0.254371 Objective Loss 0.254371 LR 0.001000 Time 0.022125 +2023-10-02 20:58:21,472 - Epoch: [61][ 780/ 1236] Overall Loss 0.254365 Objective Loss 0.254365 LR 0.001000 Time 0.022104 +2023-10-02 20:58:21,677 - Epoch: [61][ 790/ 1236] Overall Loss 0.254684 Objective Loss 0.254684 LR 0.001000 Time 0.022084 +2023-10-02 20:58:21,882 - Epoch: [61][ 800/ 1236] Overall Loss 0.254693 Objective Loss 0.254693 LR 0.001000 Time 0.022064 +2023-10-02 20:58:22,088 - Epoch: [61][ 810/ 1236] Overall Loss 0.254499 Objective Loss 0.254499 LR 0.001000 Time 0.022045 +2023-10-02 20:58:22,294 - Epoch: [61][ 820/ 1236] Overall Loss 0.254796 Objective Loss 0.254796 LR 0.001000 Time 0.022027 +2023-10-02 20:58:22,500 - Epoch: [61][ 830/ 1236] Overall Loss 0.254961 Objective Loss 0.254961 LR 0.001000 Time 0.022010 +2023-10-02 20:58:22,706 - Epoch: [61][ 840/ 1236] Overall Loss 0.254973 Objective Loss 0.254973 LR 0.001000 Time 0.021992 +2023-10-02 20:58:22,912 - Epoch: [61][ 850/ 1236] Overall Loss 0.254770 Objective Loss 0.254770 LR 0.001000 Time 0.021975 +2023-10-02 20:58:23,119 - Epoch: [61][ 860/ 1236] Overall Loss 0.254676 Objective Loss 0.254676 LR 0.001000 Time 0.021960 +2023-10-02 20:58:23,334 - Epoch: [61][ 870/ 1236] Overall Loss 0.254560 Objective Loss 0.254560 LR 0.001000 Time 0.021954 +2023-10-02 20:58:23,545 - Epoch: [61][ 880/ 1236] Overall Loss 0.254834 Objective Loss 0.254834 LR 0.001000 Time 0.021944 +2023-10-02 20:58:23,753 - Epoch: [61][ 890/ 1236] Overall Loss 0.254932 Objective Loss 0.254932 LR 0.001000 Time 0.021931 +2023-10-02 20:58:23,963 - Epoch: [61][ 900/ 1236] Overall Loss 0.255246 Objective Loss 0.255246 LR 0.001000 Time 0.021920 +2023-10-02 20:58:24,171 - Epoch: [61][ 910/ 1236] Overall Loss 0.255391 Objective Loss 0.255391 LR 0.001000 Time 0.021907 +2023-10-02 20:58:24,381 - Epoch: [61][ 920/ 1236] Overall Loss 0.255559 Objective Loss 0.255559 LR 0.001000 Time 0.021897 +2023-10-02 20:58:24,588 - Epoch: [61][ 930/ 1236] Overall Loss 0.255505 Objective Loss 0.255505 LR 0.001000 Time 0.021884 +2023-10-02 20:58:24,799 - Epoch: [61][ 940/ 1236] Overall Loss 0.255392 Objective Loss 0.255392 LR 0.001000 Time 0.021875 +2023-10-02 20:58:25,006 - Epoch: [61][ 950/ 1236] Overall Loss 0.255684 Objective Loss 0.255684 LR 0.001000 Time 0.021863 +2023-10-02 20:58:25,216 - Epoch: [61][ 960/ 1236] Overall Loss 0.255652 Objective Loss 0.255652 LR 0.001000 Time 0.021853 +2023-10-02 20:58:25,423 - Epoch: [61][ 970/ 1236] Overall Loss 0.255482 Objective Loss 0.255482 LR 0.001000 Time 0.021842 +2023-10-02 20:58:25,633 - Epoch: [61][ 980/ 1236] Overall Loss 0.255493 Objective Loss 0.255493 LR 0.001000 Time 0.021832 +2023-10-02 20:58:25,841 - Epoch: [61][ 990/ 1236] Overall Loss 0.255201 Objective Loss 0.255201 LR 0.001000 Time 0.021821 +2023-10-02 20:58:26,051 - Epoch: [61][ 1000/ 1236] Overall Loss 0.255244 Objective Loss 0.255244 LR 0.001000 Time 0.021813 +2023-10-02 20:58:26,259 - Epoch: [61][ 1010/ 1236] Overall Loss 0.255394 Objective Loss 0.255394 LR 0.001000 Time 0.021802 +2023-10-02 20:58:26,468 - Epoch: [61][ 1020/ 1236] Overall Loss 0.255433 Objective Loss 0.255433 LR 0.001000 Time 0.021794 +2023-10-02 20:58:26,676 - Epoch: [61][ 1030/ 1236] Overall Loss 0.255810 Objective Loss 0.255810 LR 0.001000 Time 0.021783 +2023-10-02 20:58:26,887 - Epoch: [61][ 1040/ 1236] Overall Loss 0.256121 Objective Loss 0.256121 LR 0.001000 Time 0.021776 +2023-10-02 20:58:27,094 - Epoch: [61][ 1050/ 1236] Overall Loss 0.256213 Objective Loss 0.256213 LR 0.001000 Time 0.021766 +2023-10-02 20:58:27,304 - Epoch: [61][ 1060/ 1236] Overall Loss 0.256382 Objective Loss 0.256382 LR 0.001000 Time 0.021758 +2023-10-02 20:58:27,511 - Epoch: [61][ 1070/ 1236] Overall Loss 0.256251 Objective Loss 0.256251 LR 0.001000 Time 0.021748 +2023-10-02 20:58:27,722 - Epoch: [61][ 1080/ 1236] Overall Loss 0.256150 Objective Loss 0.256150 LR 0.001000 Time 0.021742 +2023-10-02 20:58:27,930 - Epoch: [61][ 1090/ 1236] Overall Loss 0.256000 Objective Loss 0.256000 LR 0.001000 Time 0.021733 +2023-10-02 20:58:28,141 - Epoch: [61][ 1100/ 1236] Overall Loss 0.256261 Objective Loss 0.256261 LR 0.001000 Time 0.021726 +2023-10-02 20:58:28,348 - Epoch: [61][ 1110/ 1236] Overall Loss 0.256446 Objective Loss 0.256446 LR 0.001000 Time 0.021717 +2023-10-02 20:58:28,559 - Epoch: [61][ 1120/ 1236] Overall Loss 0.256567 Objective Loss 0.256567 LR 0.001000 Time 0.021711 +2023-10-02 20:58:28,766 - Epoch: [61][ 1130/ 1236] Overall Loss 0.256369 Objective Loss 0.256369 LR 0.001000 Time 0.021702 +2023-10-02 20:58:28,977 - Epoch: [61][ 1140/ 1236] Overall Loss 0.256439 Objective Loss 0.256439 LR 0.001000 Time 0.021696 +2023-10-02 20:58:29,184 - Epoch: [61][ 1150/ 1236] Overall Loss 0.256555 Objective Loss 0.256555 LR 0.001000 Time 0.021688 +2023-10-02 20:58:29,395 - Epoch: [61][ 1160/ 1236] Overall Loss 0.256604 Objective Loss 0.256604 LR 0.001000 Time 0.021682 +2023-10-02 20:58:29,602 - Epoch: [61][ 1170/ 1236] Overall Loss 0.256725 Objective Loss 0.256725 LR 0.001000 Time 0.021674 +2023-10-02 20:58:29,811 - Epoch: [61][ 1180/ 1236] Overall Loss 0.256472 Objective Loss 0.256472 LR 0.001000 Time 0.021667 +2023-10-02 20:58:30,018 - Epoch: [61][ 1190/ 1236] Overall Loss 0.256745 Objective Loss 0.256745 LR 0.001000 Time 0.021658 +2023-10-02 20:58:30,225 - Epoch: [61][ 1200/ 1236] Overall Loss 0.257023 Objective Loss 0.257023 LR 0.001000 Time 0.021651 +2023-10-02 20:58:30,432 - Epoch: [61][ 1210/ 1236] Overall Loss 0.257066 Objective Loss 0.257066 LR 0.001000 Time 0.021642 +2023-10-02 20:58:30,640 - Epoch: [61][ 1220/ 1236] Overall Loss 0.257245 Objective Loss 0.257245 LR 0.001000 Time 0.021635 +2023-10-02 20:58:30,899 - Epoch: [61][ 1230/ 1236] Overall Loss 0.257393 Objective Loss 0.257393 LR 0.001000 Time 0.021670 +2023-10-02 20:58:31,022 - Epoch: [61][ 1236/ 1236] Overall Loss 0.257335 Objective Loss 0.257335 Top1 87.169043 Top5 98.574338 LR 0.001000 Time 0.021663 +2023-10-02 20:58:31,157 - --- validate (epoch=61)----------- +2023-10-02 20:58:31,157 - 29943 samples (256 per mini-batch) +2023-10-02 20:58:31,619 - Epoch: [61][ 10/ 117] Loss 0.353325 Top1 84.531250 Top5 98.437500 +2023-10-02 20:58:31,770 - Epoch: [61][ 20/ 117] Loss 0.350074 Top1 83.593750 Top5 98.281250 +2023-10-02 20:58:31,920 - Epoch: [61][ 30/ 117] Loss 0.325808 Top1 83.463542 Top5 98.320312 +2023-10-02 20:58:32,070 - Epoch: [61][ 40/ 117] Loss 0.333368 Top1 83.271484 Top5 98.183594 +2023-10-02 20:58:32,220 - Epoch: [61][ 50/ 117] Loss 0.328872 Top1 83.507812 Top5 98.195312 +2023-10-02 20:58:32,369 - Epoch: [61][ 60/ 117] Loss 0.331912 Top1 83.372396 Top5 98.144531 +2023-10-02 20:58:32,520 - Epoch: [61][ 70/ 117] Loss 0.330988 Top1 83.337054 Top5 98.108259 +2023-10-02 20:58:32,670 - Epoch: [61][ 80/ 117] Loss 0.327807 Top1 83.491211 Top5 98.139648 +2023-10-02 20:58:32,821 - Epoch: [61][ 90/ 117] Loss 0.327619 Top1 83.511285 Top5 98.120660 +2023-10-02 20:58:32,972 - Epoch: [61][ 100/ 117] Loss 0.327064 Top1 83.460938 Top5 98.078125 +2023-10-02 20:58:33,130 - Epoch: [61][ 110/ 117] Loss 0.329797 Top1 83.299006 Top5 98.043324 +2023-10-02 20:58:33,218 - Epoch: [61][ 117/ 117] Loss 0.330293 Top1 83.204756 Top5 98.062986 +2023-10-02 20:58:33,365 - ==> Top1: 83.205 Top5: 98.063 Loss: 0.330 + +2023-10-02 20:58:33,366 - ==> Confusion: +[[ 916 0 4 1 5 2 0 0 6 84 1 0 2 1 6 2 3 6 0 1 10] + [ 0 1041 0 0 7 21 3 18 1 1 1 2 1 0 1 4 7 0 15 3 5] + [ 5 0 961 10 2 0 21 6 0 4 1 1 7 3 1 9 1 2 16 2 4] + [ 0 1 15 975 0 1 0 3 2 2 8 0 8 4 31 3 4 3 12 1 16] + [ 24 2 1 0 968 5 0 0 1 11 0 0 3 0 13 5 10 0 0 1 6] + [ 4 49 1 2 3 946 1 23 0 8 2 12 8 17 3 1 5 4 8 7 12] + [ 1 3 32 0 0 0 1119 2 0 0 3 3 2 1 0 6 0 1 1 11 6] + [ 5 25 21 1 3 19 6 1034 1 1 9 10 4 5 2 1 1 0 55 7 8] + [ 17 3 0 1 2 0 0 0 975 31 14 1 4 13 17 4 0 1 2 1 3] + [ 63 3 1 0 10 2 1 0 35 965 2 1 0 16 4 4 0 0 1 3 8] + [ 0 2 8 11 0 1 3 1 13 1 954 2 1 29 4 3 1 0 8 3 8] + [ 1 1 0 0 1 7 0 0 0 0 0 947 40 8 0 4 0 18 0 6 2] + [ 0 0 0 5 2 2 0 0 3 3 1 47 945 3 1 10 2 34 1 4 5] + [ 1 0 0 0 1 4 0 0 16 16 2 7 3 1053 6 1 0 1 0 1 7] + [ 5 0 3 13 10 0 0 0 21 3 0 0 2 0 1032 0 0 5 4 0 3] + [ 0 0 1 2 4 0 0 0 0 1 0 10 8 0 0 1065 18 19 2 1 3] + [ 2 11 0 0 6 5 1 0 2 0 0 7 3 3 3 8 1095 1 2 3 9] + [ 0 0 0 4 0 0 2 0 0 0 0 2 15 0 6 5 4 999 1 0 0] + [ 2 5 4 18 0 0 1 11 4 0 5 3 4 0 23 0 1 0 976 0 11] + [ 0 3 3 0 0 3 13 8 0 1 1 17 8 3 1 2 8 2 7 1064 8] + [ 136 222 141 118 101 117 37 79 134 83 170 148 402 284 174 77 134 106 183 175 4884]] + +2023-10-02 20:58:33,367 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:58:33,367 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:58:33,373 - + +2023-10-02 20:58:33,373 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:58:34,518 - Epoch: [62][ 10/ 1236] Overall Loss 0.252146 Objective Loss 0.252146 LR 0.001000 Time 0.114443 +2023-10-02 20:58:34,728 - Epoch: [62][ 20/ 1236] Overall Loss 0.247471 Objective Loss 0.247471 LR 0.001000 Time 0.067664 +2023-10-02 20:58:34,937 - Epoch: [62][ 30/ 1236] Overall Loss 0.243777 Objective Loss 0.243777 LR 0.001000 Time 0.052085 +2023-10-02 20:58:35,145 - Epoch: [62][ 40/ 1236] Overall Loss 0.239659 Objective Loss 0.239659 LR 0.001000 Time 0.044241 +2023-10-02 20:58:35,355 - Epoch: [62][ 50/ 1236] Overall Loss 0.246447 Objective Loss 0.246447 LR 0.001000 Time 0.039582 +2023-10-02 20:58:35,563 - Epoch: [62][ 60/ 1236] Overall Loss 0.251240 Objective Loss 0.251240 LR 0.001000 Time 0.036448 +2023-10-02 20:58:35,773 - Epoch: [62][ 70/ 1236] Overall Loss 0.250040 Objective Loss 0.250040 LR 0.001000 Time 0.034232 +2023-10-02 20:58:35,980 - Epoch: [62][ 80/ 1236] Overall Loss 0.250291 Objective Loss 0.250291 LR 0.001000 Time 0.032539 +2023-10-02 20:58:36,190 - Epoch: [62][ 90/ 1236] Overall Loss 0.252776 Objective Loss 0.252776 LR 0.001000 Time 0.031248 +2023-10-02 20:58:36,397 - Epoch: [62][ 100/ 1236] Overall Loss 0.255057 Objective Loss 0.255057 LR 0.001000 Time 0.030191 +2023-10-02 20:58:36,611 - Epoch: [62][ 110/ 1236] Overall Loss 0.253562 Objective Loss 0.253562 LR 0.001000 Time 0.029388 +2023-10-02 20:58:36,825 - Epoch: [62][ 120/ 1236] Overall Loss 0.253994 Objective Loss 0.253994 LR 0.001000 Time 0.028709 +2023-10-02 20:58:37,045 - Epoch: [62][ 130/ 1236] Overall Loss 0.255439 Objective Loss 0.255439 LR 0.001000 Time 0.028189 +2023-10-02 20:58:37,259 - Epoch: [62][ 140/ 1236] Overall Loss 0.255136 Objective Loss 0.255136 LR 0.001000 Time 0.027695 +2023-10-02 20:58:37,479 - Epoch: [62][ 150/ 1236] Overall Loss 0.254808 Objective Loss 0.254808 LR 0.001000 Time 0.027312 +2023-10-02 20:58:37,694 - Epoch: [62][ 160/ 1236] Overall Loss 0.253354 Objective Loss 0.253354 LR 0.001000 Time 0.026937 +2023-10-02 20:58:37,909 - Epoch: [62][ 170/ 1236] Overall Loss 0.252902 Objective Loss 0.252902 LR 0.001000 Time 0.026609 +2023-10-02 20:58:38,120 - Epoch: [62][ 180/ 1236] Overall Loss 0.253322 Objective Loss 0.253322 LR 0.001000 Time 0.026295 +2023-10-02 20:58:38,331 - Epoch: [62][ 190/ 1236] Overall Loss 0.251408 Objective Loss 0.251408 LR 0.001000 Time 0.026013 +2023-10-02 20:58:38,551 - Epoch: [62][ 200/ 1236] Overall Loss 0.253077 Objective Loss 0.253077 LR 0.001000 Time 0.025808 +2023-10-02 20:58:38,765 - Epoch: [62][ 210/ 1236] Overall Loss 0.253600 Objective Loss 0.253600 LR 0.001000 Time 0.025589 +2023-10-02 20:58:38,970 - Epoch: [62][ 220/ 1236] Overall Loss 0.252090 Objective Loss 0.252090 LR 0.001000 Time 0.025359 +2023-10-02 20:58:39,178 - Epoch: [62][ 230/ 1236] Overall Loss 0.252752 Objective Loss 0.252752 LR 0.001000 Time 0.025157 +2023-10-02 20:58:39,383 - Epoch: [62][ 240/ 1236] Overall Loss 0.252666 Objective Loss 0.252666 LR 0.001000 Time 0.024963 +2023-10-02 20:58:39,590 - Epoch: [62][ 250/ 1236] Overall Loss 0.252072 Objective Loss 0.252072 LR 0.001000 Time 0.024791 +2023-10-02 20:58:39,795 - Epoch: [62][ 260/ 1236] Overall Loss 0.252461 Objective Loss 0.252461 LR 0.001000 Time 0.024625 +2023-10-02 20:58:40,002 - Epoch: [62][ 270/ 1236] Overall Loss 0.252782 Objective Loss 0.252782 LR 0.001000 Time 0.024480 +2023-10-02 20:58:40,207 - Epoch: [62][ 280/ 1236] Overall Loss 0.252270 Objective Loss 0.252270 LR 0.001000 Time 0.024335 +2023-10-02 20:58:40,414 - Epoch: [62][ 290/ 1236] Overall Loss 0.252050 Objective Loss 0.252050 LR 0.001000 Time 0.024210 +2023-10-02 20:58:40,619 - Epoch: [62][ 300/ 1236] Overall Loss 0.251768 Objective Loss 0.251768 LR 0.001000 Time 0.024085 +2023-10-02 20:58:40,827 - Epoch: [62][ 310/ 1236] Overall Loss 0.250329 Objective Loss 0.250329 LR 0.001000 Time 0.023976 +2023-10-02 20:58:41,032 - Epoch: [62][ 320/ 1236] Overall Loss 0.249677 Objective Loss 0.249677 LR 0.001000 Time 0.023868 +2023-10-02 20:58:41,239 - Epoch: [62][ 330/ 1236] Overall Loss 0.250242 Objective Loss 0.250242 LR 0.001000 Time 0.023770 +2023-10-02 20:58:41,444 - Epoch: [62][ 340/ 1236] Overall Loss 0.250226 Objective Loss 0.250226 LR 0.001000 Time 0.023673 +2023-10-02 20:58:41,653 - Epoch: [62][ 350/ 1236] Overall Loss 0.249666 Objective Loss 0.249666 LR 0.001000 Time 0.023592 +2023-10-02 20:58:41,860 - Epoch: [62][ 360/ 1236] Overall Loss 0.249448 Objective Loss 0.249448 LR 0.001000 Time 0.023511 +2023-10-02 20:58:42,067 - Epoch: [62][ 370/ 1236] Overall Loss 0.249644 Objective Loss 0.249644 LR 0.001000 Time 0.023434 +2023-10-02 20:58:42,272 - Epoch: [62][ 380/ 1236] Overall Loss 0.249963 Objective Loss 0.249963 LR 0.001000 Time 0.023357 +2023-10-02 20:58:42,481 - Epoch: [62][ 390/ 1236] Overall Loss 0.250073 Objective Loss 0.250073 LR 0.001000 Time 0.023292 +2023-10-02 20:58:42,687 - Epoch: [62][ 400/ 1236] Overall Loss 0.250016 Objective Loss 0.250016 LR 0.001000 Time 0.023225 +2023-10-02 20:58:42,896 - Epoch: [62][ 410/ 1236] Overall Loss 0.249743 Objective Loss 0.249743 LR 0.001000 Time 0.023167 +2023-10-02 20:58:43,102 - Epoch: [62][ 420/ 1236] Overall Loss 0.249781 Objective Loss 0.249781 LR 0.001000 Time 0.023106 +2023-10-02 20:58:43,309 - Epoch: [62][ 430/ 1236] Overall Loss 0.249436 Objective Loss 0.249436 LR 0.001000 Time 0.023048 +2023-10-02 20:58:43,514 - Epoch: [62][ 440/ 1236] Overall Loss 0.249310 Objective Loss 0.249310 LR 0.001000 Time 0.022990 +2023-10-02 20:58:43,721 - Epoch: [62][ 450/ 1236] Overall Loss 0.248747 Objective Loss 0.248747 LR 0.001000 Time 0.022938 +2023-10-02 20:58:43,925 - Epoch: [62][ 460/ 1236] Overall Loss 0.248747 Objective Loss 0.248747 LR 0.001000 Time 0.022884 +2023-10-02 20:58:44,132 - Epoch: [62][ 470/ 1236] Overall Loss 0.248584 Objective Loss 0.248584 LR 0.001000 Time 0.022836 +2023-10-02 20:58:44,337 - Epoch: [62][ 480/ 1236] Overall Loss 0.248717 Objective Loss 0.248717 LR 0.001000 Time 0.022787 +2023-10-02 20:58:44,544 - Epoch: [62][ 490/ 1236] Overall Loss 0.248907 Objective Loss 0.248907 LR 0.001000 Time 0.022743 +2023-10-02 20:58:44,748 - Epoch: [62][ 500/ 1236] Overall Loss 0.249046 Objective Loss 0.249046 LR 0.001000 Time 0.022697 +2023-10-02 20:58:44,955 - Epoch: [62][ 510/ 1236] Overall Loss 0.249227 Objective Loss 0.249227 LR 0.001000 Time 0.022656 +2023-10-02 20:58:45,160 - Epoch: [62][ 520/ 1236] Overall Loss 0.249508 Objective Loss 0.249508 LR 0.001000 Time 0.022614 +2023-10-02 20:58:45,367 - Epoch: [62][ 530/ 1236] Overall Loss 0.249362 Objective Loss 0.249362 LR 0.001000 Time 0.022577 +2023-10-02 20:58:45,572 - Epoch: [62][ 540/ 1236] Overall Loss 0.249359 Objective Loss 0.249359 LR 0.001000 Time 0.022539 +2023-10-02 20:58:45,779 - Epoch: [62][ 550/ 1236] Overall Loss 0.248754 Objective Loss 0.248754 LR 0.001000 Time 0.022505 +2023-10-02 20:58:45,985 - Epoch: [62][ 560/ 1236] Overall Loss 0.249342 Objective Loss 0.249342 LR 0.001000 Time 0.022469 +2023-10-02 20:58:46,191 - Epoch: [62][ 570/ 1236] Overall Loss 0.249099 Objective Loss 0.249099 LR 0.001000 Time 0.022437 +2023-10-02 20:58:46,396 - Epoch: [62][ 580/ 1236] Overall Loss 0.248821 Objective Loss 0.248821 LR 0.001000 Time 0.022403 +2023-10-02 20:58:46,603 - Epoch: [62][ 590/ 1236] Overall Loss 0.248912 Objective Loss 0.248912 LR 0.001000 Time 0.022373 +2023-10-02 20:58:46,808 - Epoch: [62][ 600/ 1236] Overall Loss 0.248965 Objective Loss 0.248965 LR 0.001000 Time 0.022341 +2023-10-02 20:58:47,015 - Epoch: [62][ 610/ 1236] Overall Loss 0.249095 Objective Loss 0.249095 LR 0.001000 Time 0.022313 +2023-10-02 20:58:47,219 - Epoch: [62][ 620/ 1236] Overall Loss 0.249099 Objective Loss 0.249099 LR 0.001000 Time 0.022283 +2023-10-02 20:58:47,426 - Epoch: [62][ 630/ 1236] Overall Loss 0.249329 Objective Loss 0.249329 LR 0.001000 Time 0.022257 +2023-10-02 20:58:47,631 - Epoch: [62][ 640/ 1236] Overall Loss 0.249066 Objective Loss 0.249066 LR 0.001000 Time 0.022228 +2023-10-02 20:58:47,837 - Epoch: [62][ 650/ 1236] Overall Loss 0.248879 Objective Loss 0.248879 LR 0.001000 Time 0.022204 +2023-10-02 20:58:48,042 - Epoch: [62][ 660/ 1236] Overall Loss 0.248627 Objective Loss 0.248627 LR 0.001000 Time 0.022177 +2023-10-02 20:58:48,249 - Epoch: [62][ 670/ 1236] Overall Loss 0.248637 Objective Loss 0.248637 LR 0.001000 Time 0.022154 +2023-10-02 20:58:48,454 - Epoch: [62][ 680/ 1236] Overall Loss 0.249458 Objective Loss 0.249458 LR 0.001000 Time 0.022129 +2023-10-02 20:58:48,661 - Epoch: [62][ 690/ 1236] Overall Loss 0.250276 Objective Loss 0.250276 LR 0.001000 Time 0.022108 +2023-10-02 20:58:48,866 - Epoch: [62][ 700/ 1236] Overall Loss 0.250499 Objective Loss 0.250499 LR 0.001000 Time 0.022084 +2023-10-02 20:58:49,072 - Epoch: [62][ 710/ 1236] Overall Loss 0.250607 Objective Loss 0.250607 LR 0.001000 Time 0.022064 +2023-10-02 20:58:49,277 - Epoch: [62][ 720/ 1236] Overall Loss 0.250784 Objective Loss 0.250784 LR 0.001000 Time 0.022041 +2023-10-02 20:58:49,484 - Epoch: [62][ 730/ 1236] Overall Loss 0.251313 Objective Loss 0.251313 LR 0.001000 Time 0.022022 +2023-10-02 20:58:49,688 - Epoch: [62][ 740/ 1236] Overall Loss 0.251500 Objective Loss 0.251500 LR 0.001000 Time 0.022001 +2023-10-02 20:58:49,895 - Epoch: [62][ 750/ 1236] Overall Loss 0.251615 Objective Loss 0.251615 LR 0.001000 Time 0.021983 +2023-10-02 20:58:50,100 - Epoch: [62][ 760/ 1236] Overall Loss 0.251604 Objective Loss 0.251604 LR 0.001000 Time 0.021963 +2023-10-02 20:58:50,307 - Epoch: [62][ 770/ 1236] Overall Loss 0.251741 Objective Loss 0.251741 LR 0.001000 Time 0.021946 +2023-10-02 20:58:50,511 - Epoch: [62][ 780/ 1236] Overall Loss 0.251902 Objective Loss 0.251902 LR 0.001000 Time 0.021926 +2023-10-02 20:58:50,718 - Epoch: [62][ 790/ 1236] Overall Loss 0.252100 Objective Loss 0.252100 LR 0.001000 Time 0.021910 +2023-10-02 20:58:50,923 - Epoch: [62][ 800/ 1236] Overall Loss 0.252301 Objective Loss 0.252301 LR 0.001000 Time 0.021891 +2023-10-02 20:58:51,130 - Epoch: [62][ 810/ 1236] Overall Loss 0.252866 Objective Loss 0.252866 LR 0.001000 Time 0.021876 +2023-10-02 20:58:51,335 - Epoch: [62][ 820/ 1236] Overall Loss 0.252935 Objective Loss 0.252935 LR 0.001000 Time 0.021859 +2023-10-02 20:58:51,541 - Epoch: [62][ 830/ 1236] Overall Loss 0.252788 Objective Loss 0.252788 LR 0.001000 Time 0.021844 +2023-10-02 20:58:51,746 - Epoch: [62][ 840/ 1236] Overall Loss 0.252668 Objective Loss 0.252668 LR 0.001000 Time 0.021828 +2023-10-02 20:58:51,953 - Epoch: [62][ 850/ 1236] Overall Loss 0.252692 Objective Loss 0.252692 LR 0.001000 Time 0.021814 +2023-10-02 20:58:52,159 - Epoch: [62][ 860/ 1236] Overall Loss 0.252716 Objective Loss 0.252716 LR 0.001000 Time 0.021799 +2023-10-02 20:58:52,368 - Epoch: [62][ 870/ 1236] Overall Loss 0.252706 Objective Loss 0.252706 LR 0.001000 Time 0.021788 +2023-10-02 20:58:52,574 - Epoch: [62][ 880/ 1236] Overall Loss 0.252497 Objective Loss 0.252497 LR 0.001000 Time 0.021775 +2023-10-02 20:58:52,783 - Epoch: [62][ 890/ 1236] Overall Loss 0.252612 Objective Loss 0.252612 LR 0.001000 Time 0.021764 +2023-10-02 20:58:52,989 - Epoch: [62][ 900/ 1236] Overall Loss 0.252647 Objective Loss 0.252647 LR 0.001000 Time 0.021752 +2023-10-02 20:58:53,198 - Epoch: [62][ 910/ 1236] Overall Loss 0.252708 Objective Loss 0.252708 LR 0.001000 Time 0.021742 +2023-10-02 20:58:53,405 - Epoch: [62][ 920/ 1236] Overall Loss 0.252770 Objective Loss 0.252770 LR 0.001000 Time 0.021729 +2023-10-02 20:58:53,613 - Epoch: [62][ 930/ 1236] Overall Loss 0.252667 Objective Loss 0.252667 LR 0.001000 Time 0.021720 +2023-10-02 20:58:53,820 - Epoch: [62][ 940/ 1236] Overall Loss 0.252730 Objective Loss 0.252730 LR 0.001000 Time 0.021708 +2023-10-02 20:58:54,028 - Epoch: [62][ 950/ 1236] Overall Loss 0.252817 Objective Loss 0.252817 LR 0.001000 Time 0.021699 +2023-10-02 20:58:54,234 - Epoch: [62][ 960/ 1236] Overall Loss 0.252972 Objective Loss 0.252972 LR 0.001000 Time 0.021687 +2023-10-02 20:58:54,443 - Epoch: [62][ 970/ 1236] Overall Loss 0.252841 Objective Loss 0.252841 LR 0.001000 Time 0.021678 +2023-10-02 20:58:54,650 - Epoch: [62][ 980/ 1236] Overall Loss 0.252967 Objective Loss 0.252967 LR 0.001000 Time 0.021667 +2023-10-02 20:58:54,858 - Epoch: [62][ 990/ 1236] Overall Loss 0.252914 Objective Loss 0.252914 LR 0.001000 Time 0.021659 +2023-10-02 20:58:55,065 - Epoch: [62][ 1000/ 1236] Overall Loss 0.253014 Objective Loss 0.253014 LR 0.001000 Time 0.021649 +2023-10-02 20:58:55,273 - Epoch: [62][ 1010/ 1236] Overall Loss 0.252827 Objective Loss 0.252827 LR 0.001000 Time 0.021640 +2023-10-02 20:58:55,480 - Epoch: [62][ 1020/ 1236] Overall Loss 0.252679 Objective Loss 0.252679 LR 0.001000 Time 0.021630 +2023-10-02 20:58:55,688 - Epoch: [62][ 1030/ 1236] Overall Loss 0.252893 Objective Loss 0.252893 LR 0.001000 Time 0.021622 +2023-10-02 20:58:55,895 - Epoch: [62][ 1040/ 1236] Overall Loss 0.252837 Objective Loss 0.252837 LR 0.001000 Time 0.021613 +2023-10-02 20:58:56,103 - Epoch: [62][ 1050/ 1236] Overall Loss 0.252905 Objective Loss 0.252905 LR 0.001000 Time 0.021605 +2023-10-02 20:58:56,310 - Epoch: [62][ 1060/ 1236] Overall Loss 0.252876 Objective Loss 0.252876 LR 0.001000 Time 0.021596 +2023-10-02 20:58:56,518 - Epoch: [62][ 1070/ 1236] Overall Loss 0.252948 Objective Loss 0.252948 LR 0.001000 Time 0.021589 +2023-10-02 20:58:56,725 - Epoch: [62][ 1080/ 1236] Overall Loss 0.253149 Objective Loss 0.253149 LR 0.001000 Time 0.021580 +2023-10-02 20:58:56,933 - Epoch: [62][ 1090/ 1236] Overall Loss 0.252939 Objective Loss 0.252939 LR 0.001000 Time 0.021572 +2023-10-02 20:58:57,139 - Epoch: [62][ 1100/ 1236] Overall Loss 0.252973 Objective Loss 0.252973 LR 0.001000 Time 0.021564 +2023-10-02 20:58:57,348 - Epoch: [62][ 1110/ 1236] Overall Loss 0.253178 Objective Loss 0.253178 LR 0.001000 Time 0.021557 +2023-10-02 20:58:57,554 - Epoch: [62][ 1120/ 1236] Overall Loss 0.253334 Objective Loss 0.253334 LR 0.001000 Time 0.021548 +2023-10-02 20:58:57,762 - Epoch: [62][ 1130/ 1236] Overall Loss 0.253620 Objective Loss 0.253620 LR 0.001000 Time 0.021541 +2023-10-02 20:58:57,967 - Epoch: [62][ 1140/ 1236] Overall Loss 0.253715 Objective Loss 0.253715 LR 0.001000 Time 0.021532 +2023-10-02 20:58:58,176 - Epoch: [62][ 1150/ 1236] Overall Loss 0.253782 Objective Loss 0.253782 LR 0.001000 Time 0.021526 +2023-10-02 20:58:58,382 - Epoch: [62][ 1160/ 1236] Overall Loss 0.254030 Objective Loss 0.254030 LR 0.001000 Time 0.021518 +2023-10-02 20:58:58,590 - Epoch: [62][ 1170/ 1236] Overall Loss 0.254091 Objective Loss 0.254091 LR 0.001000 Time 0.021512 +2023-10-02 20:58:58,796 - Epoch: [62][ 1180/ 1236] Overall Loss 0.254456 Objective Loss 0.254456 LR 0.001000 Time 0.021504 +2023-10-02 20:58:59,005 - Epoch: [62][ 1190/ 1236] Overall Loss 0.254709 Objective Loss 0.254709 LR 0.001000 Time 0.021498 +2023-10-02 20:58:59,211 - Epoch: [62][ 1200/ 1236] Overall Loss 0.254794 Objective Loss 0.254794 LR 0.001000 Time 0.021490 +2023-10-02 20:58:59,421 - Epoch: [62][ 1210/ 1236] Overall Loss 0.254681 Objective Loss 0.254681 LR 0.001000 Time 0.021486 +2023-10-02 20:58:59,629 - Epoch: [62][ 1220/ 1236] Overall Loss 0.254670 Objective Loss 0.254670 LR 0.001000 Time 0.021480 +2023-10-02 20:58:59,889 - Epoch: [62][ 1230/ 1236] Overall Loss 0.254688 Objective Loss 0.254688 LR 0.001000 Time 0.021517 +2023-10-02 20:59:00,011 - Epoch: [62][ 1236/ 1236] Overall Loss 0.254767 Objective Loss 0.254767 Top1 84.114053 Top5 96.945010 LR 0.001000 Time 0.021511 +2023-10-02 20:59:00,143 - --- validate (epoch=62)----------- +2023-10-02 20:59:00,143 - 29943 samples (256 per mini-batch) +2023-10-02 20:59:00,607 - Epoch: [62][ 10/ 117] Loss 0.354664 Top1 83.281250 Top5 98.007812 +2023-10-02 20:59:00,759 - Epoch: [62][ 20/ 117] Loss 0.321311 Top1 84.472656 Top5 98.281250 +2023-10-02 20:59:00,910 - Epoch: [62][ 30/ 117] Loss 0.323595 Top1 84.453125 Top5 98.385417 +2023-10-02 20:59:01,060 - Epoch: [62][ 40/ 117] Loss 0.322518 Top1 84.248047 Top5 98.486328 +2023-10-02 20:59:01,210 - Epoch: [62][ 50/ 117] Loss 0.328139 Top1 84.101562 Top5 98.429688 +2023-10-02 20:59:01,360 - Epoch: [62][ 60/ 117] Loss 0.323962 Top1 84.069010 Top5 98.457031 +2023-10-02 20:59:01,510 - Epoch: [62][ 70/ 117] Loss 0.317306 Top1 84.241071 Top5 98.448661 +2023-10-02 20:59:01,659 - Epoch: [62][ 80/ 117] Loss 0.320008 Top1 84.174805 Top5 98.422852 +2023-10-02 20:59:01,809 - Epoch: [62][ 90/ 117] Loss 0.317405 Top1 84.179688 Top5 98.411458 +2023-10-02 20:59:01,959 - Epoch: [62][ 100/ 117] Loss 0.319024 Top1 84.191406 Top5 98.386719 +2023-10-02 20:59:02,115 - Epoch: [62][ 110/ 117] Loss 0.325520 Top1 84.009233 Top5 98.323864 +2023-10-02 20:59:02,204 - Epoch: [62][ 117/ 117] Loss 0.326209 Top1 83.969542 Top5 98.323481 +2023-10-02 20:59:02,321 - ==> Top1: 83.970 Top5: 98.323 Loss: 0.326 + +2023-10-02 20:59:02,322 - ==> Confusion: +[[ 907 5 5 0 11 3 0 0 9 68 2 2 0 2 6 1 3 1 4 0 21] + [ 0 1042 1 0 7 30 2 19 1 0 5 2 0 0 0 2 0 0 13 4 3] + [ 1 1 962 15 0 0 13 14 0 0 4 2 9 2 2 5 1 1 13 4 7] + [ 1 5 7 990 0 2 1 3 2 0 8 0 4 3 25 0 1 0 22 1 14] + [ 20 7 2 0 962 11 1 0 1 9 0 2 0 8 8 2 8 0 0 3 6] + [ 1 45 0 1 3 988 1 19 0 4 7 8 0 11 4 0 2 0 4 8 10] + [ 0 6 29 0 0 1 1112 7 0 0 3 2 0 1 1 6 0 0 2 9 12] + [ 1 24 16 1 2 26 4 1052 2 1 7 6 3 7 1 0 1 0 49 12 3] + [ 14 3 0 0 3 4 0 1 972 39 7 1 1 12 18 2 2 0 5 2 3] + [ 75 1 3 0 11 4 0 3 45 927 2 1 0 30 5 0 1 0 0 2 9] + [ 0 0 9 11 0 1 4 5 18 0 966 5 0 12 5 0 1 2 6 1 7] + [ 0 1 1 0 0 9 0 3 0 0 1 965 18 8 0 1 1 13 0 10 4] + [ 0 2 1 5 1 2 3 2 3 1 3 54 948 2 0 7 2 10 5 2 15] + [ 1 0 1 0 2 3 0 1 13 6 12 7 0 1048 6 0 2 1 1 3 12] + [ 6 1 4 14 9 0 0 0 27 3 4 1 1 1 1006 0 1 4 15 0 4] + [ 0 0 1 2 6 2 0 0 0 0 1 8 7 1 0 1065 18 8 2 5 8] + [ 1 23 1 0 9 8 1 0 0 0 0 5 3 2 4 9 1068 1 1 8 17] + [ 0 0 1 5 0 0 1 0 2 1 0 10 27 2 5 7 1 968 4 0 4] + [ 1 4 2 8 0 0 1 14 3 0 4 0 0 0 10 0 0 1 1008 1 11] + [ 0 3 5 1 0 2 8 12 0 1 0 13 7 2 0 0 3 0 3 1085 7] + [ 105 191 120 98 77 200 26 73 125 88 210 154 310 283 141 49 82 40 224 207 5102]] + +2023-10-02 20:59:02,323 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:59:02,323 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:59:02,329 - + +2023-10-02 20:59:02,330 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:59:03,345 - Epoch: [63][ 10/ 1236] Overall Loss 0.259317 Objective Loss 0.259317 LR 0.001000 Time 0.101476 +2023-10-02 20:59:03,552 - Epoch: [63][ 20/ 1236] Overall Loss 0.254901 Objective Loss 0.254901 LR 0.001000 Time 0.061062 +2023-10-02 20:59:03,759 - Epoch: [63][ 30/ 1236] Overall Loss 0.253556 Objective Loss 0.253556 LR 0.001000 Time 0.047596 +2023-10-02 20:59:03,966 - Epoch: [63][ 40/ 1236] Overall Loss 0.249422 Objective Loss 0.249422 LR 0.001000 Time 0.040857 +2023-10-02 20:59:04,171 - Epoch: [63][ 50/ 1236] Overall Loss 0.245180 Objective Loss 0.245180 LR 0.001000 Time 0.036792 +2023-10-02 20:59:04,377 - Epoch: [63][ 60/ 1236] Overall Loss 0.243364 Objective Loss 0.243364 LR 0.001000 Time 0.034093 +2023-10-02 20:59:04,583 - Epoch: [63][ 70/ 1236] Overall Loss 0.243928 Objective Loss 0.243928 LR 0.001000 Time 0.032151 +2023-10-02 20:59:04,790 - Epoch: [63][ 80/ 1236] Overall Loss 0.242139 Objective Loss 0.242139 LR 0.001000 Time 0.030723 +2023-10-02 20:59:04,994 - Epoch: [63][ 90/ 1236] Overall Loss 0.245079 Objective Loss 0.245079 LR 0.001000 Time 0.029566 +2023-10-02 20:59:05,201 - Epoch: [63][ 100/ 1236] Overall Loss 0.247655 Objective Loss 0.247655 LR 0.001000 Time 0.028683 +2023-10-02 20:59:05,405 - Epoch: [63][ 110/ 1236] Overall Loss 0.249052 Objective Loss 0.249052 LR 0.001000 Time 0.027928 +2023-10-02 20:59:05,613 - Epoch: [63][ 120/ 1236] Overall Loss 0.249424 Objective Loss 0.249424 LR 0.001000 Time 0.027329 +2023-10-02 20:59:05,816 - Epoch: [63][ 130/ 1236] Overall Loss 0.249534 Objective Loss 0.249534 LR 0.001000 Time 0.026784 +2023-10-02 20:59:06,024 - Epoch: [63][ 140/ 1236] Overall Loss 0.249507 Objective Loss 0.249507 LR 0.001000 Time 0.026354 +2023-10-02 20:59:06,227 - Epoch: [63][ 150/ 1236] Overall Loss 0.251201 Objective Loss 0.251201 LR 0.001000 Time 0.025953 +2023-10-02 20:59:06,435 - Epoch: [63][ 160/ 1236] Overall Loss 0.251247 Objective Loss 0.251247 LR 0.001000 Time 0.025628 +2023-10-02 20:59:06,639 - Epoch: [63][ 170/ 1236] Overall Loss 0.250947 Objective Loss 0.250947 LR 0.001000 Time 0.025317 +2023-10-02 20:59:06,846 - Epoch: [63][ 180/ 1236] Overall Loss 0.251423 Objective Loss 0.251423 LR 0.001000 Time 0.025061 +2023-10-02 20:59:07,050 - Epoch: [63][ 190/ 1236] Overall Loss 0.250505 Objective Loss 0.250505 LR 0.001000 Time 0.024810 +2023-10-02 20:59:07,257 - Epoch: [63][ 200/ 1236] Overall Loss 0.249544 Objective Loss 0.249544 LR 0.001000 Time 0.024608 +2023-10-02 20:59:07,461 - Epoch: [63][ 210/ 1236] Overall Loss 0.248341 Objective Loss 0.248341 LR 0.001000 Time 0.024403 +2023-10-02 20:59:07,669 - Epoch: [63][ 220/ 1236] Overall Loss 0.248553 Objective Loss 0.248553 LR 0.001000 Time 0.024238 +2023-10-02 20:59:07,873 - Epoch: [63][ 230/ 1236] Overall Loss 0.248335 Objective Loss 0.248335 LR 0.001000 Time 0.024070 +2023-10-02 20:59:08,080 - Epoch: [63][ 240/ 1236] Overall Loss 0.249688 Objective Loss 0.249688 LR 0.001000 Time 0.023931 +2023-10-02 20:59:08,289 - Epoch: [63][ 250/ 1236] Overall Loss 0.250138 Objective Loss 0.250138 LR 0.001000 Time 0.023808 +2023-10-02 20:59:08,495 - Epoch: [63][ 260/ 1236] Overall Loss 0.248937 Objective Loss 0.248937 LR 0.001000 Time 0.023683 +2023-10-02 20:59:08,704 - Epoch: [63][ 270/ 1236] Overall Loss 0.248735 Objective Loss 0.248735 LR 0.001000 Time 0.023579 +2023-10-02 20:59:08,910 - Epoch: [63][ 280/ 1236] Overall Loss 0.248889 Objective Loss 0.248889 LR 0.001000 Time 0.023471 +2023-10-02 20:59:09,119 - Epoch: [63][ 290/ 1236] Overall Loss 0.249601 Objective Loss 0.249601 LR 0.001000 Time 0.023381 +2023-10-02 20:59:09,325 - Epoch: [63][ 300/ 1236] Overall Loss 0.249989 Objective Loss 0.249989 LR 0.001000 Time 0.023288 +2023-10-02 20:59:09,534 - Epoch: [63][ 310/ 1236] Overall Loss 0.250199 Objective Loss 0.250199 LR 0.001000 Time 0.023210 +2023-10-02 20:59:09,740 - Epoch: [63][ 320/ 1236] Overall Loss 0.250554 Objective Loss 0.250554 LR 0.001000 Time 0.023128 +2023-10-02 20:59:09,950 - Epoch: [63][ 330/ 1236] Overall Loss 0.250442 Objective Loss 0.250442 LR 0.001000 Time 0.023061 +2023-10-02 20:59:10,156 - Epoch: [63][ 340/ 1236] Overall Loss 0.250565 Objective Loss 0.250565 LR 0.001000 Time 0.022989 +2023-10-02 20:59:10,365 - Epoch: [63][ 350/ 1236] Overall Loss 0.250813 Objective Loss 0.250813 LR 0.001000 Time 0.022929 +2023-10-02 20:59:10,571 - Epoch: [63][ 360/ 1236] Overall Loss 0.250884 Objective Loss 0.250884 LR 0.001000 Time 0.022864 +2023-10-02 20:59:10,781 - Epoch: [63][ 370/ 1236] Overall Loss 0.250697 Objective Loss 0.250697 LR 0.001000 Time 0.022813 +2023-10-02 20:59:10,987 - Epoch: [63][ 380/ 1236] Overall Loss 0.250644 Objective Loss 0.250644 LR 0.001000 Time 0.022753 +2023-10-02 20:59:11,197 - Epoch: [63][ 390/ 1236] Overall Loss 0.250287 Objective Loss 0.250287 LR 0.001000 Time 0.022707 +2023-10-02 20:59:11,403 - Epoch: [63][ 400/ 1236] Overall Loss 0.250555 Objective Loss 0.250555 LR 0.001000 Time 0.022654 +2023-10-02 20:59:11,612 - Epoch: [63][ 410/ 1236] Overall Loss 0.250634 Objective Loss 0.250634 LR 0.001000 Time 0.022610 +2023-10-02 20:59:11,819 - Epoch: [63][ 420/ 1236] Overall Loss 0.250696 Objective Loss 0.250696 LR 0.001000 Time 0.022564 +2023-10-02 20:59:12,028 - Epoch: [63][ 430/ 1236] Overall Loss 0.250470 Objective Loss 0.250470 LR 0.001000 Time 0.022525 +2023-10-02 20:59:12,236 - Epoch: [63][ 440/ 1236] Overall Loss 0.250269 Objective Loss 0.250269 LR 0.001000 Time 0.022483 +2023-10-02 20:59:12,446 - Epoch: [63][ 450/ 1236] Overall Loss 0.250134 Objective Loss 0.250134 LR 0.001000 Time 0.022450 +2023-10-02 20:59:12,652 - Epoch: [63][ 460/ 1236] Overall Loss 0.249540 Objective Loss 0.249540 LR 0.001000 Time 0.022410 +2023-10-02 20:59:12,861 - Epoch: [63][ 470/ 1236] Overall Loss 0.249479 Objective Loss 0.249479 LR 0.001000 Time 0.022377 +2023-10-02 20:59:13,069 - Epoch: [63][ 480/ 1236] Overall Loss 0.249532 Objective Loss 0.249532 LR 0.001000 Time 0.022340 +2023-10-02 20:59:13,278 - Epoch: [63][ 490/ 1236] Overall Loss 0.249561 Objective Loss 0.249561 LR 0.001000 Time 0.022311 +2023-10-02 20:59:13,485 - Epoch: [63][ 500/ 1236] Overall Loss 0.249123 Objective Loss 0.249123 LR 0.001000 Time 0.022276 +2023-10-02 20:59:13,694 - Epoch: [63][ 510/ 1236] Overall Loss 0.249535 Objective Loss 0.249535 LR 0.001000 Time 0.022249 +2023-10-02 20:59:13,902 - Epoch: [63][ 520/ 1236] Overall Loss 0.249222 Objective Loss 0.249222 LR 0.001000 Time 0.022217 +2023-10-02 20:59:14,112 - Epoch: [63][ 530/ 1236] Overall Loss 0.249097 Objective Loss 0.249097 LR 0.001000 Time 0.022194 +2023-10-02 20:59:14,318 - Epoch: [63][ 540/ 1236] Overall Loss 0.249708 Objective Loss 0.249708 LR 0.001000 Time 0.022164 +2023-10-02 20:59:14,528 - Epoch: [63][ 550/ 1236] Overall Loss 0.249410 Objective Loss 0.249410 LR 0.001000 Time 0.022142 +2023-10-02 20:59:14,734 - Epoch: [63][ 560/ 1236] Overall Loss 0.249236 Objective Loss 0.249236 LR 0.001000 Time 0.022115 +2023-10-02 20:59:14,943 - Epoch: [63][ 570/ 1236] Overall Loss 0.249182 Objective Loss 0.249182 LR 0.001000 Time 0.022093 +2023-10-02 20:59:15,151 - Epoch: [63][ 580/ 1236] Overall Loss 0.249351 Objective Loss 0.249351 LR 0.001000 Time 0.022067 +2023-10-02 20:59:15,360 - Epoch: [63][ 590/ 1236] Overall Loss 0.249606 Objective Loss 0.249606 LR 0.001000 Time 0.022047 +2023-10-02 20:59:15,567 - Epoch: [63][ 600/ 1236] Overall Loss 0.249605 Objective Loss 0.249605 LR 0.001000 Time 0.022024 +2023-10-02 20:59:15,776 - Epoch: [63][ 610/ 1236] Overall Loss 0.249340 Objective Loss 0.249340 LR 0.001000 Time 0.022006 +2023-10-02 20:59:15,984 - Epoch: [63][ 620/ 1236] Overall Loss 0.249415 Objective Loss 0.249415 LR 0.001000 Time 0.021983 +2023-10-02 20:59:16,193 - Epoch: [63][ 630/ 1236] Overall Loss 0.249426 Objective Loss 0.249426 LR 0.001000 Time 0.021966 +2023-10-02 20:59:16,401 - Epoch: [63][ 640/ 1236] Overall Loss 0.249413 Objective Loss 0.249413 LR 0.001000 Time 0.021946 +2023-10-02 20:59:16,610 - Epoch: [63][ 650/ 1236] Overall Loss 0.249170 Objective Loss 0.249170 LR 0.001000 Time 0.021930 +2023-10-02 20:59:16,817 - Epoch: [63][ 660/ 1236] Overall Loss 0.249509 Objective Loss 0.249509 LR 0.001000 Time 0.021912 +2023-10-02 20:59:17,027 - Epoch: [63][ 670/ 1236] Overall Loss 0.249217 Objective Loss 0.249217 LR 0.001000 Time 0.021897 +2023-10-02 20:59:17,235 - Epoch: [63][ 680/ 1236] Overall Loss 0.249596 Objective Loss 0.249596 LR 0.001000 Time 0.021879 +2023-10-02 20:59:17,446 - Epoch: [63][ 690/ 1236] Overall Loss 0.249175 Objective Loss 0.249175 LR 0.001000 Time 0.021866 +2023-10-02 20:59:17,652 - Epoch: [63][ 700/ 1236] Overall Loss 0.249379 Objective Loss 0.249379 LR 0.001000 Time 0.021848 +2023-10-02 20:59:17,862 - Epoch: [63][ 710/ 1236] Overall Loss 0.249422 Objective Loss 0.249422 LR 0.001000 Time 0.021836 +2023-10-02 20:59:18,069 - Epoch: [63][ 720/ 1236] Overall Loss 0.249502 Objective Loss 0.249502 LR 0.001000 Time 0.021819 +2023-10-02 20:59:18,278 - Epoch: [63][ 730/ 1236] Overall Loss 0.249420 Objective Loss 0.249420 LR 0.001000 Time 0.021806 +2023-10-02 20:59:18,485 - Epoch: [63][ 740/ 1236] Overall Loss 0.249599 Objective Loss 0.249599 LR 0.001000 Time 0.021790 +2023-10-02 20:59:18,694 - Epoch: [63][ 750/ 1236] Overall Loss 0.250066 Objective Loss 0.250066 LR 0.001000 Time 0.021778 +2023-10-02 20:59:18,902 - Epoch: [63][ 760/ 1236] Overall Loss 0.250531 Objective Loss 0.250531 LR 0.001000 Time 0.021764 +2023-10-02 20:59:19,111 - Epoch: [63][ 770/ 1236] Overall Loss 0.250867 Objective Loss 0.250867 LR 0.001000 Time 0.021752 +2023-10-02 20:59:19,319 - Epoch: [63][ 780/ 1236] Overall Loss 0.251008 Objective Loss 0.251008 LR 0.001000 Time 0.021740 +2023-10-02 20:59:19,528 - Epoch: [63][ 790/ 1236] Overall Loss 0.251755 Objective Loss 0.251755 LR 0.001000 Time 0.021729 +2023-10-02 20:59:19,736 - Epoch: [63][ 800/ 1236] Overall Loss 0.252091 Objective Loss 0.252091 LR 0.001000 Time 0.021718 +2023-10-02 20:59:19,946 - Epoch: [63][ 810/ 1236] Overall Loss 0.252167 Objective Loss 0.252167 LR 0.001000 Time 0.021708 +2023-10-02 20:59:20,154 - Epoch: [63][ 820/ 1236] Overall Loss 0.252130 Objective Loss 0.252130 LR 0.001000 Time 0.021695 +2023-10-02 20:59:20,364 - Epoch: [63][ 830/ 1236] Overall Loss 0.252247 Objective Loss 0.252247 LR 0.001000 Time 0.021686 +2023-10-02 20:59:20,572 - Epoch: [63][ 840/ 1236] Overall Loss 0.252467 Objective Loss 0.252467 LR 0.001000 Time 0.021675 +2023-10-02 20:59:20,782 - Epoch: [63][ 850/ 1236] Overall Loss 0.252660 Objective Loss 0.252660 LR 0.001000 Time 0.021667 +2023-10-02 20:59:20,990 - Epoch: [63][ 860/ 1236] Overall Loss 0.252903 Objective Loss 0.252903 LR 0.001000 Time 0.021655 +2023-10-02 20:59:21,200 - Epoch: [63][ 870/ 1236] Overall Loss 0.253117 Objective Loss 0.253117 LR 0.001000 Time 0.021646 +2023-10-02 20:59:21,408 - Epoch: [63][ 880/ 1236] Overall Loss 0.253266 Objective Loss 0.253266 LR 0.001000 Time 0.021635 +2023-10-02 20:59:21,621 - Epoch: [63][ 890/ 1236] Overall Loss 0.253401 Objective Loss 0.253401 LR 0.001000 Time 0.021631 +2023-10-02 20:59:21,828 - Epoch: [63][ 900/ 1236] Overall Loss 0.253061 Objective Loss 0.253061 LR 0.001000 Time 0.021620 +2023-10-02 20:59:22,039 - Epoch: [63][ 910/ 1236] Overall Loss 0.253238 Objective Loss 0.253238 LR 0.001000 Time 0.021614 +2023-10-02 20:59:22,244 - Epoch: [63][ 920/ 1236] Overall Loss 0.252946 Objective Loss 0.252946 LR 0.001000 Time 0.021602 +2023-10-02 20:59:22,455 - Epoch: [63][ 930/ 1236] Overall Loss 0.252717 Objective Loss 0.252717 LR 0.001000 Time 0.021596 +2023-10-02 20:59:22,661 - Epoch: [63][ 940/ 1236] Overall Loss 0.252779 Objective Loss 0.252779 LR 0.001000 Time 0.021585 +2023-10-02 20:59:22,871 - Epoch: [63][ 950/ 1236] Overall Loss 0.252895 Objective Loss 0.252895 LR 0.001000 Time 0.021579 +2023-10-02 20:59:23,077 - Epoch: [63][ 960/ 1236] Overall Loss 0.253024 Objective Loss 0.253024 LR 0.001000 Time 0.021569 +2023-10-02 20:59:23,288 - Epoch: [63][ 970/ 1236] Overall Loss 0.252982 Objective Loss 0.252982 LR 0.001000 Time 0.021563 +2023-10-02 20:59:23,494 - Epoch: [63][ 980/ 1236] Overall Loss 0.253095 Objective Loss 0.253095 LR 0.001000 Time 0.021553 +2023-10-02 20:59:23,704 - Epoch: [63][ 990/ 1236] Overall Loss 0.253193 Objective Loss 0.253193 LR 0.001000 Time 0.021547 +2023-10-02 20:59:23,910 - Epoch: [63][ 1000/ 1236] Overall Loss 0.252995 Objective Loss 0.252995 LR 0.001000 Time 0.021537 +2023-10-02 20:59:24,120 - Epoch: [63][ 1010/ 1236] Overall Loss 0.252977 Objective Loss 0.252977 LR 0.001000 Time 0.021532 +2023-10-02 20:59:24,326 - Epoch: [63][ 1020/ 1236] Overall Loss 0.252991 Objective Loss 0.252991 LR 0.001000 Time 0.021522 +2023-10-02 20:59:24,536 - Epoch: [63][ 1030/ 1236] Overall Loss 0.252911 Objective Loss 0.252911 LR 0.001000 Time 0.021517 +2023-10-02 20:59:24,742 - Epoch: [63][ 1040/ 1236] Overall Loss 0.252997 Objective Loss 0.252997 LR 0.001000 Time 0.021507 +2023-10-02 20:59:24,950 - Epoch: [63][ 1050/ 1236] Overall Loss 0.253370 Objective Loss 0.253370 LR 0.001000 Time 0.021500 +2023-10-02 20:59:25,155 - Epoch: [63][ 1060/ 1236] Overall Loss 0.253361 Objective Loss 0.253361 LR 0.001000 Time 0.021491 +2023-10-02 20:59:25,362 - Epoch: [63][ 1070/ 1236] Overall Loss 0.253410 Objective Loss 0.253410 LR 0.001000 Time 0.021483 +2023-10-02 20:59:25,569 - Epoch: [63][ 1080/ 1236] Overall Loss 0.253801 Objective Loss 0.253801 LR 0.001000 Time 0.021476 +2023-10-02 20:59:25,777 - Epoch: [63][ 1090/ 1236] Overall Loss 0.253742 Objective Loss 0.253742 LR 0.001000 Time 0.021469 +2023-10-02 20:59:25,983 - Epoch: [63][ 1100/ 1236] Overall Loss 0.253493 Objective Loss 0.253493 LR 0.001000 Time 0.021461 +2023-10-02 20:59:26,190 - Epoch: [63][ 1110/ 1236] Overall Loss 0.253538 Objective Loss 0.253538 LR 0.001000 Time 0.021454 +2023-10-02 20:59:26,396 - Epoch: [63][ 1120/ 1236] Overall Loss 0.253304 Objective Loss 0.253304 LR 0.001000 Time 0.021446 +2023-10-02 20:59:26,603 - Epoch: [63][ 1130/ 1236] Overall Loss 0.253523 Objective Loss 0.253523 LR 0.001000 Time 0.021439 +2023-10-02 20:59:26,810 - Epoch: [63][ 1140/ 1236] Overall Loss 0.253756 Objective Loss 0.253756 LR 0.001000 Time 0.021432 +2023-10-02 20:59:27,017 - Epoch: [63][ 1150/ 1236] Overall Loss 0.253972 Objective Loss 0.253972 LR 0.001000 Time 0.021425 +2023-10-02 20:59:27,223 - Epoch: [63][ 1160/ 1236] Overall Loss 0.254032 Objective Loss 0.254032 LR 0.001000 Time 0.021417 +2023-10-02 20:59:27,433 - Epoch: [63][ 1170/ 1236] Overall Loss 0.254303 Objective Loss 0.254303 LR 0.001000 Time 0.021414 +2023-10-02 20:59:27,644 - Epoch: [63][ 1180/ 1236] Overall Loss 0.254511 Objective Loss 0.254511 LR 0.001000 Time 0.021410 +2023-10-02 20:59:27,851 - Epoch: [63][ 1190/ 1236] Overall Loss 0.254747 Objective Loss 0.254747 LR 0.001000 Time 0.021404 +2023-10-02 20:59:28,057 - Epoch: [63][ 1200/ 1236] Overall Loss 0.254621 Objective Loss 0.254621 LR 0.001000 Time 0.021396 +2023-10-02 20:59:28,264 - Epoch: [63][ 1210/ 1236] Overall Loss 0.254758 Objective Loss 0.254758 LR 0.001000 Time 0.021390 +2023-10-02 20:59:28,470 - Epoch: [63][ 1220/ 1236] Overall Loss 0.254978 Objective Loss 0.254978 LR 0.001000 Time 0.021382 +2023-10-02 20:59:28,729 - Epoch: [63][ 1230/ 1236] Overall Loss 0.255443 Objective Loss 0.255443 LR 0.001000 Time 0.021418 +2023-10-02 20:59:28,850 - Epoch: [63][ 1236/ 1236] Overall Loss 0.255426 Objective Loss 0.255426 Top1 88.187373 Top5 99.592668 LR 0.001000 Time 0.021412 +2023-10-02 20:59:28,997 - --- validate (epoch=63)----------- +2023-10-02 20:59:28,997 - 29943 samples (256 per mini-batch) +2023-10-02 20:59:29,492 - Epoch: [63][ 10/ 117] Loss 0.339853 Top1 82.851562 Top5 98.281250 +2023-10-02 20:59:29,647 - Epoch: [63][ 20/ 117] Loss 0.364777 Top1 82.421875 Top5 98.183594 +2023-10-02 20:59:29,799 - Epoch: [63][ 30/ 117] Loss 0.355404 Top1 82.682292 Top5 98.190104 +2023-10-02 20:59:29,952 - Epoch: [63][ 40/ 117] Loss 0.358467 Top1 82.646484 Top5 98.076172 +2023-10-02 20:59:30,102 - Epoch: [63][ 50/ 117] Loss 0.351276 Top1 82.937500 Top5 98.093750 +2023-10-02 20:59:30,255 - Epoch: [63][ 60/ 117] Loss 0.344795 Top1 83.085938 Top5 98.092448 +2023-10-02 20:59:30,410 - Epoch: [63][ 70/ 117] Loss 0.345141 Top1 83.286830 Top5 98.125000 +2023-10-02 20:59:30,567 - Epoch: [63][ 80/ 117] Loss 0.342354 Top1 83.374023 Top5 98.144531 +2023-10-02 20:59:30,727 - Epoch: [63][ 90/ 117] Loss 0.343338 Top1 83.346354 Top5 98.146701 +2023-10-02 20:59:30,888 - Epoch: [63][ 100/ 117] Loss 0.339386 Top1 83.519531 Top5 98.164062 +2023-10-02 20:59:31,054 - Epoch: [63][ 110/ 117] Loss 0.339229 Top1 83.522727 Top5 98.174716 +2023-10-02 20:59:31,147 - Epoch: [63][ 117/ 117] Loss 0.340728 Top1 83.488628 Top5 98.176535 +2023-10-02 20:59:31,244 - ==> Top1: 83.489 Top5: 98.177 Loss: 0.341 + +2023-10-02 20:59:31,245 - ==> Confusion: +[[ 905 1 1 1 3 2 0 3 9 84 2 1 1 3 8 1 4 6 2 0 13] + [ 1 1040 2 1 4 22 2 20 3 1 2 1 1 0 1 4 3 0 16 1 6] + [ 5 0 941 8 6 0 25 8 0 3 3 2 4 1 4 7 0 4 16 5 14] + [ 2 5 14 944 0 4 0 1 7 3 9 0 10 5 30 2 1 7 28 1 16] + [ 17 10 1 0 960 6 0 0 0 13 3 4 0 5 12 4 11 0 0 2 2] + [ 1 48 0 0 0 962 1 21 0 11 1 11 6 9 6 0 4 3 10 6 16] + [ 0 7 28 1 0 0 1120 8 0 0 4 4 0 0 1 3 0 0 0 6 9] + [ 3 24 10 0 3 27 5 1047 0 4 2 9 3 3 0 0 1 3 52 13 9] + [ 15 1 0 0 2 1 0 2 974 46 8 4 2 7 16 2 3 2 1 1 2] + [ 66 0 1 1 7 2 1 0 32 981 1 0 0 9 8 1 1 2 0 1 5] + [ 0 4 10 7 1 2 3 2 23 0 937 2 0 19 7 1 1 3 15 4 12] + [ 0 1 1 0 1 16 0 3 1 1 0 901 65 4 0 6 3 18 0 11 3] + [ 0 1 3 4 0 3 1 2 2 2 3 28 969 1 1 7 0 23 2 4 12] + [ 1 1 1 1 3 11 1 0 12 19 6 7 1 1029 7 0 2 1 0 3 13] + [ 12 0 5 11 3 0 0 0 24 3 4 1 4 4 1006 0 0 6 11 0 7] + [ 0 0 1 1 6 0 2 0 0 1 0 9 10 1 0 1058 15 15 2 4 9] + [ 1 15 0 0 5 5 2 0 1 1 0 4 1 3 2 9 1101 1 0 1 9] + [ 0 0 1 2 0 0 2 0 0 1 0 3 18 0 0 8 1 992 2 1 7] + [ 3 4 6 10 0 0 0 15 5 1 4 1 0 0 11 0 1 0 996 1 10] + [ 0 2 6 0 0 5 6 9 1 1 1 13 3 3 0 5 8 3 0 1074 12] + [ 134 210 115 48 97 177 45 82 118 91 152 109 349 262 159 77 160 84 202 172 5062]] + +2023-10-02 20:59:31,246 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 20:59:31,247 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 20:59:31,253 - + +2023-10-02 20:59:31,253 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 20:59:32,287 - Epoch: [64][ 10/ 1236] Overall Loss 0.260358 Objective Loss 0.260358 LR 0.001000 Time 0.103302 +2023-10-02 20:59:32,497 - Epoch: [64][ 20/ 1236] Overall Loss 0.284805 Objective Loss 0.284805 LR 0.001000 Time 0.062126 +2023-10-02 20:59:32,706 - Epoch: [64][ 30/ 1236] Overall Loss 0.287232 Objective Loss 0.287232 LR 0.001000 Time 0.048333 +2023-10-02 20:59:32,917 - Epoch: [64][ 40/ 1236] Overall Loss 0.280291 Objective Loss 0.280291 LR 0.001000 Time 0.041531 +2023-10-02 20:59:33,125 - Epoch: [64][ 50/ 1236] Overall Loss 0.274425 Objective Loss 0.274425 LR 0.001000 Time 0.037377 +2023-10-02 20:59:33,335 - Epoch: [64][ 60/ 1236] Overall Loss 0.268688 Objective Loss 0.268688 LR 0.001000 Time 0.034643 +2023-10-02 20:59:33,544 - Epoch: [64][ 70/ 1236] Overall Loss 0.265408 Objective Loss 0.265408 LR 0.001000 Time 0.032683 +2023-10-02 20:59:33,755 - Epoch: [64][ 80/ 1236] Overall Loss 0.266175 Objective Loss 0.266175 LR 0.001000 Time 0.031231 +2023-10-02 20:59:33,963 - Epoch: [64][ 90/ 1236] Overall Loss 0.266151 Objective Loss 0.266151 LR 0.001000 Time 0.030070 +2023-10-02 20:59:34,173 - Epoch: [64][ 100/ 1236] Overall Loss 0.264737 Objective Loss 0.264737 LR 0.001000 Time 0.029159 +2023-10-02 20:59:34,383 - Epoch: [64][ 110/ 1236] Overall Loss 0.263954 Objective Loss 0.263954 LR 0.001000 Time 0.028399 +2023-10-02 20:59:34,594 - Epoch: [64][ 120/ 1236] Overall Loss 0.262256 Objective Loss 0.262256 LR 0.001000 Time 0.027787 +2023-10-02 20:59:34,802 - Epoch: [64][ 130/ 1236] Overall Loss 0.264114 Objective Loss 0.264114 LR 0.001000 Time 0.027251 +2023-10-02 20:59:35,012 - Epoch: [64][ 140/ 1236] Overall Loss 0.264043 Objective Loss 0.264043 LR 0.001000 Time 0.026802 +2023-10-02 20:59:35,222 - Epoch: [64][ 150/ 1236] Overall Loss 0.262575 Objective Loss 0.262575 LR 0.001000 Time 0.026401 +2023-10-02 20:59:35,432 - Epoch: [64][ 160/ 1236] Overall Loss 0.261815 Objective Loss 0.261815 LR 0.001000 Time 0.026062 +2023-10-02 20:59:35,641 - Epoch: [64][ 170/ 1236] Overall Loss 0.261592 Objective Loss 0.261592 LR 0.001000 Time 0.025754 +2023-10-02 20:59:35,851 - Epoch: [64][ 180/ 1236] Overall Loss 0.263272 Objective Loss 0.263272 LR 0.001000 Time 0.025487 +2023-10-02 20:59:36,061 - Epoch: [64][ 190/ 1236] Overall Loss 0.262849 Objective Loss 0.262849 LR 0.001000 Time 0.025240 +2023-10-02 20:59:36,270 - Epoch: [64][ 200/ 1236] Overall Loss 0.263093 Objective Loss 0.263093 LR 0.001000 Time 0.025024 +2023-10-02 20:59:36,479 - Epoch: [64][ 210/ 1236] Overall Loss 0.263311 Objective Loss 0.263311 LR 0.001000 Time 0.024818 +2023-10-02 20:59:36,687 - Epoch: [64][ 220/ 1236] Overall Loss 0.261605 Objective Loss 0.261605 LR 0.001000 Time 0.024636 +2023-10-02 20:59:36,894 - Epoch: [64][ 230/ 1236] Overall Loss 0.262368 Objective Loss 0.262368 LR 0.001000 Time 0.024460 +2023-10-02 20:59:37,103 - Epoch: [64][ 240/ 1236] Overall Loss 0.260602 Objective Loss 0.260602 LR 0.001000 Time 0.024310 +2023-10-02 20:59:37,312 - Epoch: [64][ 250/ 1236] Overall Loss 0.260149 Objective Loss 0.260149 LR 0.001000 Time 0.024167 +2023-10-02 20:59:37,522 - Epoch: [64][ 260/ 1236] Overall Loss 0.259637 Objective Loss 0.259637 LR 0.001000 Time 0.024045 +2023-10-02 20:59:37,730 - Epoch: [64][ 270/ 1236] Overall Loss 0.260113 Objective Loss 0.260113 LR 0.001000 Time 0.023920 +2023-10-02 20:59:37,940 - Epoch: [64][ 280/ 1236] Overall Loss 0.259596 Objective Loss 0.259596 LR 0.001000 Time 0.023812 +2023-10-02 20:59:38,147 - Epoch: [64][ 290/ 1236] Overall Loss 0.259914 Objective Loss 0.259914 LR 0.001000 Time 0.023701 +2023-10-02 20:59:38,357 - Epoch: [64][ 300/ 1236] Overall Loss 0.259512 Objective Loss 0.259512 LR 0.001000 Time 0.023609 +2023-10-02 20:59:38,565 - Epoch: [64][ 310/ 1236] Overall Loss 0.259229 Objective Loss 0.259229 LR 0.001000 Time 0.023514 +2023-10-02 20:59:38,774 - Epoch: [64][ 320/ 1236] Overall Loss 0.259293 Objective Loss 0.259293 LR 0.001000 Time 0.023432 +2023-10-02 20:59:38,982 - Epoch: [64][ 330/ 1236] Overall Loss 0.259902 Objective Loss 0.259902 LR 0.001000 Time 0.023348 +2023-10-02 20:59:39,192 - Epoch: [64][ 340/ 1236] Overall Loss 0.259959 Objective Loss 0.259959 LR 0.001000 Time 0.023277 +2023-10-02 20:59:39,401 - Epoch: [64][ 350/ 1236] Overall Loss 0.259439 Objective Loss 0.259439 LR 0.001000 Time 0.023204 +2023-10-02 20:59:39,613 - Epoch: [64][ 360/ 1236] Overall Loss 0.258993 Objective Loss 0.258993 LR 0.001000 Time 0.023148 +2023-10-02 20:59:39,820 - Epoch: [64][ 370/ 1236] Overall Loss 0.257831 Objective Loss 0.257831 LR 0.001000 Time 0.023083 +2023-10-02 20:59:40,032 - Epoch: [64][ 380/ 1236] Overall Loss 0.257762 Objective Loss 0.257762 LR 0.001000 Time 0.023031 +2023-10-02 20:59:40,239 - Epoch: [64][ 390/ 1236] Overall Loss 0.257623 Objective Loss 0.257623 LR 0.001000 Time 0.022972 +2023-10-02 20:59:40,451 - Epoch: [64][ 400/ 1236] Overall Loss 0.258130 Objective Loss 0.258130 LR 0.001000 Time 0.022926 +2023-10-02 20:59:40,658 - Epoch: [64][ 410/ 1236] Overall Loss 0.257755 Objective Loss 0.257755 LR 0.001000 Time 0.022873 +2023-10-02 20:59:40,870 - Epoch: [64][ 420/ 1236] Overall Loss 0.257973 Objective Loss 0.257973 LR 0.001000 Time 0.022831 +2023-10-02 20:59:41,079 - Epoch: [64][ 430/ 1236] Overall Loss 0.257443 Objective Loss 0.257443 LR 0.001000 Time 0.022783 +2023-10-02 20:59:41,291 - Epoch: [64][ 440/ 1236] Overall Loss 0.257819 Objective Loss 0.257819 LR 0.001000 Time 0.022746 +2023-10-02 20:59:41,498 - Epoch: [64][ 450/ 1236] Overall Loss 0.257128 Objective Loss 0.257128 LR 0.001000 Time 0.022701 +2023-10-02 20:59:41,710 - Epoch: [64][ 460/ 1236] Overall Loss 0.256643 Objective Loss 0.256643 LR 0.001000 Time 0.022667 +2023-10-02 20:59:41,917 - Epoch: [64][ 470/ 1236] Overall Loss 0.256209 Objective Loss 0.256209 LR 0.001000 Time 0.022625 +2023-10-02 20:59:42,129 - Epoch: [64][ 480/ 1236] Overall Loss 0.256765 Objective Loss 0.256765 LR 0.001000 Time 0.022594 +2023-10-02 20:59:42,337 - Epoch: [64][ 490/ 1236] Overall Loss 0.257064 Objective Loss 0.257064 LR 0.001000 Time 0.022556 +2023-10-02 20:59:42,549 - Epoch: [64][ 500/ 1236] Overall Loss 0.256739 Objective Loss 0.256739 LR 0.001000 Time 0.022528 +2023-10-02 20:59:42,756 - Epoch: [64][ 510/ 1236] Overall Loss 0.256606 Objective Loss 0.256606 LR 0.001000 Time 0.022493 +2023-10-02 20:59:42,968 - Epoch: [64][ 520/ 1236] Overall Loss 0.256362 Objective Loss 0.256362 LR 0.001000 Time 0.022467 +2023-10-02 20:59:43,176 - Epoch: [64][ 530/ 1236] Overall Loss 0.256285 Objective Loss 0.256285 LR 0.001000 Time 0.022435 +2023-10-02 20:59:43,388 - Epoch: [64][ 540/ 1236] Overall Loss 0.256672 Objective Loss 0.256672 LR 0.001000 Time 0.022411 +2023-10-02 20:59:43,595 - Epoch: [64][ 550/ 1236] Overall Loss 0.257313 Objective Loss 0.257313 LR 0.001000 Time 0.022380 +2023-10-02 20:59:43,807 - Epoch: [64][ 560/ 1236] Overall Loss 0.257556 Objective Loss 0.257556 LR 0.001000 Time 0.022358 +2023-10-02 20:59:44,014 - Epoch: [64][ 570/ 1236] Overall Loss 0.257848 Objective Loss 0.257848 LR 0.001000 Time 0.022329 +2023-10-02 20:59:44,226 - Epoch: [64][ 580/ 1236] Overall Loss 0.258368 Objective Loss 0.258368 LR 0.001000 Time 0.022309 +2023-10-02 20:59:44,433 - Epoch: [64][ 590/ 1236] Overall Loss 0.258606 Objective Loss 0.258606 LR 0.001000 Time 0.022282 +2023-10-02 20:59:44,645 - Epoch: [64][ 600/ 1236] Overall Loss 0.258643 Objective Loss 0.258643 LR 0.001000 Time 0.022263 +2023-10-02 20:59:44,853 - Epoch: [64][ 610/ 1236] Overall Loss 0.259236 Objective Loss 0.259236 LR 0.001000 Time 0.022238 +2023-10-02 20:59:45,064 - Epoch: [64][ 620/ 1236] Overall Loss 0.260689 Objective Loss 0.260689 LR 0.001000 Time 0.022220 +2023-10-02 20:59:45,272 - Epoch: [64][ 630/ 1236] Overall Loss 0.260720 Objective Loss 0.260720 LR 0.001000 Time 0.022197 +2023-10-02 20:59:45,483 - Epoch: [64][ 640/ 1236] Overall Loss 0.261117 Objective Loss 0.261117 LR 0.001000 Time 0.022178 +2023-10-02 20:59:45,691 - Epoch: [64][ 650/ 1236] Overall Loss 0.261130 Objective Loss 0.261130 LR 0.001000 Time 0.022158 +2023-10-02 20:59:45,903 - Epoch: [64][ 660/ 1236] Overall Loss 0.261939 Objective Loss 0.261939 LR 0.001000 Time 0.022143 +2023-10-02 20:59:46,111 - Epoch: [64][ 670/ 1236] Overall Loss 0.262200 Objective Loss 0.262200 LR 0.001000 Time 0.022121 +2023-10-02 20:59:46,323 - Epoch: [64][ 680/ 1236] Overall Loss 0.262228 Objective Loss 0.262228 LR 0.001000 Time 0.022108 +2023-10-02 20:59:46,530 - Epoch: [64][ 690/ 1236] Overall Loss 0.262413 Objective Loss 0.262413 LR 0.001000 Time 0.022088 +2023-10-02 20:59:46,743 - Epoch: [64][ 700/ 1236] Overall Loss 0.262524 Objective Loss 0.262524 LR 0.001000 Time 0.022076 +2023-10-02 20:59:46,954 - Epoch: [64][ 710/ 1236] Overall Loss 0.262327 Objective Loss 0.262327 LR 0.001000 Time 0.022061 +2023-10-02 20:59:47,179 - Epoch: [64][ 720/ 1236] Overall Loss 0.262721 Objective Loss 0.262721 LR 0.001000 Time 0.022067 +2023-10-02 20:59:47,388 - Epoch: [64][ 730/ 1236] Overall Loss 0.262792 Objective Loss 0.262792 LR 0.001000 Time 0.022051 +2023-10-02 20:59:47,600 - Epoch: [64][ 740/ 1236] Overall Loss 0.262640 Objective Loss 0.262640 LR 0.001000 Time 0.022038 +2023-10-02 20:59:47,820 - Epoch: [64][ 750/ 1236] Overall Loss 0.262682 Objective Loss 0.262682 LR 0.001000 Time 0.022038 +2023-10-02 20:59:48,039 - Epoch: [64][ 760/ 1236] Overall Loss 0.262553 Objective Loss 0.262553 LR 0.001000 Time 0.022035 +2023-10-02 20:59:48,248 - Epoch: [64][ 770/ 1236] Overall Loss 0.262393 Objective Loss 0.262393 LR 0.001000 Time 0.022020 +2023-10-02 20:59:48,459 - Epoch: [64][ 780/ 1236] Overall Loss 0.262430 Objective Loss 0.262430 LR 0.001000 Time 0.022008 +2023-10-02 20:59:48,667 - Epoch: [64][ 790/ 1236] Overall Loss 0.262279 Objective Loss 0.262279 LR 0.001000 Time 0.021993 +2023-10-02 20:59:48,878 - Epoch: [64][ 800/ 1236] Overall Loss 0.261876 Objective Loss 0.261876 LR 0.001000 Time 0.021981 +2023-10-02 20:59:49,087 - Epoch: [64][ 810/ 1236] Overall Loss 0.262005 Objective Loss 0.262005 LR 0.001000 Time 0.021967 +2023-10-02 20:59:49,298 - Epoch: [64][ 820/ 1236] Overall Loss 0.262657 Objective Loss 0.262657 LR 0.001000 Time 0.021956 +2023-10-02 20:59:49,506 - Epoch: [64][ 830/ 1236] Overall Loss 0.262543 Objective Loss 0.262543 LR 0.001000 Time 0.021942 +2023-10-02 20:59:49,717 - Epoch: [64][ 840/ 1236] Overall Loss 0.262674 Objective Loss 0.262674 LR 0.001000 Time 0.021931 +2023-10-02 20:59:49,925 - Epoch: [64][ 850/ 1236] Overall Loss 0.262470 Objective Loss 0.262470 LR 0.001000 Time 0.021918 +2023-10-02 20:59:50,136 - Epoch: [64][ 860/ 1236] Overall Loss 0.261922 Objective Loss 0.261922 LR 0.001000 Time 0.021908 +2023-10-02 20:59:50,345 - Epoch: [64][ 870/ 1236] Overall Loss 0.261866 Objective Loss 0.261866 LR 0.001000 Time 0.021895 +2023-10-02 20:59:50,556 - Epoch: [64][ 880/ 1236] Overall Loss 0.261995 Objective Loss 0.261995 LR 0.001000 Time 0.021886 +2023-10-02 20:59:50,765 - Epoch: [64][ 890/ 1236] Overall Loss 0.261908 Objective Loss 0.261908 LR 0.001000 Time 0.021874 +2023-10-02 20:59:50,976 - Epoch: [64][ 900/ 1236] Overall Loss 0.261749 Objective Loss 0.261749 LR 0.001000 Time 0.021865 +2023-10-02 20:59:51,184 - Epoch: [64][ 910/ 1236] Overall Loss 0.261665 Objective Loss 0.261665 LR 0.001000 Time 0.021853 +2023-10-02 20:59:51,395 - Epoch: [64][ 920/ 1236] Overall Loss 0.261555 Objective Loss 0.261555 LR 0.001000 Time 0.021845 +2023-10-02 20:59:51,604 - Epoch: [64][ 930/ 1236] Overall Loss 0.261577 Objective Loss 0.261577 LR 0.001000 Time 0.021834 +2023-10-02 20:59:51,815 - Epoch: [64][ 940/ 1236] Overall Loss 0.261527 Objective Loss 0.261527 LR 0.001000 Time 0.021826 +2023-10-02 20:59:52,024 - Epoch: [64][ 950/ 1236] Overall Loss 0.261377 Objective Loss 0.261377 LR 0.001000 Time 0.021815 +2023-10-02 20:59:52,235 - Epoch: [64][ 960/ 1236] Overall Loss 0.261159 Objective Loss 0.261159 LR 0.001000 Time 0.021807 +2023-10-02 20:59:52,443 - Epoch: [64][ 970/ 1236] Overall Loss 0.261125 Objective Loss 0.261125 LR 0.001000 Time 0.021797 +2023-10-02 20:59:52,654 - Epoch: [64][ 980/ 1236] Overall Loss 0.261031 Objective Loss 0.261031 LR 0.001000 Time 0.021790 +2023-10-02 20:59:52,863 - Epoch: [64][ 990/ 1236] Overall Loss 0.261037 Objective Loss 0.261037 LR 0.001000 Time 0.021780 +2023-10-02 20:59:53,074 - Epoch: [64][ 1000/ 1236] Overall Loss 0.261113 Objective Loss 0.261113 LR 0.001000 Time 0.021773 +2023-10-02 20:59:53,282 - Epoch: [64][ 1010/ 1236] Overall Loss 0.261338 Objective Loss 0.261338 LR 0.001000 Time 0.021763 +2023-10-02 20:59:53,493 - Epoch: [64][ 1020/ 1236] Overall Loss 0.261057 Objective Loss 0.261057 LR 0.001000 Time 0.021756 +2023-10-02 20:59:53,701 - Epoch: [64][ 1030/ 1236] Overall Loss 0.261354 Objective Loss 0.261354 LR 0.001000 Time 0.021747 +2023-10-02 20:59:53,912 - Epoch: [64][ 1040/ 1236] Overall Loss 0.261385 Objective Loss 0.261385 LR 0.001000 Time 0.021740 +2023-10-02 20:59:54,121 - Epoch: [64][ 1050/ 1236] Overall Loss 0.261273 Objective Loss 0.261273 LR 0.001000 Time 0.021731 +2023-10-02 20:59:54,332 - Epoch: [64][ 1060/ 1236] Overall Loss 0.261537 Objective Loss 0.261537 LR 0.001000 Time 0.021725 +2023-10-02 20:59:54,540 - Epoch: [64][ 1070/ 1236] Overall Loss 0.261612 Objective Loss 0.261612 LR 0.001000 Time 0.021716 +2023-10-02 20:59:54,751 - Epoch: [64][ 1080/ 1236] Overall Loss 0.261879 Objective Loss 0.261879 LR 0.001000 Time 0.021710 +2023-10-02 20:59:54,959 - Epoch: [64][ 1090/ 1236] Overall Loss 0.262246 Objective Loss 0.262246 LR 0.001000 Time 0.021702 +2023-10-02 20:59:55,170 - Epoch: [64][ 1100/ 1236] Overall Loss 0.262240 Objective Loss 0.262240 LR 0.001000 Time 0.021696 +2023-10-02 20:59:55,379 - Epoch: [64][ 1110/ 1236] Overall Loss 0.262264 Objective Loss 0.262264 LR 0.001000 Time 0.021688 +2023-10-02 20:59:55,590 - Epoch: [64][ 1120/ 1236] Overall Loss 0.262445 Objective Loss 0.262445 LR 0.001000 Time 0.021682 +2023-10-02 20:59:55,798 - Epoch: [64][ 1130/ 1236] Overall Loss 0.262712 Objective Loss 0.262712 LR 0.001000 Time 0.021675 +2023-10-02 20:59:56,009 - Epoch: [64][ 1140/ 1236] Overall Loss 0.262950 Objective Loss 0.262950 LR 0.001000 Time 0.021669 +2023-10-02 20:59:56,217 - Epoch: [64][ 1150/ 1236] Overall Loss 0.263276 Objective Loss 0.263276 LR 0.001000 Time 0.021662 +2023-10-02 20:59:56,429 - Epoch: [64][ 1160/ 1236] Overall Loss 0.263277 Objective Loss 0.263277 LR 0.001000 Time 0.021657 +2023-10-02 20:59:56,637 - Epoch: [64][ 1170/ 1236] Overall Loss 0.263703 Objective Loss 0.263703 LR 0.001000 Time 0.021650 +2023-10-02 20:59:56,849 - Epoch: [64][ 1180/ 1236] Overall Loss 0.263557 Objective Loss 0.263557 LR 0.001000 Time 0.021645 +2023-10-02 20:59:57,057 - Epoch: [64][ 1190/ 1236] Overall Loss 0.263524 Objective Loss 0.263524 LR 0.001000 Time 0.021638 +2023-10-02 20:59:57,268 - Epoch: [64][ 1200/ 1236] Overall Loss 0.263457 Objective Loss 0.263457 LR 0.001000 Time 0.021633 +2023-10-02 20:59:57,477 - Epoch: [64][ 1210/ 1236] Overall Loss 0.263793 Objective Loss 0.263793 LR 0.001000 Time 0.021626 +2023-10-02 20:59:57,688 - Epoch: [64][ 1220/ 1236] Overall Loss 0.264000 Objective Loss 0.264000 LR 0.001000 Time 0.021622 +2023-10-02 20:59:57,953 - Epoch: [64][ 1230/ 1236] Overall Loss 0.264010 Objective Loss 0.264010 LR 0.001000 Time 0.021662 +2023-10-02 20:59:58,077 - Epoch: [64][ 1236/ 1236] Overall Loss 0.264216 Objective Loss 0.264216 Top1 86.150713 Top5 97.352342 LR 0.001000 Time 0.021656 +2023-10-02 20:59:58,227 - --- validate (epoch=64)----------- +2023-10-02 20:59:58,227 - 29943 samples (256 per mini-batch) +2023-10-02 20:59:58,717 - Epoch: [64][ 10/ 117] Loss 0.321759 Top1 84.492188 Top5 98.320312 +2023-10-02 20:59:58,864 - Epoch: [64][ 20/ 117] Loss 0.328982 Top1 84.121094 Top5 98.476562 +2023-10-02 20:59:59,011 - Epoch: [64][ 30/ 117] Loss 0.321942 Top1 84.322917 Top5 98.463542 +2023-10-02 20:59:59,158 - Epoch: [64][ 40/ 117] Loss 0.326887 Top1 83.867188 Top5 98.339844 +2023-10-02 20:59:59,305 - Epoch: [64][ 50/ 117] Loss 0.335401 Top1 83.828125 Top5 98.343750 +2023-10-02 20:59:59,453 - Epoch: [64][ 60/ 117] Loss 0.334954 Top1 83.932292 Top5 98.339844 +2023-10-02 20:59:59,601 - Epoch: [64][ 70/ 117] Loss 0.337292 Top1 84.001116 Top5 98.320312 +2023-10-02 20:59:59,752 - Epoch: [64][ 80/ 117] Loss 0.335927 Top1 84.028320 Top5 98.276367 +2023-10-02 20:59:59,901 - Epoch: [64][ 90/ 117] Loss 0.334511 Top1 84.036458 Top5 98.250868 +2023-10-02 21:00:00,048 - Epoch: [64][ 100/ 117] Loss 0.341283 Top1 83.914062 Top5 98.203125 +2023-10-02 21:00:00,203 - Epoch: [64][ 110/ 117] Loss 0.341814 Top1 83.842330 Top5 98.210227 +2023-10-02 21:00:00,293 - Epoch: [64][ 117/ 117] Loss 0.344936 Top1 83.765822 Top5 98.206593 +2023-10-02 21:00:00,423 - ==> Top1: 83.766 Top5: 98.207 Loss: 0.345 + +2023-10-02 21:00:00,424 - ==> Confusion: +[[ 964 2 2 0 5 5 0 0 8 35 2 2 0 2 3 1 2 0 1 1 15] + [ 0 1053 0 0 5 31 1 10 2 0 1 1 0 0 5 3 2 0 8 3 6] + [ 6 0 961 8 3 3 20 6 1 1 5 0 5 1 3 5 0 1 9 8 10] + [ 4 7 16 966 0 1 1 2 1 0 9 0 8 3 30 2 1 6 16 1 15] + [ 38 10 1 0 946 8 0 0 1 7 1 1 0 4 13 5 6 0 0 3 6] + [ 5 47 1 4 4 969 0 21 4 4 6 4 6 15 7 0 0 0 5 4 10] + [ 0 9 36 0 0 2 1107 5 0 0 8 3 0 0 1 5 0 0 0 6 9] + [ 3 35 23 0 2 31 3 1020 2 1 8 10 4 3 1 1 1 0 46 15 9] + [ 23 5 1 2 2 4 0 1 970 30 13 2 1 13 11 1 4 1 2 0 3] + [ 169 0 0 0 4 5 1 0 50 835 4 1 0 26 11 0 1 0 0 1 11] + [ 4 0 7 5 2 2 1 2 15 1 968 3 0 12 6 1 1 2 10 2 9] + [ 1 0 4 0 0 9 0 1 0 0 1 945 26 4 0 6 4 18 0 8 8] + [ 2 1 3 5 1 5 0 1 3 0 2 45 938 2 5 9 1 25 1 5 14] + [ 1 0 2 0 3 7 0 0 11 7 5 3 0 1058 4 1 2 1 0 2 12] + [ 13 2 2 15 5 0 1 0 24 1 2 0 4 3 1013 0 1 3 9 1 2] + [ 0 0 0 0 8 1 2 0 0 0 1 2 4 1 0 1077 9 16 2 4 7] + [ 3 19 2 0 5 9 1 1 0 0 0 5 3 2 5 10 1075 0 1 9 11] + [ 0 1 0 3 0 1 3 0 0 1 0 4 11 0 6 6 3 994 0 0 5] + [ 4 8 4 10 0 1 0 11 6 0 2 1 0 0 17 0 2 0 985 0 17] + [ 0 0 4 1 2 6 9 6 0 0 2 13 5 0 0 2 3 1 0 1096 2] + [ 182 205 140 63 71 162 37 69 93 60 180 123 321 324 142 71 103 69 133 215 5142]] + +2023-10-02 21:00:00,425 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:00:00,425 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:00:00,431 - + +2023-10-02 21:00:00,431 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:00:01,578 - Epoch: [65][ 10/ 1236] Overall Loss 0.259019 Objective Loss 0.259019 LR 0.001000 Time 0.114638 +2023-10-02 21:00:01,787 - Epoch: [65][ 20/ 1236] Overall Loss 0.261808 Objective Loss 0.261808 LR 0.001000 Time 0.067731 +2023-10-02 21:00:01,995 - Epoch: [65][ 30/ 1236] Overall Loss 0.261735 Objective Loss 0.261735 LR 0.001000 Time 0.052043 +2023-10-02 21:00:02,204 - Epoch: [65][ 40/ 1236] Overall Loss 0.264199 Objective Loss 0.264199 LR 0.001000 Time 0.044250 +2023-10-02 21:00:02,412 - Epoch: [65][ 50/ 1236] Overall Loss 0.256181 Objective Loss 0.256181 LR 0.001000 Time 0.039534 +2023-10-02 21:00:02,621 - Epoch: [65][ 60/ 1236] Overall Loss 0.249242 Objective Loss 0.249242 LR 0.001000 Time 0.036424 +2023-10-02 21:00:02,829 - Epoch: [65][ 70/ 1236] Overall Loss 0.252164 Objective Loss 0.252164 LR 0.001000 Time 0.034171 +2023-10-02 21:00:03,038 - Epoch: [65][ 80/ 1236] Overall Loss 0.251175 Objective Loss 0.251175 LR 0.001000 Time 0.032509 +2023-10-02 21:00:03,247 - Epoch: [65][ 90/ 1236] Overall Loss 0.250643 Objective Loss 0.250643 LR 0.001000 Time 0.031195 +2023-10-02 21:00:03,456 - Epoch: [65][ 100/ 1236] Overall Loss 0.249530 Objective Loss 0.249530 LR 0.001000 Time 0.030162 +2023-10-02 21:00:03,664 - Epoch: [65][ 110/ 1236] Overall Loss 0.250420 Objective Loss 0.250420 LR 0.001000 Time 0.029300 +2023-10-02 21:00:03,873 - Epoch: [65][ 120/ 1236] Overall Loss 0.251570 Objective Loss 0.251570 LR 0.001000 Time 0.028598 +2023-10-02 21:00:04,081 - Epoch: [65][ 130/ 1236] Overall Loss 0.252231 Objective Loss 0.252231 LR 0.001000 Time 0.027989 +2023-10-02 21:00:04,290 - Epoch: [65][ 140/ 1236] Overall Loss 0.251925 Objective Loss 0.251925 LR 0.001000 Time 0.027482 +2023-10-02 21:00:04,499 - Epoch: [65][ 150/ 1236] Overall Loss 0.250724 Objective Loss 0.250724 LR 0.001000 Time 0.027029 +2023-10-02 21:00:04,707 - Epoch: [65][ 160/ 1236] Overall Loss 0.249725 Objective Loss 0.249725 LR 0.001000 Time 0.026639 +2023-10-02 21:00:04,915 - Epoch: [65][ 170/ 1236] Overall Loss 0.249785 Objective Loss 0.249785 LR 0.001000 Time 0.026289 +2023-10-02 21:00:05,124 - Epoch: [65][ 180/ 1236] Overall Loss 0.251919 Objective Loss 0.251919 LR 0.001000 Time 0.025987 +2023-10-02 21:00:05,332 - Epoch: [65][ 190/ 1236] Overall Loss 0.252111 Objective Loss 0.252111 LR 0.001000 Time 0.025705 +2023-10-02 21:00:05,540 - Epoch: [65][ 200/ 1236] Overall Loss 0.251280 Objective Loss 0.251280 LR 0.001000 Time 0.025461 +2023-10-02 21:00:05,748 - Epoch: [65][ 210/ 1236] Overall Loss 0.252211 Objective Loss 0.252211 LR 0.001000 Time 0.025231 +2023-10-02 21:00:05,957 - Epoch: [65][ 220/ 1236] Overall Loss 0.252297 Objective Loss 0.252297 LR 0.001000 Time 0.025031 +2023-10-02 21:00:06,165 - Epoch: [65][ 230/ 1236] Overall Loss 0.253421 Objective Loss 0.253421 LR 0.001000 Time 0.024840 +2023-10-02 21:00:06,373 - Epoch: [65][ 240/ 1236] Overall Loss 0.253102 Objective Loss 0.253102 LR 0.001000 Time 0.024673 +2023-10-02 21:00:06,581 - Epoch: [65][ 250/ 1236] Overall Loss 0.254461 Objective Loss 0.254461 LR 0.001000 Time 0.024512 +2023-10-02 21:00:06,789 - Epoch: [65][ 260/ 1236] Overall Loss 0.256440 Objective Loss 0.256440 LR 0.001000 Time 0.024369 +2023-10-02 21:00:06,997 - Epoch: [65][ 270/ 1236] Overall Loss 0.256139 Objective Loss 0.256139 LR 0.001000 Time 0.024230 +2023-10-02 21:00:07,205 - Epoch: [65][ 280/ 1236] Overall Loss 0.256270 Objective Loss 0.256270 LR 0.001000 Time 0.024107 +2023-10-02 21:00:07,413 - Epoch: [65][ 290/ 1236] Overall Loss 0.255768 Objective Loss 0.255768 LR 0.001000 Time 0.023988 +2023-10-02 21:00:07,622 - Epoch: [65][ 300/ 1236] Overall Loss 0.256470 Objective Loss 0.256470 LR 0.001000 Time 0.023882 +2023-10-02 21:00:07,830 - Epoch: [65][ 310/ 1236] Overall Loss 0.258293 Objective Loss 0.258293 LR 0.001000 Time 0.023778 +2023-10-02 21:00:08,038 - Epoch: [65][ 320/ 1236] Overall Loss 0.258522 Objective Loss 0.258522 LR 0.001000 Time 0.023685 +2023-10-02 21:00:08,246 - Epoch: [65][ 330/ 1236] Overall Loss 0.259365 Objective Loss 0.259365 LR 0.001000 Time 0.023593 +2023-10-02 21:00:08,455 - Epoch: [65][ 340/ 1236] Overall Loss 0.259089 Objective Loss 0.259089 LR 0.001000 Time 0.023513 +2023-10-02 21:00:08,663 - Epoch: [65][ 350/ 1236] Overall Loss 0.259745 Objective Loss 0.259745 LR 0.001000 Time 0.023431 +2023-10-02 21:00:08,872 - Epoch: [65][ 360/ 1236] Overall Loss 0.259950 Objective Loss 0.259950 LR 0.001000 Time 0.023358 +2023-10-02 21:00:09,081 - Epoch: [65][ 370/ 1236] Overall Loss 0.259369 Objective Loss 0.259369 LR 0.001000 Time 0.023292 +2023-10-02 21:00:09,290 - Epoch: [65][ 380/ 1236] Overall Loss 0.258597 Objective Loss 0.258597 LR 0.001000 Time 0.023229 +2023-10-02 21:00:09,499 - Epoch: [65][ 390/ 1236] Overall Loss 0.259014 Objective Loss 0.259014 LR 0.001000 Time 0.023169 +2023-10-02 21:00:09,709 - Epoch: [65][ 400/ 1236] Overall Loss 0.258800 Objective Loss 0.258800 LR 0.001000 Time 0.023112 +2023-10-02 21:00:09,918 - Epoch: [65][ 410/ 1236] Overall Loss 0.258865 Objective Loss 0.258865 LR 0.001000 Time 0.023058 +2023-10-02 21:00:10,127 - Epoch: [65][ 420/ 1236] Overall Loss 0.258946 Objective Loss 0.258946 LR 0.001000 Time 0.023007 +2023-10-02 21:00:10,336 - Epoch: [65][ 430/ 1236] Overall Loss 0.258792 Objective Loss 0.258792 LR 0.001000 Time 0.022957 +2023-10-02 21:00:10,547 - Epoch: [65][ 440/ 1236] Overall Loss 0.258862 Objective Loss 0.258862 LR 0.001000 Time 0.022912 +2023-10-02 21:00:10,756 - Epoch: [65][ 450/ 1236] Overall Loss 0.258842 Objective Loss 0.258842 LR 0.001000 Time 0.022867 +2023-10-02 21:00:10,965 - Epoch: [65][ 460/ 1236] Overall Loss 0.258842 Objective Loss 0.258842 LR 0.001000 Time 0.022825 +2023-10-02 21:00:11,174 - Epoch: [65][ 470/ 1236] Overall Loss 0.259009 Objective Loss 0.259009 LR 0.001000 Time 0.022783 +2023-10-02 21:00:11,384 - Epoch: [65][ 480/ 1236] Overall Loss 0.259183 Objective Loss 0.259183 LR 0.001000 Time 0.022744 +2023-10-02 21:00:11,593 - Epoch: [65][ 490/ 1236] Overall Loss 0.259681 Objective Loss 0.259681 LR 0.001000 Time 0.022706 +2023-10-02 21:00:11,803 - Epoch: [65][ 500/ 1236] Overall Loss 0.259689 Objective Loss 0.259689 LR 0.001000 Time 0.022671 +2023-10-02 21:00:12,012 - Epoch: [65][ 510/ 1236] Overall Loss 0.259857 Objective Loss 0.259857 LR 0.001000 Time 0.022636 +2023-10-02 21:00:12,221 - Epoch: [65][ 520/ 1236] Overall Loss 0.260353 Objective Loss 0.260353 LR 0.001000 Time 0.022603 +2023-10-02 21:00:12,430 - Epoch: [65][ 530/ 1236] Overall Loss 0.261028 Objective Loss 0.261028 LR 0.001000 Time 0.022570 +2023-10-02 21:00:12,640 - Epoch: [65][ 540/ 1236] Overall Loss 0.260860 Objective Loss 0.260860 LR 0.001000 Time 0.022540 +2023-10-02 21:00:12,849 - Epoch: [65][ 550/ 1236] Overall Loss 0.260962 Objective Loss 0.260962 LR 0.001000 Time 0.022510 +2023-10-02 21:00:13,059 - Epoch: [65][ 560/ 1236] Overall Loss 0.260973 Objective Loss 0.260973 LR 0.001000 Time 0.022481 +2023-10-02 21:00:13,268 - Epoch: [65][ 570/ 1236] Overall Loss 0.261299 Objective Loss 0.261299 LR 0.001000 Time 0.022452 +2023-10-02 21:00:13,477 - Epoch: [65][ 580/ 1236] Overall Loss 0.261360 Objective Loss 0.261360 LR 0.001000 Time 0.022426 +2023-10-02 21:00:13,686 - Epoch: [65][ 590/ 1236] Overall Loss 0.261333 Objective Loss 0.261333 LR 0.001000 Time 0.022400 +2023-10-02 21:00:13,896 - Epoch: [65][ 600/ 1236] Overall Loss 0.260884 Objective Loss 0.260884 LR 0.001000 Time 0.022375 +2023-10-02 21:00:14,105 - Epoch: [65][ 610/ 1236] Overall Loss 0.261029 Objective Loss 0.261029 LR 0.001000 Time 0.022351 +2023-10-02 21:00:14,315 - Epoch: [65][ 620/ 1236] Overall Loss 0.260749 Objective Loss 0.260749 LR 0.001000 Time 0.022328 +2023-10-02 21:00:14,524 - Epoch: [65][ 630/ 1236] Overall Loss 0.260817 Objective Loss 0.260817 LR 0.001000 Time 0.022306 +2023-10-02 21:00:14,734 - Epoch: [65][ 640/ 1236] Overall Loss 0.260672 Objective Loss 0.260672 LR 0.001000 Time 0.022284 +2023-10-02 21:00:14,943 - Epoch: [65][ 650/ 1236] Overall Loss 0.260524 Objective Loss 0.260524 LR 0.001000 Time 0.022263 +2023-10-02 21:00:15,153 - Epoch: [65][ 660/ 1236] Overall Loss 0.260415 Objective Loss 0.260415 LR 0.001000 Time 0.022243 +2023-10-02 21:00:15,362 - Epoch: [65][ 670/ 1236] Overall Loss 0.260404 Objective Loss 0.260404 LR 0.001000 Time 0.022223 +2023-10-02 21:00:15,572 - Epoch: [65][ 680/ 1236] Overall Loss 0.260699 Objective Loss 0.260699 LR 0.001000 Time 0.022204 +2023-10-02 21:00:15,781 - Epoch: [65][ 690/ 1236] Overall Loss 0.260706 Objective Loss 0.260706 LR 0.001000 Time 0.022185 +2023-10-02 21:00:15,991 - Epoch: [65][ 700/ 1236] Overall Loss 0.260766 Objective Loss 0.260766 LR 0.001000 Time 0.022167 +2023-10-02 21:00:16,200 - Epoch: [65][ 710/ 1236] Overall Loss 0.261121 Objective Loss 0.261121 LR 0.001000 Time 0.022149 +2023-10-02 21:00:16,410 - Epoch: [65][ 720/ 1236] Overall Loss 0.261035 Objective Loss 0.261035 LR 0.001000 Time 0.022132 +2023-10-02 21:00:16,619 - Epoch: [65][ 730/ 1236] Overall Loss 0.261016 Objective Loss 0.261016 LR 0.001000 Time 0.022115 +2023-10-02 21:00:16,828 - Epoch: [65][ 740/ 1236] Overall Loss 0.261001 Objective Loss 0.261001 LR 0.001000 Time 0.022099 +2023-10-02 21:00:17,037 - Epoch: [65][ 750/ 1236] Overall Loss 0.261033 Objective Loss 0.261033 LR 0.001000 Time 0.022082 +2023-10-02 21:00:17,247 - Epoch: [65][ 760/ 1236] Overall Loss 0.261322 Objective Loss 0.261322 LR 0.001000 Time 0.022068 +2023-10-02 21:00:17,457 - Epoch: [65][ 770/ 1236] Overall Loss 0.261547 Objective Loss 0.261547 LR 0.001000 Time 0.022052 +2023-10-02 21:00:17,666 - Epoch: [65][ 780/ 1236] Overall Loss 0.261424 Objective Loss 0.261424 LR 0.001000 Time 0.022038 +2023-10-02 21:00:17,875 - Epoch: [65][ 790/ 1236] Overall Loss 0.261518 Objective Loss 0.261518 LR 0.001000 Time 0.022023 +2023-10-02 21:00:18,085 - Epoch: [65][ 800/ 1236] Overall Loss 0.261496 Objective Loss 0.261496 LR 0.001000 Time 0.022009 +2023-10-02 21:00:18,294 - Epoch: [65][ 810/ 1236] Overall Loss 0.261030 Objective Loss 0.261030 LR 0.001000 Time 0.021996 +2023-10-02 21:00:18,504 - Epoch: [65][ 820/ 1236] Overall Loss 0.261264 Objective Loss 0.261264 LR 0.001000 Time 0.021983 +2023-10-02 21:00:18,713 - Epoch: [65][ 830/ 1236] Overall Loss 0.261416 Objective Loss 0.261416 LR 0.001000 Time 0.021970 +2023-10-02 21:00:18,923 - Epoch: [65][ 840/ 1236] Overall Loss 0.260975 Objective Loss 0.260975 LR 0.001000 Time 0.021957 +2023-10-02 21:00:19,132 - Epoch: [65][ 850/ 1236] Overall Loss 0.261101 Objective Loss 0.261101 LR 0.001000 Time 0.021945 +2023-10-02 21:00:19,342 - Epoch: [65][ 860/ 1236] Overall Loss 0.261225 Objective Loss 0.261225 LR 0.001000 Time 0.021933 +2023-10-02 21:00:19,551 - Epoch: [65][ 870/ 1236] Overall Loss 0.261204 Objective Loss 0.261204 LR 0.001000 Time 0.021921 +2023-10-02 21:00:19,761 - Epoch: [65][ 880/ 1236] Overall Loss 0.261466 Objective Loss 0.261466 LR 0.001000 Time 0.021910 +2023-10-02 21:00:19,970 - Epoch: [65][ 890/ 1236] Overall Loss 0.261475 Objective Loss 0.261475 LR 0.001000 Time 0.021899 +2023-10-02 21:00:20,180 - Epoch: [65][ 900/ 1236] Overall Loss 0.261779 Objective Loss 0.261779 LR 0.001000 Time 0.021888 +2023-10-02 21:00:20,389 - Epoch: [65][ 910/ 1236] Overall Loss 0.261903 Objective Loss 0.261903 LR 0.001000 Time 0.021877 +2023-10-02 21:00:20,599 - Epoch: [65][ 920/ 1236] Overall Loss 0.262147 Objective Loss 0.262147 LR 0.001000 Time 0.021867 +2023-10-02 21:00:20,808 - Epoch: [65][ 930/ 1236] Overall Loss 0.262086 Objective Loss 0.262086 LR 0.001000 Time 0.021857 +2023-10-02 21:00:21,018 - Epoch: [65][ 940/ 1236] Overall Loss 0.261867 Objective Loss 0.261867 LR 0.001000 Time 0.021847 +2023-10-02 21:00:21,227 - Epoch: [65][ 950/ 1236] Overall Loss 0.261568 Objective Loss 0.261568 LR 0.001000 Time 0.021837 +2023-10-02 21:00:21,437 - Epoch: [65][ 960/ 1236] Overall Loss 0.261547 Objective Loss 0.261547 LR 0.001000 Time 0.021828 +2023-10-02 21:00:21,646 - Epoch: [65][ 970/ 1236] Overall Loss 0.261993 Objective Loss 0.261993 LR 0.001000 Time 0.021818 +2023-10-02 21:00:21,856 - Epoch: [65][ 980/ 1236] Overall Loss 0.262235 Objective Loss 0.262235 LR 0.001000 Time 0.021809 +2023-10-02 21:00:22,065 - Epoch: [65][ 990/ 1236] Overall Loss 0.262134 Objective Loss 0.262134 LR 0.001000 Time 0.021799 +2023-10-02 21:00:22,275 - Epoch: [65][ 1000/ 1236] Overall Loss 0.262148 Objective Loss 0.262148 LR 0.001000 Time 0.021791 +2023-10-02 21:00:22,484 - Epoch: [65][ 1010/ 1236] Overall Loss 0.262055 Objective Loss 0.262055 LR 0.001000 Time 0.021782 +2023-10-02 21:00:22,694 - Epoch: [65][ 1020/ 1236] Overall Loss 0.262136 Objective Loss 0.262136 LR 0.001000 Time 0.021774 +2023-10-02 21:00:22,903 - Epoch: [65][ 1030/ 1236] Overall Loss 0.262239 Objective Loss 0.262239 LR 0.001000 Time 0.021765 +2023-10-02 21:00:23,113 - Epoch: [65][ 1040/ 1236] Overall Loss 0.262269 Objective Loss 0.262269 LR 0.001000 Time 0.021757 +2023-10-02 21:00:23,322 - Epoch: [65][ 1050/ 1236] Overall Loss 0.262319 Objective Loss 0.262319 LR 0.001000 Time 0.021749 +2023-10-02 21:00:23,532 - Epoch: [65][ 1060/ 1236] Overall Loss 0.262229 Objective Loss 0.262229 LR 0.001000 Time 0.021741 +2023-10-02 21:00:23,741 - Epoch: [65][ 1070/ 1236] Overall Loss 0.262398 Objective Loss 0.262398 LR 0.001000 Time 0.021733 +2023-10-02 21:00:23,951 - Epoch: [65][ 1080/ 1236] Overall Loss 0.262307 Objective Loss 0.262307 LR 0.001000 Time 0.021726 +2023-10-02 21:00:24,160 - Epoch: [65][ 1090/ 1236] Overall Loss 0.262877 Objective Loss 0.262877 LR 0.001000 Time 0.021718 +2023-10-02 21:00:24,370 - Epoch: [65][ 1100/ 1236] Overall Loss 0.262986 Objective Loss 0.262986 LR 0.001000 Time 0.021711 +2023-10-02 21:00:24,579 - Epoch: [65][ 1110/ 1236] Overall Loss 0.263380 Objective Loss 0.263380 LR 0.001000 Time 0.021704 +2023-10-02 21:00:24,789 - Epoch: [65][ 1120/ 1236] Overall Loss 0.263351 Objective Loss 0.263351 LR 0.001000 Time 0.021697 +2023-10-02 21:00:24,998 - Epoch: [65][ 1130/ 1236] Overall Loss 0.263192 Objective Loss 0.263192 LR 0.001000 Time 0.021690 +2023-10-02 21:00:25,208 - Epoch: [65][ 1140/ 1236] Overall Loss 0.263349 Objective Loss 0.263349 LR 0.001000 Time 0.021683 +2023-10-02 21:00:25,417 - Epoch: [65][ 1150/ 1236] Overall Loss 0.263275 Objective Loss 0.263275 LR 0.001000 Time 0.021677 +2023-10-02 21:00:25,627 - Epoch: [65][ 1160/ 1236] Overall Loss 0.263058 Objective Loss 0.263058 LR 0.001000 Time 0.021670 +2023-10-02 21:00:25,836 - Epoch: [65][ 1170/ 1236] Overall Loss 0.262850 Objective Loss 0.262850 LR 0.001000 Time 0.021664 +2023-10-02 21:00:26,046 - Epoch: [65][ 1180/ 1236] Overall Loss 0.262953 Objective Loss 0.262953 LR 0.001000 Time 0.021657 +2023-10-02 21:00:26,255 - Epoch: [65][ 1190/ 1236] Overall Loss 0.262778 Objective Loss 0.262778 LR 0.001000 Time 0.021651 +2023-10-02 21:00:26,465 - Epoch: [65][ 1200/ 1236] Overall Loss 0.263014 Objective Loss 0.263014 LR 0.001000 Time 0.021645 +2023-10-02 21:00:26,674 - Epoch: [65][ 1210/ 1236] Overall Loss 0.262984 Objective Loss 0.262984 LR 0.001000 Time 0.021639 +2023-10-02 21:00:26,884 - Epoch: [65][ 1220/ 1236] Overall Loss 0.262621 Objective Loss 0.262621 LR 0.001000 Time 0.021633 +2023-10-02 21:00:27,147 - Epoch: [65][ 1230/ 1236] Overall Loss 0.262853 Objective Loss 0.262853 LR 0.001000 Time 0.021671 +2023-10-02 21:00:27,269 - Epoch: [65][ 1236/ 1236] Overall Loss 0.262810 Objective Loss 0.262810 Top1 86.150713 Top5 98.778004 LR 0.001000 Time 0.021665 +2023-10-02 21:00:27,410 - --- validate (epoch=65)----------- +2023-10-02 21:00:27,410 - 29943 samples (256 per mini-batch) +2023-10-02 21:00:27,895 - Epoch: [65][ 10/ 117] Loss 0.329781 Top1 83.320312 Top5 98.437500 +2023-10-02 21:00:28,046 - Epoch: [65][ 20/ 117] Loss 0.346074 Top1 83.300781 Top5 98.085938 +2023-10-02 21:00:28,198 - Epoch: [65][ 30/ 117] Loss 0.345264 Top1 83.515625 Top5 98.125000 +2023-10-02 21:00:28,349 - Epoch: [65][ 40/ 117] Loss 0.347226 Top1 83.535156 Top5 98.203125 +2023-10-02 21:00:28,508 - Epoch: [65][ 50/ 117] Loss 0.343972 Top1 83.601562 Top5 98.226562 +2023-10-02 21:00:28,670 - Epoch: [65][ 60/ 117] Loss 0.343466 Top1 83.815104 Top5 98.222656 +2023-10-02 21:00:28,830 - Epoch: [65][ 70/ 117] Loss 0.344405 Top1 83.872768 Top5 98.247768 +2023-10-02 21:00:28,993 - Epoch: [65][ 80/ 117] Loss 0.344111 Top1 83.764648 Top5 98.242188 +2023-10-02 21:00:29,150 - Epoch: [65][ 90/ 117] Loss 0.344324 Top1 83.654514 Top5 98.233507 +2023-10-02 21:00:29,310 - Epoch: [65][ 100/ 117] Loss 0.349724 Top1 83.503906 Top5 98.207031 +2023-10-02 21:00:29,477 - Epoch: [65][ 110/ 117] Loss 0.348278 Top1 83.533381 Top5 98.227983 +2023-10-02 21:00:29,567 - Epoch: [65][ 117/ 117] Loss 0.348824 Top1 83.501987 Top5 98.209932 +2023-10-02 21:00:29,710 - ==> Top1: 83.502 Top5: 98.210 Loss: 0.349 + +2023-10-02 21:00:29,711 - ==> Confusion: +[[ 951 1 2 2 4 3 0 0 10 45 1 1 0 3 9 1 2 2 0 0 13] + [ 1 1003 0 2 6 39 1 19 3 1 3 1 0 0 2 5 11 0 24 2 8] + [ 2 0 979 10 0 0 23 4 0 1 1 1 7 1 1 3 3 2 8 4 6] + [ 2 0 15 980 0 3 1 1 0 0 1 0 12 1 18 0 2 8 22 1 22] + [ 31 4 2 0 947 6 0 0 0 13 2 1 1 7 7 5 13 0 1 3 7] + [ 4 33 0 1 2 968 2 20 4 3 1 13 2 20 7 2 3 2 6 7 16] + [ 0 4 44 2 0 1 1115 3 0 0 2 2 0 0 0 2 0 2 0 6 8] + [ 4 18 38 0 7 30 3 1019 4 3 2 9 5 4 1 2 0 1 48 12 8] + [ 19 4 2 1 2 1 0 0 978 29 8 3 3 9 21 0 5 2 0 0 2] + [ 138 1 2 2 2 1 0 0 46 864 1 0 0 30 16 1 0 5 0 0 10] + [ 4 0 16 11 0 0 6 2 24 1 936 3 0 22 5 0 0 3 10 2 8] + [ 2 0 1 1 1 8 0 5 0 1 0 961 15 11 0 2 1 15 0 6 5] + [ 0 0 2 6 1 4 1 1 3 0 0 57 941 2 0 5 0 23 4 5 13] + [ 1 0 2 0 0 6 0 0 16 13 3 11 0 1043 7 0 2 1 0 1 13] + [ 7 0 5 33 0 0 0 0 22 2 1 0 4 4 1008 0 0 2 6 0 7] + [ 1 0 0 2 6 0 4 1 0 0 0 9 12 1 0 1051 10 20 1 3 13] + [ 3 6 4 1 4 3 0 0 2 0 0 7 2 1 2 10 1095 1 0 5 15] + [ 0 0 0 3 0 0 2 0 0 0 0 6 21 0 2 4 3 992 0 1 4] + [ 1 1 8 15 0 0 1 14 3 0 1 1 4 0 18 0 0 0 983 3 15] + [ 0 3 5 3 0 3 12 6 0 0 1 24 4 2 0 4 14 1 5 1051 14] + [ 187 124 173 124 72 161 35 80 116 56 138 144 325 298 163 41 150 96 131 153 5138]] + +2023-10-02 21:00:29,712 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:00:29,712 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:00:29,718 - + +2023-10-02 21:00:29,718 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:00:30,736 - Epoch: [66][ 10/ 1236] Overall Loss 0.268913 Objective Loss 0.268913 LR 0.001000 Time 0.101774 +2023-10-02 21:00:30,944 - Epoch: [66][ 20/ 1236] Overall Loss 0.266466 Objective Loss 0.266466 LR 0.001000 Time 0.061268 +2023-10-02 21:00:31,152 - Epoch: [66][ 30/ 1236] Overall Loss 0.252273 Objective Loss 0.252273 LR 0.001000 Time 0.047760 +2023-10-02 21:00:31,362 - Epoch: [66][ 40/ 1236] Overall Loss 0.250244 Objective Loss 0.250244 LR 0.001000 Time 0.041047 +2023-10-02 21:00:31,570 - Epoch: [66][ 50/ 1236] Overall Loss 0.250008 Objective Loss 0.250008 LR 0.001000 Time 0.036972 +2023-10-02 21:00:31,779 - Epoch: [66][ 60/ 1236] Overall Loss 0.253494 Objective Loss 0.253494 LR 0.001000 Time 0.034285 +2023-10-02 21:00:31,987 - Epoch: [66][ 70/ 1236] Overall Loss 0.253224 Objective Loss 0.253224 LR 0.001000 Time 0.032339 +2023-10-02 21:00:32,196 - Epoch: [66][ 80/ 1236] Overall Loss 0.253040 Objective Loss 0.253040 LR 0.001000 Time 0.030907 +2023-10-02 21:00:32,404 - Epoch: [66][ 90/ 1236] Overall Loss 0.255581 Objective Loss 0.255581 LR 0.001000 Time 0.029770 +2023-10-02 21:00:32,613 - Epoch: [66][ 100/ 1236] Overall Loss 0.252169 Objective Loss 0.252169 LR 0.001000 Time 0.028884 +2023-10-02 21:00:32,820 - Epoch: [66][ 110/ 1236] Overall Loss 0.250428 Objective Loss 0.250428 LR 0.001000 Time 0.028126 +2023-10-02 21:00:33,030 - Epoch: [66][ 120/ 1236] Overall Loss 0.250996 Objective Loss 0.250996 LR 0.001000 Time 0.027523 +2023-10-02 21:00:33,238 - Epoch: [66][ 130/ 1236] Overall Loss 0.247989 Objective Loss 0.247989 LR 0.001000 Time 0.026996 +2023-10-02 21:00:33,447 - Epoch: [66][ 140/ 1236] Overall Loss 0.247290 Objective Loss 0.247290 LR 0.001000 Time 0.026560 +2023-10-02 21:00:33,655 - Epoch: [66][ 150/ 1236] Overall Loss 0.249654 Objective Loss 0.249654 LR 0.001000 Time 0.026166 +2023-10-02 21:00:33,867 - Epoch: [66][ 160/ 1236] Overall Loss 0.248723 Objective Loss 0.248723 LR 0.001000 Time 0.025852 +2023-10-02 21:00:34,074 - Epoch: [66][ 170/ 1236] Overall Loss 0.250349 Objective Loss 0.250349 LR 0.001000 Time 0.025552 +2023-10-02 21:00:34,285 - Epoch: [66][ 180/ 1236] Overall Loss 0.251821 Objective Loss 0.251821 LR 0.001000 Time 0.025299 +2023-10-02 21:00:34,494 - Epoch: [66][ 190/ 1236] Overall Loss 0.253485 Objective Loss 0.253485 LR 0.001000 Time 0.025063 +2023-10-02 21:00:34,705 - Epoch: [66][ 200/ 1236] Overall Loss 0.254784 Objective Loss 0.254784 LR 0.001000 Time 0.024860 +2023-10-02 21:00:34,914 - Epoch: [66][ 210/ 1236] Overall Loss 0.255355 Objective Loss 0.255355 LR 0.001000 Time 0.024671 +2023-10-02 21:00:35,126 - Epoch: [66][ 220/ 1236] Overall Loss 0.255885 Objective Loss 0.255885 LR 0.001000 Time 0.024506 +2023-10-02 21:00:35,334 - Epoch: [66][ 230/ 1236] Overall Loss 0.256343 Objective Loss 0.256343 LR 0.001000 Time 0.024344 +2023-10-02 21:00:35,546 - Epoch: [66][ 240/ 1236] Overall Loss 0.256211 Objective Loss 0.256211 LR 0.001000 Time 0.024210 +2023-10-02 21:00:35,753 - Epoch: [66][ 250/ 1236] Overall Loss 0.254491 Objective Loss 0.254491 LR 0.001000 Time 0.024071 +2023-10-02 21:00:35,965 - Epoch: [66][ 260/ 1236] Overall Loss 0.254458 Objective Loss 0.254458 LR 0.001000 Time 0.023958 +2023-10-02 21:00:36,173 - Epoch: [66][ 270/ 1236] Overall Loss 0.255238 Objective Loss 0.255238 LR 0.001000 Time 0.023841 +2023-10-02 21:00:36,385 - Epoch: [66][ 280/ 1236] Overall Loss 0.255440 Objective Loss 0.255440 LR 0.001000 Time 0.023744 +2023-10-02 21:00:36,593 - Epoch: [66][ 290/ 1236] Overall Loss 0.255533 Objective Loss 0.255533 LR 0.001000 Time 0.023643 +2023-10-02 21:00:36,805 - Epoch: [66][ 300/ 1236] Overall Loss 0.255449 Objective Loss 0.255449 LR 0.001000 Time 0.023559 +2023-10-02 21:00:37,013 - Epoch: [66][ 310/ 1236] Overall Loss 0.256481 Objective Loss 0.256481 LR 0.001000 Time 0.023469 +2023-10-02 21:00:37,222 - Epoch: [66][ 320/ 1236] Overall Loss 0.256764 Objective Loss 0.256764 LR 0.001000 Time 0.023390 +2023-10-02 21:00:37,430 - Epoch: [66][ 330/ 1236] Overall Loss 0.257255 Objective Loss 0.257255 LR 0.001000 Time 0.023307 +2023-10-02 21:00:37,638 - Epoch: [66][ 340/ 1236] Overall Loss 0.256117 Objective Loss 0.256117 LR 0.001000 Time 0.023232 +2023-10-02 21:00:37,846 - Epoch: [66][ 350/ 1236] Overall Loss 0.256207 Objective Loss 0.256207 LR 0.001000 Time 0.023158 +2023-10-02 21:00:38,055 - Epoch: [66][ 360/ 1236] Overall Loss 0.256789 Objective Loss 0.256789 LR 0.001000 Time 0.023094 +2023-10-02 21:00:38,263 - Epoch: [66][ 370/ 1236] Overall Loss 0.257650 Objective Loss 0.257650 LR 0.001000 Time 0.023027 +2023-10-02 21:00:38,472 - Epoch: [66][ 380/ 1236] Overall Loss 0.258138 Objective Loss 0.258138 LR 0.001000 Time 0.022971 +2023-10-02 21:00:38,680 - Epoch: [66][ 390/ 1236] Overall Loss 0.258362 Objective Loss 0.258362 LR 0.001000 Time 0.022913 +2023-10-02 21:00:38,889 - Epoch: [66][ 400/ 1236] Overall Loss 0.257960 Objective Loss 0.257960 LR 0.001000 Time 0.022861 +2023-10-02 21:00:39,098 - Epoch: [66][ 410/ 1236] Overall Loss 0.258112 Objective Loss 0.258112 LR 0.001000 Time 0.022810 +2023-10-02 21:00:39,308 - Epoch: [66][ 420/ 1236] Overall Loss 0.258144 Objective Loss 0.258144 LR 0.001000 Time 0.022765 +2023-10-02 21:00:39,516 - Epoch: [66][ 430/ 1236] Overall Loss 0.258408 Objective Loss 0.258408 LR 0.001000 Time 0.022717 +2023-10-02 21:00:39,724 - Epoch: [66][ 440/ 1236] Overall Loss 0.259088 Objective Loss 0.259088 LR 0.001000 Time 0.022673 +2023-10-02 21:00:39,932 - Epoch: [66][ 450/ 1236] Overall Loss 0.258644 Objective Loss 0.258644 LR 0.001000 Time 0.022630 +2023-10-02 21:00:40,142 - Epoch: [66][ 460/ 1236] Overall Loss 0.258659 Objective Loss 0.258659 LR 0.001000 Time 0.022593 +2023-10-02 21:00:40,350 - Epoch: [66][ 470/ 1236] Overall Loss 0.258802 Objective Loss 0.258802 LR 0.001000 Time 0.022551 +2023-10-02 21:00:40,559 - Epoch: [66][ 480/ 1236] Overall Loss 0.258437 Objective Loss 0.258437 LR 0.001000 Time 0.022518 +2023-10-02 21:00:40,767 - Epoch: [66][ 490/ 1236] Overall Loss 0.257943 Objective Loss 0.257943 LR 0.001000 Time 0.022480 +2023-10-02 21:00:40,976 - Epoch: [66][ 500/ 1236] Overall Loss 0.257877 Objective Loss 0.257877 LR 0.001000 Time 0.022446 +2023-10-02 21:00:41,184 - Epoch: [66][ 510/ 1236] Overall Loss 0.258180 Objective Loss 0.258180 LR 0.001000 Time 0.022412 +2023-10-02 21:00:41,393 - Epoch: [66][ 520/ 1236] Overall Loss 0.258123 Objective Loss 0.258123 LR 0.001000 Time 0.022381 +2023-10-02 21:00:41,601 - Epoch: [66][ 530/ 1236] Overall Loss 0.258076 Objective Loss 0.258076 LR 0.001000 Time 0.022351 +2023-10-02 21:00:41,810 - Epoch: [66][ 540/ 1236] Overall Loss 0.258632 Objective Loss 0.258632 LR 0.001000 Time 0.022324 +2023-10-02 21:00:42,019 - Epoch: [66][ 550/ 1236] Overall Loss 0.258241 Objective Loss 0.258241 LR 0.001000 Time 0.022295 +2023-10-02 21:00:42,231 - Epoch: [66][ 560/ 1236] Overall Loss 0.258083 Objective Loss 0.258083 LR 0.001000 Time 0.022275 +2023-10-02 21:00:42,439 - Epoch: [66][ 570/ 1236] Overall Loss 0.258021 Objective Loss 0.258021 LR 0.001000 Time 0.022248 +2023-10-02 21:00:42,651 - Epoch: [66][ 580/ 1236] Overall Loss 0.258952 Objective Loss 0.258952 LR 0.001000 Time 0.022230 +2023-10-02 21:00:42,859 - Epoch: [66][ 590/ 1236] Overall Loss 0.258982 Objective Loss 0.258982 LR 0.001000 Time 0.022205 +2023-10-02 21:00:43,071 - Epoch: [66][ 600/ 1236] Overall Loss 0.259505 Objective Loss 0.259505 LR 0.001000 Time 0.022188 +2023-10-02 21:00:43,279 - Epoch: [66][ 610/ 1236] Overall Loss 0.259438 Objective Loss 0.259438 LR 0.001000 Time 0.022165 +2023-10-02 21:00:43,489 - Epoch: [66][ 620/ 1236] Overall Loss 0.258721 Objective Loss 0.258721 LR 0.001000 Time 0.022147 +2023-10-02 21:00:43,699 - Epoch: [66][ 630/ 1236] Overall Loss 0.259105 Objective Loss 0.259105 LR 0.001000 Time 0.022127 +2023-10-02 21:00:43,911 - Epoch: [66][ 640/ 1236] Overall Loss 0.258930 Objective Loss 0.258930 LR 0.001000 Time 0.022112 +2023-10-02 21:00:44,119 - Epoch: [66][ 650/ 1236] Overall Loss 0.259325 Objective Loss 0.259325 LR 0.001000 Time 0.022091 +2023-10-02 21:00:44,331 - Epoch: [66][ 660/ 1236] Overall Loss 0.259221 Objective Loss 0.259221 LR 0.001000 Time 0.022077 +2023-10-02 21:00:44,539 - Epoch: [66][ 670/ 1236] Overall Loss 0.259227 Objective Loss 0.259227 LR 0.001000 Time 0.022058 +2023-10-02 21:00:44,751 - Epoch: [66][ 680/ 1236] Overall Loss 0.259009 Objective Loss 0.259009 LR 0.001000 Time 0.022045 +2023-10-02 21:00:44,959 - Epoch: [66][ 690/ 1236] Overall Loss 0.259276 Objective Loss 0.259276 LR 0.001000 Time 0.022026 +2023-10-02 21:00:45,170 - Epoch: [66][ 700/ 1236] Overall Loss 0.259498 Objective Loss 0.259498 LR 0.001000 Time 0.022013 +2023-10-02 21:00:45,378 - Epoch: [66][ 710/ 1236] Overall Loss 0.259463 Objective Loss 0.259463 LR 0.001000 Time 0.021995 +2023-10-02 21:00:45,590 - Epoch: [66][ 720/ 1236] Overall Loss 0.259586 Objective Loss 0.259586 LR 0.001000 Time 0.021984 +2023-10-02 21:00:45,798 - Epoch: [66][ 730/ 1236] Overall Loss 0.259551 Objective Loss 0.259551 LR 0.001000 Time 0.021967 +2023-10-02 21:00:46,010 - Epoch: [66][ 740/ 1236] Overall Loss 0.259898 Objective Loss 0.259898 LR 0.001000 Time 0.021956 +2023-10-02 21:00:46,218 - Epoch: [66][ 750/ 1236] Overall Loss 0.260094 Objective Loss 0.260094 LR 0.001000 Time 0.021940 +2023-10-02 21:00:46,429 - Epoch: [66][ 760/ 1236] Overall Loss 0.259988 Objective Loss 0.259988 LR 0.001000 Time 0.021930 +2023-10-02 21:00:46,638 - Epoch: [66][ 770/ 1236] Overall Loss 0.259898 Objective Loss 0.259898 LR 0.001000 Time 0.021916 +2023-10-02 21:00:46,850 - Epoch: [66][ 780/ 1236] Overall Loss 0.259991 Objective Loss 0.259991 LR 0.001000 Time 0.021906 +2023-10-02 21:00:47,058 - Epoch: [66][ 790/ 1236] Overall Loss 0.260056 Objective Loss 0.260056 LR 0.001000 Time 0.021891 +2023-10-02 21:00:47,270 - Epoch: [66][ 800/ 1236] Overall Loss 0.259742 Objective Loss 0.259742 LR 0.001000 Time 0.021882 +2023-10-02 21:00:47,478 - Epoch: [66][ 810/ 1236] Overall Loss 0.259554 Objective Loss 0.259554 LR 0.001000 Time 0.021869 +2023-10-02 21:00:47,688 - Epoch: [66][ 820/ 1236] Overall Loss 0.259579 Objective Loss 0.259579 LR 0.001000 Time 0.021858 +2023-10-02 21:00:47,897 - Epoch: [66][ 830/ 1236] Overall Loss 0.259830 Objective Loss 0.259830 LR 0.001000 Time 0.021845 +2023-10-02 21:00:48,109 - Epoch: [66][ 840/ 1236] Overall Loss 0.260003 Objective Loss 0.260003 LR 0.001000 Time 0.021837 +2023-10-02 21:00:48,317 - Epoch: [66][ 850/ 1236] Overall Loss 0.259837 Objective Loss 0.259837 LR 0.001000 Time 0.021824 +2023-10-02 21:00:48,529 - Epoch: [66][ 860/ 1236] Overall Loss 0.259831 Objective Loss 0.259831 LR 0.001000 Time 0.021816 +2023-10-02 21:00:48,737 - Epoch: [66][ 870/ 1236] Overall Loss 0.259662 Objective Loss 0.259662 LR 0.001000 Time 0.021804 +2023-10-02 21:00:48,949 - Epoch: [66][ 880/ 1236] Overall Loss 0.260157 Objective Loss 0.260157 LR 0.001000 Time 0.021797 +2023-10-02 21:00:49,157 - Epoch: [66][ 890/ 1236] Overall Loss 0.260378 Objective Loss 0.260378 LR 0.001000 Time 0.021785 +2023-10-02 21:00:49,367 - Epoch: [66][ 900/ 1236] Overall Loss 0.260412 Objective Loss 0.260412 LR 0.001000 Time 0.021777 +2023-10-02 21:00:49,576 - Epoch: [66][ 910/ 1236] Overall Loss 0.260692 Objective Loss 0.260692 LR 0.001000 Time 0.021766 +2023-10-02 21:00:49,788 - Epoch: [66][ 920/ 1236] Overall Loss 0.260597 Objective Loss 0.260597 LR 0.001000 Time 0.021760 +2023-10-02 21:00:49,996 - Epoch: [66][ 930/ 1236] Overall Loss 0.260589 Objective Loss 0.260589 LR 0.001000 Time 0.021749 +2023-10-02 21:00:50,209 - Epoch: [66][ 940/ 1236] Overall Loss 0.260773 Objective Loss 0.260773 LR 0.001000 Time 0.021743 +2023-10-02 21:00:50,418 - Epoch: [66][ 950/ 1236] Overall Loss 0.260850 Objective Loss 0.260850 LR 0.001000 Time 0.021734 +2023-10-02 21:00:50,630 - Epoch: [66][ 960/ 1236] Overall Loss 0.260901 Objective Loss 0.260901 LR 0.001000 Time 0.021728 +2023-10-02 21:00:50,839 - Epoch: [66][ 970/ 1236] Overall Loss 0.261064 Objective Loss 0.261064 LR 0.001000 Time 0.021719 +2023-10-02 21:00:51,050 - Epoch: [66][ 980/ 1236] Overall Loss 0.261110 Objective Loss 0.261110 LR 0.001000 Time 0.021713 +2023-10-02 21:00:51,260 - Epoch: [66][ 990/ 1236] Overall Loss 0.261344 Objective Loss 0.261344 LR 0.001000 Time 0.021705 +2023-10-02 21:00:51,473 - Epoch: [66][ 1000/ 1236] Overall Loss 0.260954 Objective Loss 0.260954 LR 0.001000 Time 0.021701 +2023-10-02 21:00:51,682 - Epoch: [66][ 1010/ 1236] Overall Loss 0.261024 Objective Loss 0.261024 LR 0.001000 Time 0.021693 +2023-10-02 21:00:51,893 - Epoch: [66][ 1020/ 1236] Overall Loss 0.261480 Objective Loss 0.261480 LR 0.001000 Time 0.021687 +2023-10-02 21:00:52,103 - Epoch: [66][ 1030/ 1236] Overall Loss 0.261452 Objective Loss 0.261452 LR 0.001000 Time 0.021679 +2023-10-02 21:00:52,314 - Epoch: [66][ 1040/ 1236] Overall Loss 0.261506 Objective Loss 0.261506 LR 0.001000 Time 0.021674 +2023-10-02 21:00:52,524 - Epoch: [66][ 1050/ 1236] Overall Loss 0.261685 Objective Loss 0.261685 LR 0.001000 Time 0.021666 +2023-10-02 21:00:52,737 - Epoch: [66][ 1060/ 1236] Overall Loss 0.261346 Objective Loss 0.261346 LR 0.001000 Time 0.021662 +2023-10-02 21:00:52,946 - Epoch: [66][ 1070/ 1236] Overall Loss 0.261249 Objective Loss 0.261249 LR 0.001000 Time 0.021654 +2023-10-02 21:00:53,157 - Epoch: [66][ 1080/ 1236] Overall Loss 0.261276 Objective Loss 0.261276 LR 0.001000 Time 0.021649 +2023-10-02 21:00:53,367 - Epoch: [66][ 1090/ 1236] Overall Loss 0.261177 Objective Loss 0.261177 LR 0.001000 Time 0.021643 +2023-10-02 21:00:53,579 - Epoch: [66][ 1100/ 1236] Overall Loss 0.261467 Objective Loss 0.261467 LR 0.001000 Time 0.021639 +2023-10-02 21:00:53,789 - Epoch: [66][ 1110/ 1236] Overall Loss 0.261528 Objective Loss 0.261528 LR 0.001000 Time 0.021633 +2023-10-02 21:00:54,000 - Epoch: [66][ 1120/ 1236] Overall Loss 0.261476 Objective Loss 0.261476 LR 0.001000 Time 0.021628 +2023-10-02 21:00:54,210 - Epoch: [66][ 1130/ 1236] Overall Loss 0.261502 Objective Loss 0.261502 LR 0.001000 Time 0.021622 +2023-10-02 21:00:54,423 - Epoch: [66][ 1140/ 1236] Overall Loss 0.261657 Objective Loss 0.261657 LR 0.001000 Time 0.021618 +2023-10-02 21:00:54,632 - Epoch: [66][ 1150/ 1236] Overall Loss 0.261793 Objective Loss 0.261793 LR 0.001000 Time 0.021612 +2023-10-02 21:00:54,843 - Epoch: [66][ 1160/ 1236] Overall Loss 0.261818 Objective Loss 0.261818 LR 0.001000 Time 0.021607 +2023-10-02 21:00:55,053 - Epoch: [66][ 1170/ 1236] Overall Loss 0.261724 Objective Loss 0.261724 LR 0.001000 Time 0.021602 +2023-10-02 21:00:55,265 - Epoch: [66][ 1180/ 1236] Overall Loss 0.261897 Objective Loss 0.261897 LR 0.001000 Time 0.021598 +2023-10-02 21:00:55,474 - Epoch: [66][ 1190/ 1236] Overall Loss 0.261841 Objective Loss 0.261841 LR 0.001000 Time 0.021592 +2023-10-02 21:00:55,686 - Epoch: [66][ 1200/ 1236] Overall Loss 0.261715 Objective Loss 0.261715 LR 0.001000 Time 0.021589 +2023-10-02 21:00:55,896 - Epoch: [66][ 1210/ 1236] Overall Loss 0.261618 Objective Loss 0.261618 LR 0.001000 Time 0.021583 +2023-10-02 21:00:56,107 - Epoch: [66][ 1220/ 1236] Overall Loss 0.261572 Objective Loss 0.261572 LR 0.001000 Time 0.021579 +2023-10-02 21:00:56,371 - Epoch: [66][ 1230/ 1236] Overall Loss 0.261548 Objective Loss 0.261548 LR 0.001000 Time 0.021618 +2023-10-02 21:00:56,495 - Epoch: [66][ 1236/ 1236] Overall Loss 0.261590 Objective Loss 0.261590 Top1 84.317719 Top5 98.167006 LR 0.001000 Time 0.021613 +2023-10-02 21:00:56,636 - --- validate (epoch=66)----------- +2023-10-02 21:00:56,637 - 29943 samples (256 per mini-batch) +2023-10-02 21:00:57,135 - Epoch: [66][ 10/ 117] Loss 0.339346 Top1 82.890625 Top5 98.046875 +2023-10-02 21:00:57,291 - Epoch: [66][ 20/ 117] Loss 0.343360 Top1 82.832031 Top5 98.261719 +2023-10-02 21:00:57,445 - Epoch: [66][ 30/ 117] Loss 0.339901 Top1 83.007812 Top5 98.125000 +2023-10-02 21:00:57,601 - Epoch: [66][ 40/ 117] Loss 0.346282 Top1 83.251953 Top5 98.056641 +2023-10-02 21:00:57,755 - Epoch: [66][ 50/ 117] Loss 0.346906 Top1 83.187500 Top5 97.968750 +2023-10-02 21:00:57,911 - Epoch: [66][ 60/ 117] Loss 0.337104 Top1 83.242188 Top5 98.001302 +2023-10-02 21:00:58,065 - Epoch: [66][ 70/ 117] Loss 0.327781 Top1 83.353795 Top5 98.063616 +2023-10-02 21:00:58,219 - Epoch: [66][ 80/ 117] Loss 0.323563 Top1 83.525391 Top5 98.046875 +2023-10-02 21:00:58,370 - Epoch: [66][ 90/ 117] Loss 0.325700 Top1 83.641493 Top5 98.098958 +2023-10-02 21:00:58,524 - Epoch: [66][ 100/ 117] Loss 0.326185 Top1 83.792969 Top5 98.093750 +2023-10-02 21:00:58,682 - Epoch: [66][ 110/ 117] Loss 0.326519 Top1 83.817472 Top5 98.132102 +2023-10-02 21:00:58,772 - Epoch: [66][ 117/ 117] Loss 0.326295 Top1 83.845974 Top5 98.106402 +2023-10-02 21:00:58,919 - ==> Top1: 83.846 Top5: 98.106 Loss: 0.326 + +2023-10-02 21:00:58,920 - ==> Confusion: +[[ 890 2 2 0 13 4 0 5 7 101 2 0 0 1 2 2 5 3 1 0 10] + [ 1 1052 1 0 6 17 4 24 1 1 2 1 0 0 0 4 1 0 7 5 4] + [ 5 0 954 14 2 0 31 11 0 0 4 1 4 0 0 7 0 1 8 6 8] + [ 1 3 15 976 0 2 2 5 3 3 6 0 13 3 18 1 0 5 15 2 16] + [ 21 10 2 0 965 3 1 0 2 16 0 0 0 1 4 8 8 0 1 3 5] + [ 2 46 1 1 5 951 0 36 1 7 5 4 8 15 8 0 2 0 5 5 14] + [ 1 4 19 0 0 1 1126 7 0 0 5 3 1 0 0 9 0 1 0 6 8] + [ 2 18 17 0 4 13 6 1074 0 3 5 8 5 1 2 1 2 2 34 12 9] + [ 16 7 0 1 2 1 0 2 964 46 12 1 0 8 14 2 4 2 3 1 3] + [ 66 0 1 0 6 1 1 0 22 993 3 1 0 14 2 0 0 1 2 1 5] + [ 1 3 10 5 0 0 7 3 7 3 974 1 0 12 6 0 1 3 4 3 10] + [ 1 0 4 0 1 6 0 5 0 4 0 905 54 10 0 3 1 24 0 12 5] + [ 0 0 5 1 2 0 2 1 0 1 3 24 967 2 0 8 3 26 3 9 11] + [ 0 0 1 0 7 9 2 3 16 17 6 2 1 1031 7 0 1 2 0 3 11] + [ 8 1 3 19 9 0 0 0 18 10 3 1 4 2 998 0 0 7 5 0 13] + [ 0 0 1 1 5 1 1 0 0 0 0 4 9 1 0 1073 9 14 3 6 6] + [ 0 11 0 0 11 6 1 2 2 0 2 5 3 0 2 12 1079 1 0 12 12] + [ 1 0 0 2 0 0 3 0 0 1 0 2 21 1 1 5 0 998 1 0 2] + [ 0 5 5 14 2 0 2 19 3 1 6 1 1 0 10 0 0 0 986 2 11] + [ 0 2 1 2 2 0 12 11 0 1 2 8 2 1 3 2 4 0 2 1088 9] + [ 127 195 147 74 123 111 65 132 103 170 183 97 308 296 104 77 76 79 155 221 5062]] + +2023-10-02 21:00:58,921 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:00:58,921 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:00:58,928 - + +2023-10-02 21:00:58,928 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:01:00,088 - Epoch: [67][ 10/ 1236] Overall Loss 0.254180 Objective Loss 0.254180 LR 0.001000 Time 0.115943 +2023-10-02 21:01:00,297 - Epoch: [67][ 20/ 1236] Overall Loss 0.270272 Objective Loss 0.270272 LR 0.001000 Time 0.068429 +2023-10-02 21:01:00,508 - Epoch: [67][ 30/ 1236] Overall Loss 0.253163 Objective Loss 0.253163 LR 0.001000 Time 0.052620 +2023-10-02 21:01:00,718 - Epoch: [67][ 40/ 1236] Overall Loss 0.251840 Objective Loss 0.251840 LR 0.001000 Time 0.044723 +2023-10-02 21:01:00,928 - Epoch: [67][ 50/ 1236] Overall Loss 0.252135 Objective Loss 0.252135 LR 0.001000 Time 0.039949 +2023-10-02 21:01:01,137 - Epoch: [67][ 60/ 1236] Overall Loss 0.251295 Objective Loss 0.251295 LR 0.001000 Time 0.036760 +2023-10-02 21:01:01,346 - Epoch: [67][ 70/ 1236] Overall Loss 0.249720 Objective Loss 0.249720 LR 0.001000 Time 0.034481 +2023-10-02 21:01:01,556 - Epoch: [67][ 80/ 1236] Overall Loss 0.246607 Objective Loss 0.246607 LR 0.001000 Time 0.032787 +2023-10-02 21:01:01,765 - Epoch: [67][ 90/ 1236] Overall Loss 0.245175 Objective Loss 0.245175 LR 0.001000 Time 0.031456 +2023-10-02 21:01:01,975 - Epoch: [67][ 100/ 1236] Overall Loss 0.245522 Objective Loss 0.245522 LR 0.001000 Time 0.030406 +2023-10-02 21:01:02,185 - Epoch: [67][ 110/ 1236] Overall Loss 0.248036 Objective Loss 0.248036 LR 0.001000 Time 0.029532 +2023-10-02 21:01:02,394 - Epoch: [67][ 120/ 1236] Overall Loss 0.250655 Objective Loss 0.250655 LR 0.001000 Time 0.028816 +2023-10-02 21:01:02,604 - Epoch: [67][ 130/ 1236] Overall Loss 0.250471 Objective Loss 0.250471 LR 0.001000 Time 0.028202 +2023-10-02 21:01:02,815 - Epoch: [67][ 140/ 1236] Overall Loss 0.248685 Objective Loss 0.248685 LR 0.001000 Time 0.027691 +2023-10-02 21:01:03,023 - Epoch: [67][ 150/ 1236] Overall Loss 0.249033 Objective Loss 0.249033 LR 0.001000 Time 0.027223 +2023-10-02 21:01:03,233 - Epoch: [67][ 160/ 1236] Overall Loss 0.248186 Objective Loss 0.248186 LR 0.001000 Time 0.026831 +2023-10-02 21:01:03,441 - Epoch: [67][ 170/ 1236] Overall Loss 0.248724 Objective Loss 0.248724 LR 0.001000 Time 0.026469 +2023-10-02 21:01:03,651 - Epoch: [67][ 180/ 1236] Overall Loss 0.250243 Objective Loss 0.250243 LR 0.001000 Time 0.026162 +2023-10-02 21:01:03,859 - Epoch: [67][ 190/ 1236] Overall Loss 0.250803 Objective Loss 0.250803 LR 0.001000 Time 0.025871 +2023-10-02 21:01:04,069 - Epoch: [67][ 200/ 1236] Overall Loss 0.249589 Objective Loss 0.249589 LR 0.001000 Time 0.025625 +2023-10-02 21:01:04,277 - Epoch: [67][ 210/ 1236] Overall Loss 0.249055 Objective Loss 0.249055 LR 0.001000 Time 0.025387 +2023-10-02 21:01:04,486 - Epoch: [67][ 220/ 1236] Overall Loss 0.248498 Objective Loss 0.248498 LR 0.001000 Time 0.025184 +2023-10-02 21:01:04,694 - Epoch: [67][ 230/ 1236] Overall Loss 0.247527 Objective Loss 0.247527 LR 0.001000 Time 0.024987 +2023-10-02 21:01:04,904 - Epoch: [67][ 240/ 1236] Overall Loss 0.248998 Objective Loss 0.248998 LR 0.001000 Time 0.024818 +2023-10-02 21:01:05,112 - Epoch: [67][ 250/ 1236] Overall Loss 0.248005 Objective Loss 0.248005 LR 0.001000 Time 0.024652 +2023-10-02 21:01:05,322 - Epoch: [67][ 260/ 1236] Overall Loss 0.247984 Objective Loss 0.247984 LR 0.001000 Time 0.024510 +2023-10-02 21:01:05,530 - Epoch: [67][ 270/ 1236] Overall Loss 0.248750 Objective Loss 0.248750 LR 0.001000 Time 0.024367 +2023-10-02 21:01:05,740 - Epoch: [67][ 280/ 1236] Overall Loss 0.249247 Objective Loss 0.249247 LR 0.001000 Time 0.024245 +2023-10-02 21:01:05,948 - Epoch: [67][ 290/ 1236] Overall Loss 0.250066 Objective Loss 0.250066 LR 0.001000 Time 0.024121 +2023-10-02 21:01:06,157 - Epoch: [67][ 300/ 1236] Overall Loss 0.249804 Objective Loss 0.249804 LR 0.001000 Time 0.024015 +2023-10-02 21:01:06,365 - Epoch: [67][ 310/ 1236] Overall Loss 0.250368 Objective Loss 0.250368 LR 0.001000 Time 0.023907 +2023-10-02 21:01:06,575 - Epoch: [67][ 320/ 1236] Overall Loss 0.250069 Objective Loss 0.250069 LR 0.001000 Time 0.023814 +2023-10-02 21:01:06,783 - Epoch: [67][ 330/ 1236] Overall Loss 0.250210 Objective Loss 0.250210 LR 0.001000 Time 0.023717 +2023-10-02 21:01:06,993 - Epoch: [67][ 340/ 1236] Overall Loss 0.250276 Objective Loss 0.250276 LR 0.001000 Time 0.023637 +2023-10-02 21:01:07,202 - Epoch: [67][ 350/ 1236] Overall Loss 0.251079 Objective Loss 0.251079 LR 0.001000 Time 0.023557 +2023-10-02 21:01:07,414 - Epoch: [67][ 360/ 1236] Overall Loss 0.250621 Objective Loss 0.250621 LR 0.001000 Time 0.023491 +2023-10-02 21:01:07,622 - Epoch: [67][ 370/ 1236] Overall Loss 0.250774 Objective Loss 0.250774 LR 0.001000 Time 0.023418 +2023-10-02 21:01:07,834 - Epoch: [67][ 380/ 1236] Overall Loss 0.251150 Objective Loss 0.251150 LR 0.001000 Time 0.023360 +2023-10-02 21:01:08,041 - Epoch: [67][ 390/ 1236] Overall Loss 0.252074 Objective Loss 0.252074 LR 0.001000 Time 0.023289 +2023-10-02 21:01:08,249 - Epoch: [67][ 400/ 1236] Overall Loss 0.252284 Objective Loss 0.252284 LR 0.001000 Time 0.023228 +2023-10-02 21:01:08,457 - Epoch: [67][ 410/ 1236] Overall Loss 0.251854 Objective Loss 0.251854 LR 0.001000 Time 0.023164 +2023-10-02 21:01:08,666 - Epoch: [67][ 420/ 1236] Overall Loss 0.252060 Objective Loss 0.252060 LR 0.001000 Time 0.023108 +2023-10-02 21:01:08,873 - Epoch: [67][ 430/ 1236] Overall Loss 0.252036 Objective Loss 0.252036 LR 0.001000 Time 0.023049 +2023-10-02 21:01:09,083 - Epoch: [67][ 440/ 1236] Overall Loss 0.252196 Objective Loss 0.252196 LR 0.001000 Time 0.023003 +2023-10-02 21:01:09,289 - Epoch: [67][ 450/ 1236] Overall Loss 0.252204 Objective Loss 0.252204 LR 0.001000 Time 0.022949 +2023-10-02 21:01:09,499 - Epoch: [67][ 460/ 1236] Overall Loss 0.252195 Objective Loss 0.252195 LR 0.001000 Time 0.022905 +2023-10-02 21:01:09,705 - Epoch: [67][ 470/ 1236] Overall Loss 0.252513 Objective Loss 0.252513 LR 0.001000 Time 0.022856 +2023-10-02 21:01:09,913 - Epoch: [67][ 480/ 1236] Overall Loss 0.252949 Objective Loss 0.252949 LR 0.001000 Time 0.022812 +2023-10-02 21:01:10,120 - Epoch: [67][ 490/ 1236] Overall Loss 0.253389 Objective Loss 0.253389 LR 0.001000 Time 0.022769 +2023-10-02 21:01:10,330 - Epoch: [67][ 500/ 1236] Overall Loss 0.253251 Objective Loss 0.253251 LR 0.001000 Time 0.022733 +2023-10-02 21:01:10,537 - Epoch: [67][ 510/ 1236] Overall Loss 0.253535 Objective Loss 0.253535 LR 0.001000 Time 0.022692 +2023-10-02 21:01:10,747 - Epoch: [67][ 520/ 1236] Overall Loss 0.253208 Objective Loss 0.253208 LR 0.001000 Time 0.022659 +2023-10-02 21:01:10,953 - Epoch: [67][ 530/ 1236] Overall Loss 0.253481 Objective Loss 0.253481 LR 0.001000 Time 0.022620 +2023-10-02 21:01:11,162 - Epoch: [67][ 540/ 1236] Overall Loss 0.253862 Objective Loss 0.253862 LR 0.001000 Time 0.022587 +2023-10-02 21:01:11,370 - Epoch: [67][ 550/ 1236] Overall Loss 0.253856 Objective Loss 0.253856 LR 0.001000 Time 0.022551 +2023-10-02 21:01:11,580 - Epoch: [67][ 560/ 1236] Overall Loss 0.254658 Objective Loss 0.254658 LR 0.001000 Time 0.022524 +2023-10-02 21:01:11,786 - Epoch: [67][ 570/ 1236] Overall Loss 0.255339 Objective Loss 0.255339 LR 0.001000 Time 0.022490 +2023-10-02 21:01:11,996 - Epoch: [67][ 580/ 1236] Overall Loss 0.255836 Objective Loss 0.255836 LR 0.001000 Time 0.022464 +2023-10-02 21:01:12,203 - Epoch: [67][ 590/ 1236] Overall Loss 0.256284 Objective Loss 0.256284 LR 0.001000 Time 0.022432 +2023-10-02 21:01:12,413 - Epoch: [67][ 600/ 1236] Overall Loss 0.256268 Objective Loss 0.256268 LR 0.001000 Time 0.022408 +2023-10-02 21:01:12,619 - Epoch: [67][ 610/ 1236] Overall Loss 0.256782 Objective Loss 0.256782 LR 0.001000 Time 0.022378 +2023-10-02 21:01:12,829 - Epoch: [67][ 620/ 1236] Overall Loss 0.256858 Objective Loss 0.256858 LR 0.001000 Time 0.022355 +2023-10-02 21:01:13,035 - Epoch: [67][ 630/ 1236] Overall Loss 0.256956 Objective Loss 0.256956 LR 0.001000 Time 0.022328 +2023-10-02 21:01:13,245 - Epoch: [67][ 640/ 1236] Overall Loss 0.256984 Objective Loss 0.256984 LR 0.001000 Time 0.022306 +2023-10-02 21:01:13,452 - Epoch: [67][ 650/ 1236] Overall Loss 0.257437 Objective Loss 0.257437 LR 0.001000 Time 0.022280 +2023-10-02 21:01:13,661 - Epoch: [67][ 660/ 1236] Overall Loss 0.257637 Objective Loss 0.257637 LR 0.001000 Time 0.022259 +2023-10-02 21:01:13,868 - Epoch: [67][ 670/ 1236] Overall Loss 0.258128 Objective Loss 0.258128 LR 0.001000 Time 0.022234 +2023-10-02 21:01:14,078 - Epoch: [67][ 680/ 1236] Overall Loss 0.258799 Objective Loss 0.258799 LR 0.001000 Time 0.022216 +2023-10-02 21:01:14,285 - Epoch: [67][ 690/ 1236] Overall Loss 0.259357 Objective Loss 0.259357 LR 0.001000 Time 0.022192 +2023-10-02 21:01:14,493 - Epoch: [67][ 700/ 1236] Overall Loss 0.258931 Objective Loss 0.258931 LR 0.001000 Time 0.022173 +2023-10-02 21:01:14,700 - Epoch: [67][ 710/ 1236] Overall Loss 0.259194 Objective Loss 0.259194 LR 0.001000 Time 0.022150 +2023-10-02 21:01:14,909 - Epoch: [67][ 720/ 1236] Overall Loss 0.259476 Objective Loss 0.259476 LR 0.001000 Time 0.022131 +2023-10-02 21:01:15,116 - Epoch: [67][ 730/ 1236] Overall Loss 0.259383 Objective Loss 0.259383 LR 0.001000 Time 0.022110 +2023-10-02 21:01:15,325 - Epoch: [67][ 740/ 1236] Overall Loss 0.259292 Objective Loss 0.259292 LR 0.001000 Time 0.022093 +2023-10-02 21:01:15,532 - Epoch: [67][ 750/ 1236] Overall Loss 0.259285 Objective Loss 0.259285 LR 0.001000 Time 0.022073 +2023-10-02 21:01:15,741 - Epoch: [67][ 760/ 1236] Overall Loss 0.259638 Objective Loss 0.259638 LR 0.001000 Time 0.022057 +2023-10-02 21:01:15,949 - Epoch: [67][ 770/ 1236] Overall Loss 0.259906 Objective Loss 0.259906 LR 0.001000 Time 0.022038 +2023-10-02 21:01:16,157 - Epoch: [67][ 780/ 1236] Overall Loss 0.259718 Objective Loss 0.259718 LR 0.001000 Time 0.022023 +2023-10-02 21:01:16,365 - Epoch: [67][ 790/ 1236] Overall Loss 0.260122 Objective Loss 0.260122 LR 0.001000 Time 0.022005 +2023-10-02 21:01:16,574 - Epoch: [67][ 800/ 1236] Overall Loss 0.260123 Objective Loss 0.260123 LR 0.001000 Time 0.021990 +2023-10-02 21:01:16,781 - Epoch: [67][ 810/ 1236] Overall Loss 0.260129 Objective Loss 0.260129 LR 0.001000 Time 0.021973 +2023-10-02 21:01:16,990 - Epoch: [67][ 820/ 1236] Overall Loss 0.260150 Objective Loss 0.260150 LR 0.001000 Time 0.021960 +2023-10-02 21:01:17,198 - Epoch: [67][ 830/ 1236] Overall Loss 0.260358 Objective Loss 0.260358 LR 0.001000 Time 0.021943 +2023-10-02 21:01:17,406 - Epoch: [67][ 840/ 1236] Overall Loss 0.260448 Objective Loss 0.260448 LR 0.001000 Time 0.021930 +2023-10-02 21:01:17,614 - Epoch: [67][ 850/ 1236] Overall Loss 0.260515 Objective Loss 0.260515 LR 0.001000 Time 0.021914 +2023-10-02 21:01:17,822 - Epoch: [67][ 860/ 1236] Overall Loss 0.260535 Objective Loss 0.260535 LR 0.001000 Time 0.021902 +2023-10-02 21:01:18,030 - Epoch: [67][ 870/ 1236] Overall Loss 0.260578 Objective Loss 0.260578 LR 0.001000 Time 0.021887 +2023-10-02 21:01:18,238 - Epoch: [67][ 880/ 1236] Overall Loss 0.260534 Objective Loss 0.260534 LR 0.001000 Time 0.021875 +2023-10-02 21:01:18,446 - Epoch: [67][ 890/ 1236] Overall Loss 0.261121 Objective Loss 0.261121 LR 0.001000 Time 0.021860 +2023-10-02 21:01:18,654 - Epoch: [67][ 900/ 1236] Overall Loss 0.260853 Objective Loss 0.260853 LR 0.001000 Time 0.021849 +2023-10-02 21:01:18,862 - Epoch: [67][ 910/ 1236] Overall Loss 0.260615 Objective Loss 0.260615 LR 0.001000 Time 0.021835 +2023-10-02 21:01:19,071 - Epoch: [67][ 920/ 1236] Overall Loss 0.260681 Objective Loss 0.260681 LR 0.001000 Time 0.021824 +2023-10-02 21:01:19,278 - Epoch: [67][ 930/ 1236] Overall Loss 0.260806 Objective Loss 0.260806 LR 0.001000 Time 0.021811 +2023-10-02 21:01:19,487 - Epoch: [67][ 940/ 1236] Overall Loss 0.260984 Objective Loss 0.260984 LR 0.001000 Time 0.021801 +2023-10-02 21:01:19,695 - Epoch: [67][ 950/ 1236] Overall Loss 0.260815 Objective Loss 0.260815 LR 0.001000 Time 0.021789 +2023-10-02 21:01:19,904 - Epoch: [67][ 960/ 1236] Overall Loss 0.260818 Objective Loss 0.260818 LR 0.001000 Time 0.021779 +2023-10-02 21:01:20,111 - Epoch: [67][ 970/ 1236] Overall Loss 0.260939 Objective Loss 0.260939 LR 0.001000 Time 0.021767 +2023-10-02 21:01:20,320 - Epoch: [67][ 980/ 1236] Overall Loss 0.260733 Objective Loss 0.260733 LR 0.001000 Time 0.021757 +2023-10-02 21:01:20,528 - Epoch: [67][ 990/ 1236] Overall Loss 0.261081 Objective Loss 0.261081 LR 0.001000 Time 0.021746 +2023-10-02 21:01:20,737 - Epoch: [67][ 1000/ 1236] Overall Loss 0.261091 Objective Loss 0.261091 LR 0.001000 Time 0.021737 +2023-10-02 21:01:20,944 - Epoch: [67][ 1010/ 1236] Overall Loss 0.261009 Objective Loss 0.261009 LR 0.001000 Time 0.021726 +2023-10-02 21:01:21,152 - Epoch: [67][ 1020/ 1236] Overall Loss 0.261053 Objective Loss 0.261053 LR 0.001000 Time 0.021716 +2023-10-02 21:01:21,360 - Epoch: [67][ 1030/ 1236] Overall Loss 0.261192 Objective Loss 0.261192 LR 0.001000 Time 0.021705 +2023-10-02 21:01:21,568 - Epoch: [67][ 1040/ 1236] Overall Loss 0.261297 Objective Loss 0.261297 LR 0.001000 Time 0.021697 +2023-10-02 21:01:21,776 - Epoch: [67][ 1050/ 1236] Overall Loss 0.261263 Objective Loss 0.261263 LR 0.001000 Time 0.021686 +2023-10-02 21:01:21,984 - Epoch: [67][ 1060/ 1236] Overall Loss 0.261488 Objective Loss 0.261488 LR 0.001000 Time 0.021678 +2023-10-02 21:01:22,192 - Epoch: [67][ 1070/ 1236] Overall Loss 0.261348 Objective Loss 0.261348 LR 0.001000 Time 0.021668 +2023-10-02 21:01:22,401 - Epoch: [67][ 1080/ 1236] Overall Loss 0.261332 Objective Loss 0.261332 LR 0.001000 Time 0.021661 +2023-10-02 21:01:22,608 - Epoch: [67][ 1090/ 1236] Overall Loss 0.261313 Objective Loss 0.261313 LR 0.001000 Time 0.021651 +2023-10-02 21:01:22,817 - Epoch: [67][ 1100/ 1236] Overall Loss 0.261169 Objective Loss 0.261169 LR 0.001000 Time 0.021644 +2023-10-02 21:01:23,022 - Epoch: [67][ 1110/ 1236] Overall Loss 0.261079 Objective Loss 0.261079 LR 0.001000 Time 0.021632 +2023-10-02 21:01:23,230 - Epoch: [67][ 1120/ 1236] Overall Loss 0.261030 Objective Loss 0.261030 LR 0.001000 Time 0.021625 +2023-10-02 21:01:23,436 - Epoch: [67][ 1130/ 1236] Overall Loss 0.260972 Objective Loss 0.260972 LR 0.001000 Time 0.021614 +2023-10-02 21:01:23,640 - Epoch: [67][ 1140/ 1236] Overall Loss 0.260843 Objective Loss 0.260843 LR 0.001000 Time 0.021603 +2023-10-02 21:01:23,844 - Epoch: [67][ 1150/ 1236] Overall Loss 0.261229 Objective Loss 0.261229 LR 0.001000 Time 0.021592 +2023-10-02 21:01:24,048 - Epoch: [67][ 1160/ 1236] Overall Loss 0.261270 Objective Loss 0.261270 LR 0.001000 Time 0.021582 +2023-10-02 21:01:24,252 - Epoch: [67][ 1170/ 1236] Overall Loss 0.261245 Objective Loss 0.261245 LR 0.001000 Time 0.021571 +2023-10-02 21:01:24,456 - Epoch: [67][ 1180/ 1236] Overall Loss 0.261481 Objective Loss 0.261481 LR 0.001000 Time 0.021561 +2023-10-02 21:01:24,660 - Epoch: [67][ 1190/ 1236] Overall Loss 0.261637 Objective Loss 0.261637 LR 0.001000 Time 0.021551 +2023-10-02 21:01:24,864 - Epoch: [67][ 1200/ 1236] Overall Loss 0.261579 Objective Loss 0.261579 LR 0.001000 Time 0.021541 +2023-10-02 21:01:25,067 - Epoch: [67][ 1210/ 1236] Overall Loss 0.261738 Objective Loss 0.261738 LR 0.001000 Time 0.021531 +2023-10-02 21:01:25,271 - Epoch: [67][ 1220/ 1236] Overall Loss 0.261822 Objective Loss 0.261822 LR 0.001000 Time 0.021522 +2023-10-02 21:01:25,529 - Epoch: [67][ 1230/ 1236] Overall Loss 0.261968 Objective Loss 0.261968 LR 0.001000 Time 0.021556 +2023-10-02 21:01:25,653 - Epoch: [67][ 1236/ 1236] Overall Loss 0.262119 Objective Loss 0.262119 Top1 84.317719 Top5 98.574338 LR 0.001000 Time 0.021551 +2023-10-02 21:01:25,786 - --- validate (epoch=67)----------- +2023-10-02 21:01:25,786 - 29943 samples (256 per mini-batch) +2023-10-02 21:01:26,279 - Epoch: [67][ 10/ 117] Loss 0.322784 Top1 83.593750 Top5 97.929688 +2023-10-02 21:01:26,430 - Epoch: [67][ 20/ 117] Loss 0.353960 Top1 83.476562 Top5 97.890625 +2023-10-02 21:01:26,581 - Epoch: [67][ 30/ 117] Loss 0.349587 Top1 82.825521 Top5 97.929688 +2023-10-02 21:01:26,733 - Epoch: [67][ 40/ 117] Loss 0.353088 Top1 82.978516 Top5 97.949219 +2023-10-02 21:01:26,885 - Epoch: [67][ 50/ 117] Loss 0.344278 Top1 83.250000 Top5 97.992188 +2023-10-02 21:01:27,037 - Epoch: [67][ 60/ 117] Loss 0.342732 Top1 83.261719 Top5 98.033854 +2023-10-02 21:01:27,188 - Epoch: [67][ 70/ 117] Loss 0.339496 Top1 83.097098 Top5 98.046875 +2023-10-02 21:01:27,340 - Epoch: [67][ 80/ 117] Loss 0.338990 Top1 83.242188 Top5 98.041992 +2023-10-02 21:01:27,492 - Epoch: [67][ 90/ 117] Loss 0.335834 Top1 83.289931 Top5 98.077257 +2023-10-02 21:01:27,643 - Epoch: [67][ 100/ 117] Loss 0.337729 Top1 83.171875 Top5 98.046875 +2023-10-02 21:01:27,802 - Epoch: [67][ 110/ 117] Loss 0.345523 Top1 82.975852 Top5 98.000710 +2023-10-02 21:01:27,892 - Epoch: [67][ 117/ 117] Loss 0.345219 Top1 83.011054 Top5 98.002872 +2023-10-02 21:01:28,040 - ==> Top1: 83.011 Top5: 98.003 Loss: 0.345 + +2023-10-02 21:01:28,040 - ==> Confusion: +[[ 913 1 10 0 12 3 0 1 3 70 1 1 0 5 6 6 3 2 0 0 13] + [ 1 1031 3 2 11 28 1 21 0 3 1 1 0 0 2 6 6 0 7 1 6] + [ 4 0 955 15 2 2 33 9 0 2 1 1 7 3 3 5 0 1 4 1 8] + [ 2 1 12 980 2 3 3 3 0 0 1 0 12 4 36 1 2 4 10 0 13] + [ 21 4 2 0 965 4 1 2 1 9 0 3 0 3 6 8 11 0 0 2 8] + [ 5 50 0 3 5 948 4 22 3 5 3 19 3 11 7 2 5 0 3 4 14] + [ 1 4 28 1 0 1 1123 7 0 0 1 1 0 0 0 7 0 1 1 9 6] + [ 2 20 24 1 3 17 6 1054 0 6 6 7 2 1 3 2 4 3 38 14 5] + [ 21 4 1 3 2 1 0 0 917 67 10 1 1 14 30 1 8 1 4 1 2] + [ 91 0 0 1 6 0 0 0 13 970 2 0 0 12 9 1 1 2 1 4 6] + [ 4 3 17 15 2 2 4 4 16 2 938 4 0 16 7 1 1 2 3 3 9] + [ 0 2 6 0 1 8 0 3 0 0 0 938 26 6 0 10 3 18 0 7 7] + [ 0 1 5 1 2 2 4 2 1 3 3 41 940 5 4 21 0 16 2 4 11] + [ 1 0 1 0 4 8 2 0 16 16 3 4 2 1033 7 1 3 1 0 4 13] + [ 14 0 3 19 7 0 0 0 12 3 0 1 4 4 1019 0 1 1 7 0 6] + [ 0 0 1 0 5 2 6 0 0 1 0 3 5 0 0 1082 9 8 2 4 6] + [ 2 4 2 0 5 10 0 1 1 0 1 4 1 1 4 14 1099 0 1 2 9] + [ 0 0 0 4 0 0 3 0 1 1 0 3 18 3 9 13 4 976 0 0 3] + [ 1 7 9 19 0 0 1 29 6 0 3 0 2 1 27 0 0 0 954 0 9] + [ 0 2 2 5 1 3 16 2 0 1 1 10 2 2 0 10 11 3 2 1072 7] + [ 170 194 166 107 113 138 61 104 67 109 166 104 330 272 196 101 109 53 158 238 4949]] + +2023-10-02 21:01:28,042 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:01:28,042 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:01:28,048 - + +2023-10-02 21:01:28,048 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:01:29,086 - Epoch: [68][ 10/ 1236] Overall Loss 0.229218 Objective Loss 0.229218 LR 0.001000 Time 0.103718 +2023-10-02 21:01:29,292 - Epoch: [68][ 20/ 1236] Overall Loss 0.240025 Objective Loss 0.240025 LR 0.001000 Time 0.062150 +2023-10-02 21:01:29,497 - Epoch: [68][ 30/ 1236] Overall Loss 0.244179 Objective Loss 0.244179 LR 0.001000 Time 0.048233 +2023-10-02 21:01:29,704 - Epoch: [68][ 40/ 1236] Overall Loss 0.248572 Objective Loss 0.248572 LR 0.001000 Time 0.041334 +2023-10-02 21:01:29,909 - Epoch: [68][ 50/ 1236] Overall Loss 0.252337 Objective Loss 0.252337 LR 0.001000 Time 0.037142 +2023-10-02 21:01:30,116 - Epoch: [68][ 60/ 1236] Overall Loss 0.258101 Objective Loss 0.258101 LR 0.001000 Time 0.034390 +2023-10-02 21:01:30,321 - Epoch: [68][ 70/ 1236] Overall Loss 0.258193 Objective Loss 0.258193 LR 0.001000 Time 0.032388 +2023-10-02 21:01:30,528 - Epoch: [68][ 80/ 1236] Overall Loss 0.258379 Objective Loss 0.258379 LR 0.001000 Time 0.030919 +2023-10-02 21:01:30,733 - Epoch: [68][ 90/ 1236] Overall Loss 0.259478 Objective Loss 0.259478 LR 0.001000 Time 0.029749 +2023-10-02 21:01:30,940 - Epoch: [68][ 100/ 1236] Overall Loss 0.262758 Objective Loss 0.262758 LR 0.001000 Time 0.028838 +2023-10-02 21:01:31,145 - Epoch: [68][ 110/ 1236] Overall Loss 0.261368 Objective Loss 0.261368 LR 0.001000 Time 0.028071 +2023-10-02 21:01:31,352 - Epoch: [68][ 120/ 1236] Overall Loss 0.260712 Objective Loss 0.260712 LR 0.001000 Time 0.027449 +2023-10-02 21:01:31,557 - Epoch: [68][ 130/ 1236] Overall Loss 0.259883 Objective Loss 0.259883 LR 0.001000 Time 0.026903 +2023-10-02 21:01:31,765 - Epoch: [68][ 140/ 1236] Overall Loss 0.260308 Objective Loss 0.260308 LR 0.001000 Time 0.026465 +2023-10-02 21:01:31,969 - Epoch: [68][ 150/ 1236] Overall Loss 0.260581 Objective Loss 0.260581 LR 0.001000 Time 0.026062 +2023-10-02 21:01:32,177 - Epoch: [68][ 160/ 1236] Overall Loss 0.260434 Objective Loss 0.260434 LR 0.001000 Time 0.025730 +2023-10-02 21:01:32,381 - Epoch: [68][ 170/ 1236] Overall Loss 0.261030 Objective Loss 0.261030 LR 0.001000 Time 0.025415 +2023-10-02 21:01:32,586 - Epoch: [68][ 180/ 1236] Overall Loss 0.260250 Objective Loss 0.260250 LR 0.001000 Time 0.025141 +2023-10-02 21:01:32,792 - Epoch: [68][ 190/ 1236] Overall Loss 0.259347 Objective Loss 0.259347 LR 0.001000 Time 0.024896 +2023-10-02 21:01:32,998 - Epoch: [68][ 200/ 1236] Overall Loss 0.259748 Objective Loss 0.259748 LR 0.001000 Time 0.024684 +2023-10-02 21:01:33,204 - Epoch: [68][ 210/ 1236] Overall Loss 0.260630 Objective Loss 0.260630 LR 0.001000 Time 0.024480 +2023-10-02 21:01:33,411 - Epoch: [68][ 220/ 1236] Overall Loss 0.260462 Objective Loss 0.260462 LR 0.001000 Time 0.024306 +2023-10-02 21:01:33,616 - Epoch: [68][ 230/ 1236] Overall Loss 0.260950 Objective Loss 0.260950 LR 0.001000 Time 0.024136 +2023-10-02 21:01:33,824 - Epoch: [68][ 240/ 1236] Overall Loss 0.261322 Objective Loss 0.261322 LR 0.001000 Time 0.023996 +2023-10-02 21:01:34,028 - Epoch: [68][ 250/ 1236] Overall Loss 0.260463 Objective Loss 0.260463 LR 0.001000 Time 0.023851 +2023-10-02 21:01:34,235 - Epoch: [68][ 260/ 1236] Overall Loss 0.260908 Objective Loss 0.260908 LR 0.001000 Time 0.023727 +2023-10-02 21:01:34,439 - Epoch: [68][ 270/ 1236] Overall Loss 0.260598 Objective Loss 0.260598 LR 0.001000 Time 0.023604 +2023-10-02 21:01:34,646 - Epoch: [68][ 280/ 1236] Overall Loss 0.261657 Objective Loss 0.261657 LR 0.001000 Time 0.023497 +2023-10-02 21:01:34,851 - Epoch: [68][ 290/ 1236] Overall Loss 0.261677 Objective Loss 0.261677 LR 0.001000 Time 0.023394 +2023-10-02 21:01:35,059 - Epoch: [68][ 300/ 1236] Overall Loss 0.261618 Objective Loss 0.261618 LR 0.001000 Time 0.023306 +2023-10-02 21:01:35,262 - Epoch: [68][ 310/ 1236] Overall Loss 0.261198 Objective Loss 0.261198 LR 0.001000 Time 0.023210 +2023-10-02 21:01:35,468 - Epoch: [68][ 320/ 1236] Overall Loss 0.261257 Objective Loss 0.261257 LR 0.001000 Time 0.023128 +2023-10-02 21:01:35,673 - Epoch: [68][ 330/ 1236] Overall Loss 0.261689 Objective Loss 0.261689 LR 0.001000 Time 0.023041 +2023-10-02 21:01:35,881 - Epoch: [68][ 340/ 1236] Overall Loss 0.261949 Objective Loss 0.261949 LR 0.001000 Time 0.022975 +2023-10-02 21:01:36,085 - Epoch: [68][ 350/ 1236] Overall Loss 0.262161 Objective Loss 0.262161 LR 0.001000 Time 0.022901 +2023-10-02 21:01:36,293 - Epoch: [68][ 360/ 1236] Overall Loss 0.262124 Objective Loss 0.262124 LR 0.001000 Time 0.022842 +2023-10-02 21:01:36,498 - Epoch: [68][ 370/ 1236] Overall Loss 0.262783 Objective Loss 0.262783 LR 0.001000 Time 0.022777 +2023-10-02 21:01:36,705 - Epoch: [68][ 380/ 1236] Overall Loss 0.263217 Objective Loss 0.263217 LR 0.001000 Time 0.022721 +2023-10-02 21:01:36,910 - Epoch: [68][ 390/ 1236] Overall Loss 0.263372 Objective Loss 0.263372 LR 0.001000 Time 0.022662 +2023-10-02 21:01:37,117 - Epoch: [68][ 400/ 1236] Overall Loss 0.263173 Objective Loss 0.263173 LR 0.001000 Time 0.022612 +2023-10-02 21:01:37,323 - Epoch: [68][ 410/ 1236] Overall Loss 0.263711 Objective Loss 0.263711 LR 0.001000 Time 0.022558 +2023-10-02 21:01:37,530 - Epoch: [68][ 420/ 1236] Overall Loss 0.263477 Objective Loss 0.263477 LR 0.001000 Time 0.022513 +2023-10-02 21:01:37,735 - Epoch: [68][ 430/ 1236] Overall Loss 0.263115 Objective Loss 0.263115 LR 0.001000 Time 0.022465 +2023-10-02 21:01:37,942 - Epoch: [68][ 440/ 1236] Overall Loss 0.263364 Objective Loss 0.263364 LR 0.001000 Time 0.022423 +2023-10-02 21:01:38,148 - Epoch: [68][ 450/ 1236] Overall Loss 0.263876 Objective Loss 0.263876 LR 0.001000 Time 0.022378 +2023-10-02 21:01:38,356 - Epoch: [68][ 460/ 1236] Overall Loss 0.263901 Objective Loss 0.263901 LR 0.001000 Time 0.022344 +2023-10-02 21:01:38,560 - Epoch: [68][ 470/ 1236] Overall Loss 0.263509 Objective Loss 0.263509 LR 0.001000 Time 0.022303 +2023-10-02 21:01:38,767 - Epoch: [68][ 480/ 1236] Overall Loss 0.263182 Objective Loss 0.263182 LR 0.001000 Time 0.022269 +2023-10-02 21:01:38,973 - Epoch: [68][ 490/ 1236] Overall Loss 0.263739 Objective Loss 0.263739 LR 0.001000 Time 0.022231 +2023-10-02 21:01:39,181 - Epoch: [68][ 500/ 1236] Overall Loss 0.263433 Objective Loss 0.263433 LR 0.001000 Time 0.022202 +2023-10-02 21:01:39,385 - Epoch: [68][ 510/ 1236] Overall Loss 0.263759 Objective Loss 0.263759 LR 0.001000 Time 0.022166 +2023-10-02 21:01:39,592 - Epoch: [68][ 520/ 1236] Overall Loss 0.263272 Objective Loss 0.263272 LR 0.001000 Time 0.022138 +2023-10-02 21:01:39,798 - Epoch: [68][ 530/ 1236] Overall Loss 0.263369 Objective Loss 0.263369 LR 0.001000 Time 0.022105 +2023-10-02 21:01:40,005 - Epoch: [68][ 540/ 1236] Overall Loss 0.263244 Objective Loss 0.263244 LR 0.001000 Time 0.022079 +2023-10-02 21:01:40,211 - Epoch: [68][ 550/ 1236] Overall Loss 0.263682 Objective Loss 0.263682 LR 0.001000 Time 0.022049 +2023-10-02 21:01:40,418 - Epoch: [68][ 560/ 1236] Overall Loss 0.263698 Objective Loss 0.263698 LR 0.001000 Time 0.022024 +2023-10-02 21:01:40,623 - Epoch: [68][ 570/ 1236] Overall Loss 0.262958 Objective Loss 0.262958 LR 0.001000 Time 0.021995 +2023-10-02 21:01:40,830 - Epoch: [68][ 580/ 1236] Overall Loss 0.262417 Objective Loss 0.262417 LR 0.001000 Time 0.021973 +2023-10-02 21:01:41,036 - Epoch: [68][ 590/ 1236] Overall Loss 0.262364 Objective Loss 0.262364 LR 0.001000 Time 0.021946 +2023-10-02 21:01:41,243 - Epoch: [68][ 600/ 1236] Overall Loss 0.262610 Objective Loss 0.262610 LR 0.001000 Time 0.021925 +2023-10-02 21:01:41,449 - Epoch: [68][ 610/ 1236] Overall Loss 0.262975 Objective Loss 0.262975 LR 0.001000 Time 0.021900 +2023-10-02 21:01:41,656 - Epoch: [68][ 620/ 1236] Overall Loss 0.262808 Objective Loss 0.262808 LR 0.001000 Time 0.021880 +2023-10-02 21:01:41,861 - Epoch: [68][ 630/ 1236] Overall Loss 0.262720 Objective Loss 0.262720 LR 0.001000 Time 0.021857 +2023-10-02 21:01:42,068 - Epoch: [68][ 640/ 1236] Overall Loss 0.263018 Objective Loss 0.263018 LR 0.001000 Time 0.021839 +2023-10-02 21:01:42,274 - Epoch: [68][ 650/ 1236] Overall Loss 0.263177 Objective Loss 0.263177 LR 0.001000 Time 0.021817 +2023-10-02 21:01:42,481 - Epoch: [68][ 660/ 1236] Overall Loss 0.262933 Objective Loss 0.262933 LR 0.001000 Time 0.021800 +2023-10-02 21:01:42,687 - Epoch: [68][ 670/ 1236] Overall Loss 0.263474 Objective Loss 0.263474 LR 0.001000 Time 0.021779 +2023-10-02 21:01:42,894 - Epoch: [68][ 680/ 1236] Overall Loss 0.263602 Objective Loss 0.263602 LR 0.001000 Time 0.021763 +2023-10-02 21:01:43,099 - Epoch: [68][ 690/ 1236] Overall Loss 0.262953 Objective Loss 0.262953 LR 0.001000 Time 0.021743 +2023-10-02 21:01:43,306 - Epoch: [68][ 700/ 1236] Overall Loss 0.263151 Objective Loss 0.263151 LR 0.001000 Time 0.021728 +2023-10-02 21:01:43,512 - Epoch: [68][ 710/ 1236] Overall Loss 0.262751 Objective Loss 0.262751 LR 0.001000 Time 0.021709 +2023-10-02 21:01:43,719 - Epoch: [68][ 720/ 1236] Overall Loss 0.263367 Objective Loss 0.263367 LR 0.001000 Time 0.021695 +2023-10-02 21:01:43,925 - Epoch: [68][ 730/ 1236] Overall Loss 0.263553 Objective Loss 0.263553 LR 0.001000 Time 0.021677 +2023-10-02 21:01:44,132 - Epoch: [68][ 740/ 1236] Overall Loss 0.263772 Objective Loss 0.263772 LR 0.001000 Time 0.021664 +2023-10-02 21:01:44,337 - Epoch: [68][ 750/ 1236] Overall Loss 0.263794 Objective Loss 0.263794 LR 0.001000 Time 0.021647 +2023-10-02 21:01:44,545 - Epoch: [68][ 760/ 1236] Overall Loss 0.264073 Objective Loss 0.264073 LR 0.001000 Time 0.021634 +2023-10-02 21:01:44,749 - Epoch: [68][ 770/ 1236] Overall Loss 0.264099 Objective Loss 0.264099 LR 0.001000 Time 0.021619 +2023-10-02 21:01:44,957 - Epoch: [68][ 780/ 1236] Overall Loss 0.263716 Objective Loss 0.263716 LR 0.001000 Time 0.021608 +2023-10-02 21:01:45,162 - Epoch: [68][ 790/ 1236] Overall Loss 0.263837 Objective Loss 0.263837 LR 0.001000 Time 0.021593 +2023-10-02 21:01:45,370 - Epoch: [68][ 800/ 1236] Overall Loss 0.263935 Objective Loss 0.263935 LR 0.001000 Time 0.021583 +2023-10-02 21:01:45,575 - Epoch: [68][ 810/ 1236] Overall Loss 0.263813 Objective Loss 0.263813 LR 0.001000 Time 0.021569 +2023-10-02 21:01:45,782 - Epoch: [68][ 820/ 1236] Overall Loss 0.263774 Objective Loss 0.263774 LR 0.001000 Time 0.021558 +2023-10-02 21:01:45,988 - Epoch: [68][ 830/ 1236] Overall Loss 0.263679 Objective Loss 0.263679 LR 0.001000 Time 0.021544 +2023-10-02 21:01:46,195 - Epoch: [68][ 840/ 1236] Overall Loss 0.264021 Objective Loss 0.264021 LR 0.001000 Time 0.021534 +2023-10-02 21:01:46,400 - Epoch: [68][ 850/ 1236] Overall Loss 0.264205 Objective Loss 0.264205 LR 0.001000 Time 0.021521 +2023-10-02 21:01:46,607 - Epoch: [68][ 860/ 1236] Overall Loss 0.263762 Objective Loss 0.263762 LR 0.001000 Time 0.021511 +2023-10-02 21:01:46,813 - Epoch: [68][ 870/ 1236] Overall Loss 0.263655 Objective Loss 0.263655 LR 0.001000 Time 0.021499 +2023-10-02 21:01:47,020 - Epoch: [68][ 880/ 1236] Overall Loss 0.264099 Objective Loss 0.264099 LR 0.001000 Time 0.021489 +2023-10-02 21:01:47,226 - Epoch: [68][ 890/ 1236] Overall Loss 0.264106 Objective Loss 0.264106 LR 0.001000 Time 0.021477 +2023-10-02 21:01:47,433 - Epoch: [68][ 900/ 1236] Overall Loss 0.264044 Objective Loss 0.264044 LR 0.001000 Time 0.021468 +2023-10-02 21:01:47,639 - Epoch: [68][ 910/ 1236] Overall Loss 0.263915 Objective Loss 0.263915 LR 0.001000 Time 0.021457 +2023-10-02 21:01:47,845 - Epoch: [68][ 920/ 1236] Overall Loss 0.264145 Objective Loss 0.264145 LR 0.001000 Time 0.021448 +2023-10-02 21:01:48,051 - Epoch: [68][ 930/ 1236] Overall Loss 0.263995 Objective Loss 0.263995 LR 0.001000 Time 0.021437 +2023-10-02 21:01:48,258 - Epoch: [68][ 940/ 1236] Overall Loss 0.264204 Objective Loss 0.264204 LR 0.001000 Time 0.021429 +2023-10-02 21:01:48,463 - Epoch: [68][ 950/ 1236] Overall Loss 0.264412 Objective Loss 0.264412 LR 0.001000 Time 0.021417 +2023-10-02 21:01:48,670 - Epoch: [68][ 960/ 1236] Overall Loss 0.264841 Objective Loss 0.264841 LR 0.001000 Time 0.021410 +2023-10-02 21:01:48,876 - Epoch: [68][ 970/ 1236] Overall Loss 0.264751 Objective Loss 0.264751 LR 0.001000 Time 0.021399 +2023-10-02 21:01:49,083 - Epoch: [68][ 980/ 1236] Overall Loss 0.264574 Objective Loss 0.264574 LR 0.001000 Time 0.021392 +2023-10-02 21:01:49,289 - Epoch: [68][ 990/ 1236] Overall Loss 0.264757 Objective Loss 0.264757 LR 0.001000 Time 0.021382 +2023-10-02 21:01:49,496 - Epoch: [68][ 1000/ 1236] Overall Loss 0.264571 Objective Loss 0.264571 LR 0.001000 Time 0.021375 +2023-10-02 21:01:49,702 - Epoch: [68][ 1010/ 1236] Overall Loss 0.264420 Objective Loss 0.264420 LR 0.001000 Time 0.021366 +2023-10-02 21:01:49,909 - Epoch: [68][ 1020/ 1236] Overall Loss 0.264495 Objective Loss 0.264495 LR 0.001000 Time 0.021360 +2023-10-02 21:01:50,116 - Epoch: [68][ 1030/ 1236] Overall Loss 0.264618 Objective Loss 0.264618 LR 0.001000 Time 0.021352 +2023-10-02 21:01:50,324 - Epoch: [68][ 1040/ 1236] Overall Loss 0.264700 Objective Loss 0.264700 LR 0.001000 Time 0.021346 +2023-10-02 21:01:50,531 - Epoch: [68][ 1050/ 1236] Overall Loss 0.264855 Objective Loss 0.264855 LR 0.001000 Time 0.021338 +2023-10-02 21:01:50,740 - Epoch: [68][ 1060/ 1236] Overall Loss 0.264843 Objective Loss 0.264843 LR 0.001000 Time 0.021333 +2023-10-02 21:01:50,947 - Epoch: [68][ 1070/ 1236] Overall Loss 0.265092 Objective Loss 0.265092 LR 0.001000 Time 0.021326 +2023-10-02 21:01:51,155 - Epoch: [68][ 1080/ 1236] Overall Loss 0.264868 Objective Loss 0.264868 LR 0.001000 Time 0.021321 +2023-10-02 21:01:51,362 - Epoch: [68][ 1090/ 1236] Overall Loss 0.264638 Objective Loss 0.264638 LR 0.001000 Time 0.021314 +2023-10-02 21:01:51,570 - Epoch: [68][ 1100/ 1236] Overall Loss 0.264889 Objective Loss 0.264889 LR 0.001000 Time 0.021309 +2023-10-02 21:01:51,777 - Epoch: [68][ 1110/ 1236] Overall Loss 0.264769 Objective Loss 0.264769 LR 0.001000 Time 0.021303 +2023-10-02 21:01:51,985 - Epoch: [68][ 1120/ 1236] Overall Loss 0.264851 Objective Loss 0.264851 LR 0.001000 Time 0.021298 +2023-10-02 21:01:52,192 - Epoch: [68][ 1130/ 1236] Overall Loss 0.264665 Objective Loss 0.264665 LR 0.001000 Time 0.021291 +2023-10-02 21:01:52,400 - Epoch: [68][ 1140/ 1236] Overall Loss 0.264359 Objective Loss 0.264359 LR 0.001000 Time 0.021287 +2023-10-02 21:01:52,607 - Epoch: [68][ 1150/ 1236] Overall Loss 0.264117 Objective Loss 0.264117 LR 0.001000 Time 0.021280 +2023-10-02 21:01:52,815 - Epoch: [68][ 1160/ 1236] Overall Loss 0.264042 Objective Loss 0.264042 LR 0.001000 Time 0.021276 +2023-10-02 21:01:53,022 - Epoch: [68][ 1170/ 1236] Overall Loss 0.264107 Objective Loss 0.264107 LR 0.001000 Time 0.021271 +2023-10-02 21:01:53,230 - Epoch: [68][ 1180/ 1236] Overall Loss 0.264022 Objective Loss 0.264022 LR 0.001000 Time 0.021266 +2023-10-02 21:01:53,437 - Epoch: [68][ 1190/ 1236] Overall Loss 0.263822 Objective Loss 0.263822 LR 0.001000 Time 0.021261 +2023-10-02 21:01:53,645 - Epoch: [68][ 1200/ 1236] Overall Loss 0.263901 Objective Loss 0.263901 LR 0.001000 Time 0.021257 +2023-10-02 21:01:53,853 - Epoch: [68][ 1210/ 1236] Overall Loss 0.263984 Objective Loss 0.263984 LR 0.001000 Time 0.021253 +2023-10-02 21:01:54,061 - Epoch: [68][ 1220/ 1236] Overall Loss 0.263975 Objective Loss 0.263975 LR 0.001000 Time 0.021249 +2023-10-02 21:01:54,322 - Epoch: [68][ 1230/ 1236] Overall Loss 0.263977 Objective Loss 0.263977 LR 0.001000 Time 0.021288 +2023-10-02 21:01:54,444 - Epoch: [68][ 1236/ 1236] Overall Loss 0.264075 Objective Loss 0.264075 Top1 85.947047 Top5 97.963340 LR 0.001000 Time 0.021283 +2023-10-02 21:01:54,582 - --- validate (epoch=68)----------- +2023-10-02 21:01:54,582 - 29943 samples (256 per mini-batch) +2023-10-02 21:01:55,071 - Epoch: [68][ 10/ 117] Loss 0.338527 Top1 82.929688 Top5 98.281250 +2023-10-02 21:01:55,225 - Epoch: [68][ 20/ 117] Loss 0.381037 Top1 82.656250 Top5 98.105469 +2023-10-02 21:01:55,378 - Epoch: [68][ 30/ 117] Loss 0.373187 Top1 82.864583 Top5 98.151042 +2023-10-02 21:01:55,531 - Epoch: [68][ 40/ 117] Loss 0.367285 Top1 83.212891 Top5 98.193359 +2023-10-02 21:01:55,684 - Epoch: [68][ 50/ 117] Loss 0.360211 Top1 83.515625 Top5 98.234375 +2023-10-02 21:01:55,839 - Epoch: [68][ 60/ 117] Loss 0.357377 Top1 83.437500 Top5 98.235677 +2023-10-02 21:01:55,992 - Epoch: [68][ 70/ 117] Loss 0.358433 Top1 83.387277 Top5 98.281250 +2023-10-02 21:01:56,146 - Epoch: [68][ 80/ 117] Loss 0.358178 Top1 83.334961 Top5 98.227539 +2023-10-02 21:01:56,297 - Epoch: [68][ 90/ 117] Loss 0.362259 Top1 83.203125 Top5 98.242188 +2023-10-02 21:01:56,450 - Epoch: [68][ 100/ 117] Loss 0.360719 Top1 83.250000 Top5 98.238281 +2023-10-02 21:01:56,607 - Epoch: [68][ 110/ 117] Loss 0.360741 Top1 83.227983 Top5 98.263494 +2023-10-02 21:01:56,697 - Epoch: [68][ 117/ 117] Loss 0.356774 Top1 83.284908 Top5 98.273386 +2023-10-02 21:01:56,837 - ==> Top1: 83.285 Top5: 98.273 Loss: 0.357 + +2023-10-02 21:01:56,838 - ==> Confusion: +[[ 935 1 1 1 8 3 0 2 6 59 3 4 0 3 8 1 0 1 0 0 14] + [ 1 1036 5 0 6 32 0 21 0 0 3 2 2 2 1 3 2 0 6 2 7] + [ 6 0 952 15 1 2 15 12 0 0 5 1 5 2 1 11 1 2 11 3 11] + [ 1 3 17 939 1 2 0 4 3 0 11 0 21 6 38 2 0 10 15 0 16] + [ 31 4 2 1 948 7 0 1 0 10 1 0 0 10 7 6 10 0 0 1 11] + [ 2 35 0 1 3 954 2 24 2 9 5 14 5 27 6 1 6 0 2 5 13] + [ 0 5 45 1 0 1 1091 8 0 0 9 1 2 0 0 6 0 1 1 13 7] + [ 3 15 15 1 8 25 3 1073 5 3 5 6 7 4 0 0 1 0 25 8 11] + [ 16 4 1 3 3 4 0 0 946 52 8 1 10 17 19 0 0 2 1 0 2] + [ 117 1 0 0 6 4 1 0 26 906 2 1 1 32 7 1 0 1 0 3 10] + [ 2 0 11 7 1 1 2 2 17 2 966 1 0 16 7 0 2 0 7 1 8] + [ 0 1 3 0 1 8 0 1 0 0 0 960 28 15 0 2 0 11 1 2 2] + [ 0 0 5 2 3 1 0 3 0 0 1 50 954 6 0 2 1 19 1 6 14] + [ 0 0 1 0 3 3 0 2 13 10 2 6 0 1063 4 0 0 1 0 0 11] + [ 11 2 5 19 6 0 0 0 22 5 2 0 6 4 994 0 0 3 10 0 12] + [ 0 0 2 3 6 0 0 0 0 0 0 11 11 1 0 1052 11 18 0 9 10] + [ 1 16 0 0 7 7 0 1 2 0 0 7 7 6 5 9 1067 1 3 4 18] + [ 0 0 1 2 0 0 1 0 0 0 0 9 28 6 3 6 1 975 1 5 0] + [ 3 5 11 14 0 1 0 25 10 0 6 0 8 0 15 0 1 0 954 2 13] + [ 0 2 4 6 0 5 9 25 0 0 2 16 5 4 1 0 5 1 2 1050 15] + [ 154 161 129 98 90 161 35 133 85 71 163 128 334 344 111 88 74 80 166 177 5123]] + +2023-10-02 21:01:56,839 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:01:56,839 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:01:56,846 - + +2023-10-02 21:01:56,846 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:01:57,870 - Epoch: [69][ 10/ 1236] Overall Loss 0.290830 Objective Loss 0.290830 LR 0.001000 Time 0.102346 +2023-10-02 21:01:58,078 - Epoch: [69][ 20/ 1236] Overall Loss 0.281226 Objective Loss 0.281226 LR 0.001000 Time 0.061579 +2023-10-02 21:01:58,286 - Epoch: [69][ 30/ 1236] Overall Loss 0.265408 Objective Loss 0.265408 LR 0.001000 Time 0.047918 +2023-10-02 21:01:58,494 - Epoch: [69][ 40/ 1236] Overall Loss 0.263645 Objective Loss 0.263645 LR 0.001000 Time 0.041145 +2023-10-02 21:01:58,702 - Epoch: [69][ 50/ 1236] Overall Loss 0.256286 Objective Loss 0.256286 LR 0.001000 Time 0.037027 +2023-10-02 21:01:58,910 - Epoch: [69][ 60/ 1236] Overall Loss 0.251604 Objective Loss 0.251604 LR 0.001000 Time 0.034324 +2023-10-02 21:01:59,117 - Epoch: [69][ 70/ 1236] Overall Loss 0.256637 Objective Loss 0.256637 LR 0.001000 Time 0.032359 +2023-10-02 21:01:59,327 - Epoch: [69][ 80/ 1236] Overall Loss 0.257924 Objective Loss 0.257924 LR 0.001000 Time 0.030930 +2023-10-02 21:01:59,532 - Epoch: [69][ 90/ 1236] Overall Loss 0.257115 Objective Loss 0.257115 LR 0.001000 Time 0.029776 +2023-10-02 21:01:59,742 - Epoch: [69][ 100/ 1236] Overall Loss 0.256984 Objective Loss 0.256984 LR 0.001000 Time 0.028888 +2023-10-02 21:01:59,947 - Epoch: [69][ 110/ 1236] Overall Loss 0.256803 Objective Loss 0.256803 LR 0.001000 Time 0.028130 +2023-10-02 21:02:00,157 - Epoch: [69][ 120/ 1236] Overall Loss 0.255454 Objective Loss 0.255454 LR 0.001000 Time 0.027534 +2023-10-02 21:02:00,363 - Epoch: [69][ 130/ 1236] Overall Loss 0.255970 Objective Loss 0.255970 LR 0.001000 Time 0.026996 +2023-10-02 21:02:00,572 - Epoch: [69][ 140/ 1236] Overall Loss 0.256757 Objective Loss 0.256757 LR 0.001000 Time 0.026553 +2023-10-02 21:02:00,779 - Epoch: [69][ 150/ 1236] Overall Loss 0.256106 Objective Loss 0.256106 LR 0.001000 Time 0.026155 +2023-10-02 21:02:00,987 - Epoch: [69][ 160/ 1236] Overall Loss 0.258281 Objective Loss 0.258281 LR 0.001000 Time 0.025818 +2023-10-02 21:02:01,194 - Epoch: [69][ 170/ 1236] Overall Loss 0.256114 Objective Loss 0.256114 LR 0.001000 Time 0.025517 +2023-10-02 21:02:01,404 - Epoch: [69][ 180/ 1236] Overall Loss 0.255868 Objective Loss 0.255868 LR 0.001000 Time 0.025264 +2023-10-02 21:02:01,610 - Epoch: [69][ 190/ 1236] Overall Loss 0.256117 Objective Loss 0.256117 LR 0.001000 Time 0.025016 +2023-10-02 21:02:01,820 - Epoch: [69][ 200/ 1236] Overall Loss 0.255144 Objective Loss 0.255144 LR 0.001000 Time 0.024812 +2023-10-02 21:02:02,026 - Epoch: [69][ 210/ 1236] Overall Loss 0.256845 Objective Loss 0.256845 LR 0.001000 Time 0.024611 +2023-10-02 21:02:02,235 - Epoch: [69][ 220/ 1236] Overall Loss 0.257600 Objective Loss 0.257600 LR 0.001000 Time 0.024440 +2023-10-02 21:02:02,440 - Epoch: [69][ 230/ 1236] Overall Loss 0.256693 Objective Loss 0.256693 LR 0.001000 Time 0.024269 +2023-10-02 21:02:02,648 - Epoch: [69][ 240/ 1236] Overall Loss 0.257435 Objective Loss 0.257435 LR 0.001000 Time 0.024122 +2023-10-02 21:02:02,855 - Epoch: [69][ 250/ 1236] Overall Loss 0.258405 Objective Loss 0.258405 LR 0.001000 Time 0.023985 +2023-10-02 21:02:03,065 - Epoch: [69][ 260/ 1236] Overall Loss 0.258372 Objective Loss 0.258372 LR 0.001000 Time 0.023868 +2023-10-02 21:02:03,271 - Epoch: [69][ 270/ 1236] Overall Loss 0.258531 Objective Loss 0.258531 LR 0.001000 Time 0.023746 +2023-10-02 21:02:03,481 - Epoch: [69][ 280/ 1236] Overall Loss 0.258092 Objective Loss 0.258092 LR 0.001000 Time 0.023646 +2023-10-02 21:02:03,686 - Epoch: [69][ 290/ 1236] Overall Loss 0.258393 Objective Loss 0.258393 LR 0.001000 Time 0.023540 +2023-10-02 21:02:03,896 - Epoch: [69][ 300/ 1236] Overall Loss 0.257554 Objective Loss 0.257554 LR 0.001000 Time 0.023454 +2023-10-02 21:02:04,102 - Epoch: [69][ 310/ 1236] Overall Loss 0.257891 Objective Loss 0.257891 LR 0.001000 Time 0.023361 +2023-10-02 21:02:04,310 - Epoch: [69][ 320/ 1236] Overall Loss 0.258297 Objective Loss 0.258297 LR 0.001000 Time 0.023279 +2023-10-02 21:02:04,517 - Epoch: [69][ 330/ 1236] Overall Loss 0.258868 Objective Loss 0.258868 LR 0.001000 Time 0.023194 +2023-10-02 21:02:04,727 - Epoch: [69][ 340/ 1236] Overall Loss 0.258741 Objective Loss 0.258741 LR 0.001000 Time 0.023129 +2023-10-02 21:02:04,933 - Epoch: [69][ 350/ 1236] Overall Loss 0.258885 Objective Loss 0.258885 LR 0.001000 Time 0.023056 +2023-10-02 21:02:05,141 - Epoch: [69][ 360/ 1236] Overall Loss 0.257738 Objective Loss 0.257738 LR 0.001000 Time 0.022995 +2023-10-02 21:02:05,349 - Epoch: [69][ 370/ 1236] Overall Loss 0.257935 Objective Loss 0.257935 LR 0.001000 Time 0.022931 +2023-10-02 21:02:05,558 - Epoch: [69][ 380/ 1236] Overall Loss 0.258232 Objective Loss 0.258232 LR 0.001000 Time 0.022875 +2023-10-02 21:02:05,765 - Epoch: [69][ 390/ 1236] Overall Loss 0.257972 Objective Loss 0.257972 LR 0.001000 Time 0.022817 +2023-10-02 21:02:05,974 - Epoch: [69][ 400/ 1236] Overall Loss 0.256981 Objective Loss 0.256981 LR 0.001000 Time 0.022769 +2023-10-02 21:02:06,182 - Epoch: [69][ 410/ 1236] Overall Loss 0.256170 Objective Loss 0.256170 LR 0.001000 Time 0.022716 +2023-10-02 21:02:06,391 - Epoch: [69][ 420/ 1236] Overall Loss 0.255907 Objective Loss 0.255907 LR 0.001000 Time 0.022671 +2023-10-02 21:02:06,598 - Epoch: [69][ 430/ 1236] Overall Loss 0.256396 Objective Loss 0.256396 LR 0.001000 Time 0.022623 +2023-10-02 21:02:06,807 - Epoch: [69][ 440/ 1236] Overall Loss 0.256692 Objective Loss 0.256692 LR 0.001000 Time 0.022583 +2023-10-02 21:02:07,015 - Epoch: [69][ 450/ 1236] Overall Loss 0.255967 Objective Loss 0.255967 LR 0.001000 Time 0.022539 +2023-10-02 21:02:07,224 - Epoch: [69][ 460/ 1236] Overall Loss 0.256227 Objective Loss 0.256227 LR 0.001000 Time 0.022503 +2023-10-02 21:02:07,432 - Epoch: [69][ 470/ 1236] Overall Loss 0.256037 Objective Loss 0.256037 LR 0.001000 Time 0.022462 +2023-10-02 21:02:07,640 - Epoch: [69][ 480/ 1236] Overall Loss 0.255508 Objective Loss 0.255508 LR 0.001000 Time 0.022429 +2023-10-02 21:02:07,848 - Epoch: [69][ 490/ 1236] Overall Loss 0.254455 Objective Loss 0.254455 LR 0.001000 Time 0.022391 +2023-10-02 21:02:08,057 - Epoch: [69][ 500/ 1236] Overall Loss 0.254539 Objective Loss 0.254539 LR 0.001000 Time 0.022360 +2023-10-02 21:02:08,264 - Epoch: [69][ 510/ 1236] Overall Loss 0.254767 Objective Loss 0.254767 LR 0.001000 Time 0.022326 +2023-10-02 21:02:08,473 - Epoch: [69][ 520/ 1236] Overall Loss 0.255227 Objective Loss 0.255227 LR 0.001000 Time 0.022298 +2023-10-02 21:02:08,681 - Epoch: [69][ 530/ 1236] Overall Loss 0.254912 Objective Loss 0.254912 LR 0.001000 Time 0.022266 +2023-10-02 21:02:08,889 - Epoch: [69][ 540/ 1236] Overall Loss 0.255035 Objective Loss 0.255035 LR 0.001000 Time 0.022240 +2023-10-02 21:02:09,097 - Epoch: [69][ 550/ 1236] Overall Loss 0.255101 Objective Loss 0.255101 LR 0.001000 Time 0.022210 +2023-10-02 21:02:09,306 - Epoch: [69][ 560/ 1236] Overall Loss 0.255057 Objective Loss 0.255057 LR 0.001000 Time 0.022185 +2023-10-02 21:02:09,513 - Epoch: [69][ 570/ 1236] Overall Loss 0.255538 Objective Loss 0.255538 LR 0.001000 Time 0.022157 +2023-10-02 21:02:09,722 - Epoch: [69][ 580/ 1236] Overall Loss 0.255790 Objective Loss 0.255790 LR 0.001000 Time 0.022134 +2023-10-02 21:02:09,929 - Epoch: [69][ 590/ 1236] Overall Loss 0.256124 Objective Loss 0.256124 LR 0.001000 Time 0.022108 +2023-10-02 21:02:10,138 - Epoch: [69][ 600/ 1236] Overall Loss 0.256012 Objective Loss 0.256012 LR 0.001000 Time 0.022087 +2023-10-02 21:02:10,345 - Epoch: [69][ 610/ 1236] Overall Loss 0.255948 Objective Loss 0.255948 LR 0.001000 Time 0.022063 +2023-10-02 21:02:10,555 - Epoch: [69][ 620/ 1236] Overall Loss 0.256018 Objective Loss 0.256018 LR 0.001000 Time 0.022044 +2023-10-02 21:02:10,762 - Epoch: [69][ 630/ 1236] Overall Loss 0.255945 Objective Loss 0.255945 LR 0.001000 Time 0.022021 +2023-10-02 21:02:10,973 - Epoch: [69][ 640/ 1236] Overall Loss 0.256179 Objective Loss 0.256179 LR 0.001000 Time 0.022005 +2023-10-02 21:02:11,179 - Epoch: [69][ 650/ 1236] Overall Loss 0.256003 Objective Loss 0.256003 LR 0.001000 Time 0.021984 +2023-10-02 21:02:11,388 - Epoch: [69][ 660/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.021967 +2023-10-02 21:02:11,596 - Epoch: [69][ 670/ 1236] Overall Loss 0.256800 Objective Loss 0.256800 LR 0.001000 Time 0.021946 +2023-10-02 21:02:11,804 - Epoch: [69][ 680/ 1236] Overall Loss 0.256696 Objective Loss 0.256696 LR 0.001000 Time 0.021930 +2023-10-02 21:02:12,012 - Epoch: [69][ 690/ 1236] Overall Loss 0.256973 Objective Loss 0.256973 LR 0.001000 Time 0.021911 +2023-10-02 21:02:12,221 - Epoch: [69][ 700/ 1236] Overall Loss 0.256936 Objective Loss 0.256936 LR 0.001000 Time 0.021896 +2023-10-02 21:02:12,428 - Epoch: [69][ 710/ 1236] Overall Loss 0.256957 Objective Loss 0.256957 LR 0.001000 Time 0.021877 +2023-10-02 21:02:12,637 - Epoch: [69][ 720/ 1236] Overall Loss 0.257287 Objective Loss 0.257287 LR 0.001000 Time 0.021863 +2023-10-02 21:02:12,845 - Epoch: [69][ 730/ 1236] Overall Loss 0.257522 Objective Loss 0.257522 LR 0.001000 Time 0.021846 +2023-10-02 21:02:13,053 - Epoch: [69][ 740/ 1236] Overall Loss 0.257527 Objective Loss 0.257527 LR 0.001000 Time 0.021832 +2023-10-02 21:02:13,261 - Epoch: [69][ 750/ 1236] Overall Loss 0.257585 Objective Loss 0.257585 LR 0.001000 Time 0.021815 +2023-10-02 21:02:13,469 - Epoch: [69][ 760/ 1236] Overall Loss 0.257884 Objective Loss 0.257884 LR 0.001000 Time 0.021802 +2023-10-02 21:02:13,677 - Epoch: [69][ 770/ 1236] Overall Loss 0.258055 Objective Loss 0.258055 LR 0.001000 Time 0.021787 +2023-10-02 21:02:13,885 - Epoch: [69][ 780/ 1236] Overall Loss 0.257750 Objective Loss 0.257750 LR 0.001000 Time 0.021775 +2023-10-02 21:02:14,093 - Epoch: [69][ 790/ 1236] Overall Loss 0.257586 Objective Loss 0.257586 LR 0.001000 Time 0.021760 +2023-10-02 21:02:14,302 - Epoch: [69][ 800/ 1236] Overall Loss 0.257230 Objective Loss 0.257230 LR 0.001000 Time 0.021749 +2023-10-02 21:02:14,510 - Epoch: [69][ 810/ 1236] Overall Loss 0.257417 Objective Loss 0.257417 LR 0.001000 Time 0.021734 +2023-10-02 21:02:14,719 - Epoch: [69][ 820/ 1236] Overall Loss 0.257500 Objective Loss 0.257500 LR 0.001000 Time 0.021724 +2023-10-02 21:02:14,926 - Epoch: [69][ 830/ 1236] Overall Loss 0.256920 Objective Loss 0.256920 LR 0.001000 Time 0.021710 +2023-10-02 21:02:15,135 - Epoch: [69][ 840/ 1236] Overall Loss 0.256584 Objective Loss 0.256584 LR 0.001000 Time 0.021700 +2023-10-02 21:02:15,343 - Epoch: [69][ 850/ 1236] Overall Loss 0.256517 Objective Loss 0.256517 LR 0.001000 Time 0.021687 +2023-10-02 21:02:15,552 - Epoch: [69][ 860/ 1236] Overall Loss 0.256624 Objective Loss 0.256624 LR 0.001000 Time 0.021678 +2023-10-02 21:02:15,759 - Epoch: [69][ 870/ 1236] Overall Loss 0.256450 Objective Loss 0.256450 LR 0.001000 Time 0.021665 +2023-10-02 21:02:15,968 - Epoch: [69][ 880/ 1236] Overall Loss 0.256259 Objective Loss 0.256259 LR 0.001000 Time 0.021656 +2023-10-02 21:02:16,176 - Epoch: [69][ 890/ 1236] Overall Loss 0.256073 Objective Loss 0.256073 LR 0.001000 Time 0.021644 +2023-10-02 21:02:16,385 - Epoch: [69][ 900/ 1236] Overall Loss 0.256458 Objective Loss 0.256458 LR 0.001000 Time 0.021636 +2023-10-02 21:02:16,592 - Epoch: [69][ 910/ 1236] Overall Loss 0.256678 Objective Loss 0.256678 LR 0.001000 Time 0.021624 +2023-10-02 21:02:16,801 - Epoch: [69][ 920/ 1236] Overall Loss 0.256119 Objective Loss 0.256119 LR 0.001000 Time 0.021616 +2023-10-02 21:02:17,009 - Epoch: [69][ 930/ 1236] Overall Loss 0.256139 Objective Loss 0.256139 LR 0.001000 Time 0.021605 +2023-10-02 21:02:17,218 - Epoch: [69][ 940/ 1236] Overall Loss 0.256272 Objective Loss 0.256272 LR 0.001000 Time 0.021597 +2023-10-02 21:02:17,425 - Epoch: [69][ 950/ 1236] Overall Loss 0.256148 Objective Loss 0.256148 LR 0.001000 Time 0.021587 +2023-10-02 21:02:17,634 - Epoch: [69][ 960/ 1236] Overall Loss 0.256056 Objective Loss 0.256056 LR 0.001000 Time 0.021579 +2023-10-02 21:02:17,842 - Epoch: [69][ 970/ 1236] Overall Loss 0.256076 Objective Loss 0.256076 LR 0.001000 Time 0.021569 +2023-10-02 21:02:18,051 - Epoch: [69][ 980/ 1236] Overall Loss 0.256254 Objective Loss 0.256254 LR 0.001000 Time 0.021562 +2023-10-02 21:02:18,258 - Epoch: [69][ 990/ 1236] Overall Loss 0.255871 Objective Loss 0.255871 LR 0.001000 Time 0.021552 +2023-10-02 21:02:18,467 - Epoch: [69][ 1000/ 1236] Overall Loss 0.255878 Objective Loss 0.255878 LR 0.001000 Time 0.021545 +2023-10-02 21:02:18,675 - Epoch: [69][ 1010/ 1236] Overall Loss 0.255944 Objective Loss 0.255944 LR 0.001000 Time 0.021536 +2023-10-02 21:02:18,883 - Epoch: [69][ 1020/ 1236] Overall Loss 0.256076 Objective Loss 0.256076 LR 0.001000 Time 0.021529 +2023-10-02 21:02:19,091 - Epoch: [69][ 1030/ 1236] Overall Loss 0.256270 Objective Loss 0.256270 LR 0.001000 Time 0.021520 +2023-10-02 21:02:19,300 - Epoch: [69][ 1040/ 1236] Overall Loss 0.256284 Objective Loss 0.256284 LR 0.001000 Time 0.021514 +2023-10-02 21:02:19,508 - Epoch: [69][ 1050/ 1236] Overall Loss 0.256439 Objective Loss 0.256439 LR 0.001000 Time 0.021505 +2023-10-02 21:02:19,716 - Epoch: [69][ 1060/ 1236] Overall Loss 0.256406 Objective Loss 0.256406 LR 0.001000 Time 0.021499 +2023-10-02 21:02:19,924 - Epoch: [69][ 1070/ 1236] Overall Loss 0.256100 Objective Loss 0.256100 LR 0.001000 Time 0.021490 +2023-10-02 21:02:20,132 - Epoch: [69][ 1080/ 1236] Overall Loss 0.256029 Objective Loss 0.256029 LR 0.001000 Time 0.021484 +2023-10-02 21:02:20,340 - Epoch: [69][ 1090/ 1236] Overall Loss 0.255879 Objective Loss 0.255879 LR 0.001000 Time 0.021476 +2023-10-02 21:02:20,549 - Epoch: [69][ 1100/ 1236] Overall Loss 0.255833 Objective Loss 0.255833 LR 0.001000 Time 0.021471 +2023-10-02 21:02:20,757 - Epoch: [69][ 1110/ 1236] Overall Loss 0.256295 Objective Loss 0.256295 LR 0.001000 Time 0.021463 +2023-10-02 21:02:20,966 - Epoch: [69][ 1120/ 1236] Overall Loss 0.256842 Objective Loss 0.256842 LR 0.001000 Time 0.021458 +2023-10-02 21:02:21,174 - Epoch: [69][ 1130/ 1236] Overall Loss 0.257225 Objective Loss 0.257225 LR 0.001000 Time 0.021450 +2023-10-02 21:02:21,383 - Epoch: [69][ 1140/ 1236] Overall Loss 0.257105 Objective Loss 0.257105 LR 0.001000 Time 0.021445 +2023-10-02 21:02:21,591 - Epoch: [69][ 1150/ 1236] Overall Loss 0.257043 Objective Loss 0.257043 LR 0.001000 Time 0.021438 +2023-10-02 21:02:21,799 - Epoch: [69][ 1160/ 1236] Overall Loss 0.257156 Objective Loss 0.257156 LR 0.001000 Time 0.021433 +2023-10-02 21:02:22,007 - Epoch: [69][ 1170/ 1236] Overall Loss 0.257505 Objective Loss 0.257505 LR 0.001000 Time 0.021426 +2023-10-02 21:02:22,216 - Epoch: [69][ 1180/ 1236] Overall Loss 0.257540 Objective Loss 0.257540 LR 0.001000 Time 0.021421 +2023-10-02 21:02:22,424 - Epoch: [69][ 1190/ 1236] Overall Loss 0.257437 Objective Loss 0.257437 LR 0.001000 Time 0.021414 +2023-10-02 21:02:22,633 - Epoch: [69][ 1200/ 1236] Overall Loss 0.257385 Objective Loss 0.257385 LR 0.001000 Time 0.021410 +2023-10-02 21:02:22,841 - Epoch: [69][ 1210/ 1236] Overall Loss 0.257501 Objective Loss 0.257501 LR 0.001000 Time 0.021404 +2023-10-02 21:02:23,050 - Epoch: [69][ 1220/ 1236] Overall Loss 0.257568 Objective Loss 0.257568 LR 0.001000 Time 0.021399 +2023-10-02 21:02:23,312 - Epoch: [69][ 1230/ 1236] Overall Loss 0.257565 Objective Loss 0.257565 LR 0.001000 Time 0.021437 +2023-10-02 21:02:23,435 - Epoch: [69][ 1236/ 1236] Overall Loss 0.257816 Objective Loss 0.257816 Top1 85.132383 Top5 98.167006 LR 0.001000 Time 0.021433 +2023-10-02 21:02:23,567 - --- validate (epoch=69)----------- +2023-10-02 21:02:23,567 - 29943 samples (256 per mini-batch) +2023-10-02 21:02:24,057 - Epoch: [69][ 10/ 117] Loss 0.332223 Top1 82.773438 Top5 97.929688 +2023-10-02 21:02:24,204 - Epoch: [69][ 20/ 117] Loss 0.337624 Top1 83.457031 Top5 98.046875 +2023-10-02 21:02:24,352 - Epoch: [69][ 30/ 117] Loss 0.327782 Top1 83.945312 Top5 98.125000 +2023-10-02 21:02:24,498 - Epoch: [69][ 40/ 117] Loss 0.338306 Top1 83.554688 Top5 98.085938 +2023-10-02 21:02:24,644 - Epoch: [69][ 50/ 117] Loss 0.336108 Top1 83.679688 Top5 98.117188 +2023-10-02 21:02:24,790 - Epoch: [69][ 60/ 117] Loss 0.331365 Top1 83.652344 Top5 98.072917 +2023-10-02 21:02:24,936 - Epoch: [69][ 70/ 117] Loss 0.332816 Top1 83.660714 Top5 98.108259 +2023-10-02 21:02:25,083 - Epoch: [69][ 80/ 117] Loss 0.333894 Top1 83.583984 Top5 98.115234 +2023-10-02 21:02:25,231 - Epoch: [69][ 90/ 117] Loss 0.333408 Top1 83.615451 Top5 98.120660 +2023-10-02 21:02:25,378 - Epoch: [69][ 100/ 117] Loss 0.332757 Top1 83.660156 Top5 98.125000 +2023-10-02 21:02:25,533 - Epoch: [69][ 110/ 117] Loss 0.333649 Top1 83.739347 Top5 98.146307 +2023-10-02 21:02:25,623 - Epoch: [69][ 117/ 117] Loss 0.328246 Top1 83.936145 Top5 98.176535 +2023-10-02 21:02:25,766 - ==> Top1: 83.936 Top5: 98.177 Loss: 0.328 + +2023-10-02 21:02:25,767 - ==> Confusion: +[[ 946 1 3 0 6 3 0 0 10 53 1 0 1 2 2 1 4 3 1 0 13] + [ 0 1021 2 0 6 38 0 30 3 1 1 0 1 0 1 3 1 0 16 2 5] + [ 7 1 964 11 3 0 14 10 1 2 2 0 9 1 2 5 1 1 12 3 7] + [ 1 3 11 956 0 2 0 2 8 0 7 0 14 3 34 2 0 0 29 1 16] + [ 24 7 1 1 973 7 0 0 0 9 2 0 0 3 6 4 8 0 0 0 5] + [ 1 41 3 0 4 983 1 24 1 5 4 7 5 9 5 1 4 0 3 3 12] + [ 1 4 33 0 0 5 1114 9 0 0 4 2 0 0 0 4 0 0 1 7 7] + [ 3 13 10 1 3 24 0 1075 2 4 3 4 6 4 3 0 0 0 50 7 6] + [ 14 1 4 0 2 4 0 1 952 40 16 2 4 16 25 0 3 0 4 0 1] + [ 116 1 1 0 11 6 1 0 41 898 2 1 0 21 7 2 1 1 1 2 6] + [ 2 3 9 9 2 2 3 3 18 0 962 3 1 17 3 1 1 0 6 1 7] + [ 0 2 5 0 2 17 0 0 0 0 1 953 17 5 0 3 1 17 0 7 5] + [ 1 0 3 3 0 5 2 0 1 0 3 34 968 1 2 11 0 14 1 2 17] + [ 0 1 1 0 4 10 0 1 16 9 5 5 3 1035 7 0 2 1 0 3 16] + [ 13 1 3 24 6 0 1 0 16 3 0 0 3 4 1009 0 1 3 10 0 4] + [ 0 0 1 1 3 4 2 0 0 0 0 7 5 1 2 1066 12 12 1 8 9] + [ 2 14 0 0 7 11 1 0 2 0 1 4 2 3 3 11 1083 0 1 8 8] + [ 0 0 0 4 0 0 3 0 0 1 1 4 30 3 8 12 2 966 1 1 2] + [ 2 2 4 11 1 1 0 16 6 1 2 0 3 0 10 3 0 0 997 0 9] + [ 0 2 9 1 0 5 9 9 0 1 0 15 6 2 1 4 7 1 4 1063 13] + [ 195 163 136 69 106 183 41 104 103 69 148 96 316 272 145 45 144 49 175 197 5149]] + +2023-10-02 21:02:25,768 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:02:25,768 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:02:25,774 - + +2023-10-02 21:02:25,774 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:02:26,793 - Epoch: [70][ 10/ 1236] Overall Loss 0.242273 Objective Loss 0.242273 LR 0.001000 Time 0.101820 +2023-10-02 21:02:27,002 - Epoch: [70][ 20/ 1236] Overall Loss 0.236850 Objective Loss 0.236850 LR 0.001000 Time 0.061329 +2023-10-02 21:02:27,210 - Epoch: [70][ 30/ 1236] Overall Loss 0.239197 Objective Loss 0.239197 LR 0.001000 Time 0.047800 +2023-10-02 21:02:27,418 - Epoch: [70][ 40/ 1236] Overall Loss 0.246118 Objective Loss 0.246118 LR 0.001000 Time 0.041046 +2023-10-02 21:02:27,625 - Epoch: [70][ 50/ 1236] Overall Loss 0.250725 Objective Loss 0.250725 LR 0.001000 Time 0.036950 +2023-10-02 21:02:27,833 - Epoch: [70][ 60/ 1236] Overall Loss 0.245722 Objective Loss 0.245722 LR 0.001000 Time 0.034252 +2023-10-02 21:02:28,040 - Epoch: [70][ 70/ 1236] Overall Loss 0.241885 Objective Loss 0.241885 LR 0.001000 Time 0.032320 +2023-10-02 21:02:28,249 - Epoch: [70][ 80/ 1236] Overall Loss 0.244838 Objective Loss 0.244838 LR 0.001000 Time 0.030879 +2023-10-02 21:02:28,456 - Epoch: [70][ 90/ 1236] Overall Loss 0.244915 Objective Loss 0.244915 LR 0.001000 Time 0.029744 +2023-10-02 21:02:28,664 - Epoch: [70][ 100/ 1236] Overall Loss 0.245642 Objective Loss 0.245642 LR 0.001000 Time 0.028850 +2023-10-02 21:02:28,872 - Epoch: [70][ 110/ 1236] Overall Loss 0.246553 Objective Loss 0.246553 LR 0.001000 Time 0.028101 +2023-10-02 21:02:29,080 - Epoch: [70][ 120/ 1236] Overall Loss 0.245468 Objective Loss 0.245468 LR 0.001000 Time 0.027494 +2023-10-02 21:02:29,287 - Epoch: [70][ 130/ 1236] Overall Loss 0.244198 Objective Loss 0.244198 LR 0.001000 Time 0.026964 +2023-10-02 21:02:29,496 - Epoch: [70][ 140/ 1236] Overall Loss 0.245937 Objective Loss 0.245937 LR 0.001000 Time 0.026525 +2023-10-02 21:02:29,703 - Epoch: [70][ 150/ 1236] Overall Loss 0.244721 Objective Loss 0.244721 LR 0.001000 Time 0.026129 +2023-10-02 21:02:29,912 - Epoch: [70][ 160/ 1236] Overall Loss 0.245870 Objective Loss 0.245870 LR 0.001000 Time 0.025797 +2023-10-02 21:02:30,119 - Epoch: [70][ 170/ 1236] Overall Loss 0.245348 Objective Loss 0.245348 LR 0.001000 Time 0.025490 +2023-10-02 21:02:30,328 - Epoch: [70][ 180/ 1236] Overall Loss 0.246766 Objective Loss 0.246766 LR 0.001000 Time 0.025230 +2023-10-02 21:02:30,535 - Epoch: [70][ 190/ 1236] Overall Loss 0.246980 Objective Loss 0.246980 LR 0.001000 Time 0.024984 +2023-10-02 21:02:30,743 - Epoch: [70][ 200/ 1236] Overall Loss 0.246835 Objective Loss 0.246835 LR 0.001000 Time 0.024775 +2023-10-02 21:02:30,951 - Epoch: [70][ 210/ 1236] Overall Loss 0.247889 Objective Loss 0.247889 LR 0.001000 Time 0.024576 +2023-10-02 21:02:31,159 - Epoch: [70][ 220/ 1236] Overall Loss 0.247150 Objective Loss 0.247150 LR 0.001000 Time 0.024405 +2023-10-02 21:02:31,366 - Epoch: [70][ 230/ 1236] Overall Loss 0.246948 Objective Loss 0.246948 LR 0.001000 Time 0.024239 +2023-10-02 21:02:31,575 - Epoch: [70][ 240/ 1236] Overall Loss 0.246470 Objective Loss 0.246470 LR 0.001000 Time 0.024097 +2023-10-02 21:02:31,782 - Epoch: [70][ 250/ 1236] Overall Loss 0.246265 Objective Loss 0.246265 LR 0.001000 Time 0.023955 +2023-10-02 21:02:31,993 - Epoch: [70][ 260/ 1236] Overall Loss 0.246718 Objective Loss 0.246718 LR 0.001000 Time 0.023843 +2023-10-02 21:02:32,199 - Epoch: [70][ 270/ 1236] Overall Loss 0.246738 Objective Loss 0.246738 LR 0.001000 Time 0.023724 +2023-10-02 21:02:32,407 - Epoch: [70][ 280/ 1236] Overall Loss 0.246313 Objective Loss 0.246313 LR 0.001000 Time 0.023618 +2023-10-02 21:02:32,614 - Epoch: [70][ 290/ 1236] Overall Loss 0.247169 Objective Loss 0.247169 LR 0.001000 Time 0.023516 +2023-10-02 21:02:32,823 - Epoch: [70][ 300/ 1236] Overall Loss 0.247366 Objective Loss 0.247366 LR 0.001000 Time 0.023427 +2023-10-02 21:02:33,030 - Epoch: [70][ 310/ 1236] Overall Loss 0.247360 Objective Loss 0.247360 LR 0.001000 Time 0.023336 +2023-10-02 21:02:33,239 - Epoch: [70][ 320/ 1236] Overall Loss 0.248494 Objective Loss 0.248494 LR 0.001000 Time 0.023257 +2023-10-02 21:02:33,446 - Epoch: [70][ 330/ 1236] Overall Loss 0.248353 Objective Loss 0.248353 LR 0.001000 Time 0.023176 +2023-10-02 21:02:33,656 - Epoch: [70][ 340/ 1236] Overall Loss 0.248454 Objective Loss 0.248454 LR 0.001000 Time 0.023111 +2023-10-02 21:02:33,863 - Epoch: [70][ 350/ 1236] Overall Loss 0.248492 Objective Loss 0.248492 LR 0.001000 Time 0.023039 +2023-10-02 21:02:34,073 - Epoch: [70][ 360/ 1236] Overall Loss 0.249197 Objective Loss 0.249197 LR 0.001000 Time 0.022983 +2023-10-02 21:02:34,279 - Epoch: [70][ 370/ 1236] Overall Loss 0.249894 Objective Loss 0.249894 LR 0.001000 Time 0.022918 +2023-10-02 21:02:34,488 - Epoch: [70][ 380/ 1236] Overall Loss 0.250650 Objective Loss 0.250650 LR 0.001000 Time 0.022863 +2023-10-02 21:02:34,695 - Epoch: [70][ 390/ 1236] Overall Loss 0.250721 Objective Loss 0.250721 LR 0.001000 Time 0.022808 +2023-10-02 21:02:34,905 - Epoch: [70][ 400/ 1236] Overall Loss 0.250766 Objective Loss 0.250766 LR 0.001000 Time 0.022762 +2023-10-02 21:02:35,112 - Epoch: [70][ 410/ 1236] Overall Loss 0.250511 Objective Loss 0.250511 LR 0.001000 Time 0.022710 +2023-10-02 21:02:35,320 - Epoch: [70][ 420/ 1236] Overall Loss 0.249923 Objective Loss 0.249923 LR 0.001000 Time 0.022665 +2023-10-02 21:02:35,528 - Epoch: [70][ 430/ 1236] Overall Loss 0.250284 Objective Loss 0.250284 LR 0.001000 Time 0.022620 +2023-10-02 21:02:35,736 - Epoch: [70][ 440/ 1236] Overall Loss 0.250317 Objective Loss 0.250317 LR 0.001000 Time 0.022579 +2023-10-02 21:02:35,945 - Epoch: [70][ 450/ 1236] Overall Loss 0.250572 Objective Loss 0.250572 LR 0.001000 Time 0.022539 +2023-10-02 21:02:36,153 - Epoch: [70][ 460/ 1236] Overall Loss 0.251313 Objective Loss 0.251313 LR 0.001000 Time 0.022501 +2023-10-02 21:02:36,361 - Epoch: [70][ 470/ 1236] Overall Loss 0.251959 Objective Loss 0.251959 LR 0.001000 Time 0.022464 +2023-10-02 21:02:36,569 - Epoch: [70][ 480/ 1236] Overall Loss 0.252479 Objective Loss 0.252479 LR 0.001000 Time 0.022430 +2023-10-02 21:02:36,777 - Epoch: [70][ 490/ 1236] Overall Loss 0.252421 Objective Loss 0.252421 LR 0.001000 Time 0.022395 +2023-10-02 21:02:36,986 - Epoch: [70][ 500/ 1236] Overall Loss 0.252382 Objective Loss 0.252382 LR 0.001000 Time 0.022364 +2023-10-02 21:02:37,193 - Epoch: [70][ 510/ 1236] Overall Loss 0.252479 Objective Loss 0.252479 LR 0.001000 Time 0.022329 +2023-10-02 21:02:37,402 - Epoch: [70][ 520/ 1236] Overall Loss 0.252131 Objective Loss 0.252131 LR 0.001000 Time 0.022301 +2023-10-02 21:02:37,610 - Epoch: [70][ 530/ 1236] Overall Loss 0.252997 Objective Loss 0.252997 LR 0.001000 Time 0.022271 +2023-10-02 21:02:37,818 - Epoch: [70][ 540/ 1236] Overall Loss 0.253473 Objective Loss 0.253473 LR 0.001000 Time 0.022244 +2023-10-02 21:02:38,026 - Epoch: [70][ 550/ 1236] Overall Loss 0.253555 Objective Loss 0.253555 LR 0.001000 Time 0.022215 +2023-10-02 21:02:38,234 - Epoch: [70][ 560/ 1236] Overall Loss 0.253636 Objective Loss 0.253636 LR 0.001000 Time 0.022189 +2023-10-02 21:02:38,442 - Epoch: [70][ 570/ 1236] Overall Loss 0.253433 Objective Loss 0.253433 LR 0.001000 Time 0.022164 +2023-10-02 21:02:38,650 - Epoch: [70][ 580/ 1236] Overall Loss 0.253110 Objective Loss 0.253110 LR 0.001000 Time 0.022140 +2023-10-02 21:02:38,858 - Epoch: [70][ 590/ 1236] Overall Loss 0.253257 Objective Loss 0.253257 LR 0.001000 Time 0.022117 +2023-10-02 21:02:39,067 - Epoch: [70][ 600/ 1236] Overall Loss 0.253457 Objective Loss 0.253457 LR 0.001000 Time 0.022096 +2023-10-02 21:02:39,275 - Epoch: [70][ 610/ 1236] Overall Loss 0.253109 Objective Loss 0.253109 LR 0.001000 Time 0.022074 +2023-10-02 21:02:39,484 - Epoch: [70][ 620/ 1236] Overall Loss 0.253545 Objective Loss 0.253545 LR 0.001000 Time 0.022054 +2023-10-02 21:02:39,691 - Epoch: [70][ 630/ 1236] Overall Loss 0.253922 Objective Loss 0.253922 LR 0.001000 Time 0.022033 +2023-10-02 21:02:39,900 - Epoch: [70][ 640/ 1236] Overall Loss 0.253443 Objective Loss 0.253443 LR 0.001000 Time 0.022014 +2023-10-02 21:02:40,108 - Epoch: [70][ 650/ 1236] Overall Loss 0.253402 Objective Loss 0.253402 LR 0.001000 Time 0.021995 +2023-10-02 21:02:40,316 - Epoch: [70][ 660/ 1236] Overall Loss 0.253260 Objective Loss 0.253260 LR 0.001000 Time 0.021977 +2023-10-02 21:02:40,524 - Epoch: [70][ 670/ 1236] Overall Loss 0.253758 Objective Loss 0.253758 LR 0.001000 Time 0.021959 +2023-10-02 21:02:40,733 - Epoch: [70][ 680/ 1236] Overall Loss 0.253747 Objective Loss 0.253747 LR 0.001000 Time 0.021942 +2023-10-02 21:02:40,941 - Epoch: [70][ 690/ 1236] Overall Loss 0.253531 Objective Loss 0.253531 LR 0.001000 Time 0.021925 +2023-10-02 21:02:41,149 - Epoch: [70][ 700/ 1236] Overall Loss 0.253654 Objective Loss 0.253654 LR 0.001000 Time 0.021909 +2023-10-02 21:02:41,357 - Epoch: [70][ 710/ 1236] Overall Loss 0.253096 Objective Loss 0.253096 LR 0.001000 Time 0.021893 +2023-10-02 21:02:41,566 - Epoch: [70][ 720/ 1236] Overall Loss 0.252767 Objective Loss 0.252767 LR 0.001000 Time 0.021878 +2023-10-02 21:02:41,773 - Epoch: [70][ 730/ 1236] Overall Loss 0.253318 Objective Loss 0.253318 LR 0.001000 Time 0.021862 +2023-10-02 21:02:41,984 - Epoch: [70][ 740/ 1236] Overall Loss 0.253183 Objective Loss 0.253183 LR 0.001000 Time 0.021851 +2023-10-02 21:02:42,197 - Epoch: [70][ 750/ 1236] Overall Loss 0.252898 Objective Loss 0.252898 LR 0.001000 Time 0.021844 +2023-10-02 21:02:42,411 - Epoch: [70][ 760/ 1236] Overall Loss 0.253262 Objective Loss 0.253262 LR 0.001000 Time 0.021838 +2023-10-02 21:02:42,623 - Epoch: [70][ 770/ 1236] Overall Loss 0.253104 Objective Loss 0.253104 LR 0.001000 Time 0.021829 +2023-10-02 21:02:42,836 - Epoch: [70][ 780/ 1236] Overall Loss 0.253249 Objective Loss 0.253249 LR 0.001000 Time 0.021821 +2023-10-02 21:02:43,050 - Epoch: [70][ 790/ 1236] Overall Loss 0.253079 Objective Loss 0.253079 LR 0.001000 Time 0.021815 +2023-10-02 21:02:43,261 - Epoch: [70][ 800/ 1236] Overall Loss 0.253478 Objective Loss 0.253478 LR 0.001000 Time 0.021806 +2023-10-02 21:02:43,474 - Epoch: [70][ 810/ 1236] Overall Loss 0.253801 Objective Loss 0.253801 LR 0.001000 Time 0.021800 +2023-10-02 21:02:43,689 - Epoch: [70][ 820/ 1236] Overall Loss 0.254232 Objective Loss 0.254232 LR 0.001000 Time 0.021796 +2023-10-02 21:02:43,903 - Epoch: [70][ 830/ 1236] Overall Loss 0.254524 Objective Loss 0.254524 LR 0.001000 Time 0.021790 +2023-10-02 21:02:44,118 - Epoch: [70][ 840/ 1236] Overall Loss 0.254718 Objective Loss 0.254718 LR 0.001000 Time 0.021784 +2023-10-02 21:02:44,329 - Epoch: [70][ 850/ 1236] Overall Loss 0.254854 Objective Loss 0.254854 LR 0.001000 Time 0.021776 +2023-10-02 21:02:44,544 - Epoch: [70][ 860/ 1236] Overall Loss 0.255116 Objective Loss 0.255116 LR 0.001000 Time 0.021772 +2023-10-02 21:02:44,758 - Epoch: [70][ 870/ 1236] Overall Loss 0.254980 Objective Loss 0.254980 LR 0.001000 Time 0.021767 +2023-10-02 21:02:44,972 - Epoch: [70][ 880/ 1236] Overall Loss 0.255141 Objective Loss 0.255141 LR 0.001000 Time 0.021763 +2023-10-02 21:02:45,186 - Epoch: [70][ 890/ 1236] Overall Loss 0.255197 Objective Loss 0.255197 LR 0.001000 Time 0.021758 +2023-10-02 21:02:45,398 - Epoch: [70][ 900/ 1236] Overall Loss 0.255363 Objective Loss 0.255363 LR 0.001000 Time 0.021752 +2023-10-02 21:02:45,608 - Epoch: [70][ 910/ 1236] Overall Loss 0.255492 Objective Loss 0.255492 LR 0.001000 Time 0.021743 +2023-10-02 21:02:45,823 - Epoch: [70][ 920/ 1236] Overall Loss 0.255868 Objective Loss 0.255868 LR 0.001000 Time 0.021740 +2023-10-02 21:02:46,036 - Epoch: [70][ 930/ 1236] Overall Loss 0.255807 Objective Loss 0.255807 LR 0.001000 Time 0.021735 +2023-10-02 21:02:46,249 - Epoch: [70][ 940/ 1236] Overall Loss 0.256151 Objective Loss 0.256151 LR 0.001000 Time 0.021730 +2023-10-02 21:02:46,463 - Epoch: [70][ 950/ 1236] Overall Loss 0.256606 Objective Loss 0.256606 LR 0.001000 Time 0.021726 +2023-10-02 21:02:46,676 - Epoch: [70][ 960/ 1236] Overall Loss 0.256659 Objective Loss 0.256659 LR 0.001000 Time 0.021721 +2023-10-02 21:02:46,896 - Epoch: [70][ 970/ 1236] Overall Loss 0.256734 Objective Loss 0.256734 LR 0.001000 Time 0.021723 +2023-10-02 21:02:47,106 - Epoch: [70][ 980/ 1236] Overall Loss 0.256951 Objective Loss 0.256951 LR 0.001000 Time 0.021716 +2023-10-02 21:02:47,315 - Epoch: [70][ 990/ 1236] Overall Loss 0.257093 Objective Loss 0.257093 LR 0.001000 Time 0.021707 +2023-10-02 21:02:47,525 - Epoch: [70][ 1000/ 1236] Overall Loss 0.257294 Objective Loss 0.257294 LR 0.001000 Time 0.021700 +2023-10-02 21:02:47,735 - Epoch: [70][ 1010/ 1236] Overall Loss 0.257550 Objective Loss 0.257550 LR 0.001000 Time 0.021692 +2023-10-02 21:02:47,944 - Epoch: [70][ 1020/ 1236] Overall Loss 0.257727 Objective Loss 0.257727 LR 0.001000 Time 0.021685 +2023-10-02 21:02:48,153 - Epoch: [70][ 1030/ 1236] Overall Loss 0.257696 Objective Loss 0.257696 LR 0.001000 Time 0.021677 +2023-10-02 21:02:48,362 - Epoch: [70][ 1040/ 1236] Overall Loss 0.257802 Objective Loss 0.257802 LR 0.001000 Time 0.021669 +2023-10-02 21:02:48,571 - Epoch: [70][ 1050/ 1236] Overall Loss 0.257798 Objective Loss 0.257798 LR 0.001000 Time 0.021661 +2023-10-02 21:02:48,781 - Epoch: [70][ 1060/ 1236] Overall Loss 0.257845 Objective Loss 0.257845 LR 0.001000 Time 0.021654 +2023-10-02 21:02:48,989 - Epoch: [70][ 1070/ 1236] Overall Loss 0.257765 Objective Loss 0.257765 LR 0.001000 Time 0.021647 +2023-10-02 21:02:49,199 - Epoch: [70][ 1080/ 1236] Overall Loss 0.257802 Objective Loss 0.257802 LR 0.001000 Time 0.021640 +2023-10-02 21:02:49,408 - Epoch: [70][ 1090/ 1236] Overall Loss 0.258024 Objective Loss 0.258024 LR 0.001000 Time 0.021633 +2023-10-02 21:02:49,617 - Epoch: [70][ 1100/ 1236] Overall Loss 0.258287 Objective Loss 0.258287 LR 0.001000 Time 0.021626 +2023-10-02 21:02:49,826 - Epoch: [70][ 1110/ 1236] Overall Loss 0.258461 Objective Loss 0.258461 LR 0.001000 Time 0.021619 +2023-10-02 21:02:50,036 - Epoch: [70][ 1120/ 1236] Overall Loss 0.258534 Objective Loss 0.258534 LR 0.001000 Time 0.021613 +2023-10-02 21:02:50,245 - Epoch: [70][ 1130/ 1236] Overall Loss 0.258771 Objective Loss 0.258771 LR 0.001000 Time 0.021607 +2023-10-02 21:02:50,455 - Epoch: [70][ 1140/ 1236] Overall Loss 0.258546 Objective Loss 0.258546 LR 0.001000 Time 0.021601 +2023-10-02 21:02:50,664 - Epoch: [70][ 1150/ 1236] Overall Loss 0.258379 Objective Loss 0.258379 LR 0.001000 Time 0.021594 +2023-10-02 21:02:50,873 - Epoch: [70][ 1160/ 1236] Overall Loss 0.258314 Objective Loss 0.258314 LR 0.001000 Time 0.021588 +2023-10-02 21:02:51,082 - Epoch: [70][ 1170/ 1236] Overall Loss 0.258606 Objective Loss 0.258606 LR 0.001000 Time 0.021582 +2023-10-02 21:02:51,291 - Epoch: [70][ 1180/ 1236] Overall Loss 0.258779 Objective Loss 0.258779 LR 0.001000 Time 0.021576 +2023-10-02 21:02:51,501 - Epoch: [70][ 1190/ 1236] Overall Loss 0.258751 Objective Loss 0.258751 LR 0.001000 Time 0.021571 +2023-10-02 21:02:51,711 - Epoch: [70][ 1200/ 1236] Overall Loss 0.258794 Objective Loss 0.258794 LR 0.001000 Time 0.021566 +2023-10-02 21:02:51,921 - Epoch: [70][ 1210/ 1236] Overall Loss 0.258863 Objective Loss 0.258863 LR 0.001000 Time 0.021561 +2023-10-02 21:02:52,130 - Epoch: [70][ 1220/ 1236] Overall Loss 0.258976 Objective Loss 0.258976 LR 0.001000 Time 0.021555 +2023-10-02 21:02:52,393 - Epoch: [70][ 1230/ 1236] Overall Loss 0.259289 Objective Loss 0.259289 LR 0.001000 Time 0.021593 +2023-10-02 21:02:52,514 - Epoch: [70][ 1236/ 1236] Overall Loss 0.259344 Objective Loss 0.259344 Top1 85.132383 Top5 98.981670 LR 0.001000 Time 0.021587 +2023-10-02 21:02:52,643 - --- validate (epoch=70)----------- +2023-10-02 21:02:52,644 - 29943 samples (256 per mini-batch) +2023-10-02 21:02:53,129 - Epoch: [70][ 10/ 117] Loss 0.339231 Top1 82.890625 Top5 98.203125 +2023-10-02 21:02:53,282 - Epoch: [70][ 20/ 117] Loss 0.354561 Top1 82.695312 Top5 98.222656 +2023-10-02 21:02:53,433 - Epoch: [70][ 30/ 117] Loss 0.342118 Top1 83.333333 Top5 98.177083 +2023-10-02 21:02:53,583 - Epoch: [70][ 40/ 117] Loss 0.337035 Top1 83.496094 Top5 98.212891 +2023-10-02 21:02:53,733 - Epoch: [70][ 50/ 117] Loss 0.337534 Top1 83.593750 Top5 98.257812 +2023-10-02 21:02:53,883 - Epoch: [70][ 60/ 117] Loss 0.336973 Top1 83.574219 Top5 98.261719 +2023-10-02 21:02:54,034 - Epoch: [70][ 70/ 117] Loss 0.335900 Top1 83.526786 Top5 98.253348 +2023-10-02 21:02:54,185 - Epoch: [70][ 80/ 117] Loss 0.336947 Top1 83.554688 Top5 98.261719 +2023-10-02 21:02:54,335 - Epoch: [70][ 90/ 117] Loss 0.336140 Top1 83.615451 Top5 98.259549 +2023-10-02 21:02:54,486 - Epoch: [70][ 100/ 117] Loss 0.336480 Top1 83.488281 Top5 98.210938 +2023-10-02 21:02:54,644 - Epoch: [70][ 110/ 117] Loss 0.333197 Top1 83.650568 Top5 98.227983 +2023-10-02 21:02:54,733 - Epoch: [70][ 117/ 117] Loss 0.334486 Top1 83.712387 Top5 98.196573 +2023-10-02 21:02:54,875 - ==> Top1: 83.712 Top5: 98.197 Loss: 0.334 + +2023-10-02 21:02:54,876 - ==> Confusion: +[[ 912 2 2 0 17 3 0 0 10 69 2 0 1 2 5 1 3 1 0 0 20] + [ 0 1047 2 2 2 25 0 24 2 0 3 1 2 1 0 4 1 0 11 2 2] + [ 4 2 938 23 2 2 23 10 0 0 2 2 9 1 2 5 0 1 12 3 15] + [ 1 4 9 991 2 5 0 3 3 1 2 0 5 1 25 2 1 4 11 3 16] + [ 22 15 2 0 952 10 0 1 0 12 1 1 0 3 7 5 10 0 1 1 7] + [ 1 43 1 2 1 981 0 27 2 1 3 7 0 15 5 0 2 0 7 9 9] + [ 0 6 29 1 0 0 1109 12 0 1 7 0 0 0 0 3 0 0 2 8 13] + [ 2 16 14 2 3 20 0 1066 0 1 4 4 3 7 1 0 1 0 60 7 7] + [ 11 7 0 1 2 7 0 1 968 37 12 0 2 11 19 2 2 2 2 1 2] + [ 92 1 4 0 7 1 3 0 42 921 1 1 0 25 8 0 0 0 2 2 9] + [ 2 1 10 14 1 2 0 6 18 0 955 2 0 15 5 0 0 1 10 3 8] + [ 1 1 2 0 1 17 1 2 0 0 1 914 36 8 0 4 3 12 1 17 14] + [ 0 1 4 2 0 5 1 3 1 2 2 40 961 5 3 7 3 8 0 7 13] + [ 0 0 1 0 1 15 0 1 11 9 6 6 0 1042 8 1 3 1 0 1 13] + [ 8 2 3 27 6 0 0 0 23 2 2 0 4 1 1000 0 1 3 5 0 14] + [ 0 0 2 2 5 0 3 2 0 1 0 9 6 1 1 1050 18 14 2 5 13] + [ 1 22 2 1 5 5 0 0 2 0 0 4 1 1 2 5 1089 1 2 7 11] + [ 0 0 0 4 1 0 3 0 0 2 0 6 21 0 4 3 3 981 0 1 9] + [ 1 3 2 22 2 0 0 16 4 0 2 0 1 0 10 0 0 0 997 0 8] + [ 0 2 6 2 2 5 9 12 0 0 0 6 7 0 0 2 7 0 4 1071 17] + [ 112 253 102 103 76 213 43 102 108 71 163 79 334 264 135 54 149 52 168 203 5121]] + +2023-10-02 21:02:54,877 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:02:54,877 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:02:54,883 - + +2023-10-02 21:02:54,883 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:02:56,005 - Epoch: [71][ 10/ 1236] Overall Loss 0.234183 Objective Loss 0.234183 LR 0.001000 Time 0.112075 +2023-10-02 21:02:56,211 - Epoch: [71][ 20/ 1236] Overall Loss 0.247134 Objective Loss 0.247134 LR 0.001000 Time 0.066332 +2023-10-02 21:02:56,416 - Epoch: [71][ 30/ 1236] Overall Loss 0.260913 Objective Loss 0.260913 LR 0.001000 Time 0.051007 +2023-10-02 21:02:56,621 - Epoch: [71][ 40/ 1236] Overall Loss 0.271530 Objective Loss 0.271530 LR 0.001000 Time 0.043374 +2023-10-02 21:02:56,826 - Epoch: [71][ 50/ 1236] Overall Loss 0.271519 Objective Loss 0.271519 LR 0.001000 Time 0.038769 +2023-10-02 21:02:57,031 - Epoch: [71][ 60/ 1236] Overall Loss 0.265666 Objective Loss 0.265666 LR 0.001000 Time 0.035720 +2023-10-02 21:02:57,236 - Epoch: [71][ 70/ 1236] Overall Loss 0.265415 Objective Loss 0.265415 LR 0.001000 Time 0.033525 +2023-10-02 21:02:57,443 - Epoch: [71][ 80/ 1236] Overall Loss 0.267826 Objective Loss 0.267826 LR 0.001000 Time 0.031918 +2023-10-02 21:02:57,647 - Epoch: [71][ 90/ 1236] Overall Loss 0.265646 Objective Loss 0.265646 LR 0.001000 Time 0.030631 +2023-10-02 21:02:57,853 - Epoch: [71][ 100/ 1236] Overall Loss 0.265715 Objective Loss 0.265715 LR 0.001000 Time 0.029632 +2023-10-02 21:02:58,058 - Epoch: [71][ 110/ 1236] Overall Loss 0.265009 Objective Loss 0.265009 LR 0.001000 Time 0.028796 +2023-10-02 21:02:58,263 - Epoch: [71][ 120/ 1236] Overall Loss 0.263799 Objective Loss 0.263799 LR 0.001000 Time 0.028104 +2023-10-02 21:02:58,469 - Epoch: [71][ 130/ 1236] Overall Loss 0.262938 Objective Loss 0.262938 LR 0.001000 Time 0.027511 +2023-10-02 21:02:58,676 - Epoch: [71][ 140/ 1236] Overall Loss 0.262431 Objective Loss 0.262431 LR 0.001000 Time 0.027022 +2023-10-02 21:02:58,880 - Epoch: [71][ 150/ 1236] Overall Loss 0.261689 Objective Loss 0.261689 LR 0.001000 Time 0.026581 +2023-10-02 21:02:59,087 - Epoch: [71][ 160/ 1236] Overall Loss 0.260882 Objective Loss 0.260882 LR 0.001000 Time 0.026210 +2023-10-02 21:02:59,291 - Epoch: [71][ 170/ 1236] Overall Loss 0.261494 Objective Loss 0.261494 LR 0.001000 Time 0.025867 +2023-10-02 21:02:59,496 - Epoch: [71][ 180/ 1236] Overall Loss 0.261730 Objective Loss 0.261730 LR 0.001000 Time 0.025569 +2023-10-02 21:02:59,702 - Epoch: [71][ 190/ 1236] Overall Loss 0.260282 Objective Loss 0.260282 LR 0.001000 Time 0.025303 +2023-10-02 21:02:59,907 - Epoch: [71][ 200/ 1236] Overall Loss 0.259130 Objective Loss 0.259130 LR 0.001000 Time 0.025061 +2023-10-02 21:03:00,112 - Epoch: [71][ 210/ 1236] Overall Loss 0.259196 Objective Loss 0.259196 LR 0.001000 Time 0.024839 +2023-10-02 21:03:00,318 - Epoch: [71][ 220/ 1236] Overall Loss 0.258471 Objective Loss 0.258471 LR 0.001000 Time 0.024645 +2023-10-02 21:03:00,523 - Epoch: [71][ 230/ 1236] Overall Loss 0.257566 Objective Loss 0.257566 LR 0.001000 Time 0.024459 +2023-10-02 21:03:00,729 - Epoch: [71][ 240/ 1236] Overall Loss 0.257264 Objective Loss 0.257264 LR 0.001000 Time 0.024297 +2023-10-02 21:03:00,935 - Epoch: [71][ 250/ 1236] Overall Loss 0.256869 Objective Loss 0.256869 LR 0.001000 Time 0.024141 +2023-10-02 21:03:01,142 - Epoch: [71][ 260/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.024009 +2023-10-02 21:03:01,346 - Epoch: [71][ 270/ 1236] Overall Loss 0.256319 Objective Loss 0.256319 LR 0.001000 Time 0.023874 +2023-10-02 21:03:01,552 - Epoch: [71][ 280/ 1236] Overall Loss 0.256334 Objective Loss 0.256334 LR 0.001000 Time 0.023756 +2023-10-02 21:03:01,758 - Epoch: [71][ 290/ 1236] Overall Loss 0.256234 Objective Loss 0.256234 LR 0.001000 Time 0.023642 +2023-10-02 21:03:01,965 - Epoch: [71][ 300/ 1236] Overall Loss 0.255887 Objective Loss 0.255887 LR 0.001000 Time 0.023543 +2023-10-02 21:03:02,169 - Epoch: [71][ 310/ 1236] Overall Loss 0.255517 Objective Loss 0.255517 LR 0.001000 Time 0.023441 +2023-10-02 21:03:02,376 - Epoch: [71][ 320/ 1236] Overall Loss 0.256141 Objective Loss 0.256141 LR 0.001000 Time 0.023354 +2023-10-02 21:03:02,581 - Epoch: [71][ 330/ 1236] Overall Loss 0.257109 Objective Loss 0.257109 LR 0.001000 Time 0.023267 +2023-10-02 21:03:02,787 - Epoch: [71][ 340/ 1236] Overall Loss 0.256701 Objective Loss 0.256701 LR 0.001000 Time 0.023188 +2023-10-02 21:03:02,993 - Epoch: [71][ 350/ 1236] Overall Loss 0.256459 Objective Loss 0.256459 LR 0.001000 Time 0.023110 +2023-10-02 21:03:03,200 - Epoch: [71][ 360/ 1236] Overall Loss 0.256416 Objective Loss 0.256416 LR 0.001000 Time 0.023041 +2023-10-02 21:03:03,406 - Epoch: [71][ 370/ 1236] Overall Loss 0.257109 Objective Loss 0.257109 LR 0.001000 Time 0.022972 +2023-10-02 21:03:03,613 - Epoch: [71][ 380/ 1236] Overall Loss 0.256855 Objective Loss 0.256855 LR 0.001000 Time 0.022912 +2023-10-02 21:03:03,818 - Epoch: [71][ 390/ 1236] Overall Loss 0.256661 Objective Loss 0.256661 LR 0.001000 Time 0.022850 +2023-10-02 21:03:04,024 - Epoch: [71][ 400/ 1236] Overall Loss 0.256789 Objective Loss 0.256789 LR 0.001000 Time 0.022793 +2023-10-02 21:03:04,230 - Epoch: [71][ 410/ 1236] Overall Loss 0.256921 Objective Loss 0.256921 LR 0.001000 Time 0.022735 +2023-10-02 21:03:04,438 - Epoch: [71][ 420/ 1236] Overall Loss 0.258120 Objective Loss 0.258120 LR 0.001000 Time 0.022687 +2023-10-02 21:03:04,643 - Epoch: [71][ 430/ 1236] Overall Loss 0.257953 Objective Loss 0.257953 LR 0.001000 Time 0.022636 +2023-10-02 21:03:04,849 - Epoch: [71][ 440/ 1236] Overall Loss 0.259256 Objective Loss 0.259256 LR 0.001000 Time 0.022589 +2023-10-02 21:03:05,055 - Epoch: [71][ 450/ 1236] Overall Loss 0.259434 Objective Loss 0.259434 LR 0.001000 Time 0.022542 +2023-10-02 21:03:05,262 - Epoch: [71][ 460/ 1236] Overall Loss 0.259559 Objective Loss 0.259559 LR 0.001000 Time 0.022502 +2023-10-02 21:03:05,467 - Epoch: [71][ 470/ 1236] Overall Loss 0.260087 Objective Loss 0.260087 LR 0.001000 Time 0.022458 +2023-10-02 21:03:05,673 - Epoch: [71][ 480/ 1236] Overall Loss 0.259772 Objective Loss 0.259772 LR 0.001000 Time 0.022419 +2023-10-02 21:03:05,879 - Epoch: [71][ 490/ 1236] Overall Loss 0.260589 Objective Loss 0.260589 LR 0.001000 Time 0.022379 +2023-10-02 21:03:06,085 - Epoch: [71][ 500/ 1236] Overall Loss 0.260163 Objective Loss 0.260163 LR 0.001000 Time 0.022343 +2023-10-02 21:03:06,291 - Epoch: [71][ 510/ 1236] Overall Loss 0.260010 Objective Loss 0.260010 LR 0.001000 Time 0.022306 +2023-10-02 21:03:06,497 - Epoch: [71][ 520/ 1236] Overall Loss 0.260150 Objective Loss 0.260150 LR 0.001000 Time 0.022273 +2023-10-02 21:03:06,704 - Epoch: [71][ 530/ 1236] Overall Loss 0.260266 Objective Loss 0.260266 LR 0.001000 Time 0.022239 +2023-10-02 21:03:06,910 - Epoch: [71][ 540/ 1236] Overall Loss 0.259797 Objective Loss 0.259797 LR 0.001000 Time 0.022208 +2023-10-02 21:03:07,116 - Epoch: [71][ 550/ 1236] Overall Loss 0.259571 Objective Loss 0.259571 LR 0.001000 Time 0.022177 +2023-10-02 21:03:07,322 - Epoch: [71][ 560/ 1236] Overall Loss 0.259652 Objective Loss 0.259652 LR 0.001000 Time 0.022149 +2023-10-02 21:03:07,528 - Epoch: [71][ 570/ 1236] Overall Loss 0.260520 Objective Loss 0.260520 LR 0.001000 Time 0.022119 +2023-10-02 21:03:07,734 - Epoch: [71][ 580/ 1236] Overall Loss 0.261417 Objective Loss 0.261417 LR 0.001000 Time 0.022092 +2023-10-02 21:03:07,940 - Epoch: [71][ 590/ 1236] Overall Loss 0.261540 Objective Loss 0.261540 LR 0.001000 Time 0.022064 +2023-10-02 21:03:08,146 - Epoch: [71][ 600/ 1236] Overall Loss 0.261761 Objective Loss 0.261761 LR 0.001000 Time 0.022039 +2023-10-02 21:03:08,352 - Epoch: [71][ 610/ 1236] Overall Loss 0.261901 Objective Loss 0.261901 LR 0.001000 Time 0.022014 +2023-10-02 21:03:08,560 - Epoch: [71][ 620/ 1236] Overall Loss 0.262094 Objective Loss 0.262094 LR 0.001000 Time 0.021993 +2023-10-02 21:03:08,765 - Epoch: [71][ 630/ 1236] Overall Loss 0.261824 Objective Loss 0.261824 LR 0.001000 Time 0.021969 +2023-10-02 21:03:08,972 - Epoch: [71][ 640/ 1236] Overall Loss 0.262194 Objective Loss 0.262194 LR 0.001000 Time 0.021949 +2023-10-02 21:03:09,177 - Epoch: [71][ 650/ 1236] Overall Loss 0.262080 Objective Loss 0.262080 LR 0.001000 Time 0.021926 +2023-10-02 21:03:09,383 - Epoch: [71][ 660/ 1236] Overall Loss 0.261898 Objective Loss 0.261898 LR 0.001000 Time 0.021905 +2023-10-02 21:03:09,589 - Epoch: [71][ 670/ 1236] Overall Loss 0.262014 Objective Loss 0.262014 LR 0.001000 Time 0.021884 +2023-10-02 21:03:09,795 - Epoch: [71][ 680/ 1236] Overall Loss 0.262266 Objective Loss 0.262266 LR 0.001000 Time 0.021865 +2023-10-02 21:03:10,001 - Epoch: [71][ 690/ 1236] Overall Loss 0.262885 Objective Loss 0.262885 LR 0.001000 Time 0.021844 +2023-10-02 21:03:10,209 - Epoch: [71][ 700/ 1236] Overall Loss 0.262972 Objective Loss 0.262972 LR 0.001000 Time 0.021828 +2023-10-02 21:03:10,413 - Epoch: [71][ 710/ 1236] Overall Loss 0.263266 Objective Loss 0.263266 LR 0.001000 Time 0.021808 +2023-10-02 21:03:10,619 - Epoch: [71][ 720/ 1236] Overall Loss 0.263279 Objective Loss 0.263279 LR 0.001000 Time 0.021791 +2023-10-02 21:03:10,825 - Epoch: [71][ 730/ 1236] Overall Loss 0.263267 Objective Loss 0.263267 LR 0.001000 Time 0.021773 +2023-10-02 21:03:11,033 - Epoch: [71][ 740/ 1236] Overall Loss 0.263225 Objective Loss 0.263225 LR 0.001000 Time 0.021759 +2023-10-02 21:03:11,238 - Epoch: [71][ 750/ 1236] Overall Loss 0.263452 Objective Loss 0.263452 LR 0.001000 Time 0.021741 +2023-10-02 21:03:11,445 - Epoch: [71][ 760/ 1236] Overall Loss 0.263837 Objective Loss 0.263837 LR 0.001000 Time 0.021728 +2023-10-02 21:03:11,650 - Epoch: [71][ 770/ 1236] Overall Loss 0.263694 Objective Loss 0.263694 LR 0.001000 Time 0.021711 +2023-10-02 21:03:11,857 - Epoch: [71][ 780/ 1236] Overall Loss 0.263356 Objective Loss 0.263356 LR 0.001000 Time 0.021698 +2023-10-02 21:03:12,064 - Epoch: [71][ 790/ 1236] Overall Loss 0.263963 Objective Loss 0.263963 LR 0.001000 Time 0.021683 +2023-10-02 21:03:12,272 - Epoch: [71][ 800/ 1236] Overall Loss 0.264667 Objective Loss 0.264667 LR 0.001000 Time 0.021672 +2023-10-02 21:03:12,478 - Epoch: [71][ 810/ 1236] Overall Loss 0.264656 Objective Loss 0.264656 LR 0.001000 Time 0.021657 +2023-10-02 21:03:12,686 - Epoch: [71][ 820/ 1236] Overall Loss 0.265315 Objective Loss 0.265315 LR 0.001000 Time 0.021646 +2023-10-02 21:03:12,893 - Epoch: [71][ 830/ 1236] Overall Loss 0.265262 Objective Loss 0.265262 LR 0.001000 Time 0.021633 +2023-10-02 21:03:13,101 - Epoch: [71][ 840/ 1236] Overall Loss 0.265125 Objective Loss 0.265125 LR 0.001000 Time 0.021622 +2023-10-02 21:03:13,308 - Epoch: [71][ 850/ 1236] Overall Loss 0.265574 Objective Loss 0.265574 LR 0.001000 Time 0.021610 +2023-10-02 21:03:13,516 - Epoch: [71][ 860/ 1236] Overall Loss 0.265714 Objective Loss 0.265714 LR 0.001000 Time 0.021601 +2023-10-02 21:03:13,722 - Epoch: [71][ 870/ 1236] Overall Loss 0.265790 Objective Loss 0.265790 LR 0.001000 Time 0.021587 +2023-10-02 21:03:13,930 - Epoch: [71][ 880/ 1236] Overall Loss 0.266135 Objective Loss 0.266135 LR 0.001000 Time 0.021578 +2023-10-02 21:03:14,137 - Epoch: [71][ 890/ 1236] Overall Loss 0.266080 Objective Loss 0.266080 LR 0.001000 Time 0.021567 +2023-10-02 21:03:14,346 - Epoch: [71][ 900/ 1236] Overall Loss 0.266581 Objective Loss 0.266581 LR 0.001000 Time 0.021559 +2023-10-02 21:03:14,554 - Epoch: [71][ 910/ 1236] Overall Loss 0.266896 Objective Loss 0.266896 LR 0.001000 Time 0.021548 +2023-10-02 21:03:14,762 - Epoch: [71][ 920/ 1236] Overall Loss 0.267204 Objective Loss 0.267204 LR 0.001000 Time 0.021540 +2023-10-02 21:03:14,970 - Epoch: [71][ 930/ 1236] Overall Loss 0.267129 Objective Loss 0.267129 LR 0.001000 Time 0.021530 +2023-10-02 21:03:15,178 - Epoch: [71][ 940/ 1236] Overall Loss 0.267243 Objective Loss 0.267243 LR 0.001000 Time 0.021522 +2023-10-02 21:03:15,385 - Epoch: [71][ 950/ 1236] Overall Loss 0.267291 Objective Loss 0.267291 LR 0.001000 Time 0.021512 +2023-10-02 21:03:15,594 - Epoch: [71][ 960/ 1236] Overall Loss 0.267623 Objective Loss 0.267623 LR 0.001000 Time 0.021505 +2023-10-02 21:03:15,801 - Epoch: [71][ 970/ 1236] Overall Loss 0.267251 Objective Loss 0.267251 LR 0.001000 Time 0.021496 +2023-10-02 21:03:16,010 - Epoch: [71][ 980/ 1236] Overall Loss 0.267429 Objective Loss 0.267429 LR 0.001000 Time 0.021489 +2023-10-02 21:03:16,217 - Epoch: [71][ 990/ 1236] Overall Loss 0.267143 Objective Loss 0.267143 LR 0.001000 Time 0.021480 +2023-10-02 21:03:16,426 - Epoch: [71][ 1000/ 1236] Overall Loss 0.267151 Objective Loss 0.267151 LR 0.001000 Time 0.021473 +2023-10-02 21:03:16,633 - Epoch: [71][ 1010/ 1236] Overall Loss 0.267188 Objective Loss 0.267188 LR 0.001000 Time 0.021464 +2023-10-02 21:03:16,841 - Epoch: [71][ 1020/ 1236] Overall Loss 0.267437 Objective Loss 0.267437 LR 0.001000 Time 0.021458 +2023-10-02 21:03:17,048 - Epoch: [71][ 1030/ 1236] Overall Loss 0.267628 Objective Loss 0.267628 LR 0.001000 Time 0.021449 +2023-10-02 21:03:17,255 - Epoch: [71][ 1040/ 1236] Overall Loss 0.267607 Objective Loss 0.267607 LR 0.001000 Time 0.021442 +2023-10-02 21:03:17,461 - Epoch: [71][ 1050/ 1236] Overall Loss 0.267752 Objective Loss 0.267752 LR 0.001000 Time 0.021432 +2023-10-02 21:03:17,667 - Epoch: [71][ 1060/ 1236] Overall Loss 0.267661 Objective Loss 0.267661 LR 0.001000 Time 0.021424 +2023-10-02 21:03:17,874 - Epoch: [71][ 1070/ 1236] Overall Loss 0.268073 Objective Loss 0.268073 LR 0.001000 Time 0.021415 +2023-10-02 21:03:18,081 - Epoch: [71][ 1080/ 1236] Overall Loss 0.268302 Objective Loss 0.268302 LR 0.001000 Time 0.021409 +2023-10-02 21:03:18,286 - Epoch: [71][ 1090/ 1236] Overall Loss 0.268236 Objective Loss 0.268236 LR 0.001000 Time 0.021400 +2023-10-02 21:03:18,494 - Epoch: [71][ 1100/ 1236] Overall Loss 0.268593 Objective Loss 0.268593 LR 0.001000 Time 0.021394 +2023-10-02 21:03:18,699 - Epoch: [71][ 1110/ 1236] Overall Loss 0.268581 Objective Loss 0.268581 LR 0.001000 Time 0.021386 +2023-10-02 21:03:18,905 - Epoch: [71][ 1120/ 1236] Overall Loss 0.268441 Objective Loss 0.268441 LR 0.001000 Time 0.021379 +2023-10-02 21:03:19,111 - Epoch: [71][ 1130/ 1236] Overall Loss 0.268556 Objective Loss 0.268556 LR 0.001000 Time 0.021371 +2023-10-02 21:03:19,319 - Epoch: [71][ 1140/ 1236] Overall Loss 0.268683 Objective Loss 0.268683 LR 0.001000 Time 0.021365 +2023-10-02 21:03:19,524 - Epoch: [71][ 1150/ 1236] Overall Loss 0.268695 Objective Loss 0.268695 LR 0.001000 Time 0.021358 +2023-10-02 21:03:19,730 - Epoch: [71][ 1160/ 1236] Overall Loss 0.268685 Objective Loss 0.268685 LR 0.001000 Time 0.021351 +2023-10-02 21:03:19,936 - Epoch: [71][ 1170/ 1236] Overall Loss 0.268905 Objective Loss 0.268905 LR 0.001000 Time 0.021343 +2023-10-02 21:03:20,142 - Epoch: [71][ 1180/ 1236] Overall Loss 0.269090 Objective Loss 0.269090 LR 0.001000 Time 0.021337 +2023-10-02 21:03:20,348 - Epoch: [71][ 1190/ 1236] Overall Loss 0.268885 Objective Loss 0.268885 LR 0.001000 Time 0.021329 +2023-10-02 21:03:20,556 - Epoch: [71][ 1200/ 1236] Overall Loss 0.268771 Objective Loss 0.268771 LR 0.001000 Time 0.021324 +2023-10-02 21:03:20,761 - Epoch: [71][ 1210/ 1236] Overall Loss 0.268806 Objective Loss 0.268806 LR 0.001000 Time 0.021317 +2023-10-02 21:03:20,968 - Epoch: [71][ 1220/ 1236] Overall Loss 0.269112 Objective Loss 0.269112 LR 0.001000 Time 0.021312 +2023-10-02 21:03:21,225 - Epoch: [71][ 1230/ 1236] Overall Loss 0.269255 Objective Loss 0.269255 LR 0.001000 Time 0.021348 +2023-10-02 21:03:21,347 - Epoch: [71][ 1236/ 1236] Overall Loss 0.269333 Objective Loss 0.269333 Top1 86.354379 Top5 99.592668 LR 0.001000 Time 0.021342 +2023-10-02 21:03:21,495 - --- validate (epoch=71)----------- +2023-10-02 21:03:21,495 - 29943 samples (256 per mini-batch) +2023-10-02 21:03:21,990 - Epoch: [71][ 10/ 117] Loss 0.333884 Top1 82.734375 Top5 97.773438 +2023-10-02 21:03:22,141 - Epoch: [71][ 20/ 117] Loss 0.321669 Top1 83.125000 Top5 98.027344 +2023-10-02 21:03:22,290 - Epoch: [71][ 30/ 117] Loss 0.325455 Top1 83.059896 Top5 98.072917 +2023-10-02 21:03:22,441 - Epoch: [71][ 40/ 117] Loss 0.337654 Top1 82.890625 Top5 98.037109 +2023-10-02 21:03:22,589 - Epoch: [71][ 50/ 117] Loss 0.335443 Top1 83.289062 Top5 98.070312 +2023-10-02 21:03:22,739 - Epoch: [71][ 60/ 117] Loss 0.335951 Top1 83.404948 Top5 98.085938 +2023-10-02 21:03:22,889 - Epoch: [71][ 70/ 117] Loss 0.335537 Top1 83.437500 Top5 98.136161 +2023-10-02 21:03:23,039 - Epoch: [71][ 80/ 117] Loss 0.338609 Top1 83.520508 Top5 98.139648 +2023-10-02 21:03:23,189 - Epoch: [71][ 90/ 117] Loss 0.340009 Top1 83.355035 Top5 98.172743 +2023-10-02 21:03:23,339 - Epoch: [71][ 100/ 117] Loss 0.337125 Top1 83.382812 Top5 98.183594 +2023-10-02 21:03:23,497 - Epoch: [71][ 110/ 117] Loss 0.335083 Top1 83.444602 Top5 98.210227 +2023-10-02 21:03:23,585 - Epoch: [71][ 117/ 117] Loss 0.337171 Top1 83.401797 Top5 98.189894 +2023-10-02 21:03:23,722 - ==> Top1: 83.402 Top5: 98.190 Loss: 0.337 + +2023-10-02 21:03:23,722 - ==> Confusion: +[[ 932 1 1 0 5 4 0 0 9 70 2 0 0 3 8 0 1 3 0 0 11] + [ 2 1042 0 0 6 27 1 19 4 0 3 1 0 0 2 3 1 0 11 4 5] + [ 7 0 963 4 1 0 22 12 0 5 3 2 7 2 2 5 2 1 7 4 7] + [ 0 5 16 951 0 5 2 5 8 1 3 0 13 3 39 3 0 2 17 1 15] + [ 35 5 1 0 949 6 0 0 1 18 0 1 0 3 6 6 7 0 0 3 9] + [ 1 48 0 1 2 970 2 31 3 5 3 6 2 12 4 1 3 1 2 7 12] + [ 0 6 29 0 0 2 1110 9 0 0 1 4 1 0 1 9 2 0 0 10 7] + [ 4 23 15 0 3 21 1 1066 1 5 3 6 1 2 4 0 3 0 40 9 11] + [ 14 2 2 0 0 0 0 1 983 53 4 0 0 10 9 0 4 2 1 1 3] + [ 84 0 0 0 4 0 0 0 35 961 2 1 0 15 5 0 1 2 0 2 7] + [ 2 1 10 9 0 2 1 5 33 1 945 2 1 14 5 0 2 0 11 2 7] + [ 1 1 1 1 4 11 0 3 0 0 1 955 24 7 0 1 0 13 0 8 4] + [ 0 1 3 4 0 0 1 1 4 4 4 54 946 1 4 11 4 8 2 7 9] + [ 1 0 0 0 0 5 0 2 19 10 5 8 0 1049 4 0 2 1 0 2 11] + [ 19 3 2 16 5 1 0 0 34 2 0 0 4 3 992 0 1 5 6 1 7] + [ 0 0 1 2 3 0 0 0 0 1 0 8 9 1 1 1064 20 9 3 7 5] + [ 1 18 1 1 9 6 0 1 1 0 0 9 2 2 3 12 1086 0 1 1 7] + [ 2 0 0 2 1 3 4 0 1 2 0 13 35 1 5 4 1 959 0 0 5] + [ 2 4 5 7 2 0 0 20 4 0 0 0 0 0 22 0 0 0 992 0 10] + [ 0 4 3 1 0 7 5 14 0 1 1 13 4 3 0 6 4 1 0 1077 8] + [ 169 182 130 69 104 156 34 108 152 124 141 158 360 324 140 70 90 63 158 192 4981]] + +2023-10-02 21:03:23,723 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:03:23,724 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:03:23,730 - + +2023-10-02 21:03:23,730 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:03:24,726 - Epoch: [72][ 10/ 1236] Overall Loss 0.265709 Objective Loss 0.265709 LR 0.001000 Time 0.099543 +2023-10-02 21:03:24,931 - Epoch: [72][ 20/ 1236] Overall Loss 0.255715 Objective Loss 0.255715 LR 0.001000 Time 0.060034 +2023-10-02 21:03:25,137 - Epoch: [72][ 30/ 1236] Overall Loss 0.258235 Objective Loss 0.258235 LR 0.001000 Time 0.046880 +2023-10-02 21:03:25,344 - Epoch: [72][ 40/ 1236] Overall Loss 0.261036 Objective Loss 0.261036 LR 0.001000 Time 0.040322 +2023-10-02 21:03:25,548 - Epoch: [72][ 50/ 1236] Overall Loss 0.251679 Objective Loss 0.251679 LR 0.001000 Time 0.036337 +2023-10-02 21:03:25,755 - Epoch: [72][ 60/ 1236] Overall Loss 0.249674 Objective Loss 0.249674 LR 0.001000 Time 0.033727 +2023-10-02 21:03:25,959 - Epoch: [72][ 70/ 1236] Overall Loss 0.249801 Objective Loss 0.249801 LR 0.001000 Time 0.031817 +2023-10-02 21:03:26,166 - Epoch: [72][ 80/ 1236] Overall Loss 0.246551 Objective Loss 0.246551 LR 0.001000 Time 0.030425 +2023-10-02 21:03:26,370 - Epoch: [72][ 90/ 1236] Overall Loss 0.247640 Objective Loss 0.247640 LR 0.001000 Time 0.029309 +2023-10-02 21:03:26,577 - Epoch: [72][ 100/ 1236] Overall Loss 0.250672 Objective Loss 0.250672 LR 0.001000 Time 0.028442 +2023-10-02 21:03:26,787 - Epoch: [72][ 110/ 1236] Overall Loss 0.248449 Objective Loss 0.248449 LR 0.001000 Time 0.027760 +2023-10-02 21:03:26,993 - Epoch: [72][ 120/ 1236] Overall Loss 0.250005 Objective Loss 0.250005 LR 0.001000 Time 0.027163 +2023-10-02 21:03:27,202 - Epoch: [72][ 130/ 1236] Overall Loss 0.251733 Objective Loss 0.251733 LR 0.001000 Time 0.026679 +2023-10-02 21:03:27,410 - Epoch: [72][ 140/ 1236] Overall Loss 0.253493 Objective Loss 0.253493 LR 0.001000 Time 0.026248 +2023-10-02 21:03:27,620 - Epoch: [72][ 150/ 1236] Overall Loss 0.255048 Objective Loss 0.255048 LR 0.001000 Time 0.025895 +2023-10-02 21:03:27,835 - Epoch: [72][ 160/ 1236] Overall Loss 0.254900 Objective Loss 0.254900 LR 0.001000 Time 0.025611 +2023-10-02 21:03:28,048 - Epoch: [72][ 170/ 1236] Overall Loss 0.256118 Objective Loss 0.256118 LR 0.001000 Time 0.025352 +2023-10-02 21:03:28,263 - Epoch: [72][ 180/ 1236] Overall Loss 0.255203 Objective Loss 0.255203 LR 0.001000 Time 0.025138 +2023-10-02 21:03:28,476 - Epoch: [72][ 190/ 1236] Overall Loss 0.254792 Objective Loss 0.254792 LR 0.001000 Time 0.024932 +2023-10-02 21:03:28,691 - Epoch: [72][ 200/ 1236] Overall Loss 0.254456 Objective Loss 0.254456 LR 0.001000 Time 0.024753 +2023-10-02 21:03:28,903 - Epoch: [72][ 210/ 1236] Overall Loss 0.255272 Objective Loss 0.255272 LR 0.001000 Time 0.024582 +2023-10-02 21:03:29,119 - Epoch: [72][ 220/ 1236] Overall Loss 0.253179 Objective Loss 0.253179 LR 0.001000 Time 0.024442 +2023-10-02 21:03:29,331 - Epoch: [72][ 230/ 1236] Overall Loss 0.252128 Objective Loss 0.252128 LR 0.001000 Time 0.024301 +2023-10-02 21:03:29,546 - Epoch: [72][ 240/ 1236] Overall Loss 0.251069 Objective Loss 0.251069 LR 0.001000 Time 0.024183 +2023-10-02 21:03:29,758 - Epoch: [72][ 250/ 1236] Overall Loss 0.251442 Objective Loss 0.251442 LR 0.001000 Time 0.024062 +2023-10-02 21:03:29,972 - Epoch: [72][ 260/ 1236] Overall Loss 0.251586 Objective Loss 0.251586 LR 0.001000 Time 0.023955 +2023-10-02 21:03:30,186 - Epoch: [72][ 270/ 1236] Overall Loss 0.252309 Objective Loss 0.252309 LR 0.001000 Time 0.023859 +2023-10-02 21:03:30,399 - Epoch: [72][ 280/ 1236] Overall Loss 0.251842 Objective Loss 0.251842 LR 0.001000 Time 0.023765 +2023-10-02 21:03:30,613 - Epoch: [72][ 290/ 1236] Overall Loss 0.251197 Objective Loss 0.251197 LR 0.001000 Time 0.023681 +2023-10-02 21:03:30,827 - Epoch: [72][ 300/ 1236] Overall Loss 0.251662 Objective Loss 0.251662 LR 0.001000 Time 0.023605 +2023-10-02 21:03:31,040 - Epoch: [72][ 310/ 1236] Overall Loss 0.252444 Objective Loss 0.252444 LR 0.001000 Time 0.023528 +2023-10-02 21:03:31,255 - Epoch: [72][ 320/ 1236] Overall Loss 0.251881 Objective Loss 0.251881 LR 0.001000 Time 0.023461 +2023-10-02 21:03:31,467 - Epoch: [72][ 330/ 1236] Overall Loss 0.250821 Objective Loss 0.250821 LR 0.001000 Time 0.023392 +2023-10-02 21:03:31,682 - Epoch: [72][ 340/ 1236] Overall Loss 0.249821 Objective Loss 0.249821 LR 0.001000 Time 0.023335 +2023-10-02 21:03:31,894 - Epoch: [72][ 350/ 1236] Overall Loss 0.249444 Objective Loss 0.249444 LR 0.001000 Time 0.023273 +2023-10-02 21:03:32,107 - Epoch: [72][ 360/ 1236] Overall Loss 0.250744 Objective Loss 0.250744 LR 0.001000 Time 0.023213 +2023-10-02 21:03:32,320 - Epoch: [72][ 370/ 1236] Overall Loss 0.250356 Objective Loss 0.250356 LR 0.001000 Time 0.023157 +2023-10-02 21:03:32,532 - Epoch: [72][ 380/ 1236] Overall Loss 0.250649 Objective Loss 0.250649 LR 0.001000 Time 0.023103 +2023-10-02 21:03:32,746 - Epoch: [72][ 390/ 1236] Overall Loss 0.249872 Objective Loss 0.249872 LR 0.001000 Time 0.023053 +2023-10-02 21:03:32,958 - Epoch: [72][ 400/ 1236] Overall Loss 0.250543 Objective Loss 0.250543 LR 0.001000 Time 0.023004 +2023-10-02 21:03:33,171 - Epoch: [72][ 410/ 1236] Overall Loss 0.250631 Objective Loss 0.250631 LR 0.001000 Time 0.022958 +2023-10-02 21:03:33,384 - Epoch: [72][ 420/ 1236] Overall Loss 0.251189 Objective Loss 0.251189 LR 0.001000 Time 0.022916 +2023-10-02 21:03:33,597 - Epoch: [72][ 430/ 1236] Overall Loss 0.251422 Objective Loss 0.251422 LR 0.001000 Time 0.022879 +2023-10-02 21:03:33,808 - Epoch: [72][ 440/ 1236] Overall Loss 0.251565 Objective Loss 0.251565 LR 0.001000 Time 0.022835 +2023-10-02 21:03:34,020 - Epoch: [72][ 450/ 1236] Overall Loss 0.252172 Objective Loss 0.252172 LR 0.001000 Time 0.022798 +2023-10-02 21:03:34,231 - Epoch: [72][ 460/ 1236] Overall Loss 0.252526 Objective Loss 0.252526 LR 0.001000 Time 0.022757 +2023-10-02 21:03:34,443 - Epoch: [72][ 470/ 1236] Overall Loss 0.252314 Objective Loss 0.252314 LR 0.001000 Time 0.022723 +2023-10-02 21:03:34,654 - Epoch: [72][ 480/ 1236] Overall Loss 0.252715 Objective Loss 0.252715 LR 0.001000 Time 0.022688 +2023-10-02 21:03:34,866 - Epoch: [72][ 490/ 1236] Overall Loss 0.252906 Objective Loss 0.252906 LR 0.001000 Time 0.022654 +2023-10-02 21:03:35,077 - Epoch: [72][ 500/ 1236] Overall Loss 0.253384 Objective Loss 0.253384 LR 0.001000 Time 0.022622 +2023-10-02 21:03:35,289 - Epoch: [72][ 510/ 1236] Overall Loss 0.253460 Objective Loss 0.253460 LR 0.001000 Time 0.022593 +2023-10-02 21:03:35,501 - Epoch: [72][ 520/ 1236] Overall Loss 0.254094 Objective Loss 0.254094 LR 0.001000 Time 0.022565 +2023-10-02 21:03:35,713 - Epoch: [72][ 530/ 1236] Overall Loss 0.253722 Objective Loss 0.253722 LR 0.001000 Time 0.022538 +2023-10-02 21:03:35,924 - Epoch: [72][ 540/ 1236] Overall Loss 0.253568 Objective Loss 0.253568 LR 0.001000 Time 0.022511 +2023-10-02 21:03:36,136 - Epoch: [72][ 550/ 1236] Overall Loss 0.254122 Objective Loss 0.254122 LR 0.001000 Time 0.022487 +2023-10-02 21:03:36,348 - Epoch: [72][ 560/ 1236] Overall Loss 0.254120 Objective Loss 0.254120 LR 0.001000 Time 0.022462 +2023-10-02 21:03:36,559 - Epoch: [72][ 570/ 1236] Overall Loss 0.253752 Objective Loss 0.253752 LR 0.001000 Time 0.022438 +2023-10-02 21:03:36,771 - Epoch: [72][ 580/ 1236] Overall Loss 0.253942 Objective Loss 0.253942 LR 0.001000 Time 0.022414 +2023-10-02 21:03:36,982 - Epoch: [72][ 590/ 1236] Overall Loss 0.254510 Objective Loss 0.254510 LR 0.001000 Time 0.022390 +2023-10-02 21:03:37,193 - Epoch: [72][ 600/ 1236] Overall Loss 0.255104 Objective Loss 0.255104 LR 0.001000 Time 0.022368 +2023-10-02 21:03:37,405 - Epoch: [72][ 610/ 1236] Overall Loss 0.255458 Objective Loss 0.255458 LR 0.001000 Time 0.022348 +2023-10-02 21:03:37,616 - Epoch: [72][ 620/ 1236] Overall Loss 0.255500 Objective Loss 0.255500 LR 0.001000 Time 0.022327 +2023-10-02 21:03:37,828 - Epoch: [72][ 630/ 1236] Overall Loss 0.255910 Objective Loss 0.255910 LR 0.001000 Time 0.022309 +2023-10-02 21:03:38,039 - Epoch: [72][ 640/ 1236] Overall Loss 0.256211 Objective Loss 0.256211 LR 0.001000 Time 0.022290 +2023-10-02 21:03:38,252 - Epoch: [72][ 650/ 1236] Overall Loss 0.256344 Objective Loss 0.256344 LR 0.001000 Time 0.022273 +2023-10-02 21:03:38,463 - Epoch: [72][ 660/ 1236] Overall Loss 0.256301 Objective Loss 0.256301 LR 0.001000 Time 0.022255 +2023-10-02 21:03:38,676 - Epoch: [72][ 670/ 1236] Overall Loss 0.256552 Objective Loss 0.256552 LR 0.001000 Time 0.022239 +2023-10-02 21:03:38,887 - Epoch: [72][ 680/ 1236] Overall Loss 0.256878 Objective Loss 0.256878 LR 0.001000 Time 0.022222 +2023-10-02 21:03:39,099 - Epoch: [72][ 690/ 1236] Overall Loss 0.257134 Objective Loss 0.257134 LR 0.001000 Time 0.022206 +2023-10-02 21:03:39,310 - Epoch: [72][ 700/ 1236] Overall Loss 0.257562 Objective Loss 0.257562 LR 0.001000 Time 0.022190 +2023-10-02 21:03:39,521 - Epoch: [72][ 710/ 1236] Overall Loss 0.257473 Objective Loss 0.257473 LR 0.001000 Time 0.022173 +2023-10-02 21:03:39,733 - Epoch: [72][ 720/ 1236] Overall Loss 0.257651 Objective Loss 0.257651 LR 0.001000 Time 0.022157 +2023-10-02 21:03:39,945 - Epoch: [72][ 730/ 1236] Overall Loss 0.257646 Objective Loss 0.257646 LR 0.001000 Time 0.022144 +2023-10-02 21:03:40,156 - Epoch: [72][ 740/ 1236] Overall Loss 0.258238 Objective Loss 0.258238 LR 0.001000 Time 0.022129 +2023-10-02 21:03:40,367 - Epoch: [72][ 750/ 1236] Overall Loss 0.258295 Objective Loss 0.258295 LR 0.001000 Time 0.022114 +2023-10-02 21:03:40,579 - Epoch: [72][ 760/ 1236] Overall Loss 0.258153 Objective Loss 0.258153 LR 0.001000 Time 0.022100 +2023-10-02 21:03:40,791 - Epoch: [72][ 770/ 1236] Overall Loss 0.258413 Objective Loss 0.258413 LR 0.001000 Time 0.022087 +2023-10-02 21:03:41,002 - Epoch: [72][ 780/ 1236] Overall Loss 0.259021 Objective Loss 0.259021 LR 0.001000 Time 0.022074 +2023-10-02 21:03:41,214 - Epoch: [72][ 790/ 1236] Overall Loss 0.258761 Objective Loss 0.258761 LR 0.001000 Time 0.022063 +2023-10-02 21:03:41,426 - Epoch: [72][ 800/ 1236] Overall Loss 0.258679 Objective Loss 0.258679 LR 0.001000 Time 0.022049 +2023-10-02 21:03:41,638 - Epoch: [72][ 810/ 1236] Overall Loss 0.259267 Objective Loss 0.259267 LR 0.001000 Time 0.022038 +2023-10-02 21:03:41,849 - Epoch: [72][ 820/ 1236] Overall Loss 0.259249 Objective Loss 0.259249 LR 0.001000 Time 0.022026 +2023-10-02 21:03:42,061 - Epoch: [72][ 830/ 1236] Overall Loss 0.259533 Objective Loss 0.259533 LR 0.001000 Time 0.022016 +2023-10-02 21:03:42,272 - Epoch: [72][ 840/ 1236] Overall Loss 0.259542 Objective Loss 0.259542 LR 0.001000 Time 0.022003 +2023-10-02 21:03:42,485 - Epoch: [72][ 850/ 1236] Overall Loss 0.259744 Objective Loss 0.259744 LR 0.001000 Time 0.021994 +2023-10-02 21:03:42,696 - Epoch: [72][ 860/ 1236] Overall Loss 0.259652 Objective Loss 0.259652 LR 0.001000 Time 0.021983 +2023-10-02 21:03:42,908 - Epoch: [72][ 870/ 1236] Overall Loss 0.259834 Objective Loss 0.259834 LR 0.001000 Time 0.021973 +2023-10-02 21:03:43,119 - Epoch: [72][ 880/ 1236] Overall Loss 0.260104 Objective Loss 0.260104 LR 0.001000 Time 0.021963 +2023-10-02 21:03:43,331 - Epoch: [72][ 890/ 1236] Overall Loss 0.260017 Objective Loss 0.260017 LR 0.001000 Time 0.021952 +2023-10-02 21:03:43,542 - Epoch: [72][ 900/ 1236] Overall Loss 0.260191 Objective Loss 0.260191 LR 0.001000 Time 0.021943 +2023-10-02 21:03:43,754 - Epoch: [72][ 910/ 1236] Overall Loss 0.259995 Objective Loss 0.259995 LR 0.001000 Time 0.021934 +2023-10-02 21:03:43,966 - Epoch: [72][ 920/ 1236] Overall Loss 0.260089 Objective Loss 0.260089 LR 0.001000 Time 0.021925 +2023-10-02 21:03:44,178 - Epoch: [72][ 930/ 1236] Overall Loss 0.260432 Objective Loss 0.260432 LR 0.001000 Time 0.021917 +2023-10-02 21:03:44,389 - Epoch: [72][ 940/ 1236] Overall Loss 0.260530 Objective Loss 0.260530 LR 0.001000 Time 0.021908 +2023-10-02 21:03:44,602 - Epoch: [72][ 950/ 1236] Overall Loss 0.260757 Objective Loss 0.260757 LR 0.001000 Time 0.021900 +2023-10-02 21:03:44,813 - Epoch: [72][ 960/ 1236] Overall Loss 0.260937 Objective Loss 0.260937 LR 0.001000 Time 0.021891 +2023-10-02 21:03:45,025 - Epoch: [72][ 970/ 1236] Overall Loss 0.261213 Objective Loss 0.261213 LR 0.001000 Time 0.021883 +2023-10-02 21:03:45,236 - Epoch: [72][ 980/ 1236] Overall Loss 0.261166 Objective Loss 0.261166 LR 0.001000 Time 0.021875 +2023-10-02 21:03:45,448 - Epoch: [72][ 990/ 1236] Overall Loss 0.261329 Objective Loss 0.261329 LR 0.001000 Time 0.021866 +2023-10-02 21:03:45,659 - Epoch: [72][ 1000/ 1236] Overall Loss 0.261456 Objective Loss 0.261456 LR 0.001000 Time 0.021858 +2023-10-02 21:03:45,871 - Epoch: [72][ 1010/ 1236] Overall Loss 0.261904 Objective Loss 0.261904 LR 0.001000 Time 0.021851 +2023-10-02 21:03:46,082 - Epoch: [72][ 1020/ 1236] Overall Loss 0.262183 Objective Loss 0.262183 LR 0.001000 Time 0.021842 +2023-10-02 21:03:46,294 - Epoch: [72][ 1030/ 1236] Overall Loss 0.262282 Objective Loss 0.262282 LR 0.001000 Time 0.021835 +2023-10-02 21:03:46,505 - Epoch: [72][ 1040/ 1236] Overall Loss 0.262302 Objective Loss 0.262302 LR 0.001000 Time 0.021827 +2023-10-02 21:03:46,717 - Epoch: [72][ 1050/ 1236] Overall Loss 0.262345 Objective Loss 0.262345 LR 0.001000 Time 0.021820 +2023-10-02 21:03:46,928 - Epoch: [72][ 1060/ 1236] Overall Loss 0.262332 Objective Loss 0.262332 LR 0.001000 Time 0.021812 +2023-10-02 21:03:47,140 - Epoch: [72][ 1070/ 1236] Overall Loss 0.262279 Objective Loss 0.262279 LR 0.001000 Time 0.021805 +2023-10-02 21:03:47,351 - Epoch: [72][ 1080/ 1236] Overall Loss 0.262365 Objective Loss 0.262365 LR 0.001000 Time 0.021799 +2023-10-02 21:03:47,563 - Epoch: [72][ 1090/ 1236] Overall Loss 0.262412 Objective Loss 0.262412 LR 0.001000 Time 0.021793 +2023-10-02 21:03:47,775 - Epoch: [72][ 1100/ 1236] Overall Loss 0.262524 Objective Loss 0.262524 LR 0.001000 Time 0.021786 +2023-10-02 21:03:47,987 - Epoch: [72][ 1110/ 1236] Overall Loss 0.262795 Objective Loss 0.262795 LR 0.001000 Time 0.021780 +2023-10-02 21:03:48,198 - Epoch: [72][ 1120/ 1236] Overall Loss 0.263248 Objective Loss 0.263248 LR 0.001000 Time 0.021773 +2023-10-02 21:03:48,411 - Epoch: [72][ 1130/ 1236] Overall Loss 0.263166 Objective Loss 0.263166 LR 0.001000 Time 0.021767 +2023-10-02 21:03:48,622 - Epoch: [72][ 1140/ 1236] Overall Loss 0.262978 Objective Loss 0.262978 LR 0.001000 Time 0.021761 +2023-10-02 21:03:48,834 - Epoch: [72][ 1150/ 1236] Overall Loss 0.262862 Objective Loss 0.262862 LR 0.001000 Time 0.021756 +2023-10-02 21:03:49,046 - Epoch: [72][ 1160/ 1236] Overall Loss 0.263033 Objective Loss 0.263033 LR 0.001000 Time 0.021750 +2023-10-02 21:03:49,258 - Epoch: [72][ 1170/ 1236] Overall Loss 0.263123 Objective Loss 0.263123 LR 0.001000 Time 0.021745 +2023-10-02 21:03:49,469 - Epoch: [72][ 1180/ 1236] Overall Loss 0.263513 Objective Loss 0.263513 LR 0.001000 Time 0.021738 +2023-10-02 21:03:49,681 - Epoch: [72][ 1190/ 1236] Overall Loss 0.263543 Objective Loss 0.263543 LR 0.001000 Time 0.021733 +2023-10-02 21:03:49,891 - Epoch: [72][ 1200/ 1236] Overall Loss 0.263712 Objective Loss 0.263712 LR 0.001000 Time 0.021727 +2023-10-02 21:03:50,104 - Epoch: [72][ 1210/ 1236] Overall Loss 0.263943 Objective Loss 0.263943 LR 0.001000 Time 0.021723 +2023-10-02 21:03:50,315 - Epoch: [72][ 1220/ 1236] Overall Loss 0.263977 Objective Loss 0.263977 LR 0.001000 Time 0.021716 +2023-10-02 21:03:50,580 - Epoch: [72][ 1230/ 1236] Overall Loss 0.263938 Objective Loss 0.263938 LR 0.001000 Time 0.021755 +2023-10-02 21:03:50,703 - Epoch: [72][ 1236/ 1236] Overall Loss 0.263908 Objective Loss 0.263908 Top1 86.354379 Top5 98.574338 LR 0.001000 Time 0.021749 +2023-10-02 21:03:50,835 - --- validate (epoch=72)----------- +2023-10-02 21:03:50,835 - 29943 samples (256 per mini-batch) +2023-10-02 21:03:51,339 - Epoch: [72][ 10/ 117] Loss 0.303566 Top1 83.867188 Top5 98.671875 +2023-10-02 21:03:51,485 - Epoch: [72][ 20/ 117] Loss 0.326254 Top1 83.671875 Top5 98.339844 +2023-10-02 21:03:51,634 - Epoch: [72][ 30/ 117] Loss 0.337606 Top1 83.515625 Top5 98.177083 +2023-10-02 21:03:51,784 - Epoch: [72][ 40/ 117] Loss 0.336337 Top1 83.554688 Top5 98.232422 +2023-10-02 21:03:51,931 - Epoch: [72][ 50/ 117] Loss 0.342747 Top1 83.414062 Top5 98.242188 +2023-10-02 21:03:52,078 - Epoch: [72][ 60/ 117] Loss 0.336507 Top1 83.444010 Top5 98.274740 +2023-10-02 21:03:52,225 - Epoch: [72][ 70/ 117] Loss 0.334073 Top1 83.454241 Top5 98.286830 +2023-10-02 21:03:52,374 - Epoch: [72][ 80/ 117] Loss 0.336361 Top1 83.603516 Top5 98.247070 +2023-10-02 21:03:52,524 - Epoch: [72][ 90/ 117] Loss 0.338644 Top1 83.559028 Top5 98.237847 +2023-10-02 21:03:52,673 - Epoch: [72][ 100/ 117] Loss 0.337934 Top1 83.546875 Top5 98.273438 +2023-10-02 21:03:52,831 - Epoch: [72][ 110/ 117] Loss 0.335912 Top1 83.487216 Top5 98.263494 +2023-10-02 21:03:52,921 - Epoch: [72][ 117/ 117] Loss 0.334991 Top1 83.485289 Top5 98.229970 +2023-10-02 21:03:53,038 - ==> Top1: 83.485 Top5: 98.230 Loss: 0.335 + +2023-10-02 21:03:53,039 - ==> Confusion: +[[ 956 1 6 1 7 3 0 1 7 37 2 0 1 3 6 2 1 1 0 0 15] + [ 0 1028 2 2 9 35 0 27 1 1 0 1 0 0 1 3 1 1 8 2 9] + [ 3 0 966 9 3 0 27 11 0 0 4 0 7 1 3 7 2 1 6 1 5] + [ 1 2 19 975 0 2 0 2 2 0 5 0 7 2 35 1 1 3 14 1 17] + [ 31 5 2 2 963 5 1 0 0 11 1 0 1 3 4 5 7 1 1 1 6] + [ 4 38 1 2 2 962 0 42 2 3 1 6 1 12 7 2 4 1 4 5 17] + [ 0 3 30 0 0 1 1125 7 0 0 7 2 0 1 0 7 0 0 1 4 3] + [ 1 15 17 3 5 23 4 1066 3 1 2 2 4 3 4 1 5 1 46 5 7] + [ 27 6 1 1 1 3 0 2 939 56 9 3 1 14 15 2 2 2 2 0 3] + [ 153 2 1 1 10 0 0 0 28 882 2 3 0 14 12 1 1 0 2 0 7] + [ 3 0 16 4 1 3 1 8 22 1 956 1 0 12 6 1 3 1 7 0 7] + [ 2 3 1 0 0 12 0 6 0 0 0 924 36 10 1 4 2 14 1 12 7] + [ 0 0 6 4 0 1 2 3 1 0 4 35 959 5 2 9 1 15 5 7 9] + [ 2 0 5 0 5 9 1 2 11 17 5 6 1 1032 7 0 0 0 0 1 15] + [ 13 0 4 21 4 1 0 0 14 1 0 0 2 0 1026 0 3 2 5 0 5] + [ 0 0 2 1 5 2 2 1 0 0 0 7 6 0 1 1076 12 7 1 5 6] + [ 1 14 1 2 11 8 0 1 2 0 0 4 1 0 3 10 1085 0 3 4 11] + [ 1 0 1 9 0 0 2 0 2 0 0 6 34 1 3 18 2 956 1 0 2] + [ 4 3 9 20 1 0 0 19 2 1 4 0 0 0 14 0 0 0 979 0 12] + [ 0 4 5 3 2 2 12 20 0 0 2 7 5 4 1 3 5 0 2 1065 10] + [ 157 154 177 102 118 146 67 121 99 65 141 94 352 291 184 68 114 41 164 172 5078]] + +2023-10-02 21:03:53,040 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:03:53,040 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:03:53,047 - + +2023-10-02 21:03:53,047 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:03:54,092 - Epoch: [73][ 10/ 1236] Overall Loss 0.252944 Objective Loss 0.252944 LR 0.001000 Time 0.104527 +2023-10-02 21:03:54,304 - Epoch: [73][ 20/ 1236] Overall Loss 0.244291 Objective Loss 0.244291 LR 0.001000 Time 0.062799 +2023-10-02 21:03:54,516 - Epoch: [73][ 30/ 1236] Overall Loss 0.257771 Objective Loss 0.257771 LR 0.001000 Time 0.048922 +2023-10-02 21:03:54,727 - Epoch: [73][ 40/ 1236] Overall Loss 0.258229 Objective Loss 0.258229 LR 0.001000 Time 0.041963 +2023-10-02 21:03:54,939 - Epoch: [73][ 50/ 1236] Overall Loss 0.258324 Objective Loss 0.258324 LR 0.001000 Time 0.037774 +2023-10-02 21:03:55,150 - Epoch: [73][ 60/ 1236] Overall Loss 0.258935 Objective Loss 0.258935 LR 0.001000 Time 0.035002 +2023-10-02 21:03:55,362 - Epoch: [73][ 70/ 1236] Overall Loss 0.259523 Objective Loss 0.259523 LR 0.001000 Time 0.033004 +2023-10-02 21:03:55,574 - Epoch: [73][ 80/ 1236] Overall Loss 0.260725 Objective Loss 0.260725 LR 0.001000 Time 0.031521 +2023-10-02 21:03:55,785 - Epoch: [73][ 90/ 1236] Overall Loss 0.261480 Objective Loss 0.261480 LR 0.001000 Time 0.030355 +2023-10-02 21:03:55,997 - Epoch: [73][ 100/ 1236] Overall Loss 0.260154 Objective Loss 0.260154 LR 0.001000 Time 0.029429 +2023-10-02 21:03:56,209 - Epoch: [73][ 110/ 1236] Overall Loss 0.260968 Objective Loss 0.260968 LR 0.001000 Time 0.028666 +2023-10-02 21:03:56,420 - Epoch: [73][ 120/ 1236] Overall Loss 0.262157 Objective Loss 0.262157 LR 0.001000 Time 0.028038 +2023-10-02 21:03:56,632 - Epoch: [73][ 130/ 1236] Overall Loss 0.262556 Objective Loss 0.262556 LR 0.001000 Time 0.027498 +2023-10-02 21:03:56,845 - Epoch: [73][ 140/ 1236] Overall Loss 0.263922 Objective Loss 0.263922 LR 0.001000 Time 0.027027 +2023-10-02 21:03:57,056 - Epoch: [73][ 150/ 1236] Overall Loss 0.261454 Objective Loss 0.261454 LR 0.001000 Time 0.026625 +2023-10-02 21:03:57,272 - Epoch: [73][ 160/ 1236] Overall Loss 0.260362 Objective Loss 0.260362 LR 0.001000 Time 0.026308 +2023-10-02 21:03:57,483 - Epoch: [73][ 170/ 1236] Overall Loss 0.260822 Objective Loss 0.260822 LR 0.001000 Time 0.026000 +2023-10-02 21:03:57,698 - Epoch: [73][ 180/ 1236] Overall Loss 0.259917 Objective Loss 0.259917 LR 0.001000 Time 0.025746 +2023-10-02 21:03:57,909 - Epoch: [73][ 190/ 1236] Overall Loss 0.258367 Objective Loss 0.258367 LR 0.001000 Time 0.025499 +2023-10-02 21:03:58,122 - Epoch: [73][ 200/ 1236] Overall Loss 0.258307 Objective Loss 0.258307 LR 0.001000 Time 0.025291 +2023-10-02 21:03:58,335 - Epoch: [73][ 210/ 1236] Overall Loss 0.258955 Objective Loss 0.258955 LR 0.001000 Time 0.025091 +2023-10-02 21:03:58,548 - Epoch: [73][ 220/ 1236] Overall Loss 0.257046 Objective Loss 0.257046 LR 0.001000 Time 0.024917 +2023-10-02 21:03:58,759 - Epoch: [73][ 230/ 1236] Overall Loss 0.257424 Objective Loss 0.257424 LR 0.001000 Time 0.024750 +2023-10-02 21:03:58,974 - Epoch: [73][ 240/ 1236] Overall Loss 0.257281 Objective Loss 0.257281 LR 0.001000 Time 0.024612 +2023-10-02 21:03:59,184 - Epoch: [73][ 250/ 1236] Overall Loss 0.256951 Objective Loss 0.256951 LR 0.001000 Time 0.024470 +2023-10-02 21:03:59,399 - Epoch: [73][ 260/ 1236] Overall Loss 0.256727 Objective Loss 0.256727 LR 0.001000 Time 0.024354 +2023-10-02 21:03:59,610 - Epoch: [73][ 270/ 1236] Overall Loss 0.257543 Objective Loss 0.257543 LR 0.001000 Time 0.024230 +2023-10-02 21:03:59,825 - Epoch: [73][ 280/ 1236] Overall Loss 0.257990 Objective Loss 0.257990 LR 0.001000 Time 0.024132 +2023-10-02 21:04:00,035 - Epoch: [73][ 290/ 1236] Overall Loss 0.257773 Objective Loss 0.257773 LR 0.001000 Time 0.024025 +2023-10-02 21:04:00,250 - Epoch: [73][ 300/ 1236] Overall Loss 0.257872 Objective Loss 0.257872 LR 0.001000 Time 0.023940 +2023-10-02 21:04:00,461 - Epoch: [73][ 310/ 1236] Overall Loss 0.258254 Objective Loss 0.258254 LR 0.001000 Time 0.023847 +2023-10-02 21:04:00,675 - Epoch: [73][ 320/ 1236] Overall Loss 0.258524 Objective Loss 0.258524 LR 0.001000 Time 0.023767 +2023-10-02 21:04:00,887 - Epoch: [73][ 330/ 1236] Overall Loss 0.259188 Objective Loss 0.259188 LR 0.001000 Time 0.023688 +2023-10-02 21:04:01,100 - Epoch: [73][ 340/ 1236] Overall Loss 0.258193 Objective Loss 0.258193 LR 0.001000 Time 0.023620 +2023-10-02 21:04:01,309 - Epoch: [73][ 350/ 1236] Overall Loss 0.257557 Objective Loss 0.257557 LR 0.001000 Time 0.023539 +2023-10-02 21:04:01,521 - Epoch: [73][ 360/ 1236] Overall Loss 0.256898 Objective Loss 0.256898 LR 0.001000 Time 0.023474 +2023-10-02 21:04:01,729 - Epoch: [73][ 370/ 1236] Overall Loss 0.256379 Objective Loss 0.256379 LR 0.001000 Time 0.023401 +2023-10-02 21:04:01,941 - Epoch: [73][ 380/ 1236] Overall Loss 0.256080 Objective Loss 0.256080 LR 0.001000 Time 0.023343 +2023-10-02 21:04:02,149 - Epoch: [73][ 390/ 1236] Overall Loss 0.256241 Objective Loss 0.256241 LR 0.001000 Time 0.023277 +2023-10-02 21:04:02,361 - Epoch: [73][ 400/ 1236] Overall Loss 0.256548 Objective Loss 0.256548 LR 0.001000 Time 0.023224 +2023-10-02 21:04:02,569 - Epoch: [73][ 410/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.023164 +2023-10-02 21:04:02,781 - Epoch: [73][ 420/ 1236] Overall Loss 0.256561 Objective Loss 0.256561 LR 0.001000 Time 0.023117 +2023-10-02 21:04:02,989 - Epoch: [73][ 430/ 1236] Overall Loss 0.256449 Objective Loss 0.256449 LR 0.001000 Time 0.023062 +2023-10-02 21:04:03,201 - Epoch: [73][ 440/ 1236] Overall Loss 0.256953 Objective Loss 0.256953 LR 0.001000 Time 0.023019 +2023-10-02 21:04:03,409 - Epoch: [73][ 450/ 1236] Overall Loss 0.257913 Objective Loss 0.257913 LR 0.001000 Time 0.022969 +2023-10-02 21:04:03,621 - Epoch: [73][ 460/ 1236] Overall Loss 0.258239 Objective Loss 0.258239 LR 0.001000 Time 0.022930 +2023-10-02 21:04:03,829 - Epoch: [73][ 470/ 1236] Overall Loss 0.258677 Objective Loss 0.258677 LR 0.001000 Time 0.022884 +2023-10-02 21:04:04,041 - Epoch: [73][ 480/ 1236] Overall Loss 0.259493 Objective Loss 0.259493 LR 0.001000 Time 0.022849 +2023-10-02 21:04:04,249 - Epoch: [73][ 490/ 1236] Overall Loss 0.259328 Objective Loss 0.259328 LR 0.001000 Time 0.022806 +2023-10-02 21:04:04,462 - Epoch: [73][ 500/ 1236] Overall Loss 0.259812 Objective Loss 0.259812 LR 0.001000 Time 0.022774 +2023-10-02 21:04:04,670 - Epoch: [73][ 510/ 1236] Overall Loss 0.259882 Objective Loss 0.259882 LR 0.001000 Time 0.022735 +2023-10-02 21:04:04,882 - Epoch: [73][ 520/ 1236] Overall Loss 0.259924 Objective Loss 0.259924 LR 0.001000 Time 0.022706 +2023-10-02 21:04:05,090 - Epoch: [73][ 530/ 1236] Overall Loss 0.260022 Objective Loss 0.260022 LR 0.001000 Time 0.022669 +2023-10-02 21:04:05,302 - Epoch: [73][ 540/ 1236] Overall Loss 0.259928 Objective Loss 0.259928 LR 0.001000 Time 0.022642 +2023-10-02 21:04:05,510 - Epoch: [73][ 550/ 1236] Overall Loss 0.260329 Objective Loss 0.260329 LR 0.001000 Time 0.022607 +2023-10-02 21:04:05,722 - Epoch: [73][ 560/ 1236] Overall Loss 0.260049 Objective Loss 0.260049 LR 0.001000 Time 0.022581 +2023-10-02 21:04:05,930 - Epoch: [73][ 570/ 1236] Overall Loss 0.260353 Objective Loss 0.260353 LR 0.001000 Time 0.022550 +2023-10-02 21:04:06,142 - Epoch: [73][ 580/ 1236] Overall Loss 0.260229 Objective Loss 0.260229 LR 0.001000 Time 0.022526 +2023-10-02 21:04:06,350 - Epoch: [73][ 590/ 1236] Overall Loss 0.260318 Objective Loss 0.260318 LR 0.001000 Time 0.022497 +2023-10-02 21:04:06,562 - Epoch: [73][ 600/ 1236] Overall Loss 0.260151 Objective Loss 0.260151 LR 0.001000 Time 0.022474 +2023-10-02 21:04:06,770 - Epoch: [73][ 610/ 1236] Overall Loss 0.259630 Objective Loss 0.259630 LR 0.001000 Time 0.022446 +2023-10-02 21:04:06,982 - Epoch: [73][ 620/ 1236] Overall Loss 0.259307 Objective Loss 0.259307 LR 0.001000 Time 0.022426 +2023-10-02 21:04:07,190 - Epoch: [73][ 630/ 1236] Overall Loss 0.259296 Objective Loss 0.259296 LR 0.001000 Time 0.022399 +2023-10-02 21:04:07,403 - Epoch: [73][ 640/ 1236] Overall Loss 0.259652 Objective Loss 0.259652 LR 0.001000 Time 0.022381 +2023-10-02 21:04:07,611 - Epoch: [73][ 650/ 1236] Overall Loss 0.259662 Objective Loss 0.259662 LR 0.001000 Time 0.022356 +2023-10-02 21:04:07,823 - Epoch: [73][ 660/ 1236] Overall Loss 0.259796 Objective Loss 0.259796 LR 0.001000 Time 0.022339 +2023-10-02 21:04:08,031 - Epoch: [73][ 670/ 1236] Overall Loss 0.259932 Objective Loss 0.259932 LR 0.001000 Time 0.022315 +2023-10-02 21:04:08,243 - Epoch: [73][ 680/ 1236] Overall Loss 0.259895 Objective Loss 0.259895 LR 0.001000 Time 0.022299 +2023-10-02 21:04:08,452 - Epoch: [73][ 690/ 1236] Overall Loss 0.259750 Objective Loss 0.259750 LR 0.001000 Time 0.022277 +2023-10-02 21:04:08,663 - Epoch: [73][ 700/ 1236] Overall Loss 0.259884 Objective Loss 0.259884 LR 0.001000 Time 0.022261 +2023-10-02 21:04:08,871 - Epoch: [73][ 710/ 1236] Overall Loss 0.259865 Objective Loss 0.259865 LR 0.001000 Time 0.022240 +2023-10-02 21:04:09,083 - Epoch: [73][ 720/ 1236] Overall Loss 0.260275 Objective Loss 0.260275 LR 0.001000 Time 0.022225 +2023-10-02 21:04:09,292 - Epoch: [73][ 730/ 1236] Overall Loss 0.260435 Objective Loss 0.260435 LR 0.001000 Time 0.022205 +2023-10-02 21:04:09,504 - Epoch: [73][ 740/ 1236] Overall Loss 0.260514 Objective Loss 0.260514 LR 0.001000 Time 0.022192 +2023-10-02 21:04:09,712 - Epoch: [73][ 750/ 1236] Overall Loss 0.260116 Objective Loss 0.260116 LR 0.001000 Time 0.022173 +2023-10-02 21:04:09,924 - Epoch: [73][ 760/ 1236] Overall Loss 0.260129 Objective Loss 0.260129 LR 0.001000 Time 0.022160 +2023-10-02 21:04:10,132 - Epoch: [73][ 770/ 1236] Overall Loss 0.259919 Objective Loss 0.259919 LR 0.001000 Time 0.022142 +2023-10-02 21:04:10,344 - Epoch: [73][ 780/ 1236] Overall Loss 0.260034 Objective Loss 0.260034 LR 0.001000 Time 0.022130 +2023-10-02 21:04:10,553 - Epoch: [73][ 790/ 1236] Overall Loss 0.260433 Objective Loss 0.260433 LR 0.001000 Time 0.022113 +2023-10-02 21:04:10,765 - Epoch: [73][ 800/ 1236] Overall Loss 0.260710 Objective Loss 0.260710 LR 0.001000 Time 0.022101 +2023-10-02 21:04:10,973 - Epoch: [73][ 810/ 1236] Overall Loss 0.260310 Objective Loss 0.260310 LR 0.001000 Time 0.022085 +2023-10-02 21:04:11,185 - Epoch: [73][ 820/ 1236] Overall Loss 0.260327 Objective Loss 0.260327 LR 0.001000 Time 0.022074 +2023-10-02 21:04:11,394 - Epoch: [73][ 830/ 1236] Overall Loss 0.260965 Objective Loss 0.260965 LR 0.001000 Time 0.022059 +2023-10-02 21:04:11,606 - Epoch: [73][ 840/ 1236] Overall Loss 0.260833 Objective Loss 0.260833 LR 0.001000 Time 0.022049 +2023-10-02 21:04:11,814 - Epoch: [73][ 850/ 1236] Overall Loss 0.260764 Objective Loss 0.260764 LR 0.001000 Time 0.022034 +2023-10-02 21:04:12,026 - Epoch: [73][ 860/ 1236] Overall Loss 0.260888 Objective Loss 0.260888 LR 0.001000 Time 0.022024 +2023-10-02 21:04:12,234 - Epoch: [73][ 870/ 1236] Overall Loss 0.260662 Objective Loss 0.260662 LR 0.001000 Time 0.022010 +2023-10-02 21:04:12,446 - Epoch: [73][ 880/ 1236] Overall Loss 0.260512 Objective Loss 0.260512 LR 0.001000 Time 0.021999 +2023-10-02 21:04:12,656 - Epoch: [73][ 890/ 1236] Overall Loss 0.260306 Objective Loss 0.260306 LR 0.001000 Time 0.021988 +2023-10-02 21:04:12,863 - Epoch: [73][ 900/ 1236] Overall Loss 0.260228 Objective Loss 0.260228 LR 0.001000 Time 0.021974 +2023-10-02 21:04:13,073 - Epoch: [73][ 910/ 1236] Overall Loss 0.260214 Objective Loss 0.260214 LR 0.001000 Time 0.021963 +2023-10-02 21:04:13,281 - Epoch: [73][ 920/ 1236] Overall Loss 0.260396 Objective Loss 0.260396 LR 0.001000 Time 0.021950 +2023-10-02 21:04:13,490 - Epoch: [73][ 930/ 1236] Overall Loss 0.260638 Objective Loss 0.260638 LR 0.001000 Time 0.021938 +2023-10-02 21:04:13,698 - Epoch: [73][ 940/ 1236] Overall Loss 0.260975 Objective Loss 0.260975 LR 0.001000 Time 0.021926 +2023-10-02 21:04:13,909 - Epoch: [73][ 950/ 1236] Overall Loss 0.261078 Objective Loss 0.261078 LR 0.001000 Time 0.021916 +2023-10-02 21:04:14,116 - Epoch: [73][ 960/ 1236] Overall Loss 0.261074 Objective Loss 0.261074 LR 0.001000 Time 0.021903 +2023-10-02 21:04:14,325 - Epoch: [73][ 970/ 1236] Overall Loss 0.261342 Objective Loss 0.261342 LR 0.001000 Time 0.021893 +2023-10-02 21:04:14,533 - Epoch: [73][ 980/ 1236] Overall Loss 0.261225 Objective Loss 0.261225 LR 0.001000 Time 0.021882 +2023-10-02 21:04:14,743 - Epoch: [73][ 990/ 1236] Overall Loss 0.261050 Objective Loss 0.261050 LR 0.001000 Time 0.021873 +2023-10-02 21:04:14,951 - Epoch: [73][ 1000/ 1236] Overall Loss 0.261225 Objective Loss 0.261225 LR 0.001000 Time 0.021861 +2023-10-02 21:04:15,161 - Epoch: [73][ 1010/ 1236] Overall Loss 0.261253 Objective Loss 0.261253 LR 0.001000 Time 0.021852 +2023-10-02 21:04:15,369 - Epoch: [73][ 1020/ 1236] Overall Loss 0.261551 Objective Loss 0.261551 LR 0.001000 Time 0.021841 +2023-10-02 21:04:15,579 - Epoch: [73][ 1030/ 1236] Overall Loss 0.261698 Objective Loss 0.261698 LR 0.001000 Time 0.021833 +2023-10-02 21:04:15,786 - Epoch: [73][ 1040/ 1236] Overall Loss 0.261687 Objective Loss 0.261687 LR 0.001000 Time 0.021822 +2023-10-02 21:04:15,995 - Epoch: [73][ 1050/ 1236] Overall Loss 0.261434 Objective Loss 0.261434 LR 0.001000 Time 0.021813 +2023-10-02 21:04:16,203 - Epoch: [73][ 1060/ 1236] Overall Loss 0.261476 Objective Loss 0.261476 LR 0.001000 Time 0.021802 +2023-10-02 21:04:16,414 - Epoch: [73][ 1070/ 1236] Overall Loss 0.261124 Objective Loss 0.261124 LR 0.001000 Time 0.021795 +2023-10-02 21:04:16,621 - Epoch: [73][ 1080/ 1236] Overall Loss 0.260907 Objective Loss 0.260907 LR 0.001000 Time 0.021785 +2023-10-02 21:04:16,832 - Epoch: [73][ 1090/ 1236] Overall Loss 0.261025 Objective Loss 0.261025 LR 0.001000 Time 0.021778 +2023-10-02 21:04:17,045 - Epoch: [73][ 1100/ 1236] Overall Loss 0.261066 Objective Loss 0.261066 LR 0.001000 Time 0.021773 +2023-10-02 21:04:17,259 - Epoch: [73][ 1110/ 1236] Overall Loss 0.260998 Objective Loss 0.260998 LR 0.001000 Time 0.021770 +2023-10-02 21:04:17,470 - Epoch: [73][ 1120/ 1236] Overall Loss 0.261205 Objective Loss 0.261205 LR 0.001000 Time 0.021762 +2023-10-02 21:04:17,681 - Epoch: [73][ 1130/ 1236] Overall Loss 0.261167 Objective Loss 0.261167 LR 0.001000 Time 0.021756 +2023-10-02 21:04:17,890 - Epoch: [73][ 1140/ 1236] Overall Loss 0.261216 Objective Loss 0.261216 LR 0.001000 Time 0.021749 +2023-10-02 21:04:18,102 - Epoch: [73][ 1150/ 1236] Overall Loss 0.261557 Objective Loss 0.261557 LR 0.001000 Time 0.021743 +2023-10-02 21:04:18,311 - Epoch: [73][ 1160/ 1236] Overall Loss 0.261465 Objective Loss 0.261465 LR 0.001000 Time 0.021735 +2023-10-02 21:04:18,522 - Epoch: [73][ 1170/ 1236] Overall Loss 0.261589 Objective Loss 0.261589 LR 0.001000 Time 0.021730 +2023-10-02 21:04:18,731 - Epoch: [73][ 1180/ 1236] Overall Loss 0.261838 Objective Loss 0.261838 LR 0.001000 Time 0.021723 +2023-10-02 21:04:18,942 - Epoch: [73][ 1190/ 1236] Overall Loss 0.261772 Objective Loss 0.261772 LR 0.001000 Time 0.021717 +2023-10-02 21:04:19,152 - Epoch: [73][ 1200/ 1236] Overall Loss 0.261957 Objective Loss 0.261957 LR 0.001000 Time 0.021710 +2023-10-02 21:04:19,363 - Epoch: [73][ 1210/ 1236] Overall Loss 0.262224 Objective Loss 0.262224 LR 0.001000 Time 0.021705 +2023-10-02 21:04:19,572 - Epoch: [73][ 1220/ 1236] Overall Loss 0.262130 Objective Loss 0.262130 LR 0.001000 Time 0.021699 +2023-10-02 21:04:19,835 - Epoch: [73][ 1230/ 1236] Overall Loss 0.261933 Objective Loss 0.261933 LR 0.001000 Time 0.021736 +2023-10-02 21:04:19,956 - Epoch: [73][ 1236/ 1236] Overall Loss 0.261959 Objective Loss 0.261959 Top1 87.780041 Top5 97.963340 LR 0.001000 Time 0.021728 +2023-10-02 21:04:20,081 - --- validate (epoch=73)----------- +2023-10-02 21:04:20,081 - 29943 samples (256 per mini-batch) +2023-10-02 21:04:20,580 - Epoch: [73][ 10/ 117] Loss 0.296837 Top1 83.593750 Top5 98.242188 +2023-10-02 21:04:20,743 - Epoch: [73][ 20/ 117] Loss 0.304092 Top1 83.457031 Top5 98.222656 +2023-10-02 21:04:20,900 - Epoch: [73][ 30/ 117] Loss 0.311974 Top1 83.333333 Top5 98.242188 +2023-10-02 21:04:21,063 - Epoch: [73][ 40/ 117] Loss 0.319419 Top1 83.144531 Top5 98.134766 +2023-10-02 21:04:21,220 - Epoch: [73][ 50/ 117] Loss 0.321100 Top1 83.085938 Top5 98.140625 +2023-10-02 21:04:21,374 - Epoch: [73][ 60/ 117] Loss 0.325861 Top1 83.085938 Top5 98.098958 +2023-10-02 21:04:21,525 - Epoch: [73][ 70/ 117] Loss 0.325862 Top1 83.320312 Top5 98.136161 +2023-10-02 21:04:21,677 - Epoch: [73][ 80/ 117] Loss 0.328166 Top1 83.403320 Top5 98.110352 +2023-10-02 21:04:21,828 - Epoch: [73][ 90/ 117] Loss 0.328872 Top1 83.519965 Top5 98.090278 +2023-10-02 21:04:21,982 - Epoch: [73][ 100/ 117] Loss 0.328685 Top1 83.519531 Top5 98.093750 +2023-10-02 21:04:22,140 - Epoch: [73][ 110/ 117] Loss 0.328313 Top1 83.355824 Top5 98.117898 +2023-10-02 21:04:22,228 - Epoch: [73][ 117/ 117] Loss 0.330283 Top1 83.345022 Top5 98.086364 +2023-10-02 21:04:22,364 - ==> Top1: 83.345 Top5: 98.086 Loss: 0.330 + +2023-10-02 21:04:22,365 - ==> Confusion: +[[ 927 2 4 0 13 3 0 0 11 60 2 0 1 4 8 1 3 1 1 0 9] + [ 0 1057 2 1 3 27 2 11 4 0 3 1 0 1 0 3 0 0 7 3 6] + [ 3 2 975 7 5 1 21 8 0 1 0 0 7 3 1 4 1 2 7 2 6] + [ 2 4 17 966 0 6 2 1 1 2 11 0 7 5 37 2 0 1 10 0 15] + [ 27 5 1 0 971 6 0 0 2 9 1 2 0 0 7 5 7 0 0 0 7] + [ 2 43 2 0 4 989 1 16 5 4 6 3 1 16 7 1 0 0 5 2 9] + [ 0 4 29 0 0 4 1122 2 0 0 5 3 0 0 0 8 1 0 0 4 9] + [ 1 25 17 1 4 44 5 1025 0 4 5 8 4 5 2 0 2 1 46 10 9] + [ 15 5 1 0 2 3 0 0 962 51 8 1 1 9 19 3 1 1 4 1 2] + [ 91 0 4 0 7 0 0 0 32 943 0 0 0 19 9 2 1 1 1 2 7] + [ 1 4 11 4 2 4 1 2 15 2 970 2 0 12 5 2 2 0 5 2 7] + [ 1 3 2 0 2 11 0 3 0 1 0 954 18 8 0 4 5 16 0 4 3] + [ 0 2 3 3 1 2 1 1 1 0 5 56 934 4 5 11 0 23 0 7 9] + [ 2 0 2 0 2 4 0 0 21 18 6 4 1 1043 8 0 0 0 0 0 8] + [ 10 3 4 19 6 0 0 0 20 2 2 0 3 1 1008 0 1 3 10 0 9] + [ 0 0 2 0 7 2 1 0 0 0 0 10 5 2 1 1068 18 8 1 3 6] + [ 1 18 2 0 7 9 0 0 1 2 1 8 0 1 2 9 1085 0 0 6 9] + [ 1 1 0 2 0 0 3 0 0 3 1 6 11 2 2 8 3 993 1 0 1] + [ 3 5 8 17 1 2 0 13 5 0 5 0 0 0 12 0 1 0 986 0 10] + [ 0 7 3 3 0 12 6 3 0 1 1 22 6 5 0 4 4 1 3 1064 7] + [ 131 229 171 93 92 191 54 71 115 108 193 134 372 288 133 61 131 66 183 175 4914]] + +2023-10-02 21:04:22,366 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:04:22,366 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:04:22,372 - + +2023-10-02 21:04:22,372 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:04:23,491 - Epoch: [74][ 10/ 1236] Overall Loss 0.275515 Objective Loss 0.275515 LR 0.001000 Time 0.111867 +2023-10-02 21:04:23,698 - Epoch: [74][ 20/ 1236] Overall Loss 0.268826 Objective Loss 0.268826 LR 0.001000 Time 0.066246 +2023-10-02 21:04:23,904 - Epoch: [74][ 30/ 1236] Overall Loss 0.264774 Objective Loss 0.264774 LR 0.001000 Time 0.050977 +2023-10-02 21:04:24,112 - Epoch: [74][ 40/ 1236] Overall Loss 0.272006 Objective Loss 0.272006 LR 0.001000 Time 0.043429 +2023-10-02 21:04:24,316 - Epoch: [74][ 50/ 1236] Overall Loss 0.263855 Objective Loss 0.263855 LR 0.001000 Time 0.038825 +2023-10-02 21:04:24,525 - Epoch: [74][ 60/ 1236] Overall Loss 0.257975 Objective Loss 0.257975 LR 0.001000 Time 0.035819 +2023-10-02 21:04:24,729 - Epoch: [74][ 70/ 1236] Overall Loss 0.256031 Objective Loss 0.256031 LR 0.001000 Time 0.033618 +2023-10-02 21:04:24,937 - Epoch: [74][ 80/ 1236] Overall Loss 0.253882 Objective Loss 0.253882 LR 0.001000 Time 0.032011 +2023-10-02 21:04:25,141 - Epoch: [74][ 90/ 1236] Overall Loss 0.255121 Objective Loss 0.255121 LR 0.001000 Time 0.030723 +2023-10-02 21:04:25,350 - Epoch: [74][ 100/ 1236] Overall Loss 0.254470 Objective Loss 0.254470 LR 0.001000 Time 0.029731 +2023-10-02 21:04:25,554 - Epoch: [74][ 110/ 1236] Overall Loss 0.255835 Objective Loss 0.255835 LR 0.001000 Time 0.028882 +2023-10-02 21:04:25,762 - Epoch: [74][ 120/ 1236] Overall Loss 0.257529 Objective Loss 0.257529 LR 0.001000 Time 0.028206 +2023-10-02 21:04:25,966 - Epoch: [74][ 130/ 1236] Overall Loss 0.257599 Objective Loss 0.257599 LR 0.001000 Time 0.027606 +2023-10-02 21:04:26,175 - Epoch: [74][ 140/ 1236] Overall Loss 0.258336 Objective Loss 0.258336 LR 0.001000 Time 0.027120 +2023-10-02 21:04:26,379 - Epoch: [74][ 150/ 1236] Overall Loss 0.257772 Objective Loss 0.257772 LR 0.001000 Time 0.026673 +2023-10-02 21:04:26,587 - Epoch: [74][ 160/ 1236] Overall Loss 0.257168 Objective Loss 0.257168 LR 0.001000 Time 0.026302 +2023-10-02 21:04:26,791 - Epoch: [74][ 170/ 1236] Overall Loss 0.257354 Objective Loss 0.257354 LR 0.001000 Time 0.025954 +2023-10-02 21:04:26,999 - Epoch: [74][ 180/ 1236] Overall Loss 0.256003 Objective Loss 0.256003 LR 0.001000 Time 0.025667 +2023-10-02 21:04:27,203 - Epoch: [74][ 190/ 1236] Overall Loss 0.255507 Objective Loss 0.255507 LR 0.001000 Time 0.025389 +2023-10-02 21:04:27,411 - Epoch: [74][ 200/ 1236] Overall Loss 0.256450 Objective Loss 0.256450 LR 0.001000 Time 0.025157 +2023-10-02 21:04:27,615 - Epoch: [74][ 210/ 1236] Overall Loss 0.256227 Objective Loss 0.256227 LR 0.001000 Time 0.024930 +2023-10-02 21:04:27,823 - Epoch: [74][ 220/ 1236] Overall Loss 0.256033 Objective Loss 0.256033 LR 0.001000 Time 0.024738 +2023-10-02 21:04:28,027 - Epoch: [74][ 230/ 1236] Overall Loss 0.255866 Objective Loss 0.255866 LR 0.001000 Time 0.024550 +2023-10-02 21:04:28,234 - Epoch: [74][ 240/ 1236] Overall Loss 0.256299 Objective Loss 0.256299 LR 0.001000 Time 0.024390 +2023-10-02 21:04:28,439 - Epoch: [74][ 250/ 1236] Overall Loss 0.256109 Objective Loss 0.256109 LR 0.001000 Time 0.024231 +2023-10-02 21:04:28,647 - Epoch: [74][ 260/ 1236] Overall Loss 0.256180 Objective Loss 0.256180 LR 0.001000 Time 0.024097 +2023-10-02 21:04:28,851 - Epoch: [74][ 270/ 1236] Overall Loss 0.256730 Objective Loss 0.256730 LR 0.001000 Time 0.023960 +2023-10-02 21:04:29,059 - Epoch: [74][ 280/ 1236] Overall Loss 0.256900 Objective Loss 0.256900 LR 0.001000 Time 0.023845 +2023-10-02 21:04:29,262 - Epoch: [74][ 290/ 1236] Overall Loss 0.256216 Objective Loss 0.256216 LR 0.001000 Time 0.023725 +2023-10-02 21:04:29,471 - Epoch: [74][ 300/ 1236] Overall Loss 0.256445 Objective Loss 0.256445 LR 0.001000 Time 0.023627 +2023-10-02 21:04:29,674 - Epoch: [74][ 310/ 1236] Overall Loss 0.256649 Objective Loss 0.256649 LR 0.001000 Time 0.023521 +2023-10-02 21:04:29,882 - Epoch: [74][ 320/ 1236] Overall Loss 0.257610 Objective Loss 0.257610 LR 0.001000 Time 0.023434 +2023-10-02 21:04:30,086 - Epoch: [74][ 330/ 1236] Overall Loss 0.257114 Objective Loss 0.257114 LR 0.001000 Time 0.023342 +2023-10-02 21:04:30,294 - Epoch: [74][ 340/ 1236] Overall Loss 0.256984 Objective Loss 0.256984 LR 0.001000 Time 0.023266 +2023-10-02 21:04:30,498 - Epoch: [74][ 350/ 1236] Overall Loss 0.256543 Objective Loss 0.256543 LR 0.001000 Time 0.023184 +2023-10-02 21:04:30,707 - Epoch: [74][ 360/ 1236] Overall Loss 0.256521 Objective Loss 0.256521 LR 0.001000 Time 0.023118 +2023-10-02 21:04:30,911 - Epoch: [74][ 370/ 1236] Overall Loss 0.256174 Objective Loss 0.256174 LR 0.001000 Time 0.023045 +2023-10-02 21:04:31,119 - Epoch: [74][ 380/ 1236] Overall Loss 0.256655 Objective Loss 0.256655 LR 0.001000 Time 0.022985 +2023-10-02 21:04:31,324 - Epoch: [74][ 390/ 1236] Overall Loss 0.257520 Objective Loss 0.257520 LR 0.001000 Time 0.022920 +2023-10-02 21:04:31,532 - Epoch: [74][ 400/ 1236] Overall Loss 0.257204 Objective Loss 0.257204 LR 0.001000 Time 0.022867 +2023-10-02 21:04:31,739 - Epoch: [74][ 410/ 1236] Overall Loss 0.257566 Objective Loss 0.257566 LR 0.001000 Time 0.022812 +2023-10-02 21:04:31,947 - Epoch: [74][ 420/ 1236] Overall Loss 0.257664 Objective Loss 0.257664 LR 0.001000 Time 0.022764 +2023-10-02 21:04:32,152 - Epoch: [74][ 430/ 1236] Overall Loss 0.257595 Objective Loss 0.257595 LR 0.001000 Time 0.022709 +2023-10-02 21:04:32,360 - Epoch: [74][ 440/ 1236] Overall Loss 0.257959 Objective Loss 0.257959 LR 0.001000 Time 0.022666 +2023-10-02 21:04:32,564 - Epoch: [74][ 450/ 1236] Overall Loss 0.257711 Objective Loss 0.257711 LR 0.001000 Time 0.022616 +2023-10-02 21:04:32,773 - Epoch: [74][ 460/ 1236] Overall Loss 0.257860 Objective Loss 0.257860 LR 0.001000 Time 0.022576 +2023-10-02 21:04:32,977 - Epoch: [74][ 470/ 1236] Overall Loss 0.258126 Objective Loss 0.258126 LR 0.001000 Time 0.022531 +2023-10-02 21:04:33,185 - Epoch: [74][ 480/ 1236] Overall Loss 0.258269 Objective Loss 0.258269 LR 0.001000 Time 0.022494 +2023-10-02 21:04:33,390 - Epoch: [74][ 490/ 1236] Overall Loss 0.257765 Objective Loss 0.257765 LR 0.001000 Time 0.022452 +2023-10-02 21:04:33,598 - Epoch: [74][ 500/ 1236] Overall Loss 0.257610 Objective Loss 0.257610 LR 0.001000 Time 0.022419 +2023-10-02 21:04:33,803 - Epoch: [74][ 510/ 1236] Overall Loss 0.257426 Objective Loss 0.257426 LR 0.001000 Time 0.022381 +2023-10-02 21:04:34,012 - Epoch: [74][ 520/ 1236] Overall Loss 0.258040 Objective Loss 0.258040 LR 0.001000 Time 0.022351 +2023-10-02 21:04:34,216 - Epoch: [74][ 530/ 1236] Overall Loss 0.258191 Objective Loss 0.258191 LR 0.001000 Time 0.022315 +2023-10-02 21:04:34,425 - Epoch: [74][ 540/ 1236] Overall Loss 0.257861 Objective Loss 0.257861 LR 0.001000 Time 0.022287 +2023-10-02 21:04:34,630 - Epoch: [74][ 550/ 1236] Overall Loss 0.257641 Objective Loss 0.257641 LR 0.001000 Time 0.022253 +2023-10-02 21:04:34,838 - Epoch: [74][ 560/ 1236] Overall Loss 0.257385 Objective Loss 0.257385 LR 0.001000 Time 0.022227 +2023-10-02 21:04:35,043 - Epoch: [74][ 570/ 1236] Overall Loss 0.257595 Objective Loss 0.257595 LR 0.001000 Time 0.022197 +2023-10-02 21:04:35,251 - Epoch: [74][ 580/ 1236] Overall Loss 0.257473 Objective Loss 0.257473 LR 0.001000 Time 0.022173 +2023-10-02 21:04:35,456 - Epoch: [74][ 590/ 1236] Overall Loss 0.256832 Objective Loss 0.256832 LR 0.001000 Time 0.022144 +2023-10-02 21:04:35,665 - Epoch: [74][ 600/ 1236] Overall Loss 0.256834 Objective Loss 0.256834 LR 0.001000 Time 0.022122 +2023-10-02 21:04:35,869 - Epoch: [74][ 610/ 1236] Overall Loss 0.256550 Objective Loss 0.256550 LR 0.001000 Time 0.022094 +2023-10-02 21:04:36,077 - Epoch: [74][ 620/ 1236] Overall Loss 0.256712 Objective Loss 0.256712 LR 0.001000 Time 0.022073 +2023-10-02 21:04:36,282 - Epoch: [74][ 630/ 1236] Overall Loss 0.256531 Objective Loss 0.256531 LR 0.001000 Time 0.022046 +2023-10-02 21:04:36,490 - Epoch: [74][ 640/ 1236] Overall Loss 0.256470 Objective Loss 0.256470 LR 0.001000 Time 0.022027 +2023-10-02 21:04:36,695 - Epoch: [74][ 650/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.022003 +2023-10-02 21:04:36,903 - Epoch: [74][ 660/ 1236] Overall Loss 0.257046 Objective Loss 0.257046 LR 0.001000 Time 0.021985 +2023-10-02 21:04:37,108 - Epoch: [74][ 670/ 1236] Overall Loss 0.256879 Objective Loss 0.256879 LR 0.001000 Time 0.021962 +2023-10-02 21:04:37,317 - Epoch: [74][ 680/ 1236] Overall Loss 0.256809 Objective Loss 0.256809 LR 0.001000 Time 0.021945 +2023-10-02 21:04:37,521 - Epoch: [74][ 690/ 1236] Overall Loss 0.256810 Objective Loss 0.256810 LR 0.001000 Time 0.021923 +2023-10-02 21:04:37,730 - Epoch: [74][ 700/ 1236] Overall Loss 0.256706 Objective Loss 0.256706 LR 0.001000 Time 0.021907 +2023-10-02 21:04:37,934 - Epoch: [74][ 710/ 1236] Overall Loss 0.256418 Objective Loss 0.256418 LR 0.001000 Time 0.021886 +2023-10-02 21:04:38,142 - Epoch: [74][ 720/ 1236] Overall Loss 0.256452 Objective Loss 0.256452 LR 0.001000 Time 0.021871 +2023-10-02 21:04:38,347 - Epoch: [74][ 730/ 1236] Overall Loss 0.256202 Objective Loss 0.256202 LR 0.001000 Time 0.021851 +2023-10-02 21:04:38,555 - Epoch: [74][ 740/ 1236] Overall Loss 0.256234 Objective Loss 0.256234 LR 0.001000 Time 0.021837 +2023-10-02 21:04:38,760 - Epoch: [74][ 750/ 1236] Overall Loss 0.256386 Objective Loss 0.256386 LR 0.001000 Time 0.021818 +2023-10-02 21:04:38,968 - Epoch: [74][ 760/ 1236] Overall Loss 0.256319 Objective Loss 0.256319 LR 0.001000 Time 0.021805 +2023-10-02 21:04:39,173 - Epoch: [74][ 770/ 1236] Overall Loss 0.256815 Objective Loss 0.256815 LR 0.001000 Time 0.021787 +2023-10-02 21:04:39,381 - Epoch: [74][ 780/ 1236] Overall Loss 0.256460 Objective Loss 0.256460 LR 0.001000 Time 0.021774 +2023-10-02 21:04:39,585 - Epoch: [74][ 790/ 1236] Overall Loss 0.256794 Objective Loss 0.256794 LR 0.001000 Time 0.021757 +2023-10-02 21:04:39,794 - Epoch: [74][ 800/ 1236] Overall Loss 0.256958 Objective Loss 0.256958 LR 0.001000 Time 0.021745 +2023-10-02 21:04:39,998 - Epoch: [74][ 810/ 1236] Overall Loss 0.257202 Objective Loss 0.257202 LR 0.001000 Time 0.021729 +2023-10-02 21:04:40,206 - Epoch: [74][ 820/ 1236] Overall Loss 0.257190 Objective Loss 0.257190 LR 0.001000 Time 0.021717 +2023-10-02 21:04:40,411 - Epoch: [74][ 830/ 1236] Overall Loss 0.257380 Objective Loss 0.257380 LR 0.001000 Time 0.021702 +2023-10-02 21:04:40,619 - Epoch: [74][ 840/ 1236] Overall Loss 0.257063 Objective Loss 0.257063 LR 0.001000 Time 0.021691 +2023-10-02 21:04:40,824 - Epoch: [74][ 850/ 1236] Overall Loss 0.256909 Objective Loss 0.256909 LR 0.001000 Time 0.021676 +2023-10-02 21:04:41,032 - Epoch: [74][ 860/ 1236] Overall Loss 0.256898 Objective Loss 0.256898 LR 0.001000 Time 0.021666 +2023-10-02 21:04:41,237 - Epoch: [74][ 870/ 1236] Overall Loss 0.257100 Objective Loss 0.257100 LR 0.001000 Time 0.021652 +2023-10-02 21:04:41,446 - Epoch: [74][ 880/ 1236] Overall Loss 0.257380 Objective Loss 0.257380 LR 0.001000 Time 0.021643 +2023-10-02 21:04:41,651 - Epoch: [74][ 890/ 1236] Overall Loss 0.257389 Objective Loss 0.257389 LR 0.001000 Time 0.021630 +2023-10-02 21:04:41,860 - Epoch: [74][ 900/ 1236] Overall Loss 0.257251 Objective Loss 0.257251 LR 0.001000 Time 0.021621 +2023-10-02 21:04:42,064 - Epoch: [74][ 910/ 1236] Overall Loss 0.256974 Objective Loss 0.256974 LR 0.001000 Time 0.021608 +2023-10-02 21:04:42,273 - Epoch: [74][ 920/ 1236] Overall Loss 0.257095 Objective Loss 0.257095 LR 0.001000 Time 0.021599 +2023-10-02 21:04:42,477 - Epoch: [74][ 930/ 1236] Overall Loss 0.256967 Objective Loss 0.256967 LR 0.001000 Time 0.021587 +2023-10-02 21:04:42,686 - Epoch: [74][ 940/ 1236] Overall Loss 0.256826 Objective Loss 0.256826 LR 0.001000 Time 0.021578 +2023-10-02 21:04:42,890 - Epoch: [74][ 950/ 1236] Overall Loss 0.256496 Objective Loss 0.256496 LR 0.001000 Time 0.021566 +2023-10-02 21:04:43,098 - Epoch: [74][ 960/ 1236] Overall Loss 0.256567 Objective Loss 0.256567 LR 0.001000 Time 0.021558 +2023-10-02 21:04:43,302 - Epoch: [74][ 970/ 1236] Overall Loss 0.256835 Objective Loss 0.256835 LR 0.001000 Time 0.021546 +2023-10-02 21:04:43,511 - Epoch: [74][ 980/ 1236] Overall Loss 0.256650 Objective Loss 0.256650 LR 0.001000 Time 0.021539 +2023-10-02 21:04:43,715 - Epoch: [74][ 990/ 1236] Overall Loss 0.257097 Objective Loss 0.257097 LR 0.001000 Time 0.021527 +2023-10-02 21:04:43,924 - Epoch: [74][ 1000/ 1236] Overall Loss 0.257203 Objective Loss 0.257203 LR 0.001000 Time 0.021520 +2023-10-02 21:04:44,128 - Epoch: [74][ 1010/ 1236] Overall Loss 0.257360 Objective Loss 0.257360 LR 0.001000 Time 0.021509 +2023-10-02 21:04:44,336 - Epoch: [74][ 1020/ 1236] Overall Loss 0.257210 Objective Loss 0.257210 LR 0.001000 Time 0.021502 +2023-10-02 21:04:44,541 - Epoch: [74][ 1030/ 1236] Overall Loss 0.257420 Objective Loss 0.257420 LR 0.001000 Time 0.021492 +2023-10-02 21:04:44,749 - Epoch: [74][ 1040/ 1236] Overall Loss 0.257454 Objective Loss 0.257454 LR 0.001000 Time 0.021485 +2023-10-02 21:04:44,953 - Epoch: [74][ 1050/ 1236] Overall Loss 0.257459 Objective Loss 0.257459 LR 0.001000 Time 0.021475 +2023-10-02 21:04:45,162 - Epoch: [74][ 1060/ 1236] Overall Loss 0.257595 Objective Loss 0.257595 LR 0.001000 Time 0.021468 +2023-10-02 21:04:45,366 - Epoch: [74][ 1070/ 1236] Overall Loss 0.257879 Objective Loss 0.257879 LR 0.001000 Time 0.021458 +2023-10-02 21:04:45,575 - Epoch: [74][ 1080/ 1236] Overall Loss 0.257891 Objective Loss 0.257891 LR 0.001000 Time 0.021452 +2023-10-02 21:04:45,779 - Epoch: [74][ 1090/ 1236] Overall Loss 0.257918 Objective Loss 0.257918 LR 0.001000 Time 0.021443 +2023-10-02 21:04:45,987 - Epoch: [74][ 1100/ 1236] Overall Loss 0.258117 Objective Loss 0.258117 LR 0.001000 Time 0.021437 +2023-10-02 21:04:46,192 - Epoch: [74][ 1110/ 1236] Overall Loss 0.258468 Objective Loss 0.258468 LR 0.001000 Time 0.021428 +2023-10-02 21:04:46,400 - Epoch: [74][ 1120/ 1236] Overall Loss 0.258771 Objective Loss 0.258771 LR 0.001000 Time 0.021422 +2023-10-02 21:04:46,605 - Epoch: [74][ 1130/ 1236] Overall Loss 0.258634 Objective Loss 0.258634 LR 0.001000 Time 0.021414 +2023-10-02 21:04:46,814 - Epoch: [74][ 1140/ 1236] Overall Loss 0.259108 Objective Loss 0.259108 LR 0.001000 Time 0.021409 +2023-10-02 21:04:47,018 - Epoch: [74][ 1150/ 1236] Overall Loss 0.259193 Objective Loss 0.259193 LR 0.001000 Time 0.021400 +2023-10-02 21:04:47,227 - Epoch: [74][ 1160/ 1236] Overall Loss 0.259438 Objective Loss 0.259438 LR 0.001000 Time 0.021395 +2023-10-02 21:04:47,431 - Epoch: [74][ 1170/ 1236] Overall Loss 0.259471 Objective Loss 0.259471 LR 0.001000 Time 0.021387 +2023-10-02 21:04:47,640 - Epoch: [74][ 1180/ 1236] Overall Loss 0.259583 Objective Loss 0.259583 LR 0.001000 Time 0.021382 +2023-10-02 21:04:47,844 - Epoch: [74][ 1190/ 1236] Overall Loss 0.259525 Objective Loss 0.259525 LR 0.001000 Time 0.021374 +2023-10-02 21:04:48,053 - Epoch: [74][ 1200/ 1236] Overall Loss 0.259722 Objective Loss 0.259722 LR 0.001000 Time 0.021369 +2023-10-02 21:04:48,258 - Epoch: [74][ 1210/ 1236] Overall Loss 0.259498 Objective Loss 0.259498 LR 0.001000 Time 0.021362 +2023-10-02 21:04:48,466 - Epoch: [74][ 1220/ 1236] Overall Loss 0.259688 Objective Loss 0.259688 LR 0.001000 Time 0.021357 +2023-10-02 21:04:48,724 - Epoch: [74][ 1230/ 1236] Overall Loss 0.259712 Objective Loss 0.259712 LR 0.001000 Time 0.021393 +2023-10-02 21:04:48,846 - Epoch: [74][ 1236/ 1236] Overall Loss 0.259587 Objective Loss 0.259587 Top1 86.354379 Top5 98.778004 LR 0.001000 Time 0.021388 +2023-10-02 21:04:48,975 - --- validate (epoch=74)----------- +2023-10-02 21:04:48,976 - 29943 samples (256 per mini-batch) +2023-10-02 21:04:49,447 - Epoch: [74][ 10/ 117] Loss 0.345801 Top1 83.984375 Top5 98.164062 +2023-10-02 21:04:49,599 - Epoch: [74][ 20/ 117] Loss 0.333075 Top1 83.886719 Top5 98.125000 +2023-10-02 21:04:49,748 - Epoch: [74][ 30/ 117] Loss 0.335069 Top1 83.632812 Top5 98.046875 +2023-10-02 21:04:49,899 - Epoch: [74][ 40/ 117] Loss 0.329951 Top1 83.632812 Top5 98.066406 +2023-10-02 21:04:50,048 - Epoch: [74][ 50/ 117] Loss 0.327315 Top1 83.757812 Top5 98.187500 +2023-10-02 21:04:50,199 - Epoch: [74][ 60/ 117] Loss 0.324868 Top1 83.704427 Top5 98.183594 +2023-10-02 21:04:50,348 - Epoch: [74][ 70/ 117] Loss 0.327630 Top1 83.705357 Top5 98.169643 +2023-10-02 21:04:50,500 - Epoch: [74][ 80/ 117] Loss 0.325426 Top1 83.701172 Top5 98.173828 +2023-10-02 21:04:50,649 - Epoch: [74][ 90/ 117] Loss 0.324251 Top1 83.663194 Top5 98.168403 +2023-10-02 21:04:50,802 - Epoch: [74][ 100/ 117] Loss 0.324947 Top1 83.617188 Top5 98.136719 +2023-10-02 21:04:50,960 - Epoch: [74][ 110/ 117] Loss 0.328176 Top1 83.469460 Top5 98.082386 +2023-10-02 21:04:51,049 - Epoch: [74][ 117/ 117] Loss 0.331083 Top1 83.471930 Top5 98.039609 +2023-10-02 21:04:51,144 - ==> Top1: 83.472 Top5: 98.040 Loss: 0.331 + +2023-10-02 21:04:51,145 - ==> Confusion: +[[ 944 1 6 1 7 5 0 0 8 54 3 0 5 1 0 1 1 2 0 0 11] + [ 0 1024 2 0 7 36 1 27 3 0 3 2 0 0 0 6 4 0 9 3 4] + [ 2 0 937 10 2 0 46 7 0 1 1 2 8 4 2 8 4 0 6 4 12] + [ 0 3 16 966 1 4 2 3 4 1 11 0 10 5 34 1 1 1 8 2 16] + [ 31 7 1 0 961 3 1 1 1 9 4 1 0 3 5 6 10 0 1 2 3] + [ 3 35 1 1 2 989 0 25 0 6 6 3 2 13 3 1 4 0 3 7 12] + [ 0 0 22 0 0 0 1125 7 0 0 3 1 0 1 0 14 0 0 0 12 6] + [ 1 13 23 0 6 28 12 1053 1 0 10 5 3 0 3 2 1 3 37 11 6] + [ 15 1 0 1 2 2 0 1 977 31 14 2 2 18 14 1 1 1 1 2 3] + [ 99 1 2 0 7 3 0 1 30 931 2 0 0 22 6 2 0 1 0 4 8] + [ 2 1 11 6 0 3 4 1 11 0 971 2 2 17 4 2 1 2 1 6 6] + [ 0 1 2 0 0 11 0 2 0 0 0 970 20 9 0 2 0 14 0 3 1] + [ 0 0 1 1 1 4 4 0 2 1 1 50 934 2 3 14 3 23 1 5 18] + [ 1 0 1 0 2 16 0 1 14 9 4 8 1 1036 7 0 1 1 0 1 16] + [ 11 1 5 13 7 0 0 0 33 6 3 0 5 4 992 0 2 4 7 0 8] + [ 0 0 3 0 5 0 2 0 0 0 0 7 7 0 1 1066 18 11 2 4 8] + [ 1 13 1 0 5 6 0 0 1 0 0 8 2 3 3 8 1089 0 1 10 10] + [ 0 0 0 1 1 0 2 0 1 0 0 8 18 3 7 3 0 988 0 2 4] + [ 2 6 2 17 0 0 0 21 6 0 8 1 2 0 12 0 0 0 977 4 10] + [ 0 4 4 0 0 4 6 6 0 0 0 22 4 1 0 7 8 1 2 1077 6] + [ 156 170 128 81 88 183 60 96 106 101 234 156 290 298 129 84 123 73 125 237 4987]] + +2023-10-02 21:04:51,146 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:04:51,146 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:04:51,152 - + +2023-10-02 21:04:51,152 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:04:52,157 - Epoch: [75][ 10/ 1236] Overall Loss 0.249456 Objective Loss 0.249456 LR 0.001000 Time 0.100436 +2023-10-02 21:04:52,363 - Epoch: [75][ 20/ 1236] Overall Loss 0.258334 Objective Loss 0.258334 LR 0.001000 Time 0.060486 +2023-10-02 21:04:52,568 - Epoch: [75][ 30/ 1236] Overall Loss 0.275203 Objective Loss 0.275203 LR 0.001000 Time 0.047132 +2023-10-02 21:04:52,773 - Epoch: [75][ 40/ 1236] Overall Loss 0.273007 Objective Loss 0.273007 LR 0.001000 Time 0.040477 +2023-10-02 21:04:52,977 - Epoch: [75][ 50/ 1236] Overall Loss 0.274781 Objective Loss 0.274781 LR 0.001000 Time 0.036438 +2023-10-02 21:04:53,183 - Epoch: [75][ 60/ 1236] Overall Loss 0.266955 Objective Loss 0.266955 LR 0.001000 Time 0.033787 +2023-10-02 21:04:53,387 - Epoch: [75][ 70/ 1236] Overall Loss 0.264937 Objective Loss 0.264937 LR 0.001000 Time 0.031853 +2023-10-02 21:04:53,594 - Epoch: [75][ 80/ 1236] Overall Loss 0.260508 Objective Loss 0.260508 LR 0.001000 Time 0.030458 +2023-10-02 21:04:53,797 - Epoch: [75][ 90/ 1236] Overall Loss 0.257410 Objective Loss 0.257410 LR 0.001000 Time 0.029325 +2023-10-02 21:04:54,004 - Epoch: [75][ 100/ 1236] Overall Loss 0.256209 Objective Loss 0.256209 LR 0.001000 Time 0.028457 +2023-10-02 21:04:54,207 - Epoch: [75][ 110/ 1236] Overall Loss 0.253936 Objective Loss 0.253936 LR 0.001000 Time 0.027713 +2023-10-02 21:04:54,414 - Epoch: [75][ 120/ 1236] Overall Loss 0.253657 Objective Loss 0.253657 LR 0.001000 Time 0.027126 +2023-10-02 21:04:54,617 - Epoch: [75][ 130/ 1236] Overall Loss 0.256229 Objective Loss 0.256229 LR 0.001000 Time 0.026600 +2023-10-02 21:04:54,824 - Epoch: [75][ 140/ 1236] Overall Loss 0.256594 Objective Loss 0.256594 LR 0.001000 Time 0.026176 +2023-10-02 21:04:55,027 - Epoch: [75][ 150/ 1236] Overall Loss 0.255997 Objective Loss 0.255997 LR 0.001000 Time 0.025782 +2023-10-02 21:04:55,234 - Epoch: [75][ 160/ 1236] Overall Loss 0.257240 Objective Loss 0.257240 LR 0.001000 Time 0.025463 +2023-10-02 21:04:55,437 - Epoch: [75][ 170/ 1236] Overall Loss 0.257477 Objective Loss 0.257477 LR 0.001000 Time 0.025157 +2023-10-02 21:04:55,644 - Epoch: [75][ 180/ 1236] Overall Loss 0.257254 Objective Loss 0.257254 LR 0.001000 Time 0.024909 +2023-10-02 21:04:55,847 - Epoch: [75][ 190/ 1236] Overall Loss 0.256751 Objective Loss 0.256751 LR 0.001000 Time 0.024665 +2023-10-02 21:04:56,054 - Epoch: [75][ 200/ 1236] Overall Loss 0.256066 Objective Loss 0.256066 LR 0.001000 Time 0.024465 +2023-10-02 21:04:56,257 - Epoch: [75][ 210/ 1236] Overall Loss 0.256337 Objective Loss 0.256337 LR 0.001000 Time 0.024266 +2023-10-02 21:04:56,464 - Epoch: [75][ 220/ 1236] Overall Loss 0.255442 Objective Loss 0.255442 LR 0.001000 Time 0.024104 +2023-10-02 21:04:56,667 - Epoch: [75][ 230/ 1236] Overall Loss 0.255022 Objective Loss 0.255022 LR 0.001000 Time 0.023937 +2023-10-02 21:04:56,874 - Epoch: [75][ 240/ 1236] Overall Loss 0.254807 Objective Loss 0.254807 LR 0.001000 Time 0.023801 +2023-10-02 21:04:57,077 - Epoch: [75][ 250/ 1236] Overall Loss 0.254407 Objective Loss 0.254407 LR 0.001000 Time 0.023660 +2023-10-02 21:04:57,284 - Epoch: [75][ 260/ 1236] Overall Loss 0.253510 Objective Loss 0.253510 LR 0.001000 Time 0.023545 +2023-10-02 21:04:57,487 - Epoch: [75][ 270/ 1236] Overall Loss 0.253130 Objective Loss 0.253130 LR 0.001000 Time 0.023424 +2023-10-02 21:04:57,694 - Epoch: [75][ 280/ 1236] Overall Loss 0.252021 Objective Loss 0.252021 LR 0.001000 Time 0.023325 +2023-10-02 21:04:57,897 - Epoch: [75][ 290/ 1236] Overall Loss 0.251526 Objective Loss 0.251526 LR 0.001000 Time 0.023220 +2023-10-02 21:04:58,104 - Epoch: [75][ 300/ 1236] Overall Loss 0.251689 Objective Loss 0.251689 LR 0.001000 Time 0.023135 +2023-10-02 21:04:58,307 - Epoch: [75][ 310/ 1236] Overall Loss 0.251928 Objective Loss 0.251928 LR 0.001000 Time 0.023043 +2023-10-02 21:04:58,514 - Epoch: [75][ 320/ 1236] Overall Loss 0.251818 Objective Loss 0.251818 LR 0.001000 Time 0.022969 +2023-10-02 21:04:58,717 - Epoch: [75][ 330/ 1236] Overall Loss 0.252102 Objective Loss 0.252102 LR 0.001000 Time 0.022888 +2023-10-02 21:04:58,924 - Epoch: [75][ 340/ 1236] Overall Loss 0.252622 Objective Loss 0.252622 LR 0.001000 Time 0.022823 +2023-10-02 21:04:59,127 - Epoch: [75][ 350/ 1236] Overall Loss 0.252914 Objective Loss 0.252914 LR 0.001000 Time 0.022750 +2023-10-02 21:04:59,335 - Epoch: [75][ 360/ 1236] Overall Loss 0.252601 Objective Loss 0.252601 LR 0.001000 Time 0.022693 +2023-10-02 21:04:59,538 - Epoch: [75][ 370/ 1236] Overall Loss 0.252452 Objective Loss 0.252452 LR 0.001000 Time 0.022629 +2023-10-02 21:04:59,744 - Epoch: [75][ 380/ 1236] Overall Loss 0.252855 Objective Loss 0.252855 LR 0.001000 Time 0.022575 +2023-10-02 21:04:59,949 - Epoch: [75][ 390/ 1236] Overall Loss 0.253163 Objective Loss 0.253163 LR 0.001000 Time 0.022516 +2023-10-02 21:05:00,155 - Epoch: [75][ 400/ 1236] Overall Loss 0.253599 Objective Loss 0.253599 LR 0.001000 Time 0.022468 +2023-10-02 21:05:00,359 - Epoch: [75][ 410/ 1236] Overall Loss 0.254190 Objective Loss 0.254190 LR 0.001000 Time 0.022415 +2023-10-02 21:05:00,565 - Epoch: [75][ 420/ 1236] Overall Loss 0.254638 Objective Loss 0.254638 LR 0.001000 Time 0.022371 +2023-10-02 21:05:00,770 - Epoch: [75][ 430/ 1236] Overall Loss 0.255089 Objective Loss 0.255089 LR 0.001000 Time 0.022323 +2023-10-02 21:05:00,974 - Epoch: [75][ 440/ 1236] Overall Loss 0.255006 Objective Loss 0.255006 LR 0.001000 Time 0.022280 +2023-10-02 21:05:01,179 - Epoch: [75][ 450/ 1236] Overall Loss 0.255098 Objective Loss 0.255098 LR 0.001000 Time 0.022238 +2023-10-02 21:05:01,384 - Epoch: [75][ 460/ 1236] Overall Loss 0.254881 Objective Loss 0.254881 LR 0.001000 Time 0.022202 +2023-10-02 21:05:01,589 - Epoch: [75][ 470/ 1236] Overall Loss 0.255465 Objective Loss 0.255465 LR 0.001000 Time 0.022162 +2023-10-02 21:05:01,795 - Epoch: [75][ 480/ 1236] Overall Loss 0.255940 Objective Loss 0.255940 LR 0.001000 Time 0.022128 +2023-10-02 21:05:02,000 - Epoch: [75][ 490/ 1236] Overall Loss 0.256208 Objective Loss 0.256208 LR 0.001000 Time 0.022091 +2023-10-02 21:05:02,206 - Epoch: [75][ 500/ 1236] Overall Loss 0.256103 Objective Loss 0.256103 LR 0.001000 Time 0.022061 +2023-10-02 21:05:02,410 - Epoch: [75][ 510/ 1236] Overall Loss 0.256552 Objective Loss 0.256552 LR 0.001000 Time 0.022027 +2023-10-02 21:05:02,617 - Epoch: [75][ 520/ 1236] Overall Loss 0.256381 Objective Loss 0.256381 LR 0.001000 Time 0.021999 +2023-10-02 21:05:02,821 - Epoch: [75][ 530/ 1236] Overall Loss 0.256300 Objective Loss 0.256300 LR 0.001000 Time 0.021967 +2023-10-02 21:05:03,027 - Epoch: [75][ 540/ 1236] Overall Loss 0.256331 Objective Loss 0.256331 LR 0.001000 Time 0.021941 +2023-10-02 21:05:03,232 - Epoch: [75][ 550/ 1236] Overall Loss 0.256857 Objective Loss 0.256857 LR 0.001000 Time 0.021911 +2023-10-02 21:05:03,438 - Epoch: [75][ 560/ 1236] Overall Loss 0.256912 Objective Loss 0.256912 LR 0.001000 Time 0.021887 +2023-10-02 21:05:03,642 - Epoch: [75][ 570/ 1236] Overall Loss 0.257057 Objective Loss 0.257057 LR 0.001000 Time 0.021860 +2023-10-02 21:05:03,848 - Epoch: [75][ 580/ 1236] Overall Loss 0.257021 Objective Loss 0.257021 LR 0.001000 Time 0.021837 +2023-10-02 21:05:04,052 - Epoch: [75][ 590/ 1236] Overall Loss 0.256592 Objective Loss 0.256592 LR 0.001000 Time 0.021811 +2023-10-02 21:05:04,258 - Epoch: [75][ 600/ 1236] Overall Loss 0.256454 Objective Loss 0.256454 LR 0.001000 Time 0.021790 +2023-10-02 21:05:04,463 - Epoch: [75][ 610/ 1236] Overall Loss 0.256321 Objective Loss 0.256321 LR 0.001000 Time 0.021767 +2023-10-02 21:05:04,669 - Epoch: [75][ 620/ 1236] Overall Loss 0.256139 Objective Loss 0.256139 LR 0.001000 Time 0.021747 +2023-10-02 21:05:04,874 - Epoch: [75][ 630/ 1236] Overall Loss 0.256414 Objective Loss 0.256414 LR 0.001000 Time 0.021725 +2023-10-02 21:05:05,080 - Epoch: [75][ 640/ 1236] Overall Loss 0.256531 Objective Loss 0.256531 LR 0.001000 Time 0.021707 +2023-10-02 21:05:05,295 - Epoch: [75][ 650/ 1236] Overall Loss 0.256591 Objective Loss 0.256591 LR 0.001000 Time 0.021701 +2023-10-02 21:05:05,511 - Epoch: [75][ 660/ 1236] Overall Loss 0.256724 Objective Loss 0.256724 LR 0.001000 Time 0.021698 +2023-10-02 21:05:05,731 - Epoch: [75][ 670/ 1236] Overall Loss 0.256749 Objective Loss 0.256749 LR 0.001000 Time 0.021702 +2023-10-02 21:05:05,946 - Epoch: [75][ 680/ 1236] Overall Loss 0.256682 Objective Loss 0.256682 LR 0.001000 Time 0.021700 +2023-10-02 21:05:06,157 - Epoch: [75][ 690/ 1236] Overall Loss 0.256817 Objective Loss 0.256817 LR 0.001000 Time 0.021690 +2023-10-02 21:05:06,364 - Epoch: [75][ 700/ 1236] Overall Loss 0.256513 Objective Loss 0.256513 LR 0.001000 Time 0.021675 +2023-10-02 21:05:06,569 - Epoch: [75][ 710/ 1236] Overall Loss 0.257112 Objective Loss 0.257112 LR 0.001000 Time 0.021657 +2023-10-02 21:05:06,776 - Epoch: [75][ 720/ 1236] Overall Loss 0.256400 Objective Loss 0.256400 LR 0.001000 Time 0.021644 +2023-10-02 21:05:06,982 - Epoch: [75][ 730/ 1236] Overall Loss 0.256103 Objective Loss 0.256103 LR 0.001000 Time 0.021627 +2023-10-02 21:05:07,190 - Epoch: [75][ 740/ 1236] Overall Loss 0.256026 Objective Loss 0.256026 LR 0.001000 Time 0.021615 +2023-10-02 21:05:07,394 - Epoch: [75][ 750/ 1236] Overall Loss 0.256377 Objective Loss 0.256377 LR 0.001000 Time 0.021599 +2023-10-02 21:05:07,601 - Epoch: [75][ 760/ 1236] Overall Loss 0.256608 Objective Loss 0.256608 LR 0.001000 Time 0.021587 +2023-10-02 21:05:07,807 - Epoch: [75][ 770/ 1236] Overall Loss 0.256722 Objective Loss 0.256722 LR 0.001000 Time 0.021571 +2023-10-02 21:05:08,015 - Epoch: [75][ 780/ 1236] Overall Loss 0.256989 Objective Loss 0.256989 LR 0.001000 Time 0.021562 +2023-10-02 21:05:08,219 - Epoch: [75][ 790/ 1236] Overall Loss 0.257032 Objective Loss 0.257032 LR 0.001000 Time 0.021547 +2023-10-02 21:05:08,428 - Epoch: [75][ 800/ 1236] Overall Loss 0.257195 Objective Loss 0.257195 LR 0.001000 Time 0.021538 +2023-10-02 21:05:08,632 - Epoch: [75][ 810/ 1236] Overall Loss 0.256884 Objective Loss 0.256884 LR 0.001000 Time 0.021524 +2023-10-02 21:05:08,839 - Epoch: [75][ 820/ 1236] Overall Loss 0.257332 Objective Loss 0.257332 LR 0.001000 Time 0.021513 +2023-10-02 21:05:09,044 - Epoch: [75][ 830/ 1236] Overall Loss 0.257383 Objective Loss 0.257383 LR 0.001000 Time 0.021500 +2023-10-02 21:05:09,251 - Epoch: [75][ 840/ 1236] Overall Loss 0.257064 Objective Loss 0.257064 LR 0.001000 Time 0.021490 +2023-10-02 21:05:09,456 - Epoch: [75][ 850/ 1236] Overall Loss 0.257021 Objective Loss 0.257021 LR 0.001000 Time 0.021476 +2023-10-02 21:05:09,663 - Epoch: [75][ 860/ 1236] Overall Loss 0.257355 Objective Loss 0.257355 LR 0.001000 Time 0.021467 +2023-10-02 21:05:09,869 - Epoch: [75][ 870/ 1236] Overall Loss 0.257360 Objective Loss 0.257360 LR 0.001000 Time 0.021455 +2023-10-02 21:05:10,076 - Epoch: [75][ 880/ 1236] Overall Loss 0.257299 Objective Loss 0.257299 LR 0.001000 Time 0.021446 +2023-10-02 21:05:10,281 - Epoch: [75][ 890/ 1236] Overall Loss 0.257286 Objective Loss 0.257286 LR 0.001000 Time 0.021434 +2023-10-02 21:05:10,489 - Epoch: [75][ 900/ 1236] Overall Loss 0.257218 Objective Loss 0.257218 LR 0.001000 Time 0.021426 +2023-10-02 21:05:10,694 - Epoch: [75][ 910/ 1236] Overall Loss 0.257466 Objective Loss 0.257466 LR 0.001000 Time 0.021414 +2023-10-02 21:05:10,901 - Epoch: [75][ 920/ 1236] Overall Loss 0.257364 Objective Loss 0.257364 LR 0.001000 Time 0.021406 +2023-10-02 21:05:11,106 - Epoch: [75][ 930/ 1236] Overall Loss 0.257343 Objective Loss 0.257343 LR 0.001000 Time 0.021395 +2023-10-02 21:05:11,313 - Epoch: [75][ 940/ 1236] Overall Loss 0.257436 Objective Loss 0.257436 LR 0.001000 Time 0.021387 +2023-10-02 21:05:11,518 - Epoch: [75][ 950/ 1236] Overall Loss 0.257875 Objective Loss 0.257875 LR 0.001000 Time 0.021377 +2023-10-02 21:05:11,725 - Epoch: [75][ 960/ 1236] Overall Loss 0.258218 Objective Loss 0.258218 LR 0.001000 Time 0.021369 +2023-10-02 21:05:11,930 - Epoch: [75][ 970/ 1236] Overall Loss 0.258571 Objective Loss 0.258571 LR 0.001000 Time 0.021359 +2023-10-02 21:05:12,137 - Epoch: [75][ 980/ 1236] Overall Loss 0.258537 Objective Loss 0.258537 LR 0.001000 Time 0.021352 +2023-10-02 21:05:12,343 - Epoch: [75][ 990/ 1236] Overall Loss 0.258714 Objective Loss 0.258714 LR 0.001000 Time 0.021342 +2023-10-02 21:05:12,550 - Epoch: [75][ 1000/ 1236] Overall Loss 0.259060 Objective Loss 0.259060 LR 0.001000 Time 0.021336 +2023-10-02 21:05:12,755 - Epoch: [75][ 1010/ 1236] Overall Loss 0.259259 Objective Loss 0.259259 LR 0.001000 Time 0.021326 +2023-10-02 21:05:12,964 - Epoch: [75][ 1020/ 1236] Overall Loss 0.259534 Objective Loss 0.259534 LR 0.001000 Time 0.021321 +2023-10-02 21:05:13,168 - Epoch: [75][ 1030/ 1236] Overall Loss 0.259697 Objective Loss 0.259697 LR 0.001000 Time 0.021312 +2023-10-02 21:05:13,376 - Epoch: [75][ 1040/ 1236] Overall Loss 0.260011 Objective Loss 0.260011 LR 0.001000 Time 0.021307 +2023-10-02 21:05:13,580 - Epoch: [75][ 1050/ 1236] Overall Loss 0.259705 Objective Loss 0.259705 LR 0.001000 Time 0.021299 +2023-10-02 21:05:13,788 - Epoch: [75][ 1060/ 1236] Overall Loss 0.259744 Objective Loss 0.259744 LR 0.001000 Time 0.021293 +2023-10-02 21:05:13,993 - Epoch: [75][ 1070/ 1236] Overall Loss 0.259831 Objective Loss 0.259831 LR 0.001000 Time 0.021284 +2023-10-02 21:05:14,200 - Epoch: [75][ 1080/ 1236] Overall Loss 0.259838 Objective Loss 0.259838 LR 0.001000 Time 0.021279 +2023-10-02 21:05:14,406 - Epoch: [75][ 1090/ 1236] Overall Loss 0.259647 Objective Loss 0.259647 LR 0.001000 Time 0.021271 +2023-10-02 21:05:14,614 - Epoch: [75][ 1100/ 1236] Overall Loss 0.259621 Objective Loss 0.259621 LR 0.001000 Time 0.021267 +2023-10-02 21:05:14,818 - Epoch: [75][ 1110/ 1236] Overall Loss 0.259729 Objective Loss 0.259729 LR 0.001000 Time 0.021259 +2023-10-02 21:05:15,025 - Epoch: [75][ 1120/ 1236] Overall Loss 0.259693 Objective Loss 0.259693 LR 0.001000 Time 0.021254 +2023-10-02 21:05:15,231 - Epoch: [75][ 1130/ 1236] Overall Loss 0.259708 Objective Loss 0.259708 LR 0.001000 Time 0.021246 +2023-10-02 21:05:15,437 - Epoch: [75][ 1140/ 1236] Overall Loss 0.259582 Objective Loss 0.259582 LR 0.001000 Time 0.021240 +2023-10-02 21:05:15,643 - Epoch: [75][ 1150/ 1236] Overall Loss 0.259528 Objective Loss 0.259528 LR 0.001000 Time 0.021234 +2023-10-02 21:05:15,851 - Epoch: [75][ 1160/ 1236] Overall Loss 0.259785 Objective Loss 0.259785 LR 0.001000 Time 0.021230 +2023-10-02 21:05:16,055 - Epoch: [75][ 1170/ 1236] Overall Loss 0.259982 Objective Loss 0.259982 LR 0.001000 Time 0.021223 +2023-10-02 21:05:16,262 - Epoch: [75][ 1180/ 1236] Overall Loss 0.260296 Objective Loss 0.260296 LR 0.001000 Time 0.021218 +2023-10-02 21:05:16,467 - Epoch: [75][ 1190/ 1236] Overall Loss 0.260175 Objective Loss 0.260175 LR 0.001000 Time 0.021212 +2023-10-02 21:05:16,674 - Epoch: [75][ 1200/ 1236] Overall Loss 0.260120 Objective Loss 0.260120 LR 0.001000 Time 0.021207 +2023-10-02 21:05:16,880 - Epoch: [75][ 1210/ 1236] Overall Loss 0.260169 Objective Loss 0.260169 LR 0.001000 Time 0.021201 +2023-10-02 21:05:17,087 - Epoch: [75][ 1220/ 1236] Overall Loss 0.260275 Objective Loss 0.260275 LR 0.001000 Time 0.021196 +2023-10-02 21:05:17,346 - Epoch: [75][ 1230/ 1236] Overall Loss 0.260346 Objective Loss 0.260346 LR 0.001000 Time 0.021233 +2023-10-02 21:05:17,468 - Epoch: [75][ 1236/ 1236] Overall Loss 0.260471 Objective Loss 0.260471 Top1 83.706721 Top5 97.759674 LR 0.001000 Time 0.021229 +2023-10-02 21:05:17,607 - --- validate (epoch=75)----------- +2023-10-02 21:05:17,607 - 29943 samples (256 per mini-batch) +2023-10-02 21:05:18,086 - Epoch: [75][ 10/ 117] Loss 0.351455 Top1 82.421875 Top5 97.695312 +2023-10-02 21:05:18,237 - Epoch: [75][ 20/ 117] Loss 0.325256 Top1 83.281250 Top5 98.085938 +2023-10-02 21:05:18,387 - Epoch: [75][ 30/ 117] Loss 0.329773 Top1 83.229167 Top5 98.046875 +2023-10-02 21:05:18,539 - Epoch: [75][ 40/ 117] Loss 0.334818 Top1 83.359375 Top5 97.978516 +2023-10-02 21:05:18,688 - Epoch: [75][ 50/ 117] Loss 0.335119 Top1 83.500000 Top5 98.000000 +2023-10-02 21:05:18,841 - Epoch: [75][ 60/ 117] Loss 0.326317 Top1 83.769531 Top5 98.059896 +2023-10-02 21:05:18,990 - Epoch: [75][ 70/ 117] Loss 0.327207 Top1 83.872768 Top5 98.097098 +2023-10-02 21:05:19,141 - Epoch: [75][ 80/ 117] Loss 0.328053 Top1 83.842773 Top5 98.134766 +2023-10-02 21:05:19,290 - Epoch: [75][ 90/ 117] Loss 0.335180 Top1 83.650174 Top5 98.138021 +2023-10-02 21:05:19,441 - Epoch: [75][ 100/ 117] Loss 0.337280 Top1 83.570312 Top5 98.140625 +2023-10-02 21:05:19,597 - Epoch: [75][ 110/ 117] Loss 0.337993 Top1 83.512074 Top5 98.117898 +2023-10-02 21:05:19,686 - Epoch: [75][ 117/ 117] Loss 0.335478 Top1 83.622216 Top5 98.143139 +2023-10-02 21:05:19,812 - ==> Top1: 83.622 Top5: 98.143 Loss: 0.335 + +2023-10-02 21:05:19,813 - ==> Confusion: +[[ 925 0 4 1 5 4 0 0 4 72 2 0 2 2 4 6 3 1 0 0 15] + [ 0 1020 2 0 6 35 3 24 3 1 2 3 0 1 0 5 2 0 12 4 8] + [ 5 0 952 10 4 1 37 4 0 1 1 5 5 1 0 9 1 1 6 4 9] + [ 1 3 13 968 0 7 3 1 4 2 9 0 9 5 25 4 1 4 10 2 18] + [ 20 4 1 1 968 6 2 0 0 14 0 1 0 2 7 8 10 0 0 2 4] + [ 0 24 1 1 5 988 1 30 0 4 2 10 1 12 5 2 4 0 6 8 12] + [ 0 3 18 0 0 1 1140 1 0 0 4 4 0 0 0 8 0 0 1 4 7] + [ 4 12 17 0 8 40 9 1041 1 4 8 9 2 3 0 3 1 1 38 9 8] + [ 20 3 3 1 2 5 1 1 962 41 6 2 1 16 17 0 1 0 0 3 4] + [ 97 0 2 2 7 2 1 1 20 951 2 0 0 12 8 3 1 0 0 1 9] + [ 3 0 9 10 2 4 8 5 13 2 950 4 0 19 7 0 0 3 3 4 7] + [ 2 1 5 0 0 7 0 2 0 0 1 943 22 6 0 13 4 18 0 9 2] + [ 1 1 2 3 1 5 6 1 0 0 1 46 949 2 1 17 1 12 1 4 14] + [ 1 0 1 0 2 12 1 1 3 17 6 7 0 1051 3 4 1 1 0 0 8] + [ 9 1 6 18 2 0 0 0 20 4 5 0 7 6 997 1 4 3 6 0 12] + [ 1 0 2 0 4 1 1 0 0 0 0 5 6 0 1 1092 6 5 1 6 3] + [ 2 8 1 0 3 9 0 0 0 0 0 3 2 1 4 20 1088 0 0 7 13] + [ 0 0 1 4 0 1 2 0 0 1 0 7 19 0 4 23 4 968 0 0 4] + [ 3 4 6 21 1 2 0 20 5 1 4 0 2 0 9 0 0 0 980 1 9] + [ 0 0 1 3 1 4 15 6 0 1 1 11 2 2 0 8 6 1 0 1078 12] + [ 164 132 156 82 103 218 75 74 61 80 189 148 302 307 130 147 110 59 151 189 5028]] + +2023-10-02 21:05:19,814 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:05:19,814 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:05:19,820 - + +2023-10-02 21:05:19,820 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:05:20,947 - Epoch: [76][ 10/ 1236] Overall Loss 0.272281 Objective Loss 0.272281 LR 0.001000 Time 0.112634 +2023-10-02 21:05:21,155 - Epoch: [76][ 20/ 1236] Overall Loss 0.277483 Objective Loss 0.277483 LR 0.001000 Time 0.066709 +2023-10-02 21:05:21,362 - Epoch: [76][ 30/ 1236] Overall Loss 0.275478 Objective Loss 0.275478 LR 0.001000 Time 0.051306 +2023-10-02 21:05:21,570 - Epoch: [76][ 40/ 1236] Overall Loss 0.269119 Objective Loss 0.269119 LR 0.001000 Time 0.043671 +2023-10-02 21:05:21,776 - Epoch: [76][ 50/ 1236] Overall Loss 0.264460 Objective Loss 0.264460 LR 0.001000 Time 0.039027 +2023-10-02 21:05:21,984 - Epoch: [76][ 60/ 1236] Overall Loss 0.257265 Objective Loss 0.257265 LR 0.001000 Time 0.035984 +2023-10-02 21:05:22,190 - Epoch: [76][ 70/ 1236] Overall Loss 0.254568 Objective Loss 0.254568 LR 0.001000 Time 0.033765 +2023-10-02 21:05:22,398 - Epoch: [76][ 80/ 1236] Overall Loss 0.250173 Objective Loss 0.250173 LR 0.001000 Time 0.032139 +2023-10-02 21:05:22,602 - Epoch: [76][ 90/ 1236] Overall Loss 0.251653 Objective Loss 0.251653 LR 0.001000 Time 0.030820 +2023-10-02 21:05:22,812 - Epoch: [76][ 100/ 1236] Overall Loss 0.254034 Objective Loss 0.254034 LR 0.001000 Time 0.029838 +2023-10-02 21:05:23,018 - Epoch: [76][ 110/ 1236] Overall Loss 0.251931 Objective Loss 0.251931 LR 0.001000 Time 0.028994 +2023-10-02 21:05:23,227 - Epoch: [76][ 120/ 1236] Overall Loss 0.252942 Objective Loss 0.252942 LR 0.001000 Time 0.028317 +2023-10-02 21:05:23,434 - Epoch: [76][ 130/ 1236] Overall Loss 0.252086 Objective Loss 0.252086 LR 0.001000 Time 0.027718 +2023-10-02 21:05:23,642 - Epoch: [76][ 140/ 1236] Overall Loss 0.251716 Objective Loss 0.251716 LR 0.001000 Time 0.027222 +2023-10-02 21:05:23,848 - Epoch: [76][ 150/ 1236] Overall Loss 0.251764 Objective Loss 0.251764 LR 0.001000 Time 0.026774 +2023-10-02 21:05:24,056 - Epoch: [76][ 160/ 1236] Overall Loss 0.252525 Objective Loss 0.252525 LR 0.001000 Time 0.026398 +2023-10-02 21:05:24,262 - Epoch: [76][ 170/ 1236] Overall Loss 0.253670 Objective Loss 0.253670 LR 0.001000 Time 0.026048 +2023-10-02 21:05:24,471 - Epoch: [76][ 180/ 1236] Overall Loss 0.252794 Objective Loss 0.252794 LR 0.001000 Time 0.025761 +2023-10-02 21:05:24,677 - Epoch: [76][ 190/ 1236] Overall Loss 0.251513 Objective Loss 0.251513 LR 0.001000 Time 0.025484 +2023-10-02 21:05:24,886 - Epoch: [76][ 200/ 1236] Overall Loss 0.251784 Objective Loss 0.251784 LR 0.001000 Time 0.025257 +2023-10-02 21:05:25,091 - Epoch: [76][ 210/ 1236] Overall Loss 0.250947 Objective Loss 0.250947 LR 0.001000 Time 0.025028 +2023-10-02 21:05:25,299 - Epoch: [76][ 220/ 1236] Overall Loss 0.250625 Objective Loss 0.250625 LR 0.001000 Time 0.024835 +2023-10-02 21:05:25,505 - Epoch: [76][ 230/ 1236] Overall Loss 0.251213 Objective Loss 0.251213 LR 0.001000 Time 0.024646 +2023-10-02 21:05:25,715 - Epoch: [76][ 240/ 1236] Overall Loss 0.252161 Objective Loss 0.252161 LR 0.001000 Time 0.024491 +2023-10-02 21:05:25,920 - Epoch: [76][ 250/ 1236] Overall Loss 0.252288 Objective Loss 0.252288 LR 0.001000 Time 0.024330 +2023-10-02 21:05:26,129 - Epoch: [76][ 260/ 1236] Overall Loss 0.252745 Objective Loss 0.252745 LR 0.001000 Time 0.024198 +2023-10-02 21:05:26,334 - Epoch: [76][ 270/ 1236] Overall Loss 0.252747 Objective Loss 0.252747 LR 0.001000 Time 0.024061 +2023-10-02 21:05:26,544 - Epoch: [76][ 280/ 1236] Overall Loss 0.252986 Objective Loss 0.252986 LR 0.001000 Time 0.023948 +2023-10-02 21:05:26,749 - Epoch: [76][ 290/ 1236] Overall Loss 0.252182 Objective Loss 0.252182 LR 0.001000 Time 0.023828 +2023-10-02 21:05:26,958 - Epoch: [76][ 300/ 1236] Overall Loss 0.252190 Objective Loss 0.252190 LR 0.001000 Time 0.023731 +2023-10-02 21:05:27,163 - Epoch: [76][ 310/ 1236] Overall Loss 0.253007 Objective Loss 0.253007 LR 0.001000 Time 0.023626 +2023-10-02 21:05:27,373 - Epoch: [76][ 320/ 1236] Overall Loss 0.253925 Objective Loss 0.253925 LR 0.001000 Time 0.023541 +2023-10-02 21:05:27,578 - Epoch: [76][ 330/ 1236] Overall Loss 0.253664 Objective Loss 0.253664 LR 0.001000 Time 0.023448 +2023-10-02 21:05:27,787 - Epoch: [76][ 340/ 1236] Overall Loss 0.253494 Objective Loss 0.253494 LR 0.001000 Time 0.023374 +2023-10-02 21:05:27,992 - Epoch: [76][ 350/ 1236] Overall Loss 0.253404 Objective Loss 0.253404 LR 0.001000 Time 0.023292 +2023-10-02 21:05:28,202 - Epoch: [76][ 360/ 1236] Overall Loss 0.253244 Objective Loss 0.253244 LR 0.001000 Time 0.023225 +2023-10-02 21:05:28,408 - Epoch: [76][ 370/ 1236] Overall Loss 0.253861 Objective Loss 0.253861 LR 0.001000 Time 0.023153 +2023-10-02 21:05:28,618 - Epoch: [76][ 380/ 1236] Overall Loss 0.254045 Objective Loss 0.254045 LR 0.001000 Time 0.023097 +2023-10-02 21:05:28,824 - Epoch: [76][ 390/ 1236] Overall Loss 0.253791 Objective Loss 0.253791 LR 0.001000 Time 0.023032 +2023-10-02 21:05:29,033 - Epoch: [76][ 400/ 1236] Overall Loss 0.254211 Objective Loss 0.254211 LR 0.001000 Time 0.022978 +2023-10-02 21:05:29,240 - Epoch: [76][ 410/ 1236] Overall Loss 0.254602 Objective Loss 0.254602 LR 0.001000 Time 0.022920 +2023-10-02 21:05:29,451 - Epoch: [76][ 420/ 1236] Overall Loss 0.254862 Objective Loss 0.254862 LR 0.001000 Time 0.022874 +2023-10-02 21:05:29,657 - Epoch: [76][ 430/ 1236] Overall Loss 0.255704 Objective Loss 0.255704 LR 0.001000 Time 0.022821 +2023-10-02 21:05:29,866 - Epoch: [76][ 440/ 1236] Overall Loss 0.256177 Objective Loss 0.256177 LR 0.001000 Time 0.022777 +2023-10-02 21:05:30,073 - Epoch: [76][ 450/ 1236] Overall Loss 0.255817 Objective Loss 0.255817 LR 0.001000 Time 0.022728 +2023-10-02 21:05:30,284 - Epoch: [76][ 460/ 1236] Overall Loss 0.255935 Objective Loss 0.255935 LR 0.001000 Time 0.022690 +2023-10-02 21:05:30,490 - Epoch: [76][ 470/ 1236] Overall Loss 0.256068 Objective Loss 0.256068 LR 0.001000 Time 0.022645 +2023-10-02 21:05:30,700 - Epoch: [76][ 480/ 1236] Overall Loss 0.255794 Objective Loss 0.255794 LR 0.001000 Time 0.022611 +2023-10-02 21:05:30,906 - Epoch: [76][ 490/ 1236] Overall Loss 0.255371 Objective Loss 0.255371 LR 0.001000 Time 0.022570 +2023-10-02 21:05:31,117 - Epoch: [76][ 500/ 1236] Overall Loss 0.255928 Objective Loss 0.255928 LR 0.001000 Time 0.022539 +2023-10-02 21:05:31,323 - Epoch: [76][ 510/ 1236] Overall Loss 0.256271 Objective Loss 0.256271 LR 0.001000 Time 0.022500 +2023-10-02 21:05:31,533 - Epoch: [76][ 520/ 1236] Overall Loss 0.256092 Objective Loss 0.256092 LR 0.001000 Time 0.022472 +2023-10-02 21:05:31,739 - Epoch: [76][ 530/ 1236] Overall Loss 0.256317 Objective Loss 0.256317 LR 0.001000 Time 0.022436 +2023-10-02 21:05:31,949 - Epoch: [76][ 540/ 1236] Overall Loss 0.256902 Objective Loss 0.256902 LR 0.001000 Time 0.022409 +2023-10-02 21:05:32,155 - Epoch: [76][ 550/ 1236] Overall Loss 0.257349 Objective Loss 0.257349 LR 0.001000 Time 0.022375 +2023-10-02 21:05:32,366 - Epoch: [76][ 560/ 1236] Overall Loss 0.257663 Objective Loss 0.257663 LR 0.001000 Time 0.022351 +2023-10-02 21:05:32,572 - Epoch: [76][ 570/ 1236] Overall Loss 0.258127 Objective Loss 0.258127 LR 0.001000 Time 0.022320 +2023-10-02 21:05:32,781 - Epoch: [76][ 580/ 1236] Overall Loss 0.258210 Objective Loss 0.258210 LR 0.001000 Time 0.022296 +2023-10-02 21:05:32,989 - Epoch: [76][ 590/ 1236] Overall Loss 0.258383 Objective Loss 0.258383 LR 0.001000 Time 0.022266 +2023-10-02 21:05:33,199 - Epoch: [76][ 600/ 1236] Overall Loss 0.258888 Objective Loss 0.258888 LR 0.001000 Time 0.022245 +2023-10-02 21:05:33,405 - Epoch: [76][ 610/ 1236] Overall Loss 0.258934 Objective Loss 0.258934 LR 0.001000 Time 0.022218 +2023-10-02 21:05:33,616 - Epoch: [76][ 620/ 1236] Overall Loss 0.259091 Objective Loss 0.259091 LR 0.001000 Time 0.022199 +2023-10-02 21:05:33,821 - Epoch: [76][ 630/ 1236] Overall Loss 0.258882 Objective Loss 0.258882 LR 0.001000 Time 0.022173 +2023-10-02 21:05:34,032 - Epoch: [76][ 640/ 1236] Overall Loss 0.258543 Objective Loss 0.258543 LR 0.001000 Time 0.022155 +2023-10-02 21:05:34,238 - Epoch: [76][ 650/ 1236] Overall Loss 0.258569 Objective Loss 0.258569 LR 0.001000 Time 0.022131 +2023-10-02 21:05:34,449 - Epoch: [76][ 660/ 1236] Overall Loss 0.258631 Objective Loss 0.258631 LR 0.001000 Time 0.022114 +2023-10-02 21:05:34,655 - Epoch: [76][ 670/ 1236] Overall Loss 0.258407 Objective Loss 0.258407 LR 0.001000 Time 0.022091 +2023-10-02 21:05:34,865 - Epoch: [76][ 680/ 1236] Overall Loss 0.258475 Objective Loss 0.258475 LR 0.001000 Time 0.022075 +2023-10-02 21:05:35,072 - Epoch: [76][ 690/ 1236] Overall Loss 0.258625 Objective Loss 0.258625 LR 0.001000 Time 0.022054 +2023-10-02 21:05:35,282 - Epoch: [76][ 700/ 1236] Overall Loss 0.258750 Objective Loss 0.258750 LR 0.001000 Time 0.022039 +2023-10-02 21:05:35,488 - Epoch: [76][ 710/ 1236] Overall Loss 0.258838 Objective Loss 0.258838 LR 0.001000 Time 0.022018 +2023-10-02 21:05:35,699 - Epoch: [76][ 720/ 1236] Overall Loss 0.258693 Objective Loss 0.258693 LR 0.001000 Time 0.022005 +2023-10-02 21:05:35,905 - Epoch: [76][ 730/ 1236] Overall Loss 0.258690 Objective Loss 0.258690 LR 0.001000 Time 0.021985 +2023-10-02 21:05:36,115 - Epoch: [76][ 740/ 1236] Overall Loss 0.258892 Objective Loss 0.258892 LR 0.001000 Time 0.021972 +2023-10-02 21:05:36,321 - Epoch: [76][ 750/ 1236] Overall Loss 0.259013 Objective Loss 0.259013 LR 0.001000 Time 0.021953 +2023-10-02 21:05:36,531 - Epoch: [76][ 760/ 1236] Overall Loss 0.259130 Objective Loss 0.259130 LR 0.001000 Time 0.021941 +2023-10-02 21:05:36,737 - Epoch: [76][ 770/ 1236] Overall Loss 0.259040 Objective Loss 0.259040 LR 0.001000 Time 0.021923 +2023-10-02 21:05:36,948 - Epoch: [76][ 780/ 1236] Overall Loss 0.258970 Objective Loss 0.258970 LR 0.001000 Time 0.021911 +2023-10-02 21:05:37,154 - Epoch: [76][ 790/ 1236] Overall Loss 0.258928 Objective Loss 0.258928 LR 0.001000 Time 0.021894 +2023-10-02 21:05:37,364 - Epoch: [76][ 800/ 1236] Overall Loss 0.258807 Objective Loss 0.258807 LR 0.001000 Time 0.021883 +2023-10-02 21:05:37,570 - Epoch: [76][ 810/ 1236] Overall Loss 0.258650 Objective Loss 0.258650 LR 0.001000 Time 0.021867 +2023-10-02 21:05:37,781 - Epoch: [76][ 820/ 1236] Overall Loss 0.259194 Objective Loss 0.259194 LR 0.001000 Time 0.021857 +2023-10-02 21:05:37,987 - Epoch: [76][ 830/ 1236] Overall Loss 0.259161 Objective Loss 0.259161 LR 0.001000 Time 0.021842 +2023-10-02 21:05:38,197 - Epoch: [76][ 840/ 1236] Overall Loss 0.259543 Objective Loss 0.259543 LR 0.001000 Time 0.021831 +2023-10-02 21:05:38,404 - Epoch: [76][ 850/ 1236] Overall Loss 0.259285 Objective Loss 0.259285 LR 0.001000 Time 0.021816 +2023-10-02 21:05:38,614 - Epoch: [76][ 860/ 1236] Overall Loss 0.259628 Objective Loss 0.259628 LR 0.001000 Time 0.021805 +2023-10-02 21:05:38,821 - Epoch: [76][ 870/ 1236] Overall Loss 0.259710 Objective Loss 0.259710 LR 0.001000 Time 0.021791 +2023-10-02 21:05:39,030 - Epoch: [76][ 880/ 1236] Overall Loss 0.259699 Objective Loss 0.259699 LR 0.001000 Time 0.021781 +2023-10-02 21:05:39,237 - Epoch: [76][ 890/ 1236] Overall Loss 0.259587 Objective Loss 0.259587 LR 0.001000 Time 0.021768 +2023-10-02 21:05:39,447 - Epoch: [76][ 900/ 1236] Overall Loss 0.259611 Objective Loss 0.259611 LR 0.001000 Time 0.021758 +2023-10-02 21:05:39,654 - Epoch: [76][ 910/ 1236] Overall Loss 0.259620 Objective Loss 0.259620 LR 0.001000 Time 0.021745 +2023-10-02 21:05:39,865 - Epoch: [76][ 920/ 1236] Overall Loss 0.259713 Objective Loss 0.259713 LR 0.001000 Time 0.021738 +2023-10-02 21:05:40,071 - Epoch: [76][ 930/ 1236] Overall Loss 0.260003 Objective Loss 0.260003 LR 0.001000 Time 0.021725 +2023-10-02 21:05:40,282 - Epoch: [76][ 940/ 1236] Overall Loss 0.260070 Objective Loss 0.260070 LR 0.001000 Time 0.021718 +2023-10-02 21:05:40,488 - Epoch: [76][ 950/ 1236] Overall Loss 0.260243 Objective Loss 0.260243 LR 0.001000 Time 0.021706 +2023-10-02 21:05:40,697 - Epoch: [76][ 960/ 1236] Overall Loss 0.260087 Objective Loss 0.260087 LR 0.001000 Time 0.021698 +2023-10-02 21:05:40,905 - Epoch: [76][ 970/ 1236] Overall Loss 0.259850 Objective Loss 0.259850 LR 0.001000 Time 0.021686 +2023-10-02 21:05:41,115 - Epoch: [76][ 980/ 1236] Overall Loss 0.260423 Objective Loss 0.260423 LR 0.001000 Time 0.021679 +2023-10-02 21:05:41,321 - Epoch: [76][ 990/ 1236] Overall Loss 0.260108 Objective Loss 0.260108 LR 0.001000 Time 0.021668 +2023-10-02 21:05:41,530 - Epoch: [76][ 1000/ 1236] Overall Loss 0.260319 Objective Loss 0.260319 LR 0.001000 Time 0.021660 +2023-10-02 21:05:41,738 - Epoch: [76][ 1010/ 1236] Overall Loss 0.260462 Objective Loss 0.260462 LR 0.001000 Time 0.021650 +2023-10-02 21:05:41,948 - Epoch: [76][ 1020/ 1236] Overall Loss 0.260466 Objective Loss 0.260466 LR 0.001000 Time 0.021643 +2023-10-02 21:05:42,155 - Epoch: [76][ 1030/ 1236] Overall Loss 0.260862 Objective Loss 0.260862 LR 0.001000 Time 0.021633 +2023-10-02 21:05:42,364 - Epoch: [76][ 1040/ 1236] Overall Loss 0.261100 Objective Loss 0.261100 LR 0.001000 Time 0.021626 +2023-10-02 21:05:42,572 - Epoch: [76][ 1050/ 1236] Overall Loss 0.261072 Objective Loss 0.261072 LR 0.001000 Time 0.021616 +2023-10-02 21:05:42,781 - Epoch: [76][ 1060/ 1236] Overall Loss 0.261560 Objective Loss 0.261560 LR 0.001000 Time 0.021609 +2023-10-02 21:05:42,988 - Epoch: [76][ 1070/ 1236] Overall Loss 0.261673 Objective Loss 0.261673 LR 0.001000 Time 0.021601 +2023-10-02 21:05:43,198 - Epoch: [76][ 1080/ 1236] Overall Loss 0.261909 Objective Loss 0.261909 LR 0.001000 Time 0.021594 +2023-10-02 21:05:43,405 - Epoch: [76][ 1090/ 1236] Overall Loss 0.262134 Objective Loss 0.262134 LR 0.001000 Time 0.021585 +2023-10-02 21:05:43,614 - Epoch: [76][ 1100/ 1236] Overall Loss 0.261998 Objective Loss 0.261998 LR 0.001000 Time 0.021579 +2023-10-02 21:05:43,822 - Epoch: [76][ 1110/ 1236] Overall Loss 0.261940 Objective Loss 0.261940 LR 0.001000 Time 0.021570 +2023-10-02 21:05:44,032 - Epoch: [76][ 1120/ 1236] Overall Loss 0.262042 Objective Loss 0.262042 LR 0.001000 Time 0.021565 +2023-10-02 21:05:44,239 - Epoch: [76][ 1130/ 1236] Overall Loss 0.262097 Objective Loss 0.262097 LR 0.001000 Time 0.021556 +2023-10-02 21:05:44,448 - Epoch: [76][ 1140/ 1236] Overall Loss 0.262104 Objective Loss 0.262104 LR 0.001000 Time 0.021551 +2023-10-02 21:05:44,656 - Epoch: [76][ 1150/ 1236] Overall Loss 0.262160 Objective Loss 0.262160 LR 0.001000 Time 0.021542 +2023-10-02 21:05:44,866 - Epoch: [76][ 1160/ 1236] Overall Loss 0.262393 Objective Loss 0.262393 LR 0.001000 Time 0.021538 +2023-10-02 21:05:45,073 - Epoch: [76][ 1170/ 1236] Overall Loss 0.262319 Objective Loss 0.262319 LR 0.001000 Time 0.021530 +2023-10-02 21:05:45,282 - Epoch: [76][ 1180/ 1236] Overall Loss 0.262405 Objective Loss 0.262405 LR 0.001000 Time 0.021525 +2023-10-02 21:05:45,489 - Epoch: [76][ 1190/ 1236] Overall Loss 0.262381 Objective Loss 0.262381 LR 0.001000 Time 0.021517 +2023-10-02 21:05:45,699 - Epoch: [76][ 1200/ 1236] Overall Loss 0.262665 Objective Loss 0.262665 LR 0.001000 Time 0.021512 +2023-10-02 21:05:45,906 - Epoch: [76][ 1210/ 1236] Overall Loss 0.262995 Objective Loss 0.262995 LR 0.001000 Time 0.021504 +2023-10-02 21:05:46,116 - Epoch: [76][ 1220/ 1236] Overall Loss 0.262885 Objective Loss 0.262885 LR 0.001000 Time 0.021500 +2023-10-02 21:05:46,375 - Epoch: [76][ 1230/ 1236] Overall Loss 0.262663 Objective Loss 0.262663 LR 0.001000 Time 0.021535 +2023-10-02 21:05:46,497 - Epoch: [76][ 1236/ 1236] Overall Loss 0.262785 Objective Loss 0.262785 Top1 87.780041 Top5 98.778004 LR 0.001000 Time 0.021529 +2023-10-02 21:05:46,620 - --- validate (epoch=76)----------- +2023-10-02 21:05:46,620 - 29943 samples (256 per mini-batch) +2023-10-02 21:05:47,102 - Epoch: [76][ 10/ 117] Loss 0.357800 Top1 83.085938 Top5 98.320312 +2023-10-02 21:05:47,253 - Epoch: [76][ 20/ 117] Loss 0.369764 Top1 82.519531 Top5 97.910156 +2023-10-02 21:05:47,404 - Epoch: [76][ 30/ 117] Loss 0.346501 Top1 83.046875 Top5 97.890625 +2023-10-02 21:05:47,554 - Epoch: [76][ 40/ 117] Loss 0.341135 Top1 83.291016 Top5 97.939453 +2023-10-02 21:05:47,705 - Epoch: [76][ 50/ 117] Loss 0.334866 Top1 83.546875 Top5 97.921875 +2023-10-02 21:05:47,856 - Epoch: [76][ 60/ 117] Loss 0.333008 Top1 83.639323 Top5 97.916667 +2023-10-02 21:05:48,007 - Epoch: [76][ 70/ 117] Loss 0.334430 Top1 83.498884 Top5 97.974330 +2023-10-02 21:05:48,158 - Epoch: [76][ 80/ 117] Loss 0.338059 Top1 83.437500 Top5 97.973633 +2023-10-02 21:05:48,309 - Epoch: [76][ 90/ 117] Loss 0.334584 Top1 83.411458 Top5 97.994792 +2023-10-02 21:05:48,460 - Epoch: [76][ 100/ 117] Loss 0.333632 Top1 83.527344 Top5 98.015625 +2023-10-02 21:05:48,617 - Epoch: [76][ 110/ 117] Loss 0.331752 Top1 83.572443 Top5 98.036222 +2023-10-02 21:05:48,706 - Epoch: [76][ 117/ 117] Loss 0.329663 Top1 83.668971 Top5 98.049628 +2023-10-02 21:05:48,820 - ==> Top1: 83.669 Top5: 98.050 Loss: 0.330 + +2023-10-02 21:05:48,820 - ==> Confusion: +[[ 912 0 4 0 7 4 0 1 8 90 2 0 1 2 3 0 3 1 0 0 12] + [ 2 1037 0 1 10 32 0 17 4 2 3 0 0 1 1 4 2 0 7 2 6] + [ 4 0 947 12 5 0 35 17 1 3 3 1 8 2 2 5 1 0 4 1 5] + [ 1 3 13 964 1 5 1 5 11 4 10 0 8 4 29 2 0 2 13 0 13] + [ 17 4 0 0 971 7 0 0 4 18 1 1 0 4 3 7 8 0 1 0 4] + [ 3 33 1 1 1 986 0 25 5 5 1 3 3 17 4 1 1 0 2 10 14] + [ 0 1 28 0 0 0 1132 5 0 1 4 1 0 0 0 3 0 0 1 9 6] + [ 1 21 10 0 4 36 7 1063 3 3 3 7 4 8 0 0 1 0 33 5 9] + [ 18 3 1 0 2 1 0 1 983 42 7 1 1 14 13 1 1 0 0 0 0] + [ 65 4 0 0 6 2 0 0 27 989 2 0 0 11 3 0 1 0 0 2 7] + [ 1 1 5 6 0 5 2 6 20 2 956 6 1 16 6 0 2 0 3 0 15] + [ 2 0 2 1 2 11 0 4 0 2 0 949 24 10 0 2 2 11 0 7 6] + [ 0 1 2 4 0 6 1 3 3 1 2 58 930 8 4 7 3 16 1 5 13] + [ 1 0 3 0 3 9 0 1 20 16 7 3 0 1036 6 0 1 0 0 0 13] + [ 14 1 4 11 7 0 0 0 25 8 6 0 3 2 1005 0 0 3 6 0 6] + [ 0 0 2 0 4 1 3 0 0 0 0 9 9 1 1 1066 16 10 2 4 6] + [ 0 14 2 0 6 8 0 0 1 0 0 3 2 3 5 8 1097 0 1 5 6] + [ 0 0 1 4 1 1 3 0 2 1 1 6 21 1 7 7 3 974 1 0 4] + [ 1 8 5 18 1 1 1 19 6 1 5 0 1 0 7 0 0 0 975 2 17] + [ 0 2 2 0 1 5 13 16 0 0 2 11 5 3 0 4 10 0 1 1069 8] + [ 159 190 136 86 100 179 55 110 144 140 175 107 315 287 145 67 149 50 124 175 5012]] + +2023-10-02 21:05:48,821 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:05:48,821 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:05:48,830 - + +2023-10-02 21:05:48,831 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:05:49,843 - Epoch: [77][ 10/ 1236] Overall Loss 0.250315 Objective Loss 0.250315 LR 0.001000 Time 0.101178 +2023-10-02 21:05:50,049 - Epoch: [77][ 20/ 1236] Overall Loss 0.259340 Objective Loss 0.259340 LR 0.001000 Time 0.060887 +2023-10-02 21:05:50,255 - Epoch: [77][ 30/ 1236] Overall Loss 0.263720 Objective Loss 0.263720 LR 0.001000 Time 0.047385 +2023-10-02 21:05:50,462 - Epoch: [77][ 40/ 1236] Overall Loss 0.271982 Objective Loss 0.271982 LR 0.001000 Time 0.040720 +2023-10-02 21:05:50,666 - Epoch: [77][ 50/ 1236] Overall Loss 0.266298 Objective Loss 0.266298 LR 0.001000 Time 0.036652 +2023-10-02 21:05:50,874 - Epoch: [77][ 60/ 1236] Overall Loss 0.267378 Objective Loss 0.267378 LR 0.001000 Time 0.034001 +2023-10-02 21:05:51,078 - Epoch: [77][ 70/ 1236] Overall Loss 0.261540 Objective Loss 0.261540 LR 0.001000 Time 0.032050 +2023-10-02 21:05:51,285 - Epoch: [77][ 80/ 1236] Overall Loss 0.263133 Objective Loss 0.263133 LR 0.001000 Time 0.030637 +2023-10-02 21:05:51,489 - Epoch: [77][ 90/ 1236] Overall Loss 0.261726 Objective Loss 0.261726 LR 0.001000 Time 0.029496 +2023-10-02 21:05:51,696 - Epoch: [77][ 100/ 1236] Overall Loss 0.259762 Objective Loss 0.259762 LR 0.001000 Time 0.028614 +2023-10-02 21:05:51,900 - Epoch: [77][ 110/ 1236] Overall Loss 0.261309 Objective Loss 0.261309 LR 0.001000 Time 0.027863 +2023-10-02 21:05:52,108 - Epoch: [77][ 120/ 1236] Overall Loss 0.262549 Objective Loss 0.262549 LR 0.001000 Time 0.027271 +2023-10-02 21:05:52,315 - Epoch: [77][ 130/ 1236] Overall Loss 0.259011 Objective Loss 0.259011 LR 0.001000 Time 0.026759 +2023-10-02 21:05:52,525 - Epoch: [77][ 140/ 1236] Overall Loss 0.259079 Objective Loss 0.259079 LR 0.001000 Time 0.026346 +2023-10-02 21:05:52,736 - Epoch: [77][ 150/ 1236] Overall Loss 0.258923 Objective Loss 0.258923 LR 0.001000 Time 0.025995 +2023-10-02 21:05:52,954 - Epoch: [77][ 160/ 1236] Overall Loss 0.256135 Objective Loss 0.256135 LR 0.001000 Time 0.025735 +2023-10-02 21:05:53,168 - Epoch: [77][ 170/ 1236] Overall Loss 0.254922 Objective Loss 0.254922 LR 0.001000 Time 0.025476 +2023-10-02 21:05:53,387 - Epoch: [77][ 180/ 1236] Overall Loss 0.253312 Objective Loss 0.253312 LR 0.001000 Time 0.025274 +2023-10-02 21:05:53,601 - Epoch: [77][ 190/ 1236] Overall Loss 0.254602 Objective Loss 0.254602 LR 0.001000 Time 0.025067 +2023-10-02 21:05:53,819 - Epoch: [77][ 200/ 1236] Overall Loss 0.256132 Objective Loss 0.256132 LR 0.001000 Time 0.024906 +2023-10-02 21:05:54,033 - Epoch: [77][ 210/ 1236] Overall Loss 0.256072 Objective Loss 0.256072 LR 0.001000 Time 0.024734 +2023-10-02 21:05:54,252 - Epoch: [77][ 220/ 1236] Overall Loss 0.255877 Objective Loss 0.255877 LR 0.001000 Time 0.024604 +2023-10-02 21:05:54,466 - Epoch: [77][ 230/ 1236] Overall Loss 0.256722 Objective Loss 0.256722 LR 0.001000 Time 0.024463 +2023-10-02 21:05:54,676 - Epoch: [77][ 240/ 1236] Overall Loss 0.256434 Objective Loss 0.256434 LR 0.001000 Time 0.024320 +2023-10-02 21:05:54,883 - Epoch: [77][ 250/ 1236] Overall Loss 0.255633 Objective Loss 0.255633 LR 0.001000 Time 0.024168 +2023-10-02 21:05:55,091 - Epoch: [77][ 260/ 1236] Overall Loss 0.256372 Objective Loss 0.256372 LR 0.001000 Time 0.024036 +2023-10-02 21:05:55,297 - Epoch: [77][ 270/ 1236] Overall Loss 0.256245 Objective Loss 0.256245 LR 0.001000 Time 0.023906 +2023-10-02 21:05:55,505 - Epoch: [77][ 280/ 1236] Overall Loss 0.256664 Objective Loss 0.256664 LR 0.001000 Time 0.023792 +2023-10-02 21:05:55,712 - Epoch: [77][ 290/ 1236] Overall Loss 0.257111 Objective Loss 0.257111 LR 0.001000 Time 0.023680 +2023-10-02 21:05:55,919 - Epoch: [77][ 300/ 1236] Overall Loss 0.257064 Objective Loss 0.257064 LR 0.001000 Time 0.023579 +2023-10-02 21:05:56,124 - Epoch: [77][ 310/ 1236] Overall Loss 0.257526 Objective Loss 0.257526 LR 0.001000 Time 0.023479 +2023-10-02 21:05:56,332 - Epoch: [77][ 320/ 1236] Overall Loss 0.257877 Objective Loss 0.257877 LR 0.001000 Time 0.023394 +2023-10-02 21:05:56,542 - Epoch: [77][ 330/ 1236] Overall Loss 0.257571 Objective Loss 0.257571 LR 0.001000 Time 0.023322 +2023-10-02 21:05:56,753 - Epoch: [77][ 340/ 1236] Overall Loss 0.257871 Objective Loss 0.257871 LR 0.001000 Time 0.023252 +2023-10-02 21:05:56,964 - Epoch: [77][ 350/ 1236] Overall Loss 0.257540 Objective Loss 0.257540 LR 0.001000 Time 0.023184 +2023-10-02 21:05:57,172 - Epoch: [77][ 360/ 1236] Overall Loss 0.257206 Objective Loss 0.257206 LR 0.001000 Time 0.023117 +2023-10-02 21:05:57,379 - Epoch: [77][ 370/ 1236] Overall Loss 0.257336 Objective Loss 0.257336 LR 0.001000 Time 0.023048 +2023-10-02 21:05:57,587 - Epoch: [77][ 380/ 1236] Overall Loss 0.257677 Objective Loss 0.257677 LR 0.001000 Time 0.022990 +2023-10-02 21:05:57,792 - Epoch: [77][ 390/ 1236] Overall Loss 0.257961 Objective Loss 0.257961 LR 0.001000 Time 0.022924 +2023-10-02 21:05:58,001 - Epoch: [77][ 400/ 1236] Overall Loss 0.257452 Objective Loss 0.257452 LR 0.001000 Time 0.022872 +2023-10-02 21:05:58,205 - Epoch: [77][ 410/ 1236] Overall Loss 0.257966 Objective Loss 0.257966 LR 0.001000 Time 0.022812 +2023-10-02 21:05:58,414 - Epoch: [77][ 420/ 1236] Overall Loss 0.257640 Objective Loss 0.257640 LR 0.001000 Time 0.022765 +2023-10-02 21:05:58,618 - Epoch: [77][ 430/ 1236] Overall Loss 0.257538 Objective Loss 0.257538 LR 0.001000 Time 0.022710 +2023-10-02 21:05:58,827 - Epoch: [77][ 440/ 1236] Overall Loss 0.258285 Objective Loss 0.258285 LR 0.001000 Time 0.022668 +2023-10-02 21:05:59,032 - Epoch: [77][ 450/ 1236] Overall Loss 0.258480 Objective Loss 0.258480 LR 0.001000 Time 0.022618 +2023-10-02 21:05:59,240 - Epoch: [77][ 460/ 1236] Overall Loss 0.259066 Objective Loss 0.259066 LR 0.001000 Time 0.022579 +2023-10-02 21:05:59,445 - Epoch: [77][ 470/ 1236] Overall Loss 0.258911 Objective Loss 0.258911 LR 0.001000 Time 0.022534 +2023-10-02 21:05:59,654 - Epoch: [77][ 480/ 1236] Overall Loss 0.259605 Objective Loss 0.259605 LR 0.001000 Time 0.022499 +2023-10-02 21:05:59,858 - Epoch: [77][ 490/ 1236] Overall Loss 0.259375 Objective Loss 0.259375 LR 0.001000 Time 0.022457 +2023-10-02 21:06:00,067 - Epoch: [77][ 500/ 1236] Overall Loss 0.259606 Objective Loss 0.259606 LR 0.001000 Time 0.022424 +2023-10-02 21:06:00,272 - Epoch: [77][ 510/ 1236] Overall Loss 0.259545 Objective Loss 0.259545 LR 0.001000 Time 0.022385 +2023-10-02 21:06:00,480 - Epoch: [77][ 520/ 1236] Overall Loss 0.258926 Objective Loss 0.258926 LR 0.001000 Time 0.022355 +2023-10-02 21:06:00,685 - Epoch: [77][ 530/ 1236] Overall Loss 0.259013 Objective Loss 0.259013 LR 0.001000 Time 0.022319 +2023-10-02 21:06:00,894 - Epoch: [77][ 540/ 1236] Overall Loss 0.259096 Objective Loss 0.259096 LR 0.001000 Time 0.022292 +2023-10-02 21:06:01,098 - Epoch: [77][ 550/ 1236] Overall Loss 0.259366 Objective Loss 0.259366 LR 0.001000 Time 0.022258 +2023-10-02 21:06:01,307 - Epoch: [77][ 560/ 1236] Overall Loss 0.259312 Objective Loss 0.259312 LR 0.001000 Time 0.022233 +2023-10-02 21:06:01,512 - Epoch: [77][ 570/ 1236] Overall Loss 0.259404 Objective Loss 0.259404 LR 0.001000 Time 0.022201 +2023-10-02 21:06:01,720 - Epoch: [77][ 580/ 1236] Overall Loss 0.259350 Objective Loss 0.259350 LR 0.001000 Time 0.022177 +2023-10-02 21:06:01,925 - Epoch: [77][ 590/ 1236] Overall Loss 0.259219 Objective Loss 0.259219 LR 0.001000 Time 0.022147 +2023-10-02 21:06:02,133 - Epoch: [77][ 600/ 1236] Overall Loss 0.258957 Objective Loss 0.258957 LR 0.001000 Time 0.022126 +2023-10-02 21:06:02,338 - Epoch: [77][ 610/ 1236] Overall Loss 0.259226 Objective Loss 0.259226 LR 0.001000 Time 0.022098 +2023-10-02 21:06:02,546 - Epoch: [77][ 620/ 1236] Overall Loss 0.259058 Objective Loss 0.259058 LR 0.001000 Time 0.022077 +2023-10-02 21:06:02,751 - Epoch: [77][ 630/ 1236] Overall Loss 0.258782 Objective Loss 0.258782 LR 0.001000 Time 0.022051 +2023-10-02 21:06:02,960 - Epoch: [77][ 640/ 1236] Overall Loss 0.258719 Objective Loss 0.258719 LR 0.001000 Time 0.022032 +2023-10-02 21:06:03,165 - Epoch: [77][ 650/ 1236] Overall Loss 0.258253 Objective Loss 0.258253 LR 0.001000 Time 0.022008 +2023-10-02 21:06:03,373 - Epoch: [77][ 660/ 1236] Overall Loss 0.258358 Objective Loss 0.258358 LR 0.001000 Time 0.021990 +2023-10-02 21:06:03,578 - Epoch: [77][ 670/ 1236] Overall Loss 0.257939 Objective Loss 0.257939 LR 0.001000 Time 0.021967 +2023-10-02 21:06:03,787 - Epoch: [77][ 680/ 1236] Overall Loss 0.257556 Objective Loss 0.257556 LR 0.001000 Time 0.021951 +2023-10-02 21:06:03,992 - Epoch: [77][ 690/ 1236] Overall Loss 0.257676 Objective Loss 0.257676 LR 0.001000 Time 0.021930 +2023-10-02 21:06:04,201 - Epoch: [77][ 700/ 1236] Overall Loss 0.257765 Objective Loss 0.257765 LR 0.001000 Time 0.021914 +2023-10-02 21:06:04,406 - Epoch: [77][ 710/ 1236] Overall Loss 0.257996 Objective Loss 0.257996 LR 0.001000 Time 0.021894 +2023-10-02 21:06:04,614 - Epoch: [77][ 720/ 1236] Overall Loss 0.258086 Objective Loss 0.258086 LR 0.001000 Time 0.021879 +2023-10-02 21:06:04,819 - Epoch: [77][ 730/ 1236] Overall Loss 0.258465 Objective Loss 0.258465 LR 0.001000 Time 0.021859 +2023-10-02 21:06:05,028 - Epoch: [77][ 740/ 1236] Overall Loss 0.258894 Objective Loss 0.258894 LR 0.001000 Time 0.021845 +2023-10-02 21:06:05,233 - Epoch: [77][ 750/ 1236] Overall Loss 0.258857 Objective Loss 0.258857 LR 0.001000 Time 0.021827 +2023-10-02 21:06:05,441 - Epoch: [77][ 760/ 1236] Overall Loss 0.258633 Objective Loss 0.258633 LR 0.001000 Time 0.021814 +2023-10-02 21:06:05,646 - Epoch: [77][ 770/ 1236] Overall Loss 0.258625 Objective Loss 0.258625 LR 0.001000 Time 0.021796 +2023-10-02 21:06:05,855 - Epoch: [77][ 780/ 1236] Overall Loss 0.258929 Objective Loss 0.258929 LR 0.001000 Time 0.021784 +2023-10-02 21:06:06,060 - Epoch: [77][ 790/ 1236] Overall Loss 0.258754 Objective Loss 0.258754 LR 0.001000 Time 0.021767 +2023-10-02 21:06:06,269 - Epoch: [77][ 800/ 1236] Overall Loss 0.258737 Objective Loss 0.258737 LR 0.001000 Time 0.021756 +2023-10-02 21:06:06,473 - Epoch: [77][ 810/ 1236] Overall Loss 0.258868 Objective Loss 0.258868 LR 0.001000 Time 0.021740 +2023-10-02 21:06:06,682 - Epoch: [77][ 820/ 1236] Overall Loss 0.258837 Objective Loss 0.258837 LR 0.001000 Time 0.021729 +2023-10-02 21:06:06,887 - Epoch: [77][ 830/ 1236] Overall Loss 0.259379 Objective Loss 0.259379 LR 0.001000 Time 0.021714 +2023-10-02 21:06:07,096 - Epoch: [77][ 840/ 1236] Overall Loss 0.259811 Objective Loss 0.259811 LR 0.001000 Time 0.021703 +2023-10-02 21:06:07,301 - Epoch: [77][ 850/ 1236] Overall Loss 0.259968 Objective Loss 0.259968 LR 0.001000 Time 0.021689 +2023-10-02 21:06:07,510 - Epoch: [77][ 860/ 1236] Overall Loss 0.260269 Objective Loss 0.260269 LR 0.001000 Time 0.021679 +2023-10-02 21:06:07,714 - Epoch: [77][ 870/ 1236] Overall Loss 0.260207 Objective Loss 0.260207 LR 0.001000 Time 0.021664 +2023-10-02 21:06:07,923 - Epoch: [77][ 880/ 1236] Overall Loss 0.260298 Objective Loss 0.260298 LR 0.001000 Time 0.021655 +2023-10-02 21:06:08,128 - Epoch: [77][ 890/ 1236] Overall Loss 0.260433 Objective Loss 0.260433 LR 0.001000 Time 0.021642 +2023-10-02 21:06:08,337 - Epoch: [77][ 900/ 1236] Overall Loss 0.261241 Objective Loss 0.261241 LR 0.001000 Time 0.021633 +2023-10-02 21:06:08,542 - Epoch: [77][ 910/ 1236] Overall Loss 0.261229 Objective Loss 0.261229 LR 0.001000 Time 0.021620 +2023-10-02 21:06:08,750 - Epoch: [77][ 920/ 1236] Overall Loss 0.261578 Objective Loss 0.261578 LR 0.001000 Time 0.021612 +2023-10-02 21:06:08,955 - Epoch: [77][ 930/ 1236] Overall Loss 0.262072 Objective Loss 0.262072 LR 0.001000 Time 0.021599 +2023-10-02 21:06:09,164 - Epoch: [77][ 940/ 1236] Overall Loss 0.262289 Objective Loss 0.262289 LR 0.001000 Time 0.021591 +2023-10-02 21:06:09,369 - Epoch: [77][ 950/ 1236] Overall Loss 0.262504 Objective Loss 0.262504 LR 0.001000 Time 0.021579 +2023-10-02 21:06:09,578 - Epoch: [77][ 960/ 1236] Overall Loss 0.262712 Objective Loss 0.262712 LR 0.001000 Time 0.021572 +2023-10-02 21:06:09,783 - Epoch: [77][ 970/ 1236] Overall Loss 0.263043 Objective Loss 0.263043 LR 0.001000 Time 0.021560 +2023-10-02 21:06:09,992 - Epoch: [77][ 980/ 1236] Overall Loss 0.262989 Objective Loss 0.262989 LR 0.001000 Time 0.021554 +2023-10-02 21:06:10,197 - Epoch: [77][ 990/ 1236] Overall Loss 0.263351 Objective Loss 0.263351 LR 0.001000 Time 0.021543 +2023-10-02 21:06:10,405 - Epoch: [77][ 1000/ 1236] Overall Loss 0.263567 Objective Loss 0.263567 LR 0.001000 Time 0.021535 +2023-10-02 21:06:10,611 - Epoch: [77][ 1010/ 1236] Overall Loss 0.263790 Objective Loss 0.263790 LR 0.001000 Time 0.021524 +2023-10-02 21:06:10,820 - Epoch: [77][ 1020/ 1236] Overall Loss 0.263850 Objective Loss 0.263850 LR 0.001000 Time 0.021518 +2023-10-02 21:06:11,026 - Epoch: [77][ 1030/ 1236] Overall Loss 0.264353 Objective Loss 0.264353 LR 0.001000 Time 0.021508 +2023-10-02 21:06:11,234 - Epoch: [77][ 1040/ 1236] Overall Loss 0.264681 Objective Loss 0.264681 LR 0.001000 Time 0.021502 +2023-10-02 21:06:11,439 - Epoch: [77][ 1050/ 1236] Overall Loss 0.264871 Objective Loss 0.264871 LR 0.001000 Time 0.021492 +2023-10-02 21:06:11,648 - Epoch: [77][ 1060/ 1236] Overall Loss 0.264916 Objective Loss 0.264916 LR 0.001000 Time 0.021486 +2023-10-02 21:06:11,853 - Epoch: [77][ 1070/ 1236] Overall Loss 0.265015 Objective Loss 0.265015 LR 0.001000 Time 0.021476 +2023-10-02 21:06:12,062 - Epoch: [77][ 1080/ 1236] Overall Loss 0.265423 Objective Loss 0.265423 LR 0.001000 Time 0.021471 +2023-10-02 21:06:12,268 - Epoch: [77][ 1090/ 1236] Overall Loss 0.265458 Objective Loss 0.265458 LR 0.001000 Time 0.021462 +2023-10-02 21:06:12,476 - Epoch: [77][ 1100/ 1236] Overall Loss 0.265743 Objective Loss 0.265743 LR 0.001000 Time 0.021456 +2023-10-02 21:06:12,681 - Epoch: [77][ 1110/ 1236] Overall Loss 0.265872 Objective Loss 0.265872 LR 0.001000 Time 0.021447 +2023-10-02 21:06:12,890 - Epoch: [77][ 1120/ 1236] Overall Loss 0.266126 Objective Loss 0.266126 LR 0.001000 Time 0.021442 +2023-10-02 21:06:13,095 - Epoch: [77][ 1130/ 1236] Overall Loss 0.266359 Objective Loss 0.266359 LR 0.001000 Time 0.021434 +2023-10-02 21:06:13,303 - Epoch: [77][ 1140/ 1236] Overall Loss 0.266603 Objective Loss 0.266603 LR 0.001000 Time 0.021427 +2023-10-02 21:06:13,509 - Epoch: [77][ 1150/ 1236] Overall Loss 0.266675 Objective Loss 0.266675 LR 0.001000 Time 0.021419 +2023-10-02 21:06:13,718 - Epoch: [77][ 1160/ 1236] Overall Loss 0.266523 Objective Loss 0.266523 LR 0.001000 Time 0.021414 +2023-10-02 21:06:13,923 - Epoch: [77][ 1170/ 1236] Overall Loss 0.266617 Objective Loss 0.266617 LR 0.001000 Time 0.021406 +2023-10-02 21:06:14,131 - Epoch: [77][ 1180/ 1236] Overall Loss 0.266877 Objective Loss 0.266877 LR 0.001000 Time 0.021401 +2023-10-02 21:06:14,337 - Epoch: [77][ 1190/ 1236] Overall Loss 0.267038 Objective Loss 0.267038 LR 0.001000 Time 0.021393 +2023-10-02 21:06:14,545 - Epoch: [77][ 1200/ 1236] Overall Loss 0.267087 Objective Loss 0.267087 LR 0.001000 Time 0.021387 +2023-10-02 21:06:14,751 - Epoch: [77][ 1210/ 1236] Overall Loss 0.267171 Objective Loss 0.267171 LR 0.001000 Time 0.021380 +2023-10-02 21:06:14,959 - Epoch: [77][ 1220/ 1236] Overall Loss 0.267178 Objective Loss 0.267178 LR 0.001000 Time 0.021375 +2023-10-02 21:06:15,217 - Epoch: [77][ 1230/ 1236] Overall Loss 0.267182 Objective Loss 0.267182 LR 0.001000 Time 0.021410 +2023-10-02 21:06:15,338 - Epoch: [77][ 1236/ 1236] Overall Loss 0.267403 Objective Loss 0.267403 Top1 82.688391 Top5 97.759674 LR 0.001000 Time 0.021404 +2023-10-02 21:06:15,483 - --- validate (epoch=77)----------- +2023-10-02 21:06:15,484 - 29943 samples (256 per mini-batch) +2023-10-02 21:06:15,938 - Epoch: [77][ 10/ 117] Loss 0.333331 Top1 83.437500 Top5 98.281250 +2023-10-02 21:06:16,090 - Epoch: [77][ 20/ 117] Loss 0.343565 Top1 83.261719 Top5 98.203125 +2023-10-02 21:06:16,239 - Epoch: [77][ 30/ 117] Loss 0.348508 Top1 82.968750 Top5 98.190104 +2023-10-02 21:06:16,391 - Epoch: [77][ 40/ 117] Loss 0.352515 Top1 82.929688 Top5 98.183594 +2023-10-02 21:06:16,542 - Epoch: [77][ 50/ 117] Loss 0.343168 Top1 83.257812 Top5 98.179688 +2023-10-02 21:06:16,693 - Epoch: [77][ 60/ 117] Loss 0.344483 Top1 83.222656 Top5 98.196615 +2023-10-02 21:06:16,844 - Epoch: [77][ 70/ 117] Loss 0.348242 Top1 83.297991 Top5 98.164062 +2023-10-02 21:06:16,996 - Epoch: [77][ 80/ 117] Loss 0.348108 Top1 83.212891 Top5 98.149414 +2023-10-02 21:06:17,148 - Epoch: [77][ 90/ 117] Loss 0.347163 Top1 83.333333 Top5 98.159722 +2023-10-02 21:06:17,299 - Epoch: [77][ 100/ 117] Loss 0.344822 Top1 83.355469 Top5 98.179688 +2023-10-02 21:06:17,458 - Epoch: [77][ 110/ 117] Loss 0.341609 Top1 83.327415 Top5 98.174716 +2023-10-02 21:06:17,546 - Epoch: [77][ 117/ 117] Loss 0.339104 Top1 83.324984 Top5 98.193234 +2023-10-02 21:06:17,682 - ==> Top1: 83.325 Top5: 98.193 Loss: 0.339 + +2023-10-02 21:06:17,682 - ==> Confusion: +[[ 937 0 3 0 9 5 0 0 7 61 0 0 0 2 6 0 6 0 1 0 13] + [ 0 1034 4 0 9 43 0 15 1 1 1 1 0 0 0 3 2 0 5 4 8] + [ 4 0 939 11 4 1 37 10 0 1 1 2 7 2 0 5 2 1 10 7 12] + [ 2 5 14 936 0 10 3 0 10 1 12 0 8 3 39 2 1 9 16 1 17] + [ 22 5 1 1 970 5 1 0 0 13 2 1 0 3 7 3 8 0 1 2 5] + [ 0 38 0 0 1 993 1 20 0 7 2 7 4 8 7 2 4 0 5 5 12] + [ 0 3 20 1 0 1 1129 4 0 0 4 3 1 0 0 4 1 1 2 9 8] + [ 4 26 18 0 4 43 5 1034 1 2 3 5 2 3 2 0 1 1 44 14 6] + [ 19 2 1 1 2 3 0 1 960 48 9 0 2 8 23 0 3 0 2 1 4] + [ 102 3 1 0 10 3 0 0 30 927 2 0 0 17 11 2 0 1 0 5 5] + [ 4 3 5 11 2 3 8 5 12 1 958 3 1 13 5 0 0 0 6 5 8] + [ 3 0 0 0 1 11 0 3 0 1 0 941 36 8 0 1 1 19 1 3 6] + [ 0 1 5 2 0 1 2 1 2 1 1 50 956 2 4 4 0 19 1 6 10] + [ 0 0 1 0 0 11 1 3 14 15 6 4 3 1035 6 2 0 0 0 1 17] + [ 12 0 5 11 1 1 0 0 13 0 3 0 2 6 1027 0 2 2 7 0 9] + [ 0 0 1 0 6 1 4 0 0 2 0 9 8 0 0 1044 19 26 2 7 5] + [ 0 19 1 1 9 9 1 0 1 0 0 6 0 3 5 7 1083 0 1 7 8] + [ 0 0 1 8 0 1 2 0 2 0 0 4 16 1 3 4 0 992 0 1 3] + [ 2 9 3 23 1 1 2 17 5 0 3 0 0 1 17 0 0 0 972 0 12] + [ 1 4 4 1 0 5 5 11 0 1 1 18 2 0 0 2 6 1 2 1083 5] + [ 159 201 123 84 118 219 48 101 110 110 187 107 310 264 167 51 114 67 168 197 5000]] + +2023-10-02 21:06:17,684 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:06:17,684 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:06:17,690 - + +2023-10-02 21:06:17,690 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:06:18,698 - Epoch: [78][ 10/ 1236] Overall Loss 0.271905 Objective Loss 0.271905 LR 0.001000 Time 0.100767 +2023-10-02 21:06:18,904 - Epoch: [78][ 20/ 1236] Overall Loss 0.257620 Objective Loss 0.257620 LR 0.001000 Time 0.060657 +2023-10-02 21:06:19,109 - Epoch: [78][ 30/ 1236] Overall Loss 0.260546 Objective Loss 0.260546 LR 0.001000 Time 0.047217 +2023-10-02 21:06:19,315 - Epoch: [78][ 40/ 1236] Overall Loss 0.262236 Objective Loss 0.262236 LR 0.001000 Time 0.040568 +2023-10-02 21:06:19,520 - Epoch: [78][ 50/ 1236] Overall Loss 0.257958 Objective Loss 0.257958 LR 0.001000 Time 0.036527 +2023-10-02 21:06:19,728 - Epoch: [78][ 60/ 1236] Overall Loss 0.260885 Objective Loss 0.260885 LR 0.001000 Time 0.033896 +2023-10-02 21:06:19,932 - Epoch: [78][ 70/ 1236] Overall Loss 0.266173 Objective Loss 0.266173 LR 0.001000 Time 0.031964 +2023-10-02 21:06:20,140 - Epoch: [78][ 80/ 1236] Overall Loss 0.269286 Objective Loss 0.269286 LR 0.001000 Time 0.030560 +2023-10-02 21:06:20,344 - Epoch: [78][ 90/ 1236] Overall Loss 0.273237 Objective Loss 0.273237 LR 0.001000 Time 0.029427 +2023-10-02 21:06:20,551 - Epoch: [78][ 100/ 1236] Overall Loss 0.270702 Objective Loss 0.270702 LR 0.001000 Time 0.028559 +2023-10-02 21:06:20,755 - Epoch: [78][ 110/ 1236] Overall Loss 0.267840 Objective Loss 0.267840 LR 0.001000 Time 0.027815 +2023-10-02 21:06:20,963 - Epoch: [78][ 120/ 1236] Overall Loss 0.266626 Objective Loss 0.266626 LR 0.001000 Time 0.027228 +2023-10-02 21:06:21,167 - Epoch: [78][ 130/ 1236] Overall Loss 0.266949 Objective Loss 0.266949 LR 0.001000 Time 0.026700 +2023-10-02 21:06:21,375 - Epoch: [78][ 140/ 1236] Overall Loss 0.268521 Objective Loss 0.268521 LR 0.001000 Time 0.026275 +2023-10-02 21:06:21,579 - Epoch: [78][ 150/ 1236] Overall Loss 0.268165 Objective Loss 0.268165 LR 0.001000 Time 0.025881 +2023-10-02 21:06:21,785 - Epoch: [78][ 160/ 1236] Overall Loss 0.269434 Objective Loss 0.269434 LR 0.001000 Time 0.025549 +2023-10-02 21:06:21,990 - Epoch: [78][ 170/ 1236] Overall Loss 0.269821 Objective Loss 0.269821 LR 0.001000 Time 0.025241 +2023-10-02 21:06:22,196 - Epoch: [78][ 180/ 1236] Overall Loss 0.270044 Objective Loss 0.270044 LR 0.001000 Time 0.024985 +2023-10-02 21:06:22,401 - Epoch: [78][ 190/ 1236] Overall Loss 0.271013 Objective Loss 0.271013 LR 0.001000 Time 0.024742 +2023-10-02 21:06:22,608 - Epoch: [78][ 200/ 1236] Overall Loss 0.270119 Objective Loss 0.270119 LR 0.001000 Time 0.024537 +2023-10-02 21:06:22,813 - Epoch: [78][ 210/ 1236] Overall Loss 0.268895 Objective Loss 0.268895 LR 0.001000 Time 0.024339 +2023-10-02 21:06:23,021 - Epoch: [78][ 220/ 1236] Overall Loss 0.268847 Objective Loss 0.268847 LR 0.001000 Time 0.024176 +2023-10-02 21:06:23,225 - Epoch: [78][ 230/ 1236] Overall Loss 0.268811 Objective Loss 0.268811 LR 0.001000 Time 0.024009 +2023-10-02 21:06:23,431 - Epoch: [78][ 240/ 1236] Overall Loss 0.268477 Objective Loss 0.268477 LR 0.001000 Time 0.023867 +2023-10-02 21:06:23,636 - Epoch: [78][ 250/ 1236] Overall Loss 0.268463 Objective Loss 0.268463 LR 0.001000 Time 0.023726 +2023-10-02 21:06:23,842 - Epoch: [78][ 260/ 1236] Overall Loss 0.267688 Objective Loss 0.267688 LR 0.001000 Time 0.023606 +2023-10-02 21:06:24,048 - Epoch: [78][ 270/ 1236] Overall Loss 0.266843 Objective Loss 0.266843 LR 0.001000 Time 0.023486 +2023-10-02 21:06:24,255 - Epoch: [78][ 280/ 1236] Overall Loss 0.266348 Objective Loss 0.266348 LR 0.001000 Time 0.023387 +2023-10-02 21:06:24,459 - Epoch: [78][ 290/ 1236] Overall Loss 0.266277 Objective Loss 0.266277 LR 0.001000 Time 0.023283 +2023-10-02 21:06:24,667 - Epoch: [78][ 300/ 1236] Overall Loss 0.266922 Objective Loss 0.266922 LR 0.001000 Time 0.023198 +2023-10-02 21:06:24,869 - Epoch: [78][ 310/ 1236] Overall Loss 0.267407 Objective Loss 0.267407 LR 0.001000 Time 0.023102 +2023-10-02 21:06:25,077 - Epoch: [78][ 320/ 1236] Overall Loss 0.266946 Objective Loss 0.266946 LR 0.001000 Time 0.023029 +2023-10-02 21:06:25,281 - Epoch: [78][ 330/ 1236] Overall Loss 0.266581 Objective Loss 0.266581 LR 0.001000 Time 0.022949 +2023-10-02 21:06:25,489 - Epoch: [78][ 340/ 1236] Overall Loss 0.266483 Objective Loss 0.266483 LR 0.001000 Time 0.022884 +2023-10-02 21:06:25,693 - Epoch: [78][ 350/ 1236] Overall Loss 0.267040 Objective Loss 0.267040 LR 0.001000 Time 0.022812 +2023-10-02 21:06:25,901 - Epoch: [78][ 360/ 1236] Overall Loss 0.267352 Objective Loss 0.267352 LR 0.001000 Time 0.022756 +2023-10-02 21:06:26,105 - Epoch: [78][ 370/ 1236] Overall Loss 0.267009 Objective Loss 0.267009 LR 0.001000 Time 0.022691 +2023-10-02 21:06:26,313 - Epoch: [78][ 380/ 1236] Overall Loss 0.266770 Objective Loss 0.266770 LR 0.001000 Time 0.022640 +2023-10-02 21:06:26,517 - Epoch: [78][ 390/ 1236] Overall Loss 0.266671 Objective Loss 0.266671 LR 0.001000 Time 0.022583 +2023-10-02 21:06:26,725 - Epoch: [78][ 400/ 1236] Overall Loss 0.266790 Objective Loss 0.266790 LR 0.001000 Time 0.022537 +2023-10-02 21:06:26,929 - Epoch: [78][ 410/ 1236] Overall Loss 0.266307 Objective Loss 0.266307 LR 0.001000 Time 0.022485 +2023-10-02 21:06:27,137 - Epoch: [78][ 420/ 1236] Overall Loss 0.266125 Objective Loss 0.266125 LR 0.001000 Time 0.022444 +2023-10-02 21:06:27,341 - Epoch: [78][ 430/ 1236] Overall Loss 0.266192 Objective Loss 0.266192 LR 0.001000 Time 0.022395 +2023-10-02 21:06:27,549 - Epoch: [78][ 440/ 1236] Overall Loss 0.266572 Objective Loss 0.266572 LR 0.001000 Time 0.022358 +2023-10-02 21:06:27,753 - Epoch: [78][ 450/ 1236] Overall Loss 0.266188 Objective Loss 0.266188 LR 0.001000 Time 0.022314 +2023-10-02 21:06:27,961 - Epoch: [78][ 460/ 1236] Overall Loss 0.266230 Objective Loss 0.266230 LR 0.001000 Time 0.022280 +2023-10-02 21:06:28,165 - Epoch: [78][ 470/ 1236] Overall Loss 0.265671 Objective Loss 0.265671 LR 0.001000 Time 0.022240 +2023-10-02 21:06:28,373 - Epoch: [78][ 480/ 1236] Overall Loss 0.265657 Objective Loss 0.265657 LR 0.001000 Time 0.022209 +2023-10-02 21:06:28,577 - Epoch: [78][ 490/ 1236] Overall Loss 0.265850 Objective Loss 0.265850 LR 0.001000 Time 0.022172 +2023-10-02 21:06:28,785 - Epoch: [78][ 500/ 1236] Overall Loss 0.266044 Objective Loss 0.266044 LR 0.001000 Time 0.022144 +2023-10-02 21:06:28,989 - Epoch: [78][ 510/ 1236] Overall Loss 0.265954 Objective Loss 0.265954 LR 0.001000 Time 0.022110 +2023-10-02 21:06:29,197 - Epoch: [78][ 520/ 1236] Overall Loss 0.265793 Objective Loss 0.265793 LR 0.001000 Time 0.022084 +2023-10-02 21:06:29,401 - Epoch: [78][ 530/ 1236] Overall Loss 0.265926 Objective Loss 0.265926 LR 0.001000 Time 0.022051 +2023-10-02 21:06:29,609 - Epoch: [78][ 540/ 1236] Overall Loss 0.265947 Objective Loss 0.265947 LR 0.001000 Time 0.022027 +2023-10-02 21:06:29,813 - Epoch: [78][ 550/ 1236] Overall Loss 0.265712 Objective Loss 0.265712 LR 0.001000 Time 0.021998 +2023-10-02 21:06:30,021 - Epoch: [78][ 560/ 1236] Overall Loss 0.265745 Objective Loss 0.265745 LR 0.001000 Time 0.021976 +2023-10-02 21:06:30,226 - Epoch: [78][ 570/ 1236] Overall Loss 0.265173 Objective Loss 0.265173 LR 0.001000 Time 0.021949 +2023-10-02 21:06:30,432 - Epoch: [78][ 580/ 1236] Overall Loss 0.265068 Objective Loss 0.265068 LR 0.001000 Time 0.021926 +2023-10-02 21:06:30,641 - Epoch: [78][ 590/ 1236] Overall Loss 0.264869 Objective Loss 0.264869 LR 0.001000 Time 0.021908 +2023-10-02 21:06:30,848 - Epoch: [78][ 600/ 1236] Overall Loss 0.264643 Objective Loss 0.264643 LR 0.001000 Time 0.021887 +2023-10-02 21:06:31,057 - Epoch: [78][ 610/ 1236] Overall Loss 0.264658 Objective Loss 0.264658 LR 0.001000 Time 0.021871 +2023-10-02 21:06:31,264 - Epoch: [78][ 620/ 1236] Overall Loss 0.264595 Objective Loss 0.264595 LR 0.001000 Time 0.021851 +2023-10-02 21:06:31,474 - Epoch: [78][ 630/ 1236] Overall Loss 0.264703 Objective Loss 0.264703 LR 0.001000 Time 0.021836 +2023-10-02 21:06:31,681 - Epoch: [78][ 640/ 1236] Overall Loss 0.264286 Objective Loss 0.264286 LR 0.001000 Time 0.021818 +2023-10-02 21:06:31,890 - Epoch: [78][ 650/ 1236] Overall Loss 0.264034 Objective Loss 0.264034 LR 0.001000 Time 0.021804 +2023-10-02 21:06:32,097 - Epoch: [78][ 660/ 1236] Overall Loss 0.264274 Objective Loss 0.264274 LR 0.001000 Time 0.021786 +2023-10-02 21:06:32,306 - Epoch: [78][ 670/ 1236] Overall Loss 0.264279 Objective Loss 0.264279 LR 0.001000 Time 0.021773 +2023-10-02 21:06:32,513 - Epoch: [78][ 680/ 1236] Overall Loss 0.264286 Objective Loss 0.264286 LR 0.001000 Time 0.021756 +2023-10-02 21:06:32,722 - Epoch: [78][ 690/ 1236] Overall Loss 0.264469 Objective Loss 0.264469 LR 0.001000 Time 0.021743 +2023-10-02 21:06:32,929 - Epoch: [78][ 700/ 1236] Overall Loss 0.264279 Objective Loss 0.264279 LR 0.001000 Time 0.021728 +2023-10-02 21:06:33,138 - Epoch: [78][ 710/ 1236] Overall Loss 0.263830 Objective Loss 0.263830 LR 0.001000 Time 0.021714 +2023-10-02 21:06:33,345 - Epoch: [78][ 720/ 1236] Overall Loss 0.263896 Objective Loss 0.263896 LR 0.001000 Time 0.021700 +2023-10-02 21:06:33,554 - Epoch: [78][ 730/ 1236] Overall Loss 0.263712 Objective Loss 0.263712 LR 0.001000 Time 0.021689 +2023-10-02 21:06:33,761 - Epoch: [78][ 740/ 1236] Overall Loss 0.264126 Objective Loss 0.264126 LR 0.001000 Time 0.021675 +2023-10-02 21:06:33,970 - Epoch: [78][ 750/ 1236] Overall Loss 0.264185 Objective Loss 0.264185 LR 0.001000 Time 0.021664 +2023-10-02 21:06:34,177 - Epoch: [78][ 760/ 1236] Overall Loss 0.264299 Objective Loss 0.264299 LR 0.001000 Time 0.021651 +2023-10-02 21:06:34,386 - Epoch: [78][ 770/ 1236] Overall Loss 0.263926 Objective Loss 0.263926 LR 0.001000 Time 0.021641 +2023-10-02 21:06:34,593 - Epoch: [78][ 780/ 1236] Overall Loss 0.264132 Objective Loss 0.264132 LR 0.001000 Time 0.021629 +2023-10-02 21:06:34,803 - Epoch: [78][ 790/ 1236] Overall Loss 0.264225 Objective Loss 0.264225 LR 0.001000 Time 0.021619 +2023-10-02 21:06:35,010 - Epoch: [78][ 800/ 1236] Overall Loss 0.263985 Objective Loss 0.263985 LR 0.001000 Time 0.021609 +2023-10-02 21:06:35,220 - Epoch: [78][ 810/ 1236] Overall Loss 0.264216 Objective Loss 0.264216 LR 0.001000 Time 0.021600 +2023-10-02 21:06:35,428 - Epoch: [78][ 820/ 1236] Overall Loss 0.264428 Objective Loss 0.264428 LR 0.001000 Time 0.021590 +2023-10-02 21:06:35,638 - Epoch: [78][ 830/ 1236] Overall Loss 0.264468 Objective Loss 0.264468 LR 0.001000 Time 0.021582 +2023-10-02 21:06:35,845 - Epoch: [78][ 840/ 1236] Overall Loss 0.264480 Objective Loss 0.264480 LR 0.001000 Time 0.021572 +2023-10-02 21:06:36,055 - Epoch: [78][ 850/ 1236] Overall Loss 0.264231 Objective Loss 0.264231 LR 0.001000 Time 0.021564 +2023-10-02 21:06:36,262 - Epoch: [78][ 860/ 1236] Overall Loss 0.264099 Objective Loss 0.264099 LR 0.001000 Time 0.021554 +2023-10-02 21:06:36,472 - Epoch: [78][ 870/ 1236] Overall Loss 0.264043 Objective Loss 0.264043 LR 0.001000 Time 0.021547 +2023-10-02 21:06:36,679 - Epoch: [78][ 880/ 1236] Overall Loss 0.264152 Objective Loss 0.264152 LR 0.001000 Time 0.021538 +2023-10-02 21:06:36,889 - Epoch: [78][ 890/ 1236] Overall Loss 0.264069 Objective Loss 0.264069 LR 0.001000 Time 0.021531 +2023-10-02 21:06:37,097 - Epoch: [78][ 900/ 1236] Overall Loss 0.264367 Objective Loss 0.264367 LR 0.001000 Time 0.021522 +2023-10-02 21:06:37,307 - Epoch: [78][ 910/ 1236] Overall Loss 0.264159 Objective Loss 0.264159 LR 0.001000 Time 0.021516 +2023-10-02 21:06:37,514 - Epoch: [78][ 920/ 1236] Overall Loss 0.264708 Objective Loss 0.264708 LR 0.001000 Time 0.021507 +2023-10-02 21:06:37,724 - Epoch: [78][ 930/ 1236] Overall Loss 0.264861 Objective Loss 0.264861 LR 0.001000 Time 0.021501 +2023-10-02 21:06:37,931 - Epoch: [78][ 940/ 1236] Overall Loss 0.265034 Objective Loss 0.265034 LR 0.001000 Time 0.021493 +2023-10-02 21:06:38,141 - Epoch: [78][ 950/ 1236] Overall Loss 0.265069 Objective Loss 0.265069 LR 0.001000 Time 0.021487 +2023-10-02 21:06:38,349 - Epoch: [78][ 960/ 1236] Overall Loss 0.264915 Objective Loss 0.264915 LR 0.001000 Time 0.021479 +2023-10-02 21:06:38,558 - Epoch: [78][ 970/ 1236] Overall Loss 0.264572 Objective Loss 0.264572 LR 0.001000 Time 0.021474 +2023-10-02 21:06:38,766 - Epoch: [78][ 980/ 1236] Overall Loss 0.264683 Objective Loss 0.264683 LR 0.001000 Time 0.021466 +2023-10-02 21:06:38,976 - Epoch: [78][ 990/ 1236] Overall Loss 0.264735 Objective Loss 0.264735 LR 0.001000 Time 0.021461 +2023-10-02 21:06:39,183 - Epoch: [78][ 1000/ 1236] Overall Loss 0.264769 Objective Loss 0.264769 LR 0.001000 Time 0.021453 +2023-10-02 21:06:39,393 - Epoch: [78][ 1010/ 1236] Overall Loss 0.264574 Objective Loss 0.264574 LR 0.001000 Time 0.021448 +2023-10-02 21:06:39,601 - Epoch: [78][ 1020/ 1236] Overall Loss 0.264462 Objective Loss 0.264462 LR 0.001000 Time 0.021441 +2023-10-02 21:06:39,810 - Epoch: [78][ 1030/ 1236] Overall Loss 0.264237 Objective Loss 0.264237 LR 0.001000 Time 0.021436 +2023-10-02 21:06:40,018 - Epoch: [78][ 1040/ 1236] Overall Loss 0.264209 Objective Loss 0.264209 LR 0.001000 Time 0.021430 +2023-10-02 21:06:40,228 - Epoch: [78][ 1050/ 1236] Overall Loss 0.264214 Objective Loss 0.264214 LR 0.001000 Time 0.021425 +2023-10-02 21:06:40,435 - Epoch: [78][ 1060/ 1236] Overall Loss 0.264296 Objective Loss 0.264296 LR 0.001000 Time 0.021418 +2023-10-02 21:06:40,645 - Epoch: [78][ 1070/ 1236] Overall Loss 0.264472 Objective Loss 0.264472 LR 0.001000 Time 0.021414 +2023-10-02 21:06:40,852 - Epoch: [78][ 1080/ 1236] Overall Loss 0.264950 Objective Loss 0.264950 LR 0.001000 Time 0.021407 +2023-10-02 21:06:41,062 - Epoch: [78][ 1090/ 1236] Overall Loss 0.265023 Objective Loss 0.265023 LR 0.001000 Time 0.021403 +2023-10-02 21:06:41,270 - Epoch: [78][ 1100/ 1236] Overall Loss 0.265305 Objective Loss 0.265305 LR 0.001000 Time 0.021397 +2023-10-02 21:06:41,480 - Epoch: [78][ 1110/ 1236] Overall Loss 0.265359 Objective Loss 0.265359 LR 0.001000 Time 0.021393 +2023-10-02 21:06:41,688 - Epoch: [78][ 1120/ 1236] Overall Loss 0.265476 Objective Loss 0.265476 LR 0.001000 Time 0.021387 +2023-10-02 21:06:41,897 - Epoch: [78][ 1130/ 1236] Overall Loss 0.265550 Objective Loss 0.265550 LR 0.001000 Time 0.021383 +2023-10-02 21:06:42,105 - Epoch: [78][ 1140/ 1236] Overall Loss 0.265943 Objective Loss 0.265943 LR 0.001000 Time 0.021378 +2023-10-02 21:06:42,315 - Epoch: [78][ 1150/ 1236] Overall Loss 0.265846 Objective Loss 0.265846 LR 0.001000 Time 0.021374 +2023-10-02 21:06:42,523 - Epoch: [78][ 1160/ 1236] Overall Loss 0.265908 Objective Loss 0.265908 LR 0.001000 Time 0.021368 +2023-10-02 21:06:42,732 - Epoch: [78][ 1170/ 1236] Overall Loss 0.266171 Objective Loss 0.266171 LR 0.001000 Time 0.021365 +2023-10-02 21:06:42,940 - Epoch: [78][ 1180/ 1236] Overall Loss 0.266316 Objective Loss 0.266316 LR 0.001000 Time 0.021360 +2023-10-02 21:06:43,150 - Epoch: [78][ 1190/ 1236] Overall Loss 0.266470 Objective Loss 0.266470 LR 0.001000 Time 0.021356 +2023-10-02 21:06:43,357 - Epoch: [78][ 1200/ 1236] Overall Loss 0.266489 Objective Loss 0.266489 LR 0.001000 Time 0.021351 +2023-10-02 21:06:43,567 - Epoch: [78][ 1210/ 1236] Overall Loss 0.266360 Objective Loss 0.266360 LR 0.001000 Time 0.021348 +2023-10-02 21:06:43,775 - Epoch: [78][ 1220/ 1236] Overall Loss 0.266469 Objective Loss 0.266469 LR 0.001000 Time 0.021343 +2023-10-02 21:06:44,036 - Epoch: [78][ 1230/ 1236] Overall Loss 0.266453 Objective Loss 0.266453 LR 0.001000 Time 0.021381 +2023-10-02 21:06:44,158 - Epoch: [78][ 1236/ 1236] Overall Loss 0.266505 Objective Loss 0.266505 Top1 84.521385 Top5 97.963340 LR 0.001000 Time 0.021375 +2023-10-02 21:06:44,298 - --- validate (epoch=78)----------- +2023-10-02 21:06:44,299 - 29943 samples (256 per mini-batch) +2023-10-02 21:06:44,787 - Epoch: [78][ 10/ 117] Loss 0.313065 Top1 84.101562 Top5 98.359375 +2023-10-02 21:06:44,950 - Epoch: [78][ 20/ 117] Loss 0.326306 Top1 83.457031 Top5 98.183594 +2023-10-02 21:06:45,108 - Epoch: [78][ 30/ 117] Loss 0.339924 Top1 83.320312 Top5 98.203125 +2023-10-02 21:06:45,269 - Epoch: [78][ 40/ 117] Loss 0.343950 Top1 83.759766 Top5 98.291016 +2023-10-02 21:06:45,426 - Epoch: [78][ 50/ 117] Loss 0.348062 Top1 83.742188 Top5 98.242188 +2023-10-02 21:06:45,586 - Epoch: [78][ 60/ 117] Loss 0.341937 Top1 83.971354 Top5 98.281250 +2023-10-02 21:06:45,744 - Epoch: [78][ 70/ 117] Loss 0.343830 Top1 84.012277 Top5 98.331473 +2023-10-02 21:06:45,904 - Epoch: [78][ 80/ 117] Loss 0.345352 Top1 83.916016 Top5 98.315430 +2023-10-02 21:06:46,061 - Epoch: [78][ 90/ 117] Loss 0.343822 Top1 83.967014 Top5 98.315972 +2023-10-02 21:06:46,222 - Epoch: [78][ 100/ 117] Loss 0.345544 Top1 83.953125 Top5 98.281250 +2023-10-02 21:06:46,388 - Epoch: [78][ 110/ 117] Loss 0.347188 Top1 83.888494 Top5 98.288352 +2023-10-02 21:06:46,477 - Epoch: [78][ 117/ 117] Loss 0.345313 Top1 83.902749 Top5 98.300104 +2023-10-02 21:06:46,601 - ==> Top1: 83.903 Top5: 98.300 Loss: 0.345 + +2023-10-02 21:06:46,601 - ==> Confusion: +[[ 920 1 7 0 6 5 0 0 11 69 2 1 0 1 7 0 1 0 1 0 18] + [ 0 1057 3 0 7 14 1 17 3 1 2 0 0 1 2 4 2 0 8 2 7] + [ 3 0 978 5 5 2 23 7 1 0 0 0 3 3 1 1 1 1 9 2 11] + [ 0 3 24 940 0 4 1 4 7 1 10 0 2 5 41 2 1 3 21 2 18] + [ 27 5 2 0 968 5 0 1 0 11 0 1 0 5 12 2 9 0 0 2 0] + [ 3 54 2 2 5 957 2 25 2 10 2 4 1 8 6 1 5 1 8 2 16] + [ 0 8 28 1 0 1 1118 7 0 0 3 3 0 0 1 6 0 1 2 3 9] + [ 4 10 17 1 4 25 10 1065 0 2 3 5 2 4 3 0 0 1 50 7 5] + [ 13 2 0 0 2 1 0 1 977 48 8 1 1 10 15 1 2 4 2 0 1] + [ 83 1 1 0 7 4 0 0 30 955 1 0 0 12 10 1 0 1 0 1 12] + [ 2 0 12 6 2 3 1 4 24 2 945 3 0 19 7 1 0 1 9 1 11] + [ 1 1 2 0 2 17 0 6 0 0 2 939 16 12 0 2 4 15 0 7 9] + [ 1 3 3 6 0 4 4 0 3 1 2 57 899 10 9 8 6 24 6 3 19] + [ 1 1 1 0 5 7 0 1 11 31 1 0 0 1044 6 0 0 1 0 0 9] + [ 13 1 3 11 7 0 0 0 21 4 2 1 1 4 1012 0 3 0 12 0 6] + [ 0 0 4 0 4 2 3 0 0 0 0 6 7 1 0 1048 24 14 3 7 11] + [ 1 15 0 0 6 8 1 2 1 1 0 1 0 1 3 5 1103 0 0 5 8] + [ 0 1 0 5 0 1 5 0 1 2 0 4 6 2 7 9 1 983 0 2 9] + [ 1 5 5 13 0 1 1 17 6 1 3 0 2 0 8 0 1 0 993 0 11] + [ 0 6 5 3 1 9 11 12 0 0 1 8 4 2 0 3 13 1 3 1057 13] + [ 153 202 145 67 111 137 50 104 107 138 117 110 261 253 182 39 156 54 200 154 5165]] + +2023-10-02 21:06:46,602 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:06:46,603 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:06:46,608 - + +2023-10-02 21:06:46,609 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:06:47,620 - Epoch: [79][ 10/ 1236] Overall Loss 0.288999 Objective Loss 0.288999 LR 0.001000 Time 0.101044 +2023-10-02 21:06:47,828 - Epoch: [79][ 20/ 1236] Overall Loss 0.285833 Objective Loss 0.285833 LR 0.001000 Time 0.060938 +2023-10-02 21:06:48,035 - Epoch: [79][ 30/ 1236] Overall Loss 0.278953 Objective Loss 0.278953 LR 0.001000 Time 0.047516 +2023-10-02 21:06:48,244 - Epoch: [79][ 40/ 1236] Overall Loss 0.277965 Objective Loss 0.277965 LR 0.001000 Time 0.040858 +2023-10-02 21:06:48,450 - Epoch: [79][ 50/ 1236] Overall Loss 0.272341 Objective Loss 0.272341 LR 0.001000 Time 0.036799 +2023-10-02 21:06:48,659 - Epoch: [79][ 60/ 1236] Overall Loss 0.269204 Objective Loss 0.269204 LR 0.001000 Time 0.034147 +2023-10-02 21:06:48,865 - Epoch: [79][ 70/ 1236] Overall Loss 0.273268 Objective Loss 0.273268 LR 0.001000 Time 0.032205 +2023-10-02 21:06:49,074 - Epoch: [79][ 80/ 1236] Overall Loss 0.270493 Objective Loss 0.270493 LR 0.001000 Time 0.030790 +2023-10-02 21:06:49,280 - Epoch: [79][ 90/ 1236] Overall Loss 0.268478 Objective Loss 0.268478 LR 0.001000 Time 0.029653 +2023-10-02 21:06:49,488 - Epoch: [79][ 100/ 1236] Overall Loss 0.263678 Objective Loss 0.263678 LR 0.001000 Time 0.028762 +2023-10-02 21:06:49,694 - Epoch: [79][ 110/ 1236] Overall Loss 0.261437 Objective Loss 0.261437 LR 0.001000 Time 0.028010 +2023-10-02 21:06:49,903 - Epoch: [79][ 120/ 1236] Overall Loss 0.259312 Objective Loss 0.259312 LR 0.001000 Time 0.027415 +2023-10-02 21:06:50,109 - Epoch: [79][ 130/ 1236] Overall Loss 0.259154 Objective Loss 0.259154 LR 0.001000 Time 0.026884 +2023-10-02 21:06:50,318 - Epoch: [79][ 140/ 1236] Overall Loss 0.257110 Objective Loss 0.257110 LR 0.001000 Time 0.026458 +2023-10-02 21:06:50,523 - Epoch: [79][ 150/ 1236] Overall Loss 0.257644 Objective Loss 0.257644 LR 0.001000 Time 0.026061 +2023-10-02 21:06:50,732 - Epoch: [79][ 160/ 1236] Overall Loss 0.257885 Objective Loss 0.257885 LR 0.001000 Time 0.025737 +2023-10-02 21:06:50,938 - Epoch: [79][ 170/ 1236] Overall Loss 0.258252 Objective Loss 0.258252 LR 0.001000 Time 0.025432 +2023-10-02 21:06:51,147 - Epoch: [79][ 180/ 1236] Overall Loss 0.258651 Objective Loss 0.258651 LR 0.001000 Time 0.025178 +2023-10-02 21:06:51,352 - Epoch: [79][ 190/ 1236] Overall Loss 0.256694 Objective Loss 0.256694 LR 0.001000 Time 0.024932 +2023-10-02 21:06:51,561 - Epoch: [79][ 200/ 1236] Overall Loss 0.259424 Objective Loss 0.259424 LR 0.001000 Time 0.024729 +2023-10-02 21:06:51,767 - Epoch: [79][ 210/ 1236] Overall Loss 0.260519 Objective Loss 0.260519 LR 0.001000 Time 0.024528 +2023-10-02 21:06:51,976 - Epoch: [79][ 220/ 1236] Overall Loss 0.260451 Objective Loss 0.260451 LR 0.001000 Time 0.024363 +2023-10-02 21:06:52,182 - Epoch: [79][ 230/ 1236] Overall Loss 0.259940 Objective Loss 0.259940 LR 0.001000 Time 0.024196 +2023-10-02 21:06:52,389 - Epoch: [79][ 240/ 1236] Overall Loss 0.258971 Objective Loss 0.258971 LR 0.001000 Time 0.024053 +2023-10-02 21:06:52,596 - Epoch: [79][ 250/ 1236] Overall Loss 0.259288 Objective Loss 0.259288 LR 0.001000 Time 0.023910 +2023-10-02 21:06:52,805 - Epoch: [79][ 260/ 1236] Overall Loss 0.259421 Objective Loss 0.259421 LR 0.001000 Time 0.023795 +2023-10-02 21:06:53,011 - Epoch: [79][ 270/ 1236] Overall Loss 0.259128 Objective Loss 0.259128 LR 0.001000 Time 0.023674 +2023-10-02 21:06:53,220 - Epoch: [79][ 280/ 1236] Overall Loss 0.259429 Objective Loss 0.259429 LR 0.001000 Time 0.023575 +2023-10-02 21:06:53,425 - Epoch: [79][ 290/ 1236] Overall Loss 0.259035 Objective Loss 0.259035 LR 0.001000 Time 0.023469 +2023-10-02 21:06:53,633 - Epoch: [79][ 300/ 1236] Overall Loss 0.258650 Objective Loss 0.258650 LR 0.001000 Time 0.023378 +2023-10-02 21:06:53,840 - Epoch: [79][ 310/ 1236] Overall Loss 0.259575 Objective Loss 0.259575 LR 0.001000 Time 0.023286 +2023-10-02 21:06:54,049 - Epoch: [79][ 320/ 1236] Overall Loss 0.259609 Objective Loss 0.259609 LR 0.001000 Time 0.023212 +2023-10-02 21:06:54,255 - Epoch: [79][ 330/ 1236] Overall Loss 0.259143 Objective Loss 0.259143 LR 0.001000 Time 0.023132 +2023-10-02 21:06:54,466 - Epoch: [79][ 340/ 1236] Overall Loss 0.258601 Objective Loss 0.258601 LR 0.001000 Time 0.023070 +2023-10-02 21:06:54,672 - Epoch: [79][ 350/ 1236] Overall Loss 0.258954 Objective Loss 0.258954 LR 0.001000 Time 0.023000 +2023-10-02 21:06:54,882 - Epoch: [79][ 360/ 1236] Overall Loss 0.258901 Objective Loss 0.258901 LR 0.001000 Time 0.022944 +2023-10-02 21:06:55,088 - Epoch: [79][ 370/ 1236] Overall Loss 0.258077 Objective Loss 0.258077 LR 0.001000 Time 0.022880 +2023-10-02 21:06:55,299 - Epoch: [79][ 380/ 1236] Overall Loss 0.258227 Objective Loss 0.258227 LR 0.001000 Time 0.022831 +2023-10-02 21:06:55,505 - Epoch: [79][ 390/ 1236] Overall Loss 0.258350 Objective Loss 0.258350 LR 0.001000 Time 0.022774 +2023-10-02 21:06:55,715 - Epoch: [79][ 400/ 1236] Overall Loss 0.258418 Objective Loss 0.258418 LR 0.001000 Time 0.022730 +2023-10-02 21:06:55,922 - Epoch: [79][ 410/ 1236] Overall Loss 0.258168 Objective Loss 0.258168 LR 0.001000 Time 0.022677 +2023-10-02 21:06:56,132 - Epoch: [79][ 420/ 1236] Overall Loss 0.257865 Objective Loss 0.257865 LR 0.001000 Time 0.022638 +2023-10-02 21:06:56,338 - Epoch: [79][ 430/ 1236] Overall Loss 0.258066 Objective Loss 0.258066 LR 0.001000 Time 0.022590 +2023-10-02 21:06:56,549 - Epoch: [79][ 440/ 1236] Overall Loss 0.257761 Objective Loss 0.257761 LR 0.001000 Time 0.022554 +2023-10-02 21:06:56,755 - Epoch: [79][ 450/ 1236] Overall Loss 0.257559 Objective Loss 0.257559 LR 0.001000 Time 0.022511 +2023-10-02 21:06:56,965 - Epoch: [79][ 460/ 1236] Overall Loss 0.257138 Objective Loss 0.257138 LR 0.001000 Time 0.022478 +2023-10-02 21:06:57,172 - Epoch: [79][ 470/ 1236] Overall Loss 0.257688 Objective Loss 0.257688 LR 0.001000 Time 0.022438 +2023-10-02 21:06:57,382 - Epoch: [79][ 480/ 1236] Overall Loss 0.257526 Objective Loss 0.257526 LR 0.001000 Time 0.022408 +2023-10-02 21:06:57,588 - Epoch: [79][ 490/ 1236] Overall Loss 0.257577 Objective Loss 0.257577 LR 0.001000 Time 0.022370 +2023-10-02 21:06:57,798 - Epoch: [79][ 500/ 1236] Overall Loss 0.257722 Objective Loss 0.257722 LR 0.001000 Time 0.022343 +2023-10-02 21:06:58,005 - Epoch: [79][ 510/ 1236] Overall Loss 0.258242 Objective Loss 0.258242 LR 0.001000 Time 0.022309 +2023-10-02 21:06:58,215 - Epoch: [79][ 520/ 1236] Overall Loss 0.258234 Objective Loss 0.258234 LR 0.001000 Time 0.022284 +2023-10-02 21:06:58,421 - Epoch: [79][ 530/ 1236] Overall Loss 0.258162 Objective Loss 0.258162 LR 0.001000 Time 0.022253 +2023-10-02 21:06:58,632 - Epoch: [79][ 540/ 1236] Overall Loss 0.257859 Objective Loss 0.257859 LR 0.001000 Time 0.022230 +2023-10-02 21:06:58,838 - Epoch: [79][ 550/ 1236] Overall Loss 0.257379 Objective Loss 0.257379 LR 0.001000 Time 0.022200 +2023-10-02 21:06:59,049 - Epoch: [79][ 560/ 1236] Overall Loss 0.256565 Objective Loss 0.256565 LR 0.001000 Time 0.022179 +2023-10-02 21:06:59,255 - Epoch: [79][ 570/ 1236] Overall Loss 0.256820 Objective Loss 0.256820 LR 0.001000 Time 0.022151 +2023-10-02 21:06:59,466 - Epoch: [79][ 580/ 1236] Overall Loss 0.256948 Objective Loss 0.256948 LR 0.001000 Time 0.022132 +2023-10-02 21:06:59,672 - Epoch: [79][ 590/ 1236] Overall Loss 0.257025 Objective Loss 0.257025 LR 0.001000 Time 0.022106 +2023-10-02 21:06:59,882 - Epoch: [79][ 600/ 1236] Overall Loss 0.256808 Objective Loss 0.256808 LR 0.001000 Time 0.022088 +2023-10-02 21:07:00,089 - Epoch: [79][ 610/ 1236] Overall Loss 0.257266 Objective Loss 0.257266 LR 0.001000 Time 0.022064 +2023-10-02 21:07:00,299 - Epoch: [79][ 620/ 1236] Overall Loss 0.257800 Objective Loss 0.257800 LR 0.001000 Time 0.022046 +2023-10-02 21:07:00,505 - Epoch: [79][ 630/ 1236] Overall Loss 0.258278 Objective Loss 0.258278 LR 0.001000 Time 0.022024 +2023-10-02 21:07:00,716 - Epoch: [79][ 640/ 1236] Overall Loss 0.257809 Objective Loss 0.257809 LR 0.001000 Time 0.022008 +2023-10-02 21:07:00,923 - Epoch: [79][ 650/ 1236] Overall Loss 0.257578 Objective Loss 0.257578 LR 0.001000 Time 0.021987 +2023-10-02 21:07:01,133 - Epoch: [79][ 660/ 1236] Overall Loss 0.257617 Objective Loss 0.257617 LR 0.001000 Time 0.021973 +2023-10-02 21:07:01,340 - Epoch: [79][ 670/ 1236] Overall Loss 0.257778 Objective Loss 0.257778 LR 0.001000 Time 0.021952 +2023-10-02 21:07:01,550 - Epoch: [79][ 680/ 1236] Overall Loss 0.257749 Objective Loss 0.257749 LR 0.001000 Time 0.021938 +2023-10-02 21:07:01,757 - Epoch: [79][ 690/ 1236] Overall Loss 0.257566 Objective Loss 0.257566 LR 0.001000 Time 0.021919 +2023-10-02 21:07:01,967 - Epoch: [79][ 700/ 1236] Overall Loss 0.257626 Objective Loss 0.257626 LR 0.001000 Time 0.021906 +2023-10-02 21:07:02,174 - Epoch: [79][ 710/ 1236] Overall Loss 0.257749 Objective Loss 0.257749 LR 0.001000 Time 0.021888 +2023-10-02 21:07:02,384 - Epoch: [79][ 720/ 1236] Overall Loss 0.258151 Objective Loss 0.258151 LR 0.001000 Time 0.021876 +2023-10-02 21:07:02,591 - Epoch: [79][ 730/ 1236] Overall Loss 0.257984 Objective Loss 0.257984 LR 0.001000 Time 0.021859 +2023-10-02 21:07:02,801 - Epoch: [79][ 740/ 1236] Overall Loss 0.257840 Objective Loss 0.257840 LR 0.001000 Time 0.021848 +2023-10-02 21:07:03,008 - Epoch: [79][ 750/ 1236] Overall Loss 0.257533 Objective Loss 0.257533 LR 0.001000 Time 0.021831 +2023-10-02 21:07:03,218 - Epoch: [79][ 760/ 1236] Overall Loss 0.257932 Objective Loss 0.257932 LR 0.001000 Time 0.021821 +2023-10-02 21:07:03,425 - Epoch: [79][ 770/ 1236] Overall Loss 0.257593 Objective Loss 0.257593 LR 0.001000 Time 0.021805 +2023-10-02 21:07:03,635 - Epoch: [79][ 780/ 1236] Overall Loss 0.258030 Objective Loss 0.258030 LR 0.001000 Time 0.021795 +2023-10-02 21:07:03,841 - Epoch: [79][ 790/ 1236] Overall Loss 0.257847 Objective Loss 0.257847 LR 0.001000 Time 0.021780 +2023-10-02 21:07:04,052 - Epoch: [79][ 800/ 1236] Overall Loss 0.257736 Objective Loss 0.257736 LR 0.001000 Time 0.021771 +2023-10-02 21:07:04,259 - Epoch: [79][ 810/ 1236] Overall Loss 0.257856 Objective Loss 0.257856 LR 0.001000 Time 0.021757 +2023-10-02 21:07:04,469 - Epoch: [79][ 820/ 1236] Overall Loss 0.257839 Objective Loss 0.257839 LR 0.001000 Time 0.021747 +2023-10-02 21:07:04,676 - Epoch: [79][ 830/ 1236] Overall Loss 0.258069 Objective Loss 0.258069 LR 0.001000 Time 0.021734 +2023-10-02 21:07:04,886 - Epoch: [79][ 840/ 1236] Overall Loss 0.258182 Objective Loss 0.258182 LR 0.001000 Time 0.021725 +2023-10-02 21:07:05,093 - Epoch: [79][ 850/ 1236] Overall Loss 0.258087 Objective Loss 0.258087 LR 0.001000 Time 0.021712 +2023-10-02 21:07:05,302 - Epoch: [79][ 860/ 1236] Overall Loss 0.257846 Objective Loss 0.257846 LR 0.001000 Time 0.021703 +2023-10-02 21:07:05,509 - Epoch: [79][ 870/ 1236] Overall Loss 0.258355 Objective Loss 0.258355 LR 0.001000 Time 0.021691 +2023-10-02 21:07:05,720 - Epoch: [79][ 880/ 1236] Overall Loss 0.258360 Objective Loss 0.258360 LR 0.001000 Time 0.021684 +2023-10-02 21:07:05,926 - Epoch: [79][ 890/ 1236] Overall Loss 0.258057 Objective Loss 0.258057 LR 0.001000 Time 0.021672 +2023-10-02 21:07:06,137 - Epoch: [79][ 900/ 1236] Overall Loss 0.258290 Objective Loss 0.258290 LR 0.001000 Time 0.021664 +2023-10-02 21:07:06,343 - Epoch: [79][ 910/ 1236] Overall Loss 0.258780 Objective Loss 0.258780 LR 0.001000 Time 0.021653 +2023-10-02 21:07:06,553 - Epoch: [79][ 920/ 1236] Overall Loss 0.258835 Objective Loss 0.258835 LR 0.001000 Time 0.021646 +2023-10-02 21:07:06,760 - Epoch: [79][ 930/ 1236] Overall Loss 0.258936 Objective Loss 0.258936 LR 0.001000 Time 0.021635 +2023-10-02 21:07:06,971 - Epoch: [79][ 940/ 1236] Overall Loss 0.258813 Objective Loss 0.258813 LR 0.001000 Time 0.021628 +2023-10-02 21:07:07,177 - Epoch: [79][ 950/ 1236] Overall Loss 0.258992 Objective Loss 0.258992 LR 0.001000 Time 0.021618 +2023-10-02 21:07:07,388 - Epoch: [79][ 960/ 1236] Overall Loss 0.258980 Objective Loss 0.258980 LR 0.001000 Time 0.021612 +2023-10-02 21:07:07,594 - Epoch: [79][ 970/ 1236] Overall Loss 0.259069 Objective Loss 0.259069 LR 0.001000 Time 0.021602 +2023-10-02 21:07:07,805 - Epoch: [79][ 980/ 1236] Overall Loss 0.259109 Objective Loss 0.259109 LR 0.001000 Time 0.021596 +2023-10-02 21:07:08,012 - Epoch: [79][ 990/ 1236] Overall Loss 0.259335 Objective Loss 0.259335 LR 0.001000 Time 0.021586 +2023-10-02 21:07:08,222 - Epoch: [79][ 1000/ 1236] Overall Loss 0.259313 Objective Loss 0.259313 LR 0.001000 Time 0.021581 +2023-10-02 21:07:08,429 - Epoch: [79][ 1010/ 1236] Overall Loss 0.259196 Objective Loss 0.259196 LR 0.001000 Time 0.021571 +2023-10-02 21:07:08,639 - Epoch: [79][ 1020/ 1236] Overall Loss 0.259187 Objective Loss 0.259187 LR 0.001000 Time 0.021566 +2023-10-02 21:07:08,846 - Epoch: [79][ 1030/ 1236] Overall Loss 0.259197 Objective Loss 0.259197 LR 0.001000 Time 0.021556 +2023-10-02 21:07:09,056 - Epoch: [79][ 1040/ 1236] Overall Loss 0.259388 Objective Loss 0.259388 LR 0.001000 Time 0.021551 +2023-10-02 21:07:09,263 - Epoch: [79][ 1050/ 1236] Overall Loss 0.259326 Objective Loss 0.259326 LR 0.001000 Time 0.021542 +2023-10-02 21:07:09,473 - Epoch: [79][ 1060/ 1236] Overall Loss 0.259316 Objective Loss 0.259316 LR 0.001000 Time 0.021537 +2023-10-02 21:07:09,680 - Epoch: [79][ 1070/ 1236] Overall Loss 0.259205 Objective Loss 0.259205 LR 0.001000 Time 0.021529 +2023-10-02 21:07:09,891 - Epoch: [79][ 1080/ 1236] Overall Loss 0.259472 Objective Loss 0.259472 LR 0.001000 Time 0.021525 +2023-10-02 21:07:10,098 - Epoch: [79][ 1090/ 1236] Overall Loss 0.259715 Objective Loss 0.259715 LR 0.001000 Time 0.021517 +2023-10-02 21:07:10,309 - Epoch: [79][ 1100/ 1236] Overall Loss 0.259762 Objective Loss 0.259762 LR 0.001000 Time 0.021513 +2023-10-02 21:07:10,515 - Epoch: [79][ 1110/ 1236] Overall Loss 0.259714 Objective Loss 0.259714 LR 0.001000 Time 0.021505 +2023-10-02 21:07:10,726 - Epoch: [79][ 1120/ 1236] Overall Loss 0.259844 Objective Loss 0.259844 LR 0.001000 Time 0.021500 +2023-10-02 21:07:10,932 - Epoch: [79][ 1130/ 1236] Overall Loss 0.259961 Objective Loss 0.259961 LR 0.001000 Time 0.021493 +2023-10-02 21:07:11,143 - Epoch: [79][ 1140/ 1236] Overall Loss 0.259927 Objective Loss 0.259927 LR 0.001000 Time 0.021489 +2023-10-02 21:07:11,350 - Epoch: [79][ 1150/ 1236] Overall Loss 0.259707 Objective Loss 0.259707 LR 0.001000 Time 0.021481 +2023-10-02 21:07:11,560 - Epoch: [79][ 1160/ 1236] Overall Loss 0.259844 Objective Loss 0.259844 LR 0.001000 Time 0.021477 +2023-10-02 21:07:11,767 - Epoch: [79][ 1170/ 1236] Overall Loss 0.259675 Objective Loss 0.259675 LR 0.001000 Time 0.021470 +2023-10-02 21:07:11,977 - Epoch: [79][ 1180/ 1236] Overall Loss 0.259742 Objective Loss 0.259742 LR 0.001000 Time 0.021466 +2023-10-02 21:07:12,184 - Epoch: [79][ 1190/ 1236] Overall Loss 0.259771 Objective Loss 0.259771 LR 0.001000 Time 0.021459 +2023-10-02 21:07:12,394 - Epoch: [79][ 1200/ 1236] Overall Loss 0.259744 Objective Loss 0.259744 LR 0.001000 Time 0.021456 +2023-10-02 21:07:12,601 - Epoch: [79][ 1210/ 1236] Overall Loss 0.259686 Objective Loss 0.259686 LR 0.001000 Time 0.021449 +2023-10-02 21:07:12,813 - Epoch: [79][ 1220/ 1236] Overall Loss 0.259633 Objective Loss 0.259633 LR 0.001000 Time 0.021447 +2023-10-02 21:07:13,080 - Epoch: [79][ 1230/ 1236] Overall Loss 0.259510 Objective Loss 0.259510 LR 0.001000 Time 0.021489 +2023-10-02 21:07:13,202 - Epoch: [79][ 1236/ 1236] Overall Loss 0.259466 Objective Loss 0.259466 Top1 86.761711 Top5 98.981670 LR 0.001000 Time 0.021483 +2023-10-02 21:07:13,345 - --- validate (epoch=79)----------- +2023-10-02 21:07:13,345 - 29943 samples (256 per mini-batch) +2023-10-02 21:07:13,842 - Epoch: [79][ 10/ 117] Loss 0.336148 Top1 83.945312 Top5 98.007812 +2023-10-02 21:07:14,001 - Epoch: [79][ 20/ 117] Loss 0.323960 Top1 84.335938 Top5 98.320312 +2023-10-02 21:07:14,161 - Epoch: [79][ 30/ 117] Loss 0.332864 Top1 84.257812 Top5 98.281250 +2023-10-02 21:07:14,319 - Epoch: [79][ 40/ 117] Loss 0.339728 Top1 84.091797 Top5 98.222656 +2023-10-02 21:07:14,479 - Epoch: [79][ 50/ 117] Loss 0.334862 Top1 84.039062 Top5 98.226562 +2023-10-02 21:07:14,636 - Epoch: [79][ 60/ 117] Loss 0.338564 Top1 83.984375 Top5 98.268229 +2023-10-02 21:07:14,796 - Epoch: [79][ 70/ 117] Loss 0.341045 Top1 83.989955 Top5 98.231027 +2023-10-02 21:07:14,955 - Epoch: [79][ 80/ 117] Loss 0.337136 Top1 84.082031 Top5 98.271484 +2023-10-02 21:07:15,116 - Epoch: [79][ 90/ 117] Loss 0.340101 Top1 84.019097 Top5 98.255208 +2023-10-02 21:07:15,268 - Epoch: [79][ 100/ 117] Loss 0.338579 Top1 84.003906 Top5 98.265625 +2023-10-02 21:07:15,425 - Epoch: [79][ 110/ 117] Loss 0.340450 Top1 84.023438 Top5 98.238636 +2023-10-02 21:07:15,513 - Epoch: [79][ 117/ 117] Loss 0.341527 Top1 84.009618 Top5 98.229970 +2023-10-02 21:07:15,648 - ==> Top1: 84.010 Top5: 98.230 Loss: 0.342 + +2023-10-02 21:07:15,649 - ==> Confusion: +[[ 950 0 3 0 8 2 0 1 4 55 1 1 0 2 5 2 0 1 0 0 15] + [ 5 1040 2 1 5 35 0 15 0 1 1 2 1 2 0 5 0 0 8 1 7] + [ 3 1 977 9 4 0 16 10 0 0 2 2 2 2 3 4 1 0 8 3 9] + [ 2 1 24 938 0 4 1 0 11 0 9 0 11 3 37 3 2 3 17 0 23] + [ 27 4 4 1 974 5 0 0 0 11 0 0 0 4 5 5 5 0 0 1 4] + [ 3 45 2 1 4 956 0 35 3 6 3 6 2 12 4 2 2 0 4 8 18] + [ 1 3 31 0 1 1 1123 5 0 0 5 1 0 1 1 3 0 0 2 7 6] + [ 3 15 21 0 7 21 8 1060 4 2 5 4 3 2 2 0 4 0 42 7 8] + [ 23 2 2 0 4 2 0 1 962 50 8 1 0 16 13 0 2 0 0 2 1] + [ 143 1 3 1 8 3 2 1 25 888 1 0 0 22 8 0 3 0 0 1 9] + [ 4 1 11 10 1 3 5 2 16 0 955 3 0 16 5 0 3 0 8 1 9] + [ 1 2 3 0 1 12 0 3 0 1 0 953 16 8 0 3 0 12 1 11 8] + [ 0 1 4 4 1 4 4 0 2 0 0 72 913 5 7 12 3 15 1 7 13] + [ 4 0 4 0 3 7 0 2 11 14 2 6 0 1048 7 0 0 1 0 0 10] + [ 14 0 1 11 10 0 0 0 35 2 6 1 2 5 1001 0 1 0 7 0 5] + [ 0 0 4 2 5 0 1 0 0 0 0 9 5 0 0 1068 13 9 2 10 6] + [ 3 12 3 0 6 6 1 0 2 1 0 8 2 2 3 11 1078 0 0 8 15] + [ 1 1 2 4 0 1 2 0 1 1 0 12 25 2 3 11 0 964 1 2 5] + [ 3 13 10 20 1 0 0 19 7 1 4 1 2 0 16 1 0 0 960 0 10] + [ 0 2 5 2 0 7 5 8 0 1 2 9 4 0 0 4 4 1 1 1095 2] + [ 177 141 166 65 95 132 37 87 96 67 146 141 290 290 158 70 78 50 131 236 5252]] + +2023-10-02 21:07:15,650 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:07:15,650 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:07:15,656 - + +2023-10-02 21:07:15,656 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:07:16,788 - Epoch: [80][ 10/ 1236] Overall Loss 0.274538 Objective Loss 0.274538 LR 0.001000 Time 0.113101 +2023-10-02 21:07:16,996 - Epoch: [80][ 20/ 1236] Overall Loss 0.267609 Objective Loss 0.267609 LR 0.001000 Time 0.066932 +2023-10-02 21:07:17,203 - Epoch: [80][ 30/ 1236] Overall Loss 0.271584 Objective Loss 0.271584 LR 0.001000 Time 0.051493 +2023-10-02 21:07:17,412 - Epoch: [80][ 40/ 1236] Overall Loss 0.264522 Objective Loss 0.264522 LR 0.001000 Time 0.043833 +2023-10-02 21:07:17,618 - Epoch: [80][ 50/ 1236] Overall Loss 0.255926 Objective Loss 0.255926 LR 0.001000 Time 0.039178 +2023-10-02 21:07:17,827 - Epoch: [80][ 60/ 1236] Overall Loss 0.253786 Objective Loss 0.253786 LR 0.001000 Time 0.036134 +2023-10-02 21:07:18,033 - Epoch: [80][ 70/ 1236] Overall Loss 0.254055 Objective Loss 0.254055 LR 0.001000 Time 0.033910 +2023-10-02 21:07:18,242 - Epoch: [80][ 80/ 1236] Overall Loss 0.256289 Objective Loss 0.256289 LR 0.001000 Time 0.032283 +2023-10-02 21:07:18,448 - Epoch: [80][ 90/ 1236] Overall Loss 0.255459 Objective Loss 0.255459 LR 0.001000 Time 0.030982 +2023-10-02 21:07:18,657 - Epoch: [80][ 100/ 1236] Overall Loss 0.257117 Objective Loss 0.257117 LR 0.001000 Time 0.029964 +2023-10-02 21:07:18,864 - Epoch: [80][ 110/ 1236] Overall Loss 0.256637 Objective Loss 0.256637 LR 0.001000 Time 0.029126 +2023-10-02 21:07:19,073 - Epoch: [80][ 120/ 1236] Overall Loss 0.255449 Objective Loss 0.255449 LR 0.001000 Time 0.028433 +2023-10-02 21:07:19,281 - Epoch: [80][ 130/ 1236] Overall Loss 0.255259 Objective Loss 0.255259 LR 0.001000 Time 0.027838 +2023-10-02 21:07:19,489 - Epoch: [80][ 140/ 1236] Overall Loss 0.253339 Objective Loss 0.253339 LR 0.001000 Time 0.027334 +2023-10-02 21:07:19,696 - Epoch: [80][ 150/ 1236] Overall Loss 0.253449 Objective Loss 0.253449 LR 0.001000 Time 0.026880 +2023-10-02 21:07:19,904 - Epoch: [80][ 160/ 1236] Overall Loss 0.251767 Objective Loss 0.251767 LR 0.001000 Time 0.026495 +2023-10-02 21:07:20,113 - Epoch: [80][ 170/ 1236] Overall Loss 0.252970 Objective Loss 0.252970 LR 0.001000 Time 0.026158 +2023-10-02 21:07:20,323 - Epoch: [80][ 180/ 1236] Overall Loss 0.252807 Objective Loss 0.252807 LR 0.001000 Time 0.025869 +2023-10-02 21:07:20,531 - Epoch: [80][ 190/ 1236] Overall Loss 0.253432 Objective Loss 0.253432 LR 0.001000 Time 0.025594 +2023-10-02 21:07:20,741 - Epoch: [80][ 200/ 1236] Overall Loss 0.253215 Objective Loss 0.253215 LR 0.001000 Time 0.025362 +2023-10-02 21:07:20,949 - Epoch: [80][ 210/ 1236] Overall Loss 0.254760 Objective Loss 0.254760 LR 0.001000 Time 0.025139 +2023-10-02 21:07:21,159 - Epoch: [80][ 220/ 1236] Overall Loss 0.254879 Objective Loss 0.254879 LR 0.001000 Time 0.024950 +2023-10-02 21:07:21,367 - Epoch: [80][ 230/ 1236] Overall Loss 0.254610 Objective Loss 0.254610 LR 0.001000 Time 0.024764 +2023-10-02 21:07:21,577 - Epoch: [80][ 240/ 1236] Overall Loss 0.254507 Objective Loss 0.254507 LR 0.001000 Time 0.024606 +2023-10-02 21:07:21,785 - Epoch: [80][ 250/ 1236] Overall Loss 0.254872 Objective Loss 0.254872 LR 0.001000 Time 0.024447 +2023-10-02 21:07:21,996 - Epoch: [80][ 260/ 1236] Overall Loss 0.253486 Objective Loss 0.253486 LR 0.001000 Time 0.024315 +2023-10-02 21:07:22,204 - Epoch: [80][ 270/ 1236] Overall Loss 0.253702 Objective Loss 0.253702 LR 0.001000 Time 0.024181 +2023-10-02 21:07:22,414 - Epoch: [80][ 280/ 1236] Overall Loss 0.252602 Objective Loss 0.252602 LR 0.001000 Time 0.024065 +2023-10-02 21:07:22,622 - Epoch: [80][ 290/ 1236] Overall Loss 0.252032 Objective Loss 0.252032 LR 0.001000 Time 0.023952 +2023-10-02 21:07:22,832 - Epoch: [80][ 300/ 1236] Overall Loss 0.252717 Objective Loss 0.252717 LR 0.001000 Time 0.023853 +2023-10-02 21:07:23,040 - Epoch: [80][ 310/ 1236] Overall Loss 0.253167 Objective Loss 0.253167 LR 0.001000 Time 0.023750 +2023-10-02 21:07:23,250 - Epoch: [80][ 320/ 1236] Overall Loss 0.252542 Objective Loss 0.252542 LR 0.001000 Time 0.023664 +2023-10-02 21:07:23,459 - Epoch: [80][ 330/ 1236] Overall Loss 0.252151 Objective Loss 0.252151 LR 0.001000 Time 0.023574 +2023-10-02 21:07:23,669 - Epoch: [80][ 340/ 1236] Overall Loss 0.252680 Objective Loss 0.252680 LR 0.001000 Time 0.023498 +2023-10-02 21:07:23,878 - Epoch: [80][ 350/ 1236] Overall Loss 0.252912 Objective Loss 0.252912 LR 0.001000 Time 0.023419 +2023-10-02 21:07:24,089 - Epoch: [80][ 360/ 1236] Overall Loss 0.252258 Objective Loss 0.252258 LR 0.001000 Time 0.023353 +2023-10-02 21:07:24,295 - Epoch: [80][ 370/ 1236] Overall Loss 0.252511 Objective Loss 0.252511 LR 0.001000 Time 0.023279 +2023-10-02 21:07:24,504 - Epoch: [80][ 380/ 1236] Overall Loss 0.252051 Objective Loss 0.252051 LR 0.001000 Time 0.023217 +2023-10-02 21:07:24,710 - Epoch: [80][ 390/ 1236] Overall Loss 0.251735 Objective Loss 0.251735 LR 0.001000 Time 0.023149 +2023-10-02 21:07:24,916 - Epoch: [80][ 400/ 1236] Overall Loss 0.251965 Objective Loss 0.251965 LR 0.001000 Time 0.023084 +2023-10-02 21:07:25,122 - Epoch: [80][ 410/ 1236] Overall Loss 0.252193 Objective Loss 0.252193 LR 0.001000 Time 0.023019 +2023-10-02 21:07:25,327 - Epoch: [80][ 420/ 1236] Overall Loss 0.251491 Objective Loss 0.251491 LR 0.001000 Time 0.022959 +2023-10-02 21:07:25,531 - Epoch: [80][ 430/ 1236] Overall Loss 0.251332 Objective Loss 0.251332 LR 0.001000 Time 0.022897 +2023-10-02 21:07:25,737 - Epoch: [80][ 440/ 1236] Overall Loss 0.251844 Objective Loss 0.251844 LR 0.001000 Time 0.022844 +2023-10-02 21:07:25,942 - Epoch: [80][ 450/ 1236] Overall Loss 0.252118 Objective Loss 0.252118 LR 0.001000 Time 0.022788 +2023-10-02 21:07:26,148 - Epoch: [80][ 460/ 1236] Overall Loss 0.251942 Objective Loss 0.251942 LR 0.001000 Time 0.022740 +2023-10-02 21:07:26,354 - Epoch: [80][ 470/ 1236] Overall Loss 0.252665 Objective Loss 0.252665 LR 0.001000 Time 0.022691 +2023-10-02 21:07:26,558 - Epoch: [80][ 480/ 1236] Overall Loss 0.253017 Objective Loss 0.253017 LR 0.001000 Time 0.022643 +2023-10-02 21:07:26,763 - Epoch: [80][ 490/ 1236] Overall Loss 0.253210 Objective Loss 0.253210 LR 0.001000 Time 0.022596 +2023-10-02 21:07:26,969 - Epoch: [80][ 500/ 1236] Overall Loss 0.253044 Objective Loss 0.253044 LR 0.001000 Time 0.022555 +2023-10-02 21:07:27,174 - Epoch: [80][ 510/ 1236] Overall Loss 0.253613 Objective Loss 0.253613 LR 0.001000 Time 0.022513 +2023-10-02 21:07:27,380 - Epoch: [80][ 520/ 1236] Overall Loss 0.253629 Objective Loss 0.253629 LR 0.001000 Time 0.022475 +2023-10-02 21:07:27,585 - Epoch: [80][ 530/ 1236] Overall Loss 0.254153 Objective Loss 0.254153 LR 0.001000 Time 0.022435 +2023-10-02 21:07:27,791 - Epoch: [80][ 540/ 1236] Overall Loss 0.254061 Objective Loss 0.254061 LR 0.001000 Time 0.022400 +2023-10-02 21:07:27,996 - Epoch: [80][ 550/ 1236] Overall Loss 0.254423 Objective Loss 0.254423 LR 0.001000 Time 0.022363 +2023-10-02 21:07:28,202 - Epoch: [80][ 560/ 1236] Overall Loss 0.254430 Objective Loss 0.254430 LR 0.001000 Time 0.022331 +2023-10-02 21:07:28,405 - Epoch: [80][ 570/ 1236] Overall Loss 0.254864 Objective Loss 0.254864 LR 0.001000 Time 0.022293 +2023-10-02 21:07:28,610 - Epoch: [80][ 580/ 1236] Overall Loss 0.254820 Objective Loss 0.254820 LR 0.001000 Time 0.022261 +2023-10-02 21:07:28,815 - Epoch: [80][ 590/ 1236] Overall Loss 0.255242 Objective Loss 0.255242 LR 0.001000 Time 0.022230 +2023-10-02 21:07:29,020 - Epoch: [80][ 600/ 1236] Overall Loss 0.255708 Objective Loss 0.255708 LR 0.001000 Time 0.022202 +2023-10-02 21:07:29,226 - Epoch: [80][ 610/ 1236] Overall Loss 0.255807 Objective Loss 0.255807 LR 0.001000 Time 0.022172 +2023-10-02 21:07:29,431 - Epoch: [80][ 620/ 1236] Overall Loss 0.255929 Objective Loss 0.255929 LR 0.001000 Time 0.022146 +2023-10-02 21:07:29,637 - Epoch: [80][ 630/ 1236] Overall Loss 0.255784 Objective Loss 0.255784 LR 0.001000 Time 0.022118 +2023-10-02 21:07:29,842 - Epoch: [80][ 640/ 1236] Overall Loss 0.255600 Objective Loss 0.255600 LR 0.001000 Time 0.022093 +2023-10-02 21:07:30,047 - Epoch: [80][ 650/ 1236] Overall Loss 0.255863 Objective Loss 0.255863 LR 0.001000 Time 0.022066 +2023-10-02 21:07:30,253 - Epoch: [80][ 660/ 1236] Overall Loss 0.255695 Objective Loss 0.255695 LR 0.001000 Time 0.022044 +2023-10-02 21:07:30,458 - Epoch: [80][ 670/ 1236] Overall Loss 0.255972 Objective Loss 0.255972 LR 0.001000 Time 0.022018 +2023-10-02 21:07:30,664 - Epoch: [80][ 680/ 1236] Overall Loss 0.256547 Objective Loss 0.256547 LR 0.001000 Time 0.021997 +2023-10-02 21:07:30,869 - Epoch: [80][ 690/ 1236] Overall Loss 0.256776 Objective Loss 0.256776 LR 0.001000 Time 0.021973 +2023-10-02 21:07:31,075 - Epoch: [80][ 700/ 1236] Overall Loss 0.256913 Objective Loss 0.256913 LR 0.001000 Time 0.021953 +2023-10-02 21:07:31,280 - Epoch: [80][ 710/ 1236] Overall Loss 0.257379 Objective Loss 0.257379 LR 0.001000 Time 0.021930 +2023-10-02 21:07:31,485 - Epoch: [80][ 720/ 1236] Overall Loss 0.257714 Objective Loss 0.257714 LR 0.001000 Time 0.021909 +2023-10-02 21:07:31,690 - Epoch: [80][ 730/ 1236] Overall Loss 0.257691 Objective Loss 0.257691 LR 0.001000 Time 0.021888 +2023-10-02 21:07:31,895 - Epoch: [80][ 740/ 1236] Overall Loss 0.257745 Objective Loss 0.257745 LR 0.001000 Time 0.021870 +2023-10-02 21:07:32,099 - Epoch: [80][ 750/ 1236] Overall Loss 0.257978 Objective Loss 0.257978 LR 0.001000 Time 0.021848 +2023-10-02 21:07:32,305 - Epoch: [80][ 760/ 1236] Overall Loss 0.258081 Objective Loss 0.258081 LR 0.001000 Time 0.021831 +2023-10-02 21:07:32,510 - Epoch: [80][ 770/ 1236] Overall Loss 0.257920 Objective Loss 0.257920 LR 0.001000 Time 0.021812 +2023-10-02 21:07:32,715 - Epoch: [80][ 780/ 1236] Overall Loss 0.257819 Objective Loss 0.257819 LR 0.001000 Time 0.021795 +2023-10-02 21:07:32,921 - Epoch: [80][ 790/ 1236] Overall Loss 0.257832 Objective Loss 0.257832 LR 0.001000 Time 0.021777 +2023-10-02 21:07:33,126 - Epoch: [80][ 800/ 1236] Overall Loss 0.258022 Objective Loss 0.258022 LR 0.001000 Time 0.021762 +2023-10-02 21:07:33,332 - Epoch: [80][ 810/ 1236] Overall Loss 0.257970 Objective Loss 0.257970 LR 0.001000 Time 0.021745 +2023-10-02 21:07:33,536 - Epoch: [80][ 820/ 1236] Overall Loss 0.257945 Objective Loss 0.257945 LR 0.001000 Time 0.021728 +2023-10-02 21:07:33,741 - Epoch: [80][ 830/ 1236] Overall Loss 0.258474 Objective Loss 0.258474 LR 0.001000 Time 0.021712 +2023-10-02 21:07:33,947 - Epoch: [80][ 840/ 1236] Overall Loss 0.258592 Objective Loss 0.258592 LR 0.001000 Time 0.021698 +2023-10-02 21:07:34,152 - Epoch: [80][ 850/ 1236] Overall Loss 0.259481 Objective Loss 0.259481 LR 0.001000 Time 0.021682 +2023-10-02 21:07:34,356 - Epoch: [80][ 860/ 1236] Overall Loss 0.259826 Objective Loss 0.259826 LR 0.001000 Time 0.021668 +2023-10-02 21:07:34,562 - Epoch: [80][ 870/ 1236] Overall Loss 0.260038 Objective Loss 0.260038 LR 0.001000 Time 0.021654 +2023-10-02 21:07:34,767 - Epoch: [80][ 880/ 1236] Overall Loss 0.260472 Objective Loss 0.260472 LR 0.001000 Time 0.021642 +2023-10-02 21:07:34,972 - Epoch: [80][ 890/ 1236] Overall Loss 0.260790 Objective Loss 0.260790 LR 0.001000 Time 0.021627 +2023-10-02 21:07:35,178 - Epoch: [80][ 900/ 1236] Overall Loss 0.261152 Objective Loss 0.261152 LR 0.001000 Time 0.021615 +2023-10-02 21:07:35,383 - Epoch: [80][ 910/ 1236] Overall Loss 0.261360 Objective Loss 0.261360 LR 0.001000 Time 0.021601 +2023-10-02 21:07:35,589 - Epoch: [80][ 920/ 1236] Overall Loss 0.261124 Objective Loss 0.261124 LR 0.001000 Time 0.021590 +2023-10-02 21:07:35,795 - Epoch: [80][ 930/ 1236] Overall Loss 0.261260 Objective Loss 0.261260 LR 0.001000 Time 0.021577 +2023-10-02 21:07:35,999 - Epoch: [80][ 940/ 1236] Overall Loss 0.261032 Objective Loss 0.261032 LR 0.001000 Time 0.021564 +2023-10-02 21:07:36,204 - Epoch: [80][ 950/ 1236] Overall Loss 0.261179 Objective Loss 0.261179 LR 0.001000 Time 0.021552 +2023-10-02 21:07:36,410 - Epoch: [80][ 960/ 1236] Overall Loss 0.261056 Objective Loss 0.261056 LR 0.001000 Time 0.021541 +2023-10-02 21:07:36,615 - Epoch: [80][ 970/ 1236] Overall Loss 0.261285 Objective Loss 0.261285 LR 0.001000 Time 0.021529 +2023-10-02 21:07:36,821 - Epoch: [80][ 980/ 1236] Overall Loss 0.261359 Objective Loss 0.261359 LR 0.001000 Time 0.021519 +2023-10-02 21:07:37,026 - Epoch: [80][ 990/ 1236] Overall Loss 0.261508 Objective Loss 0.261508 LR 0.001000 Time 0.021507 +2023-10-02 21:07:37,231 - Epoch: [80][ 1000/ 1236] Overall Loss 0.261534 Objective Loss 0.261534 LR 0.001000 Time 0.021498 +2023-10-02 21:07:37,437 - Epoch: [80][ 1010/ 1236] Overall Loss 0.261510 Objective Loss 0.261510 LR 0.001000 Time 0.021487 +2023-10-02 21:07:37,642 - Epoch: [80][ 1020/ 1236] Overall Loss 0.261491 Objective Loss 0.261491 LR 0.001000 Time 0.021477 +2023-10-02 21:07:37,847 - Epoch: [80][ 1030/ 1236] Overall Loss 0.261280 Objective Loss 0.261280 LR 0.001000 Time 0.021467 +2023-10-02 21:07:38,053 - Epoch: [80][ 1040/ 1236] Overall Loss 0.261507 Objective Loss 0.261507 LR 0.001000 Time 0.021458 +2023-10-02 21:07:38,258 - Epoch: [80][ 1050/ 1236] Overall Loss 0.261520 Objective Loss 0.261520 LR 0.001000 Time 0.021447 +2023-10-02 21:07:38,464 - Epoch: [80][ 1060/ 1236] Overall Loss 0.261351 Objective Loss 0.261351 LR 0.001000 Time 0.021439 +2023-10-02 21:07:38,669 - Epoch: [80][ 1070/ 1236] Overall Loss 0.261269 Objective Loss 0.261269 LR 0.001000 Time 0.021429 +2023-10-02 21:07:38,875 - Epoch: [80][ 1080/ 1236] Overall Loss 0.261564 Objective Loss 0.261564 LR 0.001000 Time 0.021420 +2023-10-02 21:07:39,080 - Epoch: [80][ 1090/ 1236] Overall Loss 0.261602 Objective Loss 0.261602 LR 0.001000 Time 0.021411 +2023-10-02 21:07:39,286 - Epoch: [80][ 1100/ 1236] Overall Loss 0.261784 Objective Loss 0.261784 LR 0.001000 Time 0.021403 +2023-10-02 21:07:39,491 - Epoch: [80][ 1110/ 1236] Overall Loss 0.261647 Objective Loss 0.261647 LR 0.001000 Time 0.021393 +2023-10-02 21:07:39,697 - Epoch: [80][ 1120/ 1236] Overall Loss 0.261575 Objective Loss 0.261575 LR 0.001000 Time 0.021386 +2023-10-02 21:07:39,901 - Epoch: [80][ 1130/ 1236] Overall Loss 0.261610 Objective Loss 0.261610 LR 0.001000 Time 0.021376 +2023-10-02 21:07:40,107 - Epoch: [80][ 1140/ 1236] Overall Loss 0.261498 Objective Loss 0.261498 LR 0.001000 Time 0.021369 +2023-10-02 21:07:40,312 - Epoch: [80][ 1150/ 1236] Overall Loss 0.261571 Objective Loss 0.261571 LR 0.001000 Time 0.021360 +2023-10-02 21:07:40,518 - Epoch: [80][ 1160/ 1236] Overall Loss 0.261371 Objective Loss 0.261371 LR 0.001000 Time 0.021353 +2023-10-02 21:07:40,723 - Epoch: [80][ 1170/ 1236] Overall Loss 0.261379 Objective Loss 0.261379 LR 0.001000 Time 0.021345 +2023-10-02 21:07:40,929 - Epoch: [80][ 1180/ 1236] Overall Loss 0.261364 Objective Loss 0.261364 LR 0.001000 Time 0.021338 +2023-10-02 21:07:41,134 - Epoch: [80][ 1190/ 1236] Overall Loss 0.261348 Objective Loss 0.261348 LR 0.001000 Time 0.021330 +2023-10-02 21:07:41,340 - Epoch: [80][ 1200/ 1236] Overall Loss 0.261079 Objective Loss 0.261079 LR 0.001000 Time 0.021323 +2023-10-02 21:07:41,545 - Epoch: [80][ 1210/ 1236] Overall Loss 0.261248 Objective Loss 0.261248 LR 0.001000 Time 0.021315 +2023-10-02 21:07:41,751 - Epoch: [80][ 1220/ 1236] Overall Loss 0.261090 Objective Loss 0.261090 LR 0.001000 Time 0.021309 +2023-10-02 21:07:42,010 - Epoch: [80][ 1230/ 1236] Overall Loss 0.261377 Objective Loss 0.261377 LR 0.001000 Time 0.021345 +2023-10-02 21:07:42,130 - Epoch: [80][ 1236/ 1236] Overall Loss 0.261483 Objective Loss 0.261483 Top1 85.743381 Top5 98.167006 LR 0.001000 Time 0.021339 +2023-10-02 21:07:42,263 - --- validate (epoch=80)----------- +2023-10-02 21:07:42,263 - 29943 samples (256 per mini-batch) +2023-10-02 21:07:42,754 - Epoch: [80][ 10/ 117] Loss 0.343643 Top1 83.554688 Top5 98.085938 +2023-10-02 21:07:42,916 - Epoch: [80][ 20/ 117] Loss 0.339002 Top1 84.042969 Top5 98.125000 +2023-10-02 21:07:43,065 - Epoch: [80][ 30/ 117] Loss 0.331721 Top1 84.296875 Top5 98.281250 +2023-10-02 21:07:43,216 - Epoch: [80][ 40/ 117] Loss 0.333806 Top1 84.218750 Top5 98.271484 +2023-10-02 21:07:43,365 - Epoch: [80][ 50/ 117] Loss 0.334683 Top1 83.937500 Top5 98.281250 +2023-10-02 21:07:43,516 - Epoch: [80][ 60/ 117] Loss 0.339477 Top1 83.593750 Top5 98.203125 +2023-10-02 21:07:43,665 - Epoch: [80][ 70/ 117] Loss 0.336101 Top1 83.593750 Top5 98.141741 +2023-10-02 21:07:43,816 - Epoch: [80][ 80/ 117] Loss 0.340549 Top1 83.510742 Top5 98.125000 +2023-10-02 21:07:43,965 - Epoch: [80][ 90/ 117] Loss 0.336169 Top1 83.589410 Top5 98.172743 +2023-10-02 21:07:44,117 - Epoch: [80][ 100/ 117] Loss 0.333576 Top1 83.652344 Top5 98.140625 +2023-10-02 21:07:44,274 - Epoch: [80][ 110/ 117] Loss 0.332327 Top1 83.686080 Top5 98.142756 +2023-10-02 21:07:44,362 - Epoch: [80][ 117/ 117] Loss 0.332875 Top1 83.662292 Top5 98.136459 +2023-10-02 21:07:44,500 - ==> Top1: 83.662 Top5: 98.136 Loss: 0.333 + +2023-10-02 21:07:44,501 - ==> Confusion: +[[ 913 4 2 0 5 2 0 2 12 78 2 0 0 2 10 0 1 2 1 0 14] + [ 1 1057 1 1 4 25 0 15 3 1 2 3 0 1 1 3 0 0 6 2 5] + [ 5 0 970 10 2 0 14 9 0 2 2 2 6 2 1 2 2 1 14 3 9] + [ 0 2 10 954 0 5 1 6 3 3 8 0 6 3 45 1 0 3 22 1 16] + [ 19 14 1 1 945 3 0 0 0 13 2 2 0 6 17 6 11 1 1 2 6] + [ 3 60 1 1 2 956 1 26 6 9 2 14 2 7 8 1 1 0 2 5 9] + [ 1 6 27 0 0 3 1125 4 0 1 4 2 0 0 1 2 0 0 1 8 6] + [ 4 27 12 1 0 27 8 1070 3 2 8 5 1 2 1 1 0 2 27 11 6] + [ 14 5 0 0 1 2 0 1 977 33 11 2 6 8 19 0 1 2 3 1 3] + [ 76 1 1 0 5 2 2 0 49 926 1 1 1 24 18 1 0 0 1 2 8] + [ 0 1 9 6 0 2 4 4 16 1 972 6 0 10 7 0 1 1 4 2 7] + [ 1 2 2 0 0 7 0 3 0 1 1 964 19 10 0 4 0 14 0 6 1] + [ 0 0 0 3 2 2 2 3 3 0 4 53 942 5 3 8 1 17 5 8 7] + [ 2 0 1 1 1 6 0 0 16 14 6 12 0 1033 7 0 0 0 0 5 15] + [ 9 2 0 10 1 0 0 1 24 1 3 0 3 3 1020 0 3 1 8 1 11] + [ 0 1 2 3 6 0 4 0 0 0 1 8 9 1 1 1045 15 18 2 9 9] + [ 0 22 1 0 5 9 3 1 2 0 0 7 1 0 4 13 1080 0 0 6 7] + [ 0 0 0 4 1 0 3 0 1 2 0 11 25 1 10 3 0 974 0 0 3] + [ 1 9 7 10 0 0 1 19 8 0 6 1 5 0 9 0 0 0 980 2 10] + [ 0 1 3 2 0 4 7 14 0 1 2 15 3 1 0 3 2 1 0 1089 4] + [ 111 287 115 79 91 141 46 96 112 82 193 135 375 229 186 35 112 64 132 225 5059]] + +2023-10-02 21:07:44,502 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:07:44,502 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:07:44,508 - + +2023-10-02 21:07:44,508 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:07:45,515 - Epoch: [81][ 10/ 1236] Overall Loss 0.271282 Objective Loss 0.271282 LR 0.001000 Time 0.100663 +2023-10-02 21:07:45,722 - Epoch: [81][ 20/ 1236] Overall Loss 0.267955 Objective Loss 0.267955 LR 0.001000 Time 0.060674 +2023-10-02 21:07:45,928 - Epoch: [81][ 30/ 1236] Overall Loss 0.266070 Objective Loss 0.266070 LR 0.001000 Time 0.047277 +2023-10-02 21:07:46,137 - Epoch: [81][ 40/ 1236] Overall Loss 0.265589 Objective Loss 0.265589 LR 0.001000 Time 0.040668 +2023-10-02 21:07:46,342 - Epoch: [81][ 50/ 1236] Overall Loss 0.265486 Objective Loss 0.265486 LR 0.001000 Time 0.036621 +2023-10-02 21:07:46,551 - Epoch: [81][ 60/ 1236] Overall Loss 0.263914 Objective Loss 0.263914 LR 0.001000 Time 0.033992 +2023-10-02 21:07:46,755 - Epoch: [81][ 70/ 1236] Overall Loss 0.267856 Objective Loss 0.267856 LR 0.001000 Time 0.032055 +2023-10-02 21:07:46,964 - Epoch: [81][ 80/ 1236] Overall Loss 0.268235 Objective Loss 0.268235 LR 0.001000 Time 0.030653 +2023-10-02 21:07:47,169 - Epoch: [81][ 90/ 1236] Overall Loss 0.266993 Objective Loss 0.266993 LR 0.001000 Time 0.029519 +2023-10-02 21:07:47,378 - Epoch: [81][ 100/ 1236] Overall Loss 0.269550 Objective Loss 0.269550 LR 0.001000 Time 0.028653 +2023-10-02 21:07:47,582 - Epoch: [81][ 110/ 1236] Overall Loss 0.270421 Objective Loss 0.270421 LR 0.001000 Time 0.027904 +2023-10-02 21:07:47,789 - Epoch: [81][ 120/ 1236] Overall Loss 0.268483 Objective Loss 0.268483 LR 0.001000 Time 0.027303 +2023-10-02 21:07:47,994 - Epoch: [81][ 130/ 1236] Overall Loss 0.267771 Objective Loss 0.267771 LR 0.001000 Time 0.026769 +2023-10-02 21:07:48,203 - Epoch: [81][ 140/ 1236] Overall Loss 0.267685 Objective Loss 0.267685 LR 0.001000 Time 0.026344 +2023-10-02 21:07:48,407 - Epoch: [81][ 150/ 1236] Overall Loss 0.268071 Objective Loss 0.268071 LR 0.001000 Time 0.025949 +2023-10-02 21:07:48,615 - Epoch: [81][ 160/ 1236] Overall Loss 0.268898 Objective Loss 0.268898 LR 0.001000 Time 0.025626 +2023-10-02 21:07:48,819 - Epoch: [81][ 170/ 1236] Overall Loss 0.269178 Objective Loss 0.269178 LR 0.001000 Time 0.025317 +2023-10-02 21:07:49,027 - Epoch: [81][ 180/ 1236] Overall Loss 0.269425 Objective Loss 0.269425 LR 0.001000 Time 0.025061 +2023-10-02 21:07:49,231 - Epoch: [81][ 190/ 1236] Overall Loss 0.270088 Objective Loss 0.270088 LR 0.001000 Time 0.024816 +2023-10-02 21:07:49,439 - Epoch: [81][ 200/ 1236] Overall Loss 0.269615 Objective Loss 0.269615 LR 0.001000 Time 0.024614 +2023-10-02 21:07:49,643 - Epoch: [81][ 210/ 1236] Overall Loss 0.269171 Objective Loss 0.269171 LR 0.001000 Time 0.024412 +2023-10-02 21:07:49,851 - Epoch: [81][ 220/ 1236] Overall Loss 0.268230 Objective Loss 0.268230 LR 0.001000 Time 0.024247 +2023-10-02 21:07:50,055 - Epoch: [81][ 230/ 1236] Overall Loss 0.268662 Objective Loss 0.268662 LR 0.001000 Time 0.024079 +2023-10-02 21:07:50,264 - Epoch: [81][ 240/ 1236] Overall Loss 0.267739 Objective Loss 0.267739 LR 0.001000 Time 0.023942 +2023-10-02 21:07:50,467 - Epoch: [81][ 250/ 1236] Overall Loss 0.266624 Objective Loss 0.266624 LR 0.001000 Time 0.023797 +2023-10-02 21:07:50,675 - Epoch: [81][ 260/ 1236] Overall Loss 0.267269 Objective Loss 0.267269 LR 0.001000 Time 0.023680 +2023-10-02 21:07:50,879 - Epoch: [81][ 270/ 1236] Overall Loss 0.265969 Objective Loss 0.265969 LR 0.001000 Time 0.023558 +2023-10-02 21:07:51,087 - Epoch: [81][ 280/ 1236] Overall Loss 0.266216 Objective Loss 0.266216 LR 0.001000 Time 0.023460 +2023-10-02 21:07:51,292 - Epoch: [81][ 290/ 1236] Overall Loss 0.265573 Objective Loss 0.265573 LR 0.001000 Time 0.023354 +2023-10-02 21:07:51,500 - Epoch: [81][ 300/ 1236] Overall Loss 0.264978 Objective Loss 0.264978 LR 0.001000 Time 0.023268 +2023-10-02 21:07:51,703 - Epoch: [81][ 310/ 1236] Overall Loss 0.264392 Objective Loss 0.264392 LR 0.001000 Time 0.023173 +2023-10-02 21:07:51,911 - Epoch: [81][ 320/ 1236] Overall Loss 0.264045 Objective Loss 0.264045 LR 0.001000 Time 0.023099 +2023-10-02 21:07:52,115 - Epoch: [81][ 330/ 1236] Overall Loss 0.264137 Objective Loss 0.264137 LR 0.001000 Time 0.023015 +2023-10-02 21:07:52,323 - Epoch: [81][ 340/ 1236] Overall Loss 0.264040 Objective Loss 0.264040 LR 0.001000 Time 0.022949 +2023-10-02 21:07:52,528 - Epoch: [81][ 350/ 1236] Overall Loss 0.264320 Objective Loss 0.264320 LR 0.001000 Time 0.022877 +2023-10-02 21:07:52,736 - Epoch: [81][ 360/ 1236] Overall Loss 0.264496 Objective Loss 0.264496 LR 0.001000 Time 0.022819 +2023-10-02 21:07:52,941 - Epoch: [81][ 370/ 1236] Overall Loss 0.264707 Objective Loss 0.264707 LR 0.001000 Time 0.022755 +2023-10-02 21:07:53,150 - Epoch: [81][ 380/ 1236] Overall Loss 0.264947 Objective Loss 0.264947 LR 0.001000 Time 0.022706 +2023-10-02 21:07:53,355 - Epoch: [81][ 390/ 1236] Overall Loss 0.263880 Objective Loss 0.263880 LR 0.001000 Time 0.022649 +2023-10-02 21:07:53,564 - Epoch: [81][ 400/ 1236] Overall Loss 0.263597 Objective Loss 0.263597 LR 0.001000 Time 0.022604 +2023-10-02 21:07:53,769 - Epoch: [81][ 410/ 1236] Overall Loss 0.263706 Objective Loss 0.263706 LR 0.001000 Time 0.022552 +2023-10-02 21:07:53,978 - Epoch: [81][ 420/ 1236] Overall Loss 0.263866 Objective Loss 0.263866 LR 0.001000 Time 0.022511 +2023-10-02 21:07:54,183 - Epoch: [81][ 430/ 1236] Overall Loss 0.263808 Objective Loss 0.263808 LR 0.001000 Time 0.022464 +2023-10-02 21:07:54,391 - Epoch: [81][ 440/ 1236] Overall Loss 0.263640 Objective Loss 0.263640 LR 0.001000 Time 0.022427 +2023-10-02 21:07:54,596 - Epoch: [81][ 450/ 1236] Overall Loss 0.263717 Objective Loss 0.263717 LR 0.001000 Time 0.022383 +2023-10-02 21:07:54,805 - Epoch: [81][ 460/ 1236] Overall Loss 0.263703 Objective Loss 0.263703 LR 0.001000 Time 0.022350 +2023-10-02 21:07:55,010 - Epoch: [81][ 470/ 1236] Overall Loss 0.263835 Objective Loss 0.263835 LR 0.001000 Time 0.022310 +2023-10-02 21:07:55,219 - Epoch: [81][ 480/ 1236] Overall Loss 0.264379 Objective Loss 0.264379 LR 0.001000 Time 0.022279 +2023-10-02 21:07:55,425 - Epoch: [81][ 490/ 1236] Overall Loss 0.264618 Objective Loss 0.264618 LR 0.001000 Time 0.022242 +2023-10-02 21:07:55,634 - Epoch: [81][ 500/ 1236] Overall Loss 0.264935 Objective Loss 0.264935 LR 0.001000 Time 0.022215 +2023-10-02 21:07:55,839 - Epoch: [81][ 510/ 1236] Overall Loss 0.265057 Objective Loss 0.265057 LR 0.001000 Time 0.022180 +2023-10-02 21:07:56,048 - Epoch: [81][ 520/ 1236] Overall Loss 0.264946 Objective Loss 0.264946 LR 0.001000 Time 0.022155 +2023-10-02 21:07:56,253 - Epoch: [81][ 530/ 1236] Overall Loss 0.265016 Objective Loss 0.265016 LR 0.001000 Time 0.022123 +2023-10-02 21:07:56,462 - Epoch: [81][ 540/ 1236] Overall Loss 0.265065 Objective Loss 0.265065 LR 0.001000 Time 0.022100 +2023-10-02 21:07:56,667 - Epoch: [81][ 550/ 1236] Overall Loss 0.265037 Objective Loss 0.265037 LR 0.001000 Time 0.022071 +2023-10-02 21:07:56,876 - Epoch: [81][ 560/ 1236] Overall Loss 0.264746 Objective Loss 0.264746 LR 0.001000 Time 0.022049 +2023-10-02 21:07:57,081 - Epoch: [81][ 570/ 1236] Overall Loss 0.264474 Objective Loss 0.264474 LR 0.001000 Time 0.022021 +2023-10-02 21:07:57,290 - Epoch: [81][ 580/ 1236] Overall Loss 0.264454 Objective Loss 0.264454 LR 0.001000 Time 0.022001 +2023-10-02 21:07:57,495 - Epoch: [81][ 590/ 1236] Overall Loss 0.264749 Objective Loss 0.264749 LR 0.001000 Time 0.021976 +2023-10-02 21:07:57,704 - Epoch: [81][ 600/ 1236] Overall Loss 0.265100 Objective Loss 0.265100 LR 0.001000 Time 0.021957 +2023-10-02 21:07:57,909 - Epoch: [81][ 610/ 1236] Overall Loss 0.265266 Objective Loss 0.265266 LR 0.001000 Time 0.021933 +2023-10-02 21:07:58,118 - Epoch: [81][ 620/ 1236] Overall Loss 0.264975 Objective Loss 0.264975 LR 0.001000 Time 0.021916 +2023-10-02 21:07:58,323 - Epoch: [81][ 630/ 1236] Overall Loss 0.265313 Objective Loss 0.265313 LR 0.001000 Time 0.021893 +2023-10-02 21:07:58,532 - Epoch: [81][ 640/ 1236] Overall Loss 0.265407 Objective Loss 0.265407 LR 0.001000 Time 0.021877 +2023-10-02 21:07:58,738 - Epoch: [81][ 650/ 1236] Overall Loss 0.265429 Objective Loss 0.265429 LR 0.001000 Time 0.021856 +2023-10-02 21:07:58,947 - Epoch: [81][ 660/ 1236] Overall Loss 0.265614 Objective Loss 0.265614 LR 0.001000 Time 0.021842 +2023-10-02 21:07:59,152 - Epoch: [81][ 670/ 1236] Overall Loss 0.265578 Objective Loss 0.265578 LR 0.001000 Time 0.021821 +2023-10-02 21:07:59,361 - Epoch: [81][ 680/ 1236] Overall Loss 0.265773 Objective Loss 0.265773 LR 0.001000 Time 0.021807 +2023-10-02 21:07:59,566 - Epoch: [81][ 690/ 1236] Overall Loss 0.265784 Objective Loss 0.265784 LR 0.001000 Time 0.021788 +2023-10-02 21:07:59,775 - Epoch: [81][ 700/ 1236] Overall Loss 0.265764 Objective Loss 0.265764 LR 0.001000 Time 0.021775 +2023-10-02 21:07:59,980 - Epoch: [81][ 710/ 1236] Overall Loss 0.265692 Objective Loss 0.265692 LR 0.001000 Time 0.021757 +2023-10-02 21:08:00,190 - Epoch: [81][ 720/ 1236] Overall Loss 0.265728 Objective Loss 0.265728 LR 0.001000 Time 0.021745 +2023-10-02 21:08:00,395 - Epoch: [81][ 730/ 1236] Overall Loss 0.265869 Objective Loss 0.265869 LR 0.001000 Time 0.021728 +2023-10-02 21:08:00,604 - Epoch: [81][ 740/ 1236] Overall Loss 0.265972 Objective Loss 0.265972 LR 0.001000 Time 0.021716 +2023-10-02 21:08:00,809 - Epoch: [81][ 750/ 1236] Overall Loss 0.266165 Objective Loss 0.266165 LR 0.001000 Time 0.021700 +2023-10-02 21:08:01,018 - Epoch: [81][ 760/ 1236] Overall Loss 0.265957 Objective Loss 0.265957 LR 0.001000 Time 0.021689 +2023-10-02 21:08:01,223 - Epoch: [81][ 770/ 1236] Overall Loss 0.265765 Objective Loss 0.265765 LR 0.001000 Time 0.021673 +2023-10-02 21:08:01,432 - Epoch: [81][ 780/ 1236] Overall Loss 0.265609 Objective Loss 0.265609 LR 0.001000 Time 0.021663 +2023-10-02 21:08:01,638 - Epoch: [81][ 790/ 1236] Overall Loss 0.265128 Objective Loss 0.265128 LR 0.001000 Time 0.021648 +2023-10-02 21:08:01,847 - Epoch: [81][ 800/ 1236] Overall Loss 0.265009 Objective Loss 0.265009 LR 0.001000 Time 0.021639 +2023-10-02 21:08:02,052 - Epoch: [81][ 810/ 1236] Overall Loss 0.264575 Objective Loss 0.264575 LR 0.001000 Time 0.021624 +2023-10-02 21:08:02,261 - Epoch: [81][ 820/ 1236] Overall Loss 0.264300 Objective Loss 0.264300 LR 0.001000 Time 0.021616 +2023-10-02 21:08:02,466 - Epoch: [81][ 830/ 1236] Overall Loss 0.264402 Objective Loss 0.264402 LR 0.001000 Time 0.021602 +2023-10-02 21:08:02,675 - Epoch: [81][ 840/ 1236] Overall Loss 0.264476 Objective Loss 0.264476 LR 0.001000 Time 0.021593 +2023-10-02 21:08:02,881 - Epoch: [81][ 850/ 1236] Overall Loss 0.264194 Objective Loss 0.264194 LR 0.001000 Time 0.021581 +2023-10-02 21:08:03,090 - Epoch: [81][ 860/ 1236] Overall Loss 0.264011 Objective Loss 0.264011 LR 0.001000 Time 0.021572 +2023-10-02 21:08:03,295 - Epoch: [81][ 870/ 1236] Overall Loss 0.263829 Objective Loss 0.263829 LR 0.001000 Time 0.021560 +2023-10-02 21:08:03,504 - Epoch: [81][ 880/ 1236] Overall Loss 0.264010 Objective Loss 0.264010 LR 0.001000 Time 0.021552 +2023-10-02 21:08:03,709 - Epoch: [81][ 890/ 1236] Overall Loss 0.264001 Objective Loss 0.264001 LR 0.001000 Time 0.021540 +2023-10-02 21:08:03,919 - Epoch: [81][ 900/ 1236] Overall Loss 0.264081 Objective Loss 0.264081 LR 0.001000 Time 0.021533 +2023-10-02 21:08:04,124 - Epoch: [81][ 910/ 1236] Overall Loss 0.264318 Objective Loss 0.264318 LR 0.001000 Time 0.021522 +2023-10-02 21:08:04,333 - Epoch: [81][ 920/ 1236] Overall Loss 0.264039 Objective Loss 0.264039 LR 0.001000 Time 0.021514 +2023-10-02 21:08:04,537 - Epoch: [81][ 930/ 1236] Overall Loss 0.264244 Objective Loss 0.264244 LR 0.001000 Time 0.021503 +2023-10-02 21:08:04,745 - Epoch: [81][ 940/ 1236] Overall Loss 0.264022 Objective Loss 0.264022 LR 0.001000 Time 0.021495 +2023-10-02 21:08:04,951 - Epoch: [81][ 950/ 1236] Overall Loss 0.264122 Objective Loss 0.264122 LR 0.001000 Time 0.021484 +2023-10-02 21:08:05,160 - Epoch: [81][ 960/ 1236] Overall Loss 0.264542 Objective Loss 0.264542 LR 0.001000 Time 0.021477 +2023-10-02 21:08:05,365 - Epoch: [81][ 970/ 1236] Overall Loss 0.264922 Objective Loss 0.264922 LR 0.001000 Time 0.021467 +2023-10-02 21:08:05,573 - Epoch: [81][ 980/ 1236] Overall Loss 0.264691 Objective Loss 0.264691 LR 0.001000 Time 0.021460 +2023-10-02 21:08:05,778 - Epoch: [81][ 990/ 1236] Overall Loss 0.264337 Objective Loss 0.264337 LR 0.001000 Time 0.021450 +2023-10-02 21:08:05,987 - Epoch: [81][ 1000/ 1236] Overall Loss 0.264415 Objective Loss 0.264415 LR 0.001000 Time 0.021444 +2023-10-02 21:08:06,192 - Epoch: [81][ 1010/ 1236] Overall Loss 0.264461 Objective Loss 0.264461 LR 0.001000 Time 0.021434 +2023-10-02 21:08:06,400 - Epoch: [81][ 1020/ 1236] Overall Loss 0.264480 Objective Loss 0.264480 LR 0.001000 Time 0.021428 +2023-10-02 21:08:06,605 - Epoch: [81][ 1030/ 1236] Overall Loss 0.264155 Objective Loss 0.264155 LR 0.001000 Time 0.021418 +2023-10-02 21:08:06,814 - Epoch: [81][ 1040/ 1236] Overall Loss 0.264031 Objective Loss 0.264031 LR 0.001000 Time 0.021413 +2023-10-02 21:08:07,019 - Epoch: [81][ 1050/ 1236] Overall Loss 0.263681 Objective Loss 0.263681 LR 0.001000 Time 0.021404 +2023-10-02 21:08:07,227 - Epoch: [81][ 1060/ 1236] Overall Loss 0.263510 Objective Loss 0.263510 LR 0.001000 Time 0.021399 +2023-10-02 21:08:07,432 - Epoch: [81][ 1070/ 1236] Overall Loss 0.263561 Objective Loss 0.263561 LR 0.001000 Time 0.021390 +2023-10-02 21:08:07,641 - Epoch: [81][ 1080/ 1236] Overall Loss 0.263503 Objective Loss 0.263503 LR 0.001000 Time 0.021385 +2023-10-02 21:08:07,846 - Epoch: [81][ 1090/ 1236] Overall Loss 0.263292 Objective Loss 0.263292 LR 0.001000 Time 0.021376 +2023-10-02 21:08:08,055 - Epoch: [81][ 1100/ 1236] Overall Loss 0.263155 Objective Loss 0.263155 LR 0.001000 Time 0.021371 +2023-10-02 21:08:08,260 - Epoch: [81][ 1110/ 1236] Overall Loss 0.263028 Objective Loss 0.263028 LR 0.001000 Time 0.021363 +2023-10-02 21:08:08,468 - Epoch: [81][ 1120/ 1236] Overall Loss 0.263208 Objective Loss 0.263208 LR 0.001000 Time 0.021358 +2023-10-02 21:08:08,673 - Epoch: [81][ 1130/ 1236] Overall Loss 0.263546 Objective Loss 0.263546 LR 0.001000 Time 0.021350 +2023-10-02 21:08:08,882 - Epoch: [81][ 1140/ 1236] Overall Loss 0.263630 Objective Loss 0.263630 LR 0.001000 Time 0.021346 +2023-10-02 21:08:09,086 - Epoch: [81][ 1150/ 1236] Overall Loss 0.263670 Objective Loss 0.263670 LR 0.001000 Time 0.021338 +2023-10-02 21:08:09,294 - Epoch: [81][ 1160/ 1236] Overall Loss 0.263554 Objective Loss 0.263554 LR 0.001000 Time 0.021332 +2023-10-02 21:08:09,500 - Epoch: [81][ 1170/ 1236] Overall Loss 0.263604 Objective Loss 0.263604 LR 0.001000 Time 0.021325 +2023-10-02 21:08:09,709 - Epoch: [81][ 1180/ 1236] Overall Loss 0.263380 Objective Loss 0.263380 LR 0.001000 Time 0.021321 +2023-10-02 21:08:09,914 - Epoch: [81][ 1190/ 1236] Overall Loss 0.263359 Objective Loss 0.263359 LR 0.001000 Time 0.021314 +2023-10-02 21:08:10,123 - Epoch: [81][ 1200/ 1236] Overall Loss 0.263516 Objective Loss 0.263516 LR 0.001000 Time 0.021310 +2023-10-02 21:08:10,327 - Epoch: [81][ 1210/ 1236] Overall Loss 0.263706 Objective Loss 0.263706 LR 0.001000 Time 0.021303 +2023-10-02 21:08:10,536 - Epoch: [81][ 1220/ 1236] Overall Loss 0.263898 Objective Loss 0.263898 LR 0.001000 Time 0.021299 +2023-10-02 21:08:10,794 - Epoch: [81][ 1230/ 1236] Overall Loss 0.263582 Objective Loss 0.263582 LR 0.001000 Time 0.021336 +2023-10-02 21:08:10,916 - Epoch: [81][ 1236/ 1236] Overall Loss 0.263534 Objective Loss 0.263534 Top1 85.743381 Top5 98.167006 LR 0.001000 Time 0.021330 +2023-10-02 21:08:11,062 - --- validate (epoch=81)----------- +2023-10-02 21:08:11,063 - 29943 samples (256 per mini-batch) +2023-10-02 21:08:11,558 - Epoch: [81][ 10/ 117] Loss 0.312743 Top1 85.156250 Top5 98.320312 +2023-10-02 21:08:11,709 - Epoch: [81][ 20/ 117] Loss 0.327623 Top1 84.472656 Top5 98.300781 +2023-10-02 21:08:11,858 - Epoch: [81][ 30/ 117] Loss 0.336392 Top1 84.101562 Top5 98.177083 +2023-10-02 21:08:12,008 - Epoch: [81][ 40/ 117] Loss 0.348134 Top1 83.583984 Top5 98.007812 +2023-10-02 21:08:12,157 - Epoch: [81][ 50/ 117] Loss 0.346791 Top1 83.437500 Top5 97.984375 +2023-10-02 21:08:12,307 - Epoch: [81][ 60/ 117] Loss 0.342355 Top1 83.515625 Top5 98.072917 +2023-10-02 21:08:12,457 - Epoch: [81][ 70/ 117] Loss 0.343258 Top1 83.493304 Top5 98.080357 +2023-10-02 21:08:12,608 - Epoch: [81][ 80/ 117] Loss 0.345107 Top1 83.310547 Top5 98.076172 +2023-10-02 21:08:12,757 - Epoch: [81][ 90/ 117] Loss 0.346861 Top1 83.229167 Top5 98.059896 +2023-10-02 21:08:12,907 - Epoch: [81][ 100/ 117] Loss 0.346241 Top1 83.238281 Top5 98.082031 +2023-10-02 21:08:13,064 - Epoch: [81][ 110/ 117] Loss 0.347388 Top1 83.242188 Top5 98.082386 +2023-10-02 21:08:13,152 - Epoch: [81][ 117/ 117] Loss 0.347751 Top1 83.284908 Top5 98.083024 +2023-10-02 21:08:13,297 - ==> Top1: 83.285 Top5: 98.083 Loss: 0.348 + +2023-10-02 21:08:13,298 - ==> Confusion: +[[ 950 1 2 0 5 2 0 0 7 47 1 0 1 2 5 0 5 3 1 0 18] + [ 2 1041 1 1 7 26 1 26 5 2 1 0 0 0 0 4 0 0 9 2 3] + [ 1 1 964 9 2 0 22 9 1 2 2 0 5 3 0 5 1 2 13 3 11] + [ 1 2 12 974 0 5 0 3 4 1 9 0 2 3 29 1 2 4 19 3 15] + [ 36 11 3 0 957 4 0 0 0 9 1 0 1 3 7 5 6 1 2 0 4] + [ 2 53 1 5 1 953 1 36 2 7 2 1 5 9 5 3 2 3 3 7 15] + [ 0 5 31 0 0 1 1115 3 0 0 5 0 1 0 1 7 0 1 2 12 7] + [ 2 16 20 0 7 18 2 1070 0 0 4 1 2 1 4 0 2 0 55 8 6] + [ 17 10 1 1 3 3 3 1 967 45 9 2 0 7 12 0 2 2 3 0 1] + [ 139 0 4 1 8 1 2 1 28 901 1 1 0 10 7 1 0 2 0 2 10] + [ 3 1 10 10 2 1 3 6 19 0 954 1 0 14 5 0 0 5 9 1 9] + [ 2 1 5 0 1 12 1 10 0 2 0 929 31 4 0 3 1 18 1 8 6] + [ 0 2 4 6 0 0 1 1 2 1 4 24 965 4 3 4 1 18 8 6 14] + [ 1 0 3 0 5 13 2 3 18 11 10 7 1 1009 10 2 2 1 1 4 16] + [ 12 0 6 17 4 0 0 0 25 5 2 0 3 3 990 0 1 4 20 0 9] + [ 0 0 3 2 6 0 2 0 0 1 0 6 12 2 0 1060 12 12 3 4 9] + [ 3 22 1 0 7 4 0 2 2 0 0 7 0 2 4 11 1079 0 1 5 11] + [ 0 0 0 6 1 0 1 0 0 0 0 4 16 1 6 4 0 987 2 4 6] + [ 3 3 4 12 0 0 0 19 2 0 2 0 3 0 7 0 0 0 1006 0 7] + [ 0 2 2 1 1 9 12 8 0 0 2 10 3 0 0 3 5 3 1 1083 7] + [ 177 228 157 118 85 117 41 124 94 80 165 86 323 284 198 71 100 74 219 180 4984]] + +2023-10-02 21:08:13,299 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:08:13,299 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:08:13,305 - + +2023-10-02 21:08:13,305 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:08:14,316 - Epoch: [82][ 10/ 1236] Overall Loss 0.268080 Objective Loss 0.268080 LR 0.001000 Time 0.100977 +2023-10-02 21:08:14,521 - Epoch: [82][ 20/ 1236] Overall Loss 0.264305 Objective Loss 0.264305 LR 0.001000 Time 0.060716 +2023-10-02 21:08:14,725 - Epoch: [82][ 30/ 1236] Overall Loss 0.260978 Objective Loss 0.260978 LR 0.001000 Time 0.047284 +2023-10-02 21:08:14,930 - Epoch: [82][ 40/ 1236] Overall Loss 0.257098 Objective Loss 0.257098 LR 0.001000 Time 0.040574 +2023-10-02 21:08:15,134 - Epoch: [82][ 50/ 1236] Overall Loss 0.258769 Objective Loss 0.258769 LR 0.001000 Time 0.036539 +2023-10-02 21:08:15,340 - Epoch: [82][ 60/ 1236] Overall Loss 0.254702 Objective Loss 0.254702 LR 0.001000 Time 0.033880 +2023-10-02 21:08:15,543 - Epoch: [82][ 70/ 1236] Overall Loss 0.253493 Objective Loss 0.253493 LR 0.001000 Time 0.031937 +2023-10-02 21:08:15,750 - Epoch: [82][ 80/ 1236] Overall Loss 0.254119 Objective Loss 0.254119 LR 0.001000 Time 0.030519 +2023-10-02 21:08:15,953 - Epoch: [82][ 90/ 1236] Overall Loss 0.258222 Objective Loss 0.258222 LR 0.001000 Time 0.029382 +2023-10-02 21:08:16,159 - Epoch: [82][ 100/ 1236] Overall Loss 0.262329 Objective Loss 0.262329 LR 0.001000 Time 0.028503 +2023-10-02 21:08:16,362 - Epoch: [82][ 110/ 1236] Overall Loss 0.261875 Objective Loss 0.261875 LR 0.001000 Time 0.027755 +2023-10-02 21:08:16,568 - Epoch: [82][ 120/ 1236] Overall Loss 0.263420 Objective Loss 0.263420 LR 0.001000 Time 0.027158 +2023-10-02 21:08:16,771 - Epoch: [82][ 130/ 1236] Overall Loss 0.263523 Objective Loss 0.263523 LR 0.001000 Time 0.026628 +2023-10-02 21:08:16,977 - Epoch: [82][ 140/ 1236] Overall Loss 0.262906 Objective Loss 0.262906 LR 0.001000 Time 0.026198 +2023-10-02 21:08:17,181 - Epoch: [82][ 150/ 1236] Overall Loss 0.262974 Objective Loss 0.262974 LR 0.001000 Time 0.025805 +2023-10-02 21:08:17,386 - Epoch: [82][ 160/ 1236] Overall Loss 0.261104 Objective Loss 0.261104 LR 0.001000 Time 0.025474 +2023-10-02 21:08:17,589 - Epoch: [82][ 170/ 1236] Overall Loss 0.261064 Objective Loss 0.261064 LR 0.001000 Time 0.025169 +2023-10-02 21:08:17,794 - Epoch: [82][ 180/ 1236] Overall Loss 0.261460 Objective Loss 0.261460 LR 0.001000 Time 0.024908 +2023-10-02 21:08:17,999 - Epoch: [82][ 190/ 1236] Overall Loss 0.259872 Objective Loss 0.259872 LR 0.001000 Time 0.024665 +2023-10-02 21:08:18,205 - Epoch: [82][ 200/ 1236] Overall Loss 0.259596 Objective Loss 0.259596 LR 0.001000 Time 0.024462 +2023-10-02 21:08:18,408 - Epoch: [82][ 210/ 1236] Overall Loss 0.261600 Objective Loss 0.261600 LR 0.001000 Time 0.024264 +2023-10-02 21:08:18,613 - Epoch: [82][ 220/ 1236] Overall Loss 0.260980 Objective Loss 0.260980 LR 0.001000 Time 0.024088 +2023-10-02 21:08:18,817 - Epoch: [82][ 230/ 1236] Overall Loss 0.259791 Objective Loss 0.259791 LR 0.001000 Time 0.023924 +2023-10-02 21:08:19,022 - Epoch: [82][ 240/ 1236] Overall Loss 0.260623 Objective Loss 0.260623 LR 0.001000 Time 0.023780 +2023-10-02 21:08:19,227 - Epoch: [82][ 250/ 1236] Overall Loss 0.260568 Objective Loss 0.260568 LR 0.001000 Time 0.023641 +2023-10-02 21:08:19,432 - Epoch: [82][ 260/ 1236] Overall Loss 0.260300 Objective Loss 0.260300 LR 0.001000 Time 0.023521 +2023-10-02 21:08:19,637 - Epoch: [82][ 270/ 1236] Overall Loss 0.260883 Objective Loss 0.260883 LR 0.001000 Time 0.023402 +2023-10-02 21:08:19,843 - Epoch: [82][ 280/ 1236] Overall Loss 0.259878 Objective Loss 0.259878 LR 0.001000 Time 0.023301 +2023-10-02 21:08:20,045 - Epoch: [82][ 290/ 1236] Overall Loss 0.260432 Objective Loss 0.260432 LR 0.001000 Time 0.023195 +2023-10-02 21:08:20,251 - Epoch: [82][ 300/ 1236] Overall Loss 0.260759 Objective Loss 0.260759 LR 0.001000 Time 0.023105 +2023-10-02 21:08:20,455 - Epoch: [82][ 310/ 1236] Overall Loss 0.261757 Objective Loss 0.261757 LR 0.001000 Time 0.023014 +2023-10-02 21:08:20,660 - Epoch: [82][ 320/ 1236] Overall Loss 0.261173 Objective Loss 0.261173 LR 0.001000 Time 0.022935 +2023-10-02 21:08:20,865 - Epoch: [82][ 330/ 1236] Overall Loss 0.260877 Objective Loss 0.260877 LR 0.001000 Time 0.022855 +2023-10-02 21:08:21,070 - Epoch: [82][ 340/ 1236] Overall Loss 0.259905 Objective Loss 0.259905 LR 0.001000 Time 0.022786 +2023-10-02 21:08:21,275 - Epoch: [82][ 350/ 1236] Overall Loss 0.259851 Objective Loss 0.259851 LR 0.001000 Time 0.022715 +2023-10-02 21:08:21,482 - Epoch: [82][ 360/ 1236] Overall Loss 0.260437 Objective Loss 0.260437 LR 0.001000 Time 0.022658 +2023-10-02 21:08:21,685 - Epoch: [82][ 370/ 1236] Overall Loss 0.260572 Objective Loss 0.260572 LR 0.001000 Time 0.022595 +2023-10-02 21:08:21,890 - Epoch: [82][ 380/ 1236] Overall Loss 0.260803 Objective Loss 0.260803 LR 0.001000 Time 0.022539 +2023-10-02 21:08:22,095 - Epoch: [82][ 390/ 1236] Overall Loss 0.259849 Objective Loss 0.259849 LR 0.001000 Time 0.022486 +2023-10-02 21:08:22,302 - Epoch: [82][ 400/ 1236] Overall Loss 0.259739 Objective Loss 0.259739 LR 0.001000 Time 0.022439 +2023-10-02 21:08:22,505 - Epoch: [82][ 410/ 1236] Overall Loss 0.259548 Objective Loss 0.259548 LR 0.001000 Time 0.022388 +2023-10-02 21:08:22,710 - Epoch: [82][ 420/ 1236] Overall Loss 0.260035 Objective Loss 0.260035 LR 0.001000 Time 0.022343 +2023-10-02 21:08:22,915 - Epoch: [82][ 430/ 1236] Overall Loss 0.260394 Objective Loss 0.260394 LR 0.001000 Time 0.022298 +2023-10-02 21:08:23,120 - Epoch: [82][ 440/ 1236] Overall Loss 0.260918 Objective Loss 0.260918 LR 0.001000 Time 0.022257 +2023-10-02 21:08:23,325 - Epoch: [82][ 450/ 1236] Overall Loss 0.261243 Objective Loss 0.261243 LR 0.001000 Time 0.022217 +2023-10-02 21:08:23,530 - Epoch: [82][ 460/ 1236] Overall Loss 0.261728 Objective Loss 0.261728 LR 0.001000 Time 0.022179 +2023-10-02 21:08:23,735 - Epoch: [82][ 470/ 1236] Overall Loss 0.262143 Objective Loss 0.262143 LR 0.001000 Time 0.022142 +2023-10-02 21:08:23,941 - Epoch: [82][ 480/ 1236] Overall Loss 0.262143 Objective Loss 0.262143 LR 0.001000 Time 0.022110 +2023-10-02 21:08:24,145 - Epoch: [82][ 490/ 1236] Overall Loss 0.262299 Objective Loss 0.262299 LR 0.001000 Time 0.022074 +2023-10-02 21:08:24,351 - Epoch: [82][ 500/ 1236] Overall Loss 0.262712 Objective Loss 0.262712 LR 0.001000 Time 0.022044 +2023-10-02 21:08:24,555 - Epoch: [82][ 510/ 1236] Overall Loss 0.262864 Objective Loss 0.262864 LR 0.001000 Time 0.022011 +2023-10-02 21:08:24,761 - Epoch: [82][ 520/ 1236] Overall Loss 0.263180 Objective Loss 0.263180 LR 0.001000 Time 0.021984 +2023-10-02 21:08:24,965 - Epoch: [82][ 530/ 1236] Overall Loss 0.263235 Objective Loss 0.263235 LR 0.001000 Time 0.021953 +2023-10-02 21:08:25,171 - Epoch: [82][ 540/ 1236] Overall Loss 0.263875 Objective Loss 0.263875 LR 0.001000 Time 0.021928 +2023-10-02 21:08:25,375 - Epoch: [82][ 550/ 1236] Overall Loss 0.263810 Objective Loss 0.263810 LR 0.001000 Time 0.021899 +2023-10-02 21:08:25,581 - Epoch: [82][ 560/ 1236] Overall Loss 0.263964 Objective Loss 0.263964 LR 0.001000 Time 0.021876 +2023-10-02 21:08:25,785 - Epoch: [82][ 570/ 1236] Overall Loss 0.264455 Objective Loss 0.264455 LR 0.001000 Time 0.021849 +2023-10-02 21:08:25,991 - Epoch: [82][ 580/ 1236] Overall Loss 0.264873 Objective Loss 0.264873 LR 0.001000 Time 0.021828 +2023-10-02 21:08:26,195 - Epoch: [82][ 590/ 1236] Overall Loss 0.264574 Objective Loss 0.264574 LR 0.001000 Time 0.021802 +2023-10-02 21:08:26,401 - Epoch: [82][ 600/ 1236] Overall Loss 0.264232 Objective Loss 0.264232 LR 0.001000 Time 0.021782 +2023-10-02 21:08:26,604 - Epoch: [82][ 610/ 1236] Overall Loss 0.264282 Objective Loss 0.264282 LR 0.001000 Time 0.021758 +2023-10-02 21:08:26,811 - Epoch: [82][ 620/ 1236] Overall Loss 0.264116 Objective Loss 0.264116 LR 0.001000 Time 0.021740 +2023-10-02 21:08:27,015 - Epoch: [82][ 630/ 1236] Overall Loss 0.264451 Objective Loss 0.264451 LR 0.001000 Time 0.021717 +2023-10-02 21:08:27,221 - Epoch: [82][ 640/ 1236] Overall Loss 0.264435 Objective Loss 0.264435 LR 0.001000 Time 0.021700 +2023-10-02 21:08:27,425 - Epoch: [82][ 650/ 1236] Overall Loss 0.264191 Objective Loss 0.264191 LR 0.001000 Time 0.021679 +2023-10-02 21:08:27,631 - Epoch: [82][ 660/ 1236] Overall Loss 0.264551 Objective Loss 0.264551 LR 0.001000 Time 0.021663 +2023-10-02 21:08:27,834 - Epoch: [82][ 670/ 1236] Overall Loss 0.264300 Objective Loss 0.264300 LR 0.001000 Time 0.021643 +2023-10-02 21:08:28,041 - Epoch: [82][ 680/ 1236] Overall Loss 0.264175 Objective Loss 0.264175 LR 0.001000 Time 0.021628 +2023-10-02 21:08:28,244 - Epoch: [82][ 690/ 1236] Overall Loss 0.264235 Objective Loss 0.264235 LR 0.001000 Time 0.021609 +2023-10-02 21:08:28,451 - Epoch: [82][ 700/ 1236] Overall Loss 0.264475 Objective Loss 0.264475 LR 0.001000 Time 0.021595 +2023-10-02 21:08:28,654 - Epoch: [82][ 710/ 1236] Overall Loss 0.263994 Objective Loss 0.263994 LR 0.001000 Time 0.021577 +2023-10-02 21:08:28,861 - Epoch: [82][ 720/ 1236] Overall Loss 0.264142 Objective Loss 0.264142 LR 0.001000 Time 0.021563 +2023-10-02 21:08:29,064 - Epoch: [82][ 730/ 1236] Overall Loss 0.264384 Objective Loss 0.264384 LR 0.001000 Time 0.021546 +2023-10-02 21:08:29,270 - Epoch: [82][ 740/ 1236] Overall Loss 0.264018 Objective Loss 0.264018 LR 0.001000 Time 0.021533 +2023-10-02 21:08:29,474 - Epoch: [82][ 750/ 1236] Overall Loss 0.263801 Objective Loss 0.263801 LR 0.001000 Time 0.021517 +2023-10-02 21:08:29,680 - Epoch: [82][ 760/ 1236] Overall Loss 0.263858 Objective Loss 0.263858 LR 0.001000 Time 0.021505 +2023-10-02 21:08:29,884 - Epoch: [82][ 770/ 1236] Overall Loss 0.263743 Objective Loss 0.263743 LR 0.001000 Time 0.021490 +2023-10-02 21:08:30,090 - Epoch: [82][ 780/ 1236] Overall Loss 0.263521 Objective Loss 0.263521 LR 0.001000 Time 0.021479 +2023-10-02 21:08:30,294 - Epoch: [82][ 790/ 1236] Overall Loss 0.263730 Objective Loss 0.263730 LR 0.001000 Time 0.021464 +2023-10-02 21:08:30,500 - Epoch: [82][ 800/ 1236] Overall Loss 0.263760 Objective Loss 0.263760 LR 0.001000 Time 0.021454 +2023-10-02 21:08:30,704 - Epoch: [82][ 810/ 1236] Overall Loss 0.263843 Objective Loss 0.263843 LR 0.001000 Time 0.021440 +2023-10-02 21:08:30,910 - Epoch: [82][ 820/ 1236] Overall Loss 0.263536 Objective Loss 0.263536 LR 0.001000 Time 0.021430 +2023-10-02 21:08:31,114 - Epoch: [82][ 830/ 1236] Overall Loss 0.263713 Objective Loss 0.263713 LR 0.001000 Time 0.021416 +2023-10-02 21:08:31,320 - Epoch: [82][ 840/ 1236] Overall Loss 0.263554 Objective Loss 0.263554 LR 0.001000 Time 0.021407 +2023-10-02 21:08:31,524 - Epoch: [82][ 850/ 1236] Overall Loss 0.263583 Objective Loss 0.263583 LR 0.001000 Time 0.021394 +2023-10-02 21:08:31,730 - Epoch: [82][ 860/ 1236] Overall Loss 0.263801 Objective Loss 0.263801 LR 0.001000 Time 0.021385 +2023-10-02 21:08:31,934 - Epoch: [82][ 870/ 1236] Overall Loss 0.264063 Objective Loss 0.264063 LR 0.001000 Time 0.021373 +2023-10-02 21:08:32,140 - Epoch: [82][ 880/ 1236] Overall Loss 0.263814 Objective Loss 0.263814 LR 0.001000 Time 0.021364 +2023-10-02 21:08:32,344 - Epoch: [82][ 890/ 1236] Overall Loss 0.263729 Objective Loss 0.263729 LR 0.001000 Time 0.021352 +2023-10-02 21:08:32,550 - Epoch: [82][ 900/ 1236] Overall Loss 0.263500 Objective Loss 0.263500 LR 0.001000 Time 0.021344 +2023-10-02 21:08:32,753 - Epoch: [82][ 910/ 1236] Overall Loss 0.263428 Objective Loss 0.263428 LR 0.001000 Time 0.021333 +2023-10-02 21:08:32,960 - Epoch: [82][ 920/ 1236] Overall Loss 0.263410 Objective Loss 0.263410 LR 0.001000 Time 0.021325 +2023-10-02 21:08:33,163 - Epoch: [82][ 930/ 1236] Overall Loss 0.262960 Objective Loss 0.262960 LR 0.001000 Time 0.021314 +2023-10-02 21:08:33,370 - Epoch: [82][ 940/ 1236] Overall Loss 0.262648 Objective Loss 0.262648 LR 0.001000 Time 0.021307 +2023-10-02 21:08:33,573 - Epoch: [82][ 950/ 1236] Overall Loss 0.262847 Objective Loss 0.262847 LR 0.001000 Time 0.021297 +2023-10-02 21:08:33,780 - Epoch: [82][ 960/ 1236] Overall Loss 0.262885 Objective Loss 0.262885 LR 0.001000 Time 0.021290 +2023-10-02 21:08:33,983 - Epoch: [82][ 970/ 1236] Overall Loss 0.263104 Objective Loss 0.263104 LR 0.001000 Time 0.021279 +2023-10-02 21:08:34,190 - Epoch: [82][ 980/ 1236] Overall Loss 0.263060 Objective Loss 0.263060 LR 0.001000 Time 0.021273 +2023-10-02 21:08:34,393 - Epoch: [82][ 990/ 1236] Overall Loss 0.263064 Objective Loss 0.263064 LR 0.001000 Time 0.021263 +2023-10-02 21:08:34,600 - Epoch: [82][ 1000/ 1236] Overall Loss 0.263030 Objective Loss 0.263030 LR 0.001000 Time 0.021257 +2023-10-02 21:08:34,803 - Epoch: [82][ 1010/ 1236] Overall Loss 0.263032 Objective Loss 0.263032 LR 0.001000 Time 0.021247 +2023-10-02 21:08:35,010 - Epoch: [82][ 1020/ 1236] Overall Loss 0.263061 Objective Loss 0.263061 LR 0.001000 Time 0.021242 +2023-10-02 21:08:35,214 - Epoch: [82][ 1030/ 1236] Overall Loss 0.262876 Objective Loss 0.262876 LR 0.001000 Time 0.021233 +2023-10-02 21:08:35,420 - Epoch: [82][ 1040/ 1236] Overall Loss 0.262634 Objective Loss 0.262634 LR 0.001000 Time 0.021227 +2023-10-02 21:08:35,624 - Epoch: [82][ 1050/ 1236] Overall Loss 0.262600 Objective Loss 0.262600 LR 0.001000 Time 0.021218 +2023-10-02 21:08:35,830 - Epoch: [82][ 1060/ 1236] Overall Loss 0.262620 Objective Loss 0.262620 LR 0.001000 Time 0.021212 +2023-10-02 21:08:36,034 - Epoch: [82][ 1070/ 1236] Overall Loss 0.262332 Objective Loss 0.262332 LR 0.001000 Time 0.021204 +2023-10-02 21:08:36,240 - Epoch: [82][ 1080/ 1236] Overall Loss 0.262478 Objective Loss 0.262478 LR 0.001000 Time 0.021199 +2023-10-02 21:08:36,444 - Epoch: [82][ 1090/ 1236] Overall Loss 0.262486 Objective Loss 0.262486 LR 0.001000 Time 0.021191 +2023-10-02 21:08:36,651 - Epoch: [82][ 1100/ 1236] Overall Loss 0.262618 Objective Loss 0.262618 LR 0.001000 Time 0.021186 +2023-10-02 21:08:36,854 - Epoch: [82][ 1110/ 1236] Overall Loss 0.262827 Objective Loss 0.262827 LR 0.001000 Time 0.021178 +2023-10-02 21:08:37,061 - Epoch: [82][ 1120/ 1236] Overall Loss 0.263290 Objective Loss 0.263290 LR 0.001000 Time 0.021173 +2023-10-02 21:08:37,264 - Epoch: [82][ 1130/ 1236] Overall Loss 0.263700 Objective Loss 0.263700 LR 0.001000 Time 0.021166 +2023-10-02 21:08:37,469 - Epoch: [82][ 1140/ 1236] Overall Loss 0.264036 Objective Loss 0.264036 LR 0.001000 Time 0.021160 +2023-10-02 21:08:37,674 - Epoch: [82][ 1150/ 1236] Overall Loss 0.264018 Objective Loss 0.264018 LR 0.001000 Time 0.021153 +2023-10-02 21:08:37,880 - Epoch: [82][ 1160/ 1236] Overall Loss 0.264249 Objective Loss 0.264249 LR 0.001000 Time 0.021147 +2023-10-02 21:08:38,084 - Epoch: [82][ 1170/ 1236] Overall Loss 0.264400 Objective Loss 0.264400 LR 0.001000 Time 0.021141 +2023-10-02 21:08:38,291 - Epoch: [82][ 1180/ 1236] Overall Loss 0.264549 Objective Loss 0.264549 LR 0.001000 Time 0.021137 +2023-10-02 21:08:38,495 - Epoch: [82][ 1190/ 1236] Overall Loss 0.264528 Objective Loss 0.264528 LR 0.001000 Time 0.021130 +2023-10-02 21:08:38,700 - Epoch: [82][ 1200/ 1236] Overall Loss 0.264785 Objective Loss 0.264785 LR 0.001000 Time 0.021125 +2023-10-02 21:08:38,904 - Epoch: [82][ 1210/ 1236] Overall Loss 0.264888 Objective Loss 0.264888 LR 0.001000 Time 0.021119 +2023-10-02 21:08:39,110 - Epoch: [82][ 1220/ 1236] Overall Loss 0.264845 Objective Loss 0.264845 LR 0.001000 Time 0.021114 +2023-10-02 21:08:39,367 - Epoch: [82][ 1230/ 1236] Overall Loss 0.264812 Objective Loss 0.264812 LR 0.001000 Time 0.021151 +2023-10-02 21:08:39,487 - Epoch: [82][ 1236/ 1236] Overall Loss 0.264832 Objective Loss 0.264832 Top1 84.928717 Top5 97.759674 LR 0.001000 Time 0.021146 +2023-10-02 21:08:39,619 - --- validate (epoch=82)----------- +2023-10-02 21:08:39,620 - 29943 samples (256 per mini-batch) +2023-10-02 21:08:40,110 - Epoch: [82][ 10/ 117] Loss 0.305027 Top1 82.929688 Top5 98.125000 +2023-10-02 21:08:40,263 - Epoch: [82][ 20/ 117] Loss 0.350636 Top1 82.285156 Top5 98.007812 +2023-10-02 21:08:40,416 - Epoch: [82][ 30/ 117] Loss 0.345559 Top1 82.799479 Top5 97.786458 +2023-10-02 21:08:40,568 - Epoch: [82][ 40/ 117] Loss 0.346840 Top1 82.861328 Top5 97.939453 +2023-10-02 21:08:40,722 - Epoch: [82][ 50/ 117] Loss 0.344868 Top1 82.812500 Top5 97.929688 +2023-10-02 21:08:40,874 - Epoch: [82][ 60/ 117] Loss 0.348057 Top1 82.936198 Top5 97.884115 +2023-10-02 21:08:41,027 - Epoch: [82][ 70/ 117] Loss 0.347083 Top1 82.829241 Top5 97.918527 +2023-10-02 21:08:41,177 - Epoch: [82][ 80/ 117] Loss 0.347464 Top1 82.963867 Top5 97.949219 +2023-10-02 21:08:41,328 - Epoch: [82][ 90/ 117] Loss 0.344362 Top1 83.055556 Top5 97.968750 +2023-10-02 21:08:41,479 - Epoch: [82][ 100/ 117] Loss 0.339757 Top1 83.218750 Top5 98.054688 +2023-10-02 21:08:41,636 - Epoch: [82][ 110/ 117] Loss 0.338247 Top1 83.203125 Top5 98.039773 +2023-10-02 21:08:41,724 - Epoch: [82][ 117/ 117] Loss 0.336980 Top1 83.254851 Top5 98.042948 +2023-10-02 21:08:41,868 - ==> Top1: 83.255 Top5: 98.043 Loss: 0.337 + +2023-10-02 21:08:41,869 - ==> Confusion: +[[ 942 2 1 1 7 5 0 1 6 47 2 0 2 1 6 3 3 1 2 0 18] + [ 0 1061 1 2 4 22 0 14 2 1 0 1 1 0 3 4 3 0 4 1 7] + [ 3 1 960 20 2 0 12 8 0 1 1 0 12 1 1 3 0 2 12 5 12] + [ 2 5 11 973 0 5 0 2 10 0 5 0 8 4 24 1 0 6 20 0 13] + [ 27 9 2 0 971 3 0 0 0 5 1 1 3 3 5 4 12 0 1 1 2] + [ 2 53 0 2 9 953 1 31 0 3 1 10 5 7 5 0 9 1 6 9 9] + [ 0 7 38 1 1 1 1110 4 0 0 2 1 0 0 1 7 0 0 4 5 9] + [ 4 27 22 0 5 19 6 1035 1 3 5 1 6 4 6 0 2 3 54 7 8] + [ 22 5 0 0 1 1 0 1 961 32 12 1 5 16 18 0 4 2 2 1 5] + [ 130 0 1 0 8 2 1 0 38 887 2 1 0 19 15 3 2 1 0 0 9] + [ 1 3 8 7 0 5 2 7 10 1 956 0 4 14 2 0 2 4 10 1 16] + [ 1 1 2 0 3 8 0 4 0 1 0 945 29 4 0 4 8 14 0 7 4] + [ 0 0 2 4 1 3 1 1 2 1 2 48 953 1 2 7 5 14 5 5 11] + [ 0 0 2 0 5 13 1 0 11 9 10 13 3 1015 8 2 4 3 0 5 15] + [ 7 0 3 21 9 1 0 0 18 1 5 0 3 3 996 0 1 3 17 0 13] + [ 0 0 0 4 3 0 0 0 1 1 2 3 10 0 1 1061 20 13 1 6 8] + [ 1 18 2 0 4 5 0 0 2 1 0 3 2 1 2 9 1093 0 3 6 9] + [ 1 2 0 3 0 0 3 0 0 1 0 5 22 2 7 6 0 985 0 0 1] + [ 1 6 3 19 2 0 0 13 2 0 3 1 2 0 10 1 1 0 991 0 13] + [ 0 2 6 1 1 4 9 9 0 0 0 11 8 2 1 1 10 1 5 1073 8] + [ 162 254 134 107 105 130 52 98 84 46 141 114 386 227 150 61 160 81 219 186 5008]] + +2023-10-02 21:08:41,870 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:08:41,870 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:08:41,876 - + +2023-10-02 21:08:41,876 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:08:42,999 - Epoch: [83][ 10/ 1236] Overall Loss 0.276382 Objective Loss 0.276382 LR 0.001000 Time 0.112281 +2023-10-02 21:08:43,205 - Epoch: [83][ 20/ 1236] Overall Loss 0.273376 Objective Loss 0.273376 LR 0.001000 Time 0.066404 +2023-10-02 21:08:43,410 - Epoch: [83][ 30/ 1236] Overall Loss 0.259633 Objective Loss 0.259633 LR 0.001000 Time 0.051089 +2023-10-02 21:08:43,615 - Epoch: [83][ 40/ 1236] Overall Loss 0.265038 Objective Loss 0.265038 LR 0.001000 Time 0.043436 +2023-10-02 21:08:43,820 - Epoch: [83][ 50/ 1236] Overall Loss 0.264643 Objective Loss 0.264643 LR 0.001000 Time 0.038837 +2023-10-02 21:08:44,025 - Epoch: [83][ 60/ 1236] Overall Loss 0.266255 Objective Loss 0.266255 LR 0.001000 Time 0.035777 +2023-10-02 21:08:44,229 - Epoch: [83][ 70/ 1236] Overall Loss 0.260395 Objective Loss 0.260395 LR 0.001000 Time 0.033584 +2023-10-02 21:08:44,436 - Epoch: [83][ 80/ 1236] Overall Loss 0.261403 Objective Loss 0.261403 LR 0.001000 Time 0.031964 +2023-10-02 21:08:44,639 - Epoch: [83][ 90/ 1236] Overall Loss 0.259847 Objective Loss 0.259847 LR 0.001000 Time 0.030667 +2023-10-02 21:08:44,846 - Epoch: [83][ 100/ 1236] Overall Loss 0.257731 Objective Loss 0.257731 LR 0.001000 Time 0.029664 +2023-10-02 21:08:45,049 - Epoch: [83][ 110/ 1236] Overall Loss 0.258822 Objective Loss 0.258822 LR 0.001000 Time 0.028813 +2023-10-02 21:08:45,255 - Epoch: [83][ 120/ 1236] Overall Loss 0.257462 Objective Loss 0.257462 LR 0.001000 Time 0.028128 +2023-10-02 21:08:45,458 - Epoch: [83][ 130/ 1236] Overall Loss 0.256991 Objective Loss 0.256991 LR 0.001000 Time 0.027524 +2023-10-02 21:08:45,665 - Epoch: [83][ 140/ 1236] Overall Loss 0.258171 Objective Loss 0.258171 LR 0.001000 Time 0.027032 +2023-10-02 21:08:45,868 - Epoch: [83][ 150/ 1236] Overall Loss 0.259340 Objective Loss 0.259340 LR 0.001000 Time 0.026582 +2023-10-02 21:08:46,074 - Epoch: [83][ 160/ 1236] Overall Loss 0.259057 Objective Loss 0.259057 LR 0.001000 Time 0.026208 +2023-10-02 21:08:46,277 - Epoch: [83][ 170/ 1236] Overall Loss 0.259287 Objective Loss 0.259287 LR 0.001000 Time 0.025859 +2023-10-02 21:08:46,484 - Epoch: [83][ 180/ 1236] Overall Loss 0.256281 Objective Loss 0.256281 LR 0.001000 Time 0.025570 +2023-10-02 21:08:46,687 - Epoch: [83][ 190/ 1236] Overall Loss 0.258665 Objective Loss 0.258665 LR 0.001000 Time 0.025292 +2023-10-02 21:08:46,894 - Epoch: [83][ 200/ 1236] Overall Loss 0.257176 Objective Loss 0.257176 LR 0.001000 Time 0.025059 +2023-10-02 21:08:47,097 - Epoch: [83][ 210/ 1236] Overall Loss 0.257679 Objective Loss 0.257679 LR 0.001000 Time 0.024832 +2023-10-02 21:08:47,304 - Epoch: [83][ 220/ 1236] Overall Loss 0.258750 Objective Loss 0.258750 LR 0.001000 Time 0.024641 +2023-10-02 21:08:47,507 - Epoch: [83][ 230/ 1236] Overall Loss 0.259195 Objective Loss 0.259195 LR 0.001000 Time 0.024453 +2023-10-02 21:08:47,713 - Epoch: [83][ 240/ 1236] Overall Loss 0.260950 Objective Loss 0.260950 LR 0.001000 Time 0.024291 +2023-10-02 21:08:47,916 - Epoch: [83][ 250/ 1236] Overall Loss 0.259812 Objective Loss 0.259812 LR 0.001000 Time 0.024128 +2023-10-02 21:08:48,122 - Epoch: [83][ 260/ 1236] Overall Loss 0.259862 Objective Loss 0.259862 LR 0.001000 Time 0.023994 +2023-10-02 21:08:48,325 - Epoch: [83][ 270/ 1236] Overall Loss 0.259456 Objective Loss 0.259456 LR 0.001000 Time 0.023856 +2023-10-02 21:08:48,532 - Epoch: [83][ 280/ 1236] Overall Loss 0.259962 Objective Loss 0.259962 LR 0.001000 Time 0.023742 +2023-10-02 21:08:48,735 - Epoch: [83][ 290/ 1236] Overall Loss 0.261266 Objective Loss 0.261266 LR 0.001000 Time 0.023623 +2023-10-02 21:08:48,942 - Epoch: [83][ 300/ 1236] Overall Loss 0.261479 Objective Loss 0.261479 LR 0.001000 Time 0.023521 +2023-10-02 21:08:49,145 - Epoch: [83][ 310/ 1236] Overall Loss 0.262868 Objective Loss 0.262868 LR 0.001000 Time 0.023417 +2023-10-02 21:08:49,351 - Epoch: [83][ 320/ 1236] Overall Loss 0.263193 Objective Loss 0.263193 LR 0.001000 Time 0.023330 +2023-10-02 21:08:49,555 - Epoch: [83][ 330/ 1236] Overall Loss 0.263296 Objective Loss 0.263296 LR 0.001000 Time 0.023240 +2023-10-02 21:08:49,762 - Epoch: [83][ 340/ 1236] Overall Loss 0.263936 Objective Loss 0.263936 LR 0.001000 Time 0.023164 +2023-10-02 21:08:49,966 - Epoch: [83][ 350/ 1236] Overall Loss 0.262761 Objective Loss 0.262761 LR 0.001000 Time 0.023083 +2023-10-02 21:08:50,171 - Epoch: [83][ 360/ 1236] Overall Loss 0.263216 Objective Loss 0.263216 LR 0.001000 Time 0.023011 +2023-10-02 21:08:50,375 - Epoch: [83][ 370/ 1236] Overall Loss 0.263197 Objective Loss 0.263197 LR 0.001000 Time 0.022942 +2023-10-02 21:08:50,582 - Epoch: [83][ 380/ 1236] Overall Loss 0.263420 Objective Loss 0.263420 LR 0.001000 Time 0.022882 +2023-10-02 21:08:50,786 - Epoch: [83][ 390/ 1236] Overall Loss 0.264436 Objective Loss 0.264436 LR 0.001000 Time 0.022816 +2023-10-02 21:08:50,993 - Epoch: [83][ 400/ 1236] Overall Loss 0.265615 Objective Loss 0.265615 LR 0.001000 Time 0.022762 +2023-10-02 21:08:51,196 - Epoch: [83][ 410/ 1236] Overall Loss 0.266591 Objective Loss 0.266591 LR 0.001000 Time 0.022702 +2023-10-02 21:08:51,403 - Epoch: [83][ 420/ 1236] Overall Loss 0.266744 Objective Loss 0.266744 LR 0.001000 Time 0.022654 +2023-10-02 21:08:51,606 - Epoch: [83][ 430/ 1236] Overall Loss 0.266675 Objective Loss 0.266675 LR 0.001000 Time 0.022599 +2023-10-02 21:08:51,812 - Epoch: [83][ 440/ 1236] Overall Loss 0.267486 Objective Loss 0.267486 LR 0.001000 Time 0.022552 +2023-10-02 21:08:52,017 - Epoch: [83][ 450/ 1236] Overall Loss 0.268035 Objective Loss 0.268035 LR 0.001000 Time 0.022505 +2023-10-02 21:08:52,223 - Epoch: [83][ 460/ 1236] Overall Loss 0.269218 Objective Loss 0.269218 LR 0.001000 Time 0.022465 +2023-10-02 21:08:52,427 - Epoch: [83][ 470/ 1236] Overall Loss 0.269307 Objective Loss 0.269307 LR 0.001000 Time 0.022419 +2023-10-02 21:08:52,632 - Epoch: [83][ 480/ 1236] Overall Loss 0.268841 Objective Loss 0.268841 LR 0.001000 Time 0.022379 +2023-10-02 21:08:52,837 - Epoch: [83][ 490/ 1236] Overall Loss 0.269033 Objective Loss 0.269033 LR 0.001000 Time 0.022340 +2023-10-02 21:08:53,042 - Epoch: [83][ 500/ 1236] Overall Loss 0.268389 Objective Loss 0.268389 LR 0.001000 Time 0.022303 +2023-10-02 21:08:53,247 - Epoch: [83][ 510/ 1236] Overall Loss 0.268204 Objective Loss 0.268204 LR 0.001000 Time 0.022267 +2023-10-02 21:08:53,453 - Epoch: [83][ 520/ 1236] Overall Loss 0.267935 Objective Loss 0.267935 LR 0.001000 Time 0.022234 +2023-10-02 21:08:53,658 - Epoch: [83][ 530/ 1236] Overall Loss 0.267749 Objective Loss 0.267749 LR 0.001000 Time 0.022200 +2023-10-02 21:08:53,863 - Epoch: [83][ 540/ 1236] Overall Loss 0.268135 Objective Loss 0.268135 LR 0.001000 Time 0.022169 +2023-10-02 21:08:54,068 - Epoch: [83][ 550/ 1236] Overall Loss 0.268729 Objective Loss 0.268729 LR 0.001000 Time 0.022138 +2023-10-02 21:08:54,274 - Epoch: [83][ 560/ 1236] Overall Loss 0.268876 Objective Loss 0.268876 LR 0.001000 Time 0.022109 +2023-10-02 21:08:54,478 - Epoch: [83][ 570/ 1236] Overall Loss 0.269418 Objective Loss 0.269418 LR 0.001000 Time 0.022080 +2023-10-02 21:08:54,685 - Epoch: [83][ 580/ 1236] Overall Loss 0.269545 Objective Loss 0.269545 LR 0.001000 Time 0.022056 +2023-10-02 21:08:54,889 - Epoch: [83][ 590/ 1236] Overall Loss 0.270008 Objective Loss 0.270008 LR 0.001000 Time 0.022027 +2023-10-02 21:08:55,094 - Epoch: [83][ 600/ 1236] Overall Loss 0.270243 Objective Loss 0.270243 LR 0.001000 Time 0.022001 +2023-10-02 21:08:55,299 - Epoch: [83][ 610/ 1236] Overall Loss 0.270615 Objective Loss 0.270615 LR 0.001000 Time 0.021976 +2023-10-02 21:08:55,506 - Epoch: [83][ 620/ 1236] Overall Loss 0.270832 Objective Loss 0.270832 LR 0.001000 Time 0.021955 +2023-10-02 21:08:55,710 - Epoch: [83][ 630/ 1236] Overall Loss 0.271006 Objective Loss 0.271006 LR 0.001000 Time 0.021929 +2023-10-02 21:08:55,915 - Epoch: [83][ 640/ 1236] Overall Loss 0.271166 Objective Loss 0.271166 LR 0.001000 Time 0.021907 +2023-10-02 21:08:56,120 - Epoch: [83][ 650/ 1236] Overall Loss 0.271644 Objective Loss 0.271644 LR 0.001000 Time 0.021885 +2023-10-02 21:08:56,326 - Epoch: [83][ 660/ 1236] Overall Loss 0.271778 Objective Loss 0.271778 LR 0.001000 Time 0.021864 +2023-10-02 21:08:56,531 - Epoch: [83][ 670/ 1236] Overall Loss 0.272833 Objective Loss 0.272833 LR 0.001000 Time 0.021843 +2023-10-02 21:08:56,736 - Epoch: [83][ 680/ 1236] Overall Loss 0.273212 Objective Loss 0.273212 LR 0.001000 Time 0.021824 +2023-10-02 21:08:56,941 - Epoch: [83][ 690/ 1236] Overall Loss 0.273487 Objective Loss 0.273487 LR 0.001000 Time 0.021804 +2023-10-02 21:08:57,146 - Epoch: [83][ 700/ 1236] Overall Loss 0.273735 Objective Loss 0.273735 LR 0.001000 Time 0.021786 +2023-10-02 21:08:57,351 - Epoch: [83][ 710/ 1236] Overall Loss 0.273705 Objective Loss 0.273705 LR 0.001000 Time 0.021767 +2023-10-02 21:08:57,558 - Epoch: [83][ 720/ 1236] Overall Loss 0.274312 Objective Loss 0.274312 LR 0.001000 Time 0.021751 +2023-10-02 21:08:57,761 - Epoch: [83][ 730/ 1236] Overall Loss 0.274649 Objective Loss 0.274649 LR 0.001000 Time 0.021731 +2023-10-02 21:08:57,967 - Epoch: [83][ 740/ 1236] Overall Loss 0.275382 Objective Loss 0.275382 LR 0.001000 Time 0.021715 +2023-10-02 21:08:58,172 - Epoch: [83][ 750/ 1236] Overall Loss 0.275855 Objective Loss 0.275855 LR 0.001000 Time 0.021698 +2023-10-02 21:08:58,378 - Epoch: [83][ 760/ 1236] Overall Loss 0.275731 Objective Loss 0.275731 LR 0.001000 Time 0.021684 +2023-10-02 21:08:58,582 - Epoch: [83][ 770/ 1236] Overall Loss 0.275756 Objective Loss 0.275756 LR 0.001000 Time 0.021667 +2023-10-02 21:08:58,787 - Epoch: [83][ 780/ 1236] Overall Loss 0.275758 Objective Loss 0.275758 LR 0.001000 Time 0.021652 +2023-10-02 21:08:58,992 - Epoch: [83][ 790/ 1236] Overall Loss 0.276125 Objective Loss 0.276125 LR 0.001000 Time 0.021637 +2023-10-02 21:08:59,198 - Epoch: [83][ 800/ 1236] Overall Loss 0.276279 Objective Loss 0.276279 LR 0.001000 Time 0.021623 +2023-10-02 21:08:59,402 - Epoch: [83][ 810/ 1236] Overall Loss 0.276191 Objective Loss 0.276191 LR 0.001000 Time 0.021608 +2023-10-02 21:08:59,608 - Epoch: [83][ 820/ 1236] Overall Loss 0.275973 Objective Loss 0.275973 LR 0.001000 Time 0.021595 +2023-10-02 21:08:59,813 - Epoch: [83][ 830/ 1236] Overall Loss 0.275597 Objective Loss 0.275597 LR 0.001000 Time 0.021582 +2023-10-02 21:09:00,018 - Epoch: [83][ 840/ 1236] Overall Loss 0.275590 Objective Loss 0.275590 LR 0.001000 Time 0.021569 +2023-10-02 21:09:00,223 - Epoch: [83][ 850/ 1236] Overall Loss 0.275756 Objective Loss 0.275756 LR 0.001000 Time 0.021556 +2023-10-02 21:09:00,429 - Epoch: [83][ 860/ 1236] Overall Loss 0.276081 Objective Loss 0.276081 LR 0.001000 Time 0.021544 +2023-10-02 21:09:00,634 - Epoch: [83][ 870/ 1236] Overall Loss 0.275968 Objective Loss 0.275968 LR 0.001000 Time 0.021531 +2023-10-02 21:09:00,839 - Epoch: [83][ 880/ 1236] Overall Loss 0.276274 Objective Loss 0.276274 LR 0.001000 Time 0.021520 +2023-10-02 21:09:01,044 - Epoch: [83][ 890/ 1236] Overall Loss 0.276256 Objective Loss 0.276256 LR 0.001000 Time 0.021508 +2023-10-02 21:09:01,249 - Epoch: [83][ 900/ 1236] Overall Loss 0.276161 Objective Loss 0.276161 LR 0.001000 Time 0.021497 +2023-10-02 21:09:01,454 - Epoch: [83][ 910/ 1236] Overall Loss 0.276036 Objective Loss 0.276036 LR 0.001000 Time 0.021486 +2023-10-02 21:09:01,660 - Epoch: [83][ 920/ 1236] Overall Loss 0.276245 Objective Loss 0.276245 LR 0.001000 Time 0.021475 +2023-10-02 21:09:01,865 - Epoch: [83][ 930/ 1236] Overall Loss 0.276135 Objective Loss 0.276135 LR 0.001000 Time 0.021464 +2023-10-02 21:09:02,070 - Epoch: [83][ 940/ 1236] Overall Loss 0.276212 Objective Loss 0.276212 LR 0.001000 Time 0.021454 +2023-10-02 21:09:02,275 - Epoch: [83][ 950/ 1236] Overall Loss 0.276760 Objective Loss 0.276760 LR 0.001000 Time 0.021444 +2023-10-02 21:09:02,481 - Epoch: [83][ 960/ 1236] Overall Loss 0.276976 Objective Loss 0.276976 LR 0.001000 Time 0.021434 +2023-10-02 21:09:02,685 - Epoch: [83][ 970/ 1236] Overall Loss 0.277208 Objective Loss 0.277208 LR 0.001000 Time 0.021424 +2023-10-02 21:09:02,891 - Epoch: [83][ 980/ 1236] Overall Loss 0.276896 Objective Loss 0.276896 LR 0.001000 Time 0.021415 +2023-10-02 21:09:03,096 - Epoch: [83][ 990/ 1236] Overall Loss 0.276992 Objective Loss 0.276992 LR 0.001000 Time 0.021405 +2023-10-02 21:09:03,301 - Epoch: [83][ 1000/ 1236] Overall Loss 0.277086 Objective Loss 0.277086 LR 0.001000 Time 0.021396 +2023-10-02 21:09:03,506 - Epoch: [83][ 1010/ 1236] Overall Loss 0.277147 Objective Loss 0.277147 LR 0.001000 Time 0.021387 +2023-10-02 21:09:03,712 - Epoch: [83][ 1020/ 1236] Overall Loss 0.277572 Objective Loss 0.277572 LR 0.001000 Time 0.021379 +2023-10-02 21:09:03,917 - Epoch: [83][ 1030/ 1236] Overall Loss 0.277687 Objective Loss 0.277687 LR 0.001000 Time 0.021370 +2023-10-02 21:09:04,122 - Epoch: [83][ 1040/ 1236] Overall Loss 0.277833 Objective Loss 0.277833 LR 0.001000 Time 0.021361 +2023-10-02 21:09:04,327 - Epoch: [83][ 1050/ 1236] Overall Loss 0.278106 Objective Loss 0.278106 LR 0.001000 Time 0.021353 +2023-10-02 21:09:04,533 - Epoch: [83][ 1060/ 1236] Overall Loss 0.277846 Objective Loss 0.277846 LR 0.001000 Time 0.021345 +2023-10-02 21:09:04,737 - Epoch: [83][ 1070/ 1236] Overall Loss 0.277999 Objective Loss 0.277999 LR 0.001000 Time 0.021337 +2023-10-02 21:09:04,943 - Epoch: [83][ 1080/ 1236] Overall Loss 0.277987 Objective Loss 0.277987 LR 0.001000 Time 0.021329 +2023-10-02 21:09:05,148 - Epoch: [83][ 1090/ 1236] Overall Loss 0.277752 Objective Loss 0.277752 LR 0.001000 Time 0.021321 +2023-10-02 21:09:05,353 - Epoch: [83][ 1100/ 1236] Overall Loss 0.278168 Objective Loss 0.278168 LR 0.001000 Time 0.021314 +2023-10-02 21:09:05,558 - Epoch: [83][ 1110/ 1236] Overall Loss 0.278547 Objective Loss 0.278547 LR 0.001000 Time 0.021307 +2023-10-02 21:09:05,764 - Epoch: [83][ 1120/ 1236] Overall Loss 0.278424 Objective Loss 0.278424 LR 0.001000 Time 0.021300 +2023-10-02 21:09:05,969 - Epoch: [83][ 1130/ 1236] Overall Loss 0.278151 Objective Loss 0.278151 LR 0.001000 Time 0.021292 +2023-10-02 21:09:06,175 - Epoch: [83][ 1140/ 1236] Overall Loss 0.278168 Objective Loss 0.278168 LR 0.001000 Time 0.021286 +2023-10-02 21:09:06,379 - Epoch: [83][ 1150/ 1236] Overall Loss 0.278015 Objective Loss 0.278015 LR 0.001000 Time 0.021278 +2023-10-02 21:09:06,585 - Epoch: [83][ 1160/ 1236] Overall Loss 0.278053 Objective Loss 0.278053 LR 0.001000 Time 0.021272 +2023-10-02 21:09:06,790 - Epoch: [83][ 1170/ 1236] Overall Loss 0.277785 Objective Loss 0.277785 LR 0.001000 Time 0.021265 +2023-10-02 21:09:06,995 - Epoch: [83][ 1180/ 1236] Overall Loss 0.277421 Objective Loss 0.277421 LR 0.001000 Time 0.021258 +2023-10-02 21:09:07,200 - Epoch: [83][ 1190/ 1236] Overall Loss 0.277669 Objective Loss 0.277669 LR 0.001000 Time 0.021252 +2023-10-02 21:09:07,405 - Epoch: [83][ 1200/ 1236] Overall Loss 0.277725 Objective Loss 0.277725 LR 0.001000 Time 0.021245 +2023-10-02 21:09:07,610 - Epoch: [83][ 1210/ 1236] Overall Loss 0.277580 Objective Loss 0.277580 LR 0.001000 Time 0.021239 +2023-10-02 21:09:07,817 - Epoch: [83][ 1220/ 1236] Overall Loss 0.277842 Objective Loss 0.277842 LR 0.001000 Time 0.021234 +2023-10-02 21:09:08,073 - Epoch: [83][ 1230/ 1236] Overall Loss 0.277993 Objective Loss 0.277993 LR 0.001000 Time 0.021269 +2023-10-02 21:09:08,194 - Epoch: [83][ 1236/ 1236] Overall Loss 0.277961 Objective Loss 0.277961 Top1 85.947047 Top5 98.370672 LR 0.001000 Time 0.021264 +2023-10-02 21:09:08,340 - --- validate (epoch=83)----------- +2023-10-02 21:09:08,340 - 29943 samples (256 per mini-batch) +2023-10-02 21:09:08,829 - Epoch: [83][ 10/ 117] Loss 0.328945 Top1 83.632812 Top5 98.320312 +2023-10-02 21:09:08,991 - Epoch: [83][ 20/ 117] Loss 0.342015 Top1 83.671875 Top5 98.281250 +2023-10-02 21:09:09,150 - Epoch: [83][ 30/ 117] Loss 0.332660 Top1 83.802083 Top5 98.255208 +2023-10-02 21:09:09,309 - Epoch: [83][ 40/ 117] Loss 0.332480 Top1 83.896484 Top5 98.330078 +2023-10-02 21:09:09,463 - Epoch: [83][ 50/ 117] Loss 0.336647 Top1 83.843750 Top5 98.265625 +2023-10-02 21:09:09,615 - Epoch: [83][ 60/ 117] Loss 0.331672 Top1 84.095052 Top5 98.313802 +2023-10-02 21:09:09,766 - Epoch: [83][ 70/ 117] Loss 0.333684 Top1 84.095982 Top5 98.264509 +2023-10-02 21:09:09,917 - Epoch: [83][ 80/ 117] Loss 0.335208 Top1 84.003906 Top5 98.247070 +2023-10-02 21:09:10,067 - Epoch: [83][ 90/ 117] Loss 0.334874 Top1 83.910590 Top5 98.177083 +2023-10-02 21:09:10,217 - Epoch: [83][ 100/ 117] Loss 0.334990 Top1 83.925781 Top5 98.132812 +2023-10-02 21:09:10,373 - Epoch: [83][ 110/ 117] Loss 0.334902 Top1 83.884943 Top5 98.146307 +2023-10-02 21:09:10,462 - Epoch: [83][ 117/ 117] Loss 0.334120 Top1 83.919447 Top5 98.159837 +2023-10-02 21:09:10,607 - ==> Top1: 83.919 Top5: 98.160 Loss: 0.334 + +2023-10-02 21:09:10,608 - ==> Confusion: +[[ 932 0 3 2 8 3 0 0 7 64 1 0 1 2 7 1 4 1 2 0 12] + [ 1 1033 4 2 4 17 1 31 3 0 4 1 0 1 1 4 1 0 11 5 7] + [ 3 0 963 14 3 0 25 4 0 1 3 1 9 3 2 3 2 1 11 1 7] + [ 1 2 13 981 3 4 1 3 3 3 7 0 5 2 32 0 1 6 10 1 11] + [ 28 9 1 0 961 3 0 1 1 12 3 1 1 3 7 3 10 1 1 2 2] + [ 4 57 0 0 3 938 2 36 3 4 4 4 4 13 3 0 5 1 11 11 13] + [ 0 1 27 3 0 0 1118 7 0 1 8 2 0 0 0 2 1 1 2 11 7] + [ 3 16 14 1 3 20 8 1067 3 2 1 8 4 3 4 0 3 0 37 7 14] + [ 23 2 0 0 4 1 0 2 973 30 12 2 5 6 15 0 2 1 8 2 1] + [ 107 1 0 1 5 3 1 1 33 933 4 1 0 13 7 0 0 0 0 1 8] + [ 2 0 9 12 0 3 2 6 16 2 955 1 0 15 4 0 3 1 5 0 17] + [ 2 3 2 1 1 10 0 3 0 0 0 924 37 8 0 1 3 16 0 16 8] + [ 0 2 1 5 0 4 1 2 2 1 2 48 941 1 2 5 2 20 2 5 22] + [ 3 0 1 0 1 8 0 1 21 11 4 5 2 1040 5 1 1 1 0 2 12] + [ 15 0 5 23 4 0 0 0 19 3 3 0 4 2 986 0 1 2 24 0 10] + [ 0 0 1 1 6 0 3 0 0 0 0 5 7 2 0 1047 19 21 2 6 14] + [ 1 23 0 1 7 6 0 1 3 1 0 4 3 1 4 4 1083 0 0 8 11] + [ 1 1 0 2 0 0 1 0 2 1 0 5 28 1 5 6 0 977 3 4 1] + [ 3 5 7 16 0 0 0 19 1 1 9 3 0 0 10 0 0 0 981 0 13] + [ 0 2 3 1 1 4 7 6 0 2 2 8 4 0 0 4 6 0 0 1096 6] + [ 160 151 142 123 85 147 46 117 121 96 183 89 272 268 141 39 106 74 149 197 5199]] + +2023-10-02 21:09:10,609 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:09:10,609 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:09:10,615 - + +2023-10-02 21:09:10,615 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:09:11,627 - Epoch: [84][ 10/ 1236] Overall Loss 0.259497 Objective Loss 0.259497 LR 0.001000 Time 0.101141 +2023-10-02 21:09:11,832 - Epoch: [84][ 20/ 1236] Overall Loss 0.268185 Objective Loss 0.268185 LR 0.001000 Time 0.060813 +2023-10-02 21:09:12,037 - Epoch: [84][ 30/ 1236] Overall Loss 0.276650 Objective Loss 0.276650 LR 0.001000 Time 0.047348 +2023-10-02 21:09:12,244 - Epoch: [84][ 40/ 1236] Overall Loss 0.264321 Objective Loss 0.264321 LR 0.001000 Time 0.040674 +2023-10-02 21:09:12,447 - Epoch: [84][ 50/ 1236] Overall Loss 0.264631 Objective Loss 0.264631 LR 0.001000 Time 0.036595 +2023-10-02 21:09:12,654 - Epoch: [84][ 60/ 1236] Overall Loss 0.265645 Objective Loss 0.265645 LR 0.001000 Time 0.033938 +2023-10-02 21:09:12,857 - Epoch: [84][ 70/ 1236] Overall Loss 0.268244 Objective Loss 0.268244 LR 0.001000 Time 0.031988 +2023-10-02 21:09:13,064 - Epoch: [84][ 80/ 1236] Overall Loss 0.267818 Objective Loss 0.267818 LR 0.001000 Time 0.030571 +2023-10-02 21:09:13,267 - Epoch: [84][ 90/ 1236] Overall Loss 0.267178 Objective Loss 0.267178 LR 0.001000 Time 0.029430 +2023-10-02 21:09:13,474 - Epoch: [84][ 100/ 1236] Overall Loss 0.264864 Objective Loss 0.264864 LR 0.001000 Time 0.028553 +2023-10-02 21:09:13,677 - Epoch: [84][ 110/ 1236] Overall Loss 0.268827 Objective Loss 0.268827 LR 0.001000 Time 0.027802 +2023-10-02 21:09:13,884 - Epoch: [84][ 120/ 1236] Overall Loss 0.271825 Objective Loss 0.271825 LR 0.001000 Time 0.027205 +2023-10-02 21:09:14,087 - Epoch: [84][ 130/ 1236] Overall Loss 0.272379 Objective Loss 0.272379 LR 0.001000 Time 0.026675 +2023-10-02 21:09:14,294 - Epoch: [84][ 140/ 1236] Overall Loss 0.270990 Objective Loss 0.270990 LR 0.001000 Time 0.026245 +2023-10-02 21:09:14,497 - Epoch: [84][ 150/ 1236] Overall Loss 0.273662 Objective Loss 0.273662 LR 0.001000 Time 0.025849 +2023-10-02 21:09:14,704 - Epoch: [84][ 160/ 1236] Overall Loss 0.273036 Objective Loss 0.273036 LR 0.001000 Time 0.025523 +2023-10-02 21:09:14,907 - Epoch: [84][ 170/ 1236] Overall Loss 0.272270 Objective Loss 0.272270 LR 0.001000 Time 0.025214 +2023-10-02 21:09:15,112 - Epoch: [84][ 180/ 1236] Overall Loss 0.273711 Objective Loss 0.273711 LR 0.001000 Time 0.024954 +2023-10-02 21:09:15,316 - Epoch: [84][ 190/ 1236] Overall Loss 0.273436 Objective Loss 0.273436 LR 0.001000 Time 0.024709 +2023-10-02 21:09:15,522 - Epoch: [84][ 200/ 1236] Overall Loss 0.272637 Objective Loss 0.272637 LR 0.001000 Time 0.024506 +2023-10-02 21:09:15,726 - Epoch: [84][ 210/ 1236] Overall Loss 0.270004 Objective Loss 0.270004 LR 0.001000 Time 0.024307 +2023-10-02 21:09:15,932 - Epoch: [84][ 220/ 1236] Overall Loss 0.270371 Objective Loss 0.270371 LR 0.001000 Time 0.024139 +2023-10-02 21:09:16,136 - Epoch: [84][ 230/ 1236] Overall Loss 0.268955 Objective Loss 0.268955 LR 0.001000 Time 0.023974 +2023-10-02 21:09:16,343 - Epoch: [84][ 240/ 1236] Overall Loss 0.269038 Objective Loss 0.269038 LR 0.001000 Time 0.023834 +2023-10-02 21:09:16,546 - Epoch: [84][ 250/ 1236] Overall Loss 0.269404 Objective Loss 0.269404 LR 0.001000 Time 0.023695 +2023-10-02 21:09:16,753 - Epoch: [84][ 260/ 1236] Overall Loss 0.270022 Objective Loss 0.270022 LR 0.001000 Time 0.023577 +2023-10-02 21:09:16,956 - Epoch: [84][ 270/ 1236] Overall Loss 0.269292 Objective Loss 0.269292 LR 0.001000 Time 0.023453 +2023-10-02 21:09:17,162 - Epoch: [84][ 280/ 1236] Overall Loss 0.269822 Objective Loss 0.269822 LR 0.001000 Time 0.023352 +2023-10-02 21:09:17,365 - Epoch: [84][ 290/ 1236] Overall Loss 0.269921 Objective Loss 0.269921 LR 0.001000 Time 0.023247 +2023-10-02 21:09:17,572 - Epoch: [84][ 300/ 1236] Overall Loss 0.271322 Objective Loss 0.271322 LR 0.001000 Time 0.023160 +2023-10-02 21:09:17,775 - Epoch: [84][ 310/ 1236] Overall Loss 0.271506 Objective Loss 0.271506 LR 0.001000 Time 0.023066 +2023-10-02 21:09:17,980 - Epoch: [84][ 320/ 1236] Overall Loss 0.270983 Objective Loss 0.270983 LR 0.001000 Time 0.022987 +2023-10-02 21:09:18,184 - Epoch: [84][ 330/ 1236] Overall Loss 0.270587 Objective Loss 0.270587 LR 0.001000 Time 0.022906 +2023-10-02 21:09:18,391 - Epoch: [84][ 340/ 1236] Overall Loss 0.270106 Objective Loss 0.270106 LR 0.001000 Time 0.022840 +2023-10-02 21:09:18,595 - Epoch: [84][ 350/ 1236] Overall Loss 0.270857 Objective Loss 0.270857 LR 0.001000 Time 0.022769 +2023-10-02 21:09:18,801 - Epoch: [84][ 360/ 1236] Overall Loss 0.271079 Objective Loss 0.271079 LR 0.001000 Time 0.022710 +2023-10-02 21:09:19,005 - Epoch: [84][ 370/ 1236] Overall Loss 0.270535 Objective Loss 0.270535 LR 0.001000 Time 0.022646 +2023-10-02 21:09:19,212 - Epoch: [84][ 380/ 1236] Overall Loss 0.271746 Objective Loss 0.271746 LR 0.001000 Time 0.022593 +2023-10-02 21:09:19,415 - Epoch: [84][ 390/ 1236] Overall Loss 0.270426 Objective Loss 0.270426 LR 0.001000 Time 0.022535 +2023-10-02 21:09:19,622 - Epoch: [84][ 400/ 1236] Overall Loss 0.269764 Objective Loss 0.269764 LR 0.001000 Time 0.022489 +2023-10-02 21:09:19,826 - Epoch: [84][ 410/ 1236] Overall Loss 0.269590 Objective Loss 0.269590 LR 0.001000 Time 0.022435 +2023-10-02 21:09:20,033 - Epoch: [84][ 420/ 1236] Overall Loss 0.269537 Objective Loss 0.269537 LR 0.001000 Time 0.022394 +2023-10-02 21:09:20,236 - Epoch: [84][ 430/ 1236] Overall Loss 0.269659 Objective Loss 0.269659 LR 0.001000 Time 0.022345 +2023-10-02 21:09:20,443 - Epoch: [84][ 440/ 1236] Overall Loss 0.268887 Objective Loss 0.268887 LR 0.001000 Time 0.022307 +2023-10-02 21:09:20,647 - Epoch: [84][ 450/ 1236] Overall Loss 0.268825 Objective Loss 0.268825 LR 0.001000 Time 0.022264 +2023-10-02 21:09:20,854 - Epoch: [84][ 460/ 1236] Overall Loss 0.268663 Objective Loss 0.268663 LR 0.001000 Time 0.022229 +2023-10-02 21:09:21,057 - Epoch: [84][ 470/ 1236] Overall Loss 0.269134 Objective Loss 0.269134 LR 0.001000 Time 0.022188 +2023-10-02 21:09:21,264 - Epoch: [84][ 480/ 1236] Overall Loss 0.269325 Objective Loss 0.269325 LR 0.001000 Time 0.022156 +2023-10-02 21:09:21,468 - Epoch: [84][ 490/ 1236] Overall Loss 0.269226 Objective Loss 0.269226 LR 0.001000 Time 0.022119 +2023-10-02 21:09:21,675 - Epoch: [84][ 500/ 1236] Overall Loss 0.269222 Objective Loss 0.269222 LR 0.001000 Time 0.022090 +2023-10-02 21:09:21,878 - Epoch: [84][ 510/ 1236] Overall Loss 0.268996 Objective Loss 0.268996 LR 0.001000 Time 0.022055 +2023-10-02 21:09:22,085 - Epoch: [84][ 520/ 1236] Overall Loss 0.268591 Objective Loss 0.268591 LR 0.001000 Time 0.022029 +2023-10-02 21:09:22,289 - Epoch: [84][ 530/ 1236] Overall Loss 0.268728 Objective Loss 0.268728 LR 0.001000 Time 0.021997 +2023-10-02 21:09:22,496 - Epoch: [84][ 540/ 1236] Overall Loss 0.269056 Objective Loss 0.269056 LR 0.001000 Time 0.021972 +2023-10-02 21:09:22,699 - Epoch: [84][ 550/ 1236] Overall Loss 0.268574 Objective Loss 0.268574 LR 0.001000 Time 0.021943 +2023-10-02 21:09:22,906 - Epoch: [84][ 560/ 1236] Overall Loss 0.268422 Objective Loss 0.268422 LR 0.001000 Time 0.021920 +2023-10-02 21:09:23,110 - Epoch: [84][ 570/ 1236] Overall Loss 0.268694 Objective Loss 0.268694 LR 0.001000 Time 0.021891 +2023-10-02 21:09:23,317 - Epoch: [84][ 580/ 1236] Overall Loss 0.268808 Objective Loss 0.268808 LR 0.001000 Time 0.021870 +2023-10-02 21:09:23,520 - Epoch: [84][ 590/ 1236] Overall Loss 0.269080 Objective Loss 0.269080 LR 0.001000 Time 0.021844 +2023-10-02 21:09:23,726 - Epoch: [84][ 600/ 1236] Overall Loss 0.269316 Objective Loss 0.269316 LR 0.001000 Time 0.021823 +2023-10-02 21:09:23,931 - Epoch: [84][ 610/ 1236] Overall Loss 0.269201 Objective Loss 0.269201 LR 0.001000 Time 0.021798 +2023-10-02 21:09:24,138 - Epoch: [84][ 620/ 1236] Overall Loss 0.268840 Objective Loss 0.268840 LR 0.001000 Time 0.021780 +2023-10-02 21:09:24,342 - Epoch: [84][ 630/ 1236] Overall Loss 0.269044 Objective Loss 0.269044 LR 0.001000 Time 0.021757 +2023-10-02 21:09:24,549 - Epoch: [84][ 640/ 1236] Overall Loss 0.268604 Objective Loss 0.268604 LR 0.001000 Time 0.021741 +2023-10-02 21:09:24,753 - Epoch: [84][ 650/ 1236] Overall Loss 0.268554 Objective Loss 0.268554 LR 0.001000 Time 0.021719 +2023-10-02 21:09:24,958 - Epoch: [84][ 660/ 1236] Overall Loss 0.268244 Objective Loss 0.268244 LR 0.001000 Time 0.021701 +2023-10-02 21:09:25,163 - Epoch: [84][ 670/ 1236] Overall Loss 0.267734 Objective Loss 0.267734 LR 0.001000 Time 0.021683 +2023-10-02 21:09:25,370 - Epoch: [84][ 680/ 1236] Overall Loss 0.267822 Objective Loss 0.267822 LR 0.001000 Time 0.021668 +2023-10-02 21:09:25,574 - Epoch: [84][ 690/ 1236] Overall Loss 0.267614 Objective Loss 0.267614 LR 0.001000 Time 0.021649 +2023-10-02 21:09:25,779 - Epoch: [84][ 700/ 1236] Overall Loss 0.267446 Objective Loss 0.267446 LR 0.001000 Time 0.021633 +2023-10-02 21:09:25,984 - Epoch: [84][ 710/ 1236] Overall Loss 0.267389 Objective Loss 0.267389 LR 0.001000 Time 0.021614 +2023-10-02 21:09:26,190 - Epoch: [84][ 720/ 1236] Overall Loss 0.267569 Objective Loss 0.267569 LR 0.001000 Time 0.021600 +2023-10-02 21:09:26,395 - Epoch: [84][ 730/ 1236] Overall Loss 0.267415 Objective Loss 0.267415 LR 0.001000 Time 0.021584 +2023-10-02 21:09:26,601 - Epoch: [84][ 740/ 1236] Overall Loss 0.266976 Objective Loss 0.266976 LR 0.001000 Time 0.021570 +2023-10-02 21:09:26,805 - Epoch: [84][ 750/ 1236] Overall Loss 0.266681 Objective Loss 0.266681 LR 0.001000 Time 0.021555 +2023-10-02 21:09:27,011 - Epoch: [84][ 760/ 1236] Overall Loss 0.266676 Objective Loss 0.266676 LR 0.001000 Time 0.021542 +2023-10-02 21:09:27,216 - Epoch: [84][ 770/ 1236] Overall Loss 0.266277 Objective Loss 0.266277 LR 0.001000 Time 0.021527 +2023-10-02 21:09:27,422 - Epoch: [84][ 780/ 1236] Overall Loss 0.266083 Objective Loss 0.266083 LR 0.001000 Time 0.021515 +2023-10-02 21:09:27,627 - Epoch: [84][ 790/ 1236] Overall Loss 0.266001 Objective Loss 0.266001 LR 0.001000 Time 0.021502 +2023-10-02 21:09:27,832 - Epoch: [84][ 800/ 1236] Overall Loss 0.266207 Objective Loss 0.266207 LR 0.001000 Time 0.021490 +2023-10-02 21:09:28,037 - Epoch: [84][ 810/ 1236] Overall Loss 0.265949 Objective Loss 0.265949 LR 0.001000 Time 0.021477 +2023-10-02 21:09:28,243 - Epoch: [84][ 820/ 1236] Overall Loss 0.265575 Objective Loss 0.265575 LR 0.001000 Time 0.021466 +2023-10-02 21:09:28,448 - Epoch: [84][ 830/ 1236] Overall Loss 0.265526 Objective Loss 0.265526 LR 0.001000 Time 0.021454 +2023-10-02 21:09:28,655 - Epoch: [84][ 840/ 1236] Overall Loss 0.265397 Objective Loss 0.265397 LR 0.001000 Time 0.021444 +2023-10-02 21:09:28,859 - Epoch: [84][ 850/ 1236] Overall Loss 0.265003 Objective Loss 0.265003 LR 0.001000 Time 0.021431 +2023-10-02 21:09:29,066 - Epoch: [84][ 860/ 1236] Overall Loss 0.265015 Objective Loss 0.265015 LR 0.001000 Time 0.021422 +2023-10-02 21:09:29,269 - Epoch: [84][ 870/ 1236] Overall Loss 0.264747 Objective Loss 0.264747 LR 0.001000 Time 0.021410 +2023-10-02 21:09:29,475 - Epoch: [84][ 880/ 1236] Overall Loss 0.264561 Objective Loss 0.264561 LR 0.001000 Time 0.021400 +2023-10-02 21:09:29,680 - Epoch: [84][ 890/ 1236] Overall Loss 0.264286 Objective Loss 0.264286 LR 0.001000 Time 0.021390 +2023-10-02 21:09:29,886 - Epoch: [84][ 900/ 1236] Overall Loss 0.264232 Objective Loss 0.264232 LR 0.001000 Time 0.021381 +2023-10-02 21:09:30,091 - Epoch: [84][ 910/ 1236] Overall Loss 0.264051 Objective Loss 0.264051 LR 0.001000 Time 0.021370 +2023-10-02 21:09:30,298 - Epoch: [84][ 920/ 1236] Overall Loss 0.263755 Objective Loss 0.263755 LR 0.001000 Time 0.021363 +2023-10-02 21:09:30,501 - Epoch: [84][ 930/ 1236] Overall Loss 0.263731 Objective Loss 0.263731 LR 0.001000 Time 0.021352 +2023-10-02 21:09:30,707 - Epoch: [84][ 940/ 1236] Overall Loss 0.263460 Objective Loss 0.263460 LR 0.001000 Time 0.021343 +2023-10-02 21:09:30,912 - Epoch: [84][ 950/ 1236] Overall Loss 0.263987 Objective Loss 0.263987 LR 0.001000 Time 0.021332 +2023-10-02 21:09:31,117 - Epoch: [84][ 960/ 1236] Overall Loss 0.263945 Objective Loss 0.263945 LR 0.001000 Time 0.021324 +2023-10-02 21:09:31,322 - Epoch: [84][ 970/ 1236] Overall Loss 0.263756 Objective Loss 0.263756 LR 0.001000 Time 0.021314 +2023-10-02 21:09:31,528 - Epoch: [84][ 980/ 1236] Overall Loss 0.263819 Objective Loss 0.263819 LR 0.001000 Time 0.021306 +2023-10-02 21:09:31,733 - Epoch: [84][ 990/ 1236] Overall Loss 0.263895 Objective Loss 0.263895 LR 0.001000 Time 0.021298 +2023-10-02 21:09:31,940 - Epoch: [84][ 1000/ 1236] Overall Loss 0.264146 Objective Loss 0.264146 LR 0.001000 Time 0.021291 +2023-10-02 21:09:32,144 - Epoch: [84][ 1010/ 1236] Overall Loss 0.264110 Objective Loss 0.264110 LR 0.001000 Time 0.021282 +2023-10-02 21:09:32,349 - Epoch: [84][ 1020/ 1236] Overall Loss 0.264455 Objective Loss 0.264455 LR 0.001000 Time 0.021275 +2023-10-02 21:09:32,554 - Epoch: [84][ 1030/ 1236] Overall Loss 0.264532 Objective Loss 0.264532 LR 0.001000 Time 0.021267 +2023-10-02 21:09:32,761 - Epoch: [84][ 1040/ 1236] Overall Loss 0.264706 Objective Loss 0.264706 LR 0.001000 Time 0.021261 +2023-10-02 21:09:32,965 - Epoch: [84][ 1050/ 1236] Overall Loss 0.264959 Objective Loss 0.264959 LR 0.001000 Time 0.021252 +2023-10-02 21:09:33,172 - Epoch: [84][ 1060/ 1236] Overall Loss 0.264929 Objective Loss 0.264929 LR 0.001000 Time 0.021247 +2023-10-02 21:09:33,375 - Epoch: [84][ 1070/ 1236] Overall Loss 0.264915 Objective Loss 0.264915 LR 0.001000 Time 0.021238 +2023-10-02 21:09:33,582 - Epoch: [84][ 1080/ 1236] Overall Loss 0.265084 Objective Loss 0.265084 LR 0.001000 Time 0.021232 +2023-10-02 21:09:33,787 - Epoch: [84][ 1090/ 1236] Overall Loss 0.265199 Objective Loss 0.265199 LR 0.001000 Time 0.021225 +2023-10-02 21:09:33,992 - Epoch: [84][ 1100/ 1236] Overall Loss 0.265181 Objective Loss 0.265181 LR 0.001000 Time 0.021219 +2023-10-02 21:09:34,197 - Epoch: [84][ 1110/ 1236] Overall Loss 0.265499 Objective Loss 0.265499 LR 0.001000 Time 0.021212 +2023-10-02 21:09:34,403 - Epoch: [84][ 1120/ 1236] Overall Loss 0.265450 Objective Loss 0.265450 LR 0.001000 Time 0.021206 +2023-10-02 21:09:34,608 - Epoch: [84][ 1130/ 1236] Overall Loss 0.265303 Objective Loss 0.265303 LR 0.001000 Time 0.021200 +2023-10-02 21:09:34,814 - Epoch: [84][ 1140/ 1236] Overall Loss 0.265223 Objective Loss 0.265223 LR 0.001000 Time 0.021194 +2023-10-02 21:09:35,019 - Epoch: [84][ 1150/ 1236] Overall Loss 0.265149 Objective Loss 0.265149 LR 0.001000 Time 0.021188 +2023-10-02 21:09:35,225 - Epoch: [84][ 1160/ 1236] Overall Loss 0.265039 Objective Loss 0.265039 LR 0.001000 Time 0.021183 +2023-10-02 21:09:35,430 - Epoch: [84][ 1170/ 1236] Overall Loss 0.264895 Objective Loss 0.264895 LR 0.001000 Time 0.021177 +2023-10-02 21:09:35,636 - Epoch: [84][ 1180/ 1236] Overall Loss 0.264970 Objective Loss 0.264970 LR 0.001000 Time 0.021171 +2023-10-02 21:09:35,841 - Epoch: [84][ 1190/ 1236] Overall Loss 0.265013 Objective Loss 0.265013 LR 0.001000 Time 0.021165 +2023-10-02 21:09:36,047 - Epoch: [84][ 1200/ 1236] Overall Loss 0.264853 Objective Loss 0.264853 LR 0.001000 Time 0.021160 +2023-10-02 21:09:36,252 - Epoch: [84][ 1210/ 1236] Overall Loss 0.264813 Objective Loss 0.264813 LR 0.001000 Time 0.021155 +2023-10-02 21:09:36,458 - Epoch: [84][ 1220/ 1236] Overall Loss 0.264795 Objective Loss 0.264795 LR 0.001000 Time 0.021150 +2023-10-02 21:09:36,715 - Epoch: [84][ 1230/ 1236] Overall Loss 0.264696 Objective Loss 0.264696 LR 0.001000 Time 0.021187 +2023-10-02 21:09:36,836 - Epoch: [84][ 1236/ 1236] Overall Loss 0.264593 Objective Loss 0.264593 Top1 89.205703 Top5 98.370672 LR 0.001000 Time 0.021181 +2023-10-02 21:09:36,974 - --- validate (epoch=84)----------- +2023-10-02 21:09:36,974 - 29943 samples (256 per mini-batch) +2023-10-02 21:09:37,463 - Epoch: [84][ 10/ 117] Loss 0.338639 Top1 83.789062 Top5 97.617188 +2023-10-02 21:09:37,613 - Epoch: [84][ 20/ 117] Loss 0.328701 Top1 83.105469 Top5 97.871094 +2023-10-02 21:09:37,761 - Epoch: [84][ 30/ 117] Loss 0.322809 Top1 83.255208 Top5 98.085938 +2023-10-02 21:09:37,912 - Epoch: [84][ 40/ 117] Loss 0.319836 Top1 83.378906 Top5 98.173828 +2023-10-02 21:09:38,062 - Epoch: [84][ 50/ 117] Loss 0.321125 Top1 83.476562 Top5 98.195312 +2023-10-02 21:09:38,212 - Epoch: [84][ 60/ 117] Loss 0.321345 Top1 83.496094 Top5 98.190104 +2023-10-02 21:09:38,362 - Epoch: [84][ 70/ 117] Loss 0.319512 Top1 83.565848 Top5 98.169643 +2023-10-02 21:09:38,512 - Epoch: [84][ 80/ 117] Loss 0.321343 Top1 83.569336 Top5 98.188477 +2023-10-02 21:09:38,662 - Epoch: [84][ 90/ 117] Loss 0.321457 Top1 83.654514 Top5 98.177083 +2023-10-02 21:09:38,812 - Epoch: [84][ 100/ 117] Loss 0.319676 Top1 83.582031 Top5 98.164062 +2023-10-02 21:09:38,969 - Epoch: [84][ 110/ 117] Loss 0.323006 Top1 83.480114 Top5 98.164062 +2023-10-02 21:09:39,058 - Epoch: [84][ 117/ 117] Loss 0.323241 Top1 83.458571 Top5 98.173196 +2023-10-02 21:09:39,187 - ==> Top1: 83.459 Top5: 98.173 Loss: 0.323 + +2023-10-02 21:09:39,188 - ==> Confusion: +[[ 916 2 1 1 8 3 0 0 14 70 2 2 2 2 5 0 5 3 1 0 13] + [ 1 1035 0 2 5 28 1 27 2 1 1 1 1 0 1 3 2 0 14 4 2] + [ 3 1 948 16 3 0 18 11 0 1 4 0 11 5 1 4 1 0 19 2 8] + [ 2 2 9 964 0 7 0 6 5 1 9 0 15 3 33 0 0 6 12 1 14] + [ 26 8 1 1 959 4 0 0 1 16 1 2 2 2 4 4 9 0 1 3 6] + [ 3 36 0 2 1 987 1 29 2 5 2 3 6 5 2 0 3 1 7 6 15] + [ 0 4 25 0 0 0 1123 10 0 0 5 2 0 0 0 6 0 0 3 7 6] + [ 2 13 10 0 0 20 7 1075 1 2 2 8 8 3 1 0 0 0 40 19 7] + [ 14 2 2 1 1 3 0 0 995 41 5 1 4 5 9 1 0 0 2 1 2] + [ 78 0 0 1 6 6 0 0 36 958 2 3 0 11 3 1 0 0 0 3 11] + [ 2 3 5 7 0 3 1 5 22 2 961 4 0 14 6 1 0 1 7 0 9] + [ 0 0 2 0 1 29 0 2 0 0 0 918 41 14 0 2 0 15 0 4 7] + [ 0 2 2 3 0 4 1 0 2 1 2 28 982 1 3 5 1 13 2 5 11] + [ 1 0 3 1 0 19 0 0 22 17 7 5 0 1025 5 0 1 0 0 1 12] + [ 11 1 1 15 2 0 0 0 27 4 1 0 7 6 999 0 1 2 16 0 8] + [ 0 0 2 1 5 0 0 0 0 0 1 6 13 2 1 1060 16 17 0 6 4] + [ 2 20 0 0 6 6 0 0 1 1 0 5 4 2 3 6 1081 2 2 8 12] + [ 0 0 0 7 0 0 2 0 1 0 0 6 31 3 7 1 0 973 2 0 5] + [ 4 3 2 18 0 1 0 16 5 0 1 0 3 0 10 0 0 0 996 0 9] + [ 0 1 2 4 1 8 9 6 0 0 2 12 8 3 0 2 2 1 1 1083 7] + [ 135 177 103 101 106 162 37 110 176 108 156 98 423 257 136 47 111 85 240 185 4952]] + +2023-10-02 21:09:39,189 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:09:39,189 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:09:39,195 - + +2023-10-02 21:09:39,195 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:09:40,320 - Epoch: [85][ 10/ 1236] Overall Loss 0.259641 Objective Loss 0.259641 LR 0.001000 Time 0.112434 +2023-10-02 21:09:40,529 - Epoch: [85][ 20/ 1236] Overall Loss 0.243061 Objective Loss 0.243061 LR 0.001000 Time 0.066645 +2023-10-02 21:09:40,736 - Epoch: [85][ 30/ 1236] Overall Loss 0.236500 Objective Loss 0.236500 LR 0.001000 Time 0.051317 +2023-10-02 21:09:40,944 - Epoch: [85][ 40/ 1236] Overall Loss 0.242555 Objective Loss 0.242555 LR 0.001000 Time 0.043670 +2023-10-02 21:09:41,150 - Epoch: [85][ 50/ 1236] Overall Loss 0.239108 Objective Loss 0.239108 LR 0.001000 Time 0.039042 +2023-10-02 21:09:41,358 - Epoch: [85][ 60/ 1236] Overall Loss 0.240933 Objective Loss 0.240933 LR 0.001000 Time 0.035992 +2023-10-02 21:09:41,565 - Epoch: [85][ 70/ 1236] Overall Loss 0.242155 Objective Loss 0.242155 LR 0.001000 Time 0.033780 +2023-10-02 21:09:41,773 - Epoch: [85][ 80/ 1236] Overall Loss 0.243436 Objective Loss 0.243436 LR 0.001000 Time 0.032151 +2023-10-02 21:09:41,979 - Epoch: [85][ 90/ 1236] Overall Loss 0.241055 Objective Loss 0.241055 LR 0.001000 Time 0.030858 +2023-10-02 21:09:42,188 - Epoch: [85][ 100/ 1236] Overall Loss 0.245817 Objective Loss 0.245817 LR 0.001000 Time 0.029858 +2023-10-02 21:09:42,393 - Epoch: [85][ 110/ 1236] Overall Loss 0.244718 Objective Loss 0.244718 LR 0.001000 Time 0.029007 +2023-10-02 21:09:42,601 - Epoch: [85][ 120/ 1236] Overall Loss 0.245527 Objective Loss 0.245527 LR 0.001000 Time 0.028318 +2023-10-02 21:09:42,807 - Epoch: [85][ 130/ 1236] Overall Loss 0.246513 Objective Loss 0.246513 LR 0.001000 Time 0.027715 +2023-10-02 21:09:43,015 - Epoch: [85][ 140/ 1236] Overall Loss 0.246425 Objective Loss 0.246425 LR 0.001000 Time 0.027218 +2023-10-02 21:09:43,221 - Epoch: [85][ 150/ 1236] Overall Loss 0.244624 Objective Loss 0.244624 LR 0.001000 Time 0.026769 +2023-10-02 21:09:43,430 - Epoch: [85][ 160/ 1236] Overall Loss 0.244324 Objective Loss 0.244324 LR 0.001000 Time 0.026399 +2023-10-02 21:09:43,635 - Epoch: [85][ 170/ 1236] Overall Loss 0.244861 Objective Loss 0.244861 LR 0.001000 Time 0.026052 +2023-10-02 21:09:43,843 - Epoch: [85][ 180/ 1236] Overall Loss 0.244319 Objective Loss 0.244319 LR 0.001000 Time 0.025756 +2023-10-02 21:09:44,050 - Epoch: [85][ 190/ 1236] Overall Loss 0.244630 Objective Loss 0.244630 LR 0.001000 Time 0.025481 +2023-10-02 21:09:44,259 - Epoch: [85][ 200/ 1236] Overall Loss 0.246412 Objective Loss 0.246412 LR 0.001000 Time 0.025250 +2023-10-02 21:09:44,464 - Epoch: [85][ 210/ 1236] Overall Loss 0.246829 Objective Loss 0.246829 LR 0.001000 Time 0.025025 +2023-10-02 21:09:44,673 - Epoch: [85][ 220/ 1236] Overall Loss 0.248358 Objective Loss 0.248358 LR 0.001000 Time 0.024836 +2023-10-02 21:09:44,879 - Epoch: [85][ 230/ 1236] Overall Loss 0.248594 Objective Loss 0.248594 LR 0.001000 Time 0.024648 +2023-10-02 21:09:45,088 - Epoch: [85][ 240/ 1236] Overall Loss 0.249036 Objective Loss 0.249036 LR 0.001000 Time 0.024491 +2023-10-02 21:09:45,293 - Epoch: [85][ 250/ 1236] Overall Loss 0.248956 Objective Loss 0.248956 LR 0.001000 Time 0.024332 +2023-10-02 21:09:45,502 - Epoch: [85][ 260/ 1236] Overall Loss 0.249844 Objective Loss 0.249844 LR 0.001000 Time 0.024199 +2023-10-02 21:09:45,707 - Epoch: [85][ 270/ 1236] Overall Loss 0.249592 Objective Loss 0.249592 LR 0.001000 Time 0.024062 +2023-10-02 21:09:45,916 - Epoch: [85][ 280/ 1236] Overall Loss 0.250192 Objective Loss 0.250192 LR 0.001000 Time 0.023948 +2023-10-02 21:09:46,122 - Epoch: [85][ 290/ 1236] Overall Loss 0.252019 Objective Loss 0.252019 LR 0.001000 Time 0.023830 +2023-10-02 21:09:46,331 - Epoch: [85][ 300/ 1236] Overall Loss 0.250895 Objective Loss 0.250895 LR 0.001000 Time 0.023731 +2023-10-02 21:09:46,536 - Epoch: [85][ 310/ 1236] Overall Loss 0.250814 Objective Loss 0.250814 LR 0.001000 Time 0.023627 +2023-10-02 21:09:46,745 - Epoch: [85][ 320/ 1236] Overall Loss 0.250538 Objective Loss 0.250538 LR 0.001000 Time 0.023541 +2023-10-02 21:09:46,951 - Epoch: [85][ 330/ 1236] Overall Loss 0.251146 Objective Loss 0.251146 LR 0.001000 Time 0.023450 +2023-10-02 21:09:47,160 - Epoch: [85][ 340/ 1236] Overall Loss 0.251094 Objective Loss 0.251094 LR 0.001000 Time 0.023374 +2023-10-02 21:09:47,365 - Epoch: [85][ 350/ 1236] Overall Loss 0.251221 Objective Loss 0.251221 LR 0.001000 Time 0.023292 +2023-10-02 21:09:47,574 - Epoch: [85][ 360/ 1236] Overall Loss 0.251940 Objective Loss 0.251940 LR 0.001000 Time 0.023225 +2023-10-02 21:09:47,780 - Epoch: [85][ 370/ 1236] Overall Loss 0.251637 Objective Loss 0.251637 LR 0.001000 Time 0.023153 +2023-10-02 21:09:47,988 - Epoch: [85][ 380/ 1236] Overall Loss 0.251521 Objective Loss 0.251521 LR 0.001000 Time 0.023090 +2023-10-02 21:09:48,195 - Epoch: [85][ 390/ 1236] Overall Loss 0.251696 Objective Loss 0.251696 LR 0.001000 Time 0.023025 +2023-10-02 21:09:48,403 - Epoch: [85][ 400/ 1236] Overall Loss 0.251656 Objective Loss 0.251656 LR 0.001000 Time 0.022968 +2023-10-02 21:09:48,610 - Epoch: [85][ 410/ 1236] Overall Loss 0.251737 Objective Loss 0.251737 LR 0.001000 Time 0.022909 +2023-10-02 21:09:48,818 - Epoch: [85][ 420/ 1236] Overall Loss 0.251499 Objective Loss 0.251499 LR 0.001000 Time 0.022858 +2023-10-02 21:09:49,025 - Epoch: [85][ 430/ 1236] Overall Loss 0.251520 Objective Loss 0.251520 LR 0.001000 Time 0.022804 +2023-10-02 21:09:49,233 - Epoch: [85][ 440/ 1236] Overall Loss 0.251553 Objective Loss 0.251553 LR 0.001000 Time 0.022758 +2023-10-02 21:09:49,440 - Epoch: [85][ 450/ 1236] Overall Loss 0.251080 Objective Loss 0.251080 LR 0.001000 Time 0.022709 +2023-10-02 21:09:49,649 - Epoch: [85][ 460/ 1236] Overall Loss 0.251772 Objective Loss 0.251772 LR 0.001000 Time 0.022670 +2023-10-02 21:09:49,855 - Epoch: [85][ 470/ 1236] Overall Loss 0.251837 Objective Loss 0.251837 LR 0.001000 Time 0.022625 +2023-10-02 21:09:50,063 - Epoch: [85][ 480/ 1236] Overall Loss 0.252261 Objective Loss 0.252261 LR 0.001000 Time 0.022586 +2023-10-02 21:09:50,270 - Epoch: [85][ 490/ 1236] Overall Loss 0.252294 Objective Loss 0.252294 LR 0.001000 Time 0.022544 +2023-10-02 21:09:50,478 - Epoch: [85][ 500/ 1236] Overall Loss 0.252660 Objective Loss 0.252660 LR 0.001000 Time 0.022508 +2023-10-02 21:09:50,685 - Epoch: [85][ 510/ 1236] Overall Loss 0.253415 Objective Loss 0.253415 LR 0.001000 Time 0.022470 +2023-10-02 21:09:50,893 - Epoch: [85][ 520/ 1236] Overall Loss 0.253842 Objective Loss 0.253842 LR 0.001000 Time 0.022437 +2023-10-02 21:09:51,100 - Epoch: [85][ 530/ 1236] Overall Loss 0.254282 Objective Loss 0.254282 LR 0.001000 Time 0.022402 +2023-10-02 21:09:51,308 - Epoch: [85][ 540/ 1236] Overall Loss 0.254782 Objective Loss 0.254782 LR 0.001000 Time 0.022371 +2023-10-02 21:09:51,515 - Epoch: [85][ 550/ 1236] Overall Loss 0.254800 Objective Loss 0.254800 LR 0.001000 Time 0.022338 +2023-10-02 21:09:51,723 - Epoch: [85][ 560/ 1236] Overall Loss 0.254806 Objective Loss 0.254806 LR 0.001000 Time 0.022310 +2023-10-02 21:09:51,930 - Epoch: [85][ 570/ 1236] Overall Loss 0.255167 Objective Loss 0.255167 LR 0.001000 Time 0.022279 +2023-10-02 21:09:52,138 - Epoch: [85][ 580/ 1236] Overall Loss 0.255369 Objective Loss 0.255369 LR 0.001000 Time 0.022254 +2023-10-02 21:09:52,345 - Epoch: [85][ 590/ 1236] Overall Loss 0.255216 Objective Loss 0.255216 LR 0.001000 Time 0.022225 +2023-10-02 21:09:52,553 - Epoch: [85][ 600/ 1236] Overall Loss 0.254806 Objective Loss 0.254806 LR 0.001000 Time 0.022200 +2023-10-02 21:09:52,760 - Epoch: [85][ 610/ 1236] Overall Loss 0.254899 Objective Loss 0.254899 LR 0.001000 Time 0.022173 +2023-10-02 21:09:52,968 - Epoch: [85][ 620/ 1236] Overall Loss 0.255056 Objective Loss 0.255056 LR 0.001000 Time 0.022150 +2023-10-02 21:09:53,175 - Epoch: [85][ 630/ 1236] Overall Loss 0.255380 Objective Loss 0.255380 LR 0.001000 Time 0.022125 +2023-10-02 21:09:53,383 - Epoch: [85][ 640/ 1236] Overall Loss 0.255287 Objective Loss 0.255287 LR 0.001000 Time 0.022104 +2023-10-02 21:09:53,590 - Epoch: [85][ 650/ 1236] Overall Loss 0.255122 Objective Loss 0.255122 LR 0.001000 Time 0.022081 +2023-10-02 21:09:53,798 - Epoch: [85][ 660/ 1236] Overall Loss 0.255137 Objective Loss 0.255137 LR 0.001000 Time 0.022061 +2023-10-02 21:09:54,005 - Epoch: [85][ 670/ 1236] Overall Loss 0.255228 Objective Loss 0.255228 LR 0.001000 Time 0.022038 +2023-10-02 21:09:54,213 - Epoch: [85][ 680/ 1236] Overall Loss 0.255183 Objective Loss 0.255183 LR 0.001000 Time 0.022019 +2023-10-02 21:09:54,420 - Epoch: [85][ 690/ 1236] Overall Loss 0.254958 Objective Loss 0.254958 LR 0.001000 Time 0.021998 +2023-10-02 21:09:54,628 - Epoch: [85][ 700/ 1236] Overall Loss 0.255159 Objective Loss 0.255159 LR 0.001000 Time 0.021981 +2023-10-02 21:09:54,835 - Epoch: [85][ 710/ 1236] Overall Loss 0.255484 Objective Loss 0.255484 LR 0.001000 Time 0.021961 +2023-10-02 21:09:55,043 - Epoch: [85][ 720/ 1236] Overall Loss 0.255529 Objective Loss 0.255529 LR 0.001000 Time 0.021944 +2023-10-02 21:09:55,250 - Epoch: [85][ 730/ 1236] Overall Loss 0.255945 Objective Loss 0.255945 LR 0.001000 Time 0.021925 +2023-10-02 21:09:55,458 - Epoch: [85][ 740/ 1236] Overall Loss 0.256624 Objective Loss 0.256624 LR 0.001000 Time 0.021909 +2023-10-02 21:09:55,665 - Epoch: [85][ 750/ 1236] Overall Loss 0.256973 Objective Loss 0.256973 LR 0.001000 Time 0.021891 +2023-10-02 21:09:55,873 - Epoch: [85][ 760/ 1236] Overall Loss 0.257155 Objective Loss 0.257155 LR 0.001000 Time 0.021876 +2023-10-02 21:09:56,080 - Epoch: [85][ 770/ 1236] Overall Loss 0.257850 Objective Loss 0.257850 LR 0.001000 Time 0.021859 +2023-10-02 21:09:56,288 - Epoch: [85][ 780/ 1236] Overall Loss 0.258759 Objective Loss 0.258759 LR 0.001000 Time 0.021845 +2023-10-02 21:09:56,495 - Epoch: [85][ 790/ 1236] Overall Loss 0.259688 Objective Loss 0.259688 LR 0.001000 Time 0.021829 +2023-10-02 21:09:56,703 - Epoch: [85][ 800/ 1236] Overall Loss 0.259880 Objective Loss 0.259880 LR 0.001000 Time 0.021816 +2023-10-02 21:09:56,911 - Epoch: [85][ 810/ 1236] Overall Loss 0.260347 Objective Loss 0.260347 LR 0.001000 Time 0.021800 +2023-10-02 21:09:57,119 - Epoch: [85][ 820/ 1236] Overall Loss 0.260613 Objective Loss 0.260613 LR 0.001000 Time 0.021787 +2023-10-02 21:09:57,326 - Epoch: [85][ 830/ 1236] Overall Loss 0.260446 Objective Loss 0.260446 LR 0.001000 Time 0.021772 +2023-10-02 21:09:57,534 - Epoch: [85][ 840/ 1236] Overall Loss 0.260260 Objective Loss 0.260260 LR 0.001000 Time 0.021760 +2023-10-02 21:09:57,741 - Epoch: [85][ 850/ 1236] Overall Loss 0.260411 Objective Loss 0.260411 LR 0.001000 Time 0.021746 +2023-10-02 21:09:57,949 - Epoch: [85][ 860/ 1236] Overall Loss 0.260443 Objective Loss 0.260443 LR 0.001000 Time 0.021735 +2023-10-02 21:09:58,156 - Epoch: [85][ 870/ 1236] Overall Loss 0.260311 Objective Loss 0.260311 LR 0.001000 Time 0.021721 +2023-10-02 21:09:58,364 - Epoch: [85][ 880/ 1236] Overall Loss 0.260898 Objective Loss 0.260898 LR 0.001000 Time 0.021710 +2023-10-02 21:09:58,571 - Epoch: [85][ 890/ 1236] Overall Loss 0.260909 Objective Loss 0.260909 LR 0.001000 Time 0.021697 +2023-10-02 21:09:58,779 - Epoch: [85][ 900/ 1236] Overall Loss 0.261224 Objective Loss 0.261224 LR 0.001000 Time 0.021687 +2023-10-02 21:09:58,986 - Epoch: [85][ 910/ 1236] Overall Loss 0.261901 Objective Loss 0.261901 LR 0.001000 Time 0.021675 +2023-10-02 21:09:59,195 - Epoch: [85][ 920/ 1236] Overall Loss 0.262567 Objective Loss 0.262567 LR 0.001000 Time 0.021665 +2023-10-02 21:09:59,401 - Epoch: [85][ 930/ 1236] Overall Loss 0.262720 Objective Loss 0.262720 LR 0.001000 Time 0.021653 +2023-10-02 21:09:59,609 - Epoch: [85][ 940/ 1236] Overall Loss 0.263196 Objective Loss 0.263196 LR 0.001000 Time 0.021643 +2023-10-02 21:09:59,816 - Epoch: [85][ 950/ 1236] Overall Loss 0.263403 Objective Loss 0.263403 LR 0.001000 Time 0.021632 +2023-10-02 21:10:00,024 - Epoch: [85][ 960/ 1236] Overall Loss 0.263875 Objective Loss 0.263875 LR 0.001000 Time 0.021623 +2023-10-02 21:10:00,231 - Epoch: [85][ 970/ 1236] Overall Loss 0.264389 Objective Loss 0.264389 LR 0.001000 Time 0.021613 +2023-10-02 21:10:00,439 - Epoch: [85][ 980/ 1236] Overall Loss 0.264205 Objective Loss 0.264205 LR 0.001000 Time 0.021604 +2023-10-02 21:10:00,646 - Epoch: [85][ 990/ 1236] Overall Loss 0.264461 Objective Loss 0.264461 LR 0.001000 Time 0.021593 +2023-10-02 21:10:00,856 - Epoch: [85][ 1000/ 1236] Overall Loss 0.264364 Objective Loss 0.264364 LR 0.001000 Time 0.021587 +2023-10-02 21:10:01,062 - Epoch: [85][ 1010/ 1236] Overall Loss 0.264768 Objective Loss 0.264768 LR 0.001000 Time 0.021577 +2023-10-02 21:10:01,270 - Epoch: [85][ 1020/ 1236] Overall Loss 0.265124 Objective Loss 0.265124 LR 0.001000 Time 0.021569 +2023-10-02 21:10:01,477 - Epoch: [85][ 1030/ 1236] Overall Loss 0.265370 Objective Loss 0.265370 LR 0.001000 Time 0.021559 +2023-10-02 21:10:01,685 - Epoch: [85][ 1040/ 1236] Overall Loss 0.265895 Objective Loss 0.265895 LR 0.001000 Time 0.021552 +2023-10-02 21:10:01,893 - Epoch: [85][ 1050/ 1236] Overall Loss 0.266114 Objective Loss 0.266114 LR 0.001000 Time 0.021542 +2023-10-02 21:10:02,101 - Epoch: [85][ 1060/ 1236] Overall Loss 0.266031 Objective Loss 0.266031 LR 0.001000 Time 0.021535 +2023-10-02 21:10:02,308 - Epoch: [85][ 1070/ 1236] Overall Loss 0.266365 Objective Loss 0.266365 LR 0.001000 Time 0.021527 +2023-10-02 21:10:02,516 - Epoch: [85][ 1080/ 1236] Overall Loss 0.266321 Objective Loss 0.266321 LR 0.001000 Time 0.021521 +2023-10-02 21:10:02,724 - Epoch: [85][ 1090/ 1236] Overall Loss 0.266289 Objective Loss 0.266289 LR 0.001000 Time 0.021512 +2023-10-02 21:10:02,931 - Epoch: [85][ 1100/ 1236] Overall Loss 0.266378 Objective Loss 0.266378 LR 0.001000 Time 0.021505 +2023-10-02 21:10:03,139 - Epoch: [85][ 1110/ 1236] Overall Loss 0.266392 Objective Loss 0.266392 LR 0.001000 Time 0.021498 +2023-10-02 21:10:03,347 - Epoch: [85][ 1120/ 1236] Overall Loss 0.266501 Objective Loss 0.266501 LR 0.001000 Time 0.021491 +2023-10-02 21:10:03,554 - Epoch: [85][ 1130/ 1236] Overall Loss 0.266463 Objective Loss 0.266463 LR 0.001000 Time 0.021483 +2023-10-02 21:10:03,763 - Epoch: [85][ 1140/ 1236] Overall Loss 0.266720 Objective Loss 0.266720 LR 0.001000 Time 0.021477 +2023-10-02 21:10:03,970 - Epoch: [85][ 1150/ 1236] Overall Loss 0.267012 Objective Loss 0.267012 LR 0.001000 Time 0.021471 +2023-10-02 21:10:04,179 - Epoch: [85][ 1160/ 1236] Overall Loss 0.267445 Objective Loss 0.267445 LR 0.001000 Time 0.021465 +2023-10-02 21:10:04,386 - Epoch: [85][ 1170/ 1236] Overall Loss 0.267555 Objective Loss 0.267555 LR 0.001000 Time 0.021459 +2023-10-02 21:10:04,595 - Epoch: [85][ 1180/ 1236] Overall Loss 0.267981 Objective Loss 0.267981 LR 0.001000 Time 0.021453 +2023-10-02 21:10:04,802 - Epoch: [85][ 1190/ 1236] Overall Loss 0.268082 Objective Loss 0.268082 LR 0.001000 Time 0.021446 +2023-10-02 21:10:05,010 - Epoch: [85][ 1200/ 1236] Overall Loss 0.268127 Objective Loss 0.268127 LR 0.001000 Time 0.021440 +2023-10-02 21:10:05,218 - Epoch: [85][ 1210/ 1236] Overall Loss 0.268310 Objective Loss 0.268310 LR 0.001000 Time 0.021433 +2023-10-02 21:10:05,426 - Epoch: [85][ 1220/ 1236] Overall Loss 0.268467 Objective Loss 0.268467 LR 0.001000 Time 0.021428 +2023-10-02 21:10:05,685 - Epoch: [85][ 1230/ 1236] Overall Loss 0.268557 Objective Loss 0.268557 LR 0.001000 Time 0.021463 +2023-10-02 21:10:05,807 - Epoch: [85][ 1236/ 1236] Overall Loss 0.268821 Objective Loss 0.268821 Top1 84.114053 Top5 98.778004 LR 0.001000 Time 0.021457 +2023-10-02 21:10:05,948 - --- validate (epoch=85)----------- +2023-10-02 21:10:05,949 - 29943 samples (256 per mini-batch) +2023-10-02 21:10:06,457 - Epoch: [85][ 10/ 117] Loss 0.318326 Top1 83.359375 Top5 98.476562 +2023-10-02 21:10:06,615 - Epoch: [85][ 20/ 117] Loss 0.315595 Top1 83.906250 Top5 98.242188 +2023-10-02 21:10:06,779 - Epoch: [85][ 30/ 117] Loss 0.311286 Top1 83.750000 Top5 98.268229 +2023-10-02 21:10:06,938 - Epoch: [85][ 40/ 117] Loss 0.322818 Top1 83.398438 Top5 98.212891 +2023-10-02 21:10:07,100 - Epoch: [85][ 50/ 117] Loss 0.325332 Top1 83.640625 Top5 98.171875 +2023-10-02 21:10:07,258 - Epoch: [85][ 60/ 117] Loss 0.322979 Top1 83.671875 Top5 98.177083 +2023-10-02 21:10:07,421 - Epoch: [85][ 70/ 117] Loss 0.332005 Top1 83.392857 Top5 98.108259 +2023-10-02 21:10:07,580 - Epoch: [85][ 80/ 117] Loss 0.335213 Top1 83.417969 Top5 98.061523 +2023-10-02 21:10:07,738 - Epoch: [85][ 90/ 117] Loss 0.336179 Top1 83.450521 Top5 97.999132 +2023-10-02 21:10:07,889 - Epoch: [85][ 100/ 117] Loss 0.333322 Top1 83.457031 Top5 98.027344 +2023-10-02 21:10:08,046 - Epoch: [85][ 110/ 117] Loss 0.334761 Top1 83.416193 Top5 98.014915 +2023-10-02 21:10:08,134 - Epoch: [85][ 117/ 117] Loss 0.336379 Top1 83.371740 Top5 98.026250 +2023-10-02 21:10:08,275 - ==> Top1: 83.372 Top5: 98.026 Loss: 0.336 + +2023-10-02 21:10:08,275 - ==> Confusion: +[[ 890 1 6 0 11 5 0 0 16 80 3 2 1 4 8 2 3 0 1 0 17] + [ 0 1044 0 1 6 34 1 19 2 0 2 1 0 0 2 3 2 0 6 4 4] + [ 3 1 945 6 5 0 25 14 0 4 7 2 8 3 4 3 1 2 12 1 10] + [ 1 3 13 952 1 6 2 1 11 0 12 0 10 7 29 1 2 4 19 0 15] + [ 20 3 1 0 967 2 0 0 1 15 2 3 3 3 8 5 11 1 1 2 2] + [ 3 39 1 1 4 973 2 28 2 5 3 11 3 18 2 1 1 0 4 3 12] + [ 0 5 19 0 0 5 1134 6 0 0 2 1 0 0 0 2 0 0 4 8 5] + [ 3 16 12 0 5 21 4 1076 1 4 5 11 3 2 0 0 3 0 38 7 7] + [ 16 5 0 1 2 3 0 0 981 37 12 3 4 10 6 1 1 1 3 0 3] + [ 77 3 1 1 2 1 0 1 41 932 3 0 0 36 11 2 2 1 0 1 4] + [ 1 1 5 4 0 1 2 5 14 1 976 3 1 15 6 0 2 0 4 1 11] + [ 0 1 2 0 3 16 0 1 0 2 0 925 37 10 0 1 1 21 0 7 8] + [ 0 1 2 0 1 2 0 1 2 0 3 38 946 8 3 8 2 29 0 5 17] + [ 0 0 3 0 3 9 2 0 14 13 7 6 2 1040 5 2 3 0 0 2 8] + [ 13 0 2 19 4 0 0 0 22 0 10 0 3 5 1001 0 2 2 11 0 7] + [ 0 0 4 1 5 0 3 0 0 1 0 9 6 0 2 1065 10 17 1 3 7] + [ 1 23 0 2 6 4 0 1 3 1 0 3 2 4 3 9 1083 0 0 7 9] + [ 0 0 0 5 0 0 3 0 1 1 0 2 13 1 8 6 1 991 2 1 3] + [ 2 6 3 16 1 0 1 16 2 0 6 0 4 0 7 0 0 0 994 0 10] + [ 0 1 3 3 0 3 14 11 0 0 5 15 4 4 1 2 6 1 1 1070 8] + [ 137 206 131 89 100 158 53 107 148 105 209 110 318 324 129 72 114 66 194 156 4979]] + +2023-10-02 21:10:08,277 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:10:08,277 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:10:08,283 - + +2023-10-02 21:10:08,283 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:10:09,288 - Epoch: [86][ 10/ 1236] Overall Loss 0.246649 Objective Loss 0.246649 LR 0.001000 Time 0.100486 +2023-10-02 21:10:09,494 - Epoch: [86][ 20/ 1236] Overall Loss 0.263681 Objective Loss 0.263681 LR 0.001000 Time 0.060519 +2023-10-02 21:10:09,699 - Epoch: [86][ 30/ 1236] Overall Loss 0.267134 Objective Loss 0.267134 LR 0.001000 Time 0.047142 +2023-10-02 21:10:09,906 - Epoch: [86][ 40/ 1236] Overall Loss 0.263167 Objective Loss 0.263167 LR 0.001000 Time 0.040521 +2023-10-02 21:10:10,111 - Epoch: [86][ 50/ 1236] Overall Loss 0.260973 Objective Loss 0.260973 LR 0.001000 Time 0.036497 +2023-10-02 21:10:10,317 - Epoch: [86][ 60/ 1236] Overall Loss 0.260751 Objective Loss 0.260751 LR 0.001000 Time 0.033852 +2023-10-02 21:10:10,522 - Epoch: [86][ 70/ 1236] Overall Loss 0.255680 Objective Loss 0.255680 LR 0.001000 Time 0.031932 +2023-10-02 21:10:10,728 - Epoch: [86][ 80/ 1236] Overall Loss 0.257152 Objective Loss 0.257152 LR 0.001000 Time 0.030522 +2023-10-02 21:10:10,933 - Epoch: [86][ 90/ 1236] Overall Loss 0.256942 Objective Loss 0.256942 LR 0.001000 Time 0.029397 +2023-10-02 21:10:11,140 - Epoch: [86][ 100/ 1236] Overall Loss 0.255145 Objective Loss 0.255145 LR 0.001000 Time 0.028525 +2023-10-02 21:10:11,344 - Epoch: [86][ 110/ 1236] Overall Loss 0.256184 Objective Loss 0.256184 LR 0.001000 Time 0.027785 +2023-10-02 21:10:11,551 - Epoch: [86][ 120/ 1236] Overall Loss 0.258228 Objective Loss 0.258228 LR 0.001000 Time 0.027193 +2023-10-02 21:10:11,755 - Epoch: [86][ 130/ 1236] Overall Loss 0.258900 Objective Loss 0.258900 LR 0.001000 Time 0.026671 +2023-10-02 21:10:11,962 - Epoch: [86][ 140/ 1236] Overall Loss 0.260346 Objective Loss 0.260346 LR 0.001000 Time 0.026242 +2023-10-02 21:10:12,166 - Epoch: [86][ 150/ 1236] Overall Loss 0.260490 Objective Loss 0.260490 LR 0.001000 Time 0.025852 +2023-10-02 21:10:12,372 - Epoch: [86][ 160/ 1236] Overall Loss 0.262615 Objective Loss 0.262615 LR 0.001000 Time 0.025522 +2023-10-02 21:10:12,576 - Epoch: [86][ 170/ 1236] Overall Loss 0.261758 Objective Loss 0.261758 LR 0.001000 Time 0.025219 +2023-10-02 21:10:12,783 - Epoch: [86][ 180/ 1236] Overall Loss 0.261691 Objective Loss 0.261691 LR 0.001000 Time 0.024966 +2023-10-02 21:10:12,987 - Epoch: [86][ 190/ 1236] Overall Loss 0.263833 Objective Loss 0.263833 LR 0.001000 Time 0.024724 +2023-10-02 21:10:13,194 - Epoch: [86][ 200/ 1236] Overall Loss 0.264357 Objective Loss 0.264357 LR 0.001000 Time 0.024520 +2023-10-02 21:10:13,398 - Epoch: [86][ 210/ 1236] Overall Loss 0.266269 Objective Loss 0.266269 LR 0.001000 Time 0.024324 +2023-10-02 21:10:13,605 - Epoch: [86][ 220/ 1236] Overall Loss 0.267358 Objective Loss 0.267358 LR 0.001000 Time 0.024156 +2023-10-02 21:10:13,809 - Epoch: [86][ 230/ 1236] Overall Loss 0.267098 Objective Loss 0.267098 LR 0.001000 Time 0.023993 +2023-10-02 21:10:14,016 - Epoch: [86][ 240/ 1236] Overall Loss 0.267681 Objective Loss 0.267681 LR 0.001000 Time 0.023853 +2023-10-02 21:10:14,220 - Epoch: [86][ 250/ 1236] Overall Loss 0.267745 Objective Loss 0.267745 LR 0.001000 Time 0.023716 +2023-10-02 21:10:14,426 - Epoch: [86][ 260/ 1236] Overall Loss 0.267138 Objective Loss 0.267138 LR 0.001000 Time 0.023594 +2023-10-02 21:10:14,632 - Epoch: [86][ 270/ 1236] Overall Loss 0.267080 Objective Loss 0.267080 LR 0.001000 Time 0.023475 +2023-10-02 21:10:14,839 - Epoch: [86][ 280/ 1236] Overall Loss 0.267355 Objective Loss 0.267355 LR 0.001000 Time 0.023376 +2023-10-02 21:10:15,042 - Epoch: [86][ 290/ 1236] Overall Loss 0.268518 Objective Loss 0.268518 LR 0.001000 Time 0.023269 +2023-10-02 21:10:15,249 - Epoch: [86][ 300/ 1236] Overall Loss 0.268961 Objective Loss 0.268961 LR 0.001000 Time 0.023183 +2023-10-02 21:10:15,453 - Epoch: [86][ 310/ 1236] Overall Loss 0.268921 Objective Loss 0.268921 LR 0.001000 Time 0.023093 +2023-10-02 21:10:15,660 - Epoch: [86][ 320/ 1236] Overall Loss 0.268142 Objective Loss 0.268142 LR 0.001000 Time 0.023016 +2023-10-02 21:10:15,865 - Epoch: [86][ 330/ 1236] Overall Loss 0.268105 Objective Loss 0.268105 LR 0.001000 Time 0.022938 +2023-10-02 21:10:16,072 - Epoch: [86][ 340/ 1236] Overall Loss 0.267629 Objective Loss 0.267629 LR 0.001000 Time 0.022871 +2023-10-02 21:10:16,276 - Epoch: [86][ 350/ 1236] Overall Loss 0.268230 Objective Loss 0.268230 LR 0.001000 Time 0.022800 +2023-10-02 21:10:16,482 - Epoch: [86][ 360/ 1236] Overall Loss 0.268733 Objective Loss 0.268733 LR 0.001000 Time 0.022739 +2023-10-02 21:10:16,688 - Epoch: [86][ 370/ 1236] Overall Loss 0.268717 Objective Loss 0.268717 LR 0.001000 Time 0.022676 +2023-10-02 21:10:16,893 - Epoch: [86][ 380/ 1236] Overall Loss 0.270277 Objective Loss 0.270277 LR 0.001000 Time 0.022620 +2023-10-02 21:10:17,098 - Epoch: [86][ 390/ 1236] Overall Loss 0.270025 Objective Loss 0.270025 LR 0.001000 Time 0.022564 +2023-10-02 21:10:17,305 - Epoch: [86][ 400/ 1236] Overall Loss 0.270585 Objective Loss 0.270585 LR 0.001000 Time 0.022518 +2023-10-02 21:10:17,510 - Epoch: [86][ 410/ 1236] Overall Loss 0.270680 Objective Loss 0.270680 LR 0.001000 Time 0.022466 +2023-10-02 21:10:17,716 - Epoch: [86][ 420/ 1236] Overall Loss 0.270759 Objective Loss 0.270759 LR 0.001000 Time 0.022421 +2023-10-02 21:10:17,922 - Epoch: [86][ 430/ 1236] Overall Loss 0.270985 Objective Loss 0.270985 LR 0.001000 Time 0.022378 +2023-10-02 21:10:18,128 - Epoch: [86][ 440/ 1236] Overall Loss 0.270854 Objective Loss 0.270854 LR 0.001000 Time 0.022337 +2023-10-02 21:10:18,334 - Epoch: [86][ 450/ 1236] Overall Loss 0.270309 Objective Loss 0.270309 LR 0.001000 Time 0.022297 +2023-10-02 21:10:18,539 - Epoch: [86][ 460/ 1236] Overall Loss 0.270462 Objective Loss 0.270462 LR 0.001000 Time 0.022259 +2023-10-02 21:10:18,745 - Epoch: [86][ 470/ 1236] Overall Loss 0.270664 Objective Loss 0.270664 LR 0.001000 Time 0.022220 +2023-10-02 21:10:18,951 - Epoch: [86][ 480/ 1236] Overall Loss 0.271177 Objective Loss 0.271177 LR 0.001000 Time 0.022185 +2023-10-02 21:10:19,157 - Epoch: [86][ 490/ 1236] Overall Loss 0.271733 Objective Loss 0.271733 LR 0.001000 Time 0.022152 +2023-10-02 21:10:19,363 - Epoch: [86][ 500/ 1236] Overall Loss 0.272388 Objective Loss 0.272388 LR 0.001000 Time 0.022120 +2023-10-02 21:10:19,568 - Epoch: [86][ 510/ 1236] Overall Loss 0.272013 Objective Loss 0.272013 LR 0.001000 Time 0.022086 +2023-10-02 21:10:19,774 - Epoch: [86][ 520/ 1236] Overall Loss 0.272514 Objective Loss 0.272514 LR 0.001000 Time 0.022057 +2023-10-02 21:10:19,980 - Epoch: [86][ 530/ 1236] Overall Loss 0.272319 Objective Loss 0.272319 LR 0.001000 Time 0.022026 +2023-10-02 21:10:20,187 - Epoch: [86][ 540/ 1236] Overall Loss 0.271857 Objective Loss 0.271857 LR 0.001000 Time 0.022002 +2023-10-02 21:10:20,392 - Epoch: [86][ 550/ 1236] Overall Loss 0.272362 Objective Loss 0.272362 LR 0.001000 Time 0.021974 +2023-10-02 21:10:20,599 - Epoch: [86][ 560/ 1236] Overall Loss 0.272529 Objective Loss 0.272529 LR 0.001000 Time 0.021951 +2023-10-02 21:10:20,803 - Epoch: [86][ 570/ 1236] Overall Loss 0.272662 Objective Loss 0.272662 LR 0.001000 Time 0.021924 +2023-10-02 21:10:21,011 - Epoch: [86][ 580/ 1236] Overall Loss 0.273170 Objective Loss 0.273170 LR 0.001000 Time 0.021903 +2023-10-02 21:10:21,215 - Epoch: [86][ 590/ 1236] Overall Loss 0.273934 Objective Loss 0.273934 LR 0.001000 Time 0.021878 +2023-10-02 21:10:21,422 - Epoch: [86][ 600/ 1236] Overall Loss 0.274812 Objective Loss 0.274812 LR 0.001000 Time 0.021858 +2023-10-02 21:10:21,627 - Epoch: [86][ 610/ 1236] Overall Loss 0.275155 Objective Loss 0.275155 LR 0.001000 Time 0.021834 +2023-10-02 21:10:21,833 - Epoch: [86][ 620/ 1236] Overall Loss 0.275439 Objective Loss 0.275439 LR 0.001000 Time 0.021814 +2023-10-02 21:10:22,039 - Epoch: [86][ 630/ 1236] Overall Loss 0.275519 Objective Loss 0.275519 LR 0.001000 Time 0.021794 +2023-10-02 21:10:22,246 - Epoch: [86][ 640/ 1236] Overall Loss 0.275604 Objective Loss 0.275604 LR 0.001000 Time 0.021777 +2023-10-02 21:10:22,451 - Epoch: [86][ 650/ 1236] Overall Loss 0.275891 Objective Loss 0.275891 LR 0.001000 Time 0.021756 +2023-10-02 21:10:22,657 - Epoch: [86][ 660/ 1236] Overall Loss 0.276450 Objective Loss 0.276450 LR 0.001000 Time 0.021738 +2023-10-02 21:10:22,863 - Epoch: [86][ 670/ 1236] Overall Loss 0.276488 Objective Loss 0.276488 LR 0.001000 Time 0.021719 +2023-10-02 21:10:23,069 - Epoch: [86][ 680/ 1236] Overall Loss 0.276629 Objective Loss 0.276629 LR 0.001000 Time 0.021702 +2023-10-02 21:10:23,275 - Epoch: [86][ 690/ 1236] Overall Loss 0.276516 Objective Loss 0.276516 LR 0.001000 Time 0.021686 +2023-10-02 21:10:23,481 - Epoch: [86][ 700/ 1236] Overall Loss 0.276936 Objective Loss 0.276936 LR 0.001000 Time 0.021670 +2023-10-02 21:10:23,686 - Epoch: [86][ 710/ 1236] Overall Loss 0.276703 Objective Loss 0.276703 LR 0.001000 Time 0.021653 +2023-10-02 21:10:23,893 - Epoch: [86][ 720/ 1236] Overall Loss 0.277026 Objective Loss 0.277026 LR 0.001000 Time 0.021639 +2023-10-02 21:10:24,097 - Epoch: [86][ 730/ 1236] Overall Loss 0.277102 Objective Loss 0.277102 LR 0.001000 Time 0.021622 +2023-10-02 21:10:24,304 - Epoch: [86][ 740/ 1236] Overall Loss 0.276990 Objective Loss 0.276990 LR 0.001000 Time 0.021609 +2023-10-02 21:10:24,508 - Epoch: [86][ 750/ 1236] Overall Loss 0.276848 Objective Loss 0.276848 LR 0.001000 Time 0.021593 +2023-10-02 21:10:24,715 - Epoch: [86][ 760/ 1236] Overall Loss 0.276983 Objective Loss 0.276983 LR 0.001000 Time 0.021581 +2023-10-02 21:10:24,920 - Epoch: [86][ 770/ 1236] Overall Loss 0.276938 Objective Loss 0.276938 LR 0.001000 Time 0.021566 +2023-10-02 21:10:25,126 - Epoch: [86][ 780/ 1236] Overall Loss 0.276906 Objective Loss 0.276906 LR 0.001000 Time 0.021554 +2023-10-02 21:10:25,331 - Epoch: [86][ 790/ 1236] Overall Loss 0.276860 Objective Loss 0.276860 LR 0.001000 Time 0.021539 +2023-10-02 21:10:25,537 - Epoch: [86][ 800/ 1236] Overall Loss 0.277078 Objective Loss 0.277078 LR 0.001000 Time 0.021528 +2023-10-02 21:10:25,742 - Epoch: [86][ 810/ 1236] Overall Loss 0.277079 Objective Loss 0.277079 LR 0.001000 Time 0.021514 +2023-10-02 21:10:25,948 - Epoch: [86][ 820/ 1236] Overall Loss 0.277394 Objective Loss 0.277394 LR 0.001000 Time 0.021504 +2023-10-02 21:10:26,153 - Epoch: [86][ 830/ 1236] Overall Loss 0.277492 Objective Loss 0.277492 LR 0.001000 Time 0.021491 +2023-10-02 21:10:26,359 - Epoch: [86][ 840/ 1236] Overall Loss 0.277384 Objective Loss 0.277384 LR 0.001000 Time 0.021480 +2023-10-02 21:10:26,565 - Epoch: [86][ 850/ 1236] Overall Loss 0.277173 Objective Loss 0.277173 LR 0.001000 Time 0.021467 +2023-10-02 21:10:26,770 - Epoch: [86][ 860/ 1236] Overall Loss 0.276921 Objective Loss 0.276921 LR 0.001000 Time 0.021456 +2023-10-02 21:10:26,976 - Epoch: [86][ 870/ 1236] Overall Loss 0.276923 Objective Loss 0.276923 LR 0.001000 Time 0.021444 +2023-10-02 21:10:27,183 - Epoch: [86][ 880/ 1236] Overall Loss 0.276694 Objective Loss 0.276694 LR 0.001000 Time 0.021436 +2023-10-02 21:10:27,387 - Epoch: [86][ 890/ 1236] Overall Loss 0.276712 Objective Loss 0.276712 LR 0.001000 Time 0.021424 +2023-10-02 21:10:27,593 - Epoch: [86][ 900/ 1236] Overall Loss 0.276608 Objective Loss 0.276608 LR 0.001000 Time 0.021414 +2023-10-02 21:10:27,799 - Epoch: [86][ 910/ 1236] Overall Loss 0.276404 Objective Loss 0.276404 LR 0.001000 Time 0.021404 +2023-10-02 21:10:28,006 - Epoch: [86][ 920/ 1236] Overall Loss 0.276265 Objective Loss 0.276265 LR 0.001000 Time 0.021396 +2023-10-02 21:10:28,210 - Epoch: [86][ 930/ 1236] Overall Loss 0.276051 Objective Loss 0.276051 LR 0.001000 Time 0.021385 +2023-10-02 21:10:28,417 - Epoch: [86][ 940/ 1236] Overall Loss 0.275676 Objective Loss 0.275676 LR 0.001000 Time 0.021378 +2023-10-02 21:10:28,622 - Epoch: [86][ 950/ 1236] Overall Loss 0.275670 Objective Loss 0.275670 LR 0.001000 Time 0.021367 +2023-10-02 21:10:28,828 - Epoch: [86][ 960/ 1236] Overall Loss 0.275682 Objective Loss 0.275682 LR 0.001000 Time 0.021360 +2023-10-02 21:10:29,033 - Epoch: [86][ 970/ 1236] Overall Loss 0.275592 Objective Loss 0.275592 LR 0.001000 Time 0.021350 +2023-10-02 21:10:29,240 - Epoch: [86][ 980/ 1236] Overall Loss 0.275603 Objective Loss 0.275603 LR 0.001000 Time 0.021343 +2023-10-02 21:10:29,444 - Epoch: [86][ 990/ 1236] Overall Loss 0.275581 Objective Loss 0.275581 LR 0.001000 Time 0.021333 +2023-10-02 21:10:29,650 - Epoch: [86][ 1000/ 1236] Overall Loss 0.275702 Objective Loss 0.275702 LR 0.001000 Time 0.021326 +2023-10-02 21:10:29,856 - Epoch: [86][ 1010/ 1236] Overall Loss 0.275773 Objective Loss 0.275773 LR 0.001000 Time 0.021317 +2023-10-02 21:10:30,063 - Epoch: [86][ 1020/ 1236] Overall Loss 0.275772 Objective Loss 0.275772 LR 0.001000 Time 0.021311 +2023-10-02 21:10:30,267 - Epoch: [86][ 1030/ 1236] Overall Loss 0.275831 Objective Loss 0.275831 LR 0.001000 Time 0.021302 +2023-10-02 21:10:30,474 - Epoch: [86][ 1040/ 1236] Overall Loss 0.276083 Objective Loss 0.276083 LR 0.001000 Time 0.021296 +2023-10-02 21:10:30,679 - Epoch: [86][ 1050/ 1236] Overall Loss 0.275965 Objective Loss 0.275965 LR 0.001000 Time 0.021288 +2023-10-02 21:10:30,886 - Epoch: [86][ 1060/ 1236] Overall Loss 0.275728 Objective Loss 0.275728 LR 0.001000 Time 0.021282 +2023-10-02 21:10:31,091 - Epoch: [86][ 1070/ 1236] Overall Loss 0.275684 Objective Loss 0.275684 LR 0.001000 Time 0.021274 +2023-10-02 21:10:31,297 - Epoch: [86][ 1080/ 1236] Overall Loss 0.275551 Objective Loss 0.275551 LR 0.001000 Time 0.021268 +2023-10-02 21:10:31,502 - Epoch: [86][ 1090/ 1236] Overall Loss 0.275490 Objective Loss 0.275490 LR 0.001000 Time 0.021261 +2023-10-02 21:10:31,709 - Epoch: [86][ 1100/ 1236] Overall Loss 0.275402 Objective Loss 0.275402 LR 0.001000 Time 0.021255 +2023-10-02 21:10:31,914 - Epoch: [86][ 1110/ 1236] Overall Loss 0.275415 Objective Loss 0.275415 LR 0.001000 Time 0.021248 +2023-10-02 21:10:32,122 - Epoch: [86][ 1120/ 1236] Overall Loss 0.275169 Objective Loss 0.275169 LR 0.001000 Time 0.021243 +2023-10-02 21:10:32,326 - Epoch: [86][ 1130/ 1236] Overall Loss 0.275425 Objective Loss 0.275425 LR 0.001000 Time 0.021236 +2023-10-02 21:10:32,533 - Epoch: [86][ 1140/ 1236] Overall Loss 0.275619 Objective Loss 0.275619 LR 0.001000 Time 0.021231 +2023-10-02 21:10:32,738 - Epoch: [86][ 1150/ 1236] Overall Loss 0.275448 Objective Loss 0.275448 LR 0.001000 Time 0.021224 +2023-10-02 21:10:32,945 - Epoch: [86][ 1160/ 1236] Overall Loss 0.275288 Objective Loss 0.275288 LR 0.001000 Time 0.021219 +2023-10-02 21:10:33,150 - Epoch: [86][ 1170/ 1236] Overall Loss 0.275634 Objective Loss 0.275634 LR 0.001000 Time 0.021213 +2023-10-02 21:10:33,357 - Epoch: [86][ 1180/ 1236] Overall Loss 0.275422 Objective Loss 0.275422 LR 0.001000 Time 0.021208 +2023-10-02 21:10:33,561 - Epoch: [86][ 1190/ 1236] Overall Loss 0.275345 Objective Loss 0.275345 LR 0.001000 Time 0.021201 +2023-10-02 21:10:33,768 - Epoch: [86][ 1200/ 1236] Overall Loss 0.275322 Objective Loss 0.275322 LR 0.001000 Time 0.021197 +2023-10-02 21:10:33,973 - Epoch: [86][ 1210/ 1236] Overall Loss 0.275181 Objective Loss 0.275181 LR 0.001000 Time 0.021191 +2023-10-02 21:10:34,178 - Epoch: [86][ 1220/ 1236] Overall Loss 0.275250 Objective Loss 0.275250 LR 0.001000 Time 0.021185 +2023-10-02 21:10:34,436 - Epoch: [86][ 1230/ 1236] Overall Loss 0.275429 Objective Loss 0.275429 LR 0.001000 Time 0.021221 +2023-10-02 21:10:34,557 - Epoch: [86][ 1236/ 1236] Overall Loss 0.275522 Objective Loss 0.275522 Top1 87.169043 Top5 98.370672 LR 0.001000 Time 0.021216 +2023-10-02 21:10:34,691 - --- validate (epoch=86)----------- +2023-10-02 21:10:34,692 - 29943 samples (256 per mini-batch) +2023-10-02 21:10:35,171 - Epoch: [86][ 10/ 117] Loss 0.372785 Top1 83.632812 Top5 98.359375 +2023-10-02 21:10:35,322 - Epoch: [86][ 20/ 117] Loss 0.354750 Top1 84.199219 Top5 98.222656 +2023-10-02 21:10:35,479 - Epoch: [86][ 30/ 117] Loss 0.362807 Top1 83.880208 Top5 98.059896 +2023-10-02 21:10:35,640 - Epoch: [86][ 40/ 117] Loss 0.352922 Top1 84.121094 Top5 98.027344 +2023-10-02 21:10:35,797 - Epoch: [86][ 50/ 117] Loss 0.340924 Top1 84.304688 Top5 98.085938 +2023-10-02 21:10:35,958 - Epoch: [86][ 60/ 117] Loss 0.336429 Top1 84.375000 Top5 98.131510 +2023-10-02 21:10:36,114 - Epoch: [86][ 70/ 117] Loss 0.334719 Top1 84.358259 Top5 98.125000 +2023-10-02 21:10:36,275 - Epoch: [86][ 80/ 117] Loss 0.336324 Top1 84.248047 Top5 98.149414 +2023-10-02 21:10:36,431 - Epoch: [86][ 90/ 117] Loss 0.340242 Top1 84.079861 Top5 98.129340 +2023-10-02 21:10:36,582 - Epoch: [86][ 100/ 117] Loss 0.339932 Top1 84.042969 Top5 98.140625 +2023-10-02 21:10:36,738 - Epoch: [86][ 110/ 117] Loss 0.338181 Top1 84.051847 Top5 98.160511 +2023-10-02 21:10:36,827 - Epoch: [86][ 117/ 117] Loss 0.337974 Top1 83.972882 Top5 98.196573 +2023-10-02 21:10:36,960 - ==> Top1: 83.973 Top5: 98.197 Loss: 0.338 + +2023-10-02 21:10:36,961 - ==> Confusion: +[[ 898 1 6 1 5 1 0 0 13 86 1 1 1 3 7 3 6 1 1 0 15] + [ 0 1031 2 1 4 20 0 23 2 1 1 0 1 0 3 5 8 0 12 4 13] + [ 9 0 944 17 0 0 35 9 0 1 1 1 7 1 1 4 1 1 11 4 9] + [ 3 2 13 951 0 2 1 3 8 0 10 0 7 7 35 1 1 9 13 0 23] + [ 19 8 2 0 950 4 0 0 1 21 1 3 1 2 7 9 15 0 1 1 5] + [ 1 41 1 2 5 947 3 28 2 5 2 11 4 18 1 1 7 2 8 6 21] + [ 1 0 24 0 0 1 1124 6 0 1 2 1 0 1 0 2 0 2 2 14 10] + [ 1 14 24 0 2 27 7 1052 0 5 5 7 7 6 3 1 2 2 36 9 8] + [ 19 1 1 1 1 2 0 1 949 48 16 2 2 11 24 2 2 2 3 1 1] + [ 69 1 0 1 5 0 0 0 16 971 1 0 0 29 10 2 0 1 0 3 10] + [ 0 2 9 9 0 2 5 5 16 1 954 3 0 13 5 2 2 1 6 2 16] + [ 0 0 4 1 1 6 1 1 1 2 1 926 42 13 0 4 4 14 0 7 7] + [ 1 1 2 6 1 1 4 2 3 0 2 34 958 5 5 8 3 16 1 7 8] + [ 2 0 2 0 2 4 1 0 4 19 3 2 1 1056 6 0 4 0 0 3 10] + [ 10 1 4 9 4 1 0 0 8 3 2 0 1 4 1040 0 1 2 7 0 4] + [ 0 0 1 0 2 0 3 0 0 0 1 6 9 0 2 1073 16 9 0 5 7] + [ 2 11 1 0 1 6 2 0 2 2 0 6 0 3 2 11 1088 2 2 7 13] + [ 0 0 2 2 1 0 3 0 0 0 0 6 40 1 8 8 2 960 1 1 3] + [ 3 4 4 11 0 0 0 13 3 0 5 1 1 0 16 0 1 0 987 1 18] + [ 0 2 5 2 2 2 6 5 0 0 2 11 4 2 0 6 7 1 2 1082 11] + [ 127 163 116 80 52 113 53 99 79 107 176 104 329 321 168 55 164 51 129 216 5203]] + +2023-10-02 21:10:36,962 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:10:36,962 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:10:36,969 - + +2023-10-02 21:10:36,969 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:10:37,976 - Epoch: [87][ 10/ 1236] Overall Loss 0.248055 Objective Loss 0.248055 LR 0.001000 Time 0.100677 +2023-10-02 21:10:38,185 - Epoch: [87][ 20/ 1236] Overall Loss 0.258051 Objective Loss 0.258051 LR 0.001000 Time 0.060748 +2023-10-02 21:10:38,392 - Epoch: [87][ 30/ 1236] Overall Loss 0.263890 Objective Loss 0.263890 LR 0.001000 Time 0.047386 +2023-10-02 21:10:38,601 - Epoch: [87][ 40/ 1236] Overall Loss 0.266439 Objective Loss 0.266439 LR 0.001000 Time 0.040763 +2023-10-02 21:10:38,807 - Epoch: [87][ 50/ 1236] Overall Loss 0.264137 Objective Loss 0.264137 LR 0.001000 Time 0.036715 +2023-10-02 21:10:39,016 - Epoch: [87][ 60/ 1236] Overall Loss 0.257922 Objective Loss 0.257922 LR 0.001000 Time 0.034075 +2023-10-02 21:10:39,221 - Epoch: [87][ 70/ 1236] Overall Loss 0.256798 Objective Loss 0.256798 LR 0.001000 Time 0.032138 +2023-10-02 21:10:39,429 - Epoch: [87][ 80/ 1236] Overall Loss 0.259085 Objective Loss 0.259085 LR 0.001000 Time 0.030722 +2023-10-02 21:10:39,636 - Epoch: [87][ 90/ 1236] Overall Loss 0.258851 Objective Loss 0.258851 LR 0.001000 Time 0.029601 +2023-10-02 21:10:39,846 - Epoch: [87][ 100/ 1236] Overall Loss 0.257663 Objective Loss 0.257663 LR 0.001000 Time 0.028732 +2023-10-02 21:10:40,051 - Epoch: [87][ 110/ 1236] Overall Loss 0.258681 Objective Loss 0.258681 LR 0.001000 Time 0.027987 +2023-10-02 21:10:40,259 - Epoch: [87][ 120/ 1236] Overall Loss 0.259787 Objective Loss 0.259787 LR 0.001000 Time 0.027387 +2023-10-02 21:10:40,466 - Epoch: [87][ 130/ 1236] Overall Loss 0.262380 Objective Loss 0.262380 LR 0.001000 Time 0.026860 +2023-10-02 21:10:40,674 - Epoch: [87][ 140/ 1236] Overall Loss 0.263289 Objective Loss 0.263289 LR 0.001000 Time 0.026426 +2023-10-02 21:10:40,881 - Epoch: [87][ 150/ 1236] Overall Loss 0.261804 Objective Loss 0.261804 LR 0.001000 Time 0.026034 +2023-10-02 21:10:41,089 - Epoch: [87][ 160/ 1236] Overall Loss 0.264250 Objective Loss 0.264250 LR 0.001000 Time 0.025706 +2023-10-02 21:10:41,296 - Epoch: [87][ 170/ 1236] Overall Loss 0.265708 Objective Loss 0.265708 LR 0.001000 Time 0.025402 +2023-10-02 21:10:41,504 - Epoch: [87][ 180/ 1236] Overall Loss 0.266274 Objective Loss 0.266274 LR 0.001000 Time 0.025144 +2023-10-02 21:10:41,708 - Epoch: [87][ 190/ 1236] Overall Loss 0.268619 Objective Loss 0.268619 LR 0.001000 Time 0.024894 +2023-10-02 21:10:41,917 - Epoch: [87][ 200/ 1236] Overall Loss 0.267666 Objective Loss 0.267666 LR 0.001000 Time 0.024685 +2023-10-02 21:10:42,121 - Epoch: [87][ 210/ 1236] Overall Loss 0.268026 Objective Loss 0.268026 LR 0.001000 Time 0.024478 +2023-10-02 21:10:42,328 - Epoch: [87][ 220/ 1236] Overall Loss 0.267689 Objective Loss 0.267689 LR 0.001000 Time 0.024307 +2023-10-02 21:10:42,533 - Epoch: [87][ 230/ 1236] Overall Loss 0.266098 Objective Loss 0.266098 LR 0.001000 Time 0.024137 +2023-10-02 21:10:42,738 - Epoch: [87][ 240/ 1236] Overall Loss 0.264970 Objective Loss 0.264970 LR 0.001000 Time 0.023986 +2023-10-02 21:10:42,942 - Epoch: [87][ 250/ 1236] Overall Loss 0.263475 Objective Loss 0.263475 LR 0.001000 Time 0.023836 +2023-10-02 21:10:43,149 - Epoch: [87][ 260/ 1236] Overall Loss 0.261690 Objective Loss 0.261690 LR 0.001000 Time 0.023713 +2023-10-02 21:10:43,354 - Epoch: [87][ 270/ 1236] Overall Loss 0.261937 Objective Loss 0.261937 LR 0.001000 Time 0.023594 +2023-10-02 21:10:43,562 - Epoch: [87][ 280/ 1236] Overall Loss 0.261852 Objective Loss 0.261852 LR 0.001000 Time 0.023494 +2023-10-02 21:10:43,767 - Epoch: [87][ 290/ 1236] Overall Loss 0.263061 Objective Loss 0.263061 LR 0.001000 Time 0.023389 +2023-10-02 21:10:43,975 - Epoch: [87][ 300/ 1236] Overall Loss 0.262329 Objective Loss 0.262329 LR 0.001000 Time 0.023304 +2023-10-02 21:10:44,180 - Epoch: [87][ 310/ 1236] Overall Loss 0.263384 Objective Loss 0.263384 LR 0.001000 Time 0.023212 +2023-10-02 21:10:44,387 - Epoch: [87][ 320/ 1236] Overall Loss 0.263792 Objective Loss 0.263792 LR 0.001000 Time 0.023132 +2023-10-02 21:10:44,592 - Epoch: [87][ 330/ 1236] Overall Loss 0.263437 Objective Loss 0.263437 LR 0.001000 Time 0.023051 +2023-10-02 21:10:44,801 - Epoch: [87][ 340/ 1236] Overall Loss 0.263229 Objective Loss 0.263229 LR 0.001000 Time 0.022985 +2023-10-02 21:10:45,005 - Epoch: [87][ 350/ 1236] Overall Loss 0.263509 Objective Loss 0.263509 LR 0.001000 Time 0.022912 +2023-10-02 21:10:45,214 - Epoch: [87][ 360/ 1236] Overall Loss 0.263210 Objective Loss 0.263210 LR 0.001000 Time 0.022854 +2023-10-02 21:10:45,418 - Epoch: [87][ 370/ 1236] Overall Loss 0.262984 Objective Loss 0.262984 LR 0.001000 Time 0.022789 +2023-10-02 21:10:45,627 - Epoch: [87][ 380/ 1236] Overall Loss 0.263085 Objective Loss 0.263085 LR 0.001000 Time 0.022737 +2023-10-02 21:10:45,832 - Epoch: [87][ 390/ 1236] Overall Loss 0.262586 Objective Loss 0.262586 LR 0.001000 Time 0.022679 +2023-10-02 21:10:46,040 - Epoch: [87][ 400/ 1236] Overall Loss 0.262545 Objective Loss 0.262545 LR 0.001000 Time 0.022633 +2023-10-02 21:10:46,245 - Epoch: [87][ 410/ 1236] Overall Loss 0.262422 Objective Loss 0.262422 LR 0.001000 Time 0.022579 +2023-10-02 21:10:46,454 - Epoch: [87][ 420/ 1236] Overall Loss 0.261798 Objective Loss 0.261798 LR 0.001000 Time 0.022538 +2023-10-02 21:10:46,658 - Epoch: [87][ 430/ 1236] Overall Loss 0.261824 Objective Loss 0.261824 LR 0.001000 Time 0.022489 +2023-10-02 21:10:46,867 - Epoch: [87][ 440/ 1236] Overall Loss 0.262265 Objective Loss 0.262265 LR 0.001000 Time 0.022452 +2023-10-02 21:10:47,072 - Epoch: [87][ 450/ 1236] Overall Loss 0.262652 Objective Loss 0.262652 LR 0.001000 Time 0.022407 +2023-10-02 21:10:47,280 - Epoch: [87][ 460/ 1236] Overall Loss 0.262485 Objective Loss 0.262485 LR 0.001000 Time 0.022373 +2023-10-02 21:10:47,485 - Epoch: [87][ 470/ 1236] Overall Loss 0.262581 Objective Loss 0.262581 LR 0.001000 Time 0.022332 +2023-10-02 21:10:47,694 - Epoch: [87][ 480/ 1236] Overall Loss 0.262188 Objective Loss 0.262188 LR 0.001000 Time 0.022301 +2023-10-02 21:10:47,899 - Epoch: [87][ 490/ 1236] Overall Loss 0.262021 Objective Loss 0.262021 LR 0.001000 Time 0.022263 +2023-10-02 21:10:48,108 - Epoch: [87][ 500/ 1236] Overall Loss 0.261847 Objective Loss 0.261847 LR 0.001000 Time 0.022235 +2023-10-02 21:10:48,312 - Epoch: [87][ 510/ 1236] Overall Loss 0.262276 Objective Loss 0.262276 LR 0.001000 Time 0.022200 +2023-10-02 21:10:48,521 - Epoch: [87][ 520/ 1236] Overall Loss 0.262226 Objective Loss 0.262226 LR 0.001000 Time 0.022174 +2023-10-02 21:10:48,726 - Epoch: [87][ 530/ 1236] Overall Loss 0.262105 Objective Loss 0.262105 LR 0.001000 Time 0.022142 +2023-10-02 21:10:48,935 - Epoch: [87][ 540/ 1236] Overall Loss 0.261602 Objective Loss 0.261602 LR 0.001000 Time 0.022117 +2023-10-02 21:10:49,139 - Epoch: [87][ 550/ 1236] Overall Loss 0.261294 Objective Loss 0.261294 LR 0.001000 Time 0.022087 +2023-10-02 21:10:49,348 - Epoch: [87][ 560/ 1236] Overall Loss 0.261724 Objective Loss 0.261724 LR 0.001000 Time 0.022065 +2023-10-02 21:10:49,553 - Epoch: [87][ 570/ 1236] Overall Loss 0.261620 Objective Loss 0.261620 LR 0.001000 Time 0.022037 +2023-10-02 21:10:49,762 - Epoch: [87][ 580/ 1236] Overall Loss 0.261631 Objective Loss 0.261631 LR 0.001000 Time 0.022016 +2023-10-02 21:10:49,966 - Epoch: [87][ 590/ 1236] Overall Loss 0.261746 Objective Loss 0.261746 LR 0.001000 Time 0.021990 +2023-10-02 21:10:50,175 - Epoch: [87][ 600/ 1236] Overall Loss 0.261893 Objective Loss 0.261893 LR 0.001000 Time 0.021970 +2023-10-02 21:10:50,380 - Epoch: [87][ 610/ 1236] Overall Loss 0.262294 Objective Loss 0.262294 LR 0.001000 Time 0.021946 +2023-10-02 21:10:50,589 - Epoch: [87][ 620/ 1236] Overall Loss 0.262228 Objective Loss 0.262228 LR 0.001000 Time 0.021928 +2023-10-02 21:10:50,794 - Epoch: [87][ 630/ 1236] Overall Loss 0.262236 Objective Loss 0.262236 LR 0.001000 Time 0.021905 +2023-10-02 21:10:51,003 - Epoch: [87][ 640/ 1236] Overall Loss 0.262469 Objective Loss 0.262469 LR 0.001000 Time 0.021888 +2023-10-02 21:10:51,207 - Epoch: [87][ 650/ 1236] Overall Loss 0.262232 Objective Loss 0.262232 LR 0.001000 Time 0.021866 +2023-10-02 21:10:51,415 - Epoch: [87][ 660/ 1236] Overall Loss 0.262293 Objective Loss 0.262293 LR 0.001000 Time 0.021849 +2023-10-02 21:10:51,621 - Epoch: [87][ 670/ 1236] Overall Loss 0.262717 Objective Loss 0.262717 LR 0.001000 Time 0.021828 +2023-10-02 21:10:51,830 - Epoch: [87][ 680/ 1236] Overall Loss 0.263053 Objective Loss 0.263053 LR 0.001000 Time 0.021814 +2023-10-02 21:10:52,035 - Epoch: [87][ 690/ 1236] Overall Loss 0.263352 Objective Loss 0.263352 LR 0.001000 Time 0.021794 +2023-10-02 21:10:52,243 - Epoch: [87][ 700/ 1236] Overall Loss 0.263078 Objective Loss 0.263078 LR 0.001000 Time 0.021781 +2023-10-02 21:10:52,448 - Epoch: [87][ 710/ 1236] Overall Loss 0.263502 Objective Loss 0.263502 LR 0.001000 Time 0.021762 +2023-10-02 21:10:52,655 - Epoch: [87][ 720/ 1236] Overall Loss 0.263620 Objective Loss 0.263620 LR 0.001000 Time 0.021747 +2023-10-02 21:10:52,860 - Epoch: [87][ 730/ 1236] Overall Loss 0.263574 Objective Loss 0.263574 LR 0.001000 Time 0.021729 +2023-10-02 21:10:53,068 - Epoch: [87][ 740/ 1236] Overall Loss 0.263282 Objective Loss 0.263282 LR 0.001000 Time 0.021715 +2023-10-02 21:10:53,272 - Epoch: [87][ 750/ 1236] Overall Loss 0.263116 Objective Loss 0.263116 LR 0.001000 Time 0.021699 +2023-10-02 21:10:53,480 - Epoch: [87][ 760/ 1236] Overall Loss 0.263334 Objective Loss 0.263334 LR 0.001000 Time 0.021686 +2023-10-02 21:10:53,686 - Epoch: [87][ 770/ 1236] Overall Loss 0.263357 Objective Loss 0.263357 LR 0.001000 Time 0.021671 +2023-10-02 21:10:53,894 - Epoch: [87][ 780/ 1236] Overall Loss 0.263271 Objective Loss 0.263271 LR 0.001000 Time 0.021661 +2023-10-02 21:10:54,099 - Epoch: [87][ 790/ 1236] Overall Loss 0.263548 Objective Loss 0.263548 LR 0.001000 Time 0.021645 +2023-10-02 21:10:54,307 - Epoch: [87][ 800/ 1236] Overall Loss 0.263584 Objective Loss 0.263584 LR 0.001000 Time 0.021634 +2023-10-02 21:10:54,513 - Epoch: [87][ 810/ 1236] Overall Loss 0.263763 Objective Loss 0.263763 LR 0.001000 Time 0.021621 +2023-10-02 21:10:54,721 - Epoch: [87][ 820/ 1236] Overall Loss 0.263996 Objective Loss 0.263996 LR 0.001000 Time 0.021611 +2023-10-02 21:10:54,926 - Epoch: [87][ 830/ 1236] Overall Loss 0.264748 Objective Loss 0.264748 LR 0.001000 Time 0.021597 +2023-10-02 21:10:55,135 - Epoch: [87][ 840/ 1236] Overall Loss 0.265013 Objective Loss 0.265013 LR 0.001000 Time 0.021588 +2023-10-02 21:10:55,340 - Epoch: [87][ 850/ 1236] Overall Loss 0.264855 Objective Loss 0.264855 LR 0.001000 Time 0.021575 +2023-10-02 21:10:55,549 - Epoch: [87][ 860/ 1236] Overall Loss 0.264647 Objective Loss 0.264647 LR 0.001000 Time 0.021566 +2023-10-02 21:10:55,754 - Epoch: [87][ 870/ 1236] Overall Loss 0.264581 Objective Loss 0.264581 LR 0.001000 Time 0.021554 +2023-10-02 21:10:55,962 - Epoch: [87][ 880/ 1236] Overall Loss 0.264665 Objective Loss 0.264665 LR 0.001000 Time 0.021546 +2023-10-02 21:10:56,167 - Epoch: [87][ 890/ 1236] Overall Loss 0.264417 Objective Loss 0.264417 LR 0.001000 Time 0.021533 +2023-10-02 21:10:56,375 - Epoch: [87][ 900/ 1236] Overall Loss 0.264386 Objective Loss 0.264386 LR 0.001000 Time 0.021524 +2023-10-02 21:10:56,581 - Epoch: [87][ 910/ 1236] Overall Loss 0.264531 Objective Loss 0.264531 LR 0.001000 Time 0.021514 +2023-10-02 21:10:56,790 - Epoch: [87][ 920/ 1236] Overall Loss 0.264406 Objective Loss 0.264406 LR 0.001000 Time 0.021507 +2023-10-02 21:10:56,995 - Epoch: [87][ 930/ 1236] Overall Loss 0.264178 Objective Loss 0.264178 LR 0.001000 Time 0.021496 +2023-10-02 21:10:57,204 - Epoch: [87][ 940/ 1236] Overall Loss 0.264377 Objective Loss 0.264377 LR 0.001000 Time 0.021489 +2023-10-02 21:10:57,408 - Epoch: [87][ 950/ 1236] Overall Loss 0.264643 Objective Loss 0.264643 LR 0.001000 Time 0.021478 +2023-10-02 21:10:57,616 - Epoch: [87][ 960/ 1236] Overall Loss 0.264804 Objective Loss 0.264804 LR 0.001000 Time 0.021470 +2023-10-02 21:10:57,822 - Epoch: [87][ 970/ 1236] Overall Loss 0.265147 Objective Loss 0.265147 LR 0.001000 Time 0.021461 +2023-10-02 21:10:58,031 - Epoch: [87][ 980/ 1236] Overall Loss 0.265469 Objective Loss 0.265469 LR 0.001000 Time 0.021455 +2023-10-02 21:10:58,235 - Epoch: [87][ 990/ 1236] Overall Loss 0.265458 Objective Loss 0.265458 LR 0.001000 Time 0.021444 +2023-10-02 21:10:58,444 - Epoch: [87][ 1000/ 1236] Overall Loss 0.265483 Objective Loss 0.265483 LR 0.001000 Time 0.021439 +2023-10-02 21:10:58,649 - Epoch: [87][ 1010/ 1236] Overall Loss 0.265709 Objective Loss 0.265709 LR 0.001000 Time 0.021429 +2023-10-02 21:10:58,858 - Epoch: [87][ 1020/ 1236] Overall Loss 0.265823 Objective Loss 0.265823 LR 0.001000 Time 0.021423 +2023-10-02 21:10:59,063 - Epoch: [87][ 1030/ 1236] Overall Loss 0.265537 Objective Loss 0.265537 LR 0.001000 Time 0.021414 +2023-10-02 21:10:59,272 - Epoch: [87][ 1040/ 1236] Overall Loss 0.265642 Objective Loss 0.265642 LR 0.001000 Time 0.021409 +2023-10-02 21:10:59,476 - Epoch: [87][ 1050/ 1236] Overall Loss 0.265869 Objective Loss 0.265869 LR 0.001000 Time 0.021399 +2023-10-02 21:10:59,685 - Epoch: [87][ 1060/ 1236] Overall Loss 0.265922 Objective Loss 0.265922 LR 0.001000 Time 0.021394 +2023-10-02 21:10:59,890 - Epoch: [87][ 1070/ 1236] Overall Loss 0.265798 Objective Loss 0.265798 LR 0.001000 Time 0.021385 +2023-10-02 21:11:00,099 - Epoch: [87][ 1080/ 1236] Overall Loss 0.265982 Objective Loss 0.265982 LR 0.001000 Time 0.021380 +2023-10-02 21:11:00,304 - Epoch: [87][ 1090/ 1236] Overall Loss 0.266131 Objective Loss 0.266131 LR 0.001000 Time 0.021372 +2023-10-02 21:11:00,512 - Epoch: [87][ 1100/ 1236] Overall Loss 0.266041 Objective Loss 0.266041 LR 0.001000 Time 0.021367 +2023-10-02 21:11:00,717 - Epoch: [87][ 1110/ 1236] Overall Loss 0.266319 Objective Loss 0.266319 LR 0.001000 Time 0.021359 +2023-10-02 21:11:00,926 - Epoch: [87][ 1120/ 1236] Overall Loss 0.266113 Objective Loss 0.266113 LR 0.001000 Time 0.021355 +2023-10-02 21:11:01,131 - Epoch: [87][ 1130/ 1236] Overall Loss 0.265896 Objective Loss 0.265896 LR 0.001000 Time 0.021347 +2023-10-02 21:11:01,339 - Epoch: [87][ 1140/ 1236] Overall Loss 0.265768 Objective Loss 0.265768 LR 0.001000 Time 0.021342 +2023-10-02 21:11:01,545 - Epoch: [87][ 1150/ 1236] Overall Loss 0.265791 Objective Loss 0.265791 LR 0.001000 Time 0.021335 +2023-10-02 21:11:01,752 - Epoch: [87][ 1160/ 1236] Overall Loss 0.265749 Objective Loss 0.265749 LR 0.001000 Time 0.021330 +2023-10-02 21:11:01,959 - Epoch: [87][ 1170/ 1236] Overall Loss 0.265835 Objective Loss 0.265835 LR 0.001000 Time 0.021323 +2023-10-02 21:11:02,168 - Epoch: [87][ 1180/ 1236] Overall Loss 0.266260 Objective Loss 0.266260 LR 0.001000 Time 0.021319 +2023-10-02 21:11:02,373 - Epoch: [87][ 1190/ 1236] Overall Loss 0.266144 Objective Loss 0.266144 LR 0.001000 Time 0.021312 +2023-10-02 21:11:02,582 - Epoch: [87][ 1200/ 1236] Overall Loss 0.266262 Objective Loss 0.266262 LR 0.001000 Time 0.021309 +2023-10-02 21:11:02,786 - Epoch: [87][ 1210/ 1236] Overall Loss 0.266208 Objective Loss 0.266208 LR 0.001000 Time 0.021301 +2023-10-02 21:11:02,995 - Epoch: [87][ 1220/ 1236] Overall Loss 0.266499 Objective Loss 0.266499 LR 0.001000 Time 0.021298 +2023-10-02 21:11:03,251 - Epoch: [87][ 1230/ 1236] Overall Loss 0.266646 Objective Loss 0.266646 LR 0.001000 Time 0.021333 +2023-10-02 21:11:03,373 - Epoch: [87][ 1236/ 1236] Overall Loss 0.266586 Objective Loss 0.266586 Top1 88.187373 Top5 97.963340 LR 0.001000 Time 0.021327 +2023-10-02 21:11:03,520 - --- validate (epoch=87)----------- +2023-10-02 21:11:03,520 - 29943 samples (256 per mini-batch) +2023-10-02 21:11:04,009 - Epoch: [87][ 10/ 117] Loss 0.349570 Top1 81.914062 Top5 98.164062 +2023-10-02 21:11:04,156 - Epoch: [87][ 20/ 117] Loss 0.353139 Top1 82.011719 Top5 98.339844 +2023-10-02 21:11:04,301 - Epoch: [87][ 30/ 117] Loss 0.363250 Top1 81.822917 Top5 98.216146 +2023-10-02 21:11:04,446 - Epoch: [87][ 40/ 117] Loss 0.358007 Top1 82.060547 Top5 98.173828 +2023-10-02 21:11:04,591 - Epoch: [87][ 50/ 117] Loss 0.349176 Top1 82.460938 Top5 98.195312 +2023-10-02 21:11:04,735 - Epoch: [87][ 60/ 117] Loss 0.348837 Top1 82.500000 Top5 98.216146 +2023-10-02 21:11:04,883 - Epoch: [87][ 70/ 117] Loss 0.344367 Top1 82.745536 Top5 98.180804 +2023-10-02 21:11:05,031 - Epoch: [87][ 80/ 117] Loss 0.348221 Top1 82.636719 Top5 98.173828 +2023-10-02 21:11:05,179 - Epoch: [87][ 90/ 117] Loss 0.352028 Top1 82.638889 Top5 98.125000 +2023-10-02 21:11:05,325 - Epoch: [87][ 100/ 117] Loss 0.348010 Top1 82.812500 Top5 98.089844 +2023-10-02 21:11:05,478 - Epoch: [87][ 110/ 117] Loss 0.349108 Top1 82.901278 Top5 98.135653 +2023-10-02 21:11:05,566 - Epoch: [87][ 117/ 117] Loss 0.349795 Top1 82.954280 Top5 98.136459 +2023-10-02 21:11:05,709 - ==> Top1: 82.954 Top5: 98.136 Loss: 0.350 + +2023-10-02 21:11:05,710 - ==> Confusion: +[[ 955 3 6 0 10 4 0 0 3 36 1 0 1 3 8 2 2 1 0 1 14] + [ 0 1027 1 2 5 35 0 18 2 0 2 0 1 1 1 7 4 1 10 3 11] + [ 3 0 965 5 4 0 16 5 0 1 2 1 9 2 1 11 0 1 11 3 16] + [ 1 2 16 964 0 11 0 4 0 0 2 0 5 4 22 2 1 6 29 0 20] + [ 30 11 2 0 951 6 1 0 0 11 2 0 1 2 8 9 9 1 1 0 5] + [ 3 30 2 0 4 992 3 21 2 5 2 9 4 9 3 1 4 0 6 2 14] + [ 1 0 26 0 0 1 1118 4 0 0 3 6 1 2 0 8 0 2 1 8 10] + [ 3 21 19 1 5 40 8 1048 1 1 3 2 3 5 5 0 2 2 30 7 12] + [ 34 6 1 3 1 5 0 1 936 38 13 1 2 13 20 1 3 1 7 1 2] + [ 143 2 0 2 9 2 4 0 29 861 1 1 0 41 7 2 0 0 0 5 10] + [ 2 0 10 8 0 1 5 5 9 0 961 1 0 16 7 1 0 3 8 2 14] + [ 2 2 2 0 2 7 0 1 0 1 0 959 19 10 0 2 2 10 0 7 9] + [ 0 0 1 6 0 3 2 0 1 0 1 53 945 8 3 12 1 18 2 4 8] + [ 0 0 1 1 2 15 0 0 7 9 4 7 2 1043 6 4 3 1 0 2 12] + [ 16 2 5 23 5 0 0 0 13 4 4 0 3 4 990 0 4 5 11 1 11] + [ 0 0 2 0 4 0 0 0 0 0 1 11 6 0 0 1073 17 8 2 6 4] + [ 2 17 0 0 3 6 1 0 1 2 0 4 1 1 2 17 1085 0 1 2 16] + [ 0 0 1 4 0 0 2 0 2 0 1 12 33 2 1 9 0 966 0 1 4] + [ 1 5 6 8 1 0 1 24 2 0 7 0 2 0 7 0 0 0 987 3 14] + [ 0 4 8 0 1 5 13 13 0 0 0 16 6 0 1 6 10 1 0 1061 7] + [ 129 192 155 91 69 211 42 97 95 50 182 158 380 342 138 95 116 66 165 180 4952]] + +2023-10-02 21:11:05,711 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:11:05,711 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:11:05,718 - + +2023-10-02 21:11:05,718 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:11:06,732 - Epoch: [88][ 10/ 1236] Overall Loss 0.278916 Objective Loss 0.278916 LR 0.001000 Time 0.101352 +2023-10-02 21:11:06,938 - Epoch: [88][ 20/ 1236] Overall Loss 0.252229 Objective Loss 0.252229 LR 0.001000 Time 0.060964 +2023-10-02 21:11:07,143 - Epoch: [88][ 30/ 1236] Overall Loss 0.273083 Objective Loss 0.273083 LR 0.001000 Time 0.047439 +2023-10-02 21:11:07,349 - Epoch: [88][ 40/ 1236] Overall Loss 0.274966 Objective Loss 0.274966 LR 0.001000 Time 0.040719 +2023-10-02 21:11:07,555 - Epoch: [88][ 50/ 1236] Overall Loss 0.270151 Objective Loss 0.270151 LR 0.001000 Time 0.036660 +2023-10-02 21:11:07,762 - Epoch: [88][ 60/ 1236] Overall Loss 0.264609 Objective Loss 0.264609 LR 0.001000 Time 0.033995 +2023-10-02 21:11:07,966 - Epoch: [88][ 70/ 1236] Overall Loss 0.266735 Objective Loss 0.266735 LR 0.001000 Time 0.032053 +2023-10-02 21:11:08,172 - Epoch: [88][ 80/ 1236] Overall Loss 0.267192 Objective Loss 0.267192 LR 0.001000 Time 0.030619 +2023-10-02 21:11:08,378 - Epoch: [88][ 90/ 1236] Overall Loss 0.265319 Objective Loss 0.265319 LR 0.001000 Time 0.029487 +2023-10-02 21:11:08,585 - Epoch: [88][ 100/ 1236] Overall Loss 0.266346 Objective Loss 0.266346 LR 0.001000 Time 0.028607 +2023-10-02 21:11:08,789 - Epoch: [88][ 110/ 1236] Overall Loss 0.267150 Objective Loss 0.267150 LR 0.001000 Time 0.027857 +2023-10-02 21:11:08,996 - Epoch: [88][ 120/ 1236] Overall Loss 0.266742 Objective Loss 0.266742 LR 0.001000 Time 0.027262 +2023-10-02 21:11:09,201 - Epoch: [88][ 130/ 1236] Overall Loss 0.267914 Objective Loss 0.267914 LR 0.001000 Time 0.026734 +2023-10-02 21:11:09,408 - Epoch: [88][ 140/ 1236] Overall Loss 0.267981 Objective Loss 0.267981 LR 0.001000 Time 0.026304 +2023-10-02 21:11:09,613 - Epoch: [88][ 150/ 1236] Overall Loss 0.271153 Objective Loss 0.271153 LR 0.001000 Time 0.025912 +2023-10-02 21:11:09,820 - Epoch: [88][ 160/ 1236] Overall Loss 0.272317 Objective Loss 0.272317 LR 0.001000 Time 0.025588 +2023-10-02 21:11:10,025 - Epoch: [88][ 170/ 1236] Overall Loss 0.273481 Objective Loss 0.273481 LR 0.001000 Time 0.025286 +2023-10-02 21:11:10,233 - Epoch: [88][ 180/ 1236] Overall Loss 0.272679 Objective Loss 0.272679 LR 0.001000 Time 0.025033 +2023-10-02 21:11:10,437 - Epoch: [88][ 190/ 1236] Overall Loss 0.270933 Objective Loss 0.270933 LR 0.001000 Time 0.024790 +2023-10-02 21:11:10,644 - Epoch: [88][ 200/ 1236] Overall Loss 0.270598 Objective Loss 0.270598 LR 0.001000 Time 0.024585 +2023-10-02 21:11:10,849 - Epoch: [88][ 210/ 1236] Overall Loss 0.270909 Objective Loss 0.270909 LR 0.001000 Time 0.024387 +2023-10-02 21:11:11,056 - Epoch: [88][ 220/ 1236] Overall Loss 0.269223 Objective Loss 0.269223 LR 0.001000 Time 0.024219 +2023-10-02 21:11:11,261 - Epoch: [88][ 230/ 1236] Overall Loss 0.268720 Objective Loss 0.268720 LR 0.001000 Time 0.024054 +2023-10-02 21:11:11,467 - Epoch: [88][ 240/ 1236] Overall Loss 0.268293 Objective Loss 0.268293 LR 0.001000 Time 0.023911 +2023-10-02 21:11:11,673 - Epoch: [88][ 250/ 1236] Overall Loss 0.268413 Objective Loss 0.268413 LR 0.001000 Time 0.023771 +2023-10-02 21:11:11,880 - Epoch: [88][ 260/ 1236] Overall Loss 0.268275 Objective Loss 0.268275 LR 0.001000 Time 0.023654 +2023-10-02 21:11:12,085 - Epoch: [88][ 270/ 1236] Overall Loss 0.267724 Objective Loss 0.267724 LR 0.001000 Time 0.023533 +2023-10-02 21:11:12,292 - Epoch: [88][ 280/ 1236] Overall Loss 0.267644 Objective Loss 0.267644 LR 0.001000 Time 0.023432 +2023-10-02 21:11:12,496 - Epoch: [88][ 290/ 1236] Overall Loss 0.267557 Objective Loss 0.267557 LR 0.001000 Time 0.023327 +2023-10-02 21:11:12,702 - Epoch: [88][ 300/ 1236] Overall Loss 0.266935 Objective Loss 0.266935 LR 0.001000 Time 0.023236 +2023-10-02 21:11:12,908 - Epoch: [88][ 310/ 1236] Overall Loss 0.266344 Objective Loss 0.266344 LR 0.001000 Time 0.023144 +2023-10-02 21:11:13,115 - Epoch: [88][ 320/ 1236] Overall Loss 0.266231 Objective Loss 0.266231 LR 0.001000 Time 0.023068 +2023-10-02 21:11:13,320 - Epoch: [88][ 330/ 1236] Overall Loss 0.266234 Objective Loss 0.266234 LR 0.001000 Time 0.022989 +2023-10-02 21:11:13,527 - Epoch: [88][ 340/ 1236] Overall Loss 0.266965 Objective Loss 0.266965 LR 0.001000 Time 0.022922 +2023-10-02 21:11:13,732 - Epoch: [88][ 350/ 1236] Overall Loss 0.267810 Objective Loss 0.267810 LR 0.001000 Time 0.022851 +2023-10-02 21:11:13,939 - Epoch: [88][ 360/ 1236] Overall Loss 0.267600 Objective Loss 0.267600 LR 0.001000 Time 0.022789 +2023-10-02 21:11:14,145 - Epoch: [88][ 370/ 1236] Overall Loss 0.267676 Objective Loss 0.267676 LR 0.001000 Time 0.022726 +2023-10-02 21:11:14,353 - Epoch: [88][ 380/ 1236] Overall Loss 0.267649 Objective Loss 0.267649 LR 0.001000 Time 0.022674 +2023-10-02 21:11:14,557 - Epoch: [88][ 390/ 1236] Overall Loss 0.267275 Objective Loss 0.267275 LR 0.001000 Time 0.022617 +2023-10-02 21:11:14,765 - Epoch: [88][ 400/ 1236] Overall Loss 0.268186 Objective Loss 0.268186 LR 0.001000 Time 0.022570 +2023-10-02 21:11:14,970 - Epoch: [88][ 410/ 1236] Overall Loss 0.267569 Objective Loss 0.267569 LR 0.001000 Time 0.022519 +2023-10-02 21:11:15,178 - Epoch: [88][ 420/ 1236] Overall Loss 0.268380 Objective Loss 0.268380 LR 0.001000 Time 0.022476 +2023-10-02 21:11:15,382 - Epoch: [88][ 430/ 1236] Overall Loss 0.267921 Objective Loss 0.267921 LR 0.001000 Time 0.022429 +2023-10-02 21:11:15,590 - Epoch: [88][ 440/ 1236] Overall Loss 0.267271 Objective Loss 0.267271 LR 0.001000 Time 0.022390 +2023-10-02 21:11:15,794 - Epoch: [88][ 450/ 1236] Overall Loss 0.267412 Objective Loss 0.267412 LR 0.001000 Time 0.022346 +2023-10-02 21:11:16,002 - Epoch: [88][ 460/ 1236] Overall Loss 0.267670 Objective Loss 0.267670 LR 0.001000 Time 0.022311 +2023-10-02 21:11:16,207 - Epoch: [88][ 470/ 1236] Overall Loss 0.267897 Objective Loss 0.267897 LR 0.001000 Time 0.022272 +2023-10-02 21:11:16,414 - Epoch: [88][ 480/ 1236] Overall Loss 0.267865 Objective Loss 0.267865 LR 0.001000 Time 0.022239 +2023-10-02 21:11:16,619 - Epoch: [88][ 490/ 1236] Overall Loss 0.267384 Objective Loss 0.267384 LR 0.001000 Time 0.022203 +2023-10-02 21:11:16,827 - Epoch: [88][ 500/ 1236] Overall Loss 0.267178 Objective Loss 0.267178 LR 0.001000 Time 0.022173 +2023-10-02 21:11:17,031 - Epoch: [88][ 510/ 1236] Overall Loss 0.266598 Objective Loss 0.266598 LR 0.001000 Time 0.022139 +2023-10-02 21:11:17,239 - Epoch: [88][ 520/ 1236] Overall Loss 0.266413 Objective Loss 0.266413 LR 0.001000 Time 0.022113 +2023-10-02 21:11:17,443 - Epoch: [88][ 530/ 1236] Overall Loss 0.265949 Objective Loss 0.265949 LR 0.001000 Time 0.022081 +2023-10-02 21:11:17,651 - Epoch: [88][ 540/ 1236] Overall Loss 0.266142 Objective Loss 0.266142 LR 0.001000 Time 0.022056 +2023-10-02 21:11:17,856 - Epoch: [88][ 550/ 1236] Overall Loss 0.266270 Objective Loss 0.266270 LR 0.001000 Time 0.022027 +2023-10-02 21:11:18,064 - Epoch: [88][ 560/ 1236] Overall Loss 0.266302 Objective Loss 0.266302 LR 0.001000 Time 0.022004 +2023-10-02 21:11:18,269 - Epoch: [88][ 570/ 1236] Overall Loss 0.266407 Objective Loss 0.266407 LR 0.001000 Time 0.021977 +2023-10-02 21:11:18,475 - Epoch: [88][ 580/ 1236] Overall Loss 0.266422 Objective Loss 0.266422 LR 0.001000 Time 0.021954 +2023-10-02 21:11:18,681 - Epoch: [88][ 590/ 1236] Overall Loss 0.266524 Objective Loss 0.266524 LR 0.001000 Time 0.021928 +2023-10-02 21:11:18,887 - Epoch: [88][ 600/ 1236] Overall Loss 0.265842 Objective Loss 0.265842 LR 0.001000 Time 0.021906 +2023-10-02 21:11:19,093 - Epoch: [88][ 610/ 1236] Overall Loss 0.265818 Objective Loss 0.265818 LR 0.001000 Time 0.021882 +2023-10-02 21:11:19,301 - Epoch: [88][ 620/ 1236] Overall Loss 0.265633 Objective Loss 0.265633 LR 0.001000 Time 0.021863 +2023-10-02 21:11:19,506 - Epoch: [88][ 630/ 1236] Overall Loss 0.265651 Objective Loss 0.265651 LR 0.001000 Time 0.021841 +2023-10-02 21:11:19,714 - Epoch: [88][ 640/ 1236] Overall Loss 0.265800 Objective Loss 0.265800 LR 0.001000 Time 0.021824 +2023-10-02 21:11:19,918 - Epoch: [88][ 650/ 1236] Overall Loss 0.266320 Objective Loss 0.266320 LR 0.001000 Time 0.021803 +2023-10-02 21:11:20,126 - Epoch: [88][ 660/ 1236] Overall Loss 0.266068 Objective Loss 0.266068 LR 0.001000 Time 0.021786 +2023-10-02 21:11:20,331 - Epoch: [88][ 670/ 1236] Overall Loss 0.266832 Objective Loss 0.266832 LR 0.001000 Time 0.021766 +2023-10-02 21:11:20,538 - Epoch: [88][ 680/ 1236] Overall Loss 0.266499 Objective Loss 0.266499 LR 0.001000 Time 0.021751 +2023-10-02 21:11:20,743 - Epoch: [88][ 690/ 1236] Overall Loss 0.266306 Objective Loss 0.266306 LR 0.001000 Time 0.021732 +2023-10-02 21:11:20,951 - Epoch: [88][ 700/ 1236] Overall Loss 0.266313 Objective Loss 0.266313 LR 0.001000 Time 0.021718 +2023-10-02 21:11:21,156 - Epoch: [88][ 710/ 1236] Overall Loss 0.266215 Objective Loss 0.266215 LR 0.001000 Time 0.021701 +2023-10-02 21:11:21,364 - Epoch: [88][ 720/ 1236] Overall Loss 0.266227 Objective Loss 0.266227 LR 0.001000 Time 0.021687 +2023-10-02 21:11:21,568 - Epoch: [88][ 730/ 1236] Overall Loss 0.265966 Objective Loss 0.265966 LR 0.001000 Time 0.021670 +2023-10-02 21:11:21,776 - Epoch: [88][ 740/ 1236] Overall Loss 0.265729 Objective Loss 0.265729 LR 0.001000 Time 0.021658 +2023-10-02 21:11:21,980 - Epoch: [88][ 750/ 1236] Overall Loss 0.265754 Objective Loss 0.265754 LR 0.001000 Time 0.021641 +2023-10-02 21:11:22,188 - Epoch: [88][ 760/ 1236] Overall Loss 0.265803 Objective Loss 0.265803 LR 0.001000 Time 0.021630 +2023-10-02 21:11:22,393 - Epoch: [88][ 770/ 1236] Overall Loss 0.265893 Objective Loss 0.265893 LR 0.001000 Time 0.021614 +2023-10-02 21:11:22,601 - Epoch: [88][ 780/ 1236] Overall Loss 0.265794 Objective Loss 0.265794 LR 0.001000 Time 0.021603 +2023-10-02 21:11:22,805 - Epoch: [88][ 790/ 1236] Overall Loss 0.265755 Objective Loss 0.265755 LR 0.001000 Time 0.021588 +2023-10-02 21:11:23,013 - Epoch: [88][ 800/ 1236] Overall Loss 0.265630 Objective Loss 0.265630 LR 0.001000 Time 0.021578 +2023-10-02 21:11:23,218 - Epoch: [88][ 810/ 1236] Overall Loss 0.265628 Objective Loss 0.265628 LR 0.001000 Time 0.021564 +2023-10-02 21:11:23,426 - Epoch: [88][ 820/ 1236] Overall Loss 0.265601 Objective Loss 0.265601 LR 0.001000 Time 0.021554 +2023-10-02 21:11:23,630 - Epoch: [88][ 830/ 1236] Overall Loss 0.265514 Objective Loss 0.265514 LR 0.001000 Time 0.021540 +2023-10-02 21:11:23,837 - Epoch: [88][ 840/ 1236] Overall Loss 0.265608 Objective Loss 0.265608 LR 0.001000 Time 0.021529 +2023-10-02 21:11:24,043 - Epoch: [88][ 850/ 1236] Overall Loss 0.265805 Objective Loss 0.265805 LR 0.001000 Time 0.021517 +2023-10-02 21:11:24,250 - Epoch: [88][ 860/ 1236] Overall Loss 0.265614 Objective Loss 0.265614 LR 0.001000 Time 0.021507 +2023-10-02 21:11:24,455 - Epoch: [88][ 870/ 1236] Overall Loss 0.265693 Objective Loss 0.265693 LR 0.001000 Time 0.021495 +2023-10-02 21:11:24,663 - Epoch: [88][ 880/ 1236] Overall Loss 0.265750 Objective Loss 0.265750 LR 0.001000 Time 0.021487 +2023-10-02 21:11:24,868 - Epoch: [88][ 890/ 1236] Overall Loss 0.265724 Objective Loss 0.265724 LR 0.001000 Time 0.021475 +2023-10-02 21:11:25,076 - Epoch: [88][ 900/ 1236] Overall Loss 0.266000 Objective Loss 0.266000 LR 0.001000 Time 0.021467 +2023-10-02 21:11:25,280 - Epoch: [88][ 910/ 1236] Overall Loss 0.266184 Objective Loss 0.266184 LR 0.001000 Time 0.021456 +2023-10-02 21:11:25,488 - Epoch: [88][ 920/ 1236] Overall Loss 0.266417 Objective Loss 0.266417 LR 0.001000 Time 0.021448 +2023-10-02 21:11:25,693 - Epoch: [88][ 930/ 1236] Overall Loss 0.266367 Objective Loss 0.266367 LR 0.001000 Time 0.021438 +2023-10-02 21:11:25,900 - Epoch: [88][ 940/ 1236] Overall Loss 0.266609 Objective Loss 0.266609 LR 0.001000 Time 0.021429 +2023-10-02 21:11:26,106 - Epoch: [88][ 950/ 1236] Overall Loss 0.266622 Objective Loss 0.266622 LR 0.001000 Time 0.021419 +2023-10-02 21:11:26,313 - Epoch: [88][ 960/ 1236] Overall Loss 0.266819 Objective Loss 0.266819 LR 0.001000 Time 0.021412 +2023-10-02 21:11:26,521 - Epoch: [88][ 970/ 1236] Overall Loss 0.266812 Objective Loss 0.266812 LR 0.001000 Time 0.021405 +2023-10-02 21:11:26,738 - Epoch: [88][ 980/ 1236] Overall Loss 0.266483 Objective Loss 0.266483 LR 0.001000 Time 0.021407 +2023-10-02 21:11:26,950 - Epoch: [88][ 990/ 1236] Overall Loss 0.266896 Objective Loss 0.266896 LR 0.001000 Time 0.021405 +2023-10-02 21:11:27,167 - Epoch: [88][ 1000/ 1236] Overall Loss 0.267188 Objective Loss 0.267188 LR 0.001000 Time 0.021407 +2023-10-02 21:11:27,379 - Epoch: [88][ 1010/ 1236] Overall Loss 0.266965 Objective Loss 0.266965 LR 0.001000 Time 0.021405 +2023-10-02 21:11:27,595 - Epoch: [88][ 1020/ 1236] Overall Loss 0.267165 Objective Loss 0.267165 LR 0.001000 Time 0.021407 +2023-10-02 21:11:27,808 - Epoch: [88][ 1030/ 1236] Overall Loss 0.266949 Objective Loss 0.266949 LR 0.001000 Time 0.021405 +2023-10-02 21:11:28,025 - Epoch: [88][ 1040/ 1236] Overall Loss 0.266998 Objective Loss 0.266998 LR 0.001000 Time 0.021407 +2023-10-02 21:11:28,237 - Epoch: [88][ 1050/ 1236] Overall Loss 0.266925 Objective Loss 0.266925 LR 0.001000 Time 0.021405 +2023-10-02 21:11:28,453 - Epoch: [88][ 1060/ 1236] Overall Loss 0.267084 Objective Loss 0.267084 LR 0.001000 Time 0.021408 +2023-10-02 21:11:28,666 - Epoch: [88][ 1070/ 1236] Overall Loss 0.267107 Objective Loss 0.267107 LR 0.001000 Time 0.021405 +2023-10-02 21:11:28,882 - Epoch: [88][ 1080/ 1236] Overall Loss 0.267109 Objective Loss 0.267109 LR 0.001000 Time 0.021407 +2023-10-02 21:11:29,094 - Epoch: [88][ 1090/ 1236] Overall Loss 0.267109 Objective Loss 0.267109 LR 0.001000 Time 0.021405 +2023-10-02 21:11:29,311 - Epoch: [88][ 1100/ 1236] Overall Loss 0.267237 Objective Loss 0.267237 LR 0.001000 Time 0.021407 +2023-10-02 21:11:29,523 - Epoch: [88][ 1110/ 1236] Overall Loss 0.267406 Objective Loss 0.267406 LR 0.001000 Time 0.021405 +2023-10-02 21:11:29,740 - Epoch: [88][ 1120/ 1236] Overall Loss 0.267305 Objective Loss 0.267305 LR 0.001000 Time 0.021408 +2023-10-02 21:11:29,952 - Epoch: [88][ 1130/ 1236] Overall Loss 0.267551 Objective Loss 0.267551 LR 0.001000 Time 0.021406 +2023-10-02 21:11:30,169 - Epoch: [88][ 1140/ 1236] Overall Loss 0.267858 Objective Loss 0.267858 LR 0.001000 Time 0.021408 +2023-10-02 21:11:30,381 - Epoch: [88][ 1150/ 1236] Overall Loss 0.267636 Objective Loss 0.267636 LR 0.001000 Time 0.021406 +2023-10-02 21:11:30,598 - Epoch: [88][ 1160/ 1236] Overall Loss 0.267589 Objective Loss 0.267589 LR 0.001000 Time 0.021408 +2023-10-02 21:11:30,810 - Epoch: [88][ 1170/ 1236] Overall Loss 0.267652 Objective Loss 0.267652 LR 0.001000 Time 0.021406 +2023-10-02 21:11:31,026 - Epoch: [88][ 1180/ 1236] Overall Loss 0.267710 Objective Loss 0.267710 LR 0.001000 Time 0.021407 +2023-10-02 21:11:31,232 - Epoch: [88][ 1190/ 1236] Overall Loss 0.267692 Objective Loss 0.267692 LR 0.001000 Time 0.021400 +2023-10-02 21:11:31,438 - Epoch: [88][ 1200/ 1236] Overall Loss 0.267740 Objective Loss 0.267740 LR 0.001000 Time 0.021393 +2023-10-02 21:11:31,644 - Epoch: [88][ 1210/ 1236] Overall Loss 0.267733 Objective Loss 0.267733 LR 0.001000 Time 0.021385 +2023-10-02 21:11:31,851 - Epoch: [88][ 1220/ 1236] Overall Loss 0.267717 Objective Loss 0.267717 LR 0.001000 Time 0.021379 +2023-10-02 21:11:32,111 - Epoch: [88][ 1230/ 1236] Overall Loss 0.267870 Objective Loss 0.267870 LR 0.001000 Time 0.021415 +2023-10-02 21:11:32,232 - Epoch: [88][ 1236/ 1236] Overall Loss 0.267780 Objective Loss 0.267780 Top1 87.780041 Top5 98.167006 LR 0.001000 Time 0.021409 +2023-10-02 21:11:32,382 - --- validate (epoch=88)----------- +2023-10-02 21:11:32,382 - 29943 samples (256 per mini-batch) +2023-10-02 21:11:32,870 - Epoch: [88][ 10/ 117] Loss 0.320748 Top1 83.828125 Top5 98.046875 +2023-10-02 21:11:33,021 - Epoch: [88][ 20/ 117] Loss 0.331090 Top1 83.339844 Top5 98.085938 +2023-10-02 21:11:33,172 - Epoch: [88][ 30/ 117] Loss 0.336979 Top1 83.359375 Top5 98.020833 +2023-10-02 21:11:33,324 - Epoch: [88][ 40/ 117] Loss 0.332972 Top1 83.417969 Top5 98.046875 +2023-10-02 21:11:33,475 - Epoch: [88][ 50/ 117] Loss 0.347821 Top1 83.304688 Top5 98.007812 +2023-10-02 21:11:33,625 - Epoch: [88][ 60/ 117] Loss 0.345757 Top1 83.567708 Top5 98.079427 +2023-10-02 21:11:33,779 - Epoch: [88][ 70/ 117] Loss 0.341861 Top1 83.582589 Top5 98.152902 +2023-10-02 21:11:33,936 - Epoch: [88][ 80/ 117] Loss 0.342313 Top1 83.554688 Top5 98.149414 +2023-10-02 21:11:34,093 - Epoch: [88][ 90/ 117] Loss 0.340865 Top1 83.606771 Top5 98.146701 +2023-10-02 21:11:34,251 - Epoch: [88][ 100/ 117] Loss 0.337498 Top1 83.613281 Top5 98.144531 +2023-10-02 21:11:34,414 - Epoch: [88][ 110/ 117] Loss 0.338206 Top1 83.487216 Top5 98.181818 +2023-10-02 21:11:34,503 - Epoch: [88][ 117/ 117] Loss 0.336375 Top1 83.592158 Top5 98.176535 +2023-10-02 21:11:34,644 - ==> Top1: 83.592 Top5: 98.177 Loss: 0.336 + +2023-10-02 21:11:34,645 - ==> Confusion: +[[ 898 1 4 1 9 5 0 1 8 91 4 0 0 4 7 1 2 0 0 0 14] + [ 1 1006 0 2 5 48 1 36 4 2 3 1 1 1 2 3 0 0 8 2 5] + [ 4 2 952 18 1 0 26 11 0 4 6 1 6 4 0 4 1 1 8 2 5] + [ 1 2 15 957 0 4 1 6 9 1 12 0 5 4 36 0 0 4 15 1 16] + [ 19 7 2 0 956 9 0 1 1 21 2 1 2 4 7 6 4 1 0 1 6] + [ 2 28 0 2 3 997 1 19 2 4 3 6 2 20 6 0 2 0 7 6 6] + [ 0 6 26 0 0 2 1117 6 0 0 5 1 0 1 0 4 0 0 4 13 6] + [ 0 13 19 0 2 43 4 1058 1 5 7 6 1 5 3 1 3 1 29 7 10] + [ 16 2 2 0 2 2 0 0 968 47 7 3 3 14 15 2 2 0 4 0 0] + [ 65 1 0 1 6 4 1 0 30 964 7 1 0 27 4 0 0 0 0 1 7] + [ 1 1 7 5 2 2 2 3 20 1 969 1 0 17 2 0 3 1 3 1 12] + [ 1 0 4 0 0 17 0 5 0 1 0 956 20 9 0 1 1 12 0 4 4] + [ 1 0 0 5 0 3 1 2 0 3 2 48 958 4 5 6 1 5 3 7 14] + [ 2 0 0 0 2 11 0 3 9 7 4 4 0 1059 6 0 1 1 1 0 9] + [ 7 0 5 18 5 2 0 0 25 3 4 0 2 6 1003 0 2 3 9 0 7] + [ 0 0 2 2 3 0 2 0 0 0 0 6 9 2 0 1063 19 8 2 5 11] + [ 1 15 1 1 10 9 0 0 3 1 0 7 2 3 3 11 1075 0 1 5 13] + [ 0 1 0 5 1 1 2 0 0 2 1 7 40 3 1 6 0 960 2 0 6] + [ 3 2 3 12 0 2 1 22 7 1 9 0 1 0 10 0 0 0 982 1 12] + [ 0 1 3 1 0 6 6 9 0 1 7 12 4 2 0 2 1 0 3 1089 5] + [ 143 167 130 99 69 220 42 117 99 91 193 127 328 350 141 42 115 50 157 182 5043]] + +2023-10-02 21:11:34,646 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:11:34,646 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:11:34,652 - + +2023-10-02 21:11:34,652 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:11:35,776 - Epoch: [89][ 10/ 1236] Overall Loss 0.271372 Objective Loss 0.271372 LR 0.001000 Time 0.112305 +2023-10-02 21:11:35,982 - Epoch: [89][ 20/ 1236] Overall Loss 0.275662 Objective Loss 0.275662 LR 0.001000 Time 0.066409 +2023-10-02 21:11:36,185 - Epoch: [89][ 30/ 1236] Overall Loss 0.268183 Objective Loss 0.268183 LR 0.001000 Time 0.051007 +2023-10-02 21:11:36,390 - Epoch: [89][ 40/ 1236] Overall Loss 0.265550 Objective Loss 0.265550 LR 0.001000 Time 0.043382 +2023-10-02 21:11:36,593 - Epoch: [89][ 50/ 1236] Overall Loss 0.264939 Objective Loss 0.264939 LR 0.001000 Time 0.038739 +2023-10-02 21:11:36,799 - Epoch: [89][ 60/ 1236] Overall Loss 0.262244 Objective Loss 0.262244 LR 0.001000 Time 0.035710 +2023-10-02 21:11:37,003 - Epoch: [89][ 70/ 1236] Overall Loss 0.264496 Objective Loss 0.264496 LR 0.001000 Time 0.033501 +2023-10-02 21:11:37,210 - Epoch: [89][ 80/ 1236] Overall Loss 0.265784 Objective Loss 0.265784 LR 0.001000 Time 0.031895 +2023-10-02 21:11:37,414 - Epoch: [89][ 90/ 1236] Overall Loss 0.265791 Objective Loss 0.265791 LR 0.001000 Time 0.030609 +2023-10-02 21:11:37,620 - Epoch: [89][ 100/ 1236] Overall Loss 0.268629 Objective Loss 0.268629 LR 0.001000 Time 0.029612 +2023-10-02 21:11:37,824 - Epoch: [89][ 110/ 1236] Overall Loss 0.267172 Objective Loss 0.267172 LR 0.001000 Time 0.028771 +2023-10-02 21:11:38,031 - Epoch: [89][ 120/ 1236] Overall Loss 0.265273 Objective Loss 0.265273 LR 0.001000 Time 0.028093 +2023-10-02 21:11:38,234 - Epoch: [89][ 130/ 1236] Overall Loss 0.265036 Objective Loss 0.265036 LR 0.001000 Time 0.027494 +2023-10-02 21:11:38,441 - Epoch: [89][ 140/ 1236] Overall Loss 0.264377 Objective Loss 0.264377 LR 0.001000 Time 0.027006 +2023-10-02 21:11:38,645 - Epoch: [89][ 150/ 1236] Overall Loss 0.265396 Objective Loss 0.265396 LR 0.001000 Time 0.026560 +2023-10-02 21:11:38,851 - Epoch: [89][ 160/ 1236] Overall Loss 0.264967 Objective Loss 0.264967 LR 0.001000 Time 0.026191 +2023-10-02 21:11:39,055 - Epoch: [89][ 170/ 1236] Overall Loss 0.264359 Objective Loss 0.264359 LR 0.001000 Time 0.025845 +2023-10-02 21:11:39,261 - Epoch: [89][ 180/ 1236] Overall Loss 0.263351 Objective Loss 0.263351 LR 0.001000 Time 0.025554 +2023-10-02 21:11:39,464 - Epoch: [89][ 190/ 1236] Overall Loss 0.264481 Objective Loss 0.264481 LR 0.001000 Time 0.025277 +2023-10-02 21:11:39,671 - Epoch: [89][ 200/ 1236] Overall Loss 0.264538 Objective Loss 0.264538 LR 0.001000 Time 0.025047 +2023-10-02 21:11:39,875 - Epoch: [89][ 210/ 1236] Overall Loss 0.263639 Objective Loss 0.263639 LR 0.001000 Time 0.024821 +2023-10-02 21:11:40,082 - Epoch: [89][ 220/ 1236] Overall Loss 0.262558 Objective Loss 0.262558 LR 0.001000 Time 0.024632 +2023-10-02 21:11:40,285 - Epoch: [89][ 230/ 1236] Overall Loss 0.262671 Objective Loss 0.262671 LR 0.001000 Time 0.024444 +2023-10-02 21:11:40,492 - Epoch: [89][ 240/ 1236] Overall Loss 0.262137 Objective Loss 0.262137 LR 0.001000 Time 0.024287 +2023-10-02 21:11:40,695 - Epoch: [89][ 250/ 1236] Overall Loss 0.262500 Objective Loss 0.262500 LR 0.001000 Time 0.024127 +2023-10-02 21:11:40,902 - Epoch: [89][ 260/ 1236] Overall Loss 0.262609 Objective Loss 0.262609 LR 0.001000 Time 0.023994 +2023-10-02 21:11:41,106 - Epoch: [89][ 270/ 1236] Overall Loss 0.262935 Objective Loss 0.262935 LR 0.001000 Time 0.023858 +2023-10-02 21:11:41,313 - Epoch: [89][ 280/ 1236] Overall Loss 0.264479 Objective Loss 0.264479 LR 0.001000 Time 0.023745 +2023-10-02 21:11:41,517 - Epoch: [89][ 290/ 1236] Overall Loss 0.263589 Objective Loss 0.263589 LR 0.001000 Time 0.023627 +2023-10-02 21:11:41,723 - Epoch: [89][ 300/ 1236] Overall Loss 0.264499 Objective Loss 0.264499 LR 0.001000 Time 0.023528 +2023-10-02 21:11:41,927 - Epoch: [89][ 310/ 1236] Overall Loss 0.263844 Objective Loss 0.263844 LR 0.001000 Time 0.023424 +2023-10-02 21:11:42,134 - Epoch: [89][ 320/ 1236] Overall Loss 0.264725 Objective Loss 0.264725 LR 0.001000 Time 0.023339 +2023-10-02 21:11:42,338 - Epoch: [89][ 330/ 1236] Overall Loss 0.265422 Objective Loss 0.265422 LR 0.001000 Time 0.023249 +2023-10-02 21:11:42,544 - Epoch: [89][ 340/ 1236] Overall Loss 0.265735 Objective Loss 0.265735 LR 0.001000 Time 0.023169 +2023-10-02 21:11:42,749 - Epoch: [89][ 350/ 1236] Overall Loss 0.266194 Objective Loss 0.266194 LR 0.001000 Time 0.023089 +2023-10-02 21:11:42,954 - Epoch: [89][ 360/ 1236] Overall Loss 0.266327 Objective Loss 0.266327 LR 0.001000 Time 0.023018 +2023-10-02 21:11:43,160 - Epoch: [89][ 370/ 1236] Overall Loss 0.266623 Objective Loss 0.266623 LR 0.001000 Time 0.022947 +2023-10-02 21:11:43,367 - Epoch: [89][ 380/ 1236] Overall Loss 0.266059 Objective Loss 0.266059 LR 0.001000 Time 0.022886 +2023-10-02 21:11:43,570 - Epoch: [89][ 390/ 1236] Overall Loss 0.266210 Objective Loss 0.266210 LR 0.001000 Time 0.022821 +2023-10-02 21:11:43,777 - Epoch: [89][ 400/ 1236] Overall Loss 0.265749 Objective Loss 0.265749 LR 0.001000 Time 0.022767 +2023-10-02 21:11:43,982 - Epoch: [89][ 410/ 1236] Overall Loss 0.266336 Objective Loss 0.266336 LR 0.001000 Time 0.022710 +2023-10-02 21:11:44,189 - Epoch: [89][ 420/ 1236] Overall Loss 0.266500 Objective Loss 0.266500 LR 0.001000 Time 0.022661 +2023-10-02 21:11:44,392 - Epoch: [89][ 430/ 1236] Overall Loss 0.266023 Objective Loss 0.266023 LR 0.001000 Time 0.022608 +2023-10-02 21:11:44,596 - Epoch: [89][ 440/ 1236] Overall Loss 0.265795 Objective Loss 0.265795 LR 0.001000 Time 0.022556 +2023-10-02 21:11:44,801 - Epoch: [89][ 450/ 1236] Overall Loss 0.266002 Objective Loss 0.266002 LR 0.001000 Time 0.022510 +2023-10-02 21:11:45,007 - Epoch: [89][ 460/ 1236] Overall Loss 0.265673 Objective Loss 0.265673 LR 0.001000 Time 0.022467 +2023-10-02 21:11:45,212 - Epoch: [89][ 470/ 1236] Overall Loss 0.265479 Objective Loss 0.265479 LR 0.001000 Time 0.022423 +2023-10-02 21:11:45,419 - Epoch: [89][ 480/ 1236] Overall Loss 0.265698 Objective Loss 0.265698 LR 0.001000 Time 0.022386 +2023-10-02 21:11:45,623 - Epoch: [89][ 490/ 1236] Overall Loss 0.265792 Objective Loss 0.265792 LR 0.001000 Time 0.022345 +2023-10-02 21:11:45,829 - Epoch: [89][ 500/ 1236] Overall Loss 0.266149 Objective Loss 0.266149 LR 0.001000 Time 0.022309 +2023-10-02 21:11:46,034 - Epoch: [89][ 510/ 1236] Overall Loss 0.266221 Objective Loss 0.266221 LR 0.001000 Time 0.022270 +2023-10-02 21:11:46,241 - Epoch: [89][ 520/ 1236] Overall Loss 0.266623 Objective Loss 0.266623 LR 0.001000 Time 0.022240 +2023-10-02 21:11:46,445 - Epoch: [89][ 530/ 1236] Overall Loss 0.267348 Objective Loss 0.267348 LR 0.001000 Time 0.022205 +2023-10-02 21:11:46,651 - Epoch: [89][ 540/ 1236] Overall Loss 0.267126 Objective Loss 0.267126 LR 0.001000 Time 0.022174 +2023-10-02 21:11:46,856 - Epoch: [89][ 550/ 1236] Overall Loss 0.267576 Objective Loss 0.267576 LR 0.001000 Time 0.022141 +2023-10-02 21:11:47,062 - Epoch: [89][ 560/ 1236] Overall Loss 0.267721 Objective Loss 0.267721 LR 0.001000 Time 0.022113 +2023-10-02 21:11:47,267 - Epoch: [89][ 570/ 1236] Overall Loss 0.267002 Objective Loss 0.267002 LR 0.001000 Time 0.022082 +2023-10-02 21:11:47,472 - Epoch: [89][ 580/ 1236] Overall Loss 0.267059 Objective Loss 0.267059 LR 0.001000 Time 0.022055 +2023-10-02 21:11:47,677 - Epoch: [89][ 590/ 1236] Overall Loss 0.267308 Objective Loss 0.267308 LR 0.001000 Time 0.022026 +2023-10-02 21:11:47,884 - Epoch: [89][ 600/ 1236] Overall Loss 0.267425 Objective Loss 0.267425 LR 0.001000 Time 0.022004 +2023-10-02 21:11:48,087 - Epoch: [89][ 610/ 1236] Overall Loss 0.266969 Objective Loss 0.266969 LR 0.001000 Time 0.021976 +2023-10-02 21:11:48,295 - Epoch: [89][ 620/ 1236] Overall Loss 0.267207 Objective Loss 0.267207 LR 0.001000 Time 0.021955 +2023-10-02 21:11:48,497 - Epoch: [89][ 630/ 1236] Overall Loss 0.266746 Objective Loss 0.266746 LR 0.001000 Time 0.021928 +2023-10-02 21:11:48,703 - Epoch: [89][ 640/ 1236] Overall Loss 0.266992 Objective Loss 0.266992 LR 0.001000 Time 0.021906 +2023-10-02 21:11:48,908 - Epoch: [89][ 650/ 1236] Overall Loss 0.267682 Objective Loss 0.267682 LR 0.001000 Time 0.021883 +2023-10-02 21:11:49,115 - Epoch: [89][ 660/ 1236] Overall Loss 0.267665 Objective Loss 0.267665 LR 0.001000 Time 0.021864 +2023-10-02 21:11:49,320 - Epoch: [89][ 670/ 1236] Overall Loss 0.267564 Objective Loss 0.267564 LR 0.001000 Time 0.021842 +2023-10-02 21:11:49,524 - Epoch: [89][ 680/ 1236] Overall Loss 0.267967 Objective Loss 0.267967 LR 0.001000 Time 0.021822 +2023-10-02 21:11:49,729 - Epoch: [89][ 690/ 1236] Overall Loss 0.267950 Objective Loss 0.267950 LR 0.001000 Time 0.021802 +2023-10-02 21:11:49,936 - Epoch: [89][ 700/ 1236] Overall Loss 0.267986 Objective Loss 0.267986 LR 0.001000 Time 0.021786 +2023-10-02 21:11:50,144 - Epoch: [89][ 710/ 1236] Overall Loss 0.267973 Objective Loss 0.267973 LR 0.001000 Time 0.021771 +2023-10-02 21:11:50,358 - Epoch: [89][ 720/ 1236] Overall Loss 0.267861 Objective Loss 0.267861 LR 0.001000 Time 0.021765 +2023-10-02 21:11:50,562 - Epoch: [89][ 730/ 1236] Overall Loss 0.268244 Objective Loss 0.268244 LR 0.001000 Time 0.021745 +2023-10-02 21:11:50,769 - Epoch: [89][ 740/ 1236] Overall Loss 0.267938 Objective Loss 0.267938 LR 0.001000 Time 0.021730 +2023-10-02 21:11:50,975 - Epoch: [89][ 750/ 1236] Overall Loss 0.267782 Objective Loss 0.267782 LR 0.001000 Time 0.021714 +2023-10-02 21:11:51,182 - Epoch: [89][ 760/ 1236] Overall Loss 0.267412 Objective Loss 0.267412 LR 0.001000 Time 0.021700 +2023-10-02 21:11:51,388 - Epoch: [89][ 770/ 1236] Overall Loss 0.267274 Objective Loss 0.267274 LR 0.001000 Time 0.021684 +2023-10-02 21:11:51,595 - Epoch: [89][ 780/ 1236] Overall Loss 0.267548 Objective Loss 0.267548 LR 0.001000 Time 0.021670 +2023-10-02 21:11:51,801 - Epoch: [89][ 790/ 1236] Overall Loss 0.267742 Objective Loss 0.267742 LR 0.001000 Time 0.021655 +2023-10-02 21:11:52,008 - Epoch: [89][ 800/ 1236] Overall Loss 0.268056 Objective Loss 0.268056 LR 0.001000 Time 0.021642 +2023-10-02 21:11:52,214 - Epoch: [89][ 810/ 1236] Overall Loss 0.268126 Objective Loss 0.268126 LR 0.001000 Time 0.021628 +2023-10-02 21:11:52,421 - Epoch: [89][ 820/ 1236] Overall Loss 0.267548 Objective Loss 0.267548 LR 0.001000 Time 0.021616 +2023-10-02 21:11:52,627 - Epoch: [89][ 830/ 1236] Overall Loss 0.267518 Objective Loss 0.267518 LR 0.001000 Time 0.021602 +2023-10-02 21:11:52,834 - Epoch: [89][ 840/ 1236] Overall Loss 0.267322 Objective Loss 0.267322 LR 0.001000 Time 0.021591 +2023-10-02 21:11:53,040 - Epoch: [89][ 850/ 1236] Overall Loss 0.267316 Objective Loss 0.267316 LR 0.001000 Time 0.021577 +2023-10-02 21:11:53,247 - Epoch: [89][ 860/ 1236] Overall Loss 0.267086 Objective Loss 0.267086 LR 0.001000 Time 0.021567 +2023-10-02 21:11:53,453 - Epoch: [89][ 870/ 1236] Overall Loss 0.266924 Objective Loss 0.266924 LR 0.001000 Time 0.021554 +2023-10-02 21:11:53,660 - Epoch: [89][ 880/ 1236] Overall Loss 0.266975 Objective Loss 0.266975 LR 0.001000 Time 0.021544 +2023-10-02 21:11:53,866 - Epoch: [89][ 890/ 1236] Overall Loss 0.266951 Objective Loss 0.266951 LR 0.001000 Time 0.021532 +2023-10-02 21:11:54,073 - Epoch: [89][ 900/ 1236] Overall Loss 0.266672 Objective Loss 0.266672 LR 0.001000 Time 0.021522 +2023-10-02 21:11:54,280 - Epoch: [89][ 910/ 1236] Overall Loss 0.266962 Objective Loss 0.266962 LR 0.001000 Time 0.021511 +2023-10-02 21:11:54,487 - Epoch: [89][ 920/ 1236] Overall Loss 0.266848 Objective Loss 0.266848 LR 0.001000 Time 0.021502 +2023-10-02 21:11:54,691 - Epoch: [89][ 930/ 1236] Overall Loss 0.267243 Objective Loss 0.267243 LR 0.001000 Time 0.021489 +2023-10-02 21:11:54,898 - Epoch: [89][ 940/ 1236] Overall Loss 0.267450 Objective Loss 0.267450 LR 0.001000 Time 0.021480 +2023-10-02 21:11:55,103 - Epoch: [89][ 950/ 1236] Overall Loss 0.267703 Objective Loss 0.267703 LR 0.001000 Time 0.021470 +2023-10-02 21:11:55,310 - Epoch: [89][ 960/ 1236] Overall Loss 0.267661 Objective Loss 0.267661 LR 0.001000 Time 0.021461 +2023-10-02 21:11:55,516 - Epoch: [89][ 970/ 1236] Overall Loss 0.267223 Objective Loss 0.267223 LR 0.001000 Time 0.021451 +2023-10-02 21:11:55,723 - Epoch: [89][ 980/ 1236] Overall Loss 0.266961 Objective Loss 0.266961 LR 0.001000 Time 0.021442 +2023-10-02 21:11:55,929 - Epoch: [89][ 990/ 1236] Overall Loss 0.266843 Objective Loss 0.266843 LR 0.001000 Time 0.021432 +2023-10-02 21:11:56,136 - Epoch: [89][ 1000/ 1236] Overall Loss 0.266788 Objective Loss 0.266788 LR 0.001000 Time 0.021425 +2023-10-02 21:11:56,342 - Epoch: [89][ 1010/ 1236] Overall Loss 0.266818 Objective Loss 0.266818 LR 0.001000 Time 0.021415 +2023-10-02 21:11:56,549 - Epoch: [89][ 1020/ 1236] Overall Loss 0.266966 Objective Loss 0.266966 LR 0.001000 Time 0.021408 +2023-10-02 21:11:56,755 - Epoch: [89][ 1030/ 1236] Overall Loss 0.267255 Objective Loss 0.267255 LR 0.001000 Time 0.021399 +2023-10-02 21:11:56,962 - Epoch: [89][ 1040/ 1236] Overall Loss 0.267303 Objective Loss 0.267303 LR 0.001000 Time 0.021391 +2023-10-02 21:11:57,168 - Epoch: [89][ 1050/ 1236] Overall Loss 0.267211 Objective Loss 0.267211 LR 0.001000 Time 0.021383 +2023-10-02 21:11:57,375 - Epoch: [89][ 1060/ 1236] Overall Loss 0.267238 Objective Loss 0.267238 LR 0.001000 Time 0.021376 +2023-10-02 21:11:57,581 - Epoch: [89][ 1070/ 1236] Overall Loss 0.267417 Objective Loss 0.267417 LR 0.001000 Time 0.021367 +2023-10-02 21:11:57,788 - Epoch: [89][ 1080/ 1236] Overall Loss 0.267344 Objective Loss 0.267344 LR 0.001000 Time 0.021361 +2023-10-02 21:11:57,993 - Epoch: [89][ 1090/ 1236] Overall Loss 0.267075 Objective Loss 0.267075 LR 0.001000 Time 0.021351 +2023-10-02 21:11:58,200 - Epoch: [89][ 1100/ 1236] Overall Loss 0.267093 Objective Loss 0.267093 LR 0.001000 Time 0.021345 +2023-10-02 21:11:58,406 - Epoch: [89][ 1110/ 1236] Overall Loss 0.266842 Objective Loss 0.266842 LR 0.001000 Time 0.021338 +2023-10-02 21:11:58,613 - Epoch: [89][ 1120/ 1236] Overall Loss 0.266696 Objective Loss 0.266696 LR 0.001000 Time 0.021331 +2023-10-02 21:11:58,819 - Epoch: [89][ 1130/ 1236] Overall Loss 0.266494 Objective Loss 0.266494 LR 0.001000 Time 0.021324 +2023-10-02 21:11:59,026 - Epoch: [89][ 1140/ 1236] Overall Loss 0.266401 Objective Loss 0.266401 LR 0.001000 Time 0.021318 +2023-10-02 21:11:59,233 - Epoch: [89][ 1150/ 1236] Overall Loss 0.266245 Objective Loss 0.266245 LR 0.001000 Time 0.021311 +2023-10-02 21:11:59,439 - Epoch: [89][ 1160/ 1236] Overall Loss 0.266344 Objective Loss 0.266344 LR 0.001000 Time 0.021305 +2023-10-02 21:11:59,646 - Epoch: [89][ 1170/ 1236] Overall Loss 0.266457 Objective Loss 0.266457 LR 0.001000 Time 0.021298 +2023-10-02 21:11:59,853 - Epoch: [89][ 1180/ 1236] Overall Loss 0.266486 Objective Loss 0.266486 LR 0.001000 Time 0.021293 +2023-10-02 21:12:00,059 - Epoch: [89][ 1190/ 1236] Overall Loss 0.266884 Objective Loss 0.266884 LR 0.001000 Time 0.021286 +2023-10-02 21:12:00,266 - Epoch: [89][ 1200/ 1236] Overall Loss 0.267134 Objective Loss 0.267134 LR 0.001000 Time 0.021281 +2023-10-02 21:12:00,473 - Epoch: [89][ 1210/ 1236] Overall Loss 0.267194 Objective Loss 0.267194 LR 0.001000 Time 0.021274 +2023-10-02 21:12:00,679 - Epoch: [89][ 1220/ 1236] Overall Loss 0.267397 Objective Loss 0.267397 LR 0.001000 Time 0.021269 +2023-10-02 21:12:00,938 - Epoch: [89][ 1230/ 1236] Overall Loss 0.267383 Objective Loss 0.267383 LR 0.001000 Time 0.021305 +2023-10-02 21:12:01,060 - Epoch: [89][ 1236/ 1236] Overall Loss 0.267319 Objective Loss 0.267319 Top1 84.521385 Top5 98.167006 LR 0.001000 Time 0.021300 +2023-10-02 21:12:01,187 - --- validate (epoch=89)----------- +2023-10-02 21:12:01,188 - 29943 samples (256 per mini-batch) +2023-10-02 21:12:01,689 - Epoch: [89][ 10/ 117] Loss 0.311406 Top1 82.812500 Top5 98.398438 +2023-10-02 21:12:01,843 - Epoch: [89][ 20/ 117] Loss 0.341704 Top1 82.265625 Top5 97.988281 +2023-10-02 21:12:01,996 - Epoch: [89][ 30/ 117] Loss 0.339940 Top1 82.382812 Top5 98.007812 +2023-10-02 21:12:02,149 - Epoch: [89][ 40/ 117] Loss 0.354769 Top1 81.611328 Top5 97.812500 +2023-10-02 21:12:02,301 - Epoch: [89][ 50/ 117] Loss 0.349194 Top1 81.789062 Top5 97.796875 +2023-10-02 21:12:02,454 - Epoch: [89][ 60/ 117] Loss 0.345724 Top1 81.725260 Top5 97.792969 +2023-10-02 21:12:02,606 - Epoch: [89][ 70/ 117] Loss 0.345059 Top1 81.847098 Top5 97.845982 +2023-10-02 21:12:02,759 - Epoch: [89][ 80/ 117] Loss 0.348908 Top1 81.733398 Top5 97.836914 +2023-10-02 21:12:02,911 - Epoch: [89][ 90/ 117] Loss 0.344542 Top1 81.857639 Top5 97.868924 +2023-10-02 21:12:03,063 - Epoch: [89][ 100/ 117] Loss 0.340741 Top1 82.015625 Top5 97.871094 +2023-10-02 21:12:03,220 - Epoch: [89][ 110/ 117] Loss 0.338344 Top1 82.130682 Top5 97.894176 +2023-10-02 21:12:03,309 - Epoch: [89][ 117/ 117] Loss 0.336436 Top1 82.189493 Top5 97.916040 +2023-10-02 21:12:03,412 - ==> Top1: 82.189 Top5: 97.916 Loss: 0.336 + +2023-10-02 21:12:03,413 - ==> Confusion: +[[ 943 0 4 2 1 5 0 0 10 66 1 0 0 3 6 0 1 0 0 0 8] + [ 0 1045 0 1 9 24 1 15 4 2 4 1 0 0 3 5 0 0 13 2 2] + [ 6 0 945 18 0 1 28 10 0 4 3 2 9 1 1 8 1 1 9 2 7] + [ 1 3 13 944 0 3 2 3 5 1 17 1 8 3 47 1 2 5 16 0 14] + [ 34 8 1 0 946 6 0 0 1 17 4 1 0 2 12 5 10 0 1 1 1] + [ 5 52 2 2 4 966 0 19 3 3 1 12 1 15 3 0 4 0 5 9 10] + [ 0 3 24 0 0 1 1117 5 0 0 5 2 1 1 0 10 0 3 0 15 4] + [ 7 19 13 0 2 31 7 1040 1 3 6 8 1 2 1 4 4 1 55 8 5] + [ 17 5 0 0 0 1 0 2 967 37 18 2 3 7 18 2 2 0 6 1 1] + [ 93 0 2 0 5 2 0 0 43 925 4 1 0 26 7 4 0 1 0 2 4] + [ 1 1 11 3 0 2 2 3 13 1 977 2 1 14 6 0 1 2 6 4 3] + [ 1 0 3 0 1 13 0 4 0 1 0 943 35 6 0 5 0 17 1 5 0] + [ 0 0 1 3 0 0 3 0 1 0 5 45 961 0 3 11 1 16 3 5 10] + [ 2 0 1 0 1 12 1 0 15 12 10 9 1 1039 6 0 1 1 0 1 7] + [ 10 3 2 9 2 0 0 0 21 1 7 0 4 3 1023 0 3 2 9 0 2] + [ 0 0 0 2 3 0 1 0 0 0 1 6 5 0 2 1075 12 12 1 8 6] + [ 2 24 1 1 8 6 0 0 1 1 2 3 2 3 3 12 1078 1 2 6 5] + [ 0 0 0 3 0 0 1 0 1 3 1 5 19 0 5 6 0 988 1 2 3] + [ 3 3 2 10 2 1 2 13 5 0 4 0 2 0 9 0 0 0 1003 0 9] + [ 0 2 2 0 1 1 7 7 0 0 2 13 6 3 0 4 6 0 0 1094 4] + [ 170 212 157 102 76 181 44 101 132 107 254 124 367 339 208 83 108 79 216 254 4591]] + +2023-10-02 21:12:03,414 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:12:03,414 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:12:03,420 - + +2023-10-02 21:12:03,420 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:12:04,437 - Epoch: [90][ 10/ 1236] Overall Loss 0.245287 Objective Loss 0.245287 LR 0.001000 Time 0.101608 +2023-10-02 21:12:04,646 - Epoch: [90][ 20/ 1236] Overall Loss 0.244397 Objective Loss 0.244397 LR 0.001000 Time 0.061254 +2023-10-02 21:12:04,854 - Epoch: [90][ 30/ 1236] Overall Loss 0.241839 Objective Loss 0.241839 LR 0.001000 Time 0.047743 +2023-10-02 21:12:05,063 - Epoch: [90][ 40/ 1236] Overall Loss 0.243193 Objective Loss 0.243193 LR 0.001000 Time 0.041021 +2023-10-02 21:12:05,270 - Epoch: [90][ 50/ 1236] Overall Loss 0.246748 Objective Loss 0.246748 LR 0.001000 Time 0.036950 +2023-10-02 21:12:05,479 - Epoch: [90][ 60/ 1236] Overall Loss 0.247038 Objective Loss 0.247038 LR 0.001000 Time 0.034267 +2023-10-02 21:12:05,685 - Epoch: [90][ 70/ 1236] Overall Loss 0.246366 Objective Loss 0.246366 LR 0.001000 Time 0.032321 +2023-10-02 21:12:05,894 - Epoch: [90][ 80/ 1236] Overall Loss 0.249328 Objective Loss 0.249328 LR 0.001000 Time 0.030883 +2023-10-02 21:12:06,101 - Epoch: [90][ 90/ 1236] Overall Loss 0.251500 Objective Loss 0.251500 LR 0.001000 Time 0.029746 +2023-10-02 21:12:06,310 - Epoch: [90][ 100/ 1236] Overall Loss 0.253605 Objective Loss 0.253605 LR 0.001000 Time 0.028858 +2023-10-02 21:12:06,516 - Epoch: [90][ 110/ 1236] Overall Loss 0.255806 Objective Loss 0.255806 LR 0.001000 Time 0.028098 +2023-10-02 21:12:06,726 - Epoch: [90][ 120/ 1236] Overall Loss 0.256078 Objective Loss 0.256078 LR 0.001000 Time 0.027502 +2023-10-02 21:12:06,931 - Epoch: [90][ 130/ 1236] Overall Loss 0.257716 Objective Loss 0.257716 LR 0.001000 Time 0.026963 +2023-10-02 21:12:07,140 - Epoch: [90][ 140/ 1236] Overall Loss 0.256710 Objective Loss 0.256710 LR 0.001000 Time 0.026524 +2023-10-02 21:12:07,347 - Epoch: [90][ 150/ 1236] Overall Loss 0.255887 Objective Loss 0.255887 LR 0.001000 Time 0.026126 +2023-10-02 21:12:07,556 - Epoch: [90][ 160/ 1236] Overall Loss 0.254039 Objective Loss 0.254039 LR 0.001000 Time 0.025801 +2023-10-02 21:12:07,762 - Epoch: [90][ 170/ 1236] Overall Loss 0.254969 Objective Loss 0.254969 LR 0.001000 Time 0.025490 +2023-10-02 21:12:07,971 - Epoch: [90][ 180/ 1236] Overall Loss 0.254226 Objective Loss 0.254226 LR 0.001000 Time 0.025238 +2023-10-02 21:12:08,177 - Epoch: [90][ 190/ 1236] Overall Loss 0.254118 Objective Loss 0.254118 LR 0.001000 Time 0.024991 +2023-10-02 21:12:08,386 - Epoch: [90][ 200/ 1236] Overall Loss 0.254291 Objective Loss 0.254291 LR 0.001000 Time 0.024784 +2023-10-02 21:12:08,593 - Epoch: [90][ 210/ 1236] Overall Loss 0.254955 Objective Loss 0.254955 LR 0.001000 Time 0.024581 +2023-10-02 21:12:08,801 - Epoch: [90][ 220/ 1236] Overall Loss 0.255996 Objective Loss 0.255996 LR 0.001000 Time 0.024411 +2023-10-02 21:12:09,008 - Epoch: [90][ 230/ 1236] Overall Loss 0.256200 Objective Loss 0.256200 LR 0.001000 Time 0.024242 +2023-10-02 21:12:09,218 - Epoch: [90][ 240/ 1236] Overall Loss 0.255911 Objective Loss 0.255911 LR 0.001000 Time 0.024106 +2023-10-02 21:12:09,424 - Epoch: [90][ 250/ 1236] Overall Loss 0.257148 Objective Loss 0.257148 LR 0.001000 Time 0.023964 +2023-10-02 21:12:09,634 - Epoch: [90][ 260/ 1236] Overall Loss 0.257940 Objective Loss 0.257940 LR 0.001000 Time 0.023849 +2023-10-02 21:12:09,839 - Epoch: [90][ 270/ 1236] Overall Loss 0.257860 Objective Loss 0.257860 LR 0.001000 Time 0.023725 +2023-10-02 21:12:10,049 - Epoch: [90][ 280/ 1236] Overall Loss 0.257549 Objective Loss 0.257549 LR 0.001000 Time 0.023625 +2023-10-02 21:12:10,256 - Epoch: [90][ 290/ 1236] Overall Loss 0.258345 Objective Loss 0.258345 LR 0.001000 Time 0.023519 +2023-10-02 21:12:10,465 - Epoch: [90][ 300/ 1236] Overall Loss 0.258562 Objective Loss 0.258562 LR 0.001000 Time 0.023430 +2023-10-02 21:12:10,672 - Epoch: [90][ 310/ 1236] Overall Loss 0.259218 Objective Loss 0.259218 LR 0.001000 Time 0.023337 +2023-10-02 21:12:10,881 - Epoch: [90][ 320/ 1236] Overall Loss 0.260825 Objective Loss 0.260825 LR 0.001000 Time 0.023259 +2023-10-02 21:12:11,088 - Epoch: [90][ 330/ 1236] Overall Loss 0.260878 Objective Loss 0.260878 LR 0.001000 Time 0.023178 +2023-10-02 21:12:11,297 - Epoch: [90][ 340/ 1236] Overall Loss 0.261041 Objective Loss 0.261041 LR 0.001000 Time 0.023109 +2023-10-02 21:12:11,503 - Epoch: [90][ 350/ 1236] Overall Loss 0.262013 Objective Loss 0.262013 LR 0.001000 Time 0.023036 +2023-10-02 21:12:11,712 - Epoch: [90][ 360/ 1236] Overall Loss 0.262648 Objective Loss 0.262648 LR 0.001000 Time 0.022975 +2023-10-02 21:12:11,919 - Epoch: [90][ 370/ 1236] Overall Loss 0.263387 Objective Loss 0.263387 LR 0.001000 Time 0.022910 +2023-10-02 21:12:12,128 - Epoch: [90][ 380/ 1236] Overall Loss 0.263926 Objective Loss 0.263926 LR 0.001000 Time 0.022856 +2023-10-02 21:12:12,336 - Epoch: [90][ 390/ 1236] Overall Loss 0.264576 Objective Loss 0.264576 LR 0.001000 Time 0.022801 +2023-10-02 21:12:12,544 - Epoch: [90][ 400/ 1236] Overall Loss 0.265010 Objective Loss 0.265010 LR 0.001000 Time 0.022752 +2023-10-02 21:12:12,751 - Epoch: [90][ 410/ 1236] Overall Loss 0.265306 Objective Loss 0.265306 LR 0.001000 Time 0.022701 +2023-10-02 21:12:12,960 - Epoch: [90][ 420/ 1236] Overall Loss 0.265723 Objective Loss 0.265723 LR 0.001000 Time 0.022657 +2023-10-02 21:12:13,167 - Epoch: [90][ 430/ 1236] Overall Loss 0.265780 Objective Loss 0.265780 LR 0.001000 Time 0.022611 +2023-10-02 21:12:13,376 - Epoch: [90][ 440/ 1236] Overall Loss 0.265948 Objective Loss 0.265948 LR 0.001000 Time 0.022571 +2023-10-02 21:12:13,583 - Epoch: [90][ 450/ 1236] Overall Loss 0.265510 Objective Loss 0.265510 LR 0.001000 Time 0.022529 +2023-10-02 21:12:13,793 - Epoch: [90][ 460/ 1236] Overall Loss 0.265616 Objective Loss 0.265616 LR 0.001000 Time 0.022495 +2023-10-02 21:12:13,999 - Epoch: [90][ 470/ 1236] Overall Loss 0.265748 Objective Loss 0.265748 LR 0.001000 Time 0.022454 +2023-10-02 21:12:14,208 - Epoch: [90][ 480/ 1236] Overall Loss 0.265502 Objective Loss 0.265502 LR 0.001000 Time 0.022421 +2023-10-02 21:12:14,415 - Epoch: [90][ 490/ 1236] Overall Loss 0.265774 Objective Loss 0.265774 LR 0.001000 Time 0.022383 +2023-10-02 21:12:14,624 - Epoch: [90][ 500/ 1236] Overall Loss 0.266003 Objective Loss 0.266003 LR 0.001000 Time 0.022352 +2023-10-02 21:12:14,831 - Epoch: [90][ 510/ 1236] Overall Loss 0.265528 Objective Loss 0.265528 LR 0.001000 Time 0.022320 +2023-10-02 21:12:15,041 - Epoch: [90][ 520/ 1236] Overall Loss 0.266130 Objective Loss 0.266130 LR 0.001000 Time 0.022293 +2023-10-02 21:12:15,248 - Epoch: [90][ 530/ 1236] Overall Loss 0.266052 Objective Loss 0.266052 LR 0.001000 Time 0.022262 +2023-10-02 21:12:15,457 - Epoch: [90][ 540/ 1236] Overall Loss 0.266223 Objective Loss 0.266223 LR 0.001000 Time 0.022236 +2023-10-02 21:12:15,664 - Epoch: [90][ 550/ 1236] Overall Loss 0.265813 Objective Loss 0.265813 LR 0.001000 Time 0.022208 +2023-10-02 21:12:15,873 - Epoch: [90][ 560/ 1236] Overall Loss 0.265922 Objective Loss 0.265922 LR 0.001000 Time 0.022184 +2023-10-02 21:12:16,080 - Epoch: [90][ 570/ 1236] Overall Loss 0.265511 Objective Loss 0.265511 LR 0.001000 Time 0.022158 +2023-10-02 21:12:16,289 - Epoch: [90][ 580/ 1236] Overall Loss 0.265732 Objective Loss 0.265732 LR 0.001000 Time 0.022135 +2023-10-02 21:12:16,496 - Epoch: [90][ 590/ 1236] Overall Loss 0.266117 Objective Loss 0.266117 LR 0.001000 Time 0.022111 +2023-10-02 21:12:16,705 - Epoch: [90][ 600/ 1236] Overall Loss 0.266186 Objective Loss 0.266186 LR 0.001000 Time 0.022090 +2023-10-02 21:12:16,912 - Epoch: [90][ 610/ 1236] Overall Loss 0.266323 Objective Loss 0.266323 LR 0.001000 Time 0.022068 +2023-10-02 21:12:17,121 - Epoch: [90][ 620/ 1236] Overall Loss 0.266284 Objective Loss 0.266284 LR 0.001000 Time 0.022048 +2023-10-02 21:12:17,328 - Epoch: [90][ 630/ 1236] Overall Loss 0.266076 Objective Loss 0.266076 LR 0.001000 Time 0.022026 +2023-10-02 21:12:17,538 - Epoch: [90][ 640/ 1236] Overall Loss 0.265992 Objective Loss 0.265992 LR 0.001000 Time 0.022010 +2023-10-02 21:12:17,745 - Epoch: [90][ 650/ 1236] Overall Loss 0.266022 Objective Loss 0.266022 LR 0.001000 Time 0.021988 +2023-10-02 21:12:17,953 - Epoch: [90][ 660/ 1236] Overall Loss 0.265647 Objective Loss 0.265647 LR 0.001000 Time 0.021970 +2023-10-02 21:12:18,160 - Epoch: [90][ 670/ 1236] Overall Loss 0.265642 Objective Loss 0.265642 LR 0.001000 Time 0.021951 +2023-10-02 21:12:18,370 - Epoch: [90][ 680/ 1236] Overall Loss 0.266408 Objective Loss 0.266408 LR 0.001000 Time 0.021937 +2023-10-02 21:12:18,576 - Epoch: [90][ 690/ 1236] Overall Loss 0.266242 Objective Loss 0.266242 LR 0.001000 Time 0.021917 +2023-10-02 21:12:18,785 - Epoch: [90][ 700/ 1236] Overall Loss 0.266290 Objective Loss 0.266290 LR 0.001000 Time 0.021902 +2023-10-02 21:12:18,990 - Epoch: [90][ 710/ 1236] Overall Loss 0.266721 Objective Loss 0.266721 LR 0.001000 Time 0.021881 +2023-10-02 21:12:19,198 - Epoch: [90][ 720/ 1236] Overall Loss 0.267010 Objective Loss 0.267010 LR 0.001000 Time 0.021866 +2023-10-02 21:12:19,403 - Epoch: [90][ 730/ 1236] Overall Loss 0.266818 Objective Loss 0.266818 LR 0.001000 Time 0.021846 +2023-10-02 21:12:19,611 - Epoch: [90][ 740/ 1236] Overall Loss 0.266755 Objective Loss 0.266755 LR 0.001000 Time 0.021832 +2023-10-02 21:12:19,815 - Epoch: [90][ 750/ 1236] Overall Loss 0.266743 Objective Loss 0.266743 LR 0.001000 Time 0.021813 +2023-10-02 21:12:20,022 - Epoch: [90][ 760/ 1236] Overall Loss 0.266348 Objective Loss 0.266348 LR 0.001000 Time 0.021798 +2023-10-02 21:12:20,228 - Epoch: [90][ 770/ 1236] Overall Loss 0.266312 Objective Loss 0.266312 LR 0.001000 Time 0.021781 +2023-10-02 21:12:20,435 - Epoch: [90][ 780/ 1236] Overall Loss 0.266467 Objective Loss 0.266467 LR 0.001000 Time 0.021767 +2023-10-02 21:12:20,640 - Epoch: [90][ 790/ 1236] Overall Loss 0.266249 Objective Loss 0.266249 LR 0.001000 Time 0.021751 +2023-10-02 21:12:20,847 - Epoch: [90][ 800/ 1236] Overall Loss 0.265725 Objective Loss 0.265725 LR 0.001000 Time 0.021738 +2023-10-02 21:12:21,053 - Epoch: [90][ 810/ 1236] Overall Loss 0.265481 Objective Loss 0.265481 LR 0.001000 Time 0.021723 +2023-10-02 21:12:21,261 - Epoch: [90][ 820/ 1236] Overall Loss 0.265812 Objective Loss 0.265812 LR 0.001000 Time 0.021712 +2023-10-02 21:12:21,469 - Epoch: [90][ 830/ 1236] Overall Loss 0.265608 Objective Loss 0.265608 LR 0.001000 Time 0.021699 +2023-10-02 21:12:21,677 - Epoch: [90][ 840/ 1236] Overall Loss 0.265792 Objective Loss 0.265792 LR 0.001000 Time 0.021687 +2023-10-02 21:12:21,883 - Epoch: [90][ 850/ 1236] Overall Loss 0.266011 Objective Loss 0.266011 LR 0.001000 Time 0.021673 +2023-10-02 21:12:22,089 - Epoch: [90][ 860/ 1236] Overall Loss 0.266009 Objective Loss 0.266009 LR 0.001000 Time 0.021661 +2023-10-02 21:12:22,295 - Epoch: [90][ 870/ 1236] Overall Loss 0.265900 Objective Loss 0.265900 LR 0.001000 Time 0.021648 +2023-10-02 21:12:22,502 - Epoch: [90][ 880/ 1236] Overall Loss 0.265495 Objective Loss 0.265495 LR 0.001000 Time 0.021637 +2023-10-02 21:12:22,708 - Epoch: [90][ 890/ 1236] Overall Loss 0.265822 Objective Loss 0.265822 LR 0.001000 Time 0.021624 +2023-10-02 21:12:22,915 - Epoch: [90][ 900/ 1236] Overall Loss 0.266199 Objective Loss 0.266199 LR 0.001000 Time 0.021614 +2023-10-02 21:12:23,121 - Epoch: [90][ 910/ 1236] Overall Loss 0.266366 Objective Loss 0.266366 LR 0.001000 Time 0.021602 +2023-10-02 21:12:23,328 - Epoch: [90][ 920/ 1236] Overall Loss 0.266466 Objective Loss 0.266466 LR 0.001000 Time 0.021592 +2023-10-02 21:12:23,533 - Epoch: [90][ 930/ 1236] Overall Loss 0.266681 Objective Loss 0.266681 LR 0.001000 Time 0.021580 +2023-10-02 21:12:23,740 - Epoch: [90][ 940/ 1236] Overall Loss 0.267252 Objective Loss 0.267252 LR 0.001000 Time 0.021571 +2023-10-02 21:12:23,946 - Epoch: [90][ 950/ 1236] Overall Loss 0.267008 Objective Loss 0.267008 LR 0.001000 Time 0.021560 +2023-10-02 21:12:24,153 - Epoch: [90][ 960/ 1236] Overall Loss 0.267440 Objective Loss 0.267440 LR 0.001000 Time 0.021550 +2023-10-02 21:12:24,358 - Epoch: [90][ 970/ 1236] Overall Loss 0.267772 Objective Loss 0.267772 LR 0.001000 Time 0.021540 +2023-10-02 21:12:24,565 - Epoch: [90][ 980/ 1236] Overall Loss 0.268055 Objective Loss 0.268055 LR 0.001000 Time 0.021531 +2023-10-02 21:12:24,771 - Epoch: [90][ 990/ 1236] Overall Loss 0.268058 Objective Loss 0.268058 LR 0.001000 Time 0.021521 +2023-10-02 21:12:24,978 - Epoch: [90][ 1000/ 1236] Overall Loss 0.268242 Objective Loss 0.268242 LR 0.001000 Time 0.021512 +2023-10-02 21:12:25,184 - Epoch: [90][ 1010/ 1236] Overall Loss 0.268148 Objective Loss 0.268148 LR 0.001000 Time 0.021503 +2023-10-02 21:12:25,391 - Epoch: [90][ 1020/ 1236] Overall Loss 0.268128 Objective Loss 0.268128 LR 0.001000 Time 0.021495 +2023-10-02 21:12:25,596 - Epoch: [90][ 1030/ 1236] Overall Loss 0.268362 Objective Loss 0.268362 LR 0.001000 Time 0.021485 +2023-10-02 21:12:25,803 - Epoch: [90][ 1040/ 1236] Overall Loss 0.268629 Objective Loss 0.268629 LR 0.001000 Time 0.021477 +2023-10-02 21:12:26,009 - Epoch: [90][ 1050/ 1236] Overall Loss 0.268724 Objective Loss 0.268724 LR 0.001000 Time 0.021468 +2023-10-02 21:12:26,216 - Epoch: [90][ 1060/ 1236] Overall Loss 0.268874 Objective Loss 0.268874 LR 0.001000 Time 0.021461 +2023-10-02 21:12:26,422 - Epoch: [90][ 1070/ 1236] Overall Loss 0.268948 Objective Loss 0.268948 LR 0.001000 Time 0.021451 +2023-10-02 21:12:26,628 - Epoch: [90][ 1080/ 1236] Overall Loss 0.268755 Objective Loss 0.268755 LR 0.001000 Time 0.021443 +2023-10-02 21:12:26,834 - Epoch: [90][ 1090/ 1236] Overall Loss 0.268762 Objective Loss 0.268762 LR 0.001000 Time 0.021436 +2023-10-02 21:12:27,041 - Epoch: [90][ 1100/ 1236] Overall Loss 0.268999 Objective Loss 0.268999 LR 0.001000 Time 0.021428 +2023-10-02 21:12:27,247 - Epoch: [90][ 1110/ 1236] Overall Loss 0.268458 Objective Loss 0.268458 LR 0.001000 Time 0.021421 +2023-10-02 21:12:27,454 - Epoch: [90][ 1120/ 1236] Overall Loss 0.268686 Objective Loss 0.268686 LR 0.001000 Time 0.021414 +2023-10-02 21:12:27,660 - Epoch: [90][ 1130/ 1236] Overall Loss 0.268464 Objective Loss 0.268464 LR 0.001000 Time 0.021406 +2023-10-02 21:12:27,868 - Epoch: [90][ 1140/ 1236] Overall Loss 0.268332 Objective Loss 0.268332 LR 0.001000 Time 0.021401 +2023-10-02 21:12:28,073 - Epoch: [90][ 1150/ 1236] Overall Loss 0.268551 Objective Loss 0.268551 LR 0.001000 Time 0.021392 +2023-10-02 21:12:28,280 - Epoch: [90][ 1160/ 1236] Overall Loss 0.268616 Objective Loss 0.268616 LR 0.001000 Time 0.021386 +2023-10-02 21:12:28,485 - Epoch: [90][ 1170/ 1236] Overall Loss 0.268719 Objective Loss 0.268719 LR 0.001000 Time 0.021378 +2023-10-02 21:12:28,691 - Epoch: [90][ 1180/ 1236] Overall Loss 0.268939 Objective Loss 0.268939 LR 0.001000 Time 0.021372 +2023-10-02 21:12:28,897 - Epoch: [90][ 1190/ 1236] Overall Loss 0.268961 Objective Loss 0.268961 LR 0.001000 Time 0.021365 +2023-10-02 21:12:29,104 - Epoch: [90][ 1200/ 1236] Overall Loss 0.268813 Objective Loss 0.268813 LR 0.001000 Time 0.021359 +2023-10-02 21:12:29,309 - Epoch: [90][ 1210/ 1236] Overall Loss 0.268711 Objective Loss 0.268711 LR 0.001000 Time 0.021352 +2023-10-02 21:12:29,517 - Epoch: [90][ 1220/ 1236] Overall Loss 0.269011 Objective Loss 0.269011 LR 0.001000 Time 0.021347 +2023-10-02 21:12:29,774 - Epoch: [90][ 1230/ 1236] Overall Loss 0.269066 Objective Loss 0.269066 LR 0.001000 Time 0.021381 +2023-10-02 21:12:29,895 - Epoch: [90][ 1236/ 1236] Overall Loss 0.269189 Objective Loss 0.269189 Top1 82.281059 Top5 97.352342 LR 0.001000 Time 0.021375 +2023-10-02 21:12:30,015 - --- validate (epoch=90)----------- +2023-10-02 21:12:30,015 - 29943 samples (256 per mini-batch) +2023-10-02 21:12:30,509 - Epoch: [90][ 10/ 117] Loss 0.359815 Top1 82.851562 Top5 97.734375 +2023-10-02 21:12:30,676 - Epoch: [90][ 20/ 117] Loss 0.350199 Top1 82.949219 Top5 97.714844 +2023-10-02 21:12:30,844 - Epoch: [90][ 30/ 117] Loss 0.353999 Top1 82.734375 Top5 97.578125 +2023-10-02 21:12:31,009 - Epoch: [90][ 40/ 117] Loss 0.345825 Top1 82.587891 Top5 97.705078 +2023-10-02 21:12:31,176 - Epoch: [90][ 50/ 117] Loss 0.344012 Top1 82.492188 Top5 97.812500 +2023-10-02 21:12:31,338 - Epoch: [90][ 60/ 117] Loss 0.345827 Top1 82.148438 Top5 97.786458 +2023-10-02 21:12:31,499 - Epoch: [90][ 70/ 117] Loss 0.342917 Top1 82.237723 Top5 97.767857 +2023-10-02 21:12:31,658 - Epoch: [90][ 80/ 117] Loss 0.344774 Top1 82.177734 Top5 97.807617 +2023-10-02 21:12:31,819 - Epoch: [90][ 90/ 117] Loss 0.348163 Top1 82.048611 Top5 97.782118 +2023-10-02 21:12:31,979 - Epoch: [90][ 100/ 117] Loss 0.348944 Top1 81.996094 Top5 97.777344 +2023-10-02 21:12:32,148 - Epoch: [90][ 110/ 117] Loss 0.349540 Top1 81.953125 Top5 97.755682 +2023-10-02 21:12:32,237 - Epoch: [90][ 117/ 117] Loss 0.349730 Top1 81.949037 Top5 97.749057 +2023-10-02 21:12:32,336 - ==> Top1: 81.949 Top5: 97.749 Loss: 0.350 + +2023-10-02 21:12:32,337 - ==> Confusion: +[[ 931 8 4 0 19 2 0 4 4 41 2 2 0 2 9 3 4 3 1 0 11] + [ 4 1051 2 0 4 32 1 19 2 1 1 1 1 1 1 3 0 0 5 0 2] + [ 2 2 956 9 2 1 32 8 0 4 2 0 5 4 1 5 0 1 11 4 7] + [ 2 4 13 943 1 4 2 2 6 1 12 0 3 5 31 2 1 9 35 0 13] + [ 17 10 0 1 981 6 1 1 0 8 0 0 0 2 6 8 6 1 2 0 0] + [ 2 44 1 2 4 972 0 26 2 4 3 5 5 15 3 1 0 2 5 8 12] + [ 0 1 21 1 0 1 1130 9 0 0 6 1 1 0 0 6 0 2 2 8 2] + [ 2 21 13 0 4 42 3 1053 2 0 6 5 1 6 2 3 1 0 40 6 8] + [ 21 8 3 2 2 2 0 0 951 34 22 1 1 10 16 3 2 1 10 0 0] + [ 122 2 1 2 10 2 1 0 36 892 4 0 0 28 8 3 0 0 0 2 6] + [ 3 2 13 4 2 0 1 3 11 1 975 1 0 10 3 3 0 3 9 3 6] + [ 0 1 3 0 0 15 0 6 0 1 1 936 34 5 0 2 1 15 0 10 5] + [ 0 0 3 7 0 4 3 5 1 1 3 59 925 0 7 8 1 22 9 4 6] + [ 1 1 3 0 5 12 0 0 18 13 7 5 0 1033 6 2 1 0 0 1 11] + [ 10 3 2 20 13 0 0 0 21 4 7 1 2 5 979 0 3 4 18 0 9] + [ 0 0 1 0 6 1 2 0 0 0 0 8 7 0 0 1068 17 14 1 7 2] + [ 1 21 0 0 8 11 0 0 3 1 1 5 1 2 1 14 1076 0 1 6 9] + [ 0 2 0 0 0 0 0 0 0 1 0 4 13 2 1 10 0 998 3 2 2] + [ 1 7 5 11 1 2 1 14 1 0 5 1 1 0 8 0 0 1 999 0 10] + [ 0 4 3 1 0 6 11 19 0 0 4 13 4 0 0 3 6 1 1 1072 4] + [ 130 309 163 80 120 221 66 104 153 67 244 130 375 292 145 97 91 66 234 201 4617]] + +2023-10-02 21:12:32,338 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:12:32,338 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:12:32,345 - + +2023-10-02 21:12:32,345 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:12:33,353 - Epoch: [91][ 10/ 1236] Overall Loss 0.249520 Objective Loss 0.249520 LR 0.001000 Time 0.100789 +2023-10-02 21:12:33,562 - Epoch: [91][ 20/ 1236] Overall Loss 0.252578 Objective Loss 0.252578 LR 0.001000 Time 0.060793 +2023-10-02 21:12:33,773 - Epoch: [91][ 30/ 1236] Overall Loss 0.265598 Objective Loss 0.265598 LR 0.001000 Time 0.047527 +2023-10-02 21:12:33,991 - Epoch: [91][ 40/ 1236] Overall Loss 0.264350 Objective Loss 0.264350 LR 0.001000 Time 0.041094 +2023-10-02 21:12:34,214 - Epoch: [91][ 50/ 1236] Overall Loss 0.264468 Objective Loss 0.264468 LR 0.001000 Time 0.037313 +2023-10-02 21:12:34,432 - Epoch: [91][ 60/ 1236] Overall Loss 0.262453 Objective Loss 0.262453 LR 0.001000 Time 0.034728 +2023-10-02 21:12:34,644 - Epoch: [91][ 70/ 1236] Overall Loss 0.265719 Objective Loss 0.265719 LR 0.001000 Time 0.032786 +2023-10-02 21:12:34,852 - Epoch: [91][ 80/ 1236] Overall Loss 0.270974 Objective Loss 0.270974 LR 0.001000 Time 0.031283 +2023-10-02 21:12:35,063 - Epoch: [91][ 90/ 1236] Overall Loss 0.269735 Objective Loss 0.269735 LR 0.001000 Time 0.030154 +2023-10-02 21:12:35,272 - Epoch: [91][ 100/ 1236] Overall Loss 0.267194 Objective Loss 0.267194 LR 0.001000 Time 0.029215 +2023-10-02 21:12:35,483 - Epoch: [91][ 110/ 1236] Overall Loss 0.263900 Objective Loss 0.263900 LR 0.001000 Time 0.028476 +2023-10-02 21:12:35,695 - Epoch: [91][ 120/ 1236] Overall Loss 0.260807 Objective Loss 0.260807 LR 0.001000 Time 0.027850 +2023-10-02 21:12:35,911 - Epoch: [91][ 130/ 1236] Overall Loss 0.260952 Objective Loss 0.260952 LR 0.001000 Time 0.027365 +2023-10-02 21:12:36,124 - Epoch: [91][ 140/ 1236] Overall Loss 0.260945 Objective Loss 0.260945 LR 0.001000 Time 0.026926 +2023-10-02 21:12:36,336 - Epoch: [91][ 150/ 1236] Overall Loss 0.260190 Objective Loss 0.260190 LR 0.001000 Time 0.026533 +2023-10-02 21:12:36,543 - Epoch: [91][ 160/ 1236] Overall Loss 0.260603 Objective Loss 0.260603 LR 0.001000 Time 0.026164 +2023-10-02 21:12:36,752 - Epoch: [91][ 170/ 1236] Overall Loss 0.260713 Objective Loss 0.260713 LR 0.001000 Time 0.025851 +2023-10-02 21:12:36,961 - Epoch: [91][ 180/ 1236] Overall Loss 0.262686 Objective Loss 0.262686 LR 0.001000 Time 0.025577 +2023-10-02 21:12:37,168 - Epoch: [91][ 190/ 1236] Overall Loss 0.262324 Objective Loss 0.262324 LR 0.001000 Time 0.025320 +2023-10-02 21:12:37,376 - Epoch: [91][ 200/ 1236] Overall Loss 0.261217 Objective Loss 0.261217 LR 0.001000 Time 0.025092 +2023-10-02 21:12:37,583 - Epoch: [91][ 210/ 1236] Overall Loss 0.261830 Objective Loss 0.261830 LR 0.001000 Time 0.024880 +2023-10-02 21:12:37,790 - Epoch: [91][ 220/ 1236] Overall Loss 0.261323 Objective Loss 0.261323 LR 0.001000 Time 0.024691 +2023-10-02 21:12:37,997 - Epoch: [91][ 230/ 1236] Overall Loss 0.261250 Objective Loss 0.261250 LR 0.001000 Time 0.024515 +2023-10-02 21:12:38,205 - Epoch: [91][ 240/ 1236] Overall Loss 0.261165 Objective Loss 0.261165 LR 0.001000 Time 0.024358 +2023-10-02 21:12:38,412 - Epoch: [91][ 250/ 1236] Overall Loss 0.261827 Objective Loss 0.261827 LR 0.001000 Time 0.024210 +2023-10-02 21:12:38,620 - Epoch: [91][ 260/ 1236] Overall Loss 0.262225 Objective Loss 0.262225 LR 0.001000 Time 0.024081 +2023-10-02 21:12:38,829 - Epoch: [91][ 270/ 1236] Overall Loss 0.262184 Objective Loss 0.262184 LR 0.001000 Time 0.023962 +2023-10-02 21:12:39,039 - Epoch: [91][ 280/ 1236] Overall Loss 0.261403 Objective Loss 0.261403 LR 0.001000 Time 0.023853 +2023-10-02 21:12:39,247 - Epoch: [91][ 290/ 1236] Overall Loss 0.261263 Objective Loss 0.261263 LR 0.001000 Time 0.023747 +2023-10-02 21:12:39,455 - Epoch: [91][ 300/ 1236] Overall Loss 0.261784 Objective Loss 0.261784 LR 0.001000 Time 0.023648 +2023-10-02 21:12:39,661 - Epoch: [91][ 310/ 1236] Overall Loss 0.260994 Objective Loss 0.260994 LR 0.001000 Time 0.023550 +2023-10-02 21:12:39,872 - Epoch: [91][ 320/ 1236] Overall Loss 0.260919 Objective Loss 0.260919 LR 0.001000 Time 0.023470 +2023-10-02 21:12:40,084 - Epoch: [91][ 330/ 1236] Overall Loss 0.260937 Objective Loss 0.260937 LR 0.001000 Time 0.023401 +2023-10-02 21:12:40,295 - Epoch: [91][ 340/ 1236] Overall Loss 0.260501 Objective Loss 0.260501 LR 0.001000 Time 0.023331 +2023-10-02 21:12:40,505 - Epoch: [91][ 350/ 1236] Overall Loss 0.260101 Objective Loss 0.260101 LR 0.001000 Time 0.023264 +2023-10-02 21:12:40,716 - Epoch: [91][ 360/ 1236] Overall Loss 0.259098 Objective Loss 0.259098 LR 0.001000 Time 0.023201 +2023-10-02 21:12:40,925 - Epoch: [91][ 370/ 1236] Overall Loss 0.259228 Objective Loss 0.259228 LR 0.001000 Time 0.023140 +2023-10-02 21:12:41,136 - Epoch: [91][ 380/ 1236] Overall Loss 0.259352 Objective Loss 0.259352 LR 0.001000 Time 0.023084 +2023-10-02 21:12:41,346 - Epoch: [91][ 390/ 1236] Overall Loss 0.259390 Objective Loss 0.259390 LR 0.001000 Time 0.023030 +2023-10-02 21:12:41,557 - Epoch: [91][ 400/ 1236] Overall Loss 0.259698 Objective Loss 0.259698 LR 0.001000 Time 0.022980 +2023-10-02 21:12:41,767 - Epoch: [91][ 410/ 1236] Overall Loss 0.260024 Objective Loss 0.260024 LR 0.001000 Time 0.022931 +2023-10-02 21:12:41,977 - Epoch: [91][ 420/ 1236] Overall Loss 0.260795 Objective Loss 0.260795 LR 0.001000 Time 0.022885 +2023-10-02 21:12:42,187 - Epoch: [91][ 430/ 1236] Overall Loss 0.260956 Objective Loss 0.260956 LR 0.001000 Time 0.022839 +2023-10-02 21:12:42,397 - Epoch: [91][ 440/ 1236] Overall Loss 0.260952 Objective Loss 0.260952 LR 0.001000 Time 0.022798 +2023-10-02 21:12:42,607 - Epoch: [91][ 450/ 1236] Overall Loss 0.260960 Objective Loss 0.260960 LR 0.001000 Time 0.022757 +2023-10-02 21:12:42,818 - Epoch: [91][ 460/ 1236] Overall Loss 0.260280 Objective Loss 0.260280 LR 0.001000 Time 0.022719 +2023-10-02 21:12:43,028 - Epoch: [91][ 470/ 1236] Overall Loss 0.260212 Objective Loss 0.260212 LR 0.001000 Time 0.022682 +2023-10-02 21:12:43,238 - Epoch: [91][ 480/ 1236] Overall Loss 0.260626 Objective Loss 0.260626 LR 0.001000 Time 0.022647 +2023-10-02 21:12:43,448 - Epoch: [91][ 490/ 1236] Overall Loss 0.260637 Objective Loss 0.260637 LR 0.001000 Time 0.022612 +2023-10-02 21:12:43,658 - Epoch: [91][ 500/ 1236] Overall Loss 0.260714 Objective Loss 0.260714 LR 0.001000 Time 0.022579 +2023-10-02 21:12:43,868 - Epoch: [91][ 510/ 1236] Overall Loss 0.260565 Objective Loss 0.260565 LR 0.001000 Time 0.022547 +2023-10-02 21:12:44,078 - Epoch: [91][ 520/ 1236] Overall Loss 0.260509 Objective Loss 0.260509 LR 0.001000 Time 0.022517 +2023-10-02 21:12:44,288 - Epoch: [91][ 530/ 1236] Overall Loss 0.260353 Objective Loss 0.260353 LR 0.001000 Time 0.022487 +2023-10-02 21:12:44,498 - Epoch: [91][ 540/ 1236] Overall Loss 0.259662 Objective Loss 0.259662 LR 0.001000 Time 0.022460 +2023-10-02 21:12:44,708 - Epoch: [91][ 550/ 1236] Overall Loss 0.260610 Objective Loss 0.260610 LR 0.001000 Time 0.022432 +2023-10-02 21:12:44,918 - Epoch: [91][ 560/ 1236] Overall Loss 0.261414 Objective Loss 0.261414 LR 0.001000 Time 0.022406 +2023-10-02 21:12:45,129 - Epoch: [91][ 570/ 1236] Overall Loss 0.261375 Objective Loss 0.261375 LR 0.001000 Time 0.022382 +2023-10-02 21:12:45,339 - Epoch: [91][ 580/ 1236] Overall Loss 0.261419 Objective Loss 0.261419 LR 0.001000 Time 0.022358 +2023-10-02 21:12:45,549 - Epoch: [91][ 590/ 1236] Overall Loss 0.261776 Objective Loss 0.261776 LR 0.001000 Time 0.022334 +2023-10-02 21:12:45,759 - Epoch: [91][ 600/ 1236] Overall Loss 0.261956 Objective Loss 0.261956 LR 0.001000 Time 0.022311 +2023-10-02 21:12:45,969 - Epoch: [91][ 610/ 1236] Overall Loss 0.262055 Objective Loss 0.262055 LR 0.001000 Time 0.022289 +2023-10-02 21:12:46,179 - Epoch: [91][ 620/ 1236] Overall Loss 0.261806 Objective Loss 0.261806 LR 0.001000 Time 0.022268 +2023-10-02 21:12:46,389 - Epoch: [91][ 630/ 1236] Overall Loss 0.262585 Objective Loss 0.262585 LR 0.001000 Time 0.022247 +2023-10-02 21:12:46,599 - Epoch: [91][ 640/ 1236] Overall Loss 0.262647 Objective Loss 0.262647 LR 0.001000 Time 0.022227 +2023-10-02 21:12:46,809 - Epoch: [91][ 650/ 1236] Overall Loss 0.262841 Objective Loss 0.262841 LR 0.001000 Time 0.022208 +2023-10-02 21:12:47,019 - Epoch: [91][ 660/ 1236] Overall Loss 0.262634 Objective Loss 0.262634 LR 0.001000 Time 0.022189 +2023-10-02 21:12:47,229 - Epoch: [91][ 670/ 1236] Overall Loss 0.262839 Objective Loss 0.262839 LR 0.001000 Time 0.022171 +2023-10-02 21:12:47,439 - Epoch: [91][ 680/ 1236] Overall Loss 0.262744 Objective Loss 0.262744 LR 0.001000 Time 0.022153 +2023-10-02 21:12:47,650 - Epoch: [91][ 690/ 1236] Overall Loss 0.263133 Objective Loss 0.263133 LR 0.001000 Time 0.022137 +2023-10-02 21:12:47,860 - Epoch: [91][ 700/ 1236] Overall Loss 0.263181 Objective Loss 0.263181 LR 0.001000 Time 0.022120 +2023-10-02 21:12:48,070 - Epoch: [91][ 710/ 1236] Overall Loss 0.263450 Objective Loss 0.263450 LR 0.001000 Time 0.022104 +2023-10-02 21:12:48,280 - Epoch: [91][ 720/ 1236] Overall Loss 0.263668 Objective Loss 0.263668 LR 0.001000 Time 0.022089 +2023-10-02 21:12:48,491 - Epoch: [91][ 730/ 1236] Overall Loss 0.263715 Objective Loss 0.263715 LR 0.001000 Time 0.022074 +2023-10-02 21:12:48,701 - Epoch: [91][ 740/ 1236] Overall Loss 0.263622 Objective Loss 0.263622 LR 0.001000 Time 0.022059 +2023-10-02 21:12:48,911 - Epoch: [91][ 750/ 1236] Overall Loss 0.263381 Objective Loss 0.263381 LR 0.001000 Time 0.022045 +2023-10-02 21:12:49,121 - Epoch: [91][ 760/ 1236] Overall Loss 0.263853 Objective Loss 0.263853 LR 0.001000 Time 0.022031 +2023-10-02 21:12:49,332 - Epoch: [91][ 770/ 1236] Overall Loss 0.263811 Objective Loss 0.263811 LR 0.001000 Time 0.022017 +2023-10-02 21:12:49,542 - Epoch: [91][ 780/ 1236] Overall Loss 0.263930 Objective Loss 0.263930 LR 0.001000 Time 0.022004 +2023-10-02 21:12:49,752 - Epoch: [91][ 790/ 1236] Overall Loss 0.263979 Objective Loss 0.263979 LR 0.001000 Time 0.021991 +2023-10-02 21:12:49,962 - Epoch: [91][ 800/ 1236] Overall Loss 0.264020 Objective Loss 0.264020 LR 0.001000 Time 0.021978 +2023-10-02 21:12:50,172 - Epoch: [91][ 810/ 1236] Overall Loss 0.263921 Objective Loss 0.263921 LR 0.001000 Time 0.021966 +2023-10-02 21:12:50,382 - Epoch: [91][ 820/ 1236] Overall Loss 0.263973 Objective Loss 0.263973 LR 0.001000 Time 0.021954 +2023-10-02 21:12:50,593 - Epoch: [91][ 830/ 1236] Overall Loss 0.264118 Objective Loss 0.264118 LR 0.001000 Time 0.021943 +2023-10-02 21:12:50,803 - Epoch: [91][ 840/ 1236] Overall Loss 0.264408 Objective Loss 0.264408 LR 0.001000 Time 0.021931 +2023-10-02 21:12:51,013 - Epoch: [91][ 850/ 1236] Overall Loss 0.264892 Objective Loss 0.264892 LR 0.001000 Time 0.021920 +2023-10-02 21:12:51,223 - Epoch: [91][ 860/ 1236] Overall Loss 0.264713 Objective Loss 0.264713 LR 0.001000 Time 0.021909 +2023-10-02 21:12:51,433 - Epoch: [91][ 870/ 1236] Overall Loss 0.264711 Objective Loss 0.264711 LR 0.001000 Time 0.021898 +2023-10-02 21:12:51,643 - Epoch: [91][ 880/ 1236] Overall Loss 0.264464 Objective Loss 0.264464 LR 0.001000 Time 0.021887 +2023-10-02 21:12:51,853 - Epoch: [91][ 890/ 1236] Overall Loss 0.264555 Objective Loss 0.264555 LR 0.001000 Time 0.021877 +2023-10-02 21:12:52,063 - Epoch: [91][ 900/ 1236] Overall Loss 0.265001 Objective Loss 0.265001 LR 0.001000 Time 0.021867 +2023-10-02 21:12:52,273 - Epoch: [91][ 910/ 1236] Overall Loss 0.264794 Objective Loss 0.264794 LR 0.001000 Time 0.021857 +2023-10-02 21:12:52,484 - Epoch: [91][ 920/ 1236] Overall Loss 0.264640 Objective Loss 0.264640 LR 0.001000 Time 0.021848 +2023-10-02 21:12:52,694 - Epoch: [91][ 930/ 1236] Overall Loss 0.264982 Objective Loss 0.264982 LR 0.001000 Time 0.021839 +2023-10-02 21:12:52,904 - Epoch: [91][ 940/ 1236] Overall Loss 0.264792 Objective Loss 0.264792 LR 0.001000 Time 0.021829 +2023-10-02 21:12:53,114 - Epoch: [91][ 950/ 1236] Overall Loss 0.264519 Objective Loss 0.264519 LR 0.001000 Time 0.021821 +2023-10-02 21:12:53,325 - Epoch: [91][ 960/ 1236] Overall Loss 0.264337 Objective Loss 0.264337 LR 0.001000 Time 0.021813 +2023-10-02 21:12:53,536 - Epoch: [91][ 970/ 1236] Overall Loss 0.264521 Objective Loss 0.264521 LR 0.001000 Time 0.021804 +2023-10-02 21:12:53,746 - Epoch: [91][ 980/ 1236] Overall Loss 0.264934 Objective Loss 0.264934 LR 0.001000 Time 0.021796 +2023-10-02 21:12:53,956 - Epoch: [91][ 990/ 1236] Overall Loss 0.265200 Objective Loss 0.265200 LR 0.001000 Time 0.021788 +2023-10-02 21:12:54,166 - Epoch: [91][ 1000/ 1236] Overall Loss 0.265100 Objective Loss 0.265100 LR 0.001000 Time 0.021780 +2023-10-02 21:12:54,376 - Epoch: [91][ 1010/ 1236] Overall Loss 0.265103 Objective Loss 0.265103 LR 0.001000 Time 0.021772 +2023-10-02 21:12:54,587 - Epoch: [91][ 1020/ 1236] Overall Loss 0.264904 Objective Loss 0.264904 LR 0.001000 Time 0.021765 +2023-10-02 21:12:54,798 - Epoch: [91][ 1030/ 1236] Overall Loss 0.265288 Objective Loss 0.265288 LR 0.001000 Time 0.021758 +2023-10-02 21:12:55,009 - Epoch: [91][ 1040/ 1236] Overall Loss 0.265508 Objective Loss 0.265508 LR 0.001000 Time 0.021751 +2023-10-02 21:12:55,220 - Epoch: [91][ 1050/ 1236] Overall Loss 0.265770 Objective Loss 0.265770 LR 0.001000 Time 0.021744 +2023-10-02 21:12:55,431 - Epoch: [91][ 1060/ 1236] Overall Loss 0.265844 Objective Loss 0.265844 LR 0.001000 Time 0.021738 +2023-10-02 21:12:55,641 - Epoch: [91][ 1070/ 1236] Overall Loss 0.265544 Objective Loss 0.265544 LR 0.001000 Time 0.021731 +2023-10-02 21:12:55,852 - Epoch: [91][ 1080/ 1236] Overall Loss 0.266246 Objective Loss 0.266246 LR 0.001000 Time 0.021725 +2023-10-02 21:12:56,063 - Epoch: [91][ 1090/ 1236] Overall Loss 0.266753 Objective Loss 0.266753 LR 0.001000 Time 0.021718 +2023-10-02 21:12:56,274 - Epoch: [91][ 1100/ 1236] Overall Loss 0.266784 Objective Loss 0.266784 LR 0.001000 Time 0.021713 +2023-10-02 21:12:56,485 - Epoch: [91][ 1110/ 1236] Overall Loss 0.266703 Objective Loss 0.266703 LR 0.001000 Time 0.021707 +2023-10-02 21:12:56,696 - Epoch: [91][ 1120/ 1236] Overall Loss 0.266833 Objective Loss 0.266833 LR 0.001000 Time 0.021701 +2023-10-02 21:12:56,906 - Epoch: [91][ 1130/ 1236] Overall Loss 0.266953 Objective Loss 0.266953 LR 0.001000 Time 0.021695 +2023-10-02 21:12:57,117 - Epoch: [91][ 1140/ 1236] Overall Loss 0.267115 Objective Loss 0.267115 LR 0.001000 Time 0.021689 +2023-10-02 21:12:57,328 - Epoch: [91][ 1150/ 1236] Overall Loss 0.267222 Objective Loss 0.267222 LR 0.001000 Time 0.021684 +2023-10-02 21:12:57,539 - Epoch: [91][ 1160/ 1236] Overall Loss 0.267459 Objective Loss 0.267459 LR 0.001000 Time 0.021678 +2023-10-02 21:12:57,750 - Epoch: [91][ 1170/ 1236] Overall Loss 0.267553 Objective Loss 0.267553 LR 0.001000 Time 0.021673 +2023-10-02 21:12:57,961 - Epoch: [91][ 1180/ 1236] Overall Loss 0.267765 Objective Loss 0.267765 LR 0.001000 Time 0.021668 +2023-10-02 21:12:58,172 - Epoch: [91][ 1190/ 1236] Overall Loss 0.267669 Objective Loss 0.267669 LR 0.001000 Time 0.021663 +2023-10-02 21:12:58,383 - Epoch: [91][ 1200/ 1236] Overall Loss 0.268182 Objective Loss 0.268182 LR 0.001000 Time 0.021658 +2023-10-02 21:12:58,593 - Epoch: [91][ 1210/ 1236] Overall Loss 0.268386 Objective Loss 0.268386 LR 0.001000 Time 0.021652 +2023-10-02 21:12:58,804 - Epoch: [91][ 1220/ 1236] Overall Loss 0.268642 Objective Loss 0.268642 LR 0.001000 Time 0.021647 +2023-10-02 21:12:59,068 - Epoch: [91][ 1230/ 1236] Overall Loss 0.269108 Objective Loss 0.269108 LR 0.001000 Time 0.021685 +2023-10-02 21:12:59,190 - Epoch: [91][ 1236/ 1236] Overall Loss 0.269181 Objective Loss 0.269181 Top1 85.539715 Top5 98.370672 LR 0.001000 Time 0.021678 +2023-10-02 21:12:59,320 - --- validate (epoch=91)----------- +2023-10-02 21:12:59,320 - 29943 samples (256 per mini-batch) +2023-10-02 21:12:59,811 - Epoch: [91][ 10/ 117] Loss 0.327789 Top1 82.421875 Top5 98.164062 +2023-10-02 21:12:59,972 - Epoch: [91][ 20/ 117] Loss 0.362253 Top1 82.265625 Top5 98.066406 +2023-10-02 21:13:00,131 - Epoch: [91][ 30/ 117] Loss 0.361513 Top1 82.513021 Top5 98.072917 +2023-10-02 21:13:00,291 - Epoch: [91][ 40/ 117] Loss 0.364560 Top1 82.705078 Top5 98.066406 +2023-10-02 21:13:00,448 - Epoch: [91][ 50/ 117] Loss 0.355043 Top1 82.898438 Top5 98.179688 +2023-10-02 21:13:00,608 - Epoch: [91][ 60/ 117] Loss 0.355822 Top1 82.903646 Top5 98.125000 +2023-10-02 21:13:00,763 - Epoch: [91][ 70/ 117] Loss 0.349638 Top1 83.074777 Top5 98.175223 +2023-10-02 21:13:00,923 - Epoch: [91][ 80/ 117] Loss 0.347952 Top1 83.149414 Top5 98.100586 +2023-10-02 21:13:01,079 - Epoch: [91][ 90/ 117] Loss 0.349995 Top1 83.129340 Top5 98.059896 +2023-10-02 21:13:01,238 - Epoch: [91][ 100/ 117] Loss 0.349265 Top1 83.187500 Top5 98.070312 +2023-10-02 21:13:01,403 - Epoch: [91][ 110/ 117] Loss 0.348258 Top1 83.117898 Top5 98.057528 +2023-10-02 21:13:01,492 - Epoch: [91][ 117/ 117] Loss 0.347967 Top1 83.161340 Top5 98.056307 +2023-10-02 21:13:01,641 - ==> Top1: 83.161 Top5: 98.056 Loss: 0.348 + +2023-10-02 21:13:01,642 - ==> Confusion: +[[ 922 2 10 0 7 2 0 0 7 56 2 0 1 2 13 3 7 4 1 0 11] + [ 1 1033 0 0 4 32 5 15 0 1 1 1 1 0 2 4 4 0 21 2 4] + [ 5 1 953 10 0 0 32 6 0 1 2 0 11 0 1 5 4 2 13 1 9] + [ 1 5 21 962 1 3 4 3 2 0 3 0 8 5 32 2 2 6 15 0 14] + [ 30 11 4 1 942 7 0 1 1 15 1 0 2 1 6 7 12 0 1 2 6] + [ 1 23 0 4 1 988 2 20 1 4 1 11 8 13 7 2 3 1 5 6 15] + [ 0 0 24 0 0 1 1131 5 0 1 4 1 0 0 0 6 0 1 2 7 8] + [ 5 13 28 0 5 33 8 1022 3 3 3 5 7 5 4 0 1 1 49 11 12] + [ 17 3 4 2 2 5 0 0 937 38 14 1 5 15 32 1 2 3 6 0 2] + [ 124 0 3 3 7 2 2 0 30 898 2 2 1 23 9 1 0 1 0 2 9] + [ 2 1 10 15 1 3 11 1 11 1 954 3 1 10 9 1 2 3 5 1 8] + [ 0 0 4 0 0 9 0 3 0 1 0 929 46 10 0 8 2 11 0 8 4] + [ 0 1 2 1 0 0 3 1 1 0 1 36 975 3 4 9 3 12 1 6 9] + [ 3 0 3 0 3 13 0 0 14 12 8 3 3 1037 9 0 1 1 0 1 8] + [ 11 3 5 17 6 0 0 0 9 1 1 0 6 2 1018 1 3 3 9 0 6] + [ 0 0 2 0 5 0 0 0 0 0 0 6 9 0 0 1076 10 12 3 6 5] + [ 3 6 0 0 3 10 0 1 0 0 0 5 3 3 5 17 1092 1 1 5 6] + [ 0 0 1 3 0 0 3 0 1 1 0 4 16 0 1 5 2 1000 0 0 1] + [ 2 4 4 16 1 0 1 8 2 0 3 2 4 1 13 0 2 0 994 2 9] + [ 0 1 6 0 1 2 19 5 0 0 0 9 8 1 0 5 12 1 2 1071 9] + [ 126 156 195 115 70 175 62 82 79 90 140 126 361 282 205 76 132 77 210 179 4967]] + +2023-10-02 21:13:01,643 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:13:01,643 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:13:01,649 - + +2023-10-02 21:13:01,649 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:13:02,772 - Epoch: [92][ 10/ 1236] Overall Loss 0.262019 Objective Loss 0.262019 LR 0.001000 Time 0.112236 +2023-10-02 21:13:02,981 - Epoch: [92][ 20/ 1236] Overall Loss 0.263593 Objective Loss 0.263593 LR 0.001000 Time 0.066549 +2023-10-02 21:13:03,188 - Epoch: [92][ 30/ 1236] Overall Loss 0.268016 Objective Loss 0.268016 LR 0.001000 Time 0.051231 +2023-10-02 21:13:03,398 - Epoch: [92][ 40/ 1236] Overall Loss 0.258856 Objective Loss 0.258856 LR 0.001000 Time 0.043661 +2023-10-02 21:13:03,604 - Epoch: [92][ 50/ 1236] Overall Loss 0.258361 Objective Loss 0.258361 LR 0.001000 Time 0.039047 +2023-10-02 21:13:03,814 - Epoch: [92][ 60/ 1236] Overall Loss 0.254051 Objective Loss 0.254051 LR 0.001000 Time 0.036032 +2023-10-02 21:13:04,020 - Epoch: [92][ 70/ 1236] Overall Loss 0.254723 Objective Loss 0.254723 LR 0.001000 Time 0.033825 +2023-10-02 21:13:04,230 - Epoch: [92][ 80/ 1236] Overall Loss 0.254967 Objective Loss 0.254967 LR 0.001000 Time 0.032219 +2023-10-02 21:13:04,435 - Epoch: [92][ 90/ 1236] Overall Loss 0.256907 Objective Loss 0.256907 LR 0.001000 Time 0.030914 +2023-10-02 21:13:04,646 - Epoch: [92][ 100/ 1236] Overall Loss 0.257856 Objective Loss 0.257856 LR 0.001000 Time 0.029926 +2023-10-02 21:13:04,853 - Epoch: [92][ 110/ 1236] Overall Loss 0.262425 Objective Loss 0.262425 LR 0.001000 Time 0.029090 +2023-10-02 21:13:05,063 - Epoch: [92][ 120/ 1236] Overall Loss 0.262561 Objective Loss 0.262561 LR 0.001000 Time 0.028409 +2023-10-02 21:13:05,269 - Epoch: [92][ 130/ 1236] Overall Loss 0.263081 Objective Loss 0.263081 LR 0.001000 Time 0.027811 +2023-10-02 21:13:05,481 - Epoch: [92][ 140/ 1236] Overall Loss 0.259954 Objective Loss 0.259954 LR 0.001000 Time 0.027332 +2023-10-02 21:13:05,687 - Epoch: [92][ 150/ 1236] Overall Loss 0.260013 Objective Loss 0.260013 LR 0.001000 Time 0.026885 +2023-10-02 21:13:05,899 - Epoch: [92][ 160/ 1236] Overall Loss 0.259605 Objective Loss 0.259605 LR 0.001000 Time 0.026525 +2023-10-02 21:13:06,106 - Epoch: [92][ 170/ 1236] Overall Loss 0.261090 Objective Loss 0.261090 LR 0.001000 Time 0.026184 +2023-10-02 21:13:06,317 - Epoch: [92][ 180/ 1236] Overall Loss 0.261293 Objective Loss 0.261293 LR 0.001000 Time 0.025895 +2023-10-02 21:13:06,530 - Epoch: [92][ 190/ 1236] Overall Loss 0.261970 Objective Loss 0.261970 LR 0.001000 Time 0.025654 +2023-10-02 21:13:06,740 - Epoch: [92][ 200/ 1236] Overall Loss 0.262702 Objective Loss 0.262702 LR 0.001000 Time 0.025420 +2023-10-02 21:13:06,954 - Epoch: [92][ 210/ 1236] Overall Loss 0.263993 Objective Loss 0.263993 LR 0.001000 Time 0.025225 +2023-10-02 21:13:07,163 - Epoch: [92][ 220/ 1236] Overall Loss 0.262507 Objective Loss 0.262507 LR 0.001000 Time 0.025030 +2023-10-02 21:13:07,377 - Epoch: [92][ 230/ 1236] Overall Loss 0.262304 Objective Loss 0.262304 LR 0.001000 Time 0.024868 +2023-10-02 21:13:07,586 - Epoch: [92][ 240/ 1236] Overall Loss 0.262819 Objective Loss 0.262819 LR 0.001000 Time 0.024705 +2023-10-02 21:13:07,800 - Epoch: [92][ 250/ 1236] Overall Loss 0.262458 Objective Loss 0.262458 LR 0.001000 Time 0.024569 +2023-10-02 21:13:08,009 - Epoch: [92][ 260/ 1236] Overall Loss 0.261900 Objective Loss 0.261900 LR 0.001000 Time 0.024429 +2023-10-02 21:13:08,223 - Epoch: [92][ 270/ 1236] Overall Loss 0.261266 Objective Loss 0.261266 LR 0.001000 Time 0.024313 +2023-10-02 21:13:08,432 - Epoch: [92][ 280/ 1236] Overall Loss 0.261239 Objective Loss 0.261239 LR 0.001000 Time 0.024193 +2023-10-02 21:13:08,646 - Epoch: [92][ 290/ 1236] Overall Loss 0.261774 Objective Loss 0.261774 LR 0.001000 Time 0.024093 +2023-10-02 21:13:08,856 - Epoch: [92][ 300/ 1236] Overall Loss 0.261878 Objective Loss 0.261878 LR 0.001000 Time 0.023990 +2023-10-02 21:13:09,069 - Epoch: [92][ 310/ 1236] Overall Loss 0.262821 Objective Loss 0.262821 LR 0.001000 Time 0.023903 +2023-10-02 21:13:09,280 - Epoch: [92][ 320/ 1236] Overall Loss 0.263397 Objective Loss 0.263397 LR 0.001000 Time 0.023814 +2023-10-02 21:13:09,494 - Epoch: [92][ 330/ 1236] Overall Loss 0.262862 Objective Loss 0.262862 LR 0.001000 Time 0.023735 +2023-10-02 21:13:09,708 - Epoch: [92][ 340/ 1236] Overall Loss 0.262588 Objective Loss 0.262588 LR 0.001000 Time 0.023661 +2023-10-02 21:13:09,921 - Epoch: [92][ 350/ 1236] Overall Loss 0.263611 Objective Loss 0.263611 LR 0.001000 Time 0.023591 +2023-10-02 21:13:10,133 - Epoch: [92][ 360/ 1236] Overall Loss 0.263426 Objective Loss 0.263426 LR 0.001000 Time 0.023522 +2023-10-02 21:13:10,345 - Epoch: [92][ 370/ 1236] Overall Loss 0.263658 Objective Loss 0.263658 LR 0.001000 Time 0.023458 +2023-10-02 21:13:10,558 - Epoch: [92][ 380/ 1236] Overall Loss 0.264458 Objective Loss 0.264458 LR 0.001000 Time 0.023400 +2023-10-02 21:13:10,770 - Epoch: [92][ 390/ 1236] Overall Loss 0.263906 Objective Loss 0.263906 LR 0.001000 Time 0.023342 +2023-10-02 21:13:10,981 - Epoch: [92][ 400/ 1236] Overall Loss 0.263838 Objective Loss 0.263838 LR 0.001000 Time 0.023286 +2023-10-02 21:13:11,193 - Epoch: [92][ 410/ 1236] Overall Loss 0.264041 Objective Loss 0.264041 LR 0.001000 Time 0.023234 +2023-10-02 21:13:11,405 - Epoch: [92][ 420/ 1236] Overall Loss 0.263885 Objective Loss 0.263885 LR 0.001000 Time 0.023180 +2023-10-02 21:13:11,617 - Epoch: [92][ 430/ 1236] Overall Loss 0.263824 Objective Loss 0.263824 LR 0.001000 Time 0.023134 +2023-10-02 21:13:11,829 - Epoch: [92][ 440/ 1236] Overall Loss 0.263679 Objective Loss 0.263679 LR 0.001000 Time 0.023086 +2023-10-02 21:13:12,041 - Epoch: [92][ 450/ 1236] Overall Loss 0.263281 Objective Loss 0.263281 LR 0.001000 Time 0.023044 +2023-10-02 21:13:12,253 - Epoch: [92][ 460/ 1236] Overall Loss 0.262507 Objective Loss 0.262507 LR 0.001000 Time 0.023004 +2023-10-02 21:13:12,466 - Epoch: [92][ 470/ 1236] Overall Loss 0.262381 Objective Loss 0.262381 LR 0.001000 Time 0.022966 +2023-10-02 21:13:12,678 - Epoch: [92][ 480/ 1236] Overall Loss 0.262593 Objective Loss 0.262593 LR 0.001000 Time 0.022929 +2023-10-02 21:13:12,890 - Epoch: [92][ 490/ 1236] Overall Loss 0.262942 Objective Loss 0.262942 LR 0.001000 Time 0.022890 +2023-10-02 21:13:13,102 - Epoch: [92][ 500/ 1236] Overall Loss 0.262618 Objective Loss 0.262618 LR 0.001000 Time 0.022851 +2023-10-02 21:13:13,314 - Epoch: [92][ 510/ 1236] Overall Loss 0.262775 Objective Loss 0.262775 LR 0.001000 Time 0.022818 +2023-10-02 21:13:13,525 - Epoch: [92][ 520/ 1236] Overall Loss 0.263522 Objective Loss 0.263522 LR 0.001000 Time 0.022785 +2023-10-02 21:13:13,737 - Epoch: [92][ 530/ 1236] Overall Loss 0.263615 Objective Loss 0.263615 LR 0.001000 Time 0.022752 +2023-10-02 21:13:13,949 - Epoch: [92][ 540/ 1236] Overall Loss 0.264037 Objective Loss 0.264037 LR 0.001000 Time 0.022722 +2023-10-02 21:13:14,161 - Epoch: [92][ 550/ 1236] Overall Loss 0.264381 Objective Loss 0.264381 LR 0.001000 Time 0.022694 +2023-10-02 21:13:14,373 - Epoch: [92][ 560/ 1236] Overall Loss 0.263990 Objective Loss 0.263990 LR 0.001000 Time 0.022667 +2023-10-02 21:13:14,585 - Epoch: [92][ 570/ 1236] Overall Loss 0.263979 Objective Loss 0.263979 LR 0.001000 Time 0.022640 +2023-10-02 21:13:14,798 - Epoch: [92][ 580/ 1236] Overall Loss 0.264311 Objective Loss 0.264311 LR 0.001000 Time 0.022615 +2023-10-02 21:13:15,009 - Epoch: [92][ 590/ 1236] Overall Loss 0.264929 Objective Loss 0.264929 LR 0.001000 Time 0.022588 +2023-10-02 21:13:15,221 - Epoch: [92][ 600/ 1236] Overall Loss 0.264827 Objective Loss 0.264827 LR 0.001000 Time 0.022563 +2023-10-02 21:13:15,433 - Epoch: [92][ 610/ 1236] Overall Loss 0.265033 Objective Loss 0.265033 LR 0.001000 Time 0.022541 +2023-10-02 21:13:15,645 - Epoch: [92][ 620/ 1236] Overall Loss 0.265012 Objective Loss 0.265012 LR 0.001000 Time 0.022519 +2023-10-02 21:13:15,858 - Epoch: [92][ 630/ 1236] Overall Loss 0.265542 Objective Loss 0.265542 LR 0.001000 Time 0.022498 +2023-10-02 21:13:16,069 - Epoch: [92][ 640/ 1236] Overall Loss 0.266195 Objective Loss 0.266195 LR 0.001000 Time 0.022476 +2023-10-02 21:13:16,281 - Epoch: [92][ 650/ 1236] Overall Loss 0.266502 Objective Loss 0.266502 LR 0.001000 Time 0.022455 +2023-10-02 21:13:16,492 - Epoch: [92][ 660/ 1236] Overall Loss 0.266962 Objective Loss 0.266962 LR 0.001000 Time 0.022432 +2023-10-02 21:13:16,703 - Epoch: [92][ 670/ 1236] Overall Loss 0.267333 Objective Loss 0.267333 LR 0.001000 Time 0.022412 +2023-10-02 21:13:16,915 - Epoch: [92][ 680/ 1236] Overall Loss 0.267131 Objective Loss 0.267131 LR 0.001000 Time 0.022393 +2023-10-02 21:13:17,126 - Epoch: [92][ 690/ 1236] Overall Loss 0.267198 Objective Loss 0.267198 LR 0.001000 Time 0.022375 +2023-10-02 21:13:17,338 - Epoch: [92][ 700/ 1236] Overall Loss 0.266878 Objective Loss 0.266878 LR 0.001000 Time 0.022354 +2023-10-02 21:13:17,549 - Epoch: [92][ 710/ 1236] Overall Loss 0.267293 Objective Loss 0.267293 LR 0.001000 Time 0.022337 +2023-10-02 21:13:17,761 - Epoch: [92][ 720/ 1236] Overall Loss 0.267212 Objective Loss 0.267212 LR 0.001000 Time 0.022318 +2023-10-02 21:13:17,972 - Epoch: [92][ 730/ 1236] Overall Loss 0.267445 Objective Loss 0.267445 LR 0.001000 Time 0.022301 +2023-10-02 21:13:18,183 - Epoch: [92][ 740/ 1236] Overall Loss 0.267460 Objective Loss 0.267460 LR 0.001000 Time 0.022283 +2023-10-02 21:13:18,393 - Epoch: [92][ 750/ 1236] Overall Loss 0.267519 Objective Loss 0.267519 LR 0.001000 Time 0.022266 +2023-10-02 21:13:18,604 - Epoch: [92][ 760/ 1236] Overall Loss 0.267304 Objective Loss 0.267304 LR 0.001000 Time 0.022250 +2023-10-02 21:13:18,816 - Epoch: [92][ 770/ 1236] Overall Loss 0.267802 Objective Loss 0.267802 LR 0.001000 Time 0.022235 +2023-10-02 21:13:19,027 - Epoch: [92][ 780/ 1236] Overall Loss 0.267787 Objective Loss 0.267787 LR 0.001000 Time 0.022218 +2023-10-02 21:13:19,238 - Epoch: [92][ 790/ 1236] Overall Loss 0.267785 Objective Loss 0.267785 LR 0.001000 Time 0.022204 +2023-10-02 21:13:19,450 - Epoch: [92][ 800/ 1236] Overall Loss 0.267829 Objective Loss 0.267829 LR 0.001000 Time 0.022188 +2023-10-02 21:13:19,661 - Epoch: [92][ 810/ 1236] Overall Loss 0.267940 Objective Loss 0.267940 LR 0.001000 Time 0.022175 +2023-10-02 21:13:19,872 - Epoch: [92][ 820/ 1236] Overall Loss 0.268018 Objective Loss 0.268018 LR 0.001000 Time 0.022161 +2023-10-02 21:13:20,084 - Epoch: [92][ 830/ 1236] Overall Loss 0.267875 Objective Loss 0.267875 LR 0.001000 Time 0.022146 +2023-10-02 21:13:20,295 - Epoch: [92][ 840/ 1236] Overall Loss 0.267969 Objective Loss 0.267969 LR 0.001000 Time 0.022132 +2023-10-02 21:13:20,507 - Epoch: [92][ 850/ 1236] Overall Loss 0.268182 Objective Loss 0.268182 LR 0.001000 Time 0.022121 +2023-10-02 21:13:20,719 - Epoch: [92][ 860/ 1236] Overall Loss 0.268178 Objective Loss 0.268178 LR 0.001000 Time 0.022109 +2023-10-02 21:13:20,930 - Epoch: [92][ 870/ 1236] Overall Loss 0.268157 Objective Loss 0.268157 LR 0.001000 Time 0.022097 +2023-10-02 21:13:21,142 - Epoch: [92][ 880/ 1236] Overall Loss 0.268449 Objective Loss 0.268449 LR 0.001000 Time 0.022085 +2023-10-02 21:13:21,353 - Epoch: [92][ 890/ 1236] Overall Loss 0.268287 Objective Loss 0.268287 LR 0.001000 Time 0.022074 +2023-10-02 21:13:21,565 - Epoch: [92][ 900/ 1236] Overall Loss 0.268710 Objective Loss 0.268710 LR 0.001000 Time 0.022061 +2023-10-02 21:13:21,777 - Epoch: [92][ 910/ 1236] Overall Loss 0.268405 Objective Loss 0.268405 LR 0.001000 Time 0.022051 +2023-10-02 21:13:21,988 - Epoch: [92][ 920/ 1236] Overall Loss 0.268221 Objective Loss 0.268221 LR 0.001000 Time 0.022039 +2023-10-02 21:13:22,200 - Epoch: [92][ 930/ 1236] Overall Loss 0.268214 Objective Loss 0.268214 LR 0.001000 Time 0.022028 +2023-10-02 21:13:22,411 - Epoch: [92][ 940/ 1236] Overall Loss 0.268273 Objective Loss 0.268273 LR 0.001000 Time 0.022018 +2023-10-02 21:13:22,623 - Epoch: [92][ 950/ 1236] Overall Loss 0.268247 Objective Loss 0.268247 LR 0.001000 Time 0.022008 +2023-10-02 21:13:22,834 - Epoch: [92][ 960/ 1236] Overall Loss 0.268295 Objective Loss 0.268295 LR 0.001000 Time 0.021998 +2023-10-02 21:13:23,046 - Epoch: [92][ 970/ 1236] Overall Loss 0.268159 Objective Loss 0.268159 LR 0.001000 Time 0.021988 +2023-10-02 21:13:23,257 - Epoch: [92][ 980/ 1236] Overall Loss 0.268280 Objective Loss 0.268280 LR 0.001000 Time 0.021979 +2023-10-02 21:13:23,469 - Epoch: [92][ 990/ 1236] Overall Loss 0.268312 Objective Loss 0.268312 LR 0.001000 Time 0.021969 +2023-10-02 21:13:23,680 - Epoch: [92][ 1000/ 1236] Overall Loss 0.268424 Objective Loss 0.268424 LR 0.001000 Time 0.021959 +2023-10-02 21:13:23,892 - Epoch: [92][ 1010/ 1236] Overall Loss 0.268853 Objective Loss 0.268853 LR 0.001000 Time 0.021949 +2023-10-02 21:13:24,103 - Epoch: [92][ 1020/ 1236] Overall Loss 0.268882 Objective Loss 0.268882 LR 0.001000 Time 0.021941 +2023-10-02 21:13:24,315 - Epoch: [92][ 1030/ 1236] Overall Loss 0.268910 Objective Loss 0.268910 LR 0.001000 Time 0.021933 +2023-10-02 21:13:24,525 - Epoch: [92][ 1040/ 1236] Overall Loss 0.268970 Objective Loss 0.268970 LR 0.001000 Time 0.021923 +2023-10-02 21:13:24,737 - Epoch: [92][ 1050/ 1236] Overall Loss 0.268754 Objective Loss 0.268754 LR 0.001000 Time 0.021915 +2023-10-02 21:13:24,948 - Epoch: [92][ 1060/ 1236] Overall Loss 0.268661 Objective Loss 0.268661 LR 0.001000 Time 0.021907 +2023-10-02 21:13:25,160 - Epoch: [92][ 1070/ 1236] Overall Loss 0.268594 Objective Loss 0.268594 LR 0.001000 Time 0.021900 +2023-10-02 21:13:25,372 - Epoch: [92][ 1080/ 1236] Overall Loss 0.268624 Objective Loss 0.268624 LR 0.001000 Time 0.021893 +2023-10-02 21:13:25,584 - Epoch: [92][ 1090/ 1236] Overall Loss 0.268754 Objective Loss 0.268754 LR 0.001000 Time 0.021885 +2023-10-02 21:13:25,795 - Epoch: [92][ 1100/ 1236] Overall Loss 0.268723 Objective Loss 0.268723 LR 0.001000 Time 0.021877 +2023-10-02 21:13:26,007 - Epoch: [92][ 1110/ 1236] Overall Loss 0.269012 Objective Loss 0.269012 LR 0.001000 Time 0.021869 +2023-10-02 21:13:26,218 - Epoch: [92][ 1120/ 1236] Overall Loss 0.268887 Objective Loss 0.268887 LR 0.001000 Time 0.021862 +2023-10-02 21:13:26,430 - Epoch: [92][ 1130/ 1236] Overall Loss 0.268932 Objective Loss 0.268932 LR 0.001000 Time 0.021856 +2023-10-02 21:13:26,642 - Epoch: [92][ 1140/ 1236] Overall Loss 0.268958 Objective Loss 0.268958 LR 0.001000 Time 0.021849 +2023-10-02 21:13:26,854 - Epoch: [92][ 1150/ 1236] Overall Loss 0.268851 Objective Loss 0.268851 LR 0.001000 Time 0.021842 +2023-10-02 21:13:27,065 - Epoch: [92][ 1160/ 1236] Overall Loss 0.268814 Objective Loss 0.268814 LR 0.001000 Time 0.021836 +2023-10-02 21:13:27,277 - Epoch: [92][ 1170/ 1236] Overall Loss 0.268821 Objective Loss 0.268821 LR 0.001000 Time 0.021829 +2023-10-02 21:13:27,489 - Epoch: [92][ 1180/ 1236] Overall Loss 0.268705 Objective Loss 0.268705 LR 0.001000 Time 0.021823 +2023-10-02 21:13:27,701 - Epoch: [92][ 1190/ 1236] Overall Loss 0.268899 Objective Loss 0.268899 LR 0.001000 Time 0.021816 +2023-10-02 21:13:27,912 - Epoch: [92][ 1200/ 1236] Overall Loss 0.268839 Objective Loss 0.268839 LR 0.001000 Time 0.021810 +2023-10-02 21:13:28,124 - Epoch: [92][ 1210/ 1236] Overall Loss 0.268910 Objective Loss 0.268910 LR 0.001000 Time 0.021804 +2023-10-02 21:13:28,335 - Epoch: [92][ 1220/ 1236] Overall Loss 0.268964 Objective Loss 0.268964 LR 0.001000 Time 0.021798 +2023-10-02 21:13:28,600 - Epoch: [92][ 1230/ 1236] Overall Loss 0.268993 Objective Loss 0.268993 LR 0.001000 Time 0.021835 +2023-10-02 21:13:28,722 - Epoch: [92][ 1236/ 1236] Overall Loss 0.268802 Objective Loss 0.268802 Top1 88.187373 Top5 98.778004 LR 0.001000 Time 0.021828 +2023-10-02 21:13:28,862 - --- validate (epoch=92)----------- +2023-10-02 21:13:28,862 - 29943 samples (256 per mini-batch) +2023-10-02 21:13:29,353 - Epoch: [92][ 10/ 117] Loss 0.358502 Top1 84.023438 Top5 98.007812 +2023-10-02 21:13:29,500 - Epoch: [92][ 20/ 117] Loss 0.362106 Top1 83.554688 Top5 98.046875 +2023-10-02 21:13:29,647 - Epoch: [92][ 30/ 117] Loss 0.345938 Top1 83.710938 Top5 98.138021 +2023-10-02 21:13:29,793 - Epoch: [92][ 40/ 117] Loss 0.334696 Top1 84.023438 Top5 98.125000 +2023-10-02 21:13:29,939 - Epoch: [92][ 50/ 117] Loss 0.341683 Top1 83.960938 Top5 98.125000 +2023-10-02 21:13:30,086 - Epoch: [92][ 60/ 117] Loss 0.345566 Top1 83.906250 Top5 98.131510 +2023-10-02 21:13:30,232 - Epoch: [92][ 70/ 117] Loss 0.350779 Top1 83.755580 Top5 98.097098 +2023-10-02 21:13:30,379 - Epoch: [92][ 80/ 117] Loss 0.347355 Top1 83.759766 Top5 98.115234 +2023-10-02 21:13:30,527 - Epoch: [92][ 90/ 117] Loss 0.346491 Top1 83.789062 Top5 98.142361 +2023-10-02 21:13:30,674 - Epoch: [92][ 100/ 117] Loss 0.345619 Top1 83.878906 Top5 98.125000 +2023-10-02 21:13:30,829 - Epoch: [92][ 110/ 117] Loss 0.345812 Top1 83.881392 Top5 98.114347 +2023-10-02 21:13:30,919 - Epoch: [92][ 117/ 117] Loss 0.346088 Top1 83.939485 Top5 98.089704 +2023-10-02 21:13:31,060 - ==> Top1: 83.939 Top5: 98.090 Loss: 0.346 + +2023-10-02 21:13:31,060 - ==> Confusion: +[[ 945 0 4 0 7 2 0 1 3 51 2 0 3 1 5 2 6 0 0 0 18] + [ 0 1041 1 1 3 31 0 27 2 2 0 2 0 0 0 3 3 0 6 1 8] + [ 5 0 975 14 2 0 17 6 0 3 0 2 8 1 1 4 3 1 6 4 4] + [ 3 1 19 972 2 4 1 4 2 0 4 0 9 2 26 0 3 8 13 1 15] + [ 27 9 2 1 962 9 0 1 0 9 1 0 1 2 5 6 8 1 0 1 5] + [ 3 38 2 2 0 977 1 23 1 7 1 8 4 13 5 1 3 3 4 7 13] + [ 0 4 29 0 0 1 1121 4 0 0 6 2 3 0 0 4 0 0 1 9 7] + [ 4 19 18 0 3 30 9 1042 1 3 5 11 2 4 3 0 2 0 39 14 9] + [ 26 9 0 2 2 3 0 0 917 52 13 0 2 11 34 0 4 2 6 2 4] + [ 154 2 3 3 13 2 0 0 18 884 0 1 0 20 7 2 0 0 0 5 5] + [ 3 5 10 13 1 2 4 6 14 1 937 1 0 14 10 0 4 1 9 7 11] + [ 3 0 0 0 0 9 0 2 0 1 0 930 43 11 0 4 2 18 0 9 3] + [ 2 1 3 2 0 0 0 0 1 0 3 30 963 2 4 11 2 20 4 9 11] + [ 1 0 3 1 4 6 1 0 11 12 4 6 1 1041 5 1 3 1 0 6 12] + [ 8 2 5 21 9 1 0 0 10 1 3 0 4 4 1012 0 1 2 6 0 12] + [ 2 0 1 1 4 2 0 0 0 1 0 7 6 1 0 1062 20 13 0 7 7] + [ 0 16 1 0 8 11 1 0 0 3 0 8 4 0 2 8 1083 0 0 7 9] + [ 0 0 0 3 0 0 2 0 0 1 2 6 26 0 1 3 0 991 0 0 3] + [ 3 9 5 18 0 1 1 19 1 1 3 2 5 0 9 0 1 1 978 1 10] + [ 0 1 5 2 0 6 6 5 0 1 0 15 4 2 0 2 3 2 0 1091 7] + [ 138 194 140 94 96 177 33 91 48 87 121 109 321 287 139 71 149 51 137 212 5210]] + +2023-10-02 21:13:31,062 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:13:31,062 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:13:31,068 - + +2023-10-02 21:13:31,068 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:13:32,107 - Epoch: [93][ 10/ 1236] Overall Loss 0.233854 Objective Loss 0.233854 LR 0.001000 Time 0.103856 +2023-10-02 21:13:32,317 - Epoch: [93][ 20/ 1236] Overall Loss 0.246358 Objective Loss 0.246358 LR 0.001000 Time 0.062384 +2023-10-02 21:13:32,534 - Epoch: [93][ 30/ 1236] Overall Loss 0.246433 Objective Loss 0.246433 LR 0.001000 Time 0.048840 +2023-10-02 21:13:32,756 - Epoch: [93][ 40/ 1236] Overall Loss 0.254126 Objective Loss 0.254126 LR 0.001000 Time 0.042152 +2023-10-02 21:13:32,973 - Epoch: [93][ 50/ 1236] Overall Loss 0.250366 Objective Loss 0.250366 LR 0.001000 Time 0.038051 +2023-10-02 21:13:33,194 - Epoch: [93][ 60/ 1236] Overall Loss 0.250177 Objective Loss 0.250177 LR 0.001000 Time 0.035391 +2023-10-02 21:13:33,403 - Epoch: [93][ 70/ 1236] Overall Loss 0.250701 Objective Loss 0.250701 LR 0.001000 Time 0.033317 +2023-10-02 21:13:33,613 - Epoch: [93][ 80/ 1236] Overall Loss 0.251733 Objective Loss 0.251733 LR 0.001000 Time 0.031771 +2023-10-02 21:13:33,822 - Epoch: [93][ 90/ 1236] Overall Loss 0.251949 Objective Loss 0.251949 LR 0.001000 Time 0.030561 +2023-10-02 21:13:34,032 - Epoch: [93][ 100/ 1236] Overall Loss 0.250499 Objective Loss 0.250499 LR 0.001000 Time 0.029600 +2023-10-02 21:13:34,241 - Epoch: [93][ 110/ 1236] Overall Loss 0.250588 Objective Loss 0.250588 LR 0.001000 Time 0.028798 +2023-10-02 21:13:34,452 - Epoch: [93][ 120/ 1236] Overall Loss 0.251187 Objective Loss 0.251187 LR 0.001000 Time 0.028153 +2023-10-02 21:13:34,660 - Epoch: [93][ 130/ 1236] Overall Loss 0.249759 Objective Loss 0.249759 LR 0.001000 Time 0.027587 +2023-10-02 21:13:34,871 - Epoch: [93][ 140/ 1236] Overall Loss 0.248531 Objective Loss 0.248531 LR 0.001000 Time 0.027120 +2023-10-02 21:13:35,079 - Epoch: [93][ 150/ 1236] Overall Loss 0.250685 Objective Loss 0.250685 LR 0.001000 Time 0.026699 +2023-10-02 21:13:35,289 - Epoch: [93][ 160/ 1236] Overall Loss 0.254812 Objective Loss 0.254812 LR 0.001000 Time 0.026340 +2023-10-02 21:13:35,494 - Epoch: [93][ 170/ 1236] Overall Loss 0.254563 Objective Loss 0.254563 LR 0.001000 Time 0.025995 +2023-10-02 21:13:35,702 - Epoch: [93][ 180/ 1236] Overall Loss 0.255676 Objective Loss 0.255676 LR 0.001000 Time 0.025704 +2023-10-02 21:13:35,908 - Epoch: [93][ 190/ 1236] Overall Loss 0.257264 Objective Loss 0.257264 LR 0.001000 Time 0.025429 +2023-10-02 21:13:36,116 - Epoch: [93][ 200/ 1236] Overall Loss 0.257780 Objective Loss 0.257780 LR 0.001000 Time 0.025195 +2023-10-02 21:13:36,323 - Epoch: [93][ 210/ 1236] Overall Loss 0.258908 Objective Loss 0.258908 LR 0.001000 Time 0.024972 +2023-10-02 21:13:36,530 - Epoch: [93][ 220/ 1236] Overall Loss 0.258460 Objective Loss 0.258460 LR 0.001000 Time 0.024777 +2023-10-02 21:13:36,736 - Epoch: [93][ 230/ 1236] Overall Loss 0.258412 Objective Loss 0.258412 LR 0.001000 Time 0.024592 +2023-10-02 21:13:36,945 - Epoch: [93][ 240/ 1236] Overall Loss 0.258607 Objective Loss 0.258607 LR 0.001000 Time 0.024437 +2023-10-02 21:13:37,152 - Epoch: [93][ 250/ 1236] Overall Loss 0.259251 Objective Loss 0.259251 LR 0.001000 Time 0.024283 +2023-10-02 21:13:37,358 - Epoch: [93][ 260/ 1236] Overall Loss 0.260558 Objective Loss 0.260558 LR 0.001000 Time 0.024138 +2023-10-02 21:13:37,567 - Epoch: [93][ 270/ 1236] Overall Loss 0.260390 Objective Loss 0.260390 LR 0.001000 Time 0.024017 +2023-10-02 21:13:37,772 - Epoch: [93][ 280/ 1236] Overall Loss 0.262575 Objective Loss 0.262575 LR 0.001000 Time 0.023892 +2023-10-02 21:13:37,981 - Epoch: [93][ 290/ 1236] Overall Loss 0.263788 Objective Loss 0.263788 LR 0.001000 Time 0.023787 +2023-10-02 21:13:38,187 - Epoch: [93][ 300/ 1236] Overall Loss 0.263407 Objective Loss 0.263407 LR 0.001000 Time 0.023678 +2023-10-02 21:13:38,394 - Epoch: [93][ 310/ 1236] Overall Loss 0.263430 Objective Loss 0.263430 LR 0.001000 Time 0.023583 +2023-10-02 21:13:38,602 - Epoch: [93][ 320/ 1236] Overall Loss 0.263299 Objective Loss 0.263299 LR 0.001000 Time 0.023489 +2023-10-02 21:13:38,812 - Epoch: [93][ 330/ 1236] Overall Loss 0.262549 Objective Loss 0.262549 LR 0.001000 Time 0.023414 +2023-10-02 21:13:39,019 - Epoch: [93][ 340/ 1236] Overall Loss 0.261514 Objective Loss 0.261514 LR 0.001000 Time 0.023332 +2023-10-02 21:13:39,229 - Epoch: [93][ 350/ 1236] Overall Loss 0.262040 Objective Loss 0.262040 LR 0.001000 Time 0.023264 +2023-10-02 21:13:39,435 - Epoch: [93][ 360/ 1236] Overall Loss 0.262127 Objective Loss 0.262127 LR 0.001000 Time 0.023190 +2023-10-02 21:13:39,645 - Epoch: [93][ 370/ 1236] Overall Loss 0.261896 Objective Loss 0.261896 LR 0.001000 Time 0.023131 +2023-10-02 21:13:39,852 - Epoch: [93][ 380/ 1236] Overall Loss 0.262792 Objective Loss 0.262792 LR 0.001000 Time 0.023064 +2023-10-02 21:13:40,061 - Epoch: [93][ 390/ 1236] Overall Loss 0.263586 Objective Loss 0.263586 LR 0.001000 Time 0.023010 +2023-10-02 21:13:40,268 - Epoch: [93][ 400/ 1236] Overall Loss 0.264195 Objective Loss 0.264195 LR 0.001000 Time 0.022950 +2023-10-02 21:13:40,478 - Epoch: [93][ 410/ 1236] Overall Loss 0.264205 Objective Loss 0.264205 LR 0.001000 Time 0.022902 +2023-10-02 21:13:40,684 - Epoch: [93][ 420/ 1236] Overall Loss 0.264751 Objective Loss 0.264751 LR 0.001000 Time 0.022848 +2023-10-02 21:13:40,894 - Epoch: [93][ 430/ 1236] Overall Loss 0.264679 Objective Loss 0.264679 LR 0.001000 Time 0.022804 +2023-10-02 21:13:41,101 - Epoch: [93][ 440/ 1236] Overall Loss 0.264874 Objective Loss 0.264874 LR 0.001000 Time 0.022754 +2023-10-02 21:13:41,311 - Epoch: [93][ 450/ 1236] Overall Loss 0.264887 Objective Loss 0.264887 LR 0.001000 Time 0.022714 +2023-10-02 21:13:41,517 - Epoch: [93][ 460/ 1236] Overall Loss 0.264787 Objective Loss 0.264787 LR 0.001000 Time 0.022669 +2023-10-02 21:13:41,727 - Epoch: [93][ 470/ 1236] Overall Loss 0.265268 Objective Loss 0.265268 LR 0.001000 Time 0.022633 +2023-10-02 21:13:41,934 - Epoch: [93][ 480/ 1236] Overall Loss 0.265501 Objective Loss 0.265501 LR 0.001000 Time 0.022591 +2023-10-02 21:13:42,144 - Epoch: [93][ 490/ 1236] Overall Loss 0.265967 Objective Loss 0.265967 LR 0.001000 Time 0.022558 +2023-10-02 21:13:42,350 - Epoch: [93][ 500/ 1236] Overall Loss 0.265444 Objective Loss 0.265444 LR 0.001000 Time 0.022519 +2023-10-02 21:13:42,560 - Epoch: [93][ 510/ 1236] Overall Loss 0.265675 Objective Loss 0.265675 LR 0.001000 Time 0.022489 +2023-10-02 21:13:42,767 - Epoch: [93][ 520/ 1236] Overall Loss 0.264342 Objective Loss 0.264342 LR 0.001000 Time 0.022453 +2023-10-02 21:13:42,977 - Epoch: [93][ 530/ 1236] Overall Loss 0.264206 Objective Loss 0.264206 LR 0.001000 Time 0.022425 +2023-10-02 21:13:43,184 - Epoch: [93][ 540/ 1236] Overall Loss 0.264486 Objective Loss 0.264486 LR 0.001000 Time 0.022392 +2023-10-02 21:13:43,394 - Epoch: [93][ 550/ 1236] Overall Loss 0.264697 Objective Loss 0.264697 LR 0.001000 Time 0.022366 +2023-10-02 21:13:43,600 - Epoch: [93][ 560/ 1236] Overall Loss 0.265045 Objective Loss 0.265045 LR 0.001000 Time 0.022335 +2023-10-02 21:13:43,811 - Epoch: [93][ 570/ 1236] Overall Loss 0.264996 Objective Loss 0.264996 LR 0.001000 Time 0.022312 +2023-10-02 21:13:44,017 - Epoch: [93][ 580/ 1236] Overall Loss 0.265299 Objective Loss 0.265299 LR 0.001000 Time 0.022283 +2023-10-02 21:13:44,228 - Epoch: [93][ 590/ 1236] Overall Loss 0.265625 Objective Loss 0.265625 LR 0.001000 Time 0.022261 +2023-10-02 21:13:44,434 - Epoch: [93][ 600/ 1236] Overall Loss 0.265829 Objective Loss 0.265829 LR 0.001000 Time 0.022234 +2023-10-02 21:13:44,645 - Epoch: [93][ 610/ 1236] Overall Loss 0.265998 Objective Loss 0.265998 LR 0.001000 Time 0.022214 +2023-10-02 21:13:44,851 - Epoch: [93][ 620/ 1236] Overall Loss 0.266145 Objective Loss 0.266145 LR 0.001000 Time 0.022189 +2023-10-02 21:13:45,062 - Epoch: [93][ 630/ 1236] Overall Loss 0.266726 Objective Loss 0.266726 LR 0.001000 Time 0.022170 +2023-10-02 21:13:45,269 - Epoch: [93][ 640/ 1236] Overall Loss 0.266470 Objective Loss 0.266470 LR 0.001000 Time 0.022147 +2023-10-02 21:13:45,479 - Epoch: [93][ 650/ 1236] Overall Loss 0.266417 Objective Loss 0.266417 LR 0.001000 Time 0.022128 +2023-10-02 21:13:45,686 - Epoch: [93][ 660/ 1236] Overall Loss 0.266611 Objective Loss 0.266611 LR 0.001000 Time 0.022106 +2023-10-02 21:13:45,896 - Epoch: [93][ 670/ 1236] Overall Loss 0.267332 Objective Loss 0.267332 LR 0.001000 Time 0.022089 +2023-10-02 21:13:46,102 - Epoch: [93][ 680/ 1236] Overall Loss 0.267280 Objective Loss 0.267280 LR 0.001000 Time 0.022067 +2023-10-02 21:13:46,313 - Epoch: [93][ 690/ 1236] Overall Loss 0.266667 Objective Loss 0.266667 LR 0.001000 Time 0.022052 +2023-10-02 21:13:46,520 - Epoch: [93][ 700/ 1236] Overall Loss 0.266622 Objective Loss 0.266622 LR 0.001000 Time 0.022032 +2023-10-02 21:13:46,730 - Epoch: [93][ 710/ 1236] Overall Loss 0.266505 Objective Loss 0.266505 LR 0.001000 Time 0.022018 +2023-10-02 21:13:46,941 - Epoch: [93][ 720/ 1236] Overall Loss 0.266366 Objective Loss 0.266366 LR 0.001000 Time 0.022000 +2023-10-02 21:13:47,151 - Epoch: [93][ 730/ 1236] Overall Loss 0.266131 Objective Loss 0.266131 LR 0.001000 Time 0.021987 +2023-10-02 21:13:47,358 - Epoch: [93][ 740/ 1236] Overall Loss 0.266008 Objective Loss 0.266008 LR 0.001000 Time 0.021969 +2023-10-02 21:13:47,569 - Epoch: [93][ 750/ 1236] Overall Loss 0.266173 Objective Loss 0.266173 LR 0.001000 Time 0.021957 +2023-10-02 21:13:47,777 - Epoch: [93][ 760/ 1236] Overall Loss 0.266116 Objective Loss 0.266116 LR 0.001000 Time 0.021940 +2023-10-02 21:13:47,986 - Epoch: [93][ 770/ 1236] Overall Loss 0.266155 Objective Loss 0.266155 LR 0.001000 Time 0.021926 +2023-10-02 21:13:48,194 - Epoch: [93][ 780/ 1236] Overall Loss 0.265958 Objective Loss 0.265958 LR 0.001000 Time 0.021910 +2023-10-02 21:13:48,405 - Epoch: [93][ 790/ 1236] Overall Loss 0.266267 Objective Loss 0.266267 LR 0.001000 Time 0.021899 +2023-10-02 21:13:48,612 - Epoch: [93][ 800/ 1236] Overall Loss 0.266270 Objective Loss 0.266270 LR 0.001000 Time 0.021884 +2023-10-02 21:13:48,823 - Epoch: [93][ 810/ 1236] Overall Loss 0.266071 Objective Loss 0.266071 LR 0.001000 Time 0.021874 +2023-10-02 21:13:49,030 - Epoch: [93][ 820/ 1236] Overall Loss 0.266378 Objective Loss 0.266378 LR 0.001000 Time 0.021859 +2023-10-02 21:13:49,240 - Epoch: [93][ 830/ 1236] Overall Loss 0.266293 Objective Loss 0.266293 LR 0.001000 Time 0.021849 +2023-10-02 21:13:49,447 - Epoch: [93][ 840/ 1236] Overall Loss 0.266587 Objective Loss 0.266587 LR 0.001000 Time 0.021835 +2023-10-02 21:13:49,658 - Epoch: [93][ 850/ 1236] Overall Loss 0.266250 Objective Loss 0.266250 LR 0.001000 Time 0.021826 +2023-10-02 21:13:49,865 - Epoch: [93][ 860/ 1236] Overall Loss 0.266195 Objective Loss 0.266195 LR 0.001000 Time 0.021813 +2023-10-02 21:13:50,076 - Epoch: [93][ 870/ 1236] Overall Loss 0.266072 Objective Loss 0.266072 LR 0.001000 Time 0.021804 +2023-10-02 21:13:50,284 - Epoch: [93][ 880/ 1236] Overall Loss 0.266296 Objective Loss 0.266296 LR 0.001000 Time 0.021792 +2023-10-02 21:13:50,494 - Epoch: [93][ 890/ 1236] Overall Loss 0.266423 Objective Loss 0.266423 LR 0.001000 Time 0.021783 +2023-10-02 21:13:50,702 - Epoch: [93][ 900/ 1236] Overall Loss 0.266714 Objective Loss 0.266714 LR 0.001000 Time 0.021771 +2023-10-02 21:13:50,912 - Epoch: [93][ 910/ 1236] Overall Loss 0.266535 Objective Loss 0.266535 LR 0.001000 Time 0.021763 +2023-10-02 21:13:51,119 - Epoch: [93][ 920/ 1236] Overall Loss 0.266427 Objective Loss 0.266427 LR 0.001000 Time 0.021751 +2023-10-02 21:13:51,330 - Epoch: [93][ 930/ 1236] Overall Loss 0.266738 Objective Loss 0.266738 LR 0.001000 Time 0.021744 +2023-10-02 21:13:51,538 - Epoch: [93][ 940/ 1236] Overall Loss 0.267343 Objective Loss 0.267343 LR 0.001000 Time 0.021733 +2023-10-02 21:13:51,749 - Epoch: [93][ 950/ 1236] Overall Loss 0.267675 Objective Loss 0.267675 LR 0.001000 Time 0.021726 +2023-10-02 21:13:51,956 - Epoch: [93][ 960/ 1236] Overall Loss 0.267887 Objective Loss 0.267887 LR 0.001000 Time 0.021715 +2023-10-02 21:13:52,167 - Epoch: [93][ 970/ 1236] Overall Loss 0.267786 Objective Loss 0.267786 LR 0.001000 Time 0.021708 +2023-10-02 21:13:52,374 - Epoch: [93][ 980/ 1236] Overall Loss 0.267729 Objective Loss 0.267729 LR 0.001000 Time 0.021698 +2023-10-02 21:13:52,585 - Epoch: [93][ 990/ 1236] Overall Loss 0.268059 Objective Loss 0.268059 LR 0.001000 Time 0.021691 +2023-10-02 21:13:52,792 - Epoch: [93][ 1000/ 1236] Overall Loss 0.267991 Objective Loss 0.267991 LR 0.001000 Time 0.021681 +2023-10-02 21:13:53,003 - Epoch: [93][ 1010/ 1236] Overall Loss 0.268055 Objective Loss 0.268055 LR 0.001000 Time 0.021675 +2023-10-02 21:13:53,210 - Epoch: [93][ 1020/ 1236] Overall Loss 0.268137 Objective Loss 0.268137 LR 0.001000 Time 0.021665 +2023-10-02 21:13:53,421 - Epoch: [93][ 1030/ 1236] Overall Loss 0.267981 Objective Loss 0.267981 LR 0.001000 Time 0.021659 +2023-10-02 21:13:53,628 - Epoch: [93][ 1040/ 1236] Overall Loss 0.267644 Objective Loss 0.267644 LR 0.001000 Time 0.021650 +2023-10-02 21:13:53,839 - Epoch: [93][ 1050/ 1236] Overall Loss 0.267653 Objective Loss 0.267653 LR 0.001000 Time 0.021644 +2023-10-02 21:13:54,045 - Epoch: [93][ 1060/ 1236] Overall Loss 0.267406 Objective Loss 0.267406 LR 0.001000 Time 0.021635 +2023-10-02 21:13:54,257 - Epoch: [93][ 1070/ 1236] Overall Loss 0.267690 Objective Loss 0.267690 LR 0.001000 Time 0.021630 +2023-10-02 21:13:54,464 - Epoch: [93][ 1080/ 1236] Overall Loss 0.267512 Objective Loss 0.267512 LR 0.001000 Time 0.021621 +2023-10-02 21:13:54,675 - Epoch: [93][ 1090/ 1236] Overall Loss 0.267652 Objective Loss 0.267652 LR 0.001000 Time 0.021616 +2023-10-02 21:13:54,882 - Epoch: [93][ 1100/ 1236] Overall Loss 0.267652 Objective Loss 0.267652 LR 0.001000 Time 0.021607 +2023-10-02 21:13:55,093 - Epoch: [93][ 1110/ 1236] Overall Loss 0.267566 Objective Loss 0.267566 LR 0.001000 Time 0.021602 +2023-10-02 21:13:55,300 - Epoch: [93][ 1120/ 1236] Overall Loss 0.267759 Objective Loss 0.267759 LR 0.001000 Time 0.021594 +2023-10-02 21:13:55,511 - Epoch: [93][ 1130/ 1236] Overall Loss 0.267857 Objective Loss 0.267857 LR 0.001000 Time 0.021589 +2023-10-02 21:13:55,718 - Epoch: [93][ 1140/ 1236] Overall Loss 0.267992 Objective Loss 0.267992 LR 0.001000 Time 0.021582 +2023-10-02 21:13:55,929 - Epoch: [93][ 1150/ 1236] Overall Loss 0.267992 Objective Loss 0.267992 LR 0.001000 Time 0.021577 +2023-10-02 21:13:56,136 - Epoch: [93][ 1160/ 1236] Overall Loss 0.268386 Objective Loss 0.268386 LR 0.001000 Time 0.021569 +2023-10-02 21:13:56,347 - Epoch: [93][ 1170/ 1236] Overall Loss 0.268271 Objective Loss 0.268271 LR 0.001000 Time 0.021565 +2023-10-02 21:13:56,554 - Epoch: [93][ 1180/ 1236] Overall Loss 0.268179 Objective Loss 0.268179 LR 0.001000 Time 0.021557 +2023-10-02 21:13:56,765 - Epoch: [93][ 1190/ 1236] Overall Loss 0.268361 Objective Loss 0.268361 LR 0.001000 Time 0.021553 +2023-10-02 21:13:56,972 - Epoch: [93][ 1200/ 1236] Overall Loss 0.268463 Objective Loss 0.268463 LR 0.001000 Time 0.021546 +2023-10-02 21:13:57,183 - Epoch: [93][ 1210/ 1236] Overall Loss 0.268595 Objective Loss 0.268595 LR 0.001000 Time 0.021542 +2023-10-02 21:13:57,390 - Epoch: [93][ 1220/ 1236] Overall Loss 0.268488 Objective Loss 0.268488 LR 0.001000 Time 0.021535 +2023-10-02 21:13:57,652 - Epoch: [93][ 1230/ 1236] Overall Loss 0.268631 Objective Loss 0.268631 LR 0.001000 Time 0.021572 +2023-10-02 21:13:57,774 - Epoch: [93][ 1236/ 1236] Overall Loss 0.268778 Objective Loss 0.268778 Top1 82.892057 Top5 97.556008 LR 0.001000 Time 0.021566 +2023-10-02 21:13:57,926 - --- validate (epoch=93)----------- +2023-10-02 21:13:57,926 - 29943 samples (256 per mini-batch) +2023-10-02 21:13:58,416 - Epoch: [93][ 10/ 117] Loss 0.318174 Top1 83.398438 Top5 97.812500 +2023-10-02 21:13:58,565 - Epoch: [93][ 20/ 117] Loss 0.328603 Top1 83.046875 Top5 97.832031 +2023-10-02 21:13:58,714 - Epoch: [93][ 30/ 117] Loss 0.345603 Top1 82.656250 Top5 97.773438 +2023-10-02 21:13:58,861 - Epoch: [93][ 40/ 117] Loss 0.354234 Top1 82.548828 Top5 97.714844 +2023-10-02 21:13:59,008 - Epoch: [93][ 50/ 117] Loss 0.357989 Top1 82.398438 Top5 97.710938 +2023-10-02 21:13:59,153 - Epoch: [93][ 60/ 117] Loss 0.355994 Top1 82.506510 Top5 97.832031 +2023-10-02 21:13:59,298 - Epoch: [93][ 70/ 117] Loss 0.359057 Top1 82.327009 Top5 97.712054 +2023-10-02 21:13:59,443 - Epoch: [93][ 80/ 117] Loss 0.359721 Top1 82.324219 Top5 97.734375 +2023-10-02 21:13:59,589 - Epoch: [93][ 90/ 117] Loss 0.364765 Top1 82.057292 Top5 97.769097 +2023-10-02 21:13:59,734 - Epoch: [93][ 100/ 117] Loss 0.366400 Top1 82.074219 Top5 97.781250 +2023-10-02 21:13:59,889 - Epoch: [93][ 110/ 117] Loss 0.360730 Top1 82.215909 Top5 97.805398 +2023-10-02 21:13:59,977 - Epoch: [93][ 117/ 117] Loss 0.361372 Top1 82.186154 Top5 97.825869 +2023-10-02 21:14:00,077 - ==> Top1: 82.186 Top5: 97.826 Loss: 0.361 + +2023-10-02 21:14:00,078 - ==> Confusion: +[[ 914 1 5 0 7 5 0 1 9 81 3 2 1 2 4 0 5 1 0 0 9] + [ 1 1047 2 1 8 19 1 20 2 3 3 1 1 0 3 3 2 3 8 0 3] + [ 1 0 965 14 1 2 22 6 0 4 0 1 6 3 2 4 3 1 8 5 8] + [ 0 4 25 933 0 8 1 2 4 0 4 1 12 6 42 1 1 8 23 2 12] + [ 28 9 0 1 943 13 0 0 1 17 2 0 0 1 9 3 14 2 0 3 4] + [ 1 46 0 1 1 991 0 11 1 5 2 6 3 11 8 0 4 1 0 13 11] + [ 0 2 29 0 0 1 1126 4 0 0 6 1 0 0 0 2 1 1 3 11 4] + [ 7 27 26 2 2 49 14 987 2 5 6 10 3 1 4 0 2 3 37 22 9] + [ 16 2 0 2 2 4 0 0 963 50 9 1 1 13 14 1 5 0 3 0 3] + [ 98 0 0 2 0 3 1 0 26 951 1 1 1 21 5 1 3 1 0 1 3] + [ 3 3 9 16 2 6 2 1 20 4 948 1 0 9 7 0 0 3 7 4 8] + [ 0 0 2 1 1 12 0 1 0 0 0 928 41 7 0 0 1 11 0 25 5] + [ 1 0 1 1 1 2 3 1 2 2 5 58 949 3 1 4 2 9 3 9 11] + [ 1 0 3 0 4 22 2 0 18 13 8 11 1 1016 4 0 2 2 0 4 8] + [ 10 2 3 19 5 1 0 0 17 9 2 0 3 5 1002 0 2 7 8 0 6] + [ 0 0 3 1 3 1 0 0 0 0 0 7 8 0 0 1053 29 9 2 12 6] + [ 1 18 2 1 4 7 1 0 3 1 0 4 0 2 1 3 1096 1 0 11 5] + [ 1 0 1 1 0 1 2 0 2 1 0 17 34 1 1 6 2 962 0 4 2] + [ 1 2 7 8 0 1 1 13 3 2 4 1 1 0 13 0 2 0 994 4 11] + [ 0 3 7 0 0 8 11 7 0 1 1 13 2 2 0 1 5 1 1 1088 1] + [ 150 176 140 71 81 239 41 90 121 138 205 141 371 337 149 53 136 60 134 319 4753]] + +2023-10-02 21:14:00,079 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:14:00,079 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:14:00,085 - + +2023-10-02 21:14:00,086 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:14:01,207 - Epoch: [94][ 10/ 1236] Overall Loss 0.267015 Objective Loss 0.267015 LR 0.001000 Time 0.112085 +2023-10-02 21:14:01,416 - Epoch: [94][ 20/ 1236] Overall Loss 0.284918 Objective Loss 0.284918 LR 0.001000 Time 0.066448 +2023-10-02 21:14:01,624 - Epoch: [94][ 30/ 1236] Overall Loss 0.283061 Objective Loss 0.283061 LR 0.001000 Time 0.051240 +2023-10-02 21:14:01,833 - Epoch: [94][ 40/ 1236] Overall Loss 0.279847 Objective Loss 0.279847 LR 0.001000 Time 0.043641 +2023-10-02 21:14:02,041 - Epoch: [94][ 50/ 1236] Overall Loss 0.273942 Objective Loss 0.273942 LR 0.001000 Time 0.039041 +2023-10-02 21:14:02,250 - Epoch: [94][ 60/ 1236] Overall Loss 0.272079 Objective Loss 0.272079 LR 0.001000 Time 0.036025 +2023-10-02 21:14:02,457 - Epoch: [94][ 70/ 1236] Overall Loss 0.270289 Objective Loss 0.270289 LR 0.001000 Time 0.033828 +2023-10-02 21:14:02,666 - Epoch: [94][ 80/ 1236] Overall Loss 0.267957 Objective Loss 0.267957 LR 0.001000 Time 0.032206 +2023-10-02 21:14:02,874 - Epoch: [94][ 90/ 1236] Overall Loss 0.265335 Objective Loss 0.265335 LR 0.001000 Time 0.030923 +2023-10-02 21:14:03,084 - Epoch: [94][ 100/ 1236] Overall Loss 0.263573 Objective Loss 0.263573 LR 0.001000 Time 0.029930 +2023-10-02 21:14:03,291 - Epoch: [94][ 110/ 1236] Overall Loss 0.263783 Objective Loss 0.263783 LR 0.001000 Time 0.029088 +2023-10-02 21:14:03,502 - Epoch: [94][ 120/ 1236] Overall Loss 0.260787 Objective Loss 0.260787 LR 0.001000 Time 0.028415 +2023-10-02 21:14:03,709 - Epoch: [94][ 130/ 1236] Overall Loss 0.261020 Objective Loss 0.261020 LR 0.001000 Time 0.027820 +2023-10-02 21:14:03,918 - Epoch: [94][ 140/ 1236] Overall Loss 0.260901 Objective Loss 0.260901 LR 0.001000 Time 0.027324 +2023-10-02 21:14:04,126 - Epoch: [94][ 150/ 1236] Overall Loss 0.259338 Objective Loss 0.259338 LR 0.001000 Time 0.026882 +2023-10-02 21:14:04,337 - Epoch: [94][ 160/ 1236] Overall Loss 0.258291 Objective Loss 0.258291 LR 0.001000 Time 0.026519 +2023-10-02 21:14:04,544 - Epoch: [94][ 170/ 1236] Overall Loss 0.258167 Objective Loss 0.258167 LR 0.001000 Time 0.026176 +2023-10-02 21:14:04,753 - Epoch: [94][ 180/ 1236] Overall Loss 0.257404 Objective Loss 0.257404 LR 0.001000 Time 0.025882 +2023-10-02 21:14:04,962 - Epoch: [94][ 190/ 1236] Overall Loss 0.257120 Objective Loss 0.257120 LR 0.001000 Time 0.025611 +2023-10-02 21:14:05,171 - Epoch: [94][ 200/ 1236] Overall Loss 0.257991 Objective Loss 0.257991 LR 0.001000 Time 0.025372 +2023-10-02 21:14:05,377 - Epoch: [94][ 210/ 1236] Overall Loss 0.257345 Objective Loss 0.257345 LR 0.001000 Time 0.025139 +2023-10-02 21:14:05,584 - Epoch: [94][ 220/ 1236] Overall Loss 0.258691 Objective Loss 0.258691 LR 0.001000 Time 0.024936 +2023-10-02 21:14:05,792 - Epoch: [94][ 230/ 1236] Overall Loss 0.258672 Objective Loss 0.258672 LR 0.001000 Time 0.024753 +2023-10-02 21:14:06,004 - Epoch: [94][ 240/ 1236] Overall Loss 0.257551 Objective Loss 0.257551 LR 0.001000 Time 0.024602 +2023-10-02 21:14:06,215 - Epoch: [94][ 250/ 1236] Overall Loss 0.257063 Objective Loss 0.257063 LR 0.001000 Time 0.024460 +2023-10-02 21:14:06,429 - Epoch: [94][ 260/ 1236] Overall Loss 0.256648 Objective Loss 0.256648 LR 0.001000 Time 0.024341 +2023-10-02 21:14:06,639 - Epoch: [94][ 270/ 1236] Overall Loss 0.255625 Objective Loss 0.255625 LR 0.001000 Time 0.024217 +2023-10-02 21:14:06,851 - Epoch: [94][ 280/ 1236] Overall Loss 0.256717 Objective Loss 0.256717 LR 0.001000 Time 0.024106 +2023-10-02 21:14:07,059 - Epoch: [94][ 290/ 1236] Overall Loss 0.256300 Objective Loss 0.256300 LR 0.001000 Time 0.023991 +2023-10-02 21:14:07,270 - Epoch: [94][ 300/ 1236] Overall Loss 0.257211 Objective Loss 0.257211 LR 0.001000 Time 0.023895 +2023-10-02 21:14:07,478 - Epoch: [94][ 310/ 1236] Overall Loss 0.257161 Objective Loss 0.257161 LR 0.001000 Time 0.023794 +2023-10-02 21:14:07,690 - Epoch: [94][ 320/ 1236] Overall Loss 0.258766 Objective Loss 0.258766 LR 0.001000 Time 0.023711 +2023-10-02 21:14:07,897 - Epoch: [94][ 330/ 1236] Overall Loss 0.260283 Objective Loss 0.260283 LR 0.001000 Time 0.023620 +2023-10-02 21:14:08,104 - Epoch: [94][ 340/ 1236] Overall Loss 0.260460 Objective Loss 0.260460 LR 0.001000 Time 0.023534 +2023-10-02 21:14:08,312 - Epoch: [94][ 350/ 1236] Overall Loss 0.260535 Objective Loss 0.260535 LR 0.001000 Time 0.023450 +2023-10-02 21:14:08,522 - Epoch: [94][ 360/ 1236] Overall Loss 0.260644 Objective Loss 0.260644 LR 0.001000 Time 0.023382 +2023-10-02 21:14:08,729 - Epoch: [94][ 370/ 1236] Overall Loss 0.260917 Objective Loss 0.260917 LR 0.001000 Time 0.023309 +2023-10-02 21:14:08,940 - Epoch: [94][ 380/ 1236] Overall Loss 0.260367 Objective Loss 0.260367 LR 0.001000 Time 0.023250 +2023-10-02 21:14:09,147 - Epoch: [94][ 390/ 1236] Overall Loss 0.260645 Objective Loss 0.260645 LR 0.001000 Time 0.023185 +2023-10-02 21:14:09,358 - Epoch: [94][ 400/ 1236] Overall Loss 0.261028 Objective Loss 0.261028 LR 0.001000 Time 0.023131 +2023-10-02 21:14:09,565 - Epoch: [94][ 410/ 1236] Overall Loss 0.261771 Objective Loss 0.261771 LR 0.001000 Time 0.023072 +2023-10-02 21:14:09,776 - Epoch: [94][ 420/ 1236] Overall Loss 0.262536 Objective Loss 0.262536 LR 0.001000 Time 0.023024 +2023-10-02 21:14:09,984 - Epoch: [94][ 430/ 1236] Overall Loss 0.262358 Objective Loss 0.262358 LR 0.001000 Time 0.022970 +2023-10-02 21:14:10,193 - Epoch: [94][ 440/ 1236] Overall Loss 0.262790 Objective Loss 0.262790 LR 0.001000 Time 0.022923 +2023-10-02 21:14:10,402 - Epoch: [94][ 450/ 1236] Overall Loss 0.262997 Objective Loss 0.262997 LR 0.001000 Time 0.022874 +2023-10-02 21:14:10,612 - Epoch: [94][ 460/ 1236] Overall Loss 0.263566 Objective Loss 0.263566 LR 0.001000 Time 0.022835 +2023-10-02 21:14:10,820 - Epoch: [94][ 470/ 1236] Overall Loss 0.264215 Objective Loss 0.264215 LR 0.001000 Time 0.022790 +2023-10-02 21:14:11,031 - Epoch: [94][ 480/ 1236] Overall Loss 0.264169 Objective Loss 0.264169 LR 0.001000 Time 0.022754 +2023-10-02 21:14:11,238 - Epoch: [94][ 490/ 1236] Overall Loss 0.264310 Objective Loss 0.264310 LR 0.001000 Time 0.022712 +2023-10-02 21:14:11,448 - Epoch: [94][ 500/ 1236] Overall Loss 0.264476 Objective Loss 0.264476 LR 0.001000 Time 0.022677 +2023-10-02 21:14:11,657 - Epoch: [94][ 510/ 1236] Overall Loss 0.264946 Objective Loss 0.264946 LR 0.001000 Time 0.022639 +2023-10-02 21:14:11,868 - Epoch: [94][ 520/ 1236] Overall Loss 0.265589 Objective Loss 0.265589 LR 0.001000 Time 0.022609 +2023-10-02 21:14:12,075 - Epoch: [94][ 530/ 1236] Overall Loss 0.265961 Objective Loss 0.265961 LR 0.001000 Time 0.022573 +2023-10-02 21:14:12,286 - Epoch: [94][ 540/ 1236] Overall Loss 0.266412 Objective Loss 0.266412 LR 0.001000 Time 0.022545 +2023-10-02 21:14:12,493 - Epoch: [94][ 550/ 1236] Overall Loss 0.266265 Objective Loss 0.266265 LR 0.001000 Time 0.022511 +2023-10-02 21:14:12,705 - Epoch: [94][ 560/ 1236] Overall Loss 0.266549 Objective Loss 0.266549 LR 0.001000 Time 0.022486 +2023-10-02 21:14:12,912 - Epoch: [94][ 570/ 1236] Overall Loss 0.266932 Objective Loss 0.266932 LR 0.001000 Time 0.022454 +2023-10-02 21:14:13,123 - Epoch: [94][ 580/ 1236] Overall Loss 0.266244 Objective Loss 0.266244 LR 0.001000 Time 0.022431 +2023-10-02 21:14:13,330 - Epoch: [94][ 590/ 1236] Overall Loss 0.266411 Objective Loss 0.266411 LR 0.001000 Time 0.022401 +2023-10-02 21:14:13,541 - Epoch: [94][ 600/ 1236] Overall Loss 0.266610 Objective Loss 0.266610 LR 0.001000 Time 0.022379 +2023-10-02 21:14:13,748 - Epoch: [94][ 610/ 1236] Overall Loss 0.266713 Objective Loss 0.266713 LR 0.001000 Time 0.022352 +2023-10-02 21:14:13,958 - Epoch: [94][ 620/ 1236] Overall Loss 0.266736 Objective Loss 0.266736 LR 0.001000 Time 0.022329 +2023-10-02 21:14:14,167 - Epoch: [94][ 630/ 1236] Overall Loss 0.267160 Objective Loss 0.267160 LR 0.001000 Time 0.022304 +2023-10-02 21:14:14,378 - Epoch: [94][ 640/ 1236] Overall Loss 0.267282 Objective Loss 0.267282 LR 0.001000 Time 0.022285 +2023-10-02 21:14:14,585 - Epoch: [94][ 650/ 1236] Overall Loss 0.267531 Objective Loss 0.267531 LR 0.001000 Time 0.022260 +2023-10-02 21:14:14,797 - Epoch: [94][ 660/ 1236] Overall Loss 0.267988 Objective Loss 0.267988 LR 0.001000 Time 0.022243 +2023-10-02 21:14:15,003 - Epoch: [94][ 670/ 1236] Overall Loss 0.268424 Objective Loss 0.268424 LR 0.001000 Time 0.022219 +2023-10-02 21:14:15,213 - Epoch: [94][ 680/ 1236] Overall Loss 0.268149 Objective Loss 0.268149 LR 0.001000 Time 0.022200 +2023-10-02 21:14:15,419 - Epoch: [94][ 690/ 1236] Overall Loss 0.268307 Objective Loss 0.268307 LR 0.001000 Time 0.022177 +2023-10-02 21:14:15,630 - Epoch: [94][ 700/ 1236] Overall Loss 0.268354 Objective Loss 0.268354 LR 0.001000 Time 0.022160 +2023-10-02 21:14:15,836 - Epoch: [94][ 710/ 1236] Overall Loss 0.268684 Objective Loss 0.268684 LR 0.001000 Time 0.022139 +2023-10-02 21:14:16,047 - Epoch: [94][ 720/ 1236] Overall Loss 0.268746 Objective Loss 0.268746 LR 0.001000 Time 0.022123 +2023-10-02 21:14:16,253 - Epoch: [94][ 730/ 1236] Overall Loss 0.269168 Objective Loss 0.269168 LR 0.001000 Time 0.022102 +2023-10-02 21:14:16,462 - Epoch: [94][ 740/ 1236] Overall Loss 0.269630 Objective Loss 0.269630 LR 0.001000 Time 0.022086 +2023-10-02 21:14:16,670 - Epoch: [94][ 750/ 1236] Overall Loss 0.269454 Objective Loss 0.269454 LR 0.001000 Time 0.022068 +2023-10-02 21:14:16,881 - Epoch: [94][ 760/ 1236] Overall Loss 0.269377 Objective Loss 0.269377 LR 0.001000 Time 0.022054 +2023-10-02 21:14:17,087 - Epoch: [94][ 770/ 1236] Overall Loss 0.269825 Objective Loss 0.269825 LR 0.001000 Time 0.022036 +2023-10-02 21:14:17,298 - Epoch: [94][ 780/ 1236] Overall Loss 0.270163 Objective Loss 0.270163 LR 0.001000 Time 0.022023 +2023-10-02 21:14:17,504 - Epoch: [94][ 790/ 1236] Overall Loss 0.270085 Objective Loss 0.270085 LR 0.001000 Time 0.022005 +2023-10-02 21:14:17,714 - Epoch: [94][ 800/ 1236] Overall Loss 0.270001 Objective Loss 0.270001 LR 0.001000 Time 0.021992 +2023-10-02 21:14:17,921 - Epoch: [94][ 810/ 1236] Overall Loss 0.269900 Objective Loss 0.269900 LR 0.001000 Time 0.021975 +2023-10-02 21:14:18,131 - Epoch: [94][ 820/ 1236] Overall Loss 0.269835 Objective Loss 0.269835 LR 0.001000 Time 0.021963 +2023-10-02 21:14:18,337 - Epoch: [94][ 830/ 1236] Overall Loss 0.269542 Objective Loss 0.269542 LR 0.001000 Time 0.021947 +2023-10-02 21:14:18,548 - Epoch: [94][ 840/ 1236] Overall Loss 0.269402 Objective Loss 0.269402 LR 0.001000 Time 0.021936 +2023-10-02 21:14:18,754 - Epoch: [94][ 850/ 1236] Overall Loss 0.269294 Objective Loss 0.269294 LR 0.001000 Time 0.021921 +2023-10-02 21:14:18,965 - Epoch: [94][ 860/ 1236] Overall Loss 0.269373 Objective Loss 0.269373 LR 0.001000 Time 0.021910 +2023-10-02 21:14:19,171 - Epoch: [94][ 870/ 1236] Overall Loss 0.269284 Objective Loss 0.269284 LR 0.001000 Time 0.021895 +2023-10-02 21:14:19,382 - Epoch: [94][ 880/ 1236] Overall Loss 0.269051 Objective Loss 0.269051 LR 0.001000 Time 0.021885 +2023-10-02 21:14:19,588 - Epoch: [94][ 890/ 1236] Overall Loss 0.268976 Objective Loss 0.268976 LR 0.001000 Time 0.021871 +2023-10-02 21:14:19,797 - Epoch: [94][ 900/ 1236] Overall Loss 0.268774 Objective Loss 0.268774 LR 0.001000 Time 0.021860 +2023-10-02 21:14:20,005 - Epoch: [94][ 910/ 1236] Overall Loss 0.268826 Objective Loss 0.268826 LR 0.001000 Time 0.021847 +2023-10-02 21:14:20,214 - Epoch: [94][ 920/ 1236] Overall Loss 0.268742 Objective Loss 0.268742 LR 0.001000 Time 0.021836 +2023-10-02 21:14:20,422 - Epoch: [94][ 930/ 1236] Overall Loss 0.268557 Objective Loss 0.268557 LR 0.001000 Time 0.021823 +2023-10-02 21:14:20,632 - Epoch: [94][ 940/ 1236] Overall Loss 0.268406 Objective Loss 0.268406 LR 0.001000 Time 0.021814 +2023-10-02 21:14:20,839 - Epoch: [94][ 950/ 1236] Overall Loss 0.268267 Objective Loss 0.268267 LR 0.001000 Time 0.021802 +2023-10-02 21:14:21,049 - Epoch: [94][ 960/ 1236] Overall Loss 0.268279 Objective Loss 0.268279 LR 0.001000 Time 0.021794 +2023-10-02 21:14:21,256 - Epoch: [94][ 970/ 1236] Overall Loss 0.268547 Objective Loss 0.268547 LR 0.001000 Time 0.021782 +2023-10-02 21:14:21,465 - Epoch: [94][ 980/ 1236] Overall Loss 0.268547 Objective Loss 0.268547 LR 0.001000 Time 0.021773 +2023-10-02 21:14:21,673 - Epoch: [94][ 990/ 1236] Overall Loss 0.268536 Objective Loss 0.268536 LR 0.001000 Time 0.021761 +2023-10-02 21:14:21,882 - Epoch: [94][ 1000/ 1236] Overall Loss 0.268742 Objective Loss 0.268742 LR 0.001000 Time 0.021753 +2023-10-02 21:14:22,090 - Epoch: [94][ 1010/ 1236] Overall Loss 0.268860 Objective Loss 0.268860 LR 0.001000 Time 0.021741 +2023-10-02 21:14:22,300 - Epoch: [94][ 1020/ 1236] Overall Loss 0.268842 Objective Loss 0.268842 LR 0.001000 Time 0.021734 +2023-10-02 21:14:22,507 - Epoch: [94][ 1030/ 1236] Overall Loss 0.268800 Objective Loss 0.268800 LR 0.001000 Time 0.021723 +2023-10-02 21:14:22,717 - Epoch: [94][ 1040/ 1236] Overall Loss 0.268891 Objective Loss 0.268891 LR 0.001000 Time 0.021716 +2023-10-02 21:14:22,924 - Epoch: [94][ 1050/ 1236] Overall Loss 0.268801 Objective Loss 0.268801 LR 0.001000 Time 0.021706 +2023-10-02 21:14:23,134 - Epoch: [94][ 1060/ 1236] Overall Loss 0.268713 Objective Loss 0.268713 LR 0.001000 Time 0.021699 +2023-10-02 21:14:23,341 - Epoch: [94][ 1070/ 1236] Overall Loss 0.268732 Objective Loss 0.268732 LR 0.001000 Time 0.021689 +2023-10-02 21:14:23,550 - Epoch: [94][ 1080/ 1236] Overall Loss 0.268519 Objective Loss 0.268519 LR 0.001000 Time 0.021682 +2023-10-02 21:14:23,757 - Epoch: [94][ 1090/ 1236] Overall Loss 0.268981 Objective Loss 0.268981 LR 0.001000 Time 0.021672 +2023-10-02 21:14:23,967 - Epoch: [94][ 1100/ 1236] Overall Loss 0.268974 Objective Loss 0.268974 LR 0.001000 Time 0.021665 +2023-10-02 21:14:24,174 - Epoch: [94][ 1110/ 1236] Overall Loss 0.268948 Objective Loss 0.268948 LR 0.001000 Time 0.021656 +2023-10-02 21:14:24,385 - Epoch: [94][ 1120/ 1236] Overall Loss 0.269044 Objective Loss 0.269044 LR 0.001000 Time 0.021650 +2023-10-02 21:14:24,591 - Epoch: [94][ 1130/ 1236] Overall Loss 0.269169 Objective Loss 0.269169 LR 0.001000 Time 0.021641 +2023-10-02 21:14:24,800 - Epoch: [94][ 1140/ 1236] Overall Loss 0.269050 Objective Loss 0.269050 LR 0.001000 Time 0.021634 +2023-10-02 21:14:25,008 - Epoch: [94][ 1150/ 1236] Overall Loss 0.268956 Objective Loss 0.268956 LR 0.001000 Time 0.021625 +2023-10-02 21:14:25,217 - Epoch: [94][ 1160/ 1236] Overall Loss 0.269056 Objective Loss 0.269056 LR 0.001000 Time 0.021619 +2023-10-02 21:14:25,425 - Epoch: [94][ 1170/ 1236] Overall Loss 0.268924 Objective Loss 0.268924 LR 0.001000 Time 0.021611 +2023-10-02 21:14:25,634 - Epoch: [94][ 1180/ 1236] Overall Loss 0.268776 Objective Loss 0.268776 LR 0.001000 Time 0.021604 +2023-10-02 21:14:25,842 - Epoch: [94][ 1190/ 1236] Overall Loss 0.268916 Objective Loss 0.268916 LR 0.001000 Time 0.021596 +2023-10-02 21:14:26,052 - Epoch: [94][ 1200/ 1236] Overall Loss 0.268956 Objective Loss 0.268956 LR 0.001000 Time 0.021591 +2023-10-02 21:14:26,259 - Epoch: [94][ 1210/ 1236] Overall Loss 0.268780 Objective Loss 0.268780 LR 0.001000 Time 0.021583 +2023-10-02 21:14:26,469 - Epoch: [94][ 1220/ 1236] Overall Loss 0.268839 Objective Loss 0.268839 LR 0.001000 Time 0.021579 +2023-10-02 21:14:26,730 - Epoch: [94][ 1230/ 1236] Overall Loss 0.268767 Objective Loss 0.268767 LR 0.001000 Time 0.021615 +2023-10-02 21:14:26,852 - Epoch: [94][ 1236/ 1236] Overall Loss 0.268627 Objective Loss 0.268627 Top1 85.336049 Top5 97.556008 LR 0.001000 Time 0.021608 +2023-10-02 21:14:26,994 - --- validate (epoch=94)----------- +2023-10-02 21:14:26,995 - 29943 samples (256 per mini-batch) +2023-10-02 21:14:27,484 - Epoch: [94][ 10/ 117] Loss 0.338961 Top1 83.554688 Top5 98.007812 +2023-10-02 21:14:27,640 - Epoch: [94][ 20/ 117] Loss 0.332653 Top1 83.378906 Top5 98.066406 +2023-10-02 21:14:27,792 - Epoch: [94][ 30/ 117] Loss 0.337486 Top1 83.072917 Top5 97.981771 +2023-10-02 21:14:27,948 - Epoch: [94][ 40/ 117] Loss 0.342231 Top1 83.125000 Top5 97.929688 +2023-10-02 21:14:28,099 - Epoch: [94][ 50/ 117] Loss 0.346230 Top1 83.242188 Top5 97.867188 +2023-10-02 21:14:28,253 - Epoch: [94][ 60/ 117] Loss 0.344627 Top1 83.300781 Top5 97.981771 +2023-10-02 21:14:28,404 - Epoch: [94][ 70/ 117] Loss 0.346614 Top1 83.214286 Top5 97.912946 +2023-10-02 21:14:28,560 - Epoch: [94][ 80/ 117] Loss 0.347781 Top1 83.154297 Top5 97.915039 +2023-10-02 21:14:28,712 - Epoch: [94][ 90/ 117] Loss 0.347631 Top1 83.216146 Top5 97.934028 +2023-10-02 21:14:28,865 - Epoch: [94][ 100/ 117] Loss 0.345588 Top1 83.273438 Top5 97.957031 +2023-10-02 21:14:29,025 - Epoch: [94][ 110/ 117] Loss 0.345147 Top1 83.284801 Top5 97.954545 +2023-10-02 21:14:29,114 - Epoch: [94][ 117/ 117] Loss 0.341307 Top1 83.311625 Top5 97.992853 +2023-10-02 21:14:29,246 - ==> Top1: 83.312 Top5: 97.993 Loss: 0.341 + +2023-10-02 21:14:29,247 - ==> Confusion: +[[ 921 1 4 1 4 1 0 1 10 73 1 4 1 3 8 0 4 1 1 0 11] + [ 0 1053 1 2 4 19 1 21 3 1 2 1 0 0 0 2 1 0 13 2 5] + [ 3 1 950 18 4 0 20 11 0 3 3 2 5 4 2 3 1 1 11 3 11] + [ 1 5 12 979 0 3 2 7 10 1 7 0 6 4 20 1 1 5 10 1 14] + [ 28 13 0 1 946 4 1 0 1 13 2 1 0 2 6 10 12 1 1 1 7] + [ 4 57 1 1 5 944 0 35 4 4 7 6 7 17 3 0 0 0 4 5 12] + [ 0 5 26 0 0 0 1128 8 0 0 5 2 0 1 1 4 0 1 1 5 4] + [ 7 9 18 1 3 22 4 1074 2 4 1 11 1 2 2 2 1 0 34 10 10] + [ 20 2 0 0 2 4 0 4 961 44 11 1 4 15 11 2 2 0 3 0 3] + [ 85 2 3 3 6 4 0 1 24 948 3 1 0 19 11 2 1 0 0 2 4] + [ 3 1 6 9 1 1 2 2 18 1 969 2 0 9 8 0 0 2 5 1 13] + [ 0 0 2 0 1 9 0 6 0 0 0 930 47 7 0 2 1 11 0 12 7] + [ 0 0 2 8 0 1 3 1 2 0 5 46 957 4 2 10 0 8 3 4 12] + [ 1 0 2 0 1 6 0 0 11 16 11 5 0 1049 5 1 3 0 1 1 6] + [ 11 1 3 26 4 0 1 0 26 4 4 0 6 3 985 0 1 3 12 0 11] + [ 0 0 2 1 1 1 0 1 0 0 0 9 9 0 0 1070 19 10 2 5 4] + [ 1 27 1 0 2 8 0 0 1 0 1 6 2 2 2 12 1082 0 1 6 7] + [ 0 1 0 6 0 0 2 1 1 1 0 8 27 1 2 9 1 974 0 1 3] + [ 1 3 5 15 1 1 1 19 5 1 5 1 5 0 8 0 0 0 987 1 9] + [ 0 5 3 3 1 2 12 13 0 0 2 24 2 2 0 5 5 1 1 1061 10] + [ 128 227 131 120 76 150 45 141 112 70 183 122 337 305 136 72 140 64 168 200 4978]] + +2023-10-02 21:14:29,248 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:14:29,248 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:14:29,254 - + +2023-10-02 21:14:29,254 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:14:30,268 - Epoch: [95][ 10/ 1236] Overall Loss 0.248484 Objective Loss 0.248484 LR 0.001000 Time 0.101331 +2023-10-02 21:14:30,478 - Epoch: [95][ 20/ 1236] Overall Loss 0.251445 Objective Loss 0.251445 LR 0.001000 Time 0.061123 +2023-10-02 21:14:30,684 - Epoch: [95][ 30/ 1236] Overall Loss 0.249692 Objective Loss 0.249692 LR 0.001000 Time 0.047632 +2023-10-02 21:14:30,894 - Epoch: [95][ 40/ 1236] Overall Loss 0.254576 Objective Loss 0.254576 LR 0.001000 Time 0.040966 +2023-10-02 21:14:31,100 - Epoch: [95][ 50/ 1236] Overall Loss 0.255997 Objective Loss 0.255997 LR 0.001000 Time 0.036885 +2023-10-02 21:14:31,310 - Epoch: [95][ 60/ 1236] Overall Loss 0.253929 Objective Loss 0.253929 LR 0.001000 Time 0.034227 +2023-10-02 21:14:31,516 - Epoch: [95][ 70/ 1236] Overall Loss 0.254790 Objective Loss 0.254790 LR 0.001000 Time 0.032278 +2023-10-02 21:14:31,726 - Epoch: [95][ 80/ 1236] Overall Loss 0.252175 Objective Loss 0.252175 LR 0.001000 Time 0.030862 +2023-10-02 21:14:31,932 - Epoch: [95][ 90/ 1236] Overall Loss 0.250446 Objective Loss 0.250446 LR 0.001000 Time 0.029719 +2023-10-02 21:14:32,141 - Epoch: [95][ 100/ 1236] Overall Loss 0.247221 Objective Loss 0.247221 LR 0.001000 Time 0.028833 +2023-10-02 21:14:32,347 - Epoch: [95][ 110/ 1236] Overall Loss 0.251961 Objective Loss 0.251961 LR 0.001000 Time 0.028087 +2023-10-02 21:14:32,555 - Epoch: [95][ 120/ 1236] Overall Loss 0.252310 Objective Loss 0.252310 LR 0.001000 Time 0.027478 +2023-10-02 21:14:32,762 - Epoch: [95][ 130/ 1236] Overall Loss 0.253162 Objective Loss 0.253162 LR 0.001000 Time 0.026942 +2023-10-02 21:14:32,970 - Epoch: [95][ 140/ 1236] Overall Loss 0.255212 Objective Loss 0.255212 LR 0.001000 Time 0.026505 +2023-10-02 21:14:33,177 - Epoch: [95][ 150/ 1236] Overall Loss 0.255302 Objective Loss 0.255302 LR 0.001000 Time 0.026106 +2023-10-02 21:14:33,389 - Epoch: [95][ 160/ 1236] Overall Loss 0.256078 Objective Loss 0.256078 LR 0.001000 Time 0.025794 +2023-10-02 21:14:33,596 - Epoch: [95][ 170/ 1236] Overall Loss 0.255460 Objective Loss 0.255460 LR 0.001000 Time 0.025496 +2023-10-02 21:14:33,808 - Epoch: [95][ 180/ 1236] Overall Loss 0.256053 Objective Loss 0.256053 LR 0.001000 Time 0.025254 +2023-10-02 21:14:34,015 - Epoch: [95][ 190/ 1236] Overall Loss 0.256877 Objective Loss 0.256877 LR 0.001000 Time 0.025017 +2023-10-02 21:14:34,227 - Epoch: [95][ 200/ 1236] Overall Loss 0.256887 Objective Loss 0.256887 LR 0.001000 Time 0.024824 +2023-10-02 21:14:34,435 - Epoch: [95][ 210/ 1236] Overall Loss 0.258191 Objective Loss 0.258191 LR 0.001000 Time 0.024629 +2023-10-02 21:14:34,647 - Epoch: [95][ 220/ 1236] Overall Loss 0.258390 Objective Loss 0.258390 LR 0.001000 Time 0.024470 +2023-10-02 21:14:34,854 - Epoch: [95][ 230/ 1236] Overall Loss 0.257648 Objective Loss 0.257648 LR 0.001000 Time 0.024308 +2023-10-02 21:14:35,066 - Epoch: [95][ 240/ 1236] Overall Loss 0.257942 Objective Loss 0.257942 LR 0.001000 Time 0.024177 +2023-10-02 21:14:35,274 - Epoch: [95][ 250/ 1236] Overall Loss 0.257725 Objective Loss 0.257725 LR 0.001000 Time 0.024040 +2023-10-02 21:14:35,486 - Epoch: [95][ 260/ 1236] Overall Loss 0.258681 Objective Loss 0.258681 LR 0.001000 Time 0.023928 +2023-10-02 21:14:35,693 - Epoch: [95][ 270/ 1236] Overall Loss 0.259493 Objective Loss 0.259493 LR 0.001000 Time 0.023809 +2023-10-02 21:14:35,905 - Epoch: [95][ 280/ 1236] Overall Loss 0.259702 Objective Loss 0.259702 LR 0.001000 Time 0.023714 +2023-10-02 21:14:36,112 - Epoch: [95][ 290/ 1236] Overall Loss 0.259564 Objective Loss 0.259564 LR 0.001000 Time 0.023611 +2023-10-02 21:14:36,324 - Epoch: [95][ 300/ 1236] Overall Loss 0.260405 Objective Loss 0.260405 LR 0.001000 Time 0.023529 +2023-10-02 21:14:36,532 - Epoch: [95][ 310/ 1236] Overall Loss 0.262378 Objective Loss 0.262378 LR 0.001000 Time 0.023439 +2023-10-02 21:14:36,742 - Epoch: [95][ 320/ 1236] Overall Loss 0.262015 Objective Loss 0.262015 LR 0.001000 Time 0.023364 +2023-10-02 21:14:36,952 - Epoch: [95][ 330/ 1236] Overall Loss 0.262972 Objective Loss 0.262972 LR 0.001000 Time 0.023285 +2023-10-02 21:14:37,164 - Epoch: [95][ 340/ 1236] Overall Loss 0.262979 Objective Loss 0.262979 LR 0.001000 Time 0.023222 +2023-10-02 21:14:37,371 - Epoch: [95][ 350/ 1236] Overall Loss 0.262953 Objective Loss 0.262953 LR 0.001000 Time 0.023150 +2023-10-02 21:14:37,581 - Epoch: [95][ 360/ 1236] Overall Loss 0.263970 Objective Loss 0.263970 LR 0.001000 Time 0.023090 +2023-10-02 21:14:37,788 - Epoch: [95][ 370/ 1236] Overall Loss 0.264668 Objective Loss 0.264668 LR 0.001000 Time 0.023023 +2023-10-02 21:14:37,997 - Epoch: [95][ 380/ 1236] Overall Loss 0.264417 Objective Loss 0.264417 LR 0.001000 Time 0.022967 +2023-10-02 21:14:38,203 - Epoch: [95][ 390/ 1236] Overall Loss 0.264298 Objective Loss 0.264298 LR 0.001000 Time 0.022906 +2023-10-02 21:14:38,413 - Epoch: [95][ 400/ 1236] Overall Loss 0.264888 Objective Loss 0.264888 LR 0.001000 Time 0.022859 +2023-10-02 21:14:38,620 - Epoch: [95][ 410/ 1236] Overall Loss 0.264717 Objective Loss 0.264717 LR 0.001000 Time 0.022804 +2023-10-02 21:14:38,829 - Epoch: [95][ 420/ 1236] Overall Loss 0.264484 Objective Loss 0.264484 LR 0.001000 Time 0.022759 +2023-10-02 21:14:39,035 - Epoch: [95][ 430/ 1236] Overall Loss 0.264242 Objective Loss 0.264242 LR 0.001000 Time 0.022708 +2023-10-02 21:14:39,244 - Epoch: [95][ 440/ 1236] Overall Loss 0.264311 Objective Loss 0.264311 LR 0.001000 Time 0.022666 +2023-10-02 21:14:39,451 - Epoch: [95][ 450/ 1236] Overall Loss 0.264697 Objective Loss 0.264697 LR 0.001000 Time 0.022620 +2023-10-02 21:14:39,661 - Epoch: [95][ 460/ 1236] Overall Loss 0.265146 Objective Loss 0.265146 LR 0.001000 Time 0.022584 +2023-10-02 21:14:39,868 - Epoch: [95][ 470/ 1236] Overall Loss 0.264607 Objective Loss 0.264607 LR 0.001000 Time 0.022542 +2023-10-02 21:14:40,078 - Epoch: [95][ 480/ 1236] Overall Loss 0.265119 Objective Loss 0.265119 LR 0.001000 Time 0.022510 +2023-10-02 21:14:40,284 - Epoch: [95][ 490/ 1236] Overall Loss 0.264941 Objective Loss 0.264941 LR 0.001000 Time 0.022471 +2023-10-02 21:14:40,494 - Epoch: [95][ 500/ 1236] Overall Loss 0.264159 Objective Loss 0.264159 LR 0.001000 Time 0.022441 +2023-10-02 21:14:40,699 - Epoch: [95][ 510/ 1236] Overall Loss 0.264071 Objective Loss 0.264071 LR 0.001000 Time 0.022402 +2023-10-02 21:14:40,910 - Epoch: [95][ 520/ 1236] Overall Loss 0.264026 Objective Loss 0.264026 LR 0.001000 Time 0.022375 +2023-10-02 21:14:41,116 - Epoch: [95][ 530/ 1236] Overall Loss 0.263789 Objective Loss 0.263789 LR 0.001000 Time 0.022341 +2023-10-02 21:14:41,325 - Epoch: [95][ 540/ 1236] Overall Loss 0.264489 Objective Loss 0.264489 LR 0.001000 Time 0.022316 +2023-10-02 21:14:41,532 - Epoch: [95][ 550/ 1236] Overall Loss 0.265005 Objective Loss 0.265005 LR 0.001000 Time 0.022284 +2023-10-02 21:14:41,742 - Epoch: [95][ 560/ 1236] Overall Loss 0.265018 Objective Loss 0.265018 LR 0.001000 Time 0.022261 +2023-10-02 21:14:41,952 - Epoch: [95][ 570/ 1236] Overall Loss 0.264741 Objective Loss 0.264741 LR 0.001000 Time 0.022237 +2023-10-02 21:14:42,163 - Epoch: [95][ 580/ 1236] Overall Loss 0.265480 Objective Loss 0.265480 LR 0.001000 Time 0.022215 +2023-10-02 21:14:42,373 - Epoch: [95][ 590/ 1236] Overall Loss 0.265363 Objective Loss 0.265363 LR 0.001000 Time 0.022195 +2023-10-02 21:14:42,585 - Epoch: [95][ 600/ 1236] Overall Loss 0.265184 Objective Loss 0.265184 LR 0.001000 Time 0.022177 +2023-10-02 21:14:42,794 - Epoch: [95][ 610/ 1236] Overall Loss 0.265073 Objective Loss 0.265073 LR 0.001000 Time 0.022156 +2023-10-02 21:14:43,007 - Epoch: [95][ 620/ 1236] Overall Loss 0.265223 Objective Loss 0.265223 LR 0.001000 Time 0.022141 +2023-10-02 21:14:43,216 - Epoch: [95][ 630/ 1236] Overall Loss 0.265323 Objective Loss 0.265323 LR 0.001000 Time 0.022122 +2023-10-02 21:14:43,430 - Epoch: [95][ 640/ 1236] Overall Loss 0.264888 Objective Loss 0.264888 LR 0.001000 Time 0.022110 +2023-10-02 21:14:43,640 - Epoch: [95][ 650/ 1236] Overall Loss 0.265073 Objective Loss 0.265073 LR 0.001000 Time 0.022092 +2023-10-02 21:14:43,855 - Epoch: [95][ 660/ 1236] Overall Loss 0.264850 Objective Loss 0.264850 LR 0.001000 Time 0.022083 +2023-10-02 21:14:44,066 - Epoch: [95][ 670/ 1236] Overall Loss 0.264694 Objective Loss 0.264694 LR 0.001000 Time 0.022068 +2023-10-02 21:14:44,276 - Epoch: [95][ 680/ 1236] Overall Loss 0.265340 Objective Loss 0.265340 LR 0.001000 Time 0.022052 +2023-10-02 21:14:44,486 - Epoch: [95][ 690/ 1236] Overall Loss 0.265945 Objective Loss 0.265945 LR 0.001000 Time 0.022034 +2023-10-02 21:14:44,695 - Epoch: [95][ 700/ 1236] Overall Loss 0.266195 Objective Loss 0.266195 LR 0.001000 Time 0.022017 +2023-10-02 21:14:44,905 - Epoch: [95][ 710/ 1236] Overall Loss 0.266323 Objective Loss 0.266323 LR 0.001000 Time 0.022001 +2023-10-02 21:14:45,115 - Epoch: [95][ 720/ 1236] Overall Loss 0.266763 Objective Loss 0.266763 LR 0.001000 Time 0.021986 +2023-10-02 21:14:45,325 - Epoch: [95][ 730/ 1236] Overall Loss 0.266903 Objective Loss 0.266903 LR 0.001000 Time 0.021971 +2023-10-02 21:14:45,535 - Epoch: [95][ 740/ 1236] Overall Loss 0.267134 Objective Loss 0.267134 LR 0.001000 Time 0.021957 +2023-10-02 21:14:45,745 - Epoch: [95][ 750/ 1236] Overall Loss 0.267278 Objective Loss 0.267278 LR 0.001000 Time 0.021942 +2023-10-02 21:14:45,955 - Epoch: [95][ 760/ 1236] Overall Loss 0.267575 Objective Loss 0.267575 LR 0.001000 Time 0.021929 +2023-10-02 21:14:46,164 - Epoch: [95][ 770/ 1236] Overall Loss 0.268025 Objective Loss 0.268025 LR 0.001000 Time 0.021914 +2023-10-02 21:14:46,374 - Epoch: [95][ 780/ 1236] Overall Loss 0.268209 Objective Loss 0.268209 LR 0.001000 Time 0.021902 +2023-10-02 21:14:46,584 - Epoch: [95][ 790/ 1236] Overall Loss 0.268048 Objective Loss 0.268048 LR 0.001000 Time 0.021889 +2023-10-02 21:14:46,797 - Epoch: [95][ 800/ 1236] Overall Loss 0.268136 Objective Loss 0.268136 LR 0.001000 Time 0.021881 +2023-10-02 21:14:47,006 - Epoch: [95][ 810/ 1236] Overall Loss 0.267838 Objective Loss 0.267838 LR 0.001000 Time 0.021868 +2023-10-02 21:14:47,216 - Epoch: [95][ 820/ 1236] Overall Loss 0.268188 Objective Loss 0.268188 LR 0.001000 Time 0.021857 +2023-10-02 21:14:47,426 - Epoch: [95][ 830/ 1236] Overall Loss 0.268123 Objective Loss 0.268123 LR 0.001000 Time 0.021845 +2023-10-02 21:14:47,636 - Epoch: [95][ 840/ 1236] Overall Loss 0.268609 Objective Loss 0.268609 LR 0.001000 Time 0.021834 +2023-10-02 21:14:47,846 - Epoch: [95][ 850/ 1236] Overall Loss 0.268739 Objective Loss 0.268739 LR 0.001000 Time 0.021823 +2023-10-02 21:14:48,056 - Epoch: [95][ 860/ 1236] Overall Loss 0.269105 Objective Loss 0.269105 LR 0.001000 Time 0.021813 +2023-10-02 21:14:48,266 - Epoch: [95][ 870/ 1236] Overall Loss 0.269033 Objective Loss 0.269033 LR 0.001000 Time 0.021802 +2023-10-02 21:14:48,476 - Epoch: [95][ 880/ 1236] Overall Loss 0.269022 Objective Loss 0.269022 LR 0.001000 Time 0.021792 +2023-10-02 21:14:48,686 - Epoch: [95][ 890/ 1236] Overall Loss 0.269117 Objective Loss 0.269117 LR 0.001000 Time 0.021782 +2023-10-02 21:14:48,896 - Epoch: [95][ 900/ 1236] Overall Loss 0.269045 Objective Loss 0.269045 LR 0.001000 Time 0.021773 +2023-10-02 21:14:49,106 - Epoch: [95][ 910/ 1236] Overall Loss 0.269269 Objective Loss 0.269269 LR 0.001000 Time 0.021763 +2023-10-02 21:14:49,316 - Epoch: [95][ 920/ 1236] Overall Loss 0.269219 Objective Loss 0.269219 LR 0.001000 Time 0.021754 +2023-10-02 21:14:49,526 - Epoch: [95][ 930/ 1236] Overall Loss 0.268967 Objective Loss 0.268967 LR 0.001000 Time 0.021744 +2023-10-02 21:14:49,736 - Epoch: [95][ 940/ 1236] Overall Loss 0.269035 Objective Loss 0.269035 LR 0.001000 Time 0.021736 +2023-10-02 21:14:49,946 - Epoch: [95][ 950/ 1236] Overall Loss 0.269189 Objective Loss 0.269189 LR 0.001000 Time 0.021726 +2023-10-02 21:14:50,156 - Epoch: [95][ 960/ 1236] Overall Loss 0.269601 Objective Loss 0.269601 LR 0.001000 Time 0.021718 +2023-10-02 21:14:50,366 - Epoch: [95][ 970/ 1236] Overall Loss 0.269485 Objective Loss 0.269485 LR 0.001000 Time 0.021709 +2023-10-02 21:14:50,576 - Epoch: [95][ 980/ 1236] Overall Loss 0.269553 Objective Loss 0.269553 LR 0.001000 Time 0.021702 +2023-10-02 21:14:50,786 - Epoch: [95][ 990/ 1236] Overall Loss 0.269552 Objective Loss 0.269552 LR 0.001000 Time 0.021693 +2023-10-02 21:14:50,996 - Epoch: [95][ 1000/ 1236] Overall Loss 0.269819 Objective Loss 0.269819 LR 0.001000 Time 0.021686 +2023-10-02 21:14:51,206 - Epoch: [95][ 1010/ 1236] Overall Loss 0.269882 Objective Loss 0.269882 LR 0.001000 Time 0.021678 +2023-10-02 21:14:51,416 - Epoch: [95][ 1020/ 1236] Overall Loss 0.269576 Objective Loss 0.269576 LR 0.001000 Time 0.021671 +2023-10-02 21:14:51,625 - Epoch: [95][ 1030/ 1236] Overall Loss 0.269498 Objective Loss 0.269498 LR 0.001000 Time 0.021662 +2023-10-02 21:14:51,835 - Epoch: [95][ 1040/ 1236] Overall Loss 0.269345 Objective Loss 0.269345 LR 0.001000 Time 0.021655 +2023-10-02 21:14:52,044 - Epoch: [95][ 1050/ 1236] Overall Loss 0.269429 Objective Loss 0.269429 LR 0.001000 Time 0.021646 +2023-10-02 21:14:52,254 - Epoch: [95][ 1060/ 1236] Overall Loss 0.269234 Objective Loss 0.269234 LR 0.001000 Time 0.021640 +2023-10-02 21:14:52,464 - Epoch: [95][ 1070/ 1236] Overall Loss 0.269693 Objective Loss 0.269693 LR 0.001000 Time 0.021632 +2023-10-02 21:14:52,674 - Epoch: [95][ 1080/ 1236] Overall Loss 0.269841 Objective Loss 0.269841 LR 0.001000 Time 0.021626 +2023-10-02 21:14:52,884 - Epoch: [95][ 1090/ 1236] Overall Loss 0.269971 Objective Loss 0.269971 LR 0.001000 Time 0.021619 +2023-10-02 21:14:53,094 - Epoch: [95][ 1100/ 1236] Overall Loss 0.270137 Objective Loss 0.270137 LR 0.001000 Time 0.021613 +2023-10-02 21:14:53,304 - Epoch: [95][ 1110/ 1236] Overall Loss 0.270127 Objective Loss 0.270127 LR 0.001000 Time 0.021606 +2023-10-02 21:14:53,514 - Epoch: [95][ 1120/ 1236] Overall Loss 0.270054 Objective Loss 0.270054 LR 0.001000 Time 0.021600 +2023-10-02 21:14:53,724 - Epoch: [95][ 1130/ 1236] Overall Loss 0.269881 Objective Loss 0.269881 LR 0.001000 Time 0.021594 +2023-10-02 21:14:53,934 - Epoch: [95][ 1140/ 1236] Overall Loss 0.269919 Objective Loss 0.269919 LR 0.001000 Time 0.021588 +2023-10-02 21:14:54,144 - Epoch: [95][ 1150/ 1236] Overall Loss 0.270015 Objective Loss 0.270015 LR 0.001000 Time 0.021582 +2023-10-02 21:14:54,354 - Epoch: [95][ 1160/ 1236] Overall Loss 0.270325 Objective Loss 0.270325 LR 0.001000 Time 0.021576 +2023-10-02 21:14:54,564 - Epoch: [95][ 1170/ 1236] Overall Loss 0.270358 Objective Loss 0.270358 LR 0.001000 Time 0.021570 +2023-10-02 21:14:54,774 - Epoch: [95][ 1180/ 1236] Overall Loss 0.270301 Objective Loss 0.270301 LR 0.001000 Time 0.021565 +2023-10-02 21:14:54,984 - Epoch: [95][ 1190/ 1236] Overall Loss 0.270372 Objective Loss 0.270372 LR 0.001000 Time 0.021559 +2023-10-02 21:14:55,194 - Epoch: [95][ 1200/ 1236] Overall Loss 0.270190 Objective Loss 0.270190 LR 0.001000 Time 0.021554 +2023-10-02 21:14:55,404 - Epoch: [95][ 1210/ 1236] Overall Loss 0.270171 Objective Loss 0.270171 LR 0.001000 Time 0.021548 +2023-10-02 21:14:55,614 - Epoch: [95][ 1220/ 1236] Overall Loss 0.270437 Objective Loss 0.270437 LR 0.001000 Time 0.021543 +2023-10-02 21:14:55,877 - Epoch: [95][ 1230/ 1236] Overall Loss 0.270463 Objective Loss 0.270463 LR 0.001000 Time 0.021581 +2023-10-02 21:14:55,999 - Epoch: [95][ 1236/ 1236] Overall Loss 0.270416 Objective Loss 0.270416 Top1 86.354379 Top5 99.185336 LR 0.001000 Time 0.021575 +2023-10-02 21:14:56,139 - --- validate (epoch=95)----------- +2023-10-02 21:14:56,139 - 29943 samples (256 per mini-batch) +2023-10-02 21:14:56,632 - Epoch: [95][ 10/ 117] Loss 0.369026 Top1 82.890625 Top5 97.968750 +2023-10-02 21:14:56,785 - Epoch: [95][ 20/ 117] Loss 0.372887 Top1 83.046875 Top5 97.968750 +2023-10-02 21:14:56,937 - Epoch: [95][ 30/ 117] Loss 0.350614 Top1 83.346354 Top5 98.020833 +2023-10-02 21:14:57,089 - Epoch: [95][ 40/ 117] Loss 0.347896 Top1 83.261719 Top5 98.066406 +2023-10-02 21:14:57,240 - Epoch: [95][ 50/ 117] Loss 0.348020 Top1 83.226562 Top5 98.062500 +2023-10-02 21:14:57,390 - Epoch: [95][ 60/ 117] Loss 0.347203 Top1 83.281250 Top5 98.131510 +2023-10-02 21:14:57,541 - Epoch: [95][ 70/ 117] Loss 0.342331 Top1 83.314732 Top5 98.219866 +2023-10-02 21:14:57,694 - Epoch: [95][ 80/ 117] Loss 0.344070 Top1 83.374023 Top5 98.149414 +2023-10-02 21:14:57,846 - Epoch: [95][ 90/ 117] Loss 0.342843 Top1 83.506944 Top5 98.103299 +2023-10-02 21:14:57,998 - Epoch: [95][ 100/ 117] Loss 0.342272 Top1 83.523438 Top5 98.144531 +2023-10-02 21:14:58,157 - Epoch: [95][ 110/ 117] Loss 0.342591 Top1 83.504972 Top5 98.107244 +2023-10-02 21:14:58,245 - Epoch: [95][ 117/ 117] Loss 0.343017 Top1 83.578800 Top5 98.133120 +2023-10-02 21:14:58,347 - ==> Top1: 83.579 Top5: 98.133 Loss: 0.343 + +2023-10-02 21:14:58,348 - ==> Confusion: +[[ 892 2 4 0 7 4 0 2 8 91 2 2 1 4 7 0 8 0 0 0 16] + [ 0 1023 2 0 4 29 2 36 2 2 0 3 1 1 1 2 0 0 14 2 7] + [ 4 1 954 10 2 0 32 9 1 2 3 2 7 1 1 4 1 3 10 0 9] + [ 0 4 21 964 0 6 2 2 7 2 10 1 10 3 18 3 0 9 11 0 16] + [ 17 7 5 0 950 9 1 0 1 13 3 0 3 3 10 6 12 1 1 1 7] + [ 3 46 0 3 3 961 2 30 2 7 2 6 6 12 7 0 3 0 5 6 12] + [ 0 3 15 0 0 2 1132 7 0 1 6 2 0 0 0 3 1 0 4 10 5] + [ 1 8 25 0 2 22 12 1055 1 4 2 13 4 1 2 1 2 1 45 9 8] + [ 14 2 0 1 2 3 0 1 954 50 17 1 3 11 14 1 4 1 4 1 5] + [ 75 0 0 0 1 1 1 1 39 957 2 1 0 22 4 3 0 0 0 4 8] + [ 2 4 12 8 0 2 5 4 15 0 956 3 0 16 3 1 5 1 5 0 11] + [ 0 1 0 0 0 16 1 5 0 0 0 941 32 6 0 2 1 17 0 5 8] + [ 0 2 1 5 0 0 3 2 0 1 2 24 984 2 1 8 1 14 3 3 12] + [ 1 0 2 0 3 9 1 0 11 9 7 9 0 1047 3 1 2 1 0 1 12] + [ 9 0 7 24 1 1 0 0 26 5 4 0 3 5 985 0 2 5 15 0 9] + [ 2 0 2 0 5 0 2 0 0 0 0 6 8 0 0 1068 15 15 2 2 7] + [ 1 16 1 0 5 8 0 1 0 0 0 10 2 2 3 6 1092 0 1 5 8] + [ 0 0 0 5 0 0 2 0 1 1 0 6 20 0 2 1 0 995 1 0 4] + [ 3 5 3 16 0 1 0 24 4 1 5 0 5 0 15 0 1 0 974 0 11] + [ 0 1 5 3 1 5 14 16 0 0 5 18 6 1 1 5 6 0 3 1045 17] + [ 94 158 153 87 64 164 54 118 93 103 191 147 370 309 129 61 120 82 161 150 5097]] + +2023-10-02 21:14:58,349 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:14:58,349 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:14:58,355 - + +2023-10-02 21:14:58,355 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:14:59,382 - Epoch: [96][ 10/ 1236] Overall Loss 0.249029 Objective Loss 0.249029 LR 0.001000 Time 0.102682 +2023-10-02 21:14:59,591 - Epoch: [96][ 20/ 1236] Overall Loss 0.237892 Objective Loss 0.237892 LR 0.001000 Time 0.061744 +2023-10-02 21:14:59,799 - Epoch: [96][ 30/ 1236] Overall Loss 0.247540 Objective Loss 0.247540 LR 0.001000 Time 0.048094 +2023-10-02 21:15:00,008 - Epoch: [96][ 40/ 1236] Overall Loss 0.255269 Objective Loss 0.255269 LR 0.001000 Time 0.041291 +2023-10-02 21:15:00,216 - Epoch: [96][ 50/ 1236] Overall Loss 0.254319 Objective Loss 0.254319 LR 0.001000 Time 0.037188 +2023-10-02 21:15:00,426 - Epoch: [96][ 60/ 1236] Overall Loss 0.254992 Objective Loss 0.254992 LR 0.001000 Time 0.034489 +2023-10-02 21:15:00,633 - Epoch: [96][ 70/ 1236] Overall Loss 0.260225 Objective Loss 0.260225 LR 0.001000 Time 0.032513 +2023-10-02 21:15:00,844 - Epoch: [96][ 80/ 1236] Overall Loss 0.263363 Objective Loss 0.263363 LR 0.001000 Time 0.031076 +2023-10-02 21:15:01,051 - Epoch: [96][ 90/ 1236] Overall Loss 0.264365 Objective Loss 0.264365 LR 0.001000 Time 0.029921 +2023-10-02 21:15:01,261 - Epoch: [96][ 100/ 1236] Overall Loss 0.262994 Objective Loss 0.262994 LR 0.001000 Time 0.029029 +2023-10-02 21:15:01,468 - Epoch: [96][ 110/ 1236] Overall Loss 0.263496 Objective Loss 0.263496 LR 0.001000 Time 0.028270 +2023-10-02 21:15:01,678 - Epoch: [96][ 120/ 1236] Overall Loss 0.262898 Objective Loss 0.262898 LR 0.001000 Time 0.027663 +2023-10-02 21:15:01,885 - Epoch: [96][ 130/ 1236] Overall Loss 0.261537 Objective Loss 0.261537 LR 0.001000 Time 0.027125 +2023-10-02 21:15:02,095 - Epoch: [96][ 140/ 1236] Overall Loss 0.260142 Objective Loss 0.260142 LR 0.001000 Time 0.026687 +2023-10-02 21:15:02,302 - Epoch: [96][ 150/ 1236] Overall Loss 0.261223 Objective Loss 0.261223 LR 0.001000 Time 0.026285 +2023-10-02 21:15:02,512 - Epoch: [96][ 160/ 1236] Overall Loss 0.261146 Objective Loss 0.261146 LR 0.001000 Time 0.025954 +2023-10-02 21:15:02,720 - Epoch: [96][ 170/ 1236] Overall Loss 0.260320 Objective Loss 0.260320 LR 0.001000 Time 0.025645 +2023-10-02 21:15:02,930 - Epoch: [96][ 180/ 1236] Overall Loss 0.259693 Objective Loss 0.259693 LR 0.001000 Time 0.025385 +2023-10-02 21:15:03,141 - Epoch: [96][ 190/ 1236] Overall Loss 0.259312 Objective Loss 0.259312 LR 0.001000 Time 0.025159 +2023-10-02 21:15:03,351 - Epoch: [96][ 200/ 1236] Overall Loss 0.259096 Objective Loss 0.259096 LR 0.001000 Time 0.024952 +2023-10-02 21:15:03,557 - Epoch: [96][ 210/ 1236] Overall Loss 0.259719 Objective Loss 0.259719 LR 0.001000 Time 0.024739 +2023-10-02 21:15:03,763 - Epoch: [96][ 220/ 1236] Overall Loss 0.259514 Objective Loss 0.259514 LR 0.001000 Time 0.024554 +2023-10-02 21:15:03,970 - Epoch: [96][ 230/ 1236] Overall Loss 0.259741 Objective Loss 0.259741 LR 0.001000 Time 0.024383 +2023-10-02 21:15:04,177 - Epoch: [96][ 240/ 1236] Overall Loss 0.260153 Objective Loss 0.260153 LR 0.001000 Time 0.024230 +2023-10-02 21:15:04,383 - Epoch: [96][ 250/ 1236] Overall Loss 0.262651 Objective Loss 0.262651 LR 0.001000 Time 0.024078 +2023-10-02 21:15:04,592 - Epoch: [96][ 260/ 1236] Overall Loss 0.262780 Objective Loss 0.262780 LR 0.001000 Time 0.023952 +2023-10-02 21:15:04,799 - Epoch: [96][ 270/ 1236] Overall Loss 0.263187 Objective Loss 0.263187 LR 0.001000 Time 0.023827 +2023-10-02 21:15:05,006 - Epoch: [96][ 280/ 1236] Overall Loss 0.263809 Objective Loss 0.263809 LR 0.001000 Time 0.023717 +2023-10-02 21:15:05,213 - Epoch: [96][ 290/ 1236] Overall Loss 0.263568 Objective Loss 0.263568 LR 0.001000 Time 0.023607 +2023-10-02 21:15:05,422 - Epoch: [96][ 300/ 1236] Overall Loss 0.264624 Objective Loss 0.264624 LR 0.001000 Time 0.023515 +2023-10-02 21:15:05,632 - Epoch: [96][ 310/ 1236] Overall Loss 0.264867 Objective Loss 0.264867 LR 0.001000 Time 0.023428 +2023-10-02 21:15:05,841 - Epoch: [96][ 320/ 1236] Overall Loss 0.264099 Objective Loss 0.264099 LR 0.001000 Time 0.023349 +2023-10-02 21:15:06,051 - Epoch: [96][ 330/ 1236] Overall Loss 0.263345 Objective Loss 0.263345 LR 0.001000 Time 0.023272 +2023-10-02 21:15:06,259 - Epoch: [96][ 340/ 1236] Overall Loss 0.263655 Objective Loss 0.263655 LR 0.001000 Time 0.023199 +2023-10-02 21:15:06,468 - Epoch: [96][ 350/ 1236] Overall Loss 0.262888 Objective Loss 0.262888 LR 0.001000 Time 0.023134 +2023-10-02 21:15:06,678 - Epoch: [96][ 360/ 1236] Overall Loss 0.262731 Objective Loss 0.262731 LR 0.001000 Time 0.023072 +2023-10-02 21:15:06,888 - Epoch: [96][ 370/ 1236] Overall Loss 0.263193 Objective Loss 0.263193 LR 0.001000 Time 0.023014 +2023-10-02 21:15:07,098 - Epoch: [96][ 380/ 1236] Overall Loss 0.263909 Objective Loss 0.263909 LR 0.001000 Time 0.022959 +2023-10-02 21:15:07,308 - Epoch: [96][ 390/ 1236] Overall Loss 0.264081 Objective Loss 0.264081 LR 0.001000 Time 0.022906 +2023-10-02 21:15:07,518 - Epoch: [96][ 400/ 1236] Overall Loss 0.264552 Objective Loss 0.264552 LR 0.001000 Time 0.022857 +2023-10-02 21:15:07,729 - Epoch: [96][ 410/ 1236] Overall Loss 0.264901 Objective Loss 0.264901 LR 0.001000 Time 0.022810 +2023-10-02 21:15:07,939 - Epoch: [96][ 420/ 1236] Overall Loss 0.264971 Objective Loss 0.264971 LR 0.001000 Time 0.022766 +2023-10-02 21:15:08,149 - Epoch: [96][ 430/ 1236] Overall Loss 0.265745 Objective Loss 0.265745 LR 0.001000 Time 0.022722 +2023-10-02 21:15:08,360 - Epoch: [96][ 440/ 1236] Overall Loss 0.265391 Objective Loss 0.265391 LR 0.001000 Time 0.022685 +2023-10-02 21:15:08,571 - Epoch: [96][ 450/ 1236] Overall Loss 0.265777 Objective Loss 0.265777 LR 0.001000 Time 0.022646 +2023-10-02 21:15:08,781 - Epoch: [96][ 460/ 1236] Overall Loss 0.265748 Objective Loss 0.265748 LR 0.001000 Time 0.022609 +2023-10-02 21:15:08,992 - Epoch: [96][ 470/ 1236] Overall Loss 0.265831 Objective Loss 0.265831 LR 0.001000 Time 0.022573 +2023-10-02 21:15:09,202 - Epoch: [96][ 480/ 1236] Overall Loss 0.265486 Objective Loss 0.265486 LR 0.001000 Time 0.022540 +2023-10-02 21:15:09,413 - Epoch: [96][ 490/ 1236] Overall Loss 0.265133 Objective Loss 0.265133 LR 0.001000 Time 0.022507 +2023-10-02 21:15:09,623 - Epoch: [96][ 500/ 1236] Overall Loss 0.265432 Objective Loss 0.265432 LR 0.001000 Time 0.022476 +2023-10-02 21:15:09,833 - Epoch: [96][ 510/ 1236] Overall Loss 0.265374 Objective Loss 0.265374 LR 0.001000 Time 0.022447 +2023-10-02 21:15:10,043 - Epoch: [96][ 520/ 1236] Overall Loss 0.265523 Objective Loss 0.265523 LR 0.001000 Time 0.022419 +2023-10-02 21:15:10,254 - Epoch: [96][ 530/ 1236] Overall Loss 0.265288 Objective Loss 0.265288 LR 0.001000 Time 0.022394 +2023-10-02 21:15:10,464 - Epoch: [96][ 540/ 1236] Overall Loss 0.265372 Objective Loss 0.265372 LR 0.001000 Time 0.022367 +2023-10-02 21:15:10,675 - Epoch: [96][ 550/ 1236] Overall Loss 0.265121 Objective Loss 0.265121 LR 0.001000 Time 0.022343 +2023-10-02 21:15:10,884 - Epoch: [96][ 560/ 1236] Overall Loss 0.265384 Objective Loss 0.265384 LR 0.001000 Time 0.022317 +2023-10-02 21:15:11,094 - Epoch: [96][ 570/ 1236] Overall Loss 0.265313 Objective Loss 0.265313 LR 0.001000 Time 0.022294 +2023-10-02 21:15:11,305 - Epoch: [96][ 580/ 1236] Overall Loss 0.265213 Objective Loss 0.265213 LR 0.001000 Time 0.022272 +2023-10-02 21:15:11,515 - Epoch: [96][ 590/ 1236] Overall Loss 0.264995 Objective Loss 0.264995 LR 0.001000 Time 0.022250 +2023-10-02 21:15:11,726 - Epoch: [96][ 600/ 1236] Overall Loss 0.264964 Objective Loss 0.264964 LR 0.001000 Time 0.022230 +2023-10-02 21:15:11,937 - Epoch: [96][ 610/ 1236] Overall Loss 0.265406 Objective Loss 0.265406 LR 0.001000 Time 0.022211 +2023-10-02 21:15:12,146 - Epoch: [96][ 620/ 1236] Overall Loss 0.265193 Objective Loss 0.265193 LR 0.001000 Time 0.022190 +2023-10-02 21:15:12,357 - Epoch: [96][ 630/ 1236] Overall Loss 0.265808 Objective Loss 0.265808 LR 0.001000 Time 0.022172 +2023-10-02 21:15:12,567 - Epoch: [96][ 640/ 1236] Overall Loss 0.266045 Objective Loss 0.266045 LR 0.001000 Time 0.022153 +2023-10-02 21:15:12,778 - Epoch: [96][ 650/ 1236] Overall Loss 0.265903 Objective Loss 0.265903 LR 0.001000 Time 0.022136 +2023-10-02 21:15:12,988 - Epoch: [96][ 660/ 1236] Overall Loss 0.266349 Objective Loss 0.266349 LR 0.001000 Time 0.022118 +2023-10-02 21:15:13,199 - Epoch: [96][ 670/ 1236] Overall Loss 0.266811 Objective Loss 0.266811 LR 0.001000 Time 0.022102 +2023-10-02 21:15:13,408 - Epoch: [96][ 680/ 1236] Overall Loss 0.267002 Objective Loss 0.267002 LR 0.001000 Time 0.022085 +2023-10-02 21:15:13,619 - Epoch: [96][ 690/ 1236] Overall Loss 0.267178 Objective Loss 0.267178 LR 0.001000 Time 0.022070 +2023-10-02 21:15:13,829 - Epoch: [96][ 700/ 1236] Overall Loss 0.266991 Objective Loss 0.266991 LR 0.001000 Time 0.022055 +2023-10-02 21:15:14,040 - Epoch: [96][ 710/ 1236] Overall Loss 0.267384 Objective Loss 0.267384 LR 0.001000 Time 0.022040 +2023-10-02 21:15:14,249 - Epoch: [96][ 720/ 1236] Overall Loss 0.267373 Objective Loss 0.267373 LR 0.001000 Time 0.022025 +2023-10-02 21:15:14,460 - Epoch: [96][ 730/ 1236] Overall Loss 0.267141 Objective Loss 0.267141 LR 0.001000 Time 0.022011 +2023-10-02 21:15:14,669 - Epoch: [96][ 740/ 1236] Overall Loss 0.266819 Objective Loss 0.266819 LR 0.001000 Time 0.021996 +2023-10-02 21:15:14,877 - Epoch: [96][ 750/ 1236] Overall Loss 0.266765 Objective Loss 0.266765 LR 0.001000 Time 0.021980 +2023-10-02 21:15:15,086 - Epoch: [96][ 760/ 1236] Overall Loss 0.267022 Objective Loss 0.267022 LR 0.001000 Time 0.021965 +2023-10-02 21:15:15,294 - Epoch: [96][ 770/ 1236] Overall Loss 0.266621 Objective Loss 0.266621 LR 0.001000 Time 0.021948 +2023-10-02 21:15:15,503 - Epoch: [96][ 780/ 1236] Overall Loss 0.266442 Objective Loss 0.266442 LR 0.001000 Time 0.021934 +2023-10-02 21:15:15,710 - Epoch: [96][ 790/ 1236] Overall Loss 0.266100 Objective Loss 0.266100 LR 0.001000 Time 0.021916 +2023-10-02 21:15:15,919 - Epoch: [96][ 800/ 1236] Overall Loss 0.265907 Objective Loss 0.265907 LR 0.001000 Time 0.021904 +2023-10-02 21:15:16,128 - Epoch: [96][ 810/ 1236] Overall Loss 0.266218 Objective Loss 0.266218 LR 0.001000 Time 0.021889 +2023-10-02 21:15:16,338 - Epoch: [96][ 820/ 1236] Overall Loss 0.266058 Objective Loss 0.266058 LR 0.001000 Time 0.021877 +2023-10-02 21:15:16,546 - Epoch: [96][ 830/ 1236] Overall Loss 0.265862 Objective Loss 0.265862 LR 0.001000 Time 0.021863 +2023-10-02 21:15:16,756 - Epoch: [96][ 840/ 1236] Overall Loss 0.266150 Objective Loss 0.266150 LR 0.001000 Time 0.021852 +2023-10-02 21:15:16,965 - Epoch: [96][ 850/ 1236] Overall Loss 0.266305 Objective Loss 0.266305 LR 0.001000 Time 0.021839 +2023-10-02 21:15:17,175 - Epoch: [96][ 860/ 1236] Overall Loss 0.266290 Objective Loss 0.266290 LR 0.001000 Time 0.021828 +2023-10-02 21:15:17,383 - Epoch: [96][ 870/ 1236] Overall Loss 0.266487 Objective Loss 0.266487 LR 0.001000 Time 0.021815 +2023-10-02 21:15:17,594 - Epoch: [96][ 880/ 1236] Overall Loss 0.266955 Objective Loss 0.266955 LR 0.001000 Time 0.021806 +2023-10-02 21:15:17,802 - Epoch: [96][ 890/ 1236] Overall Loss 0.266880 Objective Loss 0.266880 LR 0.001000 Time 0.021795 +2023-10-02 21:15:18,012 - Epoch: [96][ 900/ 1236] Overall Loss 0.266523 Objective Loss 0.266523 LR 0.001000 Time 0.021786 +2023-10-02 21:15:18,220 - Epoch: [96][ 910/ 1236] Overall Loss 0.266433 Objective Loss 0.266433 LR 0.001000 Time 0.021773 +2023-10-02 21:15:18,430 - Epoch: [96][ 920/ 1236] Overall Loss 0.266359 Objective Loss 0.266359 LR 0.001000 Time 0.021764 +2023-10-02 21:15:18,638 - Epoch: [96][ 930/ 1236] Overall Loss 0.266547 Objective Loss 0.266547 LR 0.001000 Time 0.021752 +2023-10-02 21:15:18,848 - Epoch: [96][ 940/ 1236] Overall Loss 0.266560 Objective Loss 0.266560 LR 0.001000 Time 0.021744 +2023-10-02 21:15:19,057 - Epoch: [96][ 950/ 1236] Overall Loss 0.266350 Objective Loss 0.266350 LR 0.001000 Time 0.021733 +2023-10-02 21:15:19,266 - Epoch: [96][ 960/ 1236] Overall Loss 0.266185 Objective Loss 0.266185 LR 0.001000 Time 0.021724 +2023-10-02 21:15:19,475 - Epoch: [96][ 970/ 1236] Overall Loss 0.266074 Objective Loss 0.266074 LR 0.001000 Time 0.021714 +2023-10-02 21:15:19,684 - Epoch: [96][ 980/ 1236] Overall Loss 0.266422 Objective Loss 0.266422 LR 0.001000 Time 0.021706 +2023-10-02 21:15:19,890 - Epoch: [96][ 990/ 1236] Overall Loss 0.266286 Objective Loss 0.266286 LR 0.001000 Time 0.021692 +2023-10-02 21:15:20,093 - Epoch: [96][ 1000/ 1236] Overall Loss 0.266636 Objective Loss 0.266636 LR 0.001000 Time 0.021679 +2023-10-02 21:15:20,297 - Epoch: [96][ 1010/ 1236] Overall Loss 0.266874 Objective Loss 0.266874 LR 0.001000 Time 0.021666 +2023-10-02 21:15:20,501 - Epoch: [96][ 1020/ 1236] Overall Loss 0.267063 Objective Loss 0.267063 LR 0.001000 Time 0.021653 +2023-10-02 21:15:20,704 - Epoch: [96][ 1030/ 1236] Overall Loss 0.266977 Objective Loss 0.266977 LR 0.001000 Time 0.021640 +2023-10-02 21:15:20,909 - Epoch: [96][ 1040/ 1236] Overall Loss 0.267142 Objective Loss 0.267142 LR 0.001000 Time 0.021628 +2023-10-02 21:15:21,120 - Epoch: [96][ 1050/ 1236] Overall Loss 0.267232 Objective Loss 0.267232 LR 0.001000 Time 0.021623 +2023-10-02 21:15:21,324 - Epoch: [96][ 1060/ 1236] Overall Loss 0.267580 Objective Loss 0.267580 LR 0.001000 Time 0.021611 +2023-10-02 21:15:21,529 - Epoch: [96][ 1070/ 1236] Overall Loss 0.267625 Objective Loss 0.267625 LR 0.001000 Time 0.021600 +2023-10-02 21:15:21,733 - Epoch: [96][ 1080/ 1236] Overall Loss 0.267471 Objective Loss 0.267471 LR 0.001000 Time 0.021589 +2023-10-02 21:15:21,937 - Epoch: [96][ 1090/ 1236] Overall Loss 0.267526 Objective Loss 0.267526 LR 0.001000 Time 0.021578 +2023-10-02 21:15:22,141 - Epoch: [96][ 1100/ 1236] Overall Loss 0.267661 Objective Loss 0.267661 LR 0.001000 Time 0.021567 +2023-10-02 21:15:22,345 - Epoch: [96][ 1110/ 1236] Overall Loss 0.267884 Objective Loss 0.267884 LR 0.001000 Time 0.021557 +2023-10-02 21:15:22,550 - Epoch: [96][ 1120/ 1236] Overall Loss 0.267956 Objective Loss 0.267956 LR 0.001000 Time 0.021546 +2023-10-02 21:15:22,754 - Epoch: [96][ 1130/ 1236] Overall Loss 0.268131 Objective Loss 0.268131 LR 0.001000 Time 0.021536 +2023-10-02 21:15:22,958 - Epoch: [96][ 1140/ 1236] Overall Loss 0.268110 Objective Loss 0.268110 LR 0.001000 Time 0.021526 +2023-10-02 21:15:23,162 - Epoch: [96][ 1150/ 1236] Overall Loss 0.267923 Objective Loss 0.267923 LR 0.001000 Time 0.021516 +2023-10-02 21:15:23,366 - Epoch: [96][ 1160/ 1236] Overall Loss 0.267939 Objective Loss 0.267939 LR 0.001000 Time 0.021507 +2023-10-02 21:15:23,571 - Epoch: [96][ 1170/ 1236] Overall Loss 0.267805 Objective Loss 0.267805 LR 0.001000 Time 0.021497 +2023-10-02 21:15:23,775 - Epoch: [96][ 1180/ 1236] Overall Loss 0.267620 Objective Loss 0.267620 LR 0.001000 Time 0.021488 +2023-10-02 21:15:23,979 - Epoch: [96][ 1190/ 1236] Overall Loss 0.267617 Objective Loss 0.267617 LR 0.001000 Time 0.021479 +2023-10-02 21:15:24,183 - Epoch: [96][ 1200/ 1236] Overall Loss 0.267424 Objective Loss 0.267424 LR 0.001000 Time 0.021469 +2023-10-02 21:15:24,387 - Epoch: [96][ 1210/ 1236] Overall Loss 0.267436 Objective Loss 0.267436 LR 0.001000 Time 0.021461 +2023-10-02 21:15:24,592 - Epoch: [96][ 1220/ 1236] Overall Loss 0.267516 Objective Loss 0.267516 LR 0.001000 Time 0.021452 +2023-10-02 21:15:24,852 - Epoch: [96][ 1230/ 1236] Overall Loss 0.267563 Objective Loss 0.267563 LR 0.001000 Time 0.021489 +2023-10-02 21:15:24,976 - Epoch: [96][ 1236/ 1236] Overall Loss 0.267569 Objective Loss 0.267569 Top1 85.743381 Top5 98.370672 LR 0.001000 Time 0.021485 +2023-10-02 21:15:25,101 - --- validate (epoch=96)----------- +2023-10-02 21:15:25,101 - 29943 samples (256 per mini-batch) +2023-10-02 21:15:25,597 - Epoch: [96][ 10/ 117] Loss 0.317959 Top1 83.671875 Top5 98.007812 +2023-10-02 21:15:25,750 - Epoch: [96][ 20/ 117] Loss 0.325631 Top1 84.082031 Top5 97.910156 +2023-10-02 21:15:25,903 - Epoch: [96][ 30/ 117] Loss 0.328745 Top1 83.945312 Top5 98.072917 +2023-10-02 21:15:26,055 - Epoch: [96][ 40/ 117] Loss 0.331728 Top1 83.652344 Top5 98.164062 +2023-10-02 21:15:26,208 - Epoch: [96][ 50/ 117] Loss 0.328862 Top1 83.695312 Top5 98.101562 +2023-10-02 21:15:26,360 - Epoch: [96][ 60/ 117] Loss 0.339789 Top1 83.483073 Top5 97.994792 +2023-10-02 21:15:26,512 - Epoch: [96][ 70/ 117] Loss 0.342699 Top1 83.359375 Top5 98.002232 +2023-10-02 21:15:26,664 - Epoch: [96][ 80/ 117] Loss 0.342160 Top1 83.237305 Top5 98.027344 +2023-10-02 21:15:26,817 - Epoch: [96][ 90/ 117] Loss 0.344101 Top1 83.155382 Top5 98.016493 +2023-10-02 21:15:26,969 - Epoch: [96][ 100/ 117] Loss 0.344750 Top1 83.136719 Top5 97.992188 +2023-10-02 21:15:27,129 - Epoch: [96][ 110/ 117] Loss 0.343598 Top1 83.224432 Top5 98.014915 +2023-10-02 21:15:27,219 - Epoch: [96][ 117/ 117] Loss 0.342835 Top1 83.218114 Top5 98.022910 +2023-10-02 21:15:27,348 - ==> Top1: 83.218 Top5: 98.023 Loss: 0.343 + +2023-10-02 21:15:27,349 - ==> Confusion: +[[ 919 3 2 1 14 4 0 1 8 57 2 4 0 3 8 3 4 0 2 0 15] + [ 1 1032 0 0 6 31 3 22 1 1 3 2 1 1 3 3 0 0 11 3 7] + [ 1 3 930 14 1 0 38 13 0 0 5 0 6 1 1 10 1 1 13 7 11] + [ 1 3 16 966 0 6 5 1 5 0 12 1 5 3 27 5 0 2 17 1 13] + [ 16 6 3 1 968 12 0 0 2 9 3 0 2 2 7 3 10 2 1 0 3] + [ 3 33 0 0 3 985 1 30 1 4 4 5 3 9 7 0 4 0 2 10 12] + [ 0 4 16 0 0 2 1130 6 0 0 6 1 0 1 0 4 1 0 1 12 7] + [ 2 10 15 0 8 33 9 1056 2 1 4 5 1 3 3 1 1 2 46 9 7] + [ 19 4 3 1 0 4 0 3 945 40 18 3 3 10 20 0 2 1 8 2 3] + [ 98 0 2 0 13 5 1 0 28 911 3 3 1 28 12 2 0 0 1 3 8] + [ 3 2 9 10 0 1 3 3 11 1 954 1 2 17 6 0 2 2 11 3 12] + [ 1 0 1 0 2 18 0 5 1 0 1 930 39 3 0 4 1 14 1 11 3] + [ 0 0 2 4 0 2 3 1 1 1 3 29 964 5 3 10 2 18 4 11 5] + [ 1 0 1 0 1 14 1 0 8 8 5 5 1 1044 4 1 6 1 0 1 17] + [ 12 0 5 28 7 2 0 0 19 3 1 0 4 6 990 0 2 2 15 0 5] + [ 0 0 2 1 3 2 2 1 0 2 0 4 4 1 2 1081 14 9 0 4 2] + [ 1 15 3 1 9 7 1 0 0 0 0 4 0 1 4 12 1089 0 1 6 7] + [ 0 1 1 3 2 0 3 0 1 1 0 5 20 1 2 11 0 980 2 2 3] + [ 2 1 3 18 0 0 2 14 5 0 3 1 0 0 12 0 2 0 993 1 11] + [ 0 1 3 1 1 4 10 6 0 1 1 7 6 0 0 2 8 3 0 1091 7] + [ 115 187 115 86 87 213 68 121 81 89 195 92 379 276 145 74 136 73 186 228 4959]] + +2023-10-02 21:15:27,350 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:15:27,350 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:15:27,357 - + +2023-10-02 21:15:27,357 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:15:28,392 - Epoch: [97][ 10/ 1236] Overall Loss 0.259494 Objective Loss 0.259494 LR 0.001000 Time 0.103512 +2023-10-02 21:15:28,604 - Epoch: [97][ 20/ 1236] Overall Loss 0.266016 Objective Loss 0.266016 LR 0.001000 Time 0.062322 +2023-10-02 21:15:28,813 - Epoch: [97][ 30/ 1236] Overall Loss 0.262668 Objective Loss 0.262668 LR 0.001000 Time 0.048500 +2023-10-02 21:15:29,024 - Epoch: [97][ 40/ 1236] Overall Loss 0.259206 Objective Loss 0.259206 LR 0.001000 Time 0.041645 +2023-10-02 21:15:29,234 - Epoch: [97][ 50/ 1236] Overall Loss 0.260746 Objective Loss 0.260746 LR 0.001000 Time 0.037506 +2023-10-02 21:15:29,445 - Epoch: [97][ 60/ 1236] Overall Loss 0.260893 Objective Loss 0.260893 LR 0.001000 Time 0.034760 +2023-10-02 21:15:29,654 - Epoch: [97][ 70/ 1236] Overall Loss 0.262727 Objective Loss 0.262727 LR 0.001000 Time 0.032783 +2023-10-02 21:15:29,865 - Epoch: [97][ 80/ 1236] Overall Loss 0.265934 Objective Loss 0.265934 LR 0.001000 Time 0.031317 +2023-10-02 21:15:30,074 - Epoch: [97][ 90/ 1236] Overall Loss 0.266697 Objective Loss 0.266697 LR 0.001000 Time 0.030156 +2023-10-02 21:15:30,285 - Epoch: [97][ 100/ 1236] Overall Loss 0.266555 Objective Loss 0.266555 LR 0.001000 Time 0.029244 +2023-10-02 21:15:30,494 - Epoch: [97][ 110/ 1236] Overall Loss 0.265479 Objective Loss 0.265479 LR 0.001000 Time 0.028471 +2023-10-02 21:15:30,702 - Epoch: [97][ 120/ 1236] Overall Loss 0.262230 Objective Loss 0.262230 LR 0.001000 Time 0.027835 +2023-10-02 21:15:30,911 - Epoch: [97][ 130/ 1236] Overall Loss 0.261906 Objective Loss 0.261906 LR 0.001000 Time 0.027296 +2023-10-02 21:15:31,122 - Epoch: [97][ 140/ 1236] Overall Loss 0.263994 Objective Loss 0.263994 LR 0.001000 Time 0.026850 +2023-10-02 21:15:31,331 - Epoch: [97][ 150/ 1236] Overall Loss 0.262753 Objective Loss 0.262753 LR 0.001000 Time 0.026443 +2023-10-02 21:15:31,542 - Epoch: [97][ 160/ 1236] Overall Loss 0.263473 Objective Loss 0.263473 LR 0.001000 Time 0.026108 +2023-10-02 21:15:31,751 - Epoch: [97][ 170/ 1236] Overall Loss 0.265628 Objective Loss 0.265628 LR 0.001000 Time 0.025791 +2023-10-02 21:15:31,962 - Epoch: [97][ 180/ 1236] Overall Loss 0.264137 Objective Loss 0.264137 LR 0.001000 Time 0.025529 +2023-10-02 21:15:32,171 - Epoch: [97][ 190/ 1236] Overall Loss 0.264052 Objective Loss 0.264052 LR 0.001000 Time 0.025276 +2023-10-02 21:15:32,382 - Epoch: [97][ 200/ 1236] Overall Loss 0.265582 Objective Loss 0.265582 LR 0.001000 Time 0.025066 +2023-10-02 21:15:32,590 - Epoch: [97][ 210/ 1236] Overall Loss 0.266171 Objective Loss 0.266171 LR 0.001000 Time 0.024858 +2023-10-02 21:15:32,801 - Epoch: [97][ 220/ 1236] Overall Loss 0.266786 Objective Loss 0.266786 LR 0.001000 Time 0.024685 +2023-10-02 21:15:33,010 - Epoch: [97][ 230/ 1236] Overall Loss 0.265843 Objective Loss 0.265843 LR 0.001000 Time 0.024513 +2023-10-02 21:15:33,221 - Epoch: [97][ 240/ 1236] Overall Loss 0.265912 Objective Loss 0.265912 LR 0.001000 Time 0.024370 +2023-10-02 21:15:33,430 - Epoch: [97][ 250/ 1236] Overall Loss 0.265551 Objective Loss 0.265551 LR 0.001000 Time 0.024225 +2023-10-02 21:15:33,640 - Epoch: [97][ 260/ 1236] Overall Loss 0.265958 Objective Loss 0.265958 LR 0.001000 Time 0.024101 +2023-10-02 21:15:33,849 - Epoch: [97][ 270/ 1236] Overall Loss 0.266181 Objective Loss 0.266181 LR 0.001000 Time 0.023975 +2023-10-02 21:15:34,060 - Epoch: [97][ 280/ 1236] Overall Loss 0.265453 Objective Loss 0.265453 LR 0.001000 Time 0.023871 +2023-10-02 21:15:34,270 - Epoch: [97][ 290/ 1236] Overall Loss 0.265260 Objective Loss 0.265260 LR 0.001000 Time 0.023767 +2023-10-02 21:15:34,481 - Epoch: [97][ 300/ 1236] Overall Loss 0.265349 Objective Loss 0.265349 LR 0.001000 Time 0.023679 +2023-10-02 21:15:34,691 - Epoch: [97][ 310/ 1236] Overall Loss 0.264650 Objective Loss 0.264650 LR 0.001000 Time 0.023587 +2023-10-02 21:15:34,902 - Epoch: [97][ 320/ 1236] Overall Loss 0.265601 Objective Loss 0.265601 LR 0.001000 Time 0.023509 +2023-10-02 21:15:35,111 - Epoch: [97][ 330/ 1236] Overall Loss 0.266086 Objective Loss 0.266086 LR 0.001000 Time 0.023425 +2023-10-02 21:15:35,322 - Epoch: [97][ 340/ 1236] Overall Loss 0.266009 Objective Loss 0.266009 LR 0.001000 Time 0.023356 +2023-10-02 21:15:35,532 - Epoch: [97][ 350/ 1236] Overall Loss 0.265859 Objective Loss 0.265859 LR 0.001000 Time 0.023287 +2023-10-02 21:15:35,743 - Epoch: [97][ 360/ 1236] Overall Loss 0.265622 Objective Loss 0.265622 LR 0.001000 Time 0.023226 +2023-10-02 21:15:35,953 - Epoch: [97][ 370/ 1236] Overall Loss 0.265555 Objective Loss 0.265555 LR 0.001000 Time 0.023163 +2023-10-02 21:15:36,164 - Epoch: [97][ 380/ 1236] Overall Loss 0.265272 Objective Loss 0.265272 LR 0.001000 Time 0.023108 +2023-10-02 21:15:36,373 - Epoch: [97][ 390/ 1236] Overall Loss 0.265328 Objective Loss 0.265328 LR 0.001000 Time 0.023050 +2023-10-02 21:15:36,584 - Epoch: [97][ 400/ 1236] Overall Loss 0.265290 Objective Loss 0.265290 LR 0.001000 Time 0.023002 +2023-10-02 21:15:36,794 - Epoch: [97][ 410/ 1236] Overall Loss 0.265265 Objective Loss 0.265265 LR 0.001000 Time 0.022948 +2023-10-02 21:15:37,005 - Epoch: [97][ 420/ 1236] Overall Loss 0.264760 Objective Loss 0.264760 LR 0.001000 Time 0.022903 +2023-10-02 21:15:37,214 - Epoch: [97][ 430/ 1236] Overall Loss 0.264478 Objective Loss 0.264478 LR 0.001000 Time 0.022856 +2023-10-02 21:15:37,424 - Epoch: [97][ 440/ 1236] Overall Loss 0.264861 Objective Loss 0.264861 LR 0.001000 Time 0.022814 +2023-10-02 21:15:37,634 - Epoch: [97][ 450/ 1236] Overall Loss 0.264660 Objective Loss 0.264660 LR 0.001000 Time 0.022772 +2023-10-02 21:15:37,846 - Epoch: [97][ 460/ 1236] Overall Loss 0.264215 Objective Loss 0.264215 LR 0.001000 Time 0.022738 +2023-10-02 21:15:38,056 - Epoch: [97][ 470/ 1236] Overall Loss 0.264894 Objective Loss 0.264894 LR 0.001000 Time 0.022700 +2023-10-02 21:15:38,265 - Epoch: [97][ 480/ 1236] Overall Loss 0.264816 Objective Loss 0.264816 LR 0.001000 Time 0.022661 +2023-10-02 21:15:38,474 - Epoch: [97][ 490/ 1236] Overall Loss 0.264542 Objective Loss 0.264542 LR 0.001000 Time 0.022626 +2023-10-02 21:15:38,685 - Epoch: [97][ 500/ 1236] Overall Loss 0.264311 Objective Loss 0.264311 LR 0.001000 Time 0.022595 +2023-10-02 21:15:38,895 - Epoch: [97][ 510/ 1236] Overall Loss 0.263690 Objective Loss 0.263690 LR 0.001000 Time 0.022560 +2023-10-02 21:15:39,106 - Epoch: [97][ 520/ 1236] Overall Loss 0.263418 Objective Loss 0.263418 LR 0.001000 Time 0.022531 +2023-10-02 21:15:39,315 - Epoch: [97][ 530/ 1236] Overall Loss 0.263774 Objective Loss 0.263774 LR 0.001000 Time 0.022500 +2023-10-02 21:15:39,527 - Epoch: [97][ 540/ 1236] Overall Loss 0.263700 Objective Loss 0.263700 LR 0.001000 Time 0.022475 +2023-10-02 21:15:39,736 - Epoch: [97][ 550/ 1236] Overall Loss 0.264206 Objective Loss 0.264206 LR 0.001000 Time 0.022446 +2023-10-02 21:15:39,947 - Epoch: [97][ 560/ 1236] Overall Loss 0.263938 Objective Loss 0.263938 LR 0.001000 Time 0.022422 +2023-10-02 21:15:40,157 - Epoch: [97][ 570/ 1236] Overall Loss 0.263985 Objective Loss 0.263985 LR 0.001000 Time 0.022395 +2023-10-02 21:15:40,367 - Epoch: [97][ 580/ 1236] Overall Loss 0.263571 Objective Loss 0.263571 LR 0.001000 Time 0.022372 +2023-10-02 21:15:40,577 - Epoch: [97][ 590/ 1236] Overall Loss 0.263699 Objective Loss 0.263699 LR 0.001000 Time 0.022348 +2023-10-02 21:15:40,788 - Epoch: [97][ 600/ 1236] Overall Loss 0.264307 Objective Loss 0.264307 LR 0.001000 Time 0.022326 +2023-10-02 21:15:40,997 - Epoch: [97][ 610/ 1236] Overall Loss 0.264594 Objective Loss 0.264594 LR 0.001000 Time 0.022302 +2023-10-02 21:15:41,209 - Epoch: [97][ 620/ 1236] Overall Loss 0.265104 Objective Loss 0.265104 LR 0.001000 Time 0.022283 +2023-10-02 21:15:41,418 - Epoch: [97][ 630/ 1236] Overall Loss 0.265376 Objective Loss 0.265376 LR 0.001000 Time 0.022262 +2023-10-02 21:15:41,629 - Epoch: [97][ 640/ 1236] Overall Loss 0.265515 Objective Loss 0.265515 LR 0.001000 Time 0.022243 +2023-10-02 21:15:41,839 - Epoch: [97][ 650/ 1236] Overall Loss 0.265119 Objective Loss 0.265119 LR 0.001000 Time 0.022222 +2023-10-02 21:15:42,051 - Epoch: [97][ 660/ 1236] Overall Loss 0.265621 Objective Loss 0.265621 LR 0.001000 Time 0.022206 +2023-10-02 21:15:42,260 - Epoch: [97][ 670/ 1236] Overall Loss 0.265318 Objective Loss 0.265318 LR 0.001000 Time 0.022187 +2023-10-02 21:15:42,471 - Epoch: [97][ 680/ 1236] Overall Loss 0.264912 Objective Loss 0.264912 LR 0.001000 Time 0.022171 +2023-10-02 21:15:42,681 - Epoch: [97][ 690/ 1236] Overall Loss 0.264734 Objective Loss 0.264734 LR 0.001000 Time 0.022153 +2023-10-02 21:15:42,892 - Epoch: [97][ 700/ 1236] Overall Loss 0.264746 Objective Loss 0.264746 LR 0.001000 Time 0.022138 +2023-10-02 21:15:43,102 - Epoch: [97][ 710/ 1236] Overall Loss 0.264178 Objective Loss 0.264178 LR 0.001000 Time 0.022121 +2023-10-02 21:15:43,314 - Epoch: [97][ 720/ 1236] Overall Loss 0.264147 Objective Loss 0.264147 LR 0.001000 Time 0.022108 +2023-10-02 21:15:43,523 - Epoch: [97][ 730/ 1236] Overall Loss 0.264103 Objective Loss 0.264103 LR 0.001000 Time 0.022091 +2023-10-02 21:15:43,735 - Epoch: [97][ 740/ 1236] Overall Loss 0.264100 Objective Loss 0.264100 LR 0.001000 Time 0.022077 +2023-10-02 21:15:43,945 - Epoch: [97][ 750/ 1236] Overall Loss 0.264071 Objective Loss 0.264071 LR 0.001000 Time 0.022062 +2023-10-02 21:15:44,156 - Epoch: [97][ 760/ 1236] Overall Loss 0.264447 Objective Loss 0.264447 LR 0.001000 Time 0.022049 +2023-10-02 21:15:44,366 - Epoch: [97][ 770/ 1236] Overall Loss 0.264532 Objective Loss 0.264532 LR 0.001000 Time 0.022035 +2023-10-02 21:15:44,576 - Epoch: [97][ 780/ 1236] Overall Loss 0.264900 Objective Loss 0.264900 LR 0.001000 Time 0.022022 +2023-10-02 21:15:44,786 - Epoch: [97][ 790/ 1236] Overall Loss 0.264711 Objective Loss 0.264711 LR 0.001000 Time 0.022009 +2023-10-02 21:15:44,997 - Epoch: [97][ 800/ 1236] Overall Loss 0.264615 Objective Loss 0.264615 LR 0.001000 Time 0.021997 +2023-10-02 21:15:45,206 - Epoch: [97][ 810/ 1236] Overall Loss 0.264416 Objective Loss 0.264416 LR 0.001000 Time 0.021983 +2023-10-02 21:15:45,418 - Epoch: [97][ 820/ 1236] Overall Loss 0.264529 Objective Loss 0.264529 LR 0.001000 Time 0.021972 +2023-10-02 21:15:45,628 - Epoch: [97][ 830/ 1236] Overall Loss 0.264554 Objective Loss 0.264554 LR 0.001000 Time 0.021960 +2023-10-02 21:15:45,839 - Epoch: [97][ 840/ 1236] Overall Loss 0.264216 Objective Loss 0.264216 LR 0.001000 Time 0.021950 +2023-10-02 21:15:46,049 - Epoch: [97][ 850/ 1236] Overall Loss 0.264072 Objective Loss 0.264072 LR 0.001000 Time 0.021938 +2023-10-02 21:15:46,259 - Epoch: [97][ 860/ 1236] Overall Loss 0.264574 Objective Loss 0.264574 LR 0.001000 Time 0.021928 +2023-10-02 21:15:46,470 - Epoch: [97][ 870/ 1236] Overall Loss 0.264597 Objective Loss 0.264597 LR 0.001000 Time 0.021916 +2023-10-02 21:15:46,681 - Epoch: [97][ 880/ 1236] Overall Loss 0.264753 Objective Loss 0.264753 LR 0.001000 Time 0.021906 +2023-10-02 21:15:46,890 - Epoch: [97][ 890/ 1236] Overall Loss 0.264364 Objective Loss 0.264364 LR 0.001000 Time 0.021894 +2023-10-02 21:15:47,101 - Epoch: [97][ 900/ 1236] Overall Loss 0.264714 Objective Loss 0.264714 LR 0.001000 Time 0.021886 +2023-10-02 21:15:47,310 - Epoch: [97][ 910/ 1236] Overall Loss 0.265082 Objective Loss 0.265082 LR 0.001000 Time 0.021874 +2023-10-02 21:15:47,522 - Epoch: [97][ 920/ 1236] Overall Loss 0.265581 Objective Loss 0.265581 LR 0.001000 Time 0.021866 +2023-10-02 21:15:47,732 - Epoch: [97][ 930/ 1236] Overall Loss 0.265507 Objective Loss 0.265507 LR 0.001000 Time 0.021856 +2023-10-02 21:15:47,942 - Epoch: [97][ 940/ 1236] Overall Loss 0.265554 Objective Loss 0.265554 LR 0.001000 Time 0.021848 +2023-10-02 21:15:48,153 - Epoch: [97][ 950/ 1236] Overall Loss 0.265873 Objective Loss 0.265873 LR 0.001000 Time 0.021839 +2023-10-02 21:15:48,370 - Epoch: [97][ 960/ 1236] Overall Loss 0.266025 Objective Loss 0.266025 LR 0.001000 Time 0.021837 +2023-10-02 21:15:48,584 - Epoch: [97][ 970/ 1236] Overall Loss 0.265984 Objective Loss 0.265984 LR 0.001000 Time 0.021832 +2023-10-02 21:15:48,804 - Epoch: [97][ 980/ 1236] Overall Loss 0.266210 Objective Loss 0.266210 LR 0.001000 Time 0.021833 +2023-10-02 21:15:49,018 - Epoch: [97][ 990/ 1236] Overall Loss 0.266354 Objective Loss 0.266354 LR 0.001000 Time 0.021829 +2023-10-02 21:15:49,238 - Epoch: [97][ 1000/ 1236] Overall Loss 0.266207 Objective Loss 0.266207 LR 0.001000 Time 0.021830 +2023-10-02 21:15:49,452 - Epoch: [97][ 1010/ 1236] Overall Loss 0.266176 Objective Loss 0.266176 LR 0.001000 Time 0.021826 +2023-10-02 21:15:49,672 - Epoch: [97][ 1020/ 1236] Overall Loss 0.266129 Objective Loss 0.266129 LR 0.001000 Time 0.021827 +2023-10-02 21:15:49,886 - Epoch: [97][ 1030/ 1236] Overall Loss 0.265876 Objective Loss 0.265876 LR 0.001000 Time 0.021823 +2023-10-02 21:15:50,106 - Epoch: [97][ 1040/ 1236] Overall Loss 0.265940 Objective Loss 0.265940 LR 0.001000 Time 0.021824 +2023-10-02 21:15:50,321 - Epoch: [97][ 1050/ 1236] Overall Loss 0.266030 Objective Loss 0.266030 LR 0.001000 Time 0.021821 +2023-10-02 21:15:50,541 - Epoch: [97][ 1060/ 1236] Overall Loss 0.266204 Objective Loss 0.266204 LR 0.001000 Time 0.021822 +2023-10-02 21:15:50,755 - Epoch: [97][ 1070/ 1236] Overall Loss 0.266232 Objective Loss 0.266232 LR 0.001000 Time 0.021818 +2023-10-02 21:15:50,964 - Epoch: [97][ 1080/ 1236] Overall Loss 0.266576 Objective Loss 0.266576 LR 0.001000 Time 0.021809 +2023-10-02 21:15:51,173 - Epoch: [97][ 1090/ 1236] Overall Loss 0.266716 Objective Loss 0.266716 LR 0.001000 Time 0.021800 +2023-10-02 21:15:51,384 - Epoch: [97][ 1100/ 1236] Overall Loss 0.266656 Objective Loss 0.266656 LR 0.001000 Time 0.021794 +2023-10-02 21:15:51,593 - Epoch: [97][ 1110/ 1236] Overall Loss 0.266867 Objective Loss 0.266867 LR 0.001000 Time 0.021785 +2023-10-02 21:15:51,804 - Epoch: [97][ 1120/ 1236] Overall Loss 0.266871 Objective Loss 0.266871 LR 0.001000 Time 0.021779 +2023-10-02 21:15:52,013 - Epoch: [97][ 1130/ 1236] Overall Loss 0.266854 Objective Loss 0.266854 LR 0.001000 Time 0.021771 +2023-10-02 21:15:52,224 - Epoch: [97][ 1140/ 1236] Overall Loss 0.266907 Objective Loss 0.266907 LR 0.001000 Time 0.021765 +2023-10-02 21:15:52,433 - Epoch: [97][ 1150/ 1236] Overall Loss 0.267035 Objective Loss 0.267035 LR 0.001000 Time 0.021757 +2023-10-02 21:15:52,644 - Epoch: [97][ 1160/ 1236] Overall Loss 0.266909 Objective Loss 0.266909 LR 0.001000 Time 0.021751 +2023-10-02 21:15:52,853 - Epoch: [97][ 1170/ 1236] Overall Loss 0.267044 Objective Loss 0.267044 LR 0.001000 Time 0.021743 +2023-10-02 21:15:53,064 - Epoch: [97][ 1180/ 1236] Overall Loss 0.266893 Objective Loss 0.266893 LR 0.001000 Time 0.021738 +2023-10-02 21:15:53,273 - Epoch: [97][ 1190/ 1236] Overall Loss 0.266746 Objective Loss 0.266746 LR 0.001000 Time 0.021730 +2023-10-02 21:15:53,484 - Epoch: [97][ 1200/ 1236] Overall Loss 0.266843 Objective Loss 0.266843 LR 0.001000 Time 0.021725 +2023-10-02 21:15:53,693 - Epoch: [97][ 1210/ 1236] Overall Loss 0.267020 Objective Loss 0.267020 LR 0.001000 Time 0.021718 +2023-10-02 21:15:53,904 - Epoch: [97][ 1220/ 1236] Overall Loss 0.267039 Objective Loss 0.267039 LR 0.001000 Time 0.021712 +2023-10-02 21:15:54,167 - Epoch: [97][ 1230/ 1236] Overall Loss 0.267193 Objective Loss 0.267193 LR 0.001000 Time 0.021749 +2023-10-02 21:15:54,289 - Epoch: [97][ 1236/ 1236] Overall Loss 0.267083 Objective Loss 0.267083 Top1 86.354379 Top5 98.167006 LR 0.001000 Time 0.021742 +2023-10-02 21:15:54,434 - --- validate (epoch=97)----------- +2023-10-02 21:15:54,434 - 29943 samples (256 per mini-batch) +2023-10-02 21:15:54,911 - Epoch: [97][ 10/ 117] Loss 0.349215 Top1 83.750000 Top5 97.968750 +2023-10-02 21:15:55,066 - Epoch: [97][ 20/ 117] Loss 0.364162 Top1 83.183594 Top5 98.007812 +2023-10-02 21:15:55,217 - Epoch: [97][ 30/ 117] Loss 0.359063 Top1 83.059896 Top5 97.981771 +2023-10-02 21:15:55,371 - Epoch: [97][ 40/ 117] Loss 0.347711 Top1 83.281250 Top5 98.037109 +2023-10-02 21:15:55,523 - Epoch: [97][ 50/ 117] Loss 0.347314 Top1 83.242188 Top5 98.078125 +2023-10-02 21:15:55,677 - Epoch: [97][ 60/ 117] Loss 0.347014 Top1 83.268229 Top5 98.072917 +2023-10-02 21:15:55,833 - Epoch: [97][ 70/ 117] Loss 0.348664 Top1 83.303571 Top5 98.080357 +2023-10-02 21:15:55,992 - Epoch: [97][ 80/ 117] Loss 0.348683 Top1 83.305664 Top5 98.017578 +2023-10-02 21:15:56,151 - Epoch: [97][ 90/ 117] Loss 0.347545 Top1 83.311632 Top5 98.003472 +2023-10-02 21:15:56,310 - Epoch: [97][ 100/ 117] Loss 0.346901 Top1 83.394531 Top5 98.007812 +2023-10-02 21:15:56,477 - Epoch: [97][ 110/ 117] Loss 0.346120 Top1 83.476562 Top5 98.057528 +2023-10-02 21:15:56,567 - Epoch: [97][ 117/ 117] Loss 0.344246 Top1 83.628895 Top5 98.066326 +2023-10-02 21:15:56,694 - ==> Top1: 83.629 Top5: 98.066 Loss: 0.344 + +2023-10-02 21:15:56,695 - ==> Confusion: +[[ 933 0 3 1 7 3 1 0 11 59 4 2 1 0 4 0 5 0 2 0 14] + [ 1 1048 0 2 8 18 4 18 2 3 2 1 0 1 1 3 2 0 8 2 7] + [ 4 0 950 24 5 0 25 10 1 1 1 0 5 1 1 3 0 1 14 3 7] + [ 2 3 13 969 0 7 1 2 11 0 5 0 12 3 32 0 0 2 13 0 14] + [ 31 9 1 0 954 7 1 1 3 11 1 0 1 2 7 6 9 1 1 1 3] + [ 4 40 1 4 1 964 2 32 5 5 2 9 3 10 5 1 2 0 3 9 14] + [ 1 6 34 1 0 0 1113 6 0 0 4 1 0 1 0 6 1 0 0 9 8] + [ 7 27 6 1 6 31 7 1050 2 3 3 5 3 4 0 0 4 3 45 5 6] + [ 20 2 0 2 2 1 0 1 988 36 8 1 3 7 11 0 5 0 1 0 1] + [ 108 1 0 1 4 2 1 1 37 925 2 0 0 20 5 2 1 1 0 1 7] + [ 1 3 9 9 3 0 6 7 20 0 964 3 1 5 4 0 2 3 3 1 9] + [ 0 1 1 1 2 10 0 4 0 2 0 946 29 9 0 0 2 14 0 8 6] + [ 0 1 3 1 0 0 2 2 1 1 2 38 969 1 4 2 6 12 5 9 9] + [ 2 0 1 0 2 4 0 1 20 14 3 6 7 1016 10 1 4 1 0 3 24] + [ 14 2 4 22 4 0 0 0 39 2 2 0 3 3 987 0 1 1 7 0 10] + [ 3 0 1 1 6 0 0 0 0 1 0 8 6 0 1 1052 22 15 3 5 10] + [ 2 15 1 1 8 7 1 1 2 0 0 7 1 1 7 7 1088 0 0 2 10] + [ 0 0 0 5 0 0 2 0 2 0 0 14 25 0 3 6 1 975 0 0 5] + [ 2 6 5 22 1 2 0 14 5 0 1 2 1 0 11 0 1 0 984 0 11] + [ 0 3 2 3 1 8 13 10 0 0 1 15 4 1 0 1 10 0 2 1066 12] + [ 149 207 108 110 92 132 31 94 142 104 184 123 407 199 138 47 141 63 178 156 5100]] + +2023-10-02 21:15:56,696 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:15:56,696 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:15:56,703 - + +2023-10-02 21:15:56,703 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:15:57,844 - Epoch: [98][ 10/ 1236] Overall Loss 0.252949 Objective Loss 0.252949 LR 0.001000 Time 0.114095 +2023-10-02 21:15:58,052 - Epoch: [98][ 20/ 1236] Overall Loss 0.251882 Objective Loss 0.251882 LR 0.001000 Time 0.067435 +2023-10-02 21:15:58,260 - Epoch: [98][ 30/ 1236] Overall Loss 0.254467 Objective Loss 0.254467 LR 0.001000 Time 0.051861 +2023-10-02 21:15:58,468 - Epoch: [98][ 40/ 1236] Overall Loss 0.250590 Objective Loss 0.250590 LR 0.001000 Time 0.044098 +2023-10-02 21:15:58,676 - Epoch: [98][ 50/ 1236] Overall Loss 0.249174 Objective Loss 0.249174 LR 0.001000 Time 0.039399 +2023-10-02 21:15:58,884 - Epoch: [98][ 60/ 1236] Overall Loss 0.250451 Objective Loss 0.250451 LR 0.001000 Time 0.036299 +2023-10-02 21:15:59,092 - Epoch: [98][ 70/ 1236] Overall Loss 0.252106 Objective Loss 0.252106 LR 0.001000 Time 0.034057 +2023-10-02 21:15:59,300 - Epoch: [98][ 80/ 1236] Overall Loss 0.252513 Objective Loss 0.252513 LR 0.001000 Time 0.032402 +2023-10-02 21:15:59,507 - Epoch: [98][ 90/ 1236] Overall Loss 0.251575 Objective Loss 0.251575 LR 0.001000 Time 0.031088 +2023-10-02 21:15:59,717 - Epoch: [98][ 100/ 1236] Overall Loss 0.253616 Objective Loss 0.253616 LR 0.001000 Time 0.030072 +2023-10-02 21:15:59,923 - Epoch: [98][ 110/ 1236] Overall Loss 0.253505 Objective Loss 0.253505 LR 0.001000 Time 0.029210 +2023-10-02 21:16:00,132 - Epoch: [98][ 120/ 1236] Overall Loss 0.254913 Objective Loss 0.254913 LR 0.001000 Time 0.028510 +2023-10-02 21:16:00,339 - Epoch: [98][ 130/ 1236] Overall Loss 0.256880 Objective Loss 0.256880 LR 0.001000 Time 0.027897 +2023-10-02 21:16:00,547 - Epoch: [98][ 140/ 1236] Overall Loss 0.254926 Objective Loss 0.254926 LR 0.001000 Time 0.027394 +2023-10-02 21:16:00,755 - Epoch: [98][ 150/ 1236] Overall Loss 0.255030 Objective Loss 0.255030 LR 0.001000 Time 0.026939 +2023-10-02 21:16:00,963 - Epoch: [98][ 160/ 1236] Overall Loss 0.255291 Objective Loss 0.255291 LR 0.001000 Time 0.026553 +2023-10-02 21:16:01,170 - Epoch: [98][ 170/ 1236] Overall Loss 0.255555 Objective Loss 0.255555 LR 0.001000 Time 0.026202 +2023-10-02 21:16:01,378 - Epoch: [98][ 180/ 1236] Overall Loss 0.256310 Objective Loss 0.256310 LR 0.001000 Time 0.025903 +2023-10-02 21:16:01,585 - Epoch: [98][ 190/ 1236] Overall Loss 0.256874 Objective Loss 0.256874 LR 0.001000 Time 0.025622 +2023-10-02 21:16:01,794 - Epoch: [98][ 200/ 1236] Overall Loss 0.256059 Objective Loss 0.256059 LR 0.001000 Time 0.025380 +2023-10-02 21:16:02,000 - Epoch: [98][ 210/ 1236] Overall Loss 0.255818 Objective Loss 0.255818 LR 0.001000 Time 0.025149 +2023-10-02 21:16:02,209 - Epoch: [98][ 220/ 1236] Overall Loss 0.258268 Objective Loss 0.258268 LR 0.001000 Time 0.024954 +2023-10-02 21:16:02,417 - Epoch: [98][ 230/ 1236] Overall Loss 0.258829 Objective Loss 0.258829 LR 0.001000 Time 0.024764 +2023-10-02 21:16:02,625 - Epoch: [98][ 240/ 1236] Overall Loss 0.259156 Objective Loss 0.259156 LR 0.001000 Time 0.024601 +2023-10-02 21:16:02,833 - Epoch: [98][ 250/ 1236] Overall Loss 0.259613 Objective Loss 0.259613 LR 0.001000 Time 0.024439 +2023-10-02 21:16:03,043 - Epoch: [98][ 260/ 1236] Overall Loss 0.260683 Objective Loss 0.260683 LR 0.001000 Time 0.024308 +2023-10-02 21:16:03,252 - Epoch: [98][ 270/ 1236] Overall Loss 0.260641 Objective Loss 0.260641 LR 0.001000 Time 0.024175 +2023-10-02 21:16:03,462 - Epoch: [98][ 280/ 1236] Overall Loss 0.259484 Objective Loss 0.259484 LR 0.001000 Time 0.024063 +2023-10-02 21:16:03,671 - Epoch: [98][ 290/ 1236] Overall Loss 0.259977 Objective Loss 0.259977 LR 0.001000 Time 0.023947 +2023-10-02 21:16:03,882 - Epoch: [98][ 300/ 1236] Overall Loss 0.260237 Objective Loss 0.260237 LR 0.001000 Time 0.023850 +2023-10-02 21:16:04,090 - Epoch: [98][ 310/ 1236] Overall Loss 0.259548 Objective Loss 0.259548 LR 0.001000 Time 0.023746 +2023-10-02 21:16:04,300 - Epoch: [98][ 320/ 1236] Overall Loss 0.259656 Objective Loss 0.259656 LR 0.001000 Time 0.023660 +2023-10-02 21:16:04,509 - Epoch: [98][ 330/ 1236] Overall Loss 0.258154 Objective Loss 0.258154 LR 0.001000 Time 0.023577 +2023-10-02 21:16:04,719 - Epoch: [98][ 340/ 1236] Overall Loss 0.258365 Objective Loss 0.258365 LR 0.001000 Time 0.023501 +2023-10-02 21:16:04,929 - Epoch: [98][ 350/ 1236] Overall Loss 0.258056 Objective Loss 0.258056 LR 0.001000 Time 0.023427 +2023-10-02 21:16:05,139 - Epoch: [98][ 360/ 1236] Overall Loss 0.258120 Objective Loss 0.258120 LR 0.001000 Time 0.023358 +2023-10-02 21:16:05,350 - Epoch: [98][ 370/ 1236] Overall Loss 0.257891 Objective Loss 0.257891 LR 0.001000 Time 0.023296 +2023-10-02 21:16:05,557 - Epoch: [98][ 380/ 1236] Overall Loss 0.257225 Objective Loss 0.257225 LR 0.001000 Time 0.023225 +2023-10-02 21:16:05,767 - Epoch: [98][ 390/ 1236] Overall Loss 0.257663 Objective Loss 0.257663 LR 0.001000 Time 0.023166 +2023-10-02 21:16:05,973 - Epoch: [98][ 400/ 1236] Overall Loss 0.257539 Objective Loss 0.257539 LR 0.001000 Time 0.023103 +2023-10-02 21:16:06,182 - Epoch: [98][ 410/ 1236] Overall Loss 0.257391 Objective Loss 0.257391 LR 0.001000 Time 0.023048 +2023-10-02 21:16:06,389 - Epoch: [98][ 420/ 1236] Overall Loss 0.257752 Objective Loss 0.257752 LR 0.001000 Time 0.022991 +2023-10-02 21:16:06,598 - Epoch: [98][ 430/ 1236] Overall Loss 0.257614 Objective Loss 0.257614 LR 0.001000 Time 0.022941 +2023-10-02 21:16:06,805 - Epoch: [98][ 440/ 1236] Overall Loss 0.257308 Objective Loss 0.257308 LR 0.001000 Time 0.022889 +2023-10-02 21:16:07,013 - Epoch: [98][ 450/ 1236] Overall Loss 0.257846 Objective Loss 0.257846 LR 0.001000 Time 0.022843 +2023-10-02 21:16:07,220 - Epoch: [98][ 460/ 1236] Overall Loss 0.257838 Objective Loss 0.257838 LR 0.001000 Time 0.022796 +2023-10-02 21:16:07,429 - Epoch: [98][ 470/ 1236] Overall Loss 0.258115 Objective Loss 0.258115 LR 0.001000 Time 0.022755 +2023-10-02 21:16:07,636 - Epoch: [98][ 480/ 1236] Overall Loss 0.258221 Objective Loss 0.258221 LR 0.001000 Time 0.022710 +2023-10-02 21:16:07,845 - Epoch: [98][ 490/ 1236] Overall Loss 0.258123 Objective Loss 0.258123 LR 0.001000 Time 0.022673 +2023-10-02 21:16:08,052 - Epoch: [98][ 500/ 1236] Overall Loss 0.258138 Objective Loss 0.258138 LR 0.001000 Time 0.022633 +2023-10-02 21:16:08,261 - Epoch: [98][ 510/ 1236] Overall Loss 0.258190 Objective Loss 0.258190 LR 0.001000 Time 0.022598 +2023-10-02 21:16:08,468 - Epoch: [98][ 520/ 1236] Overall Loss 0.258136 Objective Loss 0.258136 LR 0.001000 Time 0.022560 +2023-10-02 21:16:08,676 - Epoch: [98][ 530/ 1236] Overall Loss 0.257968 Objective Loss 0.257968 LR 0.001000 Time 0.022528 +2023-10-02 21:16:08,883 - Epoch: [98][ 540/ 1236] Overall Loss 0.258067 Objective Loss 0.258067 LR 0.001000 Time 0.022493 +2023-10-02 21:16:09,092 - Epoch: [98][ 550/ 1236] Overall Loss 0.258277 Objective Loss 0.258277 LR 0.001000 Time 0.022463 +2023-10-02 21:16:09,299 - Epoch: [98][ 560/ 1236] Overall Loss 0.258916 Objective Loss 0.258916 LR 0.001000 Time 0.022431 +2023-10-02 21:16:09,508 - Epoch: [98][ 570/ 1236] Overall Loss 0.258925 Objective Loss 0.258925 LR 0.001000 Time 0.022403 +2023-10-02 21:16:09,714 - Epoch: [98][ 580/ 1236] Overall Loss 0.258925 Objective Loss 0.258925 LR 0.001000 Time 0.022373 +2023-10-02 21:16:09,923 - Epoch: [98][ 590/ 1236] Overall Loss 0.258752 Objective Loss 0.258752 LR 0.001000 Time 0.022347 +2023-10-02 21:16:10,132 - Epoch: [98][ 600/ 1236] Overall Loss 0.258978 Objective Loss 0.258978 LR 0.001000 Time 0.022322 +2023-10-02 21:16:10,345 - Epoch: [98][ 610/ 1236] Overall Loss 0.259382 Objective Loss 0.259382 LR 0.001000 Time 0.022305 +2023-10-02 21:16:10,558 - Epoch: [98][ 620/ 1236] Overall Loss 0.259658 Objective Loss 0.259658 LR 0.001000 Time 0.022288 +2023-10-02 21:16:10,772 - Epoch: [98][ 630/ 1236] Overall Loss 0.259824 Objective Loss 0.259824 LR 0.001000 Time 0.022272 +2023-10-02 21:16:10,985 - Epoch: [98][ 640/ 1236] Overall Loss 0.260069 Objective Loss 0.260069 LR 0.001000 Time 0.022257 +2023-10-02 21:16:11,198 - Epoch: [98][ 650/ 1236] Overall Loss 0.259883 Objective Loss 0.259883 LR 0.001000 Time 0.022242 +2023-10-02 21:16:11,417 - Epoch: [98][ 660/ 1236] Overall Loss 0.259969 Objective Loss 0.259969 LR 0.001000 Time 0.022235 +2023-10-02 21:16:11,635 - Epoch: [98][ 670/ 1236] Overall Loss 0.259668 Objective Loss 0.259668 LR 0.001000 Time 0.022229 +2023-10-02 21:16:11,857 - Epoch: [98][ 680/ 1236] Overall Loss 0.259381 Objective Loss 0.259381 LR 0.001000 Time 0.022228 +2023-10-02 21:16:12,076 - Epoch: [98][ 690/ 1236] Overall Loss 0.259503 Objective Loss 0.259503 LR 0.001000 Time 0.022222 +2023-10-02 21:16:12,297 - Epoch: [98][ 700/ 1236] Overall Loss 0.259838 Objective Loss 0.259838 LR 0.001000 Time 0.022220 +2023-10-02 21:16:12,516 - Epoch: [98][ 710/ 1236] Overall Loss 0.259653 Objective Loss 0.259653 LR 0.001000 Time 0.022215 +2023-10-02 21:16:12,737 - Epoch: [98][ 720/ 1236] Overall Loss 0.259714 Objective Loss 0.259714 LR 0.001000 Time 0.022213 +2023-10-02 21:16:12,956 - Epoch: [98][ 730/ 1236] Overall Loss 0.260032 Objective Loss 0.260032 LR 0.001000 Time 0.022208 +2023-10-02 21:16:13,177 - Epoch: [98][ 740/ 1236] Overall Loss 0.260018 Objective Loss 0.260018 LR 0.001000 Time 0.022206 +2023-10-02 21:16:13,395 - Epoch: [98][ 750/ 1236] Overall Loss 0.260345 Objective Loss 0.260345 LR 0.001000 Time 0.022201 +2023-10-02 21:16:13,617 - Epoch: [98][ 760/ 1236] Overall Loss 0.260451 Objective Loss 0.260451 LR 0.001000 Time 0.022199 +2023-10-02 21:16:13,835 - Epoch: [98][ 770/ 1236] Overall Loss 0.260825 Objective Loss 0.260825 LR 0.001000 Time 0.022195 +2023-10-02 21:16:14,057 - Epoch: [98][ 780/ 1236] Overall Loss 0.261115 Objective Loss 0.261115 LR 0.001000 Time 0.022193 +2023-10-02 21:16:14,275 - Epoch: [98][ 790/ 1236] Overall Loss 0.261373 Objective Loss 0.261373 LR 0.001000 Time 0.022189 +2023-10-02 21:16:14,497 - Epoch: [98][ 800/ 1236] Overall Loss 0.261680 Objective Loss 0.261680 LR 0.001000 Time 0.022188 +2023-10-02 21:16:14,716 - Epoch: [98][ 810/ 1236] Overall Loss 0.261769 Objective Loss 0.261769 LR 0.001000 Time 0.022183 +2023-10-02 21:16:14,937 - Epoch: [98][ 820/ 1236] Overall Loss 0.261898 Objective Loss 0.261898 LR 0.001000 Time 0.022182 +2023-10-02 21:16:15,156 - Epoch: [98][ 830/ 1236] Overall Loss 0.261938 Objective Loss 0.261938 LR 0.001000 Time 0.022178 +2023-10-02 21:16:15,377 - Epoch: [98][ 840/ 1236] Overall Loss 0.262214 Objective Loss 0.262214 LR 0.001000 Time 0.022177 +2023-10-02 21:16:15,596 - Epoch: [98][ 850/ 1236] Overall Loss 0.262522 Objective Loss 0.262522 LR 0.001000 Time 0.022173 +2023-10-02 21:16:15,817 - Epoch: [98][ 860/ 1236] Overall Loss 0.262542 Objective Loss 0.262542 LR 0.001000 Time 0.022172 +2023-10-02 21:16:16,036 - Epoch: [98][ 870/ 1236] Overall Loss 0.262550 Objective Loss 0.262550 LR 0.001000 Time 0.022168 +2023-10-02 21:16:16,257 - Epoch: [98][ 880/ 1236] Overall Loss 0.262535 Objective Loss 0.262535 LR 0.001000 Time 0.022167 +2023-10-02 21:16:16,476 - Epoch: [98][ 890/ 1236] Overall Loss 0.263249 Objective Loss 0.263249 LR 0.001000 Time 0.022163 +2023-10-02 21:16:16,697 - Epoch: [98][ 900/ 1236] Overall Loss 0.263648 Objective Loss 0.263648 LR 0.001000 Time 0.022162 +2023-10-02 21:16:16,915 - Epoch: [98][ 910/ 1236] Overall Loss 0.263438 Objective Loss 0.263438 LR 0.001000 Time 0.022158 +2023-10-02 21:16:17,136 - Epoch: [98][ 920/ 1236] Overall Loss 0.263650 Objective Loss 0.263650 LR 0.001000 Time 0.022157 +2023-10-02 21:16:17,355 - Epoch: [98][ 930/ 1236] Overall Loss 0.264230 Objective Loss 0.264230 LR 0.001000 Time 0.022154 +2023-10-02 21:16:17,577 - Epoch: [98][ 940/ 1236] Overall Loss 0.264624 Objective Loss 0.264624 LR 0.001000 Time 0.022154 +2023-10-02 21:16:17,795 - Epoch: [98][ 950/ 1236] Overall Loss 0.264848 Objective Loss 0.264848 LR 0.001000 Time 0.022150 +2023-10-02 21:16:18,017 - Epoch: [98][ 960/ 1236] Overall Loss 0.265164 Objective Loss 0.265164 LR 0.001000 Time 0.022150 +2023-10-02 21:16:18,236 - Epoch: [98][ 970/ 1236] Overall Loss 0.264738 Objective Loss 0.264738 LR 0.001000 Time 0.022147 +2023-10-02 21:16:18,458 - Epoch: [98][ 980/ 1236] Overall Loss 0.265150 Objective Loss 0.265150 LR 0.001000 Time 0.022147 +2023-10-02 21:16:18,677 - Epoch: [98][ 990/ 1236] Overall Loss 0.265027 Objective Loss 0.265027 LR 0.001000 Time 0.022144 +2023-10-02 21:16:18,897 - Epoch: [98][ 1000/ 1236] Overall Loss 0.265047 Objective Loss 0.265047 LR 0.001000 Time 0.022142 +2023-10-02 21:16:19,116 - Epoch: [98][ 1010/ 1236] Overall Loss 0.265084 Objective Loss 0.265084 LR 0.001000 Time 0.022140 +2023-10-02 21:16:19,338 - Epoch: [98][ 1020/ 1236] Overall Loss 0.264943 Objective Loss 0.264943 LR 0.001000 Time 0.022140 +2023-10-02 21:16:19,557 - Epoch: [98][ 1030/ 1236] Overall Loss 0.265032 Objective Loss 0.265032 LR 0.001000 Time 0.022136 +2023-10-02 21:16:19,778 - Epoch: [98][ 1040/ 1236] Overall Loss 0.264899 Objective Loss 0.264899 LR 0.001000 Time 0.022136 +2023-10-02 21:16:19,997 - Epoch: [98][ 1050/ 1236] Overall Loss 0.264985 Objective Loss 0.264985 LR 0.001000 Time 0.022134 +2023-10-02 21:16:20,219 - Epoch: [98][ 1060/ 1236] Overall Loss 0.265202 Objective Loss 0.265202 LR 0.001000 Time 0.022134 +2023-10-02 21:16:20,433 - Epoch: [98][ 1070/ 1236] Overall Loss 0.265453 Objective Loss 0.265453 LR 0.001000 Time 0.022127 +2023-10-02 21:16:20,646 - Epoch: [98][ 1080/ 1236] Overall Loss 0.265775 Objective Loss 0.265775 LR 0.001000 Time 0.022118 +2023-10-02 21:16:20,861 - Epoch: [98][ 1090/ 1236] Overall Loss 0.265895 Objective Loss 0.265895 LR 0.001000 Time 0.022112 +2023-10-02 21:16:21,072 - Epoch: [98][ 1100/ 1236] Overall Loss 0.266150 Objective Loss 0.266150 LR 0.001000 Time 0.022103 +2023-10-02 21:16:21,286 - Epoch: [98][ 1110/ 1236] Overall Loss 0.266080 Objective Loss 0.266080 LR 0.001000 Time 0.022096 +2023-10-02 21:16:21,498 - Epoch: [98][ 1120/ 1236] Overall Loss 0.266506 Objective Loss 0.266506 LR 0.001000 Time 0.022088 +2023-10-02 21:16:21,712 - Epoch: [98][ 1130/ 1236] Overall Loss 0.266421 Objective Loss 0.266421 LR 0.001000 Time 0.022081 +2023-10-02 21:16:21,923 - Epoch: [98][ 1140/ 1236] Overall Loss 0.266421 Objective Loss 0.266421 LR 0.001000 Time 0.022073 +2023-10-02 21:16:22,137 - Epoch: [98][ 1150/ 1236] Overall Loss 0.266578 Objective Loss 0.266578 LR 0.001000 Time 0.022066 +2023-10-02 21:16:22,350 - Epoch: [98][ 1160/ 1236] Overall Loss 0.266507 Objective Loss 0.266507 LR 0.001000 Time 0.022059 +2023-10-02 21:16:22,562 - Epoch: [98][ 1170/ 1236] Overall Loss 0.266745 Objective Loss 0.266745 LR 0.001000 Time 0.022051 +2023-10-02 21:16:22,775 - Epoch: [98][ 1180/ 1236] Overall Loss 0.267066 Objective Loss 0.267066 LR 0.001000 Time 0.022044 +2023-10-02 21:16:22,994 - Epoch: [98][ 1190/ 1236] Overall Loss 0.267195 Objective Loss 0.267195 LR 0.001000 Time 0.022042 +2023-10-02 21:16:23,208 - Epoch: [98][ 1200/ 1236] Overall Loss 0.267549 Objective Loss 0.267549 LR 0.001000 Time 0.022037 +2023-10-02 21:16:23,427 - Epoch: [98][ 1210/ 1236] Overall Loss 0.267680 Objective Loss 0.267680 LR 0.001000 Time 0.022035 +2023-10-02 21:16:23,642 - Epoch: [98][ 1220/ 1236] Overall Loss 0.267841 Objective Loss 0.267841 LR 0.001000 Time 0.022030 +2023-10-02 21:16:23,908 - Epoch: [98][ 1230/ 1236] Overall Loss 0.267929 Objective Loss 0.267929 LR 0.001000 Time 0.022067 +2023-10-02 21:16:24,030 - Epoch: [98][ 1236/ 1236] Overall Loss 0.268104 Objective Loss 0.268104 Top1 85.947047 Top5 97.556008 LR 0.001000 Time 0.022058 +2023-10-02 21:16:24,166 - --- validate (epoch=98)----------- +2023-10-02 21:16:24,167 - 29943 samples (256 per mini-batch) +2023-10-02 21:16:24,670 - Epoch: [98][ 10/ 117] Loss 0.347223 Top1 83.476562 Top5 98.398438 +2023-10-02 21:16:24,834 - Epoch: [98][ 20/ 117] Loss 0.342815 Top1 83.378906 Top5 98.183594 +2023-10-02 21:16:24,991 - Epoch: [98][ 30/ 117] Loss 0.341029 Top1 83.580729 Top5 98.242188 +2023-10-02 21:16:25,153 - Epoch: [98][ 40/ 117] Loss 0.347431 Top1 83.515625 Top5 98.173828 +2023-10-02 21:16:25,309 - Epoch: [98][ 50/ 117] Loss 0.348458 Top1 83.195312 Top5 98.125000 +2023-10-02 21:16:25,470 - Epoch: [98][ 60/ 117] Loss 0.341973 Top1 83.190104 Top5 98.138021 +2023-10-02 21:16:25,625 - Epoch: [98][ 70/ 117] Loss 0.340834 Top1 83.152902 Top5 98.097098 +2023-10-02 21:16:25,786 - Epoch: [98][ 80/ 117] Loss 0.339662 Top1 83.295898 Top5 98.081055 +2023-10-02 21:16:25,941 - Epoch: [98][ 90/ 117] Loss 0.342929 Top1 83.246528 Top5 98.094618 +2023-10-02 21:16:26,102 - Epoch: [98][ 100/ 117] Loss 0.341856 Top1 83.250000 Top5 98.046875 +2023-10-02 21:16:26,264 - Epoch: [98][ 110/ 117] Loss 0.343995 Top1 83.178267 Top5 98.057528 +2023-10-02 21:16:26,354 - Epoch: [98][ 117/ 117] Loss 0.342740 Top1 83.144641 Top5 98.056307 +2023-10-02 21:16:26,493 - ==> Top1: 83.145 Top5: 98.056 Loss: 0.343 + +2023-10-02 21:16:26,494 - ==> Confusion: +[[ 925 2 7 0 7 3 0 1 10 61 2 0 1 1 7 1 4 0 1 0 17] + [ 0 1025 1 2 8 28 2 30 0 0 2 0 0 0 3 4 1 0 14 5 6] + [ 3 1 970 14 0 0 17 9 0 2 5 1 7 2 4 4 1 0 7 2 7] + [ 0 3 17 950 0 9 1 5 8 0 9 0 9 2 35 1 1 0 20 0 19] + [ 17 13 4 0 957 9 0 0 2 15 2 1 2 0 10 4 10 0 1 1 2] + [ 3 30 1 1 1 1004 1 26 3 5 4 3 4 7 4 1 1 0 3 1 13] + [ 1 2 35 2 0 1 1113 8 0 0 7 2 1 0 1 4 0 0 2 6 6] + [ 1 13 16 0 6 28 4 1070 1 3 8 5 2 4 2 1 3 2 39 2 8] + [ 15 4 0 0 0 1 0 1 969 41 13 4 6 10 14 0 2 1 6 1 1] + [ 97 2 0 1 6 3 1 1 32 935 2 0 0 17 11 1 0 1 0 3 6] + [ 0 2 6 6 1 2 3 6 11 1 975 1 0 11 7 0 2 2 10 1 6] + [ 0 0 2 0 1 12 0 2 0 2 2 929 55 6 0 0 0 12 0 7 5] + [ 1 1 3 0 0 1 3 3 2 0 2 37 968 4 7 6 2 6 7 7 8] + [ 0 0 1 0 0 16 2 0 17 15 10 5 2 1029 5 0 3 1 0 0 13] + [ 9 1 5 14 5 2 0 0 30 4 4 0 3 2 998 0 0 0 16 0 8] + [ 0 1 1 3 3 1 2 0 0 0 1 8 9 0 0 1061 20 9 2 8 5] + [ 0 22 0 0 6 11 0 0 2 0 0 2 3 0 4 9 1081 0 0 11 10] + [ 0 0 1 4 0 0 2 0 4 1 0 4 40 2 3 8 3 961 1 0 4] + [ 1 4 5 15 1 0 0 17 7 1 7 1 1 0 16 0 0 0 979 0 13] + [ 0 1 5 1 2 7 12 19 0 0 5 13 5 3 0 3 3 0 4 1060 9] + [ 144 169 183 80 94 193 59 122 120 99 211 96 394 304 154 43 111 49 164 179 4937]] + +2023-10-02 21:16:26,495 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:16:26,495 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:16:26,501 - + +2023-10-02 21:16:26,501 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:16:27,551 - Epoch: [99][ 10/ 1236] Overall Loss 0.260945 Objective Loss 0.260945 LR 0.001000 Time 0.104901 +2023-10-02 21:16:27,760 - Epoch: [99][ 20/ 1236] Overall Loss 0.257438 Objective Loss 0.257438 LR 0.001000 Time 0.062917 +2023-10-02 21:16:27,968 - Epoch: [99][ 30/ 1236] Overall Loss 0.258865 Objective Loss 0.258865 LR 0.001000 Time 0.048819 +2023-10-02 21:16:28,178 - Epoch: [99][ 40/ 1236] Overall Loss 0.256146 Objective Loss 0.256146 LR 0.001000 Time 0.041865 +2023-10-02 21:16:28,385 - Epoch: [99][ 50/ 1236] Overall Loss 0.257775 Objective Loss 0.257775 LR 0.001000 Time 0.037618 +2023-10-02 21:16:28,595 - Epoch: [99][ 60/ 1236] Overall Loss 0.259139 Objective Loss 0.259139 LR 0.001000 Time 0.034851 +2023-10-02 21:16:28,802 - Epoch: [99][ 70/ 1236] Overall Loss 0.260519 Objective Loss 0.260519 LR 0.001000 Time 0.032817 +2023-10-02 21:16:29,012 - Epoch: [99][ 80/ 1236] Overall Loss 0.264382 Objective Loss 0.264382 LR 0.001000 Time 0.031344 +2023-10-02 21:16:29,219 - Epoch: [99][ 90/ 1236] Overall Loss 0.262392 Objective Loss 0.262392 LR 0.001000 Time 0.030154 +2023-10-02 21:16:29,429 - Epoch: [99][ 100/ 1236] Overall Loss 0.261820 Objective Loss 0.261820 LR 0.001000 Time 0.029239 +2023-10-02 21:16:29,635 - Epoch: [99][ 110/ 1236] Overall Loss 0.262536 Objective Loss 0.262536 LR 0.001000 Time 0.028447 +2023-10-02 21:16:29,844 - Epoch: [99][ 120/ 1236] Overall Loss 0.262738 Objective Loss 0.262738 LR 0.001000 Time 0.027816 +2023-10-02 21:16:30,051 - Epoch: [99][ 130/ 1236] Overall Loss 0.264861 Objective Loss 0.264861 LR 0.001000 Time 0.027257 +2023-10-02 21:16:30,260 - Epoch: [99][ 140/ 1236] Overall Loss 0.265259 Objective Loss 0.265259 LR 0.001000 Time 0.026803 +2023-10-02 21:16:30,467 - Epoch: [99][ 150/ 1236] Overall Loss 0.266622 Objective Loss 0.266622 LR 0.001000 Time 0.026388 +2023-10-02 21:16:30,678 - Epoch: [99][ 160/ 1236] Overall Loss 0.266925 Objective Loss 0.266925 LR 0.001000 Time 0.026053 +2023-10-02 21:16:30,888 - Epoch: [99][ 170/ 1236] Overall Loss 0.266587 Objective Loss 0.266587 LR 0.001000 Time 0.025746 +2023-10-02 21:16:31,099 - Epoch: [99][ 180/ 1236] Overall Loss 0.266601 Objective Loss 0.266601 LR 0.001000 Time 0.025487 +2023-10-02 21:16:31,305 - Epoch: [99][ 190/ 1236] Overall Loss 0.264975 Objective Loss 0.264975 LR 0.001000 Time 0.025230 +2023-10-02 21:16:31,516 - Epoch: [99][ 200/ 1236] Overall Loss 0.265286 Objective Loss 0.265286 LR 0.001000 Time 0.025020 +2023-10-02 21:16:31,723 - Epoch: [99][ 210/ 1236] Overall Loss 0.267982 Objective Loss 0.267982 LR 0.001000 Time 0.024810 +2023-10-02 21:16:31,930 - Epoch: [99][ 220/ 1236] Overall Loss 0.268170 Objective Loss 0.268170 LR 0.001000 Time 0.024622 +2023-10-02 21:16:32,137 - Epoch: [99][ 230/ 1236] Overall Loss 0.267565 Objective Loss 0.267565 LR 0.001000 Time 0.024442 +2023-10-02 21:16:32,345 - Epoch: [99][ 240/ 1236] Overall Loss 0.268017 Objective Loss 0.268017 LR 0.001000 Time 0.024289 +2023-10-02 21:16:32,554 - Epoch: [99][ 250/ 1236] Overall Loss 0.268163 Objective Loss 0.268163 LR 0.001000 Time 0.024150 +2023-10-02 21:16:32,765 - Epoch: [99][ 260/ 1236] Overall Loss 0.269791 Objective Loss 0.269791 LR 0.001000 Time 0.024031 +2023-10-02 21:16:32,974 - Epoch: [99][ 270/ 1236] Overall Loss 0.270054 Objective Loss 0.270054 LR 0.001000 Time 0.023913 +2023-10-02 21:16:33,182 - Epoch: [99][ 280/ 1236] Overall Loss 0.270983 Objective Loss 0.270983 LR 0.001000 Time 0.023800 +2023-10-02 21:16:33,389 - Epoch: [99][ 290/ 1236] Overall Loss 0.269573 Objective Loss 0.269573 LR 0.001000 Time 0.023694 +2023-10-02 21:16:33,598 - Epoch: [99][ 300/ 1236] Overall Loss 0.269236 Objective Loss 0.269236 LR 0.001000 Time 0.023598 +2023-10-02 21:16:33,805 - Epoch: [99][ 310/ 1236] Overall Loss 0.269541 Objective Loss 0.269541 LR 0.001000 Time 0.023502 +2023-10-02 21:16:34,013 - Epoch: [99][ 320/ 1236] Overall Loss 0.268752 Objective Loss 0.268752 LR 0.001000 Time 0.023417 +2023-10-02 21:16:34,222 - Epoch: [99][ 330/ 1236] Overall Loss 0.268530 Objective Loss 0.268530 LR 0.001000 Time 0.023335 +2023-10-02 21:16:34,432 - Epoch: [99][ 340/ 1236] Overall Loss 0.268889 Objective Loss 0.268889 LR 0.001000 Time 0.023265 +2023-10-02 21:16:34,640 - Epoch: [99][ 350/ 1236] Overall Loss 0.268719 Objective Loss 0.268719 LR 0.001000 Time 0.023193 +2023-10-02 21:16:34,850 - Epoch: [99][ 360/ 1236] Overall Loss 0.268948 Objective Loss 0.268948 LR 0.001000 Time 0.023133 +2023-10-02 21:16:35,058 - Epoch: [99][ 370/ 1236] Overall Loss 0.268173 Objective Loss 0.268173 LR 0.001000 Time 0.023068 +2023-10-02 21:16:35,268 - Epoch: [99][ 380/ 1236] Overall Loss 0.268090 Objective Loss 0.268090 LR 0.001000 Time 0.023014 +2023-10-02 21:16:35,475 - Epoch: [99][ 390/ 1236] Overall Loss 0.268095 Objective Loss 0.268095 LR 0.001000 Time 0.022954 +2023-10-02 21:16:35,685 - Epoch: [99][ 400/ 1236] Overall Loss 0.267781 Objective Loss 0.267781 LR 0.001000 Time 0.022905 +2023-10-02 21:16:35,893 - Epoch: [99][ 410/ 1236] Overall Loss 0.268394 Objective Loss 0.268394 LR 0.001000 Time 0.022852 +2023-10-02 21:16:36,103 - Epoch: [99][ 420/ 1236] Overall Loss 0.268495 Objective Loss 0.268495 LR 0.001000 Time 0.022807 +2023-10-02 21:16:36,312 - Epoch: [99][ 430/ 1236] Overall Loss 0.268381 Objective Loss 0.268381 LR 0.001000 Time 0.022762 +2023-10-02 21:16:36,523 - Epoch: [99][ 440/ 1236] Overall Loss 0.268793 Objective Loss 0.268793 LR 0.001000 Time 0.022724 +2023-10-02 21:16:36,732 - Epoch: [99][ 450/ 1236] Overall Loss 0.269141 Objective Loss 0.269141 LR 0.001000 Time 0.022681 +2023-10-02 21:16:36,942 - Epoch: [99][ 460/ 1236] Overall Loss 0.269177 Objective Loss 0.269177 LR 0.001000 Time 0.022643 +2023-10-02 21:16:37,151 - Epoch: [99][ 470/ 1236] Overall Loss 0.269116 Objective Loss 0.269116 LR 0.001000 Time 0.022606 +2023-10-02 21:16:37,361 - Epoch: [99][ 480/ 1236] Overall Loss 0.269602 Objective Loss 0.269602 LR 0.001000 Time 0.022573 +2023-10-02 21:16:37,570 - Epoch: [99][ 490/ 1236] Overall Loss 0.269232 Objective Loss 0.269232 LR 0.001000 Time 0.022538 +2023-10-02 21:16:37,782 - Epoch: [99][ 500/ 1236] Overall Loss 0.268765 Objective Loss 0.268765 LR 0.001000 Time 0.022509 +2023-10-02 21:16:37,989 - Epoch: [99][ 510/ 1236] Overall Loss 0.268940 Objective Loss 0.268940 LR 0.001000 Time 0.022474 +2023-10-02 21:16:38,199 - Epoch: [99][ 520/ 1236] Overall Loss 0.268655 Objective Loss 0.268655 LR 0.001000 Time 0.022445 +2023-10-02 21:16:38,409 - Epoch: [99][ 530/ 1236] Overall Loss 0.268656 Objective Loss 0.268656 LR 0.001000 Time 0.022414 +2023-10-02 21:16:38,619 - Epoch: [99][ 540/ 1236] Overall Loss 0.269016 Objective Loss 0.269016 LR 0.001000 Time 0.022387 +2023-10-02 21:16:38,827 - Epoch: [99][ 550/ 1236] Overall Loss 0.268871 Objective Loss 0.268871 LR 0.001000 Time 0.022357 +2023-10-02 21:16:39,037 - Epoch: [99][ 560/ 1236] Overall Loss 0.268963 Objective Loss 0.268963 LR 0.001000 Time 0.022332 +2023-10-02 21:16:39,246 - Epoch: [99][ 570/ 1236] Overall Loss 0.269174 Objective Loss 0.269174 LR 0.001000 Time 0.022306 +2023-10-02 21:16:39,456 - Epoch: [99][ 580/ 1236] Overall Loss 0.269691 Objective Loss 0.269691 LR 0.001000 Time 0.022283 +2023-10-02 21:16:39,665 - Epoch: [99][ 590/ 1236] Overall Loss 0.269520 Objective Loss 0.269520 LR 0.001000 Time 0.022259 +2023-10-02 21:16:39,875 - Epoch: [99][ 600/ 1236] Overall Loss 0.269156 Objective Loss 0.269156 LR 0.001000 Time 0.022237 +2023-10-02 21:16:40,084 - Epoch: [99][ 610/ 1236] Overall Loss 0.269032 Objective Loss 0.269032 LR 0.001000 Time 0.022215 +2023-10-02 21:16:40,295 - Epoch: [99][ 620/ 1236] Overall Loss 0.269161 Objective Loss 0.269161 LR 0.001000 Time 0.022196 +2023-10-02 21:16:40,504 - Epoch: [99][ 630/ 1236] Overall Loss 0.269082 Objective Loss 0.269082 LR 0.001000 Time 0.022175 +2023-10-02 21:16:40,715 - Epoch: [99][ 640/ 1236] Overall Loss 0.268985 Objective Loss 0.268985 LR 0.001000 Time 0.022159 +2023-10-02 21:16:40,923 - Epoch: [99][ 650/ 1236] Overall Loss 0.268898 Objective Loss 0.268898 LR 0.001000 Time 0.022137 +2023-10-02 21:16:41,134 - Epoch: [99][ 660/ 1236] Overall Loss 0.267580 Objective Loss 0.267580 LR 0.001000 Time 0.022121 +2023-10-02 21:16:41,342 - Epoch: [99][ 670/ 1236] Overall Loss 0.267725 Objective Loss 0.267725 LR 0.001000 Time 0.022100 +2023-10-02 21:16:41,552 - Epoch: [99][ 680/ 1236] Overall Loss 0.268031 Objective Loss 0.268031 LR 0.001000 Time 0.022084 +2023-10-02 21:16:41,761 - Epoch: [99][ 690/ 1236] Overall Loss 0.268103 Objective Loss 0.268103 LR 0.001000 Time 0.022066 +2023-10-02 21:16:41,971 - Epoch: [99][ 700/ 1236] Overall Loss 0.268253 Objective Loss 0.268253 LR 0.001000 Time 0.022051 +2023-10-02 21:16:42,180 - Epoch: [99][ 710/ 1236] Overall Loss 0.268517 Objective Loss 0.268517 LR 0.001000 Time 0.022035 +2023-10-02 21:16:42,390 - Epoch: [99][ 720/ 1236] Overall Loss 0.268525 Objective Loss 0.268525 LR 0.001000 Time 0.022020 +2023-10-02 21:16:42,599 - Epoch: [99][ 730/ 1236] Overall Loss 0.268750 Objective Loss 0.268750 LR 0.001000 Time 0.022004 +2023-10-02 21:16:42,809 - Epoch: [99][ 740/ 1236] Overall Loss 0.269297 Objective Loss 0.269297 LR 0.001000 Time 0.021990 +2023-10-02 21:16:43,018 - Epoch: [99][ 750/ 1236] Overall Loss 0.269240 Objective Loss 0.269240 LR 0.001000 Time 0.021976 +2023-10-02 21:16:43,228 - Epoch: [99][ 760/ 1236] Overall Loss 0.269988 Objective Loss 0.269988 LR 0.001000 Time 0.021962 +2023-10-02 21:16:43,437 - Epoch: [99][ 770/ 1236] Overall Loss 0.270478 Objective Loss 0.270478 LR 0.001000 Time 0.021948 +2023-10-02 21:16:43,647 - Epoch: [99][ 780/ 1236] Overall Loss 0.270638 Objective Loss 0.270638 LR 0.001000 Time 0.021935 +2023-10-02 21:16:43,856 - Epoch: [99][ 790/ 1236] Overall Loss 0.271085 Objective Loss 0.271085 LR 0.001000 Time 0.021922 +2023-10-02 21:16:44,066 - Epoch: [99][ 800/ 1236] Overall Loss 0.271521 Objective Loss 0.271521 LR 0.001000 Time 0.021910 +2023-10-02 21:16:44,275 - Epoch: [99][ 810/ 1236] Overall Loss 0.271633 Objective Loss 0.271633 LR 0.001000 Time 0.021897 +2023-10-02 21:16:44,485 - Epoch: [99][ 820/ 1236] Overall Loss 0.272078 Objective Loss 0.272078 LR 0.001000 Time 0.021886 +2023-10-02 21:16:44,694 - Epoch: [99][ 830/ 1236] Overall Loss 0.272463 Objective Loss 0.272463 LR 0.001000 Time 0.021873 +2023-10-02 21:16:44,904 - Epoch: [99][ 840/ 1236] Overall Loss 0.272519 Objective Loss 0.272519 LR 0.001000 Time 0.021863 +2023-10-02 21:16:45,113 - Epoch: [99][ 850/ 1236] Overall Loss 0.272595 Objective Loss 0.272595 LR 0.001000 Time 0.021851 +2023-10-02 21:16:45,323 - Epoch: [99][ 860/ 1236] Overall Loss 0.272427 Objective Loss 0.272427 LR 0.001000 Time 0.021841 +2023-10-02 21:16:45,532 - Epoch: [99][ 870/ 1236] Overall Loss 0.272713 Objective Loss 0.272713 LR 0.001000 Time 0.021829 +2023-10-02 21:16:45,741 - Epoch: [99][ 880/ 1236] Overall Loss 0.273019 Objective Loss 0.273019 LR 0.001000 Time 0.021819 +2023-10-02 21:16:45,950 - Epoch: [99][ 890/ 1236] Overall Loss 0.273483 Objective Loss 0.273483 LR 0.001000 Time 0.021808 +2023-10-02 21:16:46,160 - Epoch: [99][ 900/ 1236] Overall Loss 0.273610 Objective Loss 0.273610 LR 0.001000 Time 0.021799 +2023-10-02 21:16:46,369 - Epoch: [99][ 910/ 1236] Overall Loss 0.273867 Objective Loss 0.273867 LR 0.001000 Time 0.021789 +2023-10-02 21:16:46,579 - Epoch: [99][ 920/ 1236] Overall Loss 0.273843 Objective Loss 0.273843 LR 0.001000 Time 0.021780 +2023-10-02 21:16:46,789 - Epoch: [99][ 930/ 1236] Overall Loss 0.273966 Objective Loss 0.273966 LR 0.001000 Time 0.021771 +2023-10-02 21:16:46,999 - Epoch: [99][ 940/ 1236] Overall Loss 0.274215 Objective Loss 0.274215 LR 0.001000 Time 0.021762 +2023-10-02 21:16:47,208 - Epoch: [99][ 950/ 1236] Overall Loss 0.273966 Objective Loss 0.273966 LR 0.001000 Time 0.021753 +2023-10-02 21:16:47,418 - Epoch: [99][ 960/ 1236] Overall Loss 0.273917 Objective Loss 0.273917 LR 0.001000 Time 0.021745 +2023-10-02 21:16:47,627 - Epoch: [99][ 970/ 1236] Overall Loss 0.274122 Objective Loss 0.274122 LR 0.001000 Time 0.021736 +2023-10-02 21:16:47,837 - Epoch: [99][ 980/ 1236] Overall Loss 0.274129 Objective Loss 0.274129 LR 0.001000 Time 0.021728 +2023-10-02 21:16:48,046 - Epoch: [99][ 990/ 1236] Overall Loss 0.274094 Objective Loss 0.274094 LR 0.001000 Time 0.021720 +2023-10-02 21:16:48,257 - Epoch: [99][ 1000/ 1236] Overall Loss 0.274333 Objective Loss 0.274333 LR 0.001000 Time 0.021714 +2023-10-02 21:16:48,465 - Epoch: [99][ 1010/ 1236] Overall Loss 0.274358 Objective Loss 0.274358 LR 0.001000 Time 0.021704 +2023-10-02 21:16:48,675 - Epoch: [99][ 1020/ 1236] Overall Loss 0.274364 Objective Loss 0.274364 LR 0.001000 Time 0.021697 +2023-10-02 21:16:48,884 - Epoch: [99][ 1030/ 1236] Overall Loss 0.274470 Objective Loss 0.274470 LR 0.001000 Time 0.021689 +2023-10-02 21:16:49,094 - Epoch: [99][ 1040/ 1236] Overall Loss 0.274657 Objective Loss 0.274657 LR 0.001000 Time 0.021682 +2023-10-02 21:16:49,304 - Epoch: [99][ 1050/ 1236] Overall Loss 0.274462 Objective Loss 0.274462 LR 0.001000 Time 0.021675 +2023-10-02 21:16:49,514 - Epoch: [99][ 1060/ 1236] Overall Loss 0.274507 Objective Loss 0.274507 LR 0.001000 Time 0.021668 +2023-10-02 21:16:49,723 - Epoch: [99][ 1070/ 1236] Overall Loss 0.274648 Objective Loss 0.274648 LR 0.001000 Time 0.021661 +2023-10-02 21:16:49,932 - Epoch: [99][ 1080/ 1236] Overall Loss 0.274744 Objective Loss 0.274744 LR 0.001000 Time 0.021654 +2023-10-02 21:16:50,141 - Epoch: [99][ 1090/ 1236] Overall Loss 0.274749 Objective Loss 0.274749 LR 0.001000 Time 0.021647 +2023-10-02 21:16:50,351 - Epoch: [99][ 1100/ 1236] Overall Loss 0.274899 Objective Loss 0.274899 LR 0.001000 Time 0.021641 +2023-10-02 21:16:50,560 - Epoch: [99][ 1110/ 1236] Overall Loss 0.274772 Objective Loss 0.274772 LR 0.001000 Time 0.021634 +2023-10-02 21:16:50,770 - Epoch: [99][ 1120/ 1236] Overall Loss 0.274820 Objective Loss 0.274820 LR 0.001000 Time 0.021628 +2023-10-02 21:16:50,980 - Epoch: [99][ 1130/ 1236] Overall Loss 0.274615 Objective Loss 0.274615 LR 0.001000 Time 0.021622 +2023-10-02 21:16:51,190 - Epoch: [99][ 1140/ 1236] Overall Loss 0.274378 Objective Loss 0.274378 LR 0.001000 Time 0.021616 +2023-10-02 21:16:51,399 - Epoch: [99][ 1150/ 1236] Overall Loss 0.274231 Objective Loss 0.274231 LR 0.001000 Time 0.021610 +2023-10-02 21:16:51,609 - Epoch: [99][ 1160/ 1236] Overall Loss 0.274183 Objective Loss 0.274183 LR 0.001000 Time 0.021605 +2023-10-02 21:16:51,819 - Epoch: [99][ 1170/ 1236] Overall Loss 0.274092 Objective Loss 0.274092 LR 0.001000 Time 0.021599 +2023-10-02 21:16:52,030 - Epoch: [99][ 1180/ 1236] Overall Loss 0.274103 Objective Loss 0.274103 LR 0.001000 Time 0.021594 +2023-10-02 21:16:52,238 - Epoch: [99][ 1190/ 1236] Overall Loss 0.274227 Objective Loss 0.274227 LR 0.001000 Time 0.021587 +2023-10-02 21:16:52,448 - Epoch: [99][ 1200/ 1236] Overall Loss 0.274303 Objective Loss 0.274303 LR 0.001000 Time 0.021582 +2023-10-02 21:16:52,657 - Epoch: [99][ 1210/ 1236] Overall Loss 0.274598 Objective Loss 0.274598 LR 0.001000 Time 0.021577 +2023-10-02 21:16:52,867 - Epoch: [99][ 1220/ 1236] Overall Loss 0.274612 Objective Loss 0.274612 LR 0.001000 Time 0.021572 +2023-10-02 21:16:53,130 - Epoch: [99][ 1230/ 1236] Overall Loss 0.274624 Objective Loss 0.274624 LR 0.001000 Time 0.021610 +2023-10-02 21:16:53,253 - Epoch: [99][ 1236/ 1236] Overall Loss 0.274594 Objective Loss 0.274594 Top1 82.892057 Top5 98.574338 LR 0.001000 Time 0.021605 +2023-10-02 21:16:53,399 - --- validate (epoch=99)----------- +2023-10-02 21:16:53,400 - 29943 samples (256 per mini-batch) +2023-10-02 21:16:53,896 - Epoch: [99][ 10/ 117] Loss 0.350033 Top1 82.968750 Top5 97.812500 +2023-10-02 21:16:54,050 - Epoch: [99][ 20/ 117] Loss 0.359878 Top1 82.500000 Top5 97.773438 +2023-10-02 21:16:54,202 - Epoch: [99][ 30/ 117] Loss 0.361122 Top1 82.161458 Top5 97.864583 +2023-10-02 21:16:54,355 - Epoch: [99][ 40/ 117] Loss 0.349230 Top1 82.382812 Top5 97.978516 +2023-10-02 21:16:54,508 - Epoch: [99][ 50/ 117] Loss 0.354730 Top1 82.226562 Top5 97.921875 +2023-10-02 21:16:54,659 - Epoch: [99][ 60/ 117] Loss 0.355259 Top1 82.233073 Top5 97.845052 +2023-10-02 21:16:54,810 - Epoch: [99][ 70/ 117] Loss 0.355847 Top1 82.243304 Top5 97.834821 +2023-10-02 21:16:54,961 - Epoch: [99][ 80/ 117] Loss 0.356136 Top1 82.143555 Top5 97.817383 +2023-10-02 21:16:55,112 - Epoch: [99][ 90/ 117] Loss 0.357688 Top1 82.126736 Top5 97.855903 +2023-10-02 21:16:55,263 - Epoch: [99][ 100/ 117] Loss 0.353230 Top1 82.128906 Top5 97.898438 +2023-10-02 21:16:55,422 - Epoch: [99][ 110/ 117] Loss 0.353317 Top1 82.159091 Top5 97.915483 +2023-10-02 21:16:55,512 - Epoch: [99][ 117/ 117] Loss 0.352597 Top1 82.269646 Top5 97.922720 +2023-10-02 21:16:55,650 - ==> Top1: 82.270 Top5: 97.923 Loss: 0.353 + +2023-10-02 21:16:55,651 - ==> Confusion: +[[ 916 4 5 2 8 3 0 0 7 64 2 0 2 3 6 1 8 3 2 0 14] + [ 0 998 2 1 7 35 0 46 1 1 2 2 0 0 2 4 6 0 15 4 5] + [ 2 0 950 14 1 0 28 13 0 1 6 0 6 2 0 3 4 1 12 5 8] + [ 1 3 17 966 0 3 1 5 4 0 9 0 12 3 23 3 0 8 14 2 15] + [ 30 8 2 0 940 9 0 1 3 14 0 1 1 2 8 5 16 1 4 2 3] + [ 4 25 1 2 5 946 1 43 3 3 2 11 6 22 4 1 5 1 8 12 11] + [ 0 0 25 0 0 0 1119 8 0 0 8 3 1 0 0 6 1 0 2 15 3] + [ 4 11 14 1 5 27 6 1054 1 2 8 10 5 5 2 1 1 2 37 18 4] + [ 19 4 0 1 3 1 0 1 957 48 10 5 4 9 18 1 2 1 3 2 0] + [ 117 0 1 1 8 4 1 1 43 904 0 0 0 19 6 3 1 4 0 0 6] + [ 0 1 7 13 2 0 3 3 21 1 950 1 1 19 5 0 2 2 10 2 10] + [ 1 1 2 0 0 8 1 2 0 0 0 952 35 8 0 2 1 15 0 6 1] + [ 0 0 0 6 0 1 2 2 0 2 3 48 953 1 1 5 1 26 4 6 7] + [ 0 1 2 0 6 5 0 0 11 8 8 8 3 1050 6 1 1 1 1 2 5] + [ 11 4 3 32 5 0 0 0 21 2 2 1 6 4 983 1 1 0 19 0 6] + [ 0 0 1 1 5 0 1 0 0 2 2 8 11 1 1 1058 19 13 1 6 4] + [ 0 17 0 0 7 7 1 2 1 0 0 9 1 2 4 9 1085 2 0 4 10] + [ 0 1 1 3 1 0 9 0 0 0 0 2 21 0 0 5 0 992 1 0 2] + [ 1 2 3 17 0 1 1 13 4 0 6 0 6 0 10 0 1 0 996 1 6] + [ 0 3 4 1 0 2 5 7 0 0 1 19 6 4 0 2 4 0 1 1082 11] + [ 126 165 147 116 82 166 54 132 95 111 206 147 406 325 126 60 122 116 219 201 4783]] + +2023-10-02 21:16:55,652 - ==> Best [Top1: 84.290 Top5: 98.313 Sparsity:0.00 Params: 169472 on epoch: 46] +2023-10-02 21:16:55,652 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:16:55,658 - + +2023-10-02 21:16:55,659 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:16:56,677 - Epoch: [100][ 10/ 1236] Overall Loss 0.271902 Objective Loss 0.271902 LR 0.000500 Time 0.101790 +2023-10-02 21:16:56,886 - Epoch: [100][ 20/ 1236] Overall Loss 0.255857 Objective Loss 0.255857 LR 0.000500 Time 0.061320 +2023-10-02 21:16:57,094 - Epoch: [100][ 30/ 1236] Overall Loss 0.257613 Objective Loss 0.257613 LR 0.000500 Time 0.047768 +2023-10-02 21:16:57,303 - Epoch: [100][ 40/ 1236] Overall Loss 0.258402 Objective Loss 0.258402 LR 0.000500 Time 0.041042 +2023-10-02 21:16:57,511 - Epoch: [100][ 50/ 1236] Overall Loss 0.255655 Objective Loss 0.255655 LR 0.000500 Time 0.036965 +2023-10-02 21:16:57,721 - Epoch: [100][ 60/ 1236] Overall Loss 0.252014 Objective Loss 0.252014 LR 0.000500 Time 0.034306 +2023-10-02 21:16:57,928 - Epoch: [100][ 70/ 1236] Overall Loss 0.253788 Objective Loss 0.253788 LR 0.000500 Time 0.032353 +2023-10-02 21:16:58,137 - Epoch: [100][ 80/ 1236] Overall Loss 0.253432 Objective Loss 0.253432 LR 0.000500 Time 0.030920 +2023-10-02 21:16:58,345 - Epoch: [100][ 90/ 1236] Overall Loss 0.253865 Objective Loss 0.253865 LR 0.000500 Time 0.029777 +2023-10-02 21:16:58,554 - Epoch: [100][ 100/ 1236] Overall Loss 0.251728 Objective Loss 0.251728 LR 0.000500 Time 0.028890 +2023-10-02 21:16:58,762 - Epoch: [100][ 110/ 1236] Overall Loss 0.253635 Objective Loss 0.253635 LR 0.000500 Time 0.028153 +2023-10-02 21:16:58,972 - Epoch: [100][ 120/ 1236] Overall Loss 0.256113 Objective Loss 0.256113 LR 0.000500 Time 0.027548 +2023-10-02 21:16:59,180 - Epoch: [100][ 130/ 1236] Overall Loss 0.254705 Objective Loss 0.254705 LR 0.000500 Time 0.027018 +2023-10-02 21:16:59,388 - Epoch: [100][ 140/ 1236] Overall Loss 0.255412 Objective Loss 0.255412 LR 0.000500 Time 0.026577 +2023-10-02 21:16:59,596 - Epoch: [100][ 150/ 1236] Overall Loss 0.254622 Objective Loss 0.254622 LR 0.000500 Time 0.026181 +2023-10-02 21:16:59,804 - Epoch: [100][ 160/ 1236] Overall Loss 0.253229 Objective Loss 0.253229 LR 0.000500 Time 0.025842 +2023-10-02 21:17:00,010 - Epoch: [100][ 170/ 1236] Overall Loss 0.251743 Objective Loss 0.251743 LR 0.000500 Time 0.025533 +2023-10-02 21:17:00,219 - Epoch: [100][ 180/ 1236] Overall Loss 0.249600 Objective Loss 0.249600 LR 0.000500 Time 0.025275 +2023-10-02 21:17:00,425 - Epoch: [100][ 190/ 1236] Overall Loss 0.247615 Objective Loss 0.247615 LR 0.000500 Time 0.025025 +2023-10-02 21:17:00,633 - Epoch: [100][ 200/ 1236] Overall Loss 0.247339 Objective Loss 0.247339 LR 0.000500 Time 0.024813 +2023-10-02 21:17:00,840 - Epoch: [100][ 210/ 1236] Overall Loss 0.246920 Objective Loss 0.246920 LR 0.000500 Time 0.024612 +2023-10-02 21:17:01,049 - Epoch: [100][ 220/ 1236] Overall Loss 0.246450 Objective Loss 0.246450 LR 0.000500 Time 0.024440 +2023-10-02 21:17:01,256 - Epoch: [100][ 230/ 1236] Overall Loss 0.245419 Objective Loss 0.245419 LR 0.000500 Time 0.024272 +2023-10-02 21:17:01,463 - Epoch: [100][ 240/ 1236] Overall Loss 0.244421 Objective Loss 0.244421 LR 0.000500 Time 0.024122 +2023-10-02 21:17:01,671 - Epoch: [100][ 250/ 1236] Overall Loss 0.243720 Objective Loss 0.243720 LR 0.000500 Time 0.023982 +2023-10-02 21:17:01,880 - Epoch: [100][ 260/ 1236] Overall Loss 0.243470 Objective Loss 0.243470 LR 0.000500 Time 0.023864 +2023-10-02 21:17:02,087 - Epoch: [100][ 270/ 1236] Overall Loss 0.242544 Objective Loss 0.242544 LR 0.000500 Time 0.023746 +2023-10-02 21:17:02,297 - Epoch: [100][ 280/ 1236] Overall Loss 0.242118 Objective Loss 0.242118 LR 0.000500 Time 0.023645 +2023-10-02 21:17:02,506 - Epoch: [100][ 290/ 1236] Overall Loss 0.241174 Objective Loss 0.241174 LR 0.000500 Time 0.023550 +2023-10-02 21:17:02,718 - Epoch: [100][ 300/ 1236] Overall Loss 0.240967 Objective Loss 0.240967 LR 0.000500 Time 0.023470 +2023-10-02 21:17:02,926 - Epoch: [100][ 310/ 1236] Overall Loss 0.240847 Objective Loss 0.240847 LR 0.000500 Time 0.023383 +2023-10-02 21:17:03,136 - Epoch: [100][ 320/ 1236] Overall Loss 0.239930 Objective Loss 0.239930 LR 0.000500 Time 0.023309 +2023-10-02 21:17:03,345 - Epoch: [100][ 330/ 1236] Overall Loss 0.240739 Objective Loss 0.240739 LR 0.000500 Time 0.023236 +2023-10-02 21:17:03,558 - Epoch: [100][ 340/ 1236] Overall Loss 0.241820 Objective Loss 0.241820 LR 0.000500 Time 0.023175 +2023-10-02 21:17:03,765 - Epoch: [100][ 350/ 1236] Overall Loss 0.242086 Objective Loss 0.242086 LR 0.000500 Time 0.023106 +2023-10-02 21:17:03,975 - Epoch: [100][ 360/ 1236] Overall Loss 0.241839 Objective Loss 0.241839 LR 0.000500 Time 0.023046 +2023-10-02 21:17:04,183 - Epoch: [100][ 370/ 1236] Overall Loss 0.241017 Objective Loss 0.241017 LR 0.000500 Time 0.022984 +2023-10-02 21:17:04,394 - Epoch: [100][ 380/ 1236] Overall Loss 0.242112 Objective Loss 0.242112 LR 0.000500 Time 0.022934 +2023-10-02 21:17:04,601 - Epoch: [100][ 390/ 1236] Overall Loss 0.241634 Objective Loss 0.241634 LR 0.000500 Time 0.022875 +2023-10-02 21:17:04,810 - Epoch: [100][ 400/ 1236] Overall Loss 0.242093 Objective Loss 0.242093 LR 0.000500 Time 0.022827 +2023-10-02 21:17:05,018 - Epoch: [100][ 410/ 1236] Overall Loss 0.242449 Objective Loss 0.242449 LR 0.000500 Time 0.022773 +2023-10-02 21:17:05,229 - Epoch: [100][ 420/ 1236] Overall Loss 0.242342 Objective Loss 0.242342 LR 0.000500 Time 0.022733 +2023-10-02 21:17:05,436 - Epoch: [100][ 430/ 1236] Overall Loss 0.242172 Objective Loss 0.242172 LR 0.000500 Time 0.022684 +2023-10-02 21:17:05,647 - Epoch: [100][ 440/ 1236] Overall Loss 0.242025 Objective Loss 0.242025 LR 0.000500 Time 0.022648 +2023-10-02 21:17:05,853 - Epoch: [100][ 450/ 1236] Overall Loss 0.241974 Objective Loss 0.241974 LR 0.000500 Time 0.022603 +2023-10-02 21:17:06,063 - Epoch: [100][ 460/ 1236] Overall Loss 0.241514 Objective Loss 0.241514 LR 0.000500 Time 0.022567 +2023-10-02 21:17:06,271 - Epoch: [100][ 470/ 1236] Overall Loss 0.241368 Objective Loss 0.241368 LR 0.000500 Time 0.022525 +2023-10-02 21:17:06,482 - Epoch: [100][ 480/ 1236] Overall Loss 0.241056 Objective Loss 0.241056 LR 0.000500 Time 0.022495 +2023-10-02 21:17:06,689 - Epoch: [100][ 490/ 1236] Overall Loss 0.240924 Objective Loss 0.240924 LR 0.000500 Time 0.022457 +2023-10-02 21:17:06,898 - Epoch: [100][ 500/ 1236] Overall Loss 0.241227 Objective Loss 0.241227 LR 0.000500 Time 0.022427 +2023-10-02 21:17:07,106 - Epoch: [100][ 510/ 1236] Overall Loss 0.240911 Objective Loss 0.240911 LR 0.000500 Time 0.022392 +2023-10-02 21:17:07,316 - Epoch: [100][ 520/ 1236] Overall Loss 0.241289 Objective Loss 0.241289 LR 0.000500 Time 0.022365 +2023-10-02 21:17:07,524 - Epoch: [100][ 530/ 1236] Overall Loss 0.240756 Objective Loss 0.240756 LR 0.000500 Time 0.022332 +2023-10-02 21:17:07,735 - Epoch: [100][ 540/ 1236] Overall Loss 0.241070 Objective Loss 0.241070 LR 0.000500 Time 0.022309 +2023-10-02 21:17:07,942 - Epoch: [100][ 550/ 1236] Overall Loss 0.240653 Objective Loss 0.240653 LR 0.000500 Time 0.022279 +2023-10-02 21:17:08,153 - Epoch: [100][ 560/ 1236] Overall Loss 0.240217 Objective Loss 0.240217 LR 0.000500 Time 0.022257 +2023-10-02 21:17:08,360 - Epoch: [100][ 570/ 1236] Overall Loss 0.240517 Objective Loss 0.240517 LR 0.000500 Time 0.022229 +2023-10-02 21:17:08,569 - Epoch: [100][ 580/ 1236] Overall Loss 0.240076 Objective Loss 0.240076 LR 0.000500 Time 0.022206 +2023-10-02 21:17:08,777 - Epoch: [100][ 590/ 1236] Overall Loss 0.239846 Objective Loss 0.239846 LR 0.000500 Time 0.022180 +2023-10-02 21:17:08,988 - Epoch: [100][ 600/ 1236] Overall Loss 0.239683 Objective Loss 0.239683 LR 0.000500 Time 0.022161 +2023-10-02 21:17:09,195 - Epoch: [100][ 610/ 1236] Overall Loss 0.239115 Objective Loss 0.239115 LR 0.000500 Time 0.022136 +2023-10-02 21:17:09,406 - Epoch: [100][ 620/ 1236] Overall Loss 0.238893 Objective Loss 0.238893 LR 0.000500 Time 0.022119 +2023-10-02 21:17:09,613 - Epoch: [100][ 630/ 1236] Overall Loss 0.238633 Objective Loss 0.238633 LR 0.000500 Time 0.022096 +2023-10-02 21:17:09,822 - Epoch: [100][ 640/ 1236] Overall Loss 0.238435 Objective Loss 0.238435 LR 0.000500 Time 0.022078 +2023-10-02 21:17:10,030 - Epoch: [100][ 650/ 1236] Overall Loss 0.237743 Objective Loss 0.237743 LR 0.000500 Time 0.022056 +2023-10-02 21:17:10,242 - Epoch: [100][ 660/ 1236] Overall Loss 0.237407 Objective Loss 0.237407 LR 0.000500 Time 0.022041 +2023-10-02 21:17:10,452 - Epoch: [100][ 670/ 1236] Overall Loss 0.237284 Objective Loss 0.237284 LR 0.000500 Time 0.022026 +2023-10-02 21:17:10,672 - Epoch: [100][ 680/ 1236] Overall Loss 0.237572 Objective Loss 0.237572 LR 0.000500 Time 0.022026 +2023-10-02 21:17:10,887 - Epoch: [100][ 690/ 1236] Overall Loss 0.237581 Objective Loss 0.237581 LR 0.000500 Time 0.022017 +2023-10-02 21:17:11,108 - Epoch: [100][ 700/ 1236] Overall Loss 0.237446 Objective Loss 0.237446 LR 0.000500 Time 0.022017 +2023-10-02 21:17:11,322 - Epoch: [100][ 710/ 1236] Overall Loss 0.237413 Objective Loss 0.237413 LR 0.000500 Time 0.022009 +2023-10-02 21:17:11,543 - Epoch: [100][ 720/ 1236] Overall Loss 0.237245 Objective Loss 0.237245 LR 0.000500 Time 0.022010 +2023-10-02 21:17:11,758 - Epoch: [100][ 730/ 1236] Overall Loss 0.236800 Objective Loss 0.236800 LR 0.000500 Time 0.022002 +2023-10-02 21:17:11,970 - Epoch: [100][ 740/ 1236] Overall Loss 0.236860 Objective Loss 0.236860 LR 0.000500 Time 0.021991 +2023-10-02 21:17:12,179 - Epoch: [100][ 750/ 1236] Overall Loss 0.236743 Objective Loss 0.236743 LR 0.000500 Time 0.021974 +2023-10-02 21:17:12,390 - Epoch: [100][ 760/ 1236] Overall Loss 0.236558 Objective Loss 0.236558 LR 0.000500 Time 0.021962 +2023-10-02 21:17:12,599 - Epoch: [100][ 770/ 1236] Overall Loss 0.236370 Objective Loss 0.236370 LR 0.000500 Time 0.021946 +2023-10-02 21:17:12,809 - Epoch: [100][ 780/ 1236] Overall Loss 0.236325 Objective Loss 0.236325 LR 0.000500 Time 0.021934 +2023-10-02 21:17:13,018 - Epoch: [100][ 790/ 1236] Overall Loss 0.236659 Objective Loss 0.236659 LR 0.000500 Time 0.021919 +2023-10-02 21:17:13,229 - Epoch: [100][ 800/ 1236] Overall Loss 0.236378 Objective Loss 0.236378 LR 0.000500 Time 0.021908 +2023-10-02 21:17:13,437 - Epoch: [100][ 810/ 1236] Overall Loss 0.236246 Objective Loss 0.236246 LR 0.000500 Time 0.021893 +2023-10-02 21:17:13,648 - Epoch: [100][ 820/ 1236] Overall Loss 0.236555 Objective Loss 0.236555 LR 0.000500 Time 0.021882 +2023-10-02 21:17:13,857 - Epoch: [100][ 830/ 1236] Overall Loss 0.236400 Objective Loss 0.236400 LR 0.000500 Time 0.021868 +2023-10-02 21:17:14,068 - Epoch: [100][ 840/ 1236] Overall Loss 0.236364 Objective Loss 0.236364 LR 0.000500 Time 0.021860 +2023-10-02 21:17:14,276 - Epoch: [100][ 850/ 1236] Overall Loss 0.235924 Objective Loss 0.235924 LR 0.000500 Time 0.021846 +2023-10-02 21:17:14,486 - Epoch: [100][ 860/ 1236] Overall Loss 0.235660 Objective Loss 0.235660 LR 0.000500 Time 0.021837 +2023-10-02 21:17:14,695 - Epoch: [100][ 870/ 1236] Overall Loss 0.235667 Objective Loss 0.235667 LR 0.000500 Time 0.021824 +2023-10-02 21:17:14,906 - Epoch: [100][ 880/ 1236] Overall Loss 0.235591 Objective Loss 0.235591 LR 0.000500 Time 0.021815 +2023-10-02 21:17:15,115 - Epoch: [100][ 890/ 1236] Overall Loss 0.235538 Objective Loss 0.235538 LR 0.000500 Time 0.021803 +2023-10-02 21:17:15,326 - Epoch: [100][ 900/ 1236] Overall Loss 0.235365 Objective Loss 0.235365 LR 0.000500 Time 0.021795 +2023-10-02 21:17:15,535 - Epoch: [100][ 910/ 1236] Overall Loss 0.235306 Objective Loss 0.235306 LR 0.000500 Time 0.021783 +2023-10-02 21:17:15,746 - Epoch: [100][ 920/ 1236] Overall Loss 0.235367 Objective Loss 0.235367 LR 0.000500 Time 0.021776 +2023-10-02 21:17:15,955 - Epoch: [100][ 930/ 1236] Overall Loss 0.234878 Objective Loss 0.234878 LR 0.000500 Time 0.021765 +2023-10-02 21:17:16,166 - Epoch: [100][ 940/ 1236] Overall Loss 0.234876 Objective Loss 0.234876 LR 0.000500 Time 0.021757 +2023-10-02 21:17:16,375 - Epoch: [100][ 950/ 1236] Overall Loss 0.234684 Objective Loss 0.234684 LR 0.000500 Time 0.021747 +2023-10-02 21:17:16,585 - Epoch: [100][ 960/ 1236] Overall Loss 0.234832 Objective Loss 0.234832 LR 0.000500 Time 0.021739 +2023-10-02 21:17:16,794 - Epoch: [100][ 970/ 1236] Overall Loss 0.235044 Objective Loss 0.235044 LR 0.000500 Time 0.021729 +2023-10-02 21:17:17,005 - Epoch: [100][ 980/ 1236] Overall Loss 0.234841 Objective Loss 0.234841 LR 0.000500 Time 0.021721 +2023-10-02 21:17:17,214 - Epoch: [100][ 990/ 1236] Overall Loss 0.234620 Objective Loss 0.234620 LR 0.000500 Time 0.021712 +2023-10-02 21:17:17,424 - Epoch: [100][ 1000/ 1236] Overall Loss 0.234685 Objective Loss 0.234685 LR 0.000500 Time 0.021705 +2023-10-02 21:17:17,633 - Epoch: [100][ 1010/ 1236] Overall Loss 0.234407 Objective Loss 0.234407 LR 0.000500 Time 0.021695 +2023-10-02 21:17:17,844 - Epoch: [100][ 1020/ 1236] Overall Loss 0.234163 Objective Loss 0.234163 LR 0.000500 Time 0.021689 +2023-10-02 21:17:18,053 - Epoch: [100][ 1030/ 1236] Overall Loss 0.234031 Objective Loss 0.234031 LR 0.000500 Time 0.021679 +2023-10-02 21:17:18,263 - Epoch: [100][ 1040/ 1236] Overall Loss 0.233681 Objective Loss 0.233681 LR 0.000500 Time 0.021673 +2023-10-02 21:17:18,472 - Epoch: [100][ 1050/ 1236] Overall Loss 0.233636 Objective Loss 0.233636 LR 0.000500 Time 0.021664 +2023-10-02 21:17:18,683 - Epoch: [100][ 1060/ 1236] Overall Loss 0.233517 Objective Loss 0.233517 LR 0.000500 Time 0.021658 +2023-10-02 21:17:18,892 - Epoch: [100][ 1070/ 1236] Overall Loss 0.233340 Objective Loss 0.233340 LR 0.000500 Time 0.021650 +2023-10-02 21:17:19,103 - Epoch: [100][ 1080/ 1236] Overall Loss 0.233431 Objective Loss 0.233431 LR 0.000500 Time 0.021644 +2023-10-02 21:17:19,312 - Epoch: [100][ 1090/ 1236] Overall Loss 0.233166 Objective Loss 0.233166 LR 0.000500 Time 0.021636 +2023-10-02 21:17:19,523 - Epoch: [100][ 1100/ 1236] Overall Loss 0.233009 Objective Loss 0.233009 LR 0.000500 Time 0.021631 +2023-10-02 21:17:19,732 - Epoch: [100][ 1110/ 1236] Overall Loss 0.233075 Objective Loss 0.233075 LR 0.000500 Time 0.021623 +2023-10-02 21:17:19,943 - Epoch: [100][ 1120/ 1236] Overall Loss 0.233018 Objective Loss 0.233018 LR 0.000500 Time 0.021618 +2023-10-02 21:17:20,152 - Epoch: [100][ 1130/ 1236] Overall Loss 0.232901 Objective Loss 0.232901 LR 0.000500 Time 0.021610 +2023-10-02 21:17:20,363 - Epoch: [100][ 1140/ 1236] Overall Loss 0.232854 Objective Loss 0.232854 LR 0.000500 Time 0.021605 +2023-10-02 21:17:20,572 - Epoch: [100][ 1150/ 1236] Overall Loss 0.232855 Objective Loss 0.232855 LR 0.000500 Time 0.021598 +2023-10-02 21:17:20,783 - Epoch: [100][ 1160/ 1236] Overall Loss 0.232693 Objective Loss 0.232693 LR 0.000500 Time 0.021594 +2023-10-02 21:17:20,992 - Epoch: [100][ 1170/ 1236] Overall Loss 0.232526 Objective Loss 0.232526 LR 0.000500 Time 0.021586 +2023-10-02 21:17:21,203 - Epoch: [100][ 1180/ 1236] Overall Loss 0.232277 Objective Loss 0.232277 LR 0.000500 Time 0.021582 +2023-10-02 21:17:21,412 - Epoch: [100][ 1190/ 1236] Overall Loss 0.232278 Objective Loss 0.232278 LR 0.000500 Time 0.021575 +2023-10-02 21:17:21,623 - Epoch: [100][ 1200/ 1236] Overall Loss 0.232403 Objective Loss 0.232403 LR 0.000500 Time 0.021571 +2023-10-02 21:17:21,832 - Epoch: [100][ 1210/ 1236] Overall Loss 0.232316 Objective Loss 0.232316 LR 0.000500 Time 0.021564 +2023-10-02 21:17:22,043 - Epoch: [100][ 1220/ 1236] Overall Loss 0.232376 Objective Loss 0.232376 LR 0.000500 Time 0.021560 +2023-10-02 21:17:22,307 - Epoch: [100][ 1230/ 1236] Overall Loss 0.232281 Objective Loss 0.232281 LR 0.000500 Time 0.021598 +2023-10-02 21:17:22,430 - Epoch: [100][ 1236/ 1236] Overall Loss 0.232424 Objective Loss 0.232424 Top1 88.798371 Top5 98.574338 LR 0.000500 Time 0.021592 +2023-10-02 21:17:22,575 - --- validate (epoch=100)----------- +2023-10-02 21:17:22,575 - 29943 samples (256 per mini-batch) +2023-10-02 21:17:23,052 - Epoch: [100][ 10/ 117] Loss 0.316892 Top1 84.296875 Top5 98.164062 +2023-10-02 21:17:23,201 - Epoch: [100][ 20/ 117] Loss 0.340992 Top1 83.828125 Top5 98.027344 +2023-10-02 21:17:23,351 - Epoch: [100][ 30/ 117] Loss 0.336642 Top1 84.036458 Top5 98.177083 +2023-10-02 21:17:23,500 - Epoch: [100][ 40/ 117] Loss 0.327486 Top1 84.365234 Top5 98.193359 +2023-10-02 21:17:23,650 - Epoch: [100][ 50/ 117] Loss 0.324348 Top1 84.375000 Top5 98.273438 +2023-10-02 21:17:23,800 - Epoch: [100][ 60/ 117] Loss 0.322348 Top1 84.485677 Top5 98.287760 +2023-10-02 21:17:23,951 - Epoch: [100][ 70/ 117] Loss 0.325832 Top1 84.308036 Top5 98.286830 +2023-10-02 21:17:24,102 - Epoch: [100][ 80/ 117] Loss 0.319464 Top1 84.501953 Top5 98.315430 +2023-10-02 21:17:24,254 - Epoch: [100][ 90/ 117] Loss 0.320701 Top1 84.561632 Top5 98.281250 +2023-10-02 21:17:24,405 - Epoch: [100][ 100/ 117] Loss 0.321016 Top1 84.554688 Top5 98.269531 +2023-10-02 21:17:24,564 - Epoch: [100][ 110/ 117] Loss 0.323034 Top1 84.442472 Top5 98.263494 +2023-10-02 21:17:24,652 - Epoch: [100][ 117/ 117] Loss 0.320391 Top1 84.580703 Top5 98.300104 +2023-10-02 21:17:24,749 - ==> Top1: 84.581 Top5: 98.300 Loss: 0.320 + +2023-10-02 21:17:24,750 - ==> Confusion: +[[ 932 0 3 0 8 3 0 0 4 68 1 1 0 3 3 1 4 1 1 0 17] + [ 0 1043 0 1 6 27 0 30 3 0 0 1 1 0 0 3 0 0 8 4 4] + [ 9 0 961 14 2 0 27 8 0 1 0 1 7 2 0 3 1 2 8 4 6] + [ 2 3 15 961 0 4 1 2 6 0 6 0 7 5 36 3 1 5 11 1 20] + [ 28 5 1 0 959 7 1 0 1 17 1 0 2 2 4 5 10 1 0 1 5] + [ 3 32 1 0 1 1000 3 23 1 5 2 6 2 11 4 0 1 0 6 3 12] + [ 2 0 21 1 0 0 1132 5 0 0 5 0 0 0 0 3 1 0 1 14 6] + [ 1 15 19 0 2 30 8 1063 1 1 1 4 3 4 1 0 1 4 38 13 9] + [ 21 0 0 1 1 0 0 0 967 47 11 2 7 15 8 0 4 1 4 0 0] + [ 91 1 3 0 7 3 0 0 18 950 0 1 1 25 7 2 1 1 0 0 8] + [ 3 1 8 9 3 1 2 5 13 1 959 3 0 19 4 0 3 1 6 2 10] + [ 2 0 1 0 1 11 0 7 0 1 0 938 42 6 0 3 1 10 0 8 4] + [ 1 0 1 6 0 2 3 1 1 1 1 31 983 3 1 4 0 9 3 9 8] + [ 2 0 1 0 3 5 0 0 7 14 6 10 3 1050 8 1 0 0 0 3 6] + [ 21 1 4 24 5 1 0 0 23 6 1 0 2 2 999 0 2 2 3 1 4] + [ 1 0 1 2 4 0 1 0 0 0 1 7 8 0 0 1067 16 11 2 10 3] + [ 2 16 0 0 7 8 0 0 0 0 0 4 3 2 5 12 1083 0 2 5 12] + [ 0 0 0 3 0 0 4 0 2 1 0 3 30 1 5 4 0 978 0 1 6] + [ 1 6 1 14 2 1 0 13 8 0 3 2 1 0 11 0 0 0 994 1 10] + [ 0 2 3 1 1 2 6 6 0 0 0 11 4 0 1 3 3 1 0 1101 7] + [ 150 159 104 76 87 174 42 98 94 105 172 97 373 308 122 59 73 66 129 211 5206]] + +2023-10-02 21:17:24,751 - ==> Best [Top1: 84.581 Top5: 98.300 Sparsity:0.00 Params: 169472 on epoch: 100] +2023-10-02 21:17:24,752 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:17:24,765 - + +2023-10-02 21:17:24,765 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:17:25,900 - Epoch: [101][ 10/ 1236] Overall Loss 0.214485 Objective Loss 0.214485 LR 0.000500 Time 0.113447 +2023-10-02 21:17:26,108 - Epoch: [101][ 20/ 1236] Overall Loss 0.218069 Objective Loss 0.218069 LR 0.000500 Time 0.067114 +2023-10-02 21:17:26,317 - Epoch: [101][ 30/ 1236] Overall Loss 0.211024 Objective Loss 0.211024 LR 0.000500 Time 0.051700 +2023-10-02 21:17:26,526 - Epoch: [101][ 40/ 1236] Overall Loss 0.213998 Objective Loss 0.213998 LR 0.000500 Time 0.043983 +2023-10-02 21:17:26,735 - Epoch: [101][ 50/ 1236] Overall Loss 0.212017 Objective Loss 0.212017 LR 0.000500 Time 0.039352 +2023-10-02 21:17:26,943 - Epoch: [101][ 60/ 1236] Overall Loss 0.209511 Objective Loss 0.209511 LR 0.000500 Time 0.036267 +2023-10-02 21:17:27,152 - Epoch: [101][ 70/ 1236] Overall Loss 0.208344 Objective Loss 0.208344 LR 0.000500 Time 0.034067 +2023-10-02 21:17:27,361 - Epoch: [101][ 80/ 1236] Overall Loss 0.207293 Objective Loss 0.207293 LR 0.000500 Time 0.032413 +2023-10-02 21:17:27,570 - Epoch: [101][ 90/ 1236] Overall Loss 0.208268 Objective Loss 0.208268 LR 0.000500 Time 0.031131 +2023-10-02 21:17:27,779 - Epoch: [101][ 100/ 1236] Overall Loss 0.208937 Objective Loss 0.208937 LR 0.000500 Time 0.030101 +2023-10-02 21:17:27,988 - Epoch: [101][ 110/ 1236] Overall Loss 0.211271 Objective Loss 0.211271 LR 0.000500 Time 0.029266 +2023-10-02 21:17:28,196 - Epoch: [101][ 120/ 1236] Overall Loss 0.210775 Objective Loss 0.210775 LR 0.000500 Time 0.028560 +2023-10-02 21:17:28,405 - Epoch: [101][ 130/ 1236] Overall Loss 0.211400 Objective Loss 0.211400 LR 0.000500 Time 0.027956 +2023-10-02 21:17:28,614 - Epoch: [101][ 140/ 1236] Overall Loss 0.212089 Objective Loss 0.212089 LR 0.000500 Time 0.027448 +2023-10-02 21:17:28,822 - Epoch: [101][ 150/ 1236] Overall Loss 0.212885 Objective Loss 0.212885 LR 0.000500 Time 0.027000 +2023-10-02 21:17:29,031 - Epoch: [101][ 160/ 1236] Overall Loss 0.213629 Objective Loss 0.213629 LR 0.000500 Time 0.026613 +2023-10-02 21:17:29,242 - Epoch: [101][ 170/ 1236] Overall Loss 0.214686 Objective Loss 0.214686 LR 0.000500 Time 0.026282 +2023-10-02 21:17:29,454 - Epoch: [101][ 180/ 1236] Overall Loss 0.215171 Objective Loss 0.215171 LR 0.000500 Time 0.026000 +2023-10-02 21:17:29,664 - Epoch: [101][ 190/ 1236] Overall Loss 0.214351 Objective Loss 0.214351 LR 0.000500 Time 0.025734 +2023-10-02 21:17:29,877 - Epoch: [101][ 200/ 1236] Overall Loss 0.214830 Objective Loss 0.214830 LR 0.000500 Time 0.025506 +2023-10-02 21:17:30,087 - Epoch: [101][ 210/ 1236] Overall Loss 0.214074 Objective Loss 0.214074 LR 0.000500 Time 0.025289 +2023-10-02 21:17:30,300 - Epoch: [101][ 220/ 1236] Overall Loss 0.214705 Objective Loss 0.214705 LR 0.000500 Time 0.025107 +2023-10-02 21:17:30,510 - Epoch: [101][ 230/ 1236] Overall Loss 0.214353 Objective Loss 0.214353 LR 0.000500 Time 0.024927 +2023-10-02 21:17:30,723 - Epoch: [101][ 240/ 1236] Overall Loss 0.214686 Objective Loss 0.214686 LR 0.000500 Time 0.024770 +2023-10-02 21:17:30,933 - Epoch: [101][ 250/ 1236] Overall Loss 0.214375 Objective Loss 0.214375 LR 0.000500 Time 0.024618 +2023-10-02 21:17:31,147 - Epoch: [101][ 260/ 1236] Overall Loss 0.213648 Objective Loss 0.213648 LR 0.000500 Time 0.024487 +2023-10-02 21:17:31,356 - Epoch: [101][ 270/ 1236] Overall Loss 0.213330 Objective Loss 0.213330 LR 0.000500 Time 0.024355 +2023-10-02 21:17:31,569 - Epoch: [101][ 280/ 1236] Overall Loss 0.213095 Objective Loss 0.213095 LR 0.000500 Time 0.024241 +2023-10-02 21:17:31,779 - Epoch: [101][ 290/ 1236] Overall Loss 0.213498 Objective Loss 0.213498 LR 0.000500 Time 0.024127 +2023-10-02 21:17:31,987 - Epoch: [101][ 300/ 1236] Overall Loss 0.213945 Objective Loss 0.213945 LR 0.000500 Time 0.024010 +2023-10-02 21:17:32,194 - Epoch: [101][ 310/ 1236] Overall Loss 0.214660 Objective Loss 0.214660 LR 0.000500 Time 0.023903 +2023-10-02 21:17:32,402 - Epoch: [101][ 320/ 1236] Overall Loss 0.214921 Objective Loss 0.214921 LR 0.000500 Time 0.023806 +2023-10-02 21:17:32,609 - Epoch: [101][ 330/ 1236] Overall Loss 0.214912 Objective Loss 0.214912 LR 0.000500 Time 0.023712 +2023-10-02 21:17:32,818 - Epoch: [101][ 340/ 1236] Overall Loss 0.215119 Objective Loss 0.215119 LR 0.000500 Time 0.023626 +2023-10-02 21:17:33,025 - Epoch: [101][ 350/ 1236] Overall Loss 0.214659 Objective Loss 0.214659 LR 0.000500 Time 0.023544 +2023-10-02 21:17:33,234 - Epoch: [101][ 360/ 1236] Overall Loss 0.214724 Objective Loss 0.214724 LR 0.000500 Time 0.023468 +2023-10-02 21:17:33,443 - Epoch: [101][ 370/ 1236] Overall Loss 0.215172 Objective Loss 0.215172 LR 0.000500 Time 0.023398 +2023-10-02 21:17:33,653 - Epoch: [101][ 380/ 1236] Overall Loss 0.215174 Objective Loss 0.215174 LR 0.000500 Time 0.023335 +2023-10-02 21:17:33,859 - Epoch: [101][ 390/ 1236] Overall Loss 0.214736 Objective Loss 0.214736 LR 0.000500 Time 0.023264 +2023-10-02 21:17:34,069 - Epoch: [101][ 400/ 1236] Overall Loss 0.214725 Objective Loss 0.214725 LR 0.000500 Time 0.023206 +2023-10-02 21:17:34,275 - Epoch: [101][ 410/ 1236] Overall Loss 0.214406 Objective Loss 0.214406 LR 0.000500 Time 0.023142 +2023-10-02 21:17:34,484 - Epoch: [101][ 420/ 1236] Overall Loss 0.214351 Objective Loss 0.214351 LR 0.000500 Time 0.023088 +2023-10-02 21:17:34,690 - Epoch: [101][ 430/ 1236] Overall Loss 0.214194 Objective Loss 0.214194 LR 0.000500 Time 0.023030 +2023-10-02 21:17:34,900 - Epoch: [101][ 440/ 1236] Overall Loss 0.214499 Objective Loss 0.214499 LR 0.000500 Time 0.022983 +2023-10-02 21:17:35,106 - Epoch: [101][ 450/ 1236] Overall Loss 0.214469 Objective Loss 0.214469 LR 0.000500 Time 0.022929 +2023-10-02 21:17:35,314 - Epoch: [101][ 460/ 1236] Overall Loss 0.214602 Objective Loss 0.214602 LR 0.000500 Time 0.022882 +2023-10-02 21:17:35,520 - Epoch: [101][ 470/ 1236] Overall Loss 0.214589 Objective Loss 0.214589 LR 0.000500 Time 0.022834 +2023-10-02 21:17:35,729 - Epoch: [101][ 480/ 1236] Overall Loss 0.213941 Objective Loss 0.213941 LR 0.000500 Time 0.022792 +2023-10-02 21:17:35,936 - Epoch: [101][ 490/ 1236] Overall Loss 0.214112 Objective Loss 0.214112 LR 0.000500 Time 0.022749 +2023-10-02 21:17:36,144 - Epoch: [101][ 500/ 1236] Overall Loss 0.214230 Objective Loss 0.214230 LR 0.000500 Time 0.022710 +2023-10-02 21:17:36,352 - Epoch: [101][ 510/ 1236] Overall Loss 0.214847 Objective Loss 0.214847 LR 0.000500 Time 0.022673 +2023-10-02 21:17:36,561 - Epoch: [101][ 520/ 1236] Overall Loss 0.214930 Objective Loss 0.214930 LR 0.000500 Time 0.022637 +2023-10-02 21:17:36,768 - Epoch: [101][ 530/ 1236] Overall Loss 0.214825 Objective Loss 0.214825 LR 0.000500 Time 0.022599 +2023-10-02 21:17:36,978 - Epoch: [101][ 540/ 1236] Overall Loss 0.214694 Objective Loss 0.214694 LR 0.000500 Time 0.022568 +2023-10-02 21:17:37,184 - Epoch: [101][ 550/ 1236] Overall Loss 0.215738 Objective Loss 0.215738 LR 0.000500 Time 0.022532 +2023-10-02 21:17:37,394 - Epoch: [101][ 560/ 1236] Overall Loss 0.216024 Objective Loss 0.216024 LR 0.000500 Time 0.022504 +2023-10-02 21:17:37,601 - Epoch: [101][ 570/ 1236] Overall Loss 0.215591 Objective Loss 0.215591 LR 0.000500 Time 0.022471 +2023-10-02 21:17:37,811 - Epoch: [101][ 580/ 1236] Overall Loss 0.215256 Objective Loss 0.215256 LR 0.000500 Time 0.022446 +2023-10-02 21:17:38,017 - Epoch: [101][ 590/ 1236] Overall Loss 0.215626 Objective Loss 0.215626 LR 0.000500 Time 0.022413 +2023-10-02 21:17:38,226 - Epoch: [101][ 600/ 1236] Overall Loss 0.215586 Objective Loss 0.215586 LR 0.000500 Time 0.022388 +2023-10-02 21:17:38,433 - Epoch: [101][ 610/ 1236] Overall Loss 0.215400 Objective Loss 0.215400 LR 0.000500 Time 0.022359 +2023-10-02 21:17:38,644 - Epoch: [101][ 620/ 1236] Overall Loss 0.215405 Objective Loss 0.215405 LR 0.000500 Time 0.022337 +2023-10-02 21:17:38,850 - Epoch: [101][ 630/ 1236] Overall Loss 0.215195 Objective Loss 0.215195 LR 0.000500 Time 0.022310 +2023-10-02 21:17:39,060 - Epoch: [101][ 640/ 1236] Overall Loss 0.214885 Objective Loss 0.214885 LR 0.000500 Time 0.022289 +2023-10-02 21:17:39,267 - Epoch: [101][ 650/ 1236] Overall Loss 0.215709 Objective Loss 0.215709 LR 0.000500 Time 0.022263 +2023-10-02 21:17:39,476 - Epoch: [101][ 660/ 1236] Overall Loss 0.215582 Objective Loss 0.215582 LR 0.000500 Time 0.022243 +2023-10-02 21:17:39,682 - Epoch: [101][ 670/ 1236] Overall Loss 0.215432 Objective Loss 0.215432 LR 0.000500 Time 0.022219 +2023-10-02 21:17:39,893 - Epoch: [101][ 680/ 1236] Overall Loss 0.215223 Objective Loss 0.215223 LR 0.000500 Time 0.022200 +2023-10-02 21:17:40,099 - Epoch: [101][ 690/ 1236] Overall Loss 0.215911 Objective Loss 0.215911 LR 0.000500 Time 0.022178 +2023-10-02 21:17:40,309 - Epoch: [101][ 700/ 1236] Overall Loss 0.216311 Objective Loss 0.216311 LR 0.000500 Time 0.022160 +2023-10-02 21:17:40,516 - Epoch: [101][ 710/ 1236] Overall Loss 0.216200 Objective Loss 0.216200 LR 0.000500 Time 0.022139 +2023-10-02 21:17:40,725 - Epoch: [101][ 720/ 1236] Overall Loss 0.216100 Objective Loss 0.216100 LR 0.000500 Time 0.022121 +2023-10-02 21:17:40,933 - Epoch: [101][ 730/ 1236] Overall Loss 0.216360 Objective Loss 0.216360 LR 0.000500 Time 0.022103 +2023-10-02 21:17:41,143 - Epoch: [101][ 740/ 1236] Overall Loss 0.216709 Objective Loss 0.216709 LR 0.000500 Time 0.022087 +2023-10-02 21:17:41,349 - Epoch: [101][ 750/ 1236] Overall Loss 0.217080 Objective Loss 0.217080 LR 0.000500 Time 0.022068 +2023-10-02 21:17:41,558 - Epoch: [101][ 760/ 1236] Overall Loss 0.217570 Objective Loss 0.217570 LR 0.000500 Time 0.022052 +2023-10-02 21:17:41,766 - Epoch: [101][ 770/ 1236] Overall Loss 0.217675 Objective Loss 0.217675 LR 0.000500 Time 0.022033 +2023-10-02 21:17:41,974 - Epoch: [101][ 780/ 1236] Overall Loss 0.217594 Objective Loss 0.217594 LR 0.000500 Time 0.022018 +2023-10-02 21:17:42,182 - Epoch: [101][ 790/ 1236] Overall Loss 0.217924 Objective Loss 0.217924 LR 0.000500 Time 0.022002 +2023-10-02 21:17:42,391 - Epoch: [101][ 800/ 1236] Overall Loss 0.218109 Objective Loss 0.218109 LR 0.000500 Time 0.021988 +2023-10-02 21:17:42,599 - Epoch: [101][ 810/ 1236] Overall Loss 0.218074 Objective Loss 0.218074 LR 0.000500 Time 0.021973 +2023-10-02 21:17:42,808 - Epoch: [101][ 820/ 1236] Overall Loss 0.218074 Objective Loss 0.218074 LR 0.000500 Time 0.021959 +2023-10-02 21:17:43,016 - Epoch: [101][ 830/ 1236] Overall Loss 0.218426 Objective Loss 0.218426 LR 0.000500 Time 0.021943 +2023-10-02 21:17:43,226 - Epoch: [101][ 840/ 1236] Overall Loss 0.218423 Objective Loss 0.218423 LR 0.000500 Time 0.021932 +2023-10-02 21:17:43,432 - Epoch: [101][ 850/ 1236] Overall Loss 0.218595 Objective Loss 0.218595 LR 0.000500 Time 0.021917 +2023-10-02 21:17:43,642 - Epoch: [101][ 860/ 1236] Overall Loss 0.218154 Objective Loss 0.218154 LR 0.000500 Time 0.021905 +2023-10-02 21:17:43,849 - Epoch: [101][ 870/ 1236] Overall Loss 0.218082 Objective Loss 0.218082 LR 0.000500 Time 0.021890 +2023-10-02 21:17:44,058 - Epoch: [101][ 880/ 1236] Overall Loss 0.218207 Objective Loss 0.218207 LR 0.000500 Time 0.021878 +2023-10-02 21:17:44,266 - Epoch: [101][ 890/ 1236] Overall Loss 0.218274 Objective Loss 0.218274 LR 0.000500 Time 0.021866 +2023-10-02 21:17:44,475 - Epoch: [101][ 900/ 1236] Overall Loss 0.218182 Objective Loss 0.218182 LR 0.000500 Time 0.021855 +2023-10-02 21:17:44,683 - Epoch: [101][ 910/ 1236] Overall Loss 0.218316 Objective Loss 0.218316 LR 0.000500 Time 0.021843 +2023-10-02 21:17:44,892 - Epoch: [101][ 920/ 1236] Overall Loss 0.218419 Objective Loss 0.218419 LR 0.000500 Time 0.021832 +2023-10-02 21:17:45,100 - Epoch: [101][ 930/ 1236] Overall Loss 0.218511 Objective Loss 0.218511 LR 0.000500 Time 0.021820 +2023-10-02 21:17:45,309 - Epoch: [101][ 940/ 1236] Overall Loss 0.218536 Objective Loss 0.218536 LR 0.000500 Time 0.021809 +2023-10-02 21:17:45,517 - Epoch: [101][ 950/ 1236] Overall Loss 0.218363 Objective Loss 0.218363 LR 0.000500 Time 0.021798 +2023-10-02 21:17:45,726 - Epoch: [101][ 960/ 1236] Overall Loss 0.218442 Objective Loss 0.218442 LR 0.000500 Time 0.021789 +2023-10-02 21:17:45,934 - Epoch: [101][ 970/ 1236] Overall Loss 0.218513 Objective Loss 0.218513 LR 0.000500 Time 0.021777 +2023-10-02 21:17:46,143 - Epoch: [101][ 980/ 1236] Overall Loss 0.218305 Objective Loss 0.218305 LR 0.000500 Time 0.021767 +2023-10-02 21:17:46,350 - Epoch: [101][ 990/ 1236] Overall Loss 0.218513 Objective Loss 0.218513 LR 0.000500 Time 0.021757 +2023-10-02 21:17:46,559 - Epoch: [101][ 1000/ 1236] Overall Loss 0.218751 Objective Loss 0.218751 LR 0.000500 Time 0.021748 +2023-10-02 21:17:46,767 - Epoch: [101][ 1010/ 1236] Overall Loss 0.218825 Objective Loss 0.218825 LR 0.000500 Time 0.021738 +2023-10-02 21:17:46,976 - Epoch: [101][ 1020/ 1236] Overall Loss 0.218819 Objective Loss 0.218819 LR 0.000500 Time 0.021730 +2023-10-02 21:17:47,184 - Epoch: [101][ 1030/ 1236] Overall Loss 0.218699 Objective Loss 0.218699 LR 0.000500 Time 0.021720 +2023-10-02 21:17:47,394 - Epoch: [101][ 1040/ 1236] Overall Loss 0.218608 Objective Loss 0.218608 LR 0.000500 Time 0.021713 +2023-10-02 21:17:47,601 - Epoch: [101][ 1050/ 1236] Overall Loss 0.218531 Objective Loss 0.218531 LR 0.000500 Time 0.021703 +2023-10-02 21:17:47,809 - Epoch: [101][ 1060/ 1236] Overall Loss 0.218610 Objective Loss 0.218610 LR 0.000500 Time 0.021695 +2023-10-02 21:17:48,017 - Epoch: [101][ 1070/ 1236] Overall Loss 0.218513 Objective Loss 0.218513 LR 0.000500 Time 0.021686 +2023-10-02 21:17:48,226 - Epoch: [101][ 1080/ 1236] Overall Loss 0.218533 Objective Loss 0.218533 LR 0.000500 Time 0.021678 +2023-10-02 21:17:48,434 - Epoch: [101][ 1090/ 1236] Overall Loss 0.218602 Objective Loss 0.218602 LR 0.000500 Time 0.021670 +2023-10-02 21:17:48,644 - Epoch: [101][ 1100/ 1236] Overall Loss 0.218574 Objective Loss 0.218574 LR 0.000500 Time 0.021664 +2023-10-02 21:17:48,851 - Epoch: [101][ 1110/ 1236] Overall Loss 0.218362 Objective Loss 0.218362 LR 0.000500 Time 0.021654 +2023-10-02 21:17:49,059 - Epoch: [101][ 1120/ 1236] Overall Loss 0.218288 Objective Loss 0.218288 LR 0.000500 Time 0.021647 +2023-10-02 21:17:49,267 - Epoch: [101][ 1130/ 1236] Overall Loss 0.218540 Objective Loss 0.218540 LR 0.000500 Time 0.021639 +2023-10-02 21:17:49,476 - Epoch: [101][ 1140/ 1236] Overall Loss 0.218713 Objective Loss 0.218713 LR 0.000500 Time 0.021633 +2023-10-02 21:17:49,684 - Epoch: [101][ 1150/ 1236] Overall Loss 0.218931 Objective Loss 0.218931 LR 0.000500 Time 0.021625 +2023-10-02 21:17:49,893 - Epoch: [101][ 1160/ 1236] Overall Loss 0.219181 Objective Loss 0.219181 LR 0.000500 Time 0.021619 +2023-10-02 21:17:50,101 - Epoch: [101][ 1170/ 1236] Overall Loss 0.219204 Objective Loss 0.219204 LR 0.000500 Time 0.021611 +2023-10-02 21:17:50,310 - Epoch: [101][ 1180/ 1236] Overall Loss 0.219094 Objective Loss 0.219094 LR 0.000500 Time 0.021605 +2023-10-02 21:17:50,518 - Epoch: [101][ 1190/ 1236] Overall Loss 0.218997 Objective Loss 0.218997 LR 0.000500 Time 0.021598 +2023-10-02 21:17:50,727 - Epoch: [101][ 1200/ 1236] Overall Loss 0.219008 Objective Loss 0.219008 LR 0.000500 Time 0.021592 +2023-10-02 21:17:50,935 - Epoch: [101][ 1210/ 1236] Overall Loss 0.219048 Objective Loss 0.219048 LR 0.000500 Time 0.021585 +2023-10-02 21:17:51,144 - Epoch: [101][ 1220/ 1236] Overall Loss 0.218969 Objective Loss 0.218969 LR 0.000500 Time 0.021579 +2023-10-02 21:17:51,404 - Epoch: [101][ 1230/ 1236] Overall Loss 0.219236 Objective Loss 0.219236 LR 0.000500 Time 0.021615 +2023-10-02 21:17:51,526 - Epoch: [101][ 1236/ 1236] Overall Loss 0.219257 Objective Loss 0.219257 Top1 86.761711 Top5 98.574338 LR 0.000500 Time 0.021609 +2023-10-02 21:17:51,652 - --- validate (epoch=101)----------- +2023-10-02 21:17:51,652 - 29943 samples (256 per mini-batch) +2023-10-02 21:17:52,124 - Epoch: [101][ 10/ 117] Loss 0.298517 Top1 85.351562 Top5 98.398438 +2023-10-02 21:17:52,279 - Epoch: [101][ 20/ 117] Loss 0.311910 Top1 84.570312 Top5 98.242188 +2023-10-02 21:17:52,432 - Epoch: [101][ 30/ 117] Loss 0.306396 Top1 84.791667 Top5 98.307292 +2023-10-02 21:17:52,587 - Epoch: [101][ 40/ 117] Loss 0.313873 Top1 84.580078 Top5 98.300781 +2023-10-02 21:17:52,740 - Epoch: [101][ 50/ 117] Loss 0.312021 Top1 84.531250 Top5 98.328125 +2023-10-02 21:17:52,894 - Epoch: [101][ 60/ 117] Loss 0.313368 Top1 84.472656 Top5 98.287760 +2023-10-02 21:17:53,045 - Epoch: [101][ 70/ 117] Loss 0.315889 Top1 84.648438 Top5 98.242188 +2023-10-02 21:17:53,197 - Epoch: [101][ 80/ 117] Loss 0.314380 Top1 84.711914 Top5 98.276367 +2023-10-02 21:17:53,347 - Epoch: [101][ 90/ 117] Loss 0.312003 Top1 84.735243 Top5 98.285590 +2023-10-02 21:17:53,498 - Epoch: [101][ 100/ 117] Loss 0.311854 Top1 84.789062 Top5 98.296875 +2023-10-02 21:17:53,656 - Epoch: [101][ 110/ 117] Loss 0.315789 Top1 84.673295 Top5 98.259943 +2023-10-02 21:17:53,745 - Epoch: [101][ 117/ 117] Loss 0.315094 Top1 84.694252 Top5 98.266707 +2023-10-02 21:17:53,886 - ==> Top1: 84.694 Top5: 98.267 Loss: 0.315 + +2023-10-02 21:17:53,887 - ==> Confusion: +[[ 933 0 6 0 5 4 0 1 8 58 0 1 1 4 6 2 6 0 1 0 14] + [ 1 1052 1 2 4 25 0 21 0 1 2 1 0 0 0 4 2 0 7 1 7] + [ 5 0 971 8 0 0 24 6 0 0 2 0 7 2 1 7 3 2 7 3 8] + [ 2 3 16 968 1 2 2 1 1 1 6 0 12 6 39 2 0 4 7 0 16] + [ 32 6 3 0 961 2 0 0 1 12 2 0 1 0 8 6 10 0 1 1 4] + [ 2 34 1 0 3 985 3 32 0 8 1 8 2 13 5 0 2 1 3 3 10] + [ 1 2 20 1 0 1 1124 8 0 0 5 0 1 0 0 7 0 1 3 12 5] + [ 1 16 17 0 9 25 7 1064 0 4 4 2 2 1 1 2 2 5 32 14 10] + [ 14 3 0 1 2 2 2 0 968 49 11 1 1 9 14 3 5 0 4 0 0] + [ 88 1 4 0 7 0 0 0 17 963 3 1 1 13 8 3 0 0 0 3 7] + [ 3 0 10 9 1 1 0 3 15 0 968 3 1 14 4 1 4 2 7 0 7] + [ 0 0 0 0 1 10 0 4 0 1 1 940 34 10 0 6 4 14 0 6 4] + [ 0 1 1 1 0 4 3 0 0 2 1 32 976 2 2 11 3 13 2 4 10] + [ 1 0 1 0 6 5 1 0 11 17 4 2 1 1040 7 2 2 0 0 1 18] + [ 8 2 4 24 5 0 0 0 15 3 4 0 3 4 1012 0 2 2 6 0 7] + [ 0 1 2 2 3 1 2 1 0 0 1 4 8 0 1 1074 15 13 0 4 2] + [ 1 13 1 0 6 7 0 0 0 0 0 6 0 0 4 12 1094 0 0 5 12] + [ 0 0 0 5 0 0 3 0 1 1 0 1 24 0 5 6 3 984 1 1 3] + [ 3 5 1 20 1 0 1 21 0 1 4 2 3 0 12 2 1 0 983 0 8] + [ 0 1 2 2 0 1 9 6 0 0 1 16 4 0 0 5 7 1 0 1083 14] + [ 106 189 144 77 91 129 50 96 84 97 166 111 384 252 145 61 134 68 130 174 5217]] + +2023-10-02 21:17:53,888 - ==> Best [Top1: 84.694 Top5: 98.267 Sparsity:0.00 Params: 169472 on epoch: 101] +2023-10-02 21:17:53,888 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:17:53,901 - + +2023-10-02 21:17:53,901 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:17:54,926 - Epoch: [102][ 10/ 1236] Overall Loss 0.210249 Objective Loss 0.210249 LR 0.000500 Time 0.102432 +2023-10-02 21:17:55,134 - Epoch: [102][ 20/ 1236] Overall Loss 0.203557 Objective Loss 0.203557 LR 0.000500 Time 0.061570 +2023-10-02 21:17:55,342 - Epoch: [102][ 30/ 1236] Overall Loss 0.209204 Objective Loss 0.209204 LR 0.000500 Time 0.047966 +2023-10-02 21:17:55,551 - Epoch: [102][ 40/ 1236] Overall Loss 0.212654 Objective Loss 0.212654 LR 0.000500 Time 0.041201 +2023-10-02 21:17:55,758 - Epoch: [102][ 50/ 1236] Overall Loss 0.214711 Objective Loss 0.214711 LR 0.000500 Time 0.037092 +2023-10-02 21:17:55,965 - Epoch: [102][ 60/ 1236] Overall Loss 0.212993 Objective Loss 0.212993 LR 0.000500 Time 0.034350 +2023-10-02 21:17:56,173 - Epoch: [102][ 70/ 1236] Overall Loss 0.213234 Objective Loss 0.213234 LR 0.000500 Time 0.032411 +2023-10-02 21:17:56,382 - Epoch: [102][ 80/ 1236] Overall Loss 0.216965 Objective Loss 0.216965 LR 0.000500 Time 0.030970 +2023-10-02 21:17:56,589 - Epoch: [102][ 90/ 1236] Overall Loss 0.215696 Objective Loss 0.215696 LR 0.000500 Time 0.029826 +2023-10-02 21:17:56,798 - Epoch: [102][ 100/ 1236] Overall Loss 0.212672 Objective Loss 0.212672 LR 0.000500 Time 0.028933 +2023-10-02 21:17:57,005 - Epoch: [102][ 110/ 1236] Overall Loss 0.212324 Objective Loss 0.212324 LR 0.000500 Time 0.028179 +2023-10-02 21:17:57,214 - Epoch: [102][ 120/ 1236] Overall Loss 0.212074 Objective Loss 0.212074 LR 0.000500 Time 0.027571 +2023-10-02 21:17:57,421 - Epoch: [102][ 130/ 1236] Overall Loss 0.210566 Objective Loss 0.210566 LR 0.000500 Time 0.027038 +2023-10-02 21:17:57,629 - Epoch: [102][ 140/ 1236] Overall Loss 0.209741 Objective Loss 0.209741 LR 0.000500 Time 0.026589 +2023-10-02 21:17:57,837 - Epoch: [102][ 150/ 1236] Overall Loss 0.208242 Objective Loss 0.208242 LR 0.000500 Time 0.026192 +2023-10-02 21:17:58,045 - Epoch: [102][ 160/ 1236] Overall Loss 0.208908 Objective Loss 0.208908 LR 0.000500 Time 0.025854 +2023-10-02 21:17:58,255 - Epoch: [102][ 170/ 1236] Overall Loss 0.211013 Objective Loss 0.211013 LR 0.000500 Time 0.025565 +2023-10-02 21:17:58,466 - Epoch: [102][ 180/ 1236] Overall Loss 0.210681 Objective Loss 0.210681 LR 0.000500 Time 0.025311 +2023-10-02 21:17:58,676 - Epoch: [102][ 190/ 1236] Overall Loss 0.210702 Objective Loss 0.210702 LR 0.000500 Time 0.025081 +2023-10-02 21:17:58,888 - Epoch: [102][ 200/ 1236] Overall Loss 0.211188 Objective Loss 0.211188 LR 0.000500 Time 0.024877 +2023-10-02 21:17:59,095 - Epoch: [102][ 210/ 1236] Overall Loss 0.211461 Objective Loss 0.211461 LR 0.000500 Time 0.024679 +2023-10-02 21:17:59,305 - Epoch: [102][ 220/ 1236] Overall Loss 0.211501 Objective Loss 0.211501 LR 0.000500 Time 0.024509 +2023-10-02 21:17:59,513 - Epoch: [102][ 230/ 1236] Overall Loss 0.212226 Objective Loss 0.212226 LR 0.000500 Time 0.024349 +2023-10-02 21:17:59,722 - Epoch: [102][ 240/ 1236] Overall Loss 0.212400 Objective Loss 0.212400 LR 0.000500 Time 0.024204 +2023-10-02 21:17:59,928 - Epoch: [102][ 250/ 1236] Overall Loss 0.211562 Objective Loss 0.211562 LR 0.000500 Time 0.024058 +2023-10-02 21:18:00,137 - Epoch: [102][ 260/ 1236] Overall Loss 0.210718 Objective Loss 0.210718 LR 0.000500 Time 0.023933 +2023-10-02 21:18:00,346 - Epoch: [102][ 270/ 1236] Overall Loss 0.210952 Objective Loss 0.210952 LR 0.000500 Time 0.023819 +2023-10-02 21:18:00,554 - Epoch: [102][ 280/ 1236] Overall Loss 0.210251 Objective Loss 0.210251 LR 0.000500 Time 0.023707 +2023-10-02 21:18:00,763 - Epoch: [102][ 290/ 1236] Overall Loss 0.209605 Objective Loss 0.209605 LR 0.000500 Time 0.023608 +2023-10-02 21:18:00,972 - Epoch: [102][ 300/ 1236] Overall Loss 0.209712 Objective Loss 0.209712 LR 0.000500 Time 0.023515 +2023-10-02 21:18:01,182 - Epoch: [102][ 310/ 1236] Overall Loss 0.210032 Objective Loss 0.210032 LR 0.000500 Time 0.023429 +2023-10-02 21:18:01,392 - Epoch: [102][ 320/ 1236] Overall Loss 0.210980 Objective Loss 0.210980 LR 0.000500 Time 0.023351 +2023-10-02 21:18:01,603 - Epoch: [102][ 330/ 1236] Overall Loss 0.211008 Objective Loss 0.211008 LR 0.000500 Time 0.023279 +2023-10-02 21:18:01,813 - Epoch: [102][ 340/ 1236] Overall Loss 0.211615 Objective Loss 0.211615 LR 0.000500 Time 0.023212 +2023-10-02 21:18:02,024 - Epoch: [102][ 350/ 1236] Overall Loss 0.212475 Objective Loss 0.212475 LR 0.000500 Time 0.023150 +2023-10-02 21:18:02,235 - Epoch: [102][ 360/ 1236] Overall Loss 0.211519 Objective Loss 0.211519 LR 0.000500 Time 0.023092 +2023-10-02 21:18:02,442 - Epoch: [102][ 370/ 1236] Overall Loss 0.211364 Objective Loss 0.211364 LR 0.000500 Time 0.023026 +2023-10-02 21:18:02,652 - Epoch: [102][ 380/ 1236] Overall Loss 0.210805 Objective Loss 0.210805 LR 0.000500 Time 0.022974 +2023-10-02 21:18:02,860 - Epoch: [102][ 390/ 1236] Overall Loss 0.210894 Objective Loss 0.210894 LR 0.000500 Time 0.022916 +2023-10-02 21:18:03,071 - Epoch: [102][ 400/ 1236] Overall Loss 0.210849 Objective Loss 0.210849 LR 0.000500 Time 0.022870 +2023-10-02 21:18:03,278 - Epoch: [102][ 410/ 1236] Overall Loss 0.211182 Objective Loss 0.211182 LR 0.000500 Time 0.022816 +2023-10-02 21:18:03,489 - Epoch: [102][ 420/ 1236] Overall Loss 0.211007 Objective Loss 0.211007 LR 0.000500 Time 0.022774 +2023-10-02 21:18:03,696 - Epoch: [102][ 430/ 1236] Overall Loss 0.211441 Objective Loss 0.211441 LR 0.000500 Time 0.022727 +2023-10-02 21:18:03,907 - Epoch: [102][ 440/ 1236] Overall Loss 0.211348 Objective Loss 0.211348 LR 0.000500 Time 0.022688 +2023-10-02 21:18:04,114 - Epoch: [102][ 450/ 1236] Overall Loss 0.211757 Objective Loss 0.211757 LR 0.000500 Time 0.022645 +2023-10-02 21:18:04,325 - Epoch: [102][ 460/ 1236] Overall Loss 0.211845 Objective Loss 0.211845 LR 0.000500 Time 0.022610 +2023-10-02 21:18:04,533 - Epoch: [102][ 470/ 1236] Overall Loss 0.211723 Objective Loss 0.211723 LR 0.000500 Time 0.022570 +2023-10-02 21:18:04,745 - Epoch: [102][ 480/ 1236] Overall Loss 0.210929 Objective Loss 0.210929 LR 0.000500 Time 0.022541 +2023-10-02 21:18:04,954 - Epoch: [102][ 490/ 1236] Overall Loss 0.210681 Objective Loss 0.210681 LR 0.000500 Time 0.022508 +2023-10-02 21:18:05,167 - Epoch: [102][ 500/ 1236] Overall Loss 0.210658 Objective Loss 0.210658 LR 0.000500 Time 0.022479 +2023-10-02 21:18:05,374 - Epoch: [102][ 510/ 1236] Overall Loss 0.210354 Objective Loss 0.210354 LR 0.000500 Time 0.022444 +2023-10-02 21:18:05,585 - Epoch: [102][ 520/ 1236] Overall Loss 0.209866 Objective Loss 0.209866 LR 0.000500 Time 0.022418 +2023-10-02 21:18:05,793 - Epoch: [102][ 530/ 1236] Overall Loss 0.209234 Objective Loss 0.209234 LR 0.000500 Time 0.022387 +2023-10-02 21:18:06,004 - Epoch: [102][ 540/ 1236] Overall Loss 0.209383 Objective Loss 0.209383 LR 0.000500 Time 0.022362 +2023-10-02 21:18:06,214 - Epoch: [102][ 550/ 1236] Overall Loss 0.209553 Objective Loss 0.209553 LR 0.000500 Time 0.022336 +2023-10-02 21:18:06,427 - Epoch: [102][ 560/ 1236] Overall Loss 0.209590 Objective Loss 0.209590 LR 0.000500 Time 0.022315 +2023-10-02 21:18:06,637 - Epoch: [102][ 570/ 1236] Overall Loss 0.209885 Objective Loss 0.209885 LR 0.000500 Time 0.022293 +2023-10-02 21:18:06,850 - Epoch: [102][ 580/ 1236] Overall Loss 0.210140 Objective Loss 0.210140 LR 0.000500 Time 0.022274 +2023-10-02 21:18:07,059 - Epoch: [102][ 590/ 1236] Overall Loss 0.210447 Objective Loss 0.210447 LR 0.000500 Time 0.022251 +2023-10-02 21:18:07,272 - Epoch: [102][ 600/ 1236] Overall Loss 0.210735 Objective Loss 0.210735 LR 0.000500 Time 0.022234 +2023-10-02 21:18:07,482 - Epoch: [102][ 610/ 1236] Overall Loss 0.210423 Objective Loss 0.210423 LR 0.000500 Time 0.022213 +2023-10-02 21:18:07,695 - Epoch: [102][ 620/ 1236] Overall Loss 0.210178 Objective Loss 0.210178 LR 0.000500 Time 0.022197 +2023-10-02 21:18:07,904 - Epoch: [102][ 630/ 1236] Overall Loss 0.210227 Objective Loss 0.210227 LR 0.000500 Time 0.022177 +2023-10-02 21:18:08,117 - Epoch: [102][ 640/ 1236] Overall Loss 0.210375 Objective Loss 0.210375 LR 0.000500 Time 0.022162 +2023-10-02 21:18:08,326 - Epoch: [102][ 650/ 1236] Overall Loss 0.210141 Objective Loss 0.210141 LR 0.000500 Time 0.022143 +2023-10-02 21:18:08,538 - Epoch: [102][ 660/ 1236] Overall Loss 0.210445 Objective Loss 0.210445 LR 0.000500 Time 0.022127 +2023-10-02 21:18:08,749 - Epoch: [102][ 670/ 1236] Overall Loss 0.210440 Objective Loss 0.210440 LR 0.000500 Time 0.022112 +2023-10-02 21:18:08,976 - Epoch: [102][ 680/ 1236] Overall Loss 0.210603 Objective Loss 0.210603 LR 0.000500 Time 0.022120 +2023-10-02 21:18:09,206 - Epoch: [102][ 690/ 1236] Overall Loss 0.210318 Objective Loss 0.210318 LR 0.000500 Time 0.022132 +2023-10-02 21:18:09,422 - Epoch: [102][ 700/ 1236] Overall Loss 0.210346 Objective Loss 0.210346 LR 0.000500 Time 0.022124 +2023-10-02 21:18:09,640 - Epoch: [102][ 710/ 1236] Overall Loss 0.210192 Objective Loss 0.210192 LR 0.000500 Time 0.022118 +2023-10-02 21:18:09,854 - Epoch: [102][ 720/ 1236] Overall Loss 0.210263 Objective Loss 0.210263 LR 0.000500 Time 0.022108 +2023-10-02 21:18:10,068 - Epoch: [102][ 730/ 1236] Overall Loss 0.210465 Objective Loss 0.210465 LR 0.000500 Time 0.022098 +2023-10-02 21:18:10,283 - Epoch: [102][ 740/ 1236] Overall Loss 0.210801 Objective Loss 0.210801 LR 0.000500 Time 0.022089 +2023-10-02 21:18:10,497 - Epoch: [102][ 750/ 1236] Overall Loss 0.210950 Objective Loss 0.210950 LR 0.000500 Time 0.022079 +2023-10-02 21:18:10,713 - Epoch: [102][ 760/ 1236] Overall Loss 0.211171 Objective Loss 0.211171 LR 0.000500 Time 0.022072 +2023-10-02 21:18:10,933 - Epoch: [102][ 770/ 1236] Overall Loss 0.211109 Objective Loss 0.211109 LR 0.000500 Time 0.022070 +2023-10-02 21:18:11,159 - Epoch: [102][ 780/ 1236] Overall Loss 0.211205 Objective Loss 0.211205 LR 0.000500 Time 0.022076 +2023-10-02 21:18:11,378 - Epoch: [102][ 790/ 1236] Overall Loss 0.211886 Objective Loss 0.211886 LR 0.000500 Time 0.022074 +2023-10-02 21:18:11,601 - Epoch: [102][ 800/ 1236] Overall Loss 0.211738 Objective Loss 0.211738 LR 0.000500 Time 0.022077 +2023-10-02 21:18:11,814 - Epoch: [102][ 810/ 1236] Overall Loss 0.212229 Objective Loss 0.212229 LR 0.000500 Time 0.022066 +2023-10-02 21:18:12,025 - Epoch: [102][ 820/ 1236] Overall Loss 0.212314 Objective Loss 0.212314 LR 0.000500 Time 0.022054 +2023-10-02 21:18:12,238 - Epoch: [102][ 830/ 1236] Overall Loss 0.212118 Objective Loss 0.212118 LR 0.000500 Time 0.022044 +2023-10-02 21:18:12,450 - Epoch: [102][ 840/ 1236] Overall Loss 0.212168 Objective Loss 0.212168 LR 0.000500 Time 0.022033 +2023-10-02 21:18:12,669 - Epoch: [102][ 850/ 1236] Overall Loss 0.212125 Objective Loss 0.212125 LR 0.000500 Time 0.022031 +2023-10-02 21:18:12,891 - Epoch: [102][ 860/ 1236] Overall Loss 0.211809 Objective Loss 0.211809 LR 0.000500 Time 0.022033 +2023-10-02 21:18:13,111 - Epoch: [102][ 870/ 1236] Overall Loss 0.212085 Objective Loss 0.212085 LR 0.000500 Time 0.022032 +2023-10-02 21:18:13,333 - Epoch: [102][ 880/ 1236] Overall Loss 0.211879 Objective Loss 0.211879 LR 0.000500 Time 0.022034 +2023-10-02 21:18:13,553 - Epoch: [102][ 890/ 1236] Overall Loss 0.212007 Objective Loss 0.212007 LR 0.000500 Time 0.022032 +2023-10-02 21:18:13,783 - Epoch: [102][ 900/ 1236] Overall Loss 0.211906 Objective Loss 0.211906 LR 0.000500 Time 0.022043 +2023-10-02 21:18:13,996 - Epoch: [102][ 910/ 1236] Overall Loss 0.212177 Objective Loss 0.212177 LR 0.000500 Time 0.022035 +2023-10-02 21:18:14,210 - Epoch: [102][ 920/ 1236] Overall Loss 0.212338 Objective Loss 0.212338 LR 0.000500 Time 0.022026 +2023-10-02 21:18:14,423 - Epoch: [102][ 930/ 1236] Overall Loss 0.212242 Objective Loss 0.212242 LR 0.000500 Time 0.022018 +2023-10-02 21:18:14,633 - Epoch: [102][ 940/ 1236] Overall Loss 0.212642 Objective Loss 0.212642 LR 0.000500 Time 0.022007 +2023-10-02 21:18:14,844 - Epoch: [102][ 950/ 1236] Overall Loss 0.212621 Objective Loss 0.212621 LR 0.000500 Time 0.021998 +2023-10-02 21:18:15,052 - Epoch: [102][ 960/ 1236] Overall Loss 0.212620 Objective Loss 0.212620 LR 0.000500 Time 0.021984 +2023-10-02 21:18:15,263 - Epoch: [102][ 970/ 1236] Overall Loss 0.212751 Objective Loss 0.212751 LR 0.000500 Time 0.021975 +2023-10-02 21:18:15,470 - Epoch: [102][ 980/ 1236] Overall Loss 0.213099 Objective Loss 0.213099 LR 0.000500 Time 0.021962 +2023-10-02 21:18:15,694 - Epoch: [102][ 990/ 1236] Overall Loss 0.213196 Objective Loss 0.213196 LR 0.000500 Time 0.021966 +2023-10-02 21:18:15,908 - Epoch: [102][ 1000/ 1236] Overall Loss 0.213232 Objective Loss 0.213232 LR 0.000500 Time 0.021959 +2023-10-02 21:18:16,124 - Epoch: [102][ 1010/ 1236] Overall Loss 0.213182 Objective Loss 0.213182 LR 0.000500 Time 0.021953 +2023-10-02 21:18:16,338 - Epoch: [102][ 1020/ 1236] Overall Loss 0.213210 Objective Loss 0.213210 LR 0.000500 Time 0.021948 +2023-10-02 21:18:16,554 - Epoch: [102][ 1030/ 1236] Overall Loss 0.213104 Objective Loss 0.213104 LR 0.000500 Time 0.021943 +2023-10-02 21:18:16,766 - Epoch: [102][ 1040/ 1236] Overall Loss 0.213068 Objective Loss 0.213068 LR 0.000500 Time 0.021935 +2023-10-02 21:18:16,981 - Epoch: [102][ 1050/ 1236] Overall Loss 0.213181 Objective Loss 0.213181 LR 0.000500 Time 0.021930 +2023-10-02 21:18:17,193 - Epoch: [102][ 1060/ 1236] Overall Loss 0.213322 Objective Loss 0.213322 LR 0.000500 Time 0.021923 +2023-10-02 21:18:17,408 - Epoch: [102][ 1070/ 1236] Overall Loss 0.213453 Objective Loss 0.213453 LR 0.000500 Time 0.021918 +2023-10-02 21:18:17,621 - Epoch: [102][ 1080/ 1236] Overall Loss 0.213523 Objective Loss 0.213523 LR 0.000500 Time 0.021911 +2023-10-02 21:18:17,836 - Epoch: [102][ 1090/ 1236] Overall Loss 0.213557 Objective Loss 0.213557 LR 0.000500 Time 0.021906 +2023-10-02 21:18:18,048 - Epoch: [102][ 1100/ 1236] Overall Loss 0.213510 Objective Loss 0.213510 LR 0.000500 Time 0.021900 +2023-10-02 21:18:18,263 - Epoch: [102][ 1110/ 1236] Overall Loss 0.213307 Objective Loss 0.213307 LR 0.000500 Time 0.021895 +2023-10-02 21:18:18,475 - Epoch: [102][ 1120/ 1236] Overall Loss 0.213348 Objective Loss 0.213348 LR 0.000500 Time 0.021888 +2023-10-02 21:18:18,690 - Epoch: [102][ 1130/ 1236] Overall Loss 0.213228 Objective Loss 0.213228 LR 0.000500 Time 0.021884 +2023-10-02 21:18:18,902 - Epoch: [102][ 1140/ 1236] Overall Loss 0.213255 Objective Loss 0.213255 LR 0.000500 Time 0.021877 +2023-10-02 21:18:19,118 - Epoch: [102][ 1150/ 1236] Overall Loss 0.213085 Objective Loss 0.213085 LR 0.000500 Time 0.021873 +2023-10-02 21:18:19,331 - Epoch: [102][ 1160/ 1236] Overall Loss 0.212917 Objective Loss 0.212917 LR 0.000500 Time 0.021868 +2023-10-02 21:18:19,546 - Epoch: [102][ 1170/ 1236] Overall Loss 0.212988 Objective Loss 0.212988 LR 0.000500 Time 0.021864 +2023-10-02 21:18:19,759 - Epoch: [102][ 1180/ 1236] Overall Loss 0.213057 Objective Loss 0.213057 LR 0.000500 Time 0.021859 +2023-10-02 21:18:19,974 - Epoch: [102][ 1190/ 1236] Overall Loss 0.212882 Objective Loss 0.212882 LR 0.000500 Time 0.021854 +2023-10-02 21:18:20,187 - Epoch: [102][ 1200/ 1236] Overall Loss 0.212909 Objective Loss 0.212909 LR 0.000500 Time 0.021849 +2023-10-02 21:18:20,402 - Epoch: [102][ 1210/ 1236] Overall Loss 0.212902 Objective Loss 0.212902 LR 0.000500 Time 0.021845 +2023-10-02 21:18:20,615 - Epoch: [102][ 1220/ 1236] Overall Loss 0.212816 Objective Loss 0.212816 LR 0.000500 Time 0.021840 +2023-10-02 21:18:20,883 - Epoch: [102][ 1230/ 1236] Overall Loss 0.212873 Objective Loss 0.212873 LR 0.000500 Time 0.021880 +2023-10-02 21:18:21,006 - Epoch: [102][ 1236/ 1236] Overall Loss 0.213086 Objective Loss 0.213086 Top1 84.114053 Top5 98.370672 LR 0.000500 Time 0.021873 +2023-10-02 21:18:21,142 - --- validate (epoch=102)----------- +2023-10-02 21:18:21,142 - 29943 samples (256 per mini-batch) +2023-10-02 21:18:21,658 - Epoch: [102][ 10/ 117] Loss 0.289365 Top1 84.140625 Top5 98.750000 +2023-10-02 21:18:21,824 - Epoch: [102][ 20/ 117] Loss 0.302496 Top1 84.589844 Top5 98.613281 +2023-10-02 21:18:21,988 - Epoch: [102][ 30/ 117] Loss 0.311421 Top1 84.479167 Top5 98.424479 +2023-10-02 21:18:22,156 - Epoch: [102][ 40/ 117] Loss 0.321710 Top1 84.111328 Top5 98.271484 +2023-10-02 21:18:22,320 - Epoch: [102][ 50/ 117] Loss 0.320535 Top1 84.296875 Top5 98.382812 +2023-10-02 21:18:22,488 - Epoch: [102][ 60/ 117] Loss 0.315824 Top1 84.472656 Top5 98.378906 +2023-10-02 21:18:22,654 - Epoch: [102][ 70/ 117] Loss 0.315315 Top1 84.536830 Top5 98.325893 +2023-10-02 21:18:22,825 - Epoch: [102][ 80/ 117] Loss 0.311799 Top1 84.384766 Top5 98.364258 +2023-10-02 21:18:22,993 - Epoch: [102][ 90/ 117] Loss 0.313491 Top1 84.470486 Top5 98.350694 +2023-10-02 21:18:23,164 - Epoch: [102][ 100/ 117] Loss 0.314880 Top1 84.429688 Top5 98.312500 +2023-10-02 21:18:23,337 - Epoch: [102][ 110/ 117] Loss 0.315199 Top1 84.520597 Top5 98.299006 +2023-10-02 21:18:23,427 - Epoch: [102][ 117/ 117] Loss 0.317955 Top1 84.473834 Top5 98.306783 +2023-10-02 21:18:23,546 - ==> Top1: 84.474 Top5: 98.307 Loss: 0.318 + +2023-10-02 21:18:23,547 - ==> Confusion: +[[ 925 0 4 1 9 4 0 2 10 61 1 2 0 3 3 1 7 0 0 0 17] + [ 0 1061 0 2 4 13 0 23 1 0 1 0 1 0 1 4 2 0 12 3 3] + [ 1 1 976 11 1 0 21 5 0 1 2 0 7 2 1 2 1 1 12 3 8] + [ 0 1 10 967 0 3 2 1 4 0 7 0 10 3 40 1 2 7 12 0 19] + [ 25 7 2 0 963 5 0 0 0 15 2 0 2 2 5 7 10 0 2 0 3] + [ 3 59 0 0 3 960 4 25 1 5 3 13 3 13 4 0 8 0 3 1 8] + [ 1 3 22 2 0 0 1133 5 0 0 8 1 0 0 0 3 0 0 0 7 6] + [ 0 9 26 0 10 21 6 1064 2 3 5 4 4 2 0 1 2 2 42 8 7] + [ 14 2 0 2 3 0 0 0 986 33 12 2 2 9 14 2 1 0 4 0 3] + [ 83 1 1 1 7 0 2 1 33 936 0 0 1 23 13 1 0 1 0 4 11] + [ 0 3 10 7 0 1 4 3 14 1 968 2 1 11 5 1 1 1 4 2 14] + [ 1 0 1 0 1 5 0 2 0 0 0 965 25 10 0 3 2 13 0 7 0] + [ 0 1 0 4 1 0 1 1 0 1 4 36 975 0 3 9 3 8 1 9 11] + [ 0 1 2 0 5 5 3 0 17 10 4 6 0 1038 4 2 0 2 0 4 16] + [ 11 2 3 20 2 0 0 0 26 2 4 0 3 2 1008 0 2 2 8 0 6] + [ 0 0 1 3 2 0 0 0 0 1 1 9 8 0 0 1063 18 11 2 8 7] + [ 0 12 1 0 5 6 1 0 1 0 0 3 1 0 5 8 1099 0 2 5 12] + [ 0 1 1 5 0 0 2 0 0 2 0 7 25 0 5 6 1 973 1 0 9] + [ 3 4 8 13 1 0 1 13 3 0 5 2 1 1 17 0 0 0 984 2 10] + [ 0 2 4 3 0 1 16 4 0 0 1 13 6 0 1 1 11 0 0 1080 9] + [ 111 214 144 82 83 123 49 85 114 100 183 128 337 249 162 43 146 63 125 194 5170]] + +2023-10-02 21:18:23,548 - ==> Best [Top1: 84.694 Top5: 98.267 Sparsity:0.00 Params: 169472 on epoch: 101] +2023-10-02 21:18:23,548 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:18:23,554 - + +2023-10-02 21:18:23,554 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:18:24,715 - Epoch: [103][ 10/ 1236] Overall Loss 0.216047 Objective Loss 0.216047 LR 0.000500 Time 0.115978 +2023-10-02 21:18:24,925 - Epoch: [103][ 20/ 1236] Overall Loss 0.223752 Objective Loss 0.223752 LR 0.000500 Time 0.068495 +2023-10-02 21:18:25,134 - Epoch: [103][ 30/ 1236] Overall Loss 0.217735 Objective Loss 0.217735 LR 0.000500 Time 0.052625 +2023-10-02 21:18:25,345 - Epoch: [103][ 40/ 1236] Overall Loss 0.216589 Objective Loss 0.216589 LR 0.000500 Time 0.044726 +2023-10-02 21:18:25,555 - Epoch: [103][ 50/ 1236] Overall Loss 0.214693 Objective Loss 0.214693 LR 0.000500 Time 0.039975 +2023-10-02 21:18:25,765 - Epoch: [103][ 60/ 1236] Overall Loss 0.214258 Objective Loss 0.214258 LR 0.000500 Time 0.036808 +2023-10-02 21:18:25,975 - Epoch: [103][ 70/ 1236] Overall Loss 0.210931 Objective Loss 0.210931 LR 0.000500 Time 0.034545 +2023-10-02 21:18:26,186 - Epoch: [103][ 80/ 1236] Overall Loss 0.212730 Objective Loss 0.212730 LR 0.000500 Time 0.032862 +2023-10-02 21:18:26,405 - Epoch: [103][ 90/ 1236] Overall Loss 0.211280 Objective Loss 0.211280 LR 0.000500 Time 0.031639 +2023-10-02 21:18:26,619 - Epoch: [103][ 100/ 1236] Overall Loss 0.211946 Objective Loss 0.211946 LR 0.000500 Time 0.030614 +2023-10-02 21:18:26,828 - Epoch: [103][ 110/ 1236] Overall Loss 0.212733 Objective Loss 0.212733 LR 0.000500 Time 0.029726 +2023-10-02 21:18:27,037 - Epoch: [103][ 120/ 1236] Overall Loss 0.210029 Objective Loss 0.210029 LR 0.000500 Time 0.028989 +2023-10-02 21:18:27,250 - Epoch: [103][ 130/ 1236] Overall Loss 0.210058 Objective Loss 0.210058 LR 0.000500 Time 0.028394 +2023-10-02 21:18:27,460 - Epoch: [103][ 140/ 1236] Overall Loss 0.208764 Objective Loss 0.208764 LR 0.000500 Time 0.027867 +2023-10-02 21:18:27,669 - Epoch: [103][ 150/ 1236] Overall Loss 0.209092 Objective Loss 0.209092 LR 0.000500 Time 0.027394 +2023-10-02 21:18:27,879 - Epoch: [103][ 160/ 1236] Overall Loss 0.208278 Objective Loss 0.208278 LR 0.000500 Time 0.026993 +2023-10-02 21:18:28,086 - Epoch: [103][ 170/ 1236] Overall Loss 0.208127 Objective Loss 0.208127 LR 0.000500 Time 0.026625 +2023-10-02 21:18:28,297 - Epoch: [103][ 180/ 1236] Overall Loss 0.207874 Objective Loss 0.207874 LR 0.000500 Time 0.026312 +2023-10-02 21:18:28,506 - Epoch: [103][ 190/ 1236] Overall Loss 0.206753 Objective Loss 0.206753 LR 0.000500 Time 0.026026 +2023-10-02 21:18:28,718 - Epoch: [103][ 200/ 1236] Overall Loss 0.207171 Objective Loss 0.207171 LR 0.000500 Time 0.025783 +2023-10-02 21:18:28,927 - Epoch: [103][ 210/ 1236] Overall Loss 0.206979 Objective Loss 0.206979 LR 0.000500 Time 0.025550 +2023-10-02 21:18:29,138 - Epoch: [103][ 220/ 1236] Overall Loss 0.207094 Objective Loss 0.207094 LR 0.000500 Time 0.025344 +2023-10-02 21:18:29,347 - Epoch: [103][ 230/ 1236] Overall Loss 0.206629 Objective Loss 0.206629 LR 0.000500 Time 0.025152 +2023-10-02 21:18:29,558 - Epoch: [103][ 240/ 1236] Overall Loss 0.207088 Objective Loss 0.207088 LR 0.000500 Time 0.024982 +2023-10-02 21:18:29,767 - Epoch: [103][ 250/ 1236] Overall Loss 0.205401 Objective Loss 0.205401 LR 0.000500 Time 0.024811 +2023-10-02 21:18:29,978 - Epoch: [103][ 260/ 1236] Overall Loss 0.205653 Objective Loss 0.205653 LR 0.000500 Time 0.024667 +2023-10-02 21:18:30,187 - Epoch: [103][ 270/ 1236] Overall Loss 0.206050 Objective Loss 0.206050 LR 0.000500 Time 0.024520 +2023-10-02 21:18:30,397 - Epoch: [103][ 280/ 1236] Overall Loss 0.205089 Objective Loss 0.205089 LR 0.000500 Time 0.024395 +2023-10-02 21:18:30,608 - Epoch: [103][ 290/ 1236] Overall Loss 0.205826 Objective Loss 0.205826 LR 0.000500 Time 0.024277 +2023-10-02 21:18:30,821 - Epoch: [103][ 300/ 1236] Overall Loss 0.206190 Objective Loss 0.206190 LR 0.000500 Time 0.024175 +2023-10-02 21:18:31,032 - Epoch: [103][ 310/ 1236] Overall Loss 0.207097 Objective Loss 0.207097 LR 0.000500 Time 0.024071 +2023-10-02 21:18:31,244 - Epoch: [103][ 320/ 1236] Overall Loss 0.207037 Objective Loss 0.207037 LR 0.000500 Time 0.023980 +2023-10-02 21:18:31,455 - Epoch: [103][ 330/ 1236] Overall Loss 0.206971 Objective Loss 0.206971 LR 0.000500 Time 0.023891 +2023-10-02 21:18:31,669 - Epoch: [103][ 340/ 1236] Overall Loss 0.207168 Objective Loss 0.207168 LR 0.000500 Time 0.023817 +2023-10-02 21:18:31,878 - Epoch: [103][ 350/ 1236] Overall Loss 0.206725 Objective Loss 0.206725 LR 0.000500 Time 0.023735 +2023-10-02 21:18:32,090 - Epoch: [103][ 360/ 1236] Overall Loss 0.206528 Objective Loss 0.206528 LR 0.000500 Time 0.023663 +2023-10-02 21:18:32,301 - Epoch: [103][ 370/ 1236] Overall Loss 0.206660 Objective Loss 0.206660 LR 0.000500 Time 0.023593 +2023-10-02 21:18:32,513 - Epoch: [103][ 380/ 1236] Overall Loss 0.206994 Objective Loss 0.206994 LR 0.000500 Time 0.023529 +2023-10-02 21:18:32,724 - Epoch: [103][ 390/ 1236] Overall Loss 0.206998 Objective Loss 0.206998 LR 0.000500 Time 0.023467 +2023-10-02 21:18:32,936 - Epoch: [103][ 400/ 1236] Overall Loss 0.207023 Objective Loss 0.207023 LR 0.000500 Time 0.023409 +2023-10-02 21:18:33,148 - Epoch: [103][ 410/ 1236] Overall Loss 0.206772 Objective Loss 0.206772 LR 0.000500 Time 0.023353 +2023-10-02 21:18:33,360 - Epoch: [103][ 420/ 1236] Overall Loss 0.206538 Objective Loss 0.206538 LR 0.000500 Time 0.023301 +2023-10-02 21:18:33,571 - Epoch: [103][ 430/ 1236] Overall Loss 0.206945 Objective Loss 0.206945 LR 0.000500 Time 0.023249 +2023-10-02 21:18:33,782 - Epoch: [103][ 440/ 1236] Overall Loss 0.207208 Objective Loss 0.207208 LR 0.000500 Time 0.023201 +2023-10-02 21:18:33,994 - Epoch: [103][ 450/ 1236] Overall Loss 0.207334 Objective Loss 0.207334 LR 0.000500 Time 0.023155 +2023-10-02 21:18:34,206 - Epoch: [103][ 460/ 1236] Overall Loss 0.206920 Objective Loss 0.206920 LR 0.000500 Time 0.023112 +2023-10-02 21:18:34,417 - Epoch: [103][ 470/ 1236] Overall Loss 0.206492 Objective Loss 0.206492 LR 0.000500 Time 0.023068 +2023-10-02 21:18:34,629 - Epoch: [103][ 480/ 1236] Overall Loss 0.206710 Objective Loss 0.206710 LR 0.000500 Time 0.023029 +2023-10-02 21:18:34,840 - Epoch: [103][ 490/ 1236] Overall Loss 0.206981 Objective Loss 0.206981 LR 0.000500 Time 0.022989 +2023-10-02 21:18:35,052 - Epoch: [103][ 500/ 1236] Overall Loss 0.206942 Objective Loss 0.206942 LR 0.000500 Time 0.022953 +2023-10-02 21:18:35,263 - Epoch: [103][ 510/ 1236] Overall Loss 0.206725 Objective Loss 0.206725 LR 0.000500 Time 0.022914 +2023-10-02 21:18:35,475 - Epoch: [103][ 520/ 1236] Overall Loss 0.206829 Objective Loss 0.206829 LR 0.000500 Time 0.022880 +2023-10-02 21:18:35,686 - Epoch: [103][ 530/ 1236] Overall Loss 0.206772 Objective Loss 0.206772 LR 0.000500 Time 0.022844 +2023-10-02 21:18:35,898 - Epoch: [103][ 540/ 1236] Overall Loss 0.207240 Objective Loss 0.207240 LR 0.000500 Time 0.022813 +2023-10-02 21:18:36,110 - Epoch: [103][ 550/ 1236] Overall Loss 0.206935 Objective Loss 0.206935 LR 0.000500 Time 0.022782 +2023-10-02 21:18:36,322 - Epoch: [103][ 560/ 1236] Overall Loss 0.207060 Objective Loss 0.207060 LR 0.000500 Time 0.022752 +2023-10-02 21:18:36,533 - Epoch: [103][ 570/ 1236] Overall Loss 0.206928 Objective Loss 0.206928 LR 0.000500 Time 0.022723 +2023-10-02 21:18:36,745 - Epoch: [103][ 580/ 1236] Overall Loss 0.206929 Objective Loss 0.206929 LR 0.000500 Time 0.022696 +2023-10-02 21:18:36,956 - Epoch: [103][ 590/ 1236] Overall Loss 0.207269 Objective Loss 0.207269 LR 0.000500 Time 0.022669 +2023-10-02 21:18:37,168 - Epoch: [103][ 600/ 1236] Overall Loss 0.207145 Objective Loss 0.207145 LR 0.000500 Time 0.022644 +2023-10-02 21:18:37,378 - Epoch: [103][ 610/ 1236] Overall Loss 0.207078 Objective Loss 0.207078 LR 0.000500 Time 0.022616 +2023-10-02 21:18:37,590 - Epoch: [103][ 620/ 1236] Overall Loss 0.207046 Objective Loss 0.207046 LR 0.000500 Time 0.022592 +2023-10-02 21:18:37,801 - Epoch: [103][ 630/ 1236] Overall Loss 0.206826 Objective Loss 0.206826 LR 0.000500 Time 0.022569 +2023-10-02 21:18:38,013 - Epoch: [103][ 640/ 1236] Overall Loss 0.206975 Objective Loss 0.206975 LR 0.000500 Time 0.022546 +2023-10-02 21:18:38,224 - Epoch: [103][ 650/ 1236] Overall Loss 0.207041 Objective Loss 0.207041 LR 0.000500 Time 0.022524 +2023-10-02 21:18:38,436 - Epoch: [103][ 660/ 1236] Overall Loss 0.206971 Objective Loss 0.206971 LR 0.000500 Time 0.022503 +2023-10-02 21:18:38,647 - Epoch: [103][ 670/ 1236] Overall Loss 0.206601 Objective Loss 0.206601 LR 0.000500 Time 0.022482 +2023-10-02 21:18:38,859 - Epoch: [103][ 680/ 1236] Overall Loss 0.206970 Objective Loss 0.206970 LR 0.000500 Time 0.022462 +2023-10-02 21:18:39,069 - Epoch: [103][ 690/ 1236] Overall Loss 0.206986 Objective Loss 0.206986 LR 0.000500 Time 0.022442 +2023-10-02 21:18:39,281 - Epoch: [103][ 700/ 1236] Overall Loss 0.207144 Objective Loss 0.207144 LR 0.000500 Time 0.022424 +2023-10-02 21:18:39,492 - Epoch: [103][ 710/ 1236] Overall Loss 0.207305 Objective Loss 0.207305 LR 0.000500 Time 0.022405 +2023-10-02 21:18:39,706 - Epoch: [103][ 720/ 1236] Overall Loss 0.207610 Objective Loss 0.207610 LR 0.000500 Time 0.022389 +2023-10-02 21:18:39,916 - Epoch: [103][ 730/ 1236] Overall Loss 0.207801 Objective Loss 0.207801 LR 0.000500 Time 0.022370 +2023-10-02 21:18:40,128 - Epoch: [103][ 740/ 1236] Overall Loss 0.208004 Objective Loss 0.208004 LR 0.000500 Time 0.022354 +2023-10-02 21:18:40,339 - Epoch: [103][ 750/ 1236] Overall Loss 0.207841 Objective Loss 0.207841 LR 0.000500 Time 0.022337 +2023-10-02 21:18:40,551 - Epoch: [103][ 760/ 1236] Overall Loss 0.208236 Objective Loss 0.208236 LR 0.000500 Time 0.022321 +2023-10-02 21:18:40,762 - Epoch: [103][ 770/ 1236] Overall Loss 0.207943 Objective Loss 0.207943 LR 0.000500 Time 0.022306 +2023-10-02 21:18:40,976 - Epoch: [103][ 780/ 1236] Overall Loss 0.207734 Objective Loss 0.207734 LR 0.000500 Time 0.022293 +2023-10-02 21:18:41,185 - Epoch: [103][ 790/ 1236] Overall Loss 0.207878 Objective Loss 0.207878 LR 0.000500 Time 0.022276 +2023-10-02 21:18:41,398 - Epoch: [103][ 800/ 1236] Overall Loss 0.208044 Objective Loss 0.208044 LR 0.000500 Time 0.022262 +2023-10-02 21:18:41,609 - Epoch: [103][ 810/ 1236] Overall Loss 0.208042 Objective Loss 0.208042 LR 0.000500 Time 0.022248 +2023-10-02 21:18:41,821 - Epoch: [103][ 820/ 1236] Overall Loss 0.207953 Objective Loss 0.207953 LR 0.000500 Time 0.022235 +2023-10-02 21:18:42,032 - Epoch: [103][ 830/ 1236] Overall Loss 0.207828 Objective Loss 0.207828 LR 0.000500 Time 0.022221 +2023-10-02 21:18:42,246 - Epoch: [103][ 840/ 1236] Overall Loss 0.207969 Objective Loss 0.207969 LR 0.000500 Time 0.022210 +2023-10-02 21:18:42,456 - Epoch: [103][ 850/ 1236] Overall Loss 0.208070 Objective Loss 0.208070 LR 0.000500 Time 0.022196 +2023-10-02 21:18:42,668 - Epoch: [103][ 860/ 1236] Overall Loss 0.208185 Objective Loss 0.208185 LR 0.000500 Time 0.022184 +2023-10-02 21:18:42,879 - Epoch: [103][ 870/ 1236] Overall Loss 0.208482 Objective Loss 0.208482 LR 0.000500 Time 0.022172 +2023-10-02 21:18:43,091 - Epoch: [103][ 880/ 1236] Overall Loss 0.208482 Objective Loss 0.208482 LR 0.000500 Time 0.022160 +2023-10-02 21:18:43,303 - Epoch: [103][ 890/ 1236] Overall Loss 0.208506 Objective Loss 0.208506 LR 0.000500 Time 0.022147 +2023-10-02 21:18:43,515 - Epoch: [103][ 900/ 1236] Overall Loss 0.208469 Objective Loss 0.208469 LR 0.000500 Time 0.022136 +2023-10-02 21:18:43,726 - Epoch: [103][ 910/ 1236] Overall Loss 0.208413 Objective Loss 0.208413 LR 0.000500 Time 0.022125 +2023-10-02 21:18:43,938 - Epoch: [103][ 920/ 1236] Overall Loss 0.208408 Objective Loss 0.208408 LR 0.000500 Time 0.022114 +2023-10-02 21:18:44,149 - Epoch: [103][ 930/ 1236] Overall Loss 0.208436 Objective Loss 0.208436 LR 0.000500 Time 0.022103 +2023-10-02 21:18:44,361 - Epoch: [103][ 940/ 1236] Overall Loss 0.208380 Objective Loss 0.208380 LR 0.000500 Time 0.022093 +2023-10-02 21:18:44,572 - Epoch: [103][ 950/ 1236] Overall Loss 0.208311 Objective Loss 0.208311 LR 0.000500 Time 0.022083 +2023-10-02 21:18:44,786 - Epoch: [103][ 960/ 1236] Overall Loss 0.208400 Objective Loss 0.208400 LR 0.000500 Time 0.022075 +2023-10-02 21:18:44,996 - Epoch: [103][ 970/ 1236] Overall Loss 0.208417 Objective Loss 0.208417 LR 0.000500 Time 0.022063 +2023-10-02 21:18:45,208 - Epoch: [103][ 980/ 1236] Overall Loss 0.208428 Objective Loss 0.208428 LR 0.000500 Time 0.022054 +2023-10-02 21:18:45,419 - Epoch: [103][ 990/ 1236] Overall Loss 0.208585 Objective Loss 0.208585 LR 0.000500 Time 0.022044 +2023-10-02 21:18:45,632 - Epoch: [103][ 1000/ 1236] Overall Loss 0.208562 Objective Loss 0.208562 LR 0.000500 Time 0.022036 +2023-10-02 21:18:45,843 - Epoch: [103][ 1010/ 1236] Overall Loss 0.208687 Objective Loss 0.208687 LR 0.000500 Time 0.022026 +2023-10-02 21:18:46,055 - Epoch: [103][ 1020/ 1236] Overall Loss 0.208742 Objective Loss 0.208742 LR 0.000500 Time 0.022018 +2023-10-02 21:18:46,267 - Epoch: [103][ 1030/ 1236] Overall Loss 0.208996 Objective Loss 0.208996 LR 0.000500 Time 0.022009 +2023-10-02 21:18:46,478 - Epoch: [103][ 1040/ 1236] Overall Loss 0.208902 Objective Loss 0.208902 LR 0.000500 Time 0.022000 +2023-10-02 21:18:46,697 - Epoch: [103][ 1050/ 1236] Overall Loss 0.208898 Objective Loss 0.208898 LR 0.000500 Time 0.021999 +2023-10-02 21:18:46,912 - Epoch: [103][ 1060/ 1236] Overall Loss 0.208992 Objective Loss 0.208992 LR 0.000500 Time 0.021994 +2023-10-02 21:18:47,132 - Epoch: [103][ 1070/ 1236] Overall Loss 0.208939 Objective Loss 0.208939 LR 0.000500 Time 0.021994 +2023-10-02 21:18:47,348 - Epoch: [103][ 1080/ 1236] Overall Loss 0.209081 Objective Loss 0.209081 LR 0.000500 Time 0.021989 +2023-10-02 21:18:47,567 - Epoch: [103][ 1090/ 1236] Overall Loss 0.209331 Objective Loss 0.209331 LR 0.000500 Time 0.021989 +2023-10-02 21:18:47,783 - Epoch: [103][ 1100/ 1236] Overall Loss 0.209264 Objective Loss 0.209264 LR 0.000500 Time 0.021984 +2023-10-02 21:18:47,995 - Epoch: [103][ 1110/ 1236] Overall Loss 0.209118 Objective Loss 0.209118 LR 0.000500 Time 0.021978 +2023-10-02 21:18:48,204 - Epoch: [103][ 1120/ 1236] Overall Loss 0.209286 Objective Loss 0.209286 LR 0.000500 Time 0.021966 +2023-10-02 21:18:48,414 - Epoch: [103][ 1130/ 1236] Overall Loss 0.209392 Objective Loss 0.209392 LR 0.000500 Time 0.021958 +2023-10-02 21:18:48,623 - Epoch: [103][ 1140/ 1236] Overall Loss 0.209414 Objective Loss 0.209414 LR 0.000500 Time 0.021947 +2023-10-02 21:18:48,834 - Epoch: [103][ 1150/ 1236] Overall Loss 0.209448 Objective Loss 0.209448 LR 0.000500 Time 0.021939 +2023-10-02 21:18:49,043 - Epoch: [103][ 1160/ 1236] Overall Loss 0.209432 Objective Loss 0.209432 LR 0.000500 Time 0.021929 +2023-10-02 21:18:49,253 - Epoch: [103][ 1170/ 1236] Overall Loss 0.209448 Objective Loss 0.209448 LR 0.000500 Time 0.021921 +2023-10-02 21:18:49,462 - Epoch: [103][ 1180/ 1236] Overall Loss 0.209356 Objective Loss 0.209356 LR 0.000500 Time 0.021911 +2023-10-02 21:18:49,672 - Epoch: [103][ 1190/ 1236] Overall Loss 0.209589 Objective Loss 0.209589 LR 0.000500 Time 0.021903 +2023-10-02 21:18:49,881 - Epoch: [103][ 1200/ 1236] Overall Loss 0.209783 Objective Loss 0.209783 LR 0.000500 Time 0.021893 +2023-10-02 21:18:50,091 - Epoch: [103][ 1210/ 1236] Overall Loss 0.209922 Objective Loss 0.209922 LR 0.000500 Time 0.021886 +2023-10-02 21:18:50,300 - Epoch: [103][ 1220/ 1236] Overall Loss 0.209827 Objective Loss 0.209827 LR 0.000500 Time 0.021877 +2023-10-02 21:18:50,563 - Epoch: [103][ 1230/ 1236] Overall Loss 0.209765 Objective Loss 0.209765 LR 0.000500 Time 0.021912 +2023-10-02 21:18:50,684 - Epoch: [103][ 1236/ 1236] Overall Loss 0.210094 Objective Loss 0.210094 Top1 86.761711 Top5 97.963340 LR 0.000500 Time 0.021903 +2023-10-02 21:18:50,812 - --- validate (epoch=103)----------- +2023-10-02 21:18:50,812 - 29943 samples (256 per mini-batch) +2023-10-02 21:18:51,284 - Epoch: [103][ 10/ 117] Loss 0.320670 Top1 84.140625 Top5 98.085938 +2023-10-02 21:18:51,438 - Epoch: [103][ 20/ 117] Loss 0.315634 Top1 84.296875 Top5 98.281250 +2023-10-02 21:18:51,590 - Epoch: [103][ 30/ 117] Loss 0.323409 Top1 84.361979 Top5 98.268229 +2023-10-02 21:18:51,741 - Epoch: [103][ 40/ 117] Loss 0.313400 Top1 84.580078 Top5 98.378906 +2023-10-02 21:18:51,893 - Epoch: [103][ 50/ 117] Loss 0.306978 Top1 85.007812 Top5 98.445312 +2023-10-02 21:18:52,044 - Epoch: [103][ 60/ 117] Loss 0.313358 Top1 84.986979 Top5 98.424479 +2023-10-02 21:18:52,194 - Epoch: [103][ 70/ 117] Loss 0.311937 Top1 84.960938 Top5 98.443080 +2023-10-02 21:18:52,345 - Epoch: [103][ 80/ 117] Loss 0.315758 Top1 84.921875 Top5 98.378906 +2023-10-02 21:18:52,496 - Epoch: [103][ 90/ 117] Loss 0.318231 Top1 84.765625 Top5 98.372396 +2023-10-02 21:18:52,646 - Epoch: [103][ 100/ 117] Loss 0.319295 Top1 84.785156 Top5 98.386719 +2023-10-02 21:18:52,805 - Epoch: [103][ 110/ 117] Loss 0.318809 Top1 84.783381 Top5 98.384233 +2023-10-02 21:18:52,894 - Epoch: [103][ 117/ 117] Loss 0.319004 Top1 84.787763 Top5 98.366897 +2023-10-02 21:18:53,023 - ==> Top1: 84.788 Top5: 98.367 Loss: 0.319 + +2023-10-02 21:18:53,023 - ==> Confusion: +[[ 937 1 6 1 13 3 0 0 3 58 2 1 0 1 5 3 2 0 0 0 14] + [ 1 1056 2 1 6 21 3 20 2 0 1 0 0 0 0 3 2 0 5 3 5] + [ 2 0 982 11 1 0 18 2 1 1 3 0 7 3 1 5 1 1 9 3 5] + [ 3 2 23 963 0 0 2 1 1 1 5 0 9 5 35 4 1 7 12 0 15] + [ 26 4 3 0 961 6 0 0 2 12 1 0 2 0 8 8 11 0 1 1 4] + [ 1 41 3 1 7 984 0 25 3 6 1 11 2 7 5 0 1 0 5 1 12] + [ 0 3 25 2 0 0 1117 3 0 0 6 3 0 0 0 11 0 1 1 9 10] + [ 2 18 23 0 5 32 9 1057 1 5 3 7 2 3 2 1 2 1 28 10 7] + [ 20 3 1 1 2 1 0 1 978 46 7 2 1 8 11 0 5 0 1 0 1] + [ 111 1 2 0 7 2 1 0 22 934 1 1 1 17 6 3 1 0 0 2 7] + [ 5 1 11 8 1 2 3 4 14 1 962 2 0 12 7 1 2 2 5 0 10] + [ 0 0 2 2 2 6 0 4 0 1 0 961 28 5 0 3 1 14 0 3 3] + [ 1 1 2 5 1 0 1 1 2 0 1 50 956 0 5 13 1 8 2 6 12] + [ 1 0 4 0 5 3 0 0 15 15 1 6 0 1046 8 0 1 0 0 2 12] + [ 13 2 3 23 5 0 0 0 20 5 1 0 3 2 999 0 3 3 10 0 9] + [ 1 0 1 2 5 0 0 0 0 1 1 9 9 0 0 1067 18 9 2 4 5] + [ 1 15 1 0 4 6 0 0 1 0 0 3 0 2 5 9 1095 0 1 5 13] + [ 1 1 2 5 0 0 1 0 0 0 0 3 26 1 2 9 3 978 0 0 6] + [ 2 8 7 20 0 1 0 20 3 2 4 1 2 0 12 0 1 0 974 1 10] + [ 0 1 9 0 0 2 8 9 0 0 1 16 7 0 0 3 8 1 0 1077 10] + [ 137 157 195 67 102 148 41 96 87 85 163 102 333 277 129 59 111 51 110 151 5304]] + +2023-10-02 21:18:53,025 - ==> Best [Top1: 84.788 Top5: 98.367 Sparsity:0.00 Params: 169472 on epoch: 103] +2023-10-02 21:18:53,025 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:18:53,038 - + +2023-10-02 21:18:53,038 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:18:54,055 - Epoch: [104][ 10/ 1236] Overall Loss 0.196291 Objective Loss 0.196291 LR 0.000500 Time 0.101662 +2023-10-02 21:18:54,265 - Epoch: [104][ 20/ 1236] Overall Loss 0.193863 Objective Loss 0.193863 LR 0.000500 Time 0.061288 +2023-10-02 21:18:54,473 - Epoch: [104][ 30/ 1236] Overall Loss 0.199228 Objective Loss 0.199228 LR 0.000500 Time 0.047783 +2023-10-02 21:18:54,684 - Epoch: [104][ 40/ 1236] Overall Loss 0.204014 Objective Loss 0.204014 LR 0.000500 Time 0.041092 +2023-10-02 21:18:54,895 - Epoch: [104][ 50/ 1236] Overall Loss 0.202191 Objective Loss 0.202191 LR 0.000500 Time 0.037086 +2023-10-02 21:18:55,106 - Epoch: [104][ 60/ 1236] Overall Loss 0.202999 Objective Loss 0.202999 LR 0.000500 Time 0.034422 +2023-10-02 21:18:55,316 - Epoch: [104][ 70/ 1236] Overall Loss 0.205038 Objective Loss 0.205038 LR 0.000500 Time 0.032507 +2023-10-02 21:18:55,528 - Epoch: [104][ 80/ 1236] Overall Loss 0.205056 Objective Loss 0.205056 LR 0.000500 Time 0.031080 +2023-10-02 21:18:55,738 - Epoch: [104][ 90/ 1236] Overall Loss 0.204233 Objective Loss 0.204233 LR 0.000500 Time 0.029963 +2023-10-02 21:18:55,949 - Epoch: [104][ 100/ 1236] Overall Loss 0.204727 Objective Loss 0.204727 LR 0.000500 Time 0.029075 +2023-10-02 21:18:56,160 - Epoch: [104][ 110/ 1236] Overall Loss 0.203546 Objective Loss 0.203546 LR 0.000500 Time 0.028345 +2023-10-02 21:18:56,371 - Epoch: [104][ 120/ 1236] Overall Loss 0.203597 Objective Loss 0.203597 LR 0.000500 Time 0.027739 +2023-10-02 21:18:56,582 - Epoch: [104][ 130/ 1236] Overall Loss 0.202650 Objective Loss 0.202650 LR 0.000500 Time 0.027225 +2023-10-02 21:18:56,794 - Epoch: [104][ 140/ 1236] Overall Loss 0.201765 Objective Loss 0.201765 LR 0.000500 Time 0.026790 +2023-10-02 21:18:57,004 - Epoch: [104][ 150/ 1236] Overall Loss 0.202371 Objective Loss 0.202371 LR 0.000500 Time 0.026405 +2023-10-02 21:18:57,215 - Epoch: [104][ 160/ 1236] Overall Loss 0.203710 Objective Loss 0.203710 LR 0.000500 Time 0.026070 +2023-10-02 21:18:57,426 - Epoch: [104][ 170/ 1236] Overall Loss 0.204768 Objective Loss 0.204768 LR 0.000500 Time 0.025774 +2023-10-02 21:18:57,637 - Epoch: [104][ 180/ 1236] Overall Loss 0.206567 Objective Loss 0.206567 LR 0.000500 Time 0.025513 +2023-10-02 21:18:57,847 - Epoch: [104][ 190/ 1236] Overall Loss 0.206919 Objective Loss 0.206919 LR 0.000500 Time 0.025275 +2023-10-02 21:18:58,057 - Epoch: [104][ 200/ 1236] Overall Loss 0.205268 Objective Loss 0.205268 LR 0.000500 Time 0.025060 +2023-10-02 21:18:58,266 - Epoch: [104][ 210/ 1236] Overall Loss 0.205201 Objective Loss 0.205201 LR 0.000500 Time 0.024862 +2023-10-02 21:18:58,476 - Epoch: [104][ 220/ 1236] Overall Loss 0.204910 Objective Loss 0.204910 LR 0.000500 Time 0.024685 +2023-10-02 21:18:58,685 - Epoch: [104][ 230/ 1236] Overall Loss 0.204966 Objective Loss 0.204966 LR 0.000500 Time 0.024520 +2023-10-02 21:18:58,896 - Epoch: [104][ 240/ 1236] Overall Loss 0.204639 Objective Loss 0.204639 LR 0.000500 Time 0.024373 +2023-10-02 21:18:59,106 - Epoch: [104][ 250/ 1236] Overall Loss 0.204064 Objective Loss 0.204064 LR 0.000500 Time 0.024238 +2023-10-02 21:18:59,317 - Epoch: [104][ 260/ 1236] Overall Loss 0.202447 Objective Loss 0.202447 LR 0.000500 Time 0.024117 +2023-10-02 21:18:59,528 - Epoch: [104][ 270/ 1236] Overall Loss 0.202618 Objective Loss 0.202618 LR 0.000500 Time 0.024004 +2023-10-02 21:18:59,740 - Epoch: [104][ 280/ 1236] Overall Loss 0.201981 Objective Loss 0.201981 LR 0.000500 Time 0.023901 +2023-10-02 21:18:59,951 - Epoch: [104][ 290/ 1236] Overall Loss 0.202291 Objective Loss 0.202291 LR 0.000500 Time 0.023804 +2023-10-02 21:19:00,161 - Epoch: [104][ 300/ 1236] Overall Loss 0.202873 Objective Loss 0.202873 LR 0.000500 Time 0.023712 +2023-10-02 21:19:00,371 - Epoch: [104][ 310/ 1236] Overall Loss 0.203515 Objective Loss 0.203515 LR 0.000500 Time 0.023622 +2023-10-02 21:19:00,581 - Epoch: [104][ 320/ 1236] Overall Loss 0.204206 Objective Loss 0.204206 LR 0.000500 Time 0.023540 +2023-10-02 21:19:00,791 - Epoch: [104][ 330/ 1236] Overall Loss 0.203839 Objective Loss 0.203839 LR 0.000500 Time 0.023463 +2023-10-02 21:19:01,002 - Epoch: [104][ 340/ 1236] Overall Loss 0.204033 Objective Loss 0.204033 LR 0.000500 Time 0.023391 +2023-10-02 21:19:01,213 - Epoch: [104][ 350/ 1236] Overall Loss 0.203784 Objective Loss 0.203784 LR 0.000500 Time 0.023324 +2023-10-02 21:19:01,424 - Epoch: [104][ 360/ 1236] Overall Loss 0.203876 Objective Loss 0.203876 LR 0.000500 Time 0.023262 +2023-10-02 21:19:01,635 - Epoch: [104][ 370/ 1236] Overall Loss 0.203549 Objective Loss 0.203549 LR 0.000500 Time 0.023202 +2023-10-02 21:19:01,846 - Epoch: [104][ 380/ 1236] Overall Loss 0.203791 Objective Loss 0.203791 LR 0.000500 Time 0.023147 +2023-10-02 21:19:02,057 - Epoch: [104][ 390/ 1236] Overall Loss 0.203430 Objective Loss 0.203430 LR 0.000500 Time 0.023093 +2023-10-02 21:19:02,268 - Epoch: [104][ 400/ 1236] Overall Loss 0.203356 Objective Loss 0.203356 LR 0.000500 Time 0.023043 +2023-10-02 21:19:02,479 - Epoch: [104][ 410/ 1236] Overall Loss 0.203796 Objective Loss 0.203796 LR 0.000500 Time 0.022995 +2023-10-02 21:19:02,690 - Epoch: [104][ 420/ 1236] Overall Loss 0.204437 Objective Loss 0.204437 LR 0.000500 Time 0.022949 +2023-10-02 21:19:02,901 - Epoch: [104][ 430/ 1236] Overall Loss 0.204903 Objective Loss 0.204903 LR 0.000500 Time 0.022905 +2023-10-02 21:19:03,112 - Epoch: [104][ 440/ 1236] Overall Loss 0.204850 Objective Loss 0.204850 LR 0.000500 Time 0.022864 +2023-10-02 21:19:03,323 - Epoch: [104][ 450/ 1236] Overall Loss 0.205386 Objective Loss 0.205386 LR 0.000500 Time 0.022823 +2023-10-02 21:19:03,534 - Epoch: [104][ 460/ 1236] Overall Loss 0.205468 Objective Loss 0.205468 LR 0.000500 Time 0.022785 +2023-10-02 21:19:03,745 - Epoch: [104][ 470/ 1236] Overall Loss 0.205507 Objective Loss 0.205507 LR 0.000500 Time 0.022748 +2023-10-02 21:19:03,956 - Epoch: [104][ 480/ 1236] Overall Loss 0.204952 Objective Loss 0.204952 LR 0.000500 Time 0.022713 +2023-10-02 21:19:04,167 - Epoch: [104][ 490/ 1236] Overall Loss 0.204742 Objective Loss 0.204742 LR 0.000500 Time 0.022679 +2023-10-02 21:19:04,378 - Epoch: [104][ 500/ 1236] Overall Loss 0.204738 Objective Loss 0.204738 LR 0.000500 Time 0.022647 +2023-10-02 21:19:04,588 - Epoch: [104][ 510/ 1236] Overall Loss 0.204461 Objective Loss 0.204461 LR 0.000500 Time 0.022615 +2023-10-02 21:19:04,799 - Epoch: [104][ 520/ 1236] Overall Loss 0.204375 Objective Loss 0.204375 LR 0.000500 Time 0.022586 +2023-10-02 21:19:05,010 - Epoch: [104][ 530/ 1236] Overall Loss 0.204157 Objective Loss 0.204157 LR 0.000500 Time 0.022557 +2023-10-02 21:19:05,222 - Epoch: [104][ 540/ 1236] Overall Loss 0.204832 Objective Loss 0.204832 LR 0.000500 Time 0.022530 +2023-10-02 21:19:05,432 - Epoch: [104][ 550/ 1236] Overall Loss 0.204384 Objective Loss 0.204384 LR 0.000500 Time 0.022503 +2023-10-02 21:19:05,643 - Epoch: [104][ 560/ 1236] Overall Loss 0.204819 Objective Loss 0.204819 LR 0.000500 Time 0.022477 +2023-10-02 21:19:05,854 - Epoch: [104][ 570/ 1236] Overall Loss 0.205616 Objective Loss 0.205616 LR 0.000500 Time 0.022452 +2023-10-02 21:19:06,065 - Epoch: [104][ 580/ 1236] Overall Loss 0.205587 Objective Loss 0.205587 LR 0.000500 Time 0.022429 +2023-10-02 21:19:06,276 - Epoch: [104][ 590/ 1236] Overall Loss 0.205727 Objective Loss 0.205727 LR 0.000500 Time 0.022405 +2023-10-02 21:19:06,487 - Epoch: [104][ 600/ 1236] Overall Loss 0.205575 Objective Loss 0.205575 LR 0.000500 Time 0.022383 +2023-10-02 21:19:06,698 - Epoch: [104][ 610/ 1236] Overall Loss 0.205355 Objective Loss 0.205355 LR 0.000500 Time 0.022361 +2023-10-02 21:19:06,909 - Epoch: [104][ 620/ 1236] Overall Loss 0.205612 Objective Loss 0.205612 LR 0.000500 Time 0.022340 +2023-10-02 21:19:07,120 - Epoch: [104][ 630/ 1236] Overall Loss 0.205551 Objective Loss 0.205551 LR 0.000500 Time 0.022320 +2023-10-02 21:19:07,331 - Epoch: [104][ 640/ 1236] Overall Loss 0.205679 Objective Loss 0.205679 LR 0.000500 Time 0.022301 +2023-10-02 21:19:07,542 - Epoch: [104][ 650/ 1236] Overall Loss 0.205469 Objective Loss 0.205469 LR 0.000500 Time 0.022281 +2023-10-02 21:19:07,753 - Epoch: [104][ 660/ 1236] Overall Loss 0.205211 Objective Loss 0.205211 LR 0.000500 Time 0.022263 +2023-10-02 21:19:07,964 - Epoch: [104][ 670/ 1236] Overall Loss 0.205663 Objective Loss 0.205663 LR 0.000500 Time 0.022245 +2023-10-02 21:19:08,175 - Epoch: [104][ 680/ 1236] Overall Loss 0.205456 Objective Loss 0.205456 LR 0.000500 Time 0.022227 +2023-10-02 21:19:08,385 - Epoch: [104][ 690/ 1236] Overall Loss 0.205060 Objective Loss 0.205060 LR 0.000500 Time 0.022210 +2023-10-02 21:19:08,596 - Epoch: [104][ 700/ 1236] Overall Loss 0.205090 Objective Loss 0.205090 LR 0.000500 Time 0.022194 +2023-10-02 21:19:08,807 - Epoch: [104][ 710/ 1236] Overall Loss 0.205278 Objective Loss 0.205278 LR 0.000500 Time 0.022178 +2023-10-02 21:19:09,018 - Epoch: [104][ 720/ 1236] Overall Loss 0.204959 Objective Loss 0.204959 LR 0.000500 Time 0.022163 +2023-10-02 21:19:09,229 - Epoch: [104][ 730/ 1236] Overall Loss 0.204787 Objective Loss 0.204787 LR 0.000500 Time 0.022147 +2023-10-02 21:19:09,440 - Epoch: [104][ 740/ 1236] Overall Loss 0.204519 Objective Loss 0.204519 LR 0.000500 Time 0.022133 +2023-10-02 21:19:09,649 - Epoch: [104][ 750/ 1236] Overall Loss 0.204487 Objective Loss 0.204487 LR 0.000500 Time 0.022116 +2023-10-02 21:19:09,859 - Epoch: [104][ 760/ 1236] Overall Loss 0.204354 Objective Loss 0.204354 LR 0.000500 Time 0.022100 +2023-10-02 21:19:10,069 - Epoch: [104][ 770/ 1236] Overall Loss 0.204446 Objective Loss 0.204446 LR 0.000500 Time 0.022086 +2023-10-02 21:19:10,278 - Epoch: [104][ 780/ 1236] Overall Loss 0.204395 Objective Loss 0.204395 LR 0.000500 Time 0.022070 +2023-10-02 21:19:10,486 - Epoch: [104][ 790/ 1236] Overall Loss 0.204449 Objective Loss 0.204449 LR 0.000500 Time 0.022054 +2023-10-02 21:19:10,695 - Epoch: [104][ 800/ 1236] Overall Loss 0.204234 Objective Loss 0.204234 LR 0.000500 Time 0.022039 +2023-10-02 21:19:10,904 - Epoch: [104][ 810/ 1236] Overall Loss 0.204148 Objective Loss 0.204148 LR 0.000500 Time 0.022024 +2023-10-02 21:19:11,113 - Epoch: [104][ 820/ 1236] Overall Loss 0.204180 Objective Loss 0.204180 LR 0.000500 Time 0.022010 +2023-10-02 21:19:11,321 - Epoch: [104][ 830/ 1236] Overall Loss 0.204217 Objective Loss 0.204217 LR 0.000500 Time 0.021996 +2023-10-02 21:19:11,531 - Epoch: [104][ 840/ 1236] Overall Loss 0.204207 Objective Loss 0.204207 LR 0.000500 Time 0.021983 +2023-10-02 21:19:11,739 - Epoch: [104][ 850/ 1236] Overall Loss 0.204229 Objective Loss 0.204229 LR 0.000500 Time 0.021969 +2023-10-02 21:19:11,949 - Epoch: [104][ 860/ 1236] Overall Loss 0.204303 Objective Loss 0.204303 LR 0.000500 Time 0.021957 +2023-10-02 21:19:12,157 - Epoch: [104][ 870/ 1236] Overall Loss 0.204229 Objective Loss 0.204229 LR 0.000500 Time 0.021943 +2023-10-02 21:19:12,366 - Epoch: [104][ 880/ 1236] Overall Loss 0.204352 Objective Loss 0.204352 LR 0.000500 Time 0.021931 +2023-10-02 21:19:12,574 - Epoch: [104][ 890/ 1236] Overall Loss 0.204471 Objective Loss 0.204471 LR 0.000500 Time 0.021918 +2023-10-02 21:19:12,784 - Epoch: [104][ 900/ 1236] Overall Loss 0.204297 Objective Loss 0.204297 LR 0.000500 Time 0.021907 +2023-10-02 21:19:12,992 - Epoch: [104][ 910/ 1236] Overall Loss 0.204104 Objective Loss 0.204104 LR 0.000500 Time 0.021895 +2023-10-02 21:19:13,201 - Epoch: [104][ 920/ 1236] Overall Loss 0.204456 Objective Loss 0.204456 LR 0.000500 Time 0.021884 +2023-10-02 21:19:13,409 - Epoch: [104][ 930/ 1236] Overall Loss 0.204430 Objective Loss 0.204430 LR 0.000500 Time 0.021872 +2023-10-02 21:19:13,619 - Epoch: [104][ 940/ 1236] Overall Loss 0.204671 Objective Loss 0.204671 LR 0.000500 Time 0.021862 +2023-10-02 21:19:13,827 - Epoch: [104][ 950/ 1236] Overall Loss 0.204696 Objective Loss 0.204696 LR 0.000500 Time 0.021851 +2023-10-02 21:19:14,036 - Epoch: [104][ 960/ 1236] Overall Loss 0.204762 Objective Loss 0.204762 LR 0.000500 Time 0.021841 +2023-10-02 21:19:14,244 - Epoch: [104][ 970/ 1236] Overall Loss 0.204946 Objective Loss 0.204946 LR 0.000500 Time 0.021830 +2023-10-02 21:19:14,454 - Epoch: [104][ 980/ 1236] Overall Loss 0.205150 Objective Loss 0.205150 LR 0.000500 Time 0.021820 +2023-10-02 21:19:14,662 - Epoch: [104][ 990/ 1236] Overall Loss 0.204956 Objective Loss 0.204956 LR 0.000500 Time 0.021810 +2023-10-02 21:19:14,871 - Epoch: [104][ 1000/ 1236] Overall Loss 0.205081 Objective Loss 0.205081 LR 0.000500 Time 0.021801 +2023-10-02 21:19:15,079 - Epoch: [104][ 1010/ 1236] Overall Loss 0.205084 Objective Loss 0.205084 LR 0.000500 Time 0.021791 +2023-10-02 21:19:15,289 - Epoch: [104][ 1020/ 1236] Overall Loss 0.204997 Objective Loss 0.204997 LR 0.000500 Time 0.021782 +2023-10-02 21:19:15,497 - Epoch: [104][ 1030/ 1236] Overall Loss 0.204921 Objective Loss 0.204921 LR 0.000500 Time 0.021773 +2023-10-02 21:19:15,707 - Epoch: [104][ 1040/ 1236] Overall Loss 0.204718 Objective Loss 0.204718 LR 0.000500 Time 0.021764 +2023-10-02 21:19:15,915 - Epoch: [104][ 1050/ 1236] Overall Loss 0.204822 Objective Loss 0.204822 LR 0.000500 Time 0.021755 +2023-10-02 21:19:16,124 - Epoch: [104][ 1060/ 1236] Overall Loss 0.204980 Objective Loss 0.204980 LR 0.000500 Time 0.021747 +2023-10-02 21:19:16,332 - Epoch: [104][ 1070/ 1236] Overall Loss 0.204908 Objective Loss 0.204908 LR 0.000500 Time 0.021738 +2023-10-02 21:19:16,542 - Epoch: [104][ 1080/ 1236] Overall Loss 0.204901 Objective Loss 0.204901 LR 0.000500 Time 0.021730 +2023-10-02 21:19:16,750 - Epoch: [104][ 1090/ 1236] Overall Loss 0.205185 Objective Loss 0.205185 LR 0.000500 Time 0.021722 +2023-10-02 21:19:16,959 - Epoch: [104][ 1100/ 1236] Overall Loss 0.205101 Objective Loss 0.205101 LR 0.000500 Time 0.021714 +2023-10-02 21:19:17,167 - Epoch: [104][ 1110/ 1236] Overall Loss 0.205223 Objective Loss 0.205223 LR 0.000500 Time 0.021706 +2023-10-02 21:19:17,377 - Epoch: [104][ 1120/ 1236] Overall Loss 0.205215 Objective Loss 0.205215 LR 0.000500 Time 0.021699 +2023-10-02 21:19:17,585 - Epoch: [104][ 1130/ 1236] Overall Loss 0.205099 Objective Loss 0.205099 LR 0.000500 Time 0.021691 +2023-10-02 21:19:17,795 - Epoch: [104][ 1140/ 1236] Overall Loss 0.204989 Objective Loss 0.204989 LR 0.000500 Time 0.021684 +2023-10-02 21:19:18,003 - Epoch: [104][ 1150/ 1236] Overall Loss 0.205042 Objective Loss 0.205042 LR 0.000500 Time 0.021676 +2023-10-02 21:19:18,212 - Epoch: [104][ 1160/ 1236] Overall Loss 0.205227 Objective Loss 0.205227 LR 0.000500 Time 0.021670 +2023-10-02 21:19:18,421 - Epoch: [104][ 1170/ 1236] Overall Loss 0.205281 Objective Loss 0.205281 LR 0.000500 Time 0.021662 +2023-10-02 21:19:18,630 - Epoch: [104][ 1180/ 1236] Overall Loss 0.205279 Objective Loss 0.205279 LR 0.000500 Time 0.021656 +2023-10-02 21:19:18,838 - Epoch: [104][ 1190/ 1236] Overall Loss 0.205109 Objective Loss 0.205109 LR 0.000500 Time 0.021649 +2023-10-02 21:19:19,048 - Epoch: [104][ 1200/ 1236] Overall Loss 0.204991 Objective Loss 0.204991 LR 0.000500 Time 0.021643 +2023-10-02 21:19:19,256 - Epoch: [104][ 1210/ 1236] Overall Loss 0.204954 Objective Loss 0.204954 LR 0.000500 Time 0.021636 +2023-10-02 21:19:19,466 - Epoch: [104][ 1220/ 1236] Overall Loss 0.204979 Objective Loss 0.204979 LR 0.000500 Time 0.021630 +2023-10-02 21:19:19,726 - Epoch: [104][ 1230/ 1236] Overall Loss 0.205128 Objective Loss 0.205128 LR 0.000500 Time 0.021666 +2023-10-02 21:19:19,848 - Epoch: [104][ 1236/ 1236] Overall Loss 0.205184 Objective Loss 0.205184 Top1 85.539715 Top5 97.963340 LR 0.000500 Time 0.021659 +2023-10-02 21:19:19,988 - --- validate (epoch=104)----------- +2023-10-02 21:19:19,988 - 29943 samples (256 per mini-batch) +2023-10-02 21:19:20,483 - Epoch: [104][ 10/ 117] Loss 0.343156 Top1 85.195312 Top5 98.281250 +2023-10-02 21:19:20,632 - Epoch: [104][ 20/ 117] Loss 0.311301 Top1 85.273438 Top5 98.378906 +2023-10-02 21:19:20,780 - Epoch: [104][ 30/ 117] Loss 0.321853 Top1 84.414062 Top5 98.307292 +2023-10-02 21:19:20,927 - Epoch: [104][ 40/ 117] Loss 0.327723 Top1 84.306641 Top5 98.281250 +2023-10-02 21:19:21,074 - Epoch: [104][ 50/ 117] Loss 0.332423 Top1 84.359375 Top5 98.289062 +2023-10-02 21:19:21,222 - Epoch: [104][ 60/ 117] Loss 0.331924 Top1 84.401042 Top5 98.274740 +2023-10-02 21:19:21,370 - Epoch: [104][ 70/ 117] Loss 0.333787 Top1 84.302455 Top5 98.208705 +2023-10-02 21:19:21,516 - Epoch: [104][ 80/ 117] Loss 0.331275 Top1 84.394531 Top5 98.261719 +2023-10-02 21:19:21,661 - Epoch: [104][ 90/ 117] Loss 0.328636 Top1 84.414062 Top5 98.307292 +2023-10-02 21:19:21,806 - Epoch: [104][ 100/ 117] Loss 0.326709 Top1 84.476562 Top5 98.316406 +2023-10-02 21:19:21,959 - Epoch: [104][ 110/ 117] Loss 0.323344 Top1 84.602273 Top5 98.316761 +2023-10-02 21:19:22,048 - Epoch: [104][ 117/ 117] Loss 0.321986 Top1 84.650837 Top5 98.330161 +2023-10-02 21:19:22,183 - ==> Top1: 84.651 Top5: 98.330 Loss: 0.322 + +2023-10-02 21:19:22,184 - ==> Confusion: +[[ 955 1 3 0 7 5 0 1 1 56 2 0 0 2 2 0 4 0 0 0 11] + [ 0 1067 0 0 6 13 0 21 1 1 2 1 0 0 0 3 1 0 9 2 4] + [ 3 0 980 14 2 0 15 3 0 0 2 0 9 3 0 3 1 2 5 3 11] + [ 0 4 14 972 0 1 2 2 1 1 7 0 7 5 35 2 1 7 11 0 17] + [ 24 8 0 0 965 6 0 1 2 10 1 1 3 1 8 4 8 1 2 0 5] + [ 3 49 0 0 4 967 2 32 2 9 2 12 3 11 4 0 4 1 3 0 8] + [ 0 3 26 2 0 1 1129 6 0 0 5 1 0 0 0 5 0 1 0 6 6] + [ 2 15 14 0 4 18 9 1066 2 3 4 5 7 6 2 0 1 5 39 11 5] + [ 19 3 1 0 1 1 0 0 971 45 7 2 5 11 14 1 2 0 4 1 1] + [ 106 1 3 0 8 1 0 0 14 947 1 1 1 19 6 0 1 0 1 2 7] + [ 1 4 11 10 0 1 1 6 12 0 962 0 1 19 4 1 2 4 3 0 11] + [ 0 1 0 2 1 5 0 4 0 0 0 969 24 6 0 1 0 18 0 1 3] + [ 0 1 1 1 0 1 2 0 0 2 2 38 965 3 6 5 1 16 3 8 13] + [ 1 0 2 0 4 10 1 1 11 11 4 3 2 1051 4 1 0 0 0 1 12] + [ 15 3 5 17 4 0 0 0 18 6 1 0 4 3 1004 0 1 4 6 0 10] + [ 0 0 0 1 5 0 4 0 0 0 1 8 8 0 1 1062 17 18 1 4 4] + [ 1 22 0 0 10 6 1 1 0 1 0 5 0 1 6 9 1078 0 1 6 13] + [ 0 0 0 6 0 0 4 0 0 0 0 3 21 0 5 4 2 988 0 2 3] + [ 2 4 2 15 1 1 1 20 4 2 3 2 2 0 13 0 1 0 985 1 9] + [ 0 4 4 1 0 2 13 15 0 0 0 16 4 0 1 3 7 1 0 1073 8] + [ 120 246 125 74 94 141 55 110 88 88 163 112 341 312 147 45 84 76 141 152 5191]] + +2023-10-02 21:19:22,185 - ==> Best [Top1: 84.788 Top5: 98.367 Sparsity:0.00 Params: 169472 on epoch: 103] +2023-10-02 21:19:22,185 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:19:22,192 - + +2023-10-02 21:19:22,192 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:19:23,221 - Epoch: [105][ 10/ 1236] Overall Loss 0.186689 Objective Loss 0.186689 LR 0.000500 Time 0.102931 +2023-10-02 21:19:23,430 - Epoch: [105][ 20/ 1236] Overall Loss 0.194432 Objective Loss 0.194432 LR 0.000500 Time 0.061860 +2023-10-02 21:19:23,638 - Epoch: [105][ 30/ 1236] Overall Loss 0.194526 Objective Loss 0.194526 LR 0.000500 Time 0.048160 +2023-10-02 21:19:23,846 - Epoch: [105][ 40/ 1236] Overall Loss 0.192391 Objective Loss 0.192391 LR 0.000500 Time 0.041330 +2023-10-02 21:19:24,054 - Epoch: [105][ 50/ 1236] Overall Loss 0.190800 Objective Loss 0.190800 LR 0.000500 Time 0.037190 +2023-10-02 21:19:24,264 - Epoch: [105][ 60/ 1236] Overall Loss 0.191728 Objective Loss 0.191728 LR 0.000500 Time 0.034487 +2023-10-02 21:19:24,471 - Epoch: [105][ 70/ 1236] Overall Loss 0.190757 Objective Loss 0.190757 LR 0.000500 Time 0.032506 +2023-10-02 21:19:24,681 - Epoch: [105][ 80/ 1236] Overall Loss 0.190830 Objective Loss 0.190830 LR 0.000500 Time 0.031067 +2023-10-02 21:19:24,887 - Epoch: [105][ 90/ 1236] Overall Loss 0.188677 Objective Loss 0.188677 LR 0.000500 Time 0.029907 +2023-10-02 21:19:25,096 - Epoch: [105][ 100/ 1236] Overall Loss 0.191607 Objective Loss 0.191607 LR 0.000500 Time 0.029003 +2023-10-02 21:19:25,304 - Epoch: [105][ 110/ 1236] Overall Loss 0.190369 Objective Loss 0.190369 LR 0.000500 Time 0.028241 +2023-10-02 21:19:25,517 - Epoch: [105][ 120/ 1236] Overall Loss 0.188917 Objective Loss 0.188917 LR 0.000500 Time 0.027655 +2023-10-02 21:19:25,726 - Epoch: [105][ 130/ 1236] Overall Loss 0.189378 Objective Loss 0.189378 LR 0.000500 Time 0.027134 +2023-10-02 21:19:25,938 - Epoch: [105][ 140/ 1236] Overall Loss 0.189895 Objective Loss 0.189895 LR 0.000500 Time 0.026714 +2023-10-02 21:19:26,149 - Epoch: [105][ 150/ 1236] Overall Loss 0.189363 Objective Loss 0.189363 LR 0.000500 Time 0.026327 +2023-10-02 21:19:26,361 - Epoch: [105][ 160/ 1236] Overall Loss 0.189778 Objective Loss 0.189778 LR 0.000500 Time 0.026004 +2023-10-02 21:19:26,569 - Epoch: [105][ 170/ 1236] Overall Loss 0.189188 Objective Loss 0.189188 LR 0.000500 Time 0.025697 +2023-10-02 21:19:26,780 - Epoch: [105][ 180/ 1236] Overall Loss 0.190459 Objective Loss 0.190459 LR 0.000500 Time 0.025439 +2023-10-02 21:19:26,988 - Epoch: [105][ 190/ 1236] Overall Loss 0.190362 Objective Loss 0.190362 LR 0.000500 Time 0.025188 +2023-10-02 21:19:27,199 - Epoch: [105][ 200/ 1236] Overall Loss 0.190160 Objective Loss 0.190160 LR 0.000500 Time 0.024981 +2023-10-02 21:19:27,407 - Epoch: [105][ 210/ 1236] Overall Loss 0.190290 Objective Loss 0.190290 LR 0.000500 Time 0.024776 +2023-10-02 21:19:27,618 - Epoch: [105][ 220/ 1236] Overall Loss 0.189471 Objective Loss 0.189471 LR 0.000500 Time 0.024606 +2023-10-02 21:19:27,826 - Epoch: [105][ 230/ 1236] Overall Loss 0.189849 Objective Loss 0.189849 LR 0.000500 Time 0.024435 +2023-10-02 21:19:28,037 - Epoch: [105][ 240/ 1236] Overall Loss 0.189367 Objective Loss 0.189367 LR 0.000500 Time 0.024294 +2023-10-02 21:19:28,245 - Epoch: [105][ 250/ 1236] Overall Loss 0.190274 Objective Loss 0.190274 LR 0.000500 Time 0.024149 +2023-10-02 21:19:28,456 - Epoch: [105][ 260/ 1236] Overall Loss 0.191248 Objective Loss 0.191248 LR 0.000500 Time 0.024029 +2023-10-02 21:19:28,664 - Epoch: [105][ 270/ 1236] Overall Loss 0.191045 Objective Loss 0.191045 LR 0.000500 Time 0.023904 +2023-10-02 21:19:28,875 - Epoch: [105][ 280/ 1236] Overall Loss 0.191060 Objective Loss 0.191060 LR 0.000500 Time 0.023802 +2023-10-02 21:19:29,083 - Epoch: [105][ 290/ 1236] Overall Loss 0.191562 Objective Loss 0.191562 LR 0.000500 Time 0.023694 +2023-10-02 21:19:29,294 - Epoch: [105][ 300/ 1236] Overall Loss 0.192326 Objective Loss 0.192326 LR 0.000500 Time 0.023606 +2023-10-02 21:19:29,502 - Epoch: [105][ 310/ 1236] Overall Loss 0.191962 Objective Loss 0.191962 LR 0.000500 Time 0.023513 +2023-10-02 21:19:29,714 - Epoch: [105][ 320/ 1236] Overall Loss 0.192166 Objective Loss 0.192166 LR 0.000500 Time 0.023438 +2023-10-02 21:19:29,924 - Epoch: [105][ 330/ 1236] Overall Loss 0.192709 Objective Loss 0.192709 LR 0.000500 Time 0.023359 +2023-10-02 21:19:30,135 - Epoch: [105][ 340/ 1236] Overall Loss 0.193255 Objective Loss 0.193255 LR 0.000500 Time 0.023293 +2023-10-02 21:19:30,345 - Epoch: [105][ 350/ 1236] Overall Loss 0.192971 Objective Loss 0.192971 LR 0.000500 Time 0.023222 +2023-10-02 21:19:30,556 - Epoch: [105][ 360/ 1236] Overall Loss 0.193876 Objective Loss 0.193876 LR 0.000500 Time 0.023164 +2023-10-02 21:19:30,766 - Epoch: [105][ 370/ 1236] Overall Loss 0.193868 Objective Loss 0.193868 LR 0.000500 Time 0.023100 +2023-10-02 21:19:30,978 - Epoch: [105][ 380/ 1236] Overall Loss 0.194197 Objective Loss 0.194197 LR 0.000500 Time 0.023049 +2023-10-02 21:19:31,187 - Epoch: [105][ 390/ 1236] Overall Loss 0.194586 Objective Loss 0.194586 LR 0.000500 Time 0.022990 +2023-10-02 21:19:31,399 - Epoch: [105][ 400/ 1236] Overall Loss 0.194690 Objective Loss 0.194690 LR 0.000500 Time 0.022944 +2023-10-02 21:19:31,608 - Epoch: [105][ 410/ 1236] Overall Loss 0.195062 Objective Loss 0.195062 LR 0.000500 Time 0.022892 +2023-10-02 21:19:31,820 - Epoch: [105][ 420/ 1236] Overall Loss 0.196205 Objective Loss 0.196205 LR 0.000500 Time 0.022850 +2023-10-02 21:19:32,029 - Epoch: [105][ 430/ 1236] Overall Loss 0.196255 Objective Loss 0.196255 LR 0.000500 Time 0.022803 +2023-10-02 21:19:32,241 - Epoch: [105][ 440/ 1236] Overall Loss 0.197045 Objective Loss 0.197045 LR 0.000500 Time 0.022764 +2023-10-02 21:19:32,450 - Epoch: [105][ 450/ 1236] Overall Loss 0.197075 Objective Loss 0.197075 LR 0.000500 Time 0.022721 +2023-10-02 21:19:32,662 - Epoch: [105][ 460/ 1236] Overall Loss 0.196997 Objective Loss 0.196997 LR 0.000500 Time 0.022686 +2023-10-02 21:19:32,872 - Epoch: [105][ 470/ 1236] Overall Loss 0.197122 Objective Loss 0.197122 LR 0.000500 Time 0.022646 +2023-10-02 21:19:33,084 - Epoch: [105][ 480/ 1236] Overall Loss 0.197363 Objective Loss 0.197363 LR 0.000500 Time 0.022615 +2023-10-02 21:19:33,293 - Epoch: [105][ 490/ 1236] Overall Loss 0.197319 Objective Loss 0.197319 LR 0.000500 Time 0.022578 +2023-10-02 21:19:33,505 - Epoch: [105][ 500/ 1236] Overall Loss 0.197500 Objective Loss 0.197500 LR 0.000500 Time 0.022549 +2023-10-02 21:19:33,714 - Epoch: [105][ 510/ 1236] Overall Loss 0.197784 Objective Loss 0.197784 LR 0.000500 Time 0.022515 +2023-10-02 21:19:33,926 - Epoch: [105][ 520/ 1236] Overall Loss 0.197879 Objective Loss 0.197879 LR 0.000500 Time 0.022488 +2023-10-02 21:19:34,136 - Epoch: [105][ 530/ 1236] Overall Loss 0.197564 Objective Loss 0.197564 LR 0.000500 Time 0.022457 +2023-10-02 21:19:34,347 - Epoch: [105][ 540/ 1236] Overall Loss 0.197382 Objective Loss 0.197382 LR 0.000500 Time 0.022432 +2023-10-02 21:19:34,557 - Epoch: [105][ 550/ 1236] Overall Loss 0.197317 Objective Loss 0.197317 LR 0.000500 Time 0.022402 +2023-10-02 21:19:34,769 - Epoch: [105][ 560/ 1236] Overall Loss 0.198111 Objective Loss 0.198111 LR 0.000500 Time 0.022380 +2023-10-02 21:19:34,978 - Epoch: [105][ 570/ 1236] Overall Loss 0.198249 Objective Loss 0.198249 LR 0.000500 Time 0.022353 +2023-10-02 21:19:35,190 - Epoch: [105][ 580/ 1236] Overall Loss 0.198420 Objective Loss 0.198420 LR 0.000500 Time 0.022332 +2023-10-02 21:19:35,400 - Epoch: [105][ 590/ 1236] Overall Loss 0.198414 Objective Loss 0.198414 LR 0.000500 Time 0.022306 +2023-10-02 21:19:35,611 - Epoch: [105][ 600/ 1236] Overall Loss 0.198772 Objective Loss 0.198772 LR 0.000500 Time 0.022286 +2023-10-02 21:19:35,821 - Epoch: [105][ 610/ 1236] Overall Loss 0.198711 Objective Loss 0.198711 LR 0.000500 Time 0.022262 +2023-10-02 21:19:36,033 - Epoch: [105][ 620/ 1236] Overall Loss 0.199038 Objective Loss 0.199038 LR 0.000500 Time 0.022245 +2023-10-02 21:19:36,242 - Epoch: [105][ 630/ 1236] Overall Loss 0.199455 Objective Loss 0.199455 LR 0.000500 Time 0.022222 +2023-10-02 21:19:36,454 - Epoch: [105][ 640/ 1236] Overall Loss 0.199515 Objective Loss 0.199515 LR 0.000500 Time 0.022205 +2023-10-02 21:19:36,664 - Epoch: [105][ 650/ 1236] Overall Loss 0.199892 Objective Loss 0.199892 LR 0.000500 Time 0.022183 +2023-10-02 21:19:36,876 - Epoch: [105][ 660/ 1236] Overall Loss 0.199752 Objective Loss 0.199752 LR 0.000500 Time 0.022168 +2023-10-02 21:19:37,085 - Epoch: [105][ 670/ 1236] Overall Loss 0.199849 Objective Loss 0.199849 LR 0.000500 Time 0.022148 +2023-10-02 21:19:37,297 - Epoch: [105][ 680/ 1236] Overall Loss 0.199780 Objective Loss 0.199780 LR 0.000500 Time 0.022132 +2023-10-02 21:19:37,506 - Epoch: [105][ 690/ 1236] Overall Loss 0.199762 Objective Loss 0.199762 LR 0.000500 Time 0.022113 +2023-10-02 21:19:37,718 - Epoch: [105][ 700/ 1236] Overall Loss 0.199864 Objective Loss 0.199864 LR 0.000500 Time 0.022099 +2023-10-02 21:19:37,927 - Epoch: [105][ 710/ 1236] Overall Loss 0.200263 Objective Loss 0.200263 LR 0.000500 Time 0.022081 +2023-10-02 21:19:38,139 - Epoch: [105][ 720/ 1236] Overall Loss 0.200290 Objective Loss 0.200290 LR 0.000500 Time 0.022068 +2023-10-02 21:19:38,349 - Epoch: [105][ 730/ 1236] Overall Loss 0.200234 Objective Loss 0.200234 LR 0.000500 Time 0.022050 +2023-10-02 21:19:38,561 - Epoch: [105][ 740/ 1236] Overall Loss 0.200232 Objective Loss 0.200232 LR 0.000500 Time 0.022038 +2023-10-02 21:19:38,770 - Epoch: [105][ 750/ 1236] Overall Loss 0.200383 Objective Loss 0.200383 LR 0.000500 Time 0.022022 +2023-10-02 21:19:38,982 - Epoch: [105][ 760/ 1236] Overall Loss 0.200545 Objective Loss 0.200545 LR 0.000500 Time 0.022010 +2023-10-02 21:19:39,192 - Epoch: [105][ 770/ 1236] Overall Loss 0.200637 Objective Loss 0.200637 LR 0.000500 Time 0.021995 +2023-10-02 21:19:39,404 - Epoch: [105][ 780/ 1236] Overall Loss 0.200665 Objective Loss 0.200665 LR 0.000500 Time 0.021984 +2023-10-02 21:19:39,614 - Epoch: [105][ 790/ 1236] Overall Loss 0.200906 Objective Loss 0.200906 LR 0.000500 Time 0.021970 +2023-10-02 21:19:39,825 - Epoch: [105][ 800/ 1236] Overall Loss 0.200822 Objective Loss 0.200822 LR 0.000500 Time 0.021959 +2023-10-02 21:19:40,035 - Epoch: [105][ 810/ 1236] Overall Loss 0.201017 Objective Loss 0.201017 LR 0.000500 Time 0.021945 +2023-10-02 21:19:40,247 - Epoch: [105][ 820/ 1236] Overall Loss 0.200862 Objective Loss 0.200862 LR 0.000500 Time 0.021935 +2023-10-02 21:19:40,457 - Epoch: [105][ 830/ 1236] Overall Loss 0.200842 Objective Loss 0.200842 LR 0.000500 Time 0.021922 +2023-10-02 21:19:40,668 - Epoch: [105][ 840/ 1236] Overall Loss 0.200606 Objective Loss 0.200606 LR 0.000500 Time 0.021913 +2023-10-02 21:19:40,878 - Epoch: [105][ 850/ 1236] Overall Loss 0.200812 Objective Loss 0.200812 LR 0.000500 Time 0.021900 +2023-10-02 21:19:41,090 - Epoch: [105][ 860/ 1236] Overall Loss 0.201057 Objective Loss 0.201057 LR 0.000500 Time 0.021891 +2023-10-02 21:19:41,300 - Epoch: [105][ 870/ 1236] Overall Loss 0.201193 Objective Loss 0.201193 LR 0.000500 Time 0.021879 +2023-10-02 21:19:41,512 - Epoch: [105][ 880/ 1236] Overall Loss 0.201191 Objective Loss 0.201191 LR 0.000500 Time 0.021871 +2023-10-02 21:19:41,722 - Epoch: [105][ 890/ 1236] Overall Loss 0.201251 Objective Loss 0.201251 LR 0.000500 Time 0.021859 +2023-10-02 21:19:41,934 - Epoch: [105][ 900/ 1236] Overall Loss 0.201086 Objective Loss 0.201086 LR 0.000500 Time 0.021852 +2023-10-02 21:19:42,144 - Epoch: [105][ 910/ 1236] Overall Loss 0.201004 Objective Loss 0.201004 LR 0.000500 Time 0.021841 +2023-10-02 21:19:42,356 - Epoch: [105][ 920/ 1236] Overall Loss 0.200908 Objective Loss 0.200908 LR 0.000500 Time 0.021833 +2023-10-02 21:19:42,565 - Epoch: [105][ 930/ 1236] Overall Loss 0.200781 Objective Loss 0.200781 LR 0.000500 Time 0.021822 +2023-10-02 21:19:42,777 - Epoch: [105][ 940/ 1236] Overall Loss 0.201015 Objective Loss 0.201015 LR 0.000500 Time 0.021815 +2023-10-02 21:19:42,987 - Epoch: [105][ 950/ 1236] Overall Loss 0.201381 Objective Loss 0.201381 LR 0.000500 Time 0.021804 +2023-10-02 21:19:43,197 - Epoch: [105][ 960/ 1236] Overall Loss 0.201220 Objective Loss 0.201220 LR 0.000500 Time 0.021796 +2023-10-02 21:19:43,406 - Epoch: [105][ 970/ 1236] Overall Loss 0.201203 Objective Loss 0.201203 LR 0.000500 Time 0.021785 +2023-10-02 21:19:43,617 - Epoch: [105][ 980/ 1236] Overall Loss 0.201310 Objective Loss 0.201310 LR 0.000500 Time 0.021778 +2023-10-02 21:19:43,826 - Epoch: [105][ 990/ 1236] Overall Loss 0.201611 Objective Loss 0.201611 LR 0.000500 Time 0.021768 +2023-10-02 21:19:44,037 - Epoch: [105][ 1000/ 1236] Overall Loss 0.201795 Objective Loss 0.201795 LR 0.000500 Time 0.021761 +2023-10-02 21:19:44,246 - Epoch: [105][ 1010/ 1236] Overall Loss 0.201647 Objective Loss 0.201647 LR 0.000500 Time 0.021751 +2023-10-02 21:19:44,457 - Epoch: [105][ 1020/ 1236] Overall Loss 0.201544 Objective Loss 0.201544 LR 0.000500 Time 0.021744 +2023-10-02 21:19:44,666 - Epoch: [105][ 1030/ 1236] Overall Loss 0.201594 Objective Loss 0.201594 LR 0.000500 Time 0.021734 +2023-10-02 21:19:44,877 - Epoch: [105][ 1040/ 1236] Overall Loss 0.201762 Objective Loss 0.201762 LR 0.000500 Time 0.021728 +2023-10-02 21:19:45,086 - Epoch: [105][ 1050/ 1236] Overall Loss 0.201447 Objective Loss 0.201447 LR 0.000500 Time 0.021718 +2023-10-02 21:19:45,297 - Epoch: [105][ 1060/ 1236] Overall Loss 0.201452 Objective Loss 0.201452 LR 0.000500 Time 0.021712 +2023-10-02 21:19:45,505 - Epoch: [105][ 1070/ 1236] Overall Loss 0.201751 Objective Loss 0.201751 LR 0.000500 Time 0.021703 +2023-10-02 21:19:45,716 - Epoch: [105][ 1080/ 1236] Overall Loss 0.201674 Objective Loss 0.201674 LR 0.000500 Time 0.021697 +2023-10-02 21:19:45,925 - Epoch: [105][ 1090/ 1236] Overall Loss 0.201596 Objective Loss 0.201596 LR 0.000500 Time 0.021688 +2023-10-02 21:19:46,136 - Epoch: [105][ 1100/ 1236] Overall Loss 0.201728 Objective Loss 0.201728 LR 0.000500 Time 0.021682 +2023-10-02 21:19:46,345 - Epoch: [105][ 1110/ 1236] Overall Loss 0.201839 Objective Loss 0.201839 LR 0.000500 Time 0.021674 +2023-10-02 21:19:46,556 - Epoch: [105][ 1120/ 1236] Overall Loss 0.202006 Objective Loss 0.202006 LR 0.000500 Time 0.021669 +2023-10-02 21:19:46,765 - Epoch: [105][ 1130/ 1236] Overall Loss 0.202394 Objective Loss 0.202394 LR 0.000500 Time 0.021660 +2023-10-02 21:19:46,976 - Epoch: [105][ 1140/ 1236] Overall Loss 0.202505 Objective Loss 0.202505 LR 0.000500 Time 0.021655 +2023-10-02 21:19:47,185 - Epoch: [105][ 1150/ 1236] Overall Loss 0.202519 Objective Loss 0.202519 LR 0.000500 Time 0.021647 +2023-10-02 21:19:47,396 - Epoch: [105][ 1160/ 1236] Overall Loss 0.202545 Objective Loss 0.202545 LR 0.000500 Time 0.021642 +2023-10-02 21:19:47,605 - Epoch: [105][ 1170/ 1236] Overall Loss 0.202538 Objective Loss 0.202538 LR 0.000500 Time 0.021635 +2023-10-02 21:19:47,816 - Epoch: [105][ 1180/ 1236] Overall Loss 0.202655 Objective Loss 0.202655 LR 0.000500 Time 0.021630 +2023-10-02 21:19:48,025 - Epoch: [105][ 1190/ 1236] Overall Loss 0.202619 Objective Loss 0.202619 LR 0.000500 Time 0.021623 +2023-10-02 21:19:48,236 - Epoch: [105][ 1200/ 1236] Overall Loss 0.203006 Objective Loss 0.203006 LR 0.000500 Time 0.021618 +2023-10-02 21:19:48,445 - Epoch: [105][ 1210/ 1236] Overall Loss 0.203148 Objective Loss 0.203148 LR 0.000500 Time 0.021611 +2023-10-02 21:19:48,656 - Epoch: [105][ 1220/ 1236] Overall Loss 0.203095 Objective Loss 0.203095 LR 0.000500 Time 0.021606 +2023-10-02 21:19:48,920 - Epoch: [105][ 1230/ 1236] Overall Loss 0.203279 Objective Loss 0.203279 LR 0.000500 Time 0.021643 +2023-10-02 21:19:49,042 - Epoch: [105][ 1236/ 1236] Overall Loss 0.203382 Objective Loss 0.203382 Top1 87.576375 Top5 98.778004 LR 0.000500 Time 0.021637 +2023-10-02 21:19:49,189 - --- validate (epoch=105)----------- +2023-10-02 21:19:49,190 - 29943 samples (256 per mini-batch) +2023-10-02 21:19:49,672 - Epoch: [105][ 10/ 117] Loss 0.321266 Top1 85.312500 Top5 98.398438 +2023-10-02 21:19:49,818 - Epoch: [105][ 20/ 117] Loss 0.322510 Top1 85.175781 Top5 98.417969 +2023-10-02 21:19:49,964 - Epoch: [105][ 30/ 117] Loss 0.307939 Top1 85.520833 Top5 98.476562 +2023-10-02 21:19:50,109 - Epoch: [105][ 40/ 117] Loss 0.319037 Top1 85.576172 Top5 98.427734 +2023-10-02 21:19:50,255 - Epoch: [105][ 50/ 117] Loss 0.316643 Top1 85.335938 Top5 98.437500 +2023-10-02 21:19:50,401 - Epoch: [105][ 60/ 117] Loss 0.325039 Top1 85.286458 Top5 98.359375 +2023-10-02 21:19:50,546 - Epoch: [105][ 70/ 117] Loss 0.325269 Top1 85.228795 Top5 98.303571 +2023-10-02 21:19:50,692 - Epoch: [105][ 80/ 117] Loss 0.321346 Top1 85.175781 Top5 98.271484 +2023-10-02 21:19:50,837 - Epoch: [105][ 90/ 117] Loss 0.315963 Top1 85.299479 Top5 98.315972 +2023-10-02 21:19:50,982 - Epoch: [105][ 100/ 117] Loss 0.319833 Top1 85.312500 Top5 98.339844 +2023-10-02 21:19:51,135 - Epoch: [105][ 110/ 117] Loss 0.322766 Top1 85.376420 Top5 98.345170 +2023-10-02 21:19:51,224 - Epoch: [105][ 117/ 117] Loss 0.321494 Top1 85.375547 Top5 98.320142 +2023-10-02 21:19:51,354 - ==> Top1: 85.376 Top5: 98.320 Loss: 0.321 + +2023-10-02 21:19:51,354 - ==> Confusion: +[[ 950 0 4 0 6 2 0 0 8 43 2 1 1 2 5 1 5 1 2 0 17] + [ 2 1049 0 0 6 21 0 26 3 1 1 0 0 1 0 3 1 0 11 2 4] + [ 2 0 968 7 2 0 25 7 0 2 4 2 6 2 1 2 1 1 14 2 8] + [ 1 6 15 954 0 3 1 3 3 1 11 0 5 3 36 1 0 7 16 2 21] + [ 25 10 2 0 957 6 1 0 2 10 1 2 1 3 7 4 9 0 1 1 8] + [ 3 32 1 0 4 974 2 33 0 4 2 11 4 11 7 0 3 0 5 4 16] + [ 1 4 23 1 0 0 1129 6 0 0 4 2 0 0 1 2 0 1 0 10 7] + [ 1 14 13 0 2 19 9 1071 1 3 4 5 4 3 1 0 2 6 44 10 6] + [ 14 1 0 2 1 3 0 2 983 41 10 3 1 8 11 0 5 0 3 1 0] + [ 120 1 2 0 5 0 0 0 27 923 1 2 1 16 9 2 1 0 0 3 6] + [ 3 4 10 4 0 1 5 4 14 3 955 3 1 23 4 0 0 3 5 0 11] + [ 0 2 1 0 0 15 0 3 0 0 0 937 38 6 0 2 2 17 0 9 3] + [ 1 0 1 2 0 1 1 1 0 0 4 31 984 3 2 4 4 9 1 6 13] + [ 1 0 2 0 1 7 2 0 13 12 2 6 0 1055 5 1 0 2 0 1 9] + [ 11 4 3 22 2 1 0 0 28 5 0 0 3 2 998 0 3 2 7 0 10] + [ 0 0 1 1 3 0 1 0 0 1 1 7 5 0 1 1062 22 11 2 7 9] + [ 1 14 0 0 6 7 0 0 1 0 0 5 0 1 5 6 1091 0 1 6 17] + [ 0 0 2 4 0 0 1 0 0 0 0 2 24 0 4 4 0 991 0 2 4] + [ 3 3 1 11 0 0 1 17 4 0 3 1 5 0 11 1 0 0 995 0 12] + [ 0 1 5 0 1 2 6 6 0 0 1 9 6 3 0 2 9 1 2 1089 9] + [ 145 169 105 50 76 135 36 102 94 78 166 101 324 259 129 34 110 58 149 136 5449]] + +2023-10-02 21:19:51,356 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:19:51,356 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:19:51,369 - + +2023-10-02 21:19:51,369 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:19:52,387 - Epoch: [106][ 10/ 1236] Overall Loss 0.187900 Objective Loss 0.187900 LR 0.000500 Time 0.101697 +2023-10-02 21:19:52,598 - Epoch: [106][ 20/ 1236] Overall Loss 0.189289 Objective Loss 0.189289 LR 0.000500 Time 0.061398 +2023-10-02 21:19:52,805 - Epoch: [106][ 30/ 1236] Overall Loss 0.190264 Objective Loss 0.190264 LR 0.000500 Time 0.047809 +2023-10-02 21:19:53,014 - Epoch: [106][ 40/ 1236] Overall Loss 0.192601 Objective Loss 0.192601 LR 0.000500 Time 0.041085 +2023-10-02 21:19:53,220 - Epoch: [106][ 50/ 1236] Overall Loss 0.198401 Objective Loss 0.198401 LR 0.000500 Time 0.036984 +2023-10-02 21:19:53,428 - Epoch: [106][ 60/ 1236] Overall Loss 0.195971 Objective Loss 0.195971 LR 0.000500 Time 0.034273 +2023-10-02 21:19:53,635 - Epoch: [106][ 70/ 1236] Overall Loss 0.197149 Objective Loss 0.197149 LR 0.000500 Time 0.032316 +2023-10-02 21:19:53,843 - Epoch: [106][ 80/ 1236] Overall Loss 0.196521 Objective Loss 0.196521 LR 0.000500 Time 0.030875 +2023-10-02 21:19:54,050 - Epoch: [106][ 90/ 1236] Overall Loss 0.195801 Objective Loss 0.195801 LR 0.000500 Time 0.029726 +2023-10-02 21:19:54,258 - Epoch: [106][ 100/ 1236] Overall Loss 0.198955 Objective Loss 0.198955 LR 0.000500 Time 0.028836 +2023-10-02 21:19:54,466 - Epoch: [106][ 110/ 1236] Overall Loss 0.198058 Objective Loss 0.198058 LR 0.000500 Time 0.028085 +2023-10-02 21:19:54,673 - Epoch: [106][ 120/ 1236] Overall Loss 0.198110 Objective Loss 0.198110 LR 0.000500 Time 0.027466 +2023-10-02 21:19:54,880 - Epoch: [106][ 130/ 1236] Overall Loss 0.197608 Objective Loss 0.197608 LR 0.000500 Time 0.026945 +2023-10-02 21:19:55,088 - Epoch: [106][ 140/ 1236] Overall Loss 0.197601 Objective Loss 0.197601 LR 0.000500 Time 0.026507 +2023-10-02 21:19:55,295 - Epoch: [106][ 150/ 1236] Overall Loss 0.197941 Objective Loss 0.197941 LR 0.000500 Time 0.026112 +2023-10-02 21:19:55,506 - Epoch: [106][ 160/ 1236] Overall Loss 0.197140 Objective Loss 0.197140 LR 0.000500 Time 0.025797 +2023-10-02 21:19:55,714 - Epoch: [106][ 170/ 1236] Overall Loss 0.197431 Objective Loss 0.197431 LR 0.000500 Time 0.025496 +2023-10-02 21:19:55,924 - Epoch: [106][ 180/ 1236] Overall Loss 0.197275 Objective Loss 0.197275 LR 0.000500 Time 0.025249 +2023-10-02 21:19:56,132 - Epoch: [106][ 190/ 1236] Overall Loss 0.198684 Objective Loss 0.198684 LR 0.000500 Time 0.025011 +2023-10-02 21:19:56,343 - Epoch: [106][ 200/ 1236] Overall Loss 0.198950 Objective Loss 0.198950 LR 0.000500 Time 0.024815 +2023-10-02 21:19:56,550 - Epoch: [106][ 210/ 1236] Overall Loss 0.200043 Objective Loss 0.200043 LR 0.000500 Time 0.024619 +2023-10-02 21:19:56,762 - Epoch: [106][ 220/ 1236] Overall Loss 0.198873 Objective Loss 0.198873 LR 0.000500 Time 0.024459 +2023-10-02 21:19:56,969 - Epoch: [106][ 230/ 1236] Overall Loss 0.198243 Objective Loss 0.198243 LR 0.000500 Time 0.024295 +2023-10-02 21:19:57,180 - Epoch: [106][ 240/ 1236] Overall Loss 0.197865 Objective Loss 0.197865 LR 0.000500 Time 0.024162 +2023-10-02 21:19:57,387 - Epoch: [106][ 250/ 1236] Overall Loss 0.197641 Objective Loss 0.197641 LR 0.000500 Time 0.024024 +2023-10-02 21:19:57,598 - Epoch: [106][ 260/ 1236] Overall Loss 0.196221 Objective Loss 0.196221 LR 0.000500 Time 0.023911 +2023-10-02 21:19:57,806 - Epoch: [106][ 270/ 1236] Overall Loss 0.196242 Objective Loss 0.196242 LR 0.000500 Time 0.023791 +2023-10-02 21:19:58,017 - Epoch: [106][ 280/ 1236] Overall Loss 0.195746 Objective Loss 0.195746 LR 0.000500 Time 0.023694 +2023-10-02 21:19:58,224 - Epoch: [106][ 290/ 1236] Overall Loss 0.194923 Objective Loss 0.194923 LR 0.000500 Time 0.023591 +2023-10-02 21:19:58,435 - Epoch: [106][ 300/ 1236] Overall Loss 0.194936 Objective Loss 0.194936 LR 0.000500 Time 0.023509 +2023-10-02 21:19:58,643 - Epoch: [106][ 310/ 1236] Overall Loss 0.194115 Objective Loss 0.194115 LR 0.000500 Time 0.023418 +2023-10-02 21:19:58,854 - Epoch: [106][ 320/ 1236] Overall Loss 0.193539 Objective Loss 0.193539 LR 0.000500 Time 0.023345 +2023-10-02 21:19:59,060 - Epoch: [106][ 330/ 1236] Overall Loss 0.193561 Objective Loss 0.193561 LR 0.000500 Time 0.023261 +2023-10-02 21:19:59,267 - Epoch: [106][ 340/ 1236] Overall Loss 0.192868 Objective Loss 0.192868 LR 0.000500 Time 0.023186 +2023-10-02 21:19:59,474 - Epoch: [106][ 350/ 1236] Overall Loss 0.193243 Objective Loss 0.193243 LR 0.000500 Time 0.023115 +2023-10-02 21:19:59,683 - Epoch: [106][ 360/ 1236] Overall Loss 0.193088 Objective Loss 0.193088 LR 0.000500 Time 0.023051 +2023-10-02 21:19:59,890 - Epoch: [106][ 370/ 1236] Overall Loss 0.193167 Objective Loss 0.193167 LR 0.000500 Time 0.022985 +2023-10-02 21:20:00,099 - Epoch: [106][ 380/ 1236] Overall Loss 0.194012 Objective Loss 0.194012 LR 0.000500 Time 0.022928 +2023-10-02 21:20:00,306 - Epoch: [106][ 390/ 1236] Overall Loss 0.193869 Objective Loss 0.193869 LR 0.000500 Time 0.022866 +2023-10-02 21:20:00,514 - Epoch: [106][ 400/ 1236] Overall Loss 0.193827 Objective Loss 0.193827 LR 0.000500 Time 0.022815 +2023-10-02 21:20:00,722 - Epoch: [106][ 410/ 1236] Overall Loss 0.193804 Objective Loss 0.193804 LR 0.000500 Time 0.022761 +2023-10-02 21:20:00,930 - Epoch: [106][ 420/ 1236] Overall Loss 0.194360 Objective Loss 0.194360 LR 0.000500 Time 0.022715 +2023-10-02 21:20:01,137 - Epoch: [106][ 430/ 1236] Overall Loss 0.194156 Objective Loss 0.194156 LR 0.000500 Time 0.022664 +2023-10-02 21:20:01,346 - Epoch: [106][ 440/ 1236] Overall Loss 0.194471 Objective Loss 0.194471 LR 0.000500 Time 0.022623 +2023-10-02 21:20:01,553 - Epoch: [106][ 450/ 1236] Overall Loss 0.194433 Objective Loss 0.194433 LR 0.000500 Time 0.022578 +2023-10-02 21:20:01,762 - Epoch: [106][ 460/ 1236] Overall Loss 0.194781 Objective Loss 0.194781 LR 0.000500 Time 0.022540 +2023-10-02 21:20:01,970 - Epoch: [106][ 470/ 1236] Overall Loss 0.194868 Objective Loss 0.194868 LR 0.000500 Time 0.022500 +2023-10-02 21:20:02,179 - Epoch: [106][ 480/ 1236] Overall Loss 0.194978 Objective Loss 0.194978 LR 0.000500 Time 0.022466 +2023-10-02 21:20:02,387 - Epoch: [106][ 490/ 1236] Overall Loss 0.195000 Objective Loss 0.195000 LR 0.000500 Time 0.022428 +2023-10-02 21:20:02,595 - Epoch: [106][ 500/ 1236] Overall Loss 0.195749 Objective Loss 0.195749 LR 0.000500 Time 0.022396 +2023-10-02 21:20:02,803 - Epoch: [106][ 510/ 1236] Overall Loss 0.195658 Objective Loss 0.195658 LR 0.000500 Time 0.022361 +2023-10-02 21:20:03,011 - Epoch: [106][ 520/ 1236] Overall Loss 0.195450 Objective Loss 0.195450 LR 0.000500 Time 0.022331 +2023-10-02 21:20:03,219 - Epoch: [106][ 530/ 1236] Overall Loss 0.195527 Objective Loss 0.195527 LR 0.000500 Time 0.022299 +2023-10-02 21:20:03,428 - Epoch: [106][ 540/ 1236] Overall Loss 0.195835 Objective Loss 0.195835 LR 0.000500 Time 0.022272 +2023-10-02 21:20:03,636 - Epoch: [106][ 550/ 1236] Overall Loss 0.196085 Objective Loss 0.196085 LR 0.000500 Time 0.022242 +2023-10-02 21:20:03,844 - Epoch: [106][ 560/ 1236] Overall Loss 0.196063 Objective Loss 0.196063 LR 0.000500 Time 0.022217 +2023-10-02 21:20:04,052 - Epoch: [106][ 570/ 1236] Overall Loss 0.196154 Objective Loss 0.196154 LR 0.000500 Time 0.022189 +2023-10-02 21:20:04,260 - Epoch: [106][ 580/ 1236] Overall Loss 0.196096 Objective Loss 0.196096 LR 0.000500 Time 0.022165 +2023-10-02 21:20:04,468 - Epoch: [106][ 590/ 1236] Overall Loss 0.196895 Objective Loss 0.196895 LR 0.000500 Time 0.022140 +2023-10-02 21:20:04,677 - Epoch: [106][ 600/ 1236] Overall Loss 0.196796 Objective Loss 0.196796 LR 0.000500 Time 0.022118 +2023-10-02 21:20:04,886 - Epoch: [106][ 610/ 1236] Overall Loss 0.196642 Objective Loss 0.196642 LR 0.000500 Time 0.022096 +2023-10-02 21:20:05,094 - Epoch: [106][ 620/ 1236] Overall Loss 0.197597 Objective Loss 0.197597 LR 0.000500 Time 0.022076 +2023-10-02 21:20:05,303 - Epoch: [106][ 630/ 1236] Overall Loss 0.199566 Objective Loss 0.199566 LR 0.000500 Time 0.022054 +2023-10-02 21:20:05,511 - Epoch: [106][ 640/ 1236] Overall Loss 0.201194 Objective Loss 0.201194 LR 0.000500 Time 0.022035 +2023-10-02 21:20:05,720 - Epoch: [106][ 650/ 1236] Overall Loss 0.202982 Objective Loss 0.202982 LR 0.000500 Time 0.022017 +2023-10-02 21:20:05,928 - Epoch: [106][ 660/ 1236] Overall Loss 0.204022 Objective Loss 0.204022 LR 0.000500 Time 0.021998 +2023-10-02 21:20:06,137 - Epoch: [106][ 670/ 1236] Overall Loss 0.205256 Objective Loss 0.205256 LR 0.000500 Time 0.021979 +2023-10-02 21:20:06,345 - Epoch: [106][ 680/ 1236] Overall Loss 0.206652 Objective Loss 0.206652 LR 0.000500 Time 0.021962 +2023-10-02 21:20:06,553 - Epoch: [106][ 690/ 1236] Overall Loss 0.207942 Objective Loss 0.207942 LR 0.000500 Time 0.021943 +2023-10-02 21:20:06,762 - Epoch: [106][ 700/ 1236] Overall Loss 0.209402 Objective Loss 0.209402 LR 0.000500 Time 0.021927 +2023-10-02 21:20:06,970 - Epoch: [106][ 710/ 1236] Overall Loss 0.210013 Objective Loss 0.210013 LR 0.000500 Time 0.021909 +2023-10-02 21:20:07,179 - Epoch: [106][ 720/ 1236] Overall Loss 0.210859 Objective Loss 0.210859 LR 0.000500 Time 0.021894 +2023-10-02 21:20:07,387 - Epoch: [106][ 730/ 1236] Overall Loss 0.211758 Objective Loss 0.211758 LR 0.000500 Time 0.021878 +2023-10-02 21:20:07,595 - Epoch: [106][ 740/ 1236] Overall Loss 0.212559 Objective Loss 0.212559 LR 0.000500 Time 0.021863 +2023-10-02 21:20:07,804 - Epoch: [106][ 750/ 1236] Overall Loss 0.213372 Objective Loss 0.213372 LR 0.000500 Time 0.021848 +2023-10-02 21:20:08,012 - Epoch: [106][ 760/ 1236] Overall Loss 0.214226 Objective Loss 0.214226 LR 0.000500 Time 0.021834 +2023-10-02 21:20:08,221 - Epoch: [106][ 770/ 1236] Overall Loss 0.214778 Objective Loss 0.214778 LR 0.000500 Time 0.021819 +2023-10-02 21:20:08,429 - Epoch: [106][ 780/ 1236] Overall Loss 0.215583 Objective Loss 0.215583 LR 0.000500 Time 0.021806 +2023-10-02 21:20:08,638 - Epoch: [106][ 790/ 1236] Overall Loss 0.215997 Objective Loss 0.215997 LR 0.000500 Time 0.021792 +2023-10-02 21:20:08,846 - Epoch: [106][ 800/ 1236] Overall Loss 0.216595 Objective Loss 0.216595 LR 0.000500 Time 0.021781 +2023-10-02 21:20:09,055 - Epoch: [106][ 810/ 1236] Overall Loss 0.217331 Objective Loss 0.217331 LR 0.000500 Time 0.021767 +2023-10-02 21:20:09,263 - Epoch: [106][ 820/ 1236] Overall Loss 0.217891 Objective Loss 0.217891 LR 0.000500 Time 0.021755 +2023-10-02 21:20:09,472 - Epoch: [106][ 830/ 1236] Overall Loss 0.218561 Objective Loss 0.218561 LR 0.000500 Time 0.021743 +2023-10-02 21:20:09,680 - Epoch: [106][ 840/ 1236] Overall Loss 0.219068 Objective Loss 0.219068 LR 0.000500 Time 0.021732 +2023-10-02 21:20:09,889 - Epoch: [106][ 850/ 1236] Overall Loss 0.219600 Objective Loss 0.219600 LR 0.000500 Time 0.021720 +2023-10-02 21:20:10,098 - Epoch: [106][ 860/ 1236] Overall Loss 0.220223 Objective Loss 0.220223 LR 0.000500 Time 0.021710 +2023-10-02 21:20:10,306 - Epoch: [106][ 870/ 1236] Overall Loss 0.220621 Objective Loss 0.220621 LR 0.000500 Time 0.021698 +2023-10-02 21:20:10,515 - Epoch: [106][ 880/ 1236] Overall Loss 0.221254 Objective Loss 0.221254 LR 0.000500 Time 0.021688 +2023-10-02 21:20:10,723 - Epoch: [106][ 890/ 1236] Overall Loss 0.221700 Objective Loss 0.221700 LR 0.000500 Time 0.021677 +2023-10-02 21:20:10,932 - Epoch: [106][ 900/ 1236] Overall Loss 0.221948 Objective Loss 0.221948 LR 0.000500 Time 0.021667 +2023-10-02 21:20:11,141 - Epoch: [106][ 910/ 1236] Overall Loss 0.222165 Objective Loss 0.222165 LR 0.000500 Time 0.021657 +2023-10-02 21:20:11,349 - Epoch: [106][ 920/ 1236] Overall Loss 0.222787 Objective Loss 0.222787 LR 0.000500 Time 0.021648 +2023-10-02 21:20:11,558 - Epoch: [106][ 930/ 1236] Overall Loss 0.223132 Objective Loss 0.223132 LR 0.000500 Time 0.021638 +2023-10-02 21:20:11,766 - Epoch: [106][ 940/ 1236] Overall Loss 0.223401 Objective Loss 0.223401 LR 0.000500 Time 0.021629 +2023-10-02 21:20:11,975 - Epoch: [106][ 950/ 1236] Overall Loss 0.223733 Objective Loss 0.223733 LR 0.000500 Time 0.021619 +2023-10-02 21:20:12,183 - Epoch: [106][ 960/ 1236] Overall Loss 0.223936 Objective Loss 0.223936 LR 0.000500 Time 0.021611 +2023-10-02 21:20:12,392 - Epoch: [106][ 970/ 1236] Overall Loss 0.224178 Objective Loss 0.224178 LR 0.000500 Time 0.021602 +2023-10-02 21:20:12,601 - Epoch: [106][ 980/ 1236] Overall Loss 0.224389 Objective Loss 0.224389 LR 0.000500 Time 0.021594 +2023-10-02 21:20:12,809 - Epoch: [106][ 990/ 1236] Overall Loss 0.224761 Objective Loss 0.224761 LR 0.000500 Time 0.021585 +2023-10-02 21:20:13,018 - Epoch: [106][ 1000/ 1236] Overall Loss 0.225019 Objective Loss 0.225019 LR 0.000500 Time 0.021577 +2023-10-02 21:20:13,226 - Epoch: [106][ 1010/ 1236] Overall Loss 0.225363 Objective Loss 0.225363 LR 0.000500 Time 0.021568 +2023-10-02 21:20:13,434 - Epoch: [106][ 1020/ 1236] Overall Loss 0.225577 Objective Loss 0.225577 LR 0.000500 Time 0.021561 +2023-10-02 21:20:13,643 - Epoch: [106][ 1030/ 1236] Overall Loss 0.225847 Objective Loss 0.225847 LR 0.000500 Time 0.021553 +2023-10-02 21:20:13,852 - Epoch: [106][ 1040/ 1236] Overall Loss 0.225955 Objective Loss 0.225955 LR 0.000500 Time 0.021546 +2023-10-02 21:20:14,060 - Epoch: [106][ 1050/ 1236] Overall Loss 0.226210 Objective Loss 0.226210 LR 0.000500 Time 0.021538 +2023-10-02 21:20:14,269 - Epoch: [106][ 1060/ 1236] Overall Loss 0.226497 Objective Loss 0.226497 LR 0.000500 Time 0.021532 +2023-10-02 21:20:14,477 - Epoch: [106][ 1070/ 1236] Overall Loss 0.226819 Objective Loss 0.226819 LR 0.000500 Time 0.021523 +2023-10-02 21:20:14,685 - Epoch: [106][ 1080/ 1236] Overall Loss 0.227517 Objective Loss 0.227517 LR 0.000500 Time 0.021516 +2023-10-02 21:20:14,894 - Epoch: [106][ 1090/ 1236] Overall Loss 0.227562 Objective Loss 0.227562 LR 0.000500 Time 0.021509 +2023-10-02 21:20:15,103 - Epoch: [106][ 1100/ 1236] Overall Loss 0.227819 Objective Loss 0.227819 LR 0.000500 Time 0.021503 +2023-10-02 21:20:15,311 - Epoch: [106][ 1110/ 1236] Overall Loss 0.228091 Objective Loss 0.228091 LR 0.000500 Time 0.021495 +2023-10-02 21:20:15,519 - Epoch: [106][ 1120/ 1236] Overall Loss 0.228302 Objective Loss 0.228302 LR 0.000500 Time 0.021489 +2023-10-02 21:20:15,727 - Epoch: [106][ 1130/ 1236] Overall Loss 0.228583 Objective Loss 0.228583 LR 0.000500 Time 0.021482 +2023-10-02 21:20:15,936 - Epoch: [106][ 1140/ 1236] Overall Loss 0.228928 Objective Loss 0.228928 LR 0.000500 Time 0.021477 +2023-10-02 21:20:16,144 - Epoch: [106][ 1150/ 1236] Overall Loss 0.229166 Objective Loss 0.229166 LR 0.000500 Time 0.021470 +2023-10-02 21:20:16,353 - Epoch: [106][ 1160/ 1236] Overall Loss 0.229175 Objective Loss 0.229175 LR 0.000500 Time 0.021465 +2023-10-02 21:20:16,561 - Epoch: [106][ 1170/ 1236] Overall Loss 0.229277 Objective Loss 0.229277 LR 0.000500 Time 0.021458 +2023-10-02 21:20:16,770 - Epoch: [106][ 1180/ 1236] Overall Loss 0.229321 Objective Loss 0.229321 LR 0.000500 Time 0.021453 +2023-10-02 21:20:16,978 - Epoch: [106][ 1190/ 1236] Overall Loss 0.229407 Objective Loss 0.229407 LR 0.000500 Time 0.021446 +2023-10-02 21:20:17,187 - Epoch: [106][ 1200/ 1236] Overall Loss 0.229415 Objective Loss 0.229415 LR 0.000500 Time 0.021441 +2023-10-02 21:20:17,395 - Epoch: [106][ 1210/ 1236] Overall Loss 0.229640 Objective Loss 0.229640 LR 0.000500 Time 0.021435 +2023-10-02 21:20:17,604 - Epoch: [106][ 1220/ 1236] Overall Loss 0.229810 Objective Loss 0.229810 LR 0.000500 Time 0.021430 +2023-10-02 21:20:17,864 - Epoch: [106][ 1230/ 1236] Overall Loss 0.229913 Objective Loss 0.229913 LR 0.000500 Time 0.021466 +2023-10-02 21:20:17,986 - Epoch: [106][ 1236/ 1236] Overall Loss 0.229950 Objective Loss 0.229950 Top1 86.150713 Top5 98.574338 LR 0.000500 Time 0.021460 +2023-10-02 21:20:18,130 - --- validate (epoch=106)----------- +2023-10-02 21:20:18,131 - 29943 samples (256 per mini-batch) +2023-10-02 21:20:18,618 - Epoch: [106][ 10/ 117] Loss 0.283142 Top1 85.195312 Top5 98.710938 +2023-10-02 21:20:18,766 - Epoch: [106][ 20/ 117] Loss 0.309884 Top1 83.730469 Top5 98.222656 +2023-10-02 21:20:18,912 - Epoch: [106][ 30/ 117] Loss 0.312062 Top1 83.632812 Top5 98.229167 +2023-10-02 21:20:19,058 - Epoch: [106][ 40/ 117] Loss 0.305399 Top1 83.681641 Top5 98.232422 +2023-10-02 21:20:19,204 - Epoch: [106][ 50/ 117] Loss 0.303287 Top1 83.640625 Top5 98.359375 +2023-10-02 21:20:19,350 - Epoch: [106][ 60/ 117] Loss 0.309647 Top1 83.463542 Top5 98.365885 +2023-10-02 21:20:19,496 - Epoch: [106][ 70/ 117] Loss 0.307643 Top1 83.404018 Top5 98.325893 +2023-10-02 21:20:19,643 - Epoch: [106][ 80/ 117] Loss 0.304672 Top1 83.486328 Top5 98.330078 +2023-10-02 21:20:19,789 - Epoch: [106][ 90/ 117] Loss 0.305868 Top1 83.554688 Top5 98.320312 +2023-10-02 21:20:19,935 - Epoch: [106][ 100/ 117] Loss 0.304252 Top1 83.519531 Top5 98.335938 +2023-10-02 21:20:20,089 - Epoch: [106][ 110/ 117] Loss 0.306079 Top1 83.441051 Top5 98.323864 +2023-10-02 21:20:20,178 - Epoch: [106][ 117/ 117] Loss 0.304215 Top1 83.498647 Top5 98.366897 +2023-10-02 21:20:20,320 - ==> Top1: 83.499 Top5: 98.367 Loss: 0.304 + +2023-10-02 21:20:20,321 - ==> Confusion: +[[ 952 0 2 0 11 4 0 1 2 48 1 0 0 3 6 1 3 1 0 0 15] + [ 1 1047 1 2 11 25 1 20 0 0 1 1 3 0 0 3 2 0 8 0 5] + [ 3 0 982 6 2 0 16 5 0 0 2 1 8 1 0 7 2 1 9 4 7] + [ 4 2 13 952 3 1 1 0 7 1 7 0 11 5 36 4 1 4 17 2 18] + [ 24 7 1 0 972 4 0 1 0 11 1 0 2 2 8 5 7 1 1 1 2] + [ 3 29 0 0 3 998 0 26 1 3 1 7 1 16 3 2 7 0 5 3 8] + [ 0 1 26 2 0 1 1126 7 0 0 7 1 0 0 0 7 0 0 0 8 5] + [ 4 13 13 2 4 23 5 1076 2 6 3 3 3 2 1 1 0 4 40 7 6] + [ 18 1 0 1 0 2 0 1 965 43 14 3 3 16 15 1 0 0 3 2 1] + [ 120 1 0 0 8 0 1 0 20 926 0 1 1 26 5 2 0 1 0 2 5] + [ 4 3 11 3 1 1 3 5 9 1 974 2 0 14 5 1 2 2 4 1 7] + [ 1 0 1 0 0 10 0 1 0 0 0 964 26 5 0 3 0 15 0 5 4] + [ 0 1 2 3 0 1 1 0 1 1 2 31 986 2 3 6 3 13 2 5 5] + [ 0 0 3 0 4 5 2 0 6 10 4 4 0 1064 5 0 0 2 0 2 8] + [ 14 0 7 15 4 0 0 0 13 1 2 0 3 2 1021 0 4 1 6 0 8] + [ 1 1 0 0 3 0 0 0 0 0 2 7 10 0 0 1079 17 8 1 0 5] + [ 2 13 0 0 8 7 0 0 0 0 0 3 0 1 5 8 1097 0 1 5 11] + [ 0 0 1 2 0 0 2 0 0 3 0 4 25 0 7 8 1 981 0 1 3] + [ 2 8 5 12 1 0 1 24 3 1 4 1 0 0 17 0 0 0 980 0 9] + [ 0 0 4 3 0 2 7 8 0 1 0 20 2 1 1 5 7 1 0 1081 9] + [ 170 190 171 61 114 171 41 92 72 88 228 114 479 376 154 77 142 68 139 179 4779]] + +2023-10-02 21:20:20,322 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:20:20,322 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:20:20,328 - + +2023-10-02 21:20:20,328 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:20:21,456 - Epoch: [107][ 10/ 1236] Overall Loss 0.231552 Objective Loss 0.231552 LR 0.000500 Time 0.112709 +2023-10-02 21:20:21,665 - Epoch: [107][ 20/ 1236] Overall Loss 0.234323 Objective Loss 0.234323 LR 0.000500 Time 0.066797 +2023-10-02 21:20:21,873 - Epoch: [107][ 30/ 1236] Overall Loss 0.237271 Objective Loss 0.237271 LR 0.000500 Time 0.051416 +2023-10-02 21:20:22,083 - Epoch: [107][ 40/ 1236] Overall Loss 0.236546 Objective Loss 0.236546 LR 0.000500 Time 0.043814 +2023-10-02 21:20:22,290 - Epoch: [107][ 50/ 1236] Overall Loss 0.231729 Objective Loss 0.231729 LR 0.000500 Time 0.039181 +2023-10-02 21:20:22,499 - Epoch: [107][ 60/ 1236] Overall Loss 0.235184 Objective Loss 0.235184 LR 0.000500 Time 0.036131 +2023-10-02 21:20:22,707 - Epoch: [107][ 70/ 1236] Overall Loss 0.237389 Objective Loss 0.237389 LR 0.000500 Time 0.033919 +2023-10-02 21:20:22,918 - Epoch: [107][ 80/ 1236] Overall Loss 0.236501 Objective Loss 0.236501 LR 0.000500 Time 0.032309 +2023-10-02 21:20:23,125 - Epoch: [107][ 90/ 1236] Overall Loss 0.236397 Objective Loss 0.236397 LR 0.000500 Time 0.031012 +2023-10-02 21:20:23,335 - Epoch: [107][ 100/ 1236] Overall Loss 0.239129 Objective Loss 0.239129 LR 0.000500 Time 0.030012 +2023-10-02 21:20:23,542 - Epoch: [107][ 110/ 1236] Overall Loss 0.238758 Objective Loss 0.238758 LR 0.000500 Time 0.029162 +2023-10-02 21:20:23,752 - Epoch: [107][ 120/ 1236] Overall Loss 0.238045 Objective Loss 0.238045 LR 0.000500 Time 0.028482 +2023-10-02 21:20:23,959 - Epoch: [107][ 130/ 1236] Overall Loss 0.239918 Objective Loss 0.239918 LR 0.000500 Time 0.027881 +2023-10-02 21:20:24,167 - Epoch: [107][ 140/ 1236] Overall Loss 0.239389 Objective Loss 0.239389 LR 0.000500 Time 0.027372 +2023-10-02 21:20:24,374 - Epoch: [107][ 150/ 1236] Overall Loss 0.239131 Objective Loss 0.239131 LR 0.000500 Time 0.026918 +2023-10-02 21:20:24,586 - Epoch: [107][ 160/ 1236] Overall Loss 0.240310 Objective Loss 0.240310 LR 0.000500 Time 0.026562 +2023-10-02 21:20:24,796 - Epoch: [107][ 170/ 1236] Overall Loss 0.239276 Objective Loss 0.239276 LR 0.000500 Time 0.026229 +2023-10-02 21:20:25,009 - Epoch: [107][ 180/ 1236] Overall Loss 0.240392 Objective Loss 0.240392 LR 0.000500 Time 0.025954 +2023-10-02 21:20:25,218 - Epoch: [107][ 190/ 1236] Overall Loss 0.240319 Objective Loss 0.240319 LR 0.000500 Time 0.025689 +2023-10-02 21:20:25,432 - Epoch: [107][ 200/ 1236] Overall Loss 0.239957 Objective Loss 0.239957 LR 0.000500 Time 0.025471 +2023-10-02 21:20:25,641 - Epoch: [107][ 210/ 1236] Overall Loss 0.239117 Objective Loss 0.239117 LR 0.000500 Time 0.025254 +2023-10-02 21:20:25,855 - Epoch: [107][ 220/ 1236] Overall Loss 0.238719 Objective Loss 0.238719 LR 0.000500 Time 0.025074 +2023-10-02 21:20:26,064 - Epoch: [107][ 230/ 1236] Overall Loss 0.237904 Objective Loss 0.237904 LR 0.000500 Time 0.024893 +2023-10-02 21:20:26,277 - Epoch: [107][ 240/ 1236] Overall Loss 0.237877 Objective Loss 0.237877 LR 0.000500 Time 0.024744 +2023-10-02 21:20:26,487 - Epoch: [107][ 250/ 1236] Overall Loss 0.238680 Objective Loss 0.238680 LR 0.000500 Time 0.024590 +2023-10-02 21:20:26,700 - Epoch: [107][ 260/ 1236] Overall Loss 0.238409 Objective Loss 0.238409 LR 0.000500 Time 0.024463 +2023-10-02 21:20:26,909 - Epoch: [107][ 270/ 1236] Overall Loss 0.237914 Objective Loss 0.237914 LR 0.000500 Time 0.024331 +2023-10-02 21:20:27,123 - Epoch: [107][ 280/ 1236] Overall Loss 0.237121 Objective Loss 0.237121 LR 0.000500 Time 0.024223 +2023-10-02 21:20:27,332 - Epoch: [107][ 290/ 1236] Overall Loss 0.236106 Objective Loss 0.236106 LR 0.000500 Time 0.024110 +2023-10-02 21:20:27,545 - Epoch: [107][ 300/ 1236] Overall Loss 0.235913 Objective Loss 0.235913 LR 0.000500 Time 0.024016 +2023-10-02 21:20:27,754 - Epoch: [107][ 310/ 1236] Overall Loss 0.235025 Objective Loss 0.235025 LR 0.000500 Time 0.023915 +2023-10-02 21:20:27,966 - Epoch: [107][ 320/ 1236] Overall Loss 0.234334 Objective Loss 0.234334 LR 0.000500 Time 0.023829 +2023-10-02 21:20:28,178 - Epoch: [107][ 330/ 1236] Overall Loss 0.234259 Objective Loss 0.234259 LR 0.000500 Time 0.023742 +2023-10-02 21:20:28,389 - Epoch: [107][ 340/ 1236] Overall Loss 0.234024 Objective Loss 0.234024 LR 0.000500 Time 0.023666 +2023-10-02 21:20:28,597 - Epoch: [107][ 350/ 1236] Overall Loss 0.234751 Objective Loss 0.234751 LR 0.000500 Time 0.023579 +2023-10-02 21:20:28,808 - Epoch: [107][ 360/ 1236] Overall Loss 0.233948 Objective Loss 0.233948 LR 0.000500 Time 0.023509 +2023-10-02 21:20:29,015 - Epoch: [107][ 370/ 1236] Overall Loss 0.233805 Objective Loss 0.233805 LR 0.000500 Time 0.023432 +2023-10-02 21:20:29,226 - Epoch: [107][ 380/ 1236] Overall Loss 0.233262 Objective Loss 0.233262 LR 0.000500 Time 0.023370 +2023-10-02 21:20:29,433 - Epoch: [107][ 390/ 1236] Overall Loss 0.233893 Objective Loss 0.233893 LR 0.000500 Time 0.023301 +2023-10-02 21:20:29,644 - Epoch: [107][ 400/ 1236] Overall Loss 0.233778 Objective Loss 0.233778 LR 0.000500 Time 0.023246 +2023-10-02 21:20:29,851 - Epoch: [107][ 410/ 1236] Overall Loss 0.233418 Objective Loss 0.233418 LR 0.000500 Time 0.023183 +2023-10-02 21:20:30,062 - Epoch: [107][ 420/ 1236] Overall Loss 0.232717 Objective Loss 0.232717 LR 0.000500 Time 0.023133 +2023-10-02 21:20:30,269 - Epoch: [107][ 430/ 1236] Overall Loss 0.232718 Objective Loss 0.232718 LR 0.000500 Time 0.023075 +2023-10-02 21:20:30,480 - Epoch: [107][ 440/ 1236] Overall Loss 0.232878 Objective Loss 0.232878 LR 0.000500 Time 0.023030 +2023-10-02 21:20:30,687 - Epoch: [107][ 450/ 1236] Overall Loss 0.233069 Objective Loss 0.233069 LR 0.000500 Time 0.022977 +2023-10-02 21:20:30,898 - Epoch: [107][ 460/ 1236] Overall Loss 0.233497 Objective Loss 0.233497 LR 0.000500 Time 0.022935 +2023-10-02 21:20:31,104 - Epoch: [107][ 470/ 1236] Overall Loss 0.233955 Objective Loss 0.233955 LR 0.000500 Time 0.022887 +2023-10-02 21:20:31,316 - Epoch: [107][ 480/ 1236] Overall Loss 0.234188 Objective Loss 0.234188 LR 0.000500 Time 0.022849 +2023-10-02 21:20:31,522 - Epoch: [107][ 490/ 1236] Overall Loss 0.234078 Objective Loss 0.234078 LR 0.000500 Time 0.022804 +2023-10-02 21:20:31,733 - Epoch: [107][ 500/ 1236] Overall Loss 0.234185 Objective Loss 0.234185 LR 0.000500 Time 0.022769 +2023-10-02 21:20:31,940 - Epoch: [107][ 510/ 1236] Overall Loss 0.234188 Objective Loss 0.234188 LR 0.000500 Time 0.022728 +2023-10-02 21:20:32,151 - Epoch: [107][ 520/ 1236] Overall Loss 0.234375 Objective Loss 0.234375 LR 0.000500 Time 0.022696 +2023-10-02 21:20:32,358 - Epoch: [107][ 530/ 1236] Overall Loss 0.234259 Objective Loss 0.234259 LR 0.000500 Time 0.022658 +2023-10-02 21:20:32,569 - Epoch: [107][ 540/ 1236] Overall Loss 0.233877 Objective Loss 0.233877 LR 0.000500 Time 0.022629 +2023-10-02 21:20:32,776 - Epoch: [107][ 550/ 1236] Overall Loss 0.233836 Objective Loss 0.233836 LR 0.000500 Time 0.022594 +2023-10-02 21:20:32,988 - Epoch: [107][ 560/ 1236] Overall Loss 0.233762 Objective Loss 0.233762 LR 0.000500 Time 0.022567 +2023-10-02 21:20:33,195 - Epoch: [107][ 570/ 1236] Overall Loss 0.233463 Objective Loss 0.233463 LR 0.000500 Time 0.022534 +2023-10-02 21:20:33,406 - Epoch: [107][ 580/ 1236] Overall Loss 0.233042 Objective Loss 0.233042 LR 0.000500 Time 0.022509 +2023-10-02 21:20:33,613 - Epoch: [107][ 590/ 1236] Overall Loss 0.232844 Objective Loss 0.232844 LR 0.000500 Time 0.022478 +2023-10-02 21:20:33,822 - Epoch: [107][ 600/ 1236] Overall Loss 0.232876 Objective Loss 0.232876 LR 0.000500 Time 0.022452 +2023-10-02 21:20:34,031 - Epoch: [107][ 610/ 1236] Overall Loss 0.233018 Objective Loss 0.233018 LR 0.000500 Time 0.022423 +2023-10-02 21:20:34,242 - Epoch: [107][ 620/ 1236] Overall Loss 0.233158 Objective Loss 0.233158 LR 0.000500 Time 0.022401 +2023-10-02 21:20:34,449 - Epoch: [107][ 630/ 1236] Overall Loss 0.233196 Objective Loss 0.233196 LR 0.000500 Time 0.022374 +2023-10-02 21:20:34,660 - Epoch: [107][ 640/ 1236] Overall Loss 0.233295 Objective Loss 0.233295 LR 0.000500 Time 0.022354 +2023-10-02 21:20:34,867 - Epoch: [107][ 650/ 1236] Overall Loss 0.233075 Objective Loss 0.233075 LR 0.000500 Time 0.022328 +2023-10-02 21:20:35,079 - Epoch: [107][ 660/ 1236] Overall Loss 0.233129 Objective Loss 0.233129 LR 0.000500 Time 0.022310 +2023-10-02 21:20:35,286 - Epoch: [107][ 670/ 1236] Overall Loss 0.233311 Objective Loss 0.233311 LR 0.000500 Time 0.022285 +2023-10-02 21:20:35,497 - Epoch: [107][ 680/ 1236] Overall Loss 0.233073 Objective Loss 0.233073 LR 0.000500 Time 0.022267 +2023-10-02 21:20:35,704 - Epoch: [107][ 690/ 1236] Overall Loss 0.233106 Objective Loss 0.233106 LR 0.000500 Time 0.022244 +2023-10-02 21:20:35,915 - Epoch: [107][ 700/ 1236] Overall Loss 0.232950 Objective Loss 0.232950 LR 0.000500 Time 0.022228 +2023-10-02 21:20:36,122 - Epoch: [107][ 710/ 1236] Overall Loss 0.232723 Objective Loss 0.232723 LR 0.000500 Time 0.022206 +2023-10-02 21:20:36,333 - Epoch: [107][ 720/ 1236] Overall Loss 0.232860 Objective Loss 0.232860 LR 0.000500 Time 0.022190 +2023-10-02 21:20:36,540 - Epoch: [107][ 730/ 1236] Overall Loss 0.232683 Objective Loss 0.232683 LR 0.000500 Time 0.022170 +2023-10-02 21:20:36,751 - Epoch: [107][ 740/ 1236] Overall Loss 0.232732 Objective Loss 0.232732 LR 0.000500 Time 0.022155 +2023-10-02 21:20:36,958 - Epoch: [107][ 750/ 1236] Overall Loss 0.232682 Objective Loss 0.232682 LR 0.000500 Time 0.022135 +2023-10-02 21:20:37,169 - Epoch: [107][ 760/ 1236] Overall Loss 0.232783 Objective Loss 0.232783 LR 0.000500 Time 0.022121 +2023-10-02 21:20:37,376 - Epoch: [107][ 770/ 1236] Overall Loss 0.232585 Objective Loss 0.232585 LR 0.000500 Time 0.022103 +2023-10-02 21:20:37,586 - Epoch: [107][ 780/ 1236] Overall Loss 0.232551 Objective Loss 0.232551 LR 0.000500 Time 0.022088 +2023-10-02 21:20:37,795 - Epoch: [107][ 790/ 1236] Overall Loss 0.232308 Objective Loss 0.232308 LR 0.000500 Time 0.022071 +2023-10-02 21:20:38,006 - Epoch: [107][ 800/ 1236] Overall Loss 0.232142 Objective Loss 0.232142 LR 0.000500 Time 0.022058 +2023-10-02 21:20:38,213 - Epoch: [107][ 810/ 1236] Overall Loss 0.231852 Objective Loss 0.231852 LR 0.000500 Time 0.022041 +2023-10-02 21:20:38,424 - Epoch: [107][ 820/ 1236] Overall Loss 0.231601 Objective Loss 0.231601 LR 0.000500 Time 0.022030 +2023-10-02 21:20:38,632 - Epoch: [107][ 830/ 1236] Overall Loss 0.231767 Objective Loss 0.231767 LR 0.000500 Time 0.022014 +2023-10-02 21:20:38,841 - Epoch: [107][ 840/ 1236] Overall Loss 0.231625 Objective Loss 0.231625 LR 0.000500 Time 0.022001 +2023-10-02 21:20:39,050 - Epoch: [107][ 850/ 1236] Overall Loss 0.231566 Objective Loss 0.231566 LR 0.000500 Time 0.021986 +2023-10-02 21:20:39,260 - Epoch: [107][ 860/ 1236] Overall Loss 0.231856 Objective Loss 0.231856 LR 0.000500 Time 0.021974 +2023-10-02 21:20:39,468 - Epoch: [107][ 870/ 1236] Overall Loss 0.231616 Objective Loss 0.231616 LR 0.000500 Time 0.021959 +2023-10-02 21:20:39,679 - Epoch: [107][ 880/ 1236] Overall Loss 0.231562 Objective Loss 0.231562 LR 0.000500 Time 0.021949 +2023-10-02 21:20:39,887 - Epoch: [107][ 890/ 1236] Overall Loss 0.231505 Objective Loss 0.231505 LR 0.000500 Time 0.021935 +2023-10-02 21:20:40,098 - Epoch: [107][ 900/ 1236] Overall Loss 0.231513 Objective Loss 0.231513 LR 0.000500 Time 0.021926 +2023-10-02 21:20:40,305 - Epoch: [107][ 910/ 1236] Overall Loss 0.231544 Objective Loss 0.231544 LR 0.000500 Time 0.021912 +2023-10-02 21:20:40,515 - Epoch: [107][ 920/ 1236] Overall Loss 0.231629 Objective Loss 0.231629 LR 0.000500 Time 0.021902 +2023-10-02 21:20:40,723 - Epoch: [107][ 930/ 1236] Overall Loss 0.231460 Objective Loss 0.231460 LR 0.000500 Time 0.021889 +2023-10-02 21:20:40,934 - Epoch: [107][ 940/ 1236] Overall Loss 0.231151 Objective Loss 0.231151 LR 0.000500 Time 0.021880 +2023-10-02 21:20:41,141 - Epoch: [107][ 950/ 1236] Overall Loss 0.231303 Objective Loss 0.231303 LR 0.000500 Time 0.021868 +2023-10-02 21:20:41,353 - Epoch: [107][ 960/ 1236] Overall Loss 0.231427 Objective Loss 0.231427 LR 0.000500 Time 0.021860 +2023-10-02 21:20:41,560 - Epoch: [107][ 970/ 1236] Overall Loss 0.231400 Objective Loss 0.231400 LR 0.000500 Time 0.021848 +2023-10-02 21:20:41,771 - Epoch: [107][ 980/ 1236] Overall Loss 0.231428 Objective Loss 0.231428 LR 0.000500 Time 0.021840 +2023-10-02 21:20:41,978 - Epoch: [107][ 990/ 1236] Overall Loss 0.231373 Objective Loss 0.231373 LR 0.000500 Time 0.021828 +2023-10-02 21:20:42,189 - Epoch: [107][ 1000/ 1236] Overall Loss 0.231084 Objective Loss 0.231084 LR 0.000500 Time 0.021820 +2023-10-02 21:20:42,396 - Epoch: [107][ 1010/ 1236] Overall Loss 0.231047 Objective Loss 0.231047 LR 0.000500 Time 0.021809 +2023-10-02 21:20:42,607 - Epoch: [107][ 1020/ 1236] Overall Loss 0.230970 Objective Loss 0.230970 LR 0.000500 Time 0.021802 +2023-10-02 21:20:42,814 - Epoch: [107][ 1030/ 1236] Overall Loss 0.231122 Objective Loss 0.231122 LR 0.000500 Time 0.021791 +2023-10-02 21:20:43,025 - Epoch: [107][ 1040/ 1236] Overall Loss 0.231118 Objective Loss 0.231118 LR 0.000500 Time 0.021784 +2023-10-02 21:20:43,232 - Epoch: [107][ 1050/ 1236] Overall Loss 0.231409 Objective Loss 0.231409 LR 0.000500 Time 0.021774 +2023-10-02 21:20:43,442 - Epoch: [107][ 1060/ 1236] Overall Loss 0.231549 Objective Loss 0.231549 LR 0.000500 Time 0.021766 +2023-10-02 21:20:43,651 - Epoch: [107][ 1070/ 1236] Overall Loss 0.231492 Objective Loss 0.231492 LR 0.000500 Time 0.021756 +2023-10-02 21:20:43,861 - Epoch: [107][ 1080/ 1236] Overall Loss 0.231465 Objective Loss 0.231465 LR 0.000500 Time 0.021749 +2023-10-02 21:20:44,069 - Epoch: [107][ 1090/ 1236] Overall Loss 0.231241 Objective Loss 0.231241 LR 0.000500 Time 0.021739 +2023-10-02 21:20:44,279 - Epoch: [107][ 1100/ 1236] Overall Loss 0.231087 Objective Loss 0.231087 LR 0.000500 Time 0.021732 +2023-10-02 21:20:44,487 - Epoch: [107][ 1110/ 1236] Overall Loss 0.230983 Objective Loss 0.230983 LR 0.000500 Time 0.021722 +2023-10-02 21:20:44,698 - Epoch: [107][ 1120/ 1236] Overall Loss 0.231123 Objective Loss 0.231123 LR 0.000500 Time 0.021717 +2023-10-02 21:20:44,905 - Epoch: [107][ 1130/ 1236] Overall Loss 0.230982 Objective Loss 0.230982 LR 0.000500 Time 0.021707 +2023-10-02 21:20:45,115 - Epoch: [107][ 1140/ 1236] Overall Loss 0.230983 Objective Loss 0.230983 LR 0.000500 Time 0.021701 +2023-10-02 21:20:45,323 - Epoch: [107][ 1150/ 1236] Overall Loss 0.230806 Objective Loss 0.230806 LR 0.000500 Time 0.021692 +2023-10-02 21:20:45,533 - Epoch: [107][ 1160/ 1236] Overall Loss 0.230867 Objective Loss 0.230867 LR 0.000500 Time 0.021686 +2023-10-02 21:20:45,741 - Epoch: [107][ 1170/ 1236] Overall Loss 0.230681 Objective Loss 0.230681 LR 0.000500 Time 0.021677 +2023-10-02 21:20:45,953 - Epoch: [107][ 1180/ 1236] Overall Loss 0.230786 Objective Loss 0.230786 LR 0.000500 Time 0.021672 +2023-10-02 21:20:46,160 - Epoch: [107][ 1190/ 1236] Overall Loss 0.230987 Objective Loss 0.230987 LR 0.000500 Time 0.021664 +2023-10-02 21:20:46,371 - Epoch: [107][ 1200/ 1236] Overall Loss 0.231054 Objective Loss 0.231054 LR 0.000500 Time 0.021659 +2023-10-02 21:20:46,578 - Epoch: [107][ 1210/ 1236] Overall Loss 0.231335 Objective Loss 0.231335 LR 0.000500 Time 0.021651 +2023-10-02 21:20:46,789 - Epoch: [107][ 1220/ 1236] Overall Loss 0.231316 Objective Loss 0.231316 LR 0.000500 Time 0.021646 +2023-10-02 21:20:47,048 - Epoch: [107][ 1230/ 1236] Overall Loss 0.231318 Objective Loss 0.231318 LR 0.000500 Time 0.021680 +2023-10-02 21:20:47,169 - Epoch: [107][ 1236/ 1236] Overall Loss 0.231234 Objective Loss 0.231234 Top1 85.947047 Top5 97.759674 LR 0.000500 Time 0.021673 +2023-10-02 21:20:47,302 - --- validate (epoch=107)----------- +2023-10-02 21:20:47,302 - 29943 samples (256 per mini-batch) +2023-10-02 21:20:47,796 - Epoch: [107][ 10/ 117] Loss 0.296223 Top1 84.492188 Top5 98.320312 +2023-10-02 21:20:47,948 - Epoch: [107][ 20/ 117] Loss 0.282750 Top1 84.414062 Top5 98.261719 +2023-10-02 21:20:48,099 - Epoch: [107][ 30/ 117] Loss 0.277361 Top1 84.856771 Top5 98.385417 +2023-10-02 21:20:48,250 - Epoch: [107][ 40/ 117] Loss 0.288019 Top1 84.755859 Top5 98.320312 +2023-10-02 21:20:48,400 - Epoch: [107][ 50/ 117] Loss 0.291412 Top1 84.664062 Top5 98.343750 +2023-10-02 21:20:48,550 - Epoch: [107][ 60/ 117] Loss 0.297450 Top1 84.492188 Top5 98.333333 +2023-10-02 21:20:48,698 - Epoch: [107][ 70/ 117] Loss 0.297016 Top1 84.352679 Top5 98.353795 +2023-10-02 21:20:48,845 - Epoch: [107][ 80/ 117] Loss 0.296128 Top1 84.384766 Top5 98.339844 +2023-10-02 21:20:48,993 - Epoch: [107][ 90/ 117] Loss 0.294395 Top1 84.401042 Top5 98.320312 +2023-10-02 21:20:49,140 - Epoch: [107][ 100/ 117] Loss 0.294026 Top1 84.406250 Top5 98.304688 +2023-10-02 21:20:49,295 - Epoch: [107][ 110/ 117] Loss 0.296279 Top1 84.367898 Top5 98.295455 +2023-10-02 21:20:49,384 - Epoch: [107][ 117/ 117] Loss 0.298054 Top1 84.270113 Top5 98.306783 +2023-10-02 21:20:49,516 - ==> Top1: 84.270 Top5: 98.307 Loss: 0.298 + +2023-10-02 21:20:49,517 - ==> Confusion: +[[ 952 1 2 0 9 3 0 0 5 47 3 1 1 3 5 0 7 0 0 0 11] + [ 1 1038 0 1 6 25 0 28 1 0 0 1 1 0 2 4 4 1 13 2 3] + [ 4 0 984 7 3 0 14 7 0 0 1 1 8 2 1 8 1 0 7 2 6] + [ 2 4 11 968 1 1 2 1 7 0 8 1 7 3 29 2 1 5 16 0 20] + [ 31 8 0 1 970 3 0 1 2 10 0 0 2 2 3 7 7 0 1 0 2] + [ 3 28 0 1 2 996 1 23 1 4 1 10 1 12 7 1 6 0 3 5 11] + [ 0 4 35 2 0 0 1107 7 0 1 4 1 0 0 0 8 0 1 0 13 8] + [ 3 12 17 2 1 18 3 1087 0 2 2 9 4 3 2 0 2 3 34 6 8] + [ 18 3 0 0 0 2 0 0 984 35 6 2 2 12 16 3 1 0 4 0 1] + [ 110 0 0 0 6 1 1 0 19 944 0 1 1 21 7 2 1 0 0 1 4] + [ 4 3 9 8 1 2 1 5 20 2 955 2 0 16 6 1 1 4 2 1 10] + [ 0 0 1 0 0 8 0 2 0 0 0 968 16 12 0 3 3 15 0 4 3] + [ 0 2 0 3 1 1 1 1 1 0 0 38 980 1 2 7 3 12 1 6 8] + [ 0 0 1 0 1 7 2 0 14 8 2 5 0 1061 4 1 1 0 0 1 11] + [ 12 1 3 20 4 0 0 0 18 3 1 0 2 3 1010 0 3 2 6 0 13] + [ 0 0 1 0 4 0 0 0 0 0 0 6 7 0 0 1078 17 9 3 3 6] + [ 1 8 1 0 6 8 0 0 0 1 0 3 0 3 3 7 1108 0 1 4 7] + [ 0 0 0 3 0 0 2 0 2 1 0 5 26 1 4 7 0 981 0 2 4] + [ 1 3 3 11 0 1 0 27 4 0 4 3 2 1 17 0 0 0 981 0 10] + [ 0 0 4 0 0 3 2 9 0 0 0 14 5 0 0 4 9 0 1 1093 8] + [ 149 163 142 74 112 154 48 115 96 79 189 114 400 344 141 66 151 62 132 186 4988]] + +2023-10-02 21:20:49,518 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:20:49,518 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:20:49,524 - + +2023-10-02 21:20:49,524 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:20:50,554 - Epoch: [108][ 10/ 1236] Overall Loss 0.228922 Objective Loss 0.228922 LR 0.000500 Time 0.102875 +2023-10-02 21:20:50,763 - Epoch: [108][ 20/ 1236] Overall Loss 0.218270 Objective Loss 0.218270 LR 0.000500 Time 0.061907 +2023-10-02 21:20:50,974 - Epoch: [108][ 30/ 1236] Overall Loss 0.217919 Objective Loss 0.217919 LR 0.000500 Time 0.048297 +2023-10-02 21:20:51,185 - Epoch: [108][ 40/ 1236] Overall Loss 0.216838 Objective Loss 0.216838 LR 0.000500 Time 0.041473 +2023-10-02 21:20:51,395 - Epoch: [108][ 50/ 1236] Overall Loss 0.219636 Objective Loss 0.219636 LR 0.000500 Time 0.037363 +2023-10-02 21:20:51,605 - Epoch: [108][ 60/ 1236] Overall Loss 0.222390 Objective Loss 0.222390 LR 0.000500 Time 0.034617 +2023-10-02 21:20:51,809 - Epoch: [108][ 70/ 1236] Overall Loss 0.222704 Objective Loss 0.222704 LR 0.000500 Time 0.032569 +2023-10-02 21:20:52,017 - Epoch: [108][ 80/ 1236] Overall Loss 0.223570 Objective Loss 0.223570 LR 0.000500 Time 0.031094 +2023-10-02 21:20:52,223 - Epoch: [108][ 90/ 1236] Overall Loss 0.223136 Objective Loss 0.223136 LR 0.000500 Time 0.029914 +2023-10-02 21:20:52,431 - Epoch: [108][ 100/ 1236] Overall Loss 0.222518 Objective Loss 0.222518 LR 0.000500 Time 0.028999 +2023-10-02 21:20:52,637 - Epoch: [108][ 110/ 1236] Overall Loss 0.222216 Objective Loss 0.222216 LR 0.000500 Time 0.028224 +2023-10-02 21:20:52,845 - Epoch: [108][ 120/ 1236] Overall Loss 0.223593 Objective Loss 0.223593 LR 0.000500 Time 0.027604 +2023-10-02 21:20:53,051 - Epoch: [108][ 130/ 1236] Overall Loss 0.224064 Objective Loss 0.224064 LR 0.000500 Time 0.027054 +2023-10-02 21:20:53,258 - Epoch: [108][ 140/ 1236] Overall Loss 0.222840 Objective Loss 0.222840 LR 0.000500 Time 0.026597 +2023-10-02 21:20:53,464 - Epoch: [108][ 150/ 1236] Overall Loss 0.221672 Objective Loss 0.221672 LR 0.000500 Time 0.026189 +2023-10-02 21:20:53,672 - Epoch: [108][ 160/ 1236] Overall Loss 0.222353 Objective Loss 0.222353 LR 0.000500 Time 0.025846 +2023-10-02 21:20:53,877 - Epoch: [108][ 170/ 1236] Overall Loss 0.221598 Objective Loss 0.221598 LR 0.000500 Time 0.025524 +2023-10-02 21:20:54,085 - Epoch: [108][ 180/ 1236] Overall Loss 0.221401 Objective Loss 0.221401 LR 0.000500 Time 0.025261 +2023-10-02 21:20:54,289 - Epoch: [108][ 190/ 1236] Overall Loss 0.221052 Objective Loss 0.221052 LR 0.000500 Time 0.025005 +2023-10-02 21:20:54,498 - Epoch: [108][ 200/ 1236] Overall Loss 0.220918 Objective Loss 0.220918 LR 0.000500 Time 0.024795 +2023-10-02 21:20:54,702 - Epoch: [108][ 210/ 1236] Overall Loss 0.221124 Objective Loss 0.221124 LR 0.000500 Time 0.024585 +2023-10-02 21:20:54,910 - Epoch: [108][ 220/ 1236] Overall Loss 0.222399 Objective Loss 0.222399 LR 0.000500 Time 0.024413 +2023-10-02 21:20:55,114 - Epoch: [108][ 230/ 1236] Overall Loss 0.222464 Objective Loss 0.222464 LR 0.000500 Time 0.024238 +2023-10-02 21:20:55,322 - Epoch: [108][ 240/ 1236] Overall Loss 0.221884 Objective Loss 0.221884 LR 0.000500 Time 0.024090 +2023-10-02 21:20:55,527 - Epoch: [108][ 250/ 1236] Overall Loss 0.222239 Objective Loss 0.222239 LR 0.000500 Time 0.023941 +2023-10-02 21:20:55,735 - Epoch: [108][ 260/ 1236] Overall Loss 0.222360 Objective Loss 0.222360 LR 0.000500 Time 0.023821 +2023-10-02 21:20:55,939 - Epoch: [108][ 270/ 1236] Overall Loss 0.223297 Objective Loss 0.223297 LR 0.000500 Time 0.023693 +2023-10-02 21:20:56,147 - Epoch: [108][ 280/ 1236] Overall Loss 0.224158 Objective Loss 0.224158 LR 0.000500 Time 0.023589 +2023-10-02 21:20:56,352 - Epoch: [108][ 290/ 1236] Overall Loss 0.224132 Objective Loss 0.224132 LR 0.000500 Time 0.023479 +2023-10-02 21:20:56,559 - Epoch: [108][ 300/ 1236] Overall Loss 0.224280 Objective Loss 0.224280 LR 0.000500 Time 0.023386 +2023-10-02 21:20:56,764 - Epoch: [108][ 310/ 1236] Overall Loss 0.223676 Objective Loss 0.223676 LR 0.000500 Time 0.023289 +2023-10-02 21:20:56,972 - Epoch: [108][ 320/ 1236] Overall Loss 0.223773 Objective Loss 0.223773 LR 0.000500 Time 0.023211 +2023-10-02 21:20:57,177 - Epoch: [108][ 330/ 1236] Overall Loss 0.222933 Objective Loss 0.222933 LR 0.000500 Time 0.023126 +2023-10-02 21:20:57,385 - Epoch: [108][ 340/ 1236] Overall Loss 0.222688 Objective Loss 0.222688 LR 0.000500 Time 0.023058 +2023-10-02 21:20:57,589 - Epoch: [108][ 350/ 1236] Overall Loss 0.222721 Objective Loss 0.222721 LR 0.000500 Time 0.022982 +2023-10-02 21:20:57,798 - Epoch: [108][ 360/ 1236] Overall Loss 0.222787 Objective Loss 0.222787 LR 0.000500 Time 0.022922 +2023-10-02 21:20:58,002 - Epoch: [108][ 370/ 1236] Overall Loss 0.222103 Objective Loss 0.222103 LR 0.000500 Time 0.022853 +2023-10-02 21:20:58,210 - Epoch: [108][ 380/ 1236] Overall Loss 0.222134 Objective Loss 0.222134 LR 0.000500 Time 0.022799 +2023-10-02 21:20:58,415 - Epoch: [108][ 390/ 1236] Overall Loss 0.222083 Objective Loss 0.222083 LR 0.000500 Time 0.022738 +2023-10-02 21:20:58,623 - Epoch: [108][ 400/ 1236] Overall Loss 0.222281 Objective Loss 0.222281 LR 0.000500 Time 0.022690 +2023-10-02 21:20:58,828 - Epoch: [108][ 410/ 1236] Overall Loss 0.222133 Objective Loss 0.222133 LR 0.000500 Time 0.022635 +2023-10-02 21:20:59,036 - Epoch: [108][ 420/ 1236] Overall Loss 0.222129 Objective Loss 0.222129 LR 0.000500 Time 0.022591 +2023-10-02 21:20:59,241 - Epoch: [108][ 430/ 1236] Overall Loss 0.222093 Objective Loss 0.222093 LR 0.000500 Time 0.022541 +2023-10-02 21:20:59,448 - Epoch: [108][ 440/ 1236] Overall Loss 0.222176 Objective Loss 0.222176 LR 0.000500 Time 0.022499 +2023-10-02 21:20:59,654 - Epoch: [108][ 450/ 1236] Overall Loss 0.222009 Objective Loss 0.222009 LR 0.000500 Time 0.022453 +2023-10-02 21:20:59,862 - Epoch: [108][ 460/ 1236] Overall Loss 0.222148 Objective Loss 0.222148 LR 0.000500 Time 0.022418 +2023-10-02 21:21:00,067 - Epoch: [108][ 470/ 1236] Overall Loss 0.222040 Objective Loss 0.222040 LR 0.000500 Time 0.022375 +2023-10-02 21:21:00,275 - Epoch: [108][ 480/ 1236] Overall Loss 0.221880 Objective Loss 0.221880 LR 0.000500 Time 0.022342 +2023-10-02 21:21:00,480 - Epoch: [108][ 490/ 1236] Overall Loss 0.221946 Objective Loss 0.221946 LR 0.000500 Time 0.022304 +2023-10-02 21:21:00,687 - Epoch: [108][ 500/ 1236] Overall Loss 0.221880 Objective Loss 0.221880 LR 0.000500 Time 0.022271 +2023-10-02 21:21:00,893 - Epoch: [108][ 510/ 1236] Overall Loss 0.222207 Objective Loss 0.222207 LR 0.000500 Time 0.022235 +2023-10-02 21:21:01,100 - Epoch: [108][ 520/ 1236] Overall Loss 0.223069 Objective Loss 0.223069 LR 0.000500 Time 0.022206 +2023-10-02 21:21:01,306 - Epoch: [108][ 530/ 1236] Overall Loss 0.222784 Objective Loss 0.222784 LR 0.000500 Time 0.022172 +2023-10-02 21:21:01,514 - Epoch: [108][ 540/ 1236] Overall Loss 0.222237 Objective Loss 0.222237 LR 0.000500 Time 0.022146 +2023-10-02 21:21:01,718 - Epoch: [108][ 550/ 1236] Overall Loss 0.222653 Objective Loss 0.222653 LR 0.000500 Time 0.022115 +2023-10-02 21:21:01,925 - Epoch: [108][ 560/ 1236] Overall Loss 0.222697 Objective Loss 0.222697 LR 0.000500 Time 0.022089 +2023-10-02 21:21:02,131 - Epoch: [108][ 570/ 1236] Overall Loss 0.222709 Objective Loss 0.222709 LR 0.000500 Time 0.022059 +2023-10-02 21:21:02,339 - Epoch: [108][ 580/ 1236] Overall Loss 0.223030 Objective Loss 0.223030 LR 0.000500 Time 0.022038 +2023-10-02 21:21:02,544 - Epoch: [108][ 590/ 1236] Overall Loss 0.223210 Objective Loss 0.223210 LR 0.000500 Time 0.022011 +2023-10-02 21:21:02,752 - Epoch: [108][ 600/ 1236] Overall Loss 0.223122 Objective Loss 0.223122 LR 0.000500 Time 0.021991 +2023-10-02 21:21:02,957 - Epoch: [108][ 610/ 1236] Overall Loss 0.223372 Objective Loss 0.223372 LR 0.000500 Time 0.021965 +2023-10-02 21:21:03,165 - Epoch: [108][ 620/ 1236] Overall Loss 0.223117 Objective Loss 0.223117 LR 0.000500 Time 0.021947 +2023-10-02 21:21:03,370 - Epoch: [108][ 630/ 1236] Overall Loss 0.223014 Objective Loss 0.223014 LR 0.000500 Time 0.021923 +2023-10-02 21:21:03,578 - Epoch: [108][ 640/ 1236] Overall Loss 0.222844 Objective Loss 0.222844 LR 0.000500 Time 0.021905 +2023-10-02 21:21:03,783 - Epoch: [108][ 650/ 1236] Overall Loss 0.222932 Objective Loss 0.222932 LR 0.000500 Time 0.021883 +2023-10-02 21:21:03,993 - Epoch: [108][ 660/ 1236] Overall Loss 0.222817 Objective Loss 0.222817 LR 0.000500 Time 0.021863 +2023-10-02 21:21:04,198 - Epoch: [108][ 670/ 1236] Overall Loss 0.222849 Objective Loss 0.222849 LR 0.000500 Time 0.021842 +2023-10-02 21:21:04,407 - Epoch: [108][ 680/ 1236] Overall Loss 0.222938 Objective Loss 0.222938 LR 0.000500 Time 0.021827 +2023-10-02 21:21:04,611 - Epoch: [108][ 690/ 1236] Overall Loss 0.222973 Objective Loss 0.222973 LR 0.000500 Time 0.021806 +2023-10-02 21:21:04,818 - Epoch: [108][ 700/ 1236] Overall Loss 0.223078 Objective Loss 0.223078 LR 0.000500 Time 0.021790 +2023-10-02 21:21:05,024 - Epoch: [108][ 710/ 1236] Overall Loss 0.222682 Objective Loss 0.222682 LR 0.000500 Time 0.021771 +2023-10-02 21:21:05,231 - Epoch: [108][ 720/ 1236] Overall Loss 0.222725 Objective Loss 0.222725 LR 0.000500 Time 0.021755 +2023-10-02 21:21:05,437 - Epoch: [108][ 730/ 1236] Overall Loss 0.222547 Objective Loss 0.222547 LR 0.000500 Time 0.021738 +2023-10-02 21:21:05,644 - Epoch: [108][ 740/ 1236] Overall Loss 0.222668 Objective Loss 0.222668 LR 0.000500 Time 0.021723 +2023-10-02 21:21:05,850 - Epoch: [108][ 750/ 1236] Overall Loss 0.222623 Objective Loss 0.222623 LR 0.000500 Time 0.021706 +2023-10-02 21:21:06,057 - Epoch: [108][ 760/ 1236] Overall Loss 0.222845 Objective Loss 0.222845 LR 0.000500 Time 0.021692 +2023-10-02 21:21:06,263 - Epoch: [108][ 770/ 1236] Overall Loss 0.222777 Objective Loss 0.222777 LR 0.000500 Time 0.021676 +2023-10-02 21:21:06,470 - Epoch: [108][ 780/ 1236] Overall Loss 0.222707 Objective Loss 0.222707 LR 0.000500 Time 0.021663 +2023-10-02 21:21:06,676 - Epoch: [108][ 790/ 1236] Overall Loss 0.222367 Objective Loss 0.222367 LR 0.000500 Time 0.021648 +2023-10-02 21:21:06,883 - Epoch: [108][ 800/ 1236] Overall Loss 0.222432 Objective Loss 0.222432 LR 0.000500 Time 0.021636 +2023-10-02 21:21:07,089 - Epoch: [108][ 810/ 1236] Overall Loss 0.222283 Objective Loss 0.222283 LR 0.000500 Time 0.021621 +2023-10-02 21:21:07,297 - Epoch: [108][ 820/ 1236] Overall Loss 0.222368 Objective Loss 0.222368 LR 0.000500 Time 0.021611 +2023-10-02 21:21:07,502 - Epoch: [108][ 830/ 1236] Overall Loss 0.222453 Objective Loss 0.222453 LR 0.000500 Time 0.021597 +2023-10-02 21:21:07,709 - Epoch: [108][ 840/ 1236] Overall Loss 0.222572 Objective Loss 0.222572 LR 0.000500 Time 0.021586 +2023-10-02 21:21:07,914 - Epoch: [108][ 850/ 1236] Overall Loss 0.222568 Objective Loss 0.222568 LR 0.000500 Time 0.021572 +2023-10-02 21:21:08,123 - Epoch: [108][ 860/ 1236] Overall Loss 0.222336 Objective Loss 0.222336 LR 0.000500 Time 0.021563 +2023-10-02 21:21:08,328 - Epoch: [108][ 870/ 1236] Overall Loss 0.222237 Objective Loss 0.222237 LR 0.000500 Time 0.021550 +2023-10-02 21:21:08,535 - Epoch: [108][ 880/ 1236] Overall Loss 0.222226 Objective Loss 0.222226 LR 0.000500 Time 0.021540 +2023-10-02 21:21:08,741 - Epoch: [108][ 890/ 1236] Overall Loss 0.222064 Objective Loss 0.222064 LR 0.000500 Time 0.021528 +2023-10-02 21:21:08,948 - Epoch: [108][ 900/ 1236] Overall Loss 0.222217 Objective Loss 0.222217 LR 0.000500 Time 0.021519 +2023-10-02 21:21:09,154 - Epoch: [108][ 910/ 1236] Overall Loss 0.222142 Objective Loss 0.222142 LR 0.000500 Time 0.021507 +2023-10-02 21:21:09,362 - Epoch: [108][ 920/ 1236] Overall Loss 0.222221 Objective Loss 0.222221 LR 0.000500 Time 0.021498 +2023-10-02 21:21:09,568 - Epoch: [108][ 930/ 1236] Overall Loss 0.222256 Objective Loss 0.222256 LR 0.000500 Time 0.021487 +2023-10-02 21:21:09,775 - Epoch: [108][ 940/ 1236] Overall Loss 0.222211 Objective Loss 0.222211 LR 0.000500 Time 0.021478 +2023-10-02 21:21:09,981 - Epoch: [108][ 950/ 1236] Overall Loss 0.221939 Objective Loss 0.221939 LR 0.000500 Time 0.021468 +2023-10-02 21:21:10,188 - Epoch: [108][ 960/ 1236] Overall Loss 0.221756 Objective Loss 0.221756 LR 0.000500 Time 0.021460 +2023-10-02 21:21:10,394 - Epoch: [108][ 970/ 1236] Overall Loss 0.221592 Objective Loss 0.221592 LR 0.000500 Time 0.021449 +2023-10-02 21:21:10,601 - Epoch: [108][ 980/ 1236] Overall Loss 0.221569 Objective Loss 0.221569 LR 0.000500 Time 0.021441 +2023-10-02 21:21:10,807 - Epoch: [108][ 990/ 1236] Overall Loss 0.221565 Objective Loss 0.221565 LR 0.000500 Time 0.021431 +2023-10-02 21:21:11,014 - Epoch: [108][ 1000/ 1236] Overall Loss 0.221587 Objective Loss 0.221587 LR 0.000500 Time 0.021423 +2023-10-02 21:21:11,220 - Epoch: [108][ 1010/ 1236] Overall Loss 0.221367 Objective Loss 0.221367 LR 0.000500 Time 0.021414 +2023-10-02 21:21:11,427 - Epoch: [108][ 1020/ 1236] Overall Loss 0.221316 Objective Loss 0.221316 LR 0.000500 Time 0.021407 +2023-10-02 21:21:11,633 - Epoch: [108][ 1030/ 1236] Overall Loss 0.221233 Objective Loss 0.221233 LR 0.000500 Time 0.021397 +2023-10-02 21:21:11,840 - Epoch: [108][ 1040/ 1236] Overall Loss 0.221495 Objective Loss 0.221495 LR 0.000500 Time 0.021390 +2023-10-02 21:21:12,046 - Epoch: [108][ 1050/ 1236] Overall Loss 0.221426 Objective Loss 0.221426 LR 0.000500 Time 0.021381 +2023-10-02 21:21:12,253 - Epoch: [108][ 1060/ 1236] Overall Loss 0.221511 Objective Loss 0.221511 LR 0.000500 Time 0.021375 +2023-10-02 21:21:12,458 - Epoch: [108][ 1070/ 1236] Overall Loss 0.221528 Objective Loss 0.221528 LR 0.000500 Time 0.021365 +2023-10-02 21:21:12,665 - Epoch: [108][ 1080/ 1236] Overall Loss 0.221532 Objective Loss 0.221532 LR 0.000500 Time 0.021358 +2023-10-02 21:21:12,871 - Epoch: [108][ 1090/ 1236] Overall Loss 0.221475 Objective Loss 0.221475 LR 0.000500 Time 0.021350 +2023-10-02 21:21:13,078 - Epoch: [108][ 1100/ 1236] Overall Loss 0.221589 Objective Loss 0.221589 LR 0.000500 Time 0.021344 +2023-10-02 21:21:13,284 - Epoch: [108][ 1110/ 1236] Overall Loss 0.221420 Objective Loss 0.221420 LR 0.000500 Time 0.021337 +2023-10-02 21:21:13,492 - Epoch: [108][ 1120/ 1236] Overall Loss 0.221763 Objective Loss 0.221763 LR 0.000500 Time 0.021331 +2023-10-02 21:21:13,697 - Epoch: [108][ 1130/ 1236] Overall Loss 0.221746 Objective Loss 0.221746 LR 0.000500 Time 0.021324 +2023-10-02 21:21:13,905 - Epoch: [108][ 1140/ 1236] Overall Loss 0.221853 Objective Loss 0.221853 LR 0.000500 Time 0.021319 +2023-10-02 21:21:14,111 - Epoch: [108][ 1150/ 1236] Overall Loss 0.221886 Objective Loss 0.221886 LR 0.000500 Time 0.021311 +2023-10-02 21:21:14,318 - Epoch: [108][ 1160/ 1236] Overall Loss 0.221823 Objective Loss 0.221823 LR 0.000500 Time 0.021306 +2023-10-02 21:21:14,524 - Epoch: [108][ 1170/ 1236] Overall Loss 0.221721 Objective Loss 0.221721 LR 0.000500 Time 0.021298 +2023-10-02 21:21:14,731 - Epoch: [108][ 1180/ 1236] Overall Loss 0.221792 Objective Loss 0.221792 LR 0.000500 Time 0.021293 +2023-10-02 21:21:14,937 - Epoch: [108][ 1190/ 1236] Overall Loss 0.221762 Objective Loss 0.221762 LR 0.000500 Time 0.021286 +2023-10-02 21:21:15,145 - Epoch: [108][ 1200/ 1236] Overall Loss 0.221732 Objective Loss 0.221732 LR 0.000500 Time 0.021281 +2023-10-02 21:21:15,352 - Epoch: [108][ 1210/ 1236] Overall Loss 0.221646 Objective Loss 0.221646 LR 0.000500 Time 0.021276 +2023-10-02 21:21:15,563 - Epoch: [108][ 1220/ 1236] Overall Loss 0.221736 Objective Loss 0.221736 LR 0.000500 Time 0.021274 +2023-10-02 21:21:15,826 - Epoch: [108][ 1230/ 1236] Overall Loss 0.221621 Objective Loss 0.221621 LR 0.000500 Time 0.021314 +2023-10-02 21:21:15,948 - Epoch: [108][ 1236/ 1236] Overall Loss 0.221684 Objective Loss 0.221684 Top1 84.928717 Top5 98.370672 LR 0.000500 Time 0.021309 +2023-10-02 21:21:16,075 - --- validate (epoch=108)----------- +2023-10-02 21:21:16,075 - 29943 samples (256 per mini-batch) +2023-10-02 21:21:16,578 - Epoch: [108][ 10/ 117] Loss 0.288589 Top1 85.156250 Top5 98.359375 +2023-10-02 21:21:16,739 - Epoch: [108][ 20/ 117] Loss 0.294590 Top1 85.292969 Top5 98.437500 +2023-10-02 21:21:16,894 - Epoch: [108][ 30/ 117] Loss 0.304561 Top1 84.960938 Top5 98.398438 +2023-10-02 21:21:17,052 - Epoch: [108][ 40/ 117] Loss 0.297715 Top1 84.941406 Top5 98.320312 +2023-10-02 21:21:17,210 - Epoch: [108][ 50/ 117] Loss 0.290133 Top1 84.929688 Top5 98.382812 +2023-10-02 21:21:17,373 - Epoch: [108][ 60/ 117] Loss 0.294079 Top1 85.000000 Top5 98.346354 +2023-10-02 21:21:17,529 - Epoch: [108][ 70/ 117] Loss 0.289366 Top1 85.217634 Top5 98.359375 +2023-10-02 21:21:17,689 - Epoch: [108][ 80/ 117] Loss 0.294629 Top1 84.926758 Top5 98.334961 +2023-10-02 21:21:17,846 - Epoch: [108][ 90/ 117] Loss 0.291896 Top1 84.965278 Top5 98.333333 +2023-10-02 21:21:18,005 - Epoch: [108][ 100/ 117] Loss 0.293823 Top1 84.953125 Top5 98.343750 +2023-10-02 21:21:18,169 - Epoch: [108][ 110/ 117] Loss 0.295890 Top1 84.854403 Top5 98.330966 +2023-10-02 21:21:18,257 - Epoch: [108][ 117/ 117] Loss 0.295998 Top1 84.867916 Top5 98.326821 +2023-10-02 21:21:18,365 - ==> Top1: 84.868 Top5: 98.327 Loss: 0.296 + +2023-10-02 21:21:18,366 - ==> Confusion: +[[ 954 1 7 1 8 4 0 0 6 39 1 0 0 4 4 1 4 0 0 0 16] + [ 0 1061 0 0 7 22 0 19 0 1 1 1 1 1 0 4 2 0 8 1 2] + [ 4 0 981 7 1 1 15 5 0 0 1 2 8 1 3 5 2 1 7 5 7] + [ 1 3 12 977 1 3 0 1 2 0 8 1 6 1 28 1 0 6 18 1 19] + [ 23 11 1 1 965 4 0 0 0 9 2 0 2 2 9 6 8 1 1 0 5] + [ 4 33 2 0 3 1004 0 22 0 5 3 2 4 7 7 0 5 0 3 2 10] + [ 0 3 32 0 0 1 1115 6 0 0 5 0 0 1 0 8 0 1 0 11 8] + [ 2 14 19 1 3 24 4 1079 0 2 6 8 3 2 0 0 1 0 27 14 9] + [ 14 3 1 1 2 2 0 0 964 46 11 2 3 18 14 1 3 1 1 1 1] + [ 121 1 1 0 5 0 1 0 16 922 0 2 1 28 7 3 1 0 0 3 7] + [ 2 5 13 4 0 1 4 3 11 1 968 2 0 12 3 1 6 2 4 1 10] + [ 0 0 2 0 0 19 0 3 0 0 0 946 29 8 0 4 2 16 0 4 2] + [ 0 1 4 3 1 2 2 1 0 1 4 24 984 5 3 7 1 11 2 5 7] + [ 0 1 3 0 4 10 1 0 9 12 4 2 0 1056 3 1 0 0 0 2 11] + [ 11 2 4 20 8 0 1 0 23 3 2 0 3 1 1006 0 1 2 3 0 11] + [ 0 1 1 0 4 0 0 0 0 0 0 4 9 0 1 1077 15 11 2 5 4] + [ 0 15 1 1 5 8 0 0 0 0 0 2 0 3 6 9 1088 0 2 10 11] + [ 0 0 2 0 0 0 1 0 1 1 0 1 18 0 2 6 1 1002 1 1 1] + [ 3 5 5 15 1 1 0 25 3 1 3 1 2 0 19 0 0 0 974 2 8] + [ 0 1 2 1 0 2 7 5 0 0 0 15 5 0 0 5 3 1 0 1094 11] + [ 138 179 182 69 86 165 34 93 62 81 166 82 374 285 128 59 95 68 140 224 5195]] + +2023-10-02 21:21:18,367 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:21:18,367 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:21:18,373 - + +2023-10-02 21:21:18,373 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:21:19,394 - Epoch: [109][ 10/ 1236] Overall Loss 0.189643 Objective Loss 0.189643 LR 0.000500 Time 0.101989 +2023-10-02 21:21:19,601 - Epoch: [109][ 20/ 1236] Overall Loss 0.210667 Objective Loss 0.210667 LR 0.000500 Time 0.061346 +2023-10-02 21:21:19,808 - Epoch: [109][ 30/ 1236] Overall Loss 0.204814 Objective Loss 0.204814 LR 0.000500 Time 0.047771 +2023-10-02 21:21:20,016 - Epoch: [109][ 40/ 1236] Overall Loss 0.207296 Objective Loss 0.207296 LR 0.000500 Time 0.041038 +2023-10-02 21:21:20,221 - Epoch: [109][ 50/ 1236] Overall Loss 0.208255 Objective Loss 0.208255 LR 0.000500 Time 0.036923 +2023-10-02 21:21:20,431 - Epoch: [109][ 60/ 1236] Overall Loss 0.210322 Objective Loss 0.210322 LR 0.000500 Time 0.034250 +2023-10-02 21:21:20,636 - Epoch: [109][ 70/ 1236] Overall Loss 0.213116 Objective Loss 0.213116 LR 0.000500 Time 0.032285 +2023-10-02 21:21:20,844 - Epoch: [109][ 80/ 1236] Overall Loss 0.213433 Objective Loss 0.213433 LR 0.000500 Time 0.030854 +2023-10-02 21:21:21,050 - Epoch: [109][ 90/ 1236] Overall Loss 0.216994 Objective Loss 0.216994 LR 0.000500 Time 0.029710 +2023-10-02 21:21:21,259 - Epoch: [109][ 100/ 1236] Overall Loss 0.216428 Objective Loss 0.216428 LR 0.000500 Time 0.028826 +2023-10-02 21:21:21,465 - Epoch: [109][ 110/ 1236] Overall Loss 0.216786 Objective Loss 0.216786 LR 0.000500 Time 0.028075 +2023-10-02 21:21:21,673 - Epoch: [109][ 120/ 1236] Overall Loss 0.215518 Objective Loss 0.215518 LR 0.000500 Time 0.027463 +2023-10-02 21:21:21,880 - Epoch: [109][ 130/ 1236] Overall Loss 0.215284 Objective Loss 0.215284 LR 0.000500 Time 0.026938 +2023-10-02 21:21:22,088 - Epoch: [109][ 140/ 1236] Overall Loss 0.215510 Objective Loss 0.215510 LR 0.000500 Time 0.026503 +2023-10-02 21:21:22,294 - Epoch: [109][ 150/ 1236] Overall Loss 0.214090 Objective Loss 0.214090 LR 0.000500 Time 0.026106 +2023-10-02 21:21:22,504 - Epoch: [109][ 160/ 1236] Overall Loss 0.214459 Objective Loss 0.214459 LR 0.000500 Time 0.025781 +2023-10-02 21:21:22,709 - Epoch: [109][ 170/ 1236] Overall Loss 0.213151 Objective Loss 0.213151 LR 0.000500 Time 0.025472 +2023-10-02 21:21:22,917 - Epoch: [109][ 180/ 1236] Overall Loss 0.213939 Objective Loss 0.213939 LR 0.000500 Time 0.025213 +2023-10-02 21:21:23,123 - Epoch: [109][ 190/ 1236] Overall Loss 0.212778 Objective Loss 0.212778 LR 0.000500 Time 0.024961 +2023-10-02 21:21:23,333 - Epoch: [109][ 200/ 1236] Overall Loss 0.212376 Objective Loss 0.212376 LR 0.000500 Time 0.024760 +2023-10-02 21:21:23,538 - Epoch: [109][ 210/ 1236] Overall Loss 0.211866 Objective Loss 0.211866 LR 0.000500 Time 0.024556 +2023-10-02 21:21:23,746 - Epoch: [109][ 220/ 1236] Overall Loss 0.211149 Objective Loss 0.211149 LR 0.000500 Time 0.024383 +2023-10-02 21:21:23,952 - Epoch: [109][ 230/ 1236] Overall Loss 0.210345 Objective Loss 0.210345 LR 0.000500 Time 0.024212 +2023-10-02 21:21:24,160 - Epoch: [109][ 240/ 1236] Overall Loss 0.209727 Objective Loss 0.209727 LR 0.000500 Time 0.024069 +2023-10-02 21:21:24,366 - Epoch: [109][ 250/ 1236] Overall Loss 0.209347 Objective Loss 0.209347 LR 0.000500 Time 0.023925 +2023-10-02 21:21:24,574 - Epoch: [109][ 260/ 1236] Overall Loss 0.208944 Objective Loss 0.208944 LR 0.000500 Time 0.023805 +2023-10-02 21:21:24,780 - Epoch: [109][ 270/ 1236] Overall Loss 0.209126 Objective Loss 0.209126 LR 0.000500 Time 0.023680 +2023-10-02 21:21:24,988 - Epoch: [109][ 280/ 1236] Overall Loss 0.208874 Objective Loss 0.208874 LR 0.000500 Time 0.023577 +2023-10-02 21:21:25,195 - Epoch: [109][ 290/ 1236] Overall Loss 0.209626 Objective Loss 0.209626 LR 0.000500 Time 0.023470 +2023-10-02 21:21:25,403 - Epoch: [109][ 300/ 1236] Overall Loss 0.209477 Objective Loss 0.209477 LR 0.000500 Time 0.023379 +2023-10-02 21:21:25,609 - Epoch: [109][ 310/ 1236] Overall Loss 0.209663 Objective Loss 0.209663 LR 0.000500 Time 0.023285 +2023-10-02 21:21:25,816 - Epoch: [109][ 320/ 1236] Overall Loss 0.210106 Objective Loss 0.210106 LR 0.000500 Time 0.023203 +2023-10-02 21:21:26,021 - Epoch: [109][ 330/ 1236] Overall Loss 0.209592 Objective Loss 0.209592 LR 0.000500 Time 0.023118 +2023-10-02 21:21:26,231 - Epoch: [109][ 340/ 1236] Overall Loss 0.210225 Objective Loss 0.210225 LR 0.000500 Time 0.023054 +2023-10-02 21:21:26,438 - Epoch: [109][ 350/ 1236] Overall Loss 0.209591 Objective Loss 0.209591 LR 0.000500 Time 0.022982 +2023-10-02 21:21:26,646 - Epoch: [109][ 360/ 1236] Overall Loss 0.210004 Objective Loss 0.210004 LR 0.000500 Time 0.022921 +2023-10-02 21:21:26,853 - Epoch: [109][ 370/ 1236] Overall Loss 0.210351 Objective Loss 0.210351 LR 0.000500 Time 0.022857 +2023-10-02 21:21:27,062 - Epoch: [109][ 380/ 1236] Overall Loss 0.209576 Objective Loss 0.209576 LR 0.000500 Time 0.022805 +2023-10-02 21:21:27,269 - Epoch: [109][ 390/ 1236] Overall Loss 0.209478 Objective Loss 0.209478 LR 0.000500 Time 0.022751 +2023-10-02 21:21:27,478 - Epoch: [109][ 400/ 1236] Overall Loss 0.210540 Objective Loss 0.210540 LR 0.000500 Time 0.022704 +2023-10-02 21:21:27,686 - Epoch: [109][ 410/ 1236] Overall Loss 0.210645 Objective Loss 0.210645 LR 0.000500 Time 0.022656 +2023-10-02 21:21:27,895 - Epoch: [109][ 420/ 1236] Overall Loss 0.211128 Objective Loss 0.211128 LR 0.000500 Time 0.022613 +2023-10-02 21:21:28,103 - Epoch: [109][ 430/ 1236] Overall Loss 0.211636 Objective Loss 0.211636 LR 0.000500 Time 0.022570 +2023-10-02 21:21:28,312 - Epoch: [109][ 440/ 1236] Overall Loss 0.211870 Objective Loss 0.211870 LR 0.000500 Time 0.022531 +2023-10-02 21:21:28,520 - Epoch: [109][ 450/ 1236] Overall Loss 0.212192 Objective Loss 0.212192 LR 0.000500 Time 0.022493 +2023-10-02 21:21:28,729 - Epoch: [109][ 460/ 1236] Overall Loss 0.211720 Objective Loss 0.211720 LR 0.000500 Time 0.022457 +2023-10-02 21:21:28,936 - Epoch: [109][ 470/ 1236] Overall Loss 0.211321 Objective Loss 0.211321 LR 0.000500 Time 0.022417 +2023-10-02 21:21:29,145 - Epoch: [109][ 480/ 1236] Overall Loss 0.211032 Objective Loss 0.211032 LR 0.000500 Time 0.022385 +2023-10-02 21:21:29,352 - Epoch: [109][ 490/ 1236] Overall Loss 0.211115 Objective Loss 0.211115 LR 0.000500 Time 0.022350 +2023-10-02 21:21:29,563 - Epoch: [109][ 500/ 1236] Overall Loss 0.210816 Objective Loss 0.210816 LR 0.000500 Time 0.022324 +2023-10-02 21:21:29,769 - Epoch: [109][ 510/ 1236] Overall Loss 0.210981 Objective Loss 0.210981 LR 0.000500 Time 0.022290 +2023-10-02 21:21:29,978 - Epoch: [109][ 520/ 1236] Overall Loss 0.211110 Objective Loss 0.211110 LR 0.000500 Time 0.022263 +2023-10-02 21:21:30,186 - Epoch: [109][ 530/ 1236] Overall Loss 0.211052 Objective Loss 0.211052 LR 0.000500 Time 0.022235 +2023-10-02 21:21:30,396 - Epoch: [109][ 540/ 1236] Overall Loss 0.210865 Objective Loss 0.210865 LR 0.000500 Time 0.022210 +2023-10-02 21:21:30,603 - Epoch: [109][ 550/ 1236] Overall Loss 0.210917 Objective Loss 0.210917 LR 0.000500 Time 0.022181 +2023-10-02 21:21:30,812 - Epoch: [109][ 560/ 1236] Overall Loss 0.211340 Objective Loss 0.211340 LR 0.000500 Time 0.022157 +2023-10-02 21:21:31,019 - Epoch: [109][ 570/ 1236] Overall Loss 0.211525 Objective Loss 0.211525 LR 0.000500 Time 0.022131 +2023-10-02 21:21:31,228 - Epoch: [109][ 580/ 1236] Overall Loss 0.211652 Objective Loss 0.211652 LR 0.000500 Time 0.022110 +2023-10-02 21:21:31,436 - Epoch: [109][ 590/ 1236] Overall Loss 0.211679 Objective Loss 0.211679 LR 0.000500 Time 0.022086 +2023-10-02 21:21:31,645 - Epoch: [109][ 600/ 1236] Overall Loss 0.211683 Objective Loss 0.211683 LR 0.000500 Time 0.022066 +2023-10-02 21:21:31,854 - Epoch: [109][ 610/ 1236] Overall Loss 0.211754 Objective Loss 0.211754 LR 0.000500 Time 0.022045 +2023-10-02 21:21:32,066 - Epoch: [109][ 620/ 1236] Overall Loss 0.211548 Objective Loss 0.211548 LR 0.000500 Time 0.022031 +2023-10-02 21:21:32,272 - Epoch: [109][ 630/ 1236] Overall Loss 0.211743 Objective Loss 0.211743 LR 0.000500 Time 0.022008 +2023-10-02 21:21:32,483 - Epoch: [109][ 640/ 1236] Overall Loss 0.212109 Objective Loss 0.212109 LR 0.000500 Time 0.021992 +2023-10-02 21:21:32,689 - Epoch: [109][ 650/ 1236] Overall Loss 0.212456 Objective Loss 0.212456 LR 0.000500 Time 0.021971 +2023-10-02 21:21:32,899 - Epoch: [109][ 660/ 1236] Overall Loss 0.212471 Objective Loss 0.212471 LR 0.000500 Time 0.021956 +2023-10-02 21:21:33,107 - Epoch: [109][ 670/ 1236] Overall Loss 0.212481 Objective Loss 0.212481 LR 0.000500 Time 0.021938 +2023-10-02 21:21:33,317 - Epoch: [109][ 680/ 1236] Overall Loss 0.212423 Objective Loss 0.212423 LR 0.000500 Time 0.021923 +2023-10-02 21:21:33,524 - Epoch: [109][ 690/ 1236] Overall Loss 0.212429 Objective Loss 0.212429 LR 0.000500 Time 0.021906 +2023-10-02 21:21:33,735 - Epoch: [109][ 700/ 1236] Overall Loss 0.212371 Objective Loss 0.212371 LR 0.000500 Time 0.021893 +2023-10-02 21:21:33,943 - Epoch: [109][ 710/ 1236] Overall Loss 0.212137 Objective Loss 0.212137 LR 0.000500 Time 0.021875 +2023-10-02 21:21:34,153 - Epoch: [109][ 720/ 1236] Overall Loss 0.212165 Objective Loss 0.212165 LR 0.000500 Time 0.021862 +2023-10-02 21:21:34,361 - Epoch: [109][ 730/ 1236] Overall Loss 0.212049 Objective Loss 0.212049 LR 0.000500 Time 0.021848 +2023-10-02 21:21:34,572 - Epoch: [109][ 740/ 1236] Overall Loss 0.211730 Objective Loss 0.211730 LR 0.000500 Time 0.021837 +2023-10-02 21:21:34,780 - Epoch: [109][ 750/ 1236] Overall Loss 0.211601 Objective Loss 0.211601 LR 0.000500 Time 0.021823 +2023-10-02 21:21:34,990 - Epoch: [109][ 760/ 1236] Overall Loss 0.211936 Objective Loss 0.211936 LR 0.000500 Time 0.021811 +2023-10-02 21:21:35,198 - Epoch: [109][ 770/ 1236] Overall Loss 0.212019 Objective Loss 0.212019 LR 0.000500 Time 0.021798 +2023-10-02 21:21:35,408 - Epoch: [109][ 780/ 1236] Overall Loss 0.211759 Objective Loss 0.211759 LR 0.000500 Time 0.021787 +2023-10-02 21:21:35,617 - Epoch: [109][ 790/ 1236] Overall Loss 0.211577 Objective Loss 0.211577 LR 0.000500 Time 0.021775 +2023-10-02 21:21:35,826 - Epoch: [109][ 800/ 1236] Overall Loss 0.211350 Objective Loss 0.211350 LR 0.000500 Time 0.021764 +2023-10-02 21:21:36,033 - Epoch: [109][ 810/ 1236] Overall Loss 0.211310 Objective Loss 0.211310 LR 0.000500 Time 0.021750 +2023-10-02 21:21:36,243 - Epoch: [109][ 820/ 1236] Overall Loss 0.211278 Objective Loss 0.211278 LR 0.000500 Time 0.021740 +2023-10-02 21:21:36,451 - Epoch: [109][ 830/ 1236] Overall Loss 0.211189 Objective Loss 0.211189 LR 0.000500 Time 0.021729 +2023-10-02 21:21:36,668 - Epoch: [109][ 840/ 1236] Overall Loss 0.211378 Objective Loss 0.211378 LR 0.000500 Time 0.021728 +2023-10-02 21:21:36,881 - Epoch: [109][ 850/ 1236] Overall Loss 0.211274 Objective Loss 0.211274 LR 0.000500 Time 0.021723 +2023-10-02 21:21:37,104 - Epoch: [109][ 860/ 1236] Overall Loss 0.211449 Objective Loss 0.211449 LR 0.000500 Time 0.021729 +2023-10-02 21:21:37,320 - Epoch: [109][ 870/ 1236] Overall Loss 0.211491 Objective Loss 0.211491 LR 0.000500 Time 0.021727 +2023-10-02 21:21:37,543 - Epoch: [109][ 880/ 1236] Overall Loss 0.211373 Objective Loss 0.211373 LR 0.000500 Time 0.021733 +2023-10-02 21:21:37,759 - Epoch: [109][ 890/ 1236] Overall Loss 0.211596 Objective Loss 0.211596 LR 0.000500 Time 0.021731 +2023-10-02 21:21:37,974 - Epoch: [109][ 900/ 1236] Overall Loss 0.211640 Objective Loss 0.211640 LR 0.000500 Time 0.021727 +2023-10-02 21:21:38,184 - Epoch: [109][ 910/ 1236] Overall Loss 0.211699 Objective Loss 0.211699 LR 0.000500 Time 0.021719 +2023-10-02 21:21:38,398 - Epoch: [109][ 920/ 1236] Overall Loss 0.211632 Objective Loss 0.211632 LR 0.000500 Time 0.021715 +2023-10-02 21:21:38,609 - Epoch: [109][ 930/ 1236] Overall Loss 0.211638 Objective Loss 0.211638 LR 0.000500 Time 0.021707 +2023-10-02 21:21:38,822 - Epoch: [109][ 940/ 1236] Overall Loss 0.211763 Objective Loss 0.211763 LR 0.000500 Time 0.021703 +2023-10-02 21:21:39,034 - Epoch: [109][ 950/ 1236] Overall Loss 0.211850 Objective Loss 0.211850 LR 0.000500 Time 0.021696 +2023-10-02 21:21:39,249 - Epoch: [109][ 960/ 1236] Overall Loss 0.211648 Objective Loss 0.211648 LR 0.000500 Time 0.021694 +2023-10-02 21:21:39,460 - Epoch: [109][ 970/ 1236] Overall Loss 0.211735 Objective Loss 0.211735 LR 0.000500 Time 0.021687 +2023-10-02 21:21:39,675 - Epoch: [109][ 980/ 1236] Overall Loss 0.212040 Objective Loss 0.212040 LR 0.000500 Time 0.021685 +2023-10-02 21:21:39,884 - Epoch: [109][ 990/ 1236] Overall Loss 0.212095 Objective Loss 0.212095 LR 0.000500 Time 0.021677 +2023-10-02 21:21:40,099 - Epoch: [109][ 1000/ 1236] Overall Loss 0.212180 Objective Loss 0.212180 LR 0.000500 Time 0.021674 +2023-10-02 21:21:40,308 - Epoch: [109][ 1010/ 1236] Overall Loss 0.212323 Objective Loss 0.212323 LR 0.000500 Time 0.021667 +2023-10-02 21:21:40,522 - Epoch: [109][ 1020/ 1236] Overall Loss 0.212513 Objective Loss 0.212513 LR 0.000500 Time 0.021663 +2023-10-02 21:21:40,733 - Epoch: [109][ 1030/ 1236] Overall Loss 0.212668 Objective Loss 0.212668 LR 0.000500 Time 0.021658 +2023-10-02 21:21:40,947 - Epoch: [109][ 1040/ 1236] Overall Loss 0.212940 Objective Loss 0.212940 LR 0.000500 Time 0.021655 +2023-10-02 21:21:41,158 - Epoch: [109][ 1050/ 1236] Overall Loss 0.213201 Objective Loss 0.213201 LR 0.000500 Time 0.021648 +2023-10-02 21:21:41,372 - Epoch: [109][ 1060/ 1236] Overall Loss 0.213193 Objective Loss 0.213193 LR 0.000500 Time 0.021645 +2023-10-02 21:21:41,583 - Epoch: [109][ 1070/ 1236] Overall Loss 0.213266 Objective Loss 0.213266 LR 0.000500 Time 0.021639 +2023-10-02 21:21:41,798 - Epoch: [109][ 1080/ 1236] Overall Loss 0.213194 Objective Loss 0.213194 LR 0.000500 Time 0.021637 +2023-10-02 21:21:42,009 - Epoch: [109][ 1090/ 1236] Overall Loss 0.213592 Objective Loss 0.213592 LR 0.000500 Time 0.021631 +2023-10-02 21:21:42,224 - Epoch: [109][ 1100/ 1236] Overall Loss 0.213757 Objective Loss 0.213757 LR 0.000500 Time 0.021630 +2023-10-02 21:21:42,434 - Epoch: [109][ 1110/ 1236] Overall Loss 0.214049 Objective Loss 0.214049 LR 0.000500 Time 0.021624 +2023-10-02 21:21:42,648 - Epoch: [109][ 1120/ 1236] Overall Loss 0.214084 Objective Loss 0.214084 LR 0.000500 Time 0.021622 +2023-10-02 21:21:42,859 - Epoch: [109][ 1130/ 1236] Overall Loss 0.214040 Objective Loss 0.214040 LR 0.000500 Time 0.021617 +2023-10-02 21:21:43,074 - Epoch: [109][ 1140/ 1236] Overall Loss 0.214138 Objective Loss 0.214138 LR 0.000500 Time 0.021616 +2023-10-02 21:21:43,285 - Epoch: [109][ 1150/ 1236] Overall Loss 0.214430 Objective Loss 0.214430 LR 0.000500 Time 0.021610 +2023-10-02 21:21:43,499 - Epoch: [109][ 1160/ 1236] Overall Loss 0.214525 Objective Loss 0.214525 LR 0.000500 Time 0.021609 +2023-10-02 21:21:43,710 - Epoch: [109][ 1170/ 1236] Overall Loss 0.214491 Objective Loss 0.214491 LR 0.000500 Time 0.021604 +2023-10-02 21:21:43,926 - Epoch: [109][ 1180/ 1236] Overall Loss 0.214512 Objective Loss 0.214512 LR 0.000500 Time 0.021603 +2023-10-02 21:21:44,137 - Epoch: [109][ 1190/ 1236] Overall Loss 0.214693 Objective Loss 0.214693 LR 0.000500 Time 0.021599 +2023-10-02 21:21:44,352 - Epoch: [109][ 1200/ 1236] Overall Loss 0.214513 Objective Loss 0.214513 LR 0.000500 Time 0.021598 +2023-10-02 21:21:44,565 - Epoch: [109][ 1210/ 1236] Overall Loss 0.214630 Objective Loss 0.214630 LR 0.000500 Time 0.021594 +2023-10-02 21:21:44,780 - Epoch: [109][ 1220/ 1236] Overall Loss 0.214799 Objective Loss 0.214799 LR 0.000500 Time 0.021593 +2023-10-02 21:21:45,046 - Epoch: [109][ 1230/ 1236] Overall Loss 0.214755 Objective Loss 0.214755 LR 0.000500 Time 0.021632 +2023-10-02 21:21:45,168 - Epoch: [109][ 1236/ 1236] Overall Loss 0.214949 Objective Loss 0.214949 Top1 88.187373 Top5 99.389002 LR 0.000500 Time 0.021626 +2023-10-02 21:21:45,309 - --- validate (epoch=109)----------- +2023-10-02 21:21:45,309 - 29943 samples (256 per mini-batch) +2023-10-02 21:21:45,803 - Epoch: [109][ 10/ 117] Loss 0.277296 Top1 86.289062 Top5 98.554688 +2023-10-02 21:21:45,962 - Epoch: [109][ 20/ 117] Loss 0.280462 Top1 85.839844 Top5 98.359375 +2023-10-02 21:21:46,120 - Epoch: [109][ 30/ 117] Loss 0.290715 Top1 85.234375 Top5 98.476562 +2023-10-02 21:21:46,279 - Epoch: [109][ 40/ 117] Loss 0.288553 Top1 85.507812 Top5 98.457031 +2023-10-02 21:21:46,436 - Epoch: [109][ 50/ 117] Loss 0.285295 Top1 85.500000 Top5 98.390625 +2023-10-02 21:21:46,595 - Epoch: [109][ 60/ 117] Loss 0.286864 Top1 85.371094 Top5 98.365885 +2023-10-02 21:21:46,752 - Epoch: [109][ 70/ 117] Loss 0.285479 Top1 85.251116 Top5 98.381696 +2023-10-02 21:21:46,911 - Epoch: [109][ 80/ 117] Loss 0.289459 Top1 85.273438 Top5 98.383789 +2023-10-02 21:21:47,067 - Epoch: [109][ 90/ 117] Loss 0.292158 Top1 85.160590 Top5 98.328993 +2023-10-02 21:21:47,224 - Epoch: [109][ 100/ 117] Loss 0.293685 Top1 85.089844 Top5 98.347656 +2023-10-02 21:21:47,389 - Epoch: [109][ 110/ 117] Loss 0.294852 Top1 85.014205 Top5 98.341619 +2023-10-02 21:21:47,478 - Epoch: [109][ 117/ 117] Loss 0.294660 Top1 84.984804 Top5 98.356878 +2023-10-02 21:21:47,617 - ==> Top1: 84.985 Top5: 98.357 Loss: 0.295 + +2023-10-02 21:21:47,618 - ==> Confusion: +[[ 960 2 4 0 5 3 0 0 8 39 1 0 1 4 7 0 1 0 1 0 14] + [ 1 1050 0 0 4 22 0 33 0 2 1 1 0 0 0 3 0 0 10 1 3] + [ 4 0 984 9 0 0 16 12 0 1 0 0 7 1 0 2 2 1 7 2 8] + [ 1 1 6 968 0 5 0 1 3 1 6 0 5 4 30 4 2 4 28 0 20] + [ 28 7 2 1 965 2 0 0 1 9 1 0 1 3 9 6 8 1 0 0 6] + [ 3 31 0 2 2 997 1 30 2 3 2 9 1 10 5 0 4 0 3 2 9] + [ 0 2 24 2 0 2 1127 10 0 0 2 0 0 1 0 4 0 0 2 10 5] + [ 2 11 7 1 3 23 8 1084 0 3 4 7 4 0 1 2 0 1 36 12 9] + [ 16 0 1 0 1 3 0 1 984 34 12 1 1 12 12 3 4 0 3 1 0] + [ 112 1 2 0 7 3 1 0 39 910 1 0 0 22 11 1 1 0 1 1 6] + [ 2 5 7 6 0 0 3 2 9 0 980 3 0 9 6 0 1 3 5 2 10] + [ 2 0 3 0 0 6 0 3 0 0 0 961 27 10 0 0 0 14 0 4 5] + [ 0 0 2 5 2 4 1 1 1 0 0 53 960 1 2 4 1 10 5 5 11] + [ 0 0 2 0 3 10 2 0 16 9 3 6 1 1051 4 0 0 1 0 1 10] + [ 10 2 6 20 3 0 0 0 20 1 2 0 3 3 1011 0 0 4 9 0 7] + [ 0 0 1 0 7 0 0 0 0 0 0 8 6 0 0 1073 15 12 3 6 3] + [ 3 18 1 2 7 9 0 1 0 0 0 7 0 1 7 8 1078 0 2 6 11] + [ 0 0 1 1 0 0 2 0 0 1 0 5 15 1 3 4 1 999 0 2 3] + [ 1 5 3 9 1 0 0 33 3 0 2 1 1 0 9 0 0 0 991 0 9] + [ 0 1 4 2 0 1 10 10 0 0 2 13 3 1 0 2 3 1 1 1092 6] + [ 169 168 138 70 80 152 41 112 90 77 178 100 346 297 154 51 66 53 145 196 5222]] + +2023-10-02 21:21:47,619 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:21:47,619 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:21:47,625 - + +2023-10-02 21:21:47,625 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:21:48,742 - Epoch: [110][ 10/ 1236] Overall Loss 0.183211 Objective Loss 0.183211 LR 0.000500 Time 0.111612 +2023-10-02 21:21:48,953 - Epoch: [110][ 20/ 1236] Overall Loss 0.211361 Objective Loss 0.211361 LR 0.000500 Time 0.066330 +2023-10-02 21:21:49,160 - Epoch: [110][ 30/ 1236] Overall Loss 0.218516 Objective Loss 0.218516 LR 0.000500 Time 0.051090 +2023-10-02 21:21:49,367 - Epoch: [110][ 40/ 1236] Overall Loss 0.223527 Objective Loss 0.223527 LR 0.000500 Time 0.043499 +2023-10-02 21:21:49,574 - Epoch: [110][ 50/ 1236] Overall Loss 0.219987 Objective Loss 0.219987 LR 0.000500 Time 0.038923 +2023-10-02 21:21:49,782 - Epoch: [110][ 60/ 1236] Overall Loss 0.215941 Objective Loss 0.215941 LR 0.000500 Time 0.035892 +2023-10-02 21:21:49,988 - Epoch: [110][ 70/ 1236] Overall Loss 0.211326 Objective Loss 0.211326 LR 0.000500 Time 0.033705 +2023-10-02 21:21:50,195 - Epoch: [110][ 80/ 1236] Overall Loss 0.212955 Objective Loss 0.212955 LR 0.000500 Time 0.032085 +2023-10-02 21:21:50,402 - Epoch: [110][ 90/ 1236] Overall Loss 0.210944 Objective Loss 0.210944 LR 0.000500 Time 0.030807 +2023-10-02 21:21:50,609 - Epoch: [110][ 100/ 1236] Overall Loss 0.212444 Objective Loss 0.212444 LR 0.000500 Time 0.029799 +2023-10-02 21:21:50,815 - Epoch: [110][ 110/ 1236] Overall Loss 0.213251 Objective Loss 0.213251 LR 0.000500 Time 0.028962 +2023-10-02 21:21:51,023 - Epoch: [110][ 120/ 1236] Overall Loss 0.212332 Objective Loss 0.212332 LR 0.000500 Time 0.028275 +2023-10-02 21:21:51,227 - Epoch: [110][ 130/ 1236] Overall Loss 0.211687 Objective Loss 0.211687 LR 0.000500 Time 0.027664 +2023-10-02 21:21:51,432 - Epoch: [110][ 140/ 1236] Overall Loss 0.211682 Objective Loss 0.211682 LR 0.000500 Time 0.027155 +2023-10-02 21:21:51,637 - Epoch: [110][ 150/ 1236] Overall Loss 0.210942 Objective Loss 0.210942 LR 0.000500 Time 0.026707 +2023-10-02 21:21:51,843 - Epoch: [110][ 160/ 1236] Overall Loss 0.210921 Objective Loss 0.210921 LR 0.000500 Time 0.026321 +2023-10-02 21:21:52,046 - Epoch: [110][ 170/ 1236] Overall Loss 0.210776 Objective Loss 0.210776 LR 0.000500 Time 0.025970 +2023-10-02 21:21:52,252 - Epoch: [110][ 180/ 1236] Overall Loss 0.209844 Objective Loss 0.209844 LR 0.000500 Time 0.025669 +2023-10-02 21:21:52,455 - Epoch: [110][ 190/ 1236] Overall Loss 0.209999 Objective Loss 0.209999 LR 0.000500 Time 0.025385 +2023-10-02 21:21:52,661 - Epoch: [110][ 200/ 1236] Overall Loss 0.210047 Objective Loss 0.210047 LR 0.000500 Time 0.025143 +2023-10-02 21:21:52,864 - Epoch: [110][ 210/ 1236] Overall Loss 0.210092 Objective Loss 0.210092 LR 0.000500 Time 0.024913 +2023-10-02 21:21:53,070 - Epoch: [110][ 220/ 1236] Overall Loss 0.211431 Objective Loss 0.211431 LR 0.000500 Time 0.024715 +2023-10-02 21:21:53,274 - Epoch: [110][ 230/ 1236] Overall Loss 0.210886 Objective Loss 0.210886 LR 0.000500 Time 0.024523 +2023-10-02 21:21:53,479 - Epoch: [110][ 240/ 1236] Overall Loss 0.211287 Objective Loss 0.211287 LR 0.000500 Time 0.024356 +2023-10-02 21:21:53,683 - Epoch: [110][ 250/ 1236] Overall Loss 0.211749 Objective Loss 0.211749 LR 0.000500 Time 0.024195 +2023-10-02 21:21:53,888 - Epoch: [110][ 260/ 1236] Overall Loss 0.210947 Objective Loss 0.210947 LR 0.000500 Time 0.024055 +2023-10-02 21:21:54,092 - Epoch: [110][ 270/ 1236] Overall Loss 0.211455 Objective Loss 0.211455 LR 0.000500 Time 0.023916 +2023-10-02 21:21:54,296 - Epoch: [110][ 280/ 1236] Overall Loss 0.211020 Objective Loss 0.211020 LR 0.000500 Time 0.023791 +2023-10-02 21:21:54,500 - Epoch: [110][ 290/ 1236] Overall Loss 0.210107 Objective Loss 0.210107 LR 0.000500 Time 0.023671 +2023-10-02 21:21:54,705 - Epoch: [110][ 300/ 1236] Overall Loss 0.209771 Objective Loss 0.209771 LR 0.000500 Time 0.023567 +2023-10-02 21:21:54,909 - Epoch: [110][ 310/ 1236] Overall Loss 0.209936 Objective Loss 0.209936 LR 0.000500 Time 0.023461 +2023-10-02 21:21:55,114 - Epoch: [110][ 320/ 1236] Overall Loss 0.209806 Objective Loss 0.209806 LR 0.000500 Time 0.023370 +2023-10-02 21:21:55,318 - Epoch: [110][ 330/ 1236] Overall Loss 0.209145 Objective Loss 0.209145 LR 0.000500 Time 0.023278 +2023-10-02 21:21:55,524 - Epoch: [110][ 340/ 1236] Overall Loss 0.209169 Objective Loss 0.209169 LR 0.000500 Time 0.023200 +2023-10-02 21:21:55,728 - Epoch: [110][ 350/ 1236] Overall Loss 0.209299 Objective Loss 0.209299 LR 0.000500 Time 0.023117 +2023-10-02 21:21:55,934 - Epoch: [110][ 360/ 1236] Overall Loss 0.209210 Objective Loss 0.209210 LR 0.000500 Time 0.023046 +2023-10-02 21:21:56,137 - Epoch: [110][ 370/ 1236] Overall Loss 0.209057 Objective Loss 0.209057 LR 0.000500 Time 0.022973 +2023-10-02 21:21:56,343 - Epoch: [110][ 380/ 1236] Overall Loss 0.208875 Objective Loss 0.208875 LR 0.000500 Time 0.022909 +2023-10-02 21:21:56,547 - Epoch: [110][ 390/ 1236] Overall Loss 0.208749 Objective Loss 0.208749 LR 0.000500 Time 0.022842 +2023-10-02 21:21:56,752 - Epoch: [110][ 400/ 1236] Overall Loss 0.208823 Objective Loss 0.208823 LR 0.000500 Time 0.022785 +2023-10-02 21:21:56,956 - Epoch: [110][ 410/ 1236] Overall Loss 0.208844 Objective Loss 0.208844 LR 0.000500 Time 0.022725 +2023-10-02 21:21:57,162 - Epoch: [110][ 420/ 1236] Overall Loss 0.208615 Objective Loss 0.208615 LR 0.000500 Time 0.022673 +2023-10-02 21:21:57,365 - Epoch: [110][ 430/ 1236] Overall Loss 0.208674 Objective Loss 0.208674 LR 0.000500 Time 0.022618 +2023-10-02 21:21:57,571 - Epoch: [110][ 440/ 1236] Overall Loss 0.209071 Objective Loss 0.209071 LR 0.000500 Time 0.022571 +2023-10-02 21:21:57,774 - Epoch: [110][ 450/ 1236] Overall Loss 0.208822 Objective Loss 0.208822 LR 0.000500 Time 0.022520 +2023-10-02 21:21:57,980 - Epoch: [110][ 460/ 1236] Overall Loss 0.208664 Objective Loss 0.208664 LR 0.000500 Time 0.022478 +2023-10-02 21:21:58,183 - Epoch: [110][ 470/ 1236] Overall Loss 0.208446 Objective Loss 0.208446 LR 0.000500 Time 0.022431 +2023-10-02 21:21:58,389 - Epoch: [110][ 480/ 1236] Overall Loss 0.209090 Objective Loss 0.209090 LR 0.000500 Time 0.022392 +2023-10-02 21:21:58,592 - Epoch: [110][ 490/ 1236] Overall Loss 0.208876 Objective Loss 0.208876 LR 0.000500 Time 0.022350 +2023-10-02 21:21:58,798 - Epoch: [110][ 500/ 1236] Overall Loss 0.208992 Objective Loss 0.208992 LR 0.000500 Time 0.022314 +2023-10-02 21:21:59,002 - Epoch: [110][ 510/ 1236] Overall Loss 0.208780 Objective Loss 0.208780 LR 0.000500 Time 0.022275 +2023-10-02 21:21:59,208 - Epoch: [110][ 520/ 1236] Overall Loss 0.208950 Objective Loss 0.208950 LR 0.000500 Time 0.022242 +2023-10-02 21:21:59,411 - Epoch: [110][ 530/ 1236] Overall Loss 0.209325 Objective Loss 0.209325 LR 0.000500 Time 0.022206 +2023-10-02 21:21:59,617 - Epoch: [110][ 540/ 1236] Overall Loss 0.209256 Objective Loss 0.209256 LR 0.000500 Time 0.022175 +2023-10-02 21:21:59,821 - Epoch: [110][ 550/ 1236] Overall Loss 0.209482 Objective Loss 0.209482 LR 0.000500 Time 0.022142 +2023-10-02 21:22:00,027 - Epoch: [110][ 560/ 1236] Overall Loss 0.209210 Objective Loss 0.209210 LR 0.000500 Time 0.022114 +2023-10-02 21:22:00,230 - Epoch: [110][ 570/ 1236] Overall Loss 0.209441 Objective Loss 0.209441 LR 0.000500 Time 0.022082 +2023-10-02 21:22:00,436 - Epoch: [110][ 580/ 1236] Overall Loss 0.209584 Objective Loss 0.209584 LR 0.000500 Time 0.022056 +2023-10-02 21:22:00,640 - Epoch: [110][ 590/ 1236] Overall Loss 0.209944 Objective Loss 0.209944 LR 0.000500 Time 0.022026 +2023-10-02 21:22:00,845 - Epoch: [110][ 600/ 1236] Overall Loss 0.210318 Objective Loss 0.210318 LR 0.000500 Time 0.022002 +2023-10-02 21:22:01,049 - Epoch: [110][ 610/ 1236] Overall Loss 0.210195 Objective Loss 0.210195 LR 0.000500 Time 0.021974 +2023-10-02 21:22:01,255 - Epoch: [110][ 620/ 1236] Overall Loss 0.209853 Objective Loss 0.209853 LR 0.000500 Time 0.021951 +2023-10-02 21:22:01,458 - Epoch: [110][ 630/ 1236] Overall Loss 0.209770 Objective Loss 0.209770 LR 0.000500 Time 0.021926 +2023-10-02 21:22:01,664 - Epoch: [110][ 640/ 1236] Overall Loss 0.209949 Objective Loss 0.209949 LR 0.000500 Time 0.021905 +2023-10-02 21:22:01,868 - Epoch: [110][ 650/ 1236] Overall Loss 0.209723 Objective Loss 0.209723 LR 0.000500 Time 0.021881 +2023-10-02 21:22:02,074 - Epoch: [110][ 660/ 1236] Overall Loss 0.209722 Objective Loss 0.209722 LR 0.000500 Time 0.021861 +2023-10-02 21:22:02,278 - Epoch: [110][ 670/ 1236] Overall Loss 0.209652 Objective Loss 0.209652 LR 0.000500 Time 0.021838 +2023-10-02 21:22:02,484 - Epoch: [110][ 680/ 1236] Overall Loss 0.209856 Objective Loss 0.209856 LR 0.000500 Time 0.021819 +2023-10-02 21:22:02,687 - Epoch: [110][ 690/ 1236] Overall Loss 0.209167 Objective Loss 0.209167 LR 0.000500 Time 0.021797 +2023-10-02 21:22:02,893 - Epoch: [110][ 700/ 1236] Overall Loss 0.208620 Objective Loss 0.208620 LR 0.000500 Time 0.021780 +2023-10-02 21:22:03,096 - Epoch: [110][ 710/ 1236] Overall Loss 0.208989 Objective Loss 0.208989 LR 0.000500 Time 0.021759 +2023-10-02 21:22:03,302 - Epoch: [110][ 720/ 1236] Overall Loss 0.209246 Objective Loss 0.209246 LR 0.000500 Time 0.021742 +2023-10-02 21:22:03,506 - Epoch: [110][ 730/ 1236] Overall Loss 0.209407 Objective Loss 0.209407 LR 0.000500 Time 0.021723 +2023-10-02 21:22:03,712 - Epoch: [110][ 740/ 1236] Overall Loss 0.209657 Objective Loss 0.209657 LR 0.000500 Time 0.021707 +2023-10-02 21:22:03,915 - Epoch: [110][ 750/ 1236] Overall Loss 0.209604 Objective Loss 0.209604 LR 0.000500 Time 0.021689 +2023-10-02 21:22:04,121 - Epoch: [110][ 760/ 1236] Overall Loss 0.209352 Objective Loss 0.209352 LR 0.000500 Time 0.021674 +2023-10-02 21:22:04,325 - Epoch: [110][ 770/ 1236] Overall Loss 0.209533 Objective Loss 0.209533 LR 0.000500 Time 0.021657 +2023-10-02 21:22:04,531 - Epoch: [110][ 780/ 1236] Overall Loss 0.209561 Objective Loss 0.209561 LR 0.000500 Time 0.021643 +2023-10-02 21:22:04,735 - Epoch: [110][ 790/ 1236] Overall Loss 0.209841 Objective Loss 0.209841 LR 0.000500 Time 0.021626 +2023-10-02 21:22:04,941 - Epoch: [110][ 800/ 1236] Overall Loss 0.209816 Objective Loss 0.209816 LR 0.000500 Time 0.021613 +2023-10-02 21:22:05,144 - Epoch: [110][ 810/ 1236] Overall Loss 0.209754 Objective Loss 0.209754 LR 0.000500 Time 0.021597 +2023-10-02 21:22:05,350 - Epoch: [110][ 820/ 1236] Overall Loss 0.209707 Objective Loss 0.209707 LR 0.000500 Time 0.021585 +2023-10-02 21:22:05,554 - Epoch: [110][ 830/ 1236] Overall Loss 0.209564 Objective Loss 0.209564 LR 0.000500 Time 0.021569 +2023-10-02 21:22:05,760 - Epoch: [110][ 840/ 1236] Overall Loss 0.209299 Objective Loss 0.209299 LR 0.000500 Time 0.021558 +2023-10-02 21:22:05,963 - Epoch: [110][ 850/ 1236] Overall Loss 0.209297 Objective Loss 0.209297 LR 0.000500 Time 0.021543 +2023-10-02 21:22:06,169 - Epoch: [110][ 860/ 1236] Overall Loss 0.209167 Objective Loss 0.209167 LR 0.000500 Time 0.021532 +2023-10-02 21:22:06,373 - Epoch: [110][ 870/ 1236] Overall Loss 0.209167 Objective Loss 0.209167 LR 0.000500 Time 0.021518 +2023-10-02 21:22:06,579 - Epoch: [110][ 880/ 1236] Overall Loss 0.209108 Objective Loss 0.209108 LR 0.000500 Time 0.021507 +2023-10-02 21:22:06,782 - Epoch: [110][ 890/ 1236] Overall Loss 0.209033 Objective Loss 0.209033 LR 0.000500 Time 0.021494 +2023-10-02 21:22:06,988 - Epoch: [110][ 900/ 1236] Overall Loss 0.209007 Objective Loss 0.209007 LR 0.000500 Time 0.021483 +2023-10-02 21:22:07,192 - Epoch: [110][ 910/ 1236] Overall Loss 0.208909 Objective Loss 0.208909 LR 0.000500 Time 0.021471 +2023-10-02 21:22:07,398 - Epoch: [110][ 920/ 1236] Overall Loss 0.209076 Objective Loss 0.209076 LR 0.000500 Time 0.021461 +2023-10-02 21:22:07,601 - Epoch: [110][ 930/ 1236] Overall Loss 0.209183 Objective Loss 0.209183 LR 0.000500 Time 0.021449 +2023-10-02 21:22:07,807 - Epoch: [110][ 940/ 1236] Overall Loss 0.209173 Objective Loss 0.209173 LR 0.000500 Time 0.021439 +2023-10-02 21:22:08,011 - Epoch: [110][ 950/ 1236] Overall Loss 0.209015 Objective Loss 0.209015 LR 0.000500 Time 0.021427 +2023-10-02 21:22:08,217 - Epoch: [110][ 960/ 1236] Overall Loss 0.209055 Objective Loss 0.209055 LR 0.000500 Time 0.021419 +2023-10-02 21:22:08,420 - Epoch: [110][ 970/ 1236] Overall Loss 0.209438 Objective Loss 0.209438 LR 0.000500 Time 0.021407 +2023-10-02 21:22:08,626 - Epoch: [110][ 980/ 1236] Overall Loss 0.209547 Objective Loss 0.209547 LR 0.000500 Time 0.021399 +2023-10-02 21:22:08,830 - Epoch: [110][ 990/ 1236] Overall Loss 0.209520 Objective Loss 0.209520 LR 0.000500 Time 0.021388 +2023-10-02 21:22:09,036 - Epoch: [110][ 1000/ 1236] Overall Loss 0.209769 Objective Loss 0.209769 LR 0.000500 Time 0.021380 +2023-10-02 21:22:09,239 - Epoch: [110][ 1010/ 1236] Overall Loss 0.209651 Objective Loss 0.209651 LR 0.000500 Time 0.021369 +2023-10-02 21:22:09,445 - Epoch: [110][ 1020/ 1236] Overall Loss 0.209872 Objective Loss 0.209872 LR 0.000500 Time 0.021361 +2023-10-02 21:22:09,649 - Epoch: [110][ 1030/ 1236] Overall Loss 0.209979 Objective Loss 0.209979 LR 0.000500 Time 0.021351 +2023-10-02 21:22:09,854 - Epoch: [110][ 1040/ 1236] Overall Loss 0.210180 Objective Loss 0.210180 LR 0.000500 Time 0.021343 +2023-10-02 21:22:10,058 - Epoch: [110][ 1050/ 1236] Overall Loss 0.210176 Objective Loss 0.210176 LR 0.000500 Time 0.021334 +2023-10-02 21:22:10,264 - Epoch: [110][ 1060/ 1236] Overall Loss 0.210210 Objective Loss 0.210210 LR 0.000500 Time 0.021326 +2023-10-02 21:22:10,467 - Epoch: [110][ 1070/ 1236] Overall Loss 0.210242 Objective Loss 0.210242 LR 0.000500 Time 0.021317 +2023-10-02 21:22:10,673 - Epoch: [110][ 1080/ 1236] Overall Loss 0.210139 Objective Loss 0.210139 LR 0.000500 Time 0.021310 +2023-10-02 21:22:10,877 - Epoch: [110][ 1090/ 1236] Overall Loss 0.209975 Objective Loss 0.209975 LR 0.000500 Time 0.021301 +2023-10-02 21:22:11,083 - Epoch: [110][ 1100/ 1236] Overall Loss 0.210019 Objective Loss 0.210019 LR 0.000500 Time 0.021294 +2023-10-02 21:22:11,286 - Epoch: [110][ 1110/ 1236] Overall Loss 0.210052 Objective Loss 0.210052 LR 0.000500 Time 0.021285 +2023-10-02 21:22:11,492 - Epoch: [110][ 1120/ 1236] Overall Loss 0.210123 Objective Loss 0.210123 LR 0.000500 Time 0.021279 +2023-10-02 21:22:11,696 - Epoch: [110][ 1130/ 1236] Overall Loss 0.210301 Objective Loss 0.210301 LR 0.000500 Time 0.021271 +2023-10-02 21:22:11,902 - Epoch: [110][ 1140/ 1236] Overall Loss 0.210316 Objective Loss 0.210316 LR 0.000500 Time 0.021265 +2023-10-02 21:22:12,105 - Epoch: [110][ 1150/ 1236] Overall Loss 0.210357 Objective Loss 0.210357 LR 0.000500 Time 0.021256 +2023-10-02 21:22:12,311 - Epoch: [110][ 1160/ 1236] Overall Loss 0.210373 Objective Loss 0.210373 LR 0.000500 Time 0.021250 +2023-10-02 21:22:12,514 - Epoch: [110][ 1170/ 1236] Overall Loss 0.210332 Objective Loss 0.210332 LR 0.000500 Time 0.021242 +2023-10-02 21:22:12,720 - Epoch: [110][ 1180/ 1236] Overall Loss 0.210325 Objective Loss 0.210325 LR 0.000500 Time 0.021236 +2023-10-02 21:22:12,924 - Epoch: [110][ 1190/ 1236] Overall Loss 0.210142 Objective Loss 0.210142 LR 0.000500 Time 0.021229 +2023-10-02 21:22:13,130 - Epoch: [110][ 1200/ 1236] Overall Loss 0.210094 Objective Loss 0.210094 LR 0.000500 Time 0.021223 +2023-10-02 21:22:13,333 - Epoch: [110][ 1210/ 1236] Overall Loss 0.210224 Objective Loss 0.210224 LR 0.000500 Time 0.021215 +2023-10-02 21:22:13,539 - Epoch: [110][ 1220/ 1236] Overall Loss 0.210289 Objective Loss 0.210289 LR 0.000500 Time 0.021210 +2023-10-02 21:22:13,795 - Epoch: [110][ 1230/ 1236] Overall Loss 0.210483 Objective Loss 0.210483 LR 0.000500 Time 0.021245 +2023-10-02 21:22:13,915 - Epoch: [110][ 1236/ 1236] Overall Loss 0.210613 Objective Loss 0.210613 Top1 86.965377 Top5 97.963340 LR 0.000500 Time 0.021240 +2023-10-02 21:22:14,060 - --- validate (epoch=110)----------- +2023-10-02 21:22:14,060 - 29943 samples (256 per mini-batch) +2023-10-02 21:22:14,538 - Epoch: [110][ 10/ 117] Loss 0.308974 Top1 84.570312 Top5 98.242188 +2023-10-02 21:22:14,690 - Epoch: [110][ 20/ 117] Loss 0.302961 Top1 84.335938 Top5 98.339844 +2023-10-02 21:22:14,842 - Epoch: [110][ 30/ 117] Loss 0.298432 Top1 84.648438 Top5 98.268229 +2023-10-02 21:22:14,999 - Epoch: [110][ 40/ 117] Loss 0.300246 Top1 84.472656 Top5 98.242188 +2023-10-02 21:22:15,152 - Epoch: [110][ 50/ 117] Loss 0.302576 Top1 84.429688 Top5 98.257812 +2023-10-02 21:22:15,302 - Epoch: [110][ 60/ 117] Loss 0.300219 Top1 84.518229 Top5 98.268229 +2023-10-02 21:22:15,453 - Epoch: [110][ 70/ 117] Loss 0.297998 Top1 84.587054 Top5 98.337054 +2023-10-02 21:22:15,603 - Epoch: [110][ 80/ 117] Loss 0.296466 Top1 84.702148 Top5 98.339844 +2023-10-02 21:22:15,753 - Epoch: [110][ 90/ 117] Loss 0.296580 Top1 84.665799 Top5 98.276910 +2023-10-02 21:22:15,904 - Epoch: [110][ 100/ 117] Loss 0.294760 Top1 84.664062 Top5 98.312500 +2023-10-02 21:22:16,062 - Epoch: [110][ 110/ 117] Loss 0.295096 Top1 84.655540 Top5 98.345170 +2023-10-02 21:22:16,150 - Epoch: [110][ 117/ 117] Loss 0.295356 Top1 84.707611 Top5 98.346859 +2023-10-02 21:22:16,261 - ==> Top1: 84.708 Top5: 98.347 Loss: 0.295 + +2023-10-02 21:22:16,262 - ==> Confusion: +[[ 941 1 3 1 10 3 0 3 7 47 1 2 1 6 2 0 4 0 0 0 18] + [ 0 1048 0 0 3 31 1 24 0 0 2 0 3 0 0 3 2 1 10 2 1] + [ 5 0 964 7 1 1 32 8 0 0 0 1 12 1 1 2 1 1 8 5 6] + [ 1 4 10 974 0 8 4 1 4 0 7 0 6 3 20 2 0 7 21 1 16] + [ 27 10 0 0 960 5 0 0 1 13 1 0 2 1 10 4 9 1 0 1 5] + [ 3 24 0 0 3 1007 1 29 1 4 2 8 3 11 6 0 4 0 1 3 6] + [ 0 3 17 0 0 3 1143 4 0 0 4 0 0 0 0 4 0 0 1 8 4] + [ 2 12 5 0 4 23 6 1098 0 1 3 6 5 2 0 0 2 2 32 8 7] + [ 15 1 1 2 2 2 0 1 970 39 14 2 4 13 15 2 0 1 5 0 0] + [ 122 1 1 1 5 3 0 0 30 921 1 1 1 22 5 0 0 0 0 1 4] + [ 3 6 9 8 1 1 5 4 8 1 970 0 2 11 3 0 4 1 6 1 9] + [ 0 0 1 0 0 11 0 2 0 0 0 957 27 11 0 0 1 15 0 6 4] + [ 0 0 2 2 0 1 2 1 0 0 3 31 990 1 3 4 1 7 3 7 10] + [ 0 0 2 1 0 12 1 0 10 7 3 3 0 1061 3 0 0 1 0 5 10] + [ 10 1 3 22 10 0 0 0 20 3 0 0 3 3 998 0 1 4 11 0 12] + [ 0 1 1 1 3 0 1 0 0 1 0 5 7 0 0 1069 17 11 2 7 8] + [ 1 20 2 0 5 9 0 1 0 1 0 4 0 2 4 10 1082 0 0 6 14] + [ 0 0 1 1 0 1 5 0 1 1 0 4 26 0 1 6 1 987 0 0 3] + [ 2 8 3 11 1 1 0 31 2 1 1 1 0 0 9 0 0 0 986 1 10] + [ 0 4 5 2 0 6 7 8 0 0 0 8 2 2 1 1 5 1 0 1097 3] + [ 123 211 126 75 76 187 60 127 89 86 176 92 364 286 118 62 77 58 164 207 5141]] + +2023-10-02 21:22:16,263 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:22:16,264 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:22:16,269 - + +2023-10-02 21:22:16,270 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:22:17,283 - Epoch: [111][ 10/ 1236] Overall Loss 0.197082 Objective Loss 0.197082 LR 0.000500 Time 0.101321 +2023-10-02 21:22:17,493 - Epoch: [111][ 20/ 1236] Overall Loss 0.194861 Objective Loss 0.194861 LR 0.000500 Time 0.061106 +2023-10-02 21:22:17,704 - Epoch: [111][ 30/ 1236] Overall Loss 0.193591 Objective Loss 0.193591 LR 0.000500 Time 0.047722 +2023-10-02 21:22:17,922 - Epoch: [111][ 40/ 1236] Overall Loss 0.198775 Objective Loss 0.198775 LR 0.000500 Time 0.041246 +2023-10-02 21:22:18,129 - Epoch: [111][ 50/ 1236] Overall Loss 0.194175 Objective Loss 0.194175 LR 0.000500 Time 0.037130 +2023-10-02 21:22:18,339 - Epoch: [111][ 60/ 1236] Overall Loss 0.195468 Objective Loss 0.195468 LR 0.000500 Time 0.034426 +2023-10-02 21:22:18,546 - Epoch: [111][ 70/ 1236] Overall Loss 0.194982 Objective Loss 0.194982 LR 0.000500 Time 0.032449 +2023-10-02 21:22:18,756 - Epoch: [111][ 80/ 1236] Overall Loss 0.198824 Objective Loss 0.198824 LR 0.000500 Time 0.031014 +2023-10-02 21:22:18,964 - Epoch: [111][ 90/ 1236] Overall Loss 0.198766 Objective Loss 0.198766 LR 0.000500 Time 0.029855 +2023-10-02 21:22:19,174 - Epoch: [111][ 100/ 1236] Overall Loss 0.199908 Objective Loss 0.199908 LR 0.000500 Time 0.028965 +2023-10-02 21:22:19,383 - Epoch: [111][ 110/ 1236] Overall Loss 0.199828 Objective Loss 0.199828 LR 0.000500 Time 0.028223 +2023-10-02 21:22:19,593 - Epoch: [111][ 120/ 1236] Overall Loss 0.200080 Objective Loss 0.200080 LR 0.000500 Time 0.027619 +2023-10-02 21:22:19,799 - Epoch: [111][ 130/ 1236] Overall Loss 0.200495 Objective Loss 0.200495 LR 0.000500 Time 0.027068 +2023-10-02 21:22:20,007 - Epoch: [111][ 140/ 1236] Overall Loss 0.200534 Objective Loss 0.200534 LR 0.000500 Time 0.026617 +2023-10-02 21:22:20,214 - Epoch: [111][ 150/ 1236] Overall Loss 0.199940 Objective Loss 0.199940 LR 0.000500 Time 0.026212 +2023-10-02 21:22:20,421 - Epoch: [111][ 160/ 1236] Overall Loss 0.198666 Objective Loss 0.198666 LR 0.000500 Time 0.025866 +2023-10-02 21:22:20,627 - Epoch: [111][ 170/ 1236] Overall Loss 0.199047 Objective Loss 0.199047 LR 0.000500 Time 0.025547 +2023-10-02 21:22:20,834 - Epoch: [111][ 180/ 1236] Overall Loss 0.198149 Objective Loss 0.198149 LR 0.000500 Time 0.025277 +2023-10-02 21:22:21,040 - Epoch: [111][ 190/ 1236] Overall Loss 0.199195 Objective Loss 0.199195 LR 0.000500 Time 0.025021 +2023-10-02 21:22:21,248 - Epoch: [111][ 200/ 1236] Overall Loss 0.199698 Objective Loss 0.199698 LR 0.000500 Time 0.024810 +2023-10-02 21:22:21,453 - Epoch: [111][ 210/ 1236] Overall Loss 0.200007 Objective Loss 0.200007 LR 0.000500 Time 0.024602 +2023-10-02 21:22:21,661 - Epoch: [111][ 220/ 1236] Overall Loss 0.200122 Objective Loss 0.200122 LR 0.000500 Time 0.024427 +2023-10-02 21:22:21,865 - Epoch: [111][ 230/ 1236] Overall Loss 0.200617 Objective Loss 0.200617 LR 0.000500 Time 0.024250 +2023-10-02 21:22:22,072 - Epoch: [111][ 240/ 1236] Overall Loss 0.202076 Objective Loss 0.202076 LR 0.000500 Time 0.024103 +2023-10-02 21:22:22,276 - Epoch: [111][ 250/ 1236] Overall Loss 0.201848 Objective Loss 0.201848 LR 0.000500 Time 0.023954 +2023-10-02 21:22:22,484 - Epoch: [111][ 260/ 1236] Overall Loss 0.201361 Objective Loss 0.201361 LR 0.000500 Time 0.023831 +2023-10-02 21:22:22,688 - Epoch: [111][ 270/ 1236] Overall Loss 0.202272 Objective Loss 0.202272 LR 0.000500 Time 0.023703 +2023-10-02 21:22:22,894 - Epoch: [111][ 280/ 1236] Overall Loss 0.202252 Objective Loss 0.202252 LR 0.000500 Time 0.023592 +2023-10-02 21:22:23,097 - Epoch: [111][ 290/ 1236] Overall Loss 0.202816 Objective Loss 0.202816 LR 0.000500 Time 0.023477 +2023-10-02 21:22:23,305 - Epoch: [111][ 300/ 1236] Overall Loss 0.203264 Objective Loss 0.203264 LR 0.000500 Time 0.023387 +2023-10-02 21:22:23,509 - Epoch: [111][ 310/ 1236] Overall Loss 0.203349 Objective Loss 0.203349 LR 0.000500 Time 0.023290 +2023-10-02 21:22:23,718 - Epoch: [111][ 320/ 1236] Overall Loss 0.204070 Objective Loss 0.204070 LR 0.000500 Time 0.023212 +2023-10-02 21:22:23,922 - Epoch: [111][ 330/ 1236] Overall Loss 0.203670 Objective Loss 0.203670 LR 0.000500 Time 0.023127 +2023-10-02 21:22:24,130 - Epoch: [111][ 340/ 1236] Overall Loss 0.203600 Objective Loss 0.203600 LR 0.000500 Time 0.023058 +2023-10-02 21:22:24,334 - Epoch: [111][ 350/ 1236] Overall Loss 0.204081 Objective Loss 0.204081 LR 0.000500 Time 0.022980 +2023-10-02 21:22:24,542 - Epoch: [111][ 360/ 1236] Overall Loss 0.204184 Objective Loss 0.204184 LR 0.000500 Time 0.022921 +2023-10-02 21:22:24,748 - Epoch: [111][ 370/ 1236] Overall Loss 0.204835 Objective Loss 0.204835 LR 0.000500 Time 0.022854 +2023-10-02 21:22:24,956 - Epoch: [111][ 380/ 1236] Overall Loss 0.204943 Objective Loss 0.204943 LR 0.000500 Time 0.022798 +2023-10-02 21:22:25,164 - Epoch: [111][ 390/ 1236] Overall Loss 0.204696 Objective Loss 0.204696 LR 0.000500 Time 0.022742 +2023-10-02 21:22:25,373 - Epoch: [111][ 400/ 1236] Overall Loss 0.205115 Objective Loss 0.205115 LR 0.000500 Time 0.022697 +2023-10-02 21:22:25,581 - Epoch: [111][ 410/ 1236] Overall Loss 0.205546 Objective Loss 0.205546 LR 0.000500 Time 0.022649 +2023-10-02 21:22:25,791 - Epoch: [111][ 420/ 1236] Overall Loss 0.205276 Objective Loss 0.205276 LR 0.000500 Time 0.022609 +2023-10-02 21:22:25,999 - Epoch: [111][ 430/ 1236] Overall Loss 0.205103 Objective Loss 0.205103 LR 0.000500 Time 0.022566 +2023-10-02 21:22:26,209 - Epoch: [111][ 440/ 1236] Overall Loss 0.205352 Objective Loss 0.205352 LR 0.000500 Time 0.022529 +2023-10-02 21:22:26,417 - Epoch: [111][ 450/ 1236] Overall Loss 0.204922 Objective Loss 0.204922 LR 0.000500 Time 0.022489 +2023-10-02 21:22:26,627 - Epoch: [111][ 460/ 1236] Overall Loss 0.205090 Objective Loss 0.205090 LR 0.000500 Time 0.022456 +2023-10-02 21:22:26,835 - Epoch: [111][ 470/ 1236] Overall Loss 0.205179 Objective Loss 0.205179 LR 0.000500 Time 0.022417 +2023-10-02 21:22:27,044 - Epoch: [111][ 480/ 1236] Overall Loss 0.205036 Objective Loss 0.205036 LR 0.000500 Time 0.022386 +2023-10-02 21:22:27,253 - Epoch: [111][ 490/ 1236] Overall Loss 0.205135 Objective Loss 0.205135 LR 0.000500 Time 0.022353 +2023-10-02 21:22:27,462 - Epoch: [111][ 500/ 1236] Overall Loss 0.205194 Objective Loss 0.205194 LR 0.000500 Time 0.022325 +2023-10-02 21:22:27,671 - Epoch: [111][ 510/ 1236] Overall Loss 0.205270 Objective Loss 0.205270 LR 0.000500 Time 0.022295 +2023-10-02 21:22:27,880 - Epoch: [111][ 520/ 1236] Overall Loss 0.205011 Objective Loss 0.205011 LR 0.000500 Time 0.022268 +2023-10-02 21:22:28,088 - Epoch: [111][ 530/ 1236] Overall Loss 0.204556 Objective Loss 0.204556 LR 0.000500 Time 0.022239 +2023-10-02 21:22:28,297 - Epoch: [111][ 540/ 1236] Overall Loss 0.204509 Objective Loss 0.204509 LR 0.000500 Time 0.022215 +2023-10-02 21:22:28,505 - Epoch: [111][ 550/ 1236] Overall Loss 0.204233 Objective Loss 0.204233 LR 0.000500 Time 0.022189 +2023-10-02 21:22:28,715 - Epoch: [111][ 560/ 1236] Overall Loss 0.204190 Objective Loss 0.204190 LR 0.000500 Time 0.022166 +2023-10-02 21:22:28,923 - Epoch: [111][ 570/ 1236] Overall Loss 0.204373 Objective Loss 0.204373 LR 0.000500 Time 0.022142 +2023-10-02 21:22:29,133 - Epoch: [111][ 580/ 1236] Overall Loss 0.204054 Objective Loss 0.204054 LR 0.000500 Time 0.022121 +2023-10-02 21:22:29,341 - Epoch: [111][ 590/ 1236] Overall Loss 0.204011 Objective Loss 0.204011 LR 0.000500 Time 0.022097 +2023-10-02 21:22:29,550 - Epoch: [111][ 600/ 1236] Overall Loss 0.204043 Objective Loss 0.204043 LR 0.000500 Time 0.022078 +2023-10-02 21:22:29,758 - Epoch: [111][ 610/ 1236] Overall Loss 0.204070 Objective Loss 0.204070 LR 0.000500 Time 0.022057 +2023-10-02 21:22:29,968 - Epoch: [111][ 620/ 1236] Overall Loss 0.203915 Objective Loss 0.203915 LR 0.000500 Time 0.022038 +2023-10-02 21:22:30,177 - Epoch: [111][ 630/ 1236] Overall Loss 0.203829 Objective Loss 0.203829 LR 0.000500 Time 0.022019 +2023-10-02 21:22:30,386 - Epoch: [111][ 640/ 1236] Overall Loss 0.203907 Objective Loss 0.203907 LR 0.000500 Time 0.022002 +2023-10-02 21:22:30,594 - Epoch: [111][ 650/ 1236] Overall Loss 0.203551 Objective Loss 0.203551 LR 0.000500 Time 0.021983 +2023-10-02 21:22:30,804 - Epoch: [111][ 660/ 1236] Overall Loss 0.203533 Objective Loss 0.203533 LR 0.000500 Time 0.021967 +2023-10-02 21:22:31,012 - Epoch: [111][ 670/ 1236] Overall Loss 0.203407 Objective Loss 0.203407 LR 0.000500 Time 0.021949 +2023-10-02 21:22:31,221 - Epoch: [111][ 680/ 1236] Overall Loss 0.203563 Objective Loss 0.203563 LR 0.000500 Time 0.021934 +2023-10-02 21:22:31,429 - Epoch: [111][ 690/ 1236] Overall Loss 0.203708 Objective Loss 0.203708 LR 0.000500 Time 0.021916 +2023-10-02 21:22:31,639 - Epoch: [111][ 700/ 1236] Overall Loss 0.203649 Objective Loss 0.203649 LR 0.000500 Time 0.021902 +2023-10-02 21:22:31,847 - Epoch: [111][ 710/ 1236] Overall Loss 0.203222 Objective Loss 0.203222 LR 0.000500 Time 0.021886 +2023-10-02 21:22:32,057 - Epoch: [111][ 720/ 1236] Overall Loss 0.203665 Objective Loss 0.203665 LR 0.000500 Time 0.021874 +2023-10-02 21:22:32,269 - Epoch: [111][ 730/ 1236] Overall Loss 0.203344 Objective Loss 0.203344 LR 0.000500 Time 0.021862 +2023-10-02 21:22:32,479 - Epoch: [111][ 740/ 1236] Overall Loss 0.203565 Objective Loss 0.203565 LR 0.000500 Time 0.021850 +2023-10-02 21:22:32,690 - Epoch: [111][ 750/ 1236] Overall Loss 0.203293 Objective Loss 0.203293 LR 0.000500 Time 0.021837 +2023-10-02 21:22:32,901 - Epoch: [111][ 760/ 1236] Overall Loss 0.203570 Objective Loss 0.203570 LR 0.000500 Time 0.021827 +2023-10-02 21:22:33,114 - Epoch: [111][ 770/ 1236] Overall Loss 0.203629 Objective Loss 0.203629 LR 0.000500 Time 0.021820 +2023-10-02 21:22:33,326 - Epoch: [111][ 780/ 1236] Overall Loss 0.203739 Objective Loss 0.203739 LR 0.000500 Time 0.021811 +2023-10-02 21:22:33,538 - Epoch: [111][ 790/ 1236] Overall Loss 0.203973 Objective Loss 0.203973 LR 0.000500 Time 0.021801 +2023-10-02 21:22:33,750 - Epoch: [111][ 800/ 1236] Overall Loss 0.204424 Objective Loss 0.204424 LR 0.000500 Time 0.021793 +2023-10-02 21:22:33,963 - Epoch: [111][ 810/ 1236] Overall Loss 0.204585 Objective Loss 0.204585 LR 0.000500 Time 0.021786 +2023-10-02 21:22:34,174 - Epoch: [111][ 820/ 1236] Overall Loss 0.204611 Objective Loss 0.204611 LR 0.000500 Time 0.021778 +2023-10-02 21:22:34,384 - Epoch: [111][ 830/ 1236] Overall Loss 0.204463 Objective Loss 0.204463 LR 0.000500 Time 0.021769 +2023-10-02 21:22:34,593 - Epoch: [111][ 840/ 1236] Overall Loss 0.204427 Objective Loss 0.204427 LR 0.000500 Time 0.021756 +2023-10-02 21:22:34,802 - Epoch: [111][ 850/ 1236] Overall Loss 0.204711 Objective Loss 0.204711 LR 0.000500 Time 0.021745 +2023-10-02 21:22:35,011 - Epoch: [111][ 860/ 1236] Overall Loss 0.204744 Objective Loss 0.204744 LR 0.000500 Time 0.021735 +2023-10-02 21:22:35,220 - Epoch: [111][ 870/ 1236] Overall Loss 0.204629 Objective Loss 0.204629 LR 0.000500 Time 0.021725 +2023-10-02 21:22:35,429 - Epoch: [111][ 880/ 1236] Overall Loss 0.204498 Objective Loss 0.204498 LR 0.000500 Time 0.021715 +2023-10-02 21:22:35,638 - Epoch: [111][ 890/ 1236] Overall Loss 0.204653 Objective Loss 0.204653 LR 0.000500 Time 0.021706 +2023-10-02 21:22:35,847 - Epoch: [111][ 900/ 1236] Overall Loss 0.204739 Objective Loss 0.204739 LR 0.000500 Time 0.021696 +2023-10-02 21:22:36,057 - Epoch: [111][ 910/ 1236] Overall Loss 0.204751 Objective Loss 0.204751 LR 0.000500 Time 0.021688 +2023-10-02 21:22:36,265 - Epoch: [111][ 920/ 1236] Overall Loss 0.204782 Objective Loss 0.204782 LR 0.000500 Time 0.021677 +2023-10-02 21:22:36,475 - Epoch: [111][ 930/ 1236] Overall Loss 0.204894 Objective Loss 0.204894 LR 0.000500 Time 0.021669 +2023-10-02 21:22:36,684 - Epoch: [111][ 940/ 1236] Overall Loss 0.205296 Objective Loss 0.205296 LR 0.000500 Time 0.021661 +2023-10-02 21:22:36,900 - Epoch: [111][ 950/ 1236] Overall Loss 0.205413 Objective Loss 0.205413 LR 0.000500 Time 0.021660 +2023-10-02 21:22:37,109 - Epoch: [111][ 960/ 1236] Overall Loss 0.205361 Objective Loss 0.205361 LR 0.000500 Time 0.021650 +2023-10-02 21:22:37,319 - Epoch: [111][ 970/ 1236] Overall Loss 0.205808 Objective Loss 0.205808 LR 0.000500 Time 0.021643 +2023-10-02 21:22:37,528 - Epoch: [111][ 980/ 1236] Overall Loss 0.205889 Objective Loss 0.205889 LR 0.000500 Time 0.021634 +2023-10-02 21:22:37,738 - Epoch: [111][ 990/ 1236] Overall Loss 0.205941 Objective Loss 0.205941 LR 0.000500 Time 0.021627 +2023-10-02 21:22:37,947 - Epoch: [111][ 1000/ 1236] Overall Loss 0.205842 Objective Loss 0.205842 LR 0.000500 Time 0.021618 +2023-10-02 21:22:38,157 - Epoch: [111][ 1010/ 1236] Overall Loss 0.205790 Objective Loss 0.205790 LR 0.000500 Time 0.021612 +2023-10-02 21:22:38,366 - Epoch: [111][ 1020/ 1236] Overall Loss 0.205805 Objective Loss 0.205805 LR 0.000500 Time 0.021603 +2023-10-02 21:22:38,576 - Epoch: [111][ 1030/ 1236] Overall Loss 0.206019 Objective Loss 0.206019 LR 0.000500 Time 0.021597 +2023-10-02 21:22:38,785 - Epoch: [111][ 1040/ 1236] Overall Loss 0.206161 Objective Loss 0.206161 LR 0.000500 Time 0.021589 +2023-10-02 21:22:38,995 - Epoch: [111][ 1050/ 1236] Overall Loss 0.206221 Objective Loss 0.206221 LR 0.000500 Time 0.021583 +2023-10-02 21:22:39,204 - Epoch: [111][ 1060/ 1236] Overall Loss 0.206250 Objective Loss 0.206250 LR 0.000500 Time 0.021576 +2023-10-02 21:22:39,413 - Epoch: [111][ 1070/ 1236] Overall Loss 0.206341 Objective Loss 0.206341 LR 0.000500 Time 0.021569 +2023-10-02 21:22:39,621 - Epoch: [111][ 1080/ 1236] Overall Loss 0.206502 Objective Loss 0.206502 LR 0.000500 Time 0.021561 +2023-10-02 21:22:39,829 - Epoch: [111][ 1090/ 1236] Overall Loss 0.206549 Objective Loss 0.206549 LR 0.000500 Time 0.021553 +2023-10-02 21:22:40,037 - Epoch: [111][ 1100/ 1236] Overall Loss 0.206506 Objective Loss 0.206506 LR 0.000500 Time 0.021545 +2023-10-02 21:22:40,245 - Epoch: [111][ 1110/ 1236] Overall Loss 0.206350 Objective Loss 0.206350 LR 0.000500 Time 0.021538 +2023-10-02 21:22:40,453 - Epoch: [111][ 1120/ 1236] Overall Loss 0.206631 Objective Loss 0.206631 LR 0.000500 Time 0.021531 +2023-10-02 21:22:40,660 - Epoch: [111][ 1130/ 1236] Overall Loss 0.206458 Objective Loss 0.206458 LR 0.000500 Time 0.021524 +2023-10-02 21:22:40,868 - Epoch: [111][ 1140/ 1236] Overall Loss 0.206552 Objective Loss 0.206552 LR 0.000500 Time 0.021516 +2023-10-02 21:22:41,076 - Epoch: [111][ 1150/ 1236] Overall Loss 0.206758 Objective Loss 0.206758 LR 0.000500 Time 0.021509 +2023-10-02 21:22:41,283 - Epoch: [111][ 1160/ 1236] Overall Loss 0.206869 Objective Loss 0.206869 LR 0.000500 Time 0.021501 +2023-10-02 21:22:41,491 - Epoch: [111][ 1170/ 1236] Overall Loss 0.206784 Objective Loss 0.206784 LR 0.000500 Time 0.021495 +2023-10-02 21:22:41,699 - Epoch: [111][ 1180/ 1236] Overall Loss 0.206787 Objective Loss 0.206787 LR 0.000500 Time 0.021487 +2023-10-02 21:22:41,907 - Epoch: [111][ 1190/ 1236] Overall Loss 0.206629 Objective Loss 0.206629 LR 0.000500 Time 0.021481 +2023-10-02 21:22:42,115 - Epoch: [111][ 1200/ 1236] Overall Loss 0.206533 Objective Loss 0.206533 LR 0.000500 Time 0.021474 +2023-10-02 21:22:42,323 - Epoch: [111][ 1210/ 1236] Overall Loss 0.206539 Objective Loss 0.206539 LR 0.000500 Time 0.021468 +2023-10-02 21:22:42,531 - Epoch: [111][ 1220/ 1236] Overall Loss 0.206683 Objective Loss 0.206683 LR 0.000500 Time 0.021461 +2023-10-02 21:22:42,792 - Epoch: [111][ 1230/ 1236] Overall Loss 0.206729 Objective Loss 0.206729 LR 0.000500 Time 0.021498 +2023-10-02 21:22:42,914 - Epoch: [111][ 1236/ 1236] Overall Loss 0.206750 Objective Loss 0.206750 Top1 87.780041 Top5 97.759674 LR 0.000500 Time 0.021493 +2023-10-02 21:22:43,053 - --- validate (epoch=111)----------- +2023-10-02 21:22:43,053 - 29943 samples (256 per mini-batch) +2023-10-02 21:22:43,546 - Epoch: [111][ 10/ 117] Loss 0.302413 Top1 84.609375 Top5 98.476562 +2023-10-02 21:22:43,698 - Epoch: [111][ 20/ 117] Loss 0.292939 Top1 84.648438 Top5 98.593750 +2023-10-02 21:22:43,851 - Epoch: [111][ 30/ 117] Loss 0.278548 Top1 85.052083 Top5 98.515625 +2023-10-02 21:22:44,002 - Epoch: [111][ 40/ 117] Loss 0.288211 Top1 84.873047 Top5 98.398438 +2023-10-02 21:22:44,153 - Epoch: [111][ 50/ 117] Loss 0.290962 Top1 84.609375 Top5 98.320312 +2023-10-02 21:22:44,303 - Epoch: [111][ 60/ 117] Loss 0.289100 Top1 84.759115 Top5 98.339844 +2023-10-02 21:22:44,460 - Epoch: [111][ 70/ 117] Loss 0.293672 Top1 84.793527 Top5 98.320312 +2023-10-02 21:22:44,621 - Epoch: [111][ 80/ 117] Loss 0.289895 Top1 84.990234 Top5 98.339844 +2023-10-02 21:22:44,778 - Epoch: [111][ 90/ 117] Loss 0.291943 Top1 84.874132 Top5 98.311632 +2023-10-02 21:22:44,938 - Epoch: [111][ 100/ 117] Loss 0.292401 Top1 84.886719 Top5 98.300781 +2023-10-02 21:22:45,104 - Epoch: [111][ 110/ 117] Loss 0.292351 Top1 84.939631 Top5 98.338068 +2023-10-02 21:22:45,193 - Epoch: [111][ 117/ 117] Loss 0.291605 Top1 84.921351 Top5 98.336840 +2023-10-02 21:22:45,336 - ==> Top1: 84.921 Top5: 98.337 Loss: 0.292 + +2023-10-02 21:22:45,337 - ==> Confusion: +[[ 942 0 2 0 10 3 0 0 6 57 2 2 1 3 4 0 3 0 2 0 13] + [ 0 1049 0 1 4 33 1 24 0 0 2 0 3 0 1 3 1 0 4 2 3] + [ 3 0 978 7 4 1 20 9 0 0 2 0 11 2 0 3 1 1 9 3 2] + [ 2 3 14 977 0 6 4 2 1 0 3 0 9 2 28 1 0 9 16 0 12] + [ 29 8 1 0 963 4 0 1 1 7 0 0 2 1 13 3 9 0 2 2 4] + [ 3 23 0 1 1 1021 1 17 0 4 1 6 0 10 7 0 2 0 3 5 11] + [ 0 2 22 1 0 2 1132 5 0 0 5 0 1 1 0 4 0 0 1 11 4] + [ 0 16 10 1 1 32 6 1079 0 7 1 7 4 4 0 1 3 1 31 7 7] + [ 14 1 0 1 2 2 0 1 968 51 7 0 2 11 17 2 5 0 2 1 2] + [ 105 0 1 0 4 2 2 0 23 938 1 0 1 20 10 1 1 0 0 5 5] + [ 4 3 11 5 2 3 2 5 6 0 977 1 2 10 4 1 2 2 2 1 10] + [ 0 0 1 0 0 13 1 1 0 0 0 971 18 6 1 0 2 17 1 2 1] + [ 0 0 2 5 0 0 1 2 0 1 2 47 965 4 3 4 3 12 4 8 5] + [ 0 0 4 0 1 9 0 0 6 8 5 7 0 1059 5 1 1 0 1 2 10] + [ 12 2 3 15 4 0 0 0 14 2 4 0 2 2 1025 0 1 3 6 0 6] + [ 0 0 1 1 5 0 0 0 0 1 0 10 8 0 0 1069 17 12 1 4 5] + [ 2 15 1 0 3 10 0 0 0 0 0 6 2 2 5 6 1092 0 1 5 11] + [ 0 0 4 2 0 0 1 0 3 0 1 11 18 1 3 5 0 984 0 2 3] + [ 1 11 2 14 1 0 0 29 2 1 2 0 1 0 12 1 0 0 980 1 10] + [ 0 1 3 0 0 3 5 6 0 0 2 16 4 0 1 2 5 1 0 1097 6] + [ 141 173 154 75 84 181 41 95 72 87 165 125 369 301 149 60 93 56 131 191 5162]] + +2023-10-02 21:22:45,338 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:22:45,338 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:22:45,344 - + +2023-10-02 21:22:45,344 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:22:46,475 - Epoch: [112][ 10/ 1236] Overall Loss 0.213083 Objective Loss 0.213083 LR 0.000500 Time 0.113043 +2023-10-02 21:22:46,682 - Epoch: [112][ 20/ 1236] Overall Loss 0.192483 Objective Loss 0.192483 LR 0.000500 Time 0.066839 +2023-10-02 21:22:46,888 - Epoch: [112][ 30/ 1236] Overall Loss 0.191866 Objective Loss 0.191866 LR 0.000500 Time 0.051394 +2023-10-02 21:22:47,094 - Epoch: [112][ 40/ 1236] Overall Loss 0.199525 Objective Loss 0.199525 LR 0.000500 Time 0.043695 +2023-10-02 21:22:47,299 - Epoch: [112][ 50/ 1236] Overall Loss 0.200849 Objective Loss 0.200849 LR 0.000500 Time 0.039029 +2023-10-02 21:22:47,507 - Epoch: [112][ 60/ 1236] Overall Loss 0.198881 Objective Loss 0.198881 LR 0.000500 Time 0.035981 +2023-10-02 21:22:47,711 - Epoch: [112][ 70/ 1236] Overall Loss 0.200409 Objective Loss 0.200409 LR 0.000500 Time 0.033751 +2023-10-02 21:22:47,918 - Epoch: [112][ 80/ 1236] Overall Loss 0.200225 Objective Loss 0.200225 LR 0.000500 Time 0.032124 +2023-10-02 21:22:48,122 - Epoch: [112][ 90/ 1236] Overall Loss 0.203382 Objective Loss 0.203382 LR 0.000500 Time 0.030817 +2023-10-02 21:22:48,329 - Epoch: [112][ 100/ 1236] Overall Loss 0.205923 Objective Loss 0.205923 LR 0.000500 Time 0.029796 +2023-10-02 21:22:48,534 - Epoch: [112][ 110/ 1236] Overall Loss 0.205020 Objective Loss 0.205020 LR 0.000500 Time 0.028940 +2023-10-02 21:22:48,742 - Epoch: [112][ 120/ 1236] Overall Loss 0.204560 Objective Loss 0.204560 LR 0.000500 Time 0.028256 +2023-10-02 21:22:48,946 - Epoch: [112][ 130/ 1236] Overall Loss 0.204474 Objective Loss 0.204474 LR 0.000500 Time 0.027649 +2023-10-02 21:22:49,153 - Epoch: [112][ 140/ 1236] Overall Loss 0.204438 Objective Loss 0.204438 LR 0.000500 Time 0.027157 +2023-10-02 21:22:49,357 - Epoch: [112][ 150/ 1236] Overall Loss 0.203546 Objective Loss 0.203546 LR 0.000500 Time 0.026699 +2023-10-02 21:22:49,562 - Epoch: [112][ 160/ 1236] Overall Loss 0.202230 Objective Loss 0.202230 LR 0.000500 Time 0.026314 +2023-10-02 21:22:49,767 - Epoch: [112][ 170/ 1236] Overall Loss 0.200452 Objective Loss 0.200452 LR 0.000500 Time 0.025960 +2023-10-02 21:22:49,972 - Epoch: [112][ 180/ 1236] Overall Loss 0.200017 Objective Loss 0.200017 LR 0.000500 Time 0.025659 +2023-10-02 21:22:50,178 - Epoch: [112][ 190/ 1236] Overall Loss 0.200079 Objective Loss 0.200079 LR 0.000500 Time 0.025383 +2023-10-02 21:22:50,385 - Epoch: [112][ 200/ 1236] Overall Loss 0.200229 Objective Loss 0.200229 LR 0.000500 Time 0.025146 +2023-10-02 21:22:50,590 - Epoch: [112][ 210/ 1236] Overall Loss 0.201047 Objective Loss 0.201047 LR 0.000500 Time 0.024920 +2023-10-02 21:22:50,797 - Epoch: [112][ 220/ 1236] Overall Loss 0.200625 Objective Loss 0.200625 LR 0.000500 Time 0.024725 +2023-10-02 21:22:51,002 - Epoch: [112][ 230/ 1236] Overall Loss 0.200728 Objective Loss 0.200728 LR 0.000500 Time 0.024536 +2023-10-02 21:22:51,208 - Epoch: [112][ 240/ 1236] Overall Loss 0.200259 Objective Loss 0.200259 LR 0.000500 Time 0.024369 +2023-10-02 21:22:51,412 - Epoch: [112][ 250/ 1236] Overall Loss 0.200245 Objective Loss 0.200245 LR 0.000500 Time 0.024205 +2023-10-02 21:22:51,620 - Epoch: [112][ 260/ 1236] Overall Loss 0.199330 Objective Loss 0.199330 LR 0.000500 Time 0.024073 +2023-10-02 21:22:51,824 - Epoch: [112][ 270/ 1236] Overall Loss 0.199866 Objective Loss 0.199866 LR 0.000500 Time 0.023937 +2023-10-02 21:22:52,032 - Epoch: [112][ 280/ 1236] Overall Loss 0.199749 Objective Loss 0.199749 LR 0.000500 Time 0.023823 +2023-10-02 21:22:52,236 - Epoch: [112][ 290/ 1236] Overall Loss 0.199193 Objective Loss 0.199193 LR 0.000500 Time 0.023704 +2023-10-02 21:22:52,443 - Epoch: [112][ 300/ 1236] Overall Loss 0.199833 Objective Loss 0.199833 LR 0.000500 Time 0.023601 +2023-10-02 21:22:52,648 - Epoch: [112][ 310/ 1236] Overall Loss 0.200482 Objective Loss 0.200482 LR 0.000500 Time 0.023497 +2023-10-02 21:22:52,856 - Epoch: [112][ 320/ 1236] Overall Loss 0.200234 Objective Loss 0.200234 LR 0.000500 Time 0.023411 +2023-10-02 21:22:53,060 - Epoch: [112][ 330/ 1236] Overall Loss 0.200038 Objective Loss 0.200038 LR 0.000500 Time 0.023321 +2023-10-02 21:22:53,267 - Epoch: [112][ 340/ 1236] Overall Loss 0.199532 Objective Loss 0.199532 LR 0.000500 Time 0.023242 +2023-10-02 21:22:53,473 - Epoch: [112][ 350/ 1236] Overall Loss 0.199652 Objective Loss 0.199652 LR 0.000500 Time 0.023165 +2023-10-02 21:22:53,678 - Epoch: [112][ 360/ 1236] Overall Loss 0.199834 Objective Loss 0.199834 LR 0.000500 Time 0.023091 +2023-10-02 21:22:53,884 - Epoch: [112][ 370/ 1236] Overall Loss 0.199971 Objective Loss 0.199971 LR 0.000500 Time 0.023019 +2023-10-02 21:22:54,090 - Epoch: [112][ 380/ 1236] Overall Loss 0.200062 Objective Loss 0.200062 LR 0.000500 Time 0.022956 +2023-10-02 21:22:54,296 - Epoch: [112][ 390/ 1236] Overall Loss 0.199554 Objective Loss 0.199554 LR 0.000500 Time 0.022893 +2023-10-02 21:22:54,503 - Epoch: [112][ 400/ 1236] Overall Loss 0.200005 Objective Loss 0.200005 LR 0.000500 Time 0.022837 +2023-10-02 21:22:54,708 - Epoch: [112][ 410/ 1236] Overall Loss 0.199168 Objective Loss 0.199168 LR 0.000500 Time 0.022781 +2023-10-02 21:22:54,915 - Epoch: [112][ 420/ 1236] Overall Loss 0.198746 Objective Loss 0.198746 LR 0.000500 Time 0.022730 +2023-10-02 21:22:55,120 - Epoch: [112][ 430/ 1236] Overall Loss 0.198706 Objective Loss 0.198706 LR 0.000500 Time 0.022678 +2023-10-02 21:22:55,327 - Epoch: [112][ 440/ 1236] Overall Loss 0.198443 Objective Loss 0.198443 LR 0.000500 Time 0.022632 +2023-10-02 21:22:55,532 - Epoch: [112][ 450/ 1236] Overall Loss 0.198104 Objective Loss 0.198104 LR 0.000500 Time 0.022585 +2023-10-02 21:22:55,739 - Epoch: [112][ 460/ 1236] Overall Loss 0.197351 Objective Loss 0.197351 LR 0.000500 Time 0.022543 +2023-10-02 21:22:55,945 - Epoch: [112][ 470/ 1236] Overall Loss 0.197587 Objective Loss 0.197587 LR 0.000500 Time 0.022500 +2023-10-02 21:22:56,151 - Epoch: [112][ 480/ 1236] Overall Loss 0.197913 Objective Loss 0.197913 LR 0.000500 Time 0.022461 +2023-10-02 21:22:56,357 - Epoch: [112][ 490/ 1236] Overall Loss 0.197736 Objective Loss 0.197736 LR 0.000500 Time 0.022422 +2023-10-02 21:22:56,564 - Epoch: [112][ 500/ 1236] Overall Loss 0.197939 Objective Loss 0.197939 LR 0.000500 Time 0.022386 +2023-10-02 21:22:56,770 - Epoch: [112][ 510/ 1236] Overall Loss 0.198001 Objective Loss 0.198001 LR 0.000500 Time 0.022350 +2023-10-02 21:22:56,976 - Epoch: [112][ 520/ 1236] Overall Loss 0.198693 Objective Loss 0.198693 LR 0.000500 Time 0.022318 +2023-10-02 21:22:57,182 - Epoch: [112][ 530/ 1236] Overall Loss 0.198911 Objective Loss 0.198911 LR 0.000500 Time 0.022284 +2023-10-02 21:22:57,389 - Epoch: [112][ 540/ 1236] Overall Loss 0.198929 Objective Loss 0.198929 LR 0.000500 Time 0.022254 +2023-10-02 21:22:57,594 - Epoch: [112][ 550/ 1236] Overall Loss 0.198640 Objective Loss 0.198640 LR 0.000500 Time 0.022222 +2023-10-02 21:22:57,801 - Epoch: [112][ 560/ 1236] Overall Loss 0.198756 Objective Loss 0.198756 LR 0.000500 Time 0.022194 +2023-10-02 21:22:58,006 - Epoch: [112][ 570/ 1236] Overall Loss 0.198697 Objective Loss 0.198697 LR 0.000500 Time 0.022164 +2023-10-02 21:22:58,213 - Epoch: [112][ 580/ 1236] Overall Loss 0.198840 Objective Loss 0.198840 LR 0.000500 Time 0.022138 +2023-10-02 21:22:58,419 - Epoch: [112][ 590/ 1236] Overall Loss 0.199466 Objective Loss 0.199466 LR 0.000500 Time 0.022111 +2023-10-02 21:22:58,626 - Epoch: [112][ 600/ 1236] Overall Loss 0.199613 Objective Loss 0.199613 LR 0.000500 Time 0.022086 +2023-10-02 21:22:58,831 - Epoch: [112][ 610/ 1236] Overall Loss 0.199736 Objective Loss 0.199736 LR 0.000500 Time 0.022061 +2023-10-02 21:22:59,038 - Epoch: [112][ 620/ 1236] Overall Loss 0.199905 Objective Loss 0.199905 LR 0.000500 Time 0.022038 +2023-10-02 21:22:59,242 - Epoch: [112][ 630/ 1236] Overall Loss 0.199993 Objective Loss 0.199993 LR 0.000500 Time 0.022012 +2023-10-02 21:22:59,449 - Epoch: [112][ 640/ 1236] Overall Loss 0.199934 Objective Loss 0.199934 LR 0.000500 Time 0.021991 +2023-10-02 21:22:59,654 - Epoch: [112][ 650/ 1236] Overall Loss 0.199890 Objective Loss 0.199890 LR 0.000500 Time 0.021968 +2023-10-02 21:22:59,860 - Epoch: [112][ 660/ 1236] Overall Loss 0.199979 Objective Loss 0.199979 LR 0.000500 Time 0.021946 +2023-10-02 21:23:00,065 - Epoch: [112][ 670/ 1236] Overall Loss 0.200032 Objective Loss 0.200032 LR 0.000500 Time 0.021924 +2023-10-02 21:23:00,272 - Epoch: [112][ 680/ 1236] Overall Loss 0.200055 Objective Loss 0.200055 LR 0.000500 Time 0.021906 +2023-10-02 21:23:00,476 - Epoch: [112][ 690/ 1236] Overall Loss 0.199948 Objective Loss 0.199948 LR 0.000500 Time 0.021884 +2023-10-02 21:23:00,683 - Epoch: [112][ 700/ 1236] Overall Loss 0.199835 Objective Loss 0.199835 LR 0.000500 Time 0.021866 +2023-10-02 21:23:00,889 - Epoch: [112][ 710/ 1236] Overall Loss 0.199999 Objective Loss 0.199999 LR 0.000500 Time 0.021848 +2023-10-02 21:23:01,095 - Epoch: [112][ 720/ 1236] Overall Loss 0.199967 Objective Loss 0.199967 LR 0.000500 Time 0.021831 +2023-10-02 21:23:01,301 - Epoch: [112][ 730/ 1236] Overall Loss 0.200133 Objective Loss 0.200133 LR 0.000500 Time 0.021813 +2023-10-02 21:23:01,508 - Epoch: [112][ 740/ 1236] Overall Loss 0.200159 Objective Loss 0.200159 LR 0.000500 Time 0.021797 +2023-10-02 21:23:01,713 - Epoch: [112][ 750/ 1236] Overall Loss 0.200545 Objective Loss 0.200545 LR 0.000500 Time 0.021780 +2023-10-02 21:23:01,920 - Epoch: [112][ 760/ 1236] Overall Loss 0.200616 Objective Loss 0.200616 LR 0.000500 Time 0.021766 +2023-10-02 21:23:02,125 - Epoch: [112][ 770/ 1236] Overall Loss 0.200935 Objective Loss 0.200935 LR 0.000500 Time 0.021748 +2023-10-02 21:23:02,332 - Epoch: [112][ 780/ 1236] Overall Loss 0.201116 Objective Loss 0.201116 LR 0.000500 Time 0.021734 +2023-10-02 21:23:02,538 - Epoch: [112][ 790/ 1236] Overall Loss 0.201075 Objective Loss 0.201075 LR 0.000500 Time 0.021719 +2023-10-02 21:23:02,744 - Epoch: [112][ 800/ 1236] Overall Loss 0.201202 Objective Loss 0.201202 LR 0.000500 Time 0.021706 +2023-10-02 21:23:02,949 - Epoch: [112][ 810/ 1236] Overall Loss 0.201346 Objective Loss 0.201346 LR 0.000500 Time 0.021690 +2023-10-02 21:23:03,156 - Epoch: [112][ 820/ 1236] Overall Loss 0.201311 Objective Loss 0.201311 LR 0.000500 Time 0.021677 +2023-10-02 21:23:03,361 - Epoch: [112][ 830/ 1236] Overall Loss 0.201035 Objective Loss 0.201035 LR 0.000500 Time 0.021663 +2023-10-02 21:23:03,568 - Epoch: [112][ 840/ 1236] Overall Loss 0.200889 Objective Loss 0.200889 LR 0.000500 Time 0.021651 +2023-10-02 21:23:03,774 - Epoch: [112][ 850/ 1236] Overall Loss 0.200955 Objective Loss 0.200955 LR 0.000500 Time 0.021638 +2023-10-02 21:23:03,980 - Epoch: [112][ 860/ 1236] Overall Loss 0.201100 Objective Loss 0.201100 LR 0.000500 Time 0.021627 +2023-10-02 21:23:04,186 - Epoch: [112][ 870/ 1236] Overall Loss 0.201123 Objective Loss 0.201123 LR 0.000500 Time 0.021614 +2023-10-02 21:23:04,393 - Epoch: [112][ 880/ 1236] Overall Loss 0.201101 Objective Loss 0.201101 LR 0.000500 Time 0.021603 +2023-10-02 21:23:04,599 - Epoch: [112][ 890/ 1236] Overall Loss 0.201015 Objective Loss 0.201015 LR 0.000500 Time 0.021590 +2023-10-02 21:23:04,806 - Epoch: [112][ 900/ 1236] Overall Loss 0.200883 Objective Loss 0.200883 LR 0.000500 Time 0.021579 +2023-10-02 21:23:05,011 - Epoch: [112][ 910/ 1236] Overall Loss 0.201085 Objective Loss 0.201085 LR 0.000500 Time 0.021568 +2023-10-02 21:23:05,218 - Epoch: [112][ 920/ 1236] Overall Loss 0.200937 Objective Loss 0.200937 LR 0.000500 Time 0.021558 +2023-10-02 21:23:05,424 - Epoch: [112][ 930/ 1236] Overall Loss 0.200829 Objective Loss 0.200829 LR 0.000500 Time 0.021547 +2023-10-02 21:23:05,631 - Epoch: [112][ 940/ 1236] Overall Loss 0.200612 Objective Loss 0.200612 LR 0.000500 Time 0.021537 +2023-10-02 21:23:05,836 - Epoch: [112][ 950/ 1236] Overall Loss 0.200654 Objective Loss 0.200654 LR 0.000500 Time 0.021527 +2023-10-02 21:23:06,043 - Epoch: [112][ 960/ 1236] Overall Loss 0.200662 Objective Loss 0.200662 LR 0.000500 Time 0.021518 +2023-10-02 21:23:06,249 - Epoch: [112][ 970/ 1236] Overall Loss 0.200767 Objective Loss 0.200767 LR 0.000500 Time 0.021506 +2023-10-02 21:23:06,456 - Epoch: [112][ 980/ 1236] Overall Loss 0.201055 Objective Loss 0.201055 LR 0.000500 Time 0.021498 +2023-10-02 21:23:06,661 - Epoch: [112][ 990/ 1236] Overall Loss 0.201109 Objective Loss 0.201109 LR 0.000500 Time 0.021487 +2023-10-02 21:23:06,868 - Epoch: [112][ 1000/ 1236] Overall Loss 0.201176 Objective Loss 0.201176 LR 0.000500 Time 0.021478 +2023-10-02 21:23:07,073 - Epoch: [112][ 1010/ 1236] Overall Loss 0.201478 Objective Loss 0.201478 LR 0.000500 Time 0.021468 +2023-10-02 21:23:07,280 - Epoch: [112][ 1020/ 1236] Overall Loss 0.201705 Objective Loss 0.201705 LR 0.000500 Time 0.021461 +2023-10-02 21:23:07,485 - Epoch: [112][ 1030/ 1236] Overall Loss 0.201877 Objective Loss 0.201877 LR 0.000500 Time 0.021450 +2023-10-02 21:23:07,692 - Epoch: [112][ 1040/ 1236] Overall Loss 0.201865 Objective Loss 0.201865 LR 0.000500 Time 0.021443 +2023-10-02 21:23:07,898 - Epoch: [112][ 1050/ 1236] Overall Loss 0.201879 Objective Loss 0.201879 LR 0.000500 Time 0.021435 +2023-10-02 21:23:08,105 - Epoch: [112][ 1060/ 1236] Overall Loss 0.202064 Objective Loss 0.202064 LR 0.000500 Time 0.021427 +2023-10-02 21:23:08,310 - Epoch: [112][ 1070/ 1236] Overall Loss 0.202213 Objective Loss 0.202213 LR 0.000500 Time 0.021418 +2023-10-02 21:23:08,515 - Epoch: [112][ 1080/ 1236] Overall Loss 0.202161 Objective Loss 0.202161 LR 0.000500 Time 0.021409 +2023-10-02 21:23:08,721 - Epoch: [112][ 1090/ 1236] Overall Loss 0.202162 Objective Loss 0.202162 LR 0.000500 Time 0.021400 +2023-10-02 21:23:08,928 - Epoch: [112][ 1100/ 1236] Overall Loss 0.202237 Objective Loss 0.202237 LR 0.000500 Time 0.021393 +2023-10-02 21:23:09,134 - Epoch: [112][ 1110/ 1236] Overall Loss 0.202120 Objective Loss 0.202120 LR 0.000500 Time 0.021385 +2023-10-02 21:23:09,341 - Epoch: [112][ 1120/ 1236] Overall Loss 0.202119 Objective Loss 0.202119 LR 0.000500 Time 0.021378 +2023-10-02 21:23:09,547 - Epoch: [112][ 1130/ 1236] Overall Loss 0.202135 Objective Loss 0.202135 LR 0.000500 Time 0.021371 +2023-10-02 21:23:09,754 - Epoch: [112][ 1140/ 1236] Overall Loss 0.202351 Objective Loss 0.202351 LR 0.000500 Time 0.021364 +2023-10-02 21:23:09,959 - Epoch: [112][ 1150/ 1236] Overall Loss 0.202530 Objective Loss 0.202530 LR 0.000500 Time 0.021356 +2023-10-02 21:23:10,166 - Epoch: [112][ 1160/ 1236] Overall Loss 0.202791 Objective Loss 0.202791 LR 0.000500 Time 0.021350 +2023-10-02 21:23:10,372 - Epoch: [112][ 1170/ 1236] Overall Loss 0.202821 Objective Loss 0.202821 LR 0.000500 Time 0.021342 +2023-10-02 21:23:10,578 - Epoch: [112][ 1180/ 1236] Overall Loss 0.202782 Objective Loss 0.202782 LR 0.000500 Time 0.021336 +2023-10-02 21:23:10,784 - Epoch: [112][ 1190/ 1236] Overall Loss 0.202942 Objective Loss 0.202942 LR 0.000500 Time 0.021328 +2023-10-02 21:23:10,991 - Epoch: [112][ 1200/ 1236] Overall Loss 0.202904 Objective Loss 0.202904 LR 0.000500 Time 0.021323 +2023-10-02 21:23:11,197 - Epoch: [112][ 1210/ 1236] Overall Loss 0.202747 Objective Loss 0.202747 LR 0.000500 Time 0.021315 +2023-10-02 21:23:11,404 - Epoch: [112][ 1220/ 1236] Overall Loss 0.202696 Objective Loss 0.202696 LR 0.000500 Time 0.021310 +2023-10-02 21:23:11,663 - Epoch: [112][ 1230/ 1236] Overall Loss 0.202591 Objective Loss 0.202591 LR 0.000500 Time 0.021347 +2023-10-02 21:23:11,785 - Epoch: [112][ 1236/ 1236] Overall Loss 0.202635 Objective Loss 0.202635 Top1 89.613035 Top5 98.778004 LR 0.000500 Time 0.021342 +2023-10-02 21:23:11,927 - --- validate (epoch=112)----------- +2023-10-02 21:23:11,927 - 29943 samples (256 per mini-batch) +2023-10-02 21:23:12,418 - Epoch: [112][ 10/ 117] Loss 0.276640 Top1 85.429688 Top5 98.046875 +2023-10-02 21:23:12,570 - Epoch: [112][ 20/ 117] Loss 0.293249 Top1 84.902344 Top5 98.105469 +2023-10-02 21:23:12,721 - Epoch: [112][ 30/ 117] Loss 0.287416 Top1 84.882812 Top5 98.255208 +2023-10-02 21:23:12,872 - Epoch: [112][ 40/ 117] Loss 0.296873 Top1 84.619141 Top5 98.300781 +2023-10-02 21:23:13,024 - Epoch: [112][ 50/ 117] Loss 0.296617 Top1 84.648438 Top5 98.265625 +2023-10-02 21:23:13,174 - Epoch: [112][ 60/ 117] Loss 0.298047 Top1 84.433594 Top5 98.222656 +2023-10-02 21:23:13,326 - Epoch: [112][ 70/ 117] Loss 0.293923 Top1 84.481027 Top5 98.219866 +2023-10-02 21:23:13,476 - Epoch: [112][ 80/ 117] Loss 0.295963 Top1 84.404297 Top5 98.227539 +2023-10-02 21:23:13,625 - Epoch: [112][ 90/ 117] Loss 0.294943 Top1 84.440104 Top5 98.198785 +2023-10-02 21:23:13,776 - Epoch: [112][ 100/ 117] Loss 0.294530 Top1 84.410156 Top5 98.238281 +2023-10-02 21:23:13,934 - Epoch: [112][ 110/ 117] Loss 0.293724 Top1 84.467330 Top5 98.267045 +2023-10-02 21:23:14,023 - Epoch: [112][ 117/ 117] Loss 0.292555 Top1 84.497211 Top5 98.263367 +2023-10-02 21:23:14,174 - ==> Top1: 84.497 Top5: 98.263 Loss: 0.293 + +2023-10-02 21:23:14,175 - ==> Confusion: +[[ 934 1 4 1 10 3 0 0 10 56 2 2 0 3 3 1 4 0 1 0 15] + [ 0 1054 0 0 3 25 1 23 1 1 1 0 1 0 1 3 1 0 12 1 3] + [ 2 0 969 12 2 0 21 9 0 0 2 0 9 2 2 3 2 1 12 3 5] + [ 0 2 9 979 0 7 2 3 3 1 4 1 4 5 32 1 1 5 14 0 16] + [ 29 10 1 0 964 3 0 0 1 10 1 0 2 1 10 2 9 0 1 2 4] + [ 3 30 0 0 1 995 3 24 2 9 2 3 2 14 7 0 3 0 4 4 10] + [ 1 3 19 1 0 2 1145 4 0 0 2 0 0 0 0 5 0 1 1 5 2] + [ 1 12 6 2 1 27 7 1090 0 3 5 7 3 9 0 1 1 0 33 6 4] + [ 14 5 0 1 0 3 0 2 965 43 6 3 1 14 19 3 3 0 5 1 1] + [ 96 1 1 1 4 0 1 0 25 950 1 0 0 27 1 2 1 0 0 4 4] + [ 2 6 8 11 0 2 6 3 12 2 953 0 0 19 5 0 4 2 8 2 8] + [ 0 2 2 0 0 12 0 1 0 0 0 970 16 9 0 3 1 15 0 3 1] + [ 1 1 3 6 0 0 4 3 1 1 2 31 971 8 1 6 1 9 4 6 9] + [ 0 0 3 0 2 8 0 1 7 7 3 5 0 1066 6 1 1 0 0 1 8] + [ 10 0 3 20 9 0 0 0 18 5 2 0 4 2 1010 0 3 1 8 0 6] + [ 0 0 0 1 4 1 0 0 0 1 1 7 6 0 0 1070 18 13 3 6 3] + [ 1 20 0 0 4 12 0 0 0 0 0 2 0 3 4 8 1093 0 1 4 9] + [ 0 0 3 3 0 1 1 1 0 0 0 2 14 2 6 4 1 995 0 2 3] + [ 0 3 3 17 2 0 0 25 1 1 3 0 0 0 11 0 0 0 995 0 7] + [ 0 2 3 0 2 6 8 10 0 2 1 13 2 2 0 1 3 1 0 1094 2] + [ 104 232 123 80 81 173 52 111 84 87 162 110 370 348 133 57 125 66 156 212 5039]] + +2023-10-02 21:23:14,176 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:23:14,176 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:23:14,182 - + +2023-10-02 21:23:14,182 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:23:15,216 - Epoch: [113][ 10/ 1236] Overall Loss 0.178817 Objective Loss 0.178817 LR 0.000500 Time 0.103334 +2023-10-02 21:23:15,423 - Epoch: [113][ 20/ 1236] Overall Loss 0.188584 Objective Loss 0.188584 LR 0.000500 Time 0.061989 +2023-10-02 21:23:15,628 - Epoch: [113][ 30/ 1236] Overall Loss 0.194415 Objective Loss 0.194415 LR 0.000500 Time 0.048164 +2023-10-02 21:23:15,835 - Epoch: [113][ 40/ 1236] Overall Loss 0.196035 Objective Loss 0.196035 LR 0.000500 Time 0.041276 +2023-10-02 21:23:16,040 - Epoch: [113][ 50/ 1236] Overall Loss 0.198396 Objective Loss 0.198396 LR 0.000500 Time 0.037096 +2023-10-02 21:23:16,246 - Epoch: [113][ 60/ 1236] Overall Loss 0.196708 Objective Loss 0.196708 LR 0.000500 Time 0.034347 +2023-10-02 21:23:16,452 - Epoch: [113][ 70/ 1236] Overall Loss 0.197870 Objective Loss 0.197870 LR 0.000500 Time 0.032353 +2023-10-02 21:23:16,658 - Epoch: [113][ 80/ 1236] Overall Loss 0.198603 Objective Loss 0.198603 LR 0.000500 Time 0.030885 +2023-10-02 21:23:16,863 - Epoch: [113][ 90/ 1236] Overall Loss 0.198273 Objective Loss 0.198273 LR 0.000500 Time 0.029717 +2023-10-02 21:23:17,071 - Epoch: [113][ 100/ 1236] Overall Loss 0.200432 Objective Loss 0.200432 LR 0.000500 Time 0.028818 +2023-10-02 21:23:17,275 - Epoch: [113][ 110/ 1236] Overall Loss 0.201623 Objective Loss 0.201623 LR 0.000500 Time 0.028048 +2023-10-02 21:23:17,482 - Epoch: [113][ 120/ 1236] Overall Loss 0.200532 Objective Loss 0.200532 LR 0.000500 Time 0.027439 +2023-10-02 21:23:17,686 - Epoch: [113][ 130/ 1236] Overall Loss 0.202042 Objective Loss 0.202042 LR 0.000500 Time 0.026895 +2023-10-02 21:23:17,894 - Epoch: [113][ 140/ 1236] Overall Loss 0.200895 Objective Loss 0.200895 LR 0.000500 Time 0.026456 +2023-10-02 21:23:18,098 - Epoch: [113][ 150/ 1236] Overall Loss 0.200733 Objective Loss 0.200733 LR 0.000500 Time 0.026051 +2023-10-02 21:23:18,305 - Epoch: [113][ 160/ 1236] Overall Loss 0.200102 Objective Loss 0.200102 LR 0.000500 Time 0.025711 +2023-10-02 21:23:18,510 - Epoch: [113][ 170/ 1236] Overall Loss 0.200324 Objective Loss 0.200324 LR 0.000500 Time 0.025396 +2023-10-02 21:23:18,716 - Epoch: [113][ 180/ 1236] Overall Loss 0.200008 Objective Loss 0.200008 LR 0.000500 Time 0.025131 +2023-10-02 21:23:18,922 - Epoch: [113][ 190/ 1236] Overall Loss 0.200111 Objective Loss 0.200111 LR 0.000500 Time 0.024881 +2023-10-02 21:23:19,128 - Epoch: [113][ 200/ 1236] Overall Loss 0.198929 Objective Loss 0.198929 LR 0.000500 Time 0.024669 +2023-10-02 21:23:19,333 - Epoch: [113][ 210/ 1236] Overall Loss 0.199331 Objective Loss 0.199331 LR 0.000500 Time 0.024464 +2023-10-02 21:23:19,540 - Epoch: [113][ 220/ 1236] Overall Loss 0.198872 Objective Loss 0.198872 LR 0.000500 Time 0.024291 +2023-10-02 21:23:19,746 - Epoch: [113][ 230/ 1236] Overall Loss 0.199267 Objective Loss 0.199267 LR 0.000500 Time 0.024121 +2023-10-02 21:23:19,952 - Epoch: [113][ 240/ 1236] Overall Loss 0.199393 Objective Loss 0.199393 LR 0.000500 Time 0.023976 +2023-10-02 21:23:20,157 - Epoch: [113][ 250/ 1236] Overall Loss 0.199271 Objective Loss 0.199271 LR 0.000500 Time 0.023831 +2023-10-02 21:23:20,364 - Epoch: [113][ 260/ 1236] Overall Loss 0.199876 Objective Loss 0.199876 LR 0.000500 Time 0.023709 +2023-10-02 21:23:20,569 - Epoch: [113][ 270/ 1236] Overall Loss 0.200131 Objective Loss 0.200131 LR 0.000500 Time 0.023585 +2023-10-02 21:23:20,776 - Epoch: [113][ 280/ 1236] Overall Loss 0.199530 Objective Loss 0.199530 LR 0.000500 Time 0.023480 +2023-10-02 21:23:20,982 - Epoch: [113][ 290/ 1236] Overall Loss 0.199603 Objective Loss 0.199603 LR 0.000500 Time 0.023374 +2023-10-02 21:23:21,188 - Epoch: [113][ 300/ 1236] Overall Loss 0.198762 Objective Loss 0.198762 LR 0.000500 Time 0.023282 +2023-10-02 21:23:21,394 - Epoch: [113][ 310/ 1236] Overall Loss 0.199308 Objective Loss 0.199308 LR 0.000500 Time 0.023189 +2023-10-02 21:23:21,600 - Epoch: [113][ 320/ 1236] Overall Loss 0.199336 Objective Loss 0.199336 LR 0.000500 Time 0.023109 +2023-10-02 21:23:21,806 - Epoch: [113][ 330/ 1236] Overall Loss 0.199825 Objective Loss 0.199825 LR 0.000500 Time 0.023026 +2023-10-02 21:23:22,013 - Epoch: [113][ 340/ 1236] Overall Loss 0.199122 Objective Loss 0.199122 LR 0.000500 Time 0.022957 +2023-10-02 21:23:22,218 - Epoch: [113][ 350/ 1236] Overall Loss 0.199281 Objective Loss 0.199281 LR 0.000500 Time 0.022883 +2023-10-02 21:23:22,425 - Epoch: [113][ 360/ 1236] Overall Loss 0.198632 Objective Loss 0.198632 LR 0.000500 Time 0.022821 +2023-10-02 21:23:22,630 - Epoch: [113][ 370/ 1236] Overall Loss 0.198690 Objective Loss 0.198690 LR 0.000500 Time 0.022756 +2023-10-02 21:23:22,837 - Epoch: [113][ 380/ 1236] Overall Loss 0.199055 Objective Loss 0.199055 LR 0.000500 Time 0.022700 +2023-10-02 21:23:23,042 - Epoch: [113][ 390/ 1236] Overall Loss 0.198409 Objective Loss 0.198409 LR 0.000500 Time 0.022641 +2023-10-02 21:23:23,249 - Epoch: [113][ 400/ 1236] Overall Loss 0.199114 Objective Loss 0.199114 LR 0.000500 Time 0.022591 +2023-10-02 21:23:23,455 - Epoch: [113][ 410/ 1236] Overall Loss 0.199176 Objective Loss 0.199176 LR 0.000500 Time 0.022538 +2023-10-02 21:23:23,661 - Epoch: [113][ 420/ 1236] Overall Loss 0.198859 Objective Loss 0.198859 LR 0.000500 Time 0.022492 +2023-10-02 21:23:23,867 - Epoch: [113][ 430/ 1236] Overall Loss 0.199164 Objective Loss 0.199164 LR 0.000500 Time 0.022444 +2023-10-02 21:23:24,074 - Epoch: [113][ 440/ 1236] Overall Loss 0.198517 Objective Loss 0.198517 LR 0.000500 Time 0.022403 +2023-10-02 21:23:24,279 - Epoch: [113][ 450/ 1236] Overall Loss 0.199135 Objective Loss 0.199135 LR 0.000500 Time 0.022358 +2023-10-02 21:23:24,486 - Epoch: [113][ 460/ 1236] Overall Loss 0.198938 Objective Loss 0.198938 LR 0.000500 Time 0.022321 +2023-10-02 21:23:24,692 - Epoch: [113][ 470/ 1236] Overall Loss 0.198761 Objective Loss 0.198761 LR 0.000500 Time 0.022281 +2023-10-02 21:23:24,898 - Epoch: [113][ 480/ 1236] Overall Loss 0.198005 Objective Loss 0.198005 LR 0.000500 Time 0.022246 +2023-10-02 21:23:25,104 - Epoch: [113][ 490/ 1236] Overall Loss 0.198053 Objective Loss 0.198053 LR 0.000500 Time 0.022209 +2023-10-02 21:23:25,311 - Epoch: [113][ 500/ 1236] Overall Loss 0.198159 Objective Loss 0.198159 LR 0.000500 Time 0.022178 +2023-10-02 21:23:25,516 - Epoch: [113][ 510/ 1236] Overall Loss 0.198679 Objective Loss 0.198679 LR 0.000500 Time 0.022143 +2023-10-02 21:23:25,723 - Epoch: [113][ 520/ 1236] Overall Loss 0.198996 Objective Loss 0.198996 LR 0.000500 Time 0.022114 +2023-10-02 21:23:25,929 - Epoch: [113][ 530/ 1236] Overall Loss 0.199184 Objective Loss 0.199184 LR 0.000500 Time 0.022082 +2023-10-02 21:23:26,136 - Epoch: [113][ 540/ 1236] Overall Loss 0.199264 Objective Loss 0.199264 LR 0.000500 Time 0.022056 +2023-10-02 21:23:26,341 - Epoch: [113][ 550/ 1236] Overall Loss 0.199536 Objective Loss 0.199536 LR 0.000500 Time 0.022026 +2023-10-02 21:23:26,548 - Epoch: [113][ 560/ 1236] Overall Loss 0.199125 Objective Loss 0.199125 LR 0.000500 Time 0.022001 +2023-10-02 21:23:26,754 - Epoch: [113][ 570/ 1236] Overall Loss 0.198827 Objective Loss 0.198827 LR 0.000500 Time 0.021973 +2023-10-02 21:23:26,961 - Epoch: [113][ 580/ 1236] Overall Loss 0.198827 Objective Loss 0.198827 LR 0.000500 Time 0.021951 +2023-10-02 21:23:27,166 - Epoch: [113][ 590/ 1236] Overall Loss 0.198489 Objective Loss 0.198489 LR 0.000500 Time 0.021924 +2023-10-02 21:23:27,373 - Epoch: [113][ 600/ 1236] Overall Loss 0.198466 Objective Loss 0.198466 LR 0.000500 Time 0.021903 +2023-10-02 21:23:27,579 - Epoch: [113][ 610/ 1236] Overall Loss 0.198504 Objective Loss 0.198504 LR 0.000500 Time 0.021879 +2023-10-02 21:23:27,786 - Epoch: [113][ 620/ 1236] Overall Loss 0.198514 Objective Loss 0.198514 LR 0.000500 Time 0.021859 +2023-10-02 21:23:27,991 - Epoch: [113][ 630/ 1236] Overall Loss 0.198459 Objective Loss 0.198459 LR 0.000500 Time 0.021836 +2023-10-02 21:23:28,198 - Epoch: [113][ 640/ 1236] Overall Loss 0.198409 Objective Loss 0.198409 LR 0.000500 Time 0.021817 +2023-10-02 21:23:28,404 - Epoch: [113][ 650/ 1236] Overall Loss 0.198238 Objective Loss 0.198238 LR 0.000500 Time 0.021796 +2023-10-02 21:23:28,611 - Epoch: [113][ 660/ 1236] Overall Loss 0.198050 Objective Loss 0.198050 LR 0.000500 Time 0.021779 +2023-10-02 21:23:28,816 - Epoch: [113][ 670/ 1236] Overall Loss 0.197826 Objective Loss 0.197826 LR 0.000500 Time 0.021758 +2023-10-02 21:23:29,023 - Epoch: [113][ 680/ 1236] Overall Loss 0.197853 Objective Loss 0.197853 LR 0.000500 Time 0.021742 +2023-10-02 21:23:29,229 - Epoch: [113][ 690/ 1236] Overall Loss 0.197832 Objective Loss 0.197832 LR 0.000500 Time 0.021723 +2023-10-02 21:23:29,436 - Epoch: [113][ 700/ 1236] Overall Loss 0.197797 Objective Loss 0.197797 LR 0.000500 Time 0.021708 +2023-10-02 21:23:29,641 - Epoch: [113][ 710/ 1236] Overall Loss 0.197987 Objective Loss 0.197987 LR 0.000500 Time 0.021689 +2023-10-02 21:23:29,848 - Epoch: [113][ 720/ 1236] Overall Loss 0.197888 Objective Loss 0.197888 LR 0.000500 Time 0.021675 +2023-10-02 21:23:30,054 - Epoch: [113][ 730/ 1236] Overall Loss 0.198035 Objective Loss 0.198035 LR 0.000500 Time 0.021658 +2023-10-02 21:23:30,261 - Epoch: [113][ 740/ 1236] Overall Loss 0.198186 Objective Loss 0.198186 LR 0.000500 Time 0.021644 +2023-10-02 21:23:30,466 - Epoch: [113][ 750/ 1236] Overall Loss 0.198142 Objective Loss 0.198142 LR 0.000500 Time 0.021628 +2023-10-02 21:23:30,673 - Epoch: [113][ 760/ 1236] Overall Loss 0.198150 Objective Loss 0.198150 LR 0.000500 Time 0.021615 +2023-10-02 21:23:30,879 - Epoch: [113][ 770/ 1236] Overall Loss 0.198132 Objective Loss 0.198132 LR 0.000500 Time 0.021599 +2023-10-02 21:23:31,085 - Epoch: [113][ 780/ 1236] Overall Loss 0.198109 Objective Loss 0.198109 LR 0.000500 Time 0.021587 +2023-10-02 21:23:31,291 - Epoch: [113][ 790/ 1236] Overall Loss 0.197923 Objective Loss 0.197923 LR 0.000500 Time 0.021572 +2023-10-02 21:23:31,499 - Epoch: [113][ 800/ 1236] Overall Loss 0.197897 Objective Loss 0.197897 LR 0.000500 Time 0.021562 +2023-10-02 21:23:31,706 - Epoch: [113][ 810/ 1236] Overall Loss 0.198096 Objective Loss 0.198096 LR 0.000500 Time 0.021550 +2023-10-02 21:23:31,916 - Epoch: [113][ 820/ 1236] Overall Loss 0.198115 Objective Loss 0.198115 LR 0.000500 Time 0.021542 +2023-10-02 21:23:32,124 - Epoch: [113][ 830/ 1236] Overall Loss 0.198130 Objective Loss 0.198130 LR 0.000500 Time 0.021531 +2023-10-02 21:23:32,333 - Epoch: [113][ 840/ 1236] Overall Loss 0.198430 Objective Loss 0.198430 LR 0.000500 Time 0.021524 +2023-10-02 21:23:32,541 - Epoch: [113][ 850/ 1236] Overall Loss 0.198672 Objective Loss 0.198672 LR 0.000500 Time 0.021513 +2023-10-02 21:23:32,750 - Epoch: [113][ 860/ 1236] Overall Loss 0.198813 Objective Loss 0.198813 LR 0.000500 Time 0.021506 +2023-10-02 21:23:32,958 - Epoch: [113][ 870/ 1236] Overall Loss 0.199090 Objective Loss 0.199090 LR 0.000500 Time 0.021496 +2023-10-02 21:23:33,168 - Epoch: [113][ 880/ 1236] Overall Loss 0.199086 Objective Loss 0.199086 LR 0.000500 Time 0.021489 +2023-10-02 21:23:33,376 - Epoch: [113][ 890/ 1236] Overall Loss 0.198769 Objective Loss 0.198769 LR 0.000500 Time 0.021480 +2023-10-02 21:23:33,585 - Epoch: [113][ 900/ 1236] Overall Loss 0.198499 Objective Loss 0.198499 LR 0.000500 Time 0.021474 +2023-10-02 21:23:33,793 - Epoch: [113][ 910/ 1236] Overall Loss 0.198578 Objective Loss 0.198578 LR 0.000500 Time 0.021464 +2023-10-02 21:23:34,002 - Epoch: [113][ 920/ 1236] Overall Loss 0.198609 Objective Loss 0.198609 LR 0.000500 Time 0.021459 +2023-10-02 21:23:34,210 - Epoch: [113][ 930/ 1236] Overall Loss 0.198793 Objective Loss 0.198793 LR 0.000500 Time 0.021450 +2023-10-02 21:23:34,420 - Epoch: [113][ 940/ 1236] Overall Loss 0.198946 Objective Loss 0.198946 LR 0.000500 Time 0.021444 +2023-10-02 21:23:34,628 - Epoch: [113][ 950/ 1236] Overall Loss 0.199228 Objective Loss 0.199228 LR 0.000500 Time 0.021436 +2023-10-02 21:23:34,837 - Epoch: [113][ 960/ 1236] Overall Loss 0.199024 Objective Loss 0.199024 LR 0.000500 Time 0.021430 +2023-10-02 21:23:35,045 - Epoch: [113][ 970/ 1236] Overall Loss 0.199109 Objective Loss 0.199109 LR 0.000500 Time 0.021422 +2023-10-02 21:23:35,255 - Epoch: [113][ 980/ 1236] Overall Loss 0.199073 Objective Loss 0.199073 LR 0.000500 Time 0.021418 +2023-10-02 21:23:35,463 - Epoch: [113][ 990/ 1236] Overall Loss 0.199023 Objective Loss 0.199023 LR 0.000500 Time 0.021410 +2023-10-02 21:23:35,673 - Epoch: [113][ 1000/ 1236] Overall Loss 0.198829 Objective Loss 0.198829 LR 0.000500 Time 0.021405 +2023-10-02 21:23:35,881 - Epoch: [113][ 1010/ 1236] Overall Loss 0.198925 Objective Loss 0.198925 LR 0.000500 Time 0.021398 +2023-10-02 21:23:36,091 - Epoch: [113][ 1020/ 1236] Overall Loss 0.198956 Objective Loss 0.198956 LR 0.000500 Time 0.021393 +2023-10-02 21:23:36,299 - Epoch: [113][ 1030/ 1236] Overall Loss 0.199058 Objective Loss 0.199058 LR 0.000500 Time 0.021386 +2023-10-02 21:23:36,509 - Epoch: [113][ 1040/ 1236] Overall Loss 0.199251 Objective Loss 0.199251 LR 0.000500 Time 0.021382 +2023-10-02 21:23:36,716 - Epoch: [113][ 1050/ 1236] Overall Loss 0.199089 Objective Loss 0.199089 LR 0.000500 Time 0.021374 +2023-10-02 21:23:36,926 - Epoch: [113][ 1060/ 1236] Overall Loss 0.199061 Objective Loss 0.199061 LR 0.000500 Time 0.021370 +2023-10-02 21:23:37,134 - Epoch: [113][ 1070/ 1236] Overall Loss 0.198881 Objective Loss 0.198881 LR 0.000500 Time 0.021363 +2023-10-02 21:23:37,344 - Epoch: [113][ 1080/ 1236] Overall Loss 0.198770 Objective Loss 0.198770 LR 0.000500 Time 0.021360 +2023-10-02 21:23:37,552 - Epoch: [113][ 1090/ 1236] Overall Loss 0.198410 Objective Loss 0.198410 LR 0.000500 Time 0.021353 +2023-10-02 21:23:37,761 - Epoch: [113][ 1100/ 1236] Overall Loss 0.198555 Objective Loss 0.198555 LR 0.000500 Time 0.021349 +2023-10-02 21:23:37,970 - Epoch: [113][ 1110/ 1236] Overall Loss 0.198488 Objective Loss 0.198488 LR 0.000500 Time 0.021344 +2023-10-02 21:23:38,179 - Epoch: [113][ 1120/ 1236] Overall Loss 0.198612 Objective Loss 0.198612 LR 0.000500 Time 0.021340 +2023-10-02 21:23:38,388 - Epoch: [113][ 1130/ 1236] Overall Loss 0.198701 Objective Loss 0.198701 LR 0.000500 Time 0.021334 +2023-10-02 21:23:38,598 - Epoch: [113][ 1140/ 1236] Overall Loss 0.198501 Objective Loss 0.198501 LR 0.000500 Time 0.021331 +2023-10-02 21:23:38,806 - Epoch: [113][ 1150/ 1236] Overall Loss 0.198494 Objective Loss 0.198494 LR 0.000500 Time 0.021325 +2023-10-02 21:23:39,015 - Epoch: [113][ 1160/ 1236] Overall Loss 0.198579 Objective Loss 0.198579 LR 0.000500 Time 0.021322 +2023-10-02 21:23:39,223 - Epoch: [113][ 1170/ 1236] Overall Loss 0.198590 Objective Loss 0.198590 LR 0.000500 Time 0.021316 +2023-10-02 21:23:39,433 - Epoch: [113][ 1180/ 1236] Overall Loss 0.198807 Objective Loss 0.198807 LR 0.000500 Time 0.021313 +2023-10-02 21:23:39,641 - Epoch: [113][ 1190/ 1236] Overall Loss 0.199002 Objective Loss 0.199002 LR 0.000500 Time 0.021307 +2023-10-02 21:23:39,851 - Epoch: [113][ 1200/ 1236] Overall Loss 0.198847 Objective Loss 0.198847 LR 0.000500 Time 0.021304 +2023-10-02 21:23:40,059 - Epoch: [113][ 1210/ 1236] Overall Loss 0.198897 Objective Loss 0.198897 LR 0.000500 Time 0.021299 +2023-10-02 21:23:40,268 - Epoch: [113][ 1220/ 1236] Overall Loss 0.198980 Objective Loss 0.198980 LR 0.000500 Time 0.021296 +2023-10-02 21:23:40,529 - Epoch: [113][ 1230/ 1236] Overall Loss 0.198966 Objective Loss 0.198966 LR 0.000500 Time 0.021333 +2023-10-02 21:23:40,651 - Epoch: [113][ 1236/ 1236] Overall Loss 0.198879 Objective Loss 0.198879 Top1 88.798371 Top5 98.167006 LR 0.000500 Time 0.021328 +2023-10-02 21:23:40,782 - --- validate (epoch=113)----------- +2023-10-02 21:23:40,783 - 29943 samples (256 per mini-batch) +2023-10-02 21:23:41,281 - Epoch: [113][ 10/ 117] Loss 0.318500 Top1 84.492188 Top5 98.515625 +2023-10-02 21:23:41,439 - Epoch: [113][ 20/ 117] Loss 0.308783 Top1 84.785156 Top5 98.437500 +2023-10-02 21:23:41,597 - Epoch: [113][ 30/ 117] Loss 0.290398 Top1 85.182292 Top5 98.450521 +2023-10-02 21:23:41,755 - Epoch: [113][ 40/ 117] Loss 0.292638 Top1 85.312500 Top5 98.427734 +2023-10-02 21:23:41,912 - Epoch: [113][ 50/ 117] Loss 0.288962 Top1 85.429688 Top5 98.406250 +2023-10-02 21:23:42,070 - Epoch: [113][ 60/ 117] Loss 0.289287 Top1 85.559896 Top5 98.398438 +2023-10-02 21:23:42,227 - Epoch: [113][ 70/ 117] Loss 0.289745 Top1 85.418527 Top5 98.409598 +2023-10-02 21:23:42,385 - Epoch: [113][ 80/ 117] Loss 0.292575 Top1 85.366211 Top5 98.393555 +2023-10-02 21:23:42,542 - Epoch: [113][ 90/ 117] Loss 0.288815 Top1 85.286458 Top5 98.385417 +2023-10-02 21:23:42,700 - Epoch: [113][ 100/ 117] Loss 0.289253 Top1 85.394531 Top5 98.382812 +2023-10-02 21:23:42,864 - Epoch: [113][ 110/ 117] Loss 0.289690 Top1 85.287642 Top5 98.373580 +2023-10-02 21:23:42,954 - Epoch: [113][ 117/ 117] Loss 0.291270 Top1 85.302074 Top5 98.360218 +2023-10-02 21:23:43,051 - ==> Top1: 85.302 Top5: 98.360 Loss: 0.291 + +2023-10-02 21:23:43,052 - ==> Confusion: +[[ 951 2 3 0 8 3 0 0 5 47 1 2 1 2 2 0 4 0 2 0 17] + [ 0 1051 0 0 1 33 2 22 0 0 2 0 2 0 1 3 1 0 7 1 5] + [ 3 0 971 8 1 0 17 4 0 1 2 0 9 3 2 4 2 1 16 2 10] + [ 2 2 10 967 0 3 0 0 8 0 8 1 6 5 22 3 1 6 25 0 20] + [ 28 12 0 0 961 5 0 0 1 9 1 0 2 2 7 5 8 0 2 1 6] + [ 4 26 0 0 0 1009 1 26 3 5 2 4 3 8 3 0 4 0 7 0 11] + [ 1 3 26 1 0 1 1125 8 0 0 5 1 0 0 0 7 0 0 2 6 5] + [ 2 10 10 1 2 24 5 1091 0 3 4 4 3 3 0 0 2 0 41 8 5] + [ 15 2 0 1 1 3 0 0 973 45 9 4 3 12 11 0 1 1 4 3 1] + [ 110 2 1 0 2 3 0 0 28 933 0 0 0 21 8 1 2 0 0 3 5] + [ 2 5 8 4 0 3 1 4 14 1 966 0 0 12 4 1 3 1 10 2 12] + [ 0 0 0 0 0 10 0 1 0 1 0 974 22 7 1 0 1 14 1 1 2] + [ 0 0 4 4 0 0 1 2 0 1 2 28 981 1 2 7 1 15 5 5 9] + [ 0 0 2 0 1 5 1 0 12 8 5 9 0 1061 5 1 0 0 0 1 8] + [ 11 1 4 14 7 0 0 0 18 3 1 1 3 3 1012 0 1 2 14 0 6] + [ 0 0 1 1 4 0 0 0 0 0 1 5 7 0 0 1077 18 9 3 3 5] + [ 3 16 0 2 3 10 0 1 1 1 0 3 0 1 1 8 1095 0 1 4 11] + [ 0 0 0 6 0 1 3 0 0 0 0 1 22 0 4 8 1 990 0 0 2] + [ 1 2 2 8 1 0 0 25 4 0 1 0 1 0 7 0 0 0 1007 0 9] + [ 0 1 2 2 0 6 10 10 0 1 1 15 4 0 1 1 14 1 0 1073 10] + [ 128 173 113 69 65 191 39 98 82 66 170 100 383 276 129 59 101 57 150 182 5274]] + +2023-10-02 21:23:43,053 - ==> Best [Top1: 85.376 Top5: 98.320 Sparsity:0.00 Params: 169472 on epoch: 105] +2023-10-02 21:23:43,053 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:23:43,059 - + +2023-10-02 21:23:43,059 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:23:44,096 - Epoch: [114][ 10/ 1236] Overall Loss 0.195473 Objective Loss 0.195473 LR 0.000500 Time 0.103583 +2023-10-02 21:23:44,303 - Epoch: [114][ 20/ 1236] Overall Loss 0.196891 Objective Loss 0.196891 LR 0.000500 Time 0.062157 +2023-10-02 21:23:44,510 - Epoch: [114][ 30/ 1236] Overall Loss 0.188314 Objective Loss 0.188314 LR 0.000500 Time 0.048321 +2023-10-02 21:23:44,718 - Epoch: [114][ 40/ 1236] Overall Loss 0.189630 Objective Loss 0.189630 LR 0.000500 Time 0.041430 +2023-10-02 21:23:44,925 - Epoch: [114][ 50/ 1236] Overall Loss 0.190678 Objective Loss 0.190678 LR 0.000500 Time 0.037244 +2023-10-02 21:23:45,132 - Epoch: [114][ 60/ 1236] Overall Loss 0.190760 Objective Loss 0.190760 LR 0.000500 Time 0.034493 +2023-10-02 21:23:45,339 - Epoch: [114][ 70/ 1236] Overall Loss 0.191672 Objective Loss 0.191672 LR 0.000500 Time 0.032494 +2023-10-02 21:23:45,546 - Epoch: [114][ 80/ 1236] Overall Loss 0.191204 Objective Loss 0.191204 LR 0.000500 Time 0.031022 +2023-10-02 21:23:45,753 - Epoch: [114][ 90/ 1236] Overall Loss 0.190858 Objective Loss 0.190858 LR 0.000500 Time 0.029866 +2023-10-02 21:23:45,960 - Epoch: [114][ 100/ 1236] Overall Loss 0.191719 Objective Loss 0.191719 LR 0.000500 Time 0.028951 +2023-10-02 21:23:46,166 - Epoch: [114][ 110/ 1236] Overall Loss 0.190498 Objective Loss 0.190498 LR 0.000500 Time 0.028176 +2023-10-02 21:23:46,375 - Epoch: [114][ 120/ 1236] Overall Loss 0.190387 Objective Loss 0.190387 LR 0.000500 Time 0.027567 +2023-10-02 21:23:46,580 - Epoch: [114][ 130/ 1236] Overall Loss 0.191823 Objective Loss 0.191823 LR 0.000500 Time 0.027023 +2023-10-02 21:23:46,789 - Epoch: [114][ 140/ 1236] Overall Loss 0.192662 Objective Loss 0.192662 LR 0.000500 Time 0.026584 +2023-10-02 21:23:46,995 - Epoch: [114][ 150/ 1236] Overall Loss 0.193564 Objective Loss 0.193564 LR 0.000500 Time 0.026178 +2023-10-02 21:23:47,203 - Epoch: [114][ 160/ 1236] Overall Loss 0.192274 Objective Loss 0.192274 LR 0.000500 Time 0.025845 +2023-10-02 21:23:47,409 - Epoch: [114][ 170/ 1236] Overall Loss 0.193277 Objective Loss 0.193277 LR 0.000500 Time 0.025532 +2023-10-02 21:23:47,618 - Epoch: [114][ 180/ 1236] Overall Loss 0.192854 Objective Loss 0.192854 LR 0.000500 Time 0.025273 +2023-10-02 21:23:47,823 - Epoch: [114][ 190/ 1236] Overall Loss 0.193211 Objective Loss 0.193211 LR 0.000500 Time 0.025019 +2023-10-02 21:23:48,031 - Epoch: [114][ 200/ 1236] Overall Loss 0.193367 Objective Loss 0.193367 LR 0.000500 Time 0.024808 +2023-10-02 21:23:48,238 - Epoch: [114][ 210/ 1236] Overall Loss 0.194215 Objective Loss 0.194215 LR 0.000500 Time 0.024609 +2023-10-02 21:23:48,445 - Epoch: [114][ 220/ 1236] Overall Loss 0.194300 Objective Loss 0.194300 LR 0.000500 Time 0.024433 +2023-10-02 21:23:48,655 - Epoch: [114][ 230/ 1236] Overall Loss 0.194727 Objective Loss 0.194727 LR 0.000500 Time 0.024282 +2023-10-02 21:23:48,865 - Epoch: [114][ 240/ 1236] Overall Loss 0.194326 Objective Loss 0.194326 LR 0.000500 Time 0.024145 +2023-10-02 21:23:49,080 - Epoch: [114][ 250/ 1236] Overall Loss 0.193961 Objective Loss 0.193961 LR 0.000500 Time 0.024034 +2023-10-02 21:23:49,292 - Epoch: [114][ 260/ 1236] Overall Loss 0.193915 Objective Loss 0.193915 LR 0.000500 Time 0.023927 +2023-10-02 21:23:49,507 - Epoch: [114][ 270/ 1236] Overall Loss 0.194599 Objective Loss 0.194599 LR 0.000500 Time 0.023834 +2023-10-02 21:23:49,719 - Epoch: [114][ 280/ 1236] Overall Loss 0.193996 Objective Loss 0.193996 LR 0.000500 Time 0.023740 +2023-10-02 21:23:49,934 - Epoch: [114][ 290/ 1236] Overall Loss 0.193805 Objective Loss 0.193805 LR 0.000500 Time 0.023660 +2023-10-02 21:23:50,146 - Epoch: [114][ 300/ 1236] Overall Loss 0.193754 Objective Loss 0.193754 LR 0.000500 Time 0.023579 +2023-10-02 21:23:50,360 - Epoch: [114][ 310/ 1236] Overall Loss 0.193232 Objective Loss 0.193232 LR 0.000500 Time 0.023508 +2023-10-02 21:23:50,571 - Epoch: [114][ 320/ 1236] Overall Loss 0.193403 Objective Loss 0.193403 LR 0.000500 Time 0.023431 +2023-10-02 21:23:50,783 - Epoch: [114][ 330/ 1236] Overall Loss 0.192904 Objective Loss 0.192904 LR 0.000500 Time 0.023362 +2023-10-02 21:23:50,993 - Epoch: [114][ 340/ 1236] Overall Loss 0.193285 Objective Loss 0.193285 LR 0.000500 Time 0.023293 +2023-10-02 21:23:51,205 - Epoch: [114][ 350/ 1236] Overall Loss 0.192730 Objective Loss 0.192730 LR 0.000500 Time 0.023232 +2023-10-02 21:23:51,416 - Epoch: [114][ 360/ 1236] Overall Loss 0.192537 Objective Loss 0.192537 LR 0.000500 Time 0.023171 +2023-10-02 21:23:51,626 - Epoch: [114][ 370/ 1236] Overall Loss 0.192474 Objective Loss 0.192474 LR 0.000500 Time 0.023112 +2023-10-02 21:23:51,837 - Epoch: [114][ 380/ 1236] Overall Loss 0.192451 Objective Loss 0.192451 LR 0.000500 Time 0.023059 +2023-10-02 21:23:52,044 - Epoch: [114][ 390/ 1236] Overall Loss 0.192806 Objective Loss 0.192806 LR 0.000500 Time 0.022997 +2023-10-02 21:23:52,255 - Epoch: [114][ 400/ 1236] Overall Loss 0.193042 Objective Loss 0.193042 LR 0.000500 Time 0.022948 +2023-10-02 21:23:52,462 - Epoch: [114][ 410/ 1236] Overall Loss 0.193198 Objective Loss 0.193198 LR 0.000500 Time 0.022892 +2023-10-02 21:23:52,673 - Epoch: [114][ 420/ 1236] Overall Loss 0.193078 Objective Loss 0.193078 LR 0.000500 Time 0.022848 +2023-10-02 21:23:52,879 - Epoch: [114][ 430/ 1236] Overall Loss 0.192934 Objective Loss 0.192934 LR 0.000500 Time 0.022797 +2023-10-02 21:23:53,089 - Epoch: [114][ 440/ 1236] Overall Loss 0.192701 Objective Loss 0.192701 LR 0.000500 Time 0.022754 +2023-10-02 21:23:53,297 - Epoch: [114][ 450/ 1236] Overall Loss 0.192756 Objective Loss 0.192756 LR 0.000500 Time 0.022707 +2023-10-02 21:23:53,508 - Epoch: [114][ 460/ 1236] Overall Loss 0.192536 Objective Loss 0.192536 LR 0.000500 Time 0.022672 +2023-10-02 21:23:53,715 - Epoch: [114][ 470/ 1236] Overall Loss 0.192730 Objective Loss 0.192730 LR 0.000500 Time 0.022629 +2023-10-02 21:23:53,925 - Epoch: [114][ 480/ 1236] Overall Loss 0.192540 Objective Loss 0.192540 LR 0.000500 Time 0.022595 +2023-10-02 21:23:54,133 - Epoch: [114][ 490/ 1236] Overall Loss 0.192595 Objective Loss 0.192595 LR 0.000500 Time 0.022555 +2023-10-02 21:23:54,342 - Epoch: [114][ 500/ 1236] Overall Loss 0.192879 Objective Loss 0.192879 LR 0.000500 Time 0.022523 +2023-10-02 21:23:54,550 - Epoch: [114][ 510/ 1236] Overall Loss 0.193533 Objective Loss 0.193533 LR 0.000500 Time 0.022486 +2023-10-02 21:23:54,761 - Epoch: [114][ 520/ 1236] Overall Loss 0.193486 Objective Loss 0.193486 LR 0.000500 Time 0.022459 +2023-10-02 21:23:54,968 - Epoch: [114][ 530/ 1236] Overall Loss 0.193968 Objective Loss 0.193968 LR 0.000500 Time 0.022425 +2023-10-02 21:23:55,180 - Epoch: [114][ 540/ 1236] Overall Loss 0.193806 Objective Loss 0.193806 LR 0.000500 Time 0.022400 +2023-10-02 21:23:55,386 - Epoch: [114][ 550/ 1236] Overall Loss 0.194061 Objective Loss 0.194061 LR 0.000500 Time 0.022369 +2023-10-02 21:23:55,596 - Epoch: [114][ 560/ 1236] Overall Loss 0.193913 Objective Loss 0.193913 LR 0.000500 Time 0.022343 +2023-10-02 21:23:55,804 - Epoch: [114][ 570/ 1236] Overall Loss 0.193884 Objective Loss 0.193884 LR 0.000500 Time 0.022313 +2023-10-02 21:23:56,015 - Epoch: [114][ 580/ 1236] Overall Loss 0.194218 Objective Loss 0.194218 LR 0.000500 Time 0.022291 +2023-10-02 21:23:56,221 - Epoch: [114][ 590/ 1236] Overall Loss 0.194394 Objective Loss 0.194394 LR 0.000500 Time 0.022263 +2023-10-02 21:23:56,433 - Epoch: [114][ 600/ 1236] Overall Loss 0.194867 Objective Loss 0.194867 LR 0.000500 Time 0.022244 +2023-10-02 21:23:56,639 - Epoch: [114][ 610/ 1236] Overall Loss 0.194954 Objective Loss 0.194954 LR 0.000500 Time 0.022217 +2023-10-02 21:23:56,850 - Epoch: [114][ 620/ 1236] Overall Loss 0.195111 Objective Loss 0.195111 LR 0.000500 Time 0.022199 +2023-10-02 21:23:57,057 - Epoch: [114][ 630/ 1236] Overall Loss 0.195058 Objective Loss 0.195058 LR 0.000500 Time 0.022174 +2023-10-02 21:23:57,268 - Epoch: [114][ 640/ 1236] Overall Loss 0.194891 Objective Loss 0.194891 LR 0.000500 Time 0.022157 +2023-10-02 21:23:57,475 - Epoch: [114][ 650/ 1236] Overall Loss 0.194791 Objective Loss 0.194791 LR 0.000500 Time 0.022134 +2023-10-02 21:23:57,686 - Epoch: [114][ 660/ 1236] Overall Loss 0.195204 Objective Loss 0.195204 LR 0.000500 Time 0.022118 +2023-10-02 21:23:57,892 - Epoch: [114][ 670/ 1236] Overall Loss 0.195767 Objective Loss 0.195767 LR 0.000500 Time 0.022096 +2023-10-02 21:23:58,102 - Epoch: [114][ 680/ 1236] Overall Loss 0.195724 Objective Loss 0.195724 LR 0.000500 Time 0.022078 +2023-10-02 21:23:58,310 - Epoch: [114][ 690/ 1236] Overall Loss 0.195815 Objective Loss 0.195815 LR 0.000500 Time 0.022057 +2023-10-02 21:23:58,520 - Epoch: [114][ 700/ 1236] Overall Loss 0.195636 Objective Loss 0.195636 LR 0.000500 Time 0.022042 +2023-10-02 21:23:58,727 - Epoch: [114][ 710/ 1236] Overall Loss 0.196177 Objective Loss 0.196177 LR 0.000500 Time 0.022022 +2023-10-02 21:23:58,939 - Epoch: [114][ 720/ 1236] Overall Loss 0.196331 Objective Loss 0.196331 LR 0.000500 Time 0.022009 +2023-10-02 21:23:59,146 - Epoch: [114][ 730/ 1236] Overall Loss 0.196498 Objective Loss 0.196498 LR 0.000500 Time 0.021990 +2023-10-02 21:23:59,356 - Epoch: [114][ 740/ 1236] Overall Loss 0.196362 Objective Loss 0.196362 LR 0.000500 Time 0.021978 +2023-10-02 21:23:59,563 - Epoch: [114][ 750/ 1236] Overall Loss 0.196229 Objective Loss 0.196229 LR 0.000500 Time 0.021960 +2023-10-02 21:23:59,773 - Epoch: [114][ 760/ 1236] Overall Loss 0.196448 Objective Loss 0.196448 LR 0.000500 Time 0.021947 +2023-10-02 21:23:59,981 - Epoch: [114][ 770/ 1236] Overall Loss 0.196478 Objective Loss 0.196478 LR 0.000500 Time 0.021930 +2023-10-02 21:24:00,192 - Epoch: [114][ 780/ 1236] Overall Loss 0.196911 Objective Loss 0.196911 LR 0.000500 Time 0.021919 +2023-10-02 21:24:00,399 - Epoch: [114][ 790/ 1236] Overall Loss 0.197104 Objective Loss 0.197104 LR 0.000500 Time 0.021903 +2023-10-02 21:24:00,610 - Epoch: [114][ 800/ 1236] Overall Loss 0.197518 Objective Loss 0.197518 LR 0.000500 Time 0.021892 +2023-10-02 21:24:00,816 - Epoch: [114][ 810/ 1236] Overall Loss 0.197433 Objective Loss 0.197433 LR 0.000500 Time 0.021877 +2023-10-02 21:24:01,027 - Epoch: [114][ 820/ 1236] Overall Loss 0.197548 Objective Loss 0.197548 LR 0.000500 Time 0.021867 +2023-10-02 21:24:01,234 - Epoch: [114][ 830/ 1236] Overall Loss 0.197525 Objective Loss 0.197525 LR 0.000500 Time 0.021853 +2023-10-02 21:24:01,444 - Epoch: [114][ 840/ 1236] Overall Loss 0.197627 Objective Loss 0.197627 LR 0.000500 Time 0.021842 +2023-10-02 21:24:01,652 - Epoch: [114][ 850/ 1236] Overall Loss 0.197902 Objective Loss 0.197902 LR 0.000500 Time 0.021828 +2023-10-02 21:24:01,863 - Epoch: [114][ 860/ 1236] Overall Loss 0.197891 Objective Loss 0.197891 LR 0.000500 Time 0.021819 +2023-10-02 21:24:02,070 - Epoch: [114][ 870/ 1236] Overall Loss 0.197809 Objective Loss 0.197809 LR 0.000500 Time 0.021806 +2023-10-02 21:24:02,281 - Epoch: [114][ 880/ 1236] Overall Loss 0.197852 Objective Loss 0.197852 LR 0.000500 Time 0.021797 +2023-10-02 21:24:02,488 - Epoch: [114][ 890/ 1236] Overall Loss 0.197829 Objective Loss 0.197829 LR 0.000500 Time 0.021785 +2023-10-02 21:24:02,698 - Epoch: [114][ 900/ 1236] Overall Loss 0.197647 Objective Loss 0.197647 LR 0.000500 Time 0.021775 +2023-10-02 21:24:02,906 - Epoch: [114][ 910/ 1236] Overall Loss 0.197865 Objective Loss 0.197865 LR 0.000500 Time 0.021763 +2023-10-02 21:24:03,117 - Epoch: [114][ 920/ 1236] Overall Loss 0.197829 Objective Loss 0.197829 LR 0.000500 Time 0.021756 +2023-10-02 21:24:03,323 - Epoch: [114][ 930/ 1236] Overall Loss 0.197763 Objective Loss 0.197763 LR 0.000500 Time 0.021744 +2023-10-02 21:24:03,535 - Epoch: [114][ 940/ 1236] Overall Loss 0.197934 Objective Loss 0.197934 LR 0.000500 Time 0.021738 +2023-10-02 21:24:03,748 - Epoch: [114][ 950/ 1236] Overall Loss 0.198091 Objective Loss 0.198091 LR 0.000500 Time 0.021731 +2023-10-02 21:24:03,962 - Epoch: [114][ 960/ 1236] Overall Loss 0.198014 Objective Loss 0.198014 LR 0.000500 Time 0.021727 +2023-10-02 21:24:04,173 - Epoch: [114][ 970/ 1236] Overall Loss 0.198194 Objective Loss 0.198194 LR 0.000500 Time 0.021721 +2023-10-02 21:24:04,386 - Epoch: [114][ 980/ 1236] Overall Loss 0.198392 Objective Loss 0.198392 LR 0.000500 Time 0.021716 +2023-10-02 21:24:04,599 - Epoch: [114][ 990/ 1236] Overall Loss 0.198790 Objective Loss 0.198790 LR 0.000500 Time 0.021710 +2023-10-02 21:24:04,812 - Epoch: [114][ 1000/ 1236] Overall Loss 0.198700 Objective Loss 0.198700 LR 0.000500 Time 0.021706 +2023-10-02 21:24:05,025 - Epoch: [114][ 1010/ 1236] Overall Loss 0.198820 Objective Loss 0.198820 LR 0.000500 Time 0.021699 +2023-10-02 21:24:05,238 - Epoch: [114][ 1020/ 1236] Overall Loss 0.198908 Objective Loss 0.198908 LR 0.000500 Time 0.021695 +2023-10-02 21:24:05,450 - Epoch: [114][ 1030/ 1236] Overall Loss 0.198711 Objective Loss 0.198711 LR 0.000500 Time 0.021690 +2023-10-02 21:24:05,663 - Epoch: [114][ 1040/ 1236] Overall Loss 0.198662 Objective Loss 0.198662 LR 0.000500 Time 0.021685 +2023-10-02 21:24:05,876 - Epoch: [114][ 1050/ 1236] Overall Loss 0.198559 Objective Loss 0.198559 LR 0.000500 Time 0.021679 +2023-10-02 21:24:06,089 - Epoch: [114][ 1060/ 1236] Overall Loss 0.198611 Objective Loss 0.198611 LR 0.000500 Time 0.021676 +2023-10-02 21:24:06,302 - Epoch: [114][ 1070/ 1236] Overall Loss 0.198623 Objective Loss 0.198623 LR 0.000500 Time 0.021670 +2023-10-02 21:24:06,514 - Epoch: [114][ 1080/ 1236] Overall Loss 0.198482 Objective Loss 0.198482 LR 0.000500 Time 0.021667 +2023-10-02 21:24:06,727 - Epoch: [114][ 1090/ 1236] Overall Loss 0.198420 Objective Loss 0.198420 LR 0.000500 Time 0.021661 +2023-10-02 21:24:06,940 - Epoch: [114][ 1100/ 1236] Overall Loss 0.198280 Objective Loss 0.198280 LR 0.000500 Time 0.021658 +2023-10-02 21:24:07,153 - Epoch: [114][ 1110/ 1236] Overall Loss 0.198338 Objective Loss 0.198338 LR 0.000500 Time 0.021653 +2023-10-02 21:24:07,366 - Epoch: [114][ 1120/ 1236] Overall Loss 0.198323 Objective Loss 0.198323 LR 0.000500 Time 0.021649 +2023-10-02 21:24:07,578 - Epoch: [114][ 1130/ 1236] Overall Loss 0.198354 Objective Loss 0.198354 LR 0.000500 Time 0.021644 +2023-10-02 21:24:07,793 - Epoch: [114][ 1140/ 1236] Overall Loss 0.198398 Objective Loss 0.198398 LR 0.000500 Time 0.021642 +2023-10-02 21:24:08,004 - Epoch: [114][ 1150/ 1236] Overall Loss 0.198453 Objective Loss 0.198453 LR 0.000500 Time 0.021637 +2023-10-02 21:24:08,217 - Epoch: [114][ 1160/ 1236] Overall Loss 0.198539 Objective Loss 0.198539 LR 0.000500 Time 0.021634 +2023-10-02 21:24:08,429 - Epoch: [114][ 1170/ 1236] Overall Loss 0.198425 Objective Loss 0.198425 LR 0.000500 Time 0.021629 +2023-10-02 21:24:08,642 - Epoch: [114][ 1180/ 1236] Overall Loss 0.198452 Objective Loss 0.198452 LR 0.000500 Time 0.021627 +2023-10-02 21:24:08,854 - Epoch: [114][ 1190/ 1236] Overall Loss 0.198362 Objective Loss 0.198362 LR 0.000500 Time 0.021622 +2023-10-02 21:24:09,067 - Epoch: [114][ 1200/ 1236] Overall Loss 0.198391 Objective Loss 0.198391 LR 0.000500 Time 0.021618 +2023-10-02 21:24:09,279 - Epoch: [114][ 1210/ 1236] Overall Loss 0.198563 Objective Loss 0.198563 LR 0.000500 Time 0.021614 +2023-10-02 21:24:09,492 - Epoch: [114][ 1220/ 1236] Overall Loss 0.198584 Objective Loss 0.198584 LR 0.000500 Time 0.021610 +2023-10-02 21:24:09,757 - Epoch: [114][ 1230/ 1236] Overall Loss 0.198529 Objective Loss 0.198529 LR 0.000500 Time 0.021649 +2023-10-02 21:24:09,879 - Epoch: [114][ 1236/ 1236] Overall Loss 0.198626 Objective Loss 0.198626 Top1 89.409369 Top5 98.778004 LR 0.000500 Time 0.021643 +2023-10-02 21:24:10,016 - --- validate (epoch=114)----------- +2023-10-02 21:24:10,016 - 29943 samples (256 per mini-batch) +2023-10-02 21:24:10,521 - Epoch: [114][ 10/ 117] Loss 0.278555 Top1 85.468750 Top5 98.593750 +2023-10-02 21:24:10,672 - Epoch: [114][ 20/ 117] Loss 0.286344 Top1 85.781250 Top5 98.378906 +2023-10-02 21:24:10,823 - Epoch: [114][ 30/ 117] Loss 0.297060 Top1 85.364583 Top5 98.450521 +2023-10-02 21:24:10,974 - Epoch: [114][ 40/ 117] Loss 0.292425 Top1 85.605469 Top5 98.427734 +2023-10-02 21:24:11,126 - Epoch: [114][ 50/ 117] Loss 0.290226 Top1 85.625000 Top5 98.476562 +2023-10-02 21:24:11,276 - Epoch: [114][ 60/ 117] Loss 0.289025 Top1 85.820312 Top5 98.470052 +2023-10-02 21:24:11,427 - Epoch: [114][ 70/ 117] Loss 0.288623 Top1 85.708705 Top5 98.498884 +2023-10-02 21:24:11,582 - Epoch: [114][ 80/ 117] Loss 0.295647 Top1 85.590820 Top5 98.515625 +2023-10-02 21:24:11,740 - Epoch: [114][ 90/ 117] Loss 0.294961 Top1 85.646701 Top5 98.559028 +2023-10-02 21:24:11,901 - Epoch: [114][ 100/ 117] Loss 0.297371 Top1 85.582031 Top5 98.503906 +2023-10-02 21:24:12,068 - Epoch: [114][ 110/ 117] Loss 0.297797 Top1 85.603693 Top5 98.444602 +2023-10-02 21:24:12,158 - Epoch: [114][ 117/ 117] Loss 0.296842 Top1 85.602645 Top5 98.473767 +2023-10-02 21:24:12,315 - ==> Top1: 85.603 Top5: 98.474 Loss: 0.297 + +2023-10-02 21:24:12,316 - ==> Confusion: +[[ 948 3 4 1 8 2 0 0 8 49 1 0 1 3 3 0 2 0 0 0 17] + [ 0 1063 0 0 5 18 0 17 2 1 5 0 1 0 1 3 1 1 8 2 3] + [ 3 1 986 8 1 0 17 3 0 1 2 0 6 2 1 3 2 2 11 3 4] + [ 0 3 13 965 2 2 2 3 5 1 11 0 6 4 27 2 0 4 18 0 21] + [ 28 13 0 0 960 3 1 1 1 12 1 0 3 1 8 4 8 0 1 2 3] + [ 4 43 0 3 2 980 1 30 4 9 2 2 1 7 3 1 3 0 4 2 15] + [ 1 1 24 1 0 1 1139 3 0 0 5 0 0 0 0 4 0 1 0 6 5] + [ 1 16 10 0 2 18 7 1080 0 4 8 3 3 2 0 2 3 1 39 10 9] + [ 18 1 1 0 1 2 0 1 977 36 11 1 3 8 19 1 6 0 2 0 1] + [ 109 0 1 0 6 1 0 0 23 939 0 0 0 19 7 2 1 1 1 4 5] + [ 6 4 7 5 0 2 2 2 11 0 977 3 1 7 3 1 2 1 8 0 11] + [ 2 2 1 0 2 24 1 2 0 0 0 936 24 11 0 6 1 17 0 3 3] + [ 0 2 3 3 1 3 1 1 0 1 2 26 974 4 1 3 3 11 4 7 18] + [ 1 0 2 0 1 9 1 0 14 14 3 3 0 1050 4 0 0 1 0 0 16] + [ 14 0 5 15 3 1 0 0 20 2 2 0 3 3 1015 0 0 1 6 0 11] + [ 0 0 3 2 6 2 1 0 0 0 0 2 5 1 1 1074 14 12 3 4 4] + [ 2 15 0 0 4 7 0 4 1 0 0 2 1 0 5 10 1094 0 0 4 12] + [ 0 0 1 3 1 0 4 1 2 1 0 1 27 0 3 5 0 983 0 2 4] + [ 2 6 6 17 4 1 1 17 6 1 6 1 1 0 8 0 0 0 981 1 9] + [ 0 0 1 3 0 4 11 6 0 0 2 14 4 1 0 1 8 0 0 1087 10] + [ 141 192 151 64 65 137 46 75 85 92 195 72 335 242 113 51 63 58 120 184 5424]] + +2023-10-02 21:24:12,317 - ==> Best [Top1: 85.603 Top5: 98.474 Sparsity:0.00 Params: 169472 on epoch: 114] +2023-10-02 21:24:12,317 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:24:12,331 - + +2023-10-02 21:24:12,331 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:24:13,362 - Epoch: [115][ 10/ 1236] Overall Loss 0.194467 Objective Loss 0.194467 LR 0.000500 Time 0.103084 +2023-10-02 21:24:13,571 - Epoch: [115][ 20/ 1236] Overall Loss 0.191819 Objective Loss 0.191819 LR 0.000500 Time 0.061964 +2023-10-02 21:24:13,780 - Epoch: [115][ 30/ 1236] Overall Loss 0.189227 Objective Loss 0.189227 LR 0.000500 Time 0.048226 +2023-10-02 21:24:13,989 - Epoch: [115][ 40/ 1236] Overall Loss 0.183692 Objective Loss 0.183692 LR 0.000500 Time 0.041384 +2023-10-02 21:24:14,198 - Epoch: [115][ 50/ 1236] Overall Loss 0.188983 Objective Loss 0.188983 LR 0.000500 Time 0.037249 +2023-10-02 21:24:14,406 - Epoch: [115][ 60/ 1236] Overall Loss 0.188358 Objective Loss 0.188358 LR 0.000500 Time 0.034513 +2023-10-02 21:24:14,615 - Epoch: [115][ 70/ 1236] Overall Loss 0.188201 Objective Loss 0.188201 LR 0.000500 Time 0.032540 +2023-10-02 21:24:14,823 - Epoch: [115][ 80/ 1236] Overall Loss 0.185636 Objective Loss 0.185636 LR 0.000500 Time 0.031070 +2023-10-02 21:24:15,031 - Epoch: [115][ 90/ 1236] Overall Loss 0.185641 Objective Loss 0.185641 LR 0.000500 Time 0.029932 +2023-10-02 21:24:15,242 - Epoch: [115][ 100/ 1236] Overall Loss 0.186542 Objective Loss 0.186542 LR 0.000500 Time 0.029043 +2023-10-02 21:24:15,449 - Epoch: [115][ 110/ 1236] Overall Loss 0.187630 Objective Loss 0.187630 LR 0.000500 Time 0.028284 +2023-10-02 21:24:15,658 - Epoch: [115][ 120/ 1236] Overall Loss 0.187060 Objective Loss 0.187060 LR 0.000500 Time 0.027661 +2023-10-02 21:24:15,865 - Epoch: [115][ 130/ 1236] Overall Loss 0.189865 Objective Loss 0.189865 LR 0.000500 Time 0.027120 +2023-10-02 21:24:16,074 - Epoch: [115][ 140/ 1236] Overall Loss 0.189197 Objective Loss 0.189197 LR 0.000500 Time 0.026669 +2023-10-02 21:24:16,282 - Epoch: [115][ 150/ 1236] Overall Loss 0.189167 Objective Loss 0.189167 LR 0.000500 Time 0.026267 +2023-10-02 21:24:16,490 - Epoch: [115][ 160/ 1236] Overall Loss 0.189265 Objective Loss 0.189265 LR 0.000500 Time 0.025928 +2023-10-02 21:24:16,698 - Epoch: [115][ 170/ 1236] Overall Loss 0.190107 Objective Loss 0.190107 LR 0.000500 Time 0.025616 +2023-10-02 21:24:16,906 - Epoch: [115][ 180/ 1236] Overall Loss 0.191144 Objective Loss 0.191144 LR 0.000500 Time 0.025349 +2023-10-02 21:24:17,114 - Epoch: [115][ 190/ 1236] Overall Loss 0.191857 Objective Loss 0.191857 LR 0.000500 Time 0.025099 +2023-10-02 21:24:17,323 - Epoch: [115][ 200/ 1236] Overall Loss 0.192370 Objective Loss 0.192370 LR 0.000500 Time 0.024886 +2023-10-02 21:24:17,531 - Epoch: [115][ 210/ 1236] Overall Loss 0.191594 Objective Loss 0.191594 LR 0.000500 Time 0.024684 +2023-10-02 21:24:17,739 - Epoch: [115][ 220/ 1236] Overall Loss 0.191290 Objective Loss 0.191290 LR 0.000500 Time 0.024508 +2023-10-02 21:24:17,947 - Epoch: [115][ 230/ 1236] Overall Loss 0.191308 Objective Loss 0.191308 LR 0.000500 Time 0.024340 +2023-10-02 21:24:18,156 - Epoch: [115][ 240/ 1236] Overall Loss 0.191579 Objective Loss 0.191579 LR 0.000500 Time 0.024194 +2023-10-02 21:24:18,363 - Epoch: [115][ 250/ 1236] Overall Loss 0.191566 Objective Loss 0.191566 LR 0.000500 Time 0.024052 +2023-10-02 21:24:18,572 - Epoch: [115][ 260/ 1236] Overall Loss 0.191545 Objective Loss 0.191545 LR 0.000500 Time 0.023927 +2023-10-02 21:24:18,780 - Epoch: [115][ 270/ 1236] Overall Loss 0.192412 Objective Loss 0.192412 LR 0.000500 Time 0.023806 +2023-10-02 21:24:18,988 - Epoch: [115][ 280/ 1236] Overall Loss 0.193202 Objective Loss 0.193202 LR 0.000500 Time 0.023700 +2023-10-02 21:24:19,196 - Epoch: [115][ 290/ 1236] Overall Loss 0.193166 Objective Loss 0.193166 LR 0.000500 Time 0.023593 +2023-10-02 21:24:19,405 - Epoch: [115][ 300/ 1236] Overall Loss 0.193446 Objective Loss 0.193446 LR 0.000500 Time 0.023501 +2023-10-02 21:24:19,613 - Epoch: [115][ 310/ 1236] Overall Loss 0.192669 Objective Loss 0.192669 LR 0.000500 Time 0.023409 +2023-10-02 21:24:19,821 - Epoch: [115][ 320/ 1236] Overall Loss 0.192793 Objective Loss 0.192793 LR 0.000500 Time 0.023328 +2023-10-02 21:24:20,030 - Epoch: [115][ 330/ 1236] Overall Loss 0.192449 Objective Loss 0.192449 LR 0.000500 Time 0.023249 +2023-10-02 21:24:20,238 - Epoch: [115][ 340/ 1236] Overall Loss 0.192328 Objective Loss 0.192328 LR 0.000500 Time 0.023177 +2023-10-02 21:24:20,447 - Epoch: [115][ 350/ 1236] Overall Loss 0.192239 Objective Loss 0.192239 LR 0.000500 Time 0.023107 +2023-10-02 21:24:20,657 - Epoch: [115][ 360/ 1236] Overall Loss 0.192390 Objective Loss 0.192390 LR 0.000500 Time 0.023049 +2023-10-02 21:24:20,864 - Epoch: [115][ 370/ 1236] Overall Loss 0.192703 Objective Loss 0.192703 LR 0.000500 Time 0.022984 +2023-10-02 21:24:21,075 - Epoch: [115][ 380/ 1236] Overall Loss 0.192720 Objective Loss 0.192720 LR 0.000500 Time 0.022933 +2023-10-02 21:24:21,282 - Epoch: [115][ 390/ 1236] Overall Loss 0.192679 Objective Loss 0.192679 LR 0.000500 Time 0.022876 +2023-10-02 21:24:21,492 - Epoch: [115][ 400/ 1236] Overall Loss 0.192936 Objective Loss 0.192936 LR 0.000500 Time 0.022827 +2023-10-02 21:24:21,700 - Epoch: [115][ 410/ 1236] Overall Loss 0.192556 Objective Loss 0.192556 LR 0.000500 Time 0.022776 +2023-10-02 21:24:21,910 - Epoch: [115][ 420/ 1236] Overall Loss 0.192844 Objective Loss 0.192844 LR 0.000500 Time 0.022731 +2023-10-02 21:24:22,118 - Epoch: [115][ 430/ 1236] Overall Loss 0.192549 Objective Loss 0.192549 LR 0.000500 Time 0.022684 +2023-10-02 21:24:22,328 - Epoch: [115][ 440/ 1236] Overall Loss 0.193355 Objective Loss 0.193355 LR 0.000500 Time 0.022644 +2023-10-02 21:24:22,536 - Epoch: [115][ 450/ 1236] Overall Loss 0.193634 Objective Loss 0.193634 LR 0.000500 Time 0.022601 +2023-10-02 21:24:22,746 - Epoch: [115][ 460/ 1236] Overall Loss 0.193653 Objective Loss 0.193653 LR 0.000500 Time 0.022564 +2023-10-02 21:24:22,955 - Epoch: [115][ 470/ 1236] Overall Loss 0.194240 Objective Loss 0.194240 LR 0.000500 Time 0.022526 +2023-10-02 21:24:23,164 - Epoch: [115][ 480/ 1236] Overall Loss 0.194348 Objective Loss 0.194348 LR 0.000500 Time 0.022492 +2023-10-02 21:24:23,373 - Epoch: [115][ 490/ 1236] Overall Loss 0.194303 Objective Loss 0.194303 LR 0.000500 Time 0.022456 +2023-10-02 21:24:23,583 - Epoch: [115][ 500/ 1236] Overall Loss 0.194297 Objective Loss 0.194297 LR 0.000500 Time 0.022425 +2023-10-02 21:24:23,791 - Epoch: [115][ 510/ 1236] Overall Loss 0.194446 Objective Loss 0.194446 LR 0.000500 Time 0.022392 +2023-10-02 21:24:24,001 - Epoch: [115][ 520/ 1236] Overall Loss 0.194457 Objective Loss 0.194457 LR 0.000500 Time 0.022364 +2023-10-02 21:24:24,210 - Epoch: [115][ 530/ 1236] Overall Loss 0.194471 Objective Loss 0.194471 LR 0.000500 Time 0.022334 +2023-10-02 21:24:24,420 - Epoch: [115][ 540/ 1236] Overall Loss 0.194364 Objective Loss 0.194364 LR 0.000500 Time 0.022307 +2023-10-02 21:24:24,628 - Epoch: [115][ 550/ 1236] Overall Loss 0.194788 Objective Loss 0.194788 LR 0.000500 Time 0.022279 +2023-10-02 21:24:24,838 - Epoch: [115][ 560/ 1236] Overall Loss 0.194856 Objective Loss 0.194856 LR 0.000500 Time 0.022255 +2023-10-02 21:24:25,047 - Epoch: [115][ 570/ 1236] Overall Loss 0.194725 Objective Loss 0.194725 LR 0.000500 Time 0.022228 +2023-10-02 21:24:25,257 - Epoch: [115][ 580/ 1236] Overall Loss 0.194604 Objective Loss 0.194604 LR 0.000500 Time 0.022206 +2023-10-02 21:24:25,466 - Epoch: [115][ 590/ 1236] Overall Loss 0.194709 Objective Loss 0.194709 LR 0.000500 Time 0.022181 +2023-10-02 21:24:25,676 - Epoch: [115][ 600/ 1236] Overall Loss 0.194933 Objective Loss 0.194933 LR 0.000500 Time 0.022161 +2023-10-02 21:24:25,884 - Epoch: [115][ 610/ 1236] Overall Loss 0.195143 Objective Loss 0.195143 LR 0.000500 Time 0.022137 +2023-10-02 21:24:26,094 - Epoch: [115][ 620/ 1236] Overall Loss 0.195066 Objective Loss 0.195066 LR 0.000500 Time 0.022117 +2023-10-02 21:24:26,303 - Epoch: [115][ 630/ 1236] Overall Loss 0.195081 Objective Loss 0.195081 LR 0.000500 Time 0.022096 +2023-10-02 21:24:26,512 - Epoch: [115][ 640/ 1236] Overall Loss 0.195339 Objective Loss 0.195339 LR 0.000500 Time 0.022077 +2023-10-02 21:24:26,721 - Epoch: [115][ 650/ 1236] Overall Loss 0.195309 Objective Loss 0.195309 LR 0.000500 Time 0.022057 +2023-10-02 21:24:26,931 - Epoch: [115][ 660/ 1236] Overall Loss 0.195401 Objective Loss 0.195401 LR 0.000500 Time 0.022039 +2023-10-02 21:24:27,140 - Epoch: [115][ 670/ 1236] Overall Loss 0.195481 Objective Loss 0.195481 LR 0.000500 Time 0.022020 +2023-10-02 21:24:27,349 - Epoch: [115][ 680/ 1236] Overall Loss 0.195803 Objective Loss 0.195803 LR 0.000500 Time 0.022004 +2023-10-02 21:24:27,558 - Epoch: [115][ 690/ 1236] Overall Loss 0.195674 Objective Loss 0.195674 LR 0.000500 Time 0.021986 +2023-10-02 21:24:27,767 - Epoch: [115][ 700/ 1236] Overall Loss 0.195908 Objective Loss 0.195908 LR 0.000500 Time 0.021970 +2023-10-02 21:24:27,976 - Epoch: [115][ 710/ 1236] Overall Loss 0.195991 Objective Loss 0.195991 LR 0.000500 Time 0.021953 +2023-10-02 21:24:28,186 - Epoch: [115][ 720/ 1236] Overall Loss 0.196189 Objective Loss 0.196189 LR 0.000500 Time 0.021938 +2023-10-02 21:24:28,395 - Epoch: [115][ 730/ 1236] Overall Loss 0.196146 Objective Loss 0.196146 LR 0.000500 Time 0.021922 +2023-10-02 21:24:28,604 - Epoch: [115][ 740/ 1236] Overall Loss 0.196715 Objective Loss 0.196715 LR 0.000500 Time 0.021908 +2023-10-02 21:24:28,813 - Epoch: [115][ 750/ 1236] Overall Loss 0.196892 Objective Loss 0.196892 LR 0.000500 Time 0.021892 +2023-10-02 21:24:29,022 - Epoch: [115][ 760/ 1236] Overall Loss 0.197155 Objective Loss 0.197155 LR 0.000500 Time 0.021880 +2023-10-02 21:24:29,231 - Epoch: [115][ 770/ 1236] Overall Loss 0.197126 Objective Loss 0.197126 LR 0.000500 Time 0.021865 +2023-10-02 21:24:29,441 - Epoch: [115][ 780/ 1236] Overall Loss 0.197204 Objective Loss 0.197204 LR 0.000500 Time 0.021853 +2023-10-02 21:24:29,650 - Epoch: [115][ 790/ 1236] Overall Loss 0.197299 Objective Loss 0.197299 LR 0.000500 Time 0.021839 +2023-10-02 21:24:29,859 - Epoch: [115][ 800/ 1236] Overall Loss 0.197439 Objective Loss 0.197439 LR 0.000500 Time 0.021827 +2023-10-02 21:24:30,069 - Epoch: [115][ 810/ 1236] Overall Loss 0.197634 Objective Loss 0.197634 LR 0.000500 Time 0.021815 +2023-10-02 21:24:30,279 - Epoch: [115][ 820/ 1236] Overall Loss 0.197654 Objective Loss 0.197654 LR 0.000500 Time 0.021804 +2023-10-02 21:24:30,488 - Epoch: [115][ 830/ 1236] Overall Loss 0.197266 Objective Loss 0.197266 LR 0.000500 Time 0.021791 +2023-10-02 21:24:30,697 - Epoch: [115][ 840/ 1236] Overall Loss 0.197500 Objective Loss 0.197500 LR 0.000500 Time 0.021781 +2023-10-02 21:24:30,906 - Epoch: [115][ 850/ 1236] Overall Loss 0.197391 Objective Loss 0.197391 LR 0.000500 Time 0.021769 +2023-10-02 21:24:31,116 - Epoch: [115][ 860/ 1236] Overall Loss 0.197570 Objective Loss 0.197570 LR 0.000500 Time 0.021760 +2023-10-02 21:24:31,325 - Epoch: [115][ 870/ 1236] Overall Loss 0.197736 Objective Loss 0.197736 LR 0.000500 Time 0.021748 +2023-10-02 21:24:31,536 - Epoch: [115][ 880/ 1236] Overall Loss 0.197690 Objective Loss 0.197690 LR 0.000500 Time 0.021739 +2023-10-02 21:24:31,745 - Epoch: [115][ 890/ 1236] Overall Loss 0.197466 Objective Loss 0.197466 LR 0.000500 Time 0.021728 +2023-10-02 21:24:31,956 - Epoch: [115][ 900/ 1236] Overall Loss 0.197254 Objective Loss 0.197254 LR 0.000500 Time 0.021721 +2023-10-02 21:24:32,164 - Epoch: [115][ 910/ 1236] Overall Loss 0.197339 Objective Loss 0.197339 LR 0.000500 Time 0.021711 +2023-10-02 21:24:32,374 - Epoch: [115][ 920/ 1236] Overall Loss 0.197407 Objective Loss 0.197407 LR 0.000500 Time 0.021703 +2023-10-02 21:24:32,582 - Epoch: [115][ 930/ 1236] Overall Loss 0.197526 Objective Loss 0.197526 LR 0.000500 Time 0.021692 +2023-10-02 21:24:32,792 - Epoch: [115][ 940/ 1236] Overall Loss 0.197279 Objective Loss 0.197279 LR 0.000500 Time 0.021684 +2023-10-02 21:24:33,001 - Epoch: [115][ 950/ 1236] Overall Loss 0.197364 Objective Loss 0.197364 LR 0.000500 Time 0.021674 +2023-10-02 21:24:33,211 - Epoch: [115][ 960/ 1236] Overall Loss 0.197329 Objective Loss 0.197329 LR 0.000500 Time 0.021667 +2023-10-02 21:24:33,420 - Epoch: [115][ 970/ 1236] Overall Loss 0.197363 Objective Loss 0.197363 LR 0.000500 Time 0.021658 +2023-10-02 21:24:33,630 - Epoch: [115][ 980/ 1236] Overall Loss 0.197429 Objective Loss 0.197429 LR 0.000500 Time 0.021651 +2023-10-02 21:24:33,839 - Epoch: [115][ 990/ 1236] Overall Loss 0.197386 Objective Loss 0.197386 LR 0.000500 Time 0.021642 +2023-10-02 21:24:34,050 - Epoch: [115][ 1000/ 1236] Overall Loss 0.197330 Objective Loss 0.197330 LR 0.000500 Time 0.021635 +2023-10-02 21:24:34,258 - Epoch: [115][ 1010/ 1236] Overall Loss 0.197362 Objective Loss 0.197362 LR 0.000500 Time 0.021626 +2023-10-02 21:24:34,468 - Epoch: [115][ 1020/ 1236] Overall Loss 0.197508 Objective Loss 0.197508 LR 0.000500 Time 0.021619 +2023-10-02 21:24:34,678 - Epoch: [115][ 1030/ 1236] Overall Loss 0.197809 Objective Loss 0.197809 LR 0.000500 Time 0.021611 +2023-10-02 21:24:34,888 - Epoch: [115][ 1040/ 1236] Overall Loss 0.197891 Objective Loss 0.197891 LR 0.000500 Time 0.021605 +2023-10-02 21:24:35,097 - Epoch: [115][ 1050/ 1236] Overall Loss 0.197819 Objective Loss 0.197819 LR 0.000500 Time 0.021597 +2023-10-02 21:24:35,307 - Epoch: [115][ 1060/ 1236] Overall Loss 0.197633 Objective Loss 0.197633 LR 0.000500 Time 0.021591 +2023-10-02 21:24:35,516 - Epoch: [115][ 1070/ 1236] Overall Loss 0.197724 Objective Loss 0.197724 LR 0.000500 Time 0.021584 +2023-10-02 21:24:35,726 - Epoch: [115][ 1080/ 1236] Overall Loss 0.197663 Objective Loss 0.197663 LR 0.000500 Time 0.021578 +2023-10-02 21:24:35,935 - Epoch: [115][ 1090/ 1236] Overall Loss 0.197698 Objective Loss 0.197698 LR 0.000500 Time 0.021570 +2023-10-02 21:24:36,146 - Epoch: [115][ 1100/ 1236] Overall Loss 0.197597 Objective Loss 0.197597 LR 0.000500 Time 0.021565 +2023-10-02 21:24:36,355 - Epoch: [115][ 1110/ 1236] Overall Loss 0.197572 Objective Loss 0.197572 LR 0.000500 Time 0.021558 +2023-10-02 21:24:36,565 - Epoch: [115][ 1120/ 1236] Overall Loss 0.197564 Objective Loss 0.197564 LR 0.000500 Time 0.021553 +2023-10-02 21:24:36,774 - Epoch: [115][ 1130/ 1236] Overall Loss 0.197433 Objective Loss 0.197433 LR 0.000500 Time 0.021546 +2023-10-02 21:24:36,985 - Epoch: [115][ 1140/ 1236] Overall Loss 0.197523 Objective Loss 0.197523 LR 0.000500 Time 0.021542 +2023-10-02 21:24:37,191 - Epoch: [115][ 1150/ 1236] Overall Loss 0.197594 Objective Loss 0.197594 LR 0.000500 Time 0.021533 +2023-10-02 21:24:37,399 - Epoch: [115][ 1160/ 1236] Overall Loss 0.197479 Objective Loss 0.197479 LR 0.000500 Time 0.021527 +2023-10-02 21:24:37,606 - Epoch: [115][ 1170/ 1236] Overall Loss 0.197372 Objective Loss 0.197372 LR 0.000500 Time 0.021518 +2023-10-02 21:24:37,814 - Epoch: [115][ 1180/ 1236] Overall Loss 0.197234 Objective Loss 0.197234 LR 0.000500 Time 0.021512 +2023-10-02 21:24:38,022 - Epoch: [115][ 1190/ 1236] Overall Loss 0.197351 Objective Loss 0.197351 LR 0.000500 Time 0.021504 +2023-10-02 21:24:38,230 - Epoch: [115][ 1200/ 1236] Overall Loss 0.197217 Objective Loss 0.197217 LR 0.000500 Time 0.021498 +2023-10-02 21:24:38,437 - Epoch: [115][ 1210/ 1236] Overall Loss 0.197331 Objective Loss 0.197331 LR 0.000500 Time 0.021491 +2023-10-02 21:24:38,645 - Epoch: [115][ 1220/ 1236] Overall Loss 0.197399 Objective Loss 0.197399 LR 0.000500 Time 0.021485 +2023-10-02 21:24:38,907 - Epoch: [115][ 1230/ 1236] Overall Loss 0.197367 Objective Loss 0.197367 LR 0.000500 Time 0.021522 +2023-10-02 21:24:39,030 - Epoch: [115][ 1236/ 1236] Overall Loss 0.197423 Objective Loss 0.197423 Top1 87.169043 Top5 99.389002 LR 0.000500 Time 0.021517 +2023-10-02 21:24:39,169 - --- validate (epoch=115)----------- +2023-10-02 21:24:39,169 - 29943 samples (256 per mini-batch) +2023-10-02 21:24:39,668 - Epoch: [115][ 10/ 117] Loss 0.312117 Top1 85.546875 Top5 98.242188 +2023-10-02 21:24:39,824 - Epoch: [115][ 20/ 117] Loss 0.304153 Top1 85.332031 Top5 98.320312 +2023-10-02 21:24:39,977 - Epoch: [115][ 30/ 117] Loss 0.308495 Top1 85.195312 Top5 98.307292 +2023-10-02 21:24:40,130 - Epoch: [115][ 40/ 117] Loss 0.306313 Top1 85.410156 Top5 98.349609 +2023-10-02 21:24:40,283 - Epoch: [115][ 50/ 117] Loss 0.305678 Top1 85.421875 Top5 98.312500 +2023-10-02 21:24:40,438 - Epoch: [115][ 60/ 117] Loss 0.302665 Top1 85.579427 Top5 98.378906 +2023-10-02 21:24:40,591 - Epoch: [115][ 70/ 117] Loss 0.300246 Top1 85.597098 Top5 98.420759 +2023-10-02 21:24:40,743 - Epoch: [115][ 80/ 117] Loss 0.302909 Top1 85.449219 Top5 98.383789 +2023-10-02 21:24:40,896 - Epoch: [115][ 90/ 117] Loss 0.300008 Top1 85.525174 Top5 98.415799 +2023-10-02 21:24:41,048 - Epoch: [115][ 100/ 117] Loss 0.300136 Top1 85.441406 Top5 98.433594 +2023-10-02 21:24:41,208 - Epoch: [115][ 110/ 117] Loss 0.300814 Top1 85.433239 Top5 98.437500 +2023-10-02 21:24:41,297 - Epoch: [115][ 117/ 117] Loss 0.302516 Top1 85.402264 Top5 98.420332 +2023-10-02 21:24:41,391 - ==> Top1: 85.402 Top5: 98.420 Loss: 0.303 + +2023-10-02 21:24:41,391 - ==> Confusion: +[[ 933 1 6 0 9 3 0 0 4 63 1 0 1 2 4 1 3 0 2 0 17] + [ 0 1069 1 0 2 18 1 22 0 0 1 0 0 0 0 3 1 0 6 3 4] + [ 3 1 981 12 4 0 14 6 0 0 1 0 7 2 1 2 2 1 10 2 7] + [ 1 4 15 975 0 3 2 2 4 1 3 0 3 1 30 3 0 5 17 0 20] + [ 27 12 1 0 959 3 0 0 2 11 1 0 1 3 6 5 12 0 3 0 4] + [ 3 46 0 1 1 985 1 28 3 6 3 5 0 5 3 0 5 1 4 2 14] + [ 0 3 31 0 0 1 1129 2 0 0 6 0 0 0 0 3 0 0 0 8 8] + [ 0 13 17 0 3 24 7 1089 0 3 7 6 3 4 1 0 1 0 27 4 9] + [ 17 0 0 1 0 4 0 2 975 42 10 3 2 9 16 1 1 0 4 0 2] + [ 104 0 2 0 7 2 1 0 23 950 0 0 0 11 7 0 1 1 0 4 6] + [ 3 3 8 7 1 2 2 3 9 1 979 4 0 8 3 2 3 1 4 0 10] + [ 1 0 1 0 1 16 0 2 0 0 0 959 15 7 0 2 6 16 0 5 4] + [ 0 1 6 5 0 2 4 3 1 1 3 43 940 4 5 5 1 18 2 4 20] + [ 1 0 0 0 3 10 0 1 15 15 10 5 0 1037 9 0 1 1 0 0 11] + [ 12 0 5 15 3 0 0 0 18 3 2 0 2 3 1019 0 0 2 7 0 10] + [ 0 0 2 0 5 1 0 0 0 0 0 6 3 0 0 1069 17 16 3 5 7] + [ 2 18 2 0 5 5 1 0 0 1 0 3 0 1 5 8 1092 0 1 3 14] + [ 0 1 1 1 0 1 2 1 1 0 0 0 19 0 3 6 0 998 0 1 3] + [ 1 6 4 11 2 0 1 23 3 0 4 0 0 0 10 0 0 0 995 0 8] + [ 0 2 3 2 1 4 10 6 0 0 1 11 2 0 1 2 9 1 2 1081 14] + [ 147 218 167 73 71 128 48 89 102 80 165 80 304 260 148 41 66 63 121 176 5358]] + +2023-10-02 21:24:41,392 - ==> Best [Top1: 85.603 Top5: 98.474 Sparsity:0.00 Params: 169472 on epoch: 114] +2023-10-02 21:24:41,392 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:24:41,398 - + +2023-10-02 21:24:41,399 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:24:42,542 - Epoch: [116][ 10/ 1236] Overall Loss 0.207878 Objective Loss 0.207878 LR 0.000500 Time 0.114298 +2023-10-02 21:24:42,750 - Epoch: [116][ 20/ 1236] Overall Loss 0.189590 Objective Loss 0.189590 LR 0.000500 Time 0.067531 +2023-10-02 21:24:42,957 - Epoch: [116][ 30/ 1236] Overall Loss 0.194941 Objective Loss 0.194941 LR 0.000500 Time 0.051865 +2023-10-02 21:24:43,165 - Epoch: [116][ 40/ 1236] Overall Loss 0.193820 Objective Loss 0.193820 LR 0.000500 Time 0.044097 +2023-10-02 21:24:43,372 - Epoch: [116][ 50/ 1236] Overall Loss 0.190158 Objective Loss 0.190158 LR 0.000500 Time 0.039382 +2023-10-02 21:24:43,581 - Epoch: [116][ 60/ 1236] Overall Loss 0.190047 Objective Loss 0.190047 LR 0.000500 Time 0.036305 +2023-10-02 21:24:43,787 - Epoch: [116][ 70/ 1236] Overall Loss 0.189790 Objective Loss 0.189790 LR 0.000500 Time 0.034059 +2023-10-02 21:24:43,996 - Epoch: [116][ 80/ 1236] Overall Loss 0.188210 Objective Loss 0.188210 LR 0.000500 Time 0.032401 +2023-10-02 21:24:44,203 - Epoch: [116][ 90/ 1236] Overall Loss 0.186596 Objective Loss 0.186596 LR 0.000500 Time 0.031084 +2023-10-02 21:24:44,412 - Epoch: [116][ 100/ 1236] Overall Loss 0.187450 Objective Loss 0.187450 LR 0.000500 Time 0.030067 +2023-10-02 21:24:44,618 - Epoch: [116][ 110/ 1236] Overall Loss 0.187387 Objective Loss 0.187387 LR 0.000500 Time 0.029200 +2023-10-02 21:24:44,827 - Epoch: [116][ 120/ 1236] Overall Loss 0.188424 Objective Loss 0.188424 LR 0.000500 Time 0.028509 +2023-10-02 21:24:45,033 - Epoch: [116][ 130/ 1236] Overall Loss 0.187533 Objective Loss 0.187533 LR 0.000500 Time 0.027895 +2023-10-02 21:24:45,242 - Epoch: [116][ 140/ 1236] Overall Loss 0.187653 Objective Loss 0.187653 LR 0.000500 Time 0.027397 +2023-10-02 21:24:45,448 - Epoch: [116][ 150/ 1236] Overall Loss 0.188320 Objective Loss 0.188320 LR 0.000500 Time 0.026940 +2023-10-02 21:24:45,656 - Epoch: [116][ 160/ 1236] Overall Loss 0.188491 Objective Loss 0.188491 LR 0.000500 Time 0.026552 +2023-10-02 21:24:45,863 - Epoch: [116][ 170/ 1236] Overall Loss 0.189072 Objective Loss 0.189072 LR 0.000500 Time 0.026206 +2023-10-02 21:24:46,071 - Epoch: [116][ 180/ 1236] Overall Loss 0.189077 Objective Loss 0.189077 LR 0.000500 Time 0.025905 +2023-10-02 21:24:46,278 - Epoch: [116][ 190/ 1236] Overall Loss 0.190243 Objective Loss 0.190243 LR 0.000500 Time 0.025623 +2023-10-02 21:24:46,487 - Epoch: [116][ 200/ 1236] Overall Loss 0.191417 Objective Loss 0.191417 LR 0.000500 Time 0.025388 +2023-10-02 21:24:46,697 - Epoch: [116][ 210/ 1236] Overall Loss 0.191006 Objective Loss 0.191006 LR 0.000500 Time 0.025176 +2023-10-02 21:24:46,910 - Epoch: [116][ 220/ 1236] Overall Loss 0.192172 Objective Loss 0.192172 LR 0.000500 Time 0.024998 +2023-10-02 21:24:47,127 - Epoch: [116][ 230/ 1236] Overall Loss 0.192536 Objective Loss 0.192536 LR 0.000500 Time 0.024854 +2023-10-02 21:24:47,341 - Epoch: [116][ 240/ 1236] Overall Loss 0.192278 Objective Loss 0.192278 LR 0.000500 Time 0.024706 +2023-10-02 21:24:47,551 - Epoch: [116][ 250/ 1236] Overall Loss 0.191799 Objective Loss 0.191799 LR 0.000500 Time 0.024557 +2023-10-02 21:24:47,757 - Epoch: [116][ 260/ 1236] Overall Loss 0.191742 Objective Loss 0.191742 LR 0.000500 Time 0.024402 +2023-10-02 21:24:47,966 - Epoch: [116][ 270/ 1236] Overall Loss 0.191194 Objective Loss 0.191194 LR 0.000500 Time 0.024269 +2023-10-02 21:24:48,171 - Epoch: [116][ 280/ 1236] Overall Loss 0.191930 Objective Loss 0.191930 LR 0.000500 Time 0.024135 +2023-10-02 21:24:48,379 - Epoch: [116][ 290/ 1236] Overall Loss 0.192875 Objective Loss 0.192875 LR 0.000500 Time 0.024017 +2023-10-02 21:24:48,586 - Epoch: [116][ 300/ 1236] Overall Loss 0.192879 Objective Loss 0.192879 LR 0.000500 Time 0.023901 +2023-10-02 21:24:48,794 - Epoch: [116][ 310/ 1236] Overall Loss 0.193333 Objective Loss 0.193333 LR 0.000500 Time 0.023803 +2023-10-02 21:24:49,000 - Epoch: [116][ 320/ 1236] Overall Loss 0.193888 Objective Loss 0.193888 LR 0.000500 Time 0.023700 +2023-10-02 21:24:49,209 - Epoch: [116][ 330/ 1236] Overall Loss 0.193439 Objective Loss 0.193439 LR 0.000500 Time 0.023614 +2023-10-02 21:24:49,414 - Epoch: [116][ 340/ 1236] Overall Loss 0.193216 Objective Loss 0.193216 LR 0.000500 Time 0.023523 +2023-10-02 21:24:49,623 - Epoch: [116][ 350/ 1236] Overall Loss 0.193555 Objective Loss 0.193555 LR 0.000500 Time 0.023446 +2023-10-02 21:24:49,832 - Epoch: [116][ 360/ 1236] Overall Loss 0.192995 Objective Loss 0.192995 LR 0.000500 Time 0.023371 +2023-10-02 21:24:50,039 - Epoch: [116][ 370/ 1236] Overall Loss 0.192990 Objective Loss 0.192990 LR 0.000500 Time 0.023299 +2023-10-02 21:24:50,247 - Epoch: [116][ 380/ 1236] Overall Loss 0.192896 Objective Loss 0.192896 LR 0.000500 Time 0.023233 +2023-10-02 21:24:50,455 - Epoch: [116][ 390/ 1236] Overall Loss 0.193817 Objective Loss 0.193817 LR 0.000500 Time 0.023167 +2023-10-02 21:24:50,662 - Epoch: [116][ 400/ 1236] Overall Loss 0.193698 Objective Loss 0.193698 LR 0.000500 Time 0.023105 +2023-10-02 21:24:50,870 - Epoch: [116][ 410/ 1236] Overall Loss 0.193647 Objective Loss 0.193647 LR 0.000500 Time 0.023047 +2023-10-02 21:24:51,077 - Epoch: [116][ 420/ 1236] Overall Loss 0.193348 Objective Loss 0.193348 LR 0.000500 Time 0.022991 +2023-10-02 21:24:51,291 - Epoch: [116][ 430/ 1236] Overall Loss 0.193441 Objective Loss 0.193441 LR 0.000500 Time 0.022955 +2023-10-02 21:24:51,505 - Epoch: [116][ 440/ 1236] Overall Loss 0.193168 Objective Loss 0.193168 LR 0.000500 Time 0.022917 +2023-10-02 21:24:51,723 - Epoch: [116][ 450/ 1236] Overall Loss 0.192826 Objective Loss 0.192826 LR 0.000500 Time 0.022892 +2023-10-02 21:24:51,934 - Epoch: [116][ 460/ 1236] Overall Loss 0.193159 Objective Loss 0.193159 LR 0.000500 Time 0.022852 +2023-10-02 21:24:52,143 - Epoch: [116][ 470/ 1236] Overall Loss 0.193011 Objective Loss 0.193011 LR 0.000500 Time 0.022810 +2023-10-02 21:24:52,353 - Epoch: [116][ 480/ 1236] Overall Loss 0.193243 Objective Loss 0.193243 LR 0.000500 Time 0.022768 +2023-10-02 21:24:52,562 - Epoch: [116][ 490/ 1236] Overall Loss 0.193581 Objective Loss 0.193581 LR 0.000500 Time 0.022729 +2023-10-02 21:24:52,771 - Epoch: [116][ 500/ 1236] Overall Loss 0.193790 Objective Loss 0.193790 LR 0.000500 Time 0.022693 +2023-10-02 21:24:52,981 - Epoch: [116][ 510/ 1236] Overall Loss 0.194360 Objective Loss 0.194360 LR 0.000500 Time 0.022659 +2023-10-02 21:24:53,190 - Epoch: [116][ 520/ 1236] Overall Loss 0.194754 Objective Loss 0.194754 LR 0.000500 Time 0.022624 +2023-10-02 21:24:53,399 - Epoch: [116][ 530/ 1236] Overall Loss 0.195263 Objective Loss 0.195263 LR 0.000500 Time 0.022592 +2023-10-02 21:24:53,609 - Epoch: [116][ 540/ 1236] Overall Loss 0.195504 Objective Loss 0.195504 LR 0.000500 Time 0.022560 +2023-10-02 21:24:53,818 - Epoch: [116][ 550/ 1236] Overall Loss 0.195589 Objective Loss 0.195589 LR 0.000500 Time 0.022530 +2023-10-02 21:24:54,028 - Epoch: [116][ 560/ 1236] Overall Loss 0.195861 Objective Loss 0.195861 LR 0.000500 Time 0.022501 +2023-10-02 21:24:54,237 - Epoch: [116][ 570/ 1236] Overall Loss 0.195835 Objective Loss 0.195835 LR 0.000500 Time 0.022473 +2023-10-02 21:24:54,447 - Epoch: [116][ 580/ 1236] Overall Loss 0.195637 Objective Loss 0.195637 LR 0.000500 Time 0.022446 +2023-10-02 21:24:54,656 - Epoch: [116][ 590/ 1236] Overall Loss 0.195754 Objective Loss 0.195754 LR 0.000500 Time 0.022419 +2023-10-02 21:24:54,865 - Epoch: [116][ 600/ 1236] Overall Loss 0.195911 Objective Loss 0.195911 LR 0.000500 Time 0.022394 +2023-10-02 21:24:55,074 - Epoch: [116][ 610/ 1236] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000500 Time 0.022369 +2023-10-02 21:24:55,284 - Epoch: [116][ 620/ 1236] Overall Loss 0.196214 Objective Loss 0.196214 LR 0.000500 Time 0.022346 +2023-10-02 21:24:55,493 - Epoch: [116][ 630/ 1236] Overall Loss 0.195936 Objective Loss 0.195936 LR 0.000500 Time 0.022323 +2023-10-02 21:24:55,702 - Epoch: [116][ 640/ 1236] Overall Loss 0.195521 Objective Loss 0.195521 LR 0.000500 Time 0.022301 +2023-10-02 21:24:55,912 - Epoch: [116][ 650/ 1236] Overall Loss 0.195007 Objective Loss 0.195007 LR 0.000500 Time 0.022279 +2023-10-02 21:24:56,121 - Epoch: [116][ 660/ 1236] Overall Loss 0.194670 Objective Loss 0.194670 LR 0.000500 Time 0.022258 +2023-10-02 21:24:56,331 - Epoch: [116][ 670/ 1236] Overall Loss 0.194581 Objective Loss 0.194581 LR 0.000500 Time 0.022238 +2023-10-02 21:24:56,540 - Epoch: [116][ 680/ 1236] Overall Loss 0.194739 Objective Loss 0.194739 LR 0.000500 Time 0.022219 +2023-10-02 21:24:56,749 - Epoch: [116][ 690/ 1236] Overall Loss 0.195028 Objective Loss 0.195028 LR 0.000500 Time 0.022199 +2023-10-02 21:24:56,959 - Epoch: [116][ 700/ 1236] Overall Loss 0.195045 Objective Loss 0.195045 LR 0.000500 Time 0.022181 +2023-10-02 21:24:57,168 - Epoch: [116][ 710/ 1236] Overall Loss 0.195040 Objective Loss 0.195040 LR 0.000500 Time 0.022162 +2023-10-02 21:24:57,377 - Epoch: [116][ 720/ 1236] Overall Loss 0.194989 Objective Loss 0.194989 LR 0.000500 Time 0.022145 +2023-10-02 21:24:57,586 - Epoch: [116][ 730/ 1236] Overall Loss 0.195141 Objective Loss 0.195141 LR 0.000500 Time 0.022128 +2023-10-02 21:24:57,796 - Epoch: [116][ 740/ 1236] Overall Loss 0.195291 Objective Loss 0.195291 LR 0.000500 Time 0.022111 +2023-10-02 21:24:58,005 - Epoch: [116][ 750/ 1236] Overall Loss 0.195453 Objective Loss 0.195453 LR 0.000500 Time 0.022094 +2023-10-02 21:24:58,214 - Epoch: [116][ 760/ 1236] Overall Loss 0.195332 Objective Loss 0.195332 LR 0.000500 Time 0.022079 +2023-10-02 21:24:58,423 - Epoch: [116][ 770/ 1236] Overall Loss 0.195093 Objective Loss 0.195093 LR 0.000500 Time 0.022063 +2023-10-02 21:24:58,633 - Epoch: [116][ 780/ 1236] Overall Loss 0.195216 Objective Loss 0.195216 LR 0.000500 Time 0.022049 +2023-10-02 21:24:58,841 - Epoch: [116][ 790/ 1236] Overall Loss 0.195316 Objective Loss 0.195316 LR 0.000500 Time 0.022032 +2023-10-02 21:24:59,050 - Epoch: [116][ 800/ 1236] Overall Loss 0.195391 Objective Loss 0.195391 LR 0.000500 Time 0.022018 +2023-10-02 21:24:59,259 - Epoch: [116][ 810/ 1236] Overall Loss 0.195519 Objective Loss 0.195519 LR 0.000500 Time 0.022004 +2023-10-02 21:24:59,472 - Epoch: [116][ 820/ 1236] Overall Loss 0.195597 Objective Loss 0.195597 LR 0.000500 Time 0.021995 +2023-10-02 21:24:59,692 - Epoch: [116][ 830/ 1236] Overall Loss 0.195343 Objective Loss 0.195343 LR 0.000500 Time 0.021994 +2023-10-02 21:24:59,908 - Epoch: [116][ 840/ 1236] Overall Loss 0.195278 Objective Loss 0.195278 LR 0.000500 Time 0.021989 +2023-10-02 21:25:00,127 - Epoch: [116][ 850/ 1236] Overall Loss 0.195283 Objective Loss 0.195283 LR 0.000500 Time 0.021988 +2023-10-02 21:25:00,343 - Epoch: [116][ 860/ 1236] Overall Loss 0.195382 Objective Loss 0.195382 LR 0.000500 Time 0.021983 +2023-10-02 21:25:00,562 - Epoch: [116][ 870/ 1236] Overall Loss 0.195390 Objective Loss 0.195390 LR 0.000500 Time 0.021981 +2023-10-02 21:25:00,778 - Epoch: [116][ 880/ 1236] Overall Loss 0.195418 Objective Loss 0.195418 LR 0.000500 Time 0.021977 +2023-10-02 21:25:00,998 - Epoch: [116][ 890/ 1236] Overall Loss 0.195378 Objective Loss 0.195378 LR 0.000500 Time 0.021976 +2023-10-02 21:25:01,214 - Epoch: [116][ 900/ 1236] Overall Loss 0.195585 Objective Loss 0.195585 LR 0.000500 Time 0.021972 +2023-10-02 21:25:01,433 - Epoch: [116][ 910/ 1236] Overall Loss 0.195316 Objective Loss 0.195316 LR 0.000500 Time 0.021971 +2023-10-02 21:25:01,649 - Epoch: [116][ 920/ 1236] Overall Loss 0.195145 Objective Loss 0.195145 LR 0.000500 Time 0.021966 +2023-10-02 21:25:01,868 - Epoch: [116][ 930/ 1236] Overall Loss 0.195214 Objective Loss 0.195214 LR 0.000500 Time 0.021965 +2023-10-02 21:25:02,084 - Epoch: [116][ 940/ 1236] Overall Loss 0.195476 Objective Loss 0.195476 LR 0.000500 Time 0.021960 +2023-10-02 21:25:02,306 - Epoch: [116][ 950/ 1236] Overall Loss 0.195698 Objective Loss 0.195698 LR 0.000500 Time 0.021962 +2023-10-02 21:25:02,523 - Epoch: [116][ 960/ 1236] Overall Loss 0.195651 Objective Loss 0.195651 LR 0.000500 Time 0.021959 +2023-10-02 21:25:02,746 - Epoch: [116][ 970/ 1236] Overall Loss 0.195709 Objective Loss 0.195709 LR 0.000500 Time 0.021962 +2023-10-02 21:25:02,963 - Epoch: [116][ 980/ 1236] Overall Loss 0.195748 Objective Loss 0.195748 LR 0.000500 Time 0.021959 +2023-10-02 21:25:03,186 - Epoch: [116][ 990/ 1236] Overall Loss 0.195691 Objective Loss 0.195691 LR 0.000500 Time 0.021962 +2023-10-02 21:25:03,403 - Epoch: [116][ 1000/ 1236] Overall Loss 0.195777 Objective Loss 0.195777 LR 0.000500 Time 0.021959 +2023-10-02 21:25:03,625 - Epoch: [116][ 1010/ 1236] Overall Loss 0.195699 Objective Loss 0.195699 LR 0.000500 Time 0.021961 +2023-10-02 21:25:03,842 - Epoch: [116][ 1020/ 1236] Overall Loss 0.195863 Objective Loss 0.195863 LR 0.000500 Time 0.021958 +2023-10-02 21:25:04,055 - Epoch: [116][ 1030/ 1236] Overall Loss 0.195764 Objective Loss 0.195764 LR 0.000500 Time 0.021951 +2023-10-02 21:25:04,266 - Epoch: [116][ 1040/ 1236] Overall Loss 0.195674 Objective Loss 0.195674 LR 0.000500 Time 0.021942 +2023-10-02 21:25:04,479 - Epoch: [116][ 1050/ 1236] Overall Loss 0.195400 Objective Loss 0.195400 LR 0.000500 Time 0.021937 +2023-10-02 21:25:04,690 - Epoch: [116][ 1060/ 1236] Overall Loss 0.195413 Objective Loss 0.195413 LR 0.000500 Time 0.021928 +2023-10-02 21:25:04,904 - Epoch: [116][ 1070/ 1236] Overall Loss 0.195301 Objective Loss 0.195301 LR 0.000500 Time 0.021922 +2023-10-02 21:25:05,114 - Epoch: [116][ 1080/ 1236] Overall Loss 0.195273 Objective Loss 0.195273 LR 0.000500 Time 0.021914 +2023-10-02 21:25:05,328 - Epoch: [116][ 1090/ 1236] Overall Loss 0.195357 Objective Loss 0.195357 LR 0.000500 Time 0.021908 +2023-10-02 21:25:05,539 - Epoch: [116][ 1100/ 1236] Overall Loss 0.195590 Objective Loss 0.195590 LR 0.000500 Time 0.021901 +2023-10-02 21:25:05,753 - Epoch: [116][ 1110/ 1236] Overall Loss 0.195697 Objective Loss 0.195697 LR 0.000500 Time 0.021896 +2023-10-02 21:25:05,964 - Epoch: [116][ 1120/ 1236] Overall Loss 0.195549 Objective Loss 0.195549 LR 0.000500 Time 0.021888 +2023-10-02 21:25:06,178 - Epoch: [116][ 1130/ 1236] Overall Loss 0.195397 Objective Loss 0.195397 LR 0.000500 Time 0.021883 +2023-10-02 21:25:06,388 - Epoch: [116][ 1140/ 1236] Overall Loss 0.195659 Objective Loss 0.195659 LR 0.000500 Time 0.021876 +2023-10-02 21:25:06,602 - Epoch: [116][ 1150/ 1236] Overall Loss 0.195543 Objective Loss 0.195543 LR 0.000500 Time 0.021871 +2023-10-02 21:25:06,813 - Epoch: [116][ 1160/ 1236] Overall Loss 0.195889 Objective Loss 0.195889 LR 0.000500 Time 0.021864 +2023-10-02 21:25:07,026 - Epoch: [116][ 1170/ 1236] Overall Loss 0.195997 Objective Loss 0.195997 LR 0.000500 Time 0.021859 +2023-10-02 21:25:07,237 - Epoch: [116][ 1180/ 1236] Overall Loss 0.195942 Objective Loss 0.195942 LR 0.000500 Time 0.021852 +2023-10-02 21:25:07,451 - Epoch: [116][ 1190/ 1236] Overall Loss 0.196039 Objective Loss 0.196039 LR 0.000500 Time 0.021848 +2023-10-02 21:25:07,662 - Epoch: [116][ 1200/ 1236] Overall Loss 0.196044 Objective Loss 0.196044 LR 0.000500 Time 0.021841 +2023-10-02 21:25:07,876 - Epoch: [116][ 1210/ 1236] Overall Loss 0.196198 Objective Loss 0.196198 LR 0.000500 Time 0.021837 +2023-10-02 21:25:08,086 - Epoch: [116][ 1220/ 1236] Overall Loss 0.196282 Objective Loss 0.196282 LR 0.000500 Time 0.021830 +2023-10-02 21:25:08,351 - Epoch: [116][ 1230/ 1236] Overall Loss 0.196266 Objective Loss 0.196266 LR 0.000500 Time 0.021868 +2023-10-02 21:25:08,474 - Epoch: [116][ 1236/ 1236] Overall Loss 0.196140 Objective Loss 0.196140 Top1 89.409369 Top5 100.000000 LR 0.000500 Time 0.021861 +2023-10-02 21:25:08,609 - --- validate (epoch=116)----------- +2023-10-02 21:25:08,609 - 29943 samples (256 per mini-batch) +2023-10-02 21:25:09,106 - Epoch: [116][ 10/ 117] Loss 0.335152 Top1 84.921875 Top5 98.125000 +2023-10-02 21:25:09,258 - Epoch: [116][ 20/ 117] Loss 0.325167 Top1 84.414062 Top5 98.085938 +2023-10-02 21:25:09,409 - Epoch: [116][ 30/ 117] Loss 0.320431 Top1 84.375000 Top5 98.216146 +2023-10-02 21:25:09,560 - Epoch: [116][ 40/ 117] Loss 0.318354 Top1 84.550781 Top5 98.203125 +2023-10-02 21:25:09,712 - Epoch: [116][ 50/ 117] Loss 0.314412 Top1 84.585938 Top5 98.257812 +2023-10-02 21:25:09,863 - Epoch: [116][ 60/ 117] Loss 0.310152 Top1 84.635417 Top5 98.372396 +2023-10-02 21:25:10,014 - Epoch: [116][ 70/ 117] Loss 0.310889 Top1 84.670759 Top5 98.398438 +2023-10-02 21:25:10,164 - Epoch: [116][ 80/ 117] Loss 0.310368 Top1 84.624023 Top5 98.403320 +2023-10-02 21:25:10,314 - Epoch: [116][ 90/ 117] Loss 0.306253 Top1 84.809028 Top5 98.441840 +2023-10-02 21:25:10,464 - Epoch: [116][ 100/ 117] Loss 0.305249 Top1 84.890625 Top5 98.445312 +2023-10-02 21:25:10,622 - Epoch: [116][ 110/ 117] Loss 0.302077 Top1 85.053267 Top5 98.462358 +2023-10-02 21:25:10,710 - Epoch: [116][ 117/ 117] Loss 0.299225 Top1 85.181846 Top5 98.457068 +2023-10-02 21:25:10,850 - ==> Top1: 85.182 Top5: 98.457 Loss: 0.299 + +2023-10-02 21:25:10,850 - ==> Confusion: +[[ 929 1 2 0 6 2 0 0 7 64 1 3 1 3 5 1 4 1 2 0 18] + [ 0 1031 1 2 5 35 0 27 1 1 1 0 1 1 0 2 3 1 9 2 8] + [ 3 0 979 11 1 0 17 4 0 0 2 0 8 2 0 5 2 2 9 5 6] + [ 0 3 8 973 0 6 1 3 2 0 5 0 10 5 31 4 1 4 17 0 16] + [ 31 8 0 0 961 3 0 0 1 11 1 0 1 4 7 7 6 0 1 1 7] + [ 2 25 0 0 0 1009 0 22 2 4 3 5 3 14 6 2 2 1 5 2 9] + [ 1 3 22 2 0 3 1123 5 0 0 6 0 0 1 0 7 0 0 0 12 6] + [ 1 10 9 2 4 27 4 1068 2 3 6 9 5 3 1 0 0 3 43 8 10] + [ 13 0 0 1 1 4 0 0 974 46 8 3 4 12 12 2 3 1 5 0 0] + [ 87 0 2 0 4 4 0 0 31 956 0 0 0 22 5 0 1 0 0 2 5] + [ 3 1 4 8 2 3 7 5 18 1 957 2 1 16 4 1 2 2 6 1 9] + [ 0 2 1 0 0 14 0 1 0 0 0 963 23 7 0 1 1 16 0 2 4] + [ 0 0 3 5 0 1 2 1 0 0 4 43 971 0 2 8 0 11 3 3 11] + [ 0 0 1 0 0 4 0 0 12 14 2 4 0 1065 5 2 0 1 0 2 7] + [ 10 1 5 17 8 1 0 0 26 3 1 1 2 4 1002 0 1 3 8 0 8] + [ 0 0 1 0 5 0 0 0 0 0 0 7 7 0 1 1074 14 14 3 4 4] + [ 1 19 1 0 4 8 0 0 2 0 0 4 0 3 6 12 1079 0 1 6 15] + [ 0 0 0 3 1 0 3 0 0 1 0 4 26 2 4 5 1 984 0 0 4] + [ 1 6 2 18 1 1 0 18 5 0 3 2 0 0 12 1 0 1 987 0 10] + [ 0 2 1 2 1 4 9 3 0 0 0 18 4 0 0 3 5 2 0 1092 6] + [ 120 143 105 96 65 186 33 78 90 75 154 117 333 337 131 64 63 61 154 171 5329]] + +2023-10-02 21:25:10,852 - ==> Best [Top1: 85.603 Top5: 98.474 Sparsity:0.00 Params: 169472 on epoch: 114] +2023-10-02 21:25:10,852 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:25:10,858 - + +2023-10-02 21:25:10,858 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:25:11,866 - Epoch: [117][ 10/ 1236] Overall Loss 0.180886 Objective Loss 0.180886 LR 0.000500 Time 0.100729 +2023-10-02 21:25:12,077 - Epoch: [117][ 20/ 1236] Overall Loss 0.185155 Objective Loss 0.185155 LR 0.000500 Time 0.060892 +2023-10-02 21:25:12,286 - Epoch: [117][ 30/ 1236] Overall Loss 0.181201 Objective Loss 0.181201 LR 0.000500 Time 0.047512 +2023-10-02 21:25:12,497 - Epoch: [117][ 40/ 1236] Overall Loss 0.187499 Objective Loss 0.187499 LR 0.000500 Time 0.040911 +2023-10-02 21:25:12,705 - Epoch: [117][ 50/ 1236] Overall Loss 0.185832 Objective Loss 0.185832 LR 0.000500 Time 0.036879 +2023-10-02 21:25:12,916 - Epoch: [117][ 60/ 1236] Overall Loss 0.189379 Objective Loss 0.189379 LR 0.000500 Time 0.034234 +2023-10-02 21:25:13,124 - Epoch: [117][ 70/ 1236] Overall Loss 0.188374 Objective Loss 0.188374 LR 0.000500 Time 0.032294 +2023-10-02 21:25:13,334 - Epoch: [117][ 80/ 1236] Overall Loss 0.189175 Objective Loss 0.189175 LR 0.000500 Time 0.030878 +2023-10-02 21:25:13,542 - Epoch: [117][ 90/ 1236] Overall Loss 0.189014 Objective Loss 0.189014 LR 0.000500 Time 0.029760 +2023-10-02 21:25:13,754 - Epoch: [117][ 100/ 1236] Overall Loss 0.187932 Objective Loss 0.187932 LR 0.000500 Time 0.028899 +2023-10-02 21:25:13,961 - Epoch: [117][ 110/ 1236] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000500 Time 0.028150 +2023-10-02 21:25:14,176 - Epoch: [117][ 120/ 1236] Overall Loss 0.190370 Objective Loss 0.190370 LR 0.000500 Time 0.027584 +2023-10-02 21:25:14,388 - Epoch: [117][ 130/ 1236] Overall Loss 0.191806 Objective Loss 0.191806 LR 0.000500 Time 0.027089 +2023-10-02 21:25:14,601 - Epoch: [117][ 140/ 1236] Overall Loss 0.191991 Objective Loss 0.191991 LR 0.000500 Time 0.026668 +2023-10-02 21:25:14,813 - Epoch: [117][ 150/ 1236] Overall Loss 0.191816 Objective Loss 0.191816 LR 0.000500 Time 0.026293 +2023-10-02 21:25:15,027 - Epoch: [117][ 160/ 1236] Overall Loss 0.192237 Objective Loss 0.192237 LR 0.000500 Time 0.025976 +2023-10-02 21:25:15,241 - Epoch: [117][ 170/ 1236] Overall Loss 0.193577 Objective Loss 0.193577 LR 0.000500 Time 0.025691 +2023-10-02 21:25:15,455 - Epoch: [117][ 180/ 1236] Overall Loss 0.192406 Objective Loss 0.192406 LR 0.000500 Time 0.025444 +2023-10-02 21:25:15,668 - Epoch: [117][ 190/ 1236] Overall Loss 0.191687 Objective Loss 0.191687 LR 0.000500 Time 0.025216 +2023-10-02 21:25:15,882 - Epoch: [117][ 200/ 1236] Overall Loss 0.191941 Objective Loss 0.191941 LR 0.000500 Time 0.025016 +2023-10-02 21:25:16,095 - Epoch: [117][ 210/ 1236] Overall Loss 0.191558 Objective Loss 0.191558 LR 0.000500 Time 0.024831 +2023-10-02 21:25:16,309 - Epoch: [117][ 220/ 1236] Overall Loss 0.190909 Objective Loss 0.190909 LR 0.000500 Time 0.024668 +2023-10-02 21:25:16,522 - Epoch: [117][ 230/ 1236] Overall Loss 0.191137 Objective Loss 0.191137 LR 0.000500 Time 0.024513 +2023-10-02 21:25:16,736 - Epoch: [117][ 240/ 1236] Overall Loss 0.191004 Objective Loss 0.191004 LR 0.000500 Time 0.024375 +2023-10-02 21:25:16,949 - Epoch: [117][ 250/ 1236] Overall Loss 0.191962 Objective Loss 0.191962 LR 0.000500 Time 0.024245 +2023-10-02 21:25:17,162 - Epoch: [117][ 260/ 1236] Overall Loss 0.192799 Objective Loss 0.192799 LR 0.000500 Time 0.024129 +2023-10-02 21:25:17,375 - Epoch: [117][ 270/ 1236] Overall Loss 0.193044 Objective Loss 0.193044 LR 0.000500 Time 0.024016 +2023-10-02 21:25:17,589 - Epoch: [117][ 280/ 1236] Overall Loss 0.192673 Objective Loss 0.192673 LR 0.000500 Time 0.023916 +2023-10-02 21:25:17,802 - Epoch: [117][ 290/ 1236] Overall Loss 0.192812 Objective Loss 0.192812 LR 0.000500 Time 0.023819 +2023-10-02 21:25:18,015 - Epoch: [117][ 300/ 1236] Overall Loss 0.192308 Objective Loss 0.192308 LR 0.000500 Time 0.023730 +2023-10-02 21:25:18,225 - Epoch: [117][ 310/ 1236] Overall Loss 0.191788 Objective Loss 0.191788 LR 0.000500 Time 0.023642 +2023-10-02 21:25:18,436 - Epoch: [117][ 320/ 1236] Overall Loss 0.192829 Objective Loss 0.192829 LR 0.000500 Time 0.023561 +2023-10-02 21:25:18,644 - Epoch: [117][ 330/ 1236] Overall Loss 0.193527 Objective Loss 0.193527 LR 0.000500 Time 0.023477 +2023-10-02 21:25:18,855 - Epoch: [117][ 340/ 1236] Overall Loss 0.193425 Objective Loss 0.193425 LR 0.000500 Time 0.023405 +2023-10-02 21:25:19,065 - Epoch: [117][ 350/ 1236] Overall Loss 0.193004 Objective Loss 0.193004 LR 0.000500 Time 0.023334 +2023-10-02 21:25:19,278 - Epoch: [117][ 360/ 1236] Overall Loss 0.192928 Objective Loss 0.192928 LR 0.000500 Time 0.023275 +2023-10-02 21:25:19,485 - Epoch: [117][ 370/ 1236] Overall Loss 0.193126 Objective Loss 0.193126 LR 0.000500 Time 0.023205 +2023-10-02 21:25:19,696 - Epoch: [117][ 380/ 1236] Overall Loss 0.193072 Objective Loss 0.193072 LR 0.000500 Time 0.023148 +2023-10-02 21:25:19,904 - Epoch: [117][ 390/ 1236] Overall Loss 0.192922 Objective Loss 0.192922 LR 0.000500 Time 0.023089 +2023-10-02 21:25:20,116 - Epoch: [117][ 400/ 1236] Overall Loss 0.193585 Objective Loss 0.193585 LR 0.000500 Time 0.023041 +2023-10-02 21:25:20,325 - Epoch: [117][ 410/ 1236] Overall Loss 0.193646 Objective Loss 0.193646 LR 0.000500 Time 0.022986 +2023-10-02 21:25:20,535 - Epoch: [117][ 420/ 1236] Overall Loss 0.192945 Objective Loss 0.192945 LR 0.000500 Time 0.022939 +2023-10-02 21:25:20,743 - Epoch: [117][ 430/ 1236] Overall Loss 0.192997 Objective Loss 0.192997 LR 0.000500 Time 0.022887 +2023-10-02 21:25:20,952 - Epoch: [117][ 440/ 1236] Overall Loss 0.192859 Objective Loss 0.192859 LR 0.000500 Time 0.022842 +2023-10-02 21:25:21,159 - Epoch: [117][ 450/ 1236] Overall Loss 0.192746 Objective Loss 0.192746 LR 0.000500 Time 0.022794 +2023-10-02 21:25:21,368 - Epoch: [117][ 460/ 1236] Overall Loss 0.193697 Objective Loss 0.193697 LR 0.000500 Time 0.022753 +2023-10-02 21:25:21,575 - Epoch: [117][ 470/ 1236] Overall Loss 0.193578 Objective Loss 0.193578 LR 0.000500 Time 0.022709 +2023-10-02 21:25:21,785 - Epoch: [117][ 480/ 1236] Overall Loss 0.193398 Objective Loss 0.193398 LR 0.000500 Time 0.022671 +2023-10-02 21:25:21,992 - Epoch: [117][ 490/ 1236] Overall Loss 0.193113 Objective Loss 0.193113 LR 0.000500 Time 0.022630 +2023-10-02 21:25:22,201 - Epoch: [117][ 500/ 1236] Overall Loss 0.193020 Objective Loss 0.193020 LR 0.000500 Time 0.022596 +2023-10-02 21:25:22,408 - Epoch: [117][ 510/ 1236] Overall Loss 0.192696 Objective Loss 0.192696 LR 0.000500 Time 0.022558 +2023-10-02 21:25:22,618 - Epoch: [117][ 520/ 1236] Overall Loss 0.192667 Objective Loss 0.192667 LR 0.000500 Time 0.022527 +2023-10-02 21:25:22,825 - Epoch: [117][ 530/ 1236] Overall Loss 0.192995 Objective Loss 0.192995 LR 0.000500 Time 0.022492 +2023-10-02 21:25:23,035 - Epoch: [117][ 540/ 1236] Overall Loss 0.193016 Objective Loss 0.193016 LR 0.000500 Time 0.022463 +2023-10-02 21:25:23,242 - Epoch: [117][ 550/ 1236] Overall Loss 0.192934 Objective Loss 0.192934 LR 0.000500 Time 0.022430 +2023-10-02 21:25:23,451 - Epoch: [117][ 560/ 1236] Overall Loss 0.193290 Objective Loss 0.193290 LR 0.000500 Time 0.022403 +2023-10-02 21:25:23,658 - Epoch: [117][ 570/ 1236] Overall Loss 0.193633 Objective Loss 0.193633 LR 0.000500 Time 0.022373 +2023-10-02 21:25:23,867 - Epoch: [117][ 580/ 1236] Overall Loss 0.193868 Objective Loss 0.193868 LR 0.000500 Time 0.022347 +2023-10-02 21:25:24,074 - Epoch: [117][ 590/ 1236] Overall Loss 0.194200 Objective Loss 0.194200 LR 0.000500 Time 0.022318 +2023-10-02 21:25:24,284 - Epoch: [117][ 600/ 1236] Overall Loss 0.194111 Objective Loss 0.194111 LR 0.000500 Time 0.022295 +2023-10-02 21:25:24,491 - Epoch: [117][ 610/ 1236] Overall Loss 0.194331 Objective Loss 0.194331 LR 0.000500 Time 0.022268 +2023-10-02 21:25:24,700 - Epoch: [117][ 620/ 1236] Overall Loss 0.194211 Objective Loss 0.194211 LR 0.000500 Time 0.022247 +2023-10-02 21:25:24,907 - Epoch: [117][ 630/ 1236] Overall Loss 0.194135 Objective Loss 0.194135 LR 0.000500 Time 0.022222 +2023-10-02 21:25:25,117 - Epoch: [117][ 640/ 1236] Overall Loss 0.193963 Objective Loss 0.193963 LR 0.000500 Time 0.022201 +2023-10-02 21:25:25,324 - Epoch: [117][ 650/ 1236] Overall Loss 0.193880 Objective Loss 0.193880 LR 0.000500 Time 0.022178 +2023-10-02 21:25:25,533 - Epoch: [117][ 660/ 1236] Overall Loss 0.193489 Objective Loss 0.193489 LR 0.000500 Time 0.022158 +2023-10-02 21:25:25,740 - Epoch: [117][ 670/ 1236] Overall Loss 0.193320 Objective Loss 0.193320 LR 0.000500 Time 0.022136 +2023-10-02 21:25:25,949 - Epoch: [117][ 680/ 1236] Overall Loss 0.193350 Objective Loss 0.193350 LR 0.000500 Time 0.022118 +2023-10-02 21:25:26,157 - Epoch: [117][ 690/ 1236] Overall Loss 0.193633 Objective Loss 0.193633 LR 0.000500 Time 0.022097 +2023-10-02 21:25:26,366 - Epoch: [117][ 700/ 1236] Overall Loss 0.193364 Objective Loss 0.193364 LR 0.000500 Time 0.022080 +2023-10-02 21:25:26,573 - Epoch: [117][ 710/ 1236] Overall Loss 0.193322 Objective Loss 0.193322 LR 0.000500 Time 0.022060 +2023-10-02 21:25:26,782 - Epoch: [117][ 720/ 1236] Overall Loss 0.193185 Objective Loss 0.193185 LR 0.000500 Time 0.022044 +2023-10-02 21:25:26,989 - Epoch: [117][ 730/ 1236] Overall Loss 0.193288 Objective Loss 0.193288 LR 0.000500 Time 0.022024 +2023-10-02 21:25:27,198 - Epoch: [117][ 740/ 1236] Overall Loss 0.193175 Objective Loss 0.193175 LR 0.000500 Time 0.022010 +2023-10-02 21:25:27,406 - Epoch: [117][ 750/ 1236] Overall Loss 0.192789 Objective Loss 0.192789 LR 0.000500 Time 0.021992 +2023-10-02 21:25:27,615 - Epoch: [117][ 760/ 1236] Overall Loss 0.192676 Objective Loss 0.192676 LR 0.000500 Time 0.021978 +2023-10-02 21:25:27,822 - Epoch: [117][ 770/ 1236] Overall Loss 0.192388 Objective Loss 0.192388 LR 0.000500 Time 0.021960 +2023-10-02 21:25:28,031 - Epoch: [117][ 780/ 1236] Overall Loss 0.192423 Objective Loss 0.192423 LR 0.000500 Time 0.021947 +2023-10-02 21:25:28,238 - Epoch: [117][ 790/ 1236] Overall Loss 0.192432 Objective Loss 0.192432 LR 0.000500 Time 0.021931 +2023-10-02 21:25:28,448 - Epoch: [117][ 800/ 1236] Overall Loss 0.192557 Objective Loss 0.192557 LR 0.000500 Time 0.021919 +2023-10-02 21:25:28,655 - Epoch: [117][ 810/ 1236] Overall Loss 0.192736 Objective Loss 0.192736 LR 0.000500 Time 0.021904 +2023-10-02 21:25:28,865 - Epoch: [117][ 820/ 1236] Overall Loss 0.192717 Objective Loss 0.192717 LR 0.000500 Time 0.021892 +2023-10-02 21:25:29,072 - Epoch: [117][ 830/ 1236] Overall Loss 0.192919 Objective Loss 0.192919 LR 0.000500 Time 0.021877 +2023-10-02 21:25:29,281 - Epoch: [117][ 840/ 1236] Overall Loss 0.192942 Objective Loss 0.192942 LR 0.000500 Time 0.021865 +2023-10-02 21:25:29,488 - Epoch: [117][ 850/ 1236] Overall Loss 0.193228 Objective Loss 0.193228 LR 0.000500 Time 0.021851 +2023-10-02 21:25:29,698 - Epoch: [117][ 860/ 1236] Overall Loss 0.193187 Objective Loss 0.193187 LR 0.000500 Time 0.021840 +2023-10-02 21:25:29,905 - Epoch: [117][ 870/ 1236] Overall Loss 0.193082 Objective Loss 0.193082 LR 0.000500 Time 0.021827 +2023-10-02 21:25:30,114 - Epoch: [117][ 880/ 1236] Overall Loss 0.193217 Objective Loss 0.193217 LR 0.000500 Time 0.021817 +2023-10-02 21:25:30,321 - Epoch: [117][ 890/ 1236] Overall Loss 0.193203 Objective Loss 0.193203 LR 0.000500 Time 0.021804 +2023-10-02 21:25:30,531 - Epoch: [117][ 900/ 1236] Overall Loss 0.193303 Objective Loss 0.193303 LR 0.000500 Time 0.021794 +2023-10-02 21:25:30,738 - Epoch: [117][ 910/ 1236] Overall Loss 0.193199 Objective Loss 0.193199 LR 0.000500 Time 0.021782 +2023-10-02 21:25:30,948 - Epoch: [117][ 920/ 1236] Overall Loss 0.193243 Objective Loss 0.193243 LR 0.000500 Time 0.021772 +2023-10-02 21:25:31,154 - Epoch: [117][ 930/ 1236] Overall Loss 0.193446 Objective Loss 0.193446 LR 0.000500 Time 0.021760 +2023-10-02 21:25:31,364 - Epoch: [117][ 940/ 1236] Overall Loss 0.193516 Objective Loss 0.193516 LR 0.000500 Time 0.021751 +2023-10-02 21:25:31,571 - Epoch: [117][ 950/ 1236] Overall Loss 0.193350 Objective Loss 0.193350 LR 0.000500 Time 0.021740 +2023-10-02 21:25:31,781 - Epoch: [117][ 960/ 1236] Overall Loss 0.193337 Objective Loss 0.193337 LR 0.000500 Time 0.021732 +2023-10-02 21:25:31,988 - Epoch: [117][ 970/ 1236] Overall Loss 0.193371 Objective Loss 0.193371 LR 0.000500 Time 0.021721 +2023-10-02 21:25:32,197 - Epoch: [117][ 980/ 1236] Overall Loss 0.193284 Objective Loss 0.193284 LR 0.000500 Time 0.021713 +2023-10-02 21:25:32,404 - Epoch: [117][ 990/ 1236] Overall Loss 0.193547 Objective Loss 0.193547 LR 0.000500 Time 0.021702 +2023-10-02 21:25:32,614 - Epoch: [117][ 1000/ 1236] Overall Loss 0.193658 Objective Loss 0.193658 LR 0.000500 Time 0.021694 +2023-10-02 21:25:32,821 - Epoch: [117][ 1010/ 1236] Overall Loss 0.193558 Objective Loss 0.193558 LR 0.000500 Time 0.021684 +2023-10-02 21:25:33,030 - Epoch: [117][ 1020/ 1236] Overall Loss 0.193642 Objective Loss 0.193642 LR 0.000500 Time 0.021677 +2023-10-02 21:25:33,237 - Epoch: [117][ 1030/ 1236] Overall Loss 0.193588 Objective Loss 0.193588 LR 0.000500 Time 0.021667 +2023-10-02 21:25:33,447 - Epoch: [117][ 1040/ 1236] Overall Loss 0.193323 Objective Loss 0.193323 LR 0.000500 Time 0.021660 +2023-10-02 21:25:33,654 - Epoch: [117][ 1050/ 1236] Overall Loss 0.193408 Objective Loss 0.193408 LR 0.000500 Time 0.021650 +2023-10-02 21:25:33,864 - Epoch: [117][ 1060/ 1236] Overall Loss 0.193344 Objective Loss 0.193344 LR 0.000500 Time 0.021644 +2023-10-02 21:25:34,071 - Epoch: [117][ 1070/ 1236] Overall Loss 0.193402 Objective Loss 0.193402 LR 0.000500 Time 0.021635 +2023-10-02 21:25:34,281 - Epoch: [117][ 1080/ 1236] Overall Loss 0.193675 Objective Loss 0.193675 LR 0.000500 Time 0.021628 +2023-10-02 21:25:34,488 - Epoch: [117][ 1090/ 1236] Overall Loss 0.193586 Objective Loss 0.193586 LR 0.000500 Time 0.021619 +2023-10-02 21:25:34,697 - Epoch: [117][ 1100/ 1236] Overall Loss 0.193651 Objective Loss 0.193651 LR 0.000500 Time 0.021613 +2023-10-02 21:25:34,904 - Epoch: [117][ 1110/ 1236] Overall Loss 0.193652 Objective Loss 0.193652 LR 0.000500 Time 0.021605 +2023-10-02 21:25:35,114 - Epoch: [117][ 1120/ 1236] Overall Loss 0.193709 Objective Loss 0.193709 LR 0.000500 Time 0.021599 +2023-10-02 21:25:35,321 - Epoch: [117][ 1130/ 1236] Overall Loss 0.193705 Objective Loss 0.193705 LR 0.000500 Time 0.021591 +2023-10-02 21:25:35,531 - Epoch: [117][ 1140/ 1236] Overall Loss 0.193812 Objective Loss 0.193812 LR 0.000500 Time 0.021585 +2023-10-02 21:25:35,738 - Epoch: [117][ 1150/ 1236] Overall Loss 0.193686 Objective Loss 0.193686 LR 0.000500 Time 0.021577 +2023-10-02 21:25:35,947 - Epoch: [117][ 1160/ 1236] Overall Loss 0.193583 Objective Loss 0.193583 LR 0.000500 Time 0.021571 +2023-10-02 21:25:36,155 - Epoch: [117][ 1170/ 1236] Overall Loss 0.193557 Objective Loss 0.193557 LR 0.000500 Time 0.021563 +2023-10-02 21:25:36,364 - Epoch: [117][ 1180/ 1236] Overall Loss 0.193467 Objective Loss 0.193467 LR 0.000500 Time 0.021558 +2023-10-02 21:25:36,572 - Epoch: [117][ 1190/ 1236] Overall Loss 0.193365 Objective Loss 0.193365 LR 0.000500 Time 0.021551 +2023-10-02 21:25:36,781 - Epoch: [117][ 1200/ 1236] Overall Loss 0.193363 Objective Loss 0.193363 LR 0.000500 Time 0.021546 +2023-10-02 21:25:36,988 - Epoch: [117][ 1210/ 1236] Overall Loss 0.193189 Objective Loss 0.193189 LR 0.000500 Time 0.021539 +2023-10-02 21:25:37,198 - Epoch: [117][ 1220/ 1236] Overall Loss 0.192960 Objective Loss 0.192960 LR 0.000500 Time 0.021533 +2023-10-02 21:25:37,459 - Epoch: [117][ 1230/ 1236] Overall Loss 0.192938 Objective Loss 0.192938 LR 0.000500 Time 0.021571 +2023-10-02 21:25:37,582 - Epoch: [117][ 1236/ 1236] Overall Loss 0.192987 Objective Loss 0.192987 Top1 89.409369 Top5 99.389002 LR 0.000500 Time 0.021565 +2023-10-02 21:25:37,716 - --- validate (epoch=117)----------- +2023-10-02 21:25:37,716 - 29943 samples (256 per mini-batch) +2023-10-02 21:25:38,200 - Epoch: [117][ 10/ 117] Loss 0.290074 Top1 86.914062 Top5 98.554688 +2023-10-02 21:25:38,351 - Epoch: [117][ 20/ 117] Loss 0.280415 Top1 86.484375 Top5 98.750000 +2023-10-02 21:25:38,503 - Epoch: [117][ 30/ 117] Loss 0.292911 Top1 85.963542 Top5 98.541667 +2023-10-02 21:25:38,655 - Epoch: [117][ 40/ 117] Loss 0.296803 Top1 85.761719 Top5 98.535156 +2023-10-02 21:25:38,807 - Epoch: [117][ 50/ 117] Loss 0.292919 Top1 85.812500 Top5 98.546875 +2023-10-02 21:25:38,959 - Epoch: [117][ 60/ 117] Loss 0.297160 Top1 85.638021 Top5 98.528646 +2023-10-02 21:25:39,111 - Epoch: [117][ 70/ 117] Loss 0.304081 Top1 85.569196 Top5 98.476562 +2023-10-02 21:25:39,264 - Epoch: [117][ 80/ 117] Loss 0.301129 Top1 85.668945 Top5 98.481445 +2023-10-02 21:25:39,415 - Epoch: [117][ 90/ 117] Loss 0.300647 Top1 85.638021 Top5 98.467882 +2023-10-02 21:25:39,568 - Epoch: [117][ 100/ 117] Loss 0.299868 Top1 85.601562 Top5 98.496094 +2023-10-02 21:25:39,727 - Epoch: [117][ 110/ 117] Loss 0.301309 Top1 85.550426 Top5 98.487216 +2023-10-02 21:25:39,817 - Epoch: [117][ 117/ 117] Loss 0.301280 Top1 85.535851 Top5 98.493805 +2023-10-02 21:25:39,961 - ==> Top1: 85.536 Top5: 98.494 Loss: 0.301 + +2023-10-02 21:25:39,962 - ==> Confusion: +[[ 963 1 4 0 5 2 0 0 5 39 1 1 1 5 5 0 2 1 1 0 14] + [ 0 1064 0 0 2 17 1 25 1 1 3 1 2 0 0 3 0 0 7 1 3] + [ 2 0 983 3 1 0 19 11 0 2 1 1 9 1 0 4 2 2 8 3 4] + [ 0 2 15 968 1 3 3 3 3 1 4 0 9 4 27 3 0 5 19 0 19] + [ 30 8 1 0 965 3 1 0 1 12 0 0 1 5 7 4 6 0 0 2 4] + [ 4 40 1 0 1 995 2 22 2 5 1 2 5 10 4 0 3 0 7 1 11] + [ 1 1 18 0 0 1 1135 11 0 0 3 0 0 0 1 4 0 0 3 6 7] + [ 1 10 5 0 8 22 6 1107 1 1 2 4 5 4 2 0 1 2 25 3 9] + [ 19 2 0 0 2 2 0 0 956 45 9 0 2 18 20 3 3 0 6 0 2] + [ 114 2 2 0 2 1 1 0 19 935 0 0 1 22 8 1 1 1 0 1 8] + [ 4 6 5 6 1 2 4 7 12 2 962 0 0 14 5 1 3 2 6 0 11] + [ 0 1 2 0 1 8 0 5 0 0 0 943 38 13 1 0 1 17 0 3 2] + [ 0 0 3 4 0 2 2 3 0 3 1 29 979 2 2 6 0 15 4 5 8] + [ 1 0 1 0 1 5 2 1 6 7 1 5 0 1071 3 0 0 1 0 2 12] + [ 14 2 4 13 11 0 0 0 18 4 2 0 4 2 1008 0 1 3 5 0 10] + [ 0 0 2 1 5 0 0 0 0 0 0 5 9 0 2 1065 18 12 3 5 7] + [ 1 20 0 0 4 7 0 0 2 0 0 5 1 3 3 5 1085 0 1 7 17] + [ 0 0 0 0 0 0 3 1 0 0 0 4 15 1 3 8 0 1000 0 0 3] + [ 1 5 3 11 2 0 0 25 5 1 1 0 2 0 11 1 0 0 989 0 11] + [ 0 3 5 1 1 5 9 6 0 1 0 14 2 3 2 0 7 2 0 1081 10] + [ 148 206 113 67 82 157 49 131 85 62 144 83 347 287 112 46 69 71 134 154 5358]] + +2023-10-02 21:25:39,963 - ==> Best [Top1: 85.603 Top5: 98.474 Sparsity:0.00 Params: 169472 on epoch: 114] +2023-10-02 21:25:39,963 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:25:39,969 - + +2023-10-02 21:25:39,969 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:25:40,993 - Epoch: [118][ 10/ 1236] Overall Loss 0.174667 Objective Loss 0.174667 LR 0.000500 Time 0.102307 +2023-10-02 21:25:41,201 - Epoch: [118][ 20/ 1236] Overall Loss 0.181771 Objective Loss 0.181771 LR 0.000500 Time 0.061549 +2023-10-02 21:25:41,409 - Epoch: [118][ 30/ 1236] Overall Loss 0.180284 Objective Loss 0.180284 LR 0.000500 Time 0.047936 +2023-10-02 21:25:41,617 - Epoch: [118][ 40/ 1236] Overall Loss 0.185565 Objective Loss 0.185565 LR 0.000500 Time 0.041136 +2023-10-02 21:25:41,824 - Epoch: [118][ 50/ 1236] Overall Loss 0.183838 Objective Loss 0.183838 LR 0.000500 Time 0.037023 +2023-10-02 21:25:42,032 - Epoch: [118][ 60/ 1236] Overall Loss 0.182805 Objective Loss 0.182805 LR 0.000500 Time 0.034313 +2023-10-02 21:25:42,240 - Epoch: [118][ 70/ 1236] Overall Loss 0.181442 Objective Loss 0.181442 LR 0.000500 Time 0.032375 +2023-10-02 21:25:42,448 - Epoch: [118][ 80/ 1236] Overall Loss 0.180290 Objective Loss 0.180290 LR 0.000500 Time 0.030925 +2023-10-02 21:25:42,655 - Epoch: [118][ 90/ 1236] Overall Loss 0.181349 Objective Loss 0.181349 LR 0.000500 Time 0.029778 +2023-10-02 21:25:42,864 - Epoch: [118][ 100/ 1236] Overall Loss 0.181907 Objective Loss 0.181907 LR 0.000500 Time 0.028880 +2023-10-02 21:25:43,071 - Epoch: [118][ 110/ 1236] Overall Loss 0.182907 Objective Loss 0.182907 LR 0.000500 Time 0.028134 +2023-10-02 21:25:43,279 - Epoch: [118][ 120/ 1236] Overall Loss 0.183966 Objective Loss 0.183966 LR 0.000500 Time 0.027522 +2023-10-02 21:25:43,486 - Epoch: [118][ 130/ 1236] Overall Loss 0.183944 Objective Loss 0.183944 LR 0.000500 Time 0.026986 +2023-10-02 21:25:43,695 - Epoch: [118][ 140/ 1236] Overall Loss 0.184249 Objective Loss 0.184249 LR 0.000500 Time 0.026546 +2023-10-02 21:25:43,902 - Epoch: [118][ 150/ 1236] Overall Loss 0.185518 Objective Loss 0.185518 LR 0.000500 Time 0.026147 +2023-10-02 21:25:44,110 - Epoch: [118][ 160/ 1236] Overall Loss 0.185162 Objective Loss 0.185162 LR 0.000500 Time 0.025814 +2023-10-02 21:25:44,317 - Epoch: [118][ 170/ 1236] Overall Loss 0.185721 Objective Loss 0.185721 LR 0.000500 Time 0.025505 +2023-10-02 21:25:44,526 - Epoch: [118][ 180/ 1236] Overall Loss 0.186103 Objective Loss 0.186103 LR 0.000500 Time 0.025244 +2023-10-02 21:25:44,733 - Epoch: [118][ 190/ 1236] Overall Loss 0.185472 Objective Loss 0.185472 LR 0.000500 Time 0.024997 +2023-10-02 21:25:44,941 - Epoch: [118][ 200/ 1236] Overall Loss 0.186542 Objective Loss 0.186542 LR 0.000500 Time 0.024788 +2023-10-02 21:25:45,148 - Epoch: [118][ 210/ 1236] Overall Loss 0.187350 Objective Loss 0.187350 LR 0.000500 Time 0.024583 +2023-10-02 21:25:45,357 - Epoch: [118][ 220/ 1236] Overall Loss 0.187354 Objective Loss 0.187354 LR 0.000500 Time 0.024414 +2023-10-02 21:25:45,564 - Epoch: [118][ 230/ 1236] Overall Loss 0.187534 Objective Loss 0.187534 LR 0.000500 Time 0.024247 +2023-10-02 21:25:45,771 - Epoch: [118][ 240/ 1236] Overall Loss 0.187540 Objective Loss 0.187540 LR 0.000500 Time 0.024100 +2023-10-02 21:25:45,979 - Epoch: [118][ 250/ 1236] Overall Loss 0.187726 Objective Loss 0.187726 LR 0.000500 Time 0.023963 +2023-10-02 21:25:46,187 - Epoch: [118][ 260/ 1236] Overall Loss 0.188497 Objective Loss 0.188497 LR 0.000500 Time 0.023844 +2023-10-02 21:25:46,395 - Epoch: [118][ 270/ 1236] Overall Loss 0.188402 Objective Loss 0.188402 LR 0.000500 Time 0.023722 +2023-10-02 21:25:46,603 - Epoch: [118][ 280/ 1236] Overall Loss 0.188464 Objective Loss 0.188464 LR 0.000500 Time 0.023619 +2023-10-02 21:25:46,810 - Epoch: [118][ 290/ 1236] Overall Loss 0.189149 Objective Loss 0.189149 LR 0.000500 Time 0.023513 +2023-10-02 21:25:47,019 - Epoch: [118][ 300/ 1236] Overall Loss 0.188648 Objective Loss 0.188648 LR 0.000500 Time 0.023424 +2023-10-02 21:25:47,226 - Epoch: [118][ 310/ 1236] Overall Loss 0.188408 Objective Loss 0.188408 LR 0.000500 Time 0.023332 +2023-10-02 21:25:47,435 - Epoch: [118][ 320/ 1236] Overall Loss 0.188625 Objective Loss 0.188625 LR 0.000500 Time 0.023254 +2023-10-02 21:25:47,642 - Epoch: [118][ 330/ 1236] Overall Loss 0.188962 Objective Loss 0.188962 LR 0.000500 Time 0.023172 +2023-10-02 21:25:47,851 - Epoch: [118][ 340/ 1236] Overall Loss 0.189468 Objective Loss 0.189468 LR 0.000500 Time 0.023104 +2023-10-02 21:25:48,058 - Epoch: [118][ 350/ 1236] Overall Loss 0.189268 Objective Loss 0.189268 LR 0.000500 Time 0.023031 +2023-10-02 21:25:48,267 - Epoch: [118][ 360/ 1236] Overall Loss 0.189563 Objective Loss 0.189563 LR 0.000500 Time 0.022970 +2023-10-02 21:25:48,474 - Epoch: [118][ 370/ 1236] Overall Loss 0.189894 Objective Loss 0.189894 LR 0.000500 Time 0.022905 +2023-10-02 21:25:48,683 - Epoch: [118][ 380/ 1236] Overall Loss 0.189784 Objective Loss 0.189784 LR 0.000500 Time 0.022851 +2023-10-02 21:25:48,890 - Epoch: [118][ 390/ 1236] Overall Loss 0.189658 Objective Loss 0.189658 LR 0.000500 Time 0.022796 +2023-10-02 21:25:49,099 - Epoch: [118][ 400/ 1236] Overall Loss 0.189957 Objective Loss 0.189957 LR 0.000500 Time 0.022747 +2023-10-02 21:25:49,306 - Epoch: [118][ 410/ 1236] Overall Loss 0.190637 Objective Loss 0.190637 LR 0.000500 Time 0.022697 +2023-10-02 21:25:49,515 - Epoch: [118][ 420/ 1236] Overall Loss 0.190191 Objective Loss 0.190191 LR 0.000500 Time 0.022653 +2023-10-02 21:25:49,722 - Epoch: [118][ 430/ 1236] Overall Loss 0.190010 Objective Loss 0.190010 LR 0.000500 Time 0.022608 +2023-10-02 21:25:49,931 - Epoch: [118][ 440/ 1236] Overall Loss 0.189373 Objective Loss 0.189373 LR 0.000500 Time 0.022567 +2023-10-02 21:25:50,138 - Epoch: [118][ 450/ 1236] Overall Loss 0.189291 Objective Loss 0.189291 LR 0.000500 Time 0.022527 +2023-10-02 21:25:50,347 - Epoch: [118][ 460/ 1236] Overall Loss 0.188859 Objective Loss 0.188859 LR 0.000500 Time 0.022489 +2023-10-02 21:25:50,554 - Epoch: [118][ 470/ 1236] Overall Loss 0.189229 Objective Loss 0.189229 LR 0.000500 Time 0.022451 +2023-10-02 21:25:50,763 - Epoch: [118][ 480/ 1236] Overall Loss 0.189496 Objective Loss 0.189496 LR 0.000500 Time 0.022417 +2023-10-02 21:25:50,970 - Epoch: [118][ 490/ 1236] Overall Loss 0.189189 Objective Loss 0.189189 LR 0.000500 Time 0.022383 +2023-10-02 21:25:51,179 - Epoch: [118][ 500/ 1236] Overall Loss 0.189015 Objective Loss 0.189015 LR 0.000500 Time 0.022352 +2023-10-02 21:25:51,386 - Epoch: [118][ 510/ 1236] Overall Loss 0.189181 Objective Loss 0.189181 LR 0.000500 Time 0.022317 +2023-10-02 21:25:51,595 - Epoch: [118][ 520/ 1236] Overall Loss 0.189358 Objective Loss 0.189358 LR 0.000500 Time 0.022288 +2023-10-02 21:25:51,802 - Epoch: [118][ 530/ 1236] Overall Loss 0.189216 Objective Loss 0.189216 LR 0.000500 Time 0.022258 +2023-10-02 21:25:52,011 - Epoch: [118][ 540/ 1236] Overall Loss 0.189174 Objective Loss 0.189174 LR 0.000500 Time 0.022232 +2023-10-02 21:25:52,218 - Epoch: [118][ 550/ 1236] Overall Loss 0.189040 Objective Loss 0.189040 LR 0.000500 Time 0.022202 +2023-10-02 21:25:52,427 - Epoch: [118][ 560/ 1236] Overall Loss 0.188887 Objective Loss 0.188887 LR 0.000500 Time 0.022178 +2023-10-02 21:25:52,634 - Epoch: [118][ 570/ 1236] Overall Loss 0.189180 Objective Loss 0.189180 LR 0.000500 Time 0.022152 +2023-10-02 21:25:52,843 - Epoch: [118][ 580/ 1236] Overall Loss 0.188957 Objective Loss 0.188957 LR 0.000500 Time 0.022129 +2023-10-02 21:25:53,050 - Epoch: [118][ 590/ 1236] Overall Loss 0.188724 Objective Loss 0.188724 LR 0.000500 Time 0.022105 +2023-10-02 21:25:53,259 - Epoch: [118][ 600/ 1236] Overall Loss 0.188453 Objective Loss 0.188453 LR 0.000500 Time 0.022084 +2023-10-02 21:25:53,466 - Epoch: [118][ 610/ 1236] Overall Loss 0.188475 Objective Loss 0.188475 LR 0.000500 Time 0.022061 +2023-10-02 21:25:53,675 - Epoch: [118][ 620/ 1236] Overall Loss 0.188084 Objective Loss 0.188084 LR 0.000500 Time 0.022041 +2023-10-02 21:25:53,883 - Epoch: [118][ 630/ 1236] Overall Loss 0.187901 Objective Loss 0.187901 LR 0.000500 Time 0.022021 +2023-10-02 21:25:54,091 - Epoch: [118][ 640/ 1236] Overall Loss 0.188119 Objective Loss 0.188119 LR 0.000500 Time 0.022002 +2023-10-02 21:25:54,299 - Epoch: [118][ 650/ 1236] Overall Loss 0.188238 Objective Loss 0.188238 LR 0.000500 Time 0.021982 +2023-10-02 21:25:54,507 - Epoch: [118][ 660/ 1236] Overall Loss 0.188226 Objective Loss 0.188226 LR 0.000500 Time 0.021965 +2023-10-02 21:25:54,715 - Epoch: [118][ 670/ 1236] Overall Loss 0.188061 Objective Loss 0.188061 LR 0.000500 Time 0.021947 +2023-10-02 21:25:54,923 - Epoch: [118][ 680/ 1236] Overall Loss 0.187965 Objective Loss 0.187965 LR 0.000500 Time 0.021930 +2023-10-02 21:25:55,131 - Epoch: [118][ 690/ 1236] Overall Loss 0.188015 Objective Loss 0.188015 LR 0.000500 Time 0.021913 +2023-10-02 21:25:55,339 - Epoch: [118][ 700/ 1236] Overall Loss 0.187599 Objective Loss 0.187599 LR 0.000500 Time 0.021896 +2023-10-02 21:25:55,547 - Epoch: [118][ 710/ 1236] Overall Loss 0.187469 Objective Loss 0.187469 LR 0.000500 Time 0.021881 +2023-10-02 21:25:55,757 - Epoch: [118][ 720/ 1236] Overall Loss 0.187720 Objective Loss 0.187720 LR 0.000500 Time 0.021868 +2023-10-02 21:25:55,963 - Epoch: [118][ 730/ 1236] Overall Loss 0.187754 Objective Loss 0.187754 LR 0.000500 Time 0.021850 +2023-10-02 21:25:56,172 - Epoch: [118][ 740/ 1236] Overall Loss 0.187646 Objective Loss 0.187646 LR 0.000500 Time 0.021836 +2023-10-02 21:25:56,379 - Epoch: [118][ 750/ 1236] Overall Loss 0.187407 Objective Loss 0.187407 LR 0.000500 Time 0.021821 +2023-10-02 21:25:56,588 - Epoch: [118][ 760/ 1236] Overall Loss 0.187504 Objective Loss 0.187504 LR 0.000500 Time 0.021808 +2023-10-02 21:25:56,795 - Epoch: [118][ 770/ 1236] Overall Loss 0.187496 Objective Loss 0.187496 LR 0.000500 Time 0.021794 +2023-10-02 21:25:57,004 - Epoch: [118][ 780/ 1236] Overall Loss 0.187450 Objective Loss 0.187450 LR 0.000500 Time 0.021782 +2023-10-02 21:25:57,212 - Epoch: [118][ 790/ 1236] Overall Loss 0.187764 Objective Loss 0.187764 LR 0.000500 Time 0.021769 +2023-10-02 21:25:57,420 - Epoch: [118][ 800/ 1236] Overall Loss 0.188288 Objective Loss 0.188288 LR 0.000500 Time 0.021756 +2023-10-02 21:25:57,628 - Epoch: [118][ 810/ 1236] Overall Loss 0.188382 Objective Loss 0.188382 LR 0.000500 Time 0.021744 +2023-10-02 21:25:57,836 - Epoch: [118][ 820/ 1236] Overall Loss 0.188469 Objective Loss 0.188469 LR 0.000500 Time 0.021733 +2023-10-02 21:25:58,044 - Epoch: [118][ 830/ 1236] Overall Loss 0.188607 Objective Loss 0.188607 LR 0.000500 Time 0.021721 +2023-10-02 21:25:58,254 - Epoch: [118][ 840/ 1236] Overall Loss 0.188805 Objective Loss 0.188805 LR 0.000500 Time 0.021712 +2023-10-02 21:25:58,460 - Epoch: [118][ 850/ 1236] Overall Loss 0.188725 Objective Loss 0.188725 LR 0.000500 Time 0.021699 +2023-10-02 21:25:58,669 - Epoch: [118][ 860/ 1236] Overall Loss 0.188755 Objective Loss 0.188755 LR 0.000500 Time 0.021688 +2023-10-02 21:25:58,876 - Epoch: [118][ 870/ 1236] Overall Loss 0.188753 Objective Loss 0.188753 LR 0.000500 Time 0.021677 +2023-10-02 21:25:59,085 - Epoch: [118][ 880/ 1236] Overall Loss 0.188850 Objective Loss 0.188850 LR 0.000500 Time 0.021668 +2023-10-02 21:25:59,292 - Epoch: [118][ 890/ 1236] Overall Loss 0.188833 Objective Loss 0.188833 LR 0.000500 Time 0.021656 +2023-10-02 21:25:59,501 - Epoch: [118][ 900/ 1236] Overall Loss 0.189079 Objective Loss 0.189079 LR 0.000500 Time 0.021646 +2023-10-02 21:25:59,708 - Epoch: [118][ 910/ 1236] Overall Loss 0.188965 Objective Loss 0.188965 LR 0.000500 Time 0.021636 +2023-10-02 21:25:59,917 - Epoch: [118][ 920/ 1236] Overall Loss 0.189083 Objective Loss 0.189083 LR 0.000500 Time 0.021627 +2023-10-02 21:26:00,125 - Epoch: [118][ 930/ 1236] Overall Loss 0.189129 Objective Loss 0.189129 LR 0.000500 Time 0.021618 +2023-10-02 21:26:00,333 - Epoch: [118][ 940/ 1236] Overall Loss 0.189378 Objective Loss 0.189378 LR 0.000500 Time 0.021609 +2023-10-02 21:26:00,540 - Epoch: [118][ 950/ 1236] Overall Loss 0.189359 Objective Loss 0.189359 LR 0.000500 Time 0.021600 +2023-10-02 21:26:00,749 - Epoch: [118][ 960/ 1236] Overall Loss 0.189506 Objective Loss 0.189506 LR 0.000500 Time 0.021592 +2023-10-02 21:26:00,957 - Epoch: [118][ 970/ 1236] Overall Loss 0.189357 Objective Loss 0.189357 LR 0.000500 Time 0.021583 +2023-10-02 21:26:01,166 - Epoch: [118][ 980/ 1236] Overall Loss 0.189550 Objective Loss 0.189550 LR 0.000500 Time 0.021576 +2023-10-02 21:26:01,373 - Epoch: [118][ 990/ 1236] Overall Loss 0.189196 Objective Loss 0.189196 LR 0.000500 Time 0.021567 +2023-10-02 21:26:01,582 - Epoch: [118][ 1000/ 1236] Overall Loss 0.189192 Objective Loss 0.189192 LR 0.000500 Time 0.021560 +2023-10-02 21:26:01,789 - Epoch: [118][ 1010/ 1236] Overall Loss 0.189289 Objective Loss 0.189289 LR 0.000500 Time 0.021551 +2023-10-02 21:26:01,998 - Epoch: [118][ 1020/ 1236] Overall Loss 0.189329 Objective Loss 0.189329 LR 0.000500 Time 0.021544 +2023-10-02 21:26:02,206 - Epoch: [118][ 1030/ 1236] Overall Loss 0.189213 Objective Loss 0.189213 LR 0.000500 Time 0.021537 +2023-10-02 21:26:02,414 - Epoch: [118][ 1040/ 1236] Overall Loss 0.189337 Objective Loss 0.189337 LR 0.000500 Time 0.021530 +2023-10-02 21:26:02,622 - Epoch: [118][ 1050/ 1236] Overall Loss 0.189415 Objective Loss 0.189415 LR 0.000500 Time 0.021521 +2023-10-02 21:26:02,832 - Epoch: [118][ 1060/ 1236] Overall Loss 0.189491 Objective Loss 0.189491 LR 0.000500 Time 0.021516 +2023-10-02 21:26:03,038 - Epoch: [118][ 1070/ 1236] Overall Loss 0.189484 Objective Loss 0.189484 LR 0.000500 Time 0.021507 +2023-10-02 21:26:03,247 - Epoch: [118][ 1080/ 1236] Overall Loss 0.189457 Objective Loss 0.189457 LR 0.000500 Time 0.021501 +2023-10-02 21:26:03,454 - Epoch: [118][ 1090/ 1236] Overall Loss 0.189331 Objective Loss 0.189331 LR 0.000500 Time 0.021494 +2023-10-02 21:26:03,663 - Epoch: [118][ 1100/ 1236] Overall Loss 0.189144 Objective Loss 0.189144 LR 0.000500 Time 0.021488 +2023-10-02 21:26:03,871 - Epoch: [118][ 1110/ 1236] Overall Loss 0.189194 Objective Loss 0.189194 LR 0.000500 Time 0.021481 +2023-10-02 21:26:04,080 - Epoch: [118][ 1120/ 1236] Overall Loss 0.189367 Objective Loss 0.189367 LR 0.000500 Time 0.021475 +2023-10-02 21:26:04,287 - Epoch: [118][ 1130/ 1236] Overall Loss 0.189298 Objective Loss 0.189298 LR 0.000500 Time 0.021469 +2023-10-02 21:26:04,496 - Epoch: [118][ 1140/ 1236] Overall Loss 0.189312 Objective Loss 0.189312 LR 0.000500 Time 0.021463 +2023-10-02 21:26:04,704 - Epoch: [118][ 1150/ 1236] Overall Loss 0.189134 Objective Loss 0.189134 LR 0.000500 Time 0.021457 +2023-10-02 21:26:04,912 - Epoch: [118][ 1160/ 1236] Overall Loss 0.189268 Objective Loss 0.189268 LR 0.000500 Time 0.021452 +2023-10-02 21:26:05,120 - Epoch: [118][ 1170/ 1236] Overall Loss 0.189467 Objective Loss 0.189467 LR 0.000500 Time 0.021446 +2023-10-02 21:26:05,329 - Epoch: [118][ 1180/ 1236] Overall Loss 0.189455 Objective Loss 0.189455 LR 0.000500 Time 0.021441 +2023-10-02 21:26:05,536 - Epoch: [118][ 1190/ 1236] Overall Loss 0.189176 Objective Loss 0.189176 LR 0.000500 Time 0.021435 +2023-10-02 21:26:05,745 - Epoch: [118][ 1200/ 1236] Overall Loss 0.189051 Objective Loss 0.189051 LR 0.000500 Time 0.021429 +2023-10-02 21:26:05,953 - Epoch: [118][ 1210/ 1236] Overall Loss 0.189244 Objective Loss 0.189244 LR 0.000500 Time 0.021424 +2023-10-02 21:26:06,162 - Epoch: [118][ 1220/ 1236] Overall Loss 0.189265 Objective Loss 0.189265 LR 0.000500 Time 0.021419 +2023-10-02 21:26:06,425 - Epoch: [118][ 1230/ 1236] Overall Loss 0.189313 Objective Loss 0.189313 LR 0.000500 Time 0.021459 +2023-10-02 21:26:06,548 - Epoch: [118][ 1236/ 1236] Overall Loss 0.189286 Objective Loss 0.189286 Top1 88.798371 Top5 98.981670 LR 0.000500 Time 0.021454 +2023-10-02 21:26:06,678 - --- validate (epoch=118)----------- +2023-10-02 21:26:06,678 - 29943 samples (256 per mini-batch) +2023-10-02 21:26:07,159 - Epoch: [118][ 10/ 117] Loss 0.285467 Top1 85.742188 Top5 98.554688 +2023-10-02 21:26:07,325 - Epoch: [118][ 20/ 117] Loss 0.289291 Top1 86.113281 Top5 98.496094 +2023-10-02 21:26:07,486 - Epoch: [118][ 30/ 117] Loss 0.287381 Top1 86.041667 Top5 98.541667 +2023-10-02 21:26:07,651 - Epoch: [118][ 40/ 117] Loss 0.293263 Top1 86.005859 Top5 98.515625 +2023-10-02 21:26:07,810 - Epoch: [118][ 50/ 117] Loss 0.289808 Top1 86.125000 Top5 98.515625 +2023-10-02 21:26:07,975 - Epoch: [118][ 60/ 117] Loss 0.294633 Top1 85.957031 Top5 98.457031 +2023-10-02 21:26:08,133 - Epoch: [118][ 70/ 117] Loss 0.293691 Top1 86.015625 Top5 98.482143 +2023-10-02 21:26:08,297 - Epoch: [118][ 80/ 117] Loss 0.293654 Top1 85.937500 Top5 98.540039 +2023-10-02 21:26:08,447 - Epoch: [118][ 90/ 117] Loss 0.293083 Top1 85.920139 Top5 98.563368 +2023-10-02 21:26:08,600 - Epoch: [118][ 100/ 117] Loss 0.293041 Top1 85.898438 Top5 98.582031 +2023-10-02 21:26:08,757 - Epoch: [118][ 110/ 117] Loss 0.298658 Top1 85.777699 Top5 98.554688 +2023-10-02 21:26:08,847 - Epoch: [118][ 117/ 117] Loss 0.298746 Top1 85.762950 Top5 98.527202 +2023-10-02 21:26:08,992 - ==> Top1: 85.763 Top5: 98.527 Loss: 0.299 + +2023-10-02 21:26:08,993 - ==> Confusion: +[[ 932 0 2 1 6 2 0 0 7 64 1 0 1 3 4 2 6 1 2 0 16] + [ 0 1049 0 0 4 23 0 29 2 1 0 1 0 1 1 3 2 0 9 3 3] + [ 2 1 968 14 3 0 28 6 0 2 1 2 5 2 1 5 1 0 4 2 9] + [ 0 1 16 976 1 2 3 2 2 0 2 1 4 3 33 5 1 6 13 0 18] + [ 29 10 1 0 961 4 0 0 2 13 0 0 2 1 8 4 10 0 0 1 4] + [ 3 33 0 0 4 997 3 25 4 5 0 5 1 8 4 1 1 1 2 4 15] + [ 0 3 18 0 0 1 1137 7 0 0 5 0 0 0 0 3 0 1 2 10 4] + [ 0 10 8 1 1 28 6 1088 0 4 3 10 3 5 2 1 0 0 26 11 11] + [ 15 0 0 2 1 3 0 0 974 45 9 3 3 12 14 0 3 1 2 2 0] + [ 87 2 1 0 7 2 1 0 22 970 0 1 0 15 3 0 1 0 0 1 6] + [ 1 1 10 8 0 2 4 3 10 1 965 2 0 18 5 0 3 2 6 2 10] + [ 0 1 1 0 1 12 0 1 0 0 0 963 23 12 0 0 1 14 0 5 1] + [ 0 0 1 4 1 2 2 1 1 2 2 35 967 4 0 7 3 17 1 9 9] + [ 0 0 0 0 1 11 1 0 9 12 3 6 0 1060 4 0 0 1 0 2 9] + [ 16 0 5 20 6 0 0 0 15 3 2 0 3 0 1008 0 0 2 6 1 14] + [ 0 0 1 0 5 1 2 0 0 1 0 8 6 0 0 1068 19 11 2 5 5] + [ 1 13 0 0 3 8 1 0 0 0 0 4 0 4 3 5 1102 0 2 7 8] + [ 0 0 1 2 1 2 1 0 2 2 0 3 10 2 4 4 0 996 0 2 6] + [ 2 8 4 16 1 1 1 24 4 2 2 0 1 0 14 0 1 0 974 2 11] + [ 0 3 1 3 1 3 11 4 0 0 1 12 2 3 0 1 7 1 1 1092 6] + [ 127 165 125 62 73 178 38 92 80 100 157 93 287 278 104 42 93 52 117 209 5433]] + +2023-10-02 21:26:08,994 - ==> Best [Top1: 85.763 Top5: 98.527 Sparsity:0.00 Params: 169472 on epoch: 118] +2023-10-02 21:26:08,995 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:26:09,008 - + +2023-10-02 21:26:09,008 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:26:10,149 - Epoch: [119][ 10/ 1236] Overall Loss 0.186572 Objective Loss 0.186572 LR 0.000500 Time 0.113992 +2023-10-02 21:26:10,357 - Epoch: [119][ 20/ 1236] Overall Loss 0.187922 Objective Loss 0.187922 LR 0.000500 Time 0.067371 +2023-10-02 21:26:10,564 - Epoch: [119][ 30/ 1236] Overall Loss 0.188878 Objective Loss 0.188878 LR 0.000500 Time 0.051797 +2023-10-02 21:26:10,771 - Epoch: [119][ 40/ 1236] Overall Loss 0.192973 Objective Loss 0.192973 LR 0.000500 Time 0.044026 +2023-10-02 21:26:10,978 - Epoch: [119][ 50/ 1236] Overall Loss 0.190411 Objective Loss 0.190411 LR 0.000500 Time 0.039350 +2023-10-02 21:26:11,186 - Epoch: [119][ 60/ 1236] Overall Loss 0.185267 Objective Loss 0.185267 LR 0.000500 Time 0.036249 +2023-10-02 21:26:11,392 - Epoch: [119][ 70/ 1236] Overall Loss 0.183149 Objective Loss 0.183149 LR 0.000500 Time 0.034017 +2023-10-02 21:26:11,600 - Epoch: [119][ 80/ 1236] Overall Loss 0.186854 Objective Loss 0.186854 LR 0.000500 Time 0.032355 +2023-10-02 21:26:11,806 - Epoch: [119][ 90/ 1236] Overall Loss 0.186320 Objective Loss 0.186320 LR 0.000500 Time 0.031038 +2023-10-02 21:26:12,014 - Epoch: [119][ 100/ 1236] Overall Loss 0.187704 Objective Loss 0.187704 LR 0.000500 Time 0.030006 +2023-10-02 21:26:12,220 - Epoch: [119][ 110/ 1236] Overall Loss 0.187682 Objective Loss 0.187682 LR 0.000500 Time 0.029138 +2023-10-02 21:26:12,428 - Epoch: [119][ 120/ 1236] Overall Loss 0.185801 Objective Loss 0.185801 LR 0.000500 Time 0.028438 +2023-10-02 21:26:12,634 - Epoch: [119][ 130/ 1236] Overall Loss 0.185662 Objective Loss 0.185662 LR 0.000500 Time 0.027825 +2023-10-02 21:26:12,842 - Epoch: [119][ 140/ 1236] Overall Loss 0.184431 Objective Loss 0.184431 LR 0.000500 Time 0.027320 +2023-10-02 21:26:13,048 - Epoch: [119][ 150/ 1236] Overall Loss 0.183922 Objective Loss 0.183922 LR 0.000500 Time 0.026864 +2023-10-02 21:26:13,256 - Epoch: [119][ 160/ 1236] Overall Loss 0.184039 Objective Loss 0.184039 LR 0.000500 Time 0.026482 +2023-10-02 21:26:13,462 - Epoch: [119][ 170/ 1236] Overall Loss 0.184661 Objective Loss 0.184661 LR 0.000500 Time 0.026129 +2023-10-02 21:26:13,670 - Epoch: [119][ 180/ 1236] Overall Loss 0.184651 Objective Loss 0.184651 LR 0.000500 Time 0.025830 +2023-10-02 21:26:13,876 - Epoch: [119][ 190/ 1236] Overall Loss 0.185385 Objective Loss 0.185385 LR 0.000500 Time 0.025550 +2023-10-02 21:26:14,084 - Epoch: [119][ 200/ 1236] Overall Loss 0.185141 Objective Loss 0.185141 LR 0.000500 Time 0.025310 +2023-10-02 21:26:14,291 - Epoch: [119][ 210/ 1236] Overall Loss 0.184996 Objective Loss 0.184996 LR 0.000500 Time 0.025080 +2023-10-02 21:26:14,498 - Epoch: [119][ 220/ 1236] Overall Loss 0.185224 Objective Loss 0.185224 LR 0.000500 Time 0.024883 +2023-10-02 21:26:14,705 - Epoch: [119][ 230/ 1236] Overall Loss 0.185396 Objective Loss 0.185396 LR 0.000500 Time 0.024692 +2023-10-02 21:26:14,913 - Epoch: [119][ 240/ 1236] Overall Loss 0.185233 Objective Loss 0.185233 LR 0.000500 Time 0.024528 +2023-10-02 21:26:15,119 - Epoch: [119][ 250/ 1236] Overall Loss 0.184086 Objective Loss 0.184086 LR 0.000500 Time 0.024366 +2023-10-02 21:26:15,327 - Epoch: [119][ 260/ 1236] Overall Loss 0.184429 Objective Loss 0.184429 LR 0.000500 Time 0.024227 +2023-10-02 21:26:15,533 - Epoch: [119][ 270/ 1236] Overall Loss 0.184396 Objective Loss 0.184396 LR 0.000500 Time 0.024088 +2023-10-02 21:26:15,741 - Epoch: [119][ 280/ 1236] Overall Loss 0.184249 Objective Loss 0.184249 LR 0.000500 Time 0.023969 +2023-10-02 21:26:15,951 - Epoch: [119][ 290/ 1236] Overall Loss 0.184458 Objective Loss 0.184458 LR 0.000500 Time 0.023862 +2023-10-02 21:26:16,171 - Epoch: [119][ 300/ 1236] Overall Loss 0.184486 Objective Loss 0.184486 LR 0.000500 Time 0.023797 +2023-10-02 21:26:16,384 - Epoch: [119][ 310/ 1236] Overall Loss 0.185441 Objective Loss 0.185441 LR 0.000500 Time 0.023719 +2023-10-02 21:26:16,604 - Epoch: [119][ 320/ 1236] Overall Loss 0.184893 Objective Loss 0.184893 LR 0.000500 Time 0.023662 +2023-10-02 21:26:16,819 - Epoch: [119][ 330/ 1236] Overall Loss 0.183952 Objective Loss 0.183952 LR 0.000500 Time 0.023594 +2023-10-02 21:26:17,038 - Epoch: [119][ 340/ 1236] Overall Loss 0.184140 Objective Loss 0.184140 LR 0.000500 Time 0.023545 +2023-10-02 21:26:17,252 - Epoch: [119][ 350/ 1236] Overall Loss 0.184158 Objective Loss 0.184158 LR 0.000500 Time 0.023484 +2023-10-02 21:26:17,472 - Epoch: [119][ 360/ 1236] Overall Loss 0.184407 Objective Loss 0.184407 LR 0.000500 Time 0.023441 +2023-10-02 21:26:17,686 - Epoch: [119][ 370/ 1236] Overall Loss 0.184501 Objective Loss 0.184501 LR 0.000500 Time 0.023385 +2023-10-02 21:26:17,906 - Epoch: [119][ 380/ 1236] Overall Loss 0.184820 Objective Loss 0.184820 LR 0.000500 Time 0.023348 +2023-10-02 21:26:18,121 - Epoch: [119][ 390/ 1236] Overall Loss 0.184584 Objective Loss 0.184584 LR 0.000500 Time 0.023297 +2023-10-02 21:26:18,341 - Epoch: [119][ 400/ 1236] Overall Loss 0.185075 Objective Loss 0.185075 LR 0.000500 Time 0.023264 +2023-10-02 21:26:18,555 - Epoch: [119][ 410/ 1236] Overall Loss 0.185082 Objective Loss 0.185082 LR 0.000500 Time 0.023217 +2023-10-02 21:26:18,774 - Epoch: [119][ 420/ 1236] Overall Loss 0.184978 Objective Loss 0.184978 LR 0.000500 Time 0.023187 +2023-10-02 21:26:18,983 - Epoch: [119][ 430/ 1236] Overall Loss 0.184793 Objective Loss 0.184793 LR 0.000500 Time 0.023132 +2023-10-02 21:26:19,194 - Epoch: [119][ 440/ 1236] Overall Loss 0.185287 Objective Loss 0.185287 LR 0.000500 Time 0.023085 +2023-10-02 21:26:19,402 - Epoch: [119][ 450/ 1236] Overall Loss 0.185236 Objective Loss 0.185236 LR 0.000500 Time 0.023034 +2023-10-02 21:26:19,613 - Epoch: [119][ 460/ 1236] Overall Loss 0.185421 Objective Loss 0.185421 LR 0.000500 Time 0.022991 +2023-10-02 21:26:19,821 - Epoch: [119][ 470/ 1236] Overall Loss 0.185090 Objective Loss 0.185090 LR 0.000500 Time 0.022945 +2023-10-02 21:26:20,032 - Epoch: [119][ 480/ 1236] Overall Loss 0.185059 Objective Loss 0.185059 LR 0.000500 Time 0.022905 +2023-10-02 21:26:20,241 - Epoch: [119][ 490/ 1236] Overall Loss 0.185428 Objective Loss 0.185428 LR 0.000500 Time 0.022861 +2023-10-02 21:26:20,452 - Epoch: [119][ 500/ 1236] Overall Loss 0.185413 Objective Loss 0.185413 LR 0.000500 Time 0.022824 +2023-10-02 21:26:20,660 - Epoch: [119][ 510/ 1236] Overall Loss 0.185593 Objective Loss 0.185593 LR 0.000500 Time 0.022783 +2023-10-02 21:26:20,871 - Epoch: [119][ 520/ 1236] Overall Loss 0.185941 Objective Loss 0.185941 LR 0.000500 Time 0.022749 +2023-10-02 21:26:21,080 - Epoch: [119][ 530/ 1236] Overall Loss 0.186066 Objective Loss 0.186066 LR 0.000500 Time 0.022714 +2023-10-02 21:26:21,291 - Epoch: [119][ 540/ 1236] Overall Loss 0.186158 Objective Loss 0.186158 LR 0.000500 Time 0.022683 +2023-10-02 21:26:21,500 - Epoch: [119][ 550/ 1236] Overall Loss 0.186394 Objective Loss 0.186394 LR 0.000500 Time 0.022650 +2023-10-02 21:26:21,711 - Epoch: [119][ 560/ 1236] Overall Loss 0.186403 Objective Loss 0.186403 LR 0.000500 Time 0.022621 +2023-10-02 21:26:21,919 - Epoch: [119][ 570/ 1236] Overall Loss 0.186738 Objective Loss 0.186738 LR 0.000500 Time 0.022590 +2023-10-02 21:26:22,130 - Epoch: [119][ 580/ 1236] Overall Loss 0.186836 Objective Loss 0.186836 LR 0.000500 Time 0.022563 +2023-10-02 21:26:22,339 - Epoch: [119][ 590/ 1236] Overall Loss 0.187068 Objective Loss 0.187068 LR 0.000500 Time 0.022534 +2023-10-02 21:26:22,550 - Epoch: [119][ 600/ 1236] Overall Loss 0.187423 Objective Loss 0.187423 LR 0.000500 Time 0.022510 +2023-10-02 21:26:22,759 - Epoch: [119][ 610/ 1236] Overall Loss 0.187874 Objective Loss 0.187874 LR 0.000500 Time 0.022483 +2023-10-02 21:26:22,970 - Epoch: [119][ 620/ 1236] Overall Loss 0.188238 Objective Loss 0.188238 LR 0.000500 Time 0.022460 +2023-10-02 21:26:23,178 - Epoch: [119][ 630/ 1236] Overall Loss 0.188297 Objective Loss 0.188297 LR 0.000500 Time 0.022434 +2023-10-02 21:26:23,389 - Epoch: [119][ 640/ 1236] Overall Loss 0.188281 Objective Loss 0.188281 LR 0.000500 Time 0.022412 +2023-10-02 21:26:23,598 - Epoch: [119][ 650/ 1236] Overall Loss 0.188285 Objective Loss 0.188285 LR 0.000500 Time 0.022389 +2023-10-02 21:26:23,809 - Epoch: [119][ 660/ 1236] Overall Loss 0.188297 Objective Loss 0.188297 LR 0.000500 Time 0.022368 +2023-10-02 21:26:24,017 - Epoch: [119][ 670/ 1236] Overall Loss 0.188320 Objective Loss 0.188320 LR 0.000500 Time 0.022345 +2023-10-02 21:26:24,228 - Epoch: [119][ 680/ 1236] Overall Loss 0.188186 Objective Loss 0.188186 LR 0.000500 Time 0.022326 +2023-10-02 21:26:24,436 - Epoch: [119][ 690/ 1236] Overall Loss 0.188172 Objective Loss 0.188172 LR 0.000500 Time 0.022304 +2023-10-02 21:26:24,647 - Epoch: [119][ 700/ 1236] Overall Loss 0.188524 Objective Loss 0.188524 LR 0.000500 Time 0.022286 +2023-10-02 21:26:24,856 - Epoch: [119][ 710/ 1236] Overall Loss 0.188789 Objective Loss 0.188789 LR 0.000500 Time 0.022265 +2023-10-02 21:26:25,066 - Epoch: [119][ 720/ 1236] Overall Loss 0.188644 Objective Loss 0.188644 LR 0.000500 Time 0.022248 +2023-10-02 21:26:25,275 - Epoch: [119][ 730/ 1236] Overall Loss 0.188465 Objective Loss 0.188465 LR 0.000500 Time 0.022229 +2023-10-02 21:26:25,486 - Epoch: [119][ 740/ 1236] Overall Loss 0.188512 Objective Loss 0.188512 LR 0.000500 Time 0.022213 +2023-10-02 21:26:25,694 - Epoch: [119][ 750/ 1236] Overall Loss 0.188562 Objective Loss 0.188562 LR 0.000500 Time 0.022194 +2023-10-02 21:26:25,905 - Epoch: [119][ 760/ 1236] Overall Loss 0.188543 Objective Loss 0.188543 LR 0.000500 Time 0.022179 +2023-10-02 21:26:26,114 - Epoch: [119][ 770/ 1236] Overall Loss 0.188382 Objective Loss 0.188382 LR 0.000500 Time 0.022162 +2023-10-02 21:26:26,324 - Epoch: [119][ 780/ 1236] Overall Loss 0.188083 Objective Loss 0.188083 LR 0.000500 Time 0.022147 +2023-10-02 21:26:26,534 - Epoch: [119][ 790/ 1236] Overall Loss 0.187977 Objective Loss 0.187977 LR 0.000500 Time 0.022132 +2023-10-02 21:26:26,745 - Epoch: [119][ 800/ 1236] Overall Loss 0.188347 Objective Loss 0.188347 LR 0.000500 Time 0.022118 +2023-10-02 21:26:26,953 - Epoch: [119][ 810/ 1236] Overall Loss 0.188419 Objective Loss 0.188419 LR 0.000500 Time 0.022102 +2023-10-02 21:26:27,164 - Epoch: [119][ 820/ 1236] Overall Loss 0.188539 Objective Loss 0.188539 LR 0.000500 Time 0.022089 +2023-10-02 21:26:27,373 - Epoch: [119][ 830/ 1236] Overall Loss 0.188866 Objective Loss 0.188866 LR 0.000500 Time 0.022074 +2023-10-02 21:26:27,583 - Epoch: [119][ 840/ 1236] Overall Loss 0.189133 Objective Loss 0.189133 LR 0.000500 Time 0.022062 +2023-10-02 21:26:27,792 - Epoch: [119][ 850/ 1236] Overall Loss 0.189305 Objective Loss 0.189305 LR 0.000500 Time 0.022046 +2023-10-02 21:26:28,003 - Epoch: [119][ 860/ 1236] Overall Loss 0.189284 Objective Loss 0.189284 LR 0.000500 Time 0.022035 +2023-10-02 21:26:28,212 - Epoch: [119][ 870/ 1236] Overall Loss 0.189140 Objective Loss 0.189140 LR 0.000500 Time 0.022021 +2023-10-02 21:26:28,423 - Epoch: [119][ 880/ 1236] Overall Loss 0.189123 Objective Loss 0.189123 LR 0.000500 Time 0.022010 +2023-10-02 21:26:28,632 - Epoch: [119][ 890/ 1236] Overall Loss 0.189157 Objective Loss 0.189157 LR 0.000500 Time 0.021997 +2023-10-02 21:26:28,842 - Epoch: [119][ 900/ 1236] Overall Loss 0.189351 Objective Loss 0.189351 LR 0.000500 Time 0.021987 +2023-10-02 21:26:29,051 - Epoch: [119][ 910/ 1236] Overall Loss 0.189526 Objective Loss 0.189526 LR 0.000500 Time 0.021974 +2023-10-02 21:26:29,260 - Epoch: [119][ 920/ 1236] Overall Loss 0.189368 Objective Loss 0.189368 LR 0.000500 Time 0.021962 +2023-10-02 21:26:29,467 - Epoch: [119][ 930/ 1236] Overall Loss 0.189328 Objective Loss 0.189328 LR 0.000500 Time 0.021947 +2023-10-02 21:26:29,676 - Epoch: [119][ 940/ 1236] Overall Loss 0.189336 Objective Loss 0.189336 LR 0.000500 Time 0.021935 +2023-10-02 21:26:29,884 - Epoch: [119][ 950/ 1236] Overall Loss 0.189303 Objective Loss 0.189303 LR 0.000500 Time 0.021922 +2023-10-02 21:26:30,093 - Epoch: [119][ 960/ 1236] Overall Loss 0.188770 Objective Loss 0.188770 LR 0.000500 Time 0.021910 +2023-10-02 21:26:30,300 - Epoch: [119][ 970/ 1236] Overall Loss 0.188777 Objective Loss 0.188777 LR 0.000500 Time 0.021897 +2023-10-02 21:26:30,509 - Epoch: [119][ 980/ 1236] Overall Loss 0.189110 Objective Loss 0.189110 LR 0.000500 Time 0.021886 +2023-10-02 21:26:30,716 - Epoch: [119][ 990/ 1236] Overall Loss 0.189215 Objective Loss 0.189215 LR 0.000500 Time 0.021874 +2023-10-02 21:26:30,925 - Epoch: [119][ 1000/ 1236] Overall Loss 0.189149 Objective Loss 0.189149 LR 0.000500 Time 0.021864 +2023-10-02 21:26:31,133 - Epoch: [119][ 1010/ 1236] Overall Loss 0.189370 Objective Loss 0.189370 LR 0.000500 Time 0.021853 +2023-10-02 21:26:31,342 - Epoch: [119][ 1020/ 1236] Overall Loss 0.189505 Objective Loss 0.189505 LR 0.000500 Time 0.021843 +2023-10-02 21:26:31,549 - Epoch: [119][ 1030/ 1236] Overall Loss 0.189474 Objective Loss 0.189474 LR 0.000500 Time 0.021832 +2023-10-02 21:26:31,760 - Epoch: [119][ 1040/ 1236] Overall Loss 0.189509 Objective Loss 0.189509 LR 0.000500 Time 0.021824 +2023-10-02 21:26:31,966 - Epoch: [119][ 1050/ 1236] Overall Loss 0.189239 Objective Loss 0.189239 LR 0.000500 Time 0.021813 +2023-10-02 21:26:32,175 - Epoch: [119][ 1060/ 1236] Overall Loss 0.189212 Objective Loss 0.189212 LR 0.000500 Time 0.021804 +2023-10-02 21:26:32,382 - Epoch: [119][ 1070/ 1236] Overall Loss 0.189325 Objective Loss 0.189325 LR 0.000500 Time 0.021793 +2023-10-02 21:26:32,593 - Epoch: [119][ 1080/ 1236] Overall Loss 0.189341 Objective Loss 0.189341 LR 0.000500 Time 0.021786 +2023-10-02 21:26:32,799 - Epoch: [119][ 1090/ 1236] Overall Loss 0.189350 Objective Loss 0.189350 LR 0.000500 Time 0.021775 +2023-10-02 21:26:33,008 - Epoch: [119][ 1100/ 1236] Overall Loss 0.189280 Objective Loss 0.189280 LR 0.000500 Time 0.021767 +2023-10-02 21:26:33,215 - Epoch: [119][ 1110/ 1236] Overall Loss 0.189148 Objective Loss 0.189148 LR 0.000500 Time 0.021757 +2023-10-02 21:26:33,424 - Epoch: [119][ 1120/ 1236] Overall Loss 0.189159 Objective Loss 0.189159 LR 0.000500 Time 0.021749 +2023-10-02 21:26:33,632 - Epoch: [119][ 1130/ 1236] Overall Loss 0.189259 Objective Loss 0.189259 LR 0.000500 Time 0.021740 +2023-10-02 21:26:33,840 - Epoch: [119][ 1140/ 1236] Overall Loss 0.189583 Objective Loss 0.189583 LR 0.000500 Time 0.021732 +2023-10-02 21:26:34,048 - Epoch: [119][ 1150/ 1236] Overall Loss 0.189365 Objective Loss 0.189365 LR 0.000500 Time 0.021724 +2023-10-02 21:26:34,257 - Epoch: [119][ 1160/ 1236] Overall Loss 0.189475 Objective Loss 0.189475 LR 0.000500 Time 0.021716 +2023-10-02 21:26:34,464 - Epoch: [119][ 1170/ 1236] Overall Loss 0.189465 Objective Loss 0.189465 LR 0.000500 Time 0.021708 +2023-10-02 21:26:34,673 - Epoch: [119][ 1180/ 1236] Overall Loss 0.189470 Objective Loss 0.189470 LR 0.000500 Time 0.021700 +2023-10-02 21:26:34,885 - Epoch: [119][ 1190/ 1236] Overall Loss 0.189467 Objective Loss 0.189467 LR 0.000500 Time 0.021696 +2023-10-02 21:26:35,094 - Epoch: [119][ 1200/ 1236] Overall Loss 0.189536 Objective Loss 0.189536 LR 0.000500 Time 0.021689 +2023-10-02 21:26:35,302 - Epoch: [119][ 1210/ 1236] Overall Loss 0.189456 Objective Loss 0.189456 LR 0.000500 Time 0.021681 +2023-10-02 21:26:35,511 - Epoch: [119][ 1220/ 1236] Overall Loss 0.189380 Objective Loss 0.189380 LR 0.000500 Time 0.021674 +2023-10-02 21:26:35,774 - Epoch: [119][ 1230/ 1236] Overall Loss 0.189341 Objective Loss 0.189341 LR 0.000500 Time 0.021712 +2023-10-02 21:26:35,897 - Epoch: [119][ 1236/ 1236] Overall Loss 0.189365 Objective Loss 0.189365 Top1 89.002037 Top5 98.574338 LR 0.000500 Time 0.021706 +2023-10-02 21:26:36,036 - --- validate (epoch=119)----------- +2023-10-02 21:26:36,037 - 29943 samples (256 per mini-batch) +2023-10-02 21:26:36,541 - Epoch: [119][ 10/ 117] Loss 0.277364 Top1 86.484375 Top5 98.710938 +2023-10-02 21:26:36,706 - Epoch: [119][ 20/ 117] Loss 0.283598 Top1 86.035156 Top5 98.457031 +2023-10-02 21:26:36,863 - Epoch: [119][ 30/ 117] Loss 0.286012 Top1 86.171875 Top5 98.424479 +2023-10-02 21:26:37,027 - Epoch: [119][ 40/ 117] Loss 0.290934 Top1 86.191406 Top5 98.388672 +2023-10-02 21:26:37,187 - Epoch: [119][ 50/ 117] Loss 0.304369 Top1 85.796875 Top5 98.289062 +2023-10-02 21:26:37,351 - Epoch: [119][ 60/ 117] Loss 0.304162 Top1 85.820312 Top5 98.294271 +2023-10-02 21:26:37,510 - Epoch: [119][ 70/ 117] Loss 0.302207 Top1 85.764509 Top5 98.286830 +2023-10-02 21:26:37,671 - Epoch: [119][ 80/ 117] Loss 0.299472 Top1 85.698242 Top5 98.320312 +2023-10-02 21:26:37,831 - Epoch: [119][ 90/ 117] Loss 0.301213 Top1 85.512153 Top5 98.320312 +2023-10-02 21:26:37,991 - Epoch: [119][ 100/ 117] Loss 0.298675 Top1 85.621094 Top5 98.343750 +2023-10-02 21:26:38,158 - Epoch: [119][ 110/ 117] Loss 0.297581 Top1 85.593040 Top5 98.352273 +2023-10-02 21:26:38,249 - Epoch: [119][ 117/ 117] Loss 0.296327 Top1 85.632702 Top5 98.393615 +2023-10-02 21:26:38,397 - ==> Top1: 85.633 Top5: 98.394 Loss: 0.296 + +2023-10-02 21:26:38,398 - ==> Confusion: +[[ 934 1 3 1 8 3 0 0 6 61 2 0 1 2 3 2 4 1 2 0 16] + [ 0 1057 0 1 6 20 1 23 1 0 1 0 0 0 0 3 1 0 12 3 2] + [ 2 0 978 12 4 0 22 4 0 1 2 1 5 1 1 3 0 1 11 2 6] + [ 1 1 10 982 1 4 2 5 3 0 1 0 4 3 24 5 1 4 19 1 18] + [ 20 6 1 1 978 5 0 1 0 11 0 0 1 2 9 3 10 0 0 2 0] + [ 3 27 1 1 3 994 3 30 1 5 1 9 1 8 4 2 1 1 7 2 12] + [ 0 1 21 0 0 2 1142 4 0 0 5 0 0 0 0 3 0 0 1 8 4] + [ 0 8 16 0 2 26 8 1090 0 2 3 4 3 4 2 0 0 1 36 7 6] + [ 18 1 0 1 1 2 0 1 983 33 11 3 1 8 13 4 3 2 4 0 0] + [ 98 2 2 0 8 4 0 0 28 950 0 0 0 13 7 0 1 1 1 0 4] + [ 3 1 7 12 1 0 3 2 18 0 965 2 0 13 4 0 1 2 10 1 8] + [ 0 1 2 0 0 11 0 4 0 1 0 962 21 10 0 0 0 17 0 4 2] + [ 0 0 3 3 0 2 4 2 1 2 1 40 969 2 2 4 2 13 3 5 10] + [ 0 0 0 0 4 9 2 0 10 10 7 7 0 1047 5 0 1 2 0 1 14] + [ 10 0 4 18 8 1 0 0 17 1 4 0 3 2 1011 0 4 2 9 0 7] + [ 0 0 1 0 3 0 0 0 0 1 0 8 7 0 0 1070 16 13 2 10 3] + [ 2 16 0 0 3 5 0 0 0 2 0 3 0 2 3 7 1100 1 1 6 10] + [ 0 0 1 0 0 1 3 0 0 0 0 8 23 0 3 4 2 990 1 1 1] + [ 1 3 2 15 1 0 2 21 3 2 2 0 1 0 9 0 1 0 996 0 9] + [ 0 3 1 2 0 4 11 8 0 1 0 15 0 2 0 1 6 1 0 1093 4] + [ 121 160 119 95 71 145 40 105 104 71 176 121 304 233 116 61 88 72 157 196 5350]] + +2023-10-02 21:26:38,399 - ==> Best [Top1: 85.763 Top5: 98.527 Sparsity:0.00 Params: 169472 on epoch: 118] +2023-10-02 21:26:38,399 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:26:38,405 - + +2023-10-02 21:26:38,405 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:26:39,441 - Epoch: [120][ 10/ 1236] Overall Loss 0.180223 Objective Loss 0.180223 LR 0.000500 Time 0.103494 +2023-10-02 21:26:39,649 - Epoch: [120][ 20/ 1236] Overall Loss 0.179181 Objective Loss 0.179181 LR 0.000500 Time 0.062130 +2023-10-02 21:26:39,855 - Epoch: [120][ 30/ 1236] Overall Loss 0.177015 Objective Loss 0.177015 LR 0.000500 Time 0.048287 +2023-10-02 21:26:40,063 - Epoch: [120][ 40/ 1236] Overall Loss 0.181081 Objective Loss 0.181081 LR 0.000500 Time 0.041395 +2023-10-02 21:26:40,270 - Epoch: [120][ 50/ 1236] Overall Loss 0.187816 Objective Loss 0.187816 LR 0.000500 Time 0.037245 +2023-10-02 21:26:40,477 - Epoch: [120][ 60/ 1236] Overall Loss 0.187726 Objective Loss 0.187726 LR 0.000500 Time 0.034488 +2023-10-02 21:26:40,684 - Epoch: [120][ 70/ 1236] Overall Loss 0.185035 Objective Loss 0.185035 LR 0.000500 Time 0.032513 +2023-10-02 21:26:40,892 - Epoch: [120][ 80/ 1236] Overall Loss 0.188106 Objective Loss 0.188106 LR 0.000500 Time 0.031037 +2023-10-02 21:26:41,098 - Epoch: [120][ 90/ 1236] Overall Loss 0.185527 Objective Loss 0.185527 LR 0.000500 Time 0.029883 +2023-10-02 21:26:41,306 - Epoch: [120][ 100/ 1236] Overall Loss 0.186568 Objective Loss 0.186568 LR 0.000500 Time 0.028968 +2023-10-02 21:26:41,513 - Epoch: [120][ 110/ 1236] Overall Loss 0.187001 Objective Loss 0.187001 LR 0.000500 Time 0.028212 +2023-10-02 21:26:41,721 - Epoch: [120][ 120/ 1236] Overall Loss 0.186394 Objective Loss 0.186394 LR 0.000500 Time 0.027591 +2023-10-02 21:26:41,927 - Epoch: [120][ 130/ 1236] Overall Loss 0.185487 Objective Loss 0.185487 LR 0.000500 Time 0.027045 +2023-10-02 21:26:42,135 - Epoch: [120][ 140/ 1236] Overall Loss 0.185300 Objective Loss 0.185300 LR 0.000500 Time 0.026594 +2023-10-02 21:26:42,342 - Epoch: [120][ 150/ 1236] Overall Loss 0.186247 Objective Loss 0.186247 LR 0.000500 Time 0.026188 +2023-10-02 21:26:42,554 - Epoch: [120][ 160/ 1236] Overall Loss 0.185829 Objective Loss 0.185829 LR 0.000500 Time 0.025875 +2023-10-02 21:26:42,768 - Epoch: [120][ 170/ 1236] Overall Loss 0.186303 Objective Loss 0.186303 LR 0.000500 Time 0.025600 +2023-10-02 21:26:42,979 - Epoch: [120][ 180/ 1236] Overall Loss 0.186442 Objective Loss 0.186442 LR 0.000500 Time 0.025351 +2023-10-02 21:26:43,193 - Epoch: [120][ 190/ 1236] Overall Loss 0.185851 Objective Loss 0.185851 LR 0.000500 Time 0.025133 +2023-10-02 21:26:43,405 - Epoch: [120][ 200/ 1236] Overall Loss 0.184559 Objective Loss 0.184559 LR 0.000500 Time 0.024933 +2023-10-02 21:26:43,617 - Epoch: [120][ 210/ 1236] Overall Loss 0.184869 Objective Loss 0.184869 LR 0.000500 Time 0.024750 +2023-10-02 21:26:43,830 - Epoch: [120][ 220/ 1236] Overall Loss 0.184156 Objective Loss 0.184156 LR 0.000500 Time 0.024586 +2023-10-02 21:26:44,043 - Epoch: [120][ 230/ 1236] Overall Loss 0.184931 Objective Loss 0.184931 LR 0.000500 Time 0.024435 +2023-10-02 21:26:44,256 - Epoch: [120][ 240/ 1236] Overall Loss 0.185603 Objective Loss 0.185603 LR 0.000500 Time 0.024296 +2023-10-02 21:26:44,468 - Epoch: [120][ 250/ 1236] Overall Loss 0.185377 Objective Loss 0.185377 LR 0.000500 Time 0.024167 +2023-10-02 21:26:44,681 - Epoch: [120][ 260/ 1236] Overall Loss 0.184980 Objective Loss 0.184980 LR 0.000500 Time 0.024050 +2023-10-02 21:26:44,894 - Epoch: [120][ 270/ 1236] Overall Loss 0.185833 Objective Loss 0.185833 LR 0.000500 Time 0.023943 +2023-10-02 21:26:45,106 - Epoch: [120][ 280/ 1236] Overall Loss 0.186772 Objective Loss 0.186772 LR 0.000500 Time 0.023842 +2023-10-02 21:26:45,318 - Epoch: [120][ 290/ 1236] Overall Loss 0.185817 Objective Loss 0.185817 LR 0.000500 Time 0.023746 +2023-10-02 21:26:45,530 - Epoch: [120][ 300/ 1236] Overall Loss 0.185482 Objective Loss 0.185482 LR 0.000500 Time 0.023657 +2023-10-02 21:26:45,743 - Epoch: [120][ 310/ 1236] Overall Loss 0.186944 Objective Loss 0.186944 LR 0.000500 Time 0.023573 +2023-10-02 21:26:45,955 - Epoch: [120][ 320/ 1236] Overall Loss 0.186679 Objective Loss 0.186679 LR 0.000500 Time 0.023496 +2023-10-02 21:26:46,170 - Epoch: [120][ 330/ 1236] Overall Loss 0.186044 Objective Loss 0.186044 LR 0.000500 Time 0.023429 +2023-10-02 21:26:46,385 - Epoch: [120][ 340/ 1236] Overall Loss 0.186258 Objective Loss 0.186258 LR 0.000500 Time 0.023368 +2023-10-02 21:26:46,598 - Epoch: [120][ 350/ 1236] Overall Loss 0.186392 Objective Loss 0.186392 LR 0.000500 Time 0.023303 +2023-10-02 21:26:46,818 - Epoch: [120][ 360/ 1236] Overall Loss 0.186236 Objective Loss 0.186236 LR 0.000500 Time 0.023267 +2023-10-02 21:26:47,038 - Epoch: [120][ 370/ 1236] Overall Loss 0.186182 Objective Loss 0.186182 LR 0.000500 Time 0.023229 +2023-10-02 21:26:47,261 - Epoch: [120][ 380/ 1236] Overall Loss 0.186478 Objective Loss 0.186478 LR 0.000500 Time 0.023203 +2023-10-02 21:26:47,479 - Epoch: [120][ 390/ 1236] Overall Loss 0.185908 Objective Loss 0.185908 LR 0.000500 Time 0.023168 +2023-10-02 21:26:47,702 - Epoch: [120][ 400/ 1236] Overall Loss 0.185520 Objective Loss 0.185520 LR 0.000500 Time 0.023145 +2023-10-02 21:26:47,921 - Epoch: [120][ 410/ 1236] Overall Loss 0.185795 Objective Loss 0.185795 LR 0.000500 Time 0.023113 +2023-10-02 21:26:48,144 - Epoch: [120][ 420/ 1236] Overall Loss 0.185492 Objective Loss 0.185492 LR 0.000500 Time 0.023092 +2023-10-02 21:26:48,363 - Epoch: [120][ 430/ 1236] Overall Loss 0.185877 Objective Loss 0.185877 LR 0.000500 Time 0.023063 +2023-10-02 21:26:48,585 - Epoch: [120][ 440/ 1236] Overall Loss 0.185927 Objective Loss 0.185927 LR 0.000500 Time 0.023045 +2023-10-02 21:26:48,804 - Epoch: [120][ 450/ 1236] Overall Loss 0.185820 Objective Loss 0.185820 LR 0.000500 Time 0.023018 +2023-10-02 21:26:49,027 - Epoch: [120][ 460/ 1236] Overall Loss 0.185997 Objective Loss 0.185997 LR 0.000500 Time 0.023001 +2023-10-02 21:26:49,246 - Epoch: [120][ 470/ 1236] Overall Loss 0.186404 Objective Loss 0.186404 LR 0.000500 Time 0.022976 +2023-10-02 21:26:49,469 - Epoch: [120][ 480/ 1236] Overall Loss 0.185766 Objective Loss 0.185766 LR 0.000500 Time 0.022961 +2023-10-02 21:26:49,688 - Epoch: [120][ 490/ 1236] Overall Loss 0.185876 Objective Loss 0.185876 LR 0.000500 Time 0.022939 +2023-10-02 21:26:49,911 - Epoch: [120][ 500/ 1236] Overall Loss 0.186007 Objective Loss 0.186007 LR 0.000500 Time 0.022925 +2023-10-02 21:26:50,130 - Epoch: [120][ 510/ 1236] Overall Loss 0.185627 Objective Loss 0.185627 LR 0.000500 Time 0.022904 +2023-10-02 21:26:50,353 - Epoch: [120][ 520/ 1236] Overall Loss 0.185756 Objective Loss 0.185756 LR 0.000500 Time 0.022892 +2023-10-02 21:26:50,572 - Epoch: [120][ 530/ 1236] Overall Loss 0.185787 Objective Loss 0.185787 LR 0.000500 Time 0.022872 +2023-10-02 21:26:50,783 - Epoch: [120][ 540/ 1236] Overall Loss 0.185740 Objective Loss 0.185740 LR 0.000500 Time 0.022838 +2023-10-02 21:26:50,995 - Epoch: [120][ 550/ 1236] Overall Loss 0.185876 Objective Loss 0.185876 LR 0.000500 Time 0.022806 +2023-10-02 21:26:51,207 - Epoch: [120][ 560/ 1236] Overall Loss 0.185975 Objective Loss 0.185975 LR 0.000500 Time 0.022776 +2023-10-02 21:26:51,419 - Epoch: [120][ 570/ 1236] Overall Loss 0.185975 Objective Loss 0.185975 LR 0.000500 Time 0.022746 +2023-10-02 21:26:51,629 - Epoch: [120][ 580/ 1236] Overall Loss 0.186406 Objective Loss 0.186406 LR 0.000500 Time 0.022716 +2023-10-02 21:26:51,839 - Epoch: [120][ 590/ 1236] Overall Loss 0.186567 Objective Loss 0.186567 LR 0.000500 Time 0.022684 +2023-10-02 21:26:52,049 - Epoch: [120][ 600/ 1236] Overall Loss 0.186234 Objective Loss 0.186234 LR 0.000500 Time 0.022655 +2023-10-02 21:26:52,258 - Epoch: [120][ 610/ 1236] Overall Loss 0.186464 Objective Loss 0.186464 LR 0.000500 Time 0.022624 +2023-10-02 21:26:52,468 - Epoch: [120][ 620/ 1236] Overall Loss 0.186318 Objective Loss 0.186318 LR 0.000500 Time 0.022597 +2023-10-02 21:26:52,677 - Epoch: [120][ 630/ 1236] Overall Loss 0.186600 Objective Loss 0.186600 LR 0.000500 Time 0.022568 +2023-10-02 21:26:52,887 - Epoch: [120][ 640/ 1236] Overall Loss 0.186309 Objective Loss 0.186309 LR 0.000500 Time 0.022543 +2023-10-02 21:26:53,097 - Epoch: [120][ 650/ 1236] Overall Loss 0.186771 Objective Loss 0.186771 LR 0.000500 Time 0.022517 +2023-10-02 21:26:53,307 - Epoch: [120][ 660/ 1236] Overall Loss 0.187240 Objective Loss 0.187240 LR 0.000500 Time 0.022493 +2023-10-02 21:26:53,517 - Epoch: [120][ 670/ 1236] Overall Loss 0.187129 Objective Loss 0.187129 LR 0.000500 Time 0.022468 +2023-10-02 21:26:53,728 - Epoch: [120][ 680/ 1236] Overall Loss 0.186988 Objective Loss 0.186988 LR 0.000500 Time 0.022448 +2023-10-02 21:26:53,942 - Epoch: [120][ 690/ 1236] Overall Loss 0.187366 Objective Loss 0.187366 LR 0.000500 Time 0.022431 +2023-10-02 21:26:54,155 - Epoch: [120][ 700/ 1236] Overall Loss 0.187385 Objective Loss 0.187385 LR 0.000500 Time 0.022414 +2023-10-02 21:26:54,369 - Epoch: [120][ 710/ 1236] Overall Loss 0.187237 Objective Loss 0.187237 LR 0.000500 Time 0.022398 +2023-10-02 21:26:54,582 - Epoch: [120][ 720/ 1236] Overall Loss 0.187204 Objective Loss 0.187204 LR 0.000500 Time 0.022382 +2023-10-02 21:26:54,796 - Epoch: [120][ 730/ 1236] Overall Loss 0.187025 Objective Loss 0.187025 LR 0.000500 Time 0.022368 +2023-10-02 21:26:55,009 - Epoch: [120][ 740/ 1236] Overall Loss 0.186898 Objective Loss 0.186898 LR 0.000500 Time 0.022353 +2023-10-02 21:26:55,223 - Epoch: [120][ 750/ 1236] Overall Loss 0.186773 Objective Loss 0.186773 LR 0.000500 Time 0.022338 +2023-10-02 21:26:55,436 - Epoch: [120][ 760/ 1236] Overall Loss 0.186750 Objective Loss 0.186750 LR 0.000500 Time 0.022324 +2023-10-02 21:26:55,650 - Epoch: [120][ 770/ 1236] Overall Loss 0.186870 Objective Loss 0.186870 LR 0.000500 Time 0.022310 +2023-10-02 21:26:55,863 - Epoch: [120][ 780/ 1236] Overall Loss 0.186937 Objective Loss 0.186937 LR 0.000500 Time 0.022297 +2023-10-02 21:26:56,076 - Epoch: [120][ 790/ 1236] Overall Loss 0.187095 Objective Loss 0.187095 LR 0.000500 Time 0.022282 +2023-10-02 21:26:56,287 - Epoch: [120][ 800/ 1236] Overall Loss 0.187215 Objective Loss 0.187215 LR 0.000500 Time 0.022267 +2023-10-02 21:26:56,497 - Epoch: [120][ 810/ 1236] Overall Loss 0.187402 Objective Loss 0.187402 LR 0.000500 Time 0.022249 +2023-10-02 21:26:56,708 - Epoch: [120][ 820/ 1236] Overall Loss 0.187593 Objective Loss 0.187593 LR 0.000500 Time 0.022235 +2023-10-02 21:26:56,919 - Epoch: [120][ 830/ 1236] Overall Loss 0.187895 Objective Loss 0.187895 LR 0.000500 Time 0.022220 +2023-10-02 21:26:57,130 - Epoch: [120][ 840/ 1236] Overall Loss 0.188162 Objective Loss 0.188162 LR 0.000500 Time 0.022206 +2023-10-02 21:26:57,340 - Epoch: [120][ 850/ 1236] Overall Loss 0.188141 Objective Loss 0.188141 LR 0.000500 Time 0.022193 +2023-10-02 21:26:57,551 - Epoch: [120][ 860/ 1236] Overall Loss 0.187955 Objective Loss 0.187955 LR 0.000500 Time 0.022179 +2023-10-02 21:26:57,762 - Epoch: [120][ 870/ 1236] Overall Loss 0.188005 Objective Loss 0.188005 LR 0.000500 Time 0.022166 +2023-10-02 21:26:57,973 - Epoch: [120][ 880/ 1236] Overall Loss 0.188018 Objective Loss 0.188018 LR 0.000500 Time 0.022153 +2023-10-02 21:26:58,183 - Epoch: [120][ 890/ 1236] Overall Loss 0.188261 Objective Loss 0.188261 LR 0.000500 Time 0.022141 +2023-10-02 21:26:58,394 - Epoch: [120][ 900/ 1236] Overall Loss 0.188250 Objective Loss 0.188250 LR 0.000500 Time 0.022128 +2023-10-02 21:26:58,605 - Epoch: [120][ 910/ 1236] Overall Loss 0.188429 Objective Loss 0.188429 LR 0.000500 Time 0.022116 +2023-10-02 21:26:58,816 - Epoch: [120][ 920/ 1236] Overall Loss 0.188332 Objective Loss 0.188332 LR 0.000500 Time 0.022105 +2023-10-02 21:26:59,027 - Epoch: [120][ 930/ 1236] Overall Loss 0.188366 Objective Loss 0.188366 LR 0.000500 Time 0.022093 +2023-10-02 21:26:59,237 - Epoch: [120][ 940/ 1236] Overall Loss 0.188060 Objective Loss 0.188060 LR 0.000500 Time 0.022082 +2023-10-02 21:26:59,448 - Epoch: [120][ 950/ 1236] Overall Loss 0.188053 Objective Loss 0.188053 LR 0.000500 Time 0.022071 +2023-10-02 21:26:59,659 - Epoch: [120][ 960/ 1236] Overall Loss 0.188104 Objective Loss 0.188104 LR 0.000500 Time 0.022060 +2023-10-02 21:26:59,870 - Epoch: [120][ 970/ 1236] Overall Loss 0.188087 Objective Loss 0.188087 LR 0.000500 Time 0.022050 +2023-10-02 21:27:00,081 - Epoch: [120][ 980/ 1236] Overall Loss 0.188012 Objective Loss 0.188012 LR 0.000500 Time 0.022040 +2023-10-02 21:27:00,291 - Epoch: [120][ 990/ 1236] Overall Loss 0.188337 Objective Loss 0.188337 LR 0.000500 Time 0.022029 +2023-10-02 21:27:00,502 - Epoch: [120][ 1000/ 1236] Overall Loss 0.188218 Objective Loss 0.188218 LR 0.000500 Time 0.022020 +2023-10-02 21:27:00,717 - Epoch: [120][ 1010/ 1236] Overall Loss 0.188402 Objective Loss 0.188402 LR 0.000500 Time 0.022014 +2023-10-02 21:27:00,939 - Epoch: [120][ 1020/ 1236] Overall Loss 0.188177 Objective Loss 0.188177 LR 0.000500 Time 0.022015 +2023-10-02 21:27:01,157 - Epoch: [120][ 1030/ 1236] Overall Loss 0.188063 Objective Loss 0.188063 LR 0.000500 Time 0.022013 +2023-10-02 21:27:01,378 - Epoch: [120][ 1040/ 1236] Overall Loss 0.187736 Objective Loss 0.187736 LR 0.000500 Time 0.022014 +2023-10-02 21:27:01,597 - Epoch: [120][ 1050/ 1236] Overall Loss 0.188003 Objective Loss 0.188003 LR 0.000500 Time 0.022012 +2023-10-02 21:27:01,812 - Epoch: [120][ 1060/ 1236] Overall Loss 0.188098 Objective Loss 0.188098 LR 0.000500 Time 0.022007 +2023-10-02 21:27:02,023 - Epoch: [120][ 1070/ 1236] Overall Loss 0.188233 Objective Loss 0.188233 LR 0.000500 Time 0.021997 +2023-10-02 21:27:02,245 - Epoch: [120][ 1080/ 1236] Overall Loss 0.188143 Objective Loss 0.188143 LR 0.000500 Time 0.021998 +2023-10-02 21:27:02,463 - Epoch: [120][ 1090/ 1236] Overall Loss 0.188072 Objective Loss 0.188072 LR 0.000500 Time 0.021996 +2023-10-02 21:27:02,685 - Epoch: [120][ 1100/ 1236] Overall Loss 0.188106 Objective Loss 0.188106 LR 0.000500 Time 0.021997 +2023-10-02 21:27:02,903 - Epoch: [120][ 1110/ 1236] Overall Loss 0.188391 Objective Loss 0.188391 LR 0.000500 Time 0.021995 +2023-10-02 21:27:03,125 - Epoch: [120][ 1120/ 1236] Overall Loss 0.188272 Objective Loss 0.188272 LR 0.000500 Time 0.021996 +2023-10-02 21:27:03,343 - Epoch: [120][ 1130/ 1236] Overall Loss 0.188303 Objective Loss 0.188303 LR 0.000500 Time 0.021995 +2023-10-02 21:27:03,564 - Epoch: [120][ 1140/ 1236] Overall Loss 0.188308 Objective Loss 0.188308 LR 0.000500 Time 0.021996 +2023-10-02 21:27:03,783 - Epoch: [120][ 1150/ 1236] Overall Loss 0.188458 Objective Loss 0.188458 LR 0.000500 Time 0.021994 +2023-10-02 21:27:04,005 - Epoch: [120][ 1160/ 1236] Overall Loss 0.188519 Objective Loss 0.188519 LR 0.000500 Time 0.021995 +2023-10-02 21:27:04,223 - Epoch: [120][ 1170/ 1236] Overall Loss 0.188484 Objective Loss 0.188484 LR 0.000500 Time 0.021993 +2023-10-02 21:27:04,445 - Epoch: [120][ 1180/ 1236] Overall Loss 0.188468 Objective Loss 0.188468 LR 0.000500 Time 0.021994 +2023-10-02 21:27:04,663 - Epoch: [120][ 1190/ 1236] Overall Loss 0.188453 Objective Loss 0.188453 LR 0.000500 Time 0.021993 +2023-10-02 21:27:04,885 - Epoch: [120][ 1200/ 1236] Overall Loss 0.188507 Objective Loss 0.188507 LR 0.000500 Time 0.021994 +2023-10-02 21:27:05,103 - Epoch: [120][ 1210/ 1236] Overall Loss 0.188456 Objective Loss 0.188456 LR 0.000500 Time 0.021992 +2023-10-02 21:27:05,325 - Epoch: [120][ 1220/ 1236] Overall Loss 0.188559 Objective Loss 0.188559 LR 0.000500 Time 0.021993 +2023-10-02 21:27:05,598 - Epoch: [120][ 1230/ 1236] Overall Loss 0.188532 Objective Loss 0.188532 LR 0.000500 Time 0.022036 +2023-10-02 21:27:05,721 - Epoch: [120][ 1236/ 1236] Overall Loss 0.188526 Objective Loss 0.188526 Top1 89.613035 Top5 99.389002 LR 0.000500 Time 0.022028 +2023-10-02 21:27:05,851 - --- validate (epoch=120)----------- +2023-10-02 21:27:05,851 - 29943 samples (256 per mini-batch) +2023-10-02 21:27:06,324 - Epoch: [120][ 10/ 117] Loss 0.313157 Top1 85.585938 Top5 98.203125 +2023-10-02 21:27:06,476 - Epoch: [120][ 20/ 117] Loss 0.304061 Top1 85.449219 Top5 98.398438 +2023-10-02 21:27:06,628 - Epoch: [120][ 30/ 117] Loss 0.301906 Top1 85.572917 Top5 98.580729 +2023-10-02 21:27:06,781 - Epoch: [120][ 40/ 117] Loss 0.300058 Top1 85.585938 Top5 98.505859 +2023-10-02 21:27:06,932 - Epoch: [120][ 50/ 117] Loss 0.299688 Top1 85.789062 Top5 98.523438 +2023-10-02 21:27:07,084 - Epoch: [120][ 60/ 117] Loss 0.304997 Top1 85.768229 Top5 98.450521 +2023-10-02 21:27:07,236 - Epoch: [120][ 70/ 117] Loss 0.299036 Top1 86.015625 Top5 98.504464 +2023-10-02 21:27:07,388 - Epoch: [120][ 80/ 117] Loss 0.299401 Top1 85.991211 Top5 98.471680 +2023-10-02 21:27:07,541 - Epoch: [120][ 90/ 117] Loss 0.299153 Top1 85.889757 Top5 98.485243 +2023-10-02 21:27:07,694 - Epoch: [120][ 100/ 117] Loss 0.297403 Top1 85.839844 Top5 98.445312 +2023-10-02 21:27:07,853 - Epoch: [120][ 110/ 117] Loss 0.295766 Top1 85.809659 Top5 98.430398 +2023-10-02 21:27:07,942 - Epoch: [120][ 117/ 117] Loss 0.296928 Top1 85.769629 Top5 98.433691 +2023-10-02 21:27:08,082 - ==> Top1: 85.770 Top5: 98.434 Loss: 0.297 + +2023-10-02 21:27:08,083 - ==> Confusion: +[[ 947 0 4 0 2 3 0 0 6 54 1 1 1 4 4 1 5 0 1 0 16] + [ 0 1050 0 1 8 25 0 22 1 0 0 0 3 0 1 4 1 0 7 1 7] + [ 4 1 971 16 3 0 22 5 0 0 1 1 7 2 1 3 0 1 10 4 4] + [ 2 2 9 979 1 5 3 0 7 0 4 0 8 2 30 2 0 4 10 1 20] + [ 29 4 1 0 971 5 0 0 1 11 0 0 1 3 7 4 9 0 0 3 1] + [ 3 34 0 0 2 1004 1 20 1 5 1 3 3 12 5 1 2 0 3 1 15] + [ 0 3 20 0 0 1 1135 6 0 0 5 0 0 0 0 4 0 1 1 8 7] + [ 2 12 18 2 2 36 5 1054 0 5 5 7 3 5 0 1 0 1 32 15 13] + [ 16 0 0 0 1 5 0 2 980 40 11 3 4 7 14 0 4 0 1 0 1] + [ 97 2 0 0 7 4 0 0 20 945 2 0 1 27 5 1 1 2 0 0 5] + [ 1 0 7 12 1 3 0 2 22 2 955 2 2 15 4 0 3 1 6 4 11] + [ 0 0 1 0 0 10 0 1 0 1 0 960 21 11 0 1 2 17 1 5 4] + [ 0 1 1 4 2 2 2 3 1 2 1 29 992 2 1 4 2 7 2 2 8] + [ 0 0 0 0 0 6 0 0 11 11 2 4 1 1067 4 1 1 0 0 0 11] + [ 13 1 3 16 5 0 1 0 23 2 3 0 1 4 1014 0 2 1 4 0 8] + [ 0 0 1 0 6 1 0 0 0 0 0 3 8 1 0 1076 16 13 2 2 5] + [ 1 15 0 0 3 7 0 1 0 1 0 6 1 2 4 9 1101 0 1 2 7] + [ 0 0 0 1 0 2 2 0 0 1 0 3 21 1 5 7 1 990 2 0 2] + [ 4 3 2 12 0 1 1 21 5 1 3 1 3 0 10 0 0 0 993 0 8] + [ 0 4 3 5 0 3 11 6 0 0 0 16 3 1 1 2 10 1 1 1076 9] + [ 135 156 128 78 69 134 32 64 106 81 146 112 325 279 133 66 88 61 131 159 5422]] + +2023-10-02 21:27:08,084 - ==> Best [Top1: 85.770 Top5: 98.434 Sparsity:0.00 Params: 169472 on epoch: 120] +2023-10-02 21:27:08,084 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:27:08,098 - + +2023-10-02 21:27:08,098 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:27:09,237 - Epoch: [121][ 10/ 1236] Overall Loss 0.175561 Objective Loss 0.175561 LR 0.000500 Time 0.113812 +2023-10-02 21:27:09,445 - Epoch: [121][ 20/ 1236] Overall Loss 0.177328 Objective Loss 0.177328 LR 0.000500 Time 0.067331 +2023-10-02 21:27:09,654 - Epoch: [121][ 30/ 1236] Overall Loss 0.178078 Objective Loss 0.178078 LR 0.000500 Time 0.051815 +2023-10-02 21:27:09,860 - Epoch: [121][ 40/ 1236] Overall Loss 0.180857 Objective Loss 0.180857 LR 0.000500 Time 0.044022 +2023-10-02 21:27:10,067 - Epoch: [121][ 50/ 1236] Overall Loss 0.182416 Objective Loss 0.182416 LR 0.000500 Time 0.039330 +2023-10-02 21:27:10,275 - Epoch: [121][ 60/ 1236] Overall Loss 0.182594 Objective Loss 0.182594 LR 0.000500 Time 0.036235 +2023-10-02 21:27:10,482 - Epoch: [121][ 70/ 1236] Overall Loss 0.185178 Objective Loss 0.185178 LR 0.000500 Time 0.033996 +2023-10-02 21:27:10,690 - Epoch: [121][ 80/ 1236] Overall Loss 0.187692 Objective Loss 0.187692 LR 0.000500 Time 0.032343 +2023-10-02 21:27:10,900 - Epoch: [121][ 90/ 1236] Overall Loss 0.189701 Objective Loss 0.189701 LR 0.000500 Time 0.031056 +2023-10-02 21:27:11,108 - Epoch: [121][ 100/ 1236] Overall Loss 0.189763 Objective Loss 0.189763 LR 0.000500 Time 0.030036 +2023-10-02 21:27:11,317 - Epoch: [121][ 110/ 1236] Overall Loss 0.189229 Objective Loss 0.189229 LR 0.000500 Time 0.029189 +2023-10-02 21:27:11,527 - Epoch: [121][ 120/ 1236] Overall Loss 0.187886 Objective Loss 0.187886 LR 0.000500 Time 0.028505 +2023-10-02 21:27:11,734 - Epoch: [121][ 130/ 1236] Overall Loss 0.187968 Objective Loss 0.187968 LR 0.000500 Time 0.027904 +2023-10-02 21:27:11,944 - Epoch: [121][ 140/ 1236] Overall Loss 0.187155 Objective Loss 0.187155 LR 0.000500 Time 0.027410 +2023-10-02 21:27:12,151 - Epoch: [121][ 150/ 1236] Overall Loss 0.187923 Objective Loss 0.187923 LR 0.000500 Time 0.026958 +2023-10-02 21:27:12,362 - Epoch: [121][ 160/ 1236] Overall Loss 0.188657 Objective Loss 0.188657 LR 0.000500 Time 0.026590 +2023-10-02 21:27:12,571 - Epoch: [121][ 170/ 1236] Overall Loss 0.188191 Objective Loss 0.188191 LR 0.000500 Time 0.026253 +2023-10-02 21:27:12,783 - Epoch: [121][ 180/ 1236] Overall Loss 0.188087 Objective Loss 0.188087 LR 0.000500 Time 0.025973 +2023-10-02 21:27:12,992 - Epoch: [121][ 190/ 1236] Overall Loss 0.189520 Objective Loss 0.189520 LR 0.000500 Time 0.025704 +2023-10-02 21:27:13,203 - Epoch: [121][ 200/ 1236] Overall Loss 0.188645 Objective Loss 0.188645 LR 0.000500 Time 0.025468 +2023-10-02 21:27:13,413 - Epoch: [121][ 210/ 1236] Overall Loss 0.189046 Objective Loss 0.189046 LR 0.000500 Time 0.025249 +2023-10-02 21:27:13,625 - Epoch: [121][ 220/ 1236] Overall Loss 0.189762 Objective Loss 0.189762 LR 0.000500 Time 0.025065 +2023-10-02 21:27:13,833 - Epoch: [121][ 230/ 1236] Overall Loss 0.189651 Objective Loss 0.189651 LR 0.000500 Time 0.024880 +2023-10-02 21:27:14,046 - Epoch: [121][ 240/ 1236] Overall Loss 0.190787 Objective Loss 0.190787 LR 0.000500 Time 0.024728 +2023-10-02 21:27:14,257 - Epoch: [121][ 250/ 1236] Overall Loss 0.189357 Objective Loss 0.189357 LR 0.000500 Time 0.024581 +2023-10-02 21:27:14,472 - Epoch: [121][ 260/ 1236] Overall Loss 0.189207 Objective Loss 0.189207 LR 0.000500 Time 0.024458 +2023-10-02 21:27:14,683 - Epoch: [121][ 270/ 1236] Overall Loss 0.188947 Objective Loss 0.188947 LR 0.000500 Time 0.024333 +2023-10-02 21:27:14,898 - Epoch: [121][ 280/ 1236] Overall Loss 0.189073 Objective Loss 0.189073 LR 0.000500 Time 0.024231 +2023-10-02 21:27:15,110 - Epoch: [121][ 290/ 1236] Overall Loss 0.189128 Objective Loss 0.189128 LR 0.000500 Time 0.024123 +2023-10-02 21:27:15,324 - Epoch: [121][ 300/ 1236] Overall Loss 0.188553 Objective Loss 0.188553 LR 0.000500 Time 0.024030 +2023-10-02 21:27:15,532 - Epoch: [121][ 310/ 1236] Overall Loss 0.188234 Objective Loss 0.188234 LR 0.000500 Time 0.023924 +2023-10-02 21:27:15,741 - Epoch: [121][ 320/ 1236] Overall Loss 0.187839 Objective Loss 0.187839 LR 0.000500 Time 0.023829 +2023-10-02 21:27:15,950 - Epoch: [121][ 330/ 1236] Overall Loss 0.188081 Objective Loss 0.188081 LR 0.000500 Time 0.023736 +2023-10-02 21:27:16,159 - Epoch: [121][ 340/ 1236] Overall Loss 0.187979 Objective Loss 0.187979 LR 0.000500 Time 0.023652 +2023-10-02 21:27:16,368 - Epoch: [121][ 350/ 1236] Overall Loss 0.187678 Objective Loss 0.187678 LR 0.000500 Time 0.023568 +2023-10-02 21:27:16,577 - Epoch: [121][ 360/ 1236] Overall Loss 0.187532 Objective Loss 0.187532 LR 0.000500 Time 0.023493 +2023-10-02 21:27:16,786 - Epoch: [121][ 370/ 1236] Overall Loss 0.186939 Objective Loss 0.186939 LR 0.000500 Time 0.023419 +2023-10-02 21:27:16,996 - Epoch: [121][ 380/ 1236] Overall Loss 0.186987 Objective Loss 0.186987 LR 0.000500 Time 0.023352 +2023-10-02 21:27:17,204 - Epoch: [121][ 390/ 1236] Overall Loss 0.187106 Objective Loss 0.187106 LR 0.000500 Time 0.023285 +2023-10-02 21:27:17,413 - Epoch: [121][ 400/ 1236] Overall Loss 0.187023 Objective Loss 0.187023 LR 0.000500 Time 0.023225 +2023-10-02 21:27:17,622 - Epoch: [121][ 410/ 1236] Overall Loss 0.186967 Objective Loss 0.186967 LR 0.000500 Time 0.023163 +2023-10-02 21:27:17,831 - Epoch: [121][ 420/ 1236] Overall Loss 0.186184 Objective Loss 0.186184 LR 0.000500 Time 0.023110 +2023-10-02 21:27:18,040 - Epoch: [121][ 430/ 1236] Overall Loss 0.186219 Objective Loss 0.186219 LR 0.000500 Time 0.023054 +2023-10-02 21:27:18,249 - Epoch: [121][ 440/ 1236] Overall Loss 0.186176 Objective Loss 0.186176 LR 0.000500 Time 0.023004 +2023-10-02 21:27:18,457 - Epoch: [121][ 450/ 1236] Overall Loss 0.186538 Objective Loss 0.186538 LR 0.000500 Time 0.022953 +2023-10-02 21:27:18,667 - Epoch: [121][ 460/ 1236] Overall Loss 0.187126 Objective Loss 0.187126 LR 0.000500 Time 0.022908 +2023-10-02 21:27:18,876 - Epoch: [121][ 470/ 1236] Overall Loss 0.187547 Objective Loss 0.187547 LR 0.000500 Time 0.022862 +2023-10-02 21:27:19,085 - Epoch: [121][ 480/ 1236] Overall Loss 0.187327 Objective Loss 0.187327 LR 0.000500 Time 0.022821 +2023-10-02 21:27:19,293 - Epoch: [121][ 490/ 1236] Overall Loss 0.186936 Objective Loss 0.186936 LR 0.000500 Time 0.022778 +2023-10-02 21:27:19,503 - Epoch: [121][ 500/ 1236] Overall Loss 0.187023 Objective Loss 0.187023 LR 0.000500 Time 0.022740 +2023-10-02 21:27:19,711 - Epoch: [121][ 510/ 1236] Overall Loss 0.187162 Objective Loss 0.187162 LR 0.000500 Time 0.022700 +2023-10-02 21:27:19,921 - Epoch: [121][ 520/ 1236] Overall Loss 0.186697 Objective Loss 0.186697 LR 0.000500 Time 0.022666 +2023-10-02 21:27:20,130 - Epoch: [121][ 530/ 1236] Overall Loss 0.187075 Objective Loss 0.187075 LR 0.000500 Time 0.022629 +2023-10-02 21:27:20,339 - Epoch: [121][ 540/ 1236] Overall Loss 0.187138 Objective Loss 0.187138 LR 0.000500 Time 0.022597 +2023-10-02 21:27:20,547 - Epoch: [121][ 550/ 1236] Overall Loss 0.187030 Objective Loss 0.187030 LR 0.000500 Time 0.022563 +2023-10-02 21:27:20,757 - Epoch: [121][ 560/ 1236] Overall Loss 0.186610 Objective Loss 0.186610 LR 0.000500 Time 0.022533 +2023-10-02 21:27:20,965 - Epoch: [121][ 570/ 1236] Overall Loss 0.187056 Objective Loss 0.187056 LR 0.000500 Time 0.022502 +2023-10-02 21:27:21,175 - Epoch: [121][ 580/ 1236] Overall Loss 0.187145 Objective Loss 0.187145 LR 0.000500 Time 0.022474 +2023-10-02 21:27:21,384 - Epoch: [121][ 590/ 1236] Overall Loss 0.187136 Objective Loss 0.187136 LR 0.000500 Time 0.022444 +2023-10-02 21:27:21,593 - Epoch: [121][ 600/ 1236] Overall Loss 0.187001 Objective Loss 0.187001 LR 0.000500 Time 0.022418 +2023-10-02 21:27:21,802 - Epoch: [121][ 610/ 1236] Overall Loss 0.187131 Objective Loss 0.187131 LR 0.000500 Time 0.022391 +2023-10-02 21:27:22,011 - Epoch: [121][ 620/ 1236] Overall Loss 0.186784 Objective Loss 0.186784 LR 0.000500 Time 0.022367 +2023-10-02 21:27:22,220 - Epoch: [121][ 630/ 1236] Overall Loss 0.187080 Objective Loss 0.187080 LR 0.000500 Time 0.022341 +2023-10-02 21:27:22,429 - Epoch: [121][ 640/ 1236] Overall Loss 0.187199 Objective Loss 0.187199 LR 0.000500 Time 0.022318 +2023-10-02 21:27:22,638 - Epoch: [121][ 650/ 1236] Overall Loss 0.187279 Objective Loss 0.187279 LR 0.000500 Time 0.022294 +2023-10-02 21:27:22,848 - Epoch: [121][ 660/ 1236] Overall Loss 0.186869 Objective Loss 0.186869 LR 0.000500 Time 0.022273 +2023-10-02 21:27:23,056 - Epoch: [121][ 670/ 1236] Overall Loss 0.187022 Objective Loss 0.187022 LR 0.000500 Time 0.022250 +2023-10-02 21:27:23,267 - Epoch: [121][ 680/ 1236] Overall Loss 0.187044 Objective Loss 0.187044 LR 0.000500 Time 0.022232 +2023-10-02 21:27:23,474 - Epoch: [121][ 690/ 1236] Overall Loss 0.186959 Objective Loss 0.186959 LR 0.000500 Time 0.022210 +2023-10-02 21:27:23,684 - Epoch: [121][ 700/ 1236] Overall Loss 0.187046 Objective Loss 0.187046 LR 0.000500 Time 0.022191 +2023-10-02 21:27:23,892 - Epoch: [121][ 710/ 1236] Overall Loss 0.187542 Objective Loss 0.187542 LR 0.000500 Time 0.022171 +2023-10-02 21:27:24,101 - Epoch: [121][ 720/ 1236] Overall Loss 0.187364 Objective Loss 0.187364 LR 0.000500 Time 0.022153 +2023-10-02 21:27:24,310 - Epoch: [121][ 730/ 1236] Overall Loss 0.187026 Objective Loss 0.187026 LR 0.000500 Time 0.022133 +2023-10-02 21:27:24,519 - Epoch: [121][ 740/ 1236] Overall Loss 0.187292 Objective Loss 0.187292 LR 0.000500 Time 0.022116 +2023-10-02 21:27:24,728 - Epoch: [121][ 750/ 1236] Overall Loss 0.187683 Objective Loss 0.187683 LR 0.000500 Time 0.022098 +2023-10-02 21:27:24,938 - Epoch: [121][ 760/ 1236] Overall Loss 0.187519 Objective Loss 0.187519 LR 0.000500 Time 0.022082 +2023-10-02 21:27:25,146 - Epoch: [121][ 770/ 1236] Overall Loss 0.187726 Objective Loss 0.187726 LR 0.000500 Time 0.022064 +2023-10-02 21:27:25,357 - Epoch: [121][ 780/ 1236] Overall Loss 0.187740 Objective Loss 0.187740 LR 0.000500 Time 0.022051 +2023-10-02 21:27:25,564 - Epoch: [121][ 790/ 1236] Overall Loss 0.187603 Objective Loss 0.187603 LR 0.000500 Time 0.022034 +2023-10-02 21:27:25,774 - Epoch: [121][ 800/ 1236] Overall Loss 0.187710 Objective Loss 0.187710 LR 0.000500 Time 0.022020 +2023-10-02 21:27:25,983 - Epoch: [121][ 810/ 1236] Overall Loss 0.187586 Objective Loss 0.187586 LR 0.000500 Time 0.022005 +2023-10-02 21:27:26,192 - Epoch: [121][ 820/ 1236] Overall Loss 0.187518 Objective Loss 0.187518 LR 0.000500 Time 0.021992 +2023-10-02 21:27:26,401 - Epoch: [121][ 830/ 1236] Overall Loss 0.187373 Objective Loss 0.187373 LR 0.000500 Time 0.021976 +2023-10-02 21:27:26,611 - Epoch: [121][ 840/ 1236] Overall Loss 0.187554 Objective Loss 0.187554 LR 0.000500 Time 0.021964 +2023-10-02 21:27:26,820 - Epoch: [121][ 850/ 1236] Overall Loss 0.187645 Objective Loss 0.187645 LR 0.000500 Time 0.021949 +2023-10-02 21:27:27,029 - Epoch: [121][ 860/ 1236] Overall Loss 0.187286 Objective Loss 0.187286 LR 0.000500 Time 0.021937 +2023-10-02 21:27:27,237 - Epoch: [121][ 870/ 1236] Overall Loss 0.187417 Objective Loss 0.187417 LR 0.000500 Time 0.021923 +2023-10-02 21:27:27,447 - Epoch: [121][ 880/ 1236] Overall Loss 0.187391 Objective Loss 0.187391 LR 0.000500 Time 0.021911 +2023-10-02 21:27:27,656 - Epoch: [121][ 890/ 1236] Overall Loss 0.187457 Objective Loss 0.187457 LR 0.000500 Time 0.021898 +2023-10-02 21:27:27,865 - Epoch: [121][ 900/ 1236] Overall Loss 0.187567 Objective Loss 0.187567 LR 0.000500 Time 0.021887 +2023-10-02 21:27:28,074 - Epoch: [121][ 910/ 1236] Overall Loss 0.187495 Objective Loss 0.187495 LR 0.000500 Time 0.021875 +2023-10-02 21:27:28,283 - Epoch: [121][ 920/ 1236] Overall Loss 0.187533 Objective Loss 0.187533 LR 0.000500 Time 0.021864 +2023-10-02 21:27:28,493 - Epoch: [121][ 930/ 1236] Overall Loss 0.187650 Objective Loss 0.187650 LR 0.000500 Time 0.021853 +2023-10-02 21:27:28,702 - Epoch: [121][ 940/ 1236] Overall Loss 0.187935 Objective Loss 0.187935 LR 0.000500 Time 0.021843 +2023-10-02 21:27:28,911 - Epoch: [121][ 950/ 1236] Overall Loss 0.187995 Objective Loss 0.187995 LR 0.000500 Time 0.021830 +2023-10-02 21:27:29,120 - Epoch: [121][ 960/ 1236] Overall Loss 0.187828 Objective Loss 0.187828 LR 0.000500 Time 0.021821 +2023-10-02 21:27:29,328 - Epoch: [121][ 970/ 1236] Overall Loss 0.187885 Objective Loss 0.187885 LR 0.000500 Time 0.021809 +2023-10-02 21:27:29,538 - Epoch: [121][ 980/ 1236] Overall Loss 0.187751 Objective Loss 0.187751 LR 0.000500 Time 0.021800 +2023-10-02 21:27:29,746 - Epoch: [121][ 990/ 1236] Overall Loss 0.187822 Objective Loss 0.187822 LR 0.000500 Time 0.021788 +2023-10-02 21:27:29,955 - Epoch: [121][ 1000/ 1236] Overall Loss 0.187727 Objective Loss 0.187727 LR 0.000500 Time 0.021779 +2023-10-02 21:27:30,164 - Epoch: [121][ 1010/ 1236] Overall Loss 0.187631 Objective Loss 0.187631 LR 0.000500 Time 0.021768 +2023-10-02 21:27:30,373 - Epoch: [121][ 1020/ 1236] Overall Loss 0.187848 Objective Loss 0.187848 LR 0.000500 Time 0.021760 +2023-10-02 21:27:30,581 - Epoch: [121][ 1030/ 1236] Overall Loss 0.187875 Objective Loss 0.187875 LR 0.000500 Time 0.021749 +2023-10-02 21:27:30,790 - Epoch: [121][ 1040/ 1236] Overall Loss 0.187740 Objective Loss 0.187740 LR 0.000500 Time 0.021741 +2023-10-02 21:27:31,000 - Epoch: [121][ 1050/ 1236] Overall Loss 0.187481 Objective Loss 0.187481 LR 0.000500 Time 0.021733 +2023-10-02 21:27:31,210 - Epoch: [121][ 1060/ 1236] Overall Loss 0.187775 Objective Loss 0.187775 LR 0.000500 Time 0.021726 +2023-10-02 21:27:31,419 - Epoch: [121][ 1070/ 1236] Overall Loss 0.187716 Objective Loss 0.187716 LR 0.000500 Time 0.021718 +2023-10-02 21:27:31,629 - Epoch: [121][ 1080/ 1236] Overall Loss 0.187633 Objective Loss 0.187633 LR 0.000500 Time 0.021711 +2023-10-02 21:27:31,838 - Epoch: [121][ 1090/ 1236] Overall Loss 0.187877 Objective Loss 0.187877 LR 0.000500 Time 0.021704 +2023-10-02 21:27:32,048 - Epoch: [121][ 1100/ 1236] Overall Loss 0.188014 Objective Loss 0.188014 LR 0.000500 Time 0.021697 +2023-10-02 21:27:32,257 - Epoch: [121][ 1110/ 1236] Overall Loss 0.187946 Objective Loss 0.187946 LR 0.000500 Time 0.021688 +2023-10-02 21:27:32,467 - Epoch: [121][ 1120/ 1236] Overall Loss 0.188080 Objective Loss 0.188080 LR 0.000500 Time 0.021682 +2023-10-02 21:27:32,676 - Epoch: [121][ 1130/ 1236] Overall Loss 0.188043 Objective Loss 0.188043 LR 0.000500 Time 0.021675 +2023-10-02 21:27:32,887 - Epoch: [121][ 1140/ 1236] Overall Loss 0.188299 Objective Loss 0.188299 LR 0.000500 Time 0.021669 +2023-10-02 21:27:33,096 - Epoch: [121][ 1150/ 1236] Overall Loss 0.188404 Objective Loss 0.188404 LR 0.000500 Time 0.021662 +2023-10-02 21:27:33,306 - Epoch: [121][ 1160/ 1236] Overall Loss 0.188315 Objective Loss 0.188315 LR 0.000500 Time 0.021656 +2023-10-02 21:27:33,515 - Epoch: [121][ 1170/ 1236] Overall Loss 0.188355 Objective Loss 0.188355 LR 0.000500 Time 0.021650 +2023-10-02 21:27:33,725 - Epoch: [121][ 1180/ 1236] Overall Loss 0.188309 Objective Loss 0.188309 LR 0.000500 Time 0.021644 +2023-10-02 21:27:33,935 - Epoch: [121][ 1190/ 1236] Overall Loss 0.188157 Objective Loss 0.188157 LR 0.000500 Time 0.021638 +2023-10-02 21:27:34,145 - Epoch: [121][ 1200/ 1236] Overall Loss 0.188300 Objective Loss 0.188300 LR 0.000500 Time 0.021633 +2023-10-02 21:27:34,354 - Epoch: [121][ 1210/ 1236] Overall Loss 0.188058 Objective Loss 0.188058 LR 0.000500 Time 0.021626 +2023-10-02 21:27:34,565 - Epoch: [121][ 1220/ 1236] Overall Loss 0.188068 Objective Loss 0.188068 LR 0.000500 Time 0.021622 +2023-10-02 21:27:34,827 - Epoch: [121][ 1230/ 1236] Overall Loss 0.188193 Objective Loss 0.188193 LR 0.000500 Time 0.021658 +2023-10-02 21:27:34,949 - Epoch: [121][ 1236/ 1236] Overall Loss 0.188350 Objective Loss 0.188350 Top1 90.224033 Top5 98.778004 LR 0.000500 Time 0.021652 +2023-10-02 21:27:35,084 - --- validate (epoch=121)----------- +2023-10-02 21:27:35,084 - 29943 samples (256 per mini-batch) +2023-10-02 21:27:35,586 - Epoch: [121][ 10/ 117] Loss 0.301305 Top1 85.078125 Top5 98.710938 +2023-10-02 21:27:35,738 - Epoch: [121][ 20/ 117] Loss 0.302714 Top1 85.605469 Top5 98.496094 +2023-10-02 21:27:35,890 - Epoch: [121][ 30/ 117] Loss 0.304069 Top1 85.703125 Top5 98.515625 +2023-10-02 21:27:36,045 - Epoch: [121][ 40/ 117] Loss 0.302986 Top1 85.751953 Top5 98.457031 +2023-10-02 21:27:36,196 - Epoch: [121][ 50/ 117] Loss 0.300586 Top1 85.617188 Top5 98.406250 +2023-10-02 21:27:36,353 - Epoch: [121][ 60/ 117] Loss 0.293101 Top1 85.774740 Top5 98.457031 +2023-10-02 21:27:36,507 - Epoch: [121][ 70/ 117] Loss 0.287225 Top1 85.837054 Top5 98.437500 +2023-10-02 21:27:36,663 - Epoch: [121][ 80/ 117] Loss 0.295044 Top1 85.766602 Top5 98.427734 +2023-10-02 21:27:36,818 - Epoch: [121][ 90/ 117] Loss 0.291530 Top1 85.846354 Top5 98.446181 +2023-10-02 21:27:36,976 - Epoch: [121][ 100/ 117] Loss 0.292034 Top1 85.808594 Top5 98.445312 +2023-10-02 21:27:37,137 - Epoch: [121][ 110/ 117] Loss 0.292955 Top1 85.827415 Top5 98.419744 +2023-10-02 21:27:37,227 - Epoch: [121][ 117/ 117] Loss 0.294373 Top1 85.816384 Top5 98.413653 +2023-10-02 21:27:37,385 - ==> Top1: 85.816 Top5: 98.414 Loss: 0.294 + +2023-10-02 21:27:37,386 - ==> Confusion: +[[ 934 1 5 0 6 2 0 1 7 73 2 0 1 1 2 0 3 0 0 0 12] + [ 0 1055 1 1 4 24 0 22 0 2 1 0 0 0 0 4 2 0 10 2 3] + [ 2 0 991 6 0 0 13 8 1 1 1 1 8 2 0 3 2 1 7 4 5] + [ 0 3 18 968 3 4 0 1 3 3 6 0 8 2 24 2 0 3 18 0 23] + [ 21 4 0 0 983 2 0 0 1 12 0 0 1 2 5 3 8 0 1 2 5] + [ 4 42 1 1 5 983 0 25 2 9 1 3 2 9 4 0 4 1 3 3 14] + [ 0 2 38 0 0 2 1114 9 0 0 4 0 0 1 0 3 0 1 1 11 5] + [ 2 21 17 0 7 22 5 1070 1 5 2 4 2 4 1 1 1 4 29 11 9] + [ 15 3 0 2 2 2 0 1 978 47 12 2 2 6 9 1 2 0 3 1 1] + [ 73 0 1 1 5 4 0 1 26 983 0 1 0 11 3 1 0 0 0 2 7] + [ 3 3 11 5 1 1 3 4 12 0 963 2 0 19 4 0 0 6 4 1 11] + [ 2 0 3 0 0 15 0 3 0 2 0 968 14 8 0 0 1 15 0 2 2] + [ 0 1 1 4 1 3 2 1 1 0 1 39 966 3 4 9 1 14 1 4 12] + [ 1 0 1 0 3 7 0 1 16 24 1 7 0 1044 4 0 0 0 0 3 7] + [ 13 0 5 12 4 0 0 0 26 6 3 0 4 1 1009 0 0 1 8 0 9] + [ 0 0 2 0 4 0 0 0 0 2 1 5 8 0 0 1071 18 9 2 7 5] + [ 2 18 2 1 3 6 0 0 0 0 0 5 0 2 3 6 1094 0 2 6 11] + [ 0 0 2 1 0 1 1 0 0 1 0 2 16 0 3 5 1 999 0 0 6] + [ 2 6 5 13 1 1 0 25 6 2 5 0 1 0 7 1 0 0 981 1 11] + [ 0 1 4 2 0 5 8 13 0 1 1 17 3 1 0 1 4 1 0 1085 5] + [ 134 164 158 65 66 114 34 98 94 112 168 94 310 237 106 45 92 53 101 203 5457]] + +2023-10-02 21:27:37,387 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:27:37,388 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:27:37,401 - + +2023-10-02 21:27:37,401 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:27:38,444 - Epoch: [122][ 10/ 1236] Overall Loss 0.196969 Objective Loss 0.196969 LR 0.000500 Time 0.104264 +2023-10-02 21:27:38,653 - Epoch: [122][ 20/ 1236] Overall Loss 0.191288 Objective Loss 0.191288 LR 0.000500 Time 0.062561 +2023-10-02 21:27:38,862 - Epoch: [122][ 30/ 1236] Overall Loss 0.188693 Objective Loss 0.188693 LR 0.000500 Time 0.048598 +2023-10-02 21:27:39,072 - Epoch: [122][ 40/ 1236] Overall Loss 0.188326 Objective Loss 0.188326 LR 0.000500 Time 0.041695 +2023-10-02 21:27:39,279 - Epoch: [122][ 50/ 1236] Overall Loss 0.184543 Objective Loss 0.184543 LR 0.000500 Time 0.037489 +2023-10-02 21:27:39,489 - Epoch: [122][ 60/ 1236] Overall Loss 0.180867 Objective Loss 0.180867 LR 0.000500 Time 0.034736 +2023-10-02 21:27:39,696 - Epoch: [122][ 70/ 1236] Overall Loss 0.182556 Objective Loss 0.182556 LR 0.000500 Time 0.032725 +2023-10-02 21:27:39,906 - Epoch: [122][ 80/ 1236] Overall Loss 0.177291 Objective Loss 0.177291 LR 0.000500 Time 0.031257 +2023-10-02 21:27:40,113 - Epoch: [122][ 90/ 1236] Overall Loss 0.179228 Objective Loss 0.179228 LR 0.000500 Time 0.030080 +2023-10-02 21:27:40,323 - Epoch: [122][ 100/ 1236] Overall Loss 0.180150 Objective Loss 0.180150 LR 0.000500 Time 0.029174 +2023-10-02 21:27:40,531 - Epoch: [122][ 110/ 1236] Overall Loss 0.180143 Objective Loss 0.180143 LR 0.000500 Time 0.028405 +2023-10-02 21:27:40,744 - Epoch: [122][ 120/ 1236] Overall Loss 0.180904 Objective Loss 0.180904 LR 0.000500 Time 0.027813 +2023-10-02 21:27:40,954 - Epoch: [122][ 130/ 1236] Overall Loss 0.181346 Objective Loss 0.181346 LR 0.000500 Time 0.027286 +2023-10-02 21:27:41,167 - Epoch: [122][ 140/ 1236] Overall Loss 0.181133 Objective Loss 0.181133 LR 0.000500 Time 0.026860 +2023-10-02 21:27:41,375 - Epoch: [122][ 150/ 1236] Overall Loss 0.180523 Objective Loss 0.180523 LR 0.000500 Time 0.026455 +2023-10-02 21:27:41,588 - Epoch: [122][ 160/ 1236] Overall Loss 0.180359 Objective Loss 0.180359 LR 0.000500 Time 0.026125 +2023-10-02 21:27:41,802 - Epoch: [122][ 170/ 1236] Overall Loss 0.181259 Objective Loss 0.181259 LR 0.000500 Time 0.025838 +2023-10-02 21:27:42,016 - Epoch: [122][ 180/ 1236] Overall Loss 0.180859 Objective Loss 0.180859 LR 0.000500 Time 0.025582 +2023-10-02 21:27:42,230 - Epoch: [122][ 190/ 1236] Overall Loss 0.179568 Objective Loss 0.179568 LR 0.000500 Time 0.025353 +2023-10-02 21:27:42,444 - Epoch: [122][ 200/ 1236] Overall Loss 0.180260 Objective Loss 0.180260 LR 0.000500 Time 0.025147 +2023-10-02 21:27:42,658 - Epoch: [122][ 210/ 1236] Overall Loss 0.179844 Objective Loss 0.179844 LR 0.000500 Time 0.024961 +2023-10-02 21:27:42,872 - Epoch: [122][ 220/ 1236] Overall Loss 0.179638 Objective Loss 0.179638 LR 0.000500 Time 0.024792 +2023-10-02 21:27:43,086 - Epoch: [122][ 230/ 1236] Overall Loss 0.179950 Objective Loss 0.179950 LR 0.000500 Time 0.024637 +2023-10-02 21:27:43,300 - Epoch: [122][ 240/ 1236] Overall Loss 0.179786 Objective Loss 0.179786 LR 0.000500 Time 0.024495 +2023-10-02 21:27:43,514 - Epoch: [122][ 250/ 1236] Overall Loss 0.178382 Objective Loss 0.178382 LR 0.000500 Time 0.024364 +2023-10-02 21:27:43,727 - Epoch: [122][ 260/ 1236] Overall Loss 0.178450 Objective Loss 0.178450 LR 0.000500 Time 0.024242 +2023-10-02 21:27:43,941 - Epoch: [122][ 270/ 1236] Overall Loss 0.178031 Objective Loss 0.178031 LR 0.000500 Time 0.024130 +2023-10-02 21:27:44,155 - Epoch: [122][ 280/ 1236] Overall Loss 0.178368 Objective Loss 0.178368 LR 0.000500 Time 0.024026 +2023-10-02 21:27:44,369 - Epoch: [122][ 290/ 1236] Overall Loss 0.178463 Objective Loss 0.178463 LR 0.000500 Time 0.023929 +2023-10-02 21:27:44,582 - Epoch: [122][ 300/ 1236] Overall Loss 0.177902 Objective Loss 0.177902 LR 0.000500 Time 0.023838 +2023-10-02 21:27:44,794 - Epoch: [122][ 310/ 1236] Overall Loss 0.177583 Objective Loss 0.177583 LR 0.000500 Time 0.023748 +2023-10-02 21:27:45,005 - Epoch: [122][ 320/ 1236] Overall Loss 0.177748 Objective Loss 0.177748 LR 0.000500 Time 0.023660 +2023-10-02 21:27:45,217 - Epoch: [122][ 330/ 1236] Overall Loss 0.178081 Objective Loss 0.178081 LR 0.000500 Time 0.023579 +2023-10-02 21:27:45,429 - Epoch: [122][ 340/ 1236] Overall Loss 0.178104 Objective Loss 0.178104 LR 0.000500 Time 0.023506 +2023-10-02 21:27:45,638 - Epoch: [122][ 350/ 1236] Overall Loss 0.178148 Objective Loss 0.178148 LR 0.000500 Time 0.023431 +2023-10-02 21:27:45,849 - Epoch: [122][ 360/ 1236] Overall Loss 0.178152 Objective Loss 0.178152 LR 0.000500 Time 0.023364 +2023-10-02 21:27:46,056 - Epoch: [122][ 370/ 1236] Overall Loss 0.178781 Objective Loss 0.178781 LR 0.000500 Time 0.023291 +2023-10-02 21:27:46,265 - Epoch: [122][ 380/ 1236] Overall Loss 0.179312 Objective Loss 0.179312 LR 0.000500 Time 0.023228 +2023-10-02 21:27:46,473 - Epoch: [122][ 390/ 1236] Overall Loss 0.179094 Objective Loss 0.179094 LR 0.000500 Time 0.023162 +2023-10-02 21:27:46,684 - Epoch: [122][ 400/ 1236] Overall Loss 0.179236 Objective Loss 0.179236 LR 0.000500 Time 0.023109 +2023-10-02 21:27:46,891 - Epoch: [122][ 410/ 1236] Overall Loss 0.179678 Objective Loss 0.179678 LR 0.000500 Time 0.023049 +2023-10-02 21:27:47,102 - Epoch: [122][ 420/ 1236] Overall Loss 0.179290 Objective Loss 0.179290 LR 0.000500 Time 0.023002 +2023-10-02 21:27:47,309 - Epoch: [122][ 430/ 1236] Overall Loss 0.180097 Objective Loss 0.180097 LR 0.000500 Time 0.022948 +2023-10-02 21:27:47,519 - Epoch: [122][ 440/ 1236] Overall Loss 0.180258 Objective Loss 0.180258 LR 0.000500 Time 0.022904 +2023-10-02 21:27:47,727 - Epoch: [122][ 450/ 1236] Overall Loss 0.180272 Objective Loss 0.180272 LR 0.000500 Time 0.022855 +2023-10-02 21:27:47,937 - Epoch: [122][ 460/ 1236] Overall Loss 0.180360 Objective Loss 0.180360 LR 0.000500 Time 0.022815 +2023-10-02 21:27:48,145 - Epoch: [122][ 470/ 1236] Overall Loss 0.180303 Objective Loss 0.180303 LR 0.000500 Time 0.022772 +2023-10-02 21:27:48,357 - Epoch: [122][ 480/ 1236] Overall Loss 0.180562 Objective Loss 0.180562 LR 0.000500 Time 0.022738 +2023-10-02 21:27:48,565 - Epoch: [122][ 490/ 1236] Overall Loss 0.180725 Objective Loss 0.180725 LR 0.000500 Time 0.022699 +2023-10-02 21:27:48,775 - Epoch: [122][ 500/ 1236] Overall Loss 0.180595 Objective Loss 0.180595 LR 0.000500 Time 0.022664 +2023-10-02 21:27:48,984 - Epoch: [122][ 510/ 1236] Overall Loss 0.181510 Objective Loss 0.181510 LR 0.000500 Time 0.022627 +2023-10-02 21:27:49,192 - Epoch: [122][ 520/ 1236] Overall Loss 0.181563 Objective Loss 0.181563 LR 0.000500 Time 0.022592 +2023-10-02 21:27:49,401 - Epoch: [122][ 530/ 1236] Overall Loss 0.181490 Objective Loss 0.181490 LR 0.000500 Time 0.022559 +2023-10-02 21:27:49,608 - Epoch: [122][ 540/ 1236] Overall Loss 0.181426 Objective Loss 0.181426 LR 0.000500 Time 0.022525 +2023-10-02 21:27:49,818 - Epoch: [122][ 550/ 1236] Overall Loss 0.181385 Objective Loss 0.181385 LR 0.000500 Time 0.022497 +2023-10-02 21:27:50,026 - Epoch: [122][ 560/ 1236] Overall Loss 0.181426 Objective Loss 0.181426 LR 0.000500 Time 0.022465 +2023-10-02 21:27:50,234 - Epoch: [122][ 570/ 1236] Overall Loss 0.181297 Objective Loss 0.181297 LR 0.000500 Time 0.022435 +2023-10-02 21:27:50,440 - Epoch: [122][ 580/ 1236] Overall Loss 0.181142 Objective Loss 0.181142 LR 0.000500 Time 0.022403 +2023-10-02 21:27:50,647 - Epoch: [122][ 590/ 1236] Overall Loss 0.181132 Objective Loss 0.181132 LR 0.000500 Time 0.022373 +2023-10-02 21:27:50,853 - Epoch: [122][ 600/ 1236] Overall Loss 0.181196 Objective Loss 0.181196 LR 0.000500 Time 0.022343 +2023-10-02 21:27:51,060 - Epoch: [122][ 610/ 1236] Overall Loss 0.181124 Objective Loss 0.181124 LR 0.000500 Time 0.022315 +2023-10-02 21:27:51,265 - Epoch: [122][ 620/ 1236] Overall Loss 0.181302 Objective Loss 0.181302 LR 0.000500 Time 0.022286 +2023-10-02 21:27:51,472 - Epoch: [122][ 630/ 1236] Overall Loss 0.181006 Objective Loss 0.181006 LR 0.000500 Time 0.022261 +2023-10-02 21:27:51,679 - Epoch: [122][ 640/ 1236] Overall Loss 0.180840 Objective Loss 0.180840 LR 0.000500 Time 0.022235 +2023-10-02 21:27:51,889 - Epoch: [122][ 650/ 1236] Overall Loss 0.181059 Objective Loss 0.181059 LR 0.000500 Time 0.022215 +2023-10-02 21:27:52,095 - Epoch: [122][ 660/ 1236] Overall Loss 0.181094 Objective Loss 0.181094 LR 0.000500 Time 0.022190 +2023-10-02 21:27:52,300 - Epoch: [122][ 670/ 1236] Overall Loss 0.181120 Objective Loss 0.181120 LR 0.000500 Time 0.022165 +2023-10-02 21:27:52,505 - Epoch: [122][ 680/ 1236] Overall Loss 0.181052 Objective Loss 0.181052 LR 0.000500 Time 0.022139 +2023-10-02 21:27:52,711 - Epoch: [122][ 690/ 1236] Overall Loss 0.181278 Objective Loss 0.181278 LR 0.000500 Time 0.022116 +2023-10-02 21:27:52,915 - Epoch: [122][ 700/ 1236] Overall Loss 0.181525 Objective Loss 0.181525 LR 0.000500 Time 0.022092 +2023-10-02 21:27:53,121 - Epoch: [122][ 710/ 1236] Overall Loss 0.181247 Objective Loss 0.181247 LR 0.000500 Time 0.022070 +2023-10-02 21:27:53,325 - Epoch: [122][ 720/ 1236] Overall Loss 0.181145 Objective Loss 0.181145 LR 0.000500 Time 0.022046 +2023-10-02 21:27:53,531 - Epoch: [122][ 730/ 1236] Overall Loss 0.181215 Objective Loss 0.181215 LR 0.000500 Time 0.022026 +2023-10-02 21:27:53,736 - Epoch: [122][ 740/ 1236] Overall Loss 0.181156 Objective Loss 0.181156 LR 0.000500 Time 0.022005 +2023-10-02 21:27:53,942 - Epoch: [122][ 750/ 1236] Overall Loss 0.181246 Objective Loss 0.181246 LR 0.000500 Time 0.021985 +2023-10-02 21:27:54,147 - Epoch: [122][ 760/ 1236] Overall Loss 0.181050 Objective Loss 0.181050 LR 0.000500 Time 0.021963 +2023-10-02 21:27:54,352 - Epoch: [122][ 770/ 1236] Overall Loss 0.180770 Objective Loss 0.180770 LR 0.000500 Time 0.021945 +2023-10-02 21:27:54,557 - Epoch: [122][ 780/ 1236] Overall Loss 0.180935 Objective Loss 0.180935 LR 0.000500 Time 0.021925 +2023-10-02 21:27:54,763 - Epoch: [122][ 790/ 1236] Overall Loss 0.180958 Objective Loss 0.180958 LR 0.000500 Time 0.021908 +2023-10-02 21:27:54,967 - Epoch: [122][ 800/ 1236] Overall Loss 0.181397 Objective Loss 0.181397 LR 0.000500 Time 0.021889 +2023-10-02 21:27:55,174 - Epoch: [122][ 810/ 1236] Overall Loss 0.181378 Objective Loss 0.181378 LR 0.000500 Time 0.021874 +2023-10-02 21:27:55,380 - Epoch: [122][ 820/ 1236] Overall Loss 0.181581 Objective Loss 0.181581 LR 0.000500 Time 0.021857 +2023-10-02 21:27:55,587 - Epoch: [122][ 830/ 1236] Overall Loss 0.181865 Objective Loss 0.181865 LR 0.000500 Time 0.021842 +2023-10-02 21:27:55,792 - Epoch: [122][ 840/ 1236] Overall Loss 0.182196 Objective Loss 0.182196 LR 0.000500 Time 0.021827 +2023-10-02 21:27:55,999 - Epoch: [122][ 850/ 1236] Overall Loss 0.182582 Objective Loss 0.182582 LR 0.000500 Time 0.021813 +2023-10-02 21:27:56,205 - Epoch: [122][ 860/ 1236] Overall Loss 0.182692 Objective Loss 0.182692 LR 0.000500 Time 0.021799 +2023-10-02 21:27:56,412 - Epoch: [122][ 870/ 1236] Overall Loss 0.182731 Objective Loss 0.182731 LR 0.000500 Time 0.021785 +2023-10-02 21:27:56,618 - Epoch: [122][ 880/ 1236] Overall Loss 0.182681 Objective Loss 0.182681 LR 0.000500 Time 0.021771 +2023-10-02 21:27:56,825 - Epoch: [122][ 890/ 1236] Overall Loss 0.183020 Objective Loss 0.183020 LR 0.000500 Time 0.021758 +2023-10-02 21:27:57,031 - Epoch: [122][ 900/ 1236] Overall Loss 0.183178 Objective Loss 0.183178 LR 0.000500 Time 0.021745 +2023-10-02 21:27:57,238 - Epoch: [122][ 910/ 1236] Overall Loss 0.183315 Objective Loss 0.183315 LR 0.000500 Time 0.021733 +2023-10-02 21:27:57,444 - Epoch: [122][ 920/ 1236] Overall Loss 0.183576 Objective Loss 0.183576 LR 0.000500 Time 0.021720 +2023-10-02 21:27:57,651 - Epoch: [122][ 930/ 1236] Overall Loss 0.183526 Objective Loss 0.183526 LR 0.000500 Time 0.021709 +2023-10-02 21:27:57,856 - Epoch: [122][ 940/ 1236] Overall Loss 0.183817 Objective Loss 0.183817 LR 0.000500 Time 0.021696 +2023-10-02 21:27:58,064 - Epoch: [122][ 950/ 1236] Overall Loss 0.184008 Objective Loss 0.184008 LR 0.000500 Time 0.021686 +2023-10-02 21:27:58,270 - Epoch: [122][ 960/ 1236] Overall Loss 0.183885 Objective Loss 0.183885 LR 0.000500 Time 0.021674 +2023-10-02 21:27:58,477 - Epoch: [122][ 970/ 1236] Overall Loss 0.184122 Objective Loss 0.184122 LR 0.000500 Time 0.021663 +2023-10-02 21:27:58,683 - Epoch: [122][ 980/ 1236] Overall Loss 0.184077 Objective Loss 0.184077 LR 0.000500 Time 0.021652 +2023-10-02 21:27:58,897 - Epoch: [122][ 990/ 1236] Overall Loss 0.184290 Objective Loss 0.184290 LR 0.000500 Time 0.021650 +2023-10-02 21:27:59,110 - Epoch: [122][ 1000/ 1236] Overall Loss 0.184449 Objective Loss 0.184449 LR 0.000500 Time 0.021645 +2023-10-02 21:27:59,317 - Epoch: [122][ 1010/ 1236] Overall Loss 0.184397 Objective Loss 0.184397 LR 0.000500 Time 0.021636 +2023-10-02 21:27:59,523 - Epoch: [122][ 1020/ 1236] Overall Loss 0.184441 Objective Loss 0.184441 LR 0.000500 Time 0.021624 +2023-10-02 21:27:59,730 - Epoch: [122][ 1030/ 1236] Overall Loss 0.184482 Objective Loss 0.184482 LR 0.000500 Time 0.021615 +2023-10-02 21:27:59,936 - Epoch: [122][ 1040/ 1236] Overall Loss 0.184491 Objective Loss 0.184491 LR 0.000500 Time 0.021604 +2023-10-02 21:28:00,143 - Epoch: [122][ 1050/ 1236] Overall Loss 0.184582 Objective Loss 0.184582 LR 0.000500 Time 0.021594 +2023-10-02 21:28:00,349 - Epoch: [122][ 1060/ 1236] Overall Loss 0.184640 Objective Loss 0.184640 LR 0.000500 Time 0.021585 +2023-10-02 21:28:00,556 - Epoch: [122][ 1070/ 1236] Overall Loss 0.184755 Objective Loss 0.184755 LR 0.000500 Time 0.021576 +2023-10-02 21:28:00,762 - Epoch: [122][ 1080/ 1236] Overall Loss 0.184827 Objective Loss 0.184827 LR 0.000500 Time 0.021566 +2023-10-02 21:28:00,969 - Epoch: [122][ 1090/ 1236] Overall Loss 0.184769 Objective Loss 0.184769 LR 0.000500 Time 0.021559 +2023-10-02 21:28:01,176 - Epoch: [122][ 1100/ 1236] Overall Loss 0.185083 Objective Loss 0.185083 LR 0.000500 Time 0.021549 +2023-10-02 21:28:01,383 - Epoch: [122][ 1110/ 1236] Overall Loss 0.184844 Objective Loss 0.184844 LR 0.000500 Time 0.021541 +2023-10-02 21:28:01,589 - Epoch: [122][ 1120/ 1236] Overall Loss 0.184878 Objective Loss 0.184878 LR 0.000500 Time 0.021533 +2023-10-02 21:28:01,796 - Epoch: [122][ 1130/ 1236] Overall Loss 0.184837 Objective Loss 0.184837 LR 0.000500 Time 0.021525 +2023-10-02 21:28:02,003 - Epoch: [122][ 1140/ 1236] Overall Loss 0.184820 Objective Loss 0.184820 LR 0.000500 Time 0.021517 +2023-10-02 21:28:02,210 - Epoch: [122][ 1150/ 1236] Overall Loss 0.185136 Objective Loss 0.185136 LR 0.000500 Time 0.021509 +2023-10-02 21:28:02,416 - Epoch: [122][ 1160/ 1236] Overall Loss 0.185096 Objective Loss 0.185096 LR 0.000500 Time 0.021500 +2023-10-02 21:28:02,623 - Epoch: [122][ 1170/ 1236] Overall Loss 0.185147 Objective Loss 0.185147 LR 0.000500 Time 0.021493 +2023-10-02 21:28:02,829 - Epoch: [122][ 1180/ 1236] Overall Loss 0.185243 Objective Loss 0.185243 LR 0.000500 Time 0.021486 +2023-10-02 21:28:03,036 - Epoch: [122][ 1190/ 1236] Overall Loss 0.185473 Objective Loss 0.185473 LR 0.000500 Time 0.021479 +2023-10-02 21:28:03,243 - Epoch: [122][ 1200/ 1236] Overall Loss 0.185390 Objective Loss 0.185390 LR 0.000500 Time 0.021471 +2023-10-02 21:28:03,450 - Epoch: [122][ 1210/ 1236] Overall Loss 0.185308 Objective Loss 0.185308 LR 0.000500 Time 0.021465 +2023-10-02 21:28:03,656 - Epoch: [122][ 1220/ 1236] Overall Loss 0.185437 Objective Loss 0.185437 LR 0.000500 Time 0.021457 +2023-10-02 21:28:03,913 - Epoch: [122][ 1230/ 1236] Overall Loss 0.185255 Objective Loss 0.185255 LR 0.000500 Time 0.021492 +2023-10-02 21:28:04,035 - Epoch: [122][ 1236/ 1236] Overall Loss 0.185206 Objective Loss 0.185206 Top1 87.780041 Top5 98.778004 LR 0.000500 Time 0.021486 +2023-10-02 21:28:04,168 - --- validate (epoch=122)----------- +2023-10-02 21:28:04,168 - 29943 samples (256 per mini-batch) +2023-10-02 21:28:04,666 - Epoch: [122][ 10/ 117] Loss 0.280436 Top1 86.367188 Top5 98.632812 +2023-10-02 21:28:04,820 - Epoch: [122][ 20/ 117] Loss 0.270874 Top1 86.054688 Top5 98.750000 +2023-10-02 21:28:04,974 - Epoch: [122][ 30/ 117] Loss 0.286548 Top1 85.494792 Top5 98.684896 +2023-10-02 21:28:05,126 - Epoch: [122][ 40/ 117] Loss 0.291611 Top1 85.439453 Top5 98.496094 +2023-10-02 21:28:05,278 - Epoch: [122][ 50/ 117] Loss 0.290919 Top1 85.476562 Top5 98.492188 +2023-10-02 21:28:05,431 - Epoch: [122][ 60/ 117] Loss 0.288714 Top1 85.462240 Top5 98.496094 +2023-10-02 21:28:05,586 - Epoch: [122][ 70/ 117] Loss 0.287647 Top1 85.585938 Top5 98.510045 +2023-10-02 21:28:05,739 - Epoch: [122][ 80/ 117] Loss 0.285820 Top1 85.698242 Top5 98.500977 +2023-10-02 21:28:05,892 - Epoch: [122][ 90/ 117] Loss 0.289719 Top1 85.603299 Top5 98.502604 +2023-10-02 21:28:06,044 - Epoch: [122][ 100/ 117] Loss 0.291976 Top1 85.675781 Top5 98.496094 +2023-10-02 21:28:06,203 - Epoch: [122][ 110/ 117] Loss 0.292851 Top1 85.717330 Top5 98.501420 +2023-10-02 21:28:06,292 - Epoch: [122][ 117/ 117] Loss 0.293465 Top1 85.789667 Top5 98.497145 +2023-10-02 21:28:06,439 - ==> Top1: 85.790 Top5: 98.497 Loss: 0.293 + +2023-10-02 21:28:06,440 - ==> Confusion: +[[ 948 0 4 1 6 2 0 2 4 53 2 1 1 1 5 0 3 1 0 0 16] + [ 0 1048 0 0 4 39 2 18 0 2 1 0 1 0 0 3 1 0 5 2 5] + [ 2 0 982 12 3 1 13 5 0 0 1 4 5 3 1 3 1 1 8 5 6] + [ 1 2 8 990 0 2 2 3 3 1 4 0 7 1 26 2 1 6 8 1 21] + [ 19 5 2 1 968 8 0 0 0 10 1 0 1 3 6 7 13 0 0 0 6] + [ 3 26 0 1 5 1003 3 25 1 4 1 6 3 8 4 0 4 2 2 3 12] + [ 1 2 24 2 0 2 1129 4 0 0 5 0 0 0 0 4 0 0 0 9 9] + [ 0 13 10 1 6 31 6 1075 0 2 1 5 6 5 1 0 2 0 36 8 10] + [ 15 2 1 2 1 5 0 1 975 44 10 2 3 11 11 1 3 0 1 1 0] + [ 95 1 1 0 5 3 0 1 16 947 1 0 0 28 8 4 0 2 1 0 6] + [ 2 2 10 8 2 3 2 3 10 2 967 1 1 18 3 1 0 1 5 2 10] + [ 0 1 1 0 1 12 0 3 0 0 0 963 20 11 0 0 0 17 0 3 3] + [ 0 0 3 3 2 5 1 0 0 0 3 36 982 1 2 4 0 10 2 3 11] + [ 0 0 1 0 2 10 0 0 5 12 4 10 0 1052 4 3 1 0 0 1 14] + [ 10 1 6 29 3 1 0 0 17 4 3 0 2 2 1001 0 2 3 9 0 8] + [ 0 0 1 0 6 0 0 0 0 1 0 6 6 0 1 1077 15 8 3 4 6] + [ 2 15 0 1 4 10 1 0 0 0 0 7 2 1 4 7 1086 0 2 6 13] + [ 0 0 2 3 0 1 2 0 1 0 0 2 29 2 3 7 1 979 1 1 4] + [ 2 8 5 21 0 1 1 25 5 0 3 1 1 0 12 0 0 0 972 0 11] + [ 0 3 1 2 0 5 6 7 0 1 1 17 3 0 0 0 7 1 0 1091 7] + [ 108 153 137 90 84 158 30 82 95 67 155 101 339 266 111 63 95 46 97 175 5453]] + +2023-10-02 21:28:06,441 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:28:06,441 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:28:06,447 - + +2023-10-02 21:28:06,447 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:28:07,459 - Epoch: [123][ 10/ 1236] Overall Loss 0.181339 Objective Loss 0.181339 LR 0.000500 Time 0.101196 +2023-10-02 21:28:07,665 - Epoch: [123][ 20/ 1236] Overall Loss 0.192410 Objective Loss 0.192410 LR 0.000500 Time 0.060862 +2023-10-02 21:28:07,869 - Epoch: [123][ 30/ 1236] Overall Loss 0.187620 Objective Loss 0.187620 LR 0.000500 Time 0.047336 +2023-10-02 21:28:08,076 - Epoch: [123][ 40/ 1236] Overall Loss 0.181320 Objective Loss 0.181320 LR 0.000500 Time 0.040655 +2023-10-02 21:28:08,279 - Epoch: [123][ 50/ 1236] Overall Loss 0.182059 Objective Loss 0.182059 LR 0.000500 Time 0.036577 +2023-10-02 21:28:08,484 - Epoch: [123][ 60/ 1236] Overall Loss 0.182063 Objective Loss 0.182063 LR 0.000500 Time 0.033895 +2023-10-02 21:28:08,688 - Epoch: [123][ 70/ 1236] Overall Loss 0.180441 Objective Loss 0.180441 LR 0.000500 Time 0.031945 +2023-10-02 21:28:08,893 - Epoch: [123][ 80/ 1236] Overall Loss 0.180443 Objective Loss 0.180443 LR 0.000500 Time 0.030512 +2023-10-02 21:28:09,097 - Epoch: [123][ 90/ 1236] Overall Loss 0.180574 Objective Loss 0.180574 LR 0.000500 Time 0.029373 +2023-10-02 21:28:09,302 - Epoch: [123][ 100/ 1236] Overall Loss 0.182616 Objective Loss 0.182616 LR 0.000500 Time 0.028485 +2023-10-02 21:28:09,506 - Epoch: [123][ 110/ 1236] Overall Loss 0.182253 Objective Loss 0.182253 LR 0.000500 Time 0.027736 +2023-10-02 21:28:09,712 - Epoch: [123][ 120/ 1236] Overall Loss 0.181583 Objective Loss 0.181583 LR 0.000500 Time 0.027136 +2023-10-02 21:28:09,916 - Epoch: [123][ 130/ 1236] Overall Loss 0.181697 Objective Loss 0.181697 LR 0.000500 Time 0.026604 +2023-10-02 21:28:10,121 - Epoch: [123][ 140/ 1236] Overall Loss 0.181481 Objective Loss 0.181481 LR 0.000500 Time 0.026170 +2023-10-02 21:28:10,325 - Epoch: [123][ 150/ 1236] Overall Loss 0.180905 Objective Loss 0.180905 LR 0.000500 Time 0.025773 +2023-10-02 21:28:10,532 - Epoch: [123][ 160/ 1236] Overall Loss 0.179961 Objective Loss 0.179961 LR 0.000500 Time 0.025452 +2023-10-02 21:28:10,734 - Epoch: [123][ 170/ 1236] Overall Loss 0.179890 Objective Loss 0.179890 LR 0.000500 Time 0.025145 +2023-10-02 21:28:10,940 - Epoch: [123][ 180/ 1236] Overall Loss 0.180103 Objective Loss 0.180103 LR 0.000500 Time 0.024888 +2023-10-02 21:28:11,143 - Epoch: [123][ 190/ 1236] Overall Loss 0.181379 Objective Loss 0.181379 LR 0.000500 Time 0.024643 +2023-10-02 21:28:11,350 - Epoch: [123][ 200/ 1236] Overall Loss 0.181921 Objective Loss 0.181921 LR 0.000500 Time 0.024443 +2023-10-02 21:28:11,553 - Epoch: [123][ 210/ 1236] Overall Loss 0.182178 Objective Loss 0.182178 LR 0.000500 Time 0.024243 +2023-10-02 21:28:11,760 - Epoch: [123][ 220/ 1236] Overall Loss 0.182867 Objective Loss 0.182867 LR 0.000500 Time 0.024079 +2023-10-02 21:28:11,962 - Epoch: [123][ 230/ 1236] Overall Loss 0.183787 Objective Loss 0.183787 LR 0.000500 Time 0.023912 +2023-10-02 21:28:12,168 - Epoch: [123][ 240/ 1236] Overall Loss 0.183160 Objective Loss 0.183160 LR 0.000500 Time 0.023771 +2023-10-02 21:28:12,372 - Epoch: [123][ 250/ 1236] Overall Loss 0.183161 Objective Loss 0.183161 LR 0.000500 Time 0.023630 +2023-10-02 21:28:12,577 - Epoch: [123][ 260/ 1236] Overall Loss 0.182532 Objective Loss 0.182532 LR 0.000500 Time 0.023510 +2023-10-02 21:28:12,781 - Epoch: [123][ 270/ 1236] Overall Loss 0.182090 Objective Loss 0.182090 LR 0.000500 Time 0.023389 +2023-10-02 21:28:12,988 - Epoch: [123][ 280/ 1236] Overall Loss 0.182431 Objective Loss 0.182431 LR 0.000500 Time 0.023291 +2023-10-02 21:28:13,190 - Epoch: [123][ 290/ 1236] Overall Loss 0.182059 Objective Loss 0.182059 LR 0.000500 Time 0.023185 +2023-10-02 21:28:13,397 - Epoch: [123][ 300/ 1236] Overall Loss 0.181905 Objective Loss 0.181905 LR 0.000500 Time 0.023100 +2023-10-02 21:28:13,599 - Epoch: [123][ 310/ 1236] Overall Loss 0.181396 Objective Loss 0.181396 LR 0.000500 Time 0.023007 +2023-10-02 21:28:13,805 - Epoch: [123][ 320/ 1236] Overall Loss 0.182008 Objective Loss 0.182008 LR 0.000500 Time 0.022930 +2023-10-02 21:28:14,009 - Epoch: [123][ 330/ 1236] Overall Loss 0.181325 Objective Loss 0.181325 LR 0.000500 Time 0.022849 +2023-10-02 21:28:14,216 - Epoch: [123][ 340/ 1236] Overall Loss 0.182168 Objective Loss 0.182168 LR 0.000500 Time 0.022785 +2023-10-02 21:28:14,419 - Epoch: [123][ 350/ 1236] Overall Loss 0.181512 Objective Loss 0.181512 LR 0.000500 Time 0.022712 +2023-10-02 21:28:14,625 - Epoch: [123][ 360/ 1236] Overall Loss 0.181976 Objective Loss 0.181976 LR 0.000500 Time 0.022655 +2023-10-02 21:28:14,828 - Epoch: [123][ 370/ 1236] Overall Loss 0.181879 Objective Loss 0.181879 LR 0.000500 Time 0.022590 +2023-10-02 21:28:15,034 - Epoch: [123][ 380/ 1236] Overall Loss 0.181546 Objective Loss 0.181546 LR 0.000500 Time 0.022537 +2023-10-02 21:28:15,239 - Epoch: [123][ 390/ 1236] Overall Loss 0.181163 Objective Loss 0.181163 LR 0.000500 Time 0.022479 +2023-10-02 21:28:15,444 - Epoch: [123][ 400/ 1236] Overall Loss 0.181574 Objective Loss 0.181574 LR 0.000500 Time 0.022430 +2023-10-02 21:28:15,649 - Epoch: [123][ 410/ 1236] Overall Loss 0.181735 Objective Loss 0.181735 LR 0.000500 Time 0.022379 +2023-10-02 21:28:15,854 - Epoch: [123][ 420/ 1236] Overall Loss 0.181664 Objective Loss 0.181664 LR 0.000500 Time 0.022334 +2023-10-02 21:28:16,058 - Epoch: [123][ 430/ 1236] Overall Loss 0.181308 Objective Loss 0.181308 LR 0.000500 Time 0.022286 +2023-10-02 21:28:16,264 - Epoch: [123][ 440/ 1236] Overall Loss 0.181559 Objective Loss 0.181559 LR 0.000500 Time 0.022246 +2023-10-02 21:28:16,468 - Epoch: [123][ 450/ 1236] Overall Loss 0.181902 Objective Loss 0.181902 LR 0.000500 Time 0.022202 +2023-10-02 21:28:16,675 - Epoch: [123][ 460/ 1236] Overall Loss 0.182287 Objective Loss 0.182287 LR 0.000500 Time 0.022168 +2023-10-02 21:28:16,878 - Epoch: [123][ 470/ 1236] Overall Loss 0.182193 Objective Loss 0.182193 LR 0.000500 Time 0.022129 +2023-10-02 21:28:17,084 - Epoch: [123][ 480/ 1236] Overall Loss 0.182224 Objective Loss 0.182224 LR 0.000500 Time 0.022096 +2023-10-02 21:28:17,290 - Epoch: [123][ 490/ 1236] Overall Loss 0.181975 Objective Loss 0.181975 LR 0.000500 Time 0.022062 +2023-10-02 21:28:17,497 - Epoch: [123][ 500/ 1236] Overall Loss 0.182391 Objective Loss 0.182391 LR 0.000500 Time 0.022035 +2023-10-02 21:28:17,701 - Epoch: [123][ 510/ 1236] Overall Loss 0.182709 Objective Loss 0.182709 LR 0.000500 Time 0.022002 +2023-10-02 21:28:17,909 - Epoch: [123][ 520/ 1236] Overall Loss 0.182785 Objective Loss 0.182785 LR 0.000500 Time 0.021978 +2023-10-02 21:28:18,114 - Epoch: [123][ 530/ 1236] Overall Loss 0.182920 Objective Loss 0.182920 LR 0.000500 Time 0.021949 +2023-10-02 21:28:18,321 - Epoch: [123][ 540/ 1236] Overall Loss 0.183013 Objective Loss 0.183013 LR 0.000500 Time 0.021927 +2023-10-02 21:28:18,526 - Epoch: [123][ 550/ 1236] Overall Loss 0.183080 Objective Loss 0.183080 LR 0.000500 Time 0.021899 +2023-10-02 21:28:18,734 - Epoch: [123][ 560/ 1236] Overall Loss 0.183491 Objective Loss 0.183491 LR 0.000500 Time 0.021878 +2023-10-02 21:28:18,938 - Epoch: [123][ 570/ 1236] Overall Loss 0.183036 Objective Loss 0.183036 LR 0.000500 Time 0.021852 +2023-10-02 21:28:19,146 - Epoch: [123][ 580/ 1236] Overall Loss 0.183260 Objective Loss 0.183260 LR 0.000500 Time 0.021833 +2023-10-02 21:28:19,350 - Epoch: [123][ 590/ 1236] Overall Loss 0.183498 Objective Loss 0.183498 LR 0.000500 Time 0.021809 +2023-10-02 21:28:19,556 - Epoch: [123][ 600/ 1236] Overall Loss 0.183591 Objective Loss 0.183591 LR 0.000500 Time 0.021789 +2023-10-02 21:28:19,762 - Epoch: [123][ 610/ 1236] Overall Loss 0.183670 Objective Loss 0.183670 LR 0.000500 Time 0.021766 +2023-10-02 21:28:19,968 - Epoch: [123][ 620/ 1236] Overall Loss 0.183988 Objective Loss 0.183988 LR 0.000500 Time 0.021748 +2023-10-02 21:28:20,174 - Epoch: [123][ 630/ 1236] Overall Loss 0.183994 Objective Loss 0.183994 LR 0.000500 Time 0.021727 +2023-10-02 21:28:20,382 - Epoch: [123][ 640/ 1236] Overall Loss 0.183970 Objective Loss 0.183970 LR 0.000500 Time 0.021712 +2023-10-02 21:28:20,586 - Epoch: [123][ 650/ 1236] Overall Loss 0.184036 Objective Loss 0.184036 LR 0.000500 Time 0.021691 +2023-10-02 21:28:20,794 - Epoch: [123][ 660/ 1236] Overall Loss 0.184049 Objective Loss 0.184049 LR 0.000500 Time 0.021677 +2023-10-02 21:28:20,998 - Epoch: [123][ 670/ 1236] Overall Loss 0.183969 Objective Loss 0.183969 LR 0.000500 Time 0.021658 +2023-10-02 21:28:21,206 - Epoch: [123][ 680/ 1236] Overall Loss 0.183724 Objective Loss 0.183724 LR 0.000500 Time 0.021645 +2023-10-02 21:28:21,411 - Epoch: [123][ 690/ 1236] Overall Loss 0.183867 Objective Loss 0.183867 LR 0.000500 Time 0.021627 +2023-10-02 21:28:21,617 - Epoch: [123][ 700/ 1236] Overall Loss 0.183776 Objective Loss 0.183776 LR 0.000500 Time 0.021613 +2023-10-02 21:28:21,823 - Epoch: [123][ 710/ 1236] Overall Loss 0.183823 Objective Loss 0.183823 LR 0.000500 Time 0.021596 +2023-10-02 21:28:22,029 - Epoch: [123][ 720/ 1236] Overall Loss 0.183720 Objective Loss 0.183720 LR 0.000500 Time 0.021582 +2023-10-02 21:28:22,235 - Epoch: [123][ 730/ 1236] Overall Loss 0.183841 Objective Loss 0.183841 LR 0.000500 Time 0.021566 +2023-10-02 21:28:22,443 - Epoch: [123][ 740/ 1236] Overall Loss 0.184096 Objective Loss 0.184096 LR 0.000500 Time 0.021555 +2023-10-02 21:28:22,647 - Epoch: [123][ 750/ 1236] Overall Loss 0.184446 Objective Loss 0.184446 LR 0.000500 Time 0.021540 +2023-10-02 21:28:22,853 - Epoch: [123][ 760/ 1236] Overall Loss 0.184718 Objective Loss 0.184718 LR 0.000500 Time 0.021528 +2023-10-02 21:28:23,060 - Epoch: [123][ 770/ 1236] Overall Loss 0.184668 Objective Loss 0.184668 LR 0.000500 Time 0.021514 +2023-10-02 21:28:23,268 - Epoch: [123][ 780/ 1236] Overall Loss 0.184875 Objective Loss 0.184875 LR 0.000500 Time 0.021505 +2023-10-02 21:28:23,473 - Epoch: [123][ 790/ 1236] Overall Loss 0.185067 Objective Loss 0.185067 LR 0.000500 Time 0.021492 +2023-10-02 21:28:23,682 - Epoch: [123][ 800/ 1236] Overall Loss 0.185321 Objective Loss 0.185321 LR 0.000500 Time 0.021484 +2023-10-02 21:28:23,887 - Epoch: [123][ 810/ 1236] Overall Loss 0.185090 Objective Loss 0.185090 LR 0.000500 Time 0.021471 +2023-10-02 21:28:24,095 - Epoch: [123][ 820/ 1236] Overall Loss 0.185046 Objective Loss 0.185046 LR 0.000500 Time 0.021463 +2023-10-02 21:28:24,300 - Epoch: [123][ 830/ 1236] Overall Loss 0.185082 Objective Loss 0.185082 LR 0.000500 Time 0.021451 +2023-10-02 21:28:24,509 - Epoch: [123][ 840/ 1236] Overall Loss 0.185229 Objective Loss 0.185229 LR 0.000500 Time 0.021444 +2023-10-02 21:28:24,714 - Epoch: [123][ 850/ 1236] Overall Loss 0.184989 Objective Loss 0.184989 LR 0.000500 Time 0.021432 +2023-10-02 21:28:24,922 - Epoch: [123][ 860/ 1236] Overall Loss 0.185458 Objective Loss 0.185458 LR 0.000500 Time 0.021425 +2023-10-02 21:28:25,127 - Epoch: [123][ 870/ 1236] Overall Loss 0.185898 Objective Loss 0.185898 LR 0.000500 Time 0.021414 +2023-10-02 21:28:25,336 - Epoch: [123][ 880/ 1236] Overall Loss 0.185838 Objective Loss 0.185838 LR 0.000500 Time 0.021407 +2023-10-02 21:28:25,541 - Epoch: [123][ 890/ 1236] Overall Loss 0.185622 Objective Loss 0.185622 LR 0.000500 Time 0.021397 +2023-10-02 21:28:25,749 - Epoch: [123][ 900/ 1236] Overall Loss 0.185668 Objective Loss 0.185668 LR 0.000500 Time 0.021391 +2023-10-02 21:28:25,954 - Epoch: [123][ 910/ 1236] Overall Loss 0.185709 Objective Loss 0.185709 LR 0.000500 Time 0.021380 +2023-10-02 21:28:26,163 - Epoch: [123][ 920/ 1236] Overall Loss 0.186005 Objective Loss 0.186005 LR 0.000500 Time 0.021374 +2023-10-02 21:28:26,368 - Epoch: [123][ 930/ 1236] Overall Loss 0.185925 Objective Loss 0.185925 LR 0.000500 Time 0.021364 +2023-10-02 21:28:26,576 - Epoch: [123][ 940/ 1236] Overall Loss 0.186018 Objective Loss 0.186018 LR 0.000500 Time 0.021359 +2023-10-02 21:28:26,781 - Epoch: [123][ 950/ 1236] Overall Loss 0.186210 Objective Loss 0.186210 LR 0.000500 Time 0.021349 +2023-10-02 21:28:26,990 - Epoch: [123][ 960/ 1236] Overall Loss 0.186139 Objective Loss 0.186139 LR 0.000500 Time 0.021344 +2023-10-02 21:28:27,196 - Epoch: [123][ 970/ 1236] Overall Loss 0.185864 Objective Loss 0.185864 LR 0.000500 Time 0.021336 +2023-10-02 21:28:27,405 - Epoch: [123][ 980/ 1236] Overall Loss 0.185882 Objective Loss 0.185882 LR 0.000500 Time 0.021331 +2023-10-02 21:28:27,609 - Epoch: [123][ 990/ 1236] Overall Loss 0.185987 Objective Loss 0.185987 LR 0.000500 Time 0.021322 +2023-10-02 21:28:27,818 - Epoch: [123][ 1000/ 1236] Overall Loss 0.185915 Objective Loss 0.185915 LR 0.000500 Time 0.021317 +2023-10-02 21:28:28,023 - Epoch: [123][ 1010/ 1236] Overall Loss 0.185864 Objective Loss 0.185864 LR 0.000500 Time 0.021309 +2023-10-02 21:28:28,232 - Epoch: [123][ 1020/ 1236] Overall Loss 0.185890 Objective Loss 0.185890 LR 0.000500 Time 0.021304 +2023-10-02 21:28:28,437 - Epoch: [123][ 1030/ 1236] Overall Loss 0.185780 Objective Loss 0.185780 LR 0.000500 Time 0.021296 +2023-10-02 21:28:28,645 - Epoch: [123][ 1040/ 1236] Overall Loss 0.185748 Objective Loss 0.185748 LR 0.000500 Time 0.021291 +2023-10-02 21:28:28,850 - Epoch: [123][ 1050/ 1236] Overall Loss 0.185782 Objective Loss 0.185782 LR 0.000500 Time 0.021284 +2023-10-02 21:28:29,059 - Epoch: [123][ 1060/ 1236] Overall Loss 0.185553 Objective Loss 0.185553 LR 0.000500 Time 0.021279 +2023-10-02 21:28:29,264 - Epoch: [123][ 1070/ 1236] Overall Loss 0.185459 Objective Loss 0.185459 LR 0.000500 Time 0.021272 +2023-10-02 21:28:29,473 - Epoch: [123][ 1080/ 1236] Overall Loss 0.185675 Objective Loss 0.185675 LR 0.000500 Time 0.021268 +2023-10-02 21:28:29,678 - Epoch: [123][ 1090/ 1236] Overall Loss 0.185729 Objective Loss 0.185729 LR 0.000500 Time 0.021261 +2023-10-02 21:28:29,886 - Epoch: [123][ 1100/ 1236] Overall Loss 0.185802 Objective Loss 0.185802 LR 0.000500 Time 0.021257 +2023-10-02 21:28:30,092 - Epoch: [123][ 1110/ 1236] Overall Loss 0.185705 Objective Loss 0.185705 LR 0.000500 Time 0.021250 +2023-10-02 21:28:30,300 - Epoch: [123][ 1120/ 1236] Overall Loss 0.185986 Objective Loss 0.185986 LR 0.000500 Time 0.021246 +2023-10-02 21:28:30,505 - Epoch: [123][ 1130/ 1236] Overall Loss 0.186120 Objective Loss 0.186120 LR 0.000500 Time 0.021239 +2023-10-02 21:28:30,713 - Epoch: [123][ 1140/ 1236] Overall Loss 0.185987 Objective Loss 0.185987 LR 0.000500 Time 0.021234 +2023-10-02 21:28:30,919 - Epoch: [123][ 1150/ 1236] Overall Loss 0.185908 Objective Loss 0.185908 LR 0.000500 Time 0.021228 +2023-10-02 21:28:31,128 - Epoch: [123][ 1160/ 1236] Overall Loss 0.185894 Objective Loss 0.185894 LR 0.000500 Time 0.021225 +2023-10-02 21:28:31,333 - Epoch: [123][ 1170/ 1236] Overall Loss 0.185976 Objective Loss 0.185976 LR 0.000500 Time 0.021218 +2023-10-02 21:28:31,541 - Epoch: [123][ 1180/ 1236] Overall Loss 0.186038 Objective Loss 0.186038 LR 0.000500 Time 0.021215 +2023-10-02 21:28:31,746 - Epoch: [123][ 1190/ 1236] Overall Loss 0.186202 Objective Loss 0.186202 LR 0.000500 Time 0.021208 +2023-10-02 21:28:31,955 - Epoch: [123][ 1200/ 1236] Overall Loss 0.186012 Objective Loss 0.186012 LR 0.000500 Time 0.021205 +2023-10-02 21:28:32,160 - Epoch: [123][ 1210/ 1236] Overall Loss 0.185891 Objective Loss 0.185891 LR 0.000500 Time 0.021200 +2023-10-02 21:28:32,369 - Epoch: [123][ 1220/ 1236] Overall Loss 0.186101 Objective Loss 0.186101 LR 0.000500 Time 0.021196 +2023-10-02 21:28:32,628 - Epoch: [123][ 1230/ 1236] Overall Loss 0.185858 Objective Loss 0.185858 LR 0.000500 Time 0.021234 +2023-10-02 21:28:32,749 - Epoch: [123][ 1236/ 1236] Overall Loss 0.185920 Objective Loss 0.185920 Top1 88.798371 Top5 98.981670 LR 0.000500 Time 0.021229 +2023-10-02 21:28:32,890 - --- validate (epoch=123)----------- +2023-10-02 21:28:32,891 - 29943 samples (256 per mini-batch) +2023-10-02 21:28:33,381 - Epoch: [123][ 10/ 117] Loss 0.271757 Top1 85.468750 Top5 98.632812 +2023-10-02 21:28:33,540 - Epoch: [123][ 20/ 117] Loss 0.265345 Top1 85.820312 Top5 98.750000 +2023-10-02 21:28:33,693 - Epoch: [123][ 30/ 117] Loss 0.276878 Top1 85.820312 Top5 98.658854 +2023-10-02 21:28:33,851 - Epoch: [123][ 40/ 117] Loss 0.282919 Top1 85.761719 Top5 98.593750 +2023-10-02 21:28:34,001 - Epoch: [123][ 50/ 117] Loss 0.299250 Top1 85.640625 Top5 98.414062 +2023-10-02 21:28:34,149 - Epoch: [123][ 60/ 117] Loss 0.296581 Top1 85.611979 Top5 98.378906 +2023-10-02 21:28:34,298 - Epoch: [123][ 70/ 117] Loss 0.296452 Top1 85.742188 Top5 98.387277 +2023-10-02 21:28:34,450 - Epoch: [123][ 80/ 117] Loss 0.295911 Top1 85.673828 Top5 98.422852 +2023-10-02 21:28:34,602 - Epoch: [123][ 90/ 117] Loss 0.297315 Top1 85.651042 Top5 98.407118 +2023-10-02 21:28:34,755 - Epoch: [123][ 100/ 117] Loss 0.294378 Top1 85.617188 Top5 98.445312 +2023-10-02 21:28:34,913 - Epoch: [123][ 110/ 117] Loss 0.292205 Top1 85.628551 Top5 98.473011 +2023-10-02 21:28:35,001 - Epoch: [123][ 117/ 117] Loss 0.292550 Top1 85.605985 Top5 98.493805 +2023-10-02 21:28:35,143 - ==> Top1: 85.606 Top5: 98.494 Loss: 0.293 + +2023-10-02 21:28:35,144 - ==> Confusion: +[[ 949 0 6 0 7 2 0 1 5 52 1 0 1 3 2 1 2 0 1 0 17] + [ 0 1045 1 0 2 23 3 33 0 1 2 2 0 0 0 4 0 0 10 2 3] + [ 3 1 985 8 1 0 17 7 0 1 2 0 8 2 0 3 0 1 8 1 8] + [ 0 2 13 977 0 5 1 6 1 1 2 0 8 2 21 2 0 5 24 1 18] + [ 26 5 1 0 969 6 0 0 1 8 0 0 3 1 8 4 7 0 1 3 7] + [ 3 30 0 5 2 1006 1 31 0 3 2 3 1 7 3 0 1 0 5 3 10] + [ 1 0 28 1 0 1 1133 5 0 0 3 0 1 0 0 3 0 0 1 10 4] + [ 4 12 8 1 3 22 5 1085 1 2 4 3 3 6 1 0 0 1 36 12 9] + [ 16 1 0 0 1 4 0 2 973 49 13 1 3 7 10 0 1 0 2 3 3] + [ 95 0 1 3 6 3 0 0 21 965 0 0 0 12 5 1 0 0 0 0 7] + [ 2 2 7 7 2 3 2 5 14 2 961 2 0 15 5 0 1 3 8 0 12] + [ 0 1 2 0 1 11 0 2 0 1 0 957 28 9 0 0 0 16 0 5 2] + [ 1 0 0 3 0 1 3 3 1 0 2 34 983 2 0 4 1 10 3 4 13] + [ 1 0 2 0 4 11 0 0 14 17 2 10 0 1041 5 0 0 1 0 0 11] + [ 11 0 7 19 2 1 0 0 22 4 1 0 3 2 996 0 0 5 16 0 12] + [ 0 0 1 1 5 1 0 0 0 1 1 5 7 0 0 1066 16 10 3 9 8] + [ 1 11 0 0 5 9 2 1 2 0 0 4 2 2 3 4 1095 1 2 5 12] + [ 0 0 0 1 0 0 3 0 0 0 0 2 22 1 2 5 0 997 1 2 2] + [ 1 5 3 14 0 0 0 20 6 0 2 1 0 0 7 0 1 0 999 1 8] + [ 0 1 2 3 0 3 6 6 0 0 1 13 1 1 0 0 7 1 0 1102 5] + [ 148 134 138 104 76 153 34 107 99 94 132 85 339 249 113 41 78 74 168 190 5349]] + +2023-10-02 21:28:35,145 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:28:35,145 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:28:35,152 - + +2023-10-02 21:28:35,152 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:28:36,138 - Epoch: [124][ 10/ 1236] Overall Loss 0.156656 Objective Loss 0.156656 LR 0.000500 Time 0.098615 +2023-10-02 21:28:36,344 - Epoch: [124][ 20/ 1236] Overall Loss 0.173510 Objective Loss 0.173510 LR 0.000500 Time 0.059567 +2023-10-02 21:28:36,548 - Epoch: [124][ 30/ 1236] Overall Loss 0.174698 Objective Loss 0.174698 LR 0.000500 Time 0.046518 +2023-10-02 21:28:36,754 - Epoch: [124][ 40/ 1236] Overall Loss 0.172876 Objective Loss 0.172876 LR 0.000500 Time 0.040010 +2023-10-02 21:28:36,959 - Epoch: [124][ 50/ 1236] Overall Loss 0.176873 Objective Loss 0.176873 LR 0.000500 Time 0.036087 +2023-10-02 21:28:37,167 - Epoch: [124][ 60/ 1236] Overall Loss 0.174161 Objective Loss 0.174161 LR 0.000500 Time 0.033529 +2023-10-02 21:28:37,373 - Epoch: [124][ 70/ 1236] Overall Loss 0.175959 Objective Loss 0.175959 LR 0.000500 Time 0.031680 +2023-10-02 21:28:37,581 - Epoch: [124][ 80/ 1236] Overall Loss 0.175722 Objective Loss 0.175722 LR 0.000500 Time 0.030314 +2023-10-02 21:28:37,787 - Epoch: [124][ 90/ 1236] Overall Loss 0.177463 Objective Loss 0.177463 LR 0.000500 Time 0.029232 +2023-10-02 21:28:37,994 - Epoch: [124][ 100/ 1236] Overall Loss 0.176355 Objective Loss 0.176355 LR 0.000500 Time 0.028381 +2023-10-02 21:28:38,199 - Epoch: [124][ 110/ 1236] Overall Loss 0.175987 Objective Loss 0.175987 LR 0.000500 Time 0.027660 +2023-10-02 21:28:38,408 - Epoch: [124][ 120/ 1236] Overall Loss 0.175786 Objective Loss 0.175786 LR 0.000500 Time 0.027091 +2023-10-02 21:28:38,612 - Epoch: [124][ 130/ 1236] Overall Loss 0.176417 Objective Loss 0.176417 LR 0.000500 Time 0.026579 +2023-10-02 21:28:38,820 - Epoch: [124][ 140/ 1236] Overall Loss 0.176306 Objective Loss 0.176306 LR 0.000500 Time 0.026160 +2023-10-02 21:28:39,026 - Epoch: [124][ 150/ 1236] Overall Loss 0.175357 Objective Loss 0.175357 LR 0.000500 Time 0.025778 +2023-10-02 21:28:39,234 - Epoch: [124][ 160/ 1236] Overall Loss 0.174598 Objective Loss 0.174598 LR 0.000500 Time 0.025466 +2023-10-02 21:28:39,440 - Epoch: [124][ 170/ 1236] Overall Loss 0.172825 Objective Loss 0.172825 LR 0.000500 Time 0.025172 +2023-10-02 21:28:39,649 - Epoch: [124][ 180/ 1236] Overall Loss 0.173667 Objective Loss 0.173667 LR 0.000500 Time 0.024934 +2023-10-02 21:28:39,854 - Epoch: [124][ 190/ 1236] Overall Loss 0.174068 Objective Loss 0.174068 LR 0.000500 Time 0.024698 +2023-10-02 21:28:40,063 - Epoch: [124][ 200/ 1236] Overall Loss 0.173998 Objective Loss 0.173998 LR 0.000500 Time 0.024507 +2023-10-02 21:28:40,268 - Epoch: [124][ 210/ 1236] Overall Loss 0.174986 Objective Loss 0.174986 LR 0.000500 Time 0.024314 +2023-10-02 21:28:40,476 - Epoch: [124][ 220/ 1236] Overall Loss 0.174777 Objective Loss 0.174777 LR 0.000500 Time 0.024152 +2023-10-02 21:28:40,682 - Epoch: [124][ 230/ 1236] Overall Loss 0.174343 Objective Loss 0.174343 LR 0.000500 Time 0.023991 +2023-10-02 21:28:40,890 - Epoch: [124][ 240/ 1236] Overall Loss 0.174845 Objective Loss 0.174845 LR 0.000500 Time 0.023857 +2023-10-02 21:28:41,096 - Epoch: [124][ 250/ 1236] Overall Loss 0.175403 Objective Loss 0.175403 LR 0.000500 Time 0.023722 +2023-10-02 21:28:41,305 - Epoch: [124][ 260/ 1236] Overall Loss 0.174766 Objective Loss 0.174766 LR 0.000500 Time 0.023613 +2023-10-02 21:28:41,510 - Epoch: [124][ 270/ 1236] Overall Loss 0.174849 Objective Loss 0.174849 LR 0.000500 Time 0.023496 +2023-10-02 21:28:41,719 - Epoch: [124][ 280/ 1236] Overall Loss 0.175465 Objective Loss 0.175465 LR 0.000500 Time 0.023403 +2023-10-02 21:28:41,924 - Epoch: [124][ 290/ 1236] Overall Loss 0.175273 Objective Loss 0.175273 LR 0.000500 Time 0.023302 +2023-10-02 21:28:42,132 - Epoch: [124][ 300/ 1236] Overall Loss 0.175667 Objective Loss 0.175667 LR 0.000500 Time 0.023217 +2023-10-02 21:28:42,338 - Epoch: [124][ 310/ 1236] Overall Loss 0.174755 Objective Loss 0.174755 LR 0.000500 Time 0.023128 +2023-10-02 21:28:42,546 - Epoch: [124][ 320/ 1236] Overall Loss 0.174938 Objective Loss 0.174938 LR 0.000500 Time 0.023054 +2023-10-02 21:28:42,752 - Epoch: [124][ 330/ 1236] Overall Loss 0.174870 Objective Loss 0.174870 LR 0.000500 Time 0.022980 +2023-10-02 21:28:42,960 - Epoch: [124][ 340/ 1236] Overall Loss 0.175473 Objective Loss 0.175473 LR 0.000500 Time 0.022914 +2023-10-02 21:28:43,166 - Epoch: [124][ 350/ 1236] Overall Loss 0.175776 Objective Loss 0.175776 LR 0.000500 Time 0.022848 +2023-10-02 21:28:43,376 - Epoch: [124][ 360/ 1236] Overall Loss 0.175793 Objective Loss 0.175793 LR 0.000500 Time 0.022794 +2023-10-02 21:28:43,581 - Epoch: [124][ 370/ 1236] Overall Loss 0.176466 Objective Loss 0.176466 LR 0.000500 Time 0.022731 +2023-10-02 21:28:43,790 - Epoch: [124][ 380/ 1236] Overall Loss 0.177140 Objective Loss 0.177140 LR 0.000500 Time 0.022682 +2023-10-02 21:28:43,995 - Epoch: [124][ 390/ 1236] Overall Loss 0.176965 Objective Loss 0.176965 LR 0.000500 Time 0.022625 +2023-10-02 21:28:44,204 - Epoch: [124][ 400/ 1236] Overall Loss 0.177326 Objective Loss 0.177326 LR 0.000500 Time 0.022581 +2023-10-02 21:28:44,409 - Epoch: [124][ 410/ 1236] Overall Loss 0.177559 Objective Loss 0.177559 LR 0.000500 Time 0.022530 +2023-10-02 21:28:44,618 - Epoch: [124][ 420/ 1236] Overall Loss 0.177570 Objective Loss 0.177570 LR 0.000500 Time 0.022491 +2023-10-02 21:28:44,823 - Epoch: [124][ 430/ 1236] Overall Loss 0.177424 Objective Loss 0.177424 LR 0.000500 Time 0.022444 +2023-10-02 21:28:45,032 - Epoch: [124][ 440/ 1236] Overall Loss 0.177637 Objective Loss 0.177637 LR 0.000500 Time 0.022408 +2023-10-02 21:28:45,237 - Epoch: [124][ 450/ 1236] Overall Loss 0.177638 Objective Loss 0.177638 LR 0.000500 Time 0.022366 +2023-10-02 21:28:45,445 - Epoch: [124][ 460/ 1236] Overall Loss 0.177915 Objective Loss 0.177915 LR 0.000500 Time 0.022331 +2023-10-02 21:28:45,651 - Epoch: [124][ 470/ 1236] Overall Loss 0.177948 Objective Loss 0.177948 LR 0.000500 Time 0.022294 +2023-10-02 21:28:45,860 - Epoch: [124][ 480/ 1236] Overall Loss 0.177884 Objective Loss 0.177884 LR 0.000500 Time 0.022264 +2023-10-02 21:28:46,065 - Epoch: [124][ 490/ 1236] Overall Loss 0.178374 Objective Loss 0.178374 LR 0.000500 Time 0.022228 +2023-10-02 21:28:46,273 - Epoch: [124][ 500/ 1236] Overall Loss 0.178593 Objective Loss 0.178593 LR 0.000500 Time 0.022199 +2023-10-02 21:28:46,480 - Epoch: [124][ 510/ 1236] Overall Loss 0.179319 Objective Loss 0.179319 LR 0.000500 Time 0.022167 +2023-10-02 21:28:46,689 - Epoch: [124][ 520/ 1236] Overall Loss 0.179165 Objective Loss 0.179165 LR 0.000500 Time 0.022142 +2023-10-02 21:28:46,894 - Epoch: [124][ 530/ 1236] Overall Loss 0.179081 Objective Loss 0.179081 LR 0.000500 Time 0.022111 +2023-10-02 21:28:47,103 - Epoch: [124][ 540/ 1236] Overall Loss 0.179033 Objective Loss 0.179033 LR 0.000500 Time 0.022088 +2023-10-02 21:28:47,308 - Epoch: [124][ 550/ 1236] Overall Loss 0.178937 Objective Loss 0.178937 LR 0.000500 Time 0.022059 +2023-10-02 21:28:47,517 - Epoch: [124][ 560/ 1236] Overall Loss 0.178992 Objective Loss 0.178992 LR 0.000500 Time 0.022038 +2023-10-02 21:28:47,722 - Epoch: [124][ 570/ 1236] Overall Loss 0.178896 Objective Loss 0.178896 LR 0.000500 Time 0.022011 +2023-10-02 21:28:47,930 - Epoch: [124][ 580/ 1236] Overall Loss 0.178824 Objective Loss 0.178824 LR 0.000500 Time 0.021989 +2023-10-02 21:28:48,136 - Epoch: [124][ 590/ 1236] Overall Loss 0.178718 Objective Loss 0.178718 LR 0.000500 Time 0.021966 +2023-10-02 21:28:48,344 - Epoch: [124][ 600/ 1236] Overall Loss 0.179029 Objective Loss 0.179029 LR 0.000500 Time 0.021946 +2023-10-02 21:28:48,550 - Epoch: [124][ 610/ 1236] Overall Loss 0.179156 Objective Loss 0.179156 LR 0.000500 Time 0.021923 +2023-10-02 21:28:48,758 - Epoch: [124][ 620/ 1236] Overall Loss 0.179289 Objective Loss 0.179289 LR 0.000500 Time 0.021904 +2023-10-02 21:28:48,965 - Epoch: [124][ 630/ 1236] Overall Loss 0.179225 Objective Loss 0.179225 LR 0.000500 Time 0.021884 +2023-10-02 21:28:49,174 - Epoch: [124][ 640/ 1236] Overall Loss 0.179332 Objective Loss 0.179332 LR 0.000500 Time 0.021869 +2023-10-02 21:28:49,379 - Epoch: [124][ 650/ 1236] Overall Loss 0.179584 Objective Loss 0.179584 LR 0.000500 Time 0.021848 +2023-10-02 21:28:49,588 - Epoch: [124][ 660/ 1236] Overall Loss 0.179896 Objective Loss 0.179896 LR 0.000500 Time 0.021833 +2023-10-02 21:28:49,793 - Epoch: [124][ 670/ 1236] Overall Loss 0.179274 Objective Loss 0.179274 LR 0.000500 Time 0.021812 +2023-10-02 21:28:50,002 - Epoch: [124][ 680/ 1236] Overall Loss 0.179220 Objective Loss 0.179220 LR 0.000500 Time 0.021798 +2023-10-02 21:28:50,207 - Epoch: [124][ 690/ 1236] Overall Loss 0.179005 Objective Loss 0.179005 LR 0.000500 Time 0.021779 +2023-10-02 21:28:50,415 - Epoch: [124][ 700/ 1236] Overall Loss 0.179211 Objective Loss 0.179211 LR 0.000500 Time 0.021765 +2023-10-02 21:28:50,621 - Epoch: [124][ 710/ 1236] Overall Loss 0.179241 Objective Loss 0.179241 LR 0.000500 Time 0.021749 +2023-10-02 21:28:50,829 - Epoch: [124][ 720/ 1236] Overall Loss 0.179142 Objective Loss 0.179142 LR 0.000500 Time 0.021735 +2023-10-02 21:28:51,036 - Epoch: [124][ 730/ 1236] Overall Loss 0.179202 Objective Loss 0.179202 LR 0.000500 Time 0.021719 +2023-10-02 21:28:51,245 - Epoch: [124][ 740/ 1236] Overall Loss 0.179168 Objective Loss 0.179168 LR 0.000500 Time 0.021708 +2023-10-02 21:28:51,450 - Epoch: [124][ 750/ 1236] Overall Loss 0.179306 Objective Loss 0.179306 LR 0.000500 Time 0.021692 +2023-10-02 21:28:51,658 - Epoch: [124][ 760/ 1236] Overall Loss 0.179595 Objective Loss 0.179595 LR 0.000500 Time 0.021680 +2023-10-02 21:28:51,864 - Epoch: [124][ 770/ 1236] Overall Loss 0.179876 Objective Loss 0.179876 LR 0.000500 Time 0.021666 +2023-10-02 21:28:52,072 - Epoch: [124][ 780/ 1236] Overall Loss 0.179824 Objective Loss 0.179824 LR 0.000500 Time 0.021654 +2023-10-02 21:28:52,279 - Epoch: [124][ 790/ 1236] Overall Loss 0.179752 Objective Loss 0.179752 LR 0.000500 Time 0.021641 +2023-10-02 21:28:52,488 - Epoch: [124][ 800/ 1236] Overall Loss 0.179702 Objective Loss 0.179702 LR 0.000500 Time 0.021632 +2023-10-02 21:28:52,693 - Epoch: [124][ 810/ 1236] Overall Loss 0.180099 Objective Loss 0.180099 LR 0.000500 Time 0.021618 +2023-10-02 21:28:52,901 - Epoch: [124][ 820/ 1236] Overall Loss 0.180066 Objective Loss 0.180066 LR 0.000500 Time 0.021607 +2023-10-02 21:28:53,108 - Epoch: [124][ 830/ 1236] Overall Loss 0.180218 Objective Loss 0.180218 LR 0.000500 Time 0.021596 +2023-10-02 21:28:53,316 - Epoch: [124][ 840/ 1236] Overall Loss 0.180429 Objective Loss 0.180429 LR 0.000500 Time 0.021586 +2023-10-02 21:28:53,522 - Epoch: [124][ 850/ 1236] Overall Loss 0.180488 Objective Loss 0.180488 LR 0.000500 Time 0.021573 +2023-10-02 21:28:53,730 - Epoch: [124][ 860/ 1236] Overall Loss 0.180516 Objective Loss 0.180516 LR 0.000500 Time 0.021563 +2023-10-02 21:28:53,936 - Epoch: [124][ 870/ 1236] Overall Loss 0.180404 Objective Loss 0.180404 LR 0.000500 Time 0.021552 +2023-10-02 21:28:54,145 - Epoch: [124][ 880/ 1236] Overall Loss 0.180375 Objective Loss 0.180375 LR 0.000500 Time 0.021545 +2023-10-02 21:28:54,351 - Epoch: [124][ 890/ 1236] Overall Loss 0.180523 Objective Loss 0.180523 LR 0.000500 Time 0.021533 +2023-10-02 21:28:54,560 - Epoch: [124][ 900/ 1236] Overall Loss 0.180598 Objective Loss 0.180598 LR 0.000500 Time 0.021526 +2023-10-02 21:28:54,765 - Epoch: [124][ 910/ 1236] Overall Loss 0.180895 Objective Loss 0.180895 LR 0.000500 Time 0.021514 +2023-10-02 21:28:54,974 - Epoch: [124][ 920/ 1236] Overall Loss 0.181032 Objective Loss 0.181032 LR 0.000500 Time 0.021507 +2023-10-02 21:28:55,179 - Epoch: [124][ 930/ 1236] Overall Loss 0.180908 Objective Loss 0.180908 LR 0.000500 Time 0.021496 +2023-10-02 21:28:55,387 - Epoch: [124][ 940/ 1236] Overall Loss 0.181098 Objective Loss 0.181098 LR 0.000500 Time 0.021488 +2023-10-02 21:28:55,593 - Epoch: [124][ 950/ 1236] Overall Loss 0.180954 Objective Loss 0.180954 LR 0.000500 Time 0.021479 +2023-10-02 21:28:55,801 - Epoch: [124][ 960/ 1236] Overall Loss 0.181294 Objective Loss 0.181294 LR 0.000500 Time 0.021471 +2023-10-02 21:28:56,008 - Epoch: [124][ 970/ 1236] Overall Loss 0.181409 Objective Loss 0.181409 LR 0.000500 Time 0.021463 +2023-10-02 21:28:56,217 - Epoch: [124][ 980/ 1236] Overall Loss 0.181253 Objective Loss 0.181253 LR 0.000500 Time 0.021457 +2023-10-02 21:28:56,422 - Epoch: [124][ 990/ 1236] Overall Loss 0.181273 Objective Loss 0.181273 LR 0.000500 Time 0.021447 +2023-10-02 21:28:56,630 - Epoch: [124][ 1000/ 1236] Overall Loss 0.181103 Objective Loss 0.181103 LR 0.000500 Time 0.021440 +2023-10-02 21:28:56,836 - Epoch: [124][ 1010/ 1236] Overall Loss 0.181113 Objective Loss 0.181113 LR 0.000500 Time 0.021432 +2023-10-02 21:28:57,046 - Epoch: [124][ 1020/ 1236] Overall Loss 0.181097 Objective Loss 0.181097 LR 0.000500 Time 0.021427 +2023-10-02 21:28:57,254 - Epoch: [124][ 1030/ 1236] Overall Loss 0.180900 Objective Loss 0.180900 LR 0.000500 Time 0.021421 +2023-10-02 21:28:57,464 - Epoch: [124][ 1040/ 1236] Overall Loss 0.180801 Objective Loss 0.180801 LR 0.000500 Time 0.021416 +2023-10-02 21:28:57,673 - Epoch: [124][ 1050/ 1236] Overall Loss 0.180897 Objective Loss 0.180897 LR 0.000500 Time 0.021411 +2023-10-02 21:28:57,883 - Epoch: [124][ 1060/ 1236] Overall Loss 0.180902 Objective Loss 0.180902 LR 0.000500 Time 0.021407 +2023-10-02 21:28:58,093 - Epoch: [124][ 1070/ 1236] Overall Loss 0.181209 Objective Loss 0.181209 LR 0.000500 Time 0.021401 +2023-10-02 21:28:58,303 - Epoch: [124][ 1080/ 1236] Overall Loss 0.181127 Objective Loss 0.181127 LR 0.000500 Time 0.021398 +2023-10-02 21:28:58,513 - Epoch: [124][ 1090/ 1236] Overall Loss 0.181285 Objective Loss 0.181285 LR 0.000500 Time 0.021394 +2023-10-02 21:28:58,723 - Epoch: [124][ 1100/ 1236] Overall Loss 0.181527 Objective Loss 0.181527 LR 0.000500 Time 0.021390 +2023-10-02 21:28:58,933 - Epoch: [124][ 1110/ 1236] Overall Loss 0.181673 Objective Loss 0.181673 LR 0.000500 Time 0.021386 +2023-10-02 21:28:59,143 - Epoch: [124][ 1120/ 1236] Overall Loss 0.181860 Objective Loss 0.181860 LR 0.000500 Time 0.021382 +2023-10-02 21:28:59,352 - Epoch: [124][ 1130/ 1236] Overall Loss 0.181838 Objective Loss 0.181838 LR 0.000500 Time 0.021378 +2023-10-02 21:28:59,562 - Epoch: [124][ 1140/ 1236] Overall Loss 0.181979 Objective Loss 0.181979 LR 0.000500 Time 0.021374 +2023-10-02 21:28:59,772 - Epoch: [124][ 1150/ 1236] Overall Loss 0.181961 Objective Loss 0.181961 LR 0.000500 Time 0.021370 +2023-10-02 21:28:59,981 - Epoch: [124][ 1160/ 1236] Overall Loss 0.181890 Objective Loss 0.181890 LR 0.000500 Time 0.021366 +2023-10-02 21:29:00,191 - Epoch: [124][ 1170/ 1236] Overall Loss 0.181961 Objective Loss 0.181961 LR 0.000500 Time 0.021363 +2023-10-02 21:29:00,401 - Epoch: [124][ 1180/ 1236] Overall Loss 0.181970 Objective Loss 0.181970 LR 0.000500 Time 0.021360 +2023-10-02 21:29:00,611 - Epoch: [124][ 1190/ 1236] Overall Loss 0.182300 Objective Loss 0.182300 LR 0.000500 Time 0.021356 +2023-10-02 21:29:00,821 - Epoch: [124][ 1200/ 1236] Overall Loss 0.182224 Objective Loss 0.182224 LR 0.000500 Time 0.021353 +2023-10-02 21:29:01,031 - Epoch: [124][ 1210/ 1236] Overall Loss 0.182104 Objective Loss 0.182104 LR 0.000500 Time 0.021350 +2023-10-02 21:29:01,241 - Epoch: [124][ 1220/ 1236] Overall Loss 0.181967 Objective Loss 0.181967 LR 0.000500 Time 0.021347 +2023-10-02 21:29:01,501 - Epoch: [124][ 1230/ 1236] Overall Loss 0.181814 Objective Loss 0.181814 LR 0.000500 Time 0.021385 +2023-10-02 21:29:01,623 - Epoch: [124][ 1236/ 1236] Overall Loss 0.182016 Objective Loss 0.182016 Top1 87.780041 Top5 97.963340 LR 0.000500 Time 0.021379 +2023-10-02 21:29:01,754 - --- validate (epoch=124)----------- +2023-10-02 21:29:01,754 - 29943 samples (256 per mini-batch) +2023-10-02 21:29:02,249 - Epoch: [124][ 10/ 117] Loss 0.306860 Top1 84.648438 Top5 98.125000 +2023-10-02 21:29:02,400 - Epoch: [124][ 20/ 117] Loss 0.295688 Top1 85.664062 Top5 98.085938 +2023-10-02 21:29:02,552 - Epoch: [124][ 30/ 117] Loss 0.298685 Top1 85.716146 Top5 98.138021 +2023-10-02 21:29:02,702 - Epoch: [124][ 40/ 117] Loss 0.292044 Top1 85.878906 Top5 98.330078 +2023-10-02 21:29:02,853 - Epoch: [124][ 50/ 117] Loss 0.302200 Top1 85.546875 Top5 98.304688 +2023-10-02 21:29:03,004 - Epoch: [124][ 60/ 117] Loss 0.303214 Top1 85.625000 Top5 98.313802 +2023-10-02 21:29:03,155 - Epoch: [124][ 70/ 117] Loss 0.303120 Top1 85.641741 Top5 98.297991 +2023-10-02 21:29:03,305 - Epoch: [124][ 80/ 117] Loss 0.299418 Top1 85.810547 Top5 98.325195 +2023-10-02 21:29:03,455 - Epoch: [124][ 90/ 117] Loss 0.295992 Top1 85.746528 Top5 98.346354 +2023-10-02 21:29:03,606 - Epoch: [124][ 100/ 117] Loss 0.300544 Top1 85.750000 Top5 98.316406 +2023-10-02 21:29:03,763 - Epoch: [124][ 110/ 117] Loss 0.300583 Top1 85.756392 Top5 98.341619 +2023-10-02 21:29:03,851 - Epoch: [124][ 117/ 117] Loss 0.298982 Top1 85.782988 Top5 98.333500 +2023-10-02 21:29:03,997 - ==> Top1: 85.783 Top5: 98.334 Loss: 0.299 + +2023-10-02 21:29:03,997 - ==> Confusion: +[[ 928 0 6 2 5 2 0 0 11 67 1 0 0 1 6 1 2 0 2 0 16] + [ 0 1073 1 0 4 14 3 15 1 1 1 0 1 0 0 3 0 0 7 2 5] + [ 1 0 994 6 3 0 15 3 1 1 2 2 6 3 0 3 0 1 10 2 3] + [ 0 4 15 964 3 3 0 0 4 2 4 0 5 2 36 4 0 10 13 0 20] + [ 31 9 1 1 960 4 0 0 2 12 0 0 2 3 8 3 8 1 0 0 5] + [ 4 52 0 4 1 967 2 30 5 4 1 2 1 10 5 1 3 1 4 4 15] + [ 0 0 30 1 0 2 1131 4 0 0 5 0 0 0 0 3 0 1 2 6 6] + [ 2 12 18 0 2 19 4 1067 1 4 1 2 4 6 2 0 2 2 51 8 11] + [ 15 1 0 2 2 1 0 0 986 42 7 2 0 8 11 1 4 1 0 4 2] + [ 71 0 2 1 1 5 0 0 26 977 1 1 0 15 5 1 1 1 0 2 9] + [ 4 3 8 4 0 1 3 5 15 0 969 1 2 12 4 0 4 2 8 0 8] + [ 0 1 2 0 1 12 1 3 0 0 0 954 27 9 0 1 2 15 0 4 3] + [ 0 2 3 2 0 1 0 0 0 0 2 29 985 1 1 7 3 11 5 4 12] + [ 0 1 0 0 2 5 1 0 9 13 7 7 0 1049 5 3 0 0 0 3 14] + [ 12 0 5 15 5 0 0 0 21 2 3 1 2 0 1020 0 0 2 7 0 6] + [ 0 1 1 0 4 0 0 0 0 0 0 6 7 0 0 1082 13 8 2 6 4] + [ 2 17 2 0 4 7 0 0 0 0 0 6 0 2 4 11 1092 0 0 5 9] + [ 0 0 0 1 0 0 3 0 0 0 0 3 22 0 2 6 1 993 0 2 5] + [ 1 5 5 20 0 0 0 16 5 0 2 0 1 0 14 0 1 0 983 0 15] + [ 0 4 4 1 0 2 13 4 0 1 1 10 6 1 0 0 11 0 2 1080 12] + [ 124 161 132 74 66 124 49 78 97 85 166 91 329 247 149 50 93 69 139 150 5432]] + +2023-10-02 21:29:03,999 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:29:03,999 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:29:04,005 - + +2023-10-02 21:29:04,005 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:29:05,131 - Epoch: [125][ 10/ 1236] Overall Loss 0.171626 Objective Loss 0.171626 LR 0.000500 Time 0.112547 +2023-10-02 21:29:05,337 - Epoch: [125][ 20/ 1236] Overall Loss 0.174057 Objective Loss 0.174057 LR 0.000500 Time 0.066561 +2023-10-02 21:29:05,543 - Epoch: [125][ 30/ 1236] Overall Loss 0.183082 Objective Loss 0.183082 LR 0.000500 Time 0.051220 +2023-10-02 21:29:05,749 - Epoch: [125][ 40/ 1236] Overall Loss 0.184987 Objective Loss 0.184987 LR 0.000500 Time 0.043560 +2023-10-02 21:29:05,954 - Epoch: [125][ 50/ 1236] Overall Loss 0.183531 Objective Loss 0.183531 LR 0.000500 Time 0.038956 +2023-10-02 21:29:06,161 - Epoch: [125][ 60/ 1236] Overall Loss 0.183881 Objective Loss 0.183881 LR 0.000500 Time 0.035897 +2023-10-02 21:29:06,367 - Epoch: [125][ 70/ 1236] Overall Loss 0.185226 Objective Loss 0.185226 LR 0.000500 Time 0.033704 +2023-10-02 21:29:06,572 - Epoch: [125][ 80/ 1236] Overall Loss 0.184273 Objective Loss 0.184273 LR 0.000500 Time 0.032057 +2023-10-02 21:29:06,778 - Epoch: [125][ 90/ 1236] Overall Loss 0.183440 Objective Loss 0.183440 LR 0.000500 Time 0.030774 +2023-10-02 21:29:06,984 - Epoch: [125][ 100/ 1236] Overall Loss 0.182992 Objective Loss 0.182992 LR 0.000500 Time 0.029761 +2023-10-02 21:29:07,189 - Epoch: [125][ 110/ 1236] Overall Loss 0.183469 Objective Loss 0.183469 LR 0.000500 Time 0.028906 +2023-10-02 21:29:07,396 - Epoch: [125][ 120/ 1236] Overall Loss 0.182337 Objective Loss 0.182337 LR 0.000500 Time 0.028220 +2023-10-02 21:29:07,601 - Epoch: [125][ 130/ 1236] Overall Loss 0.181355 Objective Loss 0.181355 LR 0.000500 Time 0.027615 +2023-10-02 21:29:07,808 - Epoch: [125][ 140/ 1236] Overall Loss 0.182117 Objective Loss 0.182117 LR 0.000500 Time 0.027116 +2023-10-02 21:29:08,013 - Epoch: [125][ 150/ 1236] Overall Loss 0.180497 Objective Loss 0.180497 LR 0.000500 Time 0.026667 +2023-10-02 21:29:08,220 - Epoch: [125][ 160/ 1236] Overall Loss 0.179455 Objective Loss 0.179455 LR 0.000500 Time 0.026290 +2023-10-02 21:29:08,425 - Epoch: [125][ 170/ 1236] Overall Loss 0.179478 Objective Loss 0.179478 LR 0.000500 Time 0.025942 +2023-10-02 21:29:08,634 - Epoch: [125][ 180/ 1236] Overall Loss 0.180182 Objective Loss 0.180182 LR 0.000500 Time 0.025656 +2023-10-02 21:29:08,838 - Epoch: [125][ 190/ 1236] Overall Loss 0.179446 Objective Loss 0.179446 LR 0.000500 Time 0.025380 +2023-10-02 21:29:09,046 - Epoch: [125][ 200/ 1236] Overall Loss 0.178988 Objective Loss 0.178988 LR 0.000500 Time 0.025150 +2023-10-02 21:29:09,250 - Epoch: [125][ 210/ 1236] Overall Loss 0.179889 Objective Loss 0.179889 LR 0.000500 Time 0.024924 +2023-10-02 21:29:09,458 - Epoch: [125][ 220/ 1236] Overall Loss 0.180472 Objective Loss 0.180472 LR 0.000500 Time 0.024736 +2023-10-02 21:29:09,663 - Epoch: [125][ 230/ 1236] Overall Loss 0.180560 Objective Loss 0.180560 LR 0.000500 Time 0.024549 +2023-10-02 21:29:09,871 - Epoch: [125][ 240/ 1236] Overall Loss 0.180725 Objective Loss 0.180725 LR 0.000500 Time 0.024389 +2023-10-02 21:29:10,076 - Epoch: [125][ 250/ 1236] Overall Loss 0.181065 Objective Loss 0.181065 LR 0.000500 Time 0.024231 +2023-10-02 21:29:10,283 - Epoch: [125][ 260/ 1236] Overall Loss 0.181232 Objective Loss 0.181232 LR 0.000500 Time 0.024094 +2023-10-02 21:29:10,489 - Epoch: [125][ 270/ 1236] Overall Loss 0.180996 Objective Loss 0.180996 LR 0.000500 Time 0.023958 +2023-10-02 21:29:10,696 - Epoch: [125][ 280/ 1236] Overall Loss 0.181634 Objective Loss 0.181634 LR 0.000500 Time 0.023840 +2023-10-02 21:29:10,901 - Epoch: [125][ 290/ 1236] Overall Loss 0.181707 Objective Loss 0.181707 LR 0.000500 Time 0.023721 +2023-10-02 21:29:11,108 - Epoch: [125][ 300/ 1236] Overall Loss 0.182225 Objective Loss 0.182225 LR 0.000500 Time 0.023618 +2023-10-02 21:29:11,314 - Epoch: [125][ 310/ 1236] Overall Loss 0.181658 Objective Loss 0.181658 LR 0.000500 Time 0.023515 +2023-10-02 21:29:11,520 - Epoch: [125][ 320/ 1236] Overall Loss 0.180837 Objective Loss 0.180837 LR 0.000500 Time 0.023425 +2023-10-02 21:29:11,726 - Epoch: [125][ 330/ 1236] Overall Loss 0.180552 Objective Loss 0.180552 LR 0.000500 Time 0.023332 +2023-10-02 21:29:11,932 - Epoch: [125][ 340/ 1236] Overall Loss 0.180584 Objective Loss 0.180584 LR 0.000500 Time 0.023254 +2023-10-02 21:29:12,138 - Epoch: [125][ 350/ 1236] Overall Loss 0.180941 Objective Loss 0.180941 LR 0.000500 Time 0.023172 +2023-10-02 21:29:12,345 - Epoch: [125][ 360/ 1236] Overall Loss 0.180812 Objective Loss 0.180812 LR 0.000500 Time 0.023102 +2023-10-02 21:29:12,551 - Epoch: [125][ 370/ 1236] Overall Loss 0.180535 Objective Loss 0.180535 LR 0.000500 Time 0.023031 +2023-10-02 21:29:12,758 - Epoch: [125][ 380/ 1236] Overall Loss 0.179806 Objective Loss 0.179806 LR 0.000500 Time 0.022968 +2023-10-02 21:29:12,963 - Epoch: [125][ 390/ 1236] Overall Loss 0.179579 Objective Loss 0.179579 LR 0.000500 Time 0.022905 +2023-10-02 21:29:13,170 - Epoch: [125][ 400/ 1236] Overall Loss 0.179236 Objective Loss 0.179236 LR 0.000500 Time 0.022849 +2023-10-02 21:29:13,376 - Epoch: [125][ 410/ 1236] Overall Loss 0.179008 Objective Loss 0.179008 LR 0.000500 Time 0.022792 +2023-10-02 21:29:13,582 - Epoch: [125][ 420/ 1236] Overall Loss 0.179263 Objective Loss 0.179263 LR 0.000500 Time 0.022740 +2023-10-02 21:29:13,786 - Epoch: [125][ 430/ 1236] Overall Loss 0.179075 Objective Loss 0.179075 LR 0.000500 Time 0.022685 +2023-10-02 21:29:13,993 - Epoch: [125][ 440/ 1236] Overall Loss 0.179493 Objective Loss 0.179493 LR 0.000500 Time 0.022639 +2023-10-02 21:29:14,199 - Epoch: [125][ 450/ 1236] Overall Loss 0.179759 Objective Loss 0.179759 LR 0.000500 Time 0.022592 +2023-10-02 21:29:14,405 - Epoch: [125][ 460/ 1236] Overall Loss 0.179955 Objective Loss 0.179955 LR 0.000500 Time 0.022549 +2023-10-02 21:29:14,610 - Epoch: [125][ 470/ 1236] Overall Loss 0.179882 Objective Loss 0.179882 LR 0.000500 Time 0.022506 +2023-10-02 21:29:14,817 - Epoch: [125][ 480/ 1236] Overall Loss 0.179922 Objective Loss 0.179922 LR 0.000500 Time 0.022467 +2023-10-02 21:29:15,022 - Epoch: [125][ 490/ 1236] Overall Loss 0.179440 Objective Loss 0.179440 LR 0.000500 Time 0.022427 +2023-10-02 21:29:15,229 - Epoch: [125][ 500/ 1236] Overall Loss 0.179379 Objective Loss 0.179379 LR 0.000500 Time 0.022391 +2023-10-02 21:29:15,435 - Epoch: [125][ 510/ 1236] Overall Loss 0.179539 Objective Loss 0.179539 LR 0.000500 Time 0.022356 +2023-10-02 21:29:15,642 - Epoch: [125][ 520/ 1236] Overall Loss 0.179397 Objective Loss 0.179397 LR 0.000500 Time 0.022322 +2023-10-02 21:29:15,845 - Epoch: [125][ 530/ 1236] Overall Loss 0.179516 Objective Loss 0.179516 LR 0.000500 Time 0.022283 +2023-10-02 21:29:16,047 - Epoch: [125][ 540/ 1236] Overall Loss 0.179634 Objective Loss 0.179634 LR 0.000500 Time 0.022245 +2023-10-02 21:29:16,250 - Epoch: [125][ 550/ 1236] Overall Loss 0.179967 Objective Loss 0.179967 LR 0.000500 Time 0.022208 +2023-10-02 21:29:16,452 - Epoch: [125][ 560/ 1236] Overall Loss 0.180124 Objective Loss 0.180124 LR 0.000500 Time 0.022173 +2023-10-02 21:29:16,654 - Epoch: [125][ 570/ 1236] Overall Loss 0.180126 Objective Loss 0.180126 LR 0.000500 Time 0.022138 +2023-10-02 21:29:16,857 - Epoch: [125][ 580/ 1236] Overall Loss 0.180135 Objective Loss 0.180135 LR 0.000500 Time 0.022105 +2023-10-02 21:29:17,059 - Epoch: [125][ 590/ 1236] Overall Loss 0.179898 Objective Loss 0.179898 LR 0.000500 Time 0.022073 +2023-10-02 21:29:17,262 - Epoch: [125][ 600/ 1236] Overall Loss 0.179449 Objective Loss 0.179449 LR 0.000500 Time 0.022042 +2023-10-02 21:29:17,464 - Epoch: [125][ 610/ 1236] Overall Loss 0.179351 Objective Loss 0.179351 LR 0.000500 Time 0.022012 +2023-10-02 21:29:17,667 - Epoch: [125][ 620/ 1236] Overall Loss 0.179622 Objective Loss 0.179622 LR 0.000500 Time 0.021983 +2023-10-02 21:29:17,869 - Epoch: [125][ 630/ 1236] Overall Loss 0.179870 Objective Loss 0.179870 LR 0.000500 Time 0.021955 +2023-10-02 21:29:18,072 - Epoch: [125][ 640/ 1236] Overall Loss 0.179813 Objective Loss 0.179813 LR 0.000500 Time 0.021929 +2023-10-02 21:29:18,274 - Epoch: [125][ 650/ 1236] Overall Loss 0.179830 Objective Loss 0.179830 LR 0.000500 Time 0.021902 +2023-10-02 21:29:18,477 - Epoch: [125][ 660/ 1236] Overall Loss 0.179942 Objective Loss 0.179942 LR 0.000500 Time 0.021877 +2023-10-02 21:29:18,679 - Epoch: [125][ 670/ 1236] Overall Loss 0.179826 Objective Loss 0.179826 LR 0.000500 Time 0.021852 +2023-10-02 21:29:18,882 - Epoch: [125][ 680/ 1236] Overall Loss 0.179955 Objective Loss 0.179955 LR 0.000500 Time 0.021828 +2023-10-02 21:29:19,084 - Epoch: [125][ 690/ 1236] Overall Loss 0.179966 Objective Loss 0.179966 LR 0.000500 Time 0.021805 +2023-10-02 21:29:19,287 - Epoch: [125][ 700/ 1236] Overall Loss 0.180033 Objective Loss 0.180033 LR 0.000500 Time 0.021783 +2023-10-02 21:29:19,489 - Epoch: [125][ 710/ 1236] Overall Loss 0.180027 Objective Loss 0.180027 LR 0.000500 Time 0.021760 +2023-10-02 21:29:19,692 - Epoch: [125][ 720/ 1236] Overall Loss 0.180056 Objective Loss 0.180056 LR 0.000500 Time 0.021739 +2023-10-02 21:29:19,894 - Epoch: [125][ 730/ 1236] Overall Loss 0.180006 Objective Loss 0.180006 LR 0.000500 Time 0.021718 +2023-10-02 21:29:20,096 - Epoch: [125][ 740/ 1236] Overall Loss 0.179984 Objective Loss 0.179984 LR 0.000500 Time 0.021697 +2023-10-02 21:29:20,298 - Epoch: [125][ 750/ 1236] Overall Loss 0.180235 Objective Loss 0.180235 LR 0.000500 Time 0.021677 +2023-10-02 21:29:20,501 - Epoch: [125][ 760/ 1236] Overall Loss 0.180276 Objective Loss 0.180276 LR 0.000500 Time 0.021659 +2023-10-02 21:29:20,703 - Epoch: [125][ 770/ 1236] Overall Loss 0.180280 Objective Loss 0.180280 LR 0.000500 Time 0.021640 +2023-10-02 21:29:20,906 - Epoch: [125][ 780/ 1236] Overall Loss 0.180249 Objective Loss 0.180249 LR 0.000500 Time 0.021622 +2023-10-02 21:29:21,108 - Epoch: [125][ 790/ 1236] Overall Loss 0.180535 Objective Loss 0.180535 LR 0.000500 Time 0.021604 +2023-10-02 21:29:21,311 - Epoch: [125][ 800/ 1236] Overall Loss 0.180310 Objective Loss 0.180310 LR 0.000500 Time 0.021587 +2023-10-02 21:29:21,513 - Epoch: [125][ 810/ 1236] Overall Loss 0.180571 Objective Loss 0.180571 LR 0.000500 Time 0.021569 +2023-10-02 21:29:21,716 - Epoch: [125][ 820/ 1236] Overall Loss 0.180586 Objective Loss 0.180586 LR 0.000500 Time 0.021553 +2023-10-02 21:29:21,918 - Epoch: [125][ 830/ 1236] Overall Loss 0.180709 Objective Loss 0.180709 LR 0.000500 Time 0.021536 +2023-10-02 21:29:22,120 - Epoch: [125][ 840/ 1236] Overall Loss 0.180935 Objective Loss 0.180935 LR 0.000500 Time 0.021521 +2023-10-02 21:29:22,322 - Epoch: [125][ 850/ 1236] Overall Loss 0.180682 Objective Loss 0.180682 LR 0.000500 Time 0.021505 +2023-10-02 21:29:22,525 - Epoch: [125][ 860/ 1236] Overall Loss 0.180832 Objective Loss 0.180832 LR 0.000500 Time 0.021490 +2023-10-02 21:29:22,727 - Epoch: [125][ 870/ 1236] Overall Loss 0.181118 Objective Loss 0.181118 LR 0.000500 Time 0.021475 +2023-10-02 21:29:22,929 - Epoch: [125][ 880/ 1236] Overall Loss 0.181195 Objective Loss 0.181195 LR 0.000500 Time 0.021461 +2023-10-02 21:29:23,131 - Epoch: [125][ 890/ 1236] Overall Loss 0.180876 Objective Loss 0.180876 LR 0.000500 Time 0.021447 +2023-10-02 21:29:23,334 - Epoch: [125][ 900/ 1236] Overall Loss 0.181108 Objective Loss 0.181108 LR 0.000500 Time 0.021433 +2023-10-02 21:29:23,536 - Epoch: [125][ 910/ 1236] Overall Loss 0.181089 Objective Loss 0.181089 LR 0.000500 Time 0.021419 +2023-10-02 21:29:23,739 - Epoch: [125][ 920/ 1236] Overall Loss 0.181157 Objective Loss 0.181157 LR 0.000500 Time 0.021406 +2023-10-02 21:29:23,941 - Epoch: [125][ 930/ 1236] Overall Loss 0.180926 Objective Loss 0.180926 LR 0.000500 Time 0.021393 +2023-10-02 21:29:24,143 - Epoch: [125][ 940/ 1236] Overall Loss 0.180891 Objective Loss 0.180891 LR 0.000500 Time 0.021380 +2023-10-02 21:29:24,345 - Epoch: [125][ 950/ 1236] Overall Loss 0.181048 Objective Loss 0.181048 LR 0.000500 Time 0.021368 +2023-10-02 21:29:24,548 - Epoch: [125][ 960/ 1236] Overall Loss 0.181105 Objective Loss 0.181105 LR 0.000500 Time 0.021356 +2023-10-02 21:29:24,750 - Epoch: [125][ 970/ 1236] Overall Loss 0.181346 Objective Loss 0.181346 LR 0.000500 Time 0.021344 +2023-10-02 21:29:24,952 - Epoch: [125][ 980/ 1236] Overall Loss 0.181524 Objective Loss 0.181524 LR 0.000500 Time 0.021333 +2023-10-02 21:29:25,154 - Epoch: [125][ 990/ 1236] Overall Loss 0.181478 Objective Loss 0.181478 LR 0.000500 Time 0.021321 +2023-10-02 21:29:25,357 - Epoch: [125][ 1000/ 1236] Overall Loss 0.181545 Objective Loss 0.181545 LR 0.000500 Time 0.021310 +2023-10-02 21:29:25,559 - Epoch: [125][ 1010/ 1236] Overall Loss 0.181471 Objective Loss 0.181471 LR 0.000500 Time 0.021299 +2023-10-02 21:29:25,762 - Epoch: [125][ 1020/ 1236] Overall Loss 0.181588 Objective Loss 0.181588 LR 0.000500 Time 0.021289 +2023-10-02 21:29:25,964 - Epoch: [125][ 1030/ 1236] Overall Loss 0.181602 Objective Loss 0.181602 LR 0.000500 Time 0.021278 +2023-10-02 21:29:26,166 - Epoch: [125][ 1040/ 1236] Overall Loss 0.181814 Objective Loss 0.181814 LR 0.000500 Time 0.021267 +2023-10-02 21:29:26,369 - Epoch: [125][ 1050/ 1236] Overall Loss 0.181997 Objective Loss 0.181997 LR 0.000500 Time 0.021258 +2023-10-02 21:29:26,571 - Epoch: [125][ 1060/ 1236] Overall Loss 0.182111 Objective Loss 0.182111 LR 0.000500 Time 0.021248 +2023-10-02 21:29:26,773 - Epoch: [125][ 1070/ 1236] Overall Loss 0.182060 Objective Loss 0.182060 LR 0.000500 Time 0.021238 +2023-10-02 21:29:26,976 - Epoch: [125][ 1080/ 1236] Overall Loss 0.182218 Objective Loss 0.182218 LR 0.000500 Time 0.021229 +2023-10-02 21:29:27,178 - Epoch: [125][ 1090/ 1236] Overall Loss 0.182324 Objective Loss 0.182324 LR 0.000500 Time 0.021219 +2023-10-02 21:29:27,381 - Epoch: [125][ 1100/ 1236] Overall Loss 0.182538 Objective Loss 0.182538 LR 0.000500 Time 0.021210 +2023-10-02 21:29:27,582 - Epoch: [125][ 1110/ 1236] Overall Loss 0.182476 Objective Loss 0.182476 LR 0.000500 Time 0.021201 +2023-10-02 21:29:27,785 - Epoch: [125][ 1120/ 1236] Overall Loss 0.182648 Objective Loss 0.182648 LR 0.000500 Time 0.021192 +2023-10-02 21:29:27,987 - Epoch: [125][ 1130/ 1236] Overall Loss 0.182637 Objective Loss 0.182637 LR 0.000500 Time 0.021183 +2023-10-02 21:29:28,190 - Epoch: [125][ 1140/ 1236] Overall Loss 0.182993 Objective Loss 0.182993 LR 0.000500 Time 0.021175 +2023-10-02 21:29:28,393 - Epoch: [125][ 1150/ 1236] Overall Loss 0.182857 Objective Loss 0.182857 LR 0.000500 Time 0.021167 +2023-10-02 21:29:28,595 - Epoch: [125][ 1160/ 1236] Overall Loss 0.182875 Objective Loss 0.182875 LR 0.000500 Time 0.021159 +2023-10-02 21:29:28,798 - Epoch: [125][ 1170/ 1236] Overall Loss 0.182901 Objective Loss 0.182901 LR 0.000500 Time 0.021151 +2023-10-02 21:29:29,000 - Epoch: [125][ 1180/ 1236] Overall Loss 0.183029 Objective Loss 0.183029 LR 0.000500 Time 0.021143 +2023-10-02 21:29:29,203 - Epoch: [125][ 1190/ 1236] Overall Loss 0.182881 Objective Loss 0.182881 LR 0.000500 Time 0.021135 +2023-10-02 21:29:29,405 - Epoch: [125][ 1200/ 1236] Overall Loss 0.182870 Objective Loss 0.182870 LR 0.000500 Time 0.021127 +2023-10-02 21:29:29,607 - Epoch: [125][ 1210/ 1236] Overall Loss 0.182876 Objective Loss 0.182876 LR 0.000500 Time 0.021120 +2023-10-02 21:29:29,810 - Epoch: [125][ 1220/ 1236] Overall Loss 0.183021 Objective Loss 0.183021 LR 0.000500 Time 0.021112 +2023-10-02 21:29:30,064 - Epoch: [125][ 1230/ 1236] Overall Loss 0.183104 Objective Loss 0.183104 LR 0.000500 Time 0.021147 +2023-10-02 21:29:30,185 - Epoch: [125][ 1236/ 1236] Overall Loss 0.183082 Objective Loss 0.183082 Top1 88.391039 Top5 98.167006 LR 0.000500 Time 0.021142 +2023-10-02 21:29:30,333 - --- validate (epoch=125)----------- +2023-10-02 21:29:30,333 - 29943 samples (256 per mini-batch) +2023-10-02 21:29:30,818 - Epoch: [125][ 10/ 117] Loss 0.342826 Top1 84.804688 Top5 98.046875 +2023-10-02 21:29:30,971 - Epoch: [125][ 20/ 117] Loss 0.310651 Top1 85.546875 Top5 98.183594 +2023-10-02 21:29:31,124 - Epoch: [125][ 30/ 117] Loss 0.310373 Top1 85.442708 Top5 98.098958 +2023-10-02 21:29:31,277 - Epoch: [125][ 40/ 117] Loss 0.306756 Top1 85.390625 Top5 97.968750 +2023-10-02 21:29:31,431 - Epoch: [125][ 50/ 117] Loss 0.301506 Top1 85.476562 Top5 98.078125 +2023-10-02 21:29:31,583 - Epoch: [125][ 60/ 117] Loss 0.307977 Top1 85.416667 Top5 98.059896 +2023-10-02 21:29:31,735 - Epoch: [125][ 70/ 117] Loss 0.311130 Top1 85.312500 Top5 98.108259 +2023-10-02 21:29:31,887 - Epoch: [125][ 80/ 117] Loss 0.306666 Top1 85.297852 Top5 98.168945 +2023-10-02 21:29:32,041 - Epoch: [125][ 90/ 117] Loss 0.304119 Top1 85.334201 Top5 98.203125 +2023-10-02 21:29:32,192 - Epoch: [125][ 100/ 117] Loss 0.300800 Top1 85.398438 Top5 98.242188 +2023-10-02 21:29:32,351 - Epoch: [125][ 110/ 117] Loss 0.303365 Top1 85.344460 Top5 98.274148 +2023-10-02 21:29:32,439 - Epoch: [125][ 117/ 117] Loss 0.302505 Top1 85.348829 Top5 98.283405 +2023-10-02 21:29:32,578 - ==> Top1: 85.349 Top5: 98.283 Loss: 0.303 + +2023-10-02 21:29:32,579 - ==> Confusion: +[[ 936 2 2 0 7 3 0 0 9 59 1 0 1 2 6 0 4 2 3 0 13] + [ 0 1074 0 0 7 9 1 13 3 1 1 1 0 0 1 4 0 0 10 2 4] + [ 4 0 981 11 0 0 16 7 0 0 1 1 9 3 2 3 2 1 8 2 5] + [ 2 1 12 982 1 2 1 1 2 0 3 0 3 3 36 3 0 3 15 1 18] + [ 22 9 1 0 968 2 0 0 2 9 0 0 3 4 11 3 8 1 2 2 3] + [ 5 54 1 0 1 963 1 29 4 6 7 6 0 8 6 1 4 0 4 5 11] + [ 0 2 31 1 0 0 1132 6 0 0 2 0 0 1 0 1 0 0 2 10 3] + [ 3 17 16 0 2 18 2 1073 1 3 5 2 6 5 1 0 2 2 40 12 8] + [ 14 4 0 0 1 3 3 3 986 38 7 1 2 7 13 1 0 0 4 1 1] + [ 94 2 1 0 7 2 0 1 31 942 1 0 1 16 10 1 0 0 1 1 8] + [ 1 5 8 7 0 1 3 2 11 0 975 1 0 11 6 1 1 2 7 2 9] + [ 0 1 3 0 2 12 1 6 0 0 0 948 26 9 0 0 1 16 0 7 3] + [ 1 0 3 3 2 0 1 1 3 1 0 40 975 1 3 4 1 8 5 7 9] + [ 1 0 0 0 3 11 0 0 16 9 5 7 0 1046 6 0 0 1 0 2 12] + [ 8 1 3 13 4 0 0 0 22 1 1 0 3 1 1024 0 0 2 12 0 6] + [ 0 1 1 2 3 0 2 0 0 0 0 8 9 0 1 1060 19 13 2 7 6] + [ 0 13 2 0 4 5 1 1 0 0 0 4 0 3 5 4 1100 0 1 9 9] + [ 0 0 1 1 0 1 3 0 1 0 0 3 20 2 4 2 2 995 0 0 3] + [ 2 3 6 13 0 1 0 16 4 0 4 2 1 0 14 0 0 0 988 0 14] + [ 0 4 4 7 0 3 10 9 0 2 2 11 2 2 1 0 7 0 0 1082 6] + [ 120 172 135 105 72 113 54 74 112 81 190 82 322 250 164 48 116 63 133 173 5326]] + +2023-10-02 21:29:32,580 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:29:32,580 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:29:32,586 - + +2023-10-02 21:29:32,586 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:29:33,609 - Epoch: [126][ 10/ 1236] Overall Loss 0.185029 Objective Loss 0.185029 LR 0.000500 Time 0.102246 +2023-10-02 21:29:33,815 - Epoch: [126][ 20/ 1236] Overall Loss 0.193458 Objective Loss 0.193458 LR 0.000500 Time 0.061402 +2023-10-02 21:29:34,019 - Epoch: [126][ 30/ 1236] Overall Loss 0.186151 Objective Loss 0.186151 LR 0.000500 Time 0.047707 +2023-10-02 21:29:34,224 - Epoch: [126][ 40/ 1236] Overall Loss 0.182996 Objective Loss 0.182996 LR 0.000500 Time 0.040907 +2023-10-02 21:29:34,427 - Epoch: [126][ 50/ 1236] Overall Loss 0.184633 Objective Loss 0.184633 LR 0.000500 Time 0.036776 +2023-10-02 21:29:34,633 - Epoch: [126][ 60/ 1236] Overall Loss 0.180873 Objective Loss 0.180873 LR 0.000500 Time 0.034072 +2023-10-02 21:29:34,836 - Epoch: [126][ 70/ 1236] Overall Loss 0.179865 Objective Loss 0.179865 LR 0.000500 Time 0.032105 +2023-10-02 21:29:35,041 - Epoch: [126][ 80/ 1236] Overall Loss 0.180373 Objective Loss 0.180373 LR 0.000500 Time 0.030654 +2023-10-02 21:29:35,244 - Epoch: [126][ 90/ 1236] Overall Loss 0.181353 Objective Loss 0.181353 LR 0.000500 Time 0.029496 +2023-10-02 21:29:35,450 - Epoch: [126][ 100/ 1236] Overall Loss 0.181152 Objective Loss 0.181152 LR 0.000500 Time 0.028605 +2023-10-02 21:29:35,654 - Epoch: [126][ 110/ 1236] Overall Loss 0.179908 Objective Loss 0.179908 LR 0.000500 Time 0.027851 +2023-10-02 21:29:35,858 - Epoch: [126][ 120/ 1236] Overall Loss 0.180278 Objective Loss 0.180278 LR 0.000500 Time 0.027230 +2023-10-02 21:29:36,062 - Epoch: [126][ 130/ 1236] Overall Loss 0.180351 Objective Loss 0.180351 LR 0.000500 Time 0.026701 +2023-10-02 21:29:36,267 - Epoch: [126][ 140/ 1236] Overall Loss 0.178900 Objective Loss 0.178900 LR 0.000500 Time 0.026261 +2023-10-02 21:29:36,471 - Epoch: [126][ 150/ 1236] Overall Loss 0.179254 Objective Loss 0.179254 LR 0.000500 Time 0.025867 +2023-10-02 21:29:36,677 - Epoch: [126][ 160/ 1236] Overall Loss 0.180586 Objective Loss 0.180586 LR 0.000500 Time 0.025535 +2023-10-02 21:29:36,881 - Epoch: [126][ 170/ 1236] Overall Loss 0.181389 Objective Loss 0.181389 LR 0.000500 Time 0.025231 +2023-10-02 21:29:37,086 - Epoch: [126][ 180/ 1236] Overall Loss 0.182404 Objective Loss 0.182404 LR 0.000500 Time 0.024965 +2023-10-02 21:29:37,290 - Epoch: [126][ 190/ 1236] Overall Loss 0.181603 Objective Loss 0.181603 LR 0.000500 Time 0.024727 +2023-10-02 21:29:37,497 - Epoch: [126][ 200/ 1236] Overall Loss 0.181321 Objective Loss 0.181321 LR 0.000500 Time 0.024520 +2023-10-02 21:29:37,701 - Epoch: [126][ 210/ 1236] Overall Loss 0.182399 Objective Loss 0.182399 LR 0.000500 Time 0.024324 +2023-10-02 21:29:37,905 - Epoch: [126][ 220/ 1236] Overall Loss 0.183241 Objective Loss 0.183241 LR 0.000500 Time 0.024145 +2023-10-02 21:29:38,110 - Epoch: [126][ 230/ 1236] Overall Loss 0.183123 Objective Loss 0.183123 LR 0.000500 Time 0.023985 +2023-10-02 21:29:38,317 - Epoch: [126][ 240/ 1236] Overall Loss 0.182545 Objective Loss 0.182545 LR 0.000500 Time 0.023844 +2023-10-02 21:29:38,521 - Epoch: [126][ 250/ 1236] Overall Loss 0.182085 Objective Loss 0.182085 LR 0.000500 Time 0.023707 +2023-10-02 21:29:38,726 - Epoch: [126][ 260/ 1236] Overall Loss 0.181910 Objective Loss 0.181910 LR 0.000500 Time 0.023581 +2023-10-02 21:29:38,929 - Epoch: [126][ 270/ 1236] Overall Loss 0.182311 Objective Loss 0.182311 LR 0.000500 Time 0.023460 +2023-10-02 21:29:39,135 - Epoch: [126][ 280/ 1236] Overall Loss 0.182284 Objective Loss 0.182284 LR 0.000500 Time 0.023357 +2023-10-02 21:29:39,340 - Epoch: [126][ 290/ 1236] Overall Loss 0.181914 Objective Loss 0.181914 LR 0.000500 Time 0.023256 +2023-10-02 21:29:39,546 - Epoch: [126][ 300/ 1236] Overall Loss 0.181636 Objective Loss 0.181636 LR 0.000500 Time 0.023166 +2023-10-02 21:29:39,751 - Epoch: [126][ 310/ 1236] Overall Loss 0.181516 Objective Loss 0.181516 LR 0.000500 Time 0.023078 +2023-10-02 21:29:39,956 - Epoch: [126][ 320/ 1236] Overall Loss 0.181148 Objective Loss 0.181148 LR 0.000500 Time 0.023000 +2023-10-02 21:29:40,161 - Epoch: [126][ 330/ 1236] Overall Loss 0.181171 Objective Loss 0.181171 LR 0.000500 Time 0.022922 +2023-10-02 21:29:40,368 - Epoch: [126][ 340/ 1236] Overall Loss 0.181504 Objective Loss 0.181504 LR 0.000500 Time 0.022854 +2023-10-02 21:29:40,572 - Epoch: [126][ 350/ 1236] Overall Loss 0.181656 Objective Loss 0.181656 LR 0.000500 Time 0.022785 +2023-10-02 21:29:40,779 - Epoch: [126][ 360/ 1236] Overall Loss 0.182040 Objective Loss 0.182040 LR 0.000500 Time 0.022724 +2023-10-02 21:29:40,983 - Epoch: [126][ 370/ 1236] Overall Loss 0.181754 Objective Loss 0.181754 LR 0.000500 Time 0.022662 +2023-10-02 21:29:41,190 - Epoch: [126][ 380/ 1236] Overall Loss 0.181900 Objective Loss 0.181900 LR 0.000500 Time 0.022609 +2023-10-02 21:29:41,394 - Epoch: [126][ 390/ 1236] Overall Loss 0.182222 Objective Loss 0.182222 LR 0.000500 Time 0.022553 +2023-10-02 21:29:41,601 - Epoch: [126][ 400/ 1236] Overall Loss 0.182305 Objective Loss 0.182305 LR 0.000500 Time 0.022504 +2023-10-02 21:29:41,805 - Epoch: [126][ 410/ 1236] Overall Loss 0.182083 Objective Loss 0.182083 LR 0.000500 Time 0.022454 +2023-10-02 21:29:42,012 - Epoch: [126][ 420/ 1236] Overall Loss 0.181849 Objective Loss 0.181849 LR 0.000500 Time 0.022410 +2023-10-02 21:29:42,216 - Epoch: [126][ 430/ 1236] Overall Loss 0.181746 Objective Loss 0.181746 LR 0.000500 Time 0.022364 +2023-10-02 21:29:42,423 - Epoch: [126][ 440/ 1236] Overall Loss 0.181215 Objective Loss 0.181215 LR 0.000500 Time 0.022324 +2023-10-02 21:29:42,627 - Epoch: [126][ 450/ 1236] Overall Loss 0.181446 Objective Loss 0.181446 LR 0.000500 Time 0.022282 +2023-10-02 21:29:42,833 - Epoch: [126][ 460/ 1236] Overall Loss 0.181570 Objective Loss 0.181570 LR 0.000500 Time 0.022244 +2023-10-02 21:29:43,038 - Epoch: [126][ 470/ 1236] Overall Loss 0.181957 Objective Loss 0.181957 LR 0.000500 Time 0.022206 +2023-10-02 21:29:43,244 - Epoch: [126][ 480/ 1236] Overall Loss 0.181960 Objective Loss 0.181960 LR 0.000500 Time 0.022172 +2023-10-02 21:29:43,448 - Epoch: [126][ 490/ 1236] Overall Loss 0.182314 Objective Loss 0.182314 LR 0.000500 Time 0.022136 +2023-10-02 21:29:43,655 - Epoch: [126][ 500/ 1236] Overall Loss 0.182080 Objective Loss 0.182080 LR 0.000500 Time 0.022106 +2023-10-02 21:29:43,860 - Epoch: [126][ 510/ 1236] Overall Loss 0.182184 Objective Loss 0.182184 LR 0.000500 Time 0.022073 +2023-10-02 21:29:44,066 - Epoch: [126][ 520/ 1236] Overall Loss 0.182412 Objective Loss 0.182412 LR 0.000500 Time 0.022045 +2023-10-02 21:29:44,271 - Epoch: [126][ 530/ 1236] Overall Loss 0.182827 Objective Loss 0.182827 LR 0.000500 Time 0.022015 +2023-10-02 21:29:44,477 - Epoch: [126][ 540/ 1236] Overall Loss 0.183018 Objective Loss 0.183018 LR 0.000500 Time 0.021989 +2023-10-02 21:29:44,682 - Epoch: [126][ 550/ 1236] Overall Loss 0.182674 Objective Loss 0.182674 LR 0.000500 Time 0.021961 +2023-10-02 21:29:44,888 - Epoch: [126][ 560/ 1236] Overall Loss 0.182505 Objective Loss 0.182505 LR 0.000500 Time 0.021937 +2023-10-02 21:29:45,093 - Epoch: [126][ 570/ 1236] Overall Loss 0.182497 Objective Loss 0.182497 LR 0.000500 Time 0.021911 +2023-10-02 21:29:45,299 - Epoch: [126][ 580/ 1236] Overall Loss 0.182697 Objective Loss 0.182697 LR 0.000500 Time 0.021888 +2023-10-02 21:29:45,504 - Epoch: [126][ 590/ 1236] Overall Loss 0.182601 Objective Loss 0.182601 LR 0.000500 Time 0.021864 +2023-10-02 21:29:45,710 - Epoch: [126][ 600/ 1236] Overall Loss 0.183039 Objective Loss 0.183039 LR 0.000500 Time 0.021842 +2023-10-02 21:29:45,915 - Epoch: [126][ 610/ 1236] Overall Loss 0.183087 Objective Loss 0.183087 LR 0.000500 Time 0.021819 +2023-10-02 21:29:46,122 - Epoch: [126][ 620/ 1236] Overall Loss 0.182660 Objective Loss 0.182660 LR 0.000500 Time 0.021800 +2023-10-02 21:29:46,326 - Epoch: [126][ 630/ 1236] Overall Loss 0.182612 Objective Loss 0.182612 LR 0.000500 Time 0.021778 +2023-10-02 21:29:46,533 - Epoch: [126][ 640/ 1236] Overall Loss 0.182497 Objective Loss 0.182497 LR 0.000500 Time 0.021761 +2023-10-02 21:29:46,738 - Epoch: [126][ 650/ 1236] Overall Loss 0.182194 Objective Loss 0.182194 LR 0.000500 Time 0.021740 +2023-10-02 21:29:46,944 - Epoch: [126][ 660/ 1236] Overall Loss 0.181835 Objective Loss 0.181835 LR 0.000500 Time 0.021724 +2023-10-02 21:29:47,149 - Epoch: [126][ 670/ 1236] Overall Loss 0.181800 Objective Loss 0.181800 LR 0.000500 Time 0.021704 +2023-10-02 21:29:47,356 - Epoch: [126][ 680/ 1236] Overall Loss 0.182099 Objective Loss 0.182099 LR 0.000500 Time 0.021688 +2023-10-02 21:29:47,560 - Epoch: [126][ 690/ 1236] Overall Loss 0.182001 Objective Loss 0.182001 LR 0.000500 Time 0.021670 +2023-10-02 21:29:47,766 - Epoch: [126][ 700/ 1236] Overall Loss 0.182062 Objective Loss 0.182062 LR 0.000500 Time 0.021655 +2023-10-02 21:29:47,971 - Epoch: [126][ 710/ 1236] Overall Loss 0.181961 Objective Loss 0.181961 LR 0.000500 Time 0.021638 +2023-10-02 21:29:48,178 - Epoch: [126][ 720/ 1236] Overall Loss 0.181859 Objective Loss 0.181859 LR 0.000500 Time 0.021624 +2023-10-02 21:29:48,382 - Epoch: [126][ 730/ 1236] Overall Loss 0.181809 Objective Loss 0.181809 LR 0.000500 Time 0.021607 +2023-10-02 21:29:48,589 - Epoch: [126][ 740/ 1236] Overall Loss 0.181657 Objective Loss 0.181657 LR 0.000500 Time 0.021594 +2023-10-02 21:29:48,793 - Epoch: [126][ 750/ 1236] Overall Loss 0.181726 Objective Loss 0.181726 LR 0.000500 Time 0.021579 +2023-10-02 21:29:49,000 - Epoch: [126][ 760/ 1236] Overall Loss 0.181872 Objective Loss 0.181872 LR 0.000500 Time 0.021566 +2023-10-02 21:29:49,205 - Epoch: [126][ 770/ 1236] Overall Loss 0.181609 Objective Loss 0.181609 LR 0.000500 Time 0.021551 +2023-10-02 21:29:49,411 - Epoch: [126][ 780/ 1236] Overall Loss 0.181668 Objective Loss 0.181668 LR 0.000500 Time 0.021539 +2023-10-02 21:29:49,616 - Epoch: [126][ 790/ 1236] Overall Loss 0.181288 Objective Loss 0.181288 LR 0.000500 Time 0.021526 +2023-10-02 21:29:49,823 - Epoch: [126][ 800/ 1236] Overall Loss 0.181393 Objective Loss 0.181393 LR 0.000500 Time 0.021515 +2023-10-02 21:29:50,027 - Epoch: [126][ 810/ 1236] Overall Loss 0.181411 Objective Loss 0.181411 LR 0.000500 Time 0.021501 +2023-10-02 21:29:50,234 - Epoch: [126][ 820/ 1236] Overall Loss 0.181395 Objective Loss 0.181395 LR 0.000500 Time 0.021490 +2023-10-02 21:29:50,438 - Epoch: [126][ 830/ 1236] Overall Loss 0.181234 Objective Loss 0.181234 LR 0.000500 Time 0.021478 +2023-10-02 21:29:50,645 - Epoch: [126][ 840/ 1236] Overall Loss 0.181434 Objective Loss 0.181434 LR 0.000500 Time 0.021467 +2023-10-02 21:29:50,849 - Epoch: [126][ 850/ 1236] Overall Loss 0.181581 Objective Loss 0.181581 LR 0.000500 Time 0.021455 +2023-10-02 21:29:51,056 - Epoch: [126][ 860/ 1236] Overall Loss 0.181726 Objective Loss 0.181726 LR 0.000500 Time 0.021445 +2023-10-02 21:29:51,260 - Epoch: [126][ 870/ 1236] Overall Loss 0.181508 Objective Loss 0.181508 LR 0.000500 Time 0.021434 +2023-10-02 21:29:51,467 - Epoch: [126][ 880/ 1236] Overall Loss 0.181334 Objective Loss 0.181334 LR 0.000500 Time 0.021425 +2023-10-02 21:29:51,672 - Epoch: [126][ 890/ 1236] Overall Loss 0.181258 Objective Loss 0.181258 LR 0.000500 Time 0.021414 +2023-10-02 21:29:51,878 - Epoch: [126][ 900/ 1236] Overall Loss 0.180935 Objective Loss 0.180935 LR 0.000500 Time 0.021405 +2023-10-02 21:29:52,083 - Epoch: [126][ 910/ 1236] Overall Loss 0.181058 Objective Loss 0.181058 LR 0.000500 Time 0.021394 +2023-10-02 21:29:52,290 - Epoch: [126][ 920/ 1236] Overall Loss 0.181045 Objective Loss 0.181045 LR 0.000500 Time 0.021386 +2023-10-02 21:29:52,494 - Epoch: [126][ 930/ 1236] Overall Loss 0.180899 Objective Loss 0.180899 LR 0.000500 Time 0.021375 +2023-10-02 21:29:52,701 - Epoch: [126][ 940/ 1236] Overall Loss 0.180686 Objective Loss 0.180686 LR 0.000500 Time 0.021367 +2023-10-02 21:29:52,905 - Epoch: [126][ 950/ 1236] Overall Loss 0.180597 Objective Loss 0.180597 LR 0.000500 Time 0.021357 +2023-10-02 21:29:53,111 - Epoch: [126][ 960/ 1236] Overall Loss 0.180647 Objective Loss 0.180647 LR 0.000500 Time 0.021349 +2023-10-02 21:29:53,316 - Epoch: [126][ 970/ 1236] Overall Loss 0.180813 Objective Loss 0.180813 LR 0.000500 Time 0.021340 +2023-10-02 21:29:53,523 - Epoch: [126][ 980/ 1236] Overall Loss 0.181001 Objective Loss 0.181001 LR 0.000500 Time 0.021333 +2023-10-02 21:29:53,727 - Epoch: [126][ 990/ 1236] Overall Loss 0.180933 Objective Loss 0.180933 LR 0.000500 Time 0.021323 +2023-10-02 21:29:53,934 - Epoch: [126][ 1000/ 1236] Overall Loss 0.180751 Objective Loss 0.180751 LR 0.000500 Time 0.021317 +2023-10-02 21:29:54,138 - Epoch: [126][ 1010/ 1236] Overall Loss 0.180594 Objective Loss 0.180594 LR 0.000500 Time 0.021308 +2023-10-02 21:29:54,344 - Epoch: [126][ 1020/ 1236] Overall Loss 0.180643 Objective Loss 0.180643 LR 0.000500 Time 0.021301 +2023-10-02 21:29:54,549 - Epoch: [126][ 1030/ 1236] Overall Loss 0.180666 Objective Loss 0.180666 LR 0.000500 Time 0.021293 +2023-10-02 21:29:54,756 - Epoch: [126][ 1040/ 1236] Overall Loss 0.180657 Objective Loss 0.180657 LR 0.000500 Time 0.021286 +2023-10-02 21:29:54,961 - Epoch: [126][ 1050/ 1236] Overall Loss 0.180548 Objective Loss 0.180548 LR 0.000500 Time 0.021278 +2023-10-02 21:29:55,167 - Epoch: [126][ 1060/ 1236] Overall Loss 0.180674 Objective Loss 0.180674 LR 0.000500 Time 0.021272 +2023-10-02 21:29:55,372 - Epoch: [126][ 1070/ 1236] Overall Loss 0.180862 Objective Loss 0.180862 LR 0.000500 Time 0.021264 +2023-10-02 21:29:55,579 - Epoch: [126][ 1080/ 1236] Overall Loss 0.180780 Objective Loss 0.180780 LR 0.000500 Time 0.021259 +2023-10-02 21:29:55,783 - Epoch: [126][ 1090/ 1236] Overall Loss 0.180957 Objective Loss 0.180957 LR 0.000500 Time 0.021251 +2023-10-02 21:29:55,990 - Epoch: [126][ 1100/ 1236] Overall Loss 0.180777 Objective Loss 0.180777 LR 0.000500 Time 0.021245 +2023-10-02 21:29:56,194 - Epoch: [126][ 1110/ 1236] Overall Loss 0.180904 Objective Loss 0.180904 LR 0.000500 Time 0.021238 +2023-10-02 21:29:56,401 - Epoch: [126][ 1120/ 1236] Overall Loss 0.181063 Objective Loss 0.181063 LR 0.000500 Time 0.021233 +2023-10-02 21:29:56,606 - Epoch: [126][ 1130/ 1236] Overall Loss 0.181027 Objective Loss 0.181027 LR 0.000500 Time 0.021226 +2023-10-02 21:29:56,812 - Epoch: [126][ 1140/ 1236] Overall Loss 0.180921 Objective Loss 0.180921 LR 0.000500 Time 0.021220 +2023-10-02 21:29:57,017 - Epoch: [126][ 1150/ 1236] Overall Loss 0.180840 Objective Loss 0.180840 LR 0.000500 Time 0.021214 +2023-10-02 21:29:57,224 - Epoch: [126][ 1160/ 1236] Overall Loss 0.181044 Objective Loss 0.181044 LR 0.000500 Time 0.021209 +2023-10-02 21:29:57,429 - Epoch: [126][ 1170/ 1236] Overall Loss 0.181079 Objective Loss 0.181079 LR 0.000500 Time 0.021203 +2023-10-02 21:29:57,635 - Epoch: [126][ 1180/ 1236] Overall Loss 0.181103 Objective Loss 0.181103 LR 0.000500 Time 0.021197 +2023-10-02 21:29:57,840 - Epoch: [126][ 1190/ 1236] Overall Loss 0.180925 Objective Loss 0.180925 LR 0.000500 Time 0.021191 +2023-10-02 21:29:58,047 - Epoch: [126][ 1200/ 1236] Overall Loss 0.181073 Objective Loss 0.181073 LR 0.000500 Time 0.021186 +2023-10-02 21:29:58,252 - Epoch: [126][ 1210/ 1236] Overall Loss 0.181211 Objective Loss 0.181211 LR 0.000500 Time 0.021180 +2023-10-02 21:29:58,458 - Epoch: [126][ 1220/ 1236] Overall Loss 0.181295 Objective Loss 0.181295 LR 0.000500 Time 0.021176 +2023-10-02 21:29:58,715 - Epoch: [126][ 1230/ 1236] Overall Loss 0.181492 Objective Loss 0.181492 LR 0.000500 Time 0.021212 +2023-10-02 21:29:58,836 - Epoch: [126][ 1236/ 1236] Overall Loss 0.181594 Objective Loss 0.181594 Top1 88.187373 Top5 99.185336 LR 0.000500 Time 0.021207 +2023-10-02 21:29:58,972 - --- validate (epoch=126)----------- +2023-10-02 21:29:58,972 - 29943 samples (256 per mini-batch) +2023-10-02 21:29:59,470 - Epoch: [126][ 10/ 117] Loss 0.319591 Top1 85.312500 Top5 98.671875 +2023-10-02 21:29:59,621 - Epoch: [126][ 20/ 117] Loss 0.300004 Top1 85.937500 Top5 98.496094 +2023-10-02 21:29:59,773 - Epoch: [126][ 30/ 117] Loss 0.300497 Top1 85.377604 Top5 98.567708 +2023-10-02 21:29:59,924 - Epoch: [126][ 40/ 117] Loss 0.299609 Top1 85.244141 Top5 98.466797 +2023-10-02 21:30:00,075 - Epoch: [126][ 50/ 117] Loss 0.296690 Top1 85.257812 Top5 98.507812 +2023-10-02 21:30:00,225 - Epoch: [126][ 60/ 117] Loss 0.298570 Top1 85.494792 Top5 98.470052 +2023-10-02 21:30:00,377 - Epoch: [126][ 70/ 117] Loss 0.303888 Top1 85.452009 Top5 98.459821 +2023-10-02 21:30:00,528 - Epoch: [126][ 80/ 117] Loss 0.304612 Top1 85.488281 Top5 98.452148 +2023-10-02 21:30:00,681 - Epoch: [126][ 90/ 117] Loss 0.301678 Top1 85.555556 Top5 98.450521 +2023-10-02 21:30:00,832 - Epoch: [126][ 100/ 117] Loss 0.296609 Top1 85.691406 Top5 98.414062 +2023-10-02 21:30:00,991 - Epoch: [126][ 110/ 117] Loss 0.294710 Top1 85.664062 Top5 98.441051 +2023-10-02 21:30:01,079 - Epoch: [126][ 117/ 117] Loss 0.294924 Top1 85.702835 Top5 98.437030 +2023-10-02 21:30:01,223 - ==> Top1: 85.703 Top5: 98.437 Loss: 0.295 + +2023-10-02 21:30:01,223 - ==> Confusion: +[[ 941 1 3 1 5 2 0 0 6 62 1 0 0 3 2 2 3 0 1 0 17] + [ 0 1051 1 0 5 28 3 18 2 1 1 2 2 0 1 2 0 0 9 1 4] + [ 2 0 967 10 1 0 18 8 0 0 2 1 9 4 0 4 1 1 14 5 9] + [ 0 3 11 974 1 3 0 6 3 2 3 0 9 6 27 3 0 6 14 0 18] + [ 24 6 1 0 969 3 0 0 1 10 1 0 0 4 6 4 12 0 1 2 6] + [ 3 39 0 0 4 992 3 24 2 4 3 5 2 12 5 1 4 0 3 0 10] + [ 0 1 19 1 0 1 1133 6 0 0 5 0 0 1 0 3 0 0 3 10 8] + [ 0 11 6 0 4 26 5 1079 1 2 3 7 4 5 3 0 0 2 39 9 12] + [ 18 2 0 1 0 3 0 1 979 35 13 2 5 12 13 1 1 1 0 0 2] + [ 92 2 1 0 4 2 0 1 26 955 0 0 0 23 5 1 2 0 0 2 3] + [ 2 2 6 6 0 2 1 5 13 1 972 2 1 15 4 0 0 3 7 3 8] + [ 0 0 1 0 0 9 0 2 0 1 0 967 26 6 0 1 1 14 1 4 2] + [ 0 2 1 1 0 0 2 0 2 1 2 35 982 1 1 5 0 17 2 5 9] + [ 1 0 0 0 1 5 3 0 10 11 5 6 0 1058 3 1 1 1 0 0 13] + [ 10 2 6 15 5 0 0 0 22 1 4 0 3 2 1006 0 2 3 9 0 11] + [ 0 0 1 1 4 0 0 0 0 0 0 5 9 1 0 1069 16 9 2 10 7] + [ 1 12 0 0 4 8 0 2 0 0 0 5 0 1 5 8 1095 0 0 9 11] + [ 0 0 1 2 0 0 1 1 1 0 1 6 18 1 2 3 0 998 0 2 1] + [ 0 4 4 12 0 0 0 21 5 0 5 1 3 0 10 0 0 1 988 4 10] + [ 0 2 2 3 0 5 6 8 0 0 1 15 1 0 1 1 5 1 0 1095 6] + [ 127 156 108 71 69 133 31 87 81 86 161 108 334 286 109 50 81 70 154 211 5392]] + +2023-10-02 21:30:01,225 - ==> Best [Top1: 85.816 Top5: 98.414 Sparsity:0.00 Params: 169472 on epoch: 121] +2023-10-02 21:30:01,225 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:30:01,231 - + +2023-10-02 21:30:01,231 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:30:02,240 - Epoch: [127][ 10/ 1236] Overall Loss 0.160806 Objective Loss 0.160806 LR 0.000500 Time 0.100820 +2023-10-02 21:30:02,445 - Epoch: [127][ 20/ 1236] Overall Loss 0.177847 Objective Loss 0.177847 LR 0.000500 Time 0.060668 +2023-10-02 21:30:02,650 - Epoch: [127][ 30/ 1236] Overall Loss 0.178793 Objective Loss 0.178793 LR 0.000500 Time 0.047258 +2023-10-02 21:30:02,855 - Epoch: [127][ 40/ 1236] Overall Loss 0.177479 Objective Loss 0.177479 LR 0.000500 Time 0.040570 +2023-10-02 21:30:03,060 - Epoch: [127][ 50/ 1236] Overall Loss 0.180303 Objective Loss 0.180303 LR 0.000500 Time 0.036516 +2023-10-02 21:30:03,265 - Epoch: [127][ 60/ 1236] Overall Loss 0.178099 Objective Loss 0.178099 LR 0.000500 Time 0.033847 +2023-10-02 21:30:03,470 - Epoch: [127][ 70/ 1236] Overall Loss 0.180200 Objective Loss 0.180200 LR 0.000500 Time 0.031930 +2023-10-02 21:30:03,675 - Epoch: [127][ 80/ 1236] Overall Loss 0.179812 Objective Loss 0.179812 LR 0.000500 Time 0.030505 +2023-10-02 21:30:03,878 - Epoch: [127][ 90/ 1236] Overall Loss 0.179681 Objective Loss 0.179681 LR 0.000500 Time 0.029362 +2023-10-02 21:30:04,085 - Epoch: [127][ 100/ 1236] Overall Loss 0.179503 Objective Loss 0.179503 LR 0.000500 Time 0.028494 +2023-10-02 21:30:04,287 - Epoch: [127][ 110/ 1236] Overall Loss 0.179928 Objective Loss 0.179928 LR 0.000500 Time 0.027739 +2023-10-02 21:30:04,494 - Epoch: [127][ 120/ 1236] Overall Loss 0.179517 Objective Loss 0.179517 LR 0.000500 Time 0.027150 +2023-10-02 21:30:04,697 - Epoch: [127][ 130/ 1236] Overall Loss 0.179787 Objective Loss 0.179787 LR 0.000500 Time 0.026621 +2023-10-02 21:30:04,904 - Epoch: [127][ 140/ 1236] Overall Loss 0.179836 Objective Loss 0.179836 LR 0.000500 Time 0.026195 +2023-10-02 21:30:05,107 - Epoch: [127][ 150/ 1236] Overall Loss 0.179489 Objective Loss 0.179489 LR 0.000500 Time 0.025800 +2023-10-02 21:30:05,314 - Epoch: [127][ 160/ 1236] Overall Loss 0.179592 Objective Loss 0.179592 LR 0.000500 Time 0.025478 +2023-10-02 21:30:05,516 - Epoch: [127][ 170/ 1236] Overall Loss 0.181244 Objective Loss 0.181244 LR 0.000500 Time 0.025171 +2023-10-02 21:30:05,724 - Epoch: [127][ 180/ 1236] Overall Loss 0.182339 Objective Loss 0.182339 LR 0.000500 Time 0.024921 +2023-10-02 21:30:05,926 - Epoch: [127][ 190/ 1236] Overall Loss 0.182342 Objective Loss 0.182342 LR 0.000500 Time 0.024675 +2023-10-02 21:30:06,133 - Epoch: [127][ 200/ 1236] Overall Loss 0.182449 Objective Loss 0.182449 LR 0.000500 Time 0.024475 +2023-10-02 21:30:06,336 - Epoch: [127][ 210/ 1236] Overall Loss 0.181868 Objective Loss 0.181868 LR 0.000500 Time 0.024274 +2023-10-02 21:30:06,543 - Epoch: [127][ 220/ 1236] Overall Loss 0.181877 Objective Loss 0.181877 LR 0.000500 Time 0.024110 +2023-10-02 21:30:06,746 - Epoch: [127][ 230/ 1236] Overall Loss 0.181340 Objective Loss 0.181340 LR 0.000500 Time 0.023943 +2023-10-02 21:30:06,953 - Epoch: [127][ 240/ 1236] Overall Loss 0.181284 Objective Loss 0.181284 LR 0.000500 Time 0.023807 +2023-10-02 21:30:07,156 - Epoch: [127][ 250/ 1236] Overall Loss 0.180973 Objective Loss 0.180973 LR 0.000500 Time 0.023665 +2023-10-02 21:30:07,363 - Epoch: [127][ 260/ 1236] Overall Loss 0.180350 Objective Loss 0.180350 LR 0.000500 Time 0.023551 +2023-10-02 21:30:07,566 - Epoch: [127][ 270/ 1236] Overall Loss 0.180094 Objective Loss 0.180094 LR 0.000500 Time 0.023429 +2023-10-02 21:30:07,773 - Epoch: [127][ 280/ 1236] Overall Loss 0.179761 Objective Loss 0.179761 LR 0.000500 Time 0.023331 +2023-10-02 21:30:07,976 - Epoch: [127][ 290/ 1236] Overall Loss 0.180479 Objective Loss 0.180479 LR 0.000500 Time 0.023225 +2023-10-02 21:30:08,183 - Epoch: [127][ 300/ 1236] Overall Loss 0.180407 Objective Loss 0.180407 LR 0.000500 Time 0.023141 +2023-10-02 21:30:08,386 - Epoch: [127][ 310/ 1236] Overall Loss 0.180012 Objective Loss 0.180012 LR 0.000500 Time 0.023048 +2023-10-02 21:30:08,593 - Epoch: [127][ 320/ 1236] Overall Loss 0.180371 Objective Loss 0.180371 LR 0.000500 Time 0.022973 +2023-10-02 21:30:08,796 - Epoch: [127][ 330/ 1236] Overall Loss 0.179953 Objective Loss 0.179953 LR 0.000500 Time 0.022891 +2023-10-02 21:30:09,003 - Epoch: [127][ 340/ 1236] Overall Loss 0.180047 Objective Loss 0.180047 LR 0.000500 Time 0.022827 +2023-10-02 21:30:09,207 - Epoch: [127][ 350/ 1236] Overall Loss 0.179581 Objective Loss 0.179581 LR 0.000500 Time 0.022754 +2023-10-02 21:30:09,414 - Epoch: [127][ 360/ 1236] Overall Loss 0.179598 Objective Loss 0.179598 LR 0.000500 Time 0.022697 +2023-10-02 21:30:09,617 - Epoch: [127][ 370/ 1236] Overall Loss 0.180066 Objective Loss 0.180066 LR 0.000500 Time 0.022633 +2023-10-02 21:30:09,823 - Epoch: [127][ 380/ 1236] Overall Loss 0.180396 Objective Loss 0.180396 LR 0.000500 Time 0.022578 +2023-10-02 21:30:10,025 - Epoch: [127][ 390/ 1236] Overall Loss 0.180148 Objective Loss 0.180148 LR 0.000500 Time 0.022513 +2023-10-02 21:30:10,231 - Epoch: [127][ 400/ 1236] Overall Loss 0.180604 Objective Loss 0.180604 LR 0.000500 Time 0.022465 +2023-10-02 21:30:10,433 - Epoch: [127][ 410/ 1236] Overall Loss 0.180382 Objective Loss 0.180382 LR 0.000500 Time 0.022406 +2023-10-02 21:30:10,640 - Epoch: [127][ 420/ 1236] Overall Loss 0.180504 Objective Loss 0.180504 LR 0.000500 Time 0.022363 +2023-10-02 21:30:10,844 - Epoch: [127][ 430/ 1236] Overall Loss 0.180685 Objective Loss 0.180685 LR 0.000500 Time 0.022316 +2023-10-02 21:30:11,050 - Epoch: [127][ 440/ 1236] Overall Loss 0.180739 Objective Loss 0.180739 LR 0.000500 Time 0.022276 +2023-10-02 21:30:11,255 - Epoch: [127][ 450/ 1236] Overall Loss 0.180611 Objective Loss 0.180611 LR 0.000500 Time 0.022232 +2023-10-02 21:30:11,460 - Epoch: [127][ 460/ 1236] Overall Loss 0.180085 Objective Loss 0.180085 LR 0.000500 Time 0.022194 +2023-10-02 21:30:11,665 - Epoch: [127][ 470/ 1236] Overall Loss 0.179986 Objective Loss 0.179986 LR 0.000500 Time 0.022154 +2023-10-02 21:30:11,871 - Epoch: [127][ 480/ 1236] Overall Loss 0.179759 Objective Loss 0.179759 LR 0.000500 Time 0.022121 +2023-10-02 21:30:12,074 - Epoch: [127][ 490/ 1236] Overall Loss 0.179630 Objective Loss 0.179630 LR 0.000500 Time 0.022081 +2023-10-02 21:30:12,278 - Epoch: [127][ 500/ 1236] Overall Loss 0.179666 Objective Loss 0.179666 LR 0.000500 Time 0.022048 +2023-10-02 21:30:12,486 - Epoch: [127][ 510/ 1236] Overall Loss 0.179757 Objective Loss 0.179757 LR 0.000500 Time 0.022019 +2023-10-02 21:30:12,693 - Epoch: [127][ 520/ 1236] Overall Loss 0.179825 Objective Loss 0.179825 LR 0.000500 Time 0.021994 +2023-10-02 21:30:12,903 - Epoch: [127][ 530/ 1236] Overall Loss 0.180147 Objective Loss 0.180147 LR 0.000500 Time 0.021974 +2023-10-02 21:30:13,111 - Epoch: [127][ 540/ 1236] Overall Loss 0.179759 Objective Loss 0.179759 LR 0.000500 Time 0.021950 +2023-10-02 21:30:13,323 - Epoch: [127][ 550/ 1236] Overall Loss 0.179538 Objective Loss 0.179538 LR 0.000500 Time 0.021935 +2023-10-02 21:30:13,529 - Epoch: [127][ 560/ 1236] Overall Loss 0.179166 Objective Loss 0.179166 LR 0.000500 Time 0.021912 +2023-10-02 21:30:13,738 - Epoch: [127][ 570/ 1236] Overall Loss 0.179146 Objective Loss 0.179146 LR 0.000500 Time 0.021893 +2023-10-02 21:30:13,946 - Epoch: [127][ 580/ 1236] Overall Loss 0.179029 Objective Loss 0.179029 LR 0.000500 Time 0.021871 +2023-10-02 21:30:14,156 - Epoch: [127][ 590/ 1236] Overall Loss 0.178458 Objective Loss 0.178458 LR 0.000500 Time 0.021856 +2023-10-02 21:30:14,362 - Epoch: [127][ 600/ 1236] Overall Loss 0.177935 Objective Loss 0.177935 LR 0.000500 Time 0.021834 +2023-10-02 21:30:14,572 - Epoch: [127][ 610/ 1236] Overall Loss 0.178444 Objective Loss 0.178444 LR 0.000500 Time 0.021820 +2023-10-02 21:30:14,779 - Epoch: [127][ 620/ 1236] Overall Loss 0.177868 Objective Loss 0.177868 LR 0.000500 Time 0.021801 +2023-10-02 21:30:14,993 - Epoch: [127][ 630/ 1236] Overall Loss 0.177814 Objective Loss 0.177814 LR 0.000500 Time 0.021794 +2023-10-02 21:30:15,207 - Epoch: [127][ 640/ 1236] Overall Loss 0.177763 Objective Loss 0.177763 LR 0.000500 Time 0.021786 +2023-10-02 21:30:15,422 - Epoch: [127][ 650/ 1236] Overall Loss 0.178029 Objective Loss 0.178029 LR 0.000500 Time 0.021780 +2023-10-02 21:30:15,636 - Epoch: [127][ 660/ 1236] Overall Loss 0.178300 Objective Loss 0.178300 LR 0.000500 Time 0.021775 +2023-10-02 21:30:15,851 - Epoch: [127][ 670/ 1236] Overall Loss 0.178300 Objective Loss 0.178300 LR 0.000500 Time 0.021768 +2023-10-02 21:30:16,065 - Epoch: [127][ 680/ 1236] Overall Loss 0.178498 Objective Loss 0.178498 LR 0.000500 Time 0.021760 +2023-10-02 21:30:16,280 - Epoch: [127][ 690/ 1236] Overall Loss 0.178825 Objective Loss 0.178825 LR 0.000500 Time 0.021754 +2023-10-02 21:30:16,494 - Epoch: [127][ 700/ 1236] Overall Loss 0.178944 Objective Loss 0.178944 LR 0.000500 Time 0.021748 +2023-10-02 21:30:16,710 - Epoch: [127][ 710/ 1236] Overall Loss 0.179038 Objective Loss 0.179038 LR 0.000500 Time 0.021745 +2023-10-02 21:30:16,922 - Epoch: [127][ 720/ 1236] Overall Loss 0.178862 Objective Loss 0.178862 LR 0.000500 Time 0.021738 +2023-10-02 21:30:17,136 - Epoch: [127][ 730/ 1236] Overall Loss 0.179040 Objective Loss 0.179040 LR 0.000500 Time 0.021733 +2023-10-02 21:30:17,350 - Epoch: [127][ 740/ 1236] Overall Loss 0.179046 Objective Loss 0.179046 LR 0.000500 Time 0.021726 +2023-10-02 21:30:17,564 - Epoch: [127][ 750/ 1236] Overall Loss 0.179031 Objective Loss 0.179031 LR 0.000500 Time 0.021721 +2023-10-02 21:30:17,778 - Epoch: [127][ 760/ 1236] Overall Loss 0.179204 Objective Loss 0.179204 LR 0.000500 Time 0.021716 +2023-10-02 21:30:17,994 - Epoch: [127][ 770/ 1236] Overall Loss 0.179253 Objective Loss 0.179253 LR 0.000500 Time 0.021712 +2023-10-02 21:30:18,208 - Epoch: [127][ 780/ 1236] Overall Loss 0.179314 Objective Loss 0.179314 LR 0.000500 Time 0.021708 +2023-10-02 21:30:18,423 - Epoch: [127][ 790/ 1236] Overall Loss 0.179375 Objective Loss 0.179375 LR 0.000500 Time 0.021705 +2023-10-02 21:30:18,638 - Epoch: [127][ 800/ 1236] Overall Loss 0.179515 Objective Loss 0.179515 LR 0.000500 Time 0.021701 +2023-10-02 21:30:18,853 - Epoch: [127][ 810/ 1236] Overall Loss 0.179375 Objective Loss 0.179375 LR 0.000500 Time 0.021699 +2023-10-02 21:30:19,072 - Epoch: [127][ 820/ 1236] Overall Loss 0.179320 Objective Loss 0.179320 LR 0.000500 Time 0.021700 +2023-10-02 21:30:19,294 - Epoch: [127][ 830/ 1236] Overall Loss 0.179146 Objective Loss 0.179146 LR 0.000500 Time 0.021706 +2023-10-02 21:30:19,514 - Epoch: [127][ 840/ 1236] Overall Loss 0.179086 Objective Loss 0.179086 LR 0.000500 Time 0.021710 +2023-10-02 21:30:19,739 - Epoch: [127][ 850/ 1236] Overall Loss 0.179423 Objective Loss 0.179423 LR 0.000500 Time 0.021716 +2023-10-02 21:30:19,950 - Epoch: [127][ 860/ 1236] Overall Loss 0.179549 Objective Loss 0.179549 LR 0.000500 Time 0.021708 +2023-10-02 21:30:20,163 - Epoch: [127][ 870/ 1236] Overall Loss 0.179613 Objective Loss 0.179613 LR 0.000500 Time 0.021702 +2023-10-02 21:30:20,376 - Epoch: [127][ 880/ 1236] Overall Loss 0.179722 Objective Loss 0.179722 LR 0.000500 Time 0.021696 +2023-10-02 21:30:20,590 - Epoch: [127][ 890/ 1236] Overall Loss 0.179785 Objective Loss 0.179785 LR 0.000500 Time 0.021692 +2023-10-02 21:30:20,802 - Epoch: [127][ 900/ 1236] Overall Loss 0.179740 Objective Loss 0.179740 LR 0.000500 Time 0.021686 +2023-10-02 21:30:21,016 - Epoch: [127][ 910/ 1236] Overall Loss 0.179678 Objective Loss 0.179678 LR 0.000500 Time 0.021683 +2023-10-02 21:30:21,228 - Epoch: [127][ 920/ 1236] Overall Loss 0.179477 Objective Loss 0.179477 LR 0.000500 Time 0.021677 +2023-10-02 21:30:21,443 - Epoch: [127][ 930/ 1236] Overall Loss 0.179177 Objective Loss 0.179177 LR 0.000500 Time 0.021674 +2023-10-02 21:30:21,655 - Epoch: [127][ 940/ 1236] Overall Loss 0.179144 Objective Loss 0.179144 LR 0.000500 Time 0.021669 +2023-10-02 21:30:21,870 - Epoch: [127][ 950/ 1236] Overall Loss 0.179157 Objective Loss 0.179157 LR 0.000500 Time 0.021667 +2023-10-02 21:30:22,081 - Epoch: [127][ 960/ 1236] Overall Loss 0.179229 Objective Loss 0.179229 LR 0.000500 Time 0.021661 +2023-10-02 21:30:22,294 - Epoch: [127][ 970/ 1236] Overall Loss 0.179259 Objective Loss 0.179259 LR 0.000500 Time 0.021656 +2023-10-02 21:30:22,507 - Epoch: [127][ 980/ 1236] Overall Loss 0.179568 Objective Loss 0.179568 LR 0.000500 Time 0.021651 +2023-10-02 21:30:22,722 - Epoch: [127][ 990/ 1236] Overall Loss 0.179695 Objective Loss 0.179695 LR 0.000500 Time 0.021649 +2023-10-02 21:30:22,934 - Epoch: [127][ 1000/ 1236] Overall Loss 0.179638 Objective Loss 0.179638 LR 0.000500 Time 0.021644 +2023-10-02 21:30:23,149 - Epoch: [127][ 1010/ 1236] Overall Loss 0.179712 Objective Loss 0.179712 LR 0.000500 Time 0.021643 +2023-10-02 21:30:23,361 - Epoch: [127][ 1020/ 1236] Overall Loss 0.179857 Objective Loss 0.179857 LR 0.000500 Time 0.021638 +2023-10-02 21:30:23,574 - Epoch: [127][ 1030/ 1236] Overall Loss 0.179997 Objective Loss 0.179997 LR 0.000500 Time 0.021633 +2023-10-02 21:30:23,788 - Epoch: [127][ 1040/ 1236] Overall Loss 0.180036 Objective Loss 0.180036 LR 0.000500 Time 0.021629 +2023-10-02 21:30:24,002 - Epoch: [127][ 1050/ 1236] Overall Loss 0.179978 Objective Loss 0.179978 LR 0.000500 Time 0.021627 +2023-10-02 21:30:24,214 - Epoch: [127][ 1060/ 1236] Overall Loss 0.179989 Objective Loss 0.179989 LR 0.000500 Time 0.021622 +2023-10-02 21:30:24,429 - Epoch: [127][ 1070/ 1236] Overall Loss 0.179962 Objective Loss 0.179962 LR 0.000500 Time 0.021621 +2023-10-02 21:30:24,640 - Epoch: [127][ 1080/ 1236] Overall Loss 0.179970 Objective Loss 0.179970 LR 0.000500 Time 0.021616 +2023-10-02 21:30:24,855 - Epoch: [127][ 1090/ 1236] Overall Loss 0.180262 Objective Loss 0.180262 LR 0.000500 Time 0.021614 +2023-10-02 21:30:25,067 - Epoch: [127][ 1100/ 1236] Overall Loss 0.180287 Objective Loss 0.180287 LR 0.000500 Time 0.021610 +2023-10-02 21:30:25,281 - Epoch: [127][ 1110/ 1236] Overall Loss 0.180267 Objective Loss 0.180267 LR 0.000500 Time 0.021607 +2023-10-02 21:30:25,492 - Epoch: [127][ 1120/ 1236] Overall Loss 0.180322 Objective Loss 0.180322 LR 0.000500 Time 0.021603 +2023-10-02 21:30:25,706 - Epoch: [127][ 1130/ 1236] Overall Loss 0.180221 Objective Loss 0.180221 LR 0.000500 Time 0.021601 +2023-10-02 21:30:25,918 - Epoch: [127][ 1140/ 1236] Overall Loss 0.180100 Objective Loss 0.180100 LR 0.000500 Time 0.021597 +2023-10-02 21:30:26,133 - Epoch: [127][ 1150/ 1236] Overall Loss 0.180147 Objective Loss 0.180147 LR 0.000500 Time 0.021596 +2023-10-02 21:30:26,345 - Epoch: [127][ 1160/ 1236] Overall Loss 0.180174 Objective Loss 0.180174 LR 0.000500 Time 0.021592 +2023-10-02 21:30:26,559 - Epoch: [127][ 1170/ 1236] Overall Loss 0.180217 Objective Loss 0.180217 LR 0.000500 Time 0.021590 +2023-10-02 21:30:26,773 - Epoch: [127][ 1180/ 1236] Overall Loss 0.180197 Objective Loss 0.180197 LR 0.000500 Time 0.021587 +2023-10-02 21:30:26,987 - Epoch: [127][ 1190/ 1236] Overall Loss 0.180059 Objective Loss 0.180059 LR 0.000500 Time 0.021585 +2023-10-02 21:30:27,200 - Epoch: [127][ 1200/ 1236] Overall Loss 0.179940 Objective Loss 0.179940 LR 0.000500 Time 0.021582 +2023-10-02 21:30:27,415 - Epoch: [127][ 1210/ 1236] Overall Loss 0.179871 Objective Loss 0.179871 LR 0.000500 Time 0.021580 +2023-10-02 21:30:27,627 - Epoch: [127][ 1220/ 1236] Overall Loss 0.179944 Objective Loss 0.179944 LR 0.000500 Time 0.021577 +2023-10-02 21:30:27,892 - Epoch: [127][ 1230/ 1236] Overall Loss 0.180071 Objective Loss 0.180071 LR 0.000500 Time 0.021617 +2023-10-02 21:30:28,014 - Epoch: [127][ 1236/ 1236] Overall Loss 0.179944 Objective Loss 0.179944 Top1 91.242363 Top5 98.981670 LR 0.000500 Time 0.021611 +2023-10-02 21:30:28,140 - --- validate (epoch=127)----------- +2023-10-02 21:30:28,140 - 29943 samples (256 per mini-batch) +2023-10-02 21:30:28,639 - Epoch: [127][ 10/ 117] Loss 0.281323 Top1 86.328125 Top5 98.710938 +2023-10-02 21:30:28,802 - Epoch: [127][ 20/ 117] Loss 0.300915 Top1 86.601562 Top5 98.750000 +2023-10-02 21:30:28,962 - Epoch: [127][ 30/ 117] Loss 0.314719 Top1 86.106771 Top5 98.593750 +2023-10-02 21:30:29,123 - Epoch: [127][ 40/ 117] Loss 0.302703 Top1 86.318359 Top5 98.623047 +2023-10-02 21:30:29,283 - Epoch: [127][ 50/ 117] Loss 0.301006 Top1 86.421875 Top5 98.617188 +2023-10-02 21:30:29,438 - Epoch: [127][ 60/ 117] Loss 0.299100 Top1 86.412760 Top5 98.606771 +2023-10-02 21:30:29,590 - Epoch: [127][ 70/ 117] Loss 0.301526 Top1 86.227679 Top5 98.593750 +2023-10-02 21:30:29,742 - Epoch: [127][ 80/ 117] Loss 0.299952 Top1 86.269531 Top5 98.623047 +2023-10-02 21:30:29,893 - Epoch: [127][ 90/ 117] Loss 0.302934 Top1 86.236979 Top5 98.615451 +2023-10-02 21:30:30,046 - Epoch: [127][ 100/ 117] Loss 0.305841 Top1 86.210938 Top5 98.589844 +2023-10-02 21:30:30,204 - Epoch: [127][ 110/ 117] Loss 0.303482 Top1 86.250000 Top5 98.565341 +2023-10-02 21:30:30,293 - Epoch: [127][ 117/ 117] Loss 0.302619 Top1 86.247203 Top5 98.573957 +2023-10-02 21:30:30,424 - ==> Top1: 86.247 Top5: 98.574 Loss: 0.303 + +2023-10-02 21:30:30,425 - ==> Confusion: +[[ 945 0 1 0 4 2 0 0 8 48 1 1 1 4 4 1 7 0 1 0 22] + [ 0 1042 0 0 9 27 1 22 1 1 0 0 2 0 1 2 3 0 12 1 7] + [ 1 0 974 6 2 0 14 5 0 1 2 0 9 2 1 5 1 1 15 4 13] + [ 1 2 9 983 0 2 0 1 1 0 2 0 10 3 24 2 3 5 18 1 22] + [ 31 3 0 0 974 2 0 0 2 6 0 0 2 3 9 3 9 0 0 2 4] + [ 5 39 1 1 3 999 0 19 2 2 1 8 0 7 5 1 3 0 5 1 14] + [ 1 3 26 1 0 1 1119 5 0 0 3 2 0 1 0 4 1 2 2 9 11] + [ 1 12 11 0 7 26 4 1068 1 2 3 4 4 5 2 2 1 1 45 10 9] + [ 13 6 0 0 1 3 0 0 968 38 6 2 4 13 25 0 1 2 3 1 3] + [ 104 1 0 0 7 5 0 0 22 932 1 0 0 24 10 1 1 2 0 0 9] + [ 3 5 11 10 0 4 3 6 13 0 951 0 1 15 5 0 5 2 6 1 12] + [ 0 0 1 0 0 6 0 2 0 0 0 990 11 7 0 1 0 12 0 3 2] + [ 0 1 2 1 1 1 0 1 1 1 0 36 969 3 1 5 2 17 4 3 19] + [ 1 0 1 0 3 9 0 0 8 5 4 10 1 1053 6 2 0 0 0 2 14] + [ 16 0 4 17 6 0 0 0 13 3 2 0 3 3 1016 0 0 1 10 0 7] + [ 0 0 1 1 5 0 0 0 0 1 0 8 10 0 0 1063 20 9 2 5 9] + [ 0 12 0 0 5 7 0 0 0 0 0 3 0 1 5 7 1106 0 2 4 9] + [ 0 0 1 2 0 0 1 0 0 0 0 9 23 1 3 6 0 988 0 0 4] + [ 1 6 2 12 2 2 0 17 6 1 3 0 1 0 14 0 0 0 989 0 12] + [ 0 2 1 2 0 4 4 7 0 1 0 16 6 2 0 0 12 0 1 1083 11] + [ 130 152 89 83 68 141 34 88 76 67 124 98 305 244 133 43 114 49 116 138 5613]] + +2023-10-02 21:30:30,426 - ==> Best [Top1: 86.247 Top5: 98.574 Sparsity:0.00 Params: 169472 on epoch: 127] +2023-10-02 21:30:30,426 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:30:30,440 - + +2023-10-02 21:30:30,440 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:30:31,583 - Epoch: [128][ 10/ 1236] Overall Loss 0.169792 Objective Loss 0.169792 LR 0.000500 Time 0.114223 +2023-10-02 21:30:31,790 - Epoch: [128][ 20/ 1236] Overall Loss 0.176827 Objective Loss 0.176827 LR 0.000500 Time 0.067463 +2023-10-02 21:30:31,997 - Epoch: [128][ 30/ 1236] Overall Loss 0.176562 Objective Loss 0.176562 LR 0.000500 Time 0.051848 +2023-10-02 21:30:32,204 - Epoch: [128][ 40/ 1236] Overall Loss 0.181203 Objective Loss 0.181203 LR 0.000500 Time 0.044064 +2023-10-02 21:30:32,411 - Epoch: [128][ 50/ 1236] Overall Loss 0.182320 Objective Loss 0.182320 LR 0.000500 Time 0.039374 +2023-10-02 21:30:32,618 - Epoch: [128][ 60/ 1236] Overall Loss 0.182052 Objective Loss 0.182052 LR 0.000500 Time 0.036269 +2023-10-02 21:30:32,825 - Epoch: [128][ 70/ 1236] Overall Loss 0.179187 Objective Loss 0.179187 LR 0.000500 Time 0.034034 +2023-10-02 21:30:33,039 - Epoch: [128][ 80/ 1236] Overall Loss 0.180495 Objective Loss 0.180495 LR 0.000500 Time 0.032457 +2023-10-02 21:30:33,252 - Epoch: [128][ 90/ 1236] Overall Loss 0.182692 Objective Loss 0.182692 LR 0.000500 Time 0.031213 +2023-10-02 21:30:33,471 - Epoch: [128][ 100/ 1236] Overall Loss 0.182469 Objective Loss 0.182469 LR 0.000500 Time 0.030271 +2023-10-02 21:30:33,683 - Epoch: [128][ 110/ 1236] Overall Loss 0.182694 Objective Loss 0.182694 LR 0.000500 Time 0.029451 +2023-10-02 21:30:33,902 - Epoch: [128][ 120/ 1236] Overall Loss 0.180606 Objective Loss 0.180606 LR 0.000500 Time 0.028815 +2023-10-02 21:30:34,115 - Epoch: [128][ 130/ 1236] Overall Loss 0.178357 Objective Loss 0.178357 LR 0.000500 Time 0.028233 +2023-10-02 21:30:34,333 - Epoch: [128][ 140/ 1236] Overall Loss 0.178455 Objective Loss 0.178455 LR 0.000500 Time 0.027772 +2023-10-02 21:30:34,546 - Epoch: [128][ 150/ 1236] Overall Loss 0.177814 Objective Loss 0.177814 LR 0.000500 Time 0.027339 +2023-10-02 21:30:34,765 - Epoch: [128][ 160/ 1236] Overall Loss 0.176797 Objective Loss 0.176797 LR 0.000500 Time 0.026993 +2023-10-02 21:30:34,978 - Epoch: [128][ 170/ 1236] Overall Loss 0.177620 Objective Loss 0.177620 LR 0.000500 Time 0.026657 +2023-10-02 21:30:35,196 - Epoch: [128][ 180/ 1236] Overall Loss 0.177460 Objective Loss 0.177460 LR 0.000500 Time 0.026388 +2023-10-02 21:30:35,410 - Epoch: [128][ 190/ 1236] Overall Loss 0.178581 Objective Loss 0.178581 LR 0.000500 Time 0.026120 +2023-10-02 21:30:35,628 - Epoch: [128][ 200/ 1236] Overall Loss 0.179327 Objective Loss 0.179327 LR 0.000500 Time 0.025905 +2023-10-02 21:30:35,841 - Epoch: [128][ 210/ 1236] Overall Loss 0.179902 Objective Loss 0.179902 LR 0.000500 Time 0.025684 +2023-10-02 21:30:36,060 - Epoch: [128][ 220/ 1236] Overall Loss 0.179238 Objective Loss 0.179238 LR 0.000500 Time 0.025509 +2023-10-02 21:30:36,273 - Epoch: [128][ 230/ 1236] Overall Loss 0.178853 Objective Loss 0.178853 LR 0.000500 Time 0.025324 +2023-10-02 21:30:36,491 - Epoch: [128][ 240/ 1236] Overall Loss 0.180217 Objective Loss 0.180217 LR 0.000500 Time 0.025178 +2023-10-02 21:30:36,705 - Epoch: [128][ 250/ 1236] Overall Loss 0.180363 Objective Loss 0.180363 LR 0.000500 Time 0.025023 +2023-10-02 21:30:36,923 - Epoch: [128][ 260/ 1236] Overall Loss 0.180385 Objective Loss 0.180385 LR 0.000500 Time 0.024899 +2023-10-02 21:30:37,136 - Epoch: [128][ 270/ 1236] Overall Loss 0.180981 Objective Loss 0.180981 LR 0.000500 Time 0.024765 +2023-10-02 21:30:37,347 - Epoch: [128][ 280/ 1236] Overall Loss 0.181561 Objective Loss 0.181561 LR 0.000500 Time 0.024633 +2023-10-02 21:30:37,554 - Epoch: [128][ 290/ 1236] Overall Loss 0.180994 Objective Loss 0.180994 LR 0.000500 Time 0.024491 +2023-10-02 21:30:37,762 - Epoch: [128][ 300/ 1236] Overall Loss 0.181107 Objective Loss 0.181107 LR 0.000500 Time 0.024366 +2023-10-02 21:30:37,969 - Epoch: [128][ 310/ 1236] Overall Loss 0.181462 Objective Loss 0.181462 LR 0.000500 Time 0.024246 +2023-10-02 21:30:38,176 - Epoch: [128][ 320/ 1236] Overall Loss 0.180685 Objective Loss 0.180685 LR 0.000500 Time 0.024136 +2023-10-02 21:30:38,383 - Epoch: [128][ 330/ 1236] Overall Loss 0.180800 Objective Loss 0.180800 LR 0.000500 Time 0.024027 +2023-10-02 21:30:38,591 - Epoch: [128][ 340/ 1236] Overall Loss 0.180114 Objective Loss 0.180114 LR 0.000500 Time 0.023930 +2023-10-02 21:30:38,797 - Epoch: [128][ 350/ 1236] Overall Loss 0.180399 Objective Loss 0.180399 LR 0.000500 Time 0.023832 +2023-10-02 21:30:39,005 - Epoch: [128][ 360/ 1236] Overall Loss 0.180398 Objective Loss 0.180398 LR 0.000500 Time 0.023747 +2023-10-02 21:30:39,212 - Epoch: [128][ 370/ 1236] Overall Loss 0.179945 Objective Loss 0.179945 LR 0.000500 Time 0.023664 +2023-10-02 21:30:39,420 - Epoch: [128][ 380/ 1236] Overall Loss 0.179342 Objective Loss 0.179342 LR 0.000500 Time 0.023587 +2023-10-02 21:30:39,627 - Epoch: [128][ 390/ 1236] Overall Loss 0.179551 Objective Loss 0.179551 LR 0.000500 Time 0.023513 +2023-10-02 21:30:39,835 - Epoch: [128][ 400/ 1236] Overall Loss 0.179271 Objective Loss 0.179271 LR 0.000500 Time 0.023444 +2023-10-02 21:30:40,042 - Epoch: [128][ 410/ 1236] Overall Loss 0.179269 Objective Loss 0.179269 LR 0.000500 Time 0.023375 +2023-10-02 21:30:40,250 - Epoch: [128][ 420/ 1236] Overall Loss 0.179814 Objective Loss 0.179814 LR 0.000500 Time 0.023314 +2023-10-02 21:30:40,458 - Epoch: [128][ 430/ 1236] Overall Loss 0.180014 Objective Loss 0.180014 LR 0.000500 Time 0.023253 +2023-10-02 21:30:40,665 - Epoch: [128][ 440/ 1236] Overall Loss 0.179989 Objective Loss 0.179989 LR 0.000500 Time 0.023196 +2023-10-02 21:30:40,872 - Epoch: [128][ 450/ 1236] Overall Loss 0.179810 Objective Loss 0.179810 LR 0.000500 Time 0.023137 +2023-10-02 21:30:41,080 - Epoch: [128][ 460/ 1236] Overall Loss 0.179887 Objective Loss 0.179887 LR 0.000500 Time 0.023085 +2023-10-02 21:30:41,287 - Epoch: [128][ 470/ 1236] Overall Loss 0.179707 Objective Loss 0.179707 LR 0.000500 Time 0.023034 +2023-10-02 21:30:41,495 - Epoch: [128][ 480/ 1236] Overall Loss 0.179842 Objective Loss 0.179842 LR 0.000500 Time 0.022986 +2023-10-02 21:30:41,702 - Epoch: [128][ 490/ 1236] Overall Loss 0.180014 Objective Loss 0.180014 LR 0.000500 Time 0.022939 +2023-10-02 21:30:41,910 - Epoch: [128][ 500/ 1236] Overall Loss 0.180439 Objective Loss 0.180439 LR 0.000500 Time 0.022896 +2023-10-02 21:30:42,117 - Epoch: [128][ 510/ 1236] Overall Loss 0.180247 Objective Loss 0.180247 LR 0.000500 Time 0.022851 +2023-10-02 21:30:42,325 - Epoch: [128][ 520/ 1236] Overall Loss 0.180573 Objective Loss 0.180573 LR 0.000500 Time 0.022812 +2023-10-02 21:30:42,531 - Epoch: [128][ 530/ 1236] Overall Loss 0.180338 Objective Loss 0.180338 LR 0.000500 Time 0.022769 +2023-10-02 21:30:42,739 - Epoch: [128][ 540/ 1236] Overall Loss 0.179845 Objective Loss 0.179845 LR 0.000500 Time 0.022732 +2023-10-02 21:30:42,946 - Epoch: [128][ 550/ 1236] Overall Loss 0.179529 Objective Loss 0.179529 LR 0.000500 Time 0.022693 +2023-10-02 21:30:43,154 - Epoch: [128][ 560/ 1236] Overall Loss 0.179549 Objective Loss 0.179549 LR 0.000500 Time 0.022658 +2023-10-02 21:30:43,361 - Epoch: [128][ 570/ 1236] Overall Loss 0.179771 Objective Loss 0.179771 LR 0.000500 Time 0.022623 +2023-10-02 21:30:43,570 - Epoch: [128][ 580/ 1236] Overall Loss 0.179618 Objective Loss 0.179618 LR 0.000500 Time 0.022591 +2023-10-02 21:30:43,776 - Epoch: [128][ 590/ 1236] Overall Loss 0.179632 Objective Loss 0.179632 LR 0.000500 Time 0.022558 +2023-10-02 21:30:43,985 - Epoch: [128][ 600/ 1236] Overall Loss 0.179951 Objective Loss 0.179951 LR 0.000500 Time 0.022529 +2023-10-02 21:30:44,191 - Epoch: [128][ 610/ 1236] Overall Loss 0.179546 Objective Loss 0.179546 LR 0.000500 Time 0.022498 +2023-10-02 21:30:44,400 - Epoch: [128][ 620/ 1236] Overall Loss 0.179301 Objective Loss 0.179301 LR 0.000500 Time 0.022470 +2023-10-02 21:30:44,606 - Epoch: [128][ 630/ 1236] Overall Loss 0.179075 Objective Loss 0.179075 LR 0.000500 Time 0.022441 +2023-10-02 21:30:44,815 - Epoch: [128][ 640/ 1236] Overall Loss 0.178945 Objective Loss 0.178945 LR 0.000500 Time 0.022415 +2023-10-02 21:30:45,022 - Epoch: [128][ 650/ 1236] Overall Loss 0.179057 Objective Loss 0.179057 LR 0.000500 Time 0.022388 +2023-10-02 21:30:45,230 - Epoch: [128][ 660/ 1236] Overall Loss 0.178759 Objective Loss 0.178759 LR 0.000500 Time 0.022364 +2023-10-02 21:30:45,437 - Epoch: [128][ 670/ 1236] Overall Loss 0.178642 Objective Loss 0.178642 LR 0.000500 Time 0.022338 +2023-10-02 21:30:45,645 - Epoch: [128][ 680/ 1236] Overall Loss 0.178761 Objective Loss 0.178761 LR 0.000500 Time 0.022315 +2023-10-02 21:30:45,852 - Epoch: [128][ 690/ 1236] Overall Loss 0.178495 Objective Loss 0.178495 LR 0.000500 Time 0.022291 +2023-10-02 21:30:46,060 - Epoch: [128][ 700/ 1236] Overall Loss 0.178139 Objective Loss 0.178139 LR 0.000500 Time 0.022270 +2023-10-02 21:30:46,267 - Epoch: [128][ 710/ 1236] Overall Loss 0.178295 Objective Loss 0.178295 LR 0.000500 Time 0.022247 +2023-10-02 21:30:46,475 - Epoch: [128][ 720/ 1236] Overall Loss 0.178196 Objective Loss 0.178196 LR 0.000500 Time 0.022226 +2023-10-02 21:30:46,682 - Epoch: [128][ 730/ 1236] Overall Loss 0.178237 Objective Loss 0.178237 LR 0.000500 Time 0.022205 +2023-10-02 21:30:46,890 - Epoch: [128][ 740/ 1236] Overall Loss 0.177982 Objective Loss 0.177982 LR 0.000500 Time 0.022185 +2023-10-02 21:30:47,096 - Epoch: [128][ 750/ 1236] Overall Loss 0.178078 Objective Loss 0.178078 LR 0.000500 Time 0.022165 +2023-10-02 21:30:47,305 - Epoch: [128][ 760/ 1236] Overall Loss 0.177687 Objective Loss 0.177687 LR 0.000500 Time 0.022147 +2023-10-02 21:30:47,512 - Epoch: [128][ 770/ 1236] Overall Loss 0.177519 Objective Loss 0.177519 LR 0.000500 Time 0.022128 +2023-10-02 21:30:47,720 - Epoch: [128][ 780/ 1236] Overall Loss 0.177647 Objective Loss 0.177647 LR 0.000500 Time 0.022110 +2023-10-02 21:30:47,927 - Epoch: [128][ 790/ 1236] Overall Loss 0.177655 Objective Loss 0.177655 LR 0.000500 Time 0.022092 +2023-10-02 21:30:48,135 - Epoch: [128][ 800/ 1236] Overall Loss 0.177855 Objective Loss 0.177855 LR 0.000500 Time 0.022075 +2023-10-02 21:30:48,342 - Epoch: [128][ 810/ 1236] Overall Loss 0.177513 Objective Loss 0.177513 LR 0.000500 Time 0.022058 +2023-10-02 21:30:48,550 - Epoch: [128][ 820/ 1236] Overall Loss 0.177754 Objective Loss 0.177754 LR 0.000500 Time 0.022042 +2023-10-02 21:30:48,757 - Epoch: [128][ 830/ 1236] Overall Loss 0.178065 Objective Loss 0.178065 LR 0.000500 Time 0.022025 +2023-10-02 21:30:48,965 - Epoch: [128][ 840/ 1236] Overall Loss 0.178109 Objective Loss 0.178109 LR 0.000500 Time 0.022011 +2023-10-02 21:30:49,172 - Epoch: [128][ 850/ 1236] Overall Loss 0.178290 Objective Loss 0.178290 LR 0.000500 Time 0.021995 +2023-10-02 21:30:49,380 - Epoch: [128][ 860/ 1236] Overall Loss 0.178363 Objective Loss 0.178363 LR 0.000500 Time 0.021981 +2023-10-02 21:30:49,587 - Epoch: [128][ 870/ 1236] Overall Loss 0.178277 Objective Loss 0.178277 LR 0.000500 Time 0.021966 +2023-10-02 21:30:49,795 - Epoch: [128][ 880/ 1236] Overall Loss 0.178252 Objective Loss 0.178252 LR 0.000500 Time 0.021952 +2023-10-02 21:30:50,002 - Epoch: [128][ 890/ 1236] Overall Loss 0.178320 Objective Loss 0.178320 LR 0.000500 Time 0.021937 +2023-10-02 21:30:50,210 - Epoch: [128][ 900/ 1236] Overall Loss 0.178419 Objective Loss 0.178419 LR 0.000500 Time 0.021924 +2023-10-02 21:30:50,418 - Epoch: [128][ 910/ 1236] Overall Loss 0.178413 Objective Loss 0.178413 LR 0.000500 Time 0.021911 +2023-10-02 21:30:50,626 - Epoch: [128][ 920/ 1236] Overall Loss 0.178721 Objective Loss 0.178721 LR 0.000500 Time 0.021899 +2023-10-02 21:30:50,833 - Epoch: [128][ 930/ 1236] Overall Loss 0.178813 Objective Loss 0.178813 LR 0.000500 Time 0.021886 +2023-10-02 21:30:51,041 - Epoch: [128][ 940/ 1236] Overall Loss 0.178744 Objective Loss 0.178744 LR 0.000500 Time 0.021874 +2023-10-02 21:30:51,248 - Epoch: [128][ 950/ 1236] Overall Loss 0.178950 Objective Loss 0.178950 LR 0.000500 Time 0.021861 +2023-10-02 21:30:51,456 - Epoch: [128][ 960/ 1236] Overall Loss 0.179024 Objective Loss 0.179024 LR 0.000500 Time 0.021850 +2023-10-02 21:30:51,663 - Epoch: [128][ 970/ 1236] Overall Loss 0.179100 Objective Loss 0.179100 LR 0.000500 Time 0.021838 +2023-10-02 21:30:51,871 - Epoch: [128][ 980/ 1236] Overall Loss 0.179127 Objective Loss 0.179127 LR 0.000500 Time 0.021827 +2023-10-02 21:30:52,078 - Epoch: [128][ 990/ 1236] Overall Loss 0.179244 Objective Loss 0.179244 LR 0.000500 Time 0.021815 +2023-10-02 21:30:52,286 - Epoch: [128][ 1000/ 1236] Overall Loss 0.179514 Objective Loss 0.179514 LR 0.000500 Time 0.021805 +2023-10-02 21:30:52,493 - Epoch: [128][ 1010/ 1236] Overall Loss 0.179329 Objective Loss 0.179329 LR 0.000500 Time 0.021794 +2023-10-02 21:30:52,702 - Epoch: [128][ 1020/ 1236] Overall Loss 0.179182 Objective Loss 0.179182 LR 0.000500 Time 0.021784 +2023-10-02 21:30:52,909 - Epoch: [128][ 1030/ 1236] Overall Loss 0.179263 Objective Loss 0.179263 LR 0.000500 Time 0.021773 +2023-10-02 21:30:53,117 - Epoch: [128][ 1040/ 1236] Overall Loss 0.179454 Objective Loss 0.179454 LR 0.000500 Time 0.021763 +2023-10-02 21:30:53,324 - Epoch: [128][ 1050/ 1236] Overall Loss 0.179802 Objective Loss 0.179802 LR 0.000500 Time 0.021753 +2023-10-02 21:30:53,532 - Epoch: [128][ 1060/ 1236] Overall Loss 0.179792 Objective Loss 0.179792 LR 0.000500 Time 0.021744 +2023-10-02 21:30:53,739 - Epoch: [128][ 1070/ 1236] Overall Loss 0.179879 Objective Loss 0.179879 LR 0.000500 Time 0.021734 +2023-10-02 21:30:53,947 - Epoch: [128][ 1080/ 1236] Overall Loss 0.179819 Objective Loss 0.179819 LR 0.000500 Time 0.021725 +2023-10-02 21:30:54,154 - Epoch: [128][ 1090/ 1236] Overall Loss 0.180051 Objective Loss 0.180051 LR 0.000500 Time 0.021715 +2023-10-02 21:30:54,362 - Epoch: [128][ 1100/ 1236] Overall Loss 0.180124 Objective Loss 0.180124 LR 0.000500 Time 0.021706 +2023-10-02 21:30:54,569 - Epoch: [128][ 1110/ 1236] Overall Loss 0.180384 Objective Loss 0.180384 LR 0.000500 Time 0.021697 +2023-10-02 21:30:54,777 - Epoch: [128][ 1120/ 1236] Overall Loss 0.180375 Objective Loss 0.180375 LR 0.000500 Time 0.021689 +2023-10-02 21:30:54,984 - Epoch: [128][ 1130/ 1236] Overall Loss 0.180428 Objective Loss 0.180428 LR 0.000500 Time 0.021680 +2023-10-02 21:30:55,192 - Epoch: [128][ 1140/ 1236] Overall Loss 0.180454 Objective Loss 0.180454 LR 0.000500 Time 0.021672 +2023-10-02 21:30:55,399 - Epoch: [128][ 1150/ 1236] Overall Loss 0.180528 Objective Loss 0.180528 LR 0.000500 Time 0.021663 +2023-10-02 21:30:55,607 - Epoch: [128][ 1160/ 1236] Overall Loss 0.180592 Objective Loss 0.180592 LR 0.000500 Time 0.021655 +2023-10-02 21:30:55,814 - Epoch: [128][ 1170/ 1236] Overall Loss 0.180651 Objective Loss 0.180651 LR 0.000500 Time 0.021647 +2023-10-02 21:30:56,022 - Epoch: [128][ 1180/ 1236] Overall Loss 0.180510 Objective Loss 0.180510 LR 0.000500 Time 0.021639 +2023-10-02 21:30:56,229 - Epoch: [128][ 1190/ 1236] Overall Loss 0.180588 Objective Loss 0.180588 LR 0.000500 Time 0.021631 +2023-10-02 21:30:56,437 - Epoch: [128][ 1200/ 1236] Overall Loss 0.180651 Objective Loss 0.180651 LR 0.000500 Time 0.021624 +2023-10-02 21:30:56,644 - Epoch: [128][ 1210/ 1236] Overall Loss 0.180724 Objective Loss 0.180724 LR 0.000500 Time 0.021616 +2023-10-02 21:30:56,852 - Epoch: [128][ 1220/ 1236] Overall Loss 0.180619 Objective Loss 0.180619 LR 0.000500 Time 0.021609 +2023-10-02 21:30:57,114 - Epoch: [128][ 1230/ 1236] Overall Loss 0.180476 Objective Loss 0.180476 LR 0.000500 Time 0.021644 +2023-10-02 21:30:57,236 - Epoch: [128][ 1236/ 1236] Overall Loss 0.180498 Objective Loss 0.180498 Top1 89.816701 Top5 98.778004 LR 0.000500 Time 0.021638 +2023-10-02 21:30:57,372 - --- validate (epoch=128)----------- +2023-10-02 21:30:57,372 - 29943 samples (256 per mini-batch) +2023-10-02 21:30:57,872 - Epoch: [128][ 10/ 117] Loss 0.316221 Top1 85.390625 Top5 98.203125 +2023-10-02 21:30:58,025 - Epoch: [128][ 20/ 117] Loss 0.307786 Top1 85.429688 Top5 98.320312 +2023-10-02 21:30:58,177 - Epoch: [128][ 30/ 117] Loss 0.322913 Top1 85.299479 Top5 98.281250 +2023-10-02 21:30:58,329 - Epoch: [128][ 40/ 117] Loss 0.310231 Top1 85.722656 Top5 98.291016 +2023-10-02 21:30:58,480 - Epoch: [128][ 50/ 117] Loss 0.309896 Top1 85.492188 Top5 98.265625 +2023-10-02 21:30:58,631 - Epoch: [128][ 60/ 117] Loss 0.308059 Top1 85.397135 Top5 98.300781 +2023-10-02 21:30:58,781 - Epoch: [128][ 70/ 117] Loss 0.306856 Top1 85.357143 Top5 98.353795 +2023-10-02 21:30:58,932 - Epoch: [128][ 80/ 117] Loss 0.307443 Top1 85.390625 Top5 98.359375 +2023-10-02 21:30:59,083 - Epoch: [128][ 90/ 117] Loss 0.307885 Top1 85.425347 Top5 98.376736 +2023-10-02 21:30:59,234 - Epoch: [128][ 100/ 117] Loss 0.304937 Top1 85.488281 Top5 98.414062 +2023-10-02 21:30:59,392 - Epoch: [128][ 110/ 117] Loss 0.303399 Top1 85.454545 Top5 98.448153 +2023-10-02 21:30:59,481 - Epoch: [128][ 117/ 117] Loss 0.301665 Top1 85.562569 Top5 98.447049 +2023-10-02 21:30:59,627 - ==> Top1: 85.563 Top5: 98.447 Loss: 0.302 + +2023-10-02 21:30:59,627 - ==> Confusion: +[[ 970 0 5 2 4 1 0 0 6 36 1 1 1 1 3 3 4 1 0 0 11] + [ 0 1034 2 0 5 34 1 24 2 1 1 0 2 0 0 4 1 0 11 1 8] + [ 3 0 990 6 0 0 10 8 0 1 0 0 9 1 0 6 1 1 8 2 10] + [ 2 1 16 998 0 0 1 1 2 0 1 0 1 0 13 6 1 7 15 2 22] + [ 34 8 1 0 957 8 0 0 0 7 2 0 1 4 6 5 9 0 1 1 6] + [ 4 29 1 2 1 1019 0 15 1 2 2 6 4 8 4 0 1 0 3 5 9] + [ 0 4 44 0 0 2 1108 6 0 0 3 0 0 1 0 6 0 1 3 10 3] + [ 6 10 13 1 2 27 5 1071 1 1 1 2 5 4 2 0 2 2 47 10 6] + [ 24 2 0 3 2 5 0 2 961 40 13 4 3 12 13 0 3 1 1 0 0] + [ 127 3 0 1 7 6 0 0 22 912 1 0 0 21 4 4 2 1 0 2 6] + [ 4 2 17 10 0 1 2 5 8 1 964 1 1 13 2 0 2 1 9 2 8] + [ 0 0 1 0 0 12 0 1 0 0 0 970 20 5 0 4 1 14 0 5 2] + [ 0 1 3 2 0 4 0 0 0 0 3 38 974 0 2 8 2 15 1 6 9] + [ 4 0 3 0 2 9 0 0 8 10 4 5 0 1054 5 1 0 0 0 4 10] + [ 18 1 4 27 4 1 0 0 18 8 1 0 3 3 984 0 2 4 12 0 11] + [ 0 0 3 0 5 0 0 0 0 0 0 6 8 0 1 1069 15 12 3 6 6] + [ 1 16 1 0 5 7 0 0 0 0 0 6 0 1 5 7 1097 0 3 2 10] + [ 0 1 1 2 0 0 2 0 1 1 0 8 20 1 2 5 1 990 0 1 2] + [ 2 6 3 13 1 0 1 21 4 0 2 0 1 0 11 0 0 2 987 0 14] + [ 0 1 6 0 0 2 8 4 0 1 1 9 4 1 0 2 8 2 0 1097 6] + [ 164 120 153 96 66 178 31 83 69 70 139 110 363 223 99 56 105 55 137 174 5414]] + +2023-10-02 21:30:59,629 - ==> Best [Top1: 86.247 Top5: 98.574 Sparsity:0.00 Params: 169472 on epoch: 127] +2023-10-02 21:30:59,629 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:30:59,635 - + +2023-10-02 21:30:59,635 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:31:00,657 - Epoch: [129][ 10/ 1236] Overall Loss 0.162417 Objective Loss 0.162417 LR 0.000500 Time 0.102158 +2023-10-02 21:31:00,865 - Epoch: [129][ 20/ 1236] Overall Loss 0.181872 Objective Loss 0.181872 LR 0.000500 Time 0.061435 +2023-10-02 21:31:01,071 - Epoch: [129][ 30/ 1236] Overall Loss 0.193539 Objective Loss 0.193539 LR 0.000500 Time 0.047815 +2023-10-02 21:31:01,277 - Epoch: [129][ 40/ 1236] Overall Loss 0.190058 Objective Loss 0.190058 LR 0.000500 Time 0.041014 +2023-10-02 21:31:01,484 - Epoch: [129][ 50/ 1236] Overall Loss 0.194382 Objective Loss 0.194382 LR 0.000500 Time 0.036933 +2023-10-02 21:31:01,690 - Epoch: [129][ 60/ 1236] Overall Loss 0.188935 Objective Loss 0.188935 LR 0.000500 Time 0.034216 +2023-10-02 21:31:01,896 - Epoch: [129][ 70/ 1236] Overall Loss 0.186848 Objective Loss 0.186848 LR 0.000500 Time 0.032260 +2023-10-02 21:31:02,102 - Epoch: [129][ 80/ 1236] Overall Loss 0.183109 Objective Loss 0.183109 LR 0.000500 Time 0.030805 +2023-10-02 21:31:02,307 - Epoch: [129][ 90/ 1236] Overall Loss 0.183049 Objective Loss 0.183049 LR 0.000500 Time 0.029645 +2023-10-02 21:31:02,514 - Epoch: [129][ 100/ 1236] Overall Loss 0.179575 Objective Loss 0.179575 LR 0.000500 Time 0.028748 +2023-10-02 21:31:02,720 - Epoch: [129][ 110/ 1236] Overall Loss 0.179384 Objective Loss 0.179384 LR 0.000500 Time 0.027986 +2023-10-02 21:31:02,925 - Epoch: [129][ 120/ 1236] Overall Loss 0.182085 Objective Loss 0.182085 LR 0.000500 Time 0.027365 +2023-10-02 21:31:03,131 - Epoch: [129][ 130/ 1236] Overall Loss 0.182767 Objective Loss 0.182767 LR 0.000500 Time 0.026837 +2023-10-02 21:31:03,338 - Epoch: [129][ 140/ 1236] Overall Loss 0.182914 Objective Loss 0.182914 LR 0.000500 Time 0.026396 +2023-10-02 21:31:03,543 - Epoch: [129][ 150/ 1236] Overall Loss 0.182873 Objective Loss 0.182873 LR 0.000500 Time 0.025996 +2023-10-02 21:31:03,750 - Epoch: [129][ 160/ 1236] Overall Loss 0.182478 Objective Loss 0.182478 LR 0.000500 Time 0.025662 +2023-10-02 21:31:03,955 - Epoch: [129][ 170/ 1236] Overall Loss 0.183186 Objective Loss 0.183186 LR 0.000500 Time 0.025352 +2023-10-02 21:31:04,162 - Epoch: [129][ 180/ 1236] Overall Loss 0.182277 Objective Loss 0.182277 LR 0.000500 Time 0.025092 +2023-10-02 21:31:04,368 - Epoch: [129][ 190/ 1236] Overall Loss 0.183444 Objective Loss 0.183444 LR 0.000500 Time 0.024844 +2023-10-02 21:31:04,576 - Epoch: [129][ 200/ 1236] Overall Loss 0.182348 Objective Loss 0.182348 LR 0.000500 Time 0.024643 +2023-10-02 21:31:04,780 - Epoch: [129][ 210/ 1236] Overall Loss 0.183886 Objective Loss 0.183886 LR 0.000500 Time 0.024440 +2023-10-02 21:31:04,988 - Epoch: [129][ 220/ 1236] Overall Loss 0.184247 Objective Loss 0.184247 LR 0.000500 Time 0.024270 +2023-10-02 21:31:05,193 - Epoch: [129][ 230/ 1236] Overall Loss 0.184886 Objective Loss 0.184886 LR 0.000500 Time 0.024101 +2023-10-02 21:31:05,400 - Epoch: [129][ 240/ 1236] Overall Loss 0.185425 Objective Loss 0.185425 LR 0.000500 Time 0.023959 +2023-10-02 21:31:05,606 - Epoch: [129][ 250/ 1236] Overall Loss 0.185053 Objective Loss 0.185053 LR 0.000500 Time 0.023816 +2023-10-02 21:31:05,814 - Epoch: [129][ 260/ 1236] Overall Loss 0.185496 Objective Loss 0.185496 LR 0.000500 Time 0.023702 +2023-10-02 21:31:06,018 - Epoch: [129][ 270/ 1236] Overall Loss 0.185293 Objective Loss 0.185293 LR 0.000500 Time 0.023579 +2023-10-02 21:31:06,225 - Epoch: [129][ 280/ 1236] Overall Loss 0.185820 Objective Loss 0.185820 LR 0.000500 Time 0.023475 +2023-10-02 21:31:06,431 - Epoch: [129][ 290/ 1236] Overall Loss 0.185816 Objective Loss 0.185816 LR 0.000500 Time 0.023369 +2023-10-02 21:31:06,639 - Epoch: [129][ 300/ 1236] Overall Loss 0.185815 Objective Loss 0.185815 LR 0.000500 Time 0.023284 +2023-10-02 21:31:06,843 - Epoch: [129][ 310/ 1236] Overall Loss 0.184896 Objective Loss 0.184896 LR 0.000500 Time 0.023190 +2023-10-02 21:31:07,051 - Epoch: [129][ 320/ 1236] Overall Loss 0.184090 Objective Loss 0.184090 LR 0.000500 Time 0.023112 +2023-10-02 21:31:07,256 - Epoch: [129][ 330/ 1236] Overall Loss 0.184301 Objective Loss 0.184301 LR 0.000500 Time 0.023030 +2023-10-02 21:31:07,463 - Epoch: [129][ 340/ 1236] Overall Loss 0.185049 Objective Loss 0.185049 LR 0.000500 Time 0.022959 +2023-10-02 21:31:07,667 - Epoch: [129][ 350/ 1236] Overall Loss 0.184145 Objective Loss 0.184145 LR 0.000500 Time 0.022887 +2023-10-02 21:31:07,876 - Epoch: [129][ 360/ 1236] Overall Loss 0.183759 Objective Loss 0.183759 LR 0.000500 Time 0.022830 +2023-10-02 21:31:08,080 - Epoch: [129][ 370/ 1236] Overall Loss 0.183322 Objective Loss 0.183322 LR 0.000500 Time 0.022765 +2023-10-02 21:31:08,286 - Epoch: [129][ 380/ 1236] Overall Loss 0.182535 Objective Loss 0.182535 LR 0.000500 Time 0.022706 +2023-10-02 21:31:08,490 - Epoch: [129][ 390/ 1236] Overall Loss 0.182341 Objective Loss 0.182341 LR 0.000500 Time 0.022647 +2023-10-02 21:31:08,697 - Epoch: [129][ 400/ 1236] Overall Loss 0.181926 Objective Loss 0.181926 LR 0.000500 Time 0.022598 +2023-10-02 21:31:08,903 - Epoch: [129][ 410/ 1236] Overall Loss 0.181149 Objective Loss 0.181149 LR 0.000500 Time 0.022545 +2023-10-02 21:31:09,110 - Epoch: [129][ 420/ 1236] Overall Loss 0.180768 Objective Loss 0.180768 LR 0.000500 Time 0.022500 +2023-10-02 21:31:09,316 - Epoch: [129][ 430/ 1236] Overall Loss 0.180698 Objective Loss 0.180698 LR 0.000500 Time 0.022454 +2023-10-02 21:31:09,523 - Epoch: [129][ 440/ 1236] Overall Loss 0.181146 Objective Loss 0.181146 LR 0.000500 Time 0.022413 +2023-10-02 21:31:09,728 - Epoch: [129][ 450/ 1236] Overall Loss 0.181148 Objective Loss 0.181148 LR 0.000500 Time 0.022372 +2023-10-02 21:31:09,934 - Epoch: [129][ 460/ 1236] Overall Loss 0.181129 Objective Loss 0.181129 LR 0.000500 Time 0.022333 +2023-10-02 21:31:10,139 - Epoch: [129][ 470/ 1236] Overall Loss 0.180425 Objective Loss 0.180425 LR 0.000500 Time 0.022292 +2023-10-02 21:31:10,345 - Epoch: [129][ 480/ 1236] Overall Loss 0.180368 Objective Loss 0.180368 LR 0.000500 Time 0.022257 +2023-10-02 21:31:10,550 - Epoch: [129][ 490/ 1236] Overall Loss 0.180392 Objective Loss 0.180392 LR 0.000500 Time 0.022218 +2023-10-02 21:31:10,757 - Epoch: [129][ 500/ 1236] Overall Loss 0.180484 Objective Loss 0.180484 LR 0.000500 Time 0.022186 +2023-10-02 21:31:10,961 - Epoch: [129][ 510/ 1236] Overall Loss 0.180164 Objective Loss 0.180164 LR 0.000500 Time 0.022150 +2023-10-02 21:31:11,167 - Epoch: [129][ 520/ 1236] Overall Loss 0.179657 Objective Loss 0.179657 LR 0.000500 Time 0.022118 +2023-10-02 21:31:11,373 - Epoch: [129][ 530/ 1236] Overall Loss 0.179594 Objective Loss 0.179594 LR 0.000500 Time 0.022088 +2023-10-02 21:31:11,579 - Epoch: [129][ 540/ 1236] Overall Loss 0.179640 Objective Loss 0.179640 LR 0.000500 Time 0.022061 +2023-10-02 21:31:11,785 - Epoch: [129][ 550/ 1236] Overall Loss 0.179465 Objective Loss 0.179465 LR 0.000500 Time 0.022033 +2023-10-02 21:31:11,992 - Epoch: [129][ 560/ 1236] Overall Loss 0.179587 Objective Loss 0.179587 LR 0.000500 Time 0.022009 +2023-10-02 21:31:12,197 - Epoch: [129][ 570/ 1236] Overall Loss 0.179639 Objective Loss 0.179639 LR 0.000500 Time 0.021983 +2023-10-02 21:31:12,403 - Epoch: [129][ 580/ 1236] Overall Loss 0.179568 Objective Loss 0.179568 LR 0.000500 Time 0.021958 +2023-10-02 21:31:12,609 - Epoch: [129][ 590/ 1236] Overall Loss 0.179312 Objective Loss 0.179312 LR 0.000500 Time 0.021934 +2023-10-02 21:31:12,815 - Epoch: [129][ 600/ 1236] Overall Loss 0.179325 Objective Loss 0.179325 LR 0.000500 Time 0.021911 +2023-10-02 21:31:13,020 - Epoch: [129][ 610/ 1236] Overall Loss 0.179280 Objective Loss 0.179280 LR 0.000500 Time 0.021889 +2023-10-02 21:31:13,227 - Epoch: [129][ 620/ 1236] Overall Loss 0.179124 Objective Loss 0.179124 LR 0.000500 Time 0.021868 +2023-10-02 21:31:13,432 - Epoch: [129][ 630/ 1236] Overall Loss 0.179159 Objective Loss 0.179159 LR 0.000500 Time 0.021847 +2023-10-02 21:31:13,638 - Epoch: [129][ 640/ 1236] Overall Loss 0.179425 Objective Loss 0.179425 LR 0.000500 Time 0.021826 +2023-10-02 21:31:13,842 - Epoch: [129][ 650/ 1236] Overall Loss 0.179536 Objective Loss 0.179536 LR 0.000500 Time 0.021805 +2023-10-02 21:31:14,049 - Epoch: [129][ 660/ 1236] Overall Loss 0.179424 Objective Loss 0.179424 LR 0.000500 Time 0.021787 +2023-10-02 21:31:14,254 - Epoch: [129][ 670/ 1236] Overall Loss 0.179507 Objective Loss 0.179507 LR 0.000500 Time 0.021765 +2023-10-02 21:31:14,460 - Epoch: [129][ 680/ 1236] Overall Loss 0.179255 Objective Loss 0.179255 LR 0.000500 Time 0.021748 +2023-10-02 21:31:14,665 - Epoch: [129][ 690/ 1236] Overall Loss 0.179083 Objective Loss 0.179083 LR 0.000500 Time 0.021729 +2023-10-02 21:31:14,872 - Epoch: [129][ 700/ 1236] Overall Loss 0.178995 Objective Loss 0.178995 LR 0.000500 Time 0.021714 +2023-10-02 21:31:15,077 - Epoch: [129][ 710/ 1236] Overall Loss 0.179037 Objective Loss 0.179037 LR 0.000500 Time 0.021697 +2023-10-02 21:31:15,282 - Epoch: [129][ 720/ 1236] Overall Loss 0.179090 Objective Loss 0.179090 LR 0.000500 Time 0.021680 +2023-10-02 21:31:15,488 - Epoch: [129][ 730/ 1236] Overall Loss 0.178653 Objective Loss 0.178653 LR 0.000500 Time 0.021664 +2023-10-02 21:31:15,694 - Epoch: [129][ 740/ 1236] Overall Loss 0.178791 Objective Loss 0.178791 LR 0.000500 Time 0.021650 +2023-10-02 21:31:15,899 - Epoch: [129][ 750/ 1236] Overall Loss 0.178656 Objective Loss 0.178656 LR 0.000500 Time 0.021634 +2023-10-02 21:31:16,105 - Epoch: [129][ 760/ 1236] Overall Loss 0.178908 Objective Loss 0.178908 LR 0.000500 Time 0.021620 +2023-10-02 21:31:16,310 - Epoch: [129][ 770/ 1236] Overall Loss 0.178814 Objective Loss 0.178814 LR 0.000500 Time 0.021605 +2023-10-02 21:31:16,516 - Epoch: [129][ 780/ 1236] Overall Loss 0.178757 Objective Loss 0.178757 LR 0.000500 Time 0.021592 +2023-10-02 21:31:16,721 - Epoch: [129][ 790/ 1236] Overall Loss 0.178931 Objective Loss 0.178931 LR 0.000500 Time 0.021577 +2023-10-02 21:31:16,928 - Epoch: [129][ 800/ 1236] Overall Loss 0.179186 Objective Loss 0.179186 LR 0.000500 Time 0.021565 +2023-10-02 21:31:17,133 - Epoch: [129][ 810/ 1236] Overall Loss 0.179359 Objective Loss 0.179359 LR 0.000500 Time 0.021552 +2023-10-02 21:31:17,339 - Epoch: [129][ 820/ 1236] Overall Loss 0.179489 Objective Loss 0.179489 LR 0.000500 Time 0.021539 +2023-10-02 21:31:17,545 - Epoch: [129][ 830/ 1236] Overall Loss 0.179567 Objective Loss 0.179567 LR 0.000500 Time 0.021527 +2023-10-02 21:31:17,750 - Epoch: [129][ 840/ 1236] Overall Loss 0.179605 Objective Loss 0.179605 LR 0.000500 Time 0.021515 +2023-10-02 21:31:17,955 - Epoch: [129][ 850/ 1236] Overall Loss 0.179423 Objective Loss 0.179423 LR 0.000500 Time 0.021502 +2023-10-02 21:31:18,162 - Epoch: [129][ 860/ 1236] Overall Loss 0.179543 Objective Loss 0.179543 LR 0.000500 Time 0.021493 +2023-10-02 21:31:18,366 - Epoch: [129][ 870/ 1236] Overall Loss 0.179455 Objective Loss 0.179455 LR 0.000500 Time 0.021480 +2023-10-02 21:31:18,570 - Epoch: [129][ 880/ 1236] Overall Loss 0.179560 Objective Loss 0.179560 LR 0.000500 Time 0.021468 +2023-10-02 21:31:18,775 - Epoch: [129][ 890/ 1236] Overall Loss 0.179246 Objective Loss 0.179246 LR 0.000500 Time 0.021456 +2023-10-02 21:31:18,982 - Epoch: [129][ 900/ 1236] Overall Loss 0.179006 Objective Loss 0.179006 LR 0.000500 Time 0.021448 +2023-10-02 21:31:19,187 - Epoch: [129][ 910/ 1236] Overall Loss 0.178948 Objective Loss 0.178948 LR 0.000500 Time 0.021435 +2023-10-02 21:31:19,393 - Epoch: [129][ 920/ 1236] Overall Loss 0.178880 Objective Loss 0.178880 LR 0.000500 Time 0.021427 +2023-10-02 21:31:19,598 - Epoch: [129][ 930/ 1236] Overall Loss 0.178937 Objective Loss 0.178937 LR 0.000500 Time 0.021415 +2023-10-02 21:31:19,804 - Epoch: [129][ 940/ 1236] Overall Loss 0.178950 Objective Loss 0.178950 LR 0.000500 Time 0.021406 +2023-10-02 21:31:20,010 - Epoch: [129][ 950/ 1236] Overall Loss 0.178756 Objective Loss 0.178756 LR 0.000500 Time 0.021396 +2023-10-02 21:31:20,214 - Epoch: [129][ 960/ 1236] Overall Loss 0.178703 Objective Loss 0.178703 LR 0.000500 Time 0.021386 +2023-10-02 21:31:20,419 - Epoch: [129][ 970/ 1236] Overall Loss 0.178525 Objective Loss 0.178525 LR 0.000500 Time 0.021377 +2023-10-02 21:31:20,625 - Epoch: [129][ 980/ 1236] Overall Loss 0.178605 Objective Loss 0.178605 LR 0.000500 Time 0.021369 +2023-10-02 21:31:20,831 - Epoch: [129][ 990/ 1236] Overall Loss 0.178448 Objective Loss 0.178448 LR 0.000500 Time 0.021361 +2023-10-02 21:31:21,038 - Epoch: [129][ 1000/ 1236] Overall Loss 0.178521 Objective Loss 0.178521 LR 0.000500 Time 0.021353 +2023-10-02 21:31:21,243 - Epoch: [129][ 1010/ 1236] Overall Loss 0.178693 Objective Loss 0.178693 LR 0.000500 Time 0.021344 +2023-10-02 21:31:21,449 - Epoch: [129][ 1020/ 1236] Overall Loss 0.178503 Objective Loss 0.178503 LR 0.000500 Time 0.021337 +2023-10-02 21:31:21,654 - Epoch: [129][ 1030/ 1236] Overall Loss 0.178334 Objective Loss 0.178334 LR 0.000500 Time 0.021329 +2023-10-02 21:31:21,861 - Epoch: [129][ 1040/ 1236] Overall Loss 0.178262 Objective Loss 0.178262 LR 0.000500 Time 0.021322 +2023-10-02 21:31:22,066 - Epoch: [129][ 1050/ 1236] Overall Loss 0.178374 Objective Loss 0.178374 LR 0.000500 Time 0.021314 +2023-10-02 21:31:22,272 - Epoch: [129][ 1060/ 1236] Overall Loss 0.178465 Objective Loss 0.178465 LR 0.000500 Time 0.021307 +2023-10-02 21:31:22,477 - Epoch: [129][ 1070/ 1236] Overall Loss 0.178664 Objective Loss 0.178664 LR 0.000500 Time 0.021299 +2023-10-02 21:31:22,682 - Epoch: [129][ 1080/ 1236] Overall Loss 0.178648 Objective Loss 0.178648 LR 0.000500 Time 0.021292 +2023-10-02 21:31:22,888 - Epoch: [129][ 1090/ 1236] Overall Loss 0.178393 Objective Loss 0.178393 LR 0.000500 Time 0.021285 +2023-10-02 21:31:23,095 - Epoch: [129][ 1100/ 1236] Overall Loss 0.178246 Objective Loss 0.178246 LR 0.000500 Time 0.021280 +2023-10-02 21:31:23,301 - Epoch: [129][ 1110/ 1236] Overall Loss 0.178231 Objective Loss 0.178231 LR 0.000500 Time 0.021272 +2023-10-02 21:31:23,506 - Epoch: [129][ 1120/ 1236] Overall Loss 0.178058 Objective Loss 0.178058 LR 0.000500 Time 0.021265 +2023-10-02 21:31:23,711 - Epoch: [129][ 1130/ 1236] Overall Loss 0.177906 Objective Loss 0.177906 LR 0.000500 Time 0.021258 +2023-10-02 21:31:23,918 - Epoch: [129][ 1140/ 1236] Overall Loss 0.177841 Objective Loss 0.177841 LR 0.000500 Time 0.021253 +2023-10-02 21:31:24,124 - Epoch: [129][ 1150/ 1236] Overall Loss 0.177850 Objective Loss 0.177850 LR 0.000500 Time 0.021247 +2023-10-02 21:31:24,330 - Epoch: [129][ 1160/ 1236] Overall Loss 0.177750 Objective Loss 0.177750 LR 0.000500 Time 0.021241 +2023-10-02 21:31:24,535 - Epoch: [129][ 1170/ 1236] Overall Loss 0.177580 Objective Loss 0.177580 LR 0.000500 Time 0.021234 +2023-10-02 21:31:24,741 - Epoch: [129][ 1180/ 1236] Overall Loss 0.177810 Objective Loss 0.177810 LR 0.000500 Time 0.021229 +2023-10-02 21:31:24,947 - Epoch: [129][ 1190/ 1236] Overall Loss 0.177875 Objective Loss 0.177875 LR 0.000500 Time 0.021222 +2023-10-02 21:31:25,154 - Epoch: [129][ 1200/ 1236] Overall Loss 0.177812 Objective Loss 0.177812 LR 0.000500 Time 0.021217 +2023-10-02 21:31:25,359 - Epoch: [129][ 1210/ 1236] Overall Loss 0.177667 Objective Loss 0.177667 LR 0.000500 Time 0.021211 +2023-10-02 21:31:25,565 - Epoch: [129][ 1220/ 1236] Overall Loss 0.177759 Objective Loss 0.177759 LR 0.000500 Time 0.021206 +2023-10-02 21:31:25,824 - Epoch: [129][ 1230/ 1236] Overall Loss 0.177709 Objective Loss 0.177709 LR 0.000500 Time 0.021244 +2023-10-02 21:31:25,945 - Epoch: [129][ 1236/ 1236] Overall Loss 0.177731 Objective Loss 0.177731 Top1 89.205703 Top5 98.778004 LR 0.000500 Time 0.021239 +2023-10-02 21:31:26,086 - --- validate (epoch=129)----------- +2023-10-02 21:31:26,087 - 29943 samples (256 per mini-batch) +2023-10-02 21:31:26,582 - Epoch: [129][ 10/ 117] Loss 0.248216 Top1 86.875000 Top5 99.023438 +2023-10-02 21:31:26,744 - Epoch: [129][ 20/ 117] Loss 0.276092 Top1 86.484375 Top5 98.730469 +2023-10-02 21:31:26,902 - Epoch: [129][ 30/ 117] Loss 0.277305 Top1 86.328125 Top5 98.632812 +2023-10-02 21:31:27,063 - Epoch: [129][ 40/ 117] Loss 0.280828 Top1 86.171875 Top5 98.750000 +2023-10-02 21:31:27,221 - Epoch: [129][ 50/ 117] Loss 0.284280 Top1 86.101562 Top5 98.718750 +2023-10-02 21:31:27,383 - Epoch: [129][ 60/ 117] Loss 0.287741 Top1 86.113281 Top5 98.697917 +2023-10-02 21:31:27,540 - Epoch: [129][ 70/ 117] Loss 0.293316 Top1 86.021205 Top5 98.671875 +2023-10-02 21:31:27,700 - Epoch: [129][ 80/ 117] Loss 0.296072 Top1 86.059570 Top5 98.642578 +2023-10-02 21:31:27,857 - Epoch: [129][ 90/ 117] Loss 0.296281 Top1 86.228299 Top5 98.606771 +2023-10-02 21:31:28,018 - Epoch: [129][ 100/ 117] Loss 0.298137 Top1 86.222656 Top5 98.613281 +2023-10-02 21:31:28,184 - Epoch: [129][ 110/ 117] Loss 0.298845 Top1 86.239347 Top5 98.597301 +2023-10-02 21:31:28,273 - Epoch: [129][ 117/ 117] Loss 0.297146 Top1 86.237184 Top5 98.587316 +2023-10-02 21:31:28,393 - ==> Top1: 86.237 Top5: 98.587 Loss: 0.297 + +2023-10-02 21:31:28,394 - ==> Confusion: +[[ 941 0 6 1 2 2 0 0 7 64 1 0 2 2 4 1 2 1 0 0 14] + [ 1 1042 0 2 3 32 1 21 2 1 0 0 1 0 2 4 0 0 14 1 4] + [ 1 0 973 14 0 0 19 4 0 3 5 0 5 2 1 5 1 2 10 2 9] + [ 2 2 8 994 1 1 1 2 5 1 7 1 4 3 21 3 1 4 10 1 17] + [ 32 2 0 1 961 7 0 0 1 10 1 0 0 2 9 6 10 0 2 1 5] + [ 3 36 1 3 4 1004 0 18 1 7 2 4 0 9 6 1 1 0 2 2 12] + [ 0 3 27 0 0 2 1136 6 0 0 3 1 0 1 0 3 0 0 1 4 4] + [ 1 10 10 0 4 22 9 1082 1 3 1 3 4 6 0 0 3 3 42 7 7] + [ 18 2 1 0 1 2 0 0 982 29 13 1 3 13 16 0 1 0 2 1 4] + [ 84 0 1 0 3 5 0 0 27 951 1 0 0 24 8 3 1 1 1 1 8] + [ 4 2 8 10 0 1 4 4 13 0 970 0 0 8 5 0 4 3 9 0 8] + [ 0 0 2 0 0 13 0 1 0 0 1 959 22 10 0 1 1 15 0 6 4] + [ 0 0 1 5 0 0 2 0 0 0 3 32 972 3 1 9 2 13 3 6 16] + [ 0 0 1 0 1 9 1 0 11 9 4 3 0 1058 5 1 1 1 0 0 14] + [ 16 0 4 22 5 1 0 0 19 2 6 0 2 1 1002 0 1 3 8 0 9] + [ 0 1 0 3 6 0 0 0 1 0 1 5 8 0 0 1068 15 15 1 3 7] + [ 1 11 2 0 7 10 0 0 0 0 0 5 1 2 5 10 1089 0 1 5 12] + [ 0 0 2 6 0 0 0 0 4 0 0 2 26 1 3 4 2 985 0 0 3] + [ 5 5 4 14 1 1 1 17 4 0 3 0 0 0 10 1 0 1 990 0 11] + [ 0 0 4 1 2 3 11 7 0 0 0 10 4 3 1 4 9 1 0 1078 14] + [ 122 101 103 104 56 138 51 84 86 71 150 78 348 261 122 54 79 55 105 152 5585]] + +2023-10-02 21:31:28,396 - ==> Best [Top1: 86.247 Top5: 98.574 Sparsity:0.00 Params: 169472 on epoch: 127] +2023-10-02 21:31:28,396 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:31:28,402 - + +2023-10-02 21:31:28,402 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:31:29,559 - Epoch: [130][ 10/ 1236] Overall Loss 0.168978 Objective Loss 0.168978 LR 0.000500 Time 0.115650 +2023-10-02 21:31:29,766 - Epoch: [130][ 20/ 1236] Overall Loss 0.178291 Objective Loss 0.178291 LR 0.000500 Time 0.068153 +2023-10-02 21:31:29,971 - Epoch: [130][ 30/ 1236] Overall Loss 0.175141 Objective Loss 0.175141 LR 0.000500 Time 0.052281 +2023-10-02 21:31:30,180 - Epoch: [130][ 40/ 1236] Overall Loss 0.170050 Objective Loss 0.170050 LR 0.000500 Time 0.044406 +2023-10-02 21:31:30,388 - Epoch: [130][ 50/ 1236] Overall Loss 0.167255 Objective Loss 0.167255 LR 0.000500 Time 0.039693 +2023-10-02 21:31:30,598 - Epoch: [130][ 60/ 1236] Overall Loss 0.168480 Objective Loss 0.168480 LR 0.000500 Time 0.036574 +2023-10-02 21:31:30,807 - Epoch: [130][ 70/ 1236] Overall Loss 0.170044 Objective Loss 0.170044 LR 0.000500 Time 0.034331 +2023-10-02 21:31:31,017 - Epoch: [130][ 80/ 1236] Overall Loss 0.172541 Objective Loss 0.172541 LR 0.000500 Time 0.032660 +2023-10-02 21:31:31,226 - Epoch: [130][ 90/ 1236] Overall Loss 0.174420 Objective Loss 0.174420 LR 0.000500 Time 0.031349 +2023-10-02 21:31:31,436 - Epoch: [130][ 100/ 1236] Overall Loss 0.173531 Objective Loss 0.173531 LR 0.000500 Time 0.030307 +2023-10-02 21:31:31,645 - Epoch: [130][ 110/ 1236] Overall Loss 0.171903 Objective Loss 0.171903 LR 0.000500 Time 0.029449 +2023-10-02 21:31:31,855 - Epoch: [130][ 120/ 1236] Overall Loss 0.173114 Objective Loss 0.173114 LR 0.000500 Time 0.028742 +2023-10-02 21:31:32,065 - Epoch: [130][ 130/ 1236] Overall Loss 0.173937 Objective Loss 0.173937 LR 0.000500 Time 0.028128 +2023-10-02 21:31:32,277 - Epoch: [130][ 140/ 1236] Overall Loss 0.174565 Objective Loss 0.174565 LR 0.000500 Time 0.027629 +2023-10-02 21:31:32,485 - Epoch: [130][ 150/ 1236] Overall Loss 0.175379 Objective Loss 0.175379 LR 0.000500 Time 0.027172 +2023-10-02 21:31:32,697 - Epoch: [130][ 160/ 1236] Overall Loss 0.175197 Objective Loss 0.175197 LR 0.000500 Time 0.026797 +2023-10-02 21:31:32,905 - Epoch: [130][ 170/ 1236] Overall Loss 0.176190 Objective Loss 0.176190 LR 0.000500 Time 0.026442 +2023-10-02 21:31:33,117 - Epoch: [130][ 180/ 1236] Overall Loss 0.177026 Objective Loss 0.177026 LR 0.000500 Time 0.026149 +2023-10-02 21:31:33,325 - Epoch: [130][ 190/ 1236] Overall Loss 0.176839 Objective Loss 0.176839 LR 0.000500 Time 0.025866 +2023-10-02 21:31:33,537 - Epoch: [130][ 200/ 1236] Overall Loss 0.176227 Objective Loss 0.176227 LR 0.000500 Time 0.025632 +2023-10-02 21:31:33,745 - Epoch: [130][ 210/ 1236] Overall Loss 0.175938 Objective Loss 0.175938 LR 0.000500 Time 0.025400 +2023-10-02 21:31:33,957 - Epoch: [130][ 220/ 1236] Overall Loss 0.175570 Objective Loss 0.175570 LR 0.000500 Time 0.025208 +2023-10-02 21:31:34,165 - Epoch: [130][ 230/ 1236] Overall Loss 0.174807 Objective Loss 0.174807 LR 0.000500 Time 0.025016 +2023-10-02 21:31:34,377 - Epoch: [130][ 240/ 1236] Overall Loss 0.175162 Objective Loss 0.175162 LR 0.000500 Time 0.024855 +2023-10-02 21:31:34,585 - Epoch: [130][ 250/ 1236] Overall Loss 0.174682 Objective Loss 0.174682 LR 0.000500 Time 0.024691 +2023-10-02 21:31:34,797 - Epoch: [130][ 260/ 1236] Overall Loss 0.174828 Objective Loss 0.174828 LR 0.000500 Time 0.024554 +2023-10-02 21:31:35,004 - Epoch: [130][ 270/ 1236] Overall Loss 0.175116 Objective Loss 0.175116 LR 0.000500 Time 0.024413 +2023-10-02 21:31:35,216 - Epoch: [130][ 280/ 1236] Overall Loss 0.175493 Objective Loss 0.175493 LR 0.000500 Time 0.024297 +2023-10-02 21:31:35,424 - Epoch: [130][ 290/ 1236] Overall Loss 0.175697 Objective Loss 0.175697 LR 0.000500 Time 0.024176 +2023-10-02 21:31:35,636 - Epoch: [130][ 300/ 1236] Overall Loss 0.175231 Objective Loss 0.175231 LR 0.000500 Time 0.024075 +2023-10-02 21:31:35,844 - Epoch: [130][ 310/ 1236] Overall Loss 0.174858 Objective Loss 0.174858 LR 0.000500 Time 0.023969 +2023-10-02 21:31:36,056 - Epoch: [130][ 320/ 1236] Overall Loss 0.175405 Objective Loss 0.175405 LR 0.000500 Time 0.023881 +2023-10-02 21:31:36,264 - Epoch: [130][ 330/ 1236] Overall Loss 0.175709 Objective Loss 0.175709 LR 0.000500 Time 0.023787 +2023-10-02 21:31:36,476 - Epoch: [130][ 340/ 1236] Overall Loss 0.175179 Objective Loss 0.175179 LR 0.000500 Time 0.023711 +2023-10-02 21:31:36,684 - Epoch: [130][ 350/ 1236] Overall Loss 0.175216 Objective Loss 0.175216 LR 0.000500 Time 0.023626 +2023-10-02 21:31:36,896 - Epoch: [130][ 360/ 1236] Overall Loss 0.174797 Objective Loss 0.174797 LR 0.000500 Time 0.023557 +2023-10-02 21:31:37,104 - Epoch: [130][ 370/ 1236] Overall Loss 0.174660 Objective Loss 0.174660 LR 0.000500 Time 0.023482 +2023-10-02 21:31:37,314 - Epoch: [130][ 380/ 1236] Overall Loss 0.174787 Objective Loss 0.174787 LR 0.000500 Time 0.023416 +2023-10-02 21:31:37,523 - Epoch: [130][ 390/ 1236] Overall Loss 0.174548 Objective Loss 0.174548 LR 0.000500 Time 0.023351 +2023-10-02 21:31:37,733 - Epoch: [130][ 400/ 1236] Overall Loss 0.174564 Objective Loss 0.174564 LR 0.000500 Time 0.023292 +2023-10-02 21:31:37,942 - Epoch: [130][ 410/ 1236] Overall Loss 0.175192 Objective Loss 0.175192 LR 0.000500 Time 0.023233 +2023-10-02 21:31:38,152 - Epoch: [130][ 420/ 1236] Overall Loss 0.175169 Objective Loss 0.175169 LR 0.000500 Time 0.023179 +2023-10-02 21:31:38,361 - Epoch: [130][ 430/ 1236] Overall Loss 0.175486 Objective Loss 0.175486 LR 0.000500 Time 0.023126 +2023-10-02 21:31:38,571 - Epoch: [130][ 440/ 1236] Overall Loss 0.175484 Objective Loss 0.175484 LR 0.000500 Time 0.023075 +2023-10-02 21:31:38,780 - Epoch: [130][ 450/ 1236] Overall Loss 0.176284 Objective Loss 0.176284 LR 0.000500 Time 0.023027 +2023-10-02 21:31:38,990 - Epoch: [130][ 460/ 1236] Overall Loss 0.175772 Objective Loss 0.175772 LR 0.000500 Time 0.022983 +2023-10-02 21:31:39,199 - Epoch: [130][ 470/ 1236] Overall Loss 0.176221 Objective Loss 0.176221 LR 0.000500 Time 0.022938 +2023-10-02 21:31:39,410 - Epoch: [130][ 480/ 1236] Overall Loss 0.176350 Objective Loss 0.176350 LR 0.000500 Time 0.022898 +2023-10-02 21:31:39,619 - Epoch: [130][ 490/ 1236] Overall Loss 0.176769 Objective Loss 0.176769 LR 0.000500 Time 0.022857 +2023-10-02 21:31:39,829 - Epoch: [130][ 500/ 1236] Overall Loss 0.176584 Objective Loss 0.176584 LR 0.000500 Time 0.022820 +2023-10-02 21:31:40,038 - Epoch: [130][ 510/ 1236] Overall Loss 0.177005 Objective Loss 0.177005 LR 0.000500 Time 0.022782 +2023-10-02 21:31:40,248 - Epoch: [130][ 520/ 1236] Overall Loss 0.176734 Objective Loss 0.176734 LR 0.000500 Time 0.022748 +2023-10-02 21:31:40,458 - Epoch: [130][ 530/ 1236] Overall Loss 0.176811 Objective Loss 0.176811 LR 0.000500 Time 0.022713 +2023-10-02 21:31:40,668 - Epoch: [130][ 540/ 1236] Overall Loss 0.177191 Objective Loss 0.177191 LR 0.000500 Time 0.022681 +2023-10-02 21:31:40,876 - Epoch: [130][ 550/ 1236] Overall Loss 0.177263 Objective Loss 0.177263 LR 0.000500 Time 0.022646 +2023-10-02 21:31:41,086 - Epoch: [130][ 560/ 1236] Overall Loss 0.177577 Objective Loss 0.177577 LR 0.000500 Time 0.022617 +2023-10-02 21:31:41,296 - Epoch: [130][ 570/ 1236] Overall Loss 0.176814 Objective Loss 0.176814 LR 0.000500 Time 0.022587 +2023-10-02 21:31:41,505 - Epoch: [130][ 580/ 1236] Overall Loss 0.176426 Objective Loss 0.176426 LR 0.000500 Time 0.022559 +2023-10-02 21:31:41,714 - Epoch: [130][ 590/ 1236] Overall Loss 0.176151 Objective Loss 0.176151 LR 0.000500 Time 0.022530 +2023-10-02 21:31:41,925 - Epoch: [130][ 600/ 1236] Overall Loss 0.176045 Objective Loss 0.176045 LR 0.000500 Time 0.022505 +2023-10-02 21:31:42,134 - Epoch: [130][ 610/ 1236] Overall Loss 0.175696 Objective Loss 0.175696 LR 0.000500 Time 0.022477 +2023-10-02 21:31:42,345 - Epoch: [130][ 620/ 1236] Overall Loss 0.176012 Objective Loss 0.176012 LR 0.000500 Time 0.022453 +2023-10-02 21:31:42,554 - Epoch: [130][ 630/ 1236] Overall Loss 0.176149 Objective Loss 0.176149 LR 0.000500 Time 0.022426 +2023-10-02 21:31:42,764 - Epoch: [130][ 640/ 1236] Overall Loss 0.176164 Objective Loss 0.176164 LR 0.000500 Time 0.022404 +2023-10-02 21:31:42,973 - Epoch: [130][ 650/ 1236] Overall Loss 0.176471 Objective Loss 0.176471 LR 0.000500 Time 0.022381 +2023-10-02 21:31:43,184 - Epoch: [130][ 660/ 1236] Overall Loss 0.176514 Objective Loss 0.176514 LR 0.000500 Time 0.022360 +2023-10-02 21:31:43,393 - Epoch: [130][ 670/ 1236] Overall Loss 0.176849 Objective Loss 0.176849 LR 0.000500 Time 0.022338 +2023-10-02 21:31:43,604 - Epoch: [130][ 680/ 1236] Overall Loss 0.177386 Objective Loss 0.177386 LR 0.000500 Time 0.022320 +2023-10-02 21:31:43,812 - Epoch: [130][ 690/ 1236] Overall Loss 0.177802 Objective Loss 0.177802 LR 0.000500 Time 0.022298 +2023-10-02 21:31:44,023 - Epoch: [130][ 700/ 1236] Overall Loss 0.177832 Objective Loss 0.177832 LR 0.000500 Time 0.022279 +2023-10-02 21:31:44,232 - Epoch: [130][ 710/ 1236] Overall Loss 0.177766 Objective Loss 0.177766 LR 0.000500 Time 0.022260 +2023-10-02 21:31:44,442 - Epoch: [130][ 720/ 1236] Overall Loss 0.177475 Objective Loss 0.177475 LR 0.000500 Time 0.022242 +2023-10-02 21:31:44,651 - Epoch: [130][ 730/ 1236] Overall Loss 0.177566 Objective Loss 0.177566 LR 0.000500 Time 0.022224 +2023-10-02 21:31:44,862 - Epoch: [130][ 740/ 1236] Overall Loss 0.177709 Objective Loss 0.177709 LR 0.000500 Time 0.022207 +2023-10-02 21:31:45,071 - Epoch: [130][ 750/ 1236] Overall Loss 0.177763 Objective Loss 0.177763 LR 0.000500 Time 0.022190 +2023-10-02 21:31:45,282 - Epoch: [130][ 760/ 1236] Overall Loss 0.177812 Objective Loss 0.177812 LR 0.000500 Time 0.022175 +2023-10-02 21:31:45,491 - Epoch: [130][ 770/ 1236] Overall Loss 0.177636 Objective Loss 0.177636 LR 0.000500 Time 0.022158 +2023-10-02 21:31:45,701 - Epoch: [130][ 780/ 1236] Overall Loss 0.177576 Objective Loss 0.177576 LR 0.000500 Time 0.022143 +2023-10-02 21:31:45,910 - Epoch: [130][ 790/ 1236] Overall Loss 0.177783 Objective Loss 0.177783 LR 0.000500 Time 0.022127 +2023-10-02 21:31:46,121 - Epoch: [130][ 800/ 1236] Overall Loss 0.177593 Objective Loss 0.177593 LR 0.000500 Time 0.022113 +2023-10-02 21:31:46,330 - Epoch: [130][ 810/ 1236] Overall Loss 0.177996 Objective Loss 0.177996 LR 0.000500 Time 0.022099 +2023-10-02 21:31:46,541 - Epoch: [130][ 820/ 1236] Overall Loss 0.177811 Objective Loss 0.177811 LR 0.000500 Time 0.022086 +2023-10-02 21:31:46,750 - Epoch: [130][ 830/ 1236] Overall Loss 0.177554 Objective Loss 0.177554 LR 0.000500 Time 0.022071 +2023-10-02 21:31:46,960 - Epoch: [130][ 840/ 1236] Overall Loss 0.177437 Objective Loss 0.177437 LR 0.000500 Time 0.022058 +2023-10-02 21:31:47,169 - Epoch: [130][ 850/ 1236] Overall Loss 0.177535 Objective Loss 0.177535 LR 0.000500 Time 0.022044 +2023-10-02 21:31:47,380 - Epoch: [130][ 860/ 1236] Overall Loss 0.177555 Objective Loss 0.177555 LR 0.000500 Time 0.022032 +2023-10-02 21:31:47,589 - Epoch: [130][ 870/ 1236] Overall Loss 0.177486 Objective Loss 0.177486 LR 0.000500 Time 0.022019 +2023-10-02 21:31:47,799 - Epoch: [130][ 880/ 1236] Overall Loss 0.177419 Objective Loss 0.177419 LR 0.000500 Time 0.022007 +2023-10-02 21:31:48,012 - Epoch: [130][ 890/ 1236] Overall Loss 0.177330 Objective Loss 0.177330 LR 0.000500 Time 0.021999 +2023-10-02 21:31:48,224 - Epoch: [130][ 900/ 1236] Overall Loss 0.177340 Objective Loss 0.177340 LR 0.000500 Time 0.021989 +2023-10-02 21:31:48,437 - Epoch: [130][ 910/ 1236] Overall Loss 0.177390 Objective Loss 0.177390 LR 0.000500 Time 0.021981 +2023-10-02 21:31:48,650 - Epoch: [130][ 920/ 1236] Overall Loss 0.177383 Objective Loss 0.177383 LR 0.000500 Time 0.021973 +2023-10-02 21:31:48,863 - Epoch: [130][ 930/ 1236] Overall Loss 0.177574 Objective Loss 0.177574 LR 0.000500 Time 0.021965 +2023-10-02 21:31:49,076 - Epoch: [130][ 940/ 1236] Overall Loss 0.177802 Objective Loss 0.177802 LR 0.000500 Time 0.021958 +2023-10-02 21:31:49,289 - Epoch: [130][ 950/ 1236] Overall Loss 0.177615 Objective Loss 0.177615 LR 0.000500 Time 0.021950 +2023-10-02 21:31:49,501 - Epoch: [130][ 960/ 1236] Overall Loss 0.177617 Objective Loss 0.177617 LR 0.000500 Time 0.021942 +2023-10-02 21:31:49,714 - Epoch: [130][ 970/ 1236] Overall Loss 0.177625 Objective Loss 0.177625 LR 0.000500 Time 0.021935 +2023-10-02 21:31:49,927 - Epoch: [130][ 980/ 1236] Overall Loss 0.177512 Objective Loss 0.177512 LR 0.000500 Time 0.021928 +2023-10-02 21:31:50,140 - Epoch: [130][ 990/ 1236] Overall Loss 0.177693 Objective Loss 0.177693 LR 0.000500 Time 0.021921 +2023-10-02 21:31:50,352 - Epoch: [130][ 1000/ 1236] Overall Loss 0.177772 Objective Loss 0.177772 LR 0.000500 Time 0.021914 +2023-10-02 21:31:50,565 - Epoch: [130][ 1010/ 1236] Overall Loss 0.177741 Objective Loss 0.177741 LR 0.000500 Time 0.021908 +2023-10-02 21:31:50,778 - Epoch: [130][ 1020/ 1236] Overall Loss 0.177994 Objective Loss 0.177994 LR 0.000500 Time 0.021901 +2023-10-02 21:31:50,990 - Epoch: [130][ 1030/ 1236] Overall Loss 0.177964 Objective Loss 0.177964 LR 0.000500 Time 0.021895 +2023-10-02 21:31:51,203 - Epoch: [130][ 1040/ 1236] Overall Loss 0.178116 Objective Loss 0.178116 LR 0.000500 Time 0.021889 +2023-10-02 21:31:51,416 - Epoch: [130][ 1050/ 1236] Overall Loss 0.178140 Objective Loss 0.178140 LR 0.000500 Time 0.021883 +2023-10-02 21:31:51,629 - Epoch: [130][ 1060/ 1236] Overall Loss 0.178295 Objective Loss 0.178295 LR 0.000500 Time 0.021877 +2023-10-02 21:31:51,842 - Epoch: [130][ 1070/ 1236] Overall Loss 0.178123 Objective Loss 0.178123 LR 0.000500 Time 0.021870 +2023-10-02 21:31:52,054 - Epoch: [130][ 1080/ 1236] Overall Loss 0.178087 Objective Loss 0.178087 LR 0.000500 Time 0.021864 +2023-10-02 21:31:52,267 - Epoch: [130][ 1090/ 1236] Overall Loss 0.178179 Objective Loss 0.178179 LR 0.000500 Time 0.021859 +2023-10-02 21:31:52,480 - Epoch: [130][ 1100/ 1236] Overall Loss 0.178346 Objective Loss 0.178346 LR 0.000500 Time 0.021853 +2023-10-02 21:31:52,692 - Epoch: [130][ 1110/ 1236] Overall Loss 0.178406 Objective Loss 0.178406 LR 0.000500 Time 0.021847 +2023-10-02 21:31:52,905 - Epoch: [130][ 1120/ 1236] Overall Loss 0.178568 Objective Loss 0.178568 LR 0.000500 Time 0.021842 +2023-10-02 21:31:53,118 - Epoch: [130][ 1130/ 1236] Overall Loss 0.178740 Objective Loss 0.178740 LR 0.000500 Time 0.021837 +2023-10-02 21:31:53,331 - Epoch: [130][ 1140/ 1236] Overall Loss 0.178631 Objective Loss 0.178631 LR 0.000500 Time 0.021832 +2023-10-02 21:31:53,544 - Epoch: [130][ 1150/ 1236] Overall Loss 0.178696 Objective Loss 0.178696 LR 0.000500 Time 0.021827 +2023-10-02 21:31:53,756 - Epoch: [130][ 1160/ 1236] Overall Loss 0.178858 Objective Loss 0.178858 LR 0.000500 Time 0.021821 +2023-10-02 21:31:53,969 - Epoch: [130][ 1170/ 1236] Overall Loss 0.178668 Objective Loss 0.178668 LR 0.000500 Time 0.021817 +2023-10-02 21:31:54,182 - Epoch: [130][ 1180/ 1236] Overall Loss 0.178703 Objective Loss 0.178703 LR 0.000500 Time 0.021811 +2023-10-02 21:31:54,394 - Epoch: [130][ 1190/ 1236] Overall Loss 0.178955 Objective Loss 0.178955 LR 0.000500 Time 0.021807 +2023-10-02 21:31:54,607 - Epoch: [130][ 1200/ 1236] Overall Loss 0.178893 Objective Loss 0.178893 LR 0.000500 Time 0.021802 +2023-10-02 21:31:54,820 - Epoch: [130][ 1210/ 1236] Overall Loss 0.178904 Objective Loss 0.178904 LR 0.000500 Time 0.021798 +2023-10-02 21:31:55,034 - Epoch: [130][ 1220/ 1236] Overall Loss 0.178998 Objective Loss 0.178998 LR 0.000500 Time 0.021794 +2023-10-02 21:31:55,298 - Epoch: [130][ 1230/ 1236] Overall Loss 0.178955 Objective Loss 0.178955 LR 0.000500 Time 0.021831 +2023-10-02 21:31:55,421 - Epoch: [130][ 1236/ 1236] Overall Loss 0.179018 Objective Loss 0.179018 Top1 90.224033 Top5 99.592668 LR 0.000500 Time 0.021825 +2023-10-02 21:31:55,555 - --- validate (epoch=130)----------- +2023-10-02 21:31:55,556 - 29943 samples (256 per mini-batch) +2023-10-02 21:31:56,039 - Epoch: [130][ 10/ 117] Loss 0.331235 Top1 85.351562 Top5 98.476562 +2023-10-02 21:31:56,192 - Epoch: [130][ 20/ 117] Loss 0.314198 Top1 86.250000 Top5 98.632812 +2023-10-02 21:31:56,344 - Epoch: [130][ 30/ 117] Loss 0.314203 Top1 86.263021 Top5 98.476562 +2023-10-02 21:31:56,498 - Epoch: [130][ 40/ 117] Loss 0.311608 Top1 86.240234 Top5 98.525391 +2023-10-02 21:31:56,650 - Epoch: [130][ 50/ 117] Loss 0.310900 Top1 86.117188 Top5 98.484375 +2023-10-02 21:31:56,801 - Epoch: [130][ 60/ 117] Loss 0.306584 Top1 86.126302 Top5 98.522135 +2023-10-02 21:31:56,955 - Epoch: [130][ 70/ 117] Loss 0.307166 Top1 86.077009 Top5 98.543527 +2023-10-02 21:31:57,108 - Epoch: [130][ 80/ 117] Loss 0.302463 Top1 86.240234 Top5 98.569336 +2023-10-02 21:31:57,261 - Epoch: [130][ 90/ 117] Loss 0.303918 Top1 86.206597 Top5 98.546007 +2023-10-02 21:31:57,414 - Epoch: [130][ 100/ 117] Loss 0.303604 Top1 86.062500 Top5 98.562500 +2023-10-02 21:31:57,573 - Epoch: [130][ 110/ 117] Loss 0.305566 Top1 85.976562 Top5 98.547585 +2023-10-02 21:31:57,663 - Epoch: [130][ 117/ 117] Loss 0.303680 Top1 86.026784 Top5 98.540560 +2023-10-02 21:31:57,809 - ==> Top1: 86.027 Top5: 98.541 Loss: 0.304 + +2023-10-02 21:31:57,810 - ==> Confusion: +[[ 941 1 4 0 7 3 0 0 3 63 1 0 0 2 3 1 4 0 0 0 17] + [ 1 1056 2 0 7 22 2 15 2 2 1 0 1 0 2 1 1 0 9 4 3] + [ 4 0 975 6 6 0 14 9 0 2 4 2 4 2 0 4 1 1 7 3 12] + [ 3 3 14 950 1 0 1 3 11 1 10 0 9 3 38 3 0 6 14 0 19] + [ 27 6 0 0 966 5 0 0 2 13 1 0 1 0 10 5 6 0 1 1 6] + [ 4 41 1 0 6 990 0 16 5 7 3 6 1 6 7 0 5 0 3 2 13] + [ 0 4 20 0 0 1 1131 6 0 1 4 1 0 0 0 6 0 1 2 8 6] + [ 3 16 11 0 10 24 8 1059 1 4 2 2 3 5 0 1 0 3 41 10 15] + [ 16 1 0 1 5 2 0 0 984 37 8 2 0 10 13 4 0 0 3 1 2] + [ 90 0 0 0 7 3 0 0 23 957 1 1 0 21 8 1 0 1 0 0 6] + [ 6 5 9 4 0 1 3 4 16 0 968 1 0 14 5 0 2 2 5 0 8] + [ 1 1 3 0 2 10 0 2 0 0 0 965 21 10 0 0 0 11 0 5 4] + [ 1 1 1 1 3 1 0 2 1 1 1 42 961 2 4 8 4 15 2 4 13] + [ 0 0 1 0 3 10 0 0 14 12 3 6 0 1050 5 1 1 1 0 0 12] + [ 13 0 4 14 1 0 1 0 20 3 2 0 3 1 1018 0 2 3 7 0 9] + [ 0 0 2 1 6 0 1 0 0 1 1 5 6 0 0 1072 13 15 1 5 5] + [ 0 13 0 0 6 7 0 0 0 0 0 6 0 3 4 9 1099 0 0 5 9] + [ 0 1 0 2 0 1 2 0 2 0 0 3 23 0 4 6 0 991 0 1 2] + [ 2 4 5 8 1 0 1 17 5 1 2 0 2 0 10 0 0 0 999 0 11] + [ 0 1 3 1 2 6 9 6 0 1 2 8 3 4 0 1 10 0 0 1087 8] + [ 150 140 96 73 77 101 42 73 107 90 150 97 304 237 136 59 105 59 103 166 5540]] + +2023-10-02 21:31:57,811 - ==> Best [Top1: 86.247 Top5: 98.574 Sparsity:0.00 Params: 169472 on epoch: 127] +2023-10-02 21:31:57,812 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:31:57,818 - + +2023-10-02 21:31:57,818 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:31:58,885 - Epoch: [131][ 10/ 1236] Overall Loss 0.175308 Objective Loss 0.175308 LR 0.000500 Time 0.106664 +2023-10-02 21:31:59,096 - Epoch: [131][ 20/ 1236] Overall Loss 0.166562 Objective Loss 0.166562 LR 0.000500 Time 0.063859 +2023-10-02 21:31:59,306 - Epoch: [131][ 30/ 1236] Overall Loss 0.180410 Objective Loss 0.180410 LR 0.000500 Time 0.049577 +2023-10-02 21:31:59,518 - Epoch: [131][ 40/ 1236] Overall Loss 0.181336 Objective Loss 0.181336 LR 0.000500 Time 0.042462 +2023-10-02 21:31:59,729 - Epoch: [131][ 50/ 1236] Overall Loss 0.183462 Objective Loss 0.183462 LR 0.000500 Time 0.038170 +2023-10-02 21:31:59,941 - Epoch: [131][ 60/ 1236] Overall Loss 0.181307 Objective Loss 0.181307 LR 0.000500 Time 0.035330 +2023-10-02 21:32:00,152 - Epoch: [131][ 70/ 1236] Overall Loss 0.177265 Objective Loss 0.177265 LR 0.000500 Time 0.033283 +2023-10-02 21:32:00,364 - Epoch: [131][ 80/ 1236] Overall Loss 0.178203 Objective Loss 0.178203 LR 0.000500 Time 0.031764 +2023-10-02 21:32:00,576 - Epoch: [131][ 90/ 1236] Overall Loss 0.177182 Objective Loss 0.177182 LR 0.000500 Time 0.030571 +2023-10-02 21:32:00,789 - Epoch: [131][ 100/ 1236] Overall Loss 0.178717 Objective Loss 0.178717 LR 0.000500 Time 0.029641 +2023-10-02 21:32:00,998 - Epoch: [131][ 110/ 1236] Overall Loss 0.178868 Objective Loss 0.178868 LR 0.000500 Time 0.028844 +2023-10-02 21:32:01,210 - Epoch: [131][ 120/ 1236] Overall Loss 0.178156 Objective Loss 0.178156 LR 0.000500 Time 0.028210 +2023-10-02 21:32:01,419 - Epoch: [131][ 130/ 1236] Overall Loss 0.179268 Objective Loss 0.179268 LR 0.000500 Time 0.027644 +2023-10-02 21:32:01,632 - Epoch: [131][ 140/ 1236] Overall Loss 0.179308 Objective Loss 0.179308 LR 0.000500 Time 0.027189 +2023-10-02 21:32:01,841 - Epoch: [131][ 150/ 1236] Overall Loss 0.179015 Objective Loss 0.179015 LR 0.000500 Time 0.026766 +2023-10-02 21:32:02,054 - Epoch: [131][ 160/ 1236] Overall Loss 0.179420 Objective Loss 0.179420 LR 0.000500 Time 0.026421 +2023-10-02 21:32:02,262 - Epoch: [131][ 170/ 1236] Overall Loss 0.179455 Objective Loss 0.179455 LR 0.000500 Time 0.026093 +2023-10-02 21:32:02,475 - Epoch: [131][ 180/ 1236] Overall Loss 0.179799 Objective Loss 0.179799 LR 0.000500 Time 0.025824 +2023-10-02 21:32:02,684 - Epoch: [131][ 190/ 1236] Overall Loss 0.178307 Objective Loss 0.178307 LR 0.000500 Time 0.025563 +2023-10-02 21:32:02,897 - Epoch: [131][ 200/ 1236] Overall Loss 0.179714 Objective Loss 0.179714 LR 0.000500 Time 0.025347 +2023-10-02 21:32:03,106 - Epoch: [131][ 210/ 1236] Overall Loss 0.179409 Objective Loss 0.179409 LR 0.000500 Time 0.025134 +2023-10-02 21:32:03,319 - Epoch: [131][ 220/ 1236] Overall Loss 0.179663 Objective Loss 0.179663 LR 0.000500 Time 0.024957 +2023-10-02 21:32:03,528 - Epoch: [131][ 230/ 1236] Overall Loss 0.179895 Objective Loss 0.179895 LR 0.000500 Time 0.024779 +2023-10-02 21:32:03,738 - Epoch: [131][ 240/ 1236] Overall Loss 0.179994 Objective Loss 0.179994 LR 0.000500 Time 0.024622 +2023-10-02 21:32:03,947 - Epoch: [131][ 250/ 1236] Overall Loss 0.179284 Objective Loss 0.179284 LR 0.000500 Time 0.024468 +2023-10-02 21:32:04,157 - Epoch: [131][ 260/ 1236] Overall Loss 0.180047 Objective Loss 0.180047 LR 0.000500 Time 0.024333 +2023-10-02 21:32:04,366 - Epoch: [131][ 270/ 1236] Overall Loss 0.179595 Objective Loss 0.179595 LR 0.000500 Time 0.024200 +2023-10-02 21:32:04,575 - Epoch: [131][ 280/ 1236] Overall Loss 0.179852 Objective Loss 0.179852 LR 0.000500 Time 0.024082 +2023-10-02 21:32:04,784 - Epoch: [131][ 290/ 1236] Overall Loss 0.179204 Objective Loss 0.179204 LR 0.000500 Time 0.023968 +2023-10-02 21:32:04,995 - Epoch: [131][ 300/ 1236] Overall Loss 0.179514 Objective Loss 0.179514 LR 0.000500 Time 0.023870 +2023-10-02 21:32:05,204 - Epoch: [131][ 310/ 1236] Overall Loss 0.179057 Objective Loss 0.179057 LR 0.000500 Time 0.023775 +2023-10-02 21:32:05,415 - Epoch: [131][ 320/ 1236] Overall Loss 0.178494 Objective Loss 0.178494 LR 0.000500 Time 0.023689 +2023-10-02 21:32:05,624 - Epoch: [131][ 330/ 1236] Overall Loss 0.178431 Objective Loss 0.178431 LR 0.000500 Time 0.023601 +2023-10-02 21:32:05,835 - Epoch: [131][ 340/ 1236] Overall Loss 0.178387 Objective Loss 0.178387 LR 0.000500 Time 0.023526 +2023-10-02 21:32:06,044 - Epoch: [131][ 350/ 1236] Overall Loss 0.177823 Objective Loss 0.177823 LR 0.000500 Time 0.023451 +2023-10-02 21:32:06,255 - Epoch: [131][ 360/ 1236] Overall Loss 0.177693 Objective Loss 0.177693 LR 0.000500 Time 0.023384 +2023-10-02 21:32:06,464 - Epoch: [131][ 370/ 1236] Overall Loss 0.176916 Objective Loss 0.176916 LR 0.000500 Time 0.023313 +2023-10-02 21:32:06,675 - Epoch: [131][ 380/ 1236] Overall Loss 0.177346 Objective Loss 0.177346 LR 0.000500 Time 0.023253 +2023-10-02 21:32:06,884 - Epoch: [131][ 390/ 1236] Overall Loss 0.177166 Objective Loss 0.177166 LR 0.000500 Time 0.023190 +2023-10-02 21:32:07,095 - Epoch: [131][ 400/ 1236] Overall Loss 0.177454 Objective Loss 0.177454 LR 0.000500 Time 0.023136 +2023-10-02 21:32:07,304 - Epoch: [131][ 410/ 1236] Overall Loss 0.176726 Objective Loss 0.176726 LR 0.000500 Time 0.023079 +2023-10-02 21:32:07,515 - Epoch: [131][ 420/ 1236] Overall Loss 0.176325 Objective Loss 0.176325 LR 0.000500 Time 0.023030 +2023-10-02 21:32:07,724 - Epoch: [131][ 430/ 1236] Overall Loss 0.176319 Objective Loss 0.176319 LR 0.000500 Time 0.022978 +2023-10-02 21:32:07,935 - Epoch: [131][ 440/ 1236] Overall Loss 0.176338 Objective Loss 0.176338 LR 0.000500 Time 0.022933 +2023-10-02 21:32:08,144 - Epoch: [131][ 450/ 1236] Overall Loss 0.176433 Objective Loss 0.176433 LR 0.000500 Time 0.022885 +2023-10-02 21:32:08,354 - Epoch: [131][ 460/ 1236] Overall Loss 0.176391 Objective Loss 0.176391 LR 0.000500 Time 0.022844 +2023-10-02 21:32:08,565 - Epoch: [131][ 470/ 1236] Overall Loss 0.176481 Objective Loss 0.176481 LR 0.000500 Time 0.022802 +2023-10-02 21:32:08,775 - Epoch: [131][ 480/ 1236] Overall Loss 0.176671 Objective Loss 0.176671 LR 0.000500 Time 0.022764 +2023-10-02 21:32:08,985 - Epoch: [131][ 490/ 1236] Overall Loss 0.176584 Objective Loss 0.176584 LR 0.000500 Time 0.022726 +2023-10-02 21:32:09,195 - Epoch: [131][ 500/ 1236] Overall Loss 0.176997 Objective Loss 0.176997 LR 0.000500 Time 0.022691 +2023-10-02 21:32:09,405 - Epoch: [131][ 510/ 1236] Overall Loss 0.177068 Objective Loss 0.177068 LR 0.000500 Time 0.022655 +2023-10-02 21:32:09,616 - Epoch: [131][ 520/ 1236] Overall Loss 0.177737 Objective Loss 0.177737 LR 0.000500 Time 0.022624 +2023-10-02 21:32:09,825 - Epoch: [131][ 530/ 1236] Overall Loss 0.177771 Objective Loss 0.177771 LR 0.000500 Time 0.022589 +2023-10-02 21:32:10,036 - Epoch: [131][ 540/ 1236] Overall Loss 0.177853 Objective Loss 0.177853 LR 0.000500 Time 0.022560 +2023-10-02 21:32:10,245 - Epoch: [131][ 550/ 1236] Overall Loss 0.177438 Objective Loss 0.177438 LR 0.000500 Time 0.022529 +2023-10-02 21:32:10,456 - Epoch: [131][ 560/ 1236] Overall Loss 0.177553 Objective Loss 0.177553 LR 0.000500 Time 0.022501 +2023-10-02 21:32:10,666 - Epoch: [131][ 570/ 1236] Overall Loss 0.178020 Objective Loss 0.178020 LR 0.000500 Time 0.022472 +2023-10-02 21:32:10,876 - Epoch: [131][ 580/ 1236] Overall Loss 0.178284 Objective Loss 0.178284 LR 0.000500 Time 0.022447 +2023-10-02 21:32:11,086 - Epoch: [131][ 590/ 1236] Overall Loss 0.178296 Objective Loss 0.178296 LR 0.000500 Time 0.022419 +2023-10-02 21:32:11,296 - Epoch: [131][ 600/ 1236] Overall Loss 0.178481 Objective Loss 0.178481 LR 0.000500 Time 0.022395 +2023-10-02 21:32:11,506 - Epoch: [131][ 610/ 1236] Overall Loss 0.178414 Objective Loss 0.178414 LR 0.000500 Time 0.022371 +2023-10-02 21:32:11,716 - Epoch: [131][ 620/ 1236] Overall Loss 0.178515 Objective Loss 0.178515 LR 0.000500 Time 0.022350 +2023-10-02 21:32:11,926 - Epoch: [131][ 630/ 1236] Overall Loss 0.178315 Objective Loss 0.178315 LR 0.000500 Time 0.022326 +2023-10-02 21:32:12,137 - Epoch: [131][ 640/ 1236] Overall Loss 0.178184 Objective Loss 0.178184 LR 0.000500 Time 0.022306 +2023-10-02 21:32:12,347 - Epoch: [131][ 650/ 1236] Overall Loss 0.177921 Objective Loss 0.177921 LR 0.000500 Time 0.022283 +2023-10-02 21:32:12,557 - Epoch: [131][ 660/ 1236] Overall Loss 0.177982 Objective Loss 0.177982 LR 0.000500 Time 0.022263 +2023-10-02 21:32:12,767 - Epoch: [131][ 670/ 1236] Overall Loss 0.178168 Objective Loss 0.178168 LR 0.000500 Time 0.022242 +2023-10-02 21:32:12,977 - Epoch: [131][ 680/ 1236] Overall Loss 0.178224 Objective Loss 0.178224 LR 0.000500 Time 0.022224 +2023-10-02 21:32:13,187 - Epoch: [131][ 690/ 1236] Overall Loss 0.178233 Objective Loss 0.178233 LR 0.000500 Time 0.022204 +2023-10-02 21:32:13,398 - Epoch: [131][ 700/ 1236] Overall Loss 0.178521 Objective Loss 0.178521 LR 0.000500 Time 0.022187 +2023-10-02 21:32:13,608 - Epoch: [131][ 710/ 1236] Overall Loss 0.178379 Objective Loss 0.178379 LR 0.000500 Time 0.022168 +2023-10-02 21:32:13,819 - Epoch: [131][ 720/ 1236] Overall Loss 0.178300 Objective Loss 0.178300 LR 0.000500 Time 0.022153 +2023-10-02 21:32:14,028 - Epoch: [131][ 730/ 1236] Overall Loss 0.178141 Objective Loss 0.178141 LR 0.000500 Time 0.022135 +2023-10-02 21:32:14,239 - Epoch: [131][ 740/ 1236] Overall Loss 0.177831 Objective Loss 0.177831 LR 0.000500 Time 0.022120 +2023-10-02 21:32:14,449 - Epoch: [131][ 750/ 1236] Overall Loss 0.177701 Objective Loss 0.177701 LR 0.000500 Time 0.022102 +2023-10-02 21:32:14,659 - Epoch: [131][ 760/ 1236] Overall Loss 0.177373 Objective Loss 0.177373 LR 0.000500 Time 0.022088 +2023-10-02 21:32:14,869 - Epoch: [131][ 770/ 1236] Overall Loss 0.177372 Objective Loss 0.177372 LR 0.000500 Time 0.022071 +2023-10-02 21:32:15,079 - Epoch: [131][ 780/ 1236] Overall Loss 0.177186 Objective Loss 0.177186 LR 0.000500 Time 0.022058 +2023-10-02 21:32:15,289 - Epoch: [131][ 790/ 1236] Overall Loss 0.177125 Objective Loss 0.177125 LR 0.000500 Time 0.022042 +2023-10-02 21:32:15,499 - Epoch: [131][ 800/ 1236] Overall Loss 0.177234 Objective Loss 0.177234 LR 0.000500 Time 0.022029 +2023-10-02 21:32:15,709 - Epoch: [131][ 810/ 1236] Overall Loss 0.177541 Objective Loss 0.177541 LR 0.000500 Time 0.022014 +2023-10-02 21:32:15,919 - Epoch: [131][ 820/ 1236] Overall Loss 0.177382 Objective Loss 0.177382 LR 0.000500 Time 0.022002 +2023-10-02 21:32:16,128 - Epoch: [131][ 830/ 1236] Overall Loss 0.177569 Objective Loss 0.177569 LR 0.000500 Time 0.021987 +2023-10-02 21:32:16,339 - Epoch: [131][ 840/ 1236] Overall Loss 0.177714 Objective Loss 0.177714 LR 0.000500 Time 0.021976 +2023-10-02 21:32:16,549 - Epoch: [131][ 850/ 1236] Overall Loss 0.177739 Objective Loss 0.177739 LR 0.000500 Time 0.021962 +2023-10-02 21:32:16,759 - Epoch: [131][ 860/ 1236] Overall Loss 0.177692 Objective Loss 0.177692 LR 0.000500 Time 0.021951 +2023-10-02 21:32:16,969 - Epoch: [131][ 870/ 1236] Overall Loss 0.177823 Objective Loss 0.177823 LR 0.000500 Time 0.021938 +2023-10-02 21:32:17,179 - Epoch: [131][ 880/ 1236] Overall Loss 0.178159 Objective Loss 0.178159 LR 0.000500 Time 0.021927 +2023-10-02 21:32:17,389 - Epoch: [131][ 890/ 1236] Overall Loss 0.178280 Objective Loss 0.178280 LR 0.000500 Time 0.021915 +2023-10-02 21:32:17,599 - Epoch: [131][ 900/ 1236] Overall Loss 0.178109 Objective Loss 0.178109 LR 0.000500 Time 0.021905 +2023-10-02 21:32:17,809 - Epoch: [131][ 910/ 1236] Overall Loss 0.178039 Objective Loss 0.178039 LR 0.000500 Time 0.021893 +2023-10-02 21:32:18,019 - Epoch: [131][ 920/ 1236] Overall Loss 0.177957 Objective Loss 0.177957 LR 0.000500 Time 0.021883 +2023-10-02 21:32:18,229 - Epoch: [131][ 930/ 1236] Overall Loss 0.177899 Objective Loss 0.177899 LR 0.000500 Time 0.021872 +2023-10-02 21:32:18,439 - Epoch: [131][ 940/ 1236] Overall Loss 0.178066 Objective Loss 0.178066 LR 0.000500 Time 0.021862 +2023-10-02 21:32:18,649 - Epoch: [131][ 950/ 1236] Overall Loss 0.177849 Objective Loss 0.177849 LR 0.000500 Time 0.021851 +2023-10-02 21:32:18,859 - Epoch: [131][ 960/ 1236] Overall Loss 0.177937 Objective Loss 0.177937 LR 0.000500 Time 0.021843 +2023-10-02 21:32:19,069 - Epoch: [131][ 970/ 1236] Overall Loss 0.177904 Objective Loss 0.177904 LR 0.000500 Time 0.021832 +2023-10-02 21:32:19,279 - Epoch: [131][ 980/ 1236] Overall Loss 0.177817 Objective Loss 0.177817 LR 0.000500 Time 0.021823 +2023-10-02 21:32:19,489 - Epoch: [131][ 990/ 1236] Overall Loss 0.177825 Objective Loss 0.177825 LR 0.000500 Time 0.021813 +2023-10-02 21:32:19,699 - Epoch: [131][ 1000/ 1236] Overall Loss 0.177644 Objective Loss 0.177644 LR 0.000500 Time 0.021805 +2023-10-02 21:32:19,909 - Epoch: [131][ 1010/ 1236] Overall Loss 0.177696 Objective Loss 0.177696 LR 0.000500 Time 0.021795 +2023-10-02 21:32:20,119 - Epoch: [131][ 1020/ 1236] Overall Loss 0.177757 Objective Loss 0.177757 LR 0.000500 Time 0.021788 +2023-10-02 21:32:20,329 - Epoch: [131][ 1030/ 1236] Overall Loss 0.177706 Objective Loss 0.177706 LR 0.000500 Time 0.021778 +2023-10-02 21:32:20,539 - Epoch: [131][ 1040/ 1236] Overall Loss 0.177638 Objective Loss 0.177638 LR 0.000500 Time 0.021771 +2023-10-02 21:32:20,749 - Epoch: [131][ 1050/ 1236] Overall Loss 0.177603 Objective Loss 0.177603 LR 0.000500 Time 0.021761 +2023-10-02 21:32:20,959 - Epoch: [131][ 1060/ 1236] Overall Loss 0.177333 Objective Loss 0.177333 LR 0.000500 Time 0.021754 +2023-10-02 21:32:21,169 - Epoch: [131][ 1070/ 1236] Overall Loss 0.177358 Objective Loss 0.177358 LR 0.000500 Time 0.021746 +2023-10-02 21:32:21,380 - Epoch: [131][ 1080/ 1236] Overall Loss 0.177329 Objective Loss 0.177329 LR 0.000500 Time 0.021739 +2023-10-02 21:32:21,588 - Epoch: [131][ 1090/ 1236] Overall Loss 0.177405 Objective Loss 0.177405 LR 0.000500 Time 0.021729 +2023-10-02 21:32:21,799 - Epoch: [131][ 1100/ 1236] Overall Loss 0.177326 Objective Loss 0.177326 LR 0.000500 Time 0.021723 +2023-10-02 21:32:22,008 - Epoch: [131][ 1110/ 1236] Overall Loss 0.177199 Objective Loss 0.177199 LR 0.000500 Time 0.021715 +2023-10-02 21:32:22,218 - Epoch: [131][ 1120/ 1236] Overall Loss 0.177138 Objective Loss 0.177138 LR 0.000500 Time 0.021708 +2023-10-02 21:32:22,428 - Epoch: [131][ 1130/ 1236] Overall Loss 0.177225 Objective Loss 0.177225 LR 0.000500 Time 0.021700 +2023-10-02 21:32:22,639 - Epoch: [131][ 1140/ 1236] Overall Loss 0.177181 Objective Loss 0.177181 LR 0.000500 Time 0.021694 +2023-10-02 21:32:22,848 - Epoch: [131][ 1150/ 1236] Overall Loss 0.177043 Objective Loss 0.177043 LR 0.000500 Time 0.021686 +2023-10-02 21:32:23,058 - Epoch: [131][ 1160/ 1236] Overall Loss 0.176966 Objective Loss 0.176966 LR 0.000500 Time 0.021681 +2023-10-02 21:32:23,268 - Epoch: [131][ 1170/ 1236] Overall Loss 0.177018 Objective Loss 0.177018 LR 0.000500 Time 0.021673 +2023-10-02 21:32:23,478 - Epoch: [131][ 1180/ 1236] Overall Loss 0.176944 Objective Loss 0.176944 LR 0.000500 Time 0.021668 +2023-10-02 21:32:23,688 - Epoch: [131][ 1190/ 1236] Overall Loss 0.177033 Objective Loss 0.177033 LR 0.000500 Time 0.021660 +2023-10-02 21:32:23,898 - Epoch: [131][ 1200/ 1236] Overall Loss 0.177102 Objective Loss 0.177102 LR 0.000500 Time 0.021655 +2023-10-02 21:32:24,108 - Epoch: [131][ 1210/ 1236] Overall Loss 0.177181 Objective Loss 0.177181 LR 0.000500 Time 0.021648 +2023-10-02 21:32:24,318 - Epoch: [131][ 1220/ 1236] Overall Loss 0.177121 Objective Loss 0.177121 LR 0.000500 Time 0.021643 +2023-10-02 21:32:24,582 - Epoch: [131][ 1230/ 1236] Overall Loss 0.177244 Objective Loss 0.177244 LR 0.000500 Time 0.021680 +2023-10-02 21:32:24,705 - Epoch: [131][ 1236/ 1236] Overall Loss 0.177219 Objective Loss 0.177219 Top1 89.002037 Top5 97.963340 LR 0.000500 Time 0.021674 +2023-10-02 21:32:24,855 - --- validate (epoch=131)----------- +2023-10-02 21:32:24,856 - 29943 samples (256 per mini-batch) +2023-10-02 21:32:25,363 - Epoch: [131][ 10/ 117] Loss 0.351897 Top1 85.859375 Top5 97.851562 +2023-10-02 21:32:25,520 - Epoch: [131][ 20/ 117] Loss 0.328346 Top1 86.269531 Top5 98.222656 +2023-10-02 21:32:25,673 - Epoch: [131][ 30/ 117] Loss 0.310235 Top1 86.614583 Top5 98.372396 +2023-10-02 21:32:25,829 - Epoch: [131][ 40/ 117] Loss 0.298811 Top1 86.855469 Top5 98.496094 +2023-10-02 21:32:25,980 - Epoch: [131][ 50/ 117] Loss 0.301054 Top1 86.648438 Top5 98.539062 +2023-10-02 21:32:26,136 - Epoch: [131][ 60/ 117] Loss 0.296723 Top1 86.660156 Top5 98.593750 +2023-10-02 21:32:26,291 - Epoch: [131][ 70/ 117] Loss 0.300976 Top1 86.623884 Top5 98.549107 +2023-10-02 21:32:26,449 - Epoch: [131][ 80/ 117] Loss 0.300780 Top1 86.557617 Top5 98.510742 +2023-10-02 21:32:26,610 - Epoch: [131][ 90/ 117] Loss 0.293636 Top1 86.675347 Top5 98.550347 +2023-10-02 21:32:26,781 - Epoch: [131][ 100/ 117] Loss 0.293235 Top1 86.664062 Top5 98.566406 +2023-10-02 21:32:26,954 - Epoch: [131][ 110/ 117] Loss 0.293966 Top1 86.626420 Top5 98.565341 +2023-10-02 21:32:27,043 - Epoch: [131][ 117/ 117] Loss 0.294109 Top1 86.624587 Top5 98.563938 +2023-10-02 21:32:27,174 - ==> Top1: 86.625 Top5: 98.564 Loss: 0.294 + +2023-10-02 21:32:27,175 - ==> Confusion: +[[ 946 1 3 0 7 3 0 0 8 57 2 0 0 1 5 0 1 0 1 0 15] + [ 0 1066 1 1 8 15 2 16 1 1 0 1 0 0 1 2 2 1 8 2 3] + [ 7 0 990 5 3 0 15 3 0 2 1 1 5 2 1 3 0 1 10 3 4] + [ 1 2 15 985 2 0 0 3 4 1 0 0 3 0 29 2 0 7 14 0 21] + [ 26 6 1 1 976 3 0 0 1 7 0 0 1 3 8 5 6 0 1 1 4] + [ 6 43 0 2 4 973 2 28 2 3 1 7 1 11 5 0 5 1 4 3 15] + [ 1 4 24 0 0 2 1131 2 0 0 5 2 0 0 0 7 0 0 1 8 4] + [ 2 21 13 1 10 25 6 1050 1 5 1 1 5 7 0 0 0 1 50 10 9] + [ 14 2 0 1 2 4 0 1 977 32 8 2 2 11 21 1 6 0 1 1 3] + [ 88 1 0 0 9 1 0 0 22 947 1 0 0 24 11 2 1 0 0 3 9] + [ 2 1 11 11 2 1 5 3 13 0 958 2 0 15 8 0 1 2 6 1 11] + [ 0 1 2 0 1 8 0 4 0 1 0 954 28 8 0 1 0 13 0 10 4] + [ 1 1 4 5 0 0 3 2 1 2 0 29 988 0 0 7 2 5 1 4 13] + [ 1 0 1 0 2 4 3 0 7 8 3 4 0 1069 4 0 0 1 0 3 9] + [ 11 0 5 15 3 1 0 0 13 2 2 0 1 1 1025 0 0 3 10 0 9] + [ 0 0 1 1 6 0 0 0 0 1 1 7 7 0 0 1076 14 7 2 6 5] + [ 1 13 1 2 8 8 0 0 0 0 0 5 2 1 5 12 1090 0 0 2 11] + [ 0 0 1 0 0 0 2 0 0 1 0 5 31 0 1 8 0 983 0 1 5] + [ 2 8 4 17 1 0 1 13 5 0 3 1 0 0 8 0 0 0 992 0 13] + [ 0 1 3 2 0 2 11 5 0 1 0 13 3 2 0 0 9 0 0 1094 6] + [ 118 155 106 82 84 110 35 65 82 61 142 90 289 252 112 50 82 48 119 155 5668]] + +2023-10-02 21:32:27,176 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:32:27,176 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:32:27,189 - + +2023-10-02 21:32:27,190 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:32:28,240 - Epoch: [132][ 10/ 1236] Overall Loss 0.175643 Objective Loss 0.175643 LR 0.000500 Time 0.104961 +2023-10-02 21:32:28,452 - Epoch: [132][ 20/ 1236] Overall Loss 0.175600 Objective Loss 0.175600 LR 0.000500 Time 0.063058 +2023-10-02 21:32:28,664 - Epoch: [132][ 30/ 1236] Overall Loss 0.170605 Objective Loss 0.170605 LR 0.000500 Time 0.049116 +2023-10-02 21:32:28,877 - Epoch: [132][ 40/ 1236] Overall Loss 0.168193 Objective Loss 0.168193 LR 0.000500 Time 0.042144 +2023-10-02 21:32:29,090 - Epoch: [132][ 50/ 1236] Overall Loss 0.170725 Objective Loss 0.170725 LR 0.000500 Time 0.037936 +2023-10-02 21:32:29,303 - Epoch: [132][ 60/ 1236] Overall Loss 0.169159 Objective Loss 0.169159 LR 0.000500 Time 0.035156 +2023-10-02 21:32:29,515 - Epoch: [132][ 70/ 1236] Overall Loss 0.167259 Objective Loss 0.167259 LR 0.000500 Time 0.033152 +2023-10-02 21:32:29,728 - Epoch: [132][ 80/ 1236] Overall Loss 0.167300 Objective Loss 0.167300 LR 0.000500 Time 0.031662 +2023-10-02 21:32:29,941 - Epoch: [132][ 90/ 1236] Overall Loss 0.164027 Objective Loss 0.164027 LR 0.000500 Time 0.030491 +2023-10-02 21:32:30,151 - Epoch: [132][ 100/ 1236] Overall Loss 0.164321 Objective Loss 0.164321 LR 0.000500 Time 0.029542 +2023-10-02 21:32:30,361 - Epoch: [132][ 110/ 1236] Overall Loss 0.165150 Objective Loss 0.165150 LR 0.000500 Time 0.028750 +2023-10-02 21:32:30,572 - Epoch: [132][ 120/ 1236] Overall Loss 0.163492 Objective Loss 0.163492 LR 0.000500 Time 0.028111 +2023-10-02 21:32:30,780 - Epoch: [132][ 130/ 1236] Overall Loss 0.166344 Objective Loss 0.166344 LR 0.000500 Time 0.027541 +2023-10-02 21:32:30,989 - Epoch: [132][ 140/ 1236] Overall Loss 0.165065 Objective Loss 0.165065 LR 0.000500 Time 0.027066 +2023-10-02 21:32:31,210 - Epoch: [132][ 150/ 1236] Overall Loss 0.163448 Objective Loss 0.163448 LR 0.000500 Time 0.026724 +2023-10-02 21:32:31,427 - Epoch: [132][ 160/ 1236] Overall Loss 0.164305 Objective Loss 0.164305 LR 0.000500 Time 0.026403 +2023-10-02 21:32:31,644 - Epoch: [132][ 170/ 1236] Overall Loss 0.163495 Objective Loss 0.163495 LR 0.000500 Time 0.026124 +2023-10-02 21:32:31,852 - Epoch: [132][ 180/ 1236] Overall Loss 0.165324 Objective Loss 0.165324 LR 0.000500 Time 0.025827 +2023-10-02 21:32:32,062 - Epoch: [132][ 190/ 1236] Overall Loss 0.165201 Objective Loss 0.165201 LR 0.000500 Time 0.025575 +2023-10-02 21:32:32,271 - Epoch: [132][ 200/ 1236] Overall Loss 0.165900 Objective Loss 0.165900 LR 0.000500 Time 0.025335 +2023-10-02 21:32:32,481 - Epoch: [132][ 210/ 1236] Overall Loss 0.167210 Objective Loss 0.167210 LR 0.000500 Time 0.025127 +2023-10-02 21:32:32,689 - Epoch: [132][ 220/ 1236] Overall Loss 0.167633 Objective Loss 0.167633 LR 0.000500 Time 0.024931 +2023-10-02 21:32:32,898 - Epoch: [132][ 230/ 1236] Overall Loss 0.168144 Objective Loss 0.168144 LR 0.000500 Time 0.024754 +2023-10-02 21:32:33,106 - Epoch: [132][ 240/ 1236] Overall Loss 0.167964 Objective Loss 0.167964 LR 0.000500 Time 0.024588 +2023-10-02 21:32:33,316 - Epoch: [132][ 250/ 1236] Overall Loss 0.167584 Objective Loss 0.167584 LR 0.000500 Time 0.024444 +2023-10-02 21:32:33,525 - Epoch: [132][ 260/ 1236] Overall Loss 0.167452 Objective Loss 0.167452 LR 0.000500 Time 0.024307 +2023-10-02 21:32:33,737 - Epoch: [132][ 270/ 1236] Overall Loss 0.167130 Objective Loss 0.167130 LR 0.000500 Time 0.024191 +2023-10-02 21:32:33,948 - Epoch: [132][ 280/ 1236] Overall Loss 0.167126 Objective Loss 0.167126 LR 0.000500 Time 0.024079 +2023-10-02 21:32:34,160 - Epoch: [132][ 290/ 1236] Overall Loss 0.166442 Objective Loss 0.166442 LR 0.000500 Time 0.023980 +2023-10-02 21:32:34,371 - Epoch: [132][ 300/ 1236] Overall Loss 0.167065 Objective Loss 0.167065 LR 0.000500 Time 0.023881 +2023-10-02 21:32:34,583 - Epoch: [132][ 310/ 1236] Overall Loss 0.167244 Objective Loss 0.167244 LR 0.000500 Time 0.023794 +2023-10-02 21:32:34,794 - Epoch: [132][ 320/ 1236] Overall Loss 0.167228 Objective Loss 0.167228 LR 0.000500 Time 0.023708 +2023-10-02 21:32:35,006 - Epoch: [132][ 330/ 1236] Overall Loss 0.167151 Objective Loss 0.167151 LR 0.000500 Time 0.023631 +2023-10-02 21:32:35,217 - Epoch: [132][ 340/ 1236] Overall Loss 0.168221 Objective Loss 0.168221 LR 0.000500 Time 0.023556 +2023-10-02 21:32:35,428 - Epoch: [132][ 350/ 1236] Overall Loss 0.167629 Objective Loss 0.167629 LR 0.000500 Time 0.023485 +2023-10-02 21:32:35,639 - Epoch: [132][ 360/ 1236] Overall Loss 0.168069 Objective Loss 0.168069 LR 0.000500 Time 0.023419 +2023-10-02 21:32:35,851 - Epoch: [132][ 370/ 1236] Overall Loss 0.168773 Objective Loss 0.168773 LR 0.000500 Time 0.023356 +2023-10-02 21:32:36,062 - Epoch: [132][ 380/ 1236] Overall Loss 0.169271 Objective Loss 0.169271 LR 0.000500 Time 0.023295 +2023-10-02 21:32:36,274 - Epoch: [132][ 390/ 1236] Overall Loss 0.170165 Objective Loss 0.170165 LR 0.000500 Time 0.023240 +2023-10-02 21:32:36,485 - Epoch: [132][ 400/ 1236] Overall Loss 0.170234 Objective Loss 0.170234 LR 0.000500 Time 0.023186 +2023-10-02 21:32:36,696 - Epoch: [132][ 410/ 1236] Overall Loss 0.169915 Objective Loss 0.169915 LR 0.000500 Time 0.023134 +2023-10-02 21:32:36,907 - Epoch: [132][ 420/ 1236] Overall Loss 0.169905 Objective Loss 0.169905 LR 0.000500 Time 0.023086 +2023-10-02 21:32:37,118 - Epoch: [132][ 430/ 1236] Overall Loss 0.170398 Objective Loss 0.170398 LR 0.000500 Time 0.023039 +2023-10-02 21:32:37,329 - Epoch: [132][ 440/ 1236] Overall Loss 0.170698 Objective Loss 0.170698 LR 0.000500 Time 0.022994 +2023-10-02 21:32:37,541 - Epoch: [132][ 450/ 1236] Overall Loss 0.171153 Objective Loss 0.171153 LR 0.000500 Time 0.022952 +2023-10-02 21:32:37,752 - Epoch: [132][ 460/ 1236] Overall Loss 0.170999 Objective Loss 0.170999 LR 0.000500 Time 0.022911 +2023-10-02 21:32:37,964 - Epoch: [132][ 470/ 1236] Overall Loss 0.171048 Objective Loss 0.171048 LR 0.000500 Time 0.022873 +2023-10-02 21:32:38,175 - Epoch: [132][ 480/ 1236] Overall Loss 0.170490 Objective Loss 0.170490 LR 0.000500 Time 0.022837 +2023-10-02 21:32:38,387 - Epoch: [132][ 490/ 1236] Overall Loss 0.170359 Objective Loss 0.170359 LR 0.000500 Time 0.022802 +2023-10-02 21:32:38,598 - Epoch: [132][ 500/ 1236] Overall Loss 0.170236 Objective Loss 0.170236 LR 0.000500 Time 0.022768 +2023-10-02 21:32:38,811 - Epoch: [132][ 510/ 1236] Overall Loss 0.170463 Objective Loss 0.170463 LR 0.000500 Time 0.022738 +2023-10-02 21:32:39,022 - Epoch: [132][ 520/ 1236] Overall Loss 0.170304 Objective Loss 0.170304 LR 0.000500 Time 0.022706 +2023-10-02 21:32:39,235 - Epoch: [132][ 530/ 1236] Overall Loss 0.170244 Objective Loss 0.170244 LR 0.000500 Time 0.022677 +2023-10-02 21:32:39,446 - Epoch: [132][ 540/ 1236] Overall Loss 0.170329 Objective Loss 0.170329 LR 0.000500 Time 0.022648 +2023-10-02 21:32:39,658 - Epoch: [132][ 550/ 1236] Overall Loss 0.170076 Objective Loss 0.170076 LR 0.000500 Time 0.022621 +2023-10-02 21:32:39,869 - Epoch: [132][ 560/ 1236] Overall Loss 0.170249 Objective Loss 0.170249 LR 0.000500 Time 0.022594 +2023-10-02 21:32:40,081 - Epoch: [132][ 570/ 1236] Overall Loss 0.170386 Objective Loss 0.170386 LR 0.000500 Time 0.022568 +2023-10-02 21:32:40,293 - Epoch: [132][ 580/ 1236] Overall Loss 0.170463 Objective Loss 0.170463 LR 0.000500 Time 0.022543 +2023-10-02 21:32:40,505 - Epoch: [132][ 590/ 1236] Overall Loss 0.170910 Objective Loss 0.170910 LR 0.000500 Time 0.022520 +2023-10-02 21:32:40,717 - Epoch: [132][ 600/ 1236] Overall Loss 0.170864 Objective Loss 0.170864 LR 0.000500 Time 0.022497 +2023-10-02 21:32:40,928 - Epoch: [132][ 610/ 1236] Overall Loss 0.170886 Objective Loss 0.170886 LR 0.000500 Time 0.022474 +2023-10-02 21:32:41,139 - Epoch: [132][ 620/ 1236] Overall Loss 0.171195 Objective Loss 0.171195 LR 0.000500 Time 0.022452 +2023-10-02 21:32:41,351 - Epoch: [132][ 630/ 1236] Overall Loss 0.171457 Objective Loss 0.171457 LR 0.000500 Time 0.022431 +2023-10-02 21:32:41,563 - Epoch: [132][ 640/ 1236] Overall Loss 0.171487 Objective Loss 0.171487 LR 0.000500 Time 0.022410 +2023-10-02 21:32:41,775 - Epoch: [132][ 650/ 1236] Overall Loss 0.171274 Objective Loss 0.171274 LR 0.000500 Time 0.022391 +2023-10-02 21:32:41,985 - Epoch: [132][ 660/ 1236] Overall Loss 0.171727 Objective Loss 0.171727 LR 0.000500 Time 0.022370 +2023-10-02 21:32:42,197 - Epoch: [132][ 670/ 1236] Overall Loss 0.171623 Objective Loss 0.171623 LR 0.000500 Time 0.022352 +2023-10-02 21:32:42,408 - Epoch: [132][ 680/ 1236] Overall Loss 0.172068 Objective Loss 0.172068 LR 0.000500 Time 0.022333 +2023-10-02 21:32:42,620 - Epoch: [132][ 690/ 1236] Overall Loss 0.172145 Objective Loss 0.172145 LR 0.000500 Time 0.022316 +2023-10-02 21:32:42,831 - Epoch: [132][ 700/ 1236] Overall Loss 0.172340 Objective Loss 0.172340 LR 0.000500 Time 0.022297 +2023-10-02 21:32:43,043 - Epoch: [132][ 710/ 1236] Overall Loss 0.172358 Objective Loss 0.172358 LR 0.000500 Time 0.022281 +2023-10-02 21:32:43,254 - Epoch: [132][ 720/ 1236] Overall Loss 0.172479 Objective Loss 0.172479 LR 0.000500 Time 0.022264 +2023-10-02 21:32:43,466 - Epoch: [132][ 730/ 1236] Overall Loss 0.172351 Objective Loss 0.172351 LR 0.000500 Time 0.022249 +2023-10-02 21:32:43,676 - Epoch: [132][ 740/ 1236] Overall Loss 0.172566 Objective Loss 0.172566 LR 0.000500 Time 0.022233 +2023-10-02 21:32:43,888 - Epoch: [132][ 750/ 1236] Overall Loss 0.172633 Objective Loss 0.172633 LR 0.000500 Time 0.022218 +2023-10-02 21:32:44,099 - Epoch: [132][ 760/ 1236] Overall Loss 0.172463 Objective Loss 0.172463 LR 0.000500 Time 0.022202 +2023-10-02 21:32:44,307 - Epoch: [132][ 770/ 1236] Overall Loss 0.172659 Objective Loss 0.172659 LR 0.000500 Time 0.022183 +2023-10-02 21:32:44,517 - Epoch: [132][ 780/ 1236] Overall Loss 0.173009 Objective Loss 0.173009 LR 0.000500 Time 0.022167 +2023-10-02 21:32:44,725 - Epoch: [132][ 790/ 1236] Overall Loss 0.172869 Objective Loss 0.172869 LR 0.000500 Time 0.022148 +2023-10-02 21:32:44,935 - Epoch: [132][ 800/ 1236] Overall Loss 0.172745 Objective Loss 0.172745 LR 0.000500 Time 0.022133 +2023-10-02 21:32:45,143 - Epoch: [132][ 810/ 1236] Overall Loss 0.173013 Objective Loss 0.173013 LR 0.000500 Time 0.022115 +2023-10-02 21:32:45,354 - Epoch: [132][ 820/ 1236] Overall Loss 0.173094 Objective Loss 0.173094 LR 0.000500 Time 0.022102 +2023-10-02 21:32:45,562 - Epoch: [132][ 830/ 1236] Overall Loss 0.173147 Objective Loss 0.173147 LR 0.000500 Time 0.022086 +2023-10-02 21:32:45,772 - Epoch: [132][ 840/ 1236] Overall Loss 0.173027 Objective Loss 0.173027 LR 0.000500 Time 0.022072 +2023-10-02 21:32:45,980 - Epoch: [132][ 850/ 1236] Overall Loss 0.173086 Objective Loss 0.173086 LR 0.000500 Time 0.022056 +2023-10-02 21:32:46,190 - Epoch: [132][ 860/ 1236] Overall Loss 0.172975 Objective Loss 0.172975 LR 0.000500 Time 0.022043 +2023-10-02 21:32:46,398 - Epoch: [132][ 870/ 1236] Overall Loss 0.173196 Objective Loss 0.173196 LR 0.000500 Time 0.022027 +2023-10-02 21:32:46,608 - Epoch: [132][ 880/ 1236] Overall Loss 0.173073 Objective Loss 0.173073 LR 0.000500 Time 0.022015 +2023-10-02 21:32:46,816 - Epoch: [132][ 890/ 1236] Overall Loss 0.173182 Objective Loss 0.173182 LR 0.000500 Time 0.022000 +2023-10-02 21:32:47,027 - Epoch: [132][ 900/ 1236] Overall Loss 0.173084 Objective Loss 0.173084 LR 0.000500 Time 0.021989 +2023-10-02 21:32:47,234 - Epoch: [132][ 910/ 1236] Overall Loss 0.172833 Objective Loss 0.172833 LR 0.000500 Time 0.021975 +2023-10-02 21:32:47,445 - Epoch: [132][ 920/ 1236] Overall Loss 0.172957 Objective Loss 0.172957 LR 0.000500 Time 0.021965 +2023-10-02 21:32:47,653 - Epoch: [132][ 930/ 1236] Overall Loss 0.172989 Objective Loss 0.172989 LR 0.000500 Time 0.021952 +2023-10-02 21:32:47,862 - Epoch: [132][ 940/ 1236] Overall Loss 0.172920 Objective Loss 0.172920 LR 0.000500 Time 0.021941 +2023-10-02 21:32:48,071 - Epoch: [132][ 950/ 1236] Overall Loss 0.173129 Objective Loss 0.173129 LR 0.000500 Time 0.021928 +2023-10-02 21:32:48,281 - Epoch: [132][ 960/ 1236] Overall Loss 0.173136 Objective Loss 0.173136 LR 0.000500 Time 0.021918 +2023-10-02 21:32:48,489 - Epoch: [132][ 970/ 1236] Overall Loss 0.173369 Objective Loss 0.173369 LR 0.000500 Time 0.021905 +2023-10-02 21:32:48,700 - Epoch: [132][ 980/ 1236] Overall Loss 0.173565 Objective Loss 0.173565 LR 0.000500 Time 0.021897 +2023-10-02 21:32:48,908 - Epoch: [132][ 990/ 1236] Overall Loss 0.173505 Objective Loss 0.173505 LR 0.000500 Time 0.021885 +2023-10-02 21:32:49,119 - Epoch: [132][ 1000/ 1236] Overall Loss 0.173613 Objective Loss 0.173613 LR 0.000500 Time 0.021877 +2023-10-02 21:32:49,328 - Epoch: [132][ 1010/ 1236] Overall Loss 0.173760 Objective Loss 0.173760 LR 0.000500 Time 0.021866 +2023-10-02 21:32:49,539 - Epoch: [132][ 1020/ 1236] Overall Loss 0.173875 Objective Loss 0.173875 LR 0.000500 Time 0.021858 +2023-10-02 21:32:49,748 - Epoch: [132][ 1030/ 1236] Overall Loss 0.174081 Objective Loss 0.174081 LR 0.000500 Time 0.021848 +2023-10-02 21:32:49,959 - Epoch: [132][ 1040/ 1236] Overall Loss 0.174160 Objective Loss 0.174160 LR 0.000500 Time 0.021840 +2023-10-02 21:32:50,168 - Epoch: [132][ 1050/ 1236] Overall Loss 0.174287 Objective Loss 0.174287 LR 0.000500 Time 0.021831 +2023-10-02 21:32:50,379 - Epoch: [132][ 1060/ 1236] Overall Loss 0.174203 Objective Loss 0.174203 LR 0.000500 Time 0.021824 +2023-10-02 21:32:50,588 - Epoch: [132][ 1070/ 1236] Overall Loss 0.174398 Objective Loss 0.174398 LR 0.000500 Time 0.021815 +2023-10-02 21:32:50,798 - Epoch: [132][ 1080/ 1236] Overall Loss 0.174528 Objective Loss 0.174528 LR 0.000500 Time 0.021808 +2023-10-02 21:32:51,007 - Epoch: [132][ 1090/ 1236] Overall Loss 0.174402 Objective Loss 0.174402 LR 0.000500 Time 0.021799 +2023-10-02 21:32:51,218 - Epoch: [132][ 1100/ 1236] Overall Loss 0.174397 Objective Loss 0.174397 LR 0.000500 Time 0.021792 +2023-10-02 21:32:51,427 - Epoch: [132][ 1110/ 1236] Overall Loss 0.174334 Objective Loss 0.174334 LR 0.000500 Time 0.021783 +2023-10-02 21:32:51,638 - Epoch: [132][ 1120/ 1236] Overall Loss 0.174378 Objective Loss 0.174378 LR 0.000500 Time 0.021776 +2023-10-02 21:32:51,847 - Epoch: [132][ 1130/ 1236] Overall Loss 0.174451 Objective Loss 0.174451 LR 0.000500 Time 0.021768 +2023-10-02 21:32:52,058 - Epoch: [132][ 1140/ 1236] Overall Loss 0.174633 Objective Loss 0.174633 LR 0.000500 Time 0.021762 +2023-10-02 21:32:52,267 - Epoch: [132][ 1150/ 1236] Overall Loss 0.174672 Objective Loss 0.174672 LR 0.000500 Time 0.021755 +2023-10-02 21:32:52,478 - Epoch: [132][ 1160/ 1236] Overall Loss 0.174726 Objective Loss 0.174726 LR 0.000500 Time 0.021748 +2023-10-02 21:32:52,687 - Epoch: [132][ 1170/ 1236] Overall Loss 0.174923 Objective Loss 0.174923 LR 0.000500 Time 0.021741 +2023-10-02 21:32:52,898 - Epoch: [132][ 1180/ 1236] Overall Loss 0.174875 Objective Loss 0.174875 LR 0.000500 Time 0.021735 +2023-10-02 21:32:53,107 - Epoch: [132][ 1190/ 1236] Overall Loss 0.174875 Objective Loss 0.174875 LR 0.000500 Time 0.021728 +2023-10-02 21:32:53,318 - Epoch: [132][ 1200/ 1236] Overall Loss 0.175217 Objective Loss 0.175217 LR 0.000500 Time 0.021722 +2023-10-02 21:32:53,527 - Epoch: [132][ 1210/ 1236] Overall Loss 0.175056 Objective Loss 0.175056 LR 0.000500 Time 0.021716 +2023-10-02 21:32:53,738 - Epoch: [132][ 1220/ 1236] Overall Loss 0.175051 Objective Loss 0.175051 LR 0.000500 Time 0.021710 +2023-10-02 21:32:54,002 - Epoch: [132][ 1230/ 1236] Overall Loss 0.175060 Objective Loss 0.175060 LR 0.000500 Time 0.021748 +2023-10-02 21:32:54,126 - Epoch: [132][ 1236/ 1236] Overall Loss 0.175156 Objective Loss 0.175156 Top1 86.354379 Top5 98.778004 LR 0.000500 Time 0.021742 +2023-10-02 21:32:54,261 - --- validate (epoch=132)----------- +2023-10-02 21:32:54,261 - 29943 samples (256 per mini-batch) +2023-10-02 21:32:54,779 - Epoch: [132][ 10/ 117] Loss 0.331953 Top1 85.117188 Top5 98.515625 +2023-10-02 21:32:54,941 - Epoch: [132][ 20/ 117] Loss 0.321579 Top1 85.976562 Top5 98.359375 +2023-10-02 21:32:55,106 - Epoch: [132][ 30/ 117] Loss 0.313304 Top1 86.145833 Top5 98.515625 +2023-10-02 21:32:55,268 - Epoch: [132][ 40/ 117] Loss 0.314147 Top1 86.240234 Top5 98.388672 +2023-10-02 21:32:55,431 - Epoch: [132][ 50/ 117] Loss 0.309779 Top1 86.226562 Top5 98.406250 +2023-10-02 21:32:55,592 - Epoch: [132][ 60/ 117] Loss 0.306509 Top1 86.210938 Top5 98.444010 +2023-10-02 21:32:55,758 - Epoch: [132][ 70/ 117] Loss 0.306682 Top1 86.216518 Top5 98.454241 +2023-10-02 21:32:55,918 - Epoch: [132][ 80/ 117] Loss 0.301039 Top1 86.347656 Top5 98.476562 +2023-10-02 21:32:56,080 - Epoch: [132][ 90/ 117] Loss 0.304288 Top1 86.319444 Top5 98.498264 +2023-10-02 21:32:56,239 - Epoch: [132][ 100/ 117] Loss 0.301769 Top1 86.308594 Top5 98.464844 +2023-10-02 21:32:56,404 - Epoch: [132][ 110/ 117] Loss 0.303844 Top1 86.306818 Top5 98.497869 +2023-10-02 21:32:56,494 - Epoch: [132][ 117/ 117] Loss 0.305229 Top1 86.310657 Top5 98.490465 +2023-10-02 21:32:56,595 - ==> Top1: 86.311 Top5: 98.490 Loss: 0.305 + +2023-10-02 21:32:56,596 - ==> Confusion: +[[ 942 0 3 2 8 3 0 0 2 56 1 1 1 3 6 1 2 1 0 0 18] + [ 0 1047 0 0 4 31 2 15 0 0 1 1 0 0 5 3 2 0 14 2 4] + [ 2 0 973 9 1 0 18 7 0 0 3 1 6 4 1 4 2 1 9 3 12] + [ 1 3 11 993 1 1 1 2 3 1 5 0 3 4 28 1 1 3 9 1 17] + [ 24 7 1 0 963 7 0 0 1 7 1 1 0 1 11 6 11 0 0 1 8] + [ 1 30 0 0 4 998 4 15 2 6 2 5 2 9 6 1 5 1 5 2 18] + [ 0 3 19 0 0 3 1129 3 0 0 6 1 0 0 1 5 0 2 0 10 9] + [ 2 12 8 1 7 26 8 1058 0 5 3 4 3 4 1 0 2 1 49 10 14] + [ 21 2 0 3 2 4 0 0 960 37 9 2 2 12 24 1 4 1 3 0 2] + [ 93 1 2 0 6 2 1 1 21 954 0 1 0 14 8 1 2 1 0 2 9] + [ 4 6 7 14 0 1 4 2 12 0 959 0 0 16 6 0 3 1 6 2 10] + [ 0 2 0 0 0 9 1 1 0 1 0 964 22 6 0 2 0 18 0 3 6] + [ 1 0 2 3 0 2 2 1 0 0 6 35 953 0 3 9 1 19 2 5 24] + [ 0 0 1 0 3 7 1 0 10 10 2 3 0 1070 3 0 0 1 0 0 8] + [ 7 0 4 16 5 1 0 0 6 3 3 0 1 4 1032 0 1 6 7 0 5] + [ 0 0 1 1 6 1 0 0 0 0 1 7 7 1 0 1069 13 13 2 5 7] + [ 0 12 0 1 4 10 1 0 1 0 1 4 0 3 5 8 1089 0 0 5 17] + [ 0 0 2 2 1 0 2 0 0 1 0 5 14 0 0 2 0 1003 0 1 5] + [ 0 3 1 20 1 0 1 18 4 0 2 0 1 0 15 0 0 0 992 0 10] + [ 0 0 0 4 0 3 6 3 0 1 3 15 3 1 2 1 8 1 0 1090 11] + [ 111 143 85 86 60 144 46 93 96 76 159 90 262 250 148 56 65 67 122 140 5606]] + +2023-10-02 21:32:56,597 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:32:56,597 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:32:56,603 - + +2023-10-02 21:32:56,603 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:32:57,642 - Epoch: [133][ 10/ 1236] Overall Loss 0.155160 Objective Loss 0.155160 LR 0.000500 Time 0.103783 +2023-10-02 21:32:57,852 - Epoch: [133][ 20/ 1236] Overall Loss 0.166386 Objective Loss 0.166386 LR 0.000500 Time 0.062368 +2023-10-02 21:32:58,061 - Epoch: [133][ 30/ 1236] Overall Loss 0.160331 Objective Loss 0.160331 LR 0.000500 Time 0.048536 +2023-10-02 21:32:58,271 - Epoch: [133][ 40/ 1236] Overall Loss 0.164066 Objective Loss 0.164066 LR 0.000500 Time 0.041645 +2023-10-02 21:32:58,480 - Epoch: [133][ 50/ 1236] Overall Loss 0.165902 Objective Loss 0.165902 LR 0.000500 Time 0.037493 +2023-10-02 21:32:58,690 - Epoch: [133][ 60/ 1236] Overall Loss 0.172672 Objective Loss 0.172672 LR 0.000500 Time 0.034741 +2023-10-02 21:32:58,899 - Epoch: [133][ 70/ 1236] Overall Loss 0.170799 Objective Loss 0.170799 LR 0.000500 Time 0.032758 +2023-10-02 21:32:59,109 - Epoch: [133][ 80/ 1236] Overall Loss 0.169170 Objective Loss 0.169170 LR 0.000500 Time 0.031284 +2023-10-02 21:32:59,318 - Epoch: [133][ 90/ 1236] Overall Loss 0.169862 Objective Loss 0.169862 LR 0.000500 Time 0.030129 +2023-10-02 21:32:59,528 - Epoch: [133][ 100/ 1236] Overall Loss 0.169830 Objective Loss 0.169830 LR 0.000500 Time 0.029218 +2023-10-02 21:32:59,737 - Epoch: [133][ 110/ 1236] Overall Loss 0.169426 Objective Loss 0.169426 LR 0.000500 Time 0.028457 +2023-10-02 21:32:59,946 - Epoch: [133][ 120/ 1236] Overall Loss 0.168124 Objective Loss 0.168124 LR 0.000500 Time 0.027827 +2023-10-02 21:33:00,154 - Epoch: [133][ 130/ 1236] Overall Loss 0.167452 Objective Loss 0.167452 LR 0.000500 Time 0.027273 +2023-10-02 21:33:00,364 - Epoch: [133][ 140/ 1236] Overall Loss 0.167023 Objective Loss 0.167023 LR 0.000500 Time 0.026819 +2023-10-02 21:33:00,571 - Epoch: [133][ 150/ 1236] Overall Loss 0.168633 Objective Loss 0.168633 LR 0.000500 Time 0.026404 +2023-10-02 21:33:00,780 - Epoch: [133][ 160/ 1236] Overall Loss 0.169285 Objective Loss 0.169285 LR 0.000500 Time 0.026060 +2023-10-02 21:33:00,988 - Epoch: [133][ 170/ 1236] Overall Loss 0.169425 Objective Loss 0.169425 LR 0.000500 Time 0.025739 +2023-10-02 21:33:01,198 - Epoch: [133][ 180/ 1236] Overall Loss 0.169097 Objective Loss 0.169097 LR 0.000500 Time 0.025472 +2023-10-02 21:33:01,405 - Epoch: [133][ 190/ 1236] Overall Loss 0.169016 Objective Loss 0.169016 LR 0.000500 Time 0.025217 +2023-10-02 21:33:01,614 - Epoch: [133][ 200/ 1236] Overall Loss 0.168675 Objective Loss 0.168675 LR 0.000500 Time 0.024997 +2023-10-02 21:33:01,822 - Epoch: [133][ 210/ 1236] Overall Loss 0.170142 Objective Loss 0.170142 LR 0.000500 Time 0.024790 +2023-10-02 21:33:02,031 - Epoch: [133][ 220/ 1236] Overall Loss 0.169674 Objective Loss 0.169674 LR 0.000500 Time 0.024614 +2023-10-02 21:33:02,239 - Epoch: [133][ 230/ 1236] Overall Loss 0.170479 Objective Loss 0.170479 LR 0.000500 Time 0.024443 +2023-10-02 21:33:02,449 - Epoch: [133][ 240/ 1236] Overall Loss 0.170608 Objective Loss 0.170608 LR 0.000500 Time 0.024295 +2023-10-02 21:33:02,657 - Epoch: [133][ 250/ 1236] Overall Loss 0.171079 Objective Loss 0.171079 LR 0.000500 Time 0.024149 +2023-10-02 21:33:02,866 - Epoch: [133][ 260/ 1236] Overall Loss 0.171454 Objective Loss 0.171454 LR 0.000500 Time 0.024025 +2023-10-02 21:33:03,074 - Epoch: [133][ 270/ 1236] Overall Loss 0.171816 Objective Loss 0.171816 LR 0.000500 Time 0.023899 +2023-10-02 21:33:03,283 - Epoch: [133][ 280/ 1236] Overall Loss 0.171664 Objective Loss 0.171664 LR 0.000500 Time 0.023792 +2023-10-02 21:33:03,491 - Epoch: [133][ 290/ 1236] Overall Loss 0.172284 Objective Loss 0.172284 LR 0.000500 Time 0.023683 +2023-10-02 21:33:03,700 - Epoch: [133][ 300/ 1236] Overall Loss 0.172979 Objective Loss 0.172979 LR 0.000500 Time 0.023591 +2023-10-02 21:33:03,909 - Epoch: [133][ 310/ 1236] Overall Loss 0.172808 Objective Loss 0.172808 LR 0.000500 Time 0.023496 +2023-10-02 21:33:04,118 - Epoch: [133][ 320/ 1236] Overall Loss 0.173157 Objective Loss 0.173157 LR 0.000500 Time 0.023416 +2023-10-02 21:33:04,326 - Epoch: [133][ 330/ 1236] Overall Loss 0.172626 Objective Loss 0.172626 LR 0.000500 Time 0.023333 +2023-10-02 21:33:04,536 - Epoch: [133][ 340/ 1236] Overall Loss 0.172061 Objective Loss 0.172061 LR 0.000500 Time 0.023261 +2023-10-02 21:33:04,744 - Epoch: [133][ 350/ 1236] Overall Loss 0.172432 Objective Loss 0.172432 LR 0.000500 Time 0.023187 +2023-10-02 21:33:04,953 - Epoch: [133][ 360/ 1236] Overall Loss 0.172700 Objective Loss 0.172700 LR 0.000500 Time 0.023124 +2023-10-02 21:33:05,161 - Epoch: [133][ 370/ 1236] Overall Loss 0.172628 Objective Loss 0.172628 LR 0.000500 Time 0.023057 +2023-10-02 21:33:05,371 - Epoch: [133][ 380/ 1236] Overall Loss 0.172301 Objective Loss 0.172301 LR 0.000500 Time 0.023001 +2023-10-02 21:33:05,579 - Epoch: [133][ 390/ 1236] Overall Loss 0.172395 Objective Loss 0.172395 LR 0.000500 Time 0.022941 +2023-10-02 21:33:05,788 - Epoch: [133][ 400/ 1236] Overall Loss 0.172593 Objective Loss 0.172593 LR 0.000500 Time 0.022890 +2023-10-02 21:33:05,996 - Epoch: [133][ 410/ 1236] Overall Loss 0.172978 Objective Loss 0.172978 LR 0.000500 Time 0.022835 +2023-10-02 21:33:06,206 - Epoch: [133][ 420/ 1236] Overall Loss 0.173012 Objective Loss 0.173012 LR 0.000500 Time 0.022790 +2023-10-02 21:33:06,414 - Epoch: [133][ 430/ 1236] Overall Loss 0.173150 Objective Loss 0.173150 LR 0.000500 Time 0.022740 +2023-10-02 21:33:06,623 - Epoch: [133][ 440/ 1236] Overall Loss 0.173096 Objective Loss 0.173096 LR 0.000500 Time 0.022699 +2023-10-02 21:33:06,831 - Epoch: [133][ 450/ 1236] Overall Loss 0.173640 Objective Loss 0.173640 LR 0.000500 Time 0.022654 +2023-10-02 21:33:07,041 - Epoch: [133][ 460/ 1236] Overall Loss 0.174012 Objective Loss 0.174012 LR 0.000500 Time 0.022616 +2023-10-02 21:33:07,249 - Epoch: [133][ 470/ 1236] Overall Loss 0.173788 Objective Loss 0.173788 LR 0.000500 Time 0.022574 +2023-10-02 21:33:07,459 - Epoch: [133][ 480/ 1236] Overall Loss 0.173845 Objective Loss 0.173845 LR 0.000500 Time 0.022540 +2023-10-02 21:33:07,667 - Epoch: [133][ 490/ 1236] Overall Loss 0.173396 Objective Loss 0.173396 LR 0.000500 Time 0.022502 +2023-10-02 21:33:07,876 - Epoch: [133][ 500/ 1236] Overall Loss 0.173314 Objective Loss 0.173314 LR 0.000500 Time 0.022470 +2023-10-02 21:33:08,084 - Epoch: [133][ 510/ 1236] Overall Loss 0.173036 Objective Loss 0.173036 LR 0.000500 Time 0.022435 +2023-10-02 21:33:08,294 - Epoch: [133][ 520/ 1236] Overall Loss 0.173007 Objective Loss 0.173007 LR 0.000500 Time 0.022406 +2023-10-02 21:33:08,502 - Epoch: [133][ 530/ 1236] Overall Loss 0.173273 Objective Loss 0.173273 LR 0.000500 Time 0.022374 +2023-10-02 21:33:08,712 - Epoch: [133][ 540/ 1236] Overall Loss 0.173446 Objective Loss 0.173446 LR 0.000500 Time 0.022347 +2023-10-02 21:33:08,920 - Epoch: [133][ 550/ 1236] Overall Loss 0.173709 Objective Loss 0.173709 LR 0.000500 Time 0.022317 +2023-10-02 21:33:09,130 - Epoch: [133][ 560/ 1236] Overall Loss 0.173921 Objective Loss 0.173921 LR 0.000500 Time 0.022292 +2023-10-02 21:33:09,338 - Epoch: [133][ 570/ 1236] Overall Loss 0.174054 Objective Loss 0.174054 LR 0.000500 Time 0.022263 +2023-10-02 21:33:09,547 - Epoch: [133][ 580/ 1236] Overall Loss 0.173936 Objective Loss 0.173936 LR 0.000500 Time 0.022240 +2023-10-02 21:33:09,755 - Epoch: [133][ 590/ 1236] Overall Loss 0.174172 Objective Loss 0.174172 LR 0.000500 Time 0.022213 +2023-10-02 21:33:09,965 - Epoch: [133][ 600/ 1236] Overall Loss 0.174176 Objective Loss 0.174176 LR 0.000500 Time 0.022192 +2023-10-02 21:33:10,173 - Epoch: [133][ 610/ 1236] Overall Loss 0.174217 Objective Loss 0.174217 LR 0.000500 Time 0.022166 +2023-10-02 21:33:10,382 - Epoch: [133][ 620/ 1236] Overall Loss 0.174432 Objective Loss 0.174432 LR 0.000500 Time 0.022146 +2023-10-02 21:33:10,591 - Epoch: [133][ 630/ 1236] Overall Loss 0.174746 Objective Loss 0.174746 LR 0.000500 Time 0.022123 +2023-10-02 21:33:10,800 - Epoch: [133][ 640/ 1236] Overall Loss 0.174886 Objective Loss 0.174886 LR 0.000500 Time 0.022104 +2023-10-02 21:33:11,008 - Epoch: [133][ 650/ 1236] Overall Loss 0.174776 Objective Loss 0.174776 LR 0.000500 Time 0.022082 +2023-10-02 21:33:11,218 - Epoch: [133][ 660/ 1236] Overall Loss 0.175013 Objective Loss 0.175013 LR 0.000500 Time 0.022065 +2023-10-02 21:33:11,426 - Epoch: [133][ 670/ 1236] Overall Loss 0.174966 Objective Loss 0.174966 LR 0.000500 Time 0.022044 +2023-10-02 21:33:11,636 - Epoch: [133][ 680/ 1236] Overall Loss 0.175002 Objective Loss 0.175002 LR 0.000500 Time 0.022028 +2023-10-02 21:33:11,844 - Epoch: [133][ 690/ 1236] Overall Loss 0.174883 Objective Loss 0.174883 LR 0.000500 Time 0.022008 +2023-10-02 21:33:12,053 - Epoch: [133][ 700/ 1236] Overall Loss 0.174944 Objective Loss 0.174944 LR 0.000500 Time 0.021992 +2023-10-02 21:33:12,262 - Epoch: [133][ 710/ 1236] Overall Loss 0.174834 Objective Loss 0.174834 LR 0.000500 Time 0.021974 +2023-10-02 21:33:12,471 - Epoch: [133][ 720/ 1236] Overall Loss 0.175025 Objective Loss 0.175025 LR 0.000500 Time 0.021959 +2023-10-02 21:33:12,679 - Epoch: [133][ 730/ 1236] Overall Loss 0.175100 Objective Loss 0.175100 LR 0.000500 Time 0.021942 +2023-10-02 21:33:12,889 - Epoch: [133][ 740/ 1236] Overall Loss 0.174989 Objective Loss 0.174989 LR 0.000500 Time 0.021928 +2023-10-02 21:33:13,097 - Epoch: [133][ 750/ 1236] Overall Loss 0.174981 Objective Loss 0.174981 LR 0.000500 Time 0.021911 +2023-10-02 21:33:13,306 - Epoch: [133][ 760/ 1236] Overall Loss 0.175083 Objective Loss 0.175083 LR 0.000500 Time 0.021898 +2023-10-02 21:33:13,515 - Epoch: [133][ 770/ 1236] Overall Loss 0.175367 Objective Loss 0.175367 LR 0.000500 Time 0.021882 +2023-10-02 21:33:13,724 - Epoch: [133][ 780/ 1236] Overall Loss 0.175370 Objective Loss 0.175370 LR 0.000500 Time 0.021870 +2023-10-02 21:33:13,932 - Epoch: [133][ 790/ 1236] Overall Loss 0.175430 Objective Loss 0.175430 LR 0.000500 Time 0.021854 +2023-10-02 21:33:14,142 - Epoch: [133][ 800/ 1236] Overall Loss 0.175503 Objective Loss 0.175503 LR 0.000500 Time 0.021843 +2023-10-02 21:33:14,350 - Epoch: [133][ 810/ 1236] Overall Loss 0.175486 Objective Loss 0.175486 LR 0.000500 Time 0.021828 +2023-10-02 21:33:14,559 - Epoch: [133][ 820/ 1236] Overall Loss 0.175540 Objective Loss 0.175540 LR 0.000500 Time 0.021817 +2023-10-02 21:33:14,768 - Epoch: [133][ 830/ 1236] Overall Loss 0.175671 Objective Loss 0.175671 LR 0.000500 Time 0.021804 +2023-10-02 21:33:14,977 - Epoch: [133][ 840/ 1236] Overall Loss 0.176014 Objective Loss 0.176014 LR 0.000500 Time 0.021793 +2023-10-02 21:33:15,186 - Epoch: [133][ 850/ 1236] Overall Loss 0.176057 Objective Loss 0.176057 LR 0.000500 Time 0.021780 +2023-10-02 21:33:15,395 - Epoch: [133][ 860/ 1236] Overall Loss 0.176321 Objective Loss 0.176321 LR 0.000500 Time 0.021770 +2023-10-02 21:33:15,604 - Epoch: [133][ 870/ 1236] Overall Loss 0.176384 Objective Loss 0.176384 LR 0.000500 Time 0.021758 +2023-10-02 21:33:15,813 - Epoch: [133][ 880/ 1236] Overall Loss 0.176502 Objective Loss 0.176502 LR 0.000500 Time 0.021749 +2023-10-02 21:33:16,022 - Epoch: [133][ 890/ 1236] Overall Loss 0.176128 Objective Loss 0.176128 LR 0.000500 Time 0.021736 +2023-10-02 21:33:16,231 - Epoch: [133][ 900/ 1236] Overall Loss 0.175878 Objective Loss 0.175878 LR 0.000500 Time 0.021728 +2023-10-02 21:33:16,440 - Epoch: [133][ 910/ 1236] Overall Loss 0.176291 Objective Loss 0.176291 LR 0.000500 Time 0.021716 +2023-10-02 21:33:16,649 - Epoch: [133][ 920/ 1236] Overall Loss 0.176094 Objective Loss 0.176094 LR 0.000500 Time 0.021708 +2023-10-02 21:33:16,857 - Epoch: [133][ 930/ 1236] Overall Loss 0.176272 Objective Loss 0.176272 LR 0.000500 Time 0.021696 +2023-10-02 21:33:17,067 - Epoch: [133][ 940/ 1236] Overall Loss 0.176043 Objective Loss 0.176043 LR 0.000500 Time 0.021689 +2023-10-02 21:33:17,276 - Epoch: [133][ 950/ 1236] Overall Loss 0.176041 Objective Loss 0.176041 LR 0.000500 Time 0.021678 +2023-10-02 21:33:17,485 - Epoch: [133][ 960/ 1236] Overall Loss 0.176343 Objective Loss 0.176343 LR 0.000500 Time 0.021670 +2023-10-02 21:33:17,693 - Epoch: [133][ 970/ 1236] Overall Loss 0.176376 Objective Loss 0.176376 LR 0.000500 Time 0.021660 +2023-10-02 21:33:17,903 - Epoch: [133][ 980/ 1236] Overall Loss 0.176512 Objective Loss 0.176512 LR 0.000500 Time 0.021652 +2023-10-02 21:33:18,111 - Epoch: [133][ 990/ 1236] Overall Loss 0.176559 Objective Loss 0.176559 LR 0.000500 Time 0.021642 +2023-10-02 21:33:18,320 - Epoch: [133][ 1000/ 1236] Overall Loss 0.176761 Objective Loss 0.176761 LR 0.000500 Time 0.021635 +2023-10-02 21:33:18,528 - Epoch: [133][ 1010/ 1236] Overall Loss 0.176589 Objective Loss 0.176589 LR 0.000500 Time 0.021625 +2023-10-02 21:33:18,738 - Epoch: [133][ 1020/ 1236] Overall Loss 0.176634 Objective Loss 0.176634 LR 0.000500 Time 0.021618 +2023-10-02 21:33:18,946 - Epoch: [133][ 1030/ 1236] Overall Loss 0.176774 Objective Loss 0.176774 LR 0.000500 Time 0.021609 +2023-10-02 21:33:19,156 - Epoch: [133][ 1040/ 1236] Overall Loss 0.176741 Objective Loss 0.176741 LR 0.000500 Time 0.021603 +2023-10-02 21:33:19,364 - Epoch: [133][ 1050/ 1236] Overall Loss 0.176798 Objective Loss 0.176798 LR 0.000500 Time 0.021594 +2023-10-02 21:33:19,573 - Epoch: [133][ 1060/ 1236] Overall Loss 0.176744 Objective Loss 0.176744 LR 0.000500 Time 0.021587 +2023-10-02 21:33:19,781 - Epoch: [133][ 1070/ 1236] Overall Loss 0.176774 Objective Loss 0.176774 LR 0.000500 Time 0.021579 +2023-10-02 21:33:19,991 - Epoch: [133][ 1080/ 1236] Overall Loss 0.176797 Objective Loss 0.176797 LR 0.000500 Time 0.021573 +2023-10-02 21:33:20,199 - Epoch: [133][ 1090/ 1236] Overall Loss 0.176935 Objective Loss 0.176935 LR 0.000500 Time 0.021564 +2023-10-02 21:33:20,409 - Epoch: [133][ 1100/ 1236] Overall Loss 0.177112 Objective Loss 0.177112 LR 0.000500 Time 0.021558 +2023-10-02 21:33:20,617 - Epoch: [133][ 1110/ 1236] Overall Loss 0.177084 Objective Loss 0.177084 LR 0.000500 Time 0.021550 +2023-10-02 21:33:20,826 - Epoch: [133][ 1120/ 1236] Overall Loss 0.176918 Objective Loss 0.176918 LR 0.000500 Time 0.021545 +2023-10-02 21:33:21,035 - Epoch: [133][ 1130/ 1236] Overall Loss 0.176680 Objective Loss 0.176680 LR 0.000500 Time 0.021537 +2023-10-02 21:33:21,244 - Epoch: [133][ 1140/ 1236] Overall Loss 0.176599 Objective Loss 0.176599 LR 0.000500 Time 0.021532 +2023-10-02 21:33:21,452 - Epoch: [133][ 1150/ 1236] Overall Loss 0.176468 Objective Loss 0.176468 LR 0.000500 Time 0.021525 +2023-10-02 21:33:21,662 - Epoch: [133][ 1160/ 1236] Overall Loss 0.176417 Objective Loss 0.176417 LR 0.000500 Time 0.021519 +2023-10-02 21:33:21,870 - Epoch: [133][ 1170/ 1236] Overall Loss 0.176511 Objective Loss 0.176511 LR 0.000500 Time 0.021512 +2023-10-02 21:33:22,080 - Epoch: [133][ 1180/ 1236] Overall Loss 0.176456 Objective Loss 0.176456 LR 0.000500 Time 0.021507 +2023-10-02 21:33:22,288 - Epoch: [133][ 1190/ 1236] Overall Loss 0.176488 Objective Loss 0.176488 LR 0.000500 Time 0.021500 +2023-10-02 21:33:22,497 - Epoch: [133][ 1200/ 1236] Overall Loss 0.176301 Objective Loss 0.176301 LR 0.000500 Time 0.021495 +2023-10-02 21:33:22,706 - Epoch: [133][ 1210/ 1236] Overall Loss 0.176466 Objective Loss 0.176466 LR 0.000500 Time 0.021489 +2023-10-02 21:33:22,915 - Epoch: [133][ 1220/ 1236] Overall Loss 0.176519 Objective Loss 0.176519 LR 0.000500 Time 0.021484 +2023-10-02 21:33:23,180 - Epoch: [133][ 1230/ 1236] Overall Loss 0.176729 Objective Loss 0.176729 LR 0.000500 Time 0.021523 +2023-10-02 21:33:23,303 - Epoch: [133][ 1236/ 1236] Overall Loss 0.176822 Objective Loss 0.176822 Top1 87.983707 Top5 98.574338 LR 0.000500 Time 0.021518 +2023-10-02 21:33:23,431 - --- validate (epoch=133)----------- +2023-10-02 21:33:23,431 - 29943 samples (256 per mini-batch) +2023-10-02 21:33:23,925 - Epoch: [133][ 10/ 117] Loss 0.307433 Top1 87.265625 Top5 98.476562 +2023-10-02 21:33:24,078 - Epoch: [133][ 20/ 117] Loss 0.313873 Top1 86.152344 Top5 98.554688 +2023-10-02 21:33:24,228 - Epoch: [133][ 30/ 117] Loss 0.313667 Top1 85.989583 Top5 98.489583 +2023-10-02 21:33:24,380 - Epoch: [133][ 40/ 117] Loss 0.304699 Top1 86.171875 Top5 98.535156 +2023-10-02 21:33:24,530 - Epoch: [133][ 50/ 117] Loss 0.304300 Top1 86.140625 Top5 98.554688 +2023-10-02 21:33:24,681 - Epoch: [133][ 60/ 117] Loss 0.302856 Top1 86.217448 Top5 98.541667 +2023-10-02 21:33:24,831 - Epoch: [133][ 70/ 117] Loss 0.301525 Top1 86.244420 Top5 98.526786 +2023-10-02 21:33:24,987 - Epoch: [133][ 80/ 117] Loss 0.304255 Top1 86.113281 Top5 98.471680 +2023-10-02 21:33:25,143 - Epoch: [133][ 90/ 117] Loss 0.304957 Top1 86.076389 Top5 98.446181 +2023-10-02 21:33:25,303 - Epoch: [133][ 100/ 117] Loss 0.301075 Top1 86.238281 Top5 98.464844 +2023-10-02 21:33:25,469 - Epoch: [133][ 110/ 117] Loss 0.302178 Top1 86.285511 Top5 98.487216 +2023-10-02 21:33:25,559 - Epoch: [133][ 117/ 117] Loss 0.300837 Top1 86.230505 Top5 98.477107 +2023-10-02 21:33:25,697 - ==> Top1: 86.231 Top5: 98.477 Loss: 0.301 + +2023-10-02 21:33:25,698 - ==> Confusion: +[[ 939 0 3 0 8 3 0 0 4 66 2 1 1 2 3 0 2 1 0 0 15] + [ 0 1038 0 0 5 39 1 23 4 0 1 1 0 0 3 3 0 0 7 2 4] + [ 3 0 969 7 2 0 24 10 0 3 2 1 5 2 1 4 0 1 9 4 9] + [ 2 4 16 973 0 2 1 2 6 0 4 1 4 5 28 3 0 7 14 0 17] + [ 23 7 0 0 971 7 0 0 1 12 1 0 0 2 11 4 5 0 0 2 4] + [ 1 27 0 1 1 1011 1 18 1 4 4 4 1 12 5 1 3 1 4 2 14] + [ 0 7 25 0 0 2 1128 7 0 0 3 1 0 0 0 2 0 2 1 7 6] + [ 2 15 9 0 4 27 6 1080 1 4 3 2 2 6 1 0 0 2 35 8 11] + [ 15 2 0 1 3 2 0 0 978 40 8 2 1 11 14 2 1 3 2 0 4] + [ 79 2 1 0 7 2 0 0 22 973 0 0 0 16 7 1 0 0 0 2 7] + [ 3 0 10 5 0 2 4 4 14 1 972 1 0 11 6 1 1 2 4 1 11] + [ 0 1 3 0 0 17 0 3 0 0 0 972 12 4 0 1 0 14 0 4 4] + [ 0 0 2 4 0 2 3 1 0 0 3 43 958 1 2 8 1 17 2 7 14] + [ 1 0 0 0 3 11 0 0 8 6 4 5 0 1058 6 0 0 1 0 3 13] + [ 14 1 3 18 2 0 0 0 23 1 3 0 3 0 1013 0 1 2 9 0 8] + [ 0 0 1 1 7 0 0 0 0 0 0 6 7 0 0 1066 15 15 3 7 6] + [ 1 13 0 0 4 8 1 1 0 0 0 4 0 4 7 10 1090 0 0 6 12] + [ 0 0 1 0 0 0 2 0 0 0 0 4 17 1 2 8 0 1000 0 0 3] + [ 1 3 5 12 0 1 1 27 8 1 3 1 1 0 16 0 0 0 975 0 13] + [ 0 0 3 2 1 5 8 6 0 0 0 9 2 1 0 0 6 0 1 1103 5] + [ 118 124 113 85 72 139 41 84 96 79 154 105 272 264 123 55 74 77 104 173 5553]] + +2023-10-02 21:33:25,699 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:33:25,699 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:33:25,706 - + +2023-10-02 21:33:25,706 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:33:26,854 - Epoch: [134][ 10/ 1236] Overall Loss 0.166156 Objective Loss 0.166156 LR 0.000500 Time 0.114773 +2023-10-02 21:33:27,067 - Epoch: [134][ 20/ 1236] Overall Loss 0.164695 Objective Loss 0.164695 LR 0.000500 Time 0.068001 +2023-10-02 21:33:27,277 - Epoch: [134][ 30/ 1236] Overall Loss 0.162665 Objective Loss 0.162665 LR 0.000500 Time 0.052331 +2023-10-02 21:33:27,490 - Epoch: [134][ 40/ 1236] Overall Loss 0.160035 Objective Loss 0.160035 LR 0.000500 Time 0.044555 +2023-10-02 21:33:27,700 - Epoch: [134][ 50/ 1236] Overall Loss 0.162907 Objective Loss 0.162907 LR 0.000500 Time 0.039813 +2023-10-02 21:33:27,912 - Epoch: [134][ 60/ 1236] Overall Loss 0.160641 Objective Loss 0.160641 LR 0.000500 Time 0.036705 +2023-10-02 21:33:28,122 - Epoch: [134][ 70/ 1236] Overall Loss 0.164140 Objective Loss 0.164140 LR 0.000500 Time 0.034444 +2023-10-02 21:33:28,334 - Epoch: [134][ 80/ 1236] Overall Loss 0.160809 Objective Loss 0.160809 LR 0.000500 Time 0.032789 +2023-10-02 21:33:28,544 - Epoch: [134][ 90/ 1236] Overall Loss 0.163275 Objective Loss 0.163275 LR 0.000500 Time 0.031478 +2023-10-02 21:33:28,756 - Epoch: [134][ 100/ 1236] Overall Loss 0.162911 Objective Loss 0.162911 LR 0.000500 Time 0.030440 +2023-10-02 21:33:28,966 - Epoch: [134][ 110/ 1236] Overall Loss 0.164491 Objective Loss 0.164491 LR 0.000500 Time 0.029581 +2023-10-02 21:33:29,177 - Epoch: [134][ 120/ 1236] Overall Loss 0.164770 Objective Loss 0.164770 LR 0.000500 Time 0.028869 +2023-10-02 21:33:29,386 - Epoch: [134][ 130/ 1236] Overall Loss 0.166312 Objective Loss 0.166312 LR 0.000500 Time 0.028260 +2023-10-02 21:33:29,598 - Epoch: [134][ 140/ 1236] Overall Loss 0.165973 Objective Loss 0.165973 LR 0.000500 Time 0.027746 +2023-10-02 21:33:29,807 - Epoch: [134][ 150/ 1236] Overall Loss 0.167296 Objective Loss 0.167296 LR 0.000500 Time 0.027292 +2023-10-02 21:33:30,019 - Epoch: [134][ 160/ 1236] Overall Loss 0.166964 Objective Loss 0.166964 LR 0.000500 Time 0.026906 +2023-10-02 21:33:30,229 - Epoch: [134][ 170/ 1236] Overall Loss 0.166088 Objective Loss 0.166088 LR 0.000500 Time 0.026556 +2023-10-02 21:33:30,440 - Epoch: [134][ 180/ 1236] Overall Loss 0.165495 Objective Loss 0.165495 LR 0.000500 Time 0.026251 +2023-10-02 21:33:30,649 - Epoch: [134][ 190/ 1236] Overall Loss 0.165666 Objective Loss 0.165666 LR 0.000500 Time 0.025972 +2023-10-02 21:33:30,862 - Epoch: [134][ 200/ 1236] Overall Loss 0.164973 Objective Loss 0.164973 LR 0.000500 Time 0.025735 +2023-10-02 21:33:31,076 - Epoch: [134][ 210/ 1236] Overall Loss 0.165787 Objective Loss 0.165787 LR 0.000500 Time 0.025501 +2023-10-02 21:33:31,289 - Epoch: [134][ 220/ 1236] Overall Loss 0.167683 Objective Loss 0.167683 LR 0.000500 Time 0.025307 +2023-10-02 21:33:31,498 - Epoch: [134][ 230/ 1236] Overall Loss 0.167159 Objective Loss 0.167159 LR 0.000500 Time 0.025113 +2023-10-02 21:33:31,709 - Epoch: [134][ 240/ 1236] Overall Loss 0.167036 Objective Loss 0.167036 LR 0.000500 Time 0.024945 +2023-10-02 21:33:31,919 - Epoch: [134][ 250/ 1236] Overall Loss 0.167630 Objective Loss 0.167630 LR 0.000500 Time 0.024786 +2023-10-02 21:33:32,130 - Epoch: [134][ 260/ 1236] Overall Loss 0.168440 Objective Loss 0.168440 LR 0.000500 Time 0.024643 +2023-10-02 21:33:32,340 - Epoch: [134][ 270/ 1236] Overall Loss 0.168224 Objective Loss 0.168224 LR 0.000500 Time 0.024506 +2023-10-02 21:33:32,552 - Epoch: [134][ 280/ 1236] Overall Loss 0.168244 Objective Loss 0.168244 LR 0.000500 Time 0.024388 +2023-10-02 21:33:32,761 - Epoch: [134][ 290/ 1236] Overall Loss 0.167961 Objective Loss 0.167961 LR 0.000500 Time 0.024266 +2023-10-02 21:33:32,973 - Epoch: [134][ 300/ 1236] Overall Loss 0.168440 Objective Loss 0.168440 LR 0.000500 Time 0.024163 +2023-10-02 21:33:33,180 - Epoch: [134][ 310/ 1236] Overall Loss 0.167874 Objective Loss 0.167874 LR 0.000500 Time 0.024052 +2023-10-02 21:33:33,392 - Epoch: [134][ 320/ 1236] Overall Loss 0.167872 Objective Loss 0.167872 LR 0.000500 Time 0.023961 +2023-10-02 21:33:33,600 - Epoch: [134][ 330/ 1236] Overall Loss 0.167656 Objective Loss 0.167656 LR 0.000500 Time 0.023864 +2023-10-02 21:33:33,811 - Epoch: [134][ 340/ 1236] Overall Loss 0.167633 Objective Loss 0.167633 LR 0.000500 Time 0.023783 +2023-10-02 21:33:34,019 - Epoch: [134][ 350/ 1236] Overall Loss 0.167682 Objective Loss 0.167682 LR 0.000500 Time 0.023696 +2023-10-02 21:33:34,231 - Epoch: [134][ 360/ 1236] Overall Loss 0.167849 Objective Loss 0.167849 LR 0.000500 Time 0.023626 +2023-10-02 21:33:34,439 - Epoch: [134][ 370/ 1236] Overall Loss 0.167599 Objective Loss 0.167599 LR 0.000500 Time 0.023548 +2023-10-02 21:33:34,650 - Epoch: [134][ 380/ 1236] Overall Loss 0.167689 Objective Loss 0.167689 LR 0.000500 Time 0.023484 +2023-10-02 21:33:34,858 - Epoch: [134][ 390/ 1236] Overall Loss 0.167467 Objective Loss 0.167467 LR 0.000500 Time 0.023414 +2023-10-02 21:33:35,069 - Epoch: [134][ 400/ 1236] Overall Loss 0.167629 Objective Loss 0.167629 LR 0.000500 Time 0.023356 +2023-10-02 21:33:35,277 - Epoch: [134][ 410/ 1236] Overall Loss 0.167925 Objective Loss 0.167925 LR 0.000500 Time 0.023292 +2023-10-02 21:33:35,489 - Epoch: [134][ 420/ 1236] Overall Loss 0.168203 Objective Loss 0.168203 LR 0.000500 Time 0.023241 +2023-10-02 21:33:35,696 - Epoch: [134][ 430/ 1236] Overall Loss 0.168647 Objective Loss 0.168647 LR 0.000500 Time 0.023183 +2023-10-02 21:33:35,908 - Epoch: [134][ 440/ 1236] Overall Loss 0.168739 Objective Loss 0.168739 LR 0.000500 Time 0.023136 +2023-10-02 21:33:36,116 - Epoch: [134][ 450/ 1236] Overall Loss 0.168901 Objective Loss 0.168901 LR 0.000500 Time 0.023083 +2023-10-02 21:33:36,328 - Epoch: [134][ 460/ 1236] Overall Loss 0.169002 Objective Loss 0.169002 LR 0.000500 Time 0.023041 +2023-10-02 21:33:36,535 - Epoch: [134][ 470/ 1236] Overall Loss 0.168798 Objective Loss 0.168798 LR 0.000500 Time 0.022992 +2023-10-02 21:33:36,747 - Epoch: [134][ 480/ 1236] Overall Loss 0.168856 Objective Loss 0.168856 LR 0.000500 Time 0.022953 +2023-10-02 21:33:36,954 - Epoch: [134][ 490/ 1236] Overall Loss 0.169322 Objective Loss 0.169322 LR 0.000500 Time 0.022908 +2023-10-02 21:33:37,166 - Epoch: [134][ 500/ 1236] Overall Loss 0.169420 Objective Loss 0.169420 LR 0.000500 Time 0.022872 +2023-10-02 21:33:37,374 - Epoch: [134][ 510/ 1236] Overall Loss 0.169664 Objective Loss 0.169664 LR 0.000500 Time 0.022830 +2023-10-02 21:33:37,586 - Epoch: [134][ 520/ 1236] Overall Loss 0.169705 Objective Loss 0.169705 LR 0.000500 Time 0.022798 +2023-10-02 21:33:37,793 - Epoch: [134][ 530/ 1236] Overall Loss 0.169479 Objective Loss 0.169479 LR 0.000500 Time 0.022759 +2023-10-02 21:33:38,005 - Epoch: [134][ 540/ 1236] Overall Loss 0.169906 Objective Loss 0.169906 LR 0.000500 Time 0.022729 +2023-10-02 21:33:38,213 - Epoch: [134][ 550/ 1236] Overall Loss 0.170150 Objective Loss 0.170150 LR 0.000500 Time 0.022693 +2023-10-02 21:33:38,425 - Epoch: [134][ 560/ 1236] Overall Loss 0.170364 Objective Loss 0.170364 LR 0.000500 Time 0.022666 +2023-10-02 21:33:38,633 - Epoch: [134][ 570/ 1236] Overall Loss 0.170359 Objective Loss 0.170359 LR 0.000500 Time 0.022633 +2023-10-02 21:33:38,845 - Epoch: [134][ 580/ 1236] Overall Loss 0.170216 Objective Loss 0.170216 LR 0.000500 Time 0.022608 +2023-10-02 21:33:39,053 - Epoch: [134][ 590/ 1236] Overall Loss 0.170290 Objective Loss 0.170290 LR 0.000500 Time 0.022576 +2023-10-02 21:33:39,264 - Epoch: [134][ 600/ 1236] Overall Loss 0.170438 Objective Loss 0.170438 LR 0.000500 Time 0.022551 +2023-10-02 21:33:39,473 - Epoch: [134][ 610/ 1236] Overall Loss 0.170316 Objective Loss 0.170316 LR 0.000500 Time 0.022521 +2023-10-02 21:33:39,684 - Epoch: [134][ 620/ 1236] Overall Loss 0.170241 Objective Loss 0.170241 LR 0.000500 Time 0.022499 +2023-10-02 21:33:39,892 - Epoch: [134][ 630/ 1236] Overall Loss 0.170577 Objective Loss 0.170577 LR 0.000500 Time 0.022471 +2023-10-02 21:33:40,104 - Epoch: [134][ 640/ 1236] Overall Loss 0.170671 Objective Loss 0.170671 LR 0.000500 Time 0.022451 +2023-10-02 21:33:40,312 - Epoch: [134][ 650/ 1236] Overall Loss 0.170722 Objective Loss 0.170722 LR 0.000500 Time 0.022425 +2023-10-02 21:33:40,524 - Epoch: [134][ 660/ 1236] Overall Loss 0.171014 Objective Loss 0.171014 LR 0.000500 Time 0.022406 +2023-10-02 21:33:40,732 - Epoch: [134][ 670/ 1236] Overall Loss 0.171001 Objective Loss 0.171001 LR 0.000500 Time 0.022381 +2023-10-02 21:33:40,944 - Epoch: [134][ 680/ 1236] Overall Loss 0.171119 Objective Loss 0.171119 LR 0.000500 Time 0.022363 +2023-10-02 21:33:41,152 - Epoch: [134][ 690/ 1236] Overall Loss 0.171044 Objective Loss 0.171044 LR 0.000500 Time 0.022340 +2023-10-02 21:33:41,364 - Epoch: [134][ 700/ 1236] Overall Loss 0.170812 Objective Loss 0.170812 LR 0.000500 Time 0.022324 +2023-10-02 21:33:41,572 - Epoch: [134][ 710/ 1236] Overall Loss 0.170930 Objective Loss 0.170930 LR 0.000500 Time 0.022302 +2023-10-02 21:33:41,784 - Epoch: [134][ 720/ 1236] Overall Loss 0.171076 Objective Loss 0.171076 LR 0.000500 Time 0.022286 +2023-10-02 21:33:41,992 - Epoch: [134][ 730/ 1236] Overall Loss 0.171121 Objective Loss 0.171121 LR 0.000500 Time 0.022265 +2023-10-02 21:33:42,203 - Epoch: [134][ 740/ 1236] Overall Loss 0.171254 Objective Loss 0.171254 LR 0.000500 Time 0.022250 +2023-10-02 21:33:42,411 - Epoch: [134][ 750/ 1236] Overall Loss 0.171317 Objective Loss 0.171317 LR 0.000500 Time 0.022230 +2023-10-02 21:33:42,623 - Epoch: [134][ 760/ 1236] Overall Loss 0.171283 Objective Loss 0.171283 LR 0.000500 Time 0.022216 +2023-10-02 21:33:42,831 - Epoch: [134][ 770/ 1236] Overall Loss 0.171573 Objective Loss 0.171573 LR 0.000500 Time 0.022197 +2023-10-02 21:33:43,043 - Epoch: [134][ 780/ 1236] Overall Loss 0.171792 Objective Loss 0.171792 LR 0.000500 Time 0.022184 +2023-10-02 21:33:43,251 - Epoch: [134][ 790/ 1236] Overall Loss 0.171826 Objective Loss 0.171826 LR 0.000500 Time 0.022166 +2023-10-02 21:33:43,463 - Epoch: [134][ 800/ 1236] Overall Loss 0.171734 Objective Loss 0.171734 LR 0.000500 Time 0.022153 +2023-10-02 21:33:43,671 - Epoch: [134][ 810/ 1236] Overall Loss 0.171862 Objective Loss 0.171862 LR 0.000500 Time 0.022136 +2023-10-02 21:33:43,883 - Epoch: [134][ 820/ 1236] Overall Loss 0.171862 Objective Loss 0.171862 LR 0.000500 Time 0.022124 +2023-10-02 21:33:44,090 - Epoch: [134][ 830/ 1236] Overall Loss 0.171944 Objective Loss 0.171944 LR 0.000500 Time 0.022107 +2023-10-02 21:33:44,302 - Epoch: [134][ 840/ 1236] Overall Loss 0.172227 Objective Loss 0.172227 LR 0.000500 Time 0.022096 +2023-10-02 21:33:44,510 - Epoch: [134][ 850/ 1236] Overall Loss 0.172211 Objective Loss 0.172211 LR 0.000500 Time 0.022080 +2023-10-02 21:33:44,722 - Epoch: [134][ 860/ 1236] Overall Loss 0.172262 Objective Loss 0.172262 LR 0.000500 Time 0.022069 +2023-10-02 21:33:44,930 - Epoch: [134][ 870/ 1236] Overall Loss 0.172479 Objective Loss 0.172479 LR 0.000500 Time 0.022054 +2023-10-02 21:33:45,142 - Epoch: [134][ 880/ 1236] Overall Loss 0.172671 Objective Loss 0.172671 LR 0.000500 Time 0.022044 +2023-10-02 21:33:45,350 - Epoch: [134][ 890/ 1236] Overall Loss 0.172787 Objective Loss 0.172787 LR 0.000500 Time 0.022030 +2023-10-02 21:33:45,562 - Epoch: [134][ 900/ 1236] Overall Loss 0.172775 Objective Loss 0.172775 LR 0.000500 Time 0.022021 +2023-10-02 21:33:45,770 - Epoch: [134][ 910/ 1236] Overall Loss 0.172881 Objective Loss 0.172881 LR 0.000500 Time 0.022007 +2023-10-02 21:33:45,982 - Epoch: [134][ 920/ 1236] Overall Loss 0.172949 Objective Loss 0.172949 LR 0.000500 Time 0.021998 +2023-10-02 21:33:46,190 - Epoch: [134][ 930/ 1236] Overall Loss 0.172897 Objective Loss 0.172897 LR 0.000500 Time 0.021985 +2023-10-02 21:33:46,402 - Epoch: [134][ 940/ 1236] Overall Loss 0.172916 Objective Loss 0.172916 LR 0.000500 Time 0.021976 +2023-10-02 21:33:46,610 - Epoch: [134][ 950/ 1236] Overall Loss 0.172918 Objective Loss 0.172918 LR 0.000500 Time 0.021963 +2023-10-02 21:33:46,822 - Epoch: [134][ 960/ 1236] Overall Loss 0.172955 Objective Loss 0.172955 LR 0.000500 Time 0.021955 +2023-10-02 21:33:47,030 - Epoch: [134][ 970/ 1236] Overall Loss 0.172925 Objective Loss 0.172925 LR 0.000500 Time 0.021943 +2023-10-02 21:33:47,242 - Epoch: [134][ 980/ 1236] Overall Loss 0.172898 Objective Loss 0.172898 LR 0.000500 Time 0.021935 +2023-10-02 21:33:47,450 - Epoch: [134][ 990/ 1236] Overall Loss 0.172975 Objective Loss 0.172975 LR 0.000500 Time 0.021923 +2023-10-02 21:33:47,662 - Epoch: [134][ 1000/ 1236] Overall Loss 0.173312 Objective Loss 0.173312 LR 0.000500 Time 0.021916 +2023-10-02 21:33:47,870 - Epoch: [134][ 1010/ 1236] Overall Loss 0.173457 Objective Loss 0.173457 LR 0.000500 Time 0.021904 +2023-10-02 21:33:48,082 - Epoch: [134][ 1020/ 1236] Overall Loss 0.173648 Objective Loss 0.173648 LR 0.000500 Time 0.021897 +2023-10-02 21:33:48,290 - Epoch: [134][ 1030/ 1236] Overall Loss 0.173572 Objective Loss 0.173572 LR 0.000500 Time 0.021886 +2023-10-02 21:33:48,508 - Epoch: [134][ 1040/ 1236] Overall Loss 0.173601 Objective Loss 0.173601 LR 0.000500 Time 0.021884 +2023-10-02 21:33:48,723 - Epoch: [134][ 1050/ 1236] Overall Loss 0.173671 Objective Loss 0.173671 LR 0.000500 Time 0.021881 +2023-10-02 21:33:48,945 - Epoch: [134][ 1060/ 1236] Overall Loss 0.173887 Objective Loss 0.173887 LR 0.000500 Time 0.021884 +2023-10-02 21:33:49,161 - Epoch: [134][ 1070/ 1236] Overall Loss 0.173820 Objective Loss 0.173820 LR 0.000500 Time 0.021881 +2023-10-02 21:33:49,383 - Epoch: [134][ 1080/ 1236] Overall Loss 0.173972 Objective Loss 0.173972 LR 0.000500 Time 0.021883 +2023-10-02 21:33:49,599 - Epoch: [134][ 1090/ 1236] Overall Loss 0.174042 Objective Loss 0.174042 LR 0.000500 Time 0.021880 +2023-10-02 21:33:49,820 - Epoch: [134][ 1100/ 1236] Overall Loss 0.174243 Objective Loss 0.174243 LR 0.000500 Time 0.021882 +2023-10-02 21:33:50,036 - Epoch: [134][ 1110/ 1236] Overall Loss 0.174347 Objective Loss 0.174347 LR 0.000500 Time 0.021879 +2023-10-02 21:33:50,258 - Epoch: [134][ 1120/ 1236] Overall Loss 0.174360 Objective Loss 0.174360 LR 0.000500 Time 0.021881 +2023-10-02 21:33:50,474 - Epoch: [134][ 1130/ 1236] Overall Loss 0.174336 Objective Loss 0.174336 LR 0.000500 Time 0.021878 +2023-10-02 21:33:50,695 - Epoch: [134][ 1140/ 1236] Overall Loss 0.174514 Objective Loss 0.174514 LR 0.000500 Time 0.021880 +2023-10-02 21:33:50,911 - Epoch: [134][ 1150/ 1236] Overall Loss 0.174607 Objective Loss 0.174607 LR 0.000500 Time 0.021878 +2023-10-02 21:33:51,133 - Epoch: [134][ 1160/ 1236] Overall Loss 0.174803 Objective Loss 0.174803 LR 0.000500 Time 0.021880 +2023-10-02 21:33:51,349 - Epoch: [134][ 1170/ 1236] Overall Loss 0.174943 Objective Loss 0.174943 LR 0.000500 Time 0.021877 +2023-10-02 21:33:51,570 - Epoch: [134][ 1180/ 1236] Overall Loss 0.175017 Objective Loss 0.175017 LR 0.000500 Time 0.021879 +2023-10-02 21:33:51,783 - Epoch: [134][ 1190/ 1236] Overall Loss 0.175352 Objective Loss 0.175352 LR 0.000500 Time 0.021874 +2023-10-02 21:33:51,995 - Epoch: [134][ 1200/ 1236] Overall Loss 0.175496 Objective Loss 0.175496 LR 0.000500 Time 0.021868 +2023-10-02 21:33:52,203 - Epoch: [134][ 1210/ 1236] Overall Loss 0.175443 Objective Loss 0.175443 LR 0.000500 Time 0.021859 +2023-10-02 21:33:52,415 - Epoch: [134][ 1220/ 1236] Overall Loss 0.175351 Objective Loss 0.175351 LR 0.000500 Time 0.021854 +2023-10-02 21:33:52,678 - Epoch: [134][ 1230/ 1236] Overall Loss 0.175534 Objective Loss 0.175534 LR 0.000500 Time 0.021889 +2023-10-02 21:33:52,801 - Epoch: [134][ 1236/ 1236] Overall Loss 0.175596 Objective Loss 0.175596 Top1 90.020367 Top5 99.592668 LR 0.000500 Time 0.021882 +2023-10-02 21:33:52,947 - --- validate (epoch=134)----------- +2023-10-02 21:33:52,948 - 29943 samples (256 per mini-batch) +2023-10-02 21:33:53,451 - Epoch: [134][ 10/ 117] Loss 0.287555 Top1 86.757812 Top5 98.476562 +2023-10-02 21:33:53,605 - Epoch: [134][ 20/ 117] Loss 0.305850 Top1 86.503906 Top5 98.378906 +2023-10-02 21:33:53,757 - Epoch: [134][ 30/ 117] Loss 0.301389 Top1 86.393229 Top5 98.450521 +2023-10-02 21:33:53,908 - Epoch: [134][ 40/ 117] Loss 0.295919 Top1 86.240234 Top5 98.515625 +2023-10-02 21:33:54,063 - Epoch: [134][ 50/ 117] Loss 0.297057 Top1 86.015625 Top5 98.539062 +2023-10-02 21:33:54,217 - Epoch: [134][ 60/ 117] Loss 0.302502 Top1 85.904948 Top5 98.483073 +2023-10-02 21:33:54,373 - Epoch: [134][ 70/ 117] Loss 0.301529 Top1 85.864955 Top5 98.487723 +2023-10-02 21:33:54,527 - Epoch: [134][ 80/ 117] Loss 0.300609 Top1 85.893555 Top5 98.491211 +2023-10-02 21:33:54,683 - Epoch: [134][ 90/ 117] Loss 0.298798 Top1 85.963542 Top5 98.502604 +2023-10-02 21:33:54,837 - Epoch: [134][ 100/ 117] Loss 0.295676 Top1 86.007812 Top5 98.484375 +2023-10-02 21:33:55,000 - Epoch: [134][ 110/ 117] Loss 0.297660 Top1 85.980114 Top5 98.473011 +2023-10-02 21:33:55,090 - Epoch: [134][ 117/ 117] Loss 0.298755 Top1 85.949972 Top5 98.493805 +2023-10-02 21:33:55,228 - ==> Top1: 85.950 Top5: 98.494 Loss: 0.299 + +2023-10-02 21:33:55,229 - ==> Confusion: +[[ 952 1 3 0 4 2 0 0 8 47 1 0 1 1 6 2 2 0 1 0 19] + [ 0 1073 0 0 2 16 1 19 1 1 1 0 1 0 2 2 2 1 6 1 2] + [ 5 0 971 11 1 0 15 5 0 0 3 0 7 1 1 4 3 2 16 2 9] + [ 3 3 8 998 2 1 1 1 0 0 2 0 6 2 20 4 1 5 14 0 18] + [ 25 5 1 1 973 3 0 0 1 6 1 0 1 2 5 5 13 0 0 1 7] + [ 2 42 1 2 4 994 0 24 3 3 1 3 3 9 3 0 4 2 3 2 11] + [ 0 5 22 1 0 2 1136 5 0 0 5 0 0 0 0 3 0 0 0 7 5] + [ 1 19 6 0 5 21 8 1080 1 3 3 2 2 3 2 0 3 0 44 7 8] + [ 15 2 0 0 2 4 0 0 978 34 14 1 2 11 14 0 6 2 2 0 2] + [ 107 0 1 1 7 2 0 0 30 934 0 0 0 14 7 4 1 0 0 2 9] + [ 0 5 11 12 1 3 1 5 10 1 957 1 1 15 3 0 4 4 8 0 11] + [ 0 4 3 0 0 15 0 1 0 1 0 958 16 8 0 0 1 16 0 6 6] + [ 0 0 2 4 1 2 4 0 0 0 3 20 981 0 2 8 4 12 0 8 17] + [ 1 0 1 0 2 12 0 0 12 12 6 4 0 1048 3 1 0 1 0 2 14] + [ 16 0 5 19 6 1 0 0 15 1 4 0 2 3 1004 0 3 3 12 0 7] + [ 0 0 2 2 4 0 1 0 0 1 0 4 7 0 0 1072 18 9 2 5 7] + [ 0 20 1 0 7 5 0 0 0 0 0 4 0 1 4 8 1092 0 1 4 14] + [ 1 0 0 1 0 0 3 0 1 0 0 4 21 2 2 6 0 993 0 1 3] + [ 1 4 2 16 0 0 0 19 5 2 1 0 2 0 6 0 0 0 1002 0 8] + [ 0 4 3 1 1 3 7 7 0 0 1 10 4 3 1 1 7 1 1 1087 10] + [ 103 193 109 96 59 158 45 104 99 68 130 72 324 253 111 54 108 63 148 155 5453]] + +2023-10-02 21:33:55,230 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:33:55,230 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:33:55,236 - + +2023-10-02 21:33:55,236 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:33:56,267 - Epoch: [135][ 10/ 1236] Overall Loss 0.155870 Objective Loss 0.155870 LR 0.000500 Time 0.103053 +2023-10-02 21:33:56,479 - Epoch: [135][ 20/ 1236] Overall Loss 0.162595 Objective Loss 0.162595 LR 0.000500 Time 0.062084 +2023-10-02 21:33:56,688 - Epoch: [135][ 30/ 1236] Overall Loss 0.176300 Objective Loss 0.176300 LR 0.000500 Time 0.048318 +2023-10-02 21:33:56,900 - Epoch: [135][ 40/ 1236] Overall Loss 0.170801 Objective Loss 0.170801 LR 0.000500 Time 0.041530 +2023-10-02 21:33:57,108 - Epoch: [135][ 50/ 1236] Overall Loss 0.169534 Objective Loss 0.169534 LR 0.000500 Time 0.037373 +2023-10-02 21:33:57,320 - Epoch: [135][ 60/ 1236] Overall Loss 0.171547 Objective Loss 0.171547 LR 0.000500 Time 0.034674 +2023-10-02 21:33:57,528 - Epoch: [135][ 70/ 1236] Overall Loss 0.170476 Objective Loss 0.170476 LR 0.000500 Time 0.032686 +2023-10-02 21:33:57,740 - Epoch: [135][ 80/ 1236] Overall Loss 0.171069 Objective Loss 0.171069 LR 0.000500 Time 0.031248 +2023-10-02 21:33:57,947 - Epoch: [135][ 90/ 1236] Overall Loss 0.174414 Objective Loss 0.174414 LR 0.000500 Time 0.030076 +2023-10-02 21:33:58,157 - Epoch: [135][ 100/ 1236] Overall Loss 0.173407 Objective Loss 0.173407 LR 0.000500 Time 0.029157 +2023-10-02 21:33:58,364 - Epoch: [135][ 110/ 1236] Overall Loss 0.174282 Objective Loss 0.174282 LR 0.000500 Time 0.028380 +2023-10-02 21:33:58,577 - Epoch: [135][ 120/ 1236] Overall Loss 0.173771 Objective Loss 0.173771 LR 0.000500 Time 0.027782 +2023-10-02 21:33:58,790 - Epoch: [135][ 130/ 1236] Overall Loss 0.173055 Objective Loss 0.173055 LR 0.000500 Time 0.027272 +2023-10-02 21:33:59,003 - Epoch: [135][ 140/ 1236] Overall Loss 0.172939 Objective Loss 0.172939 LR 0.000500 Time 0.026836 +2023-10-02 21:33:59,216 - Epoch: [135][ 150/ 1236] Overall Loss 0.172523 Objective Loss 0.172523 LR 0.000500 Time 0.026458 +2023-10-02 21:33:59,430 - Epoch: [135][ 160/ 1236] Overall Loss 0.172218 Objective Loss 0.172218 LR 0.000500 Time 0.026127 +2023-10-02 21:33:59,643 - Epoch: [135][ 170/ 1236] Overall Loss 0.172976 Objective Loss 0.172976 LR 0.000500 Time 0.025840 +2023-10-02 21:33:59,857 - Epoch: [135][ 180/ 1236] Overall Loss 0.171737 Objective Loss 0.171737 LR 0.000500 Time 0.025582 +2023-10-02 21:34:00,069 - Epoch: [135][ 190/ 1236] Overall Loss 0.171074 Objective Loss 0.171074 LR 0.000500 Time 0.025344 +2023-10-02 21:34:00,282 - Epoch: [135][ 200/ 1236] Overall Loss 0.171983 Objective Loss 0.171983 LR 0.000500 Time 0.025142 +2023-10-02 21:34:00,495 - Epoch: [135][ 210/ 1236] Overall Loss 0.172759 Objective Loss 0.172759 LR 0.000500 Time 0.024948 +2023-10-02 21:34:00,709 - Epoch: [135][ 220/ 1236] Overall Loss 0.172473 Objective Loss 0.172473 LR 0.000500 Time 0.024777 +2023-10-02 21:34:00,921 - Epoch: [135][ 230/ 1236] Overall Loss 0.171640 Objective Loss 0.171640 LR 0.000500 Time 0.024616 +2023-10-02 21:34:01,134 - Epoch: [135][ 240/ 1236] Overall Loss 0.170706 Objective Loss 0.170706 LR 0.000500 Time 0.024473 +2023-10-02 21:34:01,347 - Epoch: [135][ 250/ 1236] Overall Loss 0.171251 Objective Loss 0.171251 LR 0.000500 Time 0.024338 +2023-10-02 21:34:01,561 - Epoch: [135][ 260/ 1236] Overall Loss 0.170656 Objective Loss 0.170656 LR 0.000500 Time 0.024216 +2023-10-02 21:34:01,773 - Epoch: [135][ 270/ 1236] Overall Loss 0.170629 Objective Loss 0.170629 LR 0.000500 Time 0.024099 +2023-10-02 21:34:01,986 - Epoch: [135][ 280/ 1236] Overall Loss 0.170589 Objective Loss 0.170589 LR 0.000500 Time 0.023993 +2023-10-02 21:34:02,199 - Epoch: [135][ 290/ 1236] Overall Loss 0.171465 Objective Loss 0.171465 LR 0.000500 Time 0.023900 +2023-10-02 21:34:02,416 - Epoch: [135][ 300/ 1236] Overall Loss 0.171135 Objective Loss 0.171135 LR 0.000500 Time 0.023826 +2023-10-02 21:34:02,630 - Epoch: [135][ 310/ 1236] Overall Loss 0.170714 Objective Loss 0.170714 LR 0.000500 Time 0.023743 +2023-10-02 21:34:02,846 - Epoch: [135][ 320/ 1236] Overall Loss 0.170950 Objective Loss 0.170950 LR 0.000500 Time 0.023677 +2023-10-02 21:34:03,060 - Epoch: [135][ 330/ 1236] Overall Loss 0.170955 Objective Loss 0.170955 LR 0.000500 Time 0.023605 +2023-10-02 21:34:03,276 - Epoch: [135][ 340/ 1236] Overall Loss 0.170925 Objective Loss 0.170925 LR 0.000500 Time 0.023546 +2023-10-02 21:34:03,485 - Epoch: [135][ 350/ 1236] Overall Loss 0.170763 Objective Loss 0.170763 LR 0.000500 Time 0.023470 +2023-10-02 21:34:03,699 - Epoch: [135][ 360/ 1236] Overall Loss 0.170553 Objective Loss 0.170553 LR 0.000500 Time 0.023410 +2023-10-02 21:34:03,908 - Epoch: [135][ 370/ 1236] Overall Loss 0.170382 Objective Loss 0.170382 LR 0.000500 Time 0.023342 +2023-10-02 21:34:04,122 - Epoch: [135][ 380/ 1236] Overall Loss 0.170323 Objective Loss 0.170323 LR 0.000500 Time 0.023288 +2023-10-02 21:34:04,331 - Epoch: [135][ 390/ 1236] Overall Loss 0.170151 Objective Loss 0.170151 LR 0.000500 Time 0.023226 +2023-10-02 21:34:04,544 - Epoch: [135][ 400/ 1236] Overall Loss 0.169925 Objective Loss 0.169925 LR 0.000500 Time 0.023179 +2023-10-02 21:34:04,753 - Epoch: [135][ 410/ 1236] Overall Loss 0.169711 Objective Loss 0.169711 LR 0.000500 Time 0.023123 +2023-10-02 21:34:04,966 - Epoch: [135][ 420/ 1236] Overall Loss 0.170414 Objective Loss 0.170414 LR 0.000500 Time 0.023079 +2023-10-02 21:34:05,176 - Epoch: [135][ 430/ 1236] Overall Loss 0.170396 Objective Loss 0.170396 LR 0.000500 Time 0.023028 +2023-10-02 21:34:05,388 - Epoch: [135][ 440/ 1236] Overall Loss 0.170821 Objective Loss 0.170821 LR 0.000500 Time 0.022987 +2023-10-02 21:34:05,598 - Epoch: [135][ 450/ 1236] Overall Loss 0.170414 Objective Loss 0.170414 LR 0.000500 Time 0.022940 +2023-10-02 21:34:05,812 - Epoch: [135][ 460/ 1236] Overall Loss 0.171098 Objective Loss 0.171098 LR 0.000500 Time 0.022905 +2023-10-02 21:34:06,021 - Epoch: [135][ 470/ 1236] Overall Loss 0.171236 Objective Loss 0.171236 LR 0.000500 Time 0.022861 +2023-10-02 21:34:06,234 - Epoch: [135][ 480/ 1236] Overall Loss 0.171342 Objective Loss 0.171342 LR 0.000500 Time 0.022829 +2023-10-02 21:34:06,443 - Epoch: [135][ 490/ 1236] Overall Loss 0.171607 Objective Loss 0.171607 LR 0.000500 Time 0.022789 +2023-10-02 21:34:06,657 - Epoch: [135][ 500/ 1236] Overall Loss 0.171840 Objective Loss 0.171840 LR 0.000500 Time 0.022760 +2023-10-02 21:34:06,866 - Epoch: [135][ 510/ 1236] Overall Loss 0.171692 Objective Loss 0.171692 LR 0.000500 Time 0.022723 +2023-10-02 21:34:07,079 - Epoch: [135][ 520/ 1236] Overall Loss 0.172294 Objective Loss 0.172294 LR 0.000500 Time 0.022696 +2023-10-02 21:34:07,289 - Epoch: [135][ 530/ 1236] Overall Loss 0.172279 Objective Loss 0.172279 LR 0.000500 Time 0.022662 +2023-10-02 21:34:07,502 - Epoch: [135][ 540/ 1236] Overall Loss 0.172232 Objective Loss 0.172232 LR 0.000500 Time 0.022637 +2023-10-02 21:34:07,711 - Epoch: [135][ 550/ 1236] Overall Loss 0.172126 Objective Loss 0.172126 LR 0.000500 Time 0.022605 +2023-10-02 21:34:07,925 - Epoch: [135][ 560/ 1236] Overall Loss 0.172133 Objective Loss 0.172133 LR 0.000500 Time 0.022583 +2023-10-02 21:34:08,134 - Epoch: [135][ 570/ 1236] Overall Loss 0.172353 Objective Loss 0.172353 LR 0.000500 Time 0.022553 +2023-10-02 21:34:08,348 - Epoch: [135][ 580/ 1236] Overall Loss 0.172111 Objective Loss 0.172111 LR 0.000500 Time 0.022531 +2023-10-02 21:34:08,557 - Epoch: [135][ 590/ 1236] Overall Loss 0.171953 Objective Loss 0.171953 LR 0.000500 Time 0.022504 +2023-10-02 21:34:08,770 - Epoch: [135][ 600/ 1236] Overall Loss 0.171690 Objective Loss 0.171690 LR 0.000500 Time 0.022484 +2023-10-02 21:34:08,980 - Epoch: [135][ 610/ 1236] Overall Loss 0.171772 Objective Loss 0.171772 LR 0.000500 Time 0.022458 +2023-10-02 21:34:09,193 - Epoch: [135][ 620/ 1236] Overall Loss 0.172110 Objective Loss 0.172110 LR 0.000500 Time 0.022440 +2023-10-02 21:34:09,403 - Epoch: [135][ 630/ 1236] Overall Loss 0.172082 Objective Loss 0.172082 LR 0.000500 Time 0.022415 +2023-10-02 21:34:09,616 - Epoch: [135][ 640/ 1236] Overall Loss 0.171918 Objective Loss 0.171918 LR 0.000500 Time 0.022399 +2023-10-02 21:34:09,826 - Epoch: [135][ 650/ 1236] Overall Loss 0.171563 Objective Loss 0.171563 LR 0.000500 Time 0.022376 +2023-10-02 21:34:10,040 - Epoch: [135][ 660/ 1236] Overall Loss 0.171493 Objective Loss 0.171493 LR 0.000500 Time 0.022360 +2023-10-02 21:34:10,249 - Epoch: [135][ 670/ 1236] Overall Loss 0.171599 Objective Loss 0.171599 LR 0.000500 Time 0.022338 +2023-10-02 21:34:10,462 - Epoch: [135][ 680/ 1236] Overall Loss 0.171861 Objective Loss 0.171861 LR 0.000500 Time 0.022323 +2023-10-02 21:34:10,672 - Epoch: [135][ 690/ 1236] Overall Loss 0.172411 Objective Loss 0.172411 LR 0.000500 Time 0.022303 +2023-10-02 21:34:10,885 - Epoch: [135][ 700/ 1236] Overall Loss 0.172060 Objective Loss 0.172060 LR 0.000500 Time 0.022289 +2023-10-02 21:34:11,094 - Epoch: [135][ 710/ 1236] Overall Loss 0.172177 Objective Loss 0.172177 LR 0.000500 Time 0.022269 +2023-10-02 21:34:11,308 - Epoch: [135][ 720/ 1236] Overall Loss 0.172184 Objective Loss 0.172184 LR 0.000500 Time 0.022256 +2023-10-02 21:34:11,518 - Epoch: [135][ 730/ 1236] Overall Loss 0.172497 Objective Loss 0.172497 LR 0.000500 Time 0.022238 +2023-10-02 21:34:11,731 - Epoch: [135][ 740/ 1236] Overall Loss 0.172725 Objective Loss 0.172725 LR 0.000500 Time 0.022225 +2023-10-02 21:34:11,941 - Epoch: [135][ 750/ 1236] Overall Loss 0.172692 Objective Loss 0.172692 LR 0.000500 Time 0.022208 +2023-10-02 21:34:12,154 - Epoch: [135][ 760/ 1236] Overall Loss 0.172742 Objective Loss 0.172742 LR 0.000500 Time 0.022197 +2023-10-02 21:34:12,364 - Epoch: [135][ 770/ 1236] Overall Loss 0.173109 Objective Loss 0.173109 LR 0.000500 Time 0.022180 +2023-10-02 21:34:12,577 - Epoch: [135][ 780/ 1236] Overall Loss 0.172852 Objective Loss 0.172852 LR 0.000500 Time 0.022169 +2023-10-02 21:34:12,787 - Epoch: [135][ 790/ 1236] Overall Loss 0.172844 Objective Loss 0.172844 LR 0.000500 Time 0.022153 +2023-10-02 21:34:13,000 - Epoch: [135][ 800/ 1236] Overall Loss 0.173067 Objective Loss 0.173067 LR 0.000500 Time 0.022143 +2023-10-02 21:34:13,210 - Epoch: [135][ 810/ 1236] Overall Loss 0.173398 Objective Loss 0.173398 LR 0.000500 Time 0.022128 +2023-10-02 21:34:13,424 - Epoch: [135][ 820/ 1236] Overall Loss 0.173663 Objective Loss 0.173663 LR 0.000500 Time 0.022118 +2023-10-02 21:34:13,633 - Epoch: [135][ 830/ 1236] Overall Loss 0.173568 Objective Loss 0.173568 LR 0.000500 Time 0.022103 +2023-10-02 21:34:13,846 - Epoch: [135][ 840/ 1236] Overall Loss 0.173660 Objective Loss 0.173660 LR 0.000500 Time 0.022094 +2023-10-02 21:34:14,056 - Epoch: [135][ 850/ 1236] Overall Loss 0.173697 Objective Loss 0.173697 LR 0.000500 Time 0.022080 +2023-10-02 21:34:14,269 - Epoch: [135][ 860/ 1236] Overall Loss 0.173697 Objective Loss 0.173697 LR 0.000500 Time 0.022071 +2023-10-02 21:34:14,479 - Epoch: [135][ 870/ 1236] Overall Loss 0.173788 Objective Loss 0.173788 LR 0.000500 Time 0.022058 +2023-10-02 21:34:14,693 - Epoch: [135][ 880/ 1236] Overall Loss 0.173916 Objective Loss 0.173916 LR 0.000500 Time 0.022050 +2023-10-02 21:34:14,902 - Epoch: [135][ 890/ 1236] Overall Loss 0.173929 Objective Loss 0.173929 LR 0.000500 Time 0.022037 +2023-10-02 21:34:15,115 - Epoch: [135][ 900/ 1236] Overall Loss 0.174117 Objective Loss 0.174117 LR 0.000500 Time 0.022028 +2023-10-02 21:34:15,325 - Epoch: [135][ 910/ 1236] Overall Loss 0.173982 Objective Loss 0.173982 LR 0.000500 Time 0.022016 +2023-10-02 21:34:15,539 - Epoch: [135][ 920/ 1236] Overall Loss 0.174007 Objective Loss 0.174007 LR 0.000500 Time 0.022009 +2023-10-02 21:34:15,749 - Epoch: [135][ 930/ 1236] Overall Loss 0.174152 Objective Loss 0.174152 LR 0.000500 Time 0.021998 +2023-10-02 21:34:15,963 - Epoch: [135][ 940/ 1236] Overall Loss 0.174104 Objective Loss 0.174104 LR 0.000500 Time 0.021991 +2023-10-02 21:34:16,172 - Epoch: [135][ 950/ 1236] Overall Loss 0.174308 Objective Loss 0.174308 LR 0.000500 Time 0.021979 +2023-10-02 21:34:16,386 - Epoch: [135][ 960/ 1236] Overall Loss 0.174412 Objective Loss 0.174412 LR 0.000500 Time 0.021973 +2023-10-02 21:34:16,596 - Epoch: [135][ 970/ 1236] Overall Loss 0.174416 Objective Loss 0.174416 LR 0.000500 Time 0.021962 +2023-10-02 21:34:16,810 - Epoch: [135][ 980/ 1236] Overall Loss 0.174484 Objective Loss 0.174484 LR 0.000500 Time 0.021956 +2023-10-02 21:34:17,020 - Epoch: [135][ 990/ 1236] Overall Loss 0.174735 Objective Loss 0.174735 LR 0.000500 Time 0.021946 +2023-10-02 21:34:17,233 - Epoch: [135][ 1000/ 1236] Overall Loss 0.174847 Objective Loss 0.174847 LR 0.000500 Time 0.021940 +2023-10-02 21:34:17,443 - Epoch: [135][ 1010/ 1236] Overall Loss 0.174996 Objective Loss 0.174996 LR 0.000500 Time 0.021930 +2023-10-02 21:34:17,657 - Epoch: [135][ 1020/ 1236] Overall Loss 0.174981 Objective Loss 0.174981 LR 0.000500 Time 0.021924 +2023-10-02 21:34:17,867 - Epoch: [135][ 1030/ 1236] Overall Loss 0.174998 Objective Loss 0.174998 LR 0.000500 Time 0.021915 +2023-10-02 21:34:18,080 - Epoch: [135][ 1040/ 1236] Overall Loss 0.175011 Objective Loss 0.175011 LR 0.000500 Time 0.021909 +2023-10-02 21:34:18,290 - Epoch: [135][ 1050/ 1236] Overall Loss 0.174980 Objective Loss 0.174980 LR 0.000500 Time 0.021900 +2023-10-02 21:34:18,503 - Epoch: [135][ 1060/ 1236] Overall Loss 0.174980 Objective Loss 0.174980 LR 0.000500 Time 0.021894 +2023-10-02 21:34:18,714 - Epoch: [135][ 1070/ 1236] Overall Loss 0.174876 Objective Loss 0.174876 LR 0.000500 Time 0.021885 +2023-10-02 21:34:18,928 - Epoch: [135][ 1080/ 1236] Overall Loss 0.174922 Objective Loss 0.174922 LR 0.000500 Time 0.021880 +2023-10-02 21:34:19,137 - Epoch: [135][ 1090/ 1236] Overall Loss 0.174887 Objective Loss 0.174887 LR 0.000500 Time 0.021871 +2023-10-02 21:34:19,350 - Epoch: [135][ 1100/ 1236] Overall Loss 0.174862 Objective Loss 0.174862 LR 0.000500 Time 0.021865 +2023-10-02 21:34:19,560 - Epoch: [135][ 1110/ 1236] Overall Loss 0.175005 Objective Loss 0.175005 LR 0.000500 Time 0.021857 +2023-10-02 21:34:19,774 - Epoch: [135][ 1120/ 1236] Overall Loss 0.174804 Objective Loss 0.174804 LR 0.000500 Time 0.021852 +2023-10-02 21:34:19,984 - Epoch: [135][ 1130/ 1236] Overall Loss 0.174907 Objective Loss 0.174907 LR 0.000500 Time 0.021844 +2023-10-02 21:34:20,197 - Epoch: [135][ 1140/ 1236] Overall Loss 0.174900 Objective Loss 0.174900 LR 0.000500 Time 0.021839 +2023-10-02 21:34:20,408 - Epoch: [135][ 1150/ 1236] Overall Loss 0.174917 Objective Loss 0.174917 LR 0.000500 Time 0.021831 +2023-10-02 21:34:20,622 - Epoch: [135][ 1160/ 1236] Overall Loss 0.174849 Objective Loss 0.174849 LR 0.000500 Time 0.021827 +2023-10-02 21:34:20,831 - Epoch: [135][ 1170/ 1236] Overall Loss 0.174988 Objective Loss 0.174988 LR 0.000500 Time 0.021819 +2023-10-02 21:34:21,044 - Epoch: [135][ 1180/ 1236] Overall Loss 0.175270 Objective Loss 0.175270 LR 0.000500 Time 0.021814 +2023-10-02 21:34:21,254 - Epoch: [135][ 1190/ 1236] Overall Loss 0.175488 Objective Loss 0.175488 LR 0.000500 Time 0.021807 +2023-10-02 21:34:21,469 - Epoch: [135][ 1200/ 1236] Overall Loss 0.175582 Objective Loss 0.175582 LR 0.000500 Time 0.021803 +2023-10-02 21:34:21,678 - Epoch: [135][ 1210/ 1236] Overall Loss 0.175562 Objective Loss 0.175562 LR 0.000500 Time 0.021796 +2023-10-02 21:34:21,892 - Epoch: [135][ 1220/ 1236] Overall Loss 0.175565 Objective Loss 0.175565 LR 0.000500 Time 0.021793 +2023-10-02 21:34:22,158 - Epoch: [135][ 1230/ 1236] Overall Loss 0.175856 Objective Loss 0.175856 LR 0.000500 Time 0.021831 +2023-10-02 21:34:22,280 - Epoch: [135][ 1236/ 1236] Overall Loss 0.175761 Objective Loss 0.175761 Top1 89.816701 Top5 98.574338 LR 0.000500 Time 0.021824 +2023-10-02 21:34:22,420 - --- validate (epoch=135)----------- +2023-10-02 21:34:22,420 - 29943 samples (256 per mini-batch) +2023-10-02 21:34:22,919 - Epoch: [135][ 10/ 117] Loss 0.314630 Top1 84.921875 Top5 98.007812 +2023-10-02 21:34:23,071 - Epoch: [135][ 20/ 117] Loss 0.319110 Top1 85.000000 Top5 97.890625 +2023-10-02 21:34:23,223 - Epoch: [135][ 30/ 117] Loss 0.307949 Top1 85.455729 Top5 98.072917 +2023-10-02 21:34:23,375 - Epoch: [135][ 40/ 117] Loss 0.309450 Top1 85.332031 Top5 98.144531 +2023-10-02 21:34:23,526 - Epoch: [135][ 50/ 117] Loss 0.314841 Top1 85.296875 Top5 98.148438 +2023-10-02 21:34:23,678 - Epoch: [135][ 60/ 117] Loss 0.309879 Top1 85.325521 Top5 98.196615 +2023-10-02 21:34:23,831 - Epoch: [135][ 70/ 117] Loss 0.308538 Top1 85.373884 Top5 98.258929 +2023-10-02 21:34:23,983 - Epoch: [135][ 80/ 117] Loss 0.305290 Top1 85.410156 Top5 98.330078 +2023-10-02 21:34:24,135 - Epoch: [135][ 90/ 117] Loss 0.305990 Top1 85.308160 Top5 98.333333 +2023-10-02 21:34:24,287 - Epoch: [135][ 100/ 117] Loss 0.302954 Top1 85.421875 Top5 98.351562 +2023-10-02 21:34:24,447 - Epoch: [135][ 110/ 117] Loss 0.300937 Top1 85.422585 Top5 98.380682 +2023-10-02 21:34:24,537 - Epoch: [135][ 117/ 117] Loss 0.299833 Top1 85.482417 Top5 98.406973 +2023-10-02 21:34:24,675 - ==> Top1: 85.482 Top5: 98.407 Loss: 0.300 + +2023-10-02 21:34:24,676 - ==> Confusion: +[[ 943 1 4 0 10 3 0 1 2 56 2 0 1 0 3 1 3 0 2 0 18] + [ 1 1071 0 0 4 20 2 14 1 0 1 1 1 0 1 3 0 0 9 1 1] + [ 4 1 969 13 2 0 16 6 0 1 4 1 7 3 0 3 0 1 13 2 10] + [ 2 3 10 1005 0 0 0 1 2 0 5 1 3 1 15 5 1 1 13 0 21] + [ 27 5 1 0 965 4 0 1 0 12 2 1 0 5 8 6 7 0 1 0 5] + [ 3 32 1 1 1 999 1 25 1 4 1 7 2 10 5 0 4 0 3 4 12] + [ 0 4 22 1 0 1 1132 5 0 0 2 1 0 1 0 6 0 1 3 6 6] + [ 3 16 14 1 8 24 5 1054 2 3 2 7 3 5 2 1 0 0 53 7 8] + [ 16 2 0 1 1 1 0 1 972 43 8 2 1 12 20 2 1 0 4 1 1] + [ 84 1 3 1 7 3 1 1 28 949 0 1 0 18 12 1 0 0 0 1 8] + [ 3 2 10 11 0 2 4 4 11 0 955 1 1 16 6 0 2 2 11 2 10] + [ 0 4 1 0 0 7 0 4 0 1 0 954 31 4 0 7 0 14 0 4 4] + [ 0 0 1 4 0 0 1 0 0 0 3 34 983 1 5 5 2 14 1 3 11] + [ 1 0 1 0 1 8 0 0 9 12 1 8 0 1064 3 0 1 1 0 2 7] + [ 13 1 5 23 3 1 0 0 14 1 3 0 1 4 1015 0 0 2 11 0 4] + [ 0 0 0 1 6 0 0 0 0 0 0 4 10 1 2 1071 16 9 3 5 6] + [ 0 15 0 1 6 6 1 0 1 0 0 4 0 3 4 10 1093 0 1 4 12] + [ 0 0 0 2 0 0 4 0 2 0 0 3 29 1 1 7 1 985 0 0 3] + [ 2 4 5 19 0 0 1 13 3 1 3 1 1 0 4 0 0 0 1002 0 9] + [ 0 0 6 1 1 2 9 7 0 0 0 9 6 1 3 4 7 0 1 1084 11] + [ 133 187 131 118 96 146 40 72 107 72 133 88 340 324 139 59 63 49 156 121 5331]] + +2023-10-02 21:34:24,678 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:34:24,678 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:34:24,684 - + +2023-10-02 21:34:24,684 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:34:25,732 - Epoch: [136][ 10/ 1236] Overall Loss 0.180259 Objective Loss 0.180259 LR 0.000500 Time 0.104780 +2023-10-02 21:34:25,943 - Epoch: [136][ 20/ 1236] Overall Loss 0.165650 Objective Loss 0.165650 LR 0.000500 Time 0.062922 +2023-10-02 21:34:26,153 - Epoch: [136][ 30/ 1236] Overall Loss 0.171222 Objective Loss 0.171222 LR 0.000500 Time 0.048937 +2023-10-02 21:34:26,364 - Epoch: [136][ 40/ 1236] Overall Loss 0.170404 Objective Loss 0.170404 LR 0.000500 Time 0.041960 +2023-10-02 21:34:26,573 - Epoch: [136][ 50/ 1236] Overall Loss 0.171544 Objective Loss 0.171544 LR 0.000500 Time 0.037742 +2023-10-02 21:34:26,783 - Epoch: [136][ 60/ 1236] Overall Loss 0.172702 Objective Loss 0.172702 LR 0.000500 Time 0.034952 +2023-10-02 21:34:26,991 - Epoch: [136][ 70/ 1236] Overall Loss 0.170848 Objective Loss 0.170848 LR 0.000500 Time 0.032917 +2023-10-02 21:34:27,201 - Epoch: [136][ 80/ 1236] Overall Loss 0.171766 Objective Loss 0.171766 LR 0.000500 Time 0.031427 +2023-10-02 21:34:27,409 - Epoch: [136][ 90/ 1236] Overall Loss 0.171917 Objective Loss 0.171917 LR 0.000500 Time 0.030244 +2023-10-02 21:34:27,620 - Epoch: [136][ 100/ 1236] Overall Loss 0.172364 Objective Loss 0.172364 LR 0.000500 Time 0.029319 +2023-10-02 21:34:27,829 - Epoch: [136][ 110/ 1236] Overall Loss 0.172714 Objective Loss 0.172714 LR 0.000500 Time 0.028554 +2023-10-02 21:34:28,041 - Epoch: [136][ 120/ 1236] Overall Loss 0.172561 Objective Loss 0.172561 LR 0.000500 Time 0.027936 +2023-10-02 21:34:28,253 - Epoch: [136][ 130/ 1236] Overall Loss 0.175675 Objective Loss 0.175675 LR 0.000500 Time 0.027411 +2023-10-02 21:34:28,464 - Epoch: [136][ 140/ 1236] Overall Loss 0.174491 Objective Loss 0.174491 LR 0.000500 Time 0.026953 +2023-10-02 21:34:28,671 - Epoch: [136][ 150/ 1236] Overall Loss 0.173428 Objective Loss 0.173428 LR 0.000500 Time 0.026529 +2023-10-02 21:34:28,878 - Epoch: [136][ 160/ 1236] Overall Loss 0.173599 Objective Loss 0.173599 LR 0.000500 Time 0.026167 +2023-10-02 21:34:29,084 - Epoch: [136][ 170/ 1236] Overall Loss 0.173067 Objective Loss 0.173067 LR 0.000500 Time 0.025836 +2023-10-02 21:34:29,292 - Epoch: [136][ 180/ 1236] Overall Loss 0.174549 Objective Loss 0.174549 LR 0.000500 Time 0.025553 +2023-10-02 21:34:29,498 - Epoch: [136][ 190/ 1236] Overall Loss 0.175576 Objective Loss 0.175576 LR 0.000500 Time 0.025290 +2023-10-02 21:34:29,705 - Epoch: [136][ 200/ 1236] Overall Loss 0.174993 Objective Loss 0.174993 LR 0.000500 Time 0.025063 +2023-10-02 21:34:29,910 - Epoch: [136][ 210/ 1236] Overall Loss 0.174688 Objective Loss 0.174688 LR 0.000500 Time 0.024845 +2023-10-02 21:34:30,118 - Epoch: [136][ 220/ 1236] Overall Loss 0.173874 Objective Loss 0.173874 LR 0.000500 Time 0.024657 +2023-10-02 21:34:30,324 - Epoch: [136][ 230/ 1236] Overall Loss 0.174021 Objective Loss 0.174021 LR 0.000500 Time 0.024479 +2023-10-02 21:34:30,531 - Epoch: [136][ 240/ 1236] Overall Loss 0.173531 Objective Loss 0.173531 LR 0.000500 Time 0.024323 +2023-10-02 21:34:30,737 - Epoch: [136][ 250/ 1236] Overall Loss 0.172488 Objective Loss 0.172488 LR 0.000500 Time 0.024173 +2023-10-02 21:34:30,945 - Epoch: [136][ 260/ 1236] Overall Loss 0.172634 Objective Loss 0.172634 LR 0.000500 Time 0.024041 +2023-10-02 21:34:31,151 - Epoch: [136][ 270/ 1236] Overall Loss 0.172348 Objective Loss 0.172348 LR 0.000500 Time 0.023912 +2023-10-02 21:34:31,358 - Epoch: [136][ 280/ 1236] Overall Loss 0.172577 Objective Loss 0.172577 LR 0.000500 Time 0.023797 +2023-10-02 21:34:31,564 - Epoch: [136][ 290/ 1236] Overall Loss 0.172138 Objective Loss 0.172138 LR 0.000500 Time 0.023684 +2023-10-02 21:34:31,772 - Epoch: [136][ 300/ 1236] Overall Loss 0.171779 Objective Loss 0.171779 LR 0.000500 Time 0.023587 +2023-10-02 21:34:31,978 - Epoch: [136][ 310/ 1236] Overall Loss 0.171480 Objective Loss 0.171480 LR 0.000500 Time 0.023490 +2023-10-02 21:34:32,186 - Epoch: [136][ 320/ 1236] Overall Loss 0.172413 Objective Loss 0.172413 LR 0.000500 Time 0.023405 +2023-10-02 21:34:32,392 - Epoch: [136][ 330/ 1236] Overall Loss 0.172215 Objective Loss 0.172215 LR 0.000500 Time 0.023319 +2023-10-02 21:34:32,600 - Epoch: [136][ 340/ 1236] Overall Loss 0.172183 Objective Loss 0.172183 LR 0.000500 Time 0.023244 +2023-10-02 21:34:32,806 - Epoch: [136][ 350/ 1236] Overall Loss 0.172130 Objective Loss 0.172130 LR 0.000500 Time 0.023168 +2023-10-02 21:34:33,016 - Epoch: [136][ 360/ 1236] Overall Loss 0.171456 Objective Loss 0.171456 LR 0.000500 Time 0.023107 +2023-10-02 21:34:33,225 - Epoch: [136][ 370/ 1236] Overall Loss 0.171899 Objective Loss 0.171899 LR 0.000500 Time 0.023047 +2023-10-02 21:34:33,436 - Epoch: [136][ 380/ 1236] Overall Loss 0.172053 Objective Loss 0.172053 LR 0.000500 Time 0.022994 +2023-10-02 21:34:33,645 - Epoch: [136][ 390/ 1236] Overall Loss 0.172170 Objective Loss 0.172170 LR 0.000500 Time 0.022940 +2023-10-02 21:34:33,856 - Epoch: [136][ 400/ 1236] Overall Loss 0.172630 Objective Loss 0.172630 LR 0.000500 Time 0.022893 +2023-10-02 21:34:34,065 - Epoch: [136][ 410/ 1236] Overall Loss 0.172401 Objective Loss 0.172401 LR 0.000500 Time 0.022844 +2023-10-02 21:34:34,276 - Epoch: [136][ 420/ 1236] Overall Loss 0.172529 Objective Loss 0.172529 LR 0.000500 Time 0.022801 +2023-10-02 21:34:34,485 - Epoch: [136][ 430/ 1236] Overall Loss 0.172765 Objective Loss 0.172765 LR 0.000500 Time 0.022756 +2023-10-02 21:34:34,696 - Epoch: [136][ 440/ 1236] Overall Loss 0.173138 Objective Loss 0.173138 LR 0.000500 Time 0.022718 +2023-10-02 21:34:34,905 - Epoch: [136][ 450/ 1236] Overall Loss 0.172934 Objective Loss 0.172934 LR 0.000500 Time 0.022677 +2023-10-02 21:34:35,116 - Epoch: [136][ 460/ 1236] Overall Loss 0.172749 Objective Loss 0.172749 LR 0.000500 Time 0.022641 +2023-10-02 21:34:35,325 - Epoch: [136][ 470/ 1236] Overall Loss 0.172712 Objective Loss 0.172712 LR 0.000500 Time 0.022604 +2023-10-02 21:34:35,536 - Epoch: [136][ 480/ 1236] Overall Loss 0.172218 Objective Loss 0.172218 LR 0.000500 Time 0.022572 +2023-10-02 21:34:35,745 - Epoch: [136][ 490/ 1236] Overall Loss 0.173058 Objective Loss 0.173058 LR 0.000500 Time 0.022537 +2023-10-02 21:34:35,956 - Epoch: [136][ 500/ 1236] Overall Loss 0.173063 Objective Loss 0.173063 LR 0.000500 Time 0.022508 +2023-10-02 21:34:36,165 - Epoch: [136][ 510/ 1236] Overall Loss 0.173038 Objective Loss 0.173038 LR 0.000500 Time 0.022476 +2023-10-02 21:34:36,376 - Epoch: [136][ 520/ 1236] Overall Loss 0.173204 Objective Loss 0.173204 LR 0.000500 Time 0.022448 +2023-10-02 21:34:36,587 - Epoch: [136][ 530/ 1236] Overall Loss 0.173615 Objective Loss 0.173615 LR 0.000500 Time 0.022423 +2023-10-02 21:34:36,798 - Epoch: [136][ 540/ 1236] Overall Loss 0.173530 Objective Loss 0.173530 LR 0.000500 Time 0.022397 +2023-10-02 21:34:37,008 - Epoch: [136][ 550/ 1236] Overall Loss 0.173580 Objective Loss 0.173580 LR 0.000500 Time 0.022371 +2023-10-02 21:34:37,219 - Epoch: [136][ 560/ 1236] Overall Loss 0.173215 Objective Loss 0.173215 LR 0.000500 Time 0.022347 +2023-10-02 21:34:37,428 - Epoch: [136][ 570/ 1236] Overall Loss 0.173294 Objective Loss 0.173294 LR 0.000500 Time 0.022323 +2023-10-02 21:34:37,639 - Epoch: [136][ 580/ 1236] Overall Loss 0.173320 Objective Loss 0.173320 LR 0.000500 Time 0.022300 +2023-10-02 21:34:37,848 - Epoch: [136][ 590/ 1236] Overall Loss 0.173332 Objective Loss 0.173332 LR 0.000500 Time 0.022276 +2023-10-02 21:34:38,058 - Epoch: [136][ 600/ 1236] Overall Loss 0.172822 Objective Loss 0.172822 LR 0.000500 Time 0.022255 +2023-10-02 21:34:38,268 - Epoch: [136][ 610/ 1236] Overall Loss 0.172524 Objective Loss 0.172524 LR 0.000500 Time 0.022233 +2023-10-02 21:34:38,479 - Epoch: [136][ 620/ 1236] Overall Loss 0.172689 Objective Loss 0.172689 LR 0.000500 Time 0.022214 +2023-10-02 21:34:38,689 - Epoch: [136][ 630/ 1236] Overall Loss 0.172923 Objective Loss 0.172923 LR 0.000500 Time 0.022194 +2023-10-02 21:34:38,899 - Epoch: [136][ 640/ 1236] Overall Loss 0.173036 Objective Loss 0.173036 LR 0.000500 Time 0.022175 +2023-10-02 21:34:39,109 - Epoch: [136][ 650/ 1236] Overall Loss 0.172854 Objective Loss 0.172854 LR 0.000500 Time 0.022157 +2023-10-02 21:34:39,320 - Epoch: [136][ 660/ 1236] Overall Loss 0.173177 Objective Loss 0.173177 LR 0.000500 Time 0.022140 +2023-10-02 21:34:39,530 - Epoch: [136][ 670/ 1236] Overall Loss 0.173155 Objective Loss 0.173155 LR 0.000500 Time 0.022122 +2023-10-02 21:34:39,740 - Epoch: [136][ 680/ 1236] Overall Loss 0.173262 Objective Loss 0.173262 LR 0.000500 Time 0.022106 +2023-10-02 21:34:39,950 - Epoch: [136][ 690/ 1236] Overall Loss 0.173388 Objective Loss 0.173388 LR 0.000500 Time 0.022089 +2023-10-02 21:34:40,161 - Epoch: [136][ 700/ 1236] Overall Loss 0.173533 Objective Loss 0.173533 LR 0.000500 Time 0.022074 +2023-10-02 21:34:40,371 - Epoch: [136][ 710/ 1236] Overall Loss 0.173291 Objective Loss 0.173291 LR 0.000500 Time 0.022059 +2023-10-02 21:34:40,581 - Epoch: [136][ 720/ 1236] Overall Loss 0.173242 Objective Loss 0.173242 LR 0.000500 Time 0.022043 +2023-10-02 21:34:40,791 - Epoch: [136][ 730/ 1236] Overall Loss 0.173461 Objective Loss 0.173461 LR 0.000500 Time 0.022028 +2023-10-02 21:34:41,001 - Epoch: [136][ 740/ 1236] Overall Loss 0.173166 Objective Loss 0.173166 LR 0.000500 Time 0.022014 +2023-10-02 21:34:41,210 - Epoch: [136][ 750/ 1236] Overall Loss 0.173326 Objective Loss 0.173326 LR 0.000500 Time 0.021999 +2023-10-02 21:34:41,421 - Epoch: [136][ 760/ 1236] Overall Loss 0.173592 Objective Loss 0.173592 LR 0.000500 Time 0.021986 +2023-10-02 21:34:41,630 - Epoch: [136][ 770/ 1236] Overall Loss 0.173638 Objective Loss 0.173638 LR 0.000500 Time 0.021972 +2023-10-02 21:34:41,840 - Epoch: [136][ 780/ 1236] Overall Loss 0.173618 Objective Loss 0.173618 LR 0.000500 Time 0.021959 +2023-10-02 21:34:42,050 - Epoch: [136][ 790/ 1236] Overall Loss 0.173926 Objective Loss 0.173926 LR 0.000500 Time 0.021947 +2023-10-02 21:34:42,260 - Epoch: [136][ 800/ 1236] Overall Loss 0.173829 Objective Loss 0.173829 LR 0.000500 Time 0.021935 +2023-10-02 21:34:42,470 - Epoch: [136][ 810/ 1236] Overall Loss 0.173820 Objective Loss 0.173820 LR 0.000500 Time 0.021923 +2023-10-02 21:34:42,681 - Epoch: [136][ 820/ 1236] Overall Loss 0.173888 Objective Loss 0.173888 LR 0.000500 Time 0.021912 +2023-10-02 21:34:42,891 - Epoch: [136][ 830/ 1236] Overall Loss 0.173765 Objective Loss 0.173765 LR 0.000500 Time 0.021900 +2023-10-02 21:34:43,101 - Epoch: [136][ 840/ 1236] Overall Loss 0.173867 Objective Loss 0.173867 LR 0.000500 Time 0.021889 +2023-10-02 21:34:43,311 - Epoch: [136][ 850/ 1236] Overall Loss 0.173658 Objective Loss 0.173658 LR 0.000500 Time 0.021878 +2023-10-02 21:34:43,521 - Epoch: [136][ 860/ 1236] Overall Loss 0.173985 Objective Loss 0.173985 LR 0.000500 Time 0.021868 +2023-10-02 21:34:43,732 - Epoch: [136][ 870/ 1236] Overall Loss 0.174068 Objective Loss 0.174068 LR 0.000500 Time 0.021858 +2023-10-02 21:34:43,942 - Epoch: [136][ 880/ 1236] Overall Loss 0.174364 Objective Loss 0.174364 LR 0.000500 Time 0.021848 +2023-10-02 21:34:44,152 - Epoch: [136][ 890/ 1236] Overall Loss 0.174469 Objective Loss 0.174469 LR 0.000500 Time 0.021838 +2023-10-02 21:34:44,362 - Epoch: [136][ 900/ 1236] Overall Loss 0.174533 Objective Loss 0.174533 LR 0.000500 Time 0.021829 +2023-10-02 21:34:44,572 - Epoch: [136][ 910/ 1236] Overall Loss 0.174551 Objective Loss 0.174551 LR 0.000500 Time 0.021819 +2023-10-02 21:34:44,783 - Epoch: [136][ 920/ 1236] Overall Loss 0.174542 Objective Loss 0.174542 LR 0.000500 Time 0.021811 +2023-10-02 21:34:44,993 - Epoch: [136][ 930/ 1236] Overall Loss 0.174532 Objective Loss 0.174532 LR 0.000500 Time 0.021802 +2023-10-02 21:34:45,201 - Epoch: [136][ 940/ 1236] Overall Loss 0.174536 Objective Loss 0.174536 LR 0.000500 Time 0.021791 +2023-10-02 21:34:45,411 - Epoch: [136][ 950/ 1236] Overall Loss 0.174565 Objective Loss 0.174565 LR 0.000500 Time 0.021783 +2023-10-02 21:34:45,622 - Epoch: [136][ 960/ 1236] Overall Loss 0.174563 Objective Loss 0.174563 LR 0.000500 Time 0.021774 +2023-10-02 21:34:45,832 - Epoch: [136][ 970/ 1236] Overall Loss 0.174434 Objective Loss 0.174434 LR 0.000500 Time 0.021766 +2023-10-02 21:34:46,042 - Epoch: [136][ 980/ 1236] Overall Loss 0.174520 Objective Loss 0.174520 LR 0.000500 Time 0.021758 +2023-10-02 21:34:46,252 - Epoch: [136][ 990/ 1236] Overall Loss 0.174613 Objective Loss 0.174613 LR 0.000500 Time 0.021750 +2023-10-02 21:34:46,463 - Epoch: [136][ 1000/ 1236] Overall Loss 0.174721 Objective Loss 0.174721 LR 0.000500 Time 0.021743 +2023-10-02 21:34:46,672 - Epoch: [136][ 1010/ 1236] Overall Loss 0.174854 Objective Loss 0.174854 LR 0.000500 Time 0.021735 +2023-10-02 21:34:46,883 - Epoch: [136][ 1020/ 1236] Overall Loss 0.174891 Objective Loss 0.174891 LR 0.000500 Time 0.021728 +2023-10-02 21:34:47,093 - Epoch: [136][ 1030/ 1236] Overall Loss 0.174796 Objective Loss 0.174796 LR 0.000500 Time 0.021720 +2023-10-02 21:34:47,303 - Epoch: [136][ 1040/ 1236] Overall Loss 0.174926 Objective Loss 0.174926 LR 0.000500 Time 0.021713 +2023-10-02 21:34:47,513 - Epoch: [136][ 1050/ 1236] Overall Loss 0.174608 Objective Loss 0.174608 LR 0.000500 Time 0.021706 +2023-10-02 21:34:47,724 - Epoch: [136][ 1060/ 1236] Overall Loss 0.174681 Objective Loss 0.174681 LR 0.000500 Time 0.021700 +2023-10-02 21:34:47,934 - Epoch: [136][ 1070/ 1236] Overall Loss 0.174675 Objective Loss 0.174675 LR 0.000500 Time 0.021693 +2023-10-02 21:34:48,142 - Epoch: [136][ 1080/ 1236] Overall Loss 0.174867 Objective Loss 0.174867 LR 0.000500 Time 0.021685 +2023-10-02 21:34:48,352 - Epoch: [136][ 1090/ 1236] Overall Loss 0.174752 Objective Loss 0.174752 LR 0.000500 Time 0.021678 +2023-10-02 21:34:48,562 - Epoch: [136][ 1100/ 1236] Overall Loss 0.174540 Objective Loss 0.174540 LR 0.000500 Time 0.021672 +2023-10-02 21:34:48,772 - Epoch: [136][ 1110/ 1236] Overall Loss 0.174687 Objective Loss 0.174687 LR 0.000500 Time 0.021665 +2023-10-02 21:34:48,983 - Epoch: [136][ 1120/ 1236] Overall Loss 0.174728 Objective Loss 0.174728 LR 0.000500 Time 0.021659 +2023-10-02 21:34:49,192 - Epoch: [136][ 1130/ 1236] Overall Loss 0.174610 Objective Loss 0.174610 LR 0.000500 Time 0.021653 +2023-10-02 21:34:49,403 - Epoch: [136][ 1140/ 1236] Overall Loss 0.174482 Objective Loss 0.174482 LR 0.000500 Time 0.021647 +2023-10-02 21:34:49,613 - Epoch: [136][ 1150/ 1236] Overall Loss 0.174598 Objective Loss 0.174598 LR 0.000500 Time 0.021641 +2023-10-02 21:34:49,823 - Epoch: [136][ 1160/ 1236] Overall Loss 0.174538 Objective Loss 0.174538 LR 0.000500 Time 0.021635 +2023-10-02 21:34:50,033 - Epoch: [136][ 1170/ 1236] Overall Loss 0.174545 Objective Loss 0.174545 LR 0.000500 Time 0.021630 +2023-10-02 21:34:50,243 - Epoch: [136][ 1180/ 1236] Overall Loss 0.174681 Objective Loss 0.174681 LR 0.000500 Time 0.021624 +2023-10-02 21:34:50,453 - Epoch: [136][ 1190/ 1236] Overall Loss 0.174734 Objective Loss 0.174734 LR 0.000500 Time 0.021618 +2023-10-02 21:34:50,663 - Epoch: [136][ 1200/ 1236] Overall Loss 0.174618 Objective Loss 0.174618 LR 0.000500 Time 0.021613 +2023-10-02 21:34:50,873 - Epoch: [136][ 1210/ 1236] Overall Loss 0.174540 Objective Loss 0.174540 LR 0.000500 Time 0.021607 +2023-10-02 21:34:51,083 - Epoch: [136][ 1220/ 1236] Overall Loss 0.174208 Objective Loss 0.174208 LR 0.000500 Time 0.021602 +2023-10-02 21:34:51,346 - Epoch: [136][ 1230/ 1236] Overall Loss 0.174246 Objective Loss 0.174246 LR 0.000500 Time 0.021640 +2023-10-02 21:34:51,469 - Epoch: [136][ 1236/ 1236] Overall Loss 0.174229 Objective Loss 0.174229 Top1 89.613035 Top5 98.370672 LR 0.000500 Time 0.021634 +2023-10-02 21:34:51,606 - --- validate (epoch=136)----------- +2023-10-02 21:34:51,606 - 29943 samples (256 per mini-batch) +2023-10-02 21:34:52,110 - Epoch: [136][ 10/ 117] Loss 0.308892 Top1 84.687500 Top5 98.906250 +2023-10-02 21:34:52,264 - Epoch: [136][ 20/ 117] Loss 0.317529 Top1 84.902344 Top5 98.671875 +2023-10-02 21:34:52,416 - Epoch: [136][ 30/ 117] Loss 0.310208 Top1 85.091146 Top5 98.697917 +2023-10-02 21:34:52,570 - Epoch: [136][ 40/ 117] Loss 0.310297 Top1 85.322266 Top5 98.564453 +2023-10-02 21:34:52,721 - Epoch: [136][ 50/ 117] Loss 0.304563 Top1 85.421875 Top5 98.593750 +2023-10-02 21:34:52,874 - Epoch: [136][ 60/ 117] Loss 0.304433 Top1 85.266927 Top5 98.567708 +2023-10-02 21:34:53,024 - Epoch: [136][ 70/ 117] Loss 0.306242 Top1 85.312500 Top5 98.526786 +2023-10-02 21:34:53,177 - Epoch: [136][ 80/ 117] Loss 0.312364 Top1 85.297852 Top5 98.447266 +2023-10-02 21:34:53,328 - Epoch: [136][ 90/ 117] Loss 0.307645 Top1 85.434028 Top5 98.463542 +2023-10-02 21:34:53,482 - Epoch: [136][ 100/ 117] Loss 0.305618 Top1 85.562500 Top5 98.464844 +2023-10-02 21:34:53,642 - Epoch: [136][ 110/ 117] Loss 0.303907 Top1 85.596591 Top5 98.476562 +2023-10-02 21:34:53,731 - Epoch: [136][ 117/ 117] Loss 0.304536 Top1 85.672778 Top5 98.480446 +2023-10-02 21:34:53,862 - ==> Top1: 85.673 Top5: 98.480 Loss: 0.305 + +2023-10-02 21:34:53,862 - ==> Confusion: +[[ 918 3 3 1 8 2 0 0 5 76 1 2 1 0 5 0 3 0 2 0 20] + [ 0 1069 0 0 3 17 2 22 0 0 1 0 2 1 1 2 1 0 4 3 3] + [ 3 0 973 8 3 0 16 14 0 2 3 1 5 2 0 4 0 2 8 4 8] + [ 0 3 16 963 1 2 2 4 7 0 11 0 7 4 26 2 0 3 14 0 24] + [ 25 6 0 1 966 3 0 1 1 14 1 0 1 5 4 5 7 0 0 2 8] + [ 2 35 0 1 2 988 1 29 5 5 2 8 1 9 3 0 1 1 3 1 19] + [ 0 5 30 0 0 2 1123 7 0 0 5 1 0 1 0 4 0 1 1 7 4] + [ 2 14 8 0 3 24 4 1096 1 6 3 6 3 3 0 0 0 1 27 9 8] + [ 18 5 0 0 1 1 0 1 993 29 8 3 2 9 11 1 4 1 1 1 0] + [ 77 3 0 0 5 2 1 0 32 971 0 0 0 11 9 2 0 0 0 1 5] + [ 1 3 11 4 1 1 5 2 18 3 964 3 0 14 4 0 3 2 6 0 8] + [ 2 0 1 0 0 9 0 3 0 0 0 970 19 7 0 1 0 14 0 5 4] + [ 0 2 1 0 0 2 2 0 1 0 3 42 966 3 4 8 1 16 2 3 12] + [ 1 0 0 0 0 4 0 0 20 12 2 6 0 1051 3 0 2 1 0 2 15] + [ 14 3 4 14 4 0 0 0 29 3 2 0 1 4 999 0 2 5 9 0 8] + [ 0 0 1 0 4 0 0 0 0 1 1 4 6 0 1 1073 17 14 2 3 7] + [ 0 15 2 0 4 9 0 0 1 2 0 9 0 2 4 9 1088 0 2 5 9] + [ 0 1 0 2 0 0 3 0 0 1 0 1 14 0 1 5 0 1008 0 0 2] + [ 2 7 6 9 0 0 0 28 8 0 2 0 1 0 10 1 0 0 985 0 9] + [ 0 2 3 2 1 3 7 11 0 0 1 18 5 1 0 2 5 1 0 1079 11] + [ 116 184 116 59 70 134 29 115 122 103 166 104 333 267 108 63 89 61 109 147 5410]] + +2023-10-02 21:34:53,864 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:34:53,864 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:34:53,870 - + +2023-10-02 21:34:53,870 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:34:55,010 - Epoch: [137][ 10/ 1236] Overall Loss 0.155538 Objective Loss 0.155538 LR 0.000500 Time 0.113921 +2023-10-02 21:34:55,218 - Epoch: [137][ 20/ 1236] Overall Loss 0.154374 Objective Loss 0.154374 LR 0.000500 Time 0.067330 +2023-10-02 21:34:55,423 - Epoch: [137][ 30/ 1236] Overall Loss 0.162703 Objective Loss 0.162703 LR 0.000500 Time 0.051706 +2023-10-02 21:34:55,630 - Epoch: [137][ 40/ 1236] Overall Loss 0.163041 Objective Loss 0.163041 LR 0.000500 Time 0.043962 +2023-10-02 21:34:55,837 - Epoch: [137][ 50/ 1236] Overall Loss 0.166728 Objective Loss 0.166728 LR 0.000500 Time 0.039269 +2023-10-02 21:34:56,045 - Epoch: [137][ 60/ 1236] Overall Loss 0.170250 Objective Loss 0.170250 LR 0.000500 Time 0.036195 +2023-10-02 21:34:56,251 - Epoch: [137][ 70/ 1236] Overall Loss 0.167024 Objective Loss 0.167024 LR 0.000500 Time 0.033956 +2023-10-02 21:34:56,459 - Epoch: [137][ 80/ 1236] Overall Loss 0.168440 Objective Loss 0.168440 LR 0.000500 Time 0.032317 +2023-10-02 21:34:56,665 - Epoch: [137][ 90/ 1236] Overall Loss 0.165937 Objective Loss 0.165937 LR 0.000500 Time 0.031003 +2023-10-02 21:34:56,874 - Epoch: [137][ 100/ 1236] Overall Loss 0.167855 Objective Loss 0.167855 LR 0.000500 Time 0.029988 +2023-10-02 21:34:57,079 - Epoch: [137][ 110/ 1236] Overall Loss 0.168825 Objective Loss 0.168825 LR 0.000500 Time 0.029125 +2023-10-02 21:34:57,287 - Epoch: [137][ 120/ 1236] Overall Loss 0.170804 Objective Loss 0.170804 LR 0.000500 Time 0.028435 +2023-10-02 21:34:57,492 - Epoch: [137][ 130/ 1236] Overall Loss 0.172717 Objective Loss 0.172717 LR 0.000500 Time 0.027822 +2023-10-02 21:34:57,701 - Epoch: [137][ 140/ 1236] Overall Loss 0.173111 Objective Loss 0.173111 LR 0.000500 Time 0.027324 +2023-10-02 21:34:57,906 - Epoch: [137][ 150/ 1236] Overall Loss 0.172595 Objective Loss 0.172595 LR 0.000500 Time 0.026869 +2023-10-02 21:34:58,115 - Epoch: [137][ 160/ 1236] Overall Loss 0.172930 Objective Loss 0.172930 LR 0.000500 Time 0.026488 +2023-10-02 21:34:58,321 - Epoch: [137][ 170/ 1236] Overall Loss 0.171825 Objective Loss 0.171825 LR 0.000500 Time 0.026143 +2023-10-02 21:34:58,530 - Epoch: [137][ 180/ 1236] Overall Loss 0.171398 Objective Loss 0.171398 LR 0.000500 Time 0.025850 +2023-10-02 21:34:58,740 - Epoch: [137][ 190/ 1236] Overall Loss 0.171083 Objective Loss 0.171083 LR 0.000500 Time 0.025594 +2023-10-02 21:34:58,949 - Epoch: [137][ 200/ 1236] Overall Loss 0.171498 Objective Loss 0.171498 LR 0.000500 Time 0.025355 +2023-10-02 21:34:59,158 - Epoch: [137][ 210/ 1236] Overall Loss 0.172391 Objective Loss 0.172391 LR 0.000500 Time 0.025144 +2023-10-02 21:34:59,367 - Epoch: [137][ 220/ 1236] Overall Loss 0.173384 Objective Loss 0.173384 LR 0.000500 Time 0.024948 +2023-10-02 21:34:59,577 - Epoch: [137][ 230/ 1236] Overall Loss 0.173279 Objective Loss 0.173279 LR 0.000500 Time 0.024777 +2023-10-02 21:34:59,786 - Epoch: [137][ 240/ 1236] Overall Loss 0.172809 Objective Loss 0.172809 LR 0.000500 Time 0.024613 +2023-10-02 21:34:59,996 - Epoch: [137][ 250/ 1236] Overall Loss 0.172811 Objective Loss 0.172811 LR 0.000500 Time 0.024468 +2023-10-02 21:35:00,204 - Epoch: [137][ 260/ 1236] Overall Loss 0.172774 Objective Loss 0.172774 LR 0.000500 Time 0.024326 +2023-10-02 21:35:00,414 - Epoch: [137][ 270/ 1236] Overall Loss 0.173605 Objective Loss 0.173605 LR 0.000500 Time 0.024202 +2023-10-02 21:35:00,624 - Epoch: [137][ 280/ 1236] Overall Loss 0.173705 Objective Loss 0.173705 LR 0.000500 Time 0.024086 +2023-10-02 21:35:00,834 - Epoch: [137][ 290/ 1236] Overall Loss 0.173877 Objective Loss 0.173877 LR 0.000500 Time 0.023976 +2023-10-02 21:35:01,041 - Epoch: [137][ 300/ 1236] Overall Loss 0.174366 Objective Loss 0.174366 LR 0.000500 Time 0.023868 +2023-10-02 21:35:01,254 - Epoch: [137][ 310/ 1236] Overall Loss 0.173680 Objective Loss 0.173680 LR 0.000500 Time 0.023782 +2023-10-02 21:35:01,464 - Epoch: [137][ 320/ 1236] Overall Loss 0.173167 Objective Loss 0.173167 LR 0.000500 Time 0.023695 +2023-10-02 21:35:01,677 - Epoch: [137][ 330/ 1236] Overall Loss 0.173784 Objective Loss 0.173784 LR 0.000500 Time 0.023621 +2023-10-02 21:35:01,887 - Epoch: [137][ 340/ 1236] Overall Loss 0.173724 Objective Loss 0.173724 LR 0.000500 Time 0.023544 +2023-10-02 21:35:02,100 - Epoch: [137][ 350/ 1236] Overall Loss 0.173646 Objective Loss 0.173646 LR 0.000500 Time 0.023478 +2023-10-02 21:35:02,309 - Epoch: [137][ 360/ 1236] Overall Loss 0.172607 Objective Loss 0.172607 LR 0.000500 Time 0.023405 +2023-10-02 21:35:02,520 - Epoch: [137][ 370/ 1236] Overall Loss 0.172652 Objective Loss 0.172652 LR 0.000500 Time 0.023342 +2023-10-02 21:35:02,729 - Epoch: [137][ 380/ 1236] Overall Loss 0.172751 Objective Loss 0.172751 LR 0.000500 Time 0.023277 +2023-10-02 21:35:02,940 - Epoch: [137][ 390/ 1236] Overall Loss 0.173699 Objective Loss 0.173699 LR 0.000500 Time 0.023221 +2023-10-02 21:35:03,149 - Epoch: [137][ 400/ 1236] Overall Loss 0.172872 Objective Loss 0.172872 LR 0.000500 Time 0.023163 +2023-10-02 21:35:03,361 - Epoch: [137][ 410/ 1236] Overall Loss 0.172823 Objective Loss 0.172823 LR 0.000500 Time 0.023112 +2023-10-02 21:35:03,570 - Epoch: [137][ 420/ 1236] Overall Loss 0.172498 Objective Loss 0.172498 LR 0.000500 Time 0.023059 +2023-10-02 21:35:03,781 - Epoch: [137][ 430/ 1236] Overall Loss 0.172279 Objective Loss 0.172279 LR 0.000500 Time 0.023013 +2023-10-02 21:35:03,990 - Epoch: [137][ 440/ 1236] Overall Loss 0.172488 Objective Loss 0.172488 LR 0.000500 Time 0.022964 +2023-10-02 21:35:04,201 - Epoch: [137][ 450/ 1236] Overall Loss 0.172357 Objective Loss 0.172357 LR 0.000500 Time 0.022923 +2023-10-02 21:35:04,410 - Epoch: [137][ 460/ 1236] Overall Loss 0.172557 Objective Loss 0.172557 LR 0.000500 Time 0.022879 +2023-10-02 21:35:04,623 - Epoch: [137][ 470/ 1236] Overall Loss 0.172547 Objective Loss 0.172547 LR 0.000500 Time 0.022843 +2023-10-02 21:35:04,832 - Epoch: [137][ 480/ 1236] Overall Loss 0.172650 Objective Loss 0.172650 LR 0.000500 Time 0.022802 +2023-10-02 21:35:05,045 - Epoch: [137][ 490/ 1236] Overall Loss 0.172474 Objective Loss 0.172474 LR 0.000500 Time 0.022771 +2023-10-02 21:35:05,253 - Epoch: [137][ 500/ 1236] Overall Loss 0.172691 Objective Loss 0.172691 LR 0.000500 Time 0.022730 +2023-10-02 21:35:05,465 - Epoch: [137][ 510/ 1236] Overall Loss 0.172215 Objective Loss 0.172215 LR 0.000500 Time 0.022700 +2023-10-02 21:35:05,673 - Epoch: [137][ 520/ 1236] Overall Loss 0.172127 Objective Loss 0.172127 LR 0.000500 Time 0.022662 +2023-10-02 21:35:05,885 - Epoch: [137][ 530/ 1236] Overall Loss 0.172021 Objective Loss 0.172021 LR 0.000500 Time 0.022635 +2023-10-02 21:35:06,095 - Epoch: [137][ 540/ 1236] Overall Loss 0.171710 Objective Loss 0.171710 LR 0.000500 Time 0.022603 +2023-10-02 21:35:06,305 - Epoch: [137][ 550/ 1236] Overall Loss 0.171983 Objective Loss 0.171983 LR 0.000500 Time 0.022574 +2023-10-02 21:35:06,514 - Epoch: [137][ 560/ 1236] Overall Loss 0.172173 Objective Loss 0.172173 LR 0.000500 Time 0.022544 +2023-10-02 21:35:06,727 - Epoch: [137][ 570/ 1236] Overall Loss 0.172090 Objective Loss 0.172090 LR 0.000500 Time 0.022520 +2023-10-02 21:35:06,936 - Epoch: [137][ 580/ 1236] Overall Loss 0.171922 Objective Loss 0.171922 LR 0.000500 Time 0.022492 +2023-10-02 21:35:07,148 - Epoch: [137][ 590/ 1236] Overall Loss 0.171975 Objective Loss 0.171975 LR 0.000500 Time 0.022470 +2023-10-02 21:35:07,357 - Epoch: [137][ 600/ 1236] Overall Loss 0.171913 Objective Loss 0.171913 LR 0.000500 Time 0.022444 +2023-10-02 21:35:07,570 - Epoch: [137][ 610/ 1236] Overall Loss 0.171577 Objective Loss 0.171577 LR 0.000500 Time 0.022424 +2023-10-02 21:35:07,779 - Epoch: [137][ 620/ 1236] Overall Loss 0.171493 Objective Loss 0.171493 LR 0.000500 Time 0.022398 +2023-10-02 21:35:07,991 - Epoch: [137][ 630/ 1236] Overall Loss 0.171056 Objective Loss 0.171056 LR 0.000500 Time 0.022380 +2023-10-02 21:35:08,200 - Epoch: [137][ 640/ 1236] Overall Loss 0.171114 Objective Loss 0.171114 LR 0.000500 Time 0.022356 +2023-10-02 21:35:08,411 - Epoch: [137][ 650/ 1236] Overall Loss 0.171263 Objective Loss 0.171263 LR 0.000500 Time 0.022335 +2023-10-02 21:35:08,620 - Epoch: [137][ 660/ 1236] Overall Loss 0.171092 Objective Loss 0.171092 LR 0.000500 Time 0.022314 +2023-10-02 21:35:08,832 - Epoch: [137][ 670/ 1236] Overall Loss 0.171282 Objective Loss 0.171282 LR 0.000500 Time 0.022297 +2023-10-02 21:35:09,041 - Epoch: [137][ 680/ 1236] Overall Loss 0.171382 Objective Loss 0.171382 LR 0.000500 Time 0.022276 +2023-10-02 21:35:09,254 - Epoch: [137][ 690/ 1236] Overall Loss 0.171523 Objective Loss 0.171523 LR 0.000500 Time 0.022260 +2023-10-02 21:35:09,463 - Epoch: [137][ 700/ 1236] Overall Loss 0.171790 Objective Loss 0.171790 LR 0.000500 Time 0.022240 +2023-10-02 21:35:09,675 - Epoch: [137][ 710/ 1236] Overall Loss 0.171736 Objective Loss 0.171736 LR 0.000500 Time 0.022225 +2023-10-02 21:35:09,885 - Epoch: [137][ 720/ 1236] Overall Loss 0.171628 Objective Loss 0.171628 LR 0.000500 Time 0.022207 +2023-10-02 21:35:10,097 - Epoch: [137][ 730/ 1236] Overall Loss 0.171506 Objective Loss 0.171506 LR 0.000500 Time 0.022194 +2023-10-02 21:35:10,308 - Epoch: [137][ 740/ 1236] Overall Loss 0.171943 Objective Loss 0.171943 LR 0.000500 Time 0.022178 +2023-10-02 21:35:10,517 - Epoch: [137][ 750/ 1236] Overall Loss 0.172041 Objective Loss 0.172041 LR 0.000500 Time 0.022160 +2023-10-02 21:35:10,729 - Epoch: [137][ 760/ 1236] Overall Loss 0.171933 Objective Loss 0.171933 LR 0.000500 Time 0.022146 +2023-10-02 21:35:10,940 - Epoch: [137][ 770/ 1236] Overall Loss 0.171905 Objective Loss 0.171905 LR 0.000500 Time 0.022133 +2023-10-02 21:35:11,151 - Epoch: [137][ 780/ 1236] Overall Loss 0.171894 Objective Loss 0.171894 LR 0.000500 Time 0.022119 +2023-10-02 21:35:11,363 - Epoch: [137][ 790/ 1236] Overall Loss 0.171860 Objective Loss 0.171860 LR 0.000500 Time 0.022105 +2023-10-02 21:35:11,574 - Epoch: [137][ 800/ 1236] Overall Loss 0.171869 Objective Loss 0.171869 LR 0.000500 Time 0.022092 +2023-10-02 21:35:11,786 - Epoch: [137][ 810/ 1236] Overall Loss 0.171675 Objective Loss 0.171675 LR 0.000500 Time 0.022081 +2023-10-02 21:35:11,997 - Epoch: [137][ 820/ 1236] Overall Loss 0.171838 Objective Loss 0.171838 LR 0.000500 Time 0.022068 +2023-10-02 21:35:12,209 - Epoch: [137][ 830/ 1236] Overall Loss 0.172120 Objective Loss 0.172120 LR 0.000500 Time 0.022057 +2023-10-02 21:35:12,419 - Epoch: [137][ 840/ 1236] Overall Loss 0.172023 Objective Loss 0.172023 LR 0.000500 Time 0.022045 +2023-10-02 21:35:12,632 - Epoch: [137][ 850/ 1236] Overall Loss 0.171882 Objective Loss 0.171882 LR 0.000500 Time 0.022035 +2023-10-02 21:35:12,842 - Epoch: [137][ 860/ 1236] Overall Loss 0.171658 Objective Loss 0.171658 LR 0.000500 Time 0.022023 +2023-10-02 21:35:13,055 - Epoch: [137][ 870/ 1236] Overall Loss 0.171336 Objective Loss 0.171336 LR 0.000500 Time 0.022014 +2023-10-02 21:35:13,265 - Epoch: [137][ 880/ 1236] Overall Loss 0.171406 Objective Loss 0.171406 LR 0.000500 Time 0.022003 +2023-10-02 21:35:13,477 - Epoch: [137][ 890/ 1236] Overall Loss 0.171472 Objective Loss 0.171472 LR 0.000500 Time 0.021993 +2023-10-02 21:35:13,688 - Epoch: [137][ 900/ 1236] Overall Loss 0.171607 Objective Loss 0.171607 LR 0.000500 Time 0.021983 +2023-10-02 21:35:13,900 - Epoch: [137][ 910/ 1236] Overall Loss 0.171692 Objective Loss 0.171692 LR 0.000500 Time 0.021973 +2023-10-02 21:35:14,111 - Epoch: [137][ 920/ 1236] Overall Loss 0.171989 Objective Loss 0.171989 LR 0.000500 Time 0.021963 +2023-10-02 21:35:14,323 - Epoch: [137][ 930/ 1236] Overall Loss 0.172099 Objective Loss 0.172099 LR 0.000500 Time 0.021954 +2023-10-02 21:35:14,534 - Epoch: [137][ 940/ 1236] Overall Loss 0.171982 Objective Loss 0.171982 LR 0.000500 Time 0.021944 +2023-10-02 21:35:14,746 - Epoch: [137][ 950/ 1236] Overall Loss 0.171876 Objective Loss 0.171876 LR 0.000500 Time 0.021936 +2023-10-02 21:35:14,957 - Epoch: [137][ 960/ 1236] Overall Loss 0.171929 Objective Loss 0.171929 LR 0.000500 Time 0.021927 +2023-10-02 21:35:15,169 - Epoch: [137][ 970/ 1236] Overall Loss 0.172110 Objective Loss 0.172110 LR 0.000500 Time 0.021920 +2023-10-02 21:35:15,380 - Epoch: [137][ 980/ 1236] Overall Loss 0.172012 Objective Loss 0.172012 LR 0.000500 Time 0.021910 +2023-10-02 21:35:15,592 - Epoch: [137][ 990/ 1236] Overall Loss 0.171844 Objective Loss 0.171844 LR 0.000500 Time 0.021903 +2023-10-02 21:35:15,802 - Epoch: [137][ 1000/ 1236] Overall Loss 0.171651 Objective Loss 0.171651 LR 0.000500 Time 0.021894 +2023-10-02 21:35:16,015 - Epoch: [137][ 1010/ 1236] Overall Loss 0.171477 Objective Loss 0.171477 LR 0.000500 Time 0.021888 +2023-10-02 21:35:16,225 - Epoch: [137][ 1020/ 1236] Overall Loss 0.171387 Objective Loss 0.171387 LR 0.000500 Time 0.021879 +2023-10-02 21:35:16,438 - Epoch: [137][ 1030/ 1236] Overall Loss 0.171289 Objective Loss 0.171289 LR 0.000500 Time 0.021873 +2023-10-02 21:35:16,648 - Epoch: [137][ 1040/ 1236] Overall Loss 0.171140 Objective Loss 0.171140 LR 0.000500 Time 0.021865 +2023-10-02 21:35:16,861 - Epoch: [137][ 1050/ 1236] Overall Loss 0.171307 Objective Loss 0.171307 LR 0.000500 Time 0.021858 +2023-10-02 21:35:17,071 - Epoch: [137][ 1060/ 1236] Overall Loss 0.171338 Objective Loss 0.171338 LR 0.000500 Time 0.021850 +2023-10-02 21:35:17,283 - Epoch: [137][ 1070/ 1236] Overall Loss 0.171513 Objective Loss 0.171513 LR 0.000500 Time 0.021843 +2023-10-02 21:35:17,494 - Epoch: [137][ 1080/ 1236] Overall Loss 0.171520 Objective Loss 0.171520 LR 0.000500 Time 0.021836 +2023-10-02 21:35:17,706 - Epoch: [137][ 1090/ 1236] Overall Loss 0.171784 Objective Loss 0.171784 LR 0.000500 Time 0.021829 +2023-10-02 21:35:17,917 - Epoch: [137][ 1100/ 1236] Overall Loss 0.171930 Objective Loss 0.171930 LR 0.000500 Time 0.021822 +2023-10-02 21:35:18,129 - Epoch: [137][ 1110/ 1236] Overall Loss 0.171937 Objective Loss 0.171937 LR 0.000500 Time 0.021816 +2023-10-02 21:35:18,340 - Epoch: [137][ 1120/ 1236] Overall Loss 0.172030 Objective Loss 0.172030 LR 0.000500 Time 0.021809 +2023-10-02 21:35:18,551 - Epoch: [137][ 1130/ 1236] Overall Loss 0.172143 Objective Loss 0.172143 LR 0.000500 Time 0.021804 +2023-10-02 21:35:18,763 - Epoch: [137][ 1140/ 1236] Overall Loss 0.172231 Objective Loss 0.172231 LR 0.000500 Time 0.021797 +2023-10-02 21:35:18,973 - Epoch: [137][ 1150/ 1236] Overall Loss 0.172219 Objective Loss 0.172219 LR 0.000500 Time 0.021789 +2023-10-02 21:35:19,182 - Epoch: [137][ 1160/ 1236] Overall Loss 0.172316 Objective Loss 0.172316 LR 0.000500 Time 0.021781 +2023-10-02 21:35:19,390 - Epoch: [137][ 1170/ 1236] Overall Loss 0.172256 Objective Loss 0.172256 LR 0.000500 Time 0.021773 +2023-10-02 21:35:19,600 - Epoch: [137][ 1180/ 1236] Overall Loss 0.172229 Objective Loss 0.172229 LR 0.000500 Time 0.021766 +2023-10-02 21:35:19,808 - Epoch: [137][ 1190/ 1236] Overall Loss 0.172166 Objective Loss 0.172166 LR 0.000500 Time 0.021757 +2023-10-02 21:35:20,018 - Epoch: [137][ 1200/ 1236] Overall Loss 0.172457 Objective Loss 0.172457 LR 0.000500 Time 0.021750 +2023-10-02 21:35:20,226 - Epoch: [137][ 1210/ 1236] Overall Loss 0.172364 Objective Loss 0.172364 LR 0.000500 Time 0.021741 +2023-10-02 21:35:20,435 - Epoch: [137][ 1220/ 1236] Overall Loss 0.172434 Objective Loss 0.172434 LR 0.000500 Time 0.021734 +2023-10-02 21:35:20,697 - Epoch: [137][ 1230/ 1236] Overall Loss 0.172447 Objective Loss 0.172447 LR 0.000500 Time 0.021769 +2023-10-02 21:35:20,819 - Epoch: [137][ 1236/ 1236] Overall Loss 0.172371 Objective Loss 0.172371 Top1 89.002037 Top5 99.389002 LR 0.000500 Time 0.021762 +2023-10-02 21:35:20,955 - --- validate (epoch=137)----------- +2023-10-02 21:35:20,955 - 29943 samples (256 per mini-batch) +2023-10-02 21:35:21,446 - Epoch: [137][ 10/ 117] Loss 0.345492 Top1 84.726562 Top5 98.164062 +2023-10-02 21:35:21,602 - Epoch: [137][ 20/ 117] Loss 0.328598 Top1 84.472656 Top5 98.320312 +2023-10-02 21:35:21,756 - Epoch: [137][ 30/ 117] Loss 0.326259 Top1 84.609375 Top5 98.333333 +2023-10-02 21:35:21,912 - Epoch: [137][ 40/ 117] Loss 0.328127 Top1 84.697266 Top5 98.183594 +2023-10-02 21:35:22,066 - Epoch: [137][ 50/ 117] Loss 0.320379 Top1 84.742188 Top5 98.257812 +2023-10-02 21:35:22,221 - Epoch: [137][ 60/ 117] Loss 0.310220 Top1 84.902344 Top5 98.294271 +2023-10-02 21:35:22,375 - Epoch: [137][ 70/ 117] Loss 0.307326 Top1 84.949777 Top5 98.348214 +2023-10-02 21:35:22,531 - Epoch: [137][ 80/ 117] Loss 0.309284 Top1 84.990234 Top5 98.330078 +2023-10-02 21:35:22,686 - Epoch: [137][ 90/ 117] Loss 0.303051 Top1 85.065104 Top5 98.355035 +2023-10-02 21:35:22,841 - Epoch: [137][ 100/ 117] Loss 0.303321 Top1 85.121094 Top5 98.355469 +2023-10-02 21:35:23,001 - Epoch: [137][ 110/ 117] Loss 0.303402 Top1 85.149148 Top5 98.348722 +2023-10-02 21:35:23,090 - Epoch: [137][ 117/ 117] Loss 0.305851 Top1 85.074976 Top5 98.350199 +2023-10-02 21:35:23,227 - ==> Top1: 85.075 Top5: 98.350 Loss: 0.306 + +2023-10-02 21:35:23,227 - ==> Confusion: +[[ 954 2 5 1 9 3 0 1 6 41 2 1 0 1 6 1 3 0 1 0 13] + [ 0 1030 2 1 5 36 0 25 1 0 2 2 1 0 0 3 7 0 7 4 5] + [ 5 0 977 13 2 0 12 12 0 2 1 0 8 3 1 4 0 1 5 2 8] + [ 0 4 17 986 1 0 1 3 8 0 2 0 9 2 21 5 1 0 12 2 15] + [ 25 5 0 1 962 5 0 0 0 9 2 0 1 4 13 3 14 0 0 0 6] + [ 4 27 0 2 2 1003 1 30 2 4 2 3 3 8 6 1 4 0 1 3 10] + [ 0 2 26 0 0 2 1132 4 0 0 5 0 0 0 0 6 0 0 0 9 5] + [ 4 11 17 0 7 23 4 1078 1 3 6 4 2 6 2 0 0 1 33 9 7] + [ 17 2 0 0 3 1 0 0 971 38 12 1 2 15 17 1 5 0 2 0 2] + [ 123 1 1 0 15 6 1 0 22 901 0 0 1 24 12 2 1 0 0 0 9] + [ 3 1 9 11 0 0 2 4 7 0 973 1 0 13 9 0 1 5 7 1 6] + [ 1 0 3 0 1 11 0 3 0 0 0 955 30 8 0 1 1 12 0 4 5] + [ 1 1 1 3 0 1 1 0 1 1 4 32 976 3 5 8 2 10 3 7 8] + [ 1 0 0 0 4 10 1 1 13 10 2 7 0 1057 5 0 0 1 0 1 6] + [ 12 1 3 14 5 0 0 0 13 0 2 0 2 5 1026 0 3 1 7 0 7] + [ 0 0 0 1 4 0 0 0 0 0 1 6 9 0 0 1072 17 10 3 5 6] + [ 1 11 1 0 3 8 0 0 0 0 0 5 1 4 4 5 1106 0 1 5 6] + [ 0 0 0 3 0 0 2 0 0 1 0 5 32 2 3 8 1 977 0 2 2] + [ 2 6 3 11 0 1 0 22 7 0 3 0 0 0 14 0 0 0 990 1 8] + [ 0 0 1 0 2 3 8 5 0 1 0 15 5 2 0 0 11 0 0 1097 2] + [ 123 137 129 86 86 153 36 93 92 66 192 94 383 266 152 73 147 50 106 190 5251]] + +2023-10-02 21:35:23,229 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:35:23,229 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:35:23,235 - + +2023-10-02 21:35:23,235 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:35:24,259 - Epoch: [138][ 10/ 1236] Overall Loss 0.154466 Objective Loss 0.154466 LR 0.000500 Time 0.102342 +2023-10-02 21:35:24,468 - Epoch: [138][ 20/ 1236] Overall Loss 0.167241 Objective Loss 0.167241 LR 0.000500 Time 0.061623 +2023-10-02 21:35:24,676 - Epoch: [138][ 30/ 1236] Overall Loss 0.171546 Objective Loss 0.171546 LR 0.000500 Time 0.047956 +2023-10-02 21:35:24,886 - Epoch: [138][ 40/ 1236] Overall Loss 0.172058 Objective Loss 0.172058 LR 0.000500 Time 0.041216 +2023-10-02 21:35:25,093 - Epoch: [138][ 50/ 1236] Overall Loss 0.173990 Objective Loss 0.173990 LR 0.000500 Time 0.037095 +2023-10-02 21:35:25,303 - Epoch: [138][ 60/ 1236] Overall Loss 0.174247 Objective Loss 0.174247 LR 0.000500 Time 0.034413 +2023-10-02 21:35:25,509 - Epoch: [138][ 70/ 1236] Overall Loss 0.176206 Objective Loss 0.176206 LR 0.000500 Time 0.032438 +2023-10-02 21:35:25,720 - Epoch: [138][ 80/ 1236] Overall Loss 0.173099 Objective Loss 0.173099 LR 0.000500 Time 0.031013 +2023-10-02 21:35:25,926 - Epoch: [138][ 90/ 1236] Overall Loss 0.173059 Objective Loss 0.173059 LR 0.000500 Time 0.029854 +2023-10-02 21:35:26,136 - Epoch: [138][ 100/ 1236] Overall Loss 0.172188 Objective Loss 0.172188 LR 0.000500 Time 0.028966 +2023-10-02 21:35:26,343 - Epoch: [138][ 110/ 1236] Overall Loss 0.171314 Objective Loss 0.171314 LR 0.000500 Time 0.028214 +2023-10-02 21:35:26,552 - Epoch: [138][ 120/ 1236] Overall Loss 0.171008 Objective Loss 0.171008 LR 0.000500 Time 0.027597 +2023-10-02 21:35:26,759 - Epoch: [138][ 130/ 1236] Overall Loss 0.170227 Objective Loss 0.170227 LR 0.000500 Time 0.027056 +2023-10-02 21:35:26,969 - Epoch: [138][ 140/ 1236] Overall Loss 0.169131 Objective Loss 0.169131 LR 0.000500 Time 0.026620 +2023-10-02 21:35:27,174 - Epoch: [138][ 150/ 1236] Overall Loss 0.169187 Objective Loss 0.169187 LR 0.000500 Time 0.026214 +2023-10-02 21:35:27,383 - Epoch: [138][ 160/ 1236] Overall Loss 0.169470 Objective Loss 0.169470 LR 0.000500 Time 0.025878 +2023-10-02 21:35:27,590 - Epoch: [138][ 170/ 1236] Overall Loss 0.170998 Objective Loss 0.170998 LR 0.000500 Time 0.025564 +2023-10-02 21:35:27,800 - Epoch: [138][ 180/ 1236] Overall Loss 0.170938 Objective Loss 0.170938 LR 0.000500 Time 0.025310 +2023-10-02 21:35:28,006 - Epoch: [138][ 190/ 1236] Overall Loss 0.169883 Objective Loss 0.169883 LR 0.000500 Time 0.025059 +2023-10-02 21:35:28,214 - Epoch: [138][ 200/ 1236] Overall Loss 0.169735 Objective Loss 0.169735 LR 0.000500 Time 0.024845 +2023-10-02 21:35:28,420 - Epoch: [138][ 210/ 1236] Overall Loss 0.170184 Objective Loss 0.170184 LR 0.000500 Time 0.024638 +2023-10-02 21:35:28,629 - Epoch: [138][ 220/ 1236] Overall Loss 0.170119 Objective Loss 0.170119 LR 0.000500 Time 0.024464 +2023-10-02 21:35:28,836 - Epoch: [138][ 230/ 1236] Overall Loss 0.170421 Objective Loss 0.170421 LR 0.000500 Time 0.024294 +2023-10-02 21:35:29,044 - Epoch: [138][ 240/ 1236] Overall Loss 0.169365 Objective Loss 0.169365 LR 0.000500 Time 0.024149 +2023-10-02 21:35:29,251 - Epoch: [138][ 250/ 1236] Overall Loss 0.168965 Objective Loss 0.168965 LR 0.000500 Time 0.024004 +2023-10-02 21:35:29,461 - Epoch: [138][ 260/ 1236] Overall Loss 0.170083 Objective Loss 0.170083 LR 0.000500 Time 0.023887 +2023-10-02 21:35:29,666 - Epoch: [138][ 270/ 1236] Overall Loss 0.170379 Objective Loss 0.170379 LR 0.000500 Time 0.023762 +2023-10-02 21:35:29,875 - Epoch: [138][ 280/ 1236] Overall Loss 0.170199 Objective Loss 0.170199 LR 0.000500 Time 0.023658 +2023-10-02 21:35:30,082 - Epoch: [138][ 290/ 1236] Overall Loss 0.170366 Objective Loss 0.170366 LR 0.000500 Time 0.023550 +2023-10-02 21:35:30,292 - Epoch: [138][ 300/ 1236] Overall Loss 0.170172 Objective Loss 0.170172 LR 0.000500 Time 0.023465 +2023-10-02 21:35:30,497 - Epoch: [138][ 310/ 1236] Overall Loss 0.170192 Objective Loss 0.170192 LR 0.000500 Time 0.023371 +2023-10-02 21:35:30,708 - Epoch: [138][ 320/ 1236] Overall Loss 0.169278 Objective Loss 0.169278 LR 0.000500 Time 0.023299 +2023-10-02 21:35:30,915 - Epoch: [138][ 330/ 1236] Overall Loss 0.169614 Objective Loss 0.169614 LR 0.000500 Time 0.023218 +2023-10-02 21:35:31,126 - Epoch: [138][ 340/ 1236] Overall Loss 0.169262 Objective Loss 0.169262 LR 0.000500 Time 0.023154 +2023-10-02 21:35:31,333 - Epoch: [138][ 350/ 1236] Overall Loss 0.169469 Objective Loss 0.169469 LR 0.000500 Time 0.023083 +2023-10-02 21:35:31,543 - Epoch: [138][ 360/ 1236] Overall Loss 0.169239 Objective Loss 0.169239 LR 0.000500 Time 0.023026 +2023-10-02 21:35:31,750 - Epoch: [138][ 370/ 1236] Overall Loss 0.168944 Objective Loss 0.168944 LR 0.000500 Time 0.022962 +2023-10-02 21:35:31,961 - Epoch: [138][ 380/ 1236] Overall Loss 0.169006 Objective Loss 0.169006 LR 0.000500 Time 0.022911 +2023-10-02 21:35:32,167 - Epoch: [138][ 390/ 1236] Overall Loss 0.169615 Objective Loss 0.169615 LR 0.000500 Time 0.022852 +2023-10-02 21:35:32,378 - Epoch: [138][ 400/ 1236] Overall Loss 0.169349 Objective Loss 0.169349 LR 0.000500 Time 0.022807 +2023-10-02 21:35:32,584 - Epoch: [138][ 410/ 1236] Overall Loss 0.169142 Objective Loss 0.169142 LR 0.000500 Time 0.022754 +2023-10-02 21:35:32,795 - Epoch: [138][ 420/ 1236] Overall Loss 0.169176 Objective Loss 0.169176 LR 0.000500 Time 0.022713 +2023-10-02 21:35:33,002 - Epoch: [138][ 430/ 1236] Overall Loss 0.169243 Objective Loss 0.169243 LR 0.000500 Time 0.022665 +2023-10-02 21:35:33,213 - Epoch: [138][ 440/ 1236] Overall Loss 0.168997 Objective Loss 0.168997 LR 0.000500 Time 0.022628 +2023-10-02 21:35:33,419 - Epoch: [138][ 450/ 1236] Overall Loss 0.169221 Objective Loss 0.169221 LR 0.000500 Time 0.022584 +2023-10-02 21:35:33,630 - Epoch: [138][ 460/ 1236] Overall Loss 0.169229 Objective Loss 0.169229 LR 0.000500 Time 0.022550 +2023-10-02 21:35:33,836 - Epoch: [138][ 470/ 1236] Overall Loss 0.169024 Objective Loss 0.169024 LR 0.000500 Time 0.022509 +2023-10-02 21:35:34,047 - Epoch: [138][ 480/ 1236] Overall Loss 0.169156 Objective Loss 0.169156 LR 0.000500 Time 0.022478 +2023-10-02 21:35:34,253 - Epoch: [138][ 490/ 1236] Overall Loss 0.168832 Objective Loss 0.168832 LR 0.000500 Time 0.022440 +2023-10-02 21:35:34,470 - Epoch: [138][ 500/ 1236] Overall Loss 0.169221 Objective Loss 0.169221 LR 0.000500 Time 0.022425 +2023-10-02 21:35:34,685 - Epoch: [138][ 510/ 1236] Overall Loss 0.169578 Objective Loss 0.169578 LR 0.000500 Time 0.022405 +2023-10-02 21:35:34,905 - Epoch: [138][ 520/ 1236] Overall Loss 0.169414 Objective Loss 0.169414 LR 0.000500 Time 0.022397 +2023-10-02 21:35:35,120 - Epoch: [138][ 530/ 1236] Overall Loss 0.169127 Objective Loss 0.169127 LR 0.000500 Time 0.022379 +2023-10-02 21:35:35,340 - Epoch: [138][ 540/ 1236] Overall Loss 0.169216 Objective Loss 0.169216 LR 0.000500 Time 0.022372 +2023-10-02 21:35:35,555 - Epoch: [138][ 550/ 1236] Overall Loss 0.169960 Objective Loss 0.169960 LR 0.000500 Time 0.022355 +2023-10-02 21:35:35,775 - Epoch: [138][ 560/ 1236] Overall Loss 0.169900 Objective Loss 0.169900 LR 0.000500 Time 0.022349 +2023-10-02 21:35:35,990 - Epoch: [138][ 570/ 1236] Overall Loss 0.169939 Objective Loss 0.169939 LR 0.000500 Time 0.022332 +2023-10-02 21:35:36,210 - Epoch: [138][ 580/ 1236] Overall Loss 0.169962 Objective Loss 0.169962 LR 0.000500 Time 0.022327 +2023-10-02 21:35:36,425 - Epoch: [138][ 590/ 1236] Overall Loss 0.170368 Objective Loss 0.170368 LR 0.000500 Time 0.022311 +2023-10-02 21:35:36,645 - Epoch: [138][ 600/ 1236] Overall Loss 0.170664 Objective Loss 0.170664 LR 0.000500 Time 0.022306 +2023-10-02 21:35:36,860 - Epoch: [138][ 610/ 1236] Overall Loss 0.170938 Objective Loss 0.170938 LR 0.000500 Time 0.022293 +2023-10-02 21:35:37,080 - Epoch: [138][ 620/ 1236] Overall Loss 0.171103 Objective Loss 0.171103 LR 0.000500 Time 0.022288 +2023-10-02 21:35:37,295 - Epoch: [138][ 630/ 1236] Overall Loss 0.171069 Objective Loss 0.171069 LR 0.000500 Time 0.022274 +2023-10-02 21:35:37,516 - Epoch: [138][ 640/ 1236] Overall Loss 0.170929 Objective Loss 0.170929 LR 0.000500 Time 0.022270 +2023-10-02 21:35:37,730 - Epoch: [138][ 650/ 1236] Overall Loss 0.171039 Objective Loss 0.171039 LR 0.000500 Time 0.022257 +2023-10-02 21:35:37,950 - Epoch: [138][ 660/ 1236] Overall Loss 0.171474 Objective Loss 0.171474 LR 0.000500 Time 0.022253 +2023-10-02 21:35:38,165 - Epoch: [138][ 670/ 1236] Overall Loss 0.171371 Objective Loss 0.171371 LR 0.000500 Time 0.022241 +2023-10-02 21:35:38,385 - Epoch: [138][ 680/ 1236] Overall Loss 0.171106 Objective Loss 0.171106 LR 0.000500 Time 0.022237 +2023-10-02 21:35:38,600 - Epoch: [138][ 690/ 1236] Overall Loss 0.171272 Objective Loss 0.171272 LR 0.000500 Time 0.022226 +2023-10-02 21:35:38,821 - Epoch: [138][ 700/ 1236] Overall Loss 0.171682 Objective Loss 0.171682 LR 0.000500 Time 0.022222 +2023-10-02 21:35:39,038 - Epoch: [138][ 710/ 1236] Overall Loss 0.171598 Objective Loss 0.171598 LR 0.000500 Time 0.022215 +2023-10-02 21:35:39,262 - Epoch: [138][ 720/ 1236] Overall Loss 0.171852 Objective Loss 0.171852 LR 0.000500 Time 0.022215 +2023-10-02 21:35:39,481 - Epoch: [138][ 730/ 1236] Overall Loss 0.172060 Objective Loss 0.172060 LR 0.000500 Time 0.022210 +2023-10-02 21:35:39,705 - Epoch: [138][ 740/ 1236] Overall Loss 0.171895 Objective Loss 0.171895 LR 0.000500 Time 0.022212 +2023-10-02 21:35:39,924 - Epoch: [138][ 750/ 1236] Overall Loss 0.171821 Objective Loss 0.171821 LR 0.000500 Time 0.022208 +2023-10-02 21:35:40,148 - Epoch: [138][ 760/ 1236] Overall Loss 0.172027 Objective Loss 0.172027 LR 0.000500 Time 0.022209 +2023-10-02 21:35:40,367 - Epoch: [138][ 770/ 1236] Overall Loss 0.171826 Objective Loss 0.171826 LR 0.000500 Time 0.022205 +2023-10-02 21:35:40,590 - Epoch: [138][ 780/ 1236] Overall Loss 0.171987 Objective Loss 0.171987 LR 0.000500 Time 0.022207 +2023-10-02 21:35:40,809 - Epoch: [138][ 790/ 1236] Overall Loss 0.172105 Objective Loss 0.172105 LR 0.000500 Time 0.022202 +2023-10-02 21:35:41,023 - Epoch: [138][ 800/ 1236] Overall Loss 0.171745 Objective Loss 0.171745 LR 0.000500 Time 0.022192 +2023-10-02 21:35:41,234 - Epoch: [138][ 810/ 1236] Overall Loss 0.171648 Objective Loss 0.171648 LR 0.000500 Time 0.022177 +2023-10-02 21:35:41,448 - Epoch: [138][ 820/ 1236] Overall Loss 0.171671 Objective Loss 0.171671 LR 0.000500 Time 0.022167 +2023-10-02 21:35:41,659 - Epoch: [138][ 830/ 1236] Overall Loss 0.171553 Objective Loss 0.171553 LR 0.000500 Time 0.022154 +2023-10-02 21:35:41,873 - Epoch: [138][ 840/ 1236] Overall Loss 0.171736 Objective Loss 0.171736 LR 0.000500 Time 0.022145 +2023-10-02 21:35:42,090 - Epoch: [138][ 850/ 1236] Overall Loss 0.171797 Objective Loss 0.171797 LR 0.000500 Time 0.022132 +2023-10-02 21:35:42,304 - Epoch: [138][ 860/ 1236] Overall Loss 0.171814 Objective Loss 0.171814 LR 0.000500 Time 0.022123 +2023-10-02 21:35:42,515 - Epoch: [138][ 870/ 1236] Overall Loss 0.171657 Objective Loss 0.171657 LR 0.000500 Time 0.022111 +2023-10-02 21:35:42,729 - Epoch: [138][ 880/ 1236] Overall Loss 0.171770 Objective Loss 0.171770 LR 0.000500 Time 0.022103 +2023-10-02 21:35:42,940 - Epoch: [138][ 890/ 1236] Overall Loss 0.172055 Objective Loss 0.172055 LR 0.000500 Time 0.022091 +2023-10-02 21:35:43,154 - Epoch: [138][ 900/ 1236] Overall Loss 0.171907 Objective Loss 0.171907 LR 0.000500 Time 0.022083 +2023-10-02 21:35:43,365 - Epoch: [138][ 910/ 1236] Overall Loss 0.171782 Objective Loss 0.171782 LR 0.000500 Time 0.022071 +2023-10-02 21:35:43,579 - Epoch: [138][ 920/ 1236] Overall Loss 0.172213 Objective Loss 0.172213 LR 0.000500 Time 0.022064 +2023-10-02 21:35:43,790 - Epoch: [138][ 930/ 1236] Overall Loss 0.172305 Objective Loss 0.172305 LR 0.000500 Time 0.022053 +2023-10-02 21:35:44,004 - Epoch: [138][ 940/ 1236] Overall Loss 0.172491 Objective Loss 0.172491 LR 0.000500 Time 0.022046 +2023-10-02 21:35:44,215 - Epoch: [138][ 950/ 1236] Overall Loss 0.172453 Objective Loss 0.172453 LR 0.000500 Time 0.022035 +2023-10-02 21:35:44,430 - Epoch: [138][ 960/ 1236] Overall Loss 0.172355 Objective Loss 0.172355 LR 0.000500 Time 0.022030 +2023-10-02 21:35:44,642 - Epoch: [138][ 970/ 1236] Overall Loss 0.172424 Objective Loss 0.172424 LR 0.000500 Time 0.022021 +2023-10-02 21:35:44,856 - Epoch: [138][ 980/ 1236] Overall Loss 0.172269 Objective Loss 0.172269 LR 0.000500 Time 0.022014 +2023-10-02 21:35:45,067 - Epoch: [138][ 990/ 1236] Overall Loss 0.172340 Objective Loss 0.172340 LR 0.000500 Time 0.022004 +2023-10-02 21:35:45,281 - Epoch: [138][ 1000/ 1236] Overall Loss 0.172569 Objective Loss 0.172569 LR 0.000500 Time 0.021998 +2023-10-02 21:35:45,492 - Epoch: [138][ 1010/ 1236] Overall Loss 0.172409 Objective Loss 0.172409 LR 0.000500 Time 0.021989 +2023-10-02 21:35:45,705 - Epoch: [138][ 1020/ 1236] Overall Loss 0.172529 Objective Loss 0.172529 LR 0.000500 Time 0.021982 +2023-10-02 21:35:45,918 - Epoch: [138][ 1030/ 1236] Overall Loss 0.172402 Objective Loss 0.172402 LR 0.000500 Time 0.021973 +2023-10-02 21:35:46,131 - Epoch: [138][ 1040/ 1236] Overall Loss 0.172543 Objective Loss 0.172543 LR 0.000500 Time 0.021967 +2023-10-02 21:35:46,343 - Epoch: [138][ 1050/ 1236] Overall Loss 0.172403 Objective Loss 0.172403 LR 0.000500 Time 0.021958 +2023-10-02 21:35:46,556 - Epoch: [138][ 1060/ 1236] Overall Loss 0.172475 Objective Loss 0.172475 LR 0.000500 Time 0.021951 +2023-10-02 21:35:46,768 - Epoch: [138][ 1070/ 1236] Overall Loss 0.172195 Objective Loss 0.172195 LR 0.000500 Time 0.021943 +2023-10-02 21:35:46,981 - Epoch: [138][ 1080/ 1236] Overall Loss 0.172281 Objective Loss 0.172281 LR 0.000500 Time 0.021936 +2023-10-02 21:35:47,193 - Epoch: [138][ 1090/ 1236] Overall Loss 0.172186 Objective Loss 0.172186 LR 0.000500 Time 0.021928 +2023-10-02 21:35:47,406 - Epoch: [138][ 1100/ 1236] Overall Loss 0.172052 Objective Loss 0.172052 LR 0.000500 Time 0.021922 +2023-10-02 21:35:47,618 - Epoch: [138][ 1110/ 1236] Overall Loss 0.172137 Objective Loss 0.172137 LR 0.000500 Time 0.021914 +2023-10-02 21:35:47,831 - Epoch: [138][ 1120/ 1236] Overall Loss 0.172056 Objective Loss 0.172056 LR 0.000500 Time 0.021908 +2023-10-02 21:35:48,043 - Epoch: [138][ 1130/ 1236] Overall Loss 0.171978 Objective Loss 0.171978 LR 0.000500 Time 0.021901 +2023-10-02 21:35:48,256 - Epoch: [138][ 1140/ 1236] Overall Loss 0.171865 Objective Loss 0.171865 LR 0.000500 Time 0.021895 +2023-10-02 21:35:48,468 - Epoch: [138][ 1150/ 1236] Overall Loss 0.171885 Objective Loss 0.171885 LR 0.000500 Time 0.021887 +2023-10-02 21:35:48,681 - Epoch: [138][ 1160/ 1236] Overall Loss 0.171994 Objective Loss 0.171994 LR 0.000500 Time 0.021882 +2023-10-02 21:35:48,893 - Epoch: [138][ 1170/ 1236] Overall Loss 0.172111 Objective Loss 0.172111 LR 0.000500 Time 0.021875 +2023-10-02 21:35:49,107 - Epoch: [138][ 1180/ 1236] Overall Loss 0.171969 Objective Loss 0.171969 LR 0.000500 Time 0.021871 +2023-10-02 21:35:49,318 - Epoch: [138][ 1190/ 1236] Overall Loss 0.171962 Objective Loss 0.171962 LR 0.000500 Time 0.021864 +2023-10-02 21:35:49,531 - Epoch: [138][ 1200/ 1236] Overall Loss 0.172217 Objective Loss 0.172217 LR 0.000500 Time 0.021859 +2023-10-02 21:35:49,743 - Epoch: [138][ 1210/ 1236] Overall Loss 0.172393 Objective Loss 0.172393 LR 0.000500 Time 0.021852 +2023-10-02 21:35:49,956 - Epoch: [138][ 1220/ 1236] Overall Loss 0.172249 Objective Loss 0.172249 LR 0.000500 Time 0.021847 +2023-10-02 21:35:50,220 - Epoch: [138][ 1230/ 1236] Overall Loss 0.172405 Objective Loss 0.172405 LR 0.000500 Time 0.021883 +2023-10-02 21:35:50,343 - Epoch: [138][ 1236/ 1236] Overall Loss 0.172421 Objective Loss 0.172421 Top1 90.224033 Top5 99.185336 LR 0.000500 Time 0.021876 +2023-10-02 21:35:50,466 - --- validate (epoch=138)----------- +2023-10-02 21:35:50,466 - 29943 samples (256 per mini-batch) +2023-10-02 21:35:50,945 - Epoch: [138][ 10/ 117] Loss 0.327367 Top1 86.015625 Top5 98.593750 +2023-10-02 21:35:51,097 - Epoch: [138][ 20/ 117] Loss 0.327450 Top1 85.507812 Top5 98.339844 +2023-10-02 21:35:51,249 - Epoch: [138][ 30/ 117] Loss 0.314974 Top1 85.598958 Top5 98.372396 +2023-10-02 21:35:51,401 - Epoch: [138][ 40/ 117] Loss 0.315050 Top1 85.664062 Top5 98.437500 +2023-10-02 21:35:51,553 - Epoch: [138][ 50/ 117] Loss 0.308967 Top1 85.750000 Top5 98.414062 +2023-10-02 21:35:51,707 - Epoch: [138][ 60/ 117] Loss 0.307627 Top1 85.774740 Top5 98.398438 +2023-10-02 21:35:51,860 - Epoch: [138][ 70/ 117] Loss 0.311569 Top1 85.837054 Top5 98.415179 +2023-10-02 21:35:52,013 - Epoch: [138][ 80/ 117] Loss 0.306331 Top1 85.976562 Top5 98.447266 +2023-10-02 21:35:52,167 - Epoch: [138][ 90/ 117] Loss 0.305676 Top1 86.011285 Top5 98.459201 +2023-10-02 21:35:52,320 - Epoch: [138][ 100/ 117] Loss 0.301666 Top1 86.257812 Top5 98.468750 +2023-10-02 21:35:52,479 - Epoch: [138][ 110/ 117] Loss 0.300429 Top1 86.289062 Top5 98.476562 +2023-10-02 21:35:52,569 - Epoch: [138][ 117/ 117] Loss 0.301958 Top1 86.313997 Top5 98.463748 +2023-10-02 21:35:52,713 - ==> Top1: 86.314 Top5: 98.464 Loss: 0.302 + +2023-10-02 21:35:52,714 - ==> Confusion: +[[ 946 1 4 0 9 3 0 1 3 46 2 1 0 2 6 3 2 1 1 0 19] + [ 1 1058 0 1 3 26 2 14 1 0 1 1 3 0 0 3 1 0 6 1 9] + [ 2 1 975 10 1 1 17 4 0 0 2 3 8 3 1 4 2 2 11 4 5] + [ 1 4 12 981 1 0 3 0 5 0 3 1 5 2 27 4 1 3 14 2 20] + [ 27 6 1 1 964 6 0 0 2 5 0 1 0 3 9 6 10 0 2 1 6] + [ 2 42 1 1 2 996 3 23 1 4 1 7 2 7 5 2 3 1 2 1 10] + [ 0 3 20 1 0 2 1138 3 0 0 2 1 0 0 0 5 0 1 0 7 8] + [ 2 12 18 0 8 25 8 1060 1 2 7 3 4 3 3 1 0 2 43 8 8] + [ 25 0 1 0 5 2 0 1 946 44 14 3 2 14 19 1 5 1 3 0 3] + [ 101 0 1 0 5 3 0 0 20 942 1 1 0 19 12 3 1 0 0 0 10] + [ 4 4 6 7 0 1 4 5 7 0 976 1 1 9 10 1 1 3 6 1 6] + [ 0 1 1 0 0 14 0 2 0 0 0 960 21 8 0 2 0 15 0 8 3] + [ 0 0 4 4 0 1 4 0 2 0 3 34 969 1 3 9 0 19 0 4 11] + [ 0 0 0 0 2 8 0 0 17 7 5 7 0 1051 7 0 0 1 0 2 12] + [ 12 1 4 19 3 0 0 0 19 2 2 0 1 1 1018 0 0 2 7 0 10] + [ 0 0 2 1 3 0 0 1 0 0 1 5 9 0 0 1073 17 12 1 4 5] + [ 1 18 1 1 3 8 1 0 0 0 0 5 1 2 3 10 1089 0 1 6 11] + [ 0 0 0 2 0 0 3 0 0 1 0 1 18 3 3 7 1 996 0 1 2] + [ 2 5 6 17 0 2 0 11 1 1 2 1 1 0 14 0 0 1 992 1 11] + [ 0 2 2 1 1 4 13 6 0 0 1 14 3 1 0 1 7 1 0 1086 9] + [ 108 148 110 87 65 114 31 81 91 64 166 94 270 251 113 70 72 56 111 174 5629]] + +2023-10-02 21:35:52,715 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:35:52,715 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:35:52,721 - + +2023-10-02 21:35:52,721 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:35:53,873 - Epoch: [139][ 10/ 1236] Overall Loss 0.163737 Objective Loss 0.163737 LR 0.000500 Time 0.115140 +2023-10-02 21:35:54,082 - Epoch: [139][ 20/ 1236] Overall Loss 0.158030 Objective Loss 0.158030 LR 0.000500 Time 0.068007 +2023-10-02 21:35:54,291 - Epoch: [139][ 30/ 1236] Overall Loss 0.154851 Objective Loss 0.154851 LR 0.000500 Time 0.052283 +2023-10-02 21:35:54,500 - Epoch: [139][ 40/ 1236] Overall Loss 0.158090 Objective Loss 0.158090 LR 0.000500 Time 0.044435 +2023-10-02 21:35:54,709 - Epoch: [139][ 50/ 1236] Overall Loss 0.163864 Objective Loss 0.163864 LR 0.000500 Time 0.039710 +2023-10-02 21:35:54,918 - Epoch: [139][ 60/ 1236] Overall Loss 0.161344 Objective Loss 0.161344 LR 0.000500 Time 0.036582 +2023-10-02 21:35:55,127 - Epoch: [139][ 70/ 1236] Overall Loss 0.158717 Objective Loss 0.158717 LR 0.000500 Time 0.034334 +2023-10-02 21:35:55,337 - Epoch: [139][ 80/ 1236] Overall Loss 0.162896 Objective Loss 0.162896 LR 0.000500 Time 0.032656 +2023-10-02 21:35:55,545 - Epoch: [139][ 90/ 1236] Overall Loss 0.160700 Objective Loss 0.160700 LR 0.000500 Time 0.031344 +2023-10-02 21:35:55,755 - Epoch: [139][ 100/ 1236] Overall Loss 0.161339 Objective Loss 0.161339 LR 0.000500 Time 0.030307 +2023-10-02 21:35:55,965 - Epoch: [139][ 110/ 1236] Overall Loss 0.163222 Objective Loss 0.163222 LR 0.000500 Time 0.029459 +2023-10-02 21:35:56,175 - Epoch: [139][ 120/ 1236] Overall Loss 0.164125 Objective Loss 0.164125 LR 0.000500 Time 0.028748 +2023-10-02 21:35:56,384 - Epoch: [139][ 130/ 1236] Overall Loss 0.163764 Objective Loss 0.163764 LR 0.000500 Time 0.028138 +2023-10-02 21:35:56,591 - Epoch: [139][ 140/ 1236] Overall Loss 0.165003 Objective Loss 0.165003 LR 0.000500 Time 0.027610 +2023-10-02 21:35:56,799 - Epoch: [139][ 150/ 1236] Overall Loss 0.165111 Objective Loss 0.165111 LR 0.000500 Time 0.027144 +2023-10-02 21:35:57,008 - Epoch: [139][ 160/ 1236] Overall Loss 0.164133 Objective Loss 0.164133 LR 0.000500 Time 0.026752 +2023-10-02 21:35:57,215 - Epoch: [139][ 170/ 1236] Overall Loss 0.163647 Objective Loss 0.163647 LR 0.000500 Time 0.026384 +2023-10-02 21:35:57,423 - Epoch: [139][ 180/ 1236] Overall Loss 0.162894 Objective Loss 0.162894 LR 0.000500 Time 0.026076 +2023-10-02 21:35:57,631 - Epoch: [139][ 190/ 1236] Overall Loss 0.164794 Objective Loss 0.164794 LR 0.000500 Time 0.025790 +2023-10-02 21:35:57,840 - Epoch: [139][ 200/ 1236] Overall Loss 0.165832 Objective Loss 0.165832 LR 0.000500 Time 0.025543 +2023-10-02 21:35:58,048 - Epoch: [139][ 210/ 1236] Overall Loss 0.167151 Objective Loss 0.167151 LR 0.000500 Time 0.025310 +2023-10-02 21:35:58,255 - Epoch: [139][ 220/ 1236] Overall Loss 0.166943 Objective Loss 0.166943 LR 0.000500 Time 0.025101 +2023-10-02 21:35:58,463 - Epoch: [139][ 230/ 1236] Overall Loss 0.167737 Objective Loss 0.167737 LR 0.000500 Time 0.024912 +2023-10-02 21:35:58,672 - Epoch: [139][ 240/ 1236] Overall Loss 0.167082 Objective Loss 0.167082 LR 0.000500 Time 0.024742 +2023-10-02 21:35:58,880 - Epoch: [139][ 250/ 1236] Overall Loss 0.167515 Objective Loss 0.167515 LR 0.000500 Time 0.024578 +2023-10-02 21:35:59,089 - Epoch: [139][ 260/ 1236] Overall Loss 0.168174 Objective Loss 0.168174 LR 0.000500 Time 0.024435 +2023-10-02 21:35:59,296 - Epoch: [139][ 270/ 1236] Overall Loss 0.168485 Objective Loss 0.168485 LR 0.000500 Time 0.024294 +2023-10-02 21:35:59,504 - Epoch: [139][ 280/ 1236] Overall Loss 0.168800 Objective Loss 0.168800 LR 0.000500 Time 0.024168 +2023-10-02 21:35:59,711 - Epoch: [139][ 290/ 1236] Overall Loss 0.169348 Objective Loss 0.169348 LR 0.000500 Time 0.024041 +2023-10-02 21:35:59,920 - Epoch: [139][ 300/ 1236] Overall Loss 0.169768 Objective Loss 0.169768 LR 0.000500 Time 0.023936 +2023-10-02 21:36:00,128 - Epoch: [139][ 310/ 1236] Overall Loss 0.170169 Objective Loss 0.170169 LR 0.000500 Time 0.023830 +2023-10-02 21:36:00,337 - Epoch: [139][ 320/ 1236] Overall Loss 0.170935 Objective Loss 0.170935 LR 0.000500 Time 0.023737 +2023-10-02 21:36:00,543 - Epoch: [139][ 330/ 1236] Overall Loss 0.170892 Objective Loss 0.170892 LR 0.000500 Time 0.023639 +2023-10-02 21:36:00,754 - Epoch: [139][ 340/ 1236] Overall Loss 0.171223 Objective Loss 0.171223 LR 0.000500 Time 0.023561 +2023-10-02 21:36:00,962 - Epoch: [139][ 350/ 1236] Overall Loss 0.170259 Objective Loss 0.170259 LR 0.000500 Time 0.023478 +2023-10-02 21:36:01,173 - Epoch: [139][ 360/ 1236] Overall Loss 0.169635 Objective Loss 0.169635 LR 0.000500 Time 0.023411 +2023-10-02 21:36:01,382 - Epoch: [139][ 370/ 1236] Overall Loss 0.169155 Objective Loss 0.169155 LR 0.000500 Time 0.023344 +2023-10-02 21:36:01,594 - Epoch: [139][ 380/ 1236] Overall Loss 0.168838 Objective Loss 0.168838 LR 0.000500 Time 0.023287 +2023-10-02 21:36:01,802 - Epoch: [139][ 390/ 1236] Overall Loss 0.169403 Objective Loss 0.169403 LR 0.000500 Time 0.023223 +2023-10-02 21:36:02,013 - Epoch: [139][ 400/ 1236] Overall Loss 0.169743 Objective Loss 0.169743 LR 0.000500 Time 0.023168 +2023-10-02 21:36:02,222 - Epoch: [139][ 410/ 1236] Overall Loss 0.169251 Objective Loss 0.169251 LR 0.000500 Time 0.023113 +2023-10-02 21:36:02,433 - Epoch: [139][ 420/ 1236] Overall Loss 0.169422 Objective Loss 0.169422 LR 0.000500 Time 0.023064 +2023-10-02 21:36:02,642 - Epoch: [139][ 430/ 1236] Overall Loss 0.170068 Objective Loss 0.170068 LR 0.000500 Time 0.023013 +2023-10-02 21:36:02,853 - Epoch: [139][ 440/ 1236] Overall Loss 0.170045 Objective Loss 0.170045 LR 0.000500 Time 0.022969 +2023-10-02 21:36:03,062 - Epoch: [139][ 450/ 1236] Overall Loss 0.170044 Objective Loss 0.170044 LR 0.000500 Time 0.022923 +2023-10-02 21:36:03,273 - Epoch: [139][ 460/ 1236] Overall Loss 0.169955 Objective Loss 0.169955 LR 0.000500 Time 0.022882 +2023-10-02 21:36:03,483 - Epoch: [139][ 470/ 1236] Overall Loss 0.169742 Objective Loss 0.169742 LR 0.000500 Time 0.022840 +2023-10-02 21:36:03,694 - Epoch: [139][ 480/ 1236] Overall Loss 0.169818 Objective Loss 0.169818 LR 0.000500 Time 0.022803 +2023-10-02 21:36:03,903 - Epoch: [139][ 490/ 1236] Overall Loss 0.169927 Objective Loss 0.169927 LR 0.000500 Time 0.022765 +2023-10-02 21:36:04,113 - Epoch: [139][ 500/ 1236] Overall Loss 0.169682 Objective Loss 0.169682 LR 0.000500 Time 0.022730 +2023-10-02 21:36:04,322 - Epoch: [139][ 510/ 1236] Overall Loss 0.169484 Objective Loss 0.169484 LR 0.000500 Time 0.022693 +2023-10-02 21:36:04,533 - Epoch: [139][ 520/ 1236] Overall Loss 0.169560 Objective Loss 0.169560 LR 0.000500 Time 0.022662 +2023-10-02 21:36:04,743 - Epoch: [139][ 530/ 1236] Overall Loss 0.169404 Objective Loss 0.169404 LR 0.000500 Time 0.022629 +2023-10-02 21:36:04,953 - Epoch: [139][ 540/ 1236] Overall Loss 0.169536 Objective Loss 0.169536 LR 0.000500 Time 0.022600 +2023-10-02 21:36:05,162 - Epoch: [139][ 550/ 1236] Overall Loss 0.169737 Objective Loss 0.169737 LR 0.000500 Time 0.022568 +2023-10-02 21:36:05,373 - Epoch: [139][ 560/ 1236] Overall Loss 0.169446 Objective Loss 0.169446 LR 0.000500 Time 0.022541 +2023-10-02 21:36:05,583 - Epoch: [139][ 570/ 1236] Overall Loss 0.169471 Objective Loss 0.169471 LR 0.000500 Time 0.022512 +2023-10-02 21:36:05,793 - Epoch: [139][ 580/ 1236] Overall Loss 0.169195 Objective Loss 0.169195 LR 0.000500 Time 0.022487 +2023-10-02 21:36:06,003 - Epoch: [139][ 590/ 1236] Overall Loss 0.169491 Objective Loss 0.169491 LR 0.000500 Time 0.022461 +2023-10-02 21:36:06,214 - Epoch: [139][ 600/ 1236] Overall Loss 0.169390 Objective Loss 0.169390 LR 0.000500 Time 0.022438 +2023-10-02 21:36:06,424 - Epoch: [139][ 610/ 1236] Overall Loss 0.169690 Objective Loss 0.169690 LR 0.000500 Time 0.022413 +2023-10-02 21:36:06,634 - Epoch: [139][ 620/ 1236] Overall Loss 0.169466 Objective Loss 0.169466 LR 0.000500 Time 0.022391 +2023-10-02 21:36:06,843 - Epoch: [139][ 630/ 1236] Overall Loss 0.169608 Objective Loss 0.169608 LR 0.000500 Time 0.022367 +2023-10-02 21:36:07,056 - Epoch: [139][ 640/ 1236] Overall Loss 0.169685 Objective Loss 0.169685 LR 0.000500 Time 0.022349 +2023-10-02 21:36:07,264 - Epoch: [139][ 650/ 1236] Overall Loss 0.169898 Objective Loss 0.169898 LR 0.000500 Time 0.022324 +2023-10-02 21:36:07,475 - Epoch: [139][ 660/ 1236] Overall Loss 0.169731 Objective Loss 0.169731 LR 0.000500 Time 0.022305 +2023-10-02 21:36:07,684 - Epoch: [139][ 670/ 1236] Overall Loss 0.169995 Objective Loss 0.169995 LR 0.000500 Time 0.022285 +2023-10-02 21:36:07,895 - Epoch: [139][ 680/ 1236] Overall Loss 0.169867 Objective Loss 0.169867 LR 0.000500 Time 0.022267 +2023-10-02 21:36:08,105 - Epoch: [139][ 690/ 1236] Overall Loss 0.169936 Objective Loss 0.169936 LR 0.000500 Time 0.022247 +2023-10-02 21:36:08,317 - Epoch: [139][ 700/ 1236] Overall Loss 0.169676 Objective Loss 0.169676 LR 0.000500 Time 0.022232 +2023-10-02 21:36:08,525 - Epoch: [139][ 710/ 1236] Overall Loss 0.169752 Objective Loss 0.169752 LR 0.000500 Time 0.022212 +2023-10-02 21:36:08,735 - Epoch: [139][ 720/ 1236] Overall Loss 0.169811 Objective Loss 0.169811 LR 0.000500 Time 0.022195 +2023-10-02 21:36:08,945 - Epoch: [139][ 730/ 1236] Overall Loss 0.169634 Objective Loss 0.169634 LR 0.000500 Time 0.022178 +2023-10-02 21:36:09,157 - Epoch: [139][ 740/ 1236] Overall Loss 0.170012 Objective Loss 0.170012 LR 0.000500 Time 0.022164 +2023-10-02 21:36:09,365 - Epoch: [139][ 750/ 1236] Overall Loss 0.170038 Objective Loss 0.170038 LR 0.000500 Time 0.022145 +2023-10-02 21:36:09,575 - Epoch: [139][ 760/ 1236] Overall Loss 0.170153 Objective Loss 0.170153 LR 0.000500 Time 0.022131 +2023-10-02 21:36:09,785 - Epoch: [139][ 770/ 1236] Overall Loss 0.170231 Objective Loss 0.170231 LR 0.000500 Time 0.022115 +2023-10-02 21:36:09,995 - Epoch: [139][ 780/ 1236] Overall Loss 0.170493 Objective Loss 0.170493 LR 0.000500 Time 0.022101 +2023-10-02 21:36:10,205 - Epoch: [139][ 790/ 1236] Overall Loss 0.170683 Objective Loss 0.170683 LR 0.000500 Time 0.022086 +2023-10-02 21:36:10,416 - Epoch: [139][ 800/ 1236] Overall Loss 0.170854 Objective Loss 0.170854 LR 0.000500 Time 0.022073 +2023-10-02 21:36:10,625 - Epoch: [139][ 810/ 1236] Overall Loss 0.170724 Objective Loss 0.170724 LR 0.000500 Time 0.022058 +2023-10-02 21:36:10,836 - Epoch: [139][ 820/ 1236] Overall Loss 0.170788 Objective Loss 0.170788 LR 0.000500 Time 0.022046 +2023-10-02 21:36:11,046 - Epoch: [139][ 830/ 1236] Overall Loss 0.170879 Objective Loss 0.170879 LR 0.000500 Time 0.022032 +2023-10-02 21:36:11,258 - Epoch: [139][ 840/ 1236] Overall Loss 0.171160 Objective Loss 0.171160 LR 0.000500 Time 0.022022 +2023-10-02 21:36:11,466 - Epoch: [139][ 850/ 1236] Overall Loss 0.171406 Objective Loss 0.171406 LR 0.000500 Time 0.022007 +2023-10-02 21:36:11,676 - Epoch: [139][ 860/ 1236] Overall Loss 0.171208 Objective Loss 0.171208 LR 0.000500 Time 0.021996 +2023-10-02 21:36:11,886 - Epoch: [139][ 870/ 1236] Overall Loss 0.171056 Objective Loss 0.171056 LR 0.000500 Time 0.021983 +2023-10-02 21:36:12,097 - Epoch: [139][ 880/ 1236] Overall Loss 0.171140 Objective Loss 0.171140 LR 0.000500 Time 0.021973 +2023-10-02 21:36:12,307 - Epoch: [139][ 890/ 1236] Overall Loss 0.171322 Objective Loss 0.171322 LR 0.000500 Time 0.021961 +2023-10-02 21:36:12,517 - Epoch: [139][ 900/ 1236] Overall Loss 0.171488 Objective Loss 0.171488 LR 0.000500 Time 0.021952 +2023-10-02 21:36:12,727 - Epoch: [139][ 910/ 1236] Overall Loss 0.171669 Objective Loss 0.171669 LR 0.000500 Time 0.021940 +2023-10-02 21:36:12,938 - Epoch: [139][ 920/ 1236] Overall Loss 0.171715 Objective Loss 0.171715 LR 0.000500 Time 0.021930 +2023-10-02 21:36:13,147 - Epoch: [139][ 930/ 1236] Overall Loss 0.171835 Objective Loss 0.171835 LR 0.000500 Time 0.021920 +2023-10-02 21:36:13,358 - Epoch: [139][ 940/ 1236] Overall Loss 0.171796 Objective Loss 0.171796 LR 0.000500 Time 0.021911 +2023-10-02 21:36:13,568 - Epoch: [139][ 950/ 1236] Overall Loss 0.171885 Objective Loss 0.171885 LR 0.000500 Time 0.021899 +2023-10-02 21:36:13,779 - Epoch: [139][ 960/ 1236] Overall Loss 0.171979 Objective Loss 0.171979 LR 0.000500 Time 0.021890 +2023-10-02 21:36:13,988 - Epoch: [139][ 970/ 1236] Overall Loss 0.171896 Objective Loss 0.171896 LR 0.000500 Time 0.021881 +2023-10-02 21:36:14,201 - Epoch: [139][ 980/ 1236] Overall Loss 0.171787 Objective Loss 0.171787 LR 0.000500 Time 0.021874 +2023-10-02 21:36:14,409 - Epoch: [139][ 990/ 1236] Overall Loss 0.171712 Objective Loss 0.171712 LR 0.000500 Time 0.021863 +2023-10-02 21:36:14,621 - Epoch: [139][ 1000/ 1236] Overall Loss 0.172076 Objective Loss 0.172076 LR 0.000500 Time 0.021856 +2023-10-02 21:36:14,829 - Epoch: [139][ 1010/ 1236] Overall Loss 0.172121 Objective Loss 0.172121 LR 0.000500 Time 0.021845 +2023-10-02 21:36:15,040 - Epoch: [139][ 1020/ 1236] Overall Loss 0.172115 Objective Loss 0.172115 LR 0.000500 Time 0.021837 +2023-10-02 21:36:15,249 - Epoch: [139][ 1030/ 1236] Overall Loss 0.172073 Objective Loss 0.172073 LR 0.000500 Time 0.021828 +2023-10-02 21:36:15,460 - Epoch: [139][ 1040/ 1236] Overall Loss 0.171821 Objective Loss 0.171821 LR 0.000500 Time 0.021821 +2023-10-02 21:36:15,670 - Epoch: [139][ 1050/ 1236] Overall Loss 0.171638 Objective Loss 0.171638 LR 0.000500 Time 0.021812 +2023-10-02 21:36:15,882 - Epoch: [139][ 1060/ 1236] Overall Loss 0.171594 Objective Loss 0.171594 LR 0.000500 Time 0.021807 +2023-10-02 21:36:16,090 - Epoch: [139][ 1070/ 1236] Overall Loss 0.171560 Objective Loss 0.171560 LR 0.000500 Time 0.021797 +2023-10-02 21:36:16,300 - Epoch: [139][ 1080/ 1236] Overall Loss 0.171564 Objective Loss 0.171564 LR 0.000500 Time 0.021789 +2023-10-02 21:36:16,507 - Epoch: [139][ 1090/ 1236] Overall Loss 0.171654 Objective Loss 0.171654 LR 0.000500 Time 0.021779 +2023-10-02 21:36:16,717 - Epoch: [139][ 1100/ 1236] Overall Loss 0.171791 Objective Loss 0.171791 LR 0.000500 Time 0.021771 +2023-10-02 21:36:16,923 - Epoch: [139][ 1110/ 1236] Overall Loss 0.171911 Objective Loss 0.171911 LR 0.000500 Time 0.021761 +2023-10-02 21:36:17,133 - Epoch: [139][ 1120/ 1236] Overall Loss 0.172020 Objective Loss 0.172020 LR 0.000500 Time 0.021754 +2023-10-02 21:36:17,340 - Epoch: [139][ 1130/ 1236] Overall Loss 0.172060 Objective Loss 0.172060 LR 0.000500 Time 0.021744 +2023-10-02 21:36:17,550 - Epoch: [139][ 1140/ 1236] Overall Loss 0.172034 Objective Loss 0.172034 LR 0.000500 Time 0.021738 +2023-10-02 21:36:17,757 - Epoch: [139][ 1150/ 1236] Overall Loss 0.171999 Objective Loss 0.171999 LR 0.000500 Time 0.021728 +2023-10-02 21:36:17,967 - Epoch: [139][ 1160/ 1236] Overall Loss 0.172449 Objective Loss 0.172449 LR 0.000500 Time 0.021722 +2023-10-02 21:36:18,175 - Epoch: [139][ 1170/ 1236] Overall Loss 0.172645 Objective Loss 0.172645 LR 0.000500 Time 0.021713 +2023-10-02 21:36:18,384 - Epoch: [139][ 1180/ 1236] Overall Loss 0.172772 Objective Loss 0.172772 LR 0.000500 Time 0.021706 +2023-10-02 21:36:18,591 - Epoch: [139][ 1190/ 1236] Overall Loss 0.172775 Objective Loss 0.172775 LR 0.000500 Time 0.021698 +2023-10-02 21:36:18,801 - Epoch: [139][ 1200/ 1236] Overall Loss 0.172830 Objective Loss 0.172830 LR 0.000500 Time 0.021692 +2023-10-02 21:36:19,008 - Epoch: [139][ 1210/ 1236] Overall Loss 0.172895 Objective Loss 0.172895 LR 0.000500 Time 0.021683 +2023-10-02 21:36:19,218 - Epoch: [139][ 1220/ 1236] Overall Loss 0.172699 Objective Loss 0.172699 LR 0.000500 Time 0.021677 +2023-10-02 21:36:19,480 - Epoch: [139][ 1230/ 1236] Overall Loss 0.172502 Objective Loss 0.172502 LR 0.000500 Time 0.021713 +2023-10-02 21:36:19,602 - Epoch: [139][ 1236/ 1236] Overall Loss 0.172317 Objective Loss 0.172317 Top1 92.464358 Top5 99.185336 LR 0.000500 Time 0.021707 +2023-10-02 21:36:19,745 - --- validate (epoch=139)----------- +2023-10-02 21:36:19,746 - 29943 samples (256 per mini-batch) +2023-10-02 21:36:20,247 - Epoch: [139][ 10/ 117] Loss 0.261640 Top1 86.445312 Top5 98.671875 +2023-10-02 21:36:20,401 - Epoch: [139][ 20/ 117] Loss 0.284653 Top1 86.718750 Top5 98.730469 +2023-10-02 21:36:20,553 - Epoch: [139][ 30/ 117] Loss 0.299379 Top1 86.145833 Top5 98.567708 +2023-10-02 21:36:20,706 - Epoch: [139][ 40/ 117] Loss 0.302140 Top1 86.250000 Top5 98.554688 +2023-10-02 21:36:20,857 - Epoch: [139][ 50/ 117] Loss 0.303505 Top1 86.265625 Top5 98.554688 +2023-10-02 21:36:21,009 - Epoch: [139][ 60/ 117] Loss 0.302232 Top1 86.354167 Top5 98.626302 +2023-10-02 21:36:21,160 - Epoch: [139][ 70/ 117] Loss 0.304649 Top1 86.389509 Top5 98.638393 +2023-10-02 21:36:21,311 - Epoch: [139][ 80/ 117] Loss 0.308405 Top1 86.289062 Top5 98.647461 +2023-10-02 21:36:21,463 - Epoch: [139][ 90/ 117] Loss 0.305644 Top1 86.236979 Top5 98.624132 +2023-10-02 21:36:21,616 - Epoch: [139][ 100/ 117] Loss 0.306168 Top1 86.183594 Top5 98.593750 +2023-10-02 21:36:21,776 - Epoch: [139][ 110/ 117] Loss 0.307822 Top1 86.147017 Top5 98.561790 +2023-10-02 21:36:21,865 - Epoch: [139][ 117/ 117] Loss 0.305708 Top1 86.090238 Top5 98.547240 +2023-10-02 21:36:22,017 - ==> Top1: 86.090 Top5: 98.547 Loss: 0.306 + +2023-10-02 21:36:22,017 - ==> Confusion: +[[ 948 0 5 0 8 2 0 0 11 45 1 1 1 0 4 2 2 2 1 0 17] + [ 0 1043 1 1 6 23 2 22 3 1 1 1 3 0 0 3 6 0 9 0 6] + [ 3 0 980 11 0 0 15 7 0 1 1 0 6 2 0 3 1 1 14 3 8] + [ 1 1 12 978 1 2 0 2 3 0 5 0 6 4 32 2 1 3 15 3 18] + [ 27 4 1 0 966 3 0 0 1 11 1 0 1 2 9 7 13 0 0 2 2] + [ 1 23 3 1 3 988 0 33 3 3 2 7 4 11 4 0 6 2 5 6 11] + [ 0 2 31 0 0 1 1128 4 0 0 2 0 1 0 0 6 0 0 3 6 7] + [ 5 9 15 1 8 19 5 1068 3 3 2 1 2 5 3 2 1 1 47 9 9] + [ 17 0 0 1 1 2 0 1 984 31 9 1 2 12 14 2 4 1 4 2 1] + [ 103 0 2 1 7 0 1 0 27 937 2 2 0 18 9 3 0 1 0 1 5] + [ 6 2 12 6 1 0 2 0 12 0 972 0 0 12 4 2 1 1 13 0 7] + [ 1 0 0 0 0 17 0 2 0 0 0 953 24 11 0 4 2 12 0 8 1] + [ 0 0 4 1 1 2 1 0 0 0 3 40 967 2 3 15 1 8 3 8 9] + [ 0 0 0 0 2 9 0 0 19 8 2 7 0 1049 5 0 1 1 0 2 14] + [ 9 0 5 12 4 0 0 0 22 1 4 0 3 1 1017 0 1 2 10 0 10] + [ 0 0 2 0 4 0 1 1 1 0 0 6 8 0 0 1073 18 5 0 9 6] + [ 2 8 0 0 4 8 2 0 1 0 0 5 0 2 4 7 1103 0 1 4 10] + [ 0 0 1 1 1 0 1 0 0 0 1 0 27 3 2 6 0 990 0 0 5] + [ 2 1 3 11 0 0 1 9 4 0 3 1 2 0 14 1 1 0 999 1 15] + [ 0 1 6 1 1 2 8 7 0 0 0 12 6 2 1 2 11 0 0 1085 7] + [ 114 116 153 66 60 126 29 104 88 67 141 89 331 255 126 62 89 53 141 146 5549]] + +2023-10-02 21:36:22,019 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:36:22,019 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:36:22,025 - + +2023-10-02 21:36:22,025 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:36:23,063 - Epoch: [140][ 10/ 1236] Overall Loss 0.167670 Objective Loss 0.167670 LR 0.000250 Time 0.103750 +2023-10-02 21:36:23,271 - Epoch: [140][ 20/ 1236] Overall Loss 0.154277 Objective Loss 0.154277 LR 0.000250 Time 0.062244 +2023-10-02 21:36:23,477 - Epoch: [140][ 30/ 1236] Overall Loss 0.158243 Objective Loss 0.158243 LR 0.000250 Time 0.048325 +2023-10-02 21:36:23,685 - Epoch: [140][ 40/ 1236] Overall Loss 0.157839 Objective Loss 0.157839 LR 0.000250 Time 0.041420 +2023-10-02 21:36:23,891 - Epoch: [140][ 50/ 1236] Overall Loss 0.158367 Objective Loss 0.158367 LR 0.000250 Time 0.037235 +2023-10-02 21:36:24,099 - Epoch: [140][ 60/ 1236] Overall Loss 0.157473 Objective Loss 0.157473 LR 0.000250 Time 0.034482 +2023-10-02 21:36:24,305 - Epoch: [140][ 70/ 1236] Overall Loss 0.154963 Objective Loss 0.154963 LR 0.000250 Time 0.032480 +2023-10-02 21:36:24,514 - Epoch: [140][ 80/ 1236] Overall Loss 0.156872 Objective Loss 0.156872 LR 0.000250 Time 0.031034 +2023-10-02 21:36:24,720 - Epoch: [140][ 90/ 1236] Overall Loss 0.155022 Objective Loss 0.155022 LR 0.000250 Time 0.029865 +2023-10-02 21:36:24,928 - Epoch: [140][ 100/ 1236] Overall Loss 0.153878 Objective Loss 0.153878 LR 0.000250 Time 0.028962 +2023-10-02 21:36:25,133 - Epoch: [140][ 110/ 1236] Overall Loss 0.153274 Objective Loss 0.153274 LR 0.000250 Time 0.028190 +2023-10-02 21:36:25,341 - Epoch: [140][ 120/ 1236] Overall Loss 0.155722 Objective Loss 0.155722 LR 0.000250 Time 0.027571 +2023-10-02 21:36:25,546 - Epoch: [140][ 130/ 1236] Overall Loss 0.154248 Objective Loss 0.154248 LR 0.000250 Time 0.027021 +2023-10-02 21:36:25,754 - Epoch: [140][ 140/ 1236] Overall Loss 0.155515 Objective Loss 0.155515 LR 0.000250 Time 0.026580 +2023-10-02 21:36:25,959 - Epoch: [140][ 150/ 1236] Overall Loss 0.155096 Objective Loss 0.155096 LR 0.000250 Time 0.026173 +2023-10-02 21:36:26,169 - Epoch: [140][ 160/ 1236] Overall Loss 0.154259 Objective Loss 0.154259 LR 0.000250 Time 0.025843 +2023-10-02 21:36:26,374 - Epoch: [140][ 170/ 1236] Overall Loss 0.154371 Objective Loss 0.154371 LR 0.000250 Time 0.025527 +2023-10-02 21:36:26,583 - Epoch: [140][ 180/ 1236] Overall Loss 0.154060 Objective Loss 0.154060 LR 0.000250 Time 0.025271 +2023-10-02 21:36:26,788 - Epoch: [140][ 190/ 1236] Overall Loss 0.155347 Objective Loss 0.155347 LR 0.000250 Time 0.025017 +2023-10-02 21:36:26,997 - Epoch: [140][ 200/ 1236] Overall Loss 0.154938 Objective Loss 0.154938 LR 0.000250 Time 0.024810 +2023-10-02 21:36:27,202 - Epoch: [140][ 210/ 1236] Overall Loss 0.156261 Objective Loss 0.156261 LR 0.000250 Time 0.024604 +2023-10-02 21:36:27,411 - Epoch: [140][ 220/ 1236] Overall Loss 0.156540 Objective Loss 0.156540 LR 0.000250 Time 0.024435 +2023-10-02 21:36:27,616 - Epoch: [140][ 230/ 1236] Overall Loss 0.155879 Objective Loss 0.155879 LR 0.000250 Time 0.024264 +2023-10-02 21:36:27,824 - Epoch: [140][ 240/ 1236] Overall Loss 0.155976 Objective Loss 0.155976 LR 0.000250 Time 0.024118 +2023-10-02 21:36:28,031 - Epoch: [140][ 250/ 1236] Overall Loss 0.156590 Objective Loss 0.156590 LR 0.000250 Time 0.023972 +2023-10-02 21:36:28,240 - Epoch: [140][ 260/ 1236] Overall Loss 0.156725 Objective Loss 0.156725 LR 0.000250 Time 0.023855 +2023-10-02 21:36:28,445 - Epoch: [140][ 270/ 1236] Overall Loss 0.157145 Objective Loss 0.157145 LR 0.000250 Time 0.023730 +2023-10-02 21:36:28,654 - Epoch: [140][ 280/ 1236] Overall Loss 0.156920 Objective Loss 0.156920 LR 0.000250 Time 0.023628 +2023-10-02 21:36:28,860 - Epoch: [140][ 290/ 1236] Overall Loss 0.156627 Objective Loss 0.156627 LR 0.000250 Time 0.023521 +2023-10-02 21:36:29,068 - Epoch: [140][ 300/ 1236] Overall Loss 0.157839 Objective Loss 0.157839 LR 0.000250 Time 0.023431 +2023-10-02 21:36:29,275 - Epoch: [140][ 310/ 1236] Overall Loss 0.158446 Objective Loss 0.158446 LR 0.000250 Time 0.023337 +2023-10-02 21:36:29,483 - Epoch: [140][ 320/ 1236] Overall Loss 0.157879 Objective Loss 0.157879 LR 0.000250 Time 0.023257 +2023-10-02 21:36:29,690 - Epoch: [140][ 330/ 1236] Overall Loss 0.156968 Objective Loss 0.156968 LR 0.000250 Time 0.023178 +2023-10-02 21:36:29,898 - Epoch: [140][ 340/ 1236] Overall Loss 0.157172 Objective Loss 0.157172 LR 0.000250 Time 0.023106 +2023-10-02 21:36:30,104 - Epoch: [140][ 350/ 1236] Overall Loss 0.157338 Objective Loss 0.157338 LR 0.000250 Time 0.023032 +2023-10-02 21:36:30,312 - Epoch: [140][ 360/ 1236] Overall Loss 0.157537 Objective Loss 0.157537 LR 0.000250 Time 0.022969 +2023-10-02 21:36:30,519 - Epoch: [140][ 370/ 1236] Overall Loss 0.158377 Objective Loss 0.158377 LR 0.000250 Time 0.022906 +2023-10-02 21:36:30,728 - Epoch: [140][ 380/ 1236] Overall Loss 0.158269 Objective Loss 0.158269 LR 0.000250 Time 0.022853 +2023-10-02 21:36:30,933 - Epoch: [140][ 390/ 1236] Overall Loss 0.158386 Objective Loss 0.158386 LR 0.000250 Time 0.022792 +2023-10-02 21:36:31,141 - Epoch: [140][ 400/ 1236] Overall Loss 0.158924 Objective Loss 0.158924 LR 0.000250 Time 0.022742 +2023-10-02 21:36:31,348 - Epoch: [140][ 410/ 1236] Overall Loss 0.159246 Objective Loss 0.159246 LR 0.000250 Time 0.022687 +2023-10-02 21:36:31,556 - Epoch: [140][ 420/ 1236] Overall Loss 0.159200 Objective Loss 0.159200 LR 0.000250 Time 0.022642 +2023-10-02 21:36:31,763 - Epoch: [140][ 430/ 1236] Overall Loss 0.159296 Objective Loss 0.159296 LR 0.000250 Time 0.022593 +2023-10-02 21:36:31,971 - Epoch: [140][ 440/ 1236] Overall Loss 0.159336 Objective Loss 0.159336 LR 0.000250 Time 0.022552 +2023-10-02 21:36:32,178 - Epoch: [140][ 450/ 1236] Overall Loss 0.158885 Objective Loss 0.158885 LR 0.000250 Time 0.022507 +2023-10-02 21:36:32,387 - Epoch: [140][ 460/ 1236] Overall Loss 0.158304 Objective Loss 0.158304 LR 0.000250 Time 0.022472 +2023-10-02 21:36:32,592 - Epoch: [140][ 470/ 1236] Overall Loss 0.158927 Objective Loss 0.158927 LR 0.000250 Time 0.022430 +2023-10-02 21:36:32,800 - Epoch: [140][ 480/ 1236] Overall Loss 0.159415 Objective Loss 0.159415 LR 0.000250 Time 0.022395 +2023-10-02 21:36:33,007 - Epoch: [140][ 490/ 1236] Overall Loss 0.159270 Objective Loss 0.159270 LR 0.000250 Time 0.022357 +2023-10-02 21:36:33,217 - Epoch: [140][ 500/ 1236] Overall Loss 0.158997 Objective Loss 0.158997 LR 0.000250 Time 0.022328 +2023-10-02 21:36:33,422 - Epoch: [140][ 510/ 1236] Overall Loss 0.158575 Objective Loss 0.158575 LR 0.000250 Time 0.022293 +2023-10-02 21:36:33,630 - Epoch: [140][ 520/ 1236] Overall Loss 0.158360 Objective Loss 0.158360 LR 0.000250 Time 0.022264 +2023-10-02 21:36:33,837 - Epoch: [140][ 530/ 1236] Overall Loss 0.158152 Objective Loss 0.158152 LR 0.000250 Time 0.022231 +2023-10-02 21:36:34,046 - Epoch: [140][ 540/ 1236] Overall Loss 0.158248 Objective Loss 0.158248 LR 0.000250 Time 0.022207 +2023-10-02 21:36:34,252 - Epoch: [140][ 550/ 1236] Overall Loss 0.158319 Objective Loss 0.158319 LR 0.000250 Time 0.022176 +2023-10-02 21:36:34,461 - Epoch: [140][ 560/ 1236] Overall Loss 0.158115 Objective Loss 0.158115 LR 0.000250 Time 0.022154 +2023-10-02 21:36:34,667 - Epoch: [140][ 570/ 1236] Overall Loss 0.157519 Objective Loss 0.157519 LR 0.000250 Time 0.022125 +2023-10-02 21:36:34,875 - Epoch: [140][ 580/ 1236] Overall Loss 0.157405 Objective Loss 0.157405 LR 0.000250 Time 0.022102 +2023-10-02 21:36:35,081 - Epoch: [140][ 590/ 1236] Overall Loss 0.157356 Objective Loss 0.157356 LR 0.000250 Time 0.022074 +2023-10-02 21:36:35,289 - Epoch: [140][ 600/ 1236] Overall Loss 0.157047 Objective Loss 0.157047 LR 0.000250 Time 0.022053 +2023-10-02 21:36:35,496 - Epoch: [140][ 610/ 1236] Overall Loss 0.156805 Objective Loss 0.156805 LR 0.000250 Time 0.022028 +2023-10-02 21:36:35,706 - Epoch: [140][ 620/ 1236] Overall Loss 0.156600 Objective Loss 0.156600 LR 0.000250 Time 0.022009 +2023-10-02 21:36:35,911 - Epoch: [140][ 630/ 1236] Overall Loss 0.156666 Objective Loss 0.156666 LR 0.000250 Time 0.021986 +2023-10-02 21:36:36,121 - Epoch: [140][ 640/ 1236] Overall Loss 0.156754 Objective Loss 0.156754 LR 0.000250 Time 0.021970 +2023-10-02 21:36:36,326 - Epoch: [140][ 650/ 1236] Overall Loss 0.156364 Objective Loss 0.156364 LR 0.000250 Time 0.021948 +2023-10-02 21:36:36,536 - Epoch: [140][ 660/ 1236] Overall Loss 0.156218 Objective Loss 0.156218 LR 0.000250 Time 0.021932 +2023-10-02 21:36:36,741 - Epoch: [140][ 670/ 1236] Overall Loss 0.156386 Objective Loss 0.156386 LR 0.000250 Time 0.021911 +2023-10-02 21:36:36,949 - Epoch: [140][ 680/ 1236] Overall Loss 0.156459 Objective Loss 0.156459 LR 0.000250 Time 0.021894 +2023-10-02 21:36:37,156 - Epoch: [140][ 690/ 1236] Overall Loss 0.156417 Objective Loss 0.156417 LR 0.000250 Time 0.021874 +2023-10-02 21:36:37,364 - Epoch: [140][ 700/ 1236] Overall Loss 0.156773 Objective Loss 0.156773 LR 0.000250 Time 0.021859 +2023-10-02 21:36:37,571 - Epoch: [140][ 710/ 1236] Overall Loss 0.156785 Objective Loss 0.156785 LR 0.000250 Time 0.021840 +2023-10-02 21:36:37,781 - Epoch: [140][ 720/ 1236] Overall Loss 0.156881 Objective Loss 0.156881 LR 0.000250 Time 0.021827 +2023-10-02 21:36:37,986 - Epoch: [140][ 730/ 1236] Overall Loss 0.157402 Objective Loss 0.157402 LR 0.000250 Time 0.021809 +2023-10-02 21:36:38,194 - Epoch: [140][ 740/ 1236] Overall Loss 0.157523 Objective Loss 0.157523 LR 0.000250 Time 0.021795 +2023-10-02 21:36:38,401 - Epoch: [140][ 750/ 1236] Overall Loss 0.157516 Objective Loss 0.157516 LR 0.000250 Time 0.021779 +2023-10-02 21:36:38,609 - Epoch: [140][ 760/ 1236] Overall Loss 0.157366 Objective Loss 0.157366 LR 0.000250 Time 0.021766 +2023-10-02 21:36:38,816 - Epoch: [140][ 770/ 1236] Overall Loss 0.157213 Objective Loss 0.157213 LR 0.000250 Time 0.021749 +2023-10-02 21:36:39,025 - Epoch: [140][ 780/ 1236] Overall Loss 0.157574 Objective Loss 0.157574 LR 0.000250 Time 0.021738 +2023-10-02 21:36:39,231 - Epoch: [140][ 790/ 1236] Overall Loss 0.157650 Objective Loss 0.157650 LR 0.000250 Time 0.021722 +2023-10-02 21:36:39,441 - Epoch: [140][ 800/ 1236] Overall Loss 0.157780 Objective Loss 0.157780 LR 0.000250 Time 0.021712 +2023-10-02 21:36:39,646 - Epoch: [140][ 810/ 1236] Overall Loss 0.157679 Objective Loss 0.157679 LR 0.000250 Time 0.021697 +2023-10-02 21:36:39,855 - Epoch: [140][ 820/ 1236] Overall Loss 0.157535 Objective Loss 0.157535 LR 0.000250 Time 0.021686 +2023-10-02 21:36:40,061 - Epoch: [140][ 830/ 1236] Overall Loss 0.157528 Objective Loss 0.157528 LR 0.000250 Time 0.021672 +2023-10-02 21:36:40,269 - Epoch: [140][ 840/ 1236] Overall Loss 0.157827 Objective Loss 0.157827 LR 0.000250 Time 0.021662 +2023-10-02 21:36:40,476 - Epoch: [140][ 850/ 1236] Overall Loss 0.157683 Objective Loss 0.157683 LR 0.000250 Time 0.021648 +2023-10-02 21:36:40,684 - Epoch: [140][ 860/ 1236] Overall Loss 0.157676 Objective Loss 0.157676 LR 0.000250 Time 0.021638 +2023-10-02 21:36:40,891 - Epoch: [140][ 870/ 1236] Overall Loss 0.157751 Objective Loss 0.157751 LR 0.000250 Time 0.021625 +2023-10-02 21:36:41,100 - Epoch: [140][ 880/ 1236] Overall Loss 0.157596 Objective Loss 0.157596 LR 0.000250 Time 0.021616 +2023-10-02 21:36:41,306 - Epoch: [140][ 890/ 1236] Overall Loss 0.157473 Objective Loss 0.157473 LR 0.000250 Time 0.021604 +2023-10-02 21:36:41,514 - Epoch: [140][ 900/ 1236] Overall Loss 0.157355 Objective Loss 0.157355 LR 0.000250 Time 0.021595 +2023-10-02 21:36:41,721 - Epoch: [140][ 910/ 1236] Overall Loss 0.157365 Objective Loss 0.157365 LR 0.000250 Time 0.021583 +2023-10-02 21:36:41,931 - Epoch: [140][ 920/ 1236] Overall Loss 0.157279 Objective Loss 0.157279 LR 0.000250 Time 0.021576 +2023-10-02 21:36:42,136 - Epoch: [140][ 930/ 1236] Overall Loss 0.157297 Objective Loss 0.157297 LR 0.000250 Time 0.021565 +2023-10-02 21:36:42,345 - Epoch: [140][ 940/ 1236] Overall Loss 0.157249 Objective Loss 0.157249 LR 0.000250 Time 0.021556 +2023-10-02 21:36:42,551 - Epoch: [140][ 950/ 1236] Overall Loss 0.157172 Objective Loss 0.157172 LR 0.000250 Time 0.021545 +2023-10-02 21:36:42,760 - Epoch: [140][ 960/ 1236] Overall Loss 0.157103 Objective Loss 0.157103 LR 0.000250 Time 0.021538 +2023-10-02 21:36:42,966 - Epoch: [140][ 970/ 1236] Overall Loss 0.156932 Objective Loss 0.156932 LR 0.000250 Time 0.021527 +2023-10-02 21:36:43,174 - Epoch: [140][ 980/ 1236] Overall Loss 0.156673 Objective Loss 0.156673 LR 0.000250 Time 0.021519 +2023-10-02 21:36:43,381 - Epoch: [140][ 990/ 1236] Overall Loss 0.156570 Objective Loss 0.156570 LR 0.000250 Time 0.021509 +2023-10-02 21:36:43,591 - Epoch: [140][ 1000/ 1236] Overall Loss 0.156305 Objective Loss 0.156305 LR 0.000250 Time 0.021504 +2023-10-02 21:36:43,796 - Epoch: [140][ 1010/ 1236] Overall Loss 0.156521 Objective Loss 0.156521 LR 0.000250 Time 0.021494 +2023-10-02 21:36:44,006 - Epoch: [140][ 1020/ 1236] Overall Loss 0.156517 Objective Loss 0.156517 LR 0.000250 Time 0.021488 +2023-10-02 21:36:44,212 - Epoch: [140][ 1030/ 1236] Overall Loss 0.156552 Objective Loss 0.156552 LR 0.000250 Time 0.021479 +2023-10-02 21:36:44,420 - Epoch: [140][ 1040/ 1236] Overall Loss 0.156441 Objective Loss 0.156441 LR 0.000250 Time 0.021472 +2023-10-02 21:36:44,626 - Epoch: [140][ 1050/ 1236] Overall Loss 0.156452 Objective Loss 0.156452 LR 0.000250 Time 0.021463 +2023-10-02 21:36:44,835 - Epoch: [140][ 1060/ 1236] Overall Loss 0.156584 Objective Loss 0.156584 LR 0.000250 Time 0.021457 +2023-10-02 21:36:45,042 - Epoch: [140][ 1070/ 1236] Overall Loss 0.156671 Objective Loss 0.156671 LR 0.000250 Time 0.021448 +2023-10-02 21:36:45,250 - Epoch: [140][ 1080/ 1236] Overall Loss 0.156662 Objective Loss 0.156662 LR 0.000250 Time 0.021442 +2023-10-02 21:36:45,456 - Epoch: [140][ 1090/ 1236] Overall Loss 0.156655 Objective Loss 0.156655 LR 0.000250 Time 0.021434 +2023-10-02 21:36:45,666 - Epoch: [140][ 1100/ 1236] Overall Loss 0.156531 Objective Loss 0.156531 LR 0.000250 Time 0.021429 +2023-10-02 21:36:45,871 - Epoch: [140][ 1110/ 1236] Overall Loss 0.156345 Objective Loss 0.156345 LR 0.000250 Time 0.021421 +2023-10-02 21:36:46,081 - Epoch: [140][ 1120/ 1236] Overall Loss 0.156398 Objective Loss 0.156398 LR 0.000250 Time 0.021416 +2023-10-02 21:36:46,286 - Epoch: [140][ 1130/ 1236] Overall Loss 0.156372 Objective Loss 0.156372 LR 0.000250 Time 0.021408 +2023-10-02 21:36:46,495 - Epoch: [140][ 1140/ 1236] Overall Loss 0.156349 Objective Loss 0.156349 LR 0.000250 Time 0.021403 +2023-10-02 21:36:46,701 - Epoch: [140][ 1150/ 1236] Overall Loss 0.156442 Objective Loss 0.156442 LR 0.000250 Time 0.021395 +2023-10-02 21:36:46,910 - Epoch: [140][ 1160/ 1236] Overall Loss 0.156529 Objective Loss 0.156529 LR 0.000250 Time 0.021390 +2023-10-02 21:36:47,116 - Epoch: [140][ 1170/ 1236] Overall Loss 0.156478 Objective Loss 0.156478 LR 0.000250 Time 0.021383 +2023-10-02 21:36:47,325 - Epoch: [140][ 1180/ 1236] Overall Loss 0.156487 Objective Loss 0.156487 LR 0.000250 Time 0.021378 +2023-10-02 21:36:47,532 - Epoch: [140][ 1190/ 1236] Overall Loss 0.156505 Objective Loss 0.156505 LR 0.000250 Time 0.021371 +2023-10-02 21:36:47,740 - Epoch: [140][ 1200/ 1236] Overall Loss 0.156457 Objective Loss 0.156457 LR 0.000250 Time 0.021366 +2023-10-02 21:36:47,947 - Epoch: [140][ 1210/ 1236] Overall Loss 0.156455 Objective Loss 0.156455 LR 0.000250 Time 0.021359 +2023-10-02 21:36:48,155 - Epoch: [140][ 1220/ 1236] Overall Loss 0.156356 Objective Loss 0.156356 LR 0.000250 Time 0.021354 +2023-10-02 21:36:48,415 - Epoch: [140][ 1230/ 1236] Overall Loss 0.156433 Objective Loss 0.156433 LR 0.000250 Time 0.021391 +2023-10-02 21:36:48,537 - Epoch: [140][ 1236/ 1236] Overall Loss 0.156558 Objective Loss 0.156558 Top1 87.780041 Top5 98.574338 LR 0.000250 Time 0.021386 +2023-10-02 21:36:48,678 - --- validate (epoch=140)----------- +2023-10-02 21:36:48,678 - 29943 samples (256 per mini-batch) +2023-10-02 21:36:49,176 - Epoch: [140][ 10/ 117] Loss 0.321890 Top1 86.562500 Top5 98.320312 +2023-10-02 21:36:49,328 - Epoch: [140][ 20/ 117] Loss 0.299488 Top1 86.699219 Top5 98.554688 +2023-10-02 21:36:49,479 - Epoch: [140][ 30/ 117] Loss 0.297913 Top1 86.302083 Top5 98.541667 +2023-10-02 21:36:49,630 - Epoch: [140][ 40/ 117] Loss 0.289043 Top1 86.367188 Top5 98.613281 +2023-10-02 21:36:49,780 - Epoch: [140][ 50/ 117] Loss 0.291618 Top1 86.265625 Top5 98.640625 +2023-10-02 21:36:49,932 - Epoch: [140][ 60/ 117] Loss 0.293242 Top1 86.432292 Top5 98.665365 +2023-10-02 21:36:50,083 - Epoch: [140][ 70/ 117] Loss 0.294118 Top1 86.372768 Top5 98.671875 +2023-10-02 21:36:50,233 - Epoch: [140][ 80/ 117] Loss 0.291804 Top1 86.362305 Top5 98.632812 +2023-10-02 21:36:50,386 - Epoch: [140][ 90/ 117] Loss 0.292604 Top1 86.401910 Top5 98.589410 +2023-10-02 21:36:50,539 - Epoch: [140][ 100/ 117] Loss 0.291762 Top1 86.406250 Top5 98.566406 +2023-10-02 21:36:50,699 - Epoch: [140][ 110/ 117] Loss 0.291499 Top1 86.381392 Top5 98.583097 +2023-10-02 21:36:50,788 - Epoch: [140][ 117/ 117] Loss 0.293266 Top1 86.320676 Top5 98.580637 +2023-10-02 21:36:50,929 - ==> Top1: 86.321 Top5: 98.581 Loss: 0.293 + +2023-10-02 21:36:50,930 - ==> Confusion: +[[ 947 0 2 1 8 2 0 1 8 54 3 1 0 1 5 2 2 0 0 0 13] + [ 0 1063 1 0 4 20 1 18 1 0 0 1 3 0 1 2 3 0 10 0 3] + [ 3 0 993 4 3 0 9 11 0 1 4 0 7 3 0 4 0 1 5 3 5] + [ 0 4 16 982 1 0 1 5 3 1 7 0 4 3 25 1 1 3 13 1 18] + [ 24 3 1 1 968 7 0 0 2 13 0 0 2 0 12 3 9 0 0 1 4] + [ 1 37 0 1 2 992 1 21 2 5 2 10 3 9 3 1 4 1 4 3 14] + [ 0 4 30 3 0 2 1131 2 0 0 3 2 0 0 0 3 0 0 0 6 5] + [ 4 10 13 0 2 24 4 1076 1 4 3 7 2 4 3 1 1 1 44 8 6] + [ 17 0 0 0 1 1 0 2 979 39 9 3 1 11 14 1 4 1 2 2 2] + [ 82 1 2 0 6 1 0 0 26 960 1 4 0 16 8 3 0 0 0 1 8] + [ 5 1 12 5 2 0 5 1 8 0 982 1 1 7 3 0 1 2 8 1 8] + [ 0 3 1 0 0 17 0 3 0 0 0 954 24 8 0 3 2 14 0 5 1] + [ 0 1 1 4 1 2 2 1 1 0 3 31 973 1 6 12 1 8 2 4 14] + [ 1 0 0 0 1 6 0 1 14 12 3 9 0 1050 4 0 3 1 0 1 13] + [ 14 0 4 17 8 0 0 0 24 2 4 0 2 1 1009 0 1 2 9 0 4] + [ 0 0 2 1 5 0 1 0 0 0 1 6 8 0 0 1071 17 10 3 5 4] + [ 2 12 0 0 4 7 1 0 0 0 0 6 0 2 4 8 1101 0 1 3 10] + [ 0 0 0 4 1 0 1 0 0 0 1 3 19 2 1 6 0 994 0 0 6] + [ 1 4 5 15 0 1 1 13 3 0 4 2 1 0 14 0 1 0 992 1 10] + [ 0 1 3 1 2 3 8 9 0 0 1 13 6 2 0 1 9 0 1 1086 6] + [ 135 137 131 92 73 136 25 77 99 77 163 88 300 256 114 46 92 61 118 141 5544]] + +2023-10-02 21:36:50,931 - ==> Best [Top1: 86.625 Top5: 98.564 Sparsity:0.00 Params: 169472 on epoch: 131] +2023-10-02 21:36:50,931 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:36:50,937 - + +2023-10-02 21:36:50,937 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:36:51,972 - Epoch: [141][ 10/ 1236] Overall Loss 0.150974 Objective Loss 0.150974 LR 0.000250 Time 0.103396 +2023-10-02 21:36:52,180 - Epoch: [141][ 20/ 1236] Overall Loss 0.148878 Objective Loss 0.148878 LR 0.000250 Time 0.062077 +2023-10-02 21:36:52,387 - Epoch: [141][ 30/ 1236] Overall Loss 0.150836 Objective Loss 0.150836 LR 0.000250 Time 0.048237 +2023-10-02 21:36:52,596 - Epoch: [141][ 40/ 1236] Overall Loss 0.146180 Objective Loss 0.146180 LR 0.000250 Time 0.041398 +2023-10-02 21:36:52,802 - Epoch: [141][ 50/ 1236] Overall Loss 0.144841 Objective Loss 0.144841 LR 0.000250 Time 0.037228 +2023-10-02 21:36:53,011 - Epoch: [141][ 60/ 1236] Overall Loss 0.148903 Objective Loss 0.148903 LR 0.000250 Time 0.034505 +2023-10-02 21:36:53,217 - Epoch: [141][ 70/ 1236] Overall Loss 0.148392 Objective Loss 0.148392 LR 0.000250 Time 0.032512 +2023-10-02 21:36:53,426 - Epoch: [141][ 80/ 1236] Overall Loss 0.149153 Objective Loss 0.149153 LR 0.000250 Time 0.031058 +2023-10-02 21:36:53,632 - Epoch: [141][ 90/ 1236] Overall Loss 0.149500 Objective Loss 0.149500 LR 0.000250 Time 0.029891 +2023-10-02 21:36:53,841 - Epoch: [141][ 100/ 1236] Overall Loss 0.149324 Objective Loss 0.149324 LR 0.000250 Time 0.028990 +2023-10-02 21:36:54,046 - Epoch: [141][ 110/ 1236] Overall Loss 0.150791 Objective Loss 0.150791 LR 0.000250 Time 0.028218 +2023-10-02 21:36:54,255 - Epoch: [141][ 120/ 1236] Overall Loss 0.151186 Objective Loss 0.151186 LR 0.000250 Time 0.027605 +2023-10-02 21:36:54,461 - Epoch: [141][ 130/ 1236] Overall Loss 0.153288 Objective Loss 0.153288 LR 0.000250 Time 0.027062 +2023-10-02 21:36:54,668 - Epoch: [141][ 140/ 1236] Overall Loss 0.152296 Objective Loss 0.152296 LR 0.000250 Time 0.026608 +2023-10-02 21:36:54,875 - Epoch: [141][ 150/ 1236] Overall Loss 0.152431 Objective Loss 0.152431 LR 0.000250 Time 0.026212 +2023-10-02 21:36:55,082 - Epoch: [141][ 160/ 1236] Overall Loss 0.151500 Objective Loss 0.151500 LR 0.000250 Time 0.025858 +2023-10-02 21:36:55,290 - Epoch: [141][ 170/ 1236] Overall Loss 0.150265 Objective Loss 0.150265 LR 0.000250 Time 0.025560 +2023-10-02 21:36:55,496 - Epoch: [141][ 180/ 1236] Overall Loss 0.150106 Objective Loss 0.150106 LR 0.000250 Time 0.025280 +2023-10-02 21:36:55,704 - Epoch: [141][ 190/ 1236] Overall Loss 0.150258 Objective Loss 0.150258 LR 0.000250 Time 0.025046 +2023-10-02 21:36:55,910 - Epoch: [141][ 200/ 1236] Overall Loss 0.150221 Objective Loss 0.150221 LR 0.000250 Time 0.024818 +2023-10-02 21:36:56,116 - Epoch: [141][ 210/ 1236] Overall Loss 0.149315 Objective Loss 0.149315 LR 0.000250 Time 0.024617 +2023-10-02 21:36:56,322 - Epoch: [141][ 220/ 1236] Overall Loss 0.151034 Objective Loss 0.151034 LR 0.000250 Time 0.024430 +2023-10-02 21:36:56,531 - Epoch: [141][ 230/ 1236] Overall Loss 0.151649 Objective Loss 0.151649 LR 0.000250 Time 0.024274 +2023-10-02 21:36:56,737 - Epoch: [141][ 240/ 1236] Overall Loss 0.151742 Objective Loss 0.151742 LR 0.000250 Time 0.024118 +2023-10-02 21:36:56,944 - Epoch: [141][ 250/ 1236] Overall Loss 0.151559 Objective Loss 0.151559 LR 0.000250 Time 0.023983 +2023-10-02 21:36:57,151 - Epoch: [141][ 260/ 1236] Overall Loss 0.152286 Objective Loss 0.152286 LR 0.000250 Time 0.023855 +2023-10-02 21:36:57,359 - Epoch: [141][ 270/ 1236] Overall Loss 0.152322 Objective Loss 0.152322 LR 0.000250 Time 0.023738 +2023-10-02 21:36:57,565 - Epoch: [141][ 280/ 1236] Overall Loss 0.152521 Objective Loss 0.152521 LR 0.000250 Time 0.023623 +2023-10-02 21:36:57,772 - Epoch: [141][ 290/ 1236] Overall Loss 0.152440 Objective Loss 0.152440 LR 0.000250 Time 0.023522 +2023-10-02 21:36:57,979 - Epoch: [141][ 300/ 1236] Overall Loss 0.152835 Objective Loss 0.152835 LR 0.000250 Time 0.023422 +2023-10-02 21:36:58,187 - Epoch: [141][ 310/ 1236] Overall Loss 0.153287 Objective Loss 0.153287 LR 0.000250 Time 0.023336 +2023-10-02 21:36:58,394 - Epoch: [141][ 320/ 1236] Overall Loss 0.152749 Objective Loss 0.152749 LR 0.000250 Time 0.023247 +2023-10-02 21:36:58,603 - Epoch: [141][ 330/ 1236] Overall Loss 0.152344 Objective Loss 0.152344 LR 0.000250 Time 0.023175 +2023-10-02 21:36:58,808 - Epoch: [141][ 340/ 1236] Overall Loss 0.152440 Objective Loss 0.152440 LR 0.000250 Time 0.023096 +2023-10-02 21:36:59,016 - Epoch: [141][ 350/ 1236] Overall Loss 0.152491 Objective Loss 0.152491 LR 0.000250 Time 0.023030 +2023-10-02 21:36:59,223 - Epoch: [141][ 360/ 1236] Overall Loss 0.152738 Objective Loss 0.152738 LR 0.000250 Time 0.022961 +2023-10-02 21:36:59,432 - Epoch: [141][ 370/ 1236] Overall Loss 0.152804 Objective Loss 0.152804 LR 0.000250 Time 0.022905 +2023-10-02 21:36:59,638 - Epoch: [141][ 380/ 1236] Overall Loss 0.152969 Objective Loss 0.152969 LR 0.000250 Time 0.022843 +2023-10-02 21:36:59,847 - Epoch: [141][ 390/ 1236] Overall Loss 0.152775 Objective Loss 0.152775 LR 0.000250 Time 0.022793 +2023-10-02 21:37:00,053 - Epoch: [141][ 400/ 1236] Overall Loss 0.152946 Objective Loss 0.152946 LR 0.000250 Time 0.022737 +2023-10-02 21:37:00,262 - Epoch: [141][ 410/ 1236] Overall Loss 0.152949 Objective Loss 0.152949 LR 0.000250 Time 0.022692 +2023-10-02 21:37:00,468 - Epoch: [141][ 420/ 1236] Overall Loss 0.152739 Objective Loss 0.152739 LR 0.000250 Time 0.022641 +2023-10-02 21:37:00,676 - Epoch: [141][ 430/ 1236] Overall Loss 0.152505 Objective Loss 0.152505 LR 0.000250 Time 0.022597 +2023-10-02 21:37:00,883 - Epoch: [141][ 440/ 1236] Overall Loss 0.152233 Objective Loss 0.152233 LR 0.000250 Time 0.022551 +2023-10-02 21:37:01,092 - Epoch: [141][ 450/ 1236] Overall Loss 0.151629 Objective Loss 0.151629 LR 0.000250 Time 0.022514 +2023-10-02 21:37:01,298 - Epoch: [141][ 460/ 1236] Overall Loss 0.151477 Objective Loss 0.151477 LR 0.000250 Time 0.022471 +2023-10-02 21:37:01,506 - Epoch: [141][ 470/ 1236] Overall Loss 0.151178 Objective Loss 0.151178 LR 0.000250 Time 0.022435 +2023-10-02 21:37:01,713 - Epoch: [141][ 480/ 1236] Overall Loss 0.151187 Objective Loss 0.151187 LR 0.000250 Time 0.022395 +2023-10-02 21:37:01,921 - Epoch: [141][ 490/ 1236] Overall Loss 0.151057 Objective Loss 0.151057 LR 0.000250 Time 0.022363 +2023-10-02 21:37:02,128 - Epoch: [141][ 500/ 1236] Overall Loss 0.151346 Objective Loss 0.151346 LR 0.000250 Time 0.022326 +2023-10-02 21:37:02,337 - Epoch: [141][ 510/ 1236] Overall Loss 0.151288 Objective Loss 0.151288 LR 0.000250 Time 0.022298 +2023-10-02 21:37:02,543 - Epoch: [141][ 520/ 1236] Overall Loss 0.151529 Objective Loss 0.151529 LR 0.000250 Time 0.022264 +2023-10-02 21:37:02,751 - Epoch: [141][ 530/ 1236] Overall Loss 0.151621 Objective Loss 0.151621 LR 0.000250 Time 0.022236 +2023-10-02 21:37:02,958 - Epoch: [141][ 540/ 1236] Overall Loss 0.151730 Objective Loss 0.151730 LR 0.000250 Time 0.022205 +2023-10-02 21:37:03,167 - Epoch: [141][ 550/ 1236] Overall Loss 0.151968 Objective Loss 0.151968 LR 0.000250 Time 0.022181 +2023-10-02 21:37:03,373 - Epoch: [141][ 560/ 1236] Overall Loss 0.151930 Objective Loss 0.151930 LR 0.000250 Time 0.022152 +2023-10-02 21:37:03,582 - Epoch: [141][ 570/ 1236] Overall Loss 0.151750 Objective Loss 0.151750 LR 0.000250 Time 0.022130 +2023-10-02 21:37:03,788 - Epoch: [141][ 580/ 1236] Overall Loss 0.151518 Objective Loss 0.151518 LR 0.000250 Time 0.022102 +2023-10-02 21:37:03,997 - Epoch: [141][ 590/ 1236] Overall Loss 0.151737 Objective Loss 0.151737 LR 0.000250 Time 0.022082 +2023-10-02 21:37:04,203 - Epoch: [141][ 600/ 1236] Overall Loss 0.151767 Objective Loss 0.151767 LR 0.000250 Time 0.022057 +2023-10-02 21:37:04,411 - Epoch: [141][ 610/ 1236] Overall Loss 0.151776 Objective Loss 0.151776 LR 0.000250 Time 0.022035 +2023-10-02 21:37:04,618 - Epoch: [141][ 620/ 1236] Overall Loss 0.152004 Objective Loss 0.152004 LR 0.000250 Time 0.022011 +2023-10-02 21:37:04,827 - Epoch: [141][ 630/ 1236] Overall Loss 0.152145 Objective Loss 0.152145 LR 0.000250 Time 0.021994 +2023-10-02 21:37:05,033 - Epoch: [141][ 640/ 1236] Overall Loss 0.152185 Objective Loss 0.152185 LR 0.000250 Time 0.021971 +2023-10-02 21:37:05,241 - Epoch: [141][ 650/ 1236] Overall Loss 0.152470 Objective Loss 0.152470 LR 0.000250 Time 0.021953 +2023-10-02 21:37:05,448 - Epoch: [141][ 660/ 1236] Overall Loss 0.152482 Objective Loss 0.152482 LR 0.000250 Time 0.021931 +2023-10-02 21:37:05,657 - Epoch: [141][ 670/ 1236] Overall Loss 0.152154 Objective Loss 0.152154 LR 0.000250 Time 0.021916 +2023-10-02 21:37:05,863 - Epoch: [141][ 680/ 1236] Overall Loss 0.152378 Objective Loss 0.152378 LR 0.000250 Time 0.021896 +2023-10-02 21:37:06,072 - Epoch: [141][ 690/ 1236] Overall Loss 0.152491 Objective Loss 0.152491 LR 0.000250 Time 0.021881 +2023-10-02 21:37:06,278 - Epoch: [141][ 700/ 1236] Overall Loss 0.152435 Objective Loss 0.152435 LR 0.000250 Time 0.021862 +2023-10-02 21:37:06,487 - Epoch: [141][ 710/ 1236] Overall Loss 0.152494 Objective Loss 0.152494 LR 0.000250 Time 0.021848 +2023-10-02 21:37:06,693 - Epoch: [141][ 720/ 1236] Overall Loss 0.152419 Objective Loss 0.152419 LR 0.000250 Time 0.021830 +2023-10-02 21:37:06,901 - Epoch: [141][ 730/ 1236] Overall Loss 0.152268 Objective Loss 0.152268 LR 0.000250 Time 0.021816 +2023-10-02 21:37:07,108 - Epoch: [141][ 740/ 1236] Overall Loss 0.152276 Objective Loss 0.152276 LR 0.000250 Time 0.021799 +2023-10-02 21:37:07,317 - Epoch: [141][ 750/ 1236] Overall Loss 0.152314 Objective Loss 0.152314 LR 0.000250 Time 0.021787 +2023-10-02 21:37:07,523 - Epoch: [141][ 760/ 1236] Overall Loss 0.152318 Objective Loss 0.152318 LR 0.000250 Time 0.021770 +2023-10-02 21:37:07,732 - Epoch: [141][ 770/ 1236] Overall Loss 0.152387 Objective Loss 0.152387 LR 0.000250 Time 0.021759 +2023-10-02 21:37:07,938 - Epoch: [141][ 780/ 1236] Overall Loss 0.152287 Objective Loss 0.152287 LR 0.000250 Time 0.021743 +2023-10-02 21:37:08,146 - Epoch: [141][ 790/ 1236] Overall Loss 0.152003 Objective Loss 0.152003 LR 0.000250 Time 0.021731 +2023-10-02 21:37:08,353 - Epoch: [141][ 800/ 1236] Overall Loss 0.151835 Objective Loss 0.151835 LR 0.000250 Time 0.021716 +2023-10-02 21:37:08,562 - Epoch: [141][ 810/ 1236] Overall Loss 0.151734 Objective Loss 0.151734 LR 0.000250 Time 0.021706 +2023-10-02 21:37:08,768 - Epoch: [141][ 820/ 1236] Overall Loss 0.151779 Objective Loss 0.151779 LR 0.000250 Time 0.021692 +2023-10-02 21:37:08,976 - Epoch: [141][ 830/ 1236] Overall Loss 0.151796 Objective Loss 0.151796 LR 0.000250 Time 0.021681 +2023-10-02 21:37:09,183 - Epoch: [141][ 840/ 1236] Overall Loss 0.151668 Objective Loss 0.151668 LR 0.000250 Time 0.021667 +2023-10-02 21:37:09,391 - Epoch: [141][ 850/ 1236] Overall Loss 0.151818 Objective Loss 0.151818 LR 0.000250 Time 0.021657 +2023-10-02 21:37:09,598 - Epoch: [141][ 860/ 1236] Overall Loss 0.151711 Objective Loss 0.151711 LR 0.000250 Time 0.021644 +2023-10-02 21:37:09,806 - Epoch: [141][ 870/ 1236] Overall Loss 0.151711 Objective Loss 0.151711 LR 0.000250 Time 0.021634 +2023-10-02 21:37:10,013 - Epoch: [141][ 880/ 1236] Overall Loss 0.151560 Objective Loss 0.151560 LR 0.000250 Time 0.021622 +2023-10-02 21:37:10,221 - Epoch: [141][ 890/ 1236] Overall Loss 0.151655 Objective Loss 0.151655 LR 0.000250 Time 0.021612 +2023-10-02 21:37:10,428 - Epoch: [141][ 900/ 1236] Overall Loss 0.151519 Objective Loss 0.151519 LR 0.000250 Time 0.021600 +2023-10-02 21:37:10,636 - Epoch: [141][ 910/ 1236] Overall Loss 0.151576 Objective Loss 0.151576 LR 0.000250 Time 0.021591 +2023-10-02 21:37:10,843 - Epoch: [141][ 920/ 1236] Overall Loss 0.151629 Objective Loss 0.151629 LR 0.000250 Time 0.021580 +2023-10-02 21:37:11,051 - Epoch: [141][ 930/ 1236] Overall Loss 0.151806 Objective Loss 0.151806 LR 0.000250 Time 0.021571 +2023-10-02 21:37:11,258 - Epoch: [141][ 940/ 1236] Overall Loss 0.151808 Objective Loss 0.151808 LR 0.000250 Time 0.021560 +2023-10-02 21:37:11,466 - Epoch: [141][ 950/ 1236] Overall Loss 0.151801 Objective Loss 0.151801 LR 0.000250 Time 0.021552 +2023-10-02 21:37:11,673 - Epoch: [141][ 960/ 1236] Overall Loss 0.151566 Objective Loss 0.151566 LR 0.000250 Time 0.021541 +2023-10-02 21:37:11,881 - Epoch: [141][ 970/ 1236] Overall Loss 0.151615 Objective Loss 0.151615 LR 0.000250 Time 0.021533 +2023-10-02 21:37:12,088 - Epoch: [141][ 980/ 1236] Overall Loss 0.151731 Objective Loss 0.151731 LR 0.000250 Time 0.021523 +2023-10-02 21:37:12,296 - Epoch: [141][ 990/ 1236] Overall Loss 0.151825 Objective Loss 0.151825 LR 0.000250 Time 0.021516 +2023-10-02 21:37:12,503 - Epoch: [141][ 1000/ 1236] Overall Loss 0.151782 Objective Loss 0.151782 LR 0.000250 Time 0.021506 +2023-10-02 21:37:12,711 - Epoch: [141][ 1010/ 1236] Overall Loss 0.151757 Objective Loss 0.151757 LR 0.000250 Time 0.021498 +2023-10-02 21:37:12,918 - Epoch: [141][ 1020/ 1236] Overall Loss 0.151739 Objective Loss 0.151739 LR 0.000250 Time 0.021489 +2023-10-02 21:37:13,126 - Epoch: [141][ 1030/ 1236] Overall Loss 0.151848 Objective Loss 0.151848 LR 0.000250 Time 0.021482 +2023-10-02 21:37:13,333 - Epoch: [141][ 1040/ 1236] Overall Loss 0.151617 Objective Loss 0.151617 LR 0.000250 Time 0.021473 +2023-10-02 21:37:13,541 - Epoch: [141][ 1050/ 1236] Overall Loss 0.151690 Objective Loss 0.151690 LR 0.000250 Time 0.021467 +2023-10-02 21:37:13,748 - Epoch: [141][ 1060/ 1236] Overall Loss 0.151632 Objective Loss 0.151632 LR 0.000250 Time 0.021458 +2023-10-02 21:37:13,956 - Epoch: [141][ 1070/ 1236] Overall Loss 0.151495 Objective Loss 0.151495 LR 0.000250 Time 0.021452 +2023-10-02 21:37:14,163 - Epoch: [141][ 1080/ 1236] Overall Loss 0.151535 Objective Loss 0.151535 LR 0.000250 Time 0.021443 +2023-10-02 21:37:14,371 - Epoch: [141][ 1090/ 1236] Overall Loss 0.151511 Objective Loss 0.151511 LR 0.000250 Time 0.021437 +2023-10-02 21:37:14,578 - Epoch: [141][ 1100/ 1236] Overall Loss 0.151469 Objective Loss 0.151469 LR 0.000250 Time 0.021429 +2023-10-02 21:37:14,786 - Epoch: [141][ 1110/ 1236] Overall Loss 0.151430 Objective Loss 0.151430 LR 0.000250 Time 0.021423 +2023-10-02 21:37:14,993 - Epoch: [141][ 1120/ 1236] Overall Loss 0.151343 Objective Loss 0.151343 LR 0.000250 Time 0.021415 +2023-10-02 21:37:15,201 - Epoch: [141][ 1130/ 1236] Overall Loss 0.151353 Objective Loss 0.151353 LR 0.000250 Time 0.021409 +2023-10-02 21:37:15,408 - Epoch: [141][ 1140/ 1236] Overall Loss 0.151317 Objective Loss 0.151317 LR 0.000250 Time 0.021402 +2023-10-02 21:37:15,616 - Epoch: [141][ 1150/ 1236] Overall Loss 0.151421 Objective Loss 0.151421 LR 0.000250 Time 0.021396 +2023-10-02 21:37:15,823 - Epoch: [141][ 1160/ 1236] Overall Loss 0.151676 Objective Loss 0.151676 LR 0.000250 Time 0.021389 +2023-10-02 21:37:16,032 - Epoch: [141][ 1170/ 1236] Overall Loss 0.151658 Objective Loss 0.151658 LR 0.000250 Time 0.021385 +2023-10-02 21:37:16,238 - Epoch: [141][ 1180/ 1236] Overall Loss 0.151739 Objective Loss 0.151739 LR 0.000250 Time 0.021378 +2023-10-02 21:37:16,446 - Epoch: [141][ 1190/ 1236] Overall Loss 0.151780 Objective Loss 0.151780 LR 0.000250 Time 0.021373 +2023-10-02 21:37:16,653 - Epoch: [141][ 1200/ 1236] Overall Loss 0.151789 Objective Loss 0.151789 LR 0.000250 Time 0.021366 +2023-10-02 21:37:16,861 - Epoch: [141][ 1210/ 1236] Overall Loss 0.151750 Objective Loss 0.151750 LR 0.000250 Time 0.021361 +2023-10-02 21:37:17,068 - Epoch: [141][ 1220/ 1236] Overall Loss 0.151679 Objective Loss 0.151679 LR 0.000250 Time 0.021354 +2023-10-02 21:37:17,329 - Epoch: [141][ 1230/ 1236] Overall Loss 0.151747 Objective Loss 0.151747 LR 0.000250 Time 0.021392 +2023-10-02 21:37:17,450 - Epoch: [141][ 1236/ 1236] Overall Loss 0.151753 Objective Loss 0.151753 Top1 89.613035 Top5 98.574338 LR 0.000250 Time 0.021386 +2023-10-02 21:37:17,589 - --- validate (epoch=141)----------- +2023-10-02 21:37:17,589 - 29943 samples (256 per mini-batch) +2023-10-02 21:37:18,063 - Epoch: [141][ 10/ 117] Loss 0.278650 Top1 86.835938 Top5 98.632812 +2023-10-02 21:37:18,209 - Epoch: [141][ 20/ 117] Loss 0.290928 Top1 86.816406 Top5 98.613281 +2023-10-02 21:37:18,355 - Epoch: [141][ 30/ 117] Loss 0.291883 Top1 86.406250 Top5 98.684896 +2023-10-02 21:37:18,501 - Epoch: [141][ 40/ 117] Loss 0.292543 Top1 86.582031 Top5 98.691406 +2023-10-02 21:37:18,648 - Epoch: [141][ 50/ 117] Loss 0.296541 Top1 86.679688 Top5 98.695312 +2023-10-02 21:37:18,795 - Epoch: [141][ 60/ 117] Loss 0.295858 Top1 86.731771 Top5 98.710938 +2023-10-02 21:37:18,942 - Epoch: [141][ 70/ 117] Loss 0.296165 Top1 86.774554 Top5 98.666295 +2023-10-02 21:37:19,089 - Epoch: [141][ 80/ 117] Loss 0.298046 Top1 86.738281 Top5 98.666992 +2023-10-02 21:37:19,235 - Epoch: [141][ 90/ 117] Loss 0.297885 Top1 86.783854 Top5 98.632812 +2023-10-02 21:37:19,381 - Epoch: [141][ 100/ 117] Loss 0.293453 Top1 86.808594 Top5 98.652344 +2023-10-02 21:37:19,535 - Epoch: [141][ 110/ 117] Loss 0.295530 Top1 86.690341 Top5 98.615057 +2023-10-02 21:37:19,624 - Epoch: [141][ 117/ 117] Loss 0.299151 Top1 86.644625 Top5 98.600675 +2023-10-02 21:37:19,765 - ==> Top1: 86.645 Top5: 98.601 Loss: 0.299 + +2023-10-02 21:37:19,766 - ==> Confusion: +[[ 958 1 1 0 7 2 0 2 5 41 2 1 0 2 5 1 2 1 1 0 18] + [ 1 1055 0 0 4 19 1 23 1 1 0 3 1 0 0 3 1 0 10 2 6] + [ 4 0 984 5 1 0 10 7 0 1 1 0 6 2 1 4 2 1 14 5 8] + [ 0 4 12 972 0 0 3 3 4 0 4 0 7 2 27 4 1 5 17 1 23] + [ 27 3 1 1 971 9 0 0 3 7 0 0 1 1 7 5 8 0 0 1 5] + [ 4 33 1 1 2 991 1 28 3 4 1 9 3 6 4 0 2 1 6 1 15] + [ 0 4 31 0 0 2 1132 5 0 0 1 0 0 0 0 5 0 0 2 5 4] + [ 2 6 9 2 3 18 7 1093 0 3 2 5 2 6 2 0 0 1 39 9 9] + [ 19 2 0 1 3 2 0 2 970 38 10 3 0 12 14 1 3 1 3 2 3] + [ 114 1 2 1 6 2 0 1 27 923 1 1 0 19 6 2 0 1 0 2 10] + [ 4 3 11 5 0 2 6 2 9 1 957 2 1 21 5 0 1 2 6 3 12] + [ 0 1 1 0 1 15 0 4 0 0 0 963 20 6 0 3 1 15 0 3 2] + [ 0 0 4 1 0 2 0 1 0 0 0 31 985 2 4 9 0 10 4 5 10] + [ 0 0 0 0 4 4 0 0 9 9 2 7 0 1062 4 1 2 1 0 0 14] + [ 14 0 5 19 4 0 0 0 19 2 2 0 3 2 1005 0 1 1 13 0 11] + [ 0 0 2 2 5 0 0 0 0 0 1 6 7 0 0 1074 13 10 2 6 6] + [ 1 15 1 0 3 6 0 0 0 0 0 4 0 3 3 9 1099 0 1 4 12] + [ 0 0 1 1 0 1 2 0 0 1 0 4 18 1 3 4 0 995 0 2 5] + [ 2 4 3 11 1 1 0 15 4 1 0 0 2 0 7 0 0 0 1005 0 12] + [ 0 2 1 2 1 3 9 4 0 0 0 13 5 0 0 2 8 0 2 1094 6] + [ 128 126 122 80 65 121 34 103 69 54 115 91 315 248 114 55 83 47 126 153 5656]] + +2023-10-02 21:37:19,767 - ==> Best [Top1: 86.645 Top5: 98.601 Sparsity:0.00 Params: 169472 on epoch: 141] +2023-10-02 21:37:19,768 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:37:19,781 - + +2023-10-02 21:37:19,781 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:37:20,797 - Epoch: [142][ 10/ 1236] Overall Loss 0.153523 Objective Loss 0.153523 LR 0.000250 Time 0.101571 +2023-10-02 21:37:21,005 - Epoch: [142][ 20/ 1236] Overall Loss 0.147828 Objective Loss 0.147828 LR 0.000250 Time 0.061137 +2023-10-02 21:37:21,211 - Epoch: [142][ 30/ 1236] Overall Loss 0.147934 Objective Loss 0.147934 LR 0.000250 Time 0.047606 +2023-10-02 21:37:21,418 - Epoch: [142][ 40/ 1236] Overall Loss 0.147756 Objective Loss 0.147756 LR 0.000250 Time 0.040884 +2023-10-02 21:37:21,623 - Epoch: [142][ 50/ 1236] Overall Loss 0.147723 Objective Loss 0.147723 LR 0.000250 Time 0.036801 +2023-10-02 21:37:21,831 - Epoch: [142][ 60/ 1236] Overall Loss 0.150225 Objective Loss 0.150225 LR 0.000250 Time 0.034118 +2023-10-02 21:37:22,036 - Epoch: [142][ 70/ 1236] Overall Loss 0.147949 Objective Loss 0.147949 LR 0.000250 Time 0.032168 +2023-10-02 21:37:22,243 - Epoch: [142][ 80/ 1236] Overall Loss 0.146887 Objective Loss 0.146887 LR 0.000250 Time 0.030732 +2023-10-02 21:37:22,448 - Epoch: [142][ 90/ 1236] Overall Loss 0.144842 Objective Loss 0.144842 LR 0.000250 Time 0.029597 +2023-10-02 21:37:22,656 - Epoch: [142][ 100/ 1236] Overall Loss 0.144569 Objective Loss 0.144569 LR 0.000250 Time 0.028708 +2023-10-02 21:37:22,861 - Epoch: [142][ 110/ 1236] Overall Loss 0.145315 Objective Loss 0.145315 LR 0.000250 Time 0.027963 +2023-10-02 21:37:23,068 - Epoch: [142][ 120/ 1236] Overall Loss 0.145481 Objective Loss 0.145481 LR 0.000250 Time 0.027352 +2023-10-02 21:37:23,273 - Epoch: [142][ 130/ 1236] Overall Loss 0.145653 Objective Loss 0.145653 LR 0.000250 Time 0.026823 +2023-10-02 21:37:23,480 - Epoch: [142][ 140/ 1236] Overall Loss 0.144956 Objective Loss 0.144956 LR 0.000250 Time 0.026380 +2023-10-02 21:37:23,685 - Epoch: [142][ 150/ 1236] Overall Loss 0.145512 Objective Loss 0.145512 LR 0.000250 Time 0.025988 +2023-10-02 21:37:23,895 - Epoch: [142][ 160/ 1236] Overall Loss 0.145200 Objective Loss 0.145200 LR 0.000250 Time 0.025672 +2023-10-02 21:37:24,105 - Epoch: [142][ 170/ 1236] Overall Loss 0.145056 Objective Loss 0.145056 LR 0.000250 Time 0.025396 +2023-10-02 21:37:24,315 - Epoch: [142][ 180/ 1236] Overall Loss 0.145294 Objective Loss 0.145294 LR 0.000250 Time 0.025149 +2023-10-02 21:37:24,525 - Epoch: [142][ 190/ 1236] Overall Loss 0.145265 Objective Loss 0.145265 LR 0.000250 Time 0.024928 +2023-10-02 21:37:24,735 - Epoch: [142][ 200/ 1236] Overall Loss 0.145291 Objective Loss 0.145291 LR 0.000250 Time 0.024729 +2023-10-02 21:37:24,945 - Epoch: [142][ 210/ 1236] Overall Loss 0.144747 Objective Loss 0.144747 LR 0.000250 Time 0.024549 +2023-10-02 21:37:25,155 - Epoch: [142][ 220/ 1236] Overall Loss 0.145369 Objective Loss 0.145369 LR 0.000250 Time 0.024386 +2023-10-02 21:37:25,365 - Epoch: [142][ 230/ 1236] Overall Loss 0.145044 Objective Loss 0.145044 LR 0.000250 Time 0.024238 +2023-10-02 21:37:25,575 - Epoch: [142][ 240/ 1236] Overall Loss 0.146448 Objective Loss 0.146448 LR 0.000250 Time 0.024101 +2023-10-02 21:37:25,785 - Epoch: [142][ 250/ 1236] Overall Loss 0.146672 Objective Loss 0.146672 LR 0.000250 Time 0.023976 +2023-10-02 21:37:25,995 - Epoch: [142][ 260/ 1236] Overall Loss 0.146053 Objective Loss 0.146053 LR 0.000250 Time 0.023860 +2023-10-02 21:37:26,205 - Epoch: [142][ 270/ 1236] Overall Loss 0.146164 Objective Loss 0.146164 LR 0.000250 Time 0.023752 +2023-10-02 21:37:26,415 - Epoch: [142][ 280/ 1236] Overall Loss 0.145979 Objective Loss 0.145979 LR 0.000250 Time 0.023652 +2023-10-02 21:37:26,625 - Epoch: [142][ 290/ 1236] Overall Loss 0.146118 Objective Loss 0.146118 LR 0.000250 Time 0.023559 +2023-10-02 21:37:26,833 - Epoch: [142][ 300/ 1236] Overall Loss 0.146028 Objective Loss 0.146028 LR 0.000250 Time 0.023467 +2023-10-02 21:37:27,036 - Epoch: [142][ 310/ 1236] Overall Loss 0.145937 Objective Loss 0.145937 LR 0.000250 Time 0.023363 +2023-10-02 21:37:27,239 - Epoch: [142][ 320/ 1236] Overall Loss 0.145564 Objective Loss 0.145564 LR 0.000250 Time 0.023268 +2023-10-02 21:37:27,442 - Epoch: [142][ 330/ 1236] Overall Loss 0.145355 Objective Loss 0.145355 LR 0.000250 Time 0.023177 +2023-10-02 21:37:27,645 - Epoch: [142][ 340/ 1236] Overall Loss 0.145673 Objective Loss 0.145673 LR 0.000250 Time 0.023092 +2023-10-02 21:37:27,849 - Epoch: [142][ 350/ 1236] Overall Loss 0.145901 Objective Loss 0.145901 LR 0.000250 Time 0.023013 +2023-10-02 21:37:28,055 - Epoch: [142][ 360/ 1236] Overall Loss 0.146225 Objective Loss 0.146225 LR 0.000250 Time 0.022945 +2023-10-02 21:37:28,260 - Epoch: [142][ 370/ 1236] Overall Loss 0.146612 Objective Loss 0.146612 LR 0.000250 Time 0.022878 +2023-10-02 21:37:28,467 - Epoch: [142][ 380/ 1236] Overall Loss 0.146802 Objective Loss 0.146802 LR 0.000250 Time 0.022819 +2023-10-02 21:37:28,671 - Epoch: [142][ 390/ 1236] Overall Loss 0.146922 Objective Loss 0.146922 LR 0.000250 Time 0.022758 +2023-10-02 21:37:28,877 - Epoch: [142][ 400/ 1236] Overall Loss 0.147004 Objective Loss 0.147004 LR 0.000250 Time 0.022702 +2023-10-02 21:37:29,082 - Epoch: [142][ 410/ 1236] Overall Loss 0.146667 Objective Loss 0.146667 LR 0.000250 Time 0.022647 +2023-10-02 21:37:29,287 - Epoch: [142][ 420/ 1236] Overall Loss 0.146896 Objective Loss 0.146896 LR 0.000250 Time 0.022597 +2023-10-02 21:37:29,492 - Epoch: [142][ 430/ 1236] Overall Loss 0.146885 Objective Loss 0.146885 LR 0.000250 Time 0.022547 +2023-10-02 21:37:29,698 - Epoch: [142][ 440/ 1236] Overall Loss 0.146847 Objective Loss 0.146847 LR 0.000250 Time 0.022501 +2023-10-02 21:37:29,903 - Epoch: [142][ 450/ 1236] Overall Loss 0.146682 Objective Loss 0.146682 LR 0.000250 Time 0.022455 +2023-10-02 21:37:30,109 - Epoch: [142][ 460/ 1236] Overall Loss 0.147092 Objective Loss 0.147092 LR 0.000250 Time 0.022414 +2023-10-02 21:37:30,314 - Epoch: [142][ 470/ 1236] Overall Loss 0.146828 Objective Loss 0.146828 LR 0.000250 Time 0.022372 +2023-10-02 21:37:30,520 - Epoch: [142][ 480/ 1236] Overall Loss 0.146613 Objective Loss 0.146613 LR 0.000250 Time 0.022334 +2023-10-02 21:37:30,725 - Epoch: [142][ 490/ 1236] Overall Loss 0.146307 Objective Loss 0.146307 LR 0.000250 Time 0.022296 +2023-10-02 21:37:30,930 - Epoch: [142][ 500/ 1236] Overall Loss 0.146256 Objective Loss 0.146256 LR 0.000250 Time 0.022261 +2023-10-02 21:37:31,136 - Epoch: [142][ 510/ 1236] Overall Loss 0.146243 Objective Loss 0.146243 LR 0.000250 Time 0.022227 +2023-10-02 21:37:31,342 - Epoch: [142][ 520/ 1236] Overall Loss 0.146399 Objective Loss 0.146399 LR 0.000250 Time 0.022195 +2023-10-02 21:37:31,547 - Epoch: [142][ 530/ 1236] Overall Loss 0.146603 Objective Loss 0.146603 LR 0.000250 Time 0.022163 +2023-10-02 21:37:31,754 - Epoch: [142][ 540/ 1236] Overall Loss 0.146616 Objective Loss 0.146616 LR 0.000250 Time 0.022135 +2023-10-02 21:37:31,959 - Epoch: [142][ 550/ 1236] Overall Loss 0.146814 Objective Loss 0.146814 LR 0.000250 Time 0.022103 +2023-10-02 21:37:32,166 - Epoch: [142][ 560/ 1236] Overall Loss 0.147045 Objective Loss 0.147045 LR 0.000250 Time 0.022078 +2023-10-02 21:37:32,370 - Epoch: [142][ 570/ 1236] Overall Loss 0.146853 Objective Loss 0.146853 LR 0.000250 Time 0.022049 +2023-10-02 21:37:32,576 - Epoch: [142][ 580/ 1236] Overall Loss 0.147102 Objective Loss 0.147102 LR 0.000250 Time 0.022023 +2023-10-02 21:37:32,781 - Epoch: [142][ 590/ 1236] Overall Loss 0.147203 Objective Loss 0.147203 LR 0.000250 Time 0.021996 +2023-10-02 21:37:32,987 - Epoch: [142][ 600/ 1236] Overall Loss 0.147270 Objective Loss 0.147270 LR 0.000250 Time 0.021972 +2023-10-02 21:37:33,192 - Epoch: [142][ 610/ 1236] Overall Loss 0.147387 Objective Loss 0.147387 LR 0.000250 Time 0.021947 +2023-10-02 21:37:33,399 - Epoch: [142][ 620/ 1236] Overall Loss 0.147563 Objective Loss 0.147563 LR 0.000250 Time 0.021927 +2023-10-02 21:37:33,604 - Epoch: [142][ 630/ 1236] Overall Loss 0.147524 Objective Loss 0.147524 LR 0.000250 Time 0.021904 +2023-10-02 21:37:33,811 - Epoch: [142][ 640/ 1236] Overall Loss 0.147668 Objective Loss 0.147668 LR 0.000250 Time 0.021884 +2023-10-02 21:37:34,016 - Epoch: [142][ 650/ 1236] Overall Loss 0.147483 Objective Loss 0.147483 LR 0.000250 Time 0.021862 +2023-10-02 21:37:34,222 - Epoch: [142][ 660/ 1236] Overall Loss 0.147566 Objective Loss 0.147566 LR 0.000250 Time 0.021843 +2023-10-02 21:37:34,427 - Epoch: [142][ 670/ 1236] Overall Loss 0.147609 Objective Loss 0.147609 LR 0.000250 Time 0.021823 +2023-10-02 21:37:34,633 - Epoch: [142][ 680/ 1236] Overall Loss 0.147694 Objective Loss 0.147694 LR 0.000250 Time 0.021804 +2023-10-02 21:37:34,838 - Epoch: [142][ 690/ 1236] Overall Loss 0.147658 Objective Loss 0.147658 LR 0.000250 Time 0.021785 +2023-10-02 21:37:35,044 - Epoch: [142][ 700/ 1236] Overall Loss 0.147885 Objective Loss 0.147885 LR 0.000250 Time 0.021767 +2023-10-02 21:37:35,249 - Epoch: [142][ 710/ 1236] Overall Loss 0.147913 Objective Loss 0.147913 LR 0.000250 Time 0.021748 +2023-10-02 21:37:35,455 - Epoch: [142][ 720/ 1236] Overall Loss 0.147693 Objective Loss 0.147693 LR 0.000250 Time 0.021732 +2023-10-02 21:37:35,660 - Epoch: [142][ 730/ 1236] Overall Loss 0.147531 Objective Loss 0.147531 LR 0.000250 Time 0.021714 +2023-10-02 21:37:35,867 - Epoch: [142][ 740/ 1236] Overall Loss 0.147592 Objective Loss 0.147592 LR 0.000250 Time 0.021700 +2023-10-02 21:37:36,071 - Epoch: [142][ 750/ 1236] Overall Loss 0.147589 Objective Loss 0.147589 LR 0.000250 Time 0.021683 +2023-10-02 21:37:36,278 - Epoch: [142][ 760/ 1236] Overall Loss 0.147601 Objective Loss 0.147601 LR 0.000250 Time 0.021669 +2023-10-02 21:37:36,483 - Epoch: [142][ 770/ 1236] Overall Loss 0.147638 Objective Loss 0.147638 LR 0.000250 Time 0.021654 +2023-10-02 21:37:36,689 - Epoch: [142][ 780/ 1236] Overall Loss 0.147752 Objective Loss 0.147752 LR 0.000250 Time 0.021640 +2023-10-02 21:37:36,895 - Epoch: [142][ 790/ 1236] Overall Loss 0.147615 Objective Loss 0.147615 LR 0.000250 Time 0.021625 +2023-10-02 21:37:37,100 - Epoch: [142][ 800/ 1236] Overall Loss 0.147632 Objective Loss 0.147632 LR 0.000250 Time 0.021611 +2023-10-02 21:37:37,306 - Epoch: [142][ 810/ 1236] Overall Loss 0.147516 Objective Loss 0.147516 LR 0.000250 Time 0.021598 +2023-10-02 21:37:37,511 - Epoch: [142][ 820/ 1236] Overall Loss 0.147402 Objective Loss 0.147402 LR 0.000250 Time 0.021585 +2023-10-02 21:37:37,717 - Epoch: [142][ 830/ 1236] Overall Loss 0.147100 Objective Loss 0.147100 LR 0.000250 Time 0.021572 +2023-10-02 21:37:37,924 - Epoch: [142][ 840/ 1236] Overall Loss 0.147074 Objective Loss 0.147074 LR 0.000250 Time 0.021561 +2023-10-02 21:37:38,128 - Epoch: [142][ 850/ 1236] Overall Loss 0.147284 Objective Loss 0.147284 LR 0.000250 Time 0.021548 +2023-10-02 21:37:38,335 - Epoch: [142][ 860/ 1236] Overall Loss 0.147343 Objective Loss 0.147343 LR 0.000250 Time 0.021537 +2023-10-02 21:37:38,540 - Epoch: [142][ 870/ 1236] Overall Loss 0.147667 Objective Loss 0.147667 LR 0.000250 Time 0.021525 +2023-10-02 21:37:38,746 - Epoch: [142][ 880/ 1236] Overall Loss 0.147574 Objective Loss 0.147574 LR 0.000250 Time 0.021514 +2023-10-02 21:37:38,951 - Epoch: [142][ 890/ 1236] Overall Loss 0.147692 Objective Loss 0.147692 LR 0.000250 Time 0.021502 +2023-10-02 21:37:39,157 - Epoch: [142][ 900/ 1236] Overall Loss 0.147941 Objective Loss 0.147941 LR 0.000250 Time 0.021492 +2023-10-02 21:37:39,363 - Epoch: [142][ 910/ 1236] Overall Loss 0.147902 Objective Loss 0.147902 LR 0.000250 Time 0.021481 +2023-10-02 21:37:39,569 - Epoch: [142][ 920/ 1236] Overall Loss 0.147702 Objective Loss 0.147702 LR 0.000250 Time 0.021471 +2023-10-02 21:37:39,774 - Epoch: [142][ 930/ 1236] Overall Loss 0.147700 Objective Loss 0.147700 LR 0.000250 Time 0.021460 +2023-10-02 21:37:39,981 - Epoch: [142][ 940/ 1236] Overall Loss 0.147911 Objective Loss 0.147911 LR 0.000250 Time 0.021452 +2023-10-02 21:37:40,186 - Epoch: [142][ 950/ 1236] Overall Loss 0.147979 Objective Loss 0.147979 LR 0.000250 Time 0.021441 +2023-10-02 21:37:40,393 - Epoch: [142][ 960/ 1236] Overall Loss 0.148158 Objective Loss 0.148158 LR 0.000250 Time 0.021433 +2023-10-02 21:37:40,598 - Epoch: [142][ 970/ 1236] Overall Loss 0.148313 Objective Loss 0.148313 LR 0.000250 Time 0.021423 +2023-10-02 21:37:40,805 - Epoch: [142][ 980/ 1236] Overall Loss 0.148448 Objective Loss 0.148448 LR 0.000250 Time 0.021415 +2023-10-02 21:37:41,009 - Epoch: [142][ 990/ 1236] Overall Loss 0.148559 Objective Loss 0.148559 LR 0.000250 Time 0.021405 +2023-10-02 21:37:41,215 - Epoch: [142][ 1000/ 1236] Overall Loss 0.148638 Objective Loss 0.148638 LR 0.000250 Time 0.021397 +2023-10-02 21:37:41,420 - Epoch: [142][ 1010/ 1236] Overall Loss 0.148618 Objective Loss 0.148618 LR 0.000250 Time 0.021387 +2023-10-02 21:37:41,627 - Epoch: [142][ 1020/ 1236] Overall Loss 0.148593 Objective Loss 0.148593 LR 0.000250 Time 0.021380 +2023-10-02 21:37:41,832 - Epoch: [142][ 1030/ 1236] Overall Loss 0.148668 Objective Loss 0.148668 LR 0.000250 Time 0.021371 +2023-10-02 21:37:42,038 - Epoch: [142][ 1040/ 1236] Overall Loss 0.148652 Objective Loss 0.148652 LR 0.000250 Time 0.021363 +2023-10-02 21:37:42,243 - Epoch: [142][ 1050/ 1236] Overall Loss 0.148647 Objective Loss 0.148647 LR 0.000250 Time 0.021354 +2023-10-02 21:37:42,449 - Epoch: [142][ 1060/ 1236] Overall Loss 0.148643 Objective Loss 0.148643 LR 0.000250 Time 0.021347 +2023-10-02 21:37:42,654 - Epoch: [142][ 1070/ 1236] Overall Loss 0.148506 Objective Loss 0.148506 LR 0.000250 Time 0.021339 +2023-10-02 21:37:42,860 - Epoch: [142][ 1080/ 1236] Overall Loss 0.148491 Objective Loss 0.148491 LR 0.000250 Time 0.021332 +2023-10-02 21:37:43,066 - Epoch: [142][ 1090/ 1236] Overall Loss 0.148545 Objective Loss 0.148545 LR 0.000250 Time 0.021324 +2023-10-02 21:37:43,272 - Epoch: [142][ 1100/ 1236] Overall Loss 0.148495 Objective Loss 0.148495 LR 0.000250 Time 0.021318 +2023-10-02 21:37:43,477 - Epoch: [142][ 1110/ 1236] Overall Loss 0.148589 Objective Loss 0.148589 LR 0.000250 Time 0.021310 +2023-10-02 21:37:43,683 - Epoch: [142][ 1120/ 1236] Overall Loss 0.148450 Objective Loss 0.148450 LR 0.000250 Time 0.021303 +2023-10-02 21:37:43,889 - Epoch: [142][ 1130/ 1236] Overall Loss 0.148301 Objective Loss 0.148301 LR 0.000250 Time 0.021297 +2023-10-02 21:37:44,096 - Epoch: [142][ 1140/ 1236] Overall Loss 0.148297 Objective Loss 0.148297 LR 0.000250 Time 0.021291 +2023-10-02 21:37:44,301 - Epoch: [142][ 1150/ 1236] Overall Loss 0.148448 Objective Loss 0.148448 LR 0.000250 Time 0.021284 +2023-10-02 21:37:44,507 - Epoch: [142][ 1160/ 1236] Overall Loss 0.148507 Objective Loss 0.148507 LR 0.000250 Time 0.021278 +2023-10-02 21:37:44,712 - Epoch: [142][ 1170/ 1236] Overall Loss 0.148420 Objective Loss 0.148420 LR 0.000250 Time 0.021271 +2023-10-02 21:37:44,918 - Epoch: [142][ 1180/ 1236] Overall Loss 0.148403 Objective Loss 0.148403 LR 0.000250 Time 0.021265 +2023-10-02 21:37:45,123 - Epoch: [142][ 1190/ 1236] Overall Loss 0.148578 Objective Loss 0.148578 LR 0.000250 Time 0.021258 +2023-10-02 21:37:45,330 - Epoch: [142][ 1200/ 1236] Overall Loss 0.148403 Objective Loss 0.148403 LR 0.000250 Time 0.021253 +2023-10-02 21:37:45,535 - Epoch: [142][ 1210/ 1236] Overall Loss 0.148384 Objective Loss 0.148384 LR 0.000250 Time 0.021247 +2023-10-02 21:37:45,741 - Epoch: [142][ 1220/ 1236] Overall Loss 0.148385 Objective Loss 0.148385 LR 0.000250 Time 0.021241 +2023-10-02 21:37:46,000 - Epoch: [142][ 1230/ 1236] Overall Loss 0.148291 Objective Loss 0.148291 LR 0.000250 Time 0.021279 +2023-10-02 21:37:46,122 - Epoch: [142][ 1236/ 1236] Overall Loss 0.148318 Objective Loss 0.148318 Top1 88.391039 Top5 98.981670 LR 0.000250 Time 0.021274 +2023-10-02 21:37:46,263 - --- validate (epoch=142)----------- +2023-10-02 21:37:46,263 - 29943 samples (256 per mini-batch) +2023-10-02 21:37:46,759 - Epoch: [142][ 10/ 117] Loss 0.285440 Top1 87.851562 Top5 98.593750 +2023-10-02 21:37:46,920 - Epoch: [142][ 20/ 117] Loss 0.273622 Top1 87.539062 Top5 98.613281 +2023-10-02 21:37:47,079 - Epoch: [142][ 30/ 117] Loss 0.284767 Top1 87.513021 Top5 98.554688 +2023-10-02 21:37:47,241 - Epoch: [142][ 40/ 117] Loss 0.293645 Top1 87.128906 Top5 98.486328 +2023-10-02 21:37:47,399 - Epoch: [142][ 50/ 117] Loss 0.295395 Top1 86.953125 Top5 98.476562 +2023-10-02 21:37:47,560 - Epoch: [142][ 60/ 117] Loss 0.291388 Top1 86.901042 Top5 98.476562 +2023-10-02 21:37:47,718 - Epoch: [142][ 70/ 117] Loss 0.289725 Top1 86.936384 Top5 98.554688 +2023-10-02 21:37:47,879 - Epoch: [142][ 80/ 117] Loss 0.293718 Top1 86.875000 Top5 98.544922 +2023-10-02 21:37:48,037 - Epoch: [142][ 90/ 117] Loss 0.294308 Top1 86.835938 Top5 98.550347 +2023-10-02 21:37:48,199 - Epoch: [142][ 100/ 117] Loss 0.293491 Top1 86.835938 Top5 98.566406 +2023-10-02 21:37:48,365 - Epoch: [142][ 110/ 117] Loss 0.294041 Top1 86.825284 Top5 98.579545 +2023-10-02 21:37:48,455 - Epoch: [142][ 117/ 117] Loss 0.292872 Top1 86.875063 Top5 98.567278 +2023-10-02 21:37:48,602 - ==> Top1: 86.875 Top5: 98.567 Loss: 0.293 + +2023-10-02 21:37:48,603 - ==> Confusion: +[[ 942 1 5 1 8 4 0 0 8 52 2 0 0 0 5 1 2 0 2 0 17] + [ 0 1066 0 1 4 14 0 24 2 1 0 0 0 0 1 3 0 0 8 2 5] + [ 2 1 981 11 0 0 12 7 0 1 2 0 7 2 0 3 2 2 14 3 6] + [ 1 4 13 984 1 2 0 1 0 1 4 0 4 3 21 3 1 7 17 0 22] + [ 25 8 1 1 961 5 0 0 1 13 0 0 2 2 11 3 9 1 0 1 6] + [ 2 37 2 4 4 992 0 24 2 4 0 5 2 9 4 0 1 1 2 3 18] + [ 0 3 24 2 0 2 1135 6 0 0 4 1 0 0 0 3 0 0 1 7 3] + [ 2 9 8 2 6 18 5 1089 0 5 4 0 3 6 4 1 0 3 37 8 8] + [ 17 0 1 2 3 1 0 2 974 33 14 0 2 16 17 0 1 0 1 4 1] + [ 100 1 2 0 4 5 0 0 24 948 0 0 0 16 7 1 0 1 0 1 9] + [ 4 3 11 9 0 1 6 2 9 0 979 2 0 6 2 0 0 2 4 3 10] + [ 1 2 0 0 0 18 0 4 0 0 0 965 12 8 0 1 0 16 0 6 2] + [ 0 1 3 5 0 1 1 0 1 0 2 36 974 1 3 5 1 14 1 8 11] + [ 0 0 0 0 4 13 0 0 12 7 3 8 0 1052 3 0 0 1 0 0 16] + [ 11 1 5 18 4 0 0 0 20 3 2 0 2 0 1011 0 1 3 10 0 10] + [ 0 1 2 1 7 0 0 0 0 0 0 3 5 0 0 1071 16 13 2 6 7] + [ 0 19 2 1 5 4 2 0 0 0 0 6 0 2 3 7 1097 0 1 1 11] + [ 0 0 2 1 0 0 1 0 0 0 0 1 12 0 0 5 0 1013 0 1 2] + [ 2 3 3 17 1 2 0 23 5 0 1 0 2 0 7 0 0 2 987 1 12] + [ 0 1 2 1 1 2 11 4 0 0 1 12 2 3 0 1 8 0 0 1098 5] + [ 107 157 132 92 68 103 33 83 67 71 147 84 299 249 100 39 84 52 102 142 5694]] + +2023-10-02 21:37:48,604 - ==> Best [Top1: 86.875 Top5: 98.567 Sparsity:0.00 Params: 169472 on epoch: 142] +2023-10-02 21:37:48,604 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:37:48,618 - + +2023-10-02 21:37:48,618 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:37:49,781 - Epoch: [143][ 10/ 1236] Overall Loss 0.139332 Objective Loss 0.139332 LR 0.000250 Time 0.116275 +2023-10-02 21:37:49,989 - Epoch: [143][ 20/ 1236] Overall Loss 0.134807 Objective Loss 0.134807 LR 0.000250 Time 0.068509 +2023-10-02 21:37:50,196 - Epoch: [143][ 30/ 1236] Overall Loss 0.135254 Objective Loss 0.135254 LR 0.000250 Time 0.052555 +2023-10-02 21:37:50,403 - Epoch: [143][ 40/ 1236] Overall Loss 0.141107 Objective Loss 0.141107 LR 0.000250 Time 0.044595 +2023-10-02 21:37:50,610 - Epoch: [143][ 50/ 1236] Overall Loss 0.140725 Objective Loss 0.140725 LR 0.000250 Time 0.039797 +2023-10-02 21:37:50,817 - Epoch: [143][ 60/ 1236] Overall Loss 0.139815 Objective Loss 0.139815 LR 0.000250 Time 0.036619 +2023-10-02 21:37:51,024 - Epoch: [143][ 70/ 1236] Overall Loss 0.142383 Objective Loss 0.142383 LR 0.000250 Time 0.034330 +2023-10-02 21:37:51,232 - Epoch: [143][ 80/ 1236] Overall Loss 0.141152 Objective Loss 0.141152 LR 0.000250 Time 0.032645 +2023-10-02 21:37:51,438 - Epoch: [143][ 90/ 1236] Overall Loss 0.140374 Objective Loss 0.140374 LR 0.000250 Time 0.031296 +2023-10-02 21:37:51,646 - Epoch: [143][ 100/ 1236] Overall Loss 0.140863 Objective Loss 0.140863 LR 0.000250 Time 0.030242 +2023-10-02 21:37:51,852 - Epoch: [143][ 110/ 1236] Overall Loss 0.142234 Objective Loss 0.142234 LR 0.000250 Time 0.029356 +2023-10-02 21:37:52,060 - Epoch: [143][ 120/ 1236] Overall Loss 0.142762 Objective Loss 0.142762 LR 0.000250 Time 0.028637 +2023-10-02 21:37:52,266 - Epoch: [143][ 130/ 1236] Overall Loss 0.142839 Objective Loss 0.142839 LR 0.000250 Time 0.028010 +2023-10-02 21:37:52,473 - Epoch: [143][ 140/ 1236] Overall Loss 0.144574 Objective Loss 0.144574 LR 0.000250 Time 0.027489 +2023-10-02 21:37:52,680 - Epoch: [143][ 150/ 1236] Overall Loss 0.143883 Objective Loss 0.143883 LR 0.000250 Time 0.027021 +2023-10-02 21:37:52,888 - Epoch: [143][ 160/ 1236] Overall Loss 0.142939 Objective Loss 0.142939 LR 0.000250 Time 0.026630 +2023-10-02 21:37:53,094 - Epoch: [143][ 170/ 1236] Overall Loss 0.143556 Objective Loss 0.143556 LR 0.000250 Time 0.026270 +2023-10-02 21:37:53,302 - Epoch: [143][ 180/ 1236] Overall Loss 0.144039 Objective Loss 0.144039 LR 0.000250 Time 0.025964 +2023-10-02 21:37:53,509 - Epoch: [143][ 190/ 1236] Overall Loss 0.143536 Objective Loss 0.143536 LR 0.000250 Time 0.025679 +2023-10-02 21:37:53,717 - Epoch: [143][ 200/ 1236] Overall Loss 0.143081 Objective Loss 0.143081 LR 0.000250 Time 0.025432 +2023-10-02 21:37:53,923 - Epoch: [143][ 210/ 1236] Overall Loss 0.142725 Objective Loss 0.142725 LR 0.000250 Time 0.025196 +2023-10-02 21:37:54,131 - Epoch: [143][ 220/ 1236] Overall Loss 0.142445 Objective Loss 0.142445 LR 0.000250 Time 0.024995 +2023-10-02 21:37:54,338 - Epoch: [143][ 230/ 1236] Overall Loss 0.142647 Objective Loss 0.142647 LR 0.000250 Time 0.024802 +2023-10-02 21:37:54,547 - Epoch: [143][ 240/ 1236] Overall Loss 0.142621 Objective Loss 0.142621 LR 0.000250 Time 0.024637 +2023-10-02 21:37:54,751 - Epoch: [143][ 250/ 1236] Overall Loss 0.142844 Objective Loss 0.142844 LR 0.000250 Time 0.024466 +2023-10-02 21:37:54,957 - Epoch: [143][ 260/ 1236] Overall Loss 0.142079 Objective Loss 0.142079 LR 0.000250 Time 0.024318 +2023-10-02 21:37:55,164 - Epoch: [143][ 270/ 1236] Overall Loss 0.142150 Objective Loss 0.142150 LR 0.000250 Time 0.024176 +2023-10-02 21:37:55,372 - Epoch: [143][ 280/ 1236] Overall Loss 0.142409 Objective Loss 0.142409 LR 0.000250 Time 0.024054 +2023-10-02 21:37:55,578 - Epoch: [143][ 290/ 1236] Overall Loss 0.141422 Objective Loss 0.141422 LR 0.000250 Time 0.023931 +2023-10-02 21:37:55,786 - Epoch: [143][ 300/ 1236] Overall Loss 0.141432 Objective Loss 0.141432 LR 0.000250 Time 0.023826 +2023-10-02 21:37:55,993 - Epoch: [143][ 310/ 1236] Overall Loss 0.141438 Objective Loss 0.141438 LR 0.000250 Time 0.023718 +2023-10-02 21:37:56,201 - Epoch: [143][ 320/ 1236] Overall Loss 0.141665 Objective Loss 0.141665 LR 0.000250 Time 0.023626 +2023-10-02 21:37:56,407 - Epoch: [143][ 330/ 1236] Overall Loss 0.141486 Objective Loss 0.141486 LR 0.000250 Time 0.023532 +2023-10-02 21:37:56,615 - Epoch: [143][ 340/ 1236] Overall Loss 0.140977 Objective Loss 0.140977 LR 0.000250 Time 0.023450 +2023-10-02 21:37:56,822 - Epoch: [143][ 350/ 1236] Overall Loss 0.141283 Objective Loss 0.141283 LR 0.000250 Time 0.023367 +2023-10-02 21:37:57,031 - Epoch: [143][ 360/ 1236] Overall Loss 0.141562 Objective Loss 0.141562 LR 0.000250 Time 0.023297 +2023-10-02 21:37:57,237 - Epoch: [143][ 370/ 1236] Overall Loss 0.141065 Objective Loss 0.141065 LR 0.000250 Time 0.023221 +2023-10-02 21:37:57,446 - Epoch: [143][ 380/ 1236] Overall Loss 0.141397 Objective Loss 0.141397 LR 0.000250 Time 0.023158 +2023-10-02 21:37:57,652 - Epoch: [143][ 390/ 1236] Overall Loss 0.141417 Objective Loss 0.141417 LR 0.000250 Time 0.023093 +2023-10-02 21:37:57,860 - Epoch: [143][ 400/ 1236] Overall Loss 0.141789 Objective Loss 0.141789 LR 0.000250 Time 0.023035 +2023-10-02 21:37:58,067 - Epoch: [143][ 410/ 1236] Overall Loss 0.142514 Objective Loss 0.142514 LR 0.000250 Time 0.022976 +2023-10-02 21:37:58,275 - Epoch: [143][ 420/ 1236] Overall Loss 0.142487 Objective Loss 0.142487 LR 0.000250 Time 0.022923 +2023-10-02 21:37:58,481 - Epoch: [143][ 430/ 1236] Overall Loss 0.142384 Objective Loss 0.142384 LR 0.000250 Time 0.022870 +2023-10-02 21:37:58,689 - Epoch: [143][ 440/ 1236] Overall Loss 0.142196 Objective Loss 0.142196 LR 0.000250 Time 0.022822 +2023-10-02 21:37:58,895 - Epoch: [143][ 450/ 1236] Overall Loss 0.141992 Objective Loss 0.141992 LR 0.000250 Time 0.022772 +2023-10-02 21:37:59,103 - Epoch: [143][ 460/ 1236] Overall Loss 0.142328 Objective Loss 0.142328 LR 0.000250 Time 0.022729 +2023-10-02 21:37:59,310 - Epoch: [143][ 470/ 1236] Overall Loss 0.142784 Objective Loss 0.142784 LR 0.000250 Time 0.022684 +2023-10-02 21:37:59,518 - Epoch: [143][ 480/ 1236] Overall Loss 0.142829 Objective Loss 0.142829 LR 0.000250 Time 0.022644 +2023-10-02 21:37:59,724 - Epoch: [143][ 490/ 1236] Overall Loss 0.142980 Objective Loss 0.142980 LR 0.000250 Time 0.022603 +2023-10-02 21:37:59,932 - Epoch: [143][ 500/ 1236] Overall Loss 0.143641 Objective Loss 0.143641 LR 0.000250 Time 0.022566 +2023-10-02 21:38:00,139 - Epoch: [143][ 510/ 1236] Overall Loss 0.143810 Objective Loss 0.143810 LR 0.000250 Time 0.022528 +2023-10-02 21:38:00,347 - Epoch: [143][ 520/ 1236] Overall Loss 0.144401 Objective Loss 0.144401 LR 0.000250 Time 0.022495 +2023-10-02 21:38:00,554 - Epoch: [143][ 530/ 1236] Overall Loss 0.144893 Objective Loss 0.144893 LR 0.000250 Time 0.022461 +2023-10-02 21:38:00,762 - Epoch: [143][ 540/ 1236] Overall Loss 0.145561 Objective Loss 0.145561 LR 0.000250 Time 0.022430 +2023-10-02 21:38:00,969 - Epoch: [143][ 550/ 1236] Overall Loss 0.145661 Objective Loss 0.145661 LR 0.000250 Time 0.022397 +2023-10-02 21:38:01,177 - Epoch: [143][ 560/ 1236] Overall Loss 0.145546 Objective Loss 0.145546 LR 0.000250 Time 0.022368 +2023-10-02 21:38:01,384 - Epoch: [143][ 570/ 1236] Overall Loss 0.145331 Objective Loss 0.145331 LR 0.000250 Time 0.022338 +2023-10-02 21:38:01,592 - Epoch: [143][ 580/ 1236] Overall Loss 0.145527 Objective Loss 0.145527 LR 0.000250 Time 0.022311 +2023-10-02 21:38:01,799 - Epoch: [143][ 590/ 1236] Overall Loss 0.145817 Objective Loss 0.145817 LR 0.000250 Time 0.022283 +2023-10-02 21:38:02,006 - Epoch: [143][ 600/ 1236] Overall Loss 0.145887 Objective Loss 0.145887 LR 0.000250 Time 0.022257 +2023-10-02 21:38:02,212 - Epoch: [143][ 610/ 1236] Overall Loss 0.145908 Objective Loss 0.145908 LR 0.000250 Time 0.022229 +2023-10-02 21:38:02,421 - Epoch: [143][ 620/ 1236] Overall Loss 0.145632 Objective Loss 0.145632 LR 0.000250 Time 0.022206 +2023-10-02 21:38:02,627 - Epoch: [143][ 630/ 1236] Overall Loss 0.145440 Objective Loss 0.145440 LR 0.000250 Time 0.022181 +2023-10-02 21:38:02,836 - Epoch: [143][ 640/ 1236] Overall Loss 0.145525 Objective Loss 0.145525 LR 0.000250 Time 0.022160 +2023-10-02 21:38:03,042 - Epoch: [143][ 650/ 1236] Overall Loss 0.145594 Objective Loss 0.145594 LR 0.000250 Time 0.022136 +2023-10-02 21:38:03,250 - Epoch: [143][ 660/ 1236] Overall Loss 0.145679 Objective Loss 0.145679 LR 0.000250 Time 0.022114 +2023-10-02 21:38:03,456 - Epoch: [143][ 670/ 1236] Overall Loss 0.145900 Objective Loss 0.145900 LR 0.000250 Time 0.022092 +2023-10-02 21:38:03,664 - Epoch: [143][ 680/ 1236] Overall Loss 0.145972 Objective Loss 0.145972 LR 0.000250 Time 0.022073 +2023-10-02 21:38:03,871 - Epoch: [143][ 690/ 1236] Overall Loss 0.145839 Objective Loss 0.145839 LR 0.000250 Time 0.022052 +2023-10-02 21:38:04,079 - Epoch: [143][ 700/ 1236] Overall Loss 0.145646 Objective Loss 0.145646 LR 0.000250 Time 0.022034 +2023-10-02 21:38:04,286 - Epoch: [143][ 710/ 1236] Overall Loss 0.145506 Objective Loss 0.145506 LR 0.000250 Time 0.022015 +2023-10-02 21:38:04,494 - Epoch: [143][ 720/ 1236] Overall Loss 0.145494 Objective Loss 0.145494 LR 0.000250 Time 0.021998 +2023-10-02 21:38:04,701 - Epoch: [143][ 730/ 1236] Overall Loss 0.145425 Objective Loss 0.145425 LR 0.000250 Time 0.021979 +2023-10-02 21:38:04,909 - Epoch: [143][ 740/ 1236] Overall Loss 0.145135 Objective Loss 0.145135 LR 0.000250 Time 0.021963 +2023-10-02 21:38:05,116 - Epoch: [143][ 750/ 1236] Overall Loss 0.145291 Objective Loss 0.145291 LR 0.000250 Time 0.021946 +2023-10-02 21:38:05,325 - Epoch: [143][ 760/ 1236] Overall Loss 0.145394 Objective Loss 0.145394 LR 0.000250 Time 0.021931 +2023-10-02 21:38:05,531 - Epoch: [143][ 770/ 1236] Overall Loss 0.145199 Objective Loss 0.145199 LR 0.000250 Time 0.021914 +2023-10-02 21:38:05,739 - Epoch: [143][ 780/ 1236] Overall Loss 0.145044 Objective Loss 0.145044 LR 0.000250 Time 0.021899 +2023-10-02 21:38:05,946 - Epoch: [143][ 790/ 1236] Overall Loss 0.145182 Objective Loss 0.145182 LR 0.000250 Time 0.021884 +2023-10-02 21:38:06,154 - Epoch: [143][ 800/ 1236] Overall Loss 0.145201 Objective Loss 0.145201 LR 0.000250 Time 0.021870 +2023-10-02 21:38:06,361 - Epoch: [143][ 810/ 1236] Overall Loss 0.145162 Objective Loss 0.145162 LR 0.000250 Time 0.021854 +2023-10-02 21:38:06,569 - Epoch: [143][ 820/ 1236] Overall Loss 0.145141 Objective Loss 0.145141 LR 0.000250 Time 0.021841 +2023-10-02 21:38:06,776 - Epoch: [143][ 830/ 1236] Overall Loss 0.145414 Objective Loss 0.145414 LR 0.000250 Time 0.021827 +2023-10-02 21:38:06,984 - Epoch: [143][ 840/ 1236] Overall Loss 0.145231 Objective Loss 0.145231 LR 0.000250 Time 0.021815 +2023-10-02 21:38:07,191 - Epoch: [143][ 850/ 1236] Overall Loss 0.145189 Objective Loss 0.145189 LR 0.000250 Time 0.021801 +2023-10-02 21:38:07,399 - Epoch: [143][ 860/ 1236] Overall Loss 0.145347 Objective Loss 0.145347 LR 0.000250 Time 0.021789 +2023-10-02 21:38:07,606 - Epoch: [143][ 870/ 1236] Overall Loss 0.145438 Objective Loss 0.145438 LR 0.000250 Time 0.021775 +2023-10-02 21:38:07,814 - Epoch: [143][ 880/ 1236] Overall Loss 0.145239 Objective Loss 0.145239 LR 0.000250 Time 0.021764 +2023-10-02 21:38:08,021 - Epoch: [143][ 890/ 1236] Overall Loss 0.145143 Objective Loss 0.145143 LR 0.000250 Time 0.021751 +2023-10-02 21:38:08,229 - Epoch: [143][ 900/ 1236] Overall Loss 0.144873 Objective Loss 0.144873 LR 0.000250 Time 0.021741 +2023-10-02 21:38:08,436 - Epoch: [143][ 910/ 1236] Overall Loss 0.144818 Objective Loss 0.144818 LR 0.000250 Time 0.021729 +2023-10-02 21:38:08,645 - Epoch: [143][ 920/ 1236] Overall Loss 0.144913 Objective Loss 0.144913 LR 0.000250 Time 0.021719 +2023-10-02 21:38:08,851 - Epoch: [143][ 930/ 1236] Overall Loss 0.145029 Objective Loss 0.145029 LR 0.000250 Time 0.021706 +2023-10-02 21:38:09,059 - Epoch: [143][ 940/ 1236] Overall Loss 0.145410 Objective Loss 0.145410 LR 0.000250 Time 0.021696 +2023-10-02 21:38:09,266 - Epoch: [143][ 950/ 1236] Overall Loss 0.145370 Objective Loss 0.145370 LR 0.000250 Time 0.021683 +2023-10-02 21:38:09,474 - Epoch: [143][ 960/ 1236] Overall Loss 0.145391 Objective Loss 0.145391 LR 0.000250 Time 0.021674 +2023-10-02 21:38:09,681 - Epoch: [143][ 970/ 1236] Overall Loss 0.145511 Objective Loss 0.145511 LR 0.000250 Time 0.021662 +2023-10-02 21:38:09,891 - Epoch: [143][ 980/ 1236] Overall Loss 0.145802 Objective Loss 0.145802 LR 0.000250 Time 0.021655 +2023-10-02 21:38:10,095 - Epoch: [143][ 990/ 1236] Overall Loss 0.145458 Objective Loss 0.145458 LR 0.000250 Time 0.021643 +2023-10-02 21:38:10,304 - Epoch: [143][ 1000/ 1236] Overall Loss 0.145656 Objective Loss 0.145656 LR 0.000250 Time 0.021634 +2023-10-02 21:38:10,511 - Epoch: [143][ 1010/ 1236] Overall Loss 0.145714 Objective Loss 0.145714 LR 0.000250 Time 0.021623 +2023-10-02 21:38:10,719 - Epoch: [143][ 1020/ 1236] Overall Loss 0.145690 Objective Loss 0.145690 LR 0.000250 Time 0.021615 +2023-10-02 21:38:10,926 - Epoch: [143][ 1030/ 1236] Overall Loss 0.145873 Objective Loss 0.145873 LR 0.000250 Time 0.021604 +2023-10-02 21:38:11,133 - Epoch: [143][ 1040/ 1236] Overall Loss 0.146114 Objective Loss 0.146114 LR 0.000250 Time 0.021596 +2023-10-02 21:38:11,339 - Epoch: [143][ 1050/ 1236] Overall Loss 0.145999 Objective Loss 0.145999 LR 0.000250 Time 0.021585 +2023-10-02 21:38:11,549 - Epoch: [143][ 1060/ 1236] Overall Loss 0.146146 Objective Loss 0.146146 LR 0.000250 Time 0.021579 +2023-10-02 21:38:11,754 - Epoch: [143][ 1070/ 1236] Overall Loss 0.146141 Objective Loss 0.146141 LR 0.000250 Time 0.021569 +2023-10-02 21:38:11,963 - Epoch: [143][ 1080/ 1236] Overall Loss 0.146169 Objective Loss 0.146169 LR 0.000250 Time 0.021562 +2023-10-02 21:38:12,170 - Epoch: [143][ 1090/ 1236] Overall Loss 0.146242 Objective Loss 0.146242 LR 0.000250 Time 0.021553 +2023-10-02 21:38:12,378 - Epoch: [143][ 1100/ 1236] Overall Loss 0.146269 Objective Loss 0.146269 LR 0.000250 Time 0.021545 +2023-10-02 21:38:12,584 - Epoch: [143][ 1110/ 1236] Overall Loss 0.146391 Objective Loss 0.146391 LR 0.000250 Time 0.021536 +2023-10-02 21:38:12,793 - Epoch: [143][ 1120/ 1236] Overall Loss 0.146353 Objective Loss 0.146353 LR 0.000250 Time 0.021529 +2023-10-02 21:38:12,999 - Epoch: [143][ 1130/ 1236] Overall Loss 0.146281 Objective Loss 0.146281 LR 0.000250 Time 0.021521 +2023-10-02 21:38:13,209 - Epoch: [143][ 1140/ 1236] Overall Loss 0.146109 Objective Loss 0.146109 LR 0.000250 Time 0.021515 +2023-10-02 21:38:13,415 - Epoch: [143][ 1150/ 1236] Overall Loss 0.146118 Objective Loss 0.146118 LR 0.000250 Time 0.021507 +2023-10-02 21:38:13,623 - Epoch: [143][ 1160/ 1236] Overall Loss 0.146411 Objective Loss 0.146411 LR 0.000250 Time 0.021501 +2023-10-02 21:38:13,830 - Epoch: [143][ 1170/ 1236] Overall Loss 0.146485 Objective Loss 0.146485 LR 0.000250 Time 0.021492 +2023-10-02 21:38:14,038 - Epoch: [143][ 1180/ 1236] Overall Loss 0.146326 Objective Loss 0.146326 LR 0.000250 Time 0.021486 +2023-10-02 21:38:14,244 - Epoch: [143][ 1190/ 1236] Overall Loss 0.146487 Objective Loss 0.146487 LR 0.000250 Time 0.021478 +2023-10-02 21:38:14,452 - Epoch: [143][ 1200/ 1236] Overall Loss 0.146681 Objective Loss 0.146681 LR 0.000250 Time 0.021472 +2023-10-02 21:38:14,659 - Epoch: [143][ 1210/ 1236] Overall Loss 0.146723 Objective Loss 0.146723 LR 0.000250 Time 0.021464 +2023-10-02 21:38:14,867 - Epoch: [143][ 1220/ 1236] Overall Loss 0.146731 Objective Loss 0.146731 LR 0.000250 Time 0.021459 +2023-10-02 21:38:15,129 - Epoch: [143][ 1230/ 1236] Overall Loss 0.146889 Objective Loss 0.146889 LR 0.000250 Time 0.021496 +2023-10-02 21:38:15,251 - Epoch: [143][ 1236/ 1236] Overall Loss 0.146930 Objective Loss 0.146930 Top1 91.446029 Top5 99.185336 LR 0.000250 Time 0.021490 +2023-10-02 21:38:15,383 - --- validate (epoch=143)----------- +2023-10-02 21:38:15,383 - 29943 samples (256 per mini-batch) +2023-10-02 21:38:15,885 - Epoch: [143][ 10/ 117] Loss 0.307705 Top1 85.156250 Top5 98.789062 +2023-10-02 21:38:16,032 - Epoch: [143][ 20/ 117] Loss 0.302224 Top1 85.625000 Top5 98.535156 +2023-10-02 21:38:16,179 - Epoch: [143][ 30/ 117] Loss 0.293169 Top1 86.184896 Top5 98.645833 +2023-10-02 21:38:16,326 - Epoch: [143][ 40/ 117] Loss 0.297607 Top1 86.357422 Top5 98.583984 +2023-10-02 21:38:16,473 - Epoch: [143][ 50/ 117] Loss 0.293465 Top1 86.546875 Top5 98.546875 +2023-10-02 21:38:16,620 - Epoch: [143][ 60/ 117] Loss 0.290128 Top1 86.634115 Top5 98.535156 +2023-10-02 21:38:16,766 - Epoch: [143][ 70/ 117] Loss 0.298601 Top1 86.456473 Top5 98.454241 +2023-10-02 21:38:16,913 - Epoch: [143][ 80/ 117] Loss 0.297328 Top1 86.484375 Top5 98.427734 +2023-10-02 21:38:17,059 - Epoch: [143][ 90/ 117] Loss 0.299464 Top1 86.614583 Top5 98.450521 +2023-10-02 21:38:17,205 - Epoch: [143][ 100/ 117] Loss 0.303760 Top1 86.464844 Top5 98.433594 +2023-10-02 21:38:17,359 - Epoch: [143][ 110/ 117] Loss 0.299590 Top1 86.526989 Top5 98.441051 +2023-10-02 21:38:17,449 - Epoch: [143][ 117/ 117] Loss 0.297470 Top1 86.551114 Top5 98.487126 +2023-10-02 21:38:17,595 - ==> Top1: 86.551 Top5: 98.487 Loss: 0.297 + +2023-10-02 21:38:17,596 - ==> Confusion: +[[ 963 0 3 0 4 2 0 0 4 48 3 0 1 0 3 2 1 0 0 0 16] + [ 0 1065 0 1 3 17 1 21 3 1 1 0 0 0 0 2 1 0 10 2 3] + [ 2 2 984 9 0 0 16 5 0 2 1 0 7 2 0 3 1 1 10 4 7] + [ 3 2 14 977 0 0 4 2 5 0 3 0 4 2 29 4 0 5 17 1 17] + [ 30 5 0 1 972 6 0 0 1 8 0 0 0 1 8 6 6 0 0 0 6] + [ 2 42 2 3 6 995 0 19 0 7 2 4 4 3 4 1 1 1 4 1 15] + [ 0 4 25 2 0 2 1140 0 0 0 2 1 0 0 0 3 0 0 2 6 4] + [ 1 11 12 0 6 22 3 1087 1 3 4 4 3 5 1 2 1 2 33 8 9] + [ 16 0 0 2 3 1 0 2 980 40 9 2 1 8 11 3 1 2 2 4 2] + [ 93 1 0 1 6 3 0 0 24 962 0 1 0 10 8 3 0 0 0 1 6] + [ 3 0 10 9 0 1 4 1 11 2 977 1 1 7 2 0 3 3 11 0 7] + [ 0 2 3 0 1 16 0 5 0 0 0 958 24 5 0 2 1 11 0 4 3] + [ 0 0 5 7 1 0 1 0 0 2 3 28 984 0 3 6 0 12 2 5 9] + [ 0 0 0 0 4 8 0 0 8 11 3 7 1 1051 3 2 2 1 0 2 16] + [ 14 0 4 19 5 1 0 0 18 3 4 0 1 1 1005 0 1 4 10 0 11] + [ 0 0 1 1 7 0 0 0 0 1 0 6 7 0 0 1074 14 9 2 7 5] + [ 2 17 1 1 4 6 1 0 0 1 0 4 0 2 4 10 1091 0 1 4 12] + [ 0 0 2 3 0 0 2 0 0 1 0 4 23 1 2 4 0 990 0 1 5] + [ 3 3 4 16 0 1 1 14 5 1 2 1 2 0 13 0 0 1 989 1 11] + [ 0 1 4 4 1 2 6 6 0 0 1 12 8 1 1 1 6 0 0 1090 8] + [ 124 137 117 89 68 117 30 91 95 92 149 82 345 227 114 60 63 62 108 154 5581]] + +2023-10-02 21:38:17,597 - ==> Best [Top1: 86.875 Top5: 98.567 Sparsity:0.00 Params: 169472 on epoch: 142] +2023-10-02 21:38:17,597 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:38:17,604 - + +2023-10-02 21:38:17,604 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:38:18,652 - Epoch: [144][ 10/ 1236] Overall Loss 0.150466 Objective Loss 0.150466 LR 0.000250 Time 0.104756 +2023-10-02 21:38:18,860 - Epoch: [144][ 20/ 1236] Overall Loss 0.149828 Objective Loss 0.149828 LR 0.000250 Time 0.062773 +2023-10-02 21:38:19,066 - Epoch: [144][ 30/ 1236] Overall Loss 0.148843 Objective Loss 0.148843 LR 0.000250 Time 0.048708 +2023-10-02 21:38:19,274 - Epoch: [144][ 40/ 1236] Overall Loss 0.142430 Objective Loss 0.142430 LR 0.000250 Time 0.041718 +2023-10-02 21:38:19,480 - Epoch: [144][ 50/ 1236] Overall Loss 0.145410 Objective Loss 0.145410 LR 0.000250 Time 0.037500 +2023-10-02 21:38:19,688 - Epoch: [144][ 60/ 1236] Overall Loss 0.141671 Objective Loss 0.141671 LR 0.000250 Time 0.034704 +2023-10-02 21:38:19,895 - Epoch: [144][ 70/ 1236] Overall Loss 0.139134 Objective Loss 0.139134 LR 0.000250 Time 0.032693 +2023-10-02 21:38:20,103 - Epoch: [144][ 80/ 1236] Overall Loss 0.139705 Objective Loss 0.139705 LR 0.000250 Time 0.031208 +2023-10-02 21:38:20,317 - Epoch: [144][ 90/ 1236] Overall Loss 0.139655 Objective Loss 0.139655 LR 0.000250 Time 0.030109 +2023-10-02 21:38:20,534 - Epoch: [144][ 100/ 1236] Overall Loss 0.140969 Objective Loss 0.140969 LR 0.000250 Time 0.029272 +2023-10-02 21:38:20,757 - Epoch: [144][ 110/ 1236] Overall Loss 0.140513 Objective Loss 0.140513 LR 0.000250 Time 0.028633 +2023-10-02 21:38:20,976 - Epoch: [144][ 120/ 1236] Overall Loss 0.139687 Objective Loss 0.139687 LR 0.000250 Time 0.028067 +2023-10-02 21:38:21,195 - Epoch: [144][ 130/ 1236] Overall Loss 0.140101 Objective Loss 0.140101 LR 0.000250 Time 0.027588 +2023-10-02 21:38:21,409 - Epoch: [144][ 140/ 1236] Overall Loss 0.140288 Objective Loss 0.140288 LR 0.000250 Time 0.027142 +2023-10-02 21:38:21,624 - Epoch: [144][ 150/ 1236] Overall Loss 0.140056 Objective Loss 0.140056 LR 0.000250 Time 0.026761 +2023-10-02 21:38:21,835 - Epoch: [144][ 160/ 1236] Overall Loss 0.141018 Objective Loss 0.141018 LR 0.000250 Time 0.026405 +2023-10-02 21:38:22,047 - Epoch: [144][ 170/ 1236] Overall Loss 0.141150 Objective Loss 0.141150 LR 0.000250 Time 0.026094 +2023-10-02 21:38:22,258 - Epoch: [144][ 180/ 1236] Overall Loss 0.141206 Objective Loss 0.141206 LR 0.000250 Time 0.025818 +2023-10-02 21:38:22,470 - Epoch: [144][ 190/ 1236] Overall Loss 0.141827 Objective Loss 0.141827 LR 0.000250 Time 0.025571 +2023-10-02 21:38:22,682 - Epoch: [144][ 200/ 1236] Overall Loss 0.141158 Objective Loss 0.141158 LR 0.000250 Time 0.025346 +2023-10-02 21:38:22,893 - Epoch: [144][ 210/ 1236] Overall Loss 0.140814 Objective Loss 0.140814 LR 0.000250 Time 0.025145 +2023-10-02 21:38:23,104 - Epoch: [144][ 220/ 1236] Overall Loss 0.141452 Objective Loss 0.141452 LR 0.000250 Time 0.024954 +2023-10-02 21:38:23,316 - Epoch: [144][ 230/ 1236] Overall Loss 0.142159 Objective Loss 0.142159 LR 0.000250 Time 0.024788 +2023-10-02 21:38:23,528 - Epoch: [144][ 240/ 1236] Overall Loss 0.142808 Objective Loss 0.142808 LR 0.000250 Time 0.024634 +2023-10-02 21:38:23,740 - Epoch: [144][ 250/ 1236] Overall Loss 0.143368 Objective Loss 0.143368 LR 0.000250 Time 0.024494 +2023-10-02 21:38:23,950 - Epoch: [144][ 260/ 1236] Overall Loss 0.142715 Objective Loss 0.142715 LR 0.000250 Time 0.024361 +2023-10-02 21:38:24,162 - Epoch: [144][ 270/ 1236] Overall Loss 0.142469 Objective Loss 0.142469 LR 0.000250 Time 0.024241 +2023-10-02 21:38:24,373 - Epoch: [144][ 280/ 1236] Overall Loss 0.142513 Objective Loss 0.142513 LR 0.000250 Time 0.024128 +2023-10-02 21:38:24,586 - Epoch: [144][ 290/ 1236] Overall Loss 0.142840 Objective Loss 0.142840 LR 0.000250 Time 0.024026 +2023-10-02 21:38:24,797 - Epoch: [144][ 300/ 1236] Overall Loss 0.142540 Objective Loss 0.142540 LR 0.000250 Time 0.023927 +2023-10-02 21:38:25,009 - Epoch: [144][ 310/ 1236] Overall Loss 0.142458 Objective Loss 0.142458 LR 0.000250 Time 0.023838 +2023-10-02 21:38:25,220 - Epoch: [144][ 320/ 1236] Overall Loss 0.142950 Objective Loss 0.142950 LR 0.000250 Time 0.023752 +2023-10-02 21:38:25,432 - Epoch: [144][ 330/ 1236] Overall Loss 0.142884 Objective Loss 0.142884 LR 0.000250 Time 0.023673 +2023-10-02 21:38:25,643 - Epoch: [144][ 340/ 1236] Overall Loss 0.142175 Objective Loss 0.142175 LR 0.000250 Time 0.023596 +2023-10-02 21:38:25,855 - Epoch: [144][ 350/ 1236] Overall Loss 0.142078 Objective Loss 0.142078 LR 0.000250 Time 0.023525 +2023-10-02 21:38:26,063 - Epoch: [144][ 360/ 1236] Overall Loss 0.142228 Objective Loss 0.142228 LR 0.000250 Time 0.023448 +2023-10-02 21:38:26,274 - Epoch: [144][ 370/ 1236] Overall Loss 0.142286 Objective Loss 0.142286 LR 0.000250 Time 0.023384 +2023-10-02 21:38:26,482 - Epoch: [144][ 380/ 1236] Overall Loss 0.142397 Objective Loss 0.142397 LR 0.000250 Time 0.023316 +2023-10-02 21:38:26,694 - Epoch: [144][ 390/ 1236] Overall Loss 0.142777 Objective Loss 0.142777 LR 0.000250 Time 0.023260 +2023-10-02 21:38:26,902 - Epoch: [144][ 400/ 1236] Overall Loss 0.143004 Objective Loss 0.143004 LR 0.000250 Time 0.023198 +2023-10-02 21:38:27,113 - Epoch: [144][ 410/ 1236] Overall Loss 0.142363 Objective Loss 0.142363 LR 0.000250 Time 0.023144 +2023-10-02 21:38:27,322 - Epoch: [144][ 420/ 1236] Overall Loss 0.142322 Objective Loss 0.142322 LR 0.000250 Time 0.023091 +2023-10-02 21:38:27,533 - Epoch: [144][ 430/ 1236] Overall Loss 0.142796 Objective Loss 0.142796 LR 0.000250 Time 0.023043 +2023-10-02 21:38:27,742 - Epoch: [144][ 440/ 1236] Overall Loss 0.143353 Objective Loss 0.143353 LR 0.000250 Time 0.022994 +2023-10-02 21:38:27,953 - Epoch: [144][ 450/ 1236] Overall Loss 0.143481 Objective Loss 0.143481 LR 0.000250 Time 0.022950 +2023-10-02 21:38:28,162 - Epoch: [144][ 460/ 1236] Overall Loss 0.143241 Objective Loss 0.143241 LR 0.000250 Time 0.022905 +2023-10-02 21:38:28,374 - Epoch: [144][ 470/ 1236] Overall Loss 0.143689 Objective Loss 0.143689 LR 0.000250 Time 0.022868 +2023-10-02 21:38:28,582 - Epoch: [144][ 480/ 1236] Overall Loss 0.143335 Objective Loss 0.143335 LR 0.000250 Time 0.022825 +2023-10-02 21:38:28,794 - Epoch: [144][ 490/ 1236] Overall Loss 0.143216 Objective Loss 0.143216 LR 0.000250 Time 0.022790 +2023-10-02 21:38:29,002 - Epoch: [144][ 500/ 1236] Overall Loss 0.143539 Objective Loss 0.143539 LR 0.000250 Time 0.022750 +2023-10-02 21:38:29,214 - Epoch: [144][ 510/ 1236] Overall Loss 0.143739 Objective Loss 0.143739 LR 0.000250 Time 0.022718 +2023-10-02 21:38:29,423 - Epoch: [144][ 520/ 1236] Overall Loss 0.143887 Objective Loss 0.143887 LR 0.000250 Time 0.022682 +2023-10-02 21:38:29,635 - Epoch: [144][ 530/ 1236] Overall Loss 0.144054 Objective Loss 0.144054 LR 0.000250 Time 0.022654 +2023-10-02 21:38:29,843 - Epoch: [144][ 540/ 1236] Overall Loss 0.144134 Objective Loss 0.144134 LR 0.000250 Time 0.022618 +2023-10-02 21:38:30,053 - Epoch: [144][ 550/ 1236] Overall Loss 0.144110 Objective Loss 0.144110 LR 0.000250 Time 0.022588 +2023-10-02 21:38:30,263 - Epoch: [144][ 560/ 1236] Overall Loss 0.143857 Objective Loss 0.143857 LR 0.000250 Time 0.022556 +2023-10-02 21:38:30,474 - Epoch: [144][ 570/ 1236] Overall Loss 0.143600 Objective Loss 0.143600 LR 0.000250 Time 0.022531 +2023-10-02 21:38:30,682 - Epoch: [144][ 580/ 1236] Overall Loss 0.143856 Objective Loss 0.143856 LR 0.000250 Time 0.022501 +2023-10-02 21:38:30,894 - Epoch: [144][ 590/ 1236] Overall Loss 0.144097 Objective Loss 0.144097 LR 0.000250 Time 0.022477 +2023-10-02 21:38:31,107 - Epoch: [144][ 600/ 1236] Overall Loss 0.144109 Objective Loss 0.144109 LR 0.000250 Time 0.022449 +2023-10-02 21:38:31,317 - Epoch: [144][ 610/ 1236] Overall Loss 0.143899 Objective Loss 0.143899 LR 0.000250 Time 0.022424 +2023-10-02 21:38:31,526 - Epoch: [144][ 620/ 1236] Overall Loss 0.144134 Objective Loss 0.144134 LR 0.000250 Time 0.022398 +2023-10-02 21:38:31,738 - Epoch: [144][ 630/ 1236] Overall Loss 0.144065 Objective Loss 0.144065 LR 0.000250 Time 0.022378 +2023-10-02 21:38:31,947 - Epoch: [144][ 640/ 1236] Overall Loss 0.144066 Objective Loss 0.144066 LR 0.000250 Time 0.022353 +2023-10-02 21:38:32,158 - Epoch: [144][ 650/ 1236] Overall Loss 0.144058 Objective Loss 0.144058 LR 0.000250 Time 0.022334 +2023-10-02 21:38:32,367 - Epoch: [144][ 660/ 1236] Overall Loss 0.143950 Objective Loss 0.143950 LR 0.000250 Time 0.022311 +2023-10-02 21:38:32,578 - Epoch: [144][ 670/ 1236] Overall Loss 0.143797 Objective Loss 0.143797 LR 0.000250 Time 0.022293 +2023-10-02 21:38:32,789 - Epoch: [144][ 680/ 1236] Overall Loss 0.143734 Objective Loss 0.143734 LR 0.000250 Time 0.022273 +2023-10-02 21:38:33,003 - Epoch: [144][ 690/ 1236] Overall Loss 0.143833 Objective Loss 0.143833 LR 0.000250 Time 0.022259 +2023-10-02 21:38:33,212 - Epoch: [144][ 700/ 1236] Overall Loss 0.143655 Objective Loss 0.143655 LR 0.000250 Time 0.022239 +2023-10-02 21:38:33,423 - Epoch: [144][ 710/ 1236] Overall Loss 0.143334 Objective Loss 0.143334 LR 0.000250 Time 0.022223 +2023-10-02 21:38:33,635 - Epoch: [144][ 720/ 1236] Overall Loss 0.143158 Objective Loss 0.143158 LR 0.000250 Time 0.022206 +2023-10-02 21:38:33,848 - Epoch: [144][ 730/ 1236] Overall Loss 0.143318 Objective Loss 0.143318 LR 0.000250 Time 0.022191 +2023-10-02 21:38:34,060 - Epoch: [144][ 740/ 1236] Overall Loss 0.143306 Objective Loss 0.143306 LR 0.000250 Time 0.022176 +2023-10-02 21:38:34,272 - Epoch: [144][ 750/ 1236] Overall Loss 0.143361 Objective Loss 0.143361 LR 0.000250 Time 0.022162 +2023-10-02 21:38:34,484 - Epoch: [144][ 760/ 1236] Overall Loss 0.143034 Objective Loss 0.143034 LR 0.000250 Time 0.022148 +2023-10-02 21:38:34,697 - Epoch: [144][ 770/ 1236] Overall Loss 0.142955 Objective Loss 0.142955 LR 0.000250 Time 0.022134 +2023-10-02 21:38:34,908 - Epoch: [144][ 780/ 1236] Overall Loss 0.143134 Objective Loss 0.143134 LR 0.000250 Time 0.022119 +2023-10-02 21:38:35,121 - Epoch: [144][ 790/ 1236] Overall Loss 0.143225 Objective Loss 0.143225 LR 0.000250 Time 0.022107 +2023-10-02 21:38:35,333 - Epoch: [144][ 800/ 1236] Overall Loss 0.143139 Objective Loss 0.143139 LR 0.000250 Time 0.022095 +2023-10-02 21:38:35,546 - Epoch: [144][ 810/ 1236] Overall Loss 0.143368 Objective Loss 0.143368 LR 0.000250 Time 0.022085 +2023-10-02 21:38:35,757 - Epoch: [144][ 820/ 1236] Overall Loss 0.143117 Objective Loss 0.143117 LR 0.000250 Time 0.022072 +2023-10-02 21:38:35,969 - Epoch: [144][ 830/ 1236] Overall Loss 0.143428 Objective Loss 0.143428 LR 0.000250 Time 0.022061 +2023-10-02 21:38:36,181 - Epoch: [144][ 840/ 1236] Overall Loss 0.143517 Objective Loss 0.143517 LR 0.000250 Time 0.022048 +2023-10-02 21:38:36,394 - Epoch: [144][ 850/ 1236] Overall Loss 0.143478 Objective Loss 0.143478 LR 0.000250 Time 0.022039 +2023-10-02 21:38:36,603 - Epoch: [144][ 860/ 1236] Overall Loss 0.143753 Objective Loss 0.143753 LR 0.000250 Time 0.022025 +2023-10-02 21:38:36,815 - Epoch: [144][ 870/ 1236] Overall Loss 0.143867 Objective Loss 0.143867 LR 0.000250 Time 0.022015 +2023-10-02 21:38:37,023 - Epoch: [144][ 880/ 1236] Overall Loss 0.144108 Objective Loss 0.144108 LR 0.000250 Time 0.022002 +2023-10-02 21:38:37,234 - Epoch: [144][ 890/ 1236] Overall Loss 0.144384 Objective Loss 0.144384 LR 0.000250 Time 0.021991 +2023-10-02 21:38:37,444 - Epoch: [144][ 900/ 1236] Overall Loss 0.144511 Objective Loss 0.144511 LR 0.000250 Time 0.021978 +2023-10-02 21:38:37,653 - Epoch: [144][ 910/ 1236] Overall Loss 0.144749 Objective Loss 0.144749 LR 0.000250 Time 0.021965 +2023-10-02 21:38:37,861 - Epoch: [144][ 920/ 1236] Overall Loss 0.144570 Objective Loss 0.144570 LR 0.000250 Time 0.021952 +2023-10-02 21:38:38,070 - Epoch: [144][ 930/ 1236] Overall Loss 0.144513 Objective Loss 0.144513 LR 0.000250 Time 0.021940 +2023-10-02 21:38:38,277 - Epoch: [144][ 940/ 1236] Overall Loss 0.144459 Objective Loss 0.144459 LR 0.000250 Time 0.021927 +2023-10-02 21:38:38,486 - Epoch: [144][ 950/ 1236] Overall Loss 0.144619 Objective Loss 0.144619 LR 0.000250 Time 0.021916 +2023-10-02 21:38:38,694 - Epoch: [144][ 960/ 1236] Overall Loss 0.144614 Objective Loss 0.144614 LR 0.000250 Time 0.021904 +2023-10-02 21:38:38,903 - Epoch: [144][ 970/ 1236] Overall Loss 0.144740 Objective Loss 0.144740 LR 0.000250 Time 0.021893 +2023-10-02 21:38:39,111 - Epoch: [144][ 980/ 1236] Overall Loss 0.144872 Objective Loss 0.144872 LR 0.000250 Time 0.021881 +2023-10-02 21:38:39,320 - Epoch: [144][ 990/ 1236] Overall Loss 0.144763 Objective Loss 0.144763 LR 0.000250 Time 0.021871 +2023-10-02 21:38:39,528 - Epoch: [144][ 1000/ 1236] Overall Loss 0.144697 Objective Loss 0.144697 LR 0.000250 Time 0.021860 +2023-10-02 21:38:39,737 - Epoch: [144][ 1010/ 1236] Overall Loss 0.144612 Objective Loss 0.144612 LR 0.000250 Time 0.021850 +2023-10-02 21:38:39,944 - Epoch: [144][ 1020/ 1236] Overall Loss 0.144825 Objective Loss 0.144825 LR 0.000250 Time 0.021839 +2023-10-02 21:38:40,153 - Epoch: [144][ 1030/ 1236] Overall Loss 0.144749 Objective Loss 0.144749 LR 0.000250 Time 0.021830 +2023-10-02 21:38:40,361 - Epoch: [144][ 1040/ 1236] Overall Loss 0.144845 Objective Loss 0.144845 LR 0.000250 Time 0.021819 +2023-10-02 21:38:40,570 - Epoch: [144][ 1050/ 1236] Overall Loss 0.144795 Objective Loss 0.144795 LR 0.000250 Time 0.021810 +2023-10-02 21:38:40,778 - Epoch: [144][ 1060/ 1236] Overall Loss 0.144671 Objective Loss 0.144671 LR 0.000250 Time 0.021800 +2023-10-02 21:38:40,987 - Epoch: [144][ 1070/ 1236] Overall Loss 0.144639 Objective Loss 0.144639 LR 0.000250 Time 0.021791 +2023-10-02 21:38:41,195 - Epoch: [144][ 1080/ 1236] Overall Loss 0.144548 Objective Loss 0.144548 LR 0.000250 Time 0.021782 +2023-10-02 21:38:41,404 - Epoch: [144][ 1090/ 1236] Overall Loss 0.144703 Objective Loss 0.144703 LR 0.000250 Time 0.021773 +2023-10-02 21:38:41,612 - Epoch: [144][ 1100/ 1236] Overall Loss 0.144486 Objective Loss 0.144486 LR 0.000250 Time 0.021764 +2023-10-02 21:38:41,821 - Epoch: [144][ 1110/ 1236] Overall Loss 0.144520 Objective Loss 0.144520 LR 0.000250 Time 0.021756 +2023-10-02 21:38:42,028 - Epoch: [144][ 1120/ 1236] Overall Loss 0.144477 Objective Loss 0.144477 LR 0.000250 Time 0.021747 +2023-10-02 21:38:42,238 - Epoch: [144][ 1130/ 1236] Overall Loss 0.144453 Objective Loss 0.144453 LR 0.000250 Time 0.021740 +2023-10-02 21:38:42,446 - Epoch: [144][ 1140/ 1236] Overall Loss 0.144519 Objective Loss 0.144519 LR 0.000250 Time 0.021731 +2023-10-02 21:38:42,655 - Epoch: [144][ 1150/ 1236] Overall Loss 0.144607 Objective Loss 0.144607 LR 0.000250 Time 0.021724 +2023-10-02 21:38:42,863 - Epoch: [144][ 1160/ 1236] Overall Loss 0.144715 Objective Loss 0.144715 LR 0.000250 Time 0.021715 +2023-10-02 21:38:43,072 - Epoch: [144][ 1170/ 1236] Overall Loss 0.144866 Objective Loss 0.144866 LR 0.000250 Time 0.021708 +2023-10-02 21:38:43,280 - Epoch: [144][ 1180/ 1236] Overall Loss 0.144984 Objective Loss 0.144984 LR 0.000250 Time 0.021700 +2023-10-02 21:38:43,489 - Epoch: [144][ 1190/ 1236] Overall Loss 0.144944 Objective Loss 0.144944 LR 0.000250 Time 0.021693 +2023-10-02 21:38:43,696 - Epoch: [144][ 1200/ 1236] Overall Loss 0.144973 Objective Loss 0.144973 LR 0.000250 Time 0.021685 +2023-10-02 21:38:43,906 - Epoch: [144][ 1210/ 1236] Overall Loss 0.144976 Objective Loss 0.144976 LR 0.000250 Time 0.021678 +2023-10-02 21:38:44,113 - Epoch: [144][ 1220/ 1236] Overall Loss 0.144983 Objective Loss 0.144983 LR 0.000250 Time 0.021671 +2023-10-02 21:38:44,376 - Epoch: [144][ 1230/ 1236] Overall Loss 0.145016 Objective Loss 0.145016 LR 0.000250 Time 0.021708 +2023-10-02 21:38:44,499 - Epoch: [144][ 1236/ 1236] Overall Loss 0.145030 Objective Loss 0.145030 Top1 91.242363 Top5 98.981670 LR 0.000250 Time 0.021702 +2023-10-02 21:38:44,622 - --- validate (epoch=144)----------- +2023-10-02 21:38:44,623 - 29943 samples (256 per mini-batch) +2023-10-02 21:38:45,127 - Epoch: [144][ 10/ 117] Loss 0.266527 Top1 86.640625 Top5 98.710938 +2023-10-02 21:38:45,280 - Epoch: [144][ 20/ 117] Loss 0.289979 Top1 86.210938 Top5 98.691406 +2023-10-02 21:38:45,433 - Epoch: [144][ 30/ 117] Loss 0.291309 Top1 86.354167 Top5 98.697917 +2023-10-02 21:38:45,585 - Epoch: [144][ 40/ 117] Loss 0.289128 Top1 86.259766 Top5 98.632812 +2023-10-02 21:38:45,737 - Epoch: [144][ 50/ 117] Loss 0.288752 Top1 86.500000 Top5 98.609375 +2023-10-02 21:38:45,889 - Epoch: [144][ 60/ 117] Loss 0.290267 Top1 86.562500 Top5 98.606771 +2023-10-02 21:38:46,040 - Epoch: [144][ 70/ 117] Loss 0.291468 Top1 86.506696 Top5 98.577009 +2023-10-02 21:38:46,193 - Epoch: [144][ 80/ 117] Loss 0.289022 Top1 86.630859 Top5 98.583984 +2023-10-02 21:38:46,344 - Epoch: [144][ 90/ 117] Loss 0.290546 Top1 86.584201 Top5 98.580729 +2023-10-02 21:38:46,497 - Epoch: [144][ 100/ 117] Loss 0.291262 Top1 86.582031 Top5 98.605469 +2023-10-02 21:38:46,655 - Epoch: [144][ 110/ 117] Loss 0.293350 Top1 86.594460 Top5 98.579545 +2023-10-02 21:38:46,745 - Epoch: [144][ 117/ 117] Loss 0.295650 Top1 86.607888 Top5 98.563938 +2023-10-02 21:38:46,881 - ==> Top1: 86.608 Top5: 98.564 Loss: 0.296 + +2023-10-02 21:38:46,882 - ==> Confusion: +[[ 960 0 2 0 10 1 0 0 7 43 3 0 2 0 4 1 2 0 0 0 15] + [ 0 1061 0 0 7 18 2 22 3 0 0 0 1 0 1 3 0 0 6 3 4] + [ 2 0 987 4 1 0 14 9 0 2 1 0 7 2 1 2 2 2 9 3 8] + [ 1 5 13 988 1 1 2 3 2 1 5 0 3 2 25 2 1 4 12 1 17] + [ 26 2 0 0 971 4 1 0 2 11 0 0 2 1 8 3 10 0 1 3 5] + [ 4 37 1 2 5 991 0 25 0 4 1 6 2 8 4 0 4 0 3 3 16] + [ 0 3 24 0 0 1 1135 4 0 0 4 1 0 0 0 3 0 0 2 9 5] + [ 3 13 13 1 5 20 3 1080 0 3 5 5 4 4 3 0 0 1 36 12 7] + [ 17 0 0 1 3 3 0 1 977 42 10 2 2 12 11 0 2 0 2 1 3] + [ 93 1 2 0 7 5 0 0 21 948 1 1 0 18 10 3 0 0 0 0 9] + [ 5 2 10 7 0 1 6 2 11 0 978 1 0 8 3 0 3 1 4 0 11] + [ 0 1 0 0 0 12 0 3 0 0 0 963 25 8 0 4 0 13 0 3 3] + [ 0 3 1 2 1 1 1 0 0 0 3 32 983 0 4 10 1 7 3 7 9] + [ 1 0 0 0 3 8 0 0 10 10 4 8 0 1055 3 0 1 1 0 4 11] + [ 13 0 7 13 4 0 0 0 20 1 5 0 2 2 1013 0 1 3 10 0 7] + [ 0 0 1 0 5 0 0 0 0 1 1 4 10 0 0 1074 16 8 3 6 5] + [ 0 17 2 0 4 5 0 0 0 0 0 5 0 3 3 7 1097 0 1 7 10] + [ 0 0 1 0 1 0 3 0 0 0 0 5 21 1 1 7 1 992 0 1 4] + [ 3 3 3 18 0 0 0 17 7 0 1 1 2 0 7 0 0 0 994 1 11] + [ 0 0 3 1 1 1 11 3 0 0 1 12 5 1 0 2 5 0 1 1097 8] + [ 123 124 112 70 76 134 32 81 72 72 168 86 336 238 126 50 105 45 106 160 5589]] + +2023-10-02 21:38:46,883 - ==> Best [Top1: 86.875 Top5: 98.567 Sparsity:0.00 Params: 169472 on epoch: 142] +2023-10-02 21:38:46,883 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:38:46,889 - + +2023-10-02 21:38:46,889 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:38:47,930 - Epoch: [145][ 10/ 1236] Overall Loss 0.145159 Objective Loss 0.145159 LR 0.000250 Time 0.103965 +2023-10-02 21:38:48,137 - Epoch: [145][ 20/ 1236] Overall Loss 0.149477 Objective Loss 0.149477 LR 0.000250 Time 0.062332 +2023-10-02 21:38:48,345 - Epoch: [145][ 30/ 1236] Overall Loss 0.150215 Objective Loss 0.150215 LR 0.000250 Time 0.048425 +2023-10-02 21:38:48,553 - Epoch: [145][ 40/ 1236] Overall Loss 0.145001 Objective Loss 0.145001 LR 0.000250 Time 0.041521 +2023-10-02 21:38:48,761 - Epoch: [145][ 50/ 1236] Overall Loss 0.147707 Objective Loss 0.147707 LR 0.000250 Time 0.037345 +2023-10-02 21:38:48,970 - Epoch: [145][ 60/ 1236] Overall Loss 0.148004 Objective Loss 0.148004 LR 0.000250 Time 0.034594 +2023-10-02 21:38:49,177 - Epoch: [145][ 70/ 1236] Overall Loss 0.145637 Objective Loss 0.145637 LR 0.000250 Time 0.032595 +2023-10-02 21:38:49,387 - Epoch: [145][ 80/ 1236] Overall Loss 0.144412 Objective Loss 0.144412 LR 0.000250 Time 0.031135 +2023-10-02 21:38:49,593 - Epoch: [145][ 90/ 1236] Overall Loss 0.145279 Objective Loss 0.145279 LR 0.000250 Time 0.029966 +2023-10-02 21:38:49,803 - Epoch: [145][ 100/ 1236] Overall Loss 0.145453 Objective Loss 0.145453 LR 0.000250 Time 0.029063 +2023-10-02 21:38:50,009 - Epoch: [145][ 110/ 1236] Overall Loss 0.143351 Objective Loss 0.143351 LR 0.000250 Time 0.028293 +2023-10-02 21:38:50,217 - Epoch: [145][ 120/ 1236] Overall Loss 0.145446 Objective Loss 0.145446 LR 0.000250 Time 0.027667 +2023-10-02 21:38:50,424 - Epoch: [145][ 130/ 1236] Overall Loss 0.146693 Objective Loss 0.146693 LR 0.000250 Time 0.027120 +2023-10-02 21:38:50,632 - Epoch: [145][ 140/ 1236] Overall Loss 0.145463 Objective Loss 0.145463 LR 0.000250 Time 0.026669 +2023-10-02 21:38:50,839 - Epoch: [145][ 150/ 1236] Overall Loss 0.146598 Objective Loss 0.146598 LR 0.000250 Time 0.026261 +2023-10-02 21:38:51,048 - Epoch: [145][ 160/ 1236] Overall Loss 0.145359 Objective Loss 0.145359 LR 0.000250 Time 0.025923 +2023-10-02 21:38:51,255 - Epoch: [145][ 170/ 1236] Overall Loss 0.146382 Objective Loss 0.146382 LR 0.000250 Time 0.025605 +2023-10-02 21:38:51,464 - Epoch: [145][ 180/ 1236] Overall Loss 0.146615 Objective Loss 0.146615 LR 0.000250 Time 0.025339 +2023-10-02 21:38:51,671 - Epoch: [145][ 190/ 1236] Overall Loss 0.146617 Objective Loss 0.146617 LR 0.000250 Time 0.025089 +2023-10-02 21:38:51,880 - Epoch: [145][ 200/ 1236] Overall Loss 0.147319 Objective Loss 0.147319 LR 0.000250 Time 0.024880 +2023-10-02 21:38:52,087 - Epoch: [145][ 210/ 1236] Overall Loss 0.147377 Objective Loss 0.147377 LR 0.000250 Time 0.024679 +2023-10-02 21:38:52,296 - Epoch: [145][ 220/ 1236] Overall Loss 0.147467 Objective Loss 0.147467 LR 0.000250 Time 0.024507 +2023-10-02 21:38:52,504 - Epoch: [145][ 230/ 1236] Overall Loss 0.147196 Objective Loss 0.147196 LR 0.000250 Time 0.024343 +2023-10-02 21:38:52,712 - Epoch: [145][ 240/ 1236] Overall Loss 0.147639 Objective Loss 0.147639 LR 0.000250 Time 0.024194 +2023-10-02 21:38:52,922 - Epoch: [145][ 250/ 1236] Overall Loss 0.147040 Objective Loss 0.147040 LR 0.000250 Time 0.024064 +2023-10-02 21:38:53,130 - Epoch: [145][ 260/ 1236] Overall Loss 0.146509 Objective Loss 0.146509 LR 0.000250 Time 0.023940 +2023-10-02 21:38:53,338 - Epoch: [145][ 270/ 1236] Overall Loss 0.146495 Objective Loss 0.146495 LR 0.000250 Time 0.023821 +2023-10-02 21:38:53,545 - Epoch: [145][ 280/ 1236] Overall Loss 0.145105 Objective Loss 0.145105 LR 0.000250 Time 0.023709 +2023-10-02 21:38:53,753 - Epoch: [145][ 290/ 1236] Overall Loss 0.145384 Objective Loss 0.145384 LR 0.000250 Time 0.023606 +2023-10-02 21:38:53,960 - Epoch: [145][ 300/ 1236] Overall Loss 0.145096 Objective Loss 0.145096 LR 0.000250 Time 0.023508 +2023-10-02 21:38:54,167 - Epoch: [145][ 310/ 1236] Overall Loss 0.144898 Objective Loss 0.144898 LR 0.000250 Time 0.023419 +2023-10-02 21:38:54,374 - Epoch: [145][ 320/ 1236] Overall Loss 0.144562 Objective Loss 0.144562 LR 0.000250 Time 0.023332 +2023-10-02 21:38:54,582 - Epoch: [145][ 330/ 1236] Overall Loss 0.145057 Objective Loss 0.145057 LR 0.000250 Time 0.023255 +2023-10-02 21:38:54,789 - Epoch: [145][ 340/ 1236] Overall Loss 0.145436 Objective Loss 0.145436 LR 0.000250 Time 0.023177 +2023-10-02 21:38:54,997 - Epoch: [145][ 350/ 1236] Overall Loss 0.145926 Objective Loss 0.145926 LR 0.000250 Time 0.023109 +2023-10-02 21:38:55,203 - Epoch: [145][ 360/ 1236] Overall Loss 0.145775 Objective Loss 0.145775 LR 0.000250 Time 0.023040 +2023-10-02 21:38:55,412 - Epoch: [145][ 370/ 1236] Overall Loss 0.145507 Objective Loss 0.145507 LR 0.000250 Time 0.022979 +2023-10-02 21:38:55,619 - Epoch: [145][ 380/ 1236] Overall Loss 0.145359 Objective Loss 0.145359 LR 0.000250 Time 0.022919 +2023-10-02 21:38:55,827 - Epoch: [145][ 390/ 1236] Overall Loss 0.145008 Objective Loss 0.145008 LR 0.000250 Time 0.022865 +2023-10-02 21:38:56,034 - Epoch: [145][ 400/ 1236] Overall Loss 0.145049 Objective Loss 0.145049 LR 0.000250 Time 0.022810 +2023-10-02 21:38:56,243 - Epoch: [145][ 410/ 1236] Overall Loss 0.145385 Objective Loss 0.145385 LR 0.000250 Time 0.022761 +2023-10-02 21:38:56,449 - Epoch: [145][ 420/ 1236] Overall Loss 0.145382 Objective Loss 0.145382 LR 0.000250 Time 0.022710 +2023-10-02 21:38:56,658 - Epoch: [145][ 430/ 1236] Overall Loss 0.145246 Objective Loss 0.145246 LR 0.000250 Time 0.022667 +2023-10-02 21:38:56,865 - Epoch: [145][ 440/ 1236] Overall Loss 0.145300 Objective Loss 0.145300 LR 0.000250 Time 0.022621 +2023-10-02 21:38:57,073 - Epoch: [145][ 450/ 1236] Overall Loss 0.145388 Objective Loss 0.145388 LR 0.000250 Time 0.022580 +2023-10-02 21:38:57,280 - Epoch: [145][ 460/ 1236] Overall Loss 0.145245 Objective Loss 0.145245 LR 0.000250 Time 0.022538 +2023-10-02 21:38:57,488 - Epoch: [145][ 470/ 1236] Overall Loss 0.145080 Objective Loss 0.145080 LR 0.000250 Time 0.022501 +2023-10-02 21:38:57,695 - Epoch: [145][ 480/ 1236] Overall Loss 0.145197 Objective Loss 0.145197 LR 0.000250 Time 0.022462 +2023-10-02 21:38:57,903 - Epoch: [145][ 490/ 1236] Overall Loss 0.145011 Objective Loss 0.145011 LR 0.000250 Time 0.022428 +2023-10-02 21:38:58,110 - Epoch: [145][ 500/ 1236] Overall Loss 0.145015 Objective Loss 0.145015 LR 0.000250 Time 0.022392 +2023-10-02 21:38:58,318 - Epoch: [145][ 510/ 1236] Overall Loss 0.144908 Objective Loss 0.144908 LR 0.000250 Time 0.022362 +2023-10-02 21:38:58,526 - Epoch: [145][ 520/ 1236] Overall Loss 0.144532 Objective Loss 0.144532 LR 0.000250 Time 0.022330 +2023-10-02 21:38:58,734 - Epoch: [145][ 530/ 1236] Overall Loss 0.144568 Objective Loss 0.144568 LR 0.000250 Time 0.022301 +2023-10-02 21:38:58,941 - Epoch: [145][ 540/ 1236] Overall Loss 0.144751 Objective Loss 0.144751 LR 0.000250 Time 0.022270 +2023-10-02 21:38:59,149 - Epoch: [145][ 550/ 1236] Overall Loss 0.144579 Objective Loss 0.144579 LR 0.000250 Time 0.022244 +2023-10-02 21:38:59,356 - Epoch: [145][ 560/ 1236] Overall Loss 0.144318 Objective Loss 0.144318 LR 0.000250 Time 0.022216 +2023-10-02 21:38:59,565 - Epoch: [145][ 570/ 1236] Overall Loss 0.144450 Objective Loss 0.144450 LR 0.000250 Time 0.022192 +2023-10-02 21:38:59,772 - Epoch: [145][ 580/ 1236] Overall Loss 0.144243 Objective Loss 0.144243 LR 0.000250 Time 0.022166 +2023-10-02 21:38:59,981 - Epoch: [145][ 590/ 1236] Overall Loss 0.144612 Objective Loss 0.144612 LR 0.000250 Time 0.022143 +2023-10-02 21:39:00,188 - Epoch: [145][ 600/ 1236] Overall Loss 0.144635 Objective Loss 0.144635 LR 0.000250 Time 0.022118 +2023-10-02 21:39:00,396 - Epoch: [145][ 610/ 1236] Overall Loss 0.144521 Objective Loss 0.144521 LR 0.000250 Time 0.022097 +2023-10-02 21:39:00,603 - Epoch: [145][ 620/ 1236] Overall Loss 0.144240 Objective Loss 0.144240 LR 0.000250 Time 0.022074 +2023-10-02 21:39:00,812 - Epoch: [145][ 630/ 1236] Overall Loss 0.144309 Objective Loss 0.144309 LR 0.000250 Time 0.022055 +2023-10-02 21:39:01,019 - Epoch: [145][ 640/ 1236] Overall Loss 0.144301 Objective Loss 0.144301 LR 0.000250 Time 0.022033 +2023-10-02 21:39:01,227 - Epoch: [145][ 650/ 1236] Overall Loss 0.144155 Objective Loss 0.144155 LR 0.000250 Time 0.022014 +2023-10-02 21:39:01,434 - Epoch: [145][ 660/ 1236] Overall Loss 0.143939 Objective Loss 0.143939 LR 0.000250 Time 0.021993 +2023-10-02 21:39:01,642 - Epoch: [145][ 670/ 1236] Overall Loss 0.143815 Objective Loss 0.143815 LR 0.000250 Time 0.021976 +2023-10-02 21:39:01,850 - Epoch: [145][ 680/ 1236] Overall Loss 0.143702 Objective Loss 0.143702 LR 0.000250 Time 0.021957 +2023-10-02 21:39:02,058 - Epoch: [145][ 690/ 1236] Overall Loss 0.143603 Objective Loss 0.143603 LR 0.000250 Time 0.021940 +2023-10-02 21:39:02,265 - Epoch: [145][ 700/ 1236] Overall Loss 0.143913 Objective Loss 0.143913 LR 0.000250 Time 0.021921 +2023-10-02 21:39:02,473 - Epoch: [145][ 710/ 1236] Overall Loss 0.143989 Objective Loss 0.143989 LR 0.000250 Time 0.021906 +2023-10-02 21:39:02,681 - Epoch: [145][ 720/ 1236] Overall Loss 0.143906 Objective Loss 0.143906 LR 0.000250 Time 0.021889 +2023-10-02 21:39:02,889 - Epoch: [145][ 730/ 1236] Overall Loss 0.144021 Objective Loss 0.144021 LR 0.000250 Time 0.021875 +2023-10-02 21:39:03,096 - Epoch: [145][ 740/ 1236] Overall Loss 0.143938 Objective Loss 0.143938 LR 0.000250 Time 0.021858 +2023-10-02 21:39:03,305 - Epoch: [145][ 750/ 1236] Overall Loss 0.143854 Objective Loss 0.143854 LR 0.000250 Time 0.021844 +2023-10-02 21:39:03,512 - Epoch: [145][ 760/ 1236] Overall Loss 0.143738 Objective Loss 0.143738 LR 0.000250 Time 0.021829 +2023-10-02 21:39:03,720 - Epoch: [145][ 770/ 1236] Overall Loss 0.143769 Objective Loss 0.143769 LR 0.000250 Time 0.021816 +2023-10-02 21:39:03,927 - Epoch: [145][ 780/ 1236] Overall Loss 0.143668 Objective Loss 0.143668 LR 0.000250 Time 0.021801 +2023-10-02 21:39:04,136 - Epoch: [145][ 790/ 1236] Overall Loss 0.143656 Objective Loss 0.143656 LR 0.000250 Time 0.021789 +2023-10-02 21:39:04,343 - Epoch: [145][ 800/ 1236] Overall Loss 0.143784 Objective Loss 0.143784 LR 0.000250 Time 0.021775 +2023-10-02 21:39:04,551 - Epoch: [145][ 810/ 1236] Overall Loss 0.143717 Objective Loss 0.143717 LR 0.000250 Time 0.021763 +2023-10-02 21:39:04,759 - Epoch: [145][ 820/ 1236] Overall Loss 0.143812 Objective Loss 0.143812 LR 0.000250 Time 0.021750 +2023-10-02 21:39:04,967 - Epoch: [145][ 830/ 1236] Overall Loss 0.143737 Objective Loss 0.143737 LR 0.000250 Time 0.021739 +2023-10-02 21:39:05,174 - Epoch: [145][ 840/ 1236] Overall Loss 0.143871 Objective Loss 0.143871 LR 0.000250 Time 0.021726 +2023-10-02 21:39:05,383 - Epoch: [145][ 850/ 1236] Overall Loss 0.143743 Objective Loss 0.143743 LR 0.000250 Time 0.021716 +2023-10-02 21:39:05,591 - Epoch: [145][ 860/ 1236] Overall Loss 0.144036 Objective Loss 0.144036 LR 0.000250 Time 0.021704 +2023-10-02 21:39:05,799 - Epoch: [145][ 870/ 1236] Overall Loss 0.144051 Objective Loss 0.144051 LR 0.000250 Time 0.021694 +2023-10-02 21:39:06,007 - Epoch: [145][ 880/ 1236] Overall Loss 0.144109 Objective Loss 0.144109 LR 0.000250 Time 0.021683 +2023-10-02 21:39:06,215 - Epoch: [145][ 890/ 1236] Overall Loss 0.144233 Objective Loss 0.144233 LR 0.000250 Time 0.021674 +2023-10-02 21:39:06,423 - Epoch: [145][ 900/ 1236] Overall Loss 0.144299 Objective Loss 0.144299 LR 0.000250 Time 0.021663 +2023-10-02 21:39:06,632 - Epoch: [145][ 910/ 1236] Overall Loss 0.144220 Objective Loss 0.144220 LR 0.000250 Time 0.021654 +2023-10-02 21:39:06,839 - Epoch: [145][ 920/ 1236] Overall Loss 0.144238 Objective Loss 0.144238 LR 0.000250 Time 0.021644 +2023-10-02 21:39:07,047 - Epoch: [145][ 930/ 1236] Overall Loss 0.144013 Objective Loss 0.144013 LR 0.000250 Time 0.021635 +2023-10-02 21:39:07,255 - Epoch: [145][ 940/ 1236] Overall Loss 0.143977 Objective Loss 0.143977 LR 0.000250 Time 0.021625 +2023-10-02 21:39:07,463 - Epoch: [145][ 950/ 1236] Overall Loss 0.143955 Objective Loss 0.143955 LR 0.000250 Time 0.021616 +2023-10-02 21:39:07,671 - Epoch: [145][ 960/ 1236] Overall Loss 0.143815 Objective Loss 0.143815 LR 0.000250 Time 0.021607 +2023-10-02 21:39:07,880 - Epoch: [145][ 970/ 1236] Overall Loss 0.143795 Objective Loss 0.143795 LR 0.000250 Time 0.021599 +2023-10-02 21:39:08,087 - Epoch: [145][ 980/ 1236] Overall Loss 0.143809 Objective Loss 0.143809 LR 0.000250 Time 0.021590 +2023-10-02 21:39:08,296 - Epoch: [145][ 990/ 1236] Overall Loss 0.143727 Objective Loss 0.143727 LR 0.000250 Time 0.021583 +2023-10-02 21:39:08,503 - Epoch: [145][ 1000/ 1236] Overall Loss 0.143687 Objective Loss 0.143687 LR 0.000250 Time 0.021574 +2023-10-02 21:39:08,712 - Epoch: [145][ 1010/ 1236] Overall Loss 0.143570 Objective Loss 0.143570 LR 0.000250 Time 0.021567 +2023-10-02 21:39:08,920 - Epoch: [145][ 1020/ 1236] Overall Loss 0.143427 Objective Loss 0.143427 LR 0.000250 Time 0.021559 +2023-10-02 21:39:09,129 - Epoch: [145][ 1030/ 1236] Overall Loss 0.143353 Objective Loss 0.143353 LR 0.000250 Time 0.021552 +2023-10-02 21:39:09,336 - Epoch: [145][ 1040/ 1236] Overall Loss 0.143358 Objective Loss 0.143358 LR 0.000250 Time 0.021544 +2023-10-02 21:39:09,545 - Epoch: [145][ 1050/ 1236] Overall Loss 0.143538 Objective Loss 0.143538 LR 0.000250 Time 0.021537 +2023-10-02 21:39:09,753 - Epoch: [145][ 1060/ 1236] Overall Loss 0.143591 Objective Loss 0.143591 LR 0.000250 Time 0.021530 +2023-10-02 21:39:09,962 - Epoch: [145][ 1070/ 1236] Overall Loss 0.143823 Objective Loss 0.143823 LR 0.000250 Time 0.021524 +2023-10-02 21:39:10,170 - Epoch: [145][ 1080/ 1236] Overall Loss 0.143916 Objective Loss 0.143916 LR 0.000250 Time 0.021517 +2023-10-02 21:39:10,379 - Epoch: [145][ 1090/ 1236] Overall Loss 0.143874 Objective Loss 0.143874 LR 0.000250 Time 0.021511 +2023-10-02 21:39:10,587 - Epoch: [145][ 1100/ 1236] Overall Loss 0.143794 Objective Loss 0.143794 LR 0.000250 Time 0.021504 +2023-10-02 21:39:10,796 - Epoch: [145][ 1110/ 1236] Overall Loss 0.143942 Objective Loss 0.143942 LR 0.000250 Time 0.021498 +2023-10-02 21:39:11,004 - Epoch: [145][ 1120/ 1236] Overall Loss 0.143721 Objective Loss 0.143721 LR 0.000250 Time 0.021492 +2023-10-02 21:39:11,213 - Epoch: [145][ 1130/ 1236] Overall Loss 0.143729 Objective Loss 0.143729 LR 0.000250 Time 0.021486 +2023-10-02 21:39:11,420 - Epoch: [145][ 1140/ 1236] Overall Loss 0.143554 Objective Loss 0.143554 LR 0.000250 Time 0.021479 +2023-10-02 21:39:11,628 - Epoch: [145][ 1150/ 1236] Overall Loss 0.143598 Objective Loss 0.143598 LR 0.000250 Time 0.021473 +2023-10-02 21:39:11,835 - Epoch: [145][ 1160/ 1236] Overall Loss 0.143402 Objective Loss 0.143402 LR 0.000250 Time 0.021466 +2023-10-02 21:39:12,043 - Epoch: [145][ 1170/ 1236] Overall Loss 0.143432 Objective Loss 0.143432 LR 0.000250 Time 0.021460 +2023-10-02 21:39:12,249 - Epoch: [145][ 1180/ 1236] Overall Loss 0.143523 Objective Loss 0.143523 LR 0.000250 Time 0.021453 +2023-10-02 21:39:12,457 - Epoch: [145][ 1190/ 1236] Overall Loss 0.143635 Objective Loss 0.143635 LR 0.000250 Time 0.021447 +2023-10-02 21:39:12,664 - Epoch: [145][ 1200/ 1236] Overall Loss 0.143505 Objective Loss 0.143505 LR 0.000250 Time 0.021440 +2023-10-02 21:39:12,872 - Epoch: [145][ 1210/ 1236] Overall Loss 0.143368 Objective Loss 0.143368 LR 0.000250 Time 0.021435 +2023-10-02 21:39:13,078 - Epoch: [145][ 1220/ 1236] Overall Loss 0.143327 Objective Loss 0.143327 LR 0.000250 Time 0.021428 +2023-10-02 21:39:13,336 - Epoch: [145][ 1230/ 1236] Overall Loss 0.143293 Objective Loss 0.143293 LR 0.000250 Time 0.021463 +2023-10-02 21:39:13,457 - Epoch: [145][ 1236/ 1236] Overall Loss 0.143388 Objective Loss 0.143388 Top1 90.631365 Top5 99.185336 LR 0.000250 Time 0.021456 +2023-10-02 21:39:13,603 - --- validate (epoch=145)----------- +2023-10-02 21:39:13,603 - 29943 samples (256 per mini-batch) +2023-10-02 21:39:14,098 - Epoch: [145][ 10/ 117] Loss 0.314677 Top1 86.367188 Top5 98.046875 +2023-10-02 21:39:14,249 - Epoch: [145][ 20/ 117] Loss 0.296708 Top1 86.601562 Top5 98.417969 +2023-10-02 21:39:14,399 - Epoch: [145][ 30/ 117] Loss 0.290487 Top1 86.835938 Top5 98.541667 +2023-10-02 21:39:14,548 - Epoch: [145][ 40/ 117] Loss 0.289252 Top1 86.796875 Top5 98.662109 +2023-10-02 21:39:14,699 - Epoch: [145][ 50/ 117] Loss 0.288183 Top1 86.914062 Top5 98.656250 +2023-10-02 21:39:14,849 - Epoch: [145][ 60/ 117] Loss 0.287024 Top1 86.875000 Top5 98.697917 +2023-10-02 21:39:15,000 - Epoch: [145][ 70/ 117] Loss 0.293310 Top1 86.729911 Top5 98.638393 +2023-10-02 21:39:15,150 - Epoch: [145][ 80/ 117] Loss 0.293297 Top1 86.870117 Top5 98.657227 +2023-10-02 21:39:15,300 - Epoch: [145][ 90/ 117] Loss 0.291925 Top1 86.870660 Top5 98.680556 +2023-10-02 21:39:15,449 - Epoch: [145][ 100/ 117] Loss 0.290365 Top1 86.941406 Top5 98.679688 +2023-10-02 21:39:15,606 - Epoch: [145][ 110/ 117] Loss 0.292129 Top1 87.002841 Top5 98.689631 +2023-10-02 21:39:15,695 - Epoch: [145][ 117/ 117] Loss 0.292449 Top1 87.015329 Top5 98.664129 +2023-10-02 21:39:15,841 - ==> Top1: 87.015 Top5: 98.664 Loss: 0.292 + +2023-10-02 21:39:15,842 - ==> Confusion: +[[ 951 0 0 2 5 4 0 1 8 55 2 0 0 0 4 0 2 0 0 0 16] + [ 0 1054 0 0 4 23 0 25 3 1 1 1 0 0 0 3 1 0 9 3 3] + [ 3 1 987 6 0 0 7 8 0 2 3 0 6 2 0 3 1 2 12 3 10] + [ 2 3 12 991 1 1 0 1 3 0 5 0 6 2 24 1 1 2 14 0 20] + [ 36 4 0 0 954 7 0 0 3 7 0 0 0 3 12 5 9 0 0 2 8] + [ 4 30 1 1 1 1000 1 29 1 4 2 4 3 10 4 0 2 0 5 0 14] + [ 0 2 25 2 0 1 1139 4 0 0 5 1 0 0 0 2 0 0 1 5 4] + [ 2 8 14 1 4 17 3 1081 0 2 6 1 4 4 2 1 0 2 49 7 10] + [ 16 1 0 0 2 2 0 1 984 33 13 1 0 16 10 1 4 0 1 2 2] + [ 90 0 0 0 6 5 1 0 22 950 2 3 0 21 4 4 1 0 0 0 10] + [ 4 1 9 10 0 1 1 3 8 1 979 0 1 11 5 0 4 1 5 0 9] + [ 0 0 0 0 0 16 0 4 0 0 0 965 16 8 0 3 1 14 0 3 5] + [ 0 1 3 4 0 1 3 1 1 0 3 29 976 1 3 5 1 9 3 7 17] + [ 1 0 0 0 1 10 0 0 9 8 3 7 0 1060 4 0 0 1 0 1 14] + [ 11 1 3 19 4 1 0 0 23 0 4 0 3 3 1009 0 0 2 9 0 9] + [ 0 0 1 1 6 0 0 0 0 1 1 4 7 0 1 1074 13 10 3 7 5] + [ 1 12 1 1 6 4 1 0 0 0 0 4 0 1 4 9 1098 0 3 4 12] + [ 0 0 2 3 0 0 1 0 0 0 0 1 21 1 3 6 0 995 0 1 4] + [ 3 3 3 18 1 0 0 14 5 1 2 2 1 0 9 0 0 0 995 0 11] + [ 0 0 5 2 2 3 6 3 0 0 0 9 5 3 1 1 8 0 1 1094 9] + [ 117 128 100 81 48 109 29 90 65 70 157 83 305 264 108 47 71 48 117 149 5719]] + +2023-10-02 21:39:15,843 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:39:15,843 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:39:15,856 - + +2023-10-02 21:39:15,857 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:39:16,974 - Epoch: [146][ 10/ 1236] Overall Loss 0.149172 Objective Loss 0.149172 LR 0.000250 Time 0.111722 +2023-10-02 21:39:17,181 - Epoch: [146][ 20/ 1236] Overall Loss 0.139670 Objective Loss 0.139670 LR 0.000250 Time 0.066187 +2023-10-02 21:39:17,387 - Epoch: [146][ 30/ 1236] Overall Loss 0.139249 Objective Loss 0.139249 LR 0.000250 Time 0.050916 +2023-10-02 21:39:17,593 - Epoch: [146][ 40/ 1236] Overall Loss 0.139777 Objective Loss 0.139777 LR 0.000250 Time 0.043339 +2023-10-02 21:39:17,798 - Epoch: [146][ 50/ 1236] Overall Loss 0.137545 Objective Loss 0.137545 LR 0.000250 Time 0.038740 +2023-10-02 21:39:18,005 - Epoch: [146][ 60/ 1236] Overall Loss 0.136275 Objective Loss 0.136275 LR 0.000250 Time 0.035735 +2023-10-02 21:39:18,209 - Epoch: [146][ 70/ 1236] Overall Loss 0.136137 Objective Loss 0.136137 LR 0.000250 Time 0.033537 +2023-10-02 21:39:18,415 - Epoch: [146][ 80/ 1236] Overall Loss 0.138328 Objective Loss 0.138328 LR 0.000250 Time 0.031919 +2023-10-02 21:39:18,620 - Epoch: [146][ 90/ 1236] Overall Loss 0.141916 Objective Loss 0.141916 LR 0.000250 Time 0.030637 +2023-10-02 21:39:18,828 - Epoch: [146][ 100/ 1236] Overall Loss 0.141722 Objective Loss 0.141722 LR 0.000250 Time 0.029647 +2023-10-02 21:39:19,032 - Epoch: [146][ 110/ 1236] Overall Loss 0.143437 Objective Loss 0.143437 LR 0.000250 Time 0.028801 +2023-10-02 21:39:19,238 - Epoch: [146][ 120/ 1236] Overall Loss 0.143351 Objective Loss 0.143351 LR 0.000250 Time 0.028117 +2023-10-02 21:39:19,443 - Epoch: [146][ 130/ 1236] Overall Loss 0.142795 Objective Loss 0.142795 LR 0.000250 Time 0.027518 +2023-10-02 21:39:19,651 - Epoch: [146][ 140/ 1236] Overall Loss 0.142918 Objective Loss 0.142918 LR 0.000250 Time 0.027039 +2023-10-02 21:39:19,855 - Epoch: [146][ 150/ 1236] Overall Loss 0.142295 Objective Loss 0.142295 LR 0.000250 Time 0.026596 +2023-10-02 21:39:20,063 - Epoch: [146][ 160/ 1236] Overall Loss 0.142755 Objective Loss 0.142755 LR 0.000250 Time 0.026229 +2023-10-02 21:39:20,267 - Epoch: [146][ 170/ 1236] Overall Loss 0.142715 Objective Loss 0.142715 LR 0.000250 Time 0.025884 +2023-10-02 21:39:20,473 - Epoch: [146][ 180/ 1236] Overall Loss 0.142758 Objective Loss 0.142758 LR 0.000250 Time 0.025591 +2023-10-02 21:39:20,678 - Epoch: [146][ 190/ 1236] Overall Loss 0.143978 Objective Loss 0.143978 LR 0.000250 Time 0.025316 +2023-10-02 21:39:20,886 - Epoch: [146][ 200/ 1236] Overall Loss 0.144180 Objective Loss 0.144180 LR 0.000250 Time 0.025088 +2023-10-02 21:39:21,090 - Epoch: [146][ 210/ 1236] Overall Loss 0.144104 Objective Loss 0.144104 LR 0.000250 Time 0.024861 +2023-10-02 21:39:21,296 - Epoch: [146][ 220/ 1236] Overall Loss 0.145165 Objective Loss 0.145165 LR 0.000250 Time 0.024669 +2023-10-02 21:39:21,501 - Epoch: [146][ 230/ 1236] Overall Loss 0.144791 Objective Loss 0.144791 LR 0.000250 Time 0.024482 +2023-10-02 21:39:21,710 - Epoch: [146][ 240/ 1236] Overall Loss 0.144612 Objective Loss 0.144612 LR 0.000250 Time 0.024328 +2023-10-02 21:39:21,914 - Epoch: [146][ 250/ 1236] Overall Loss 0.144868 Objective Loss 0.144868 LR 0.000250 Time 0.024170 +2023-10-02 21:39:22,122 - Epoch: [146][ 260/ 1236] Overall Loss 0.145032 Objective Loss 0.145032 LR 0.000250 Time 0.024039 +2023-10-02 21:39:22,325 - Epoch: [146][ 270/ 1236] Overall Loss 0.145250 Objective Loss 0.145250 LR 0.000250 Time 0.023902 +2023-10-02 21:39:22,533 - Epoch: [146][ 280/ 1236] Overall Loss 0.144947 Objective Loss 0.144947 LR 0.000250 Time 0.023789 +2023-10-02 21:39:22,737 - Epoch: [146][ 290/ 1236] Overall Loss 0.144482 Objective Loss 0.144482 LR 0.000250 Time 0.023673 +2023-10-02 21:39:22,945 - Epoch: [146][ 300/ 1236] Overall Loss 0.144427 Objective Loss 0.144427 LR 0.000250 Time 0.023575 +2023-10-02 21:39:23,149 - Epoch: [146][ 310/ 1236] Overall Loss 0.144018 Objective Loss 0.144018 LR 0.000250 Time 0.023472 +2023-10-02 21:39:23,356 - Epoch: [146][ 320/ 1236] Overall Loss 0.143557 Objective Loss 0.143557 LR 0.000250 Time 0.023382 +2023-10-02 21:39:23,561 - Epoch: [146][ 330/ 1236] Overall Loss 0.143120 Objective Loss 0.143120 LR 0.000250 Time 0.023295 +2023-10-02 21:39:23,768 - Epoch: [146][ 340/ 1236] Overall Loss 0.142895 Objective Loss 0.142895 LR 0.000250 Time 0.023218 +2023-10-02 21:39:23,973 - Epoch: [146][ 350/ 1236] Overall Loss 0.142610 Objective Loss 0.142610 LR 0.000250 Time 0.023137 +2023-10-02 21:39:24,180 - Epoch: [146][ 360/ 1236] Overall Loss 0.142443 Objective Loss 0.142443 LR 0.000250 Time 0.023067 +2023-10-02 21:39:24,385 - Epoch: [146][ 370/ 1236] Overall Loss 0.142758 Objective Loss 0.142758 LR 0.000250 Time 0.022995 +2023-10-02 21:39:24,593 - Epoch: [146][ 380/ 1236] Overall Loss 0.142581 Objective Loss 0.142581 LR 0.000250 Time 0.022936 +2023-10-02 21:39:24,797 - Epoch: [146][ 390/ 1236] Overall Loss 0.142411 Objective Loss 0.142411 LR 0.000250 Time 0.022870 +2023-10-02 21:39:25,005 - Epoch: [146][ 400/ 1236] Overall Loss 0.142485 Objective Loss 0.142485 LR 0.000250 Time 0.022818 +2023-10-02 21:39:25,209 - Epoch: [146][ 410/ 1236] Overall Loss 0.142352 Objective Loss 0.142352 LR 0.000250 Time 0.022759 +2023-10-02 21:39:25,417 - Epoch: [146][ 420/ 1236] Overall Loss 0.142671 Objective Loss 0.142671 LR 0.000250 Time 0.022711 +2023-10-02 21:39:25,621 - Epoch: [146][ 430/ 1236] Overall Loss 0.142819 Objective Loss 0.142819 LR 0.000250 Time 0.022657 +2023-10-02 21:39:25,829 - Epoch: [146][ 440/ 1236] Overall Loss 0.142910 Objective Loss 0.142910 LR 0.000250 Time 0.022615 +2023-10-02 21:39:26,034 - Epoch: [146][ 450/ 1236] Overall Loss 0.142970 Objective Loss 0.142970 LR 0.000250 Time 0.022565 +2023-10-02 21:39:26,240 - Epoch: [146][ 460/ 1236] Overall Loss 0.143000 Objective Loss 0.143000 LR 0.000250 Time 0.022523 +2023-10-02 21:39:26,446 - Epoch: [146][ 470/ 1236] Overall Loss 0.143262 Objective Loss 0.143262 LR 0.000250 Time 0.022481 +2023-10-02 21:39:26,654 - Epoch: [146][ 480/ 1236] Overall Loss 0.143455 Objective Loss 0.143455 LR 0.000250 Time 0.022445 +2023-10-02 21:39:26,859 - Epoch: [146][ 490/ 1236] Overall Loss 0.143401 Objective Loss 0.143401 LR 0.000250 Time 0.022404 +2023-10-02 21:39:27,065 - Epoch: [146][ 500/ 1236] Overall Loss 0.143241 Objective Loss 0.143241 LR 0.000250 Time 0.022369 +2023-10-02 21:39:27,271 - Epoch: [146][ 510/ 1236] Overall Loss 0.143133 Objective Loss 0.143133 LR 0.000250 Time 0.022331 +2023-10-02 21:39:27,477 - Epoch: [146][ 520/ 1236] Overall Loss 0.142587 Objective Loss 0.142587 LR 0.000250 Time 0.022298 +2023-10-02 21:39:27,683 - Epoch: [146][ 530/ 1236] Overall Loss 0.142930 Objective Loss 0.142930 LR 0.000250 Time 0.022265 +2023-10-02 21:39:27,891 - Epoch: [146][ 540/ 1236] Overall Loss 0.142519 Objective Loss 0.142519 LR 0.000250 Time 0.022237 +2023-10-02 21:39:28,096 - Epoch: [146][ 550/ 1236] Overall Loss 0.142490 Objective Loss 0.142490 LR 0.000250 Time 0.022204 +2023-10-02 21:39:28,304 - Epoch: [146][ 560/ 1236] Overall Loss 0.142438 Objective Loss 0.142438 LR 0.000250 Time 0.022179 +2023-10-02 21:39:28,508 - Epoch: [146][ 570/ 1236] Overall Loss 0.142760 Objective Loss 0.142760 LR 0.000250 Time 0.022147 +2023-10-02 21:39:28,714 - Epoch: [146][ 580/ 1236] Overall Loss 0.142577 Objective Loss 0.142577 LR 0.000250 Time 0.022121 +2023-10-02 21:39:28,920 - Epoch: [146][ 590/ 1236] Overall Loss 0.142650 Objective Loss 0.142650 LR 0.000250 Time 0.022094 +2023-10-02 21:39:29,127 - Epoch: [146][ 600/ 1236] Overall Loss 0.142491 Objective Loss 0.142491 LR 0.000250 Time 0.022070 +2023-10-02 21:39:29,332 - Epoch: [146][ 610/ 1236] Overall Loss 0.142612 Objective Loss 0.142612 LR 0.000250 Time 0.022042 +2023-10-02 21:39:29,539 - Epoch: [146][ 620/ 1236] Overall Loss 0.142462 Objective Loss 0.142462 LR 0.000250 Time 0.022020 +2023-10-02 21:39:29,745 - Epoch: [146][ 630/ 1236] Overall Loss 0.142698 Objective Loss 0.142698 LR 0.000250 Time 0.021997 +2023-10-02 21:39:29,952 - Epoch: [146][ 640/ 1236] Overall Loss 0.142621 Objective Loss 0.142621 LR 0.000250 Time 0.021976 +2023-10-02 21:39:30,157 - Epoch: [146][ 650/ 1236] Overall Loss 0.142621 Objective Loss 0.142621 LR 0.000250 Time 0.021953 +2023-10-02 21:39:30,364 - Epoch: [146][ 660/ 1236] Overall Loss 0.142598 Objective Loss 0.142598 LR 0.000250 Time 0.021934 +2023-10-02 21:39:30,570 - Epoch: [146][ 670/ 1236] Overall Loss 0.142477 Objective Loss 0.142477 LR 0.000250 Time 0.021913 +2023-10-02 21:39:30,776 - Epoch: [146][ 680/ 1236] Overall Loss 0.142478 Objective Loss 0.142478 LR 0.000250 Time 0.021894 +2023-10-02 21:39:30,982 - Epoch: [146][ 690/ 1236] Overall Loss 0.142385 Objective Loss 0.142385 LR 0.000250 Time 0.021875 +2023-10-02 21:39:31,189 - Epoch: [146][ 700/ 1236] Overall Loss 0.142375 Objective Loss 0.142375 LR 0.000250 Time 0.021857 +2023-10-02 21:39:31,394 - Epoch: [146][ 710/ 1236] Overall Loss 0.142469 Objective Loss 0.142469 LR 0.000250 Time 0.021838 +2023-10-02 21:39:31,601 - Epoch: [146][ 720/ 1236] Overall Loss 0.142331 Objective Loss 0.142331 LR 0.000250 Time 0.021821 +2023-10-02 21:39:31,806 - Epoch: [146][ 730/ 1236] Overall Loss 0.142342 Objective Loss 0.142342 LR 0.000250 Time 0.021804 +2023-10-02 21:39:32,013 - Epoch: [146][ 740/ 1236] Overall Loss 0.142152 Objective Loss 0.142152 LR 0.000250 Time 0.021788 +2023-10-02 21:39:32,219 - Epoch: [146][ 750/ 1236] Overall Loss 0.142144 Objective Loss 0.142144 LR 0.000250 Time 0.021771 +2023-10-02 21:39:32,425 - Epoch: [146][ 760/ 1236] Overall Loss 0.142062 Objective Loss 0.142062 LR 0.000250 Time 0.021756 +2023-10-02 21:39:32,631 - Epoch: [146][ 770/ 1236] Overall Loss 0.142174 Objective Loss 0.142174 LR 0.000250 Time 0.021740 +2023-10-02 21:39:32,839 - Epoch: [146][ 780/ 1236] Overall Loss 0.142161 Objective Loss 0.142161 LR 0.000250 Time 0.021728 +2023-10-02 21:39:33,043 - Epoch: [146][ 790/ 1236] Overall Loss 0.141918 Objective Loss 0.141918 LR 0.000250 Time 0.021711 +2023-10-02 21:39:33,250 - Epoch: [146][ 800/ 1236] Overall Loss 0.141967 Objective Loss 0.141967 LR 0.000250 Time 0.021698 +2023-10-02 21:39:33,455 - Epoch: [146][ 810/ 1236] Overall Loss 0.142118 Objective Loss 0.142118 LR 0.000250 Time 0.021683 +2023-10-02 21:39:33,663 - Epoch: [146][ 820/ 1236] Overall Loss 0.142097 Objective Loss 0.142097 LR 0.000250 Time 0.021672 +2023-10-02 21:39:33,868 - Epoch: [146][ 830/ 1236] Overall Loss 0.142093 Objective Loss 0.142093 LR 0.000250 Time 0.021657 +2023-10-02 21:39:34,076 - Epoch: [146][ 840/ 1236] Overall Loss 0.142315 Objective Loss 0.142315 LR 0.000250 Time 0.021647 +2023-10-02 21:39:34,280 - Epoch: [146][ 850/ 1236] Overall Loss 0.142393 Objective Loss 0.142393 LR 0.000250 Time 0.021632 +2023-10-02 21:39:34,487 - Epoch: [146][ 860/ 1236] Overall Loss 0.142054 Objective Loss 0.142054 LR 0.000250 Time 0.021620 +2023-10-02 21:39:34,693 - Epoch: [146][ 870/ 1236] Overall Loss 0.142301 Objective Loss 0.142301 LR 0.000250 Time 0.021608 +2023-10-02 21:39:34,899 - Epoch: [146][ 880/ 1236] Overall Loss 0.142271 Objective Loss 0.142271 LR 0.000250 Time 0.021597 +2023-10-02 21:39:35,105 - Epoch: [146][ 890/ 1236] Overall Loss 0.142262 Objective Loss 0.142262 LR 0.000250 Time 0.021585 +2023-10-02 21:39:35,312 - Epoch: [146][ 900/ 1236] Overall Loss 0.142180 Objective Loss 0.142180 LR 0.000250 Time 0.021574 +2023-10-02 21:39:35,517 - Epoch: [146][ 910/ 1236] Overall Loss 0.142243 Objective Loss 0.142243 LR 0.000250 Time 0.021562 +2023-10-02 21:39:35,724 - Epoch: [146][ 920/ 1236] Overall Loss 0.142446 Objective Loss 0.142446 LR 0.000250 Time 0.021552 +2023-10-02 21:39:35,930 - Epoch: [146][ 930/ 1236] Overall Loss 0.142623 Objective Loss 0.142623 LR 0.000250 Time 0.021541 +2023-10-02 21:39:36,137 - Epoch: [146][ 940/ 1236] Overall Loss 0.142594 Objective Loss 0.142594 LR 0.000250 Time 0.021532 +2023-10-02 21:39:36,342 - Epoch: [146][ 950/ 1236] Overall Loss 0.142739 Objective Loss 0.142739 LR 0.000250 Time 0.021521 +2023-10-02 21:39:36,549 - Epoch: [146][ 960/ 1236] Overall Loss 0.142661 Objective Loss 0.142661 LR 0.000250 Time 0.021512 +2023-10-02 21:39:36,755 - Epoch: [146][ 970/ 1236] Overall Loss 0.142736 Objective Loss 0.142736 LR 0.000250 Time 0.021502 +2023-10-02 21:39:36,962 - Epoch: [146][ 980/ 1236] Overall Loss 0.142800 Objective Loss 0.142800 LR 0.000250 Time 0.021493 +2023-10-02 21:39:37,167 - Epoch: [146][ 990/ 1236] Overall Loss 0.142752 Objective Loss 0.142752 LR 0.000250 Time 0.021484 +2023-10-02 21:39:37,374 - Epoch: [146][ 1000/ 1236] Overall Loss 0.142713 Objective Loss 0.142713 LR 0.000250 Time 0.021475 +2023-10-02 21:39:37,580 - Epoch: [146][ 1010/ 1236] Overall Loss 0.142782 Objective Loss 0.142782 LR 0.000250 Time 0.021466 +2023-10-02 21:39:37,787 - Epoch: [146][ 1020/ 1236] Overall Loss 0.142544 Objective Loss 0.142544 LR 0.000250 Time 0.021458 +2023-10-02 21:39:37,992 - Epoch: [146][ 1030/ 1236] Overall Loss 0.142503 Objective Loss 0.142503 LR 0.000250 Time 0.021449 +2023-10-02 21:39:38,199 - Epoch: [146][ 1040/ 1236] Overall Loss 0.142457 Objective Loss 0.142457 LR 0.000250 Time 0.021441 +2023-10-02 21:39:38,404 - Epoch: [146][ 1050/ 1236] Overall Loss 0.142404 Objective Loss 0.142404 LR 0.000250 Time 0.021433 +2023-10-02 21:39:38,611 - Epoch: [146][ 1060/ 1236] Overall Loss 0.142386 Objective Loss 0.142386 LR 0.000250 Time 0.021425 +2023-10-02 21:39:38,817 - Epoch: [146][ 1070/ 1236] Overall Loss 0.142527 Objective Loss 0.142527 LR 0.000250 Time 0.021417 +2023-10-02 21:39:39,023 - Epoch: [146][ 1080/ 1236] Overall Loss 0.142474 Objective Loss 0.142474 LR 0.000250 Time 0.021410 +2023-10-02 21:39:39,229 - Epoch: [146][ 1090/ 1236] Overall Loss 0.142612 Objective Loss 0.142612 LR 0.000250 Time 0.021402 +2023-10-02 21:39:39,436 - Epoch: [146][ 1100/ 1236] Overall Loss 0.142659 Objective Loss 0.142659 LR 0.000250 Time 0.021395 +2023-10-02 21:39:39,641 - Epoch: [146][ 1110/ 1236] Overall Loss 0.142850 Objective Loss 0.142850 LR 0.000250 Time 0.021387 +2023-10-02 21:39:39,849 - Epoch: [146][ 1120/ 1236] Overall Loss 0.142786 Objective Loss 0.142786 LR 0.000250 Time 0.021381 +2023-10-02 21:39:40,054 - Epoch: [146][ 1130/ 1236] Overall Loss 0.142873 Objective Loss 0.142873 LR 0.000250 Time 0.021373 +2023-10-02 21:39:40,261 - Epoch: [146][ 1140/ 1236] Overall Loss 0.142766 Objective Loss 0.142766 LR 0.000250 Time 0.021367 +2023-10-02 21:39:40,466 - Epoch: [146][ 1150/ 1236] Overall Loss 0.142772 Objective Loss 0.142772 LR 0.000250 Time 0.021359 +2023-10-02 21:39:40,673 - Epoch: [146][ 1160/ 1236] Overall Loss 0.142740 Objective Loss 0.142740 LR 0.000250 Time 0.021353 +2023-10-02 21:39:40,879 - Epoch: [146][ 1170/ 1236] Overall Loss 0.142700 Objective Loss 0.142700 LR 0.000250 Time 0.021346 +2023-10-02 21:39:41,085 - Epoch: [146][ 1180/ 1236] Overall Loss 0.142787 Objective Loss 0.142787 LR 0.000250 Time 0.021340 +2023-10-02 21:39:41,291 - Epoch: [146][ 1190/ 1236] Overall Loss 0.142776 Objective Loss 0.142776 LR 0.000250 Time 0.021334 +2023-10-02 21:39:41,498 - Epoch: [146][ 1200/ 1236] Overall Loss 0.142746 Objective Loss 0.142746 LR 0.000250 Time 0.021328 +2023-10-02 21:39:41,703 - Epoch: [146][ 1210/ 1236] Overall Loss 0.142650 Objective Loss 0.142650 LR 0.000250 Time 0.021321 +2023-10-02 21:39:41,910 - Epoch: [146][ 1220/ 1236] Overall Loss 0.142573 Objective Loss 0.142573 LR 0.000250 Time 0.021316 +2023-10-02 21:39:42,169 - Epoch: [146][ 1230/ 1236] Overall Loss 0.142506 Objective Loss 0.142506 LR 0.000250 Time 0.021352 +2023-10-02 21:39:42,289 - Epoch: [146][ 1236/ 1236] Overall Loss 0.142555 Objective Loss 0.142555 Top1 90.224033 Top5 99.185336 LR 0.000250 Time 0.021346 +2023-10-02 21:39:42,425 - --- validate (epoch=146)----------- +2023-10-02 21:39:42,425 - 29943 samples (256 per mini-batch) +2023-10-02 21:39:42,918 - Epoch: [146][ 10/ 117] Loss 0.264507 Top1 87.500000 Top5 98.906250 +2023-10-02 21:39:43,070 - Epoch: [146][ 20/ 117] Loss 0.287703 Top1 86.796875 Top5 98.867188 +2023-10-02 21:39:43,221 - Epoch: [146][ 30/ 117] Loss 0.287234 Top1 86.757812 Top5 98.815104 +2023-10-02 21:39:43,372 - Epoch: [146][ 40/ 117] Loss 0.295883 Top1 86.718750 Top5 98.701172 +2023-10-02 21:39:43,522 - Epoch: [146][ 50/ 117] Loss 0.290055 Top1 86.937500 Top5 98.695312 +2023-10-02 21:39:43,673 - Epoch: [146][ 60/ 117] Loss 0.292277 Top1 86.920573 Top5 98.665365 +2023-10-02 21:39:43,824 - Epoch: [146][ 70/ 117] Loss 0.293737 Top1 86.791295 Top5 98.666295 +2023-10-02 21:39:43,974 - Epoch: [146][ 80/ 117] Loss 0.293190 Top1 86.796875 Top5 98.637695 +2023-10-02 21:39:44,125 - Epoch: [146][ 90/ 117] Loss 0.291667 Top1 86.731771 Top5 98.641493 +2023-10-02 21:39:44,276 - Epoch: [146][ 100/ 117] Loss 0.292532 Top1 86.734375 Top5 98.664062 +2023-10-02 21:39:44,434 - Epoch: [146][ 110/ 117] Loss 0.294699 Top1 86.622869 Top5 98.636364 +2023-10-02 21:39:44,522 - Epoch: [146][ 117/ 117] Loss 0.293889 Top1 86.571152 Top5 98.627392 +2023-10-02 21:39:44,655 - ==> Top1: 86.571 Top5: 98.627 Loss: 0.294 + +2023-10-02 21:39:44,656 - ==> Confusion: +[[ 953 1 0 1 9 2 0 1 8 47 2 1 1 0 5 0 1 0 0 0 18] + [ 1 1052 2 1 4 27 1 19 1 1 1 1 0 0 0 3 1 0 10 1 5] + [ 2 0 984 11 1 0 10 8 0 2 2 1 9 2 1 3 2 1 9 1 7] + [ 2 5 9 988 0 0 0 1 4 1 4 0 6 4 22 2 1 2 16 1 21] + [ 29 1 0 1 974 5 0 0 1 10 0 1 0 2 10 3 10 0 0 0 3] + [ 2 32 1 2 6 999 1 20 1 3 2 6 3 9 3 0 7 0 4 3 12] + [ 0 3 39 3 0 2 1112 6 0 0 4 1 0 1 0 6 0 0 1 7 6] + [ 2 10 10 0 6 26 5 1078 0 4 5 3 3 6 1 1 0 1 41 9 7] + [ 16 0 0 1 4 3 0 0 974 44 12 2 2 11 10 1 4 0 1 1 3] + [ 99 0 1 1 7 2 0 0 23 946 2 3 0 16 7 4 1 0 0 0 7] + [ 5 4 7 9 0 1 1 2 8 1 983 1 0 11 3 0 0 2 7 0 8] + [ 0 3 0 0 1 16 0 3 0 0 0 976 12 4 0 5 0 10 0 2 3] + [ 1 1 3 2 0 2 1 0 1 2 4 35 979 1 2 9 0 5 0 7 13] + [ 1 0 0 0 3 9 1 0 8 10 4 9 1 1052 4 0 0 1 0 1 15] + [ 10 0 6 19 4 1 0 0 20 0 2 0 2 1 1013 0 1 5 10 0 7] + [ 0 0 1 1 5 0 0 0 0 1 1 5 7 0 0 1075 15 10 3 5 5] + [ 0 13 1 1 6 5 0 0 0 0 0 5 0 2 4 9 1101 0 1 4 9] + [ 0 1 0 2 1 0 2 0 0 2 1 4 24 1 4 6 0 985 0 2 3] + [ 2 4 3 15 0 0 0 19 5 2 0 1 2 0 9 0 0 0 996 0 10] + [ 0 0 5 3 1 4 7 7 0 0 0 15 5 1 1 1 8 1 0 1085 8] + [ 128 135 109 88 76 115 26 80 74 81 172 96 287 276 124 49 69 45 133 125 5617]] + +2023-10-02 21:39:44,657 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:39:44,657 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:39:44,663 - + +2023-10-02 21:39:44,663 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:39:45,670 - Epoch: [147][ 10/ 1236] Overall Loss 0.153848 Objective Loss 0.153848 LR 0.000250 Time 0.100671 +2023-10-02 21:39:45,876 - Epoch: [147][ 20/ 1236] Overall Loss 0.153230 Objective Loss 0.153230 LR 0.000250 Time 0.060605 +2023-10-02 21:39:46,080 - Epoch: [147][ 30/ 1236] Overall Loss 0.149974 Objective Loss 0.149974 LR 0.000250 Time 0.047189 +2023-10-02 21:39:46,285 - Epoch: [147][ 40/ 1236] Overall Loss 0.151003 Objective Loss 0.151003 LR 0.000250 Time 0.040521 +2023-10-02 21:39:46,489 - Epoch: [147][ 50/ 1236] Overall Loss 0.148648 Objective Loss 0.148648 LR 0.000250 Time 0.036489 +2023-10-02 21:39:46,694 - Epoch: [147][ 60/ 1236] Overall Loss 0.147124 Objective Loss 0.147124 LR 0.000250 Time 0.033820 +2023-10-02 21:39:46,898 - Epoch: [147][ 70/ 1236] Overall Loss 0.144755 Objective Loss 0.144755 LR 0.000250 Time 0.031898 +2023-10-02 21:39:47,104 - Epoch: [147][ 80/ 1236] Overall Loss 0.144206 Objective Loss 0.144206 LR 0.000250 Time 0.030473 +2023-10-02 21:39:47,307 - Epoch: [147][ 90/ 1236] Overall Loss 0.143820 Objective Loss 0.143820 LR 0.000250 Time 0.029348 +2023-10-02 21:39:47,514 - Epoch: [147][ 100/ 1236] Overall Loss 0.142084 Objective Loss 0.142084 LR 0.000250 Time 0.028476 +2023-10-02 21:39:47,716 - Epoch: [147][ 110/ 1236] Overall Loss 0.140051 Objective Loss 0.140051 LR 0.000250 Time 0.027722 +2023-10-02 21:39:47,923 - Epoch: [147][ 120/ 1236] Overall Loss 0.139256 Objective Loss 0.139256 LR 0.000250 Time 0.027131 +2023-10-02 21:39:48,125 - Epoch: [147][ 130/ 1236] Overall Loss 0.137349 Objective Loss 0.137349 LR 0.000250 Time 0.026599 +2023-10-02 21:39:48,332 - Epoch: [147][ 140/ 1236] Overall Loss 0.137998 Objective Loss 0.137998 LR 0.000250 Time 0.026172 +2023-10-02 21:39:48,534 - Epoch: [147][ 150/ 1236] Overall Loss 0.138458 Objective Loss 0.138458 LR 0.000250 Time 0.025776 +2023-10-02 21:39:48,740 - Epoch: [147][ 160/ 1236] Overall Loss 0.138007 Objective Loss 0.138007 LR 0.000250 Time 0.025446 +2023-10-02 21:39:48,943 - Epoch: [147][ 170/ 1236] Overall Loss 0.137493 Objective Loss 0.137493 LR 0.000250 Time 0.025138 +2023-10-02 21:39:49,148 - Epoch: [147][ 180/ 1236] Overall Loss 0.138418 Objective Loss 0.138418 LR 0.000250 Time 0.024881 +2023-10-02 21:39:49,352 - Epoch: [147][ 190/ 1236] Overall Loss 0.138471 Objective Loss 0.138471 LR 0.000250 Time 0.024636 +2023-10-02 21:39:49,558 - Epoch: [147][ 200/ 1236] Overall Loss 0.139149 Objective Loss 0.139149 LR 0.000250 Time 0.024429 +2023-10-02 21:39:49,761 - Epoch: [147][ 210/ 1236] Overall Loss 0.138984 Objective Loss 0.138984 LR 0.000250 Time 0.024229 +2023-10-02 21:39:49,967 - Epoch: [147][ 220/ 1236] Overall Loss 0.138756 Objective Loss 0.138756 LR 0.000250 Time 0.024061 +2023-10-02 21:39:50,171 - Epoch: [147][ 230/ 1236] Overall Loss 0.139287 Objective Loss 0.139287 LR 0.000250 Time 0.023894 +2023-10-02 21:39:50,376 - Epoch: [147][ 240/ 1236] Overall Loss 0.138925 Objective Loss 0.138925 LR 0.000250 Time 0.023753 +2023-10-02 21:39:50,580 - Epoch: [147][ 250/ 1236] Overall Loss 0.139129 Objective Loss 0.139129 LR 0.000250 Time 0.023613 +2023-10-02 21:39:50,785 - Epoch: [147][ 260/ 1236] Overall Loss 0.139013 Objective Loss 0.139013 LR 0.000250 Time 0.023493 +2023-10-02 21:39:50,989 - Epoch: [147][ 270/ 1236] Overall Loss 0.139157 Objective Loss 0.139157 LR 0.000250 Time 0.023372 +2023-10-02 21:39:51,194 - Epoch: [147][ 280/ 1236] Overall Loss 0.138893 Objective Loss 0.138893 LR 0.000250 Time 0.023270 +2023-10-02 21:39:51,398 - Epoch: [147][ 290/ 1236] Overall Loss 0.138130 Objective Loss 0.138130 LR 0.000250 Time 0.023165 +2023-10-02 21:39:51,603 - Epoch: [147][ 300/ 1236] Overall Loss 0.138242 Objective Loss 0.138242 LR 0.000250 Time 0.023076 +2023-10-02 21:39:51,807 - Epoch: [147][ 310/ 1236] Overall Loss 0.137736 Objective Loss 0.137736 LR 0.000250 Time 0.022984 +2023-10-02 21:39:52,013 - Epoch: [147][ 320/ 1236] Overall Loss 0.137797 Objective Loss 0.137797 LR 0.000250 Time 0.022907 +2023-10-02 21:39:52,216 - Epoch: [147][ 330/ 1236] Overall Loss 0.138008 Objective Loss 0.138008 LR 0.000250 Time 0.022826 +2023-10-02 21:39:52,422 - Epoch: [147][ 340/ 1236] Overall Loss 0.138195 Objective Loss 0.138195 LR 0.000250 Time 0.022759 +2023-10-02 21:39:52,627 - Epoch: [147][ 350/ 1236] Overall Loss 0.138256 Objective Loss 0.138256 LR 0.000250 Time 0.022689 +2023-10-02 21:39:52,834 - Epoch: [147][ 360/ 1236] Overall Loss 0.138466 Objective Loss 0.138466 LR 0.000250 Time 0.022633 +2023-10-02 21:39:53,037 - Epoch: [147][ 370/ 1236] Overall Loss 0.138078 Objective Loss 0.138078 LR 0.000250 Time 0.022569 +2023-10-02 21:39:53,242 - Epoch: [147][ 380/ 1236] Overall Loss 0.138061 Objective Loss 0.138061 LR 0.000250 Time 0.022515 +2023-10-02 21:39:53,446 - Epoch: [147][ 390/ 1236] Overall Loss 0.138277 Objective Loss 0.138277 LR 0.000250 Time 0.022460 +2023-10-02 21:39:53,652 - Epoch: [147][ 400/ 1236] Overall Loss 0.138258 Objective Loss 0.138258 LR 0.000250 Time 0.022412 +2023-10-02 21:39:53,856 - Epoch: [147][ 410/ 1236] Overall Loss 0.137884 Objective Loss 0.137884 LR 0.000250 Time 0.022362 +2023-10-02 21:39:54,061 - Epoch: [147][ 420/ 1236] Overall Loss 0.138349 Objective Loss 0.138349 LR 0.000250 Time 0.022318 +2023-10-02 21:39:54,265 - Epoch: [147][ 430/ 1236] Overall Loss 0.138440 Objective Loss 0.138440 LR 0.000250 Time 0.022273 +2023-10-02 21:39:54,471 - Epoch: [147][ 440/ 1236] Overall Loss 0.138582 Objective Loss 0.138582 LR 0.000250 Time 0.022232 +2023-10-02 21:39:54,675 - Epoch: [147][ 450/ 1236] Overall Loss 0.138329 Objective Loss 0.138329 LR 0.000250 Time 0.022191 +2023-10-02 21:39:54,880 - Epoch: [147][ 460/ 1236] Overall Loss 0.138393 Objective Loss 0.138393 LR 0.000250 Time 0.022155 +2023-10-02 21:39:55,085 - Epoch: [147][ 470/ 1236] Overall Loss 0.138499 Objective Loss 0.138499 LR 0.000250 Time 0.022118 +2023-10-02 21:39:55,290 - Epoch: [147][ 480/ 1236] Overall Loss 0.138691 Objective Loss 0.138691 LR 0.000250 Time 0.022084 +2023-10-02 21:39:55,494 - Epoch: [147][ 490/ 1236] Overall Loss 0.138764 Objective Loss 0.138764 LR 0.000250 Time 0.022050 +2023-10-02 21:39:55,700 - Epoch: [147][ 500/ 1236] Overall Loss 0.138872 Objective Loss 0.138872 LR 0.000250 Time 0.022019 +2023-10-02 21:39:55,904 - Epoch: [147][ 510/ 1236] Overall Loss 0.138939 Objective Loss 0.138939 LR 0.000250 Time 0.021987 +2023-10-02 21:39:56,109 - Epoch: [147][ 520/ 1236] Overall Loss 0.138836 Objective Loss 0.138836 LR 0.000250 Time 0.021958 +2023-10-02 21:39:56,313 - Epoch: [147][ 530/ 1236] Overall Loss 0.138791 Objective Loss 0.138791 LR 0.000250 Time 0.021928 +2023-10-02 21:39:56,518 - Epoch: [147][ 540/ 1236] Overall Loss 0.138663 Objective Loss 0.138663 LR 0.000250 Time 0.021902 +2023-10-02 21:39:56,722 - Epoch: [147][ 550/ 1236] Overall Loss 0.138791 Objective Loss 0.138791 LR 0.000250 Time 0.021874 +2023-10-02 21:39:56,928 - Epoch: [147][ 560/ 1236] Overall Loss 0.138642 Objective Loss 0.138642 LR 0.000250 Time 0.021850 +2023-10-02 21:39:57,132 - Epoch: [147][ 570/ 1236] Overall Loss 0.138885 Objective Loss 0.138885 LR 0.000250 Time 0.021824 +2023-10-02 21:39:57,337 - Epoch: [147][ 580/ 1236] Overall Loss 0.139314 Objective Loss 0.139314 LR 0.000250 Time 0.021801 +2023-10-02 21:39:57,541 - Epoch: [147][ 590/ 1236] Overall Loss 0.139622 Objective Loss 0.139622 LR 0.000250 Time 0.021777 +2023-10-02 21:39:57,747 - Epoch: [147][ 600/ 1236] Overall Loss 0.139507 Objective Loss 0.139507 LR 0.000250 Time 0.021756 +2023-10-02 21:39:57,951 - Epoch: [147][ 610/ 1236] Overall Loss 0.139756 Objective Loss 0.139756 LR 0.000250 Time 0.021733 +2023-10-02 21:39:58,156 - Epoch: [147][ 620/ 1236] Overall Loss 0.139946 Objective Loss 0.139946 LR 0.000250 Time 0.021713 +2023-10-02 21:39:58,360 - Epoch: [147][ 630/ 1236] Overall Loss 0.140141 Objective Loss 0.140141 LR 0.000250 Time 0.021692 +2023-10-02 21:39:58,566 - Epoch: [147][ 640/ 1236] Overall Loss 0.139996 Objective Loss 0.139996 LR 0.000250 Time 0.021674 +2023-10-02 21:39:58,770 - Epoch: [147][ 650/ 1236] Overall Loss 0.140418 Objective Loss 0.140418 LR 0.000250 Time 0.021655 +2023-10-02 21:39:58,976 - Epoch: [147][ 660/ 1236] Overall Loss 0.140463 Objective Loss 0.140463 LR 0.000250 Time 0.021637 +2023-10-02 21:39:59,180 - Epoch: [147][ 670/ 1236] Overall Loss 0.140410 Objective Loss 0.140410 LR 0.000250 Time 0.021619 +2023-10-02 21:39:59,385 - Epoch: [147][ 680/ 1236] Overall Loss 0.140433 Objective Loss 0.140433 LR 0.000250 Time 0.021603 +2023-10-02 21:39:59,589 - Epoch: [147][ 690/ 1236] Overall Loss 0.140684 Objective Loss 0.140684 LR 0.000250 Time 0.021585 +2023-10-02 21:39:59,795 - Epoch: [147][ 700/ 1236] Overall Loss 0.140923 Objective Loss 0.140923 LR 0.000250 Time 0.021570 +2023-10-02 21:39:59,999 - Epoch: [147][ 710/ 1236] Overall Loss 0.141298 Objective Loss 0.141298 LR 0.000250 Time 0.021553 +2023-10-02 21:40:00,204 - Epoch: [147][ 720/ 1236] Overall Loss 0.141161 Objective Loss 0.141161 LR 0.000250 Time 0.021539 +2023-10-02 21:40:00,408 - Epoch: [147][ 730/ 1236] Overall Loss 0.141501 Objective Loss 0.141501 LR 0.000250 Time 0.021523 +2023-10-02 21:40:00,614 - Epoch: [147][ 740/ 1236] Overall Loss 0.141359 Objective Loss 0.141359 LR 0.000250 Time 0.021509 +2023-10-02 21:40:00,818 - Epoch: [147][ 750/ 1236] Overall Loss 0.141661 Objective Loss 0.141661 LR 0.000250 Time 0.021494 +2023-10-02 21:40:01,025 - Epoch: [147][ 760/ 1236] Overall Loss 0.141926 Objective Loss 0.141926 LR 0.000250 Time 0.021483 +2023-10-02 21:40:01,228 - Epoch: [147][ 770/ 1236] Overall Loss 0.142030 Objective Loss 0.142030 LR 0.000250 Time 0.021467 +2023-10-02 21:40:01,433 - Epoch: [147][ 780/ 1236] Overall Loss 0.142142 Objective Loss 0.142142 LR 0.000250 Time 0.021455 +2023-10-02 21:40:01,637 - Epoch: [147][ 790/ 1236] Overall Loss 0.142208 Objective Loss 0.142208 LR 0.000250 Time 0.021441 +2023-10-02 21:40:01,843 - Epoch: [147][ 800/ 1236] Overall Loss 0.142016 Objective Loss 0.142016 LR 0.000250 Time 0.021430 +2023-10-02 21:40:02,047 - Epoch: [147][ 810/ 1236] Overall Loss 0.142195 Objective Loss 0.142195 LR 0.000250 Time 0.021417 +2023-10-02 21:40:02,254 - Epoch: [147][ 820/ 1236] Overall Loss 0.142213 Objective Loss 0.142213 LR 0.000250 Time 0.021407 +2023-10-02 21:40:02,456 - Epoch: [147][ 830/ 1236] Overall Loss 0.142237 Objective Loss 0.142237 LR 0.000250 Time 0.021393 +2023-10-02 21:40:02,662 - Epoch: [147][ 840/ 1236] Overall Loss 0.142116 Objective Loss 0.142116 LR 0.000250 Time 0.021383 +2023-10-02 21:40:02,866 - Epoch: [147][ 850/ 1236] Overall Loss 0.142106 Objective Loss 0.142106 LR 0.000250 Time 0.021371 +2023-10-02 21:40:03,073 - Epoch: [147][ 860/ 1236] Overall Loss 0.142218 Objective Loss 0.142218 LR 0.000250 Time 0.021363 +2023-10-02 21:40:03,276 - Epoch: [147][ 870/ 1236] Overall Loss 0.142199 Objective Loss 0.142199 LR 0.000250 Time 0.021350 +2023-10-02 21:40:03,481 - Epoch: [147][ 880/ 1236] Overall Loss 0.142472 Objective Loss 0.142472 LR 0.000250 Time 0.021341 +2023-10-02 21:40:03,685 - Epoch: [147][ 890/ 1236] Overall Loss 0.142560 Objective Loss 0.142560 LR 0.000250 Time 0.021330 +2023-10-02 21:40:03,891 - Epoch: [147][ 900/ 1236] Overall Loss 0.142641 Objective Loss 0.142641 LR 0.000250 Time 0.021321 +2023-10-02 21:40:04,095 - Epoch: [147][ 910/ 1236] Overall Loss 0.142432 Objective Loss 0.142432 LR 0.000250 Time 0.021311 +2023-10-02 21:40:04,301 - Epoch: [147][ 920/ 1236] Overall Loss 0.142348 Objective Loss 0.142348 LR 0.000250 Time 0.021302 +2023-10-02 21:40:04,505 - Epoch: [147][ 930/ 1236] Overall Loss 0.142366 Objective Loss 0.142366 LR 0.000250 Time 0.021292 +2023-10-02 21:40:04,712 - Epoch: [147][ 940/ 1236] Overall Loss 0.142444 Objective Loss 0.142444 LR 0.000250 Time 0.021286 +2023-10-02 21:40:04,915 - Epoch: [147][ 950/ 1236] Overall Loss 0.142510 Objective Loss 0.142510 LR 0.000250 Time 0.021275 +2023-10-02 21:40:05,120 - Epoch: [147][ 960/ 1236] Overall Loss 0.142596 Objective Loss 0.142596 LR 0.000250 Time 0.021267 +2023-10-02 21:40:05,324 - Epoch: [147][ 970/ 1236] Overall Loss 0.142497 Objective Loss 0.142497 LR 0.000250 Time 0.021258 +2023-10-02 21:40:05,530 - Epoch: [147][ 980/ 1236] Overall Loss 0.142746 Objective Loss 0.142746 LR 0.000250 Time 0.021251 +2023-10-02 21:40:05,734 - Epoch: [147][ 990/ 1236] Overall Loss 0.142622 Objective Loss 0.142622 LR 0.000250 Time 0.021242 +2023-10-02 21:40:05,940 - Epoch: [147][ 1000/ 1236] Overall Loss 0.142581 Objective Loss 0.142581 LR 0.000250 Time 0.021235 +2023-10-02 21:40:06,144 - Epoch: [147][ 1010/ 1236] Overall Loss 0.142601 Objective Loss 0.142601 LR 0.000250 Time 0.021227 +2023-10-02 21:40:06,350 - Epoch: [147][ 1020/ 1236] Overall Loss 0.142702 Objective Loss 0.142702 LR 0.000250 Time 0.021220 +2023-10-02 21:40:06,554 - Epoch: [147][ 1030/ 1236] Overall Loss 0.142644 Objective Loss 0.142644 LR 0.000250 Time 0.021212 +2023-10-02 21:40:06,759 - Epoch: [147][ 1040/ 1236] Overall Loss 0.142532 Objective Loss 0.142532 LR 0.000250 Time 0.021205 +2023-10-02 21:40:06,963 - Epoch: [147][ 1050/ 1236] Overall Loss 0.142465 Objective Loss 0.142465 LR 0.000250 Time 0.021197 +2023-10-02 21:40:07,169 - Epoch: [147][ 1060/ 1236] Overall Loss 0.142431 Objective Loss 0.142431 LR 0.000250 Time 0.021191 +2023-10-02 21:40:07,373 - Epoch: [147][ 1070/ 1236] Overall Loss 0.142609 Objective Loss 0.142609 LR 0.000250 Time 0.021183 +2023-10-02 21:40:07,580 - Epoch: [147][ 1080/ 1236] Overall Loss 0.142638 Objective Loss 0.142638 LR 0.000250 Time 0.021178 +2023-10-02 21:40:07,783 - Epoch: [147][ 1090/ 1236] Overall Loss 0.142575 Objective Loss 0.142575 LR 0.000250 Time 0.021170 +2023-10-02 21:40:07,989 - Epoch: [147][ 1100/ 1236] Overall Loss 0.142738 Objective Loss 0.142738 LR 0.000250 Time 0.021165 +2023-10-02 21:40:08,192 - Epoch: [147][ 1110/ 1236] Overall Loss 0.142804 Objective Loss 0.142804 LR 0.000250 Time 0.021157 +2023-10-02 21:40:08,398 - Epoch: [147][ 1120/ 1236] Overall Loss 0.142787 Objective Loss 0.142787 LR 0.000250 Time 0.021152 +2023-10-02 21:40:08,602 - Epoch: [147][ 1130/ 1236] Overall Loss 0.142912 Objective Loss 0.142912 LR 0.000250 Time 0.021145 +2023-10-02 21:40:08,808 - Epoch: [147][ 1140/ 1236] Overall Loss 0.142823 Objective Loss 0.142823 LR 0.000250 Time 0.021140 +2023-10-02 21:40:09,012 - Epoch: [147][ 1150/ 1236] Overall Loss 0.142745 Objective Loss 0.142745 LR 0.000250 Time 0.021133 +2023-10-02 21:40:09,217 - Epoch: [147][ 1160/ 1236] Overall Loss 0.142823 Objective Loss 0.142823 LR 0.000250 Time 0.021127 +2023-10-02 21:40:09,421 - Epoch: [147][ 1170/ 1236] Overall Loss 0.142630 Objective Loss 0.142630 LR 0.000250 Time 0.021121 +2023-10-02 21:40:09,627 - Epoch: [147][ 1180/ 1236] Overall Loss 0.142538 Objective Loss 0.142538 LR 0.000250 Time 0.021116 +2023-10-02 21:40:09,831 - Epoch: [147][ 1190/ 1236] Overall Loss 0.142488 Objective Loss 0.142488 LR 0.000250 Time 0.021110 +2023-10-02 21:40:10,036 - Epoch: [147][ 1200/ 1236] Overall Loss 0.142604 Objective Loss 0.142604 LR 0.000250 Time 0.021105 +2023-10-02 21:40:10,240 - Epoch: [147][ 1210/ 1236] Overall Loss 0.142600 Objective Loss 0.142600 LR 0.000250 Time 0.021098 +2023-10-02 21:40:10,446 - Epoch: [147][ 1220/ 1236] Overall Loss 0.142608 Objective Loss 0.142608 LR 0.000250 Time 0.021093 +2023-10-02 21:40:10,702 - Epoch: [147][ 1230/ 1236] Overall Loss 0.142684 Objective Loss 0.142684 LR 0.000250 Time 0.021128 +2023-10-02 21:40:10,821 - Epoch: [147][ 1236/ 1236] Overall Loss 0.142632 Objective Loss 0.142632 Top1 92.260692 Top5 99.185336 LR 0.000250 Time 0.021123 +2023-10-02 21:40:10,966 - --- validate (epoch=147)----------- +2023-10-02 21:40:10,966 - 29943 samples (256 per mini-batch) +2023-10-02 21:40:11,475 - Epoch: [147][ 10/ 117] Loss 0.317467 Top1 85.976562 Top5 98.593750 +2023-10-02 21:40:11,632 - Epoch: [147][ 20/ 117] Loss 0.303300 Top1 86.093750 Top5 98.652344 +2023-10-02 21:40:11,789 - Epoch: [147][ 30/ 117] Loss 0.300055 Top1 86.263021 Top5 98.567708 +2023-10-02 21:40:11,945 - Epoch: [147][ 40/ 117] Loss 0.297973 Top1 86.435547 Top5 98.623047 +2023-10-02 21:40:12,101 - Epoch: [147][ 50/ 117] Loss 0.293822 Top1 86.679688 Top5 98.671875 +2023-10-02 21:40:12,257 - Epoch: [147][ 60/ 117] Loss 0.290629 Top1 86.848958 Top5 98.652344 +2023-10-02 21:40:12,413 - Epoch: [147][ 70/ 117] Loss 0.293243 Top1 86.919643 Top5 98.554688 +2023-10-02 21:40:12,569 - Epoch: [147][ 80/ 117] Loss 0.290772 Top1 86.860352 Top5 98.603516 +2023-10-02 21:40:12,724 - Epoch: [147][ 90/ 117] Loss 0.292562 Top1 86.805556 Top5 98.645833 +2023-10-02 21:40:12,880 - Epoch: [147][ 100/ 117] Loss 0.289459 Top1 86.964844 Top5 98.636719 +2023-10-02 21:40:13,044 - Epoch: [147][ 110/ 117] Loss 0.291094 Top1 86.917614 Top5 98.657670 +2023-10-02 21:40:13,132 - Epoch: [147][ 117/ 117] Loss 0.290761 Top1 86.908459 Top5 98.670808 +2023-10-02 21:40:13,228 - ==> Top1: 86.908 Top5: 98.671 Loss: 0.291 + +2023-10-02 21:40:13,229 - ==> Confusion: +[[ 931 0 2 0 8 2 0 0 7 65 2 1 1 1 7 0 3 1 0 0 19] + [ 0 1063 1 1 6 17 0 21 1 0 1 1 0 0 1 3 1 0 10 2 2] + [ 3 2 980 8 0 0 13 6 0 2 3 2 8 3 0 3 1 2 9 4 7] + [ 1 5 14 980 0 1 0 1 3 0 6 0 10 6 26 4 1 2 11 0 18] + [ 23 6 0 0 968 5 1 0 0 10 1 0 2 4 9 3 11 0 0 1 6] + [ 3 38 0 0 3 996 1 20 2 5 1 5 2 10 2 1 3 0 5 1 18] + [ 0 1 24 0 0 3 1140 4 0 0 4 1 0 0 0 4 0 0 1 4 5] + [ 1 10 11 2 6 19 4 1091 0 5 4 4 5 5 1 0 0 3 32 6 9] + [ 15 0 0 1 1 3 0 0 971 41 12 3 1 15 14 0 3 1 4 3 1] + [ 76 0 1 0 4 4 0 0 28 963 2 2 0 23 7 3 0 0 0 1 5] + [ 2 2 6 9 0 2 5 4 12 2 973 2 1 8 4 0 2 4 5 3 7] + [ 0 1 2 0 0 17 0 2 0 1 0 949 28 10 0 0 1 16 0 4 4] + [ 1 1 1 3 1 2 2 0 0 0 3 21 997 0 2 6 1 10 1 5 11] + [ 0 0 1 0 2 6 0 0 10 4 2 5 1 1067 4 0 0 1 0 1 15] + [ 11 0 5 15 3 0 0 0 18 1 1 0 3 5 1019 0 0 3 7 0 10] + [ 0 0 1 1 6 0 0 0 0 0 0 3 6 0 0 1077 16 12 2 6 4] + [ 1 13 0 0 4 6 0 0 0 0 0 4 0 2 4 9 1104 0 0 4 10] + [ 0 0 0 2 0 0 2 0 0 0 0 2 22 1 2 5 0 996 0 2 4] + [ 1 2 2 15 1 1 1 17 3 2 3 0 1 0 10 0 0 2 995 0 12] + [ 0 0 3 1 1 1 12 5 0 0 0 15 3 1 0 3 7 2 1 1091 6] + [ 93 135 99 95 52 112 36 74 75 72 173 69 328 250 130 53 74 57 112 144 5672]] + +2023-10-02 21:40:13,230 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:40:13,230 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:40:13,237 - + +2023-10-02 21:40:13,237 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:40:14,379 - Epoch: [148][ 10/ 1236] Overall Loss 0.140491 Objective Loss 0.140491 LR 0.000250 Time 0.114203 +2023-10-02 21:40:14,586 - Epoch: [148][ 20/ 1236] Overall Loss 0.132516 Objective Loss 0.132516 LR 0.000250 Time 0.067439 +2023-10-02 21:40:14,793 - Epoch: [148][ 30/ 1236] Overall Loss 0.140189 Objective Loss 0.140189 LR 0.000250 Time 0.051842 +2023-10-02 21:40:15,002 - Epoch: [148][ 40/ 1236] Overall Loss 0.140860 Objective Loss 0.140860 LR 0.000250 Time 0.044097 +2023-10-02 21:40:15,207 - Epoch: [148][ 50/ 1236] Overall Loss 0.139819 Objective Loss 0.139819 LR 0.000250 Time 0.039372 +2023-10-02 21:40:15,415 - Epoch: [148][ 60/ 1236] Overall Loss 0.138205 Objective Loss 0.138205 LR 0.000250 Time 0.036273 +2023-10-02 21:40:15,620 - Epoch: [148][ 70/ 1236] Overall Loss 0.138069 Objective Loss 0.138069 LR 0.000250 Time 0.034016 +2023-10-02 21:40:15,828 - Epoch: [148][ 80/ 1236] Overall Loss 0.139210 Objective Loss 0.139210 LR 0.000250 Time 0.032361 +2023-10-02 21:40:16,035 - Epoch: [148][ 90/ 1236] Overall Loss 0.138572 Objective Loss 0.138572 LR 0.000250 Time 0.031054 +2023-10-02 21:40:16,242 - Epoch: [148][ 100/ 1236] Overall Loss 0.135677 Objective Loss 0.135677 LR 0.000250 Time 0.030024 +2023-10-02 21:40:16,449 - Epoch: [148][ 110/ 1236] Overall Loss 0.136060 Objective Loss 0.136060 LR 0.000250 Time 0.029175 +2023-10-02 21:40:16,657 - Epoch: [148][ 120/ 1236] Overall Loss 0.134743 Objective Loss 0.134743 LR 0.000250 Time 0.028473 +2023-10-02 21:40:16,864 - Epoch: [148][ 130/ 1236] Overall Loss 0.134948 Objective Loss 0.134948 LR 0.000250 Time 0.027858 +2023-10-02 21:40:17,072 - Epoch: [148][ 140/ 1236] Overall Loss 0.136826 Objective Loss 0.136826 LR 0.000250 Time 0.027355 +2023-10-02 21:40:17,278 - Epoch: [148][ 150/ 1236] Overall Loss 0.136659 Objective Loss 0.136659 LR 0.000250 Time 0.026894 +2023-10-02 21:40:17,487 - Epoch: [148][ 160/ 1236] Overall Loss 0.135969 Objective Loss 0.135969 LR 0.000250 Time 0.026520 +2023-10-02 21:40:17,692 - Epoch: [148][ 170/ 1236] Overall Loss 0.136176 Objective Loss 0.136176 LR 0.000250 Time 0.026162 +2023-10-02 21:40:17,901 - Epoch: [148][ 180/ 1236] Overall Loss 0.136293 Objective Loss 0.136293 LR 0.000250 Time 0.025871 +2023-10-02 21:40:18,106 - Epoch: [148][ 190/ 1236] Overall Loss 0.136012 Objective Loss 0.136012 LR 0.000250 Time 0.025585 +2023-10-02 21:40:18,316 - Epoch: [148][ 200/ 1236] Overall Loss 0.135721 Objective Loss 0.135721 LR 0.000250 Time 0.025352 +2023-10-02 21:40:18,520 - Epoch: [148][ 210/ 1236] Overall Loss 0.135611 Objective Loss 0.135611 LR 0.000250 Time 0.025118 +2023-10-02 21:40:18,728 - Epoch: [148][ 220/ 1236] Overall Loss 0.135552 Objective Loss 0.135552 LR 0.000250 Time 0.024920 +2023-10-02 21:40:18,934 - Epoch: [148][ 230/ 1236] Overall Loss 0.135350 Objective Loss 0.135350 LR 0.000250 Time 0.024727 +2023-10-02 21:40:19,142 - Epoch: [148][ 240/ 1236] Overall Loss 0.136253 Objective Loss 0.136253 LR 0.000250 Time 0.024562 +2023-10-02 21:40:19,348 - Epoch: [148][ 250/ 1236] Overall Loss 0.136080 Objective Loss 0.136080 LR 0.000250 Time 0.024398 +2023-10-02 21:40:19,557 - Epoch: [148][ 260/ 1236] Overall Loss 0.136907 Objective Loss 0.136907 LR 0.000250 Time 0.024259 +2023-10-02 21:40:19,762 - Epoch: [148][ 270/ 1236] Overall Loss 0.137693 Objective Loss 0.137693 LR 0.000250 Time 0.024117 +2023-10-02 21:40:19,970 - Epoch: [148][ 280/ 1236] Overall Loss 0.137350 Objective Loss 0.137350 LR 0.000250 Time 0.023998 +2023-10-02 21:40:20,176 - Epoch: [148][ 290/ 1236] Overall Loss 0.137352 Objective Loss 0.137352 LR 0.000250 Time 0.023876 +2023-10-02 21:40:20,385 - Epoch: [148][ 300/ 1236] Overall Loss 0.138084 Objective Loss 0.138084 LR 0.000250 Time 0.023773 +2023-10-02 21:40:20,591 - Epoch: [148][ 310/ 1236] Overall Loss 0.138279 Objective Loss 0.138279 LR 0.000250 Time 0.023666 +2023-10-02 21:40:20,799 - Epoch: [148][ 320/ 1236] Overall Loss 0.138610 Objective Loss 0.138610 LR 0.000250 Time 0.023575 +2023-10-02 21:40:21,005 - Epoch: [148][ 330/ 1236] Overall Loss 0.139051 Objective Loss 0.139051 LR 0.000250 Time 0.023482 +2023-10-02 21:40:21,213 - Epoch: [148][ 340/ 1236] Overall Loss 0.138547 Objective Loss 0.138547 LR 0.000250 Time 0.023403 +2023-10-02 21:40:21,419 - Epoch: [148][ 350/ 1236] Overall Loss 0.138232 Objective Loss 0.138232 LR 0.000250 Time 0.023319 +2023-10-02 21:40:21,628 - Epoch: [148][ 360/ 1236] Overall Loss 0.138265 Objective Loss 0.138265 LR 0.000250 Time 0.023248 +2023-10-02 21:40:21,834 - Epoch: [148][ 370/ 1236] Overall Loss 0.138594 Objective Loss 0.138594 LR 0.000250 Time 0.023174 +2023-10-02 21:40:22,043 - Epoch: [148][ 380/ 1236] Overall Loss 0.138574 Objective Loss 0.138574 LR 0.000250 Time 0.023114 +2023-10-02 21:40:22,248 - Epoch: [148][ 390/ 1236] Overall Loss 0.138588 Objective Loss 0.138588 LR 0.000250 Time 0.023046 +2023-10-02 21:40:22,456 - Epoch: [148][ 400/ 1236] Overall Loss 0.139374 Objective Loss 0.139374 LR 0.000250 Time 0.022990 +2023-10-02 21:40:22,663 - Epoch: [148][ 410/ 1236] Overall Loss 0.139541 Objective Loss 0.139541 LR 0.000250 Time 0.022932 +2023-10-02 21:40:22,870 - Epoch: [148][ 420/ 1236] Overall Loss 0.140009 Objective Loss 0.140009 LR 0.000250 Time 0.022878 +2023-10-02 21:40:23,076 - Epoch: [148][ 430/ 1236] Overall Loss 0.140316 Objective Loss 0.140316 LR 0.000250 Time 0.022825 +2023-10-02 21:40:23,283 - Epoch: [148][ 440/ 1236] Overall Loss 0.140883 Objective Loss 0.140883 LR 0.000250 Time 0.022777 +2023-10-02 21:40:23,490 - Epoch: [148][ 450/ 1236] Overall Loss 0.140413 Objective Loss 0.140413 LR 0.000250 Time 0.022728 +2023-10-02 21:40:23,699 - Epoch: [148][ 460/ 1236] Overall Loss 0.140006 Objective Loss 0.140006 LR 0.000250 Time 0.022689 +2023-10-02 21:40:23,905 - Epoch: [148][ 470/ 1236] Overall Loss 0.139673 Objective Loss 0.139673 LR 0.000250 Time 0.022643 +2023-10-02 21:40:24,114 - Epoch: [148][ 480/ 1236] Overall Loss 0.139629 Objective Loss 0.139629 LR 0.000250 Time 0.022607 +2023-10-02 21:40:24,319 - Epoch: [148][ 490/ 1236] Overall Loss 0.139620 Objective Loss 0.139620 LR 0.000250 Time 0.022564 +2023-10-02 21:40:24,528 - Epoch: [148][ 500/ 1236] Overall Loss 0.139415 Objective Loss 0.139415 LR 0.000250 Time 0.022528 +2023-10-02 21:40:24,734 - Epoch: [148][ 510/ 1236] Overall Loss 0.139153 Objective Loss 0.139153 LR 0.000250 Time 0.022491 +2023-10-02 21:40:24,942 - Epoch: [148][ 520/ 1236] Overall Loss 0.139195 Objective Loss 0.139195 LR 0.000250 Time 0.022458 +2023-10-02 21:40:25,148 - Epoch: [148][ 530/ 1236] Overall Loss 0.139055 Objective Loss 0.139055 LR 0.000250 Time 0.022419 +2023-10-02 21:40:25,356 - Epoch: [148][ 540/ 1236] Overall Loss 0.139094 Objective Loss 0.139094 LR 0.000250 Time 0.022389 +2023-10-02 21:40:25,562 - Epoch: [148][ 550/ 1236] Overall Loss 0.139125 Objective Loss 0.139125 LR 0.000250 Time 0.022357 +2023-10-02 21:40:25,772 - Epoch: [148][ 560/ 1236] Overall Loss 0.139316 Objective Loss 0.139316 LR 0.000250 Time 0.022331 +2023-10-02 21:40:25,977 - Epoch: [148][ 570/ 1236] Overall Loss 0.139679 Objective Loss 0.139679 LR 0.000250 Time 0.022299 +2023-10-02 21:40:26,185 - Epoch: [148][ 580/ 1236] Overall Loss 0.139483 Objective Loss 0.139483 LR 0.000250 Time 0.022272 +2023-10-02 21:40:26,391 - Epoch: [148][ 590/ 1236] Overall Loss 0.139476 Objective Loss 0.139476 LR 0.000250 Time 0.022244 +2023-10-02 21:40:26,600 - Epoch: [148][ 600/ 1236] Overall Loss 0.139774 Objective Loss 0.139774 LR 0.000250 Time 0.022220 +2023-10-02 21:40:26,806 - Epoch: [148][ 610/ 1236] Overall Loss 0.140038 Objective Loss 0.140038 LR 0.000250 Time 0.022194 +2023-10-02 21:40:27,015 - Epoch: [148][ 620/ 1236] Overall Loss 0.139882 Objective Loss 0.139882 LR 0.000250 Time 0.022172 +2023-10-02 21:40:27,221 - Epoch: [148][ 630/ 1236] Overall Loss 0.139667 Objective Loss 0.139667 LR 0.000250 Time 0.022147 +2023-10-02 21:40:27,430 - Epoch: [148][ 640/ 1236] Overall Loss 0.139605 Objective Loss 0.139605 LR 0.000250 Time 0.022126 +2023-10-02 21:40:27,636 - Epoch: [148][ 650/ 1236] Overall Loss 0.139628 Objective Loss 0.139628 LR 0.000250 Time 0.022103 +2023-10-02 21:40:27,844 - Epoch: [148][ 660/ 1236] Overall Loss 0.139717 Objective Loss 0.139717 LR 0.000250 Time 0.022083 +2023-10-02 21:40:28,051 - Epoch: [148][ 670/ 1236] Overall Loss 0.139555 Objective Loss 0.139555 LR 0.000250 Time 0.022062 +2023-10-02 21:40:28,259 - Epoch: [148][ 680/ 1236] Overall Loss 0.139634 Objective Loss 0.139634 LR 0.000250 Time 0.022043 +2023-10-02 21:40:28,465 - Epoch: [148][ 690/ 1236] Overall Loss 0.139809 Objective Loss 0.139809 LR 0.000250 Time 0.022022 +2023-10-02 21:40:28,673 - Epoch: [148][ 700/ 1236] Overall Loss 0.140096 Objective Loss 0.140096 LR 0.000250 Time 0.022004 +2023-10-02 21:40:28,880 - Epoch: [148][ 710/ 1236] Overall Loss 0.140002 Objective Loss 0.140002 LR 0.000250 Time 0.021985 +2023-10-02 21:40:29,088 - Epoch: [148][ 720/ 1236] Overall Loss 0.139948 Objective Loss 0.139948 LR 0.000250 Time 0.021968 +2023-10-02 21:40:29,295 - Epoch: [148][ 730/ 1236] Overall Loss 0.140017 Objective Loss 0.140017 LR 0.000250 Time 0.021950 +2023-10-02 21:40:29,503 - Epoch: [148][ 740/ 1236] Overall Loss 0.140345 Objective Loss 0.140345 LR 0.000250 Time 0.021935 +2023-10-02 21:40:29,710 - Epoch: [148][ 750/ 1236] Overall Loss 0.140374 Objective Loss 0.140374 LR 0.000250 Time 0.021917 +2023-10-02 21:40:29,918 - Epoch: [148][ 760/ 1236] Overall Loss 0.140297 Objective Loss 0.140297 LR 0.000250 Time 0.021902 +2023-10-02 21:40:30,125 - Epoch: [148][ 770/ 1236] Overall Loss 0.140186 Objective Loss 0.140186 LR 0.000250 Time 0.021886 +2023-10-02 21:40:30,333 - Epoch: [148][ 780/ 1236] Overall Loss 0.140598 Objective Loss 0.140598 LR 0.000250 Time 0.021872 +2023-10-02 21:40:30,540 - Epoch: [148][ 790/ 1236] Overall Loss 0.140450 Objective Loss 0.140450 LR 0.000250 Time 0.021857 +2023-10-02 21:40:30,747 - Epoch: [148][ 800/ 1236] Overall Loss 0.140398 Objective Loss 0.140398 LR 0.000250 Time 0.021842 +2023-10-02 21:40:30,954 - Epoch: [148][ 810/ 1236] Overall Loss 0.140418 Objective Loss 0.140418 LR 0.000250 Time 0.021827 +2023-10-02 21:40:31,162 - Epoch: [148][ 820/ 1236] Overall Loss 0.140348 Objective Loss 0.140348 LR 0.000250 Time 0.021815 +2023-10-02 21:40:31,369 - Epoch: [148][ 830/ 1236] Overall Loss 0.140546 Objective Loss 0.140546 LR 0.000250 Time 0.021801 +2023-10-02 21:40:31,577 - Epoch: [148][ 840/ 1236] Overall Loss 0.140396 Objective Loss 0.140396 LR 0.000250 Time 0.021789 +2023-10-02 21:40:31,784 - Epoch: [148][ 850/ 1236] Overall Loss 0.140522 Objective Loss 0.140522 LR 0.000250 Time 0.021775 +2023-10-02 21:40:31,992 - Epoch: [148][ 860/ 1236] Overall Loss 0.140555 Objective Loss 0.140555 LR 0.000250 Time 0.021764 +2023-10-02 21:40:32,199 - Epoch: [148][ 870/ 1236] Overall Loss 0.140440 Objective Loss 0.140440 LR 0.000250 Time 0.021751 +2023-10-02 21:40:32,407 - Epoch: [148][ 880/ 1236] Overall Loss 0.140190 Objective Loss 0.140190 LR 0.000250 Time 0.021740 +2023-10-02 21:40:32,614 - Epoch: [148][ 890/ 1236] Overall Loss 0.140255 Objective Loss 0.140255 LR 0.000250 Time 0.021728 +2023-10-02 21:40:32,821 - Epoch: [148][ 900/ 1236] Overall Loss 0.140575 Objective Loss 0.140575 LR 0.000250 Time 0.021716 +2023-10-02 21:40:33,028 - Epoch: [148][ 910/ 1236] Overall Loss 0.140425 Objective Loss 0.140425 LR 0.000250 Time 0.021705 +2023-10-02 21:40:33,236 - Epoch: [148][ 920/ 1236] Overall Loss 0.140453 Objective Loss 0.140453 LR 0.000250 Time 0.021695 +2023-10-02 21:40:33,443 - Epoch: [148][ 930/ 1236] Overall Loss 0.140460 Objective Loss 0.140460 LR 0.000250 Time 0.021683 +2023-10-02 21:40:33,651 - Epoch: [148][ 940/ 1236] Overall Loss 0.140514 Objective Loss 0.140514 LR 0.000250 Time 0.021674 +2023-10-02 21:40:33,858 - Epoch: [148][ 950/ 1236] Overall Loss 0.140461 Objective Loss 0.140461 LR 0.000250 Time 0.021663 +2023-10-02 21:40:34,066 - Epoch: [148][ 960/ 1236] Overall Loss 0.140414 Objective Loss 0.140414 LR 0.000250 Time 0.021654 +2023-10-02 21:40:34,273 - Epoch: [148][ 970/ 1236] Overall Loss 0.140549 Objective Loss 0.140549 LR 0.000250 Time 0.021643 +2023-10-02 21:40:34,481 - Epoch: [148][ 980/ 1236] Overall Loss 0.140667 Objective Loss 0.140667 LR 0.000250 Time 0.021635 +2023-10-02 21:40:34,688 - Epoch: [148][ 990/ 1236] Overall Loss 0.140790 Objective Loss 0.140790 LR 0.000250 Time 0.021625 +2023-10-02 21:40:34,896 - Epoch: [148][ 1000/ 1236] Overall Loss 0.140784 Objective Loss 0.140784 LR 0.000250 Time 0.021617 +2023-10-02 21:40:35,103 - Epoch: [148][ 1010/ 1236] Overall Loss 0.140654 Objective Loss 0.140654 LR 0.000250 Time 0.021607 +2023-10-02 21:40:35,310 - Epoch: [148][ 1020/ 1236] Overall Loss 0.140844 Objective Loss 0.140844 LR 0.000250 Time 0.021598 +2023-10-02 21:40:35,517 - Epoch: [148][ 1030/ 1236] Overall Loss 0.140926 Objective Loss 0.140926 LR 0.000250 Time 0.021589 +2023-10-02 21:40:35,725 - Epoch: [148][ 1040/ 1236] Overall Loss 0.140992 Objective Loss 0.140992 LR 0.000250 Time 0.021581 +2023-10-02 21:40:35,932 - Epoch: [148][ 1050/ 1236] Overall Loss 0.141293 Objective Loss 0.141293 LR 0.000250 Time 0.021573 +2023-10-02 21:40:36,140 - Epoch: [148][ 1060/ 1236] Overall Loss 0.141334 Objective Loss 0.141334 LR 0.000250 Time 0.021565 +2023-10-02 21:40:36,347 - Epoch: [148][ 1070/ 1236] Overall Loss 0.141489 Objective Loss 0.141489 LR 0.000250 Time 0.021557 +2023-10-02 21:40:36,555 - Epoch: [148][ 1080/ 1236] Overall Loss 0.141330 Objective Loss 0.141330 LR 0.000250 Time 0.021550 +2023-10-02 21:40:36,762 - Epoch: [148][ 1090/ 1236] Overall Loss 0.141549 Objective Loss 0.141549 LR 0.000250 Time 0.021541 +2023-10-02 21:40:36,971 - Epoch: [148][ 1100/ 1236] Overall Loss 0.141524 Objective Loss 0.141524 LR 0.000250 Time 0.021536 +2023-10-02 21:40:37,177 - Epoch: [148][ 1110/ 1236] Overall Loss 0.141649 Objective Loss 0.141649 LR 0.000250 Time 0.021527 +2023-10-02 21:40:37,384 - Epoch: [148][ 1120/ 1236] Overall Loss 0.141741 Objective Loss 0.141741 LR 0.000250 Time 0.021519 +2023-10-02 21:40:37,591 - Epoch: [148][ 1130/ 1236] Overall Loss 0.141571 Objective Loss 0.141571 LR 0.000250 Time 0.021511 +2023-10-02 21:40:37,798 - Epoch: [148][ 1140/ 1236] Overall Loss 0.141517 Objective Loss 0.141517 LR 0.000250 Time 0.021504 +2023-10-02 21:40:38,005 - Epoch: [148][ 1150/ 1236] Overall Loss 0.141526 Objective Loss 0.141526 LR 0.000250 Time 0.021497 +2023-10-02 21:40:38,213 - Epoch: [148][ 1160/ 1236] Overall Loss 0.141517 Objective Loss 0.141517 LR 0.000250 Time 0.021491 +2023-10-02 21:40:38,420 - Epoch: [148][ 1170/ 1236] Overall Loss 0.141643 Objective Loss 0.141643 LR 0.000250 Time 0.021484 +2023-10-02 21:40:38,627 - Epoch: [148][ 1180/ 1236] Overall Loss 0.141606 Objective Loss 0.141606 LR 0.000250 Time 0.021477 +2023-10-02 21:40:38,834 - Epoch: [148][ 1190/ 1236] Overall Loss 0.141556 Objective Loss 0.141556 LR 0.000250 Time 0.021470 +2023-10-02 21:40:39,042 - Epoch: [148][ 1200/ 1236] Overall Loss 0.141575 Objective Loss 0.141575 LR 0.000250 Time 0.021464 +2023-10-02 21:40:39,249 - Epoch: [148][ 1210/ 1236] Overall Loss 0.141597 Objective Loss 0.141597 LR 0.000250 Time 0.021458 +2023-10-02 21:40:39,458 - Epoch: [148][ 1220/ 1236] Overall Loss 0.141386 Objective Loss 0.141386 LR 0.000250 Time 0.021452 +2023-10-02 21:40:39,716 - Epoch: [148][ 1230/ 1236] Overall Loss 0.141466 Objective Loss 0.141466 LR 0.000250 Time 0.021488 +2023-10-02 21:40:39,837 - Epoch: [148][ 1236/ 1236] Overall Loss 0.141559 Objective Loss 0.141559 Top1 89.409369 Top5 98.167006 LR 0.000250 Time 0.021482 +2023-10-02 21:40:39,973 - --- validate (epoch=148)----------- +2023-10-02 21:40:39,973 - 29943 samples (256 per mini-batch) +2023-10-02 21:40:40,470 - Epoch: [148][ 10/ 117] Loss 0.295461 Top1 86.562500 Top5 98.789062 +2023-10-02 21:40:40,624 - Epoch: [148][ 20/ 117] Loss 0.286600 Top1 86.855469 Top5 98.789062 +2023-10-02 21:40:40,775 - Epoch: [148][ 30/ 117] Loss 0.280447 Top1 86.966146 Top5 98.867188 +2023-10-02 21:40:40,930 - Epoch: [148][ 40/ 117] Loss 0.288608 Top1 87.011719 Top5 98.701172 +2023-10-02 21:40:41,082 - Epoch: [148][ 50/ 117] Loss 0.292008 Top1 87.039062 Top5 98.718750 +2023-10-02 21:40:41,236 - Epoch: [148][ 60/ 117] Loss 0.293492 Top1 86.861979 Top5 98.717448 +2023-10-02 21:40:41,389 - Epoch: [148][ 70/ 117] Loss 0.294970 Top1 86.886161 Top5 98.671875 +2023-10-02 21:40:41,543 - Epoch: [148][ 80/ 117] Loss 0.294663 Top1 86.943359 Top5 98.720703 +2023-10-02 21:40:41,695 - Epoch: [148][ 90/ 117] Loss 0.295613 Top1 86.931424 Top5 98.715278 +2023-10-02 21:40:41,843 - Epoch: [148][ 100/ 117] Loss 0.293639 Top1 86.902344 Top5 98.722656 +2023-10-02 21:40:41,999 - Epoch: [148][ 110/ 117] Loss 0.294054 Top1 86.878551 Top5 98.710938 +2023-10-02 21:40:42,088 - Epoch: [148][ 117/ 117] Loss 0.294923 Top1 86.841666 Top5 98.704205 +2023-10-02 21:40:42,235 - ==> Top1: 86.842 Top5: 98.704 Loss: 0.295 + +2023-10-02 21:40:42,236 - ==> Confusion: +[[ 953 0 1 0 4 1 0 0 8 46 1 0 2 2 10 1 2 0 1 0 18] + [ 0 1077 1 1 4 12 0 14 1 1 0 1 0 0 0 3 0 0 8 2 6] + [ 4 1 991 10 1 1 6 6 0 1 1 0 6 3 1 3 0 2 9 3 7] + [ 1 4 11 989 0 1 1 1 4 0 2 0 5 4 26 2 0 5 11 1 21] + [ 27 6 0 0 969 5 0 0 1 10 0 0 1 4 10 3 8 0 0 2 4] + [ 3 43 0 2 1 993 1 21 1 3 0 7 0 11 4 0 3 1 5 3 14] + [ 0 4 29 3 0 1 1130 2 0 0 2 0 0 0 0 3 0 1 2 9 5] + [ 0 16 15 1 9 22 2 1082 0 3 1 2 4 4 2 0 0 2 38 6 9] + [ 17 4 0 0 2 4 0 0 978 42 6 2 2 9 12 1 6 0 3 0 1] + [ 104 1 1 1 8 3 0 0 24 936 1 2 0 14 14 2 1 1 0 0 6] + [ 3 5 6 12 0 1 3 2 14 2 967 1 0 9 9 0 2 3 5 0 9] + [ 0 0 2 0 0 21 0 1 0 0 0 961 17 5 0 3 0 14 0 7 4] + [ 0 0 1 4 0 1 2 1 0 0 6 33 972 1 1 9 2 11 1 3 20] + [ 1 0 1 0 1 12 1 0 14 15 5 6 1 1036 6 0 0 1 0 2 17] + [ 10 0 5 11 4 0 0 0 14 1 2 0 2 4 1025 0 1 3 10 0 9] + [ 0 0 1 1 6 0 0 0 0 0 0 6 6 0 2 1075 14 8 2 7 6] + [ 0 15 2 0 3 4 1 0 1 0 0 6 0 2 4 9 1096 0 3 4 11] + [ 0 0 0 3 0 0 2 0 0 0 0 6 17 0 2 8 1 993 0 2 4] + [ 3 4 3 18 0 1 1 17 3 2 1 0 2 0 11 0 0 0 991 0 11] + [ 0 0 5 2 2 3 7 5 0 0 0 15 5 2 1 2 4 1 0 1089 9] + [ 107 166 124 94 53 100 18 75 86 59 123 99 302 245 138 44 71 54 123 124 5700]] + +2023-10-02 21:40:42,237 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:40:42,237 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:40:42,243 - + +2023-10-02 21:40:42,243 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:40:43,270 - Epoch: [149][ 10/ 1236] Overall Loss 0.129541 Objective Loss 0.129541 LR 0.000250 Time 0.102637 +2023-10-02 21:40:43,477 - Epoch: [149][ 20/ 1236] Overall Loss 0.127803 Objective Loss 0.127803 LR 0.000250 Time 0.061641 +2023-10-02 21:40:43,684 - Epoch: [149][ 30/ 1236] Overall Loss 0.134727 Objective Loss 0.134727 LR 0.000250 Time 0.047982 +2023-10-02 21:40:43,892 - Epoch: [149][ 40/ 1236] Overall Loss 0.131830 Objective Loss 0.131830 LR 0.000250 Time 0.041166 +2023-10-02 21:40:44,096 - Epoch: [149][ 50/ 1236] Overall Loss 0.128717 Objective Loss 0.128717 LR 0.000250 Time 0.037021 +2023-10-02 21:40:44,304 - Epoch: [149][ 60/ 1236] Overall Loss 0.131166 Objective Loss 0.131166 LR 0.000250 Time 0.034309 +2023-10-02 21:40:44,509 - Epoch: [149][ 70/ 1236] Overall Loss 0.128841 Objective Loss 0.128841 LR 0.000250 Time 0.032328 +2023-10-02 21:40:44,717 - Epoch: [149][ 80/ 1236] Overall Loss 0.129570 Objective Loss 0.129570 LR 0.000250 Time 0.030880 +2023-10-02 21:40:44,919 - Epoch: [149][ 90/ 1236] Overall Loss 0.132697 Objective Loss 0.132697 LR 0.000250 Time 0.029700 +2023-10-02 21:40:45,126 - Epoch: [149][ 100/ 1236] Overall Loss 0.133616 Objective Loss 0.133616 LR 0.000250 Time 0.028792 +2023-10-02 21:40:45,331 - Epoch: [149][ 110/ 1236] Overall Loss 0.135453 Objective Loss 0.135453 LR 0.000250 Time 0.028038 +2023-10-02 21:40:45,539 - Epoch: [149][ 120/ 1236] Overall Loss 0.135949 Objective Loss 0.135949 LR 0.000250 Time 0.027428 +2023-10-02 21:40:45,743 - Epoch: [149][ 130/ 1236] Overall Loss 0.135131 Objective Loss 0.135131 LR 0.000250 Time 0.026889 +2023-10-02 21:40:45,950 - Epoch: [149][ 140/ 1236] Overall Loss 0.135479 Objective Loss 0.135479 LR 0.000250 Time 0.026446 +2023-10-02 21:40:46,155 - Epoch: [149][ 150/ 1236] Overall Loss 0.136220 Objective Loss 0.136220 LR 0.000250 Time 0.026048 +2023-10-02 21:40:46,363 - Epoch: [149][ 160/ 1236] Overall Loss 0.136221 Objective Loss 0.136221 LR 0.000250 Time 0.025715 +2023-10-02 21:40:46,567 - Epoch: [149][ 170/ 1236] Overall Loss 0.137612 Objective Loss 0.137612 LR 0.000250 Time 0.025403 +2023-10-02 21:40:46,774 - Epoch: [149][ 180/ 1236] Overall Loss 0.138583 Objective Loss 0.138583 LR 0.000250 Time 0.025140 +2023-10-02 21:40:46,980 - Epoch: [149][ 190/ 1236] Overall Loss 0.139186 Objective Loss 0.139186 LR 0.000250 Time 0.024898 +2023-10-02 21:40:47,187 - Epoch: [149][ 200/ 1236] Overall Loss 0.138781 Objective Loss 0.138781 LR 0.000250 Time 0.024689 +2023-10-02 21:40:47,392 - Epoch: [149][ 210/ 1236] Overall Loss 0.138775 Objective Loss 0.138775 LR 0.000250 Time 0.024485 +2023-10-02 21:40:47,599 - Epoch: [149][ 220/ 1236] Overall Loss 0.139105 Objective Loss 0.139105 LR 0.000250 Time 0.024313 +2023-10-02 21:40:47,804 - Epoch: [149][ 230/ 1236] Overall Loss 0.140634 Objective Loss 0.140634 LR 0.000250 Time 0.024143 +2023-10-02 21:40:48,012 - Epoch: [149][ 240/ 1236] Overall Loss 0.140952 Objective Loss 0.140952 LR 0.000250 Time 0.024003 +2023-10-02 21:40:48,217 - Epoch: [149][ 250/ 1236] Overall Loss 0.140604 Objective Loss 0.140604 LR 0.000250 Time 0.023864 +2023-10-02 21:40:48,427 - Epoch: [149][ 260/ 1236] Overall Loss 0.140326 Objective Loss 0.140326 LR 0.000250 Time 0.023752 +2023-10-02 21:40:48,636 - Epoch: [149][ 270/ 1236] Overall Loss 0.140481 Objective Loss 0.140481 LR 0.000250 Time 0.023645 +2023-10-02 21:40:48,851 - Epoch: [149][ 280/ 1236] Overall Loss 0.140486 Objective Loss 0.140486 LR 0.000250 Time 0.023567 +2023-10-02 21:40:49,071 - Epoch: [149][ 290/ 1236] Overall Loss 0.140165 Objective Loss 0.140165 LR 0.000250 Time 0.023511 +2023-10-02 21:40:49,282 - Epoch: [149][ 300/ 1236] Overall Loss 0.139368 Objective Loss 0.139368 LR 0.000250 Time 0.023430 +2023-10-02 21:40:49,493 - Epoch: [149][ 310/ 1236] Overall Loss 0.140210 Objective Loss 0.140210 LR 0.000250 Time 0.023354 +2023-10-02 21:40:49,701 - Epoch: [149][ 320/ 1236] Overall Loss 0.139780 Objective Loss 0.139780 LR 0.000250 Time 0.023274 +2023-10-02 21:40:49,912 - Epoch: [149][ 330/ 1236] Overall Loss 0.139410 Objective Loss 0.139410 LR 0.000250 Time 0.023207 +2023-10-02 21:40:50,121 - Epoch: [149][ 340/ 1236] Overall Loss 0.139675 Objective Loss 0.139675 LR 0.000250 Time 0.023138 +2023-10-02 21:40:50,332 - Epoch: [149][ 350/ 1236] Overall Loss 0.139918 Objective Loss 0.139918 LR 0.000250 Time 0.023080 +2023-10-02 21:40:50,540 - Epoch: [149][ 360/ 1236] Overall Loss 0.140144 Objective Loss 0.140144 LR 0.000250 Time 0.023016 +2023-10-02 21:40:50,751 - Epoch: [149][ 370/ 1236] Overall Loss 0.140601 Objective Loss 0.140601 LR 0.000250 Time 0.022963 +2023-10-02 21:40:50,960 - Epoch: [149][ 380/ 1236] Overall Loss 0.140819 Objective Loss 0.140819 LR 0.000250 Time 0.022907 +2023-10-02 21:40:51,179 - Epoch: [149][ 390/ 1236] Overall Loss 0.140534 Objective Loss 0.140534 LR 0.000250 Time 0.022879 +2023-10-02 21:40:51,395 - Epoch: [149][ 400/ 1236] Overall Loss 0.140290 Objective Loss 0.140290 LR 0.000250 Time 0.022847 +2023-10-02 21:40:51,616 - Epoch: [149][ 410/ 1236] Overall Loss 0.140128 Objective Loss 0.140128 LR 0.000250 Time 0.022828 +2023-10-02 21:40:51,831 - Epoch: [149][ 420/ 1236] Overall Loss 0.139838 Objective Loss 0.139838 LR 0.000250 Time 0.022797 +2023-10-02 21:40:52,049 - Epoch: [149][ 430/ 1236] Overall Loss 0.139687 Objective Loss 0.139687 LR 0.000250 Time 0.022771 +2023-10-02 21:40:52,264 - Epoch: [149][ 440/ 1236] Overall Loss 0.139824 Objective Loss 0.139824 LR 0.000250 Time 0.022743 +2023-10-02 21:40:52,485 - Epoch: [149][ 450/ 1236] Overall Loss 0.139779 Objective Loss 0.139779 LR 0.000250 Time 0.022727 +2023-10-02 21:40:52,700 - Epoch: [149][ 460/ 1236] Overall Loss 0.140058 Objective Loss 0.140058 LR 0.000250 Time 0.022701 +2023-10-02 21:40:52,921 - Epoch: [149][ 470/ 1236] Overall Loss 0.140125 Objective Loss 0.140125 LR 0.000250 Time 0.022687 +2023-10-02 21:40:53,137 - Epoch: [149][ 480/ 1236] Overall Loss 0.140168 Objective Loss 0.140168 LR 0.000250 Time 0.022663 +2023-10-02 21:40:53,361 - Epoch: [149][ 490/ 1236] Overall Loss 0.140044 Objective Loss 0.140044 LR 0.000250 Time 0.022656 +2023-10-02 21:40:53,576 - Epoch: [149][ 500/ 1236] Overall Loss 0.139711 Objective Loss 0.139711 LR 0.000250 Time 0.022633 +2023-10-02 21:40:53,797 - Epoch: [149][ 510/ 1236] Overall Loss 0.139551 Objective Loss 0.139551 LR 0.000250 Time 0.022622 +2023-10-02 21:40:54,013 - Epoch: [149][ 520/ 1236] Overall Loss 0.139508 Objective Loss 0.139508 LR 0.000250 Time 0.022601 +2023-10-02 21:40:54,234 - Epoch: [149][ 530/ 1236] Overall Loss 0.139655 Objective Loss 0.139655 LR 0.000250 Time 0.022590 +2023-10-02 21:40:54,449 - Epoch: [149][ 540/ 1236] Overall Loss 0.139372 Objective Loss 0.139372 LR 0.000250 Time 0.022570 +2023-10-02 21:40:54,670 - Epoch: [149][ 550/ 1236] Overall Loss 0.139562 Objective Loss 0.139562 LR 0.000250 Time 0.022560 +2023-10-02 21:40:54,885 - Epoch: [149][ 560/ 1236] Overall Loss 0.139526 Objective Loss 0.139526 LR 0.000250 Time 0.022542 +2023-10-02 21:40:55,106 - Epoch: [149][ 570/ 1236] Overall Loss 0.139795 Objective Loss 0.139795 LR 0.000250 Time 0.022534 +2023-10-02 21:40:55,322 - Epoch: [149][ 580/ 1236] Overall Loss 0.139769 Objective Loss 0.139769 LR 0.000250 Time 0.022516 +2023-10-02 21:40:55,542 - Epoch: [149][ 590/ 1236] Overall Loss 0.139684 Objective Loss 0.139684 LR 0.000250 Time 0.022508 +2023-10-02 21:40:55,758 - Epoch: [149][ 600/ 1236] Overall Loss 0.139460 Objective Loss 0.139460 LR 0.000250 Time 0.022491 +2023-10-02 21:40:55,979 - Epoch: [149][ 610/ 1236] Overall Loss 0.139487 Objective Loss 0.139487 LR 0.000250 Time 0.022484 +2023-10-02 21:40:56,194 - Epoch: [149][ 620/ 1236] Overall Loss 0.139697 Objective Loss 0.139697 LR 0.000250 Time 0.022468 +2023-10-02 21:40:56,415 - Epoch: [149][ 630/ 1236] Overall Loss 0.139690 Objective Loss 0.139690 LR 0.000250 Time 0.022462 +2023-10-02 21:40:56,631 - Epoch: [149][ 640/ 1236] Overall Loss 0.139642 Objective Loss 0.139642 LR 0.000250 Time 0.022447 +2023-10-02 21:40:56,852 - Epoch: [149][ 650/ 1236] Overall Loss 0.139931 Objective Loss 0.139931 LR 0.000250 Time 0.022441 +2023-10-02 21:40:57,067 - Epoch: [149][ 660/ 1236] Overall Loss 0.139604 Objective Loss 0.139604 LR 0.000250 Time 0.022427 +2023-10-02 21:40:57,288 - Epoch: [149][ 670/ 1236] Overall Loss 0.139767 Objective Loss 0.139767 LR 0.000250 Time 0.022422 +2023-10-02 21:40:57,504 - Epoch: [149][ 680/ 1236] Overall Loss 0.139650 Objective Loss 0.139650 LR 0.000250 Time 0.022408 +2023-10-02 21:40:57,725 - Epoch: [149][ 690/ 1236] Overall Loss 0.139751 Objective Loss 0.139751 LR 0.000250 Time 0.022403 +2023-10-02 21:40:57,940 - Epoch: [149][ 700/ 1236] Overall Loss 0.139585 Objective Loss 0.139585 LR 0.000250 Time 0.022390 +2023-10-02 21:40:58,161 - Epoch: [149][ 710/ 1236] Overall Loss 0.139932 Objective Loss 0.139932 LR 0.000250 Time 0.022385 +2023-10-02 21:40:58,376 - Epoch: [149][ 720/ 1236] Overall Loss 0.140057 Objective Loss 0.140057 LR 0.000250 Time 0.022373 +2023-10-02 21:40:58,587 - Epoch: [149][ 730/ 1236] Overall Loss 0.140071 Objective Loss 0.140071 LR 0.000250 Time 0.022354 +2023-10-02 21:40:58,794 - Epoch: [149][ 740/ 1236] Overall Loss 0.140150 Objective Loss 0.140150 LR 0.000250 Time 0.022333 +2023-10-02 21:40:59,005 - Epoch: [149][ 750/ 1236] Overall Loss 0.139955 Objective Loss 0.139955 LR 0.000250 Time 0.022315 +2023-10-02 21:40:59,213 - Epoch: [149][ 760/ 1236] Overall Loss 0.140028 Objective Loss 0.140028 LR 0.000250 Time 0.022295 +2023-10-02 21:40:59,424 - Epoch: [149][ 770/ 1236] Overall Loss 0.140053 Objective Loss 0.140053 LR 0.000250 Time 0.022279 +2023-10-02 21:40:59,632 - Epoch: [149][ 780/ 1236] Overall Loss 0.140222 Objective Loss 0.140222 LR 0.000250 Time 0.022259 +2023-10-02 21:40:59,842 - Epoch: [149][ 790/ 1236] Overall Loss 0.140230 Objective Loss 0.140230 LR 0.000250 Time 0.022244 +2023-10-02 21:41:00,050 - Epoch: [149][ 800/ 1236] Overall Loss 0.140390 Objective Loss 0.140390 LR 0.000250 Time 0.022225 +2023-10-02 21:41:00,261 - Epoch: [149][ 810/ 1236] Overall Loss 0.140296 Objective Loss 0.140296 LR 0.000250 Time 0.022211 +2023-10-02 21:41:00,469 - Epoch: [149][ 820/ 1236] Overall Loss 0.140530 Objective Loss 0.140530 LR 0.000250 Time 0.022193 +2023-10-02 21:41:00,680 - Epoch: [149][ 830/ 1236] Overall Loss 0.140445 Objective Loss 0.140445 LR 0.000250 Time 0.022179 +2023-10-02 21:41:00,888 - Epoch: [149][ 840/ 1236] Overall Loss 0.140355 Objective Loss 0.140355 LR 0.000250 Time 0.022162 +2023-10-02 21:41:01,098 - Epoch: [149][ 850/ 1236] Overall Loss 0.140425 Objective Loss 0.140425 LR 0.000250 Time 0.022149 +2023-10-02 21:41:01,306 - Epoch: [149][ 860/ 1236] Overall Loss 0.140492 Objective Loss 0.140492 LR 0.000250 Time 0.022132 +2023-10-02 21:41:01,519 - Epoch: [149][ 870/ 1236] Overall Loss 0.140381 Objective Loss 0.140381 LR 0.000250 Time 0.022123 +2023-10-02 21:41:01,731 - Epoch: [149][ 880/ 1236] Overall Loss 0.140494 Objective Loss 0.140494 LR 0.000250 Time 0.022112 +2023-10-02 21:41:01,943 - Epoch: [149][ 890/ 1236] Overall Loss 0.140515 Objective Loss 0.140515 LR 0.000250 Time 0.022101 +2023-10-02 21:41:02,156 - Epoch: [149][ 900/ 1236] Overall Loss 0.140400 Objective Loss 0.140400 LR 0.000250 Time 0.022091 +2023-10-02 21:41:02,367 - Epoch: [149][ 910/ 1236] Overall Loss 0.140327 Objective Loss 0.140327 LR 0.000250 Time 0.022081 +2023-10-02 21:41:02,579 - Epoch: [149][ 920/ 1236] Overall Loss 0.140330 Objective Loss 0.140330 LR 0.000250 Time 0.022070 +2023-10-02 21:41:02,791 - Epoch: [149][ 930/ 1236] Overall Loss 0.140257 Objective Loss 0.140257 LR 0.000250 Time 0.022060 +2023-10-02 21:41:03,002 - Epoch: [149][ 940/ 1236] Overall Loss 0.140020 Objective Loss 0.140020 LR 0.000250 Time 0.022051 +2023-10-02 21:41:03,214 - Epoch: [149][ 950/ 1236] Overall Loss 0.140156 Objective Loss 0.140156 LR 0.000250 Time 0.022041 +2023-10-02 21:41:03,426 - Epoch: [149][ 960/ 1236] Overall Loss 0.140241 Objective Loss 0.140241 LR 0.000250 Time 0.022032 +2023-10-02 21:41:03,638 - Epoch: [149][ 970/ 1236] Overall Loss 0.140312 Objective Loss 0.140312 LR 0.000250 Time 0.022022 +2023-10-02 21:41:03,849 - Epoch: [149][ 980/ 1236] Overall Loss 0.140107 Objective Loss 0.140107 LR 0.000250 Time 0.022013 +2023-10-02 21:41:04,059 - Epoch: [149][ 990/ 1236] Overall Loss 0.140006 Objective Loss 0.140006 LR 0.000250 Time 0.022002 +2023-10-02 21:41:04,268 - Epoch: [149][ 1000/ 1236] Overall Loss 0.140235 Objective Loss 0.140235 LR 0.000250 Time 0.021991 +2023-10-02 21:41:04,478 - Epoch: [149][ 1010/ 1236] Overall Loss 0.140264 Objective Loss 0.140264 LR 0.000250 Time 0.021981 +2023-10-02 21:41:04,687 - Epoch: [149][ 1020/ 1236] Overall Loss 0.140161 Objective Loss 0.140161 LR 0.000250 Time 0.021970 +2023-10-02 21:41:04,898 - Epoch: [149][ 1030/ 1236] Overall Loss 0.140268 Objective Loss 0.140268 LR 0.000250 Time 0.021961 +2023-10-02 21:41:05,107 - Epoch: [149][ 1040/ 1236] Overall Loss 0.140480 Objective Loss 0.140480 LR 0.000250 Time 0.021951 +2023-10-02 21:41:05,318 - Epoch: [149][ 1050/ 1236] Overall Loss 0.140512 Objective Loss 0.140512 LR 0.000250 Time 0.021942 +2023-10-02 21:41:05,527 - Epoch: [149][ 1060/ 1236] Overall Loss 0.140274 Objective Loss 0.140274 LR 0.000250 Time 0.021932 +2023-10-02 21:41:05,737 - Epoch: [149][ 1070/ 1236] Overall Loss 0.140479 Objective Loss 0.140479 LR 0.000250 Time 0.021923 +2023-10-02 21:41:05,946 - Epoch: [149][ 1080/ 1236] Overall Loss 0.140486 Objective Loss 0.140486 LR 0.000250 Time 0.021913 +2023-10-02 21:41:06,157 - Epoch: [149][ 1090/ 1236] Overall Loss 0.140563 Objective Loss 0.140563 LR 0.000250 Time 0.021905 +2023-10-02 21:41:06,366 - Epoch: [149][ 1100/ 1236] Overall Loss 0.140487 Objective Loss 0.140487 LR 0.000250 Time 0.021896 +2023-10-02 21:41:06,576 - Epoch: [149][ 1110/ 1236] Overall Loss 0.140616 Objective Loss 0.140616 LR 0.000250 Time 0.021887 +2023-10-02 21:41:06,785 - Epoch: [149][ 1120/ 1236] Overall Loss 0.140631 Objective Loss 0.140631 LR 0.000250 Time 0.021879 +2023-10-02 21:41:06,996 - Epoch: [149][ 1130/ 1236] Overall Loss 0.140520 Objective Loss 0.140520 LR 0.000250 Time 0.021871 +2023-10-02 21:41:07,205 - Epoch: [149][ 1140/ 1236] Overall Loss 0.140554 Objective Loss 0.140554 LR 0.000250 Time 0.021863 +2023-10-02 21:41:07,416 - Epoch: [149][ 1150/ 1236] Overall Loss 0.140425 Objective Loss 0.140425 LR 0.000250 Time 0.021855 +2023-10-02 21:41:07,625 - Epoch: [149][ 1160/ 1236] Overall Loss 0.140518 Objective Loss 0.140518 LR 0.000250 Time 0.021847 +2023-10-02 21:41:07,835 - Epoch: [149][ 1170/ 1236] Overall Loss 0.140405 Objective Loss 0.140405 LR 0.000250 Time 0.021840 +2023-10-02 21:41:08,045 - Epoch: [149][ 1180/ 1236] Overall Loss 0.140358 Objective Loss 0.140358 LR 0.000250 Time 0.021832 +2023-10-02 21:41:08,255 - Epoch: [149][ 1190/ 1236] Overall Loss 0.140219 Objective Loss 0.140219 LR 0.000250 Time 0.021825 +2023-10-02 21:41:08,464 - Epoch: [149][ 1200/ 1236] Overall Loss 0.140236 Objective Loss 0.140236 LR 0.000250 Time 0.021817 +2023-10-02 21:41:08,675 - Epoch: [149][ 1210/ 1236] Overall Loss 0.140183 Objective Loss 0.140183 LR 0.000250 Time 0.021810 +2023-10-02 21:41:08,884 - Epoch: [149][ 1220/ 1236] Overall Loss 0.140029 Objective Loss 0.140029 LR 0.000250 Time 0.021803 +2023-10-02 21:41:09,148 - Epoch: [149][ 1230/ 1236] Overall Loss 0.140000 Objective Loss 0.140000 LR 0.000250 Time 0.021840 +2023-10-02 21:41:09,271 - Epoch: [149][ 1236/ 1236] Overall Loss 0.140027 Objective Loss 0.140027 Top1 91.242363 Top5 99.389002 LR 0.000250 Time 0.021833 +2023-10-02 21:41:09,413 - --- validate (epoch=149)----------- +2023-10-02 21:41:09,413 - 29943 samples (256 per mini-batch) +2023-10-02 21:41:09,916 - Epoch: [149][ 10/ 117] Loss 0.293315 Top1 86.875000 Top5 98.671875 +2023-10-02 21:41:10,071 - Epoch: [149][ 20/ 117] Loss 0.302133 Top1 86.679688 Top5 98.593750 +2023-10-02 21:41:10,225 - Epoch: [149][ 30/ 117] Loss 0.297709 Top1 87.187500 Top5 98.710938 +2023-10-02 21:41:10,379 - Epoch: [149][ 40/ 117] Loss 0.301201 Top1 86.914062 Top5 98.681641 +2023-10-02 21:41:10,533 - Epoch: [149][ 50/ 117] Loss 0.294424 Top1 86.804688 Top5 98.695312 +2023-10-02 21:41:10,691 - Epoch: [149][ 60/ 117] Loss 0.296438 Top1 86.705729 Top5 98.626302 +2023-10-02 21:41:10,850 - Epoch: [149][ 70/ 117] Loss 0.301674 Top1 86.640625 Top5 98.582589 +2023-10-02 21:41:11,011 - Epoch: [149][ 80/ 117] Loss 0.297945 Top1 86.738281 Top5 98.593750 +2023-10-02 21:41:11,169 - Epoch: [149][ 90/ 117] Loss 0.299115 Top1 86.770833 Top5 98.606771 +2023-10-02 21:41:11,329 - Epoch: [149][ 100/ 117] Loss 0.294950 Top1 86.925781 Top5 98.640625 +2023-10-02 21:41:11,495 - Epoch: [149][ 110/ 117] Loss 0.298176 Top1 86.825284 Top5 98.650568 +2023-10-02 21:41:11,584 - Epoch: [149][ 117/ 117] Loss 0.297999 Top1 86.818288 Top5 98.664129 +2023-10-02 21:41:11,705 - ==> Top1: 86.818 Top5: 98.664 Loss: 0.298 + +2023-10-02 21:41:11,706 - ==> Confusion: +[[ 952 2 4 0 7 2 0 0 6 44 3 0 1 1 6 1 3 0 0 0 18] + [ 0 1060 1 0 2 25 1 21 0 1 0 1 0 0 0 3 1 0 9 1 5] + [ 3 1 984 5 1 0 16 9 0 0 2 1 10 2 0 3 1 2 6 2 8] + [ 0 3 15 986 0 0 3 1 4 0 3 0 6 5 22 3 1 3 8 2 24] + [ 26 6 1 1 963 8 0 0 1 10 1 0 0 5 7 6 9 0 0 0 6] + [ 3 27 0 2 2 1005 1 20 1 5 3 7 1 8 5 1 3 0 2 2 18] + [ 0 2 25 0 0 1 1139 6 0 0 3 1 0 0 0 3 0 0 1 6 4] + [ 1 6 11 0 5 30 5 1076 1 2 7 6 5 5 2 0 1 0 32 9 14] + [ 18 0 0 1 3 3 0 1 976 36 10 0 3 15 13 0 6 0 2 0 2] + [ 113 1 3 0 7 5 0 0 29 923 0 1 0 23 5 1 1 1 0 0 6] + [ 2 3 10 6 0 1 2 2 13 1 974 1 0 14 2 0 4 2 5 1 10] + [ 0 0 0 0 0 15 0 2 0 0 0 961 23 7 0 2 0 15 0 4 6] + [ 0 1 1 3 0 1 1 0 0 1 4 38 970 2 2 9 2 12 2 5 14] + [ 0 0 0 0 2 8 0 0 6 7 5 5 0 1066 4 0 0 1 0 1 14] + [ 9 0 6 18 5 1 0 0 20 2 3 0 4 4 1006 0 0 1 11 0 11] + [ 0 0 1 0 5 0 1 0 0 0 1 6 6 0 1 1068 18 11 0 8 8] + [ 1 11 0 0 4 5 1 0 0 0 0 6 0 2 4 9 1102 0 1 4 11] + [ 1 0 1 1 1 0 2 0 0 1 0 3 16 1 1 2 0 1000 0 1 7] + [ 2 5 6 17 1 1 1 23 3 0 2 0 4 0 7 0 0 1 982 0 13] + [ 0 1 3 1 1 4 7 5 0 1 0 17 3 3 1 1 7 1 1 1085 10] + [ 104 114 112 65 53 134 30 84 84 52 169 79 297 278 113 44 91 48 97 139 5718]] + +2023-10-02 21:41:11,707 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:41:11,707 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:41:11,713 - + +2023-10-02 21:41:11,713 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:41:12,750 - Epoch: [150][ 10/ 1236] Overall Loss 0.142007 Objective Loss 0.142007 LR 0.000250 Time 0.103642 +2023-10-02 21:41:12,960 - Epoch: [150][ 20/ 1236] Overall Loss 0.137998 Objective Loss 0.137998 LR 0.000250 Time 0.062306 +2023-10-02 21:41:13,169 - Epoch: [150][ 30/ 1236] Overall Loss 0.135861 Objective Loss 0.135861 LR 0.000250 Time 0.048433 +2023-10-02 21:41:13,380 - Epoch: [150][ 40/ 1236] Overall Loss 0.138147 Objective Loss 0.138147 LR 0.000250 Time 0.041594 +2023-10-02 21:41:13,588 - Epoch: [150][ 50/ 1236] Overall Loss 0.134931 Objective Loss 0.134931 LR 0.000250 Time 0.037440 +2023-10-02 21:41:13,799 - Epoch: [150][ 60/ 1236] Overall Loss 0.136581 Objective Loss 0.136581 LR 0.000250 Time 0.034712 +2023-10-02 21:41:14,011 - Epoch: [150][ 70/ 1236] Overall Loss 0.136251 Objective Loss 0.136251 LR 0.000250 Time 0.032751 +2023-10-02 21:41:14,222 - Epoch: [150][ 80/ 1236] Overall Loss 0.136236 Objective Loss 0.136236 LR 0.000250 Time 0.031290 +2023-10-02 21:41:14,433 - Epoch: [150][ 90/ 1236] Overall Loss 0.136630 Objective Loss 0.136630 LR 0.000250 Time 0.030144 +2023-10-02 21:41:14,644 - Epoch: [150][ 100/ 1236] Overall Loss 0.138763 Objective Loss 0.138763 LR 0.000250 Time 0.029239 +2023-10-02 21:41:14,856 - Epoch: [150][ 110/ 1236] Overall Loss 0.138078 Objective Loss 0.138078 LR 0.000250 Time 0.028493 +2023-10-02 21:41:15,069 - Epoch: [150][ 120/ 1236] Overall Loss 0.137702 Objective Loss 0.137702 LR 0.000250 Time 0.027896 +2023-10-02 21:41:15,281 - Epoch: [150][ 130/ 1236] Overall Loss 0.139316 Objective Loss 0.139316 LR 0.000250 Time 0.027374 +2023-10-02 21:41:15,495 - Epoch: [150][ 140/ 1236] Overall Loss 0.139867 Objective Loss 0.139867 LR 0.000250 Time 0.026944 +2023-10-02 21:41:15,706 - Epoch: [150][ 150/ 1236] Overall Loss 0.139729 Objective Loss 0.139729 LR 0.000250 Time 0.026553 +2023-10-02 21:41:15,920 - Epoch: [150][ 160/ 1236] Overall Loss 0.140201 Objective Loss 0.140201 LR 0.000250 Time 0.026229 +2023-10-02 21:41:16,131 - Epoch: [150][ 170/ 1236] Overall Loss 0.140574 Objective Loss 0.140574 LR 0.000250 Time 0.025926 +2023-10-02 21:41:16,344 - Epoch: [150][ 180/ 1236] Overall Loss 0.140488 Objective Loss 0.140488 LR 0.000250 Time 0.025672 +2023-10-02 21:41:16,556 - Epoch: [150][ 190/ 1236] Overall Loss 0.140097 Objective Loss 0.140097 LR 0.000250 Time 0.025431 +2023-10-02 21:41:16,770 - Epoch: [150][ 200/ 1236] Overall Loss 0.139300 Objective Loss 0.139300 LR 0.000250 Time 0.025228 +2023-10-02 21:41:16,981 - Epoch: [150][ 210/ 1236] Overall Loss 0.138208 Objective Loss 0.138208 LR 0.000250 Time 0.025030 +2023-10-02 21:41:17,193 - Epoch: [150][ 220/ 1236] Overall Loss 0.138501 Objective Loss 0.138501 LR 0.000250 Time 0.024855 +2023-10-02 21:41:17,403 - Epoch: [150][ 230/ 1236] Overall Loss 0.138297 Objective Loss 0.138297 LR 0.000250 Time 0.024683 +2023-10-02 21:41:17,614 - Epoch: [150][ 240/ 1236] Overall Loss 0.138107 Objective Loss 0.138107 LR 0.000250 Time 0.024533 +2023-10-02 21:41:17,822 - Epoch: [150][ 250/ 1236] Overall Loss 0.138348 Objective Loss 0.138348 LR 0.000250 Time 0.024375 +2023-10-02 21:41:18,031 - Epoch: [150][ 260/ 1236] Overall Loss 0.138466 Objective Loss 0.138466 LR 0.000250 Time 0.024242 +2023-10-02 21:41:18,239 - Epoch: [150][ 270/ 1236] Overall Loss 0.138421 Objective Loss 0.138421 LR 0.000250 Time 0.024107 +2023-10-02 21:41:18,448 - Epoch: [150][ 280/ 1236] Overall Loss 0.138393 Objective Loss 0.138393 LR 0.000250 Time 0.023994 +2023-10-02 21:41:18,656 - Epoch: [150][ 290/ 1236] Overall Loss 0.138559 Objective Loss 0.138559 LR 0.000250 Time 0.023876 +2023-10-02 21:41:18,865 - Epoch: [150][ 300/ 1236] Overall Loss 0.138875 Objective Loss 0.138875 LR 0.000250 Time 0.023777 +2023-10-02 21:41:19,072 - Epoch: [150][ 310/ 1236] Overall Loss 0.139219 Objective Loss 0.139219 LR 0.000250 Time 0.023674 +2023-10-02 21:41:19,282 - Epoch: [150][ 320/ 1236] Overall Loss 0.139497 Objective Loss 0.139497 LR 0.000250 Time 0.023588 +2023-10-02 21:41:19,489 - Epoch: [150][ 330/ 1236] Overall Loss 0.139818 Objective Loss 0.139818 LR 0.000250 Time 0.023498 +2023-10-02 21:41:19,699 - Epoch: [150][ 340/ 1236] Overall Loss 0.139714 Objective Loss 0.139714 LR 0.000250 Time 0.023423 +2023-10-02 21:41:19,908 - Epoch: [150][ 350/ 1236] Overall Loss 0.139331 Objective Loss 0.139331 LR 0.000250 Time 0.023351 +2023-10-02 21:41:20,120 - Epoch: [150][ 360/ 1236] Overall Loss 0.139209 Objective Loss 0.139209 LR 0.000250 Time 0.023289 +2023-10-02 21:41:20,328 - Epoch: [150][ 370/ 1236] Overall Loss 0.139479 Objective Loss 0.139479 LR 0.000250 Time 0.023220 +2023-10-02 21:41:20,539 - Epoch: [150][ 380/ 1236] Overall Loss 0.139060 Objective Loss 0.139060 LR 0.000250 Time 0.023164 +2023-10-02 21:41:20,746 - Epoch: [150][ 390/ 1236] Overall Loss 0.139189 Objective Loss 0.139189 LR 0.000250 Time 0.023101 +2023-10-02 21:41:20,958 - Epoch: [150][ 400/ 1236] Overall Loss 0.139593 Objective Loss 0.139593 LR 0.000250 Time 0.023052 +2023-10-02 21:41:21,169 - Epoch: [150][ 410/ 1236] Overall Loss 0.139425 Objective Loss 0.139425 LR 0.000250 Time 0.023003 +2023-10-02 21:41:21,390 - Epoch: [150][ 420/ 1236] Overall Loss 0.139547 Objective Loss 0.139547 LR 0.000250 Time 0.022981 +2023-10-02 21:41:21,605 - Epoch: [150][ 430/ 1236] Overall Loss 0.140241 Objective Loss 0.140241 LR 0.000250 Time 0.022946 +2023-10-02 21:41:21,826 - Epoch: [150][ 440/ 1236] Overall Loss 0.139977 Objective Loss 0.139977 LR 0.000250 Time 0.022926 +2023-10-02 21:41:22,041 - Epoch: [150][ 450/ 1236] Overall Loss 0.139543 Objective Loss 0.139543 LR 0.000250 Time 0.022894 +2023-10-02 21:41:22,260 - Epoch: [150][ 460/ 1236] Overall Loss 0.140051 Objective Loss 0.140051 LR 0.000250 Time 0.022871 +2023-10-02 21:41:22,467 - Epoch: [150][ 470/ 1236] Overall Loss 0.139924 Objective Loss 0.139924 LR 0.000250 Time 0.022825 +2023-10-02 21:41:22,678 - Epoch: [150][ 480/ 1236] Overall Loss 0.140045 Objective Loss 0.140045 LR 0.000250 Time 0.022789 +2023-10-02 21:41:22,886 - Epoch: [150][ 490/ 1236] Overall Loss 0.140016 Objective Loss 0.140016 LR 0.000250 Time 0.022746 +2023-10-02 21:41:23,097 - Epoch: [150][ 500/ 1236] Overall Loss 0.140064 Objective Loss 0.140064 LR 0.000250 Time 0.022713 +2023-10-02 21:41:23,304 - Epoch: [150][ 510/ 1236] Overall Loss 0.139824 Objective Loss 0.139824 LR 0.000250 Time 0.022674 +2023-10-02 21:41:23,516 - Epoch: [150][ 520/ 1236] Overall Loss 0.139853 Objective Loss 0.139853 LR 0.000250 Time 0.022644 +2023-10-02 21:41:23,723 - Epoch: [150][ 530/ 1236] Overall Loss 0.139715 Objective Loss 0.139715 LR 0.000250 Time 0.022607 +2023-10-02 21:41:23,934 - Epoch: [150][ 540/ 1236] Overall Loss 0.139698 Objective Loss 0.139698 LR 0.000250 Time 0.022579 +2023-10-02 21:41:24,141 - Epoch: [150][ 550/ 1236] Overall Loss 0.139926 Objective Loss 0.139926 LR 0.000250 Time 0.022545 +2023-10-02 21:41:24,353 - Epoch: [150][ 560/ 1236] Overall Loss 0.140074 Objective Loss 0.140074 LR 0.000250 Time 0.022519 +2023-10-02 21:41:24,560 - Epoch: [150][ 570/ 1236] Overall Loss 0.140054 Objective Loss 0.140054 LR 0.000250 Time 0.022487 +2023-10-02 21:41:24,771 - Epoch: [150][ 580/ 1236] Overall Loss 0.140000 Objective Loss 0.140000 LR 0.000250 Time 0.022463 +2023-10-02 21:41:24,978 - Epoch: [150][ 590/ 1236] Overall Loss 0.139577 Objective Loss 0.139577 LR 0.000250 Time 0.022433 +2023-10-02 21:41:25,188 - Epoch: [150][ 600/ 1236] Overall Loss 0.139531 Objective Loss 0.139531 LR 0.000250 Time 0.022408 +2023-10-02 21:41:25,396 - Epoch: [150][ 610/ 1236] Overall Loss 0.139532 Objective Loss 0.139532 LR 0.000250 Time 0.022380 +2023-10-02 21:41:25,608 - Epoch: [150][ 620/ 1236] Overall Loss 0.139935 Objective Loss 0.139935 LR 0.000250 Time 0.022359 +2023-10-02 21:41:25,815 - Epoch: [150][ 630/ 1236] Overall Loss 0.139796 Objective Loss 0.139796 LR 0.000250 Time 0.022333 +2023-10-02 21:41:26,027 - Epoch: [150][ 640/ 1236] Overall Loss 0.139894 Objective Loss 0.139894 LR 0.000250 Time 0.022314 +2023-10-02 21:41:26,234 - Epoch: [150][ 650/ 1236] Overall Loss 0.139727 Objective Loss 0.139727 LR 0.000250 Time 0.022290 +2023-10-02 21:41:26,446 - Epoch: [150][ 660/ 1236] Overall Loss 0.139980 Objective Loss 0.139980 LR 0.000250 Time 0.022272 +2023-10-02 21:41:26,653 - Epoch: [150][ 670/ 1236] Overall Loss 0.139933 Objective Loss 0.139933 LR 0.000250 Time 0.022249 +2023-10-02 21:41:26,864 - Epoch: [150][ 680/ 1236] Overall Loss 0.140210 Objective Loss 0.140210 LR 0.000250 Time 0.022232 +2023-10-02 21:41:27,072 - Epoch: [150][ 690/ 1236] Overall Loss 0.140269 Objective Loss 0.140269 LR 0.000250 Time 0.022210 +2023-10-02 21:41:27,283 - Epoch: [150][ 700/ 1236] Overall Loss 0.140466 Objective Loss 0.140466 LR 0.000250 Time 0.022194 +2023-10-02 21:41:27,490 - Epoch: [150][ 710/ 1236] Overall Loss 0.140751 Objective Loss 0.140751 LR 0.000250 Time 0.022173 +2023-10-02 21:41:27,702 - Epoch: [150][ 720/ 1236] Overall Loss 0.140566 Objective Loss 0.140566 LR 0.000250 Time 0.022158 +2023-10-02 21:41:27,909 - Epoch: [150][ 730/ 1236] Overall Loss 0.140338 Objective Loss 0.140338 LR 0.000250 Time 0.022138 +2023-10-02 21:41:28,119 - Epoch: [150][ 740/ 1236] Overall Loss 0.140592 Objective Loss 0.140592 LR 0.000250 Time 0.022123 +2023-10-02 21:41:28,328 - Epoch: [150][ 750/ 1236] Overall Loss 0.140867 Objective Loss 0.140867 LR 0.000250 Time 0.022104 +2023-10-02 21:41:28,539 - Epoch: [150][ 760/ 1236] Overall Loss 0.140773 Objective Loss 0.140773 LR 0.000250 Time 0.022090 +2023-10-02 21:41:28,748 - Epoch: [150][ 770/ 1236] Overall Loss 0.140838 Objective Loss 0.140838 LR 0.000250 Time 0.022075 +2023-10-02 21:41:28,958 - Epoch: [150][ 780/ 1236] Overall Loss 0.140940 Objective Loss 0.140940 LR 0.000250 Time 0.022061 +2023-10-02 21:41:29,168 - Epoch: [150][ 790/ 1236] Overall Loss 0.141219 Objective Loss 0.141219 LR 0.000250 Time 0.022045 +2023-10-02 21:41:29,377 - Epoch: [150][ 800/ 1236] Overall Loss 0.141333 Objective Loss 0.141333 LR 0.000250 Time 0.022031 +2023-10-02 21:41:29,587 - Epoch: [150][ 810/ 1236] Overall Loss 0.141486 Objective Loss 0.141486 LR 0.000250 Time 0.022016 +2023-10-02 21:41:29,797 - Epoch: [150][ 820/ 1236] Overall Loss 0.141526 Objective Loss 0.141526 LR 0.000250 Time 0.022004 +2023-10-02 21:41:30,007 - Epoch: [150][ 830/ 1236] Overall Loss 0.141597 Objective Loss 0.141597 LR 0.000250 Time 0.021989 +2023-10-02 21:41:30,217 - Epoch: [150][ 840/ 1236] Overall Loss 0.141446 Objective Loss 0.141446 LR 0.000250 Time 0.021977 +2023-10-02 21:41:30,426 - Epoch: [150][ 850/ 1236] Overall Loss 0.141301 Objective Loss 0.141301 LR 0.000250 Time 0.021963 +2023-10-02 21:41:30,636 - Epoch: [150][ 860/ 1236] Overall Loss 0.141402 Objective Loss 0.141402 LR 0.000250 Time 0.021951 +2023-10-02 21:41:30,846 - Epoch: [150][ 870/ 1236] Overall Loss 0.141262 Objective Loss 0.141262 LR 0.000250 Time 0.021938 +2023-10-02 21:41:31,055 - Epoch: [150][ 880/ 1236] Overall Loss 0.141119 Objective Loss 0.141119 LR 0.000250 Time 0.021927 +2023-10-02 21:41:31,265 - Epoch: [150][ 890/ 1236] Overall Loss 0.141142 Objective Loss 0.141142 LR 0.000250 Time 0.021914 +2023-10-02 21:41:31,475 - Epoch: [150][ 900/ 1236] Overall Loss 0.141332 Objective Loss 0.141332 LR 0.000250 Time 0.021904 +2023-10-02 21:41:31,685 - Epoch: [150][ 910/ 1236] Overall Loss 0.141330 Objective Loss 0.141330 LR 0.000250 Time 0.021892 +2023-10-02 21:41:31,895 - Epoch: [150][ 920/ 1236] Overall Loss 0.141415 Objective Loss 0.141415 LR 0.000250 Time 0.021882 +2023-10-02 21:41:32,104 - Epoch: [150][ 930/ 1236] Overall Loss 0.141371 Objective Loss 0.141371 LR 0.000250 Time 0.021871 +2023-10-02 21:41:32,314 - Epoch: [150][ 940/ 1236] Overall Loss 0.141220 Objective Loss 0.141220 LR 0.000250 Time 0.021861 +2023-10-02 21:41:32,524 - Epoch: [150][ 950/ 1236] Overall Loss 0.141320 Objective Loss 0.141320 LR 0.000250 Time 0.021850 +2023-10-02 21:41:32,733 - Epoch: [150][ 960/ 1236] Overall Loss 0.141114 Objective Loss 0.141114 LR 0.000250 Time 0.021841 +2023-10-02 21:41:32,943 - Epoch: [150][ 970/ 1236] Overall Loss 0.141042 Objective Loss 0.141042 LR 0.000250 Time 0.021831 +2023-10-02 21:41:33,153 - Epoch: [150][ 980/ 1236] Overall Loss 0.141030 Objective Loss 0.141030 LR 0.000250 Time 0.021822 +2023-10-02 21:41:33,362 - Epoch: [150][ 990/ 1236] Overall Loss 0.141194 Objective Loss 0.141194 LR 0.000250 Time 0.021811 +2023-10-02 21:41:33,572 - Epoch: [150][ 1000/ 1236] Overall Loss 0.141074 Objective Loss 0.141074 LR 0.000250 Time 0.021803 +2023-10-02 21:41:33,782 - Epoch: [150][ 1010/ 1236] Overall Loss 0.140875 Objective Loss 0.140875 LR 0.000250 Time 0.021793 +2023-10-02 21:41:33,992 - Epoch: [150][ 1020/ 1236] Overall Loss 0.140706 Objective Loss 0.140706 LR 0.000250 Time 0.021785 +2023-10-02 21:41:34,201 - Epoch: [150][ 1030/ 1236] Overall Loss 0.140642 Objective Loss 0.140642 LR 0.000250 Time 0.021775 +2023-10-02 21:41:34,412 - Epoch: [150][ 1040/ 1236] Overall Loss 0.140753 Objective Loss 0.140753 LR 0.000250 Time 0.021768 +2023-10-02 21:41:34,621 - Epoch: [150][ 1050/ 1236] Overall Loss 0.140533 Objective Loss 0.140533 LR 0.000250 Time 0.021758 +2023-10-02 21:41:34,831 - Epoch: [150][ 1060/ 1236] Overall Loss 0.140461 Objective Loss 0.140461 LR 0.000250 Time 0.021751 +2023-10-02 21:41:35,041 - Epoch: [150][ 1070/ 1236] Overall Loss 0.140516 Objective Loss 0.140516 LR 0.000250 Time 0.021742 +2023-10-02 21:41:35,251 - Epoch: [150][ 1080/ 1236] Overall Loss 0.140416 Objective Loss 0.140416 LR 0.000250 Time 0.021735 +2023-10-02 21:41:35,461 - Epoch: [150][ 1090/ 1236] Overall Loss 0.140275 Objective Loss 0.140275 LR 0.000250 Time 0.021727 +2023-10-02 21:41:35,671 - Epoch: [150][ 1100/ 1236] Overall Loss 0.140279 Objective Loss 0.140279 LR 0.000250 Time 0.021720 +2023-10-02 21:41:35,881 - Epoch: [150][ 1110/ 1236] Overall Loss 0.140377 Objective Loss 0.140377 LR 0.000250 Time 0.021712 +2023-10-02 21:41:36,091 - Epoch: [150][ 1120/ 1236] Overall Loss 0.140279 Objective Loss 0.140279 LR 0.000250 Time 0.021705 +2023-10-02 21:41:36,301 - Epoch: [150][ 1130/ 1236] Overall Loss 0.140199 Objective Loss 0.140199 LR 0.000250 Time 0.021698 +2023-10-02 21:41:36,511 - Epoch: [150][ 1140/ 1236] Overall Loss 0.140194 Objective Loss 0.140194 LR 0.000250 Time 0.021691 +2023-10-02 21:41:36,720 - Epoch: [150][ 1150/ 1236] Overall Loss 0.140263 Objective Loss 0.140263 LR 0.000250 Time 0.021684 +2023-10-02 21:41:36,931 - Epoch: [150][ 1160/ 1236] Overall Loss 0.140295 Objective Loss 0.140295 LR 0.000250 Time 0.021678 +2023-10-02 21:41:37,140 - Epoch: [150][ 1170/ 1236] Overall Loss 0.140330 Objective Loss 0.140330 LR 0.000250 Time 0.021670 +2023-10-02 21:41:37,351 - Epoch: [150][ 1180/ 1236] Overall Loss 0.140262 Objective Loss 0.140262 LR 0.000250 Time 0.021664 +2023-10-02 21:41:37,560 - Epoch: [150][ 1190/ 1236] Overall Loss 0.140235 Objective Loss 0.140235 LR 0.000250 Time 0.021657 +2023-10-02 21:41:37,771 - Epoch: [150][ 1200/ 1236] Overall Loss 0.140427 Objective Loss 0.140427 LR 0.000250 Time 0.021652 +2023-10-02 21:41:37,981 - Epoch: [150][ 1210/ 1236] Overall Loss 0.140477 Objective Loss 0.140477 LR 0.000250 Time 0.021645 +2023-10-02 21:41:38,191 - Epoch: [150][ 1220/ 1236] Overall Loss 0.140428 Objective Loss 0.140428 LR 0.000250 Time 0.021640 +2023-10-02 21:41:38,454 - Epoch: [150][ 1230/ 1236] Overall Loss 0.140363 Objective Loss 0.140363 LR 0.000250 Time 0.021677 +2023-10-02 21:41:38,577 - Epoch: [150][ 1236/ 1236] Overall Loss 0.140331 Objective Loss 0.140331 Top1 90.020367 Top5 98.981670 LR 0.000250 Time 0.021671 +2023-10-02 21:41:38,725 - --- validate (epoch=150)----------- +2023-10-02 21:41:38,725 - 29943 samples (256 per mini-batch) +2023-10-02 21:41:39,223 - Epoch: [150][ 10/ 117] Loss 0.335423 Top1 86.562500 Top5 98.554688 +2023-10-02 21:41:39,377 - Epoch: [150][ 20/ 117] Loss 0.336899 Top1 86.347656 Top5 98.398438 +2023-10-02 21:41:39,538 - Epoch: [150][ 30/ 117] Loss 0.327635 Top1 86.210938 Top5 98.476562 +2023-10-02 21:41:39,699 - Epoch: [150][ 40/ 117] Loss 0.320930 Top1 86.289062 Top5 98.349609 +2023-10-02 21:41:39,852 - Epoch: [150][ 50/ 117] Loss 0.310797 Top1 86.468750 Top5 98.414062 +2023-10-02 21:41:40,003 - Epoch: [150][ 60/ 117] Loss 0.305424 Top1 86.549479 Top5 98.528646 +2023-10-02 21:41:40,154 - Epoch: [150][ 70/ 117] Loss 0.303801 Top1 86.601562 Top5 98.537946 +2023-10-02 21:41:40,312 - Epoch: [150][ 80/ 117] Loss 0.304211 Top1 86.557617 Top5 98.549805 +2023-10-02 21:41:40,469 - Epoch: [150][ 90/ 117] Loss 0.305100 Top1 86.770833 Top5 98.572049 +2023-10-02 21:41:40,627 - Epoch: [150][ 100/ 117] Loss 0.298715 Top1 86.785156 Top5 98.593750 +2023-10-02 21:41:40,785 - Epoch: [150][ 110/ 117] Loss 0.296125 Top1 86.921165 Top5 98.622159 +2023-10-02 21:41:40,875 - Epoch: [150][ 117/ 117] Loss 0.297337 Top1 86.868383 Top5 98.607354 +2023-10-02 21:41:41,030 - ==> Top1: 86.868 Top5: 98.607 Loss: 0.297 + +2023-10-02 21:41:41,031 - ==> Confusion: +[[ 960 0 2 0 5 3 0 0 7 56 1 0 0 1 2 1 0 0 0 0 12] + [ 0 1049 0 1 7 22 0 26 0 1 0 2 0 0 1 3 2 0 8 2 7] + [ 4 0 991 7 1 0 15 6 0 1 1 0 8 2 0 4 0 2 7 2 5] + [ 3 3 14 996 1 2 2 2 1 1 5 1 3 4 16 2 1 3 10 2 17] + [ 30 5 1 1 966 5 1 0 3 12 1 0 0 3 5 5 6 0 0 0 6] + [ 3 31 0 2 1 1002 0 24 1 6 2 9 1 10 4 0 3 0 3 1 13] + [ 0 2 26 1 0 2 1129 4 0 0 4 1 0 0 0 3 0 1 1 9 8] + [ 1 5 20 1 6 23 3 1086 1 3 5 2 4 5 3 1 1 2 31 5 10] + [ 19 0 0 1 2 3 0 2 972 46 7 3 1 12 11 0 2 2 2 3 1] + [ 111 1 0 0 5 4 0 0 21 944 1 1 0 18 2 2 0 0 0 2 7] + [ 2 2 10 8 0 1 3 3 13 1 971 1 0 13 2 0 3 2 5 1 12] + [ 0 1 0 0 1 12 0 4 0 0 0 973 13 8 0 3 0 15 0 2 3] + [ 0 0 2 3 1 2 1 0 0 1 4 43 962 3 1 6 2 13 3 6 15] + [ 0 0 1 0 3 5 0 0 8 9 3 4 0 1066 3 0 0 1 0 0 16] + [ 16 0 6 17 3 1 0 0 23 3 0 0 2 6 1000 0 1 3 11 0 9] + [ 0 0 1 1 7 0 0 0 0 0 0 6 7 0 0 1074 14 10 3 6 5] + [ 0 12 1 1 5 6 0 1 0 1 0 4 0 4 5 9 1093 0 1 6 12] + [ 1 1 0 2 0 0 3 0 1 2 0 4 19 2 1 6 0 994 0 0 2] + [ 4 4 3 22 0 1 1 23 5 1 5 0 1 0 3 0 0 1 983 0 11] + [ 0 0 1 2 0 2 8 8 0 0 1 15 3 3 1 2 8 0 1 1092 5] + [ 139 118 128 73 55 122 31 87 76 75 154 87 283 264 98 56 60 53 91 147 5708]] + +2023-10-02 21:41:41,033 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:41:41,033 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:41:41,039 - + +2023-10-02 21:41:41,039 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:41:42,065 - Epoch: [151][ 10/ 1236] Overall Loss 0.133658 Objective Loss 0.133658 LR 0.000250 Time 0.102530 +2023-10-02 21:41:42,278 - Epoch: [151][ 20/ 1236] Overall Loss 0.145180 Objective Loss 0.145180 LR 0.000250 Time 0.061890 +2023-10-02 21:41:42,491 - Epoch: [151][ 30/ 1236] Overall Loss 0.140027 Objective Loss 0.140027 LR 0.000250 Time 0.048363 +2023-10-02 21:41:42,703 - Epoch: [151][ 40/ 1236] Overall Loss 0.141006 Objective Loss 0.141006 LR 0.000250 Time 0.041566 +2023-10-02 21:41:42,916 - Epoch: [151][ 50/ 1236] Overall Loss 0.139903 Objective Loss 0.139903 LR 0.000250 Time 0.037500 +2023-10-02 21:41:43,128 - Epoch: [151][ 60/ 1236] Overall Loss 0.138124 Objective Loss 0.138124 LR 0.000250 Time 0.034784 +2023-10-02 21:41:43,342 - Epoch: [151][ 70/ 1236] Overall Loss 0.136398 Objective Loss 0.136398 LR 0.000250 Time 0.032864 +2023-10-02 21:41:43,555 - Epoch: [151][ 80/ 1236] Overall Loss 0.137391 Objective Loss 0.137391 LR 0.000250 Time 0.031411 +2023-10-02 21:41:43,766 - Epoch: [151][ 90/ 1236] Overall Loss 0.136956 Objective Loss 0.136956 LR 0.000250 Time 0.030270 +2023-10-02 21:41:43,978 - Epoch: [151][ 100/ 1236] Overall Loss 0.138969 Objective Loss 0.138969 LR 0.000250 Time 0.029352 +2023-10-02 21:41:44,186 - Epoch: [151][ 110/ 1236] Overall Loss 0.139879 Objective Loss 0.139879 LR 0.000250 Time 0.028567 +2023-10-02 21:41:44,396 - Epoch: [151][ 120/ 1236] Overall Loss 0.142731 Objective Loss 0.142731 LR 0.000250 Time 0.027927 +2023-10-02 21:41:44,603 - Epoch: [151][ 130/ 1236] Overall Loss 0.141853 Objective Loss 0.141853 LR 0.000250 Time 0.027366 +2023-10-02 21:41:44,813 - Epoch: [151][ 140/ 1236] Overall Loss 0.141835 Objective Loss 0.141835 LR 0.000250 Time 0.026905 +2023-10-02 21:41:45,021 - Epoch: [151][ 150/ 1236] Overall Loss 0.142126 Objective Loss 0.142126 LR 0.000250 Time 0.026487 +2023-10-02 21:41:45,231 - Epoch: [151][ 160/ 1236] Overall Loss 0.142285 Objective Loss 0.142285 LR 0.000250 Time 0.026141 +2023-10-02 21:41:45,439 - Epoch: [151][ 170/ 1236] Overall Loss 0.142766 Objective Loss 0.142766 LR 0.000250 Time 0.025818 +2023-10-02 21:41:45,648 - Epoch: [151][ 180/ 1236] Overall Loss 0.143633 Objective Loss 0.143633 LR 0.000250 Time 0.025545 +2023-10-02 21:41:45,855 - Epoch: [151][ 190/ 1236] Overall Loss 0.144429 Objective Loss 0.144429 LR 0.000250 Time 0.025282 +2023-10-02 21:41:46,065 - Epoch: [151][ 200/ 1236] Overall Loss 0.145126 Objective Loss 0.145126 LR 0.000250 Time 0.025065 +2023-10-02 21:41:46,275 - Epoch: [151][ 210/ 1236] Overall Loss 0.145084 Objective Loss 0.145084 LR 0.000250 Time 0.024866 +2023-10-02 21:41:46,488 - Epoch: [151][ 220/ 1236] Overall Loss 0.144698 Objective Loss 0.144698 LR 0.000250 Time 0.024700 +2023-10-02 21:41:46,697 - Epoch: [151][ 230/ 1236] Overall Loss 0.143758 Objective Loss 0.143758 LR 0.000250 Time 0.024535 +2023-10-02 21:41:46,909 - Epoch: [151][ 240/ 1236] Overall Loss 0.143394 Objective Loss 0.143394 LR 0.000250 Time 0.024398 +2023-10-02 21:41:47,119 - Epoch: [151][ 250/ 1236] Overall Loss 0.143337 Objective Loss 0.143337 LR 0.000250 Time 0.024258 +2023-10-02 21:41:47,331 - Epoch: [151][ 260/ 1236] Overall Loss 0.142718 Objective Loss 0.142718 LR 0.000250 Time 0.024142 +2023-10-02 21:41:47,541 - Epoch: [151][ 270/ 1236] Overall Loss 0.142728 Objective Loss 0.142728 LR 0.000250 Time 0.024021 +2023-10-02 21:41:47,753 - Epoch: [151][ 280/ 1236] Overall Loss 0.142697 Objective Loss 0.142697 LR 0.000250 Time 0.023922 +2023-10-02 21:41:47,963 - Epoch: [151][ 290/ 1236] Overall Loss 0.142381 Objective Loss 0.142381 LR 0.000250 Time 0.023818 +2023-10-02 21:41:48,174 - Epoch: [151][ 300/ 1236] Overall Loss 0.142086 Objective Loss 0.142086 LR 0.000250 Time 0.023727 +2023-10-02 21:41:48,385 - Epoch: [151][ 310/ 1236] Overall Loss 0.141931 Objective Loss 0.141931 LR 0.000250 Time 0.023637 +2023-10-02 21:41:48,597 - Epoch: [151][ 320/ 1236] Overall Loss 0.141857 Objective Loss 0.141857 LR 0.000250 Time 0.023562 +2023-10-02 21:41:48,807 - Epoch: [151][ 330/ 1236] Overall Loss 0.141667 Objective Loss 0.141667 LR 0.000250 Time 0.023482 +2023-10-02 21:41:49,019 - Epoch: [151][ 340/ 1236] Overall Loss 0.142320 Objective Loss 0.142320 LR 0.000250 Time 0.023413 +2023-10-02 21:41:49,228 - Epoch: [151][ 350/ 1236] Overall Loss 0.141577 Objective Loss 0.141577 LR 0.000250 Time 0.023338 +2023-10-02 21:41:49,438 - Epoch: [151][ 360/ 1236] Overall Loss 0.141659 Objective Loss 0.141659 LR 0.000250 Time 0.023273 +2023-10-02 21:41:49,648 - Epoch: [151][ 370/ 1236] Overall Loss 0.141775 Objective Loss 0.141775 LR 0.000250 Time 0.023209 +2023-10-02 21:41:49,858 - Epoch: [151][ 380/ 1236] Overall Loss 0.141960 Objective Loss 0.141960 LR 0.000250 Time 0.023150 +2023-10-02 21:41:50,067 - Epoch: [151][ 390/ 1236] Overall Loss 0.142054 Objective Loss 0.142054 LR 0.000250 Time 0.023093 +2023-10-02 21:41:50,277 - Epoch: [151][ 400/ 1236] Overall Loss 0.141730 Objective Loss 0.141730 LR 0.000250 Time 0.023040 +2023-10-02 21:41:50,487 - Epoch: [151][ 410/ 1236] Overall Loss 0.141918 Objective Loss 0.141918 LR 0.000250 Time 0.022988 +2023-10-02 21:41:50,696 - Epoch: [151][ 420/ 1236] Overall Loss 0.142046 Objective Loss 0.142046 LR 0.000250 Time 0.022940 +2023-10-02 21:41:50,906 - Epoch: [151][ 430/ 1236] Overall Loss 0.141534 Objective Loss 0.141534 LR 0.000250 Time 0.022893 +2023-10-02 21:41:51,116 - Epoch: [151][ 440/ 1236] Overall Loss 0.141835 Objective Loss 0.141835 LR 0.000250 Time 0.022850 +2023-10-02 21:41:51,326 - Epoch: [151][ 450/ 1236] Overall Loss 0.141814 Objective Loss 0.141814 LR 0.000250 Time 0.022807 +2023-10-02 21:41:51,536 - Epoch: [151][ 460/ 1236] Overall Loss 0.141545 Objective Loss 0.141545 LR 0.000250 Time 0.022767 +2023-10-02 21:41:51,745 - Epoch: [151][ 470/ 1236] Overall Loss 0.141707 Objective Loss 0.141707 LR 0.000250 Time 0.022727 +2023-10-02 21:41:51,955 - Epoch: [151][ 480/ 1236] Overall Loss 0.142089 Objective Loss 0.142089 LR 0.000250 Time 0.022691 +2023-10-02 21:41:52,165 - Epoch: [151][ 490/ 1236] Overall Loss 0.142096 Objective Loss 0.142096 LR 0.000250 Time 0.022655 +2023-10-02 21:41:52,375 - Epoch: [151][ 500/ 1236] Overall Loss 0.141863 Objective Loss 0.141863 LR 0.000250 Time 0.022622 +2023-10-02 21:41:52,584 - Epoch: [151][ 510/ 1236] Overall Loss 0.141810 Objective Loss 0.141810 LR 0.000250 Time 0.022588 +2023-10-02 21:41:52,794 - Epoch: [151][ 520/ 1236] Overall Loss 0.141959 Objective Loss 0.141959 LR 0.000250 Time 0.022557 +2023-10-02 21:41:53,004 - Epoch: [151][ 530/ 1236] Overall Loss 0.141998 Objective Loss 0.141998 LR 0.000250 Time 0.022526 +2023-10-02 21:41:53,214 - Epoch: [151][ 540/ 1236] Overall Loss 0.142124 Objective Loss 0.142124 LR 0.000250 Time 0.022498 +2023-10-02 21:41:53,424 - Epoch: [151][ 550/ 1236] Overall Loss 0.141942 Objective Loss 0.141942 LR 0.000250 Time 0.022470 +2023-10-02 21:41:53,634 - Epoch: [151][ 560/ 1236] Overall Loss 0.141913 Objective Loss 0.141913 LR 0.000250 Time 0.022443 +2023-10-02 21:41:53,845 - Epoch: [151][ 570/ 1236] Overall Loss 0.141847 Objective Loss 0.141847 LR 0.000250 Time 0.022418 +2023-10-02 21:41:54,055 - Epoch: [151][ 580/ 1236] Overall Loss 0.142213 Objective Loss 0.142213 LR 0.000250 Time 0.022394 +2023-10-02 21:41:54,265 - Epoch: [151][ 590/ 1236] Overall Loss 0.142003 Objective Loss 0.142003 LR 0.000250 Time 0.022369 +2023-10-02 21:41:54,475 - Epoch: [151][ 600/ 1236] Overall Loss 0.141894 Objective Loss 0.141894 LR 0.000250 Time 0.022347 +2023-10-02 21:41:54,685 - Epoch: [151][ 610/ 1236] Overall Loss 0.142102 Objective Loss 0.142102 LR 0.000250 Time 0.022324 +2023-10-02 21:41:54,895 - Epoch: [151][ 620/ 1236] Overall Loss 0.142036 Objective Loss 0.142036 LR 0.000250 Time 0.022302 +2023-10-02 21:41:55,104 - Epoch: [151][ 630/ 1236] Overall Loss 0.141994 Objective Loss 0.141994 LR 0.000250 Time 0.022279 +2023-10-02 21:41:55,314 - Epoch: [151][ 640/ 1236] Overall Loss 0.142239 Objective Loss 0.142239 LR 0.000250 Time 0.022258 +2023-10-02 21:41:55,522 - Epoch: [151][ 650/ 1236] Overall Loss 0.142117 Objective Loss 0.142117 LR 0.000250 Time 0.022237 +2023-10-02 21:41:55,732 - Epoch: [151][ 660/ 1236] Overall Loss 0.142140 Objective Loss 0.142140 LR 0.000250 Time 0.022217 +2023-10-02 21:41:55,941 - Epoch: [151][ 670/ 1236] Overall Loss 0.142090 Objective Loss 0.142090 LR 0.000250 Time 0.022196 +2023-10-02 21:41:56,150 - Epoch: [151][ 680/ 1236] Overall Loss 0.141931 Objective Loss 0.141931 LR 0.000250 Time 0.022178 +2023-10-02 21:41:56,359 - Epoch: [151][ 690/ 1236] Overall Loss 0.141840 Objective Loss 0.141840 LR 0.000250 Time 0.022158 +2023-10-02 21:41:56,569 - Epoch: [151][ 700/ 1236] Overall Loss 0.141636 Objective Loss 0.141636 LR 0.000250 Time 0.022141 +2023-10-02 21:41:56,777 - Epoch: [151][ 710/ 1236] Overall Loss 0.141554 Objective Loss 0.141554 LR 0.000250 Time 0.022122 +2023-10-02 21:41:56,987 - Epoch: [151][ 720/ 1236] Overall Loss 0.141481 Objective Loss 0.141481 LR 0.000250 Time 0.022105 +2023-10-02 21:41:57,195 - Epoch: [151][ 730/ 1236] Overall Loss 0.141243 Objective Loss 0.141243 LR 0.000250 Time 0.022088 +2023-10-02 21:41:57,405 - Epoch: [151][ 740/ 1236] Overall Loss 0.141277 Objective Loss 0.141277 LR 0.000250 Time 0.022072 +2023-10-02 21:41:57,613 - Epoch: [151][ 750/ 1236] Overall Loss 0.141205 Objective Loss 0.141205 LR 0.000250 Time 0.022055 +2023-10-02 21:41:57,823 - Epoch: [151][ 760/ 1236] Overall Loss 0.141112 Objective Loss 0.141112 LR 0.000250 Time 0.022041 +2023-10-02 21:41:58,031 - Epoch: [151][ 770/ 1236] Overall Loss 0.141041 Objective Loss 0.141041 LR 0.000250 Time 0.022025 +2023-10-02 21:41:58,241 - Epoch: [151][ 780/ 1236] Overall Loss 0.140804 Objective Loss 0.140804 LR 0.000250 Time 0.022011 +2023-10-02 21:41:58,450 - Epoch: [151][ 790/ 1236] Overall Loss 0.140804 Objective Loss 0.140804 LR 0.000250 Time 0.021996 +2023-10-02 21:41:58,660 - Epoch: [151][ 800/ 1236] Overall Loss 0.140675 Objective Loss 0.140675 LR 0.000250 Time 0.021983 +2023-10-02 21:41:58,868 - Epoch: [151][ 810/ 1236] Overall Loss 0.140572 Objective Loss 0.140572 LR 0.000250 Time 0.021969 +2023-10-02 21:41:59,078 - Epoch: [151][ 820/ 1236] Overall Loss 0.140499 Objective Loss 0.140499 LR 0.000250 Time 0.021957 +2023-10-02 21:41:59,287 - Epoch: [151][ 830/ 1236] Overall Loss 0.140519 Objective Loss 0.140519 LR 0.000250 Time 0.021943 +2023-10-02 21:41:59,496 - Epoch: [151][ 840/ 1236] Overall Loss 0.140618 Objective Loss 0.140618 LR 0.000250 Time 0.021930 +2023-10-02 21:41:59,704 - Epoch: [151][ 850/ 1236] Overall Loss 0.140752 Objective Loss 0.140752 LR 0.000250 Time 0.021917 +2023-10-02 21:41:59,913 - Epoch: [151][ 860/ 1236] Overall Loss 0.140694 Objective Loss 0.140694 LR 0.000250 Time 0.021905 +2023-10-02 21:42:00,121 - Epoch: [151][ 870/ 1236] Overall Loss 0.140608 Objective Loss 0.140608 LR 0.000250 Time 0.021892 +2023-10-02 21:42:00,331 - Epoch: [151][ 880/ 1236] Overall Loss 0.140574 Objective Loss 0.140574 LR 0.000250 Time 0.021881 +2023-10-02 21:42:00,539 - Epoch: [151][ 890/ 1236] Overall Loss 0.140706 Objective Loss 0.140706 LR 0.000250 Time 0.021869 +2023-10-02 21:42:00,748 - Epoch: [151][ 900/ 1236] Overall Loss 0.140638 Objective Loss 0.140638 LR 0.000250 Time 0.021858 +2023-10-02 21:42:00,956 - Epoch: [151][ 910/ 1236] Overall Loss 0.140556 Objective Loss 0.140556 LR 0.000250 Time 0.021846 +2023-10-02 21:42:01,166 - Epoch: [151][ 920/ 1236] Overall Loss 0.140511 Objective Loss 0.140511 LR 0.000250 Time 0.021837 +2023-10-02 21:42:01,375 - Epoch: [151][ 930/ 1236] Overall Loss 0.140467 Objective Loss 0.140467 LR 0.000250 Time 0.021826 +2023-10-02 21:42:01,584 - Epoch: [151][ 940/ 1236] Overall Loss 0.140252 Objective Loss 0.140252 LR 0.000250 Time 0.021816 +2023-10-02 21:42:01,792 - Epoch: [151][ 950/ 1236] Overall Loss 0.140012 Objective Loss 0.140012 LR 0.000250 Time 0.021805 +2023-10-02 21:42:02,002 - Epoch: [151][ 960/ 1236] Overall Loss 0.140067 Objective Loss 0.140067 LR 0.000250 Time 0.021795 +2023-10-02 21:42:02,210 - Epoch: [151][ 970/ 1236] Overall Loss 0.140218 Objective Loss 0.140218 LR 0.000250 Time 0.021785 +2023-10-02 21:42:02,419 - Epoch: [151][ 980/ 1236] Overall Loss 0.140053 Objective Loss 0.140053 LR 0.000250 Time 0.021776 +2023-10-02 21:42:02,628 - Epoch: [151][ 990/ 1236] Overall Loss 0.140149 Objective Loss 0.140149 LR 0.000250 Time 0.021766 +2023-10-02 21:42:02,837 - Epoch: [151][ 1000/ 1236] Overall Loss 0.140166 Objective Loss 0.140166 LR 0.000250 Time 0.021758 +2023-10-02 21:42:03,045 - Epoch: [151][ 1010/ 1236] Overall Loss 0.140313 Objective Loss 0.140313 LR 0.000250 Time 0.021748 +2023-10-02 21:42:03,254 - Epoch: [151][ 1020/ 1236] Overall Loss 0.140364 Objective Loss 0.140364 LR 0.000250 Time 0.021740 +2023-10-02 21:42:03,462 - Epoch: [151][ 1030/ 1236] Overall Loss 0.140172 Objective Loss 0.140172 LR 0.000250 Time 0.021731 +2023-10-02 21:42:03,672 - Epoch: [151][ 1040/ 1236] Overall Loss 0.140328 Objective Loss 0.140328 LR 0.000250 Time 0.021723 +2023-10-02 21:42:03,880 - Epoch: [151][ 1050/ 1236] Overall Loss 0.140552 Objective Loss 0.140552 LR 0.000250 Time 0.021714 +2023-10-02 21:42:04,089 - Epoch: [151][ 1060/ 1236] Overall Loss 0.140640 Objective Loss 0.140640 LR 0.000250 Time 0.021705 +2023-10-02 21:42:04,297 - Epoch: [151][ 1070/ 1236] Overall Loss 0.140464 Objective Loss 0.140464 LR 0.000250 Time 0.021697 +2023-10-02 21:42:04,506 - Epoch: [151][ 1080/ 1236] Overall Loss 0.140501 Objective Loss 0.140501 LR 0.000250 Time 0.021689 +2023-10-02 21:42:04,714 - Epoch: [151][ 1090/ 1236] Overall Loss 0.140663 Objective Loss 0.140663 LR 0.000250 Time 0.021681 +2023-10-02 21:42:04,923 - Epoch: [151][ 1100/ 1236] Overall Loss 0.140772 Objective Loss 0.140772 LR 0.000250 Time 0.021674 +2023-10-02 21:42:05,132 - Epoch: [151][ 1110/ 1236] Overall Loss 0.140690 Objective Loss 0.140690 LR 0.000250 Time 0.021666 +2023-10-02 21:42:05,341 - Epoch: [151][ 1120/ 1236] Overall Loss 0.140514 Objective Loss 0.140514 LR 0.000250 Time 0.021659 +2023-10-02 21:42:05,549 - Epoch: [151][ 1130/ 1236] Overall Loss 0.140447 Objective Loss 0.140447 LR 0.000250 Time 0.021651 +2023-10-02 21:42:05,758 - Epoch: [151][ 1140/ 1236] Overall Loss 0.140619 Objective Loss 0.140619 LR 0.000250 Time 0.021645 +2023-10-02 21:42:05,967 - Epoch: [151][ 1150/ 1236] Overall Loss 0.140668 Objective Loss 0.140668 LR 0.000250 Time 0.021637 +2023-10-02 21:42:06,176 - Epoch: [151][ 1160/ 1236] Overall Loss 0.140575 Objective Loss 0.140575 LR 0.000250 Time 0.021631 +2023-10-02 21:42:06,384 - Epoch: [151][ 1170/ 1236] Overall Loss 0.140590 Objective Loss 0.140590 LR 0.000250 Time 0.021624 +2023-10-02 21:42:06,593 - Epoch: [151][ 1180/ 1236] Overall Loss 0.140494 Objective Loss 0.140494 LR 0.000250 Time 0.021617 +2023-10-02 21:42:06,802 - Epoch: [151][ 1190/ 1236] Overall Loss 0.140586 Objective Loss 0.140586 LR 0.000250 Time 0.021611 +2023-10-02 21:42:07,011 - Epoch: [151][ 1200/ 1236] Overall Loss 0.140597 Objective Loss 0.140597 LR 0.000250 Time 0.021605 +2023-10-02 21:42:07,219 - Epoch: [151][ 1210/ 1236] Overall Loss 0.140844 Objective Loss 0.140844 LR 0.000250 Time 0.021598 +2023-10-02 21:42:07,429 - Epoch: [151][ 1220/ 1236] Overall Loss 0.140754 Objective Loss 0.140754 LR 0.000250 Time 0.021592 +2023-10-02 21:42:07,690 - Epoch: [151][ 1230/ 1236] Overall Loss 0.140797 Objective Loss 0.140797 LR 0.000250 Time 0.021629 +2023-10-02 21:42:07,812 - Epoch: [151][ 1236/ 1236] Overall Loss 0.140842 Objective Loss 0.140842 Top1 91.853360 Top5 98.778004 LR 0.000250 Time 0.021622 +2023-10-02 21:42:07,947 - --- validate (epoch=151)----------- +2023-10-02 21:42:07,948 - 29943 samples (256 per mini-batch) +2023-10-02 21:42:08,422 - Epoch: [151][ 10/ 117] Loss 0.271918 Top1 86.992188 Top5 98.359375 +2023-10-02 21:42:08,570 - Epoch: [151][ 20/ 117] Loss 0.274565 Top1 87.265625 Top5 98.515625 +2023-10-02 21:42:08,720 - Epoch: [151][ 30/ 117] Loss 0.289421 Top1 86.901042 Top5 98.606771 +2023-10-02 21:42:08,868 - Epoch: [151][ 40/ 117] Loss 0.292209 Top1 86.757812 Top5 98.603516 +2023-10-02 21:42:09,016 - Epoch: [151][ 50/ 117] Loss 0.295547 Top1 86.781250 Top5 98.546875 +2023-10-02 21:42:09,163 - Epoch: [151][ 60/ 117] Loss 0.293721 Top1 86.842448 Top5 98.567708 +2023-10-02 21:42:09,309 - Epoch: [151][ 70/ 117] Loss 0.294199 Top1 86.953125 Top5 98.616071 +2023-10-02 21:42:09,455 - Epoch: [151][ 80/ 117] Loss 0.297189 Top1 86.875000 Top5 98.579102 +2023-10-02 21:42:09,601 - Epoch: [151][ 90/ 117] Loss 0.295131 Top1 86.957465 Top5 98.576389 +2023-10-02 21:42:09,747 - Epoch: [151][ 100/ 117] Loss 0.293879 Top1 86.917969 Top5 98.589844 +2023-10-02 21:42:09,901 - Epoch: [151][ 110/ 117] Loss 0.294310 Top1 86.967330 Top5 98.568892 +2023-10-02 21:42:09,989 - Epoch: [151][ 117/ 117] Loss 0.291836 Top1 86.975253 Top5 98.597335 +2023-10-02 21:42:10,131 - ==> Top1: 86.975 Top5: 98.597 Loss: 0.292 + +2023-10-02 21:42:10,132 - ==> Confusion: +[[ 938 1 3 0 6 2 0 0 9 62 3 0 0 1 3 1 5 0 1 0 15] + [ 0 1075 1 1 3 11 0 15 2 1 0 0 1 0 1 3 2 0 7 2 6] + [ 1 1 981 7 0 0 18 4 0 2 1 0 9 3 0 3 2 2 11 2 9] + [ 1 3 12 980 1 0 1 2 5 0 6 0 4 3 27 5 1 3 12 2 21] + [ 20 4 0 1 981 5 0 0 2 9 0 0 0 2 10 5 9 0 0 0 2] + [ 2 34 1 2 3 1001 1 21 1 6 2 5 3 7 4 0 3 1 5 1 13] + [ 0 3 24 1 0 3 1129 5 0 0 4 1 0 0 0 5 0 0 1 8 7] + [ 1 14 16 0 5 21 5 1069 2 1 4 4 2 3 2 1 1 2 43 10 12] + [ 14 2 0 1 2 5 0 1 979 36 14 1 2 8 14 1 3 0 2 1 3] + [ 77 0 2 2 4 4 0 0 22 969 1 1 0 20 5 3 0 0 0 1 8] + [ 3 1 10 7 0 2 3 2 14 2 972 1 0 8 7 0 3 1 9 1 7] + [ 0 0 0 0 0 15 1 4 0 0 0 961 22 7 0 2 0 16 0 3 4] + [ 0 0 1 2 1 1 2 0 0 1 2 34 973 2 2 7 0 13 5 7 15] + [ 0 0 0 0 2 7 1 0 15 9 4 7 0 1052 4 1 0 1 0 2 14] + [ 12 0 5 14 5 1 0 0 16 2 1 0 2 3 1022 0 0 2 9 0 7] + [ 0 0 1 1 7 0 0 0 0 1 0 6 6 0 0 1074 14 11 3 6 4] + [ 0 15 1 0 4 4 1 0 0 1 0 4 0 4 3 8 1103 0 1 1 11] + [ 0 0 1 0 1 0 1 0 0 1 0 1 16 2 2 4 0 1004 0 1 4] + [ 3 5 2 19 0 1 1 15 3 1 2 0 2 0 7 0 0 0 996 1 10] + [ 0 0 3 3 1 1 7 6 0 0 0 12 5 3 1 2 12 0 2 1086 8] + [ 106 130 105 75 57 109 25 81 95 68 166 94 262 228 133 50 87 63 122 151 5698]] + +2023-10-02 21:42:10,133 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:42:10,133 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:42:10,139 - + +2023-10-02 21:42:10,139 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:42:11,263 - Epoch: [152][ 10/ 1236] Overall Loss 0.142047 Objective Loss 0.142047 LR 0.000250 Time 0.112410 +2023-10-02 21:42:11,472 - Epoch: [152][ 20/ 1236] Overall Loss 0.138461 Objective Loss 0.138461 LR 0.000250 Time 0.066605 +2023-10-02 21:42:11,679 - Epoch: [152][ 30/ 1236] Overall Loss 0.142733 Objective Loss 0.142733 LR 0.000250 Time 0.051294 +2023-10-02 21:42:11,886 - Epoch: [152][ 40/ 1236] Overall Loss 0.139006 Objective Loss 0.139006 LR 0.000250 Time 0.043653 +2023-10-02 21:42:12,093 - Epoch: [152][ 50/ 1236] Overall Loss 0.138829 Objective Loss 0.138829 LR 0.000250 Time 0.039041 +2023-10-02 21:42:12,300 - Epoch: [152][ 60/ 1236] Overall Loss 0.139514 Objective Loss 0.139514 LR 0.000250 Time 0.035988 +2023-10-02 21:42:12,507 - Epoch: [152][ 70/ 1236] Overall Loss 0.141608 Objective Loss 0.141608 LR 0.000250 Time 0.033792 +2023-10-02 21:42:12,715 - Epoch: [152][ 80/ 1236] Overall Loss 0.144531 Objective Loss 0.144531 LR 0.000250 Time 0.032168 +2023-10-02 21:42:12,920 - Epoch: [152][ 90/ 1236] Overall Loss 0.141313 Objective Loss 0.141313 LR 0.000250 Time 0.030866 +2023-10-02 21:42:13,128 - Epoch: [152][ 100/ 1236] Overall Loss 0.139095 Objective Loss 0.139095 LR 0.000250 Time 0.029863 +2023-10-02 21:42:13,334 - Epoch: [152][ 110/ 1236] Overall Loss 0.139102 Objective Loss 0.139102 LR 0.000250 Time 0.029010 +2023-10-02 21:42:13,542 - Epoch: [152][ 120/ 1236] Overall Loss 0.138437 Objective Loss 0.138437 LR 0.000250 Time 0.028327 +2023-10-02 21:42:13,747 - Epoch: [152][ 130/ 1236] Overall Loss 0.137742 Objective Loss 0.137742 LR 0.000250 Time 0.027721 +2023-10-02 21:42:13,955 - Epoch: [152][ 140/ 1236] Overall Loss 0.137803 Objective Loss 0.137803 LR 0.000250 Time 0.027228 +2023-10-02 21:42:14,160 - Epoch: [152][ 150/ 1236] Overall Loss 0.137697 Objective Loss 0.137697 LR 0.000250 Time 0.026776 +2023-10-02 21:42:14,369 - Epoch: [152][ 160/ 1236] Overall Loss 0.137804 Objective Loss 0.137804 LR 0.000250 Time 0.026404 +2023-10-02 21:42:14,573 - Epoch: [152][ 170/ 1236] Overall Loss 0.138774 Objective Loss 0.138774 LR 0.000250 Time 0.026053 +2023-10-02 21:42:14,782 - Epoch: [152][ 180/ 1236] Overall Loss 0.138825 Objective Loss 0.138825 LR 0.000250 Time 0.025762 +2023-10-02 21:42:14,986 - Epoch: [152][ 190/ 1236] Overall Loss 0.138971 Objective Loss 0.138971 LR 0.000250 Time 0.025483 +2023-10-02 21:42:15,196 - Epoch: [152][ 200/ 1236] Overall Loss 0.138305 Objective Loss 0.138305 LR 0.000250 Time 0.025253 +2023-10-02 21:42:15,401 - Epoch: [152][ 210/ 1236] Overall Loss 0.139335 Objective Loss 0.139335 LR 0.000250 Time 0.025026 +2023-10-02 21:42:15,609 - Epoch: [152][ 220/ 1236] Overall Loss 0.138879 Objective Loss 0.138879 LR 0.000250 Time 0.024832 +2023-10-02 21:42:15,815 - Epoch: [152][ 230/ 1236] Overall Loss 0.138568 Objective Loss 0.138568 LR 0.000250 Time 0.024644 +2023-10-02 21:42:16,024 - Epoch: [152][ 240/ 1236] Overall Loss 0.138367 Objective Loss 0.138367 LR 0.000250 Time 0.024486 +2023-10-02 21:42:16,229 - Epoch: [152][ 250/ 1236] Overall Loss 0.138205 Objective Loss 0.138205 LR 0.000250 Time 0.024327 +2023-10-02 21:42:16,439 - Epoch: [152][ 260/ 1236] Overall Loss 0.137509 Objective Loss 0.137509 LR 0.000250 Time 0.024195 +2023-10-02 21:42:16,643 - Epoch: [152][ 270/ 1236] Overall Loss 0.137057 Objective Loss 0.137057 LR 0.000250 Time 0.024057 +2023-10-02 21:42:16,852 - Epoch: [152][ 280/ 1236] Overall Loss 0.136436 Objective Loss 0.136436 LR 0.000250 Time 0.023942 +2023-10-02 21:42:17,059 - Epoch: [152][ 290/ 1236] Overall Loss 0.136708 Objective Loss 0.136708 LR 0.000250 Time 0.023823 +2023-10-02 21:42:17,266 - Epoch: [152][ 300/ 1236] Overall Loss 0.137340 Objective Loss 0.137340 LR 0.000250 Time 0.023720 +2023-10-02 21:42:17,473 - Epoch: [152][ 310/ 1236] Overall Loss 0.137244 Objective Loss 0.137244 LR 0.000250 Time 0.023616 +2023-10-02 21:42:17,682 - Epoch: [152][ 320/ 1236] Overall Loss 0.136766 Objective Loss 0.136766 LR 0.000250 Time 0.023529 +2023-10-02 21:42:17,887 - Epoch: [152][ 330/ 1236] Overall Loss 0.137238 Objective Loss 0.137238 LR 0.000250 Time 0.023438 +2023-10-02 21:42:18,095 - Epoch: [152][ 340/ 1236] Overall Loss 0.137281 Objective Loss 0.137281 LR 0.000250 Time 0.023358 +2023-10-02 21:42:18,302 - Epoch: [152][ 350/ 1236] Overall Loss 0.137062 Objective Loss 0.137062 LR 0.000250 Time 0.023279 +2023-10-02 21:42:18,510 - Epoch: [152][ 360/ 1236] Overall Loss 0.136519 Objective Loss 0.136519 LR 0.000250 Time 0.023209 +2023-10-02 21:42:18,717 - Epoch: [152][ 370/ 1236] Overall Loss 0.136611 Objective Loss 0.136611 LR 0.000250 Time 0.023137 +2023-10-02 21:42:18,925 - Epoch: [152][ 380/ 1236] Overall Loss 0.136464 Objective Loss 0.136464 LR 0.000250 Time 0.023074 +2023-10-02 21:42:19,131 - Epoch: [152][ 390/ 1236] Overall Loss 0.136406 Objective Loss 0.136406 LR 0.000250 Time 0.023012 +2023-10-02 21:42:19,345 - Epoch: [152][ 400/ 1236] Overall Loss 0.136113 Objective Loss 0.136113 LR 0.000250 Time 0.022971 +2023-10-02 21:42:19,558 - Epoch: [152][ 410/ 1236] Overall Loss 0.136220 Objective Loss 0.136220 LR 0.000250 Time 0.022927 +2023-10-02 21:42:19,774 - Epoch: [152][ 420/ 1236] Overall Loss 0.136612 Objective Loss 0.136612 LR 0.000250 Time 0.022896 +2023-10-02 21:42:19,986 - Epoch: [152][ 430/ 1236] Overall Loss 0.136504 Objective Loss 0.136504 LR 0.000250 Time 0.022856 +2023-10-02 21:42:20,203 - Epoch: [152][ 440/ 1236] Overall Loss 0.136647 Objective Loss 0.136647 LR 0.000250 Time 0.022828 +2023-10-02 21:42:20,415 - Epoch: [152][ 450/ 1236] Overall Loss 0.136700 Objective Loss 0.136700 LR 0.000250 Time 0.022790 +2023-10-02 21:42:20,631 - Epoch: [152][ 460/ 1236] Overall Loss 0.136748 Objective Loss 0.136748 LR 0.000250 Time 0.022765 +2023-10-02 21:42:20,843 - Epoch: [152][ 470/ 1236] Overall Loss 0.136934 Objective Loss 0.136934 LR 0.000250 Time 0.022732 +2023-10-02 21:42:21,060 - Epoch: [152][ 480/ 1236] Overall Loss 0.136840 Objective Loss 0.136840 LR 0.000250 Time 0.022708 +2023-10-02 21:42:21,272 - Epoch: [152][ 490/ 1236] Overall Loss 0.136762 Objective Loss 0.136762 LR 0.000250 Time 0.022677 +2023-10-02 21:42:21,478 - Epoch: [152][ 500/ 1236] Overall Loss 0.136785 Objective Loss 0.136785 LR 0.000250 Time 0.022635 +2023-10-02 21:42:21,684 - Epoch: [152][ 510/ 1236] Overall Loss 0.136709 Objective Loss 0.136709 LR 0.000250 Time 0.022595 +2023-10-02 21:42:21,893 - Epoch: [152][ 520/ 1236] Overall Loss 0.137089 Objective Loss 0.137089 LR 0.000250 Time 0.022561 +2023-10-02 21:42:22,103 - Epoch: [152][ 530/ 1236] Overall Loss 0.136859 Objective Loss 0.136859 LR 0.000250 Time 0.022530 +2023-10-02 21:42:22,312 - Epoch: [152][ 540/ 1236] Overall Loss 0.137099 Objective Loss 0.137099 LR 0.000250 Time 0.022498 +2023-10-02 21:42:22,523 - Epoch: [152][ 550/ 1236] Overall Loss 0.137193 Objective Loss 0.137193 LR 0.000250 Time 0.022471 +2023-10-02 21:42:22,732 - Epoch: [152][ 560/ 1236] Overall Loss 0.137316 Objective Loss 0.137316 LR 0.000250 Time 0.022442 +2023-10-02 21:42:22,942 - Epoch: [152][ 570/ 1236] Overall Loss 0.137218 Objective Loss 0.137218 LR 0.000250 Time 0.022418 +2023-10-02 21:42:23,152 - Epoch: [152][ 580/ 1236] Overall Loss 0.137071 Objective Loss 0.137071 LR 0.000250 Time 0.022392 +2023-10-02 21:42:23,362 - Epoch: [152][ 590/ 1236] Overall Loss 0.136993 Objective Loss 0.136993 LR 0.000250 Time 0.022368 +2023-10-02 21:42:23,572 - Epoch: [152][ 600/ 1236] Overall Loss 0.136884 Objective Loss 0.136884 LR 0.000250 Time 0.022345 +2023-10-02 21:42:23,783 - Epoch: [152][ 610/ 1236] Overall Loss 0.136787 Objective Loss 0.136787 LR 0.000250 Time 0.022324 +2023-10-02 21:42:23,993 - Epoch: [152][ 620/ 1236] Overall Loss 0.136595 Objective Loss 0.136595 LR 0.000250 Time 0.022301 +2023-10-02 21:42:24,203 - Epoch: [152][ 630/ 1236] Overall Loss 0.136675 Objective Loss 0.136675 LR 0.000250 Time 0.022279 +2023-10-02 21:42:24,412 - Epoch: [152][ 640/ 1236] Overall Loss 0.136969 Objective Loss 0.136969 LR 0.000250 Time 0.022258 +2023-10-02 21:42:24,622 - Epoch: [152][ 650/ 1236] Overall Loss 0.136843 Objective Loss 0.136843 LR 0.000250 Time 0.022239 +2023-10-02 21:42:24,831 - Epoch: [152][ 660/ 1236] Overall Loss 0.136920 Objective Loss 0.136920 LR 0.000250 Time 0.022217 +2023-10-02 21:42:25,041 - Epoch: [152][ 670/ 1236] Overall Loss 0.136925 Objective Loss 0.136925 LR 0.000250 Time 0.022198 +2023-10-02 21:42:25,250 - Epoch: [152][ 680/ 1236] Overall Loss 0.137106 Objective Loss 0.137106 LR 0.000250 Time 0.022179 +2023-10-02 21:42:25,462 - Epoch: [152][ 690/ 1236] Overall Loss 0.137149 Objective Loss 0.137149 LR 0.000250 Time 0.022163 +2023-10-02 21:42:25,670 - Epoch: [152][ 700/ 1236] Overall Loss 0.137171 Objective Loss 0.137171 LR 0.000250 Time 0.022144 +2023-10-02 21:42:25,879 - Epoch: [152][ 710/ 1236] Overall Loss 0.137068 Objective Loss 0.137068 LR 0.000250 Time 0.022124 +2023-10-02 21:42:26,088 - Epoch: [152][ 720/ 1236] Overall Loss 0.137193 Objective Loss 0.137193 LR 0.000250 Time 0.022107 +2023-10-02 21:42:26,297 - Epoch: [152][ 730/ 1236] Overall Loss 0.137197 Objective Loss 0.137197 LR 0.000250 Time 0.022090 +2023-10-02 21:42:26,506 - Epoch: [152][ 740/ 1236] Overall Loss 0.136920 Objective Loss 0.136920 LR 0.000250 Time 0.022073 +2023-10-02 21:42:26,717 - Epoch: [152][ 750/ 1236] Overall Loss 0.136940 Objective Loss 0.136940 LR 0.000250 Time 0.022059 +2023-10-02 21:42:26,925 - Epoch: [152][ 760/ 1236] Overall Loss 0.137012 Objective Loss 0.137012 LR 0.000250 Time 0.022043 +2023-10-02 21:42:27,136 - Epoch: [152][ 770/ 1236] Overall Loss 0.137111 Objective Loss 0.137111 LR 0.000250 Time 0.022030 +2023-10-02 21:42:27,345 - Epoch: [152][ 780/ 1236] Overall Loss 0.137118 Objective Loss 0.137118 LR 0.000250 Time 0.022015 +2023-10-02 21:42:27,556 - Epoch: [152][ 790/ 1236] Overall Loss 0.137237 Objective Loss 0.137237 LR 0.000250 Time 0.022002 +2023-10-02 21:42:27,766 - Epoch: [152][ 800/ 1236] Overall Loss 0.137617 Objective Loss 0.137617 LR 0.000250 Time 0.021989 +2023-10-02 21:42:27,977 - Epoch: [152][ 810/ 1236] Overall Loss 0.137728 Objective Loss 0.137728 LR 0.000250 Time 0.021977 +2023-10-02 21:42:28,185 - Epoch: [152][ 820/ 1236] Overall Loss 0.137533 Objective Loss 0.137533 LR 0.000250 Time 0.021964 +2023-10-02 21:42:28,397 - Epoch: [152][ 830/ 1236] Overall Loss 0.137912 Objective Loss 0.137912 LR 0.000250 Time 0.021954 +2023-10-02 21:42:28,606 - Epoch: [152][ 840/ 1236] Overall Loss 0.138009 Objective Loss 0.138009 LR 0.000250 Time 0.021941 +2023-10-02 21:42:28,816 - Epoch: [152][ 850/ 1236] Overall Loss 0.138043 Objective Loss 0.138043 LR 0.000250 Time 0.021929 +2023-10-02 21:42:29,026 - Epoch: [152][ 860/ 1236] Overall Loss 0.138073 Objective Loss 0.138073 LR 0.000250 Time 0.021917 +2023-10-02 21:42:29,237 - Epoch: [152][ 870/ 1236] Overall Loss 0.138083 Objective Loss 0.138083 LR 0.000250 Time 0.021908 +2023-10-02 21:42:29,445 - Epoch: [152][ 880/ 1236] Overall Loss 0.138231 Objective Loss 0.138231 LR 0.000250 Time 0.021895 +2023-10-02 21:42:29,655 - Epoch: [152][ 890/ 1236] Overall Loss 0.138201 Objective Loss 0.138201 LR 0.000250 Time 0.021885 +2023-10-02 21:42:29,863 - Epoch: [152][ 900/ 1236] Overall Loss 0.138232 Objective Loss 0.138232 LR 0.000250 Time 0.021873 +2023-10-02 21:42:30,073 - Epoch: [152][ 910/ 1236] Overall Loss 0.138326 Objective Loss 0.138326 LR 0.000250 Time 0.021862 +2023-10-02 21:42:30,284 - Epoch: [152][ 920/ 1236] Overall Loss 0.138358 Objective Loss 0.138358 LR 0.000250 Time 0.021854 +2023-10-02 21:42:30,493 - Epoch: [152][ 930/ 1236] Overall Loss 0.138259 Objective Loss 0.138259 LR 0.000250 Time 0.021843 +2023-10-02 21:42:30,703 - Epoch: [152][ 940/ 1236] Overall Loss 0.138229 Objective Loss 0.138229 LR 0.000250 Time 0.021834 +2023-10-02 21:42:30,913 - Epoch: [152][ 950/ 1236] Overall Loss 0.138212 Objective Loss 0.138212 LR 0.000250 Time 0.021824 +2023-10-02 21:42:31,124 - Epoch: [152][ 960/ 1236] Overall Loss 0.138282 Objective Loss 0.138282 LR 0.000250 Time 0.021817 +2023-10-02 21:42:31,333 - Epoch: [152][ 970/ 1236] Overall Loss 0.138335 Objective Loss 0.138335 LR 0.000250 Time 0.021807 +2023-10-02 21:42:31,545 - Epoch: [152][ 980/ 1236] Overall Loss 0.138334 Objective Loss 0.138334 LR 0.000250 Time 0.021800 +2023-10-02 21:42:31,752 - Epoch: [152][ 990/ 1236] Overall Loss 0.138367 Objective Loss 0.138367 LR 0.000250 Time 0.021789 +2023-10-02 21:42:31,962 - Epoch: [152][ 1000/ 1236] Overall Loss 0.138089 Objective Loss 0.138089 LR 0.000250 Time 0.021781 +2023-10-02 21:42:32,171 - Epoch: [152][ 1010/ 1236] Overall Loss 0.138008 Objective Loss 0.138008 LR 0.000250 Time 0.021772 +2023-10-02 21:42:32,383 - Epoch: [152][ 1020/ 1236] Overall Loss 0.138164 Objective Loss 0.138164 LR 0.000250 Time 0.021765 +2023-10-02 21:42:32,592 - Epoch: [152][ 1030/ 1236] Overall Loss 0.138266 Objective Loss 0.138266 LR 0.000250 Time 0.021756 +2023-10-02 21:42:32,803 - Epoch: [152][ 1040/ 1236] Overall Loss 0.138242 Objective Loss 0.138242 LR 0.000250 Time 0.021749 +2023-10-02 21:42:33,011 - Epoch: [152][ 1050/ 1236] Overall Loss 0.138396 Objective Loss 0.138396 LR 0.000250 Time 0.021740 +2023-10-02 21:42:33,222 - Epoch: [152][ 1060/ 1236] Overall Loss 0.138410 Objective Loss 0.138410 LR 0.000250 Time 0.021734 +2023-10-02 21:42:33,431 - Epoch: [152][ 1070/ 1236] Overall Loss 0.138209 Objective Loss 0.138209 LR 0.000250 Time 0.021725 +2023-10-02 21:42:33,642 - Epoch: [152][ 1080/ 1236] Overall Loss 0.138181 Objective Loss 0.138181 LR 0.000250 Time 0.021719 +2023-10-02 21:42:33,850 - Epoch: [152][ 1090/ 1236] Overall Loss 0.138157 Objective Loss 0.138157 LR 0.000250 Time 0.021711 +2023-10-02 21:42:34,060 - Epoch: [152][ 1100/ 1236] Overall Loss 0.138460 Objective Loss 0.138460 LR 0.000250 Time 0.021703 +2023-10-02 21:42:34,269 - Epoch: [152][ 1110/ 1236] Overall Loss 0.138504 Objective Loss 0.138504 LR 0.000250 Time 0.021695 +2023-10-02 21:42:34,478 - Epoch: [152][ 1120/ 1236] Overall Loss 0.138714 Objective Loss 0.138714 LR 0.000250 Time 0.021688 +2023-10-02 21:42:34,688 - Epoch: [152][ 1130/ 1236] Overall Loss 0.138766 Objective Loss 0.138766 LR 0.000250 Time 0.021680 +2023-10-02 21:42:34,897 - Epoch: [152][ 1140/ 1236] Overall Loss 0.138931 Objective Loss 0.138931 LR 0.000250 Time 0.021673 +2023-10-02 21:42:35,107 - Epoch: [152][ 1150/ 1236] Overall Loss 0.138911 Objective Loss 0.138911 LR 0.000250 Time 0.021666 +2023-10-02 21:42:35,317 - Epoch: [152][ 1160/ 1236] Overall Loss 0.138944 Objective Loss 0.138944 LR 0.000250 Time 0.021660 +2023-10-02 21:42:35,527 - Epoch: [152][ 1170/ 1236] Overall Loss 0.138907 Objective Loss 0.138907 LR 0.000250 Time 0.021653 +2023-10-02 21:42:35,736 - Epoch: [152][ 1180/ 1236] Overall Loss 0.138937 Objective Loss 0.138937 LR 0.000250 Time 0.021647 +2023-10-02 21:42:35,946 - Epoch: [152][ 1190/ 1236] Overall Loss 0.138923 Objective Loss 0.138923 LR 0.000250 Time 0.021640 +2023-10-02 21:42:36,155 - Epoch: [152][ 1200/ 1236] Overall Loss 0.138862 Objective Loss 0.138862 LR 0.000250 Time 0.021634 +2023-10-02 21:42:36,365 - Epoch: [152][ 1210/ 1236] Overall Loss 0.138916 Objective Loss 0.138916 LR 0.000250 Time 0.021627 +2023-10-02 21:42:36,573 - Epoch: [152][ 1220/ 1236] Overall Loss 0.138896 Objective Loss 0.138896 LR 0.000250 Time 0.021620 +2023-10-02 21:42:36,834 - Epoch: [152][ 1230/ 1236] Overall Loss 0.138885 Objective Loss 0.138885 LR 0.000250 Time 0.021656 +2023-10-02 21:42:36,955 - Epoch: [152][ 1236/ 1236] Overall Loss 0.138950 Objective Loss 0.138950 Top1 91.446029 Top5 98.778004 LR 0.000250 Time 0.021648 +2023-10-02 21:42:37,088 - --- validate (epoch=152)----------- +2023-10-02 21:42:37,089 - 29943 samples (256 per mini-batch) +2023-10-02 21:42:37,586 - Epoch: [152][ 10/ 117] Loss 0.275546 Top1 87.656250 Top5 98.750000 +2023-10-02 21:42:37,752 - Epoch: [152][ 20/ 117] Loss 0.282170 Top1 87.500000 Top5 98.730469 +2023-10-02 21:42:37,912 - Epoch: [152][ 30/ 117] Loss 0.294209 Top1 87.083333 Top5 98.606771 +2023-10-02 21:42:38,076 - Epoch: [152][ 40/ 117] Loss 0.285631 Top1 87.197266 Top5 98.652344 +2023-10-02 21:42:38,236 - Epoch: [152][ 50/ 117] Loss 0.298323 Top1 86.812500 Top5 98.578125 +2023-10-02 21:42:38,399 - Epoch: [152][ 60/ 117] Loss 0.294430 Top1 86.998698 Top5 98.613281 +2023-10-02 21:42:38,557 - Epoch: [152][ 70/ 117] Loss 0.293830 Top1 87.064732 Top5 98.610491 +2023-10-02 21:42:38,718 - Epoch: [152][ 80/ 117] Loss 0.292872 Top1 87.075195 Top5 98.637695 +2023-10-02 21:42:38,874 - Epoch: [152][ 90/ 117] Loss 0.294600 Top1 87.031250 Top5 98.624132 +2023-10-02 21:42:39,035 - Epoch: [152][ 100/ 117] Loss 0.295680 Top1 86.996094 Top5 98.597656 +2023-10-02 21:42:39,198 - Epoch: [152][ 110/ 117] Loss 0.296035 Top1 87.056108 Top5 98.600852 +2023-10-02 21:42:39,287 - Epoch: [152][ 117/ 117] Loss 0.295442 Top1 87.008650 Top5 98.604014 +2023-10-02 21:42:39,422 - ==> Top1: 87.009 Top5: 98.604 Loss: 0.295 + +2023-10-02 21:42:39,423 - ==> Confusion: +[[ 941 1 3 0 9 2 0 0 4 62 2 0 1 2 3 1 3 0 0 0 16] + [ 0 1074 2 1 4 17 0 17 3 1 0 1 0 0 0 3 1 0 4 1 2] + [ 3 0 987 8 1 0 10 6 0 2 2 0 9 2 1 3 2 2 5 4 9] + [ 3 3 14 987 1 2 1 1 2 0 2 0 4 2 33 3 2 4 8 0 17] + [ 23 3 1 1 980 5 0 0 2 14 1 0 1 1 5 4 6 0 0 1 2] + [ 2 34 1 2 2 1000 1 22 1 6 1 5 2 10 5 0 4 1 2 1 14] + [ 0 3 25 1 0 2 1129 7 0 0 2 1 0 0 0 4 0 0 2 6 9] + [ 1 18 14 0 8 24 4 1074 1 1 4 2 5 5 1 1 0 1 37 8 9] + [ 16 1 0 1 2 3 0 1 985 35 7 1 1 12 13 1 5 0 2 0 3] + [ 92 1 0 3 6 3 0 0 20 962 0 1 0 14 7 2 0 0 0 0 8] + [ 3 4 11 9 0 0 6 2 15 2 963 0 0 9 5 0 2 2 7 3 10] + [ 0 0 0 0 1 18 0 2 0 0 0 963 17 6 0 2 1 16 0 4 5] + [ 0 2 1 1 0 2 1 0 2 1 3 31 983 2 2 8 0 10 1 6 12] + [ 0 0 0 0 2 7 0 0 13 12 3 8 0 1050 4 1 0 1 0 1 17] + [ 14 0 3 15 5 0 0 0 17 1 3 0 2 1 1023 0 0 3 7 0 7] + [ 0 0 2 1 5 1 0 0 0 1 1 4 6 0 0 1075 15 9 2 7 5] + [ 0 17 2 0 6 7 0 0 0 1 0 3 0 1 4 8 1096 0 2 3 11] + [ 0 0 2 2 0 1 1 0 0 0 0 3 23 1 0 6 0 993 0 0 6] + [ 0 8 3 13 0 1 1 19 5 3 2 1 2 0 11 0 1 1 985 1 11] + [ 0 1 4 2 0 4 6 4 0 0 0 15 3 1 0 2 9 1 0 1092 8] + [ 98 166 105 68 74 103 24 71 89 85 135 79 301 236 133 48 87 49 116 127 5711]] + +2023-10-02 21:42:39,425 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:42:39,425 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:42:39,431 - + +2023-10-02 21:42:39,431 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:42:40,433 - Epoch: [153][ 10/ 1236] Overall Loss 0.099480 Objective Loss 0.099480 LR 0.000250 Time 0.100182 +2023-10-02 21:42:40,640 - Epoch: [153][ 20/ 1236] Overall Loss 0.119905 Objective Loss 0.119905 LR 0.000250 Time 0.060423 +2023-10-02 21:42:40,848 - Epoch: [153][ 30/ 1236] Overall Loss 0.124002 Objective Loss 0.124002 LR 0.000250 Time 0.047202 +2023-10-02 21:42:41,055 - Epoch: [153][ 40/ 1236] Overall Loss 0.126300 Objective Loss 0.126300 LR 0.000250 Time 0.040578 +2023-10-02 21:42:41,263 - Epoch: [153][ 50/ 1236] Overall Loss 0.125182 Objective Loss 0.125182 LR 0.000250 Time 0.036606 +2023-10-02 21:42:41,472 - Epoch: [153][ 60/ 1236] Overall Loss 0.125897 Objective Loss 0.125897 LR 0.000250 Time 0.033983 +2023-10-02 21:42:41,678 - Epoch: [153][ 70/ 1236] Overall Loss 0.128673 Objective Loss 0.128673 LR 0.000250 Time 0.032072 +2023-10-02 21:42:41,886 - Epoch: [153][ 80/ 1236] Overall Loss 0.129838 Objective Loss 0.129838 LR 0.000250 Time 0.030654 +2023-10-02 21:42:42,093 - Epoch: [153][ 90/ 1236] Overall Loss 0.128170 Objective Loss 0.128170 LR 0.000250 Time 0.029534 +2023-10-02 21:42:42,299 - Epoch: [153][ 100/ 1236] Overall Loss 0.129121 Objective Loss 0.129121 LR 0.000250 Time 0.028642 +2023-10-02 21:42:42,506 - Epoch: [153][ 110/ 1236] Overall Loss 0.128698 Objective Loss 0.128698 LR 0.000250 Time 0.027903 +2023-10-02 21:42:42,715 - Epoch: [153][ 120/ 1236] Overall Loss 0.129767 Objective Loss 0.129767 LR 0.000250 Time 0.027321 +2023-10-02 21:42:42,923 - Epoch: [153][ 130/ 1236] Overall Loss 0.129318 Objective Loss 0.129318 LR 0.000250 Time 0.026813 +2023-10-02 21:42:43,134 - Epoch: [153][ 140/ 1236] Overall Loss 0.129019 Objective Loss 0.129019 LR 0.000250 Time 0.026401 +2023-10-02 21:42:43,341 - Epoch: [153][ 150/ 1236] Overall Loss 0.128258 Objective Loss 0.128258 LR 0.000250 Time 0.026020 +2023-10-02 21:42:43,549 - Epoch: [153][ 160/ 1236] Overall Loss 0.128632 Objective Loss 0.128632 LR 0.000250 Time 0.025696 +2023-10-02 21:42:43,758 - Epoch: [153][ 170/ 1236] Overall Loss 0.129197 Objective Loss 0.129197 LR 0.000250 Time 0.025411 +2023-10-02 21:42:43,969 - Epoch: [153][ 180/ 1236] Overall Loss 0.129677 Objective Loss 0.129677 LR 0.000250 Time 0.025170 +2023-10-02 21:42:44,177 - Epoch: [153][ 190/ 1236] Overall Loss 0.129701 Objective Loss 0.129701 LR 0.000250 Time 0.024936 +2023-10-02 21:42:44,384 - Epoch: [153][ 200/ 1236] Overall Loss 0.129333 Objective Loss 0.129333 LR 0.000250 Time 0.024724 +2023-10-02 21:42:44,587 - Epoch: [153][ 210/ 1236] Overall Loss 0.129598 Objective Loss 0.129598 LR 0.000250 Time 0.024515 +2023-10-02 21:42:44,793 - Epoch: [153][ 220/ 1236] Overall Loss 0.130526 Objective Loss 0.130526 LR 0.000250 Time 0.024332 +2023-10-02 21:42:44,997 - Epoch: [153][ 230/ 1236] Overall Loss 0.131271 Objective Loss 0.131271 LR 0.000250 Time 0.024157 +2023-10-02 21:42:45,202 - Epoch: [153][ 240/ 1236] Overall Loss 0.132753 Objective Loss 0.132753 LR 0.000250 Time 0.024005 +2023-10-02 21:42:45,407 - Epoch: [153][ 250/ 1236] Overall Loss 0.132466 Objective Loss 0.132466 LR 0.000250 Time 0.023857 +2023-10-02 21:42:45,611 - Epoch: [153][ 260/ 1236] Overall Loss 0.132254 Objective Loss 0.132254 LR 0.000250 Time 0.023721 +2023-10-02 21:42:45,815 - Epoch: [153][ 270/ 1236] Overall Loss 0.131995 Objective Loss 0.131995 LR 0.000250 Time 0.023594 +2023-10-02 21:42:46,022 - Epoch: [153][ 280/ 1236] Overall Loss 0.131780 Objective Loss 0.131780 LR 0.000250 Time 0.023488 +2023-10-02 21:42:46,225 - Epoch: [153][ 290/ 1236] Overall Loss 0.131920 Objective Loss 0.131920 LR 0.000250 Time 0.023378 +2023-10-02 21:42:46,430 - Epoch: [153][ 300/ 1236] Overall Loss 0.131929 Objective Loss 0.131929 LR 0.000250 Time 0.023281 +2023-10-02 21:42:46,634 - Epoch: [153][ 310/ 1236] Overall Loss 0.131753 Objective Loss 0.131753 LR 0.000250 Time 0.023185 +2023-10-02 21:42:46,840 - Epoch: [153][ 320/ 1236] Overall Loss 0.131373 Objective Loss 0.131373 LR 0.000250 Time 0.023102 +2023-10-02 21:42:47,043 - Epoch: [153][ 330/ 1236] Overall Loss 0.132228 Objective Loss 0.132228 LR 0.000250 Time 0.023018 +2023-10-02 21:42:47,249 - Epoch: [153][ 340/ 1236] Overall Loss 0.132073 Objective Loss 0.132073 LR 0.000250 Time 0.022944 +2023-10-02 21:42:47,453 - Epoch: [153][ 350/ 1236] Overall Loss 0.132225 Objective Loss 0.132225 LR 0.000250 Time 0.022872 +2023-10-02 21:42:47,659 - Epoch: [153][ 360/ 1236] Overall Loss 0.132527 Objective Loss 0.132527 LR 0.000250 Time 0.022807 +2023-10-02 21:42:47,864 - Epoch: [153][ 370/ 1236] Overall Loss 0.132801 Objective Loss 0.132801 LR 0.000250 Time 0.022743 +2023-10-02 21:42:48,069 - Epoch: [153][ 380/ 1236] Overall Loss 0.132990 Objective Loss 0.132990 LR 0.000250 Time 0.022684 +2023-10-02 21:42:48,274 - Epoch: [153][ 390/ 1236] Overall Loss 0.133155 Objective Loss 0.133155 LR 0.000250 Time 0.022627 +2023-10-02 21:42:48,479 - Epoch: [153][ 400/ 1236] Overall Loss 0.133509 Objective Loss 0.133509 LR 0.000250 Time 0.022573 +2023-10-02 21:42:48,684 - Epoch: [153][ 410/ 1236] Overall Loss 0.133681 Objective Loss 0.133681 LR 0.000250 Time 0.022522 +2023-10-02 21:42:48,889 - Epoch: [153][ 420/ 1236] Overall Loss 0.133917 Objective Loss 0.133917 LR 0.000250 Time 0.022473 +2023-10-02 21:42:49,094 - Epoch: [153][ 430/ 1236] Overall Loss 0.133938 Objective Loss 0.133938 LR 0.000250 Time 0.022426 +2023-10-02 21:42:49,299 - Epoch: [153][ 440/ 1236] Overall Loss 0.134235 Objective Loss 0.134235 LR 0.000250 Time 0.022382 +2023-10-02 21:42:49,504 - Epoch: [153][ 450/ 1236] Overall Loss 0.134125 Objective Loss 0.134125 LR 0.000250 Time 0.022339 +2023-10-02 21:42:49,709 - Epoch: [153][ 460/ 1236] Overall Loss 0.134431 Objective Loss 0.134431 LR 0.000250 Time 0.022299 +2023-10-02 21:42:49,914 - Epoch: [153][ 470/ 1236] Overall Loss 0.134221 Objective Loss 0.134221 LR 0.000250 Time 0.022260 +2023-10-02 21:42:50,119 - Epoch: [153][ 480/ 1236] Overall Loss 0.134302 Objective Loss 0.134302 LR 0.000250 Time 0.022223 +2023-10-02 21:42:50,324 - Epoch: [153][ 490/ 1236] Overall Loss 0.134643 Objective Loss 0.134643 LR 0.000250 Time 0.022187 +2023-10-02 21:42:50,529 - Epoch: [153][ 500/ 1236] Overall Loss 0.134757 Objective Loss 0.134757 LR 0.000250 Time 0.022153 +2023-10-02 21:42:50,734 - Epoch: [153][ 510/ 1236] Overall Loss 0.134680 Objective Loss 0.134680 LR 0.000250 Time 0.022120 +2023-10-02 21:42:50,940 - Epoch: [153][ 520/ 1236] Overall Loss 0.135118 Objective Loss 0.135118 LR 0.000250 Time 0.022089 +2023-10-02 21:42:51,144 - Epoch: [153][ 530/ 1236] Overall Loss 0.135197 Objective Loss 0.135197 LR 0.000250 Time 0.022058 +2023-10-02 21:42:51,350 - Epoch: [153][ 540/ 1236] Overall Loss 0.135122 Objective Loss 0.135122 LR 0.000250 Time 0.022029 +2023-10-02 21:42:51,555 - Epoch: [153][ 550/ 1236] Overall Loss 0.134837 Objective Loss 0.134837 LR 0.000250 Time 0.022001 +2023-10-02 21:42:51,760 - Epoch: [153][ 560/ 1236] Overall Loss 0.134719 Objective Loss 0.134719 LR 0.000250 Time 0.021974 +2023-10-02 21:42:51,965 - Epoch: [153][ 570/ 1236] Overall Loss 0.134556 Objective Loss 0.134556 LR 0.000250 Time 0.021947 +2023-10-02 21:42:52,170 - Epoch: [153][ 580/ 1236] Overall Loss 0.134592 Objective Loss 0.134592 LR 0.000250 Time 0.021922 +2023-10-02 21:42:52,375 - Epoch: [153][ 590/ 1236] Overall Loss 0.134509 Objective Loss 0.134509 LR 0.000250 Time 0.021895 +2023-10-02 21:42:52,581 - Epoch: [153][ 600/ 1236] Overall Loss 0.134640 Objective Loss 0.134640 LR 0.000250 Time 0.021872 +2023-10-02 21:42:52,785 - Epoch: [153][ 610/ 1236] Overall Loss 0.134932 Objective Loss 0.134932 LR 0.000250 Time 0.021847 +2023-10-02 21:42:52,991 - Epoch: [153][ 620/ 1236] Overall Loss 0.135576 Objective Loss 0.135576 LR 0.000250 Time 0.021825 +2023-10-02 21:42:53,195 - Epoch: [153][ 630/ 1236] Overall Loss 0.135902 Objective Loss 0.135902 LR 0.000250 Time 0.021803 +2023-10-02 21:42:53,401 - Epoch: [153][ 640/ 1236] Overall Loss 0.135923 Objective Loss 0.135923 LR 0.000250 Time 0.021783 +2023-10-02 21:42:53,606 - Epoch: [153][ 650/ 1236] Overall Loss 0.135881 Objective Loss 0.135881 LR 0.000250 Time 0.021763 +2023-10-02 21:42:53,811 - Epoch: [153][ 660/ 1236] Overall Loss 0.135713 Objective Loss 0.135713 LR 0.000250 Time 0.021744 +2023-10-02 21:42:54,016 - Epoch: [153][ 670/ 1236] Overall Loss 0.136001 Objective Loss 0.136001 LR 0.000250 Time 0.021725 +2023-10-02 21:42:54,221 - Epoch: [153][ 680/ 1236] Overall Loss 0.135753 Objective Loss 0.135753 LR 0.000250 Time 0.021707 +2023-10-02 21:42:54,426 - Epoch: [153][ 690/ 1236] Overall Loss 0.135537 Objective Loss 0.135537 LR 0.000250 Time 0.021688 +2023-10-02 21:42:54,631 - Epoch: [153][ 700/ 1236] Overall Loss 0.135409 Objective Loss 0.135409 LR 0.000250 Time 0.021671 +2023-10-02 21:42:54,836 - Epoch: [153][ 710/ 1236] Overall Loss 0.135246 Objective Loss 0.135246 LR 0.000250 Time 0.021654 +2023-10-02 21:42:55,042 - Epoch: [153][ 720/ 1236] Overall Loss 0.135254 Objective Loss 0.135254 LR 0.000250 Time 0.021638 +2023-10-02 21:42:55,246 - Epoch: [153][ 730/ 1236] Overall Loss 0.135489 Objective Loss 0.135489 LR 0.000250 Time 0.021622 +2023-10-02 21:42:55,452 - Epoch: [153][ 740/ 1236] Overall Loss 0.135555 Objective Loss 0.135555 LR 0.000250 Time 0.021607 +2023-10-02 21:42:55,657 - Epoch: [153][ 750/ 1236] Overall Loss 0.135511 Objective Loss 0.135511 LR 0.000250 Time 0.021592 +2023-10-02 21:42:55,862 - Epoch: [153][ 760/ 1236] Overall Loss 0.135774 Objective Loss 0.135774 LR 0.000250 Time 0.021578 +2023-10-02 21:42:56,067 - Epoch: [153][ 770/ 1236] Overall Loss 0.135749 Objective Loss 0.135749 LR 0.000250 Time 0.021563 +2023-10-02 21:42:56,273 - Epoch: [153][ 780/ 1236] Overall Loss 0.135781 Objective Loss 0.135781 LR 0.000250 Time 0.021550 +2023-10-02 21:42:56,478 - Epoch: [153][ 790/ 1236] Overall Loss 0.135861 Objective Loss 0.135861 LR 0.000250 Time 0.021536 +2023-10-02 21:42:56,683 - Epoch: [153][ 800/ 1236] Overall Loss 0.135849 Objective Loss 0.135849 LR 0.000250 Time 0.021523 +2023-10-02 21:42:56,888 - Epoch: [153][ 810/ 1236] Overall Loss 0.136224 Objective Loss 0.136224 LR 0.000250 Time 0.021511 +2023-10-02 21:42:57,094 - Epoch: [153][ 820/ 1236] Overall Loss 0.136292 Objective Loss 0.136292 LR 0.000250 Time 0.021498 +2023-10-02 21:42:57,299 - Epoch: [153][ 830/ 1236] Overall Loss 0.136325 Objective Loss 0.136325 LR 0.000250 Time 0.021485 +2023-10-02 21:42:57,505 - Epoch: [153][ 840/ 1236] Overall Loss 0.136505 Objective Loss 0.136505 LR 0.000250 Time 0.021473 +2023-10-02 21:42:57,710 - Epoch: [153][ 850/ 1236] Overall Loss 0.136669 Objective Loss 0.136669 LR 0.000250 Time 0.021462 +2023-10-02 21:42:57,915 - Epoch: [153][ 860/ 1236] Overall Loss 0.136897 Objective Loss 0.136897 LR 0.000250 Time 0.021451 +2023-10-02 21:42:58,120 - Epoch: [153][ 870/ 1236] Overall Loss 0.136887 Objective Loss 0.136887 LR 0.000250 Time 0.021438 +2023-10-02 21:42:58,326 - Epoch: [153][ 880/ 1236] Overall Loss 0.136772 Objective Loss 0.136772 LR 0.000250 Time 0.021427 +2023-10-02 21:42:58,531 - Epoch: [153][ 890/ 1236] Overall Loss 0.136831 Objective Loss 0.136831 LR 0.000250 Time 0.021417 +2023-10-02 21:42:58,736 - Epoch: [153][ 900/ 1236] Overall Loss 0.136643 Objective Loss 0.136643 LR 0.000250 Time 0.021407 +2023-10-02 21:42:58,941 - Epoch: [153][ 910/ 1236] Overall Loss 0.136842 Objective Loss 0.136842 LR 0.000250 Time 0.021396 +2023-10-02 21:42:59,147 - Epoch: [153][ 920/ 1236] Overall Loss 0.136890 Objective Loss 0.136890 LR 0.000250 Time 0.021387 +2023-10-02 21:42:59,352 - Epoch: [153][ 930/ 1236] Overall Loss 0.136789 Objective Loss 0.136789 LR 0.000250 Time 0.021375 +2023-10-02 21:42:59,557 - Epoch: [153][ 940/ 1236] Overall Loss 0.136920 Objective Loss 0.136920 LR 0.000250 Time 0.021366 +2023-10-02 21:42:59,762 - Epoch: [153][ 950/ 1236] Overall Loss 0.136942 Objective Loss 0.136942 LR 0.000250 Time 0.021357 +2023-10-02 21:42:59,968 - Epoch: [153][ 960/ 1236] Overall Loss 0.137079 Objective Loss 0.137079 LR 0.000250 Time 0.021348 +2023-10-02 21:43:00,173 - Epoch: [153][ 970/ 1236] Overall Loss 0.137095 Objective Loss 0.137095 LR 0.000250 Time 0.021340 +2023-10-02 21:43:00,379 - Epoch: [153][ 980/ 1236] Overall Loss 0.136946 Objective Loss 0.136946 LR 0.000250 Time 0.021331 +2023-10-02 21:43:00,584 - Epoch: [153][ 990/ 1236] Overall Loss 0.136766 Objective Loss 0.136766 LR 0.000250 Time 0.021323 +2023-10-02 21:43:00,789 - Epoch: [153][ 1000/ 1236] Overall Loss 0.137053 Objective Loss 0.137053 LR 0.000250 Time 0.021314 +2023-10-02 21:43:00,994 - Epoch: [153][ 1010/ 1236] Overall Loss 0.137197 Objective Loss 0.137197 LR 0.000250 Time 0.021306 +2023-10-02 21:43:01,200 - Epoch: [153][ 1020/ 1236] Overall Loss 0.137101 Objective Loss 0.137101 LR 0.000250 Time 0.021298 +2023-10-02 21:43:01,405 - Epoch: [153][ 1030/ 1236] Overall Loss 0.137164 Objective Loss 0.137164 LR 0.000250 Time 0.021289 +2023-10-02 21:43:01,610 - Epoch: [153][ 1040/ 1236] Overall Loss 0.137407 Objective Loss 0.137407 LR 0.000250 Time 0.021282 +2023-10-02 21:43:01,816 - Epoch: [153][ 1050/ 1236] Overall Loss 0.137119 Objective Loss 0.137119 LR 0.000250 Time 0.021274 +2023-10-02 21:43:02,021 - Epoch: [153][ 1060/ 1236] Overall Loss 0.136968 Objective Loss 0.136968 LR 0.000250 Time 0.021267 +2023-10-02 21:43:02,226 - Epoch: [153][ 1070/ 1236] Overall Loss 0.136962 Objective Loss 0.136962 LR 0.000250 Time 0.021260 +2023-10-02 21:43:02,431 - Epoch: [153][ 1080/ 1236] Overall Loss 0.137057 Objective Loss 0.137057 LR 0.000250 Time 0.021253 +2023-10-02 21:43:02,637 - Epoch: [153][ 1090/ 1236] Overall Loss 0.137209 Objective Loss 0.137209 LR 0.000250 Time 0.021245 +2023-10-02 21:43:02,842 - Epoch: [153][ 1100/ 1236] Overall Loss 0.137133 Objective Loss 0.137133 LR 0.000250 Time 0.021238 +2023-10-02 21:43:03,047 - Epoch: [153][ 1110/ 1236] Overall Loss 0.137409 Objective Loss 0.137409 LR 0.000250 Time 0.021231 +2023-10-02 21:43:03,253 - Epoch: [153][ 1120/ 1236] Overall Loss 0.137361 Objective Loss 0.137361 LR 0.000250 Time 0.021225 +2023-10-02 21:43:03,458 - Epoch: [153][ 1130/ 1236] Overall Loss 0.137372 Objective Loss 0.137372 LR 0.000250 Time 0.021218 +2023-10-02 21:43:03,663 - Epoch: [153][ 1140/ 1236] Overall Loss 0.137185 Objective Loss 0.137185 LR 0.000250 Time 0.021212 +2023-10-02 21:43:03,868 - Epoch: [153][ 1150/ 1236] Overall Loss 0.137350 Objective Loss 0.137350 LR 0.000250 Time 0.021204 +2023-10-02 21:43:04,074 - Epoch: [153][ 1160/ 1236] Overall Loss 0.137346 Objective Loss 0.137346 LR 0.000250 Time 0.021198 +2023-10-02 21:43:04,279 - Epoch: [153][ 1170/ 1236] Overall Loss 0.137272 Objective Loss 0.137272 LR 0.000250 Time 0.021191 +2023-10-02 21:43:04,484 - Epoch: [153][ 1180/ 1236] Overall Loss 0.137273 Objective Loss 0.137273 LR 0.000250 Time 0.021185 +2023-10-02 21:43:04,689 - Epoch: [153][ 1190/ 1236] Overall Loss 0.137177 Objective Loss 0.137177 LR 0.000250 Time 0.021179 +2023-10-02 21:43:04,895 - Epoch: [153][ 1200/ 1236] Overall Loss 0.137373 Objective Loss 0.137373 LR 0.000250 Time 0.021174 +2023-10-02 21:43:05,100 - Epoch: [153][ 1210/ 1236] Overall Loss 0.137293 Objective Loss 0.137293 LR 0.000250 Time 0.021167 +2023-10-02 21:43:05,305 - Epoch: [153][ 1220/ 1236] Overall Loss 0.137324 Objective Loss 0.137324 LR 0.000250 Time 0.021161 +2023-10-02 21:43:05,562 - Epoch: [153][ 1230/ 1236] Overall Loss 0.137363 Objective Loss 0.137363 LR 0.000250 Time 0.021198 +2023-10-02 21:43:05,684 - Epoch: [153][ 1236/ 1236] Overall Loss 0.137385 Objective Loss 0.137385 Top1 90.427699 Top5 99.389002 LR 0.000250 Time 0.021193 +2023-10-02 21:43:05,836 - --- validate (epoch=153)----------- +2023-10-02 21:43:05,836 - 29943 samples (256 per mini-batch) +2023-10-02 21:43:06,336 - Epoch: [153][ 10/ 117] Loss 0.278713 Top1 87.070312 Top5 98.828125 +2023-10-02 21:43:06,488 - Epoch: [153][ 20/ 117] Loss 0.275623 Top1 87.089844 Top5 98.808594 +2023-10-02 21:43:06,639 - Epoch: [153][ 30/ 117] Loss 0.278421 Top1 87.070312 Top5 98.776042 +2023-10-02 21:43:06,789 - Epoch: [153][ 40/ 117] Loss 0.277976 Top1 87.031250 Top5 98.720703 +2023-10-02 21:43:06,940 - Epoch: [153][ 50/ 117] Loss 0.285137 Top1 87.078125 Top5 98.757812 +2023-10-02 21:43:07,089 - Epoch: [153][ 60/ 117] Loss 0.295205 Top1 87.096354 Top5 98.704427 +2023-10-02 21:43:07,239 - Epoch: [153][ 70/ 117] Loss 0.292436 Top1 87.092634 Top5 98.727679 +2023-10-02 21:43:07,389 - Epoch: [153][ 80/ 117] Loss 0.291248 Top1 87.133789 Top5 98.710938 +2023-10-02 21:43:07,538 - Epoch: [153][ 90/ 117] Loss 0.291961 Top1 87.105035 Top5 98.654514 +2023-10-02 21:43:07,689 - Epoch: [153][ 100/ 117] Loss 0.294793 Top1 87.027344 Top5 98.648438 +2023-10-02 21:43:07,845 - Epoch: [153][ 110/ 117] Loss 0.298929 Top1 86.953125 Top5 98.632812 +2023-10-02 21:43:07,934 - Epoch: [153][ 117/ 117] Loss 0.297489 Top1 86.918478 Top5 98.637411 +2023-10-02 21:43:08,079 - ==> Top1: 86.918 Top5: 98.637 Loss: 0.297 + +2023-10-02 21:43:08,080 - ==> Confusion: +[[ 950 1 1 1 5 2 0 0 3 60 2 1 1 1 4 0 0 0 0 0 18] + [ 0 1061 1 0 3 20 1 18 0 0 0 0 1 0 2 3 1 0 12 2 6] + [ 2 0 982 9 0 0 12 7 0 1 2 0 9 4 0 2 1 2 11 3 9] + [ 1 2 11 988 1 2 1 1 3 0 3 0 4 5 26 1 1 4 13 2 20] + [ 27 2 0 0 978 5 1 0 2 9 0 0 0 3 6 5 8 0 0 1 3] + [ 3 35 1 2 5 995 0 22 3 7 2 8 1 9 3 0 1 0 4 2 13] + [ 0 3 32 0 0 1 1123 5 0 0 4 0 0 0 0 5 0 0 2 8 8] + [ 1 8 11 2 6 24 6 1068 1 2 7 3 4 4 1 0 1 1 49 8 11] + [ 14 1 0 2 3 2 0 3 979 39 11 1 1 12 10 1 3 0 3 1 3] + [ 84 1 0 0 8 2 1 0 24 957 0 1 1 23 6 2 1 0 0 1 7] + [ 2 2 10 6 0 1 2 2 16 2 974 1 1 9 3 0 2 1 6 2 11] + [ 0 0 1 0 1 12 0 5 0 0 0 966 16 10 0 3 0 14 1 2 4] + [ 2 1 2 3 0 2 1 1 0 1 5 38 973 2 1 7 1 10 2 5 11] + [ 0 0 0 0 2 7 0 0 15 6 1 7 0 1059 4 0 0 1 0 1 16] + [ 14 1 3 17 5 0 0 0 23 2 5 0 3 2 1005 0 2 3 11 0 5] + [ 0 0 2 2 4 0 0 0 0 0 1 5 6 0 0 1073 15 9 3 9 5] + [ 0 23 2 0 5 6 1 0 0 1 0 6 0 2 4 7 1087 0 1 4 12] + [ 0 0 1 1 1 0 1 0 0 0 0 3 19 1 2 5 1 997 0 1 5] + [ 2 6 2 12 0 1 0 15 3 2 5 0 1 0 8 0 0 0 999 2 10] + [ 0 0 4 2 0 4 11 5 0 1 1 12 2 2 1 0 6 0 0 1090 11] + [ 115 132 108 79 74 107 33 84 79 72 157 83 288 260 97 44 60 45 113 153 5722]] + +2023-10-02 21:43:08,081 - ==> Best [Top1: 87.015 Top5: 98.664 Sparsity:0.00 Params: 169472 on epoch: 145] +2023-10-02 21:43:08,081 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:43:08,087 - + +2023-10-02 21:43:08,087 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:43:09,101 - Epoch: [154][ 10/ 1236] Overall Loss 0.114227 Objective Loss 0.114227 LR 0.000250 Time 0.101345 +2023-10-02 21:43:09,308 - Epoch: [154][ 20/ 1236] Overall Loss 0.106988 Objective Loss 0.106988 LR 0.000250 Time 0.060958 +2023-10-02 21:43:09,515 - Epoch: [154][ 30/ 1236] Overall Loss 0.110170 Objective Loss 0.110170 LR 0.000250 Time 0.047539 +2023-10-02 21:43:09,724 - Epoch: [154][ 40/ 1236] Overall Loss 0.116887 Objective Loss 0.116887 LR 0.000250 Time 0.040879 +2023-10-02 21:43:09,932 - Epoch: [154][ 50/ 1236] Overall Loss 0.120770 Objective Loss 0.120770 LR 0.000250 Time 0.036843 +2023-10-02 21:43:10,141 - Epoch: [154][ 60/ 1236] Overall Loss 0.121909 Objective Loss 0.121909 LR 0.000250 Time 0.034187 +2023-10-02 21:43:10,348 - Epoch: [154][ 70/ 1236] Overall Loss 0.122183 Objective Loss 0.122183 LR 0.000250 Time 0.032249 +2023-10-02 21:43:10,558 - Epoch: [154][ 80/ 1236] Overall Loss 0.121226 Objective Loss 0.121226 LR 0.000250 Time 0.030835 +2023-10-02 21:43:10,764 - Epoch: [154][ 90/ 1236] Overall Loss 0.120887 Objective Loss 0.120887 LR 0.000250 Time 0.029701 +2023-10-02 21:43:10,974 - Epoch: [154][ 100/ 1236] Overall Loss 0.123333 Objective Loss 0.123333 LR 0.000250 Time 0.028823 +2023-10-02 21:43:11,180 - Epoch: [154][ 110/ 1236] Overall Loss 0.124772 Objective Loss 0.124772 LR 0.000250 Time 0.028077 +2023-10-02 21:43:11,387 - Epoch: [154][ 120/ 1236] Overall Loss 0.126941 Objective Loss 0.126941 LR 0.000250 Time 0.027455 +2023-10-02 21:43:11,590 - Epoch: [154][ 130/ 1236] Overall Loss 0.127660 Objective Loss 0.127660 LR 0.000250 Time 0.026907 +2023-10-02 21:43:11,794 - Epoch: [154][ 140/ 1236] Overall Loss 0.128614 Objective Loss 0.128614 LR 0.000250 Time 0.026440 +2023-10-02 21:43:11,998 - Epoch: [154][ 150/ 1236] Overall Loss 0.130517 Objective Loss 0.130517 LR 0.000250 Time 0.026036 +2023-10-02 21:43:12,210 - Epoch: [154][ 160/ 1236] Overall Loss 0.131052 Objective Loss 0.131052 LR 0.000250 Time 0.025731 +2023-10-02 21:43:12,420 - Epoch: [154][ 170/ 1236] Overall Loss 0.132153 Objective Loss 0.132153 LR 0.000250 Time 0.025449 +2023-10-02 21:43:12,633 - Epoch: [154][ 180/ 1236] Overall Loss 0.132671 Objective Loss 0.132671 LR 0.000250 Time 0.025209 +2023-10-02 21:43:12,843 - Epoch: [154][ 190/ 1236] Overall Loss 0.133710 Objective Loss 0.133710 LR 0.000250 Time 0.024984 +2023-10-02 21:43:13,054 - Epoch: [154][ 200/ 1236] Overall Loss 0.133988 Objective Loss 0.133988 LR 0.000250 Time 0.024781 +2023-10-02 21:43:13,266 - Epoch: [154][ 210/ 1236] Overall Loss 0.132825 Objective Loss 0.132825 LR 0.000250 Time 0.024605 +2023-10-02 21:43:13,476 - Epoch: [154][ 220/ 1236] Overall Loss 0.133022 Objective Loss 0.133022 LR 0.000250 Time 0.024437 +2023-10-02 21:43:13,688 - Epoch: [154][ 230/ 1236] Overall Loss 0.133859 Objective Loss 0.133859 LR 0.000250 Time 0.024291 +2023-10-02 21:43:13,897 - Epoch: [154][ 240/ 1236] Overall Loss 0.133624 Objective Loss 0.133624 LR 0.000250 Time 0.024145 +2023-10-02 21:43:14,107 - Epoch: [154][ 250/ 1236] Overall Loss 0.132768 Objective Loss 0.132768 LR 0.000250 Time 0.024018 +2023-10-02 21:43:14,317 - Epoch: [154][ 260/ 1236] Overall Loss 0.132876 Objective Loss 0.132876 LR 0.000250 Time 0.023894 +2023-10-02 21:43:14,527 - Epoch: [154][ 270/ 1236] Overall Loss 0.133751 Objective Loss 0.133751 LR 0.000250 Time 0.023782 +2023-10-02 21:43:14,737 - Epoch: [154][ 280/ 1236] Overall Loss 0.133472 Objective Loss 0.133472 LR 0.000250 Time 0.023676 +2023-10-02 21:43:14,947 - Epoch: [154][ 290/ 1236] Overall Loss 0.133368 Objective Loss 0.133368 LR 0.000250 Time 0.023578 +2023-10-02 21:43:15,157 - Epoch: [154][ 300/ 1236] Overall Loss 0.133816 Objective Loss 0.133816 LR 0.000250 Time 0.023486 +2023-10-02 21:43:15,367 - Epoch: [154][ 310/ 1236] Overall Loss 0.133605 Objective Loss 0.133605 LR 0.000250 Time 0.023401 +2023-10-02 21:43:15,577 - Epoch: [154][ 320/ 1236] Overall Loss 0.133298 Objective Loss 0.133298 LR 0.000250 Time 0.023320 +2023-10-02 21:43:15,787 - Epoch: [154][ 330/ 1236] Overall Loss 0.133448 Objective Loss 0.133448 LR 0.000250 Time 0.023245 +2023-10-02 21:43:15,997 - Epoch: [154][ 340/ 1236] Overall Loss 0.133928 Objective Loss 0.133928 LR 0.000250 Time 0.023174 +2023-10-02 21:43:16,205 - Epoch: [154][ 350/ 1236] Overall Loss 0.133487 Objective Loss 0.133487 LR 0.000250 Time 0.023102 +2023-10-02 21:43:16,410 - Epoch: [154][ 360/ 1236] Overall Loss 0.133480 Objective Loss 0.133480 LR 0.000250 Time 0.023027 +2023-10-02 21:43:16,614 - Epoch: [154][ 370/ 1236] Overall Loss 0.133584 Objective Loss 0.133584 LR 0.000250 Time 0.022955 +2023-10-02 21:43:16,818 - Epoch: [154][ 380/ 1236] Overall Loss 0.133922 Objective Loss 0.133922 LR 0.000250 Time 0.022888 +2023-10-02 21:43:17,022 - Epoch: [154][ 390/ 1236] Overall Loss 0.134313 Objective Loss 0.134313 LR 0.000250 Time 0.022822 +2023-10-02 21:43:17,226 - Epoch: [154][ 400/ 1236] Overall Loss 0.133786 Objective Loss 0.133786 LR 0.000250 Time 0.022761 +2023-10-02 21:43:17,430 - Epoch: [154][ 410/ 1236] Overall Loss 0.133594 Objective Loss 0.133594 LR 0.000250 Time 0.022703 +2023-10-02 21:43:17,634 - Epoch: [154][ 420/ 1236] Overall Loss 0.133598 Objective Loss 0.133598 LR 0.000250 Time 0.022648 +2023-10-02 21:43:17,838 - Epoch: [154][ 430/ 1236] Overall Loss 0.133478 Objective Loss 0.133478 LR 0.000250 Time 0.022594 +2023-10-02 21:43:18,042 - Epoch: [154][ 440/ 1236] Overall Loss 0.133367 Objective Loss 0.133367 LR 0.000250 Time 0.022543 +2023-10-02 21:43:18,246 - Epoch: [154][ 450/ 1236] Overall Loss 0.133275 Objective Loss 0.133275 LR 0.000250 Time 0.022495 +2023-10-02 21:43:18,450 - Epoch: [154][ 460/ 1236] Overall Loss 0.134016 Objective Loss 0.134016 LR 0.000250 Time 0.022448 +2023-10-02 21:43:18,653 - Epoch: [154][ 470/ 1236] Overall Loss 0.134529 Objective Loss 0.134529 LR 0.000250 Time 0.022403 +2023-10-02 21:43:18,857 - Epoch: [154][ 480/ 1236] Overall Loss 0.134538 Objective Loss 0.134538 LR 0.000250 Time 0.022360 +2023-10-02 21:43:19,061 - Epoch: [154][ 490/ 1236] Overall Loss 0.134538 Objective Loss 0.134538 LR 0.000250 Time 0.022319 +2023-10-02 21:43:19,265 - Epoch: [154][ 500/ 1236] Overall Loss 0.134623 Objective Loss 0.134623 LR 0.000250 Time 0.022280 +2023-10-02 21:43:19,469 - Epoch: [154][ 510/ 1236] Overall Loss 0.134424 Objective Loss 0.134424 LR 0.000250 Time 0.022242 +2023-10-02 21:43:19,673 - Epoch: [154][ 520/ 1236] Overall Loss 0.134321 Objective Loss 0.134321 LR 0.000250 Time 0.022206 +2023-10-02 21:43:19,877 - Epoch: [154][ 530/ 1236] Overall Loss 0.134462 Objective Loss 0.134462 LR 0.000250 Time 0.022171 +2023-10-02 21:43:20,081 - Epoch: [154][ 540/ 1236] Overall Loss 0.134515 Objective Loss 0.134515 LR 0.000250 Time 0.022138 +2023-10-02 21:43:20,285 - Epoch: [154][ 550/ 1236] Overall Loss 0.134496 Objective Loss 0.134496 LR 0.000250 Time 0.022105 +2023-10-02 21:43:20,489 - Epoch: [154][ 560/ 1236] Overall Loss 0.134558 Objective Loss 0.134558 LR 0.000250 Time 0.022075 +2023-10-02 21:43:20,696 - Epoch: [154][ 570/ 1236] Overall Loss 0.134341 Objective Loss 0.134341 LR 0.000250 Time 0.022049 +2023-10-02 21:43:20,903 - Epoch: [154][ 580/ 1236] Overall Loss 0.134144 Objective Loss 0.134144 LR 0.000250 Time 0.022026 +2023-10-02 21:43:21,111 - Epoch: [154][ 590/ 1236] Overall Loss 0.134185 Objective Loss 0.134185 LR 0.000250 Time 0.022004 +2023-10-02 21:43:21,319 - Epoch: [154][ 600/ 1236] Overall Loss 0.134411 Objective Loss 0.134411 LR 0.000250 Time 0.021984 +2023-10-02 21:43:21,526 - Epoch: [154][ 610/ 1236] Overall Loss 0.134424 Objective Loss 0.134424 LR 0.000250 Time 0.021963 +2023-10-02 21:43:21,734 - Epoch: [154][ 620/ 1236] Overall Loss 0.134330 Objective Loss 0.134330 LR 0.000250 Time 0.021943 +2023-10-02 21:43:21,942 - Epoch: [154][ 630/ 1236] Overall Loss 0.134121 Objective Loss 0.134121 LR 0.000250 Time 0.021924 +2023-10-02 21:43:22,150 - Epoch: [154][ 640/ 1236] Overall Loss 0.134479 Objective Loss 0.134479 LR 0.000250 Time 0.021906 +2023-10-02 21:43:22,358 - Epoch: [154][ 650/ 1236] Overall Loss 0.134628 Objective Loss 0.134628 LR 0.000250 Time 0.021888 +2023-10-02 21:43:22,566 - Epoch: [154][ 660/ 1236] Overall Loss 0.134612 Objective Loss 0.134612 LR 0.000250 Time 0.021871 +2023-10-02 21:43:22,773 - Epoch: [154][ 670/ 1236] Overall Loss 0.134581 Objective Loss 0.134581 LR 0.000250 Time 0.021854 +2023-10-02 21:43:22,982 - Epoch: [154][ 680/ 1236] Overall Loss 0.134841 Objective Loss 0.134841 LR 0.000250 Time 0.021839 +2023-10-02 21:43:23,189 - Epoch: [154][ 690/ 1236] Overall Loss 0.134877 Objective Loss 0.134877 LR 0.000250 Time 0.021822 +2023-10-02 21:43:23,398 - Epoch: [154][ 700/ 1236] Overall Loss 0.134925 Objective Loss 0.134925 LR 0.000250 Time 0.021808 +2023-10-02 21:43:23,605 - Epoch: [154][ 710/ 1236] Overall Loss 0.135233 Objective Loss 0.135233 LR 0.000250 Time 0.021792 +2023-10-02 21:43:23,814 - Epoch: [154][ 720/ 1236] Overall Loss 0.135213 Objective Loss 0.135213 LR 0.000250 Time 0.021779 +2023-10-02 21:43:24,021 - Epoch: [154][ 730/ 1236] Overall Loss 0.135136 Objective Loss 0.135136 LR 0.000250 Time 0.021764 +2023-10-02 21:43:24,229 - Epoch: [154][ 740/ 1236] Overall Loss 0.135115 Objective Loss 0.135115 LR 0.000250 Time 0.021751 +2023-10-02 21:43:24,437 - Epoch: [154][ 750/ 1236] Overall Loss 0.135222 Objective Loss 0.135222 LR 0.000250 Time 0.021738 +2023-10-02 21:43:24,645 - Epoch: [154][ 760/ 1236] Overall Loss 0.135170 Objective Loss 0.135170 LR 0.000250 Time 0.021725 +2023-10-02 21:43:24,853 - Epoch: [154][ 770/ 1236] Overall Loss 0.135320 Objective Loss 0.135320 LR 0.000250 Time 0.021712 +2023-10-02 21:43:25,061 - Epoch: [154][ 780/ 1236] Overall Loss 0.135481 Objective Loss 0.135481 LR 0.000250 Time 0.021700 +2023-10-02 21:43:25,269 - Epoch: [154][ 790/ 1236] Overall Loss 0.135456 Objective Loss 0.135456 LR 0.000250 Time 0.021688 +2023-10-02 21:43:25,477 - Epoch: [154][ 800/ 1236] Overall Loss 0.135667 Objective Loss 0.135667 LR 0.000250 Time 0.021676 +2023-10-02 21:43:25,685 - Epoch: [154][ 810/ 1236] Overall Loss 0.135798 Objective Loss 0.135798 LR 0.000250 Time 0.021665 +2023-10-02 21:43:25,893 - Epoch: [154][ 820/ 1236] Overall Loss 0.135808 Objective Loss 0.135808 LR 0.000250 Time 0.021654 +2023-10-02 21:43:26,100 - Epoch: [154][ 830/ 1236] Overall Loss 0.135829 Objective Loss 0.135829 LR 0.000250 Time 0.021643 +2023-10-02 21:43:26,308 - Epoch: [154][ 840/ 1236] Overall Loss 0.135866 Objective Loss 0.135866 LR 0.000250 Time 0.021632 +2023-10-02 21:43:26,518 - Epoch: [154][ 850/ 1236] Overall Loss 0.136056 Objective Loss 0.136056 LR 0.000250 Time 0.021624 +2023-10-02 21:43:26,727 - Epoch: [154][ 860/ 1236] Overall Loss 0.136158 Objective Loss 0.136158 LR 0.000250 Time 0.021616 +2023-10-02 21:43:26,937 - Epoch: [154][ 870/ 1236] Overall Loss 0.136092 Objective Loss 0.136092 LR 0.000250 Time 0.021608 +2023-10-02 21:43:27,146 - Epoch: [154][ 880/ 1236] Overall Loss 0.136285 Objective Loss 0.136285 LR 0.000250 Time 0.021599 +2023-10-02 21:43:27,356 - Epoch: [154][ 890/ 1236] Overall Loss 0.136315 Objective Loss 0.136315 LR 0.000250 Time 0.021592 +2023-10-02 21:43:27,565 - Epoch: [154][ 900/ 1236] Overall Loss 0.136202 Objective Loss 0.136202 LR 0.000250 Time 0.021584 +2023-10-02 21:43:27,775 - Epoch: [154][ 910/ 1236] Overall Loss 0.136232 Objective Loss 0.136232 LR 0.000250 Time 0.021577 +2023-10-02 21:43:27,985 - Epoch: [154][ 920/ 1236] Overall Loss 0.136302 Objective Loss 0.136302 LR 0.000250 Time 0.021571 +2023-10-02 21:43:28,195 - Epoch: [154][ 930/ 1236] Overall Loss 0.136252 Objective Loss 0.136252 LR 0.000250 Time 0.021564 +2023-10-02 21:43:28,404 - Epoch: [154][ 940/ 1236] Overall Loss 0.136290 Objective Loss 0.136290 LR 0.000250 Time 0.021557 +2023-10-02 21:43:28,614 - Epoch: [154][ 950/ 1236] Overall Loss 0.136655 Objective Loss 0.136655 LR 0.000250 Time 0.021550 +2023-10-02 21:43:28,823 - Epoch: [154][ 960/ 1236] Overall Loss 0.136620 Objective Loss 0.136620 LR 0.000250 Time 0.021543 +2023-10-02 21:43:29,033 - Epoch: [154][ 970/ 1236] Overall Loss 0.136786 Objective Loss 0.136786 LR 0.000250 Time 0.021537 +2023-10-02 21:43:29,242 - Epoch: [154][ 980/ 1236] Overall Loss 0.136756 Objective Loss 0.136756 LR 0.000250 Time 0.021531 +2023-10-02 21:43:29,452 - Epoch: [154][ 990/ 1236] Overall Loss 0.137116 Objective Loss 0.137116 LR 0.000250 Time 0.021525 +2023-10-02 21:43:29,661 - Epoch: [154][ 1000/ 1236] Overall Loss 0.137101 Objective Loss 0.137101 LR 0.000250 Time 0.021518 +2023-10-02 21:43:29,872 - Epoch: [154][ 1010/ 1236] Overall Loss 0.136954 Objective Loss 0.136954 LR 0.000250 Time 0.021513 +2023-10-02 21:43:30,081 - Epoch: [154][ 1020/ 1236] Overall Loss 0.137139 Objective Loss 0.137139 LR 0.000250 Time 0.021507 +2023-10-02 21:43:30,291 - Epoch: [154][ 1030/ 1236] Overall Loss 0.137254 Objective Loss 0.137254 LR 0.000250 Time 0.021502 +2023-10-02 21:43:30,500 - Epoch: [154][ 1040/ 1236] Overall Loss 0.137044 Objective Loss 0.137044 LR 0.000250 Time 0.021496 +2023-10-02 21:43:30,711 - Epoch: [154][ 1050/ 1236] Overall Loss 0.137127 Objective Loss 0.137127 LR 0.000250 Time 0.021491 +2023-10-02 21:43:30,920 - Epoch: [154][ 1060/ 1236] Overall Loss 0.137341 Objective Loss 0.137341 LR 0.000250 Time 0.021485 +2023-10-02 21:43:31,131 - Epoch: [154][ 1070/ 1236] Overall Loss 0.137378 Objective Loss 0.137378 LR 0.000250 Time 0.021481 +2023-10-02 21:43:31,340 - Epoch: [154][ 1080/ 1236] Overall Loss 0.137373 Objective Loss 0.137373 LR 0.000250 Time 0.021476 +2023-10-02 21:43:31,550 - Epoch: [154][ 1090/ 1236] Overall Loss 0.137268 Objective Loss 0.137268 LR 0.000250 Time 0.021472 +2023-10-02 21:43:31,760 - Epoch: [154][ 1100/ 1236] Overall Loss 0.137157 Objective Loss 0.137157 LR 0.000250 Time 0.021466 +2023-10-02 21:43:31,970 - Epoch: [154][ 1110/ 1236] Overall Loss 0.137306 Objective Loss 0.137306 LR 0.000250 Time 0.021462 +2023-10-02 21:43:32,179 - Epoch: [154][ 1120/ 1236] Overall Loss 0.137397 Objective Loss 0.137397 LR 0.000250 Time 0.021457 +2023-10-02 21:43:32,389 - Epoch: [154][ 1130/ 1236] Overall Loss 0.137505 Objective Loss 0.137505 LR 0.000250 Time 0.021453 +2023-10-02 21:43:32,599 - Epoch: [154][ 1140/ 1236] Overall Loss 0.137629 Objective Loss 0.137629 LR 0.000250 Time 0.021448 +2023-10-02 21:43:32,809 - Epoch: [154][ 1150/ 1236] Overall Loss 0.137678 Objective Loss 0.137678 LR 0.000250 Time 0.021444 +2023-10-02 21:43:33,018 - Epoch: [154][ 1160/ 1236] Overall Loss 0.137781 Objective Loss 0.137781 LR 0.000250 Time 0.021439 +2023-10-02 21:43:33,229 - Epoch: [154][ 1170/ 1236] Overall Loss 0.137884 Objective Loss 0.137884 LR 0.000250 Time 0.021435 +2023-10-02 21:43:33,438 - Epoch: [154][ 1180/ 1236] Overall Loss 0.137990 Objective Loss 0.137990 LR 0.000250 Time 0.021431 +2023-10-02 21:43:33,648 - Epoch: [154][ 1190/ 1236] Overall Loss 0.137931 Objective Loss 0.137931 LR 0.000250 Time 0.021427 +2023-10-02 21:43:33,857 - Epoch: [154][ 1200/ 1236] Overall Loss 0.137924 Objective Loss 0.137924 LR 0.000250 Time 0.021422 +2023-10-02 21:43:34,067 - Epoch: [154][ 1210/ 1236] Overall Loss 0.137812 Objective Loss 0.137812 LR 0.000250 Time 0.021419 +2023-10-02 21:43:34,277 - Epoch: [154][ 1220/ 1236] Overall Loss 0.137661 Objective Loss 0.137661 LR 0.000250 Time 0.021415 +2023-10-02 21:43:34,538 - Epoch: [154][ 1230/ 1236] Overall Loss 0.137548 Objective Loss 0.137548 LR 0.000250 Time 0.021452 +2023-10-02 21:43:34,658 - Epoch: [154][ 1236/ 1236] Overall Loss 0.137527 Objective Loss 0.137527 Top1 90.427699 Top5 98.981670 LR 0.000250 Time 0.021446 +2023-10-02 21:43:34,787 - --- validate (epoch=154)----------- +2023-10-02 21:43:34,787 - 29943 samples (256 per mini-batch) +2023-10-02 21:43:35,264 - Epoch: [154][ 10/ 117] Loss 0.304171 Top1 87.070312 Top5 98.437500 +2023-10-02 21:43:35,416 - Epoch: [154][ 20/ 117] Loss 0.300458 Top1 87.167969 Top5 98.476562 +2023-10-02 21:43:35,566 - Epoch: [154][ 30/ 117] Loss 0.289500 Top1 87.252604 Top5 98.541667 +2023-10-02 21:43:35,717 - Epoch: [154][ 40/ 117] Loss 0.286808 Top1 87.470703 Top5 98.505859 +2023-10-02 21:43:35,867 - Epoch: [154][ 50/ 117] Loss 0.288353 Top1 87.273438 Top5 98.531250 +2023-10-02 21:43:36,018 - Epoch: [154][ 60/ 117] Loss 0.289672 Top1 87.233073 Top5 98.548177 +2023-10-02 21:43:36,169 - Epoch: [154][ 70/ 117] Loss 0.294942 Top1 87.120536 Top5 98.521205 +2023-10-02 21:43:36,319 - Epoch: [154][ 80/ 117] Loss 0.293626 Top1 87.094727 Top5 98.496094 +2023-10-02 21:43:36,470 - Epoch: [154][ 90/ 117] Loss 0.299354 Top1 87.061632 Top5 98.502604 +2023-10-02 21:43:36,619 - Epoch: [154][ 100/ 117] Loss 0.297323 Top1 87.074219 Top5 98.539062 +2023-10-02 21:43:36,778 - Epoch: [154][ 110/ 117] Loss 0.296084 Top1 87.127131 Top5 98.558239 +2023-10-02 21:43:36,867 - Epoch: [154][ 117/ 117] Loss 0.295179 Top1 87.115519 Top5 98.570618 +2023-10-02 21:43:37,014 - ==> Top1: 87.116 Top5: 98.571 Loss: 0.295 + +2023-10-02 21:43:37,015 - ==> Confusion: +[[ 948 0 2 1 6 2 0 0 3 52 2 2 1 2 4 0 2 0 2 0 21] + [ 0 1063 1 0 2 18 2 21 1 0 0 2 1 0 1 3 1 0 9 2 4] + [ 1 0 978 4 3 0 17 8 0 2 1 0 9 2 1 2 1 1 12 3 11] + [ 1 4 15 992 0 1 2 1 1 0 2 0 6 4 22 3 1 5 11 1 17] + [ 25 2 0 1 972 5 0 0 0 12 0 0 0 3 13 3 9 0 0 1 4] + [ 4 34 2 3 3 979 0 29 3 7 2 6 3 9 4 0 2 0 4 3 19] + [ 0 2 23 0 0 1 1137 7 0 0 3 0 0 0 0 5 0 0 1 6 6] + [ 1 9 10 1 7 17 5 1076 1 3 5 1 4 5 1 1 0 2 53 8 8] + [ 13 0 0 1 3 3 0 0 980 36 10 1 0 15 12 3 5 0 3 1 3] + [ 85 1 2 0 6 1 0 0 26 955 0 0 0 21 9 3 1 0 0 0 9] + [ 1 2 9 9 0 1 2 1 11 1 975 1 2 10 5 0 0 1 8 3 11] + [ 0 1 1 0 1 7 0 6 0 0 0 972 19 6 0 2 0 16 0 3 1] + [ 1 2 1 3 0 1 1 0 0 3 1 31 976 1 4 7 1 14 3 8 10] + [ 0 0 0 0 3 4 0 0 8 8 1 3 1 1068 5 1 0 2 0 2 13] + [ 13 0 4 15 6 1 0 0 18 2 2 0 2 3 1017 0 1 1 11 0 5] + [ 0 0 2 1 6 0 0 0 0 0 0 5 9 0 0 1074 15 10 2 6 4] + [ 0 19 1 0 3 6 0 0 1 0 0 5 0 2 3 8 1096 0 1 5 11] + [ 0 0 2 0 0 0 1 0 0 0 0 5 12 0 4 7 1 1002 0 1 3] + [ 3 4 2 15 0 1 0 11 3 1 1 0 2 0 11 0 0 0 1002 0 12] + [ 0 1 7 2 2 1 7 4 0 0 0 16 4 1 2 1 9 2 1 1083 9] + [ 106 129 116 77 59 101 41 81 69 67 139 89 273 262 131 46 70 46 120 143 5740]] + +2023-10-02 21:43:37,016 - ==> Best [Top1: 87.116 Top5: 98.571 Sparsity:0.00 Params: 169472 on epoch: 154] +2023-10-02 21:43:37,016 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:43:37,030 - + +2023-10-02 21:43:37,030 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:43:38,197 - Epoch: [155][ 10/ 1236] Overall Loss 0.141614 Objective Loss 0.141614 LR 0.000250 Time 0.116643 +2023-10-02 21:43:38,406 - Epoch: [155][ 20/ 1236] Overall Loss 0.136760 Objective Loss 0.136760 LR 0.000250 Time 0.068753 +2023-10-02 21:43:38,616 - Epoch: [155][ 30/ 1236] Overall Loss 0.131508 Objective Loss 0.131508 LR 0.000250 Time 0.052814 +2023-10-02 21:43:38,826 - Epoch: [155][ 40/ 1236] Overall Loss 0.132464 Objective Loss 0.132464 LR 0.000250 Time 0.044850 +2023-10-02 21:43:39,036 - Epoch: [155][ 50/ 1236] Overall Loss 0.132484 Objective Loss 0.132484 LR 0.000250 Time 0.040046 +2023-10-02 21:43:39,245 - Epoch: [155][ 60/ 1236] Overall Loss 0.132516 Objective Loss 0.132516 LR 0.000250 Time 0.036860 +2023-10-02 21:43:39,455 - Epoch: [155][ 70/ 1236] Overall Loss 0.133875 Objective Loss 0.133875 LR 0.000250 Time 0.034566 +2023-10-02 21:43:39,665 - Epoch: [155][ 80/ 1236] Overall Loss 0.134657 Objective Loss 0.134657 LR 0.000250 Time 0.032863 +2023-10-02 21:43:39,874 - Epoch: [155][ 90/ 1236] Overall Loss 0.134589 Objective Loss 0.134589 LR 0.000250 Time 0.031524 +2023-10-02 21:43:40,084 - Epoch: [155][ 100/ 1236] Overall Loss 0.133490 Objective Loss 0.133490 LR 0.000250 Time 0.030465 +2023-10-02 21:43:40,295 - Epoch: [155][ 110/ 1236] Overall Loss 0.135050 Objective Loss 0.135050 LR 0.000250 Time 0.029600 +2023-10-02 21:43:40,508 - Epoch: [155][ 120/ 1236] Overall Loss 0.134118 Objective Loss 0.134118 LR 0.000250 Time 0.028908 +2023-10-02 21:43:40,717 - Epoch: [155][ 130/ 1236] Overall Loss 0.133893 Objective Loss 0.133893 LR 0.000250 Time 0.028290 +2023-10-02 21:43:40,926 - Epoch: [155][ 140/ 1236] Overall Loss 0.135070 Objective Loss 0.135070 LR 0.000250 Time 0.027762 +2023-10-02 21:43:41,143 - Epoch: [155][ 150/ 1236] Overall Loss 0.134144 Objective Loss 0.134144 LR 0.000250 Time 0.027346 +2023-10-02 21:43:41,358 - Epoch: [155][ 160/ 1236] Overall Loss 0.133653 Objective Loss 0.133653 LR 0.000250 Time 0.026977 +2023-10-02 21:43:41,575 - Epoch: [155][ 170/ 1236] Overall Loss 0.134250 Objective Loss 0.134250 LR 0.000250 Time 0.026658 +2023-10-02 21:43:41,789 - Epoch: [155][ 180/ 1236] Overall Loss 0.133598 Objective Loss 0.133598 LR 0.000250 Time 0.026365 +2023-10-02 21:43:42,005 - Epoch: [155][ 190/ 1236] Overall Loss 0.133547 Objective Loss 0.133547 LR 0.000250 Time 0.026111 +2023-10-02 21:43:42,221 - Epoch: [155][ 200/ 1236] Overall Loss 0.133768 Objective Loss 0.133768 LR 0.000250 Time 0.025879 +2023-10-02 21:43:42,436 - Epoch: [155][ 210/ 1236] Overall Loss 0.133592 Objective Loss 0.133592 LR 0.000250 Time 0.025670 +2023-10-02 21:43:42,651 - Epoch: [155][ 220/ 1236] Overall Loss 0.133596 Objective Loss 0.133596 LR 0.000250 Time 0.025475 +2023-10-02 21:43:42,866 - Epoch: [155][ 230/ 1236] Overall Loss 0.133464 Objective Loss 0.133464 LR 0.000250 Time 0.025301 +2023-10-02 21:43:43,081 - Epoch: [155][ 240/ 1236] Overall Loss 0.132803 Objective Loss 0.132803 LR 0.000250 Time 0.025139 +2023-10-02 21:43:43,296 - Epoch: [155][ 250/ 1236] Overall Loss 0.132433 Objective Loss 0.132433 LR 0.000250 Time 0.024990 +2023-10-02 21:43:43,511 - Epoch: [155][ 260/ 1236] Overall Loss 0.132213 Objective Loss 0.132213 LR 0.000250 Time 0.024850 +2023-10-02 21:43:43,726 - Epoch: [155][ 270/ 1236] Overall Loss 0.132823 Objective Loss 0.132823 LR 0.000250 Time 0.024725 +2023-10-02 21:43:43,942 - Epoch: [155][ 280/ 1236] Overall Loss 0.132637 Objective Loss 0.132637 LR 0.000250 Time 0.024611 +2023-10-02 21:43:44,158 - Epoch: [155][ 290/ 1236] Overall Loss 0.132621 Objective Loss 0.132621 LR 0.000250 Time 0.024503 +2023-10-02 21:43:44,369 - Epoch: [155][ 300/ 1236] Overall Loss 0.132022 Objective Loss 0.132022 LR 0.000250 Time 0.024384 +2023-10-02 21:43:44,580 - Epoch: [155][ 310/ 1236] Overall Loss 0.131797 Objective Loss 0.131797 LR 0.000250 Time 0.024278 +2023-10-02 21:43:44,789 - Epoch: [155][ 320/ 1236] Overall Loss 0.131993 Objective Loss 0.131993 LR 0.000250 Time 0.024171 +2023-10-02 21:43:45,000 - Epoch: [155][ 330/ 1236] Overall Loss 0.132145 Objective Loss 0.132145 LR 0.000250 Time 0.024079 +2023-10-02 21:43:45,209 - Epoch: [155][ 340/ 1236] Overall Loss 0.132040 Objective Loss 0.132040 LR 0.000250 Time 0.023984 +2023-10-02 21:43:45,420 - Epoch: [155][ 350/ 1236] Overall Loss 0.132712 Objective Loss 0.132712 LR 0.000250 Time 0.023901 +2023-10-02 21:43:45,629 - Epoch: [155][ 360/ 1236] Overall Loss 0.133262 Objective Loss 0.133262 LR 0.000250 Time 0.023816 +2023-10-02 21:43:45,841 - Epoch: [155][ 370/ 1236] Overall Loss 0.132822 Objective Loss 0.132822 LR 0.000250 Time 0.023743 +2023-10-02 21:43:46,050 - Epoch: [155][ 380/ 1236] Overall Loss 0.133472 Objective Loss 0.133472 LR 0.000250 Time 0.023669 +2023-10-02 21:43:46,272 - Epoch: [155][ 390/ 1236] Overall Loss 0.133856 Objective Loss 0.133856 LR 0.000250 Time 0.023631 +2023-10-02 21:43:46,488 - Epoch: [155][ 400/ 1236] Overall Loss 0.133803 Objective Loss 0.133803 LR 0.000250 Time 0.023579 +2023-10-02 21:43:46,711 - Epoch: [155][ 410/ 1236] Overall Loss 0.133860 Objective Loss 0.133860 LR 0.000250 Time 0.023545 +2023-10-02 21:43:46,927 - Epoch: [155][ 420/ 1236] Overall Loss 0.133887 Objective Loss 0.133887 LR 0.000250 Time 0.023498 +2023-10-02 21:43:47,149 - Epoch: [155][ 430/ 1236] Overall Loss 0.134218 Objective Loss 0.134218 LR 0.000250 Time 0.023468 +2023-10-02 21:43:47,365 - Epoch: [155][ 440/ 1236] Overall Loss 0.134431 Objective Loss 0.134431 LR 0.000250 Time 0.023424 +2023-10-02 21:43:47,584 - Epoch: [155][ 450/ 1236] Overall Loss 0.134564 Objective Loss 0.134564 LR 0.000250 Time 0.023389 +2023-10-02 21:43:47,793 - Epoch: [155][ 460/ 1236] Overall Loss 0.134592 Objective Loss 0.134592 LR 0.000250 Time 0.023334 +2023-10-02 21:43:48,004 - Epoch: [155][ 470/ 1236] Overall Loss 0.134745 Objective Loss 0.134745 LR 0.000250 Time 0.023288 +2023-10-02 21:43:48,213 - Epoch: [155][ 480/ 1236] Overall Loss 0.134846 Objective Loss 0.134846 LR 0.000250 Time 0.023237 +2023-10-02 21:43:48,425 - Epoch: [155][ 490/ 1236] Overall Loss 0.134801 Objective Loss 0.134801 LR 0.000250 Time 0.023194 +2023-10-02 21:43:48,634 - Epoch: [155][ 500/ 1236] Overall Loss 0.135069 Objective Loss 0.135069 LR 0.000250 Time 0.023147 +2023-10-02 21:43:48,845 - Epoch: [155][ 510/ 1236] Overall Loss 0.135451 Objective Loss 0.135451 LR 0.000250 Time 0.023108 +2023-10-02 21:43:49,054 - Epoch: [155][ 520/ 1236] Overall Loss 0.135633 Objective Loss 0.135633 LR 0.000250 Time 0.023065 +2023-10-02 21:43:49,270 - Epoch: [155][ 530/ 1236] Overall Loss 0.136109 Objective Loss 0.136109 LR 0.000250 Time 0.023036 +2023-10-02 21:43:49,480 - Epoch: [155][ 540/ 1236] Overall Loss 0.136351 Objective Loss 0.136351 LR 0.000250 Time 0.022997 +2023-10-02 21:43:49,692 - Epoch: [155][ 550/ 1236] Overall Loss 0.136109 Objective Loss 0.136109 LR 0.000250 Time 0.022964 +2023-10-02 21:43:49,902 - Epoch: [155][ 560/ 1236] Overall Loss 0.135885 Objective Loss 0.135885 LR 0.000250 Time 0.022928 +2023-10-02 21:43:50,114 - Epoch: [155][ 570/ 1236] Overall Loss 0.136183 Objective Loss 0.136183 LR 0.000250 Time 0.022898 +2023-10-02 21:43:50,323 - Epoch: [155][ 580/ 1236] Overall Loss 0.136240 Objective Loss 0.136240 LR 0.000250 Time 0.022863 +2023-10-02 21:43:50,536 - Epoch: [155][ 590/ 1236] Overall Loss 0.136422 Objective Loss 0.136422 LR 0.000250 Time 0.022835 +2023-10-02 21:43:50,746 - Epoch: [155][ 600/ 1236] Overall Loss 0.136378 Objective Loss 0.136378 LR 0.000250 Time 0.022804 +2023-10-02 21:43:50,958 - Epoch: [155][ 610/ 1236] Overall Loss 0.136453 Objective Loss 0.136453 LR 0.000250 Time 0.022777 +2023-10-02 21:43:51,168 - Epoch: [155][ 620/ 1236] Overall Loss 0.136298 Objective Loss 0.136298 LR 0.000250 Time 0.022748 +2023-10-02 21:43:51,380 - Epoch: [155][ 630/ 1236] Overall Loss 0.136199 Objective Loss 0.136199 LR 0.000250 Time 0.022723 +2023-10-02 21:43:51,590 - Epoch: [155][ 640/ 1236] Overall Loss 0.136428 Objective Loss 0.136428 LR 0.000250 Time 0.022696 +2023-10-02 21:43:51,801 - Epoch: [155][ 650/ 1236] Overall Loss 0.136300 Objective Loss 0.136300 LR 0.000250 Time 0.022671 +2023-10-02 21:43:52,011 - Epoch: [155][ 660/ 1236] Overall Loss 0.136284 Objective Loss 0.136284 LR 0.000250 Time 0.022645 +2023-10-02 21:43:52,224 - Epoch: [155][ 670/ 1236] Overall Loss 0.136047 Objective Loss 0.136047 LR 0.000250 Time 0.022624 +2023-10-02 21:43:52,433 - Epoch: [155][ 680/ 1236] Overall Loss 0.136215 Objective Loss 0.136215 LR 0.000250 Time 0.022599 +2023-10-02 21:43:52,645 - Epoch: [155][ 690/ 1236] Overall Loss 0.136209 Objective Loss 0.136209 LR 0.000250 Time 0.022578 +2023-10-02 21:43:52,858 - Epoch: [155][ 700/ 1236] Overall Loss 0.135992 Objective Loss 0.135992 LR 0.000250 Time 0.022558 +2023-10-02 21:43:53,074 - Epoch: [155][ 710/ 1236] Overall Loss 0.135741 Objective Loss 0.135741 LR 0.000250 Time 0.022545 +2023-10-02 21:43:53,285 - Epoch: [155][ 720/ 1236] Overall Loss 0.135835 Objective Loss 0.135835 LR 0.000250 Time 0.022525 +2023-10-02 21:43:53,501 - Epoch: [155][ 730/ 1236] Overall Loss 0.135945 Objective Loss 0.135945 LR 0.000250 Time 0.022511 +2023-10-02 21:43:53,712 - Epoch: [155][ 740/ 1236] Overall Loss 0.136214 Objective Loss 0.136214 LR 0.000250 Time 0.022491 +2023-10-02 21:43:53,928 - Epoch: [155][ 750/ 1236] Overall Loss 0.136479 Objective Loss 0.136479 LR 0.000250 Time 0.022479 +2023-10-02 21:43:54,139 - Epoch: [155][ 760/ 1236] Overall Loss 0.136479 Objective Loss 0.136479 LR 0.000250 Time 0.022461 +2023-10-02 21:43:54,354 - Epoch: [155][ 770/ 1236] Overall Loss 0.136369 Objective Loss 0.136369 LR 0.000250 Time 0.022449 +2023-10-02 21:43:54,565 - Epoch: [155][ 780/ 1236] Overall Loss 0.136599 Objective Loss 0.136599 LR 0.000250 Time 0.022431 +2023-10-02 21:43:54,780 - Epoch: [155][ 790/ 1236] Overall Loss 0.136611 Objective Loss 0.136611 LR 0.000250 Time 0.022418 +2023-10-02 21:43:54,991 - Epoch: [155][ 800/ 1236] Overall Loss 0.136535 Objective Loss 0.136535 LR 0.000250 Time 0.022401 +2023-10-02 21:43:55,206 - Epoch: [155][ 810/ 1236] Overall Loss 0.136327 Objective Loss 0.136327 LR 0.000250 Time 0.022390 +2023-10-02 21:43:55,417 - Epoch: [155][ 820/ 1236] Overall Loss 0.136441 Objective Loss 0.136441 LR 0.000250 Time 0.022374 +2023-10-02 21:43:55,632 - Epoch: [155][ 830/ 1236] Overall Loss 0.136468 Objective Loss 0.136468 LR 0.000250 Time 0.022363 +2023-10-02 21:43:55,843 - Epoch: [155][ 840/ 1236] Overall Loss 0.136472 Objective Loss 0.136472 LR 0.000250 Time 0.022348 +2023-10-02 21:43:56,058 - Epoch: [155][ 850/ 1236] Overall Loss 0.136531 Objective Loss 0.136531 LR 0.000250 Time 0.022338 +2023-10-02 21:43:56,269 - Epoch: [155][ 860/ 1236] Overall Loss 0.136562 Objective Loss 0.136562 LR 0.000250 Time 0.022323 +2023-10-02 21:43:56,485 - Epoch: [155][ 870/ 1236] Overall Loss 0.136829 Objective Loss 0.136829 LR 0.000250 Time 0.022314 +2023-10-02 21:43:56,696 - Epoch: [155][ 880/ 1236] Overall Loss 0.136923 Objective Loss 0.136923 LR 0.000250 Time 0.022299 +2023-10-02 21:43:56,911 - Epoch: [155][ 890/ 1236] Overall Loss 0.136801 Objective Loss 0.136801 LR 0.000250 Time 0.022291 +2023-10-02 21:43:57,123 - Epoch: [155][ 900/ 1236] Overall Loss 0.137207 Objective Loss 0.137207 LR 0.000250 Time 0.022277 +2023-10-02 21:43:57,338 - Epoch: [155][ 910/ 1236] Overall Loss 0.137227 Objective Loss 0.137227 LR 0.000250 Time 0.022269 +2023-10-02 21:43:57,549 - Epoch: [155][ 920/ 1236] Overall Loss 0.137484 Objective Loss 0.137484 LR 0.000250 Time 0.022256 +2023-10-02 21:43:57,764 - Epoch: [155][ 930/ 1236] Overall Loss 0.137492 Objective Loss 0.137492 LR 0.000250 Time 0.022248 +2023-10-02 21:43:57,976 - Epoch: [155][ 940/ 1236] Overall Loss 0.137464 Objective Loss 0.137464 LR 0.000250 Time 0.022235 +2023-10-02 21:43:58,191 - Epoch: [155][ 950/ 1236] Overall Loss 0.137494 Objective Loss 0.137494 LR 0.000250 Time 0.022228 +2023-10-02 21:43:58,402 - Epoch: [155][ 960/ 1236] Overall Loss 0.137466 Objective Loss 0.137466 LR 0.000250 Time 0.022215 +2023-10-02 21:43:58,617 - Epoch: [155][ 970/ 1236] Overall Loss 0.137144 Objective Loss 0.137144 LR 0.000250 Time 0.022208 +2023-10-02 21:43:58,829 - Epoch: [155][ 980/ 1236] Overall Loss 0.137276 Objective Loss 0.137276 LR 0.000250 Time 0.022197 +2023-10-02 21:43:59,044 - Epoch: [155][ 990/ 1236] Overall Loss 0.137130 Objective Loss 0.137130 LR 0.000250 Time 0.022190 +2023-10-02 21:43:59,255 - Epoch: [155][ 1000/ 1236] Overall Loss 0.136996 Objective Loss 0.136996 LR 0.000250 Time 0.022179 +2023-10-02 21:43:59,471 - Epoch: [155][ 1010/ 1236] Overall Loss 0.137068 Objective Loss 0.137068 LR 0.000250 Time 0.022173 +2023-10-02 21:43:59,682 - Epoch: [155][ 1020/ 1236] Overall Loss 0.137241 Objective Loss 0.137241 LR 0.000250 Time 0.022162 +2023-10-02 21:43:59,897 - Epoch: [155][ 1030/ 1236] Overall Loss 0.137273 Objective Loss 0.137273 LR 0.000250 Time 0.022155 +2023-10-02 21:44:00,108 - Epoch: [155][ 1040/ 1236] Overall Loss 0.137423 Objective Loss 0.137423 LR 0.000250 Time 0.022145 +2023-10-02 21:44:00,324 - Epoch: [155][ 1050/ 1236] Overall Loss 0.137325 Objective Loss 0.137325 LR 0.000250 Time 0.022139 +2023-10-02 21:44:00,535 - Epoch: [155][ 1060/ 1236] Overall Loss 0.137363 Objective Loss 0.137363 LR 0.000250 Time 0.022129 +2023-10-02 21:44:00,750 - Epoch: [155][ 1070/ 1236] Overall Loss 0.137253 Objective Loss 0.137253 LR 0.000250 Time 0.022123 +2023-10-02 21:44:00,961 - Epoch: [155][ 1080/ 1236] Overall Loss 0.137123 Objective Loss 0.137123 LR 0.000250 Time 0.022113 +2023-10-02 21:44:01,177 - Epoch: [155][ 1090/ 1236] Overall Loss 0.137059 Objective Loss 0.137059 LR 0.000250 Time 0.022108 +2023-10-02 21:44:01,388 - Epoch: [155][ 1100/ 1236] Overall Loss 0.137081 Objective Loss 0.137081 LR 0.000250 Time 0.022098 +2023-10-02 21:44:01,603 - Epoch: [155][ 1110/ 1236] Overall Loss 0.137045 Objective Loss 0.137045 LR 0.000250 Time 0.022093 +2023-10-02 21:44:01,814 - Epoch: [155][ 1120/ 1236] Overall Loss 0.137109 Objective Loss 0.137109 LR 0.000250 Time 0.022084 +2023-10-02 21:44:02,029 - Epoch: [155][ 1130/ 1236] Overall Loss 0.137058 Objective Loss 0.137058 LR 0.000250 Time 0.022079 +2023-10-02 21:44:02,240 - Epoch: [155][ 1140/ 1236] Overall Loss 0.137018 Objective Loss 0.137018 LR 0.000250 Time 0.022070 +2023-10-02 21:44:02,456 - Epoch: [155][ 1150/ 1236] Overall Loss 0.136939 Objective Loss 0.136939 LR 0.000250 Time 0.022065 +2023-10-02 21:44:02,667 - Epoch: [155][ 1160/ 1236] Overall Loss 0.136999 Objective Loss 0.136999 LR 0.000250 Time 0.022056 +2023-10-02 21:44:02,882 - Epoch: [155][ 1170/ 1236] Overall Loss 0.137088 Objective Loss 0.137088 LR 0.000250 Time 0.022051 +2023-10-02 21:44:03,093 - Epoch: [155][ 1180/ 1236] Overall Loss 0.137040 Objective Loss 0.137040 LR 0.000250 Time 0.022043 +2023-10-02 21:44:03,308 - Epoch: [155][ 1190/ 1236] Overall Loss 0.136993 Objective Loss 0.136993 LR 0.000250 Time 0.022039 +2023-10-02 21:44:03,519 - Epoch: [155][ 1200/ 1236] Overall Loss 0.137093 Objective Loss 0.137093 LR 0.000250 Time 0.022030 +2023-10-02 21:44:03,735 - Epoch: [155][ 1210/ 1236] Overall Loss 0.136940 Objective Loss 0.136940 LR 0.000250 Time 0.022026 +2023-10-02 21:44:03,946 - Epoch: [155][ 1220/ 1236] Overall Loss 0.136887 Objective Loss 0.136887 LR 0.000250 Time 0.022018 +2023-10-02 21:44:04,213 - Epoch: [155][ 1230/ 1236] Overall Loss 0.136776 Objective Loss 0.136776 LR 0.000250 Time 0.022056 +2023-10-02 21:44:04,336 - Epoch: [155][ 1236/ 1236] Overall Loss 0.136754 Objective Loss 0.136754 Top1 90.427699 Top5 99.592668 LR 0.000250 Time 0.022048 +2023-10-02 21:44:04,467 - --- validate (epoch=155)----------- +2023-10-02 21:44:04,468 - 29943 samples (256 per mini-batch) +2023-10-02 21:44:04,972 - Epoch: [155][ 10/ 117] Loss 0.297547 Top1 87.304688 Top5 98.554688 +2023-10-02 21:44:05,128 - Epoch: [155][ 20/ 117] Loss 0.287460 Top1 87.578125 Top5 98.730469 +2023-10-02 21:44:05,284 - Epoch: [155][ 30/ 117] Loss 0.295301 Top1 87.018229 Top5 98.671875 +2023-10-02 21:44:05,439 - Epoch: [155][ 40/ 117] Loss 0.294668 Top1 87.070312 Top5 98.701172 +2023-10-02 21:44:05,593 - Epoch: [155][ 50/ 117] Loss 0.289331 Top1 87.296875 Top5 98.765625 +2023-10-02 21:44:05,748 - Epoch: [155][ 60/ 117] Loss 0.291293 Top1 87.389323 Top5 98.723958 +2023-10-02 21:44:05,898 - Epoch: [155][ 70/ 117] Loss 0.287894 Top1 87.516741 Top5 98.733259 +2023-10-02 21:44:06,047 - Epoch: [155][ 80/ 117] Loss 0.293318 Top1 87.509766 Top5 98.715820 +2023-10-02 21:44:06,194 - Epoch: [155][ 90/ 117] Loss 0.295475 Top1 87.439236 Top5 98.702257 +2023-10-02 21:44:06,340 - Epoch: [155][ 100/ 117] Loss 0.298402 Top1 87.417969 Top5 98.675781 +2023-10-02 21:44:06,494 - Epoch: [155][ 110/ 117] Loss 0.299423 Top1 87.372159 Top5 98.661222 +2023-10-02 21:44:06,584 - Epoch: [155][ 117/ 117] Loss 0.301372 Top1 87.285843 Top5 98.637411 +2023-10-02 21:44:06,679 - ==> Top1: 87.286 Top5: 98.637 Loss: 0.301 + +2023-10-02 21:44:06,680 - ==> Confusion: +[[ 953 0 1 0 6 2 0 0 2 54 2 0 1 1 4 3 3 0 1 0 17] + [ 0 1074 1 2 5 12 0 17 1 2 0 2 0 0 1 3 0 0 4 1 6] + [ 2 0 991 9 0 0 7 9 0 0 1 1 10 2 0 3 3 2 6 4 6] + [ 1 4 11 987 1 2 2 2 2 1 3 0 5 4 27 2 1 6 8 1 19] + [ 24 2 0 1 978 3 1 0 1 12 2 0 1 3 4 4 8 0 0 0 6] + [ 4 31 0 2 4 998 2 21 1 5 2 6 0 14 3 2 3 0 1 3 14] + [ 0 3 31 3 0 2 1123 5 0 0 2 0 0 0 0 5 0 1 1 8 7] + [ 5 13 15 1 5 33 5 1063 0 5 2 3 4 5 2 0 0 3 38 7 9] + [ 17 1 0 1 2 1 0 2 970 42 10 4 0 15 14 1 2 0 2 1 4] + [ 97 1 1 1 5 4 0 0 25 946 1 2 0 18 6 3 0 0 0 1 8] + [ 3 4 11 11 0 1 2 4 11 0 965 1 0 14 5 1 3 0 3 1 13] + [ 0 0 2 0 0 14 0 2 0 0 0 969 15 7 0 2 1 14 0 3 6] + [ 0 1 1 3 0 0 2 0 0 1 4 34 970 1 1 7 1 10 0 7 25] + [ 0 0 0 0 3 7 0 0 7 8 4 5 0 1064 3 0 0 1 0 2 15] + [ 13 0 5 19 2 1 0 0 20 1 2 0 2 2 1012 0 2 1 11 0 8] + [ 0 0 1 1 5 0 0 1 0 0 1 5 7 0 0 1076 15 11 2 6 3] + [ 0 14 0 0 6 7 0 0 0 0 0 4 0 3 2 11 1095 0 1 7 11] + [ 0 1 1 1 1 0 1 0 0 0 0 3 19 0 3 5 0 997 0 2 4] + [ 2 6 2 20 0 1 0 15 3 1 4 2 3 0 9 0 0 1 987 0 12] + [ 0 0 3 1 0 2 7 5 0 0 1 16 5 2 0 1 8 0 2 1092 7] + [ 103 138 115 83 66 94 29 59 89 59 146 77 253 269 96 56 82 49 85 131 5826]] + +2023-10-02 21:44:06,681 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:44:06,681 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:44:06,695 - + +2023-10-02 21:44:06,695 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:44:07,723 - Epoch: [156][ 10/ 1236] Overall Loss 0.108417 Objective Loss 0.108417 LR 0.000250 Time 0.102792 +2023-10-02 21:44:07,934 - Epoch: [156][ 20/ 1236] Overall Loss 0.122925 Objective Loss 0.122925 LR 0.000250 Time 0.061887 +2023-10-02 21:44:08,144 - Epoch: [156][ 30/ 1236] Overall Loss 0.126483 Objective Loss 0.126483 LR 0.000250 Time 0.048262 +2023-10-02 21:44:08,353 - Epoch: [156][ 40/ 1236] Overall Loss 0.125498 Objective Loss 0.125498 LR 0.000250 Time 0.041403 +2023-10-02 21:44:08,561 - Epoch: [156][ 50/ 1236] Overall Loss 0.128416 Objective Loss 0.128416 LR 0.000250 Time 0.037284 +2023-10-02 21:44:08,771 - Epoch: [156][ 60/ 1236] Overall Loss 0.134506 Objective Loss 0.134506 LR 0.000250 Time 0.034560 +2023-10-02 21:44:08,979 - Epoch: [156][ 70/ 1236] Overall Loss 0.133922 Objective Loss 0.133922 LR 0.000250 Time 0.032584 +2023-10-02 21:44:09,189 - Epoch: [156][ 80/ 1236] Overall Loss 0.133554 Objective Loss 0.133554 LR 0.000250 Time 0.031125 +2023-10-02 21:44:09,398 - Epoch: [156][ 90/ 1236] Overall Loss 0.133850 Objective Loss 0.133850 LR 0.000250 Time 0.029982 +2023-10-02 21:44:09,608 - Epoch: [156][ 100/ 1236] Overall Loss 0.134902 Objective Loss 0.134902 LR 0.000250 Time 0.029082 +2023-10-02 21:44:09,816 - Epoch: [156][ 110/ 1236] Overall Loss 0.133416 Objective Loss 0.133416 LR 0.000250 Time 0.028334 +2023-10-02 21:44:10,026 - Epoch: [156][ 120/ 1236] Overall Loss 0.135566 Objective Loss 0.135566 LR 0.000250 Time 0.027715 +2023-10-02 21:44:10,234 - Epoch: [156][ 130/ 1236] Overall Loss 0.136934 Objective Loss 0.136934 LR 0.000250 Time 0.027180 +2023-10-02 21:44:10,443 - Epoch: [156][ 140/ 1236] Overall Loss 0.136504 Objective Loss 0.136504 LR 0.000250 Time 0.026733 +2023-10-02 21:44:10,652 - Epoch: [156][ 150/ 1236] Overall Loss 0.137839 Objective Loss 0.137839 LR 0.000250 Time 0.026341 +2023-10-02 21:44:10,864 - Epoch: [156][ 160/ 1236] Overall Loss 0.136750 Objective Loss 0.136750 LR 0.000250 Time 0.026016 +2023-10-02 21:44:11,074 - Epoch: [156][ 170/ 1236] Overall Loss 0.138049 Objective Loss 0.138049 LR 0.000250 Time 0.025713 +2023-10-02 21:44:11,285 - Epoch: [156][ 180/ 1236] Overall Loss 0.138451 Objective Loss 0.138451 LR 0.000250 Time 0.025457 +2023-10-02 21:44:11,496 - Epoch: [156][ 190/ 1236] Overall Loss 0.138212 Objective Loss 0.138212 LR 0.000250 Time 0.025215 +2023-10-02 21:44:11,707 - Epoch: [156][ 200/ 1236] Overall Loss 0.137642 Objective Loss 0.137642 LR 0.000250 Time 0.025012 +2023-10-02 21:44:11,918 - Epoch: [156][ 210/ 1236] Overall Loss 0.136937 Objective Loss 0.136937 LR 0.000250 Time 0.024816 +2023-10-02 21:44:12,131 - Epoch: [156][ 220/ 1236] Overall Loss 0.138033 Objective Loss 0.138033 LR 0.000250 Time 0.024654 +2023-10-02 21:44:12,340 - Epoch: [156][ 230/ 1236] Overall Loss 0.137731 Objective Loss 0.137731 LR 0.000250 Time 0.024491 +2023-10-02 21:44:12,553 - Epoch: [156][ 240/ 1236] Overall Loss 0.137609 Objective Loss 0.137609 LR 0.000250 Time 0.024356 +2023-10-02 21:44:12,762 - Epoch: [156][ 250/ 1236] Overall Loss 0.137496 Objective Loss 0.137496 LR 0.000250 Time 0.024217 +2023-10-02 21:44:12,975 - Epoch: [156][ 260/ 1236] Overall Loss 0.137031 Objective Loss 0.137031 LR 0.000250 Time 0.024104 +2023-10-02 21:44:13,184 - Epoch: [156][ 270/ 1236] Overall Loss 0.136630 Objective Loss 0.136630 LR 0.000250 Time 0.023983 +2023-10-02 21:44:13,396 - Epoch: [156][ 280/ 1236] Overall Loss 0.136478 Objective Loss 0.136478 LR 0.000250 Time 0.023885 +2023-10-02 21:44:13,605 - Epoch: [156][ 290/ 1236] Overall Loss 0.136174 Objective Loss 0.136174 LR 0.000250 Time 0.023780 +2023-10-02 21:44:13,817 - Epoch: [156][ 300/ 1236] Overall Loss 0.136244 Objective Loss 0.136244 LR 0.000250 Time 0.023692 +2023-10-02 21:44:14,027 - Epoch: [156][ 310/ 1236] Overall Loss 0.135477 Objective Loss 0.135477 LR 0.000250 Time 0.023604 +2023-10-02 21:44:14,238 - Epoch: [156][ 320/ 1236] Overall Loss 0.135172 Objective Loss 0.135172 LR 0.000250 Time 0.023526 +2023-10-02 21:44:14,449 - Epoch: [156][ 330/ 1236] Overall Loss 0.135219 Objective Loss 0.135219 LR 0.000250 Time 0.023447 +2023-10-02 21:44:14,659 - Epoch: [156][ 340/ 1236] Overall Loss 0.135108 Objective Loss 0.135108 LR 0.000250 Time 0.023374 +2023-10-02 21:44:14,868 - Epoch: [156][ 350/ 1236] Overall Loss 0.135289 Objective Loss 0.135289 LR 0.000250 Time 0.023303 +2023-10-02 21:44:15,078 - Epoch: [156][ 360/ 1236] Overall Loss 0.135487 Objective Loss 0.135487 LR 0.000250 Time 0.023238 +2023-10-02 21:44:15,286 - Epoch: [156][ 370/ 1236] Overall Loss 0.135493 Objective Loss 0.135493 LR 0.000250 Time 0.023173 +2023-10-02 21:44:15,497 - Epoch: [156][ 380/ 1236] Overall Loss 0.135057 Objective Loss 0.135057 LR 0.000250 Time 0.023116 +2023-10-02 21:44:15,705 - Epoch: [156][ 390/ 1236] Overall Loss 0.134394 Objective Loss 0.134394 LR 0.000250 Time 0.023054 +2023-10-02 21:44:15,916 - Epoch: [156][ 400/ 1236] Overall Loss 0.134126 Objective Loss 0.134126 LR 0.000250 Time 0.023003 +2023-10-02 21:44:16,124 - Epoch: [156][ 410/ 1236] Overall Loss 0.134553 Objective Loss 0.134553 LR 0.000250 Time 0.022950 +2023-10-02 21:44:16,334 - Epoch: [156][ 420/ 1236] Overall Loss 0.134585 Objective Loss 0.134585 LR 0.000250 Time 0.022902 +2023-10-02 21:44:16,543 - Epoch: [156][ 430/ 1236] Overall Loss 0.134863 Objective Loss 0.134863 LR 0.000250 Time 0.022855 +2023-10-02 21:44:16,753 - Epoch: [156][ 440/ 1236] Overall Loss 0.135696 Objective Loss 0.135696 LR 0.000250 Time 0.022812 +2023-10-02 21:44:16,962 - Epoch: [156][ 450/ 1236] Overall Loss 0.135423 Objective Loss 0.135423 LR 0.000250 Time 0.022765 +2023-10-02 21:44:17,172 - Epoch: [156][ 460/ 1236] Overall Loss 0.135553 Objective Loss 0.135553 LR 0.000250 Time 0.022726 +2023-10-02 21:44:17,380 - Epoch: [156][ 470/ 1236] Overall Loss 0.135564 Objective Loss 0.135564 LR 0.000250 Time 0.022684 +2023-10-02 21:44:17,590 - Epoch: [156][ 480/ 1236] Overall Loss 0.135693 Objective Loss 0.135693 LR 0.000250 Time 0.022648 +2023-10-02 21:44:17,799 - Epoch: [156][ 490/ 1236] Overall Loss 0.135819 Objective Loss 0.135819 LR 0.000250 Time 0.022612 +2023-10-02 21:44:18,010 - Epoch: [156][ 500/ 1236] Overall Loss 0.136144 Objective Loss 0.136144 LR 0.000250 Time 0.022580 +2023-10-02 21:44:18,219 - Epoch: [156][ 510/ 1236] Overall Loss 0.136649 Objective Loss 0.136649 LR 0.000250 Time 0.022546 +2023-10-02 21:44:18,429 - Epoch: [156][ 520/ 1236] Overall Loss 0.136224 Objective Loss 0.136224 LR 0.000250 Time 0.022516 +2023-10-02 21:44:18,638 - Epoch: [156][ 530/ 1236] Overall Loss 0.136281 Objective Loss 0.136281 LR 0.000250 Time 0.022484 +2023-10-02 21:44:18,848 - Epoch: [156][ 540/ 1236] Overall Loss 0.136160 Objective Loss 0.136160 LR 0.000250 Time 0.022457 +2023-10-02 21:44:19,057 - Epoch: [156][ 550/ 1236] Overall Loss 0.135902 Objective Loss 0.135902 LR 0.000250 Time 0.022428 +2023-10-02 21:44:19,267 - Epoch: [156][ 560/ 1236] Overall Loss 0.135683 Objective Loss 0.135683 LR 0.000250 Time 0.022402 +2023-10-02 21:44:19,476 - Epoch: [156][ 570/ 1236] Overall Loss 0.135453 Objective Loss 0.135453 LR 0.000250 Time 0.022375 +2023-10-02 21:44:19,686 - Epoch: [156][ 580/ 1236] Overall Loss 0.135356 Objective Loss 0.135356 LR 0.000250 Time 0.022351 +2023-10-02 21:44:19,895 - Epoch: [156][ 590/ 1236] Overall Loss 0.135324 Objective Loss 0.135324 LR 0.000250 Time 0.022326 +2023-10-02 21:44:20,105 - Epoch: [156][ 600/ 1236] Overall Loss 0.135348 Objective Loss 0.135348 LR 0.000250 Time 0.022303 +2023-10-02 21:44:20,314 - Epoch: [156][ 610/ 1236] Overall Loss 0.135490 Objective Loss 0.135490 LR 0.000250 Time 0.022280 +2023-10-02 21:44:20,524 - Epoch: [156][ 620/ 1236] Overall Loss 0.135559 Objective Loss 0.135559 LR 0.000250 Time 0.022259 +2023-10-02 21:44:20,733 - Epoch: [156][ 630/ 1236] Overall Loss 0.135608 Objective Loss 0.135608 LR 0.000250 Time 0.022237 +2023-10-02 21:44:20,943 - Epoch: [156][ 640/ 1236] Overall Loss 0.135927 Objective Loss 0.135927 LR 0.000250 Time 0.022218 +2023-10-02 21:44:21,152 - Epoch: [156][ 650/ 1236] Overall Loss 0.136103 Objective Loss 0.136103 LR 0.000250 Time 0.022197 +2023-10-02 21:44:21,362 - Epoch: [156][ 660/ 1236] Overall Loss 0.135686 Objective Loss 0.135686 LR 0.000250 Time 0.022178 +2023-10-02 21:44:21,571 - Epoch: [156][ 670/ 1236] Overall Loss 0.135692 Objective Loss 0.135692 LR 0.000250 Time 0.022158 +2023-10-02 21:44:21,781 - Epoch: [156][ 680/ 1236] Overall Loss 0.135883 Objective Loss 0.135883 LR 0.000250 Time 0.022141 +2023-10-02 21:44:21,990 - Epoch: [156][ 690/ 1236] Overall Loss 0.135836 Objective Loss 0.135836 LR 0.000250 Time 0.022123 +2023-10-02 21:44:22,200 - Epoch: [156][ 700/ 1236] Overall Loss 0.135737 Objective Loss 0.135737 LR 0.000250 Time 0.022106 +2023-10-02 21:44:22,409 - Epoch: [156][ 710/ 1236] Overall Loss 0.135610 Objective Loss 0.135610 LR 0.000250 Time 0.022088 +2023-10-02 21:44:22,619 - Epoch: [156][ 720/ 1236] Overall Loss 0.135512 Objective Loss 0.135512 LR 0.000250 Time 0.022073 +2023-10-02 21:44:22,828 - Epoch: [156][ 730/ 1236] Overall Loss 0.135524 Objective Loss 0.135524 LR 0.000250 Time 0.022057 +2023-10-02 21:44:23,038 - Epoch: [156][ 740/ 1236] Overall Loss 0.135545 Objective Loss 0.135545 LR 0.000250 Time 0.022042 +2023-10-02 21:44:23,246 - Epoch: [156][ 750/ 1236] Overall Loss 0.135856 Objective Loss 0.135856 LR 0.000250 Time 0.022025 +2023-10-02 21:44:23,456 - Epoch: [156][ 760/ 1236] Overall Loss 0.135800 Objective Loss 0.135800 LR 0.000250 Time 0.022011 +2023-10-02 21:44:23,665 - Epoch: [156][ 770/ 1236] Overall Loss 0.135811 Objective Loss 0.135811 LR 0.000250 Time 0.021996 +2023-10-02 21:44:23,875 - Epoch: [156][ 780/ 1236] Overall Loss 0.135887 Objective Loss 0.135887 LR 0.000250 Time 0.021983 +2023-10-02 21:44:24,084 - Epoch: [156][ 790/ 1236] Overall Loss 0.135868 Objective Loss 0.135868 LR 0.000250 Time 0.021969 +2023-10-02 21:44:24,294 - Epoch: [156][ 800/ 1236] Overall Loss 0.135786 Objective Loss 0.135786 LR 0.000250 Time 0.021957 +2023-10-02 21:44:24,502 - Epoch: [156][ 810/ 1236] Overall Loss 0.135544 Objective Loss 0.135544 LR 0.000250 Time 0.021942 +2023-10-02 21:44:24,712 - Epoch: [156][ 820/ 1236] Overall Loss 0.135741 Objective Loss 0.135741 LR 0.000250 Time 0.021930 +2023-10-02 21:44:24,921 - Epoch: [156][ 830/ 1236] Overall Loss 0.135893 Objective Loss 0.135893 LR 0.000250 Time 0.021918 +2023-10-02 21:44:25,130 - Epoch: [156][ 840/ 1236] Overall Loss 0.135930 Objective Loss 0.135930 LR 0.000250 Time 0.021905 +2023-10-02 21:44:25,339 - Epoch: [156][ 850/ 1236] Overall Loss 0.136056 Objective Loss 0.136056 LR 0.000250 Time 0.021891 +2023-10-02 21:44:25,549 - Epoch: [156][ 860/ 1236] Overall Loss 0.136148 Objective Loss 0.136148 LR 0.000250 Time 0.021881 +2023-10-02 21:44:25,758 - Epoch: [156][ 870/ 1236] Overall Loss 0.136047 Objective Loss 0.136047 LR 0.000250 Time 0.021869 +2023-10-02 21:44:25,969 - Epoch: [156][ 880/ 1236] Overall Loss 0.136054 Objective Loss 0.136054 LR 0.000250 Time 0.021859 +2023-10-02 21:44:26,177 - Epoch: [156][ 890/ 1236] Overall Loss 0.136159 Objective Loss 0.136159 LR 0.000250 Time 0.021848 +2023-10-02 21:44:26,388 - Epoch: [156][ 900/ 1236] Overall Loss 0.136224 Objective Loss 0.136224 LR 0.000250 Time 0.021839 +2023-10-02 21:44:26,596 - Epoch: [156][ 910/ 1236] Overall Loss 0.136258 Objective Loss 0.136258 LR 0.000250 Time 0.021828 +2023-10-02 21:44:26,807 - Epoch: [156][ 920/ 1236] Overall Loss 0.136190 Objective Loss 0.136190 LR 0.000250 Time 0.021819 +2023-10-02 21:44:27,016 - Epoch: [156][ 930/ 1236] Overall Loss 0.136262 Objective Loss 0.136262 LR 0.000250 Time 0.021809 +2023-10-02 21:44:27,227 - Epoch: [156][ 940/ 1236] Overall Loss 0.136377 Objective Loss 0.136377 LR 0.000250 Time 0.021801 +2023-10-02 21:44:27,436 - Epoch: [156][ 950/ 1236] Overall Loss 0.136287 Objective Loss 0.136287 LR 0.000250 Time 0.021790 +2023-10-02 21:44:27,646 - Epoch: [156][ 960/ 1236] Overall Loss 0.136016 Objective Loss 0.136016 LR 0.000250 Time 0.021781 +2023-10-02 21:44:27,854 - Epoch: [156][ 970/ 1236] Overall Loss 0.136065 Objective Loss 0.136065 LR 0.000250 Time 0.021770 +2023-10-02 21:44:28,064 - Epoch: [156][ 980/ 1236] Overall Loss 0.136215 Objective Loss 0.136215 LR 0.000250 Time 0.021760 +2023-10-02 21:44:28,273 - Epoch: [156][ 990/ 1236] Overall Loss 0.136176 Objective Loss 0.136176 LR 0.000250 Time 0.021751 +2023-10-02 21:44:28,483 - Epoch: [156][ 1000/ 1236] Overall Loss 0.136250 Objective Loss 0.136250 LR 0.000250 Time 0.021744 +2023-10-02 21:44:28,692 - Epoch: [156][ 1010/ 1236] Overall Loss 0.136124 Objective Loss 0.136124 LR 0.000250 Time 0.021735 +2023-10-02 21:44:28,903 - Epoch: [156][ 1020/ 1236] Overall Loss 0.136122 Objective Loss 0.136122 LR 0.000250 Time 0.021728 +2023-10-02 21:44:29,112 - Epoch: [156][ 1030/ 1236] Overall Loss 0.136058 Objective Loss 0.136058 LR 0.000250 Time 0.021719 +2023-10-02 21:44:29,322 - Epoch: [156][ 1040/ 1236] Overall Loss 0.135978 Objective Loss 0.135978 LR 0.000250 Time 0.021712 +2023-10-02 21:44:29,531 - Epoch: [156][ 1050/ 1236] Overall Loss 0.135987 Objective Loss 0.135987 LR 0.000250 Time 0.021704 +2023-10-02 21:44:29,740 - Epoch: [156][ 1060/ 1236] Overall Loss 0.136118 Objective Loss 0.136118 LR 0.000250 Time 0.021696 +2023-10-02 21:44:29,949 - Epoch: [156][ 1070/ 1236] Overall Loss 0.136152 Objective Loss 0.136152 LR 0.000250 Time 0.021688 +2023-10-02 21:44:30,159 - Epoch: [156][ 1080/ 1236] Overall Loss 0.136071 Objective Loss 0.136071 LR 0.000250 Time 0.021682 +2023-10-02 21:44:30,368 - Epoch: [156][ 1090/ 1236] Overall Loss 0.136156 Objective Loss 0.136156 LR 0.000250 Time 0.021674 +2023-10-02 21:44:30,578 - Epoch: [156][ 1100/ 1236] Overall Loss 0.136254 Objective Loss 0.136254 LR 0.000250 Time 0.021668 +2023-10-02 21:44:30,787 - Epoch: [156][ 1110/ 1236] Overall Loss 0.136163 Objective Loss 0.136163 LR 0.000250 Time 0.021661 +2023-10-02 21:44:30,998 - Epoch: [156][ 1120/ 1236] Overall Loss 0.136144 Objective Loss 0.136144 LR 0.000250 Time 0.021655 +2023-10-02 21:44:31,207 - Epoch: [156][ 1130/ 1236] Overall Loss 0.136170 Objective Loss 0.136170 LR 0.000250 Time 0.021648 +2023-10-02 21:44:31,417 - Epoch: [156][ 1140/ 1236] Overall Loss 0.136142 Objective Loss 0.136142 LR 0.000250 Time 0.021643 +2023-10-02 21:44:31,626 - Epoch: [156][ 1150/ 1236] Overall Loss 0.136138 Objective Loss 0.136138 LR 0.000250 Time 0.021636 +2023-10-02 21:44:31,836 - Epoch: [156][ 1160/ 1236] Overall Loss 0.136171 Objective Loss 0.136171 LR 0.000250 Time 0.021630 +2023-10-02 21:44:32,045 - Epoch: [156][ 1170/ 1236] Overall Loss 0.136457 Objective Loss 0.136457 LR 0.000250 Time 0.021624 +2023-10-02 21:44:32,256 - Epoch: [156][ 1180/ 1236] Overall Loss 0.136453 Objective Loss 0.136453 LR 0.000250 Time 0.021619 +2023-10-02 21:44:32,465 - Epoch: [156][ 1190/ 1236] Overall Loss 0.136358 Objective Loss 0.136358 LR 0.000250 Time 0.021612 +2023-10-02 21:44:32,675 - Epoch: [156][ 1200/ 1236] Overall Loss 0.136416 Objective Loss 0.136416 LR 0.000250 Time 0.021607 +2023-10-02 21:44:32,885 - Epoch: [156][ 1210/ 1236] Overall Loss 0.136308 Objective Loss 0.136308 LR 0.000250 Time 0.021601 +2023-10-02 21:44:33,095 - Epoch: [156][ 1220/ 1236] Overall Loss 0.136337 Objective Loss 0.136337 LR 0.000250 Time 0.021597 +2023-10-02 21:44:33,359 - Epoch: [156][ 1230/ 1236] Overall Loss 0.136333 Objective Loss 0.136333 LR 0.000250 Time 0.021636 +2023-10-02 21:44:33,482 - Epoch: [156][ 1236/ 1236] Overall Loss 0.136397 Objective Loss 0.136397 Top1 90.224033 Top5 98.981670 LR 0.000250 Time 0.021630 +2023-10-02 21:44:33,613 - --- validate (epoch=156)----------- +2023-10-02 21:44:33,613 - 29943 samples (256 per mini-batch) +2023-10-02 21:44:34,110 - Epoch: [156][ 10/ 117] Loss 0.255918 Top1 88.398438 Top5 99.101562 +2023-10-02 21:44:34,263 - Epoch: [156][ 20/ 117] Loss 0.296669 Top1 87.832031 Top5 98.789062 +2023-10-02 21:44:34,416 - Epoch: [156][ 30/ 117] Loss 0.297237 Top1 87.617188 Top5 98.763021 +2023-10-02 21:44:34,567 - Epoch: [156][ 40/ 117] Loss 0.294691 Top1 87.382812 Top5 98.769531 +2023-10-02 21:44:34,719 - Epoch: [156][ 50/ 117] Loss 0.297175 Top1 87.187500 Top5 98.679688 +2023-10-02 21:44:34,872 - Epoch: [156][ 60/ 117] Loss 0.296341 Top1 87.187500 Top5 98.710938 +2023-10-02 21:44:35,024 - Epoch: [156][ 70/ 117] Loss 0.302076 Top1 87.114955 Top5 98.677455 +2023-10-02 21:44:35,174 - Epoch: [156][ 80/ 117] Loss 0.305417 Top1 87.128906 Top5 98.618164 +2023-10-02 21:44:35,326 - Epoch: [156][ 90/ 117] Loss 0.302875 Top1 87.061632 Top5 98.580729 +2023-10-02 21:44:35,477 - Epoch: [156][ 100/ 117] Loss 0.302759 Top1 86.988281 Top5 98.566406 +2023-10-02 21:44:35,635 - Epoch: [156][ 110/ 117] Loss 0.301034 Top1 86.953125 Top5 98.593750 +2023-10-02 21:44:35,725 - Epoch: [156][ 117/ 117] Loss 0.300491 Top1 87.015329 Top5 98.607354 +2023-10-02 21:44:35,862 - ==> Top1: 87.015 Top5: 98.607 Loss: 0.300 + +2023-10-02 21:44:35,862 - ==> Confusion: +[[ 962 1 3 0 8 2 0 0 4 45 2 0 1 1 4 0 1 1 0 0 15] + [ 0 1061 2 0 6 17 0 21 1 1 0 1 3 0 0 2 1 0 8 1 6] + [ 3 1 992 4 2 0 14 8 1 0 1 1 5 4 1 2 0 2 6 2 7] + [ 0 2 14 983 2 1 2 1 5 0 5 0 5 4 28 1 1 5 9 0 21] + [ 27 3 0 1 970 5 0 0 0 12 1 0 1 2 11 5 10 0 0 0 2] + [ 3 29 0 1 4 1002 2 22 2 5 1 7 0 9 3 0 1 1 4 3 17] + [ 1 2 27 0 0 1 1135 4 0 0 3 0 0 1 0 5 0 0 1 6 5] + [ 3 17 9 2 5 23 5 1068 1 3 6 4 5 5 1 0 0 3 42 6 10] + [ 19 1 0 0 2 2 0 0 975 38 9 0 3 18 13 0 3 0 1 1 4] + [ 111 1 1 0 6 3 0 0 18 940 0 1 0 18 9 2 1 0 0 0 8] + [ 5 0 11 5 0 1 4 1 11 0 974 3 0 12 5 0 2 2 5 1 11] + [ 0 0 2 0 1 13 0 3 0 2 0 971 14 6 0 1 0 15 0 3 4] + [ 0 1 3 3 0 1 0 0 2 0 6 40 966 0 2 8 0 13 2 6 15] + [ 0 0 2 0 3 8 0 0 10 10 3 7 0 1058 4 0 0 1 0 0 13] + [ 11 0 4 14 4 1 0 0 16 1 2 0 4 2 1021 0 1 1 8 0 11] + [ 0 0 1 1 6 0 0 0 0 0 1 6 6 0 0 1073 15 10 2 6 7] + [ 0 15 3 1 6 6 1 0 0 0 0 4 0 3 4 8 1092 0 1 3 14] + [ 0 0 1 1 0 0 1 0 0 0 0 3 20 1 2 3 1 1001 0 1 3] + [ 2 4 4 13 1 1 0 17 3 2 3 2 2 0 15 0 1 0 987 0 11] + [ 0 0 2 0 0 2 7 8 0 0 1 11 6 3 1 2 8 0 1 1092 8] + [ 140 128 121 84 64 94 33 62 81 56 158 81 302 241 118 51 57 60 108 134 5732]] + +2023-10-02 21:44:35,864 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:44:35,864 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:44:35,870 - + +2023-10-02 21:44:35,870 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:44:37,026 - Epoch: [157][ 10/ 1236] Overall Loss 0.126171 Objective Loss 0.126171 LR 0.000250 Time 0.115549 +2023-10-02 21:44:37,235 - Epoch: [157][ 20/ 1236] Overall Loss 0.130705 Objective Loss 0.130705 LR 0.000250 Time 0.068216 +2023-10-02 21:44:37,444 - Epoch: [157][ 30/ 1236] Overall Loss 0.125874 Objective Loss 0.125874 LR 0.000250 Time 0.052417 +2023-10-02 21:44:37,653 - Epoch: [157][ 40/ 1236] Overall Loss 0.128940 Objective Loss 0.128940 LR 0.000250 Time 0.044538 +2023-10-02 21:44:37,861 - Epoch: [157][ 50/ 1236] Overall Loss 0.131490 Objective Loss 0.131490 LR 0.000250 Time 0.039791 +2023-10-02 21:44:38,071 - Epoch: [157][ 60/ 1236] Overall Loss 0.128019 Objective Loss 0.128019 LR 0.000250 Time 0.036650 +2023-10-02 21:44:38,279 - Epoch: [157][ 70/ 1236] Overall Loss 0.130915 Objective Loss 0.130915 LR 0.000250 Time 0.034387 +2023-10-02 21:44:38,489 - Epoch: [157][ 80/ 1236] Overall Loss 0.131202 Objective Loss 0.131202 LR 0.000250 Time 0.032703 +2023-10-02 21:44:38,697 - Epoch: [157][ 90/ 1236] Overall Loss 0.131009 Objective Loss 0.131009 LR 0.000250 Time 0.031380 +2023-10-02 21:44:38,907 - Epoch: [157][ 100/ 1236] Overall Loss 0.132022 Objective Loss 0.132022 LR 0.000250 Time 0.030338 +2023-10-02 21:44:39,116 - Epoch: [157][ 110/ 1236] Overall Loss 0.131693 Objective Loss 0.131693 LR 0.000250 Time 0.029475 +2023-10-02 21:44:39,324 - Epoch: [157][ 120/ 1236] Overall Loss 0.131558 Objective Loss 0.131558 LR 0.000250 Time 0.028753 +2023-10-02 21:44:39,532 - Epoch: [157][ 130/ 1236] Overall Loss 0.129941 Objective Loss 0.129941 LR 0.000250 Time 0.028127 +2023-10-02 21:44:39,741 - Epoch: [157][ 140/ 1236] Overall Loss 0.129803 Objective Loss 0.129803 LR 0.000250 Time 0.027607 +2023-10-02 21:44:39,949 - Epoch: [157][ 150/ 1236] Overall Loss 0.130712 Objective Loss 0.130712 LR 0.000250 Time 0.027145 +2023-10-02 21:44:40,161 - Epoch: [157][ 160/ 1236] Overall Loss 0.130530 Objective Loss 0.130530 LR 0.000250 Time 0.026774 +2023-10-02 21:44:40,370 - Epoch: [157][ 170/ 1236] Overall Loss 0.131043 Objective Loss 0.131043 LR 0.000250 Time 0.026425 +2023-10-02 21:44:40,581 - Epoch: [157][ 180/ 1236] Overall Loss 0.131520 Objective Loss 0.131520 LR 0.000250 Time 0.026128 +2023-10-02 21:44:40,791 - Epoch: [157][ 190/ 1236] Overall Loss 0.131428 Objective Loss 0.131428 LR 0.000250 Time 0.025850 +2023-10-02 21:44:41,003 - Epoch: [157][ 200/ 1236] Overall Loss 0.131673 Objective Loss 0.131673 LR 0.000250 Time 0.025619 +2023-10-02 21:44:41,212 - Epoch: [157][ 210/ 1236] Overall Loss 0.131223 Objective Loss 0.131223 LR 0.000250 Time 0.025392 +2023-10-02 21:44:41,425 - Epoch: [157][ 220/ 1236] Overall Loss 0.131003 Objective Loss 0.131003 LR 0.000250 Time 0.025203 +2023-10-02 21:44:41,634 - Epoch: [157][ 230/ 1236] Overall Loss 0.131191 Objective Loss 0.131191 LR 0.000250 Time 0.025013 +2023-10-02 21:44:41,846 - Epoch: [157][ 240/ 1236] Overall Loss 0.131172 Objective Loss 0.131172 LR 0.000250 Time 0.024855 +2023-10-02 21:44:42,055 - Epoch: [157][ 250/ 1236] Overall Loss 0.131430 Objective Loss 0.131430 LR 0.000250 Time 0.024696 +2023-10-02 21:44:42,266 - Epoch: [157][ 260/ 1236] Overall Loss 0.131546 Objective Loss 0.131546 LR 0.000250 Time 0.024556 +2023-10-02 21:44:42,476 - Epoch: [157][ 270/ 1236] Overall Loss 0.131716 Objective Loss 0.131716 LR 0.000250 Time 0.024419 +2023-10-02 21:44:42,688 - Epoch: [157][ 280/ 1236] Overall Loss 0.132599 Objective Loss 0.132599 LR 0.000250 Time 0.024304 +2023-10-02 21:44:42,897 - Epoch: [157][ 290/ 1236] Overall Loss 0.132909 Objective Loss 0.132909 LR 0.000250 Time 0.024184 +2023-10-02 21:44:43,107 - Epoch: [157][ 300/ 1236] Overall Loss 0.133301 Objective Loss 0.133301 LR 0.000250 Time 0.024076 +2023-10-02 21:44:43,315 - Epoch: [157][ 310/ 1236] Overall Loss 0.133707 Objective Loss 0.133707 LR 0.000250 Time 0.023972 +2023-10-02 21:44:43,525 - Epoch: [157][ 320/ 1236] Overall Loss 0.133462 Objective Loss 0.133462 LR 0.000250 Time 0.023878 +2023-10-02 21:44:43,734 - Epoch: [157][ 330/ 1236] Overall Loss 0.133025 Objective Loss 0.133025 LR 0.000250 Time 0.023785 +2023-10-02 21:44:43,945 - Epoch: [157][ 340/ 1236] Overall Loss 0.133156 Objective Loss 0.133156 LR 0.000250 Time 0.023705 +2023-10-02 21:44:44,157 - Epoch: [157][ 350/ 1236] Overall Loss 0.133795 Objective Loss 0.133795 LR 0.000250 Time 0.023631 +2023-10-02 21:44:44,370 - Epoch: [157][ 360/ 1236] Overall Loss 0.133854 Objective Loss 0.133854 LR 0.000250 Time 0.023563 +2023-10-02 21:44:44,582 - Epoch: [157][ 370/ 1236] Overall Loss 0.134379 Objective Loss 0.134379 LR 0.000250 Time 0.023500 +2023-10-02 21:44:44,794 - Epoch: [157][ 380/ 1236] Overall Loss 0.134358 Objective Loss 0.134358 LR 0.000250 Time 0.023439 +2023-10-02 21:44:45,007 - Epoch: [157][ 390/ 1236] Overall Loss 0.134554 Objective Loss 0.134554 LR 0.000250 Time 0.023383 +2023-10-02 21:44:45,219 - Epoch: [157][ 400/ 1236] Overall Loss 0.134706 Objective Loss 0.134706 LR 0.000250 Time 0.023326 +2023-10-02 21:44:45,431 - Epoch: [157][ 410/ 1236] Overall Loss 0.134999 Objective Loss 0.134999 LR 0.000250 Time 0.023275 +2023-10-02 21:44:45,643 - Epoch: [157][ 420/ 1236] Overall Loss 0.135174 Objective Loss 0.135174 LR 0.000250 Time 0.023225 +2023-10-02 21:44:45,856 - Epoch: [157][ 430/ 1236] Overall Loss 0.135085 Objective Loss 0.135085 LR 0.000250 Time 0.023179 +2023-10-02 21:44:46,068 - Epoch: [157][ 440/ 1236] Overall Loss 0.135362 Objective Loss 0.135362 LR 0.000250 Time 0.023134 +2023-10-02 21:44:46,281 - Epoch: [157][ 450/ 1236] Overall Loss 0.135373 Objective Loss 0.135373 LR 0.000250 Time 0.023091 +2023-10-02 21:44:46,493 - Epoch: [157][ 460/ 1236] Overall Loss 0.135534 Objective Loss 0.135534 LR 0.000250 Time 0.023050 +2023-10-02 21:44:46,706 - Epoch: [157][ 470/ 1236] Overall Loss 0.135506 Objective Loss 0.135506 LR 0.000250 Time 0.023011 +2023-10-02 21:44:46,918 - Epoch: [157][ 480/ 1236] Overall Loss 0.135578 Objective Loss 0.135578 LR 0.000250 Time 0.022973 +2023-10-02 21:44:47,131 - Epoch: [157][ 490/ 1236] Overall Loss 0.135523 Objective Loss 0.135523 LR 0.000250 Time 0.022938 +2023-10-02 21:44:47,343 - Epoch: [157][ 500/ 1236] Overall Loss 0.135676 Objective Loss 0.135676 LR 0.000250 Time 0.022902 +2023-10-02 21:44:47,555 - Epoch: [157][ 510/ 1236] Overall Loss 0.135620 Objective Loss 0.135620 LR 0.000250 Time 0.022870 +2023-10-02 21:44:47,767 - Epoch: [157][ 520/ 1236] Overall Loss 0.135784 Objective Loss 0.135784 LR 0.000250 Time 0.022837 +2023-10-02 21:44:47,980 - Epoch: [157][ 530/ 1236] Overall Loss 0.135475 Objective Loss 0.135475 LR 0.000250 Time 0.022808 +2023-10-02 21:44:48,192 - Epoch: [157][ 540/ 1236] Overall Loss 0.135180 Objective Loss 0.135180 LR 0.000250 Time 0.022777 +2023-10-02 21:44:48,405 - Epoch: [157][ 550/ 1236] Overall Loss 0.135158 Objective Loss 0.135158 LR 0.000250 Time 0.022749 +2023-10-02 21:44:48,618 - Epoch: [157][ 560/ 1236] Overall Loss 0.135052 Objective Loss 0.135052 LR 0.000250 Time 0.022722 +2023-10-02 21:44:48,831 - Epoch: [157][ 570/ 1236] Overall Loss 0.134927 Objective Loss 0.134927 LR 0.000250 Time 0.022697 +2023-10-02 21:44:49,043 - Epoch: [157][ 580/ 1236] Overall Loss 0.135205 Objective Loss 0.135205 LR 0.000250 Time 0.022671 +2023-10-02 21:44:49,256 - Epoch: [157][ 590/ 1236] Overall Loss 0.134796 Objective Loss 0.134796 LR 0.000250 Time 0.022647 +2023-10-02 21:44:49,468 - Epoch: [157][ 600/ 1236] Overall Loss 0.135085 Objective Loss 0.135085 LR 0.000250 Time 0.022621 +2023-10-02 21:44:49,680 - Epoch: [157][ 610/ 1236] Overall Loss 0.135430 Objective Loss 0.135430 LR 0.000250 Time 0.022598 +2023-10-02 21:44:49,892 - Epoch: [157][ 620/ 1236] Overall Loss 0.135499 Objective Loss 0.135499 LR 0.000250 Time 0.022576 +2023-10-02 21:44:50,105 - Epoch: [157][ 630/ 1236] Overall Loss 0.135554 Objective Loss 0.135554 LR 0.000250 Time 0.022555 +2023-10-02 21:44:50,317 - Epoch: [157][ 640/ 1236] Overall Loss 0.135766 Objective Loss 0.135766 LR 0.000250 Time 0.022533 +2023-10-02 21:44:50,530 - Epoch: [157][ 650/ 1236] Overall Loss 0.135949 Objective Loss 0.135949 LR 0.000250 Time 0.022513 +2023-10-02 21:44:50,742 - Epoch: [157][ 660/ 1236] Overall Loss 0.135710 Objective Loss 0.135710 LR 0.000250 Time 0.022492 +2023-10-02 21:44:50,954 - Epoch: [157][ 670/ 1236] Overall Loss 0.135936 Objective Loss 0.135936 LR 0.000250 Time 0.022474 +2023-10-02 21:44:51,166 - Epoch: [157][ 680/ 1236] Overall Loss 0.135921 Objective Loss 0.135921 LR 0.000250 Time 0.022455 +2023-10-02 21:44:51,379 - Epoch: [157][ 690/ 1236] Overall Loss 0.136040 Objective Loss 0.136040 LR 0.000250 Time 0.022437 +2023-10-02 21:44:51,591 - Epoch: [157][ 700/ 1236] Overall Loss 0.136094 Objective Loss 0.136094 LR 0.000250 Time 0.022419 +2023-10-02 21:44:51,804 - Epoch: [157][ 710/ 1236] Overall Loss 0.136348 Objective Loss 0.136348 LR 0.000250 Time 0.022401 +2023-10-02 21:44:52,016 - Epoch: [157][ 720/ 1236] Overall Loss 0.136183 Objective Loss 0.136183 LR 0.000250 Time 0.022384 +2023-10-02 21:44:52,229 - Epoch: [157][ 730/ 1236] Overall Loss 0.136352 Objective Loss 0.136352 LR 0.000250 Time 0.022369 +2023-10-02 21:44:52,441 - Epoch: [157][ 740/ 1236] Overall Loss 0.136019 Objective Loss 0.136019 LR 0.000250 Time 0.022352 +2023-10-02 21:44:52,653 - Epoch: [157][ 750/ 1236] Overall Loss 0.135847 Objective Loss 0.135847 LR 0.000250 Time 0.022335 +2023-10-02 21:44:52,865 - Epoch: [157][ 760/ 1236] Overall Loss 0.135961 Objective Loss 0.135961 LR 0.000250 Time 0.022320 +2023-10-02 21:44:53,078 - Epoch: [157][ 770/ 1236] Overall Loss 0.135966 Objective Loss 0.135966 LR 0.000250 Time 0.022304 +2023-10-02 21:44:53,290 - Epoch: [157][ 780/ 1236] Overall Loss 0.135957 Objective Loss 0.135957 LR 0.000250 Time 0.022290 +2023-10-02 21:44:53,503 - Epoch: [157][ 790/ 1236] Overall Loss 0.135964 Objective Loss 0.135964 LR 0.000250 Time 0.022277 +2023-10-02 21:44:53,716 - Epoch: [157][ 800/ 1236] Overall Loss 0.135985 Objective Loss 0.135985 LR 0.000250 Time 0.022264 +2023-10-02 21:44:53,928 - Epoch: [157][ 810/ 1236] Overall Loss 0.136032 Objective Loss 0.136032 LR 0.000250 Time 0.022251 +2023-10-02 21:44:54,140 - Epoch: [157][ 820/ 1236] Overall Loss 0.136337 Objective Loss 0.136337 LR 0.000250 Time 0.022238 +2023-10-02 21:44:54,352 - Epoch: [157][ 830/ 1236] Overall Loss 0.136254 Objective Loss 0.136254 LR 0.000250 Time 0.022225 +2023-10-02 21:44:54,565 - Epoch: [157][ 840/ 1236] Overall Loss 0.136116 Objective Loss 0.136116 LR 0.000250 Time 0.022213 +2023-10-02 21:44:54,777 - Epoch: [157][ 850/ 1236] Overall Loss 0.136143 Objective Loss 0.136143 LR 0.000250 Time 0.022200 +2023-10-02 21:44:54,989 - Epoch: [157][ 860/ 1236] Overall Loss 0.136321 Objective Loss 0.136321 LR 0.000250 Time 0.022188 +2023-10-02 21:44:55,202 - Epoch: [157][ 870/ 1236] Overall Loss 0.136417 Objective Loss 0.136417 LR 0.000250 Time 0.022177 +2023-10-02 21:44:55,414 - Epoch: [157][ 880/ 1236] Overall Loss 0.136400 Objective Loss 0.136400 LR 0.000250 Time 0.022166 +2023-10-02 21:44:55,627 - Epoch: [157][ 890/ 1236] Overall Loss 0.136333 Objective Loss 0.136333 LR 0.000250 Time 0.022156 +2023-10-02 21:44:55,839 - Epoch: [157][ 900/ 1236] Overall Loss 0.136334 Objective Loss 0.136334 LR 0.000250 Time 0.022145 +2023-10-02 21:44:56,053 - Epoch: [157][ 910/ 1236] Overall Loss 0.136162 Objective Loss 0.136162 LR 0.000250 Time 0.022135 +2023-10-02 21:44:56,265 - Epoch: [157][ 920/ 1236] Overall Loss 0.136116 Objective Loss 0.136116 LR 0.000250 Time 0.022125 +2023-10-02 21:44:56,478 - Epoch: [157][ 930/ 1236] Overall Loss 0.136234 Objective Loss 0.136234 LR 0.000250 Time 0.022116 +2023-10-02 21:44:56,690 - Epoch: [157][ 940/ 1236] Overall Loss 0.136306 Objective Loss 0.136306 LR 0.000250 Time 0.022105 +2023-10-02 21:44:56,903 - Epoch: [157][ 950/ 1236] Overall Loss 0.136375 Objective Loss 0.136375 LR 0.000250 Time 0.022097 +2023-10-02 21:44:57,115 - Epoch: [157][ 960/ 1236] Overall Loss 0.136412 Objective Loss 0.136412 LR 0.000250 Time 0.022088 +2023-10-02 21:44:57,328 - Epoch: [157][ 970/ 1236] Overall Loss 0.136403 Objective Loss 0.136403 LR 0.000250 Time 0.022079 +2023-10-02 21:44:57,540 - Epoch: [157][ 980/ 1236] Overall Loss 0.136437 Objective Loss 0.136437 LR 0.000250 Time 0.022070 +2023-10-02 21:44:57,754 - Epoch: [157][ 990/ 1236] Overall Loss 0.136495 Objective Loss 0.136495 LR 0.000250 Time 0.022061 +2023-10-02 21:44:57,966 - Epoch: [157][ 1000/ 1236] Overall Loss 0.136497 Objective Loss 0.136497 LR 0.000250 Time 0.022052 +2023-10-02 21:44:58,178 - Epoch: [157][ 1010/ 1236] Overall Loss 0.136609 Objective Loss 0.136609 LR 0.000250 Time 0.022044 +2023-10-02 21:44:58,391 - Epoch: [157][ 1020/ 1236] Overall Loss 0.136611 Objective Loss 0.136611 LR 0.000250 Time 0.022035 +2023-10-02 21:44:58,604 - Epoch: [157][ 1030/ 1236] Overall Loss 0.136629 Objective Loss 0.136629 LR 0.000250 Time 0.022028 +2023-10-02 21:44:58,814 - Epoch: [157][ 1040/ 1236] Overall Loss 0.136579 Objective Loss 0.136579 LR 0.000250 Time 0.022018 +2023-10-02 21:44:59,023 - Epoch: [157][ 1050/ 1236] Overall Loss 0.136499 Objective Loss 0.136499 LR 0.000250 Time 0.022008 +2023-10-02 21:44:59,234 - Epoch: [157][ 1060/ 1236] Overall Loss 0.136491 Objective Loss 0.136491 LR 0.000250 Time 0.021999 +2023-10-02 21:44:59,444 - Epoch: [157][ 1070/ 1236] Overall Loss 0.136531 Objective Loss 0.136531 LR 0.000250 Time 0.021988 +2023-10-02 21:44:59,655 - Epoch: [157][ 1080/ 1236] Overall Loss 0.136504 Objective Loss 0.136504 LR 0.000250 Time 0.021979 +2023-10-02 21:44:59,864 - Epoch: [157][ 1090/ 1236] Overall Loss 0.136460 Objective Loss 0.136460 LR 0.000250 Time 0.021969 +2023-10-02 21:45:00,075 - Epoch: [157][ 1100/ 1236] Overall Loss 0.136474 Objective Loss 0.136474 LR 0.000250 Time 0.021961 +2023-10-02 21:45:00,285 - Epoch: [157][ 1110/ 1236] Overall Loss 0.136524 Objective Loss 0.136524 LR 0.000250 Time 0.021952 +2023-10-02 21:45:00,496 - Epoch: [157][ 1120/ 1236] Overall Loss 0.136806 Objective Loss 0.136806 LR 0.000250 Time 0.021944 +2023-10-02 21:45:00,706 - Epoch: [157][ 1130/ 1236] Overall Loss 0.136933 Objective Loss 0.136933 LR 0.000250 Time 0.021935 +2023-10-02 21:45:00,917 - Epoch: [157][ 1140/ 1236] Overall Loss 0.137019 Objective Loss 0.137019 LR 0.000250 Time 0.021928 +2023-10-02 21:45:01,127 - Epoch: [157][ 1150/ 1236] Overall Loss 0.136963 Objective Loss 0.136963 LR 0.000250 Time 0.021919 +2023-10-02 21:45:01,338 - Epoch: [157][ 1160/ 1236] Overall Loss 0.137163 Objective Loss 0.137163 LR 0.000250 Time 0.021912 +2023-10-02 21:45:01,548 - Epoch: [157][ 1170/ 1236] Overall Loss 0.137101 Objective Loss 0.137101 LR 0.000250 Time 0.021904 +2023-10-02 21:45:01,759 - Epoch: [157][ 1180/ 1236] Overall Loss 0.137045 Objective Loss 0.137045 LR 0.000250 Time 0.021897 +2023-10-02 21:45:01,968 - Epoch: [157][ 1190/ 1236] Overall Loss 0.137049 Objective Loss 0.137049 LR 0.000250 Time 0.021888 +2023-10-02 21:45:02,193 - Epoch: [157][ 1200/ 1236] Overall Loss 0.137066 Objective Loss 0.137066 LR 0.000250 Time 0.021893 +2023-10-02 21:45:02,418 - Epoch: [157][ 1210/ 1236] Overall Loss 0.137174 Objective Loss 0.137174 LR 0.000250 Time 0.021898 +2023-10-02 21:45:02,643 - Epoch: [157][ 1220/ 1236] Overall Loss 0.137335 Objective Loss 0.137335 LR 0.000250 Time 0.021903 +2023-10-02 21:45:02,912 - Epoch: [157][ 1230/ 1236] Overall Loss 0.137326 Objective Loss 0.137326 LR 0.000250 Time 0.021943 +2023-10-02 21:45:03,035 - Epoch: [157][ 1236/ 1236] Overall Loss 0.137342 Objective Loss 0.137342 Top1 90.835031 Top5 98.574338 LR 0.000250 Time 0.021936 +2023-10-02 21:45:03,189 - --- validate (epoch=157)----------- +2023-10-02 21:45:03,189 - 29943 samples (256 per mini-batch) +2023-10-02 21:45:03,688 - Epoch: [157][ 10/ 117] Loss 0.283298 Top1 86.953125 Top5 98.242188 +2023-10-02 21:45:03,842 - Epoch: [157][ 20/ 117] Loss 0.297830 Top1 86.933594 Top5 98.496094 +2023-10-02 21:45:03,994 - Epoch: [157][ 30/ 117] Loss 0.304124 Top1 87.018229 Top5 98.502604 +2023-10-02 21:45:04,146 - Epoch: [157][ 40/ 117] Loss 0.303291 Top1 87.148438 Top5 98.593750 +2023-10-02 21:45:04,298 - Epoch: [157][ 50/ 117] Loss 0.300076 Top1 87.085938 Top5 98.585938 +2023-10-02 21:45:04,450 - Epoch: [157][ 60/ 117] Loss 0.300845 Top1 87.076823 Top5 98.626302 +2023-10-02 21:45:04,602 - Epoch: [157][ 70/ 117] Loss 0.301287 Top1 87.260045 Top5 98.616071 +2023-10-02 21:45:04,754 - Epoch: [157][ 80/ 117] Loss 0.297043 Top1 87.353516 Top5 98.662109 +2023-10-02 21:45:04,906 - Epoch: [157][ 90/ 117] Loss 0.295260 Top1 87.348090 Top5 98.667535 +2023-10-02 21:45:05,059 - Epoch: [157][ 100/ 117] Loss 0.297151 Top1 87.269531 Top5 98.679688 +2023-10-02 21:45:05,219 - Epoch: [157][ 110/ 117] Loss 0.299214 Top1 87.159091 Top5 98.664773 +2023-10-02 21:45:05,310 - Epoch: [157][ 117/ 117] Loss 0.300365 Top1 87.138897 Top5 98.650770 +2023-10-02 21:45:05,409 - ==> Top1: 87.139 Top5: 98.651 Loss: 0.300 + +2023-10-02 21:45:05,410 - ==> Confusion: +[[ 932 0 4 1 6 2 0 0 7 65 2 1 1 0 3 0 3 0 1 0 22] + [ 0 1065 1 0 4 21 0 18 0 1 0 1 1 0 0 3 0 0 8 1 7] + [ 2 1 982 5 0 0 19 8 1 1 1 1 6 2 1 3 2 2 7 4 8] + [ 1 3 15 979 1 2 3 2 4 1 5 0 5 2 26 3 1 4 10 1 21] + [ 15 2 0 1 973 5 0 0 1 19 0 0 2 2 10 4 12 0 0 1 3] + [ 3 30 0 1 5 1003 2 16 3 5 1 5 3 6 6 0 1 0 6 1 19] + [ 0 3 22 0 0 1 1139 2 0 0 4 0 0 0 0 6 0 0 2 8 4] + [ 0 9 17 0 3 26 6 1074 2 3 2 6 3 4 3 0 0 1 43 7 9] + [ 17 1 0 1 1 2 0 1 976 41 10 0 2 12 12 1 4 1 3 1 3] + [ 81 0 1 0 6 4 0 0 22 972 1 1 0 15 6 2 1 0 0 2 5] + [ 2 1 10 6 0 1 4 1 10 1 976 2 0 9 3 0 2 3 9 2 11] + [ 0 2 2 0 0 13 0 3 0 0 0 959 27 6 0 1 0 13 1 3 5] + [ 0 2 1 2 1 1 1 0 0 1 3 29 975 2 0 12 2 12 1 6 17] + [ 0 0 1 0 2 6 0 1 13 12 3 9 0 1054 3 0 0 0 0 1 14] + [ 14 1 5 16 4 1 0 0 21 2 1 0 3 5 1006 0 1 3 9 0 9] + [ 0 1 2 1 5 0 0 0 0 0 1 5 8 0 0 1073 13 9 2 8 6] + [ 0 19 1 1 4 6 1 0 1 0 0 3 0 1 2 13 1092 0 1 4 12] + [ 1 0 0 1 1 0 2 0 0 0 0 6 21 2 2 10 0 987 0 0 5] + [ 1 4 3 15 1 1 0 19 4 2 3 1 2 0 13 0 0 0 989 0 10] + [ 0 0 5 1 3 1 9 4 0 0 1 10 6 2 1 0 8 0 1 1093 7] + [ 102 136 106 73 57 102 30 78 81 76 147 71 302 247 110 52 64 40 107 131 5793]] + +2023-10-02 21:45:05,411 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:45:05,411 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:45:05,417 - + +2023-10-02 21:45:05,418 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:45:06,468 - Epoch: [158][ 10/ 1236] Overall Loss 0.108198 Objective Loss 0.108198 LR 0.000250 Time 0.105017 +2023-10-02 21:45:06,679 - Epoch: [158][ 20/ 1236] Overall Loss 0.117951 Objective Loss 0.117951 LR 0.000250 Time 0.063020 +2023-10-02 21:45:06,889 - Epoch: [158][ 30/ 1236] Overall Loss 0.126816 Objective Loss 0.126816 LR 0.000250 Time 0.049008 +2023-10-02 21:45:07,099 - Epoch: [158][ 40/ 1236] Overall Loss 0.125242 Objective Loss 0.125242 LR 0.000250 Time 0.042001 +2023-10-02 21:45:07,310 - Epoch: [158][ 50/ 1236] Overall Loss 0.125928 Objective Loss 0.125928 LR 0.000250 Time 0.037782 +2023-10-02 21:45:07,520 - Epoch: [158][ 60/ 1236] Overall Loss 0.125689 Objective Loss 0.125689 LR 0.000250 Time 0.034985 +2023-10-02 21:45:07,730 - Epoch: [158][ 70/ 1236] Overall Loss 0.124178 Objective Loss 0.124178 LR 0.000250 Time 0.032984 +2023-10-02 21:45:07,940 - Epoch: [158][ 80/ 1236] Overall Loss 0.124246 Objective Loss 0.124246 LR 0.000250 Time 0.031482 +2023-10-02 21:45:08,150 - Epoch: [158][ 90/ 1236] Overall Loss 0.125003 Objective Loss 0.125003 LR 0.000250 Time 0.030313 +2023-10-02 21:45:08,360 - Epoch: [158][ 100/ 1236] Overall Loss 0.125349 Objective Loss 0.125349 LR 0.000250 Time 0.029381 +2023-10-02 21:45:08,570 - Epoch: [158][ 110/ 1236] Overall Loss 0.125518 Objective Loss 0.125518 LR 0.000250 Time 0.028614 +2023-10-02 21:45:08,780 - Epoch: [158][ 120/ 1236] Overall Loss 0.127507 Objective Loss 0.127507 LR 0.000250 Time 0.027979 +2023-10-02 21:45:08,990 - Epoch: [158][ 130/ 1236] Overall Loss 0.128059 Objective Loss 0.128059 LR 0.000250 Time 0.027436 +2023-10-02 21:45:09,200 - Epoch: [158][ 140/ 1236] Overall Loss 0.128843 Objective Loss 0.128843 LR 0.000250 Time 0.026980 +2023-10-02 21:45:09,409 - Epoch: [158][ 150/ 1236] Overall Loss 0.129672 Objective Loss 0.129672 LR 0.000250 Time 0.026571 +2023-10-02 21:45:09,619 - Epoch: [158][ 160/ 1236] Overall Loss 0.129163 Objective Loss 0.129163 LR 0.000250 Time 0.026222 +2023-10-02 21:45:09,828 - Epoch: [158][ 170/ 1236] Overall Loss 0.129117 Objective Loss 0.129117 LR 0.000250 Time 0.025898 +2023-10-02 21:45:10,038 - Epoch: [158][ 180/ 1236] Overall Loss 0.129207 Objective Loss 0.129207 LR 0.000250 Time 0.025626 +2023-10-02 21:45:10,247 - Epoch: [158][ 190/ 1236] Overall Loss 0.129271 Objective Loss 0.129271 LR 0.000250 Time 0.025367 +2023-10-02 21:45:10,459 - Epoch: [158][ 200/ 1236] Overall Loss 0.129073 Objective Loss 0.129073 LR 0.000250 Time 0.025157 +2023-10-02 21:45:10,669 - Epoch: [158][ 210/ 1236] Overall Loss 0.129269 Objective Loss 0.129269 LR 0.000250 Time 0.024956 +2023-10-02 21:45:10,879 - Epoch: [158][ 220/ 1236] Overall Loss 0.129312 Objective Loss 0.129312 LR 0.000250 Time 0.024777 +2023-10-02 21:45:11,088 - Epoch: [158][ 230/ 1236] Overall Loss 0.129194 Objective Loss 0.129194 LR 0.000250 Time 0.024602 +2023-10-02 21:45:11,299 - Epoch: [158][ 240/ 1236] Overall Loss 0.129548 Objective Loss 0.129548 LR 0.000250 Time 0.024452 +2023-10-02 21:45:11,508 - Epoch: [158][ 250/ 1236] Overall Loss 0.129905 Objective Loss 0.129905 LR 0.000250 Time 0.024303 +2023-10-02 21:45:11,718 - Epoch: [158][ 260/ 1236] Overall Loss 0.130328 Objective Loss 0.130328 LR 0.000250 Time 0.024177 +2023-10-02 21:45:11,927 - Epoch: [158][ 270/ 1236] Overall Loss 0.130518 Objective Loss 0.130518 LR 0.000250 Time 0.024051 +2023-10-02 21:45:12,138 - Epoch: [158][ 280/ 1236] Overall Loss 0.130928 Objective Loss 0.130928 LR 0.000250 Time 0.023942 +2023-10-02 21:45:12,346 - Epoch: [158][ 290/ 1236] Overall Loss 0.131174 Objective Loss 0.131174 LR 0.000250 Time 0.023831 +2023-10-02 21:45:12,557 - Epoch: [158][ 300/ 1236] Overall Loss 0.131728 Objective Loss 0.131728 LR 0.000250 Time 0.023737 +2023-10-02 21:45:12,765 - Epoch: [158][ 310/ 1236] Overall Loss 0.131601 Objective Loss 0.131601 LR 0.000250 Time 0.023639 +2023-10-02 21:45:12,976 - Epoch: [158][ 320/ 1236] Overall Loss 0.131478 Objective Loss 0.131478 LR 0.000250 Time 0.023557 +2023-10-02 21:45:13,185 - Epoch: [158][ 330/ 1236] Overall Loss 0.131238 Objective Loss 0.131238 LR 0.000250 Time 0.023472 +2023-10-02 21:45:13,396 - Epoch: [158][ 340/ 1236] Overall Loss 0.131984 Objective Loss 0.131984 LR 0.000250 Time 0.023404 +2023-10-02 21:45:13,604 - Epoch: [158][ 350/ 1236] Overall Loss 0.131819 Objective Loss 0.131819 LR 0.000250 Time 0.023328 +2023-10-02 21:45:13,817 - Epoch: [158][ 360/ 1236] Overall Loss 0.132726 Objective Loss 0.132726 LR 0.000250 Time 0.023271 +2023-10-02 21:45:14,025 - Epoch: [158][ 370/ 1236] Overall Loss 0.132926 Objective Loss 0.132926 LR 0.000250 Time 0.023204 +2023-10-02 21:45:14,236 - Epoch: [158][ 380/ 1236] Overall Loss 0.133539 Objective Loss 0.133539 LR 0.000250 Time 0.023147 +2023-10-02 21:45:14,445 - Epoch: [158][ 390/ 1236] Overall Loss 0.133489 Objective Loss 0.133489 LR 0.000250 Time 0.023089 +2023-10-02 21:45:14,657 - Epoch: [158][ 400/ 1236] Overall Loss 0.132707 Objective Loss 0.132707 LR 0.000250 Time 0.023041 +2023-10-02 21:45:14,866 - Epoch: [158][ 410/ 1236] Overall Loss 0.132549 Objective Loss 0.132549 LR 0.000250 Time 0.022986 +2023-10-02 21:45:15,077 - Epoch: [158][ 420/ 1236] Overall Loss 0.133264 Objective Loss 0.133264 LR 0.000250 Time 0.022941 +2023-10-02 21:45:15,287 - Epoch: [158][ 430/ 1236] Overall Loss 0.132894 Objective Loss 0.132894 LR 0.000250 Time 0.022896 +2023-10-02 21:45:15,497 - Epoch: [158][ 440/ 1236] Overall Loss 0.132964 Objective Loss 0.132964 LR 0.000250 Time 0.022852 +2023-10-02 21:45:15,707 - Epoch: [158][ 450/ 1236] Overall Loss 0.133094 Objective Loss 0.133094 LR 0.000250 Time 0.022810 +2023-10-02 21:45:15,917 - Epoch: [158][ 460/ 1236] Overall Loss 0.133416 Objective Loss 0.133416 LR 0.000250 Time 0.022771 +2023-10-02 21:45:16,128 - Epoch: [158][ 470/ 1236] Overall Loss 0.133143 Objective Loss 0.133143 LR 0.000250 Time 0.022733 +2023-10-02 21:45:16,338 - Epoch: [158][ 480/ 1236] Overall Loss 0.132838 Objective Loss 0.132838 LR 0.000250 Time 0.022698 +2023-10-02 21:45:16,548 - Epoch: [158][ 490/ 1236] Overall Loss 0.132874 Objective Loss 0.132874 LR 0.000250 Time 0.022661 +2023-10-02 21:45:16,758 - Epoch: [158][ 500/ 1236] Overall Loss 0.132786 Objective Loss 0.132786 LR 0.000250 Time 0.022629 +2023-10-02 21:45:16,968 - Epoch: [158][ 510/ 1236] Overall Loss 0.132724 Objective Loss 0.132724 LR 0.000250 Time 0.022596 +2023-10-02 21:45:17,179 - Epoch: [158][ 520/ 1236] Overall Loss 0.132702 Objective Loss 0.132702 LR 0.000250 Time 0.022567 +2023-10-02 21:45:17,389 - Epoch: [158][ 530/ 1236] Overall Loss 0.132903 Objective Loss 0.132903 LR 0.000250 Time 0.022536 +2023-10-02 21:45:17,599 - Epoch: [158][ 540/ 1236] Overall Loss 0.132926 Objective Loss 0.132926 LR 0.000250 Time 0.022508 +2023-10-02 21:45:17,809 - Epoch: [158][ 550/ 1236] Overall Loss 0.132819 Objective Loss 0.132819 LR 0.000250 Time 0.022479 +2023-10-02 21:45:18,019 - Epoch: [158][ 560/ 1236] Overall Loss 0.132630 Objective Loss 0.132630 LR 0.000250 Time 0.022453 +2023-10-02 21:45:18,229 - Epoch: [158][ 570/ 1236] Overall Loss 0.132380 Objective Loss 0.132380 LR 0.000250 Time 0.022427 +2023-10-02 21:45:18,440 - Epoch: [158][ 580/ 1236] Overall Loss 0.132634 Objective Loss 0.132634 LR 0.000250 Time 0.022404 +2023-10-02 21:45:18,649 - Epoch: [158][ 590/ 1236] Overall Loss 0.132548 Objective Loss 0.132548 LR 0.000250 Time 0.022378 +2023-10-02 21:45:18,860 - Epoch: [158][ 600/ 1236] Overall Loss 0.132967 Objective Loss 0.132967 LR 0.000250 Time 0.022356 +2023-10-02 21:45:19,070 - Epoch: [158][ 610/ 1236] Overall Loss 0.133146 Objective Loss 0.133146 LR 0.000250 Time 0.022333 +2023-10-02 21:45:19,281 - Epoch: [158][ 620/ 1236] Overall Loss 0.133214 Objective Loss 0.133214 LR 0.000250 Time 0.022313 +2023-10-02 21:45:19,491 - Epoch: [158][ 630/ 1236] Overall Loss 0.133042 Objective Loss 0.133042 LR 0.000250 Time 0.022291 +2023-10-02 21:45:19,702 - Epoch: [158][ 640/ 1236] Overall Loss 0.133162 Objective Loss 0.133162 LR 0.000250 Time 0.022272 +2023-10-02 21:45:19,912 - Epoch: [158][ 650/ 1236] Overall Loss 0.133076 Objective Loss 0.133076 LR 0.000250 Time 0.022252 +2023-10-02 21:45:20,123 - Epoch: [158][ 660/ 1236] Overall Loss 0.133231 Objective Loss 0.133231 LR 0.000250 Time 0.022235 +2023-10-02 21:45:20,334 - Epoch: [158][ 670/ 1236] Overall Loss 0.133229 Objective Loss 0.133229 LR 0.000250 Time 0.022216 +2023-10-02 21:45:20,544 - Epoch: [158][ 680/ 1236] Overall Loss 0.133145 Objective Loss 0.133145 LR 0.000250 Time 0.022199 +2023-10-02 21:45:20,754 - Epoch: [158][ 690/ 1236] Overall Loss 0.133361 Objective Loss 0.133361 LR 0.000250 Time 0.022181 +2023-10-02 21:45:20,966 - Epoch: [158][ 700/ 1236] Overall Loss 0.133394 Objective Loss 0.133394 LR 0.000250 Time 0.022166 +2023-10-02 21:45:21,176 - Epoch: [158][ 710/ 1236] Overall Loss 0.133321 Objective Loss 0.133321 LR 0.000250 Time 0.022149 +2023-10-02 21:45:21,387 - Epoch: [158][ 720/ 1236] Overall Loss 0.133270 Objective Loss 0.133270 LR 0.000250 Time 0.022134 +2023-10-02 21:45:21,597 - Epoch: [158][ 730/ 1236] Overall Loss 0.133333 Objective Loss 0.133333 LR 0.000250 Time 0.022118 +2023-10-02 21:45:21,809 - Epoch: [158][ 740/ 1236] Overall Loss 0.133430 Objective Loss 0.133430 LR 0.000250 Time 0.022106 +2023-10-02 21:45:22,019 - Epoch: [158][ 750/ 1236] Overall Loss 0.133534 Objective Loss 0.133534 LR 0.000250 Time 0.022090 +2023-10-02 21:45:22,231 - Epoch: [158][ 760/ 1236] Overall Loss 0.133698 Objective Loss 0.133698 LR 0.000250 Time 0.022078 +2023-10-02 21:45:22,441 - Epoch: [158][ 770/ 1236] Overall Loss 0.133615 Objective Loss 0.133615 LR 0.000250 Time 0.022064 +2023-10-02 21:45:22,652 - Epoch: [158][ 780/ 1236] Overall Loss 0.133361 Objective Loss 0.133361 LR 0.000250 Time 0.022051 +2023-10-02 21:45:22,862 - Epoch: [158][ 790/ 1236] Overall Loss 0.133669 Objective Loss 0.133669 LR 0.000250 Time 0.022037 +2023-10-02 21:45:23,074 - Epoch: [158][ 800/ 1236] Overall Loss 0.133867 Objective Loss 0.133867 LR 0.000250 Time 0.022027 +2023-10-02 21:45:23,283 - Epoch: [158][ 810/ 1236] Overall Loss 0.133766 Objective Loss 0.133766 LR 0.000250 Time 0.022013 +2023-10-02 21:45:23,497 - Epoch: [158][ 820/ 1236] Overall Loss 0.133624 Objective Loss 0.133624 LR 0.000250 Time 0.022005 +2023-10-02 21:45:23,707 - Epoch: [158][ 830/ 1236] Overall Loss 0.133916 Objective Loss 0.133916 LR 0.000250 Time 0.021992 +2023-10-02 21:45:23,919 - Epoch: [158][ 840/ 1236] Overall Loss 0.134067 Objective Loss 0.134067 LR 0.000250 Time 0.021982 +2023-10-02 21:45:24,129 - Epoch: [158][ 850/ 1236] Overall Loss 0.134155 Objective Loss 0.134155 LR 0.000250 Time 0.021970 +2023-10-02 21:45:24,340 - Epoch: [158][ 860/ 1236] Overall Loss 0.134216 Objective Loss 0.134216 LR 0.000250 Time 0.021960 +2023-10-02 21:45:24,550 - Epoch: [158][ 870/ 1236] Overall Loss 0.134587 Objective Loss 0.134587 LR 0.000250 Time 0.021948 +2023-10-02 21:45:24,761 - Epoch: [158][ 880/ 1236] Overall Loss 0.134409 Objective Loss 0.134409 LR 0.000250 Time 0.021938 +2023-10-02 21:45:24,971 - Epoch: [158][ 890/ 1236] Overall Loss 0.134386 Objective Loss 0.134386 LR 0.000250 Time 0.021927 +2023-10-02 21:45:25,186 - Epoch: [158][ 900/ 1236] Overall Loss 0.134449 Objective Loss 0.134449 LR 0.000250 Time 0.021923 +2023-10-02 21:45:25,396 - Epoch: [158][ 910/ 1236] Overall Loss 0.134405 Objective Loss 0.134405 LR 0.000250 Time 0.021912 +2023-10-02 21:45:25,607 - Epoch: [158][ 920/ 1236] Overall Loss 0.134484 Objective Loss 0.134484 LR 0.000250 Time 0.021903 +2023-10-02 21:45:25,817 - Epoch: [158][ 930/ 1236] Overall Loss 0.134364 Objective Loss 0.134364 LR 0.000250 Time 0.021891 +2023-10-02 21:45:26,028 - Epoch: [158][ 940/ 1236] Overall Loss 0.134208 Objective Loss 0.134208 LR 0.000250 Time 0.021883 +2023-10-02 21:45:26,238 - Epoch: [158][ 950/ 1236] Overall Loss 0.134261 Objective Loss 0.134261 LR 0.000250 Time 0.021872 +2023-10-02 21:45:26,449 - Epoch: [158][ 960/ 1236] Overall Loss 0.134313 Objective Loss 0.134313 LR 0.000250 Time 0.021864 +2023-10-02 21:45:26,659 - Epoch: [158][ 970/ 1236] Overall Loss 0.134295 Objective Loss 0.134295 LR 0.000250 Time 0.021854 +2023-10-02 21:45:26,869 - Epoch: [158][ 980/ 1236] Overall Loss 0.134300 Objective Loss 0.134300 LR 0.000250 Time 0.021846 +2023-10-02 21:45:27,080 - Epoch: [158][ 990/ 1236] Overall Loss 0.134431 Objective Loss 0.134431 LR 0.000250 Time 0.021837 +2023-10-02 21:45:27,292 - Epoch: [158][ 1000/ 1236] Overall Loss 0.134363 Objective Loss 0.134363 LR 0.000250 Time 0.021830 +2023-10-02 21:45:27,502 - Epoch: [158][ 1010/ 1236] Overall Loss 0.134372 Objective Loss 0.134372 LR 0.000250 Time 0.021822 +2023-10-02 21:45:27,716 - Epoch: [158][ 1020/ 1236] Overall Loss 0.134360 Objective Loss 0.134360 LR 0.000250 Time 0.021817 +2023-10-02 21:45:27,926 - Epoch: [158][ 1030/ 1236] Overall Loss 0.134602 Objective Loss 0.134602 LR 0.000250 Time 0.021809 +2023-10-02 21:45:28,138 - Epoch: [158][ 1040/ 1236] Overall Loss 0.134751 Objective Loss 0.134751 LR 0.000250 Time 0.021802 +2023-10-02 21:45:28,348 - Epoch: [158][ 1050/ 1236] Overall Loss 0.134878 Objective Loss 0.134878 LR 0.000250 Time 0.021794 +2023-10-02 21:45:28,560 - Epoch: [158][ 1060/ 1236] Overall Loss 0.134831 Objective Loss 0.134831 LR 0.000250 Time 0.021789 +2023-10-02 21:45:28,769 - Epoch: [158][ 1070/ 1236] Overall Loss 0.134944 Objective Loss 0.134944 LR 0.000250 Time 0.021780 +2023-10-02 21:45:28,984 - Epoch: [158][ 1080/ 1236] Overall Loss 0.135067 Objective Loss 0.135067 LR 0.000250 Time 0.021777 +2023-10-02 21:45:29,193 - Epoch: [158][ 1090/ 1236] Overall Loss 0.135030 Objective Loss 0.135030 LR 0.000250 Time 0.021769 +2023-10-02 21:45:29,405 - Epoch: [158][ 1100/ 1236] Overall Loss 0.135010 Objective Loss 0.135010 LR 0.000250 Time 0.021763 +2023-10-02 21:45:29,613 - Epoch: [158][ 1110/ 1236] Overall Loss 0.135200 Objective Loss 0.135200 LR 0.000250 Time 0.021755 +2023-10-02 21:45:29,826 - Epoch: [158][ 1120/ 1236] Overall Loss 0.135065 Objective Loss 0.135065 LR 0.000250 Time 0.021750 +2023-10-02 21:45:30,033 - Epoch: [158][ 1130/ 1236] Overall Loss 0.135060 Objective Loss 0.135060 LR 0.000250 Time 0.021741 +2023-10-02 21:45:30,246 - Epoch: [158][ 1140/ 1236] Overall Loss 0.135131 Objective Loss 0.135131 LR 0.000250 Time 0.021736 +2023-10-02 21:45:30,453 - Epoch: [158][ 1150/ 1236] Overall Loss 0.135177 Objective Loss 0.135177 LR 0.000250 Time 0.021728 +2023-10-02 21:45:30,664 - Epoch: [158][ 1160/ 1236] Overall Loss 0.135045 Objective Loss 0.135045 LR 0.000250 Time 0.021722 +2023-10-02 21:45:30,873 - Epoch: [158][ 1170/ 1236] Overall Loss 0.135066 Objective Loss 0.135066 LR 0.000250 Time 0.021715 +2023-10-02 21:45:31,085 - Epoch: [158][ 1180/ 1236] Overall Loss 0.134931 Objective Loss 0.134931 LR 0.000250 Time 0.021710 +2023-10-02 21:45:31,293 - Epoch: [158][ 1190/ 1236] Overall Loss 0.135002 Objective Loss 0.135002 LR 0.000250 Time 0.021702 +2023-10-02 21:45:31,505 - Epoch: [158][ 1200/ 1236] Overall Loss 0.134947 Objective Loss 0.134947 LR 0.000250 Time 0.021698 +2023-10-02 21:45:31,713 - Epoch: [158][ 1210/ 1236] Overall Loss 0.134827 Objective Loss 0.134827 LR 0.000250 Time 0.021690 +2023-10-02 21:45:31,925 - Epoch: [158][ 1220/ 1236] Overall Loss 0.134771 Objective Loss 0.134771 LR 0.000250 Time 0.021686 +2023-10-02 21:45:32,189 - Epoch: [158][ 1230/ 1236] Overall Loss 0.134886 Objective Loss 0.134886 LR 0.000250 Time 0.021723 +2023-10-02 21:45:32,312 - Epoch: [158][ 1236/ 1236] Overall Loss 0.134808 Objective Loss 0.134808 Top1 89.816701 Top5 98.574338 LR 0.000250 Time 0.021718 +2023-10-02 21:45:32,447 - --- validate (epoch=158)----------- +2023-10-02 21:45:32,447 - 29943 samples (256 per mini-batch) +2023-10-02 21:45:32,956 - Epoch: [158][ 10/ 117] Loss 0.275061 Top1 87.539062 Top5 98.671875 +2023-10-02 21:45:33,122 - Epoch: [158][ 20/ 117] Loss 0.304373 Top1 86.679688 Top5 98.535156 +2023-10-02 21:45:33,286 - Epoch: [158][ 30/ 117] Loss 0.315575 Top1 86.653646 Top5 98.554688 +2023-10-02 21:45:33,451 - Epoch: [158][ 40/ 117] Loss 0.320146 Top1 86.640625 Top5 98.544922 +2023-10-02 21:45:33,613 - Epoch: [158][ 50/ 117] Loss 0.310157 Top1 86.734375 Top5 98.593750 +2023-10-02 21:45:33,775 - Epoch: [158][ 60/ 117] Loss 0.308740 Top1 86.940104 Top5 98.593750 +2023-10-02 21:45:33,935 - Epoch: [158][ 70/ 117] Loss 0.307971 Top1 86.914062 Top5 98.582589 +2023-10-02 21:45:34,098 - Epoch: [158][ 80/ 117] Loss 0.303060 Top1 86.987305 Top5 98.623047 +2023-10-02 21:45:34,258 - Epoch: [158][ 90/ 117] Loss 0.301223 Top1 87.009549 Top5 98.611111 +2023-10-02 21:45:34,421 - Epoch: [158][ 100/ 117] Loss 0.299201 Top1 87.011719 Top5 98.632812 +2023-10-02 21:45:34,593 - Epoch: [158][ 110/ 117] Loss 0.297699 Top1 87.017045 Top5 98.639915 +2023-10-02 21:45:34,683 - Epoch: [158][ 117/ 117] Loss 0.300757 Top1 87.015329 Top5 98.630732 +2023-10-02 21:45:34,802 - ==> Top1: 87.015 Top5: 98.631 Loss: 0.301 + +2023-10-02 21:45:34,803 - ==> Confusion: +[[ 934 0 4 2 11 3 0 0 6 59 1 0 0 0 6 0 4 0 1 0 19] + [ 0 1072 0 2 5 7 1 17 0 1 1 1 1 0 0 3 3 0 7 4 6] + [ 1 1 988 10 1 0 8 9 0 1 0 0 7 2 1 3 1 2 9 4 8] + [ 1 3 12 983 2 0 2 1 3 0 4 0 4 3 28 2 2 6 12 2 19] + [ 17 7 0 1 975 6 0 0 0 12 1 1 1 3 8 4 10 0 0 1 3] + [ 3 43 0 0 3 978 2 27 1 4 2 7 1 10 5 1 3 0 4 2 20] + [ 0 2 33 0 0 1 1123 4 0 0 4 0 0 0 0 7 0 1 2 7 7] + [ 1 13 13 0 7 19 4 1080 1 3 3 3 3 4 2 0 1 4 41 6 10] + [ 15 3 0 0 3 4 0 0 970 35 10 1 1 19 14 2 5 1 2 0 4] + [ 78 1 2 1 10 1 0 0 23 955 0 1 0 24 10 0 1 0 0 1 11] + [ 1 2 9 6 0 1 2 1 8 1 976 2 1 13 6 0 2 4 6 1 11] + [ 0 0 1 0 0 14 0 1 0 0 0 959 22 8 0 2 2 14 0 8 4] + [ 0 0 3 3 1 2 1 0 0 1 3 32 978 4 2 5 1 9 1 5 17] + [ 1 0 0 0 2 3 0 0 7 4 1 6 0 1070 5 0 0 1 0 1 18] + [ 10 1 5 8 3 1 0 0 14 1 3 0 2 5 1023 0 1 1 11 0 12] + [ 0 0 2 2 5 0 1 0 0 1 1 5 5 0 0 1067 18 9 3 8 7] + [ 0 16 3 0 3 6 0 0 0 0 0 3 0 1 3 10 1102 0 2 3 9] + [ 0 0 1 0 0 0 1 0 0 0 0 4 24 2 2 8 0 991 0 0 5] + [ 3 5 2 12 2 0 1 13 4 1 2 1 2 0 11 0 0 0 997 0 12] + [ 0 0 6 1 2 0 6 9 0 0 1 12 4 2 2 2 8 0 0 1088 9] + [ 94 134 120 65 63 87 20 79 71 54 154 77 303 266 120 53 92 55 126 126 5746]] + +2023-10-02 21:45:34,804 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:45:34,804 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:45:34,810 - + +2023-10-02 21:45:34,810 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:45:35,854 - Epoch: [159][ 10/ 1236] Overall Loss 0.113464 Objective Loss 0.113464 LR 0.000250 Time 0.104349 +2023-10-02 21:45:36,064 - Epoch: [159][ 20/ 1236] Overall Loss 0.118308 Objective Loss 0.118308 LR 0.000250 Time 0.062625 +2023-10-02 21:45:36,274 - Epoch: [159][ 30/ 1236] Overall Loss 0.119971 Objective Loss 0.119971 LR 0.000250 Time 0.048735 +2023-10-02 21:45:36,484 - Epoch: [159][ 40/ 1236] Overall Loss 0.120342 Objective Loss 0.120342 LR 0.000250 Time 0.041793 +2023-10-02 21:45:36,693 - Epoch: [159][ 50/ 1236] Overall Loss 0.124065 Objective Loss 0.124065 LR 0.000250 Time 0.037585 +2023-10-02 21:45:36,903 - Epoch: [159][ 60/ 1236] Overall Loss 0.123810 Objective Loss 0.123810 LR 0.000250 Time 0.034819 +2023-10-02 21:45:37,113 - Epoch: [159][ 70/ 1236] Overall Loss 0.123865 Objective Loss 0.123865 LR 0.000250 Time 0.032822 +2023-10-02 21:45:37,323 - Epoch: [159][ 80/ 1236] Overall Loss 0.125095 Objective Loss 0.125095 LR 0.000250 Time 0.031348 +2023-10-02 21:45:37,533 - Epoch: [159][ 90/ 1236] Overall Loss 0.125501 Objective Loss 0.125501 LR 0.000250 Time 0.030188 +2023-10-02 21:45:37,745 - Epoch: [159][ 100/ 1236] Overall Loss 0.126063 Objective Loss 0.126063 LR 0.000250 Time 0.029283 +2023-10-02 21:45:37,955 - Epoch: [159][ 110/ 1236] Overall Loss 0.128340 Objective Loss 0.128340 LR 0.000250 Time 0.028519 +2023-10-02 21:45:38,165 - Epoch: [159][ 120/ 1236] Overall Loss 0.127944 Objective Loss 0.127944 LR 0.000250 Time 0.027890 +2023-10-02 21:45:38,375 - Epoch: [159][ 130/ 1236] Overall Loss 0.127994 Objective Loss 0.127994 LR 0.000250 Time 0.027358 +2023-10-02 21:45:38,588 - Epoch: [159][ 140/ 1236] Overall Loss 0.128379 Objective Loss 0.128379 LR 0.000250 Time 0.026923 +2023-10-02 21:45:38,800 - Epoch: [159][ 150/ 1236] Overall Loss 0.128641 Objective Loss 0.128641 LR 0.000250 Time 0.026527 +2023-10-02 21:45:39,013 - Epoch: [159][ 160/ 1236] Overall Loss 0.129350 Objective Loss 0.129350 LR 0.000250 Time 0.026198 +2023-10-02 21:45:39,224 - Epoch: [159][ 170/ 1236] Overall Loss 0.128686 Objective Loss 0.128686 LR 0.000250 Time 0.025893 +2023-10-02 21:45:39,437 - Epoch: [159][ 180/ 1236] Overall Loss 0.130460 Objective Loss 0.130460 LR 0.000250 Time 0.025636 +2023-10-02 21:45:39,649 - Epoch: [159][ 190/ 1236] Overall Loss 0.131068 Objective Loss 0.131068 LR 0.000250 Time 0.025392 +2023-10-02 21:45:39,862 - Epoch: [159][ 200/ 1236] Overall Loss 0.130952 Objective Loss 0.130952 LR 0.000250 Time 0.025187 +2023-10-02 21:45:40,074 - Epoch: [159][ 210/ 1236] Overall Loss 0.130522 Objective Loss 0.130522 LR 0.000250 Time 0.024988 +2023-10-02 21:45:40,287 - Epoch: [159][ 220/ 1236] Overall Loss 0.130718 Objective Loss 0.130718 LR 0.000250 Time 0.024817 +2023-10-02 21:45:40,498 - Epoch: [159][ 230/ 1236] Overall Loss 0.130608 Objective Loss 0.130608 LR 0.000250 Time 0.024652 +2023-10-02 21:45:40,711 - Epoch: [159][ 240/ 1236] Overall Loss 0.131779 Objective Loss 0.131779 LR 0.000250 Time 0.024510 +2023-10-02 21:45:40,923 - Epoch: [159][ 250/ 1236] Overall Loss 0.132443 Objective Loss 0.132443 LR 0.000250 Time 0.024369 +2023-10-02 21:45:41,135 - Epoch: [159][ 260/ 1236] Overall Loss 0.131977 Objective Loss 0.131977 LR 0.000250 Time 0.024249 +2023-10-02 21:45:41,347 - Epoch: [159][ 270/ 1236] Overall Loss 0.132645 Objective Loss 0.132645 LR 0.000250 Time 0.024128 +2023-10-02 21:45:41,560 - Epoch: [159][ 280/ 1236] Overall Loss 0.132976 Objective Loss 0.132976 LR 0.000250 Time 0.024026 +2023-10-02 21:45:41,771 - Epoch: [159][ 290/ 1236] Overall Loss 0.132836 Objective Loss 0.132836 LR 0.000250 Time 0.023921 +2023-10-02 21:45:41,984 - Epoch: [159][ 300/ 1236] Overall Loss 0.133236 Objective Loss 0.133236 LR 0.000250 Time 0.023831 +2023-10-02 21:45:42,196 - Epoch: [159][ 310/ 1236] Overall Loss 0.133241 Objective Loss 0.133241 LR 0.000250 Time 0.023741 +2023-10-02 21:45:42,409 - Epoch: [159][ 320/ 1236] Overall Loss 0.133191 Objective Loss 0.133191 LR 0.000250 Time 0.023664 +2023-10-02 21:45:42,621 - Epoch: [159][ 330/ 1236] Overall Loss 0.133212 Objective Loss 0.133212 LR 0.000250 Time 0.023584 +2023-10-02 21:45:42,834 - Epoch: [159][ 340/ 1236] Overall Loss 0.133471 Objective Loss 0.133471 LR 0.000250 Time 0.023515 +2023-10-02 21:45:43,043 - Epoch: [159][ 350/ 1236] Overall Loss 0.133379 Objective Loss 0.133379 LR 0.000250 Time 0.023440 +2023-10-02 21:45:43,253 - Epoch: [159][ 360/ 1236] Overall Loss 0.133653 Objective Loss 0.133653 LR 0.000250 Time 0.023372 +2023-10-02 21:45:43,463 - Epoch: [159][ 370/ 1236] Overall Loss 0.133557 Objective Loss 0.133557 LR 0.000250 Time 0.023306 +2023-10-02 21:45:43,673 - Epoch: [159][ 380/ 1236] Overall Loss 0.133224 Objective Loss 0.133224 LR 0.000250 Time 0.023245 +2023-10-02 21:45:43,883 - Epoch: [159][ 390/ 1236] Overall Loss 0.133980 Objective Loss 0.133980 LR 0.000250 Time 0.023186 +2023-10-02 21:45:44,093 - Epoch: [159][ 400/ 1236] Overall Loss 0.133957 Objective Loss 0.133957 LR 0.000250 Time 0.023131 +2023-10-02 21:45:44,303 - Epoch: [159][ 410/ 1236] Overall Loss 0.133801 Objective Loss 0.133801 LR 0.000250 Time 0.023078 +2023-10-02 21:45:44,513 - Epoch: [159][ 420/ 1236] Overall Loss 0.134049 Objective Loss 0.134049 LR 0.000250 Time 0.023028 +2023-10-02 21:45:44,723 - Epoch: [159][ 430/ 1236] Overall Loss 0.133716 Objective Loss 0.133716 LR 0.000250 Time 0.022979 +2023-10-02 21:45:44,933 - Epoch: [159][ 440/ 1236] Overall Loss 0.133512 Objective Loss 0.133512 LR 0.000250 Time 0.022934 +2023-10-02 21:45:45,143 - Epoch: [159][ 450/ 1236] Overall Loss 0.133388 Objective Loss 0.133388 LR 0.000250 Time 0.022890 +2023-10-02 21:45:45,352 - Epoch: [159][ 460/ 1236] Overall Loss 0.133099 Objective Loss 0.133099 LR 0.000250 Time 0.022847 +2023-10-02 21:45:45,562 - Epoch: [159][ 470/ 1236] Overall Loss 0.133198 Objective Loss 0.133198 LR 0.000250 Time 0.022806 +2023-10-02 21:45:45,772 - Epoch: [159][ 480/ 1236] Overall Loss 0.133511 Objective Loss 0.133511 LR 0.000250 Time 0.022768 +2023-10-02 21:45:45,982 - Epoch: [159][ 490/ 1236] Overall Loss 0.133612 Objective Loss 0.133612 LR 0.000250 Time 0.022731 +2023-10-02 21:45:46,192 - Epoch: [159][ 500/ 1236] Overall Loss 0.133429 Objective Loss 0.133429 LR 0.000250 Time 0.022697 +2023-10-02 21:45:46,402 - Epoch: [159][ 510/ 1236] Overall Loss 0.133512 Objective Loss 0.133512 LR 0.000250 Time 0.022662 +2023-10-02 21:45:46,612 - Epoch: [159][ 520/ 1236] Overall Loss 0.133621 Objective Loss 0.133621 LR 0.000250 Time 0.022630 +2023-10-02 21:45:46,822 - Epoch: [159][ 530/ 1236] Overall Loss 0.133874 Objective Loss 0.133874 LR 0.000250 Time 0.022598 +2023-10-02 21:45:47,032 - Epoch: [159][ 540/ 1236] Overall Loss 0.133484 Objective Loss 0.133484 LR 0.000250 Time 0.022568 +2023-10-02 21:45:47,242 - Epoch: [159][ 550/ 1236] Overall Loss 0.133707 Objective Loss 0.133707 LR 0.000250 Time 0.022539 +2023-10-02 21:45:47,452 - Epoch: [159][ 560/ 1236] Overall Loss 0.133717 Objective Loss 0.133717 LR 0.000250 Time 0.022511 +2023-10-02 21:45:47,662 - Epoch: [159][ 570/ 1236] Overall Loss 0.133797 Objective Loss 0.133797 LR 0.000250 Time 0.022483 +2023-10-02 21:45:47,872 - Epoch: [159][ 580/ 1236] Overall Loss 0.133909 Objective Loss 0.133909 LR 0.000250 Time 0.022458 +2023-10-02 21:45:48,082 - Epoch: [159][ 590/ 1236] Overall Loss 0.133439 Objective Loss 0.133439 LR 0.000250 Time 0.022432 +2023-10-02 21:45:48,293 - Epoch: [159][ 600/ 1236] Overall Loss 0.133571 Objective Loss 0.133571 LR 0.000250 Time 0.022408 +2023-10-02 21:45:48,502 - Epoch: [159][ 610/ 1236] Overall Loss 0.133669 Objective Loss 0.133669 LR 0.000250 Time 0.022384 +2023-10-02 21:45:48,712 - Epoch: [159][ 620/ 1236] Overall Loss 0.133285 Objective Loss 0.133285 LR 0.000250 Time 0.022362 +2023-10-02 21:45:48,922 - Epoch: [159][ 630/ 1236] Overall Loss 0.133224 Objective Loss 0.133224 LR 0.000250 Time 0.022339 +2023-10-02 21:45:49,133 - Epoch: [159][ 640/ 1236] Overall Loss 0.133203 Objective Loss 0.133203 LR 0.000250 Time 0.022318 +2023-10-02 21:45:49,342 - Epoch: [159][ 650/ 1236] Overall Loss 0.132966 Objective Loss 0.132966 LR 0.000250 Time 0.022297 +2023-10-02 21:45:49,552 - Epoch: [159][ 660/ 1236] Overall Loss 0.133093 Objective Loss 0.133093 LR 0.000250 Time 0.022277 +2023-10-02 21:45:49,762 - Epoch: [159][ 670/ 1236] Overall Loss 0.133219 Objective Loss 0.133219 LR 0.000250 Time 0.022257 +2023-10-02 21:45:49,972 - Epoch: [159][ 680/ 1236] Overall Loss 0.132950 Objective Loss 0.132950 LR 0.000250 Time 0.022238 +2023-10-02 21:45:50,182 - Epoch: [159][ 690/ 1236] Overall Loss 0.132978 Objective Loss 0.132978 LR 0.000250 Time 0.022220 +2023-10-02 21:45:50,392 - Epoch: [159][ 700/ 1236] Overall Loss 0.133255 Objective Loss 0.133255 LR 0.000250 Time 0.022202 +2023-10-02 21:45:50,602 - Epoch: [159][ 710/ 1236] Overall Loss 0.133512 Objective Loss 0.133512 LR 0.000250 Time 0.022184 +2023-10-02 21:45:50,812 - Epoch: [159][ 720/ 1236] Overall Loss 0.133693 Objective Loss 0.133693 LR 0.000250 Time 0.022168 +2023-10-02 21:45:51,022 - Epoch: [159][ 730/ 1236] Overall Loss 0.133813 Objective Loss 0.133813 LR 0.000250 Time 0.022151 +2023-10-02 21:45:51,232 - Epoch: [159][ 740/ 1236] Overall Loss 0.133899 Objective Loss 0.133899 LR 0.000250 Time 0.022135 +2023-10-02 21:45:51,442 - Epoch: [159][ 750/ 1236] Overall Loss 0.133866 Objective Loss 0.133866 LR 0.000250 Time 0.022119 +2023-10-02 21:45:51,652 - Epoch: [159][ 760/ 1236] Overall Loss 0.134006 Objective Loss 0.134006 LR 0.000250 Time 0.022104 +2023-10-02 21:45:51,862 - Epoch: [159][ 770/ 1236] Overall Loss 0.133701 Objective Loss 0.133701 LR 0.000250 Time 0.022089 +2023-10-02 21:45:52,072 - Epoch: [159][ 780/ 1236] Overall Loss 0.133787 Objective Loss 0.133787 LR 0.000250 Time 0.022075 +2023-10-02 21:45:52,282 - Epoch: [159][ 790/ 1236] Overall Loss 0.134087 Objective Loss 0.134087 LR 0.000250 Time 0.022061 +2023-10-02 21:45:52,492 - Epoch: [159][ 800/ 1236] Overall Loss 0.134132 Objective Loss 0.134132 LR 0.000250 Time 0.022047 +2023-10-02 21:45:52,702 - Epoch: [159][ 810/ 1236] Overall Loss 0.134229 Objective Loss 0.134229 LR 0.000250 Time 0.022034 +2023-10-02 21:45:52,912 - Epoch: [159][ 820/ 1236] Overall Loss 0.134434 Objective Loss 0.134434 LR 0.000250 Time 0.022021 +2023-10-02 21:45:53,122 - Epoch: [159][ 830/ 1236] Overall Loss 0.134295 Objective Loss 0.134295 LR 0.000250 Time 0.022008 +2023-10-02 21:45:53,332 - Epoch: [159][ 840/ 1236] Overall Loss 0.134526 Objective Loss 0.134526 LR 0.000250 Time 0.021996 +2023-10-02 21:45:53,542 - Epoch: [159][ 850/ 1236] Overall Loss 0.134466 Objective Loss 0.134466 LR 0.000250 Time 0.021983 +2023-10-02 21:45:53,752 - Epoch: [159][ 860/ 1236] Overall Loss 0.134719 Objective Loss 0.134719 LR 0.000250 Time 0.021972 +2023-10-02 21:45:53,962 - Epoch: [159][ 870/ 1236] Overall Loss 0.134642 Objective Loss 0.134642 LR 0.000250 Time 0.021960 +2023-10-02 21:45:54,172 - Epoch: [159][ 880/ 1236] Overall Loss 0.134505 Objective Loss 0.134505 LR 0.000250 Time 0.021949 +2023-10-02 21:45:54,382 - Epoch: [159][ 890/ 1236] Overall Loss 0.134509 Objective Loss 0.134509 LR 0.000250 Time 0.021938 +2023-10-02 21:45:54,592 - Epoch: [159][ 900/ 1236] Overall Loss 0.134508 Objective Loss 0.134508 LR 0.000250 Time 0.021927 +2023-10-02 21:45:54,802 - Epoch: [159][ 910/ 1236] Overall Loss 0.134263 Objective Loss 0.134263 LR 0.000250 Time 0.021916 +2023-10-02 21:45:55,012 - Epoch: [159][ 920/ 1236] Overall Loss 0.134295 Objective Loss 0.134295 LR 0.000250 Time 0.021906 +2023-10-02 21:45:55,222 - Epoch: [159][ 930/ 1236] Overall Loss 0.134398 Objective Loss 0.134398 LR 0.000250 Time 0.021896 +2023-10-02 21:45:55,433 - Epoch: [159][ 940/ 1236] Overall Loss 0.134526 Objective Loss 0.134526 LR 0.000250 Time 0.021886 +2023-10-02 21:45:55,642 - Epoch: [159][ 950/ 1236] Overall Loss 0.134385 Objective Loss 0.134385 LR 0.000250 Time 0.021877 +2023-10-02 21:45:55,853 - Epoch: [159][ 960/ 1236] Overall Loss 0.134438 Objective Loss 0.134438 LR 0.000250 Time 0.021867 +2023-10-02 21:45:56,062 - Epoch: [159][ 970/ 1236] Overall Loss 0.134445 Objective Loss 0.134445 LR 0.000250 Time 0.021858 +2023-10-02 21:45:56,273 - Epoch: [159][ 980/ 1236] Overall Loss 0.134375 Objective Loss 0.134375 LR 0.000250 Time 0.021849 +2023-10-02 21:45:56,482 - Epoch: [159][ 990/ 1236] Overall Loss 0.134288 Objective Loss 0.134288 LR 0.000250 Time 0.021840 +2023-10-02 21:45:56,692 - Epoch: [159][ 1000/ 1236] Overall Loss 0.134116 Objective Loss 0.134116 LR 0.000250 Time 0.021831 +2023-10-02 21:45:56,902 - Epoch: [159][ 1010/ 1236] Overall Loss 0.134149 Objective Loss 0.134149 LR 0.000250 Time 0.021822 +2023-10-02 21:45:57,112 - Epoch: [159][ 1020/ 1236] Overall Loss 0.134056 Objective Loss 0.134056 LR 0.000250 Time 0.021814 +2023-10-02 21:45:57,322 - Epoch: [159][ 1030/ 1236] Overall Loss 0.134154 Objective Loss 0.134154 LR 0.000250 Time 0.021806 +2023-10-02 21:45:57,532 - Epoch: [159][ 1040/ 1236] Overall Loss 0.134270 Objective Loss 0.134270 LR 0.000250 Time 0.021798 +2023-10-02 21:45:57,742 - Epoch: [159][ 1050/ 1236] Overall Loss 0.134323 Objective Loss 0.134323 LR 0.000250 Time 0.021789 +2023-10-02 21:45:57,952 - Epoch: [159][ 1060/ 1236] Overall Loss 0.134151 Objective Loss 0.134151 LR 0.000250 Time 0.021782 +2023-10-02 21:45:58,162 - Epoch: [159][ 1070/ 1236] Overall Loss 0.134229 Objective Loss 0.134229 LR 0.000250 Time 0.021774 +2023-10-02 21:45:58,373 - Epoch: [159][ 1080/ 1236] Overall Loss 0.134261 Objective Loss 0.134261 LR 0.000250 Time 0.021767 +2023-10-02 21:45:58,582 - Epoch: [159][ 1090/ 1236] Overall Loss 0.133975 Objective Loss 0.133975 LR 0.000250 Time 0.021759 +2023-10-02 21:45:58,793 - Epoch: [159][ 1100/ 1236] Overall Loss 0.133991 Objective Loss 0.133991 LR 0.000250 Time 0.021752 +2023-10-02 21:45:59,002 - Epoch: [159][ 1110/ 1236] Overall Loss 0.133971 Objective Loss 0.133971 LR 0.000250 Time 0.021745 +2023-10-02 21:45:59,212 - Epoch: [159][ 1120/ 1236] Overall Loss 0.133904 Objective Loss 0.133904 LR 0.000250 Time 0.021738 +2023-10-02 21:45:59,422 - Epoch: [159][ 1130/ 1236] Overall Loss 0.133892 Objective Loss 0.133892 LR 0.000250 Time 0.021731 +2023-10-02 21:45:59,633 - Epoch: [159][ 1140/ 1236] Overall Loss 0.133916 Objective Loss 0.133916 LR 0.000250 Time 0.021725 +2023-10-02 21:45:59,842 - Epoch: [159][ 1150/ 1236] Overall Loss 0.134112 Objective Loss 0.134112 LR 0.000250 Time 0.021718 +2023-10-02 21:46:00,053 - Epoch: [159][ 1160/ 1236] Overall Loss 0.134132 Objective Loss 0.134132 LR 0.000250 Time 0.021712 +2023-10-02 21:46:00,263 - Epoch: [159][ 1170/ 1236] Overall Loss 0.134099 Objective Loss 0.134099 LR 0.000250 Time 0.021706 +2023-10-02 21:46:00,473 - Epoch: [159][ 1180/ 1236] Overall Loss 0.134113 Objective Loss 0.134113 LR 0.000250 Time 0.021700 +2023-10-02 21:46:00,683 - Epoch: [159][ 1190/ 1236] Overall Loss 0.134070 Objective Loss 0.134070 LR 0.000250 Time 0.021693 +2023-10-02 21:46:00,893 - Epoch: [159][ 1200/ 1236] Overall Loss 0.134099 Objective Loss 0.134099 LR 0.000250 Time 0.021688 +2023-10-02 21:46:01,104 - Epoch: [159][ 1210/ 1236] Overall Loss 0.134086 Objective Loss 0.134086 LR 0.000250 Time 0.021682 +2023-10-02 21:46:01,314 - Epoch: [159][ 1220/ 1236] Overall Loss 0.134123 Objective Loss 0.134123 LR 0.000250 Time 0.021676 +2023-10-02 21:46:01,579 - Epoch: [159][ 1230/ 1236] Overall Loss 0.134048 Objective Loss 0.134048 LR 0.000250 Time 0.021715 +2023-10-02 21:46:01,702 - Epoch: [159][ 1236/ 1236] Overall Loss 0.134078 Objective Loss 0.134078 Top1 89.816701 Top5 98.778004 LR 0.000250 Time 0.021709 +2023-10-02 21:46:01,847 - --- validate (epoch=159)----------- +2023-10-02 21:46:01,848 - 29943 samples (256 per mini-batch) +2023-10-02 21:46:02,344 - Epoch: [159][ 10/ 117] Loss 0.273380 Top1 87.382812 Top5 98.789062 +2023-10-02 21:46:02,501 - Epoch: [159][ 20/ 117] Loss 0.293350 Top1 86.875000 Top5 98.535156 +2023-10-02 21:46:02,657 - Epoch: [159][ 30/ 117] Loss 0.311827 Top1 86.588542 Top5 98.502604 +2023-10-02 21:46:02,812 - Epoch: [159][ 40/ 117] Loss 0.318897 Top1 86.503906 Top5 98.515625 +2023-10-02 21:46:02,968 - Epoch: [159][ 50/ 117] Loss 0.317385 Top1 86.703125 Top5 98.562500 +2023-10-02 21:46:03,122 - Epoch: [159][ 60/ 117] Loss 0.313718 Top1 86.718750 Top5 98.522135 +2023-10-02 21:46:03,279 - Epoch: [159][ 70/ 117] Loss 0.313742 Top1 86.819196 Top5 98.504464 +2023-10-02 21:46:03,438 - Epoch: [159][ 80/ 117] Loss 0.305815 Top1 86.967773 Top5 98.559570 +2023-10-02 21:46:03,597 - Epoch: [159][ 90/ 117] Loss 0.302795 Top1 86.987847 Top5 98.606771 +2023-10-02 21:46:03,756 - Epoch: [159][ 100/ 117] Loss 0.300623 Top1 87.011719 Top5 98.621094 +2023-10-02 21:46:03,923 - Epoch: [159][ 110/ 117] Loss 0.298046 Top1 86.999290 Top5 98.565341 +2023-10-02 21:46:04,013 - Epoch: [159][ 117/ 117] Loss 0.299751 Top1 86.898440 Top5 98.570618 +2023-10-02 21:46:04,110 - ==> Top1: 86.898 Top5: 98.571 Loss: 0.300 + +2023-10-02 21:46:04,111 - ==> Confusion: +[[ 950 1 1 2 5 4 0 0 6 51 2 1 0 1 3 1 5 0 1 0 16] + [ 0 1074 0 2 2 15 0 16 0 1 0 0 1 0 0 3 1 0 6 1 9] + [ 1 1 984 4 1 0 16 9 0 2 5 0 6 2 1 2 1 2 11 3 5] + [ 3 4 14 985 1 0 3 1 4 0 3 0 5 3 22 1 0 5 15 0 20] + [ 21 3 1 0 981 5 0 0 2 9 3 0 2 2 8 3 7 0 0 1 2] + [ 5 43 0 1 2 988 1 25 4 5 0 4 2 10 4 0 3 1 5 2 11] + [ 0 3 24 0 0 2 1135 6 0 0 4 1 0 0 0 2 0 0 2 9 3] + [ 2 14 10 0 4 16 6 1088 1 2 3 3 5 7 1 0 0 1 42 6 7] + [ 17 3 0 2 1 3 0 1 973 38 13 2 1 10 15 2 2 0 2 1 3] + [ 104 0 1 1 5 4 0 0 26 939 2 2 0 18 4 3 0 0 0 1 9] + [ 1 2 8 7 0 2 5 1 11 1 980 2 0 9 5 0 2 1 3 3 10] + [ 0 1 0 0 0 19 1 2 0 0 0 964 9 7 0 2 2 16 0 7 5] + [ 0 1 4 4 0 3 3 0 2 0 3 39 962 2 1 5 1 15 2 3 18] + [ 0 0 2 0 5 6 0 0 13 9 5 7 0 1049 4 0 0 1 0 1 17] + [ 18 2 4 18 4 0 0 0 19 2 2 0 1 1 1004 0 1 1 14 0 10] + [ 0 0 2 1 5 0 2 0 0 0 1 6 7 0 0 1065 17 11 3 8 6] + [ 1 20 0 0 6 5 0 0 0 0 0 3 0 1 2 5 1096 0 1 4 17] + [ 0 0 1 0 0 0 3 0 0 1 0 3 9 2 1 4 0 1009 0 1 4] + [ 2 5 2 14 1 2 0 21 3 0 0 1 2 0 7 0 0 0 997 0 11] + [ 0 0 5 2 1 2 5 5 0 1 0 12 4 3 0 0 8 1 1 1097 5] + [ 119 173 120 65 61 99 29 83 80 67 143 72 278 246 100 40 97 63 127 143 5700]] + +2023-10-02 21:46:04,113 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:46:04,113 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:46:04,119 - + +2023-10-02 21:46:04,119 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:46:05,147 - Epoch: [160][ 10/ 1236] Overall Loss 0.125745 Objective Loss 0.125745 LR 0.000250 Time 0.102779 +2023-10-02 21:46:05,357 - Epoch: [160][ 20/ 1236] Overall Loss 0.125097 Objective Loss 0.125097 LR 0.000250 Time 0.061861 +2023-10-02 21:46:05,567 - Epoch: [160][ 30/ 1236] Overall Loss 0.128697 Objective Loss 0.128697 LR 0.000250 Time 0.048222 +2023-10-02 21:46:05,777 - Epoch: [160][ 40/ 1236] Overall Loss 0.136321 Objective Loss 0.136321 LR 0.000250 Time 0.041418 +2023-10-02 21:46:05,987 - Epoch: [160][ 50/ 1236] Overall Loss 0.139600 Objective Loss 0.139600 LR 0.000250 Time 0.037319 +2023-10-02 21:46:06,198 - Epoch: [160][ 60/ 1236] Overall Loss 0.139876 Objective Loss 0.139876 LR 0.000250 Time 0.034602 +2023-10-02 21:46:06,408 - Epoch: [160][ 70/ 1236] Overall Loss 0.134889 Objective Loss 0.134889 LR 0.000250 Time 0.032662 +2023-10-02 21:46:06,619 - Epoch: [160][ 80/ 1236] Overall Loss 0.134135 Objective Loss 0.134135 LR 0.000250 Time 0.031205 +2023-10-02 21:46:06,829 - Epoch: [160][ 90/ 1236] Overall Loss 0.134369 Objective Loss 0.134369 LR 0.000250 Time 0.030073 +2023-10-02 21:46:07,040 - Epoch: [160][ 100/ 1236] Overall Loss 0.134611 Objective Loss 0.134611 LR 0.000250 Time 0.029166 +2023-10-02 21:46:07,250 - Epoch: [160][ 110/ 1236] Overall Loss 0.132953 Objective Loss 0.132953 LR 0.000250 Time 0.028426 +2023-10-02 21:46:07,462 - Epoch: [160][ 120/ 1236] Overall Loss 0.130767 Objective Loss 0.130767 LR 0.000250 Time 0.027814 +2023-10-02 21:46:07,673 - Epoch: [160][ 130/ 1236] Overall Loss 0.130760 Objective Loss 0.130760 LR 0.000250 Time 0.027290 +2023-10-02 21:46:07,886 - Epoch: [160][ 140/ 1236] Overall Loss 0.132592 Objective Loss 0.132592 LR 0.000250 Time 0.026856 +2023-10-02 21:46:08,097 - Epoch: [160][ 150/ 1236] Overall Loss 0.133830 Objective Loss 0.133830 LR 0.000250 Time 0.026465 +2023-10-02 21:46:08,310 - Epoch: [160][ 160/ 1236] Overall Loss 0.133586 Objective Loss 0.133586 LR 0.000250 Time 0.026138 +2023-10-02 21:46:08,521 - Epoch: [160][ 170/ 1236] Overall Loss 0.133192 Objective Loss 0.133192 LR 0.000250 Time 0.025835 +2023-10-02 21:46:08,734 - Epoch: [160][ 180/ 1236] Overall Loss 0.133661 Objective Loss 0.133661 LR 0.000250 Time 0.025578 +2023-10-02 21:46:08,945 - Epoch: [160][ 190/ 1236] Overall Loss 0.132236 Objective Loss 0.132236 LR 0.000250 Time 0.025337 +2023-10-02 21:46:09,157 - Epoch: [160][ 200/ 1236] Overall Loss 0.132675 Objective Loss 0.132675 LR 0.000250 Time 0.025126 +2023-10-02 21:46:09,368 - Epoch: [160][ 210/ 1236] Overall Loss 0.133980 Objective Loss 0.133980 LR 0.000250 Time 0.024937 +2023-10-02 21:46:09,580 - Epoch: [160][ 220/ 1236] Overall Loss 0.133238 Objective Loss 0.133238 LR 0.000250 Time 0.024762 +2023-10-02 21:46:09,791 - Epoch: [160][ 230/ 1236] Overall Loss 0.133892 Objective Loss 0.133892 LR 0.000250 Time 0.024599 +2023-10-02 21:46:10,004 - Epoch: [160][ 240/ 1236] Overall Loss 0.133835 Objective Loss 0.133835 LR 0.000250 Time 0.024460 +2023-10-02 21:46:10,216 - Epoch: [160][ 250/ 1236] Overall Loss 0.133379 Objective Loss 0.133379 LR 0.000250 Time 0.024321 +2023-10-02 21:46:10,429 - Epoch: [160][ 260/ 1236] Overall Loss 0.133069 Objective Loss 0.133069 LR 0.000250 Time 0.024205 +2023-10-02 21:46:10,640 - Epoch: [160][ 270/ 1236] Overall Loss 0.133198 Objective Loss 0.133198 LR 0.000250 Time 0.024090 +2023-10-02 21:46:10,853 - Epoch: [160][ 280/ 1236] Overall Loss 0.133547 Objective Loss 0.133547 LR 0.000250 Time 0.023988 +2023-10-02 21:46:11,065 - Epoch: [160][ 290/ 1236] Overall Loss 0.133575 Objective Loss 0.133575 LR 0.000250 Time 0.023886 +2023-10-02 21:46:11,275 - Epoch: [160][ 300/ 1236] Overall Loss 0.133789 Objective Loss 0.133789 LR 0.000250 Time 0.023789 +2023-10-02 21:46:11,485 - Epoch: [160][ 310/ 1236] Overall Loss 0.133689 Objective Loss 0.133689 LR 0.000250 Time 0.023693 +2023-10-02 21:46:11,693 - Epoch: [160][ 320/ 1236] Overall Loss 0.133334 Objective Loss 0.133334 LR 0.000250 Time 0.023604 +2023-10-02 21:46:11,903 - Epoch: [160][ 330/ 1236] Overall Loss 0.133342 Objective Loss 0.133342 LR 0.000250 Time 0.023523 +2023-10-02 21:46:12,114 - Epoch: [160][ 340/ 1236] Overall Loss 0.133348 Objective Loss 0.133348 LR 0.000250 Time 0.023451 +2023-10-02 21:46:12,325 - Epoch: [160][ 350/ 1236] Overall Loss 0.133237 Objective Loss 0.133237 LR 0.000250 Time 0.023381 +2023-10-02 21:46:12,535 - Epoch: [160][ 360/ 1236] Overall Loss 0.132886 Objective Loss 0.132886 LR 0.000250 Time 0.023314 +2023-10-02 21:46:12,746 - Epoch: [160][ 370/ 1236] Overall Loss 0.133657 Objective Loss 0.133657 LR 0.000250 Time 0.023253 +2023-10-02 21:46:12,956 - Epoch: [160][ 380/ 1236] Overall Loss 0.133029 Objective Loss 0.133029 LR 0.000250 Time 0.023193 +2023-10-02 21:46:13,166 - Epoch: [160][ 390/ 1236] Overall Loss 0.132666 Objective Loss 0.132666 LR 0.000250 Time 0.023137 +2023-10-02 21:46:13,377 - Epoch: [160][ 400/ 1236] Overall Loss 0.132522 Objective Loss 0.132522 LR 0.000250 Time 0.023084 +2023-10-02 21:46:13,588 - Epoch: [160][ 410/ 1236] Overall Loss 0.131909 Objective Loss 0.131909 LR 0.000250 Time 0.023034 +2023-10-02 21:46:13,798 - Epoch: [160][ 420/ 1236] Overall Loss 0.132288 Objective Loss 0.132288 LR 0.000250 Time 0.022985 +2023-10-02 21:46:14,009 - Epoch: [160][ 430/ 1236] Overall Loss 0.132203 Objective Loss 0.132203 LR 0.000250 Time 0.022940 +2023-10-02 21:46:14,220 - Epoch: [160][ 440/ 1236] Overall Loss 0.132332 Objective Loss 0.132332 LR 0.000250 Time 0.022898 +2023-10-02 21:46:14,429 - Epoch: [160][ 450/ 1236] Overall Loss 0.131891 Objective Loss 0.131891 LR 0.000250 Time 0.022854 +2023-10-02 21:46:14,640 - Epoch: [160][ 460/ 1236] Overall Loss 0.132023 Objective Loss 0.132023 LR 0.000250 Time 0.022815 +2023-10-02 21:46:14,851 - Epoch: [160][ 470/ 1236] Overall Loss 0.131926 Objective Loss 0.131926 LR 0.000250 Time 0.022777 +2023-10-02 21:46:15,062 - Epoch: [160][ 480/ 1236] Overall Loss 0.131735 Objective Loss 0.131735 LR 0.000250 Time 0.022741 +2023-10-02 21:46:15,273 - Epoch: [160][ 490/ 1236] Overall Loss 0.131123 Objective Loss 0.131123 LR 0.000250 Time 0.022706 +2023-10-02 21:46:15,484 - Epoch: [160][ 500/ 1236] Overall Loss 0.130990 Objective Loss 0.130990 LR 0.000250 Time 0.022674 +2023-10-02 21:46:15,695 - Epoch: [160][ 510/ 1236] Overall Loss 0.130924 Objective Loss 0.130924 LR 0.000250 Time 0.022642 +2023-10-02 21:46:15,906 - Epoch: [160][ 520/ 1236] Overall Loss 0.130736 Objective Loss 0.130736 LR 0.000250 Time 0.022611 +2023-10-02 21:46:16,117 - Epoch: [160][ 530/ 1236] Overall Loss 0.130940 Objective Loss 0.130940 LR 0.000250 Time 0.022582 +2023-10-02 21:46:16,327 - Epoch: [160][ 540/ 1236] Overall Loss 0.130817 Objective Loss 0.130817 LR 0.000250 Time 0.022553 +2023-10-02 21:46:16,538 - Epoch: [160][ 550/ 1236] Overall Loss 0.130735 Objective Loss 0.130735 LR 0.000250 Time 0.022525 +2023-10-02 21:46:16,749 - Epoch: [160][ 560/ 1236] Overall Loss 0.130973 Objective Loss 0.130973 LR 0.000250 Time 0.022500 +2023-10-02 21:46:16,960 - Epoch: [160][ 570/ 1236] Overall Loss 0.131284 Objective Loss 0.131284 LR 0.000250 Time 0.022474 +2023-10-02 21:46:17,171 - Epoch: [160][ 580/ 1236] Overall Loss 0.131339 Objective Loss 0.131339 LR 0.000250 Time 0.022450 +2023-10-02 21:46:17,382 - Epoch: [160][ 590/ 1236] Overall Loss 0.131347 Objective Loss 0.131347 LR 0.000250 Time 0.022426 +2023-10-02 21:46:17,593 - Epoch: [160][ 600/ 1236] Overall Loss 0.131195 Objective Loss 0.131195 LR 0.000250 Time 0.022404 +2023-10-02 21:46:17,803 - Epoch: [160][ 610/ 1236] Overall Loss 0.131480 Objective Loss 0.131480 LR 0.000250 Time 0.022381 +2023-10-02 21:46:18,014 - Epoch: [160][ 620/ 1236] Overall Loss 0.131430 Objective Loss 0.131430 LR 0.000250 Time 0.022360 +2023-10-02 21:46:18,225 - Epoch: [160][ 630/ 1236] Overall Loss 0.131471 Objective Loss 0.131471 LR 0.000250 Time 0.022337 +2023-10-02 21:46:18,436 - Epoch: [160][ 640/ 1236] Overall Loss 0.131442 Objective Loss 0.131442 LR 0.000250 Time 0.022317 +2023-10-02 21:46:18,647 - Epoch: [160][ 650/ 1236] Overall Loss 0.131377 Objective Loss 0.131377 LR 0.000250 Time 0.022297 +2023-10-02 21:46:18,858 - Epoch: [160][ 660/ 1236] Overall Loss 0.131343 Objective Loss 0.131343 LR 0.000250 Time 0.022278 +2023-10-02 21:46:19,069 - Epoch: [160][ 670/ 1236] Overall Loss 0.131976 Objective Loss 0.131976 LR 0.000250 Time 0.022260 +2023-10-02 21:46:19,280 - Epoch: [160][ 680/ 1236] Overall Loss 0.131745 Objective Loss 0.131745 LR 0.000250 Time 0.022243 +2023-10-02 21:46:19,490 - Epoch: [160][ 690/ 1236] Overall Loss 0.131734 Objective Loss 0.131734 LR 0.000250 Time 0.022225 +2023-10-02 21:46:19,701 - Epoch: [160][ 700/ 1236] Overall Loss 0.131627 Objective Loss 0.131627 LR 0.000250 Time 0.022208 +2023-10-02 21:46:19,912 - Epoch: [160][ 710/ 1236] Overall Loss 0.131809 Objective Loss 0.131809 LR 0.000250 Time 0.022191 +2023-10-02 21:46:20,122 - Epoch: [160][ 720/ 1236] Overall Loss 0.131945 Objective Loss 0.131945 LR 0.000250 Time 0.022175 +2023-10-02 21:46:20,333 - Epoch: [160][ 730/ 1236] Overall Loss 0.131943 Objective Loss 0.131943 LR 0.000250 Time 0.022159 +2023-10-02 21:46:20,544 - Epoch: [160][ 740/ 1236] Overall Loss 0.132025 Objective Loss 0.132025 LR 0.000250 Time 0.022144 +2023-10-02 21:46:20,755 - Epoch: [160][ 750/ 1236] Overall Loss 0.131956 Objective Loss 0.131956 LR 0.000250 Time 0.022130 +2023-10-02 21:46:20,966 - Epoch: [160][ 760/ 1236] Overall Loss 0.132319 Objective Loss 0.132319 LR 0.000250 Time 0.022116 +2023-10-02 21:46:21,177 - Epoch: [160][ 770/ 1236] Overall Loss 0.132248 Objective Loss 0.132248 LR 0.000250 Time 0.022102 +2023-10-02 21:46:21,388 - Epoch: [160][ 780/ 1236] Overall Loss 0.132478 Objective Loss 0.132478 LR 0.000250 Time 0.022089 +2023-10-02 21:46:21,599 - Epoch: [160][ 790/ 1236] Overall Loss 0.132645 Objective Loss 0.132645 LR 0.000250 Time 0.022076 +2023-10-02 21:46:21,810 - Epoch: [160][ 800/ 1236] Overall Loss 0.132676 Objective Loss 0.132676 LR 0.000250 Time 0.022063 +2023-10-02 21:46:22,021 - Epoch: [160][ 810/ 1236] Overall Loss 0.132960 Objective Loss 0.132960 LR 0.000250 Time 0.022049 +2023-10-02 21:46:22,232 - Epoch: [160][ 820/ 1236] Overall Loss 0.132826 Objective Loss 0.132826 LR 0.000250 Time 0.022038 +2023-10-02 21:46:22,443 - Epoch: [160][ 830/ 1236] Overall Loss 0.133079 Objective Loss 0.133079 LR 0.000250 Time 0.022026 +2023-10-02 21:46:22,654 - Epoch: [160][ 840/ 1236] Overall Loss 0.133373 Objective Loss 0.133373 LR 0.000250 Time 0.022014 +2023-10-02 21:46:22,865 - Epoch: [160][ 850/ 1236] Overall Loss 0.133564 Objective Loss 0.133564 LR 0.000250 Time 0.022003 +2023-10-02 21:46:23,076 - Epoch: [160][ 860/ 1236] Overall Loss 0.133630 Objective Loss 0.133630 LR 0.000250 Time 0.021992 +2023-10-02 21:46:23,287 - Epoch: [160][ 870/ 1236] Overall Loss 0.133525 Objective Loss 0.133525 LR 0.000250 Time 0.021980 +2023-10-02 21:46:23,498 - Epoch: [160][ 880/ 1236] Overall Loss 0.133709 Objective Loss 0.133709 LR 0.000250 Time 0.021970 +2023-10-02 21:46:23,709 - Epoch: [160][ 890/ 1236] Overall Loss 0.134145 Objective Loss 0.134145 LR 0.000250 Time 0.021958 +2023-10-02 21:46:23,920 - Epoch: [160][ 900/ 1236] Overall Loss 0.134208 Objective Loss 0.134208 LR 0.000250 Time 0.021948 +2023-10-02 21:46:24,131 - Epoch: [160][ 910/ 1236] Overall Loss 0.134148 Objective Loss 0.134148 LR 0.000250 Time 0.021938 +2023-10-02 21:46:24,342 - Epoch: [160][ 920/ 1236] Overall Loss 0.134250 Objective Loss 0.134250 LR 0.000250 Time 0.021929 +2023-10-02 21:46:24,553 - Epoch: [160][ 930/ 1236] Overall Loss 0.134331 Objective Loss 0.134331 LR 0.000250 Time 0.021920 +2023-10-02 21:46:24,764 - Epoch: [160][ 940/ 1236] Overall Loss 0.134569 Objective Loss 0.134569 LR 0.000250 Time 0.021911 +2023-10-02 21:46:24,975 - Epoch: [160][ 950/ 1236] Overall Loss 0.134526 Objective Loss 0.134526 LR 0.000250 Time 0.021900 +2023-10-02 21:46:25,186 - Epoch: [160][ 960/ 1236] Overall Loss 0.134392 Objective Loss 0.134392 LR 0.000250 Time 0.021891 +2023-10-02 21:46:25,396 - Epoch: [160][ 970/ 1236] Overall Loss 0.134341 Objective Loss 0.134341 LR 0.000250 Time 0.021882 +2023-10-02 21:46:25,607 - Epoch: [160][ 980/ 1236] Overall Loss 0.134288 Objective Loss 0.134288 LR 0.000250 Time 0.021874 +2023-10-02 21:46:25,818 - Epoch: [160][ 990/ 1236] Overall Loss 0.134181 Objective Loss 0.134181 LR 0.000250 Time 0.021866 +2023-10-02 21:46:26,028 - Epoch: [160][ 1000/ 1236] Overall Loss 0.134176 Objective Loss 0.134176 LR 0.000250 Time 0.021857 +2023-10-02 21:46:26,239 - Epoch: [160][ 1010/ 1236] Overall Loss 0.134214 Objective Loss 0.134214 LR 0.000250 Time 0.021849 +2023-10-02 21:46:26,451 - Epoch: [160][ 1020/ 1236] Overall Loss 0.134212 Objective Loss 0.134212 LR 0.000250 Time 0.021841 +2023-10-02 21:46:26,661 - Epoch: [160][ 1030/ 1236] Overall Loss 0.134231 Objective Loss 0.134231 LR 0.000250 Time 0.021833 +2023-10-02 21:46:26,872 - Epoch: [160][ 1040/ 1236] Overall Loss 0.134076 Objective Loss 0.134076 LR 0.000250 Time 0.021826 +2023-10-02 21:46:27,083 - Epoch: [160][ 1050/ 1236] Overall Loss 0.134291 Objective Loss 0.134291 LR 0.000250 Time 0.021818 +2023-10-02 21:46:27,294 - Epoch: [160][ 1060/ 1236] Overall Loss 0.134360 Objective Loss 0.134360 LR 0.000250 Time 0.021811 +2023-10-02 21:46:27,505 - Epoch: [160][ 1070/ 1236] Overall Loss 0.134203 Objective Loss 0.134203 LR 0.000250 Time 0.021805 +2023-10-02 21:46:27,717 - Epoch: [160][ 1080/ 1236] Overall Loss 0.134300 Objective Loss 0.134300 LR 0.000250 Time 0.021798 +2023-10-02 21:46:27,927 - Epoch: [160][ 1090/ 1236] Overall Loss 0.134391 Objective Loss 0.134391 LR 0.000250 Time 0.021791 +2023-10-02 21:46:28,139 - Epoch: [160][ 1100/ 1236] Overall Loss 0.134140 Objective Loss 0.134140 LR 0.000250 Time 0.021785 +2023-10-02 21:46:28,350 - Epoch: [160][ 1110/ 1236] Overall Loss 0.134327 Objective Loss 0.134327 LR 0.000250 Time 0.021778 +2023-10-02 21:46:28,561 - Epoch: [160][ 1120/ 1236] Overall Loss 0.134267 Objective Loss 0.134267 LR 0.000250 Time 0.021772 +2023-10-02 21:46:28,772 - Epoch: [160][ 1130/ 1236] Overall Loss 0.134331 Objective Loss 0.134331 LR 0.000250 Time 0.021766 +2023-10-02 21:46:28,983 - Epoch: [160][ 1140/ 1236] Overall Loss 0.134350 Objective Loss 0.134350 LR 0.000250 Time 0.021760 +2023-10-02 21:46:29,194 - Epoch: [160][ 1150/ 1236] Overall Loss 0.134370 Objective Loss 0.134370 LR 0.000250 Time 0.021754 +2023-10-02 21:46:29,405 - Epoch: [160][ 1160/ 1236] Overall Loss 0.134306 Objective Loss 0.134306 LR 0.000250 Time 0.021748 +2023-10-02 21:46:29,616 - Epoch: [160][ 1170/ 1236] Overall Loss 0.134292 Objective Loss 0.134292 LR 0.000250 Time 0.021742 +2023-10-02 21:46:29,827 - Epoch: [160][ 1180/ 1236] Overall Loss 0.134402 Objective Loss 0.134402 LR 0.000250 Time 0.021736 +2023-10-02 21:46:30,038 - Epoch: [160][ 1190/ 1236] Overall Loss 0.134434 Objective Loss 0.134434 LR 0.000250 Time 0.021729 +2023-10-02 21:46:30,249 - Epoch: [160][ 1200/ 1236] Overall Loss 0.134528 Objective Loss 0.134528 LR 0.000250 Time 0.021724 +2023-10-02 21:46:30,459 - Epoch: [160][ 1210/ 1236] Overall Loss 0.134531 Objective Loss 0.134531 LR 0.000250 Time 0.021718 +2023-10-02 21:46:30,670 - Epoch: [160][ 1220/ 1236] Overall Loss 0.134721 Objective Loss 0.134721 LR 0.000250 Time 0.021712 +2023-10-02 21:46:30,937 - Epoch: [160][ 1230/ 1236] Overall Loss 0.134692 Objective Loss 0.134692 LR 0.000250 Time 0.021753 +2023-10-02 21:46:31,060 - Epoch: [160][ 1236/ 1236] Overall Loss 0.134602 Objective Loss 0.134602 Top1 91.446029 Top5 98.981670 LR 0.000250 Time 0.021747 +2023-10-02 21:46:31,188 - --- validate (epoch=160)----------- +2023-10-02 21:46:31,188 - 29943 samples (256 per mini-batch) +2023-10-02 21:46:31,702 - Epoch: [160][ 10/ 117] Loss 0.310133 Top1 87.109375 Top5 98.398438 +2023-10-02 21:46:31,856 - Epoch: [160][ 20/ 117] Loss 0.296378 Top1 87.890625 Top5 98.691406 +2023-10-02 21:46:32,007 - Epoch: [160][ 30/ 117] Loss 0.302955 Top1 87.265625 Top5 98.750000 +2023-10-02 21:46:32,161 - Epoch: [160][ 40/ 117] Loss 0.295443 Top1 87.226562 Top5 98.671875 +2023-10-02 21:46:32,312 - Epoch: [160][ 50/ 117] Loss 0.299079 Top1 87.226562 Top5 98.656250 +2023-10-02 21:46:32,468 - Epoch: [160][ 60/ 117] Loss 0.300435 Top1 87.109375 Top5 98.704427 +2023-10-02 21:46:32,618 - Epoch: [160][ 70/ 117] Loss 0.304266 Top1 87.098214 Top5 98.677455 +2023-10-02 21:46:32,772 - Epoch: [160][ 80/ 117] Loss 0.300598 Top1 87.104492 Top5 98.666992 +2023-10-02 21:46:32,925 - Epoch: [160][ 90/ 117] Loss 0.303829 Top1 87.005208 Top5 98.615451 +2023-10-02 21:46:33,081 - Epoch: [160][ 100/ 117] Loss 0.303340 Top1 86.980469 Top5 98.625000 +2023-10-02 21:46:33,241 - Epoch: [160][ 110/ 117] Loss 0.302605 Top1 87.073864 Top5 98.611506 +2023-10-02 21:46:33,331 - Epoch: [160][ 117/ 117] Loss 0.300420 Top1 87.052066 Top5 98.650770 +2023-10-02 21:46:33,476 - ==> Top1: 87.052 Top5: 98.651 Loss: 0.300 + +2023-10-02 21:46:33,477 - ==> Confusion: +[[ 944 1 2 1 5 3 0 0 6 58 1 1 0 0 6 1 4 0 1 0 16] + [ 0 1066 1 1 1 22 0 19 1 1 0 0 0 0 1 3 1 0 7 2 5] + [ 2 0 988 9 0 0 12 5 0 2 2 0 9 3 0 2 1 1 9 3 8] + [ 1 3 13 985 1 1 2 3 3 0 3 0 4 3 26 2 1 5 11 0 22] + [ 25 2 0 0 973 8 1 0 0 7 1 0 0 3 9 6 8 0 0 1 6] + [ 3 40 1 0 0 1003 0 22 1 5 2 6 1 4 5 0 4 1 4 0 14] + [ 0 3 31 0 0 1 1131 3 0 0 2 1 0 0 0 3 0 1 3 7 5] + [ 3 15 10 1 3 22 6 1068 2 4 1 9 4 5 1 0 1 0 43 10 10] + [ 13 4 0 2 2 4 0 2 990 32 7 1 1 9 12 0 3 0 4 1 2] + [ 94 1 2 0 6 6 0 0 30 952 0 1 0 15 5 0 0 0 0 0 7] + [ 2 1 10 11 0 1 2 1 13 2 964 1 2 11 4 0 2 2 7 1 16] + [ 0 0 1 0 0 10 0 5 0 0 0 953 32 8 0 2 1 15 0 4 4] + [ 0 2 1 2 0 1 1 0 1 0 4 22 998 2 3 8 0 9 1 5 8] + [ 0 0 0 0 4 7 0 0 17 6 4 9 0 1044 4 0 0 1 0 3 20] + [ 13 1 4 19 4 1 0 0 20 2 2 0 3 2 1012 0 0 1 10 0 7] + [ 0 0 2 2 4 0 0 0 0 1 1 5 7 0 0 1071 14 11 3 5 8] + [ 0 13 0 0 4 8 0 0 1 0 0 5 0 2 4 7 1100 0 2 4 11] + [ 0 1 1 2 1 0 1 0 0 1 0 3 22 2 2 4 0 992 0 2 4] + [ 1 5 2 18 1 0 0 21 3 1 1 1 2 1 11 0 0 0 987 0 13] + [ 0 0 5 2 1 4 7 2 0 0 0 14 6 3 0 3 9 0 2 1084 10] + [ 113 150 110 65 57 110 28 74 81 59 127 75 369 217 95 44 80 55 112 123 5761]] + +2023-10-02 21:46:33,478 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:46:33,479 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:46:33,485 - + +2023-10-02 21:46:33,485 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:46:34,643 - Epoch: [161][ 10/ 1236] Overall Loss 0.155332 Objective Loss 0.155332 LR 0.000250 Time 0.115696 +2023-10-02 21:46:34,853 - Epoch: [161][ 20/ 1236] Overall Loss 0.138359 Objective Loss 0.138359 LR 0.000250 Time 0.068335 +2023-10-02 21:46:35,062 - Epoch: [161][ 30/ 1236] Overall Loss 0.135112 Objective Loss 0.135112 LR 0.000250 Time 0.052524 +2023-10-02 21:46:35,272 - Epoch: [161][ 40/ 1236] Overall Loss 0.134855 Objective Loss 0.134855 LR 0.000250 Time 0.044642 +2023-10-02 21:46:35,482 - Epoch: [161][ 50/ 1236] Overall Loss 0.137848 Objective Loss 0.137848 LR 0.000250 Time 0.039875 +2023-10-02 21:46:35,693 - Epoch: [161][ 60/ 1236] Overall Loss 0.137611 Objective Loss 0.137611 LR 0.000250 Time 0.036732 +2023-10-02 21:46:35,902 - Epoch: [161][ 70/ 1236] Overall Loss 0.135126 Objective Loss 0.135126 LR 0.000250 Time 0.034453 +2023-10-02 21:46:36,112 - Epoch: [161][ 80/ 1236] Overall Loss 0.134483 Objective Loss 0.134483 LR 0.000250 Time 0.032768 +2023-10-02 21:46:36,322 - Epoch: [161][ 90/ 1236] Overall Loss 0.134699 Objective Loss 0.134699 LR 0.000250 Time 0.031445 +2023-10-02 21:46:36,537 - Epoch: [161][ 100/ 1236] Overall Loss 0.134406 Objective Loss 0.134406 LR 0.000250 Time 0.030433 +2023-10-02 21:46:36,751 - Epoch: [161][ 110/ 1236] Overall Loss 0.135874 Objective Loss 0.135874 LR 0.000250 Time 0.029592 +2023-10-02 21:46:36,964 - Epoch: [161][ 120/ 1236] Overall Loss 0.136105 Objective Loss 0.136105 LR 0.000250 Time 0.028893 +2023-10-02 21:46:37,177 - Epoch: [161][ 130/ 1236] Overall Loss 0.134357 Objective Loss 0.134357 LR 0.000250 Time 0.028295 +2023-10-02 21:46:37,389 - Epoch: [161][ 140/ 1236] Overall Loss 0.135797 Objective Loss 0.135797 LR 0.000250 Time 0.027781 +2023-10-02 21:46:37,596 - Epoch: [161][ 150/ 1236] Overall Loss 0.136107 Objective Loss 0.136107 LR 0.000250 Time 0.027312 +2023-10-02 21:46:37,806 - Epoch: [161][ 160/ 1236] Overall Loss 0.136544 Objective Loss 0.136544 LR 0.000250 Time 0.026913 +2023-10-02 21:46:38,014 - Epoch: [161][ 170/ 1236] Overall Loss 0.136219 Objective Loss 0.136219 LR 0.000250 Time 0.026545 +2023-10-02 21:46:38,224 - Epoch: [161][ 180/ 1236] Overall Loss 0.135768 Objective Loss 0.135768 LR 0.000250 Time 0.026234 +2023-10-02 21:46:38,431 - Epoch: [161][ 190/ 1236] Overall Loss 0.135240 Objective Loss 0.135240 LR 0.000250 Time 0.025937 +2023-10-02 21:46:38,641 - Epoch: [161][ 200/ 1236] Overall Loss 0.135396 Objective Loss 0.135396 LR 0.000250 Time 0.025687 +2023-10-02 21:46:38,849 - Epoch: [161][ 210/ 1236] Overall Loss 0.134663 Objective Loss 0.134663 LR 0.000250 Time 0.025446 +2023-10-02 21:46:39,058 - Epoch: [161][ 220/ 1236] Overall Loss 0.134315 Objective Loss 0.134315 LR 0.000250 Time 0.025240 +2023-10-02 21:46:39,266 - Epoch: [161][ 230/ 1236] Overall Loss 0.134729 Objective Loss 0.134729 LR 0.000250 Time 0.025041 +2023-10-02 21:46:39,476 - Epoch: [161][ 240/ 1236] Overall Loss 0.134819 Objective Loss 0.134819 LR 0.000250 Time 0.024869 +2023-10-02 21:46:39,684 - Epoch: [161][ 250/ 1236] Overall Loss 0.134436 Objective Loss 0.134436 LR 0.000250 Time 0.024701 +2023-10-02 21:46:39,893 - Epoch: [161][ 260/ 1236] Overall Loss 0.134841 Objective Loss 0.134841 LR 0.000250 Time 0.024555 +2023-10-02 21:46:40,102 - Epoch: [161][ 270/ 1236] Overall Loss 0.134450 Objective Loss 0.134450 LR 0.000250 Time 0.024414 +2023-10-02 21:46:40,310 - Epoch: [161][ 280/ 1236] Overall Loss 0.135100 Objective Loss 0.135100 LR 0.000250 Time 0.024284 +2023-10-02 21:46:40,519 - Epoch: [161][ 290/ 1236] Overall Loss 0.135303 Objective Loss 0.135303 LR 0.000250 Time 0.024161 +2023-10-02 21:46:40,731 - Epoch: [161][ 300/ 1236] Overall Loss 0.135603 Objective Loss 0.135603 LR 0.000250 Time 0.024059 +2023-10-02 21:46:40,941 - Epoch: [161][ 310/ 1236] Overall Loss 0.135321 Objective Loss 0.135321 LR 0.000250 Time 0.023960 +2023-10-02 21:46:41,153 - Epoch: [161][ 320/ 1236] Overall Loss 0.134872 Objective Loss 0.134872 LR 0.000250 Time 0.023873 +2023-10-02 21:46:41,365 - Epoch: [161][ 330/ 1236] Overall Loss 0.134707 Objective Loss 0.134707 LR 0.000250 Time 0.023786 +2023-10-02 21:46:41,579 - Epoch: [161][ 340/ 1236] Overall Loss 0.134502 Objective Loss 0.134502 LR 0.000250 Time 0.023716 +2023-10-02 21:46:41,787 - Epoch: [161][ 350/ 1236] Overall Loss 0.134536 Objective Loss 0.134536 LR 0.000250 Time 0.023634 +2023-10-02 21:46:41,998 - Epoch: [161][ 360/ 1236] Overall Loss 0.134424 Objective Loss 0.134424 LR 0.000250 Time 0.023560 +2023-10-02 21:46:42,208 - Epoch: [161][ 370/ 1236] Overall Loss 0.134605 Objective Loss 0.134605 LR 0.000250 Time 0.023487 +2023-10-02 21:46:42,418 - Epoch: [161][ 380/ 1236] Overall Loss 0.134521 Objective Loss 0.134521 LR 0.000250 Time 0.023422 +2023-10-02 21:46:42,629 - Epoch: [161][ 390/ 1236] Overall Loss 0.134240 Objective Loss 0.134240 LR 0.000250 Time 0.023360 +2023-10-02 21:46:42,839 - Epoch: [161][ 400/ 1236] Overall Loss 0.133958 Objective Loss 0.133958 LR 0.000250 Time 0.023301 +2023-10-02 21:46:43,049 - Epoch: [161][ 410/ 1236] Overall Loss 0.133857 Objective Loss 0.133857 LR 0.000250 Time 0.023244 +2023-10-02 21:46:43,260 - Epoch: [161][ 420/ 1236] Overall Loss 0.134381 Objective Loss 0.134381 LR 0.000250 Time 0.023192 +2023-10-02 21:46:43,470 - Epoch: [161][ 430/ 1236] Overall Loss 0.134323 Objective Loss 0.134323 LR 0.000250 Time 0.023138 +2023-10-02 21:46:43,680 - Epoch: [161][ 440/ 1236] Overall Loss 0.134241 Objective Loss 0.134241 LR 0.000250 Time 0.023089 +2023-10-02 21:46:43,891 - Epoch: [161][ 450/ 1236] Overall Loss 0.134068 Objective Loss 0.134068 LR 0.000250 Time 0.023040 +2023-10-02 21:46:44,103 - Epoch: [161][ 460/ 1236] Overall Loss 0.133901 Objective Loss 0.133901 LR 0.000250 Time 0.023000 +2023-10-02 21:46:44,312 - Epoch: [161][ 470/ 1236] Overall Loss 0.133590 Objective Loss 0.133590 LR 0.000250 Time 0.022954 +2023-10-02 21:46:44,522 - Epoch: [161][ 480/ 1236] Overall Loss 0.133522 Objective Loss 0.133522 LR 0.000250 Time 0.022914 +2023-10-02 21:46:44,733 - Epoch: [161][ 490/ 1236] Overall Loss 0.133513 Objective Loss 0.133513 LR 0.000250 Time 0.022872 +2023-10-02 21:46:44,943 - Epoch: [161][ 500/ 1236] Overall Loss 0.133870 Objective Loss 0.133870 LR 0.000250 Time 0.022835 +2023-10-02 21:46:45,154 - Epoch: [161][ 510/ 1236] Overall Loss 0.133630 Objective Loss 0.133630 LR 0.000250 Time 0.022797 +2023-10-02 21:46:45,364 - Epoch: [161][ 520/ 1236] Overall Loss 0.133684 Objective Loss 0.133684 LR 0.000250 Time 0.022763 +2023-10-02 21:46:45,575 - Epoch: [161][ 530/ 1236] Overall Loss 0.133558 Objective Loss 0.133558 LR 0.000250 Time 0.022731 +2023-10-02 21:46:45,786 - Epoch: [161][ 540/ 1236] Overall Loss 0.133367 Objective Loss 0.133367 LR 0.000250 Time 0.022699 +2023-10-02 21:46:45,996 - Epoch: [161][ 550/ 1236] Overall Loss 0.133153 Objective Loss 0.133153 LR 0.000250 Time 0.022666 +2023-10-02 21:46:46,210 - Epoch: [161][ 560/ 1236] Overall Loss 0.133046 Objective Loss 0.133046 LR 0.000250 Time 0.022643 +2023-10-02 21:46:46,421 - Epoch: [161][ 570/ 1236] Overall Loss 0.133092 Objective Loss 0.133092 LR 0.000250 Time 0.022612 +2023-10-02 21:46:46,631 - Epoch: [161][ 580/ 1236] Overall Loss 0.133195 Objective Loss 0.133195 LR 0.000250 Time 0.022585 +2023-10-02 21:46:46,842 - Epoch: [161][ 590/ 1236] Overall Loss 0.133105 Objective Loss 0.133105 LR 0.000250 Time 0.022556 +2023-10-02 21:46:47,052 - Epoch: [161][ 600/ 1236] Overall Loss 0.132981 Objective Loss 0.132981 LR 0.000250 Time 0.022530 +2023-10-02 21:46:47,263 - Epoch: [161][ 610/ 1236] Overall Loss 0.133227 Objective Loss 0.133227 LR 0.000250 Time 0.022503 +2023-10-02 21:46:47,473 - Epoch: [161][ 620/ 1236] Overall Loss 0.133392 Objective Loss 0.133392 LR 0.000250 Time 0.022480 +2023-10-02 21:46:47,684 - Epoch: [161][ 630/ 1236] Overall Loss 0.133285 Objective Loss 0.133285 LR 0.000250 Time 0.022454 +2023-10-02 21:46:47,894 - Epoch: [161][ 640/ 1236] Overall Loss 0.133218 Objective Loss 0.133218 LR 0.000250 Time 0.022432 +2023-10-02 21:46:48,105 - Epoch: [161][ 650/ 1236] Overall Loss 0.133255 Objective Loss 0.133255 LR 0.000250 Time 0.022408 +2023-10-02 21:46:48,315 - Epoch: [161][ 660/ 1236] Overall Loss 0.133142 Objective Loss 0.133142 LR 0.000250 Time 0.022387 +2023-10-02 21:46:48,526 - Epoch: [161][ 670/ 1236] Overall Loss 0.133225 Objective Loss 0.133225 LR 0.000250 Time 0.022365 +2023-10-02 21:46:48,738 - Epoch: [161][ 680/ 1236] Overall Loss 0.133097 Objective Loss 0.133097 LR 0.000250 Time 0.022347 +2023-10-02 21:46:48,947 - Epoch: [161][ 690/ 1236] Overall Loss 0.133034 Objective Loss 0.133034 LR 0.000250 Time 0.022326 +2023-10-02 21:46:49,157 - Epoch: [161][ 700/ 1236] Overall Loss 0.133370 Objective Loss 0.133370 LR 0.000250 Time 0.022307 +2023-10-02 21:46:49,368 - Epoch: [161][ 710/ 1236] Overall Loss 0.133226 Objective Loss 0.133226 LR 0.000250 Time 0.022287 +2023-10-02 21:46:49,579 - Epoch: [161][ 720/ 1236] Overall Loss 0.133169 Objective Loss 0.133169 LR 0.000250 Time 0.022270 +2023-10-02 21:46:49,789 - Epoch: [161][ 730/ 1236] Overall Loss 0.132959 Objective Loss 0.132959 LR 0.000250 Time 0.022253 +2023-10-02 21:46:50,001 - Epoch: [161][ 740/ 1236] Overall Loss 0.132948 Objective Loss 0.132948 LR 0.000250 Time 0.022238 +2023-10-02 21:46:50,210 - Epoch: [161][ 750/ 1236] Overall Loss 0.133034 Objective Loss 0.133034 LR 0.000250 Time 0.022220 +2023-10-02 21:46:50,420 - Epoch: [161][ 760/ 1236] Overall Loss 0.133322 Objective Loss 0.133322 LR 0.000250 Time 0.022204 +2023-10-02 21:46:50,631 - Epoch: [161][ 770/ 1236] Overall Loss 0.133553 Objective Loss 0.133553 LR 0.000250 Time 0.022188 +2023-10-02 21:46:50,841 - Epoch: [161][ 780/ 1236] Overall Loss 0.133546 Objective Loss 0.133546 LR 0.000250 Time 0.022173 +2023-10-02 21:46:51,052 - Epoch: [161][ 790/ 1236] Overall Loss 0.133538 Objective Loss 0.133538 LR 0.000250 Time 0.022157 +2023-10-02 21:46:51,264 - Epoch: [161][ 800/ 1236] Overall Loss 0.133818 Objective Loss 0.133818 LR 0.000250 Time 0.022145 +2023-10-02 21:46:51,473 - Epoch: [161][ 810/ 1236] Overall Loss 0.133665 Objective Loss 0.133665 LR 0.000250 Time 0.022129 +2023-10-02 21:46:51,683 - Epoch: [161][ 820/ 1236] Overall Loss 0.133683 Objective Loss 0.133683 LR 0.000250 Time 0.022115 +2023-10-02 21:46:51,894 - Epoch: [161][ 830/ 1236] Overall Loss 0.133879 Objective Loss 0.133879 LR 0.000250 Time 0.022101 +2023-10-02 21:46:52,104 - Epoch: [161][ 840/ 1236] Overall Loss 0.133847 Objective Loss 0.133847 LR 0.000250 Time 0.022088 +2023-10-02 21:46:52,315 - Epoch: [161][ 850/ 1236] Overall Loss 0.133967 Objective Loss 0.133967 LR 0.000250 Time 0.022074 +2023-10-02 21:46:52,525 - Epoch: [161][ 860/ 1236] Overall Loss 0.134043 Objective Loss 0.134043 LR 0.000250 Time 0.022061 +2023-10-02 21:46:52,735 - Epoch: [161][ 870/ 1236] Overall Loss 0.133899 Objective Loss 0.133899 LR 0.000250 Time 0.022047 +2023-10-02 21:46:52,946 - Epoch: [161][ 880/ 1236] Overall Loss 0.133792 Objective Loss 0.133792 LR 0.000250 Time 0.022035 +2023-10-02 21:46:53,156 - Epoch: [161][ 890/ 1236] Overall Loss 0.133838 Objective Loss 0.133838 LR 0.000250 Time 0.022023 +2023-10-02 21:46:53,367 - Epoch: [161][ 900/ 1236] Overall Loss 0.133909 Objective Loss 0.133909 LR 0.000250 Time 0.022012 +2023-10-02 21:46:53,577 - Epoch: [161][ 910/ 1236] Overall Loss 0.134014 Objective Loss 0.134014 LR 0.000250 Time 0.021999 +2023-10-02 21:46:53,787 - Epoch: [161][ 920/ 1236] Overall Loss 0.133990 Objective Loss 0.133990 LR 0.000250 Time 0.021988 +2023-10-02 21:46:53,997 - Epoch: [161][ 930/ 1236] Overall Loss 0.133825 Objective Loss 0.133825 LR 0.000250 Time 0.021977 +2023-10-02 21:46:54,207 - Epoch: [161][ 940/ 1236] Overall Loss 0.133710 Objective Loss 0.133710 LR 0.000250 Time 0.021966 +2023-10-02 21:46:54,418 - Epoch: [161][ 950/ 1236] Overall Loss 0.133648 Objective Loss 0.133648 LR 0.000250 Time 0.021956 +2023-10-02 21:46:54,628 - Epoch: [161][ 960/ 1236] Overall Loss 0.133583 Objective Loss 0.133583 LR 0.000250 Time 0.021946 +2023-10-02 21:46:54,838 - Epoch: [161][ 970/ 1236] Overall Loss 0.133575 Objective Loss 0.133575 LR 0.000250 Time 0.021936 +2023-10-02 21:46:55,049 - Epoch: [161][ 980/ 1236] Overall Loss 0.133461 Objective Loss 0.133461 LR 0.000250 Time 0.021927 +2023-10-02 21:46:55,259 - Epoch: [161][ 990/ 1236] Overall Loss 0.133422 Objective Loss 0.133422 LR 0.000250 Time 0.021917 +2023-10-02 21:46:55,469 - Epoch: [161][ 1000/ 1236] Overall Loss 0.133371 Objective Loss 0.133371 LR 0.000250 Time 0.021907 +2023-10-02 21:46:55,679 - Epoch: [161][ 1010/ 1236] Overall Loss 0.133521 Objective Loss 0.133521 LR 0.000250 Time 0.021898 +2023-10-02 21:46:55,889 - Epoch: [161][ 1020/ 1236] Overall Loss 0.133611 Objective Loss 0.133611 LR 0.000250 Time 0.021889 +2023-10-02 21:46:56,100 - Epoch: [161][ 1030/ 1236] Overall Loss 0.133563 Objective Loss 0.133563 LR 0.000250 Time 0.021881 +2023-10-02 21:46:56,310 - Epoch: [161][ 1040/ 1236] Overall Loss 0.133426 Objective Loss 0.133426 LR 0.000250 Time 0.021872 +2023-10-02 21:46:56,521 - Epoch: [161][ 1050/ 1236] Overall Loss 0.133718 Objective Loss 0.133718 LR 0.000250 Time 0.021864 +2023-10-02 21:46:56,731 - Epoch: [161][ 1060/ 1236] Overall Loss 0.133901 Objective Loss 0.133901 LR 0.000250 Time 0.021856 +2023-10-02 21:46:56,942 - Epoch: [161][ 1070/ 1236] Overall Loss 0.133912 Objective Loss 0.133912 LR 0.000250 Time 0.021848 +2023-10-02 21:46:57,152 - Epoch: [161][ 1080/ 1236] Overall Loss 0.133860 Objective Loss 0.133860 LR 0.000250 Time 0.021840 +2023-10-02 21:46:57,362 - Epoch: [161][ 1090/ 1236] Overall Loss 0.133943 Objective Loss 0.133943 LR 0.000250 Time 0.021831 +2023-10-02 21:46:57,572 - Epoch: [161][ 1100/ 1236] Overall Loss 0.134050 Objective Loss 0.134050 LR 0.000250 Time 0.021823 +2023-10-02 21:46:57,783 - Epoch: [161][ 1110/ 1236] Overall Loss 0.133986 Objective Loss 0.133986 LR 0.000250 Time 0.021815 +2023-10-02 21:46:57,993 - Epoch: [161][ 1120/ 1236] Overall Loss 0.133876 Objective Loss 0.133876 LR 0.000250 Time 0.021807 +2023-10-02 21:46:58,203 - Epoch: [161][ 1130/ 1236] Overall Loss 0.133977 Objective Loss 0.133977 LR 0.000250 Time 0.021800 +2023-10-02 21:46:58,414 - Epoch: [161][ 1140/ 1236] Overall Loss 0.133974 Objective Loss 0.133974 LR 0.000250 Time 0.021793 +2023-10-02 21:46:58,624 - Epoch: [161][ 1150/ 1236] Overall Loss 0.134041 Objective Loss 0.134041 LR 0.000250 Time 0.021786 +2023-10-02 21:46:58,834 - Epoch: [161][ 1160/ 1236] Overall Loss 0.133873 Objective Loss 0.133873 LR 0.000250 Time 0.021780 +2023-10-02 21:46:59,045 - Epoch: [161][ 1170/ 1236] Overall Loss 0.133810 Objective Loss 0.133810 LR 0.000250 Time 0.021773 +2023-10-02 21:46:59,255 - Epoch: [161][ 1180/ 1236] Overall Loss 0.133945 Objective Loss 0.133945 LR 0.000250 Time 0.021766 +2023-10-02 21:46:59,465 - Epoch: [161][ 1190/ 1236] Overall Loss 0.133956 Objective Loss 0.133956 LR 0.000250 Time 0.021759 +2023-10-02 21:46:59,675 - Epoch: [161][ 1200/ 1236] Overall Loss 0.133979 Objective Loss 0.133979 LR 0.000250 Time 0.021753 +2023-10-02 21:46:59,885 - Epoch: [161][ 1210/ 1236] Overall Loss 0.134025 Objective Loss 0.134025 LR 0.000250 Time 0.021746 +2023-10-02 21:47:00,095 - Epoch: [161][ 1220/ 1236] Overall Loss 0.133889 Objective Loss 0.133889 LR 0.000250 Time 0.021740 +2023-10-02 21:47:00,359 - Epoch: [161][ 1230/ 1236] Overall Loss 0.133914 Objective Loss 0.133914 LR 0.000250 Time 0.021778 +2023-10-02 21:47:00,482 - Epoch: [161][ 1236/ 1236] Overall Loss 0.134022 Objective Loss 0.134022 Top1 91.242363 Top5 99.389002 LR 0.000250 Time 0.021771 +2023-10-02 21:47:00,630 - --- validate (epoch=161)----------- +2023-10-02 21:47:00,631 - 29943 samples (256 per mini-batch) +2023-10-02 21:47:01,128 - Epoch: [161][ 10/ 117] Loss 0.285693 Top1 87.773438 Top5 98.828125 +2023-10-02 21:47:01,281 - Epoch: [161][ 20/ 117] Loss 0.291824 Top1 87.207031 Top5 98.496094 +2023-10-02 21:47:01,431 - Epoch: [161][ 30/ 117] Loss 0.297089 Top1 87.018229 Top5 98.502604 +2023-10-02 21:47:01,584 - Epoch: [161][ 40/ 117] Loss 0.291206 Top1 87.060547 Top5 98.564453 +2023-10-02 21:47:01,734 - Epoch: [161][ 50/ 117] Loss 0.300632 Top1 86.773438 Top5 98.539062 +2023-10-02 21:47:01,888 - Epoch: [161][ 60/ 117] Loss 0.303426 Top1 86.692708 Top5 98.535156 +2023-10-02 21:47:02,039 - Epoch: [161][ 70/ 117] Loss 0.306259 Top1 86.685268 Top5 98.604911 +2023-10-02 21:47:02,192 - Epoch: [161][ 80/ 117] Loss 0.309277 Top1 86.669922 Top5 98.579102 +2023-10-02 21:47:02,342 - Epoch: [161][ 90/ 117] Loss 0.306538 Top1 86.731771 Top5 98.628472 +2023-10-02 21:47:02,496 - Epoch: [161][ 100/ 117] Loss 0.304454 Top1 86.753906 Top5 98.613281 +2023-10-02 21:47:02,653 - Epoch: [161][ 110/ 117] Loss 0.302310 Top1 86.825284 Top5 98.636364 +2023-10-02 21:47:02,743 - Epoch: [161][ 117/ 117] Loss 0.301935 Top1 86.821628 Top5 98.640751 +2023-10-02 21:47:02,883 - ==> Top1: 86.822 Top5: 98.641 Loss: 0.302 + +2023-10-02 21:47:02,884 - ==> Confusion: +[[ 952 0 2 0 8 2 0 0 1 56 1 0 1 1 2 1 3 1 1 0 18] + [ 0 1065 2 0 6 16 3 18 0 1 0 0 0 0 1 3 3 0 8 2 3] + [ 3 1 980 6 2 0 16 8 0 3 0 1 7 2 1 2 0 1 13 5 5] + [ 2 4 14 978 1 0 3 4 4 0 3 0 4 4 24 2 1 4 17 1 19] + [ 22 3 1 0 977 6 1 0 0 13 1 0 0 2 7 3 9 0 1 3 1] + [ 6 37 0 0 1 980 2 27 2 4 2 8 1 11 5 0 6 1 4 1 18] + [ 0 3 27 1 0 1 1128 8 0 0 2 1 0 1 0 4 0 1 2 8 4] + [ 2 11 7 2 4 18 5 1090 3 1 4 5 3 3 1 0 0 1 41 11 6] + [ 16 2 0 1 3 3 0 3 973 40 8 1 1 10 16 0 3 1 3 3 2] + [ 95 0 1 0 6 5 0 0 25 958 0 1 0 15 4 1 1 0 0 1 6] + [ 2 5 11 7 0 0 3 2 13 0 973 1 0 9 2 0 3 3 6 0 13] + [ 0 1 1 0 1 10 0 5 0 0 0 966 19 4 0 2 1 15 0 7 3] + [ 0 2 1 6 0 0 0 1 0 2 3 32 981 0 1 7 4 8 0 7 13] + [ 1 0 1 0 3 6 0 0 7 12 4 8 0 1055 3 2 0 0 0 2 15] + [ 14 0 5 15 4 0 0 0 16 3 2 0 2 2 1016 0 0 1 10 0 11] + [ 0 0 2 1 5 0 1 0 0 0 2 5 6 0 0 1073 16 10 2 7 4] + [ 0 15 0 0 7 6 0 0 0 0 0 4 0 1 4 7 1100 0 2 6 9] + [ 0 0 0 1 1 0 1 0 0 1 0 4 23 2 1 6 2 991 0 2 3] + [ 3 3 1 18 0 0 0 19 2 2 2 0 0 0 7 0 1 0 1001 0 9] + [ 0 0 3 3 1 2 6 6 0 1 2 9 5 2 0 2 8 0 0 1093 9] + [ 123 155 105 71 73 86 25 89 82 78 138 73 317 242 125 49 100 46 126 135 5667]] + +2023-10-02 21:47:02,885 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:47:02,885 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:47:02,892 - + +2023-10-02 21:47:02,892 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:47:03,944 - Epoch: [162][ 10/ 1236] Overall Loss 0.118399 Objective Loss 0.118399 LR 0.000250 Time 0.105234 +2023-10-02 21:47:04,154 - Epoch: [162][ 20/ 1236] Overall Loss 0.121443 Objective Loss 0.121443 LR 0.000250 Time 0.063060 +2023-10-02 21:47:04,363 - Epoch: [162][ 30/ 1236] Overall Loss 0.117802 Objective Loss 0.117802 LR 0.000250 Time 0.049000 +2023-10-02 21:47:04,572 - Epoch: [162][ 40/ 1236] Overall Loss 0.120482 Objective Loss 0.120482 LR 0.000250 Time 0.041958 +2023-10-02 21:47:04,780 - Epoch: [162][ 50/ 1236] Overall Loss 0.120656 Objective Loss 0.120656 LR 0.000250 Time 0.037707 +2023-10-02 21:47:04,990 - Epoch: [162][ 60/ 1236] Overall Loss 0.124966 Objective Loss 0.124966 LR 0.000250 Time 0.034911 +2023-10-02 21:47:05,198 - Epoch: [162][ 70/ 1236] Overall Loss 0.124140 Objective Loss 0.124140 LR 0.000250 Time 0.032882 +2023-10-02 21:47:05,408 - Epoch: [162][ 80/ 1236] Overall Loss 0.124951 Objective Loss 0.124951 LR 0.000250 Time 0.031386 +2023-10-02 21:47:05,616 - Epoch: [162][ 90/ 1236] Overall Loss 0.126872 Objective Loss 0.126872 LR 0.000250 Time 0.030193 +2023-10-02 21:47:05,824 - Epoch: [162][ 100/ 1236] Overall Loss 0.125930 Objective Loss 0.125930 LR 0.000250 Time 0.029258 +2023-10-02 21:47:06,032 - Epoch: [162][ 110/ 1236] Overall Loss 0.126897 Objective Loss 0.126897 LR 0.000250 Time 0.028476 +2023-10-02 21:47:06,241 - Epoch: [162][ 120/ 1236] Overall Loss 0.126854 Objective Loss 0.126854 LR 0.000250 Time 0.027841 +2023-10-02 21:47:06,449 - Epoch: [162][ 130/ 1236] Overall Loss 0.126518 Objective Loss 0.126518 LR 0.000250 Time 0.027299 +2023-10-02 21:47:06,659 - Epoch: [162][ 140/ 1236] Overall Loss 0.128914 Objective Loss 0.128914 LR 0.000250 Time 0.026840 +2023-10-02 21:47:06,867 - Epoch: [162][ 150/ 1236] Overall Loss 0.128603 Objective Loss 0.128603 LR 0.000250 Time 0.026428 +2023-10-02 21:47:07,075 - Epoch: [162][ 160/ 1236] Overall Loss 0.129120 Objective Loss 0.129120 LR 0.000250 Time 0.026075 +2023-10-02 21:47:07,283 - Epoch: [162][ 170/ 1236] Overall Loss 0.129042 Objective Loss 0.129042 LR 0.000250 Time 0.025763 +2023-10-02 21:47:07,491 - Epoch: [162][ 180/ 1236] Overall Loss 0.130305 Objective Loss 0.130305 LR 0.000250 Time 0.025485 +2023-10-02 21:47:07,698 - Epoch: [162][ 190/ 1236] Overall Loss 0.130157 Objective Loss 0.130157 LR 0.000250 Time 0.025232 +2023-10-02 21:47:07,907 - Epoch: [162][ 200/ 1236] Overall Loss 0.130775 Objective Loss 0.130775 LR 0.000250 Time 0.025013 +2023-10-02 21:47:08,114 - Epoch: [162][ 210/ 1236] Overall Loss 0.130145 Objective Loss 0.130145 LR 0.000250 Time 0.024802 +2023-10-02 21:47:08,323 - Epoch: [162][ 220/ 1236] Overall Loss 0.130138 Objective Loss 0.130138 LR 0.000250 Time 0.024622 +2023-10-02 21:47:08,530 - Epoch: [162][ 230/ 1236] Overall Loss 0.129596 Objective Loss 0.129596 LR 0.000250 Time 0.024453 +2023-10-02 21:47:08,736 - Epoch: [162][ 240/ 1236] Overall Loss 0.129474 Objective Loss 0.129474 LR 0.000250 Time 0.024292 +2023-10-02 21:47:08,944 - Epoch: [162][ 250/ 1236] Overall Loss 0.129374 Objective Loss 0.129374 LR 0.000250 Time 0.024150 +2023-10-02 21:47:09,151 - Epoch: [162][ 260/ 1236] Overall Loss 0.129449 Objective Loss 0.129449 LR 0.000250 Time 0.024015 +2023-10-02 21:47:09,358 - Epoch: [162][ 270/ 1236] Overall Loss 0.129566 Objective Loss 0.129566 LR 0.000250 Time 0.023893 +2023-10-02 21:47:09,565 - Epoch: [162][ 280/ 1236] Overall Loss 0.129855 Objective Loss 0.129855 LR 0.000250 Time 0.023776 +2023-10-02 21:47:09,772 - Epoch: [162][ 290/ 1236] Overall Loss 0.130795 Objective Loss 0.130795 LR 0.000250 Time 0.023671 +2023-10-02 21:47:09,979 - Epoch: [162][ 300/ 1236] Overall Loss 0.130515 Objective Loss 0.130515 LR 0.000250 Time 0.023569 +2023-10-02 21:47:10,189 - Epoch: [162][ 310/ 1236] Overall Loss 0.130547 Objective Loss 0.130547 LR 0.000250 Time 0.023486 +2023-10-02 21:47:10,401 - Epoch: [162][ 320/ 1236] Overall Loss 0.131447 Objective Loss 0.131447 LR 0.000250 Time 0.023412 +2023-10-02 21:47:10,613 - Epoch: [162][ 330/ 1236] Overall Loss 0.132013 Objective Loss 0.132013 LR 0.000250 Time 0.023343 +2023-10-02 21:47:10,824 - Epoch: [162][ 340/ 1236] Overall Loss 0.131805 Objective Loss 0.131805 LR 0.000250 Time 0.023273 +2023-10-02 21:47:11,035 - Epoch: [162][ 350/ 1236] Overall Loss 0.131744 Objective Loss 0.131744 LR 0.000250 Time 0.023210 +2023-10-02 21:47:11,245 - Epoch: [162][ 360/ 1236] Overall Loss 0.132062 Objective Loss 0.132062 LR 0.000250 Time 0.023148 +2023-10-02 21:47:11,456 - Epoch: [162][ 370/ 1236] Overall Loss 0.131784 Objective Loss 0.131784 LR 0.000250 Time 0.023091 +2023-10-02 21:47:11,667 - Epoch: [162][ 380/ 1236] Overall Loss 0.132284 Objective Loss 0.132284 LR 0.000250 Time 0.023038 +2023-10-02 21:47:11,879 - Epoch: [162][ 390/ 1236] Overall Loss 0.132159 Objective Loss 0.132159 LR 0.000250 Time 0.022988 +2023-10-02 21:47:12,085 - Epoch: [162][ 400/ 1236] Overall Loss 0.132482 Objective Loss 0.132482 LR 0.000250 Time 0.022927 +2023-10-02 21:47:12,291 - Epoch: [162][ 410/ 1236] Overall Loss 0.132246 Objective Loss 0.132246 LR 0.000250 Time 0.022871 +2023-10-02 21:47:12,497 - Epoch: [162][ 420/ 1236] Overall Loss 0.132540 Objective Loss 0.132540 LR 0.000250 Time 0.022814 +2023-10-02 21:47:12,704 - Epoch: [162][ 430/ 1236] Overall Loss 0.132957 Objective Loss 0.132957 LR 0.000250 Time 0.022762 +2023-10-02 21:47:12,910 - Epoch: [162][ 440/ 1236] Overall Loss 0.133051 Objective Loss 0.133051 LR 0.000250 Time 0.022710 +2023-10-02 21:47:13,116 - Epoch: [162][ 450/ 1236] Overall Loss 0.133201 Objective Loss 0.133201 LR 0.000250 Time 0.022664 +2023-10-02 21:47:13,322 - Epoch: [162][ 460/ 1236] Overall Loss 0.132995 Objective Loss 0.132995 LR 0.000250 Time 0.022616 +2023-10-02 21:47:13,528 - Epoch: [162][ 470/ 1236] Overall Loss 0.133200 Objective Loss 0.133200 LR 0.000250 Time 0.022573 +2023-10-02 21:47:13,734 - Epoch: [162][ 480/ 1236] Overall Loss 0.133014 Objective Loss 0.133014 LR 0.000250 Time 0.022528 +2023-10-02 21:47:13,941 - Epoch: [162][ 490/ 1236] Overall Loss 0.132664 Objective Loss 0.132664 LR 0.000250 Time 0.022490 +2023-10-02 21:47:14,147 - Epoch: [162][ 500/ 1236] Overall Loss 0.132426 Objective Loss 0.132426 LR 0.000250 Time 0.022449 +2023-10-02 21:47:14,354 - Epoch: [162][ 510/ 1236] Overall Loss 0.132406 Objective Loss 0.132406 LR 0.000250 Time 0.022413 +2023-10-02 21:47:14,560 - Epoch: [162][ 520/ 1236] Overall Loss 0.132080 Objective Loss 0.132080 LR 0.000250 Time 0.022375 +2023-10-02 21:47:14,766 - Epoch: [162][ 530/ 1236] Overall Loss 0.131835 Objective Loss 0.131835 LR 0.000250 Time 0.022343 +2023-10-02 21:47:14,973 - Epoch: [162][ 540/ 1236] Overall Loss 0.132293 Objective Loss 0.132293 LR 0.000250 Time 0.022308 +2023-10-02 21:47:15,179 - Epoch: [162][ 550/ 1236] Overall Loss 0.132105 Objective Loss 0.132105 LR 0.000250 Time 0.022278 +2023-10-02 21:47:15,385 - Epoch: [162][ 560/ 1236] Overall Loss 0.132127 Objective Loss 0.132127 LR 0.000250 Time 0.022245 +2023-10-02 21:47:15,592 - Epoch: [162][ 570/ 1236] Overall Loss 0.132311 Objective Loss 0.132311 LR 0.000250 Time 0.022217 +2023-10-02 21:47:15,798 - Epoch: [162][ 580/ 1236] Overall Loss 0.132337 Objective Loss 0.132337 LR 0.000250 Time 0.022187 +2023-10-02 21:47:16,004 - Epoch: [162][ 590/ 1236] Overall Loss 0.132017 Objective Loss 0.132017 LR 0.000250 Time 0.022160 +2023-10-02 21:47:16,211 - Epoch: [162][ 600/ 1236] Overall Loss 0.132153 Objective Loss 0.132153 LR 0.000250 Time 0.022132 +2023-10-02 21:47:16,417 - Epoch: [162][ 610/ 1236] Overall Loss 0.132233 Objective Loss 0.132233 LR 0.000250 Time 0.022107 +2023-10-02 21:47:16,623 - Epoch: [162][ 620/ 1236] Overall Loss 0.132266 Objective Loss 0.132266 LR 0.000250 Time 0.022080 +2023-10-02 21:47:16,830 - Epoch: [162][ 630/ 1236] Overall Loss 0.132184 Objective Loss 0.132184 LR 0.000250 Time 0.022057 +2023-10-02 21:47:17,036 - Epoch: [162][ 640/ 1236] Overall Loss 0.132151 Objective Loss 0.132151 LR 0.000250 Time 0.022032 +2023-10-02 21:47:17,242 - Epoch: [162][ 650/ 1236] Overall Loss 0.132542 Objective Loss 0.132542 LR 0.000250 Time 0.022010 +2023-10-02 21:47:17,448 - Epoch: [162][ 660/ 1236] Overall Loss 0.132352 Objective Loss 0.132352 LR 0.000250 Time 0.021987 +2023-10-02 21:47:17,655 - Epoch: [162][ 670/ 1236] Overall Loss 0.132209 Objective Loss 0.132209 LR 0.000250 Time 0.021966 +2023-10-02 21:47:17,861 - Epoch: [162][ 680/ 1236] Overall Loss 0.132115 Objective Loss 0.132115 LR 0.000250 Time 0.021944 +2023-10-02 21:47:18,067 - Epoch: [162][ 690/ 1236] Overall Loss 0.132144 Objective Loss 0.132144 LR 0.000250 Time 0.021925 +2023-10-02 21:47:18,273 - Epoch: [162][ 700/ 1236] Overall Loss 0.132456 Objective Loss 0.132456 LR 0.000250 Time 0.021904 +2023-10-02 21:47:18,480 - Epoch: [162][ 710/ 1236] Overall Loss 0.132624 Objective Loss 0.132624 LR 0.000250 Time 0.021885 +2023-10-02 21:47:18,686 - Epoch: [162][ 720/ 1236] Overall Loss 0.132544 Objective Loss 0.132544 LR 0.000250 Time 0.021865 +2023-10-02 21:47:18,892 - Epoch: [162][ 730/ 1236] Overall Loss 0.132486 Objective Loss 0.132486 LR 0.000250 Time 0.021848 +2023-10-02 21:47:19,098 - Epoch: [162][ 740/ 1236] Overall Loss 0.132419 Objective Loss 0.132419 LR 0.000250 Time 0.021829 +2023-10-02 21:47:19,305 - Epoch: [162][ 750/ 1236] Overall Loss 0.132385 Objective Loss 0.132385 LR 0.000250 Time 0.021813 +2023-10-02 21:47:19,511 - Epoch: [162][ 760/ 1236] Overall Loss 0.132187 Objective Loss 0.132187 LR 0.000250 Time 0.021795 +2023-10-02 21:47:19,717 - Epoch: [162][ 770/ 1236] Overall Loss 0.131978 Objective Loss 0.131978 LR 0.000250 Time 0.021780 +2023-10-02 21:47:19,923 - Epoch: [162][ 780/ 1236] Overall Loss 0.132389 Objective Loss 0.132389 LR 0.000250 Time 0.021763 +2023-10-02 21:47:20,130 - Epoch: [162][ 790/ 1236] Overall Loss 0.132381 Objective Loss 0.132381 LR 0.000250 Time 0.021748 +2023-10-02 21:47:20,336 - Epoch: [162][ 800/ 1236] Overall Loss 0.132537 Objective Loss 0.132537 LR 0.000250 Time 0.021732 +2023-10-02 21:47:20,542 - Epoch: [162][ 810/ 1236] Overall Loss 0.132440 Objective Loss 0.132440 LR 0.000250 Time 0.021719 +2023-10-02 21:47:20,748 - Epoch: [162][ 820/ 1236] Overall Loss 0.132402 Objective Loss 0.132402 LR 0.000250 Time 0.021703 +2023-10-02 21:47:20,955 - Epoch: [162][ 830/ 1236] Overall Loss 0.132287 Objective Loss 0.132287 LR 0.000250 Time 0.021690 +2023-10-02 21:47:21,161 - Epoch: [162][ 840/ 1236] Overall Loss 0.132270 Objective Loss 0.132270 LR 0.000250 Time 0.021675 +2023-10-02 21:47:21,367 - Epoch: [162][ 850/ 1236] Overall Loss 0.132197 Objective Loss 0.132197 LR 0.000250 Time 0.021663 +2023-10-02 21:47:21,573 - Epoch: [162][ 860/ 1236] Overall Loss 0.132024 Objective Loss 0.132024 LR 0.000250 Time 0.021649 +2023-10-02 21:47:21,780 - Epoch: [162][ 870/ 1236] Overall Loss 0.131729 Objective Loss 0.131729 LR 0.000250 Time 0.021638 +2023-10-02 21:47:21,986 - Epoch: [162][ 880/ 1236] Overall Loss 0.131675 Objective Loss 0.131675 LR 0.000250 Time 0.021624 +2023-10-02 21:47:22,193 - Epoch: [162][ 890/ 1236] Overall Loss 0.131582 Objective Loss 0.131582 LR 0.000250 Time 0.021613 +2023-10-02 21:47:22,399 - Epoch: [162][ 900/ 1236] Overall Loss 0.131581 Objective Loss 0.131581 LR 0.000250 Time 0.021600 +2023-10-02 21:47:22,606 - Epoch: [162][ 910/ 1236] Overall Loss 0.131916 Objective Loss 0.131916 LR 0.000250 Time 0.021591 +2023-10-02 21:47:22,811 - Epoch: [162][ 920/ 1236] Overall Loss 0.131981 Objective Loss 0.131981 LR 0.000250 Time 0.021578 +2023-10-02 21:47:23,018 - Epoch: [162][ 930/ 1236] Overall Loss 0.132114 Objective Loss 0.132114 LR 0.000250 Time 0.021568 +2023-10-02 21:47:23,224 - Epoch: [162][ 940/ 1236] Overall Loss 0.132199 Objective Loss 0.132199 LR 0.000250 Time 0.021556 +2023-10-02 21:47:23,432 - Epoch: [162][ 950/ 1236] Overall Loss 0.132254 Objective Loss 0.132254 LR 0.000250 Time 0.021548 +2023-10-02 21:47:23,637 - Epoch: [162][ 960/ 1236] Overall Loss 0.132296 Objective Loss 0.132296 LR 0.000250 Time 0.021536 +2023-10-02 21:47:23,844 - Epoch: [162][ 970/ 1236] Overall Loss 0.132146 Objective Loss 0.132146 LR 0.000250 Time 0.021527 +2023-10-02 21:47:24,050 - Epoch: [162][ 980/ 1236] Overall Loss 0.132385 Objective Loss 0.132385 LR 0.000250 Time 0.021517 +2023-10-02 21:47:24,256 - Epoch: [162][ 990/ 1236] Overall Loss 0.132703 Objective Loss 0.132703 LR 0.000250 Time 0.021508 +2023-10-02 21:47:24,462 - Epoch: [162][ 1000/ 1236] Overall Loss 0.132850 Objective Loss 0.132850 LR 0.000250 Time 0.021497 +2023-10-02 21:47:24,670 - Epoch: [162][ 1010/ 1236] Overall Loss 0.132849 Objective Loss 0.132849 LR 0.000250 Time 0.021490 +2023-10-02 21:47:24,875 - Epoch: [162][ 1020/ 1236] Overall Loss 0.132915 Objective Loss 0.132915 LR 0.000250 Time 0.021480 +2023-10-02 21:47:25,083 - Epoch: [162][ 1030/ 1236] Overall Loss 0.133074 Objective Loss 0.133074 LR 0.000250 Time 0.021473 +2023-10-02 21:47:25,288 - Epoch: [162][ 1040/ 1236] Overall Loss 0.133060 Objective Loss 0.133060 LR 0.000250 Time 0.021463 +2023-10-02 21:47:25,496 - Epoch: [162][ 1050/ 1236] Overall Loss 0.133023 Objective Loss 0.133023 LR 0.000250 Time 0.021456 +2023-10-02 21:47:25,701 - Epoch: [162][ 1060/ 1236] Overall Loss 0.133023 Objective Loss 0.133023 LR 0.000250 Time 0.021447 +2023-10-02 21:47:25,908 - Epoch: [162][ 1070/ 1236] Overall Loss 0.133207 Objective Loss 0.133207 LR 0.000250 Time 0.021440 +2023-10-02 21:47:26,114 - Epoch: [162][ 1080/ 1236] Overall Loss 0.133475 Objective Loss 0.133475 LR 0.000250 Time 0.021432 +2023-10-02 21:47:26,320 - Epoch: [162][ 1090/ 1236] Overall Loss 0.133454 Objective Loss 0.133454 LR 0.000250 Time 0.021424 +2023-10-02 21:47:26,526 - Epoch: [162][ 1100/ 1236] Overall Loss 0.133453 Objective Loss 0.133453 LR 0.000250 Time 0.021416 +2023-10-02 21:47:26,733 - Epoch: [162][ 1110/ 1236] Overall Loss 0.133399 Objective Loss 0.133399 LR 0.000250 Time 0.021409 +2023-10-02 21:47:26,939 - Epoch: [162][ 1120/ 1236] Overall Loss 0.133350 Objective Loss 0.133350 LR 0.000250 Time 0.021400 +2023-10-02 21:47:27,146 - Epoch: [162][ 1130/ 1236] Overall Loss 0.133353 Objective Loss 0.133353 LR 0.000250 Time 0.021393 +2023-10-02 21:47:27,352 - Epoch: [162][ 1140/ 1236] Overall Loss 0.133465 Objective Loss 0.133465 LR 0.000250 Time 0.021385 +2023-10-02 21:47:27,559 - Epoch: [162][ 1150/ 1236] Overall Loss 0.133424 Objective Loss 0.133424 LR 0.000250 Time 0.021379 +2023-10-02 21:47:27,765 - Epoch: [162][ 1160/ 1236] Overall Loss 0.133434 Objective Loss 0.133434 LR 0.000250 Time 0.021371 +2023-10-02 21:47:27,972 - Epoch: [162][ 1170/ 1236] Overall Loss 0.133447 Objective Loss 0.133447 LR 0.000250 Time 0.021365 +2023-10-02 21:47:28,178 - Epoch: [162][ 1180/ 1236] Overall Loss 0.133569 Objective Loss 0.133569 LR 0.000250 Time 0.021357 +2023-10-02 21:47:28,386 - Epoch: [162][ 1190/ 1236] Overall Loss 0.133667 Objective Loss 0.133667 LR 0.000250 Time 0.021352 +2023-10-02 21:47:28,590 - Epoch: [162][ 1200/ 1236] Overall Loss 0.133499 Objective Loss 0.133499 LR 0.000250 Time 0.021344 +2023-10-02 21:47:28,798 - Epoch: [162][ 1210/ 1236] Overall Loss 0.133603 Objective Loss 0.133603 LR 0.000250 Time 0.021339 +2023-10-02 21:47:29,003 - Epoch: [162][ 1220/ 1236] Overall Loss 0.133758 Objective Loss 0.133758 LR 0.000250 Time 0.021332 +2023-10-02 21:47:29,260 - Epoch: [162][ 1230/ 1236] Overall Loss 0.133683 Objective Loss 0.133683 LR 0.000250 Time 0.021367 +2023-10-02 21:47:29,381 - Epoch: [162][ 1236/ 1236] Overall Loss 0.133650 Objective Loss 0.133650 Top1 91.242363 Top5 99.389002 LR 0.000250 Time 0.021362 +2023-10-02 21:47:29,514 - --- validate (epoch=162)----------- +2023-10-02 21:47:29,514 - 29943 samples (256 per mini-batch) +2023-10-02 21:47:30,014 - Epoch: [162][ 10/ 117] Loss 0.264840 Top1 87.851562 Top5 99.062500 +2023-10-02 21:47:30,165 - Epoch: [162][ 20/ 117] Loss 0.271886 Top1 87.890625 Top5 99.003906 +2023-10-02 21:47:30,315 - Epoch: [162][ 30/ 117] Loss 0.274168 Top1 87.669271 Top5 98.789062 +2023-10-02 21:47:30,465 - Epoch: [162][ 40/ 117] Loss 0.287786 Top1 87.412109 Top5 98.642578 +2023-10-02 21:47:30,614 - Epoch: [162][ 50/ 117] Loss 0.296272 Top1 87.273438 Top5 98.648438 +2023-10-02 21:47:30,764 - Epoch: [162][ 60/ 117] Loss 0.297460 Top1 87.220052 Top5 98.613281 +2023-10-02 21:47:30,915 - Epoch: [162][ 70/ 117] Loss 0.299461 Top1 87.265625 Top5 98.638393 +2023-10-02 21:47:31,064 - Epoch: [162][ 80/ 117] Loss 0.297094 Top1 87.285156 Top5 98.671875 +2023-10-02 21:47:31,213 - Epoch: [162][ 90/ 117] Loss 0.297277 Top1 87.278646 Top5 98.697917 +2023-10-02 21:47:31,363 - Epoch: [162][ 100/ 117] Loss 0.294893 Top1 87.343750 Top5 98.722656 +2023-10-02 21:47:31,519 - Epoch: [162][ 110/ 117] Loss 0.297285 Top1 87.212358 Top5 98.707386 +2023-10-02 21:47:31,608 - Epoch: [162][ 117/ 117] Loss 0.298767 Top1 87.205691 Top5 98.700865 +2023-10-02 21:47:31,758 - ==> Top1: 87.206 Top5: 98.701 Loss: 0.299 + +2023-10-02 21:47:31,758 - ==> Confusion: +[[ 939 0 4 1 5 2 0 0 4 58 2 0 1 1 5 2 3 1 1 0 21] + [ 0 1063 0 1 4 25 2 18 1 0 0 1 2 0 1 3 2 0 2 2 4] + [ 1 2 984 5 1 1 16 8 0 1 2 0 6 3 1 3 2 1 7 3 9] + [ 1 3 12 987 2 4 2 0 1 0 6 1 3 2 27 0 0 5 13 1 19] + [ 24 3 2 0 969 6 1 0 2 10 1 0 1 3 9 4 12 0 0 0 3] + [ 3 28 0 3 2 1002 1 14 2 4 1 9 0 13 6 1 4 0 2 4 17] + [ 0 4 25 0 0 2 1129 4 0 0 3 0 0 0 0 6 0 1 2 6 9] + [ 4 14 12 1 5 27 7 1066 1 2 5 3 4 6 1 0 0 3 34 12 11] + [ 14 2 0 1 2 2 0 1 980 32 12 0 0 16 14 1 4 1 3 1 3] + [ 84 2 2 0 6 4 0 0 28 952 0 1 0 19 9 0 1 0 0 2 9] + [ 1 1 7 9 0 1 6 0 8 1 980 2 1 9 6 0 1 1 2 3 14] + [ 0 0 1 0 0 19 0 3 0 0 0 955 24 5 0 1 1 13 0 7 6] + [ 0 2 1 3 0 1 1 0 1 0 3 31 985 1 1 6 1 13 0 7 11] + [ 2 0 0 0 1 7 0 0 10 9 3 4 0 1060 4 0 0 1 0 1 17] + [ 11 0 4 12 4 0 0 0 13 3 0 0 3 3 1029 0 0 1 7 0 11] + [ 0 0 2 1 4 1 0 0 0 0 1 5 8 0 0 1066 19 12 2 8 5] + [ 1 16 1 0 5 7 1 0 0 0 0 4 0 1 3 6 1101 0 1 4 10] + [ 0 0 1 2 1 0 1 0 0 0 0 1 19 2 3 4 0 999 0 2 3] + [ 1 7 3 16 1 1 0 16 3 1 3 1 3 0 12 0 0 0 990 0 10] + [ 0 1 4 1 2 6 7 4 0 0 0 10 2 3 0 0 8 0 0 1095 9] + [ 103 127 101 74 63 123 24 68 62 56 157 69 305 252 124 33 84 55 98 146 5781]] + +2023-10-02 21:47:31,760 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:47:31,760 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:47:31,766 - + +2023-10-02 21:47:31,766 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:47:32,783 - Epoch: [163][ 10/ 1236] Overall Loss 0.120407 Objective Loss 0.120407 LR 0.000250 Time 0.101672 +2023-10-02 21:47:32,989 - Epoch: [163][ 20/ 1236] Overall Loss 0.126246 Objective Loss 0.126246 LR 0.000250 Time 0.061100 +2023-10-02 21:47:33,194 - Epoch: [163][ 30/ 1236] Overall Loss 0.124340 Objective Loss 0.124340 LR 0.000250 Time 0.047573 +2023-10-02 21:47:33,401 - Epoch: [163][ 40/ 1236] Overall Loss 0.122454 Objective Loss 0.122454 LR 0.000250 Time 0.040828 +2023-10-02 21:47:33,606 - Epoch: [163][ 50/ 1236] Overall Loss 0.121295 Objective Loss 0.121295 LR 0.000250 Time 0.036731 +2023-10-02 21:47:33,811 - Epoch: [163][ 60/ 1236] Overall Loss 0.123569 Objective Loss 0.123569 LR 0.000250 Time 0.034022 +2023-10-02 21:47:34,016 - Epoch: [163][ 70/ 1236] Overall Loss 0.124865 Objective Loss 0.124865 LR 0.000250 Time 0.032085 +2023-10-02 21:47:34,221 - Epoch: [163][ 80/ 1236] Overall Loss 0.126164 Objective Loss 0.126164 LR 0.000250 Time 0.030643 +2023-10-02 21:47:34,427 - Epoch: [163][ 90/ 1236] Overall Loss 0.127745 Objective Loss 0.127745 LR 0.000250 Time 0.029515 +2023-10-02 21:47:34,633 - Epoch: [163][ 100/ 1236] Overall Loss 0.127797 Objective Loss 0.127797 LR 0.000250 Time 0.028621 +2023-10-02 21:47:34,837 - Epoch: [163][ 110/ 1236] Overall Loss 0.126176 Objective Loss 0.126176 LR 0.000250 Time 0.027878 +2023-10-02 21:47:35,044 - Epoch: [163][ 120/ 1236] Overall Loss 0.126362 Objective Loss 0.126362 LR 0.000250 Time 0.027274 +2023-10-02 21:47:35,248 - Epoch: [163][ 130/ 1236] Overall Loss 0.126918 Objective Loss 0.126918 LR 0.000250 Time 0.026744 +2023-10-02 21:47:35,454 - Epoch: [163][ 140/ 1236] Overall Loss 0.127740 Objective Loss 0.127740 LR 0.000250 Time 0.026300 +2023-10-02 21:47:35,659 - Epoch: [163][ 150/ 1236] Overall Loss 0.127974 Objective Loss 0.127974 LR 0.000250 Time 0.025907 +2023-10-02 21:47:35,865 - Epoch: [163][ 160/ 1236] Overall Loss 0.128588 Objective Loss 0.128588 LR 0.000250 Time 0.025571 +2023-10-02 21:47:36,070 - Epoch: [163][ 170/ 1236] Overall Loss 0.128253 Objective Loss 0.128253 LR 0.000250 Time 0.025267 +2023-10-02 21:47:36,276 - Epoch: [163][ 180/ 1236] Overall Loss 0.129335 Objective Loss 0.129335 LR 0.000250 Time 0.025004 +2023-10-02 21:47:36,480 - Epoch: [163][ 190/ 1236] Overall Loss 0.129562 Objective Loss 0.129562 LR 0.000250 Time 0.024759 +2023-10-02 21:47:36,687 - Epoch: [163][ 200/ 1236] Overall Loss 0.128437 Objective Loss 0.128437 LR 0.000250 Time 0.024556 +2023-10-02 21:47:36,891 - Epoch: [163][ 210/ 1236] Overall Loss 0.127904 Objective Loss 0.127904 LR 0.000250 Time 0.024358 +2023-10-02 21:47:37,097 - Epoch: [163][ 220/ 1236] Overall Loss 0.128360 Objective Loss 0.128360 LR 0.000250 Time 0.024183 +2023-10-02 21:47:37,302 - Epoch: [163][ 230/ 1236] Overall Loss 0.128107 Objective Loss 0.128107 LR 0.000250 Time 0.024019 +2023-10-02 21:47:37,508 - Epoch: [163][ 240/ 1236] Overall Loss 0.127898 Objective Loss 0.127898 LR 0.000250 Time 0.023873 +2023-10-02 21:47:37,712 - Epoch: [163][ 250/ 1236] Overall Loss 0.128190 Objective Loss 0.128190 LR 0.000250 Time 0.023734 +2023-10-02 21:47:37,919 - Epoch: [163][ 260/ 1236] Overall Loss 0.128411 Objective Loss 0.128411 LR 0.000250 Time 0.023618 +2023-10-02 21:47:38,123 - Epoch: [163][ 270/ 1236] Overall Loss 0.128877 Objective Loss 0.128877 LR 0.000250 Time 0.023497 +2023-10-02 21:47:38,330 - Epoch: [163][ 280/ 1236] Overall Loss 0.129347 Objective Loss 0.129347 LR 0.000250 Time 0.023396 +2023-10-02 21:47:38,534 - Epoch: [163][ 290/ 1236] Overall Loss 0.130077 Objective Loss 0.130077 LR 0.000250 Time 0.023292 +2023-10-02 21:47:38,749 - Epoch: [163][ 300/ 1236] Overall Loss 0.130551 Objective Loss 0.130551 LR 0.000250 Time 0.023229 +2023-10-02 21:47:38,962 - Epoch: [163][ 310/ 1236] Overall Loss 0.130581 Objective Loss 0.130581 LR 0.000250 Time 0.023165 +2023-10-02 21:47:39,179 - Epoch: [163][ 320/ 1236] Overall Loss 0.130736 Objective Loss 0.130736 LR 0.000250 Time 0.023120 +2023-10-02 21:47:39,393 - Epoch: [163][ 330/ 1236] Overall Loss 0.131010 Objective Loss 0.131010 LR 0.000250 Time 0.023065 +2023-10-02 21:47:39,611 - Epoch: [163][ 340/ 1236] Overall Loss 0.130335 Objective Loss 0.130335 LR 0.000250 Time 0.023027 +2023-10-02 21:47:39,825 - Epoch: [163][ 350/ 1236] Overall Loss 0.130488 Objective Loss 0.130488 LR 0.000250 Time 0.022980 +2023-10-02 21:47:40,043 - Epoch: [163][ 360/ 1236] Overall Loss 0.130841 Objective Loss 0.130841 LR 0.000250 Time 0.022948 +2023-10-02 21:47:40,257 - Epoch: [163][ 370/ 1236] Overall Loss 0.130779 Objective Loss 0.130779 LR 0.000250 Time 0.022904 +2023-10-02 21:47:40,468 - Epoch: [163][ 380/ 1236] Overall Loss 0.130728 Objective Loss 0.130728 LR 0.000250 Time 0.022857 +2023-10-02 21:47:40,675 - Epoch: [163][ 390/ 1236] Overall Loss 0.130446 Objective Loss 0.130446 LR 0.000250 Time 0.022801 +2023-10-02 21:47:40,884 - Epoch: [163][ 400/ 1236] Overall Loss 0.130917 Objective Loss 0.130917 LR 0.000250 Time 0.022751 +2023-10-02 21:47:41,090 - Epoch: [163][ 410/ 1236] Overall Loss 0.130421 Objective Loss 0.130421 LR 0.000250 Time 0.022699 +2023-10-02 21:47:41,299 - Epoch: [163][ 420/ 1236] Overall Loss 0.130254 Objective Loss 0.130254 LR 0.000250 Time 0.022654 +2023-10-02 21:47:41,505 - Epoch: [163][ 430/ 1236] Overall Loss 0.130487 Objective Loss 0.130487 LR 0.000250 Time 0.022607 +2023-10-02 21:47:41,714 - Epoch: [163][ 440/ 1236] Overall Loss 0.130443 Objective Loss 0.130443 LR 0.000250 Time 0.022566 +2023-10-02 21:47:41,920 - Epoch: [163][ 450/ 1236] Overall Loss 0.130826 Objective Loss 0.130826 LR 0.000250 Time 0.022523 +2023-10-02 21:47:42,129 - Epoch: [163][ 460/ 1236] Overall Loss 0.130783 Objective Loss 0.130783 LR 0.000250 Time 0.022486 +2023-10-02 21:47:42,336 - Epoch: [163][ 470/ 1236] Overall Loss 0.131255 Objective Loss 0.131255 LR 0.000250 Time 0.022447 +2023-10-02 21:47:42,544 - Epoch: [163][ 480/ 1236] Overall Loss 0.131303 Objective Loss 0.131303 LR 0.000250 Time 0.022414 +2023-10-02 21:47:42,751 - Epoch: [163][ 490/ 1236] Overall Loss 0.131040 Objective Loss 0.131040 LR 0.000250 Time 0.022378 +2023-10-02 21:47:42,960 - Epoch: [163][ 500/ 1236] Overall Loss 0.130851 Objective Loss 0.130851 LR 0.000250 Time 0.022346 +2023-10-02 21:47:43,166 - Epoch: [163][ 510/ 1236] Overall Loss 0.130872 Objective Loss 0.130872 LR 0.000250 Time 0.022313 +2023-10-02 21:47:43,375 - Epoch: [163][ 520/ 1236] Overall Loss 0.130867 Objective Loss 0.130867 LR 0.000250 Time 0.022284 +2023-10-02 21:47:43,581 - Epoch: [163][ 530/ 1236] Overall Loss 0.131086 Objective Loss 0.131086 LR 0.000250 Time 0.022253 +2023-10-02 21:47:43,790 - Epoch: [163][ 540/ 1236] Overall Loss 0.131274 Objective Loss 0.131274 LR 0.000250 Time 0.022227 +2023-10-02 21:47:43,997 - Epoch: [163][ 550/ 1236] Overall Loss 0.131400 Objective Loss 0.131400 LR 0.000250 Time 0.022198 +2023-10-02 21:47:44,205 - Epoch: [163][ 560/ 1236] Overall Loss 0.131505 Objective Loss 0.131505 LR 0.000250 Time 0.022173 +2023-10-02 21:47:44,412 - Epoch: [163][ 570/ 1236] Overall Loss 0.131427 Objective Loss 0.131427 LR 0.000250 Time 0.022146 +2023-10-02 21:47:44,621 - Epoch: [163][ 580/ 1236] Overall Loss 0.131427 Objective Loss 0.131427 LR 0.000250 Time 0.022123 +2023-10-02 21:47:44,827 - Epoch: [163][ 590/ 1236] Overall Loss 0.131400 Objective Loss 0.131400 LR 0.000250 Time 0.022098 +2023-10-02 21:47:45,036 - Epoch: [163][ 600/ 1236] Overall Loss 0.131466 Objective Loss 0.131466 LR 0.000250 Time 0.022076 +2023-10-02 21:47:45,242 - Epoch: [163][ 610/ 1236] Overall Loss 0.131566 Objective Loss 0.131566 LR 0.000250 Time 0.022053 +2023-10-02 21:47:45,451 - Epoch: [163][ 620/ 1236] Overall Loss 0.131598 Objective Loss 0.131598 LR 0.000250 Time 0.022033 +2023-10-02 21:47:45,659 - Epoch: [163][ 630/ 1236] Overall Loss 0.131756 Objective Loss 0.131756 LR 0.000250 Time 0.022012 +2023-10-02 21:47:45,867 - Epoch: [163][ 640/ 1236] Overall Loss 0.131597 Objective Loss 0.131597 LR 0.000250 Time 0.021993 +2023-10-02 21:47:46,073 - Epoch: [163][ 650/ 1236] Overall Loss 0.131514 Objective Loss 0.131514 LR 0.000250 Time 0.021972 +2023-10-02 21:47:46,282 - Epoch: [163][ 660/ 1236] Overall Loss 0.131540 Objective Loss 0.131540 LR 0.000250 Time 0.021955 +2023-10-02 21:47:46,489 - Epoch: [163][ 670/ 1236] Overall Loss 0.131511 Objective Loss 0.131511 LR 0.000250 Time 0.021936 +2023-10-02 21:47:46,698 - Epoch: [163][ 680/ 1236] Overall Loss 0.131413 Objective Loss 0.131413 LR 0.000250 Time 0.021919 +2023-10-02 21:47:46,904 - Epoch: [163][ 690/ 1236] Overall Loss 0.131651 Objective Loss 0.131651 LR 0.000250 Time 0.021901 +2023-10-02 21:47:47,113 - Epoch: [163][ 700/ 1236] Overall Loss 0.132103 Objective Loss 0.132103 LR 0.000250 Time 0.021886 +2023-10-02 21:47:47,319 - Epoch: [163][ 710/ 1236] Overall Loss 0.132174 Objective Loss 0.132174 LR 0.000250 Time 0.021868 +2023-10-02 21:47:47,528 - Epoch: [163][ 720/ 1236] Overall Loss 0.132399 Objective Loss 0.132399 LR 0.000250 Time 0.021854 +2023-10-02 21:47:47,735 - Epoch: [163][ 730/ 1236] Overall Loss 0.132483 Objective Loss 0.132483 LR 0.000250 Time 0.021837 +2023-10-02 21:47:47,944 - Epoch: [163][ 740/ 1236] Overall Loss 0.132566 Objective Loss 0.132566 LR 0.000250 Time 0.021823 +2023-10-02 21:47:48,150 - Epoch: [163][ 750/ 1236] Overall Loss 0.132416 Objective Loss 0.132416 LR 0.000250 Time 0.021808 +2023-10-02 21:47:48,359 - Epoch: [163][ 760/ 1236] Overall Loss 0.132642 Objective Loss 0.132642 LR 0.000250 Time 0.021795 +2023-10-02 21:47:48,566 - Epoch: [163][ 770/ 1236] Overall Loss 0.132566 Objective Loss 0.132566 LR 0.000250 Time 0.021780 +2023-10-02 21:47:48,775 - Epoch: [163][ 780/ 1236] Overall Loss 0.132583 Objective Loss 0.132583 LR 0.000250 Time 0.021768 +2023-10-02 21:47:48,982 - Epoch: [163][ 790/ 1236] Overall Loss 0.132840 Objective Loss 0.132840 LR 0.000250 Time 0.021754 +2023-10-02 21:47:49,190 - Epoch: [163][ 800/ 1236] Overall Loss 0.132755 Objective Loss 0.132755 LR 0.000250 Time 0.021743 +2023-10-02 21:47:49,397 - Epoch: [163][ 810/ 1236] Overall Loss 0.132792 Objective Loss 0.132792 LR 0.000250 Time 0.021729 +2023-10-02 21:47:49,606 - Epoch: [163][ 820/ 1236] Overall Loss 0.132703 Objective Loss 0.132703 LR 0.000250 Time 0.021718 +2023-10-02 21:47:49,813 - Epoch: [163][ 830/ 1236] Overall Loss 0.132529 Objective Loss 0.132529 LR 0.000250 Time 0.021706 +2023-10-02 21:47:50,021 - Epoch: [163][ 840/ 1236] Overall Loss 0.132423 Objective Loss 0.132423 LR 0.000250 Time 0.021695 +2023-10-02 21:47:50,228 - Epoch: [163][ 850/ 1236] Overall Loss 0.132411 Objective Loss 0.132411 LR 0.000250 Time 0.021683 +2023-10-02 21:47:50,437 - Epoch: [163][ 860/ 1236] Overall Loss 0.132378 Objective Loss 0.132378 LR 0.000250 Time 0.021673 +2023-10-02 21:47:50,643 - Epoch: [163][ 870/ 1236] Overall Loss 0.132390 Objective Loss 0.132390 LR 0.000250 Time 0.021661 +2023-10-02 21:47:50,852 - Epoch: [163][ 880/ 1236] Overall Loss 0.132321 Objective Loss 0.132321 LR 0.000250 Time 0.021651 +2023-10-02 21:47:51,058 - Epoch: [163][ 890/ 1236] Overall Loss 0.132364 Objective Loss 0.132364 LR 0.000250 Time 0.021640 +2023-10-02 21:47:51,267 - Epoch: [163][ 900/ 1236] Overall Loss 0.132212 Objective Loss 0.132212 LR 0.000250 Time 0.021631 +2023-10-02 21:47:51,474 - Epoch: [163][ 910/ 1236] Overall Loss 0.132405 Objective Loss 0.132405 LR 0.000250 Time 0.021620 +2023-10-02 21:47:51,682 - Epoch: [163][ 920/ 1236] Overall Loss 0.132622 Objective Loss 0.132622 LR 0.000250 Time 0.021611 +2023-10-02 21:47:51,889 - Epoch: [163][ 930/ 1236] Overall Loss 0.132566 Objective Loss 0.132566 LR 0.000250 Time 0.021601 +2023-10-02 21:47:52,098 - Epoch: [163][ 940/ 1236] Overall Loss 0.132580 Objective Loss 0.132580 LR 0.000250 Time 0.021593 +2023-10-02 21:47:52,305 - Epoch: [163][ 950/ 1236] Overall Loss 0.132761 Objective Loss 0.132761 LR 0.000250 Time 0.021583 +2023-10-02 21:47:52,513 - Epoch: [163][ 960/ 1236] Overall Loss 0.132822 Objective Loss 0.132822 LR 0.000250 Time 0.021575 +2023-10-02 21:47:52,720 - Epoch: [163][ 970/ 1236] Overall Loss 0.132884 Objective Loss 0.132884 LR 0.000250 Time 0.021566 +2023-10-02 21:47:52,929 - Epoch: [163][ 980/ 1236] Overall Loss 0.132831 Objective Loss 0.132831 LR 0.000250 Time 0.021558 +2023-10-02 21:47:53,135 - Epoch: [163][ 990/ 1236] Overall Loss 0.132779 Objective Loss 0.132779 LR 0.000250 Time 0.021549 +2023-10-02 21:47:53,344 - Epoch: [163][ 1000/ 1236] Overall Loss 0.132679 Objective Loss 0.132679 LR 0.000250 Time 0.021541 +2023-10-02 21:47:53,551 - Epoch: [163][ 1010/ 1236] Overall Loss 0.132646 Objective Loss 0.132646 LR 0.000250 Time 0.021532 +2023-10-02 21:47:53,759 - Epoch: [163][ 1020/ 1236] Overall Loss 0.132633 Objective Loss 0.132633 LR 0.000250 Time 0.021526 +2023-10-02 21:47:53,966 - Epoch: [163][ 1030/ 1236] Overall Loss 0.132608 Objective Loss 0.132608 LR 0.000250 Time 0.021517 +2023-10-02 21:47:54,175 - Epoch: [163][ 1040/ 1236] Overall Loss 0.132774 Objective Loss 0.132774 LR 0.000250 Time 0.021511 +2023-10-02 21:47:54,382 - Epoch: [163][ 1050/ 1236] Overall Loss 0.132717 Objective Loss 0.132717 LR 0.000250 Time 0.021502 +2023-10-02 21:47:54,590 - Epoch: [163][ 1060/ 1236] Overall Loss 0.132818 Objective Loss 0.132818 LR 0.000250 Time 0.021496 +2023-10-02 21:47:54,797 - Epoch: [163][ 1070/ 1236] Overall Loss 0.132707 Objective Loss 0.132707 LR 0.000250 Time 0.021488 +2023-10-02 21:47:55,006 - Epoch: [163][ 1080/ 1236] Overall Loss 0.132701 Objective Loss 0.132701 LR 0.000250 Time 0.021482 +2023-10-02 21:47:55,213 - Epoch: [163][ 1090/ 1236] Overall Loss 0.132541 Objective Loss 0.132541 LR 0.000250 Time 0.021475 +2023-10-02 21:47:55,422 - Epoch: [163][ 1100/ 1236] Overall Loss 0.132508 Objective Loss 0.132508 LR 0.000250 Time 0.021470 +2023-10-02 21:47:55,629 - Epoch: [163][ 1110/ 1236] Overall Loss 0.132519 Objective Loss 0.132519 LR 0.000250 Time 0.021462 +2023-10-02 21:47:55,838 - Epoch: [163][ 1120/ 1236] Overall Loss 0.132597 Objective Loss 0.132597 LR 0.000250 Time 0.021457 +2023-10-02 21:47:56,045 - Epoch: [163][ 1130/ 1236] Overall Loss 0.132760 Objective Loss 0.132760 LR 0.000250 Time 0.021450 +2023-10-02 21:47:56,254 - Epoch: [163][ 1140/ 1236] Overall Loss 0.132787 Objective Loss 0.132787 LR 0.000250 Time 0.021444 +2023-10-02 21:47:56,461 - Epoch: [163][ 1150/ 1236] Overall Loss 0.132868 Objective Loss 0.132868 LR 0.000250 Time 0.021438 +2023-10-02 21:47:56,670 - Epoch: [163][ 1160/ 1236] Overall Loss 0.132836 Objective Loss 0.132836 LR 0.000250 Time 0.021433 +2023-10-02 21:47:56,877 - Epoch: [163][ 1170/ 1236] Overall Loss 0.132655 Objective Loss 0.132655 LR 0.000250 Time 0.021427 +2023-10-02 21:47:57,086 - Epoch: [163][ 1180/ 1236] Overall Loss 0.132688 Objective Loss 0.132688 LR 0.000250 Time 0.021422 +2023-10-02 21:47:57,293 - Epoch: [163][ 1190/ 1236] Overall Loss 0.132745 Objective Loss 0.132745 LR 0.000250 Time 0.021416 +2023-10-02 21:47:57,501 - Epoch: [163][ 1200/ 1236] Overall Loss 0.132885 Objective Loss 0.132885 LR 0.000250 Time 0.021411 +2023-10-02 21:47:57,709 - Epoch: [163][ 1210/ 1236] Overall Loss 0.132959 Objective Loss 0.132959 LR 0.000250 Time 0.021405 +2023-10-02 21:47:57,918 - Epoch: [163][ 1220/ 1236] Overall Loss 0.132984 Objective Loss 0.132984 LR 0.000250 Time 0.021400 +2023-10-02 21:47:58,179 - Epoch: [163][ 1230/ 1236] Overall Loss 0.133050 Objective Loss 0.133050 LR 0.000250 Time 0.021438 +2023-10-02 21:47:58,301 - Epoch: [163][ 1236/ 1236] Overall Loss 0.133106 Objective Loss 0.133106 Top1 89.816701 Top5 98.778004 LR 0.000250 Time 0.021433 +2023-10-02 21:47:58,426 - --- validate (epoch=163)----------- +2023-10-02 21:47:58,426 - 29943 samples (256 per mini-batch) +2023-10-02 21:47:58,927 - Epoch: [163][ 10/ 117] Loss 0.283873 Top1 87.382812 Top5 98.710938 +2023-10-02 21:47:59,091 - Epoch: [163][ 20/ 117] Loss 0.301454 Top1 86.640625 Top5 98.554688 +2023-10-02 21:47:59,248 - Epoch: [163][ 30/ 117] Loss 0.300222 Top1 86.679688 Top5 98.645833 +2023-10-02 21:47:59,411 - Epoch: [163][ 40/ 117] Loss 0.298424 Top1 86.621094 Top5 98.593750 +2023-10-02 21:47:59,568 - Epoch: [163][ 50/ 117] Loss 0.301424 Top1 86.656250 Top5 98.617188 +2023-10-02 21:47:59,732 - Epoch: [163][ 60/ 117] Loss 0.302178 Top1 86.712240 Top5 98.600260 +2023-10-02 21:47:59,889 - Epoch: [163][ 70/ 117] Loss 0.305226 Top1 86.808036 Top5 98.604911 +2023-10-02 21:48:00,052 - Epoch: [163][ 80/ 117] Loss 0.301292 Top1 86.938477 Top5 98.588867 +2023-10-02 21:48:00,209 - Epoch: [163][ 90/ 117] Loss 0.302955 Top1 86.953125 Top5 98.632812 +2023-10-02 21:48:00,371 - Epoch: [163][ 100/ 117] Loss 0.305964 Top1 86.925781 Top5 98.628906 +2023-10-02 21:48:00,536 - Epoch: [163][ 110/ 117] Loss 0.302891 Top1 86.963778 Top5 98.639915 +2023-10-02 21:48:00,625 - Epoch: [163][ 117/ 117] Loss 0.302218 Top1 86.991951 Top5 98.640751 +2023-10-02 21:48:00,720 - ==> Top1: 86.992 Top5: 98.641 Loss: 0.302 + +2023-10-02 21:48:00,721 - ==> Confusion: +[[ 944 2 1 1 12 2 1 0 2 50 1 3 1 0 3 2 1 1 1 0 22] + [ 0 1050 2 2 2 37 0 14 0 0 0 0 0 0 1 3 5 0 5 2 8] + [ 2 0 973 12 2 1 19 6 0 1 1 0 5 2 0 6 1 1 9 4 11] + [ 0 3 9 985 0 2 2 1 2 0 6 0 5 4 25 3 1 5 15 1 20] + [ 25 4 0 1 974 3 0 0 1 8 2 0 0 2 7 6 12 0 1 0 4] + [ 2 22 1 2 2 1025 0 11 4 5 1 5 0 9 6 0 2 0 3 0 16] + [ 0 3 26 2 0 2 1131 1 0 0 3 2 0 0 0 5 0 0 3 7 6] + [ 2 18 13 2 5 37 7 1042 0 4 6 7 4 4 1 2 0 0 42 10 12] + [ 15 2 0 1 2 5 0 0 978 35 10 1 1 10 15 0 5 0 5 0 4] + [ 94 0 1 1 7 3 0 0 25 947 2 0 1 16 8 3 1 1 0 1 8] + [ 3 2 7 5 0 3 2 1 11 0 987 1 2 8 4 0 0 1 4 3 9] + [ 0 1 0 0 0 14 0 3 0 0 0 975 12 7 0 1 1 11 0 5 5] + [ 0 0 1 4 0 2 2 0 0 1 3 35 974 0 1 9 3 9 3 6 15] + [ 0 0 0 0 2 8 0 0 15 7 3 7 0 1053 5 1 0 1 0 1 16] + [ 11 0 4 15 6 1 0 0 18 5 3 0 1 1 1014 0 2 1 10 0 9] + [ 0 0 2 1 5 2 0 0 0 0 1 5 8 0 0 1075 14 9 1 7 4] + [ 0 10 0 0 4 8 1 0 0 0 0 3 0 3 3 8 1107 0 1 4 9] + [ 0 0 1 1 0 0 1 0 1 0 0 3 22 0 2 9 1 993 0 1 3] + [ 3 4 3 19 0 1 0 17 3 0 4 2 2 0 7 1 0 1 990 0 11] + [ 0 1 4 2 1 3 6 6 0 0 1 16 5 1 1 0 8 0 0 1088 9] + [ 94 105 102 85 59 160 31 57 76 69 170 88 294 223 115 57 94 50 93 140 5743]] + +2023-10-02 21:48:00,722 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:48:00,722 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:48:00,728 - + +2023-10-02 21:48:00,728 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:48:01,858 - Epoch: [164][ 10/ 1236] Overall Loss 0.137085 Objective Loss 0.137085 LR 0.000250 Time 0.112877 +2023-10-02 21:48:02,065 - Epoch: [164][ 20/ 1236] Overall Loss 0.124906 Objective Loss 0.124906 LR 0.000250 Time 0.066795 +2023-10-02 21:48:02,273 - Epoch: [164][ 30/ 1236] Overall Loss 0.125646 Objective Loss 0.125646 LR 0.000250 Time 0.051456 +2023-10-02 21:48:02,483 - Epoch: [164][ 40/ 1236] Overall Loss 0.127675 Objective Loss 0.127675 LR 0.000250 Time 0.043819 +2023-10-02 21:48:02,688 - Epoch: [164][ 50/ 1236] Overall Loss 0.126852 Objective Loss 0.126852 LR 0.000250 Time 0.039168 +2023-10-02 21:48:02,898 - Epoch: [164][ 60/ 1236] Overall Loss 0.130940 Objective Loss 0.130940 LR 0.000250 Time 0.036126 +2023-10-02 21:48:03,104 - Epoch: [164][ 70/ 1236] Overall Loss 0.130522 Objective Loss 0.130522 LR 0.000250 Time 0.033905 +2023-10-02 21:48:03,313 - Epoch: [164][ 80/ 1236] Overall Loss 0.128847 Objective Loss 0.128847 LR 0.000250 Time 0.032280 +2023-10-02 21:48:03,519 - Epoch: [164][ 90/ 1236] Overall Loss 0.127950 Objective Loss 0.127950 LR 0.000250 Time 0.030978 +2023-10-02 21:48:03,729 - Epoch: [164][ 100/ 1236] Overall Loss 0.130047 Objective Loss 0.130047 LR 0.000250 Time 0.029973 +2023-10-02 21:48:03,934 - Epoch: [164][ 110/ 1236] Overall Loss 0.130631 Objective Loss 0.130631 LR 0.000250 Time 0.029116 +2023-10-02 21:48:04,144 - Epoch: [164][ 120/ 1236] Overall Loss 0.130831 Objective Loss 0.130831 LR 0.000250 Time 0.028431 +2023-10-02 21:48:04,349 - Epoch: [164][ 130/ 1236] Overall Loss 0.131860 Objective Loss 0.131860 LR 0.000250 Time 0.027825 +2023-10-02 21:48:04,557 - Epoch: [164][ 140/ 1236] Overall Loss 0.131009 Objective Loss 0.131009 LR 0.000250 Time 0.027319 +2023-10-02 21:48:04,764 - Epoch: [164][ 150/ 1236] Overall Loss 0.131048 Objective Loss 0.131048 LR 0.000250 Time 0.026866 +2023-10-02 21:48:04,973 - Epoch: [164][ 160/ 1236] Overall Loss 0.131608 Objective Loss 0.131608 LR 0.000250 Time 0.026493 +2023-10-02 21:48:05,179 - Epoch: [164][ 170/ 1236] Overall Loss 0.130760 Objective Loss 0.130760 LR 0.000250 Time 0.026143 +2023-10-02 21:48:05,388 - Epoch: [164][ 180/ 1236] Overall Loss 0.131320 Objective Loss 0.131320 LR 0.000250 Time 0.025852 +2023-10-02 21:48:05,594 - Epoch: [164][ 190/ 1236] Overall Loss 0.130926 Objective Loss 0.130926 LR 0.000250 Time 0.025572 +2023-10-02 21:48:05,803 - Epoch: [164][ 200/ 1236] Overall Loss 0.130942 Objective Loss 0.130942 LR 0.000250 Time 0.025339 +2023-10-02 21:48:06,009 - Epoch: [164][ 210/ 1236] Overall Loss 0.130635 Objective Loss 0.130635 LR 0.000250 Time 0.025111 +2023-10-02 21:48:06,217 - Epoch: [164][ 220/ 1236] Overall Loss 0.130492 Objective Loss 0.130492 LR 0.000250 Time 0.024914 +2023-10-02 21:48:06,424 - Epoch: [164][ 230/ 1236] Overall Loss 0.131025 Objective Loss 0.131025 LR 0.000250 Time 0.024724 +2023-10-02 21:48:06,632 - Epoch: [164][ 240/ 1236] Overall Loss 0.130972 Objective Loss 0.130972 LR 0.000250 Time 0.024559 +2023-10-02 21:48:06,839 - Epoch: [164][ 250/ 1236] Overall Loss 0.130905 Objective Loss 0.130905 LR 0.000250 Time 0.024399 +2023-10-02 21:48:07,047 - Epoch: [164][ 260/ 1236] Overall Loss 0.130555 Objective Loss 0.130555 LR 0.000250 Time 0.024259 +2023-10-02 21:48:07,254 - Epoch: [164][ 270/ 1236] Overall Loss 0.130680 Objective Loss 0.130680 LR 0.000250 Time 0.024121 +2023-10-02 21:48:07,463 - Epoch: [164][ 280/ 1236] Overall Loss 0.130453 Objective Loss 0.130453 LR 0.000250 Time 0.024007 +2023-10-02 21:48:07,669 - Epoch: [164][ 290/ 1236] Overall Loss 0.130278 Objective Loss 0.130278 LR 0.000250 Time 0.023887 +2023-10-02 21:48:07,878 - Epoch: [164][ 300/ 1236] Overall Loss 0.130320 Objective Loss 0.130320 LR 0.000250 Time 0.023788 +2023-10-02 21:48:08,084 - Epoch: [164][ 310/ 1236] Overall Loss 0.130024 Objective Loss 0.130024 LR 0.000250 Time 0.023683 +2023-10-02 21:48:08,292 - Epoch: [164][ 320/ 1236] Overall Loss 0.129223 Objective Loss 0.129223 LR 0.000250 Time 0.023592 +2023-10-02 21:48:08,499 - Epoch: [164][ 330/ 1236] Overall Loss 0.128973 Objective Loss 0.128973 LR 0.000250 Time 0.023500 +2023-10-02 21:48:08,707 - Epoch: [164][ 340/ 1236] Overall Loss 0.129100 Objective Loss 0.129100 LR 0.000250 Time 0.023420 +2023-10-02 21:48:08,914 - Epoch: [164][ 350/ 1236] Overall Loss 0.129740 Objective Loss 0.129740 LR 0.000250 Time 0.023338 +2023-10-02 21:48:09,122 - Epoch: [164][ 360/ 1236] Overall Loss 0.129880 Objective Loss 0.129880 LR 0.000250 Time 0.023266 +2023-10-02 21:48:09,329 - Epoch: [164][ 370/ 1236] Overall Loss 0.129501 Objective Loss 0.129501 LR 0.000250 Time 0.023194 +2023-10-02 21:48:09,537 - Epoch: [164][ 380/ 1236] Overall Loss 0.129513 Objective Loss 0.129513 LR 0.000250 Time 0.023130 +2023-10-02 21:48:09,745 - Epoch: [164][ 390/ 1236] Overall Loss 0.129736 Objective Loss 0.129736 LR 0.000250 Time 0.023069 +2023-10-02 21:48:09,953 - Epoch: [164][ 400/ 1236] Overall Loss 0.130319 Objective Loss 0.130319 LR 0.000250 Time 0.023012 +2023-10-02 21:48:10,161 - Epoch: [164][ 410/ 1236] Overall Loss 0.130537 Objective Loss 0.130537 LR 0.000250 Time 0.022958 +2023-10-02 21:48:10,369 - Epoch: [164][ 420/ 1236] Overall Loss 0.130182 Objective Loss 0.130182 LR 0.000250 Time 0.022904 +2023-10-02 21:48:10,577 - Epoch: [164][ 430/ 1236] Overall Loss 0.130029 Objective Loss 0.130029 LR 0.000250 Time 0.022854 +2023-10-02 21:48:10,785 - Epoch: [164][ 440/ 1236] Overall Loss 0.130055 Objective Loss 0.130055 LR 0.000250 Time 0.022808 +2023-10-02 21:48:10,993 - Epoch: [164][ 450/ 1236] Overall Loss 0.129983 Objective Loss 0.129983 LR 0.000250 Time 0.022762 +2023-10-02 21:48:11,201 - Epoch: [164][ 460/ 1236] Overall Loss 0.129918 Objective Loss 0.129918 LR 0.000250 Time 0.022719 +2023-10-02 21:48:11,409 - Epoch: [164][ 470/ 1236] Overall Loss 0.129980 Objective Loss 0.129980 LR 0.000250 Time 0.022678 +2023-10-02 21:48:11,617 - Epoch: [164][ 480/ 1236] Overall Loss 0.129773 Objective Loss 0.129773 LR 0.000250 Time 0.022638 +2023-10-02 21:48:11,825 - Epoch: [164][ 490/ 1236] Overall Loss 0.129439 Objective Loss 0.129439 LR 0.000250 Time 0.022599 +2023-10-02 21:48:12,032 - Epoch: [164][ 500/ 1236] Overall Loss 0.129767 Objective Loss 0.129767 LR 0.000250 Time 0.022563 +2023-10-02 21:48:12,241 - Epoch: [164][ 510/ 1236] Overall Loss 0.130222 Objective Loss 0.130222 LR 0.000250 Time 0.022528 +2023-10-02 21:48:12,449 - Epoch: [164][ 520/ 1236] Overall Loss 0.130278 Objective Loss 0.130278 LR 0.000250 Time 0.022494 +2023-10-02 21:48:12,657 - Epoch: [164][ 530/ 1236] Overall Loss 0.130115 Objective Loss 0.130115 LR 0.000250 Time 0.022461 +2023-10-02 21:48:12,865 - Epoch: [164][ 540/ 1236] Overall Loss 0.130200 Objective Loss 0.130200 LR 0.000250 Time 0.022431 +2023-10-02 21:48:13,073 - Epoch: [164][ 550/ 1236] Overall Loss 0.130699 Objective Loss 0.130699 LR 0.000250 Time 0.022400 +2023-10-02 21:48:13,281 - Epoch: [164][ 560/ 1236] Overall Loss 0.130552 Objective Loss 0.130552 LR 0.000250 Time 0.022371 +2023-10-02 21:48:13,492 - Epoch: [164][ 570/ 1236] Overall Loss 0.130778 Objective Loss 0.130778 LR 0.000250 Time 0.022348 +2023-10-02 21:48:13,701 - Epoch: [164][ 580/ 1236] Overall Loss 0.130511 Objective Loss 0.130511 LR 0.000250 Time 0.022323 +2023-10-02 21:48:13,913 - Epoch: [164][ 590/ 1236] Overall Loss 0.130373 Objective Loss 0.130373 LR 0.000250 Time 0.022304 +2023-10-02 21:48:14,123 - Epoch: [164][ 600/ 1236] Overall Loss 0.130750 Objective Loss 0.130750 LR 0.000250 Time 0.022281 +2023-10-02 21:48:14,334 - Epoch: [164][ 610/ 1236] Overall Loss 0.131025 Objective Loss 0.131025 LR 0.000250 Time 0.022261 +2023-10-02 21:48:14,544 - Epoch: [164][ 620/ 1236] Overall Loss 0.131270 Objective Loss 0.131270 LR 0.000250 Time 0.022239 +2023-10-02 21:48:14,756 - Epoch: [164][ 630/ 1236] Overall Loss 0.131207 Objective Loss 0.131207 LR 0.000250 Time 0.022222 +2023-10-02 21:48:14,967 - Epoch: [164][ 640/ 1236] Overall Loss 0.131408 Objective Loss 0.131408 LR 0.000250 Time 0.022204 +2023-10-02 21:48:15,180 - Epoch: [164][ 650/ 1236] Overall Loss 0.131239 Objective Loss 0.131239 LR 0.000250 Time 0.022189 +2023-10-02 21:48:15,390 - Epoch: [164][ 660/ 1236] Overall Loss 0.131384 Objective Loss 0.131384 LR 0.000250 Time 0.022170 +2023-10-02 21:48:15,601 - Epoch: [164][ 670/ 1236] Overall Loss 0.131422 Objective Loss 0.131422 LR 0.000250 Time 0.022155 +2023-10-02 21:48:15,811 - Epoch: [164][ 680/ 1236] Overall Loss 0.131891 Objective Loss 0.131891 LR 0.000250 Time 0.022137 +2023-10-02 21:48:16,022 - Epoch: [164][ 690/ 1236] Overall Loss 0.131849 Objective Loss 0.131849 LR 0.000250 Time 0.022122 +2023-10-02 21:48:16,232 - Epoch: [164][ 700/ 1236] Overall Loss 0.132020 Objective Loss 0.132020 LR 0.000250 Time 0.022104 +2023-10-02 21:48:16,444 - Epoch: [164][ 710/ 1236] Overall Loss 0.131968 Objective Loss 0.131968 LR 0.000250 Time 0.022091 +2023-10-02 21:48:16,653 - Epoch: [164][ 720/ 1236] Overall Loss 0.132043 Objective Loss 0.132043 LR 0.000250 Time 0.022074 +2023-10-02 21:48:16,866 - Epoch: [164][ 730/ 1236] Overall Loss 0.131935 Objective Loss 0.131935 LR 0.000250 Time 0.022062 +2023-10-02 21:48:17,075 - Epoch: [164][ 740/ 1236] Overall Loss 0.132133 Objective Loss 0.132133 LR 0.000250 Time 0.022047 +2023-10-02 21:48:17,288 - Epoch: [164][ 750/ 1236] Overall Loss 0.132150 Objective Loss 0.132150 LR 0.000250 Time 0.022035 +2023-10-02 21:48:17,498 - Epoch: [164][ 760/ 1236] Overall Loss 0.132373 Objective Loss 0.132373 LR 0.000250 Time 0.022021 +2023-10-02 21:48:17,710 - Epoch: [164][ 770/ 1236] Overall Loss 0.132353 Objective Loss 0.132353 LR 0.000250 Time 0.022011 +2023-10-02 21:48:17,921 - Epoch: [164][ 780/ 1236] Overall Loss 0.132731 Objective Loss 0.132731 LR 0.000250 Time 0.021999 +2023-10-02 21:48:18,135 - Epoch: [164][ 790/ 1236] Overall Loss 0.132932 Objective Loss 0.132932 LR 0.000250 Time 0.021990 +2023-10-02 21:48:18,345 - Epoch: [164][ 800/ 1236] Overall Loss 0.132829 Objective Loss 0.132829 LR 0.000250 Time 0.021977 +2023-10-02 21:48:18,557 - Epoch: [164][ 810/ 1236] Overall Loss 0.132821 Objective Loss 0.132821 LR 0.000250 Time 0.021967 +2023-10-02 21:48:18,767 - Epoch: [164][ 820/ 1236] Overall Loss 0.133085 Objective Loss 0.133085 LR 0.000250 Time 0.021954 +2023-10-02 21:48:18,979 - Epoch: [164][ 830/ 1236] Overall Loss 0.132937 Objective Loss 0.132937 LR 0.000250 Time 0.021945 +2023-10-02 21:48:19,189 - Epoch: [164][ 840/ 1236] Overall Loss 0.133270 Objective Loss 0.133270 LR 0.000250 Time 0.021933 +2023-10-02 21:48:19,401 - Epoch: [164][ 850/ 1236] Overall Loss 0.133298 Objective Loss 0.133298 LR 0.000250 Time 0.021924 +2023-10-02 21:48:19,611 - Epoch: [164][ 860/ 1236] Overall Loss 0.133429 Objective Loss 0.133429 LR 0.000250 Time 0.021913 +2023-10-02 21:48:19,823 - Epoch: [164][ 870/ 1236] Overall Loss 0.133321 Objective Loss 0.133321 LR 0.000250 Time 0.021904 +2023-10-02 21:48:20,034 - Epoch: [164][ 880/ 1236] Overall Loss 0.133253 Objective Loss 0.133253 LR 0.000250 Time 0.021894 +2023-10-02 21:48:20,249 - Epoch: [164][ 890/ 1236] Overall Loss 0.133321 Objective Loss 0.133321 LR 0.000250 Time 0.021889 +2023-10-02 21:48:20,460 - Epoch: [164][ 900/ 1236] Overall Loss 0.133283 Objective Loss 0.133283 LR 0.000250 Time 0.021880 +2023-10-02 21:48:20,675 - Epoch: [164][ 910/ 1236] Overall Loss 0.133202 Objective Loss 0.133202 LR 0.000250 Time 0.021875 +2023-10-02 21:48:20,886 - Epoch: [164][ 920/ 1236] Overall Loss 0.132987 Objective Loss 0.132987 LR 0.000250 Time 0.021867 +2023-10-02 21:48:21,100 - Epoch: [164][ 930/ 1236] Overall Loss 0.133053 Objective Loss 0.133053 LR 0.000250 Time 0.021862 +2023-10-02 21:48:21,311 - Epoch: [164][ 940/ 1236] Overall Loss 0.132942 Objective Loss 0.132942 LR 0.000250 Time 0.021853 +2023-10-02 21:48:21,525 - Epoch: [164][ 950/ 1236] Overall Loss 0.132818 Objective Loss 0.132818 LR 0.000250 Time 0.021848 +2023-10-02 21:48:21,736 - Epoch: [164][ 960/ 1236] Overall Loss 0.132796 Objective Loss 0.132796 LR 0.000250 Time 0.021839 +2023-10-02 21:48:21,950 - Epoch: [164][ 970/ 1236] Overall Loss 0.132866 Objective Loss 0.132866 LR 0.000250 Time 0.021835 +2023-10-02 21:48:22,161 - Epoch: [164][ 980/ 1236] Overall Loss 0.132831 Objective Loss 0.132831 LR 0.000250 Time 0.021827 +2023-10-02 21:48:22,374 - Epoch: [164][ 990/ 1236] Overall Loss 0.132863 Objective Loss 0.132863 LR 0.000250 Time 0.021821 +2023-10-02 21:48:22,585 - Epoch: [164][ 1000/ 1236] Overall Loss 0.132817 Objective Loss 0.132817 LR 0.000250 Time 0.021813 +2023-10-02 21:48:22,799 - Epoch: [164][ 1010/ 1236] Overall Loss 0.132784 Objective Loss 0.132784 LR 0.000250 Time 0.021809 +2023-10-02 21:48:23,010 - Epoch: [164][ 1020/ 1236] Overall Loss 0.132796 Objective Loss 0.132796 LR 0.000250 Time 0.021801 +2023-10-02 21:48:23,224 - Epoch: [164][ 1030/ 1236] Overall Loss 0.132640 Objective Loss 0.132640 LR 0.000250 Time 0.021797 +2023-10-02 21:48:23,435 - Epoch: [164][ 1040/ 1236] Overall Loss 0.132791 Objective Loss 0.132791 LR 0.000250 Time 0.021790 +2023-10-02 21:48:23,649 - Epoch: [164][ 1050/ 1236] Overall Loss 0.132895 Objective Loss 0.132895 LR 0.000250 Time 0.021786 +2023-10-02 21:48:23,861 - Epoch: [164][ 1060/ 1236] Overall Loss 0.133033 Objective Loss 0.133033 LR 0.000250 Time 0.021779 +2023-10-02 21:48:24,075 - Epoch: [164][ 1070/ 1236] Overall Loss 0.133063 Objective Loss 0.133063 LR 0.000250 Time 0.021776 +2023-10-02 21:48:24,286 - Epoch: [164][ 1080/ 1236] Overall Loss 0.133218 Objective Loss 0.133218 LR 0.000250 Time 0.021769 +2023-10-02 21:48:24,501 - Epoch: [164][ 1090/ 1236] Overall Loss 0.133314 Objective Loss 0.133314 LR 0.000250 Time 0.021766 +2023-10-02 21:48:24,712 - Epoch: [164][ 1100/ 1236] Overall Loss 0.133365 Objective Loss 0.133365 LR 0.000250 Time 0.021760 +2023-10-02 21:48:24,925 - Epoch: [164][ 1110/ 1236] Overall Loss 0.133417 Objective Loss 0.133417 LR 0.000250 Time 0.021756 +2023-10-02 21:48:25,136 - Epoch: [164][ 1120/ 1236] Overall Loss 0.133526 Objective Loss 0.133526 LR 0.000250 Time 0.021749 +2023-10-02 21:48:25,350 - Epoch: [164][ 1130/ 1236] Overall Loss 0.133516 Objective Loss 0.133516 LR 0.000250 Time 0.021746 +2023-10-02 21:48:25,561 - Epoch: [164][ 1140/ 1236] Overall Loss 0.133673 Objective Loss 0.133673 LR 0.000250 Time 0.021740 +2023-10-02 21:48:25,775 - Epoch: [164][ 1150/ 1236] Overall Loss 0.133610 Objective Loss 0.133610 LR 0.000250 Time 0.021737 +2023-10-02 21:48:25,987 - Epoch: [164][ 1160/ 1236] Overall Loss 0.133640 Objective Loss 0.133640 LR 0.000250 Time 0.021730 +2023-10-02 21:48:26,201 - Epoch: [164][ 1170/ 1236] Overall Loss 0.133685 Objective Loss 0.133685 LR 0.000250 Time 0.021727 +2023-10-02 21:48:26,412 - Epoch: [164][ 1180/ 1236] Overall Loss 0.133594 Objective Loss 0.133594 LR 0.000250 Time 0.021721 +2023-10-02 21:48:26,624 - Epoch: [164][ 1190/ 1236] Overall Loss 0.133511 Objective Loss 0.133511 LR 0.000250 Time 0.021717 +2023-10-02 21:48:26,835 - Epoch: [164][ 1200/ 1236] Overall Loss 0.133503 Objective Loss 0.133503 LR 0.000250 Time 0.021712 +2023-10-02 21:48:27,050 - Epoch: [164][ 1210/ 1236] Overall Loss 0.133306 Objective Loss 0.133306 LR 0.000250 Time 0.021709 +2023-10-02 21:48:27,261 - Epoch: [164][ 1220/ 1236] Overall Loss 0.133256 Objective Loss 0.133256 LR 0.000250 Time 0.021704 +2023-10-02 21:48:27,526 - Epoch: [164][ 1230/ 1236] Overall Loss 0.133250 Objective Loss 0.133250 LR 0.000250 Time 0.021743 +2023-10-02 21:48:27,649 - Epoch: [164][ 1236/ 1236] Overall Loss 0.133222 Objective Loss 0.133222 Top1 90.427699 Top5 98.370672 LR 0.000250 Time 0.021736 +2023-10-02 21:48:27,787 - --- validate (epoch=164)----------- +2023-10-02 21:48:27,787 - 29943 samples (256 per mini-batch) +2023-10-02 21:48:28,282 - Epoch: [164][ 10/ 117] Loss 0.315794 Top1 86.601562 Top5 98.828125 +2023-10-02 21:48:28,435 - Epoch: [164][ 20/ 117] Loss 0.308928 Top1 87.089844 Top5 98.691406 +2023-10-02 21:48:28,587 - Epoch: [164][ 30/ 117] Loss 0.303515 Top1 87.304688 Top5 98.763021 +2023-10-02 21:48:28,739 - Epoch: [164][ 40/ 117] Loss 0.304158 Top1 87.021484 Top5 98.652344 +2023-10-02 21:48:28,890 - Epoch: [164][ 50/ 117] Loss 0.301978 Top1 86.882812 Top5 98.703125 +2023-10-02 21:48:29,043 - Epoch: [164][ 60/ 117] Loss 0.304155 Top1 86.816406 Top5 98.678385 +2023-10-02 21:48:29,195 - Epoch: [164][ 70/ 117] Loss 0.308483 Top1 86.724330 Top5 98.621652 +2023-10-02 21:48:29,348 - Epoch: [164][ 80/ 117] Loss 0.307564 Top1 86.723633 Top5 98.598633 +2023-10-02 21:48:29,500 - Epoch: [164][ 90/ 117] Loss 0.309458 Top1 86.545139 Top5 98.589410 +2023-10-02 21:48:29,652 - Epoch: [164][ 100/ 117] Loss 0.305692 Top1 86.722656 Top5 98.578125 +2023-10-02 21:48:29,810 - Epoch: [164][ 110/ 117] Loss 0.304794 Top1 86.722301 Top5 98.590199 +2023-10-02 21:48:29,900 - Epoch: [164][ 117/ 117] Loss 0.303623 Top1 86.698060 Top5 98.630732 +2023-10-02 21:48:29,997 - ==> Top1: 86.698 Top5: 98.631 Loss: 0.304 + +2023-10-02 21:48:29,998 - ==> Confusion: +[[ 931 1 6 0 4 2 0 0 7 63 1 3 1 2 6 1 1 1 1 0 19] + [ 0 1057 1 0 4 18 0 24 0 1 0 0 0 0 0 3 1 0 15 2 5] + [ 1 1 997 3 0 0 9 7 0 2 1 0 8 3 0 3 2 1 11 2 5] + [ 1 1 14 973 1 1 3 3 3 0 4 1 6 3 26 5 1 5 20 1 17] + [ 25 5 0 0 970 7 0 0 0 10 1 0 0 5 9 5 7 0 1 1 4] + [ 2 32 1 2 2 993 2 25 2 3 1 9 1 13 4 0 2 0 4 2 16] + [ 0 4 27 1 0 1 1130 4 0 0 3 1 0 0 0 6 0 1 2 6 5] + [ 3 11 11 0 7 17 5 1086 0 2 6 2 4 3 2 2 1 2 38 7 9] + [ 16 2 0 1 2 4 0 2 970 39 10 0 1 16 14 1 4 1 3 0 3] + [ 93 0 3 0 8 3 0 0 32 936 2 2 0 23 7 0 1 0 0 0 9] + [ 2 2 10 7 0 0 2 3 7 2 965 1 1 24 3 0 3 1 10 0 10] + [ 0 0 1 0 0 13 1 3 0 0 0 965 18 6 0 2 1 17 0 3 5] + [ 0 1 0 3 0 1 2 0 1 1 5 33 977 0 2 9 1 14 1 4 13] + [ 0 0 0 0 2 7 0 0 6 8 4 6 0 1068 4 0 0 2 0 2 10] + [ 12 0 4 17 2 0 0 0 14 2 1 0 2 1 1023 0 1 1 11 0 10] + [ 0 0 2 1 5 0 1 0 0 0 1 6 6 0 1 1071 16 10 3 6 5] + [ 0 17 1 0 6 5 1 0 0 0 0 5 0 4 3 9 1088 0 1 6 15] + [ 0 1 3 0 2 0 0 0 0 0 0 3 17 3 1 4 0 999 0 1 4] + [ 2 3 2 14 1 0 0 14 2 1 1 1 0 0 13 0 0 0 1003 0 11] + [ 0 0 4 1 1 1 7 8 0 0 0 13 3 3 1 2 8 1 1 1087 11] + [ 93 126 142 65 60 108 29 100 71 73 138 84 316 289 128 52 58 52 118 132 5671]] + +2023-10-02 21:48:29,999 - ==> Best [Top1: 87.286 Top5: 98.637 Sparsity:0.00 Params: 169472 on epoch: 155] +2023-10-02 21:48:29,999 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:48:30,005 - + +2023-10-02 21:48:30,005 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:48:31,055 - Epoch: [165][ 10/ 1236] Overall Loss 0.144136 Objective Loss 0.144136 LR 0.000250 Time 0.104929 +2023-10-02 21:48:31,264 - Epoch: [165][ 20/ 1236] Overall Loss 0.130190 Objective Loss 0.130190 LR 0.000250 Time 0.062907 +2023-10-02 21:48:31,472 - Epoch: [165][ 30/ 1236] Overall Loss 0.134137 Objective Loss 0.134137 LR 0.000250 Time 0.048855 +2023-10-02 21:48:31,681 - Epoch: [165][ 40/ 1236] Overall Loss 0.135374 Objective Loss 0.135374 LR 0.000250 Time 0.041867 +2023-10-02 21:48:31,889 - Epoch: [165][ 50/ 1236] Overall Loss 0.132757 Objective Loss 0.132757 LR 0.000250 Time 0.037636 +2023-10-02 21:48:32,099 - Epoch: [165][ 60/ 1236] Overall Loss 0.128197 Objective Loss 0.128197 LR 0.000250 Time 0.034869 +2023-10-02 21:48:32,306 - Epoch: [165][ 70/ 1236] Overall Loss 0.128150 Objective Loss 0.128150 LR 0.000250 Time 0.032828 +2023-10-02 21:48:32,516 - Epoch: [165][ 80/ 1236] Overall Loss 0.129046 Objective Loss 0.129046 LR 0.000250 Time 0.031355 +2023-10-02 21:48:32,723 - Epoch: [165][ 90/ 1236] Overall Loss 0.130017 Objective Loss 0.130017 LR 0.000250 Time 0.030160 +2023-10-02 21:48:32,933 - Epoch: [165][ 100/ 1236] Overall Loss 0.130492 Objective Loss 0.130492 LR 0.000250 Time 0.029249 +2023-10-02 21:48:33,140 - Epoch: [165][ 110/ 1236] Overall Loss 0.130415 Objective Loss 0.130415 LR 0.000250 Time 0.028466 +2023-10-02 21:48:33,352 - Epoch: [165][ 120/ 1236] Overall Loss 0.130734 Objective Loss 0.130734 LR 0.000250 Time 0.027856 +2023-10-02 21:48:33,560 - Epoch: [165][ 130/ 1236] Overall Loss 0.130635 Objective Loss 0.130635 LR 0.000250 Time 0.027310 +2023-10-02 21:48:33,772 - Epoch: [165][ 140/ 1236] Overall Loss 0.131132 Objective Loss 0.131132 LR 0.000250 Time 0.026872 +2023-10-02 21:48:33,980 - Epoch: [165][ 150/ 1236] Overall Loss 0.131355 Objective Loss 0.131355 LR 0.000250 Time 0.026467 +2023-10-02 21:48:34,192 - Epoch: [165][ 160/ 1236] Overall Loss 0.131045 Objective Loss 0.131045 LR 0.000250 Time 0.026136 +2023-10-02 21:48:34,400 - Epoch: [165][ 170/ 1236] Overall Loss 0.131387 Objective Loss 0.131387 LR 0.000250 Time 0.025821 +2023-10-02 21:48:34,612 - Epoch: [165][ 180/ 1236] Overall Loss 0.130667 Objective Loss 0.130667 LR 0.000250 Time 0.025562 +2023-10-02 21:48:34,820 - Epoch: [165][ 190/ 1236] Overall Loss 0.130867 Objective Loss 0.130867 LR 0.000250 Time 0.025310 +2023-10-02 21:48:35,032 - Epoch: [165][ 200/ 1236] Overall Loss 0.131271 Objective Loss 0.131271 LR 0.000250 Time 0.025103 +2023-10-02 21:48:35,240 - Epoch: [165][ 210/ 1236] Overall Loss 0.130772 Objective Loss 0.130772 LR 0.000250 Time 0.024896 +2023-10-02 21:48:35,452 - Epoch: [165][ 220/ 1236] Overall Loss 0.130431 Objective Loss 0.130431 LR 0.000250 Time 0.024727 +2023-10-02 21:48:35,660 - Epoch: [165][ 230/ 1236] Overall Loss 0.130232 Objective Loss 0.130232 LR 0.000250 Time 0.024554 +2023-10-02 21:48:35,872 - Epoch: [165][ 240/ 1236] Overall Loss 0.130681 Objective Loss 0.130681 LR 0.000250 Time 0.024414 +2023-10-02 21:48:36,080 - Epoch: [165][ 250/ 1236] Overall Loss 0.130852 Objective Loss 0.130852 LR 0.000250 Time 0.024268 +2023-10-02 21:48:36,292 - Epoch: [165][ 260/ 1236] Overall Loss 0.130962 Objective Loss 0.130962 LR 0.000250 Time 0.024150 +2023-10-02 21:48:36,500 - Epoch: [165][ 270/ 1236] Overall Loss 0.131246 Objective Loss 0.131246 LR 0.000250 Time 0.024025 +2023-10-02 21:48:36,712 - Epoch: [165][ 280/ 1236] Overall Loss 0.132072 Objective Loss 0.132072 LR 0.000250 Time 0.023922 +2023-10-02 21:48:36,920 - Epoch: [165][ 290/ 1236] Overall Loss 0.132350 Objective Loss 0.132350 LR 0.000250 Time 0.023813 +2023-10-02 21:48:37,132 - Epoch: [165][ 300/ 1236] Overall Loss 0.132210 Objective Loss 0.132210 LR 0.000250 Time 0.023726 +2023-10-02 21:48:37,339 - Epoch: [165][ 310/ 1236] Overall Loss 0.132604 Objective Loss 0.132604 LR 0.000250 Time 0.023626 +2023-10-02 21:48:37,549 - Epoch: [165][ 320/ 1236] Overall Loss 0.131915 Objective Loss 0.131915 LR 0.000250 Time 0.023546 +2023-10-02 21:48:37,756 - Epoch: [165][ 330/ 1236] Overall Loss 0.132096 Objective Loss 0.132096 LR 0.000250 Time 0.023458 +2023-10-02 21:48:37,966 - Epoch: [165][ 340/ 1236] Overall Loss 0.132084 Objective Loss 0.132084 LR 0.000250 Time 0.023384 +2023-10-02 21:48:38,173 - Epoch: [165][ 350/ 1236] Overall Loss 0.132457 Objective Loss 0.132457 LR 0.000250 Time 0.023308 +2023-10-02 21:48:38,384 - Epoch: [165][ 360/ 1236] Overall Loss 0.132792 Objective Loss 0.132792 LR 0.000250 Time 0.023246 +2023-10-02 21:48:38,591 - Epoch: [165][ 370/ 1236] Overall Loss 0.133212 Objective Loss 0.133212 LR 0.000250 Time 0.023175 +2023-10-02 21:48:38,801 - Epoch: [165][ 380/ 1236] Overall Loss 0.133350 Objective Loss 0.133350 LR 0.000250 Time 0.023116 +2023-10-02 21:48:39,008 - Epoch: [165][ 390/ 1236] Overall Loss 0.133700 Objective Loss 0.133700 LR 0.000250 Time 0.023056 +2023-10-02 21:48:39,219 - Epoch: [165][ 400/ 1236] Overall Loss 0.133379 Objective Loss 0.133379 LR 0.000250 Time 0.023006 +2023-10-02 21:48:39,426 - Epoch: [165][ 410/ 1236] Overall Loss 0.133074 Objective Loss 0.133074 LR 0.000250 Time 0.022948 +2023-10-02 21:48:39,637 - Epoch: [165][ 420/ 1236] Overall Loss 0.132723 Objective Loss 0.132723 LR 0.000250 Time 0.022903 +2023-10-02 21:48:39,843 - Epoch: [165][ 430/ 1236] Overall Loss 0.132803 Objective Loss 0.132803 LR 0.000250 Time 0.022850 +2023-10-02 21:48:40,052 - Epoch: [165][ 440/ 1236] Overall Loss 0.132417 Objective Loss 0.132417 LR 0.000250 Time 0.022805 +2023-10-02 21:48:40,260 - Epoch: [165][ 450/ 1236] Overall Loss 0.132369 Objective Loss 0.132369 LR 0.000250 Time 0.022759 +2023-10-02 21:48:40,471 - Epoch: [165][ 460/ 1236] Overall Loss 0.132009 Objective Loss 0.132009 LR 0.000250 Time 0.022722 +2023-10-02 21:48:40,677 - Epoch: [165][ 470/ 1236] Overall Loss 0.131813 Objective Loss 0.131813 LR 0.000250 Time 0.022677 +2023-10-02 21:48:40,888 - Epoch: [165][ 480/ 1236] Overall Loss 0.132114 Objective Loss 0.132114 LR 0.000250 Time 0.022644 +2023-10-02 21:48:41,095 - Epoch: [165][ 490/ 1236] Overall Loss 0.132194 Objective Loss 0.132194 LR 0.000250 Time 0.022603 +2023-10-02 21:48:41,306 - Epoch: [165][ 500/ 1236] Overall Loss 0.132239 Objective Loss 0.132239 LR 0.000250 Time 0.022572 +2023-10-02 21:48:41,512 - Epoch: [165][ 510/ 1236] Overall Loss 0.131768 Objective Loss 0.131768 LR 0.000250 Time 0.022534 +2023-10-02 21:48:41,722 - Epoch: [165][ 520/ 1236] Overall Loss 0.131771 Objective Loss 0.131771 LR 0.000250 Time 0.022504 +2023-10-02 21:48:41,930 - Epoch: [165][ 530/ 1236] Overall Loss 0.132042 Objective Loss 0.132042 LR 0.000250 Time 0.022471 +2023-10-02 21:48:42,140 - Epoch: [165][ 540/ 1236] Overall Loss 0.131932 Objective Loss 0.131932 LR 0.000250 Time 0.022442 +2023-10-02 21:48:42,347 - Epoch: [165][ 550/ 1236] Overall Loss 0.131769 Objective Loss 0.131769 LR 0.000250 Time 0.022411 +2023-10-02 21:48:42,559 - Epoch: [165][ 560/ 1236] Overall Loss 0.131615 Objective Loss 0.131615 LR 0.000250 Time 0.022387 +2023-10-02 21:48:42,765 - Epoch: [165][ 570/ 1236] Overall Loss 0.131638 Objective Loss 0.131638 LR 0.000250 Time 0.022356 +2023-10-02 21:48:42,976 - Epoch: [165][ 580/ 1236] Overall Loss 0.131591 Objective Loss 0.131591 LR 0.000250 Time 0.022335 +2023-10-02 21:48:43,183 - Epoch: [165][ 590/ 1236] Overall Loss 0.131547 Objective Loss 0.131547 LR 0.000250 Time 0.022305 +2023-10-02 21:48:43,394 - Epoch: [165][ 600/ 1236] Overall Loss 0.131588 Objective Loss 0.131588 LR 0.000250 Time 0.022285 +2023-10-02 21:48:43,600 - Epoch: [165][ 610/ 1236] Overall Loss 0.131622 Objective Loss 0.131622 LR 0.000250 Time 0.022258 +2023-10-02 21:48:43,810 - Epoch: [165][ 620/ 1236] Overall Loss 0.131430 Objective Loss 0.131430 LR 0.000250 Time 0.022236 +2023-10-02 21:48:44,018 - Epoch: [165][ 630/ 1236] Overall Loss 0.131363 Objective Loss 0.131363 LR 0.000250 Time 0.022211 +2023-10-02 21:48:44,229 - Epoch: [165][ 640/ 1236] Overall Loss 0.131396 Objective Loss 0.131396 LR 0.000250 Time 0.022193 +2023-10-02 21:48:44,435 - Epoch: [165][ 650/ 1236] Overall Loss 0.131871 Objective Loss 0.131871 LR 0.000250 Time 0.022169 +2023-10-02 21:48:44,647 - Epoch: [165][ 660/ 1236] Overall Loss 0.131935 Objective Loss 0.131935 LR 0.000250 Time 0.022153 +2023-10-02 21:48:44,853 - Epoch: [165][ 670/ 1236] Overall Loss 0.132005 Objective Loss 0.132005 LR 0.000250 Time 0.022130 +2023-10-02 21:48:45,064 - Epoch: [165][ 680/ 1236] Overall Loss 0.131972 Objective Loss 0.131972 LR 0.000250 Time 0.022115 +2023-10-02 21:48:45,271 - Epoch: [165][ 690/ 1236] Overall Loss 0.131846 Objective Loss 0.131846 LR 0.000250 Time 0.022093 +2023-10-02 21:48:45,482 - Epoch: [165][ 700/ 1236] Overall Loss 0.131835 Objective Loss 0.131835 LR 0.000250 Time 0.022079 +2023-10-02 21:48:45,688 - Epoch: [165][ 710/ 1236] Overall Loss 0.132001 Objective Loss 0.132001 LR 0.000250 Time 0.022058 +2023-10-02 21:48:45,900 - Epoch: [165][ 720/ 1236] Overall Loss 0.131909 Objective Loss 0.131909 LR 0.000250 Time 0.022045 +2023-10-02 21:48:46,106 - Epoch: [165][ 730/ 1236] Overall Loss 0.131845 Objective Loss 0.131845 LR 0.000250 Time 0.022026 +2023-10-02 21:48:46,317 - Epoch: [165][ 740/ 1236] Overall Loss 0.131951 Objective Loss 0.131951 LR 0.000250 Time 0.022013 +2023-10-02 21:48:46,524 - Epoch: [165][ 750/ 1236] Overall Loss 0.131820 Objective Loss 0.131820 LR 0.000250 Time 0.021994 +2023-10-02 21:48:46,734 - Epoch: [165][ 760/ 1236] Overall Loss 0.131933 Objective Loss 0.131933 LR 0.000250 Time 0.021981 +2023-10-02 21:48:46,942 - Epoch: [165][ 770/ 1236] Overall Loss 0.132014 Objective Loss 0.132014 LR 0.000250 Time 0.021965 +2023-10-02 21:48:47,153 - Epoch: [165][ 780/ 1236] Overall Loss 0.132007 Objective Loss 0.132007 LR 0.000250 Time 0.021954 +2023-10-02 21:48:47,360 - Epoch: [165][ 790/ 1236] Overall Loss 0.131941 Objective Loss 0.131941 LR 0.000250 Time 0.021938 +2023-10-02 21:48:47,573 - Epoch: [165][ 800/ 1236] Overall Loss 0.131803 Objective Loss 0.131803 LR 0.000250 Time 0.021929 +2023-10-02 21:48:47,783 - Epoch: [165][ 810/ 1236] Overall Loss 0.131545 Objective Loss 0.131545 LR 0.000250 Time 0.021917 +2023-10-02 21:48:47,996 - Epoch: [165][ 820/ 1236] Overall Loss 0.131495 Objective Loss 0.131495 LR 0.000250 Time 0.021910 +2023-10-02 21:48:48,205 - Epoch: [165][ 830/ 1236] Overall Loss 0.131735 Objective Loss 0.131735 LR 0.000250 Time 0.021896 +2023-10-02 21:48:48,418 - Epoch: [165][ 840/ 1236] Overall Loss 0.131652 Objective Loss 0.131652 LR 0.000250 Time 0.021889 +2023-10-02 21:48:48,629 - Epoch: [165][ 850/ 1236] Overall Loss 0.131495 Objective Loss 0.131495 LR 0.000250 Time 0.021879 +2023-10-02 21:48:48,842 - Epoch: [165][ 860/ 1236] Overall Loss 0.131579 Objective Loss 0.131579 LR 0.000250 Time 0.021872 +2023-10-02 21:48:49,052 - Epoch: [165][ 870/ 1236] Overall Loss 0.131542 Objective Loss 0.131542 LR 0.000250 Time 0.021862 +2023-10-02 21:48:49,267 - Epoch: [165][ 880/ 1236] Overall Loss 0.131605 Objective Loss 0.131605 LR 0.000250 Time 0.021857 +2023-10-02 21:48:49,477 - Epoch: [165][ 890/ 1236] Overall Loss 0.131623 Objective Loss 0.131623 LR 0.000250 Time 0.021848 +2023-10-02 21:48:49,690 - Epoch: [165][ 900/ 1236] Overall Loss 0.131728 Objective Loss 0.131728 LR 0.000250 Time 0.021841 +2023-10-02 21:48:49,902 - Epoch: [165][ 910/ 1236] Overall Loss 0.131783 Objective Loss 0.131783 LR 0.000250 Time 0.021833 +2023-10-02 21:48:50,115 - Epoch: [165][ 920/ 1236] Overall Loss 0.131858 Objective Loss 0.131858 LR 0.000250 Time 0.021827 +2023-10-02 21:48:50,325 - Epoch: [165][ 930/ 1236] Overall Loss 0.131877 Objective Loss 0.131877 LR 0.000250 Time 0.021818 +2023-10-02 21:48:50,539 - Epoch: [165][ 940/ 1236] Overall Loss 0.132051 Objective Loss 0.132051 LR 0.000250 Time 0.021813 +2023-10-02 21:48:50,749 - Epoch: [165][ 950/ 1236] Overall Loss 0.132258 Objective Loss 0.132258 LR 0.000250 Time 0.021805 +2023-10-02 21:48:50,963 - Epoch: [165][ 960/ 1236] Overall Loss 0.132002 Objective Loss 0.132002 LR 0.000250 Time 0.021799 +2023-10-02 21:48:51,174 - Epoch: [165][ 970/ 1236] Overall Loss 0.132037 Objective Loss 0.132037 LR 0.000250 Time 0.021792 +2023-10-02 21:48:51,387 - Epoch: [165][ 980/ 1236] Overall Loss 0.132177 Objective Loss 0.132177 LR 0.000250 Time 0.021786 +2023-10-02 21:48:51,597 - Epoch: [165][ 990/ 1236] Overall Loss 0.132074 Objective Loss 0.132074 LR 0.000250 Time 0.021779 +2023-10-02 21:48:51,811 - Epoch: [165][ 1000/ 1236] Overall Loss 0.132099 Objective Loss 0.132099 LR 0.000250 Time 0.021775 +2023-10-02 21:48:52,021 - Epoch: [165][ 1010/ 1236] Overall Loss 0.132089 Objective Loss 0.132089 LR 0.000250 Time 0.021767 +2023-10-02 21:48:52,236 - Epoch: [165][ 1020/ 1236] Overall Loss 0.132125 Objective Loss 0.132125 LR 0.000250 Time 0.021763 +2023-10-02 21:48:52,446 - Epoch: [165][ 1030/ 1236] Overall Loss 0.131974 Objective Loss 0.131974 LR 0.000250 Time 0.021756 +2023-10-02 21:48:52,661 - Epoch: [165][ 1040/ 1236] Overall Loss 0.132133 Objective Loss 0.132133 LR 0.000250 Time 0.021752 +2023-10-02 21:48:52,871 - Epoch: [165][ 1050/ 1236] Overall Loss 0.132053 Objective Loss 0.132053 LR 0.000250 Time 0.021745 +2023-10-02 21:48:53,085 - Epoch: [165][ 1060/ 1236] Overall Loss 0.132145 Objective Loss 0.132145 LR 0.000250 Time 0.021741 +2023-10-02 21:48:53,295 - Epoch: [165][ 1070/ 1236] Overall Loss 0.132203 Objective Loss 0.132203 LR 0.000250 Time 0.021734 +2023-10-02 21:48:53,509 - Epoch: [165][ 1080/ 1236] Overall Loss 0.132103 Objective Loss 0.132103 LR 0.000250 Time 0.021730 +2023-10-02 21:48:53,719 - Epoch: [165][ 1090/ 1236] Overall Loss 0.132074 Objective Loss 0.132074 LR 0.000250 Time 0.021723 +2023-10-02 21:48:53,933 - Epoch: [165][ 1100/ 1236] Overall Loss 0.132021 Objective Loss 0.132021 LR 0.000250 Time 0.021720 +2023-10-02 21:48:54,144 - Epoch: [165][ 1110/ 1236] Overall Loss 0.132147 Objective Loss 0.132147 LR 0.000250 Time 0.021715 +2023-10-02 21:48:54,358 - Epoch: [165][ 1120/ 1236] Overall Loss 0.132402 Objective Loss 0.132402 LR 0.000250 Time 0.021712 +2023-10-02 21:48:54,569 - Epoch: [165][ 1130/ 1236] Overall Loss 0.132400 Objective Loss 0.132400 LR 0.000250 Time 0.021706 +2023-10-02 21:48:54,782 - Epoch: [165][ 1140/ 1236] Overall Loss 0.132467 Objective Loss 0.132467 LR 0.000250 Time 0.021701 +2023-10-02 21:48:54,992 - Epoch: [165][ 1150/ 1236] Overall Loss 0.132603 Objective Loss 0.132603 LR 0.000250 Time 0.021696 +2023-10-02 21:48:55,206 - Epoch: [165][ 1160/ 1236] Overall Loss 0.132550 Objective Loss 0.132550 LR 0.000250 Time 0.021692 +2023-10-02 21:48:55,416 - Epoch: [165][ 1170/ 1236] Overall Loss 0.132544 Objective Loss 0.132544 LR 0.000250 Time 0.021686 +2023-10-02 21:48:55,630 - Epoch: [165][ 1180/ 1236] Overall Loss 0.132566 Objective Loss 0.132566 LR 0.000250 Time 0.021683 +2023-10-02 21:48:55,840 - Epoch: [165][ 1190/ 1236] Overall Loss 0.132613 Objective Loss 0.132613 LR 0.000250 Time 0.021677 +2023-10-02 21:48:56,054 - Epoch: [165][ 1200/ 1236] Overall Loss 0.132726 Objective Loss 0.132726 LR 0.000250 Time 0.021675 +2023-10-02 21:48:56,264 - Epoch: [165][ 1210/ 1236] Overall Loss 0.132713 Objective Loss 0.132713 LR 0.000250 Time 0.021669 +2023-10-02 21:48:56,477 - Epoch: [165][ 1220/ 1236] Overall Loss 0.132662 Objective Loss 0.132662 LR 0.000250 Time 0.021666 +2023-10-02 21:48:56,743 - Epoch: [165][ 1230/ 1236] Overall Loss 0.132842 Objective Loss 0.132842 LR 0.000250 Time 0.021706 +2023-10-02 21:48:56,866 - Epoch: [165][ 1236/ 1236] Overall Loss 0.132912 Objective Loss 0.132912 Top1 92.057026 Top5 99.592668 LR 0.000250 Time 0.021700 +2023-10-02 21:48:57,003 - --- validate (epoch=165)----------- +2023-10-02 21:48:57,003 - 29943 samples (256 per mini-batch) +2023-10-02 21:48:57,505 - Epoch: [165][ 10/ 117] Loss 0.289037 Top1 87.265625 Top5 98.710938 +2023-10-02 21:48:57,657 - Epoch: [165][ 20/ 117] Loss 0.306556 Top1 87.539062 Top5 98.671875 +2023-10-02 21:48:57,808 - Epoch: [165][ 30/ 117] Loss 0.298628 Top1 87.838542 Top5 98.606771 +2023-10-02 21:48:57,960 - Epoch: [165][ 40/ 117] Loss 0.295264 Top1 87.783203 Top5 98.652344 +2023-10-02 21:48:58,113 - Epoch: [165][ 50/ 117] Loss 0.298661 Top1 87.648438 Top5 98.648438 +2023-10-02 21:48:58,265 - Epoch: [165][ 60/ 117] Loss 0.300252 Top1 87.526042 Top5 98.658854 +2023-10-02 21:48:58,416 - Epoch: [165][ 70/ 117] Loss 0.302955 Top1 87.393973 Top5 98.699777 +2023-10-02 21:48:58,569 - Epoch: [165][ 80/ 117] Loss 0.302521 Top1 87.290039 Top5 98.725586 +2023-10-02 21:48:58,721 - Epoch: [165][ 90/ 117] Loss 0.299812 Top1 87.330729 Top5 98.732639 +2023-10-02 21:48:58,874 - Epoch: [165][ 100/ 117] Loss 0.299677 Top1 87.414062 Top5 98.691406 +2023-10-02 21:48:59,034 - Epoch: [165][ 110/ 117] Loss 0.300596 Top1 87.372159 Top5 98.693182 +2023-10-02 21:48:59,124 - Epoch: [165][ 117/ 117] Loss 0.301822 Top1 87.342618 Top5 98.694186 +2023-10-02 21:48:59,262 - ==> Top1: 87.343 Top5: 98.694 Loss: 0.302 + +2023-10-02 21:48:59,263 - ==> Confusion: +[[ 935 0 1 1 7 2 0 0 9 59 1 0 1 2 4 1 3 0 1 0 23] + [ 0 1071 1 0 4 15 2 19 0 1 0 0 2 0 1 3 1 0 3 1 7] + [ 1 0 976 13 1 0 18 5 0 1 3 1 7 2 0 4 1 1 9 2 11] + [ 0 1 12 986 0 0 1 2 4 0 4 1 4 2 29 4 2 4 13 1 19] + [ 19 5 0 1 971 7 0 0 2 9 0 0 1 4 7 6 11 0 0 2 5] + [ 3 33 0 2 4 997 1 21 3 4 1 5 1 9 3 0 3 1 3 4 18] + [ 0 2 31 1 0 2 1125 2 0 0 3 2 0 0 0 7 0 0 2 9 5] + [ 1 8 11 1 2 22 8 1081 1 2 5 6 2 5 2 0 1 0 39 10 11] + [ 12 1 0 0 2 4 0 2 989 31 9 1 2 10 13 1 6 0 1 0 5] + [ 76 0 2 0 5 2 0 0 30 958 2 1 1 24 4 1 1 1 0 0 11] + [ 2 4 9 12 0 1 6 1 10 1 962 0 0 18 4 0 3 1 5 2 12] + [ 0 0 2 0 1 14 0 3 0 0 0 951 28 6 0 0 2 17 0 6 5] + [ 1 1 1 3 0 0 2 0 0 0 2 22 985 1 0 6 1 11 2 5 25] + [ 0 0 0 0 3 5 1 0 12 7 4 8 2 1052 3 1 0 1 0 2 18] + [ 11 1 6 15 3 0 0 0 20 2 1 0 3 2 1021 0 0 1 6 0 9] + [ 0 0 2 1 3 1 1 0 0 0 1 4 8 0 0 1075 15 8 1 9 5] + [ 0 13 1 0 3 6 1 0 0 0 0 6 0 3 4 8 1099 0 0 5 12] + [ 0 0 1 2 0 0 2 0 0 0 0 2 23 1 2 7 0 992 0 1 5] + [ 0 6 6 22 0 0 0 19 4 1 1 1 1 0 12 0 1 0 983 0 11] + [ 0 1 4 2 0 3 9 6 0 1 1 7 6 3 3 1 7 2 0 1091 5] + [ 79 121 129 72 47 111 30 68 75 51 145 56 294 243 108 60 95 51 80 137 5853]] + +2023-10-02 21:48:59,264 - ==> Best [Top1: 87.343 Top5: 98.694 Sparsity:0.00 Params: 169472 on epoch: 165] +2023-10-02 21:48:59,264 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:48:59,278 - + +2023-10-02 21:48:59,278 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:49:00,432 - Epoch: [166][ 10/ 1236] Overall Loss 0.116359 Objective Loss 0.116359 LR 0.000250 Time 0.115374 +2023-10-02 21:49:00,643 - Epoch: [166][ 20/ 1236] Overall Loss 0.125973 Objective Loss 0.125973 LR 0.000250 Time 0.068189 +2023-10-02 21:49:00,852 - Epoch: [166][ 30/ 1236] Overall Loss 0.131621 Objective Loss 0.131621 LR 0.000250 Time 0.052366 +2023-10-02 21:49:01,061 - Epoch: [166][ 40/ 1236] Overall Loss 0.130916 Objective Loss 0.130916 LR 0.000250 Time 0.044513 +2023-10-02 21:49:01,270 - Epoch: [166][ 50/ 1236] Overall Loss 0.130521 Objective Loss 0.130521 LR 0.000250 Time 0.039746 +2023-10-02 21:49:01,481 - Epoch: [166][ 60/ 1236] Overall Loss 0.130936 Objective Loss 0.130936 LR 0.000250 Time 0.036634 +2023-10-02 21:49:01,688 - Epoch: [166][ 70/ 1236] Overall Loss 0.129336 Objective Loss 0.129336 LR 0.000250 Time 0.034358 +2023-10-02 21:49:01,898 - Epoch: [166][ 80/ 1236] Overall Loss 0.128872 Objective Loss 0.128872 LR 0.000250 Time 0.032685 +2023-10-02 21:49:02,106 - Epoch: [166][ 90/ 1236] Overall Loss 0.129630 Objective Loss 0.129630 LR 0.000250 Time 0.031346 +2023-10-02 21:49:02,315 - Epoch: [166][ 100/ 1236] Overall Loss 0.128642 Objective Loss 0.128642 LR 0.000250 Time 0.030301 +2023-10-02 21:49:02,523 - Epoch: [166][ 110/ 1236] Overall Loss 0.129271 Objective Loss 0.129271 LR 0.000250 Time 0.029422 +2023-10-02 21:49:02,733 - Epoch: [166][ 120/ 1236] Overall Loss 0.128078 Objective Loss 0.128078 LR 0.000250 Time 0.028718 +2023-10-02 21:49:02,942 - Epoch: [166][ 130/ 1236] Overall Loss 0.128679 Objective Loss 0.128679 LR 0.000250 Time 0.028104 +2023-10-02 21:49:03,149 - Epoch: [166][ 140/ 1236] Overall Loss 0.129004 Objective Loss 0.129004 LR 0.000250 Time 0.027577 +2023-10-02 21:49:03,356 - Epoch: [166][ 150/ 1236] Overall Loss 0.129057 Objective Loss 0.129057 LR 0.000250 Time 0.027107 +2023-10-02 21:49:03,565 - Epoch: [166][ 160/ 1236] Overall Loss 0.128737 Objective Loss 0.128737 LR 0.000250 Time 0.026719 +2023-10-02 21:49:03,774 - Epoch: [166][ 170/ 1236] Overall Loss 0.129079 Objective Loss 0.129079 LR 0.000250 Time 0.026367 +2023-10-02 21:49:03,984 - Epoch: [166][ 180/ 1236] Overall Loss 0.129419 Objective Loss 0.129419 LR 0.000250 Time 0.026063 +2023-10-02 21:49:04,192 - Epoch: [166][ 190/ 1236] Overall Loss 0.128952 Objective Loss 0.128952 LR 0.000250 Time 0.025778 +2023-10-02 21:49:04,401 - Epoch: [166][ 200/ 1236] Overall Loss 0.128787 Objective Loss 0.128787 LR 0.000250 Time 0.025534 +2023-10-02 21:49:04,609 - Epoch: [166][ 210/ 1236] Overall Loss 0.129120 Objective Loss 0.129120 LR 0.000250 Time 0.025302 +2023-10-02 21:49:04,819 - Epoch: [166][ 220/ 1236] Overall Loss 0.129622 Objective Loss 0.129622 LR 0.000250 Time 0.025103 +2023-10-02 21:49:05,026 - Epoch: [166][ 230/ 1236] Overall Loss 0.129156 Objective Loss 0.129156 LR 0.000250 Time 0.024909 +2023-10-02 21:49:05,240 - Epoch: [166][ 240/ 1236] Overall Loss 0.128494 Objective Loss 0.128494 LR 0.000250 Time 0.024736 +2023-10-02 21:49:05,448 - Epoch: [166][ 250/ 1236] Overall Loss 0.128338 Objective Loss 0.128338 LR 0.000250 Time 0.024572 +2023-10-02 21:49:05,657 - Epoch: [166][ 260/ 1236] Overall Loss 0.127854 Objective Loss 0.127854 LR 0.000250 Time 0.024432 +2023-10-02 21:49:05,865 - Epoch: [166][ 270/ 1236] Overall Loss 0.127741 Objective Loss 0.127741 LR 0.000250 Time 0.024292 +2023-10-02 21:49:06,078 - Epoch: [166][ 280/ 1236] Overall Loss 0.127978 Objective Loss 0.127978 LR 0.000250 Time 0.024181 +2023-10-02 21:49:06,286 - Epoch: [166][ 290/ 1236] Overall Loss 0.127948 Objective Loss 0.127948 LR 0.000250 Time 0.024060 +2023-10-02 21:49:06,497 - Epoch: [166][ 300/ 1236] Overall Loss 0.127884 Objective Loss 0.127884 LR 0.000250 Time 0.023961 +2023-10-02 21:49:06,705 - Epoch: [166][ 310/ 1236] Overall Loss 0.127916 Objective Loss 0.127916 LR 0.000250 Time 0.023857 +2023-10-02 21:49:06,915 - Epoch: [166][ 320/ 1236] Overall Loss 0.127982 Objective Loss 0.127982 LR 0.000250 Time 0.023769 +2023-10-02 21:49:07,125 - Epoch: [166][ 330/ 1236] Overall Loss 0.128315 Objective Loss 0.128315 LR 0.000250 Time 0.023678 +2023-10-02 21:49:07,335 - Epoch: [166][ 340/ 1236] Overall Loss 0.128885 Objective Loss 0.128885 LR 0.000250 Time 0.023600 +2023-10-02 21:49:07,544 - Epoch: [166][ 350/ 1236] Overall Loss 0.128832 Objective Loss 0.128832 LR 0.000250 Time 0.023518 +2023-10-02 21:49:07,754 - Epoch: [166][ 360/ 1236] Overall Loss 0.128575 Objective Loss 0.128575 LR 0.000250 Time 0.023448 +2023-10-02 21:49:07,964 - Epoch: [166][ 370/ 1236] Overall Loss 0.128696 Objective Loss 0.128696 LR 0.000250 Time 0.023376 +2023-10-02 21:49:08,174 - Epoch: [166][ 380/ 1236] Overall Loss 0.128928 Objective Loss 0.128928 LR 0.000250 Time 0.023314 +2023-10-02 21:49:08,383 - Epoch: [166][ 390/ 1236] Overall Loss 0.128891 Objective Loss 0.128891 LR 0.000250 Time 0.023248 +2023-10-02 21:49:08,594 - Epoch: [166][ 400/ 1236] Overall Loss 0.129084 Objective Loss 0.129084 LR 0.000250 Time 0.023192 +2023-10-02 21:49:08,803 - Epoch: [166][ 410/ 1236] Overall Loss 0.129457 Objective Loss 0.129457 LR 0.000250 Time 0.023133 +2023-10-02 21:49:09,013 - Epoch: [166][ 420/ 1236] Overall Loss 0.129235 Objective Loss 0.129235 LR 0.000250 Time 0.023083 +2023-10-02 21:49:09,223 - Epoch: [166][ 430/ 1236] Overall Loss 0.128872 Objective Loss 0.128872 LR 0.000250 Time 0.023029 +2023-10-02 21:49:09,433 - Epoch: [166][ 440/ 1236] Overall Loss 0.128646 Objective Loss 0.128646 LR 0.000250 Time 0.022983 +2023-10-02 21:49:09,642 - Epoch: [166][ 450/ 1236] Overall Loss 0.128389 Objective Loss 0.128389 LR 0.000250 Time 0.022932 +2023-10-02 21:49:09,852 - Epoch: [166][ 460/ 1236] Overall Loss 0.128571 Objective Loss 0.128571 LR 0.000250 Time 0.022892 +2023-10-02 21:49:10,062 - Epoch: [166][ 470/ 1236] Overall Loss 0.128345 Objective Loss 0.128345 LR 0.000250 Time 0.022846 +2023-10-02 21:49:10,272 - Epoch: [166][ 480/ 1236] Overall Loss 0.128354 Objective Loss 0.128354 LR 0.000250 Time 0.022809 +2023-10-02 21:49:10,482 - Epoch: [166][ 490/ 1236] Overall Loss 0.128800 Objective Loss 0.128800 LR 0.000250 Time 0.022768 +2023-10-02 21:49:10,693 - Epoch: [166][ 500/ 1236] Overall Loss 0.129106 Objective Loss 0.129106 LR 0.000250 Time 0.022733 +2023-10-02 21:49:10,902 - Epoch: [166][ 510/ 1236] Overall Loss 0.129096 Objective Loss 0.129096 LR 0.000250 Time 0.022695 +2023-10-02 21:49:11,112 - Epoch: [166][ 520/ 1236] Overall Loss 0.128850 Objective Loss 0.128850 LR 0.000250 Time 0.022663 +2023-10-02 21:49:11,322 - Epoch: [166][ 530/ 1236] Overall Loss 0.128526 Objective Loss 0.128526 LR 0.000250 Time 0.022627 +2023-10-02 21:49:11,532 - Epoch: [166][ 540/ 1236] Overall Loss 0.129005 Objective Loss 0.129005 LR 0.000250 Time 0.022598 +2023-10-02 21:49:11,742 - Epoch: [166][ 550/ 1236] Overall Loss 0.129177 Objective Loss 0.129177 LR 0.000250 Time 0.022565 +2023-10-02 21:49:11,952 - Epoch: [166][ 560/ 1236] Overall Loss 0.128980 Objective Loss 0.128980 LR 0.000250 Time 0.022536 +2023-10-02 21:49:12,161 - Epoch: [166][ 570/ 1236] Overall Loss 0.129244 Objective Loss 0.129244 LR 0.000250 Time 0.022505 +2023-10-02 21:49:12,372 - Epoch: [166][ 580/ 1236] Overall Loss 0.129226 Objective Loss 0.129226 LR 0.000250 Time 0.022480 +2023-10-02 21:49:12,581 - Epoch: [166][ 590/ 1236] Overall Loss 0.129200 Objective Loss 0.129200 LR 0.000250 Time 0.022451 +2023-10-02 21:49:12,791 - Epoch: [166][ 600/ 1236] Overall Loss 0.129390 Objective Loss 0.129390 LR 0.000250 Time 0.022427 +2023-10-02 21:49:13,001 - Epoch: [166][ 610/ 1236] Overall Loss 0.129613 Objective Loss 0.129613 LR 0.000250 Time 0.022400 +2023-10-02 21:49:13,213 - Epoch: [166][ 620/ 1236] Overall Loss 0.129499 Objective Loss 0.129499 LR 0.000250 Time 0.022380 +2023-10-02 21:49:13,421 - Epoch: [166][ 630/ 1236] Overall Loss 0.129523 Objective Loss 0.129523 LR 0.000250 Time 0.022355 +2023-10-02 21:49:13,632 - Epoch: [166][ 640/ 1236] Overall Loss 0.129541 Objective Loss 0.129541 LR 0.000250 Time 0.022334 +2023-10-02 21:49:13,841 - Epoch: [166][ 650/ 1236] Overall Loss 0.129525 Objective Loss 0.129525 LR 0.000250 Time 0.022310 +2023-10-02 21:49:14,053 - Epoch: [166][ 660/ 1236] Overall Loss 0.129475 Objective Loss 0.129475 LR 0.000250 Time 0.022293 +2023-10-02 21:49:14,265 - Epoch: [166][ 670/ 1236] Overall Loss 0.129405 Objective Loss 0.129405 LR 0.000250 Time 0.022275 +2023-10-02 21:49:14,476 - Epoch: [166][ 680/ 1236] Overall Loss 0.129422 Objective Loss 0.129422 LR 0.000250 Time 0.022256 +2023-10-02 21:49:14,684 - Epoch: [166][ 690/ 1236] Overall Loss 0.129593 Objective Loss 0.129593 LR 0.000250 Time 0.022234 +2023-10-02 21:49:14,892 - Epoch: [166][ 700/ 1236] Overall Loss 0.129488 Objective Loss 0.129488 LR 0.000250 Time 0.022214 +2023-10-02 21:49:15,100 - Epoch: [166][ 710/ 1236] Overall Loss 0.129311 Objective Loss 0.129311 LR 0.000250 Time 0.022194 +2023-10-02 21:49:15,308 - Epoch: [166][ 720/ 1236] Overall Loss 0.129340 Objective Loss 0.129340 LR 0.000250 Time 0.022174 +2023-10-02 21:49:15,517 - Epoch: [166][ 730/ 1236] Overall Loss 0.129249 Objective Loss 0.129249 LR 0.000250 Time 0.022156 +2023-10-02 21:49:15,724 - Epoch: [166][ 740/ 1236] Overall Loss 0.129479 Objective Loss 0.129479 LR 0.000250 Time 0.022136 +2023-10-02 21:49:15,933 - Epoch: [166][ 750/ 1236] Overall Loss 0.129518 Objective Loss 0.129518 LR 0.000250 Time 0.022119 +2023-10-02 21:49:16,140 - Epoch: [166][ 760/ 1236] Overall Loss 0.129513 Objective Loss 0.129513 LR 0.000250 Time 0.022099 +2023-10-02 21:49:16,349 - Epoch: [166][ 770/ 1236] Overall Loss 0.129509 Objective Loss 0.129509 LR 0.000250 Time 0.022082 +2023-10-02 21:49:16,556 - Epoch: [166][ 780/ 1236] Overall Loss 0.129635 Objective Loss 0.129635 LR 0.000250 Time 0.022062 +2023-10-02 21:49:16,765 - Epoch: [166][ 790/ 1236] Overall Loss 0.129536 Objective Loss 0.129536 LR 0.000250 Time 0.022047 +2023-10-02 21:49:16,972 - Epoch: [166][ 800/ 1236] Overall Loss 0.129709 Objective Loss 0.129709 LR 0.000250 Time 0.022028 +2023-10-02 21:49:17,181 - Epoch: [166][ 810/ 1236] Overall Loss 0.129867 Objective Loss 0.129867 LR 0.000250 Time 0.022014 +2023-10-02 21:49:17,388 - Epoch: [166][ 820/ 1236] Overall Loss 0.129987 Objective Loss 0.129987 LR 0.000250 Time 0.021996 +2023-10-02 21:49:17,597 - Epoch: [166][ 830/ 1236] Overall Loss 0.130214 Objective Loss 0.130214 LR 0.000250 Time 0.021982 +2023-10-02 21:49:17,804 - Epoch: [166][ 840/ 1236] Overall Loss 0.130226 Objective Loss 0.130226 LR 0.000250 Time 0.021965 +2023-10-02 21:49:18,013 - Epoch: [166][ 850/ 1236] Overall Loss 0.130286 Objective Loss 0.130286 LR 0.000250 Time 0.021952 +2023-10-02 21:49:18,220 - Epoch: [166][ 860/ 1236] Overall Loss 0.130357 Objective Loss 0.130357 LR 0.000250 Time 0.021937 +2023-10-02 21:49:18,429 - Epoch: [166][ 870/ 1236] Overall Loss 0.130263 Objective Loss 0.130263 LR 0.000250 Time 0.021924 +2023-10-02 21:49:18,636 - Epoch: [166][ 880/ 1236] Overall Loss 0.130274 Objective Loss 0.130274 LR 0.000250 Time 0.021909 +2023-10-02 21:49:18,845 - Epoch: [166][ 890/ 1236] Overall Loss 0.130288 Objective Loss 0.130288 LR 0.000250 Time 0.021896 +2023-10-02 21:49:19,052 - Epoch: [166][ 900/ 1236] Overall Loss 0.130213 Objective Loss 0.130213 LR 0.000250 Time 0.021882 +2023-10-02 21:49:19,261 - Epoch: [166][ 910/ 1236] Overall Loss 0.130350 Objective Loss 0.130350 LR 0.000250 Time 0.021870 +2023-10-02 21:49:19,468 - Epoch: [166][ 920/ 1236] Overall Loss 0.130484 Objective Loss 0.130484 LR 0.000250 Time 0.021856 +2023-10-02 21:49:19,677 - Epoch: [166][ 930/ 1236] Overall Loss 0.130587 Objective Loss 0.130587 LR 0.000250 Time 0.021845 +2023-10-02 21:49:19,885 - Epoch: [166][ 940/ 1236] Overall Loss 0.130709 Objective Loss 0.130709 LR 0.000250 Time 0.021832 +2023-10-02 21:49:20,093 - Epoch: [166][ 950/ 1236] Overall Loss 0.130850 Objective Loss 0.130850 LR 0.000250 Time 0.021822 +2023-10-02 21:49:20,301 - Epoch: [166][ 960/ 1236] Overall Loss 0.130825 Objective Loss 0.130825 LR 0.000250 Time 0.021809 +2023-10-02 21:49:20,509 - Epoch: [166][ 970/ 1236] Overall Loss 0.130877 Objective Loss 0.130877 LR 0.000250 Time 0.021799 +2023-10-02 21:49:20,717 - Epoch: [166][ 980/ 1236] Overall Loss 0.130899 Objective Loss 0.130899 LR 0.000250 Time 0.021786 +2023-10-02 21:49:20,925 - Epoch: [166][ 990/ 1236] Overall Loss 0.130891 Objective Loss 0.130891 LR 0.000250 Time 0.021777 +2023-10-02 21:49:21,133 - Epoch: [166][ 1000/ 1236] Overall Loss 0.131011 Objective Loss 0.131011 LR 0.000250 Time 0.021765 +2023-10-02 21:49:21,342 - Epoch: [166][ 1010/ 1236] Overall Loss 0.131057 Objective Loss 0.131057 LR 0.000250 Time 0.021756 +2023-10-02 21:49:21,549 - Epoch: [166][ 1020/ 1236] Overall Loss 0.130928 Objective Loss 0.130928 LR 0.000250 Time 0.021744 +2023-10-02 21:49:21,758 - Epoch: [166][ 1030/ 1236] Overall Loss 0.130903 Objective Loss 0.130903 LR 0.000250 Time 0.021735 +2023-10-02 21:49:21,965 - Epoch: [166][ 1040/ 1236] Overall Loss 0.130821 Objective Loss 0.130821 LR 0.000250 Time 0.021724 +2023-10-02 21:49:22,174 - Epoch: [166][ 1050/ 1236] Overall Loss 0.130858 Objective Loss 0.130858 LR 0.000250 Time 0.021716 +2023-10-02 21:49:22,381 - Epoch: [166][ 1060/ 1236] Overall Loss 0.131034 Objective Loss 0.131034 LR 0.000250 Time 0.021705 +2023-10-02 21:49:22,590 - Epoch: [166][ 1070/ 1236] Overall Loss 0.130897 Objective Loss 0.130897 LR 0.000250 Time 0.021697 +2023-10-02 21:49:22,797 - Epoch: [166][ 1080/ 1236] Overall Loss 0.130826 Objective Loss 0.130826 LR 0.000250 Time 0.021687 +2023-10-02 21:49:23,006 - Epoch: [166][ 1090/ 1236] Overall Loss 0.130837 Objective Loss 0.130837 LR 0.000250 Time 0.021679 +2023-10-02 21:49:23,213 - Epoch: [166][ 1100/ 1236] Overall Loss 0.130734 Objective Loss 0.130734 LR 0.000250 Time 0.021669 +2023-10-02 21:49:23,422 - Epoch: [166][ 1110/ 1236] Overall Loss 0.130884 Objective Loss 0.130884 LR 0.000250 Time 0.021662 +2023-10-02 21:49:23,630 - Epoch: [166][ 1120/ 1236] Overall Loss 0.130851 Objective Loss 0.130851 LR 0.000250 Time 0.021652 +2023-10-02 21:49:23,838 - Epoch: [166][ 1130/ 1236] Overall Loss 0.130844 Objective Loss 0.130844 LR 0.000250 Time 0.021645 +2023-10-02 21:49:24,046 - Epoch: [166][ 1140/ 1236] Overall Loss 0.130696 Objective Loss 0.130696 LR 0.000250 Time 0.021636 +2023-10-02 21:49:24,254 - Epoch: [166][ 1150/ 1236] Overall Loss 0.130604 Objective Loss 0.130604 LR 0.000250 Time 0.021629 +2023-10-02 21:49:24,462 - Epoch: [166][ 1160/ 1236] Overall Loss 0.130673 Objective Loss 0.130673 LR 0.000250 Time 0.021619 +2023-10-02 21:49:24,670 - Epoch: [166][ 1170/ 1236] Overall Loss 0.130695 Objective Loss 0.130695 LR 0.000250 Time 0.021613 +2023-10-02 21:49:24,878 - Epoch: [166][ 1180/ 1236] Overall Loss 0.130766 Objective Loss 0.130766 LR 0.000250 Time 0.021604 +2023-10-02 21:49:25,087 - Epoch: [166][ 1190/ 1236] Overall Loss 0.130707 Objective Loss 0.130707 LR 0.000250 Time 0.021598 +2023-10-02 21:49:25,294 - Epoch: [166][ 1200/ 1236] Overall Loss 0.130805 Objective Loss 0.130805 LR 0.000250 Time 0.021589 +2023-10-02 21:49:25,503 - Epoch: [166][ 1210/ 1236] Overall Loss 0.130824 Objective Loss 0.130824 LR 0.000250 Time 0.021583 +2023-10-02 21:49:25,710 - Epoch: [166][ 1220/ 1236] Overall Loss 0.130781 Objective Loss 0.130781 LR 0.000250 Time 0.021575 +2023-10-02 21:49:25,970 - Epoch: [166][ 1230/ 1236] Overall Loss 0.130691 Objective Loss 0.130691 LR 0.000250 Time 0.021610 +2023-10-02 21:49:26,091 - Epoch: [166][ 1236/ 1236] Overall Loss 0.130787 Objective Loss 0.130787 Top1 88.594705 Top5 98.981670 LR 0.000250 Time 0.021603 +2023-10-02 21:49:26,224 - --- validate (epoch=166)----------- +2023-10-02 21:49:26,224 - 29943 samples (256 per mini-batch) +2023-10-02 21:49:26,724 - Epoch: [166][ 10/ 117] Loss 0.336978 Top1 86.406250 Top5 98.281250 +2023-10-02 21:49:26,874 - Epoch: [166][ 20/ 117] Loss 0.304665 Top1 86.953125 Top5 98.515625 +2023-10-02 21:49:27,026 - Epoch: [166][ 30/ 117] Loss 0.294673 Top1 87.174479 Top5 98.580729 +2023-10-02 21:49:27,177 - Epoch: [166][ 40/ 117] Loss 0.300918 Top1 87.031250 Top5 98.593750 +2023-10-02 21:49:27,327 - Epoch: [166][ 50/ 117] Loss 0.304308 Top1 86.828125 Top5 98.585938 +2023-10-02 21:49:27,478 - Epoch: [166][ 60/ 117] Loss 0.304999 Top1 86.881510 Top5 98.587240 +2023-10-02 21:49:27,629 - Epoch: [166][ 70/ 117] Loss 0.306266 Top1 86.763393 Top5 98.549107 +2023-10-02 21:49:27,781 - Epoch: [166][ 80/ 117] Loss 0.307615 Top1 86.796875 Top5 98.583984 +2023-10-02 21:49:27,932 - Epoch: [166][ 90/ 117] Loss 0.306782 Top1 86.827257 Top5 98.611111 +2023-10-02 21:49:28,083 - Epoch: [166][ 100/ 117] Loss 0.304718 Top1 86.902344 Top5 98.628906 +2023-10-02 21:49:28,240 - Epoch: [166][ 110/ 117] Loss 0.304097 Top1 86.860795 Top5 98.647017 +2023-10-02 21:49:28,328 - Epoch: [166][ 117/ 117] Loss 0.304354 Top1 86.935177 Top5 98.637411 +2023-10-02 21:49:28,467 - ==> Top1: 86.935 Top5: 98.637 Loss: 0.304 + +2023-10-02 21:49:28,468 - ==> Confusion: +[[ 942 1 3 1 11 3 0 0 6 49 3 1 0 1 5 1 4 0 1 0 18] + [ 0 1048 0 1 4 22 0 25 2 1 1 0 1 0 1 3 2 0 13 3 4] + [ 2 1 983 2 1 0 19 5 0 1 2 2 7 2 1 2 1 2 14 2 7] + [ 1 3 16 967 0 3 4 5 5 0 4 1 8 3 27 4 1 3 15 2 17] + [ 18 5 2 1 982 4 0 0 1 9 0 0 1 5 10 1 6 0 0 3 2] + [ 3 30 0 1 3 987 2 32 4 3 3 6 1 12 6 0 2 0 4 2 15] + [ 0 5 36 0 0 1 1120 2 0 0 3 1 0 1 0 5 0 1 2 10 4] + [ 2 12 13 1 3 22 2 1076 1 2 7 6 4 3 1 1 1 0 40 10 11] + [ 16 3 0 0 1 6 0 2 972 38 14 0 3 9 11 0 6 2 2 0 4] + [ 106 0 0 1 5 5 0 0 21 946 2 0 0 19 5 1 1 0 0 1 6] + [ 1 2 11 7 1 1 3 2 8 1 982 2 0 11 2 0 3 1 4 0 11] + [ 1 1 2 0 1 10 0 7 0 1 0 971 10 6 0 0 0 17 0 4 4] + [ 0 0 2 3 0 4 2 0 0 1 6 37 975 1 0 8 1 11 1 5 11] + [ 0 0 1 0 1 8 0 0 13 9 3 7 0 1054 4 0 0 1 0 1 17] + [ 10 1 5 14 9 0 0 0 23 3 3 0 3 3 1011 0 0 1 8 0 7] + [ 0 0 3 1 5 0 0 0 0 0 1 5 8 0 0 1069 18 12 1 6 5] + [ 1 15 0 0 5 8 1 0 0 0 0 3 0 3 3 6 1103 0 2 4 7] + [ 0 0 1 3 0 0 1 0 0 1 0 5 18 2 1 4 0 998 0 0 4] + [ 2 6 3 12 0 1 0 20 3 1 3 1 2 0 10 0 2 0 990 0 12] + [ 0 0 4 0 1 2 9 4 0 1 1 15 4 3 0 0 7 1 0 1096 4] + [ 110 118 114 57 52 116 32 90 66 57 165 95 290 260 105 44 91 47 99 138 5759]] + +2023-10-02 21:49:28,469 - ==> Best [Top1: 87.343 Top5: 98.694 Sparsity:0.00 Params: 169472 on epoch: 165] +2023-10-02 21:49:28,469 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:49:28,476 - + +2023-10-02 21:49:28,476 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:49:29,502 - Epoch: [167][ 10/ 1236] Overall Loss 0.114272 Objective Loss 0.114272 LR 0.000250 Time 0.102596 +2023-10-02 21:49:29,709 - Epoch: [167][ 20/ 1236] Overall Loss 0.122713 Objective Loss 0.122713 LR 0.000250 Time 0.061617 +2023-10-02 21:49:29,915 - Epoch: [167][ 30/ 1236] Overall Loss 0.125741 Objective Loss 0.125741 LR 0.000250 Time 0.047930 +2023-10-02 21:49:30,122 - Epoch: [167][ 40/ 1236] Overall Loss 0.131960 Objective Loss 0.131960 LR 0.000250 Time 0.041108 +2023-10-02 21:49:30,327 - Epoch: [167][ 50/ 1236] Overall Loss 0.129095 Objective Loss 0.129095 LR 0.000250 Time 0.036995 +2023-10-02 21:49:30,534 - Epoch: [167][ 60/ 1236] Overall Loss 0.129994 Objective Loss 0.129994 LR 0.000250 Time 0.034267 +2023-10-02 21:49:30,740 - Epoch: [167][ 70/ 1236] Overall Loss 0.127219 Objective Loss 0.127219 LR 0.000250 Time 0.032309 +2023-10-02 21:49:30,947 - Epoch: [167][ 80/ 1236] Overall Loss 0.127320 Objective Loss 0.127320 LR 0.000250 Time 0.030851 +2023-10-02 21:49:31,153 - Epoch: [167][ 90/ 1236] Overall Loss 0.126436 Objective Loss 0.126436 LR 0.000250 Time 0.029695 +2023-10-02 21:49:31,360 - Epoch: [167][ 100/ 1236] Overall Loss 0.126219 Objective Loss 0.126219 LR 0.000250 Time 0.028792 +2023-10-02 21:49:31,564 - Epoch: [167][ 110/ 1236] Overall Loss 0.126584 Objective Loss 0.126584 LR 0.000250 Time 0.028024 +2023-10-02 21:49:31,772 - Epoch: [167][ 120/ 1236] Overall Loss 0.126737 Objective Loss 0.126737 LR 0.000250 Time 0.027413 +2023-10-02 21:49:31,978 - Epoch: [167][ 130/ 1236] Overall Loss 0.127693 Objective Loss 0.127693 LR 0.000250 Time 0.026876 +2023-10-02 21:49:32,185 - Epoch: [167][ 140/ 1236] Overall Loss 0.128553 Objective Loss 0.128553 LR 0.000250 Time 0.026434 +2023-10-02 21:49:32,391 - Epoch: [167][ 150/ 1236] Overall Loss 0.127756 Objective Loss 0.127756 LR 0.000250 Time 0.026034 +2023-10-02 21:49:32,596 - Epoch: [167][ 160/ 1236] Overall Loss 0.127365 Objective Loss 0.127365 LR 0.000250 Time 0.025689 +2023-10-02 21:49:32,802 - Epoch: [167][ 170/ 1236] Overall Loss 0.128641 Objective Loss 0.128641 LR 0.000250 Time 0.025379 +2023-10-02 21:49:33,009 - Epoch: [167][ 180/ 1236] Overall Loss 0.129534 Objective Loss 0.129534 LR 0.000250 Time 0.025119 +2023-10-02 21:49:33,215 - Epoch: [167][ 190/ 1236] Overall Loss 0.130005 Objective Loss 0.130005 LR 0.000250 Time 0.024871 +2023-10-02 21:49:33,422 - Epoch: [167][ 200/ 1236] Overall Loss 0.130499 Objective Loss 0.130499 LR 0.000250 Time 0.024662 +2023-10-02 21:49:33,627 - Epoch: [167][ 210/ 1236] Overall Loss 0.130167 Objective Loss 0.130167 LR 0.000250 Time 0.024459 +2023-10-02 21:49:33,834 - Epoch: [167][ 220/ 1236] Overall Loss 0.130808 Objective Loss 0.130808 LR 0.000250 Time 0.024285 +2023-10-02 21:49:34,038 - Epoch: [167][ 230/ 1236] Overall Loss 0.130345 Objective Loss 0.130345 LR 0.000250 Time 0.024116 +2023-10-02 21:49:34,246 - Epoch: [167][ 240/ 1236] Overall Loss 0.130169 Objective Loss 0.130169 LR 0.000250 Time 0.023973 +2023-10-02 21:49:34,451 - Epoch: [167][ 250/ 1236] Overall Loss 0.130563 Objective Loss 0.130563 LR 0.000250 Time 0.023830 +2023-10-02 21:49:34,658 - Epoch: [167][ 260/ 1236] Overall Loss 0.130475 Objective Loss 0.130475 LR 0.000250 Time 0.023709 +2023-10-02 21:49:34,864 - Epoch: [167][ 270/ 1236] Overall Loss 0.130791 Objective Loss 0.130791 LR 0.000250 Time 0.023588 +2023-10-02 21:49:35,071 - Epoch: [167][ 280/ 1236] Overall Loss 0.130877 Objective Loss 0.130877 LR 0.000250 Time 0.023483 +2023-10-02 21:49:35,277 - Epoch: [167][ 290/ 1236] Overall Loss 0.131125 Objective Loss 0.131125 LR 0.000250 Time 0.023377 +2023-10-02 21:49:35,487 - Epoch: [167][ 300/ 1236] Overall Loss 0.131060 Objective Loss 0.131060 LR 0.000250 Time 0.023298 +2023-10-02 21:49:35,694 - Epoch: [167][ 310/ 1236] Overall Loss 0.131753 Objective Loss 0.131753 LR 0.000250 Time 0.023215 +2023-10-02 21:49:35,902 - Epoch: [167][ 320/ 1236] Overall Loss 0.131708 Objective Loss 0.131708 LR 0.000250 Time 0.023138 +2023-10-02 21:49:36,112 - Epoch: [167][ 330/ 1236] Overall Loss 0.131545 Objective Loss 0.131545 LR 0.000250 Time 0.023070 +2023-10-02 21:49:36,322 - Epoch: [167][ 340/ 1236] Overall Loss 0.131380 Objective Loss 0.131380 LR 0.000250 Time 0.023009 +2023-10-02 21:49:36,531 - Epoch: [167][ 350/ 1236] Overall Loss 0.131179 Objective Loss 0.131179 LR 0.000250 Time 0.022948 +2023-10-02 21:49:36,743 - Epoch: [167][ 360/ 1236] Overall Loss 0.131255 Objective Loss 0.131255 LR 0.000250 Time 0.022898 +2023-10-02 21:49:36,953 - Epoch: [167][ 370/ 1236] Overall Loss 0.131248 Objective Loss 0.131248 LR 0.000250 Time 0.022845 +2023-10-02 21:49:37,166 - Epoch: [167][ 380/ 1236] Overall Loss 0.131267 Objective Loss 0.131267 LR 0.000250 Time 0.022803 +2023-10-02 21:49:37,375 - Epoch: [167][ 390/ 1236] Overall Loss 0.131183 Objective Loss 0.131183 LR 0.000250 Time 0.022753 +2023-10-02 21:49:37,587 - Epoch: [167][ 400/ 1236] Overall Loss 0.131135 Objective Loss 0.131135 LR 0.000250 Time 0.022716 +2023-10-02 21:49:37,797 - Epoch: [167][ 410/ 1236] Overall Loss 0.131344 Objective Loss 0.131344 LR 0.000250 Time 0.022671 +2023-10-02 21:49:38,009 - Epoch: [167][ 420/ 1236] Overall Loss 0.131694 Objective Loss 0.131694 LR 0.000250 Time 0.022636 +2023-10-02 21:49:38,219 - Epoch: [167][ 430/ 1236] Overall Loss 0.131632 Objective Loss 0.131632 LR 0.000250 Time 0.022598 +2023-10-02 21:49:38,432 - Epoch: [167][ 440/ 1236] Overall Loss 0.131772 Objective Loss 0.131772 LR 0.000250 Time 0.022567 +2023-10-02 21:49:38,641 - Epoch: [167][ 450/ 1236] Overall Loss 0.131963 Objective Loss 0.131963 LR 0.000250 Time 0.022530 +2023-10-02 21:49:38,854 - Epoch: [167][ 460/ 1236] Overall Loss 0.131897 Objective Loss 0.131897 LR 0.000250 Time 0.022501 +2023-10-02 21:49:39,063 - Epoch: [167][ 470/ 1236] Overall Loss 0.131605 Objective Loss 0.131605 LR 0.000250 Time 0.022467 +2023-10-02 21:49:39,275 - Epoch: [167][ 480/ 1236] Overall Loss 0.131508 Objective Loss 0.131508 LR 0.000250 Time 0.022440 +2023-10-02 21:49:39,485 - Epoch: [167][ 490/ 1236] Overall Loss 0.131371 Objective Loss 0.131371 LR 0.000250 Time 0.022410 +2023-10-02 21:49:39,698 - Epoch: [167][ 500/ 1236] Overall Loss 0.131329 Objective Loss 0.131329 LR 0.000250 Time 0.022387 +2023-10-02 21:49:39,908 - Epoch: [167][ 510/ 1236] Overall Loss 0.131608 Objective Loss 0.131608 LR 0.000250 Time 0.022359 +2023-10-02 21:49:40,120 - Epoch: [167][ 520/ 1236] Overall Loss 0.131828 Objective Loss 0.131828 LR 0.000250 Time 0.022336 +2023-10-02 21:49:40,330 - Epoch: [167][ 530/ 1236] Overall Loss 0.131661 Objective Loss 0.131661 LR 0.000250 Time 0.022310 +2023-10-02 21:49:40,543 - Epoch: [167][ 540/ 1236] Overall Loss 0.131709 Objective Loss 0.131709 LR 0.000250 Time 0.022290 +2023-10-02 21:49:40,752 - Epoch: [167][ 550/ 1236] Overall Loss 0.131769 Objective Loss 0.131769 LR 0.000250 Time 0.022265 +2023-10-02 21:49:40,965 - Epoch: [167][ 560/ 1236] Overall Loss 0.131940 Objective Loss 0.131940 LR 0.000250 Time 0.022246 +2023-10-02 21:49:41,175 - Epoch: [167][ 570/ 1236] Overall Loss 0.131799 Objective Loss 0.131799 LR 0.000250 Time 0.022224 +2023-10-02 21:49:41,388 - Epoch: [167][ 580/ 1236] Overall Loss 0.131698 Objective Loss 0.131698 LR 0.000250 Time 0.022207 +2023-10-02 21:49:41,597 - Epoch: [167][ 590/ 1236] Overall Loss 0.131299 Objective Loss 0.131299 LR 0.000250 Time 0.022185 +2023-10-02 21:49:41,809 - Epoch: [167][ 600/ 1236] Overall Loss 0.131515 Objective Loss 0.131515 LR 0.000250 Time 0.022168 +2023-10-02 21:49:42,019 - Epoch: [167][ 610/ 1236] Overall Loss 0.131629 Objective Loss 0.131629 LR 0.000250 Time 0.022148 +2023-10-02 21:49:42,232 - Epoch: [167][ 620/ 1236] Overall Loss 0.131614 Objective Loss 0.131614 LR 0.000250 Time 0.022133 +2023-10-02 21:49:42,441 - Epoch: [167][ 630/ 1236] Overall Loss 0.131784 Objective Loss 0.131784 LR 0.000250 Time 0.022113 +2023-10-02 21:49:42,654 - Epoch: [167][ 640/ 1236] Overall Loss 0.131751 Objective Loss 0.131751 LR 0.000250 Time 0.022101 +2023-10-02 21:49:42,864 - Epoch: [167][ 650/ 1236] Overall Loss 0.131628 Objective Loss 0.131628 LR 0.000250 Time 0.022083 +2023-10-02 21:49:43,076 - Epoch: [167][ 660/ 1236] Overall Loss 0.131565 Objective Loss 0.131565 LR 0.000250 Time 0.022069 +2023-10-02 21:49:43,287 - Epoch: [167][ 670/ 1236] Overall Loss 0.131503 Objective Loss 0.131503 LR 0.000250 Time 0.022054 +2023-10-02 21:49:43,500 - Epoch: [167][ 680/ 1236] Overall Loss 0.131277 Objective Loss 0.131277 LR 0.000250 Time 0.022042 +2023-10-02 21:49:43,709 - Epoch: [167][ 690/ 1236] Overall Loss 0.131116 Objective Loss 0.131116 LR 0.000250 Time 0.022026 +2023-10-02 21:49:43,922 - Epoch: [167][ 700/ 1236] Overall Loss 0.130955 Objective Loss 0.130955 LR 0.000250 Time 0.022013 +2023-10-02 21:49:44,132 - Epoch: [167][ 710/ 1236] Overall Loss 0.131001 Objective Loss 0.131001 LR 0.000250 Time 0.021999 +2023-10-02 21:49:44,343 - Epoch: [167][ 720/ 1236] Overall Loss 0.131156 Objective Loss 0.131156 LR 0.000250 Time 0.021987 +2023-10-02 21:49:44,553 - Epoch: [167][ 730/ 1236] Overall Loss 0.131353 Objective Loss 0.131353 LR 0.000250 Time 0.021973 +2023-10-02 21:49:44,766 - Epoch: [167][ 740/ 1236] Overall Loss 0.131423 Objective Loss 0.131423 LR 0.000250 Time 0.021963 +2023-10-02 21:49:44,975 - Epoch: [167][ 750/ 1236] Overall Loss 0.131383 Objective Loss 0.131383 LR 0.000250 Time 0.021949 +2023-10-02 21:49:45,189 - Epoch: [167][ 760/ 1236] Overall Loss 0.131428 Objective Loss 0.131428 LR 0.000250 Time 0.021940 +2023-10-02 21:49:45,398 - Epoch: [167][ 770/ 1236] Overall Loss 0.131128 Objective Loss 0.131128 LR 0.000250 Time 0.021927 +2023-10-02 21:49:45,612 - Epoch: [167][ 780/ 1236] Overall Loss 0.131261 Objective Loss 0.131261 LR 0.000250 Time 0.021919 +2023-10-02 21:49:45,821 - Epoch: [167][ 790/ 1236] Overall Loss 0.131074 Objective Loss 0.131074 LR 0.000250 Time 0.021906 +2023-10-02 21:49:46,033 - Epoch: [167][ 800/ 1236] Overall Loss 0.130968 Objective Loss 0.130968 LR 0.000250 Time 0.021897 +2023-10-02 21:49:46,243 - Epoch: [167][ 810/ 1236] Overall Loss 0.131058 Objective Loss 0.131058 LR 0.000250 Time 0.021885 +2023-10-02 21:49:46,455 - Epoch: [167][ 820/ 1236] Overall Loss 0.131025 Objective Loss 0.131025 LR 0.000250 Time 0.021876 +2023-10-02 21:49:46,664 - Epoch: [167][ 830/ 1236] Overall Loss 0.130903 Objective Loss 0.130903 LR 0.000250 Time 0.021864 +2023-10-02 21:49:46,877 - Epoch: [167][ 840/ 1236] Overall Loss 0.131003 Objective Loss 0.131003 LR 0.000250 Time 0.021857 +2023-10-02 21:49:47,087 - Epoch: [167][ 850/ 1236] Overall Loss 0.130916 Objective Loss 0.130916 LR 0.000250 Time 0.021846 +2023-10-02 21:49:47,299 - Epoch: [167][ 860/ 1236] Overall Loss 0.130998 Objective Loss 0.130998 LR 0.000250 Time 0.021838 +2023-10-02 21:49:47,508 - Epoch: [167][ 870/ 1236] Overall Loss 0.130960 Objective Loss 0.130960 LR 0.000250 Time 0.021828 +2023-10-02 21:49:47,721 - Epoch: [167][ 880/ 1236] Overall Loss 0.130739 Objective Loss 0.130739 LR 0.000250 Time 0.021820 +2023-10-02 21:49:47,930 - Epoch: [167][ 890/ 1236] Overall Loss 0.130658 Objective Loss 0.130658 LR 0.000250 Time 0.021810 +2023-10-02 21:49:48,143 - Epoch: [167][ 900/ 1236] Overall Loss 0.130667 Objective Loss 0.130667 LR 0.000250 Time 0.021803 +2023-10-02 21:49:48,352 - Epoch: [167][ 910/ 1236] Overall Loss 0.130589 Objective Loss 0.130589 LR 0.000250 Time 0.021794 +2023-10-02 21:49:48,565 - Epoch: [167][ 920/ 1236] Overall Loss 0.130460 Objective Loss 0.130460 LR 0.000250 Time 0.021788 +2023-10-02 21:49:48,775 - Epoch: [167][ 930/ 1236] Overall Loss 0.130340 Objective Loss 0.130340 LR 0.000250 Time 0.021778 +2023-10-02 21:49:48,988 - Epoch: [167][ 940/ 1236] Overall Loss 0.130562 Objective Loss 0.130562 LR 0.000250 Time 0.021773 +2023-10-02 21:49:49,197 - Epoch: [167][ 950/ 1236] Overall Loss 0.130526 Objective Loss 0.130526 LR 0.000250 Time 0.021764 +2023-10-02 21:49:49,411 - Epoch: [167][ 960/ 1236] Overall Loss 0.130468 Objective Loss 0.130468 LR 0.000250 Time 0.021759 +2023-10-02 21:49:49,620 - Epoch: [167][ 970/ 1236] Overall Loss 0.130556 Objective Loss 0.130556 LR 0.000250 Time 0.021751 +2023-10-02 21:49:49,833 - Epoch: [167][ 980/ 1236] Overall Loss 0.130595 Objective Loss 0.130595 LR 0.000250 Time 0.021745 +2023-10-02 21:49:50,042 - Epoch: [167][ 990/ 1236] Overall Loss 0.130516 Objective Loss 0.130516 LR 0.000250 Time 0.021737 +2023-10-02 21:49:50,255 - Epoch: [167][ 1000/ 1236] Overall Loss 0.130436 Objective Loss 0.130436 LR 0.000250 Time 0.021732 +2023-10-02 21:49:50,465 - Epoch: [167][ 1010/ 1236] Overall Loss 0.130397 Objective Loss 0.130397 LR 0.000250 Time 0.021724 +2023-10-02 21:49:50,678 - Epoch: [167][ 1020/ 1236] Overall Loss 0.130402 Objective Loss 0.130402 LR 0.000250 Time 0.021719 +2023-10-02 21:49:50,887 - Epoch: [167][ 1030/ 1236] Overall Loss 0.130614 Objective Loss 0.130614 LR 0.000250 Time 0.021711 +2023-10-02 21:49:51,099 - Epoch: [167][ 1040/ 1236] Overall Loss 0.130417 Objective Loss 0.130417 LR 0.000250 Time 0.021706 +2023-10-02 21:49:51,308 - Epoch: [167][ 1050/ 1236] Overall Loss 0.130366 Objective Loss 0.130366 LR 0.000250 Time 0.021698 +2023-10-02 21:49:51,522 - Epoch: [167][ 1060/ 1236] Overall Loss 0.130312 Objective Loss 0.130312 LR 0.000250 Time 0.021695 +2023-10-02 21:49:51,731 - Epoch: [167][ 1070/ 1236] Overall Loss 0.130408 Objective Loss 0.130408 LR 0.000250 Time 0.021687 +2023-10-02 21:49:51,943 - Epoch: [167][ 1080/ 1236] Overall Loss 0.130321 Objective Loss 0.130321 LR 0.000250 Time 0.021682 +2023-10-02 21:49:52,153 - Epoch: [167][ 1090/ 1236] Overall Loss 0.130390 Objective Loss 0.130390 LR 0.000250 Time 0.021676 +2023-10-02 21:49:52,366 - Epoch: [167][ 1100/ 1236] Overall Loss 0.130478 Objective Loss 0.130478 LR 0.000250 Time 0.021672 +2023-10-02 21:49:52,575 - Epoch: [167][ 1110/ 1236] Overall Loss 0.130430 Objective Loss 0.130430 LR 0.000250 Time 0.021665 +2023-10-02 21:49:52,788 - Epoch: [167][ 1120/ 1236] Overall Loss 0.130571 Objective Loss 0.130571 LR 0.000250 Time 0.021661 +2023-10-02 21:49:52,997 - Epoch: [167][ 1130/ 1236] Overall Loss 0.130582 Objective Loss 0.130582 LR 0.000250 Time 0.021654 +2023-10-02 21:49:53,210 - Epoch: [167][ 1140/ 1236] Overall Loss 0.130552 Objective Loss 0.130552 LR 0.000250 Time 0.021650 +2023-10-02 21:49:53,420 - Epoch: [167][ 1150/ 1236] Overall Loss 0.130430 Objective Loss 0.130430 LR 0.000250 Time 0.021644 +2023-10-02 21:49:53,632 - Epoch: [167][ 1160/ 1236] Overall Loss 0.130485 Objective Loss 0.130485 LR 0.000250 Time 0.021640 +2023-10-02 21:49:53,842 - Epoch: [167][ 1170/ 1236] Overall Loss 0.130530 Objective Loss 0.130530 LR 0.000250 Time 0.021635 +2023-10-02 21:49:54,054 - Epoch: [167][ 1180/ 1236] Overall Loss 0.130433 Objective Loss 0.130433 LR 0.000250 Time 0.021631 +2023-10-02 21:49:54,264 - Epoch: [167][ 1190/ 1236] Overall Loss 0.130514 Objective Loss 0.130514 LR 0.000250 Time 0.021625 +2023-10-02 21:49:54,478 - Epoch: [167][ 1200/ 1236] Overall Loss 0.130592 Objective Loss 0.130592 LR 0.000250 Time 0.021623 +2023-10-02 21:49:54,687 - Epoch: [167][ 1210/ 1236] Overall Loss 0.130500 Objective Loss 0.130500 LR 0.000250 Time 0.021617 +2023-10-02 21:49:54,900 - Epoch: [167][ 1220/ 1236] Overall Loss 0.130352 Objective Loss 0.130352 LR 0.000250 Time 0.021614 +2023-10-02 21:49:55,165 - Epoch: [167][ 1230/ 1236] Overall Loss 0.130281 Objective Loss 0.130281 LR 0.000250 Time 0.021653 +2023-10-02 21:49:55,288 - Epoch: [167][ 1236/ 1236] Overall Loss 0.130180 Objective Loss 0.130180 Top1 91.649695 Top5 98.778004 LR 0.000250 Time 0.021647 +2023-10-02 21:49:55,427 - --- validate (epoch=167)----------- +2023-10-02 21:49:55,428 - 29943 samples (256 per mini-batch) +2023-10-02 21:49:55,925 - Epoch: [167][ 10/ 117] Loss 0.304457 Top1 86.679688 Top5 98.671875 +2023-10-02 21:49:56,078 - Epoch: [167][ 20/ 117] Loss 0.304546 Top1 87.382812 Top5 98.632812 +2023-10-02 21:49:56,230 - Epoch: [167][ 30/ 117] Loss 0.305349 Top1 87.330729 Top5 98.658854 +2023-10-02 21:49:56,382 - Epoch: [167][ 40/ 117] Loss 0.312970 Top1 87.246094 Top5 98.632812 +2023-10-02 21:49:56,535 - Epoch: [167][ 50/ 117] Loss 0.316527 Top1 87.375000 Top5 98.617188 +2023-10-02 21:49:56,687 - Epoch: [167][ 60/ 117] Loss 0.313430 Top1 87.376302 Top5 98.665365 +2023-10-02 21:49:56,839 - Epoch: [167][ 70/ 117] Loss 0.313542 Top1 87.399554 Top5 98.671875 +2023-10-02 21:49:56,991 - Epoch: [167][ 80/ 117] Loss 0.310205 Top1 87.456055 Top5 98.701172 +2023-10-02 21:49:57,145 - Epoch: [167][ 90/ 117] Loss 0.304927 Top1 87.560764 Top5 98.710938 +2023-10-02 21:49:57,298 - Epoch: [167][ 100/ 117] Loss 0.307327 Top1 87.539062 Top5 98.722656 +2023-10-02 21:49:57,458 - Epoch: [167][ 110/ 117] Loss 0.307881 Top1 87.539062 Top5 98.714489 +2023-10-02 21:49:57,548 - Epoch: [167][ 117/ 117] Loss 0.306312 Top1 87.542998 Top5 98.737601 +2023-10-02 21:49:57,648 - ==> Top1: 87.543 Top5: 98.738 Loss: 0.306 + +2023-10-02 21:49:57,649 - ==> Confusion: +[[ 951 0 1 1 4 2 0 0 3 53 1 0 1 4 3 1 2 1 0 0 22] + [ 0 1063 1 0 3 19 1 18 0 0 0 0 1 0 1 3 1 0 8 4 8] + [ 2 0 984 8 1 0 10 6 0 2 1 2 6 2 1 4 1 2 10 2 12] + [ 0 2 12 977 0 2 1 5 2 1 1 1 5 4 38 3 2 3 10 1 19] + [ 29 2 0 1 974 5 0 0 1 9 0 0 0 3 8 5 5 0 1 1 6] + [ 2 33 0 2 4 1011 1 13 0 3 1 4 0 11 5 1 1 2 3 2 17] + [ 0 4 35 2 0 3 1118 5 0 0 4 0 0 0 0 5 0 1 1 6 7] + [ 2 10 14 2 5 33 5 1073 1 3 3 3 3 5 1 1 0 0 37 7 10] + [ 18 1 0 1 3 3 0 1 972 42 7 1 1 14 13 1 4 1 2 0 4] + [ 102 1 1 1 6 2 1 0 21 953 1 0 0 16 5 0 1 0 0 1 7] + [ 3 2 13 9 0 2 2 3 11 1 971 0 0 10 4 0 2 0 4 1 15] + [ 0 0 2 0 0 18 0 3 0 0 0 964 18 5 0 3 1 12 0 4 5] + [ 0 0 0 2 1 2 2 0 0 0 4 28 989 1 3 5 2 10 1 5 13] + [ 1 0 0 0 2 9 0 0 8 12 1 6 0 1054 4 1 0 1 0 3 17] + [ 10 0 5 12 3 0 0 0 16 3 2 0 1 1 1028 0 0 2 7 0 11] + [ 0 0 2 1 5 0 0 1 0 0 0 3 7 0 0 1077 13 11 2 7 5] + [ 0 17 0 1 8 5 1 0 0 0 0 3 1 2 4 7 1094 0 1 3 14] + [ 0 0 1 2 0 0 2 0 0 0 0 8 26 3 1 6 0 985 0 1 3] + [ 1 5 1 17 1 1 0 23 3 2 2 2 0 0 10 0 0 0 987 0 13] + [ 0 0 5 2 1 3 7 7 0 1 0 16 5 3 0 1 5 0 0 1090 6] + [ 107 106 95 65 60 108 22 86 55 61 114 88 313 225 109 54 74 56 75 134 5898]] + +2023-10-02 21:49:57,650 - ==> Best [Top1: 87.543 Top5: 98.738 Sparsity:0.00 Params: 169472 on epoch: 167] +2023-10-02 21:49:57,651 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:49:57,658 - + +2023-10-02 21:49:57,658 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:49:58,684 - Epoch: [168][ 10/ 1236] Overall Loss 0.144930 Objective Loss 0.144930 LR 0.000250 Time 0.102555 +2023-10-02 21:49:58,892 - Epoch: [168][ 20/ 1236] Overall Loss 0.138952 Objective Loss 0.138952 LR 0.000250 Time 0.061671 +2023-10-02 21:49:59,101 - Epoch: [168][ 30/ 1236] Overall Loss 0.141766 Objective Loss 0.141766 LR 0.000250 Time 0.048041 +2023-10-02 21:49:59,309 - Epoch: [168][ 40/ 1236] Overall Loss 0.140719 Objective Loss 0.140719 LR 0.000250 Time 0.041228 +2023-10-02 21:49:59,517 - Epoch: [168][ 50/ 1236] Overall Loss 0.138868 Objective Loss 0.138868 LR 0.000250 Time 0.037132 +2023-10-02 21:49:59,724 - Epoch: [168][ 60/ 1236] Overall Loss 0.135095 Objective Loss 0.135095 LR 0.000250 Time 0.034393 +2023-10-02 21:49:59,932 - Epoch: [168][ 70/ 1236] Overall Loss 0.135610 Objective Loss 0.135610 LR 0.000250 Time 0.032425 +2023-10-02 21:50:00,140 - Epoch: [168][ 80/ 1236] Overall Loss 0.135186 Objective Loss 0.135186 LR 0.000250 Time 0.030969 +2023-10-02 21:50:00,348 - Epoch: [168][ 90/ 1236] Overall Loss 0.133044 Objective Loss 0.133044 LR 0.000250 Time 0.029834 +2023-10-02 21:50:00,556 - Epoch: [168][ 100/ 1236] Overall Loss 0.133803 Objective Loss 0.133803 LR 0.000250 Time 0.028932 +2023-10-02 21:50:00,764 - Epoch: [168][ 110/ 1236] Overall Loss 0.133280 Objective Loss 0.133280 LR 0.000250 Time 0.028189 +2023-10-02 21:50:00,972 - Epoch: [168][ 120/ 1236] Overall Loss 0.133357 Objective Loss 0.133357 LR 0.000250 Time 0.027572 +2023-10-02 21:50:01,180 - Epoch: [168][ 130/ 1236] Overall Loss 0.134880 Objective Loss 0.134880 LR 0.000250 Time 0.027039 +2023-10-02 21:50:01,389 - Epoch: [168][ 140/ 1236] Overall Loss 0.133538 Objective Loss 0.133538 LR 0.000250 Time 0.026595 +2023-10-02 21:50:01,596 - Epoch: [168][ 150/ 1236] Overall Loss 0.132932 Objective Loss 0.132932 LR 0.000250 Time 0.026194 +2023-10-02 21:50:01,805 - Epoch: [168][ 160/ 1236] Overall Loss 0.133155 Objective Loss 0.133155 LR 0.000250 Time 0.025858 +2023-10-02 21:50:02,011 - Epoch: [168][ 170/ 1236] Overall Loss 0.133198 Objective Loss 0.133198 LR 0.000250 Time 0.025547 +2023-10-02 21:50:02,219 - Epoch: [168][ 180/ 1236] Overall Loss 0.132499 Objective Loss 0.132499 LR 0.000250 Time 0.025280 +2023-10-02 21:50:02,427 - Epoch: [168][ 190/ 1236] Overall Loss 0.131827 Objective Loss 0.131827 LR 0.000250 Time 0.025042 +2023-10-02 21:50:02,635 - Epoch: [168][ 200/ 1236] Overall Loss 0.130581 Objective Loss 0.130581 LR 0.000250 Time 0.024831 +2023-10-02 21:50:02,843 - Epoch: [168][ 210/ 1236] Overall Loss 0.131002 Objective Loss 0.131002 LR 0.000250 Time 0.024630 +2023-10-02 21:50:03,051 - Epoch: [168][ 220/ 1236] Overall Loss 0.129926 Objective Loss 0.129926 LR 0.000250 Time 0.024452 +2023-10-02 21:50:03,258 - Epoch: [168][ 230/ 1236] Overall Loss 0.128950 Objective Loss 0.128950 LR 0.000250 Time 0.024291 +2023-10-02 21:50:03,467 - Epoch: [168][ 240/ 1236] Overall Loss 0.128320 Objective Loss 0.128320 LR 0.000250 Time 0.024149 +2023-10-02 21:50:03,675 - Epoch: [168][ 250/ 1236] Overall Loss 0.128442 Objective Loss 0.128442 LR 0.000250 Time 0.024006 +2023-10-02 21:50:03,883 - Epoch: [168][ 260/ 1236] Overall Loss 0.128831 Objective Loss 0.128831 LR 0.000250 Time 0.023884 +2023-10-02 21:50:04,091 - Epoch: [168][ 270/ 1236] Overall Loss 0.129665 Objective Loss 0.129665 LR 0.000250 Time 0.023762 +2023-10-02 21:50:04,300 - Epoch: [168][ 280/ 1236] Overall Loss 0.129733 Objective Loss 0.129733 LR 0.000250 Time 0.023658 +2023-10-02 21:50:04,507 - Epoch: [168][ 290/ 1236] Overall Loss 0.129904 Objective Loss 0.129904 LR 0.000250 Time 0.023553 +2023-10-02 21:50:04,716 - Epoch: [168][ 300/ 1236] Overall Loss 0.130615 Objective Loss 0.130615 LR 0.000250 Time 0.023463 +2023-10-02 21:50:04,923 - Epoch: [168][ 310/ 1236] Overall Loss 0.130861 Objective Loss 0.130861 LR 0.000250 Time 0.023369 +2023-10-02 21:50:05,132 - Epoch: [168][ 320/ 1236] Overall Loss 0.130821 Objective Loss 0.130821 LR 0.000250 Time 0.023291 +2023-10-02 21:50:05,340 - Epoch: [168][ 330/ 1236] Overall Loss 0.131524 Objective Loss 0.131524 LR 0.000250 Time 0.023210 +2023-10-02 21:50:05,547 - Epoch: [168][ 340/ 1236] Overall Loss 0.131760 Objective Loss 0.131760 LR 0.000250 Time 0.023136 +2023-10-02 21:50:05,755 - Epoch: [168][ 350/ 1236] Overall Loss 0.132072 Objective Loss 0.132072 LR 0.000250 Time 0.023063 +2023-10-02 21:50:05,964 - Epoch: [168][ 360/ 1236] Overall Loss 0.131778 Objective Loss 0.131778 LR 0.000250 Time 0.023002 +2023-10-02 21:50:06,172 - Epoch: [168][ 370/ 1236] Overall Loss 0.131405 Objective Loss 0.131405 LR 0.000250 Time 0.022941 +2023-10-02 21:50:06,381 - Epoch: [168][ 380/ 1236] Overall Loss 0.131253 Objective Loss 0.131253 LR 0.000250 Time 0.022886 +2023-10-02 21:50:06,589 - Epoch: [168][ 390/ 1236] Overall Loss 0.131439 Objective Loss 0.131439 LR 0.000250 Time 0.022832 +2023-10-02 21:50:06,798 - Epoch: [168][ 400/ 1236] Overall Loss 0.131166 Objective Loss 0.131166 LR 0.000250 Time 0.022783 +2023-10-02 21:50:07,006 - Epoch: [168][ 410/ 1236] Overall Loss 0.131741 Objective Loss 0.131741 LR 0.000250 Time 0.022731 +2023-10-02 21:50:07,214 - Epoch: [168][ 420/ 1236] Overall Loss 0.131717 Objective Loss 0.131717 LR 0.000250 Time 0.022686 +2023-10-02 21:50:07,422 - Epoch: [168][ 430/ 1236] Overall Loss 0.131997 Objective Loss 0.131997 LR 0.000250 Time 0.022638 +2023-10-02 21:50:07,631 - Epoch: [168][ 440/ 1236] Overall Loss 0.131723 Objective Loss 0.131723 LR 0.000250 Time 0.022597 +2023-10-02 21:50:07,839 - Epoch: [168][ 450/ 1236] Overall Loss 0.131580 Objective Loss 0.131580 LR 0.000250 Time 0.022557 +2023-10-02 21:50:08,048 - Epoch: [168][ 460/ 1236] Overall Loss 0.131987 Objective Loss 0.131987 LR 0.000250 Time 0.022519 +2023-10-02 21:50:08,256 - Epoch: [168][ 470/ 1236] Overall Loss 0.132116 Objective Loss 0.132116 LR 0.000250 Time 0.022482 +2023-10-02 21:50:08,465 - Epoch: [168][ 480/ 1236] Overall Loss 0.132193 Objective Loss 0.132193 LR 0.000250 Time 0.022448 +2023-10-02 21:50:08,672 - Epoch: [168][ 490/ 1236] Overall Loss 0.132495 Objective Loss 0.132495 LR 0.000250 Time 0.022413 +2023-10-02 21:50:08,881 - Epoch: [168][ 500/ 1236] Overall Loss 0.133042 Objective Loss 0.133042 LR 0.000250 Time 0.022382 +2023-10-02 21:50:09,089 - Epoch: [168][ 510/ 1236] Overall Loss 0.133258 Objective Loss 0.133258 LR 0.000250 Time 0.022348 +2023-10-02 21:50:09,298 - Epoch: [168][ 520/ 1236] Overall Loss 0.133340 Objective Loss 0.133340 LR 0.000250 Time 0.022320 +2023-10-02 21:50:09,506 - Epoch: [168][ 530/ 1236] Overall Loss 0.133190 Objective Loss 0.133190 LR 0.000250 Time 0.022288 +2023-10-02 21:50:09,715 - Epoch: [168][ 540/ 1236] Overall Loss 0.132851 Objective Loss 0.132851 LR 0.000250 Time 0.022261 +2023-10-02 21:50:09,923 - Epoch: [168][ 550/ 1236] Overall Loss 0.132500 Objective Loss 0.132500 LR 0.000250 Time 0.022231 +2023-10-02 21:50:10,132 - Epoch: [168][ 560/ 1236] Overall Loss 0.132275 Objective Loss 0.132275 LR 0.000250 Time 0.022207 +2023-10-02 21:50:10,340 - Epoch: [168][ 570/ 1236] Overall Loss 0.132351 Objective Loss 0.132351 LR 0.000250 Time 0.022180 +2023-10-02 21:50:10,549 - Epoch: [168][ 580/ 1236] Overall Loss 0.132170 Objective Loss 0.132170 LR 0.000250 Time 0.022158 +2023-10-02 21:50:10,756 - Epoch: [168][ 590/ 1236] Overall Loss 0.131848 Objective Loss 0.131848 LR 0.000250 Time 0.022129 +2023-10-02 21:50:10,965 - Epoch: [168][ 600/ 1236] Overall Loss 0.131786 Objective Loss 0.131786 LR 0.000250 Time 0.022108 +2023-10-02 21:50:11,171 - Epoch: [168][ 610/ 1236] Overall Loss 0.132007 Objective Loss 0.132007 LR 0.000250 Time 0.022081 +2023-10-02 21:50:11,381 - Epoch: [168][ 620/ 1236] Overall Loss 0.131888 Objective Loss 0.131888 LR 0.000250 Time 0.022063 +2023-10-02 21:50:11,587 - Epoch: [168][ 630/ 1236] Overall Loss 0.131813 Objective Loss 0.131813 LR 0.000250 Time 0.022038 +2023-10-02 21:50:11,796 - Epoch: [168][ 640/ 1236] Overall Loss 0.131915 Objective Loss 0.131915 LR 0.000250 Time 0.022020 +2023-10-02 21:50:12,004 - Epoch: [168][ 650/ 1236] Overall Loss 0.131652 Objective Loss 0.131652 LR 0.000250 Time 0.021999 +2023-10-02 21:50:12,213 - Epoch: [168][ 660/ 1236] Overall Loss 0.131770 Objective Loss 0.131770 LR 0.000250 Time 0.021982 +2023-10-02 21:50:12,422 - Epoch: [168][ 670/ 1236] Overall Loss 0.132108 Objective Loss 0.132108 LR 0.000250 Time 0.021962 +2023-10-02 21:50:12,630 - Epoch: [168][ 680/ 1236] Overall Loss 0.132065 Objective Loss 0.132065 LR 0.000250 Time 0.021945 +2023-10-02 21:50:12,838 - Epoch: [168][ 690/ 1236] Overall Loss 0.132079 Objective Loss 0.132079 LR 0.000250 Time 0.021926 +2023-10-02 21:50:13,047 - Epoch: [168][ 700/ 1236] Overall Loss 0.132231 Objective Loss 0.132231 LR 0.000250 Time 0.021911 +2023-10-02 21:50:13,254 - Epoch: [168][ 710/ 1236] Overall Loss 0.132214 Objective Loss 0.132214 LR 0.000250 Time 0.021893 +2023-10-02 21:50:13,463 - Epoch: [168][ 720/ 1236] Overall Loss 0.132420 Objective Loss 0.132420 LR 0.000250 Time 0.021879 +2023-10-02 21:50:13,671 - Epoch: [168][ 730/ 1236] Overall Loss 0.132384 Objective Loss 0.132384 LR 0.000250 Time 0.021862 +2023-10-02 21:50:13,878 - Epoch: [168][ 740/ 1236] Overall Loss 0.132354 Objective Loss 0.132354 LR 0.000250 Time 0.021847 +2023-10-02 21:50:14,086 - Epoch: [168][ 750/ 1236] Overall Loss 0.132336 Objective Loss 0.132336 LR 0.000250 Time 0.021832 +2023-10-02 21:50:14,295 - Epoch: [168][ 760/ 1236] Overall Loss 0.132322 Objective Loss 0.132322 LR 0.000250 Time 0.021819 +2023-10-02 21:50:14,503 - Epoch: [168][ 770/ 1236] Overall Loss 0.132062 Objective Loss 0.132062 LR 0.000250 Time 0.021804 +2023-10-02 21:50:14,713 - Epoch: [168][ 780/ 1236] Overall Loss 0.131903 Objective Loss 0.131903 LR 0.000250 Time 0.021793 +2023-10-02 21:50:14,921 - Epoch: [168][ 790/ 1236] Overall Loss 0.131987 Objective Loss 0.131987 LR 0.000250 Time 0.021778 +2023-10-02 21:50:15,130 - Epoch: [168][ 800/ 1236] Overall Loss 0.132170 Objective Loss 0.132170 LR 0.000250 Time 0.021767 +2023-10-02 21:50:15,338 - Epoch: [168][ 810/ 1236] Overall Loss 0.132336 Objective Loss 0.132336 LR 0.000250 Time 0.021753 +2023-10-02 21:50:15,547 - Epoch: [168][ 820/ 1236] Overall Loss 0.132265 Objective Loss 0.132265 LR 0.000250 Time 0.021742 +2023-10-02 21:50:15,755 - Epoch: [168][ 830/ 1236] Overall Loss 0.132340 Objective Loss 0.132340 LR 0.000250 Time 0.021729 +2023-10-02 21:50:15,964 - Epoch: [168][ 840/ 1236] Overall Loss 0.132336 Objective Loss 0.132336 LR 0.000250 Time 0.021718 +2023-10-02 21:50:16,172 - Epoch: [168][ 850/ 1236] Overall Loss 0.132240 Objective Loss 0.132240 LR 0.000250 Time 0.021706 +2023-10-02 21:50:16,381 - Epoch: [168][ 860/ 1236] Overall Loss 0.132275 Objective Loss 0.132275 LR 0.000250 Time 0.021696 +2023-10-02 21:50:16,589 - Epoch: [168][ 870/ 1236] Overall Loss 0.132356 Objective Loss 0.132356 LR 0.000250 Time 0.021684 +2023-10-02 21:50:16,797 - Epoch: [168][ 880/ 1236] Overall Loss 0.132340 Objective Loss 0.132340 LR 0.000250 Time 0.021674 +2023-10-02 21:50:17,005 - Epoch: [168][ 890/ 1236] Overall Loss 0.132446 Objective Loss 0.132446 LR 0.000250 Time 0.021663 +2023-10-02 21:50:17,215 - Epoch: [168][ 900/ 1236] Overall Loss 0.132386 Objective Loss 0.132386 LR 0.000250 Time 0.021654 +2023-10-02 21:50:17,423 - Epoch: [168][ 910/ 1236] Overall Loss 0.132185 Objective Loss 0.132185 LR 0.000250 Time 0.021643 +2023-10-02 21:50:17,632 - Epoch: [168][ 920/ 1236] Overall Loss 0.131986 Objective Loss 0.131986 LR 0.000250 Time 0.021634 +2023-10-02 21:50:17,840 - Epoch: [168][ 930/ 1236] Overall Loss 0.132000 Objective Loss 0.132000 LR 0.000250 Time 0.021624 +2023-10-02 21:50:18,049 - Epoch: [168][ 940/ 1236] Overall Loss 0.131916 Objective Loss 0.131916 LR 0.000250 Time 0.021616 +2023-10-02 21:50:18,257 - Epoch: [168][ 950/ 1236] Overall Loss 0.131991 Objective Loss 0.131991 LR 0.000250 Time 0.021605 +2023-10-02 21:50:18,465 - Epoch: [168][ 960/ 1236] Overall Loss 0.132050 Objective Loss 0.132050 LR 0.000250 Time 0.021598 +2023-10-02 21:50:18,673 - Epoch: [168][ 970/ 1236] Overall Loss 0.132109 Objective Loss 0.132109 LR 0.000250 Time 0.021588 +2023-10-02 21:50:18,883 - Epoch: [168][ 980/ 1236] Overall Loss 0.132004 Objective Loss 0.132004 LR 0.000250 Time 0.021581 +2023-10-02 21:50:19,090 - Epoch: [168][ 990/ 1236] Overall Loss 0.131741 Objective Loss 0.131741 LR 0.000250 Time 0.021572 +2023-10-02 21:50:19,299 - Epoch: [168][ 1000/ 1236] Overall Loss 0.131834 Objective Loss 0.131834 LR 0.000250 Time 0.021565 +2023-10-02 21:50:19,509 - Epoch: [168][ 1010/ 1236] Overall Loss 0.131927 Objective Loss 0.131927 LR 0.000250 Time 0.021557 +2023-10-02 21:50:19,719 - Epoch: [168][ 1020/ 1236] Overall Loss 0.131915 Objective Loss 0.131915 LR 0.000250 Time 0.021552 +2023-10-02 21:50:19,929 - Epoch: [168][ 1030/ 1236] Overall Loss 0.131781 Objective Loss 0.131781 LR 0.000250 Time 0.021544 +2023-10-02 21:50:20,139 - Epoch: [168][ 1040/ 1236] Overall Loss 0.131849 Objective Loss 0.131849 LR 0.000250 Time 0.021540 +2023-10-02 21:50:20,349 - Epoch: [168][ 1050/ 1236] Overall Loss 0.131879 Objective Loss 0.131879 LR 0.000250 Time 0.021533 +2023-10-02 21:50:20,561 - Epoch: [168][ 1060/ 1236] Overall Loss 0.131987 Objective Loss 0.131987 LR 0.000250 Time 0.021529 +2023-10-02 21:50:20,770 - Epoch: [168][ 1070/ 1236] Overall Loss 0.132122 Objective Loss 0.132122 LR 0.000250 Time 0.021523 +2023-10-02 21:50:20,981 - Epoch: [168][ 1080/ 1236] Overall Loss 0.132027 Objective Loss 0.132027 LR 0.000250 Time 0.021519 +2023-10-02 21:50:21,191 - Epoch: [168][ 1090/ 1236] Overall Loss 0.132175 Objective Loss 0.132175 LR 0.000250 Time 0.021513 +2023-10-02 21:50:21,402 - Epoch: [168][ 1100/ 1236] Overall Loss 0.132130 Objective Loss 0.132130 LR 0.000250 Time 0.021509 +2023-10-02 21:50:21,611 - Epoch: [168][ 1110/ 1236] Overall Loss 0.132281 Objective Loss 0.132281 LR 0.000250 Time 0.021504 +2023-10-02 21:50:21,822 - Epoch: [168][ 1120/ 1236] Overall Loss 0.132315 Objective Loss 0.132315 LR 0.000250 Time 0.021500 +2023-10-02 21:50:22,032 - Epoch: [168][ 1130/ 1236] Overall Loss 0.132287 Objective Loss 0.132287 LR 0.000250 Time 0.021495 +2023-10-02 21:50:22,244 - Epoch: [168][ 1140/ 1236] Overall Loss 0.132235 Objective Loss 0.132235 LR 0.000250 Time 0.021492 +2023-10-02 21:50:22,453 - Epoch: [168][ 1150/ 1236] Overall Loss 0.132194 Objective Loss 0.132194 LR 0.000250 Time 0.021486 +2023-10-02 21:50:22,664 - Epoch: [168][ 1160/ 1236] Overall Loss 0.132307 Objective Loss 0.132307 LR 0.000250 Time 0.021483 +2023-10-02 21:50:22,874 - Epoch: [168][ 1170/ 1236] Overall Loss 0.132385 Objective Loss 0.132385 LR 0.000250 Time 0.021479 +2023-10-02 21:50:23,085 - Epoch: [168][ 1180/ 1236] Overall Loss 0.132298 Objective Loss 0.132298 LR 0.000250 Time 0.021475 +2023-10-02 21:50:23,294 - Epoch: [168][ 1190/ 1236] Overall Loss 0.132282 Objective Loss 0.132282 LR 0.000250 Time 0.021470 +2023-10-02 21:50:23,506 - Epoch: [168][ 1200/ 1236] Overall Loss 0.132163 Objective Loss 0.132163 LR 0.000250 Time 0.021467 +2023-10-02 21:50:23,716 - Epoch: [168][ 1210/ 1236] Overall Loss 0.132310 Objective Loss 0.132310 LR 0.000250 Time 0.021463 +2023-10-02 21:50:23,927 - Epoch: [168][ 1220/ 1236] Overall Loss 0.132381 Objective Loss 0.132381 LR 0.000250 Time 0.021460 +2023-10-02 21:50:24,190 - Epoch: [168][ 1230/ 1236] Overall Loss 0.132335 Objective Loss 0.132335 LR 0.000250 Time 0.021499 +2023-10-02 21:50:24,313 - Epoch: [168][ 1236/ 1236] Overall Loss 0.132411 Objective Loss 0.132411 Top1 89.816701 Top5 98.574338 LR 0.000250 Time 0.021494 +2023-10-02 21:50:24,461 - --- validate (epoch=168)----------- +2023-10-02 21:50:24,461 - 29943 samples (256 per mini-batch) +2023-10-02 21:50:24,965 - Epoch: [168][ 10/ 117] Loss 0.313104 Top1 86.875000 Top5 98.476562 +2023-10-02 21:50:25,121 - Epoch: [168][ 20/ 117] Loss 0.320308 Top1 86.972656 Top5 98.632812 +2023-10-02 21:50:25,274 - Epoch: [168][ 30/ 117] Loss 0.326498 Top1 86.861979 Top5 98.593750 +2023-10-02 21:50:25,430 - Epoch: [168][ 40/ 117] Loss 0.311966 Top1 86.884766 Top5 98.564453 +2023-10-02 21:50:25,583 - Epoch: [168][ 50/ 117] Loss 0.308303 Top1 86.750000 Top5 98.562500 +2023-10-02 21:50:25,737 - Epoch: [168][ 60/ 117] Loss 0.310742 Top1 86.627604 Top5 98.574219 +2023-10-02 21:50:25,890 - Epoch: [168][ 70/ 117] Loss 0.301579 Top1 86.796875 Top5 98.571429 +2023-10-02 21:50:26,045 - Epoch: [168][ 80/ 117] Loss 0.305916 Top1 86.674805 Top5 98.510742 +2023-10-02 21:50:26,201 - Epoch: [168][ 90/ 117] Loss 0.304094 Top1 86.692708 Top5 98.498264 +2023-10-02 21:50:26,360 - Epoch: [168][ 100/ 117] Loss 0.303097 Top1 86.699219 Top5 98.523438 +2023-10-02 21:50:26,528 - Epoch: [168][ 110/ 117] Loss 0.301628 Top1 86.832386 Top5 98.561790 +2023-10-02 21:50:26,618 - Epoch: [168][ 117/ 117] Loss 0.298969 Top1 86.848345 Top5 98.583976 +2023-10-02 21:50:26,773 - ==> Top1: 86.848 Top5: 98.584 Loss: 0.299 + +2023-10-02 21:50:26,774 - ==> Confusion: +[[ 916 1 2 0 15 2 0 0 7 68 2 0 1 1 5 2 4 0 1 0 23] + [ 0 1075 0 1 4 13 0 17 1 1 0 0 1 0 1 3 3 0 5 2 4] + [ 3 1 971 9 1 0 18 4 0 1 1 1 7 3 2 5 1 1 13 2 12] + [ 0 4 12 981 0 1 4 4 1 0 3 0 3 3 32 1 2 4 14 1 19] + [ 17 7 1 0 979 3 1 0 2 10 1 0 0 6 8 2 9 0 0 0 4] + [ 2 33 0 2 3 986 2 18 3 4 1 7 1 15 7 0 4 2 3 3 20] + [ 0 3 23 2 0 3 1130 4 0 0 4 0 0 0 0 5 0 1 2 7 7] + [ 2 11 4 1 5 23 7 1075 1 3 5 3 2 5 1 0 1 0 44 11 14] + [ 16 5 0 0 4 2 0 1 976 31 9 2 0 17 15 1 2 2 2 1 3] + [ 79 0 2 0 12 4 0 0 24 952 1 0 0 26 7 1 1 0 0 0 10] + [ 3 1 10 4 1 2 4 4 5 0 976 1 0 12 6 0 5 1 7 1 10] + [ 1 1 2 0 1 10 0 3 0 0 0 967 12 13 0 1 2 14 0 3 5] + [ 0 0 1 2 0 2 0 0 1 0 6 31 972 4 3 7 2 11 2 6 18] + [ 0 0 2 0 3 3 0 0 8 6 2 7 0 1068 4 0 0 1 0 1 14] + [ 8 0 4 16 3 0 0 0 14 2 0 0 2 5 1029 0 1 2 7 0 8] + [ 0 1 1 1 6 1 1 0 0 0 0 4 6 0 0 1073 19 10 1 5 5] + [ 0 15 0 0 2 6 0 0 0 0 0 3 0 3 2 7 1109 0 1 4 9] + [ 0 0 1 3 1 0 2 0 0 0 0 9 19 2 2 6 1 984 0 0 8] + [ 0 5 2 18 2 2 0 18 3 1 2 1 2 0 8 0 1 0 993 0 10] + [ 0 1 3 1 1 0 6 3 0 1 1 13 3 2 1 2 7 0 0 1096 11] + [ 86 163 109 78 59 95 31 65 67 61 164 85 267 287 128 45 94 52 120 152 5697]] + +2023-10-02 21:50:26,776 - ==> Best [Top1: 87.543 Top5: 98.738 Sparsity:0.00 Params: 169472 on epoch: 167] +2023-10-02 21:50:26,776 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:50:26,782 - + +2023-10-02 21:50:26,782 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:50:27,819 - Epoch: [169][ 10/ 1236] Overall Loss 0.126716 Objective Loss 0.126716 LR 0.000250 Time 0.103620 +2023-10-02 21:50:28,030 - Epoch: [169][ 20/ 1236] Overall Loss 0.127712 Objective Loss 0.127712 LR 0.000250 Time 0.062340 +2023-10-02 21:50:28,240 - Epoch: [169][ 30/ 1236] Overall Loss 0.124750 Objective Loss 0.124750 LR 0.000250 Time 0.048493 +2023-10-02 21:50:28,451 - Epoch: [169][ 40/ 1236] Overall Loss 0.122291 Objective Loss 0.122291 LR 0.000250 Time 0.041641 +2023-10-02 21:50:28,658 - Epoch: [169][ 50/ 1236] Overall Loss 0.122444 Objective Loss 0.122444 LR 0.000250 Time 0.037463 +2023-10-02 21:50:28,870 - Epoch: [169][ 60/ 1236] Overall Loss 0.123100 Objective Loss 0.123100 LR 0.000250 Time 0.034733 +2023-10-02 21:50:29,077 - Epoch: [169][ 70/ 1236] Overall Loss 0.123786 Objective Loss 0.123786 LR 0.000250 Time 0.032733 +2023-10-02 21:50:29,298 - Epoch: [169][ 80/ 1236] Overall Loss 0.123441 Objective Loss 0.123441 LR 0.000250 Time 0.031396 +2023-10-02 21:50:29,507 - Epoch: [169][ 90/ 1236] Overall Loss 0.124535 Objective Loss 0.124535 LR 0.000250 Time 0.030224 +2023-10-02 21:50:29,716 - Epoch: [169][ 100/ 1236] Overall Loss 0.125418 Objective Loss 0.125418 LR 0.000250 Time 0.029288 +2023-10-02 21:50:29,924 - Epoch: [169][ 110/ 1236] Overall Loss 0.126909 Objective Loss 0.126909 LR 0.000250 Time 0.028501 +2023-10-02 21:50:30,131 - Epoch: [169][ 120/ 1236] Overall Loss 0.128078 Objective Loss 0.128078 LR 0.000250 Time 0.027852 +2023-10-02 21:50:30,339 - Epoch: [169][ 130/ 1236] Overall Loss 0.128045 Objective Loss 0.128045 LR 0.000250 Time 0.027305 +2023-10-02 21:50:30,545 - Epoch: [169][ 140/ 1236] Overall Loss 0.127420 Objective Loss 0.127420 LR 0.000250 Time 0.026828 +2023-10-02 21:50:30,753 - Epoch: [169][ 150/ 1236] Overall Loss 0.128233 Objective Loss 0.128233 LR 0.000250 Time 0.026423 +2023-10-02 21:50:30,960 - Epoch: [169][ 160/ 1236] Overall Loss 0.128353 Objective Loss 0.128353 LR 0.000250 Time 0.026060 +2023-10-02 21:50:31,168 - Epoch: [169][ 170/ 1236] Overall Loss 0.129712 Objective Loss 0.129712 LR 0.000250 Time 0.025748 +2023-10-02 21:50:31,374 - Epoch: [169][ 180/ 1236] Overall Loss 0.129546 Objective Loss 0.129546 LR 0.000250 Time 0.025464 +2023-10-02 21:50:31,582 - Epoch: [169][ 190/ 1236] Overall Loss 0.129618 Objective Loss 0.129618 LR 0.000250 Time 0.025214 +2023-10-02 21:50:31,788 - Epoch: [169][ 200/ 1236] Overall Loss 0.129576 Objective Loss 0.129576 LR 0.000250 Time 0.024985 +2023-10-02 21:50:31,996 - Epoch: [169][ 210/ 1236] Overall Loss 0.129916 Objective Loss 0.129916 LR 0.000250 Time 0.024783 +2023-10-02 21:50:32,202 - Epoch: [169][ 220/ 1236] Overall Loss 0.129618 Objective Loss 0.129618 LR 0.000250 Time 0.024594 +2023-10-02 21:50:32,410 - Epoch: [169][ 230/ 1236] Overall Loss 0.130033 Objective Loss 0.130033 LR 0.000250 Time 0.024426 +2023-10-02 21:50:32,617 - Epoch: [169][ 240/ 1236] Overall Loss 0.129389 Objective Loss 0.129389 LR 0.000250 Time 0.024267 +2023-10-02 21:50:32,824 - Epoch: [169][ 250/ 1236] Overall Loss 0.129279 Objective Loss 0.129279 LR 0.000250 Time 0.024127 +2023-10-02 21:50:33,031 - Epoch: [169][ 260/ 1236] Overall Loss 0.128618 Objective Loss 0.128618 LR 0.000250 Time 0.023991 +2023-10-02 21:50:33,238 - Epoch: [169][ 270/ 1236] Overall Loss 0.128589 Objective Loss 0.128589 LR 0.000250 Time 0.023871 +2023-10-02 21:50:33,445 - Epoch: [169][ 280/ 1236] Overall Loss 0.128722 Objective Loss 0.128722 LR 0.000250 Time 0.023755 +2023-10-02 21:50:33,653 - Epoch: [169][ 290/ 1236] Overall Loss 0.128875 Objective Loss 0.128875 LR 0.000250 Time 0.023651 +2023-10-02 21:50:33,858 - Epoch: [169][ 300/ 1236] Overall Loss 0.128833 Objective Loss 0.128833 LR 0.000250 Time 0.023544 +2023-10-02 21:50:34,067 - Epoch: [169][ 310/ 1236] Overall Loss 0.130059 Objective Loss 0.130059 LR 0.000250 Time 0.023460 +2023-10-02 21:50:34,277 - Epoch: [169][ 320/ 1236] Overall Loss 0.130265 Objective Loss 0.130265 LR 0.000250 Time 0.023381 +2023-10-02 21:50:34,487 - Epoch: [169][ 330/ 1236] Overall Loss 0.130576 Objective Loss 0.130576 LR 0.000250 Time 0.023307 +2023-10-02 21:50:34,693 - Epoch: [169][ 340/ 1236] Overall Loss 0.130718 Objective Loss 0.130718 LR 0.000250 Time 0.023229 +2023-10-02 21:50:34,901 - Epoch: [169][ 350/ 1236] Overall Loss 0.130322 Objective Loss 0.130322 LR 0.000250 Time 0.023159 +2023-10-02 21:50:35,108 - Epoch: [169][ 360/ 1236] Overall Loss 0.129926 Objective Loss 0.129926 LR 0.000250 Time 0.023089 +2023-10-02 21:50:35,316 - Epoch: [169][ 370/ 1236] Overall Loss 0.129977 Objective Loss 0.129977 LR 0.000250 Time 0.023027 +2023-10-02 21:50:35,523 - Epoch: [169][ 380/ 1236] Overall Loss 0.129722 Objective Loss 0.129722 LR 0.000250 Time 0.022964 +2023-10-02 21:50:35,731 - Epoch: [169][ 390/ 1236] Overall Loss 0.130267 Objective Loss 0.130267 LR 0.000250 Time 0.022907 +2023-10-02 21:50:35,937 - Epoch: [169][ 400/ 1236] Overall Loss 0.130081 Objective Loss 0.130081 LR 0.000250 Time 0.022850 +2023-10-02 21:50:36,146 - Epoch: [169][ 410/ 1236] Overall Loss 0.129746 Objective Loss 0.129746 LR 0.000250 Time 0.022800 +2023-10-02 21:50:36,353 - Epoch: [169][ 420/ 1236] Overall Loss 0.129402 Objective Loss 0.129402 LR 0.000250 Time 0.022750 +2023-10-02 21:50:36,561 - Epoch: [169][ 430/ 1236] Overall Loss 0.129432 Objective Loss 0.129432 LR 0.000250 Time 0.022705 +2023-10-02 21:50:36,768 - Epoch: [169][ 440/ 1236] Overall Loss 0.129333 Objective Loss 0.129333 LR 0.000250 Time 0.022658 +2023-10-02 21:50:36,977 - Epoch: [169][ 450/ 1236] Overall Loss 0.129498 Objective Loss 0.129498 LR 0.000250 Time 0.022617 +2023-10-02 21:50:37,184 - Epoch: [169][ 460/ 1236] Overall Loss 0.129150 Objective Loss 0.129150 LR 0.000250 Time 0.022576 +2023-10-02 21:50:37,394 - Epoch: [169][ 470/ 1236] Overall Loss 0.128750 Objective Loss 0.128750 LR 0.000250 Time 0.022541 +2023-10-02 21:50:37,603 - Epoch: [169][ 480/ 1236] Overall Loss 0.128483 Objective Loss 0.128483 LR 0.000250 Time 0.022506 +2023-10-02 21:50:37,813 - Epoch: [169][ 490/ 1236] Overall Loss 0.128234 Objective Loss 0.128234 LR 0.000250 Time 0.022476 +2023-10-02 21:50:38,022 - Epoch: [169][ 500/ 1236] Overall Loss 0.128196 Objective Loss 0.128196 LR 0.000250 Time 0.022444 +2023-10-02 21:50:38,236 - Epoch: [169][ 510/ 1236] Overall Loss 0.128435 Objective Loss 0.128435 LR 0.000250 Time 0.022423 +2023-10-02 21:50:38,446 - Epoch: [169][ 520/ 1236] Overall Loss 0.128692 Objective Loss 0.128692 LR 0.000250 Time 0.022395 +2023-10-02 21:50:38,660 - Epoch: [169][ 530/ 1236] Overall Loss 0.128225 Objective Loss 0.128225 LR 0.000250 Time 0.022375 +2023-10-02 21:50:38,870 - Epoch: [169][ 540/ 1236] Overall Loss 0.128269 Objective Loss 0.128269 LR 0.000250 Time 0.022349 +2023-10-02 21:50:39,083 - Epoch: [169][ 550/ 1236] Overall Loss 0.128081 Objective Loss 0.128081 LR 0.000250 Time 0.022329 +2023-10-02 21:50:39,293 - Epoch: [169][ 560/ 1236] Overall Loss 0.128032 Objective Loss 0.128032 LR 0.000250 Time 0.022304 +2023-10-02 21:50:39,507 - Epoch: [169][ 570/ 1236] Overall Loss 0.127916 Objective Loss 0.127916 LR 0.000250 Time 0.022288 +2023-10-02 21:50:39,716 - Epoch: [169][ 580/ 1236] Overall Loss 0.127894 Objective Loss 0.127894 LR 0.000250 Time 0.022265 +2023-10-02 21:50:39,930 - Epoch: [169][ 590/ 1236] Overall Loss 0.128095 Objective Loss 0.128095 LR 0.000250 Time 0.022249 +2023-10-02 21:50:40,140 - Epoch: [169][ 600/ 1236] Overall Loss 0.128096 Objective Loss 0.128096 LR 0.000250 Time 0.022228 +2023-10-02 21:50:40,353 - Epoch: [169][ 610/ 1236] Overall Loss 0.128403 Objective Loss 0.128403 LR 0.000250 Time 0.022211 +2023-10-02 21:50:40,568 - Epoch: [169][ 620/ 1236] Overall Loss 0.128000 Objective Loss 0.128000 LR 0.000250 Time 0.022199 +2023-10-02 21:50:40,780 - Epoch: [169][ 630/ 1236] Overall Loss 0.127853 Objective Loss 0.127853 LR 0.000250 Time 0.022183 +2023-10-02 21:50:40,996 - Epoch: [169][ 640/ 1236] Overall Loss 0.127871 Objective Loss 0.127871 LR 0.000250 Time 0.022173 +2023-10-02 21:50:41,209 - Epoch: [169][ 650/ 1236] Overall Loss 0.127957 Objective Loss 0.127957 LR 0.000250 Time 0.022159 +2023-10-02 21:50:41,422 - Epoch: [169][ 660/ 1236] Overall Loss 0.127862 Objective Loss 0.127862 LR 0.000250 Time 0.022146 +2023-10-02 21:50:41,636 - Epoch: [169][ 670/ 1236] Overall Loss 0.128225 Objective Loss 0.128225 LR 0.000250 Time 0.022132 +2023-10-02 21:50:41,850 - Epoch: [169][ 680/ 1236] Overall Loss 0.128166 Objective Loss 0.128166 LR 0.000250 Time 0.022121 +2023-10-02 21:50:42,064 - Epoch: [169][ 690/ 1236] Overall Loss 0.128085 Objective Loss 0.128085 LR 0.000250 Time 0.022108 +2023-10-02 21:50:42,277 - Epoch: [169][ 700/ 1236] Overall Loss 0.128047 Objective Loss 0.128047 LR 0.000250 Time 0.022096 +2023-10-02 21:50:42,491 - Epoch: [169][ 710/ 1236] Overall Loss 0.128416 Objective Loss 0.128416 LR 0.000250 Time 0.022084 +2023-10-02 21:50:42,705 - Epoch: [169][ 720/ 1236] Overall Loss 0.128610 Objective Loss 0.128610 LR 0.000250 Time 0.022074 +2023-10-02 21:50:42,919 - Epoch: [169][ 730/ 1236] Overall Loss 0.128484 Objective Loss 0.128484 LR 0.000250 Time 0.022063 +2023-10-02 21:50:43,131 - Epoch: [169][ 740/ 1236] Overall Loss 0.128580 Objective Loss 0.128580 LR 0.000250 Time 0.022052 +2023-10-02 21:50:43,343 - Epoch: [169][ 750/ 1236] Overall Loss 0.128770 Objective Loss 0.128770 LR 0.000250 Time 0.022039 +2023-10-02 21:50:43,557 - Epoch: [169][ 760/ 1236] Overall Loss 0.128829 Objective Loss 0.128829 LR 0.000250 Time 0.022030 +2023-10-02 21:50:43,767 - Epoch: [169][ 770/ 1236] Overall Loss 0.128825 Objective Loss 0.128825 LR 0.000250 Time 0.022017 +2023-10-02 21:50:43,980 - Epoch: [169][ 780/ 1236] Overall Loss 0.128454 Objective Loss 0.128454 LR 0.000250 Time 0.022007 +2023-10-02 21:50:44,193 - Epoch: [169][ 790/ 1236] Overall Loss 0.128505 Objective Loss 0.128505 LR 0.000250 Time 0.021995 +2023-10-02 21:50:44,405 - Epoch: [169][ 800/ 1236] Overall Loss 0.128547 Objective Loss 0.128547 LR 0.000250 Time 0.021985 +2023-10-02 21:50:44,617 - Epoch: [169][ 810/ 1236] Overall Loss 0.128612 Objective Loss 0.128612 LR 0.000250 Time 0.021973 +2023-10-02 21:50:44,831 - Epoch: [169][ 820/ 1236] Overall Loss 0.128525 Objective Loss 0.128525 LR 0.000250 Time 0.021966 +2023-10-02 21:50:45,042 - Epoch: [169][ 830/ 1236] Overall Loss 0.128616 Objective Loss 0.128616 LR 0.000250 Time 0.021955 +2023-10-02 21:50:45,256 - Epoch: [169][ 840/ 1236] Overall Loss 0.128667 Objective Loss 0.128667 LR 0.000250 Time 0.021948 +2023-10-02 21:50:45,467 - Epoch: [169][ 850/ 1236] Overall Loss 0.128781 Objective Loss 0.128781 LR 0.000250 Time 0.021937 +2023-10-02 21:50:45,679 - Epoch: [169][ 860/ 1236] Overall Loss 0.128921 Objective Loss 0.128921 LR 0.000250 Time 0.021929 +2023-10-02 21:50:45,892 - Epoch: [169][ 870/ 1236] Overall Loss 0.128863 Objective Loss 0.128863 LR 0.000250 Time 0.021919 +2023-10-02 21:50:46,105 - Epoch: [169][ 880/ 1236] Overall Loss 0.128734 Objective Loss 0.128734 LR 0.000250 Time 0.021912 +2023-10-02 21:50:46,317 - Epoch: [169][ 890/ 1236] Overall Loss 0.128835 Objective Loss 0.128835 LR 0.000250 Time 0.021902 +2023-10-02 21:50:46,529 - Epoch: [169][ 900/ 1236] Overall Loss 0.129010 Objective Loss 0.129010 LR 0.000250 Time 0.021895 +2023-10-02 21:50:46,742 - Epoch: [169][ 910/ 1236] Overall Loss 0.129167 Objective Loss 0.129167 LR 0.000250 Time 0.021886 +2023-10-02 21:50:46,954 - Epoch: [169][ 920/ 1236] Overall Loss 0.129072 Objective Loss 0.129072 LR 0.000250 Time 0.021879 +2023-10-02 21:50:47,167 - Epoch: [169][ 930/ 1236] Overall Loss 0.128938 Objective Loss 0.128938 LR 0.000250 Time 0.021870 +2023-10-02 21:50:47,381 - Epoch: [169][ 940/ 1236] Overall Loss 0.128974 Objective Loss 0.128974 LR 0.000250 Time 0.021865 +2023-10-02 21:50:47,592 - Epoch: [169][ 950/ 1236] Overall Loss 0.129101 Objective Loss 0.129101 LR 0.000250 Time 0.021856 +2023-10-02 21:50:47,806 - Epoch: [169][ 960/ 1236] Overall Loss 0.129248 Objective Loss 0.129248 LR 0.000250 Time 0.021851 +2023-10-02 21:50:48,017 - Epoch: [169][ 970/ 1236] Overall Loss 0.129173 Objective Loss 0.129173 LR 0.000250 Time 0.021843 +2023-10-02 21:50:48,231 - Epoch: [169][ 980/ 1236] Overall Loss 0.128963 Objective Loss 0.128963 LR 0.000250 Time 0.021838 +2023-10-02 21:50:48,442 - Epoch: [169][ 990/ 1236] Overall Loss 0.128936 Objective Loss 0.128936 LR 0.000250 Time 0.021830 +2023-10-02 21:50:48,656 - Epoch: [169][ 1000/ 1236] Overall Loss 0.128921 Objective Loss 0.128921 LR 0.000250 Time 0.021825 +2023-10-02 21:50:48,867 - Epoch: [169][ 1010/ 1236] Overall Loss 0.128909 Objective Loss 0.128909 LR 0.000250 Time 0.021817 +2023-10-02 21:50:49,079 - Epoch: [169][ 1020/ 1236] Overall Loss 0.129162 Objective Loss 0.129162 LR 0.000250 Time 0.021812 +2023-10-02 21:50:49,292 - Epoch: [169][ 1030/ 1236] Overall Loss 0.129184 Objective Loss 0.129184 LR 0.000250 Time 0.021804 +2023-10-02 21:50:49,505 - Epoch: [169][ 1040/ 1236] Overall Loss 0.129127 Objective Loss 0.129127 LR 0.000250 Time 0.021799 +2023-10-02 21:50:49,717 - Epoch: [169][ 1050/ 1236] Overall Loss 0.129174 Objective Loss 0.129174 LR 0.000250 Time 0.021792 +2023-10-02 21:50:49,929 - Epoch: [169][ 1060/ 1236] Overall Loss 0.129182 Objective Loss 0.129182 LR 0.000250 Time 0.021787 +2023-10-02 21:50:50,142 - Epoch: [169][ 1070/ 1236] Overall Loss 0.129161 Objective Loss 0.129161 LR 0.000250 Time 0.021780 +2023-10-02 21:50:50,354 - Epoch: [169][ 1080/ 1236] Overall Loss 0.129341 Objective Loss 0.129341 LR 0.000250 Time 0.021775 +2023-10-02 21:50:50,567 - Epoch: [169][ 1090/ 1236] Overall Loss 0.129224 Objective Loss 0.129224 LR 0.000250 Time 0.021768 +2023-10-02 21:50:50,781 - Epoch: [169][ 1100/ 1236] Overall Loss 0.129281 Objective Loss 0.129281 LR 0.000250 Time 0.021765 +2023-10-02 21:50:50,992 - Epoch: [169][ 1110/ 1236] Overall Loss 0.129132 Objective Loss 0.129132 LR 0.000250 Time 0.021758 +2023-10-02 21:50:51,206 - Epoch: [169][ 1120/ 1236] Overall Loss 0.129091 Objective Loss 0.129091 LR 0.000250 Time 0.021755 +2023-10-02 21:50:51,417 - Epoch: [169][ 1130/ 1236] Overall Loss 0.129157 Objective Loss 0.129157 LR 0.000250 Time 0.021749 +2023-10-02 21:50:51,630 - Epoch: [169][ 1140/ 1236] Overall Loss 0.129153 Objective Loss 0.129153 LR 0.000250 Time 0.021744 +2023-10-02 21:50:51,842 - Epoch: [169][ 1150/ 1236] Overall Loss 0.129303 Objective Loss 0.129303 LR 0.000250 Time 0.021738 +2023-10-02 21:50:52,055 - Epoch: [169][ 1160/ 1236] Overall Loss 0.129450 Objective Loss 0.129450 LR 0.000250 Time 0.021733 +2023-10-02 21:50:52,267 - Epoch: [169][ 1170/ 1236] Overall Loss 0.129378 Objective Loss 0.129378 LR 0.000250 Time 0.021727 +2023-10-02 21:50:52,480 - Epoch: [169][ 1180/ 1236] Overall Loss 0.129476 Objective Loss 0.129476 LR 0.000250 Time 0.021723 +2023-10-02 21:50:52,692 - Epoch: [169][ 1190/ 1236] Overall Loss 0.129405 Objective Loss 0.129405 LR 0.000250 Time 0.021717 +2023-10-02 21:50:52,905 - Epoch: [169][ 1200/ 1236] Overall Loss 0.129520 Objective Loss 0.129520 LR 0.000250 Time 0.021713 +2023-10-02 21:50:53,117 - Epoch: [169][ 1210/ 1236] Overall Loss 0.129549 Objective Loss 0.129549 LR 0.000250 Time 0.021708 +2023-10-02 21:50:53,329 - Epoch: [169][ 1220/ 1236] Overall Loss 0.129530 Objective Loss 0.129530 LR 0.000250 Time 0.021703 +2023-10-02 21:50:53,593 - Epoch: [169][ 1230/ 1236] Overall Loss 0.129533 Objective Loss 0.129533 LR 0.000250 Time 0.021739 +2023-10-02 21:50:53,715 - Epoch: [169][ 1236/ 1236] Overall Loss 0.129537 Objective Loss 0.129537 Top1 90.631365 Top5 98.778004 LR 0.000250 Time 0.021732 +2023-10-02 21:50:53,853 - --- validate (epoch=169)----------- +2023-10-02 21:50:53,853 - 29943 samples (256 per mini-batch) +2023-10-02 21:50:54,341 - Epoch: [169][ 10/ 117] Loss 0.284734 Top1 86.875000 Top5 98.867188 +2023-10-02 21:50:54,496 - Epoch: [169][ 20/ 117] Loss 0.303562 Top1 86.757812 Top5 98.945312 +2023-10-02 21:50:54,648 - Epoch: [169][ 30/ 117] Loss 0.297730 Top1 86.822917 Top5 98.880208 +2023-10-02 21:50:54,802 - Epoch: [169][ 40/ 117] Loss 0.293668 Top1 87.265625 Top5 98.867188 +2023-10-02 21:50:54,951 - Epoch: [169][ 50/ 117] Loss 0.291654 Top1 87.343750 Top5 98.789062 +2023-10-02 21:50:55,104 - Epoch: [169][ 60/ 117] Loss 0.300463 Top1 87.005208 Top5 98.769531 +2023-10-02 21:50:55,254 - Epoch: [169][ 70/ 117] Loss 0.299517 Top1 86.757812 Top5 98.766741 +2023-10-02 21:50:55,407 - Epoch: [169][ 80/ 117] Loss 0.300745 Top1 86.889648 Top5 98.740234 +2023-10-02 21:50:55,558 - Epoch: [169][ 90/ 117] Loss 0.299275 Top1 86.974826 Top5 98.750000 +2023-10-02 21:50:55,715 - Epoch: [169][ 100/ 117] Loss 0.299155 Top1 87.015625 Top5 98.738281 +2023-10-02 21:50:55,878 - Epoch: [169][ 110/ 117] Loss 0.300849 Top1 86.992188 Top5 98.718040 +2023-10-02 21:50:55,966 - Epoch: [169][ 117/ 117] Loss 0.301646 Top1 86.998631 Top5 98.724243 +2023-10-02 21:50:56,106 - ==> Top1: 86.999 Top5: 98.724 Loss: 0.302 + +2023-10-02 21:50:56,107 - ==> Confusion: +[[ 944 1 3 1 6 2 0 0 8 53 1 1 1 0 5 0 3 0 1 0 20] + [ 0 1059 3 0 6 23 1 21 0 1 0 1 1 0 1 3 1 0 7 2 1] + [ 3 1 983 9 1 0 14 5 0 0 1 1 6 3 1 4 1 2 9 3 9] + [ 1 2 13 987 0 2 2 1 5 0 3 0 4 3 26 2 1 5 14 1 17] + [ 23 3 0 0 984 4 0 0 3 9 1 0 0 5 5 3 6 0 0 1 3] + [ 1 32 0 1 1 1006 2 23 2 4 0 6 1 8 5 0 2 1 4 0 17] + [ 0 5 32 3 0 1 1117 5 0 0 4 0 0 0 0 7 0 0 2 8 7] + [ 2 11 6 0 5 24 5 1078 1 2 4 5 3 5 1 0 0 1 44 10 11] + [ 16 1 0 0 4 6 0 0 982 37 10 0 2 11 13 2 0 1 3 0 1] + [ 98 1 0 0 6 3 0 0 29 945 0 1 0 17 6 3 0 0 0 0 10] + [ 1 2 8 10 1 1 2 2 13 1 975 2 0 9 4 0 0 3 4 2 13] + [ 0 0 1 0 0 17 0 3 0 0 0 959 24 5 0 2 1 15 0 3 5] + [ 0 1 2 4 0 2 1 2 0 0 3 27 985 3 2 6 0 15 0 6 9] + [ 1 1 0 0 5 10 0 0 16 8 1 8 0 1049 4 0 0 1 0 0 15] + [ 12 0 2 15 5 0 0 0 23 3 1 0 1 2 1012 0 0 1 12 0 12] + [ 0 0 2 2 6 0 2 0 0 0 0 4 8 0 0 1070 13 12 3 6 6] + [ 0 16 0 0 4 7 2 0 1 0 0 5 0 4 3 8 1093 0 1 2 15] + [ 0 0 0 1 0 0 3 0 0 1 0 5 24 2 2 4 0 992 0 0 4] + [ 2 3 2 17 0 0 0 22 4 1 3 0 1 0 8 0 1 0 993 0 11] + [ 0 0 2 0 1 2 10 6 0 1 0 14 3 4 0 0 7 1 0 1093 8] + [ 108 145 96 94 64 124 26 85 88 64 151 77 284 230 115 42 60 72 91 145 5744]] + +2023-10-02 21:50:56,109 - ==> Best [Top1: 87.543 Top5: 98.738 Sparsity:0.00 Params: 169472 on epoch: 167] +2023-10-02 21:50:56,109 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:50:56,115 - + +2023-10-02 21:50:56,115 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:50:57,249 - Epoch: [170][ 10/ 1236] Overall Loss 0.144805 Objective Loss 0.144805 LR 0.000250 Time 0.113383 +2023-10-02 21:50:57,458 - Epoch: [170][ 20/ 1236] Overall Loss 0.137577 Objective Loss 0.137577 LR 0.000250 Time 0.067096 +2023-10-02 21:50:57,664 - Epoch: [170][ 30/ 1236] Overall Loss 0.131823 Objective Loss 0.131823 LR 0.000250 Time 0.051601 +2023-10-02 21:50:57,873 - Epoch: [170][ 40/ 1236] Overall Loss 0.132088 Objective Loss 0.132088 LR 0.000250 Time 0.043900 +2023-10-02 21:50:58,080 - Epoch: [170][ 50/ 1236] Overall Loss 0.130818 Objective Loss 0.130818 LR 0.000250 Time 0.039232 +2023-10-02 21:50:58,287 - Epoch: [170][ 60/ 1236] Overall Loss 0.129068 Objective Loss 0.129068 LR 0.000250 Time 0.036149 +2023-10-02 21:50:58,495 - Epoch: [170][ 70/ 1236] Overall Loss 0.127022 Objective Loss 0.127022 LR 0.000250 Time 0.033944 +2023-10-02 21:50:58,703 - Epoch: [170][ 80/ 1236] Overall Loss 0.129751 Objective Loss 0.129751 LR 0.000250 Time 0.032298 +2023-10-02 21:50:58,909 - Epoch: [170][ 90/ 1236] Overall Loss 0.129365 Objective Loss 0.129365 LR 0.000250 Time 0.030987 +2023-10-02 21:50:59,118 - Epoch: [170][ 100/ 1236] Overall Loss 0.127181 Objective Loss 0.127181 LR 0.000250 Time 0.029971 +2023-10-02 21:50:59,325 - Epoch: [170][ 110/ 1236] Overall Loss 0.127139 Objective Loss 0.127139 LR 0.000250 Time 0.029123 +2023-10-02 21:50:59,531 - Epoch: [170][ 120/ 1236] Overall Loss 0.126980 Objective Loss 0.126980 LR 0.000250 Time 0.028412 +2023-10-02 21:50:59,738 - Epoch: [170][ 130/ 1236] Overall Loss 0.126533 Objective Loss 0.126533 LR 0.000250 Time 0.027814 +2023-10-02 21:50:59,945 - Epoch: [170][ 140/ 1236] Overall Loss 0.125893 Objective Loss 0.125893 LR 0.000250 Time 0.027303 +2023-10-02 21:51:00,151 - Epoch: [170][ 150/ 1236] Overall Loss 0.125284 Objective Loss 0.125284 LR 0.000250 Time 0.026846 +2023-10-02 21:51:00,359 - Epoch: [170][ 160/ 1236] Overall Loss 0.126048 Objective Loss 0.126048 LR 0.000250 Time 0.026467 +2023-10-02 21:51:00,566 - Epoch: [170][ 170/ 1236] Overall Loss 0.126055 Objective Loss 0.126055 LR 0.000250 Time 0.026121 +2023-10-02 21:51:00,775 - Epoch: [170][ 180/ 1236] Overall Loss 0.125879 Objective Loss 0.125879 LR 0.000250 Time 0.025827 +2023-10-02 21:51:00,982 - Epoch: [170][ 190/ 1236] Overall Loss 0.127282 Objective Loss 0.127282 LR 0.000250 Time 0.025548 +2023-10-02 21:51:01,190 - Epoch: [170][ 200/ 1236] Overall Loss 0.127114 Objective Loss 0.127114 LR 0.000250 Time 0.025310 +2023-10-02 21:51:01,397 - Epoch: [170][ 210/ 1236] Overall Loss 0.126404 Objective Loss 0.126404 LR 0.000250 Time 0.025084 +2023-10-02 21:51:01,606 - Epoch: [170][ 220/ 1236] Overall Loss 0.127116 Objective Loss 0.127116 LR 0.000250 Time 0.024891 +2023-10-02 21:51:01,813 - Epoch: [170][ 230/ 1236] Overall Loss 0.126993 Objective Loss 0.126993 LR 0.000250 Time 0.024704 +2023-10-02 21:51:02,021 - Epoch: [170][ 240/ 1236] Overall Loss 0.126503 Objective Loss 0.126503 LR 0.000250 Time 0.024541 +2023-10-02 21:51:02,228 - Epoch: [170][ 250/ 1236] Overall Loss 0.126563 Objective Loss 0.126563 LR 0.000250 Time 0.024382 +2023-10-02 21:51:02,437 - Epoch: [170][ 260/ 1236] Overall Loss 0.126270 Objective Loss 0.126270 LR 0.000250 Time 0.024244 +2023-10-02 21:51:02,643 - Epoch: [170][ 270/ 1236] Overall Loss 0.126220 Objective Loss 0.126220 LR 0.000250 Time 0.024106 +2023-10-02 21:51:02,852 - Epoch: [170][ 280/ 1236] Overall Loss 0.126287 Objective Loss 0.126287 LR 0.000250 Time 0.023988 +2023-10-02 21:51:03,058 - Epoch: [170][ 290/ 1236] Overall Loss 0.126629 Objective Loss 0.126629 LR 0.000250 Time 0.023868 +2023-10-02 21:51:03,267 - Epoch: [170][ 300/ 1236] Overall Loss 0.126579 Objective Loss 0.126579 LR 0.000250 Time 0.023766 +2023-10-02 21:51:03,473 - Epoch: [170][ 310/ 1236] Overall Loss 0.126265 Objective Loss 0.126265 LR 0.000250 Time 0.023659 +2023-10-02 21:51:03,682 - Epoch: [170][ 320/ 1236] Overall Loss 0.126560 Objective Loss 0.126560 LR 0.000250 Time 0.023573 +2023-10-02 21:51:03,889 - Epoch: [170][ 330/ 1236] Overall Loss 0.127042 Objective Loss 0.127042 LR 0.000250 Time 0.023483 +2023-10-02 21:51:04,097 - Epoch: [170][ 340/ 1236] Overall Loss 0.127268 Objective Loss 0.127268 LR 0.000250 Time 0.023405 +2023-10-02 21:51:04,304 - Epoch: [170][ 350/ 1236] Overall Loss 0.127199 Objective Loss 0.127199 LR 0.000250 Time 0.023324 +2023-10-02 21:51:04,513 - Epoch: [170][ 360/ 1236] Overall Loss 0.127898 Objective Loss 0.127898 LR 0.000250 Time 0.023254 +2023-10-02 21:51:04,720 - Epoch: [170][ 370/ 1236] Overall Loss 0.128112 Objective Loss 0.128112 LR 0.000250 Time 0.023182 +2023-10-02 21:51:04,929 - Epoch: [170][ 380/ 1236] Overall Loss 0.128209 Objective Loss 0.128209 LR 0.000250 Time 0.023120 +2023-10-02 21:51:05,137 - Epoch: [170][ 390/ 1236] Overall Loss 0.127733 Objective Loss 0.127733 LR 0.000250 Time 0.023059 +2023-10-02 21:51:05,345 - Epoch: [170][ 400/ 1236] Overall Loss 0.127910 Objective Loss 0.127910 LR 0.000250 Time 0.023003 +2023-10-02 21:51:05,553 - Epoch: [170][ 410/ 1236] Overall Loss 0.127983 Objective Loss 0.127983 LR 0.000250 Time 0.022944 +2023-10-02 21:51:05,761 - Epoch: [170][ 420/ 1236] Overall Loss 0.128215 Objective Loss 0.128215 LR 0.000250 Time 0.022893 +2023-10-02 21:51:05,969 - Epoch: [170][ 430/ 1236] Overall Loss 0.128343 Objective Loss 0.128343 LR 0.000250 Time 0.022843 +2023-10-02 21:51:06,177 - Epoch: [170][ 440/ 1236] Overall Loss 0.128171 Objective Loss 0.128171 LR 0.000250 Time 0.022797 +2023-10-02 21:51:06,385 - Epoch: [170][ 450/ 1236] Overall Loss 0.128248 Objective Loss 0.128248 LR 0.000250 Time 0.022751 +2023-10-02 21:51:06,593 - Epoch: [170][ 460/ 1236] Overall Loss 0.128422 Objective Loss 0.128422 LR 0.000250 Time 0.022709 +2023-10-02 21:51:06,799 - Epoch: [170][ 470/ 1236] Overall Loss 0.128213 Objective Loss 0.128213 LR 0.000250 Time 0.022660 +2023-10-02 21:51:07,008 - Epoch: [170][ 480/ 1236] Overall Loss 0.128433 Objective Loss 0.128433 LR 0.000250 Time 0.022622 +2023-10-02 21:51:07,215 - Epoch: [170][ 490/ 1236] Overall Loss 0.127971 Objective Loss 0.127971 LR 0.000250 Time 0.022584 +2023-10-02 21:51:07,424 - Epoch: [170][ 500/ 1236] Overall Loss 0.128128 Objective Loss 0.128128 LR 0.000250 Time 0.022549 +2023-10-02 21:51:07,632 - Epoch: [170][ 510/ 1236] Overall Loss 0.128183 Objective Loss 0.128183 LR 0.000250 Time 0.022512 +2023-10-02 21:51:07,841 - Epoch: [170][ 520/ 1236] Overall Loss 0.127931 Objective Loss 0.127931 LR 0.000250 Time 0.022479 +2023-10-02 21:51:08,049 - Epoch: [170][ 530/ 1236] Overall Loss 0.127879 Objective Loss 0.127879 LR 0.000250 Time 0.022447 +2023-10-02 21:51:08,257 - Epoch: [170][ 540/ 1236] Overall Loss 0.128108 Objective Loss 0.128108 LR 0.000250 Time 0.022417 +2023-10-02 21:51:08,465 - Epoch: [170][ 550/ 1236] Overall Loss 0.128264 Objective Loss 0.128264 LR 0.000250 Time 0.022386 +2023-10-02 21:51:08,674 - Epoch: [170][ 560/ 1236] Overall Loss 0.128181 Objective Loss 0.128181 LR 0.000250 Time 0.022359 +2023-10-02 21:51:08,882 - Epoch: [170][ 570/ 1236] Overall Loss 0.128304 Objective Loss 0.128304 LR 0.000250 Time 0.022331 +2023-10-02 21:51:09,090 - Epoch: [170][ 580/ 1236] Overall Loss 0.128166 Objective Loss 0.128166 LR 0.000250 Time 0.022304 +2023-10-02 21:51:09,298 - Epoch: [170][ 590/ 1236] Overall Loss 0.128115 Objective Loss 0.128115 LR 0.000250 Time 0.022278 +2023-10-02 21:51:09,507 - Epoch: [170][ 600/ 1236] Overall Loss 0.128183 Objective Loss 0.128183 LR 0.000250 Time 0.022254 +2023-10-02 21:51:09,714 - Epoch: [170][ 610/ 1236] Overall Loss 0.128375 Objective Loss 0.128375 LR 0.000250 Time 0.022228 +2023-10-02 21:51:09,923 - Epoch: [170][ 620/ 1236] Overall Loss 0.128391 Objective Loss 0.128391 LR 0.000250 Time 0.022206 +2023-10-02 21:51:10,131 - Epoch: [170][ 630/ 1236] Overall Loss 0.128268 Objective Loss 0.128268 LR 0.000250 Time 0.022183 +2023-10-02 21:51:10,339 - Epoch: [170][ 640/ 1236] Overall Loss 0.128275 Objective Loss 0.128275 LR 0.000250 Time 0.022162 +2023-10-02 21:51:10,547 - Epoch: [170][ 650/ 1236] Overall Loss 0.128698 Objective Loss 0.128698 LR 0.000250 Time 0.022138 +2023-10-02 21:51:10,754 - Epoch: [170][ 660/ 1236] Overall Loss 0.128547 Objective Loss 0.128547 LR 0.000250 Time 0.022116 +2023-10-02 21:51:10,961 - Epoch: [170][ 670/ 1236] Overall Loss 0.128338 Objective Loss 0.128338 LR 0.000250 Time 0.022094 +2023-10-02 21:51:11,170 - Epoch: [170][ 680/ 1236] Overall Loss 0.128524 Objective Loss 0.128524 LR 0.000250 Time 0.022076 +2023-10-02 21:51:11,378 - Epoch: [170][ 690/ 1236] Overall Loss 0.128487 Objective Loss 0.128487 LR 0.000250 Time 0.022057 +2023-10-02 21:51:11,587 - Epoch: [170][ 700/ 1236] Overall Loss 0.128630 Objective Loss 0.128630 LR 0.000250 Time 0.022040 +2023-10-02 21:51:11,794 - Epoch: [170][ 710/ 1236] Overall Loss 0.128578 Objective Loss 0.128578 LR 0.000250 Time 0.022019 +2023-10-02 21:51:12,002 - Epoch: [170][ 720/ 1236] Overall Loss 0.128513 Objective Loss 0.128513 LR 0.000250 Time 0.022002 +2023-10-02 21:51:12,210 - Epoch: [170][ 730/ 1236] Overall Loss 0.128581 Objective Loss 0.128581 LR 0.000250 Time 0.021983 +2023-10-02 21:51:12,418 - Epoch: [170][ 740/ 1236] Overall Loss 0.128348 Objective Loss 0.128348 LR 0.000250 Time 0.021967 +2023-10-02 21:51:12,627 - Epoch: [170][ 750/ 1236] Overall Loss 0.128414 Objective Loss 0.128414 LR 0.000250 Time 0.021952 +2023-10-02 21:51:12,835 - Epoch: [170][ 760/ 1236] Overall Loss 0.128633 Objective Loss 0.128633 LR 0.000250 Time 0.021937 +2023-10-02 21:51:13,043 - Epoch: [170][ 770/ 1236] Overall Loss 0.128566 Objective Loss 0.128566 LR 0.000250 Time 0.021921 +2023-10-02 21:51:13,252 - Epoch: [170][ 780/ 1236] Overall Loss 0.128498 Objective Loss 0.128498 LR 0.000250 Time 0.021908 +2023-10-02 21:51:13,460 - Epoch: [170][ 790/ 1236] Overall Loss 0.128710 Objective Loss 0.128710 LR 0.000250 Time 0.021893 +2023-10-02 21:51:13,668 - Epoch: [170][ 800/ 1236] Overall Loss 0.128644 Objective Loss 0.128644 LR 0.000250 Time 0.021879 +2023-10-02 21:51:13,876 - Epoch: [170][ 810/ 1236] Overall Loss 0.128930 Objective Loss 0.128930 LR 0.000250 Time 0.021865 +2023-10-02 21:51:14,085 - Epoch: [170][ 820/ 1236] Overall Loss 0.128843 Objective Loss 0.128843 LR 0.000250 Time 0.021853 +2023-10-02 21:51:14,292 - Epoch: [170][ 830/ 1236] Overall Loss 0.128837 Objective Loss 0.128837 LR 0.000250 Time 0.021839 +2023-10-02 21:51:14,501 - Epoch: [170][ 840/ 1236] Overall Loss 0.128822 Objective Loss 0.128822 LR 0.000250 Time 0.021828 +2023-10-02 21:51:14,709 - Epoch: [170][ 850/ 1236] Overall Loss 0.129061 Objective Loss 0.129061 LR 0.000250 Time 0.021813 +2023-10-02 21:51:14,918 - Epoch: [170][ 860/ 1236] Overall Loss 0.129112 Objective Loss 0.129112 LR 0.000250 Time 0.021802 +2023-10-02 21:51:15,125 - Epoch: [170][ 870/ 1236] Overall Loss 0.129236 Objective Loss 0.129236 LR 0.000250 Time 0.021788 +2023-10-02 21:51:15,334 - Epoch: [170][ 880/ 1236] Overall Loss 0.129330 Objective Loss 0.129330 LR 0.000250 Time 0.021777 +2023-10-02 21:51:15,542 - Epoch: [170][ 890/ 1236] Overall Loss 0.129458 Objective Loss 0.129458 LR 0.000250 Time 0.021766 +2023-10-02 21:51:15,751 - Epoch: [170][ 900/ 1236] Overall Loss 0.129263 Objective Loss 0.129263 LR 0.000250 Time 0.021755 +2023-10-02 21:51:15,958 - Epoch: [170][ 910/ 1236] Overall Loss 0.129287 Objective Loss 0.129287 LR 0.000250 Time 0.021744 +2023-10-02 21:51:16,165 - Epoch: [170][ 920/ 1236] Overall Loss 0.129330 Objective Loss 0.129330 LR 0.000250 Time 0.021732 +2023-10-02 21:51:16,372 - Epoch: [170][ 930/ 1236] Overall Loss 0.129394 Objective Loss 0.129394 LR 0.000250 Time 0.021719 +2023-10-02 21:51:16,582 - Epoch: [170][ 940/ 1236] Overall Loss 0.129204 Objective Loss 0.129204 LR 0.000250 Time 0.021712 +2023-10-02 21:51:16,790 - Epoch: [170][ 950/ 1236] Overall Loss 0.129173 Objective Loss 0.129173 LR 0.000250 Time 0.021702 +2023-10-02 21:51:17,000 - Epoch: [170][ 960/ 1236] Overall Loss 0.129283 Objective Loss 0.129283 LR 0.000250 Time 0.021694 +2023-10-02 21:51:17,207 - Epoch: [170][ 970/ 1236] Overall Loss 0.129334 Objective Loss 0.129334 LR 0.000250 Time 0.021682 +2023-10-02 21:51:17,415 - Epoch: [170][ 980/ 1236] Overall Loss 0.129348 Objective Loss 0.129348 LR 0.000250 Time 0.021673 +2023-10-02 21:51:17,622 - Epoch: [170][ 990/ 1236] Overall Loss 0.129650 Objective Loss 0.129650 LR 0.000250 Time 0.021662 +2023-10-02 21:51:17,830 - Epoch: [170][ 1000/ 1236] Overall Loss 0.129654 Objective Loss 0.129654 LR 0.000250 Time 0.021653 +2023-10-02 21:51:18,036 - Epoch: [170][ 1010/ 1236] Overall Loss 0.129849 Objective Loss 0.129849 LR 0.000250 Time 0.021643 +2023-10-02 21:51:18,244 - Epoch: [170][ 1020/ 1236] Overall Loss 0.129805 Objective Loss 0.129805 LR 0.000250 Time 0.021634 +2023-10-02 21:51:18,451 - Epoch: [170][ 1030/ 1236] Overall Loss 0.129783 Objective Loss 0.129783 LR 0.000250 Time 0.021625 +2023-10-02 21:51:18,659 - Epoch: [170][ 1040/ 1236] Overall Loss 0.129702 Objective Loss 0.129702 LR 0.000250 Time 0.021616 +2023-10-02 21:51:18,865 - Epoch: [170][ 1050/ 1236] Overall Loss 0.129787 Objective Loss 0.129787 LR 0.000250 Time 0.021607 +2023-10-02 21:51:19,073 - Epoch: [170][ 1060/ 1236] Overall Loss 0.129833 Objective Loss 0.129833 LR 0.000250 Time 0.021599 +2023-10-02 21:51:19,280 - Epoch: [170][ 1070/ 1236] Overall Loss 0.129896 Objective Loss 0.129896 LR 0.000250 Time 0.021590 +2023-10-02 21:51:19,490 - Epoch: [170][ 1080/ 1236] Overall Loss 0.129879 Objective Loss 0.129879 LR 0.000250 Time 0.021584 +2023-10-02 21:51:19,698 - Epoch: [170][ 1090/ 1236] Overall Loss 0.129922 Objective Loss 0.129922 LR 0.000250 Time 0.021576 +2023-10-02 21:51:19,908 - Epoch: [170][ 1100/ 1236] Overall Loss 0.129872 Objective Loss 0.129872 LR 0.000250 Time 0.021571 +2023-10-02 21:51:20,119 - Epoch: [170][ 1110/ 1236] Overall Loss 0.129945 Objective Loss 0.129945 LR 0.000250 Time 0.021566 +2023-10-02 21:51:20,328 - Epoch: [170][ 1120/ 1236] Overall Loss 0.129820 Objective Loss 0.129820 LR 0.000250 Time 0.021560 +2023-10-02 21:51:20,540 - Epoch: [170][ 1130/ 1236] Overall Loss 0.129991 Objective Loss 0.129991 LR 0.000250 Time 0.021556 +2023-10-02 21:51:20,749 - Epoch: [170][ 1140/ 1236] Overall Loss 0.129992 Objective Loss 0.129992 LR 0.000250 Time 0.021551 +2023-10-02 21:51:20,960 - Epoch: [170][ 1150/ 1236] Overall Loss 0.129981 Objective Loss 0.129981 LR 0.000250 Time 0.021546 +2023-10-02 21:51:21,169 - Epoch: [170][ 1160/ 1236] Overall Loss 0.129871 Objective Loss 0.129871 LR 0.000250 Time 0.021541 +2023-10-02 21:51:21,381 - Epoch: [170][ 1170/ 1236] Overall Loss 0.129934 Objective Loss 0.129934 LR 0.000250 Time 0.021537 +2023-10-02 21:51:21,590 - Epoch: [170][ 1180/ 1236] Overall Loss 0.129852 Objective Loss 0.129852 LR 0.000250 Time 0.021532 +2023-10-02 21:51:21,801 - Epoch: [170][ 1190/ 1236] Overall Loss 0.129712 Objective Loss 0.129712 LR 0.000250 Time 0.021528 +2023-10-02 21:51:22,010 - Epoch: [170][ 1200/ 1236] Overall Loss 0.129711 Objective Loss 0.129711 LR 0.000250 Time 0.021522 +2023-10-02 21:51:22,221 - Epoch: [170][ 1210/ 1236] Overall Loss 0.129738 Objective Loss 0.129738 LR 0.000250 Time 0.021519 +2023-10-02 21:51:22,430 - Epoch: [170][ 1220/ 1236] Overall Loss 0.129820 Objective Loss 0.129820 LR 0.000250 Time 0.021513 +2023-10-02 21:51:22,691 - Epoch: [170][ 1230/ 1236] Overall Loss 0.129807 Objective Loss 0.129807 LR 0.000250 Time 0.021550 +2023-10-02 21:51:22,813 - Epoch: [170][ 1236/ 1236] Overall Loss 0.129855 Objective Loss 0.129855 Top1 90.835031 Top5 99.796334 LR 0.000250 Time 0.021544 +2023-10-02 21:51:22,954 - --- validate (epoch=170)----------- +2023-10-02 21:51:22,954 - 29943 samples (256 per mini-batch) +2023-10-02 21:51:23,447 - Epoch: [170][ 10/ 117] Loss 0.313669 Top1 87.265625 Top5 98.437500 +2023-10-02 21:51:23,599 - Epoch: [170][ 20/ 117] Loss 0.314425 Top1 86.601562 Top5 98.671875 +2023-10-02 21:51:23,750 - Epoch: [170][ 30/ 117] Loss 0.307704 Top1 86.744792 Top5 98.619792 +2023-10-02 21:51:23,900 - Epoch: [170][ 40/ 117] Loss 0.305852 Top1 86.796875 Top5 98.652344 +2023-10-02 21:51:24,051 - Epoch: [170][ 50/ 117] Loss 0.304046 Top1 86.898438 Top5 98.687500 +2023-10-02 21:51:24,202 - Epoch: [170][ 60/ 117] Loss 0.296843 Top1 87.096354 Top5 98.691406 +2023-10-02 21:51:24,354 - Epoch: [170][ 70/ 117] Loss 0.299737 Top1 87.047991 Top5 98.649554 +2023-10-02 21:51:24,504 - Epoch: [170][ 80/ 117] Loss 0.298945 Top1 86.958008 Top5 98.657227 +2023-10-02 21:51:24,654 - Epoch: [170][ 90/ 117] Loss 0.304235 Top1 86.888021 Top5 98.641493 +2023-10-02 21:51:24,805 - Epoch: [170][ 100/ 117] Loss 0.307407 Top1 86.902344 Top5 98.652344 +2023-10-02 21:51:24,962 - Epoch: [170][ 110/ 117] Loss 0.310847 Top1 86.885653 Top5 98.647017 +2023-10-02 21:51:25,051 - Epoch: [170][ 117/ 117] Loss 0.308935 Top1 86.971913 Top5 98.684167 +2023-10-02 21:51:25,196 - ==> Top1: 86.972 Top5: 98.684 Loss: 0.309 + +2023-10-02 21:51:25,197 - ==> Confusion: +[[ 951 0 1 0 8 3 0 0 6 54 1 0 0 1 4 2 2 0 0 0 17] + [ 1 1064 1 1 6 17 2 15 0 3 1 0 0 0 2 3 2 0 6 1 6] + [ 2 1 983 6 2 0 12 7 0 1 2 1 9 2 1 4 1 1 9 5 7] + [ 2 3 14 985 1 1 1 0 2 1 3 0 5 3 33 1 2 4 9 1 18] + [ 24 1 1 1 976 2 0 0 1 16 0 0 2 3 8 5 8 0 0 0 2] + [ 3 42 0 2 3 980 1 20 2 5 0 8 2 12 6 1 3 0 4 2 20] + [ 0 5 30 0 0 1 1124 5 0 0 4 1 0 1 0 6 0 0 1 7 6] + [ 3 14 11 0 8 25 7 1060 1 3 4 4 3 4 1 2 0 0 45 9 14] + [ 15 1 0 1 3 3 0 0 981 40 8 0 3 10 14 0 4 0 2 1 3] + [ 104 0 0 0 11 1 1 0 22 951 0 0 1 13 6 1 0 0 0 1 7] + [ 3 3 13 3 1 1 4 1 13 0 970 0 1 14 7 0 3 2 3 1 10] + [ 0 1 2 0 1 14 0 0 0 0 0 961 21 8 0 2 1 14 0 5 5] + [ 1 0 3 4 0 0 3 0 1 2 0 28 975 0 4 7 0 17 0 6 17] + [ 0 0 0 0 4 3 0 0 9 14 1 10 1 1056 4 0 0 1 0 1 15] + [ 12 0 4 16 4 0 0 0 22 2 1 0 2 3 1019 0 0 3 7 0 6] + [ 0 0 1 1 7 0 1 0 0 0 0 4 9 0 0 1070 17 10 3 7 4] + [ 1 17 0 1 4 5 0 0 1 1 0 5 0 4 3 8 1097 0 1 2 11] + [ 0 0 0 2 0 0 2 0 0 0 0 4 21 2 1 8 0 996 0 0 2] + [ 0 4 2 20 0 0 0 11 3 2 3 0 0 0 14 0 3 0 994 0 12] + [ 0 1 4 3 1 0 7 3 0 1 2 14 4 4 0 2 8 1 0 1093 4] + [ 120 120 112 74 74 97 24 61 85 75 139 75 313 236 127 54 83 59 90 131 5756]] + +2023-10-02 21:51:25,199 - ==> Best [Top1: 87.543 Top5: 98.738 Sparsity:0.00 Params: 169472 on epoch: 167] +2023-10-02 21:51:25,199 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:51:25,205 - + +2023-10-02 21:51:25,205 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:51:26,250 - Epoch: [171][ 10/ 1236] Overall Loss 0.133164 Objective Loss 0.133164 LR 0.000250 Time 0.104462 +2023-10-02 21:51:26,458 - Epoch: [171][ 20/ 1236] Overall Loss 0.133862 Objective Loss 0.133862 LR 0.000250 Time 0.062608 +2023-10-02 21:51:26,665 - Epoch: [171][ 30/ 1236] Overall Loss 0.133993 Objective Loss 0.133993 LR 0.000250 Time 0.048637 +2023-10-02 21:51:26,873 - Epoch: [171][ 40/ 1236] Overall Loss 0.132097 Objective Loss 0.132097 LR 0.000250 Time 0.041660 +2023-10-02 21:51:27,080 - Epoch: [171][ 50/ 1236] Overall Loss 0.130612 Objective Loss 0.130612 LR 0.000250 Time 0.037458 +2023-10-02 21:51:27,288 - Epoch: [171][ 60/ 1236] Overall Loss 0.133146 Objective Loss 0.133146 LR 0.000250 Time 0.034677 +2023-10-02 21:51:27,495 - Epoch: [171][ 70/ 1236] Overall Loss 0.132604 Objective Loss 0.132604 LR 0.000250 Time 0.032656 +2023-10-02 21:51:27,702 - Epoch: [171][ 80/ 1236] Overall Loss 0.132523 Objective Loss 0.132523 LR 0.000250 Time 0.031158 +2023-10-02 21:51:27,907 - Epoch: [171][ 90/ 1236] Overall Loss 0.130260 Objective Loss 0.130260 LR 0.000250 Time 0.029959 +2023-10-02 21:51:28,114 - Epoch: [171][ 100/ 1236] Overall Loss 0.131056 Objective Loss 0.131056 LR 0.000250 Time 0.029030 +2023-10-02 21:51:28,319 - Epoch: [171][ 110/ 1236] Overall Loss 0.130373 Objective Loss 0.130373 LR 0.000250 Time 0.028243 +2023-10-02 21:51:28,526 - Epoch: [171][ 120/ 1236] Overall Loss 0.129494 Objective Loss 0.129494 LR 0.000250 Time 0.027612 +2023-10-02 21:51:28,732 - Epoch: [171][ 130/ 1236] Overall Loss 0.129390 Objective Loss 0.129390 LR 0.000250 Time 0.027056 +2023-10-02 21:51:28,939 - Epoch: [171][ 140/ 1236] Overall Loss 0.129706 Objective Loss 0.129706 LR 0.000250 Time 0.026600 +2023-10-02 21:51:29,144 - Epoch: [171][ 150/ 1236] Overall Loss 0.128823 Objective Loss 0.128823 LR 0.000250 Time 0.026186 +2023-10-02 21:51:29,351 - Epoch: [171][ 160/ 1236] Overall Loss 0.129089 Objective Loss 0.129089 LR 0.000250 Time 0.025839 +2023-10-02 21:51:29,556 - Epoch: [171][ 170/ 1236] Overall Loss 0.128521 Objective Loss 0.128521 LR 0.000250 Time 0.025518 +2023-10-02 21:51:29,763 - Epoch: [171][ 180/ 1236] Overall Loss 0.127243 Objective Loss 0.127243 LR 0.000250 Time 0.025249 +2023-10-02 21:51:29,969 - Epoch: [171][ 190/ 1236] Overall Loss 0.127032 Objective Loss 0.127032 LR 0.000250 Time 0.024993 +2023-10-02 21:51:30,176 - Epoch: [171][ 200/ 1236] Overall Loss 0.126650 Objective Loss 0.126650 LR 0.000250 Time 0.024778 +2023-10-02 21:51:30,381 - Epoch: [171][ 210/ 1236] Overall Loss 0.125957 Objective Loss 0.125957 LR 0.000250 Time 0.024568 +2023-10-02 21:51:30,588 - Epoch: [171][ 220/ 1236] Overall Loss 0.126452 Objective Loss 0.126452 LR 0.000250 Time 0.024391 +2023-10-02 21:51:30,794 - Epoch: [171][ 230/ 1236] Overall Loss 0.126445 Objective Loss 0.126445 LR 0.000250 Time 0.024218 +2023-10-02 21:51:31,001 - Epoch: [171][ 240/ 1236] Overall Loss 0.125893 Objective Loss 0.125893 LR 0.000250 Time 0.024070 +2023-10-02 21:51:31,206 - Epoch: [171][ 250/ 1236] Overall Loss 0.125426 Objective Loss 0.125426 LR 0.000250 Time 0.023922 +2023-10-02 21:51:31,413 - Epoch: [171][ 260/ 1236] Overall Loss 0.125333 Objective Loss 0.125333 LR 0.000250 Time 0.023797 +2023-10-02 21:51:31,619 - Epoch: [171][ 270/ 1236] Overall Loss 0.124708 Objective Loss 0.124708 LR 0.000250 Time 0.023672 +2023-10-02 21:51:31,826 - Epoch: [171][ 280/ 1236] Overall Loss 0.124793 Objective Loss 0.124793 LR 0.000250 Time 0.023564 +2023-10-02 21:51:32,031 - Epoch: [171][ 290/ 1236] Overall Loss 0.125708 Objective Loss 0.125708 LR 0.000250 Time 0.023455 +2023-10-02 21:51:32,238 - Epoch: [171][ 300/ 1236] Overall Loss 0.125872 Objective Loss 0.125872 LR 0.000250 Time 0.023362 +2023-10-02 21:51:32,444 - Epoch: [171][ 310/ 1236] Overall Loss 0.126592 Objective Loss 0.126592 LR 0.000250 Time 0.023266 +2023-10-02 21:51:32,651 - Epoch: [171][ 320/ 1236] Overall Loss 0.126229 Objective Loss 0.126229 LR 0.000250 Time 0.023185 +2023-10-02 21:51:32,856 - Epoch: [171][ 330/ 1236] Overall Loss 0.126380 Objective Loss 0.126380 LR 0.000250 Time 0.023101 +2023-10-02 21:51:33,064 - Epoch: [171][ 340/ 1236] Overall Loss 0.126266 Objective Loss 0.126266 LR 0.000250 Time 0.023030 +2023-10-02 21:51:33,269 - Epoch: [171][ 350/ 1236] Overall Loss 0.126114 Objective Loss 0.126114 LR 0.000250 Time 0.022955 +2023-10-02 21:51:33,477 - Epoch: [171][ 360/ 1236] Overall Loss 0.126013 Objective Loss 0.126013 LR 0.000250 Time 0.022893 +2023-10-02 21:51:33,683 - Epoch: [171][ 370/ 1236] Overall Loss 0.126255 Objective Loss 0.126255 LR 0.000250 Time 0.022828 +2023-10-02 21:51:33,891 - Epoch: [171][ 380/ 1236] Overall Loss 0.126193 Objective Loss 0.126193 LR 0.000250 Time 0.022775 +2023-10-02 21:51:34,099 - Epoch: [171][ 390/ 1236] Overall Loss 0.126505 Objective Loss 0.126505 LR 0.000250 Time 0.022722 +2023-10-02 21:51:34,308 - Epoch: [171][ 400/ 1236] Overall Loss 0.127136 Objective Loss 0.127136 LR 0.000250 Time 0.022676 +2023-10-02 21:51:34,516 - Epoch: [171][ 410/ 1236] Overall Loss 0.127082 Objective Loss 0.127082 LR 0.000250 Time 0.022627 +2023-10-02 21:51:34,725 - Epoch: [171][ 420/ 1236] Overall Loss 0.127229 Objective Loss 0.127229 LR 0.000250 Time 0.022583 +2023-10-02 21:51:34,933 - Epoch: [171][ 430/ 1236] Overall Loss 0.127549 Objective Loss 0.127549 LR 0.000250 Time 0.022538 +2023-10-02 21:51:35,142 - Epoch: [171][ 440/ 1236] Overall Loss 0.127290 Objective Loss 0.127290 LR 0.000250 Time 0.022500 +2023-10-02 21:51:35,350 - Epoch: [171][ 450/ 1236] Overall Loss 0.127503 Objective Loss 0.127503 LR 0.000250 Time 0.022459 +2023-10-02 21:51:35,559 - Epoch: [171][ 460/ 1236] Overall Loss 0.127407 Objective Loss 0.127407 LR 0.000250 Time 0.022424 +2023-10-02 21:51:35,766 - Epoch: [171][ 470/ 1236] Overall Loss 0.127499 Objective Loss 0.127499 LR 0.000250 Time 0.022386 +2023-10-02 21:51:35,977 - Epoch: [171][ 480/ 1236] Overall Loss 0.127482 Objective Loss 0.127482 LR 0.000250 Time 0.022357 +2023-10-02 21:51:36,184 - Epoch: [171][ 490/ 1236] Overall Loss 0.127593 Objective Loss 0.127593 LR 0.000250 Time 0.022322 +2023-10-02 21:51:36,393 - Epoch: [171][ 500/ 1236] Overall Loss 0.127657 Objective Loss 0.127657 LR 0.000250 Time 0.022294 +2023-10-02 21:51:36,601 - Epoch: [171][ 510/ 1236] Overall Loss 0.127831 Objective Loss 0.127831 LR 0.000250 Time 0.022261 +2023-10-02 21:51:36,811 - Epoch: [171][ 520/ 1236] Overall Loss 0.128219 Objective Loss 0.128219 LR 0.000250 Time 0.022237 +2023-10-02 21:51:37,018 - Epoch: [171][ 530/ 1236] Overall Loss 0.128287 Objective Loss 0.128287 LR 0.000250 Time 0.022207 +2023-10-02 21:51:37,227 - Epoch: [171][ 540/ 1236] Overall Loss 0.128284 Objective Loss 0.128284 LR 0.000250 Time 0.022183 +2023-10-02 21:51:37,435 - Epoch: [171][ 550/ 1236] Overall Loss 0.128495 Objective Loss 0.128495 LR 0.000250 Time 0.022155 +2023-10-02 21:51:37,651 - Epoch: [171][ 560/ 1236] Overall Loss 0.128549 Objective Loss 0.128549 LR 0.000250 Time 0.022145 +2023-10-02 21:51:37,859 - Epoch: [171][ 570/ 1236] Overall Loss 0.128517 Objective Loss 0.128517 LR 0.000250 Time 0.022121 +2023-10-02 21:51:38,069 - Epoch: [171][ 580/ 1236] Overall Loss 0.128247 Objective Loss 0.128247 LR 0.000250 Time 0.022099 +2023-10-02 21:51:38,277 - Epoch: [171][ 590/ 1236] Overall Loss 0.128267 Objective Loss 0.128267 LR 0.000250 Time 0.022075 +2023-10-02 21:51:38,486 - Epoch: [171][ 600/ 1236] Overall Loss 0.128253 Objective Loss 0.128253 LR 0.000250 Time 0.022055 +2023-10-02 21:51:38,694 - Epoch: [171][ 610/ 1236] Overall Loss 0.128470 Objective Loss 0.128470 LR 0.000250 Time 0.022032 +2023-10-02 21:51:38,905 - Epoch: [171][ 620/ 1236] Overall Loss 0.128727 Objective Loss 0.128727 LR 0.000250 Time 0.022016 +2023-10-02 21:51:39,111 - Epoch: [171][ 630/ 1236] Overall Loss 0.128561 Objective Loss 0.128561 LR 0.000250 Time 0.021994 +2023-10-02 21:51:39,321 - Epoch: [171][ 640/ 1236] Overall Loss 0.128491 Objective Loss 0.128491 LR 0.000250 Time 0.021977 +2023-10-02 21:51:39,529 - Epoch: [171][ 650/ 1236] Overall Loss 0.128505 Objective Loss 0.128505 LR 0.000250 Time 0.021957 +2023-10-02 21:51:39,740 - Epoch: [171][ 660/ 1236] Overall Loss 0.128293 Objective Loss 0.128293 LR 0.000250 Time 0.021943 +2023-10-02 21:51:39,946 - Epoch: [171][ 670/ 1236] Overall Loss 0.128389 Objective Loss 0.128389 LR 0.000250 Time 0.021923 +2023-10-02 21:51:40,154 - Epoch: [171][ 680/ 1236] Overall Loss 0.128218 Objective Loss 0.128218 LR 0.000250 Time 0.021906 +2023-10-02 21:51:40,362 - Epoch: [171][ 690/ 1236] Overall Loss 0.128058 Objective Loss 0.128058 LR 0.000250 Time 0.021887 +2023-10-02 21:51:40,570 - Epoch: [171][ 700/ 1236] Overall Loss 0.128005 Objective Loss 0.128005 LR 0.000250 Time 0.021871 +2023-10-02 21:51:40,777 - Epoch: [171][ 710/ 1236] Overall Loss 0.127696 Objective Loss 0.127696 LR 0.000250 Time 0.021853 +2023-10-02 21:51:40,985 - Epoch: [171][ 720/ 1236] Overall Loss 0.127603 Objective Loss 0.127603 LR 0.000250 Time 0.021838 +2023-10-02 21:51:41,193 - Epoch: [171][ 730/ 1236] Overall Loss 0.127523 Objective Loss 0.127523 LR 0.000250 Time 0.021821 +2023-10-02 21:51:41,401 - Epoch: [171][ 740/ 1236] Overall Loss 0.127418 Objective Loss 0.127418 LR 0.000250 Time 0.021807 +2023-10-02 21:51:41,608 - Epoch: [171][ 750/ 1236] Overall Loss 0.127483 Objective Loss 0.127483 LR 0.000250 Time 0.021790 +2023-10-02 21:51:41,816 - Epoch: [171][ 760/ 1236] Overall Loss 0.127395 Objective Loss 0.127395 LR 0.000250 Time 0.021777 +2023-10-02 21:51:42,023 - Epoch: [171][ 770/ 1236] Overall Loss 0.127330 Objective Loss 0.127330 LR 0.000250 Time 0.021762 +2023-10-02 21:51:42,232 - Epoch: [171][ 780/ 1236] Overall Loss 0.127186 Objective Loss 0.127186 LR 0.000250 Time 0.021749 +2023-10-02 21:51:42,439 - Epoch: [171][ 790/ 1236] Overall Loss 0.127344 Objective Loss 0.127344 LR 0.000250 Time 0.021734 +2023-10-02 21:51:42,647 - Epoch: [171][ 800/ 1236] Overall Loss 0.127362 Objective Loss 0.127362 LR 0.000250 Time 0.021723 +2023-10-02 21:51:42,855 - Epoch: [171][ 810/ 1236] Overall Loss 0.127333 Objective Loss 0.127333 LR 0.000250 Time 0.021708 +2023-10-02 21:51:43,063 - Epoch: [171][ 820/ 1236] Overall Loss 0.127122 Objective Loss 0.127122 LR 0.000250 Time 0.021697 +2023-10-02 21:51:43,270 - Epoch: [171][ 830/ 1236] Overall Loss 0.127140 Objective Loss 0.127140 LR 0.000250 Time 0.021684 +2023-10-02 21:51:43,479 - Epoch: [171][ 840/ 1236] Overall Loss 0.127039 Objective Loss 0.127039 LR 0.000250 Time 0.021673 +2023-10-02 21:51:43,687 - Epoch: [171][ 850/ 1236] Overall Loss 0.126983 Objective Loss 0.126983 LR 0.000250 Time 0.021662 +2023-10-02 21:51:43,897 - Epoch: [171][ 860/ 1236] Overall Loss 0.127076 Objective Loss 0.127076 LR 0.000250 Time 0.021654 +2023-10-02 21:51:44,106 - Epoch: [171][ 870/ 1236] Overall Loss 0.126998 Objective Loss 0.126998 LR 0.000250 Time 0.021643 +2023-10-02 21:51:44,316 - Epoch: [171][ 880/ 1236] Overall Loss 0.127058 Objective Loss 0.127058 LR 0.000250 Time 0.021635 +2023-10-02 21:51:44,525 - Epoch: [171][ 890/ 1236] Overall Loss 0.127057 Objective Loss 0.127057 LR 0.000250 Time 0.021625 +2023-10-02 21:51:44,735 - Epoch: [171][ 900/ 1236] Overall Loss 0.127155 Objective Loss 0.127155 LR 0.000250 Time 0.021618 +2023-10-02 21:51:44,944 - Epoch: [171][ 910/ 1236] Overall Loss 0.127369 Objective Loss 0.127369 LR 0.000250 Time 0.021608 +2023-10-02 21:51:45,154 - Epoch: [171][ 920/ 1236] Overall Loss 0.127735 Objective Loss 0.127735 LR 0.000250 Time 0.021602 +2023-10-02 21:51:45,363 - Epoch: [171][ 930/ 1236] Overall Loss 0.127933 Objective Loss 0.127933 LR 0.000250 Time 0.021592 +2023-10-02 21:51:45,574 - Epoch: [171][ 940/ 1236] Overall Loss 0.128134 Objective Loss 0.128134 LR 0.000250 Time 0.021586 +2023-10-02 21:51:45,783 - Epoch: [171][ 950/ 1236] Overall Loss 0.128218 Objective Loss 0.128218 LR 0.000250 Time 0.021577 +2023-10-02 21:51:45,993 - Epoch: [171][ 960/ 1236] Overall Loss 0.128154 Objective Loss 0.128154 LR 0.000250 Time 0.021571 +2023-10-02 21:51:46,202 - Epoch: [171][ 970/ 1236] Overall Loss 0.128218 Objective Loss 0.128218 LR 0.000250 Time 0.021562 +2023-10-02 21:51:46,412 - Epoch: [171][ 980/ 1236] Overall Loss 0.128386 Objective Loss 0.128386 LR 0.000250 Time 0.021557 +2023-10-02 21:51:46,619 - Epoch: [171][ 990/ 1236] Overall Loss 0.128455 Objective Loss 0.128455 LR 0.000250 Time 0.021548 +2023-10-02 21:51:46,829 - Epoch: [171][ 1000/ 1236] Overall Loss 0.128384 Objective Loss 0.128384 LR 0.000250 Time 0.021542 +2023-10-02 21:51:47,038 - Epoch: [171][ 1010/ 1236] Overall Loss 0.128410 Objective Loss 0.128410 LR 0.000250 Time 0.021534 +2023-10-02 21:51:47,248 - Epoch: [171][ 1020/ 1236] Overall Loss 0.128516 Objective Loss 0.128516 LR 0.000250 Time 0.021529 +2023-10-02 21:51:47,457 - Epoch: [171][ 1030/ 1236] Overall Loss 0.128553 Objective Loss 0.128553 LR 0.000250 Time 0.021521 +2023-10-02 21:51:47,667 - Epoch: [171][ 1040/ 1236] Overall Loss 0.128438 Objective Loss 0.128438 LR 0.000250 Time 0.021516 +2023-10-02 21:51:47,876 - Epoch: [171][ 1050/ 1236] Overall Loss 0.128448 Objective Loss 0.128448 LR 0.000250 Time 0.021508 +2023-10-02 21:51:48,086 - Epoch: [171][ 1060/ 1236] Overall Loss 0.128635 Objective Loss 0.128635 LR 0.000250 Time 0.021503 +2023-10-02 21:51:48,295 - Epoch: [171][ 1070/ 1236] Overall Loss 0.128466 Objective Loss 0.128466 LR 0.000250 Time 0.021496 +2023-10-02 21:51:48,507 - Epoch: [171][ 1080/ 1236] Overall Loss 0.128724 Objective Loss 0.128724 LR 0.000250 Time 0.021492 +2023-10-02 21:51:48,714 - Epoch: [171][ 1090/ 1236] Overall Loss 0.128794 Objective Loss 0.128794 LR 0.000250 Time 0.021485 +2023-10-02 21:51:48,924 - Epoch: [171][ 1100/ 1236] Overall Loss 0.128863 Objective Loss 0.128863 LR 0.000250 Time 0.021480 +2023-10-02 21:51:49,130 - Epoch: [171][ 1110/ 1236] Overall Loss 0.129022 Objective Loss 0.129022 LR 0.000250 Time 0.021472 +2023-10-02 21:51:49,337 - Epoch: [171][ 1120/ 1236] Overall Loss 0.129004 Objective Loss 0.129004 LR 0.000250 Time 0.021465 +2023-10-02 21:51:49,545 - Epoch: [171][ 1130/ 1236] Overall Loss 0.128917 Objective Loss 0.128917 LR 0.000250 Time 0.021459 +2023-10-02 21:51:49,754 - Epoch: [171][ 1140/ 1236] Overall Loss 0.129130 Objective Loss 0.129130 LR 0.000250 Time 0.021453 +2023-10-02 21:51:49,961 - Epoch: [171][ 1150/ 1236] Overall Loss 0.129091 Objective Loss 0.129091 LR 0.000250 Time 0.021446 +2023-10-02 21:51:50,169 - Epoch: [171][ 1160/ 1236] Overall Loss 0.128994 Objective Loss 0.128994 LR 0.000250 Time 0.021440 +2023-10-02 21:51:50,377 - Epoch: [171][ 1170/ 1236] Overall Loss 0.129090 Objective Loss 0.129090 LR 0.000250 Time 0.021433 +2023-10-02 21:51:50,585 - Epoch: [171][ 1180/ 1236] Overall Loss 0.129206 Objective Loss 0.129206 LR 0.000250 Time 0.021428 +2023-10-02 21:51:50,793 - Epoch: [171][ 1190/ 1236] Overall Loss 0.129140 Objective Loss 0.129140 LR 0.000250 Time 0.021420 +2023-10-02 21:51:51,001 - Epoch: [171][ 1200/ 1236] Overall Loss 0.128977 Objective Loss 0.128977 LR 0.000250 Time 0.021415 +2023-10-02 21:51:51,208 - Epoch: [171][ 1210/ 1236] Overall Loss 0.128841 Objective Loss 0.128841 LR 0.000250 Time 0.021408 +2023-10-02 21:51:51,417 - Epoch: [171][ 1220/ 1236] Overall Loss 0.128822 Objective Loss 0.128822 LR 0.000250 Time 0.021403 +2023-10-02 21:51:51,679 - Epoch: [171][ 1230/ 1236] Overall Loss 0.128855 Objective Loss 0.128855 LR 0.000250 Time 0.021442 +2023-10-02 21:51:51,802 - Epoch: [171][ 1236/ 1236] Overall Loss 0.128789 Objective Loss 0.128789 Top1 92.668024 Top5 99.389002 LR 0.000250 Time 0.021436 +2023-10-02 21:51:51,931 - --- validate (epoch=171)----------- +2023-10-02 21:51:51,931 - 29943 samples (256 per mini-batch) +2023-10-02 21:51:52,434 - Epoch: [171][ 10/ 117] Loss 0.286560 Top1 88.320312 Top5 98.632812 +2023-10-02 21:51:52,588 - Epoch: [171][ 20/ 117] Loss 0.302834 Top1 87.929688 Top5 98.691406 +2023-10-02 21:51:52,741 - Epoch: [171][ 30/ 117] Loss 0.300105 Top1 87.708333 Top5 98.763021 +2023-10-02 21:51:52,895 - Epoch: [171][ 40/ 117] Loss 0.304334 Top1 87.441406 Top5 98.779297 +2023-10-02 21:51:53,048 - Epoch: [171][ 50/ 117] Loss 0.305896 Top1 87.164062 Top5 98.710938 +2023-10-02 21:51:53,202 - Epoch: [171][ 60/ 117] Loss 0.302659 Top1 87.233073 Top5 98.730469 +2023-10-02 21:51:53,354 - Epoch: [171][ 70/ 117] Loss 0.301416 Top1 87.321429 Top5 98.716518 +2023-10-02 21:51:53,505 - Epoch: [171][ 80/ 117] Loss 0.301586 Top1 87.304688 Top5 98.730469 +2023-10-02 21:51:53,657 - Epoch: [171][ 90/ 117] Loss 0.302744 Top1 87.222222 Top5 98.728299 +2023-10-02 21:51:53,809 - Epoch: [171][ 100/ 117] Loss 0.307194 Top1 87.250000 Top5 98.660156 +2023-10-02 21:51:53,967 - Epoch: [171][ 110/ 117] Loss 0.305878 Top1 87.240767 Top5 98.668324 +2023-10-02 21:51:54,056 - Epoch: [171][ 117/ 117] Loss 0.303474 Top1 87.265805 Top5 98.674148 +2023-10-02 21:51:54,203 - ==> Top1: 87.266 Top5: 98.674 Loss: 0.303 + +2023-10-02 21:51:54,204 - ==> Confusion: +[[ 950 0 4 0 6 3 0 0 6 51 2 0 0 1 5 2 1 0 1 0 18] + [ 0 1049 1 0 3 20 0 35 0 0 0 1 0 0 0 3 2 0 8 3 6] + [ 2 0 985 4 1 1 16 6 0 2 1 1 5 2 1 3 3 1 12 3 7] + [ 0 3 14 979 1 2 2 1 1 1 2 0 3 4 33 2 1 6 12 1 21] + [ 31 2 1 0 966 5 1 0 1 10 1 0 0 4 8 6 8 0 1 1 4] + [ 4 31 0 1 1 990 1 28 2 4 1 5 2 12 7 1 3 0 3 4 16] + [ 0 2 25 2 0 2 1136 6 0 0 3 1 0 0 0 4 0 0 1 4 5] + [ 4 9 9 1 5 17 5 1101 1 2 1 3 4 4 1 0 0 2 35 7 7] + [ 18 1 0 1 3 5 0 1 977 37 6 1 1 13 13 2 4 1 2 1 2] + [ 109 0 0 0 12 1 0 0 29 929 0 1 0 16 5 3 1 0 0 2 11] + [ 3 3 7 4 0 1 4 1 13 2 975 0 1 12 5 0 3 2 4 2 11] + [ 0 0 1 0 0 14 0 4 0 0 0 973 15 6 0 2 1 11 0 4 4] + [ 0 1 1 4 1 1 3 0 0 1 4 31 973 2 2 8 1 11 3 8 13] + [ 1 0 0 0 2 5 0 0 5 15 2 7 1 1061 4 0 0 1 0 2 13] + [ 13 0 4 18 4 1 0 0 16 1 1 0 2 4 1021 0 0 1 6 0 9] + [ 0 0 1 1 6 0 1 0 0 0 2 5 7 0 0 1073 13 12 2 7 4] + [ 0 15 2 0 4 6 3 0 0 0 0 5 0 3 4 9 1094 0 0 1 15] + [ 1 0 1 2 2 0 2 0 0 1 0 5 18 0 2 6 0 994 1 2 1] + [ 2 3 1 13 0 0 1 24 4 2 2 0 1 0 5 0 1 0 995 0 14] + [ 0 2 4 2 1 0 7 4 0 0 0 13 2 2 0 2 9 0 0 1096 8] + [ 122 110 124 74 58 120 30 92 69 57 140 87 276 238 108 54 74 43 93 123 5813]] + +2023-10-02 21:51:54,205 - ==> Best [Top1: 87.543 Top5: 98.738 Sparsity:0.00 Params: 169472 on epoch: 167] +2023-10-02 21:51:54,205 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:51:54,212 - + +2023-10-02 21:51:54,212 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:51:55,236 - Epoch: [172][ 10/ 1236] Overall Loss 0.119823 Objective Loss 0.119823 LR 0.000250 Time 0.102383 +2023-10-02 21:51:55,444 - Epoch: [172][ 20/ 1236] Overall Loss 0.125404 Objective Loss 0.125404 LR 0.000250 Time 0.061586 +2023-10-02 21:51:55,652 - Epoch: [172][ 30/ 1236] Overall Loss 0.135676 Objective Loss 0.135676 LR 0.000250 Time 0.047958 +2023-10-02 21:51:55,859 - Epoch: [172][ 40/ 1236] Overall Loss 0.130376 Objective Loss 0.130376 LR 0.000250 Time 0.041152 +2023-10-02 21:51:56,066 - Epoch: [172][ 50/ 1236] Overall Loss 0.132468 Objective Loss 0.132468 LR 0.000250 Time 0.037041 +2023-10-02 21:51:56,274 - Epoch: [172][ 60/ 1236] Overall Loss 0.131314 Objective Loss 0.131314 LR 0.000250 Time 0.034326 +2023-10-02 21:51:56,480 - Epoch: [172][ 70/ 1236] Overall Loss 0.129361 Objective Loss 0.129361 LR 0.000250 Time 0.032374 +2023-10-02 21:51:56,688 - Epoch: [172][ 80/ 1236] Overall Loss 0.128720 Objective Loss 0.128720 LR 0.000250 Time 0.030922 +2023-10-02 21:51:56,895 - Epoch: [172][ 90/ 1236] Overall Loss 0.127168 Objective Loss 0.127168 LR 0.000250 Time 0.029766 +2023-10-02 21:51:57,102 - Epoch: [172][ 100/ 1236] Overall Loss 0.127321 Objective Loss 0.127321 LR 0.000250 Time 0.028858 +2023-10-02 21:51:57,309 - Epoch: [172][ 110/ 1236] Overall Loss 0.126861 Objective Loss 0.126861 LR 0.000250 Time 0.028113 +2023-10-02 21:51:57,517 - Epoch: [172][ 120/ 1236] Overall Loss 0.126550 Objective Loss 0.126550 LR 0.000250 Time 0.027502 +2023-10-02 21:51:57,724 - Epoch: [172][ 130/ 1236] Overall Loss 0.127640 Objective Loss 0.127640 LR 0.000250 Time 0.026977 +2023-10-02 21:51:57,933 - Epoch: [172][ 140/ 1236] Overall Loss 0.128601 Objective Loss 0.128601 LR 0.000250 Time 0.026535 +2023-10-02 21:51:58,138 - Epoch: [172][ 150/ 1236] Overall Loss 0.127381 Objective Loss 0.127381 LR 0.000250 Time 0.026124 +2023-10-02 21:51:58,346 - Epoch: [172][ 160/ 1236] Overall Loss 0.127004 Objective Loss 0.127004 LR 0.000250 Time 0.025791 +2023-10-02 21:51:58,553 - Epoch: [172][ 170/ 1236] Overall Loss 0.125892 Objective Loss 0.125892 LR 0.000250 Time 0.025484 +2023-10-02 21:51:58,762 - Epoch: [172][ 180/ 1236] Overall Loss 0.125179 Objective Loss 0.125179 LR 0.000250 Time 0.025222 +2023-10-02 21:51:58,968 - Epoch: [172][ 190/ 1236] Overall Loss 0.124655 Objective Loss 0.124655 LR 0.000250 Time 0.024976 +2023-10-02 21:51:59,176 - Epoch: [172][ 200/ 1236] Overall Loss 0.125012 Objective Loss 0.125012 LR 0.000250 Time 0.024765 +2023-10-02 21:51:59,383 - Epoch: [172][ 210/ 1236] Overall Loss 0.124843 Objective Loss 0.124843 LR 0.000250 Time 0.024564 +2023-10-02 21:51:59,591 - Epoch: [172][ 220/ 1236] Overall Loss 0.125150 Objective Loss 0.125150 LR 0.000250 Time 0.024392 +2023-10-02 21:51:59,798 - Epoch: [172][ 230/ 1236] Overall Loss 0.125806 Objective Loss 0.125806 LR 0.000250 Time 0.024224 +2023-10-02 21:52:00,006 - Epoch: [172][ 240/ 1236] Overall Loss 0.125881 Objective Loss 0.125881 LR 0.000250 Time 0.024081 +2023-10-02 21:52:00,213 - Epoch: [172][ 250/ 1236] Overall Loss 0.125829 Objective Loss 0.125829 LR 0.000250 Time 0.023938 +2023-10-02 21:52:00,421 - Epoch: [172][ 260/ 1236] Overall Loss 0.125753 Objective Loss 0.125753 LR 0.000250 Time 0.023817 +2023-10-02 21:52:00,628 - Epoch: [172][ 270/ 1236] Overall Loss 0.126409 Objective Loss 0.126409 LR 0.000250 Time 0.023696 +2023-10-02 21:52:00,837 - Epoch: [172][ 280/ 1236] Overall Loss 0.126183 Objective Loss 0.126183 LR 0.000250 Time 0.023592 +2023-10-02 21:52:01,043 - Epoch: [172][ 290/ 1236] Overall Loss 0.126104 Objective Loss 0.126104 LR 0.000250 Time 0.023486 +2023-10-02 21:52:01,252 - Epoch: [172][ 300/ 1236] Overall Loss 0.125716 Objective Loss 0.125716 LR 0.000250 Time 0.023396 +2023-10-02 21:52:01,458 - Epoch: [172][ 310/ 1236] Overall Loss 0.125830 Objective Loss 0.125830 LR 0.000250 Time 0.023304 +2023-10-02 21:52:01,667 - Epoch: [172][ 320/ 1236] Overall Loss 0.125721 Objective Loss 0.125721 LR 0.000250 Time 0.023225 +2023-10-02 21:52:01,874 - Epoch: [172][ 330/ 1236] Overall Loss 0.125727 Objective Loss 0.125727 LR 0.000250 Time 0.023146 +2023-10-02 21:52:02,082 - Epoch: [172][ 340/ 1236] Overall Loss 0.126183 Objective Loss 0.126183 LR 0.000250 Time 0.023076 +2023-10-02 21:52:02,289 - Epoch: [172][ 350/ 1236] Overall Loss 0.126000 Objective Loss 0.126000 LR 0.000250 Time 0.023004 +2023-10-02 21:52:02,498 - Epoch: [172][ 360/ 1236] Overall Loss 0.125929 Objective Loss 0.125929 LR 0.000250 Time 0.022942 +2023-10-02 21:52:02,705 - Epoch: [172][ 370/ 1236] Overall Loss 0.125792 Objective Loss 0.125792 LR 0.000250 Time 0.022878 +2023-10-02 21:52:02,913 - Epoch: [172][ 380/ 1236] Overall Loss 0.126190 Objective Loss 0.126190 LR 0.000250 Time 0.022822 +2023-10-02 21:52:03,120 - Epoch: [172][ 390/ 1236] Overall Loss 0.125868 Objective Loss 0.125868 LR 0.000250 Time 0.022764 +2023-10-02 21:52:03,328 - Epoch: [172][ 400/ 1236] Overall Loss 0.125674 Objective Loss 0.125674 LR 0.000250 Time 0.022714 +2023-10-02 21:52:03,535 - Epoch: [172][ 410/ 1236] Overall Loss 0.125897 Objective Loss 0.125897 LR 0.000250 Time 0.022661 +2023-10-02 21:52:03,743 - Epoch: [172][ 420/ 1236] Overall Loss 0.125438 Objective Loss 0.125438 LR 0.000250 Time 0.022616 +2023-10-02 21:52:03,950 - Epoch: [172][ 430/ 1236] Overall Loss 0.126065 Objective Loss 0.126065 LR 0.000250 Time 0.022572 +2023-10-02 21:52:04,158 - Epoch: [172][ 440/ 1236] Overall Loss 0.126283 Objective Loss 0.126283 LR 0.000250 Time 0.022531 +2023-10-02 21:52:04,364 - Epoch: [172][ 450/ 1236] Overall Loss 0.126290 Objective Loss 0.126290 LR 0.000250 Time 0.022486 +2023-10-02 21:52:04,572 - Epoch: [172][ 460/ 1236] Overall Loss 0.126323 Objective Loss 0.126323 LR 0.000250 Time 0.022449 +2023-10-02 21:52:04,778 - Epoch: [172][ 470/ 1236] Overall Loss 0.126186 Objective Loss 0.126186 LR 0.000250 Time 0.022410 +2023-10-02 21:52:04,987 - Epoch: [172][ 480/ 1236] Overall Loss 0.126190 Objective Loss 0.126190 LR 0.000250 Time 0.022377 +2023-10-02 21:52:05,194 - Epoch: [172][ 490/ 1236] Overall Loss 0.126278 Objective Loss 0.126278 LR 0.000250 Time 0.022343 +2023-10-02 21:52:05,403 - Epoch: [172][ 500/ 1236] Overall Loss 0.126557 Objective Loss 0.126557 LR 0.000250 Time 0.022312 +2023-10-02 21:52:05,608 - Epoch: [172][ 510/ 1236] Overall Loss 0.126887 Objective Loss 0.126887 LR 0.000250 Time 0.022277 +2023-10-02 21:52:05,816 - Epoch: [172][ 520/ 1236] Overall Loss 0.126620 Objective Loss 0.126620 LR 0.000250 Time 0.022249 +2023-10-02 21:52:06,024 - Epoch: [172][ 530/ 1236] Overall Loss 0.126535 Objective Loss 0.126535 LR 0.000250 Time 0.022219 +2023-10-02 21:52:06,232 - Epoch: [172][ 540/ 1236] Overall Loss 0.126743 Objective Loss 0.126743 LR 0.000250 Time 0.022192 +2023-10-02 21:52:06,439 - Epoch: [172][ 550/ 1236] Overall Loss 0.126861 Objective Loss 0.126861 LR 0.000250 Time 0.022165 +2023-10-02 21:52:06,647 - Epoch: [172][ 560/ 1236] Overall Loss 0.126813 Objective Loss 0.126813 LR 0.000250 Time 0.022140 +2023-10-02 21:52:06,854 - Epoch: [172][ 570/ 1236] Overall Loss 0.126781 Objective Loss 0.126781 LR 0.000250 Time 0.022115 +2023-10-02 21:52:07,063 - Epoch: [172][ 580/ 1236] Overall Loss 0.126948 Objective Loss 0.126948 LR 0.000250 Time 0.022092 +2023-10-02 21:52:07,269 - Epoch: [172][ 590/ 1236] Overall Loss 0.126614 Objective Loss 0.126614 LR 0.000250 Time 0.022068 +2023-10-02 21:52:07,478 - Epoch: [172][ 600/ 1236] Overall Loss 0.126891 Objective Loss 0.126891 LR 0.000250 Time 0.022047 +2023-10-02 21:52:07,685 - Epoch: [172][ 610/ 1236] Overall Loss 0.126863 Objective Loss 0.126863 LR 0.000250 Time 0.022024 +2023-10-02 21:52:07,893 - Epoch: [172][ 620/ 1236] Overall Loss 0.127212 Objective Loss 0.127212 LR 0.000250 Time 0.022004 +2023-10-02 21:52:08,100 - Epoch: [172][ 630/ 1236] Overall Loss 0.127330 Objective Loss 0.127330 LR 0.000250 Time 0.021983 +2023-10-02 21:52:08,308 - Epoch: [172][ 640/ 1236] Overall Loss 0.127348 Objective Loss 0.127348 LR 0.000250 Time 0.021964 +2023-10-02 21:52:08,515 - Epoch: [172][ 650/ 1236] Overall Loss 0.127430 Objective Loss 0.127430 LR 0.000250 Time 0.021941 +2023-10-02 21:52:08,723 - Epoch: [172][ 660/ 1236] Overall Loss 0.127441 Objective Loss 0.127441 LR 0.000250 Time 0.021924 +2023-10-02 21:52:08,930 - Epoch: [172][ 670/ 1236] Overall Loss 0.127599 Objective Loss 0.127599 LR 0.000250 Time 0.021905 +2023-10-02 21:52:09,138 - Epoch: [172][ 680/ 1236] Overall Loss 0.127648 Objective Loss 0.127648 LR 0.000250 Time 0.021889 +2023-10-02 21:52:09,345 - Epoch: [172][ 690/ 1236] Overall Loss 0.127334 Objective Loss 0.127334 LR 0.000250 Time 0.021871 +2023-10-02 21:52:09,553 - Epoch: [172][ 700/ 1236] Overall Loss 0.127331 Objective Loss 0.127331 LR 0.000250 Time 0.021855 +2023-10-02 21:52:09,760 - Epoch: [172][ 710/ 1236] Overall Loss 0.127495 Objective Loss 0.127495 LR 0.000250 Time 0.021838 +2023-10-02 21:52:09,968 - Epoch: [172][ 720/ 1236] Overall Loss 0.127684 Objective Loss 0.127684 LR 0.000250 Time 0.021823 +2023-10-02 21:52:10,175 - Epoch: [172][ 730/ 1236] Overall Loss 0.127832 Objective Loss 0.127832 LR 0.000250 Time 0.021809 +2023-10-02 21:52:10,384 - Epoch: [172][ 740/ 1236] Overall Loss 0.127679 Objective Loss 0.127679 LR 0.000250 Time 0.021795 +2023-10-02 21:52:10,591 - Epoch: [172][ 750/ 1236] Overall Loss 0.127488 Objective Loss 0.127488 LR 0.000250 Time 0.021780 +2023-10-02 21:52:10,799 - Epoch: [172][ 760/ 1236] Overall Loss 0.127359 Objective Loss 0.127359 LR 0.000250 Time 0.021767 +2023-10-02 21:52:11,006 - Epoch: [172][ 770/ 1236] Overall Loss 0.127378 Objective Loss 0.127378 LR 0.000250 Time 0.021752 +2023-10-02 21:52:11,214 - Epoch: [172][ 780/ 1236] Overall Loss 0.127428 Objective Loss 0.127428 LR 0.000250 Time 0.021740 +2023-10-02 21:52:11,421 - Epoch: [172][ 790/ 1236] Overall Loss 0.127459 Objective Loss 0.127459 LR 0.000250 Time 0.021727 +2023-10-02 21:52:11,630 - Epoch: [172][ 800/ 1236] Overall Loss 0.127536 Objective Loss 0.127536 LR 0.000250 Time 0.021715 +2023-10-02 21:52:11,837 - Epoch: [172][ 810/ 1236] Overall Loss 0.127437 Objective Loss 0.127437 LR 0.000250 Time 0.021703 +2023-10-02 21:52:12,045 - Epoch: [172][ 820/ 1236] Overall Loss 0.127413 Objective Loss 0.127413 LR 0.000250 Time 0.021691 +2023-10-02 21:52:12,253 - Epoch: [172][ 830/ 1236] Overall Loss 0.127488 Objective Loss 0.127488 LR 0.000250 Time 0.021680 +2023-10-02 21:52:12,461 - Epoch: [172][ 840/ 1236] Overall Loss 0.127419 Objective Loss 0.127419 LR 0.000250 Time 0.021669 +2023-10-02 21:52:12,668 - Epoch: [172][ 850/ 1236] Overall Loss 0.127295 Objective Loss 0.127295 LR 0.000250 Time 0.021657 +2023-10-02 21:52:12,876 - Epoch: [172][ 860/ 1236] Overall Loss 0.127228 Objective Loss 0.127228 LR 0.000250 Time 0.021647 +2023-10-02 21:52:13,083 - Epoch: [172][ 870/ 1236] Overall Loss 0.127523 Objective Loss 0.127523 LR 0.000250 Time 0.021634 +2023-10-02 21:52:13,291 - Epoch: [172][ 880/ 1236] Overall Loss 0.127682 Objective Loss 0.127682 LR 0.000250 Time 0.021624 +2023-10-02 21:52:13,499 - Epoch: [172][ 890/ 1236] Overall Loss 0.127562 Objective Loss 0.127562 LR 0.000250 Time 0.021614 +2023-10-02 21:52:13,707 - Epoch: [172][ 900/ 1236] Overall Loss 0.127758 Objective Loss 0.127758 LR 0.000250 Time 0.021605 +2023-10-02 21:52:13,914 - Epoch: [172][ 910/ 1236] Overall Loss 0.127848 Objective Loss 0.127848 LR 0.000250 Time 0.021594 +2023-10-02 21:52:14,122 - Epoch: [172][ 920/ 1236] Overall Loss 0.127724 Objective Loss 0.127724 LR 0.000250 Time 0.021584 +2023-10-02 21:52:14,329 - Epoch: [172][ 930/ 1236] Overall Loss 0.127806 Objective Loss 0.127806 LR 0.000250 Time 0.021574 +2023-10-02 21:52:14,537 - Epoch: [172][ 940/ 1236] Overall Loss 0.128006 Objective Loss 0.128006 LR 0.000250 Time 0.021565 +2023-10-02 21:52:14,744 - Epoch: [172][ 950/ 1236] Overall Loss 0.127924 Objective Loss 0.127924 LR 0.000250 Time 0.021556 +2023-10-02 21:52:14,953 - Epoch: [172][ 960/ 1236] Overall Loss 0.128111 Objective Loss 0.128111 LR 0.000250 Time 0.021548 +2023-10-02 21:52:15,160 - Epoch: [172][ 970/ 1236] Overall Loss 0.128122 Objective Loss 0.128122 LR 0.000250 Time 0.021539 +2023-10-02 21:52:15,368 - Epoch: [172][ 980/ 1236] Overall Loss 0.128223 Objective Loss 0.128223 LR 0.000250 Time 0.021531 +2023-10-02 21:52:15,575 - Epoch: [172][ 990/ 1236] Overall Loss 0.128279 Objective Loss 0.128279 LR 0.000250 Time 0.021523 +2023-10-02 21:52:15,783 - Epoch: [172][ 1000/ 1236] Overall Loss 0.128176 Objective Loss 0.128176 LR 0.000250 Time 0.021515 +2023-10-02 21:52:15,990 - Epoch: [172][ 1010/ 1236] Overall Loss 0.128251 Objective Loss 0.128251 LR 0.000250 Time 0.021507 +2023-10-02 21:52:16,199 - Epoch: [172][ 1020/ 1236] Overall Loss 0.128316 Objective Loss 0.128316 LR 0.000250 Time 0.021500 +2023-10-02 21:52:16,406 - Epoch: [172][ 1030/ 1236] Overall Loss 0.128260 Objective Loss 0.128260 LR 0.000250 Time 0.021492 +2023-10-02 21:52:16,614 - Epoch: [172][ 1040/ 1236] Overall Loss 0.128214 Objective Loss 0.128214 LR 0.000250 Time 0.021485 +2023-10-02 21:52:16,821 - Epoch: [172][ 1050/ 1236] Overall Loss 0.128226 Objective Loss 0.128226 LR 0.000250 Time 0.021476 +2023-10-02 21:52:17,029 - Epoch: [172][ 1060/ 1236] Overall Loss 0.128097 Objective Loss 0.128097 LR 0.000250 Time 0.021469 +2023-10-02 21:52:17,236 - Epoch: [172][ 1070/ 1236] Overall Loss 0.128029 Objective Loss 0.128029 LR 0.000250 Time 0.021461 +2023-10-02 21:52:17,444 - Epoch: [172][ 1080/ 1236] Overall Loss 0.127970 Objective Loss 0.127970 LR 0.000250 Time 0.021454 +2023-10-02 21:52:17,650 - Epoch: [172][ 1090/ 1236] Overall Loss 0.128108 Objective Loss 0.128108 LR 0.000250 Time 0.021446 +2023-10-02 21:52:17,857 - Epoch: [172][ 1100/ 1236] Overall Loss 0.128077 Objective Loss 0.128077 LR 0.000250 Time 0.021439 +2023-10-02 21:52:18,064 - Epoch: [172][ 1110/ 1236] Overall Loss 0.128156 Objective Loss 0.128156 LR 0.000250 Time 0.021432 +2023-10-02 21:52:18,272 - Epoch: [172][ 1120/ 1236] Overall Loss 0.128291 Objective Loss 0.128291 LR 0.000250 Time 0.021426 +2023-10-02 21:52:18,478 - Epoch: [172][ 1130/ 1236] Overall Loss 0.128214 Objective Loss 0.128214 LR 0.000250 Time 0.021418 +2023-10-02 21:52:18,686 - Epoch: [172][ 1140/ 1236] Overall Loss 0.128328 Objective Loss 0.128328 LR 0.000250 Time 0.021413 +2023-10-02 21:52:18,893 - Epoch: [172][ 1150/ 1236] Overall Loss 0.128380 Objective Loss 0.128380 LR 0.000250 Time 0.021407 +2023-10-02 21:52:19,102 - Epoch: [172][ 1160/ 1236] Overall Loss 0.128281 Objective Loss 0.128281 LR 0.000250 Time 0.021402 +2023-10-02 21:52:19,309 - Epoch: [172][ 1170/ 1236] Overall Loss 0.128387 Objective Loss 0.128387 LR 0.000250 Time 0.021395 +2023-10-02 21:52:19,517 - Epoch: [172][ 1180/ 1236] Overall Loss 0.128393 Objective Loss 0.128393 LR 0.000250 Time 0.021390 +2023-10-02 21:52:19,726 - Epoch: [172][ 1190/ 1236] Overall Loss 0.128282 Objective Loss 0.128282 LR 0.000250 Time 0.021386 +2023-10-02 21:52:19,938 - Epoch: [172][ 1200/ 1236] Overall Loss 0.128281 Objective Loss 0.128281 LR 0.000250 Time 0.021384 +2023-10-02 21:52:20,149 - Epoch: [172][ 1210/ 1236] Overall Loss 0.128282 Objective Loss 0.128282 LR 0.000250 Time 0.021381 +2023-10-02 21:52:20,359 - Epoch: [172][ 1220/ 1236] Overall Loss 0.128292 Objective Loss 0.128292 LR 0.000250 Time 0.021378 +2023-10-02 21:52:20,623 - Epoch: [172][ 1230/ 1236] Overall Loss 0.128361 Objective Loss 0.128361 LR 0.000250 Time 0.021419 +2023-10-02 21:52:20,746 - Epoch: [172][ 1236/ 1236] Overall Loss 0.128326 Objective Loss 0.128326 Top1 92.464358 Top5 99.592668 LR 0.000250 Time 0.021414 +2023-10-02 21:52:20,886 - --- validate (epoch=172)----------- +2023-10-02 21:52:20,886 - 29943 samples (256 per mini-batch) +2023-10-02 21:52:21,379 - Epoch: [172][ 10/ 117] Loss 0.286393 Top1 88.320312 Top5 98.945312 +2023-10-02 21:52:21,527 - Epoch: [172][ 20/ 117] Loss 0.291468 Top1 88.105469 Top5 98.847656 +2023-10-02 21:52:21,676 - Epoch: [172][ 30/ 117] Loss 0.296111 Top1 88.085938 Top5 98.854167 +2023-10-02 21:52:21,826 - Epoch: [172][ 40/ 117] Loss 0.304158 Top1 87.617188 Top5 98.789062 +2023-10-02 21:52:21,974 - Epoch: [172][ 50/ 117] Loss 0.295183 Top1 87.585938 Top5 98.804688 +2023-10-02 21:52:22,122 - Epoch: [172][ 60/ 117] Loss 0.299664 Top1 87.636719 Top5 98.776042 +2023-10-02 21:52:22,269 - Epoch: [172][ 70/ 117] Loss 0.301886 Top1 87.566964 Top5 98.727679 +2023-10-02 21:52:22,417 - Epoch: [172][ 80/ 117] Loss 0.297967 Top1 87.602539 Top5 98.740234 +2023-10-02 21:52:22,564 - Epoch: [172][ 90/ 117] Loss 0.298694 Top1 87.625868 Top5 98.710938 +2023-10-02 21:52:22,711 - Epoch: [172][ 100/ 117] Loss 0.303220 Top1 87.488281 Top5 98.671875 +2023-10-02 21:52:22,865 - Epoch: [172][ 110/ 117] Loss 0.303284 Top1 87.549716 Top5 98.647017 +2023-10-02 21:52:22,954 - Epoch: [172][ 117/ 117] Loss 0.300157 Top1 87.573055 Top5 98.677487 +2023-10-02 21:52:23,053 - ==> Top1: 87.573 Top5: 98.677 Loss: 0.300 + +2023-10-02 21:52:23,054 - ==> Confusion: +[[ 947 1 4 0 6 2 0 0 5 53 2 0 0 1 5 1 0 1 1 0 21] + [ 0 1072 0 2 4 12 0 23 0 0 0 0 0 0 1 2 1 0 6 2 6] + [ 1 0 994 4 1 0 10 5 0 2 0 0 6 2 1 3 1 2 13 3 8] + [ 0 4 14 978 0 0 3 4 3 0 3 1 7 3 29 1 1 3 15 1 19] + [ 24 6 1 0 976 4 0 0 0 8 1 0 1 3 9 4 7 0 0 2 4] + [ 4 37 0 1 2 980 1 38 1 5 2 4 0 7 6 0 1 0 6 4 17] + [ 0 3 32 0 0 1 1129 5 0 0 2 1 0 0 0 5 0 1 2 6 4] + [ 3 10 12 0 6 12 5 1091 1 4 3 3 3 4 1 1 0 4 39 9 7] + [ 18 4 0 1 1 5 0 1 983 35 10 1 0 7 11 1 1 1 2 3 4] + [ 98 1 1 1 9 4 0 0 32 941 1 1 0 11 6 2 0 0 0 1 10] + [ 3 2 12 8 0 2 3 1 8 2 970 0 0 11 3 0 3 3 7 1 14] + [ 0 0 1 0 0 13 0 4 0 0 0 974 12 6 0 1 2 13 0 3 6] + [ 1 0 1 3 0 0 1 1 0 2 4 34 979 1 1 5 1 10 2 6 16] + [ 1 0 1 0 3 10 0 0 7 13 4 8 1 1047 5 0 0 1 0 0 18] + [ 13 0 3 20 5 1 0 0 17 0 0 1 2 1 1012 0 1 1 13 0 11] + [ 0 0 2 2 5 1 1 0 0 0 0 5 7 0 0 1071 18 9 2 7 4] + [ 1 19 1 0 4 4 2 0 1 0 0 2 0 2 3 9 1094 0 0 4 15] + [ 0 1 0 1 0 0 2 0 0 1 0 6 18 1 4 6 0 993 0 3 2] + [ 2 6 3 14 1 1 0 19 3 1 1 0 0 0 6 0 0 0 999 0 12] + [ 0 1 3 2 1 0 8 4 0 0 2 12 3 2 0 2 13 2 0 1092 5] + [ 110 124 118 83 56 83 22 95 71 62 130 82 263 210 98 45 67 59 95 132 5900]] + +2023-10-02 21:52:23,055 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:52:23,055 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:52:23,068 - + +2023-10-02 21:52:23,069 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:52:24,212 - Epoch: [173][ 10/ 1236] Overall Loss 0.138040 Objective Loss 0.138040 LR 0.000250 Time 0.114312 +2023-10-02 21:52:24,421 - Epoch: [173][ 20/ 1236] Overall Loss 0.125744 Objective Loss 0.125744 LR 0.000250 Time 0.067542 +2023-10-02 21:52:24,628 - Epoch: [173][ 30/ 1236] Overall Loss 0.122557 Objective Loss 0.122557 LR 0.000250 Time 0.051921 +2023-10-02 21:52:24,835 - Epoch: [173][ 40/ 1236] Overall Loss 0.126266 Objective Loss 0.126266 LR 0.000250 Time 0.044125 +2023-10-02 21:52:25,042 - Epoch: [173][ 50/ 1236] Overall Loss 0.124135 Objective Loss 0.124135 LR 0.000250 Time 0.039435 +2023-10-02 21:52:25,250 - Epoch: [173][ 60/ 1236] Overall Loss 0.121096 Objective Loss 0.121096 LR 0.000250 Time 0.036314 +2023-10-02 21:52:25,457 - Epoch: [173][ 70/ 1236] Overall Loss 0.119842 Objective Loss 0.119842 LR 0.000250 Time 0.034076 +2023-10-02 21:52:25,664 - Epoch: [173][ 80/ 1236] Overall Loss 0.120390 Objective Loss 0.120390 LR 0.000250 Time 0.032405 +2023-10-02 21:52:25,871 - Epoch: [173][ 90/ 1236] Overall Loss 0.121321 Objective Loss 0.121321 LR 0.000250 Time 0.031086 +2023-10-02 21:52:26,079 - Epoch: [173][ 100/ 1236] Overall Loss 0.121961 Objective Loss 0.121961 LR 0.000250 Time 0.030052 +2023-10-02 21:52:26,285 - Epoch: [173][ 110/ 1236] Overall Loss 0.122431 Objective Loss 0.122431 LR 0.000250 Time 0.029198 +2023-10-02 21:52:26,492 - Epoch: [173][ 120/ 1236] Overall Loss 0.121143 Objective Loss 0.121143 LR 0.000250 Time 0.028480 +2023-10-02 21:52:26,698 - Epoch: [173][ 130/ 1236] Overall Loss 0.122990 Objective Loss 0.122990 LR 0.000250 Time 0.027867 +2023-10-02 21:52:26,906 - Epoch: [173][ 140/ 1236] Overall Loss 0.123954 Objective Loss 0.123954 LR 0.000250 Time 0.027358 +2023-10-02 21:52:27,113 - Epoch: [173][ 150/ 1236] Overall Loss 0.123406 Objective Loss 0.123406 LR 0.000250 Time 0.026902 +2023-10-02 21:52:27,321 - Epoch: [173][ 160/ 1236] Overall Loss 0.123817 Objective Loss 0.123817 LR 0.000250 Time 0.026519 +2023-10-02 21:52:27,528 - Epoch: [173][ 170/ 1236] Overall Loss 0.123195 Objective Loss 0.123195 LR 0.000250 Time 0.026167 +2023-10-02 21:52:27,735 - Epoch: [173][ 180/ 1236] Overall Loss 0.122734 Objective Loss 0.122734 LR 0.000250 Time 0.025866 +2023-10-02 21:52:27,943 - Epoch: [173][ 190/ 1236] Overall Loss 0.123388 Objective Loss 0.123388 LR 0.000250 Time 0.025587 +2023-10-02 21:52:28,149 - Epoch: [173][ 200/ 1236] Overall Loss 0.123398 Objective Loss 0.123398 LR 0.000250 Time 0.025340 +2023-10-02 21:52:28,356 - Epoch: [173][ 210/ 1236] Overall Loss 0.123956 Objective Loss 0.123956 LR 0.000250 Time 0.025115 +2023-10-02 21:52:28,564 - Epoch: [173][ 220/ 1236] Overall Loss 0.123775 Objective Loss 0.123775 LR 0.000250 Time 0.024917 +2023-10-02 21:52:28,770 - Epoch: [173][ 230/ 1236] Overall Loss 0.124125 Objective Loss 0.124125 LR 0.000250 Time 0.024726 +2023-10-02 21:52:28,977 - Epoch: [173][ 240/ 1236] Overall Loss 0.124479 Objective Loss 0.124479 LR 0.000250 Time 0.024553 +2023-10-02 21:52:29,183 - Epoch: [173][ 250/ 1236] Overall Loss 0.123762 Objective Loss 0.123762 LR 0.000250 Time 0.024392 +2023-10-02 21:52:29,392 - Epoch: [173][ 260/ 1236] Overall Loss 0.123467 Objective Loss 0.123467 LR 0.000250 Time 0.024254 +2023-10-02 21:52:29,598 - Epoch: [173][ 270/ 1236] Overall Loss 0.123104 Objective Loss 0.123104 LR 0.000250 Time 0.024115 +2023-10-02 21:52:29,805 - Epoch: [173][ 280/ 1236] Overall Loss 0.122838 Objective Loss 0.122838 LR 0.000250 Time 0.023991 +2023-10-02 21:52:30,012 - Epoch: [173][ 290/ 1236] Overall Loss 0.123140 Objective Loss 0.123140 LR 0.000250 Time 0.023872 +2023-10-02 21:52:30,220 - Epoch: [173][ 300/ 1236] Overall Loss 0.123644 Objective Loss 0.123644 LR 0.000250 Time 0.023768 +2023-10-02 21:52:30,426 - Epoch: [173][ 310/ 1236] Overall Loss 0.124050 Objective Loss 0.124050 LR 0.000250 Time 0.023663 +2023-10-02 21:52:30,633 - Epoch: [173][ 320/ 1236] Overall Loss 0.124981 Objective Loss 0.124981 LR 0.000250 Time 0.023569 +2023-10-02 21:52:30,841 - Epoch: [173][ 330/ 1236] Overall Loss 0.124883 Objective Loss 0.124883 LR 0.000250 Time 0.023482 +2023-10-02 21:52:31,049 - Epoch: [173][ 340/ 1236] Overall Loss 0.124935 Objective Loss 0.124935 LR 0.000250 Time 0.023403 +2023-10-02 21:52:31,256 - Epoch: [173][ 350/ 1236] Overall Loss 0.124976 Objective Loss 0.124976 LR 0.000250 Time 0.023321 +2023-10-02 21:52:31,464 - Epoch: [173][ 360/ 1236] Overall Loss 0.124584 Objective Loss 0.124584 LR 0.000250 Time 0.023250 +2023-10-02 21:52:31,671 - Epoch: [173][ 370/ 1236] Overall Loss 0.124420 Objective Loss 0.124420 LR 0.000250 Time 0.023181 +2023-10-02 21:52:31,879 - Epoch: [173][ 380/ 1236] Overall Loss 0.124413 Objective Loss 0.124413 LR 0.000250 Time 0.023117 +2023-10-02 21:52:32,087 - Epoch: [173][ 390/ 1236] Overall Loss 0.124539 Objective Loss 0.124539 LR 0.000250 Time 0.023055 +2023-10-02 21:52:32,298 - Epoch: [173][ 400/ 1236] Overall Loss 0.124589 Objective Loss 0.124589 LR 0.000250 Time 0.023004 +2023-10-02 21:52:32,509 - Epoch: [173][ 410/ 1236] Overall Loss 0.124408 Objective Loss 0.124408 LR 0.000250 Time 0.022955 +2023-10-02 21:52:32,718 - Epoch: [173][ 420/ 1236] Overall Loss 0.124570 Objective Loss 0.124570 LR 0.000250 Time 0.022900 +2023-10-02 21:52:32,929 - Epoch: [173][ 430/ 1236] Overall Loss 0.124708 Objective Loss 0.124708 LR 0.000250 Time 0.022858 +2023-10-02 21:52:33,136 - Epoch: [173][ 440/ 1236] Overall Loss 0.124685 Objective Loss 0.124685 LR 0.000250 Time 0.022808 +2023-10-02 21:52:33,347 - Epoch: [173][ 450/ 1236] Overall Loss 0.124683 Objective Loss 0.124683 LR 0.000250 Time 0.022769 +2023-10-02 21:52:33,554 - Epoch: [173][ 460/ 1236] Overall Loss 0.124515 Objective Loss 0.124515 LR 0.000250 Time 0.022724 +2023-10-02 21:52:33,763 - Epoch: [173][ 470/ 1236] Overall Loss 0.124430 Objective Loss 0.124430 LR 0.000250 Time 0.022685 +2023-10-02 21:52:33,968 - Epoch: [173][ 480/ 1236] Overall Loss 0.124420 Objective Loss 0.124420 LR 0.000250 Time 0.022639 +2023-10-02 21:52:34,175 - Epoch: [173][ 490/ 1236] Overall Loss 0.124509 Objective Loss 0.124509 LR 0.000250 Time 0.022599 +2023-10-02 21:52:34,381 - Epoch: [173][ 500/ 1236] Overall Loss 0.124674 Objective Loss 0.124674 LR 0.000250 Time 0.022556 +2023-10-02 21:52:34,590 - Epoch: [173][ 510/ 1236] Overall Loss 0.124935 Objective Loss 0.124935 LR 0.000250 Time 0.022522 +2023-10-02 21:52:34,795 - Epoch: [173][ 520/ 1236] Overall Loss 0.125295 Objective Loss 0.125295 LR 0.000250 Time 0.022483 +2023-10-02 21:52:35,003 - Epoch: [173][ 530/ 1236] Overall Loss 0.125476 Objective Loss 0.125476 LR 0.000250 Time 0.022451 +2023-10-02 21:52:35,208 - Epoch: [173][ 540/ 1236] Overall Loss 0.125434 Objective Loss 0.125434 LR 0.000250 Time 0.022414 +2023-10-02 21:52:35,416 - Epoch: [173][ 550/ 1236] Overall Loss 0.125718 Objective Loss 0.125718 LR 0.000250 Time 0.022385 +2023-10-02 21:52:35,621 - Epoch: [173][ 560/ 1236] Overall Loss 0.125726 Objective Loss 0.125726 LR 0.000250 Time 0.022350 +2023-10-02 21:52:35,829 - Epoch: [173][ 570/ 1236] Overall Loss 0.125695 Objective Loss 0.125695 LR 0.000250 Time 0.022323 +2023-10-02 21:52:36,034 - Epoch: [173][ 580/ 1236] Overall Loss 0.125650 Objective Loss 0.125650 LR 0.000250 Time 0.022291 +2023-10-02 21:52:36,243 - Epoch: [173][ 590/ 1236] Overall Loss 0.125515 Objective Loss 0.125515 LR 0.000250 Time 0.022266 +2023-10-02 21:52:36,448 - Epoch: [173][ 600/ 1236] Overall Loss 0.125539 Objective Loss 0.125539 LR 0.000250 Time 0.022236 +2023-10-02 21:52:36,656 - Epoch: [173][ 610/ 1236] Overall Loss 0.125610 Objective Loss 0.125610 LR 0.000250 Time 0.022213 +2023-10-02 21:52:36,862 - Epoch: [173][ 620/ 1236] Overall Loss 0.125516 Objective Loss 0.125516 LR 0.000250 Time 0.022185 +2023-10-02 21:52:37,070 - Epoch: [173][ 630/ 1236] Overall Loss 0.125656 Objective Loss 0.125656 LR 0.000250 Time 0.022163 +2023-10-02 21:52:37,275 - Epoch: [173][ 640/ 1236] Overall Loss 0.125945 Objective Loss 0.125945 LR 0.000250 Time 0.022137 +2023-10-02 21:52:37,483 - Epoch: [173][ 650/ 1236] Overall Loss 0.126059 Objective Loss 0.126059 LR 0.000250 Time 0.022115 +2023-10-02 21:52:37,689 - Epoch: [173][ 660/ 1236] Overall Loss 0.126128 Objective Loss 0.126128 LR 0.000250 Time 0.022090 +2023-10-02 21:52:37,896 - Epoch: [173][ 670/ 1236] Overall Loss 0.126190 Objective Loss 0.126190 LR 0.000250 Time 0.022069 +2023-10-02 21:52:38,102 - Epoch: [173][ 680/ 1236] Overall Loss 0.126031 Objective Loss 0.126031 LR 0.000250 Time 0.022046 +2023-10-02 21:52:38,311 - Epoch: [173][ 690/ 1236] Overall Loss 0.126331 Objective Loss 0.126331 LR 0.000250 Time 0.022028 +2023-10-02 21:52:38,516 - Epoch: [173][ 700/ 1236] Overall Loss 0.126227 Objective Loss 0.126227 LR 0.000250 Time 0.022006 +2023-10-02 21:52:38,723 - Epoch: [173][ 710/ 1236] Overall Loss 0.126275 Objective Loss 0.126275 LR 0.000250 Time 0.021987 +2023-10-02 21:52:38,929 - Epoch: [173][ 720/ 1236] Overall Loss 0.126380 Objective Loss 0.126380 LR 0.000250 Time 0.021966 +2023-10-02 21:52:39,138 - Epoch: [173][ 730/ 1236] Overall Loss 0.126610 Objective Loss 0.126610 LR 0.000250 Time 0.021951 +2023-10-02 21:52:39,342 - Epoch: [173][ 740/ 1236] Overall Loss 0.126371 Objective Loss 0.126371 LR 0.000250 Time 0.021930 +2023-10-02 21:52:39,551 - Epoch: [173][ 750/ 1236] Overall Loss 0.126500 Objective Loss 0.126500 LR 0.000250 Time 0.021916 +2023-10-02 21:52:39,756 - Epoch: [173][ 760/ 1236] Overall Loss 0.126749 Objective Loss 0.126749 LR 0.000250 Time 0.021897 +2023-10-02 21:52:39,965 - Epoch: [173][ 770/ 1236] Overall Loss 0.126699 Objective Loss 0.126699 LR 0.000250 Time 0.021883 +2023-10-02 21:52:40,170 - Epoch: [173][ 780/ 1236] Overall Loss 0.126629 Objective Loss 0.126629 LR 0.000250 Time 0.021865 +2023-10-02 21:52:40,377 - Epoch: [173][ 790/ 1236] Overall Loss 0.126615 Objective Loss 0.126615 LR 0.000250 Time 0.021850 +2023-10-02 21:52:40,584 - Epoch: [173][ 800/ 1236] Overall Loss 0.126877 Objective Loss 0.126877 LR 0.000250 Time 0.021834 +2023-10-02 21:52:40,792 - Epoch: [173][ 810/ 1236] Overall Loss 0.126986 Objective Loss 0.126986 LR 0.000250 Time 0.021820 +2023-10-02 21:52:40,998 - Epoch: [173][ 820/ 1236] Overall Loss 0.126815 Objective Loss 0.126815 LR 0.000250 Time 0.021804 +2023-10-02 21:52:41,206 - Epoch: [173][ 830/ 1236] Overall Loss 0.126723 Objective Loss 0.126723 LR 0.000250 Time 0.021792 +2023-10-02 21:52:41,411 - Epoch: [173][ 840/ 1236] Overall Loss 0.126607 Objective Loss 0.126607 LR 0.000250 Time 0.021776 +2023-10-02 21:52:41,619 - Epoch: [173][ 850/ 1236] Overall Loss 0.126606 Objective Loss 0.126606 LR 0.000250 Time 0.021763 +2023-10-02 21:52:41,825 - Epoch: [173][ 860/ 1236] Overall Loss 0.126688 Objective Loss 0.126688 LR 0.000250 Time 0.021749 +2023-10-02 21:52:42,033 - Epoch: [173][ 870/ 1236] Overall Loss 0.126833 Objective Loss 0.126833 LR 0.000250 Time 0.021737 +2023-10-02 21:52:42,239 - Epoch: [173][ 880/ 1236] Overall Loss 0.126836 Objective Loss 0.126836 LR 0.000250 Time 0.021723 +2023-10-02 21:52:42,448 - Epoch: [173][ 890/ 1236] Overall Loss 0.126827 Objective Loss 0.126827 LR 0.000250 Time 0.021713 +2023-10-02 21:52:42,653 - Epoch: [173][ 900/ 1236] Overall Loss 0.126783 Objective Loss 0.126783 LR 0.000250 Time 0.021699 +2023-10-02 21:52:42,861 - Epoch: [173][ 910/ 1236] Overall Loss 0.126798 Objective Loss 0.126798 LR 0.000250 Time 0.021689 +2023-10-02 21:52:43,067 - Epoch: [173][ 920/ 1236] Overall Loss 0.126729 Objective Loss 0.126729 LR 0.000250 Time 0.021675 +2023-10-02 21:52:43,274 - Epoch: [173][ 930/ 1236] Overall Loss 0.126559 Objective Loss 0.126559 LR 0.000250 Time 0.021665 +2023-10-02 21:52:43,481 - Epoch: [173][ 940/ 1236] Overall Loss 0.126658 Objective Loss 0.126658 LR 0.000250 Time 0.021653 +2023-10-02 21:52:43,690 - Epoch: [173][ 950/ 1236] Overall Loss 0.126565 Objective Loss 0.126565 LR 0.000250 Time 0.021645 +2023-10-02 21:52:43,896 - Epoch: [173][ 960/ 1236] Overall Loss 0.126765 Objective Loss 0.126765 LR 0.000250 Time 0.021633 +2023-10-02 21:52:44,103 - Epoch: [173][ 970/ 1236] Overall Loss 0.126619 Objective Loss 0.126619 LR 0.000250 Time 0.021624 +2023-10-02 21:52:44,310 - Epoch: [173][ 980/ 1236] Overall Loss 0.126716 Objective Loss 0.126716 LR 0.000250 Time 0.021612 +2023-10-02 21:52:44,517 - Epoch: [173][ 990/ 1236] Overall Loss 0.126797 Objective Loss 0.126797 LR 0.000250 Time 0.021603 +2023-10-02 21:52:44,724 - Epoch: [173][ 1000/ 1236] Overall Loss 0.126810 Objective Loss 0.126810 LR 0.000250 Time 0.021592 +2023-10-02 21:52:44,931 - Epoch: [173][ 1010/ 1236] Overall Loss 0.126860 Objective Loss 0.126860 LR 0.000250 Time 0.021583 +2023-10-02 21:52:45,138 - Epoch: [173][ 1020/ 1236] Overall Loss 0.126840 Objective Loss 0.126840 LR 0.000250 Time 0.021573 +2023-10-02 21:52:45,345 - Epoch: [173][ 1030/ 1236] Overall Loss 0.126996 Objective Loss 0.126996 LR 0.000250 Time 0.021564 +2023-10-02 21:52:45,552 - Epoch: [173][ 1040/ 1236] Overall Loss 0.126875 Objective Loss 0.126875 LR 0.000250 Time 0.021554 +2023-10-02 21:52:45,759 - Epoch: [173][ 1050/ 1236] Overall Loss 0.127036 Objective Loss 0.127036 LR 0.000250 Time 0.021546 +2023-10-02 21:52:45,966 - Epoch: [173][ 1060/ 1236] Overall Loss 0.127169 Objective Loss 0.127169 LR 0.000250 Time 0.021536 +2023-10-02 21:52:46,174 - Epoch: [173][ 1070/ 1236] Overall Loss 0.127118 Objective Loss 0.127118 LR 0.000250 Time 0.021529 +2023-10-02 21:52:46,381 - Epoch: [173][ 1080/ 1236] Overall Loss 0.127070 Objective Loss 0.127070 LR 0.000250 Time 0.021520 +2023-10-02 21:52:46,588 - Epoch: [173][ 1090/ 1236] Overall Loss 0.127033 Objective Loss 0.127033 LR 0.000250 Time 0.021513 +2023-10-02 21:52:46,795 - Epoch: [173][ 1100/ 1236] Overall Loss 0.126862 Objective Loss 0.126862 LR 0.000250 Time 0.021504 +2023-10-02 21:52:47,003 - Epoch: [173][ 1110/ 1236] Overall Loss 0.126860 Objective Loss 0.126860 LR 0.000250 Time 0.021497 +2023-10-02 21:52:47,210 - Epoch: [173][ 1120/ 1236] Overall Loss 0.126941 Objective Loss 0.126941 LR 0.000250 Time 0.021488 +2023-10-02 21:52:47,417 - Epoch: [173][ 1130/ 1236] Overall Loss 0.126991 Objective Loss 0.126991 LR 0.000250 Time 0.021481 +2023-10-02 21:52:47,624 - Epoch: [173][ 1140/ 1236] Overall Loss 0.126898 Objective Loss 0.126898 LR 0.000250 Time 0.021473 +2023-10-02 21:52:47,832 - Epoch: [173][ 1150/ 1236] Overall Loss 0.126878 Objective Loss 0.126878 LR 0.000250 Time 0.021467 +2023-10-02 21:52:48,039 - Epoch: [173][ 1160/ 1236] Overall Loss 0.126841 Objective Loss 0.126841 LR 0.000250 Time 0.021459 +2023-10-02 21:52:48,247 - Epoch: [173][ 1170/ 1236] Overall Loss 0.126813 Objective Loss 0.126813 LR 0.000250 Time 0.021453 +2023-10-02 21:52:48,454 - Epoch: [173][ 1180/ 1236] Overall Loss 0.126821 Objective Loss 0.126821 LR 0.000250 Time 0.021446 +2023-10-02 21:52:48,662 - Epoch: [173][ 1190/ 1236] Overall Loss 0.126801 Objective Loss 0.126801 LR 0.000250 Time 0.021440 +2023-10-02 21:52:48,869 - Epoch: [173][ 1200/ 1236] Overall Loss 0.126857 Objective Loss 0.126857 LR 0.000250 Time 0.021432 +2023-10-02 21:52:49,077 - Epoch: [173][ 1210/ 1236] Overall Loss 0.126922 Objective Loss 0.126922 LR 0.000250 Time 0.021427 +2023-10-02 21:52:49,284 - Epoch: [173][ 1220/ 1236] Overall Loss 0.126812 Objective Loss 0.126812 LR 0.000250 Time 0.021420 +2023-10-02 21:52:49,542 - Epoch: [173][ 1230/ 1236] Overall Loss 0.126802 Objective Loss 0.126802 LR 0.000250 Time 0.021455 +2023-10-02 21:52:49,663 - Epoch: [173][ 1236/ 1236] Overall Loss 0.126789 Objective Loss 0.126789 Top1 92.057026 Top5 99.185336 LR 0.000250 Time 0.021449 +2023-10-02 21:52:49,798 - --- validate (epoch=173)----------- +2023-10-02 21:52:49,798 - 29943 samples (256 per mini-batch) +2023-10-02 21:52:50,288 - Epoch: [173][ 10/ 117] Loss 0.279076 Top1 88.359375 Top5 98.789062 +2023-10-02 21:52:50,439 - Epoch: [173][ 20/ 117] Loss 0.302953 Top1 87.929688 Top5 98.515625 +2023-10-02 21:52:50,590 - Epoch: [173][ 30/ 117] Loss 0.289328 Top1 87.539062 Top5 98.606771 +2023-10-02 21:52:50,741 - Epoch: [173][ 40/ 117] Loss 0.287893 Top1 87.675781 Top5 98.691406 +2023-10-02 21:52:50,893 - Epoch: [173][ 50/ 117] Loss 0.293459 Top1 87.437500 Top5 98.687500 +2023-10-02 21:52:51,053 - Epoch: [173][ 60/ 117] Loss 0.296884 Top1 87.441406 Top5 98.743490 +2023-10-02 21:52:51,210 - Epoch: [173][ 70/ 117] Loss 0.298829 Top1 87.349330 Top5 98.694196 +2023-10-02 21:52:51,369 - Epoch: [173][ 80/ 117] Loss 0.298154 Top1 87.373047 Top5 98.691406 +2023-10-02 21:52:51,525 - Epoch: [173][ 90/ 117] Loss 0.297618 Top1 87.322049 Top5 98.736979 +2023-10-02 21:52:51,684 - Epoch: [173][ 100/ 117] Loss 0.302314 Top1 87.148438 Top5 98.710938 +2023-10-02 21:52:51,849 - Epoch: [173][ 110/ 117] Loss 0.305926 Top1 87.088068 Top5 98.700284 +2023-10-02 21:52:51,937 - Epoch: [173][ 117/ 117] Loss 0.305803 Top1 87.202351 Top5 98.700865 +2023-10-02 21:52:52,084 - ==> Top1: 87.202 Top5: 98.701 Loss: 0.306 + +2023-10-02 21:52:52,084 - ==> Confusion: +[[ 951 1 3 0 4 3 0 0 11 45 2 1 0 1 5 1 1 0 2 0 19] + [ 0 1068 0 1 3 12 2 19 0 2 0 1 0 0 2 2 1 0 8 4 6] + [ 2 1 981 7 0 0 14 6 0 1 1 1 7 2 1 4 3 1 12 3 9] + [ 0 5 12 982 1 2 2 2 1 0 3 0 6 2 29 0 1 5 13 1 22] + [ 28 7 0 1 972 4 0 0 0 8 1 0 0 6 8 3 7 1 0 2 2] + [ 3 43 0 2 3 991 2 23 2 4 0 8 1 8 5 0 4 1 2 3 11] + [ 0 4 26 0 0 1 1130 7 0 0 4 1 0 1 0 2 0 0 1 11 3] + [ 2 15 8 1 3 22 4 1090 2 2 3 4 4 6 1 0 0 2 35 6 8] + [ 17 3 0 2 0 5 0 3 988 34 8 2 1 9 8 0 3 1 2 1 2] + [ 108 2 0 0 7 3 0 0 27 941 0 1 0 15 6 0 1 0 0 0 8] + [ 1 3 9 11 0 2 4 1 13 2 963 1 1 13 3 0 2 3 6 3 12] + [ 0 0 0 0 0 14 0 3 0 0 0 962 21 6 0 1 1 16 0 5 6] + [ 0 1 1 1 0 1 1 1 1 1 3 32 981 0 3 5 1 11 2 7 15] + [ 0 0 0 0 3 10 0 0 12 10 1 8 1 1051 6 0 0 1 0 0 16] + [ 13 0 5 15 2 1 0 0 18 2 3 0 2 4 1016 0 1 1 10 0 8] + [ 0 0 2 2 5 0 2 0 0 0 0 6 8 0 0 1066 18 11 2 7 5] + [ 0 17 1 0 5 4 0 0 2 0 1 6 0 3 4 5 1088 0 3 8 14] + [ 0 1 0 2 0 0 2 0 0 0 0 4 24 2 4 4 0 990 0 2 3] + [ 1 5 2 18 0 1 0 21 6 2 0 0 0 0 10 0 0 0 992 0 10] + [ 0 2 2 1 1 1 8 6 0 1 0 12 4 2 0 1 8 0 0 1097 6] + [ 112 148 108 72 59 112 29 86 83 50 142 71 283 230 108 32 62 54 118 135 5811]] + +2023-10-02 21:52:52,086 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:52:52,086 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:52:52,092 - + +2023-10-02 21:52:52,092 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:52:53,112 - Epoch: [174][ 10/ 1236] Overall Loss 0.119510 Objective Loss 0.119510 LR 0.000250 Time 0.101988 +2023-10-02 21:52:53,323 - Epoch: [174][ 20/ 1236] Overall Loss 0.117707 Objective Loss 0.117707 LR 0.000250 Time 0.061509 +2023-10-02 21:52:53,532 - Epoch: [174][ 30/ 1236] Overall Loss 0.118167 Objective Loss 0.118167 LR 0.000250 Time 0.047944 +2023-10-02 21:52:53,742 - Epoch: [174][ 40/ 1236] Overall Loss 0.116033 Objective Loss 0.116033 LR 0.000250 Time 0.041208 +2023-10-02 21:52:53,951 - Epoch: [174][ 50/ 1236] Overall Loss 0.117604 Objective Loss 0.117604 LR 0.000250 Time 0.037131 +2023-10-02 21:52:54,161 - Epoch: [174][ 60/ 1236] Overall Loss 0.121767 Objective Loss 0.121767 LR 0.000250 Time 0.034443 +2023-10-02 21:52:54,370 - Epoch: [174][ 70/ 1236] Overall Loss 0.124029 Objective Loss 0.124029 LR 0.000250 Time 0.032501 +2023-10-02 21:52:54,580 - Epoch: [174][ 80/ 1236] Overall Loss 0.126313 Objective Loss 0.126313 LR 0.000250 Time 0.031046 +2023-10-02 21:52:54,789 - Epoch: [174][ 90/ 1236] Overall Loss 0.126404 Objective Loss 0.126404 LR 0.000250 Time 0.029909 +2023-10-02 21:52:54,999 - Epoch: [174][ 100/ 1236] Overall Loss 0.124323 Objective Loss 0.124323 LR 0.000250 Time 0.029018 +2023-10-02 21:52:55,208 - Epoch: [174][ 110/ 1236] Overall Loss 0.126123 Objective Loss 0.126123 LR 0.000250 Time 0.028276 +2023-10-02 21:52:55,419 - Epoch: [174][ 120/ 1236] Overall Loss 0.127226 Objective Loss 0.127226 LR 0.000250 Time 0.027671 +2023-10-02 21:52:55,628 - Epoch: [174][ 130/ 1236] Overall Loss 0.126917 Objective Loss 0.126917 LR 0.000250 Time 0.027148 +2023-10-02 21:52:55,838 - Epoch: [174][ 140/ 1236] Overall Loss 0.126146 Objective Loss 0.126146 LR 0.000250 Time 0.026698 +2023-10-02 21:52:56,046 - Epoch: [174][ 150/ 1236] Overall Loss 0.126332 Objective Loss 0.126332 LR 0.000250 Time 0.026299 +2023-10-02 21:52:56,255 - Epoch: [174][ 160/ 1236] Overall Loss 0.126468 Objective Loss 0.126468 LR 0.000250 Time 0.025963 +2023-10-02 21:52:56,466 - Epoch: [174][ 170/ 1236] Overall Loss 0.125694 Objective Loss 0.125694 LR 0.000250 Time 0.025674 +2023-10-02 21:52:56,683 - Epoch: [174][ 180/ 1236] Overall Loss 0.124481 Objective Loss 0.124481 LR 0.000250 Time 0.025452 +2023-10-02 21:52:56,889 - Epoch: [174][ 190/ 1236] Overall Loss 0.123799 Objective Loss 0.123799 LR 0.000250 Time 0.025193 +2023-10-02 21:52:57,095 - Epoch: [174][ 200/ 1236] Overall Loss 0.124329 Objective Loss 0.124329 LR 0.000250 Time 0.024962 +2023-10-02 21:52:57,300 - Epoch: [174][ 210/ 1236] Overall Loss 0.124048 Objective Loss 0.124048 LR 0.000250 Time 0.024743 +2023-10-02 21:52:57,506 - Epoch: [174][ 220/ 1236] Overall Loss 0.124345 Objective Loss 0.124345 LR 0.000250 Time 0.024554 +2023-10-02 21:52:57,712 - Epoch: [174][ 230/ 1236] Overall Loss 0.123942 Objective Loss 0.123942 LR 0.000250 Time 0.024377 +2023-10-02 21:52:57,918 - Epoch: [174][ 240/ 1236] Overall Loss 0.124256 Objective Loss 0.124256 LR 0.000250 Time 0.024219 +2023-10-02 21:52:58,124 - Epoch: [174][ 250/ 1236] Overall Loss 0.124655 Objective Loss 0.124655 LR 0.000250 Time 0.024073 +2023-10-02 21:52:58,330 - Epoch: [174][ 260/ 1236] Overall Loss 0.124950 Objective Loss 0.124950 LR 0.000250 Time 0.023939 +2023-10-02 21:52:58,536 - Epoch: [174][ 270/ 1236] Overall Loss 0.124256 Objective Loss 0.124256 LR 0.000250 Time 0.023808 +2023-10-02 21:52:58,742 - Epoch: [174][ 280/ 1236] Overall Loss 0.124591 Objective Loss 0.124591 LR 0.000250 Time 0.023691 +2023-10-02 21:52:58,947 - Epoch: [174][ 290/ 1236] Overall Loss 0.125347 Objective Loss 0.125347 LR 0.000250 Time 0.023576 +2023-10-02 21:52:59,153 - Epoch: [174][ 300/ 1236] Overall Loss 0.125593 Objective Loss 0.125593 LR 0.000250 Time 0.023478 +2023-10-02 21:52:59,359 - Epoch: [174][ 310/ 1236] Overall Loss 0.125302 Objective Loss 0.125302 LR 0.000250 Time 0.023383 +2023-10-02 21:52:59,565 - Epoch: [174][ 320/ 1236] Overall Loss 0.125470 Objective Loss 0.125470 LR 0.000250 Time 0.023295 +2023-10-02 21:52:59,771 - Epoch: [174][ 330/ 1236] Overall Loss 0.125518 Objective Loss 0.125518 LR 0.000250 Time 0.023208 +2023-10-02 21:52:59,977 - Epoch: [174][ 340/ 1236] Overall Loss 0.125672 Objective Loss 0.125672 LR 0.000250 Time 0.023131 +2023-10-02 21:53:00,183 - Epoch: [174][ 350/ 1236] Overall Loss 0.125617 Objective Loss 0.125617 LR 0.000250 Time 0.023057 +2023-10-02 21:53:00,392 - Epoch: [174][ 360/ 1236] Overall Loss 0.125599 Objective Loss 0.125599 LR 0.000250 Time 0.022996 +2023-10-02 21:53:00,601 - Epoch: [174][ 370/ 1236] Overall Loss 0.125675 Objective Loss 0.125675 LR 0.000250 Time 0.022937 +2023-10-02 21:53:00,813 - Epoch: [174][ 380/ 1236] Overall Loss 0.125858 Objective Loss 0.125858 LR 0.000250 Time 0.022890 +2023-10-02 21:53:01,021 - Epoch: [174][ 390/ 1236] Overall Loss 0.125675 Objective Loss 0.125675 LR 0.000250 Time 0.022835 +2023-10-02 21:53:01,232 - Epoch: [174][ 400/ 1236] Overall Loss 0.125686 Objective Loss 0.125686 LR 0.000250 Time 0.022793 +2023-10-02 21:53:01,442 - Epoch: [174][ 410/ 1236] Overall Loss 0.126015 Objective Loss 0.126015 LR 0.000250 Time 0.022746 +2023-10-02 21:53:01,654 - Epoch: [174][ 420/ 1236] Overall Loss 0.126103 Objective Loss 0.126103 LR 0.000250 Time 0.022708 +2023-10-02 21:53:01,863 - Epoch: [174][ 430/ 1236] Overall Loss 0.126047 Objective Loss 0.126047 LR 0.000250 Time 0.022665 +2023-10-02 21:53:02,074 - Epoch: [174][ 440/ 1236] Overall Loss 0.126280 Objective Loss 0.126280 LR 0.000250 Time 0.022628 +2023-10-02 21:53:02,293 - Epoch: [174][ 450/ 1236] Overall Loss 0.126741 Objective Loss 0.126741 LR 0.000250 Time 0.022612 +2023-10-02 21:53:02,515 - Epoch: [174][ 460/ 1236] Overall Loss 0.126776 Objective Loss 0.126776 LR 0.000250 Time 0.022601 +2023-10-02 21:53:02,734 - Epoch: [174][ 470/ 1236] Overall Loss 0.126520 Objective Loss 0.126520 LR 0.000250 Time 0.022586 +2023-10-02 21:53:02,956 - Epoch: [174][ 480/ 1236] Overall Loss 0.126625 Objective Loss 0.126625 LR 0.000250 Time 0.022577 +2023-10-02 21:53:03,174 - Epoch: [174][ 490/ 1236] Overall Loss 0.126606 Objective Loss 0.126606 LR 0.000250 Time 0.022561 +2023-10-02 21:53:03,397 - Epoch: [174][ 500/ 1236] Overall Loss 0.127193 Objective Loss 0.127193 LR 0.000250 Time 0.022552 +2023-10-02 21:53:03,615 - Epoch: [174][ 510/ 1236] Overall Loss 0.127182 Objective Loss 0.127182 LR 0.000250 Time 0.022538 +2023-10-02 21:53:03,837 - Epoch: [174][ 520/ 1236] Overall Loss 0.126970 Objective Loss 0.126970 LR 0.000250 Time 0.022530 +2023-10-02 21:53:04,055 - Epoch: [174][ 530/ 1236] Overall Loss 0.127246 Objective Loss 0.127246 LR 0.000250 Time 0.022516 +2023-10-02 21:53:04,278 - Epoch: [174][ 540/ 1236] Overall Loss 0.127314 Objective Loss 0.127314 LR 0.000250 Time 0.022509 +2023-10-02 21:53:04,496 - Epoch: [174][ 550/ 1236] Overall Loss 0.127173 Objective Loss 0.127173 LR 0.000250 Time 0.022496 +2023-10-02 21:53:04,719 - Epoch: [174][ 560/ 1236] Overall Loss 0.127087 Objective Loss 0.127087 LR 0.000250 Time 0.022490 +2023-10-02 21:53:04,937 - Epoch: [174][ 570/ 1236] Overall Loss 0.126977 Objective Loss 0.126977 LR 0.000250 Time 0.022476 +2023-10-02 21:53:05,160 - Epoch: [174][ 580/ 1236] Overall Loss 0.127117 Objective Loss 0.127117 LR 0.000250 Time 0.022472 +2023-10-02 21:53:05,378 - Epoch: [174][ 590/ 1236] Overall Loss 0.127593 Objective Loss 0.127593 LR 0.000250 Time 0.022461 +2023-10-02 21:53:05,588 - Epoch: [174][ 600/ 1236] Overall Loss 0.127398 Objective Loss 0.127398 LR 0.000250 Time 0.022435 +2023-10-02 21:53:05,797 - Epoch: [174][ 610/ 1236] Overall Loss 0.127542 Objective Loss 0.127542 LR 0.000250 Time 0.022410 +2023-10-02 21:53:06,006 - Epoch: [174][ 620/ 1236] Overall Loss 0.127814 Objective Loss 0.127814 LR 0.000250 Time 0.022385 +2023-10-02 21:53:06,215 - Epoch: [174][ 630/ 1236] Overall Loss 0.127551 Objective Loss 0.127551 LR 0.000250 Time 0.022361 +2023-10-02 21:53:06,424 - Epoch: [174][ 640/ 1236] Overall Loss 0.127531 Objective Loss 0.127531 LR 0.000250 Time 0.022335 +2023-10-02 21:53:06,633 - Epoch: [174][ 650/ 1236] Overall Loss 0.127890 Objective Loss 0.127890 LR 0.000250 Time 0.022312 +2023-10-02 21:53:06,842 - Epoch: [174][ 660/ 1236] Overall Loss 0.127909 Objective Loss 0.127909 LR 0.000250 Time 0.022290 +2023-10-02 21:53:07,051 - Epoch: [174][ 670/ 1236] Overall Loss 0.127978 Objective Loss 0.127978 LR 0.000250 Time 0.022268 +2023-10-02 21:53:07,260 - Epoch: [174][ 680/ 1236] Overall Loss 0.127823 Objective Loss 0.127823 LR 0.000250 Time 0.022247 +2023-10-02 21:53:07,469 - Epoch: [174][ 690/ 1236] Overall Loss 0.127718 Objective Loss 0.127718 LR 0.000250 Time 0.022227 +2023-10-02 21:53:07,678 - Epoch: [174][ 700/ 1236] Overall Loss 0.127841 Objective Loss 0.127841 LR 0.000250 Time 0.022207 +2023-10-02 21:53:07,887 - Epoch: [174][ 710/ 1236] Overall Loss 0.127979 Objective Loss 0.127979 LR 0.000250 Time 0.022188 +2023-10-02 21:53:08,096 - Epoch: [174][ 720/ 1236] Overall Loss 0.127676 Objective Loss 0.127676 LR 0.000250 Time 0.022169 +2023-10-02 21:53:08,305 - Epoch: [174][ 730/ 1236] Overall Loss 0.127808 Objective Loss 0.127808 LR 0.000250 Time 0.022151 +2023-10-02 21:53:08,513 - Epoch: [174][ 740/ 1236] Overall Loss 0.127893 Objective Loss 0.127893 LR 0.000250 Time 0.022132 +2023-10-02 21:53:08,722 - Epoch: [174][ 750/ 1236] Overall Loss 0.127838 Objective Loss 0.127838 LR 0.000250 Time 0.022114 +2023-10-02 21:53:08,931 - Epoch: [174][ 760/ 1236] Overall Loss 0.128075 Objective Loss 0.128075 LR 0.000250 Time 0.022098 +2023-10-02 21:53:09,140 - Epoch: [174][ 770/ 1236] Overall Loss 0.128148 Objective Loss 0.128148 LR 0.000250 Time 0.022081 +2023-10-02 21:53:09,349 - Epoch: [174][ 780/ 1236] Overall Loss 0.128153 Objective Loss 0.128153 LR 0.000250 Time 0.022066 +2023-10-02 21:53:09,558 - Epoch: [174][ 790/ 1236] Overall Loss 0.128166 Objective Loss 0.128166 LR 0.000250 Time 0.022050 +2023-10-02 21:53:09,767 - Epoch: [174][ 800/ 1236] Overall Loss 0.128344 Objective Loss 0.128344 LR 0.000250 Time 0.022035 +2023-10-02 21:53:09,976 - Epoch: [174][ 810/ 1236] Overall Loss 0.128047 Objective Loss 0.128047 LR 0.000250 Time 0.022021 +2023-10-02 21:53:10,185 - Epoch: [174][ 820/ 1236] Overall Loss 0.128013 Objective Loss 0.128013 LR 0.000250 Time 0.022007 +2023-10-02 21:53:10,395 - Epoch: [174][ 830/ 1236] Overall Loss 0.128155 Objective Loss 0.128155 LR 0.000250 Time 0.021994 +2023-10-02 21:53:10,604 - Epoch: [174][ 840/ 1236] Overall Loss 0.128139 Objective Loss 0.128139 LR 0.000250 Time 0.021978 +2023-10-02 21:53:10,814 - Epoch: [174][ 850/ 1236] Overall Loss 0.128269 Objective Loss 0.128269 LR 0.000250 Time 0.021966 +2023-10-02 21:53:11,022 - Epoch: [174][ 860/ 1236] Overall Loss 0.128447 Objective Loss 0.128447 LR 0.000250 Time 0.021952 +2023-10-02 21:53:11,231 - Epoch: [174][ 870/ 1236] Overall Loss 0.128625 Objective Loss 0.128625 LR 0.000250 Time 0.021940 +2023-10-02 21:53:11,439 - Epoch: [174][ 880/ 1236] Overall Loss 0.128690 Objective Loss 0.128690 LR 0.000250 Time 0.021927 +2023-10-02 21:53:11,648 - Epoch: [174][ 890/ 1236] Overall Loss 0.128752 Objective Loss 0.128752 LR 0.000250 Time 0.021915 +2023-10-02 21:53:11,857 - Epoch: [174][ 900/ 1236] Overall Loss 0.128711 Objective Loss 0.128711 LR 0.000250 Time 0.021902 +2023-10-02 21:53:12,066 - Epoch: [174][ 910/ 1236] Overall Loss 0.128703 Objective Loss 0.128703 LR 0.000250 Time 0.021891 +2023-10-02 21:53:12,275 - Epoch: [174][ 920/ 1236] Overall Loss 0.128794 Objective Loss 0.128794 LR 0.000250 Time 0.021878 +2023-10-02 21:53:12,484 - Epoch: [174][ 930/ 1236] Overall Loss 0.128643 Objective Loss 0.128643 LR 0.000250 Time 0.021867 +2023-10-02 21:53:12,693 - Epoch: [174][ 940/ 1236] Overall Loss 0.128757 Objective Loss 0.128757 LR 0.000250 Time 0.021856 +2023-10-02 21:53:12,902 - Epoch: [174][ 950/ 1236] Overall Loss 0.128678 Objective Loss 0.128678 LR 0.000250 Time 0.021845 +2023-10-02 21:53:13,111 - Epoch: [174][ 960/ 1236] Overall Loss 0.128811 Objective Loss 0.128811 LR 0.000250 Time 0.021835 +2023-10-02 21:53:13,319 - Epoch: [174][ 970/ 1236] Overall Loss 0.129063 Objective Loss 0.129063 LR 0.000250 Time 0.021824 +2023-10-02 21:53:13,528 - Epoch: [174][ 980/ 1236] Overall Loss 0.129117 Objective Loss 0.129117 LR 0.000250 Time 0.021814 +2023-10-02 21:53:13,738 - Epoch: [174][ 990/ 1236] Overall Loss 0.129109 Objective Loss 0.129109 LR 0.000250 Time 0.021805 +2023-10-02 21:53:13,946 - Epoch: [174][ 1000/ 1236] Overall Loss 0.129141 Objective Loss 0.129141 LR 0.000250 Time 0.021795 +2023-10-02 21:53:14,155 - Epoch: [174][ 1010/ 1236] Overall Loss 0.129049 Objective Loss 0.129049 LR 0.000250 Time 0.021786 +2023-10-02 21:53:14,364 - Epoch: [174][ 1020/ 1236] Overall Loss 0.129040 Objective Loss 0.129040 LR 0.000250 Time 0.021777 +2023-10-02 21:53:14,573 - Epoch: [174][ 1030/ 1236] Overall Loss 0.129031 Objective Loss 0.129031 LR 0.000250 Time 0.021767 +2023-10-02 21:53:14,782 - Epoch: [174][ 1040/ 1236] Overall Loss 0.128997 Objective Loss 0.128997 LR 0.000250 Time 0.021759 +2023-10-02 21:53:14,991 - Epoch: [174][ 1050/ 1236] Overall Loss 0.129113 Objective Loss 0.129113 LR 0.000250 Time 0.021750 +2023-10-02 21:53:15,199 - Epoch: [174][ 1060/ 1236] Overall Loss 0.129059 Objective Loss 0.129059 LR 0.000250 Time 0.021741 +2023-10-02 21:53:15,408 - Epoch: [174][ 1070/ 1236] Overall Loss 0.129274 Objective Loss 0.129274 LR 0.000250 Time 0.021732 +2023-10-02 21:53:15,618 - Epoch: [174][ 1080/ 1236] Overall Loss 0.129135 Objective Loss 0.129135 LR 0.000250 Time 0.021725 +2023-10-02 21:53:15,826 - Epoch: [174][ 1090/ 1236] Overall Loss 0.129190 Objective Loss 0.129190 LR 0.000250 Time 0.021716 +2023-10-02 21:53:16,035 - Epoch: [174][ 1100/ 1236] Overall Loss 0.129105 Objective Loss 0.129105 LR 0.000250 Time 0.021708 +2023-10-02 21:53:16,244 - Epoch: [174][ 1110/ 1236] Overall Loss 0.129066 Objective Loss 0.129066 LR 0.000250 Time 0.021701 +2023-10-02 21:53:16,453 - Epoch: [174][ 1120/ 1236] Overall Loss 0.129125 Objective Loss 0.129125 LR 0.000250 Time 0.021693 +2023-10-02 21:53:16,662 - Epoch: [174][ 1130/ 1236] Overall Loss 0.129067 Objective Loss 0.129067 LR 0.000250 Time 0.021685 +2023-10-02 21:53:16,870 - Epoch: [174][ 1140/ 1236] Overall Loss 0.129077 Objective Loss 0.129077 LR 0.000250 Time 0.021678 +2023-10-02 21:53:17,079 - Epoch: [174][ 1150/ 1236] Overall Loss 0.129043 Objective Loss 0.129043 LR 0.000250 Time 0.021670 +2023-10-02 21:53:17,290 - Epoch: [174][ 1160/ 1236] Overall Loss 0.128981 Objective Loss 0.128981 LR 0.000250 Time 0.021664 +2023-10-02 21:53:17,500 - Epoch: [174][ 1170/ 1236] Overall Loss 0.128929 Objective Loss 0.128929 LR 0.000250 Time 0.021658 +2023-10-02 21:53:17,709 - Epoch: [174][ 1180/ 1236] Overall Loss 0.128965 Objective Loss 0.128965 LR 0.000250 Time 0.021651 +2023-10-02 21:53:17,920 - Epoch: [174][ 1190/ 1236] Overall Loss 0.128974 Objective Loss 0.128974 LR 0.000250 Time 0.021646 +2023-10-02 21:53:18,129 - Epoch: [174][ 1200/ 1236] Overall Loss 0.129014 Objective Loss 0.129014 LR 0.000250 Time 0.021639 +2023-10-02 21:53:18,340 - Epoch: [174][ 1210/ 1236] Overall Loss 0.129012 Objective Loss 0.129012 LR 0.000250 Time 0.021634 +2023-10-02 21:53:18,549 - Epoch: [174][ 1220/ 1236] Overall Loss 0.129267 Objective Loss 0.129267 LR 0.000250 Time 0.021628 +2023-10-02 21:53:18,812 - Epoch: [174][ 1230/ 1236] Overall Loss 0.129296 Objective Loss 0.129296 LR 0.000250 Time 0.021666 +2023-10-02 21:53:18,934 - Epoch: [174][ 1236/ 1236] Overall Loss 0.129282 Objective Loss 0.129282 Top1 90.835031 Top5 98.574338 LR 0.000250 Time 0.021660 +2023-10-02 21:53:19,047 - --- validate (epoch=174)----------- +2023-10-02 21:53:19,048 - 29943 samples (256 per mini-batch) +2023-10-02 21:53:19,530 - Epoch: [174][ 10/ 117] Loss 0.313815 Top1 87.187500 Top5 99.062500 +2023-10-02 21:53:19,695 - Epoch: [174][ 20/ 117] Loss 0.301174 Top1 86.992188 Top5 98.730469 +2023-10-02 21:53:19,853 - Epoch: [174][ 30/ 117] Loss 0.302355 Top1 87.265625 Top5 98.671875 +2023-10-02 21:53:20,018 - Epoch: [174][ 40/ 117] Loss 0.297473 Top1 87.392578 Top5 98.691406 +2023-10-02 21:53:20,175 - Epoch: [174][ 50/ 117] Loss 0.304194 Top1 87.445312 Top5 98.679688 +2023-10-02 21:53:20,339 - Epoch: [174][ 60/ 117] Loss 0.301545 Top1 87.467448 Top5 98.671875 +2023-10-02 21:53:20,497 - Epoch: [174][ 70/ 117] Loss 0.307021 Top1 87.393973 Top5 98.677455 +2023-10-02 21:53:20,659 - Epoch: [174][ 80/ 117] Loss 0.306679 Top1 87.368164 Top5 98.657227 +2023-10-02 21:53:20,815 - Epoch: [174][ 90/ 117] Loss 0.310968 Top1 87.374132 Top5 98.667535 +2023-10-02 21:53:20,977 - Epoch: [174][ 100/ 117] Loss 0.306169 Top1 87.421875 Top5 98.660156 +2023-10-02 21:53:21,140 - Epoch: [174][ 110/ 117] Loss 0.306810 Top1 87.357955 Top5 98.668324 +2023-10-02 21:53:21,230 - Epoch: [174][ 117/ 117] Loss 0.310148 Top1 87.282503 Top5 98.627392 +2023-10-02 21:53:21,363 - ==> Top1: 87.283 Top5: 98.627 Loss: 0.310 + +2023-10-02 21:53:21,364 - ==> Confusion: +[[ 943 1 1 0 4 3 0 0 7 53 2 0 0 1 4 2 2 2 1 0 24] + [ 0 1072 0 1 5 17 0 18 1 1 0 0 0 0 0 3 1 0 5 3 4] + [ 2 1 990 6 0 0 13 9 0 1 0 1 5 2 1 2 2 2 9 2 8] + [ 1 4 18 970 0 2 2 4 2 0 5 0 6 4 28 3 1 5 14 0 20] + [ 27 5 0 1 971 6 0 0 0 8 0 0 0 4 9 2 12 0 1 2 2] + [ 3 45 0 0 5 995 2 16 2 5 0 4 1 10 3 0 3 0 3 1 18] + [ 0 6 27 2 0 2 1129 6 0 0 2 1 0 0 0 6 0 0 2 6 2] + [ 2 8 6 2 4 22 5 1103 2 1 1 2 2 4 1 0 0 4 35 7 7] + [ 13 3 0 1 0 7 0 4 983 36 7 2 2 10 9 0 3 2 4 1 2] + [ 91 0 0 0 5 3 0 0 29 952 1 2 0 16 5 1 0 0 0 2 12] + [ 2 1 8 9 1 1 3 2 13 2 965 1 1 14 4 0 2 4 4 3 13] + [ 0 0 1 0 0 14 0 2 0 0 0 968 22 7 0 1 0 11 0 5 4] + [ 0 1 1 2 0 2 0 1 0 0 2 35 979 2 2 7 2 13 3 5 11] + [ 2 0 0 0 3 11 0 0 6 10 1 6 0 1058 5 0 0 1 0 0 16] + [ 12 0 3 12 7 0 0 0 24 2 2 0 4 6 1003 0 0 4 8 0 14] + [ 0 0 1 1 6 1 0 0 0 0 1 5 8 0 0 1072 19 7 0 9 4] + [ 0 18 0 0 4 5 1 0 1 1 0 4 0 3 3 10 1095 0 0 5 11] + [ 0 1 1 2 0 0 1 0 1 0 0 4 25 3 1 4 1 987 0 2 5] + [ 2 7 5 8 0 1 0 27 3 2 2 0 2 0 4 0 1 1 990 0 13] + [ 0 1 3 2 1 3 6 9 0 0 0 13 3 3 0 1 10 0 1 1090 6] + [ 103 126 109 57 51 121 23 97 73 61 120 88 326 229 99 55 75 51 96 125 5820]] + +2023-10-02 21:53:21,365 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:53:21,365 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:53:21,371 - + +2023-10-02 21:53:21,371 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:53:22,542 - Epoch: [175][ 10/ 1236] Overall Loss 0.116650 Objective Loss 0.116650 LR 0.000250 Time 0.117009 +2023-10-02 21:53:22,750 - Epoch: [175][ 20/ 1236] Overall Loss 0.129106 Objective Loss 0.129106 LR 0.000250 Time 0.068900 +2023-10-02 21:53:22,959 - Epoch: [175][ 30/ 1236] Overall Loss 0.128589 Objective Loss 0.128589 LR 0.000250 Time 0.052885 +2023-10-02 21:53:23,168 - Epoch: [175][ 40/ 1236] Overall Loss 0.129404 Objective Loss 0.129404 LR 0.000250 Time 0.044884 +2023-10-02 21:53:23,379 - Epoch: [175][ 50/ 1236] Overall Loss 0.129093 Objective Loss 0.129093 LR 0.000250 Time 0.040082 +2023-10-02 21:53:23,586 - Epoch: [175][ 60/ 1236] Overall Loss 0.126413 Objective Loss 0.126413 LR 0.000250 Time 0.036859 +2023-10-02 21:53:23,793 - Epoch: [175][ 70/ 1236] Overall Loss 0.127071 Objective Loss 0.127071 LR 0.000250 Time 0.034522 +2023-10-02 21:53:24,000 - Epoch: [175][ 80/ 1236] Overall Loss 0.127760 Objective Loss 0.127760 LR 0.000250 Time 0.032797 +2023-10-02 21:53:24,207 - Epoch: [175][ 90/ 1236] Overall Loss 0.125631 Objective Loss 0.125631 LR 0.000250 Time 0.031434 +2023-10-02 21:53:24,415 - Epoch: [175][ 100/ 1236] Overall Loss 0.125330 Objective Loss 0.125330 LR 0.000250 Time 0.030365 +2023-10-02 21:53:24,621 - Epoch: [175][ 110/ 1236] Overall Loss 0.124911 Objective Loss 0.124911 LR 0.000250 Time 0.029470 +2023-10-02 21:53:24,829 - Epoch: [175][ 120/ 1236] Overall Loss 0.127409 Objective Loss 0.127409 LR 0.000250 Time 0.028745 +2023-10-02 21:53:25,035 - Epoch: [175][ 130/ 1236] Overall Loss 0.127527 Objective Loss 0.127527 LR 0.000250 Time 0.028111 +2023-10-02 21:53:25,243 - Epoch: [175][ 140/ 1236] Overall Loss 0.126986 Objective Loss 0.126986 LR 0.000250 Time 0.027585 +2023-10-02 21:53:25,450 - Epoch: [175][ 150/ 1236] Overall Loss 0.127526 Objective Loss 0.127526 LR 0.000250 Time 0.027114 +2023-10-02 21:53:25,657 - Epoch: [175][ 160/ 1236] Overall Loss 0.126931 Objective Loss 0.126931 LR 0.000250 Time 0.026715 +2023-10-02 21:53:25,865 - Epoch: [175][ 170/ 1236] Overall Loss 0.127468 Objective Loss 0.127468 LR 0.000250 Time 0.026358 +2023-10-02 21:53:26,074 - Epoch: [175][ 180/ 1236] Overall Loss 0.127471 Objective Loss 0.127471 LR 0.000250 Time 0.026051 +2023-10-02 21:53:26,284 - Epoch: [175][ 190/ 1236] Overall Loss 0.127304 Objective Loss 0.127304 LR 0.000250 Time 0.025784 +2023-10-02 21:53:26,491 - Epoch: [175][ 200/ 1236] Overall Loss 0.127778 Objective Loss 0.127778 LR 0.000250 Time 0.025529 +2023-10-02 21:53:26,701 - Epoch: [175][ 210/ 1236] Overall Loss 0.127699 Objective Loss 0.127699 LR 0.000250 Time 0.025313 +2023-10-02 21:53:26,910 - Epoch: [175][ 220/ 1236] Overall Loss 0.127349 Objective Loss 0.127349 LR 0.000250 Time 0.025109 +2023-10-02 21:53:27,120 - Epoch: [175][ 230/ 1236] Overall Loss 0.127281 Objective Loss 0.127281 LR 0.000250 Time 0.024930 +2023-10-02 21:53:27,328 - Epoch: [175][ 240/ 1236] Overall Loss 0.126801 Objective Loss 0.126801 LR 0.000250 Time 0.024757 +2023-10-02 21:53:27,539 - Epoch: [175][ 250/ 1236] Overall Loss 0.126457 Objective Loss 0.126457 LR 0.000250 Time 0.024608 +2023-10-02 21:53:27,747 - Epoch: [175][ 260/ 1236] Overall Loss 0.126227 Objective Loss 0.126227 LR 0.000250 Time 0.024461 +2023-10-02 21:53:27,957 - Epoch: [175][ 270/ 1236] Overall Loss 0.125685 Objective Loss 0.125685 LR 0.000250 Time 0.024332 +2023-10-02 21:53:28,165 - Epoch: [175][ 280/ 1236] Overall Loss 0.125343 Objective Loss 0.125343 LR 0.000250 Time 0.024205 +2023-10-02 21:53:28,375 - Epoch: [175][ 290/ 1236] Overall Loss 0.125374 Objective Loss 0.125374 LR 0.000250 Time 0.024094 +2023-10-02 21:53:28,584 - Epoch: [175][ 300/ 1236] Overall Loss 0.125871 Objective Loss 0.125871 LR 0.000250 Time 0.023984 +2023-10-02 21:53:28,794 - Epoch: [175][ 310/ 1236] Overall Loss 0.125407 Objective Loss 0.125407 LR 0.000250 Time 0.023888 +2023-10-02 21:53:29,002 - Epoch: [175][ 320/ 1236] Overall Loss 0.125120 Objective Loss 0.125120 LR 0.000250 Time 0.023792 +2023-10-02 21:53:29,213 - Epoch: [175][ 330/ 1236] Overall Loss 0.124544 Objective Loss 0.124544 LR 0.000250 Time 0.023708 +2023-10-02 21:53:29,421 - Epoch: [175][ 340/ 1236] Overall Loss 0.124789 Objective Loss 0.124789 LR 0.000250 Time 0.023622 +2023-10-02 21:53:29,632 - Epoch: [175][ 350/ 1236] Overall Loss 0.124488 Objective Loss 0.124488 LR 0.000250 Time 0.023548 +2023-10-02 21:53:29,841 - Epoch: [175][ 360/ 1236] Overall Loss 0.125006 Objective Loss 0.125006 LR 0.000250 Time 0.023474 +2023-10-02 21:53:30,053 - Epoch: [175][ 370/ 1236] Overall Loss 0.124969 Objective Loss 0.124969 LR 0.000250 Time 0.023413 +2023-10-02 21:53:30,265 - Epoch: [175][ 380/ 1236] Overall Loss 0.124710 Objective Loss 0.124710 LR 0.000250 Time 0.023352 +2023-10-02 21:53:30,478 - Epoch: [175][ 390/ 1236] Overall Loss 0.124701 Objective Loss 0.124701 LR 0.000250 Time 0.023298 +2023-10-02 21:53:30,689 - Epoch: [175][ 400/ 1236] Overall Loss 0.124914 Objective Loss 0.124914 LR 0.000250 Time 0.023243 +2023-10-02 21:53:30,902 - Epoch: [175][ 410/ 1236] Overall Loss 0.125062 Objective Loss 0.125062 LR 0.000250 Time 0.023194 +2023-10-02 21:53:31,113 - Epoch: [175][ 420/ 1236] Overall Loss 0.125229 Objective Loss 0.125229 LR 0.000250 Time 0.023144 +2023-10-02 21:53:31,326 - Epoch: [175][ 430/ 1236] Overall Loss 0.125292 Objective Loss 0.125292 LR 0.000250 Time 0.023100 +2023-10-02 21:53:31,538 - Epoch: [175][ 440/ 1236] Overall Loss 0.125115 Objective Loss 0.125115 LR 0.000250 Time 0.023055 +2023-10-02 21:53:31,751 - Epoch: [175][ 450/ 1236] Overall Loss 0.124911 Objective Loss 0.124911 LR 0.000250 Time 0.023016 +2023-10-02 21:53:31,962 - Epoch: [175][ 460/ 1236] Overall Loss 0.124935 Objective Loss 0.124935 LR 0.000250 Time 0.022974 +2023-10-02 21:53:32,175 - Epoch: [175][ 470/ 1236] Overall Loss 0.125267 Objective Loss 0.125267 LR 0.000250 Time 0.022937 +2023-10-02 21:53:32,387 - Epoch: [175][ 480/ 1236] Overall Loss 0.125650 Objective Loss 0.125650 LR 0.000250 Time 0.022900 +2023-10-02 21:53:32,599 - Epoch: [175][ 490/ 1236] Overall Loss 0.125766 Objective Loss 0.125766 LR 0.000250 Time 0.022863 +2023-10-02 21:53:32,810 - Epoch: [175][ 500/ 1236] Overall Loss 0.126211 Objective Loss 0.126211 LR 0.000250 Time 0.022829 +2023-10-02 21:53:33,023 - Epoch: [175][ 510/ 1236] Overall Loss 0.126127 Objective Loss 0.126127 LR 0.000250 Time 0.022797 +2023-10-02 21:53:33,235 - Epoch: [175][ 520/ 1236] Overall Loss 0.126266 Objective Loss 0.126266 LR 0.000250 Time 0.022766 +2023-10-02 21:53:33,447 - Epoch: [175][ 530/ 1236] Overall Loss 0.126512 Objective Loss 0.126512 LR 0.000250 Time 0.022735 +2023-10-02 21:53:33,659 - Epoch: [175][ 540/ 1236] Overall Loss 0.126641 Objective Loss 0.126641 LR 0.000250 Time 0.022706 +2023-10-02 21:53:33,871 - Epoch: [175][ 550/ 1236] Overall Loss 0.126702 Objective Loss 0.126702 LR 0.000250 Time 0.022679 +2023-10-02 21:53:34,083 - Epoch: [175][ 560/ 1236] Overall Loss 0.126522 Objective Loss 0.126522 LR 0.000250 Time 0.022651 +2023-10-02 21:53:34,296 - Epoch: [175][ 570/ 1236] Overall Loss 0.126712 Objective Loss 0.126712 LR 0.000250 Time 0.022626 +2023-10-02 21:53:34,508 - Epoch: [175][ 580/ 1236] Overall Loss 0.126818 Objective Loss 0.126818 LR 0.000250 Time 0.022601 +2023-10-02 21:53:34,720 - Epoch: [175][ 590/ 1236] Overall Loss 0.126721 Objective Loss 0.126721 LR 0.000250 Time 0.022577 +2023-10-02 21:53:34,929 - Epoch: [175][ 600/ 1236] Overall Loss 0.126655 Objective Loss 0.126655 LR 0.000250 Time 0.022549 +2023-10-02 21:53:35,138 - Epoch: [175][ 610/ 1236] Overall Loss 0.126550 Objective Loss 0.126550 LR 0.000250 Time 0.022520 +2023-10-02 21:53:35,348 - Epoch: [175][ 620/ 1236] Overall Loss 0.126772 Objective Loss 0.126772 LR 0.000250 Time 0.022495 +2023-10-02 21:53:35,556 - Epoch: [175][ 630/ 1236] Overall Loss 0.126884 Objective Loss 0.126884 LR 0.000250 Time 0.022468 +2023-10-02 21:53:35,766 - Epoch: [175][ 640/ 1236] Overall Loss 0.127083 Objective Loss 0.127083 LR 0.000250 Time 0.022445 +2023-10-02 21:53:35,975 - Epoch: [175][ 650/ 1236] Overall Loss 0.127200 Objective Loss 0.127200 LR 0.000250 Time 0.022420 +2023-10-02 21:53:36,184 - Epoch: [175][ 660/ 1236] Overall Loss 0.127290 Objective Loss 0.127290 LR 0.000250 Time 0.022397 +2023-10-02 21:53:36,393 - Epoch: [175][ 670/ 1236] Overall Loss 0.127554 Objective Loss 0.127554 LR 0.000250 Time 0.022374 +2023-10-02 21:53:36,603 - Epoch: [175][ 680/ 1236] Overall Loss 0.127432 Objective Loss 0.127432 LR 0.000250 Time 0.022353 +2023-10-02 21:53:36,811 - Epoch: [175][ 690/ 1236] Overall Loss 0.127393 Objective Loss 0.127393 LR 0.000250 Time 0.022330 +2023-10-02 21:53:37,021 - Epoch: [175][ 700/ 1236] Overall Loss 0.127417 Objective Loss 0.127417 LR 0.000250 Time 0.022311 +2023-10-02 21:53:37,229 - Epoch: [175][ 710/ 1236] Overall Loss 0.127402 Objective Loss 0.127402 LR 0.000250 Time 0.022289 +2023-10-02 21:53:37,439 - Epoch: [175][ 720/ 1236] Overall Loss 0.127141 Objective Loss 0.127141 LR 0.000250 Time 0.022271 +2023-10-02 21:53:37,647 - Epoch: [175][ 730/ 1236] Overall Loss 0.127082 Objective Loss 0.127082 LR 0.000250 Time 0.022250 +2023-10-02 21:53:37,857 - Epoch: [175][ 740/ 1236] Overall Loss 0.127218 Objective Loss 0.127218 LR 0.000250 Time 0.022233 +2023-10-02 21:53:38,065 - Epoch: [175][ 750/ 1236] Overall Loss 0.127511 Objective Loss 0.127511 LR 0.000250 Time 0.022214 +2023-10-02 21:53:38,275 - Epoch: [175][ 760/ 1236] Overall Loss 0.127550 Objective Loss 0.127550 LR 0.000250 Time 0.022197 +2023-10-02 21:53:38,484 - Epoch: [175][ 770/ 1236] Overall Loss 0.127398 Objective Loss 0.127398 LR 0.000250 Time 0.022179 +2023-10-02 21:53:38,694 - Epoch: [175][ 780/ 1236] Overall Loss 0.127448 Objective Loss 0.127448 LR 0.000250 Time 0.022164 +2023-10-02 21:53:38,902 - Epoch: [175][ 790/ 1236] Overall Loss 0.127406 Objective Loss 0.127406 LR 0.000250 Time 0.022147 +2023-10-02 21:53:39,112 - Epoch: [175][ 800/ 1236] Overall Loss 0.127668 Objective Loss 0.127668 LR 0.000250 Time 0.022132 +2023-10-02 21:53:39,321 - Epoch: [175][ 810/ 1236] Overall Loss 0.127433 Objective Loss 0.127433 LR 0.000250 Time 0.022116 +2023-10-02 21:53:39,531 - Epoch: [175][ 820/ 1236] Overall Loss 0.127298 Objective Loss 0.127298 LR 0.000250 Time 0.022102 +2023-10-02 21:53:39,739 - Epoch: [175][ 830/ 1236] Overall Loss 0.127090 Objective Loss 0.127090 LR 0.000250 Time 0.022086 +2023-10-02 21:53:39,949 - Epoch: [175][ 840/ 1236] Overall Loss 0.126927 Objective Loss 0.126927 LR 0.000250 Time 0.022073 +2023-10-02 21:53:40,157 - Epoch: [175][ 850/ 1236] Overall Loss 0.126997 Objective Loss 0.126997 LR 0.000250 Time 0.022058 +2023-10-02 21:53:40,367 - Epoch: [175][ 860/ 1236] Overall Loss 0.126819 Objective Loss 0.126819 LR 0.000250 Time 0.022045 +2023-10-02 21:53:40,575 - Epoch: [175][ 870/ 1236] Overall Loss 0.126718 Objective Loss 0.126718 LR 0.000250 Time 0.022030 +2023-10-02 21:53:40,785 - Epoch: [175][ 880/ 1236] Overall Loss 0.126861 Objective Loss 0.126861 LR 0.000250 Time 0.022018 +2023-10-02 21:53:40,994 - Epoch: [175][ 890/ 1236] Overall Loss 0.126937 Objective Loss 0.126937 LR 0.000250 Time 0.022004 +2023-10-02 21:53:41,203 - Epoch: [175][ 900/ 1236] Overall Loss 0.126814 Objective Loss 0.126814 LR 0.000250 Time 0.021993 +2023-10-02 21:53:41,412 - Epoch: [175][ 910/ 1236] Overall Loss 0.126888 Objective Loss 0.126888 LR 0.000250 Time 0.021980 +2023-10-02 21:53:41,622 - Epoch: [175][ 920/ 1236] Overall Loss 0.126956 Objective Loss 0.126956 LR 0.000250 Time 0.021969 +2023-10-02 21:53:41,830 - Epoch: [175][ 930/ 1236] Overall Loss 0.126950 Objective Loss 0.126950 LR 0.000250 Time 0.021956 +2023-10-02 21:53:42,040 - Epoch: [175][ 940/ 1236] Overall Loss 0.127227 Objective Loss 0.127227 LR 0.000250 Time 0.021946 +2023-10-02 21:53:42,249 - Epoch: [175][ 950/ 1236] Overall Loss 0.127138 Objective Loss 0.127138 LR 0.000250 Time 0.021934 +2023-10-02 21:53:42,458 - Epoch: [175][ 960/ 1236] Overall Loss 0.127172 Objective Loss 0.127172 LR 0.000250 Time 0.021924 +2023-10-02 21:53:42,667 - Epoch: [175][ 970/ 1236] Overall Loss 0.127183 Objective Loss 0.127183 LR 0.000250 Time 0.021912 +2023-10-02 21:53:42,877 - Epoch: [175][ 980/ 1236] Overall Loss 0.127119 Objective Loss 0.127119 LR 0.000250 Time 0.021903 +2023-10-02 21:53:43,085 - Epoch: [175][ 990/ 1236] Overall Loss 0.127151 Objective Loss 0.127151 LR 0.000250 Time 0.021891 +2023-10-02 21:53:43,295 - Epoch: [175][ 1000/ 1236] Overall Loss 0.127200 Objective Loss 0.127200 LR 0.000250 Time 0.021882 +2023-10-02 21:53:43,503 - Epoch: [175][ 1010/ 1236] Overall Loss 0.127122 Objective Loss 0.127122 LR 0.000250 Time 0.021871 +2023-10-02 21:53:43,713 - Epoch: [175][ 1020/ 1236] Overall Loss 0.127073 Objective Loss 0.127073 LR 0.000250 Time 0.021862 +2023-10-02 21:53:43,922 - Epoch: [175][ 1030/ 1236] Overall Loss 0.127189 Objective Loss 0.127189 LR 0.000250 Time 0.021852 +2023-10-02 21:53:44,132 - Epoch: [175][ 1040/ 1236] Overall Loss 0.127100 Objective Loss 0.127100 LR 0.000250 Time 0.021843 +2023-10-02 21:53:44,340 - Epoch: [175][ 1050/ 1236] Overall Loss 0.127074 Objective Loss 0.127074 LR 0.000250 Time 0.021833 +2023-10-02 21:53:44,550 - Epoch: [175][ 1060/ 1236] Overall Loss 0.127010 Objective Loss 0.127010 LR 0.000250 Time 0.021825 +2023-10-02 21:53:44,758 - Epoch: [175][ 1070/ 1236] Overall Loss 0.126916 Objective Loss 0.126916 LR 0.000250 Time 0.021816 +2023-10-02 21:53:44,968 - Epoch: [175][ 1080/ 1236] Overall Loss 0.126882 Objective Loss 0.126882 LR 0.000250 Time 0.021808 +2023-10-02 21:53:45,177 - Epoch: [175][ 1090/ 1236] Overall Loss 0.126819 Objective Loss 0.126819 LR 0.000250 Time 0.021799 +2023-10-02 21:53:45,387 - Epoch: [175][ 1100/ 1236] Overall Loss 0.126927 Objective Loss 0.126927 LR 0.000250 Time 0.021791 +2023-10-02 21:53:45,595 - Epoch: [175][ 1110/ 1236] Overall Loss 0.126958 Objective Loss 0.126958 LR 0.000250 Time 0.021782 +2023-10-02 21:53:45,805 - Epoch: [175][ 1120/ 1236] Overall Loss 0.126867 Objective Loss 0.126867 LR 0.000250 Time 0.021775 +2023-10-02 21:53:46,014 - Epoch: [175][ 1130/ 1236] Overall Loss 0.127014 Objective Loss 0.127014 LR 0.000250 Time 0.021767 +2023-10-02 21:53:46,224 - Epoch: [175][ 1140/ 1236] Overall Loss 0.126902 Objective Loss 0.126902 LR 0.000250 Time 0.021760 +2023-10-02 21:53:46,433 - Epoch: [175][ 1150/ 1236] Overall Loss 0.126803 Objective Loss 0.126803 LR 0.000250 Time 0.021752 +2023-10-02 21:53:46,642 - Epoch: [175][ 1160/ 1236] Overall Loss 0.126940 Objective Loss 0.126940 LR 0.000250 Time 0.021745 +2023-10-02 21:53:46,851 - Epoch: [175][ 1170/ 1236] Overall Loss 0.127064 Objective Loss 0.127064 LR 0.000250 Time 0.021737 +2023-10-02 21:53:47,061 - Epoch: [175][ 1180/ 1236] Overall Loss 0.127085 Objective Loss 0.127085 LR 0.000250 Time 0.021731 +2023-10-02 21:53:47,269 - Epoch: [175][ 1190/ 1236] Overall Loss 0.127102 Objective Loss 0.127102 LR 0.000250 Time 0.021723 +2023-10-02 21:53:47,479 - Epoch: [175][ 1200/ 1236] Overall Loss 0.127244 Objective Loss 0.127244 LR 0.000250 Time 0.021716 +2023-10-02 21:53:47,688 - Epoch: [175][ 1210/ 1236] Overall Loss 0.127158 Objective Loss 0.127158 LR 0.000250 Time 0.021709 +2023-10-02 21:53:47,898 - Epoch: [175][ 1220/ 1236] Overall Loss 0.127314 Objective Loss 0.127314 LR 0.000250 Time 0.021703 +2023-10-02 21:53:48,162 - Epoch: [175][ 1230/ 1236] Overall Loss 0.127328 Objective Loss 0.127328 LR 0.000250 Time 0.021741 +2023-10-02 21:53:48,283 - Epoch: [175][ 1236/ 1236] Overall Loss 0.127270 Objective Loss 0.127270 Top1 92.668024 Top5 99.592668 LR 0.000250 Time 0.021733 +2023-10-02 21:53:48,409 - --- validate (epoch=175)----------- +2023-10-02 21:53:48,409 - 29943 samples (256 per mini-batch) +2023-10-02 21:53:48,893 - Epoch: [175][ 10/ 117] Loss 0.281556 Top1 88.085938 Top5 98.671875 +2023-10-02 21:53:49,047 - Epoch: [175][ 20/ 117] Loss 0.302088 Top1 86.914062 Top5 98.710938 +2023-10-02 21:53:49,203 - Epoch: [175][ 30/ 117] Loss 0.295236 Top1 87.031250 Top5 98.802083 +2023-10-02 21:53:49,356 - Epoch: [175][ 40/ 117] Loss 0.295102 Top1 87.304688 Top5 98.769531 +2023-10-02 21:53:49,510 - Epoch: [175][ 50/ 117] Loss 0.295909 Top1 87.296875 Top5 98.703125 +2023-10-02 21:53:49,663 - Epoch: [175][ 60/ 117] Loss 0.295751 Top1 87.122396 Top5 98.704427 +2023-10-02 21:53:49,816 - Epoch: [175][ 70/ 117] Loss 0.293998 Top1 87.003348 Top5 98.722098 +2023-10-02 21:53:49,968 - Epoch: [175][ 80/ 117] Loss 0.300534 Top1 86.943359 Top5 98.676758 +2023-10-02 21:53:50,120 - Epoch: [175][ 90/ 117] Loss 0.301170 Top1 87.022569 Top5 98.684896 +2023-10-02 21:53:50,273 - Epoch: [175][ 100/ 117] Loss 0.299610 Top1 87.015625 Top5 98.691406 +2023-10-02 21:53:50,432 - Epoch: [175][ 110/ 117] Loss 0.298813 Top1 87.120028 Top5 98.718040 +2023-10-02 21:53:50,521 - Epoch: [175][ 117/ 117] Loss 0.301294 Top1 87.102161 Top5 98.714224 +2023-10-02 21:53:50,670 - ==> Top1: 87.102 Top5: 98.714 Loss: 0.301 + +2023-10-02 21:53:50,671 - ==> Confusion: +[[ 917 0 5 1 5 2 0 0 8 75 1 2 1 0 7 1 3 0 1 0 21] + [ 0 1074 0 0 3 14 0 16 1 1 1 0 0 0 0 3 0 0 8 2 8] + [ 1 1 987 3 1 0 13 11 0 2 1 1 7 2 1 3 3 1 7 2 9] + [ 1 3 10 970 2 4 5 2 3 1 3 0 6 2 36 2 1 6 14 0 18] + [ 18 5 0 1 973 6 0 0 3 11 1 0 0 4 9 2 12 0 0 2 3] + [ 1 34 0 0 3 991 2 25 2 6 2 6 1 11 6 1 2 1 4 3 15] + [ 0 4 23 0 0 2 1135 5 0 0 3 2 0 0 0 5 0 0 2 5 5] + [ 1 17 12 0 8 20 5 1083 2 2 2 1 3 5 0 0 0 3 38 8 8] + [ 14 2 1 1 1 4 0 3 983 37 9 1 1 10 13 1 4 0 2 1 1] + [ 81 2 1 0 9 2 0 0 27 964 0 0 0 14 4 3 2 0 0 0 10] + [ 2 3 11 4 0 1 5 2 13 1 972 2 0 11 3 0 2 2 3 3 13] + [ 0 0 1 0 1 15 0 3 0 0 0 973 12 4 0 2 0 13 0 5 6] + [ 0 0 2 3 0 1 0 2 2 0 3 37 983 0 3 6 1 11 2 4 8] + [ 0 1 0 0 3 10 0 0 10 11 2 6 0 1056 4 0 0 1 0 0 15] + [ 8 1 5 10 5 1 0 0 22 2 2 0 4 2 1023 0 0 3 7 0 6] + [ 0 0 1 0 5 2 1 0 0 1 1 6 6 0 0 1071 17 9 0 9 5] + [ 0 16 1 0 4 4 2 0 0 0 0 4 0 3 2 10 1096 0 1 5 13] + [ 0 0 1 1 0 1 2 0 0 0 0 6 17 0 3 5 0 996 0 0 6] + [ 1 5 4 14 0 0 0 18 3 1 3 0 1 0 14 0 2 0 991 0 11] + [ 0 1 4 1 0 1 9 8 0 0 2 12 5 2 1 1 6 1 0 1091 7] + [ 105 142 112 50 62 111 28 91 74 66 133 90 311 243 130 48 81 48 96 132 5752]] + +2023-10-02 21:53:50,672 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:53:50,672 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:53:50,678 - + +2023-10-02 21:53:50,678 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:53:51,696 - Epoch: [176][ 10/ 1236] Overall Loss 0.128965 Objective Loss 0.128965 LR 0.000250 Time 0.101712 +2023-10-02 21:53:51,909 - Epoch: [176][ 20/ 1236] Overall Loss 0.122106 Objective Loss 0.122106 LR 0.000250 Time 0.061468 +2023-10-02 21:53:52,121 - Epoch: [176][ 30/ 1236] Overall Loss 0.122246 Objective Loss 0.122246 LR 0.000250 Time 0.047999 +2023-10-02 21:53:52,331 - Epoch: [176][ 40/ 1236] Overall Loss 0.118327 Objective Loss 0.118327 LR 0.000250 Time 0.041258 +2023-10-02 21:53:52,543 - Epoch: [176][ 50/ 1236] Overall Loss 0.118266 Objective Loss 0.118266 LR 0.000250 Time 0.037203 +2023-10-02 21:53:52,753 - Epoch: [176][ 60/ 1236] Overall Loss 0.117361 Objective Loss 0.117361 LR 0.000250 Time 0.034510 +2023-10-02 21:53:52,965 - Epoch: [176][ 70/ 1236] Overall Loss 0.118947 Objective Loss 0.118947 LR 0.000250 Time 0.032585 +2023-10-02 21:53:53,177 - Epoch: [176][ 80/ 1236] Overall Loss 0.117392 Objective Loss 0.117392 LR 0.000250 Time 0.031158 +2023-10-02 21:53:53,389 - Epoch: [176][ 90/ 1236] Overall Loss 0.115829 Objective Loss 0.115829 LR 0.000250 Time 0.030044 +2023-10-02 21:53:53,598 - Epoch: [176][ 100/ 1236] Overall Loss 0.117563 Objective Loss 0.117563 LR 0.000250 Time 0.029133 +2023-10-02 21:53:53,808 - Epoch: [176][ 110/ 1236] Overall Loss 0.119756 Objective Loss 0.119756 LR 0.000250 Time 0.028373 +2023-10-02 21:53:54,015 - Epoch: [176][ 120/ 1236] Overall Loss 0.121115 Objective Loss 0.121115 LR 0.000250 Time 0.027738 +2023-10-02 21:53:54,223 - Epoch: [176][ 130/ 1236] Overall Loss 0.121284 Objective Loss 0.121284 LR 0.000250 Time 0.027187 +2023-10-02 21:53:54,432 - Epoch: [176][ 140/ 1236] Overall Loss 0.122209 Objective Loss 0.122209 LR 0.000250 Time 0.026738 +2023-10-02 21:53:54,641 - Epoch: [176][ 150/ 1236] Overall Loss 0.122367 Objective Loss 0.122367 LR 0.000250 Time 0.026339 +2023-10-02 21:53:54,854 - Epoch: [176][ 160/ 1236] Overall Loss 0.122461 Objective Loss 0.122461 LR 0.000250 Time 0.026022 +2023-10-02 21:53:55,063 - Epoch: [176][ 170/ 1236] Overall Loss 0.123087 Objective Loss 0.123087 LR 0.000250 Time 0.025720 +2023-10-02 21:53:55,277 - Epoch: [176][ 180/ 1236] Overall Loss 0.122460 Objective Loss 0.122460 LR 0.000250 Time 0.025475 +2023-10-02 21:53:55,486 - Epoch: [176][ 190/ 1236] Overall Loss 0.122799 Objective Loss 0.122799 LR 0.000250 Time 0.025232 +2023-10-02 21:53:55,698 - Epoch: [176][ 200/ 1236] Overall Loss 0.122322 Objective Loss 0.122322 LR 0.000250 Time 0.025031 +2023-10-02 21:53:55,908 - Epoch: [176][ 210/ 1236] Overall Loss 0.123033 Objective Loss 0.123033 LR 0.000250 Time 0.024834 +2023-10-02 21:53:56,121 - Epoch: [176][ 220/ 1236] Overall Loss 0.123284 Objective Loss 0.123284 LR 0.000250 Time 0.024669 +2023-10-02 21:53:56,333 - Epoch: [176][ 230/ 1236] Overall Loss 0.123550 Objective Loss 0.123550 LR 0.000250 Time 0.024511 +2023-10-02 21:53:56,546 - Epoch: [176][ 240/ 1236] Overall Loss 0.123803 Objective Loss 0.123803 LR 0.000250 Time 0.024377 +2023-10-02 21:53:56,755 - Epoch: [176][ 250/ 1236] Overall Loss 0.123026 Objective Loss 0.123026 LR 0.000250 Time 0.024240 +2023-10-02 21:53:56,969 - Epoch: [176][ 260/ 1236] Overall Loss 0.123451 Objective Loss 0.123451 LR 0.000250 Time 0.024126 +2023-10-02 21:53:57,178 - Epoch: [176][ 270/ 1236] Overall Loss 0.123548 Objective Loss 0.123548 LR 0.000250 Time 0.024007 +2023-10-02 21:53:57,391 - Epoch: [176][ 280/ 1236] Overall Loss 0.123871 Objective Loss 0.123871 LR 0.000250 Time 0.023910 +2023-10-02 21:53:57,601 - Epoch: [176][ 290/ 1236] Overall Loss 0.123773 Objective Loss 0.123773 LR 0.000250 Time 0.023806 +2023-10-02 21:53:57,814 - Epoch: [176][ 300/ 1236] Overall Loss 0.124823 Objective Loss 0.124823 LR 0.000250 Time 0.023723 +2023-10-02 21:53:58,023 - Epoch: [176][ 310/ 1236] Overall Loss 0.124660 Objective Loss 0.124660 LR 0.000250 Time 0.023630 +2023-10-02 21:53:58,235 - Epoch: [176][ 320/ 1236] Overall Loss 0.124633 Objective Loss 0.124633 LR 0.000250 Time 0.023552 +2023-10-02 21:53:58,446 - Epoch: [176][ 330/ 1236] Overall Loss 0.124919 Objective Loss 0.124919 LR 0.000250 Time 0.023477 +2023-10-02 21:53:58,657 - Epoch: [176][ 340/ 1236] Overall Loss 0.124851 Objective Loss 0.124851 LR 0.000250 Time 0.023407 +2023-10-02 21:53:58,866 - Epoch: [176][ 350/ 1236] Overall Loss 0.125045 Objective Loss 0.125045 LR 0.000250 Time 0.023330 +2023-10-02 21:53:59,075 - Epoch: [176][ 360/ 1236] Overall Loss 0.125416 Objective Loss 0.125416 LR 0.000250 Time 0.023263 +2023-10-02 21:53:59,285 - Epoch: [176][ 370/ 1236] Overall Loss 0.125491 Objective Loss 0.125491 LR 0.000250 Time 0.023197 +2023-10-02 21:53:59,495 - Epoch: [176][ 380/ 1236] Overall Loss 0.125788 Objective Loss 0.125788 LR 0.000250 Time 0.023139 +2023-10-02 21:53:59,705 - Epoch: [176][ 390/ 1236] Overall Loss 0.125456 Objective Loss 0.125456 LR 0.000250 Time 0.023080 +2023-10-02 21:53:59,915 - Epoch: [176][ 400/ 1236] Overall Loss 0.125452 Objective Loss 0.125452 LR 0.000250 Time 0.023028 +2023-10-02 21:54:00,123 - Epoch: [176][ 410/ 1236] Overall Loss 0.125146 Objective Loss 0.125146 LR 0.000250 Time 0.022972 +2023-10-02 21:54:00,332 - Epoch: [176][ 420/ 1236] Overall Loss 0.125155 Objective Loss 0.125155 LR 0.000250 Time 0.022922 +2023-10-02 21:54:00,540 - Epoch: [176][ 430/ 1236] Overall Loss 0.125171 Objective Loss 0.125171 LR 0.000250 Time 0.022870 +2023-10-02 21:54:00,749 - Epoch: [176][ 440/ 1236] Overall Loss 0.125884 Objective Loss 0.125884 LR 0.000250 Time 0.022824 +2023-10-02 21:54:00,958 - Epoch: [176][ 450/ 1236] Overall Loss 0.126052 Objective Loss 0.126052 LR 0.000250 Time 0.022777 +2023-10-02 21:54:01,167 - Epoch: [176][ 460/ 1236] Overall Loss 0.126360 Objective Loss 0.126360 LR 0.000250 Time 0.022736 +2023-10-02 21:54:01,376 - Epoch: [176][ 470/ 1236] Overall Loss 0.125827 Objective Loss 0.125827 LR 0.000250 Time 0.022693 +2023-10-02 21:54:01,584 - Epoch: [176][ 480/ 1236] Overall Loss 0.125645 Objective Loss 0.125645 LR 0.000250 Time 0.022655 +2023-10-02 21:54:01,793 - Epoch: [176][ 490/ 1236] Overall Loss 0.125479 Objective Loss 0.125479 LR 0.000250 Time 0.022615 +2023-10-02 21:54:02,002 - Epoch: [176][ 500/ 1236] Overall Loss 0.125512 Objective Loss 0.125512 LR 0.000250 Time 0.022581 +2023-10-02 21:54:02,211 - Epoch: [176][ 510/ 1236] Overall Loss 0.125263 Objective Loss 0.125263 LR 0.000250 Time 0.022544 +2023-10-02 21:54:02,420 - Epoch: [176][ 520/ 1236] Overall Loss 0.125158 Objective Loss 0.125158 LR 0.000250 Time 0.022512 +2023-10-02 21:54:02,629 - Epoch: [176][ 530/ 1236] Overall Loss 0.125056 Objective Loss 0.125056 LR 0.000250 Time 0.022479 +2023-10-02 21:54:02,838 - Epoch: [176][ 540/ 1236] Overall Loss 0.125268 Objective Loss 0.125268 LR 0.000250 Time 0.022449 +2023-10-02 21:54:03,046 - Epoch: [176][ 550/ 1236] Overall Loss 0.125593 Objective Loss 0.125593 LR 0.000250 Time 0.022417 +2023-10-02 21:54:03,255 - Epoch: [176][ 560/ 1236] Overall Loss 0.125705 Objective Loss 0.125705 LR 0.000250 Time 0.022389 +2023-10-02 21:54:03,464 - Epoch: [176][ 570/ 1236] Overall Loss 0.125720 Objective Loss 0.125720 LR 0.000250 Time 0.022360 +2023-10-02 21:54:03,673 - Epoch: [176][ 580/ 1236] Overall Loss 0.125932 Objective Loss 0.125932 LR 0.000250 Time 0.022334 +2023-10-02 21:54:03,881 - Epoch: [176][ 590/ 1236] Overall Loss 0.125710 Objective Loss 0.125710 LR 0.000250 Time 0.022306 +2023-10-02 21:54:04,090 - Epoch: [176][ 600/ 1236] Overall Loss 0.125803 Objective Loss 0.125803 LR 0.000250 Time 0.022282 +2023-10-02 21:54:04,299 - Epoch: [176][ 610/ 1236] Overall Loss 0.126060 Objective Loss 0.126060 LR 0.000250 Time 0.022256 +2023-10-02 21:54:04,509 - Epoch: [176][ 620/ 1236] Overall Loss 0.126198 Objective Loss 0.126198 LR 0.000250 Time 0.022236 +2023-10-02 21:54:04,717 - Epoch: [176][ 630/ 1236] Overall Loss 0.126283 Objective Loss 0.126283 LR 0.000250 Time 0.022212 +2023-10-02 21:54:04,926 - Epoch: [176][ 640/ 1236] Overall Loss 0.126313 Objective Loss 0.126313 LR 0.000250 Time 0.022192 +2023-10-02 21:54:05,134 - Epoch: [176][ 650/ 1236] Overall Loss 0.126218 Objective Loss 0.126218 LR 0.000250 Time 0.022168 +2023-10-02 21:54:05,344 - Epoch: [176][ 660/ 1236] Overall Loss 0.126285 Objective Loss 0.126285 LR 0.000250 Time 0.022149 +2023-10-02 21:54:05,552 - Epoch: [176][ 670/ 1236] Overall Loss 0.126284 Objective Loss 0.126284 LR 0.000250 Time 0.022128 +2023-10-02 21:54:05,761 - Epoch: [176][ 680/ 1236] Overall Loss 0.126478 Objective Loss 0.126478 LR 0.000250 Time 0.022110 +2023-10-02 21:54:05,970 - Epoch: [176][ 690/ 1236] Overall Loss 0.126675 Objective Loss 0.126675 LR 0.000250 Time 0.022089 +2023-10-02 21:54:06,180 - Epoch: [176][ 700/ 1236] Overall Loss 0.126385 Objective Loss 0.126385 LR 0.000250 Time 0.022074 +2023-10-02 21:54:06,388 - Epoch: [176][ 710/ 1236] Overall Loss 0.126680 Objective Loss 0.126680 LR 0.000250 Time 0.022055 +2023-10-02 21:54:06,597 - Epoch: [176][ 720/ 1236] Overall Loss 0.126506 Objective Loss 0.126506 LR 0.000250 Time 0.022038 +2023-10-02 21:54:06,805 - Epoch: [176][ 730/ 1236] Overall Loss 0.126564 Objective Loss 0.126564 LR 0.000250 Time 0.022020 +2023-10-02 21:54:07,014 - Epoch: [176][ 740/ 1236] Overall Loss 0.126410 Objective Loss 0.126410 LR 0.000250 Time 0.022004 +2023-10-02 21:54:07,223 - Epoch: [176][ 750/ 1236] Overall Loss 0.126712 Objective Loss 0.126712 LR 0.000250 Time 0.021987 +2023-10-02 21:54:07,433 - Epoch: [176][ 760/ 1236] Overall Loss 0.126806 Objective Loss 0.126806 LR 0.000250 Time 0.021974 +2023-10-02 21:54:07,641 - Epoch: [176][ 770/ 1236] Overall Loss 0.126884 Objective Loss 0.126884 LR 0.000250 Time 0.021958 +2023-10-02 21:54:07,850 - Epoch: [176][ 780/ 1236] Overall Loss 0.126885 Objective Loss 0.126885 LR 0.000250 Time 0.021945 +2023-10-02 21:54:08,059 - Epoch: [176][ 790/ 1236] Overall Loss 0.127066 Objective Loss 0.127066 LR 0.000250 Time 0.021929 +2023-10-02 21:54:08,268 - Epoch: [176][ 800/ 1236] Overall Loss 0.126971 Objective Loss 0.126971 LR 0.000250 Time 0.021916 +2023-10-02 21:54:08,477 - Epoch: [176][ 810/ 1236] Overall Loss 0.126890 Objective Loss 0.126890 LR 0.000250 Time 0.021901 +2023-10-02 21:54:08,686 - Epoch: [176][ 820/ 1236] Overall Loss 0.126877 Objective Loss 0.126877 LR 0.000250 Time 0.021889 +2023-10-02 21:54:08,895 - Epoch: [176][ 830/ 1236] Overall Loss 0.127038 Objective Loss 0.127038 LR 0.000250 Time 0.021875 +2023-10-02 21:54:09,105 - Epoch: [176][ 840/ 1236] Overall Loss 0.126872 Objective Loss 0.126872 LR 0.000250 Time 0.021864 +2023-10-02 21:54:09,313 - Epoch: [176][ 850/ 1236] Overall Loss 0.126870 Objective Loss 0.126870 LR 0.000250 Time 0.021850 +2023-10-02 21:54:09,522 - Epoch: [176][ 860/ 1236] Overall Loss 0.126859 Objective Loss 0.126859 LR 0.000250 Time 0.021839 +2023-10-02 21:54:09,731 - Epoch: [176][ 870/ 1236] Overall Loss 0.126911 Objective Loss 0.126911 LR 0.000250 Time 0.021826 +2023-10-02 21:54:09,941 - Epoch: [176][ 880/ 1236] Overall Loss 0.127040 Objective Loss 0.127040 LR 0.000250 Time 0.021816 +2023-10-02 21:54:10,149 - Epoch: [176][ 890/ 1236] Overall Loss 0.126901 Objective Loss 0.126901 LR 0.000250 Time 0.021804 +2023-10-02 21:54:10,359 - Epoch: [176][ 900/ 1236] Overall Loss 0.126945 Objective Loss 0.126945 LR 0.000250 Time 0.021794 +2023-10-02 21:54:10,567 - Epoch: [176][ 910/ 1236] Overall Loss 0.127093 Objective Loss 0.127093 LR 0.000250 Time 0.021782 +2023-10-02 21:54:10,777 - Epoch: [176][ 920/ 1236] Overall Loss 0.127028 Objective Loss 0.127028 LR 0.000250 Time 0.021772 +2023-10-02 21:54:10,985 - Epoch: [176][ 930/ 1236] Overall Loss 0.127048 Objective Loss 0.127048 LR 0.000250 Time 0.021760 +2023-10-02 21:54:11,195 - Epoch: [176][ 940/ 1236] Overall Loss 0.127004 Objective Loss 0.127004 LR 0.000250 Time 0.021753 +2023-10-02 21:54:11,403 - Epoch: [176][ 950/ 1236] Overall Loss 0.126975 Objective Loss 0.126975 LR 0.000250 Time 0.021742 +2023-10-02 21:54:11,613 - Epoch: [176][ 960/ 1236] Overall Loss 0.126877 Objective Loss 0.126877 LR 0.000250 Time 0.021733 +2023-10-02 21:54:11,821 - Epoch: [176][ 970/ 1236] Overall Loss 0.126946 Objective Loss 0.126946 LR 0.000250 Time 0.021723 +2023-10-02 21:54:12,031 - Epoch: [176][ 980/ 1236] Overall Loss 0.126900 Objective Loss 0.126900 LR 0.000250 Time 0.021715 +2023-10-02 21:54:12,239 - Epoch: [176][ 990/ 1236] Overall Loss 0.126990 Objective Loss 0.126990 LR 0.000250 Time 0.021705 +2023-10-02 21:54:12,449 - Epoch: [176][ 1000/ 1236] Overall Loss 0.126957 Objective Loss 0.126957 LR 0.000250 Time 0.021697 +2023-10-02 21:54:12,657 - Epoch: [176][ 1010/ 1236] Overall Loss 0.127000 Objective Loss 0.127000 LR 0.000250 Time 0.021687 +2023-10-02 21:54:12,867 - Epoch: [176][ 1020/ 1236] Overall Loss 0.127045 Objective Loss 0.127045 LR 0.000250 Time 0.021679 +2023-10-02 21:54:13,075 - Epoch: [176][ 1030/ 1236] Overall Loss 0.127034 Objective Loss 0.127034 LR 0.000250 Time 0.021670 +2023-10-02 21:54:13,285 - Epoch: [176][ 1040/ 1236] Overall Loss 0.127061 Objective Loss 0.127061 LR 0.000250 Time 0.021663 +2023-10-02 21:54:13,493 - Epoch: [176][ 1050/ 1236] Overall Loss 0.127271 Objective Loss 0.127271 LR 0.000250 Time 0.021653 +2023-10-02 21:54:13,703 - Epoch: [176][ 1060/ 1236] Overall Loss 0.127239 Objective Loss 0.127239 LR 0.000250 Time 0.021646 +2023-10-02 21:54:13,911 - Epoch: [176][ 1070/ 1236] Overall Loss 0.127435 Objective Loss 0.127435 LR 0.000250 Time 0.021637 +2023-10-02 21:54:14,121 - Epoch: [176][ 1080/ 1236] Overall Loss 0.127450 Objective Loss 0.127450 LR 0.000250 Time 0.021631 +2023-10-02 21:54:14,329 - Epoch: [176][ 1090/ 1236] Overall Loss 0.127334 Objective Loss 0.127334 LR 0.000250 Time 0.021622 +2023-10-02 21:54:14,539 - Epoch: [176][ 1100/ 1236] Overall Loss 0.127323 Objective Loss 0.127323 LR 0.000250 Time 0.021616 +2023-10-02 21:54:14,747 - Epoch: [176][ 1110/ 1236] Overall Loss 0.127350 Objective Loss 0.127350 LR 0.000250 Time 0.021608 +2023-10-02 21:54:14,957 - Epoch: [176][ 1120/ 1236] Overall Loss 0.127442 Objective Loss 0.127442 LR 0.000250 Time 0.021602 +2023-10-02 21:54:15,166 - Epoch: [176][ 1130/ 1236] Overall Loss 0.127490 Objective Loss 0.127490 LR 0.000250 Time 0.021594 +2023-10-02 21:54:15,375 - Epoch: [176][ 1140/ 1236] Overall Loss 0.127694 Objective Loss 0.127694 LR 0.000250 Time 0.021588 +2023-10-02 21:54:15,584 - Epoch: [176][ 1150/ 1236] Overall Loss 0.127677 Objective Loss 0.127677 LR 0.000250 Time 0.021581 +2023-10-02 21:54:15,794 - Epoch: [176][ 1160/ 1236] Overall Loss 0.127692 Objective Loss 0.127692 LR 0.000250 Time 0.021575 +2023-10-02 21:54:16,003 - Epoch: [176][ 1170/ 1236] Overall Loss 0.127832 Objective Loss 0.127832 LR 0.000250 Time 0.021568 +2023-10-02 21:54:16,212 - Epoch: [176][ 1180/ 1236] Overall Loss 0.127844 Objective Loss 0.127844 LR 0.000250 Time 0.021563 +2023-10-02 21:54:16,421 - Epoch: [176][ 1190/ 1236] Overall Loss 0.127978 Objective Loss 0.127978 LR 0.000250 Time 0.021555 +2023-10-02 21:54:16,630 - Epoch: [176][ 1200/ 1236] Overall Loss 0.128035 Objective Loss 0.128035 LR 0.000250 Time 0.021550 +2023-10-02 21:54:16,839 - Epoch: [176][ 1210/ 1236] Overall Loss 0.128043 Objective Loss 0.128043 LR 0.000250 Time 0.021543 +2023-10-02 21:54:17,049 - Epoch: [176][ 1220/ 1236] Overall Loss 0.128036 Objective Loss 0.128036 LR 0.000250 Time 0.021538 +2023-10-02 21:54:17,311 - Epoch: [176][ 1230/ 1236] Overall Loss 0.128122 Objective Loss 0.128122 LR 0.000250 Time 0.021575 +2023-10-02 21:54:17,434 - Epoch: [176][ 1236/ 1236] Overall Loss 0.128169 Objective Loss 0.128169 Top1 92.057026 Top5 98.574338 LR 0.000250 Time 0.021569 +2023-10-02 21:54:17,571 - --- validate (epoch=176)----------- +2023-10-02 21:54:17,571 - 29943 samples (256 per mini-batch) +2023-10-02 21:54:18,066 - Epoch: [176][ 10/ 117] Loss 0.284729 Top1 87.656250 Top5 98.828125 +2023-10-02 21:54:18,219 - Epoch: [176][ 20/ 117] Loss 0.302966 Top1 87.363281 Top5 98.828125 +2023-10-02 21:54:18,371 - Epoch: [176][ 30/ 117] Loss 0.315873 Top1 86.992188 Top5 98.658854 +2023-10-02 21:54:18,522 - Epoch: [176][ 40/ 117] Loss 0.311704 Top1 87.382812 Top5 98.701172 +2023-10-02 21:54:18,673 - Epoch: [176][ 50/ 117] Loss 0.308412 Top1 87.445312 Top5 98.781250 +2023-10-02 21:54:18,825 - Epoch: [176][ 60/ 117] Loss 0.304613 Top1 87.688802 Top5 98.736979 +2023-10-02 21:54:18,980 - Epoch: [176][ 70/ 117] Loss 0.308644 Top1 87.555804 Top5 98.733259 +2023-10-02 21:54:19,131 - Epoch: [176][ 80/ 117] Loss 0.309022 Top1 87.641602 Top5 98.691406 +2023-10-02 21:54:19,281 - Epoch: [176][ 90/ 117] Loss 0.310306 Top1 87.638889 Top5 98.671875 +2023-10-02 21:54:19,432 - Epoch: [176][ 100/ 117] Loss 0.308752 Top1 87.574219 Top5 98.644531 +2023-10-02 21:54:19,590 - Epoch: [176][ 110/ 117] Loss 0.309041 Top1 87.514205 Top5 98.650568 +2023-10-02 21:54:19,678 - Epoch: [176][ 117/ 117] Loss 0.309493 Top1 87.479544 Top5 98.664129 +2023-10-02 21:54:19,776 - ==> Top1: 87.480 Top5: 98.664 Loss: 0.309 + +2023-10-02 21:54:19,777 - ==> Confusion: +[[ 942 0 6 1 7 2 0 0 5 51 1 2 1 1 6 1 1 0 1 0 22] + [ 1 1068 0 1 3 13 0 22 1 1 0 1 0 0 1 2 1 0 5 3 8] + [ 1 0 989 4 2 0 11 7 0 0 3 1 5 2 1 3 3 2 10 4 8] + [ 0 3 13 981 1 2 2 2 2 1 2 0 6 2 28 3 2 5 13 0 21] + [ 19 5 0 0 978 3 0 0 0 10 0 0 1 3 13 5 7 0 0 1 5] + [ 4 38 1 2 6 982 1 25 1 6 1 5 3 11 3 0 3 0 2 3 19] + [ 0 5 28 1 0 3 1128 5 0 0 2 0 0 0 0 5 0 0 2 5 7] + [ 3 17 12 1 4 18 5 1073 2 3 4 1 5 5 1 0 0 3 43 9 9] + [ 18 2 0 1 2 2 0 1 967 38 15 1 2 12 17 0 5 1 2 0 3] + [ 103 1 0 1 9 2 0 0 27 930 0 1 0 26 7 2 2 0 0 0 8] + [ 3 1 8 6 0 1 3 1 6 0 978 2 0 13 5 0 4 1 6 2 13] + [ 0 0 2 0 0 14 0 5 0 0 0 963 19 2 0 2 1 13 0 6 8] + [ 0 0 1 3 0 0 2 2 0 1 5 31 975 1 2 7 4 11 3 3 17] + [ 0 0 0 0 3 5 0 0 9 9 1 9 0 1059 4 0 0 1 0 1 18] + [ 10 0 5 16 4 1 0 0 19 2 0 0 6 3 1013 0 1 1 10 0 10] + [ 0 0 2 2 4 1 0 0 0 0 0 3 10 0 0 1071 15 11 2 7 6] + [ 1 13 2 0 5 6 2 0 1 0 0 4 0 3 4 8 1097 0 0 2 13] + [ 0 0 1 1 0 0 2 0 0 1 0 4 18 1 3 7 1 989 0 3 7] + [ 2 3 4 16 1 0 0 19 4 1 3 0 1 0 7 0 1 0 994 0 12] + [ 0 2 4 3 2 3 9 3 0 0 2 12 5 2 0 1 7 1 0 1087 9] + [ 103 108 110 68 59 93 28 70 65 57 137 78 251 247 117 52 74 48 90 120 5930]] + +2023-10-02 21:54:19,778 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:54:19,778 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:54:19,784 - + +2023-10-02 21:54:19,784 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:54:20,805 - Epoch: [177][ 10/ 1236] Overall Loss 0.122011 Objective Loss 0.122011 LR 0.000250 Time 0.102060 +2023-10-02 21:54:21,017 - Epoch: [177][ 20/ 1236] Overall Loss 0.118609 Objective Loss 0.118609 LR 0.000250 Time 0.061579 +2023-10-02 21:54:21,226 - Epoch: [177][ 30/ 1236] Overall Loss 0.126587 Objective Loss 0.126587 LR 0.000250 Time 0.048026 +2023-10-02 21:54:21,437 - Epoch: [177][ 40/ 1236] Overall Loss 0.126435 Objective Loss 0.126435 LR 0.000250 Time 0.041288 +2023-10-02 21:54:21,647 - Epoch: [177][ 50/ 1236] Overall Loss 0.126694 Objective Loss 0.126694 LR 0.000250 Time 0.037193 +2023-10-02 21:54:21,858 - Epoch: [177][ 60/ 1236] Overall Loss 0.125219 Objective Loss 0.125219 LR 0.000250 Time 0.034509 +2023-10-02 21:54:22,067 - Epoch: [177][ 70/ 1236] Overall Loss 0.126434 Objective Loss 0.126434 LR 0.000250 Time 0.032566 +2023-10-02 21:54:22,279 - Epoch: [177][ 80/ 1236] Overall Loss 0.129169 Objective Loss 0.129169 LR 0.000250 Time 0.031138 +2023-10-02 21:54:22,489 - Epoch: [177][ 90/ 1236] Overall Loss 0.128126 Objective Loss 0.128126 LR 0.000250 Time 0.030008 +2023-10-02 21:54:22,699 - Epoch: [177][ 100/ 1236] Overall Loss 0.128036 Objective Loss 0.128036 LR 0.000250 Time 0.029106 +2023-10-02 21:54:22,907 - Epoch: [177][ 110/ 1236] Overall Loss 0.127162 Objective Loss 0.127162 LR 0.000250 Time 0.028337 +2023-10-02 21:54:23,117 - Epoch: [177][ 120/ 1236] Overall Loss 0.126887 Objective Loss 0.126887 LR 0.000250 Time 0.027723 +2023-10-02 21:54:23,327 - Epoch: [177][ 130/ 1236] Overall Loss 0.126684 Objective Loss 0.126684 LR 0.000250 Time 0.027202 +2023-10-02 21:54:23,536 - Epoch: [177][ 140/ 1236] Overall Loss 0.128186 Objective Loss 0.128186 LR 0.000250 Time 0.026753 +2023-10-02 21:54:23,745 - Epoch: [177][ 150/ 1236] Overall Loss 0.127442 Objective Loss 0.127442 LR 0.000250 Time 0.026346 +2023-10-02 21:54:23,955 - Epoch: [177][ 160/ 1236] Overall Loss 0.128068 Objective Loss 0.128068 LR 0.000250 Time 0.026016 +2023-10-02 21:54:24,162 - Epoch: [177][ 170/ 1236] Overall Loss 0.128423 Objective Loss 0.128423 LR 0.000250 Time 0.025700 +2023-10-02 21:54:24,372 - Epoch: [177][ 180/ 1236] Overall Loss 0.128380 Objective Loss 0.128380 LR 0.000250 Time 0.025437 +2023-10-02 21:54:24,580 - Epoch: [177][ 190/ 1236] Overall Loss 0.129156 Objective Loss 0.129156 LR 0.000250 Time 0.025186 +2023-10-02 21:54:24,790 - Epoch: [177][ 200/ 1236] Overall Loss 0.128959 Objective Loss 0.128959 LR 0.000250 Time 0.024976 +2023-10-02 21:54:24,998 - Epoch: [177][ 210/ 1236] Overall Loss 0.128401 Objective Loss 0.128401 LR 0.000250 Time 0.024770 +2023-10-02 21:54:25,208 - Epoch: [177][ 220/ 1236] Overall Loss 0.128683 Objective Loss 0.128683 LR 0.000250 Time 0.024597 +2023-10-02 21:54:25,416 - Epoch: [177][ 230/ 1236] Overall Loss 0.128929 Objective Loss 0.128929 LR 0.000250 Time 0.024425 +2023-10-02 21:54:25,627 - Epoch: [177][ 240/ 1236] Overall Loss 0.129839 Objective Loss 0.129839 LR 0.000250 Time 0.024283 +2023-10-02 21:54:25,835 - Epoch: [177][ 250/ 1236] Overall Loss 0.129280 Objective Loss 0.129280 LR 0.000250 Time 0.024137 +2023-10-02 21:54:26,045 - Epoch: [177][ 260/ 1236] Overall Loss 0.129695 Objective Loss 0.129695 LR 0.000250 Time 0.024015 +2023-10-02 21:54:26,253 - Epoch: [177][ 270/ 1236] Overall Loss 0.129612 Objective Loss 0.129612 LR 0.000250 Time 0.023891 +2023-10-02 21:54:26,463 - Epoch: [177][ 280/ 1236] Overall Loss 0.129321 Objective Loss 0.129321 LR 0.000250 Time 0.023787 +2023-10-02 21:54:26,671 - Epoch: [177][ 290/ 1236] Overall Loss 0.129471 Objective Loss 0.129471 LR 0.000250 Time 0.023679 +2023-10-02 21:54:26,881 - Epoch: [177][ 300/ 1236] Overall Loss 0.129324 Objective Loss 0.129324 LR 0.000250 Time 0.023589 +2023-10-02 21:54:27,089 - Epoch: [177][ 310/ 1236] Overall Loss 0.129002 Objective Loss 0.129002 LR 0.000250 Time 0.023494 +2023-10-02 21:54:27,299 - Epoch: [177][ 320/ 1236] Overall Loss 0.128671 Objective Loss 0.128671 LR 0.000250 Time 0.023415 +2023-10-02 21:54:27,507 - Epoch: [177][ 330/ 1236] Overall Loss 0.128021 Objective Loss 0.128021 LR 0.000250 Time 0.023332 +2023-10-02 21:54:27,717 - Epoch: [177][ 340/ 1236] Overall Loss 0.127521 Objective Loss 0.127521 LR 0.000250 Time 0.023263 +2023-10-02 21:54:27,930 - Epoch: [177][ 350/ 1236] Overall Loss 0.127979 Objective Loss 0.127979 LR 0.000250 Time 0.023206 +2023-10-02 21:54:28,153 - Epoch: [177][ 360/ 1236] Overall Loss 0.127906 Objective Loss 0.127906 LR 0.000250 Time 0.023178 +2023-10-02 21:54:28,369 - Epoch: [177][ 370/ 1236] Overall Loss 0.127677 Objective Loss 0.127677 LR 0.000250 Time 0.023135 +2023-10-02 21:54:28,589 - Epoch: [177][ 380/ 1236] Overall Loss 0.127853 Objective Loss 0.127853 LR 0.000250 Time 0.023104 +2023-10-02 21:54:28,805 - Epoch: [177][ 390/ 1236] Overall Loss 0.127553 Objective Loss 0.127553 LR 0.000250 Time 0.023065 +2023-10-02 21:54:29,027 - Epoch: [177][ 400/ 1236] Overall Loss 0.128039 Objective Loss 0.128039 LR 0.000250 Time 0.023041 +2023-10-02 21:54:29,242 - Epoch: [177][ 410/ 1236] Overall Loss 0.127888 Objective Loss 0.127888 LR 0.000250 Time 0.023005 +2023-10-02 21:54:29,464 - Epoch: [177][ 420/ 1236] Overall Loss 0.127651 Objective Loss 0.127651 LR 0.000250 Time 0.022984 +2023-10-02 21:54:29,680 - Epoch: [177][ 430/ 1236] Overall Loss 0.127099 Objective Loss 0.127099 LR 0.000250 Time 0.022951 +2023-10-02 21:54:29,901 - Epoch: [177][ 440/ 1236] Overall Loss 0.127168 Objective Loss 0.127168 LR 0.000250 Time 0.022932 +2023-10-02 21:54:30,117 - Epoch: [177][ 450/ 1236] Overall Loss 0.126868 Objective Loss 0.126868 LR 0.000250 Time 0.022901 +2023-10-02 21:54:30,339 - Epoch: [177][ 460/ 1236] Overall Loss 0.127087 Objective Loss 0.127087 LR 0.000250 Time 0.022884 +2023-10-02 21:54:30,555 - Epoch: [177][ 470/ 1236] Overall Loss 0.126792 Objective Loss 0.126792 LR 0.000250 Time 0.022856 +2023-10-02 21:54:30,777 - Epoch: [177][ 480/ 1236] Overall Loss 0.126946 Objective Loss 0.126946 LR 0.000250 Time 0.022841 +2023-10-02 21:54:30,992 - Epoch: [177][ 490/ 1236] Overall Loss 0.127037 Objective Loss 0.127037 LR 0.000250 Time 0.022815 +2023-10-02 21:54:31,214 - Epoch: [177][ 500/ 1236] Overall Loss 0.127567 Objective Loss 0.127567 LR 0.000250 Time 0.022802 +2023-10-02 21:54:31,430 - Epoch: [177][ 510/ 1236] Overall Loss 0.127589 Objective Loss 0.127589 LR 0.000250 Time 0.022778 +2023-10-02 21:54:31,652 - Epoch: [177][ 520/ 1236] Overall Loss 0.127820 Objective Loss 0.127820 LR 0.000250 Time 0.022765 +2023-10-02 21:54:31,868 - Epoch: [177][ 530/ 1236] Overall Loss 0.127493 Objective Loss 0.127493 LR 0.000250 Time 0.022743 +2023-10-02 21:54:32,090 - Epoch: [177][ 540/ 1236] Overall Loss 0.127599 Objective Loss 0.127599 LR 0.000250 Time 0.022732 +2023-10-02 21:54:32,306 - Epoch: [177][ 550/ 1236] Overall Loss 0.127615 Objective Loss 0.127615 LR 0.000250 Time 0.022711 +2023-10-02 21:54:32,528 - Epoch: [177][ 560/ 1236] Overall Loss 0.127647 Objective Loss 0.127647 LR 0.000250 Time 0.022700 +2023-10-02 21:54:32,744 - Epoch: [177][ 570/ 1236] Overall Loss 0.127568 Objective Loss 0.127568 LR 0.000250 Time 0.022681 +2023-10-02 21:54:32,965 - Epoch: [177][ 580/ 1236] Overall Loss 0.127497 Objective Loss 0.127497 LR 0.000250 Time 0.022671 +2023-10-02 21:54:33,181 - Epoch: [177][ 590/ 1236] Overall Loss 0.127448 Objective Loss 0.127448 LR 0.000250 Time 0.022652 +2023-10-02 21:54:33,403 - Epoch: [177][ 600/ 1236] Overall Loss 0.127501 Objective Loss 0.127501 LR 0.000250 Time 0.022643 +2023-10-02 21:54:33,619 - Epoch: [177][ 610/ 1236] Overall Loss 0.127503 Objective Loss 0.127503 LR 0.000250 Time 0.022626 +2023-10-02 21:54:33,841 - Epoch: [177][ 620/ 1236] Overall Loss 0.127281 Objective Loss 0.127281 LR 0.000250 Time 0.022618 +2023-10-02 21:54:34,056 - Epoch: [177][ 630/ 1236] Overall Loss 0.127532 Objective Loss 0.127532 LR 0.000250 Time 0.022601 +2023-10-02 21:54:34,279 - Epoch: [177][ 640/ 1236] Overall Loss 0.127843 Objective Loss 0.127843 LR 0.000250 Time 0.022594 +2023-10-02 21:54:34,493 - Epoch: [177][ 650/ 1236] Overall Loss 0.127902 Objective Loss 0.127902 LR 0.000250 Time 0.022576 +2023-10-02 21:54:34,711 - Epoch: [177][ 660/ 1236] Overall Loss 0.127991 Objective Loss 0.127991 LR 0.000250 Time 0.022564 +2023-10-02 21:54:34,933 - Epoch: [177][ 670/ 1236] Overall Loss 0.127943 Objective Loss 0.127943 LR 0.000250 Time 0.022558 +2023-10-02 21:54:35,152 - Epoch: [177][ 680/ 1236] Overall Loss 0.127959 Objective Loss 0.127959 LR 0.000250 Time 0.022547 +2023-10-02 21:54:35,375 - Epoch: [177][ 690/ 1236] Overall Loss 0.128144 Objective Loss 0.128144 LR 0.000250 Time 0.022543 +2023-10-02 21:54:35,593 - Epoch: [177][ 700/ 1236] Overall Loss 0.128270 Objective Loss 0.128270 LR 0.000250 Time 0.022532 +2023-10-02 21:54:35,815 - Epoch: [177][ 710/ 1236] Overall Loss 0.128130 Objective Loss 0.128130 LR 0.000250 Time 0.022527 +2023-10-02 21:54:36,033 - Epoch: [177][ 720/ 1236] Overall Loss 0.128240 Objective Loss 0.128240 LR 0.000250 Time 0.022517 +2023-10-02 21:54:36,256 - Epoch: [177][ 730/ 1236] Overall Loss 0.127968 Objective Loss 0.127968 LR 0.000250 Time 0.022513 +2023-10-02 21:54:36,474 - Epoch: [177][ 740/ 1236] Overall Loss 0.127813 Objective Loss 0.127813 LR 0.000250 Time 0.022504 +2023-10-02 21:54:36,697 - Epoch: [177][ 750/ 1236] Overall Loss 0.128027 Objective Loss 0.128027 LR 0.000250 Time 0.022499 +2023-10-02 21:54:36,915 - Epoch: [177][ 760/ 1236] Overall Loss 0.128163 Objective Loss 0.128163 LR 0.000250 Time 0.022490 +2023-10-02 21:54:37,137 - Epoch: [177][ 770/ 1236] Overall Loss 0.128428 Objective Loss 0.128428 LR 0.000250 Time 0.022487 +2023-10-02 21:54:37,356 - Epoch: [177][ 780/ 1236] Overall Loss 0.128512 Objective Loss 0.128512 LR 0.000250 Time 0.022478 +2023-10-02 21:54:37,579 - Epoch: [177][ 790/ 1236] Overall Loss 0.128364 Objective Loss 0.128364 LR 0.000250 Time 0.022475 +2023-10-02 21:54:37,797 - Epoch: [177][ 800/ 1236] Overall Loss 0.128276 Objective Loss 0.128276 LR 0.000250 Time 0.022466 +2023-10-02 21:54:38,019 - Epoch: [177][ 810/ 1236] Overall Loss 0.128323 Objective Loss 0.128323 LR 0.000250 Time 0.022463 +2023-10-02 21:54:38,237 - Epoch: [177][ 820/ 1236] Overall Loss 0.128263 Objective Loss 0.128263 LR 0.000250 Time 0.022455 +2023-10-02 21:54:38,454 - Epoch: [177][ 830/ 1236] Overall Loss 0.128277 Objective Loss 0.128277 LR 0.000250 Time 0.022445 +2023-10-02 21:54:38,666 - Epoch: [177][ 840/ 1236] Overall Loss 0.128406 Objective Loss 0.128406 LR 0.000250 Time 0.022430 +2023-10-02 21:54:38,879 - Epoch: [177][ 850/ 1236] Overall Loss 0.128387 Objective Loss 0.128387 LR 0.000250 Time 0.022416 +2023-10-02 21:54:39,091 - Epoch: [177][ 860/ 1236] Overall Loss 0.128146 Objective Loss 0.128146 LR 0.000250 Time 0.022401 +2023-10-02 21:54:39,304 - Epoch: [177][ 870/ 1236] Overall Loss 0.128093 Objective Loss 0.128093 LR 0.000250 Time 0.022388 +2023-10-02 21:54:39,516 - Epoch: [177][ 880/ 1236] Overall Loss 0.128081 Objective Loss 0.128081 LR 0.000250 Time 0.022374 +2023-10-02 21:54:39,729 - Epoch: [177][ 890/ 1236] Overall Loss 0.128166 Objective Loss 0.128166 LR 0.000250 Time 0.022362 +2023-10-02 21:54:39,941 - Epoch: [177][ 900/ 1236] Overall Loss 0.128030 Objective Loss 0.128030 LR 0.000250 Time 0.022349 +2023-10-02 21:54:40,154 - Epoch: [177][ 910/ 1236] Overall Loss 0.128057 Objective Loss 0.128057 LR 0.000250 Time 0.022337 +2023-10-02 21:54:40,366 - Epoch: [177][ 920/ 1236] Overall Loss 0.127993 Objective Loss 0.127993 LR 0.000250 Time 0.022324 +2023-10-02 21:54:40,579 - Epoch: [177][ 930/ 1236] Overall Loss 0.127991 Objective Loss 0.127991 LR 0.000250 Time 0.022312 +2023-10-02 21:54:40,791 - Epoch: [177][ 940/ 1236] Overall Loss 0.128018 Objective Loss 0.128018 LR 0.000250 Time 0.022300 +2023-10-02 21:54:41,003 - Epoch: [177][ 950/ 1236] Overall Loss 0.127909 Objective Loss 0.127909 LR 0.000250 Time 0.022288 +2023-10-02 21:54:41,215 - Epoch: [177][ 960/ 1236] Overall Loss 0.127842 Objective Loss 0.127842 LR 0.000250 Time 0.022277 +2023-10-02 21:54:41,428 - Epoch: [177][ 970/ 1236] Overall Loss 0.127775 Objective Loss 0.127775 LR 0.000250 Time 0.022266 +2023-10-02 21:54:41,640 - Epoch: [177][ 980/ 1236] Overall Loss 0.127603 Objective Loss 0.127603 LR 0.000250 Time 0.022255 +2023-10-02 21:54:41,853 - Epoch: [177][ 990/ 1236] Overall Loss 0.127754 Objective Loss 0.127754 LR 0.000250 Time 0.022245 +2023-10-02 21:54:42,065 - Epoch: [177][ 1000/ 1236] Overall Loss 0.127688 Objective Loss 0.127688 LR 0.000250 Time 0.022234 +2023-10-02 21:54:42,278 - Epoch: [177][ 1010/ 1236] Overall Loss 0.127809 Objective Loss 0.127809 LR 0.000250 Time 0.022225 +2023-10-02 21:54:42,490 - Epoch: [177][ 1020/ 1236] Overall Loss 0.128001 Objective Loss 0.128001 LR 0.000250 Time 0.022215 +2023-10-02 21:54:42,703 - Epoch: [177][ 1030/ 1236] Overall Loss 0.128159 Objective Loss 0.128159 LR 0.000250 Time 0.022205 +2023-10-02 21:54:42,915 - Epoch: [177][ 1040/ 1236] Overall Loss 0.128158 Objective Loss 0.128158 LR 0.000250 Time 0.022195 +2023-10-02 21:54:43,128 - Epoch: [177][ 1050/ 1236] Overall Loss 0.128200 Objective Loss 0.128200 LR 0.000250 Time 0.022186 +2023-10-02 21:54:43,340 - Epoch: [177][ 1060/ 1236] Overall Loss 0.128108 Objective Loss 0.128108 LR 0.000250 Time 0.022177 +2023-10-02 21:54:43,553 - Epoch: [177][ 1070/ 1236] Overall Loss 0.128245 Objective Loss 0.128245 LR 0.000250 Time 0.022168 +2023-10-02 21:54:43,765 - Epoch: [177][ 1080/ 1236] Overall Loss 0.128173 Objective Loss 0.128173 LR 0.000250 Time 0.022159 +2023-10-02 21:54:43,978 - Epoch: [177][ 1090/ 1236] Overall Loss 0.128016 Objective Loss 0.128016 LR 0.000250 Time 0.022151 +2023-10-02 21:54:44,190 - Epoch: [177][ 1100/ 1236] Overall Loss 0.127973 Objective Loss 0.127973 LR 0.000250 Time 0.022142 +2023-10-02 21:54:44,403 - Epoch: [177][ 1110/ 1236] Overall Loss 0.127986 Objective Loss 0.127986 LR 0.000250 Time 0.022134 +2023-10-02 21:54:44,615 - Epoch: [177][ 1120/ 1236] Overall Loss 0.127875 Objective Loss 0.127875 LR 0.000250 Time 0.022125 +2023-10-02 21:54:44,828 - Epoch: [177][ 1130/ 1236] Overall Loss 0.127862 Objective Loss 0.127862 LR 0.000250 Time 0.022118 +2023-10-02 21:54:45,040 - Epoch: [177][ 1140/ 1236] Overall Loss 0.128020 Objective Loss 0.128020 LR 0.000250 Time 0.022109 +2023-10-02 21:54:45,253 - Epoch: [177][ 1150/ 1236] Overall Loss 0.127903 Objective Loss 0.127903 LR 0.000250 Time 0.022102 +2023-10-02 21:54:45,465 - Epoch: [177][ 1160/ 1236] Overall Loss 0.127954 Objective Loss 0.127954 LR 0.000250 Time 0.022094 +2023-10-02 21:54:45,678 - Epoch: [177][ 1170/ 1236] Overall Loss 0.128110 Objective Loss 0.128110 LR 0.000250 Time 0.022087 +2023-10-02 21:54:45,890 - Epoch: [177][ 1180/ 1236] Overall Loss 0.128111 Objective Loss 0.128111 LR 0.000250 Time 0.022079 +2023-10-02 21:54:46,103 - Epoch: [177][ 1190/ 1236] Overall Loss 0.128242 Objective Loss 0.128242 LR 0.000250 Time 0.022072 +2023-10-02 21:54:46,315 - Epoch: [177][ 1200/ 1236] Overall Loss 0.128254 Objective Loss 0.128254 LR 0.000250 Time 0.022065 +2023-10-02 21:54:46,528 - Epoch: [177][ 1210/ 1236] Overall Loss 0.128329 Objective Loss 0.128329 LR 0.000250 Time 0.022058 +2023-10-02 21:54:46,740 - Epoch: [177][ 1220/ 1236] Overall Loss 0.128374 Objective Loss 0.128374 LR 0.000250 Time 0.022051 +2023-10-02 21:54:47,006 - Epoch: [177][ 1230/ 1236] Overall Loss 0.128271 Objective Loss 0.128271 LR 0.000250 Time 0.022087 +2023-10-02 21:54:47,129 - Epoch: [177][ 1236/ 1236] Overall Loss 0.128355 Objective Loss 0.128355 Top1 89.205703 Top5 98.981670 LR 0.000250 Time 0.022079 +2023-10-02 21:54:47,256 - --- validate (epoch=177)----------- +2023-10-02 21:54:47,256 - 29943 samples (256 per mini-batch) +2023-10-02 21:54:47,757 - Epoch: [177][ 10/ 117] Loss 0.313429 Top1 87.460938 Top5 98.593750 +2023-10-02 21:54:47,907 - Epoch: [177][ 20/ 117] Loss 0.292422 Top1 87.656250 Top5 98.476562 +2023-10-02 21:54:48,056 - Epoch: [177][ 30/ 117] Loss 0.298322 Top1 87.291667 Top5 98.554688 +2023-10-02 21:54:48,206 - Epoch: [177][ 40/ 117] Loss 0.302953 Top1 87.138672 Top5 98.652344 +2023-10-02 21:54:48,354 - Epoch: [177][ 50/ 117] Loss 0.300375 Top1 87.312500 Top5 98.695312 +2023-10-02 21:54:48,504 - Epoch: [177][ 60/ 117] Loss 0.300764 Top1 87.311198 Top5 98.652344 +2023-10-02 21:54:48,652 - Epoch: [177][ 70/ 117] Loss 0.302015 Top1 87.282366 Top5 98.677455 +2023-10-02 21:54:48,800 - Epoch: [177][ 80/ 117] Loss 0.302758 Top1 87.148438 Top5 98.706055 +2023-10-02 21:54:48,947 - Epoch: [177][ 90/ 117] Loss 0.304620 Top1 87.092014 Top5 98.680556 +2023-10-02 21:54:49,094 - Epoch: [177][ 100/ 117] Loss 0.301018 Top1 87.246094 Top5 98.683594 +2023-10-02 21:54:49,249 - Epoch: [177][ 110/ 117] Loss 0.302819 Top1 87.237216 Top5 98.696733 +2023-10-02 21:54:49,339 - Epoch: [177][ 117/ 117] Loss 0.303544 Top1 87.125539 Top5 98.694186 +2023-10-02 21:54:49,482 - ==> Top1: 87.126 Top5: 98.694 Loss: 0.304 + +2023-10-02 21:54:49,482 - ==> Confusion: +[[ 943 0 4 1 3 1 0 2 7 57 3 0 2 2 5 0 1 0 1 0 18] + [ 0 1069 1 1 6 12 0 23 0 1 0 1 0 0 1 3 0 0 8 3 2] + [ 1 0 992 5 0 1 12 7 0 0 2 0 9 2 1 3 2 2 9 2 6] + [ 1 3 12 990 0 0 3 3 3 1 3 1 6 1 20 2 1 5 12 0 22] + [ 25 3 0 0 972 5 0 0 0 14 0 0 0 3 13 2 9 0 0 1 3] + [ 3 40 0 1 2 985 2 29 1 5 2 7 1 7 7 0 3 0 4 3 14] + [ 0 3 27 0 0 1 1136 4 0 0 3 0 0 0 0 3 0 0 3 7 4] + [ 1 12 16 0 4 18 4 1081 1 6 5 4 5 3 1 0 0 1 37 12 7] + [ 20 4 0 1 1 4 0 2 984 31 10 1 2 6 14 1 0 1 3 3 1] + [ 100 1 0 0 8 4 0 0 23 950 0 1 0 13 7 2 0 1 0 0 9] + [ 5 3 9 5 0 1 5 2 10 0 981 2 0 10 7 0 2 1 3 0 7] + [ 0 0 3 0 0 14 0 4 0 0 0 967 11 8 0 1 0 16 0 7 4] + [ 0 0 0 5 0 2 1 2 0 2 5 26 984 0 4 7 0 9 3 6 12] + [ 1 0 1 0 4 8 0 0 14 14 3 9 0 1042 4 0 0 1 0 1 17] + [ 16 0 5 14 1 1 0 0 19 1 1 0 3 2 1018 0 1 2 10 0 7] + [ 0 1 1 3 5 1 0 0 0 0 0 9 6 0 0 1065 17 11 0 12 3] + [ 1 15 1 0 5 7 0 2 1 0 0 4 1 3 2 9 1094 0 1 4 11] + [ 0 0 1 1 0 0 1 0 0 0 0 7 22 1 4 6 1 991 0 1 2] + [ 1 4 4 14 1 0 0 22 4 1 1 0 0 0 5 0 1 0 999 1 10] + [ 0 0 3 1 0 4 6 8 0 0 1 14 2 3 0 0 7 1 0 1095 7] + [ 143 146 139 86 61 96 31 90 86 63 144 88 293 195 111 41 57 49 114 122 5750]] + +2023-10-02 21:54:49,484 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:54:49,484 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:54:49,490 - + +2023-10-02 21:54:49,490 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:54:50,530 - Epoch: [178][ 10/ 1236] Overall Loss 0.119889 Objective Loss 0.119889 LR 0.000250 Time 0.103991 +2023-10-02 21:54:50,740 - Epoch: [178][ 20/ 1236] Overall Loss 0.122913 Objective Loss 0.122913 LR 0.000250 Time 0.062448 +2023-10-02 21:54:50,949 - Epoch: [178][ 30/ 1236] Overall Loss 0.123848 Objective Loss 0.123848 LR 0.000250 Time 0.048583 +2023-10-02 21:54:51,159 - Epoch: [178][ 40/ 1236] Overall Loss 0.120348 Objective Loss 0.120348 LR 0.000250 Time 0.041676 +2023-10-02 21:54:51,368 - Epoch: [178][ 50/ 1236] Overall Loss 0.122465 Objective Loss 0.122465 LR 0.000250 Time 0.037489 +2023-10-02 21:54:51,578 - Epoch: [178][ 60/ 1236] Overall Loss 0.119471 Objective Loss 0.119471 LR 0.000250 Time 0.034736 +2023-10-02 21:54:51,786 - Epoch: [178][ 70/ 1236] Overall Loss 0.120774 Objective Loss 0.120774 LR 0.000250 Time 0.032736 +2023-10-02 21:54:51,996 - Epoch: [178][ 80/ 1236] Overall Loss 0.122537 Objective Loss 0.122537 LR 0.000250 Time 0.031255 +2023-10-02 21:54:52,204 - Epoch: [178][ 90/ 1236] Overall Loss 0.123527 Objective Loss 0.123527 LR 0.000250 Time 0.030094 +2023-10-02 21:54:52,415 - Epoch: [178][ 100/ 1236] Overall Loss 0.122570 Objective Loss 0.122570 LR 0.000250 Time 0.029190 +2023-10-02 21:54:52,624 - Epoch: [178][ 110/ 1236] Overall Loss 0.122482 Objective Loss 0.122482 LR 0.000250 Time 0.028427 +2023-10-02 21:54:52,833 - Epoch: [178][ 120/ 1236] Overall Loss 0.123983 Objective Loss 0.123983 LR 0.000250 Time 0.027798 +2023-10-02 21:54:53,041 - Epoch: [178][ 130/ 1236] Overall Loss 0.124580 Objective Loss 0.124580 LR 0.000250 Time 0.027245 +2023-10-02 21:54:53,250 - Epoch: [178][ 140/ 1236] Overall Loss 0.123586 Objective Loss 0.123586 LR 0.000250 Time 0.026790 +2023-10-02 21:54:53,460 - Epoch: [178][ 150/ 1236] Overall Loss 0.124371 Objective Loss 0.124371 LR 0.000250 Time 0.026394 +2023-10-02 21:54:53,674 - Epoch: [178][ 160/ 1236] Overall Loss 0.124778 Objective Loss 0.124778 LR 0.000250 Time 0.026078 +2023-10-02 21:54:53,884 - Epoch: [178][ 170/ 1236] Overall Loss 0.124751 Objective Loss 0.124751 LR 0.000250 Time 0.025779 +2023-10-02 21:54:54,098 - Epoch: [178][ 180/ 1236] Overall Loss 0.125312 Objective Loss 0.125312 LR 0.000250 Time 0.025533 +2023-10-02 21:54:54,308 - Epoch: [178][ 190/ 1236] Overall Loss 0.125438 Objective Loss 0.125438 LR 0.000250 Time 0.025294 +2023-10-02 21:54:54,522 - Epoch: [178][ 200/ 1236] Overall Loss 0.125687 Objective Loss 0.125687 LR 0.000250 Time 0.025098 +2023-10-02 21:54:54,732 - Epoch: [178][ 210/ 1236] Overall Loss 0.125941 Objective Loss 0.125941 LR 0.000250 Time 0.024902 +2023-10-02 21:54:54,946 - Epoch: [178][ 220/ 1236] Overall Loss 0.126468 Objective Loss 0.126468 LR 0.000250 Time 0.024742 +2023-10-02 21:54:55,156 - Epoch: [178][ 230/ 1236] Overall Loss 0.126287 Objective Loss 0.126287 LR 0.000250 Time 0.024578 +2023-10-02 21:54:55,369 - Epoch: [178][ 240/ 1236] Overall Loss 0.127124 Objective Loss 0.127124 LR 0.000250 Time 0.024439 +2023-10-02 21:54:55,580 - Epoch: [178][ 250/ 1236] Overall Loss 0.126991 Objective Loss 0.126991 LR 0.000250 Time 0.024300 +2023-10-02 21:54:55,794 - Epoch: [178][ 260/ 1236] Overall Loss 0.126906 Objective Loss 0.126906 LR 0.000250 Time 0.024187 +2023-10-02 21:54:56,004 - Epoch: [178][ 270/ 1236] Overall Loss 0.127422 Objective Loss 0.127422 LR 0.000250 Time 0.024067 +2023-10-02 21:54:56,218 - Epoch: [178][ 280/ 1236] Overall Loss 0.128484 Objective Loss 0.128484 LR 0.000250 Time 0.023971 +2023-10-02 21:54:56,428 - Epoch: [178][ 290/ 1236] Overall Loss 0.127795 Objective Loss 0.127795 LR 0.000250 Time 0.023867 +2023-10-02 21:54:56,638 - Epoch: [178][ 300/ 1236] Overall Loss 0.127807 Objective Loss 0.127807 LR 0.000250 Time 0.023771 +2023-10-02 21:54:56,848 - Epoch: [178][ 310/ 1236] Overall Loss 0.127851 Objective Loss 0.127851 LR 0.000250 Time 0.023681 +2023-10-02 21:54:57,058 - Epoch: [178][ 320/ 1236] Overall Loss 0.128017 Objective Loss 0.128017 LR 0.000250 Time 0.023597 +2023-10-02 21:54:57,268 - Epoch: [178][ 330/ 1236] Overall Loss 0.127701 Objective Loss 0.127701 LR 0.000250 Time 0.023513 +2023-10-02 21:54:57,479 - Epoch: [178][ 340/ 1236] Overall Loss 0.127559 Objective Loss 0.127559 LR 0.000250 Time 0.023441 +2023-10-02 21:54:57,690 - Epoch: [178][ 350/ 1236] Overall Loss 0.127293 Objective Loss 0.127293 LR 0.000250 Time 0.023368 +2023-10-02 21:54:57,900 - Epoch: [178][ 360/ 1236] Overall Loss 0.127404 Objective Loss 0.127404 LR 0.000250 Time 0.023302 +2023-10-02 21:54:58,109 - Epoch: [178][ 370/ 1236] Overall Loss 0.126947 Objective Loss 0.126947 LR 0.000250 Time 0.023234 +2023-10-02 21:54:58,320 - Epoch: [178][ 380/ 1236] Overall Loss 0.126691 Objective Loss 0.126691 LR 0.000250 Time 0.023175 +2023-10-02 21:54:58,529 - Epoch: [178][ 390/ 1236] Overall Loss 0.126720 Objective Loss 0.126720 LR 0.000250 Time 0.023114 +2023-10-02 21:54:58,740 - Epoch: [178][ 400/ 1236] Overall Loss 0.126225 Objective Loss 0.126225 LR 0.000250 Time 0.023062 +2023-10-02 21:54:58,949 - Epoch: [178][ 410/ 1236] Overall Loss 0.125695 Objective Loss 0.125695 LR 0.000250 Time 0.023007 +2023-10-02 21:54:59,160 - Epoch: [178][ 420/ 1236] Overall Loss 0.125813 Objective Loss 0.125813 LR 0.000250 Time 0.022959 +2023-10-02 21:54:59,369 - Epoch: [178][ 430/ 1236] Overall Loss 0.126411 Objective Loss 0.126411 LR 0.000250 Time 0.022909 +2023-10-02 21:54:59,580 - Epoch: [178][ 440/ 1236] Overall Loss 0.126220 Objective Loss 0.126220 LR 0.000250 Time 0.022866 +2023-10-02 21:54:59,789 - Epoch: [178][ 450/ 1236] Overall Loss 0.125836 Objective Loss 0.125836 LR 0.000250 Time 0.022820 +2023-10-02 21:55:00,000 - Epoch: [178][ 460/ 1236] Overall Loss 0.125825 Objective Loss 0.125825 LR 0.000250 Time 0.022781 +2023-10-02 21:55:00,209 - Epoch: [178][ 470/ 1236] Overall Loss 0.126442 Objective Loss 0.126442 LR 0.000250 Time 0.022739 +2023-10-02 21:55:00,420 - Epoch: [178][ 480/ 1236] Overall Loss 0.126634 Objective Loss 0.126634 LR 0.000250 Time 0.022703 +2023-10-02 21:55:00,629 - Epoch: [178][ 490/ 1236] Overall Loss 0.126493 Objective Loss 0.126493 LR 0.000250 Time 0.022664 +2023-10-02 21:55:00,840 - Epoch: [178][ 500/ 1236] Overall Loss 0.126730 Objective Loss 0.126730 LR 0.000250 Time 0.022631 +2023-10-02 21:55:01,049 - Epoch: [178][ 510/ 1236] Overall Loss 0.126618 Objective Loss 0.126618 LR 0.000250 Time 0.022595 +2023-10-02 21:55:01,260 - Epoch: [178][ 520/ 1236] Overall Loss 0.126672 Objective Loss 0.126672 LR 0.000250 Time 0.022565 +2023-10-02 21:55:01,469 - Epoch: [178][ 530/ 1236] Overall Loss 0.127005 Objective Loss 0.127005 LR 0.000250 Time 0.022531 +2023-10-02 21:55:01,680 - Epoch: [178][ 540/ 1236] Overall Loss 0.126570 Objective Loss 0.126570 LR 0.000250 Time 0.022504 +2023-10-02 21:55:01,889 - Epoch: [178][ 550/ 1236] Overall Loss 0.126588 Objective Loss 0.126588 LR 0.000250 Time 0.022472 +2023-10-02 21:55:02,099 - Epoch: [178][ 560/ 1236] Overall Loss 0.126741 Objective Loss 0.126741 LR 0.000250 Time 0.022445 +2023-10-02 21:55:02,308 - Epoch: [178][ 570/ 1236] Overall Loss 0.126787 Objective Loss 0.126787 LR 0.000250 Time 0.022418 +2023-10-02 21:55:02,519 - Epoch: [178][ 580/ 1236] Overall Loss 0.126793 Objective Loss 0.126793 LR 0.000250 Time 0.022394 +2023-10-02 21:55:02,728 - Epoch: [178][ 590/ 1236] Overall Loss 0.126863 Objective Loss 0.126863 LR 0.000250 Time 0.022367 +2023-10-02 21:55:02,939 - Epoch: [178][ 600/ 1236] Overall Loss 0.127028 Objective Loss 0.127028 LR 0.000250 Time 0.022344 +2023-10-02 21:55:03,148 - Epoch: [178][ 610/ 1236] Overall Loss 0.127292 Objective Loss 0.127292 LR 0.000250 Time 0.022319 +2023-10-02 21:55:03,359 - Epoch: [178][ 620/ 1236] Overall Loss 0.127143 Objective Loss 0.127143 LR 0.000250 Time 0.022298 +2023-10-02 21:55:03,568 - Epoch: [178][ 630/ 1236] Overall Loss 0.126728 Objective Loss 0.126728 LR 0.000250 Time 0.022276 +2023-10-02 21:55:03,779 - Epoch: [178][ 640/ 1236] Overall Loss 0.126652 Objective Loss 0.126652 LR 0.000250 Time 0.022257 +2023-10-02 21:55:03,988 - Epoch: [178][ 650/ 1236] Overall Loss 0.126893 Objective Loss 0.126893 LR 0.000250 Time 0.022233 +2023-10-02 21:55:04,198 - Epoch: [178][ 660/ 1236] Overall Loss 0.126817 Objective Loss 0.126817 LR 0.000250 Time 0.022214 +2023-10-02 21:55:04,407 - Epoch: [178][ 670/ 1236] Overall Loss 0.126758 Objective Loss 0.126758 LR 0.000250 Time 0.022192 +2023-10-02 21:55:04,617 - Epoch: [178][ 680/ 1236] Overall Loss 0.126590 Objective Loss 0.126590 LR 0.000250 Time 0.022174 +2023-10-02 21:55:04,825 - Epoch: [178][ 690/ 1236] Overall Loss 0.126595 Objective Loss 0.126595 LR 0.000250 Time 0.022153 +2023-10-02 21:55:05,035 - Epoch: [178][ 700/ 1236] Overall Loss 0.126957 Objective Loss 0.126957 LR 0.000250 Time 0.022135 +2023-10-02 21:55:05,244 - Epoch: [178][ 710/ 1236] Overall Loss 0.126825 Objective Loss 0.126825 LR 0.000250 Time 0.022115 +2023-10-02 21:55:05,452 - Epoch: [178][ 720/ 1236] Overall Loss 0.126989 Objective Loss 0.126989 LR 0.000250 Time 0.022098 +2023-10-02 21:55:05,661 - Epoch: [178][ 730/ 1236] Overall Loss 0.126732 Objective Loss 0.126732 LR 0.000250 Time 0.022079 +2023-10-02 21:55:05,871 - Epoch: [178][ 740/ 1236] Overall Loss 0.126591 Objective Loss 0.126591 LR 0.000250 Time 0.022063 +2023-10-02 21:55:06,080 - Epoch: [178][ 750/ 1236] Overall Loss 0.126405 Objective Loss 0.126405 LR 0.000250 Time 0.022045 +2023-10-02 21:55:06,289 - Epoch: [178][ 760/ 1236] Overall Loss 0.126637 Objective Loss 0.126637 LR 0.000250 Time 0.022031 +2023-10-02 21:55:06,498 - Epoch: [178][ 770/ 1236] Overall Loss 0.126551 Objective Loss 0.126551 LR 0.000250 Time 0.022014 +2023-10-02 21:55:06,708 - Epoch: [178][ 780/ 1236] Overall Loss 0.126470 Objective Loss 0.126470 LR 0.000250 Time 0.022001 +2023-10-02 21:55:06,917 - Epoch: [178][ 790/ 1236] Overall Loss 0.126717 Objective Loss 0.126717 LR 0.000250 Time 0.021985 +2023-10-02 21:55:07,127 - Epoch: [178][ 800/ 1236] Overall Loss 0.126530 Objective Loss 0.126530 LR 0.000250 Time 0.021972 +2023-10-02 21:55:07,336 - Epoch: [178][ 810/ 1236] Overall Loss 0.126423 Objective Loss 0.126423 LR 0.000250 Time 0.021957 +2023-10-02 21:55:07,546 - Epoch: [178][ 820/ 1236] Overall Loss 0.126570 Objective Loss 0.126570 LR 0.000250 Time 0.021944 +2023-10-02 21:55:07,755 - Epoch: [178][ 830/ 1236] Overall Loss 0.126569 Objective Loss 0.126569 LR 0.000250 Time 0.021930 +2023-10-02 21:55:07,965 - Epoch: [178][ 840/ 1236] Overall Loss 0.126763 Objective Loss 0.126763 LR 0.000250 Time 0.021918 +2023-10-02 21:55:08,173 - Epoch: [178][ 850/ 1236] Overall Loss 0.126642 Objective Loss 0.126642 LR 0.000250 Time 0.021904 +2023-10-02 21:55:08,383 - Epoch: [178][ 860/ 1236] Overall Loss 0.126458 Objective Loss 0.126458 LR 0.000250 Time 0.021893 +2023-10-02 21:55:08,592 - Epoch: [178][ 870/ 1236] Overall Loss 0.126554 Objective Loss 0.126554 LR 0.000250 Time 0.021879 +2023-10-02 21:55:08,802 - Epoch: [178][ 880/ 1236] Overall Loss 0.126490 Objective Loss 0.126490 LR 0.000250 Time 0.021869 +2023-10-02 21:55:09,011 - Epoch: [178][ 890/ 1236] Overall Loss 0.126503 Objective Loss 0.126503 LR 0.000250 Time 0.021856 +2023-10-02 21:55:09,220 - Epoch: [178][ 900/ 1236] Overall Loss 0.126472 Objective Loss 0.126472 LR 0.000250 Time 0.021846 +2023-10-02 21:55:09,429 - Epoch: [178][ 910/ 1236] Overall Loss 0.126486 Objective Loss 0.126486 LR 0.000250 Time 0.021834 +2023-10-02 21:55:09,639 - Epoch: [178][ 920/ 1236] Overall Loss 0.126404 Objective Loss 0.126404 LR 0.000250 Time 0.021824 +2023-10-02 21:55:09,848 - Epoch: [178][ 930/ 1236] Overall Loss 0.126560 Objective Loss 0.126560 LR 0.000250 Time 0.021812 +2023-10-02 21:55:10,058 - Epoch: [178][ 940/ 1236] Overall Loss 0.126413 Objective Loss 0.126413 LR 0.000250 Time 0.021803 +2023-10-02 21:55:10,267 - Epoch: [178][ 950/ 1236] Overall Loss 0.126425 Objective Loss 0.126425 LR 0.000250 Time 0.021792 +2023-10-02 21:55:10,477 - Epoch: [178][ 960/ 1236] Overall Loss 0.126379 Objective Loss 0.126379 LR 0.000250 Time 0.021783 +2023-10-02 21:55:10,685 - Epoch: [178][ 970/ 1236] Overall Loss 0.126472 Objective Loss 0.126472 LR 0.000250 Time 0.021772 +2023-10-02 21:55:10,895 - Epoch: [178][ 980/ 1236] Overall Loss 0.126373 Objective Loss 0.126373 LR 0.000250 Time 0.021764 +2023-10-02 21:55:11,104 - Epoch: [178][ 990/ 1236] Overall Loss 0.126453 Objective Loss 0.126453 LR 0.000250 Time 0.021753 +2023-10-02 21:55:11,314 - Epoch: [178][ 1000/ 1236] Overall Loss 0.126361 Objective Loss 0.126361 LR 0.000250 Time 0.021745 +2023-10-02 21:55:11,523 - Epoch: [178][ 1010/ 1236] Overall Loss 0.126166 Objective Loss 0.126166 LR 0.000250 Time 0.021735 +2023-10-02 21:55:11,733 - Epoch: [178][ 1020/ 1236] Overall Loss 0.126432 Objective Loss 0.126432 LR 0.000250 Time 0.021728 +2023-10-02 21:55:11,941 - Epoch: [178][ 1030/ 1236] Overall Loss 0.126282 Objective Loss 0.126282 LR 0.000250 Time 0.021718 +2023-10-02 21:55:12,151 - Epoch: [178][ 1040/ 1236] Overall Loss 0.126337 Objective Loss 0.126337 LR 0.000250 Time 0.021710 +2023-10-02 21:55:12,360 - Epoch: [178][ 1050/ 1236] Overall Loss 0.126284 Objective Loss 0.126284 LR 0.000250 Time 0.021701 +2023-10-02 21:55:12,570 - Epoch: [178][ 1060/ 1236] Overall Loss 0.126309 Objective Loss 0.126309 LR 0.000250 Time 0.021694 +2023-10-02 21:55:12,778 - Epoch: [178][ 1070/ 1236] Overall Loss 0.126384 Objective Loss 0.126384 LR 0.000250 Time 0.021685 +2023-10-02 21:55:12,988 - Epoch: [178][ 1080/ 1236] Overall Loss 0.126428 Objective Loss 0.126428 LR 0.000250 Time 0.021678 +2023-10-02 21:55:13,197 - Epoch: [178][ 1090/ 1236] Overall Loss 0.126413 Objective Loss 0.126413 LR 0.000250 Time 0.021669 +2023-10-02 21:55:13,406 - Epoch: [178][ 1100/ 1236] Overall Loss 0.126410 Objective Loss 0.126410 LR 0.000250 Time 0.021662 +2023-10-02 21:55:13,615 - Epoch: [178][ 1110/ 1236] Overall Loss 0.126455 Objective Loss 0.126455 LR 0.000250 Time 0.021654 +2023-10-02 21:55:13,825 - Epoch: [178][ 1120/ 1236] Overall Loss 0.126413 Objective Loss 0.126413 LR 0.000250 Time 0.021648 +2023-10-02 21:55:14,034 - Epoch: [178][ 1130/ 1236] Overall Loss 0.126503 Objective Loss 0.126503 LR 0.000250 Time 0.021640 +2023-10-02 21:55:14,244 - Epoch: [178][ 1140/ 1236] Overall Loss 0.126585 Objective Loss 0.126585 LR 0.000250 Time 0.021634 +2023-10-02 21:55:14,453 - Epoch: [178][ 1150/ 1236] Overall Loss 0.126618 Objective Loss 0.126618 LR 0.000250 Time 0.021626 +2023-10-02 21:55:14,663 - Epoch: [178][ 1160/ 1236] Overall Loss 0.126615 Objective Loss 0.126615 LR 0.000250 Time 0.021620 +2023-10-02 21:55:14,872 - Epoch: [178][ 1170/ 1236] Overall Loss 0.126620 Objective Loss 0.126620 LR 0.000250 Time 0.021613 +2023-10-02 21:55:15,081 - Epoch: [178][ 1180/ 1236] Overall Loss 0.126581 Objective Loss 0.126581 LR 0.000250 Time 0.021607 +2023-10-02 21:55:15,290 - Epoch: [178][ 1190/ 1236] Overall Loss 0.126460 Objective Loss 0.126460 LR 0.000250 Time 0.021600 +2023-10-02 21:55:15,500 - Epoch: [178][ 1200/ 1236] Overall Loss 0.126444 Objective Loss 0.126444 LR 0.000250 Time 0.021594 +2023-10-02 21:55:15,709 - Epoch: [178][ 1210/ 1236] Overall Loss 0.126604 Objective Loss 0.126604 LR 0.000250 Time 0.021587 +2023-10-02 21:55:15,918 - Epoch: [178][ 1220/ 1236] Overall Loss 0.126632 Objective Loss 0.126632 LR 0.000250 Time 0.021582 +2023-10-02 21:55:16,180 - Epoch: [178][ 1230/ 1236] Overall Loss 0.126740 Objective Loss 0.126740 LR 0.000250 Time 0.021618 +2023-10-02 21:55:16,303 - Epoch: [178][ 1236/ 1236] Overall Loss 0.126726 Objective Loss 0.126726 Top1 91.446029 Top5 99.389002 LR 0.000250 Time 0.021612 +2023-10-02 21:55:16,452 - --- validate (epoch=178)----------- +2023-10-02 21:55:16,452 - 29943 samples (256 per mini-batch) +2023-10-02 21:55:16,945 - Epoch: [178][ 10/ 117] Loss 0.291820 Top1 86.484375 Top5 98.867188 +2023-10-02 21:55:17,101 - Epoch: [178][ 20/ 117] Loss 0.276674 Top1 86.875000 Top5 99.042969 +2023-10-02 21:55:17,258 - Epoch: [178][ 30/ 117] Loss 0.294998 Top1 86.848958 Top5 98.854167 +2023-10-02 21:55:17,413 - Epoch: [178][ 40/ 117] Loss 0.299638 Top1 86.777344 Top5 98.681641 +2023-10-02 21:55:17,569 - Epoch: [178][ 50/ 117] Loss 0.301039 Top1 87.007812 Top5 98.679688 +2023-10-02 21:55:17,726 - Epoch: [178][ 60/ 117] Loss 0.310149 Top1 87.005208 Top5 98.645833 +2023-10-02 21:55:17,883 - Epoch: [178][ 70/ 117] Loss 0.308649 Top1 87.109375 Top5 98.616071 +2023-10-02 21:55:18,039 - Epoch: [178][ 80/ 117] Loss 0.305838 Top1 87.236328 Top5 98.657227 +2023-10-02 21:55:18,194 - Epoch: [178][ 90/ 117] Loss 0.304873 Top1 87.309028 Top5 98.650174 +2023-10-02 21:55:18,351 - Epoch: [178][ 100/ 117] Loss 0.300492 Top1 87.375000 Top5 98.691406 +2023-10-02 21:55:18,513 - Epoch: [178][ 110/ 117] Loss 0.303969 Top1 87.251420 Top5 98.664773 +2023-10-02 21:55:18,602 - Epoch: [178][ 117/ 117] Loss 0.302124 Top1 87.282503 Top5 98.677487 +2023-10-02 21:55:18,740 - ==> Top1: 87.283 Top5: 98.677 Loss: 0.302 + +2023-10-02 21:55:18,741 - ==> Confusion: +[[ 935 0 5 1 6 3 0 0 9 55 2 1 1 1 5 1 1 0 2 0 22] + [ 1 1066 2 1 6 13 1 18 3 1 1 0 1 0 1 3 0 0 5 1 7] + [ 1 1 986 7 0 0 16 7 0 0 1 0 10 2 3 3 2 2 8 2 5] + [ 0 1 11 993 0 1 0 3 3 1 3 0 6 3 22 3 3 4 12 2 18] + [ 22 6 1 1 976 3 0 1 2 10 2 0 0 3 7 4 6 0 0 0 6] + [ 2 40 0 1 5 988 2 21 1 6 2 9 0 8 5 1 2 2 4 0 17] + [ 0 4 26 1 0 2 1133 6 0 0 3 1 0 0 0 4 0 0 1 6 4] + [ 0 13 14 1 8 19 6 1067 1 2 5 4 4 4 0 1 0 1 46 12 10] + [ 15 2 0 1 1 3 0 1 989 32 12 2 2 10 11 0 4 1 1 0 2] + [ 82 0 0 0 5 4 0 0 40 939 0 1 0 27 6 4 1 0 0 0 10] + [ 4 0 8 7 0 1 3 1 9 1 973 1 1 14 9 2 1 1 4 3 10] + [ 0 0 2 0 1 10 0 5 0 0 0 973 10 8 0 1 0 17 0 4 4] + [ 0 1 3 1 1 1 0 1 0 2 3 37 972 1 3 11 1 10 1 4 15] + [ 1 0 0 0 3 5 0 0 13 7 4 5 0 1057 4 0 0 1 0 1 18] + [ 13 0 4 18 3 0 0 0 25 1 1 0 2 1 1018 0 1 2 6 0 6] + [ 0 0 2 1 5 1 3 0 0 0 0 6 8 0 0 1068 17 12 1 5 5] + [ 0 15 1 0 5 4 0 0 1 0 0 4 1 1 5 8 1095 0 2 6 13] + [ 0 0 0 3 1 0 3 0 0 1 0 4 16 1 3 7 0 994 0 1 4] + [ 1 3 5 16 0 1 0 19 5 2 2 1 1 0 12 0 0 0 989 0 11] + [ 0 1 4 2 1 2 3 7 0 0 2 17 3 1 2 2 7 1 0 1092 5] + [ 90 145 130 83 55 105 32 52 82 49 152 93 272 245 105 41 79 56 96 111 5832]] + +2023-10-02 21:55:18,743 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:55:18,743 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:55:18,749 - + +2023-10-02 21:55:18,749 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:55:19,885 - Epoch: [179][ 10/ 1236] Overall Loss 0.114920 Objective Loss 0.114920 LR 0.000250 Time 0.113614 +2023-10-02 21:55:20,094 - Epoch: [179][ 20/ 1236] Overall Loss 0.122611 Objective Loss 0.122611 LR 0.000250 Time 0.067215 +2023-10-02 21:55:20,303 - Epoch: [179][ 30/ 1236] Overall Loss 0.121833 Objective Loss 0.121833 LR 0.000250 Time 0.051763 +2023-10-02 21:55:20,513 - Epoch: [179][ 40/ 1236] Overall Loss 0.119486 Objective Loss 0.119486 LR 0.000250 Time 0.044059 +2023-10-02 21:55:20,721 - Epoch: [179][ 50/ 1236] Overall Loss 0.120939 Objective Loss 0.120939 LR 0.000250 Time 0.039389 +2023-10-02 21:55:20,931 - Epoch: [179][ 60/ 1236] Overall Loss 0.124164 Objective Loss 0.124164 LR 0.000250 Time 0.036310 +2023-10-02 21:55:21,140 - Epoch: [179][ 70/ 1236] Overall Loss 0.125294 Objective Loss 0.125294 LR 0.000250 Time 0.034084 +2023-10-02 21:55:21,349 - Epoch: [179][ 80/ 1236] Overall Loss 0.121504 Objective Loss 0.121504 LR 0.000250 Time 0.032432 +2023-10-02 21:55:21,556 - Epoch: [179][ 90/ 1236] Overall Loss 0.120212 Objective Loss 0.120212 LR 0.000250 Time 0.031120 +2023-10-02 21:55:21,765 - Epoch: [179][ 100/ 1236] Overall Loss 0.121564 Objective Loss 0.121564 LR 0.000250 Time 0.030091 +2023-10-02 21:55:21,972 - Epoch: [179][ 110/ 1236] Overall Loss 0.121639 Objective Loss 0.121639 LR 0.000250 Time 0.029227 +2023-10-02 21:55:22,181 - Epoch: [179][ 120/ 1236] Overall Loss 0.122595 Objective Loss 0.122595 LR 0.000250 Time 0.028529 +2023-10-02 21:55:22,388 - Epoch: [179][ 130/ 1236] Overall Loss 0.123232 Objective Loss 0.123232 LR 0.000250 Time 0.027916 +2023-10-02 21:55:22,598 - Epoch: [179][ 140/ 1236] Overall Loss 0.123654 Objective Loss 0.123654 LR 0.000250 Time 0.027414 +2023-10-02 21:55:22,805 - Epoch: [179][ 150/ 1236] Overall Loss 0.123046 Objective Loss 0.123046 LR 0.000250 Time 0.026958 +2023-10-02 21:55:23,015 - Epoch: [179][ 160/ 1236] Overall Loss 0.124004 Objective Loss 0.124004 LR 0.000250 Time 0.026583 +2023-10-02 21:55:23,222 - Epoch: [179][ 170/ 1236] Overall Loss 0.123936 Objective Loss 0.123936 LR 0.000250 Time 0.026228 +2023-10-02 21:55:23,431 - Epoch: [179][ 180/ 1236] Overall Loss 0.124287 Objective Loss 0.124287 LR 0.000250 Time 0.025929 +2023-10-02 21:55:23,638 - Epoch: [179][ 190/ 1236] Overall Loss 0.124166 Objective Loss 0.124166 LR 0.000250 Time 0.025657 +2023-10-02 21:55:23,845 - Epoch: [179][ 200/ 1236] Overall Loss 0.123681 Objective Loss 0.123681 LR 0.000250 Time 0.025406 +2023-10-02 21:55:24,053 - Epoch: [179][ 210/ 1236] Overall Loss 0.124091 Objective Loss 0.124091 LR 0.000250 Time 0.025184 +2023-10-02 21:55:24,262 - Epoch: [179][ 220/ 1236] Overall Loss 0.124726 Objective Loss 0.124726 LR 0.000250 Time 0.024988 +2023-10-02 21:55:24,471 - Epoch: [179][ 230/ 1236] Overall Loss 0.124533 Objective Loss 0.124533 LR 0.000250 Time 0.024808 +2023-10-02 21:55:24,680 - Epoch: [179][ 240/ 1236] Overall Loss 0.124411 Objective Loss 0.124411 LR 0.000250 Time 0.024643 +2023-10-02 21:55:24,887 - Epoch: [179][ 250/ 1236] Overall Loss 0.124082 Objective Loss 0.124082 LR 0.000250 Time 0.024481 +2023-10-02 21:55:25,097 - Epoch: [179][ 260/ 1236] Overall Loss 0.123650 Objective Loss 0.123650 LR 0.000250 Time 0.024345 +2023-10-02 21:55:25,304 - Epoch: [179][ 270/ 1236] Overall Loss 0.123829 Objective Loss 0.123829 LR 0.000250 Time 0.024206 +2023-10-02 21:55:25,513 - Epoch: [179][ 280/ 1236] Overall Loss 0.123388 Objective Loss 0.123388 LR 0.000250 Time 0.024086 +2023-10-02 21:55:25,720 - Epoch: [179][ 290/ 1236] Overall Loss 0.124463 Objective Loss 0.124463 LR 0.000250 Time 0.023965 +2023-10-02 21:55:25,929 - Epoch: [179][ 300/ 1236] Overall Loss 0.124250 Objective Loss 0.124250 LR 0.000250 Time 0.023862 +2023-10-02 21:55:26,137 - Epoch: [179][ 310/ 1236] Overall Loss 0.124237 Objective Loss 0.124237 LR 0.000250 Time 0.023763 +2023-10-02 21:55:26,347 - Epoch: [179][ 320/ 1236] Overall Loss 0.124348 Objective Loss 0.124348 LR 0.000250 Time 0.023675 +2023-10-02 21:55:26,556 - Epoch: [179][ 330/ 1236] Overall Loss 0.124382 Objective Loss 0.124382 LR 0.000250 Time 0.023590 +2023-10-02 21:55:26,766 - Epoch: [179][ 340/ 1236] Overall Loss 0.124744 Objective Loss 0.124744 LR 0.000250 Time 0.023512 +2023-10-02 21:55:26,976 - Epoch: [179][ 350/ 1236] Overall Loss 0.124514 Objective Loss 0.124514 LR 0.000250 Time 0.023436 +2023-10-02 21:55:27,187 - Epoch: [179][ 360/ 1236] Overall Loss 0.125085 Objective Loss 0.125085 LR 0.000250 Time 0.023370 +2023-10-02 21:55:27,397 - Epoch: [179][ 370/ 1236] Overall Loss 0.125203 Objective Loss 0.125203 LR 0.000250 Time 0.023305 +2023-10-02 21:55:27,609 - Epoch: [179][ 380/ 1236] Overall Loss 0.125683 Objective Loss 0.125683 LR 0.000250 Time 0.023250 +2023-10-02 21:55:27,818 - Epoch: [179][ 390/ 1236] Overall Loss 0.125877 Objective Loss 0.125877 LR 0.000250 Time 0.023187 +2023-10-02 21:55:28,027 - Epoch: [179][ 400/ 1236] Overall Loss 0.125816 Objective Loss 0.125816 LR 0.000250 Time 0.023130 +2023-10-02 21:55:28,236 - Epoch: [179][ 410/ 1236] Overall Loss 0.126117 Objective Loss 0.126117 LR 0.000250 Time 0.023075 +2023-10-02 21:55:28,445 - Epoch: [179][ 420/ 1236] Overall Loss 0.126021 Objective Loss 0.126021 LR 0.000250 Time 0.023024 +2023-10-02 21:55:28,659 - Epoch: [179][ 430/ 1236] Overall Loss 0.126158 Objective Loss 0.126158 LR 0.000250 Time 0.022983 +2023-10-02 21:55:28,871 - Epoch: [179][ 440/ 1236] Overall Loss 0.126513 Objective Loss 0.126513 LR 0.000250 Time 0.022942 +2023-10-02 21:55:29,084 - Epoch: [179][ 450/ 1236] Overall Loss 0.126712 Objective Loss 0.126712 LR 0.000250 Time 0.022905 +2023-10-02 21:55:29,296 - Epoch: [179][ 460/ 1236] Overall Loss 0.126937 Objective Loss 0.126937 LR 0.000250 Time 0.022868 +2023-10-02 21:55:29,510 - Epoch: [179][ 470/ 1236] Overall Loss 0.126975 Objective Loss 0.126975 LR 0.000250 Time 0.022835 +2023-10-02 21:55:29,722 - Epoch: [179][ 480/ 1236] Overall Loss 0.127333 Objective Loss 0.127333 LR 0.000250 Time 0.022800 +2023-10-02 21:55:29,935 - Epoch: [179][ 490/ 1236] Overall Loss 0.127413 Objective Loss 0.127413 LR 0.000250 Time 0.022770 +2023-10-02 21:55:30,148 - Epoch: [179][ 500/ 1236] Overall Loss 0.127435 Objective Loss 0.127435 LR 0.000250 Time 0.022738 +2023-10-02 21:55:30,361 - Epoch: [179][ 510/ 1236] Overall Loss 0.127234 Objective Loss 0.127234 LR 0.000250 Time 0.022710 +2023-10-02 21:55:30,573 - Epoch: [179][ 520/ 1236] Overall Loss 0.127452 Objective Loss 0.127452 LR 0.000250 Time 0.022680 +2023-10-02 21:55:30,787 - Epoch: [179][ 530/ 1236] Overall Loss 0.126893 Objective Loss 0.126893 LR 0.000250 Time 0.022655 +2023-10-02 21:55:31,005 - Epoch: [179][ 540/ 1236] Overall Loss 0.126865 Objective Loss 0.126865 LR 0.000250 Time 0.022639 +2023-10-02 21:55:31,221 - Epoch: [179][ 550/ 1236] Overall Loss 0.127101 Objective Loss 0.127101 LR 0.000250 Time 0.022619 +2023-10-02 21:55:31,441 - Epoch: [179][ 560/ 1236] Overall Loss 0.127200 Objective Loss 0.127200 LR 0.000250 Time 0.022608 +2023-10-02 21:55:31,657 - Epoch: [179][ 570/ 1236] Overall Loss 0.126839 Objective Loss 0.126839 LR 0.000250 Time 0.022589 +2023-10-02 21:55:31,877 - Epoch: [179][ 580/ 1236] Overall Loss 0.127002 Objective Loss 0.127002 LR 0.000250 Time 0.022579 +2023-10-02 21:55:32,093 - Epoch: [179][ 590/ 1236] Overall Loss 0.127115 Objective Loss 0.127115 LR 0.000250 Time 0.022560 +2023-10-02 21:55:32,306 - Epoch: [179][ 600/ 1236] Overall Loss 0.127521 Objective Loss 0.127521 LR 0.000250 Time 0.022539 +2023-10-02 21:55:32,515 - Epoch: [179][ 610/ 1236] Overall Loss 0.127222 Objective Loss 0.127222 LR 0.000250 Time 0.022511 +2023-10-02 21:55:32,723 - Epoch: [179][ 620/ 1236] Overall Loss 0.127215 Objective Loss 0.127215 LR 0.000250 Time 0.022483 +2023-10-02 21:55:32,928 - Epoch: [179][ 630/ 1236] Overall Loss 0.127344 Objective Loss 0.127344 LR 0.000250 Time 0.022452 +2023-10-02 21:55:33,136 - Epoch: [179][ 640/ 1236] Overall Loss 0.127564 Objective Loss 0.127564 LR 0.000250 Time 0.022425 +2023-10-02 21:55:33,343 - Epoch: [179][ 650/ 1236] Overall Loss 0.127832 Objective Loss 0.127832 LR 0.000250 Time 0.022398 +2023-10-02 21:55:33,551 - Epoch: [179][ 660/ 1236] Overall Loss 0.128080 Objective Loss 0.128080 LR 0.000250 Time 0.022374 +2023-10-02 21:55:33,758 - Epoch: [179][ 670/ 1236] Overall Loss 0.128172 Objective Loss 0.128172 LR 0.000250 Time 0.022346 +2023-10-02 21:55:33,966 - Epoch: [179][ 680/ 1236] Overall Loss 0.128235 Objective Loss 0.128235 LR 0.000250 Time 0.022323 +2023-10-02 21:55:34,173 - Epoch: [179][ 690/ 1236] Overall Loss 0.128123 Objective Loss 0.128123 LR 0.000250 Time 0.022299 +2023-10-02 21:55:34,381 - Epoch: [179][ 700/ 1236] Overall Loss 0.127889 Objective Loss 0.127889 LR 0.000250 Time 0.022277 +2023-10-02 21:55:34,588 - Epoch: [179][ 710/ 1236] Overall Loss 0.128080 Objective Loss 0.128080 LR 0.000250 Time 0.022254 +2023-10-02 21:55:34,796 - Epoch: [179][ 720/ 1236] Overall Loss 0.128110 Objective Loss 0.128110 LR 0.000250 Time 0.022234 +2023-10-02 21:55:35,002 - Epoch: [179][ 730/ 1236] Overall Loss 0.127786 Objective Loss 0.127786 LR 0.000250 Time 0.022211 +2023-10-02 21:55:35,210 - Epoch: [179][ 740/ 1236] Overall Loss 0.127713 Objective Loss 0.127713 LR 0.000250 Time 0.022191 +2023-10-02 21:55:35,417 - Epoch: [179][ 750/ 1236] Overall Loss 0.127492 Objective Loss 0.127492 LR 0.000250 Time 0.022171 +2023-10-02 21:55:35,625 - Epoch: [179][ 760/ 1236] Overall Loss 0.127496 Objective Loss 0.127496 LR 0.000250 Time 0.022153 +2023-10-02 21:55:35,833 - Epoch: [179][ 770/ 1236] Overall Loss 0.127507 Objective Loss 0.127507 LR 0.000250 Time 0.022133 +2023-10-02 21:55:36,041 - Epoch: [179][ 780/ 1236] Overall Loss 0.127478 Objective Loss 0.127478 LR 0.000250 Time 0.022115 +2023-10-02 21:55:36,248 - Epoch: [179][ 790/ 1236] Overall Loss 0.127560 Objective Loss 0.127560 LR 0.000250 Time 0.022096 +2023-10-02 21:55:36,456 - Epoch: [179][ 800/ 1236] Overall Loss 0.127362 Objective Loss 0.127362 LR 0.000250 Time 0.022079 +2023-10-02 21:55:36,662 - Epoch: [179][ 810/ 1236] Overall Loss 0.127286 Objective Loss 0.127286 LR 0.000250 Time 0.022058 +2023-10-02 21:55:36,870 - Epoch: [179][ 820/ 1236] Overall Loss 0.127265 Objective Loss 0.127265 LR 0.000250 Time 0.022043 +2023-10-02 21:55:37,077 - Epoch: [179][ 830/ 1236] Overall Loss 0.127020 Objective Loss 0.127020 LR 0.000250 Time 0.022026 +2023-10-02 21:55:37,285 - Epoch: [179][ 840/ 1236] Overall Loss 0.126910 Objective Loss 0.126910 LR 0.000250 Time 0.022012 +2023-10-02 21:55:37,493 - Epoch: [179][ 850/ 1236] Overall Loss 0.126707 Objective Loss 0.126707 LR 0.000250 Time 0.021995 +2023-10-02 21:55:37,699 - Epoch: [179][ 860/ 1236] Overall Loss 0.126403 Objective Loss 0.126403 LR 0.000250 Time 0.021979 +2023-10-02 21:55:37,906 - Epoch: [179][ 870/ 1236] Overall Loss 0.126265 Objective Loss 0.126265 LR 0.000250 Time 0.021962 +2023-10-02 21:55:38,114 - Epoch: [179][ 880/ 1236] Overall Loss 0.126226 Objective Loss 0.126226 LR 0.000250 Time 0.021949 +2023-10-02 21:55:38,320 - Epoch: [179][ 890/ 1236] Overall Loss 0.126227 Objective Loss 0.126227 LR 0.000250 Time 0.021934 +2023-10-02 21:55:38,529 - Epoch: [179][ 900/ 1236] Overall Loss 0.126102 Objective Loss 0.126102 LR 0.000250 Time 0.021921 +2023-10-02 21:55:38,736 - Epoch: [179][ 910/ 1236] Overall Loss 0.126037 Objective Loss 0.126037 LR 0.000250 Time 0.021906 +2023-10-02 21:55:38,943 - Epoch: [179][ 920/ 1236] Overall Loss 0.125864 Objective Loss 0.125864 LR 0.000250 Time 0.021893 +2023-10-02 21:55:39,150 - Epoch: [179][ 930/ 1236] Overall Loss 0.125835 Objective Loss 0.125835 LR 0.000250 Time 0.021878 +2023-10-02 21:55:39,358 - Epoch: [179][ 940/ 1236] Overall Loss 0.125745 Objective Loss 0.125745 LR 0.000250 Time 0.021866 +2023-10-02 21:55:39,565 - Epoch: [179][ 950/ 1236] Overall Loss 0.125581 Objective Loss 0.125581 LR 0.000250 Time 0.021854 +2023-10-02 21:55:39,773 - Epoch: [179][ 960/ 1236] Overall Loss 0.125427 Objective Loss 0.125427 LR 0.000250 Time 0.021842 +2023-10-02 21:55:39,981 - Epoch: [179][ 970/ 1236] Overall Loss 0.125657 Objective Loss 0.125657 LR 0.000250 Time 0.021831 +2023-10-02 21:55:40,189 - Epoch: [179][ 980/ 1236] Overall Loss 0.125685 Objective Loss 0.125685 LR 0.000250 Time 0.021820 +2023-10-02 21:55:40,396 - Epoch: [179][ 990/ 1236] Overall Loss 0.125539 Objective Loss 0.125539 LR 0.000250 Time 0.021807 +2023-10-02 21:55:40,604 - Epoch: [179][ 1000/ 1236] Overall Loss 0.125326 Objective Loss 0.125326 LR 0.000250 Time 0.021797 +2023-10-02 21:55:40,810 - Epoch: [179][ 1010/ 1236] Overall Loss 0.125289 Objective Loss 0.125289 LR 0.000250 Time 0.021784 +2023-10-02 21:55:41,018 - Epoch: [179][ 1020/ 1236] Overall Loss 0.125318 Objective Loss 0.125318 LR 0.000250 Time 0.021773 +2023-10-02 21:55:41,225 - Epoch: [179][ 1030/ 1236] Overall Loss 0.125145 Objective Loss 0.125145 LR 0.000250 Time 0.021762 +2023-10-02 21:55:41,433 - Epoch: [179][ 1040/ 1236] Overall Loss 0.125159 Objective Loss 0.125159 LR 0.000250 Time 0.021752 +2023-10-02 21:55:41,640 - Epoch: [179][ 1050/ 1236] Overall Loss 0.125262 Objective Loss 0.125262 LR 0.000250 Time 0.021742 +2023-10-02 21:55:41,848 - Epoch: [179][ 1060/ 1236] Overall Loss 0.125393 Objective Loss 0.125393 LR 0.000250 Time 0.021733 +2023-10-02 21:55:42,054 - Epoch: [179][ 1070/ 1236] Overall Loss 0.125371 Objective Loss 0.125371 LR 0.000250 Time 0.021722 +2023-10-02 21:55:42,262 - Epoch: [179][ 1080/ 1236] Overall Loss 0.125409 Objective Loss 0.125409 LR 0.000250 Time 0.021713 +2023-10-02 21:55:42,468 - Epoch: [179][ 1090/ 1236] Overall Loss 0.125483 Objective Loss 0.125483 LR 0.000250 Time 0.021703 +2023-10-02 21:55:42,677 - Epoch: [179][ 1100/ 1236] Overall Loss 0.125524 Objective Loss 0.125524 LR 0.000250 Time 0.021694 +2023-10-02 21:55:42,883 - Epoch: [179][ 1110/ 1236] Overall Loss 0.125439 Objective Loss 0.125439 LR 0.000250 Time 0.021684 +2023-10-02 21:55:43,090 - Epoch: [179][ 1120/ 1236] Overall Loss 0.125383 Objective Loss 0.125383 LR 0.000250 Time 0.021675 +2023-10-02 21:55:43,297 - Epoch: [179][ 1130/ 1236] Overall Loss 0.125303 Objective Loss 0.125303 LR 0.000250 Time 0.021667 +2023-10-02 21:55:43,506 - Epoch: [179][ 1140/ 1236] Overall Loss 0.125192 Objective Loss 0.125192 LR 0.000250 Time 0.021659 +2023-10-02 21:55:43,713 - Epoch: [179][ 1150/ 1236] Overall Loss 0.125153 Objective Loss 0.125153 LR 0.000250 Time 0.021650 +2023-10-02 21:55:43,921 - Epoch: [179][ 1160/ 1236] Overall Loss 0.125090 Objective Loss 0.125090 LR 0.000250 Time 0.021642 +2023-10-02 21:55:44,129 - Epoch: [179][ 1170/ 1236] Overall Loss 0.125115 Objective Loss 0.125115 LR 0.000250 Time 0.021634 +2023-10-02 21:55:44,337 - Epoch: [179][ 1180/ 1236] Overall Loss 0.125177 Objective Loss 0.125177 LR 0.000250 Time 0.021627 +2023-10-02 21:55:44,545 - Epoch: [179][ 1190/ 1236] Overall Loss 0.125218 Objective Loss 0.125218 LR 0.000250 Time 0.021619 +2023-10-02 21:55:44,752 - Epoch: [179][ 1200/ 1236] Overall Loss 0.125332 Objective Loss 0.125332 LR 0.000250 Time 0.021610 +2023-10-02 21:55:44,960 - Epoch: [179][ 1210/ 1236] Overall Loss 0.125452 Objective Loss 0.125452 LR 0.000250 Time 0.021603 +2023-10-02 21:55:45,165 - Epoch: [179][ 1220/ 1236] Overall Loss 0.125421 Objective Loss 0.125421 LR 0.000250 Time 0.021593 +2023-10-02 21:55:45,424 - Epoch: [179][ 1230/ 1236] Overall Loss 0.125409 Objective Loss 0.125409 LR 0.000250 Time 0.021627 +2023-10-02 21:55:45,545 - Epoch: [179][ 1236/ 1236] Overall Loss 0.125393 Objective Loss 0.125393 Top1 92.464358 Top5 99.185336 LR 0.000250 Time 0.021620 +2023-10-02 21:55:45,691 - --- validate (epoch=179)----------- +2023-10-02 21:55:45,691 - 29943 samples (256 per mini-batch) +2023-10-02 21:55:46,194 - Epoch: [179][ 10/ 117] Loss 0.333692 Top1 87.343750 Top5 98.632812 +2023-10-02 21:55:46,354 - Epoch: [179][ 20/ 117] Loss 0.304354 Top1 87.773438 Top5 98.632812 +2023-10-02 21:55:46,508 - Epoch: [179][ 30/ 117] Loss 0.293222 Top1 87.929688 Top5 98.776042 +2023-10-02 21:55:46,664 - Epoch: [179][ 40/ 117] Loss 0.290297 Top1 87.968750 Top5 98.798828 +2023-10-02 21:55:46,817 - Epoch: [179][ 50/ 117] Loss 0.294951 Top1 87.726562 Top5 98.765625 +2023-10-02 21:55:46,973 - Epoch: [179][ 60/ 117] Loss 0.301442 Top1 87.643229 Top5 98.736979 +2023-10-02 21:55:47,127 - Epoch: [179][ 70/ 117] Loss 0.304939 Top1 87.572545 Top5 98.710938 +2023-10-02 21:55:47,283 - Epoch: [179][ 80/ 117] Loss 0.302172 Top1 87.578125 Top5 98.740234 +2023-10-02 21:55:47,437 - Epoch: [179][ 90/ 117] Loss 0.310039 Top1 87.473958 Top5 98.715278 +2023-10-02 21:55:47,593 - Epoch: [179][ 100/ 117] Loss 0.314297 Top1 87.347656 Top5 98.683594 +2023-10-02 21:55:47,754 - Epoch: [179][ 110/ 117] Loss 0.313732 Top1 87.283381 Top5 98.668324 +2023-10-02 21:55:47,842 - Epoch: [179][ 117/ 117] Loss 0.313265 Top1 87.272484 Top5 98.657449 +2023-10-02 21:55:47,976 - ==> Top1: 87.272 Top5: 98.657 Loss: 0.313 + +2023-10-02 21:55:47,977 - ==> Confusion: +[[ 952 1 4 0 2 2 0 0 7 48 1 1 1 4 5 1 0 0 1 0 20] + [ 0 1062 0 0 3 19 0 24 2 1 1 0 1 0 1 3 1 0 4 1 8] + [ 1 1 985 3 0 0 19 6 0 2 1 2 4 2 1 5 2 2 11 1 8] + [ 1 4 20 967 1 1 2 4 2 1 1 0 3 1 28 3 2 4 15 2 27] + [ 33 4 1 0 965 5 1 0 0 12 0 0 0 2 9 5 11 0 0 1 1] + [ 2 40 0 1 7 986 2 27 1 4 0 6 1 7 3 1 4 1 4 2 17] + [ 0 6 25 0 0 0 1128 6 0 0 3 1 0 0 1 3 0 1 1 9 7] + [ 2 12 10 0 6 19 5 1083 1 2 5 3 5 3 1 1 0 1 40 10 9] + [ 15 4 0 1 1 4 0 1 971 33 11 2 1 11 16 3 3 2 4 2 4] + [ 113 0 0 1 6 3 0 0 30 919 1 1 0 25 5 3 1 0 0 1 10] + [ 2 1 12 5 0 1 2 6 9 1 974 2 0 10 4 1 0 1 6 4 12] + [ 0 0 1 0 1 12 0 4 0 0 0 975 12 6 0 2 1 13 0 4 4] + [ 0 1 0 2 0 0 1 0 0 2 5 40 968 1 1 10 1 12 2 3 19] + [ 2 0 0 0 4 7 0 0 8 8 3 11 0 1052 3 1 0 1 0 1 18] + [ 10 0 5 8 4 0 1 0 24 2 1 0 2 2 1023 0 1 2 9 0 7] + [ 0 1 2 1 3 1 0 0 0 1 1 5 8 0 0 1072 15 10 1 8 5] + [ 1 15 0 0 5 6 1 0 1 0 0 4 0 1 4 8 1095 0 2 6 12] + [ 0 0 1 0 1 1 3 0 0 0 0 7 21 1 1 7 0 991 0 0 4] + [ 2 3 4 12 0 0 0 23 5 1 0 0 1 0 11 0 0 1 993 0 12] + [ 0 2 4 2 1 3 4 10 0 2 2 14 3 1 0 0 7 1 0 1093 3] + [ 116 134 110 61 62 118 28 99 60 51 130 80 261 229 102 48 79 50 80 129 5878]] + +2023-10-02 21:55:47,978 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:55:47,978 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:55:47,984 - + +2023-10-02 21:55:47,984 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:55:49,005 - Epoch: [180][ 10/ 1236] Overall Loss 0.123297 Objective Loss 0.123297 LR 0.000125 Time 0.101983 +2023-10-02 21:55:49,211 - Epoch: [180][ 20/ 1236] Overall Loss 0.117263 Objective Loss 0.117263 LR 0.000125 Time 0.061265 +2023-10-02 21:55:49,416 - Epoch: [180][ 30/ 1236] Overall Loss 0.119879 Objective Loss 0.119879 LR 0.000125 Time 0.047647 +2023-10-02 21:55:49,624 - Epoch: [180][ 40/ 1236] Overall Loss 0.117814 Objective Loss 0.117814 LR 0.000125 Time 0.040913 +2023-10-02 21:55:49,828 - Epoch: [180][ 50/ 1236] Overall Loss 0.123465 Objective Loss 0.123465 LR 0.000125 Time 0.036813 +2023-10-02 21:55:50,034 - Epoch: [180][ 60/ 1236] Overall Loss 0.122865 Objective Loss 0.122865 LR 0.000125 Time 0.034105 +2023-10-02 21:55:50,240 - Epoch: [180][ 70/ 1236] Overall Loss 0.126941 Objective Loss 0.126941 LR 0.000125 Time 0.032153 +2023-10-02 21:55:50,447 - Epoch: [180][ 80/ 1236] Overall Loss 0.124658 Objective Loss 0.124658 LR 0.000125 Time 0.030721 +2023-10-02 21:55:50,651 - Epoch: [180][ 90/ 1236] Overall Loss 0.126987 Objective Loss 0.126987 LR 0.000125 Time 0.029571 +2023-10-02 21:55:50,856 - Epoch: [180][ 100/ 1236] Overall Loss 0.125618 Objective Loss 0.125618 LR 0.000125 Time 0.028664 +2023-10-02 21:55:51,062 - Epoch: [180][ 110/ 1236] Overall Loss 0.124994 Objective Loss 0.124994 LR 0.000125 Time 0.027924 +2023-10-02 21:55:51,269 - Epoch: [180][ 120/ 1236] Overall Loss 0.125340 Objective Loss 0.125340 LR 0.000125 Time 0.027321 +2023-10-02 21:55:51,474 - Epoch: [180][ 130/ 1236] Overall Loss 0.126167 Objective Loss 0.126167 LR 0.000125 Time 0.026791 +2023-10-02 21:55:51,679 - Epoch: [180][ 140/ 1236] Overall Loss 0.124969 Objective Loss 0.124969 LR 0.000125 Time 0.026339 +2023-10-02 21:55:51,884 - Epoch: [180][ 150/ 1236] Overall Loss 0.125114 Objective Loss 0.125114 LR 0.000125 Time 0.025949 +2023-10-02 21:55:52,090 - Epoch: [180][ 160/ 1236] Overall Loss 0.124400 Objective Loss 0.124400 LR 0.000125 Time 0.025617 +2023-10-02 21:55:52,295 - Epoch: [180][ 170/ 1236] Overall Loss 0.123875 Objective Loss 0.123875 LR 0.000125 Time 0.025309 +2023-10-02 21:55:52,501 - Epoch: [180][ 180/ 1236] Overall Loss 0.123116 Objective Loss 0.123116 LR 0.000125 Time 0.025049 +2023-10-02 21:55:52,706 - Epoch: [180][ 190/ 1236] Overall Loss 0.122543 Objective Loss 0.122543 LR 0.000125 Time 0.024806 +2023-10-02 21:55:52,909 - Epoch: [180][ 200/ 1236] Overall Loss 0.122366 Objective Loss 0.122366 LR 0.000125 Time 0.024583 +2023-10-02 21:55:53,115 - Epoch: [180][ 210/ 1236] Overall Loss 0.121905 Objective Loss 0.121905 LR 0.000125 Time 0.024388 +2023-10-02 21:55:53,322 - Epoch: [180][ 220/ 1236] Overall Loss 0.121455 Objective Loss 0.121455 LR 0.000125 Time 0.024222 +2023-10-02 21:55:53,527 - Epoch: [180][ 230/ 1236] Overall Loss 0.121526 Objective Loss 0.121526 LR 0.000125 Time 0.024057 +2023-10-02 21:55:53,735 - Epoch: [180][ 240/ 1236] Overall Loss 0.121191 Objective Loss 0.121191 LR 0.000125 Time 0.023919 +2023-10-02 21:55:53,939 - Epoch: [180][ 250/ 1236] Overall Loss 0.120418 Objective Loss 0.120418 LR 0.000125 Time 0.023778 +2023-10-02 21:55:54,146 - Epoch: [180][ 260/ 1236] Overall Loss 0.119719 Objective Loss 0.119719 LR 0.000125 Time 0.023658 +2023-10-02 21:55:54,350 - Epoch: [180][ 270/ 1236] Overall Loss 0.119276 Objective Loss 0.119276 LR 0.000125 Time 0.023537 +2023-10-02 21:55:54,557 - Epoch: [180][ 280/ 1236] Overall Loss 0.118952 Objective Loss 0.118952 LR 0.000125 Time 0.023436 +2023-10-02 21:55:54,762 - Epoch: [180][ 290/ 1236] Overall Loss 0.119280 Objective Loss 0.119280 LR 0.000125 Time 0.023332 +2023-10-02 21:55:54,969 - Epoch: [180][ 300/ 1236] Overall Loss 0.118876 Objective Loss 0.118876 LR 0.000125 Time 0.023245 +2023-10-02 21:55:55,174 - Epoch: [180][ 310/ 1236] Overall Loss 0.119190 Objective Loss 0.119190 LR 0.000125 Time 0.023154 +2023-10-02 21:55:55,379 - Epoch: [180][ 320/ 1236] Overall Loss 0.119169 Objective Loss 0.119169 LR 0.000125 Time 0.023069 +2023-10-02 21:55:55,585 - Epoch: [180][ 330/ 1236] Overall Loss 0.119348 Objective Loss 0.119348 LR 0.000125 Time 0.022994 +2023-10-02 21:55:55,792 - Epoch: [180][ 340/ 1236] Overall Loss 0.118716 Objective Loss 0.118716 LR 0.000125 Time 0.022927 +2023-10-02 21:55:55,997 - Epoch: [180][ 350/ 1236] Overall Loss 0.119085 Objective Loss 0.119085 LR 0.000125 Time 0.022857 +2023-10-02 21:55:56,205 - Epoch: [180][ 360/ 1236] Overall Loss 0.119516 Objective Loss 0.119516 LR 0.000125 Time 0.022798 +2023-10-02 21:55:56,410 - Epoch: [180][ 370/ 1236] Overall Loss 0.119531 Objective Loss 0.119531 LR 0.000125 Time 0.022735 +2023-10-02 21:55:56,616 - Epoch: [180][ 380/ 1236] Overall Loss 0.119563 Objective Loss 0.119563 LR 0.000125 Time 0.022679 +2023-10-02 21:55:56,822 - Epoch: [180][ 390/ 1236] Overall Loss 0.119195 Objective Loss 0.119195 LR 0.000125 Time 0.022622 +2023-10-02 21:55:57,028 - Epoch: [180][ 400/ 1236] Overall Loss 0.119084 Objective Loss 0.119084 LR 0.000125 Time 0.022571 +2023-10-02 21:55:57,238 - Epoch: [180][ 410/ 1236] Overall Loss 0.118619 Objective Loss 0.118619 LR 0.000125 Time 0.022528 +2023-10-02 21:55:57,455 - Epoch: [180][ 420/ 1236] Overall Loss 0.118628 Objective Loss 0.118628 LR 0.000125 Time 0.022507 +2023-10-02 21:55:57,667 - Epoch: [180][ 430/ 1236] Overall Loss 0.118297 Objective Loss 0.118297 LR 0.000125 Time 0.022477 +2023-10-02 21:55:57,884 - Epoch: [180][ 440/ 1236] Overall Loss 0.118114 Objective Loss 0.118114 LR 0.000125 Time 0.022458 +2023-10-02 21:55:58,090 - Epoch: [180][ 450/ 1236] Overall Loss 0.118509 Objective Loss 0.118509 LR 0.000125 Time 0.022417 +2023-10-02 21:55:58,298 - Epoch: [180][ 460/ 1236] Overall Loss 0.118651 Objective Loss 0.118651 LR 0.000125 Time 0.022381 +2023-10-02 21:55:58,503 - Epoch: [180][ 470/ 1236] Overall Loss 0.118359 Objective Loss 0.118359 LR 0.000125 Time 0.022340 +2023-10-02 21:55:58,710 - Epoch: [180][ 480/ 1236] Overall Loss 0.118444 Objective Loss 0.118444 LR 0.000125 Time 0.022304 +2023-10-02 21:55:58,916 - Epoch: [180][ 490/ 1236] Overall Loss 0.118226 Objective Loss 0.118226 LR 0.000125 Time 0.022267 +2023-10-02 21:55:59,124 - Epoch: [180][ 500/ 1236] Overall Loss 0.118208 Objective Loss 0.118208 LR 0.000125 Time 0.022237 +2023-10-02 21:55:59,329 - Epoch: [180][ 510/ 1236] Overall Loss 0.118371 Objective Loss 0.118371 LR 0.000125 Time 0.022202 +2023-10-02 21:55:59,537 - Epoch: [180][ 520/ 1236] Overall Loss 0.118413 Objective Loss 0.118413 LR 0.000125 Time 0.022175 +2023-10-02 21:55:59,742 - Epoch: [180][ 530/ 1236] Overall Loss 0.118617 Objective Loss 0.118617 LR 0.000125 Time 0.022143 +2023-10-02 21:55:59,950 - Epoch: [180][ 540/ 1236] Overall Loss 0.118660 Objective Loss 0.118660 LR 0.000125 Time 0.022117 +2023-10-02 21:56:00,155 - Epoch: [180][ 550/ 1236] Overall Loss 0.118777 Objective Loss 0.118777 LR 0.000125 Time 0.022087 +2023-10-02 21:56:00,361 - Epoch: [180][ 560/ 1236] Overall Loss 0.118691 Objective Loss 0.118691 LR 0.000125 Time 0.022061 +2023-10-02 21:56:00,568 - Epoch: [180][ 570/ 1236] Overall Loss 0.118949 Objective Loss 0.118949 LR 0.000125 Time 0.022033 +2023-10-02 21:56:00,774 - Epoch: [180][ 580/ 1236] Overall Loss 0.118656 Objective Loss 0.118656 LR 0.000125 Time 0.022008 +2023-10-02 21:56:00,980 - Epoch: [180][ 590/ 1236] Overall Loss 0.118709 Objective Loss 0.118709 LR 0.000125 Time 0.021982 +2023-10-02 21:56:01,187 - Epoch: [180][ 600/ 1236] Overall Loss 0.118496 Objective Loss 0.118496 LR 0.000125 Time 0.021960 +2023-10-02 21:56:01,393 - Epoch: [180][ 610/ 1236] Overall Loss 0.118385 Objective Loss 0.118385 LR 0.000125 Time 0.021936 +2023-10-02 21:56:01,601 - Epoch: [180][ 620/ 1236] Overall Loss 0.118720 Objective Loss 0.118720 LR 0.000125 Time 0.021917 +2023-10-02 21:56:01,806 - Epoch: [180][ 630/ 1236] Overall Loss 0.118884 Objective Loss 0.118884 LR 0.000125 Time 0.021894 +2023-10-02 21:56:02,013 - Epoch: [180][ 640/ 1236] Overall Loss 0.118867 Objective Loss 0.118867 LR 0.000125 Time 0.021874 +2023-10-02 21:56:02,219 - Epoch: [180][ 650/ 1236] Overall Loss 0.118766 Objective Loss 0.118766 LR 0.000125 Time 0.021853 +2023-10-02 21:56:02,425 - Epoch: [180][ 660/ 1236] Overall Loss 0.118800 Objective Loss 0.118800 LR 0.000125 Time 0.021834 +2023-10-02 21:56:02,632 - Epoch: [180][ 670/ 1236] Overall Loss 0.118848 Objective Loss 0.118848 LR 0.000125 Time 0.021814 +2023-10-02 21:56:02,840 - Epoch: [180][ 680/ 1236] Overall Loss 0.118890 Objective Loss 0.118890 LR 0.000125 Time 0.021799 +2023-10-02 21:56:03,045 - Epoch: [180][ 690/ 1236] Overall Loss 0.118965 Objective Loss 0.118965 LR 0.000125 Time 0.021779 +2023-10-02 21:56:03,253 - Epoch: [180][ 700/ 1236] Overall Loss 0.118721 Objective Loss 0.118721 LR 0.000125 Time 0.021765 +2023-10-02 21:56:03,458 - Epoch: [180][ 710/ 1236] Overall Loss 0.118490 Objective Loss 0.118490 LR 0.000125 Time 0.021747 +2023-10-02 21:56:03,664 - Epoch: [180][ 720/ 1236] Overall Loss 0.118485 Objective Loss 0.118485 LR 0.000125 Time 0.021731 +2023-10-02 21:56:03,870 - Epoch: [180][ 730/ 1236] Overall Loss 0.118709 Objective Loss 0.118709 LR 0.000125 Time 0.021715 +2023-10-02 21:56:04,078 - Epoch: [180][ 740/ 1236] Overall Loss 0.118739 Objective Loss 0.118739 LR 0.000125 Time 0.021702 +2023-10-02 21:56:04,282 - Epoch: [180][ 750/ 1236] Overall Loss 0.118816 Objective Loss 0.118816 LR 0.000125 Time 0.021684 +2023-10-02 21:56:04,490 - Epoch: [180][ 760/ 1236] Overall Loss 0.118758 Objective Loss 0.118758 LR 0.000125 Time 0.021672 +2023-10-02 21:56:04,695 - Epoch: [180][ 770/ 1236] Overall Loss 0.119082 Objective Loss 0.119082 LR 0.000125 Time 0.021657 +2023-10-02 21:56:04,902 - Epoch: [180][ 780/ 1236] Overall Loss 0.119294 Objective Loss 0.119294 LR 0.000125 Time 0.021644 +2023-10-02 21:56:05,108 - Epoch: [180][ 790/ 1236] Overall Loss 0.119252 Objective Loss 0.119252 LR 0.000125 Time 0.021629 +2023-10-02 21:56:05,315 - Epoch: [180][ 800/ 1236] Overall Loss 0.119130 Objective Loss 0.119130 LR 0.000125 Time 0.021617 +2023-10-02 21:56:05,521 - Epoch: [180][ 810/ 1236] Overall Loss 0.119520 Objective Loss 0.119520 LR 0.000125 Time 0.021602 +2023-10-02 21:56:05,731 - Epoch: [180][ 820/ 1236] Overall Loss 0.119538 Objective Loss 0.119538 LR 0.000125 Time 0.021594 +2023-10-02 21:56:05,939 - Epoch: [180][ 830/ 1236] Overall Loss 0.119560 Objective Loss 0.119560 LR 0.000125 Time 0.021584 +2023-10-02 21:56:06,150 - Epoch: [180][ 840/ 1236] Overall Loss 0.119587 Objective Loss 0.119587 LR 0.000125 Time 0.021578 +2023-10-02 21:56:06,359 - Epoch: [180][ 850/ 1236] Overall Loss 0.119529 Objective Loss 0.119529 LR 0.000125 Time 0.021569 +2023-10-02 21:56:06,569 - Epoch: [180][ 860/ 1236] Overall Loss 0.119439 Objective Loss 0.119439 LR 0.000125 Time 0.021563 +2023-10-02 21:56:06,779 - Epoch: [180][ 870/ 1236] Overall Loss 0.119293 Objective Loss 0.119293 LR 0.000125 Time 0.021556 +2023-10-02 21:56:06,992 - Epoch: [180][ 880/ 1236] Overall Loss 0.119133 Objective Loss 0.119133 LR 0.000125 Time 0.021553 +2023-10-02 21:56:07,203 - Epoch: [180][ 890/ 1236] Overall Loss 0.119253 Objective Loss 0.119253 LR 0.000125 Time 0.021547 +2023-10-02 21:56:07,415 - Epoch: [180][ 900/ 1236] Overall Loss 0.119221 Objective Loss 0.119221 LR 0.000125 Time 0.021542 +2023-10-02 21:56:07,625 - Epoch: [180][ 910/ 1236] Overall Loss 0.119149 Objective Loss 0.119149 LR 0.000125 Time 0.021536 +2023-10-02 21:56:07,837 - Epoch: [180][ 920/ 1236] Overall Loss 0.119198 Objective Loss 0.119198 LR 0.000125 Time 0.021532 +2023-10-02 21:56:08,048 - Epoch: [180][ 930/ 1236] Overall Loss 0.119357 Objective Loss 0.119357 LR 0.000125 Time 0.021527 +2023-10-02 21:56:08,260 - Epoch: [180][ 940/ 1236] Overall Loss 0.119527 Objective Loss 0.119527 LR 0.000125 Time 0.021523 +2023-10-02 21:56:08,471 - Epoch: [180][ 950/ 1236] Overall Loss 0.119377 Objective Loss 0.119377 LR 0.000125 Time 0.021518 +2023-10-02 21:56:08,683 - Epoch: [180][ 960/ 1236] Overall Loss 0.119243 Objective Loss 0.119243 LR 0.000125 Time 0.021515 +2023-10-02 21:56:08,896 - Epoch: [180][ 970/ 1236] Overall Loss 0.119232 Objective Loss 0.119232 LR 0.000125 Time 0.021511 +2023-10-02 21:56:09,108 - Epoch: [180][ 980/ 1236] Overall Loss 0.119239 Objective Loss 0.119239 LR 0.000125 Time 0.021508 +2023-10-02 21:56:09,322 - Epoch: [180][ 990/ 1236] Overall Loss 0.119065 Objective Loss 0.119065 LR 0.000125 Time 0.021507 +2023-10-02 21:56:09,533 - Epoch: [180][ 1000/ 1236] Overall Loss 0.119138 Objective Loss 0.119138 LR 0.000125 Time 0.021502 +2023-10-02 21:56:09,746 - Epoch: [180][ 1010/ 1236] Overall Loss 0.119206 Objective Loss 0.119206 LR 0.000125 Time 0.021500 +2023-10-02 21:56:09,957 - Epoch: [180][ 1020/ 1236] Overall Loss 0.119255 Objective Loss 0.119255 LR 0.000125 Time 0.021496 +2023-10-02 21:56:10,170 - Epoch: [180][ 1030/ 1236] Overall Loss 0.119312 Objective Loss 0.119312 LR 0.000125 Time 0.021493 +2023-10-02 21:56:10,381 - Epoch: [180][ 1040/ 1236] Overall Loss 0.119351 Objective Loss 0.119351 LR 0.000125 Time 0.021489 +2023-10-02 21:56:10,593 - Epoch: [180][ 1050/ 1236] Overall Loss 0.119479 Objective Loss 0.119479 LR 0.000125 Time 0.021486 +2023-10-02 21:56:10,805 - Epoch: [180][ 1060/ 1236] Overall Loss 0.119482 Objective Loss 0.119482 LR 0.000125 Time 0.021483 +2023-10-02 21:56:11,017 - Epoch: [180][ 1070/ 1236] Overall Loss 0.119227 Objective Loss 0.119227 LR 0.000125 Time 0.021480 +2023-10-02 21:56:11,228 - Epoch: [180][ 1080/ 1236] Overall Loss 0.119294 Objective Loss 0.119294 LR 0.000125 Time 0.021476 +2023-10-02 21:56:11,441 - Epoch: [180][ 1090/ 1236] Overall Loss 0.119235 Objective Loss 0.119235 LR 0.000125 Time 0.021474 +2023-10-02 21:56:11,652 - Epoch: [180][ 1100/ 1236] Overall Loss 0.119131 Objective Loss 0.119131 LR 0.000125 Time 0.021470 +2023-10-02 21:56:11,865 - Epoch: [180][ 1110/ 1236] Overall Loss 0.119183 Objective Loss 0.119183 LR 0.000125 Time 0.021468 +2023-10-02 21:56:12,075 - Epoch: [180][ 1120/ 1236] Overall Loss 0.119187 Objective Loss 0.119187 LR 0.000125 Time 0.021464 +2023-10-02 21:56:12,288 - Epoch: [180][ 1130/ 1236] Overall Loss 0.119182 Objective Loss 0.119182 LR 0.000125 Time 0.021462 +2023-10-02 21:56:12,500 - Epoch: [180][ 1140/ 1236] Overall Loss 0.119084 Objective Loss 0.119084 LR 0.000125 Time 0.021459 +2023-10-02 21:56:12,712 - Epoch: [180][ 1150/ 1236] Overall Loss 0.119079 Objective Loss 0.119079 LR 0.000125 Time 0.021457 +2023-10-02 21:56:12,924 - Epoch: [180][ 1160/ 1236] Overall Loss 0.119184 Objective Loss 0.119184 LR 0.000125 Time 0.021454 +2023-10-02 21:56:13,136 - Epoch: [180][ 1170/ 1236] Overall Loss 0.119302 Objective Loss 0.119302 LR 0.000125 Time 0.021452 +2023-10-02 21:56:13,347 - Epoch: [180][ 1180/ 1236] Overall Loss 0.119399 Objective Loss 0.119399 LR 0.000125 Time 0.021448 +2023-10-02 21:56:13,559 - Epoch: [180][ 1190/ 1236] Overall Loss 0.119349 Objective Loss 0.119349 LR 0.000125 Time 0.021446 +2023-10-02 21:56:13,769 - Epoch: [180][ 1200/ 1236] Overall Loss 0.119334 Objective Loss 0.119334 LR 0.000125 Time 0.021443 +2023-10-02 21:56:13,982 - Epoch: [180][ 1210/ 1236] Overall Loss 0.119473 Objective Loss 0.119473 LR 0.000125 Time 0.021440 +2023-10-02 21:56:14,192 - Epoch: [180][ 1220/ 1236] Overall Loss 0.119432 Objective Loss 0.119432 LR 0.000125 Time 0.021437 +2023-10-02 21:56:14,457 - Epoch: [180][ 1230/ 1236] Overall Loss 0.119254 Objective Loss 0.119254 LR 0.000125 Time 0.021477 +2023-10-02 21:56:14,578 - Epoch: [180][ 1236/ 1236] Overall Loss 0.119354 Objective Loss 0.119354 Top1 89.816701 Top5 98.778004 LR 0.000125 Time 0.021471 +2023-10-02 21:56:14,703 - --- validate (epoch=180)----------- +2023-10-02 21:56:14,703 - 29943 samples (256 per mini-batch) +2023-10-02 21:56:15,197 - Epoch: [180][ 10/ 117] Loss 0.352916 Top1 86.406250 Top5 98.281250 +2023-10-02 21:56:15,349 - Epoch: [180][ 20/ 117] Loss 0.338606 Top1 86.757812 Top5 98.417969 +2023-10-02 21:56:15,499 - Epoch: [180][ 30/ 117] Loss 0.325797 Top1 87.265625 Top5 98.398438 +2023-10-02 21:56:15,651 - Epoch: [180][ 40/ 117] Loss 0.307884 Top1 87.480469 Top5 98.564453 +2023-10-02 21:56:15,802 - Epoch: [180][ 50/ 117] Loss 0.298230 Top1 87.398438 Top5 98.609375 +2023-10-02 21:56:15,954 - Epoch: [180][ 60/ 117] Loss 0.300304 Top1 87.272135 Top5 98.587240 +2023-10-02 21:56:16,104 - Epoch: [180][ 70/ 117] Loss 0.303424 Top1 87.181920 Top5 98.599330 +2023-10-02 21:56:16,256 - Epoch: [180][ 80/ 117] Loss 0.306315 Top1 87.275391 Top5 98.603516 +2023-10-02 21:56:16,405 - Epoch: [180][ 90/ 117] Loss 0.301553 Top1 87.365451 Top5 98.615451 +2023-10-02 21:56:16,558 - Epoch: [180][ 100/ 117] Loss 0.302901 Top1 87.390625 Top5 98.628906 +2023-10-02 21:56:16,715 - Epoch: [180][ 110/ 117] Loss 0.307044 Top1 87.436080 Top5 98.629261 +2023-10-02 21:56:16,804 - Epoch: [180][ 117/ 117] Loss 0.306128 Top1 87.469525 Top5 98.657449 +2023-10-02 21:56:16,956 - ==> Top1: 87.470 Top5: 98.657 Loss: 0.306 + +2023-10-02 21:56:16,956 - ==> Confusion: +[[ 940 0 2 0 2 2 0 0 7 61 1 1 1 2 3 1 2 0 2 0 23] + [ 0 1074 1 0 2 12 0 20 1 2 0 1 1 0 1 3 0 0 8 1 4] + [ 1 2 984 5 0 0 16 10 0 1 2 1 5 2 1 5 2 0 9 2 8] + [ 1 2 17 977 0 0 0 0 2 1 4 1 6 3 26 3 2 5 11 1 27] + [ 22 7 1 1 969 3 1 0 0 15 1 0 1 4 6 6 7 0 0 2 4] + [ 5 41 0 0 5 983 3 25 2 7 2 5 2 10 3 0 3 1 2 1 16] + [ 0 4 21 0 0 0 1142 4 0 0 2 2 0 0 0 2 0 1 3 5 5] + [ 1 12 8 0 6 20 6 1077 2 4 6 4 3 4 0 0 0 0 45 9 11] + [ 14 2 0 1 2 2 0 1 986 36 10 2 1 9 13 1 1 1 3 3 1] + [ 92 2 0 1 4 2 0 0 33 951 1 1 0 15 4 2 0 0 0 2 9] + [ 2 2 9 5 0 1 7 3 8 2 972 1 0 13 4 0 1 2 6 3 12] + [ 0 0 2 0 1 13 0 4 1 0 0 965 14 6 0 3 1 14 0 4 7] + [ 0 0 0 1 0 0 2 2 0 3 1 28 982 0 3 7 2 12 4 7 14] + [ 1 0 0 0 4 5 0 0 13 12 4 7 0 1054 3 0 0 1 1 1 13] + [ 10 0 3 15 2 1 0 0 18 5 1 0 2 2 1017 0 1 2 12 0 10] + [ 0 0 1 1 6 1 1 0 0 1 1 6 7 0 1 1068 14 11 2 8 5] + [ 0 15 1 0 4 6 1 0 1 0 0 5 0 2 5 9 1094 0 1 6 11] + [ 0 0 0 1 0 0 2 0 0 0 0 5 21 1 2 8 0 991 0 3 4] + [ 2 4 2 12 0 0 0 21 5 2 2 0 1 0 10 0 0 0 993 0 14] + [ 0 1 6 2 0 3 8 8 0 1 2 11 5 2 2 1 6 0 0 1087 7] + [ 103 136 100 75 56 96 34 82 93 65 133 75 275 217 100 46 62 50 101 121 5885]] + +2023-10-02 21:56:16,958 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:56:16,958 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:56:16,964 - + +2023-10-02 21:56:16,964 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:56:18,000 - Epoch: [181][ 10/ 1236] Overall Loss 0.113572 Objective Loss 0.113572 LR 0.000125 Time 0.103543 +2023-10-02 21:56:18,208 - Epoch: [181][ 20/ 1236] Overall Loss 0.117661 Objective Loss 0.117661 LR 0.000125 Time 0.062140 +2023-10-02 21:56:18,415 - Epoch: [181][ 30/ 1236] Overall Loss 0.117848 Objective Loss 0.117848 LR 0.000125 Time 0.048328 +2023-10-02 21:56:18,623 - Epoch: [181][ 40/ 1236] Overall Loss 0.119299 Objective Loss 0.119299 LR 0.000125 Time 0.041431 +2023-10-02 21:56:18,830 - Epoch: [181][ 50/ 1236] Overall Loss 0.118971 Objective Loss 0.118971 LR 0.000125 Time 0.037255 +2023-10-02 21:56:19,038 - Epoch: [181][ 60/ 1236] Overall Loss 0.116600 Objective Loss 0.116600 LR 0.000125 Time 0.034506 +2023-10-02 21:56:19,244 - Epoch: [181][ 70/ 1236] Overall Loss 0.115796 Objective Loss 0.115796 LR 0.000125 Time 0.032506 +2023-10-02 21:56:19,454 - Epoch: [181][ 80/ 1236] Overall Loss 0.114776 Objective Loss 0.114776 LR 0.000125 Time 0.031062 +2023-10-02 21:56:19,663 - Epoch: [181][ 90/ 1236] Overall Loss 0.115116 Objective Loss 0.115116 LR 0.000125 Time 0.029914 +2023-10-02 21:56:19,873 - Epoch: [181][ 100/ 1236] Overall Loss 0.115541 Objective Loss 0.115541 LR 0.000125 Time 0.029019 +2023-10-02 21:56:20,081 - Epoch: [181][ 110/ 1236] Overall Loss 0.115670 Objective Loss 0.115670 LR 0.000125 Time 0.028257 +2023-10-02 21:56:20,291 - Epoch: [181][ 120/ 1236] Overall Loss 0.116471 Objective Loss 0.116471 LR 0.000125 Time 0.027647 +2023-10-02 21:56:20,498 - Epoch: [181][ 130/ 1236] Overall Loss 0.116158 Objective Loss 0.116158 LR 0.000125 Time 0.027107 +2023-10-02 21:56:20,708 - Epoch: [181][ 140/ 1236] Overall Loss 0.115545 Objective Loss 0.115545 LR 0.000125 Time 0.026668 +2023-10-02 21:56:20,916 - Epoch: [181][ 150/ 1236] Overall Loss 0.114392 Objective Loss 0.114392 LR 0.000125 Time 0.026266 +2023-10-02 21:56:21,126 - Epoch: [181][ 160/ 1236] Overall Loss 0.114795 Objective Loss 0.114795 LR 0.000125 Time 0.025936 +2023-10-02 21:56:21,335 - Epoch: [181][ 170/ 1236] Overall Loss 0.116175 Objective Loss 0.116175 LR 0.000125 Time 0.025625 +2023-10-02 21:56:21,544 - Epoch: [181][ 180/ 1236] Overall Loss 0.115624 Objective Loss 0.115624 LR 0.000125 Time 0.025366 +2023-10-02 21:56:21,753 - Epoch: [181][ 190/ 1236] Overall Loss 0.115621 Objective Loss 0.115621 LR 0.000125 Time 0.025120 +2023-10-02 21:56:21,963 - Epoch: [181][ 200/ 1236] Overall Loss 0.115176 Objective Loss 0.115176 LR 0.000125 Time 0.024913 +2023-10-02 21:56:22,171 - Epoch: [181][ 210/ 1236] Overall Loss 0.116101 Objective Loss 0.116101 LR 0.000125 Time 0.024710 +2023-10-02 21:56:22,381 - Epoch: [181][ 220/ 1236] Overall Loss 0.115867 Objective Loss 0.115867 LR 0.000125 Time 0.024541 +2023-10-02 21:56:22,590 - Epoch: [181][ 230/ 1236] Overall Loss 0.115957 Objective Loss 0.115957 LR 0.000125 Time 0.024374 +2023-10-02 21:56:22,800 - Epoch: [181][ 240/ 1236] Overall Loss 0.116042 Objective Loss 0.116042 LR 0.000125 Time 0.024233 +2023-10-02 21:56:23,008 - Epoch: [181][ 250/ 1236] Overall Loss 0.115590 Objective Loss 0.115590 LR 0.000125 Time 0.024089 +2023-10-02 21:56:23,218 - Epoch: [181][ 260/ 1236] Overall Loss 0.115889 Objective Loss 0.115889 LR 0.000125 Time 0.023969 +2023-10-02 21:56:23,426 - Epoch: [181][ 270/ 1236] Overall Loss 0.115762 Objective Loss 0.115762 LR 0.000125 Time 0.023846 +2023-10-02 21:56:23,636 - Epoch: [181][ 280/ 1236] Overall Loss 0.116032 Objective Loss 0.116032 LR 0.000125 Time 0.023743 +2023-10-02 21:56:23,844 - Epoch: [181][ 290/ 1236] Overall Loss 0.115896 Objective Loss 0.115896 LR 0.000125 Time 0.023637 +2023-10-02 21:56:24,054 - Epoch: [181][ 300/ 1236] Overall Loss 0.115734 Objective Loss 0.115734 LR 0.000125 Time 0.023549 +2023-10-02 21:56:24,263 - Epoch: [181][ 310/ 1236] Overall Loss 0.116206 Objective Loss 0.116206 LR 0.000125 Time 0.023456 +2023-10-02 21:56:24,473 - Epoch: [181][ 320/ 1236] Overall Loss 0.116131 Objective Loss 0.116131 LR 0.000125 Time 0.023378 +2023-10-02 21:56:24,681 - Epoch: [181][ 330/ 1236] Overall Loss 0.116349 Objective Loss 0.116349 LR 0.000125 Time 0.023297 +2023-10-02 21:56:24,892 - Epoch: [181][ 340/ 1236] Overall Loss 0.116396 Objective Loss 0.116396 LR 0.000125 Time 0.023230 +2023-10-02 21:56:25,103 - Epoch: [181][ 350/ 1236] Overall Loss 0.116438 Objective Loss 0.116438 LR 0.000125 Time 0.023166 +2023-10-02 21:56:25,323 - Epoch: [181][ 360/ 1236] Overall Loss 0.116326 Objective Loss 0.116326 LR 0.000125 Time 0.023132 +2023-10-02 21:56:25,530 - Epoch: [181][ 370/ 1236] Overall Loss 0.116554 Objective Loss 0.116554 LR 0.000125 Time 0.023067 +2023-10-02 21:56:25,739 - Epoch: [181][ 380/ 1236] Overall Loss 0.116418 Objective Loss 0.116418 LR 0.000125 Time 0.023009 +2023-10-02 21:56:25,947 - Epoch: [181][ 390/ 1236] Overall Loss 0.116797 Objective Loss 0.116797 LR 0.000125 Time 0.022948 +2023-10-02 21:56:26,156 - Epoch: [181][ 400/ 1236] Overall Loss 0.116510 Objective Loss 0.116510 LR 0.000125 Time 0.022894 +2023-10-02 21:56:26,363 - Epoch: [181][ 410/ 1236] Overall Loss 0.116606 Objective Loss 0.116606 LR 0.000125 Time 0.022839 +2023-10-02 21:56:26,572 - Epoch: [181][ 420/ 1236] Overall Loss 0.116849 Objective Loss 0.116849 LR 0.000125 Time 0.022791 +2023-10-02 21:56:26,780 - Epoch: [181][ 430/ 1236] Overall Loss 0.116705 Objective Loss 0.116705 LR 0.000125 Time 0.022740 +2023-10-02 21:56:26,988 - Epoch: [181][ 440/ 1236] Overall Loss 0.116292 Objective Loss 0.116292 LR 0.000125 Time 0.022697 +2023-10-02 21:56:27,196 - Epoch: [181][ 450/ 1236] Overall Loss 0.116507 Objective Loss 0.116507 LR 0.000125 Time 0.022652 +2023-10-02 21:56:27,405 - Epoch: [181][ 460/ 1236] Overall Loss 0.116611 Objective Loss 0.116611 LR 0.000125 Time 0.022612 +2023-10-02 21:56:27,613 - Epoch: [181][ 470/ 1236] Overall Loss 0.116209 Objective Loss 0.116209 LR 0.000125 Time 0.022570 +2023-10-02 21:56:27,821 - Epoch: [181][ 480/ 1236] Overall Loss 0.116120 Objective Loss 0.116120 LR 0.000125 Time 0.022534 +2023-10-02 21:56:28,029 - Epoch: [181][ 490/ 1236] Overall Loss 0.116064 Objective Loss 0.116064 LR 0.000125 Time 0.022495 +2023-10-02 21:56:28,238 - Epoch: [181][ 500/ 1236] Overall Loss 0.116197 Objective Loss 0.116197 LR 0.000125 Time 0.022462 +2023-10-02 21:56:28,447 - Epoch: [181][ 510/ 1236] Overall Loss 0.116300 Objective Loss 0.116300 LR 0.000125 Time 0.022429 +2023-10-02 21:56:28,658 - Epoch: [181][ 520/ 1236] Overall Loss 0.116544 Objective Loss 0.116544 LR 0.000125 Time 0.022402 +2023-10-02 21:56:28,867 - Epoch: [181][ 530/ 1236] Overall Loss 0.116351 Objective Loss 0.116351 LR 0.000125 Time 0.022371 +2023-10-02 21:56:29,078 - Epoch: [181][ 540/ 1236] Overall Loss 0.116609 Objective Loss 0.116609 LR 0.000125 Time 0.022347 +2023-10-02 21:56:29,287 - Epoch: [181][ 550/ 1236] Overall Loss 0.116473 Objective Loss 0.116473 LR 0.000125 Time 0.022317 +2023-10-02 21:56:29,498 - Epoch: [181][ 560/ 1236] Overall Loss 0.116542 Objective Loss 0.116542 LR 0.000125 Time 0.022294 +2023-10-02 21:56:29,707 - Epoch: [181][ 570/ 1236] Overall Loss 0.116675 Objective Loss 0.116675 LR 0.000125 Time 0.022267 +2023-10-02 21:56:29,917 - Epoch: [181][ 580/ 1236] Overall Loss 0.116647 Objective Loss 0.116647 LR 0.000125 Time 0.022246 +2023-10-02 21:56:30,127 - Epoch: [181][ 590/ 1236] Overall Loss 0.116575 Objective Loss 0.116575 LR 0.000125 Time 0.022223 +2023-10-02 21:56:30,337 - Epoch: [181][ 600/ 1236] Overall Loss 0.116225 Objective Loss 0.116225 LR 0.000125 Time 0.022203 +2023-10-02 21:56:30,548 - Epoch: [181][ 610/ 1236] Overall Loss 0.116078 Objective Loss 0.116078 LR 0.000125 Time 0.022182 +2023-10-02 21:56:30,767 - Epoch: [181][ 620/ 1236] Overall Loss 0.115940 Objective Loss 0.115940 LR 0.000125 Time 0.022177 +2023-10-02 21:56:30,974 - Epoch: [181][ 630/ 1236] Overall Loss 0.115899 Objective Loss 0.115899 LR 0.000125 Time 0.022153 +2023-10-02 21:56:31,181 - Epoch: [181][ 640/ 1236] Overall Loss 0.115712 Objective Loss 0.115712 LR 0.000125 Time 0.022129 +2023-10-02 21:56:31,388 - Epoch: [181][ 650/ 1236] Overall Loss 0.115680 Objective Loss 0.115680 LR 0.000125 Time 0.022106 +2023-10-02 21:56:31,596 - Epoch: [181][ 660/ 1236] Overall Loss 0.115719 Objective Loss 0.115719 LR 0.000125 Time 0.022086 +2023-10-02 21:56:31,804 - Epoch: [181][ 670/ 1236] Overall Loss 0.115680 Objective Loss 0.115680 LR 0.000125 Time 0.022063 +2023-10-02 21:56:32,012 - Epoch: [181][ 680/ 1236] Overall Loss 0.115688 Objective Loss 0.115688 LR 0.000125 Time 0.022044 +2023-10-02 21:56:32,219 - Epoch: [181][ 690/ 1236] Overall Loss 0.115757 Objective Loss 0.115757 LR 0.000125 Time 0.022022 +2023-10-02 21:56:32,427 - Epoch: [181][ 700/ 1236] Overall Loss 0.115476 Objective Loss 0.115476 LR 0.000125 Time 0.022004 +2023-10-02 21:56:32,634 - Epoch: [181][ 710/ 1236] Overall Loss 0.115474 Objective Loss 0.115474 LR 0.000125 Time 0.021984 +2023-10-02 21:56:32,842 - Epoch: [181][ 720/ 1236] Overall Loss 0.115316 Objective Loss 0.115316 LR 0.000125 Time 0.021967 +2023-10-02 21:56:33,049 - Epoch: [181][ 730/ 1236] Overall Loss 0.115286 Objective Loss 0.115286 LR 0.000125 Time 0.021947 +2023-10-02 21:56:33,257 - Epoch: [181][ 740/ 1236] Overall Loss 0.115323 Objective Loss 0.115323 LR 0.000125 Time 0.021931 +2023-10-02 21:56:33,464 - Epoch: [181][ 750/ 1236] Overall Loss 0.115560 Objective Loss 0.115560 LR 0.000125 Time 0.021913 +2023-10-02 21:56:33,672 - Epoch: [181][ 760/ 1236] Overall Loss 0.115603 Objective Loss 0.115603 LR 0.000125 Time 0.021898 +2023-10-02 21:56:33,879 - Epoch: [181][ 770/ 1236] Overall Loss 0.115390 Objective Loss 0.115390 LR 0.000125 Time 0.021881 +2023-10-02 21:56:34,087 - Epoch: [181][ 780/ 1236] Overall Loss 0.115421 Objective Loss 0.115421 LR 0.000125 Time 0.021867 +2023-10-02 21:56:34,294 - Epoch: [181][ 790/ 1236] Overall Loss 0.115308 Objective Loss 0.115308 LR 0.000125 Time 0.021850 +2023-10-02 21:56:34,503 - Epoch: [181][ 800/ 1236] Overall Loss 0.115332 Objective Loss 0.115332 LR 0.000125 Time 0.021837 +2023-10-02 21:56:34,710 - Epoch: [181][ 810/ 1236] Overall Loss 0.115468 Objective Loss 0.115468 LR 0.000125 Time 0.021821 +2023-10-02 21:56:34,918 - Epoch: [181][ 820/ 1236] Overall Loss 0.115591 Objective Loss 0.115591 LR 0.000125 Time 0.021808 +2023-10-02 21:56:35,125 - Epoch: [181][ 830/ 1236] Overall Loss 0.115539 Objective Loss 0.115539 LR 0.000125 Time 0.021793 +2023-10-02 21:56:35,333 - Epoch: [181][ 840/ 1236] Overall Loss 0.115456 Objective Loss 0.115456 LR 0.000125 Time 0.021781 +2023-10-02 21:56:35,540 - Epoch: [181][ 850/ 1236] Overall Loss 0.115617 Objective Loss 0.115617 LR 0.000125 Time 0.021767 +2023-10-02 21:56:35,748 - Epoch: [181][ 860/ 1236] Overall Loss 0.115478 Objective Loss 0.115478 LR 0.000125 Time 0.021756 +2023-10-02 21:56:35,956 - Epoch: [181][ 870/ 1236] Overall Loss 0.115448 Objective Loss 0.115448 LR 0.000125 Time 0.021742 +2023-10-02 21:56:36,164 - Epoch: [181][ 880/ 1236] Overall Loss 0.115458 Objective Loss 0.115458 LR 0.000125 Time 0.021731 +2023-10-02 21:56:36,371 - Epoch: [181][ 890/ 1236] Overall Loss 0.115313 Objective Loss 0.115313 LR 0.000125 Time 0.021718 +2023-10-02 21:56:36,579 - Epoch: [181][ 900/ 1236] Overall Loss 0.115151 Objective Loss 0.115151 LR 0.000125 Time 0.021708 +2023-10-02 21:56:36,787 - Epoch: [181][ 910/ 1236] Overall Loss 0.115026 Objective Loss 0.115026 LR 0.000125 Time 0.021695 +2023-10-02 21:56:36,995 - Epoch: [181][ 920/ 1236] Overall Loss 0.114917 Objective Loss 0.114917 LR 0.000125 Time 0.021686 +2023-10-02 21:56:37,204 - Epoch: [181][ 930/ 1236] Overall Loss 0.115028 Objective Loss 0.115028 LR 0.000125 Time 0.021676 +2023-10-02 21:56:37,415 - Epoch: [181][ 940/ 1236] Overall Loss 0.115050 Objective Loss 0.115050 LR 0.000125 Time 0.021669 +2023-10-02 21:56:37,622 - Epoch: [181][ 950/ 1236] Overall Loss 0.115149 Objective Loss 0.115149 LR 0.000125 Time 0.021658 +2023-10-02 21:56:37,833 - Epoch: [181][ 960/ 1236] Overall Loss 0.115120 Objective Loss 0.115120 LR 0.000125 Time 0.021652 +2023-10-02 21:56:38,040 - Epoch: [181][ 970/ 1236] Overall Loss 0.115280 Objective Loss 0.115280 LR 0.000125 Time 0.021642 +2023-10-02 21:56:38,250 - Epoch: [181][ 980/ 1236] Overall Loss 0.115303 Objective Loss 0.115303 LR 0.000125 Time 0.021635 +2023-10-02 21:56:38,457 - Epoch: [181][ 990/ 1236] Overall Loss 0.115178 Objective Loss 0.115178 LR 0.000125 Time 0.021626 +2023-10-02 21:56:38,668 - Epoch: [181][ 1000/ 1236] Overall Loss 0.115386 Objective Loss 0.115386 LR 0.000125 Time 0.021620 +2023-10-02 21:56:38,875 - Epoch: [181][ 1010/ 1236] Overall Loss 0.115462 Objective Loss 0.115462 LR 0.000125 Time 0.021611 +2023-10-02 21:56:39,086 - Epoch: [181][ 1020/ 1236] Overall Loss 0.115548 Objective Loss 0.115548 LR 0.000125 Time 0.021605 +2023-10-02 21:56:39,293 - Epoch: [181][ 1030/ 1236] Overall Loss 0.115643 Objective Loss 0.115643 LR 0.000125 Time 0.021596 +2023-10-02 21:56:39,503 - Epoch: [181][ 1040/ 1236] Overall Loss 0.115842 Objective Loss 0.115842 LR 0.000125 Time 0.021590 +2023-10-02 21:56:39,711 - Epoch: [181][ 1050/ 1236] Overall Loss 0.115853 Objective Loss 0.115853 LR 0.000125 Time 0.021582 +2023-10-02 21:56:39,921 - Epoch: [181][ 1060/ 1236] Overall Loss 0.116004 Objective Loss 0.116004 LR 0.000125 Time 0.021577 +2023-10-02 21:56:40,128 - Epoch: [181][ 1070/ 1236] Overall Loss 0.115917 Objective Loss 0.115917 LR 0.000125 Time 0.021568 +2023-10-02 21:56:40,339 - Epoch: [181][ 1080/ 1236] Overall Loss 0.116051 Objective Loss 0.116051 LR 0.000125 Time 0.021563 +2023-10-02 21:56:40,546 - Epoch: [181][ 1090/ 1236] Overall Loss 0.116018 Objective Loss 0.116018 LR 0.000125 Time 0.021555 +2023-10-02 21:56:40,757 - Epoch: [181][ 1100/ 1236] Overall Loss 0.116055 Objective Loss 0.116055 LR 0.000125 Time 0.021550 +2023-10-02 21:56:40,964 - Epoch: [181][ 1110/ 1236] Overall Loss 0.116040 Objective Loss 0.116040 LR 0.000125 Time 0.021543 +2023-10-02 21:56:41,174 - Epoch: [181][ 1120/ 1236] Overall Loss 0.116024 Objective Loss 0.116024 LR 0.000125 Time 0.021538 +2023-10-02 21:56:41,381 - Epoch: [181][ 1130/ 1236] Overall Loss 0.115935 Objective Loss 0.115935 LR 0.000125 Time 0.021530 +2023-10-02 21:56:41,592 - Epoch: [181][ 1140/ 1236] Overall Loss 0.115861 Objective Loss 0.115861 LR 0.000125 Time 0.021526 +2023-10-02 21:56:41,799 - Epoch: [181][ 1150/ 1236] Overall Loss 0.115797 Objective Loss 0.115797 LR 0.000125 Time 0.021518 +2023-10-02 21:56:42,008 - Epoch: [181][ 1160/ 1236] Overall Loss 0.115711 Objective Loss 0.115711 LR 0.000125 Time 0.021513 +2023-10-02 21:56:42,216 - Epoch: [181][ 1170/ 1236] Overall Loss 0.115793 Objective Loss 0.115793 LR 0.000125 Time 0.021506 +2023-10-02 21:56:42,427 - Epoch: [181][ 1180/ 1236] Overall Loss 0.115802 Objective Loss 0.115802 LR 0.000125 Time 0.021502 +2023-10-02 21:56:42,634 - Epoch: [181][ 1190/ 1236] Overall Loss 0.115785 Objective Loss 0.115785 LR 0.000125 Time 0.021495 +2023-10-02 21:56:42,845 - Epoch: [181][ 1200/ 1236] Overall Loss 0.115797 Objective Loss 0.115797 LR 0.000125 Time 0.021491 +2023-10-02 21:56:43,052 - Epoch: [181][ 1210/ 1236] Overall Loss 0.115791 Objective Loss 0.115791 LR 0.000125 Time 0.021484 +2023-10-02 21:56:43,263 - Epoch: [181][ 1220/ 1236] Overall Loss 0.115717 Objective Loss 0.115717 LR 0.000125 Time 0.021481 +2023-10-02 21:56:43,525 - Epoch: [181][ 1230/ 1236] Overall Loss 0.115664 Objective Loss 0.115664 LR 0.000125 Time 0.021519 +2023-10-02 21:56:43,648 - Epoch: [181][ 1236/ 1236] Overall Loss 0.115628 Objective Loss 0.115628 Top1 91.038697 Top5 99.185336 LR 0.000125 Time 0.021514 +2023-10-02 21:56:43,791 - --- validate (epoch=181)----------- +2023-10-02 21:56:43,792 - 29943 samples (256 per mini-batch) +2023-10-02 21:56:44,293 - Epoch: [181][ 10/ 117] Loss 0.325461 Top1 86.953125 Top5 98.789062 +2023-10-02 21:56:44,445 - Epoch: [181][ 20/ 117] Loss 0.313029 Top1 86.914062 Top5 98.710938 +2023-10-02 21:56:44,597 - Epoch: [181][ 30/ 117] Loss 0.316706 Top1 86.770833 Top5 98.684896 +2023-10-02 21:56:44,750 - Epoch: [181][ 40/ 117] Loss 0.302581 Top1 87.226562 Top5 98.662109 +2023-10-02 21:56:44,901 - Epoch: [181][ 50/ 117] Loss 0.304903 Top1 87.210938 Top5 98.640625 +2023-10-02 21:56:45,052 - Epoch: [181][ 60/ 117] Loss 0.302360 Top1 87.415365 Top5 98.652344 +2023-10-02 21:56:45,204 - Epoch: [181][ 70/ 117] Loss 0.303084 Top1 87.522321 Top5 98.655134 +2023-10-02 21:56:45,356 - Epoch: [181][ 80/ 117] Loss 0.303919 Top1 87.309570 Top5 98.681641 +2023-10-02 21:56:45,508 - Epoch: [181][ 90/ 117] Loss 0.304833 Top1 87.348090 Top5 98.641493 +2023-10-02 21:56:45,660 - Epoch: [181][ 100/ 117] Loss 0.303416 Top1 87.332031 Top5 98.714844 +2023-10-02 21:56:45,818 - Epoch: [181][ 110/ 117] Loss 0.306961 Top1 87.318892 Top5 98.703835 +2023-10-02 21:56:45,907 - Epoch: [181][ 117/ 117] Loss 0.306463 Top1 87.379354 Top5 98.680827 +2023-10-02 21:56:46,010 - ==> Top1: 87.379 Top5: 98.681 Loss: 0.306 + +2023-10-02 21:56:46,010 - ==> Confusion: +[[ 941 0 4 1 8 2 0 1 4 54 1 1 1 1 5 2 2 0 0 0 22] + [ 0 1068 1 0 1 18 0 22 1 0 1 1 0 0 0 3 1 0 5 4 5] + [ 1 1 976 9 0 0 20 12 0 2 0 1 7 2 2 2 2 2 7 3 7] + [ 1 3 14 990 1 0 1 3 2 0 3 0 4 2 20 3 2 5 11 2 22] + [ 19 5 1 1 971 8 0 0 2 11 1 0 0 2 8 6 9 0 0 1 5] + [ 3 35 0 2 5 990 2 26 1 5 1 6 2 8 6 0 2 1 3 3 15] + [ 0 4 18 0 0 0 1140 5 0 0 3 1 0 0 0 5 0 1 1 7 6] + [ 1 10 6 1 4 17 5 1100 2 2 4 3 3 5 2 0 0 2 34 6 11] + [ 15 5 0 1 1 4 0 3 981 32 9 1 1 12 13 0 2 1 3 2 3] + [ 100 1 0 2 7 3 0 1 33 924 0 0 1 29 5 3 1 0 0 0 9] + [ 2 1 8 6 0 1 6 4 8 3 973 2 0 9 4 0 1 2 6 3 14] + [ 0 0 1 0 1 12 0 4 1 0 0 973 11 5 0 1 0 17 0 4 5] + [ 0 1 0 2 0 1 3 1 0 2 3 35 971 0 3 7 1 19 1 8 10] + [ 0 0 1 0 3 4 0 1 11 10 1 9 0 1059 4 1 0 1 0 0 14] + [ 10 1 5 21 4 0 1 0 17 2 1 0 3 2 1011 0 0 2 10 0 11] + [ 0 0 2 1 4 1 0 0 0 1 0 5 8 0 0 1073 15 9 2 8 5] + [ 0 18 0 0 4 4 2 1 1 0 0 4 0 3 3 8 1092 0 1 6 14] + [ 0 0 0 2 1 1 2 0 0 0 0 3 15 1 1 6 0 1002 0 3 1] + [ 1 5 3 17 1 0 0 24 4 0 3 0 1 0 9 0 0 1 988 0 11] + [ 0 1 4 2 0 3 5 10 0 0 1 13 5 2 2 1 5 0 0 1095 3] + [ 88 127 103 73 63 112 29 114 69 55 125 91 286 215 97 53 65 56 99 139 5846]] + +2023-10-02 21:56:46,012 - ==> Best [Top1: 87.573 Top5: 98.677 Sparsity:0.00 Params: 169472 on epoch: 172] +2023-10-02 21:56:46,012 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:56:46,018 - + +2023-10-02 21:56:46,018 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:56:47,138 - Epoch: [182][ 10/ 1236] Overall Loss 0.122050 Objective Loss 0.122050 LR 0.000125 Time 0.111896 +2023-10-02 21:56:47,344 - Epoch: [182][ 20/ 1236] Overall Loss 0.125498 Objective Loss 0.125498 LR 0.000125 Time 0.066222 +2023-10-02 21:56:47,550 - Epoch: [182][ 30/ 1236] Overall Loss 0.121289 Objective Loss 0.121289 LR 0.000125 Time 0.051017 +2023-10-02 21:56:47,757 - Epoch: [182][ 40/ 1236] Overall Loss 0.119662 Objective Loss 0.119662 LR 0.000125 Time 0.043443 +2023-10-02 21:56:47,964 - Epoch: [182][ 50/ 1236] Overall Loss 0.119262 Objective Loss 0.119262 LR 0.000125 Time 0.038880 +2023-10-02 21:56:48,172 - Epoch: [182][ 60/ 1236] Overall Loss 0.118842 Objective Loss 0.118842 LR 0.000125 Time 0.035855 +2023-10-02 21:56:48,378 - Epoch: [182][ 70/ 1236] Overall Loss 0.117980 Objective Loss 0.117980 LR 0.000125 Time 0.033679 +2023-10-02 21:56:48,586 - Epoch: [182][ 80/ 1236] Overall Loss 0.117733 Objective Loss 0.117733 LR 0.000125 Time 0.032061 +2023-10-02 21:56:48,791 - Epoch: [182][ 90/ 1236] Overall Loss 0.116249 Objective Loss 0.116249 LR 0.000125 Time 0.030774 +2023-10-02 21:56:48,999 - Epoch: [182][ 100/ 1236] Overall Loss 0.115554 Objective Loss 0.115554 LR 0.000125 Time 0.029771 +2023-10-02 21:56:49,205 - Epoch: [182][ 110/ 1236] Overall Loss 0.114117 Objective Loss 0.114117 LR 0.000125 Time 0.028927 +2023-10-02 21:56:49,413 - Epoch: [182][ 120/ 1236] Overall Loss 0.114440 Objective Loss 0.114440 LR 0.000125 Time 0.028244 +2023-10-02 21:56:49,619 - Epoch: [182][ 130/ 1236] Overall Loss 0.115838 Objective Loss 0.115838 LR 0.000125 Time 0.027647 +2023-10-02 21:56:49,827 - Epoch: [182][ 140/ 1236] Overall Loss 0.115368 Objective Loss 0.115368 LR 0.000125 Time 0.027157 +2023-10-02 21:56:50,033 - Epoch: [182][ 150/ 1236] Overall Loss 0.114334 Objective Loss 0.114334 LR 0.000125 Time 0.026712 +2023-10-02 21:56:50,241 - Epoch: [182][ 160/ 1236] Overall Loss 0.114317 Objective Loss 0.114317 LR 0.000125 Time 0.026340 +2023-10-02 21:56:50,448 - Epoch: [182][ 170/ 1236] Overall Loss 0.116290 Objective Loss 0.116290 LR 0.000125 Time 0.025995 +2023-10-02 21:56:50,656 - Epoch: [182][ 180/ 1236] Overall Loss 0.116259 Objective Loss 0.116259 LR 0.000125 Time 0.025705 +2023-10-02 21:56:50,862 - Epoch: [182][ 190/ 1236] Overall Loss 0.115966 Objective Loss 0.115966 LR 0.000125 Time 0.025430 +2023-10-02 21:56:51,071 - Epoch: [182][ 200/ 1236] Overall Loss 0.115413 Objective Loss 0.115413 LR 0.000125 Time 0.025204 +2023-10-02 21:56:51,277 - Epoch: [182][ 210/ 1236] Overall Loss 0.114924 Objective Loss 0.114924 LR 0.000125 Time 0.024982 +2023-10-02 21:56:51,485 - Epoch: [182][ 220/ 1236] Overall Loss 0.115160 Objective Loss 0.115160 LR 0.000125 Time 0.024788 +2023-10-02 21:56:51,691 - Epoch: [182][ 230/ 1236] Overall Loss 0.115046 Objective Loss 0.115046 LR 0.000125 Time 0.024601 +2023-10-02 21:56:51,899 - Epoch: [182][ 240/ 1236] Overall Loss 0.115224 Objective Loss 0.115224 LR 0.000125 Time 0.024440 +2023-10-02 21:56:52,105 - Epoch: [182][ 250/ 1236] Overall Loss 0.114761 Objective Loss 0.114761 LR 0.000125 Time 0.024282 +2023-10-02 21:56:52,313 - Epoch: [182][ 260/ 1236] Overall Loss 0.115544 Objective Loss 0.115544 LR 0.000125 Time 0.024147 +2023-10-02 21:56:52,519 - Epoch: [182][ 270/ 1236] Overall Loss 0.115418 Objective Loss 0.115418 LR 0.000125 Time 0.024011 +2023-10-02 21:56:52,728 - Epoch: [182][ 280/ 1236] Overall Loss 0.115325 Objective Loss 0.115325 LR 0.000125 Time 0.023896 +2023-10-02 21:56:52,934 - Epoch: [182][ 290/ 1236] Overall Loss 0.115448 Objective Loss 0.115448 LR 0.000125 Time 0.023778 +2023-10-02 21:56:53,142 - Epoch: [182][ 300/ 1236] Overall Loss 0.115484 Objective Loss 0.115484 LR 0.000125 Time 0.023678 +2023-10-02 21:56:53,349 - Epoch: [182][ 310/ 1236] Overall Loss 0.115388 Objective Loss 0.115388 LR 0.000125 Time 0.023577 +2023-10-02 21:56:53,557 - Epoch: [182][ 320/ 1236] Overall Loss 0.114846 Objective Loss 0.114846 LR 0.000125 Time 0.023490 +2023-10-02 21:56:53,764 - Epoch: [182][ 330/ 1236] Overall Loss 0.114589 Objective Loss 0.114589 LR 0.000125 Time 0.023400 +2023-10-02 21:56:53,973 - Epoch: [182][ 340/ 1236] Overall Loss 0.114783 Objective Loss 0.114783 LR 0.000125 Time 0.023327 +2023-10-02 21:56:54,179 - Epoch: [182][ 350/ 1236] Overall Loss 0.115012 Objective Loss 0.115012 LR 0.000125 Time 0.023247 +2023-10-02 21:56:54,387 - Epoch: [182][ 360/ 1236] Overall Loss 0.115010 Objective Loss 0.115010 LR 0.000125 Time 0.023178 +2023-10-02 21:56:54,593 - Epoch: [182][ 370/ 1236] Overall Loss 0.114918 Objective Loss 0.114918 LR 0.000125 Time 0.023106 +2023-10-02 21:56:54,802 - Epoch: [182][ 380/ 1236] Overall Loss 0.114904 Objective Loss 0.114904 LR 0.000125 Time 0.023045 +2023-10-02 21:56:55,008 - Epoch: [182][ 390/ 1236] Overall Loss 0.114244 Objective Loss 0.114244 LR 0.000125 Time 0.022984 +2023-10-02 21:56:55,216 - Epoch: [182][ 400/ 1236] Overall Loss 0.113941 Objective Loss 0.113941 LR 0.000125 Time 0.022928 +2023-10-02 21:56:55,423 - Epoch: [182][ 410/ 1236] Overall Loss 0.113616 Objective Loss 0.113616 LR 0.000125 Time 0.022873 +2023-10-02 21:56:55,631 - Epoch: [182][ 420/ 1236] Overall Loss 0.113709 Objective Loss 0.113709 LR 0.000125 Time 0.022822 +2023-10-02 21:56:55,838 - Epoch: [182][ 430/ 1236] Overall Loss 0.114158 Objective Loss 0.114158 LR 0.000125 Time 0.022772 +2023-10-02 21:56:56,046 - Epoch: [182][ 440/ 1236] Overall Loss 0.114139 Objective Loss 0.114139 LR 0.000125 Time 0.022727 +2023-10-02 21:56:56,253 - Epoch: [182][ 450/ 1236] Overall Loss 0.114275 Objective Loss 0.114275 LR 0.000125 Time 0.022681 +2023-10-02 21:56:56,461 - Epoch: [182][ 460/ 1236] Overall Loss 0.114191 Objective Loss 0.114191 LR 0.000125 Time 0.022640 +2023-10-02 21:56:56,667 - Epoch: [182][ 470/ 1236] Overall Loss 0.114225 Objective Loss 0.114225 LR 0.000125 Time 0.022594 +2023-10-02 21:56:56,875 - Epoch: [182][ 480/ 1236] Overall Loss 0.114238 Objective Loss 0.114238 LR 0.000125 Time 0.022557 +2023-10-02 21:56:57,081 - Epoch: [182][ 490/ 1236] Overall Loss 0.114344 Objective Loss 0.114344 LR 0.000125 Time 0.022518 +2023-10-02 21:56:57,290 - Epoch: [182][ 500/ 1236] Overall Loss 0.114488 Objective Loss 0.114488 LR 0.000125 Time 0.022483 +2023-10-02 21:56:57,496 - Epoch: [182][ 510/ 1236] Overall Loss 0.114476 Objective Loss 0.114476 LR 0.000125 Time 0.022447 +2023-10-02 21:56:57,703 - Epoch: [182][ 520/ 1236] Overall Loss 0.114966 Objective Loss 0.114966 LR 0.000125 Time 0.022413 +2023-10-02 21:56:57,910 - Epoch: [182][ 530/ 1236] Overall Loss 0.114334 Objective Loss 0.114334 LR 0.000125 Time 0.022379 +2023-10-02 21:56:58,118 - Epoch: [182][ 540/ 1236] Overall Loss 0.114212 Objective Loss 0.114212 LR 0.000125 Time 0.022350 +2023-10-02 21:56:58,325 - Epoch: [182][ 550/ 1236] Overall Loss 0.114169 Objective Loss 0.114169 LR 0.000125 Time 0.022320 +2023-10-02 21:56:58,533 - Epoch: [182][ 560/ 1236] Overall Loss 0.114360 Objective Loss 0.114360 LR 0.000125 Time 0.022292 +2023-10-02 21:56:58,740 - Epoch: [182][ 570/ 1236] Overall Loss 0.114846 Objective Loss 0.114846 LR 0.000125 Time 0.022264 +2023-10-02 21:56:58,949 - Epoch: [182][ 580/ 1236] Overall Loss 0.114667 Objective Loss 0.114667 LR 0.000125 Time 0.022238 +2023-10-02 21:56:59,155 - Epoch: [182][ 590/ 1236] Overall Loss 0.114718 Objective Loss 0.114718 LR 0.000125 Time 0.022211 +2023-10-02 21:56:59,364 - Epoch: [182][ 600/ 1236] Overall Loss 0.114554 Objective Loss 0.114554 LR 0.000125 Time 0.022187 +2023-10-02 21:56:59,570 - Epoch: [182][ 610/ 1236] Overall Loss 0.114602 Objective Loss 0.114602 LR 0.000125 Time 0.022162 +2023-10-02 21:56:59,778 - Epoch: [182][ 620/ 1236] Overall Loss 0.114956 Objective Loss 0.114956 LR 0.000125 Time 0.022140 +2023-10-02 21:56:59,986 - Epoch: [182][ 630/ 1236] Overall Loss 0.115156 Objective Loss 0.115156 LR 0.000125 Time 0.022117 +2023-10-02 21:57:00,194 - Epoch: [182][ 640/ 1236] Overall Loss 0.114883 Objective Loss 0.114883 LR 0.000125 Time 0.022096 +2023-10-02 21:57:00,399 - Epoch: [182][ 650/ 1236] Overall Loss 0.114786 Objective Loss 0.114786 LR 0.000125 Time 0.022071 +2023-10-02 21:57:00,607 - Epoch: [182][ 660/ 1236] Overall Loss 0.114523 Objective Loss 0.114523 LR 0.000125 Time 0.022052 +2023-10-02 21:57:00,814 - Epoch: [182][ 670/ 1236] Overall Loss 0.114512 Objective Loss 0.114512 LR 0.000125 Time 0.022032 +2023-10-02 21:57:01,023 - Epoch: [182][ 680/ 1236] Overall Loss 0.114365 Objective Loss 0.114365 LR 0.000125 Time 0.022013 +2023-10-02 21:57:01,228 - Epoch: [182][ 690/ 1236] Overall Loss 0.114191 Objective Loss 0.114191 LR 0.000125 Time 0.021991 +2023-10-02 21:57:01,436 - Epoch: [182][ 700/ 1236] Overall Loss 0.114186 Objective Loss 0.114186 LR 0.000125 Time 0.021974 +2023-10-02 21:57:01,643 - Epoch: [182][ 710/ 1236] Overall Loss 0.114092 Objective Loss 0.114092 LR 0.000125 Time 0.021955 +2023-10-02 21:57:01,851 - Epoch: [182][ 720/ 1236] Overall Loss 0.114104 Objective Loss 0.114104 LR 0.000125 Time 0.021939 +2023-10-02 21:57:02,058 - Epoch: [182][ 730/ 1236] Overall Loss 0.113944 Objective Loss 0.113944 LR 0.000125 Time 0.021921 +2023-10-02 21:57:02,266 - Epoch: [182][ 740/ 1236] Overall Loss 0.113826 Objective Loss 0.113826 LR 0.000125 Time 0.021906 +2023-10-02 21:57:02,473 - Epoch: [182][ 750/ 1236] Overall Loss 0.113946 Objective Loss 0.113946 LR 0.000125 Time 0.021889 +2023-10-02 21:57:02,681 - Epoch: [182][ 760/ 1236] Overall Loss 0.113914 Objective Loss 0.113914 LR 0.000125 Time 0.021875 +2023-10-02 21:57:02,888 - Epoch: [182][ 770/ 1236] Overall Loss 0.114061 Objective Loss 0.114061 LR 0.000125 Time 0.021859 +2023-10-02 21:57:03,096 - Epoch: [182][ 780/ 1236] Overall Loss 0.114008 Objective Loss 0.114008 LR 0.000125 Time 0.021845 +2023-10-02 21:57:03,303 - Epoch: [182][ 790/ 1236] Overall Loss 0.114007 Objective Loss 0.114007 LR 0.000125 Time 0.021830 +2023-10-02 21:57:03,511 - Epoch: [182][ 800/ 1236] Overall Loss 0.114163 Objective Loss 0.114163 LR 0.000125 Time 0.021817 +2023-10-02 21:57:03,718 - Epoch: [182][ 810/ 1236] Overall Loss 0.114434 Objective Loss 0.114434 LR 0.000125 Time 0.021803 +2023-10-02 21:57:03,926 - Epoch: [182][ 820/ 1236] Overall Loss 0.114394 Objective Loss 0.114394 LR 0.000125 Time 0.021790 +2023-10-02 21:57:04,133 - Epoch: [182][ 830/ 1236] Overall Loss 0.114472 Objective Loss 0.114472 LR 0.000125 Time 0.021777 +2023-10-02 21:57:04,341 - Epoch: [182][ 840/ 1236] Overall Loss 0.114472 Objective Loss 0.114472 LR 0.000125 Time 0.021765 +2023-10-02 21:57:04,548 - Epoch: [182][ 850/ 1236] Overall Loss 0.114567 Objective Loss 0.114567 LR 0.000125 Time 0.021753 +2023-10-02 21:57:04,756 - Epoch: [182][ 860/ 1236] Overall Loss 0.114579 Objective Loss 0.114579 LR 0.000125 Time 0.021741 +2023-10-02 21:57:04,968 - Epoch: [182][ 870/ 1236] Overall Loss 0.114500 Objective Loss 0.114500 LR 0.000125 Time 0.021734 +2023-10-02 21:57:05,182 - Epoch: [182][ 880/ 1236] Overall Loss 0.114699 Objective Loss 0.114699 LR 0.000125 Time 0.021729 +2023-10-02 21:57:05,400 - Epoch: [182][ 890/ 1236] Overall Loss 0.114750 Objective Loss 0.114750 LR 0.000125 Time 0.021730 +2023-10-02 21:57:05,613 - Epoch: [182][ 900/ 1236] Overall Loss 0.114779 Objective Loss 0.114779 LR 0.000125 Time 0.021725 +2023-10-02 21:57:05,831 - Epoch: [182][ 910/ 1236] Overall Loss 0.114703 Objective Loss 0.114703 LR 0.000125 Time 0.021726 +2023-10-02 21:57:06,045 - Epoch: [182][ 920/ 1236] Overall Loss 0.114685 Objective Loss 0.114685 LR 0.000125 Time 0.021721 +2023-10-02 21:57:06,263 - Epoch: [182][ 930/ 1236] Overall Loss 0.114874 Objective Loss 0.114874 LR 0.000125 Time 0.021722 +2023-10-02 21:57:06,476 - Epoch: [182][ 940/ 1236] Overall Loss 0.114956 Objective Loss 0.114956 LR 0.000125 Time 0.021718 +2023-10-02 21:57:06,695 - Epoch: [182][ 950/ 1236] Overall Loss 0.115017 Objective Loss 0.115017 LR 0.000125 Time 0.021718 +2023-10-02 21:57:06,908 - Epoch: [182][ 960/ 1236] Overall Loss 0.114961 Objective Loss 0.114961 LR 0.000125 Time 0.021714 +2023-10-02 21:57:07,126 - Epoch: [182][ 970/ 1236] Overall Loss 0.114926 Objective Loss 0.114926 LR 0.000125 Time 0.021715 +2023-10-02 21:57:07,340 - Epoch: [182][ 980/ 1236] Overall Loss 0.114992 Objective Loss 0.114992 LR 0.000125 Time 0.021711 +2023-10-02 21:57:07,557 - Epoch: [182][ 990/ 1236] Overall Loss 0.114847 Objective Loss 0.114847 LR 0.000125 Time 0.021711 +2023-10-02 21:57:07,771 - Epoch: [182][ 1000/ 1236] Overall Loss 0.115039 Objective Loss 0.115039 LR 0.000125 Time 0.021707 +2023-10-02 21:57:07,989 - Epoch: [182][ 1010/ 1236] Overall Loss 0.115074 Objective Loss 0.115074 LR 0.000125 Time 0.021708 +2023-10-02 21:57:08,202 - Epoch: [182][ 1020/ 1236] Overall Loss 0.115060 Objective Loss 0.115060 LR 0.000125 Time 0.021704 +2023-10-02 21:57:08,420 - Epoch: [182][ 1030/ 1236] Overall Loss 0.115071 Objective Loss 0.115071 LR 0.000125 Time 0.021704 +2023-10-02 21:57:08,634 - Epoch: [182][ 1040/ 1236] Overall Loss 0.115104 Objective Loss 0.115104 LR 0.000125 Time 0.021701 +2023-10-02 21:57:08,852 - Epoch: [182][ 1050/ 1236] Overall Loss 0.115074 Objective Loss 0.115074 LR 0.000125 Time 0.021701 +2023-10-02 21:57:09,065 - Epoch: [182][ 1060/ 1236] Overall Loss 0.115204 Objective Loss 0.115204 LR 0.000125 Time 0.021697 +2023-10-02 21:57:09,281 - Epoch: [182][ 1070/ 1236] Overall Loss 0.115380 Objective Loss 0.115380 LR 0.000125 Time 0.021696 +2023-10-02 21:57:09,489 - Epoch: [182][ 1080/ 1236] Overall Loss 0.115368 Objective Loss 0.115368 LR 0.000125 Time 0.021687 +2023-10-02 21:57:09,700 - Epoch: [182][ 1090/ 1236] Overall Loss 0.115422 Objective Loss 0.115422 LR 0.000125 Time 0.021681 +2023-10-02 21:57:09,908 - Epoch: [182][ 1100/ 1236] Overall Loss 0.115448 Objective Loss 0.115448 LR 0.000125 Time 0.021673 +2023-10-02 21:57:10,118 - Epoch: [182][ 1110/ 1236] Overall Loss 0.115397 Objective Loss 0.115397 LR 0.000125 Time 0.021667 +2023-10-02 21:57:10,326 - Epoch: [182][ 1120/ 1236] Overall Loss 0.115205 Objective Loss 0.115205 LR 0.000125 Time 0.021659 +2023-10-02 21:57:10,537 - Epoch: [182][ 1130/ 1236] Overall Loss 0.115211 Objective Loss 0.115211 LR 0.000125 Time 0.021653 +2023-10-02 21:57:10,744 - Epoch: [182][ 1140/ 1236] Overall Loss 0.115322 Objective Loss 0.115322 LR 0.000125 Time 0.021645 +2023-10-02 21:57:10,955 - Epoch: [182][ 1150/ 1236] Overall Loss 0.115343 Objective Loss 0.115343 LR 0.000125 Time 0.021640 +2023-10-02 21:57:11,162 - Epoch: [182][ 1160/ 1236] Overall Loss 0.115424 Objective Loss 0.115424 LR 0.000125 Time 0.021632 +2023-10-02 21:57:11,372 - Epoch: [182][ 1170/ 1236] Overall Loss 0.115328 Objective Loss 0.115328 LR 0.000125 Time 0.021626 +2023-10-02 21:57:11,580 - Epoch: [182][ 1180/ 1236] Overall Loss 0.115355 Objective Loss 0.115355 LR 0.000125 Time 0.021618 +2023-10-02 21:57:11,790 - Epoch: [182][ 1190/ 1236] Overall Loss 0.115346 Objective Loss 0.115346 LR 0.000125 Time 0.021613 +2023-10-02 21:57:11,997 - Epoch: [182][ 1200/ 1236] Overall Loss 0.115234 Objective Loss 0.115234 LR 0.000125 Time 0.021605 +2023-10-02 21:57:12,207 - Epoch: [182][ 1210/ 1236] Overall Loss 0.115249 Objective Loss 0.115249 LR 0.000125 Time 0.021600 +2023-10-02 21:57:12,415 - Epoch: [182][ 1220/ 1236] Overall Loss 0.115387 Objective Loss 0.115387 LR 0.000125 Time 0.021593 +2023-10-02 21:57:12,675 - Epoch: [182][ 1230/ 1236] Overall Loss 0.115556 Objective Loss 0.115556 LR 0.000125 Time 0.021629 +2023-10-02 21:57:12,796 - Epoch: [182][ 1236/ 1236] Overall Loss 0.115529 Objective Loss 0.115529 Top1 91.446029 Top5 98.981670 LR 0.000125 Time 0.021622 +2023-10-02 21:57:12,939 - --- validate (epoch=182)----------- +2023-10-02 21:57:12,939 - 29943 samples (256 per mini-batch) +2023-10-02 21:57:13,445 - Epoch: [182][ 10/ 117] Loss 0.313455 Top1 86.875000 Top5 98.671875 +2023-10-02 21:57:13,601 - Epoch: [182][ 20/ 117] Loss 0.306530 Top1 87.714844 Top5 98.750000 +2023-10-02 21:57:13,754 - Epoch: [182][ 30/ 117] Loss 0.304449 Top1 87.955729 Top5 98.789062 +2023-10-02 21:57:13,904 - Epoch: [182][ 40/ 117] Loss 0.297687 Top1 88.085938 Top5 98.759766 +2023-10-02 21:57:14,056 - Epoch: [182][ 50/ 117] Loss 0.298127 Top1 88.101562 Top5 98.734375 +2023-10-02 21:57:14,206 - Epoch: [182][ 60/ 117] Loss 0.298454 Top1 87.955729 Top5 98.717448 +2023-10-02 21:57:14,357 - Epoch: [182][ 70/ 117] Loss 0.297182 Top1 87.907366 Top5 98.750000 +2023-10-02 21:57:14,508 - Epoch: [182][ 80/ 117] Loss 0.303370 Top1 87.670898 Top5 98.720703 +2023-10-02 21:57:14,660 - Epoch: [182][ 90/ 117] Loss 0.297660 Top1 87.812500 Top5 98.736979 +2023-10-02 21:57:14,813 - Epoch: [182][ 100/ 117] Loss 0.299467 Top1 87.726562 Top5 98.773438 +2023-10-02 21:57:14,972 - Epoch: [182][ 110/ 117] Loss 0.303161 Top1 87.620739 Top5 98.732244 +2023-10-02 21:57:15,060 - Epoch: [182][ 117/ 117] Loss 0.303602 Top1 87.683265 Top5 98.730922 +2023-10-02 21:57:15,173 - ==> Top1: 87.683 Top5: 98.731 Loss: 0.304 + +2023-10-02 21:57:15,174 - ==> Confusion: +[[ 939 1 4 1 4 3 0 0 7 56 2 1 0 4 4 2 1 1 1 0 19] + [ 1 1068 2 0 4 15 1 24 0 0 0 0 0 0 0 2 1 0 5 3 5] + [ 1 1 978 8 0 1 16 10 0 2 0 1 6 3 1 3 2 2 10 1 10] + [ 0 3 13 994 0 1 0 3 2 0 2 0 4 3 23 1 0 4 14 1 21] + [ 21 5 2 1 969 5 0 0 0 11 1 0 0 5 8 5 9 0 0 1 7] + [ 2 34 0 1 4 1003 2 18 1 5 0 7 2 8 5 0 1 1 5 1 16] + [ 0 3 20 1 0 1 1141 6 0 0 3 1 0 0 0 2 0 1 1 5 6] + [ 1 11 11 0 5 23 7 1073 2 2 7 3 2 4 1 1 0 2 44 10 9] + [ 14 1 0 1 2 4 0 2 988 29 10 1 1 12 10 1 3 1 3 1 5] + [ 85 0 0 2 8 4 0 0 26 948 1 0 0 27 7 2 0 0 0 1 8] + [ 1 2 8 6 0 1 4 2 8 1 977 1 0 13 5 0 3 3 6 1 11] + [ 0 1 1 0 0 13 0 3 0 0 0 967 16 6 0 1 0 17 0 7 3] + [ 0 1 1 4 0 1 1 1 0 2 2 28 981 2 3 8 0 11 2 6 14] + [ 0 0 0 0 4 4 0 0 8 6 4 9 0 1059 4 0 0 1 0 1 19] + [ 10 0 3 16 3 1 0 0 23 1 0 0 2 3 1016 0 1 2 8 0 12] + [ 0 0 2 1 5 1 0 0 0 1 0 6 7 0 0 1067 14 13 2 7 8] + [ 0 16 2 0 4 5 0 0 2 0 1 5 0 2 3 8 1093 0 3 5 12] + [ 0 0 1 1 0 0 1 0 0 0 0 5 20 3 3 7 1 992 1 1 2] + [ 3 2 1 18 0 0 0 14 4 1 1 1 1 0 11 0 0 1 997 0 13] + [ 0 0 1 4 0 3 11 5 0 0 1 13 5 3 1 1 6 0 1 1089 8] + [ 91 135 105 75 45 112 35 79 71 51 132 80 282 218 103 37 58 51 104 125 5916]] + +2023-10-02 21:57:15,175 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:57:15,175 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:57:15,189 - + +2023-10-02 21:57:15,189 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:57:16,201 - Epoch: [183][ 10/ 1236] Overall Loss 0.123684 Objective Loss 0.123684 LR 0.000125 Time 0.101124 +2023-10-02 21:57:16,407 - Epoch: [183][ 20/ 1236] Overall Loss 0.116784 Objective Loss 0.116784 LR 0.000125 Time 0.060850 +2023-10-02 21:57:16,612 - Epoch: [183][ 30/ 1236] Overall Loss 0.115167 Objective Loss 0.115167 LR 0.000125 Time 0.047403 +2023-10-02 21:57:16,818 - Epoch: [183][ 40/ 1236] Overall Loss 0.115292 Objective Loss 0.115292 LR 0.000125 Time 0.040686 +2023-10-02 21:57:17,023 - Epoch: [183][ 50/ 1236] Overall Loss 0.112480 Objective Loss 0.112480 LR 0.000125 Time 0.036624 +2023-10-02 21:57:17,229 - Epoch: [183][ 60/ 1236] Overall Loss 0.113157 Objective Loss 0.113157 LR 0.000125 Time 0.033944 +2023-10-02 21:57:17,434 - Epoch: [183][ 70/ 1236] Overall Loss 0.112775 Objective Loss 0.112775 LR 0.000125 Time 0.032019 +2023-10-02 21:57:17,641 - Epoch: [183][ 80/ 1236] Overall Loss 0.111255 Objective Loss 0.111255 LR 0.000125 Time 0.030604 +2023-10-02 21:57:17,844 - Epoch: [183][ 90/ 1236] Overall Loss 0.111543 Objective Loss 0.111543 LR 0.000125 Time 0.029458 +2023-10-02 21:57:18,052 - Epoch: [183][ 100/ 1236] Overall Loss 0.112429 Objective Loss 0.112429 LR 0.000125 Time 0.028586 +2023-10-02 21:57:18,255 - Epoch: [183][ 110/ 1236] Overall Loss 0.112069 Objective Loss 0.112069 LR 0.000125 Time 0.027833 +2023-10-02 21:57:18,463 - Epoch: [183][ 120/ 1236] Overall Loss 0.110761 Objective Loss 0.110761 LR 0.000125 Time 0.027238 +2023-10-02 21:57:18,666 - Epoch: [183][ 130/ 1236] Overall Loss 0.110103 Objective Loss 0.110103 LR 0.000125 Time 0.026707 +2023-10-02 21:57:18,873 - Epoch: [183][ 140/ 1236] Overall Loss 0.109126 Objective Loss 0.109126 LR 0.000125 Time 0.026274 +2023-10-02 21:57:19,078 - Epoch: [183][ 150/ 1236] Overall Loss 0.108968 Objective Loss 0.108968 LR 0.000125 Time 0.025879 +2023-10-02 21:57:19,286 - Epoch: [183][ 160/ 1236] Overall Loss 0.109754 Objective Loss 0.109754 LR 0.000125 Time 0.025559 +2023-10-02 21:57:19,490 - Epoch: [183][ 170/ 1236] Overall Loss 0.110730 Objective Loss 0.110730 LR 0.000125 Time 0.025251 +2023-10-02 21:57:19,697 - Epoch: [183][ 180/ 1236] Overall Loss 0.111306 Objective Loss 0.111306 LR 0.000125 Time 0.024999 +2023-10-02 21:57:19,900 - Epoch: [183][ 190/ 1236] Overall Loss 0.111836 Objective Loss 0.111836 LR 0.000125 Time 0.024753 +2023-10-02 21:57:20,108 - Epoch: [183][ 200/ 1236] Overall Loss 0.111467 Objective Loss 0.111467 LR 0.000125 Time 0.024552 +2023-10-02 21:57:20,312 - Epoch: [183][ 210/ 1236] Overall Loss 0.111320 Objective Loss 0.111320 LR 0.000125 Time 0.024350 +2023-10-02 21:57:20,518 - Epoch: [183][ 220/ 1236] Overall Loss 0.110735 Objective Loss 0.110735 LR 0.000125 Time 0.024180 +2023-10-02 21:57:20,723 - Epoch: [183][ 230/ 1236] Overall Loss 0.111415 Objective Loss 0.111415 LR 0.000125 Time 0.024013 +2023-10-02 21:57:20,931 - Epoch: [183][ 240/ 1236] Overall Loss 0.112118 Objective Loss 0.112118 LR 0.000125 Time 0.023877 +2023-10-02 21:57:21,134 - Epoch: [183][ 250/ 1236] Overall Loss 0.111678 Objective Loss 0.111678 LR 0.000125 Time 0.023736 +2023-10-02 21:57:21,342 - Epoch: [183][ 260/ 1236] Overall Loss 0.112303 Objective Loss 0.112303 LR 0.000125 Time 0.023619 +2023-10-02 21:57:21,546 - Epoch: [183][ 270/ 1236] Overall Loss 0.112728 Objective Loss 0.112728 LR 0.000125 Time 0.023498 +2023-10-02 21:57:21,754 - Epoch: [183][ 280/ 1236] Overall Loss 0.113374 Objective Loss 0.113374 LR 0.000125 Time 0.023401 +2023-10-02 21:57:21,957 - Epoch: [183][ 290/ 1236] Overall Loss 0.113358 Objective Loss 0.113358 LR 0.000125 Time 0.023295 +2023-10-02 21:57:22,163 - Epoch: [183][ 300/ 1236] Overall Loss 0.113644 Objective Loss 0.113644 LR 0.000125 Time 0.023205 +2023-10-02 21:57:22,369 - Epoch: [183][ 310/ 1236] Overall Loss 0.112975 Objective Loss 0.112975 LR 0.000125 Time 0.023113 +2023-10-02 21:57:22,576 - Epoch: [183][ 320/ 1236] Overall Loss 0.113094 Objective Loss 0.113094 LR 0.000125 Time 0.023039 +2023-10-02 21:57:22,780 - Epoch: [183][ 330/ 1236] Overall Loss 0.113116 Objective Loss 0.113116 LR 0.000125 Time 0.022957 +2023-10-02 21:57:22,988 - Epoch: [183][ 340/ 1236] Overall Loss 0.113767 Objective Loss 0.113767 LR 0.000125 Time 0.022892 +2023-10-02 21:57:23,192 - Epoch: [183][ 350/ 1236] Overall Loss 0.114459 Objective Loss 0.114459 LR 0.000125 Time 0.022820 +2023-10-02 21:57:23,398 - Epoch: [183][ 360/ 1236] Overall Loss 0.114163 Objective Loss 0.114163 LR 0.000125 Time 0.022759 +2023-10-02 21:57:23,604 - Epoch: [183][ 370/ 1236] Overall Loss 0.114226 Objective Loss 0.114226 LR 0.000125 Time 0.022695 +2023-10-02 21:57:23,810 - Epoch: [183][ 380/ 1236] Overall Loss 0.113923 Objective Loss 0.113923 LR 0.000125 Time 0.022640 +2023-10-02 21:57:24,015 - Epoch: [183][ 390/ 1236] Overall Loss 0.113677 Objective Loss 0.113677 LR 0.000125 Time 0.022581 +2023-10-02 21:57:24,223 - Epoch: [183][ 400/ 1236] Overall Loss 0.113748 Objective Loss 0.113748 LR 0.000125 Time 0.022535 +2023-10-02 21:57:24,427 - Epoch: [183][ 410/ 1236] Overall Loss 0.113786 Objective Loss 0.113786 LR 0.000125 Time 0.022483 +2023-10-02 21:57:24,633 - Epoch: [183][ 420/ 1236] Overall Loss 0.114055 Objective Loss 0.114055 LR 0.000125 Time 0.022437 +2023-10-02 21:57:24,838 - Epoch: [183][ 430/ 1236] Overall Loss 0.113878 Objective Loss 0.113878 LR 0.000125 Time 0.022390 +2023-10-02 21:57:25,045 - Epoch: [183][ 440/ 1236] Overall Loss 0.113978 Objective Loss 0.113978 LR 0.000125 Time 0.022349 +2023-10-02 21:57:25,250 - Epoch: [183][ 450/ 1236] Overall Loss 0.113995 Objective Loss 0.113995 LR 0.000125 Time 0.022305 +2023-10-02 21:57:25,457 - Epoch: [183][ 460/ 1236] Overall Loss 0.113750 Objective Loss 0.113750 LR 0.000125 Time 0.022270 +2023-10-02 21:57:25,661 - Epoch: [183][ 470/ 1236] Overall Loss 0.113794 Objective Loss 0.113794 LR 0.000125 Time 0.022229 +2023-10-02 21:57:25,866 - Epoch: [183][ 480/ 1236] Overall Loss 0.114242 Objective Loss 0.114242 LR 0.000125 Time 0.022193 +2023-10-02 21:57:26,072 - Epoch: [183][ 490/ 1236] Overall Loss 0.114060 Objective Loss 0.114060 LR 0.000125 Time 0.022156 +2023-10-02 21:57:26,278 - Epoch: [183][ 500/ 1236] Overall Loss 0.114291 Objective Loss 0.114291 LR 0.000125 Time 0.022125 +2023-10-02 21:57:26,482 - Epoch: [183][ 510/ 1236] Overall Loss 0.114388 Objective Loss 0.114388 LR 0.000125 Time 0.022089 +2023-10-02 21:57:26,689 - Epoch: [183][ 520/ 1236] Overall Loss 0.114466 Objective Loss 0.114466 LR 0.000125 Time 0.022061 +2023-10-02 21:57:26,892 - Epoch: [183][ 530/ 1236] Overall Loss 0.114126 Objective Loss 0.114126 LR 0.000125 Time 0.022026 +2023-10-02 21:57:27,099 - Epoch: [183][ 540/ 1236] Overall Loss 0.114046 Objective Loss 0.114046 LR 0.000125 Time 0.022000 +2023-10-02 21:57:27,305 - Epoch: [183][ 550/ 1236] Overall Loss 0.113801 Objective Loss 0.113801 LR 0.000125 Time 0.021971 +2023-10-02 21:57:27,511 - Epoch: [183][ 560/ 1236] Overall Loss 0.113726 Objective Loss 0.113726 LR 0.000125 Time 0.021946 +2023-10-02 21:57:27,717 - Epoch: [183][ 570/ 1236] Overall Loss 0.113717 Objective Loss 0.113717 LR 0.000125 Time 0.021919 +2023-10-02 21:57:27,923 - Epoch: [183][ 580/ 1236] Overall Loss 0.113799 Objective Loss 0.113799 LR 0.000125 Time 0.021896 +2023-10-02 21:57:28,128 - Epoch: [183][ 590/ 1236] Overall Loss 0.113782 Objective Loss 0.113782 LR 0.000125 Time 0.021871 +2023-10-02 21:57:28,335 - Epoch: [183][ 600/ 1236] Overall Loss 0.113857 Objective Loss 0.113857 LR 0.000125 Time 0.021850 +2023-10-02 21:57:28,540 - Epoch: [183][ 610/ 1236] Overall Loss 0.113710 Objective Loss 0.113710 LR 0.000125 Time 0.021826 +2023-10-02 21:57:28,746 - Epoch: [183][ 620/ 1236] Overall Loss 0.113760 Objective Loss 0.113760 LR 0.000125 Time 0.021805 +2023-10-02 21:57:28,952 - Epoch: [183][ 630/ 1236] Overall Loss 0.113588 Objective Loss 0.113588 LR 0.000125 Time 0.021783 +2023-10-02 21:57:29,158 - Epoch: [183][ 640/ 1236] Overall Loss 0.113551 Objective Loss 0.113551 LR 0.000125 Time 0.021765 +2023-10-02 21:57:29,364 - Epoch: [183][ 650/ 1236] Overall Loss 0.113359 Objective Loss 0.113359 LR 0.000125 Time 0.021743 +2023-10-02 21:57:29,570 - Epoch: [183][ 660/ 1236] Overall Loss 0.113321 Objective Loss 0.113321 LR 0.000125 Time 0.021727 +2023-10-02 21:57:29,777 - Epoch: [183][ 670/ 1236] Overall Loss 0.113297 Objective Loss 0.113297 LR 0.000125 Time 0.021709 +2023-10-02 21:57:29,995 - Epoch: [183][ 680/ 1236] Overall Loss 0.113458 Objective Loss 0.113458 LR 0.000125 Time 0.021709 +2023-10-02 21:57:30,216 - Epoch: [183][ 690/ 1236] Overall Loss 0.113857 Objective Loss 0.113857 LR 0.000125 Time 0.021713 +2023-10-02 21:57:30,435 - Epoch: [183][ 700/ 1236] Overall Loss 0.113637 Objective Loss 0.113637 LR 0.000125 Time 0.021714 +2023-10-02 21:57:30,656 - Epoch: [183][ 710/ 1236] Overall Loss 0.113712 Objective Loss 0.113712 LR 0.000125 Time 0.021718 +2023-10-02 21:57:30,874 - Epoch: [183][ 720/ 1236] Overall Loss 0.113763 Objective Loss 0.113763 LR 0.000125 Time 0.021718 +2023-10-02 21:57:31,095 - Epoch: [183][ 730/ 1236] Overall Loss 0.113898 Objective Loss 0.113898 LR 0.000125 Time 0.021722 +2023-10-02 21:57:31,314 - Epoch: [183][ 740/ 1236] Overall Loss 0.113997 Objective Loss 0.113997 LR 0.000125 Time 0.021723 +2023-10-02 21:57:31,527 - Epoch: [183][ 750/ 1236] Overall Loss 0.114059 Objective Loss 0.114059 LR 0.000125 Time 0.021716 +2023-10-02 21:57:31,737 - Epoch: [183][ 760/ 1236] Overall Loss 0.114248 Objective Loss 0.114248 LR 0.000125 Time 0.021706 +2023-10-02 21:57:31,947 - Epoch: [183][ 770/ 1236] Overall Loss 0.114338 Objective Loss 0.114338 LR 0.000125 Time 0.021695 +2023-10-02 21:57:32,159 - Epoch: [183][ 780/ 1236] Overall Loss 0.114450 Objective Loss 0.114450 LR 0.000125 Time 0.021687 +2023-10-02 21:57:32,370 - Epoch: [183][ 790/ 1236] Overall Loss 0.114643 Objective Loss 0.114643 LR 0.000125 Time 0.021680 +2023-10-02 21:57:32,581 - Epoch: [183][ 800/ 1236] Overall Loss 0.114609 Objective Loss 0.114609 LR 0.000125 Time 0.021671 +2023-10-02 21:57:32,791 - Epoch: [183][ 810/ 1236] Overall Loss 0.114513 Objective Loss 0.114513 LR 0.000125 Time 0.021663 +2023-10-02 21:57:33,001 - Epoch: [183][ 820/ 1236] Overall Loss 0.114576 Objective Loss 0.114576 LR 0.000125 Time 0.021654 +2023-10-02 21:57:33,212 - Epoch: [183][ 830/ 1236] Overall Loss 0.114627 Objective Loss 0.114627 LR 0.000125 Time 0.021647 +2023-10-02 21:57:33,422 - Epoch: [183][ 840/ 1236] Overall Loss 0.114885 Objective Loss 0.114885 LR 0.000125 Time 0.021640 +2023-10-02 21:57:33,633 - Epoch: [183][ 850/ 1236] Overall Loss 0.114927 Objective Loss 0.114927 LR 0.000125 Time 0.021630 +2023-10-02 21:57:33,843 - Epoch: [183][ 860/ 1236] Overall Loss 0.115151 Objective Loss 0.115151 LR 0.000125 Time 0.021623 +2023-10-02 21:57:34,053 - Epoch: [183][ 870/ 1236] Overall Loss 0.115250 Objective Loss 0.115250 LR 0.000125 Time 0.021614 +2023-10-02 21:57:34,264 - Epoch: [183][ 880/ 1236] Overall Loss 0.115280 Objective Loss 0.115280 LR 0.000125 Time 0.021607 +2023-10-02 21:57:34,474 - Epoch: [183][ 890/ 1236] Overall Loss 0.115391 Objective Loss 0.115391 LR 0.000125 Time 0.021599 +2023-10-02 21:57:34,684 - Epoch: [183][ 900/ 1236] Overall Loss 0.115493 Objective Loss 0.115493 LR 0.000125 Time 0.021592 +2023-10-02 21:57:34,894 - Epoch: [183][ 910/ 1236] Overall Loss 0.115471 Objective Loss 0.115471 LR 0.000125 Time 0.021585 +2023-10-02 21:57:35,105 - Epoch: [183][ 920/ 1236] Overall Loss 0.115434 Objective Loss 0.115434 LR 0.000125 Time 0.021579 +2023-10-02 21:57:35,315 - Epoch: [183][ 930/ 1236] Overall Loss 0.115419 Objective Loss 0.115419 LR 0.000125 Time 0.021572 +2023-10-02 21:57:35,525 - Epoch: [183][ 940/ 1236] Overall Loss 0.115364 Objective Loss 0.115364 LR 0.000125 Time 0.021566 +2023-10-02 21:57:35,735 - Epoch: [183][ 950/ 1236] Overall Loss 0.115357 Objective Loss 0.115357 LR 0.000125 Time 0.021558 +2023-10-02 21:57:35,945 - Epoch: [183][ 960/ 1236] Overall Loss 0.115368 Objective Loss 0.115368 LR 0.000125 Time 0.021552 +2023-10-02 21:57:36,156 - Epoch: [183][ 970/ 1236] Overall Loss 0.115334 Objective Loss 0.115334 LR 0.000125 Time 0.021545 +2023-10-02 21:57:36,367 - Epoch: [183][ 980/ 1236] Overall Loss 0.115434 Objective Loss 0.115434 LR 0.000125 Time 0.021540 +2023-10-02 21:57:36,577 - Epoch: [183][ 990/ 1236] Overall Loss 0.115569 Objective Loss 0.115569 LR 0.000125 Time 0.021533 +2023-10-02 21:57:36,787 - Epoch: [183][ 1000/ 1236] Overall Loss 0.115552 Objective Loss 0.115552 LR 0.000125 Time 0.021528 +2023-10-02 21:57:36,997 - Epoch: [183][ 1010/ 1236] Overall Loss 0.115396 Objective Loss 0.115396 LR 0.000125 Time 0.021521 +2023-10-02 21:57:37,208 - Epoch: [183][ 1020/ 1236] Overall Loss 0.115276 Objective Loss 0.115276 LR 0.000125 Time 0.021516 +2023-10-02 21:57:37,419 - Epoch: [183][ 1030/ 1236] Overall Loss 0.115344 Objective Loss 0.115344 LR 0.000125 Time 0.021511 +2023-10-02 21:57:37,629 - Epoch: [183][ 1040/ 1236] Overall Loss 0.115202 Objective Loss 0.115202 LR 0.000125 Time 0.021506 +2023-10-02 21:57:37,839 - Epoch: [183][ 1050/ 1236] Overall Loss 0.114974 Objective Loss 0.114974 LR 0.000125 Time 0.021500 +2023-10-02 21:57:38,049 - Epoch: [183][ 1060/ 1236] Overall Loss 0.115037 Objective Loss 0.115037 LR 0.000125 Time 0.021495 +2023-10-02 21:57:38,260 - Epoch: [183][ 1070/ 1236] Overall Loss 0.115158 Objective Loss 0.115158 LR 0.000125 Time 0.021489 +2023-10-02 21:57:38,470 - Epoch: [183][ 1080/ 1236] Overall Loss 0.115226 Objective Loss 0.115226 LR 0.000125 Time 0.021485 +2023-10-02 21:57:38,680 - Epoch: [183][ 1090/ 1236] Overall Loss 0.115334 Objective Loss 0.115334 LR 0.000125 Time 0.021479 +2023-10-02 21:57:38,891 - Epoch: [183][ 1100/ 1236] Overall Loss 0.115291 Objective Loss 0.115291 LR 0.000125 Time 0.021474 +2023-10-02 21:57:39,101 - Epoch: [183][ 1110/ 1236] Overall Loss 0.115299 Objective Loss 0.115299 LR 0.000125 Time 0.021468 +2023-10-02 21:57:39,311 - Epoch: [183][ 1120/ 1236] Overall Loss 0.115322 Objective Loss 0.115322 LR 0.000125 Time 0.021464 +2023-10-02 21:57:39,522 - Epoch: [183][ 1130/ 1236] Overall Loss 0.115425 Objective Loss 0.115425 LR 0.000125 Time 0.021459 +2023-10-02 21:57:39,732 - Epoch: [183][ 1140/ 1236] Overall Loss 0.115274 Objective Loss 0.115274 LR 0.000125 Time 0.021455 +2023-10-02 21:57:39,942 - Epoch: [183][ 1150/ 1236] Overall Loss 0.115225 Objective Loss 0.115225 LR 0.000125 Time 0.021449 +2023-10-02 21:57:40,153 - Epoch: [183][ 1160/ 1236] Overall Loss 0.115163 Objective Loss 0.115163 LR 0.000125 Time 0.021446 +2023-10-02 21:57:40,363 - Epoch: [183][ 1170/ 1236] Overall Loss 0.115042 Objective Loss 0.115042 LR 0.000125 Time 0.021442 +2023-10-02 21:57:40,574 - Epoch: [183][ 1180/ 1236] Overall Loss 0.115008 Objective Loss 0.115008 LR 0.000125 Time 0.021438 +2023-10-02 21:57:40,785 - Epoch: [183][ 1190/ 1236] Overall Loss 0.114979 Objective Loss 0.114979 LR 0.000125 Time 0.021434 +2023-10-02 21:57:40,996 - Epoch: [183][ 1200/ 1236] Overall Loss 0.115093 Objective Loss 0.115093 LR 0.000125 Time 0.021431 +2023-10-02 21:57:41,206 - Epoch: [183][ 1210/ 1236] Overall Loss 0.115064 Objective Loss 0.115064 LR 0.000125 Time 0.021427 +2023-10-02 21:57:41,417 - Epoch: [183][ 1220/ 1236] Overall Loss 0.115148 Objective Loss 0.115148 LR 0.000125 Time 0.021424 +2023-10-02 21:57:41,680 - Epoch: [183][ 1230/ 1236] Overall Loss 0.115167 Objective Loss 0.115167 LR 0.000125 Time 0.021462 +2023-10-02 21:57:41,803 - Epoch: [183][ 1236/ 1236] Overall Loss 0.115146 Objective Loss 0.115146 Top1 94.093686 Top5 99.592668 LR 0.000125 Time 0.021457 +2023-10-02 21:57:41,946 - --- validate (epoch=183)----------- +2023-10-02 21:57:41,946 - 29943 samples (256 per mini-batch) +2023-10-02 21:57:42,442 - Epoch: [183][ 10/ 117] Loss 0.292448 Top1 87.187500 Top5 98.593750 +2023-10-02 21:57:42,598 - Epoch: [183][ 20/ 117] Loss 0.292236 Top1 87.363281 Top5 98.808594 +2023-10-02 21:57:42,750 - Epoch: [183][ 30/ 117] Loss 0.295519 Top1 87.343750 Top5 98.763021 +2023-10-02 21:57:42,905 - Epoch: [183][ 40/ 117] Loss 0.299974 Top1 87.255859 Top5 98.740234 +2023-10-02 21:57:43,055 - Epoch: [183][ 50/ 117] Loss 0.301514 Top1 87.343750 Top5 98.718750 +2023-10-02 21:57:43,208 - Epoch: [183][ 60/ 117] Loss 0.301155 Top1 87.441406 Top5 98.710938 +2023-10-02 21:57:43,359 - Epoch: [183][ 70/ 117] Loss 0.298733 Top1 87.511161 Top5 98.744420 +2023-10-02 21:57:43,513 - Epoch: [183][ 80/ 117] Loss 0.304444 Top1 87.519531 Top5 98.715820 +2023-10-02 21:57:43,664 - Epoch: [183][ 90/ 117] Loss 0.305721 Top1 87.534722 Top5 98.728299 +2023-10-02 21:57:43,818 - Epoch: [183][ 100/ 117] Loss 0.301332 Top1 87.714844 Top5 98.785156 +2023-10-02 21:57:43,976 - Epoch: [183][ 110/ 117] Loss 0.302216 Top1 87.730824 Top5 98.781960 +2023-10-02 21:57:44,065 - Epoch: [183][ 117/ 117] Loss 0.301898 Top1 87.679925 Top5 98.781017 +2023-10-02 21:57:44,205 - ==> Top1: 87.680 Top5: 98.781 Loss: 0.302 + +2023-10-02 21:57:44,206 - ==> Confusion: +[[ 938 1 5 0 2 3 0 0 2 67 2 1 1 2 6 1 2 0 1 0 16] + [ 1 1059 1 2 5 19 0 22 1 1 0 0 0 0 0 3 0 0 7 3 7] + [ 1 0 992 8 0 0 12 7 0 2 1 0 6 2 0 4 2 2 9 1 7] + [ 0 2 12 993 0 0 1 2 2 0 6 0 4 2 21 3 1 5 12 2 21] + [ 20 5 0 1 976 3 0 0 1 11 0 0 0 3 9 5 8 0 1 1 6] + [ 3 35 0 3 6 998 1 19 1 6 1 6 2 5 4 0 2 1 3 2 18] + [ 0 3 23 2 0 1 1136 5 0 0 3 1 0 0 0 3 0 1 1 7 5] + [ 2 10 15 0 4 20 4 1074 2 5 6 5 0 4 1 1 0 1 44 10 10] + [ 16 2 0 1 1 3 0 2 983 35 11 1 1 9 12 0 2 1 2 2 5] + [ 95 0 2 3 4 4 0 0 29 953 0 0 1 15 5 1 0 0 0 1 6] + [ 2 2 13 10 0 2 3 2 9 2 975 1 0 11 1 0 2 2 4 2 10] + [ 0 0 0 0 2 13 0 3 0 0 0 978 9 3 0 1 0 15 0 4 7] + [ 0 0 1 4 0 0 2 0 0 2 2 37 974 1 4 6 2 11 3 6 13] + [ 1 0 1 0 3 7 0 0 13 16 4 11 0 1038 4 0 0 1 0 1 19] + [ 10 0 6 21 3 0 0 0 19 4 1 0 2 2 1016 0 1 2 7 0 7] + [ 0 0 2 1 5 1 1 0 0 0 0 6 7 0 0 1069 16 10 1 9 6] + [ 1 13 2 0 6 7 0 0 1 0 0 4 0 2 4 7 1096 0 2 6 10] + [ 0 0 1 0 1 0 2 0 1 0 0 4 15 1 2 5 0 1000 0 2 4] + [ 3 4 4 14 0 0 0 17 3 1 1 0 1 0 6 0 0 0 1004 0 10] + [ 0 0 4 5 1 2 7 3 0 1 0 13 4 2 2 0 6 0 1 1094 7] + [ 107 111 125 86 63 92 28 81 75 63 126 83 262 198 112 43 65 52 94 132 5907]] + +2023-10-02 21:57:44,207 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:57:44,207 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:57:44,213 - + +2023-10-02 21:57:44,213 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:57:45,364 - Epoch: [184][ 10/ 1236] Overall Loss 0.111831 Objective Loss 0.111831 LR 0.000125 Time 0.114997 +2023-10-02 21:57:45,573 - Epoch: [184][ 20/ 1236] Overall Loss 0.103982 Objective Loss 0.103982 LR 0.000125 Time 0.067932 +2023-10-02 21:57:45,781 - Epoch: [184][ 30/ 1236] Overall Loss 0.104793 Objective Loss 0.104793 LR 0.000125 Time 0.052222 +2023-10-02 21:57:45,990 - Epoch: [184][ 40/ 1236] Overall Loss 0.113918 Objective Loss 0.113918 LR 0.000125 Time 0.044391 +2023-10-02 21:57:46,199 - Epoch: [184][ 50/ 1236] Overall Loss 0.113087 Objective Loss 0.113087 LR 0.000125 Time 0.039655 +2023-10-02 21:57:46,409 - Epoch: [184][ 60/ 1236] Overall Loss 0.115574 Objective Loss 0.115574 LR 0.000125 Time 0.036534 +2023-10-02 21:57:46,617 - Epoch: [184][ 70/ 1236] Overall Loss 0.115607 Objective Loss 0.115607 LR 0.000125 Time 0.034273 +2023-10-02 21:57:46,826 - Epoch: [184][ 80/ 1236] Overall Loss 0.115009 Objective Loss 0.115009 LR 0.000125 Time 0.032596 +2023-10-02 21:57:47,035 - Epoch: [184][ 90/ 1236] Overall Loss 0.117475 Objective Loss 0.117475 LR 0.000125 Time 0.031293 +2023-10-02 21:57:47,246 - Epoch: [184][ 100/ 1236] Overall Loss 0.115305 Objective Loss 0.115305 LR 0.000125 Time 0.030264 +2023-10-02 21:57:47,456 - Epoch: [184][ 110/ 1236] Overall Loss 0.113529 Objective Loss 0.113529 LR 0.000125 Time 0.029409 +2023-10-02 21:57:47,665 - Epoch: [184][ 120/ 1236] Overall Loss 0.113491 Objective Loss 0.113491 LR 0.000125 Time 0.028696 +2023-10-02 21:57:47,873 - Epoch: [184][ 130/ 1236] Overall Loss 0.111898 Objective Loss 0.111898 LR 0.000125 Time 0.028081 +2023-10-02 21:57:48,082 - Epoch: [184][ 140/ 1236] Overall Loss 0.111526 Objective Loss 0.111526 LR 0.000125 Time 0.027568 +2023-10-02 21:57:48,293 - Epoch: [184][ 150/ 1236] Overall Loss 0.111366 Objective Loss 0.111366 LR 0.000125 Time 0.027121 +2023-10-02 21:57:48,503 - Epoch: [184][ 160/ 1236] Overall Loss 0.110863 Objective Loss 0.110863 LR 0.000125 Time 0.026739 +2023-10-02 21:57:48,712 - Epoch: [184][ 170/ 1236] Overall Loss 0.111936 Objective Loss 0.111936 LR 0.000125 Time 0.026388 +2023-10-02 21:57:48,924 - Epoch: [184][ 180/ 1236] Overall Loss 0.112942 Objective Loss 0.112942 LR 0.000125 Time 0.026094 +2023-10-02 21:57:49,134 - Epoch: [184][ 190/ 1236] Overall Loss 0.112922 Objective Loss 0.112922 LR 0.000125 Time 0.025819 +2023-10-02 21:57:49,346 - Epoch: [184][ 200/ 1236] Overall Loss 0.112189 Objective Loss 0.112189 LR 0.000125 Time 0.025589 +2023-10-02 21:57:49,555 - Epoch: [184][ 210/ 1236] Overall Loss 0.112106 Objective Loss 0.112106 LR 0.000125 Time 0.025363 +2023-10-02 21:57:49,767 - Epoch: [184][ 220/ 1236] Overall Loss 0.112599 Objective Loss 0.112599 LR 0.000125 Time 0.025174 +2023-10-02 21:57:49,976 - Epoch: [184][ 230/ 1236] Overall Loss 0.113007 Objective Loss 0.113007 LR 0.000125 Time 0.024986 +2023-10-02 21:57:50,189 - Epoch: [184][ 240/ 1236] Overall Loss 0.113005 Objective Loss 0.113005 LR 0.000125 Time 0.024830 +2023-10-02 21:57:50,397 - Epoch: [184][ 250/ 1236] Overall Loss 0.113354 Objective Loss 0.113354 LR 0.000125 Time 0.024671 +2023-10-02 21:57:50,610 - Epoch: [184][ 260/ 1236] Overall Loss 0.112761 Objective Loss 0.112761 LR 0.000125 Time 0.024539 +2023-10-02 21:57:50,819 - Epoch: [184][ 270/ 1236] Overall Loss 0.112473 Objective Loss 0.112473 LR 0.000125 Time 0.024402 +2023-10-02 21:57:51,031 - Epoch: [184][ 280/ 1236] Overall Loss 0.113057 Objective Loss 0.113057 LR 0.000125 Time 0.024288 +2023-10-02 21:57:51,240 - Epoch: [184][ 290/ 1236] Overall Loss 0.113091 Objective Loss 0.113091 LR 0.000125 Time 0.024169 +2023-10-02 21:57:51,450 - Epoch: [184][ 300/ 1236] Overall Loss 0.113414 Objective Loss 0.113414 LR 0.000125 Time 0.024063 +2023-10-02 21:57:51,659 - Epoch: [184][ 310/ 1236] Overall Loss 0.113295 Objective Loss 0.113295 LR 0.000125 Time 0.023959 +2023-10-02 21:57:51,869 - Epoch: [184][ 320/ 1236] Overall Loss 0.113616 Objective Loss 0.113616 LR 0.000125 Time 0.023867 +2023-10-02 21:57:52,080 - Epoch: [184][ 330/ 1236] Overall Loss 0.114040 Objective Loss 0.114040 LR 0.000125 Time 0.023776 +2023-10-02 21:57:52,290 - Epoch: [184][ 340/ 1236] Overall Loss 0.113596 Objective Loss 0.113596 LR 0.000125 Time 0.023697 +2023-10-02 21:57:52,501 - Epoch: [184][ 350/ 1236] Overall Loss 0.113425 Objective Loss 0.113425 LR 0.000125 Time 0.023616 +2023-10-02 21:57:52,711 - Epoch: [184][ 360/ 1236] Overall Loss 0.114003 Objective Loss 0.114003 LR 0.000125 Time 0.023544 +2023-10-02 21:57:52,919 - Epoch: [184][ 370/ 1236] Overall Loss 0.114306 Objective Loss 0.114306 LR 0.000125 Time 0.023465 +2023-10-02 21:57:53,127 - Epoch: [184][ 380/ 1236] Overall Loss 0.114027 Objective Loss 0.114027 LR 0.000125 Time 0.023393 +2023-10-02 21:57:53,335 - Epoch: [184][ 390/ 1236] Overall Loss 0.114410 Objective Loss 0.114410 LR 0.000125 Time 0.023324 +2023-10-02 21:57:53,545 - Epoch: [184][ 400/ 1236] Overall Loss 0.114383 Objective Loss 0.114383 LR 0.000125 Time 0.023264 +2023-10-02 21:57:53,753 - Epoch: [184][ 410/ 1236] Overall Loss 0.114399 Objective Loss 0.114399 LR 0.000125 Time 0.023202 +2023-10-02 21:57:53,963 - Epoch: [184][ 420/ 1236] Overall Loss 0.114671 Objective Loss 0.114671 LR 0.000125 Time 0.023147 +2023-10-02 21:57:54,172 - Epoch: [184][ 430/ 1236] Overall Loss 0.114574 Objective Loss 0.114574 LR 0.000125 Time 0.023091 +2023-10-02 21:57:54,381 - Epoch: [184][ 440/ 1236] Overall Loss 0.114546 Objective Loss 0.114546 LR 0.000125 Time 0.023042 +2023-10-02 21:57:54,590 - Epoch: [184][ 450/ 1236] Overall Loss 0.114153 Objective Loss 0.114153 LR 0.000125 Time 0.022991 +2023-10-02 21:57:54,800 - Epoch: [184][ 460/ 1236] Overall Loss 0.114033 Objective Loss 0.114033 LR 0.000125 Time 0.022946 +2023-10-02 21:57:55,009 - Epoch: [184][ 470/ 1236] Overall Loss 0.114051 Objective Loss 0.114051 LR 0.000125 Time 0.022899 +2023-10-02 21:57:55,218 - Epoch: [184][ 480/ 1236] Overall Loss 0.113942 Objective Loss 0.113942 LR 0.000125 Time 0.022858 +2023-10-02 21:57:55,428 - Epoch: [184][ 490/ 1236] Overall Loss 0.113537 Objective Loss 0.113537 LR 0.000125 Time 0.022815 +2023-10-02 21:57:55,637 - Epoch: [184][ 500/ 1236] Overall Loss 0.113658 Objective Loss 0.113658 LR 0.000125 Time 0.022777 +2023-10-02 21:57:55,846 - Epoch: [184][ 510/ 1236] Overall Loss 0.113901 Objective Loss 0.113901 LR 0.000125 Time 0.022738 +2023-10-02 21:57:56,056 - Epoch: [184][ 520/ 1236] Overall Loss 0.113602 Objective Loss 0.113602 LR 0.000125 Time 0.022703 +2023-10-02 21:57:56,265 - Epoch: [184][ 530/ 1236] Overall Loss 0.113435 Objective Loss 0.113435 LR 0.000125 Time 0.022666 +2023-10-02 21:57:56,474 - Epoch: [184][ 540/ 1236] Overall Loss 0.113165 Objective Loss 0.113165 LR 0.000125 Time 0.022634 +2023-10-02 21:57:56,683 - Epoch: [184][ 550/ 1236] Overall Loss 0.112850 Objective Loss 0.112850 LR 0.000125 Time 0.022600 +2023-10-02 21:57:56,893 - Epoch: [184][ 560/ 1236] Overall Loss 0.112707 Objective Loss 0.112707 LR 0.000125 Time 0.022570 +2023-10-02 21:57:57,102 - Epoch: [184][ 570/ 1236] Overall Loss 0.112650 Objective Loss 0.112650 LR 0.000125 Time 0.022538 +2023-10-02 21:57:57,311 - Epoch: [184][ 580/ 1236] Overall Loss 0.112639 Objective Loss 0.112639 LR 0.000125 Time 0.022510 +2023-10-02 21:57:57,520 - Epoch: [184][ 590/ 1236] Overall Loss 0.112754 Objective Loss 0.112754 LR 0.000125 Time 0.022480 +2023-10-02 21:57:57,730 - Epoch: [184][ 600/ 1236] Overall Loss 0.112717 Objective Loss 0.112717 LR 0.000125 Time 0.022454 +2023-10-02 21:57:57,940 - Epoch: [184][ 610/ 1236] Overall Loss 0.112692 Objective Loss 0.112692 LR 0.000125 Time 0.022427 +2023-10-02 21:57:58,149 - Epoch: [184][ 620/ 1236] Overall Loss 0.112968 Objective Loss 0.112968 LR 0.000125 Time 0.022403 +2023-10-02 21:57:58,359 - Epoch: [184][ 630/ 1236] Overall Loss 0.113034 Objective Loss 0.113034 LR 0.000125 Time 0.022378 +2023-10-02 21:57:58,568 - Epoch: [184][ 640/ 1236] Overall Loss 0.112987 Objective Loss 0.112987 LR 0.000125 Time 0.022355 +2023-10-02 21:57:58,777 - Epoch: [184][ 650/ 1236] Overall Loss 0.113019 Objective Loss 0.113019 LR 0.000125 Time 0.022330 +2023-10-02 21:57:58,986 - Epoch: [184][ 660/ 1236] Overall Loss 0.113036 Objective Loss 0.113036 LR 0.000125 Time 0.022307 +2023-10-02 21:57:59,195 - Epoch: [184][ 670/ 1236] Overall Loss 0.113161 Objective Loss 0.113161 LR 0.000125 Time 0.022285 +2023-10-02 21:57:59,405 - Epoch: [184][ 680/ 1236] Overall Loss 0.113095 Objective Loss 0.113095 LR 0.000125 Time 0.022265 +2023-10-02 21:57:59,614 - Epoch: [184][ 690/ 1236] Overall Loss 0.113179 Objective Loss 0.113179 LR 0.000125 Time 0.022243 +2023-10-02 21:57:59,823 - Epoch: [184][ 700/ 1236] Overall Loss 0.113178 Objective Loss 0.113178 LR 0.000125 Time 0.022224 +2023-10-02 21:58:00,033 - Epoch: [184][ 710/ 1236] Overall Loss 0.113153 Objective Loss 0.113153 LR 0.000125 Time 0.022203 +2023-10-02 21:58:00,242 - Epoch: [184][ 720/ 1236] Overall Loss 0.113056 Objective Loss 0.113056 LR 0.000125 Time 0.022185 +2023-10-02 21:58:00,450 - Epoch: [184][ 730/ 1236] Overall Loss 0.113302 Objective Loss 0.113302 LR 0.000125 Time 0.022164 +2023-10-02 21:58:00,659 - Epoch: [184][ 740/ 1236] Overall Loss 0.113484 Objective Loss 0.113484 LR 0.000125 Time 0.022147 +2023-10-02 21:58:00,868 - Epoch: [184][ 750/ 1236] Overall Loss 0.113330 Objective Loss 0.113330 LR 0.000125 Time 0.022128 +2023-10-02 21:58:01,078 - Epoch: [184][ 760/ 1236] Overall Loss 0.113584 Objective Loss 0.113584 LR 0.000125 Time 0.022113 +2023-10-02 21:58:01,288 - Epoch: [184][ 770/ 1236] Overall Loss 0.113611 Objective Loss 0.113611 LR 0.000125 Time 0.022097 +2023-10-02 21:58:01,497 - Epoch: [184][ 780/ 1236] Overall Loss 0.113730 Objective Loss 0.113730 LR 0.000125 Time 0.022082 +2023-10-02 21:58:01,706 - Epoch: [184][ 790/ 1236] Overall Loss 0.113744 Objective Loss 0.113744 LR 0.000125 Time 0.022065 +2023-10-02 21:58:01,916 - Epoch: [184][ 800/ 1236] Overall Loss 0.113673 Objective Loss 0.113673 LR 0.000125 Time 0.022051 +2023-10-02 21:58:02,125 - Epoch: [184][ 810/ 1236] Overall Loss 0.113587 Objective Loss 0.113587 LR 0.000125 Time 0.022035 +2023-10-02 21:58:02,335 - Epoch: [184][ 820/ 1236] Overall Loss 0.113740 Objective Loss 0.113740 LR 0.000125 Time 0.022022 +2023-10-02 21:58:02,544 - Epoch: [184][ 830/ 1236] Overall Loss 0.113780 Objective Loss 0.113780 LR 0.000125 Time 0.022007 +2023-10-02 21:58:02,754 - Epoch: [184][ 840/ 1236] Overall Loss 0.113914 Objective Loss 0.113914 LR 0.000125 Time 0.021994 +2023-10-02 21:58:02,963 - Epoch: [184][ 850/ 1236] Overall Loss 0.113875 Objective Loss 0.113875 LR 0.000125 Time 0.021980 +2023-10-02 21:58:03,172 - Epoch: [184][ 860/ 1236] Overall Loss 0.113864 Objective Loss 0.113864 LR 0.000125 Time 0.021967 +2023-10-02 21:58:03,381 - Epoch: [184][ 870/ 1236] Overall Loss 0.113990 Objective Loss 0.113990 LR 0.000125 Time 0.021953 +2023-10-02 21:58:03,591 - Epoch: [184][ 880/ 1236] Overall Loss 0.113840 Objective Loss 0.113840 LR 0.000125 Time 0.021942 +2023-10-02 21:58:03,800 - Epoch: [184][ 890/ 1236] Overall Loss 0.114150 Objective Loss 0.114150 LR 0.000125 Time 0.021929 +2023-10-02 21:58:04,010 - Epoch: [184][ 900/ 1236] Overall Loss 0.114313 Objective Loss 0.114313 LR 0.000125 Time 0.021918 +2023-10-02 21:58:04,220 - Epoch: [184][ 910/ 1236] Overall Loss 0.114419 Objective Loss 0.114419 LR 0.000125 Time 0.021906 +2023-10-02 21:58:04,430 - Epoch: [184][ 920/ 1236] Overall Loss 0.114292 Objective Loss 0.114292 LR 0.000125 Time 0.021895 +2023-10-02 21:58:04,639 - Epoch: [184][ 930/ 1236] Overall Loss 0.114189 Objective Loss 0.114189 LR 0.000125 Time 0.021883 +2023-10-02 21:58:04,849 - Epoch: [184][ 940/ 1236] Overall Loss 0.114500 Objective Loss 0.114500 LR 0.000125 Time 0.021873 +2023-10-02 21:58:05,058 - Epoch: [184][ 950/ 1236] Overall Loss 0.114632 Objective Loss 0.114632 LR 0.000125 Time 0.021861 +2023-10-02 21:58:05,267 - Epoch: [184][ 960/ 1236] Overall Loss 0.114701 Objective Loss 0.114701 LR 0.000125 Time 0.021852 +2023-10-02 21:58:05,477 - Epoch: [184][ 970/ 1236] Overall Loss 0.114627 Objective Loss 0.114627 LR 0.000125 Time 0.021840 +2023-10-02 21:58:05,686 - Epoch: [184][ 980/ 1236] Overall Loss 0.114826 Objective Loss 0.114826 LR 0.000125 Time 0.021831 +2023-10-02 21:58:05,895 - Epoch: [184][ 990/ 1236] Overall Loss 0.114791 Objective Loss 0.114791 LR 0.000125 Time 0.021820 +2023-10-02 21:58:06,105 - Epoch: [184][ 1000/ 1236] Overall Loss 0.114764 Objective Loss 0.114764 LR 0.000125 Time 0.021812 +2023-10-02 21:58:06,314 - Epoch: [184][ 1010/ 1236] Overall Loss 0.114807 Objective Loss 0.114807 LR 0.000125 Time 0.021801 +2023-10-02 21:58:06,524 - Epoch: [184][ 1020/ 1236] Overall Loss 0.114914 Objective Loss 0.114914 LR 0.000125 Time 0.021793 +2023-10-02 21:58:06,733 - Epoch: [184][ 1030/ 1236] Overall Loss 0.114758 Objective Loss 0.114758 LR 0.000125 Time 0.021783 +2023-10-02 21:58:06,943 - Epoch: [184][ 1040/ 1236] Overall Loss 0.114898 Objective Loss 0.114898 LR 0.000125 Time 0.021775 +2023-10-02 21:58:07,152 - Epoch: [184][ 1050/ 1236] Overall Loss 0.114959 Objective Loss 0.114959 LR 0.000125 Time 0.021765 +2023-10-02 21:58:07,361 - Epoch: [184][ 1060/ 1236] Overall Loss 0.115131 Objective Loss 0.115131 LR 0.000125 Time 0.021757 +2023-10-02 21:58:07,571 - Epoch: [184][ 1070/ 1236] Overall Loss 0.115071 Objective Loss 0.115071 LR 0.000125 Time 0.021748 +2023-10-02 21:58:07,781 - Epoch: [184][ 1080/ 1236] Overall Loss 0.115004 Objective Loss 0.115004 LR 0.000125 Time 0.021741 +2023-10-02 21:58:07,991 - Epoch: [184][ 1090/ 1236] Overall Loss 0.115026 Objective Loss 0.115026 LR 0.000125 Time 0.021733 +2023-10-02 21:58:08,205 - Epoch: [184][ 1100/ 1236] Overall Loss 0.114943 Objective Loss 0.114943 LR 0.000125 Time 0.021730 +2023-10-02 21:58:08,417 - Epoch: [184][ 1110/ 1236] Overall Loss 0.114940 Objective Loss 0.114940 LR 0.000125 Time 0.021724 +2023-10-02 21:58:08,631 - Epoch: [184][ 1120/ 1236] Overall Loss 0.114956 Objective Loss 0.114956 LR 0.000125 Time 0.021721 +2023-10-02 21:58:08,843 - Epoch: [184][ 1130/ 1236] Overall Loss 0.114878 Objective Loss 0.114878 LR 0.000125 Time 0.021715 +2023-10-02 21:58:09,057 - Epoch: [184][ 1140/ 1236] Overall Loss 0.115070 Objective Loss 0.115070 LR 0.000125 Time 0.021712 +2023-10-02 21:58:09,269 - Epoch: [184][ 1150/ 1236] Overall Loss 0.115109 Objective Loss 0.115109 LR 0.000125 Time 0.021706 +2023-10-02 21:58:09,483 - Epoch: [184][ 1160/ 1236] Overall Loss 0.115094 Objective Loss 0.115094 LR 0.000125 Time 0.021703 +2023-10-02 21:58:09,695 - Epoch: [184][ 1170/ 1236] Overall Loss 0.115192 Objective Loss 0.115192 LR 0.000125 Time 0.021697 +2023-10-02 21:58:09,909 - Epoch: [184][ 1180/ 1236] Overall Loss 0.115242 Objective Loss 0.115242 LR 0.000125 Time 0.021695 +2023-10-02 21:58:10,121 - Epoch: [184][ 1190/ 1236] Overall Loss 0.115234 Objective Loss 0.115234 LR 0.000125 Time 0.021690 +2023-10-02 21:58:10,335 - Epoch: [184][ 1200/ 1236] Overall Loss 0.115320 Objective Loss 0.115320 LR 0.000125 Time 0.021687 +2023-10-02 21:58:10,547 - Epoch: [184][ 1210/ 1236] Overall Loss 0.115366 Objective Loss 0.115366 LR 0.000125 Time 0.021682 +2023-10-02 21:58:10,761 - Epoch: [184][ 1220/ 1236] Overall Loss 0.115350 Objective Loss 0.115350 LR 0.000125 Time 0.021680 +2023-10-02 21:58:11,028 - Epoch: [184][ 1230/ 1236] Overall Loss 0.115481 Objective Loss 0.115481 LR 0.000125 Time 0.021719 +2023-10-02 21:58:11,150 - Epoch: [184][ 1236/ 1236] Overall Loss 0.115494 Objective Loss 0.115494 Top1 91.242363 Top5 98.778004 LR 0.000125 Time 0.021712 +2023-10-02 21:58:11,312 - --- validate (epoch=184)----------- +2023-10-02 21:58:11,312 - 29943 samples (256 per mini-batch) +2023-10-02 21:58:11,791 - Epoch: [184][ 10/ 117] Loss 0.282079 Top1 87.304688 Top5 98.750000 +2023-10-02 21:58:11,938 - Epoch: [184][ 20/ 117] Loss 0.286590 Top1 87.382812 Top5 98.769531 +2023-10-02 21:58:12,084 - Epoch: [184][ 30/ 117] Loss 0.280580 Top1 87.851562 Top5 98.736979 +2023-10-02 21:58:12,231 - Epoch: [184][ 40/ 117] Loss 0.278010 Top1 87.861328 Top5 98.798828 +2023-10-02 21:58:12,377 - Epoch: [184][ 50/ 117] Loss 0.282368 Top1 87.937500 Top5 98.835938 +2023-10-02 21:58:12,523 - Epoch: [184][ 60/ 117] Loss 0.284784 Top1 87.786458 Top5 98.860677 +2023-10-02 21:58:12,670 - Epoch: [184][ 70/ 117] Loss 0.290954 Top1 87.734375 Top5 98.828125 +2023-10-02 21:58:12,817 - Epoch: [184][ 80/ 117] Loss 0.289212 Top1 87.822266 Top5 98.813477 +2023-10-02 21:58:12,963 - Epoch: [184][ 90/ 117] Loss 0.293741 Top1 87.703993 Top5 98.797743 +2023-10-02 21:58:13,110 - Epoch: [184][ 100/ 117] Loss 0.298291 Top1 87.535156 Top5 98.757812 +2023-10-02 21:58:13,264 - Epoch: [184][ 110/ 117] Loss 0.299999 Top1 87.517756 Top5 98.718040 +2023-10-02 21:58:13,354 - Epoch: [184][ 117/ 117] Loss 0.303423 Top1 87.429449 Top5 98.697525 +2023-10-02 21:58:13,490 - ==> Top1: 87.429 Top5: 98.698 Loss: 0.303 + +2023-10-02 21:58:13,491 - ==> Confusion: +[[ 945 0 3 2 4 2 0 0 4 56 2 1 1 1 6 1 3 0 1 0 18] + [ 0 1065 0 1 5 18 0 16 0 2 0 1 0 0 3 3 0 0 6 2 9] + [ 1 0 988 8 1 0 13 7 0 2 1 0 8 2 1 3 2 2 9 2 6] + [ 0 3 15 984 2 0 0 1 4 0 4 0 5 1 27 2 1 5 10 2 23] + [ 29 4 0 1 970 5 1 0 0 12 1 0 0 4 7 6 7 0 0 0 3] + [ 3 32 0 1 3 995 3 23 1 6 2 5 3 9 6 0 0 0 4 2 18] + [ 0 3 25 1 0 1 1142 3 0 0 3 2 0 0 0 2 0 0 1 4 4] + [ 2 12 16 1 7 20 5 1075 2 2 4 2 5 5 1 0 0 1 40 9 9] + [ 15 1 0 1 2 3 0 1 991 35 11 1 0 8 10 0 1 1 3 2 3] + [ 96 2 0 1 5 4 0 0 24 958 0 0 0 14 2 3 0 0 0 1 9] + [ 1 2 10 8 0 1 4 4 12 2 965 1 0 15 5 0 2 3 4 3 11] + [ 0 2 1 0 1 16 0 3 0 0 0 976 10 7 0 1 0 14 0 1 3] + [ 0 0 1 4 0 1 2 0 0 2 3 37 976 2 3 9 1 12 1 4 10] + [ 1 0 1 0 4 7 0 0 13 14 1 12 0 1040 4 0 0 0 0 1 21] + [ 10 0 5 16 4 1 0 0 25 1 1 0 1 1 1023 0 1 1 7 0 4] + [ 0 0 1 1 6 1 1 0 0 0 0 5 7 0 0 1072 17 9 2 9 3] + [ 1 16 1 0 4 7 1 0 1 0 0 4 0 2 4 9 1092 0 3 5 11] + [ 0 1 0 2 0 0 2 0 0 0 0 3 17 1 4 9 0 994 0 2 3] + [ 3 3 4 16 0 0 0 21 4 1 4 1 1 0 9 0 0 0 990 0 11] + [ 0 0 3 3 1 2 9 3 0 1 0 13 3 3 0 1 7 1 1 1091 10] + [ 107 130 128 75 55 109 36 73 81 59 131 89 295 216 117 43 54 49 102 109 5847]] + +2023-10-02 21:58:13,492 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:58:13,492 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:58:13,498 - + +2023-10-02 21:58:13,498 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:58:14,543 - Epoch: [185][ 10/ 1236] Overall Loss 0.111820 Objective Loss 0.111820 LR 0.000125 Time 0.104434 +2023-10-02 21:58:14,753 - Epoch: [185][ 20/ 1236] Overall Loss 0.109689 Objective Loss 0.109689 LR 0.000125 Time 0.062678 +2023-10-02 21:58:14,962 - Epoch: [185][ 30/ 1236] Overall Loss 0.110868 Objective Loss 0.110868 LR 0.000125 Time 0.048733 +2023-10-02 21:58:15,171 - Epoch: [185][ 40/ 1236] Overall Loss 0.111314 Objective Loss 0.111314 LR 0.000125 Time 0.041776 +2023-10-02 21:58:15,379 - Epoch: [185][ 50/ 1236] Overall Loss 0.111793 Objective Loss 0.111793 LR 0.000125 Time 0.037557 +2023-10-02 21:58:15,588 - Epoch: [185][ 60/ 1236] Overall Loss 0.113161 Objective Loss 0.113161 LR 0.000125 Time 0.034775 +2023-10-02 21:58:15,796 - Epoch: [185][ 70/ 1236] Overall Loss 0.114676 Objective Loss 0.114676 LR 0.000125 Time 0.032759 +2023-10-02 21:58:16,006 - Epoch: [185][ 80/ 1236] Overall Loss 0.114045 Objective Loss 0.114045 LR 0.000125 Time 0.031276 +2023-10-02 21:58:16,215 - Epoch: [185][ 90/ 1236] Overall Loss 0.113903 Objective Loss 0.113903 LR 0.000125 Time 0.030106 +2023-10-02 21:58:16,425 - Epoch: [185][ 100/ 1236] Overall Loss 0.113473 Objective Loss 0.113473 LR 0.000125 Time 0.029197 +2023-10-02 21:58:16,633 - Epoch: [185][ 110/ 1236] Overall Loss 0.113015 Objective Loss 0.113015 LR 0.000125 Time 0.028424 +2023-10-02 21:58:16,843 - Epoch: [185][ 120/ 1236] Overall Loss 0.113411 Objective Loss 0.113411 LR 0.000125 Time 0.027801 +2023-10-02 21:58:17,052 - Epoch: [185][ 130/ 1236] Overall Loss 0.114188 Objective Loss 0.114188 LR 0.000125 Time 0.027254 +2023-10-02 21:58:17,262 - Epoch: [185][ 140/ 1236] Overall Loss 0.112740 Objective Loss 0.112740 LR 0.000125 Time 0.026803 +2023-10-02 21:58:17,470 - Epoch: [185][ 150/ 1236] Overall Loss 0.113991 Objective Loss 0.113991 LR 0.000125 Time 0.026397 +2023-10-02 21:58:17,680 - Epoch: [185][ 160/ 1236] Overall Loss 0.112841 Objective Loss 0.112841 LR 0.000125 Time 0.026058 +2023-10-02 21:58:17,889 - Epoch: [185][ 170/ 1236] Overall Loss 0.112156 Objective Loss 0.112156 LR 0.000125 Time 0.025741 +2023-10-02 21:58:18,098 - Epoch: [185][ 180/ 1236] Overall Loss 0.111900 Objective Loss 0.111900 LR 0.000125 Time 0.025475 +2023-10-02 21:58:18,307 - Epoch: [185][ 190/ 1236] Overall Loss 0.112287 Objective Loss 0.112287 LR 0.000125 Time 0.025225 +2023-10-02 21:58:18,517 - Epoch: [185][ 200/ 1236] Overall Loss 0.112964 Objective Loss 0.112964 LR 0.000125 Time 0.025010 +2023-10-02 21:58:18,725 - Epoch: [185][ 210/ 1236] Overall Loss 0.112747 Objective Loss 0.112747 LR 0.000125 Time 0.024804 +2023-10-02 21:58:18,935 - Epoch: [185][ 220/ 1236] Overall Loss 0.112906 Objective Loss 0.112906 LR 0.000125 Time 0.024630 +2023-10-02 21:58:19,144 - Epoch: [185][ 230/ 1236] Overall Loss 0.112937 Objective Loss 0.112937 LR 0.000125 Time 0.024461 +2023-10-02 21:58:19,354 - Epoch: [185][ 240/ 1236] Overall Loss 0.113073 Objective Loss 0.113073 LR 0.000125 Time 0.024316 +2023-10-02 21:58:19,563 - Epoch: [185][ 250/ 1236] Overall Loss 0.113223 Objective Loss 0.113223 LR 0.000125 Time 0.024172 +2023-10-02 21:58:19,773 - Epoch: [185][ 260/ 1236] Overall Loss 0.112966 Objective Loss 0.112966 LR 0.000125 Time 0.024047 +2023-10-02 21:58:19,981 - Epoch: [185][ 270/ 1236] Overall Loss 0.113310 Objective Loss 0.113310 LR 0.000125 Time 0.023923 +2023-10-02 21:58:20,191 - Epoch: [185][ 280/ 1236] Overall Loss 0.113196 Objective Loss 0.113196 LR 0.000125 Time 0.023816 +2023-10-02 21:58:20,399 - Epoch: [185][ 290/ 1236] Overall Loss 0.113032 Objective Loss 0.113032 LR 0.000125 Time 0.023708 +2023-10-02 21:58:20,610 - Epoch: [185][ 300/ 1236] Overall Loss 0.113249 Objective Loss 0.113249 LR 0.000125 Time 0.023618 +2023-10-02 21:58:20,819 - Epoch: [185][ 310/ 1236] Overall Loss 0.113932 Objective Loss 0.113932 LR 0.000125 Time 0.023532 +2023-10-02 21:58:21,031 - Epoch: [185][ 320/ 1236] Overall Loss 0.114089 Objective Loss 0.114089 LR 0.000125 Time 0.023456 +2023-10-02 21:58:21,240 - Epoch: [185][ 330/ 1236] Overall Loss 0.114369 Objective Loss 0.114369 LR 0.000125 Time 0.023374 +2023-10-02 21:58:21,451 - Epoch: [185][ 340/ 1236] Overall Loss 0.113827 Objective Loss 0.113827 LR 0.000125 Time 0.023306 +2023-10-02 21:58:21,660 - Epoch: [185][ 350/ 1236] Overall Loss 0.113494 Objective Loss 0.113494 LR 0.000125 Time 0.023233 +2023-10-02 21:58:21,873 - Epoch: [185][ 360/ 1236] Overall Loss 0.113330 Objective Loss 0.113330 LR 0.000125 Time 0.023179 +2023-10-02 21:58:22,081 - Epoch: [185][ 370/ 1236] Overall Loss 0.113157 Objective Loss 0.113157 LR 0.000125 Time 0.023113 +2023-10-02 21:58:22,293 - Epoch: [185][ 380/ 1236] Overall Loss 0.113472 Objective Loss 0.113472 LR 0.000125 Time 0.023062 +2023-10-02 21:58:22,502 - Epoch: [185][ 390/ 1236] Overall Loss 0.113470 Objective Loss 0.113470 LR 0.000125 Time 0.023007 +2023-10-02 21:58:22,715 - Epoch: [185][ 400/ 1236] Overall Loss 0.113559 Objective Loss 0.113559 LR 0.000125 Time 0.022960 +2023-10-02 21:58:22,925 - Epoch: [185][ 410/ 1236] Overall Loss 0.113271 Objective Loss 0.113271 LR 0.000125 Time 0.022908 +2023-10-02 21:58:23,136 - Epoch: [185][ 420/ 1236] Overall Loss 0.113455 Objective Loss 0.113455 LR 0.000125 Time 0.022865 +2023-10-02 21:58:23,346 - Epoch: [185][ 430/ 1236] Overall Loss 0.113220 Objective Loss 0.113220 LR 0.000125 Time 0.022818 +2023-10-02 21:58:23,558 - Epoch: [185][ 440/ 1236] Overall Loss 0.113295 Objective Loss 0.113295 LR 0.000125 Time 0.022780 +2023-10-02 21:58:23,767 - Epoch: [185][ 450/ 1236] Overall Loss 0.113222 Objective Loss 0.113222 LR 0.000125 Time 0.022738 +2023-10-02 21:58:23,978 - Epoch: [185][ 460/ 1236] Overall Loss 0.113360 Objective Loss 0.113360 LR 0.000125 Time 0.022701 +2023-10-02 21:58:24,188 - Epoch: [185][ 470/ 1236] Overall Loss 0.113218 Objective Loss 0.113218 LR 0.000125 Time 0.022662 +2023-10-02 21:58:24,401 - Epoch: [185][ 480/ 1236] Overall Loss 0.113507 Objective Loss 0.113507 LR 0.000125 Time 0.022633 +2023-10-02 21:58:24,613 - Epoch: [185][ 490/ 1236] Overall Loss 0.113617 Objective Loss 0.113617 LR 0.000125 Time 0.022601 +2023-10-02 21:58:24,826 - Epoch: [185][ 500/ 1236] Overall Loss 0.113367 Objective Loss 0.113367 LR 0.000125 Time 0.022575 +2023-10-02 21:58:25,039 - Epoch: [185][ 510/ 1236] Overall Loss 0.113226 Objective Loss 0.113226 LR 0.000125 Time 0.022547 +2023-10-02 21:58:25,251 - Epoch: [185][ 520/ 1236] Overall Loss 0.113130 Objective Loss 0.113130 LR 0.000125 Time 0.022521 +2023-10-02 21:58:25,466 - Epoch: [185][ 530/ 1236] Overall Loss 0.113350 Objective Loss 0.113350 LR 0.000125 Time 0.022500 +2023-10-02 21:58:25,679 - Epoch: [185][ 540/ 1236] Overall Loss 0.113366 Objective Loss 0.113366 LR 0.000125 Time 0.022476 +2023-10-02 21:58:25,890 - Epoch: [185][ 550/ 1236] Overall Loss 0.113449 Objective Loss 0.113449 LR 0.000125 Time 0.022449 +2023-10-02 21:58:26,102 - Epoch: [185][ 560/ 1236] Overall Loss 0.113315 Objective Loss 0.113315 LR 0.000125 Time 0.022425 +2023-10-02 21:58:26,314 - Epoch: [185][ 570/ 1236] Overall Loss 0.113143 Objective Loss 0.113143 LR 0.000125 Time 0.022401 +2023-10-02 21:58:26,526 - Epoch: [185][ 580/ 1236] Overall Loss 0.113127 Objective Loss 0.113127 LR 0.000125 Time 0.022380 +2023-10-02 21:58:26,738 - Epoch: [185][ 590/ 1236] Overall Loss 0.113369 Objective Loss 0.113369 LR 0.000125 Time 0.022357 +2023-10-02 21:58:26,950 - Epoch: [185][ 600/ 1236] Overall Loss 0.113715 Objective Loss 0.113715 LR 0.000125 Time 0.022338 +2023-10-02 21:58:27,162 - Epoch: [185][ 610/ 1236] Overall Loss 0.113706 Objective Loss 0.113706 LR 0.000125 Time 0.022316 +2023-10-02 21:58:27,375 - Epoch: [185][ 620/ 1236] Overall Loss 0.113764 Objective Loss 0.113764 LR 0.000125 Time 0.022298 +2023-10-02 21:58:27,587 - Epoch: [185][ 630/ 1236] Overall Loss 0.113800 Objective Loss 0.113800 LR 0.000125 Time 0.022279 +2023-10-02 21:58:27,799 - Epoch: [185][ 640/ 1236] Overall Loss 0.114071 Objective Loss 0.114071 LR 0.000125 Time 0.022261 +2023-10-02 21:58:28,011 - Epoch: [185][ 650/ 1236] Overall Loss 0.114070 Objective Loss 0.114070 LR 0.000125 Time 0.022243 +2023-10-02 21:58:28,223 - Epoch: [185][ 660/ 1236] Overall Loss 0.114151 Objective Loss 0.114151 LR 0.000125 Time 0.022226 +2023-10-02 21:58:28,434 - Epoch: [185][ 670/ 1236] Overall Loss 0.114084 Objective Loss 0.114084 LR 0.000125 Time 0.022207 +2023-10-02 21:58:28,646 - Epoch: [185][ 680/ 1236] Overall Loss 0.113997 Objective Loss 0.113997 LR 0.000125 Time 0.022192 +2023-10-02 21:58:28,858 - Epoch: [185][ 690/ 1236] Overall Loss 0.114097 Objective Loss 0.114097 LR 0.000125 Time 0.022175 +2023-10-02 21:58:29,070 - Epoch: [185][ 700/ 1236] Overall Loss 0.114210 Objective Loss 0.114210 LR 0.000125 Time 0.022160 +2023-10-02 21:58:29,282 - Epoch: [185][ 710/ 1236] Overall Loss 0.114224 Objective Loss 0.114224 LR 0.000125 Time 0.022145 +2023-10-02 21:58:29,494 - Epoch: [185][ 720/ 1236] Overall Loss 0.114009 Objective Loss 0.114009 LR 0.000125 Time 0.022132 +2023-10-02 21:58:29,706 - Epoch: [185][ 730/ 1236] Overall Loss 0.113915 Objective Loss 0.113915 LR 0.000125 Time 0.022117 +2023-10-02 21:58:29,918 - Epoch: [185][ 740/ 1236] Overall Loss 0.114166 Objective Loss 0.114166 LR 0.000125 Time 0.022104 +2023-10-02 21:58:30,131 - Epoch: [185][ 750/ 1236] Overall Loss 0.114088 Objective Loss 0.114088 LR 0.000125 Time 0.022090 +2023-10-02 21:58:30,343 - Epoch: [185][ 760/ 1236] Overall Loss 0.114241 Objective Loss 0.114241 LR 0.000125 Time 0.022079 +2023-10-02 21:58:30,555 - Epoch: [185][ 770/ 1236] Overall Loss 0.114430 Objective Loss 0.114430 LR 0.000125 Time 0.022065 +2023-10-02 21:58:30,767 - Epoch: [185][ 780/ 1236] Overall Loss 0.114443 Objective Loss 0.114443 LR 0.000125 Time 0.022053 +2023-10-02 21:58:30,980 - Epoch: [185][ 790/ 1236] Overall Loss 0.114562 Objective Loss 0.114562 LR 0.000125 Time 0.022043 +2023-10-02 21:58:31,192 - Epoch: [185][ 800/ 1236] Overall Loss 0.114505 Objective Loss 0.114505 LR 0.000125 Time 0.022032 +2023-10-02 21:58:31,404 - Epoch: [185][ 810/ 1236] Overall Loss 0.114442 Objective Loss 0.114442 LR 0.000125 Time 0.022020 +2023-10-02 21:58:31,616 - Epoch: [185][ 820/ 1236] Overall Loss 0.114354 Objective Loss 0.114354 LR 0.000125 Time 0.022010 +2023-10-02 21:58:31,828 - Epoch: [185][ 830/ 1236] Overall Loss 0.114375 Objective Loss 0.114375 LR 0.000125 Time 0.021999 +2023-10-02 21:58:32,040 - Epoch: [185][ 840/ 1236] Overall Loss 0.114243 Objective Loss 0.114243 LR 0.000125 Time 0.021988 +2023-10-02 21:58:32,251 - Epoch: [185][ 850/ 1236] Overall Loss 0.114288 Objective Loss 0.114288 LR 0.000125 Time 0.021977 +2023-10-02 21:58:32,457 - Epoch: [185][ 860/ 1236] Overall Loss 0.114228 Objective Loss 0.114228 LR 0.000125 Time 0.021962 +2023-10-02 21:58:32,668 - Epoch: [185][ 870/ 1236] Overall Loss 0.114083 Objective Loss 0.114083 LR 0.000125 Time 0.021951 +2023-10-02 21:58:32,875 - Epoch: [185][ 880/ 1236] Overall Loss 0.113931 Objective Loss 0.113931 LR 0.000125 Time 0.021936 +2023-10-02 21:58:33,085 - Epoch: [185][ 890/ 1236] Overall Loss 0.113923 Objective Loss 0.113923 LR 0.000125 Time 0.021926 +2023-10-02 21:58:33,292 - Epoch: [185][ 900/ 1236] Overall Loss 0.113903 Objective Loss 0.113903 LR 0.000125 Time 0.021911 +2023-10-02 21:58:33,502 - Epoch: [185][ 910/ 1236] Overall Loss 0.113929 Objective Loss 0.113929 LR 0.000125 Time 0.021901 +2023-10-02 21:58:33,709 - Epoch: [185][ 920/ 1236] Overall Loss 0.113910 Objective Loss 0.113910 LR 0.000125 Time 0.021888 +2023-10-02 21:58:33,919 - Epoch: [185][ 930/ 1236] Overall Loss 0.113787 Objective Loss 0.113787 LR 0.000125 Time 0.021878 +2023-10-02 21:58:34,127 - Epoch: [185][ 940/ 1236] Overall Loss 0.113778 Objective Loss 0.113778 LR 0.000125 Time 0.021866 +2023-10-02 21:58:34,337 - Epoch: [185][ 950/ 1236] Overall Loss 0.113839 Objective Loss 0.113839 LR 0.000125 Time 0.021856 +2023-10-02 21:58:34,543 - Epoch: [185][ 960/ 1236] Overall Loss 0.113812 Objective Loss 0.113812 LR 0.000125 Time 0.021843 +2023-10-02 21:58:34,753 - Epoch: [185][ 970/ 1236] Overall Loss 0.113758 Objective Loss 0.113758 LR 0.000125 Time 0.021834 +2023-10-02 21:58:34,960 - Epoch: [185][ 980/ 1236] Overall Loss 0.113652 Objective Loss 0.113652 LR 0.000125 Time 0.021822 +2023-10-02 21:58:35,171 - Epoch: [185][ 990/ 1236] Overall Loss 0.113593 Objective Loss 0.113593 LR 0.000125 Time 0.021814 +2023-10-02 21:58:35,378 - Epoch: [185][ 1000/ 1236] Overall Loss 0.113496 Objective Loss 0.113496 LR 0.000125 Time 0.021803 +2023-10-02 21:58:35,588 - Epoch: [185][ 1010/ 1236] Overall Loss 0.113390 Objective Loss 0.113390 LR 0.000125 Time 0.021795 +2023-10-02 21:58:35,795 - Epoch: [185][ 1020/ 1236] Overall Loss 0.113308 Objective Loss 0.113308 LR 0.000125 Time 0.021784 +2023-10-02 21:58:36,005 - Epoch: [185][ 1030/ 1236] Overall Loss 0.113340 Objective Loss 0.113340 LR 0.000125 Time 0.021776 +2023-10-02 21:58:36,212 - Epoch: [185][ 1040/ 1236] Overall Loss 0.113297 Objective Loss 0.113297 LR 0.000125 Time 0.021766 +2023-10-02 21:58:36,422 - Epoch: [185][ 1050/ 1236] Overall Loss 0.113503 Objective Loss 0.113503 LR 0.000125 Time 0.021758 +2023-10-02 21:58:36,629 - Epoch: [185][ 1060/ 1236] Overall Loss 0.113638 Objective Loss 0.113638 LR 0.000125 Time 0.021748 +2023-10-02 21:58:36,840 - Epoch: [185][ 1070/ 1236] Overall Loss 0.113611 Objective Loss 0.113611 LR 0.000125 Time 0.021741 +2023-10-02 21:58:37,046 - Epoch: [185][ 1080/ 1236] Overall Loss 0.113640 Objective Loss 0.113640 LR 0.000125 Time 0.021731 +2023-10-02 21:58:37,257 - Epoch: [185][ 1090/ 1236] Overall Loss 0.113665 Objective Loss 0.113665 LR 0.000125 Time 0.021724 +2023-10-02 21:58:37,464 - Epoch: [185][ 1100/ 1236] Overall Loss 0.113774 Objective Loss 0.113774 LR 0.000125 Time 0.021715 +2023-10-02 21:58:37,674 - Epoch: [185][ 1110/ 1236] Overall Loss 0.113769 Objective Loss 0.113769 LR 0.000125 Time 0.021708 +2023-10-02 21:58:37,881 - Epoch: [185][ 1120/ 1236] Overall Loss 0.113750 Objective Loss 0.113750 LR 0.000125 Time 0.021699 +2023-10-02 21:58:38,091 - Epoch: [185][ 1130/ 1236] Overall Loss 0.113698 Objective Loss 0.113698 LR 0.000125 Time 0.021693 +2023-10-02 21:58:38,299 - Epoch: [185][ 1140/ 1236] Overall Loss 0.113758 Objective Loss 0.113758 LR 0.000125 Time 0.021684 +2023-10-02 21:58:38,509 - Epoch: [185][ 1150/ 1236] Overall Loss 0.113754 Objective Loss 0.113754 LR 0.000125 Time 0.021678 +2023-10-02 21:58:38,716 - Epoch: [185][ 1160/ 1236] Overall Loss 0.113783 Objective Loss 0.113783 LR 0.000125 Time 0.021669 +2023-10-02 21:58:38,926 - Epoch: [185][ 1170/ 1236] Overall Loss 0.113897 Objective Loss 0.113897 LR 0.000125 Time 0.021664 +2023-10-02 21:58:39,133 - Epoch: [185][ 1180/ 1236] Overall Loss 0.113928 Objective Loss 0.113928 LR 0.000125 Time 0.021655 +2023-10-02 21:58:39,344 - Epoch: [185][ 1190/ 1236] Overall Loss 0.114066 Objective Loss 0.114066 LR 0.000125 Time 0.021650 +2023-10-02 21:58:39,551 - Epoch: [185][ 1200/ 1236] Overall Loss 0.114132 Objective Loss 0.114132 LR 0.000125 Time 0.021642 +2023-10-02 21:58:39,762 - Epoch: [185][ 1210/ 1236] Overall Loss 0.114094 Objective Loss 0.114094 LR 0.000125 Time 0.021637 +2023-10-02 21:58:39,968 - Epoch: [185][ 1220/ 1236] Overall Loss 0.114070 Objective Loss 0.114070 LR 0.000125 Time 0.021628 +2023-10-02 21:58:40,230 - Epoch: [185][ 1230/ 1236] Overall Loss 0.114197 Objective Loss 0.114197 LR 0.000125 Time 0.021665 +2023-10-02 21:58:40,351 - Epoch: [185][ 1236/ 1236] Overall Loss 0.114170 Objective Loss 0.114170 Top1 91.038697 Top5 98.574338 LR 0.000125 Time 0.021658 +2023-10-02 21:58:40,480 - --- validate (epoch=185)----------- +2023-10-02 21:58:40,481 - 29943 samples (256 per mini-batch) +2023-10-02 21:58:40,978 - Epoch: [185][ 10/ 117] Loss 0.306544 Top1 87.148438 Top5 98.632812 +2023-10-02 21:58:41,130 - Epoch: [185][ 20/ 117] Loss 0.300020 Top1 87.773438 Top5 98.750000 +2023-10-02 21:58:41,281 - Epoch: [185][ 30/ 117] Loss 0.324989 Top1 87.226562 Top5 98.580729 +2023-10-02 21:58:41,431 - Epoch: [185][ 40/ 117] Loss 0.318536 Top1 87.089844 Top5 98.623047 +2023-10-02 21:58:41,581 - Epoch: [185][ 50/ 117] Loss 0.312404 Top1 87.328125 Top5 98.640625 +2023-10-02 21:58:41,731 - Epoch: [185][ 60/ 117] Loss 0.308333 Top1 87.389323 Top5 98.671875 +2023-10-02 21:58:41,882 - Epoch: [185][ 70/ 117] Loss 0.307036 Top1 87.449777 Top5 98.655134 +2023-10-02 21:58:42,033 - Epoch: [185][ 80/ 117] Loss 0.307660 Top1 87.441406 Top5 98.647461 +2023-10-02 21:58:42,183 - Epoch: [185][ 90/ 117] Loss 0.305680 Top1 87.304688 Top5 98.680556 +2023-10-02 21:58:42,334 - Epoch: [185][ 100/ 117] Loss 0.306849 Top1 87.406250 Top5 98.664062 +2023-10-02 21:58:42,491 - Epoch: [185][ 110/ 117] Loss 0.305684 Top1 87.393466 Top5 98.664773 +2023-10-02 21:58:42,581 - Epoch: [185][ 117/ 117] Loss 0.304512 Top1 87.452827 Top5 98.690846 +2023-10-02 21:58:42,716 - ==> Top1: 87.453 Top5: 98.691 Loss: 0.305 + +2023-10-02 21:58:42,717 - ==> Confusion: +[[ 941 0 5 0 5 3 0 0 3 56 2 0 0 0 6 2 2 1 2 0 22] + [ 0 1065 0 1 3 17 1 19 0 1 0 0 0 0 0 2 0 0 10 3 9] + [ 1 2 982 8 0 0 14 7 0 2 0 1 7 2 1 3 2 1 13 1 9] + [ 0 4 14 984 2 1 1 4 3 1 4 0 11 2 20 3 1 4 11 1 18] + [ 20 6 1 1 970 7 0 0 0 9 0 0 0 3 12 6 7 0 1 1 6] + [ 2 24 0 0 3 1010 2 19 1 6 2 4 2 9 3 1 1 2 5 2 18] + [ 0 3 24 1 0 1 1131 6 0 0 3 1 0 0 0 5 0 1 1 8 6] + [ 1 10 11 0 6 24 5 1074 2 2 5 2 1 6 1 2 0 2 45 6 13] + [ 16 2 0 1 2 5 0 2 972 34 12 1 1 15 14 0 3 2 3 0 4] + [ 89 0 1 2 4 4 0 0 20 951 1 0 1 25 8 3 1 0 0 0 9] + [ 3 1 10 6 0 1 4 2 6 2 978 0 0 13 4 0 2 2 5 1 13] + [ 0 0 2 0 1 18 0 3 0 0 0 955 22 7 0 0 0 16 0 5 6] + [ 0 1 1 1 0 0 1 1 0 2 3 26 985 2 3 8 0 11 2 6 15] + [ 0 0 1 0 4 11 0 0 9 10 3 6 0 1045 4 0 0 1 0 2 23] + [ 11 0 5 16 6 1 0 0 16 4 1 0 2 1 1013 0 1 2 11 0 11] + [ 0 0 2 1 5 1 0 0 0 0 0 7 8 0 0 1071 14 9 2 9 5] + [ 0 16 1 0 6 6 0 0 0 0 0 4 0 2 3 9 1094 0 1 7 12] + [ 0 0 1 1 0 0 3 0 0 0 0 3 25 3 2 6 0 988 0 1 5] + [ 2 3 3 16 0 1 0 19 3 1 2 0 1 0 9 0 0 1 996 0 11] + [ 0 0 3 1 1 3 6 5 0 0 0 13 4 3 0 2 6 0 1 1094 10] + [ 80 119 105 83 51 141 34 87 54 53 130 65 279 218 110 54 63 45 112 135 5887]] + +2023-10-02 21:58:42,718 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:58:42,718 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:58:42,724 - + +2023-10-02 21:58:42,724 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:58:43,764 - Epoch: [186][ 10/ 1236] Overall Loss 0.129188 Objective Loss 0.129188 LR 0.000125 Time 0.103962 +2023-10-02 21:58:43,976 - Epoch: [186][ 20/ 1236] Overall Loss 0.120753 Objective Loss 0.120753 LR 0.000125 Time 0.062523 +2023-10-02 21:58:44,188 - Epoch: [186][ 30/ 1236] Overall Loss 0.115419 Objective Loss 0.115419 LR 0.000125 Time 0.048749 +2023-10-02 21:58:44,400 - Epoch: [186][ 40/ 1236] Overall Loss 0.115716 Objective Loss 0.115716 LR 0.000125 Time 0.041842 +2023-10-02 21:58:44,613 - Epoch: [186][ 50/ 1236] Overall Loss 0.112978 Objective Loss 0.112978 LR 0.000125 Time 0.037718 +2023-10-02 21:58:44,824 - Epoch: [186][ 60/ 1236] Overall Loss 0.113016 Objective Loss 0.113016 LR 0.000125 Time 0.034945 +2023-10-02 21:58:45,037 - Epoch: [186][ 70/ 1236] Overall Loss 0.113187 Objective Loss 0.113187 LR 0.000125 Time 0.032985 +2023-10-02 21:58:45,248 - Epoch: [186][ 80/ 1236] Overall Loss 0.114461 Objective Loss 0.114461 LR 0.000125 Time 0.031501 +2023-10-02 21:58:45,461 - Epoch: [186][ 90/ 1236] Overall Loss 0.114410 Objective Loss 0.114410 LR 0.000125 Time 0.030355 +2023-10-02 21:58:45,670 - Epoch: [186][ 100/ 1236] Overall Loss 0.114742 Objective Loss 0.114742 LR 0.000125 Time 0.029411 +2023-10-02 21:58:45,877 - Epoch: [186][ 110/ 1236] Overall Loss 0.114094 Objective Loss 0.114094 LR 0.000125 Time 0.028603 +2023-10-02 21:58:46,085 - Epoch: [186][ 120/ 1236] Overall Loss 0.114584 Objective Loss 0.114584 LR 0.000125 Time 0.027951 +2023-10-02 21:58:46,292 - Epoch: [186][ 130/ 1236] Overall Loss 0.114315 Objective Loss 0.114315 LR 0.000125 Time 0.027393 +2023-10-02 21:58:46,500 - Epoch: [186][ 140/ 1236] Overall Loss 0.114127 Objective Loss 0.114127 LR 0.000125 Time 0.026917 +2023-10-02 21:58:46,707 - Epoch: [186][ 150/ 1236] Overall Loss 0.114101 Objective Loss 0.114101 LR 0.000125 Time 0.026490 +2023-10-02 21:58:46,915 - Epoch: [186][ 160/ 1236] Overall Loss 0.113703 Objective Loss 0.113703 LR 0.000125 Time 0.026133 +2023-10-02 21:58:47,121 - Epoch: [186][ 170/ 1236] Overall Loss 0.113821 Objective Loss 0.113821 LR 0.000125 Time 0.025803 +2023-10-02 21:58:47,330 - Epoch: [186][ 180/ 1236] Overall Loss 0.114068 Objective Loss 0.114068 LR 0.000125 Time 0.025524 +2023-10-02 21:58:47,536 - Epoch: [186][ 190/ 1236] Overall Loss 0.113273 Objective Loss 0.113273 LR 0.000125 Time 0.025260 +2023-10-02 21:58:47,744 - Epoch: [186][ 200/ 1236] Overall Loss 0.113368 Objective Loss 0.113368 LR 0.000125 Time 0.025036 +2023-10-02 21:58:47,951 - Epoch: [186][ 210/ 1236] Overall Loss 0.113813 Objective Loss 0.113813 LR 0.000125 Time 0.024821 +2023-10-02 21:58:48,159 - Epoch: [186][ 220/ 1236] Overall Loss 0.113390 Objective Loss 0.113390 LR 0.000125 Time 0.024638 +2023-10-02 21:58:48,366 - Epoch: [186][ 230/ 1236] Overall Loss 0.113969 Objective Loss 0.113969 LR 0.000125 Time 0.024459 +2023-10-02 21:58:48,574 - Epoch: [186][ 240/ 1236] Overall Loss 0.114247 Objective Loss 0.114247 LR 0.000125 Time 0.024306 +2023-10-02 21:58:48,781 - Epoch: [186][ 250/ 1236] Overall Loss 0.114146 Objective Loss 0.114146 LR 0.000125 Time 0.024156 +2023-10-02 21:58:48,989 - Epoch: [186][ 260/ 1236] Overall Loss 0.114231 Objective Loss 0.114231 LR 0.000125 Time 0.024025 +2023-10-02 21:58:49,196 - Epoch: [186][ 270/ 1236] Overall Loss 0.114331 Objective Loss 0.114331 LR 0.000125 Time 0.023896 +2023-10-02 21:58:49,404 - Epoch: [186][ 280/ 1236] Overall Loss 0.113952 Objective Loss 0.113952 LR 0.000125 Time 0.023785 +2023-10-02 21:58:49,611 - Epoch: [186][ 290/ 1236] Overall Loss 0.113915 Objective Loss 0.113915 LR 0.000125 Time 0.023673 +2023-10-02 21:58:49,819 - Epoch: [186][ 300/ 1236] Overall Loss 0.114266 Objective Loss 0.114266 LR 0.000125 Time 0.023576 +2023-10-02 21:58:50,026 - Epoch: [186][ 310/ 1236] Overall Loss 0.114316 Objective Loss 0.114316 LR 0.000125 Time 0.023478 +2023-10-02 21:58:50,235 - Epoch: [186][ 320/ 1236] Overall Loss 0.114173 Objective Loss 0.114173 LR 0.000125 Time 0.023394 +2023-10-02 21:58:50,442 - Epoch: [186][ 330/ 1236] Overall Loss 0.114119 Objective Loss 0.114119 LR 0.000125 Time 0.023308 +2023-10-02 21:58:50,650 - Epoch: [186][ 340/ 1236] Overall Loss 0.114221 Objective Loss 0.114221 LR 0.000125 Time 0.023234 +2023-10-02 21:58:50,855 - Epoch: [186][ 350/ 1236] Overall Loss 0.113732 Objective Loss 0.113732 LR 0.000125 Time 0.023151 +2023-10-02 21:58:51,063 - Epoch: [186][ 360/ 1236] Overall Loss 0.113943 Objective Loss 0.113943 LR 0.000125 Time 0.023085 +2023-10-02 21:58:51,270 - Epoch: [186][ 370/ 1236] Overall Loss 0.114083 Objective Loss 0.114083 LR 0.000125 Time 0.023018 +2023-10-02 21:58:51,478 - Epoch: [186][ 380/ 1236] Overall Loss 0.113977 Objective Loss 0.113977 LR 0.000125 Time 0.022958 +2023-10-02 21:58:51,686 - Epoch: [186][ 390/ 1236] Overall Loss 0.113885 Objective Loss 0.113885 LR 0.000125 Time 0.022901 +2023-10-02 21:58:51,894 - Epoch: [186][ 400/ 1236] Overall Loss 0.114057 Objective Loss 0.114057 LR 0.000125 Time 0.022848 +2023-10-02 21:58:52,101 - Epoch: [186][ 410/ 1236] Overall Loss 0.114285 Objective Loss 0.114285 LR 0.000125 Time 0.022795 +2023-10-02 21:58:52,309 - Epoch: [186][ 420/ 1236] Overall Loss 0.114517 Objective Loss 0.114517 LR 0.000125 Time 0.022748 +2023-10-02 21:58:52,517 - Epoch: [186][ 430/ 1236] Overall Loss 0.114477 Objective Loss 0.114477 LR 0.000125 Time 0.022700 +2023-10-02 21:58:52,725 - Epoch: [186][ 440/ 1236] Overall Loss 0.114489 Objective Loss 0.114489 LR 0.000125 Time 0.022656 +2023-10-02 21:58:52,932 - Epoch: [186][ 450/ 1236] Overall Loss 0.114306 Objective Loss 0.114306 LR 0.000125 Time 0.022609 +2023-10-02 21:58:53,140 - Epoch: [186][ 460/ 1236] Overall Loss 0.114029 Objective Loss 0.114029 LR 0.000125 Time 0.022570 +2023-10-02 21:58:53,347 - Epoch: [186][ 470/ 1236] Overall Loss 0.114099 Objective Loss 0.114099 LR 0.000125 Time 0.022527 +2023-10-02 21:58:53,555 - Epoch: [186][ 480/ 1236] Overall Loss 0.113861 Objective Loss 0.113861 LR 0.000125 Time 0.022490 +2023-10-02 21:58:53,763 - Epoch: [186][ 490/ 1236] Overall Loss 0.113594 Objective Loss 0.113594 LR 0.000125 Time 0.022454 +2023-10-02 21:58:53,971 - Epoch: [186][ 500/ 1236] Overall Loss 0.113672 Objective Loss 0.113672 LR 0.000125 Time 0.022420 +2023-10-02 21:58:54,178 - Epoch: [186][ 510/ 1236] Overall Loss 0.113687 Objective Loss 0.113687 LR 0.000125 Time 0.022384 +2023-10-02 21:58:54,386 - Epoch: [186][ 520/ 1236] Overall Loss 0.113524 Objective Loss 0.113524 LR 0.000125 Time 0.022353 +2023-10-02 21:58:54,593 - Epoch: [186][ 530/ 1236] Overall Loss 0.113297 Objective Loss 0.113297 LR 0.000125 Time 0.022321 +2023-10-02 21:58:54,801 - Epoch: [186][ 540/ 1236] Overall Loss 0.113404 Objective Loss 0.113404 LR 0.000125 Time 0.022292 +2023-10-02 21:58:55,009 - Epoch: [186][ 550/ 1236] Overall Loss 0.113347 Objective Loss 0.113347 LR 0.000125 Time 0.022264 +2023-10-02 21:58:55,217 - Epoch: [186][ 560/ 1236] Overall Loss 0.113111 Objective Loss 0.113111 LR 0.000125 Time 0.022237 +2023-10-02 21:58:55,424 - Epoch: [186][ 570/ 1236] Overall Loss 0.113252 Objective Loss 0.113252 LR 0.000125 Time 0.022210 +2023-10-02 21:58:55,632 - Epoch: [186][ 580/ 1236] Overall Loss 0.113008 Objective Loss 0.113008 LR 0.000125 Time 0.022185 +2023-10-02 21:58:55,839 - Epoch: [186][ 590/ 1236] Overall Loss 0.112794 Objective Loss 0.112794 LR 0.000125 Time 0.022160 +2023-10-02 21:58:56,047 - Epoch: [186][ 600/ 1236] Overall Loss 0.112881 Objective Loss 0.112881 LR 0.000125 Time 0.022136 +2023-10-02 21:58:56,255 - Epoch: [186][ 610/ 1236] Overall Loss 0.112737 Objective Loss 0.112737 LR 0.000125 Time 0.022113 +2023-10-02 21:58:56,463 - Epoch: [186][ 620/ 1236] Overall Loss 0.112773 Objective Loss 0.112773 LR 0.000125 Time 0.022092 +2023-10-02 21:58:56,670 - Epoch: [186][ 630/ 1236] Overall Loss 0.113107 Objective Loss 0.113107 LR 0.000125 Time 0.022067 +2023-10-02 21:58:56,878 - Epoch: [186][ 640/ 1236] Overall Loss 0.112808 Objective Loss 0.112808 LR 0.000125 Time 0.022047 +2023-10-02 21:58:57,086 - Epoch: [186][ 650/ 1236] Overall Loss 0.112858 Objective Loss 0.112858 LR 0.000125 Time 0.022027 +2023-10-02 21:58:57,294 - Epoch: [186][ 660/ 1236] Overall Loss 0.113100 Objective Loss 0.113100 LR 0.000125 Time 0.022008 +2023-10-02 21:58:57,501 - Epoch: [186][ 670/ 1236] Overall Loss 0.113235 Objective Loss 0.113235 LR 0.000125 Time 0.021986 +2023-10-02 21:58:57,709 - Epoch: [186][ 680/ 1236] Overall Loss 0.113227 Objective Loss 0.113227 LR 0.000125 Time 0.021968 +2023-10-02 21:58:57,917 - Epoch: [186][ 690/ 1236] Overall Loss 0.113278 Objective Loss 0.113278 LR 0.000125 Time 0.021950 +2023-10-02 21:58:58,125 - Epoch: [186][ 700/ 1236] Overall Loss 0.113319 Objective Loss 0.113319 LR 0.000125 Time 0.021934 +2023-10-02 21:58:58,332 - Epoch: [186][ 710/ 1236] Overall Loss 0.113412 Objective Loss 0.113412 LR 0.000125 Time 0.021916 +2023-10-02 21:58:58,540 - Epoch: [186][ 720/ 1236] Overall Loss 0.113339 Objective Loss 0.113339 LR 0.000125 Time 0.021900 +2023-10-02 21:58:58,748 - Epoch: [186][ 730/ 1236] Overall Loss 0.113410 Objective Loss 0.113410 LR 0.000125 Time 0.021884 +2023-10-02 21:58:58,956 - Epoch: [186][ 740/ 1236] Overall Loss 0.113202 Objective Loss 0.113202 LR 0.000125 Time 0.021869 +2023-10-02 21:58:59,163 - Epoch: [186][ 750/ 1236] Overall Loss 0.113241 Objective Loss 0.113241 LR 0.000125 Time 0.021853 +2023-10-02 21:58:59,371 - Epoch: [186][ 760/ 1236] Overall Loss 0.113392 Objective Loss 0.113392 LR 0.000125 Time 0.021839 +2023-10-02 21:58:59,578 - Epoch: [186][ 770/ 1236] Overall Loss 0.113708 Objective Loss 0.113708 LR 0.000125 Time 0.021824 +2023-10-02 21:58:59,786 - Epoch: [186][ 780/ 1236] Overall Loss 0.113713 Objective Loss 0.113713 LR 0.000125 Time 0.021811 +2023-10-02 21:58:59,994 - Epoch: [186][ 790/ 1236] Overall Loss 0.113879 Objective Loss 0.113879 LR 0.000125 Time 0.021797 +2023-10-02 21:59:00,202 - Epoch: [186][ 800/ 1236] Overall Loss 0.114050 Objective Loss 0.114050 LR 0.000125 Time 0.021784 +2023-10-02 21:59:00,409 - Epoch: [186][ 810/ 1236] Overall Loss 0.114312 Objective Loss 0.114312 LR 0.000125 Time 0.021770 +2023-10-02 21:59:00,617 - Epoch: [186][ 820/ 1236] Overall Loss 0.114252 Objective Loss 0.114252 LR 0.000125 Time 0.021758 +2023-10-02 21:59:00,824 - Epoch: [186][ 830/ 1236] Overall Loss 0.114107 Objective Loss 0.114107 LR 0.000125 Time 0.021745 +2023-10-02 21:59:01,032 - Epoch: [186][ 840/ 1236] Overall Loss 0.114154 Objective Loss 0.114154 LR 0.000125 Time 0.021734 +2023-10-02 21:59:01,239 - Epoch: [186][ 850/ 1236] Overall Loss 0.114315 Objective Loss 0.114315 LR 0.000125 Time 0.021720 +2023-10-02 21:59:01,447 - Epoch: [186][ 860/ 1236] Overall Loss 0.114273 Objective Loss 0.114273 LR 0.000125 Time 0.021708 +2023-10-02 21:59:01,655 - Epoch: [186][ 870/ 1236] Overall Loss 0.114191 Objective Loss 0.114191 LR 0.000125 Time 0.021697 +2023-10-02 21:59:01,863 - Epoch: [186][ 880/ 1236] Overall Loss 0.114164 Objective Loss 0.114164 LR 0.000125 Time 0.021687 +2023-10-02 21:59:02,070 - Epoch: [186][ 890/ 1236] Overall Loss 0.113954 Objective Loss 0.113954 LR 0.000125 Time 0.021676 +2023-10-02 21:59:02,279 - Epoch: [186][ 900/ 1236] Overall Loss 0.113896 Objective Loss 0.113896 LR 0.000125 Time 0.021666 +2023-10-02 21:59:02,486 - Epoch: [186][ 910/ 1236] Overall Loss 0.114081 Objective Loss 0.114081 LR 0.000125 Time 0.021656 +2023-10-02 21:59:02,694 - Epoch: [186][ 920/ 1236] Overall Loss 0.114268 Objective Loss 0.114268 LR 0.000125 Time 0.021646 +2023-10-02 21:59:02,901 - Epoch: [186][ 930/ 1236] Overall Loss 0.114360 Objective Loss 0.114360 LR 0.000125 Time 0.021636 +2023-10-02 21:59:03,109 - Epoch: [186][ 940/ 1236] Overall Loss 0.114490 Objective Loss 0.114490 LR 0.000125 Time 0.021626 +2023-10-02 21:59:03,316 - Epoch: [186][ 950/ 1236] Overall Loss 0.114440 Objective Loss 0.114440 LR 0.000125 Time 0.021616 +2023-10-02 21:59:03,525 - Epoch: [186][ 960/ 1236] Overall Loss 0.114385 Objective Loss 0.114385 LR 0.000125 Time 0.021608 +2023-10-02 21:59:03,732 - Epoch: [186][ 970/ 1236] Overall Loss 0.114469 Objective Loss 0.114469 LR 0.000125 Time 0.021599 +2023-10-02 21:59:03,940 - Epoch: [186][ 980/ 1236] Overall Loss 0.114572 Objective Loss 0.114572 LR 0.000125 Time 0.021590 +2023-10-02 21:59:04,147 - Epoch: [186][ 990/ 1236] Overall Loss 0.114649 Objective Loss 0.114649 LR 0.000125 Time 0.021581 +2023-10-02 21:59:04,355 - Epoch: [186][ 1000/ 1236] Overall Loss 0.114845 Objective Loss 0.114845 LR 0.000125 Time 0.021573 +2023-10-02 21:59:04,563 - Epoch: [186][ 1010/ 1236] Overall Loss 0.114883 Objective Loss 0.114883 LR 0.000125 Time 0.021564 +2023-10-02 21:59:04,771 - Epoch: [186][ 1020/ 1236] Overall Loss 0.114787 Objective Loss 0.114787 LR 0.000125 Time 0.021557 +2023-10-02 21:59:04,978 - Epoch: [186][ 1030/ 1236] Overall Loss 0.114570 Objective Loss 0.114570 LR 0.000125 Time 0.021548 +2023-10-02 21:59:05,186 - Epoch: [186][ 1040/ 1236] Overall Loss 0.114372 Objective Loss 0.114372 LR 0.000125 Time 0.021541 +2023-10-02 21:59:05,394 - Epoch: [186][ 1050/ 1236] Overall Loss 0.114463 Objective Loss 0.114463 LR 0.000125 Time 0.021533 +2023-10-02 21:59:05,602 - Epoch: [186][ 1060/ 1236] Overall Loss 0.114427 Objective Loss 0.114427 LR 0.000125 Time 0.021526 +2023-10-02 21:59:05,809 - Epoch: [186][ 1070/ 1236] Overall Loss 0.114289 Objective Loss 0.114289 LR 0.000125 Time 0.021518 +2023-10-02 21:59:06,017 - Epoch: [186][ 1080/ 1236] Overall Loss 0.114138 Objective Loss 0.114138 LR 0.000125 Time 0.021511 +2023-10-02 21:59:06,225 - Epoch: [186][ 1090/ 1236] Overall Loss 0.114148 Objective Loss 0.114148 LR 0.000125 Time 0.021504 +2023-10-02 21:59:06,433 - Epoch: [186][ 1100/ 1236] Overall Loss 0.114227 Objective Loss 0.114227 LR 0.000125 Time 0.021497 +2023-10-02 21:59:06,640 - Epoch: [186][ 1110/ 1236] Overall Loss 0.114278 Objective Loss 0.114278 LR 0.000125 Time 0.021490 +2023-10-02 21:59:06,848 - Epoch: [186][ 1120/ 1236] Overall Loss 0.114101 Objective Loss 0.114101 LR 0.000125 Time 0.021484 +2023-10-02 21:59:07,056 - Epoch: [186][ 1130/ 1236] Overall Loss 0.114108 Objective Loss 0.114108 LR 0.000125 Time 0.021477 +2023-10-02 21:59:07,264 - Epoch: [186][ 1140/ 1236] Overall Loss 0.114048 Objective Loss 0.114048 LR 0.000125 Time 0.021471 +2023-10-02 21:59:07,471 - Epoch: [186][ 1150/ 1236] Overall Loss 0.114152 Objective Loss 0.114152 LR 0.000125 Time 0.021464 +2023-10-02 21:59:07,679 - Epoch: [186][ 1160/ 1236] Overall Loss 0.114117 Objective Loss 0.114117 LR 0.000125 Time 0.021458 +2023-10-02 21:59:07,886 - Epoch: [186][ 1170/ 1236] Overall Loss 0.114068 Objective Loss 0.114068 LR 0.000125 Time 0.021450 +2023-10-02 21:59:08,094 - Epoch: [186][ 1180/ 1236] Overall Loss 0.114035 Objective Loss 0.114035 LR 0.000125 Time 0.021444 +2023-10-02 21:59:08,302 - Epoch: [186][ 1190/ 1236] Overall Loss 0.114069 Objective Loss 0.114069 LR 0.000125 Time 0.021437 +2023-10-02 21:59:08,510 - Epoch: [186][ 1200/ 1236] Overall Loss 0.114261 Objective Loss 0.114261 LR 0.000125 Time 0.021432 +2023-10-02 21:59:08,717 - Epoch: [186][ 1210/ 1236] Overall Loss 0.114394 Objective Loss 0.114394 LR 0.000125 Time 0.021425 +2023-10-02 21:59:08,925 - Epoch: [186][ 1220/ 1236] Overall Loss 0.114356 Objective Loss 0.114356 LR 0.000125 Time 0.021419 +2023-10-02 21:59:09,186 - Epoch: [186][ 1230/ 1236] Overall Loss 0.114248 Objective Loss 0.114248 LR 0.000125 Time 0.021457 +2023-10-02 21:59:09,308 - Epoch: [186][ 1236/ 1236] Overall Loss 0.114198 Objective Loss 0.114198 Top1 94.093686 Top5 99.796334 LR 0.000125 Time 0.021452 +2023-10-02 21:59:09,447 - --- validate (epoch=186)----------- +2023-10-02 21:59:09,447 - 29943 samples (256 per mini-batch) +2023-10-02 21:59:09,961 - Epoch: [186][ 10/ 117] Loss 0.274460 Top1 87.617188 Top5 98.906250 +2023-10-02 21:59:10,116 - Epoch: [186][ 20/ 117] Loss 0.296223 Top1 87.519531 Top5 98.847656 +2023-10-02 21:59:10,269 - Epoch: [186][ 30/ 117] Loss 0.306720 Top1 87.343750 Top5 98.684896 +2023-10-02 21:59:10,423 - Epoch: [186][ 40/ 117] Loss 0.309556 Top1 87.255859 Top5 98.710938 +2023-10-02 21:59:10,575 - Epoch: [186][ 50/ 117] Loss 0.318087 Top1 87.187500 Top5 98.648438 +2023-10-02 21:59:10,729 - Epoch: [186][ 60/ 117] Loss 0.311435 Top1 87.447917 Top5 98.678385 +2023-10-02 21:59:10,880 - Epoch: [186][ 70/ 117] Loss 0.310022 Top1 87.410714 Top5 98.660714 +2023-10-02 21:59:11,034 - Epoch: [186][ 80/ 117] Loss 0.310513 Top1 87.382812 Top5 98.671875 +2023-10-02 21:59:11,186 - Epoch: [186][ 90/ 117] Loss 0.315093 Top1 87.382812 Top5 98.632812 +2023-10-02 21:59:11,340 - Epoch: [186][ 100/ 117] Loss 0.310572 Top1 87.437500 Top5 98.664062 +2023-10-02 21:59:11,499 - Epoch: [186][ 110/ 117] Loss 0.309083 Top1 87.524858 Top5 98.668324 +2023-10-02 21:59:11,588 - Epoch: [186][ 117/ 117] Loss 0.307616 Top1 87.532979 Top5 98.690846 +2023-10-02 21:59:11,693 - ==> Top1: 87.533 Top5: 98.691 Loss: 0.308 + +2023-10-02 21:59:11,694 - ==> Confusion: +[[ 937 0 7 0 4 3 0 1 8 60 2 0 0 1 5 1 0 1 0 0 20] + [ 0 1063 1 0 4 16 0 24 0 1 0 0 0 0 0 3 0 0 7 3 9] + [ 1 0 987 3 0 0 14 9 0 2 2 1 7 2 1 3 2 2 10 1 9] + [ 0 3 15 979 1 0 2 6 3 1 4 0 5 4 27 2 1 5 9 0 22] + [ 19 4 1 1 976 5 0 0 0 10 0 0 0 3 10 6 9 0 0 2 4] + [ 3 30 0 0 5 1000 2 18 1 5 2 10 2 7 4 0 0 0 5 2 20] + [ 0 2 27 1 0 1 1135 4 0 0 3 1 0 0 0 3 0 1 1 6 6] + [ 1 10 13 0 5 22 5 1078 2 2 3 2 3 4 1 0 0 2 43 11 11] + [ 15 2 0 1 1 3 0 2 988 31 9 2 1 11 12 0 3 1 4 1 2] + [ 89 0 1 1 6 3 0 0 25 953 0 0 1 23 5 3 0 0 0 2 7] + [ 2 1 13 6 0 3 2 5 10 3 965 2 0 12 4 0 3 1 8 1 12] + [ 0 0 1 0 0 12 0 6 1 0 0 974 11 5 0 1 0 14 0 3 7] + [ 0 0 0 1 0 1 2 0 0 1 4 40 967 2 3 8 2 14 2 8 13] + [ 0 0 0 0 3 7 0 0 11 8 5 9 0 1048 4 0 0 1 0 1 22] + [ 11 0 3 14 5 0 0 0 24 2 1 0 1 1 1017 0 2 2 9 0 9] + [ 0 0 1 1 6 0 1 0 0 0 1 6 7 0 0 1069 17 10 2 9 4] + [ 0 14 1 0 6 6 1 0 1 1 0 5 0 3 3 9 1094 0 0 6 11] + [ 0 0 0 1 0 0 2 0 0 0 0 6 16 2 3 4 1 996 0 3 4] + [ 1 2 2 16 0 1 0 24 5 0 1 0 0 0 9 0 0 2 993 0 12] + [ 0 2 3 2 1 2 10 5 0 1 1 16 2 3 0 1 7 0 1 1088 7] + [ 97 114 112 66 56 124 32 96 74 51 126 89 241 237 108 46 66 55 92 120 5903]] + +2023-10-02 21:59:11,695 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:59:11,695 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:59:11,701 - + +2023-10-02 21:59:11,701 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:59:12,736 - Epoch: [187][ 10/ 1236] Overall Loss 0.111189 Objective Loss 0.111189 LR 0.000125 Time 0.103463 +2023-10-02 21:59:12,947 - Epoch: [187][ 20/ 1236] Overall Loss 0.115206 Objective Loss 0.115206 LR 0.000125 Time 0.062228 +2023-10-02 21:59:13,154 - Epoch: [187][ 30/ 1236] Overall Loss 0.114022 Objective Loss 0.114022 LR 0.000125 Time 0.048401 +2023-10-02 21:59:13,364 - Epoch: [187][ 40/ 1236] Overall Loss 0.114950 Objective Loss 0.114950 LR 0.000125 Time 0.041527 +2023-10-02 21:59:13,572 - Epoch: [187][ 50/ 1236] Overall Loss 0.113447 Objective Loss 0.113447 LR 0.000125 Time 0.037386 +2023-10-02 21:59:13,782 - Epoch: [187][ 60/ 1236] Overall Loss 0.112792 Objective Loss 0.112792 LR 0.000125 Time 0.034644 +2023-10-02 21:59:13,990 - Epoch: [187][ 70/ 1236] Overall Loss 0.113491 Objective Loss 0.113491 LR 0.000125 Time 0.032644 +2023-10-02 21:59:14,200 - Epoch: [187][ 80/ 1236] Overall Loss 0.114920 Objective Loss 0.114920 LR 0.000125 Time 0.031179 +2023-10-02 21:59:14,408 - Epoch: [187][ 90/ 1236] Overall Loss 0.113819 Objective Loss 0.113819 LR 0.000125 Time 0.030027 +2023-10-02 21:59:14,617 - Epoch: [187][ 100/ 1236] Overall Loss 0.113873 Objective Loss 0.113873 LR 0.000125 Time 0.029115 +2023-10-02 21:59:14,826 - Epoch: [187][ 110/ 1236] Overall Loss 0.115475 Objective Loss 0.115475 LR 0.000125 Time 0.028357 +2023-10-02 21:59:15,035 - Epoch: [187][ 120/ 1236] Overall Loss 0.115021 Objective Loss 0.115021 LR 0.000125 Time 0.027734 +2023-10-02 21:59:15,240 - Epoch: [187][ 130/ 1236] Overall Loss 0.114995 Objective Loss 0.114995 LR 0.000125 Time 0.027165 +2023-10-02 21:59:15,449 - Epoch: [187][ 140/ 1236] Overall Loss 0.114450 Objective Loss 0.114450 LR 0.000125 Time 0.026721 +2023-10-02 21:59:15,657 - Epoch: [187][ 150/ 1236] Overall Loss 0.114107 Objective Loss 0.114107 LR 0.000125 Time 0.026311 +2023-10-02 21:59:15,866 - Epoch: [187][ 160/ 1236] Overall Loss 0.114445 Objective Loss 0.114445 LR 0.000125 Time 0.025970 +2023-10-02 21:59:16,074 - Epoch: [187][ 170/ 1236] Overall Loss 0.113993 Objective Loss 0.113993 LR 0.000125 Time 0.025656 +2023-10-02 21:59:16,282 - Epoch: [187][ 180/ 1236] Overall Loss 0.114223 Objective Loss 0.114223 LR 0.000125 Time 0.025386 +2023-10-02 21:59:16,490 - Epoch: [187][ 190/ 1236] Overall Loss 0.114934 Objective Loss 0.114934 LR 0.000125 Time 0.025137 +2023-10-02 21:59:16,698 - Epoch: [187][ 200/ 1236] Overall Loss 0.114508 Objective Loss 0.114508 LR 0.000125 Time 0.024920 +2023-10-02 21:59:16,906 - Epoch: [187][ 210/ 1236] Overall Loss 0.114322 Objective Loss 0.114322 LR 0.000125 Time 0.024715 +2023-10-02 21:59:17,116 - Epoch: [187][ 220/ 1236] Overall Loss 0.114762 Objective Loss 0.114762 LR 0.000125 Time 0.024543 +2023-10-02 21:59:17,324 - Epoch: [187][ 230/ 1236] Overall Loss 0.114021 Objective Loss 0.114021 LR 0.000125 Time 0.024373 +2023-10-02 21:59:17,533 - Epoch: [187][ 240/ 1236] Overall Loss 0.112975 Objective Loss 0.112975 LR 0.000125 Time 0.024229 +2023-10-02 21:59:17,740 - Epoch: [187][ 250/ 1236] Overall Loss 0.113424 Objective Loss 0.113424 LR 0.000125 Time 0.024080 +2023-10-02 21:59:17,948 - Epoch: [187][ 260/ 1236] Overall Loss 0.113982 Objective Loss 0.113982 LR 0.000125 Time 0.023954 +2023-10-02 21:59:18,155 - Epoch: [187][ 270/ 1236] Overall Loss 0.113270 Objective Loss 0.113270 LR 0.000125 Time 0.023833 +2023-10-02 21:59:18,364 - Epoch: [187][ 280/ 1236] Overall Loss 0.113494 Objective Loss 0.113494 LR 0.000125 Time 0.023727 +2023-10-02 21:59:18,575 - Epoch: [187][ 290/ 1236] Overall Loss 0.113370 Objective Loss 0.113370 LR 0.000125 Time 0.023632 +2023-10-02 21:59:18,785 - Epoch: [187][ 300/ 1236] Overall Loss 0.113981 Objective Loss 0.113981 LR 0.000125 Time 0.023542 +2023-10-02 21:59:18,995 - Epoch: [187][ 310/ 1236] Overall Loss 0.113899 Objective Loss 0.113899 LR 0.000125 Time 0.023459 +2023-10-02 21:59:19,204 - Epoch: [187][ 320/ 1236] Overall Loss 0.113907 Objective Loss 0.113907 LR 0.000125 Time 0.023377 +2023-10-02 21:59:19,415 - Epoch: [187][ 330/ 1236] Overall Loss 0.114084 Objective Loss 0.114084 LR 0.000125 Time 0.023307 +2023-10-02 21:59:19,624 - Epoch: [187][ 340/ 1236] Overall Loss 0.113915 Objective Loss 0.113915 LR 0.000125 Time 0.023235 +2023-10-02 21:59:19,834 - Epoch: [187][ 350/ 1236] Overall Loss 0.114243 Objective Loss 0.114243 LR 0.000125 Time 0.023171 +2023-10-02 21:59:20,044 - Epoch: [187][ 360/ 1236] Overall Loss 0.114108 Objective Loss 0.114108 LR 0.000125 Time 0.023109 +2023-10-02 21:59:20,256 - Epoch: [187][ 370/ 1236] Overall Loss 0.113820 Objective Loss 0.113820 LR 0.000125 Time 0.023055 +2023-10-02 21:59:20,466 - Epoch: [187][ 380/ 1236] Overall Loss 0.113691 Objective Loss 0.113691 LR 0.000125 Time 0.023002 +2023-10-02 21:59:20,678 - Epoch: [187][ 390/ 1236] Overall Loss 0.113163 Objective Loss 0.113163 LR 0.000125 Time 0.022953 +2023-10-02 21:59:20,887 - Epoch: [187][ 400/ 1236] Overall Loss 0.113121 Objective Loss 0.113121 LR 0.000125 Time 0.022901 +2023-10-02 21:59:21,098 - Epoch: [187][ 410/ 1236] Overall Loss 0.113543 Objective Loss 0.113543 LR 0.000125 Time 0.022857 +2023-10-02 21:59:21,308 - Epoch: [187][ 420/ 1236] Overall Loss 0.113446 Objective Loss 0.113446 LR 0.000125 Time 0.022811 +2023-10-02 21:59:21,520 - Epoch: [187][ 430/ 1236] Overall Loss 0.113383 Objective Loss 0.113383 LR 0.000125 Time 0.022772 +2023-10-02 21:59:21,729 - Epoch: [187][ 440/ 1236] Overall Loss 0.113317 Objective Loss 0.113317 LR 0.000125 Time 0.022730 +2023-10-02 21:59:21,942 - Epoch: [187][ 450/ 1236] Overall Loss 0.113523 Objective Loss 0.113523 LR 0.000125 Time 0.022696 +2023-10-02 21:59:22,151 - Epoch: [187][ 460/ 1236] Overall Loss 0.113570 Objective Loss 0.113570 LR 0.000125 Time 0.022657 +2023-10-02 21:59:22,364 - Epoch: [187][ 470/ 1236] Overall Loss 0.113733 Objective Loss 0.113733 LR 0.000125 Time 0.022627 +2023-10-02 21:59:22,574 - Epoch: [187][ 480/ 1236] Overall Loss 0.113852 Objective Loss 0.113852 LR 0.000125 Time 0.022592 +2023-10-02 21:59:22,786 - Epoch: [187][ 490/ 1236] Overall Loss 0.113911 Objective Loss 0.113911 LR 0.000125 Time 0.022564 +2023-10-02 21:59:22,996 - Epoch: [187][ 500/ 1236] Overall Loss 0.114099 Objective Loss 0.114099 LR 0.000125 Time 0.022532 +2023-10-02 21:59:23,208 - Epoch: [187][ 510/ 1236] Overall Loss 0.113970 Objective Loss 0.113970 LR 0.000125 Time 0.022505 +2023-10-02 21:59:23,420 - Epoch: [187][ 520/ 1236] Overall Loss 0.114104 Objective Loss 0.114104 LR 0.000125 Time 0.022479 +2023-10-02 21:59:23,646 - Epoch: [187][ 530/ 1236] Overall Loss 0.114367 Objective Loss 0.114367 LR 0.000125 Time 0.022481 +2023-10-02 21:59:23,863 - Epoch: [187][ 540/ 1236] Overall Loss 0.114309 Objective Loss 0.114309 LR 0.000125 Time 0.022466 +2023-10-02 21:59:24,079 - Epoch: [187][ 550/ 1236] Overall Loss 0.114147 Objective Loss 0.114147 LR 0.000125 Time 0.022448 +2023-10-02 21:59:24,294 - Epoch: [187][ 560/ 1236] Overall Loss 0.114096 Objective Loss 0.114096 LR 0.000125 Time 0.022431 +2023-10-02 21:59:24,510 - Epoch: [187][ 570/ 1236] Overall Loss 0.114071 Objective Loss 0.114071 LR 0.000125 Time 0.022415 +2023-10-02 21:59:24,725 - Epoch: [187][ 580/ 1236] Overall Loss 0.113861 Objective Loss 0.113861 LR 0.000125 Time 0.022400 +2023-10-02 21:59:24,941 - Epoch: [187][ 590/ 1236] Overall Loss 0.113920 Objective Loss 0.113920 LR 0.000125 Time 0.022386 +2023-10-02 21:59:25,156 - Epoch: [187][ 600/ 1236] Overall Loss 0.114155 Objective Loss 0.114155 LR 0.000125 Time 0.022370 +2023-10-02 21:59:25,373 - Epoch: [187][ 610/ 1236] Overall Loss 0.114059 Objective Loss 0.114059 LR 0.000125 Time 0.022359 +2023-10-02 21:59:25,588 - Epoch: [187][ 620/ 1236] Overall Loss 0.113873 Objective Loss 0.113873 LR 0.000125 Time 0.022344 +2023-10-02 21:59:25,804 - Epoch: [187][ 630/ 1236] Overall Loss 0.114158 Objective Loss 0.114158 LR 0.000125 Time 0.022331 +2023-10-02 21:59:26,019 - Epoch: [187][ 640/ 1236] Overall Loss 0.113975 Objective Loss 0.113975 LR 0.000125 Time 0.022318 +2023-10-02 21:59:26,235 - Epoch: [187][ 650/ 1236] Overall Loss 0.114168 Objective Loss 0.114168 LR 0.000125 Time 0.022306 +2023-10-02 21:59:26,450 - Epoch: [187][ 660/ 1236] Overall Loss 0.114126 Objective Loss 0.114126 LR 0.000125 Time 0.022293 +2023-10-02 21:59:26,666 - Epoch: [187][ 670/ 1236] Overall Loss 0.113901 Objective Loss 0.113901 LR 0.000125 Time 0.022282 +2023-10-02 21:59:26,881 - Epoch: [187][ 680/ 1236] Overall Loss 0.113880 Objective Loss 0.113880 LR 0.000125 Time 0.022270 +2023-10-02 21:59:27,097 - Epoch: [187][ 690/ 1236] Overall Loss 0.114193 Objective Loss 0.114193 LR 0.000125 Time 0.022259 +2023-10-02 21:59:27,312 - Epoch: [187][ 700/ 1236] Overall Loss 0.114234 Objective Loss 0.114234 LR 0.000125 Time 0.022248 +2023-10-02 21:59:27,527 - Epoch: [187][ 710/ 1236] Overall Loss 0.114413 Objective Loss 0.114413 LR 0.000125 Time 0.022237 +2023-10-02 21:59:27,742 - Epoch: [187][ 720/ 1236] Overall Loss 0.114440 Objective Loss 0.114440 LR 0.000125 Time 0.022226 +2023-10-02 21:59:27,957 - Epoch: [187][ 730/ 1236] Overall Loss 0.114448 Objective Loss 0.114448 LR 0.000125 Time 0.022215 +2023-10-02 21:59:28,171 - Epoch: [187][ 740/ 1236] Overall Loss 0.114238 Objective Loss 0.114238 LR 0.000125 Time 0.022204 +2023-10-02 21:59:28,386 - Epoch: [187][ 750/ 1236] Overall Loss 0.114309 Objective Loss 0.114309 LR 0.000125 Time 0.022194 +2023-10-02 21:59:28,600 - Epoch: [187][ 760/ 1236] Overall Loss 0.114436 Objective Loss 0.114436 LR 0.000125 Time 0.022183 +2023-10-02 21:59:28,814 - Epoch: [187][ 770/ 1236] Overall Loss 0.114684 Objective Loss 0.114684 LR 0.000125 Time 0.022173 +2023-10-02 21:59:29,029 - Epoch: [187][ 780/ 1236] Overall Loss 0.114650 Objective Loss 0.114650 LR 0.000125 Time 0.022163 +2023-10-02 21:59:29,243 - Epoch: [187][ 790/ 1236] Overall Loss 0.114418 Objective Loss 0.114418 LR 0.000125 Time 0.022153 +2023-10-02 21:59:29,457 - Epoch: [187][ 800/ 1236] Overall Loss 0.114475 Objective Loss 0.114475 LR 0.000125 Time 0.022143 +2023-10-02 21:59:29,671 - Epoch: [187][ 810/ 1236] Overall Loss 0.114356 Objective Loss 0.114356 LR 0.000125 Time 0.022133 +2023-10-02 21:59:29,885 - Epoch: [187][ 820/ 1236] Overall Loss 0.114417 Objective Loss 0.114417 LR 0.000125 Time 0.022125 +2023-10-02 21:59:30,100 - Epoch: [187][ 830/ 1236] Overall Loss 0.114287 Objective Loss 0.114287 LR 0.000125 Time 0.022117 +2023-10-02 21:59:30,314 - Epoch: [187][ 840/ 1236] Overall Loss 0.114294 Objective Loss 0.114294 LR 0.000125 Time 0.022108 +2023-10-02 21:59:30,529 - Epoch: [187][ 850/ 1236] Overall Loss 0.114441 Objective Loss 0.114441 LR 0.000125 Time 0.022099 +2023-10-02 21:59:30,743 - Epoch: [187][ 860/ 1236] Overall Loss 0.114420 Objective Loss 0.114420 LR 0.000125 Time 0.022092 +2023-10-02 21:59:30,958 - Epoch: [187][ 870/ 1236] Overall Loss 0.114444 Objective Loss 0.114444 LR 0.000125 Time 0.022083 +2023-10-02 21:59:31,172 - Epoch: [187][ 880/ 1236] Overall Loss 0.114300 Objective Loss 0.114300 LR 0.000125 Time 0.022075 +2023-10-02 21:59:31,387 - Epoch: [187][ 890/ 1236] Overall Loss 0.114556 Objective Loss 0.114556 LR 0.000125 Time 0.022068 +2023-10-02 21:59:31,601 - Epoch: [187][ 900/ 1236] Overall Loss 0.114518 Objective Loss 0.114518 LR 0.000125 Time 0.022061 +2023-10-02 21:59:31,816 - Epoch: [187][ 910/ 1236] Overall Loss 0.114550 Objective Loss 0.114550 LR 0.000125 Time 0.022054 +2023-10-02 21:59:32,030 - Epoch: [187][ 920/ 1236] Overall Loss 0.114558 Objective Loss 0.114558 LR 0.000125 Time 0.022046 +2023-10-02 21:59:32,244 - Epoch: [187][ 930/ 1236] Overall Loss 0.114414 Objective Loss 0.114414 LR 0.000125 Time 0.022039 +2023-10-02 21:59:32,459 - Epoch: [187][ 940/ 1236] Overall Loss 0.114303 Objective Loss 0.114303 LR 0.000125 Time 0.022033 +2023-10-02 21:59:32,674 - Epoch: [187][ 950/ 1236] Overall Loss 0.114311 Objective Loss 0.114311 LR 0.000125 Time 0.022027 +2023-10-02 21:59:32,888 - Epoch: [187][ 960/ 1236] Overall Loss 0.114398 Objective Loss 0.114398 LR 0.000125 Time 0.022020 +2023-10-02 21:59:33,103 - Epoch: [187][ 970/ 1236] Overall Loss 0.114478 Objective Loss 0.114478 LR 0.000125 Time 0.022014 +2023-10-02 21:59:33,317 - Epoch: [187][ 980/ 1236] Overall Loss 0.114617 Objective Loss 0.114617 LR 0.000125 Time 0.022008 +2023-10-02 21:59:33,532 - Epoch: [187][ 990/ 1236] Overall Loss 0.114547 Objective Loss 0.114547 LR 0.000125 Time 0.022002 +2023-10-02 21:59:33,747 - Epoch: [187][ 1000/ 1236] Overall Loss 0.114408 Objective Loss 0.114408 LR 0.000125 Time 0.021996 +2023-10-02 21:59:33,961 - Epoch: [187][ 1010/ 1236] Overall Loss 0.114497 Objective Loss 0.114497 LR 0.000125 Time 0.021990 +2023-10-02 21:59:34,175 - Epoch: [187][ 1020/ 1236] Overall Loss 0.114540 Objective Loss 0.114540 LR 0.000125 Time 0.021984 +2023-10-02 21:59:34,390 - Epoch: [187][ 1030/ 1236] Overall Loss 0.114560 Objective Loss 0.114560 LR 0.000125 Time 0.021979 +2023-10-02 21:59:34,604 - Epoch: [187][ 1040/ 1236] Overall Loss 0.114491 Objective Loss 0.114491 LR 0.000125 Time 0.021973 +2023-10-02 21:59:34,819 - Epoch: [187][ 1050/ 1236] Overall Loss 0.114376 Objective Loss 0.114376 LR 0.000125 Time 0.021968 +2023-10-02 21:59:35,034 - Epoch: [187][ 1060/ 1236] Overall Loss 0.114356 Objective Loss 0.114356 LR 0.000125 Time 0.021963 +2023-10-02 21:59:35,249 - Epoch: [187][ 1070/ 1236] Overall Loss 0.114367 Objective Loss 0.114367 LR 0.000125 Time 0.021958 +2023-10-02 21:59:35,463 - Epoch: [187][ 1080/ 1236] Overall Loss 0.114259 Objective Loss 0.114259 LR 0.000125 Time 0.021953 +2023-10-02 21:59:35,678 - Epoch: [187][ 1090/ 1236] Overall Loss 0.114291 Objective Loss 0.114291 LR 0.000125 Time 0.021948 +2023-10-02 21:59:35,892 - Epoch: [187][ 1100/ 1236] Overall Loss 0.114247 Objective Loss 0.114247 LR 0.000125 Time 0.021943 +2023-10-02 21:59:36,107 - Epoch: [187][ 1110/ 1236] Overall Loss 0.114373 Objective Loss 0.114373 LR 0.000125 Time 0.021939 +2023-10-02 21:59:36,325 - Epoch: [187][ 1120/ 1236] Overall Loss 0.114354 Objective Loss 0.114354 LR 0.000125 Time 0.021932 +2023-10-02 21:59:36,540 - Epoch: [187][ 1130/ 1236] Overall Loss 0.114202 Objective Loss 0.114202 LR 0.000125 Time 0.021928 +2023-10-02 21:59:36,755 - Epoch: [187][ 1140/ 1236] Overall Loss 0.114140 Objective Loss 0.114140 LR 0.000125 Time 0.021924 +2023-10-02 21:59:36,970 - Epoch: [187][ 1150/ 1236] Overall Loss 0.114204 Objective Loss 0.114204 LR 0.000125 Time 0.021919 +2023-10-02 21:59:37,184 - Epoch: [187][ 1160/ 1236] Overall Loss 0.114252 Objective Loss 0.114252 LR 0.000125 Time 0.021915 +2023-10-02 21:59:37,399 - Epoch: [187][ 1170/ 1236] Overall Loss 0.114326 Objective Loss 0.114326 LR 0.000125 Time 0.021911 +2023-10-02 21:59:37,617 - Epoch: [187][ 1180/ 1236] Overall Loss 0.114344 Objective Loss 0.114344 LR 0.000125 Time 0.021909 +2023-10-02 21:59:37,839 - Epoch: [187][ 1190/ 1236] Overall Loss 0.114292 Objective Loss 0.114292 LR 0.000125 Time 0.021912 +2023-10-02 21:59:38,054 - Epoch: [187][ 1200/ 1236] Overall Loss 0.114237 Objective Loss 0.114237 LR 0.000125 Time 0.021908 +2023-10-02 21:59:38,271 - Epoch: [187][ 1210/ 1236] Overall Loss 0.114320 Objective Loss 0.114320 LR 0.000125 Time 0.021906 +2023-10-02 21:59:38,486 - Epoch: [187][ 1220/ 1236] Overall Loss 0.114312 Objective Loss 0.114312 LR 0.000125 Time 0.021902 +2023-10-02 21:59:38,753 - Epoch: [187][ 1230/ 1236] Overall Loss 0.114212 Objective Loss 0.114212 LR 0.000125 Time 0.021941 +2023-10-02 21:59:38,877 - Epoch: [187][ 1236/ 1236] Overall Loss 0.114150 Objective Loss 0.114150 Top1 94.704684 Top5 99.185336 LR 0.000125 Time 0.021935 +2023-10-02 21:59:39,026 - --- validate (epoch=187)----------- +2023-10-02 21:59:39,027 - 29943 samples (256 per mini-batch) +2023-10-02 21:59:39,527 - Epoch: [187][ 10/ 117] Loss 0.306291 Top1 87.734375 Top5 98.554688 +2023-10-02 21:59:39,680 - Epoch: [187][ 20/ 117] Loss 0.329834 Top1 87.519531 Top5 98.437500 +2023-10-02 21:59:39,832 - Epoch: [187][ 30/ 117] Loss 0.319257 Top1 87.486979 Top5 98.528646 +2023-10-02 21:59:39,983 - Epoch: [187][ 40/ 117] Loss 0.310803 Top1 87.597656 Top5 98.671875 +2023-10-02 21:59:40,135 - Epoch: [187][ 50/ 117] Loss 0.312652 Top1 87.640625 Top5 98.679688 +2023-10-02 21:59:40,285 - Epoch: [187][ 60/ 117] Loss 0.315730 Top1 87.630208 Top5 98.639323 +2023-10-02 21:59:40,438 - Epoch: [187][ 70/ 117] Loss 0.313977 Top1 87.661830 Top5 98.632812 +2023-10-02 21:59:40,589 - Epoch: [187][ 80/ 117] Loss 0.310855 Top1 87.724609 Top5 98.647461 +2023-10-02 21:59:40,741 - Epoch: [187][ 90/ 117] Loss 0.311283 Top1 87.656250 Top5 98.637153 +2023-10-02 21:59:40,892 - Epoch: [187][ 100/ 117] Loss 0.307924 Top1 87.726562 Top5 98.648438 +2023-10-02 21:59:41,050 - Epoch: [187][ 110/ 117] Loss 0.307522 Top1 87.720170 Top5 98.664773 +2023-10-02 21:59:41,141 - Epoch: [187][ 117/ 117] Loss 0.310824 Top1 87.659887 Top5 98.664129 +2023-10-02 21:59:41,236 - ==> Top1: 87.660 Top5: 98.664 Loss: 0.311 + +2023-10-02 21:59:41,237 - ==> Confusion: +[[ 949 0 4 1 1 2 0 0 3 57 3 0 1 1 5 0 1 0 1 0 21] + [ 1 1058 1 1 4 18 0 26 0 1 0 0 0 0 0 3 1 0 7 3 7] + [ 1 1 988 9 0 1 12 8 0 1 1 0 8 3 0 3 2 1 9 1 7] + [ 1 1 16 978 0 0 2 1 2 1 3 1 6 3 30 3 1 4 15 1 20] + [ 25 3 0 0 977 4 0 0 0 11 0 0 1 2 8 6 7 0 1 2 3] + [ 4 29 0 0 5 990 1 24 1 6 1 8 2 10 4 0 3 0 4 1 23] + [ 0 2 30 2 0 1 1124 7 0 0 4 1 0 0 0 6 0 1 1 7 5] + [ 4 8 13 0 4 17 5 1086 1 3 5 6 4 4 1 1 0 1 34 10 11] + [ 17 2 0 1 1 5 0 2 984 31 9 2 1 13 13 0 1 1 3 0 3] + [ 99 2 2 1 5 2 0 0 19 948 0 0 1 22 6 2 0 0 0 2 8] + [ 1 0 12 9 0 2 2 2 10 2 970 1 0 12 5 1 4 3 3 1 13] + [ 0 0 0 0 2 12 0 5 0 0 0 973 16 6 0 1 0 12 0 3 5] + [ 0 0 1 0 0 0 1 2 0 1 3 32 974 2 5 7 0 13 5 7 15] + [ 0 0 2 0 4 6 0 0 13 8 4 7 0 1053 4 0 0 0 0 0 18] + [ 11 0 6 13 5 0 0 0 17 2 1 0 1 2 1022 0 1 2 11 0 7] + [ 0 0 1 1 6 0 0 0 0 0 0 5 9 0 0 1070 15 9 2 10 6] + [ 1 13 1 0 5 5 2 1 1 1 0 4 0 2 2 8 1099 0 1 4 11] + [ 0 0 2 1 0 0 1 0 0 0 0 5 17 1 2 6 0 994 0 3 6] + [ 3 2 4 12 0 0 0 21 4 1 3 1 1 0 9 0 0 1 993 0 13] + [ 0 2 2 2 2 0 6 7 0 0 1 12 5 4 1 2 6 1 1 1090 8] + [ 95 106 125 70 50 107 25 95 68 55 135 85 256 219 107 43 65 43 93 135 5928]] + +2023-10-02 21:59:41,238 - ==> Best [Top1: 87.683 Top5: 98.731 Sparsity:0.00 Params: 169472 on epoch: 182] +2023-10-02 21:59:41,238 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 21:59:41,244 - + +2023-10-02 21:59:41,245 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 21:59:42,413 - Epoch: [188][ 10/ 1236] Overall Loss 0.113399 Objective Loss 0.113399 LR 0.000125 Time 0.116764 +2023-10-02 21:59:42,624 - Epoch: [188][ 20/ 1236] Overall Loss 0.111641 Objective Loss 0.111641 LR 0.000125 Time 0.068896 +2023-10-02 21:59:42,833 - Epoch: [188][ 30/ 1236] Overall Loss 0.113819 Objective Loss 0.113819 LR 0.000125 Time 0.052909 +2023-10-02 21:59:43,045 - Epoch: [188][ 40/ 1236] Overall Loss 0.113974 Objective Loss 0.113974 LR 0.000125 Time 0.044967 +2023-10-02 21:59:43,253 - Epoch: [188][ 50/ 1236] Overall Loss 0.115395 Objective Loss 0.115395 LR 0.000125 Time 0.040123 +2023-10-02 21:59:43,465 - Epoch: [188][ 60/ 1236] Overall Loss 0.114398 Objective Loss 0.114398 LR 0.000125 Time 0.036960 +2023-10-02 21:59:43,672 - Epoch: [188][ 70/ 1236] Overall Loss 0.113180 Objective Loss 0.113180 LR 0.000125 Time 0.034645 +2023-10-02 21:59:43,884 - Epoch: [188][ 80/ 1236] Overall Loss 0.113706 Objective Loss 0.113706 LR 0.000125 Time 0.032952 +2023-10-02 21:59:44,091 - Epoch: [188][ 90/ 1236] Overall Loss 0.112315 Objective Loss 0.112315 LR 0.000125 Time 0.031595 +2023-10-02 21:59:44,303 - Epoch: [188][ 100/ 1236] Overall Loss 0.111771 Objective Loss 0.111771 LR 0.000125 Time 0.030546 +2023-10-02 21:59:44,511 - Epoch: [188][ 110/ 1236] Overall Loss 0.111858 Objective Loss 0.111858 LR 0.000125 Time 0.029656 +2023-10-02 21:59:44,721 - Epoch: [188][ 120/ 1236] Overall Loss 0.113329 Objective Loss 0.113329 LR 0.000125 Time 0.028932 +2023-10-02 21:59:44,928 - Epoch: [188][ 130/ 1236] Overall Loss 0.113919 Objective Loss 0.113919 LR 0.000125 Time 0.028293 +2023-10-02 21:59:45,137 - Epoch: [188][ 140/ 1236] Overall Loss 0.113693 Objective Loss 0.113693 LR 0.000125 Time 0.027763 +2023-10-02 21:59:45,345 - Epoch: [188][ 150/ 1236] Overall Loss 0.113280 Objective Loss 0.113280 LR 0.000125 Time 0.027287 +2023-10-02 21:59:45,554 - Epoch: [188][ 160/ 1236] Overall Loss 0.112708 Objective Loss 0.112708 LR 0.000125 Time 0.026887 +2023-10-02 21:59:45,762 - Epoch: [188][ 170/ 1236] Overall Loss 0.112604 Objective Loss 0.112604 LR 0.000125 Time 0.026519 +2023-10-02 21:59:45,971 - Epoch: [188][ 180/ 1236] Overall Loss 0.111906 Objective Loss 0.111906 LR 0.000125 Time 0.026205 +2023-10-02 21:59:46,179 - Epoch: [188][ 190/ 1236] Overall Loss 0.112152 Objective Loss 0.112152 LR 0.000125 Time 0.025912 +2023-10-02 21:59:46,388 - Epoch: [188][ 200/ 1236] Overall Loss 0.111524 Objective Loss 0.111524 LR 0.000125 Time 0.025658 +2023-10-02 21:59:46,596 - Epoch: [188][ 210/ 1236] Overall Loss 0.111542 Objective Loss 0.111542 LR 0.000125 Time 0.025424 +2023-10-02 21:59:46,805 - Epoch: [188][ 220/ 1236] Overall Loss 0.111300 Objective Loss 0.111300 LR 0.000125 Time 0.025219 +2023-10-02 21:59:47,013 - Epoch: [188][ 230/ 1236] Overall Loss 0.110905 Objective Loss 0.110905 LR 0.000125 Time 0.025019 +2023-10-02 21:59:47,222 - Epoch: [188][ 240/ 1236] Overall Loss 0.109933 Objective Loss 0.109933 LR 0.000125 Time 0.024847 +2023-10-02 21:59:47,430 - Epoch: [188][ 250/ 1236] Overall Loss 0.109986 Objective Loss 0.109986 LR 0.000125 Time 0.024678 +2023-10-02 21:59:47,639 - Epoch: [188][ 260/ 1236] Overall Loss 0.110206 Objective Loss 0.110206 LR 0.000125 Time 0.024532 +2023-10-02 21:59:47,846 - Epoch: [188][ 270/ 1236] Overall Loss 0.110446 Objective Loss 0.110446 LR 0.000125 Time 0.024383 +2023-10-02 21:59:48,055 - Epoch: [188][ 280/ 1236] Overall Loss 0.110527 Objective Loss 0.110527 LR 0.000125 Time 0.024258 +2023-10-02 21:59:48,262 - Epoch: [188][ 290/ 1236] Overall Loss 0.110684 Objective Loss 0.110684 LR 0.000125 Time 0.024132 +2023-10-02 21:59:48,472 - Epoch: [188][ 300/ 1236] Overall Loss 0.110790 Objective Loss 0.110790 LR 0.000125 Time 0.024024 +2023-10-02 21:59:48,680 - Epoch: [188][ 310/ 1236] Overall Loss 0.110640 Objective Loss 0.110640 LR 0.000125 Time 0.023916 +2023-10-02 21:59:48,889 - Epoch: [188][ 320/ 1236] Overall Loss 0.110636 Objective Loss 0.110636 LR 0.000125 Time 0.023820 +2023-10-02 21:59:49,096 - Epoch: [188][ 330/ 1236] Overall Loss 0.110712 Objective Loss 0.110712 LR 0.000125 Time 0.023724 +2023-10-02 21:59:49,306 - Epoch: [188][ 340/ 1236] Overall Loss 0.110424 Objective Loss 0.110424 LR 0.000125 Time 0.023641 +2023-10-02 21:59:49,514 - Epoch: [188][ 350/ 1236] Overall Loss 0.110290 Objective Loss 0.110290 LR 0.000125 Time 0.023555 +2023-10-02 21:59:49,725 - Epoch: [188][ 360/ 1236] Overall Loss 0.110413 Objective Loss 0.110413 LR 0.000125 Time 0.023487 +2023-10-02 21:59:49,933 - Epoch: [188][ 370/ 1236] Overall Loss 0.110087 Objective Loss 0.110087 LR 0.000125 Time 0.023413 +2023-10-02 21:59:50,145 - Epoch: [188][ 380/ 1236] Overall Loss 0.110255 Objective Loss 0.110255 LR 0.000125 Time 0.023353 +2023-10-02 21:59:50,353 - Epoch: [188][ 390/ 1236] Overall Loss 0.110518 Objective Loss 0.110518 LR 0.000125 Time 0.023287 +2023-10-02 21:59:50,563 - Epoch: [188][ 400/ 1236] Overall Loss 0.110599 Objective Loss 0.110599 LR 0.000125 Time 0.023230 +2023-10-02 21:59:50,772 - Epoch: [188][ 410/ 1236] Overall Loss 0.110449 Objective Loss 0.110449 LR 0.000125 Time 0.023169 +2023-10-02 21:59:50,984 - Epoch: [188][ 420/ 1236] Overall Loss 0.110867 Objective Loss 0.110867 LR 0.000125 Time 0.023121 +2023-10-02 21:59:51,192 - Epoch: [188][ 430/ 1236] Overall Loss 0.110890 Objective Loss 0.110890 LR 0.000125 Time 0.023066 +2023-10-02 21:59:51,404 - Epoch: [188][ 440/ 1236] Overall Loss 0.110636 Objective Loss 0.110636 LR 0.000125 Time 0.023023 +2023-10-02 21:59:51,612 - Epoch: [188][ 450/ 1236] Overall Loss 0.110441 Objective Loss 0.110441 LR 0.000125 Time 0.022973 +2023-10-02 21:59:51,824 - Epoch: [188][ 460/ 1236] Overall Loss 0.110125 Objective Loss 0.110125 LR 0.000125 Time 0.022933 +2023-10-02 21:59:52,031 - Epoch: [188][ 470/ 1236] Overall Loss 0.110610 Objective Loss 0.110610 LR 0.000125 Time 0.022887 +2023-10-02 21:59:52,243 - Epoch: [188][ 480/ 1236] Overall Loss 0.111279 Objective Loss 0.111279 LR 0.000125 Time 0.022851 +2023-10-02 21:59:52,451 - Epoch: [188][ 490/ 1236] Overall Loss 0.111363 Objective Loss 0.111363 LR 0.000125 Time 0.022808 +2023-10-02 21:59:52,663 - Epoch: [188][ 500/ 1236] Overall Loss 0.111310 Objective Loss 0.111310 LR 0.000125 Time 0.022776 +2023-10-02 21:59:52,871 - Epoch: [188][ 510/ 1236] Overall Loss 0.111297 Objective Loss 0.111297 LR 0.000125 Time 0.022736 +2023-10-02 21:59:53,083 - Epoch: [188][ 520/ 1236] Overall Loss 0.111284 Objective Loss 0.111284 LR 0.000125 Time 0.022706 +2023-10-02 21:59:53,291 - Epoch: [188][ 530/ 1236] Overall Loss 0.111080 Objective Loss 0.111080 LR 0.000125 Time 0.022669 +2023-10-02 21:59:53,502 - Epoch: [188][ 540/ 1236] Overall Loss 0.110806 Objective Loss 0.110806 LR 0.000125 Time 0.022639 +2023-10-02 21:59:53,712 - Epoch: [188][ 550/ 1236] Overall Loss 0.110776 Objective Loss 0.110776 LR 0.000125 Time 0.022606 +2023-10-02 21:59:53,924 - Epoch: [188][ 560/ 1236] Overall Loss 0.110631 Objective Loss 0.110631 LR 0.000125 Time 0.022581 +2023-10-02 21:59:54,132 - Epoch: [188][ 570/ 1236] Overall Loss 0.110834 Objective Loss 0.110834 LR 0.000125 Time 0.022549 +2023-10-02 21:59:54,343 - Epoch: [188][ 580/ 1236] Overall Loss 0.110934 Objective Loss 0.110934 LR 0.000125 Time 0.022523 +2023-10-02 21:59:54,552 - Epoch: [188][ 590/ 1236] Overall Loss 0.111022 Objective Loss 0.111022 LR 0.000125 Time 0.022493 +2023-10-02 21:59:54,764 - Epoch: [188][ 600/ 1236] Overall Loss 0.110648 Objective Loss 0.110648 LR 0.000125 Time 0.022471 +2023-10-02 21:59:54,972 - Epoch: [188][ 610/ 1236] Overall Loss 0.110814 Objective Loss 0.110814 LR 0.000125 Time 0.022443 +2023-10-02 21:59:55,184 - Epoch: [188][ 620/ 1236] Overall Loss 0.111144 Objective Loss 0.111144 LR 0.000125 Time 0.022423 +2023-10-02 21:59:55,392 - Epoch: [188][ 630/ 1236] Overall Loss 0.111155 Objective Loss 0.111155 LR 0.000125 Time 0.022397 +2023-10-02 21:59:55,604 - Epoch: [188][ 640/ 1236] Overall Loss 0.111332 Objective Loss 0.111332 LR 0.000125 Time 0.022378 +2023-10-02 21:59:55,812 - Epoch: [188][ 650/ 1236] Overall Loss 0.111180 Objective Loss 0.111180 LR 0.000125 Time 0.022353 +2023-10-02 21:59:56,024 - Epoch: [188][ 660/ 1236] Overall Loss 0.111267 Objective Loss 0.111267 LR 0.000125 Time 0.022335 +2023-10-02 21:59:56,233 - Epoch: [188][ 670/ 1236] Overall Loss 0.111218 Objective Loss 0.111218 LR 0.000125 Time 0.022312 +2023-10-02 21:59:56,445 - Epoch: [188][ 680/ 1236] Overall Loss 0.111039 Objective Loss 0.111039 LR 0.000125 Time 0.022296 +2023-10-02 21:59:56,653 - Epoch: [188][ 690/ 1236] Overall Loss 0.110974 Objective Loss 0.110974 LR 0.000125 Time 0.022274 +2023-10-02 21:59:56,865 - Epoch: [188][ 700/ 1236] Overall Loss 0.110927 Objective Loss 0.110927 LR 0.000125 Time 0.022258 +2023-10-02 21:59:57,073 - Epoch: [188][ 710/ 1236] Overall Loss 0.110834 Objective Loss 0.110834 LR 0.000125 Time 0.022237 +2023-10-02 21:59:57,285 - Epoch: [188][ 720/ 1236] Overall Loss 0.111032 Objective Loss 0.111032 LR 0.000125 Time 0.022222 +2023-10-02 21:59:57,493 - Epoch: [188][ 730/ 1236] Overall Loss 0.111033 Objective Loss 0.111033 LR 0.000125 Time 0.022202 +2023-10-02 21:59:57,704 - Epoch: [188][ 740/ 1236] Overall Loss 0.111085 Objective Loss 0.111085 LR 0.000125 Time 0.022187 +2023-10-02 21:59:57,913 - Epoch: [188][ 750/ 1236] Overall Loss 0.111081 Objective Loss 0.111081 LR 0.000125 Time 0.022168 +2023-10-02 21:59:58,124 - Epoch: [188][ 760/ 1236] Overall Loss 0.111030 Objective Loss 0.111030 LR 0.000125 Time 0.022154 +2023-10-02 21:59:58,334 - Epoch: [188][ 770/ 1236] Overall Loss 0.111260 Objective Loss 0.111260 LR 0.000125 Time 0.022136 +2023-10-02 21:59:58,544 - Epoch: [188][ 780/ 1236] Overall Loss 0.111492 Objective Loss 0.111492 LR 0.000125 Time 0.022122 +2023-10-02 21:59:58,754 - Epoch: [188][ 790/ 1236] Overall Loss 0.111432 Objective Loss 0.111432 LR 0.000125 Time 0.022105 +2023-10-02 21:59:58,966 - Epoch: [188][ 800/ 1236] Overall Loss 0.111385 Objective Loss 0.111385 LR 0.000125 Time 0.022093 +2023-10-02 21:59:59,174 - Epoch: [188][ 810/ 1236] Overall Loss 0.111341 Objective Loss 0.111341 LR 0.000125 Time 0.022077 +2023-10-02 21:59:59,386 - Epoch: [188][ 820/ 1236] Overall Loss 0.111572 Objective Loss 0.111572 LR 0.000125 Time 0.022066 +2023-10-02 21:59:59,594 - Epoch: [188][ 830/ 1236] Overall Loss 0.111562 Objective Loss 0.111562 LR 0.000125 Time 0.022050 +2023-10-02 21:59:59,806 - Epoch: [188][ 840/ 1236] Overall Loss 0.111537 Objective Loss 0.111537 LR 0.000125 Time 0.022040 +2023-10-02 22:00:00,015 - Epoch: [188][ 850/ 1236] Overall Loss 0.111561 Objective Loss 0.111561 LR 0.000125 Time 0.022026 +2023-10-02 22:00:00,227 - Epoch: [188][ 860/ 1236] Overall Loss 0.111503 Objective Loss 0.111503 LR 0.000125 Time 0.022016 +2023-10-02 22:00:00,435 - Epoch: [188][ 870/ 1236] Overall Loss 0.111364 Objective Loss 0.111364 LR 0.000125 Time 0.022002 +2023-10-02 22:00:00,646 - Epoch: [188][ 880/ 1236] Overall Loss 0.111456 Objective Loss 0.111456 LR 0.000125 Time 0.021991 +2023-10-02 22:00:00,855 - Epoch: [188][ 890/ 1236] Overall Loss 0.111381 Objective Loss 0.111381 LR 0.000125 Time 0.021977 +2023-10-02 22:00:01,066 - Epoch: [188][ 900/ 1236] Overall Loss 0.111419 Objective Loss 0.111419 LR 0.000125 Time 0.021967 +2023-10-02 22:00:01,276 - Epoch: [188][ 910/ 1236] Overall Loss 0.111572 Objective Loss 0.111572 LR 0.000125 Time 0.021954 +2023-10-02 22:00:01,487 - Epoch: [188][ 920/ 1236] Overall Loss 0.111529 Objective Loss 0.111529 LR 0.000125 Time 0.021945 +2023-10-02 22:00:01,696 - Epoch: [188][ 930/ 1236] Overall Loss 0.111519 Objective Loss 0.111519 LR 0.000125 Time 0.021933 +2023-10-02 22:00:01,907 - Epoch: [188][ 940/ 1236] Overall Loss 0.111616 Objective Loss 0.111616 LR 0.000125 Time 0.021923 +2023-10-02 22:00:02,117 - Epoch: [188][ 950/ 1236] Overall Loss 0.111599 Objective Loss 0.111599 LR 0.000125 Time 0.021911 +2023-10-02 22:00:02,329 - Epoch: [188][ 960/ 1236] Overall Loss 0.111599 Objective Loss 0.111599 LR 0.000125 Time 0.021904 +2023-10-02 22:00:02,537 - Epoch: [188][ 970/ 1236] Overall Loss 0.111669 Objective Loss 0.111669 LR 0.000125 Time 0.021892 +2023-10-02 22:00:02,749 - Epoch: [188][ 980/ 1236] Overall Loss 0.111621 Objective Loss 0.111621 LR 0.000125 Time 0.021885 +2023-10-02 22:00:02,957 - Epoch: [188][ 990/ 1236] Overall Loss 0.111722 Objective Loss 0.111722 LR 0.000125 Time 0.021874 +2023-10-02 22:00:03,168 - Epoch: [188][ 1000/ 1236] Overall Loss 0.111758 Objective Loss 0.111758 LR 0.000125 Time 0.021866 +2023-10-02 22:00:03,378 - Epoch: [188][ 1010/ 1236] Overall Loss 0.111623 Objective Loss 0.111623 LR 0.000125 Time 0.021855 +2023-10-02 22:00:03,590 - Epoch: [188][ 1020/ 1236] Overall Loss 0.111730 Objective Loss 0.111730 LR 0.000125 Time 0.021849 +2023-10-02 22:00:03,799 - Epoch: [188][ 1030/ 1236] Overall Loss 0.111670 Objective Loss 0.111670 LR 0.000125 Time 0.021839 +2023-10-02 22:00:04,009 - Epoch: [188][ 1040/ 1236] Overall Loss 0.111798 Objective Loss 0.111798 LR 0.000125 Time 0.021831 +2023-10-02 22:00:04,219 - Epoch: [188][ 1050/ 1236] Overall Loss 0.111776 Objective Loss 0.111776 LR 0.000125 Time 0.021821 +2023-10-02 22:00:04,430 - Epoch: [188][ 1060/ 1236] Overall Loss 0.111746 Objective Loss 0.111746 LR 0.000125 Time 0.021814 +2023-10-02 22:00:04,639 - Epoch: [188][ 1070/ 1236] Overall Loss 0.111666 Objective Loss 0.111666 LR 0.000125 Time 0.021805 +2023-10-02 22:00:04,850 - Epoch: [188][ 1080/ 1236] Overall Loss 0.111639 Objective Loss 0.111639 LR 0.000125 Time 0.021798 +2023-10-02 22:00:05,061 - Epoch: [188][ 1090/ 1236] Overall Loss 0.111583 Objective Loss 0.111583 LR 0.000125 Time 0.021789 +2023-10-02 22:00:05,274 - Epoch: [188][ 1100/ 1236] Overall Loss 0.111605 Objective Loss 0.111605 LR 0.000125 Time 0.021785 +2023-10-02 22:00:05,485 - Epoch: [188][ 1110/ 1236] Overall Loss 0.111574 Objective Loss 0.111574 LR 0.000125 Time 0.021777 +2023-10-02 22:00:05,698 - Epoch: [188][ 1120/ 1236] Overall Loss 0.111672 Objective Loss 0.111672 LR 0.000125 Time 0.021773 +2023-10-02 22:00:05,910 - Epoch: [188][ 1130/ 1236] Overall Loss 0.111764 Objective Loss 0.111764 LR 0.000125 Time 0.021766 +2023-10-02 22:00:06,123 - Epoch: [188][ 1140/ 1236] Overall Loss 0.111906 Objective Loss 0.111906 LR 0.000125 Time 0.021762 +2023-10-02 22:00:06,335 - Epoch: [188][ 1150/ 1236] Overall Loss 0.111860 Objective Loss 0.111860 LR 0.000125 Time 0.021755 +2023-10-02 22:00:06,548 - Epoch: [188][ 1160/ 1236] Overall Loss 0.111923 Objective Loss 0.111923 LR 0.000125 Time 0.021752 +2023-10-02 22:00:06,760 - Epoch: [188][ 1170/ 1236] Overall Loss 0.111984 Objective Loss 0.111984 LR 0.000125 Time 0.021745 +2023-10-02 22:00:06,973 - Epoch: [188][ 1180/ 1236] Overall Loss 0.111907 Objective Loss 0.111907 LR 0.000125 Time 0.021741 +2023-10-02 22:00:07,184 - Epoch: [188][ 1190/ 1236] Overall Loss 0.111954 Objective Loss 0.111954 LR 0.000125 Time 0.021735 +2023-10-02 22:00:07,399 - Epoch: [188][ 1200/ 1236] Overall Loss 0.111824 Objective Loss 0.111824 LR 0.000125 Time 0.021732 +2023-10-02 22:00:07,609 - Epoch: [188][ 1210/ 1236] Overall Loss 0.111765 Objective Loss 0.111765 LR 0.000125 Time 0.021726 +2023-10-02 22:00:07,822 - Epoch: [188][ 1220/ 1236] Overall Loss 0.111858 Objective Loss 0.111858 LR 0.000125 Time 0.021722 +2023-10-02 22:00:08,088 - Epoch: [188][ 1230/ 1236] Overall Loss 0.111873 Objective Loss 0.111873 LR 0.000125 Time 0.021760 +2023-10-02 22:00:08,210 - Epoch: [188][ 1236/ 1236] Overall Loss 0.111905 Objective Loss 0.111905 Top1 91.446029 Top5 98.778004 LR 0.000125 Time 0.021754 +2023-10-02 22:00:08,343 - --- validate (epoch=188)----------- +2023-10-02 22:00:08,344 - 29943 samples (256 per mini-batch) +2023-10-02 22:00:08,839 - Epoch: [188][ 10/ 117] Loss 0.294355 Top1 88.242188 Top5 98.593750 +2023-10-02 22:00:08,993 - Epoch: [188][ 20/ 117] Loss 0.298507 Top1 87.812500 Top5 98.593750 +2023-10-02 22:00:09,146 - Epoch: [188][ 30/ 117] Loss 0.308089 Top1 87.734375 Top5 98.502604 +2023-10-02 22:00:09,298 - Epoch: [188][ 40/ 117] Loss 0.309323 Top1 87.734375 Top5 98.603516 +2023-10-02 22:00:09,451 - Epoch: [188][ 50/ 117] Loss 0.311208 Top1 87.773438 Top5 98.656250 +2023-10-02 22:00:09,602 - Epoch: [188][ 60/ 117] Loss 0.307452 Top1 87.701823 Top5 98.658854 +2023-10-02 22:00:09,754 - Epoch: [188][ 70/ 117] Loss 0.308022 Top1 87.795759 Top5 98.727679 +2023-10-02 22:00:09,905 - Epoch: [188][ 80/ 117] Loss 0.304276 Top1 87.788086 Top5 98.740234 +2023-10-02 22:00:10,057 - Epoch: [188][ 90/ 117] Loss 0.303114 Top1 87.782118 Top5 98.710938 +2023-10-02 22:00:10,209 - Epoch: [188][ 100/ 117] Loss 0.301956 Top1 87.890625 Top5 98.703125 +2023-10-02 22:00:10,367 - Epoch: [188][ 110/ 117] Loss 0.304410 Top1 87.794744 Top5 98.696733 +2023-10-02 22:00:10,458 - Epoch: [188][ 117/ 117] Loss 0.305222 Top1 87.753398 Top5 98.700865 +2023-10-02 22:00:10,602 - ==> Top1: 87.753 Top5: 98.701 Loss: 0.305 + +2023-10-02 22:00:10,602 - ==> Confusion: +[[ 937 0 5 0 4 2 0 0 3 64 2 0 0 2 7 0 1 0 1 0 22] + [ 0 1070 0 2 4 11 2 20 0 1 0 0 1 0 0 2 0 0 4 2 12] + [ 2 0 989 5 0 0 12 9 0 2 0 1 7 2 1 2 2 2 9 2 9] + [ 1 3 17 977 0 1 0 2 4 1 4 1 6 3 26 5 1 4 13 1 19] + [ 24 4 1 0 972 4 1 0 0 13 0 0 0 4 8 6 9 0 0 0 4] + [ 5 36 0 1 5 994 2 17 1 6 2 6 3 9 4 0 2 0 4 3 16] + [ 0 2 23 1 0 0 1133 6 0 0 3 1 0 0 0 5 0 1 3 7 6] + [ 3 10 14 0 5 21 7 1073 3 2 3 2 6 5 1 1 0 1 39 10 12] + [ 15 2 0 1 1 4 0 0 974 42 11 3 0 12 12 0 0 1 3 3 5] + [ 86 1 0 2 5 3 0 0 25 955 0 2 0 20 6 2 0 0 0 2 10] + [ 2 1 9 9 0 1 3 2 6 1 972 3 2 10 6 0 1 2 4 3 16] + [ 0 0 1 0 0 12 0 7 1 0 0 964 18 4 0 2 0 14 1 4 7] + [ 0 0 1 1 0 1 2 1 0 1 3 29 981 0 2 7 1 13 2 4 19] + [ 0 0 1 0 4 5 0 0 8 9 2 7 0 1060 4 0 0 0 0 2 17] + [ 12 0 4 11 2 1 0 0 19 2 1 0 3 4 1020 0 1 2 11 0 8] + [ 0 0 2 1 5 2 1 0 0 0 0 5 8 0 0 1074 14 9 0 8 5] + [ 1 16 1 0 5 6 1 0 0 0 0 5 0 3 3 7 1093 0 2 5 13] + [ 0 0 1 1 0 0 2 0 0 0 0 4 18 1 2 5 0 994 0 4 6] + [ 2 5 3 13 0 0 0 18 5 1 2 1 1 0 14 0 0 0 990 0 13] + [ 0 1 4 3 1 1 6 5 0 0 0 14 5 2 1 0 6 2 1 1093 7] + [ 105 106 107 72 51 111 28 74 69 56 138 73 284 212 102 45 55 44 90 122 5961]] + +2023-10-02 22:00:10,604 - ==> Best [Top1: 87.753 Top5: 98.701 Sparsity:0.00 Params: 169472 on epoch: 188] +2023-10-02 22:00:10,604 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:00:10,618 - + +2023-10-02 22:00:10,618 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:00:11,657 - Epoch: [189][ 10/ 1236] Overall Loss 0.100155 Objective Loss 0.100155 LR 0.000125 Time 0.103889 +2023-10-02 22:00:11,866 - Epoch: [189][ 20/ 1236] Overall Loss 0.100511 Objective Loss 0.100511 LR 0.000125 Time 0.062356 +2023-10-02 22:00:12,074 - Epoch: [189][ 30/ 1236] Overall Loss 0.107501 Objective Loss 0.107501 LR 0.000125 Time 0.048496 +2023-10-02 22:00:12,282 - Epoch: [189][ 40/ 1236] Overall Loss 0.110072 Objective Loss 0.110072 LR 0.000125 Time 0.041572 +2023-10-02 22:00:12,490 - Epoch: [189][ 50/ 1236] Overall Loss 0.113337 Objective Loss 0.113337 LR 0.000125 Time 0.037380 +2023-10-02 22:00:12,698 - Epoch: [189][ 60/ 1236] Overall Loss 0.113758 Objective Loss 0.113758 LR 0.000125 Time 0.034610 +2023-10-02 22:00:12,906 - Epoch: [189][ 70/ 1236] Overall Loss 0.113790 Objective Loss 0.113790 LR 0.000125 Time 0.032632 +2023-10-02 22:00:13,114 - Epoch: [189][ 80/ 1236] Overall Loss 0.112711 Objective Loss 0.112711 LR 0.000125 Time 0.031152 +2023-10-02 22:00:13,322 - Epoch: [189][ 90/ 1236] Overall Loss 0.114485 Objective Loss 0.114485 LR 0.000125 Time 0.029982 +2023-10-02 22:00:13,530 - Epoch: [189][ 100/ 1236] Overall Loss 0.113936 Objective Loss 0.113936 LR 0.000125 Time 0.029063 +2023-10-02 22:00:13,738 - Epoch: [189][ 110/ 1236] Overall Loss 0.111999 Objective Loss 0.111999 LR 0.000125 Time 0.028297 +2023-10-02 22:00:13,947 - Epoch: [189][ 120/ 1236] Overall Loss 0.110455 Objective Loss 0.110455 LR 0.000125 Time 0.027674 +2023-10-02 22:00:14,154 - Epoch: [189][ 130/ 1236] Overall Loss 0.109944 Objective Loss 0.109944 LR 0.000125 Time 0.027127 +2023-10-02 22:00:14,363 - Epoch: [189][ 140/ 1236] Overall Loss 0.109846 Objective Loss 0.109846 LR 0.000125 Time 0.026677 +2023-10-02 22:00:14,570 - Epoch: [189][ 150/ 1236] Overall Loss 0.109726 Objective Loss 0.109726 LR 0.000125 Time 0.026271 +2023-10-02 22:00:14,778 - Epoch: [189][ 160/ 1236] Overall Loss 0.109733 Objective Loss 0.109733 LR 0.000125 Time 0.025930 +2023-10-02 22:00:14,986 - Epoch: [189][ 170/ 1236] Overall Loss 0.110559 Objective Loss 0.110559 LR 0.000125 Time 0.025615 +2023-10-02 22:00:15,196 - Epoch: [189][ 180/ 1236] Overall Loss 0.109057 Objective Loss 0.109057 LR 0.000125 Time 0.025357 +2023-10-02 22:00:15,402 - Epoch: [189][ 190/ 1236] Overall Loss 0.109837 Objective Loss 0.109837 LR 0.000125 Time 0.025105 +2023-10-02 22:00:15,611 - Epoch: [189][ 200/ 1236] Overall Loss 0.110471 Objective Loss 0.110471 LR 0.000125 Time 0.024894 +2023-10-02 22:00:15,818 - Epoch: [189][ 210/ 1236] Overall Loss 0.110277 Objective Loss 0.110277 LR 0.000125 Time 0.024687 +2023-10-02 22:00:16,027 - Epoch: [189][ 220/ 1236] Overall Loss 0.110649 Objective Loss 0.110649 LR 0.000125 Time 0.024513 +2023-10-02 22:00:16,234 - Epoch: [189][ 230/ 1236] Overall Loss 0.110564 Objective Loss 0.110564 LR 0.000125 Time 0.024342 +2023-10-02 22:00:16,442 - Epoch: [189][ 240/ 1236] Overall Loss 0.111240 Objective Loss 0.111240 LR 0.000125 Time 0.024193 +2023-10-02 22:00:16,650 - Epoch: [189][ 250/ 1236] Overall Loss 0.111288 Objective Loss 0.111288 LR 0.000125 Time 0.024052 +2023-10-02 22:00:16,857 - Epoch: [189][ 260/ 1236] Overall Loss 0.110570 Objective Loss 0.110570 LR 0.000125 Time 0.023925 +2023-10-02 22:00:17,065 - Epoch: [189][ 270/ 1236] Overall Loss 0.110670 Objective Loss 0.110670 LR 0.000125 Time 0.023802 +2023-10-02 22:00:17,271 - Epoch: [189][ 280/ 1236] Overall Loss 0.110853 Objective Loss 0.110853 LR 0.000125 Time 0.023688 +2023-10-02 22:00:17,479 - Epoch: [189][ 290/ 1236] Overall Loss 0.110798 Objective Loss 0.110798 LR 0.000125 Time 0.023586 +2023-10-02 22:00:17,686 - Epoch: [189][ 300/ 1236] Overall Loss 0.110286 Objective Loss 0.110286 LR 0.000125 Time 0.023487 +2023-10-02 22:00:17,893 - Epoch: [189][ 310/ 1236] Overall Loss 0.110675 Objective Loss 0.110675 LR 0.000125 Time 0.023399 +2023-10-02 22:00:18,100 - Epoch: [189][ 320/ 1236] Overall Loss 0.110893 Objective Loss 0.110893 LR 0.000125 Time 0.023312 +2023-10-02 22:00:18,307 - Epoch: [189][ 330/ 1236] Overall Loss 0.110779 Objective Loss 0.110779 LR 0.000125 Time 0.023234 +2023-10-02 22:00:18,514 - Epoch: [189][ 340/ 1236] Overall Loss 0.110820 Objective Loss 0.110820 LR 0.000125 Time 0.023157 +2023-10-02 22:00:18,722 - Epoch: [189][ 350/ 1236] Overall Loss 0.110962 Objective Loss 0.110962 LR 0.000125 Time 0.023089 +2023-10-02 22:00:18,929 - Epoch: [189][ 360/ 1236] Overall Loss 0.110844 Objective Loss 0.110844 LR 0.000125 Time 0.023021 +2023-10-02 22:00:19,138 - Epoch: [189][ 370/ 1236] Overall Loss 0.110470 Objective Loss 0.110470 LR 0.000125 Time 0.022963 +2023-10-02 22:00:19,346 - Epoch: [189][ 380/ 1236] Overall Loss 0.110053 Objective Loss 0.110053 LR 0.000125 Time 0.022905 +2023-10-02 22:00:19,554 - Epoch: [189][ 390/ 1236] Overall Loss 0.109857 Objective Loss 0.109857 LR 0.000125 Time 0.022847 +2023-10-02 22:00:19,763 - Epoch: [189][ 400/ 1236] Overall Loss 0.110087 Objective Loss 0.110087 LR 0.000125 Time 0.022798 +2023-10-02 22:00:19,972 - Epoch: [189][ 410/ 1236] Overall Loss 0.109921 Objective Loss 0.109921 LR 0.000125 Time 0.022746 +2023-10-02 22:00:20,181 - Epoch: [189][ 420/ 1236] Overall Loss 0.109631 Objective Loss 0.109631 LR 0.000125 Time 0.022702 +2023-10-02 22:00:20,388 - Epoch: [189][ 430/ 1236] Overall Loss 0.109804 Objective Loss 0.109804 LR 0.000125 Time 0.022652 +2023-10-02 22:00:20,597 - Epoch: [189][ 440/ 1236] Overall Loss 0.110006 Objective Loss 0.110006 LR 0.000125 Time 0.022612 +2023-10-02 22:00:20,806 - Epoch: [189][ 450/ 1236] Overall Loss 0.110444 Objective Loss 0.110444 LR 0.000125 Time 0.022570 +2023-10-02 22:00:21,016 - Epoch: [189][ 460/ 1236] Overall Loss 0.110482 Objective Loss 0.110482 LR 0.000125 Time 0.022534 +2023-10-02 22:00:21,224 - Epoch: [189][ 470/ 1236] Overall Loss 0.110562 Objective Loss 0.110562 LR 0.000125 Time 0.022494 +2023-10-02 22:00:21,431 - Epoch: [189][ 480/ 1236] Overall Loss 0.110703 Objective Loss 0.110703 LR 0.000125 Time 0.022457 +2023-10-02 22:00:21,640 - Epoch: [189][ 490/ 1236] Overall Loss 0.110640 Objective Loss 0.110640 LR 0.000125 Time 0.022420 +2023-10-02 22:00:21,848 - Epoch: [189][ 500/ 1236] Overall Loss 0.111128 Objective Loss 0.111128 LR 0.000125 Time 0.022387 +2023-10-02 22:00:22,057 - Epoch: [189][ 510/ 1236] Overall Loss 0.111357 Objective Loss 0.111357 LR 0.000125 Time 0.022355 +2023-10-02 22:00:22,266 - Epoch: [189][ 520/ 1236] Overall Loss 0.111519 Objective Loss 0.111519 LR 0.000125 Time 0.022327 +2023-10-02 22:00:22,475 - Epoch: [189][ 530/ 1236] Overall Loss 0.111764 Objective Loss 0.111764 LR 0.000125 Time 0.022297 +2023-10-02 22:00:22,685 - Epoch: [189][ 540/ 1236] Overall Loss 0.111756 Objective Loss 0.111756 LR 0.000125 Time 0.022271 +2023-10-02 22:00:22,893 - Epoch: [189][ 550/ 1236] Overall Loss 0.111840 Objective Loss 0.111840 LR 0.000125 Time 0.022245 +2023-10-02 22:00:23,102 - Epoch: [189][ 560/ 1236] Overall Loss 0.111731 Objective Loss 0.111731 LR 0.000125 Time 0.022220 +2023-10-02 22:00:23,311 - Epoch: [189][ 570/ 1236] Overall Loss 0.111635 Objective Loss 0.111635 LR 0.000125 Time 0.022194 +2023-10-02 22:00:23,520 - Epoch: [189][ 580/ 1236] Overall Loss 0.111809 Objective Loss 0.111809 LR 0.000125 Time 0.022171 +2023-10-02 22:00:23,728 - Epoch: [189][ 590/ 1236] Overall Loss 0.111854 Objective Loss 0.111854 LR 0.000125 Time 0.022145 +2023-10-02 22:00:23,937 - Epoch: [189][ 600/ 1236] Overall Loss 0.111691 Objective Loss 0.111691 LR 0.000125 Time 0.022124 +2023-10-02 22:00:24,144 - Epoch: [189][ 610/ 1236] Overall Loss 0.111732 Objective Loss 0.111732 LR 0.000125 Time 0.022097 +2023-10-02 22:00:24,352 - Epoch: [189][ 620/ 1236] Overall Loss 0.111922 Objective Loss 0.111922 LR 0.000125 Time 0.022076 +2023-10-02 22:00:24,560 - Epoch: [189][ 630/ 1236] Overall Loss 0.111789 Objective Loss 0.111789 LR 0.000125 Time 0.022053 +2023-10-02 22:00:24,769 - Epoch: [189][ 640/ 1236] Overall Loss 0.111789 Objective Loss 0.111789 LR 0.000125 Time 0.022034 +2023-10-02 22:00:24,976 - Epoch: [189][ 650/ 1236] Overall Loss 0.111929 Objective Loss 0.111929 LR 0.000125 Time 0.022011 +2023-10-02 22:00:25,185 - Epoch: [189][ 660/ 1236] Overall Loss 0.112075 Objective Loss 0.112075 LR 0.000125 Time 0.021994 +2023-10-02 22:00:25,393 - Epoch: [189][ 670/ 1236] Overall Loss 0.111969 Objective Loss 0.111969 LR 0.000125 Time 0.021974 +2023-10-02 22:00:25,602 - Epoch: [189][ 680/ 1236] Overall Loss 0.112028 Objective Loss 0.112028 LR 0.000125 Time 0.021958 +2023-10-02 22:00:25,810 - Epoch: [189][ 690/ 1236] Overall Loss 0.112111 Objective Loss 0.112111 LR 0.000125 Time 0.021941 +2023-10-02 22:00:26,019 - Epoch: [189][ 700/ 1236] Overall Loss 0.112292 Objective Loss 0.112292 LR 0.000125 Time 0.021925 +2023-10-02 22:00:26,228 - Epoch: [189][ 710/ 1236] Overall Loss 0.112396 Objective Loss 0.112396 LR 0.000125 Time 0.021909 +2023-10-02 22:00:26,437 - Epoch: [189][ 720/ 1236] Overall Loss 0.112171 Objective Loss 0.112171 LR 0.000125 Time 0.021895 +2023-10-02 22:00:26,646 - Epoch: [189][ 730/ 1236] Overall Loss 0.112455 Objective Loss 0.112455 LR 0.000125 Time 0.021880 +2023-10-02 22:00:26,855 - Epoch: [189][ 740/ 1236] Overall Loss 0.112430 Objective Loss 0.112430 LR 0.000125 Time 0.021867 +2023-10-02 22:00:27,064 - Epoch: [189][ 750/ 1236] Overall Loss 0.112519 Objective Loss 0.112519 LR 0.000125 Time 0.021851 +2023-10-02 22:00:27,273 - Epoch: [189][ 760/ 1236] Overall Loss 0.112525 Objective Loss 0.112525 LR 0.000125 Time 0.021838 +2023-10-02 22:00:27,482 - Epoch: [189][ 770/ 1236] Overall Loss 0.112741 Objective Loss 0.112741 LR 0.000125 Time 0.021824 +2023-10-02 22:00:27,691 - Epoch: [189][ 780/ 1236] Overall Loss 0.113051 Objective Loss 0.113051 LR 0.000125 Time 0.021812 +2023-10-02 22:00:27,900 - Epoch: [189][ 790/ 1236] Overall Loss 0.112855 Objective Loss 0.112855 LR 0.000125 Time 0.021799 +2023-10-02 22:00:28,109 - Epoch: [189][ 800/ 1236] Overall Loss 0.112880 Objective Loss 0.112880 LR 0.000125 Time 0.021788 +2023-10-02 22:00:28,318 - Epoch: [189][ 810/ 1236] Overall Loss 0.112741 Objective Loss 0.112741 LR 0.000125 Time 0.021775 +2023-10-02 22:00:28,527 - Epoch: [189][ 820/ 1236] Overall Loss 0.112701 Objective Loss 0.112701 LR 0.000125 Time 0.021764 +2023-10-02 22:00:28,737 - Epoch: [189][ 830/ 1236] Overall Loss 0.112851 Objective Loss 0.112851 LR 0.000125 Time 0.021755 +2023-10-02 22:00:28,949 - Epoch: [189][ 840/ 1236] Overall Loss 0.112710 Objective Loss 0.112710 LR 0.000125 Time 0.021747 +2023-10-02 22:00:29,159 - Epoch: [189][ 850/ 1236] Overall Loss 0.112814 Objective Loss 0.112814 LR 0.000125 Time 0.021738 +2023-10-02 22:00:29,371 - Epoch: [189][ 860/ 1236] Overall Loss 0.112947 Objective Loss 0.112947 LR 0.000125 Time 0.021731 +2023-10-02 22:00:29,581 - Epoch: [189][ 870/ 1236] Overall Loss 0.113018 Objective Loss 0.113018 LR 0.000125 Time 0.021721 +2023-10-02 22:00:29,793 - Epoch: [189][ 880/ 1236] Overall Loss 0.113072 Objective Loss 0.113072 LR 0.000125 Time 0.021714 +2023-10-02 22:00:30,004 - Epoch: [189][ 890/ 1236] Overall Loss 0.113020 Objective Loss 0.113020 LR 0.000125 Time 0.021705 +2023-10-02 22:00:30,217 - Epoch: [189][ 900/ 1236] Overall Loss 0.112915 Objective Loss 0.112915 LR 0.000125 Time 0.021700 +2023-10-02 22:00:30,429 - Epoch: [189][ 910/ 1236] Overall Loss 0.113010 Objective Loss 0.113010 LR 0.000125 Time 0.021695 +2023-10-02 22:00:30,644 - Epoch: [189][ 920/ 1236] Overall Loss 0.112941 Objective Loss 0.112941 LR 0.000125 Time 0.021692 +2023-10-02 22:00:30,858 - Epoch: [189][ 930/ 1236] Overall Loss 0.112762 Objective Loss 0.112762 LR 0.000125 Time 0.021688 +2023-10-02 22:00:31,072 - Epoch: [189][ 940/ 1236] Overall Loss 0.112867 Objective Loss 0.112867 LR 0.000125 Time 0.021685 +2023-10-02 22:00:31,286 - Epoch: [189][ 950/ 1236] Overall Loss 0.112854 Objective Loss 0.112854 LR 0.000125 Time 0.021682 +2023-10-02 22:00:31,500 - Epoch: [189][ 960/ 1236] Overall Loss 0.112721 Objective Loss 0.112721 LR 0.000125 Time 0.021678 +2023-10-02 22:00:31,714 - Epoch: [189][ 970/ 1236] Overall Loss 0.112534 Objective Loss 0.112534 LR 0.000125 Time 0.021675 +2023-10-02 22:00:31,936 - Epoch: [189][ 980/ 1236] Overall Loss 0.112711 Objective Loss 0.112711 LR 0.000125 Time 0.021680 +2023-10-02 22:00:32,158 - Epoch: [189][ 990/ 1236] Overall Loss 0.112769 Objective Loss 0.112769 LR 0.000125 Time 0.021684 +2023-10-02 22:00:32,382 - Epoch: [189][ 1000/ 1236] Overall Loss 0.112656 Objective Loss 0.112656 LR 0.000125 Time 0.021691 +2023-10-02 22:00:32,602 - Epoch: [189][ 1010/ 1236] Overall Loss 0.112579 Objective Loss 0.112579 LR 0.000125 Time 0.021694 +2023-10-02 22:00:32,827 - Epoch: [189][ 1020/ 1236] Overall Loss 0.112639 Objective Loss 0.112639 LR 0.000125 Time 0.021701 +2023-10-02 22:00:33,048 - Epoch: [189][ 1030/ 1236] Overall Loss 0.112650 Objective Loss 0.112650 LR 0.000125 Time 0.021704 +2023-10-02 22:00:33,268 - Epoch: [189][ 1040/ 1236] Overall Loss 0.112711 Objective Loss 0.112711 LR 0.000125 Time 0.021707 +2023-10-02 22:00:33,482 - Epoch: [189][ 1050/ 1236] Overall Loss 0.112674 Objective Loss 0.112674 LR 0.000125 Time 0.021704 +2023-10-02 22:00:33,696 - Epoch: [189][ 1060/ 1236] Overall Loss 0.112608 Objective Loss 0.112608 LR 0.000125 Time 0.021700 +2023-10-02 22:00:33,910 - Epoch: [189][ 1070/ 1236] Overall Loss 0.112651 Objective Loss 0.112651 LR 0.000125 Time 0.021697 +2023-10-02 22:00:34,124 - Epoch: [189][ 1080/ 1236] Overall Loss 0.112629 Objective Loss 0.112629 LR 0.000125 Time 0.021693 +2023-10-02 22:00:34,338 - Epoch: [189][ 1090/ 1236] Overall Loss 0.112693 Objective Loss 0.112693 LR 0.000125 Time 0.021690 +2023-10-02 22:00:34,551 - Epoch: [189][ 1100/ 1236] Overall Loss 0.112759 Objective Loss 0.112759 LR 0.000125 Time 0.021687 +2023-10-02 22:00:34,765 - Epoch: [189][ 1110/ 1236] Overall Loss 0.112758 Objective Loss 0.112758 LR 0.000125 Time 0.021684 +2023-10-02 22:00:34,979 - Epoch: [189][ 1120/ 1236] Overall Loss 0.112795 Objective Loss 0.112795 LR 0.000125 Time 0.021681 +2023-10-02 22:00:35,193 - Epoch: [189][ 1130/ 1236] Overall Loss 0.112755 Objective Loss 0.112755 LR 0.000125 Time 0.021678 +2023-10-02 22:00:35,406 - Epoch: [189][ 1140/ 1236] Overall Loss 0.112694 Objective Loss 0.112694 LR 0.000125 Time 0.021675 +2023-10-02 22:00:35,621 - Epoch: [189][ 1150/ 1236] Overall Loss 0.112734 Objective Loss 0.112734 LR 0.000125 Time 0.021672 +2023-10-02 22:00:35,834 - Epoch: [189][ 1160/ 1236] Overall Loss 0.112723 Objective Loss 0.112723 LR 0.000125 Time 0.021669 +2023-10-02 22:00:36,049 - Epoch: [189][ 1170/ 1236] Overall Loss 0.112712 Objective Loss 0.112712 LR 0.000125 Time 0.021667 +2023-10-02 22:00:36,262 - Epoch: [189][ 1180/ 1236] Overall Loss 0.112643 Objective Loss 0.112643 LR 0.000125 Time 0.021663 +2023-10-02 22:00:36,476 - Epoch: [189][ 1190/ 1236] Overall Loss 0.112624 Objective Loss 0.112624 LR 0.000125 Time 0.021661 +2023-10-02 22:00:36,689 - Epoch: [189][ 1200/ 1236] Overall Loss 0.112675 Objective Loss 0.112675 LR 0.000125 Time 0.021658 +2023-10-02 22:00:36,903 - Epoch: [189][ 1210/ 1236] Overall Loss 0.112777 Objective Loss 0.112777 LR 0.000125 Time 0.021655 +2023-10-02 22:00:37,117 - Epoch: [189][ 1220/ 1236] Overall Loss 0.112855 Objective Loss 0.112855 LR 0.000125 Time 0.021652 +2023-10-02 22:00:37,384 - Epoch: [189][ 1230/ 1236] Overall Loss 0.112834 Objective Loss 0.112834 LR 0.000125 Time 0.021693 +2023-10-02 22:00:37,507 - Epoch: [189][ 1236/ 1236] Overall Loss 0.112745 Objective Loss 0.112745 Top1 93.686354 Top5 99.592668 LR 0.000125 Time 0.021687 +2023-10-02 22:00:37,658 - --- validate (epoch=189)----------- +2023-10-02 22:00:37,658 - 29943 samples (256 per mini-batch) +2023-10-02 22:00:38,156 - Epoch: [189][ 10/ 117] Loss 0.278833 Top1 88.671875 Top5 98.906250 +2023-10-02 22:00:38,304 - Epoch: [189][ 20/ 117] Loss 0.296475 Top1 88.339844 Top5 98.652344 +2023-10-02 22:00:38,451 - Epoch: [189][ 30/ 117] Loss 0.296964 Top1 87.721354 Top5 98.580729 +2023-10-02 22:00:38,597 - Epoch: [189][ 40/ 117] Loss 0.295200 Top1 87.900391 Top5 98.603516 +2023-10-02 22:00:38,742 - Epoch: [189][ 50/ 117] Loss 0.298118 Top1 87.828125 Top5 98.648438 +2023-10-02 22:00:38,888 - Epoch: [189][ 60/ 117] Loss 0.302559 Top1 87.890625 Top5 98.632812 +2023-10-02 22:00:39,035 - Epoch: [189][ 70/ 117] Loss 0.299679 Top1 88.002232 Top5 98.683036 +2023-10-02 22:00:39,183 - Epoch: [189][ 80/ 117] Loss 0.306486 Top1 87.841797 Top5 98.647461 +2023-10-02 22:00:39,332 - Epoch: [189][ 90/ 117] Loss 0.306051 Top1 87.842882 Top5 98.689236 +2023-10-02 22:00:39,477 - Epoch: [189][ 100/ 117] Loss 0.300732 Top1 87.925781 Top5 98.734375 +2023-10-02 22:00:39,633 - Epoch: [189][ 110/ 117] Loss 0.303074 Top1 87.897727 Top5 98.728693 +2023-10-02 22:00:39,722 - Epoch: [189][ 117/ 117] Loss 0.306913 Top1 87.873627 Top5 98.707544 +2023-10-02 22:00:39,865 - ==> Top1: 87.874 Top5: 98.708 Loss: 0.307 + +2023-10-02 22:00:39,866 - ==> Confusion: +[[ 944 0 3 1 3 1 0 0 6 56 1 1 1 1 6 1 2 0 1 0 22] + [ 0 1071 0 1 5 12 0 17 0 0 1 0 0 0 3 3 0 0 4 3 11] + [ 1 1 985 6 0 0 12 7 0 1 2 1 7 2 1 4 3 1 10 2 10] + [ 1 4 15 986 0 0 0 1 3 1 4 0 4 2 29 2 1 5 9 1 21] + [ 23 4 0 0 977 4 0 0 0 13 0 0 0 4 7 5 7 0 0 1 5] + [ 3 37 0 1 7 997 0 17 1 5 1 5 2 8 5 0 4 0 4 1 18] + [ 0 4 26 0 0 3 1131 5 0 0 3 1 0 0 0 5 0 1 1 6 5] + [ 1 10 10 1 6 21 7 1080 1 1 5 5 4 4 1 1 0 1 42 8 9] + [ 15 3 0 1 1 5 0 3 978 34 11 1 1 13 14 0 2 1 3 1 2] + [ 87 2 1 1 6 4 0 0 29 954 0 0 0 22 6 0 1 0 0 1 5] + [ 3 3 11 9 0 2 3 3 7 1 970 1 0 8 6 0 3 2 5 1 15] + [ 0 0 0 0 0 16 0 5 0 0 0 968 15 6 0 1 1 15 0 2 6] + [ 0 1 1 2 1 1 1 2 0 1 2 28 975 1 3 9 1 13 2 4 20] + [ 0 0 1 0 4 6 0 0 13 10 1 8 0 1050 4 0 0 1 0 1 20] + [ 12 0 4 15 1 0 0 0 23 1 2 0 3 3 1021 0 1 2 5 0 8] + [ 0 0 2 1 5 1 0 0 0 0 0 3 8 0 0 1070 15 11 2 8 8] + [ 0 19 1 0 8 5 1 0 1 1 0 4 0 2 2 8 1092 0 1 4 12] + [ 0 0 0 2 1 0 3 0 0 1 0 5 18 0 3 5 0 994 0 1 5] + [ 4 3 6 17 0 0 0 17 5 1 0 1 2 0 10 0 0 0 989 0 13] + [ 0 0 6 2 2 3 7 5 0 1 0 12 3 0 2 1 6 2 0 1089 11] + [ 105 98 98 72 64 107 26 78 68 59 121 77 252 216 116 47 64 47 85 114 5991]] + +2023-10-02 22:00:39,867 - ==> Best [Top1: 87.874 Top5: 98.708 Sparsity:0.00 Params: 169472 on epoch: 189] +2023-10-02 22:00:39,867 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:00:39,881 - + +2023-10-02 22:00:39,881 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:00:40,926 - Epoch: [190][ 10/ 1236] Overall Loss 0.098963 Objective Loss 0.098963 LR 0.000063 Time 0.104492 +2023-10-02 22:00:41,136 - Epoch: [190][ 20/ 1236] Overall Loss 0.101728 Objective Loss 0.101728 LR 0.000063 Time 0.062712 +2023-10-02 22:00:41,346 - Epoch: [190][ 30/ 1236] Overall Loss 0.103116 Objective Loss 0.103116 LR 0.000063 Time 0.048741 +2023-10-02 22:00:41,556 - Epoch: [190][ 40/ 1236] Overall Loss 0.104564 Objective Loss 0.104564 LR 0.000063 Time 0.041800 +2023-10-02 22:00:41,765 - Epoch: [190][ 50/ 1236] Overall Loss 0.104550 Objective Loss 0.104550 LR 0.000063 Time 0.037599 +2023-10-02 22:00:41,976 - Epoch: [190][ 60/ 1236] Overall Loss 0.104794 Objective Loss 0.104794 LR 0.000063 Time 0.034836 +2023-10-02 22:00:42,185 - Epoch: [190][ 70/ 1236] Overall Loss 0.105929 Objective Loss 0.105929 LR 0.000063 Time 0.032828 +2023-10-02 22:00:42,396 - Epoch: [190][ 80/ 1236] Overall Loss 0.106594 Objective Loss 0.106594 LR 0.000063 Time 0.031352 +2023-10-02 22:00:42,605 - Epoch: [190][ 90/ 1236] Overall Loss 0.105661 Objective Loss 0.105661 LR 0.000063 Time 0.030180 +2023-10-02 22:00:42,816 - Epoch: [190][ 100/ 1236] Overall Loss 0.106865 Objective Loss 0.106865 LR 0.000063 Time 0.029266 +2023-10-02 22:00:43,025 - Epoch: [190][ 110/ 1236] Overall Loss 0.106993 Objective Loss 0.106993 LR 0.000063 Time 0.028495 +2023-10-02 22:00:43,235 - Epoch: [190][ 120/ 1236] Overall Loss 0.108805 Objective Loss 0.108805 LR 0.000063 Time 0.027871 +2023-10-02 22:00:43,446 - Epoch: [190][ 130/ 1236] Overall Loss 0.110037 Objective Loss 0.110037 LR 0.000063 Time 0.027338 +2023-10-02 22:00:43,659 - Epoch: [190][ 140/ 1236] Overall Loss 0.108097 Objective Loss 0.108097 LR 0.000063 Time 0.026892 +2023-10-02 22:00:43,872 - Epoch: [190][ 150/ 1236] Overall Loss 0.107883 Objective Loss 0.107883 LR 0.000063 Time 0.026512 +2023-10-02 22:00:44,084 - Epoch: [190][ 160/ 1236] Overall Loss 0.107469 Objective Loss 0.107469 LR 0.000063 Time 0.026173 +2023-10-02 22:00:44,297 - Epoch: [190][ 170/ 1236] Overall Loss 0.106426 Objective Loss 0.106426 LR 0.000063 Time 0.025884 +2023-10-02 22:00:44,511 - Epoch: [190][ 180/ 1236] Overall Loss 0.105670 Objective Loss 0.105670 LR 0.000063 Time 0.025627 +2023-10-02 22:00:44,724 - Epoch: [190][ 190/ 1236] Overall Loss 0.105543 Objective Loss 0.105543 LR 0.000063 Time 0.025397 +2023-10-02 22:00:44,937 - Epoch: [190][ 200/ 1236] Overall Loss 0.105313 Objective Loss 0.105313 LR 0.000063 Time 0.025186 +2023-10-02 22:00:45,150 - Epoch: [190][ 210/ 1236] Overall Loss 0.105089 Objective Loss 0.105089 LR 0.000063 Time 0.024999 +2023-10-02 22:00:45,363 - Epoch: [190][ 220/ 1236] Overall Loss 0.105079 Objective Loss 0.105079 LR 0.000063 Time 0.024822 +2023-10-02 22:00:45,577 - Epoch: [190][ 230/ 1236] Overall Loss 0.105301 Objective Loss 0.105301 LR 0.000063 Time 0.024672 +2023-10-02 22:00:45,791 - Epoch: [190][ 240/ 1236] Overall Loss 0.105962 Objective Loss 0.105962 LR 0.000063 Time 0.024533 +2023-10-02 22:00:46,004 - Epoch: [190][ 250/ 1236] Overall Loss 0.106116 Objective Loss 0.106116 LR 0.000063 Time 0.024402 +2023-10-02 22:00:46,217 - Epoch: [190][ 260/ 1236] Overall Loss 0.106627 Objective Loss 0.106627 LR 0.000063 Time 0.024279 +2023-10-02 22:00:46,430 - Epoch: [190][ 270/ 1236] Overall Loss 0.106417 Objective Loss 0.106417 LR 0.000063 Time 0.024162 +2023-10-02 22:00:46,642 - Epoch: [190][ 280/ 1236] Overall Loss 0.106136 Objective Loss 0.106136 LR 0.000063 Time 0.024052 +2023-10-02 22:00:46,855 - Epoch: [190][ 290/ 1236] Overall Loss 0.106662 Objective Loss 0.106662 LR 0.000063 Time 0.023950 +2023-10-02 22:00:47,067 - Epoch: [190][ 300/ 1236] Overall Loss 0.107101 Objective Loss 0.107101 LR 0.000063 Time 0.023854 +2023-10-02 22:00:47,280 - Epoch: [190][ 310/ 1236] Overall Loss 0.107202 Objective Loss 0.107202 LR 0.000063 Time 0.023765 +2023-10-02 22:00:47,493 - Epoch: [190][ 320/ 1236] Overall Loss 0.107043 Objective Loss 0.107043 LR 0.000063 Time 0.023682 +2023-10-02 22:00:47,705 - Epoch: [190][ 330/ 1236] Overall Loss 0.107315 Objective Loss 0.107315 LR 0.000063 Time 0.023605 +2023-10-02 22:00:47,926 - Epoch: [190][ 340/ 1236] Overall Loss 0.107130 Objective Loss 0.107130 LR 0.000063 Time 0.023554 +2023-10-02 22:00:48,142 - Epoch: [190][ 350/ 1236] Overall Loss 0.106565 Objective Loss 0.106565 LR 0.000063 Time 0.023498 +2023-10-02 22:00:48,363 - Epoch: [190][ 360/ 1236] Overall Loss 0.106508 Objective Loss 0.106508 LR 0.000063 Time 0.023458 +2023-10-02 22:00:48,579 - Epoch: [190][ 370/ 1236] Overall Loss 0.106420 Objective Loss 0.106420 LR 0.000063 Time 0.023407 +2023-10-02 22:00:48,800 - Epoch: [190][ 380/ 1236] Overall Loss 0.106912 Objective Loss 0.106912 LR 0.000063 Time 0.023372 +2023-10-02 22:00:49,016 - Epoch: [190][ 390/ 1236] Overall Loss 0.106999 Objective Loss 0.106999 LR 0.000063 Time 0.023325 +2023-10-02 22:00:49,237 - Epoch: [190][ 400/ 1236] Overall Loss 0.106970 Objective Loss 0.106970 LR 0.000063 Time 0.023293 +2023-10-02 22:00:49,452 - Epoch: [190][ 410/ 1236] Overall Loss 0.107243 Objective Loss 0.107243 LR 0.000063 Time 0.023250 +2023-10-02 22:00:49,673 - Epoch: [190][ 420/ 1236] Overall Loss 0.107285 Objective Loss 0.107285 LR 0.000063 Time 0.023222 +2023-10-02 22:00:49,890 - Epoch: [190][ 430/ 1236] Overall Loss 0.107460 Objective Loss 0.107460 LR 0.000063 Time 0.023184 +2023-10-02 22:00:50,110 - Epoch: [190][ 440/ 1236] Overall Loss 0.107291 Objective Loss 0.107291 LR 0.000063 Time 0.023158 +2023-10-02 22:00:50,326 - Epoch: [190][ 450/ 1236] Overall Loss 0.106944 Objective Loss 0.106944 LR 0.000063 Time 0.023123 +2023-10-02 22:00:50,547 - Epoch: [190][ 460/ 1236] Overall Loss 0.106960 Objective Loss 0.106960 LR 0.000063 Time 0.023099 +2023-10-02 22:00:50,763 - Epoch: [190][ 470/ 1236] Overall Loss 0.106977 Objective Loss 0.106977 LR 0.000063 Time 0.023066 +2023-10-02 22:00:50,984 - Epoch: [190][ 480/ 1236] Overall Loss 0.107460 Objective Loss 0.107460 LR 0.000063 Time 0.023046 +2023-10-02 22:00:51,200 - Epoch: [190][ 490/ 1236] Overall Loss 0.107821 Objective Loss 0.107821 LR 0.000063 Time 0.023016 +2023-10-02 22:00:51,421 - Epoch: [190][ 500/ 1236] Overall Loss 0.108349 Objective Loss 0.108349 LR 0.000063 Time 0.022997 +2023-10-02 22:00:51,638 - Epoch: [190][ 510/ 1236] Overall Loss 0.108737 Objective Loss 0.108737 LR 0.000063 Time 0.022969 +2023-10-02 22:00:51,859 - Epoch: [190][ 520/ 1236] Overall Loss 0.108634 Objective Loss 0.108634 LR 0.000063 Time 0.022951 +2023-10-02 22:00:52,075 - Epoch: [190][ 530/ 1236] Overall Loss 0.108688 Objective Loss 0.108688 LR 0.000063 Time 0.022926 +2023-10-02 22:00:52,296 - Epoch: [190][ 540/ 1236] Overall Loss 0.108492 Objective Loss 0.108492 LR 0.000063 Time 0.022910 +2023-10-02 22:00:52,511 - Epoch: [190][ 550/ 1236] Overall Loss 0.108306 Objective Loss 0.108306 LR 0.000063 Time 0.022885 +2023-10-02 22:00:52,733 - Epoch: [190][ 560/ 1236] Overall Loss 0.108742 Objective Loss 0.108742 LR 0.000063 Time 0.022870 +2023-10-02 22:00:52,949 - Epoch: [190][ 570/ 1236] Overall Loss 0.108725 Objective Loss 0.108725 LR 0.000063 Time 0.022847 +2023-10-02 22:00:53,170 - Epoch: [190][ 580/ 1236] Overall Loss 0.108713 Objective Loss 0.108713 LR 0.000063 Time 0.022834 +2023-10-02 22:00:53,386 - Epoch: [190][ 590/ 1236] Overall Loss 0.108368 Objective Loss 0.108368 LR 0.000063 Time 0.022813 +2023-10-02 22:00:53,607 - Epoch: [190][ 600/ 1236] Overall Loss 0.108490 Objective Loss 0.108490 LR 0.000063 Time 0.022800 +2023-10-02 22:00:53,823 - Epoch: [190][ 610/ 1236] Overall Loss 0.108578 Objective Loss 0.108578 LR 0.000063 Time 0.022780 +2023-10-02 22:00:54,043 - Epoch: [190][ 620/ 1236] Overall Loss 0.108671 Objective Loss 0.108671 LR 0.000063 Time 0.022768 +2023-10-02 22:00:54,259 - Epoch: [190][ 630/ 1236] Overall Loss 0.108942 Objective Loss 0.108942 LR 0.000063 Time 0.022749 +2023-10-02 22:00:54,481 - Epoch: [190][ 640/ 1236] Overall Loss 0.108796 Objective Loss 0.108796 LR 0.000063 Time 0.022738 +2023-10-02 22:00:54,697 - Epoch: [190][ 650/ 1236] Overall Loss 0.108621 Objective Loss 0.108621 LR 0.000063 Time 0.022720 +2023-10-02 22:00:54,918 - Epoch: [190][ 660/ 1236] Overall Loss 0.108702 Objective Loss 0.108702 LR 0.000063 Time 0.022711 +2023-10-02 22:00:55,134 - Epoch: [190][ 670/ 1236] Overall Loss 0.108857 Objective Loss 0.108857 LR 0.000063 Time 0.022694 +2023-10-02 22:00:55,355 - Epoch: [190][ 680/ 1236] Overall Loss 0.109208 Objective Loss 0.109208 LR 0.000063 Time 0.022684 +2023-10-02 22:00:55,571 - Epoch: [190][ 690/ 1236] Overall Loss 0.109496 Objective Loss 0.109496 LR 0.000063 Time 0.022668 +2023-10-02 22:00:55,792 - Epoch: [190][ 700/ 1236] Overall Loss 0.109514 Objective Loss 0.109514 LR 0.000063 Time 0.022659 +2023-10-02 22:00:56,008 - Epoch: [190][ 710/ 1236] Overall Loss 0.109464 Objective Loss 0.109464 LR 0.000063 Time 0.022644 +2023-10-02 22:00:56,229 - Epoch: [190][ 720/ 1236] Overall Loss 0.109138 Objective Loss 0.109138 LR 0.000063 Time 0.022636 +2023-10-02 22:00:56,445 - Epoch: [190][ 730/ 1236] Overall Loss 0.108990 Objective Loss 0.108990 LR 0.000063 Time 0.022622 +2023-10-02 22:00:56,666 - Epoch: [190][ 740/ 1236] Overall Loss 0.109315 Objective Loss 0.109315 LR 0.000063 Time 0.022614 +2023-10-02 22:00:56,882 - Epoch: [190][ 750/ 1236] Overall Loss 0.109284 Objective Loss 0.109284 LR 0.000063 Time 0.022600 +2023-10-02 22:00:57,103 - Epoch: [190][ 760/ 1236] Overall Loss 0.109536 Objective Loss 0.109536 LR 0.000063 Time 0.022593 +2023-10-02 22:00:57,319 - Epoch: [190][ 770/ 1236] Overall Loss 0.109593 Objective Loss 0.109593 LR 0.000063 Time 0.022580 +2023-10-02 22:00:57,540 - Epoch: [190][ 780/ 1236] Overall Loss 0.109405 Objective Loss 0.109405 LR 0.000063 Time 0.022574 +2023-10-02 22:00:57,756 - Epoch: [190][ 790/ 1236] Overall Loss 0.109348 Objective Loss 0.109348 LR 0.000063 Time 0.022561 +2023-10-02 22:00:57,978 - Epoch: [190][ 800/ 1236] Overall Loss 0.109268 Objective Loss 0.109268 LR 0.000063 Time 0.022555 +2023-10-02 22:00:58,193 - Epoch: [190][ 810/ 1236] Overall Loss 0.109233 Objective Loss 0.109233 LR 0.000063 Time 0.022543 +2023-10-02 22:00:58,414 - Epoch: [190][ 820/ 1236] Overall Loss 0.109245 Objective Loss 0.109245 LR 0.000063 Time 0.022537 +2023-10-02 22:00:58,631 - Epoch: [190][ 830/ 1236] Overall Loss 0.109382 Objective Loss 0.109382 LR 0.000063 Time 0.022525 +2023-10-02 22:00:58,848 - Epoch: [190][ 840/ 1236] Overall Loss 0.109204 Objective Loss 0.109204 LR 0.000063 Time 0.022516 +2023-10-02 22:00:59,057 - Epoch: [190][ 850/ 1236] Overall Loss 0.109396 Objective Loss 0.109396 LR 0.000063 Time 0.022495 +2023-10-02 22:00:59,268 - Epoch: [190][ 860/ 1236] Overall Loss 0.109458 Objective Loss 0.109458 LR 0.000063 Time 0.022478 +2023-10-02 22:00:59,477 - Epoch: [190][ 870/ 1236] Overall Loss 0.109717 Objective Loss 0.109717 LR 0.000063 Time 0.022458 +2023-10-02 22:00:59,688 - Epoch: [190][ 880/ 1236] Overall Loss 0.109779 Objective Loss 0.109779 LR 0.000063 Time 0.022442 +2023-10-02 22:00:59,897 - Epoch: [190][ 890/ 1236] Overall Loss 0.109773 Objective Loss 0.109773 LR 0.000063 Time 0.022424 +2023-10-02 22:01:00,107 - Epoch: [190][ 900/ 1236] Overall Loss 0.109786 Objective Loss 0.109786 LR 0.000063 Time 0.022408 +2023-10-02 22:01:00,317 - Epoch: [190][ 910/ 1236] Overall Loss 0.109719 Objective Loss 0.109719 LR 0.000063 Time 0.022390 +2023-10-02 22:01:00,527 - Epoch: [190][ 920/ 1236] Overall Loss 0.109431 Objective Loss 0.109431 LR 0.000063 Time 0.022375 +2023-10-02 22:01:00,737 - Epoch: [190][ 930/ 1236] Overall Loss 0.109546 Objective Loss 0.109546 LR 0.000063 Time 0.022358 +2023-10-02 22:01:00,947 - Epoch: [190][ 940/ 1236] Overall Loss 0.109475 Objective Loss 0.109475 LR 0.000063 Time 0.022343 +2023-10-02 22:01:01,157 - Epoch: [190][ 950/ 1236] Overall Loss 0.109572 Objective Loss 0.109572 LR 0.000063 Time 0.022327 +2023-10-02 22:01:01,367 - Epoch: [190][ 960/ 1236] Overall Loss 0.109567 Objective Loss 0.109567 LR 0.000063 Time 0.022313 +2023-10-02 22:01:01,576 - Epoch: [190][ 970/ 1236] Overall Loss 0.109686 Objective Loss 0.109686 LR 0.000063 Time 0.022299 +2023-10-02 22:01:01,787 - Epoch: [190][ 980/ 1236] Overall Loss 0.109617 Objective Loss 0.109617 LR 0.000063 Time 0.022286 +2023-10-02 22:01:01,996 - Epoch: [190][ 990/ 1236] Overall Loss 0.109668 Objective Loss 0.109668 LR 0.000063 Time 0.022271 +2023-10-02 22:01:02,206 - Epoch: [190][ 1000/ 1236] Overall Loss 0.109563 Objective Loss 0.109563 LR 0.000063 Time 0.022258 +2023-10-02 22:01:02,416 - Epoch: [190][ 1010/ 1236] Overall Loss 0.109561 Objective Loss 0.109561 LR 0.000063 Time 0.022244 +2023-10-02 22:01:02,626 - Epoch: [190][ 1020/ 1236] Overall Loss 0.109408 Objective Loss 0.109408 LR 0.000063 Time 0.022232 +2023-10-02 22:01:02,835 - Epoch: [190][ 1030/ 1236] Overall Loss 0.109527 Objective Loss 0.109527 LR 0.000063 Time 0.022219 +2023-10-02 22:01:03,046 - Epoch: [190][ 1040/ 1236] Overall Loss 0.109323 Objective Loss 0.109323 LR 0.000063 Time 0.022207 +2023-10-02 22:01:03,255 - Epoch: [190][ 1050/ 1236] Overall Loss 0.109388 Objective Loss 0.109388 LR 0.000063 Time 0.022194 +2023-10-02 22:01:03,465 - Epoch: [190][ 1060/ 1236] Overall Loss 0.109468 Objective Loss 0.109468 LR 0.000063 Time 0.022182 +2023-10-02 22:01:03,674 - Epoch: [190][ 1070/ 1236] Overall Loss 0.109587 Objective Loss 0.109587 LR 0.000063 Time 0.022170 +2023-10-02 22:01:03,885 - Epoch: [190][ 1080/ 1236] Overall Loss 0.109725 Objective Loss 0.109725 LR 0.000063 Time 0.022159 +2023-10-02 22:01:04,094 - Epoch: [190][ 1090/ 1236] Overall Loss 0.109738 Objective Loss 0.109738 LR 0.000063 Time 0.022147 +2023-10-02 22:01:04,304 - Epoch: [190][ 1100/ 1236] Overall Loss 0.109706 Objective Loss 0.109706 LR 0.000063 Time 0.022136 +2023-10-02 22:01:04,513 - Epoch: [190][ 1110/ 1236] Overall Loss 0.109623 Objective Loss 0.109623 LR 0.000063 Time 0.022125 +2023-10-02 22:01:04,723 - Epoch: [190][ 1120/ 1236] Overall Loss 0.109550 Objective Loss 0.109550 LR 0.000063 Time 0.022114 +2023-10-02 22:01:04,933 - Epoch: [190][ 1130/ 1236] Overall Loss 0.109540 Objective Loss 0.109540 LR 0.000063 Time 0.022104 +2023-10-02 22:01:05,143 - Epoch: [190][ 1140/ 1236] Overall Loss 0.109433 Objective Loss 0.109433 LR 0.000063 Time 0.022094 +2023-10-02 22:01:05,353 - Epoch: [190][ 1150/ 1236] Overall Loss 0.109447 Objective Loss 0.109447 LR 0.000063 Time 0.022084 +2023-10-02 22:01:05,564 - Epoch: [190][ 1160/ 1236] Overall Loss 0.109547 Objective Loss 0.109547 LR 0.000063 Time 0.022074 +2023-10-02 22:01:05,774 - Epoch: [190][ 1170/ 1236] Overall Loss 0.109512 Objective Loss 0.109512 LR 0.000063 Time 0.022063 +2023-10-02 22:01:05,984 - Epoch: [190][ 1180/ 1236] Overall Loss 0.109454 Objective Loss 0.109454 LR 0.000063 Time 0.022055 +2023-10-02 22:01:06,194 - Epoch: [190][ 1190/ 1236] Overall Loss 0.109617 Objective Loss 0.109617 LR 0.000063 Time 0.022045 +2023-10-02 22:01:06,405 - Epoch: [190][ 1200/ 1236] Overall Loss 0.109687 Objective Loss 0.109687 LR 0.000063 Time 0.022036 +2023-10-02 22:01:06,615 - Epoch: [190][ 1210/ 1236] Overall Loss 0.109688 Objective Loss 0.109688 LR 0.000063 Time 0.022026 +2023-10-02 22:01:06,825 - Epoch: [190][ 1220/ 1236] Overall Loss 0.109701 Objective Loss 0.109701 LR 0.000063 Time 0.022018 +2023-10-02 22:01:07,091 - Epoch: [190][ 1230/ 1236] Overall Loss 0.109595 Objective Loss 0.109595 LR 0.000063 Time 0.022054 +2023-10-02 22:01:07,214 - Epoch: [190][ 1236/ 1236] Overall Loss 0.109595 Objective Loss 0.109595 Top1 90.427699 Top5 98.370672 LR 0.000063 Time 0.022046 +2023-10-02 22:01:07,362 - --- validate (epoch=190)----------- +2023-10-02 22:01:07,363 - 29943 samples (256 per mini-batch) +2023-10-02 22:01:07,865 - Epoch: [190][ 10/ 117] Loss 0.294466 Top1 88.203125 Top5 98.398438 +2023-10-02 22:01:08,023 - Epoch: [190][ 20/ 117] Loss 0.299430 Top1 88.007812 Top5 98.554688 +2023-10-02 22:01:08,180 - Epoch: [190][ 30/ 117] Loss 0.308282 Top1 87.513021 Top5 98.580729 +2023-10-02 22:01:08,335 - Epoch: [190][ 40/ 117] Loss 0.303270 Top1 87.548828 Top5 98.671875 +2023-10-02 22:01:08,491 - Epoch: [190][ 50/ 117] Loss 0.305514 Top1 87.578125 Top5 98.773438 +2023-10-02 22:01:08,646 - Epoch: [190][ 60/ 117] Loss 0.306539 Top1 87.675781 Top5 98.763021 +2023-10-02 22:01:08,802 - Epoch: [190][ 70/ 117] Loss 0.311761 Top1 87.628348 Top5 98.761161 +2023-10-02 22:01:08,956 - Epoch: [190][ 80/ 117] Loss 0.314049 Top1 87.529297 Top5 98.750000 +2023-10-02 22:01:09,109 - Epoch: [190][ 90/ 117] Loss 0.310961 Top1 87.634549 Top5 98.741319 +2023-10-02 22:01:09,260 - Epoch: [190][ 100/ 117] Loss 0.306313 Top1 87.695312 Top5 98.753906 +2023-10-02 22:01:09,419 - Epoch: [190][ 110/ 117] Loss 0.306906 Top1 87.674006 Top5 98.725142 +2023-10-02 22:01:09,509 - Epoch: [190][ 117/ 117] Loss 0.306793 Top1 87.713322 Top5 98.734262 +2023-10-02 22:01:09,626 - ==> Top1: 87.713 Top5: 98.734 Loss: 0.307 + +2023-10-02 22:01:09,626 - ==> Confusion: +[[ 947 0 3 1 2 2 0 0 3 56 2 0 1 2 6 1 2 0 1 0 21] + [ 0 1078 1 1 5 12 0 17 0 2 0 0 0 0 2 3 0 0 3 1 6] + [ 1 0 984 8 0 0 11 9 0 2 1 0 7 2 1 2 2 2 11 3 10] + [ 1 4 11 987 0 1 0 1 2 0 2 0 3 1 31 2 1 4 14 2 22] + [ 25 4 0 0 971 4 0 0 0 11 1 0 0 5 9 4 9 0 0 2 5] + [ 4 40 0 1 4 980 1 22 1 6 1 8 2 11 6 0 4 1 5 2 17] + [ 0 3 27 1 0 2 1131 8 0 0 3 1 0 0 0 3 0 1 1 5 5] + [ 1 9 10 1 6 19 5 1086 2 2 4 4 2 3 1 1 0 1 42 7 12] + [ 15 5 0 1 1 2 0 3 989 28 10 1 1 11 11 0 1 1 3 3 3] + [ 90 1 2 1 6 4 0 0 28 952 0 0 0 14 5 2 0 1 0 2 11] + [ 3 2 8 11 0 2 3 2 14 1 961 1 2 14 3 0 2 3 4 1 16] + [ 0 0 2 0 0 13 0 5 0 0 0 970 17 5 0 1 0 15 0 2 5] + [ 0 1 1 2 0 1 1 2 0 1 2 25 982 1 3 7 1 13 2 4 19] + [ 0 0 1 0 4 9 0 0 14 13 2 8 0 1041 4 0 0 1 0 1 21] + [ 13 0 3 14 2 1 0 0 21 1 1 0 2 3 1017 0 1 2 12 0 8] + [ 0 0 2 1 5 1 0 0 0 0 0 5 7 0 0 1075 12 10 1 9 6] + [ 0 18 0 0 5 4 0 0 1 1 0 5 0 2 2 8 1095 0 2 5 13] + [ 0 0 1 2 0 0 2 0 0 0 0 4 14 1 2 6 0 999 0 2 5] + [ 4 4 2 14 0 0 0 24 3 1 0 1 1 0 7 0 0 0 995 1 11] + [ 0 1 4 3 1 2 6 7 0 0 1 11 5 1 2 1 6 2 0 1090 9] + [ 107 123 99 75 51 107 27 93 73 52 116 77 267 202 120 40 72 48 103 119 5934]] + +2023-10-02 22:01:09,628 - ==> Best [Top1: 87.874 Top5: 98.708 Sparsity:0.00 Params: 169472 on epoch: 189] +2023-10-02 22:01:09,628 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:01:09,634 - + +2023-10-02 22:01:09,634 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:01:10,771 - Epoch: [191][ 10/ 1236] Overall Loss 0.114130 Objective Loss 0.114130 LR 0.000063 Time 0.113664 +2023-10-02 22:01:10,981 - Epoch: [191][ 20/ 1236] Overall Loss 0.113616 Objective Loss 0.113616 LR 0.000063 Time 0.067282 +2023-10-02 22:01:11,190 - Epoch: [191][ 30/ 1236] Overall Loss 0.109206 Objective Loss 0.109206 LR 0.000063 Time 0.051810 +2023-10-02 22:01:11,400 - Epoch: [191][ 40/ 1236] Overall Loss 0.112908 Objective Loss 0.112908 LR 0.000063 Time 0.044099 +2023-10-02 22:01:11,609 - Epoch: [191][ 50/ 1236] Overall Loss 0.110450 Objective Loss 0.110450 LR 0.000063 Time 0.039452 +2023-10-02 22:01:11,819 - Epoch: [191][ 60/ 1236] Overall Loss 0.109644 Objective Loss 0.109644 LR 0.000063 Time 0.036371 +2023-10-02 22:01:12,029 - Epoch: [191][ 70/ 1236] Overall Loss 0.110181 Objective Loss 0.110181 LR 0.000063 Time 0.034164 +2023-10-02 22:01:12,239 - Epoch: [191][ 80/ 1236] Overall Loss 0.108263 Objective Loss 0.108263 LR 0.000063 Time 0.032515 +2023-10-02 22:01:12,448 - Epoch: [191][ 90/ 1236] Overall Loss 0.108016 Objective Loss 0.108016 LR 0.000063 Time 0.031219 +2023-10-02 22:01:12,658 - Epoch: [191][ 100/ 1236] Overall Loss 0.108573 Objective Loss 0.108573 LR 0.000063 Time 0.030194 +2023-10-02 22:01:12,868 - Epoch: [191][ 110/ 1236] Overall Loss 0.107491 Objective Loss 0.107491 LR 0.000063 Time 0.029355 +2023-10-02 22:01:13,079 - Epoch: [191][ 120/ 1236] Overall Loss 0.106133 Objective Loss 0.106133 LR 0.000063 Time 0.028663 +2023-10-02 22:01:13,289 - Epoch: [191][ 130/ 1236] Overall Loss 0.106003 Objective Loss 0.106003 LR 0.000063 Time 0.028067 +2023-10-02 22:01:13,500 - Epoch: [191][ 140/ 1236] Overall Loss 0.106403 Objective Loss 0.106403 LR 0.000063 Time 0.027566 +2023-10-02 22:01:13,710 - Epoch: [191][ 150/ 1236] Overall Loss 0.106669 Objective Loss 0.106669 LR 0.000063 Time 0.027126 +2023-10-02 22:01:13,920 - Epoch: [191][ 160/ 1236] Overall Loss 0.106792 Objective Loss 0.106792 LR 0.000063 Time 0.026741 +2023-10-02 22:01:14,129 - Epoch: [191][ 170/ 1236] Overall Loss 0.106992 Objective Loss 0.106992 LR 0.000063 Time 0.026396 +2023-10-02 22:01:14,339 - Epoch: [191][ 180/ 1236] Overall Loss 0.106144 Objective Loss 0.106144 LR 0.000063 Time 0.026096 +2023-10-02 22:01:14,549 - Epoch: [191][ 190/ 1236] Overall Loss 0.106960 Objective Loss 0.106960 LR 0.000063 Time 0.025823 +2023-10-02 22:01:14,758 - Epoch: [191][ 200/ 1236] Overall Loss 0.106750 Objective Loss 0.106750 LR 0.000063 Time 0.025577 +2023-10-02 22:01:14,967 - Epoch: [191][ 210/ 1236] Overall Loss 0.106778 Objective Loss 0.106778 LR 0.000063 Time 0.025354 +2023-10-02 22:01:15,180 - Epoch: [191][ 220/ 1236] Overall Loss 0.106147 Objective Loss 0.106147 LR 0.000063 Time 0.025166 +2023-10-02 22:01:15,391 - Epoch: [191][ 230/ 1236] Overall Loss 0.106735 Objective Loss 0.106735 LR 0.000063 Time 0.024983 +2023-10-02 22:01:15,602 - Epoch: [191][ 240/ 1236] Overall Loss 0.106512 Objective Loss 0.106512 LR 0.000063 Time 0.024821 +2023-10-02 22:01:15,813 - Epoch: [191][ 250/ 1236] Overall Loss 0.107151 Objective Loss 0.107151 LR 0.000063 Time 0.024665 +2023-10-02 22:01:16,026 - Epoch: [191][ 260/ 1236] Overall Loss 0.108117 Objective Loss 0.108117 LR 0.000063 Time 0.024533 +2023-10-02 22:01:16,237 - Epoch: [191][ 270/ 1236] Overall Loss 0.108164 Objective Loss 0.108164 LR 0.000063 Time 0.024401 +2023-10-02 22:01:16,449 - Epoch: [191][ 280/ 1236] Overall Loss 0.108087 Objective Loss 0.108087 LR 0.000063 Time 0.024286 +2023-10-02 22:01:16,660 - Epoch: [191][ 290/ 1236] Overall Loss 0.107521 Objective Loss 0.107521 LR 0.000063 Time 0.024172 +2023-10-02 22:01:16,873 - Epoch: [191][ 300/ 1236] Overall Loss 0.107798 Objective Loss 0.107798 LR 0.000063 Time 0.024074 +2023-10-02 22:01:17,085 - Epoch: [191][ 310/ 1236] Overall Loss 0.107883 Objective Loss 0.107883 LR 0.000063 Time 0.023975 +2023-10-02 22:01:17,294 - Epoch: [191][ 320/ 1236] Overall Loss 0.107849 Objective Loss 0.107849 LR 0.000063 Time 0.023880 +2023-10-02 22:01:17,502 - Epoch: [191][ 330/ 1236] Overall Loss 0.107656 Objective Loss 0.107656 LR 0.000063 Time 0.023784 +2023-10-02 22:01:17,712 - Epoch: [191][ 340/ 1236] Overall Loss 0.107470 Objective Loss 0.107470 LR 0.000063 Time 0.023701 +2023-10-02 22:01:17,922 - Epoch: [191][ 350/ 1236] Overall Loss 0.106910 Objective Loss 0.106910 LR 0.000063 Time 0.023623 +2023-10-02 22:01:18,132 - Epoch: [191][ 360/ 1236] Overall Loss 0.106874 Objective Loss 0.106874 LR 0.000063 Time 0.023550 +2023-10-02 22:01:18,342 - Epoch: [191][ 370/ 1236] Overall Loss 0.106900 Objective Loss 0.106900 LR 0.000063 Time 0.023479 +2023-10-02 22:01:18,552 - Epoch: [191][ 380/ 1236] Overall Loss 0.106950 Objective Loss 0.106950 LR 0.000063 Time 0.023413 +2023-10-02 22:01:18,762 - Epoch: [191][ 390/ 1236] Overall Loss 0.106756 Objective Loss 0.106756 LR 0.000063 Time 0.023349 +2023-10-02 22:01:18,973 - Epoch: [191][ 400/ 1236] Overall Loss 0.106847 Objective Loss 0.106847 LR 0.000063 Time 0.023291 +2023-10-02 22:01:19,183 - Epoch: [191][ 410/ 1236] Overall Loss 0.106721 Objective Loss 0.106721 LR 0.000063 Time 0.023236 +2023-10-02 22:01:19,391 - Epoch: [191][ 420/ 1236] Overall Loss 0.106312 Objective Loss 0.106312 LR 0.000063 Time 0.023176 +2023-10-02 22:01:19,601 - Epoch: [191][ 430/ 1236] Overall Loss 0.106274 Objective Loss 0.106274 LR 0.000063 Time 0.023126 +2023-10-02 22:01:19,809 - Epoch: [191][ 440/ 1236] Overall Loss 0.106782 Objective Loss 0.106782 LR 0.000063 Time 0.023071 +2023-10-02 22:01:20,018 - Epoch: [191][ 450/ 1236] Overall Loss 0.106662 Objective Loss 0.106662 LR 0.000063 Time 0.023023 +2023-10-02 22:01:20,227 - Epoch: [191][ 460/ 1236] Overall Loss 0.106705 Objective Loss 0.106705 LR 0.000063 Time 0.022973 +2023-10-02 22:01:20,437 - Epoch: [191][ 470/ 1236] Overall Loss 0.106452 Objective Loss 0.106452 LR 0.000063 Time 0.022931 +2023-10-02 22:01:20,644 - Epoch: [191][ 480/ 1236] Overall Loss 0.106162 Objective Loss 0.106162 LR 0.000063 Time 0.022884 +2023-10-02 22:01:20,855 - Epoch: [191][ 490/ 1236] Overall Loss 0.106303 Objective Loss 0.106303 LR 0.000063 Time 0.022847 +2023-10-02 22:01:21,062 - Epoch: [191][ 500/ 1236] Overall Loss 0.106058 Objective Loss 0.106058 LR 0.000063 Time 0.022804 +2023-10-02 22:01:21,273 - Epoch: [191][ 510/ 1236] Overall Loss 0.106059 Objective Loss 0.106059 LR 0.000063 Time 0.022769 +2023-10-02 22:01:21,480 - Epoch: [191][ 520/ 1236] Overall Loss 0.106151 Objective Loss 0.106151 LR 0.000063 Time 0.022730 +2023-10-02 22:01:21,691 - Epoch: [191][ 530/ 1236] Overall Loss 0.106234 Objective Loss 0.106234 LR 0.000063 Time 0.022698 +2023-10-02 22:01:21,899 - Epoch: [191][ 540/ 1236] Overall Loss 0.106268 Objective Loss 0.106268 LR 0.000063 Time 0.022661 +2023-10-02 22:01:22,109 - Epoch: [191][ 550/ 1236] Overall Loss 0.106472 Objective Loss 0.106472 LR 0.000063 Time 0.022631 +2023-10-02 22:01:22,316 - Epoch: [191][ 560/ 1236] Overall Loss 0.106316 Objective Loss 0.106316 LR 0.000063 Time 0.022596 +2023-10-02 22:01:22,526 - Epoch: [191][ 570/ 1236] Overall Loss 0.106795 Objective Loss 0.106795 LR 0.000063 Time 0.022568 +2023-10-02 22:01:22,734 - Epoch: [191][ 580/ 1236] Overall Loss 0.107070 Objective Loss 0.107070 LR 0.000063 Time 0.022536 +2023-10-02 22:01:22,944 - Epoch: [191][ 590/ 1236] Overall Loss 0.107042 Objective Loss 0.107042 LR 0.000063 Time 0.022510 +2023-10-02 22:01:23,151 - Epoch: [191][ 600/ 1236] Overall Loss 0.106912 Objective Loss 0.106912 LR 0.000063 Time 0.022480 +2023-10-02 22:01:23,362 - Epoch: [191][ 610/ 1236] Overall Loss 0.106956 Objective Loss 0.106956 LR 0.000063 Time 0.022456 +2023-10-02 22:01:23,569 - Epoch: [191][ 620/ 1236] Overall Loss 0.107024 Objective Loss 0.107024 LR 0.000063 Time 0.022427 +2023-10-02 22:01:23,780 - Epoch: [191][ 630/ 1236] Overall Loss 0.107035 Objective Loss 0.107035 LR 0.000063 Time 0.022405 +2023-10-02 22:01:23,987 - Epoch: [191][ 640/ 1236] Overall Loss 0.107216 Objective Loss 0.107216 LR 0.000063 Time 0.022379 +2023-10-02 22:01:24,198 - Epoch: [191][ 650/ 1236] Overall Loss 0.107134 Objective Loss 0.107134 LR 0.000063 Time 0.022358 +2023-10-02 22:01:24,405 - Epoch: [191][ 660/ 1236] Overall Loss 0.107564 Objective Loss 0.107564 LR 0.000063 Time 0.022333 +2023-10-02 22:01:24,615 - Epoch: [191][ 670/ 1236] Overall Loss 0.107540 Objective Loss 0.107540 LR 0.000063 Time 0.022313 +2023-10-02 22:01:24,823 - Epoch: [191][ 680/ 1236] Overall Loss 0.107636 Objective Loss 0.107636 LR 0.000063 Time 0.022289 +2023-10-02 22:01:25,033 - Epoch: [191][ 690/ 1236] Overall Loss 0.107395 Objective Loss 0.107395 LR 0.000063 Time 0.022271 +2023-10-02 22:01:25,240 - Epoch: [191][ 700/ 1236] Overall Loss 0.107449 Objective Loss 0.107449 LR 0.000063 Time 0.022248 +2023-10-02 22:01:25,451 - Epoch: [191][ 710/ 1236] Overall Loss 0.107548 Objective Loss 0.107548 LR 0.000063 Time 0.022231 +2023-10-02 22:01:25,658 - Epoch: [191][ 720/ 1236] Overall Loss 0.107512 Objective Loss 0.107512 LR 0.000063 Time 0.022210 +2023-10-02 22:01:25,869 - Epoch: [191][ 730/ 1236] Overall Loss 0.107619 Objective Loss 0.107619 LR 0.000063 Time 0.022194 +2023-10-02 22:01:26,077 - Epoch: [191][ 740/ 1236] Overall Loss 0.107455 Objective Loss 0.107455 LR 0.000063 Time 0.022174 +2023-10-02 22:01:26,287 - Epoch: [191][ 750/ 1236] Overall Loss 0.107359 Objective Loss 0.107359 LR 0.000063 Time 0.022159 +2023-10-02 22:01:26,495 - Epoch: [191][ 760/ 1236] Overall Loss 0.107413 Objective Loss 0.107413 LR 0.000063 Time 0.022140 +2023-10-02 22:01:26,705 - Epoch: [191][ 770/ 1236] Overall Loss 0.107446 Objective Loss 0.107446 LR 0.000063 Time 0.022126 +2023-10-02 22:01:26,913 - Epoch: [191][ 780/ 1236] Overall Loss 0.107280 Objective Loss 0.107280 LR 0.000063 Time 0.022108 +2023-10-02 22:01:27,122 - Epoch: [191][ 790/ 1236] Overall Loss 0.107583 Objective Loss 0.107583 LR 0.000063 Time 0.022092 +2023-10-02 22:01:27,331 - Epoch: [191][ 800/ 1236] Overall Loss 0.107671 Objective Loss 0.107671 LR 0.000063 Time 0.022075 +2023-10-02 22:01:27,541 - Epoch: [191][ 810/ 1236] Overall Loss 0.107566 Objective Loss 0.107566 LR 0.000063 Time 0.022062 +2023-10-02 22:01:27,749 - Epoch: [191][ 820/ 1236] Overall Loss 0.107383 Objective Loss 0.107383 LR 0.000063 Time 0.022045 +2023-10-02 22:01:27,958 - Epoch: [191][ 830/ 1236] Overall Loss 0.107403 Objective Loss 0.107403 LR 0.000063 Time 0.022032 +2023-10-02 22:01:28,167 - Epoch: [191][ 840/ 1236] Overall Loss 0.107551 Objective Loss 0.107551 LR 0.000063 Time 0.022016 +2023-10-02 22:01:28,377 - Epoch: [191][ 850/ 1236] Overall Loss 0.107427 Objective Loss 0.107427 LR 0.000063 Time 0.022004 +2023-10-02 22:01:28,584 - Epoch: [191][ 860/ 1236] Overall Loss 0.107424 Objective Loss 0.107424 LR 0.000063 Time 0.021989 +2023-10-02 22:01:28,794 - Epoch: [191][ 870/ 1236] Overall Loss 0.107310 Objective Loss 0.107310 LR 0.000063 Time 0.021977 +2023-10-02 22:01:29,002 - Epoch: [191][ 880/ 1236] Overall Loss 0.107071 Objective Loss 0.107071 LR 0.000063 Time 0.021962 +2023-10-02 22:01:29,213 - Epoch: [191][ 890/ 1236] Overall Loss 0.106922 Objective Loss 0.106922 LR 0.000063 Time 0.021952 +2023-10-02 22:01:29,420 - Epoch: [191][ 900/ 1236] Overall Loss 0.107022 Objective Loss 0.107022 LR 0.000063 Time 0.021938 +2023-10-02 22:01:29,631 - Epoch: [191][ 910/ 1236] Overall Loss 0.107035 Objective Loss 0.107035 LR 0.000063 Time 0.021928 +2023-10-02 22:01:29,839 - Epoch: [191][ 920/ 1236] Overall Loss 0.106998 Objective Loss 0.106998 LR 0.000063 Time 0.021915 +2023-10-02 22:01:30,049 - Epoch: [191][ 930/ 1236] Overall Loss 0.107049 Objective Loss 0.107049 LR 0.000063 Time 0.021905 +2023-10-02 22:01:30,256 - Epoch: [191][ 940/ 1236] Overall Loss 0.107255 Objective Loss 0.107255 LR 0.000063 Time 0.021892 +2023-10-02 22:01:30,467 - Epoch: [191][ 950/ 1236] Overall Loss 0.107229 Objective Loss 0.107229 LR 0.000063 Time 0.021883 +2023-10-02 22:01:30,674 - Epoch: [191][ 960/ 1236] Overall Loss 0.107313 Objective Loss 0.107313 LR 0.000063 Time 0.021871 +2023-10-02 22:01:30,885 - Epoch: [191][ 970/ 1236] Overall Loss 0.107403 Objective Loss 0.107403 LR 0.000063 Time 0.021863 +2023-10-02 22:01:31,093 - Epoch: [191][ 980/ 1236] Overall Loss 0.107432 Objective Loss 0.107432 LR 0.000063 Time 0.021851 +2023-10-02 22:01:31,303 - Epoch: [191][ 990/ 1236] Overall Loss 0.107459 Objective Loss 0.107459 LR 0.000063 Time 0.021843 +2023-10-02 22:01:31,511 - Epoch: [191][ 1000/ 1236] Overall Loss 0.107446 Objective Loss 0.107446 LR 0.000063 Time 0.021831 +2023-10-02 22:01:31,721 - Epoch: [191][ 1010/ 1236] Overall Loss 0.107589 Objective Loss 0.107589 LR 0.000063 Time 0.021823 +2023-10-02 22:01:31,929 - Epoch: [191][ 1020/ 1236] Overall Loss 0.107635 Objective Loss 0.107635 LR 0.000063 Time 0.021813 +2023-10-02 22:01:32,139 - Epoch: [191][ 1030/ 1236] Overall Loss 0.107700 Objective Loss 0.107700 LR 0.000063 Time 0.021805 +2023-10-02 22:01:32,347 - Epoch: [191][ 1040/ 1236] Overall Loss 0.107649 Objective Loss 0.107649 LR 0.000063 Time 0.021794 +2023-10-02 22:01:32,557 - Epoch: [191][ 1050/ 1236] Overall Loss 0.107672 Objective Loss 0.107672 LR 0.000063 Time 0.021787 +2023-10-02 22:01:32,764 - Epoch: [191][ 1060/ 1236] Overall Loss 0.107727 Objective Loss 0.107727 LR 0.000063 Time 0.021777 +2023-10-02 22:01:32,974 - Epoch: [191][ 1070/ 1236] Overall Loss 0.107809 Objective Loss 0.107809 LR 0.000063 Time 0.021769 +2023-10-02 22:01:33,183 - Epoch: [191][ 1080/ 1236] Overall Loss 0.107848 Objective Loss 0.107848 LR 0.000063 Time 0.021759 +2023-10-02 22:01:33,393 - Epoch: [191][ 1090/ 1236] Overall Loss 0.107817 Objective Loss 0.107817 LR 0.000063 Time 0.021753 +2023-10-02 22:01:33,601 - Epoch: [191][ 1100/ 1236] Overall Loss 0.107815 Objective Loss 0.107815 LR 0.000063 Time 0.021743 +2023-10-02 22:01:33,812 - Epoch: [191][ 1110/ 1236] Overall Loss 0.107902 Objective Loss 0.107902 LR 0.000063 Time 0.021737 +2023-10-02 22:01:34,019 - Epoch: [191][ 1120/ 1236] Overall Loss 0.107847 Objective Loss 0.107847 LR 0.000063 Time 0.021728 +2023-10-02 22:01:34,230 - Epoch: [191][ 1130/ 1236] Overall Loss 0.107918 Objective Loss 0.107918 LR 0.000063 Time 0.021722 +2023-10-02 22:01:34,438 - Epoch: [191][ 1140/ 1236] Overall Loss 0.107922 Objective Loss 0.107922 LR 0.000063 Time 0.021713 +2023-10-02 22:01:34,649 - Epoch: [191][ 1150/ 1236] Overall Loss 0.107918 Objective Loss 0.107918 LR 0.000063 Time 0.021707 +2023-10-02 22:01:34,856 - Epoch: [191][ 1160/ 1236] Overall Loss 0.107876 Objective Loss 0.107876 LR 0.000063 Time 0.021699 +2023-10-02 22:01:35,065 - Epoch: [191][ 1170/ 1236] Overall Loss 0.107828 Objective Loss 0.107828 LR 0.000063 Time 0.021692 +2023-10-02 22:01:35,274 - Epoch: [191][ 1180/ 1236] Overall Loss 0.107852 Objective Loss 0.107852 LR 0.000063 Time 0.021684 +2023-10-02 22:01:35,485 - Epoch: [191][ 1190/ 1236] Overall Loss 0.107949 Objective Loss 0.107949 LR 0.000063 Time 0.021678 +2023-10-02 22:01:35,692 - Epoch: [191][ 1200/ 1236] Overall Loss 0.107953 Objective Loss 0.107953 LR 0.000063 Time 0.021670 +2023-10-02 22:01:35,901 - Epoch: [191][ 1210/ 1236] Overall Loss 0.108024 Objective Loss 0.108024 LR 0.000063 Time 0.021664 +2023-10-02 22:01:36,110 - Epoch: [191][ 1220/ 1236] Overall Loss 0.108026 Objective Loss 0.108026 LR 0.000063 Time 0.021656 +2023-10-02 22:01:36,371 - Epoch: [191][ 1230/ 1236] Overall Loss 0.108030 Objective Loss 0.108030 LR 0.000063 Time 0.021692 +2023-10-02 22:01:36,492 - Epoch: [191][ 1236/ 1236] Overall Loss 0.108152 Objective Loss 0.108152 Top1 91.649695 Top5 98.574338 LR 0.000063 Time 0.021685 +2023-10-02 22:01:36,616 - --- validate (epoch=191)----------- +2023-10-02 22:01:36,617 - 29943 samples (256 per mini-batch) +2023-10-02 22:01:37,111 - Epoch: [191][ 10/ 117] Loss 0.299597 Top1 87.187500 Top5 98.671875 +2023-10-02 22:01:37,275 - Epoch: [191][ 20/ 117] Loss 0.296432 Top1 87.597656 Top5 98.886719 +2023-10-02 22:01:37,435 - Epoch: [191][ 30/ 117] Loss 0.296367 Top1 87.513021 Top5 98.854167 +2023-10-02 22:01:37,598 - Epoch: [191][ 40/ 117] Loss 0.302746 Top1 87.275391 Top5 98.710938 +2023-10-02 22:01:37,757 - Epoch: [191][ 50/ 117] Loss 0.301078 Top1 87.421875 Top5 98.695312 +2023-10-02 22:01:37,920 - Epoch: [191][ 60/ 117] Loss 0.296951 Top1 87.669271 Top5 98.743490 +2023-10-02 22:01:38,079 - Epoch: [191][ 70/ 117] Loss 0.301351 Top1 87.561384 Top5 98.694196 +2023-10-02 22:01:38,239 - Epoch: [191][ 80/ 117] Loss 0.303082 Top1 87.412109 Top5 98.691406 +2023-10-02 22:01:38,395 - Epoch: [191][ 90/ 117] Loss 0.305323 Top1 87.482639 Top5 98.654514 +2023-10-02 22:01:38,555 - Epoch: [191][ 100/ 117] Loss 0.302803 Top1 87.484375 Top5 98.664062 +2023-10-02 22:01:38,720 - Epoch: [191][ 110/ 117] Loss 0.302393 Top1 87.524858 Top5 98.707386 +2023-10-02 22:01:38,809 - Epoch: [191][ 117/ 117] Loss 0.302874 Top1 87.506262 Top5 98.697525 +2023-10-02 22:01:38,950 - ==> Top1: 87.506 Top5: 98.698 Loss: 0.303 + +2023-10-02 22:01:38,951 - ==> Confusion: +[[ 945 0 2 1 3 2 0 0 4 62 2 0 1 2 6 0 0 0 1 0 19] + [ 0 1077 1 1 4 12 1 15 0 1 0 0 0 0 2 3 0 0 5 3 6] + [ 3 1 982 9 1 0 13 9 0 2 1 1 7 3 1 2 2 1 8 3 7] + [ 3 4 12 988 0 1 0 1 2 1 4 0 7 3 26 3 1 4 9 1 19] + [ 25 4 1 0 971 4 1 0 1 13 0 0 0 3 12 5 9 0 0 0 1] + [ 3 37 0 1 6 999 1 18 1 5 2 5 1 10 4 0 3 0 4 1 15] + [ 0 3 25 0 0 2 1135 6 0 0 3 1 0 0 0 4 0 1 1 5 5] + [ 1 8 10 1 5 22 6 1082 2 3 6 5 3 5 1 0 0 2 33 13 10] + [ 15 2 0 1 1 2 0 2 989 36 8 1 0 9 14 0 3 1 3 0 2] + [ 91 2 1 2 5 2 0 0 27 959 0 0 0 15 6 2 0 0 0 2 5] + [ 3 2 8 7 0 2 2 3 14 1 967 1 0 12 7 0 2 2 4 3 13] + [ 0 1 0 0 0 12 0 8 0 1 0 968 13 5 0 1 1 16 0 2 7] + [ 0 1 1 1 1 0 1 1 0 2 4 30 978 2 4 8 0 10 3 4 17] + [ 0 0 0 0 4 8 0 0 13 13 2 8 0 1050 4 0 0 1 0 1 15] + [ 12 0 5 17 2 1 0 0 21 3 1 0 2 3 1022 0 1 2 6 0 3] + [ 0 0 2 1 5 2 1 0 0 0 0 5 7 0 0 1070 16 9 2 9 5] + [ 0 17 1 0 6 6 0 0 0 0 0 4 0 3 3 6 1096 0 2 6 11] + [ 0 0 0 2 1 0 2 0 0 0 0 3 19 3 4 6 0 992 0 2 4] + [ 4 3 2 17 0 0 0 17 4 1 2 1 1 0 11 0 0 0 993 0 12] + [ 1 1 4 3 0 3 8 5 0 0 1 14 5 1 2 1 6 0 0 1088 9] + [ 111 134 109 84 61 99 25 80 86 59 137 78 273 211 117 45 76 45 93 131 5851]] + +2023-10-02 22:01:38,952 - ==> Best [Top1: 87.874 Top5: 98.708 Sparsity:0.00 Params: 169472 on epoch: 189] +2023-10-02 22:01:38,952 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:01:38,958 - + +2023-10-02 22:01:38,958 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:01:39,981 - Epoch: [192][ 10/ 1236] Overall Loss 0.107263 Objective Loss 0.107263 LR 0.000063 Time 0.102235 +2023-10-02 22:01:40,191 - Epoch: [192][ 20/ 1236] Overall Loss 0.105853 Objective Loss 0.105853 LR 0.000063 Time 0.061582 +2023-10-02 22:01:40,398 - Epoch: [192][ 30/ 1236] Overall Loss 0.103674 Objective Loss 0.103674 LR 0.000063 Time 0.047966 +2023-10-02 22:01:40,608 - Epoch: [192][ 40/ 1236] Overall Loss 0.102715 Objective Loss 0.102715 LR 0.000063 Time 0.041209 +2023-10-02 22:01:40,816 - Epoch: [192][ 50/ 1236] Overall Loss 0.100632 Objective Loss 0.100632 LR 0.000063 Time 0.037121 +2023-10-02 22:01:41,026 - Epoch: [192][ 60/ 1236] Overall Loss 0.104748 Objective Loss 0.104748 LR 0.000063 Time 0.034422 +2023-10-02 22:01:41,233 - Epoch: [192][ 70/ 1236] Overall Loss 0.104547 Objective Loss 0.104547 LR 0.000063 Time 0.032467 +2023-10-02 22:01:41,443 - Epoch: [192][ 80/ 1236] Overall Loss 0.103164 Objective Loss 0.103164 LR 0.000063 Time 0.031027 +2023-10-02 22:01:41,651 - Epoch: [192][ 90/ 1236] Overall Loss 0.103354 Objective Loss 0.103354 LR 0.000063 Time 0.029886 +2023-10-02 22:01:41,859 - Epoch: [192][ 100/ 1236] Overall Loss 0.102978 Objective Loss 0.102978 LR 0.000063 Time 0.028974 +2023-10-02 22:01:42,067 - Epoch: [192][ 110/ 1236] Overall Loss 0.102820 Objective Loss 0.102820 LR 0.000063 Time 0.028221 +2023-10-02 22:01:42,277 - Epoch: [192][ 120/ 1236] Overall Loss 0.102950 Objective Loss 0.102950 LR 0.000063 Time 0.027611 +2023-10-02 22:01:42,483 - Epoch: [192][ 130/ 1236] Overall Loss 0.103834 Objective Loss 0.103834 LR 0.000063 Time 0.027071 +2023-10-02 22:01:42,693 - Epoch: [192][ 140/ 1236] Overall Loss 0.104126 Objective Loss 0.104126 LR 0.000063 Time 0.026635 +2023-10-02 22:01:42,900 - Epoch: [192][ 150/ 1236] Overall Loss 0.104766 Objective Loss 0.104766 LR 0.000063 Time 0.026237 +2023-10-02 22:01:43,110 - Epoch: [192][ 160/ 1236] Overall Loss 0.105202 Objective Loss 0.105202 LR 0.000063 Time 0.025912 +2023-10-02 22:01:43,317 - Epoch: [192][ 170/ 1236] Overall Loss 0.105784 Objective Loss 0.105784 LR 0.000063 Time 0.025604 +2023-10-02 22:01:43,528 - Epoch: [192][ 180/ 1236] Overall Loss 0.104474 Objective Loss 0.104474 LR 0.000063 Time 0.025349 +2023-10-02 22:01:43,735 - Epoch: [192][ 190/ 1236] Overall Loss 0.104044 Objective Loss 0.104044 LR 0.000063 Time 0.025103 +2023-10-02 22:01:43,945 - Epoch: [192][ 200/ 1236] Overall Loss 0.103460 Objective Loss 0.103460 LR 0.000063 Time 0.024899 +2023-10-02 22:01:44,152 - Epoch: [192][ 210/ 1236] Overall Loss 0.103194 Objective Loss 0.103194 LR 0.000063 Time 0.024697 +2023-10-02 22:01:44,363 - Epoch: [192][ 220/ 1236] Overall Loss 0.103542 Objective Loss 0.103542 LR 0.000063 Time 0.024530 +2023-10-02 22:01:44,570 - Epoch: [192][ 230/ 1236] Overall Loss 0.102757 Objective Loss 0.102757 LR 0.000063 Time 0.024363 +2023-10-02 22:01:44,780 - Epoch: [192][ 240/ 1236] Overall Loss 0.102840 Objective Loss 0.102840 LR 0.000063 Time 0.024223 +2023-10-02 22:01:44,987 - Epoch: [192][ 250/ 1236] Overall Loss 0.103147 Objective Loss 0.103147 LR 0.000063 Time 0.024082 +2023-10-02 22:01:45,196 - Epoch: [192][ 260/ 1236] Overall Loss 0.102996 Objective Loss 0.102996 LR 0.000063 Time 0.023959 +2023-10-02 22:01:45,405 - Epoch: [192][ 270/ 1236] Overall Loss 0.102537 Objective Loss 0.102537 LR 0.000063 Time 0.023837 +2023-10-02 22:01:45,615 - Epoch: [192][ 280/ 1236] Overall Loss 0.102739 Objective Loss 0.102739 LR 0.000063 Time 0.023737 +2023-10-02 22:01:45,822 - Epoch: [192][ 290/ 1236] Overall Loss 0.102731 Objective Loss 0.102731 LR 0.000063 Time 0.023631 +2023-10-02 22:01:46,032 - Epoch: [192][ 300/ 1236] Overall Loss 0.102967 Objective Loss 0.102967 LR 0.000063 Time 0.023541 +2023-10-02 22:01:46,240 - Epoch: [192][ 310/ 1236] Overall Loss 0.103082 Objective Loss 0.103082 LR 0.000063 Time 0.023449 +2023-10-02 22:01:46,449 - Epoch: [192][ 320/ 1236] Overall Loss 0.103128 Objective Loss 0.103128 LR 0.000063 Time 0.023369 +2023-10-02 22:01:46,657 - Epoch: [192][ 330/ 1236] Overall Loss 0.103059 Objective Loss 0.103059 LR 0.000063 Time 0.023289 +2023-10-02 22:01:46,866 - Epoch: [192][ 340/ 1236] Overall Loss 0.103346 Objective Loss 0.103346 LR 0.000063 Time 0.023217 +2023-10-02 22:01:47,071 - Epoch: [192][ 350/ 1236] Overall Loss 0.103894 Objective Loss 0.103894 LR 0.000063 Time 0.023140 +2023-10-02 22:01:47,279 - Epoch: [192][ 360/ 1236] Overall Loss 0.104151 Objective Loss 0.104151 LR 0.000063 Time 0.023074 +2023-10-02 22:01:47,485 - Epoch: [192][ 370/ 1236] Overall Loss 0.104579 Objective Loss 0.104579 LR 0.000063 Time 0.023006 +2023-10-02 22:01:47,693 - Epoch: [192][ 380/ 1236] Overall Loss 0.104451 Objective Loss 0.104451 LR 0.000063 Time 0.022947 +2023-10-02 22:01:47,899 - Epoch: [192][ 390/ 1236] Overall Loss 0.104079 Objective Loss 0.104079 LR 0.000063 Time 0.022886 +2023-10-02 22:01:48,107 - Epoch: [192][ 400/ 1236] Overall Loss 0.104614 Objective Loss 0.104614 LR 0.000063 Time 0.022833 +2023-10-02 22:01:48,313 - Epoch: [192][ 410/ 1236] Overall Loss 0.104727 Objective Loss 0.104727 LR 0.000063 Time 0.022779 +2023-10-02 22:01:48,522 - Epoch: [192][ 420/ 1236] Overall Loss 0.104758 Objective Loss 0.104758 LR 0.000063 Time 0.022732 +2023-10-02 22:01:48,727 - Epoch: [192][ 430/ 1236] Overall Loss 0.104988 Objective Loss 0.104988 LR 0.000063 Time 0.022680 +2023-10-02 22:01:48,934 - Epoch: [192][ 440/ 1236] Overall Loss 0.105165 Objective Loss 0.105165 LR 0.000063 Time 0.022635 +2023-10-02 22:01:49,141 - Epoch: [192][ 450/ 1236] Overall Loss 0.105363 Objective Loss 0.105363 LR 0.000063 Time 0.022590 +2023-10-02 22:01:49,348 - Epoch: [192][ 460/ 1236] Overall Loss 0.105698 Objective Loss 0.105698 LR 0.000063 Time 0.022550 +2023-10-02 22:01:49,555 - Epoch: [192][ 470/ 1236] Overall Loss 0.105708 Objective Loss 0.105708 LR 0.000063 Time 0.022506 +2023-10-02 22:01:49,762 - Epoch: [192][ 480/ 1236] Overall Loss 0.106075 Objective Loss 0.106075 LR 0.000063 Time 0.022469 +2023-10-02 22:01:49,969 - Epoch: [192][ 490/ 1236] Overall Loss 0.106212 Objective Loss 0.106212 LR 0.000063 Time 0.022430 +2023-10-02 22:01:50,176 - Epoch: [192][ 500/ 1236] Overall Loss 0.106290 Objective Loss 0.106290 LR 0.000063 Time 0.022395 +2023-10-02 22:01:50,383 - Epoch: [192][ 510/ 1236] Overall Loss 0.106433 Objective Loss 0.106433 LR 0.000063 Time 0.022358 +2023-10-02 22:01:50,592 - Epoch: [192][ 520/ 1236] Overall Loss 0.106227 Objective Loss 0.106227 LR 0.000063 Time 0.022329 +2023-10-02 22:01:50,797 - Epoch: [192][ 530/ 1236] Overall Loss 0.106208 Objective Loss 0.106208 LR 0.000063 Time 0.022295 +2023-10-02 22:01:51,005 - Epoch: [192][ 540/ 1236] Overall Loss 0.106350 Objective Loss 0.106350 LR 0.000063 Time 0.022266 +2023-10-02 22:01:51,212 - Epoch: [192][ 550/ 1236] Overall Loss 0.106414 Objective Loss 0.106414 LR 0.000063 Time 0.022234 +2023-10-02 22:01:51,419 - Epoch: [192][ 560/ 1236] Overall Loss 0.106788 Objective Loss 0.106788 LR 0.000063 Time 0.022207 +2023-10-02 22:01:51,625 - Epoch: [192][ 570/ 1236] Overall Loss 0.107157 Objective Loss 0.107157 LR 0.000063 Time 0.022177 +2023-10-02 22:01:51,834 - Epoch: [192][ 580/ 1236] Overall Loss 0.107268 Objective Loss 0.107268 LR 0.000063 Time 0.022154 +2023-10-02 22:01:52,040 - Epoch: [192][ 590/ 1236] Overall Loss 0.107207 Objective Loss 0.107207 LR 0.000063 Time 0.022127 +2023-10-02 22:01:52,248 - Epoch: [192][ 600/ 1236] Overall Loss 0.107082 Objective Loss 0.107082 LR 0.000063 Time 0.022105 +2023-10-02 22:01:52,454 - Epoch: [192][ 610/ 1236] Overall Loss 0.106937 Objective Loss 0.106937 LR 0.000063 Time 0.022079 +2023-10-02 22:01:52,661 - Epoch: [192][ 620/ 1236] Overall Loss 0.107093 Objective Loss 0.107093 LR 0.000063 Time 0.022057 +2023-10-02 22:01:52,868 - Epoch: [192][ 630/ 1236] Overall Loss 0.106955 Objective Loss 0.106955 LR 0.000063 Time 0.022033 +2023-10-02 22:01:53,077 - Epoch: [192][ 640/ 1236] Overall Loss 0.106905 Objective Loss 0.106905 LR 0.000063 Time 0.022014 +2023-10-02 22:01:53,282 - Epoch: [192][ 650/ 1236] Overall Loss 0.107213 Objective Loss 0.107213 LR 0.000063 Time 0.021991 +2023-10-02 22:01:53,489 - Epoch: [192][ 660/ 1236] Overall Loss 0.107393 Objective Loss 0.107393 LR 0.000063 Time 0.021971 +2023-10-02 22:01:53,696 - Epoch: [192][ 670/ 1236] Overall Loss 0.107232 Objective Loss 0.107232 LR 0.000063 Time 0.021950 +2023-10-02 22:01:53,903 - Epoch: [192][ 680/ 1236] Overall Loss 0.107182 Objective Loss 0.107182 LR 0.000063 Time 0.021931 +2023-10-02 22:01:54,110 - Epoch: [192][ 690/ 1236] Overall Loss 0.107304 Objective Loss 0.107304 LR 0.000063 Time 0.021911 +2023-10-02 22:01:54,319 - Epoch: [192][ 700/ 1236] Overall Loss 0.107232 Objective Loss 0.107232 LR 0.000063 Time 0.021896 +2023-10-02 22:01:54,525 - Epoch: [192][ 710/ 1236] Overall Loss 0.107276 Objective Loss 0.107276 LR 0.000063 Time 0.021877 +2023-10-02 22:01:54,732 - Epoch: [192][ 720/ 1236] Overall Loss 0.107195 Objective Loss 0.107195 LR 0.000063 Time 0.021861 +2023-10-02 22:01:54,939 - Epoch: [192][ 730/ 1236] Overall Loss 0.107115 Objective Loss 0.107115 LR 0.000063 Time 0.021843 +2023-10-02 22:01:55,148 - Epoch: [192][ 740/ 1236] Overall Loss 0.107009 Objective Loss 0.107009 LR 0.000063 Time 0.021829 +2023-10-02 22:01:55,353 - Epoch: [192][ 750/ 1236] Overall Loss 0.107218 Objective Loss 0.107218 LR 0.000063 Time 0.021812 +2023-10-02 22:01:55,562 - Epoch: [192][ 760/ 1236] Overall Loss 0.107182 Objective Loss 0.107182 LR 0.000063 Time 0.021799 +2023-10-02 22:01:55,767 - Epoch: [192][ 770/ 1236] Overall Loss 0.107017 Objective Loss 0.107017 LR 0.000063 Time 0.021782 +2023-10-02 22:01:55,974 - Epoch: [192][ 780/ 1236] Overall Loss 0.107025 Objective Loss 0.107025 LR 0.000063 Time 0.021768 +2023-10-02 22:01:56,181 - Epoch: [192][ 790/ 1236] Overall Loss 0.107283 Objective Loss 0.107283 LR 0.000063 Time 0.021752 +2023-10-02 22:01:56,390 - Epoch: [192][ 800/ 1236] Overall Loss 0.107425 Objective Loss 0.107425 LR 0.000063 Time 0.021741 +2023-10-02 22:01:56,596 - Epoch: [192][ 810/ 1236] Overall Loss 0.107367 Objective Loss 0.107367 LR 0.000063 Time 0.021726 +2023-10-02 22:01:56,804 - Epoch: [192][ 820/ 1236] Overall Loss 0.107478 Objective Loss 0.107478 LR 0.000063 Time 0.021716 +2023-10-02 22:01:57,010 - Epoch: [192][ 830/ 1236] Overall Loss 0.107482 Objective Loss 0.107482 LR 0.000063 Time 0.021701 +2023-10-02 22:01:57,217 - Epoch: [192][ 840/ 1236] Overall Loss 0.107518 Objective Loss 0.107518 LR 0.000063 Time 0.021689 +2023-10-02 22:01:57,424 - Epoch: [192][ 850/ 1236] Overall Loss 0.107642 Objective Loss 0.107642 LR 0.000063 Time 0.021676 +2023-10-02 22:01:57,632 - Epoch: [192][ 860/ 1236] Overall Loss 0.107656 Objective Loss 0.107656 LR 0.000063 Time 0.021665 +2023-10-02 22:01:57,839 - Epoch: [192][ 870/ 1236] Overall Loss 0.107721 Objective Loss 0.107721 LR 0.000063 Time 0.021652 +2023-10-02 22:01:58,048 - Epoch: [192][ 880/ 1236] Overall Loss 0.107555 Objective Loss 0.107555 LR 0.000063 Time 0.021643 +2023-10-02 22:01:58,253 - Epoch: [192][ 890/ 1236] Overall Loss 0.107536 Objective Loss 0.107536 LR 0.000063 Time 0.021631 +2023-10-02 22:01:58,462 - Epoch: [192][ 900/ 1236] Overall Loss 0.107374 Objective Loss 0.107374 LR 0.000063 Time 0.021622 +2023-10-02 22:01:58,668 - Epoch: [192][ 910/ 1236] Overall Loss 0.107519 Objective Loss 0.107519 LR 0.000063 Time 0.021610 +2023-10-02 22:01:58,877 - Epoch: [192][ 920/ 1236] Overall Loss 0.107431 Objective Loss 0.107431 LR 0.000063 Time 0.021602 +2023-10-02 22:01:59,082 - Epoch: [192][ 930/ 1236] Overall Loss 0.107562 Objective Loss 0.107562 LR 0.000063 Time 0.021590 +2023-10-02 22:01:59,289 - Epoch: [192][ 940/ 1236] Overall Loss 0.107466 Objective Loss 0.107466 LR 0.000063 Time 0.021581 +2023-10-02 22:01:59,496 - Epoch: [192][ 950/ 1236] Overall Loss 0.107640 Objective Loss 0.107640 LR 0.000063 Time 0.021570 +2023-10-02 22:01:59,703 - Epoch: [192][ 960/ 1236] Overall Loss 0.107541 Objective Loss 0.107541 LR 0.000063 Time 0.021560 +2023-10-02 22:01:59,910 - Epoch: [192][ 970/ 1236] Overall Loss 0.107621 Objective Loss 0.107621 LR 0.000063 Time 0.021550 +2023-10-02 22:02:00,118 - Epoch: [192][ 980/ 1236] Overall Loss 0.107699 Objective Loss 0.107699 LR 0.000063 Time 0.021542 +2023-10-02 22:02:00,325 - Epoch: [192][ 990/ 1236] Overall Loss 0.107606 Objective Loss 0.107606 LR 0.000063 Time 0.021532 +2023-10-02 22:02:00,534 - Epoch: [192][ 1000/ 1236] Overall Loss 0.107433 Objective Loss 0.107433 LR 0.000063 Time 0.021525 +2023-10-02 22:02:00,740 - Epoch: [192][ 1010/ 1236] Overall Loss 0.107377 Objective Loss 0.107377 LR 0.000063 Time 0.021515 +2023-10-02 22:02:00,947 - Epoch: [192][ 1020/ 1236] Overall Loss 0.107361 Objective Loss 0.107361 LR 0.000063 Time 0.021508 +2023-10-02 22:02:01,154 - Epoch: [192][ 1030/ 1236] Overall Loss 0.107327 Objective Loss 0.107327 LR 0.000063 Time 0.021498 +2023-10-02 22:02:01,361 - Epoch: [192][ 1040/ 1236] Overall Loss 0.107456 Objective Loss 0.107456 LR 0.000063 Time 0.021491 +2023-10-02 22:02:01,569 - Epoch: [192][ 1050/ 1236] Overall Loss 0.107361 Objective Loss 0.107361 LR 0.000063 Time 0.021482 +2023-10-02 22:02:01,776 - Epoch: [192][ 1060/ 1236] Overall Loss 0.107273 Objective Loss 0.107273 LR 0.000063 Time 0.021474 +2023-10-02 22:02:01,983 - Epoch: [192][ 1070/ 1236] Overall Loss 0.107386 Objective Loss 0.107386 LR 0.000063 Time 0.021465 +2023-10-02 22:02:02,191 - Epoch: [192][ 1080/ 1236] Overall Loss 0.107438 Objective Loss 0.107438 LR 0.000063 Time 0.021460 +2023-10-02 22:02:02,397 - Epoch: [192][ 1090/ 1236] Overall Loss 0.107361 Objective Loss 0.107361 LR 0.000063 Time 0.021451 +2023-10-02 22:02:02,606 - Epoch: [192][ 1100/ 1236] Overall Loss 0.107312 Objective Loss 0.107312 LR 0.000063 Time 0.021446 +2023-10-02 22:02:02,811 - Epoch: [192][ 1110/ 1236] Overall Loss 0.107285 Objective Loss 0.107285 LR 0.000063 Time 0.021438 +2023-10-02 22:02:03,021 - Epoch: [192][ 1120/ 1236] Overall Loss 0.107299 Objective Loss 0.107299 LR 0.000063 Time 0.021433 +2023-10-02 22:02:03,227 - Epoch: [192][ 1130/ 1236] Overall Loss 0.107398 Objective Loss 0.107398 LR 0.000063 Time 0.021425 +2023-10-02 22:02:03,434 - Epoch: [192][ 1140/ 1236] Overall Loss 0.107429 Objective Loss 0.107429 LR 0.000063 Time 0.021419 +2023-10-02 22:02:03,641 - Epoch: [192][ 1150/ 1236] Overall Loss 0.107378 Objective Loss 0.107378 LR 0.000063 Time 0.021412 +2023-10-02 22:02:03,849 - Epoch: [192][ 1160/ 1236] Overall Loss 0.107521 Objective Loss 0.107521 LR 0.000063 Time 0.021406 +2023-10-02 22:02:04,055 - Epoch: [192][ 1170/ 1236] Overall Loss 0.107538 Objective Loss 0.107538 LR 0.000063 Time 0.021398 +2023-10-02 22:02:04,263 - Epoch: [192][ 1180/ 1236] Overall Loss 0.107568 Objective Loss 0.107568 LR 0.000063 Time 0.021392 +2023-10-02 22:02:04,470 - Epoch: [192][ 1190/ 1236] Overall Loss 0.107536 Objective Loss 0.107536 LR 0.000063 Time 0.021385 +2023-10-02 22:02:04,677 - Epoch: [192][ 1200/ 1236] Overall Loss 0.107641 Objective Loss 0.107641 LR 0.000063 Time 0.021379 +2023-10-02 22:02:04,884 - Epoch: [192][ 1210/ 1236] Overall Loss 0.107588 Objective Loss 0.107588 LR 0.000063 Time 0.021372 +2023-10-02 22:02:05,093 - Epoch: [192][ 1220/ 1236] Overall Loss 0.107568 Objective Loss 0.107568 LR 0.000063 Time 0.021368 +2023-10-02 22:02:05,350 - Epoch: [192][ 1230/ 1236] Overall Loss 0.107453 Objective Loss 0.107453 LR 0.000063 Time 0.021403 +2023-10-02 22:02:05,472 - Epoch: [192][ 1236/ 1236] Overall Loss 0.107423 Objective Loss 0.107423 Top1 90.835031 Top5 98.167006 LR 0.000063 Time 0.021398 +2023-10-02 22:02:05,609 - --- validate (epoch=192)----------- +2023-10-02 22:02:05,609 - 29943 samples (256 per mini-batch) +2023-10-02 22:02:06,099 - Epoch: [192][ 10/ 117] Loss 0.303168 Top1 88.125000 Top5 99.023438 +2023-10-02 22:02:06,250 - Epoch: [192][ 20/ 117] Loss 0.295082 Top1 87.714844 Top5 98.847656 +2023-10-02 22:02:06,401 - Epoch: [192][ 30/ 117] Loss 0.297051 Top1 87.695312 Top5 98.776042 +2023-10-02 22:02:06,552 - Epoch: [192][ 40/ 117] Loss 0.296609 Top1 88.037109 Top5 98.759766 +2023-10-02 22:02:06,703 - Epoch: [192][ 50/ 117] Loss 0.295242 Top1 87.921875 Top5 98.781250 +2023-10-02 22:02:06,854 - Epoch: [192][ 60/ 117] Loss 0.294755 Top1 87.962240 Top5 98.795573 +2023-10-02 22:02:07,004 - Epoch: [192][ 70/ 117] Loss 0.292600 Top1 87.968750 Top5 98.805804 +2023-10-02 22:02:07,158 - Epoch: [192][ 80/ 117] Loss 0.299649 Top1 87.783203 Top5 98.798828 +2023-10-02 22:02:07,307 - Epoch: [192][ 90/ 117] Loss 0.299007 Top1 87.747396 Top5 98.810764 +2023-10-02 22:02:07,459 - Epoch: [192][ 100/ 117] Loss 0.300615 Top1 87.746094 Top5 98.820312 +2023-10-02 22:02:07,617 - Epoch: [192][ 110/ 117] Loss 0.302664 Top1 87.794744 Top5 98.792614 +2023-10-02 22:02:07,705 - Epoch: [192][ 117/ 117] Loss 0.304483 Top1 87.756738 Top5 98.774338 +2023-10-02 22:02:07,821 - ==> Top1: 87.757 Top5: 98.774 Loss: 0.304 + +2023-10-02 22:02:07,822 - ==> Confusion: +[[ 935 0 4 1 2 2 0 0 5 63 1 0 1 3 7 0 2 0 1 0 23] + [ 0 1064 0 1 2 16 1 21 0 0 0 1 1 0 2 2 1 0 6 3 10] + [ 1 0 980 6 1 0 18 10 0 1 1 1 7 2 1 2 3 2 10 1 9] + [ 1 4 12 983 2 2 0 1 3 1 6 1 7 3 27 2 1 4 10 1 18] + [ 26 4 0 1 966 4 1 0 0 12 0 0 0 5 12 4 9 0 0 0 6] + [ 3 27 0 0 5 1007 1 20 1 4 2 6 1 9 5 0 4 0 3 2 16] + [ 0 3 21 0 0 1 1140 4 0 0 2 1 0 0 0 4 0 1 2 7 5] + [ 1 8 8 0 6 25 5 1085 1 2 6 4 2 4 1 2 0 1 35 10 12] + [ 14 3 0 0 2 2 0 3 978 35 10 1 0 16 13 0 3 1 3 0 5] + [ 92 0 1 1 6 2 0 0 21 952 0 0 1 23 8 1 1 0 0 1 9] + [ 2 1 6 6 0 2 3 3 7 2 969 2 0 17 6 0 3 3 8 1 12] + [ 0 1 0 0 0 14 0 3 0 0 0 974 10 6 0 2 0 17 0 4 4] + [ 0 0 1 1 1 2 2 1 0 1 2 29 978 2 4 8 0 14 2 5 15] + [ 0 0 0 0 4 5 0 0 8 8 2 7 0 1064 4 0 0 1 0 2 14] + [ 14 0 3 12 3 1 0 0 19 3 2 0 3 2 1020 0 0 1 9 0 9] + [ 0 0 2 1 5 1 0 0 0 0 0 6 7 0 0 1063 19 11 2 10 7] + [ 0 15 1 0 4 6 1 1 0 1 0 4 0 3 3 8 1094 0 2 6 12] + [ 0 0 0 1 0 0 2 0 0 0 0 7 13 2 2 7 1 997 0 2 4] + [ 2 4 2 14 0 0 0 18 3 1 2 1 2 0 10 0 0 1 997 0 11] + [ 0 0 3 1 2 3 7 3 0 0 0 11 5 2 2 0 6 1 1 1096 9] + [ 91 105 95 63 57 115 30 84 68 52 132 84 264 228 112 42 64 52 104 128 5935]] + +2023-10-02 22:02:07,823 - ==> Best [Top1: 87.874 Top5: 98.708 Sparsity:0.00 Params: 169472 on epoch: 189] +2023-10-02 22:02:07,823 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:02:07,829 - + +2023-10-02 22:02:07,829 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:02:08,974 - Epoch: [193][ 10/ 1236] Overall Loss 0.110042 Objective Loss 0.110042 LR 0.000063 Time 0.114454 +2023-10-02 22:02:09,185 - Epoch: [193][ 20/ 1236] Overall Loss 0.106377 Objective Loss 0.106377 LR 0.000063 Time 0.067749 +2023-10-02 22:02:09,395 - Epoch: [193][ 30/ 1236] Overall Loss 0.103892 Objective Loss 0.103892 LR 0.000063 Time 0.052145 +2023-10-02 22:02:09,605 - Epoch: [193][ 40/ 1236] Overall Loss 0.105181 Objective Loss 0.105181 LR 0.000063 Time 0.044344 +2023-10-02 22:02:09,813 - Epoch: [193][ 50/ 1236] Overall Loss 0.101882 Objective Loss 0.101882 LR 0.000063 Time 0.039608 +2023-10-02 22:02:10,020 - Epoch: [193][ 60/ 1236] Overall Loss 0.101967 Objective Loss 0.101967 LR 0.000063 Time 0.036460 +2023-10-02 22:02:10,227 - Epoch: [193][ 70/ 1236] Overall Loss 0.101752 Objective Loss 0.101752 LR 0.000063 Time 0.034185 +2023-10-02 22:02:10,436 - Epoch: [193][ 80/ 1236] Overall Loss 0.101553 Objective Loss 0.101553 LR 0.000063 Time 0.032519 +2023-10-02 22:02:10,644 - Epoch: [193][ 90/ 1236] Overall Loss 0.103614 Objective Loss 0.103614 LR 0.000063 Time 0.031199 +2023-10-02 22:02:10,852 - Epoch: [193][ 100/ 1236] Overall Loss 0.105542 Objective Loss 0.105542 LR 0.000063 Time 0.030161 +2023-10-02 22:02:11,061 - Epoch: [193][ 110/ 1236] Overall Loss 0.105300 Objective Loss 0.105300 LR 0.000063 Time 0.029300 +2023-10-02 22:02:11,278 - Epoch: [193][ 120/ 1236] Overall Loss 0.105415 Objective Loss 0.105415 LR 0.000063 Time 0.028666 +2023-10-02 22:02:11,492 - Epoch: [193][ 130/ 1236] Overall Loss 0.105827 Objective Loss 0.105827 LR 0.000063 Time 0.028106 +2023-10-02 22:02:11,710 - Epoch: [193][ 140/ 1236] Overall Loss 0.105842 Objective Loss 0.105842 LR 0.000063 Time 0.027651 +2023-10-02 22:02:11,922 - Epoch: [193][ 150/ 1236] Overall Loss 0.105781 Objective Loss 0.105781 LR 0.000063 Time 0.027221 +2023-10-02 22:02:12,140 - Epoch: [193][ 160/ 1236] Overall Loss 0.106502 Objective Loss 0.106502 LR 0.000063 Time 0.026879 +2023-10-02 22:02:12,353 - Epoch: [193][ 170/ 1236] Overall Loss 0.105905 Objective Loss 0.105905 LR 0.000063 Time 0.026545 +2023-10-02 22:02:12,570 - Epoch: [193][ 180/ 1236] Overall Loss 0.106529 Objective Loss 0.106529 LR 0.000063 Time 0.026276 +2023-10-02 22:02:12,783 - Epoch: [193][ 190/ 1236] Overall Loss 0.107249 Objective Loss 0.107249 LR 0.000063 Time 0.026010 +2023-10-02 22:02:13,000 - Epoch: [193][ 200/ 1236] Overall Loss 0.108338 Objective Loss 0.108338 LR 0.000063 Time 0.025795 +2023-10-02 22:02:13,213 - Epoch: [193][ 210/ 1236] Overall Loss 0.107847 Objective Loss 0.107847 LR 0.000063 Time 0.025577 +2023-10-02 22:02:13,430 - Epoch: [193][ 220/ 1236] Overall Loss 0.108545 Objective Loss 0.108545 LR 0.000063 Time 0.025401 +2023-10-02 22:02:13,643 - Epoch: [193][ 230/ 1236] Overall Loss 0.109282 Objective Loss 0.109282 LR 0.000063 Time 0.025220 +2023-10-02 22:02:13,860 - Epoch: [193][ 240/ 1236] Overall Loss 0.109088 Objective Loss 0.109088 LR 0.000063 Time 0.025073 +2023-10-02 22:02:14,073 - Epoch: [193][ 250/ 1236] Overall Loss 0.108984 Objective Loss 0.108984 LR 0.000063 Time 0.024919 +2023-10-02 22:02:14,290 - Epoch: [193][ 260/ 1236] Overall Loss 0.108683 Objective Loss 0.108683 LR 0.000063 Time 0.024796 +2023-10-02 22:02:14,503 - Epoch: [193][ 270/ 1236] Overall Loss 0.109219 Objective Loss 0.109219 LR 0.000063 Time 0.024663 +2023-10-02 22:02:14,720 - Epoch: [193][ 280/ 1236] Overall Loss 0.109298 Objective Loss 0.109298 LR 0.000063 Time 0.024558 +2023-10-02 22:02:14,933 - Epoch: [193][ 290/ 1236] Overall Loss 0.109289 Objective Loss 0.109289 LR 0.000063 Time 0.024445 +2023-10-02 22:02:15,152 - Epoch: [193][ 300/ 1236] Overall Loss 0.109106 Objective Loss 0.109106 LR 0.000063 Time 0.024356 +2023-10-02 22:02:15,364 - Epoch: [193][ 310/ 1236] Overall Loss 0.108564 Objective Loss 0.108564 LR 0.000063 Time 0.024256 +2023-10-02 22:02:15,582 - Epoch: [193][ 320/ 1236] Overall Loss 0.108350 Objective Loss 0.108350 LR 0.000063 Time 0.024177 +2023-10-02 22:02:15,795 - Epoch: [193][ 330/ 1236] Overall Loss 0.108347 Objective Loss 0.108347 LR 0.000063 Time 0.024087 +2023-10-02 22:02:16,013 - Epoch: [193][ 340/ 1236] Overall Loss 0.108887 Objective Loss 0.108887 LR 0.000063 Time 0.024019 +2023-10-02 22:02:16,226 - Epoch: [193][ 350/ 1236] Overall Loss 0.108896 Objective Loss 0.108896 LR 0.000063 Time 0.023940 +2023-10-02 22:02:16,444 - Epoch: [193][ 360/ 1236] Overall Loss 0.109283 Objective Loss 0.109283 LR 0.000063 Time 0.023880 +2023-10-02 22:02:16,657 - Epoch: [193][ 370/ 1236] Overall Loss 0.109155 Objective Loss 0.109155 LR 0.000063 Time 0.023810 +2023-10-02 22:02:16,875 - Epoch: [193][ 380/ 1236] Overall Loss 0.109281 Objective Loss 0.109281 LR 0.000063 Time 0.023757 +2023-10-02 22:02:17,089 - Epoch: [193][ 390/ 1236] Overall Loss 0.109190 Objective Loss 0.109190 LR 0.000063 Time 0.023694 +2023-10-02 22:02:17,307 - Epoch: [193][ 400/ 1236] Overall Loss 0.108962 Objective Loss 0.108962 LR 0.000063 Time 0.023647 +2023-10-02 22:02:17,521 - Epoch: [193][ 410/ 1236] Overall Loss 0.108735 Objective Loss 0.108735 LR 0.000063 Time 0.023590 +2023-10-02 22:02:17,733 - Epoch: [193][ 420/ 1236] Overall Loss 0.108525 Objective Loss 0.108525 LR 0.000063 Time 0.023533 +2023-10-02 22:02:17,939 - Epoch: [193][ 430/ 1236] Overall Loss 0.108496 Objective Loss 0.108496 LR 0.000063 Time 0.023465 +2023-10-02 22:02:18,148 - Epoch: [193][ 440/ 1236] Overall Loss 0.108708 Objective Loss 0.108708 LR 0.000063 Time 0.023406 +2023-10-02 22:02:18,354 - Epoch: [193][ 450/ 1236] Overall Loss 0.108669 Objective Loss 0.108669 LR 0.000063 Time 0.023344 +2023-10-02 22:02:18,564 - Epoch: [193][ 460/ 1236] Overall Loss 0.108501 Objective Loss 0.108501 LR 0.000063 Time 0.023291 +2023-10-02 22:02:18,770 - Epoch: [193][ 470/ 1236] Overall Loss 0.108415 Objective Loss 0.108415 LR 0.000063 Time 0.023234 +2023-10-02 22:02:18,979 - Epoch: [193][ 480/ 1236] Overall Loss 0.108161 Objective Loss 0.108161 LR 0.000063 Time 0.023185 +2023-10-02 22:02:19,185 - Epoch: [193][ 490/ 1236] Overall Loss 0.107982 Objective Loss 0.107982 LR 0.000063 Time 0.023132 +2023-10-02 22:02:19,395 - Epoch: [193][ 500/ 1236] Overall Loss 0.107966 Objective Loss 0.107966 LR 0.000063 Time 0.023088 +2023-10-02 22:02:19,602 - Epoch: [193][ 510/ 1236] Overall Loss 0.107960 Objective Loss 0.107960 LR 0.000063 Time 0.023040 +2023-10-02 22:02:19,811 - Epoch: [193][ 520/ 1236] Overall Loss 0.108017 Objective Loss 0.108017 LR 0.000063 Time 0.023000 +2023-10-02 22:02:20,018 - Epoch: [193][ 530/ 1236] Overall Loss 0.107977 Objective Loss 0.107977 LR 0.000063 Time 0.022955 +2023-10-02 22:02:20,226 - Epoch: [193][ 540/ 1236] Overall Loss 0.108196 Objective Loss 0.108196 LR 0.000063 Time 0.022915 +2023-10-02 22:02:20,434 - Epoch: [193][ 550/ 1236] Overall Loss 0.108219 Objective Loss 0.108219 LR 0.000063 Time 0.022873 +2023-10-02 22:02:20,644 - Epoch: [193][ 560/ 1236] Overall Loss 0.108080 Objective Loss 0.108080 LR 0.000063 Time 0.022839 +2023-10-02 22:02:20,850 - Epoch: [193][ 570/ 1236] Overall Loss 0.108096 Objective Loss 0.108096 LR 0.000063 Time 0.022800 +2023-10-02 22:02:21,059 - Epoch: [193][ 580/ 1236] Overall Loss 0.108043 Objective Loss 0.108043 LR 0.000063 Time 0.022766 +2023-10-02 22:02:21,267 - Epoch: [193][ 590/ 1236] Overall Loss 0.107786 Objective Loss 0.107786 LR 0.000063 Time 0.022731 +2023-10-02 22:02:21,476 - Epoch: [193][ 600/ 1236] Overall Loss 0.107571 Objective Loss 0.107571 LR 0.000063 Time 0.022699 +2023-10-02 22:02:21,683 - Epoch: [193][ 610/ 1236] Overall Loss 0.107565 Objective Loss 0.107565 LR 0.000063 Time 0.022663 +2023-10-02 22:02:21,891 - Epoch: [193][ 620/ 1236] Overall Loss 0.107453 Objective Loss 0.107453 LR 0.000063 Time 0.022634 +2023-10-02 22:02:22,100 - Epoch: [193][ 630/ 1236] Overall Loss 0.107591 Objective Loss 0.107591 LR 0.000063 Time 0.022604 +2023-10-02 22:02:22,308 - Epoch: [193][ 640/ 1236] Overall Loss 0.107542 Objective Loss 0.107542 LR 0.000063 Time 0.022576 +2023-10-02 22:02:22,515 - Epoch: [193][ 650/ 1236] Overall Loss 0.107430 Objective Loss 0.107430 LR 0.000063 Time 0.022544 +2023-10-02 22:02:22,724 - Epoch: [193][ 660/ 1236] Overall Loss 0.107681 Objective Loss 0.107681 LR 0.000063 Time 0.022519 +2023-10-02 22:02:22,931 - Epoch: [193][ 670/ 1236] Overall Loss 0.107604 Objective Loss 0.107604 LR 0.000063 Time 0.022489 +2023-10-02 22:02:23,139 - Epoch: [193][ 680/ 1236] Overall Loss 0.107502 Objective Loss 0.107502 LR 0.000063 Time 0.022465 +2023-10-02 22:02:23,346 - Epoch: [193][ 690/ 1236] Overall Loss 0.107304 Objective Loss 0.107304 LR 0.000063 Time 0.022437 +2023-10-02 22:02:23,555 - Epoch: [193][ 700/ 1236] Overall Loss 0.107143 Objective Loss 0.107143 LR 0.000063 Time 0.022414 +2023-10-02 22:02:23,764 - Epoch: [193][ 710/ 1236] Overall Loss 0.107059 Objective Loss 0.107059 LR 0.000063 Time 0.022390 +2023-10-02 22:02:23,972 - Epoch: [193][ 720/ 1236] Overall Loss 0.106946 Objective Loss 0.106946 LR 0.000063 Time 0.022368 +2023-10-02 22:02:24,180 - Epoch: [193][ 730/ 1236] Overall Loss 0.106984 Objective Loss 0.106984 LR 0.000063 Time 0.022344 +2023-10-02 22:02:24,389 - Epoch: [193][ 740/ 1236] Overall Loss 0.106831 Objective Loss 0.106831 LR 0.000063 Time 0.022324 +2023-10-02 22:02:24,597 - Epoch: [193][ 750/ 1236] Overall Loss 0.106971 Objective Loss 0.106971 LR 0.000063 Time 0.022302 +2023-10-02 22:02:24,807 - Epoch: [193][ 760/ 1236] Overall Loss 0.107136 Objective Loss 0.107136 LR 0.000063 Time 0.022284 +2023-10-02 22:02:25,017 - Epoch: [193][ 770/ 1236] Overall Loss 0.106945 Objective Loss 0.106945 LR 0.000063 Time 0.022265 +2023-10-02 22:02:25,225 - Epoch: [193][ 780/ 1236] Overall Loss 0.106920 Objective Loss 0.106920 LR 0.000063 Time 0.022246 +2023-10-02 22:02:25,435 - Epoch: [193][ 790/ 1236] Overall Loss 0.106982 Objective Loss 0.106982 LR 0.000063 Time 0.022230 +2023-10-02 22:02:25,644 - Epoch: [193][ 800/ 1236] Overall Loss 0.107104 Objective Loss 0.107104 LR 0.000063 Time 0.022213 +2023-10-02 22:02:25,854 - Epoch: [193][ 810/ 1236] Overall Loss 0.107015 Objective Loss 0.107015 LR 0.000063 Time 0.022196 +2023-10-02 22:02:26,065 - Epoch: [193][ 820/ 1236] Overall Loss 0.106974 Objective Loss 0.106974 LR 0.000063 Time 0.022182 +2023-10-02 22:02:26,275 - Epoch: [193][ 830/ 1236] Overall Loss 0.106995 Objective Loss 0.106995 LR 0.000063 Time 0.022166 +2023-10-02 22:02:26,486 - Epoch: [193][ 840/ 1236] Overall Loss 0.106891 Objective Loss 0.106891 LR 0.000063 Time 0.022152 +2023-10-02 22:02:26,696 - Epoch: [193][ 850/ 1236] Overall Loss 0.107046 Objective Loss 0.107046 LR 0.000063 Time 0.022137 +2023-10-02 22:02:26,907 - Epoch: [193][ 860/ 1236] Overall Loss 0.106930 Objective Loss 0.106930 LR 0.000063 Time 0.022124 +2023-10-02 22:02:27,118 - Epoch: [193][ 870/ 1236] Overall Loss 0.106935 Objective Loss 0.106935 LR 0.000063 Time 0.022110 +2023-10-02 22:02:27,329 - Epoch: [193][ 880/ 1236] Overall Loss 0.106936 Objective Loss 0.106936 LR 0.000063 Time 0.022099 +2023-10-02 22:02:27,538 - Epoch: [193][ 890/ 1236] Overall Loss 0.107047 Objective Loss 0.107047 LR 0.000063 Time 0.022084 +2023-10-02 22:02:27,749 - Epoch: [193][ 900/ 1236] Overall Loss 0.107414 Objective Loss 0.107414 LR 0.000063 Time 0.022071 +2023-10-02 22:02:27,959 - Epoch: [193][ 910/ 1236] Overall Loss 0.107326 Objective Loss 0.107326 LR 0.000063 Time 0.022058 +2023-10-02 22:02:28,170 - Epoch: [193][ 920/ 1236] Overall Loss 0.107502 Objective Loss 0.107502 LR 0.000063 Time 0.022048 +2023-10-02 22:02:28,380 - Epoch: [193][ 930/ 1236] Overall Loss 0.107549 Objective Loss 0.107549 LR 0.000063 Time 0.022035 +2023-10-02 22:02:28,591 - Epoch: [193][ 940/ 1236] Overall Loss 0.107517 Objective Loss 0.107517 LR 0.000063 Time 0.022024 +2023-10-02 22:02:28,802 - Epoch: [193][ 950/ 1236] Overall Loss 0.107542 Objective Loss 0.107542 LR 0.000063 Time 0.022012 +2023-10-02 22:02:29,013 - Epoch: [193][ 960/ 1236] Overall Loss 0.107735 Objective Loss 0.107735 LR 0.000063 Time 0.022002 +2023-10-02 22:02:29,224 - Epoch: [193][ 970/ 1236] Overall Loss 0.107603 Objective Loss 0.107603 LR 0.000063 Time 0.021993 +2023-10-02 22:02:29,435 - Epoch: [193][ 980/ 1236] Overall Loss 0.107588 Objective Loss 0.107588 LR 0.000063 Time 0.021983 +2023-10-02 22:02:29,646 - Epoch: [193][ 990/ 1236] Overall Loss 0.107755 Objective Loss 0.107755 LR 0.000063 Time 0.021973 +2023-10-02 22:02:29,856 - Epoch: [193][ 1000/ 1236] Overall Loss 0.107590 Objective Loss 0.107590 LR 0.000063 Time 0.021964 +2023-10-02 22:02:30,067 - Epoch: [193][ 1010/ 1236] Overall Loss 0.107554 Objective Loss 0.107554 LR 0.000063 Time 0.021955 +2023-10-02 22:02:30,277 - Epoch: [193][ 1020/ 1236] Overall Loss 0.107375 Objective Loss 0.107375 LR 0.000063 Time 0.021945 +2023-10-02 22:02:30,488 - Epoch: [193][ 1030/ 1236] Overall Loss 0.107274 Objective Loss 0.107274 LR 0.000063 Time 0.021935 +2023-10-02 22:02:30,698 - Epoch: [193][ 1040/ 1236] Overall Loss 0.107219 Objective Loss 0.107219 LR 0.000063 Time 0.021926 +2023-10-02 22:02:30,909 - Epoch: [193][ 1050/ 1236] Overall Loss 0.107251 Objective Loss 0.107251 LR 0.000063 Time 0.021917 +2023-10-02 22:02:31,120 - Epoch: [193][ 1060/ 1236] Overall Loss 0.107250 Objective Loss 0.107250 LR 0.000063 Time 0.021909 +2023-10-02 22:02:31,331 - Epoch: [193][ 1070/ 1236] Overall Loss 0.107257 Objective Loss 0.107257 LR 0.000063 Time 0.021901 +2023-10-02 22:02:31,541 - Epoch: [193][ 1080/ 1236] Overall Loss 0.107371 Objective Loss 0.107371 LR 0.000063 Time 0.021893 +2023-10-02 22:02:31,752 - Epoch: [193][ 1090/ 1236] Overall Loss 0.107465 Objective Loss 0.107465 LR 0.000063 Time 0.021885 +2023-10-02 22:02:31,963 - Epoch: [193][ 1100/ 1236] Overall Loss 0.107467 Objective Loss 0.107467 LR 0.000063 Time 0.021878 +2023-10-02 22:02:32,175 - Epoch: [193][ 1110/ 1236] Overall Loss 0.107449 Objective Loss 0.107449 LR 0.000063 Time 0.021871 +2023-10-02 22:02:32,385 - Epoch: [193][ 1120/ 1236] Overall Loss 0.107465 Objective Loss 0.107465 LR 0.000063 Time 0.021863 +2023-10-02 22:02:32,596 - Epoch: [193][ 1130/ 1236] Overall Loss 0.107513 Objective Loss 0.107513 LR 0.000063 Time 0.021856 +2023-10-02 22:02:32,807 - Epoch: [193][ 1140/ 1236] Overall Loss 0.107481 Objective Loss 0.107481 LR 0.000063 Time 0.021849 +2023-10-02 22:02:33,018 - Epoch: [193][ 1150/ 1236] Overall Loss 0.107385 Objective Loss 0.107385 LR 0.000063 Time 0.021841 +2023-10-02 22:02:33,230 - Epoch: [193][ 1160/ 1236] Overall Loss 0.107380 Objective Loss 0.107380 LR 0.000063 Time 0.021835 +2023-10-02 22:02:33,441 - Epoch: [193][ 1170/ 1236] Overall Loss 0.107335 Objective Loss 0.107335 LR 0.000063 Time 0.021829 +2023-10-02 22:02:33,652 - Epoch: [193][ 1180/ 1236] Overall Loss 0.107250 Objective Loss 0.107250 LR 0.000063 Time 0.021822 +2023-10-02 22:02:33,863 - Epoch: [193][ 1190/ 1236] Overall Loss 0.107393 Objective Loss 0.107393 LR 0.000063 Time 0.021816 +2023-10-02 22:02:34,074 - Epoch: [193][ 1200/ 1236] Overall Loss 0.107270 Objective Loss 0.107270 LR 0.000063 Time 0.021809 +2023-10-02 22:02:34,285 - Epoch: [193][ 1210/ 1236] Overall Loss 0.107209 Objective Loss 0.107209 LR 0.000063 Time 0.021803 +2023-10-02 22:02:34,496 - Epoch: [193][ 1220/ 1236] Overall Loss 0.107174 Objective Loss 0.107174 LR 0.000063 Time 0.021797 +2023-10-02 22:02:34,757 - Epoch: [193][ 1230/ 1236] Overall Loss 0.107175 Objective Loss 0.107175 LR 0.000063 Time 0.021832 +2023-10-02 22:02:34,879 - Epoch: [193][ 1236/ 1236] Overall Loss 0.107216 Objective Loss 0.107216 Top1 92.260692 Top5 99.592668 LR 0.000063 Time 0.021824 +2023-10-02 22:02:35,022 - --- validate (epoch=193)----------- +2023-10-02 22:02:35,022 - 29943 samples (256 per mini-batch) +2023-10-02 22:02:35,507 - Epoch: [193][ 10/ 117] Loss 0.299043 Top1 87.773438 Top5 98.945312 +2023-10-02 22:02:35,659 - Epoch: [193][ 20/ 117] Loss 0.298807 Top1 87.753906 Top5 98.769531 +2023-10-02 22:02:35,811 - Epoch: [193][ 30/ 117] Loss 0.298808 Top1 87.669271 Top5 98.893229 +2023-10-02 22:02:35,965 - Epoch: [193][ 40/ 117] Loss 0.303119 Top1 87.539062 Top5 98.906250 +2023-10-02 22:02:36,117 - Epoch: [193][ 50/ 117] Loss 0.305107 Top1 87.476562 Top5 98.875000 +2023-10-02 22:02:36,270 - Epoch: [193][ 60/ 117] Loss 0.309695 Top1 87.421875 Top5 98.828125 +2023-10-02 22:02:36,423 - Epoch: [193][ 70/ 117] Loss 0.310092 Top1 87.488839 Top5 98.772321 +2023-10-02 22:02:36,585 - Epoch: [193][ 80/ 117] Loss 0.309768 Top1 87.597656 Top5 98.740234 +2023-10-02 22:02:36,742 - Epoch: [193][ 90/ 117] Loss 0.312698 Top1 87.651910 Top5 98.697917 +2023-10-02 22:02:36,902 - Epoch: [193][ 100/ 117] Loss 0.310278 Top1 87.773438 Top5 98.734375 +2023-10-02 22:02:37,068 - Epoch: [193][ 110/ 117] Loss 0.307972 Top1 87.695312 Top5 98.767756 +2023-10-02 22:02:37,156 - Epoch: [193][ 117/ 117] Loss 0.307170 Top1 87.686605 Top5 98.750960 +2023-10-02 22:02:37,304 - ==> Top1: 87.687 Top5: 98.751 Loss: 0.307 + +2023-10-02 22:02:37,305 - ==> Confusion: +[[ 938 0 2 2 5 2 0 0 8 59 1 0 1 2 6 0 2 0 1 0 21] + [ 0 1054 1 1 3 21 0 23 1 1 1 0 0 0 0 3 1 0 9 3 9] + [ 1 1 985 7 1 0 15 8 0 2 1 1 6 2 1 3 2 1 9 2 8] + [ 2 3 14 980 0 2 0 1 5 1 6 1 6 3 24 2 1 4 13 1 20] + [ 25 3 0 0 970 4 0 0 0 14 0 0 0 3 13 5 8 0 0 1 4] + [ 3 25 0 2 5 1001 1 21 1 4 2 6 0 11 5 0 2 0 3 2 22] + [ 0 3 25 0 0 1 1133 6 0 0 3 1 0 0 0 4 0 1 1 7 6] + [ 1 8 12 1 5 19 7 1084 1 5 6 5 2 4 1 1 0 1 35 10 10] + [ 14 0 0 0 2 2 0 2 990 31 12 0 1 11 12 0 3 1 4 1 3] + [ 96 0 2 0 6 3 0 0 32 942 0 1 0 20 7 2 1 0 0 0 7] + [ 3 2 8 6 0 2 2 3 9 3 975 3 0 11 5 0 0 2 4 3 12] + [ 0 0 1 0 1 13 0 7 0 0 0 964 13 5 0 2 1 17 0 4 7] + [ 0 0 1 2 1 0 2 1 0 1 2 29 972 2 2 8 1 16 3 6 19] + [ 0 0 1 0 4 5 0 0 14 11 2 7 0 1049 4 0 0 0 0 1 21] + [ 13 0 3 13 2 1 0 0 20 1 1 0 3 2 1020 0 0 2 10 0 10] + [ 0 0 1 1 6 0 1 0 0 0 1 4 6 0 0 1073 17 9 2 8 5] + [ 0 16 1 0 3 4 2 1 0 0 0 4 0 2 4 7 1100 0 1 5 11] + [ 0 0 0 1 0 0 3 0 0 0 0 2 18 2 2 7 1 997 0 1 4] + [ 3 3 2 14 0 0 0 18 5 1 2 1 1 0 8 0 0 1 997 1 11] + [ 0 1 3 2 1 3 8 6 0 0 0 12 5 1 2 1 7 2 1 1087 10] + [ 104 90 110 66 52 103 28 85 77 59 137 83 259 213 111 47 64 43 103 126 5945]] + +2023-10-02 22:02:37,306 - ==> Best [Top1: 87.874 Top5: 98.708 Sparsity:0.00 Params: 169472 on epoch: 189] +2023-10-02 22:02:37,306 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:02:37,312 - + +2023-10-02 22:02:37,312 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:02:38,341 - Epoch: [194][ 10/ 1236] Overall Loss 0.098408 Objective Loss 0.098408 LR 0.000063 Time 0.102792 +2023-10-02 22:02:38,547 - Epoch: [194][ 20/ 1236] Overall Loss 0.101957 Objective Loss 0.101957 LR 0.000063 Time 0.061701 +2023-10-02 22:02:38,754 - Epoch: [194][ 30/ 1236] Overall Loss 0.104749 Objective Loss 0.104749 LR 0.000063 Time 0.048006 +2023-10-02 22:02:38,959 - Epoch: [194][ 40/ 1236] Overall Loss 0.100898 Objective Loss 0.100898 LR 0.000063 Time 0.041134 +2023-10-02 22:02:39,166 - Epoch: [194][ 50/ 1236] Overall Loss 0.098442 Objective Loss 0.098442 LR 0.000063 Time 0.037027 +2023-10-02 22:02:39,374 - Epoch: [194][ 60/ 1236] Overall Loss 0.100308 Objective Loss 0.100308 LR 0.000063 Time 0.034325 +2023-10-02 22:02:39,582 - Epoch: [194][ 70/ 1236] Overall Loss 0.101998 Objective Loss 0.101998 LR 0.000063 Time 0.032370 +2023-10-02 22:02:39,788 - Epoch: [194][ 80/ 1236] Overall Loss 0.101617 Objective Loss 0.101617 LR 0.000063 Time 0.030902 +2023-10-02 22:02:39,996 - Epoch: [194][ 90/ 1236] Overall Loss 0.103192 Objective Loss 0.103192 LR 0.000063 Time 0.029763 +2023-10-02 22:02:40,205 - Epoch: [194][ 100/ 1236] Overall Loss 0.103102 Objective Loss 0.103102 LR 0.000063 Time 0.028874 +2023-10-02 22:02:40,412 - Epoch: [194][ 110/ 1236] Overall Loss 0.103403 Objective Loss 0.103403 LR 0.000063 Time 0.028117 +2023-10-02 22:02:40,620 - Epoch: [194][ 120/ 1236] Overall Loss 0.104348 Objective Loss 0.104348 LR 0.000063 Time 0.027501 +2023-10-02 22:02:40,827 - Epoch: [194][ 130/ 1236] Overall Loss 0.105838 Objective Loss 0.105838 LR 0.000063 Time 0.026967 +2023-10-02 22:02:41,033 - Epoch: [194][ 140/ 1236] Overall Loss 0.104733 Objective Loss 0.104733 LR 0.000063 Time 0.026511 +2023-10-02 22:02:41,240 - Epoch: [194][ 150/ 1236] Overall Loss 0.104961 Objective Loss 0.104961 LR 0.000063 Time 0.026116 +2023-10-02 22:02:41,449 - Epoch: [194][ 160/ 1236] Overall Loss 0.105281 Objective Loss 0.105281 LR 0.000063 Time 0.025785 +2023-10-02 22:02:41,656 - Epoch: [194][ 170/ 1236] Overall Loss 0.107378 Objective Loss 0.107378 LR 0.000063 Time 0.025480 +2023-10-02 22:02:41,864 - Epoch: [194][ 180/ 1236] Overall Loss 0.107717 Objective Loss 0.107717 LR 0.000063 Time 0.025217 +2023-10-02 22:02:42,070 - Epoch: [194][ 190/ 1236] Overall Loss 0.107438 Objective Loss 0.107438 LR 0.000063 Time 0.024971 +2023-10-02 22:02:42,277 - Epoch: [194][ 200/ 1236] Overall Loss 0.107896 Objective Loss 0.107896 LR 0.000063 Time 0.024758 +2023-10-02 22:02:42,484 - Epoch: [194][ 210/ 1236] Overall Loss 0.107485 Objective Loss 0.107485 LR 0.000063 Time 0.024555 +2023-10-02 22:02:42,692 - Epoch: [194][ 220/ 1236] Overall Loss 0.107504 Objective Loss 0.107504 LR 0.000063 Time 0.024384 +2023-10-02 22:02:42,899 - Epoch: [194][ 230/ 1236] Overall Loss 0.107771 Objective Loss 0.107771 LR 0.000063 Time 0.024219 +2023-10-02 22:02:43,107 - Epoch: [194][ 240/ 1236] Overall Loss 0.106923 Objective Loss 0.106923 LR 0.000063 Time 0.024073 +2023-10-02 22:02:43,314 - Epoch: [194][ 250/ 1236] Overall Loss 0.105702 Objective Loss 0.105702 LR 0.000063 Time 0.023937 +2023-10-02 22:02:43,521 - Epoch: [194][ 260/ 1236] Overall Loss 0.105560 Objective Loss 0.105560 LR 0.000063 Time 0.023813 +2023-10-02 22:02:43,727 - Epoch: [194][ 270/ 1236] Overall Loss 0.105197 Objective Loss 0.105197 LR 0.000063 Time 0.023688 +2023-10-02 22:02:43,933 - Epoch: [194][ 280/ 1236] Overall Loss 0.105526 Objective Loss 0.105526 LR 0.000063 Time 0.023577 +2023-10-02 22:02:44,139 - Epoch: [194][ 290/ 1236] Overall Loss 0.105749 Objective Loss 0.105749 LR 0.000063 Time 0.023471 +2023-10-02 22:02:44,347 - Epoch: [194][ 300/ 1236] Overall Loss 0.105784 Objective Loss 0.105784 LR 0.000063 Time 0.023380 +2023-10-02 22:02:44,555 - Epoch: [194][ 310/ 1236] Overall Loss 0.106263 Objective Loss 0.106263 LR 0.000063 Time 0.023293 +2023-10-02 22:02:44,761 - Epoch: [194][ 320/ 1236] Overall Loss 0.106224 Objective Loss 0.106224 LR 0.000063 Time 0.023209 +2023-10-02 22:02:44,967 - Epoch: [194][ 330/ 1236] Overall Loss 0.105782 Objective Loss 0.105782 LR 0.000063 Time 0.023125 +2023-10-02 22:02:45,174 - Epoch: [194][ 340/ 1236] Overall Loss 0.105885 Objective Loss 0.105885 LR 0.000063 Time 0.023055 +2023-10-02 22:02:45,381 - Epoch: [194][ 350/ 1236] Overall Loss 0.106042 Objective Loss 0.106042 LR 0.000063 Time 0.022984 +2023-10-02 22:02:45,588 - Epoch: [194][ 360/ 1236] Overall Loss 0.106263 Objective Loss 0.106263 LR 0.000063 Time 0.022918 +2023-10-02 22:02:45,795 - Epoch: [194][ 370/ 1236] Overall Loss 0.106302 Objective Loss 0.106302 LR 0.000063 Time 0.022855 +2023-10-02 22:02:46,004 - Epoch: [194][ 380/ 1236] Overall Loss 0.106884 Objective Loss 0.106884 LR 0.000063 Time 0.022803 +2023-10-02 22:02:46,213 - Epoch: [194][ 390/ 1236] Overall Loss 0.106955 Objective Loss 0.106955 LR 0.000063 Time 0.022750 +2023-10-02 22:02:46,421 - Epoch: [194][ 400/ 1236] Overall Loss 0.107118 Objective Loss 0.107118 LR 0.000063 Time 0.022701 +2023-10-02 22:02:46,632 - Epoch: [194][ 410/ 1236] Overall Loss 0.107010 Objective Loss 0.107010 LR 0.000063 Time 0.022656 +2023-10-02 22:02:46,842 - Epoch: [194][ 420/ 1236] Overall Loss 0.107101 Objective Loss 0.107101 LR 0.000063 Time 0.022618 +2023-10-02 22:02:47,054 - Epoch: [194][ 430/ 1236] Overall Loss 0.107298 Objective Loss 0.107298 LR 0.000063 Time 0.022581 +2023-10-02 22:02:47,265 - Epoch: [194][ 440/ 1236] Overall Loss 0.107211 Objective Loss 0.107211 LR 0.000063 Time 0.022547 +2023-10-02 22:02:47,476 - Epoch: [194][ 450/ 1236] Overall Loss 0.106940 Objective Loss 0.106940 LR 0.000063 Time 0.022511 +2023-10-02 22:02:47,687 - Epoch: [194][ 460/ 1236] Overall Loss 0.106867 Objective Loss 0.106867 LR 0.000063 Time 0.022479 +2023-10-02 22:02:47,894 - Epoch: [194][ 470/ 1236] Overall Loss 0.107311 Objective Loss 0.107311 LR 0.000063 Time 0.022439 +2023-10-02 22:02:48,102 - Epoch: [194][ 480/ 1236] Overall Loss 0.107308 Objective Loss 0.107308 LR 0.000063 Time 0.022405 +2023-10-02 22:02:48,311 - Epoch: [194][ 490/ 1236] Overall Loss 0.107100 Objective Loss 0.107100 LR 0.000063 Time 0.022371 +2023-10-02 22:02:48,521 - Epoch: [194][ 500/ 1236] Overall Loss 0.106933 Objective Loss 0.106933 LR 0.000063 Time 0.022342 +2023-10-02 22:02:48,730 - Epoch: [194][ 510/ 1236] Overall Loss 0.107039 Objective Loss 0.107039 LR 0.000063 Time 0.022311 +2023-10-02 22:02:48,939 - Epoch: [194][ 520/ 1236] Overall Loss 0.107058 Objective Loss 0.107058 LR 0.000063 Time 0.022284 +2023-10-02 22:02:49,147 - Epoch: [194][ 530/ 1236] Overall Loss 0.107209 Objective Loss 0.107209 LR 0.000063 Time 0.022253 +2023-10-02 22:02:49,355 - Epoch: [194][ 540/ 1236] Overall Loss 0.107146 Objective Loss 0.107146 LR 0.000063 Time 0.022225 +2023-10-02 22:02:49,561 - Epoch: [194][ 550/ 1236] Overall Loss 0.107244 Objective Loss 0.107244 LR 0.000063 Time 0.022192 +2023-10-02 22:02:49,768 - Epoch: [194][ 560/ 1236] Overall Loss 0.107243 Objective Loss 0.107243 LR 0.000063 Time 0.022166 +2023-10-02 22:02:49,977 - Epoch: [194][ 570/ 1236] Overall Loss 0.107092 Objective Loss 0.107092 LR 0.000063 Time 0.022141 +2023-10-02 22:02:50,186 - Epoch: [194][ 580/ 1236] Overall Loss 0.107045 Objective Loss 0.107045 LR 0.000063 Time 0.022119 +2023-10-02 22:02:50,395 - Epoch: [194][ 590/ 1236] Overall Loss 0.106824 Objective Loss 0.106824 LR 0.000063 Time 0.022096 +2023-10-02 22:02:50,605 - Epoch: [194][ 600/ 1236] Overall Loss 0.106657 Objective Loss 0.106657 LR 0.000063 Time 0.022077 +2023-10-02 22:02:50,817 - Epoch: [194][ 610/ 1236] Overall Loss 0.106810 Objective Loss 0.106810 LR 0.000063 Time 0.022061 +2023-10-02 22:02:51,028 - Epoch: [194][ 620/ 1236] Overall Loss 0.106524 Objective Loss 0.106524 LR 0.000063 Time 0.022044 +2023-10-02 22:02:51,240 - Epoch: [194][ 630/ 1236] Overall Loss 0.106494 Objective Loss 0.106494 LR 0.000063 Time 0.022029 +2023-10-02 22:02:51,451 - Epoch: [194][ 640/ 1236] Overall Loss 0.106476 Objective Loss 0.106476 LR 0.000063 Time 0.022013 +2023-10-02 22:02:51,663 - Epoch: [194][ 650/ 1236] Overall Loss 0.106360 Objective Loss 0.106360 LR 0.000063 Time 0.021998 +2023-10-02 22:02:51,873 - Epoch: [194][ 660/ 1236] Overall Loss 0.106355 Objective Loss 0.106355 LR 0.000063 Time 0.021984 +2023-10-02 22:02:52,085 - Epoch: [194][ 670/ 1236] Overall Loss 0.106386 Objective Loss 0.106386 LR 0.000063 Time 0.021970 +2023-10-02 22:02:52,293 - Epoch: [194][ 680/ 1236] Overall Loss 0.106314 Objective Loss 0.106314 LR 0.000063 Time 0.021952 +2023-10-02 22:02:52,501 - Epoch: [194][ 690/ 1236] Overall Loss 0.106302 Objective Loss 0.106302 LR 0.000063 Time 0.021933 +2023-10-02 22:02:52,710 - Epoch: [194][ 700/ 1236] Overall Loss 0.106104 Objective Loss 0.106104 LR 0.000063 Time 0.021918 +2023-10-02 22:02:52,919 - Epoch: [194][ 710/ 1236] Overall Loss 0.106165 Objective Loss 0.106165 LR 0.000063 Time 0.021901 +2023-10-02 22:02:53,129 - Epoch: [194][ 720/ 1236] Overall Loss 0.106368 Objective Loss 0.106368 LR 0.000063 Time 0.021888 +2023-10-02 22:02:53,341 - Epoch: [194][ 730/ 1236] Overall Loss 0.106551 Objective Loss 0.106551 LR 0.000063 Time 0.021877 +2023-10-02 22:02:53,550 - Epoch: [194][ 740/ 1236] Overall Loss 0.106528 Objective Loss 0.106528 LR 0.000063 Time 0.021864 +2023-10-02 22:02:53,762 - Epoch: [194][ 750/ 1236] Overall Loss 0.106510 Objective Loss 0.106510 LR 0.000063 Time 0.021853 +2023-10-02 22:02:53,973 - Epoch: [194][ 760/ 1236] Overall Loss 0.106655 Objective Loss 0.106655 LR 0.000063 Time 0.021842 +2023-10-02 22:02:54,185 - Epoch: [194][ 770/ 1236] Overall Loss 0.106485 Objective Loss 0.106485 LR 0.000063 Time 0.021832 +2023-10-02 22:02:54,396 - Epoch: [194][ 780/ 1236] Overall Loss 0.106622 Objective Loss 0.106622 LR 0.000063 Time 0.021822 +2023-10-02 22:02:54,608 - Epoch: [194][ 790/ 1236] Overall Loss 0.106463 Objective Loss 0.106463 LR 0.000063 Time 0.021812 +2023-10-02 22:02:54,819 - Epoch: [194][ 800/ 1236] Overall Loss 0.106528 Objective Loss 0.106528 LR 0.000063 Time 0.021803 +2023-10-02 22:02:55,031 - Epoch: [194][ 810/ 1236] Overall Loss 0.106511 Objective Loss 0.106511 LR 0.000063 Time 0.021794 +2023-10-02 22:02:55,242 - Epoch: [194][ 820/ 1236] Overall Loss 0.106537 Objective Loss 0.106537 LR 0.000063 Time 0.021785 +2023-10-02 22:02:55,454 - Epoch: [194][ 830/ 1236] Overall Loss 0.106458 Objective Loss 0.106458 LR 0.000063 Time 0.021776 +2023-10-02 22:02:55,664 - Epoch: [194][ 840/ 1236] Overall Loss 0.106443 Objective Loss 0.106443 LR 0.000063 Time 0.021767 +2023-10-02 22:02:55,876 - Epoch: [194][ 850/ 1236] Overall Loss 0.106441 Objective Loss 0.106441 LR 0.000063 Time 0.021759 +2023-10-02 22:02:56,087 - Epoch: [194][ 860/ 1236] Overall Loss 0.106316 Objective Loss 0.106316 LR 0.000063 Time 0.021750 +2023-10-02 22:02:56,299 - Epoch: [194][ 870/ 1236] Overall Loss 0.106569 Objective Loss 0.106569 LR 0.000063 Time 0.021742 +2023-10-02 22:02:56,510 - Epoch: [194][ 880/ 1236] Overall Loss 0.106642 Objective Loss 0.106642 LR 0.000063 Time 0.021734 +2023-10-02 22:02:56,722 - Epoch: [194][ 890/ 1236] Overall Loss 0.106694 Objective Loss 0.106694 LR 0.000063 Time 0.021727 +2023-10-02 22:02:56,933 - Epoch: [194][ 900/ 1236] Overall Loss 0.106586 Objective Loss 0.106586 LR 0.000063 Time 0.021719 +2023-10-02 22:02:57,145 - Epoch: [194][ 910/ 1236] Overall Loss 0.106425 Objective Loss 0.106425 LR 0.000063 Time 0.021712 +2023-10-02 22:02:57,355 - Epoch: [194][ 920/ 1236] Overall Loss 0.106473 Objective Loss 0.106473 LR 0.000063 Time 0.021705 +2023-10-02 22:02:57,567 - Epoch: [194][ 930/ 1236] Overall Loss 0.106361 Objective Loss 0.106361 LR 0.000063 Time 0.021698 +2023-10-02 22:02:57,778 - Epoch: [194][ 940/ 1236] Overall Loss 0.106560 Objective Loss 0.106560 LR 0.000063 Time 0.021691 +2023-10-02 22:02:57,989 - Epoch: [194][ 950/ 1236] Overall Loss 0.106528 Objective Loss 0.106528 LR 0.000063 Time 0.021682 +2023-10-02 22:02:58,196 - Epoch: [194][ 960/ 1236] Overall Loss 0.106413 Objective Loss 0.106413 LR 0.000063 Time 0.021673 +2023-10-02 22:02:58,404 - Epoch: [194][ 970/ 1236] Overall Loss 0.106541 Objective Loss 0.106541 LR 0.000063 Time 0.021662 +2023-10-02 22:02:58,611 - Epoch: [194][ 980/ 1236] Overall Loss 0.106511 Objective Loss 0.106511 LR 0.000063 Time 0.021652 +2023-10-02 22:02:58,818 - Epoch: [194][ 990/ 1236] Overall Loss 0.106427 Objective Loss 0.106427 LR 0.000063 Time 0.021642 +2023-10-02 22:02:59,025 - Epoch: [194][ 1000/ 1236] Overall Loss 0.106386 Objective Loss 0.106386 LR 0.000063 Time 0.021632 +2023-10-02 22:02:59,231 - Epoch: [194][ 1010/ 1236] Overall Loss 0.106362 Objective Loss 0.106362 LR 0.000063 Time 0.021620 +2023-10-02 22:02:59,439 - Epoch: [194][ 1020/ 1236] Overall Loss 0.106139 Objective Loss 0.106139 LR 0.000063 Time 0.021612 +2023-10-02 22:02:59,647 - Epoch: [194][ 1030/ 1236] Overall Loss 0.106022 Objective Loss 0.106022 LR 0.000063 Time 0.021604 +2023-10-02 22:02:59,855 - Epoch: [194][ 1040/ 1236] Overall Loss 0.106154 Objective Loss 0.106154 LR 0.000063 Time 0.021596 +2023-10-02 22:03:00,064 - Epoch: [194][ 1050/ 1236] Overall Loss 0.106071 Objective Loss 0.106071 LR 0.000063 Time 0.021588 +2023-10-02 22:03:00,272 - Epoch: [194][ 1060/ 1236] Overall Loss 0.106086 Objective Loss 0.106086 LR 0.000063 Time 0.021581 +2023-10-02 22:03:00,480 - Epoch: [194][ 1070/ 1236] Overall Loss 0.106179 Objective Loss 0.106179 LR 0.000063 Time 0.021573 +2023-10-02 22:03:00,689 - Epoch: [194][ 1080/ 1236] Overall Loss 0.106262 Objective Loss 0.106262 LR 0.000063 Time 0.021566 +2023-10-02 22:03:00,897 - Epoch: [194][ 1090/ 1236] Overall Loss 0.106448 Objective Loss 0.106448 LR 0.000063 Time 0.021558 +2023-10-02 22:03:01,105 - Epoch: [194][ 1100/ 1236] Overall Loss 0.106482 Objective Loss 0.106482 LR 0.000063 Time 0.021551 +2023-10-02 22:03:01,314 - Epoch: [194][ 1110/ 1236] Overall Loss 0.106627 Objective Loss 0.106627 LR 0.000063 Time 0.021543 +2023-10-02 22:03:01,522 - Epoch: [194][ 1120/ 1236] Overall Loss 0.106585 Objective Loss 0.106585 LR 0.000063 Time 0.021537 +2023-10-02 22:03:01,731 - Epoch: [194][ 1130/ 1236] Overall Loss 0.106687 Objective Loss 0.106687 LR 0.000063 Time 0.021529 +2023-10-02 22:03:01,938 - Epoch: [194][ 1140/ 1236] Overall Loss 0.106741 Objective Loss 0.106741 LR 0.000063 Time 0.021522 +2023-10-02 22:03:02,149 - Epoch: [194][ 1150/ 1236] Overall Loss 0.106684 Objective Loss 0.106684 LR 0.000063 Time 0.021518 +2023-10-02 22:03:02,359 - Epoch: [194][ 1160/ 1236] Overall Loss 0.106741 Objective Loss 0.106741 LR 0.000063 Time 0.021513 +2023-10-02 22:03:02,571 - Epoch: [194][ 1170/ 1236] Overall Loss 0.106850 Objective Loss 0.106850 LR 0.000063 Time 0.021510 +2023-10-02 22:03:02,781 - Epoch: [194][ 1180/ 1236] Overall Loss 0.106832 Objective Loss 0.106832 LR 0.000063 Time 0.021505 +2023-10-02 22:03:02,992 - Epoch: [194][ 1190/ 1236] Overall Loss 0.106798 Objective Loss 0.106798 LR 0.000063 Time 0.021502 +2023-10-02 22:03:03,202 - Epoch: [194][ 1200/ 1236] Overall Loss 0.106768 Objective Loss 0.106768 LR 0.000063 Time 0.021497 +2023-10-02 22:03:03,413 - Epoch: [194][ 1210/ 1236] Overall Loss 0.106740 Objective Loss 0.106740 LR 0.000063 Time 0.021493 +2023-10-02 22:03:03,623 - Epoch: [194][ 1220/ 1236] Overall Loss 0.106704 Objective Loss 0.106704 LR 0.000063 Time 0.021488 +2023-10-02 22:03:03,885 - Epoch: [194][ 1230/ 1236] Overall Loss 0.106609 Objective Loss 0.106609 LR 0.000063 Time 0.021526 +2023-10-02 22:03:04,006 - Epoch: [194][ 1236/ 1236] Overall Loss 0.106560 Objective Loss 0.106560 Top1 93.686354 Top5 99.185336 LR 0.000063 Time 0.021519 +2023-10-02 22:03:04,139 - --- validate (epoch=194)----------- +2023-10-02 22:03:04,139 - 29943 samples (256 per mini-batch) +2023-10-02 22:03:04,613 - Epoch: [194][ 10/ 117] Loss 0.317486 Top1 86.835938 Top5 98.632812 +2023-10-02 22:03:04,769 - Epoch: [194][ 20/ 117] Loss 0.313526 Top1 87.597656 Top5 98.515625 +2023-10-02 22:03:04,924 - Epoch: [194][ 30/ 117] Loss 0.309070 Top1 87.656250 Top5 98.554688 +2023-10-02 22:03:05,079 - Epoch: [194][ 40/ 117] Loss 0.299002 Top1 87.939453 Top5 98.662109 +2023-10-02 22:03:05,233 - Epoch: [194][ 50/ 117] Loss 0.293423 Top1 88.242188 Top5 98.750000 +2023-10-02 22:03:05,387 - Epoch: [194][ 60/ 117] Loss 0.305862 Top1 87.988281 Top5 98.691406 +2023-10-02 22:03:05,537 - Epoch: [194][ 70/ 117] Loss 0.301748 Top1 88.002232 Top5 98.755580 +2023-10-02 22:03:05,688 - Epoch: [194][ 80/ 117] Loss 0.305905 Top1 87.905273 Top5 98.696289 +2023-10-02 22:03:05,838 - Epoch: [194][ 90/ 117] Loss 0.303630 Top1 87.968750 Top5 98.719618 +2023-10-02 22:03:05,989 - Epoch: [194][ 100/ 117] Loss 0.300401 Top1 88.085938 Top5 98.746094 +2023-10-02 22:03:06,145 - Epoch: [194][ 110/ 117] Loss 0.302681 Top1 87.993608 Top5 98.750000 +2023-10-02 22:03:06,234 - Epoch: [194][ 117/ 117] Loss 0.303049 Top1 87.997195 Top5 98.754300 +2023-10-02 22:03:06,380 - ==> Top1: 87.997 Top5: 98.754 Loss: 0.303 + +2023-10-02 22:03:06,381 - ==> Confusion: +[[ 934 0 3 1 4 2 0 0 4 64 1 1 1 1 6 0 1 0 2 0 25] + [ 0 1067 0 0 5 17 0 18 0 1 1 0 0 0 0 3 1 0 8 3 7] + [ 3 1 984 5 0 1 15 8 0 1 2 0 5 2 1 2 2 2 11 2 9] + [ 2 3 13 985 0 1 0 1 2 0 7 1 6 3 22 2 1 5 12 1 22] + [ 23 6 0 1 968 4 1 0 1 9 1 0 1 4 11 3 9 0 0 1 7] + [ 3 28 0 0 3 1011 2 18 1 5 2 5 0 6 3 0 4 1 4 1 19] + [ 0 3 24 0 0 0 1134 6 0 0 3 1 0 0 0 5 0 1 2 7 5] + [ 0 8 11 0 5 20 7 1084 0 4 4 4 2 3 1 1 0 0 45 9 10] + [ 16 2 0 0 2 3 0 2 989 30 10 3 0 11 10 0 2 1 3 2 3] + [ 92 0 1 2 6 2 0 0 27 953 0 1 0 20 5 1 0 0 0 2 7] + [ 3 1 8 5 0 2 2 2 7 2 977 1 1 13 7 0 2 1 4 1 14] + [ 0 0 0 0 0 12 0 6 0 0 0 968 19 4 0 2 1 15 0 2 6] + [ 0 1 1 2 0 1 2 1 0 1 2 25 983 0 1 5 1 14 2 6 20] + [ 1 0 0 0 4 8 0 0 11 10 3 8 0 1047 4 0 0 1 0 1 21] + [ 12 0 4 14 3 1 0 0 21 2 2 0 3 3 1017 0 0 2 7 0 10] + [ 0 0 2 1 5 0 1 0 0 0 0 6 8 0 0 1069 15 10 2 9 6] + [ 0 12 0 0 5 6 0 0 0 0 0 4 0 3 4 7 1100 0 2 6 12] + [ 0 1 1 1 0 0 3 0 0 1 0 5 15 1 1 3 0 1001 0 2 3] + [ 2 3 2 13 1 0 0 19 5 1 0 0 1 0 7 0 0 0 1001 0 13] + [ 0 1 4 2 1 2 9 4 0 1 0 13 5 2 0 2 7 2 0 1087 10] + [ 96 109 92 60 52 121 26 76 67 52 125 85 271 220 94 41 63 50 92 123 5990]] + +2023-10-02 22:03:06,382 - ==> Best [Top1: 87.997 Top5: 98.754 Sparsity:0.00 Params: 169472 on epoch: 194] +2023-10-02 22:03:06,382 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:03:06,390 - + +2023-10-02 22:03:06,390 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:03:07,412 - Epoch: [195][ 10/ 1236] Overall Loss 0.103052 Objective Loss 0.103052 LR 0.000031 Time 0.102190 +2023-10-02 22:03:07,622 - Epoch: [195][ 20/ 1236] Overall Loss 0.099487 Objective Loss 0.099487 LR 0.000031 Time 0.061577 +2023-10-02 22:03:07,833 - Epoch: [195][ 30/ 1236] Overall Loss 0.101295 Objective Loss 0.101295 LR 0.000031 Time 0.048066 +2023-10-02 22:03:08,042 - Epoch: [195][ 40/ 1236] Overall Loss 0.104519 Objective Loss 0.104519 LR 0.000031 Time 0.041282 +2023-10-02 22:03:08,254 - Epoch: [195][ 50/ 1236] Overall Loss 0.100856 Objective Loss 0.100856 LR 0.000031 Time 0.037246 +2023-10-02 22:03:08,463 - Epoch: [195][ 60/ 1236] Overall Loss 0.101306 Objective Loss 0.101306 LR 0.000031 Time 0.034519 +2023-10-02 22:03:08,674 - Epoch: [195][ 70/ 1236] Overall Loss 0.101842 Objective Loss 0.101842 LR 0.000031 Time 0.032596 +2023-10-02 22:03:08,883 - Epoch: [195][ 80/ 1236] Overall Loss 0.103153 Objective Loss 0.103153 LR 0.000031 Time 0.031135 +2023-10-02 22:03:09,091 - Epoch: [195][ 90/ 1236] Overall Loss 0.102761 Objective Loss 0.102761 LR 0.000031 Time 0.029975 +2023-10-02 22:03:09,298 - Epoch: [195][ 100/ 1236] Overall Loss 0.103782 Objective Loss 0.103782 LR 0.000031 Time 0.029054 +2023-10-02 22:03:09,506 - Epoch: [195][ 110/ 1236] Overall Loss 0.104618 Objective Loss 0.104618 LR 0.000031 Time 0.028288 +2023-10-02 22:03:09,720 - Epoch: [195][ 120/ 1236] Overall Loss 0.104915 Objective Loss 0.104915 LR 0.000031 Time 0.027710 +2023-10-02 22:03:09,930 - Epoch: [195][ 130/ 1236] Overall Loss 0.104313 Objective Loss 0.104313 LR 0.000031 Time 0.027190 +2023-10-02 22:03:10,144 - Epoch: [195][ 140/ 1236] Overall Loss 0.102860 Objective Loss 0.102860 LR 0.000031 Time 0.026773 +2023-10-02 22:03:10,353 - Epoch: [195][ 150/ 1236] Overall Loss 0.101785 Objective Loss 0.101785 LR 0.000031 Time 0.026381 +2023-10-02 22:03:10,565 - Epoch: [195][ 160/ 1236] Overall Loss 0.102207 Objective Loss 0.102207 LR 0.000031 Time 0.026054 +2023-10-02 22:03:10,773 - Epoch: [195][ 170/ 1236] Overall Loss 0.102874 Objective Loss 0.102874 LR 0.000031 Time 0.025745 +2023-10-02 22:03:10,984 - Epoch: [195][ 180/ 1236] Overall Loss 0.103284 Objective Loss 0.103284 LR 0.000031 Time 0.025487 +2023-10-02 22:03:11,191 - Epoch: [195][ 190/ 1236] Overall Loss 0.102998 Objective Loss 0.102998 LR 0.000031 Time 0.025233 +2023-10-02 22:03:11,402 - Epoch: [195][ 200/ 1236] Overall Loss 0.103298 Objective Loss 0.103298 LR 0.000031 Time 0.025026 +2023-10-02 22:03:11,610 - Epoch: [195][ 210/ 1236] Overall Loss 0.103105 Objective Loss 0.103105 LR 0.000031 Time 0.024822 +2023-10-02 22:03:11,821 - Epoch: [195][ 220/ 1236] Overall Loss 0.103176 Objective Loss 0.103176 LR 0.000031 Time 0.024652 +2023-10-02 22:03:12,029 - Epoch: [195][ 230/ 1236] Overall Loss 0.103403 Objective Loss 0.103403 LR 0.000031 Time 0.024484 +2023-10-02 22:03:12,240 - Epoch: [195][ 240/ 1236] Overall Loss 0.103746 Objective Loss 0.103746 LR 0.000031 Time 0.024342 +2023-10-02 22:03:12,448 - Epoch: [195][ 250/ 1236] Overall Loss 0.104186 Objective Loss 0.104186 LR 0.000031 Time 0.024200 +2023-10-02 22:03:12,660 - Epoch: [195][ 260/ 1236] Overall Loss 0.104093 Objective Loss 0.104093 LR 0.000031 Time 0.024080 +2023-10-02 22:03:12,868 - Epoch: [195][ 270/ 1236] Overall Loss 0.103619 Objective Loss 0.103619 LR 0.000031 Time 0.023959 +2023-10-02 22:03:13,079 - Epoch: [195][ 280/ 1236] Overall Loss 0.104418 Objective Loss 0.104418 LR 0.000031 Time 0.023856 +2023-10-02 22:03:13,287 - Epoch: [195][ 290/ 1236] Overall Loss 0.104573 Objective Loss 0.104573 LR 0.000031 Time 0.023750 +2023-10-02 22:03:13,498 - Epoch: [195][ 300/ 1236] Overall Loss 0.105133 Objective Loss 0.105133 LR 0.000031 Time 0.023661 +2023-10-02 22:03:13,705 - Epoch: [195][ 310/ 1236] Overall Loss 0.104605 Objective Loss 0.104605 LR 0.000031 Time 0.023565 +2023-10-02 22:03:13,915 - Epoch: [195][ 320/ 1236] Overall Loss 0.104720 Objective Loss 0.104720 LR 0.000031 Time 0.023482 +2023-10-02 22:03:14,123 - Epoch: [195][ 330/ 1236] Overall Loss 0.104403 Objective Loss 0.104403 LR 0.000031 Time 0.023398 +2023-10-02 22:03:14,334 - Epoch: [195][ 340/ 1236] Overall Loss 0.104334 Objective Loss 0.104334 LR 0.000031 Time 0.023328 +2023-10-02 22:03:14,541 - Epoch: [195][ 350/ 1236] Overall Loss 0.104505 Objective Loss 0.104505 LR 0.000031 Time 0.023252 +2023-10-02 22:03:14,751 - Epoch: [195][ 360/ 1236] Overall Loss 0.104413 Objective Loss 0.104413 LR 0.000031 Time 0.023190 +2023-10-02 22:03:14,959 - Epoch: [195][ 370/ 1236] Overall Loss 0.104438 Objective Loss 0.104438 LR 0.000031 Time 0.023124 +2023-10-02 22:03:15,168 - Epoch: [195][ 380/ 1236] Overall Loss 0.104318 Objective Loss 0.104318 LR 0.000031 Time 0.023065 +2023-10-02 22:03:15,377 - Epoch: [195][ 390/ 1236] Overall Loss 0.104109 Objective Loss 0.104109 LR 0.000031 Time 0.023005 +2023-10-02 22:03:15,586 - Epoch: [195][ 400/ 1236] Overall Loss 0.104178 Objective Loss 0.104178 LR 0.000031 Time 0.022952 +2023-10-02 22:03:15,795 - Epoch: [195][ 410/ 1236] Overall Loss 0.103947 Objective Loss 0.103947 LR 0.000031 Time 0.022897 +2023-10-02 22:03:16,005 - Epoch: [195][ 420/ 1236] Overall Loss 0.103991 Objective Loss 0.103991 LR 0.000031 Time 0.022853 +2023-10-02 22:03:16,213 - Epoch: [195][ 430/ 1236] Overall Loss 0.104299 Objective Loss 0.104299 LR 0.000031 Time 0.022803 +2023-10-02 22:03:16,422 - Epoch: [195][ 440/ 1236] Overall Loss 0.104211 Objective Loss 0.104211 LR 0.000031 Time 0.022760 +2023-10-02 22:03:16,631 - Epoch: [195][ 450/ 1236] Overall Loss 0.104287 Objective Loss 0.104287 LR 0.000031 Time 0.022715 +2023-10-02 22:03:16,840 - Epoch: [195][ 460/ 1236] Overall Loss 0.103972 Objective Loss 0.103972 LR 0.000031 Time 0.022676 +2023-10-02 22:03:17,049 - Epoch: [195][ 470/ 1236] Overall Loss 0.103744 Objective Loss 0.103744 LR 0.000031 Time 0.022634 +2023-10-02 22:03:17,259 - Epoch: [195][ 480/ 1236] Overall Loss 0.103844 Objective Loss 0.103844 LR 0.000031 Time 0.022600 +2023-10-02 22:03:17,467 - Epoch: [195][ 490/ 1236] Overall Loss 0.104065 Objective Loss 0.104065 LR 0.000031 Time 0.022562 +2023-10-02 22:03:17,676 - Epoch: [195][ 500/ 1236] Overall Loss 0.104273 Objective Loss 0.104273 LR 0.000031 Time 0.022529 +2023-10-02 22:03:17,885 - Epoch: [195][ 510/ 1236] Overall Loss 0.104487 Objective Loss 0.104487 LR 0.000031 Time 0.022493 +2023-10-02 22:03:18,095 - Epoch: [195][ 520/ 1236] Overall Loss 0.104607 Objective Loss 0.104607 LR 0.000031 Time 0.022465 +2023-10-02 22:03:18,303 - Epoch: [195][ 530/ 1236] Overall Loss 0.104407 Objective Loss 0.104407 LR 0.000031 Time 0.022432 +2023-10-02 22:03:18,513 - Epoch: [195][ 540/ 1236] Overall Loss 0.104114 Objective Loss 0.104114 LR 0.000031 Time 0.022406 +2023-10-02 22:03:18,721 - Epoch: [195][ 550/ 1236] Overall Loss 0.104202 Objective Loss 0.104202 LR 0.000031 Time 0.022375 +2023-10-02 22:03:18,930 - Epoch: [195][ 560/ 1236] Overall Loss 0.103972 Objective Loss 0.103972 LR 0.000031 Time 0.022349 +2023-10-02 22:03:19,139 - Epoch: [195][ 570/ 1236] Overall Loss 0.104126 Objective Loss 0.104126 LR 0.000031 Time 0.022320 +2023-10-02 22:03:19,350 - Epoch: [195][ 580/ 1236] Overall Loss 0.103836 Objective Loss 0.103836 LR 0.000031 Time 0.022299 +2023-10-02 22:03:19,557 - Epoch: [195][ 590/ 1236] Overall Loss 0.103582 Objective Loss 0.103582 LR 0.000031 Time 0.022272 +2023-10-02 22:03:19,768 - Epoch: [195][ 600/ 1236] Overall Loss 0.103689 Objective Loss 0.103689 LR 0.000031 Time 0.022251 +2023-10-02 22:03:19,975 - Epoch: [195][ 610/ 1236] Overall Loss 0.103945 Objective Loss 0.103945 LR 0.000031 Time 0.022226 +2023-10-02 22:03:20,185 - Epoch: [195][ 620/ 1236] Overall Loss 0.103832 Objective Loss 0.103832 LR 0.000031 Time 0.022205 +2023-10-02 22:03:20,394 - Epoch: [195][ 630/ 1236] Overall Loss 0.103753 Objective Loss 0.103753 LR 0.000031 Time 0.022182 +2023-10-02 22:03:20,605 - Epoch: [195][ 640/ 1236] Overall Loss 0.103922 Objective Loss 0.103922 LR 0.000031 Time 0.022165 +2023-10-02 22:03:20,812 - Epoch: [195][ 650/ 1236] Overall Loss 0.103927 Objective Loss 0.103927 LR 0.000031 Time 0.022143 +2023-10-02 22:03:21,023 - Epoch: [195][ 660/ 1236] Overall Loss 0.103830 Objective Loss 0.103830 LR 0.000031 Time 0.022126 +2023-10-02 22:03:21,231 - Epoch: [195][ 670/ 1236] Overall Loss 0.103975 Objective Loss 0.103975 LR 0.000031 Time 0.022105 +2023-10-02 22:03:21,442 - Epoch: [195][ 680/ 1236] Overall Loss 0.104224 Objective Loss 0.104224 LR 0.000031 Time 0.022090 +2023-10-02 22:03:21,649 - Epoch: [195][ 690/ 1236] Overall Loss 0.103991 Objective Loss 0.103991 LR 0.000031 Time 0.022070 +2023-10-02 22:03:21,860 - Epoch: [195][ 700/ 1236] Overall Loss 0.104092 Objective Loss 0.104092 LR 0.000031 Time 0.022056 +2023-10-02 22:03:22,068 - Epoch: [195][ 710/ 1236] Overall Loss 0.104063 Objective Loss 0.104063 LR 0.000031 Time 0.022037 +2023-10-02 22:03:22,279 - Epoch: [195][ 720/ 1236] Overall Loss 0.104222 Objective Loss 0.104222 LR 0.000031 Time 0.022024 +2023-10-02 22:03:22,486 - Epoch: [195][ 730/ 1236] Overall Loss 0.104388 Objective Loss 0.104388 LR 0.000031 Time 0.022006 +2023-10-02 22:03:22,696 - Epoch: [195][ 740/ 1236] Overall Loss 0.104540 Objective Loss 0.104540 LR 0.000031 Time 0.021991 +2023-10-02 22:03:22,905 - Epoch: [195][ 750/ 1236] Overall Loss 0.104445 Objective Loss 0.104445 LR 0.000031 Time 0.021975 +2023-10-02 22:03:23,116 - Epoch: [195][ 760/ 1236] Overall Loss 0.104596 Objective Loss 0.104596 LR 0.000031 Time 0.021963 +2023-10-02 22:03:23,323 - Epoch: [195][ 770/ 1236] Overall Loss 0.104633 Objective Loss 0.104633 LR 0.000031 Time 0.021947 +2023-10-02 22:03:23,535 - Epoch: [195][ 780/ 1236] Overall Loss 0.104509 Objective Loss 0.104509 LR 0.000031 Time 0.021936 +2023-10-02 22:03:23,742 - Epoch: [195][ 790/ 1236] Overall Loss 0.104676 Objective Loss 0.104676 LR 0.000031 Time 0.021921 +2023-10-02 22:03:23,952 - Epoch: [195][ 800/ 1236] Overall Loss 0.104701 Objective Loss 0.104701 LR 0.000031 Time 0.021908 +2023-10-02 22:03:24,161 - Epoch: [195][ 810/ 1236] Overall Loss 0.104788 Objective Loss 0.104788 LR 0.000031 Time 0.021894 +2023-10-02 22:03:24,371 - Epoch: [195][ 820/ 1236] Overall Loss 0.104930 Objective Loss 0.104930 LR 0.000031 Time 0.021882 +2023-10-02 22:03:24,580 - Epoch: [195][ 830/ 1236] Overall Loss 0.104984 Objective Loss 0.104984 LR 0.000031 Time 0.021869 +2023-10-02 22:03:24,790 - Epoch: [195][ 840/ 1236] Overall Loss 0.104758 Objective Loss 0.104758 LR 0.000031 Time 0.021859 +2023-10-02 22:03:24,998 - Epoch: [195][ 850/ 1236] Overall Loss 0.104742 Objective Loss 0.104742 LR 0.000031 Time 0.021846 +2023-10-02 22:03:25,209 - Epoch: [195][ 860/ 1236] Overall Loss 0.104776 Objective Loss 0.104776 LR 0.000031 Time 0.021837 +2023-10-02 22:03:25,417 - Epoch: [195][ 870/ 1236] Overall Loss 0.104622 Objective Loss 0.104622 LR 0.000031 Time 0.021824 +2023-10-02 22:03:25,627 - Epoch: [195][ 880/ 1236] Overall Loss 0.104610 Objective Loss 0.104610 LR 0.000031 Time 0.021815 +2023-10-02 22:03:25,835 - Epoch: [195][ 890/ 1236] Overall Loss 0.104650 Objective Loss 0.104650 LR 0.000031 Time 0.021803 +2023-10-02 22:03:26,046 - Epoch: [195][ 900/ 1236] Overall Loss 0.104729 Objective Loss 0.104729 LR 0.000031 Time 0.021795 +2023-10-02 22:03:26,254 - Epoch: [195][ 910/ 1236] Overall Loss 0.104582 Objective Loss 0.104582 LR 0.000031 Time 0.021784 +2023-10-02 22:03:26,463 - Epoch: [195][ 920/ 1236] Overall Loss 0.104430 Objective Loss 0.104430 LR 0.000031 Time 0.021774 +2023-10-02 22:03:26,672 - Epoch: [195][ 930/ 1236] Overall Loss 0.104238 Objective Loss 0.104238 LR 0.000031 Time 0.021763 +2023-10-02 22:03:26,882 - Epoch: [195][ 940/ 1236] Overall Loss 0.104232 Objective Loss 0.104232 LR 0.000031 Time 0.021754 +2023-10-02 22:03:27,091 - Epoch: [195][ 950/ 1236] Overall Loss 0.104259 Objective Loss 0.104259 LR 0.000031 Time 0.021743 +2023-10-02 22:03:27,300 - Epoch: [195][ 960/ 1236] Overall Loss 0.104431 Objective Loss 0.104431 LR 0.000031 Time 0.021735 +2023-10-02 22:03:27,510 - Epoch: [195][ 970/ 1236] Overall Loss 0.104637 Objective Loss 0.104637 LR 0.000031 Time 0.021725 +2023-10-02 22:03:27,719 - Epoch: [195][ 980/ 1236] Overall Loss 0.104563 Objective Loss 0.104563 LR 0.000031 Time 0.021717 +2023-10-02 22:03:27,928 - Epoch: [195][ 990/ 1236] Overall Loss 0.104640 Objective Loss 0.104640 LR 0.000031 Time 0.021707 +2023-10-02 22:03:28,139 - Epoch: [195][ 1000/ 1236] Overall Loss 0.104673 Objective Loss 0.104673 LR 0.000031 Time 0.021701 +2023-10-02 22:03:28,347 - Epoch: [195][ 1010/ 1236] Overall Loss 0.104556 Objective Loss 0.104556 LR 0.000031 Time 0.021691 +2023-10-02 22:03:28,558 - Epoch: [195][ 1020/ 1236] Overall Loss 0.104622 Objective Loss 0.104622 LR 0.000031 Time 0.021686 +2023-10-02 22:03:28,766 - Epoch: [195][ 1030/ 1236] Overall Loss 0.104378 Objective Loss 0.104378 LR 0.000031 Time 0.021677 +2023-10-02 22:03:28,976 - Epoch: [195][ 1040/ 1236] Overall Loss 0.104372 Objective Loss 0.104372 LR 0.000031 Time 0.021669 +2023-10-02 22:03:29,185 - Epoch: [195][ 1050/ 1236] Overall Loss 0.104375 Objective Loss 0.104375 LR 0.000031 Time 0.021661 +2023-10-02 22:03:29,396 - Epoch: [195][ 1060/ 1236] Overall Loss 0.104334 Objective Loss 0.104334 LR 0.000031 Time 0.021655 +2023-10-02 22:03:29,603 - Epoch: [195][ 1070/ 1236] Overall Loss 0.104400 Objective Loss 0.104400 LR 0.000031 Time 0.021647 +2023-10-02 22:03:29,813 - Epoch: [195][ 1080/ 1236] Overall Loss 0.104403 Objective Loss 0.104403 LR 0.000031 Time 0.021640 +2023-10-02 22:03:30,022 - Epoch: [195][ 1090/ 1236] Overall Loss 0.104487 Objective Loss 0.104487 LR 0.000031 Time 0.021632 +2023-10-02 22:03:30,231 - Epoch: [195][ 1100/ 1236] Overall Loss 0.104410 Objective Loss 0.104410 LR 0.000031 Time 0.021625 +2023-10-02 22:03:30,441 - Epoch: [195][ 1110/ 1236] Overall Loss 0.104514 Objective Loss 0.104514 LR 0.000031 Time 0.021617 +2023-10-02 22:03:30,650 - Epoch: [195][ 1120/ 1236] Overall Loss 0.104471 Objective Loss 0.104471 LR 0.000031 Time 0.021611 +2023-10-02 22:03:30,860 - Epoch: [195][ 1130/ 1236] Overall Loss 0.104364 Objective Loss 0.104364 LR 0.000031 Time 0.021605 +2023-10-02 22:03:31,069 - Epoch: [195][ 1140/ 1236] Overall Loss 0.104290 Objective Loss 0.104290 LR 0.000031 Time 0.021599 +2023-10-02 22:03:31,278 - Epoch: [195][ 1150/ 1236] Overall Loss 0.104358 Objective Loss 0.104358 LR 0.000031 Time 0.021592 +2023-10-02 22:03:31,488 - Epoch: [195][ 1160/ 1236] Overall Loss 0.104469 Objective Loss 0.104469 LR 0.000031 Time 0.021587 +2023-10-02 22:03:31,697 - Epoch: [195][ 1170/ 1236] Overall Loss 0.104524 Objective Loss 0.104524 LR 0.000031 Time 0.021579 +2023-10-02 22:03:31,907 - Epoch: [195][ 1180/ 1236] Overall Loss 0.104607 Objective Loss 0.104607 LR 0.000031 Time 0.021574 +2023-10-02 22:03:32,117 - Epoch: [195][ 1190/ 1236] Overall Loss 0.104525 Objective Loss 0.104525 LR 0.000031 Time 0.021568 +2023-10-02 22:03:32,326 - Epoch: [195][ 1200/ 1236] Overall Loss 0.104401 Objective Loss 0.104401 LR 0.000031 Time 0.021562 +2023-10-02 22:03:32,536 - Epoch: [195][ 1210/ 1236] Overall Loss 0.104522 Objective Loss 0.104522 LR 0.000031 Time 0.021556 +2023-10-02 22:03:32,746 - Epoch: [195][ 1220/ 1236] Overall Loss 0.104505 Objective Loss 0.104505 LR 0.000031 Time 0.021551 +2023-10-02 22:03:33,007 - Epoch: [195][ 1230/ 1236] Overall Loss 0.104670 Objective Loss 0.104670 LR 0.000031 Time 0.021587 +2023-10-02 22:03:33,129 - Epoch: [195][ 1236/ 1236] Overall Loss 0.104779 Objective Loss 0.104779 Top1 92.057026 Top5 98.574338 LR 0.000031 Time 0.021581 +2023-10-02 22:03:33,264 - --- validate (epoch=195)----------- +2023-10-02 22:03:33,265 - 29943 samples (256 per mini-batch) +2023-10-02 22:03:33,750 - Epoch: [195][ 10/ 117] Loss 0.289768 Top1 87.304688 Top5 98.632812 +2023-10-02 22:03:33,901 - Epoch: [195][ 20/ 117] Loss 0.305420 Top1 87.500000 Top5 98.554688 +2023-10-02 22:03:34,051 - Epoch: [195][ 30/ 117] Loss 0.305841 Top1 87.630208 Top5 98.489583 +2023-10-02 22:03:34,202 - Epoch: [195][ 40/ 117] Loss 0.298709 Top1 87.802734 Top5 98.535156 +2023-10-02 22:03:34,353 - Epoch: [195][ 50/ 117] Loss 0.295787 Top1 87.867188 Top5 98.601562 +2023-10-02 22:03:34,506 - Epoch: [195][ 60/ 117] Loss 0.293636 Top1 87.903646 Top5 98.678385 +2023-10-02 22:03:34,656 - Epoch: [195][ 70/ 117] Loss 0.296662 Top1 87.901786 Top5 98.666295 +2023-10-02 22:03:34,809 - Epoch: [195][ 80/ 117] Loss 0.299745 Top1 87.817383 Top5 98.710938 +2023-10-02 22:03:34,960 - Epoch: [195][ 90/ 117] Loss 0.302773 Top1 87.717014 Top5 98.715278 +2023-10-02 22:03:35,112 - Epoch: [195][ 100/ 117] Loss 0.303932 Top1 87.667969 Top5 98.757812 +2023-10-02 22:03:35,270 - Epoch: [195][ 110/ 117] Loss 0.302489 Top1 87.730824 Top5 98.767756 +2023-10-02 22:03:35,358 - Epoch: [195][ 117/ 117] Loss 0.303686 Top1 87.763417 Top5 98.754300 +2023-10-02 22:03:35,496 - ==> Top1: 87.763 Top5: 98.754 Loss: 0.304 + +2023-10-02 22:03:35,497 - ==> Confusion: +[[ 936 0 3 1 3 2 0 0 7 61 1 1 1 2 5 0 3 0 1 0 23] + [ 0 1076 0 0 2 17 0 17 0 0 0 0 0 0 1 3 0 0 5 3 7] + [ 2 1 985 6 1 0 18 6 0 2 1 0 7 2 1 2 2 1 11 1 7] + [ 2 3 15 984 0 1 0 1 3 1 5 1 4 3 24 3 1 5 10 1 22] + [ 26 4 0 1 972 4 0 0 0 13 0 0 0 3 8 4 10 0 0 1 4] + [ 4 33 0 0 6 1004 2 18 1 4 2 3 0 10 3 0 4 1 3 1 17] + [ 0 3 22 0 0 1 1139 5 0 0 3 1 0 0 0 3 0 1 1 7 5] + [ 2 9 13 0 5 22 5 1087 1 4 4 2 2 3 0 0 0 1 40 7 11] + [ 16 3 0 1 1 3 0 3 982 34 9 1 1 9 14 0 3 1 3 1 4] + [ 93 1 0 2 7 2 0 0 29 950 1 0 0 17 6 2 1 1 0 0 7] + [ 3 2 7 7 0 2 2 1 10 2 976 2 0 10 7 0 3 2 4 2 11] + [ 0 0 1 0 1 11 0 6 0 0 0 970 14 4 0 2 1 15 0 4 6] + [ 0 1 1 1 0 2 2 1 0 1 4 27 977 2 1 8 1 15 2 4 18] + [ 0 0 1 0 3 7 0 0 10 8 3 9 0 1057 4 0 0 0 0 1 16] + [ 14 0 3 17 2 0 0 0 23 1 2 0 2 3 1011 0 0 2 11 0 10] + [ 0 0 2 1 5 1 1 0 0 0 0 5 8 0 0 1069 19 8 1 11 3] + [ 1 15 0 0 5 7 0 0 0 1 0 3 0 3 3 7 1097 0 1 6 12] + [ 0 1 1 4 0 0 3 0 0 0 0 4 14 0 1 5 0 998 0 2 5] + [ 2 5 2 15 0 0 0 20 4 1 1 1 1 0 10 0 0 0 994 0 12] + [ 0 1 4 2 2 2 9 6 0 1 0 11 5 2 2 1 6 0 1 1090 7] + [ 101 124 103 71 58 111 31 79 70 56 132 77 273 213 107 41 65 47 98 123 5925]] + +2023-10-02 22:03:35,498 - ==> Best [Top1: 87.997 Top5: 98.754 Sparsity:0.00 Params: 169472 on epoch: 194] +2023-10-02 22:03:35,499 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:03:35,505 - + +2023-10-02 22:03:35,505 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:03:36,517 - Epoch: [196][ 10/ 1236] Overall Loss 0.095615 Objective Loss 0.095615 LR 0.000031 Time 0.101174 +2023-10-02 22:03:36,726 - Epoch: [196][ 20/ 1236] Overall Loss 0.107532 Objective Loss 0.107532 LR 0.000031 Time 0.061025 +2023-10-02 22:03:36,933 - Epoch: [196][ 30/ 1236] Overall Loss 0.110692 Objective Loss 0.110692 LR 0.000031 Time 0.047573 +2023-10-02 22:03:37,142 - Epoch: [196][ 40/ 1236] Overall Loss 0.108543 Objective Loss 0.108543 LR 0.000031 Time 0.040895 +2023-10-02 22:03:37,349 - Epoch: [196][ 50/ 1236] Overall Loss 0.108293 Objective Loss 0.108293 LR 0.000031 Time 0.036851 +2023-10-02 22:03:37,558 - Epoch: [196][ 60/ 1236] Overall Loss 0.107727 Objective Loss 0.107727 LR 0.000031 Time 0.034188 +2023-10-02 22:03:37,765 - Epoch: [196][ 70/ 1236] Overall Loss 0.105099 Objective Loss 0.105099 LR 0.000031 Time 0.032244 +2023-10-02 22:03:37,974 - Epoch: [196][ 80/ 1236] Overall Loss 0.105199 Objective Loss 0.105199 LR 0.000031 Time 0.030819 +2023-10-02 22:03:38,181 - Epoch: [196][ 90/ 1236] Overall Loss 0.104953 Objective Loss 0.104953 LR 0.000031 Time 0.029678 +2023-10-02 22:03:38,389 - Epoch: [196][ 100/ 1236] Overall Loss 0.104007 Objective Loss 0.104007 LR 0.000031 Time 0.028785 +2023-10-02 22:03:38,596 - Epoch: [196][ 110/ 1236] Overall Loss 0.103104 Objective Loss 0.103104 LR 0.000031 Time 0.028046 +2023-10-02 22:03:38,804 - Epoch: [196][ 120/ 1236] Overall Loss 0.102398 Objective Loss 0.102398 LR 0.000031 Time 0.027437 +2023-10-02 22:03:39,010 - Epoch: [196][ 130/ 1236] Overall Loss 0.102865 Objective Loss 0.102865 LR 0.000031 Time 0.026905 +2023-10-02 22:03:39,218 - Epoch: [196][ 140/ 1236] Overall Loss 0.103182 Objective Loss 0.103182 LR 0.000031 Time 0.026467 +2023-10-02 22:03:39,425 - Epoch: [196][ 150/ 1236] Overall Loss 0.103869 Objective Loss 0.103869 LR 0.000031 Time 0.026071 +2023-10-02 22:03:39,633 - Epoch: [196][ 160/ 1236] Overall Loss 0.103032 Objective Loss 0.103032 LR 0.000031 Time 0.025739 +2023-10-02 22:03:39,840 - Epoch: [196][ 170/ 1236] Overall Loss 0.102593 Objective Loss 0.102593 LR 0.000031 Time 0.025432 +2023-10-02 22:03:40,049 - Epoch: [196][ 180/ 1236] Overall Loss 0.102930 Objective Loss 0.102930 LR 0.000031 Time 0.025182 +2023-10-02 22:03:40,258 - Epoch: [196][ 190/ 1236] Overall Loss 0.103661 Objective Loss 0.103661 LR 0.000031 Time 0.024948 +2023-10-02 22:03:40,470 - Epoch: [196][ 200/ 1236] Overall Loss 0.103288 Objective Loss 0.103288 LR 0.000031 Time 0.024757 +2023-10-02 22:03:40,677 - Epoch: [196][ 210/ 1236] Overall Loss 0.103442 Objective Loss 0.103442 LR 0.000031 Time 0.024566 +2023-10-02 22:03:40,886 - Epoch: [196][ 220/ 1236] Overall Loss 0.103339 Objective Loss 0.103339 LR 0.000031 Time 0.024397 +2023-10-02 22:03:41,095 - Epoch: [196][ 230/ 1236] Overall Loss 0.103444 Objective Loss 0.103444 LR 0.000031 Time 0.024243 +2023-10-02 22:03:41,305 - Epoch: [196][ 240/ 1236] Overall Loss 0.103554 Objective Loss 0.103554 LR 0.000031 Time 0.024107 +2023-10-02 22:03:41,514 - Epoch: [196][ 250/ 1236] Overall Loss 0.103714 Objective Loss 0.103714 LR 0.000031 Time 0.023974 +2023-10-02 22:03:41,724 - Epoch: [196][ 260/ 1236] Overall Loss 0.103034 Objective Loss 0.103034 LR 0.000031 Time 0.023858 +2023-10-02 22:03:41,933 - Epoch: [196][ 270/ 1236] Overall Loss 0.104141 Objective Loss 0.104141 LR 0.000031 Time 0.023743 +2023-10-02 22:03:42,145 - Epoch: [196][ 280/ 1236] Overall Loss 0.104036 Objective Loss 0.104036 LR 0.000031 Time 0.023649 +2023-10-02 22:03:42,352 - Epoch: [196][ 290/ 1236] Overall Loss 0.103959 Objective Loss 0.103959 LR 0.000031 Time 0.023548 +2023-10-02 22:03:42,564 - Epoch: [196][ 300/ 1236] Overall Loss 0.104144 Objective Loss 0.104144 LR 0.000031 Time 0.023468 +2023-10-02 22:03:42,772 - Epoch: [196][ 310/ 1236] Overall Loss 0.103971 Objective Loss 0.103971 LR 0.000031 Time 0.023381 +2023-10-02 22:03:42,981 - Epoch: [196][ 320/ 1236] Overall Loss 0.104314 Objective Loss 0.104314 LR 0.000031 Time 0.023304 +2023-10-02 22:03:43,189 - Epoch: [196][ 330/ 1236] Overall Loss 0.104050 Objective Loss 0.104050 LR 0.000031 Time 0.023225 +2023-10-02 22:03:43,397 - Epoch: [196][ 340/ 1236] Overall Loss 0.103686 Objective Loss 0.103686 LR 0.000031 Time 0.023152 +2023-10-02 22:03:43,604 - Epoch: [196][ 350/ 1236] Overall Loss 0.103708 Objective Loss 0.103708 LR 0.000031 Time 0.023083 +2023-10-02 22:03:43,812 - Epoch: [196][ 360/ 1236] Overall Loss 0.103535 Objective Loss 0.103535 LR 0.000031 Time 0.023017 +2023-10-02 22:03:44,020 - Epoch: [196][ 370/ 1236] Overall Loss 0.103735 Objective Loss 0.103735 LR 0.000031 Time 0.022957 +2023-10-02 22:03:44,229 - Epoch: [196][ 380/ 1236] Overall Loss 0.103345 Objective Loss 0.103345 LR 0.000031 Time 0.022902 +2023-10-02 22:03:44,438 - Epoch: [196][ 390/ 1236] Overall Loss 0.103195 Objective Loss 0.103195 LR 0.000031 Time 0.022846 +2023-10-02 22:03:44,647 - Epoch: [196][ 400/ 1236] Overall Loss 0.102983 Objective Loss 0.102983 LR 0.000031 Time 0.022796 +2023-10-02 22:03:44,855 - Epoch: [196][ 410/ 1236] Overall Loss 0.103493 Objective Loss 0.103493 LR 0.000031 Time 0.022744 +2023-10-02 22:03:45,063 - Epoch: [196][ 420/ 1236] Overall Loss 0.103433 Objective Loss 0.103433 LR 0.000031 Time 0.022699 +2023-10-02 22:03:45,272 - Epoch: [196][ 430/ 1236] Overall Loss 0.103398 Objective Loss 0.103398 LR 0.000031 Time 0.022651 +2023-10-02 22:03:45,481 - Epoch: [196][ 440/ 1236] Overall Loss 0.103178 Objective Loss 0.103178 LR 0.000031 Time 0.022611 +2023-10-02 22:03:45,689 - Epoch: [196][ 450/ 1236] Overall Loss 0.103058 Objective Loss 0.103058 LR 0.000031 Time 0.022568 +2023-10-02 22:03:45,898 - Epoch: [196][ 460/ 1236] Overall Loss 0.103217 Objective Loss 0.103217 LR 0.000031 Time 0.022530 +2023-10-02 22:03:46,107 - Epoch: [196][ 470/ 1236] Overall Loss 0.103519 Objective Loss 0.103519 LR 0.000031 Time 0.022492 +2023-10-02 22:03:46,315 - Epoch: [196][ 480/ 1236] Overall Loss 0.103811 Objective Loss 0.103811 LR 0.000031 Time 0.022457 +2023-10-02 22:03:46,528 - Epoch: [196][ 490/ 1236] Overall Loss 0.104114 Objective Loss 0.104114 LR 0.000031 Time 0.022429 +2023-10-02 22:03:46,738 - Epoch: [196][ 500/ 1236] Overall Loss 0.104136 Objective Loss 0.104136 LR 0.000031 Time 0.022400 +2023-10-02 22:03:46,949 - Epoch: [196][ 510/ 1236] Overall Loss 0.104033 Objective Loss 0.104033 LR 0.000031 Time 0.022375 +2023-10-02 22:03:47,160 - Epoch: [196][ 520/ 1236] Overall Loss 0.103810 Objective Loss 0.103810 LR 0.000031 Time 0.022347 +2023-10-02 22:03:47,373 - Epoch: [196][ 530/ 1236] Overall Loss 0.103850 Objective Loss 0.103850 LR 0.000031 Time 0.022326 +2023-10-02 22:03:47,582 - Epoch: [196][ 540/ 1236] Overall Loss 0.103625 Objective Loss 0.103625 LR 0.000031 Time 0.022299 +2023-10-02 22:03:47,795 - Epoch: [196][ 550/ 1236] Overall Loss 0.103716 Objective Loss 0.103716 LR 0.000031 Time 0.022280 +2023-10-02 22:03:48,005 - Epoch: [196][ 560/ 1236] Overall Loss 0.103633 Objective Loss 0.103633 LR 0.000031 Time 0.022256 +2023-10-02 22:03:48,215 - Epoch: [196][ 570/ 1236] Overall Loss 0.103658 Objective Loss 0.103658 LR 0.000031 Time 0.022234 +2023-10-02 22:03:48,426 - Epoch: [196][ 580/ 1236] Overall Loss 0.103537 Objective Loss 0.103537 LR 0.000031 Time 0.022211 +2023-10-02 22:03:48,636 - Epoch: [196][ 590/ 1236] Overall Loss 0.103905 Objective Loss 0.103905 LR 0.000031 Time 0.022191 +2023-10-02 22:03:48,849 - Epoch: [196][ 600/ 1236] Overall Loss 0.103928 Objective Loss 0.103928 LR 0.000031 Time 0.022173 +2023-10-02 22:03:49,061 - Epoch: [196][ 610/ 1236] Overall Loss 0.103826 Objective Loss 0.103826 LR 0.000031 Time 0.022156 +2023-10-02 22:03:49,272 - Epoch: [196][ 620/ 1236] Overall Loss 0.104109 Objective Loss 0.104109 LR 0.000031 Time 0.022137 +2023-10-02 22:03:49,484 - Epoch: [196][ 630/ 1236] Overall Loss 0.103987 Objective Loss 0.103987 LR 0.000031 Time 0.022122 +2023-10-02 22:03:49,696 - Epoch: [196][ 640/ 1236] Overall Loss 0.103863 Objective Loss 0.103863 LR 0.000031 Time 0.022104 +2023-10-02 22:03:49,909 - Epoch: [196][ 650/ 1236] Overall Loss 0.103890 Objective Loss 0.103890 LR 0.000031 Time 0.022090 +2023-10-02 22:03:50,118 - Epoch: [196][ 660/ 1236] Overall Loss 0.103893 Objective Loss 0.103893 LR 0.000031 Time 0.022070 +2023-10-02 22:03:50,330 - Epoch: [196][ 670/ 1236] Overall Loss 0.103893 Objective Loss 0.103893 LR 0.000031 Time 0.022056 +2023-10-02 22:03:50,539 - Epoch: [196][ 680/ 1236] Overall Loss 0.104359 Objective Loss 0.104359 LR 0.000031 Time 0.022039 +2023-10-02 22:03:50,751 - Epoch: [196][ 690/ 1236] Overall Loss 0.104333 Objective Loss 0.104333 LR 0.000031 Time 0.022024 +2023-10-02 22:03:50,962 - Epoch: [196][ 700/ 1236] Overall Loss 0.104172 Objective Loss 0.104172 LR 0.000031 Time 0.022008 +2023-10-02 22:03:51,175 - Epoch: [196][ 710/ 1236] Overall Loss 0.104394 Objective Loss 0.104394 LR 0.000031 Time 0.021998 +2023-10-02 22:03:51,385 - Epoch: [196][ 720/ 1236] Overall Loss 0.104246 Objective Loss 0.104246 LR 0.000031 Time 0.021984 +2023-10-02 22:03:51,597 - Epoch: [196][ 730/ 1236] Overall Loss 0.104306 Objective Loss 0.104306 LR 0.000031 Time 0.021973 +2023-10-02 22:03:51,808 - Epoch: [196][ 740/ 1236] Overall Loss 0.104270 Objective Loss 0.104270 LR 0.000031 Time 0.021959 +2023-10-02 22:03:52,022 - Epoch: [196][ 750/ 1236] Overall Loss 0.104184 Objective Loss 0.104184 LR 0.000031 Time 0.021950 +2023-10-02 22:03:52,230 - Epoch: [196][ 760/ 1236] Overall Loss 0.104254 Objective Loss 0.104254 LR 0.000031 Time 0.021935 +2023-10-02 22:03:52,442 - Epoch: [196][ 770/ 1236] Overall Loss 0.104093 Objective Loss 0.104093 LR 0.000031 Time 0.021925 +2023-10-02 22:03:52,653 - Epoch: [196][ 780/ 1236] Overall Loss 0.104016 Objective Loss 0.104016 LR 0.000031 Time 0.021913 +2023-10-02 22:03:52,865 - Epoch: [196][ 790/ 1236] Overall Loss 0.103939 Objective Loss 0.103939 LR 0.000031 Time 0.021903 +2023-10-02 22:03:53,076 - Epoch: [196][ 800/ 1236] Overall Loss 0.104046 Objective Loss 0.104046 LR 0.000031 Time 0.021891 +2023-10-02 22:03:53,289 - Epoch: [196][ 810/ 1236] Overall Loss 0.104216 Objective Loss 0.104216 LR 0.000031 Time 0.021883 +2023-10-02 22:03:53,499 - Epoch: [196][ 820/ 1236] Overall Loss 0.104164 Objective Loss 0.104164 LR 0.000031 Time 0.021872 +2023-10-02 22:03:53,711 - Epoch: [196][ 830/ 1236] Overall Loss 0.104028 Objective Loss 0.104028 LR 0.000031 Time 0.021864 +2023-10-02 22:03:53,922 - Epoch: [196][ 840/ 1236] Overall Loss 0.103985 Objective Loss 0.103985 LR 0.000031 Time 0.021853 +2023-10-02 22:03:54,134 - Epoch: [196][ 850/ 1236] Overall Loss 0.104130 Objective Loss 0.104130 LR 0.000031 Time 0.021844 +2023-10-02 22:03:54,345 - Epoch: [196][ 860/ 1236] Overall Loss 0.104185 Objective Loss 0.104185 LR 0.000031 Time 0.021834 +2023-10-02 22:03:54,556 - Epoch: [196][ 870/ 1236] Overall Loss 0.104280 Objective Loss 0.104280 LR 0.000031 Time 0.021824 +2023-10-02 22:03:54,766 - Epoch: [196][ 880/ 1236] Overall Loss 0.104345 Objective Loss 0.104345 LR 0.000031 Time 0.021814 +2023-10-02 22:03:54,977 - Epoch: [196][ 890/ 1236] Overall Loss 0.104419 Objective Loss 0.104419 LR 0.000031 Time 0.021805 +2023-10-02 22:03:55,188 - Epoch: [196][ 900/ 1236] Overall Loss 0.104283 Objective Loss 0.104283 LR 0.000031 Time 0.021795 +2023-10-02 22:03:55,398 - Epoch: [196][ 910/ 1236] Overall Loss 0.104245 Objective Loss 0.104245 LR 0.000031 Time 0.021786 +2023-10-02 22:03:55,609 - Epoch: [196][ 920/ 1236] Overall Loss 0.104390 Objective Loss 0.104390 LR 0.000031 Time 0.021777 +2023-10-02 22:03:55,819 - Epoch: [196][ 930/ 1236] Overall Loss 0.104383 Objective Loss 0.104383 LR 0.000031 Time 0.021769 +2023-10-02 22:03:56,030 - Epoch: [196][ 940/ 1236] Overall Loss 0.104477 Objective Loss 0.104477 LR 0.000031 Time 0.021759 +2023-10-02 22:03:56,241 - Epoch: [196][ 950/ 1236] Overall Loss 0.104476 Objective Loss 0.104476 LR 0.000031 Time 0.021752 +2023-10-02 22:03:56,452 - Epoch: [196][ 960/ 1236] Overall Loss 0.104472 Objective Loss 0.104472 LR 0.000031 Time 0.021745 +2023-10-02 22:03:56,663 - Epoch: [196][ 970/ 1236] Overall Loss 0.104445 Objective Loss 0.104445 LR 0.000031 Time 0.021738 +2023-10-02 22:03:56,874 - Epoch: [196][ 980/ 1236] Overall Loss 0.104484 Objective Loss 0.104484 LR 0.000031 Time 0.021729 +2023-10-02 22:03:57,084 - Epoch: [196][ 990/ 1236] Overall Loss 0.104500 Objective Loss 0.104500 LR 0.000031 Time 0.021722 +2023-10-02 22:03:57,295 - Epoch: [196][ 1000/ 1236] Overall Loss 0.104393 Objective Loss 0.104393 LR 0.000031 Time 0.021714 +2023-10-02 22:03:57,506 - Epoch: [196][ 1010/ 1236] Overall Loss 0.104302 Objective Loss 0.104302 LR 0.000031 Time 0.021707 +2023-10-02 22:03:57,716 - Epoch: [196][ 1020/ 1236] Overall Loss 0.104400 Objective Loss 0.104400 LR 0.000031 Time 0.021699 +2023-10-02 22:03:57,927 - Epoch: [196][ 1030/ 1236] Overall Loss 0.104515 Objective Loss 0.104515 LR 0.000031 Time 0.021693 +2023-10-02 22:03:58,138 - Epoch: [196][ 1040/ 1236] Overall Loss 0.104568 Objective Loss 0.104568 LR 0.000031 Time 0.021687 +2023-10-02 22:03:58,348 - Epoch: [196][ 1050/ 1236] Overall Loss 0.104388 Objective Loss 0.104388 LR 0.000031 Time 0.021680 +2023-10-02 22:03:58,559 - Epoch: [196][ 1060/ 1236] Overall Loss 0.104317 Objective Loss 0.104317 LR 0.000031 Time 0.021673 +2023-10-02 22:03:58,770 - Epoch: [196][ 1070/ 1236] Overall Loss 0.104215 Objective Loss 0.104215 LR 0.000031 Time 0.021668 +2023-10-02 22:03:58,981 - Epoch: [196][ 1080/ 1236] Overall Loss 0.104276 Objective Loss 0.104276 LR 0.000031 Time 0.021661 +2023-10-02 22:03:59,192 - Epoch: [196][ 1090/ 1236] Overall Loss 0.104334 Objective Loss 0.104334 LR 0.000031 Time 0.021655 +2023-10-02 22:03:59,402 - Epoch: [196][ 1100/ 1236] Overall Loss 0.104402 Objective Loss 0.104402 LR 0.000031 Time 0.021648 +2023-10-02 22:03:59,613 - Epoch: [196][ 1110/ 1236] Overall Loss 0.104318 Objective Loss 0.104318 LR 0.000031 Time 0.021642 +2023-10-02 22:03:59,824 - Epoch: [196][ 1120/ 1236] Overall Loss 0.104379 Objective Loss 0.104379 LR 0.000031 Time 0.021637 +2023-10-02 22:04:00,035 - Epoch: [196][ 1130/ 1236] Overall Loss 0.104365 Objective Loss 0.104365 LR 0.000031 Time 0.021632 +2023-10-02 22:04:00,246 - Epoch: [196][ 1140/ 1236] Overall Loss 0.104477 Objective Loss 0.104477 LR 0.000031 Time 0.021626 +2023-10-02 22:04:00,456 - Epoch: [196][ 1150/ 1236] Overall Loss 0.104217 Objective Loss 0.104217 LR 0.000031 Time 0.021620 +2023-10-02 22:04:00,666 - Epoch: [196][ 1160/ 1236] Overall Loss 0.104192 Objective Loss 0.104192 LR 0.000031 Time 0.021615 +2023-10-02 22:04:00,877 - Epoch: [196][ 1170/ 1236] Overall Loss 0.104174 Objective Loss 0.104174 LR 0.000031 Time 0.021610 +2023-10-02 22:04:01,088 - Epoch: [196][ 1180/ 1236] Overall Loss 0.104072 Objective Loss 0.104072 LR 0.000031 Time 0.021604 +2023-10-02 22:04:01,299 - Epoch: [196][ 1190/ 1236] Overall Loss 0.103914 Objective Loss 0.103914 LR 0.000031 Time 0.021599 +2023-10-02 22:04:01,510 - Epoch: [196][ 1200/ 1236] Overall Loss 0.103973 Objective Loss 0.103973 LR 0.000031 Time 0.021595 +2023-10-02 22:04:01,720 - Epoch: [196][ 1210/ 1236] Overall Loss 0.104027 Objective Loss 0.104027 LR 0.000031 Time 0.021590 +2023-10-02 22:04:01,931 - Epoch: [196][ 1220/ 1236] Overall Loss 0.104043 Objective Loss 0.104043 LR 0.000031 Time 0.021584 +2023-10-02 22:04:02,194 - Epoch: [196][ 1230/ 1236] Overall Loss 0.103971 Objective Loss 0.103971 LR 0.000031 Time 0.021623 +2023-10-02 22:04:02,316 - Epoch: [196][ 1236/ 1236] Overall Loss 0.103877 Objective Loss 0.103877 Top1 93.686354 Top5 99.185336 LR 0.000031 Time 0.021616 +2023-10-02 22:04:02,459 - --- validate (epoch=196)----------- +2023-10-02 22:04:02,459 - 29943 samples (256 per mini-batch) +2023-10-02 22:04:02,962 - Epoch: [196][ 10/ 117] Loss 0.312463 Top1 87.695312 Top5 98.789062 +2023-10-02 22:04:03,114 - Epoch: [196][ 20/ 117] Loss 0.301037 Top1 88.085938 Top5 98.828125 +2023-10-02 22:04:03,264 - Epoch: [196][ 30/ 117] Loss 0.290730 Top1 88.346354 Top5 98.893229 +2023-10-02 22:04:03,414 - Epoch: [196][ 40/ 117] Loss 0.291283 Top1 88.222656 Top5 98.759766 +2023-10-02 22:04:03,566 - Epoch: [196][ 50/ 117] Loss 0.293911 Top1 88.203125 Top5 98.804688 +2023-10-02 22:04:03,727 - Epoch: [196][ 60/ 117] Loss 0.297977 Top1 88.144531 Top5 98.834635 +2023-10-02 22:04:03,883 - Epoch: [196][ 70/ 117] Loss 0.299915 Top1 87.974330 Top5 98.755580 +2023-10-02 22:04:04,044 - Epoch: [196][ 80/ 117] Loss 0.302047 Top1 87.900391 Top5 98.730469 +2023-10-02 22:04:04,200 - Epoch: [196][ 90/ 117] Loss 0.302701 Top1 87.821181 Top5 98.732639 +2023-10-02 22:04:04,362 - Epoch: [196][ 100/ 117] Loss 0.299671 Top1 87.867188 Top5 98.746094 +2023-10-02 22:04:04,526 - Epoch: [196][ 110/ 117] Loss 0.301398 Top1 87.933239 Top5 98.757102 +2023-10-02 22:04:04,614 - Epoch: [196][ 117/ 117] Loss 0.303277 Top1 87.850249 Top5 98.737601 +2023-10-02 22:04:04,731 - ==> Top1: 87.850 Top5: 98.738 Loss: 0.303 + +2023-10-02 22:04:04,732 - ==> Confusion: +[[ 944 0 3 2 3 2 0 0 8 51 1 0 1 2 6 0 1 0 2 0 24] + [ 0 1077 0 0 3 13 1 18 0 2 0 0 0 0 0 3 1 0 4 2 7] + [ 1 0 986 7 2 0 17 7 0 2 2 0 6 2 1 2 2 1 9 2 7] + [ 2 3 15 985 1 1 1 0 2 1 5 1 4 2 25 2 1 5 11 1 21] + [ 24 5 0 1 972 4 1 0 0 12 0 0 0 3 8 4 10 0 0 1 5] + [ 3 34 0 1 7 1005 2 15 1 4 1 6 0 7 5 0 2 0 3 1 19] + [ 0 3 22 0 0 1 1135 5 0 0 3 0 0 0 0 7 0 1 1 7 6] + [ 0 9 10 0 5 23 8 1085 2 2 4 5 3 4 1 0 0 1 37 6 13] + [ 15 4 0 1 1 3 0 1 986 30 10 1 2 9 14 0 2 1 3 2 4] + [ 92 2 1 2 9 2 0 0 30 943 0 1 0 22 6 0 0 1 0 1 7] + [ 4 2 7 5 0 3 1 2 10 1 985 2 0 6 6 0 1 1 2 3 12] + [ 0 0 0 0 0 11 0 5 0 0 0 972 15 4 0 2 0 16 0 3 7] + [ 0 1 1 1 0 2 1 1 0 1 3 31 980 2 2 8 0 12 2 5 15] + [ 0 0 1 0 4 6 0 0 13 8 3 9 0 1050 4 1 0 0 0 1 19] + [ 12 0 3 13 5 1 0 0 21 1 1 0 2 3 1017 0 1 2 9 0 10] + [ 0 0 2 1 5 1 1 0 0 0 0 5 7 0 0 1071 17 10 2 8 4] + [ 0 14 1 0 4 6 2 0 1 1 0 4 0 3 3 8 1099 0 1 3 11] + [ 0 1 1 1 0 0 2 0 0 0 0 4 19 0 3 7 0 992 0 3 5] + [ 3 4 3 15 0 0 0 19 2 1 2 1 0 0 13 0 0 0 994 0 11] + [ 0 1 3 2 1 2 8 5 0 0 0 13 6 2 1 2 8 1 0 1089 8] + [ 98 135 97 73 52 115 27 77 65 51 141 75 268 205 114 41 74 47 100 112 5938]] + +2023-10-02 22:04:04,733 - ==> Best [Top1: 87.997 Top5: 98.754 Sparsity:0.00 Params: 169472 on epoch: 194] +2023-10-02 22:04:04,733 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:04:04,739 - + +2023-10-02 22:04:04,739 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:04:05,880 - Epoch: [197][ 10/ 1236] Overall Loss 0.100915 Objective Loss 0.100915 LR 0.000031 Time 0.114003 +2023-10-02 22:04:06,092 - Epoch: [197][ 20/ 1236] Overall Loss 0.097770 Objective Loss 0.097770 LR 0.000031 Time 0.067585 +2023-10-02 22:04:06,302 - Epoch: [197][ 30/ 1236] Overall Loss 0.099100 Objective Loss 0.099100 LR 0.000031 Time 0.052070 +2023-10-02 22:04:06,515 - Epoch: [197][ 40/ 1236] Overall Loss 0.100638 Objective Loss 0.100638 LR 0.000031 Time 0.044357 +2023-10-02 22:04:06,725 - Epoch: [197][ 50/ 1236] Overall Loss 0.102380 Objective Loss 0.102380 LR 0.000031 Time 0.039689 +2023-10-02 22:04:06,938 - Epoch: [197][ 60/ 1236] Overall Loss 0.103331 Objective Loss 0.103331 LR 0.000031 Time 0.036611 +2023-10-02 22:04:07,148 - Epoch: [197][ 70/ 1236] Overall Loss 0.102559 Objective Loss 0.102559 LR 0.000031 Time 0.034382 +2023-10-02 22:04:07,361 - Epoch: [197][ 80/ 1236] Overall Loss 0.102661 Objective Loss 0.102661 LR 0.000031 Time 0.032733 +2023-10-02 22:04:07,571 - Epoch: [197][ 90/ 1236] Overall Loss 0.101539 Objective Loss 0.101539 LR 0.000031 Time 0.031430 +2023-10-02 22:04:07,782 - Epoch: [197][ 100/ 1236] Overall Loss 0.101096 Objective Loss 0.101096 LR 0.000031 Time 0.030396 +2023-10-02 22:04:07,991 - Epoch: [197][ 110/ 1236] Overall Loss 0.101259 Objective Loss 0.101259 LR 0.000031 Time 0.029519 +2023-10-02 22:04:08,202 - Epoch: [197][ 120/ 1236] Overall Loss 0.101541 Objective Loss 0.101541 LR 0.000031 Time 0.028816 +2023-10-02 22:04:08,412 - Epoch: [197][ 130/ 1236] Overall Loss 0.102339 Objective Loss 0.102339 LR 0.000031 Time 0.028202 +2023-10-02 22:04:08,624 - Epoch: [197][ 140/ 1236] Overall Loss 0.103804 Objective Loss 0.103804 LR 0.000031 Time 0.027697 +2023-10-02 22:04:08,834 - Epoch: [197][ 150/ 1236] Overall Loss 0.103279 Objective Loss 0.103279 LR 0.000031 Time 0.027244 +2023-10-02 22:04:09,048 - Epoch: [197][ 160/ 1236] Overall Loss 0.103352 Objective Loss 0.103352 LR 0.000031 Time 0.026876 +2023-10-02 22:04:09,257 - Epoch: [197][ 170/ 1236] Overall Loss 0.103367 Objective Loss 0.103367 LR 0.000031 Time 0.026524 +2023-10-02 22:04:09,470 - Epoch: [197][ 180/ 1236] Overall Loss 0.103209 Objective Loss 0.103209 LR 0.000031 Time 0.026230 +2023-10-02 22:04:09,681 - Epoch: [197][ 190/ 1236] Overall Loss 0.103889 Objective Loss 0.103889 LR 0.000031 Time 0.025950 +2023-10-02 22:04:09,892 - Epoch: [197][ 200/ 1236] Overall Loss 0.103494 Objective Loss 0.103494 LR 0.000031 Time 0.025710 +2023-10-02 22:04:10,103 - Epoch: [197][ 210/ 1236] Overall Loss 0.103113 Objective Loss 0.103113 LR 0.000031 Time 0.025481 +2023-10-02 22:04:10,316 - Epoch: [197][ 220/ 1236] Overall Loss 0.102967 Objective Loss 0.102967 LR 0.000031 Time 0.025288 +2023-10-02 22:04:10,533 - Epoch: [197][ 230/ 1236] Overall Loss 0.103062 Objective Loss 0.103062 LR 0.000031 Time 0.025129 +2023-10-02 22:04:10,757 - Epoch: [197][ 240/ 1236] Overall Loss 0.103347 Objective Loss 0.103347 LR 0.000031 Time 0.025010 +2023-10-02 22:04:10,970 - Epoch: [197][ 250/ 1236] Overall Loss 0.103127 Objective Loss 0.103127 LR 0.000031 Time 0.024863 +2023-10-02 22:04:11,185 - Epoch: [197][ 260/ 1236] Overall Loss 0.103068 Objective Loss 0.103068 LR 0.000031 Time 0.024728 +2023-10-02 22:04:11,404 - Epoch: [197][ 270/ 1236] Overall Loss 0.102925 Objective Loss 0.102925 LR 0.000031 Time 0.024620 +2023-10-02 22:04:11,625 - Epoch: [197][ 280/ 1236] Overall Loss 0.102691 Objective Loss 0.102691 LR 0.000031 Time 0.024530 +2023-10-02 22:04:11,842 - Epoch: [197][ 290/ 1236] Overall Loss 0.101922 Objective Loss 0.101922 LR 0.000031 Time 0.024431 +2023-10-02 22:04:12,058 - Epoch: [197][ 300/ 1236] Overall Loss 0.101925 Objective Loss 0.101925 LR 0.000031 Time 0.024330 +2023-10-02 22:04:12,270 - Epoch: [197][ 310/ 1236] Overall Loss 0.101839 Objective Loss 0.101839 LR 0.000031 Time 0.024230 +2023-10-02 22:04:12,483 - Epoch: [197][ 320/ 1236] Overall Loss 0.102001 Objective Loss 0.102001 LR 0.000031 Time 0.024137 +2023-10-02 22:04:12,695 - Epoch: [197][ 330/ 1236] Overall Loss 0.101840 Objective Loss 0.101840 LR 0.000031 Time 0.024047 +2023-10-02 22:04:12,905 - Epoch: [197][ 340/ 1236] Overall Loss 0.101682 Objective Loss 0.101682 LR 0.000031 Time 0.023956 +2023-10-02 22:04:13,115 - Epoch: [197][ 350/ 1236] Overall Loss 0.101754 Objective Loss 0.101754 LR 0.000031 Time 0.023871 +2023-10-02 22:04:13,325 - Epoch: [197][ 360/ 1236] Overall Loss 0.101512 Objective Loss 0.101512 LR 0.000031 Time 0.023786 +2023-10-02 22:04:13,535 - Epoch: [197][ 370/ 1236] Overall Loss 0.101565 Objective Loss 0.101565 LR 0.000031 Time 0.023711 +2023-10-02 22:04:13,744 - Epoch: [197][ 380/ 1236] Overall Loss 0.101664 Objective Loss 0.101664 LR 0.000031 Time 0.023633 +2023-10-02 22:04:13,956 - Epoch: [197][ 390/ 1236] Overall Loss 0.101760 Objective Loss 0.101760 LR 0.000031 Time 0.023569 +2023-10-02 22:04:14,164 - Epoch: [197][ 400/ 1236] Overall Loss 0.101450 Objective Loss 0.101450 LR 0.000031 Time 0.023499 +2023-10-02 22:04:14,374 - Epoch: [197][ 410/ 1236] Overall Loss 0.101449 Objective Loss 0.101449 LR 0.000031 Time 0.023438 +2023-10-02 22:04:14,584 - Epoch: [197][ 420/ 1236] Overall Loss 0.101813 Objective Loss 0.101813 LR 0.000031 Time 0.023375 +2023-10-02 22:04:14,794 - Epoch: [197][ 430/ 1236] Overall Loss 0.102114 Objective Loss 0.102114 LR 0.000031 Time 0.023319 +2023-10-02 22:04:15,003 - Epoch: [197][ 440/ 1236] Overall Loss 0.102098 Objective Loss 0.102098 LR 0.000031 Time 0.023260 +2023-10-02 22:04:15,213 - Epoch: [197][ 450/ 1236] Overall Loss 0.102341 Objective Loss 0.102341 LR 0.000031 Time 0.023209 +2023-10-02 22:04:15,422 - Epoch: [197][ 460/ 1236] Overall Loss 0.102073 Objective Loss 0.102073 LR 0.000031 Time 0.023156 +2023-10-02 22:04:15,633 - Epoch: [197][ 470/ 1236] Overall Loss 0.102353 Objective Loss 0.102353 LR 0.000031 Time 0.023111 +2023-10-02 22:04:15,842 - Epoch: [197][ 480/ 1236] Overall Loss 0.102341 Objective Loss 0.102341 LR 0.000031 Time 0.023062 +2023-10-02 22:04:16,052 - Epoch: [197][ 490/ 1236] Overall Loss 0.102147 Objective Loss 0.102147 LR 0.000031 Time 0.023020 +2023-10-02 22:04:16,262 - Epoch: [197][ 500/ 1236] Overall Loss 0.102092 Objective Loss 0.102092 LR 0.000031 Time 0.022976 +2023-10-02 22:04:16,472 - Epoch: [197][ 510/ 1236] Overall Loss 0.101973 Objective Loss 0.101973 LR 0.000031 Time 0.022937 +2023-10-02 22:04:16,682 - Epoch: [197][ 520/ 1236] Overall Loss 0.102301 Objective Loss 0.102301 LR 0.000031 Time 0.022896 +2023-10-02 22:04:16,892 - Epoch: [197][ 530/ 1236] Overall Loss 0.102085 Objective Loss 0.102085 LR 0.000031 Time 0.022861 +2023-10-02 22:04:17,101 - Epoch: [197][ 540/ 1236] Overall Loss 0.102524 Objective Loss 0.102524 LR 0.000031 Time 0.022822 +2023-10-02 22:04:17,312 - Epoch: [197][ 550/ 1236] Overall Loss 0.102402 Objective Loss 0.102402 LR 0.000031 Time 0.022789 +2023-10-02 22:04:17,521 - Epoch: [197][ 560/ 1236] Overall Loss 0.102462 Objective Loss 0.102462 LR 0.000031 Time 0.022753 +2023-10-02 22:04:17,732 - Epoch: [197][ 570/ 1236] Overall Loss 0.102589 Objective Loss 0.102589 LR 0.000031 Time 0.022724 +2023-10-02 22:04:17,945 - Epoch: [197][ 580/ 1236] Overall Loss 0.102602 Objective Loss 0.102602 LR 0.000031 Time 0.022697 +2023-10-02 22:04:18,156 - Epoch: [197][ 590/ 1236] Overall Loss 0.102585 Objective Loss 0.102585 LR 0.000031 Time 0.022670 +2023-10-02 22:04:18,368 - Epoch: [197][ 600/ 1236] Overall Loss 0.102536 Objective Loss 0.102536 LR 0.000031 Time 0.022643 +2023-10-02 22:04:18,579 - Epoch: [197][ 610/ 1236] Overall Loss 0.102528 Objective Loss 0.102528 LR 0.000031 Time 0.022617 +2023-10-02 22:04:18,792 - Epoch: [197][ 620/ 1236] Overall Loss 0.102882 Objective Loss 0.102882 LR 0.000031 Time 0.022593 +2023-10-02 22:04:19,003 - Epoch: [197][ 630/ 1236] Overall Loss 0.102979 Objective Loss 0.102979 LR 0.000031 Time 0.022569 +2023-10-02 22:04:19,215 - Epoch: [197][ 640/ 1236] Overall Loss 0.102999 Objective Loss 0.102999 LR 0.000031 Time 0.022545 +2023-10-02 22:04:19,426 - Epoch: [197][ 650/ 1236] Overall Loss 0.102884 Objective Loss 0.102884 LR 0.000031 Time 0.022523 +2023-10-02 22:04:19,639 - Epoch: [197][ 660/ 1236] Overall Loss 0.102857 Objective Loss 0.102857 LR 0.000031 Time 0.022502 +2023-10-02 22:04:19,850 - Epoch: [197][ 670/ 1236] Overall Loss 0.102902 Objective Loss 0.102902 LR 0.000031 Time 0.022481 +2023-10-02 22:04:20,062 - Epoch: [197][ 680/ 1236] Overall Loss 0.102618 Objective Loss 0.102618 LR 0.000031 Time 0.022460 +2023-10-02 22:04:20,273 - Epoch: [197][ 690/ 1236] Overall Loss 0.102545 Objective Loss 0.102545 LR 0.000031 Time 0.022440 +2023-10-02 22:04:20,485 - Epoch: [197][ 700/ 1236] Overall Loss 0.102824 Objective Loss 0.102824 LR 0.000031 Time 0.022421 +2023-10-02 22:04:20,697 - Epoch: [197][ 710/ 1236] Overall Loss 0.102822 Objective Loss 0.102822 LR 0.000031 Time 0.022402 +2023-10-02 22:04:20,909 - Epoch: [197][ 720/ 1236] Overall Loss 0.103147 Objective Loss 0.103147 LR 0.000031 Time 0.022385 +2023-10-02 22:04:21,120 - Epoch: [197][ 730/ 1236] Overall Loss 0.103221 Objective Loss 0.103221 LR 0.000031 Time 0.022367 +2023-10-02 22:04:21,332 - Epoch: [197][ 740/ 1236] Overall Loss 0.102951 Objective Loss 0.102951 LR 0.000031 Time 0.022349 +2023-10-02 22:04:21,543 - Epoch: [197][ 750/ 1236] Overall Loss 0.102953 Objective Loss 0.102953 LR 0.000031 Time 0.022333 +2023-10-02 22:04:21,756 - Epoch: [197][ 760/ 1236] Overall Loss 0.103136 Objective Loss 0.103136 LR 0.000031 Time 0.022317 +2023-10-02 22:04:21,966 - Epoch: [197][ 770/ 1236] Overall Loss 0.103045 Objective Loss 0.103045 LR 0.000031 Time 0.022300 +2023-10-02 22:04:22,178 - Epoch: [197][ 780/ 1236] Overall Loss 0.102789 Objective Loss 0.102789 LR 0.000031 Time 0.022286 +2023-10-02 22:04:22,389 - Epoch: [197][ 790/ 1236] Overall Loss 0.102913 Objective Loss 0.102913 LR 0.000031 Time 0.022270 +2023-10-02 22:04:22,601 - Epoch: [197][ 800/ 1236] Overall Loss 0.103092 Objective Loss 0.103092 LR 0.000031 Time 0.022256 +2023-10-02 22:04:22,813 - Epoch: [197][ 810/ 1236] Overall Loss 0.103024 Objective Loss 0.103024 LR 0.000031 Time 0.022242 +2023-10-02 22:04:23,024 - Epoch: [197][ 820/ 1236] Overall Loss 0.103225 Objective Loss 0.103225 LR 0.000031 Time 0.022228 +2023-10-02 22:04:23,236 - Epoch: [197][ 830/ 1236] Overall Loss 0.103278 Objective Loss 0.103278 LR 0.000031 Time 0.022215 +2023-10-02 22:04:23,447 - Epoch: [197][ 840/ 1236] Overall Loss 0.103191 Objective Loss 0.103191 LR 0.000031 Time 0.022202 +2023-10-02 22:04:23,658 - Epoch: [197][ 850/ 1236] Overall Loss 0.103163 Objective Loss 0.103163 LR 0.000031 Time 0.022188 +2023-10-02 22:04:23,870 - Epoch: [197][ 860/ 1236] Overall Loss 0.102918 Objective Loss 0.102918 LR 0.000031 Time 0.022176 +2023-10-02 22:04:24,081 - Epoch: [197][ 870/ 1236] Overall Loss 0.102822 Objective Loss 0.102822 LR 0.000031 Time 0.022164 +2023-10-02 22:04:24,293 - Epoch: [197][ 880/ 1236] Overall Loss 0.102844 Objective Loss 0.102844 LR 0.000031 Time 0.022152 +2023-10-02 22:04:24,504 - Epoch: [197][ 890/ 1236] Overall Loss 0.102846 Objective Loss 0.102846 LR 0.000031 Time 0.022140 +2023-10-02 22:04:24,717 - Epoch: [197][ 900/ 1236] Overall Loss 0.102756 Objective Loss 0.102756 LR 0.000031 Time 0.022130 +2023-10-02 22:04:24,928 - Epoch: [197][ 910/ 1236] Overall Loss 0.102900 Objective Loss 0.102900 LR 0.000031 Time 0.022118 +2023-10-02 22:04:25,140 - Epoch: [197][ 920/ 1236] Overall Loss 0.102910 Objective Loss 0.102910 LR 0.000031 Time 0.022108 +2023-10-02 22:04:25,351 - Epoch: [197][ 930/ 1236] Overall Loss 0.103046 Objective Loss 0.103046 LR 0.000031 Time 0.022097 +2023-10-02 22:04:25,563 - Epoch: [197][ 940/ 1236] Overall Loss 0.102902 Objective Loss 0.102902 LR 0.000031 Time 0.022087 +2023-10-02 22:04:25,775 - Epoch: [197][ 950/ 1236] Overall Loss 0.102929 Objective Loss 0.102929 LR 0.000031 Time 0.022077 +2023-10-02 22:04:25,987 - Epoch: [197][ 960/ 1236] Overall Loss 0.103017 Objective Loss 0.103017 LR 0.000031 Time 0.022067 +2023-10-02 22:04:26,198 - Epoch: [197][ 970/ 1236] Overall Loss 0.102945 Objective Loss 0.102945 LR 0.000031 Time 0.022057 +2023-10-02 22:04:26,410 - Epoch: [197][ 980/ 1236] Overall Loss 0.102785 Objective Loss 0.102785 LR 0.000031 Time 0.022048 +2023-10-02 22:04:26,621 - Epoch: [197][ 990/ 1236] Overall Loss 0.102945 Objective Loss 0.102945 LR 0.000031 Time 0.022038 +2023-10-02 22:04:26,833 - Epoch: [197][ 1000/ 1236] Overall Loss 0.102969 Objective Loss 0.102969 LR 0.000031 Time 0.022029 +2023-10-02 22:04:27,044 - Epoch: [197][ 1010/ 1236] Overall Loss 0.103107 Objective Loss 0.103107 LR 0.000031 Time 0.022019 +2023-10-02 22:04:27,256 - Epoch: [197][ 1020/ 1236] Overall Loss 0.103055 Objective Loss 0.103055 LR 0.000031 Time 0.022011 +2023-10-02 22:04:27,467 - Epoch: [197][ 1030/ 1236] Overall Loss 0.103194 Objective Loss 0.103194 LR 0.000031 Time 0.022002 +2023-10-02 22:04:27,678 - Epoch: [197][ 1040/ 1236] Overall Loss 0.103031 Objective Loss 0.103031 LR 0.000031 Time 0.021993 +2023-10-02 22:04:27,889 - Epoch: [197][ 1050/ 1236] Overall Loss 0.103073 Objective Loss 0.103073 LR 0.000031 Time 0.021984 +2023-10-02 22:04:28,102 - Epoch: [197][ 1060/ 1236] Overall Loss 0.103076 Objective Loss 0.103076 LR 0.000031 Time 0.021977 +2023-10-02 22:04:28,313 - Epoch: [197][ 1070/ 1236] Overall Loss 0.103027 Objective Loss 0.103027 LR 0.000031 Time 0.021969 +2023-10-02 22:04:28,525 - Epoch: [197][ 1080/ 1236] Overall Loss 0.102893 Objective Loss 0.102893 LR 0.000031 Time 0.021962 +2023-10-02 22:04:28,737 - Epoch: [197][ 1090/ 1236] Overall Loss 0.102966 Objective Loss 0.102966 LR 0.000031 Time 0.021954 +2023-10-02 22:04:28,949 - Epoch: [197][ 1100/ 1236] Overall Loss 0.103009 Objective Loss 0.103009 LR 0.000031 Time 0.021947 +2023-10-02 22:04:29,161 - Epoch: [197][ 1110/ 1236] Overall Loss 0.102899 Objective Loss 0.102899 LR 0.000031 Time 0.021939 +2023-10-02 22:04:29,373 - Epoch: [197][ 1120/ 1236] Overall Loss 0.102831 Objective Loss 0.102831 LR 0.000031 Time 0.021932 +2023-10-02 22:04:29,584 - Epoch: [197][ 1130/ 1236] Overall Loss 0.103003 Objective Loss 0.103003 LR 0.000031 Time 0.021925 +2023-10-02 22:04:29,796 - Epoch: [197][ 1140/ 1236] Overall Loss 0.103078 Objective Loss 0.103078 LR 0.000031 Time 0.021918 +2023-10-02 22:04:30,008 - Epoch: [197][ 1150/ 1236] Overall Loss 0.103068 Objective Loss 0.103068 LR 0.000031 Time 0.021911 +2023-10-02 22:04:30,219 - Epoch: [197][ 1160/ 1236] Overall Loss 0.103214 Objective Loss 0.103214 LR 0.000031 Time 0.021905 +2023-10-02 22:04:30,430 - Epoch: [197][ 1170/ 1236] Overall Loss 0.103216 Objective Loss 0.103216 LR 0.000031 Time 0.021897 +2023-10-02 22:04:30,642 - Epoch: [197][ 1180/ 1236] Overall Loss 0.103414 Objective Loss 0.103414 LR 0.000031 Time 0.021891 +2023-10-02 22:04:30,854 - Epoch: [197][ 1190/ 1236] Overall Loss 0.103838 Objective Loss 0.103838 LR 0.000031 Time 0.021884 +2023-10-02 22:04:31,066 - Epoch: [197][ 1200/ 1236] Overall Loss 0.103846 Objective Loss 0.103846 LR 0.000031 Time 0.021879 +2023-10-02 22:04:31,277 - Epoch: [197][ 1210/ 1236] Overall Loss 0.103807 Objective Loss 0.103807 LR 0.000031 Time 0.021872 +2023-10-02 22:04:31,490 - Epoch: [197][ 1220/ 1236] Overall Loss 0.103766 Objective Loss 0.103766 LR 0.000031 Time 0.021867 +2023-10-02 22:04:31,756 - Epoch: [197][ 1230/ 1236] Overall Loss 0.103716 Objective Loss 0.103716 LR 0.000031 Time 0.021905 +2023-10-02 22:04:31,879 - Epoch: [197][ 1236/ 1236] Overall Loss 0.103712 Objective Loss 0.103712 Top1 91.446029 Top5 99.389002 LR 0.000031 Time 0.021898 +2023-10-02 22:04:32,012 - --- validate (epoch=197)----------- +2023-10-02 22:04:32,012 - 29943 samples (256 per mini-batch) +2023-10-02 22:04:32,513 - Epoch: [197][ 10/ 117] Loss 0.260637 Top1 88.007812 Top5 98.750000 +2023-10-02 22:04:32,677 - Epoch: [197][ 20/ 117] Loss 0.283251 Top1 87.988281 Top5 98.730469 +2023-10-02 22:04:32,838 - Epoch: [197][ 30/ 117] Loss 0.290530 Top1 87.812500 Top5 98.658854 +2023-10-02 22:04:33,000 - Epoch: [197][ 40/ 117] Loss 0.296199 Top1 87.822266 Top5 98.701172 +2023-10-02 22:04:33,159 - Epoch: [197][ 50/ 117] Loss 0.304720 Top1 87.757812 Top5 98.632812 +2023-10-02 22:04:33,320 - Epoch: [197][ 60/ 117] Loss 0.303529 Top1 87.721354 Top5 98.632812 +2023-10-02 22:04:33,478 - Epoch: [197][ 70/ 117] Loss 0.303865 Top1 87.734375 Top5 98.638393 +2023-10-02 22:04:33,640 - Epoch: [197][ 80/ 117] Loss 0.302702 Top1 87.802734 Top5 98.657227 +2023-10-02 22:04:33,798 - Epoch: [197][ 90/ 117] Loss 0.305047 Top1 87.825521 Top5 98.671875 +2023-10-02 22:04:33,959 - Epoch: [197][ 100/ 117] Loss 0.305843 Top1 87.738281 Top5 98.703125 +2023-10-02 22:04:34,126 - Epoch: [197][ 110/ 117] Loss 0.303268 Top1 87.752131 Top5 98.710938 +2023-10-02 22:04:34,216 - Epoch: [197][ 117/ 117] Loss 0.299764 Top1 87.786795 Top5 98.747620 +2023-10-02 22:04:34,351 - ==> Top1: 87.787 Top5: 98.748 Loss: 0.300 + +2023-10-02 22:04:34,352 - ==> Confusion: +[[ 947 0 2 1 2 2 0 0 7 54 2 1 1 0 6 0 2 0 2 0 21] + [ 0 1068 0 0 4 15 0 22 1 1 0 0 0 0 0 3 1 0 6 3 7] + [ 1 1 978 9 1 0 18 10 0 2 2 0 6 2 1 2 2 1 10 2 8] + [ 1 3 14 988 0 0 1 1 3 1 4 1 4 2 24 2 1 5 13 1 20] + [ 25 4 0 1 973 4 1 0 0 12 0 0 0 3 9 5 7 0 0 1 5] + [ 3 31 0 1 6 998 1 22 1 5 1 6 1 9 4 0 1 0 4 1 21] + [ 0 3 19 1 0 2 1139 6 0 0 3 1 0 0 0 4 0 1 1 6 5] + [ 2 8 12 0 3 21 6 1091 1 2 4 2 4 4 1 1 0 1 36 8 11] + [ 14 4 0 1 1 4 0 1 989 30 10 2 1 7 14 0 4 1 3 0 3] + [ 96 1 0 2 8 2 0 0 28 950 1 0 0 17 4 2 1 0 0 1 6] + [ 3 2 6 6 0 2 2 3 11 3 980 2 0 5 6 0 2 1 5 3 11] + [ 0 0 1 0 1 11 0 6 0 0 0 975 10 5 0 2 2 14 0 3 5] + [ 0 1 1 1 0 1 1 1 0 2 2 33 982 1 2 7 0 9 3 5 16] + [ 0 0 1 0 4 7 0 0 13 8 3 7 0 1054 4 1 0 0 0 0 17] + [ 11 0 3 14 8 1 0 0 22 2 1 0 1 3 1015 0 0 1 10 0 9] + [ 0 0 2 1 5 1 0 0 0 0 0 6 7 0 0 1071 18 9 2 7 5] + [ 0 17 1 0 5 6 0 0 0 0 0 3 0 2 4 7 1096 0 2 6 12] + [ 0 1 1 2 0 0 2 0 0 0 0 4 19 0 2 5 0 995 0 3 4] + [ 1 5 2 12 1 0 0 22 3 1 1 0 0 0 11 0 0 0 998 0 11] + [ 0 0 3 3 2 2 7 5 0 0 0 11 5 2 1 1 7 1 0 1093 9] + [ 96 115 96 76 52 122 27 96 78 50 126 75 273 210 111 45 70 44 108 129 5906]] + +2023-10-02 22:04:34,354 - ==> Best [Top1: 87.997 Top5: 98.754 Sparsity:0.00 Params: 169472 on epoch: 194] +2023-10-02 22:04:34,354 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:04:34,360 - + +2023-10-02 22:04:34,360 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:04:35,395 - Epoch: [198][ 10/ 1236] Overall Loss 0.106623 Objective Loss 0.106623 LR 0.000031 Time 0.103492 +2023-10-02 22:04:35,605 - Epoch: [198][ 20/ 1236] Overall Loss 0.105672 Objective Loss 0.105672 LR 0.000031 Time 0.062200 +2023-10-02 22:04:35,814 - Epoch: [198][ 30/ 1236] Overall Loss 0.104609 Objective Loss 0.104609 LR 0.000031 Time 0.048428 +2023-10-02 22:04:36,024 - Epoch: [198][ 40/ 1236] Overall Loss 0.106189 Objective Loss 0.106189 LR 0.000031 Time 0.041572 +2023-10-02 22:04:36,234 - Epoch: [198][ 50/ 1236] Overall Loss 0.101846 Objective Loss 0.101846 LR 0.000031 Time 0.037437 +2023-10-02 22:04:36,444 - Epoch: [198][ 60/ 1236] Overall Loss 0.099526 Objective Loss 0.099526 LR 0.000031 Time 0.034695 +2023-10-02 22:04:36,653 - Epoch: [198][ 70/ 1236] Overall Loss 0.099669 Objective Loss 0.099669 LR 0.000031 Time 0.032702 +2023-10-02 22:04:36,863 - Epoch: [198][ 80/ 1236] Overall Loss 0.096847 Objective Loss 0.096847 LR 0.000031 Time 0.031239 +2023-10-02 22:04:37,071 - Epoch: [198][ 90/ 1236] Overall Loss 0.096519 Objective Loss 0.096519 LR 0.000031 Time 0.030076 +2023-10-02 22:04:37,281 - Epoch: [198][ 100/ 1236] Overall Loss 0.097991 Objective Loss 0.097991 LR 0.000031 Time 0.029162 +2023-10-02 22:04:37,489 - Epoch: [198][ 110/ 1236] Overall Loss 0.100034 Objective Loss 0.100034 LR 0.000031 Time 0.028388 +2023-10-02 22:04:37,699 - Epoch: [198][ 120/ 1236] Overall Loss 0.099948 Objective Loss 0.099948 LR 0.000031 Time 0.027769 +2023-10-02 22:04:37,906 - Epoch: [198][ 130/ 1236] Overall Loss 0.101493 Objective Loss 0.101493 LR 0.000031 Time 0.027225 +2023-10-02 22:04:38,116 - Epoch: [198][ 140/ 1236] Overall Loss 0.102998 Objective Loss 0.102998 LR 0.000031 Time 0.026777 +2023-10-02 22:04:38,323 - Epoch: [198][ 150/ 1236] Overall Loss 0.102608 Objective Loss 0.102608 LR 0.000031 Time 0.026373 +2023-10-02 22:04:38,533 - Epoch: [198][ 160/ 1236] Overall Loss 0.102963 Objective Loss 0.102963 LR 0.000031 Time 0.026034 +2023-10-02 22:04:38,741 - Epoch: [198][ 170/ 1236] Overall Loss 0.102506 Objective Loss 0.102506 LR 0.000031 Time 0.025721 +2023-10-02 22:04:38,951 - Epoch: [198][ 180/ 1236] Overall Loss 0.101677 Objective Loss 0.101677 LR 0.000031 Time 0.025459 +2023-10-02 22:04:39,158 - Epoch: [198][ 190/ 1236] Overall Loss 0.102420 Objective Loss 0.102420 LR 0.000031 Time 0.025207 +2023-10-02 22:04:39,368 - Epoch: [198][ 200/ 1236] Overall Loss 0.102825 Objective Loss 0.102825 LR 0.000031 Time 0.024994 +2023-10-02 22:04:39,575 - Epoch: [198][ 210/ 1236] Overall Loss 0.103376 Objective Loss 0.103376 LR 0.000031 Time 0.024788 +2023-10-02 22:04:39,785 - Epoch: [198][ 220/ 1236] Overall Loss 0.103242 Objective Loss 0.103242 LR 0.000031 Time 0.024615 +2023-10-02 22:04:39,993 - Epoch: [198][ 230/ 1236] Overall Loss 0.102924 Objective Loss 0.102924 LR 0.000031 Time 0.024446 +2023-10-02 22:04:40,202 - Epoch: [198][ 240/ 1236] Overall Loss 0.102807 Objective Loss 0.102807 LR 0.000031 Time 0.024300 +2023-10-02 22:04:40,410 - Epoch: [198][ 250/ 1236] Overall Loss 0.102456 Objective Loss 0.102456 LR 0.000031 Time 0.024158 +2023-10-02 22:04:40,620 - Epoch: [198][ 260/ 1236] Overall Loss 0.103220 Objective Loss 0.103220 LR 0.000031 Time 0.024035 +2023-10-02 22:04:40,830 - Epoch: [198][ 270/ 1236] Overall Loss 0.103536 Objective Loss 0.103536 LR 0.000031 Time 0.023923 +2023-10-02 22:04:41,041 - Epoch: [198][ 280/ 1236] Overall Loss 0.103802 Objective Loss 0.103802 LR 0.000031 Time 0.023819 +2023-10-02 22:04:41,247 - Epoch: [198][ 290/ 1236] Overall Loss 0.104017 Objective Loss 0.104017 LR 0.000031 Time 0.023709 +2023-10-02 22:04:41,456 - Epoch: [198][ 300/ 1236] Overall Loss 0.103796 Objective Loss 0.103796 LR 0.000031 Time 0.023612 +2023-10-02 22:04:41,662 - Epoch: [198][ 310/ 1236] Overall Loss 0.104138 Objective Loss 0.104138 LR 0.000031 Time 0.023516 +2023-10-02 22:04:41,870 - Epoch: [198][ 320/ 1236] Overall Loss 0.104769 Objective Loss 0.104769 LR 0.000031 Time 0.023429 +2023-10-02 22:04:42,077 - Epoch: [198][ 330/ 1236] Overall Loss 0.104970 Objective Loss 0.104970 LR 0.000031 Time 0.023344 +2023-10-02 22:04:42,285 - Epoch: [198][ 340/ 1236] Overall Loss 0.105466 Objective Loss 0.105466 LR 0.000031 Time 0.023269 +2023-10-02 22:04:42,492 - Epoch: [198][ 350/ 1236] Overall Loss 0.105148 Objective Loss 0.105148 LR 0.000031 Time 0.023194 +2023-10-02 22:04:42,700 - Epoch: [198][ 360/ 1236] Overall Loss 0.104857 Objective Loss 0.104857 LR 0.000031 Time 0.023127 +2023-10-02 22:04:42,906 - Epoch: [198][ 370/ 1236] Overall Loss 0.104405 Objective Loss 0.104405 LR 0.000031 Time 0.023059 +2023-10-02 22:04:43,115 - Epoch: [198][ 380/ 1236] Overall Loss 0.104115 Objective Loss 0.104115 LR 0.000031 Time 0.023001 +2023-10-02 22:04:43,321 - Epoch: [198][ 390/ 1236] Overall Loss 0.104330 Objective Loss 0.104330 LR 0.000031 Time 0.022940 +2023-10-02 22:04:43,530 - Epoch: [198][ 400/ 1236] Overall Loss 0.103833 Objective Loss 0.103833 LR 0.000031 Time 0.022887 +2023-10-02 22:04:43,737 - Epoch: [198][ 410/ 1236] Overall Loss 0.103705 Objective Loss 0.103705 LR 0.000031 Time 0.022831 +2023-10-02 22:04:43,945 - Epoch: [198][ 420/ 1236] Overall Loss 0.103712 Objective Loss 0.103712 LR 0.000031 Time 0.022784 +2023-10-02 22:04:44,151 - Epoch: [198][ 430/ 1236] Overall Loss 0.103791 Objective Loss 0.103791 LR 0.000031 Time 0.022733 +2023-10-02 22:04:44,360 - Epoch: [198][ 440/ 1236] Overall Loss 0.103662 Objective Loss 0.103662 LR 0.000031 Time 0.022689 +2023-10-02 22:04:44,567 - Epoch: [198][ 450/ 1236] Overall Loss 0.103852 Objective Loss 0.103852 LR 0.000031 Time 0.022643 +2023-10-02 22:04:44,775 - Epoch: [198][ 460/ 1236] Overall Loss 0.103868 Objective Loss 0.103868 LR 0.000031 Time 0.022603 +2023-10-02 22:04:44,981 - Epoch: [198][ 470/ 1236] Overall Loss 0.103800 Objective Loss 0.103800 LR 0.000031 Time 0.022561 +2023-10-02 22:04:45,190 - Epoch: [198][ 480/ 1236] Overall Loss 0.103732 Objective Loss 0.103732 LR 0.000031 Time 0.022525 +2023-10-02 22:04:45,397 - Epoch: [198][ 490/ 1236] Overall Loss 0.103793 Objective Loss 0.103793 LR 0.000031 Time 0.022487 +2023-10-02 22:04:45,605 - Epoch: [198][ 500/ 1236] Overall Loss 0.104251 Objective Loss 0.104251 LR 0.000031 Time 0.022453 +2023-10-02 22:04:45,812 - Epoch: [198][ 510/ 1236] Overall Loss 0.104380 Objective Loss 0.104380 LR 0.000031 Time 0.022417 +2023-10-02 22:04:46,020 - Epoch: [198][ 520/ 1236] Overall Loss 0.104348 Objective Loss 0.104348 LR 0.000031 Time 0.022386 +2023-10-02 22:04:46,227 - Epoch: [198][ 530/ 1236] Overall Loss 0.104413 Objective Loss 0.104413 LR 0.000031 Time 0.022353 +2023-10-02 22:04:46,435 - Epoch: [198][ 540/ 1236] Overall Loss 0.104416 Objective Loss 0.104416 LR 0.000031 Time 0.022324 +2023-10-02 22:04:46,642 - Epoch: [198][ 550/ 1236] Overall Loss 0.104305 Objective Loss 0.104305 LR 0.000031 Time 0.022294 +2023-10-02 22:04:46,850 - Epoch: [198][ 560/ 1236] Overall Loss 0.104266 Objective Loss 0.104266 LR 0.000031 Time 0.022267 +2023-10-02 22:04:47,057 - Epoch: [198][ 570/ 1236] Overall Loss 0.103881 Objective Loss 0.103881 LR 0.000031 Time 0.022239 +2023-10-02 22:04:47,265 - Epoch: [198][ 580/ 1236] Overall Loss 0.103875 Objective Loss 0.103875 LR 0.000031 Time 0.022213 +2023-10-02 22:04:47,472 - Epoch: [198][ 590/ 1236] Overall Loss 0.103859 Objective Loss 0.103859 LR 0.000031 Time 0.022187 +2023-10-02 22:04:47,681 - Epoch: [198][ 600/ 1236] Overall Loss 0.103844 Objective Loss 0.103844 LR 0.000031 Time 0.022165 +2023-10-02 22:04:47,887 - Epoch: [198][ 610/ 1236] Overall Loss 0.103711 Objective Loss 0.103711 LR 0.000031 Time 0.022139 +2023-10-02 22:04:48,095 - Epoch: [198][ 620/ 1236] Overall Loss 0.103766 Objective Loss 0.103766 LR 0.000031 Time 0.022117 +2023-10-02 22:04:48,302 - Epoch: [198][ 630/ 1236] Overall Loss 0.103865 Objective Loss 0.103865 LR 0.000031 Time 0.022094 +2023-10-02 22:04:48,510 - Epoch: [198][ 640/ 1236] Overall Loss 0.103798 Objective Loss 0.103798 LR 0.000031 Time 0.022073 +2023-10-02 22:04:48,717 - Epoch: [198][ 650/ 1236] Overall Loss 0.103834 Objective Loss 0.103834 LR 0.000031 Time 0.022052 +2023-10-02 22:04:48,925 - Epoch: [198][ 660/ 1236] Overall Loss 0.103781 Objective Loss 0.103781 LR 0.000031 Time 0.022033 +2023-10-02 22:04:49,132 - Epoch: [198][ 670/ 1236] Overall Loss 0.103876 Objective Loss 0.103876 LR 0.000031 Time 0.022012 +2023-10-02 22:04:49,340 - Epoch: [198][ 680/ 1236] Overall Loss 0.103702 Objective Loss 0.103702 LR 0.000031 Time 0.021994 +2023-10-02 22:04:49,547 - Epoch: [198][ 690/ 1236] Overall Loss 0.103559 Objective Loss 0.103559 LR 0.000031 Time 0.021974 +2023-10-02 22:04:49,755 - Epoch: [198][ 700/ 1236] Overall Loss 0.103535 Objective Loss 0.103535 LR 0.000031 Time 0.021958 +2023-10-02 22:04:49,962 - Epoch: [198][ 710/ 1236] Overall Loss 0.103334 Objective Loss 0.103334 LR 0.000031 Time 0.021939 +2023-10-02 22:04:50,170 - Epoch: [198][ 720/ 1236] Overall Loss 0.103255 Objective Loss 0.103255 LR 0.000031 Time 0.021923 +2023-10-02 22:04:50,377 - Epoch: [198][ 730/ 1236] Overall Loss 0.103377 Objective Loss 0.103377 LR 0.000031 Time 0.021905 +2023-10-02 22:04:50,585 - Epoch: [198][ 740/ 1236] Overall Loss 0.103266 Objective Loss 0.103266 LR 0.000031 Time 0.021891 +2023-10-02 22:04:50,792 - Epoch: [198][ 750/ 1236] Overall Loss 0.103110 Objective Loss 0.103110 LR 0.000031 Time 0.021874 +2023-10-02 22:04:51,001 - Epoch: [198][ 760/ 1236] Overall Loss 0.102925 Objective Loss 0.102925 LR 0.000031 Time 0.021861 +2023-10-02 22:04:51,207 - Epoch: [198][ 770/ 1236] Overall Loss 0.102981 Objective Loss 0.102981 LR 0.000031 Time 0.021844 +2023-10-02 22:04:51,415 - Epoch: [198][ 780/ 1236] Overall Loss 0.103160 Objective Loss 0.103160 LR 0.000031 Time 0.021831 +2023-10-02 22:04:51,622 - Epoch: [198][ 790/ 1236] Overall Loss 0.102992 Objective Loss 0.102992 LR 0.000031 Time 0.021816 +2023-10-02 22:04:51,831 - Epoch: [198][ 800/ 1236] Overall Loss 0.102971 Objective Loss 0.102971 LR 0.000031 Time 0.021804 +2023-10-02 22:04:52,038 - Epoch: [198][ 810/ 1236] Overall Loss 0.103073 Objective Loss 0.103073 LR 0.000031 Time 0.021790 +2023-10-02 22:04:52,246 - Epoch: [198][ 820/ 1236] Overall Loss 0.103268 Objective Loss 0.103268 LR 0.000031 Time 0.021777 +2023-10-02 22:04:52,453 - Epoch: [198][ 830/ 1236] Overall Loss 0.103062 Objective Loss 0.103062 LR 0.000031 Time 0.021764 +2023-10-02 22:04:52,661 - Epoch: [198][ 840/ 1236] Overall Loss 0.102960 Objective Loss 0.102960 LR 0.000031 Time 0.021752 +2023-10-02 22:04:52,868 - Epoch: [198][ 850/ 1236] Overall Loss 0.102954 Objective Loss 0.102954 LR 0.000031 Time 0.021739 +2023-10-02 22:04:53,076 - Epoch: [198][ 860/ 1236] Overall Loss 0.102947 Objective Loss 0.102947 LR 0.000031 Time 0.021728 +2023-10-02 22:04:53,283 - Epoch: [198][ 870/ 1236] Overall Loss 0.102941 Objective Loss 0.102941 LR 0.000031 Time 0.021716 +2023-10-02 22:04:53,491 - Epoch: [198][ 880/ 1236] Overall Loss 0.103061 Objective Loss 0.103061 LR 0.000031 Time 0.021705 +2023-10-02 22:04:53,698 - Epoch: [198][ 890/ 1236] Overall Loss 0.103070 Objective Loss 0.103070 LR 0.000031 Time 0.021693 +2023-10-02 22:04:53,906 - Epoch: [198][ 900/ 1236] Overall Loss 0.103357 Objective Loss 0.103357 LR 0.000031 Time 0.021683 +2023-10-02 22:04:54,113 - Epoch: [198][ 910/ 1236] Overall Loss 0.103369 Objective Loss 0.103369 LR 0.000031 Time 0.021672 +2023-10-02 22:04:54,321 - Epoch: [198][ 920/ 1236] Overall Loss 0.103529 Objective Loss 0.103529 LR 0.000031 Time 0.021663 +2023-10-02 22:04:54,528 - Epoch: [198][ 930/ 1236] Overall Loss 0.103500 Objective Loss 0.103500 LR 0.000031 Time 0.021652 +2023-10-02 22:04:54,736 - Epoch: [198][ 940/ 1236] Overall Loss 0.103296 Objective Loss 0.103296 LR 0.000031 Time 0.021642 +2023-10-02 22:04:54,943 - Epoch: [198][ 950/ 1236] Overall Loss 0.103330 Objective Loss 0.103330 LR 0.000031 Time 0.021632 +2023-10-02 22:04:55,151 - Epoch: [198][ 960/ 1236] Overall Loss 0.103306 Objective Loss 0.103306 LR 0.000031 Time 0.021623 +2023-10-02 22:04:55,358 - Epoch: [198][ 970/ 1236] Overall Loss 0.103408 Objective Loss 0.103408 LR 0.000031 Time 0.021613 +2023-10-02 22:04:55,566 - Epoch: [198][ 980/ 1236] Overall Loss 0.103389 Objective Loss 0.103389 LR 0.000031 Time 0.021605 +2023-10-02 22:04:55,773 - Epoch: [198][ 990/ 1236] Overall Loss 0.103455 Objective Loss 0.103455 LR 0.000031 Time 0.021595 +2023-10-02 22:04:55,981 - Epoch: [198][ 1000/ 1236] Overall Loss 0.103589 Objective Loss 0.103589 LR 0.000031 Time 0.021586 +2023-10-02 22:04:56,188 - Epoch: [198][ 1010/ 1236] Overall Loss 0.103474 Objective Loss 0.103474 LR 0.000031 Time 0.021577 +2023-10-02 22:04:56,396 - Epoch: [198][ 1020/ 1236] Overall Loss 0.103561 Objective Loss 0.103561 LR 0.000031 Time 0.021569 +2023-10-02 22:04:56,603 - Epoch: [198][ 1030/ 1236] Overall Loss 0.103495 Objective Loss 0.103495 LR 0.000031 Time 0.021561 +2023-10-02 22:04:56,811 - Epoch: [198][ 1040/ 1236] Overall Loss 0.103431 Objective Loss 0.103431 LR 0.000031 Time 0.021553 +2023-10-02 22:04:57,018 - Epoch: [198][ 1050/ 1236] Overall Loss 0.103446 Objective Loss 0.103446 LR 0.000031 Time 0.021545 +2023-10-02 22:04:57,226 - Epoch: [198][ 1060/ 1236] Overall Loss 0.103526 Objective Loss 0.103526 LR 0.000031 Time 0.021538 +2023-10-02 22:04:57,433 - Epoch: [198][ 1070/ 1236] Overall Loss 0.103396 Objective Loss 0.103396 LR 0.000031 Time 0.021529 +2023-10-02 22:04:57,641 - Epoch: [198][ 1080/ 1236] Overall Loss 0.103317 Objective Loss 0.103317 LR 0.000031 Time 0.021522 +2023-10-02 22:04:57,848 - Epoch: [198][ 1090/ 1236] Overall Loss 0.103184 Objective Loss 0.103184 LR 0.000031 Time 0.021515 +2023-10-02 22:04:58,056 - Epoch: [198][ 1100/ 1236] Overall Loss 0.103242 Objective Loss 0.103242 LR 0.000031 Time 0.021508 +2023-10-02 22:04:58,263 - Epoch: [198][ 1110/ 1236] Overall Loss 0.103256 Objective Loss 0.103256 LR 0.000031 Time 0.021500 +2023-10-02 22:04:58,471 - Epoch: [198][ 1120/ 1236] Overall Loss 0.103267 Objective Loss 0.103267 LR 0.000031 Time 0.021494 +2023-10-02 22:04:58,678 - Epoch: [198][ 1130/ 1236] Overall Loss 0.103197 Objective Loss 0.103197 LR 0.000031 Time 0.021486 +2023-10-02 22:04:58,886 - Epoch: [198][ 1140/ 1236] Overall Loss 0.103028 Objective Loss 0.103028 LR 0.000031 Time 0.021480 +2023-10-02 22:04:59,093 - Epoch: [198][ 1150/ 1236] Overall Loss 0.102990 Objective Loss 0.102990 LR 0.000031 Time 0.021473 +2023-10-02 22:04:59,301 - Epoch: [198][ 1160/ 1236] Overall Loss 0.103099 Objective Loss 0.103099 LR 0.000031 Time 0.021467 +2023-10-02 22:04:59,508 - Epoch: [198][ 1170/ 1236] Overall Loss 0.103082 Objective Loss 0.103082 LR 0.000031 Time 0.021460 +2023-10-02 22:04:59,716 - Epoch: [198][ 1180/ 1236] Overall Loss 0.103136 Objective Loss 0.103136 LR 0.000031 Time 0.021454 +2023-10-02 22:04:59,923 - Epoch: [198][ 1190/ 1236] Overall Loss 0.103091 Objective Loss 0.103091 LR 0.000031 Time 0.021448 +2023-10-02 22:05:00,131 - Epoch: [198][ 1200/ 1236] Overall Loss 0.103155 Objective Loss 0.103155 LR 0.000031 Time 0.021442 +2023-10-02 22:05:00,338 - Epoch: [198][ 1210/ 1236] Overall Loss 0.103294 Objective Loss 0.103294 LR 0.000031 Time 0.021435 +2023-10-02 22:05:00,551 - Epoch: [198][ 1220/ 1236] Overall Loss 0.103347 Objective Loss 0.103347 LR 0.000031 Time 0.021433 +2023-10-02 22:05:00,821 - Epoch: [198][ 1230/ 1236] Overall Loss 0.103466 Objective Loss 0.103466 LR 0.000031 Time 0.021479 +2023-10-02 22:05:00,944 - Epoch: [198][ 1236/ 1236] Overall Loss 0.103489 Objective Loss 0.103489 Top1 94.297352 Top5 99.389002 LR 0.000031 Time 0.021474 +2023-10-02 22:05:01,088 - --- validate (epoch=198)----------- +2023-10-02 22:05:01,088 - 29943 samples (256 per mini-batch) +2023-10-02 22:05:01,566 - Epoch: [198][ 10/ 117] Loss 0.313329 Top1 88.515625 Top5 98.710938 +2023-10-02 22:05:01,719 - Epoch: [198][ 20/ 117] Loss 0.313464 Top1 87.753906 Top5 98.906250 +2023-10-02 22:05:01,871 - Epoch: [198][ 30/ 117] Loss 0.303600 Top1 88.007812 Top5 98.906250 +2023-10-02 22:05:02,023 - Epoch: [198][ 40/ 117] Loss 0.298757 Top1 88.154297 Top5 98.876953 +2023-10-02 22:05:02,176 - Epoch: [198][ 50/ 117] Loss 0.299767 Top1 88.257812 Top5 98.835938 +2023-10-02 22:05:02,328 - Epoch: [198][ 60/ 117] Loss 0.307492 Top1 88.157552 Top5 98.769531 +2023-10-02 22:05:02,481 - Epoch: [198][ 70/ 117] Loss 0.304354 Top1 88.214286 Top5 98.800223 +2023-10-02 22:05:02,640 - Epoch: [198][ 80/ 117] Loss 0.305231 Top1 88.251953 Top5 98.784180 +2023-10-02 22:05:02,800 - Epoch: [198][ 90/ 117] Loss 0.303054 Top1 88.207465 Top5 98.758681 +2023-10-02 22:05:02,960 - Epoch: [198][ 100/ 117] Loss 0.304727 Top1 88.078125 Top5 98.746094 +2023-10-02 22:05:03,127 - Epoch: [198][ 110/ 117] Loss 0.305257 Top1 88.036222 Top5 98.750000 +2023-10-02 22:05:03,217 - Epoch: [198][ 117/ 117] Loss 0.302794 Top1 88.067328 Top5 98.767659 +2023-10-02 22:05:03,314 - ==> Top1: 88.067 Top5: 98.768 Loss: 0.303 + +2023-10-02 22:05:03,315 - ==> Confusion: +[[ 942 0 2 1 2 2 0 0 6 62 1 0 1 2 5 0 1 1 2 0 20] + [ 0 1069 1 0 4 14 0 20 0 1 1 0 0 0 0 3 0 0 6 3 9] + [ 2 0 988 5 0 0 12 8 1 1 1 0 7 3 1 2 2 2 11 1 9] + [ 2 3 15 986 0 1 2 1 2 1 4 1 5 3 24 3 1 4 11 1 19] + [ 27 5 0 0 967 5 1 0 0 13 0 0 0 4 8 5 8 0 0 1 6] + [ 2 32 0 0 6 996 1 19 1 5 2 7 2 11 5 0 3 1 3 1 19] + [ 0 4 26 0 0 2 1132 5 0 0 3 1 0 0 0 3 0 1 1 7 6] + [ 1 8 11 1 5 18 6 1085 1 3 5 6 2 4 1 1 0 1 33 12 14] + [ 16 4 1 1 1 4 0 1 989 32 8 2 1 9 12 0 0 1 3 2 2] + [ 85 0 0 2 4 2 0 0 32 959 0 1 0 16 7 2 1 0 0 1 7] + [ 1 2 8 8 0 1 3 3 10 2 973 1 0 12 5 0 3 2 4 2 13] + [ 0 0 1 0 0 14 0 4 0 0 0 976 10 5 0 2 1 13 0 4 5] + [ 0 0 1 3 0 1 1 1 0 1 3 30 979 1 1 9 0 13 4 5 15] + [ 0 0 0 0 2 4 0 0 10 9 2 8 0 1062 4 0 0 1 0 1 16] + [ 13 0 3 10 2 1 0 0 21 1 2 0 2 2 1027 0 1 1 6 0 9] + [ 0 0 2 1 5 1 0 0 0 0 0 6 8 0 0 1071 15 10 1 10 4] + [ 0 13 1 0 5 5 0 0 2 0 0 5 0 2 2 8 1101 0 2 7 8] + [ 0 1 0 0 0 0 2 0 0 1 0 3 16 0 1 8 0 1001 0 2 3] + [ 2 3 1 15 0 0 0 20 4 2 2 0 0 0 12 0 1 1 994 0 11] + [ 1 1 3 3 1 2 9 3 0 1 0 13 3 2 0 0 7 0 0 1095 8] + [ 105 99 101 72 50 96 21 76 75 59 127 87 269 215 102 45 66 48 92 122 5978]] + +2023-10-02 22:05:03,317 - ==> Best [Top1: 88.067 Top5: 98.768 Sparsity:0.00 Params: 169472 on epoch: 198] +2023-10-02 22:05:03,317 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:05:03,330 - + +2023-10-02 22:05:03,330 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-02 22:05:04,367 - Epoch: [199][ 10/ 1236] Overall Loss 0.091644 Objective Loss 0.091644 LR 0.000031 Time 0.103554 +2023-10-02 22:05:04,573 - Epoch: [199][ 20/ 1236] Overall Loss 0.108640 Objective Loss 0.108640 LR 0.000031 Time 0.062079 +2023-10-02 22:05:04,777 - Epoch: [199][ 30/ 1236] Overall Loss 0.106269 Objective Loss 0.106269 LR 0.000031 Time 0.048176 +2023-10-02 22:05:04,984 - Epoch: [199][ 40/ 1236] Overall Loss 0.102446 Objective Loss 0.102446 LR 0.000031 Time 0.041295 +2023-10-02 22:05:05,189 - Epoch: [199][ 50/ 1236] Overall Loss 0.103419 Objective Loss 0.103419 LR 0.000031 Time 0.037122 +2023-10-02 22:05:05,395 - Epoch: [199][ 60/ 1236] Overall Loss 0.106589 Objective Loss 0.106589 LR 0.000031 Time 0.034375 +2023-10-02 22:05:05,601 - Epoch: [199][ 70/ 1236] Overall Loss 0.108436 Objective Loss 0.108436 LR 0.000031 Time 0.032375 +2023-10-02 22:05:05,807 - Epoch: [199][ 80/ 1236] Overall Loss 0.107285 Objective Loss 0.107285 LR 0.000031 Time 0.030910 +2023-10-02 22:05:06,013 - Epoch: [199][ 90/ 1236] Overall Loss 0.107161 Objective Loss 0.107161 LR 0.000031 Time 0.029738 +2023-10-02 22:05:06,219 - Epoch: [199][ 100/ 1236] Overall Loss 0.107836 Objective Loss 0.107836 LR 0.000031 Time 0.028830 +2023-10-02 22:05:06,425 - Epoch: [199][ 110/ 1236] Overall Loss 0.106444 Objective Loss 0.106444 LR 0.000031 Time 0.028063 +2023-10-02 22:05:06,632 - Epoch: [199][ 120/ 1236] Overall Loss 0.106050 Objective Loss 0.106050 LR 0.000031 Time 0.027446 +2023-10-02 22:05:06,837 - Epoch: [199][ 130/ 1236] Overall Loss 0.106786 Objective Loss 0.106786 LR 0.000031 Time 0.026904 +2023-10-02 22:05:07,044 - Epoch: [199][ 140/ 1236] Overall Loss 0.106644 Objective Loss 0.106644 LR 0.000031 Time 0.026457 +2023-10-02 22:05:07,249 - Epoch: [199][ 150/ 1236] Overall Loss 0.105773 Objective Loss 0.105773 LR 0.000031 Time 0.026052 +2023-10-02 22:05:07,457 - Epoch: [199][ 160/ 1236] Overall Loss 0.105699 Objective Loss 0.105699 LR 0.000031 Time 0.025717 +2023-10-02 22:05:07,662 - Epoch: [199][ 170/ 1236] Overall Loss 0.105590 Objective Loss 0.105590 LR 0.000031 Time 0.025404 +2023-10-02 22:05:07,870 - Epoch: [199][ 180/ 1236] Overall Loss 0.105780 Objective Loss 0.105780 LR 0.000031 Time 0.025143 +2023-10-02 22:05:08,075 - Epoch: [199][ 190/ 1236] Overall Loss 0.105607 Objective Loss 0.105607 LR 0.000031 Time 0.024893 +2023-10-02 22:05:08,282 - Epoch: [199][ 200/ 1236] Overall Loss 0.105671 Objective Loss 0.105671 LR 0.000031 Time 0.024682 +2023-10-02 22:05:08,488 - Epoch: [199][ 210/ 1236] Overall Loss 0.105545 Objective Loss 0.105545 LR 0.000031 Time 0.024479 +2023-10-02 22:05:08,695 - Epoch: [199][ 220/ 1236] Overall Loss 0.105009 Objective Loss 0.105009 LR 0.000031 Time 0.024306 +2023-10-02 22:05:08,901 - Epoch: [199][ 230/ 1236] Overall Loss 0.104517 Objective Loss 0.104517 LR 0.000031 Time 0.024137 +2023-10-02 22:05:09,108 - Epoch: [199][ 240/ 1236] Overall Loss 0.104908 Objective Loss 0.104908 LR 0.000031 Time 0.023992 +2023-10-02 22:05:09,313 - Epoch: [199][ 250/ 1236] Overall Loss 0.104291 Objective Loss 0.104291 LR 0.000031 Time 0.023848 +2023-10-02 22:05:09,520 - Epoch: [199][ 260/ 1236] Overall Loss 0.104765 Objective Loss 0.104765 LR 0.000031 Time 0.023726 +2023-10-02 22:05:09,726 - Epoch: [199][ 270/ 1236] Overall Loss 0.105004 Objective Loss 0.105004 LR 0.000031 Time 0.023603 +2023-10-02 22:05:09,933 - Epoch: [199][ 280/ 1236] Overall Loss 0.104530 Objective Loss 0.104530 LR 0.000031 Time 0.023498 +2023-10-02 22:05:10,138 - Epoch: [199][ 290/ 1236] Overall Loss 0.104974 Objective Loss 0.104974 LR 0.000031 Time 0.023392 +2023-10-02 22:05:10,346 - Epoch: [199][ 300/ 1236] Overall Loss 0.104435 Objective Loss 0.104435 LR 0.000031 Time 0.023302 +2023-10-02 22:05:10,551 - Epoch: [199][ 310/ 1236] Overall Loss 0.104850 Objective Loss 0.104850 LR 0.000031 Time 0.023208 +2023-10-02 22:05:10,759 - Epoch: [199][ 320/ 1236] Overall Loss 0.104822 Objective Loss 0.104822 LR 0.000031 Time 0.023130 +2023-10-02 22:05:10,965 - Epoch: [199][ 330/ 1236] Overall Loss 0.105008 Objective Loss 0.105008 LR 0.000031 Time 0.023048 +2023-10-02 22:05:11,172 - Epoch: [199][ 340/ 1236] Overall Loss 0.104563 Objective Loss 0.104563 LR 0.000031 Time 0.022978 +2023-10-02 22:05:11,377 - Epoch: [199][ 350/ 1236] Overall Loss 0.104637 Objective Loss 0.104637 LR 0.000031 Time 0.022903 +2023-10-02 22:05:11,582 - Epoch: [199][ 360/ 1236] Overall Loss 0.104748 Objective Loss 0.104748 LR 0.000031 Time 0.022836 +2023-10-02 22:05:11,788 - Epoch: [199][ 370/ 1236] Overall Loss 0.104698 Objective Loss 0.104698 LR 0.000031 Time 0.022774 +2023-10-02 22:05:11,995 - Epoch: [199][ 380/ 1236] Overall Loss 0.104694 Objective Loss 0.104694 LR 0.000031 Time 0.022719 +2023-10-02 22:05:12,200 - Epoch: [199][ 390/ 1236] Overall Loss 0.104663 Objective Loss 0.104663 LR 0.000031 Time 0.022660 +2023-10-02 22:05:12,407 - Epoch: [199][ 400/ 1236] Overall Loss 0.104678 Objective Loss 0.104678 LR 0.000031 Time 0.022609 +2023-10-02 22:05:12,613 - Epoch: [199][ 410/ 1236] Overall Loss 0.104238 Objective Loss 0.104238 LR 0.000031 Time 0.022556 +2023-10-02 22:05:12,820 - Epoch: [199][ 420/ 1236] Overall Loss 0.104425 Objective Loss 0.104425 LR 0.000031 Time 0.022511 +2023-10-02 22:05:13,025 - Epoch: [199][ 430/ 1236] Overall Loss 0.104358 Objective Loss 0.104358 LR 0.000031 Time 0.022465 +2023-10-02 22:05:13,233 - Epoch: [199][ 440/ 1236] Overall Loss 0.104209 Objective Loss 0.104209 LR 0.000031 Time 0.022424 +2023-10-02 22:05:13,439 - Epoch: [199][ 450/ 1236] Overall Loss 0.103930 Objective Loss 0.103930 LR 0.000031 Time 0.022380 +2023-10-02 22:05:13,646 - Epoch: [199][ 460/ 1236] Overall Loss 0.104109 Objective Loss 0.104109 LR 0.000031 Time 0.022343 +2023-10-02 22:05:13,851 - Epoch: [199][ 470/ 1236] Overall Loss 0.104075 Objective Loss 0.104075 LR 0.000031 Time 0.022302 +2023-10-02 22:05:14,059 - Epoch: [199][ 480/ 1236] Overall Loss 0.104485 Objective Loss 0.104485 LR 0.000031 Time 0.022269 +2023-10-02 22:05:14,264 - Epoch: [199][ 490/ 1236] Overall Loss 0.104297 Objective Loss 0.104297 LR 0.000031 Time 0.022230 +2023-10-02 22:05:14,473 - Epoch: [199][ 500/ 1236] Overall Loss 0.104150 Objective Loss 0.104150 LR 0.000031 Time 0.022203 +2023-10-02 22:05:14,683 - Epoch: [199][ 510/ 1236] Overall Loss 0.103803 Objective Loss 0.103803 LR 0.000031 Time 0.022178 +2023-10-02 22:05:14,892 - Epoch: [199][ 520/ 1236] Overall Loss 0.103644 Objective Loss 0.103644 LR 0.000031 Time 0.022153 +2023-10-02 22:05:15,103 - Epoch: [199][ 530/ 1236] Overall Loss 0.103763 Objective Loss 0.103763 LR 0.000031 Time 0.022133 +2023-10-02 22:05:15,312 - Epoch: [199][ 540/ 1236] Overall Loss 0.103728 Objective Loss 0.103728 LR 0.000031 Time 0.022109 +2023-10-02 22:05:15,523 - Epoch: [199][ 550/ 1236] Overall Loss 0.103705 Objective Loss 0.103705 LR 0.000031 Time 0.022091 +2023-10-02 22:05:15,734 - Epoch: [199][ 560/ 1236] Overall Loss 0.104011 Objective Loss 0.104011 LR 0.000031 Time 0.022073 +2023-10-02 22:05:15,947 - Epoch: [199][ 570/ 1236] Overall Loss 0.104216 Objective Loss 0.104216 LR 0.000031 Time 0.022058 +2023-10-02 22:05:16,158 - Epoch: [199][ 580/ 1236] Overall Loss 0.104102 Objective Loss 0.104102 LR 0.000031 Time 0.022040 +2023-10-02 22:05:16,376 - Epoch: [199][ 590/ 1236] Overall Loss 0.104143 Objective Loss 0.104143 LR 0.000031 Time 0.022036 +2023-10-02 22:05:16,584 - Epoch: [199][ 600/ 1236] Overall Loss 0.103697 Objective Loss 0.103697 LR 0.000031 Time 0.022014 +2023-10-02 22:05:16,793 - Epoch: [199][ 610/ 1236] Overall Loss 0.103537 Objective Loss 0.103537 LR 0.000031 Time 0.021996 +2023-10-02 22:05:17,001 - Epoch: [199][ 620/ 1236] Overall Loss 0.103457 Objective Loss 0.103457 LR 0.000031 Time 0.021976 +2023-10-02 22:05:17,211 - Epoch: [199][ 630/ 1236] Overall Loss 0.103308 Objective Loss 0.103308 LR 0.000031 Time 0.021960 +2023-10-02 22:05:17,418 - Epoch: [199][ 640/ 1236] Overall Loss 0.103239 Objective Loss 0.103239 LR 0.000031 Time 0.021940 +2023-10-02 22:05:17,628 - Epoch: [199][ 650/ 1236] Overall Loss 0.102984 Objective Loss 0.102984 LR 0.000031 Time 0.021925 +2023-10-02 22:05:17,836 - Epoch: [199][ 660/ 1236] Overall Loss 0.103194 Objective Loss 0.103194 LR 0.000031 Time 0.021907 +2023-10-02 22:05:18,045 - Epoch: [199][ 670/ 1236] Overall Loss 0.103312 Objective Loss 0.103312 LR 0.000031 Time 0.021892 +2023-10-02 22:05:18,253 - Epoch: [199][ 680/ 1236] Overall Loss 0.103303 Objective Loss 0.103303 LR 0.000031 Time 0.021875 +2023-10-02 22:05:18,463 - Epoch: [199][ 690/ 1236] Overall Loss 0.103234 Objective Loss 0.103234 LR 0.000031 Time 0.021862 +2023-10-02 22:05:18,670 - Epoch: [199][ 700/ 1236] Overall Loss 0.103170 Objective Loss 0.103170 LR 0.000031 Time 0.021845 +2023-10-02 22:05:18,880 - Epoch: [199][ 710/ 1236] Overall Loss 0.103138 Objective Loss 0.103138 LR 0.000031 Time 0.021833 +2023-10-02 22:05:19,087 - Epoch: [199][ 720/ 1236] Overall Loss 0.102973 Objective Loss 0.102973 LR 0.000031 Time 0.021816 +2023-10-02 22:05:19,297 - Epoch: [199][ 730/ 1236] Overall Loss 0.102763 Objective Loss 0.102763 LR 0.000031 Time 0.021805 +2023-10-02 22:05:19,504 - Epoch: [199][ 740/ 1236] Overall Loss 0.102638 Objective Loss 0.102638 LR 0.000031 Time 0.021789 +2023-10-02 22:05:19,714 - Epoch: [199][ 750/ 1236] Overall Loss 0.102605 Objective Loss 0.102605 LR 0.000031 Time 0.021778 +2023-10-02 22:05:19,921 - Epoch: [199][ 760/ 1236] Overall Loss 0.102721 Objective Loss 0.102721 LR 0.000031 Time 0.021764 +2023-10-02 22:05:20,131 - Epoch: [199][ 770/ 1236] Overall Loss 0.102770 Objective Loss 0.102770 LR 0.000031 Time 0.021754 +2023-10-02 22:05:20,339 - Epoch: [199][ 780/ 1236] Overall Loss 0.102613 Objective Loss 0.102613 LR 0.000031 Time 0.021740 +2023-10-02 22:05:20,549 - Epoch: [199][ 790/ 1236] Overall Loss 0.102696 Objective Loss 0.102696 LR 0.000031 Time 0.021730 +2023-10-02 22:05:20,756 - Epoch: [199][ 800/ 1236] Overall Loss 0.102686 Objective Loss 0.102686 LR 0.000031 Time 0.021718 +2023-10-02 22:05:20,966 - Epoch: [199][ 810/ 1236] Overall Loss 0.102679 Objective Loss 0.102679 LR 0.000031 Time 0.021708 +2023-10-02 22:05:21,174 - Epoch: [199][ 820/ 1236] Overall Loss 0.102541 Objective Loss 0.102541 LR 0.000031 Time 0.021696 +2023-10-02 22:05:21,384 - Epoch: [199][ 830/ 1236] Overall Loss 0.102456 Objective Loss 0.102456 LR 0.000031 Time 0.021687 +2023-10-02 22:05:21,591 - Epoch: [199][ 840/ 1236] Overall Loss 0.102491 Objective Loss 0.102491 LR 0.000031 Time 0.021675 +2023-10-02 22:05:21,801 - Epoch: [199][ 850/ 1236] Overall Loss 0.102422 Objective Loss 0.102422 LR 0.000031 Time 0.021667 +2023-10-02 22:05:22,008 - Epoch: [199][ 860/ 1236] Overall Loss 0.102342 Objective Loss 0.102342 LR 0.000031 Time 0.021656 +2023-10-02 22:05:22,219 - Epoch: [199][ 870/ 1236] Overall Loss 0.102523 Objective Loss 0.102523 LR 0.000031 Time 0.021648 +2023-10-02 22:05:22,426 - Epoch: [199][ 880/ 1236] Overall Loss 0.102493 Objective Loss 0.102493 LR 0.000031 Time 0.021637 +2023-10-02 22:05:22,636 - Epoch: [199][ 890/ 1236] Overall Loss 0.102395 Objective Loss 0.102395 LR 0.000031 Time 0.021630 +2023-10-02 22:05:22,843 - Epoch: [199][ 900/ 1236] Overall Loss 0.102572 Objective Loss 0.102572 LR 0.000031 Time 0.021620 +2023-10-02 22:05:23,054 - Epoch: [199][ 910/ 1236] Overall Loss 0.102616 Objective Loss 0.102616 LR 0.000031 Time 0.021613 +2023-10-02 22:05:23,261 - Epoch: [199][ 920/ 1236] Overall Loss 0.102686 Objective Loss 0.102686 LR 0.000031 Time 0.021603 +2023-10-02 22:05:23,471 - Epoch: [199][ 930/ 1236] Overall Loss 0.102676 Objective Loss 0.102676 LR 0.000031 Time 0.021596 +2023-10-02 22:05:23,679 - Epoch: [199][ 940/ 1236] Overall Loss 0.102754 Objective Loss 0.102754 LR 0.000031 Time 0.021587 +2023-10-02 22:05:23,889 - Epoch: [199][ 950/ 1236] Overall Loss 0.102620 Objective Loss 0.102620 LR 0.000031 Time 0.021580 +2023-10-02 22:05:24,096 - Epoch: [199][ 960/ 1236] Overall Loss 0.102572 Objective Loss 0.102572 LR 0.000031 Time 0.021571 +2023-10-02 22:05:24,306 - Epoch: [199][ 970/ 1236] Overall Loss 0.102644 Objective Loss 0.102644 LR 0.000031 Time 0.021565 +2023-10-02 22:05:24,514 - Epoch: [199][ 980/ 1236] Overall Loss 0.102660 Objective Loss 0.102660 LR 0.000031 Time 0.021556 +2023-10-02 22:05:24,724 - Epoch: [199][ 990/ 1236] Overall Loss 0.102677 Objective Loss 0.102677 LR 0.000031 Time 0.021550 +2023-10-02 22:05:24,931 - Epoch: [199][ 1000/ 1236] Overall Loss 0.102622 Objective Loss 0.102622 LR 0.000031 Time 0.021541 +2023-10-02 22:05:25,141 - Epoch: [199][ 1010/ 1236] Overall Loss 0.102610 Objective Loss 0.102610 LR 0.000031 Time 0.021536 +2023-10-02 22:05:25,349 - Epoch: [199][ 1020/ 1236] Overall Loss 0.102766 Objective Loss 0.102766 LR 0.000031 Time 0.021528 +2023-10-02 22:05:25,559 - Epoch: [199][ 1030/ 1236] Overall Loss 0.102832 Objective Loss 0.102832 LR 0.000031 Time 0.021523 +2023-10-02 22:05:25,766 - Epoch: [199][ 1040/ 1236] Overall Loss 0.102725 Objective Loss 0.102725 LR 0.000031 Time 0.021515 +2023-10-02 22:05:25,980 - Epoch: [199][ 1050/ 1236] Overall Loss 0.102725 Objective Loss 0.102725 LR 0.000031 Time 0.021513 +2023-10-02 22:05:26,192 - Epoch: [199][ 1060/ 1236] Overall Loss 0.102765 Objective Loss 0.102765 LR 0.000031 Time 0.021510 +2023-10-02 22:05:26,406 - Epoch: [199][ 1070/ 1236] Overall Loss 0.102889 Objective Loss 0.102889 LR 0.000031 Time 0.021508 +2023-10-02 22:05:26,618 - Epoch: [199][ 1080/ 1236] Overall Loss 0.102891 Objective Loss 0.102891 LR 0.000031 Time 0.021505 +2023-10-02 22:05:26,832 - Epoch: [199][ 1090/ 1236] Overall Loss 0.102895 Objective Loss 0.102895 LR 0.000031 Time 0.021504 +2023-10-02 22:05:27,044 - Epoch: [199][ 1100/ 1236] Overall Loss 0.102985 Objective Loss 0.102985 LR 0.000031 Time 0.021501 +2023-10-02 22:05:27,258 - Epoch: [199][ 1110/ 1236] Overall Loss 0.102995 Objective Loss 0.102995 LR 0.000031 Time 0.021499 +2023-10-02 22:05:27,469 - Epoch: [199][ 1120/ 1236] Overall Loss 0.103104 Objective Loss 0.103104 LR 0.000031 Time 0.021496 +2023-10-02 22:05:27,683 - Epoch: [199][ 1130/ 1236] Overall Loss 0.103184 Objective Loss 0.103184 LR 0.000031 Time 0.021494 +2023-10-02 22:05:27,895 - Epoch: [199][ 1140/ 1236] Overall Loss 0.103149 Objective Loss 0.103149 LR 0.000031 Time 0.021492 +2023-10-02 22:05:28,109 - Epoch: [199][ 1150/ 1236] Overall Loss 0.103068 Objective Loss 0.103068 LR 0.000031 Time 0.021490 +2023-10-02 22:05:28,321 - Epoch: [199][ 1160/ 1236] Overall Loss 0.103239 Objective Loss 0.103239 LR 0.000031 Time 0.021488 +2023-10-02 22:05:28,535 - Epoch: [199][ 1170/ 1236] Overall Loss 0.103245 Objective Loss 0.103245 LR 0.000031 Time 0.021486 +2023-10-02 22:05:28,748 - Epoch: [199][ 1180/ 1236] Overall Loss 0.103281 Objective Loss 0.103281 LR 0.000031 Time 0.021484 +2023-10-02 22:05:28,961 - Epoch: [199][ 1190/ 1236] Overall Loss 0.103244 Objective Loss 0.103244 LR 0.000031 Time 0.021483 +2023-10-02 22:05:29,174 - Epoch: [199][ 1200/ 1236] Overall Loss 0.103187 Objective Loss 0.103187 LR 0.000031 Time 0.021480 +2023-10-02 22:05:29,387 - Epoch: [199][ 1210/ 1236] Overall Loss 0.103251 Objective Loss 0.103251 LR 0.000031 Time 0.021479 +2023-10-02 22:05:29,599 - Epoch: [199][ 1220/ 1236] Overall Loss 0.103334 Objective Loss 0.103334 LR 0.000031 Time 0.021477 +2023-10-02 22:05:29,865 - Epoch: [199][ 1230/ 1236] Overall Loss 0.103353 Objective Loss 0.103353 LR 0.000031 Time 0.021518 +2023-10-02 22:05:29,988 - Epoch: [199][ 1236/ 1236] Overall Loss 0.103455 Objective Loss 0.103455 Top1 93.279022 Top5 99.592668 LR 0.000031 Time 0.021513 +2023-10-02 22:05:30,120 - --- validate (epoch=199)----------- +2023-10-02 22:05:30,120 - 29943 samples (256 per mini-batch) +2023-10-02 22:05:30,617 - Epoch: [199][ 10/ 117] Loss 0.329514 Top1 87.187500 Top5 98.710938 +2023-10-02 22:05:30,772 - Epoch: [199][ 20/ 117] Loss 0.325719 Top1 87.382812 Top5 98.847656 +2023-10-02 22:05:30,925 - Epoch: [199][ 30/ 117] Loss 0.305597 Top1 87.903646 Top5 98.984375 +2023-10-02 22:05:31,080 - Epoch: [199][ 40/ 117] Loss 0.301663 Top1 87.890625 Top5 98.955078 +2023-10-02 22:05:31,233 - Epoch: [199][ 50/ 117] Loss 0.299902 Top1 87.898438 Top5 98.859375 +2023-10-02 22:05:31,386 - Epoch: [199][ 60/ 117] Loss 0.303687 Top1 87.994792 Top5 98.821615 +2023-10-02 22:05:31,538 - Epoch: [199][ 70/ 117] Loss 0.308236 Top1 87.918527 Top5 98.772321 +2023-10-02 22:05:31,691 - Epoch: [199][ 80/ 117] Loss 0.308622 Top1 87.866211 Top5 98.764648 +2023-10-02 22:05:31,842 - Epoch: [199][ 90/ 117] Loss 0.306997 Top1 87.899306 Top5 98.754340 +2023-10-02 22:05:31,993 - Epoch: [199][ 100/ 117] Loss 0.302918 Top1 88.035156 Top5 98.781250 +2023-10-02 22:05:32,152 - Epoch: [199][ 110/ 117] Loss 0.301693 Top1 87.975852 Top5 98.764205 +2023-10-02 22:05:32,242 - Epoch: [199][ 117/ 117] Loss 0.302588 Top1 87.973817 Top5 98.764319 +2023-10-02 22:05:32,347 - ==> Top1: 87.974 Top5: 98.764 Loss: 0.303 + +2023-10-02 22:05:32,348 - ==> Confusion: +[[ 941 0 3 1 3 2 0 0 7 56 1 0 1 3 5 0 1 0 2 0 24] + [ 0 1075 2 1 4 13 0 18 0 0 0 0 0 0 0 3 0 0 6 3 6] + [ 2 1 986 4 0 0 15 10 0 2 2 0 6 2 1 3 2 1 9 2 8] + [ 2 2 14 979 0 0 0 2 3 1 5 1 5 3 26 2 1 5 15 1 22] + [ 24 4 0 1 973 4 1 0 0 12 0 0 1 2 9 4 8 0 0 1 6] + [ 3 37 0 1 8 986 2 22 1 5 1 8 2 9 5 0 2 0 3 1 20] + [ 0 3 24 0 0 1 1139 5 0 0 3 1 0 0 0 2 0 1 1 5 6] + [ 0 9 12 0 5 18 7 1092 0 4 4 5 3 4 1 1 0 2 33 7 11] + [ 15 2 0 1 1 4 0 2 986 31 11 1 0 10 13 0 3 1 3 1 4] + [ 86 0 0 1 5 2 0 0 27 958 1 0 0 20 6 3 1 0 0 1 8] + [ 1 3 8 6 0 2 3 2 12 2 975 1 0 12 5 0 2 1 2 2 14] + [ 0 0 1 0 0 13 0 5 0 0 0 968 17 6 0 3 0 13 0 3 6] + [ 0 2 1 2 1 2 1 1 0 1 3 28 976 0 2 9 0 13 2 4 20] + [ 0 0 0 0 3 3 0 0 12 10 3 7 0 1055 4 1 0 1 0 1 19] + [ 12 0 3 13 3 0 0 0 20 0 0 0 3 2 1021 0 0 2 8 0 14] + [ 0 0 2 1 5 1 0 0 0 0 0 6 7 0 0 1072 15 9 1 10 5] + [ 0 16 1 0 4 5 1 0 1 1 0 3 0 3 1 9 1098 0 2 4 12] + [ 0 1 0 2 0 0 2 0 0 1 0 2 21 2 1 6 0 993 0 3 4] + [ 2 5 3 15 0 0 0 19 2 1 3 0 0 0 10 0 0 0 997 0 11] + [ 0 1 3 3 1 2 7 3 0 1 0 10 4 2 0 0 8 0 0 1098 9] + [ 87 113 105 66 49 98 29 84 70 53 131 78 269 223 103 41 64 46 98 124 5974]] + +2023-10-02 22:05:32,349 - ==> Best [Top1: 88.067 Top5: 98.768 Sparsity:0.00 Params: 169472 on epoch: 198] +2023-10-02 22:05:32,349 - Saving checkpoint to: logs/2023.10.02-195228/qat_checkpoint.pth.tar +2023-10-02 22:05:32,355 - --- test --------------------- +2023-10-02 22:05:32,356 - 33015 samples (256 per mini-batch) +2023-10-02 22:05:32,975 - Test: [ 10/ 129] Loss 0.325681 Top1 86.679688 Top5 98.906250 +2023-10-02 22:05:33,139 - Test: [ 20/ 129] Loss 0.338963 Top1 86.503906 Top5 98.847656 +2023-10-02 22:05:33,298 - Test: [ 30/ 129] Loss 0.342946 Top1 86.653646 Top5 98.697917 +2023-10-02 22:05:33,460 - Test: [ 40/ 129] Loss 0.355511 Top1 86.552734 Top5 98.525391 +2023-10-02 22:05:33,618 - Test: [ 50/ 129] Loss 0.350244 Top1 86.710938 Top5 98.578125 +2023-10-02 22:05:33,780 - Test: [ 60/ 129] Loss 0.342706 Top1 86.822917 Top5 98.613281 +2023-10-02 22:05:33,939 - Test: [ 70/ 129] Loss 0.347321 Top1 86.668527 Top5 98.515625 +2023-10-02 22:05:34,101 - Test: [ 80/ 129] Loss 0.346646 Top1 86.704102 Top5 98.520508 +2023-10-02 22:05:34,260 - Test: [ 90/ 129] Loss 0.347310 Top1 86.705729 Top5 98.485243 +2023-10-02 22:05:34,422 - Test: [ 100/ 129] Loss 0.347181 Top1 86.816406 Top5 98.488281 +2023-10-02 22:05:34,580 - Test: [ 110/ 129] Loss 0.346567 Top1 86.793324 Top5 98.462358 +2023-10-02 22:05:34,742 - Test: [ 120/ 129] Loss 0.346850 Top1 86.780599 Top5 98.476562 +2023-10-02 22:05:34,871 - Test: [ 129/ 129] Loss 0.343824 Top1 86.848402 Top5 98.500682 +2023-10-02 22:05:35,020 - ==> Top1: 86.848 Top5: 98.501 Loss: 0.344 + +2023-10-02 22:05:35,020 - ==> Confusion: +[[1141 4 3 3 12 0 0 0 10 76 0 0 1 0 5 9 0 4 0 0 7] + [ 3 1069 4 3 14 39 2 32 3 0 4 3 2 7 2 2 4 1 8 1 15] + [ 2 4 1133 7 0 1 46 5 0 2 2 0 4 1 1 4 2 0 3 3 16] + [ 3 2 24 1064 1 1 1 0 1 3 4 0 10 2 33 2 2 3 12 1 19] + [ 14 11 2 1 1157 10 0 5 0 2 0 3 2 2 8 0 6 1 0 0 9] + [ 5 26 0 1 4 1084 1 39 0 2 1 14 1 8 1 2 0 1 0 8 8] + [ 0 0 10 3 1 5 1205 1 0 0 1 2 0 2 0 8 0 1 0 12 6] + [ 0 13 16 1 0 23 4 1105 0 2 0 0 0 0 0 0 1 0 20 21 9] + [ 9 3 0 1 4 9 0 0 1056 59 14 1 4 6 7 0 5 1 4 1 4] + [ 90 0 1 1 2 7 3 0 20 1052 2 0 0 14 2 2 1 0 0 0 9] + [ 4 1 9 9 1 1 4 6 15 4 1103 0 1 12 2 0 1 0 9 2 13] + [ 3 2 3 0 5 26 0 1 0 2 0 1141 33 8 1 8 3 7 0 22 7] + [ 0 0 3 9 1 3 1 0 0 1 0 32 1108 1 1 7 3 13 0 4 28] + [ 0 1 1 3 8 9 1 0 4 15 9 11 5 1097 6 2 3 1 0 1 23] + [ 5 1 0 11 5 1 3 0 24 8 5 0 3 5 1236 0 2 2 9 2 13] + [ 0 0 0 1 3 0 3 0 0 0 0 3 10 0 0 1123 8 17 0 6 8] + [ 1 3 0 3 5 1 1 0 3 0 5 4 5 2 1 20 1147 2 2 5 8] + [ 1 4 4 5 0 4 1 3 5 0 3 8 34 3 2 18 2 1114 1 2 10] + [ 0 7 15 21 0 5 0 35 2 1 2 0 1 3 13 0 0 0 1101 1 17] + [ 2 4 8 0 0 4 8 14 0 0 0 27 6 6 0 2 10 2 1 1148 12] + [ 88 150 156 60 64 138 38 138 57 37 100 67 262 279 64 46 80 70 101 188 6289]] + +2023-10-02 22:05:35,153 - +2023-10-02 22:05:35,153 - Log file for this run: /home/alicangok/Projects/AI8X/train_clean/logs/2023.10.02-195228/2023.10.02-195228.log diff --git a/trained/ai85-kws20_v3-qat8.pth.tar b/trained/ai85-kws20_v3-qat8.pth.tar index 1760c881d9f71f8be30fee724a98db0849608ee8..392a3364876a4b699cbaf4e00f70732f6ef3f759 100644 GIT binary patch literal 2070584 zcma%E30w`|`z~9_QkF_~6+*b(?s?8FTh?$zu8=it6jFD(MTkfXqL9k2?AcYe>|3($ zOZHv%eP8}(=H8fmhu^=?$9tA}pLyPM=6%o1IdkT6SCuSfU|?))@Si_3gUSZpoNqv2 zxPNm$Z>@K;kP#z0wX0~*zt~??iAarsoo8tV_YVp74bT{>j(D1|QmEEj>+j_k=&RL~ z@bwKIInv)(8yGw))T9UJ=g;~3bznzKG$k9UHKpu4>o7o2uwU3n|7PLAfxiA;zQLoy z9hwD)X+y%aUZDYjL$#XHzS>}BPBX7CZQ#gI6EDu+CoFKJpO@BuRA?~g6)rc?l(Dn( zG!nRjIqzZq8Y3OVq|?Y?Z*4CIXO$)zW5GU!%+ZO$!AYsml=UpbK)zvq-p(?OiONu` zDc4hFpfZfnl;=f-sR6v16}1{uwZ_b|wzRFI{R4*uXuW&_wV|3y0nCzsBKB^7 zc;v6f))yM<=;TD?R8AeCX&^SDA#a48-iSuL5sh_5G|_78)taV%Gs4?%cvz^`tC({% z(>GhksE`vmpm2ap(_G9b;~5?FjB=jQQOBszYKU3`&wA3$797w^*Z7hC8mcENawG>Q zLLKEwnMNrlbK=RI^<*u0vX(lsR$5JKwWiJAu4QmYGo3GKTz>JCWp+6@I>?ApE>mjS zifP)p8OXJo_B@ZDDoSM-r|G~ae5Rw0uaj2es@8Ot4l;T1AnWprl8<49Os;ewjt)ew z=_=+_@lETdZ<-t5wC*~p9$JmNTGLZHM&$Z-=^Rz9XDV`32PYYIlFMY8UScK>o~gH< zsSnT8SI6Y3)$~(q`b&4ByxH%781Nq+xm*F(ivz_xgLt07dLA#HXNZo+TdVO=YkZ}B zQQk}{ApG=1s`ev@<$Ny8kjgqu%`;Wezen7Bj}> zV#XCb<4Qf_DxPt*j&Y4vld0CM{Tri{`>!j`D05WG6pji4&3ZB82A*-Fo^cb;xLL=z zMXSkDYqF)?-|4t=avz3-{ao z_x$21a(^Y1j&jXjG0i?VgU(vbexApL1pssZ18#No?tf56cu1=`tkxWnc7H{2_s{#q zsUQ8RQszh<pGmQO9&itGTS!T#@cYg~fNF`5U8@ z`#=4Y(TNRNPV9d3Ow9P4XMCY&e91Gu(lNf)YTl?dZ>8N|@u&O0(~}jszoXKLIyuP{ zn)hO|4?NjNJ=rIo?6Z#Si&pbht@-x13-{aozyIPXa(^dB{#pHpnC7ROfvuJ^Ff=gW zsbnk$O2%=9tf<;2+yb}~EH4mqC0S0(m10uOm1fej*6&BX5u-yTCLCNDruNH8S$t1J!{~MS(&w&E6Y+{vpEx?Duye^Qk@OuSx(DUU{cLhWK!GP@I3RK;*rS*kO^oaMBf1(Rydl1b0%Qr+4~exbcz|k-$4kAR;7%y;_)H_sO_7K=t})a8(-iik04h^(P53j3OtEKW-jt>+)tk~xsEXm5vs7n_jODbP1Cwe_ z&ZKy>{_PpXAH+CErl_}~=vj;^WHLFS;1v8xL{<=1=B)rr^;S@!Duz?CRA+?~%V{}h zCe>UECZ(O@SB!Ekna(c*eub&hL80JU@o*7;YgQJ5wGGSg2Dq?rWx?vgimD`Cu(oA+ zogwX5PRq4tQq6T>QYu)BPh7Z;OzoE~x?puwQYWR8gPiNcpF(7eD=YJz(V3biE!Ue#HP?qpsp}z{>TrFTPH_YHWd?;nDN?~2$h3c%B6umkCgQ+`QEoVYB9SE{SedtE zBun*{1PN6!+$fgnED2^gEf>P1n$s{T6{~+vnmCRr>WwJ6b1K+$p30O;E|fos$O#kifA$c; zM)DZ`6e43{S()#QIF{-=BVMSA;ij@wXTvm>({j_9RC5VTO2uli=@&PHY3WVS$EsW| zqeLcG$hevO8AK)|vNCT%5=-?a%o3_%xY;b#nJ|atv|KWiYHlu*QrAN?FXK{}PB8=6 z-Hff0u{o}SOXcw*1JYQTH((x1^#;rrs$#eWEY%sXkma=8A|}<`VkV^mL~_E$En(Vv zQ;Hs19a%>>%GtQJls}Wmlyp|+P03)X-jroRRSdVBr8-kqu$-1#$)uWF#iX<;Qn9+4 zX%{m^*DFNHF3TGJOd?Y)$=2SO64a!U98NnfpEK7hBv@ZmQjnc6R7^mkU`;K-)fa_$g+3XwI3S()#RBP`YTMxIa=!yRR*&W3!J({jg{ zRCC9fl#143vqkO%)6$!uk5+zFfl>z+EGPLhh)g)e%Df4uS*kbTj8GNBon@)cgmWyX z<<2vy<}NTPbwNbaNbVxjDP{okOSzn1uaR+=c)ZAf%dE^BaD}CM1Fi~HFm#Q9hbe639WwUJ!#(EDBr>Itm3dR1uvBl#Q=uw`d&W|o zDbHC>%e`Py&Ant&Dq8=Xk#et?qTY&M(aN60`Dr)znm>uiiZ`sxTk)2qdMn-uRWaOq zmg=ndz;asdBa>?G6O+=;@jF^SGo4=s{CdEEj9ri~JY2;8m6hELQncJRmf;O>Vewif zj{DBndzMg@Ve6LM4_44w@{{G*{I&#di6N7o7QZJ;Lw$>_-Ebw?@jt9^kST;Qt0WH< zSy76W`Ccf^QhhI!5vpQ1BbMs8jaiP(XPH!UCQN#o7vmPsQp>T!dg`JFUq`lX%APG` zTzMWU;;z8TJabXsYsu<3Ur8@3PEXO9YOscsmOnO!+#{GMGT9q9v#;m_zI5~nd z=Yb+-3s&ZtEm^8(t|nB)a8@kUF;{0fHiKnS&DCU5JU+4>mztN>VrqIDeg~wCJlMJ*i*8DznFEg$DV50 zgQSde;&CEkXI6GIFxPS|ScYddcQY`I4>#w_U}D7%Z5>q(#A*L$Qd|87?I)}HGr0M%lkXvIg&A^`QnN+>J%bG-zt zhZrkzLorsT-&@e{Bi8?)MhZlI1)!%GXf55kUju;7P_CbV>@P;v{U1nv^x_5xxPfBa zU-m%wj`R)+@zT=`5@3VHFwr`uxJPs_fy+w(4H1L>;u}S+sHUqRnb ztpC5dA%txd;QR&LP%-YG?GZr31WO@MF4mN6x%_ujj1V9r#SqD;V9QS- zT#%qVN~|pDaPnqD_d+fE*)6O1gsxDR^pxfEz2uNk#?h7I9RJ6JX=TFi9^ef(cxa0%(F5 z^cUaw6%`W&NR$}zm!8)}#Uw$0vRMCrb%WsG(E@IY828Wi2%s1N6e|WvMuj66Cuqit zH6>f__&X}53Xo}Hh^3S(I5rb(nJ%a&h}9*7h^-5Sa5DtunPO#0|8)E_h!O>0k{Bo% zM0#McAetp$XN$3tmy)dw2(d6n&`%cY|4$c72AX+0pGQ|+ds9@_a zA>3L)d7W5U(%}?;M#XvoxIqk*j0!!lSX68ju$#nK$!o*r;DU2+7WB7>_5Y`l0x(Me zW{ZK=Qc)pV0@2@Eas=d7F;dbUL`cC?w+XoIVw_}Dux=4Y#SQ_sQw)>zq9T~Ul`DXD zi9vtyjbBl*TY&5lL;lk9x~SMI=~O~ zbCr__kqQnXa_n@|@vX9ewZV(tH7o)9$W1=s~KOmxkElN32s z(T(JyfV(8diS{r{-~=bWETFE4Q6f+IjVcx{R|Qak7$oxiKS2V|H34#643Tu2qE>M? z1i(!(K(eoOfpSaGzAe`NcjE<*zayaTic$Y)hXA=JK<}y?+v=X#ii?#pVc){b_2q+gZ>L2Y8AZ-OmJ28ZR-=cpZ zGSruAFDP^nD~M+2?DgEQcN-lA%}!#?^5RqRGi+Bur?Xf`H05GvV=g0}9dr@&yNdNi zEh^I2T}Ch6gT6|DbQ42Ni))b|0~J)nbj*X$xoWE9iQPb&FrYUq0GT(CjbPEZ%iG z=NurY4HT>WrDcMP4HDD`i`AuU6?DA>-60~~uuvb*QmkqGSFxNoJK$N7F9j+Io7u=^ zqx`kuj+_r)&!03nL>m~yzLn&>et!zf`3fiXV^a6ctZ<0)7ix#{wW>0a+%VxV&lIZl z58(n>vFnL_C3SBU{l|oOd4~_<0$J^^H=p|5>6C2(VH^Ls;p~|B)#nIS=06kUMzRe5 z>QncnAQ!}n`i}*EU#l8!*Xm>Ba>`XBa>mO3X$A6ma!ZDs}qL*x8b_2dql3} z#xq?peIzRjyQ3zs3{S87NQ|4v*XzF%0n)0k9q)0qrYRgL5lSmxhtmC5;wtYmvjW-whb z|4ddEwrwV|49~CoB8W@k>x*`R&0+-||7@1ia&wqebIDBpwXKe9FU4G@Am&P8Wq~V| zWkg(Qe0>quJXX+g&1X3+w}44Cw~)y&RgFk)5zG9$t^9Y4Y)>THO|+QliusqYvap?e zDa-Ku`foqdS+Qt~Y6i>e=$EmamRruGnp?r7R8v`>%7rZ(E180rY85LBRI6D=M74$$ zi>NYLUPrZ-<+R*7Ce_?}Cc{+Lk=zEB`5!H1A5pSTq@BQRWV&MdO{^@?Z)O=?PYWC8 zx3FT-#`!Fk*U@LQoR-UBQq65;QnICtO36M};I=UZG1Ycf7N~ZxjEHI{D;81Zvb>IJ z7t66-hfJ!uJxqqF8bxw@S?1p@s_0s^hF!M0JAYbyO!=PRpHQ zQq7%aGE60ltKIaU@%it{YP)9d$aUf}DCqURzj==d+O zoR+)Hq?)_Jq*PP+k8lb52$#Fc6vSKwtSoR{V;K?Gb-uoc>jo?6xNfqXmb=BIn!C+p zn2JVncUb2C(^d*4cbDmk`R}o^u;cAM%kcdA9d8f#`l20g4_QIS|A^(Z++!xyTp^Ri z+bV4Ek;~Z+8nDjgp0ML$s;8_hP(5QA5!G|PzKH4tE9j_RvYeKC#iW{h&19IWbtLzO zW&TG~`JWbWbYSuGmg$Q5-?6f=!RS59@cg zkx>8a1pjx^wav?qr>2<0lU1B1AV zOxm$7H&m5i9}|^Qm1bKhja0^}vMLi*IaPU81yx0rsme@MNmW@@MO9U0uCh>Bs;a51 zRMk~AR5ew#RJB!gRCQJLRMsjRm9472s)4GZ%1+ft)mYU;Wv^m+K3Q;&Xk#^|LGFpNHey1EXbrSFE}%H z%(D_}jBjvI2 zv%B<1xenrwaybh@@qhiq2xlp%HQi;tO_eQ}Qf|_WN;P+;!KfWZZDQJqOxMEGc3-iD7!qxgav-qp5+Jc&7cZt@XfB&Eg zSLa^?*^zcY)!Ot2T! zB)dzr9wq9orvCy+byqV1AlY62*^A%&Uq=7#E}5Vvu^3kTC!aWnKTr4j4Wc*@<$~J( z_>$RCP_q3!YW+T!)R?9CySh{-5kXz#0Kd=v$6jd&nv%x;?jR~CNp#R3W0g|S6OGtK zw8g(nb^05pc${(;lx+XoLDHjCi@&Q&bx=z|U7~~jGeWfzG$lHSf5oZ4UvaGkC5aCD zV{~dG=t*=?@v+I}Z=B+hsjZ-7``-?d9+%qvU0te!+6(Fu9rT|use_;?(Lwy{t)dR< zC@4vE&>!PbCqYl5gNlzvu7Beck42pYCENdYkn~8@1fx!z3L3D4)gVI1c_O6#Q! z4sqc6uxeMAKBanC9eMOr`0&BUpZ)5ZaI9~!WBgYOMbMzO?i{~Hsxt88e^O9@@OAxk zb^Md8P$&FWAlIL-JrWky%w1(<>^XY;+)nNW2K+zwzy2zrg#n8FDq(YnqWNH_c5;Ja zKTY`mDjYFNAon-iAaP4fvFf4UC!$i!G<@Jz!Moxj%p5Amy2s(r!#Z+?+evL+>J#LqMO%lp2B6sE);<#gF_(ryeVRmzrTpWbMv99>O{WUrJr2*`Q$Z%n~Bbr%< zpyhshBxfEbH#^zF=86Y=j}FDQ!4=T))KZduwgyIA8HP8On{aUOMSN|T0h8vg7#!h3 z`hUtJ%>x_2`LHca!rjsRU@iO{)(+`47UN3p3@kb~6+ZRjk>1k=E`878^Kcv5dQeO1 zs_>^5JY(qK^g%RbU;qjmAH$dLXW=p99F~6hfbKq<;F=q!Jl)~E(zbVj^7GyE%IzNO zl(nCXP|mn~l6>h?8$NsMB6+nv&eUlMd&PY+Zsa^t?tm4VI7gsvzzkIKjltWiU9n(y zO?v0a1iCRbhOW$vq6fD5QRNajeH0RiX&Fl}|Kk~SomGZbF@B33r>5d*v=f{S7h=u( zcWCtWGuEel#vji@h4--UkbPqjE-uMM@Tr$@{$NSXmzmRNcTK5Z{xN)SS`{BR48lu;Nw^bt9Cfen zN400MSbHS{+D139v!5aLsQ3iel1GC3Sr17)D5mv=Nd*~POD@97cT=#FEXDq58}Qzz z7han8!lCbpc-15ykE)+W#{3y@cvAuwY8}JaRZev0CntKvtv`Lx)Rji$m!)nc6R~;8 zH#F^Do|f2JlD6q`0-x${M5Fry5VAA|N1RV%{)o?T-*OArGGfuJ&Sj8BBCYl8L`jWPMPIqWPm ziP_kSC}>g^OGkQQTD66UQAFX}z3-%(eO=s*9f+kT6*zx*5q8#1MTIC8V5`OUxK-He zxEAZpm!NM-C_Xjz!KkpgWLK*##5i{`X z?nvxxj0vqzll+7eB(h3VJRI5$Bua@|5AKrF4pT|lD_P`V8AD_THHFKtdI%l2l5~8W zfL22?v2E=oj4F2(`Rfm%yE+`J%r3&r(}31r-hfuk?oR7Z4y3h}o-{1P3%8SUk!N@Y z2L|M$rf0dypc! z%{LNUc{+aVyo3flE73-dHgx!YBU-R_3S3PS$;OwtSog)}U3smZS;qZ==TtH9h!NL%SZIOs&n_=xA8dwnjB^)}}lzw5)=~ zJ)_x}ngFwE&9QVrH1_6?#`oMj%=!KZ@gEMMex4u7t=dU`8kK<0`dX;Hx*3+Wv_X^c zkI4PkIk^430Na|>r>$1o)6@??a3^*ix?~Q6%61GUSR~;@?x#t*~KL9 z@N)7+zJLVgWRk^AXOViU^JLKqW1P8H4x583Bgnf1Hg#=;Z%z8aR~C-t4fo-}Ne7yc z(u%q`_~Y=bRD@oh0l!aEF|O5M+zINAc8PaLRQ=Or#{7Ha;Lf(#F?T9Hof?C^roGX= z?-H2hY(S8IDY|6FPmG?p2y(M%EbsgkR@W=gp`;q^;PDvTfG$X@<_zVAIoMb_4R?Zn zV2;+9zHPM;yYgHVoxe6EHW%BF{ShE@k2N6|&#YCPN|}Z^d07YwHKrZ16ja^52ffg( zE^TKYgcge*z`p!Xye)W-R<{nJ)JPi){<#U~hdw~Ja+l$r_6WA!s?$b?O4Epep74FP z1gmQYqU4TQsG8sa&B{>Hd(KbN+o3${+JvK1g-z((a|S{as^dk2-I)EO2K`dngl6}s zM2~gMME<<$a7w#JE}NVuC&Ef&fvFK5+$bQmYD^_E4Z&C22&@ZlgTZfSBW_0utU9&C z=O9-M8n_)@_omIpkb*rJzvl;f#8ji7E**yLwHKsN+Y(puol&)6UGzNN z5Qpx(B`4d3!@SfX3>xqmM^|k}P^To^JQa+p<5$y}9X`_YkBpSZlYY?slQz>hSv+mH z)qpO(Z%s#f)~5po^`iHlPo@vkd(x1{b#OGm7he*`A-zNaF70}Z#x)mV`j(f9r8m!# zr*D>#Jzs|q_fiK)O4kmgajDvL#C|p1+R2U1n;%6-C8p3C6})NFhEcd^uo8LQ7opO% zWzZ}f1M6v>Vc)GiEPqZ%i|ECOKbZ#C!;^8SV*siS?4sOLcY|{L$79OMqn{~L0!lh5 z4xLfP4sU>%I$aTrD0ttVhH-~Cp-J{kRIpisYbNK=BmESDY%gL{&?+?Xk452%exNTF zBPnMu4rHE36Wh<|d3_a97GJ|D%c}HF|8BJGrylf5yOwn7!8a&4G?jFIVS$qes$h(N zJ#_5Q1snXJkgdR*KE81-)Cz^dFeGJ}{K3P4`DPs_JH1$FL+6*j+jKkd} z^$?bKg&5fqtRHTNXM1h%BDgBX%{D{b9TiF8wFDC4wTHA=){)R>2Nk`CuTi+J1fE)` zG3oXIq`S1n=s7cR@Ut(>?i?enHx zNMc@&CfYDVVt%<4d_uyJ_iQXOAGb#IsakNUUPy*dFv0yj-EqXa9|pBZ!oW`n@VeU@ z*9NPRzQqq8uVvzb)fPMnpN%c2o8p7RTYL?-qOEeP(dsK)sj+<@Ixoy=ZcxhN-gHpRylKgq?*?J#*$ zZ-iItg}}>Y(KaL=Cy#DKz@mICJa!14n{qJpN;F1qH=-wc_M{DRr_kg19qHNt3%X%h zDpK#1N4dJS+4Gen#*OTbP1haJKd~A{jJ%5!dK+1VhO}u;2bwZLPABzPhO0K)$<$7d z$mig9y(VTqX2Kz6KF*IE1DznB-fSCmt+pEe$D1Q;UJ49l}=0kqe zi2kVa9Z6M>Vqw)C7+*ULhtsc--HLEr&!34qJ+>kzAraLIhoNl~f3&qJM=OqTqrLi- zrEea_AZSe_E?=65DSOUfN%ei$apecR2AR|64Y#1eg4!6<&JPWj4MAPb4cC^}N4<8} z$lGnR@qVNsJ-gy3j5k};E>o(~BfYO+W9n4oI~Y>;=dZA0=}Zj!#-WF$4bIIs!&L`Q z3>_SS4UGcPE6xKxU1}hcJ+G!JzLTWpb@0x}7Z1+JVB^1#jI@u!-hsz(eu)k3>)U{u zPCSE>7iXc{#+A63SB}2UD^Fubl%a|BUgPew`N*A73kt8j=xMVDJ383mlUE)YIWdh? zed~zMOMP&AM-nFVO+@w%Eet#=!p43ZIFCW}#@F6--?kET^x4HQnxVpw5>A+~^$5a} z%g~g92k=RZMM7s+xIVu}3>_PzUvdo?&bLB~v){=~yIivM&_Z(hl>w~1Ya>x*hm6MW z$@PPyiQ|d8q$k%N)6C)^EzUbM+~05 z^uUOd?il*L0!9t2j+Jw};@(m_v?&~j4gOlpPp^WXj&oq)y$8zq+mX2S7Cf))LG^PE z7;&Q(%^E#`9$nXr&Kci_-t0Sx&c4u+zPdnhcF$hZ~=j}_@SVK;)^%k8mbRtR=)+K#%mNqF9N8kS~%Bms*r!+OjYC*F~^ugswuT>+;jttQKJ7NA!fTRNky39Vz7f(!3v;@!O4 z#O~sJlxX}G2Or32XDds(U|%L)E^Ce%*S3%=->cxhohjU=%dn{TPf~M89w|_EL}t4w zaLS#9mdR;2-z^HM+bdv#dOrz>t%MgbHSu9g2^1P{A}wD8kuqlWarDG)oQk=Jy-T!+ z?PZT?KdK?ZHVs31??Sc1cVIi^9hTo7MGZT8(l(J%BzsdT8J@J0RO2p^MZ0H`F$-)- zLF4VjaZnxPY%PZgPhOFC7ax-(O)NRSLreNUpGJDNNhgtel8E{BC8X3G?M- z_NyyN==(yA1*OlB(|(mueSbxqdF_a&IZik=*8}qoS>sIUAb1=c0d;gT2Hy!pndN5a zo?V8{`D#UP<}{>Zx_Z)keJO3QcnO|dsE?f$$D?b5WE3WC$L;fb5PW?hy0>pg_LNK^ zedi{T*ehvd=8gn%=43Q^Q!y2pld96Z7uD!i%g)rRH|WlOq}BksKM;Ur77c5v1Z? z6LKtTF}B2*(5VB;)633g)O@TpUFP>4mm5ujMb0XeWY?_py6LF>r9B#Nv4VGVQ#37{ zgH`ncpmnW+NheETs)ILaGVF6$EM0Q`)|5SCvf(Z$CKH#6Rn zbpveZn9`%^qFRIL((D=Z`sFlQzY^%zp8b(O{|tt0OUKMgQ6L|>Kzqyp4R-c}<&RBx zn>-)yZcIY<>SU}h?~Y{k6y<`Mh02CYo-6lz|4_~gc&qfFN0eJv*T>coKBzQyKHhE} zk9zM_=-H7+QZqOWAge z8a@e1Hn(nXYDm|mT*m%RIZ)4xMq)}$TzonQ=R*BqY0wgzjC~NLDhZDg8_DHYPssP} zw~0gjR+#>CKI*EFi>Se z-3PD1>a3nP@IDQxr$?b&iW{c)_eT7KDP;BL?PR4#DNGOXhs~Zj*gSO#)~Cu~+rKof z-7k-fLlmnsCc@BeZk*C(dw^6Ke?W`$#Q_Q$KBUa)bW2FsYCaGdFZLH)NO=qO#jqo$J7aWm|J-$OBOZ3LN3iHsXQE9X;IT>XaZwKAA8zVZd z!1_V0u;p9>^sUeVZDr|*SiTPOn$ZY2S_`zBC5~7{V%?C%2)-AAFnB?EZ6fZhTn>xY z-niO23EQ7!;pFxL#2mPSQ)fqGL_{fC;YDj&p<+AQxK0~d#-;|Xy6_f4yH%s+g?`j! zO$u!nzk-(GGHA$76Dr%)2eOXISlMzVezI}?XuJj^A6a9;w!!c-+k`m-c4P0WHNdfT z*i|kHHP{$(TTzMlstw$`$Py5PjHwlvmb3LWS%hL$?*Ne$HP>6ywUXfv}b z2)kc`j{I^3-G*+!9d~bddAZv zl-M<=ZevPQ*NF$w-69!{BGU0BDj&Nu*5D+W2JO}%$kYVm{mfX%Oxj}4_%MX|M8m+k z9Fp1EUOz=DJ~Ya~R=?$_Qf(#{FDnPj(Q@?b5Q+8QPoR@e9xjZShf8n!VO2yN9#@$L zMRpt%27}-~!W-2>>f%)Db`n;{0yEz_A-GH(WKXCG_A`ZuKO2D2M@Pb#)1v0rq1c@> z4==Yzp?<$tWRXu}sLr)S@2x;u^P1RLC!1ucZ<19n?2)_FlC6;ylKxrN*qid1gja5h z-uqWVR(=}--Q&>Vx3 zmT6~U)y-?DEbE2%=wx!yqc2wPzKQVW`QR*LP_q9FG;9UanK&z2*Z>w*zmO!?a|>9J~tnMngeE&-20cvu%8BKb-)m5XDdS+nnvOcY?#Y7 zrV2`EjXw4?cyUAeZOH|k{A58xk9DSL8(iq&5{>CKA4__)!6tC49C0i}gO?$5kz4-= z#_f)W)mR(U=Q2PiUBI(uCvjtMDweajYLEq(cc_DhXHv1f=L>wg`3Ak$RH3(eZ$j3f zQq-taANnn=BMr}}O$)9(g=;BGlsI6ICy4>T>8l8Q=|CGCYfrD_pU22}bG$s@htSYa zlp3bRvrm1|wDAg(S<;5gdz3?NW%{CY{2nweZ$h1q`=fjF)#S?Ld~!KCi(JplCcW)! z5NrK{tol|Ht%8Sx_z%Zd^M<%{+aEX255ewN3Fu|96|)SVqWsXxw8lp@RW*N(jUmI) zD?c0dV@|+gMgdyZYDkB^F2LOJ#%R;y6LB3>4mbU4phoR3C>viB^#&}0@B1(K{LF%e z*Zv5@pCu?aZW3gBUXs|Z4H2~^1_w9A!eXofxs9A*#MY5BrXNDk`{&4haTlFJm%`Lx z2b{Bm$z(2__>|d3I>+86U*ajg48BRy{4OD?ZabQMA&{2M4WxZPn$lULUSi>m)i_!q z0v_8l(0A80jGuOdtxe9t{j7XUpI?V=&S*)`^#uLiswa(Tk`1H9WwFkq1|ntM&@`$J zRCbO?YG;T2M$VY)y$W8X*W>2t7@TG6SBCrEkx3;cgXDGwEwLQOlQv;ja5if6h(cJC zcw%3@42JE#OWHnLObjMulZgAT6c3NPBh4=w)z^=J^Zb?QH^mQK_LyP%-C+1wrD2zW zCybokaW$nihSyp{g51)u$gebg_^m!Q{n?y`85z;V%5=od+eubcutE7g4@t1|W^&Hv z9=TyVl>}c;hbr(cEWeec@3|lNkyMB-a}FYBZ5+<3!qBDDWVBT+L#G-mu=%qEGOeG& zRNIn9d~HunB0cE#d~14VL<9=Lm(qq;ZqfstpVIlSbE)dk%tNXOR;uAG782wg!8^U^5e^AGS>eG>F?YC zdpq_cwJg7**A*?5-OHfKW=CnLTMpf5HJCbl?SlPR;&89$2AFJGj;+tH!#+9=P1^$x z1{Y%8{Km9bW_#Mfb}YT=QJs!n6|7u5IaS%V>N@2s_x;M*b@G)Lp2sUE9|=P5k8C}F zTZAWX**M#88{~!rX7Po1PM+d*ua~%U(3Cc_DM_>Lq#^n9GT4^C4aKNu=zjkZs%4$Q zJI@$2O*5qjE_bJgD~zCfZU#}8?Dq70Vh#dtXOMw2EHP$=J@)i=f>Cf9$VwMfQr0DY zuZEBq&vdfyg@!0kR>SRjpNOh+Tf7{+5I2u4#Jv&8_;fuAil{u&Z^93FV5u4+b`tj;T9+{FbS4yTuWOr2esL_KJ$$d;})bMuGCCK zwo6B|?@{<*Gm8WbpAG9>1~ju#U3$!^B^?!FPg^FHq?5uGsNHH7b|ivpY2!P{kB1AbUZ}a%I_%kKafIF20bQjWuB7m6`qmhVI{F*NfMC_ zxJH_k8G=(Dt6;L@2x6CJz`kC0ysprkF1?pTuO_F_87HPtiyqZzdX^`AW|TzcefF-$ zZ45QXN29Jc$*4$zC!bW9km`oQX`Tr3tbkU5mN0Xl0lVHCuz1d5WTp2- z{n1r1Y-j+PpL390dr}q+LdxL1p$U41O(hoGR^lA~jMPu859_I2;TG+UH=p}r$BJku z%3px%+HE-W(~5pK@~2h34d_DC*R(>8Oc@_Xl?NL%RGxQ!P32Si(T^w2kO{A*q0*U; zxLebTE*nsY+-oxtyUhnv6qzDr!bJB7U?5{LA;pVef&+tI`h7`c%R)i^;@y|4lOc zhc_n8+=F}BC-G&-QIxX?!II+&d^MYntUXy+V6p%PCx>EI(s$B$^(^dPcp1@(FBr-0 zb=TPZ`cBteoIIC~uT}>1UTIspZ?!Q!(r^^)56>sj1r@MlcyqjK)D!n^cfrBEZ7^rt zE>b%21X(a=3aa}agw44_xZiv!I^_;TtBA#jjX#O`+AnBqRF^K>Uzv6fTaUOI#&rBD z_Re{1E9!Ww5pA&I4Jw5u!1mmBob6JT1_jrnz3y7k31)YYSdAd`$UxZ5xr{lL-ohl| zA?W#|2x@1ByQ@s8$@^Z^X@?(;pWl+so_Q75hnm3jAwle|>1cXz461TL=s(pGPp>v2 zhc7oo?&}!j|MWq~iB^bIyJF|b4ASD=B~sPR4d;$bMfIi0XhMNE@#D!I_lu-?OhcUC z(*@smw8WrkLqJ-T#uNwFTJ&&i0tzqOCcBrvBJNddz_)6zMdt6-&x~^OB=RsW{2Y)zmvn8rYS<>wvaEzZ^?_` zt>p8rgCs1zH5pycLJ^hSm<;;blk|%pMP7dBN6OaDB*R`lSNPR!K^*TTk#pnklV|1? z@nk@09B{l$=GAP7x~msqajBzlTYDMCiIb_>l%DiW^)|TmY7#0BiN+#jJSMD}jIElN zq)*o^WM9odzy8q}70dRy1X~k>pz&l)&$VRIrR5~J zc@k-v@3jn?O}^vKCW0XgWaoL#^Rk-86-alLU?d09>s1$ zmtl`kaP2Hw-AhKt_if<*G#Wt{2II%mkud#Xjh+2xk^2KTW88xeh_%0l&O=kM)6g6t zbNq?_tYA9)>Q>s)Wh0I6Hkan!G^beyrlQwtrsHt|veWg6x;Vr#k=wqQp7AViuiBPXaW=H~t+liZ$>-apO}`w7%m@LrjNt}jlqJxXh_H(>#$vN`>=mIdfzTY)Yc zW>0(f?MQd6Fr%tZgP?t?fHH6_Tf6Oxf=fW#sg2>9_?q;cluTCbT1zgyJx-pPogt4> zw-E>V5^}-H3CGQiu}NVK3(NL6a7~4?GjhnQYHS=B@qtuZ_m2EDGsc_tKgiSc6!N`s zRfP4ZkGB^qLt$1H%Nsl=%TprA%1t(~A6FUczvPjQjvvU6J2g?l>YpYlHN8 zg+$T1BHlRJ9V_>vP51-@$cE!!%s!Iv?ifb&$wQRYBlNaj1D7Uq zaIc^%Dwb$V%T@&~m*zv;SyK82M$}V9@a_FP(k#Ut8DD$h_Qwt=VE36`5hd|F@(Nln z+JwU4DQMdH5RO#YiuG*FtWa>4RIYK0l-@LhIH&9=Gqt`i-&#fDS-l>gF16$$h z_!>wZRUIBLXOkQGBhWliftQQUkOF(Q|Lj8?95M*1KiY!aYDciuW({r~+>HrKA0i|& z8DmX4(QYBLX@SQ^dU(Mi`f=?_y8k2Ss(Tz7&7Ozc(9QUGVjx^9PeyP=BTTsIhdydA z9QEc<%PSR*OO~Qvjw4&QSOW97_1I7*83y&YBbx0)Fn(mt)-G%i-8~iu?w`SQ`;oX} z+ZmbZa-1EPOD>Ibhl|BnY&zZ%^E z4GgSP9!Cz8L5awwSbp3I^KG7ygPkhTciF5f@*U`^(eK<74WiZCvjYjf@foMovQO(yB8LQblfP`K!aPAJydJtld#IbeC zAEcVwb}T=Xh{C6rP|Mwd?%7p{N(+)Xq2lvJu3KcJ9JoLg}@sFu;fvDY*^eHCQ0wf?%@jbFdv4qV?EJ5v=W?3j>hme#u)I_ z8_hGP!l-%-UaT04_m#rgyIV8F-K);V>C)7G>K&NYxQhiZUn9yP8y{o)qfv+h7R{-I z@*N!zc>vHpbSIT;OJQ}|8{*T>6cMj_<2KhBspjT5PGgAWhlQl$t#UBj_>+`uwUZ3# zt0lLNrs7?xAb4-yjB}ErF2|?BymThWMJ4AQZk3dclra;DWV(Fx|T<~+K$NT zHx)124aCGV<8gY|GI%ekf_2NsB4h6Zf-~(=b{Bi!W849kA}lfQWqZudbj8z*)rj0T z9}%b0(b0GiqI+bKAMu88v(G0RKhGyWy^fPc!M(}lrg`}8b{gYqSEcWs)S=_!*0Htq z_0XO!i+PiNkn(N)VQZU!MD18a_xnO7x$K3}{5^Al$o7=(Cv^@N zl9}Ev@E%hM)5m@!-xK#pw~hEtjv8l)+TRD^o3^L{BSN~I!XuTY9o$%tsF6qQ6nR7xQ& zN>Z}YKuC&=B7I3DS)qvU{{Q}apXVOub)3fo`KNJ^-t>*i3ZGzdh!m^rS7N(z>%fo; z-w|E;RI}BX{1pi&N5#t zWd&|Y?9;#jrsTMY`OX>4?p$d_#faOewkyYvv>I%U6k*BJ{$YtuFoIMg(c)JF^_}PO z(j^`54g7GlvWK3H{7ro`PtsGFlO!V_Nw=m(3A_Z8*&NA{Y{b$&w0i0?!$-5(?3Ge1 zh0n_ZEj6+J<`gVxnS+yihC?@G2Nhf}1b(gK=hZcMJ;x4D>z6_OKRN3&BNNlR$Dwd=8}26N!s~lBnpK9fc0Xm-GAkFh5}b8j z>jm$o{TR0_9(ODAP&@Ppro1pEm)AE)_33aJx5%M4+78!S?V%-1K;Y4#tRh;RZ40=E z;Q@x`ITe1MD)ACrDPZNvtOlESuKJ4^2U&hSVvg`YH z;)|^YyHcpi7QTCh@-ay;P>2Q-FGNb{6lOBwJxab0WxZFsFzr?jRGhDo5kLPAemDsp zV)|%(lTI~$z9i$X0H?2G^0sHcJID7DPIP+R9!eWbVV4dW3>^Ib9#lC~sR6ztT z?i6Khhg$LIo&xsWQD#jO*DTein)jj~Fz;A9^z9l>&~rVS(ka{-o=^u09Cd>HXAu#jzMmb3@cj|kB`-cuzVzjjGhE& zzR+Op?!VwZAq@%MhmqED8W~MO^l@I{_wgkt?+k)Lu^FDbO5@?2N;;ENi>g7AELl;J zIh%YyNPQM8dG?^o*+D%cdvw3s2&?j3%s0z{$*?{eGIJQ3j{KxM6Zat8Z5OQC1|wE2 z6s7VN5J`x_%Eh%17-YeGwHuziHN%mxLy+Umc=*{v^Y3!(2_KD#Nx3xMFc&LwOA!+8 zgRUqUbR8CjZ@_Mhso=B8gm~!corLYz?Fi5s54QrIsWDCVbC4~2s=t`&hs|KCIsaTJ zvK2G#6l2WgFj%FnNB9>*oEoNvm5qzYYk@7?bq>P1zZ3@kxrm)!OFGX30qsnj5xa|v z#_60>-iYKpPt35)!JCqQ$hxG(V!ke6)e|{~=;4P&Z-Qyib#v(L^1!_9Q(=C&o97~i zNd5Z{n`VE&B>NATW-^LBe5}n9ZfdgS&jz!okWox=loBhwr^9xOjAy#CO?Z2wov!bW z!Kxd1_*0q!j}zWl`+Fg_?L36r{Qs_eFTp-s@4x{`ei!5KrB3oRY*tN2+Nor`+kG1G z^?d$$*b497f55oa>I@TAnDM!>>`JaQi}tt<&zJ(io^(Hq3aQ8C-O?gzjaLvuUdD zGO=O93rsWIjb&w8xVA+bS4&-yQ*xZ|W#d@m*?X8{Y=~EFowVXkCDMGqV(c7AmT&w8 zma;K8Er_N}LkSc*E8(WT2zJJ2kh$JP+R_^gSF7o)`9~tN+jEw^@$q98*>Y@p)-ZO% zej*beWX2Y%tF!X&8L$@Ny&=0nc;YgEH-#%1GYnxKKen-u#|EtFTL_LO{h*e62T}KE zBW9%@p{r}I(YGu;7@jQviig?s%CXuCYa;u~GC z9-j!?g-39FSR)!!5}@78@2i26SbMiIdpswAg@2pIYWy!^QrtLnXIJ9SNZuWB8HC`E zLC9W)uzDl zddy1UYY|h8D;t<#m zb(E#IonU%u0=A)O6=WQ>*}-L#*v%6|S$VVw%lCPSl+>Y+<$b74kH#Z0Ig;WZb<&ey zRZQZ2mcHV6=4YJBJd>}nOS$!Igp55CO8vyWpI@+kO&NY0p23Nsb1-Ov1G>(tV@!V` z6tbUS&j=BwmG%LlzmDUcolrRF;u+zcd+9=#ky*lLH}ZrkCg7}%($m8OhTt! zFEIsEoSL%Dp{t(8&R9zio<5%caPlmnr>J zA}Nfvr)4j1(Z~FEG`sFLO&UK5F-wP`<$@gg11(_klodBHTl+7*b^gYRYxnSMl`YgFqHs^706D(V zc>mcC;Zho~)%WJzRSD)RGmfqQ#b(Ri?|8)MIzux3jW_Vo^DvTMIWrmhkuEr;Qf`x;{=R^gOa2-3ca zpesfKK1U2O%;g0B`c%Mq$W=`El>mv4(@+;K$-Wpf7GpV|6?+VzC-5QuT&{%ZJWs6b zOhRK(A>=nlBj(p3JbkbYEAy+UV?AdS27ADJ6oZOiD{ZwarUH9STs3yW36W4-^!JCL zeJ$=@t0%>)8MN@bCRU%DfSi_z7)5%JEq+DMq&FcW?i5VkuY{iXQ|fdKqo8BAsOf7H zNu3ZPa>jDB&)fyE{t%3wmP21hszEEl1%vuGVauE25VS==pu7yH14dEi@~spQ?Mqie zZ_|_0EsVd0vcuiBK&GFyR<4~`RLgtzbq>5*v%yI=H zXUAaXU(VX>lx966X0j*sF3j;JXO8u*lDFD8Xzg2s`jJA+{#-)uoUN#@KZBIjM6hYN z5w0|DNA8RO*m#ad%8J7|?{5S4YSdMa65 z0c1yxWT8*`@!9PFqTJ*7Jo^I;8^p8w0DbiBF+**i1q>g(q}W|?v}{KsDYSOc$a7){ zGJZjZ;&Bvtax~_;_#%Hz7=CE%!f)5bn!BaA**M=q^ZR2yDHgMVqEGf5HW z%Ea(^sVXX0x07${I?~ZOMRJ4J(wC+b@~XT<*#oobMYSQhr03GCtraxs)JvK__$g`J zk08efhJ0T;!TZ0@F->v^`(C%4bt)S&(_b3+7ZeCtwjV#$qi|^(-|N1Pz=)n8`iBqn zRQ^A@I%EiTEi}X*%}2Cm=SD2HuRt8n=Tv=%ut61k4s%!q$&&-oX;y{W9G)RHOEaTs zgV?r?Yp}d=iGq#;8@k+J<#q~XnaOxwuXJ&o(G`2C$lGUkx#;&c)V69P!iq>~9TmKxsLvCO}b1B;V9WZNr5#pTQVxpx8 z>v0*u-a8IJ$IlKfl}C}#9|-+Tez?!`V~0h4C_JqMdkYH`^+w{xj|^<6GJ>Jn6|$M^ zhv2*x99j7WG4}5Thp>kzHJ)=c*zr7`{eUy7P)PwJKzOiKrM(UD|doG4z4 zmH}(Tw2i=v^WD62AdjQ>qcBqW6zbafR<-|$K)-k`P2RB?**9xYa=!(BLQ!^m+)D)M z9>$uRO@ef-AA&<(Q>g5^3kiEa3Vs%B7aW!!3-^pLxCmb2&He%W=OoE~)YhSFmI1DQ z-veXktuW5gg~y&MYUz7K(Y#9+Y^=x@-yX!;%x*z9xB!yJB5*A#j+L)_#@>arGmFc2 z*=XByY>rDJe;?*yr@{mDRR2b|{s=a?%A6J3_v71#Z&dfn9sio*VB&ZSsm*!#dLS5^ zbQ8(Cek88S>flf3V60p2L*I@%Q_c2LX!;v6&H2X6&1wVlh8N4dsmC0LSipzp+h6Rr zVvC#(#(*6&!V@v0Bc63!@bHMWko}UV;@EcU*={D#}zjTcR02Q z&+uF%tWX-~9S-8zZ+;hlpNIg7V15q}z-h`Uq}*sm$?UtB>=uDH0-jmgnL|JI1aAH6 z#ihOC%rE&L=VtHXlyNYIhjb#bax}XhY07rrw_-zT8Pi;xcH2a}+Jf4xuVjP`mm8((yh={8PX%x-*Y9#&!s7_8g-o^&HCH z+DIGpzEDGh5zn={DLZ)~<<_N8X38H*DH$O79t|vP|4t#H{es+vtAg)~bto-x8Cff) z3-+`m3BL1urg>8o>P2`)GdKv-PVjl;ydJu;WFECeFQk&!J4v-#k1n=cr)$Tq(NqgD z_(v!sYOVp&ep%sj<0xoK|Dwr1tZ-_0GSUqASz$^Ns#{7S+Ykxc(L>R=cO>e&v=J7q zj>x^^FkW09pUxQz51)w?I?D06?k87aK@t;=$q^O4o!m$x&dFiK%75fuG#(elq&Oed zL47iX7yZta8Fi3>t0LT zPh5lF-yLzN_YySM-bcdic9^A3z0#Oi`2UZ&o7a4P;d|i%%0i)n`=OZ0HGeFO> zl<`t$5Y8MLgQ?+4NENw4Ctml{mxoQ{ZKTzZLvy@w4M%(6UQ}z6C`ZcMU)(_7llbZ^t(3QbV^P?oQ zXeUNBo`HhNRpgt*Apf5Ut~?`{{kDhv%}|~JAIIx4PI&We95!Ayz~ws@l-8C?1%XyH zetrg73ZK&Jd$VbtOfo$;kin(+sW1>T<((}JgzTJ7qkf#Dg@5JYW~PTp;{#!{W(Q`+ ze5cvEQq-SoLU(2yrZeeFsYkj^&|-UB@a_I~YOYPCqV+eZal>1R?i&QfEM-bu|Cxd< zbdmRcHHLkkfCI_O7&=28jUSpwmUmL5V%?C-bH@0^5}a$OqfE~Tda@~%Bo1*7`OOI$ z_v<7DF9-9Ux@8FZ(ojk^4Y;FSCr_^7@jaou;M{&Or0Y$oH* zH)ZTiYN0b9`I>eo5W%}@ux67A3x6%fUd5e-Mk>GGFfHxjSU>DQh67ECubsXP|g9pY9~o`Wek0v(?vgT-lx%O^4LAA zk#tAKlWgWZq(4i=K=U({8@|AF&fPtp`j%8J)zCCy93~g;K=#&TyqG7+$^~*v`n?tn z=}@QEVH0TPwAHkTG)R~oDoBi6M{5tBr2p<6qde0P(y&M)>#rJA(x*=!Qx;R}a5LIs zmQCv{j`6j6GL@DYlUG+5@m+!354h0Z(diT)vXwqJ_K?i!;n?LyxEj05#QizSlr-b6Yl8%ZKkXQ?Ch3rQ~i zKnrq*LnBEJ&2e7^Q~!qJVtO9Zmp*`1^0fESGQk$kMkRBkCo%;q$L=JI>htQNrk5uRLCz_xP&IO~}qYvl;^ z=QUB2u#GB>UeW%>-{k54ghqMWAy;fLF3ea=qb`O*nlIc*HyY9OdoK(#8Fn6=j()XR zLBjAc4f`zKhcb5pP* zol!*GBzhaSMNogR$aK-2sk~Da3Jd4cNY;3SWr^vK_{BSTf8_)fmoxS&G%kJw-I{ue1)2qJri89bBA9UhQw`KlgO%?citX z;c9RSn+|y+KU_bfhmtpcNPOd9i2LkfR);UJ&(?X&qsfPTv@l}F!;RUerG7}5Re*I; z4`5!W!cO~Yv&+*s6Ylepyf;ojkmU}1S&@$~=0$iWKZ$2!(sXUbWD-~ z=D0kq0s+nfKS#fv;`87 zdAS(z+h)LUzy!myra;qmD7FRsWhO_hg%AIE2@l3b3!SGQ7phKm7p`!MrasrV^tG** zs+**tt}cr`okeuebTAfK{Gdt6>mfeH15thQxc#h$GuCt_LWv~e)piIEw)qzKyn)I2`cdY;YSv^7~Q9g zGmj|BUK#5SsAI>EhqU0CIt=VxF?8PsoahU|vY~wbxl0tM)IO8tx0O&=n~NiRIfJOJ zkNBmcIKcDIDDC4^ce$MQnn%)v+=)b*#(dqPtQqAv@TtG6(c?6MeQG=lGgaQqrHB`A^ zD*Xx=Nv4|a)VEU$`TQO~DzE^$UeyRJt-^^vOR?*pD!OxJfe+4Dw`M->{@Q@d4>)}=<^1z<7VTuo4!+c6OQdR{HhUL|-xkM6%`vzesD}j;RdI3V3mVY2 zqZ{I36!SidzI1oc@O}JFGvx#QR-R1qNk3_A>j3@L%b`@?qtqR;ff_E`Q0baT%G64r z$)}VNe#e|$%Ltq;>T!!+n@F1+P;qpTDDVQ>1~QsP=dFZ36^H`lJeXd zl5{hJUqLwhXn2uOCG!|CC z{YVLRl*iJ9yLaeyu{=6HKPIucjb!`&IfYpb=PaEQ&TU_eYNsfiPTPme$zQ0aV<_EF zTuXb3^JvVg0y@4efeOUBfna(MAlZ|`t`ZsQQ8`TZK$T8eD7S+R6IeyM&?ofq;ztg8cg33#!zznA@Y9O zOs-d3NM?5{P4FHJ-J|E}t&u-|yvc+!&vWJsDa9q9KJ@loK zTy*8!3}?jRok?e91UV@dlTwr_=bk1*an@_fe({mA&UDkOQqDe&;2HW5HJJSEC8>3T zX=$R6x-JJ%>xDQ{a$ZDNTU2P}%4EtL)5aA|Ep(Vnmby8D2{B9Q zr}uT*lJT8%)Bn&EHFfx{2%#ae?`WjwYE%}VL8eO!#2=_~4*_?FImzPLi!q!TV9@(! zgv|N6ct5w5&UZS{iry&N*>Z*=UOk|x)zy^YSx?T-W5`ij9?b`QFiOi4vOUxJI`EEI z`5d}3H;~dEuc2$VLg;PIINEM3P8-UIPHAkXff0PKm{Lz4=A_VUkwl8GP^ZtL^QoeC zFFD=XMsJ>*(qEMu0+)rykagfW;`&~qwONZf9#LYmi(li;+$Utdxt|t&G{6sITkJZd zixtz_iKL?tTU`k2!aP*Hs>hQPFK{z39?RdFk(AY1`cJKy5*^bhR4IWRMkmnEFf;O* zbA_DdHqol0Ix0`NPD5`-P)Ol5n%nz<9C?mh@KO`;Q7`F3-7(tXkxS*~y|j1GB%C|+ zl}@SLqg5gC)b?Kv#h<=NPq*Kug?fDMer13@tiH>gb53ZySc4q@SCGA4i)UHqp!dp` zcNvDlBV{V=oi^i^5zoWL=5Rh^5PVmdqeICDyNjc^dAI=kxEpQcODTl?S3qeiWRP@6 z6%*tq!*iM>e7>mTZQucn{eBbgf3%?Q*GZUo`a=JJDqekBht%2o(e_~%{yp*HT(CP< zip=NU$Q!)RGmL3ZGGnuA^;ko;BJ1xM#!eX7LEk+PgBEhu-Tn-Q$@8`DPdr`+q)>!> z9Zgo0#n}Z$(AzKwV`F8isIiVT{;9ceSM_3SW=7^}{iys5~$Iub^9 z211{UvBE}&bHa}f*M;k|?g}+al7%j{{N8tE1dOW|p*(<4Fx46}jVI&Sq)@CM_7bLl zzChFJE>cHuPGkQ`JexBdu|Ia;K6lBq{;|WhbVEcHx6;dAdu;sg2=mcysdyv>B{5UoaL)H$^I%`?nw&xz4))qP+u7af^TG0EaiH8qd=*i4|B=~%l4$po< z?T(=|Xip*SnCnVBpn#d{J?zsD#M>$Q+-1;4jpkpd_OS)Z_wi0t;V?W`A#CN(-oXiZ zB;BM8$$xh!wB;ye46`J?>F#86Yc6u80~F$bgl)#S-Kzo5c{5?ULjmoF&*PI6%(w-_j)!Hcx&u9>qMBNm|DP0u4&O0pp`Rxde{jZfC zOf9AP&K2bBbcb5R3MnC>iv<6PVg4eXhh#gVzhpK}N9$pFz6xvFw36MFb!9!t(^<(A zbtd|%AG?<+;njHq=<~d`&|)}V{}_SgbctLhbiru60vptD2Bn-kd9(8ZmbgEKN31v; z)C?fP{||{}GTi+pi$ew(bUXVtE%_*p($*0;STYEGvN8}_w$b?)!*OMx3zr@G(b3EK z{T<6N$Hfc&!5eXR#a;CMk!Gby*^o1=1yOH#;%eb{0}H4Dp#~S~K|>Wd*K4(S1Aa-Q%v$S6k5_X~;VV7qN2WBNY76 zU{y!n!fIP1Miu9yAZ0L2GaOJnY9c(ZTjJSnD@^K=_~$&Q;-Vl zy`spB&nHmv9?mfqEr7;xNj%Fer7x4`;=d>@xaw}ghPYiQt4-iuf)La>SmRrUII~-2 z#pK+~*>YEPcGOah&2DVL!FEaJFj|fEnCY=V4;{8;sR%RO(*ez2Baqi1MD2xf=nb(( z!4L^7s1fI#^gOcO*+p}jbIDcu0-f0VhKwIHQPBe}Hh|dXYA>2o(eNLKP^Oj&= zwtj)yLL2t)&@iT>H;64u=)(iyZMgC?)F0jZ&`NoRq{aD2^UQ_ZxoS*r2*y15>zFsE z8p{Va;?mnz#9=X9dou~1t{ZsQbu*SZS!2Q%MZU(aBawC2snIfrJl&hfQs)q*arU`CZvnqw zmEr|=V;q+{fTb&=F*a)#F33h=qirL0H)i8fegYoXX7hX`9Ss?gq}l3(Sw|9Dw!wMDF8!N}Dy>|)Joo}R*w&DVcQ2*SjG{kB<#?Cx zA8oul5y6|Kko0aOe4qZNA2Ma+^-ULV(|6!^&N1%2&4kKV?uCljgeN&OF`%N%zOK+` zcd9C(9 znkmLEUWus(#$&yuK77;1GCwB|7C6e7U9mA{V*Zva+(C!+I7+coiC~(wHtgv}unG4k zveCW@?1K1D>Iikjss~H4#BUjAu(aVbMGCh*M`FM`7ik{VI4P*XzD3+|*cA-90t-~0 zT#vPL5^(oLI6{`JL-Iod=)~9HqQfQZX>W$?>WdKNyq){rBajzQp{svfU^^hiCNI!p zlOKKN?w$`gb^aIqyu;s38iqTE~?v*aoWTRf3Tk zKM&MIg66d%*C7&{_w2>)K^Yi!R0ln7t66XSQf6rI7`9g~L*j@k`>dtQq&Saqpw0v9 zRLx+QI2&^!`8{eh!JcQJbI%{3nK(r=$A|M7=Mh@@&w{Qyo#MWTKlpR76IWU~aj^a` z9{$SUKHze`ChvvxUPow5ip1fnSOk8^#`;eeaV~#SF+iKY4n|Vadpcg2MgK&sSU&GBd+YLhfz)ff zC=+8kDr1;&gad61NTZpvV@RwbnsjSiXl{}y)nq-O*fcSC&i_q$65puD@dI5e%b}hF zc9^DA06P)x4SW}gseEQ5EfaxO^HJ>b(}QgPTul~lk%QmQhp<13nlZ$29DY0RgRk;Q z4D5@-=CWYy+Bp+b0>?tExdtbF)Y#t-sZhL}jgXfq_&19CZGInxili?>rwH-JdKpAG z+y3nPF)Ubf1-nikLHEx67;?lNY6Jdw{=R`Vhg@W@KObk=b1ty1+4mXgzhVo|NHL$T z?QDcaAnWoU&%!civXE=z*|kMH%iwp~pJR8y$6yN9Tg`**g;97oinm&u+uR@TU-Pw^`mXBVqrtEMe={8eyqckuc=<8R2}hNTKq(DHv83 zi}E@j#7z%Ei{^2NfAocD;2t!WJcrfRW1x{o;NlsBX|oTZihIFQPfcS+LuHv;(st}@ zO2Uqy3$T!w%KVg;ut#DB%ww=RTg1EQUO&HKhT2NJdcpJe;SbThyc9;#cQDz~5LWy> zH7b_I=hC(KYP<`p-`T;)uz_r2SD-jtogIs}XQS%WnVFL`8hS>O{bYBnTOh#>=4!JK z9*KC?KMJ`YwGn!C9kk4@!;jBW^ZXi-H=>Kop0vY~cYBf_4`J_u52Lw5 zmuz~o1)ERYKrT73=f36cR5b&9m3k4in5y(nJ_+WhdR+0v^76m zIJvb+xK+DYSkSRixYT=sP+40+7&&=1gn!H;0o`G3`1$@Px;}9ySWz#1k~JffN%LV*{2Ym&#aI@1C@uFHgdaQa32J=0 zC~&0(dg8pmi8$J{>#)H6@JrKCMbog;?F8=I-$HzJ6JqbiLpnp3W{r_yvY$6HlRMj3 zM4$mH6iVTy-G21dO0%h}53)~PeMdq~{YVq%; ze#s|DQ8Z`Cvx3;)+fq#A$`ksjcL0vhi#X@74-U(tadL|q41ZsxQCGupNu?3n<_ux8 zN8E?5>I69aN)v1vUP@(YbGRe^5+;?N!M10ya4$E*tM|FcjSppWIy2akyEoXt(ba5Z zz5)!d`7*x;)$CT{F_vJ^f~<3!U>LC<;oWm#-I52zw)co#ehdF%uabkBlwkEFL+IcI z?zsLzk@hnvxs2ia+ZWTjA4j1nFqwCx3NU8lJN)P?N5D@W2}WtKvUibe;gM)|>xC6d zJQ{~Dd#7Mw^BAT!+Lq18m0`)E!=R$4P3E<_RMLD0^|!p)tLRMjYKS+V5{A>ZAK?RVD0W@tVQ^Ls-R9`O5V|15m}Pl-9N zUB$-nJHzkm8IX!NN(W3hKYFbYLE^2LJCS>!*8dRrT8W!Rwx~f(IROzJ@vs~B5GUPN zK>f4^br*Ev<9v0Nr+FR6&0mn?mmTDATAc38?WO=*V>EO3S`VorO!GJ$3H&NpGI$+M z7IpIHPm~#P{=<6zBphEfi(=Lr;2n3(xYpNw<4#l9uDulb=1F=B%SSgLINC)<$@%_FuZ-FFM zC~QS3@BRNAmBtlL&??l8pP z4Q7FNh=ndTW}@8Ndufe66>=tUYU3X0E6l~i%n%YjN*Am+{z0&{VK|sW404`UVU^hf zEDF&@@d*XNad#2SFXdg{ZO-@Ga;F{t4o&znJPg+6=FAy?FR!3d`y6Wp=$uOnI*kGg=*j=k{0e z&v`zA>^BJ#N{0&;wLBAOE!&Ri=dL2Q;w7>z!jWt=i5?b3nr^l)fK8NUR~Wd(;^9x!+O;2 z`v#k%RM{_Ec{Z?R0rcbtQE3n(&He|tQliP8&e32x*Cm;v>`{Ew@1U!nreJwp5K;p; z7iejQe4AzTbjT{x;4pufRKCXy19=uFHJEi?6Y@D+5C1S|Efg}iSHw!4T~^9~(&1Wi zOc|g(F$U~w5!h+DY~)`{B#*Y=6wy43XBeN@`HiE6>(*RkezNB=SY-&N#+-t+g$)*3 z$g!-9nM|qGgN@7gWG{j);a)tU?B^h+W-4Im2?FMJD+c3LmkC-!8px~A7vFo5vF1k< zBA@uvCvgjbMuiI7T^P<}wS3s4VX>_G#b)-l=mmytK8G$@1-4i%9Xkvc(fn6i1rP1H zkoDssY|8utr-&EPDb0lPOwnYnP! z^Sw`tFe18=KJ(s!aY_V}lL%tZ(oI;K%uZKcl}E#O?~jqFX+BBZ!^t}j zDgKo4^8#J?r4R47&RDe93ad3l;U;g2+UqNj@~jrUpQYL5hjmEj9>IWV>C`E|8hbuI zfM&@hOzB*LyQSXbw&@Yg<(z?G&tOb?G9MF93ScM|OX z@M6xrHcJBEMsl1Oz?X9@k za|cPh6aBs{AKN7B@TK+vN{zG_|JE&Q_WXhQ^Tu=M^m^>tz8>Scrt&_iH745+Ma+FS zVa1t?!iTokg`eWng-bqc6*_vW2q%<%Bfo3|G}YLkJ5n3Hm7=`cd5<=)Q|6rJ8kDXH zf&7cp81*0p7R(s_N*3&`#vW$!>?S+AJDs^2EN4c7XNZ>_jqm#NF#JF)W^3eQ*o2+v zydVbsPuxA^8;qkd+n|504C&W%adz=g{PYRpd+m1eK6r%$p5N)^oIbi)e}!D$=F;IG z;@JP%5P@eM@NtI)gySkne^xDxEFXmKQ@ZfYb3?&_c%0=-&oA42Oi-&r?>hn0s5WK) zd5W=lw|M7g)O)sR>M)^9wv15go1AcQ^Dh=#cAUvN?}v8Yoec`}!nheuJ@EYp@`%@BFFCzuNAr2^8G7;j+en?w*6tbOi z$ke`08!lKuW-HGKYV9G|q=ZsZw&^ftIrKa4c{gJG;^|B; zK8(#P2xULdtFu#42QdAZH5{jI!i~A%+ylkuil;U}`?M5%CqAYtYp&2At7Pi+`%1!L zqG;GuLaLjA<=Yn_QITh&GMT)Kc^I>!xc_-Jcehumvwn2}6Lr^M(K%|&;ldkiTIvMv zgLb%GUWw?M60B2Ng!xU9c=$d5Nw3p!v;8ZgFHB{-N~PFu4erRl)rw{JwAjzzGgla_XqtR1}N!JTT}!@6OICr?46w%$Tqd#Y*!r@a_oQJ{Q32mj}j+n!rPEKGfH( zhZmnY^(MMuXTw!;y1$AlE~L?jEgp0;UJ3`S*P>Xpj~18=#@U6=kO*4}DSZzZ%e&*6 zUk{n*^z-La51TcmkZ(HypS8?TrI}1Kb{A90=N@_;CxQOb-_$f(18cJzsQBY5d>?k;!xk?oyqI(+S(rxOgJkbG7h^?>mDX_v38+W^7Q?!}`)*I`nZk z98)aNI?V$6k4M1h4gbwgWz;R1LF=6hX!w5}6gO@q{QujG+(lB@seA&#%l}~DsDORk z_XjT*W#Ot5K}{hQ=Hq?v+jKblnyJgK_&vjl<^bfrYNl5wI3wj^fZKj2Fg^J=R?l^X z@m?SPEF8wt+TG}QUWkM9xX-O>9(J6UK(Nv>hz?4J%NRZrwu#5$VuDiFG3-5c1{c46 z!LtjTanEQ)P&e=X*#0Ax({3m_9)!)ihq8`yy6pKVRd!C;Orsk_A@AppRr33wFn$hP zk{f7GV-3wROXbe)TKpKyzm-+M*W|5J>6G{^GHwec)9Nr%l&T>;yEHl)tU&d9dZ|;p zk5ow#;hd$6;cJSknJ6-b_mgDvP&A33M!M+_Jc?CfG5-Ci7?=yE`2DzQR*MYj$5=U` z5c&gKc}8189(;CE%pG^-o<3-QdkA~I^{{x+CsG{TK)<|(z%Do*6KykLT55*u(+}wO zeQ6AvB#-Tv6`_1=5NE~H=}r6$9EtGdcbxqgJ=+%_=4?lfrxvX3XVP=G6w;bJ5&=z} zbhk$wUyu0F{NgUuKODw7kEpOeUb-yds*p|osl`ZkGqTcC@psw_*zNg)iknCA*HaBK zzxnrcJpbX4atGF}Rbm@osIcazzX0xvljZM*=9?Kfu|OM-@{0(qXGuk(nRc#GWGF-1bRY5)WZB{0s?3IG*|zQa*l%G){v)Tb z*-5c1WAH7e`XZBcZ1ZJ9i{E1vpDQT2r^B|a0X0jc*!W56ELU2L{rMAy86#uhcc1|E z$8+EubPbC-R={P*JDPVTj~u7TAhL2MrfxRCzOw6dabN(OLYA}J8yRaI;=pW*eVM3G znf)ER6lxoCf$S3Wb0@$9Ul07+^pc9w=klE4F85!FGG*fyCGtCv) z-7D3opEU*_C-_0)TRkpqeFT$?cIc`+f#!uMe2m=;lZF_~xc>(Sx0o@9nvCQc?ab5{%<^KFAcOyg~~|dys!JIBxxxl zl924ALVb`Cno4CfwF^ll8A+0r5mE_JR!C%KWmEXw-``*U%u~-f=RWWEb-i92$lE20 z;*Xr7&l~!9=WH+re^7@<$U<2CN59g&wgJmn?H&)-w@{izRLSEEpWTntI)!Dq<+#nAGM|=CL-&| z_rd({Z1Czd4D_~OOjQ*L=ZsOR<_Z57TFrkCs{9Odz1S5azE8zJ z2{ml_@SUcu9*X4;HevMUvtE66Fhfa@!cAbzSqnv(|v=PrM)UU zVp@dksZU@oorA0)hp_qCJS3=Wg==Uu4bRV}7{$W;>DMB3fKwobgQew8^bkcm(7b)ZUyI9&% z`jsM7C*!}~op7DL8ONqs;>j*;bpBWY=hX}0Y;Ogf+oO=8Fak?9U*lEvaq4AkfT~mZe#o7RGuw*hlj$$bPk5fn_~7SMYjE=8Iw6QjQvzIXJ1H8X>9lv89j(0EL_6jf z;q)CjD1G951n+~Cw#JfkIFg|6#JRpy#>5D3u<8lwxAHTUG99Rd64p+{ULR&J~G+S zLnDVqP`l@9v}qNi)U6efHu~%qXJ=IZZN#G;;RyJn$xb{}Vfj6+2zhgubG_!EjQb7l z9mvG&UB}@qQHu)WS4iisfRT-};Vs#S;FYm3_~?OQ+>H=>VI*45E8~@YBHF&~M8cO~ zC@zn}_S5q)*ZMkqhfQP?F3T~Q#odss+KXmo&IBB9OsniJP-M;>$_s3yW1Ls=Rmp^c z*BX%XoGmnpXSpQff6>C#*;HbnOfI(k95B5By(1rBhtny1Tzv+u#e|D+M@3r>ZYZj< zA=2WkYUE8EdX|Vf*B2C)=P{MIfVaJO!E@p(zKBQd33?-I-fYo@X{vUpDvtUe#wWlS5dR^#S(rcRX7$%aV17 zaZlR*qgWl251;q3NSI|0k&`j+C@Djw)*m6Au9*C22C|1ZV3@{N3MeoT4b{jOja*$W zqVH9r{I&<8Ig4vV7p8R4_FHz`E43V_&+ovRU2%wOv4d|{2`pym3sv0 zdNx94eI&XZD?zR8crkYXn=Nr4oyptLmw67WkMVu)Lo2rH@o@Hm=UBE~?az`D_`OG5 z79;LiLSD@bm$v7DEF;huo=#)d=Tob}F`B}AcMZz6c&@P&&o9PeZ^1Wc1@~wB+kYcV z`Z97VY_YpB5X%BQu-teRe7^X@tJ)tKZrm~Km_yN$-^smP64rj=ShW2Q8Ag=S86Vz3 z*yf1!Z9EhF?kGN-y#$TK9o&JTjZYDNJWH5`=DrgA(JO*yzgbANZ72I;14zqAV)Ck9 zk`Qa8VN=|x-ba8iU=QrA3UJ`vDqQ_#2<4L%wE0juo%`&7(w#A|Xjl&~y{!mq7zE4w z-SEHq3+gf7VEgbBoc^oE6GJCAG);~||R*Dl%Tm4KFHsQI*@=~*?f8H+9%~RSU?bOj&;JM}hnPmNiEr=d! z55L8aX-!)T8EVgli+Ca)<|aYfU3XESBGPn0ZAUV*lvRLE=VfI7AOnr9FbS}Fp8_Oz& z1u>}+y?8j&nnf>)W|{Z5FblhdY=?;+>(TguK6aVT3^&9hMOP@q^Indl1B$aNDAc)> zOiL{gvf2qp`%g#Z1!c}1DWcKad~qrL8hm=b!Fh~4TmI!1&(7Q9*Z$WKV!#q6IIu(U zeC~eVhKaqZMv(Uw+#P)f&Gox*J2MgE&M84!ct8U-XJP4JBjz4%&ipw)EV5XUrBrQ$ zb&nI!vBcLlp6Rx6SvOTJTHws1A+VU8`-~K~r{~BzI zbi>*ne>~cp4~=X5+AQ_Luu&H1IR1)`YUfkN=6)EZG61^vv(R(N191i#n6W4pn*y(* zHMQL>mN19{+SNcQgJvq2#dUYA@1^z)VJQE#_DmX$lZux!X@kvyo6rW0Mzab#dNQOxH>eIUk7Dy zNVbEUdpF%mK8F)ml-Mx8i7e=xKQky-W_vTX^Y{383fAPVnl01uuFVZDZS!CrHyj6* zmC38PR zY8FPA9>Z(*{%l;83XAtuWJXWRImdZAY@RH_fmmI*aaWCni6ZOkj$@hU7qXR)Eby=9 zKhAjAfV>_5K}AUlGvBn4?O|o)`YS**Z2%tE>A*ek3GH`orRP5Rbl$Xqe&~8_`dNhv^L*|zDYQ{BWusH zt$B{Gwwyc5&mfvJC2y`MLs44+j=rj-?fwRcNScG6WlPXgmxb&1e^EtBY3|@Lwv>%7wG*7IJLc}f>UNs*^L{G&U;d6cUFA5HpAOAH zGsI3lM0KSyIPW$M@A|KS>L!1TdFhYM-&N65@d>XtIk9nk{yZ$ekge=G36JV0G<)SC zp{HmzJq>cAUpx*d=AuL`|HcUxiJc_>(Hoikv{)J%361WFsQ*_?OSH{+myOR!4u?YK zi9b$H)Z`4cKH9YHC-y{{u~N}gCTknOWYQhj$Xf+?DAhtcHW|Pv=nuWD91QDia`
RCv7PB-Vc})`o%bEg z;X2~f&mvDg1}aYlq}sh;^BzoB zSjILQ*|F9eGMF@NC1#w+<9pp2+#P!cBX2t5=b1LTap*tXuFXTlk;xc5aXP|o55eV; zBbew_9DBL7j1_mavU?6K%*t~qtDrsb-_e1XHtwvuy#>ae{ZX&7i;U89;Zo1vV+!wJ z`k#B;rK-*R|KmUTKji?dbU-KcqU zkzc2dkeOh>Cd?ntnzm@Ng}DmMob#+L`8@iR@YDc7zGYNvTul2U4pF-BmFC2!QQSWz(x2r{GY8G1 z!ubd3)>v^k>t80-dF>R|ItnGv&0xMKhWlB(A;Y^$CoZ2NnxKRY8`a@6#R)SHJK+!K z5%j6*2)cig1k~CJ+5b|6B#Bnx+lXVr-QRx%|Lt0Ie!3|w`r}1z`xEF}&LZlkJeWq! zn?YIXQ^-Ve3a$9Okes%qQhHXOP_S?%IcH3uk70A^)c*byX_F!(yxlJ-w%ikrg=&)R zPg9C(eklm4TZFBdYw4?P4_W-=zT_dB`PLLhTC+)N(dZC^|Dkb(5?I11c%2*^S zdleuG2o4e%{d5q;6uo6Ju4=Sl(_}IpHqt6i=$z3*q6y(Oy z;m~;DRQ*yx)ewkT zyWyTj1s;oE!#ox4oFDxP>)m58@p_LS5gz3aaY^J6Nqc&8M_Iz{L@cT=qei&XWCF1gO0q0-$ z(kA5_bGBbbka`P^ZtO9oXQyC+TRPpvsOi+c-bl8`5`Sz zx?UH|_-v!4zX?xIr-OQ|DJla@_-O@I42@m$>$d{SC~ zphvNsdE|qYoQG$(OP0^${?ed!bNILzaIWBYGF~J>8&oIKi|Bm1kiLgHKHVpg(PjFt zwVFglsdF)Dfj#aQcNF6PRD~NJS&c5>r!d+hx4>fZ70n?AV!8_(}?uu z(7RAqqEo5V^udc*jTGkcyx%XL2i*RC3tDDQKzZsw6by)={Ni)8dh~Oeu)_tBc1rM; zO(BJ-L`ta%qnED=$RjR`B<*@g=3OUEGwe?}C3bXOcp+>oA4Tu4`O{`Yyob?5UkBis*Ymhyx0=yrG@q_4%}^5SSLE#|JR`APWr z$`9Im4QU-mKDs2t<_>ETMNOY$RylnMF5?Y9zCwFLA$txo7jdPkW` zhS=XGgO4u!9>m>_^B((Afr*Bd;a+b$$sI$v0g!M!8$rqBcg`pX2IL;JUz4}jK04bZ^Y(edYT~& z8()&V^&Y{S=WxqSP6($yM^maqJV|erC)-upv}vmigX4{i)Z(g_Ia%?kvQTLFHI-au_}yQo@!S)o2;~1MQ8HtRK%l z{Myk9i`kq>Q{F?GJdbciMCed;L!)jCu6|t&jazHz&maETPDar0;(cVWHJl=&tO;hP z>BUMlJoqGo5o(jLf1m|YrtG26IowCFBniJypTn)&C(yNk`_xJ{VR-rjvJ}%mijElk zzAK<{AZKD!9iW=y_b5wq1V;9FBGoY#=Ji_;6l8~)`+l?a*&9WJc-Mc=hD_157bPNv z$kn1-x%_FEV*GQpZ0(YMchwt_IQ2v?#)qGbZJhZ^6$z7JfX06HG-y}`$-k?my2VA5!To~w2AjfG$_9>Y{2tGLp0+LLfgD|p>Prn&bghx9 z?kRKMt`p9#nU54<11z^`A?}nhdUwslk^b&zF&>3(kq+|LOrcOu&i2?d3bT2qnC&%y z@sL$mb9od(GVjpws?oR`Du*=RLGTsxfoxhICH*-_XO%9~{GqqV!aSD3b<8L)ls_95 z2O|4X5?24I!`{D5h~!NBY?l%E^{km>U%FzbL^gihcE-8K*$96y9crI4F#1p?9%!dP zZdDFr^>~{V zf~bE1qBQAJQPAFU5f@E}9ylh5#?8_fO?fvzrBTUbmI8&}0^Tc`?&>)M0_69^u`CiTHJ)4zv6pLw)ys z$c3cC?S&T(%GSa4{XXQ*Jck$gk}Sjj8y+mRf%}ObI&x$rH1Fs`;phmc9OJI_jt#Ww zmlp2cb%1>f=Q0{a!aIR8kUhk(h38AFe50UEv!ScB1Xqurfo{M8n8|cwn}Pz9lu_W0 zYo4JjGG^Yx?(pwTKT$w{t>}*3MA45W3q;Dh>_zXiN?5}ETl7nDH#&WKadLtzOFh?* zt-TQov&=N^@;Q!5Q6~16bH~e|%bYbj0*%x6;?}ZyY<;50)Mmej=IIPL_bf+Kwhl^; z@V(#bX7naLM~Fuw`nAo$(gFwMT4!M2=>}vs0{SeQsA`}CmK4uMiDw~>JimtD zXBu#7rVS>pn1l;t3?ycl!AElWbD83Qp=?^GEHh4% zU@8gb%rGX1S@mpY&%A8dBHL=j?1)2GV+$tcpU2+AojAL-5>-_jU|Kp5d!B5iCwuwGloj9o z1IT|tj{Run{&6_7-~JEJ6dplY{vhW1#&Ir2HPk2U=bg(tNVk0n+XN4MzaTJ8rx+Gp zAIf$&EM_Vh`mBNTvCr_V^4O+$WC)z?Sd@ej<7MDCe=`*sDq`3QFV2SZA5;!fuAds z*tk8vAid%?%46GbF0=%b_#AY)?V)Sn8*@6S7t$9diBYg5os zRsr({H>|iA0P*Ml!Itk_uRJ-6e|(m4>Dy;g43tIZ?djP6%Z|_C-SGU;8aREH$6$$5 z$k;B=Y<&Fku}fb}Z|RC+i9SgTwdcKvYZjU1?UAclyi<-`c-Zfx?!7UhlvBl#DcAO7ohuqyM zc)2MD$`*>qdC~$QM4T-hCe9vpN-&YeH;nzb4l??Z80?`1IXh?0sLw`c*dE;MeM9}O zDzPQ!jacz(b2j{|h@H2SWoP9>(KRFq;=z@?qtVOfVRGz3q#M)pS> zjMsM=zJC>iMdxfOn-xc)u%A2MvHYXQfE)6$3qp@ zk%~$b^&Up~oTrfC&xZL|Qs_SC16oc(AG*%CEO&U8i)cV-xDNP&1&HP3<_ z#HAWJzLUL!5{nM#Z>+`*Qz^#!4A}8uX=r>Bg~&yxG1*a?s0|PY{7J7|mTWe{7 zrwwunc|SCObJ$i4;m)i|dbhP5w#Ot{C?+xU|0c6ThD%ui?{?ON6rerfE*AUz#;R}y zR=VX5ii}<1<#ZPwHO;uFDa(T8#h7>MH+-4@85u>BkeTL-+D(h_D|jlVc}&3xi(aC! z52zP9(2flIGlm*x!``|5i8>Kj#pJC2ne))DRo;nLjn|HveVl<+2H$z>!4G%_s=iL1}a4zzK zjHV3ljP7I->ZjPC!cJ!Py@mPNMX^WS=1iqjis?<$X91SFoV9GtPHdXR*g-|+m*Ix7 z?~lOt=rx2Wyui29-DqA>gCT~zm(7njY{EG1C7%Sz>*H~-w}CeE4DhsjQ7q$~FN;x6 zXE%fM*w=A#%&u<>rew8Z?Z7&Os-8s^XV}ht7L0VQLHMH|ft+)PVPbX_pSUY|@}nV$ zw@Vc59QIIz@Q0$e3to!?awV<(rydq14d<-kHAUENa1xue0@`iaPGWounNEho!1Bp{(#Hv&FGnE!aVt2 zbZ7Q>9OvwS0?y7-_Ey37GyIGlF_OEX&GFswG@Z3*rzPHvLcM(=K-k_`I_vxM3V+!?nL9T6U zDCgr&TDH!RhD}POL+x+q^@4t|-mHpA;=^Gw<0&lwQ=Gq3($tq!*G! zgCpmW^sR+-d808E<#DD>-C$b#@UoELzLDN7&7dm}GHLXE7b@wLr*Z0CR-vDD@nw4; z9Q5sB#rxD=?)->#&tVA)fgqplUyg7Y`j|Z{e`!v`Ln4gbNd5! zZ}lWPI5(eaPDazxXZuNcwIYpDlNPc$L;XibHPvTTkllz(Dmt1&du`td<0REtucs4h z+-Sxe9Q9dHC(mAlt74VB0gbsAMg4|y4uNMm6~FPLYSU3G`A5cM zJITj^yK`5~q=8qmDDB-Xl6AR3U0#>yg{cm`9bZewbGu15wV52^j*&)l8riS1rXRw91Pw*1yER59o1j!G8n!hV$jDy+HUDd9Xg`UTU?aB9I+`60k7lFn zqnPzdFV^|-H@38vQ+u#6N=ERkzN9teI1gh#=ONZ`r=3I67E(03MHT+%N$;*D>GX3D zB65upXgUc~Ri`3u+DxopvJmTb_QNvcD6$=PfW{?Uq_NWH>5W+`d7c#~*++Wh7#cvo z)qSYRT!-+mMJVMxK^sLu__X!6U>-P={(dPXsc-EhrJ7Cqp6T-()O?(I8w%f~Z8%Yx zj`|bkeBU2~{@c$YsI&^(r1%aqs0a>=2cTB*3U+*cigk}4VnpUmq`DkK`X|mHS56@F zs6!OWJKkCG+iBADIppKmEi|em3T@h&l(cg!8SjdsF5XSg9`{;UH}U~(9U;#@aCmRLvJRNM?8%TSnB$3RJ z0+M)BKtqNr)9y(nLJNr@d=Z~3YZyXPRlq$PON4*;MONKX)LP*{_SNa6=RJ>Z@Hui3 zixoCDPeb5m4;)TgihKI&@p+32s+C*FucQb@Ynm~c^N@$9WMht|IkGA=`28W5{MzQ)5kT_C}x!bwLeOwasAIzIzM|j-H^hl!rN4tw2j_> zQ$e=>LZqH3K$q7!{E2vob3ZdMsBtpR>+OW&$~`C%9pcaV6m07LM4K9y(E2yxC>Wv$ zUJ`*yKYO%ZlEWL5653&Vi5`!xp>xuENbK-*`gX~K%vHA1Y1dukul1Nr9^a?jc0XD~ z1F67+uyN#SSZm~=v)B*AZd+j43^nv-YO?Wjr?J_)Ma(vG412ehvt8ET!!yG!^gFB0 z0(x|rtfwrCOSy*c(@il?{4-VY%zES$K3lr|kruyx!ifo`lw`3CIKCB+29&`|eg|gF z=iN!M+Z0l&h3l={**(e<^Ky8edXobpoX*jZ?pfTWlmHo-bcD_}L2l!FIvS)$DlbyV zQ|&ReF0zK$fpGk>%EFNEkDxDpmBv5fToLWhWSFXi#Tymyv@Md{4iu8ywOooY{Ev>m ztfj4EYe`K-iVWWjr=EV#gw?Z+NX+)PU_a!euw3+4s0sbf*-S(6MsF&llYx*qe2BCYZpB|K~j#DUF8-0@nD z0WMXj8j+5SVT%cx6;Cy9XBG{{tnE^XQ=+__yL-1*)oocyQHcXkH^Ll0AW`02h?>%zt89o-*b;WFovf7>JxB%1~PVKVSH^a*52WMpdsfFrs9I(9crQ+g9W0%x0^&? z9KuBpgJMLJ?V?1NI~7oJO%vxVl~J@^7K2wRAX>(O&YvlT(c4p)$h(lD1qrZtWPq*O zf62qmgl8O8u*uCB+xu)VYS?(pNI%K>A~nbv(TUfQ?>OgSKU5{tQNBn8%U(4KufBIu z_j?Q6TNi=&l}B*i&j7a0AA~`2iB#xwme$$2Y?5a^E+lo)~ zQt`5t`)Y2M(p;M!(wHlSvlED)(QR-vPaG*Bmi#?&9X-+MriZr(zpZNM(B*Rowvp_Q^pb5&4D@$u z<5G?Sy$RZ7m3u)Cj^A%`j*~iDdf_zMme1p!ftRF4DMIt4YT|-X1g_hKk9JAup6(2h z;ta%`zlP6)rPyxo!A$wT`ekk$y_KDD~z|(aKfDBAt0IB2$^C?9tJIaM*Aei*+C4RnjX& zdtQZDkv%Mg9~56+#(TT+Oe{c)`KZgXA*yrnsX>M*gd{Szhd0>7pC{M{dn2}S+z70z z;7nNSD_E4!h29M|%-33$g{X?bXuS`*5B@{#^A-^6}bXq?s`|tKNoMISrO8Va3kr zU&Pm4DafnUWH*BKnA+|&EXYxsJt~gGOL2j*pB2n${4JKa>@z!bqLJ+zGmZVI48WZ^ z)fk|t#5#8fYgz=pDf&FEO&l<&=UkFUOXfN``#LqxR!1Vm+8;B z6Y)3~b`Xn}Oo#K{hxC4kH#Hs!fV@bB?d|hs)e50(Qk^`ze%=HdWqu1|PDju!!xNNO zGaSw7ei(2-kLOw}a8$;eyK(&3?$|yoQ|4aAH4aqrWG-5CJCHthD+~OvlBMx`Txydz zosK9W|8--LH!}f^BJSooTLzWZF{o?(K<~Gw;q#|@*v-{sehOVU@0ZOT+b!0)}qp^O7geX?dkNNmlNm zmeMjx7;uwh=Iy2#*%5T}=1ICf*$KP&yK8bu9U5nL;YUC$rVe>dSwYKDeq;^`5Ae@4 z(3Rc1sK!jAZ6HMN!>X#62NIRB~s9?!W|@w+mPmNiX&PH$agVh;p+m}JFE+kLK_4$_FW8{|>bCY?|zJ zStj$;hP@nG4Xy9y>Ad4ozdC;7QDB{|9JF^^mfqCe-GI!+{>4>HIgy-eGLMt2y(Tuo4Xe6j-;4 z3{y?Lj$fV<@G&r=uB#76OwNXfmHJ#_A|wqWsY6I>}Ggz9@P;X*DMhVCrqJ#qi@sV z#i4NDr4G}Q$+W0Q4R0%M!AeG!wcT}R#$)F((?927b?Xf|^(?^amH9j~5ryQLra0^w zMzcLOp)0Qo0g-ylwDmG_7kMEyTno=Vt5_x3i~8I=McKbzvX{hX-Q11) zH>R*X1GX@)ce(6L_6D|2ZZqX8#N%t|2|UZ~!o#D3*bdoa=;t?!9KR@l@^8~oK6Agy zGbaHbEJ(`x0y=B{V-1t`Gqq#QYUe1CWY>~T6~do;0gu4_g0Yl}t22RcOOet#4#mf9h@7W9_7ecT}1 zunWCO=iwx_4>s4I)0b!2XnHBIsaVNo&evy-8BzGPl=HZ`mtjV59Zau&$8Ae9#!u+X zM57Sd+cNn6ra${>Gm=^IjIXTjD=c3%9v+S#sq8IxbGgsueb+#^gziFiqACgtFVUL4 z)#MR#l@iOuG3U@Y?oVo^^+r=5wQeitA7(@DLk{e&x^WKcAjCf5&+EKunz3gnj)XZN zI#GmI^Loh@#kBVHaNc#80dvj*m{HEZBOdM)d(=&M?$V$8-*-ZRbIt1buI0*31tfUX z3kjRW@m_Zg^ef|G(-(qnnSr>o=Q$N-{-is{6wvK|i`+b-Xr4l;@TRv4a^HU;_2Di4 z930R2!3>V=MKrfi66PtfsQi_Vz&kuczj{8t?Hq=gb4#%EstQ}+`4bnD@8P0h8Ep5A zMO}y+_sVtP$z6Mi6G)%gC2FE-3yK>7}lyO(OOS8YoQQ?qi zMr*l9$7`1;dS0|>@f=mrmL@*S-?S6KX2)UX%iWPB|6zLkOIp2I9wl*+IQhLC4ufZ- zcIZfKbttFB*P_^vg%!-RG>ch=Ok;mX+p%xoTOsoABXzZ01m8T*c?w&x*mVrvC-f)n zcFwY$pv$r(%-HMl(QHL_Ki1DA6Y649>6D%n_t|98Pd+1ae|3=DD_2u|h%+v%PKBVe z9Y24M#hS?~n7_G#uKpN7^Q%4ZY&iE}^*zE@&Z;Q7p@pL+WjHxrm+4x{vc6$m5c-m^ z!Zr#%hL2gsNG8f3U?s9D;j_BAN+OG!DeP^kG4cjq#O!|qn9X5+S5FP)y{bae)76E( zX*-}=kJg8|c=^)`S#!UWR_tpWzxf%TO*sQYONB*E(coUqDyaDEq-3kzkUcHK!vFrl z1C)V1q?U3Vr2G^2T;3wr2eRsgK5TslKdl53(x3_2Qf0_|>x;hnNBnPNz*OxPGr#rPYlBmVh+z2e1_(GsvTMSW*gKo%F*vd#__KPj_YTp(LGSPtS@i90MY=ijwEp*cMAyt<1z5Q>_HcJ&_ z8oFPxXXq(tMRGU%-McjMKrf`GgfMy@!6v=;VR#jRsmH<*r*{r&b9cku+8=$J6p)>F zly*P7CX~J21TTYgm{NQhhO4Szl>3^tE*y{gh1qa983v=@QJ+B?{9Y#4(A z#(AJ7Ls-#I8TQ6!BfMWMzz&~W%J3P3MgtpE&J)3<)ev^~F3|W@Z!qeuIn&uaf|Vcb zJiC?=rU@tMUJ4ILphjUP?fg_q3%4~=)Qg97uYkWluiHQ$&Z*(|{W)0o zFB8AsrQv=V_jad!q@)`=g^JB`koH@RM-!@e2SKb7Y?Vu{0eV&2M$>J=1(VY<*?JI;kqK1yE(tJ|W4^ba za@NLW)cNV4b;Tx(vEjUi`8*RdejN_{Tn^<6^7!bn4cCIpAhq;3&+)ItdOkDV^7td| zu>8Z)WmQE-i#yne_7v6`VaLp5_aLvKOsEfgjTE^>Y|`);)@$g+9-V%UA<~{yY%m7f z>f4ah_z7d56=G`DUpi_ni^_=$q3>glKOJg_<(WezE&x+|rpNEIp6p8838w0Pk_|k* zi`iEyGOLOL6gHf|rcCa$8_DxAwXV?498MP(-J>&4zmt#ACgdoU;N1R1+)i19?v2`_ z^*>@nMmO?Al};t1>%}=D({4r4r=X<>KF<4O$6sL2$W%1{I)cxC_G8-{MJ6?DFv~vF zf#2a3h#hqrea~-Dl*%nyF>NrWEUj%FN?C*pk+C938$hVcgHOsjC1zg?bj$6rdpAP}f_H0ZYF^=~>PSUP&IowxVgKH71 zvD7XcMa!mR)>j!!HJk^LOfp{HKEin`hoR3IOM9&}@O(1=t}cI%g=cp9>kK3doVz20NAnb%k2!JvoA*pI6|Hqli8Dw}M%3Ucuh? zAII8%iL(xVJ0;!oU_eZPE91{+^SCX!Sw}8ul3e*qmn*qIlmtAM@YuMz+UdcqDYwl7O|C zX`9dUCUvL@$c33SXT;ac!?#^m>20YaKc7V7v&>;cIv+rXk|%e4E+&KWBlIGyA8Jky zMt4}C^`?noZmJKmr_>RNDB>;L_ODeJw-?oN;cVv3X$xQk zp%6baGb5Sr{r&^;dFuIo@B6yW^Ej-tk)4|cgKLc3pTeD12XtOO9M@;hL4lzjK19dS zwbF?=c0E<>OR8`x`T@>p9ftdB(dFuU8|$-d*j~|&TW?zP&37hj9{T{!5B|XYh6Tse zOyk2_`tZ0%X_%*&PPYU;jI0n^3VxG=;>I0Jr?d}K7; zytjk4&p1!_#7R|u$G#dQ3k!*J;;%9~?w%O^yyr3tH#J?lyv;uO!i*QM{ z7Q-Jm!|7@kB8QB|J;_l_-0%?caf-aMyeAuP(c;t1HHgg#!-K!67*=}(R!-rF*P4Xq zE1StLwTcd>3-8C_QG)*}9N(iuU~lY*Q75O6_ujQ+c5EN5n3_uNAHpf}X&43HkU{(S zS?K?JH=f)nLRnxY+8<1ZP4}KiGIYfHFn{#w;sse{Q%oM%P4xU1u=DpMzP$SY-&y3y zC&$ZjpkqG5`t@O1udTe$GKgi)EarQLX8iYF4hHHO;-B$ktQ~EMbyZS$hpyurl5pZ=){p$7zah9Tbjfb9{^y zm)2@=nSMTozcLY-D#6CWf7*+O~mWUMBXv7ZRwO@cfK$o`HWZqGLO%S~y}`;jrnH;IH>2mlI=2 zC143zDX*g6)mEgSr9dleQW5R?2{U&o@KsM;?)YZIBgb@QHU(d8Be_vnOfrI*RVwUvfET|jTFhl;t! zX7U<&g%<7Afxkl`xyj|zZXb~i8RJX4-usi`EJup;IZWeMccNJ(X_kf0%1BZ0R!_XJ z#_f>NX#dg)Rq~T)PEZv2?vTOqclGoz#|XVt1nsPnXN^l+;65 z)5{8-E2qI^%U-njMC07ftN3+19wpujFmXW%+He07UX~ub==)2=hg?AAdk0KBG7&k~ z&LU)M8GIeS!S0t9pBncN_sfM-_{Kzp{tQH9G+m=D9ZzWg$Q$J8`hj-*45#7qc9W-OA;#V4!yYxg zIC+KzH%%VMtwim+x#df54mHQK@;J232!(;g0=zHs#FIE<$SjM* zn?Lzt9}t0aTjyeTHyy-Wc~3E42BF_|H_YhefFSYQQ=VgwPIJG|ndEyS2d~1##yZ^j zaVM6!e-{tB3m)g{i{z5k6{?R0q4lN}rUVxwQTa4JPTN7wfA>>u>08v6 zFZjnDp^c8cpb+ee{;3CHxHbssH3N}klt#I>4dmb@MVaz?tg~K)>cx8a>o`MFUT8}9 zzFOdmbvLoMen}N`_R^f%&e*hJC^_uSmiX^TrjeJ&VDY0-7&oGxZhmnh?YgPteL=YD z2geHs`2swTzae?xa*|53{*iy?8L}wKq|s#u=zVfHecGsvF+RCCeB?7^+>T?xT+vGt zyth|Nbm{k;2FfYggm9f?jA>dSICj%%zU4I7e5pch?`(9bNykJIx&4?>ta@|^l7{;t z`)bLF@0RfIpeLB<*A;hc2f=CDB&?cO3FtpWhssWP-${{f9#$*WwSOt~-}_XWIx|bU z(r}zqJ$oZ{`|yRdt-C_s${I1!@o?HxP79YGpy|PC2;vw>-Q?JL%_n?c5Q4`_a}lUz z&CM6Xc=?*0{GfRx8#O$~1}#@geU?I#-6GH+M%R^&DjfUM9}5c?(I*G-ZjY+O_Mt}H zSfa{-BQo&7^B&DOn@{fH8bIVMMCpB_ZzrPYlVC#++og-tHK!3>c@-~u2Ea1*Iawdf zr9(+Ul51IWNdCDaPL95V;TPRuKG%@Md-u?y}QvJg6)0#h$mXB3Pn`q1rFV^5| zrYTMiI*D)DM+?{rve)o;UiO%PtFy%jlw*;Yf{zfQ?x%!#4 zCP0`C_AVKTbxLy3{}qA&uNaB0^C7YyYRDsoL~yq#k!3$I2dTn?Qhdvcz3VsfpUp-r zUvXAAuJ+L1Hyt#xatfZy_uw|s(QV)2gY>tRlB2GsG&M#ZT2B6mz9dD*(JHdlvO$dg z1UwdZV`JxT+`DTxzP9)}(jJ||D79`JetH1^&fCrhA|trn&5)NF1yJRQ3S50|#cp9z2ct-%*dvvGUOJ}i8PLBew=W zW|`w-coYUNxQ?P_MYxmffuK9WL4NTTeYeh`hMxAgE3yRTukxYwZyj_+4kd8;8Vb=e z!Byd3O&H>dXXT^x!oX@bdjJyvq) zP*2*jOoDFbU4`$;7bi~kMBpz~dOz`|#Q%*t-p}!aT&aOyed zo@9kyC)`k2im^FPn3SxHW4#hdPGnho%#WbmS=%T`brxCoH>GF0=2P}fWipdZrIB6c zAnJ+-PQSEAWVH+|YnGuP=>j#I3ddoP5-)TM<~K6o?Cv=goyvpI>sJ~!mc}5X%Q39X z8v)bF#-y&fM{?^z5GnR_g66eCd=K4$s(1a!Dcy(?YDLDYC=v?-Rnb296}66Zqm+fM zB0snvv;Na$|Ai|2ReBHu)`d_*c&+8{p{MC(gA6MT8P93MUL$2nfW&FfeaV~Ehe^Y+ z8sV*}=;D=5x$yeY$r}blvPKFd|Tq7ppW0_swp| z9oFGf10zt^S50hU_8_OXm}xv3%1gW5p?<>#K{G~p@$ABo-+dA84_Cwv+uxF^z6-@U zaR5#x`r~H2F{b=}Pu~{dVQKk^K8LfkoC1yC3#6@f1wGLeB%q>s;38su3z08e}#BDuFdQt`4uIh&E6*p+6b{(AF)!^u2G3(mn$lvu7 zxnlD;PLG&@)t?ejta=_D|Ef?nBM3f|ev;$-y>M8Z1En#E2%9htO*clP@6Y+zFfUQ6 zt63yH_U@BZ`b*DhZ78gII(?MNM=Q}buQo}mvn4W&uj5PnS2VoKLBcuUrs-QmF6+Ym zbN-_1{X9I405%DRrZ=^aUBqN0kGc-)G*y;RNy@lI$e&JjN3aLnr4 z2NN_BvDds31qR`;Svdh)Yy~GduscTBc;kt+I2S0a5VOr}s>(P@<#EYW9-cwHRoltL z)sH$Y*Tj+eJ&+fp1+ywS-2Ko-0r}@?kNSIB_W3i#TeMKAaW~isr`dn!V`<&)eRMA{ zijoffBDaznYWtT!4w|d!SBw+x2(E{hy$ZfX8z6JJ8xBvnKxO-a5c^{d6i-VrT&ND) z{X-x>^Pphh?15sW24qFYPI9^kuOEKE_yK~;qj3>G9=?E4!6gJuufgHGJGdQJfzOHq zpxI%MTVF>&qjnnJJ{g0rm;ceA)5oO+W6Gtb8n>k(d*P|#4>}pED&Fm76eDI9U!JPN>tG{&nb$%4$5~hjH|D@pFq`s5tm1|9CUUTot+c0|pLFH)Q5`Y{_Su5Fu3*6m z<%XR1t_P12cfez!M`nLghJR$WAgL|~ZKnT;Oo%xz*)^2UTy8)`jd(uh?m=!<2+lj1 z!s*Zw{B2%>34tPwy!j0suIYk(3W5ne^9Px{Jw+>PqVajmJz!2HX0%+yr|_Hj`n>`N z{>t*6BvbA>av<+7HRqZUbNEYXPhJ; zHawzFMpuxwdMU=GDkARMa_pYfh-k5MeO=l@V`ughJ*7{yZOvnEusN1|c01FshrQuBO%C$k z57W}6Lb-BJPVib!(yHxpc-YhlBg@y}pKcv|`yRuqFNa`pvk3EUhry*qxS>|{WjW#Y z*b*TahQ|fV$gB)OWD2j(a+ntVh=Q+({_+=p+&-;^@&o16V*HEVS^gBf3N?J2eVsy% zHPUzC13D%#MOWEPSkjt|qLg^NzG#4m(uMGuE@qFPPa|~LZm4~UMZwl!6cxy!G=CWC zgY#%?ZUtHBPZD=cJ?!{=0?w~~L3N!sKX4D{^*uN7!uoG0le>!2lj$gY{RLV*ZF%uv zXa2@j7-_G8#LDTIU-prL0^d-D=!iAAsKZ9wZI1p(!@WM!aR1>@$Y>sd&Y6AKB6vB4 zLq%5Vm$(x}TH;losi+zuc!e8}A=u{#^!)aUZd@i>OIzWxX*GP!F2dSMun&jd#$Y)}f28fd5;F#DW{D6KV|W!_aKaWu7XfidjuU2Gnvzykk`5f>*j8NM8gJM2md8S(M^r7 zjm8c!&rnbp3sVI#v+xiz<(+j@Qg)KA{VSrHllA0yUvRthx3G(XCvW_q$MJzZx$@gL z*b3LzW~UTP-WrJEMhkGlI}05a6)1l`8*kGmv)PTw{62ON4@~I5z#c}dxuO&6`RicT z>VZh$vDmk6Df~sx{g0_&_w)`H4(gY5t4|!|Rmalj9o;c>Nj2TRE1bXQQlVhE8#h$~ zF?DVPxMU$rmQAATl}BjYyA5Pe5JB4v&yl}(BI#W}jjEQ#IO{PPGF^r9cc~x#_H)4O z-_B63HHZ5tKb%c2giN<4O!FF#(z(%CcPSgb<>h#N_b?*+hT+UlA9&ujh2?Nf462T& zNvB13WquK^O<#&S{Z?xJPw=tYFNtTB=xt}67oC-Bc%`ryqpPj)dAb#*-^@anhFlz| zorKixQj8Lx@#*%<&^iASn|}#*a#^xq4bR|?mu{RtJAi-v$>t%Q;<;~E9sXDF4F~Hq z`JtLBZ=YO){>e9CGeYp?uZBUXRCp>A65tsV4wKZQm}{60gHx`Unl1yED@_tE5?&j6a>tBdP6ZT-<4?heWJp?)z+hDY15C>;}K>U(r*c@0WoEcfryc~xiVL4*99fRE! zUWl1mDSE}vFhXQ?W;(n>h^#65mRCW~VBP!N9*-SlN3EB*hDqzhudU7`J}&(m6k zYc&1%Y%YgjS^q<47pu1oc=%~qp2#>qn0sC+#U7p0Qp)aS zRlKU3)%(q_r4439(uANT(!VcLVNj{SO6N^kf1@_f$ZUu4n=Sb9*9Mv6h2wDfE#&J8 zhtpcYOW14&AN&4%t6>kX*I&=NV{JHA%y-rueS?RcV`x|AECfjRq3xOM4%P2;?aq61JYxh}8!gY1thUBoyB!vu&-k=zeFBi}Rtp2mcX9O~W43 z1mUt>0NJG;sNC5H%L`pFxn~cKj+n=93Zi&l^Cn*4B3yHSW?=C<1=h$^<9iR3*tFc9 z3mn^Vy1RvR?6(xDdcg^4Z@n<-qx}n{b~7Gu><()vR*F7Ld6URG$@8!8*D$?mHHLLe z!;7t9XnmG~^UA4s-|8+n81L!Sjw%e8(w`@G8qWTKy1YPEi5tc4K<4iZ%G1oHFSGkW zTM~%zwtdmL;36&8^k%QqnLHxHljoi^=apOj!Slv_u?LAjSp5)II^e;7rY`0MTco^f z+H+*ow-d*Wh0DT$xVmyEI-OymOUtG7S4oJ}m*MD_`uro$iS3m8@O6(9SarP+D!o)V zbF$#K=r>~Xw@mbWA^JMiqBr>F4;Ysr+W8izq?&|p?(4urWW8(PH*0{U=YurqRo3uyYh#(+4%8e zE!+i1rmn>uS>x=n`OqMkkNrWN8wF#vG6J2IOVR65Jpz9xV??G2Dq5%WO|5-=rhW?7 z&FjaP_O&7Ed3T;WoLDl-pTmnMv#Xa54_5nvi7gK>+; z5z1l**ChB0Z!U}OSTLpJ9m&y z7ZtY5OXO9JSJ}oWk<%yJ@qZ5+ao=VuMQ_RzjK+6lIZO_}wiKTF>*X^A8^AY zab7t&x{Lx(E{4akaFp4(;nf&RT))!|wX&kLZYqVvZ!hGec)?OZ4~47tQcRan1iZ|~ z4CQLP8JUNJ3q7%}K^r;GXTf!B7KR)&WNpJOyt+pnKid+?b}wd8T-G7-D=DXI2|4un z{Wcmlr!&p?@PJlT_QI02>i9P9IXU0Vq-{UC(5j1f>EEhm8o6GAL*-rZeMmFCzYs$Y zo&F=a*f@-fJ0RN{pW}|U>nkNla{6oLjjKiy@t%x2F4jn%o!G;?P2me$A3BSu`n|xSndqAP` z3f*lO%`2}wXP=Ls`I%Y~ce@wFfvw_qHOLxIdN+xAoZtY=lk(xM+T3z95@!y(lg{%3 zYLTm@dF$$FYf_HzBo7sg#ltWemm>P_S5W#ylLPnY@RG!ABx(%cCP@^}e0h{-XF0J+ zMqi%jTa4)6zDSNfgD6YksA+i#J<~3nz2gC5q{`ryLq&?va$F^~0&~!eWufp-G=1QJ@I4rqt zlqe}{Dv($@&XY`5@Ue8H8MInEiY6C6CDp%2DDL}oy7amy*$geES6a`h`qg8aSW{2M zPg5x^e=#-ucN>MvLwKHN3LDt%=Vk6v&Nh95nX*?Tqn>=0XnOafs)ogsdu}fEkvx=K zuaATB!{5lOaAZH#Wi0D$!&A1MhS`J8(1~6M;~yq4{oWN8hPv<)E_=J7>ruW>aGX!= z#omXWc&}lCiEp~YA=g5h9u^^;qqW5=GZhuN_sk4>y| zSYgj(o}77)v`i=B$KgZxxhWrO2WO+R?Q9fYw-&qAP+Iu5nB4ph(uIBj^je%j#)_P3 z#G^|*V)ifA*`*~dd~l!dTwlnxJuaZ@l)?N~<`ZhvDzNsvuh@0WMbl&l>^T&HKbM!I zkoUmcC>1g}dg!)ruC!xXv2<%lhqS4xyVX(KK2}qX=SojZyh5|WrN~(n46kz=;H9wu z^Co;I-NvRvF%O&O1>%wAYQesUL(e2FF23r@%1s-1dyi4v*S`eKc8&O| zuEjq41wXo<5s$kn!@HDH(VN~$&dxHX!YP8;xjlmZTp>`JNkThPAg(gQ^Ro+berB8b`#Ih(P}j~4ap&+VN)rlV-qFa>_;d6E1?D2VdQvR zmm;$dN`mavNh#ZeV(+|_d|Nz_F5gL(G#Dvidz&_v6uP04r7xrjUZ8*KSYUjE2Kp#s z{v!j#=!$2IsTDpq-JxfVA0*>Cw5ik4*OI1ZGilG7$&~MrBhl&BM0d8EAtPLLOcu&v z+xNHh_snxT@yUzU%6ygW(rxfy3%0X!#_=Q*XY<;q+%B5A_3!`$Zx1 zw;_x^EJE21Zz$HdA#;&AnzMRf?Nagm8<&hOTWb(LCLNle!qMmUK&X5eiYGD)F@>k& z=fTDB@>IvkBb8KNE_kTdu2BAz3`$&gmX`IALB*U1I+wa0E6j`WY}0XkOHab!VBrPe z{&-MUhxH4;VARhyu#Ocz)+vYJG{XS(UwqK_<4LI6pAx);RBS9rgx(@;4DF$hW#={F zZlMgd4SFKm-bOXNlsqP6k(z}G{yk}@Gf9PXw6KM?=fu)XsVt8-kmCcxs_`u#u#&EQCgOkB1SQW^iY4BO96wmOmKkvyict7QI zc|bnhS5sE>Oji5ee61tz58X-~sJ8;YXvq!)f86Xu2A?o6d|pL%1GB5l^z9C;SpA z!c%{0fHIpd{f1woqws$ASKJbl#!U$e`1PU{>~Liie=z9751-$n&3lI9>I``l|0noR zq1uQO&gl(}ia32^G3FS?q4Cmbyg#@G9d;6&^U#LEOiQdjtB;LeM9;cK2445;DfjOk z_}cd2m)Co*w8xhg@$bx-N%yEuwN03 zIm1(N>f&q!zmuY|EgX{q&f;mO!`L~!4y#Kuaqh`bG)1@6GX{8P-SXnRbU{w0Z#5 zXN{)g@hn;R-k!Q=s*!zqhh+1BaWr6gg=EhOWyz=6N9pqt;Rc)54^|d0=+~hSbS5f+ zR+v1bmDdfCe#`@ZPgvoSj0GkO_dxOJe>8W3HVnMX@Skx2YUUHqH|fsLdrd(~q1da* zN5OM_2#i!#;H#-Io}8UcR~yz*Kym^-*&9Ipn&q)>NGvNz zKi+|bN-fy%TI?4D%kpkx97cK{flsGwoH%+HQ#xNk`=DlMdps9CoE#j_uY_{GD>Qs< z;NP^Ho?G;Wk=XB?93mJPnLp{jA0;Hw8Ox8aSa9D#mdt7LK%_ zwU*N*FMbwDBwembCipIr7}brU7Edo)vqg(Crk#_heSIidoERbTTvtp=IZgC&Sw1}v zEuz(?^%PsSfwt9(dFhS5eCO)`PA}@o)2`fv`lVoKG}_R+lX3K~-k5x%Eoo|w2Bma+ zC`tIe19$UIW8_}(o$znPy~+15vuPTh_R=qr)rTv|G z{X#oFR{a@k#D35sR}+pJBk@6;+nYDq;hy=yP88gj zQG9G$7~k7)0gpJtWUW&6o%I$85vSNe; zgkkEfG~9lX1_hZFxc6UAIGq1Sr#_BCmsdNWGd%?F!w#cxg%8?tPEdlSDe4+W;P3L$ zC^wjh(jq6^&U`35(F1VtLSHESY^A^F{Yll_l=_{J;JbG)@_rNvUa)wUEmXln-3IEM zmQP*&jzqI02qPR$B4+&&n1(vyN2g-gT+-$tM}?Pi!6cSZP!mq07)%nmy!WGRcu4j< zp8jtmyNqAQQLBdXiii#Pa6bvE-wxob&psq=55o^_H#F=#FAW^6ZuKOwx7CsBidLGw z7o_p|tE9U_Rk3NV6KvOPMCQ!}a1+0$XSeHUu-_JVXg1+)gy5yk8_VZM4(6m8+1OCM zfkT$n@YAUe1^eL|k6E^yy@Iv5Tk0c{pFR@kbnBK6)i5=~};{kc|=GtTtlnGzkIC7XwdR-+J^x);69ofVn8 zLfD>3LXOB^CsqGLPmh6IaoL`Se;CD6CywOUkdHX;vzM}~EKt*Th1i+Jpxae{`1W%` zN_JN$C59k-hCgOhD`4h}VPx{zoz(9yN8+=5Y-tcV;tQi#-_ei_R>xrNq5$0faTQAK zEzmCefS5mzu->cy_p6VfQ=J;$&l=9XuZQtt-%xg6W6Z`oWyCvrBb$%&;IHg`+MKxUnI0S0K7p3T zKFm1Q9mmHnfLA~`o?SnNYhn&$S}_;9 zqU;31UpB(IG#SS7|0rDWW-?5dAi{btx~#m0eow3M@!49~9tgq!?FRhRdn|S^id?_? z1DciBqUxh8JDibnqyB1^=B4u72O0d;#GO-e4<8K7- zSKkgp*GFJc>;?Qec@pXxM{%p)GrIkFi0Idt!`}Tp&GULnafOP=+0p}Aqn2TdNg>LH zeZV>Ae5^CE#}u>PkbIngVS4dcmV62!cb?#x(;kdmvxxsyUgR*3Dt;}O#JjG}<;+$a zR#DXEe_?UF^Gp#BPRimVo5%9!qOR<2>4;Ou1lM=?T_kVz`iJvQ@hv)<-TF$JX82* z%|;|9K5Z|>3QoDJW~QXeOMP6849EEYuE4tG z4hqc*MF!CYF+oGn6mkG@yH6qGcwcr7I>P-<`}3cFanL;b0jqa)V%tLsJkh@mN=AZ@ z9M(*alw1l&=ub7p_ z;q{sv>@b~=`x7V9BabRtD|%60XG3rz^fIpJ3Px`acc{GRPut8dlle;(M6Q$-Y}Q9~ zreZHeYjnl^RZlTYs}~P`Y|9sR%;XHQ%PG7s+%duj6I&h)f1Pbm7d(U7;&7^W&4IUi z1@5MQh25Xm`1xR@_#T+^#y%c6pK}~`SI$FIVKW*%WuVK+)V|XkwG{zS{+5hapHuMe z$9kMUuZUH9X2ZDA3*o)|(EFD7zb~3jwK7tP^VvKaH>;Sk_X%aNhK5n5NXF`-8r&S;o$rt>hC zKTwaqPhF9qb{i?hc~CnUhhBxQ(Dt@Qi^3&j{m^8W>H+Muy$w#!LWE;30#lX!xp+ez zn~uEAmdm4gsQPn!UX+Kc9pag5@ELCps&S8A)A{|N0It#)%S+FHCC`f^(DZW!(l(#M z)O7{8^f??Ioy0sKYzV%z4aUUjPVg3auahMQ>1KZ;-oAJXNBoOm=jMFAa`rHX1 zGm+z;a0J@DgbOWb9rP>ok@ivnk2BNJPxA?UQ>uhFqXvEEy+;3CMF@NrC4I5_r?jwG z*=oH7S^bT+x5~Frv%1++u+2llux0)ooYee=9#k!My}oFl_d;;7`*N<65f>Z$#b*CK z$a@opD-2KX{!cLe$o)ZcGgpPocR0ru;L+Y z8#V@~$sW!9HsVwB5pgd9{;oSm^L=ZmpUCFfK4_N4kdIL2hx}Ba*_qrY2;G)j8Y5#qZL&}6wpoZF<)<& zv@bHF+9~g7Tg+@oqHSQ9;*ZX6lyFB*jU{thj452#2`~G~@wuOt+@!di1p}>(y=sNx+=PaUE*f*+&*&R4&#pGw)MF3 zE)^wv?Xf}Il7^NJ2Y=7T)ds;Y4*ZU+d4)*rFh!-7Ighpo<&?EyyrlDFUa@sQn&jqF zQ=zx$i{6FTCN=JSM))F)I`ioBJ+MOOCS0P2^1Om^e9n9#Pu3Kk@T@3W5`3J7L{(7q z%S^HhD59Sa-Dq=hkz~C%*L@j%LNMRUk*a`p_R|nrBEO!Gs{0>V7j2wYDlO&Y4m!tja zP1wbXJXSvlRBNM;u8d)VFV9* z;=(D~q3k)vo0V(6A##lZ`^;lr^`ALMYkbC>zs6{`_oJU_tI=3_3pZx8;$4UaTQq;g zm~uUQ?zLRZ49byvxdQ&~B45?t1=YG^Nk>(cGiHut9bz7wsK(!BUPOnZA@1z@PrNI} zbM9o3O?O$!_R!~$6X#L#t`eiRsj_>1PyRUKGxQ50V6lHAtrz*-rq4IvyXOT?sC*Oo z(Bs%A7_;fwt2yIv3U~2LnwX)ThSk+kkZ)C@(+aA1EQS>8(SLQ;_Ff>J& zV0QHgEZQnOr*FksOi_(5SFGSiwjOLG=3V?{H#FDAQQL8IzMH;=^`{52N0B=p&b*AR zi$;^(i`&#R*$qMEOJQ@{9-0#@apOf8ncgnM-P4`9#>$lacbo8p3qR22@`(C;{Y$F~ z$70OgVK{cKo?4ytY1x`=%k-uOxSdns6VCnl-&9#1H*F?Hjn^f=f%4eomV+0wl-Myz zJ^MA35q4LyC8v@S&)%k!;^&z3{7oH&w<)*bxMZALKUDkdM`ik9JTkkC zJ*pz7u<45ErEf=&*e7`3Xu>Gr^OhVJtd*HZB(j|@A?LgrueHB|6Yy$t0b=i6g_Eb?jGbziH1A2ksos6qZNnPgJ$D&fZxo;1-ZSW{JK0ku9!8XtsxtC@Qn?^a}yX}5mL0a3&QZ+AQY3RlS-10U6 z{dg(>(Bbh;XHJ%2K!DiMB)L_6O&P666>DI(jlzlQH&`u@96b| z;mAywhBb@Tuzsh=BOTi)X>t0AQuD1m{b>OA+#1DAH$(ZyHNny^2}5Mhr|>Yl0h9Vb zRBsi1&*>i}r>?ocBrz6?*QHaTXN*KaRfp8S45lpWUQ*XHQPL8fbg5LeP`d3vmh`VY zNhdXkoM!k%9F#mjeZLm09G8Z_y@%1^{x`58Pno?$PJiyz1lTB_q)vy_sKjA6^!m4A zf!z;$U-u6UEno2}|02cbbYN$ne*D04FsI%8jg9U{piovpM|%2F)Zz1V?C3vw`La7c z>kbh-!Z!M`X+DLxhSJM-^GRpZeKOn78;{0(rVf$S(l{cIoh7Q!>0ybAz#%B#oW>Oirgmyurn=zbjzwK@3WXAzmY}1!4Xuy z$AN4zOUc%BxX7ON#+N%*NE`Ex2F*W12Oqi6y_Id0H}@AE{H%frg(}e2OedKIa){j; zj}@|apnBjUhA9dz%iR9><1NmV!EJc18^kB(9_NK=@jO0!5ZmTGp-=ur_|knK%eeL6 zFABZ+rh7Unqk^P&UpGr_D;`Tf47)B3Doc{S3ztg$&7A4h4J`~^vjlnME4cK)?Bk`R zyfTx9m$g&9w-J7ZeWpW!JrTa(Jo$fYfyG-J_P*xKO%{{5W`!Jk7GK2T1FBRvr6fA*Ip`Qor(<`W>7^_ALh`IU!v!ZK@$Z%W~u^9Yv4QvSA-F52#FSu+k*svg3wP98P@tP$9YAwaG^~WJ44NpF-s-ZIvn`gpq+xl>GkI%UOG75{u9>2t(6X*4pVO8Vz zSnYZl#^rl3;cN*ypYOu6zRU79!Cv_}BL(Z`PsgQ~g>+~sqr7nzF4xaSj9~jVZ+St- z9GFKn?dH}Q?)>S5j{uh(MoYP5>m%;i!*IRH z6m!CE)27^x_KEk@A6L`%7N8bKx8mm|J;TX}){`|QU zrN%sLIuxhpezMObW+x&cJQ-K+lRd2%R zqbt#SnjKUfRPlN9T=X*N%9lTiIZSvTPV8NW+HFG-5FABI|HsjJhvodfZ@gWSlr}<( zcB$t&@25~^D14|$ibO_|T_Lj~BMnlhw2OvHLn13Gvmzsuq9KwMBI$R3fB*Tzadgo0 zyx;eIUFZ3FX+5I}FT?RX+Me0kMKR5hW^7xRB}Em4QGV?mdd@qvPP1Z3%FLEB<-$q+ z%^A{aE2YL=JgX#m zBF7WTgAb8mtr8gwbWmX0HZ;UFKuz9^Rr~ygJZC*tEso$^Bkm3kZAH`a63jQ=0g0YK zL0#)CnqPg5oYrz~chyG3j_oDmmxkm#B^Ks`f8nvtA9O|BgWAF!&2^sYi? zdNGKN)pKM%>)xTgCdIPW@e18u8jf`aq1Zn-6&{j23no}E7~3n(pZ|-g@vSy}QvV^i z@#VXqcibs5e4`HYwteWADa4{D=@{{AGTx2ULfL#Z7Mweb8GH0$gFyz?XivbcQ~%+# z(KOJ|jeIV41J~wtA-z8VWBL!EW4<1H=64G#z0#rAV}Ju+jX1MvKHXAqz&RCvn5;U6 z&_f~U{FzIBLMgU-b|N#f+{W_3dC_r_Y-&dhhA)a2gxEHd!~=0O-MvS)OV5zUT@wma zmZStxPZHr=^|~9S)NYYZ3)lP7pL`>j9*O3T-*jAhU4py0A)H}24qDS6kztk+EBbkw znKiqy4NEK-`17oI%Q?EpT}Hc4YNPFvB?29+A+x=d?maj|=Z=rZ?aD%=f0SV_(_bKW zQ9PoTNus369HLvz@V%J#8M7v0?By@?Me789?s|+JP0!HqDj9Zzwn2573k=Mjuuxx9 zq1=U$!jL7OS#4+{+xFL=ZJJv^n)L~gvRJ{YD&v`m%t@9z%#``wo*@{m;z~Q0%Rqtm z`W{a_2(uDZY#uPCxzp#6+ulI3=r~AKoOe9iXC&3~{Kca2I_&fLJ4|D53wv%+%@*!n z!YsUJVa)6ROmnb@!fR_xzbcR2E3T45?KaHKkK`QKWALBJ8H%|%a7$W^{NlC39F0n0 zgJzd-=Hnir$IS0Sr;91Vkc&q2{?u1GTsQ`Ps}0eSG8v~jyJ$)jLrUjve9Mo<3zhA> zvo#e;3TkMSPM`q!6VP>-#p)*7uxQ6_BzPCV#nu5L{hXg>V#{{VoX!4Kk7dRkCjst8 zJmGGL`>*Gtq&yWxZaFA!G>1h>9d-Id^EcTY%E=gyPfJ44S0RawPJU!5xrFDai+IjD zA6qSspntFmrrYt~f7@_8=Ut-^&P9IlJ`~ZX63E{*lRkei!u|$Xd~o|hO=aB_`fm~~ z*y>0P13Glfj!Y9)4Q zIwQho3a%MGCy)3+*qa^)t>;yEoKgs_iCJ)bC(qg5o0yz>Av-hw3_JKIm_2kcV}sTV z#aX^zU7I|HmH0cb?l&4te@rb_8w=1e>loGy^y7QJ9NU^!3oDlz7@QD5q3_ zO-k5!v5~SeAJNC(fmHa?1tSJsfOKji^t+cL@aj~k+1b$nA7>hUp`WZ%mScub6kM_n zBZPDRWCnUMbG19`mz&I5thht`qL3NXoIp_4Wj3f;T6llRV4?n^I<|dB8tdvZWd{FR z1Qk6?AmP!DixdAMXT(WltC_?2Krr7`ShMG61?)6;j1Q>a$K3^zcz);ud?K458~YYY zk-bQ@RjqJ_~-k z#^K4P&$K#nHwZ^mu;#=hL_}y{G0%P6+9=63y;Ns1MrKToR9Jb}8P0cn zfR`s0u-0VW)8BBIscbi6TOAH?mfr$4O=&ClvM*#$9}&Abu^Ynmh4lT^5Y)!j;q!74 zmbah>1)0~Nx?vD(rSq6~>s@B#d7dq6*~)gds3##qft2%m(C_*_{(72 z(kq~w`CDkqDjhTgZbD1-L7YA_1^tdE$+@JIn$3=2+M6tV4A~5=JWt&Cl}m=XeU`9@ zgs7i7yDLy&t%0}kuX7?Kj#x8ezg5g#;L3{1ESbGTE$ScE)0k0j1Py@+WO-g07P(5C zQC&lAl}Xe#lXKjy8FZ2lLws5gc8|`+@~uLs4BVyi=yd4*;Q4l*#S2jnhI8F>swh4$ z@aBBQFRvFOZQNjZ&TOQW6PA+MfkCienU61Y#R(wDu**nmeSc^TWLHHmc z36lxxc)x7}>BP!y`eLfb{zp@UD+QojyyBW$Q|+``O* z$1~e)8W4-oCma*tf?hIKq@TtL^Nkp-^q7`*Zsrb3d$xSdV)pf-Hp}H_w9c+k7`0P{ z!qxbGDcKLZj5!}t7!1<6L0f`!*?+(N*gQWEwwv>}$9gNXshl^q-XVu>giGSeVa^|) zwU)iCO=kP7{n*qkt4K{s2g@eu!()Upz6l!1xZ{FgG3OgT^mW9E3&-(Zk>@|`RM2^Q z3`O2^;%>qbl+*uL@J#+F-JkW2mNm|!`;TUl%t>_=b5_P$?)A*PAM*aojX%-?Pz8?t;pcBr2~!Gc&MPpZJJA$6G7+C#e!eH7$a^^?yvADm8Xpf;$rGrS34->oMQHyooiYwTBp>dJK;8;;DcIoc^l_N37eP5k4#E7? zN4QOC!^vr_*b-NStOYZ9=c0jP4Y&t;`xz`x3BW8@b2O!2Cy7I==yY2GnkS31m?gv5 zxN$j<-7*~Ep|&i1^b$7P#)dV#oy1Zupip{!f0L@e;52y5{~Q?5sp3fn5|zQ z!hAB%LqYQs`M(K+sr-7je`q>$QVwT9C)Air&pAP(M>-b&oWTN@FJ`%sn#^cQG&D+l z1v3`#KH%D=Xld4k+Ll(%X%f(tt;dn*Ys#dul9)?PE*s&ri9KE;&YpJ{z%o#TE&JMr zJw?&DDz_S*^%7(wzZO+6!e^%lj1qL?Y28h{9X<{3v*!zM+Y|_gbiWd|jqMa}{PJAb zdeuePw{tXH-@SuX^)IZv%J=@m?jwiy@7-J5;B<2|TN)?M;$*KwoYv!#@ovGKMO>w;6?FFA?Z`Y))r^%#vElSu7b#*%kU9PPU$MKSmPk|XzG zcq%v|>=So?-+50OdykT>=Tg**?cw=rSIl|O_YD^{k(yONmo5I#l<@_Wlxjpn;(h4% z`vX*5p@a-6EBNv`MVki00s;3CH0{Ch|4M1}f2P=W`ZN~gJ;1Y~Q>YQ2f{hkOaq;RA z7|!v-D{D>c@2wZO9$N;_8Jaky5`>G9{4BEeGS<7tLe8F_;Sa7u&BRVP%OxW|U^`}h zoq}S6VNiTTLh-O8!U<3Pgy}wp!X=M1gdcAUg*(5E$DyawvHI3VSUp$))o?A$P_f3u zv4H-ha0KY{9&z<)2yZ3vj>~aa-_&3=^JlSh603RNV=Ox~_XiX|sk5!wo3Yg7BCgFo zk1?z6U`tE^T04!fg75FwIW(?qi!NuQQ*#dsa~L zI&nI}^DWDkD56~55?{9QtbNiNWUPEee+Hsyjg%!;du~Ir&lv1oEsvbEJeqOa2P-o6 z^0R0ysm%EaJF?QP_&IUDy(v41v z$WFv6@9XGte1XfqILj<05gM}mIqFl4I+;KuCh+~tp@aCNV8EsgY=_Dr?$dOLMP%R! z(BV0l=%xhO#{0B-k|S!ED;`YQhU?m^(0XSaE|kQfwstwY`Fte)A|C`Es2l4w| zS{D}D&SEz4`ph#yp52fh!m_1|ne0+?Y&h(SpbtUNaXby>fl_SQwhR&C>rm`a0E+d; zEoBAdFMdu%+{H1$)`xGLsIiV*j7jm4Bo3KL`F6S=b;sQT>3aCVMM7PT+Ai+a%m zoL&46e=g6&NQu3C&t}N_PhP~pvJhOi3q_63KKgvY5n|`sP_SVHmMu=f^byzbGv^U^ zedIuL=QIpbk;daC%Gmvm^LUb`;Aqqh8WH%ClspvBb*2QX#+`-zswf;=6@oFB^xF^7Ypuf+ErCE+le)=2L%&G}ubn~Ww8hu9y^bJ?_s+P3@SpG+Fwopgl# zt3XIx9Ljypn^E>$0oo;^h|?{mH>8GTb*Cw#Y6`NpC_~rv41O(8gNy$tWc*!);)oaY z`06$+EIEbjLHzwZl7CkQ)57+vayOt@(nv$x4ZYW;G&($U3`XjMo&enr9q zjxh6@i}@4Z(V=~b^zh_bDonSFvBUy1l$QK%eWi;^jCX<(5xW^y))N!=h+ zH8he@@n`zv$v@)<&MVd{C8M{M8=)ya&xe!Sb)9W zL$Sl8mFCN-;+awgJ?!&GzV1YL17@r$tUI@jelwf$b=HR2=U6}N!;rPNJEYeRw zV$~dE7xAlVlqcGA_tTxz!x68_Ij}FRVXZm`^RvNxQ@rBbhl4~|Xb4)V4(M5!^)&Tlv%D8*At-q++X z5$I03L1S}M1!oV5^3Q2DoWEw^x8xO^5J`vDrCIRT--C}1iLiYt#R8@7!qRU&MCI-9 zkmoCv7n`sJ!U0%S-oxCsaEv%>g!?Tk;j_pT!&(mD`+^MUO_So>+FLZKCjcjP2eIi_ z#h9Um2s>Ijg2lesfuEjRQCHuEh}l&rcgTWl*-1p&?uFw@e{NI@MTm(7^g=apr6LH{ z41JH)?AGIy^u4EV37WiP*yWb**YwZ5c)j9uhEVk3PudRC~ z7VTpn+kAyBnX813%bD=N4<=M9vK5vMnSuPZaR_i;3xzm_)J&jl74MH%Mq})Z`}ngw z0~008p`X42o7>7^$NQ@J%bGFuST~M-IS&_}qrAUmF}|$5k4{z{HMv52;O4YjBQY=SwZ%HFO$ZY&s6=H(6TTFJI5}<;2me_b%Z}u|2AN->;t5( ztHiLU{El?6N#HRqNw7HP3CR=>VlJPA%trDtWMXCt5@JdPTbjjDI<^!GxEFq7bqv;D znuy!eYT!2NJpwXXFn$X6xK5owyAQl4pO#YGmX&0~nzkdscLxrfnTFv@pOA%)6>cQn zfN5YiWOi{E)EYzlJ*5T9yF=N$TsQV$yD3xj`+)QJ2(g!K5Tx^g^SDp5k6-15YfGes zkH_S&J7;Y1^(4OojAHCb!YVfP?*w+MJP|Su(ZZIc@>YXn)vO|t#H_;XYlWhJ=Lt6# zFQ(}{`?X#IU>{enVQ#?@m}KBU^SyfMl)*;K=DxRI+PCS$+kKE-^c`+GgIQsQ zI{TaX0>2*yvNe$R=~9|Fwk`O?kyU{_yW|{5r%;AI4_QvS4~MwV3C3Np@|T zBU&b`5_k)?3&i~+A+>cTtIJ)&mhm(2x^+_!<^KnluV}CbI(2w;oS)$eZcydW22vU) z%7i|tOh5QIpNWiS7b^oG{?m_Kqt#G6ISd=CuA%H$43vX5z<+i)n^IV}e)DMVr%9(zJ5?Y$Q;NMVh-U5lxg{KErHzic``Jshn?Gd-1PNpuAT^1jf>ka3tIGYStHngsEC zyU5Ts6JleZBcUS$+iGTFR{dxQhjFH%Sw5shKOoG(i;3;mV$B|U_~2WG=?7n;Dy$#= zPsXy_`=6kF@J0$-6HGS8Y#{gSGJ1pg+#xdt@ohwn6YH_$ogHT+oM3DERx#1*a_rz) zUp%&&!A_MwWKHYNvK?xlaJ9>oOq8N5Z&{U)xor{bN~GDJh8H-awivIfGzCo}^7Obb zo6>w9(~HVGbUAE0JvjQ1q=$#nl)DPde`y92XkKCl{JynQR2_$psk6)!D9qd7jL-^gx@?$D zg$khjg6n0%+7IN(xtXIY&w0FGC#>vpuHGkp6FuS zcHnMbA3fNzk)GbrqA!XQ@#V@M2&+$_OX&o}UP(c0Pa4X544Bw?J0_pki*+YkVDe2L z4M7Q{myv)Ek$J4f;5!o+JH#xdpCTgG7}Fx&A>C~%voBC)CZ(4kdOidbD&nZ@=W5&z z`+)`GOWBVZOPF+eImGQ(q3Y^4)Zhu4FD9X1+7a^mJt)uXI%d=h*pf)FIlI-Fjwa7# z>c`WxaSz$8-=V@&)mwx`(}xN}SDj^>mh-IZ{avt0_(`?y0(PwJ1sjy~m*vcjVq5a_ z1qa(xX-icK_ty5nsh}74_S<0ZD0gz_S>+@8dH5!ij;TMCG4rSyEy~)6vz2FIdG0g| zduSpo-l!_9f11En*Ksat;aE0a=@QZg4nfx^1-bQu@wfCDfqR(qy>_Dc_W+7J7O=zm zV(i_Jox(}m`h;f3)UAe=Xjo-)I{*H^9l}`OY)Wiigtf1fSgb&dcN{Y?CgCs&4-I4S z`%kgHm;2d6V{N99{Tcz1F=TdaD2_Y}#{FJm{m0L+^TGSr9e!?T;0%~C_lSKNeTj`% zxx$nu1+l#oX7k*r82&5&LOI%oSodH*Qc6?sY^M%fyVIymY#*6CFCwoRE5x4D;Xd~s z8tCQj6O(wzoV|v{YUi*&Jq!;XxHNALRupE^RGCCdvE;mk zRj+8dW-C>0xJV7!j|4xJQYpk*i-JFOQRbjkNPQd()29Xq{qmX;KAoYyy?p1JXavo* z<2Zlp5}h1BjNeJ*5jN}-y*&MqdQP3BS(cv!8~83T%~^pn3^g%q;TJ#Xs8w2%=<%ph`fIHu#Fn9km3Li2LKC`Q#aC8d$ zt8$#ZH`&Zy3ME--NG-j5I1B6Aitu*HOC&FOf@PfV`QABJ==Qop*!uUmP@%&Yb#uX*1o%-}y%sq$|ALs!@we?=co@29pa&uD4uLLB~g0u3U5SSK!uMF;Jf z&xcr+kivb#-}&rw-3WF>a}vgi-lni+yyKmci)Zr^k#r^yZ|y`8bL$X=sOcgsKN}zO zOA+`t593vA@HwQIhHqGb=}P-BQq=|7Q}rOsil=qQOek)64oMyzfvaIEkRJ1wzFYf| zoqYk7EbJnQJK@M5eiNAv*Z3J^5o%x9;&)vZ#Ch)`l)DR5bxhczIW3sptteD*@fLpC zvR#<1<0(|io-9lpbc=;)=hAG~aNJFnW}g<1Vqd)!nar(I+|4Zqr&*pT^*VvvetR65 zVvMCBX|(h3Jy>lNVU;T-+3zE&Of{$(;V)L>`~AU4h@6a>dg&ND{W`WzTQnAlN+&9cSB6eomdO>3PoCvfI!;=mt{x9nRfRoE=d4fs1-^2u_m3 zn=|`4b6*L*-UiT5w1Lx0C20LjCJ$#{8gbelDyfB-5`P6&Ym?Apu7xeDIpg6#9sdkH z!ivd1(7E|3H18~gcg!7JO*UnkW5;m!me; z)XYOoTL=bhgixPp0Q)8zi1Vzx#GzvB*_Z|Oo;_f7SrnROP1??iXqc&o^r>T!E%uLG z;^vd6;~A2V5=TSY6!-oHgt4@Y+5^PCJR$TMk0eE`a_7FB2FvTI2;&TOVAu_JNkLk+giP1-7Rh$z7?P0`r>a zcKA?ai*KNPr_=yOX?Ns(D^P3`9Iukx1|SrB@Nk$bu-!MkfE$%Dr#@53a`H6la)_t05utP~@f4j7(xsVO&B;?Mg>E=AI2=BRb39AD z;nF|y@Bd7nJ4KP<7X#gC-Q3&0ge|VoW9K>?P-7p3Siv5BP=1c% zdffN6VILL?)oJd_7;>+g55wQ5@Pczo20bu?^l5eaZ`J{F;@PF_CGmuM7h1Y-C_VX} zf}XDtn6+#edoaS0IlpseA4Qrl(ZU8(IotUBMkS`O;WMb}HcFPN;wnE!$j|e~@7Qpx zo-~;KuX{dmM=c&_71CVAwt7>_%ODv0?Z2QgYq z#g=Ng`|dT$bw$`Wjk{PQosP(h|8X8q0{#B?j%I}1qwXI+Y4+f9s#sSeVD@9sGiL{! zhHu7`a)Q2s0OczW(%!h~&=>W>kDv&|_npD@d4K85o@_zQE_JxPJdYQK-;g!;5pqrh zV@~OG^qFj>)s> z|8qJiVuK&eZaDjQK9si_Ay{h^r6#4};;nmF!95Dqf8AM~(2BVQ_aR1lJbF`-u`~P# zLPgeKVX-cXq$8>3q%GoqZNT5xUPwsf@3NZ#P~`L8$c3T8yQwdPldg(dnT{T8weGZ% zmDA~TVZ1zdYZm>-^Buu(a@q>>YmOKtQbH3A-@s#zJnOzJ&N7yEAS^2cfBC&7y~`P< z9tTjfu#z)gALH|kSSa{&hVs81FkbT%H+Tp5kTLgWiD`h6^O}pqdhSE@Z^bCem8?ljQF&fXo^zymRwHVUs9S**UW7 z-9b0|E|S)&Y--yOO4%-<^xu?al)Acx4lk2}?0jwRpnpLFPa~+ib^*%=#jrt`mE?L_1)v&Ok!4@ffEks^1Vr$^j5o%`B_jB`g* zO|mwzQWKPId z!m*>Rp9~d(g}IOK3AYTrBs^IVCCpOXA*>!bTxeq@hN_Tb7&nV^bj$K^^_MqdHbs)^ zK_{@J0<3=5gp!~l97){(NA{WI_YGnyk!x9Hxeu#5uf~Q4@?80w7(5QXLq3wbM?mz4gxgwxf z#e%U@Kakx)zE61PfZB8wY^?-39QCK98QBI5<>ngK(K0Myn0tzUWK@Z{%GT!y3fNRKmXC0>;G7#E=iq z=)x@nB!`D$_}oZ@RJkDS^%XifDGCy?RXoFS9<8>UagTceYv===`mhGm*YUH>y*C)c z=XLAFUf`GYRII)|K#x3x_@FQa%br-HqE{21`}uy_`~*Z6f52oTRhG@2igM;MELSv$ z?}rP}Q7F#LeFkvLs}>8Ak|E4+fzIZ?WNyfJscy${|CBe>Y;|EQ)ld1=7D)PmtxeM0#XW5MtuHAoVl?CC8iEIKOhQ2$r&i;-<43sn;4vb z6tDevL2cQ6l2C7o_);@d|by!_^hKCg+`{H&71 z-i2UC$8${k+JTfwmH1$G82Vx+cwpF!EDKducNJ{o`(IOEYm_k0v^F^12YO-T}13KTjZNX+7yu>~vk~UA;%(t*B04jeTj*(NempevcGil~8|bF2%We z)0`4{lnoH5R$46ns z{*!o6x(BmOmqYx`JCd@gq~Eh7&|0jDt(v2-sA(xm&PXAJU8VIW`91sYYHVuZ`RDx4 zv?4f(w$F>jwVC%|Ij0|AzV6Vi^jK_aF)11O{vZ(8hHzM zK8Z88TheTZZUZu!V_^7_d+Xkfg*EA5l>9#$b4eP%H`Y;zodr&e-49>p$?pVmkoa3q z0peru+3hUby}ON_eN@Ii>xQ$8KZ}^0r!w2*R)J4a|8V!H2A_S~vEN0rSZK{F#L2#( zBL1A-#-dPN5R3Bm1-P8A2h%6Mv{^NQRz5sLJ9=YD-{Xa#^{Nt0zw;9wYbUZ^%_VHl zn!_wtxRNQUOS4whFsxd44^p$Durg%{G`J7#MQ3j_RwXBYL-p~d4-_i#Vw3@t*Uj}Zzx#c=&(1ZP3M z!MJl-@GRVe2=$5Z&^E)a8Qk0W^&oPh;}GRsir#<|aBw3`_;ep!j>EP#%CQdBPnn+wR|*eerlwEpO@(M1PS=ORkB7V-2ZtC!F1E>r69lVpFZndW3PP}zYp%8!u4%Dz(CzBC>^ zPFl!c;)~hwqu5V3F*dw)C!$Y&r7Jj3-@ho(QF$8*_Ww@z5~3*NRSn&3o{Mrx0qiyr zrWv(DF6=fkj}*~j!x{=w-$N1l-0!HALrsP==*VRk>iCgJ2k$4-`_oyJobZ}H`DBr) zUZ9|vJ09NJ@po2P3MF`t!5-Z_c;9owu_R-3ynQFg59ou}l+mnYcosZD)0ui*2RlCG zGP`b?!H!HXWF?6+Sfo4eU*ECA+ng2>eQH69ioa-)`(L_pDT$JvJ*U*8j2xU4k=XBu z-DxU_9exzcdlC?NFAUCn_B33^7u{+9VT5P^eOY^xu2B|c_$ktpKZay|KAMt*pUH#g zP`2hRN5zhRP#av13BNS3ds86EJ&?pO8zmTD@}u^-yLpCbKHgrvO}#IhXzatQ^e*u@ zRrn>*o9ZSyeI<+1S`T9L!QBv^8Ujbf06a>`gi#FNcTY&8tW@qfmX1X`i@<*AQLycE zC%dj=bTcP_X!KUHyp&4SQCmnSZ!Wp<8EbiCDD5-2EwD+~hRm~jlu18myI(N$GB{^H ztqM}7TX3#u6~^6~i=WOuP+74SKG7HOzUwB=+~R+Juml(RBK?B31L^`BG0V3A!9#^u zzQX|HE=h5|XC%!YV~8JS$FQqg6aS6dg#~E~5wNKqeMOU5#94XvSS1R1dk#{?2}%A= z=Kl_@bI?0}0D(VZ==zm>vUe&a*W`AA$lNJ3=;=L@9v?>YjTX})kpuL@KSvPIY)%dC z&*=P~7#bIsBhcgbZ_#NMG^kKokT8kihPMTNACbbP@*Ya-T8fds&+;5#41TYhicy~H z1hX?@sB^tN?F)(*6mL`Ftf-+7XbmIN$R=9Xw3MQK_i`?S5>}f%qb@@!vX%-Vhhy&r zI+e!c;x38tXDZ35$clDfJ4qu4U80vB%W3nnaoylXj-j-0J?(H*@9%FAxRCddN#*-2}wd`Tfk2~B5#ePy|{n12`cq)wr! zif8C=Poco8@FZCkD`D!&mx6%f1$6b+9isjEIBO-C%PipP=bv-Y0AQSR9S#{Hr8?G zW&brdo}CNFWzJc9?LL~F%({i;swQ+}aYBUbv1#75h@@X_Qq{31| zA5NPN(Bu~un4J;=@zEC8uJs3t+oag2Pj3bP^@R&mQ>-kT^pB8c@Mg}j`69?l87x>S zTW>l2=mfzntzbcEWSn6CA$P$M^BpwZ)RevzoD-z~Cq;k6htvGo{enB1WrDv&Ysll% zWr6i_b&?J4rJQ@`s6s!H7B3LT1+S4%ku*i2JI`eoy@KIXK(8kKu4gu=XM)cvHMw1Q%=dyY45EpjB4 z_2;puEC>*Ds)?^YPMR0x47FZ2v`P7s67Q(T9Zn`i zpNj&EigdEuB+E0QtrYfPF}VwOqH5x8>JQC>)|>$xTsRJw^i9c?f1Y+VhYHG$Yg08N)i~K>2$kov=uz>mE87`jxwV zZ|Y*ysBzf6eFtRJ4wHAEHMyq-(5ic!<-yOtt|!k?#t809wKbw1ewI-GWuFgjwr43XJ-%m#dBq*2r zkDi~%1o_R!(Eo}k?$SPTh@J|^eJj!Jsg8;D|KV^cpCQx-U~GLN-t+t6v;2JAzb?uA zpQhsj&*!RJ>Vlj(W3oO4TQ%-MeA8vTemTgn08ODrhjm zn6(b~Yiu>dAS!35@(2@-}>jb*wCre%vxWCN&Af|H$ zv{qInrT20ly+i^nYWyX5oO6R31_h&SXDY0kHMw(58OyJ@VCSMZ+%Xx+RCGUK#-DD% z3}ay4g<8ZFnc%wfOsd`!OG`53sG#j3-THou_e-L&VZR!t^3TC1_L7`U-jQ)y0Q@pa zaX#oDvOnLz;L8TIvRxOBmqxO}TJEZM^~U|iO#1lwA{ukzpnUEx?34Dfdsam({>EWe z9k~EzN>LF1R)xtuBJ5e|B=+M|BS!g*77mg-FWf)(gD~oSrSJ`{6`nmYNZ4)ENZ$O+ z|19S>w=aI)a#Q|FDbaO0xM5T2?9&#WH8uo}^Jik#D_> zl^tYFcO{uqRV4ZTI!`_S=|kFdD9;Q@uz(NiFfUmFt&1hG;BO7Ye0*5fD}9zOn+>(# z-L!Pc6Y6-OgSNC2sC&Ew4}zr;-JC&-Uq{k0&T*mmKkEJp0X5*qPVOFy{Y1+j2r_N!|*I49>;#iW%(h^`T62MFDlnheI}@0wZ5t z#i(<+Xj`)qH!e4mwT3h8*XAy?Ihq;b9X1#|_qC@i;y#j;;??c*q{;d6zdS`YF5bsP+` zoX~zVo4YPG+4gnr?1i2`Yxx|))bblBdCRhTaSkm^gyY0rz>F#A1@`G=+B zksd<^XK*|@p^N17Bd8f8#kv>lXI;vEOwj)crfXNzzEFQk_~MS4haTgJ$v*7d{E}|F zzNFO;vjlH5j!|u06?!d3vBYtni1HDIw1^1%9x|9|c%MeqkO$N{-IsQLN}~I%%V^YQ zD|)gl9X@-{Bk%Ab>I#zN52t<_{v{2K&Lf%4mXR#C_%#wAE}^?#d4lq!v$Q!c6(3#0 zpzimI%1rp|=4dgc4g92Xv-g-7p~qqtyw$tUi`a4TK@BK?79m3K4uu}pZpl26%N>)|C_RTc51LlkwE#I z6lxb&B5!gM&TC&I+ug_TGrt1wbmn8vG6mG!Z>N|ud@glX0Pn*QIBR9kbe~RPs>ya( zekqf-EYG7rW{H%D4BQt+;?IB@{(Y~)JB_pK;PO-4FHi`Rb1h{1a1=C~pOIg1Gu$2} zvuqm9((CjQx3q#x3hvR;ss1#EMj+JW4h~D#!BW;7X%W^CiT1PB9&)QW}V^P))~q8p(%m&d^i5!9sfD=Ls`@WN47u37Y>}Uex83W{p}wS zrQQ&%e?A)9+h?)!*6Y}}hm8VT4@K@Eyo)zWlkoTYU2-dSAgN=LR6c7t{8E#6$B#Q( z`$}lEa}?dzTu9Yhf+^&V63@%7K+)_E^i92zE`(%}$vqzu=6hnjO)tW`-s6fOhj$LY z(RM)udGWi!1D>z`?7f-I=kAc1CEc`Qf*p5@l%XK^0yL*(!fyR8%<#BIX@qo=Zt?4J0A&fTai4A@JdIs=Qx=hWMGd zonio8txGuHc!XWLzmHA*u#M(kv4Mt|IV&kX##RPfvrS_|@a^vw} zI(24LY|q4<;;GqkGe#%p!p(gpo`|QBMM1ehWScSfO{>w4??=(#^%tUlcn&0npLeDd z(6I5rFw#(F-s_~<6Ye*6-C0hj=4A;w7j6)=1(#F*cL_|tu~V?seXT%7u&%KZi5N6)Y{d!2ENs+HM>ed_2NT|qX-^GGb_IU)|eV+*w=2`#jL zGE0g7-jMb}OKC_uDLvDl#Kp*`uL7G?XYph$tmVib_IK zNRlWK86hKk{I2is55W8OJl*$oo#%0UJ{C-EbqNkO=0hi8CTyPQV=(X8q&|#gM=s7~ zJN4&4V}}kt*9)=NwGYm08O!=dGc(QRy$gF zUgbi(AM1a6iycexVrHLZuxiO&EHE3wUg=!NR-OylshCX;BmJ?tG!qu)si+MJ!$ZDT zDOTKznlbIHSSS!W$tDR^#A<||#vQ_ffh^${-)0)nZKf^ftHGQ#SfG(4+x_2J6ea89 zxee!EM~AZLsuAo?ohNRXU8hxtGI8hU6+G?w1=Ck1Z2Cnnc0c?H;?>>oa{GF?i+Zw8 zSB|q=j$$mwKL%s{6kxy8oZs;C z5Oa^C__y;gx-BoFoWJ)!7~cjqd*Fk|cUtwDqvEAU;i0D^*0(rep}r|hB-}7saRGb; zvP^HgK7S6V<8*&Ay|xuY%)kCG>9Y@nTb*Q$4;FIG_urd?aF5p=_ z&WyO{VTd)piEy`+Wz}NEm|6OP=JFlksVPc0xqK#S{_TNg^J>_Q7x4RgIHeo?rDs{A zQNMBy&e?`z=Fe5Q5<3^V?{{HhTq=g?okdICLmV1h0#`3DEb`J|e}dPsPQySZSG9o! zI3M8sxY^7w@dECvDYGy471^j-QTA2akcAo6fjR9H_WpN6n4oe;INPvZ*w~RF^s>_t zUf4Pkr^kmN?zanm%X?$xlPJu3z6>5_ew@o#hJY6C6%MRGUF$1+J;43VdJ~ws&mPXW z<2lt&DJ*!R2~)Gaf^U*D;AfMF^Bw;1)Vv9;+D_O^Q^ko_bCF(@ji%x(Y>#>ZFQsmX zt`5SCX)#Djk3{zP<=Cns&U_rYQE)~Qy~D3ka=6+_J>h2Pg13PYHduww9J zp{f{!mc_f+dNuyva;^bqy1j|nFhE%ArvLoE3iek6or3!VHje~o-AyD;ptV#5b%9lM+Or2yhpojF=Ud~ zBSdZ}+pRT&)hB7OiRxDDiFyhvGxA_JdxkRYhn}pfg5Oj!;@I}7U-8?o5^2N6Vk$pR zz3;Zhq%?mFa}7sBIOk}c-UGX{Wf0@J5Df(?6m z=r3aSZ{nRkUG{L?8dg4e6N{e?Hvh+B)^GG0ZcTiD5d8>FD&Fj?*>%owIZS~z!zlJy zCSs=?M@w1}od2jZ+AqzvJ>dU#!9UnKO=ah+53&vXojD?;6#vcsLF-kx=Qp$&1G_(= z!Qu(VY43v~=fBNc&7F0H)3LO50wi76V*WmUx2cmww*7ic{xJ>P-^JqDE)%xFtQ-?t zHq(=PHmGZegorEOQ_YJ)dld5aEKbf#f>&M5Tm+lw1JqEO^|7vl|8Sp?^Vxg0-E=f6s_ z%Nh3UjnJ9}TMF5y$?P4d`usGr=jIX7svE}Hzm(D{p{==nB>bzlukKPSubhGt?|)F3xCG0$B^k8Fu3F$6nMTb zbix_xnm+*!65ON4?`3N0Z=l?rjUUV#w~q79zMeJ<3@&3=2g+H{pEfKImBi>eSA5#o zf=~bM!{Pm96zVv$UEK#*c!V+%y?e->?ZMHDk@(8}^Ge$f(Swl{G_1Uh3eZi%8+xcA zrjyt7Ptl2wBD^EsK$*@ApZ~cc?;f9(OBK+gd@(q^=g*aM5ot`r7MY zKd}qln{Hzpf8X3I^<+y-7P2VcE4awHUpqO^xv#hsT4y-Ja_UW#N#3LpDcZ2oP{8?s zKeTB?AkRE5gi#)!-;a@Cub1^BbL}Rk{2+nVe#k?t#0Oa16=kLc*-&mC%7QdqS^4Qa z0d%fs$V;k`LWL%{9(|DBCDi)8M!-}Y5JS5 zcwPAxWu+0AZ^WHTm;3N=vkPuET_MS{=Jf9AS^8Wz9y=Xw(H&v5Ab+d~HfDKaIPbux zC&=L8yM@S=RKP;0A^S%Rnmpy$(Mc`HC^f~>swEWutQQe04A}aIcc6Guk`37y&fcoY zvrTIY>6h+en*UrH2R#>{<8K=o@3bKmiOcvWBF(z?J;5$nCARL#GUoF{o?S8k#&6Gp z5r4-_FCs*f?^gRczRj1g?+s#R7*bC?Ao6nG)d;pCiV?4LHZ_ zTLsNc{zsx2H>pMZJ-y#4iJ~7dG-t>}R{n243s2m{K8Z!J*W4qruO*#zzDR_q?K((Q zGvu4_&;J_to?r5(Z_y_)k@+VY}@y*V*pS&zAI)NUmN;e1Y!+i~$JEZvWu=oUSlpO-%y3E@W>uAwshcuJY`aN~ zUH2%$btdAzD!@Bs4{cqONB_)AX-{eu^=%k})Hw&~`cr9C1nFT_)=~uLt7FSqNetSW zOCcw6$tUI|6^gdfs`F3iN@)-MI(L`yMAT{8)LaT!d6hIC9HxVj6KQ*x8%6z!rO~tE z$ind&DJ2!r1lv}cZLpP$W8=u6I-l6{MC$9xCO3XQSUPzgO^Djd43AnT=Srp ztE}lXXVR#6rP9Pr{dC{IhD0>NC_m^VecPQ#q01lB;XyhGI$@5sRo0kc>Ikn9=jo20 zA(p&Z2(w&EtZ*3yW}t@Z8=vU6wmXuF_VJ#39ekE{W5~2iSfJ|#@#8gAyUq^Qd3xN% zHwI8%kH$L*JXA5J(Eg7$ClA$<1zgI zL6;O7AJO?>Ev$S`h>0=5(?nA|vl}9-kzSH@j)g$)&Vd$qF$rVHJmd}n) z2j8VV_dQ7_x>q2ll1g#u2k6td6(s$bX!QdbTH#YlmLqydjCGP(ycP^@49ARNMO0SI z^U*8PFz#XmIs-HC>=pNw6xQJA?J+F%UNF1*bQcqyv4X9UHDi+&R^reHCHy|@1zGV# z7~I>3`DMH>zxg*^XuCmgXL5$x=8@1os|FXI*Wb@O*%kK(W8P5)Vc&Qpei7oCrZ@b@ z%Hn}~6&Y@?r7N2EX>^h-DryH|%+uYpHK2jgCWzvQ-!Xcv7eQ7pH&De*KHD?jMS1&& zBi+mm_ft7H#j=2Td32<-lZFZJ&?~2pR6Xq+HO}>*pVdC3 ztC~v@BcIWQld{md)$ba<0tnXuRd-^*Pszq?-VHU3KiG-C& z20okn!C7`BYU~&aCnn&-M`swlu|a^lJ}%1qBhw&V*ekE#-t~0Mi;u4<3lHm^@&b?T5%myb*8|c63MHJ|AjApc6q|u&dsqv~Peseyq z`3(s;{cEQwgUzYMa2Ea9nMm5kn$Rx4OA|fP>9?H$4d3BMZ*`AT;EF6-+Ig3DwA`fd zfBuzE1(YUcflE;rIhm_)IU){^I!Gi0`}j+_-m zpp{%qD_tCkILo!D;RD(DSYz4G;dog1g63X}tyv*%Q8RUK~avPrst>eNRYwjwb$1aKzv>$EiDPC;7_p-vQf` zlsPDkW?jpsg;OKQ;8ikRu;NapCUedMFvZ#>12lW{Q3}}UO2$0@SRZnl2D&O~`SV7S z6HO#{{(Tww?E$>+)Nwx%@8&7*$INQ(%b2kX6Z~#Lc3&ND@b2Stl`w2yxC{onW{^+s ze)1Z5hPsPR(|Y|%k`_-VDVf70wc#siJ{bj-ZKh~D|Bgx*CQ-h6Crw~aNqnav#;;IE zjkpT-@^jtN`3C6OoCwYC!+11%4OZW?#JVAUbabm2qM!2nN0Tq+jZDB-Z7Jq)2_$@jFdC`;{^jIO8>y`(0(t>C~p- zq*ytE{%U*|%ysAzureV%{BW0sW%bg=du^l{UP&7k@@I;nDqivGmeE05%+`v+g_+8% zcyK%YCe()*7oB=zCbkmAGw$smmHy|s^_Z*ne5)?7z>>;rs{j7G%c zm(&qH4sjR8#N=1J9!r(KgA<5KLE=XA3@lgFudGWMVimw(yK3_ zG8dk z>jys?9b77iuc;PXoTp6dC-9t$)f>Ug$)SR(0~R!6LlU_z4WolA7t!mR&ZKQ>MO_|4 zAW`RnVQOAD6~*_hdsMNbu$kJfj>3+!^RWDV0OnZiLc&pJSVwbT{H7;#a_3jlm3~RC zVh`w`dKWDzi=eqB!_d#&8!!LP!t>B1WDNfg{Uds*bd>~xMmr#?dIS>RD?;`8c`D?* zwLeqAIaFb&-nkCO_x!MSizj|)7@@1@F9dNDnXG~-8>7ZqabndNefbjR20OyN!VmX; ztVCB-4m^7cVVoKOcZ0L^cw{pf_J5#Cp$Up>)KS`Vm@e_m!WiQ^ikRI>5A_=;VCYiP zWy9&AaS}ddHbHk?7xy2FG6g>4SeBNG!9~3kGFJ-~YQ6MC{~d*BUZ*aN1N5cEn0Hx$ zv$m5t7uF8DuJLRMsl!uWLRhRmLn!}drto~~JfZH^mBNj~jfD7GNoAp*sINy8qlV}p z;kXiJrZ$sYav(&f9Kp(IOK>hz0Q-G9xO!X~2HNjw?CiOCJ9-AhPwv2X342(ETql#N zwNTJq#s(7gFLDux?v}x8-Yb{o6|XZ4JZs_BT{5D<^cGXfM3} z(?&Rbro7Nz_bFRCZ!T-I6h}z22~5cVt&LikCM}1=m<6=!P`15Sw}T9OQ3vM$yMk!Jj9ElsfkrX)F`L-z#5e*}q`!(6$$x zu;V|k09(jCpMz2#PjnWHha6SX&ZdhLQhbpkCnEmM&c0__ADR&Hs@=C5{QHYOTut49}{uEyirfg!k+$43J_|1!kEY74R2~{jwZU&o= zvtha32ulP+``@mCVP7e9_VO9Z<>UOmmkpae>b(E0i*9Kns4AJ_?M4U8x<3!TLpmv4 zawWa**-jbtVRY>Ne)_p7nQpZ!l5vy`j`VNEVgFD}IT()`m!u zgJl|!rTB`3k{L9i@U37`l{!^OSdx5)J;`5xEGSLcD!8*p9`^G`BJzp?e694b!}T>; zP4OX}tVd+<%LTrCR%EMchG%~~ps&$HA^WYcaYh7AnC#~6Sq}uR5yh)-Zw0mi=ZI+! z!7%-9N=!UU35LOBF4HQoRDVeq0|xUR);L6~I-(-rCha+09aH8`We)8Rv*7Y5D zS7(aEqj}V5F&_78R>Ej$Fj4{>@Vwm^*Q~RM)@M@kpl9^PM;5!k4M+5a+a&d;Tu}Bp zfF{QKQdMUP1rLfLshjTvc_F5Pk(5vWoVqA|$q<0gId6u@@;k_4+BPkZLd~yG^soea zn=pZXr!miZhGjTN7_d!1YOTAQ~sSE(tFxSjh98R^YJS( zm77QF&jbj<>Q4xo9+}g3A8XpPl`|=$M$$Ls$AXhLv^cL{i^BRDZFU(!7e}8GB-clf znBQ@_FW5mUoI~F-A&s6}2UF#fQr?^BhixC{j0GvMlsA2Nw|y~g=E~x8x+&k~07oau zquYB2)uk&7h9&lq+hz^u-&DnRSHk9wiTH0zA)VNHnIt)D$^7FuoS*xhe58A+(rG+* z?G5IxW^ZijOXqt!KjdXO;*XOG9tIlILrqz7R~<|Rkv6n1YBoJjEEoK)^u=55U-Z4a z6B#>Hups6vEp}f|D}s#?dw)6R&D@UYBmrEn|D?y7?sPRnfpqsRBL5vBR5@xly%nw{ z4Gn49a9W=Z?vthj$)RN1IflOc>JT_sY!b-N-%HBvoVWg?i#tgg$gB1w4Rp+>R^ekh zsiuSeD$ZIF_K8auF79F;9Y$|ry^Sho|5qDtxq$QrY z)FEO-v8>VBC1fo%ie08P92&{*EjvoFx0p0x&4@gr6+ zshytn@NmTspQlBm`^nv)wtKL3u8+|@b~dvlq_JB+8a;6D`4b9NdMQ_w%!74J{q zroSIe=*XGX=zFytDq;DM)Jw*L8QhId3H&Ti2(|=E2wFvsS@%scfO1R{W-U5}gZ}({b?gyb)3s4*&E2;b^6=^S zHT1dcgL#cNtPd8ErKJ;ha4v=Ix7|>hmx+)q^^j|sfS>h7$QPLp9hvEn*D!}hn>=ba zhSEakBntg?p8SmYeBo3n-MAG_V-L3rMCPTVfoGAXKDmh#p{Y3NmJFv~gc7$!g6hxJ zg2Vo?g0{c*g2eTK)(0&|jeX;iL>`e}X@r&-at_Mll;7+PpnU{pblj&Fl`z`4Toji}j3LJ7XU9Y<$l{+1{T|2{O!~>4 zHyN`ypIRNRTe|6#{xLc-;}}_(c(X|k87zBoENfe|ka-UiFjtTJxcf8{gM!Y$NU`}ZJ3*D!~+$Vp5^1pOc*Zrisn+*w)2&K3VQ$Z#@f(*SgM`Q+5QIjR+q%R2OkA@1f%f3%xukIAE>tTnK@^c4ZiW5 z9v|h(AFFxzH75&WT-IXVcQ@`lRf5J38Qi|wNV@9>V`%?KxYsR0kMS;8?>XioJ!W5$HVpx zEn{;%|EV!sYrSq4H1oCxxj9C`$_*z(o;f?P+an)KA6KKV#uQsE6`{c0o87hAY{U59 zFnRTsogPpX&UhmyJbF@D$f+p8q2tc5yq*krtbY!(ZCkM*b`8B5*bL3uL$H%eMc1}= zGBiFTm~A@&BP7n@KSi%=)?KyGRk&4aU&M z?O*7@7&ml%x=z=Fvgpl&46GDAjm7u&LOOX89*yVy=G{9%GEJmc%5y7jdC0Lz#_0!2 zi1YSG>pWlXRF%VmqOnLh$!CxCT{vQHz~-Nc#dx=&m}b_A0}+GS-oz8gH5-HI%YV_Y zn9MTQE@n~OJM&y~6z9OUk(|$LI#W4^3QmgSi>(SK`L2iBaPFPXn@<BB6`Nf!H72RyxT0G$o|&)I+F zc~o(_ej!M3)^;b=DA~eeoCu=zdDlu-2fy#-)Bc+8f}0ZGXz?rV%a*p{-#Jqxbq_)G zaycxXFb#5T3efiBZ1YWXSa+KYd!i8nd&?q>?AB%pyH8;I%N;z=bc1GFZpGY`YuGV9 zo^$_(VCs)Bs6I7DNsuy0I*6hE)?(a!Cc*b{-lXg`1kS&MuxHT-gv?Mz_ru$`khKuM zGvny>8#k;8Y=BDSdKe67;N2Y=G>aEd;_E!x&*y({8r7g>HU;+^zEi533@cW4WcKd5 zOf|U=&qXRRTG&Cxwp>lbO>+j>GrJoB)6%m7~Pjl^Ys&b~Z8gk<^}=!()` z95j4{X+PIOUgSSC1*T)AOEj)7S%{xYZlK%Rg{|=Y53?6spb^6r1P3ekz)Sl&WSaf) zr=gije$Sx0KLeo7J(!})3ZZYJ1Y6H2s(Rr}qAtVf+PXM$nR1$LEuBIZ8#%KiKo_PH z57G7==JfK7Exsp(!sZIslCRbNg=WA-E5bsXFMnL7oB&BEwMLs1@3fphnJ z5jUTA;kmC&)?5yL8i{n}^E$fDIVgsf{%{$R!=GpWq3266H7LHPy51|8xGN8xIW;i4 zAql^X3PIsno`v!~1m%hH*kP}N@FHh~u9!=1V`BvMgL$7hkUKMCt0-}nCp`>uqp)k* z7`J>fzE!=XlW#(>!Ap}Fg^RK~m22^g&y)w>l42p64cu$lNA$^3pf{oft+~(Pr_Xz{ zVe$A~+JotL2cY;PjXTT}kmp>>`xB<UQm^VQX8S5YJx z@;=k?f?H%~BZa+hmy%3fGtKp~hWw&Q_>*FS?ise!JkUxdeNj-Dvm4>(rr@oQI5SB4 z1o3jtKc19;SAOki8ppfPiHi9AGl9Imt`wl7lRoG;V0!gfICJ+@*o1Sq@mB@)Ihhn> z>x|U--l(Wvh1Xr?7*t~mY4=0uOy7shNJs4TYNGUO_WZkR&GKcOIH%$Q;&`^XLR_CM zDII}+pHS8+_ndwFcZQ8#IF3!6Bg$4C<&)Q^d>^~s1~((lL9kGVHP4@pn9P0H=yHhm zzT>m3G2AOIe+E(-zOe0(#lnC+g8FZF=%b1ow)ZLEt(7NJ>F0aFebZUJP?>GhwO}@` zmh9Wd;m~P0O+O~^=Y8c#&MriK#&^=Jk>^Z*&X}6b`QQ@>sb-UC$DmZ+Nywy%(|Pn^rX5_opV4sr(Qw(l4R13O z&^0C%x)OWw#Kjt;Gb(BF`xNRas3ae)do-AH=ljAcsORfHN{+XNEYEITyEOv#L*EgJ zxzNnsdfNGN9<7jJFp=c%SiKT_`@I{>PF*1RjSb{9(HoYxc460(J#ZiDi=&)L(f@_- z0lz&U`;IPJH0cO=sGSm+3>yy3fE}>cPJ?FkX>40N7q=h(BYTx0R0BuE3)W#a%Z2FS zO!)5$0R<#u)Y3aR$XWkRobmW_fe5SdI*zi$|L}g44V(2al?BzFV4k(d*}SaPtgF}! zYs;@=i`7tO?&-*^^Q>5mjgaeTFjoqbkm--yO%Z>w8$}^I<|gLknTrYX{-IP&wf| zz1u9(^Artgy90~5v23-z4^!IZz|M+GGOJngST1=QyLq4flH^O=T(J-0N_OaYGXx>W zPoc}|0YW!dU_$9lXm=-|=<{F-_;-)?9~cLdr`5>L??*b%oE+jaG$W_sY@yy578COm z2QME)M(9`CB^2OTVIRC5%$dC6QdT>775kk$gr$#Yp~vTVhuCcm*7H5*^GUIob1D?u z&hdNTUhWFC{RHWk9&BCxa;EZ*_vpu*!m@);`Ru9#CBE+wIgp2}raAZ-UPNbQ$Kh9h zB!aGAM6h!{^cL8_?8hGx5wAzYJax7?%ZT+I6k!7)hoF0C6|#yXncM;kHqqRQxh6?7 zx1T8(<57dK7In66u_J5X{7L29$t*Cp5AXk+!8OTYEO6^w*5WvtscguFrXOd!^@^i- zYyxsJY9V&?DvsZ6!lhHYv2>&@yIpgNZP6)VVv8c!c%@B&OmBJxLAD&*%62pF9g#_LYWDYbWv6HJGtEf-{!3V(T6oY}zi1 z$lgQr@1+{lPsE};?JRD4+(yCIDEt|;3p}*SJ~!zzndQHb?d*nNJ~wp;(4nPPhPWf5 zh||g5usXOI8Uj&#-aZCzKUE-kTN}oYe~qWIw{Tl45{)zc>B<~yoH@J&if@zQH7FU~ z^_zicOX1EzEK1#EtUMj5FU^wfW^8&9#BJk{RE=jy7=B^H&AF=7d zu34XO{Y@dp3RHOFXbJbhB%^K5PHZc1Ma0t_+VFc4Vm10OiaD~~v0g0FT8BxixDcM;8nnI**ey{8tTIIsjV@2c@wlk=r#)Irb?0p{~mfaHbb*(n7z^QqAK zx(!bo4ABw59rjNKvon*`*sgjlw%kmdnZ}kw&o>r_E)Qp0#ilV0;dHj`f(kR`>_x@p zmGtzH9cF7cAXn9z`JNE4$mRx=51bTa%vB~cUnlCmT1KCG^XZn87ZrA=k!RR=^FZFt#BI!_l@H*-Jub?`bV%0Q>|Esh&r48Clh*} zi>QFV*NvjN?}2+uG*3C<%HQ+!=ZQI14XlFReJQqn?F{zD&7WC#im?`#KXl}pGw!@O z!t>`VIYZwN?xV8l?y)ckb+b_=mxqfAXYuv9Ge$f#q7&O%*xg7I;ru)jR)|XpP1YY} zQi&5;`r>aS@_IY;7s|1B!j)`gODcOa*M|AF6w;R+ALw5~5*m1}z2x^$7Imc%IltGC zPSGP05%Audg&b1$NFwtsXm zEx$`W2|>n2Q%K&`fy1MhB>JJ7Hg!(m{D)99it#gToR~1zC{w8Haz`jtC1TSNEMv1< zx>%TehI9Q(yx{$wb7`*ahy4Z44=S_goUbEj-1-q&2L`d$=N<6jbJLu@_1F#^_hII9ZOg=ijB5Rfo|DRIug=Ci8V?jI%R52A-#Z9`XzgY zGWyrk7vsHTd+!yU6Jd0$@H~C0I6$kKO39|+CCxh5L}w=Kq=19f^kCHkQrLKjHfQC~ ze8+jzeLJ7>R4w8AY#Ivo@50#!({YY}hD|sBP@T#&D8(f6=k7U(ynFyvU(Ud79f47_ z5_4QqV6wLikFCm46X}V^;!nvxiD#=fWMgpILu6)1un&gc@Zw=R{pIJ|+cWvu+Fyux zr3~~>58>~APocVThR}3pjxcabt}w19SvXWzN0{w6jr{pe(%s@M{dZJ?&qyrzJVydo z1+A26IugAvLhwr=oo8IRi*%F=Mm-Q`d-g14HaDVKb9V}pj0j=Bz5npDD5F^;B{8EY z5WZFO@X7^Hux+PsOP-(LU0}WJbbQvm!kNm=h-&4#ydj?j+8O7mpt6|mPQFe`l~<{% zA)786hLY{#7xYBg8GD~j#pu-WI5}7fGE4bx$L;{%lkB3o+2M3w=PoVgJ@Z{l`Of4< zGwO4@uz4;2?pqqN)Pt|Fy0(iY#i|JdH8q9lxr)Mo`>(jGB9Ntf9HiV1YxJgXL*efW zsDE%7hqBV)TGYY$0P7IH*dHru{zJ&(Nc?wQ8g~q~;CbqKq<+4J!-os8cr52)zRrZr z*=X7>Babf@>yWwSGCU?!K*vqU|J^p~A2tbt7VgLDvM&%j8;MQJT`6+SO{#usi2V*e zc-FrSt^wyU`^^a~>1w3)K~B)zZi@yT8(cg-7SlXF(9aygm^DYRGp8CKoupY}gFG8~ zrxvAWL!kKRBMN>?v8+ZBc7*pnXBq87(~>kQiN8z5_e>$vGz}?+Joo*tiALYvNC#X8 zct>m$b|0OFvmu+&)cBpA*lfazi|1io)eg&0b*>WDVs2(lFdAP8U9(=i)=*^?vkciX z)sZaCp3m)uCE-Q__c(Y-vcvYe%$alLU8>q}@LxYIlMUqa$#YoRUV-g@&v5qQDkx@t zN4EH6#!42ksGZ@gZ^vvFJgOW)@;^z%)f|Q{3lMW*FNE!G5cP|rit=(&*0jQ#XY+B- zXc25ThG6EIzckf|D1Uw~?fdTmja@B;tuha&SjvN1E}f+IW)Vz0WR0V{mcuGE5_+F3 zk=j}gvF^X{DIdyO>Nw+xflYkRe{MnMbbf6F=VNElmiSJ(>G7Ff_oR`EZxY?Qrig){ z=k&~99CkCDFxlZRxf=S@hO#^K(nJLPw$i9-v&7z4cgUe<4vl%)O-JW-Q@MmWp4^y( zVY~j3SA--gQhcz*cQ*9oxZ`x{2AG`?gUHksC|i*a@hSUo=inXWzffQbR;tX~_9rc# z?2WdA;b@MJgG6~4&)0iEY_Kxj>0Ci6@(ZbHbqx6xs-ftX5?WS#CZ7sN7-dhxL2KSK znm!*Bhrc1C#$0k7X9DXJ>$ndr4^dLLF;i+OCbv7#TehDHtFMvMp+DqmV2S7-D);v{pJ>73oHBJd?Pp0C4%8WScw?ZI}}tYP!D*0Y53iY(6JF7*d0A#%lZ zm>*cm{nFw1y?rUzf`7EK))ymRWMQb)X*?~=!ss(RE4F(s8BEV7J&)V8bcGBA#0+a&O$g>mrWTli@9osQThE2 zO86p!0|_5#aA^^Ja*C#&`HSgQ!U>AN6`JbQNN&|!dy z?ml=UVT;Vmj(Bc72E~=8$Q!$WUiJ5px3UJVwpn1hL z!7b|{I^Gmuebg|<9SVlB7~Hi0b~{QCZES$!bK~f+NCOp~lf$cx5r~w{Kpy`&hMAlZ zWdCrW^>-)JJI-C$bS;z=mTA$nv3{6){t|ZI`i}c<=4_{*FMDrk%o>ZuFnR1|)R;&)LWnMrU8p*>1qWpf zX-mEt*^2UAdVT?aRyWg{nGdMR;sEWQ#P`<|{2|BZC7w-3aW}pKlU-g>QFs`IG-+@z z#}Hf;9gFR$LPCyv;iY)7_0YHz^a9@Bh(E z$;tR?8IFDL&!f4#3fJGKVZ*3J81=XqIfA1|>8n8AVty~@-n<(NxI;K49F1p^k)Rk4 zwqqSmxA{VCgAvo@@2%Lg$5>hZH736&l3gqO28rHK%v*jQAwjjs=Q&Bq1ar2%*nl1V z$k|qE`dG3!7(Sc!qfq$($~Y4!!}k=We>Z^kbbAa{ScDx@HY0MHGR`FJ!n@^Pps?{F zo`#9CnKQ>Slaf(v;aWRbXzj&(hbY*zC1I4EC-1@)lDV!S96P6COYRi--yertma%Ya z2}O|Idf}+rG9N28@$+o74$e2ep-JAhm|q!B{mshU->Zpj-VwB^b_{}_PQi=$QB*y`jYj_U zrWs*P0>`Vxf+bTf3%2kc;WFn@=;cxoeY^&zGn1&<_XGX7Z-YL4AJmJ-VCTi5 zP%vLam%XP^^Yk#XoM%s3`>O<3wtCZ^OER?g&=&4c^TGh{j22eb(ZL@NX@x}!3^yFG4B?G)}#Q?M~$eZo|ad&2+D-WFbT z-7Ac{^bC!&hOx>QHz9PLj;hUea1-*q^REsR)zQL#@jLvIvP#%@8%vPj50R>h>q1ZuL~Pvw2w3vHyw3WsR1=qo$0 zAyotE%2VL_-z}=sQi07D?r(ZmOJQH8P@2nr(vY*l^{+9|b#sF0jS@P=^J~j?T&Iw8 zqcF(fB-~3n;8$A+z0g0D;8yQZj94FsApTjM{+UR1$;YrG z*$O{C0;|5O;_sTzH1M~SjJ%*!LYPui6W4V7$I-hFIfl1**5dZ8rS#}1{l`nTFn zI=|pv=<_tu=L-3&=Lj4x-WLe--b49KJF?d0a88O6j#(Qc;MEA+pPvLh_35~ue}hJM z?WP5loF8;Z3=YeK5Pd%pdV{v1daE5*Tihh|pnmeZ_8XP;4(wvP9@7c_i?cxw@$AGQ zo^=|D_!kxDLk@mOzd_3m&$bqGyN`=klzl#w)6@ zzH@@_f76k4^*tF}2&0cY8`yf$1KE(ks^Yt}Ac6NnMEGp~T{Gvg%d>I3+ZP(adwQI$ z%ooAb$oq7zZff{dV}h*BVa3cd+X~^99M*mC*^Q=ofXtyT);Ta!TAu)eOBB- zIQmj4{FI-PUhOWpC@+PttURK9HL#{w9!`87@kS$yR(n6DhSF6y+sdDXCft3aZ^wPF zgGhKMn~DyoVM&k|To(DjEuw$DEws!F+Qp`D*CnBY!@hykbvheb!R#n|t*4@_rI2siE0r3rSTj zox)p85IEc%S6+)Eux2AF%cOC8MjZ7$t)kM?KdD6cjjjcsg2nJ-e5X4expTMC?>TDt zZ@mTPscylMp@$JAdmRe%J|VZ=kS&R?MM}_hgle3{%Z1ah zQ-W39R%JSlW^BZoKb&QVZ2cQD30_D7l;EA^PT$TM(eLyMfpki|V8(C->u;@A0wbj#Bp$)g z(BlA$ACj1KeJ@Qvv7UCyh+>^qA31B+lK090!3^aS^iVa0WH`HKQiKG%z5O@fa2E$F z0pVX`R5_g|RkvJ{l)pexjtVqq*lWSr+C4PAbTFIu(~$K)G-CyimDx?sZQo}k3;mCp z_*~@#*VB6VW?e}K1s2e)k>vbO-cjc{-G5Pibh~~EX8Ff+e>lJEq~2wdf1P0sdsIb&ttA>%ui{c^Hpo5oFK!GmO(?ZX(%%iWBJnc4WZ?G(0uOoQb-5jb4v zr2(;C+JAg7goF4qLpy~sN{py_ivmFBYTga4C#9v#9aYMffkN#9o&uOj7C+z7@5ySsW*@@sKjI zDV%9$qm!g&Q}>QDV59l$s%s^t=zXVaFDj{V*>_6#&ADDP+CU{iIG1aMAeU_Hc8kKy z3F(+A{RnB%T<{DzkJZ?GnE|X);OEr2=ICiT34Zpoo}Y;F#BD1jOB@ zMZ#Q4-E@m0eGBQ@e-hZ_uYfZ4S@3kdfLurX7IenCQ;p9Bnzi}_MTuS#yneTjqw*aC(~zx3v^-VD;nzjgwam%EZ7?#6oXJ^Iu1i3MBy5BLZH%Q$K+#eUQqvfFZVnU*2%?gkfAje-pl zzF6YFzgp<*eMA-oLfp!EPPyjF=nb?&o=gPJbLY|{v*(Z=SPk|2E)+A0mhT(J?Y(zKLWszE&f|JC6n;t) zX_vN=Xe-jRRYq2}5-A!gNu&}DNfVWJX(K7x$^ZJl?L~d!dG7nV&ht3FUz&V3oK!|f z(R!P18XD9`N1oMEyJ`eH4$s26hN+mea3vmm*?^=jW|Q5 zxN{^^IqT2TWmTAV3+FXi_@eylJY-n8BjEXJM0iJIY%KTj7};Wf(O~3aHAc2=!{d3g zNWAYG?Kho*184kTymuqKyw+p)H_j;jC4^??45*2%Lv$vCqO$=c0{FXZ%r1PC35BNY z88r8uN6eTlIAbvdJqsi-lAp))UaWz7R3v=Y979fAA>M5|MDrYv(37TaI{4%x9p^Jd zlg87O%g;2cN7UlVizhI2?t_`w6--)j7&(WR!thu;l!kr6oLntd#xoawwb8V>sGp84 zvV!@6q4<1eJW^fs;qm4qO?lKsex?Q}h?0cV>JgCknT8veO;L1k39hc-y)(^JIE9L{ zzQ8Uxn*E0xHo4e!?j@wNnjm9PhYZsOoE)NtJ1NcdYor?%*F8mDM=zet&*#}8Q$8Qy z4)21I&@3K{1Kk@?=~M&X>7S|NdKGR*D6@CAV_5!|(QGlFZ&{qaKvh9rAl^kab57-+ zoIIYN9gF8Pp3UC8=Y0Dqktltjkj zjfx!!#Fvx7o!hiLScdlnG!fU3LGGitUsr)MBVOAvhv}|tRDU`8s|A8*J3|q@wFWjD z#hK1qQ>GuT$z~*%Qv^Fh1*bbn{PYV_2v>yPD^D0cxkvMc@-y*bLu5;>K}%yaUN)^p zQLzYSd-MJNU{HZg2T7(Z#`D182!A}8X8G&lUxXMw4N1o8AX(=4p%Rt1#!&ek6%_5| z{)jLR9(-^VmzAtpo0O2fGpwWbh%Rc3Os6FcRn(XkPa!4q1#WUvXuqu^JwL3A>o2Xa z$$JBp?2xAy`xeljjL|f#(t*T6_K}qNeoByv6-1w(N6%`DX@>`YC$=d& zTpRO803`)x&?()FN^cXUQZ3IWUr(W#Z-3AZF-2sy&4ye364cGIqX5e|YJ5^fO0CBU z_YTvXt$q-#7lCE-HR?E<&6!vmDK)f(PJJ7Rt<7&KaA*kb{7Jw8S66&9O2XZqcC0>J zf>SZS=nC(r*lI-L?s`w$<;(zyin)mCsu$R#+tBLak-V?Ehr-;#Xn%!0ZOHC{%C#0; z_iD#JaZ$GOGHfu4OoumHToR9QP5ZM53u^1~z^zq*pl!6loJduWwAHe_7Hv{Xr8YiPZv~{(1yf z7URUu0|?-L;K?a<@Ms9d^i{gVmgES6_Yb27r7Af8_7(LPUKb4IOzX3EJkUMe0yW0p zDS4(GE;Uc1^~-nD=$qD7Ve+t^huFehY4oHipRRe5^J(265j7xT`%97&aIt z$1l_6W9{_ECy94^!ZGAQA{-T?X!s3nyj#4OySsdmB|0B9=Eo2ru?{c2?o;!@k*J;{ zM8-pH7|j|3nJ7l~35O_QssuVYZ|CD*&fvUxjt)OqLuv=Mu$`U}?9Ych7Iv$Yy-M?B z?k-i>x+4zLE}w(as#Zw$N-(e1zp(1MfG@A55NS3KBf3_=($E?%4nfdABg8j_9x52$ zM3Pe!@McjM6@GX}DLY2dEYTUz+184d^%m@Ck`CLZsLo6;eZuCRv&vH!ru;GxKFD1t6mi`v%(WOJ zROaWa6Dx)y;QS((Ed39g_zq-IDIPE zoX2T|r|%OeLqrOdi=NWoCM{^0+u+}N6}ZmYhxhT~?4JBzY7__K|MRPpb@ zssucGwFu{&?eO1xQzR7(!j8#*=!DS*ICk?MW7sYFIsv4=CP#2tAO)wd3o&MYE{et% zqSbN-f6u8c?Ee}1{}to=x&72XahPDP(IC=2!FQN~D(L(_LxKn2lkH^-sgMa*F`5`n7tkKhXy0X)Q1EU*K(&% z3VIGa;oRv4eApI;^IzIYBeIsH)Qh;I;VpaC8^l&kxr_uc8%& z+6Pa>neYr^W|{QhUnzQT_i$F+E9k4n;?|^8^1b3ue=klz5`Uhn>`Fu5C!YBX|H9dE z$uvXqH}z>Qg!>T_{N-7xH(u#f?Cnmy3M*0dC;|3OWiU=J#jeQbt2;$GPuLNPd>wr`OL4e8e92y#>hDwgE%YckXJ8MHot^eBJUL!hSCvN-YM9^`{ysX z9dvj-?oP7@728tumu6CWS;aXEIxIR$li4-%9f@crru1px$vrnVX6qiNBoWO==nt-AMEwFU33-ecBu7%gR|K@$phFKx(_q*crT|??2~p3Q*(I%@6V#BV!Rts zC(ep=i!h^75f>Vc(=6o*y1PvuiniNf_wfiCEd>EJLgEc`Yd+un|XgC6H+ca@Q9 zSuk@e@?u_M4RHTC75=fwcxv3sXR*`R$T^!>`W8N4oUjIp*W@wyF3%b5u*dg5MNs7q z*X1`8QIgkB{hDtDZ5|hK#m$VxJ&|R8Y$JU7L}|sX%Yu><*-)+NN2A_XoaNp#Pu5DM zPH*VvCQ~*#Bb?E)p)4`V3?Bl@X!?r($lhTk*3Nnj)qe(Tb-x=MK6XBv4U{K{^|oFObpLEpR$D1HQj<@aMN49w$tt=s=#)cs@XrZiQ3tSb6OBeN1s4 z(ey2H12KaTa)0)j+AsJ)<`aLX4bg(E<{}z3sYjqLl>n1EDR$0Yj-6G|XX}gf_)K1# z9s79#6Rr+n=6?Ksk-QTJJ$E9Rb6`i#RlxJBfzbTC5bd+QaZ&dqIX#(5b*&K?+SG}} z$9<5vdjQt=RB(p(%^o}Y;Ydsw0yX!cAuAMRw#^iC^&*L>M&qC_=TolHWm0nYv3t&a zfx*{%g0|!)dZRfCX6gEzA$5S1Yq|v=y)CJzR*{w1u45<1Kjfb~1m8DS z5bNG2xN7-MkV}O$qDmGAZdKx}({Uv6PDFx&4>=uuD7Y(<1DS2VG5BHTI(U)7E^x1zK=fWI^P5ccMSAE3v=x`X+ts}U<7p(TvW$rE3 z%wD|^Z@i45*CP!!RTsA0$?|WV6+1rBf&G0th#gxk2R+M^bbC<}5*A(Op29^i(8-|W zx*Dop&**mK2^tk=8;9)3}r`7;FG-g~i7`Z|=pPQt&FKS)FC5^a2+PHBGp zGaW9N!?SS^l zm-J5EgpEna zlKUT=`5F4-0|z)~%ECU7yCVL5L&e)e=yBw|m;dhb%W=OIc#PhvfbRZ*^-bGdlmgjz=xy#J)xiNrx8u=ZRv!rdj_tB%Q zsYtT5hPe^Mn`w(WE4h3mG%Q{|8*}|FpvlI!fl?oX5 z_8v}hf4ZTcJ_|W0&N8n*#&cm8I!~Kn^pe5&z~^E#lPiXnkA`RNTIyiwXxDdgbmI>IRCjm_E+5Iu`t(S9ECdR$b^+3h^je0YF@Kdyt_?g5m!}m&F&KHwQk1rQ(|n*xm0|Z_?quiR^YSg4oLBiq>t4D zGJ66v$n1xeu_wyxQ?Tt!9Ts%lgNXGNycREj+c{F;j5jp*buoTqvYQ93l@M(Up0FaZgTw+Y0&+)9R;z(xLdL z`Hen>pQT~vl;Jtj5PDyY*cu~scJYZQi+l746u~`Re(ex@IhZXu*oU0L^SIHY&+3nW zy;l7Mi7OHa4IPfMyZkv_caQYi4RNIR55;Qir>Q|x>6wfR4f*U#Iv0xQ?*TE)tW!ZH zpLM%FF@}xwBIJ+aOh`)stn<2YHuE`V|4zce7d=QE_Y(S2jfkD}o^%2)3f`Jj)1rjE zeD|!4Rkx~0*7yZ}2o;#9=V!!xIf)C0+7a)20#yx3m>6-1e5Fbe-l)u`)Et4uvWZxn zVu->O6?D0ovmPXOV|*mf1zeOzlz1x)Hnu?QO&*SRW#Wv(2?YAjhc^G7HMD%D>!K$p zJ1&w$$IL_B0v%}njf8cUKAstyaZdd=s#<7^p}7v2dGIdDTu-OkrzQyT8IEOqrnqPB z3@Y;-Mk{|X8hoUPY_8>#m9{+�)|2%vlIUJWBZQdee?|Xn)FmNV^t8&MJ*F?^ymB%lG0Bq~ffPBS0dJrMc85_gl-Pcb)UMX>IQv+=s9zvylxzuqupH3O=rz+#I zST$%eG7CyE`qd+_@nY=iV-a?^ItSt=n_zBm0Sn)Kf!OQwINs*PXJ;8mxnP5yQU|OE zRsx-#j!s+t`xyO-dJUh@**P<@SLqiu^4*+DS|oH8*P>{!7T@bFMkntE{FA(xiey@cTUIUhM};(ZHN_Pwc&6{0Umc| zV_01wN~%Xgf@g}RnM{Ie;3G0=;&&jUWE|i>3-A9V5Vx57rL{Auhv$W~Z@1!A3HQt{ z4aXSmJqS%0hv4LG7(Q?vS|i`$h-N+xa@M6p^CZ;Yn1`^(+`HP{M_-u z&E8E&4M|3Woh#IDUJ%T~SbAG{MR1jQ2_6nRNcU@8XnbWWQHuwfm(Jq*aL$$9HwNDi z_tRgy-559W3isJn!jrosZ~x+X;HUCT-qI1mt%~wg8|ie|C_;O6Cl5QF7yRU;XZr|?rJyo<(BB-X)m) z;xu+%q=kbT{W6RfmC;7^IkmKjc4X=ToDalB!KMzh+ z;N9T)xH@k$CeJa#DW^p!yEh;1C(;JedkVzXrru;7R^ z)^T+dOFb&jw2kjTR-0!H`u0OgyasPsD-P%t;_|J{kpAY0DBdgFux>Zz@NAgABj<{# z-#@%xa`h*6r%AGzjmAv#?_`$Gx%IpG&b*Fu_GI@yrkkFe zJE*-J8Z(ALLr)z!E{+i4Y_8j!6E8DA0?S8CgiShx$IC_uf3}zjRfk&%+4Skc5v~xH z-$}vv5lIN-{x-+{Ks;{aP8E~sNaH-`ftx4cxi<}u^1orDY!^J-siuHxmv-1?W)OOPkh23QFxsW7dzEJWcBk0NRqcygND0bEYy5YTr z9(A0f(Xx#+>uf#c{r+pg zn5$3(*Ng=S+?<4(PtBM)>jsXC+`yxA=lK45CpN6)d*~87^nTTcf$e=#YEZ*$zWZ2O zohY1~{az?y{7BeQd_y>^yF%z1HCyP+Jq|BUexSIWC48Q;kHl<5;U9IAroTv}d)E{( z_wR4&GJ8wcp7hb)^eRe}k!K0%er#s)6c#5xgk3IIV#B5TvEj=Ja^(N!Wm|s!&02sV zCaRdb-T*iIs;JbUkKQ)Er-9TfocZyKIytjsfcq9dYJVYF^FGR4sfJLyGFq*^iKexl zq6=@tFh#$EW*B|v%*e~+F*29Ng&n7|%}1%5KX;ZN=NWa0r#$a^oU(PtVq?r|Tvz#t z(@g_#JI*_CeFC=Ns-1A6R=zO#Nr5omXOD1J`WoT2vwzvO-Ym|^;~v&u<#^>)jY$R3 z_-*xtKFx2Yt2ZTOY`CRBX?}wvI>U`J+X47EPjv7qAgW_DJJJV`Qjp- z?No#h|EwO_X~_3w{y21b30D60#GtK1F>8D^tvxdaQv}7B%5$88?H{mwWeFbr9fOWx z+~c6H%mNCP8NY)s-S96&mbWCg;~@Sg=eLNoG2f zyH6)5<687=c>cEx9y7U%?}RCC-?K+w-ZY3_Yr?)uN(@~cylapH|F0|2YknSK?o!M# z<`pjG)L>@lE8Oa9frghRcXJVZw>jd^VFBEuX5ew7Dc-I3!V;s0Sog6OqE^>1E4vZ* zpRK|3m5F38I7xnCrF8y>9JXA{rdv;3Y1RG;8sEnIeU0W&Tjzs&{Mi{k`6>-jRfdy& z0HnoL;k?TXd**FN`=^mGX&q0p?MsNRETxpk<)kFFljPj=DOm9mf%Arwn|70-Q5_Ak z?V-*)^>kiLp8CaXIA>3hR-`zPx9coArORiXt=mZg?@3R$ivpGv)6=K-Xkx_`>UP|U z)e7a96P=4*=UuSXU5j2z73}!88F`DNkXloU_hYYNv}grl-lgNam;;`C+yIXS$;fnE zjG7HLkp4Ot_uV3>!7YzwJbX$M2fon+6B#IYmD5AXa#HFPEpah z41$rdm>)hHL)Tkz-)$ku=8eNE$7;%S90G^JZ*Vxkm~x>3tG@M=9xk(j<9s13`VFu- zX()c2JWl7Vj?lY#x|qwm?FVEW(JSMG;)+vLJ8vo7D4Im;@*Qbdk_sKkji(cFlPE>& z5NYmEhP2&p+LrN@3V9Dx&gUcDTquFL75y|iAp%+3Em*`MdA8Hpj^(V}#91nL=x^5ztqw~`*5Nv55{(r9+{BYJqHoa!VBscXFidHryw z6btTnyKs`UG{Y#v^r_&{r(KYn+lW=7a!mf;X!cBj=N{_0Lu=7cEIE*kZpjvOxF%p2 z_m14uj)RNk37j5r2Y;?yMfmLBuzPCDzGYlPTlIPRz5Fkgc(#-Jvlg!tmTfM z{m(4vYqdPpNbMH{p6#OemLxbegk$}(b@1nN^v1zi=z78aGWjRoQPP?!8rSkuDs#?xOE1& zyI+9iIW|b2E05n%BhV%{}yZ9dzFW(5(J zHcymgKI05%(G6(vD8c%7ad_u%k9%tzpqyucg&O>fxalsA$u`2tI1O^wvhYLI5tz`E|x*9UV;6q zm1leBbz=J9KWHnu&N;%sBc+}2zPuh&zD2_Hj4!-g)$nqTKU_5SW47y6Ea&ep9nKl~ zwb>PFx07JmFPL)1A%{7&fKTN8h2LrN`w5Wg z_>AFa8jvHh8{&U=QLD^f8oaob)CRAC+9)eT=PSVG2cM^4SaqnVq;lU=F z6YC5KwN%LblH#+h2*H1A=FyaZLp1W(9c;~>!m53Am{`-b9&^vytA z_d-c42(&S;4dhUM=S?zkKz zG&>Z%*z>#4ml+sx(Hd@dPoic$clM}P;M1Wj z&P&ySb;CQHIP1ZjYm(Tb_s3cCLf#D-_(0LWc{kZQm#y8hn)Rk#!bZb$C_MieljLIY z?`JFQeD&F#toNv$nN3wT=jf2T1J>Tn$Bf8UWOvu%PTwh7$LB#qzgBP-k2*WF@-9O4 zb3ng}k!V=~=W8A;t2mws0}|Oy+a;`d+7)ECRU>}lTR5d!vz41Cz-iPRiYOTj`&)_l zvnLLXb`splsK$&0hhX|QjdrwY;_LZvrgWl)`Tpz0nVxZ&8OWVm4)Sd6vk!Q^L64;v zZN>&^J0$&#K!9xqRxMV7a>ZoCowLK=m7L?8H-&CYir}**H}t%Ij@7qx*rKq#2-BNM zBX%yN$t`Q(Z~Yi!#T3}|Kw@cg193!noxG1Ep+d757hlWad*Wicl=F*xmw97ZXFVAh zyThO(67?~|5LFgNF7DwpUw%AlpFhCf6_?PsdLIsrlqvJ` zbPVY!Kc`8P9#X2ZB=REHA>{ESM2_NFv#ohlwsSA%1WB;GS{)`Q(ShgEcd%RG1pZiE z!TQU@V%M%_X|Hy&TRVp_!Nh$?tl_SezI+JeO0aD89XdVIg1!war@Ra?4BfF7X#;04 zb^LS0O*;U`?QLXI!#OSs;;`X}62!LsqvIW}aDQlmyNNFm@YIF*Z}VewyA;{Mtm_E6 zCeGlxhniia5oPIs+dgt|5ar#7KNa*=PmApwuwV{@oUwRVH0UyCq`%)nr!q<@fB$OQYsF{zmL`bW_8+Py71I5Q z8MNbBJKf!siR)X%S=|B$c1LOuD-4!r*M7zDZdwao8{fvnt^0WQoqMx4r=cu3guJx| zXtr1&XQT!qxcNOTlC7inJj;1$Mmvt*8^o*4%!81mmp*7k~fP&LXS zy0H>7=BP4#zr$#F$r%IfbC^X*7+dxs5}ztMY5Cz}bTUU8zqTr&NSH(|2@|L^(Tz^H zKf}92BbZggJT_0Yh?%_fV#(ViarfqP^n4dGJB1OfXmKhw?;?o5NXEmyNPO27WiuZb zv&5w{*wJOHSna4wP?XHUv7b*dTRjY>?&~NoU8)w0!rOaRoO0QsY-5txyKEU>u42#jLptf=|EPr5+H@=C2%ARtEp;t`l#o1U4 zn|%Qt?>O(;q8V>BZqUlOe9C{Y3L&e0;G2vK&RWY+=)7xmdh7&T^6ka<0Z(>!tPdOd zVk*;`u?Jai(y>OOlB$1XgGMZbVemIWbYPcYw!~ffntTZnwS!o)%MLbly9bMl--;ME znLC^Q!%g)Ftmx-ltd3I@xVVri+j(EW%aFwd3}+`_&%if}p*%xShC`p?vEqif&~2x+ z@c6gGOx%Wd1jF{T^~WmMUpk6U3*6ZI8%gZkjY>8#G@Lo?I)`;4{4=t?9Y;7zqx0n~ zwy1SCzRDgTmHrA^@Kzf$L`2XyD3TT_9dyXtBTsK9*f0axeEwdkWJimeS$#}0n^iuD zMZWuvg*_T9U!C(zW-8<1M0q&Ij>QxAV1)5!+?xwiawbvFy|hkhWPM(##_!p zmB-pEZt&`JMEBoVo-0X#|8pl?mC2!JTkq4)EebgCw4apB#Zh>zk}d^s#^u3l2BaxUZRKc_h{M@X>7PM8k3i~p~g0XHi++~z{G*1>IxS9Y-tV?vstW zGPVcR(V7n_H0NhHO{jLHEni~EKZT+q#oxP;AE}eXChf}+s9bTk!Kj&TUc`dGh z=7>BD^c%u{{cQMcy@8ho$#{Rkga5y0;i1Y6Drzj^{rgv_jkIGAqimSd*FMZ%pMVDn z5_r|%j1@i3IJ<>m#4%|&Nu3~zjIqLPiaUi1HwOvT>{ba&e@_#R2z&8M`C!>X9hq=I2vi)hKe^HinJ7kj6o_AzBy9x+`gYW&8?_T&VVF_93`z0|0vGH3f>FH zBICaS+Nd2sY-9+}8g*f#&nZY-hQU*|8Xmdja5eE@{VU>`v+Ew_7rl+?+KguL#;-9d zMh;I4OJNwzdtUDq*iDmE)J!*ldtE1;wV8#H+?#bcTN9TCou-Xzj?l6TqoC!q8=K~* zW9Ew$Ftnb_@3zL++jy5QyLqDYXaiP${0qrNTX4wXGTE&+f~$TK&I}31{BH|T9b^de zeragTKTIp$aDQ#!Or+%-;Mdsaq<$%gL~ijO`)Dy#uam~7d!jgyb)WOUVrY1d1;zy> zKwb1Tk|z9y!;B~_*#3tuecq2>aw2SwVJDIv9EM>(_l&EJ!M(INddodWKlv_cv$O~< z`lQjl*9)maqMEMF*TRzZ+PIf|h33ra6$~#3;2rmNnrm%?fsi1Q}3YyZHG$-nVz);qdJCBx9&<^f8sW+#O zaRY*P73V4Ny(t!Xzo5_V<@BSeoWl6It~nEYj$cI+o+r}!$${i~Gl*=`Bq`i+BNczE zC7m^EiGE6Dq2Tcoxc|~)pO!H8 zce@7jR*b;(^`U6a7tt!Q0$J@-kg)Ov)fb4(1i)5u$(@ z=L}JmB8LE(4+P14y7+r69BpFxjsiS)_k)IgP{NFk8&q;=7)r;_hGAR^zOFmM`(4s( z;h|O-t4%@Jtr#pzn1|^HWKb(Ag^A{iXpYGkgw(7?+XGW<7^RGVZNEwELn-ZeVMz_X zlSwj2livMR;`}gQvM9Bpg6LxUlC6b9oVlRgu7Ssc(&)+IXo?*_6w}3G@L*;Em`ECC zxShhE@*dm>)<$>M94z6CU)h~g;1gtx@rMr5Qv20(m@`gPv}WVOTVv!uVqo*;gWw@a7FoV87_NiwMc>aiT?e8T5kY(dyd^sMkH1Tr}-S{OVy!D&w7r z+MgtM`9H$6zk=h=qbVCJN#AQCb!d&G>oImD_x`%z!m?82|7?Kbz)<$kXfd;qb!2^3 zwRo6541+vhV_M$_1n_Rb$`!7d&b=jC!-`;gH3N1%w=qmxo_*<&Wzn7KfTjUXaG!Gb z(!uciah=vL?I4Rs;na1XGtYZ<(50^mwXZ*j*&d|1719>k}4J!Y_=#lU< z?*M0-x|2?$Z(;7oxti~wgT%^2@;r%H?9Q%8dZfq$caoM*-CmT6W z;{i2t)`zmW7thn2BGEzODA2V)*A<4}Z$H!Euj}c`(GYZJKgKk67=@g(EjTeA7Y@{s z)7N4I z!TwB^jQo(w{3H9MZr87~&j)*nFol()kp>YCXmc(uP;II_U|MA_) zmS6;!PDicd4f^@!7dl6eVA8ywVs-EVx~g_z-XufBmyX5k&}EqUeFs|ALNS=n|1vDz zlS^m-&cqj@QY8jI7jiDfsu)}j_QI@f3P{jj05Jp3ALIF}qql0{x_S?y&ps7cCRI|y zUJVQ!oq~6Dc?i0{03s%4$eK1AJ39Ab%;MMB+j0z#W|<-}DVr+!4r=?!k=QdS82XDt z@Fmp{@?S-;BGd};RZCDjVL5L7Hy#DszRF=z0R!)J!WY*&aFzZ>ITC&Qz; z75aH&+2A8e!p1{9-|qE*`HF{Oj@53gs;WV_>H+LMmxf)s|Nnf4U}{nc%6T?=U}*+tHSn&g zl?f!e)-&}hO|1Rh4|XrCm1R%OV5!qbu&#|qXlCtq3ih+Y{wys7_Nd?j>!FJmRoS8& zD_CJpGMnf*jny0~YPK$soowtd-63u1_n`f{r zt{zh{N@rV6oni%Na+q#LD$DOuW}P*WkPj_^N!t|^3^e0$O9ou(R%6=AeQ71jxZV5=j}G=5ny*O}8;qU9>4H^G@nZhC}`$#(3ZX&CEqO=7yA z{aMRH3#PF-7aAWmP&RHE^p*vn!hRPzW^U&$Qb&vlF+$VQ@rb%T1C8w)kv(rNY8A!N zA(4r8-4Sf`q%J(ye1f8YJ233>z!Y8Hb#VWN7@-c+J*&vf6V=$ZcLyw*KyGAS^7$@_2|7eQ@yE{dW{qR1AdYcJ|6lp-|ZDQicwfTeBvyHpV^iua#8%Ek z{2R`4;vUA=rUh6fUW(F`0~mM448Ltv5bU0w?9)~nZ~Mitz+?e5`0I&b1FNcvBdQ-rq*0Qc|j($ z&hc*hea@y&nF0xximre&&<{L?^l}9hsM?~on!ASMGjaHA1oz8%pyo$2H3wXRWwQ7XE7eIXEKk}RJg7Seqkcf@Mo40!)C25a<3)7j6=MnZ#{v?mi6tF>G zSF%py1ITol#(W1|V!_-=s`T<0yDFZ+I8%__Zt=$x?Izghe8n4{QMnlS7T>?8LB;YW z^3pk9Owxs&D7R!Krhm|V|1bu{oFexM6P)ADyPSV}Ve2&(F@CpcN46}RxzL3r>HD)$ zCp_8fuo28UF$ei=TD*7m9Lcx(@a5uoW@memEj=H>gso+Qfcz{PrLT@<|CAs%`xM!{ zv8H8*R?yC#GTK+(PIXJ0IJ2pM^V~gX6;)Enq8f_f`JC;4yfA&lWK7BLr<)fTGz#ay zqAd|Kmh-#&n;6J)SN63o5scQ)L#0nCYC9xYz^-^!_Q{{^K1sOLn~u@m8OXg^fl}`j zo+ab?%t}X$9$o{hdu^~TdyBQb&ml912Y091Kumr+a$o*{^^vEL(7c2(7Wz z#D#Ar|HnICM#9^Zf3oEpQd$4VZe$&L4~ex4S)^?yi<{fRwthLy`tGV%zABJX496TGt>DI1#GJT*9ZwKxUf6_~8p&#kpH+OdE*)I0M>@#eJT*vb^RaQ3L zhUISWMbn?dIOk}E-IDyB>7~N5MRiPm!#kh*N8-bh0EjqzqJTzG668K11*Ls7VyHIGpPqunJ^dse^pRB4 zIN!P}5{*w{VKrwbBokDyZATXA@!5+1Um4W*so>e)2|V|t1=R!L)Ti-?x=;Gh+Z&a1 zelVv+OddrOE*=uFUGe0zX$oRPQ~BRjgy}05VBWe)ic->oB=@0iI<}v8!3&VfGm<^0 z)9C2z+jPOLpDGI;P-r#(_iXG*FKs`B@of++Z{`2mIh?Z#hsev{_8TT^ld8`1e=m2&%7CBFm)~jv-~3I$($FUT+lm%CSF>EZEElANJT{5sUs~ z&9*Bqh3&W|tSN56%#bJO4k?Dioe;boyijP}e?&MwyjI9|Rtx7e9v1fN4Hw?y_xpDS zV_+i|1k0mADD+u?uluAB!NYVzoHH?UVhmnyjCn{RXe;` z(RU+u-h2;&|NVzNwQelDD8e@1D8R0%d01&M1NRo^;p@%Y7;EqhlikFbef)LUl!jr- zkkzEcXOV|Ka?eq41$`GBqDfZCw0_nFsucC%PGb+8%9X>@UZPX zcf-Df`G$|Uz_WZO98Tg>@pp{rtA)Ml9VA_RgguHC=yy5Bt{44cca24aCB1jq_H7w# z-0L7F-*%pKh9@E;w;qM6Pa%0O8~W$RqWm4_-Hh*ssO4Q8PQC(Bw^}H0_Cr7 zF;O#PxlwD`{o;*mqYqeRj(6eN zyA-DU&nC57;cK-I?R_^0{nw&!NG6=;cx4f^g!3L_wxZgHKSM1CGl8Epd-wGKju#K% z-R{E}?>(GlhHhZLHGP>%yd9f4_a!zc?SZpx5&jrP;QO!b@VV`d_@F?Vl4%Vqt@9WZ z^czni2cX>12%V-(4B2DNRwQ^ZWK3ahX9h8|Wt`Lg&KujqYLS~?hfP7R5n6K+YG20V zk;gtV&x@tHag!0+-~mJ5jXWpef%}d>Y4&+*?&t~T-7Ig+OjQ7vq9QOOl4OMCBvBoK zmwew?7yg+GQfsl;A`ia=lOWT%7F(KQvEOqHu6IaavvCW}-nEb91Fw+%St&$C4TJM~ zb7*8mA^ceeC7${G7?{Tye2bAiqL&i?a=*^OOh{cd$B>yOXw@4Aue(u{WLrzR zfB!@ADemmJ-$&o>z9Q2tb#yOj5>}1MMGen?r;k&?VE37LvHS%cSjjVp9d8j}Y|SJj zy;#qZsjSFGl$9No!+*c{e9|-*Kd%(vw#zEqOqIjjF)Iac1CnUA{#KqfEk`Tgr8h}0 z!N~Jtu;NlK?$qu_>C0q{RGyAp>k(LND8jN+g4meiO>A0b0FysT%ty%IU1=FK^nVnc zhhL6w8^)WGN>iFbAyT6IJg!HYq$neWL^dHJiZXtrrBIO;N)jzJDTPR6Rip@MAt@sx zE9AZ2e;__RPxpOY=Xo68Z|Gk-!ZVO_t`3JW@38kAmx9BSN_@=LVy`2(HzZJnDbBis zWX`zG-h7X9Lv`3+hr!Hh%5WCVb0dyPZtxxXlq8cha7$%A3`>XOt-J_dCUn!+*kCyJ ze#Lx~c7(?A^W?247;e-+t>;xbwAm1)d@N&eXD3wu`oMV$=Wflq4OM;4yBx6}@}EoK zWuJgx?l_(lmQAAL>O5nZhVhfyxKni``@P}|eDZ@RYzF6Fw3tBe{&+N${UjY&k&h5U z7eYT!?RXiqP$%uVdxb8^M$uxU%YupMhp>HHs$iDn#P^R8baMA$`ZnB(p1oAWE}7w2 zm^%oG7K891QU*S}>)|r!6a93KL{lk$_ci~*l&Vbpyfz69d2w)!48{eXn;G6zi$%ls zL22w*^3v|2{8Sy3pC$BO<35Ii`Vb~7a2DZqtl&Kx8^20y9TE@EHaqmM5oZ$0My$-x zi7mC<$j&J0vp|D56b4A3{y;H~2kEk%y#M#0e<7ytjKRiVfi$TzpB_Eu4i=keI>9;D zlVpvkX3=5c)2zdY*D+%kbiLWl9d2y&Eh*-oSd1$(1~7A_5iG;FKYN*2fnF^K$ece+ z%S~-?;`2k^Nz-C)`Hbe%me(-f9fu!!cOd#NozG4)aABu3J};e2s~pQHC1N~&p6(^N zA9|>Dki?aoYT6N{#1iXQv&?I=*thP~V88Mo>YIkJ5WW8F zvHt+}siz1(`kQfX#cufWJ9*%xFnrBkhM}vzkzM?9RC6w&s5A=~U&Z73i$!?h*g{=4 z?^sf`w&>`Bp`saPz3f9w6stU~&t~5V$C_+a);>3cUCvEp(Sr}Or0|*SQ%FCY4==~m zs{ZU+=NCvlxdPK~$>iSO3HlTdXu&FUb>!d@XOu^a#7XS2LjiEPuaGA8!? z6l*)Qgf;X0rLu7&+!k`y*ST11j|zj{DRFF^m_gb7@~C>K5vqqJV|)D*{QD7um0qf% z-pBumb}x<-d3Pp>O#fwwp0qC!U5OlwQkFq0I<1tb}Bc6kB>_1Y0bv!6qf1K&@LDY--imj)6L?E?8hZ7Q|9` z@}^nXOzoVVcYYYpbp5m8e35ZTpHfC?8GnVO&mnYqbv^alB98qRh9Tv7o3QYMA(nqT zh_!N@Z(3dei&4h78b43?d9sX3V^r|>5YNZ)t}mGnA_tSBg4%ye=n0x=`<6c1)31X9 z0(R2xv@}7h%u0Cq;4sKMaEf!g*veq0MwlID(`#pX( zvv|{v84Z8vNF|@6)m(yUNCQl}&m+##8I$WRMH`=`h;ri+L^so|M6y3BnDOpQEYn;a zac?ToJyf1qKJ4RpJ3m}g+D99UOQ@H3@hz6#!kG0}(arahN^f3M>Fr_cu}=zn*>ISR zIm_5q3pX_9gu}!(g7S)glO{Z{`aeFqfAyZ?7kiWaq_^Z;-VBZ5W7&`BajY|4h6T0T zBjDE|df)ktjw)zDVQ&{5%}Cd2%z71Q}7GdocG||xDxDc3}tdMH<_P%3CnXm&ZZUk zu`PoYSirk`#<>bbcsE_72h*3L z@pldy4T`Zfnx7kV)A56Sr-&j6+OC^F>zeDSFN^mJ$q9`6+)`%#A;SgZ$o=Rc zsw{tiPniSRnF>edwC@$ZmipkmXC0Y4OV5q>OJ;$eAmPzd!jI zp@ZM&FID1Z-Bc{hK20@tBbdR`PCsVD+KxbVBj9nPPD&MV8Cq`|cH(8fF2p0>V3wktBcQ zy5QDghD9GZE8|TO@3GmTpZsmwAA1-kvtHwj*k$}!tpd4cuPBuPpB2WFlx#{v<51(m+4$ zS(FtNEY!*X6AqT++?w;y_D@Iqhtc@btcM*o+p#<^7V(Dh7_cH4|NL7i>H87d8uS_d z)^6-a)hU+daGF(|FlCn`6fm#hKV#PK%a}$ELk=vFX}LYDqsruN9i(F|fgyI{1h!aWT=PqfI{`-qg23jHX z`(pUD&PVlGNtiA@M?Wi;Ak!rbNf%<#ni7PKAO29!JZW5M;IpKjZy53AB~~k3hG+Uz ztm*1UU-sz3Pw6oB%00yOnPK2Z0F06OBvcH3L_2!;@7+!UDK2A3rt+QO)oo9U7bU>g z_ahwS>{*SG88h;UMr!_P+|(P&{tR?u-#x7cl;ql?*bovDuWQNhaEz0=MVhQfr&%^CamZ-UOk^WnH z9J>Z9GeLDI6E}K{G=Vz;KJKIQVJ*Tn&vg2^Ie}RFe!6(Uk(9Gi1)EcjbT7^jhW>i+ zocx9w_p8#5tL}m&-yyxVzRI7`$FX&^ABG#=qaD(UwBrhczSsxU+x|jVumT(OF%Bk+ zR-$_NQM$iy7EU?Og&pVgttmK7Yev=x{&qYMdw4wpzqCQ!mY*SXnlRop0p9(jF<37O zCS8_b?|zX}j3yaS8XeEO@(}&p9>Eqa_j9~0MQH8Iq znXs1-@MF0bZIa9;%W%S#{o63lArkeXg_wBxD9uveM3eT5iKaRxiSqPriB8Iti@t0M z6g6CM5)~wH?jMUo?%P|CEY86RzE{(?j-f;2j_^G_pHp%s^pgqp*crZ%qS$`!Ns_=g z^8&ny;tUSskMMkR3dQGmpZ`-NT;;Xc8V74OwRAKav2#C!wZrkO-Gkm{7t!O_@ia2> zC=DpMN(n{HBz2}-2(em7R~zP1^qFYdXq-#f`hpI+QLim?;mb&Xwfn@gGgMe@2b6kLmQ}Y-%2UfMS2tQ=Qp!sxa!J z1H<0a9W7&-M_!G0R_%bA8IucGlYU1@EiCCvY9 zh7{kKXq&7+&!)XF4DiI~+R+GovJO`qL$OW$1Y-Z>V$-{|$TYo6*L2k(zoVIE50F9N z>>-#U|ALI#ZA6;)V?~x*qeRn_qeW*lLq)c0nP{?SHtpnHjO`P}5Ere7ga|nlpD3jv z-aiHFDleM4KZ3ryM^V24^GK#lg`6+;V>?q#*`g0Rtam$5Z1zahyxDgz-4eQ*B?1TbFji`;`#}*&Gr4{z%jK`ss9V$Rc{U zT7~w9_X(^BTwPbnwZF9?o!T6>40Q$+-13 z9jZ7-FN&{I$+d9WaBd{*D!1UPj}?+L_F}IkcSC&T)v`1#?Cjly$FlxdHGUS7!c|iw0d(OXFeFx zz{^RrXHgclb?4IK4N-KnK!FV6zSG`&`jBZFg98%{@YauKRs)|<8-I5!7#@S0`rJ1& z@faeOp2E2OQP}-d4?W(C@$RuJKD@ExXBU6$dguYGM^ET@iaYl@Ff8rhvFhJ;Sj^8u z2O~f8?9UI_J${Xc;`N~Yzo56_HNV%YB9Jq1j}PZPQ!zU@gwKIv_ZWO=&7tDI6Y-_- z3munBrHe+l$n54%8ZHvZo-rx5)*oTJrtr+;HoV9W#>}pLSnZ$-)0Azr*zE$z=Je6n{Sw$cOB^?L zWmBvFF%*@agv`Vc$nyQ;0+ZXgIq?ph(BYlLN=EBQ7`t#d!F}F z*L1{6Zvo@3Sa2q`7=pO#n7to@3$1gJVPg&BG<_5ukmTnr-d9xPzKAhfuvLrqWpXA! zW_cA=sGCE6;11ZyXQK3X7NUklLacc!c=Z)Co{dFo?OY@T&P3~~eTdSy4^R6(su(y1 zJxk8h>eDBwpF}0){`8^mE~`jM=`wZy>yN2ka_E*WrL-$UDOKi;u(xF!-Tmc8PO4=j z^Pz#Z=oQoGvY~X0p9AJe`%uuMLp0ecfo7_XB&&zw@SCg;S7#+mPZ|ot3uiE?um(>j z^wOeZX86`_38m{x;IqgZvjX*CxMc&~Ym6lu&*wDXyqq5Me?M;^XZ1;Qrsz@b&0a2x z#Y3mVW6EIoSc&7fvmp}ytU$}dwFqK<@L9GOQ4zYxU$Byvm}k<5|9b^AKhVWsN%Z+< z)6HQoh25Q*LXEBg?Q*xGGo`xZbSg+l4fo>t=i}IO@Cp(i9>v?PK!|_#$4-|CXpr|q zWAG%b-@F8@!U>&sf6%I!48%y@g29y+*socG=P||b{#T6Jbz-=2N(E2uOG19R6pp@- zME2BXx?FjNBBy?&aeOz?{!u{5HBFQny{B7;1w1kIM$YqfIJTwWe5oxiFNCH$Yh2V%m;lX2cI_jB(ldJ-H~uB zos1{OF36%mSob>uJGAZq?5~h}J>$$@%QzHedF>){6~zv7Fth_h2%e=Ijsew`Ba+0TsJ`%tA$uji}^I zt{Y9*^SPb|I}CtA+BbUlvzbh0^+$s0E1GocKJBg6#Fjis#N7*~c^QjUD;sg{ zB`2D5{=?oW&a8I9RMsQNv6B}4F{^kK(EEe)jRzrKRteE79@9ugDUeG5cY0mNcTpC8 z%{~Ub^=py$PDK>9&0Tb;caG@C!a$M6?0e=gs{cIc=a(Acfyjf_%!c`jj==9;zZ6Bt-{#m zW>h4Jv%5LcOpH6W#Yd-7nI>n>R1ZROiUv|IJAzp#^6rEtxtBzcdj4-JIjIDVHXZop zHPQNJZR9^Zj9=CHh};o{BlTvOSn!#)dat1I++%gxTO3A`;xN4Ui>gCpAs%p-w6F1b z0QW8?dv#H*;~U|hXSFb8_(&QzeLY?s)MP2khBBq$@9};h-!a_$EX*wk$BM{X$n5)t zb(ae9;!qlyosOf~UwN*IGY&q!`Hv!prwBeD4hfC-f}nRs=Cu7)F+5)OXES)#(Df2w)(cbedNr2*)4PP30!#M`W$irgwa|Mawm3;a&s+*u)-1)>Chnda;DNmP=6L?qRv6dm$RsvbvZ5(f z%s62zQ@Syg&q%NQpD{)SdnQBKbrTHF0Do`C(ynC>ghJnO(BGVg{vMeawZ{r>oCVM| z^Nk>0Y>Bbj*6f^l3fuhtDBI6jbCvt^$)G(K6P_5cUtXrn*=PVOY;D2czgH2oNa*(Lw8|6J&6DY~(|dB?wg?~HmU8y&G&rT5 zqMWl8%zWofrlF?7#!SqkHy6uk^yP~rF;b1yUH4@dm1*X55c;<+*RIr9&exNu;?MOOru{l3MSu&Q9O5wUXx)P>TLP! zqY$(Cv*p*ZEMb0h1v&89n19X_EScK}@b|M`r9Cuv#*)R9bS9ngkiBS|%j7!dLf7am z_2-?Ehy7ew^oQ+CZ18&=PK==-@0zmEpw@NmX`o|m>8h=I=nsPgAOL4F0l z7v&g2UUL@y`!E|b#_XZDqvXwgMN4B;;cpyo8O@Bulv&xrUA%J>K=%IUa5eB2BJX=+ z`0pk3NP^L{@^~_t)IrBzhv8jo5{6DWiw?6aN>u+YtQt@P&2(ip((VDyZgGe5ivh4v z(xTB1MxdQDh))Ftpv2AzWtNMmU&t$A@!+RqB>NvW@x7kjjvy?H9f6b6ehYS|auF^g z%fxlsp?@P9+KD4jKO>0l^I51MkXUrX8nlUN-ppol7J;m2D9`fzdrH=~w4nBVAj(67V4-*co4@cpEK`G) zN){=E+hL$Z6GFF&vj_h!L%n<`te05Of(9{s&Im!}q3yWWW)6E%2rcE_Q2U+VVG!ZL zZag(%i8Eti=^#bvo$6$;;TLJAHzHt%2U{2y!%i2+u`#=v(ARW{{zPivN82c*)Hc$E zIg)g0sgw|Ks+b=4Yoqp?=jiI_k+i{9hlcg+q_*C9RMeP9-k*C(YuiQor+1a~9%qsV z?{O%u4#T~p*|7P;d)c|#h&ZHA?{*{+JDH2?<9{Oin+99rV8U9gTM**rz`Ze5_|x$o z)z7|yKeI9R@IY+KJ50?pxfio29UHn2!)L%m?q-Oh%F#OzCZ@o)Mwl?gS3{VettqUJ zZdjr>o>mwiz(JV{cLe*n1B4A@=aIvtZj>@bCNai_&AesKHq1YTfB-l4XmK(tdhf$D zEbHJInnzj*Kh5r`SyFD_NnA@*VbY`Zn6cDkwz%X47M-(Z)y6?=sYN^H|GY&1J!lr@ zddt&T=MlI&FdlLt<><#b{OpuF>1rGivbNq~|D~#m?!8K7_H*-bWBpBQ0{h=-jIB7v4%17bI{eD}g<|F-} zI!qnU(B40x^ekx*tKWBr=`MTCGUCs$Gs@hJ?t7EkPdnqdVg>H6EqWkSX!#azQ6ZlU6+HH%-#O%ybY1xE{cp9Zpax;GXzBwY0t1gQSo6LANRm zH$P6r99Bg4o=V|S^&6`F<|jJ#I7Vb2t|Ri}ZsWLPjDJT%xf?STvn3C)4PhTyj*OIO zMpZVuY^G7~LxtMC zhHUVA3DG$-5*4V75~=8)XLjw~n3+|AOAfNE-GK9qCl5sGsZjD;bCh28yHAP#%wuDqwNw%w0|r%4G6)6Ren$}+`#jArr2U?im+4M zFO~0xufF_j?s*J1&f22#7$YMaV;`*Oi*vFhKbi1rhd}EzV30*@;!pbcXIGW?m`w7}nM5c6!k%=z#mn7&~U3lCh*rpF#( zBQ;;Lv8#VF#}Y|V)2v4Jee^DNx6J|uGo;uuYe{C~dY@;CE}-zoN%V_71nZME2s>GU zS$oA;Oz;mhORj>p>0Z1`ZNvDQPq@Ck3{u(I{QS_0ky~A$zdHx3U#22?5r6mfdx>e% zMOZ4GjEq-5QI#JJ!}eHY-C7U*Te6UJ;fz+lBD`%a!J5cqC?q`O{hv%M?wE?$iGIkt z9)_JcE|ALMUAX&`ASNlzwOb_D_S&GOdE9u@yV|a}A4M5Xk!m(QMhlMNDq` zXcoEh7zSUs0oOsXJoD0yJ0@o!QgVd)`dAFMFF<6gDmL>hYtpxMd_S-OmAbk0G@Uf%Jpxq>p^E^Kz_$NpMNv++Nr zS=*Kl3=DiqTLb~7tz)2l@-XK8*oW%SQ8>frIb9PT_%@QT4YSQ@Ad=^EA+x<>wP|Zv1HGDg4vn6B?#mBvH~>)bamPzGR+w?)MjAf z+})7>GYlWA?J)0A3hwQX!{fk%=;xL@s4wK z3qQU?RdGB2+vdHbsYkJi_lx~bZAT{O^Dnvj9>?nXF~?tdIHEfQMG<2$qHYpIxzjjD z$^l{Z3Yg?xi1txanceHftoxY*OTHPwte*I@%JbDo(|n5+s+XatTnB5h-&mrVg_l1c zaz}v*iw(7AY2C7{!eJ2e`aOy@)rMgHqA0xMXONIeGtLQJkJrOhu;ut-%=7Za72oY> z*E^5?_fp}xY5}|?9>8&r3|k*xfx9b9QOh%v6GreEP+T=5WN=@R#_qx(31BtJmTVny~NY8W#%Ou^PP?k052 zWm0#W*e&B0cI>*C=vCZ(=KDF2Df-;vt~FcM*=)#YQo7nFTGI5G5>muC+vPpITsi{z<~5YMy^);mFL1+E=EVUyFle^MTwW9<=!e#hJt-_~iG4OwR7m90N;(J0bvX?!?5=)JVdkYApI(jSylr37f1tDgSdf)_!ndYu|XY1S5f| z@tyAq#cQ~+{42HASE9YW0e(-;LvOV!YLa@$an3VZH&O!nk99FcuZP|>zNL8eJv3?8 zVPdKM_+6@pcek%nhxc*1Y#2tByMibpu$-FmYiN$-4T|0GOZi&{lFN!0ls%)5q{ly^ zm;dn@{OXa2@sFl)3w9x~>JB0wD>LthdhENhE=w4a4;33HHu7c|n`oNIp7kcOCGE?Y zPJJ)d*;%9L?r;Ib3L$Q)gPVX(0$UKP)DK9U(wB(!w|Gc3W|y$)Ts4s`PGR3B!~=f<((g1u}&lqDPMa)mCOUI;OB zp6Bp9h*rlDC|yua%MKr)e^r_EQ1u@DvFarM)+~BIe>fRk8w4MvF|aRpLC;El=4!OV z8|~+`(DF3AC+M=z=ZHBu4rg1UmDqWfh9?C|NKT!P8}T++s-=ZTgHkB7oqHp*M)31x z8ODw1z@GPFY}ufvm^C{bk*Q^nanfP3_hp%+WC2_rd&AYr2RVJ?&~)n*zd!B6sGiM8 ze>4~A$($mGI&;>iG;xtcSDs7aQ=-p)8)HR5W5FJFCebyvW z4x_Z_iS(cyH2dyN+SM@;U4Q1IVcBN9vWf%K{7vdNW|33We>P(>L(-UMMGunpkA?KhRS{Rj2P45z3V)nAJ9E8& zUyqJc=I%x^)3ry5wGDJ7&2iC)^9~O3{mCo|bk5t2TKjW|-&@FY8S^>kX&^d^PScRU zN>a@)r>Eh+C@;A`Vi)JodVij=HfckbRwY_ExB1bVW!Um)5@wt@0J*XJUe+N}z2E=0d`7cfrVZSoU=R`tO0|dc zfEOqpu$@_6(PIT}iR9@bjnS^$`}<-BzWL9G{V#i%jg{h^l9L#pbr)(AW!UI*-;i^& zhV#gq1&M#-XvmB;B;+ljAA6>eM%X8ztl$hW-|G~@cZ|<0u2Mrp0xA7DA>6sK4`m*l z<(9zDiJXH+^}CqDJ#98hb`VsX0znG7$a;DkW%k8zvW&+{|3nJ0kcIqzJkP=Bf+kyS zu*`+$7;|pZmzgr~y*La7#ytD=nD72`18z+t|Mc18e4_-OJ5PkS2hhJk9f1e>aSnPg)wN1tjofr(?>mm^ za`Vs;C<(8{=`^y?jS~0nq+icWXr+ZCx!ZLLo6gt?opphI@qN&8x`km`Uvb1onr-j-2+P}(;S#!$qPdHw zT=gvd$h}48`f;?~V4E-_+Z=&w&%^9yCZaCn<3xTKoD+lI5Y8BwLHJT* zgzw4{u;rP7%O#ztY-_@_ug?&+`XXoe{9{o+ndslG08!i3Y|*}7Wg;Xxi@w>uq0E|f z_*9Svh0s*^Jjy^y$tL*Q6eCDej@y?LP?!EBVIFH9> zJRe#og8Qzki1CzV89QXzYUi2;c_4ojlHz**iMIU3I7SS0OeYAa!qr2RPcfxNPI&@~EP3$0H$a~r~qk^n> ze(MxxwiTS-O;dNrktF{a8GJYX^z~tqYHFqT2M2NYO+FpA@<5Om;Z(i`EIG$1f_u)R zHWksKb6QyN?Id+`XM6356zbETPS!eKNx!p#=DhHuCpUtr^r{VHD%G$-HI-~Rx3YFX z8MRF1PLwt4VgAb%)9!4e6qj6@ElQ;0Cpdfg_6RbGE+cozbubUw4AaO}P)RhzP5ygr zDk#H;!L9hf=fZ1uCgbxPS6KZg126GrYt@X>bwpbaY!De8S}TflE#!CeyYxd?P3*!(ao>?^p8Cw?xha=}biu`P#MHB{hGt_b<`3AC-!fgDnu zuq)S%#-4deN-vw}hWrQmdzkaP_k7358JB6Q-vLU?TS^Vz9jLv=lFX!)pgByE_Ep)@ zwa4>mqV7P_OgunSj8u@Ym0`yCu`pZKO=b(flj(#WqK0Aky>T|42uV5eH8#CI!D?}~>)nBEwg z93iHk+A-dp z>@N=^SAJH0wzog+PrpMp?c8Jg=_46RRO8GRT{gYzAjb6^K(f?6NU!jR?mF(;G|z!{ z+cMPFC*p`Q?*_CU$JropHsSC$l(r7Q#h_`BuF!{muNgLIa09{l3uv(5_sAQPZ0EzP z*e&yqvnT`E%JmzVFi;ChZ~xKcCw;WuApnQ3OR!KrFBo*|HU+;Wbbsb^hozF}Y#N4u z_Ep$ST68VRi(U=gMZ%UGA+qoup#Z;Fw2S4N=V2Qi}lgbG}U6MAAnfFKdjcq!l^Palg+f^!u)d3#TTaJzXA0BqpQz z*lf5|1VX-N0|IiVVzur&&R6Nj3eNtdqcQPte$Hp;SBB!*@JRkVJdFUc6#l&prJPq; zM{|B%PW< z`bvf5u8|~}vqD(_s{8{mg zhAKR#PwVsX(oTn+myduwK4Z5F?-XszhFQQ7j-mWAm` zMOe#T(fR2gh~n%>+x8Z*3KPu9KL*(u#!y*ehF3G?aBtdp9Gg9gJ&ak)n&&nNaRVC# zQWc}Boo#~gL|r<&Gl%L+izro<^YG><;zsis`m_B#NqR{^Sxy=Y23tdO+i3jycZiI% zwxcFB8qTubNZOYL&n;o7uwM=F2}NXgoO`x*-=^;0z}?cX$xh@=aSElhHR=W#Eo&joU4MmQb6J`u=|XzNCrE{!LGy_;Ytrh4 zmN?&O2#E;K=|Q#eabk9<^l&ohoQ)Wk%-)HparQ{noelSlER5y5j2)+s;o6^2T<)g= z!=^HJc6}F<*Ue+oUfo~|59*0}o{K~u!avaTnt?dQyJ7J=3#m?37Ms&~h%t-1u${Ny zrjI#i@MmET?~pu=O-IZOaoptj$a-@X-cL8iyt45)Z`m6BKRiND3HFRPOKVR5N;QL%B z`ulVW9Gflh=&T!#pD)CyykN*SjwE|;dFV-);qyUr_;Hre%R+gSY}-c@HYbtEAU%5d zSz8F39w>xF6jFjkHZ}c`LT~?pz>X;-{VbBy)H3M(y%O5HhhgU+T2{sgZ#WIvt@*gWu7P%GU8c5@O>%6{p_*J@s=gCKk1y_}k_TDTGyN9bUig#7KjsYV!W&p# z@d^tzj%1ts?b-A`?)G&}A?uAPWE4CeD&vn}M9mqzt5e2F-if<9c^-E=7E@{RRY9z^ zR(R{8hx2CNh031GbRq2wZH|s6eGlH-bXBDxz1$5KJDt8?&mf%#$LPDEK$l!U2-%~W z==_HDkdNWqpGl!~{xTzD34PMCHKUSoFR3wG10$9eLFf7~X7=SgCf+P!jmfH_8bf`N zcAlzeiSJWpQnHjKiwz))1Fz`zb)J(5+KOPk$2{-%gu5yGC^nHh9c&HJy?+=E3c%2> zjpW69Uem>vLf&RE4&*xF!XiT6bnYk{=1hfq&d`RoaWp2A=WtbC@w4Ao=(Z0uA`gh6F>nVI+ny@w9->F};D+YPlBTg%rdPmR3xyGaJyYI#4W zuz)6xy-7#6Ou_7zoav%72v-s|VuacU6d1inAwSa{5TY=9BKItKEa7uu4}AL*i;2Ik zkeb44N*Vl^{`F4dPE+m|-~X3pTz*QI26Mh*>P_x>I!{X~N=a<)VUn8yJW!kr@4vDr z-jhMOLq1T1%Y9h9iogWDqugz{8@K%B89q;D4SL_vQpMeRXGEm@&YTjLPo)#n)#17{ z97{T#$@2CK$~TwA5!3ZpoR)^>vEv}#Fcz7GCYWz!j`h65Q|8fz8Chmb{hA!s{@sEh z0sZkVshTR+o8mU^^Ev)U$!k=QmVphP+aExZPB6~?@TEO!B@_^@3Jtd#EpJnwuY|KY({JD2BiZ%2@!M>RDKozF9$j&vkw0wo+MA)^;g zH0xp#*-kbQAbJE}>o)7em&f*B}T(xv$~a}t?{DzLjr7fg`p!^3KmK`Xl9cK4d;x#z2VW6 zIIELdN2}oHLjHgFRtLGCtq{wxqZ2V_>6q#p>YT2EoB##vRJ~6pCOJ^A@Rc0(y6N2N zuQYAgR&tblPP3n@QLN%g8j?1JqLX5Uql3>0s^eRPzSvMo+!ajWPV&@TtwAN%ElJ5f zfhI^VrX|e-@bq6B%@O^kS93lLqqpB#&-DBjph_;W~- zV%q));{`>!bco-TW2z~ehudJ`z3CJ9R<8tb_w&XY|i6vvbZWmVW*@p6jRQlon zTnK86gXVh$w9TChJ>eG}PjbK>-(cA2j=;o=3BvEYEkbOd7->}urn4^Mbji9wxOkdp zvKuzx>UKc3!v}Lu7_vJm)0vf$1*-i|VT{gs?gx2}dn@0eBeEL%IWx7_CLEns8?pbk zDtziAc{W>;HAZ}=A9n7DPqv^ZtF-avm>u-|b79))K?gJ1Nl~wqJbozA)8*}g^SlYP zM zEP!WPrXjJwi(ZFelKmJ(L7@_ZCoN8pQS=qzXX>@uSBkJjXpQ= zNww!PJi8PG8?7vy)hNQ`1@X8&RUE;Q!*F}eK%DVCSyAAtn z_!9O9(~$Pm1EbISz&-CX3E7xIEGrs8t%|ey@wH3)Q%P!XK?Gu30Ko zu+gQLju|e-{JDKJ<fI`Zm$F0o{Z9)I`B}0t zRt;x5-0^pFHSAwqK~-%ErePN9|F+S%(}lEB_9Ml6`a>(r6zTS+n4&)k7G1_@UUD9G z`F&V*=NDA6>o9MEG+ZYnL+Q|SyycAA?NQGmXvsqS*&s^vN#cIlg>2&UFxD=a$Mo)1 zu(Ghlti{7T;MDl!-;GlpcX9@CRkGcLQtjME%4U3C|D4-OE6^BG=WQe zHC@{Hlg_Mn#-2@)h_+9|^*SMhxR%qZ5@j6GN`;9~3vBWaqxnrL)rbePg?C=CVdr{T z=;cZlk*&@qJk8+mqW3h#CXzc)F2H)}cNkB;3|9M>dR-*B7cl}Orrm=5BMDaLHJcm^kQ#<5H-Wp?M6Fq3F_j-bw4nC7>Dh4-9e6YXEKTJ0f% z$~#}#dglnXtavBXbGZxeFk|kj?#y|%D|@&6BTAIk!Ylm zld)Sk0egMx5n+2D-b<4p`lJXc2Sm^q;0m#@Fy!-I_sgsMvCX}o{@G~ZSLPe+Uet@? zw0g9kPbu3fhE;w^R`9$xLYtP7{Vf!cj5TxawI!) zH-Y16L|b=3Y_AkMly1ia<@Ri?8UH&~igNCw9h2(W$LviPbGOw)94|Tr!N)YJ8XS)9 z4Y%mykK24U_JZ<$B+&eM(@=7&1?G_k%y)(jt9F~qLaNen;@20%rz~NEoT1Q{zLAZH z?d6=+Fl0}E0jssk+2!Xuna#vy?2xJ?%NPj2NavO8eohKoUmeKy%}{32=i+fKgL4<0 z(=oH83EMt@N6JkpHg|ms(tn-fUdVN9q`9me|MJm2do z#!PoT#&!1;G?d36_7~^MehI-Hqm2;etk#0UQ{=EN&-l^3YgG1QAzqxV#4e}r_~&#U z8qzkrA3Fj66jY#l!3Y^E&fw3kSY+^Cp!bv|?BUNqmXki89iG~bK?_$nn{W=Q#xYvO z_Z$`!$-QXh=!h)A-B=gSGcBb#|E6ON?|HpTO+(1BNYpPMfjsez#`2BfqLxi83*lsn=$n7ZnO@$L?OxvWKyYumc)ZFEqsQ8@1jgKHy2ZF z&TxL3DJz!U!}Mz}u!WM3*-WWDY}L^e=t&E)#>_#U)8V-`o%0yCCW(emKSS>I^RWGD z2A|=)#_kI4#cR6@fz>z+-FyhmHQ_jGIs?Z|&d{72KM6mr=y%#j+Bt3~2Hw2HT>d!nzvywAYj$Flt zRc5jnnG)7p8O4sYOR0Q_>8*+TMSWiApz@mp21^O z;yw4sJu*t8TeZTdTf?2x4S`s3Bo=PQzUXo3Cz;I2Ok!dna~;fJ2ZD~WEE7dGRN^tR zvNlkuS3ZRM|KQpL3$~#qjEOvPWHU!7;)BjSB(DlV|KVWBN<~rbs5!=V$J*%R~TVbq7!;wYwaduC8246Y4;9OlW*4${r zt#NOGMJFJ4WexqOeT63WAH;cyS2&Q^fz5q(m}aU@3C3X*`nQqfmiE!Zk+x(edfb@S z?Zvw5qnYVOu+XH<>|@t`>pxibjAHNGJFrhk8%QEd(Z0cZ$Ya=v zL-NdXfhzlYjQgd^g;6e%1($6%k$dF;`Ye_p>S-U{SvVO{vm^20(gWW68btYR&M0w7 zf~~v;Tzns)z-bI?<$NuX(>GvQ=M5oIb9^d#1)tkjQNS4y!6S7rI`Xx#y?&Cq8ZFqD8l@XoDJ$e^sB6YFMOdC~Se8@#&DV$&A!lZf-_lC=`Eu)X%)#_~2 zEc}GnB~9=t+J^O2!%=Ubgb{CgQ8wL)J)FLnS^Rs78OI%PKIkAcp2RQ_jUj@WTjT{5 z|NZ3omMqrrI-9v&OhMnkaJGD#2D9Ddz&4&*$2>d?nEnoD5{UG2FG3`inCC&QE*)KR z6L9&7DweJB!CcV;kQ%WHLjAvK?jL>Q6FQxk`En+^-QvdP{yEArm!D+;<5#d^i+Om& z`+?JkrEuQbe#l!&LCkMFH2t@sFtZ8!WL}`^C@-Y(b5FUUG8R1!5v;q}DcBYIR&YJ| zi6F-4vY<8Opup106J4Eou-Z@qfqXNb&R)fxjbm{}dnns-pE210HC9u1A6FA}(7ibs zfuneSZD}ihTpU2HzZCl~`X$EX9YcVgII~?jpFO(a!&1wm**cqPO!HbZ%6#Y3QTwm- zda)L!x&5HvO@Yg_%eA-*u#CzCur8|6q37qj4n-{Ma|q8UL^K~ z4it}py}1FBz4Y+Hb~di@yLex$1g?~h16Ijl$-U8dsW%U|WK|%aI!HQIyZB6OBX;Vm zW6``PR57iRG%Ewi?3@ltEgW!QofZ-+xL=_qi`MTCqpwd5Ft3#7oGsmPWvw`3w>!YT zJ^}}ho`y)*ZxktCKwe=aw2L>R=GJ{!dx@}~0STrbV9Ct388OCZl6DRGaMmwF^3+yz zrr*M(lP3{3%mII1y9#uiGX!^U6bL3wI4)RR9UzFfCMU2<;SA`sesb8Y4E;cDRP;_n z_{@6RS@Dphm-o}r1fCP&yp(XWwb*@993DsISjufLHZ0``o3&;O+cn9Asr0>m1+k$x4Jqundl3>JlJ#E~@(9j}-;5X|LKi?P27r0{HBt!T;6~kDQ<#62N!0&g< zu>IFe?!Wy&ckiT9SzRu5F8fHM7X70?>;c6##?pa=+jR4@JU&}XqDr2-LW5S}OSdI# zWa65zm9# zieXU85E}b|iulEFAzT1U%98A5iaERu^#EQCx9{bpCKZ z)8)}fD*jAq|76j3w3>9k)YFK`eH8mm8Jd4;sqtC?wJ*@d`*)`3PU1QE8eeR#3q;z? zZ8XwH1C#nDq5q@}1aGylN%1ko+8D6>usJMh_Y`($9B1)oEoK3s<5~aH ztq2d!L-OH&xN++Xp85xH&w>o1?Kszc&T4$q^TyB@+@q74fzq`+S2>gCJE9Yzp`3?7 z?M*1*To&Im++n!1l1`@c%)-w@kmk&4m5vy=-<}H3x%a8)bvq3RjpsQ6by$Cz0{Imq zV9fcr!}E1mxREhSluQVIN(=%#C94N%jw9+%Hsz&OMiCXTM~m$StiB`fF|+Hfb49p>%gd+ncN zkw0ky_G{?F^HLH*5{@G;_%H?*IUr)b0#?pd$3^2vxb%I-+?iMLHBpOwZ63+8Bd?Hx z<10GPGXY97IWw@SgVYmF5`Kiy-HCD3WFJZ?Cz9xld>Q>qN+XF7T`FK@WXWe)W@{&sU+kVx}$5DF?XYmM2?9quBx2DLj62c zY&nDA?sSCQS%sKsnRMl*BKngJ5Upf~lNs|6z3B@%w!EO7ykqD**BzO0a}cW;4somf z2y_zXoHr@DuzMzPFABZ245f=1lj*3CBwHRgmko(AXW#4=GMQBiSj8zVcFAr#j`f_u zH~&E#I(!#*4b!omGwaR03Sez4%zi&57RT?Pu{T~|a@GfQUCySBd3$isXC3Fw8e<1% zZ4TEBCK;O|($Ug}mYV>QpEWS=ha_ryGbqZL@3WOhu#^=aaBM;o+C%tma>4@m1oQdY z@Ei(J^&%SXLs~mbsAhr^o%hg2_027a-Q|eKjayNdVF)(E3OkRDMT^Q7o=x2kd%F!7 z9eWTdt6r1mb6<#YW~h|y4cs378rG|yBl~YE#?;=VicSTrkerMe>3aAhF&7atN1>-( zlKUI=F|o=N?ll{s>z9l--|S(<9bM6yA#nW{44c{d*ian;Z(q(NiQEs`ZNz^@IbC17c39YBu^{2a;qUl(Fp-y_FSampwJu z$nA>ka?w=wS;CB+C>+6RqBL=Qc^;ZNdA2cq{r~S=C=a+vg`*>|M^FG$`$TkZNQR+N zFhb9!z}7iI@NjdfVC>FXK}W$IfxpLdK}YspL0OLg1Gjdd^iLqNo*#iGzY``e4T5yY zA-w33XSVM&*r>1y^#42osvnMXhxz}#|9^-$9gbU9icne?gwL`@NVb28*c5RVCp3z^ z&=}6_>wA&v&T|^sX(W1yGg(Hhz>=$c_Er^x%JFNkYSj>WKV?0QYicEN$u~6NVlnxK zmeboWYp|r~Ds+NH*w_Ry);9hg{5%$5hORpP4M{?JP%UcHA`!oK1)8D?Y2BzB6vBit zyI2uRILF88Qvp>UoBxro3xNsg6j*)X@63)%RD z?abLcn1!7h&lLG@*7S2{v3f&1+(VC`Gbs|n;nuLs+9#M&a$k_`(I!~nc||boMYbSt zovGkYALrYToq#x=mwY~557Aa5adGiEs>~XoUN3W0FXi*R-i6pdbP;NJC;oA{1^cI) z!O}O^Gby)BCVR(*doUYO+B*Xqmz{&h>$_OZ=Vb>rCE(^K5zJstxOI9DoEQ4Tu)q~R zz0@%aSE#FEF6<8Ypf!FGQm4ki-)k`zRXwCfS$Apc>nG$OE{O>jI_S}7eox~Naf7KG|yjp)zsL95UWq+W1Cc^`K_O7Fq0 z4n7ktG)6Vwd!Lypht4Q7O#J2z`?&pxcoc+=uoY1L{DD-iZ^S(FV9XgB36JbB+*p4Q z4JV}VxbHZ|yq09bXC>I<3LW-&z8)L0;xc+)d_uX&WOjRSA(Pxm>@4SzeaYUA^r5Fn ze+1_|@$-MS&;opWGn8kOJZQLzIcx_GqmuI{j+&pqtD(tw9j1)JC`V+fq+($QpY49Q zjnJEU_<711t*^@7rIu8vgu#G5K2sMT*Fy_N4-Djd)B& zs!?<|>x1#~qVsh6s1!;Kx@c_vI|^^+xd7k8lpz|3(y6%+D=Wr+lR`|qa1v{sbg^*H z9|%$f?3x#`icNM*A<~_-oReiEeh1+8pOZ-Au1DKh;_OgI9k#tR#k^Pssp9qA4Kq3nq` zWIA4x)C)QME#u#5&huWsOCH6XgSYK?2@c6DWi!&8*qcmMM%-02I5`tp>iKBv?1g)Y z0rMO#&AHmVKUTGu^U12nM^qQD{uAKTT^sxn_C)QiS2UzS9%~0|F}7eM`gotC<$xx3 zY)z%|ei3|+w8Q4jYq6fc%Y#Lhz(X>Yo`qkBoBV6oj*w%@7QZ=z{yZZ1_haRNGLC-m zMVeYWeh(LC*A&#)K6hz$fr+D%rsMwfEwH!QgU34^p&=el68bZ6b4U)-qR(+hdH{S@ z&c?T{dRjeJg5Rwf`7S#~<4aY^KzLyIXmYB2Yd{a8HyG(Jzdgl_ZG*gh;4sY^NQ@SHbYmij|Goo%?U zy@zV{NucP{Q1rKO@0G%3{OPSh$T=ytdf6LDjtRoqB?~dyhVLe7WYKfy1101~(NT*# z#x^cdlze{yM!XBdVT)MqeAxy+VIiK#R5LN7Ae8D0~6TSk}+&V zhzOI*YsSNtK>S&ihgyX(EYxij)B46|$VPl-CbAnRCZ#dc;dhy+bqkwem(Co*Rx|ZQ z^6Wr2?{a>YVq5#wn8jLRi!9Aq;NSpWyWORfPf^g|cjn@-6OcL?&2v+dSa4d6I~_J* z;?xSJP688(d-W2H4JY}UY7_Nt>FHF2d#p7o0IdM2Xfn-+d-nvK^hGLiq# z99dt?$+WVXS&F&-jY*%uyDW{=%vypIi8N_e!Y}wX7q&XvrJI+Vq zaRK+}#LULuf5!MHDu%kpj%m5sdxc+l!txsT=y?66sIASU&iCb7-^OBPlLXeR_)dGp zy=c`XMZwp$K!F1V2_n~Q6>NTPEx2)4Qjjn#hV10|4rbm>`j!8k2IG=xi0whnpp=32 z^F?@Ok%2P)d~M@*g{jB6OQ*CR6RZu{EICIunV-|;xd%ce^#S+M9HIQ@?`U_`LKJT1 zd$V>m4?GnUHIm5rN5olV^IS}a%#$F#pNK5axanghg&i)5!jCR}JWE}Fgz$y)i3YW&q4dBwm^eMA=AkPP!@HJ+ zsc&IX@(Oc(l~`V*5X&=|#^Sv#fzx&&=zJN5 zi2T#Dvycdd0(eggY;A?sXQ-( z0)leMVSGG^4NNBQ-4^KSJdDwSmvH94+gM~+gC}Wwv7`7943yttYH=5&^xs3Y^A6IS z&S86O1g_fMgz3l|Sdv@~m+lhSeQ}0T^-xUcT!;_y+~?Gt$(@-25UbV4C-3{%T&Bn7 zIB2j6cR4mH^D7DsGtt}EOGnQ*U}%yiJT=DRTgfOi&upXxZ(8Vh?IgHpZNRnNhtbls z6Z!L`;lh6>6|QGeng2(UT3t=gw?@;awmqb$Q%`Y1GvH~)&$G=(AO}yR`-x+7Od^KA z>qKhkQ|{xs42$Olu=}O&Vv@z(>A@%=^0gqd9C;kpJnJH&{TIVk{$M-b?Vrw%r)&3nX-{|` z{p!0nIcX$kQ{B3|WN_#+smeEyMBZ;>hY$@aN-(EKl{0A3ay8mF zM~9p}(`m`IW||$+Neg>Lkg~p+EMGa%o!^?6Dmf9~!>8i%U>t-j_+3VUa|UbQkerwl z3fcvDscm~CAKW-tM2GU{b<%SokvS>`IBky&3XnndC^M&IvO?4AA<45C~AJ6OR zFQykxKIG=&N9TTdlBk0TE%}m$U0eBH>CA1kyNzdN!MbeMTFzfJkV1`E5WI4(BJ4^W zgn7?WwRy|MAG(I+v$X z?a1roe5{1$nB7QZX)(=l=Uu9*?KsQ3e3kZgm|eoVh!5sMaOF6Lt2LluVilJ1j^8jW zLi|h{JdQ8Gg{M{U7HfpNZUQFropaImF+3AE9&%nz_*-ue#o{IKoBNj1hR@*Nah^HX z4aKTo*N{B21Opn}7w)SJ=bH!dYtl&!-LV_RdzWLW?p(NB{6|+Oufh;McYmm{4~O4R z=iQ0vXx%!Ab%;eUdD%GT_QHcb3$tV&`p2@Y;5eSisD;PLTc}#rgU&zF%(m?@ER9s4 z$sHnd;B- zOqr#c4ogzyJAtSX$nA4M)B8E7Juw9`MT&SmK^5zrl`yN!8Z)%*p`x~n&p;!fzt4$t zEc@A>Ix~Uyd4EC7_Xxour)0s*W$uE5GM%J4R+sZK!m-OT2Z6yG5o&LR6RiR~wYh}F zKQF;{)d>_Oa`$IM2=1lGWAqw(ynf~mmF{@#%kYNU(FyowYXh-YsrbI^0xSzIBV4fp z<9@|p<*|O+-L?d~hOa~hOw_xuOZ_%zn4r9Lv&;iUiyW@#b*=z=R{(9V;J9oYS4*ANpxv%6H((y zx+0WH;w!h3NVyPZb{~giWFPixj$l`xW`S`o?Nmua*xLExhtO$kH7vttaVPAE8-sL_ zMVPg=2%Xjp)`jv&Oq7J&ly>?(J{oUr%b=(0&$+X%7;88cN^&3Q<&r<-q?>{%IkIf8 z_-Gb;>IQDiPl4=eo>@Gn$@Xj+fEmxQzTbQb7elO(>%!OzR?E-WB7$LM*I507_3V7B zJln_5@c#{8!XBKqVHsjwn7-mR?96sSBJimo*06skTA?nMjKe!D z@o}yQ!tHk9FXug3J37Fc@B9)AyQ$D3jU@9V(6rN?=l3{F-HX>g<}M2 zB^v}D7$NY`J&DBG)7fZs?oh7^V}1dXnB0lm(3U-lF9XF`vHu=s=9EE&vsVn~FUC#3 z@l37Ki`C`rVK$i?SjS6ww$be*cTehLYC$d1)umXVU>YmwZ%4d1@0+ZSVaH{bv%#3_ zcp=N@-@Fg2x{R~1Zg^v=<4s6%{z&xZ7M`n*gtv1D_G;V0V|*6YCY8YTX(?jq7~TuJ zqam9A7H_v^lb^V;BjN{H=`%}aRWHt_C%(g&ty;|JiVW*rD#xAP9sJJMfL1jjY+M?E z>vJREQFQ=fQ|F>yY8bR$8sY2Se7sU?#L9r9aMIfg&kj+z{U^=#y`RDYhVp-YA(wmHl)2CA^k&Z$~V`o0~ibY0^NS z@*1e<(L|x9E_nRKd?T| zhjEDpJ2v?$tKRC*oUh8UcXc;lx6X-;*O|yP7M;YXle@T=Y7#W#hQsn~ARhh|Vztep z%q8kN+NB(!>-rOGvQ*gLmdD&Ba~%5(i*NUw&)#lqmC;ZZR&*QA`(iO} zWe44!HUlE+{_xdw1YH~khbb57a)KUmf+`;2_Bd; zk^Amn%|X-}Kdda{^GTWMFgdP=wzEQ5cljiQzkI`&Wo>+JS&o&{vruB|hvnO~khl99 z>W_|Pq66kE%f^AF+L$vPVGaEFlmg4SfiT%K6{?B+ZuRRnvF#sdNyTJ-UKWN9&&K^p z*g={3q2#mgBK0VA(dyL!v|1&FjxMgE7fL6{JnI5QPE~{T`Fuq|OpD{8GjlG(Sna_YGxfw9+%xgXGeihDp7q?0Rk% z+nais{SL}x-d(epo3l2nU3!#V-?)NBnf!#XR~A!x^r#W z);Yq=UwHy^ImrL}+;pKUXr|%Lz2w^Wi1L4o!-CrfarfOmES=5s*(qZnBq53mV(BdO z=~?!Sa}T$QY=&^M4^FSzfe#Ces3dq2t+3cjLazE$bZfvkC%D{reZwB?i)et_z&$*i zor5sW`q~-f1)aH?*y9t9l&QNRr>~1pdCoBC?<0+l`79$no~_=0m`(dPmA&%$j@Lbj zczJz21pE&9SXCRebD$Yg+(SAvK`^0*^-WW{C?L2@jIF< zC+;__0Wh`ATlj;1s%+4>2X8x@U*mZenwku^GCY?BgOFqx>y5t>yhRSiYA{Z-5 zc_*puIm&MIK+!b=zyI1JEm@ZTmaS*1{pT2EJZ5)va+y@WKf7DGm}LgavJ?vg_N(|N z_9O?vxJ(cPfxl= z%89l(b$T%><{d|t!UpJ7$WV;76t3;GL)ZGXa9z6tF1!mYl@&&1kt6sSViZ=3`;ghF zug3m&zZt)d8wJnTC-7xh89WTVVWuz)sxh)uZYqkma}J{?un?2=ZeqRDag=oaqcpSs zP-ESDn3`Tcj0^iZ?-+03~_wpPxo3_K{yAsnYTF&kiS+j;4VJN-VPKN*7dH#MY z?(*}yl$|+V{~nQfVwkaE&gsT0$f~7#`n`phNCH zoqn*CB&!3VA>0Yyck;||g%TTe@GC~=TVdU8LpJe44Kq}#XBNHoEM&rdNKH*4)sa6) zJ>($bwT|&j?m29EnSi$NVfbXM4rj}FOjc~duiqjpx4jjaS^4C7r-WKJj7LGvd=w@u z2X4Kl(Sjmlk6SOO^_>8NaXtttJc20hU@36}DGhO>Zxaaj78qmu(=cq3IECn06Ogki z6kq&?Gb_2_EPvPlXHUL`%Gk4*eOHmajf-NUf2vucLuSlH^l z#4Mg~3b;6u8Fnw@*(6tq`$=$2 zgskgu*3Ng6gYkc1d)ffj=}FXfk@tTyO|awC0u=9NIP_u&w%p>blR+!i)Z)zcnkBG> zrLL^k`UiBEmvVNqfR$%BF`s)wS!uO5rgLV^*@;Ftbt4sL{j1;~UWQ(SbZpdAN5#+t z(mJJv$@BluhxfqKXlX2R8c92TilcTg3)gbbV+|V7Wt0kyuzh&-xDFpLII+{l&g_ec zAv=5E2j03xqT^Q)ByJe8W6Px2{8JYYUAYDx{4wlJs2_W6y}X#tN*=2-0wjAeQE_4ZBg!RiJ}&s3BNma4AeE3;?5`?6!v?Q(|`)< z%A#<3#Z5HG*5mrdC0Lm=mI@`}DNgx4P2y~*00m2A$Cz@y(E%DM-^t%I1{l9Gfb+OR zaqwF-oL&CWy(tEua6h;ymLd8<1FlXh!JNI^{d{sg@0x0{XKGVfr%fPhI-1H%u7jm< zmdU!gRfsy;ffV;n6z7e=t3A0!H5-*^of>B|OkBVn_Bptc_7HRG{L%B}HyP};g_@Q( zY-;Oif7C6~&09nQD@{~e$3e%e9H;#6AY$h+Ts|qny*webXr2bPZsXmyCp&PA&wOWJ zUP7_OqnVER9#;H(K3f+055XPc9>^VM{mZqge!P|xf_plIG0-QGe!ta#vZ+UG{QR=8-@kJ|7R$y z9tM!r(gURYy%>(W3|Wnd0~;}9EVJLjnKSRxj9=)Tr@L)|7?}SPqH0PkbLukY>7~yK zbDBw`Eea;de22%M-(l*eMx%b@8cX=jqze&Iw4*tmJf+f%7wjBIZznFJ#K~$Dnd?uw z^p>3Fw$OQ>m9*LQf$_kwd@@=)35!+^dU5jIcT9U= z$i{Tnivo*Td6Mdp`r+OHHtR44=clT|=GcuF_4j$K?E01=X!$7`Hx+Obv&F z_+7fwG!eN8_wgjj7r`eS>H32)2pR~+%KwU?uU!Du$s6#=?>`FBYb1-EJfFFt3ZKdc z@M7j0SglOJ$aUN$;v$ZlLq%cWrGQnhKGTux#iV`54)Pyou{}Sn*gV5Tb|X2S{ivPA zUaPnuxV{w?`O<9Nv}WX-JVP}#?~RkA{*igW3tVzEVA+dxSbnDsb2=!-&Skp8drvK9 zN0y_jK^m!iZm?yei!sJD&}189l$vlB@~GpGEgZ=?+sBRmnXSY_*D4(EzsEh=v+=~I zfC62+jkn)0B{AMj9jf+=X4YAfgW@5QNEI=bQ0J_~?0mqy8r2&rk<9OKr)I~}B>QFX zj{F0~!IucvDTQRUGWz-?jg>RCG5S|KKKvQPju#?qX!|YLdfB3lbFMEqRA9a4Wz3B$ zgzJepP=DQF?7iFr@8dpUWIA_QE6Fhzd1YoYr4coK8c5K;2ak?p$jV&|@ye&vRDFs} zA9~`^i|jX63ass zNXxU=_+)D!!W&hY-cfC4-RQ-pF9>H+xhm|qydxUVai+@CI9y!#kyf`SQ|*7l=xCw@ zd`?cs@&H+$S(}Dm8NN8n^LQsoS&(j9EEwH-S>PgCD$w4UDyV#HE%n7FJ%|LU)DJ zXin)XdLPT(w!Tv#dP5(L$=e`Zo5Qx+=p0mVMtJ1G^T>Yr88?K};M3xd zwT}}qAbc0KsrC4}o6pqvcgSsFIR<9(&UC&j&K{bF`@vy2Jv$X$S^MzZBSK(tu|aS* zvsutM;ksbYxIBSc{7OMUX*a#|bb`$0^-$P!7(#jOaQx36ZT`U|vHTu2H-4e+OM`T( z@CW7P?5ACmZCO{sTGq5Elv(ZfX75#|vpXl%SRc_JUn+eGZ>1=d{i~!2td@M-Pg4NTczk}*Mv*L%7Q5wA)Z=03_MC>@ z+2b%ze*}DTHqjkLO)O8Hi97PLFtD}6@t|XvbZs@1%cC)S>@zq$t;HhMaV$tlhsBN& zVh^VtW&-m@reM*|WZ%7E=1x~xpWJSC^9IlF%-i;-h4=OTJJ@SBXXTWCN*6BoanrSfg7DE@OhrHqj0J>xp^j zK|I@U4u8pAs5{+^JyY`FCz=SoPuzW%r3xCd7jHUvzw?qGlo|{{W4@D(iXW|hbb@BP zm(b?)$E0FVLAGluNJMi3#6v^41DVgS+izn}eIC|cGQjfl9=ISHjc7i99HzS!F$p{H zfaf_&3$;joUI=-o<&u>~KPgQX#-GXzvN&vrQoW@puG2*0w>2o7y$>{d3nmCTkYIis zjkG^S6ZSu*HQz*WY~~5FH4Y$eiFA@MNh0UKS9G+2bC$G>Y0ce9_@urJz7c+C*|G~= zx!WMTA#o1+1vD^sxljvSs&wa2&}8?pM?R$M1#i0jBB=9Mm*&Z^>@Bg=@LCetuuy;e>+vVa5km&7#8BE%I1iQGCiZucrfA} zRK7Q1a-JMB`!tnR?w`n1?O)?Z&K10yHwKS4tm8bV^;ngd4E2s|%-S&mb(1*DwTIv1 zj}-83Z8?0(58~pRUfTIg4mZCTqd^T=AIZO8it9LU&;m!CR%2d$6az@x|s_nv5Cu~!V`zc^28-ZoP+-${D3OT+0}7tMSUPYTstbiq#;#3SD2E@hLvbO=bcC7D(L_i03;Z`59&}^gR~B^-KvrFFm5}nfIyI%mCdY zn)n*@iSJW_apzYQs>W1fY1SKr>(`;HKMF6MIbuajC&`3s!*msQO?oC`{JI^OlxKu_x;jYm;Y_NOGf>vJ#(zVH zu&UKBv4G!KRQ{WUh@X*Yw@gDrb_T2uTESNO0fnk6@c!jAlm(B0wX78yHrU{k*GQ<` z_dwI*Sajk5JeD28bMNg4lAVvM-kRuk3f;P$8K5H*;HnHRRhK7Jm) zYfB>hy8AMZCQrPdv8MPNgnowuYwMJ#l@MTEX7Bfjav8yb~3qGvh_Ts zCHta6DH`+npV{1e8f9YjC}M|j((5@~YkN_k*N4&FQ`zCs>g??#J`WsTNB+~av2WK* zNKO$($O%ihF4_d~m65b*OAk3m@^|SRIV8ndVT-yP_e!{2Nt9-|2MRqS!14x?!Y>kPlOlnZE>bQZZ0#3Wr1*pS34R zV~MX9wmgdG&Zn)sQ%hJI`b((yPo z(*#m6Dv)-6Nxq+s2pT=#3dD?`3x4Ka7p&6D6qLGZ3%(tYp}4-yWVikcJ($x*vg0FY zV5tEO7+t4{t5qR*t&23y1Crmf6n<9XQ2Wh)H_O(t_RVt8>?xE;6 z0qDQbfELB!Y(_*sOw(nV^fYyr{%?|?@Hj#p?y#KxupgDAy_7Z$N=fB8 zt@FJf31yUo$cQo`KO;M{($dmYX_wM8+nw)yBC80Y2q~iMLS%cN_wV<=_gvTW@42pX zUFUnA`!nv(XJDUgr!%oVI^%}TpZdkj>~G1;vl&%Pmc}LKqQxo3UNDzw6=l7hE%0R& zMVim~mm+_psy$QiB1Q0O$1t;RyoJqjLp_@rnqzF%>Gv~-^7ELr`dUoHml;gjPhTd| zE`donAkvwxS~6|6X9VAGihKiGteJIQeS(4Umjq>^?slE2xdJQy9>Mt03XDeH0;c)t zWF{fvhG2a0Qii}b#?@&blYSxp66iITYqD@)s#F+5MW#Gs!AJ_`9Wy*2E%4$>j^vbM&pq zhfvI1jC5p-*mF!sfVNHkEqR*>qi!%mDw)i!SL2zK5w(Jaq71N{k1W%8Q;gYW8YYO- zn=QC=L4zsPT)~+AT*B_Db^Qqv{tkr_Llg0wI1&)G`*5?E- z?l%fz=Ee!0&3QX5M|6*+BJY>EcdKCd*Bim-)SrS~%{v4ZntmdDT+ED^d`aYCy1}f< zOk-|~awdM?5bg7JJz>PXA2CjT;x>;BN7#I7u4H&2VaAK*Y*{9Xxg_$&pREx2v<-3u zb>7zm&2{38`~+D><&GI69%sjtr%5tvl=Se}V3*5N3jFuR!8#wZNtBy&&Y9B~yNRJhNe`8WV41z!;A|#E{%UCe2pT z#(L&$W=vZq;~>JUrO$JjW>H6yy9lR!b?j!U*z2OXdpdLRa*v>;P>o64Kb=`-IG16M zNi&0{MFKS`17_k#H)gTiab`sDE@tTiE#{hnG_&@>Q$ghFNsL%Q5Od8GuBb%%+%#`82-~)L1MiavwV#k<0#U}_N4w|yt{2}jE3Zh-tiXNf5lVSJ3)KltDIR%52wF zWFjMb1ZPLiVvI62iF#c`esz(bLSF0;)4f%WxxMb9Ao57RV5F=h;~|Xq_ixK2*f0s0A@LnMMq!m?%&)s$~9p-D1AX z{=qbAiaKR`M%#E;d}bW?WDB;6JOmFVCNl0fs{~D{Zi2}1(gN3#GEq@pw&0JcHN%VY zogNy!5@^Rp2zvTOyA!aTkyQ$1B0gI(TjsY3UUVK65QlO>Q^$3I#jc}*mieiIF#~%9 zpI_acw&2$v!N+!M#_ZTMX1czpGY-`ld*OA#<6Evwz@xp4G9&8lwlQL2hiw>x_2&iF z%4-<~d4d!S-tzg+ey`ZdRG!t7qlgTU$ zW{TJCU^e>|FXIuK$QQeVi#9!z9bcFY&;O;iiiMf*a^&5uMoEH{{g znOhhW5jR#jMw@Yq6?KWtK4HUX=h)m*6xvkJ_{=ozN@f;TXE7!FDi|~8hoa7Jd7GJ5 ziZ)wTiu{b-uLYj_9hl33bC_dtYngjz5}9$8OBstf{{+vy$1<0rmotOk4>Q^U%Nd2! z;ez3HznG&>jcoRh(zR)ko@}!;%Fssb%okC&`7Gv_PA0QD_AnFe;K2m{5p~HH#0#$d z&K2z*bub0Rjm+xpsmxlX6O7vWFi|(dG20Jo9@`%O_{DZ|(qG$$Q>E?ntFGC`WKLmn zlYN-QqW$5!&B09cf?#IjECXhYO%~HqsVK^yGPUv9^_sC*8OF3e&}V|KIWXz2;mpX+ zFlN1|gDg{OGZT7NhbfvX>O0eU#%R8iwJ8A!8=C=+$%x;~xX+Us9Y3UXMr_$gvH$(K z&l8sr6B|8R?EhS2#aJ<)9cwpk+OytbozE^G3rj1TsW!IOw$olc3L5ua)#XSbtJ!h&0%9E5t3Q$upS42HPC^GJpY?)!8D*q7QTOq^!Y>Mt;43OVici>PE!aE zc07{eu(kSdusWYWQv^b?bha!+g6JB1U4$0hZJ4x^+T3_su24=1@^>cL9W>(T-kCO7Ui^) z*Sn{nFfo~~U)GHVjdD0Caw186x*sMjQKC|kS>VyJ5A$D{fLQbc&ggzRE%SWA&JqS= zLBW1hjbsS>*%^###)v-Qli|(7AlBC)6LXjQ;K6RfF%GXt!?;8;xl5ir(eis`}xVZ8IJsI|uxQ>n@R!hoB^OS0bOyxY_ex3n;muP&y zV;xwz8-Pev#oI+?wBoV_>YdBS4XPVqy~10%y7CFv*K!!_+)AM5{T%$BuoZW z0=t^uDCTn<;lnb_Zkvowyamqoii3U2Poc}f1rV+JnENllk4x`Xffpe_a{aGUS-E}O z$P2Z^tvrMJb~wP(D^=t_wTZa6;R2ET843+sKB7kY1)RN2o0nCZgaI1naG_cYXJ0GB zU3&q)na{=2nbG(|qX>&u6TH7?3SPAu4Poq7XtQDPAB1rGq+5uO*=`KFFdu_|TjIz` zmr3@TN5u1{J9L`b;NgdzWWvr*)!6F$&P)qKj@pZ^_9+hj4MhSkzwcLM%oZ2^*$63neGG3e)ya!xP=nxb1-*ef6mi z?cU`OE#IY7RrD^m9*#u))EGKu*BvzX5W}_9hW?uOA6`FUn!NiYj(>iDjQMbr zy#APu_s%6@s(2BS{TETKS`pZd#`Jx%B&KF9{Ln4Q)m7Y!%E zR;wU(+$m?0t6IXz7#5-2mj%L|n(Kr)XC?_t4jiEtOUh|}&mGqMqa5#a|~{AR=B2;$`xmzgVj{@5!sdMW+sxx9%U*u#fjXHoI=;zEXU1K`LNqwO(@U2 z1Y7OL4r+FPiMZQTm^}F_U6*V~JENboJM<^v0$s9Zq zPILc~8;HJK5F7a3y2*7Aa#=p?Hd_^aOFt z3N7K@5sQU6txCcRS*|o;_BDFvtsiyU6HemOP?*nouLXyZ0F&b>2GP3*chC#Jep3%@x0^)eg3o6V*Io#gQPDBA*W9C zG%T~?+3hLENeGI8U#}nCC1XLJJ?m@8t0{Nz|71+!+g5XCP1nh?J`L9C(^EFaa~I00 zNub^h1N^%~j!b{2$?0r+L$XgR3Yqach2NIB2zzgyrVqNN(A%>PQE+IX5_@&+~;!4>Va<31=E(RtyrG(Gnb`C}D9R*#p* z4W8>8#%@)>xT1?RChQeyJ}4v=#&Y0oxQ&zlcY@OhGNL+t-9)fai7%Fsp#3c}*!8go4_{W}SC@|AHLq&((Ld$*x>@l!B*yZm(p-3%C2#OhYYO%~ ztDwy<9Y}`tV&K#Bh}8wdhH)qS9Gw4SssEo3)cM0?csE&x9txGEE6yD!|ICcgV9IB5 z?ba@O{)R65@5jT2gUYk%>`$^NkYW+rV#(EdMc7cIPX|jz3fpXKg}*Dbg$hO!_}DfT zey#8|b}kFxG>0>|jFWye^zeP|z^;|ha(5<-D6S>_SFdmbD`yioem#Bi59qbbzqBdp zB4+8y@I5uh(a^iIp7zwAe%3h@7jm>$g50yd(|>wd;uH4sm$r@|$y9FUt#` zU&fnycW}yxugP)sXB5Q#yliSHfF(u+W zpg^+!1wpcG5}EbiNu0C!6p?q7!Ru>hQ z!;H}EY#Hfju_EVeY{|p9Rk(k2D%|tkNfkeyL$zPF?0hW^vUf!oxqbdSMt{<>|L1E9 zdB;P+@6&PkrmFz$rg`N4nKl$ZAPL6b+F<~W(W6%upf?}P$|*(T4*wje8(s*(>pbba zsUNAP>nIqmUVwf39C6sD9j5MSri~*Xa7#u^XjqffL}JWb(Y99}v>QsOcJNlZG*BJd z2367Gume`tWPse8N91efQJ5m#4`1ySvC-u*zOea$ak^POJ@^XJ4S{SJZ^nRd;d5!W#TxxD4ox7-*Q-iF+p; zvA3~IK%e$U7UyXj(%nj%bJ^_l9vu1Bqskr>%F9k@yUv~b%3 zyqLcgZ;uK^535+3;oyK9z8|BrvU6cmycKzW&Kea%45+)?Trj$q3ASSo!tE(zgvqvE zFukpx-s8O-S`3isYucnIDrql6M}0@h8>n!zMx5?CW6xOt7O-C3;a(;0;`l& z8^}JS6|n%L9~F}yckA)w{K+^=!jY?RH^#+sLV#UT__O8`mA6Udj7n!y-3k>@{bfU5 zd-c$v%3|_%zmhQLxtP%IDW%WSMNvG}&j<>a;?RrTprH^0O^?sv-mI`R^l3 z0*B$-`H8Gy#B8v%h@^sT(cI&E@=&Id08ZiU#QpGSjLA^Mi*IFwa`tuTu_X*&sb+%F z$^%5du^VYY5;(fuq-HUvQBn0Cc_v>$9G)D-bGa$_beblPPj(|`Js)Ama!vkHMm|-~ zNJC6cBCRpn!sp)ZylRvecj=v)P%%Xb`wj3pIb6FQ4cbAyd*_(d<_|= z>j<9(-C(X`MYAE1-fPQcSN-WGYo?yZ^h2-Nfw(VpoBl{Xx!fF^Bbw03PLh}Z6ha?; zE})~@vT@r5N;V!<=VR*P!FStVFpM7!pZ6*8vzv@Dfq(4aZd6Ccu8qX15v33vm5V$7 ztVPeghWtMt7s68sq3-Su5Ez}u=oRz%RVAm<&Nq!l?UNUN{5f7Y%Q_NXCe+awuM2QG z!yNVt9*|pqWWdBH2;U9Y;Qqzw=-cta;rcCG^!l+24i{a8yoWc*fsv{_Z=lUzuMkJa zY6;$`W+JaP@X6up%x6TwU6S|QL5SD|CEnfP6fQKL!#}b4#~ljFp$jR%>E0Xg(^MIo z4-`RQmlF71X|DHCk^>DJX})AD%eTM&?y#xGUg$sd8>rX?!|C@=NpYSG1W8DP+viT4 z@WBt0x1?j|$E1eG!#30`_cpic!5S>M5RH@WpK};lSw!|8JVp%luX5woy+VyS%Xo+8 ziM;0@Loz<+OoQovLtwqo31Zr2;qTvR_*AhFf81QaZ#7g&~-3%i`?I1ZsCLh1E)@DR-%yuKsnLj=noUO|Shx!`{8* z`oIVJ)dS#~k3GzDc!(`=s{H7}VlwqnBMc=h1+C(FP_kwmRI?@c;>iK-){FsGtJIBu z^vsYa=R)|L+o{~+s?kuKcLufd4f!$iR588Yh#$539{dy9aq?d)z;E;zyuR2HznoAd z0oyC+`aVi6TB~t<-$f`I?S|6IAF0Y5d47WV23%mi8~2B`G%Q-70_gI9lu@nyUnuNdggPd{gj zoBHq3m0T~W{!)qsKKi`GB@5DeXade3IY`^TK7u&SY|P*6#&3ABnr|~7PcI*SM*27D zg7E=){N<$1Nj*si!{0gdZ^sETFwjP0$5&CG!YIxmPzh%kNekqU#{2&eofnA* zIL9m;QwP-f97B6}VXG({^}H8sqBG&$8ds3(Ttor}wS~8*wvf(~n_zC_XVhBy9_A{q zBYjW2$-R9~L1&RQd7a9av*H^+-|Ffh->H(}j^pC8o3Bv_8 z^ikK?fGvM1l8uFRLWnF*`jtb{`OG$`uU zfmJ%2QDu53UfZ%7EB3B~burT5weRL(0H7d>JjxCiF>NUkc@VyFn zIYk0uq+=oTi!{`gOG5u8eLQ-t1`O3gNPdS41Z}=We{bl)x8EH25_b!J?9ni&(};)8 z$ROI9r3Cc}im0+i8oM*t1~K_`{O!|6L;WW5ZjLkgAk8&we!Vgp8_yuex~1UNr+kM+ zRZ~$@ehdctWZ<90;k4;>45&{SLOX|I9Nn%?tv=f0sq{>IyVM=#hO1Hw!(jBy-=z^L6T z>E?i1JiVu#-Qc<&R1Y(-SZykZ(WkWQS_+jiS`BjIp=fP(hMT^o4*WM}K;jb_&=~Jc zOfG(av{M=ov|O}MAbk@oOU}ccwVyFMb{RAd%LyAt_~0|s4BF8+1)iBSqGiu_IFp+I zf`UfWJU&EC#yp3F+Fl$T_L;^w4nyJAk&w}R694*Uus(~H;5dU_xbBWV{>}P>iZMm- zefLQ)?4CkqEiNN}RlmZ&qG$D+#jk+!!;@r7*%y4Du^$WD|G={m9#nls3RKQt3CC-! zdFiedfe_I@IBb;koq%_Xn@z9WL-VLDg; zBF$T8N;0hK@YF3Ku^wK)&KQx7K6_r#8RwbOrl$ z=44cPlt3CgYG5s+3=enp;$CY-)~n?sX6)~vWwyYCHNrZ7s*d*z_p5i`?uXDhk%ROPsL_JzAm&STZJf~k}+o@S)^c>JbIT+XQDb%)N4bDvrXSd`T;3NqpUVT|Q{s{Pwl}oh7RvA@% z9a>92ok#lSiU+D5NrXjvYsq7kELb8_id&UZiNfc@H1`NWo48YC|KA|?pJXuBerki# zr$JEsVGO@ATaW&pB*J^&#$ld*BCFO&P~2fV?)znjo&8%t_0a^P+E9b)EtPa@!VDtf zWWoGMw1aHrBV3jkg)7hOq*D**z*$QM&izRRTU!T`fAtA0zTFFBN9pqFE~>EP*m?Nf z^Z;KaEu@Lcf7lw6Z=BOoO4GWgu=gV;(gwrt?3P2viCxq`{C9}M{&#=DKy?{T4;YDG zAJ4=w$9Po!6GcvDC_!MsUf7=cfK#h#BI{!UFz`$Sl*!EJtT$&;wOSFLIH&~OYR>pe z?WWzIpQj+@T|7!Gd5lNOyYWgb;g>IKV{5HC;NQ<}WK!t{kXU>dfZQdw6f)rQx+1Wc z764Nx1mh#t4xji9&;!+pFyWk{@N0l89ut9=qd5h%Mrsc|^fCgsYQCmz*OwBdS~W0> zzsgji2*98Ycy?qly{hobqXGA_vZCSE?IM&`_U zfLEL%(fnO59$r@im7@cR?8{dCmurqA3NDfY4Q)J|*hTtJE&x-Hd-%0_9mYF0kw*vh z(d^}4`X;}N^*(%z{*IqQ-P#f%Sa6e4Nn6NKPl8{nB7Sms5ORz1pe}nZZU{Y!HQvei zGG{AJ@-;#I+Ai#W{~HgN#nHo)&eiLk-AGz@?UO;K$n1@^uCu!YHY;^^0D}AoIKnCCt(o3 zKqxhCq)>H`V8FHa*qxe=X$Qz2uQkNn(SYQ%O~UJ59cbAYO|IOq z$4a$NG(@Zz8HsM(lX?d2-ArNgo_Elp--?P;E%}yN_t82%y20XsD+zaYgWgfjL{>?I z_NWb^+o(kRptp-eF1*S*_6^|qj(#i;9l?LvUz2+YIy>U|7Q*R<*%~OMb#+&D=#=qyO2kc&&HChZMqm|5==`z z7L%6M+TdG$noCnLgW9S|^#h#;pzif3vUgP`w|i_BYvdjbo~s-1tV|Vc=9T%I+0wYn zCkuc7F@l4ib#dKa1AchtBO=VN!#wK&Jovl-ZnudV%{Izl_h1*%sJjNgFNQ;^MF{MW zEM=n~&w+aDYh=CFO}2A|G(2{yge@_1aIITBY|$MAx#ByV_%m6U8=VHh$)>E1j5R#@ zvIzW?Z=>&AHNM&G0$QcY^W9nz#3ZPk4Bb!!*B?LVsxT8QlSzZK>w=I?j>A6t@AZB= zSCF2)GGO)27~YLG<+YrZ==g`6tB`*J03-jhL%9`e$~k9)ug}d<&fpi~nGr#kSuRC!=eN|eWH~)D*&Ov+ z7@T+`mOFp$3Gj3Wyqx}yD;w*;Pt!N%|8#lqyI$@=4gXXO8EM2%Ja~y5>~ALOsbjEg zO9?b5D1%wmM({oFO8ZKtV3Hq8jrZ!Zi+tWvg_sN~zF!YVJk%tsU0WMIZ&t%)Mm~@% zURdWTJ)hn*cceOdjNz!m67u|(kbe8Qia2~+kJZYZsA{>5YB>#{4Zi_L1zO?XWr!wo z4fr{p4*Z#o^+a%JKJnOTL53TCkxL#t6pFr?f#fZ5x5-1r7JFj&;~U%6H4;+iFM`7H z%DmnlF+Qino;QD)NWA^tvZiu>9hzA=RCoMJIlh>_p7WEI${_k??Zf`%-8eWu6WdD* zu?M%oUv)2-_4GfO<8}cCC!K(&qS^i8Q4iX|6q5?+?W7{A1?1b6sL8lrRLeddGy}Gg z?#^W}RCSi*Z=Q;qP8-6v&}N$Beu=g#m(l3vU99f#bxgQijt@W0guO|}v7)$vq$G-X zwz~tUemexSF7C&lSrY&aQ|Yql#UyIeX1IFP38=FYSDm#0mTXV}7gr^^+er% zW>Akw%6x1?D=wW%*ocebXjN}WZdKmF!qtQL?!-ubhUi=`>M#epf}0TVK8+m8G-Iuo zoPtM}8-eUk1h4C7A+cd6_;?gL=uNyx>~nq-&xe}u?phH!Fgpu2TX#U($z*t-tVt(0 zm*9yj=J?oL4KHTv!HIq?VzEUJ#&}u6q^At~;p7Cozr!3POEN9XK4hPR!gv9@0U2JcF8`cXqHXK~t{*SQg%ZwViugs}!XY{>M#w1F7~ zBQ1pV&E|A^GozMXm^6y}cqWr5>jZFubdF4n+KPAfRnRA0Wt{HB5qNC0Dj0t?g0|Jy zsk`nx{B9`=oPPtS``ZMk^-#?p(i&9Zl+wK-q!=jo^}VaTq%ppdwvay>>tw85BJc+M}E;SYovI8jXOB}pb8Uz zoyGg=W3afknppZw6`q>K;$N>U9GL;&o3RvR%{@q^V-h5pR2dxY^X7(SOV&%kdJ14_0YY_6idh$@CH=B#zmOxFG-5>`~C_{Iv zE~yBNqq8Q)(RHhCvx;k15m`B7vW!%~WxX$)>%S}zQRpPN%NG1^tmT}{21ToN`S9mz zHJMS~M>7^rL3ZpC5_rJ_3|=lKbL-~8w&GDhTdu&#=bFN{-Y*cG=M5I4+sK@QE%h>X zrVdeSu9E#3#dOY|VfuLV9^7)w9x|ea;PcFJ!mT%Z;n)L945&88aJ6FYxK9L`xMKre z>Cc6^QVhN2AUgZpNOaP=O?>A+V>ia0pjYAy@kdVz{*e61`DtmB%0dBDJvW@+*k}6|NZxhY7nE=6|Z(zyEFqqRe zQn5G(~N3Wxhsi#L+%>=C~2?P}_|>+>=QZ z7FUybt(CZO>sMkXd`0>W`$A~DC7`@15QRay>OYaL7Vpg|MBgUOHJY?U#(tKUo#9<`eUvy}GY2vrj*A2Ac{U@z>62&c}D5i~X@j23@( zrpq*v_zk(6_=}#C`L+K4@nt>Xe52bu-oH@W;d$$B&RK*j51xJQVERrLe$&Z7rJkWm z$3*^!(46mV&x3DM#zEW$Kk#*oBUei{(C3oKYW-OT3Kh0c@-G2;7B$iUoBOo9b}{*W zW(I6mRs`KyPgtW+O&sO>kql2%hV5aK=nu(0wmQcYl)vVY!xkANckW*L^VmA>Rb~jz zSJ{KSNh6%|Xn_N}d>~UR5R-f;$~62W;e$h@cU2|fYm><&zprG)zcs*~d&k~0l%u_a z(L^D%nr_`5Mbu7v!uGXOiP^)G{NmLid~SFqzq;)+jrLbS*J4{dq#y$Y`LjrNyd@dC zB9q>KrwUwoI{DT#NTl~Kp+@0b(de}+7LRg4y`*dGotRK`j%mTnC$*@XwUR8qJbC~U_Pt<#-0&rD ze5J5a<{cd|FA@^OQ;AZj1vaccNiR>)ga1~gv&r9H5Td-1Jg8u~uWQ~DsaN+%vhzyf zcR7wzZ{ zJ)F3#1JV;}Fhkm#t{ogt+mg!J2Den`svH9n>IdNcfH*!=u*Uw&&6xCcA}o-b4LiI> z!IW8w{6~WqNQX4=r;9sHI{cNY*NU+5o$0tiS_VDmSKzq#0~qk8gDPr7Vcq9oa(Zk5 zO-`4`)Z&psjq`6CbVYn+O(_q77tJ6WBvAg~5EtQU9}JAx7VU`&xLIHa~PE z(;D^Z;#N1j)zu2O=A9>}7kb)ltKJ<5b{~qdRs+Cs4Tn zXCi2R?Qr%_4~d5#ILYZIs`U-hxF;-o%fJNpCr!ZRDH<3(uLh^@NTF{ZRdeG?Qy|@S zEGWzea8TP$TGXb~w)N3aJunHnGvCv#-`C@_jFULYe`E!By3vqpC4=8-j zVgoLFk>holV0bAW_WB-xMK9-rq02FRA!Uo7lFC5-^?iq4xllM0{sd~4h~|_nbLoeb zLU_H-2p()24U6)s;ndtSWZvoZ7*Ki%f}^?ms=PKhkfkh4ogE3?ktq<>yBuXNPT?(Y z#AA_yKK&#o1!<6vbz!P(T*DlPXBRI)#Ew-o$Tfq`Hq`{D)z>i*JaKY7hw^uI=+8se zq|P$~f1AsL{j&w&W3EEwoCCSledSz(+BIUjbq+YqilntRZyg4e#L(=+e5~IkgG(l! z1&7``xU@b2+n|Lz-L1vDzgFW9Ev>-MCIM(WX*WKfy^n0w$|TlDWg(lBh3P7rq2s9$ zeN{7z?>275pASys0^?Xnu}9+85sY11ro-s>Lu7S{1T;n51B3jDaOY7zxpVF?X2mDq z9s$M7QCs1T(p31MatKRa<x@?O9 z)E@1j!rg{go0U$THdc^r9|~!lzyg9x!-=lm1YEb~3PkRfV=tyWB|C%)ROk>+y|&e( zx=S7>H~kh7&p*u$9$!U&*=@n^-I=e8o38intr6b7k-9@JHla4Df8> zxX0T;ddziLt}z=Ho8-~fC)04<)q4#i3oa2G?h)vQ-C+Y~cw)|n6|^@#3a=Y<;^(CW z^wOhaT(NpIJhgd9K`;vPGC6P@`3MR$SCF)W-uO6b6Zsx!1hWr~65cg7Q95odcpP!;*|G9IjvCHX;6B?Xhat2I|4Z`5s2{=@7 zjk?{Hz`l4Z5&n<{mvjE$bngzedXzzSwB*yZGmRmAo+nh+c2mWYK3pySf>Tz0Po`>hb|nq)%qOE+SBR~gwFDn(r_x`}b&U-B_#FDz6j z2HnU`61_J9>c^e~W6gZrw`m5pC~n8{u5~cJFS0=_l&2?DyU1X1H4!LJkb3?-%Wz9taH^3ZT7x zDRwTdKyQm2y3b`g_xtH;*mUkXtb3mTI)5zCV$&4#dsK<}(WCgyC*I;L^-6kaK`a-? zroi&@X8P@K8N^)6wm!N;Hl1nUx<_h)-S_KoE9)3;sx1Zo&~&ykIG!z3 z)&skB{bYgSaWrljK+Tb=SoFyQd`^90H%@OMBsH4!9x=x`MGoXd!3%nUYN3qB6!KB| zJK9f@qJO?Ak?r&9xdgKXFv0%-n5=K3XH{#-;`$oeQD;I0l2;rw7g*8RYxE#` zTP~dOl!gC39EDWtg(AL8QS>c*OW7C8ZN;G_F&M2djjk4sfJ>$wB*R>emz=7^ z+i$vvL%V4G@bGo^NPIFZEz}^-H?AkMrlmqcUn}0KdxqDuBIs-B#dz%1N$l25z>sC) zFw$fS_Oc$RV4PJyF#aYqNxp}gAZ3VpsRKS={itT{94uXN7aL!S;lxWfiTL-+;BE6C zxI_rC^;$6|eltLYe|NCtlnajE-cCPVl|#?;Vs6RaAN0!?o{PSHnyXg2OdK4xkjV2h zN%Yt|R7d48?HKq?EOr63juInDyPF(Loa9ik%a;m8-p8heTfpp76xC4;#i1MJoZr@F zREiVB1A8K|L4^N&K4s!ne=jQTlaA9<^2ySJx!exJPqeOOGAj$KAdl}R+B$=9GHe5c z6@|0mD}>l{N(o4a1Fo{Lq8D|i!mBf{sD~WSDzA+o^#hXFXL<`iezHI>D~ct4Gcc0< zKr}qp5*7I_cHw{t%n4|MsqQS-oMDB|dnNdsNB6Nc(VM*e8%$?xnvFHe(WKb+HYBI+ zhfU$*aO0j{{Byp7MifWEkBm}ITGAcsy$k7eE|~s^uOYDu3!&6H2Evd0f`;5gHr7Rg zKfKqG|8Z_F{&9}LTWy!HT+ay4S9cMQ3DM{{b00JmmC$9S+Ss>{f%=f!xG3oc6mq4I ztMmdh`Kv_yP%c`Zk%62gp^$zyj5Tl$gIL9C@+&nJ59Tcd*T?_C*3Jb`C%+Wx>+TR` zixYU7R}=myH03)bUpmz5_Hg&NF2}c@Q}M&i15{ZQnZ0VbP~<%dhKnnK7_VIi{bl#4 z+Xh*frz#FHJJlijd>u{Q@f>X>La0%_3f*XHh6Cb-B5XYeJVcs@`I8+eb+($jY-`9H zp1Myrnd(6Jr}cR0_7=E1(G(7{Qc$ImfF6n+_BmT)ND6m`R_!l_U42M<7#(h+trZjm zEu#&V^8ALeUqt1qIaqb&FQtJd*nQ*{%?lkN?3tw~3<;Lw&ENLWh&xvrB70`S!qx~J zn7$tMM0!{FJTugPKNWiCUWcOdW5M;md3=#`90o=&BBu52I7@JrwUqCrV^%N3xa~1G z-hCL#CWSz%r9Anp0CYlSKN~xuy8f2ke|-AbEIb&n55Mjz1$C7_^nQ{ec9&S=$LH0g zWZMut8&rX;dAh&ry&R*irKqu&#G z|8}t7VkpemV+}XI3i$aGl+jhm8fxl?@!0-sJhSl+U1j~2t`8JL|Bm(CIKu~+k$MsB zGD0yVv{U2_$RRJcj1_)Sjp2sHrLbFFk3WtV@f#CCGW4?0ZHq{k%PPW(fGI??{wdCO zmcmy;+xq0Lq@2RdJ2#<@<2e4uqiFgz^%)LLiGsVGau8zF zfd9R}M-Hh^VRHq{HmS%8Y%L_dS1la&&zP%K4ok+&OE z`SFS#aNQx0wwhC%r_w-p#Rsh1kcgA`pamvBnn_brGR!tU36Ja^khhN?VM$gHhW{x> z_pfq7zr#{OrJ37c|FKc_yPCJaDOZudD%XdMlDonMYD$x!BqjdWMMHko+Du&ZNL*y3 zcI3}J|ACKN)nRAKY}~kSDwI2$Lq^;TSk;k7R~;}0Ba!|np>&8mwi}?cm0yrCn_@V@ zr8J1B5aIjE&qTRl8k;w7JlQb5n#N`gLXJ-&8LgN~r^h`d(IT$y)`D?hogGWuZpYH4 z`XbIFO;tEG{3PV$X~BURRy23-cJ!0nO8+j2!`Gv-N$Y5`MWa@@S2bMdSJ`vU~G3=-a;q z`jxa`!@TKq)}B*bu=fnQbzv0_dbVKh(;`;slLa?BB9I(e90(dliO~361rC1~-~_+# zTv2>JdG4c)Cua7e%-VP?z+3dO<78T!^M;jPe2|C*tf9Xz&E%@2B|!S&9WqPpB_$Vf zNycfv2D^@shUhkB0RJg4cZ4s|l{Mvem{{|V#pQUqdlxKuZV9V!A(uTM-~($N_-rX_ zKBH3^mO709-|l=;_-zr!^8ttp+o^wJHM`zVj|?p9BY_HWtYEN+HLJM79=vZ2_#zqH zh2oHyehhlcD?zg930*1T+|ihUTLC|abjAjDF5`@|9d=RA^iQ0hUse5u7IQRulnN@@ z5uBm;4Yu!?9DXGRe9`qP%kx7e7|>)rTb;beP_Fdz<~&ik6KKR z<}V??ot?qHdmAKqPb2NAkJ(jE$AC+65G-D?5Fd)Db-Ro+v@xfPEYG}7-;7KqiO2QG z^1DqA+g#MMLTO71q--HUD3VgtUK73IAoT{I4NB(0Z znodciQ`TG4Q*Ax)_T3!3H7y!9-hV}ZF1bxi`j3&1OdCxwtf3m(@)+VZR&@5Jz|%@k z@K)pCWZ)=p(wmO`86n*B$P&zZD1)P4q+zyJI{E9b2%9GzWZx%PVMSyE7AOv5+WaCc zwOou#_Nnp(L-{myZ92mEG}KBMkFQu`M3LX*Somst>)MxSv0xrf` zy&yHxfc>+54Sl$A8r&S=hUuV>XMR`G>|G1ec_-?4iR%z(VoWn>-e#$0|y*SVdw6}cuRFL&A)JhbKlWLR=zk)RP1aCv#pqQ zsQtr*EKH-a!$1%GcbVR?KLQVXZ^A(ANh-2uHx%0aBWr_Vh-pYZAEX<=&mU39-xwJM zquPSOTM*91KQCfqtz6*qxoi@;zn$#d=|Q?r7?AAjEJ#@S67-hshF@Ll_+y9u<9Al2 z@jVqC@N8@$6k9ZNcf~6}`PxjFB{&TIMK4H)(^Qnwm!s`o_OMbz7rq|Wr*)bPKC|F) zZ@U2>FaO@*f=&YZemu{zNA^I^iy^R-+C#MGHd4D$Cy1?RUd%9C!l6PIU2`Rb%olqJ z7fuud3@(6oPC+2s`T|;$(!f{065@Fq;nulzV|c#=Rw79pfS4{42UcKAP{}S-|h@&F9U&_f!9@4tzOIfR=^JaBN>3 zhIlXG-$xIj|G*sn(2GXQQz-`pj)%mPF|cW*4W573PsXafAZ7Yx6ORqJ#!2l;%Qd5@j9=nMIK% zQ=~x>&R#o1R8lD_rAg7;r<6wNdw##a;J#evp1s$4p6C6DO-8r8ZhUZR7HzPb9_AtLm@4;}ptmx^sjUPiv+L7AL{%q`GKIu#)JZ zK?^BuU5B?F#c_hF8ry$N#LntEu#aV_+?S0!wls&sCjCh$IkcGSy2@k4Tu*pkb)LI? zv;up=4B=?qT97M|#@sGR82LCCnk2;FaW-O#GrAQac)~F7TH#J7wtb-I-lXyW#$3j| zEr-c?u7Ll+=cBApIMYyb9hN>`jIz3AB%;a+Wi~kTK{rZq=e$G2GsOY7N5?_)pb8Y_ z>%y%MEzlEQ3ED|};O~Ygm^Nh`tcAs( zhqgrFktR9mkVZGJ>cqg7ZU`FN2Wz#OA#uijI3#zI6ujI^JlY(D*(a7dC&)6pt`E>? zdL5n!-trHgH-%{^hTn1|g5Nwm9lh?nLU)pYjp|MiJExD7kN!yJE0BH5zKPS2jnZad7QCv-V{w=oWvol;{5AL-DW zBgbNLMI~NJ=)gT6#<9N{K3q`cZ7hpi!fgJlhnPj#xOC}h*tGl+^^Y&1+AFqTQ*0-R zTqO@P>#o5{*rTqjDNugQO=a_nK932VV`_{Mhxm`_^*SA)u6aJi=FwUE7_ z`%R(q^bc@qNhgJMjc{i9BA7S)0DU#j4wk%0AWEGw_`X|`FK>v&sWDdk>EL4We(?g( z-{1*|S{^#!u);&5;@SA$BguXw{Z2;2Ok$&dK;%b|RJxUQv zql4i4IyutOXh3b9TXBny7|XtLiYh2Prg=-waB5MZuza832?@LckHg=An)3}3E_jg4 z?u}#DOR}iO^bBryiZz*g_#U~jI-gH|9LKNO{|YpxpQJ^*g|pZujK)pB#Ffr5WFE%{ z$!F`oqhd4}#f{R6GUj!}_spM51a zfOUrA#A@~nVTU6;BS);zAk-XJogIoRDqr*JN{jfVo}RQQu8G`uaTAWMC?HCY#i8uW zVxkPPxOv0|*q)PRAANck) z2Y{DQ1j$oYtSp6NBlPUqHHT!BdeDk}b)#AAnK7d6Pc=o)eWgUMA3vj?k{FYIpvjy^ zyug9Y#kjLBhMn?Tk98(yEXa2dQ}QK6s~WoK4A&(>HsJuAOFv4+t+xQB@dZS$*`K_3 zx=v<2I>Fc5Y_oTMIG@~-(c%k+w2{?!^dN7p1Ps3z0d!J18Fphb)OlxueX$Z$$#qbh z8$(2oE{Kb&RGP5IFBn2??ZH{G4d16uXJxmQScFd$21+iaYeOE<|F*4!Zwefl`g|Ga zPtJssX;1jjLnq;x^CRj!FNeq(7(mjr0Zh-RLLV5)zNcEk!bb%_Hr#-yCP7OtcZTTB zyL9QBGuUA20-=68_(dNL>4)rM68G#8xtZh)MfOSHyz3va+bKsJengO2x*zb`=xyLV zvyPj2#TVYMPN6nyley%FyI{)U=Tz>FCev3J>X&!UpkalVNn2tFG`eqv8AC>ppv#dE zy8R=ab2E(W=*gykYnn+$;#oL+eI{6(_k`Kz>)__SbkI6K37-Yu<6Ab3hE{oNHsp0G z{nDbr{de~rE~QCW{_X`y{2+!4wMRf!z-4IPF`nJ?y^7~A{KQ)u#*?iT#n3nImi_O9 zQc(V`2J!7`pmVOBca*V%y~cOQkkO4;Tzdj@UOm9q=howX^FOsy_8lUZU$oO5^>=ai z_lw-wftOG^SrMMCGZVdC&tb6PTjKKW7OD1_1?m;uAh{|AbWi>U_Oc4r6~Bg$J__uQ zx-lDhTZ%c`?Ob|IT*8=;fj*`c-MXw4zYZO9mbt-H18Y~pPc~YbRJBbm=5PxM#JK*Ens3+2V*z8 z2wJ-rPTl58uL$>_$h-A;d$t-ztgav+U{MYP4v?O>>oB7-7LGK8L)8s^k!|`vI(K~v zL05#{9-kfl0G zv^nMxZ1Y+OwhvCzgwj=H$%cgxJ62P)*;`7qsm=`K(*;DumORj!U`y7Uw9}UV+VMu$ zCLCR}3@Z~D7WG@;mojr+OrwB`U0i}jYFf--#7HKtt-+4>uEs;wlVNLwaHiSB6XW8J zS`$xO(E8_zt-qJSbL%Af^=UF{I?Icmj*AETf?wFbOp-m?auCbTt)fE>@MNCue0r64S$dJ8r;Hi%hEW42o zBOJ`Q#f$W4B{v%$=|^MHeqF94TgcrVZ^S2$PGjh*QJW^<9`#H(<9-5z3QH5;wcH^7yhD@jnW1D-BfZJ+#X59>Oa zfpX{7kmTe;zRnS%8=glN9+(EM@S9)#&V&qouF5>7s<3_;XLj+=P5St~EJ!46r4n)* z;jZdeV&+@L|6a5WmQ1)z*3`ctJ+;%|ULXTI?`hCuGy$izC)TzO9LG=lK9j006Zmi6 zWEdt}jyq3zL4}}!ogO_zl)H91jCOSci#;o0YTzrF*D?cN$~WT|OJ!!TK%A|b<1frm zi{YNvCGfB}1gE-2^7NKA_rCBqKkAnpix6wa&tsEN^s-bA{@M z48fb`$>ivQd0=#SJ-8Gq5{bFhtbBnETDn|DW^6$f7ba4bj!87{p8d7S59q^diQ}j(V@lH|%8*C)XAVI^p@cq4bT0;3%;QeXkh6aHB5ao?`<| z!v6Q)kOd^~+dC>j%IJ&+RXp-72R9}@r}?$MAZ`ADt6fw9$D^w7im3(d>l?zmt{lah z*-jS!B8wHo9Kwj2H15^xROtC#P2xUQLw{2#?W8yGsqY+Gxit$<`K98QPg1z`Kr^oW zb(mgW@}1OIe}qLx#&XX8tRZw%76j)`WV_3g;pwCSa&dVgr#)1TeV%llSceEaiM>Yb z=!d)5r1cLzE2Y5A0wZv`wigq63NT83A9f`?!cDQE3^n}l3(iF2cb4pZ>2TIj`kG3e zSp)MIRf1+~0PXpH1d6uALhj-;RC<^VUiJDS;}cJ5p>8v;yviL1-d}>iMWaPm7fj^t zAH7Q6%9t_RHGQ_rN;#NV+)NuDU!slz18Hse3v$%!0BXJ!m^QC|@~D|ZPR5+3P5(;3 zQ++o5c>FPU@z(=XUpI|jlaGOs4+wwo^-vu!e6jq0e%9XgX&kFQ!V=_K(8jb;dY2@wvZkpBqnI3UV<|ayc%zZHLBP3OI3-H;(?2hF)?S zbn{sSxX%}Ikwr4(bVCfhb8^KnsTh8W`)l$m@fP`G77z07-FQ4r0oG``WATXqkegHi zTNl^T+AIDzcJ5o2tGxwog&IQ2h`Hc@;iLWWJ6fW>S7OMa7!T-B5NHdDGq{1AOLXt* z%Va{64fMJD(iUl1Y?yVIDp>YI{8KAYy5j@f$88hb(AG=> zrX)Y5?uvb+ZJ37W*sWDiR8kHKCgtS0oHUy6K1KrGPlQ7=fYUf`4SljT_GqYwKdrC9 z&_5GJ5%))l+%(i!uR|+tcN`>hy2fGaRUXAx*ux6jK*%=NCtK$?kWW1dn7B+0CnlEQ zHqm)1UwRl%Z2S)--^ubBOD>ZmxpzSP2m|HUFKbD)I-R7po-R54n_7Pkr|hwe=vw+I zSjG+!?SJnfMo&w0@bC?widEoYodEJHKEuEUQ`Tg^jaf}v!Zz;jMDpSsowfNfL`l|R z=!p(2UHF(*4sXJ{T8iwF@*A96b&2@b)k3QCZQ>|YxAm;4A~T9}>}_1~p?F6SMCt~R zVROp`zl;uU%>PX7GlFsEt||!j6yZm~6R}`|fvDK86UdiSP_1mhO_~E>xjqSu{gmN= z`T$vdUW&xrPJDDOEalHZ9BO(V8X}ER`^g#VwD35zWXywl%V@Y`J{9bf zlQC(2K9&rMk+T(R$T+(pFsFSeyZJ?ib>|OZM*aRQ*RTO+8mh6nG-Wail@J> zu&Ndn7SiO)tXI6ma{HqseM%IjnEb-tz-C%IGMHZe`w+h=GMwL;g0{x`cr9Lro_lIY zmdx0L0n0=2i*hw4tyP92AHwZ*mLpw1s{^mzxrxnJ^@vL611xL;{7);2{PKyx)6R|9 ztt`#%C7j}&WzX2l?TWE*){M3@%k0!mVRXL|5Vq zeS28QVUy%qFB#!~Q~V23c4mT#v>{FKUV$;cr=aG(au_s-6|%pzXt-b&J5+EPZ>vOA zJL|gIZDK>1Xk8B24QRuN_!jhST}v7TUu{UG0cTXA&cb!BVx#vdlvutI&J4=n$%WRm zN-cmEc|E5_p*K)bqeRGJSh3pOUAUyij2UeZm=nv}u;8mCYu>Sq@7^;7UJBXHqrJmX z;)5U2+%gL;O2pQNUudb_wV((*H=abPh%4CiQVTvk;-Kd5d$gFx=!iSxDVOeoJ6#Up ze5-J(Vr&GNjT|@a`&RsucL=l=-=`A%MzAm!!Oxl-ByUmyd3NOwP5AEsYKKq71riQe zE;)@9PPj>$bK+q7_@i+7MFoc0jc1#Wa2UIB3$EXrghPe7MkZ(|ELl7fvO1Mut5Fb& z+m5Im(9fq=-8slva0#Dpji5&)WI)+{9qd*w0EF-OK2)79`=o>G+O(KoWj^kkBG2TUJqgw< z{cE?}%uHwknag|*TCq6~Gjaau6ZBzn0!}ojflQkyP~iH&sP8WgSS^QpkB?;?=Sna< zwui>PROL=}pW$SSHHhaPH~eBXighUp=BeyM?98qJ)LAU#YTg{jU%NMwcY-cAHPw&w z4Gkx2QVogK8YOTzrw!d>d!g`-G7eGtNUeOMp{&oBHf&wOXO>lgPvC3f=4plpCcfl6 zS{CBTAQwJi!%1qHmrY&#w3*^7W43-w2rb%x*t|Ik6O*EFnQ(SWTwjIlg}LNG=RrDm z)?$1RRZp}+&(rBN0{>2bfXzRQV4b@JI!y}2X#ML@7&6`7v{1-Nx_0ybnkvvs?>l`q zx0zHb$-~48O3<^X8iOZ}Mp>=Pxaa9Lj2G>~z`PM8*jbMm?6gL3O`vUWe`4jXiOla= zJC57Ch}HZj#}r2>GSiz{xNj9hquHxze^d+yDLig*RJL}I*f7L;8)01g`|)Fd~-jL;#t;6gZL7l)(D?@#DdlY#*rMck9o zb~vp}6()%S!LU6BKfbc2TGa!%Y(W*dYcYeH*xL)d;}fim>&4r%5MTBEg2#f_EH2v? zBBwYrapk$}?7S!jIzx$VXCq#)l7hFFrh)09za-MZ9$rEam>t^$c3-sM_`f7@&}$)Y zcY0$_#yDKiC1Q_!&Y;xIaCS)3mwoTbVv6rFv2Kx|56_mx-#s&N#Y%n1QwbtrAGFzC zOL>+SEAVBc3VEOB*-#_r2cUkQPEm@$;nN4{%ibdJ-d+hG)OsP}a6JADtj2kUuW_$! zBvu9uWfMyUUq^@&B#wy!2hT8?b~uMPhMeRFtBbhPgWT^5Qa7`}K_a84m&Fm{G9x%|eW+=;d5jqYD_{ki#K&OsE~i*B<%RM&!O+BW1hSkRK`~C(l!IGa-zP=ClCTaPqx>Al%B-P3}`$fm&gor1<5&`3kwqiFC48wl?!rwO|) zY1meG+$Wa}MfEM<8!n5cR_Wl*Rf3er2gc64#YKtR(9+q04%0-yIp!PG z1n>gKL=pE#cF6Q-~f{NmKt*AybQM zaC`Dg?w4mKo)O>34Rc_0iQZ@OeDN@NCpV994&R8ntEEU3_JIWdv%?0p|M08oUrh0~ zVuM>M@#fEH*cD%o(x#PAR$qx4zngIE4r#dKwSy?CuYi<(F=$s3^heV^yyTxKXgt*z z`NtEFuHAt7wjFebaOP}rRT9ZHkKh%qMA8|%;>eWiM$oWy2QU6yIAQt$?rGY6tEZ;p_?vFxxktD|+sVM_bHb*sU|z7X6&l z9={kbRY+mS%g0o|ZzQy(ey!DL@&lAPK!cA2LU55P(@M~$9=l}uY=uO=qC=ehP&AsGY+zP2Ma(eu45-ffQekPRT{4o+h_U1vsx)u`cBFWzQ ze}&%R0s~`e6h3slZSVC*Rq*KCqS|p+$fd?Iu$8_=;*ETJ^v>SvXeEB7=A7;(y(Oop0cw~Ea z>R{Ie4j-=?O%|4RP>G=FXp(CJRyVGp1~&|nPb!I2BzibG!4EcfeK2@buVkkeXQ1ir z1E{F&2aCm5F^NQbj7m#|_gdYclrT*sxn(a-z4(x3*e@eHc5W6j=o{ej34PS#&9N_a zB^UJ8o=ImEqRsIXe4-{Wq52$f#oMiTH>3($9-KtYxZ#}cqIEcVqY3{{te*dKZYAn@ zn9_TKzgW3#6LwC`!Qy>ZbXUh+sF~k^y*+N^$*GCVLS;YBpT?6gs|Gy#;wU{9902Ff zSm8q3$xJsv6Vz1gp=6{M8)-I38rQ4@{d?2d?vaLMx8n%l{hVq87;HxuEfk`)X= zM_fEY;1z^h^F7~qGODx~CjAwm(|KK1eXtQ+wPmU1CO`XF^^@@QVGtSQkCA$V4Weyv z?Oe!{DDa(E0PE|{)b4hBM(TfjBhsDq@aw-4dM-d*)N>{QYld#c@PaJ3%}-|wD51b2 zVE5BuOiR~?WStk-Mg4BHbyqoxyEQ{g;7>YKZ3QfqZJ>Tj1Wozq7~*I)8){#e)po~4 z!x14XsLE|&CE|&As4$HG5EPExBNAz{eHYwSj^@JOQ?C4LJZGElg?qMZ;~~W~a(GJ} z9knPvH;Kjy+(BJRnvzqYFKh%CVE`9hPOhlXrDtb#}7QG zK1u=P$Fa*Svg!i0`89$re|`xg_(HIDFouPm`Bb<4AsB3}fxL6k*rKcslHaW`{c;A| z7p6`pPIyVYXcO`Hc?$St&N#!p6i4k{h&fFYMUI}+n5g0~OdU{wZM`OV>enZF`_48< zjhn=V8#a>$tF@4-ABrNeC#0!AiW{<0hOXW;590QzLgSApvgd}t#SMEx#){6dbOU=f zeys=evU3%=Fuy;I^7_D6Q&Xc~-;Nwog1K(fbc? zxDOLeY}$x7cUdySIt#E3{|b0Ii%E?O$7f5U;Jl738@z8xWp&QN%ZQ8g$*QN2_v0%} zFx<}ihY7yJLMK96HF3O{f~b|d3iE~CwUmku9NjCAv!*yPsS(N4HR&SRpczHHBE9ip zUpJVi3S2^R8R#Z&c(?6M?Y{C#USIhvw@@z)Q|}jH&{ZkAZ0k`bYQK&zZdSpZ)*|`_ zZSmtzF^G$b#V5vb;NPW)RkFibw}vg*km$wv+g@SUN;R2G$b67BDk3g<`w3ghp?UiR zrZdxvq>Xw-COske@}MJe7xan(r^T#b+BdSNz7cLmJfh!+&WAs?Tk+_HrL|jSHF46{ z`*fdaKb_<00|g-^xFI`~wB4u`_QS_ukCq=k;a#~n^CM)&z2)qmn=(@*fuOMdHg9oT z4?0DLuzHp=^Pg-DJqs>@^Lrn>JSzr^OH81!ToPV~wW8}j8`@*2!Q7twfL$(TFj_m6 zGI(;9uQUu&qmh#Z*h+u!e8M zVdrg9-n5Mu`A2}k#8Q6uhbfTAZKL8jT9`FhL0cyULg;^gIM@4!;rkmI(WM|CVIE!r z!(s(a_OAx4`_G+U?l%&OE>2>ulL)lL&4URu4zR6z=d*y?dRQ+}j#0}kAOY`_p*qKy zX8@Dy&xQ5?r`_IB5dAZ!E8(%L82uRk~|q^%8PIyse%-IOr>qbL%0c&^$-vv zqRG7)BCi{Vkp?C*rMp8#Cx!=es9FtU5riXpTef$szEYE5icKb+G0m zK+Rc!Ii~c7Hq;Cg%@9qu*GhU#WOtT;-5Z`say|6isMF9Tu8Fw~uV9H{7KmQnz_77* zNz%C&{J}_lHtcd7#;SiRygd9loew=>@dBTXC->5i(&bL4 zFwIIErvDWa{S}yYKAVk+*ic{E7}!pas72zag(o0%p8@lIV{Y#xcZSRzFWlw-&WF3! z)1afYt=6}&0n%j0!O@E^@S40TxZTR6T`lr#e*F-55Gn9>3=U#WnkC69R;^ua^&btM za*xIw-H5jijui2ezrmWYdKh)q2(sjz=+jmPPS;=3fuYjyM8ggb>*Q1Uku%BU;7KHB zq40bs8_+g`Tzc#JI8s}AjKrE-lE^tj`N`wlsN&BHH2-p~y^7!^QQP>APIZ1q?%rPt zW4`(mjWuF0=jmzA)cq(OAGZKM6(pnQYJt7crz9$P-VEbLtHY$!C$-D=3>S^oyMivW zuF=^KHV7G4iW(VFP^mjA}hw!f7{Xf;}eO)k*{RV<0mvAeVFL3cN;8F z_qIR3?lt(T)Wf+GS71WwYkuvkr&Ky}0*EzfLy=S*zpUjswUo=pP{VQT_91Kh-zR;T?CVY=OIA6dX2%^mE$s|x zb3RS_7d!DYKi|T?V)-c2uEH7W8w94=L=4JXhZ@p$=$XF~Y~8By_2N(Dne<17=YdrJCt+t%hdJE?yl71*IloQ+bYYph#8#6;r*|uw%k{2 zifhV%@QhYa1f+ldP3=dG;?H=*sKfsEahGk+)7>tnHq;6lKX%)g65X`GK4(8 za1cA{9ASf*97xyBCY=v;iS2PgH}5`+1J7pT`6(gvxs7nYocIX*j*3I*rWItlaG$T? zzLT?(ewc4I3tm1x3={S-sw(VJKm5E3^VZeSo8#qKf|d$9ruLHLO|>M4q#{AqCJhUt zFQEC)`#8EikL%mJ69-zKm=bT%( za3B66C0Y`k4h?rRXmI5}njdR{ehx+WxOxE{?s^IqHSDLS9fuL8ThAyvsRzg34yP@$ z59zn{uj%W+zr=je2@-Nw0`E(np&ldi;KtWW@LBC0Tx>rB-VSNP`)CJk4V+FVjT)rG ze=4zGhVRjD!$kIz&SF#R%P~>xFz(&*p5CcBL9K*&;Gx4wn3a(LN^P-J(`F9J3Fq#q zu1>E1y`m6!eMcv>DzJS@9&E+6*{sH!$IrcHOu0{*Ej640t3S0v{Rub7?-PNKq&KH} zb0zT=8X60uAM#nROv$5zYl+j-7HTK5~VAqHeI+F@9z zF-|z5O8J&Utn+R>v)LWX_%*e#W|Idj`*#p}G}f}zt3k|Q?iz+6li~AbeVDdA1~iOb zk%JO#u&;V8Na#-%b*;Dr>hu$A%O3?}O0~HuYXi83krF7EuLJ+B?IY0{5}@L`h`g3m zf-5S@@XBH*$@!u~PmEH4wHL3`cI~H}n&kc3VT-y+#|KaP!0H9PQ+|_-jJpU8$}70f zcP+T-vkbw!PllX*@QH4@KMrOTw~z-xowQGJ6ZucJ87`O`iWZcqh=#1wCsa-e<@dST zUur6cd*y?$Wr;NW=srr$9nOYQ$+^_0D}%_fS+M5DDhU6+njMw*V#ikr9ShNWdG8k{ z=*4{KE`^(X^6izxYs^0)ZF`!IF4n+LJ2&C}2ThL>LB%hsz}GG+w{BWKDAte zz$&MR$~>-vn!Y%wElwxZ8^4luc9Go4wMV(RhvUIBc!2J;UxxmdG|_cg1O52Jk#=^d zK~KFGs4c$;MzjH%P9CGd?O{}Piao7dre$xk>jZCo?;}||b3N_8z6V>rXJB`y6FoO^ z3(5%^<+Inr(X=j)|C0BK58DTr>LQ1)6c(bz=lPJ|U`vdBXA{mOnNBY9!`HW~`Ev?m z$n%d%+)N`oJXigKoHh`EU~!XSgJTJ#>*qsrYl)Ca{Y}#kO{M38Uyza7!&!`=5!5~1 zMyun8!^x80bkWzTB=YK7=3lpr&8`k$SySh-?OVLrga9A5`^yxZrJ#s0gYJ;+Ca@G^ zU%=4xQnI4$CJrQ*;)FzooA%1U6}3$J2@yKHr*D^_*D8z7mMe&U`k&xC2Kvd7yT8e< zf|V$35HHLHXHce8*!7CLL*rpHXg3#f!ujRG^J5BR{c!YOn$A6+U`5BR(!ddG*I{yN z49OEyMB3;Hd)?C^V6-)Svey+|T5BSjo3Rz!Zv5dr?>xhQ0@v%7cdvc<=zbb!)&+If z&w##@IsZ7tnFJbFkW+JIFyhM1n&tgt@#=zRzW?tg=InTwX>UKsnzj^>9D8?u;%-e0 zQT)m^?3_#w4#v}4vPLkc=NeT#VG1*yEQ#QFpi%ODB(PE*nu-s>$b%{n>~@qE{#0bi zyZ!Lt=T^Hj0`Kz22`~7Zvxa0nazOd`F{qLJ2}bW40IgxAP*u8=+-a7ES9_%S#(U9l zux2NC=MEDcY5tGv2vS0KfqA`s^=a_CE-m_-apR@Vv|Oq3myK~71$SnJ#LZ#DKlMRqk1PUaQ}>-41AegM-3gnc?V7&x}R(H zJ4Lp9+JwJ&P3+G1J_ zOUREk-|5@m?ZmqFjNq|thr7DVp-Z#?F6)k<;*vY)lIjQiiwYUsVb(+24#t4bq8Ncw za|M1}&w(o0$I#ZW8;*`lC((8u@aLN|Y)-u|2qVWxsOpY*GqrlY2~EEjzi1d16e}cnYh{?Z+gYsqnVP ziW(FG_1(1@ea|M*ZtK9EnN^C9+~5jtog

}-i*vT1 z%ZD4tEl9w*AKsF_`p^8{)gHKwlVopt6L7j}B6<(aL;+odYLC}p$Qo~Rk+wr2l^|$#7O7 z7lIR?5s4!w;KSGX=$k|NYN6XEIcF&b?p;N*0=-dY&QzHF*8*H`{v`{G8=%cdO%yUF z1794_rO9h7iDb%tFmLXK$;=RZ?7!221N-58%q&o7lot(RLR$ zJ1LLHHPk>(!xPdq1m9Qwb6o2zA$YfQv3Ex?oZm78Oaf*Qllm(}f8Bm!`QZUw`#BRo zn`yFFPgdd$i}~QV;yh_pkORMfa@Z+o$cC+i%?i#9JS7b%Pt5m~*7k1q7NE_0$bT}2$*~}gYYd9x=jOh}2TQWB%`=MI?JkC@ z?~Kt>*q80TS52}MBFNB|67KEm&BRXfF**O>IyvyD61J}EfX222`ar#%Os$*?FWd`Z zup}Oi+DDR3{}gxvIk3zw7sk)_fubGixT+$UMD0(+qkI$HkUJClUYx@V`YHG)ax6O@ zDFe?J$iry$A4KVl3)$^b#VwDFrrQ@9^Ojr3iUJ^0z;=?{~pbYY^FmmRgm&#SGcAn1$P{rVOY|5c&4fg;l0f?aWY~-iUBSa{JD!( z9D;8D$E18m0-n+?!X)urT&?bn^>`E4JuqN0V?E*Sa~0T~Weo!ds&HTYLUfh6zz(Eiq^SQ-m$-rL^p#ia!EdMYa9KBcKA(H}({k#XOCVj)> zM+UJ~=u{bMl8%ah8Rk?9&-<6tWbRcve11BgZl0e95sokEh)qh^^f(LdjUNF+KL4Tz z$KR*v5Qg1CHaBqjWw`LA9q@}7%&hUi!FSWyrCv$4zw8j&OgY4hP4=_5^mv2@rdf2s zH(A^;e;*v2X9~t`EsM4pIbQ>G=j;>mM5_FO+RYfIy$G_N#Np}8 zdqU=`3_5o;3E2h*Zp}|aI(Ew;+@x~N{@uC=F5vkN{$YRx$b4TzP#|b=06&Tw-lUz=SrpLZ+rl)l-;L6Yx@S3U)sr3s%uFV41 zUEc+*izebY%iCm=@(sH1&wO-z;{rYIbNJGPrC>kuH1QlW8@`=A3_)$d*e~2c>>8Wt z+O%u9_}f(MeEXYzY+R2Q#1}$J#UepBQU*0`2^c>wm8=!+4<;{zx!zbmLJVi}3YQch zf72cQLj6U2l52w-liYFpc3Ch!HVG1boubccXXBJh>*(ssg?NAYL%_h^zoW0ZOK?1T7ytqW)S z`7J#$MuIhR*|>JYJ~V%0!Geh)K78{BcV>-+U6n)0v`P#3=v_)$ZRIf7x1Rgditx?2 z1~P>JF@+0pG-ng`v%JuB24biZJF$z0~{=(}t>$M3ES)xU^ecMcGPnMG53spe0 zc%cDNM;9UIt$*k5msag9a+O9>CWq6;2E%(+dXzF=xp?d`m48y z^WVwBK64W;_87p>+A_iO_?DQ2Msj|G|H!8ow!G^SYpndblloigfYhWoNHM)nj_3V=Bo)9K_vKuQ#SYqM?=R@OB$P?1=(dQV4Cq>ZgiV3Zbf4f zYx;_OeXd0%FaIX;Cw_4nX$dgdM?$1BvXzXIb%n?|Ij|wBgFExd9WsZl0fUD@tasfZ z_TbBUc0kxul$a*NPS;}49C?GcNtp!B*5hExp$>X;rofz+UkF8ScQAf<5*r({fUO^G ziAJs2FfO_RTK81IRX!lT^O(be=_c|15(I>M(ap@ZP#ZgA*}WsQeo z*^#$l%)G$}E=(1)&pjhxV2>T`R~*BKR1}ghGikQJd>6e`FEGpNH-lwYJ?JRj0~OsI z2v#epy>TcBd>S@!5A>|T)pZ*g+8&R$2F2L%R!R0JxsFbEnT%im&-b*+7TvlA^yr;f^749*z%h}5>nHP|=xI6py~jbX`cUfnw2hbwd;XKbcOWQx4iw2s!_dr3 zh+2{dJ9%&TVtWg={cVNNv|*wp=PE&2{~gS_Y{72czJP_Z&!d#94Re@k#<=Mo*fBE# z&Uz?7Tf`;!=9_`C-xM+4vW`DroJa2NN`QCMy@jl`Iq|unNc+ASqF1X4oVcA$?#TxT z-J4?YI#ULG1iNkecUf|#rJb;@_rzCwF@M75IH!O28wok|ke|P!0iLvHz~1x-{*~xa z?L|3t%(Pd5S2pR~w2*$@=R_k;Ixv)_iACcfA32z^=VdxB{$4m4Zcg z5%{K0rPK5_q3MQ;cx03qsoBocD_?y%7u}IKE*12T<&*^lw zi6r{+zj)^fb79@Z7;5Zs4}J)_40q@wHeXF4#ygsXx8x9Wt4y#ww~2o>Ka_+V8-;&$ z#-M&h2Q2MPC%bbx@cGFsT%o@J#>y99$d|pu=;JGBo<24;J1vlM2Aa)IjjV{-c#S!$dje5me*41&nC!f?u*vF?FFg-G8AJri5!^%o`p@ zcX)}S&ThbiDUu?mU*8~i%6|e&e<;{bmKG(th0r>S32c3y6q`6}B1jZ_?V7?OAjGBr)N?IM~c?;n(I8CNkWDwNsMtkI;*8zr2C0j1ESFPl&S?c!^%u zC6ltE8CY6=2ItVD#9Gx7t$x-Boj2dnvCNoVsB^#t{#E!=WrRqxdI3sZxrJATC`0IM zAzz{xjT>egv02~E@#EBBw7h*&VC%ggYn3*^M$D5RPt!}ZX-e{GeDG;EUHz^Ma;BB>a<|2qyG15>lov^@8|M(C z+(oddyNt$2PQ=!eW)SD%jmKxt$E3^SVQ$NP>RwnK+DraEjf3;|lb~xvAQ5XigX5m3;+gPv)ax7|zsg#ndqE_c_#`rO_oUjf zS2D4Cbvs!rXzhu`&+R?(#xj2oRs4O;?bmy%YtqW9)4h-lW0OB{R`Sbs||F zw-Yt~Ey24#p5P1}dG^cSo`tha)L!`pb_(o>d;bDB%l=f-HDf!OJWm(O6HIZj#w50V zUOldiQDIXKAI4)f=SbFR31TNc659^k!MQ_iNx8xW?0nQl6{;&?RLxIRfAY#cYwQ7( z%ZNv>{R?nq21PN4YrC_oYWJZg`WeLgPlH|0?vmOsofwg11xKbGCa(R`tVwns>)u(xCB_<{oza{i<;BaOwr78OHdy;6W?@b&bpFjgF6h+1U)8Wf_ zft$AUdhIRF(*8-}T{Mq064kdR@PGriE48L8 zTYO=~@N_WxA4TWk*VEg^@zT=JkcL#El&DlHjdNdLLPQiQTSP|EB84B?I}K?O4cbdX z>)h8VN>)jfRYch%viEbIKcHT(?>gt)*Y)|l-*>cWt4JyxR-w2pZ36lnmV)NCW*Yuj z5_77Hd0vtam$~f(288s&t+WgbO{L`R0SW937vdVtFVLV2>1JlLKaAh;3_tSuru%0U zh{(inCPph2B8HnGyEz^_9{bah>5AAYnn;u_kKL(l+6soi6I!<#UmBtjLO6f#fAseA*JQ)8ENbL3M6Vgl00Z^cd`3APsy~YfK6ff|A4-t@KHV7R z?4N-%(zKgR*I$GA#Dkj}IF9R48l~zzZ?T_U4C(Kasl$hQa!Wmo*xTuI?I%0w4|kqj z_Rko~zb`{in;3FsCM9B#6X4h3-C)#p0;Y@DlB*6Bt_@tpP=)>I zAJ+1Y&E6lNmo<#R6UV~L%um!arjPzw$Nvu{Q#`lAjcIe@=P`>)Xm@fjdhX}v10_O& z;FzV{x?pW`q~i)4P1-gC`WKyd2TyPEWA!Df zYA21AQ&n*C+kZ5@ssN*{g}IgT3Pham-j8b+&~BNtWa|2>AUsP4cZ4mmk(Sv9JNQ{% z{y`7!U#pWKV3sfx-pL@Fl=*wm%6oX-=oDPju_w+F@>J_Z3z_>Y6yr04sls6|)^guM zc$;L0BZUg&!N1>>$Qh7jTKmAkRsq8un$Rd_GHM3=NAJnpLfx=F+L=7Sp0Gf46sxCV znguW|IDo1Z9)_g8N~nMJg7*KqK$oymXf&FKm*%L_^Q9sJFqFX`7qw{A6ajAUFUNHP zzH6O!A8a3vP?7dUFV~tt6@{9oiICx#Q7(5PJF>Nc=8ABl*ju>Jmp&+CHQ00TVcPv=;MEzo+Ni zX5)qX_i^rARc_rJT~Mw|p!U9+cq_4rKCxL&pS`jr4c{c;Pw>J*0i(nx}irbGB=0q62y38Sa= zgY48w$AjmmgXd{wk{@*uQ(r!X5y@ycRjbdP;(tHyHmC@$g-K(`6mQa0&iAD~mHCcf zIv)BZ0dcP~+0q{!@IYLInS3-4=Y@Zzw-S$o=%<4?`Bx6h9odTC#_Yg+l0$@-TY`p{ z2_Er^!||{TGM!#wP`|9;_4QUKgTVo zPXy&p+fa3}HurF`5|l63hP`87kh_ir^mq3GY}i^sw>2pX#vk2_u5yLMX_YH{covBt zMFIXzjRgUpU5fO-OD-?sUG4WbV8WgzENnLwT;~qp_FYs{;fefiXn7*VC z%Ww0jwXA(K^HK`iQ)zf*7`q_eCCWRYn>?GTUCNDS zKm3W|qwCN%*aaWxgp=bBDEQw{7bKVMhbzC_n5jpfP?ys2kg)tY&pVtbaDD59Z5df` zmsJp)TJRU8`c`0Twm9M2r(EF=aV}@;JtD0+4x7vpVZe7Eefnqsf9HqM#tqTL^-B>E zACeU~?^6_1{-jtQQbUWC%E`|2%g}G%e*Dszi}T+q!=ttV=zSn8cys?giE8)`i&sn) z#M!u0kBN^V=0gKS=Hw9HuX*@jV378EDsU?@qq(Xn@{FnHH9Wko437#Wg4whXSfU#O z#Y4H!bVU_7F*lrD#b=&)?xb&+0v5DtlFI?Fv3#sAnqSKT`z?Nu@v(!3m*x_?cLV~~ zyTL}Y7KjsH+ElqhOmO`&2twp3%y$mPozYXlbInU~^LQ1wm|TZ?<$he})Q%~se6G_$ zoQo`W;G~i>U|94ro{v4wEKom1+KmaH4VaIeHU%KL*9L!^{v>Q;7TF&?fqPYvhK45B znVr%c{eg%gKNHDvpFVQM#UK9SCP-Qm2F$N3P_{=-ARO#~5&P$$xbS!gk^0I+Iq%g_@yerbl?9Gw5Yl|5(|IlB2994)5 zW0&Fif3q>{b_EU0j)joAbgXD?rf>CHm|L21+}hm-IdA@xj*L*_Vm!}c`!W%to+`u% zzrICPy{gEx^)dKWAltnDeJlG)n)hHIiKAztwAs5Z>KxJN!muF~+VJKLF5F#;+j(E< zXY~e>E|&=RwL}HC_Xp!5CBP|#gXF<1Veb6R2^i&8$9xLbfEQt__}_qKv!dlwdgX>5 zx{oeFi%oU7Sbr^1cfSD#a#Ue>x(psv9tU=n)u`Z51TFJbIN`E~*fLm#^pZ9zc#a}2 z|41)gpVmC(rZIS5eGd0*%mpb5;SkVu8e&A>z?{u5@%IJ^C{@z~%e=EV+qxSIx2BSk zfpxfTy9y4ZO2G1vV0QRjJOs?t1d)BCM8CfQ&ARv*$(55>Q$GW>q)t#npM&(()5~;; zy%UCaT?Nm8N>Zxv9Uh+JdESAC>8wn3LBfnKT)+4l4G}*B5_3Mlw(=62gZ(eb#*=(L zT4(__HLIfAyiKs%(ik=t<}iYz-DJ}P9YMISq@ZUlZQfU00`DFMz{?yf&afznSX?m1 zldEDtWwj%GE0+-TT9sig|N8$moy61AeEF_cHMQE(11qj2qT#t7oLi|18hZxFM)_}O zni>y_8@uR{mlnk7cqm^E z`o=VJE6|EY&Q^u5&#%*!a?OlTq9gfIlR}eXA5l&+8bu2h(IK^9>=rrz9>Qrb&s~Y= zdrSvibp?n|xyUNY{qLBV+%%GuB|h2=|||J>!j|KrBt6uTI9c=Bd^Ab*A3zVj9P-;AGh!H2cznAl6NxlSO9 zFX@xGc{W6DZWqn-&BUI^4RrO$PtsG?!yfbUz>%{NbPb;mm#W$eTB0XuSG72m`SFYB zzip?#Mv5R#w2t>w|7aE~dCyF_Sx?p)M^VGElc4qKS|ZSVM3idx;_BB^@ZHpn_;u72 z`wEs2cbS-Gsfb=uX}bj~*OB%oK z5_#(Nfmw4(gLxNxigtQTrvq_&S>4Z1sr!W8?3NBOYo3ltg15Y&Q9;_Y*XIPDl()bQ z6U!-IMx<*^KU2z|hcOfNNwF}t4j46BAAC;9#he6iwW8Nl`&mrOSS(@fivd9@R7`<5qwwc)UwlL%ho<} zwUr_9&*wq)oL40GP(B%&;s-BYtAS|aXLfi%j_zFV&JIpjhl{U5$hxC@@c2QN6zNrx zpPU`-c(x7JjLnAky=`o!u^FA{*=v=3xseVIEWtiUHlH%LFPCN zK97V7?KFBZ?;@RXBa1x0n}Yz3bk2rSDs{II9bQV&m(pl+KV$=))t3uf&8I;A$!cow zMF@YjKcPt~;Sl~#oE?m~!1PuDQwq`ih57+*~GD*x>_jeX=r#=gp;`Oh+#2RzID z`yD+P_vI|RGyXfXPmoKN$*hDhrxtKlJ_#D1+pGoIkD2`w3+M`-&9eRea$>A<2>dj8 zPvma{l*-wq0ePcSQagW5g8oKCkdI{Ej_aq%D^3Z>-oqGJbz-(MIk#_y3 zMi!6GBpo}>+2|+qP}|-x8zVJwxK-Cght8cLns)Qy-Iu%Ui@VbqUEx$n^pydZ<7>&t z`Bc~BEvXF?g3UiB1gK3kPabVAIS!Fk{67>~E3aGF7xV_wHs!+JyHTb{d1f zfhhW}A0bNXq~PfF0<5xg$Igv!>6*a|I;6p8=Q;+6{l3k3b*mrld94jYeLS}%Z#SMd z>?BG5ylJjtBIqnjgt<0ZxZbXSdc=MuA(n~6S*nxS_Mn+`gxx~zt}s~iRuo^F^L)e4 z!*usGe$Qn$1%4?egU&NYNZF%*;}Ta*)E+uTI>oGav9MpLty~|1zyi@daJqwV>!e27}dW ziSq+}81t2)l~^NlXEcY5iQ7fo?z|`Mak`*!%K#Y}A8H@|mqyh1)3ZTN^p1fM4f!C$ zU2T=-dRNt9TACwSA@Q8n*HplXPdqoGhv!u-RUvxO-gx700x~-aapRd`GFkg5$y%ci z8zpSuTb3J;HBDr>NC>`ogsj*iFGlfAKMkIs4dTHI=&D%4Jvk=MJ#Nw9qEjz0fBMHl z-}6kqH{VH1Lr#%i^KGnjcrjG)9f?n|Q7~06lJ5GD3|+xOWcBMTTDH=@d2!_|c#?A# z^{sZ%LeItYrS&;#$a^NEx&mNm;Y9BA@v~_5(v(ZFo6fb)Q{aj$_tCn!D%{vrBHWGN zLflLTIr7$|5w4sJfIE`cv0?oJvAUYUf;LzD%FX{2n;O%vd`OMr}uk=6^iZc~l(89o}jC zF9_0KYGDWO50d@T1z-JS&}qbqiEda8C43hBi<}M|xaovCOT($I)EIcCeipo>6X1Dm zG_3ZO0(+G3y+ zTyTj-)+E7Iqjv1JjG;bj6WFE`Lztty7}G}f!(z2OE7?DHNu=NjIU0Qu&-qs~E6f)% zAI}dkqhsHI;>;UNYuiECb19wfuKP_wPZ+SN$_3D}%L>fOifjU1uAmB$@|9rAYGq1HzRaZ;Bns9Sf! zn-MjtT`UY|>dWDongY#G_+WGC`+KG*Rh#=M--q=Nyl}H?7J_g(KI!Yl$zQ+Ws_!$o zLqGB{YxQ+7s_p^zCvDBHBa!s4T^SKGn@AlGl;ceWHLgr@4yP-$iJF)elgUms&9k10 zLUqVSa$wPYayOdqYF6BW?R;m`&xd0sbcqYphl0rX!aOv0EXJ;By%?pgN|ShgmYnZt zGBsieGrMbm)ta45Ga5wcSbko9mkYv)U9+H2&lvJ`=P>I1lcI7a`T1JQU zCE>#^5vY$)AX$%iHmp@O(MUAM8&9vJ@Z%&j{BsH4PJD|AacW$58B25IA2%JjsKn>T zWKc=14gEGsa)FhqHj-CL$^OR&h)VS>(!G2keJ)$Zx}1xKJ8L}P{k6v=%k&hE$=gDh zcq9CrxqqAPaE+@mENnG zTVnX{N>wJ_i{O7-U8kv(x(ptOu4L1^O+fIGkf_Potjkp1;eI?180|OaM)!A94cQ)|E~E)vxbM+s#7=F zd+sO5J?ksH{b(*OHoi_*G|U2#P1o?Hh%%SqJD#hu&LADyili&JmCe{&0+(J?gXZk3 zu+)1dW=yN1mtHK#QPxWEp?jTR@`1?$nZ`1~A|z@V}nTn{#GhOz1y{T>$n8>JU5hea`IsJDK^qH zkz~5#NiLOpm`N{eNG1(bjKm*qCuRjP%nA#>ld(LTypYhRWs`Wf*Rq#%(t33ix}}6D zy@1Lre?mQ6#hQn1c;eQHwP>*V72f#x2(SH2!1e$~%;SV`sg5~QIVO;eNW9G!Txi4{ zDRSKV3$IZk;0(!oq(uU+Z(glAp)Mv7uNTsvM0AhqMMZ)m zt^e_7OmZ*Bsy$cebd^ms^x}BDX}N(ncwInIae{j`rD4VUCyXPmpjGp;u=x5Zycime zZZyos@y&6PekzijT9^(ddAVf&J6AG(M*|hT@tAfRx6&PFPT={=6Hzg)12?zvGbN{` z(3@0=ZLp73UH67{KXw|Ek4vD#-aGjYR*|F{I{m6wlY*5Az;HQ?&vK*4)+_ zXJ+h$b)qx)Q;lyjo8q6+Pi?EaTpS!G^iQRQ===g9lw{Au&hvVeR&9Kwg1C& zMmHH_bw-fU7Z3h^`6xSn3_nw!K^jiQ;=b)2&=}KA=NwALx3kufKg<{WC2|%My(PJ| zhOz96qsAztx&a;9W6|(tCXU;|!cC)jL_EutcJD5wW3$A$SG!tC`rm$HIHClqUk2GH z*OFm=<`D=J4#Uc^M~H8z1~{kc!I8`hgnIs9-|#%6dbxWjbUT%_1$v^le+@KR?}YWu z4N#E%7|0?m?&+ft6nk0J~1Wy7Fm9r@T{3X60Mz`)!B9MiQ( z#Pj>~{b(Kb7iQtff;;3(=39Isp~an1+QHVO&I1+oRiN6Uit4Y|;>uN_^qG(w??rk` zGVizJv}whZ*uADddFRoJ{(EG%$8v~XzX*(rw4tpm8|(gR!})WI9uGN7x%frdPDXHV}hqc48_ri~xRfcZUb z(j(=Ak~?ehZfpu%c%&|f+-J%0b<3xjJ;6E?S!8hyDQKiQp z50p4?a%Pd-sZXo8nlCBjVYM!qHvSfwY>`L4gg+wt4!(4&WEJcjt|Iro zW-}}O-Zoc7i;<3@IM}gCL=b*&D|p-R{b8}ypnd2xwRotCB_rmfq0*O{UUs5P&2DO< zbAoZ*p@HKr)>6etV-SuIg|B;S>2CMU+~HMwIaOI}?!S9_Bw6V-`I8|+&NfDp?GY_h z<5C;hzugIMzV5@25Q?U@wpd$xf|$%;N#T9MC2X3@-Imkhw9du@F5}sV7w5F%5^ipBCCgguN!tfs@D)?!#H~8e>QgehjyX+Mr6n^9 z4%e`g_Gb{6Koh2F8AlQxSF&TBZ;&19DyiJiDoptJ1tr^CSk+7FMDB+rE$I*<179cM z41?Ry@$C$mtMiq(gbA@{vp0~f+H{kg17>I1+nFGDk2yU;X-@dNms{fATY$~tg zZnGam*iwg-_T|&uSDs*L$nO`{ARNxp;exx2xX~^-F6Lq=L?!KncS(q2J30 zsca?V=E>ng-9fVKPBrfzm&ELU?}^Lt8`Q0)g1vm}A9a{3gW7B2Y?RM$1e+xr;AgEc zDwn9!qCe?$OOPCl3H{ATzFJ5(oNs2Ds~(YHp-2#D=h@U%)KS$sB@90B5u0}I5Z1L=f zFgcJI%0%!ptqY&OQv29sGEUA1!007}$z*_GZ97$4a}=_^o6`g<3vfD9565(*1cvi# zNV}*gCUa}(;o&Q!!X}GsU;U1zoIXvAqK`8wKXMtZgao*q8wHy6dAK2cD|gIIjXS$? zFJxH75Y0V&?&M<~)_K*DU0ftwx}VEuFhAmh%RexZJ>NXKh6t2eOX2TNM@VYDN*B46 z!d!)H>^XP^}NzvGC-TZ$6na@guMLgLma$j zpxO?OnIp$R1)tl!e8(8S7)x{Gci$xQcF8f*2gZ`lQ4GWgisF|aMyR{bL&xjDbw0R0!`;gGuMW-o^{m65L4ne{=O^kZ^18Hs}dB7gV zs7;^ffyP?u>!bwHzgKawMSj%tRS|U!5XNuW#t54y!aqXE#aUzNld^BLhR@5T&aogb z_jNPPr)VrdF`hxTA_xFqg=e;Qrh^ZuU8Q&H7x4+M4`W3IiM$BkL_ z2*1r$Bz>?Ry_LhjIcE}g>#HY9PYuCX>BHDyzLSQQ7Sp0v8^K*wl#@xY#+H#|sHxgO z4@#_rK+iDTd-VjJ|LGmBnJL5t=8Qw_yDQ;-ej&yd^XL1F67X6bkE_mmpryV)`4T9} z`#u`bBrTr&yb;7WCRM`jai6ICpaBLn$OwMDT>|sQ0@V1=B{@%OVf>NfP^NQ}HJCLM z-RvEB2IV#`c$F@P?vLqWt1+CGNH1n)??ri!qtMVi9v?7w!1$LR++L|n0%FYs-nA8U zB{PfLT6Lck?=J!<>LI4@iZS!c83-85b0T6w>95bmFsmR6az>Q!T-7MO>y<-)E+`=B zl1*@>dmkxso{uBLkBFz?EWtyQ$=q09Ypz4ul^F94yO38SyB-1C&ZHN6&{%Y!dXi+$O;W zF?i~u66{P%pk3lTH$p1}qqHZ{lBDzaK3W1KDxad%9w{I)LfpIWtGSMAkMY^&n^Zqk z9jAx*fs(k4pnLjPc2!j}F}&-5UeCwThnFvs|F*3ntHun0-^XZ}5XFO1^V}Z18&0_T4!soB9J*2LJU_j?ljHQcF z?p!9F&D-O`7L?)lFbPglkHA0i4RE*N5xe5p7-(?TV!tU7`1HgVO-uW+esCBq%=1a2 z^(bY;Q`z#8Q2OocG+1)#9QAaIwo&WwW$^eS3|^SZ=e_;VdFMRlN39nexDo=!Pm`!Q z?@p}_tfbr`o&nBtrQbs_I_XBSADhmB%HOqujrXI$1kq&s6t)>>S$k+10I@HxXz zSy)%3g%6#2p`a`hW*)girBpNVKh6zDzVjT;@4?izx*X5HtAw<1lTk``JJubOfx(y% z9J;&|&L|2qfmaQo(u()8O(|w#==T2yJ)S0ktD#a4t$5 z<{TAb6iWrrDDQ^z?sw1z(LHDeHBBD9ui1ktc65cqPiEY~en|Yg6iyF4!n1Ei(PYwJ z=1;vlEC3(a$<4#b2A}AwJ`2)2u7#|Ty@ZqI-C)Wcze0FgD(j%WoBX&dL=MTHgU)I3 zM9^>(mMq*04^0&WSq6oW^UeZ8oL`{1iaD~WDNO6fS`=oZ;a~VsY(HMc9L>~54ZcGj zk~9Xq_fCNNRjIJ&_G)Y#R1mB*SqN$wV@Yj5IkDuQpOyDp!oM?jC~=ujF3aUpGmkdx z+Uvmrzd&mEnFlx5m*SZ1yTC(d1=%Gs1H8Y=f_7;+^MYEi`-^7co4g{TyhfH08+#ft zE}rCFw?&npVEUx`l1*N26Km_GO^)uE1d6FAF^GMEHCEc>;L6RogP%jatRA7H=?BWD zKg1)+t!TVFlr0M9`{S;I_{eY$U8t|gO*1sWm-1qud_xD6KHb5{&as$uOqVlVItwkb zHObbYDYS0uW{`95fxo4*VcnU(AWM<)BL}P_o}X*9?#+YZzZQb&Z){<{*=`)K;tMb- zg^?UdLy40$c(?WmNlTT+RqOIdcaaAchuhQqvL-rhz7Ee7Ya@OVrm(P154*q3hEcWE zXm;*4ZCv+^OjFN>&pYeM4Qrlbf3q2%M9xBSGku)#!XAD9(}v13KJ4KqXSmq)k!T^0 zH1AmnJXt!CD?6%)r=+v2eWtf#tcxT-CTw=~A)MFDXN8R) z(|3o?!qyL+?EIUr$R}&wUrj98gK@qfp%&k$<>Er-+rPvZv3bBeJjT04KGFB#CG`74 zP3}nlXEMR{D=n~$Cj~tJb&Ik(=JDBwQqOqGcDKnLjuSb4=8j4Zlv)6J5*U*4(FLN@5zkvJawW%Z^gNQ4coOzn9u4-Y2>( z6VXIO9+*9@?1O+T-fh1Cn*DhQ!P02-xpa!iH3D1rQj%M8ECV7(R^#`1>3G1=AM3WS zgr5hi;M_bl!R94SxTEhf3D)x_dfi88@`f*@ea53EHPZqpt{5i069$R(UHQM+!0?4-E!+uP2Ug8Ee?X!@H%pg%O(r{O=jahj~BF=?1OZl zzvQ9xcxZ_cfx#0Zw63oVvY#9!liHQ=`}5=`v5O|yx9bbpbHW)auHR>C!=~c&wv$Lk zC*sPBQV`NT9-bteq({E9L}zs}{QRSbQ^!!;tQ3ozHe|7ZffjVdhUKhA_8&>_p@{ zbD(?vEOdMoA|vr~Ff_S=Tz$F(p2)Q_KF`ze=K8DjQ*RWrARq}s3hprh17kr}F@yDG z^Fez}KT)p~&@a=w(KUSw({oS~wj@cQy;d1&zi203cFowqoxyE?GO5+|1Jua9l6_ot z2R&}5u^N9IY=ja{qgbLHnEQuA!5A-CcWDJ&8_EOSq-(JG+DTF%9m8rUi9%!-kPjz! zFexs&xb8P7FjE7mMenAKL^Rkwk+ZiUrzdFcYx}hBCs2u2&%&2%!2lC*l0SN zWPawFWoRnhF!?K)aXx`~`X426{aa{_ydL>$5(;E}6EfynTv-E6n+@L-l zQWWz@K*DSYNXvqapMt@7vO2uq&_yThl;Vz79KrMBf*EjE@exnL=B2d=O6$8YhkSzWn3C|QvY z-ZwHxad8n@rCv_nTw4N?wj!`pVwCs|PexR!VPy_I zoN3PvDO7-GX)GAT_>!3=Rn&OdRyLnG&(2>EMw=qe(O4%v6x~^m7A0cbA(2P4HmeA& zpQzBgrknU&P#7W4ZBcuY6@9*>3XcDG1-e5gvNu=u;fAScc&fV(oU9HZt7$-XIuz1) z&t8lH(=JEF*^H+pdGmz-jm_Vv(u$+L7QOTbI}48-d`^ZcHNZ=ul-d`>u^*I8 z!7NY@+Qc=uBE=GHGj_&bpJH$_GXZIX5e7_b##u*V@n-*ay4K|hO}+hy6lcyRd(Z1I zQWZ_G%utez^U=bE>0?>>?0e+cji2~n=XCsXPz@Fu-J&sb#4!KuW2#WO61N9Au$EDx zINL}QUbbGNT}_JUyz~XW@iQR>yKP~e(KdGeT`4Z7qmUJ8@P?bpCyCW2PjZ<_r}48K z=x(03J~qUNn>obqmj66Oi=hFSR(=-lTgAY&(*&e8p925cU&z(ELrB>e>*J=O5F6pb zF=t)5p3klZ`2q$4yAr2Ys)ajG9$2aAD7?<$_@ z;Rv77kHQ-to(mCCP822-0>Aec7>bnT6bo?2|K_Mb^Xft_rr-pfdR++e0 z+7O࿈=pghq4o{FC*@2~2UXx`&Ad{P&_hGRLq8TB;TRSH)11|h|ZO%%FXjsjv)XPD-!b*tY z{bC13%|J2bElL0PfoVOo~EyG!a@cz|~eshj1F+q>Xy<}UO5MU)5;YM!Nw{!Jxqh9U6Va|+#TeTiAK!woH(F5rCq zA}pFI1idZdc;3qm41CVQre+3EJRW)zHiP~;J9_5%B(Unc%velO<|gb{;B0iiQ_FA1 zXzjbnxc7$}WLI2A%Q3Y!UFb;StT{T5?|KPOD1pCMg}F}xKfK823pcoIf{U+%nOw6} zylWCnyEm1Q(6`Sa=)x1)d+;Cg)j*3r4miRpkB&7=Lz+6TGiunG3(baHJ&fnAQyoTTNyi$M^7X}XR5hfFAo1u`u{#Z#U zaq2B8*wtH2uiw(;>Q~O=jD>b`oqXSTo(jR5-N_{UlQXAz$AddJ@)f_N=g~j&l`+M8 z3|F8dz>F=gNH;tt8s!)1bQc|bGocrL)y*fb!$Qd7jq}MN857(SmIPB4=|J|yWjODe z2p;?|9;c>S(;dpUNUHZRy{xtjXK5GVDx2lF(J=(uMfl$H3VZmVtjbN@a0RRRZd0D| z1YA_J3?CV%g8A$W?5oJb&fGP)e&H;PNic*<&lTWJ+I`CW8723g@V(wqeefpR!6C#I zmNjuq$fUWn^hzS5Hv2recx0G(ov&w(8I0ka--~k+Jj>Lm25EoXbnv^gm!x;Kv$un4 z$y3V)Qh)Cb#`kA7hnfB*xe7?O=!OtgnNYeZNe-Wg{3NebXMz5O4P2h%I&S!CFn8wf zU-tZ@RyyeZnw+s#!OVm~(r2FrUWZqMn!k0^yQ>}Sg~UR%8Z|>utc5G=9JpUyLAR;| zvkNtV_j_?TVUH+toy(=_G>0f>ET*X|Qt*yODs7)&1q>g)+0b{AEfO0hc|ud5&2qRr4i=k)J}4#ekp{GPQwYw3+ejr5hQnd zE;TEWCKqQK(dTmmnCaRp-OC`z`_v4z2jr8%* zQO3A)A5l2mPj`j%P~o&`?3&?2MB&IrcH=2Ic>6pWHw{h2Exrq||Ls3!Y|9~bPu)A( z_gohTk554FqQfNL>OLb|B8CNvTS=o%F4c27&D!ckOMGuKrgEiXTW~t%1epYwFRDt@zIOulkBy|y%DEBRm`khd19{8ln!tYG_RHuu$ z*Lst8l}AY3Oaq#8R8b%z97tX~drVIW<)RoDK%eh@Np$~&u$L4v$;~H1&{n*O3W=Se zpDh={iH9xp(>ggeReK-GEUBg@?^d8lM>Scq>=!k?x14s$`my(BxuBg|1~pQVChlv~ zNZ+@8u&7mv^lSd4x1*=gg|k(g?Pq*}isXIGKU=5pJ^QgZw|E&1`(2FvQ>J0v#__Ol zd@wF~*KAX?DvX8&t_F)M@(?Z5NftDmz^Nymkr|rfLH)v3+|@3P-9|`@^~CVk>;QPM z%MW6{v{PDRj+b^kpq+EXX#e9FM(N=f_JrXOov~5^$Lbkj-NK!aLJNrN^c|48U^b+ z>15f(MX=D#gZTZPMprrVo{9`192n}L-^A32bkaU_#Iw|YjuA+e9frKi$ML@5cI<7O z!p8n^$LlYDFzFMk=@#vD`pBV<)UWd)evyoddu+Uq^;ylzcxf0%7R0vB>3{q z#dFvB9NU{!4w<=yP$)!ry02PlM zgLs?!;1H<}h0poCcC9*lV^|m3Biv!Fl0RsE=z!9~??mE!6(j{b1wE@)9QduxeVi^p z*}CH>qhyZ>YGPP#7mpEKDcvHM#b;?mxY(&i+<&(;xZNKm;Ow>Yuy{`i?;(7Sz8&HC z?7~Bwtfj;~8fU--kKRTtH6z+To-a~sfGNE zy*QGj%_Ka>NS6jQ+`=}sde*v|qm!2^!|<(_RR3!X9T{_v46YD?_>rmD_9YxAp1Dlb zha+(Qie!AhT9ixX`Bk3t3n9dc=eC%qqJD}ncgNxl&Cck83H&wVn!!)@p7c&~b$TK! zOW;7|*D3a@>Q=V6LK?rnehi+i1Mn}1&%nJuLRx#DGKo&N$6UlQ*weg@f0qu^I*^Y zcMHTDr8zf`Ar!eK%?Uk@1Zmq$2Avwfy+szUshe@TT7|iP3a;FAqszFEEyp9WM%=R+ zX-<~U({+rm1K(BuLF@H$n7sEW-nM^B&aJH?yM_c{y;d2{o)`EwkWO zkP;NBUt#g$0TQAf0&Z^qaC!f4ocgO4>#t4ZtV{+_-G2f1Iz*K26)$9b`5fwpd7=Vg zvq*4Sv=X$IE{7=#q^VKjKC;~GCNt~vOR$*5LU107{^TQlVXca`Z(KPqzaTDHYZ}M3 zkCKg&!h&(q;{-`d2k7(O1;npZ5#lyUb8D*9xNCQI;h8^ny7%E`HWb$cR7x)nG7-ByFgmQ1n!!4H{PD^h>DxnV33tR>D{~$wnq*Vr%8VB zUM>_Ge+NU|__I{nJs6$;XksPTLH6v>gSgjg$;NddAfFaRIu#li8Mgv5E%^r_#m~t5 zq=&>mdI-h`nG3WhmgSmB7J5m;cXvQ;*zwTPk9)9jYyCQQKBfBUpkalKH{}jVHsZR9G zA6Y1g$Oj)gZ%}#{3Jtn9;lun`$c@fpvsM0MrdD<}f8MAJY}+0v`qBilCh~B@dO1__ zw3nL8&0uXQ&vR>?fm`a=(VNp0v9s+z`hDwa;wL;Cj8%fk+XX!~LQ3gm?9|6p5Z#M! z+eh)FM>1Y~>_$w6WWg<0Kv*N=dR;WhN_SpZhwQX=o?CrIJ-hdnfPt2lS!mIp@Bw>-YP9*=JfsWOK?& zaO=t?!~HSrG~<1^uiOr$OjWps4RV}igc!G@oab^V#S_^v$?O?Fb7stEa~hG6%cxoU z!c>o)yno%BT-vHgX=NfYaa4jYow3Yq`!Nt%r~-FWFGABqe%BE>3#^mAqn*?*l(a6w z(dZoXD>{THgWPaz{%iEo72~?AKVX$r1sNS525;kEkbS<3LG06M()Pmych%~$Vhb-& zj|Elq`yoBh*K=SdTsey#6HRHzBQf|i=>zdoI!ecdo5Ar5mq<)$3FEN6nk*|C!>085 zgHUx27>r}cq&cQ^rYXy2=H6lSuT)`GMm28dSy`^O1&raG5)#sPj!sC;1x`=~;rlA# zrHeZBdlnJB6AMV(MWpwyZK7wEC(!d+XGzreYiv#BJ&T(o`@vbY3M40oK%C-Rs?jS( z9!xpI%ukvKH<(7o)b@-3pxQ2cyArpb8eh zFoqw0MMzosPWJv>{jz2hMd*et3=?IP8zZf@|?o^P}6t>yoD#hkM(cBp~)MT zP6&isVM)X;WDJ#eYNWOOJK+8rd$6cZf^%^%S%s2kOmq2t(myH!vqybUF>WBhSbo-BKC&;=2)ZuVw~$?~w}a$r<#; zrPjLleLZAX%TnHtw+QOBGU4yW0n%6#O(z&x;orAHs8y4~K3s4NQ-@tp?ZG5W{Z9-1 zbOg{)*#XxNxxr@B-C%G!*y6%sIh=7l3_mMfgb39NYTcAdIoD#$gNjfz#nJXn$y#JpGkK(AfA%*-j)X70b5fgkFU*cTRZcxjnFB)8^}GdE4~RAng9 ze8w}H|FfV*uQh4$h!15=&f&h%^O#33prd3T?X`SHQ@3qqBJMk(_4gV|=Lt~@9X*u0 zSA)rKLz!D1j%3l0KDzXug9-L8A%)4dST61aoxk|qZLAHw`L4oZ{dEHvuOc~dly`-e^Iy>pFFiw@eRWu^4B(&cRD%PAIfb7B6{+q0Zh5FfF&2R3v3W8>h|re3ph?JC=ch zR~esc(?y%9x3R`9m^}Y)9@bSgvZ~iINp{+4HX)Ah{^o~3#-|q~W&9WtDL#SRUf_$T z!p7j4jh8`t+fGniED!oW%TU;(2G`a!;9t{Ou*ohDE@+)5?{%)z{b^d%^4ouS#oPo< z7iQ3ojxzMzs08oIFB2ivh&q+tDa7c;Ui_T3m5y#-!dA`_!$Yyw^k>{DymKcR-$&2K zk3SM2G}aW(+n8-v#;_gT(K1J2Y=QPUM{%_`b0$ti9L4KIi}b4?o1AYr#isi+fS6v8CAkY$MG zP7m$G`R7a7M-fIpaD=L) z%FuOlmq5`d2kC)@P{dt;o7zfbt+zMx+`X9DR=khds-cECX>-tOV=X=W@C0qMJ4wA- z!s*O8`OJ%@%jmU(ze)VLBwFVdjdvDkK=aRN(o|dqArlATZ-E1_BN6CuDh+ok+7X?@ zRkT9?2ATLNjJ&#dzOH|R1-t9*YO*UkhIF+A@Y$a=ux+9_*({ujmVkqqoU(n+VwaWPqtW<_=UPj&3t81c#IexU~HsE3?ubX8m4B z_wikyZ@f=ncAym*$vQ)ydNwf$tu@4A-(}uEWgz&`s0vcwcyC5qC%)Em;ofCV=KTDM z@nxF^cO>v1HcKlC{{H(+2Gu(0?`H;rA@`@?-@FgGX)}qN?P}=g;=2T6q(xY3$ktk<|fU%ace(0>!j`1Ac^Z+}-1+E~w6<4@!i zc%JG%Rg2ZrCUUEDc!%XvW2WurHJIfmfq`Yx+<=Y(_d`Jg>|40HQ}63g=*J^4Oc5v1 zamdcL;ky7?O_(+59!);3hhKDC=mq_STv+}ic(^bTZa-X(26vsnOe>4LD-6Mp9TM`7BOQ z>85WiH3ivca_I9UHBQ)efV>VqOMX4~f?4If?`2SigXl3_eg6@jIQ$; z!2tA_(n_EIR)J%YY3AZZq3r%oE}-^l0cln5q3YGnkdhWj8r@BBX)sb36=TRA7>C!! z1o8YbB_e#t9>zcLWS5>@gwKnOKzCmX{OKPcMGRoPlC@L%TPPs zS5{nd09p^Jay?_^xW@-aV0dUD?2MG;%&!zOPx<-f(h_Bae>^+8SOJQJuA=;-A*dHo z2i229-baq1re1KsE3vm zyk6@K{+-pV%O-#H2^|Cpv9%y?9Z1%@_kx~!B0ZediQktUptHZO#790lBh=(qSxkZ_WAnT)?dHfAFL62)3p^#Z>V)yxJ-W`(sBL^UKRf1V-bx z_L*Gd!AE$0)+v71K8Zf)UB=1yuE(vH7Gc2O3hJ3^$ zS9_2!r@f$bwUz#^@FJb2!_2e{e^fbUO+$i{P;KH9YTRp0ONune^soEKmAY=6!{@mU zEb^s#^IZ52aEV2p*;AZdvxv-Yl)-aJwkXih!hpBpkg$4yzP5No-o)2I;cG&;nX@ru z#Uvs;!w`>%|0YYO#KIiE8gje7oavIPV1mQj$)YQ}&?|{|q7Ic2r}y)4YjGAdj}Ie1 z)FW8`FaD5h2von}uu_^fA ziUr2%<>8*6PQaM0BHyo9($eZz8r#poPh-b}cJf6`Zg=ZFCa^_s!V zpT>lL8b10WE3{Va7Oy7y3*KVB;R$IxMK2#m3>Kv6pe4*ecwzsE)cj z_L9{f_;b?LRdjRCJyt&ZC$Xi{g0RGBu=v_RmUW$g$@&B$KV_4o-^DaC=@(-v#QVqP zgazTtn?R`67EaH;k9{efXs*_Yrb(kDurm&V2A0845$}JB3`fVDuV}R{&eWdCWreR5#5R}~z z!C@z1{;a=1pc)}BFistTyPLkkn5e@b7hlcitzXd}2J)DAs+|h%J|OLCN^C!WZaeZp zhpvqghuo=+%)gEH;G_DGeE;4;HO5a7MCQF=zId0=ooomFmzU12XxvNVmN~($ZYg5N z^Cjj#kB394d)fC1DsbyaAWYzI37Vp!#F`bx+Y#a%`(7Rkw5*`}*ang`^o?k2;Jwak zbf9Uc4|69<0UNVAnC?Oa^mG+t4rZ`?{sv-s?fHP^G|2L39_0ojB3@MT#Qi7<$Ob>>|} zAfrz!gd<@0ZEb@Zr>(W`J+VPU@z^v z_mLLW%t62QpR_no8)Qx#t2_0dEH=6~5Xl+s=9#@7cwv=3Q@=+D8`6@oVCqM5ROA7f z({6)54k7D*V8cb|f> z#^MRd2r8zNo}Q*p#MSG>Ha@0L7I!mK4m~8C)_vj~pv~XEJIRAS4^Zv6&Dz9Cz?^>% z=qCGII>vhv=Juq~i%xU!*PtODmj^s!6N%C#BeZJmemXSvEN#-bNCbv*$ZnX4LV8Ic zci#=hF#J>KzBtUZGdNl{1ZMlZo^De)`)h zq3*HmRpRCOn#xE}+9b%N^0h*EGIS=ax^o%QZU=zV$y{RkWDc2D<4r%WCQCql$vF^$+2&K}#X05?ROp=E&(9yjm>$Erp$%l;NwQlCY)-kXc_ z#2K<{xPoq$kw7aeBie_-V3|J=-yH}ep8s{Q!nPVv++alJh8<=%<|JU3R0)drDp2(! zubGORpL9>j5xD*%6@J8p)9tGcLrr2atPo!f8~FLm6a)U<;g(^M`C&dp*C|pb&2hj^ z8Vf0pZJGb9mJnliBND3rjQt{EitGPqV{Jtkb}12z%?cwmZQii1HIQl9Gaf2#mC(dB z%}i5}HZ|1a*`)U)%&WZg;I-&uG9hsZ)*3dmDgp&I#bzmkcZ!MLYI)F@`;M7D5J%Uw z{73wjIH7J>4Crqwg%Qm^ER8Q=WS3qcuPuHs53VXwtBg(bw|Y2E?6RdNCV0?Z|7N=U zVkfm+A&0-yjxw?>QDkMa3a&0mq1zn)k-*$sKD!o9>vKgx^Y|wEt>`N~+$upTbyhI- zZ}{1AeiZfoY6wc27m>6h?ceZwuJKsf# zOh_N26S@%;15ZJYZx&s(Wj62INX4!=MZCG-02Uw&Y3!Jx?^v5H1S`Grxxs@GT$AEQ+*SV$lKHa| zGoy=!<7G4({fx?+d8q3CIL>-s5eM1cTr6>hto8zAoZ+-E-wSuO>lo z$!$2kuoB+7mePlUU}lX)I&_Msk=&+-a5-&?Kv*LdcI!BU=$RsNdS50iIRAuR-zki< zR4d?0)>3!`7s%B&$q-?*0zLa5;i9(7Ah+rUnfWN2p4GAh`6I``DE~5Sapn0aO=ofN zTLa&#b^^7TNpMGLDd_m`q$B*@Mxg1(EO}5v&bB7f_r3XWskjLOcWQwC#D^gDp^Vk+ z>1Njc6%e6Si_nH=l=j?Lgua?Y_*Ih+G4Y9@X%Gg1ClA7GH97Qo6VJ8WkK^K=ALlM8 zv*hnRPjnh;hxX$?7{{u3ocq$3&X}iw76+DM)eL~XBMNv$d@<^5&%m$Gg>l377j)gC zTV`YQ7#!Lik15B(!Fb$UvZ}Y47+&R>`Bu_cA6iepjaKoy!9yUg!sqoG63Eu`0{;8E z7?$5oBY&;ZXx8$1T+#KpaKyr`?wh&^Yb`3yBB{-4Ri3|k6*qLaa?gchaFfCfddAX~yE-?N+fDX!>HOzsg_H~&jBEh6 zTn%pb*){aXq)6g&e;y}1D34QK*0STK`j~4kIEUW9Cqn!JKD%pt1Ee#?3e?}U6Cptd zxz)-0KXZkkK42;;ZoW?+o_ApUGk@X_zE`vDdIeM3u{=aEdI{?55@i# zLErbK7(z7l=hT~6&y&erWq(sQ25iRaqlNo3Aw;U%2AElKx-bDYV zCg?0N!scQX!Jlg^SSo3m`|mJ@6?{(4?W#0L6xLa|Zn?uwOg+j;KI}kC^GIsucON!4 z^un!s-temA5qWpIk#6nLM5gOJd3>w|BmV)G`zFwonfmNavnYDcpWlUoFVVhY@jc8 zJ&NRf0kOs~XsAwxB_88ZFKv|m$}J-XGrodDYaKkZQ4$QmT3B{71pkScW5SPLWMOtK zs*WpzO-FN4cxoG+mbL=VhG|magOQBk`FPgqR|JIfdx|t!Lp)S`8gIqiBaIp}=oY{0 zsF72NF`7DH{HcKYZyzQrN|Rux_g@koqm6f6Yv{+U^_<}J9}=q@32DNUxO1ELV|@Q+ zSie6G@BC1rwYEGfd6F2BTojEmuWCTn|0D!IiXunHKEuh)^69iLC-6HnJQCH3)q#jhmZVlKqVIMO%!ZQ=SRNl2*<#aW|yq(YblwTU$l z*enh?@>hwpo*%YU=|SuColw$a%yrH8Me~)LXi?u$va z!p1|G>?Pv7tO<_rJ_6aBCw!^N!Smdl|*__Ri^0Uv5EV zf{Y-iMi#t1Vu|o~eVnysKbF6cfu%lvW-0VAx5|7Qa@;V_+6x7!eoT@DKhQY%1s#^$ z3R9o2AfnBBShPM4CTTt;LEreysk=N4npTLfoZix&gM}a!cLXLLYodAklfib%RhX8t znA=?Nf-s7wU~SBPES~3xn^iX9%*PV2PV_pRXt5t>yb$4fH_LFJ1FA^j+UcCXVICI# z^&`7C8{_QFPNes{5OXN63SXG2!6%PSs5PuZr;WJ~^U?)gc5LHrwauh`>zv?Px-1wD zEhMF_=|uf_843~u83|J-AoqjFsE-DfmQG^T4)mevy+>4zljH7wt0hhrXi?joS;I(P z#b+kz^qHL;D9pQyiX)CJc$v~amUq!jIuB1SsDe{+%ejR9Y#{UAlfNDBv9Wp*e(stK zM>iyqFH-WPX7N+HD^!9Tp1PJ)XOzMNuj6RYU{@#E+YS{`7oqy2I2w8-;`9|Noc8f) zXyAAXM(q^1#_@h&oUs?4M=io$h=DEAo{+Fk9Hl*(25-{;_B2DOW)uVu z$l;M&L$tu)9b+vXL*B1?y%1c}Uk*@ou%RT{!S# zE=FkBQ`zCC*eQAwXAv#b@-yt4=Ul+TqMP^(r=jxxbewpsi@a&Oil(4RT>q3R$JtUt1?*Q1|^`C`_LFh7YmDk<&x+kVzKdnsu2Bn`VGxzs7<< ztcFDU7^BSgv*@O+g=3Yi@YuMe+?Y{O6x_}M=`>5uqS%0l<-Y*#@LbOI{T%jFjRLqY zB3$RODJWTGM|w-#!6z$96aFo=tr`BnmU>BuZ^8`brO7-bCU1HIgP!dUG%(4c@*7Uj%CE-!_lFO>ZcC>dJ6{o3y(Ic~cO_L{I0Z#| zBuG2&+Mc-C9j=(nBQrCHndZ^?Ft9EL-{nQ&UDXBb$litA^Oh;x-1l?29Yx>h(>XnK z(u>KAe}^#0xcAaMoeiYxG4B*e8HYAj%JkVt3wXSZfCZ0IsLKm^FxnuG_b&+MR_#m%ZS!M+)=(ZL~Hhx}~oa|5$Q%>d0(TIfY)J$@A# z}=vH}rIl~3_+gFksOL@=A&gbN`(<*W@DV}s(c}|Y5Y9#F) zJR82OgxYrwQ|njfY5%bSw)LYd%slRb56>j9fhFf~=)XJA@@Wm|>6DWTa`U;49V*-! z1w(F7BpJ5;r;ke7=V{GW1FUf$!hH2nENUK$G8PtKA2bdBwDI5jiU!(w(|{^T&aYeg zHU}SG_(DBHGH`}-7iAt8uy5V{(Zr?@#lt6A6{pL3=H~H#J+0Q>3OciErCzZFDW2Xj@Eo zhtDAE*OoFe?mqN=<#|wU_yorWI^gEXw+uTno0LCFW}J;@kh|K7_`|pofA9Roclw7> zJV=fcz1xQxd^c8WiZl1)RW;LZCJIvq6Ny}3IP5h(L2Ahn$hr3re*fG?rw^OM+T|KKtnRY})@UgA|tSW!Dd$r`=r z6ZgF&PDT%j#L=Cwti6g(k6H)HPh;V@Qv@_0tb>Se-5@@7988m$3btoV;M22c{#^0c zB29Q3`EO()x}3M-B5lzIP=r;40cqUB&mhv?8AAp z@sy?`JRiFYtsfNOYPJC!`rqQGf$JC`Cy7oPi}>EO0SXNV&`?cFTr{(V?yTE~Csrln zk~4Cc;+;r7E)3xAd{&|3+rPopge#=~>CecmPU9iDvCtFq~j_bQ3U|DV)u~~E! zPJC3MO~-E%3|d25Bk(u zN8qp2Muqlw<7$}&ROMMToi|;UTeWN#do@{yd%EQ*n&yOZR4I3X+f!)2p*gE7!TJw!MU7doNC8f46owz08N#+>{JKV(pYB9 zJv-dKAOhya zYc+nC9tK-(8-QWYWw<*vgs3DJuoFX!;abKPe4`#ip7o02ZH1RWs4xw{33m`ry3p4K8LLnm56C5e>QcY^vyF%pqnn@jk_-JrRg#`YF1Cu%FmH+ z+k^%8{0`zL#YQG6^fFGn?9KGpRD2CRHtg!z;jDQnz!K#W-SZB z`%&&-Cb0}X6P3C2WzfjP02bU@UfTX>LzbK>t< zpPiF1{r3x^VfB*gEIiF=>FUDmqt3AJ1q+6Yr-AfsJ#03!rPpShAS&GfFhNF_tSVud z4?%0tYqu7sEbo9qaz5xQIE2UTXMyB?aonCQ3BrzJ1aI^t@sRy(qF6Nu4jUxkNv;b@ zgw5cJv=w0Ubv_%Zx(gP4=J&rUTRC{W6nED$)FAi?xgB9dKePsLRu^w#)3YpC(O*ng ztg z_T}Fooj10-wGb!JoHZiJpxTe)Ktmp&tk6#TBu1frJxkubhjU|7e3}OD5@c zy$>Z;>Ez`h0c3nJBKJHa$*P2jaI0q}*uTCF%_F;@QYejzyby*7Q3h<5P$xg{66c+< zK_pdgko@OdLwy2nvb&FaV&?fh@Y*ek%rB3Ha>H7@dXwR9pP$BE>UORB`Zog0XQ$(4 zHi|kL9K$Q_*U-929sSZ{sG~+Gp8oic)*P75=p2}U?YX)%?zc2~U3rn{CHF&6P$YQs z_spRO*J=2|Y9fd^f&AtGrAn9L@zgN-?mrD`S}+-Yo^EAao8HoS?v+7Nbc<`|l+;vib?)RIi*~No+NH!J6w&

h|Ps}0n1z~uqHfrwfqm75 z>@0MG)eqGmL8OLihw+)SJyJxW*Nm}0ISx${&XM=Nois&9n{KxXqrcN@8DbSr;#z)@ zMc2pGt=_YdnBSYtO-$*>7oI_QrT84|*^xmv6quq`WfYq94dPW3G2+Yn!k?&rW%Py$ z$ruxJu7dwR0nNh{;vazL9c^d{m=5hCDY$1{F|}7dwRG#HE_TCY0cu+kdumlilQh7zM6EGy?B!FBz9X}X5cz&DbSzTfe&a(oA;;GP6$u`h(qGe1pdUHeTAlW17pGlo-}9*ZiyR&cHH0x3M?3D+dX zU~SnoY^oRL-VK;zyPP>HuHv8lrst`1P!DO(ctL#pQ}{m9byWCrm&RXKg4f=voU{8s z>fzYJZb;}bPd%cDE5BWUpwG|9p0jg7s)oV3pkwrA-w9MSHOKn7r(xmY4l?a`E5C<8 z3^wO~U$@DE!yolg_Pjp#Z$cL_9(99ikuywD)@01AUO*mPGow2yqo~W8XfRY-4dQ!# zv+jCAT-f(ixa9F0N-ERI{6R5pe7X%z9{fQYK15)f1EF>GH8iDg8+;YMLS29UM&s5u z#Fd|$?YPamNKQ0R6LD#3_L84XDtog>_Ilx7hgj%25s&guSK_p@$t0mQn5@1(ft5MF z4{8Qip!W?MbXd~^i|hwUT&oW&pMDoICMOY32~{f8k^&DjgUI*DJ9zBnAI#z3&m*Q2 z1pn%?Fro&rHB%YV3}@q_6+!sR{VptWk3_~i0h?VU1=kZs&@1@{e|P;2AuS2y+s%{s zD>a?1%{oC>2dqF*`ho=B^5#BwyP@-rI8f$00e8kkVt#KX-7_$QM6Ihs#qYaOT~D7w zza@C$YBvPuEax-UeT>?x`M9~w8dFn`(J#$6z^I(RE9uDL!@I(;_HGCi&yRpF^RsFA z!!m;Qo8WN4c(j+fj$?l&q1TpVYRj`5ye=o=;q9C8iSJGv|1A>ktxAUm(`FL5dp%wF zHwwS$w!s%839{hRbUd{9IX!Xoh=ug83FKgTA6E6*FeYnG!5ay6xIHNd*OP-Zeu)+a z>kYD(ly|}1{qc0=Xa$+_aU#Jn@%UCqhi=VpwAj6HFU)Z7K*{g}c&+|3Ic0Q~ypBHy zwuc1xzPA9ivXt;aS_LXk8^Qt^57M=15*lz{@%K|9jQ@B7WefQ4MX{Wq&L9cp26E|z zK}+VygKB130sLBCE^K0UM>?6ZX0T+HW~SYi05)L$oVN%LXRA6Vo z`1MUx<(dKAlz18^)lPt*IbLw^{X*KiP6A%H#4*`xyg+M~GaXkRg{R!!leOm$!C|fa zuw8xtFMIuAe{bIeF&Ku%I@Q=Gy^_q*;{7(U4``D9UGPqu3{k;;ps-&O6(8_-&7aDA z2Sfs&%4{PAXNswgXEJbyZqlS%_RQ?FkHBq08f5KwM?uSsMja5rZk|_|^OB*}(?W@v zumv_BC?>}r$J2o|=@_=~Jkc<@%sg;&LdO0vu~3|cdyhOO?(GwSD3wD@O9#2gpBbxJ zAG|VMjT_#SLZ{xfMc3ELbvyj!%=P-3apZXhE*Uoly1G~5X1===E~;o@WG%tXNSlfd zo>ge(m`F;f9N%q>rH^jJpv3SpoZ~2oHx^sfHOd?T%LBWhUH3afc_*OSJq>W;T`-dk z_d~t09n4#oMA!E{rhSS=s9l-{-bZ~vwR_A+Ma3=bl%0fE6PHkhL!Bf}b2_eZT?AJi*V8q8R^7OI zC6!kmpf(Sd5Y=;C)I_fXTEEs4b)VTNG*^|2URR9qFT06{+g053;W51znuZ#EE6Gcj zV{}KfA9cU(1L|3|Fy_D-R^2!nH=pie6<4gq%Z{S3uEYkupLK+WbPaCJ$q*vKpIOC- z6&Yu{lg}!)v0lN#tix(EcIjXu+C7=fRU}$)1wPtb(B3N~z~CSZ21PKsr==l4yAEr2 zn}gcjn1PwDOfhnYUt0sKgTSmT=$G~CC07;VvOlbK=m}!^{ zFEbuN$lF{HdTs*W726Yr#Fkz?`V#gc6R2Tkz`04 zNvA8c`WfHs9ZXcibub#Jgz#;j2|2PHCV$BUsk6t}>!(ZLi>f>rCZ`iGvkOr1`7r!@ z(8=Fznvs_bqIS0gp5A6%9owRTGRge?obTOTnX3SI)<`m$$v22m)KT_&bvKpvtRX+G z4H&ndBNmROy^Lz2Hu2Bc4QW^0VdBsi-Vt#S!o(huZw1%jS5zbX&8w8L8w@? z72>aqC%U(l;PCv(WRcrL-sv=pTl)Pye!f+L6Y?_9z%Y!I#u}nnnj1RGD_U$3jiTQ} z{K)&=XCPYXDCE8o-e)|=qB%zyk!P!#jo*w#8~_) z{uB3AEkM`AvD|3I0kFI`kLSrP;I3A@CL8iANz7x8z=$vGn{}EAk6evS{Q1(|>pSta zuO$z4-C>Alnj}W3Q}^g6u+#7oBjv3O?&0@vN1ZG;s=b7Z_(pK#y)=FOpD>(kH%1|G z3a#tA;pwj9Fn4$uLdOQf<7-n{d$o%+#iNDa|M6}SlbIZGFT<;Ep5wuj?QrJM8rX7l z960+f5mda{2Ek8|+{l?PxM*TV#=MZlLr>hXK`MZ35Z+6@?b^}s)nd+@@5X+)#Lw#Q zibAo{I8Jv!9s^|#Lg+VXL5clp7?N;>NtQ}bRuGIG-3iF}m6E8CZ8*!(4QmAv*kj<2 zx^ld;Hr@hcB>wPwOA~q{PnASQ{U)3*pW(aki0(R3$UZi|3*3ovxbYx`t}RT0C7nZ{ z(KQUEvMKniqnV#MoZ^B^Ou5_UdE73~OLbZ)NQ~em@%~)G$hBu+N4OgJ9m=E+lQwdW+%T0CJ_N?yhf-gEJ= z)D7n4jxG49(FlJeY@};9*5WhSCOmQdAANT`2ZBzUk+EYmxi1Eq_)ILG?`Ph_=AzFS z(h`P$=Om!{GFJ|@BB^VmdtIWqKPLW(fNw9vQGbGFT_8WN+`sWP*;1)(@pWS!++N@cc(fLFol&e7V1c=)PBn-YrwG zVEjU!t>MhZe)^8eikj|DNoI)ZOLqiqCC%J5I#KTTYO4zW-cXFc+hXmcxSu#xOtAj+-&A4^MWj z;=1)uVxE6JTJF7onOBUsl%jcpyH9m-kNYV&=dcpWPY=_S+&lPkQyDIoSx!D&dIB>n zEn#`ke{5g=7Pu!hkF|SyktAG9hDO6b)a_RZ?QM+3fcP84B(k0q{Pn*@0*V2K38gU z5f|Ioqq~E+{u4c@n{)_|F0e4iCvy3KOi@6{la>7DPggw**q6k@EP2QDe?3gp98b(BIL{_-nZZ~6F-NIg{o91n02lLeq?rn z$jR}7*XvJ_$y@ay;LrugtqtMdtsm)&F_HrPA}>0=wUspWC<)rkwBeIDl4-h`pfYhP zSl#r5+wMnUgG>qbUH*(#%Zl)pS{Tm;--2VcE~8Jc1U808Vv*HdBA>?3O{<0An4Ua% z&xr&nw_$4JaF6u!i@*(-`Y4+F&YgjO26X+}1(;ky2G|+qwJ1{tz zX%lWCVqK@<;)x{s-S#s1d$*9Rf96QC>dIj4SrKZi>5n?9p(s7@2#YHvxxgB0u77tV zoGnj;@|go*^Hv)AH&>zix}&7+&lotYd5-Aaodlakc5;tI_i*igZd}bAalsmHlw1KN zc)zO+Q_b`s{Dcj>9OdQX9g<|JqpUzLQ;fbj*N<5iKQTMI2CnBu@=P>6s+Si+ZMHlm z_PS2cm6=Sw&r89ZWqP=3ha#WLl!yG|i{Voh&&*oTPo^kGvRSGJh<>O6^r~NjwGDMx z$T}iZdyoG&U0^|{KW;0yMYkB6a;n!xP`X>5Q&A}+N3&JY?k2yVxv3=B-Bv{xkO1|{Wu>^ zx>Une-`B+Mz&7sc{B!tjSPo-OD08cyYjdX_|3ZTed{(4+6*qt8D4svuKz`lY2iD7V ziA>up60tnj!tqHMO@A3;kq{k>vy~|3-0xxAhvblBHe-?FMc8R24`LS%Gfzw9@$a73 z)OJ)Fom2Vek8cVos4K#%-G=(LL%4a9^3gbsHSXwnW?PAok_~Pg+~+(Iz&9O4 z>h@A-h$^M$-b>+sr}aTlQ^}4yy|G?{;$- z^=m(*A1k8A(+B=U_zaIxwewDjM)Dj1J?A#I@4_Oq-8`w{kD3y}b{DR%DP4 zi+6bR?;2S=C7)8a4lbN1>JQF8N5sl&;Nm5`23V)Wy3h)0={{2+0Cp-&QzY!6v z4L%2k3ytU*huPe(N9H&^XcI1rSxX`Z{cxc11bh5f8Z<9>K`nxdQ2yv%Zo&Fr*ew5; z3~V|>GR)?KyvS*=){;XrtxFI$?gNSS_anMtBlMllTs-PG1N^TnW%rzm!i1=OICrH8 z29BHoAKr;07IYoGjCj_-_bVX$;U2^v@}jFfN~t`ogsui_$opl*i3{z8HlE3*In5gy zwr&7YSV4sawU~81lxB^s=DmD2@X&99;L}=QuMW;dn-f==R6_@B3*+defFkD2!)tV( z=PC@gdrujri-d~#pvISKCi;^P5kBAwm6%BTeyM@4`8aM+ObCOrqv400tYE*9TV0Q) zBu<J!=VzuGK*ak$Px z4y2d$W6#5FOk9U3^loh@&wIbpFNdTAF`*$aYOoGc`7Gpmvs7}jq?3qliss5Q=5ck4 zexaqyeICQm3BUdHKy>F5`aJ75eNUutcH$ZONBkM_F#OAoS@aj8Yr-KRU5DhCjG_Cg zYQb{+I^2B9lyj|JNh2r zuUQsE(`4L;cw=mamE%uC*1B5AuPg>teGgdw+n7t=RfZo=8ldLWX0r3c8Jv8wi}om#qno5WHhs9v zHXn<@Yhu?Tn>$hWm(g_FBv)lVR500!Vaegi*KE7Akjku!@!%c=<>$w4EBu zd*e*Nr~ZG6&cmO|_l@I9_TCflfBXEbq~s{=^MKVnUXm9XS#FVpSz z8$K)wC0X0;;Nh_WWUmOJyj~E*%n87h26fouD=g4SG{W%EE6B>v!feS8Xe&6)gvi99 zTAdo4l6Zr=Zzth2_tX4wAtG3~cs)qeJ;s>zNu)w!2aFrL8XjJ6pk|3v!N_6L0#OTusIrSR_;pP|+nP-*>ZxYds(Gc~es$=z-6 zdhRGGt38OWmYVQR>N+tm3B^faFDr&CO7M$!3QX7C2sa-!qfSQ?ZoYlVLOVnQ7HkXR zzq@7Vsw~N}j%sM~{0knu5)01;)CKZQrKH8x7|!g>C4pz}65ErzSaW?(FpQTa|1~O; z3Qa#U0mQi8Q62m;?=ZfY;*Y1Y#X0YuLA1Z9jp-3)m?lUeDbLPu8cxS?%(C0KxAq+* z$+ARi#x=11cLuBdXF^=7w!r;F0WLYD&aJe2#C#klAxPJ7hqu8Dn5jR+Ch!BbcnLv! z-#*+aHvvTJlHg5xQH9a+{g7wY4i9%Vk-Vr8%vPx*Ycs3qhwTfX=)zHQqAmlMxp#na z+hRDR`4=sOCUXz?{jk0LgH+exJf`fBsTe&}i`g9wk==R7Fkv-?kb#&h2_hv@0lDopk?#qa}P*w3Xtkn`shoECNvq~AJ-Cr2|am`j%M zM|=^_Ty-O}i@hMSG!&cE`EKs+XiU!a<3_v{@#&2N`0K|SIJSo0m3g2Df1X{XE7Qkg z!`uS0bDss*zoHTiq9d_{XY{PCEyJdV-*Cd-Stz|g7n`TcL$!4cHcg8oLLnw>O6La9 zdOQ&|>gyqDj}qN7Fo&FMaWW@n_lc~s47U0_BmI@f`Tq4o-26dE@KV+pmyT-DPaE&x z{lR=x%gVxgo(>qfs0Hm#<>8^dYv|zZpAdVl6>@iE(R%^n$V(jX@dt18GR#A_Qy;)0 z^C^6b4H@eb9;NlG+K5Km3@ZF)9+&?19mx{ZU{q^7KHM;o^STs4hF_lL9e)BmQQ%9A z?FO*4!hqWy?T73B7Q>#)eek-y18$_t3O@ci4~C@x!KQt?MVlT-?#bP!5Y2=e;_9=)I!eD^3<0yFb+pHK5%=BsJFd&S z1@njJaS7+AaO?Uu;)=*rOf%bB5&WLRsll?`Q=5CN#=}S8u#@-7*lWYyqtSHqtt4{I zUa&esMey}~G`^J(g{1@C5Lujt<7XN%T4SV`XCI~$_1p_kWn2uK-sIz~>jm&|BHzX4 zJFZWf_My9^h4}#AX^1clVWXUaVQ+RDJH;~vgMISo-yMzUWXfmJUV!Nhrd-?g0y7<5R(ssrAzWE9D~k$vQb$`sx)@CMtqzam-oqfj>g8-_c6g@^b3 zh>BP^=8QB$WAPx2bW@6k>ZqPO1^fdOar2T{;CuNJ7(Mu9F(zprD2D~Y^lOb6U|C7^ z%%V}FX)1~Pz_T@c6a= z1YXCS_!_j3UW^GT(p+xe6>^N{yL!U1d8GaX18ytAR|oCPDi{}7)4co^P3*oaYEfwbnGL7k>j z+&1PIT{m+btbeJEmp_l^|JR@6Qm1uvZ2vjTNc~I-@^W#PdI@o#IiI{$R)>4d)=XgN zHZX`ufy>G-$)UBp3$f%E76+(ur{cAW`^%bdVXB?mtE_QODqHXb?dk9TuSpnu{TCgFj; zAYoS=92w#{Px6ECxpgWulOA?6o`em)oA5zsD&1(uY%~@Qd5bbYU^zFZ}eK>P*~j{g}#j z1=DwT*HX?tfuB>TvZ1RtVxF@u%#GbiCcPo(?6nPFpJHM5T_M!fTWw|(dV;d0p15N3W3hqqolZ*k2_nA6Df*Z&hi(HHf2uQ?ihziN`_xt-Kin)i=J zU#F)wWLfCQDhoP)AsHEyMnn9>;Gt|V#LB9`znv3kiVDv@s#*ZDpN)uLfB*urCxP?| zJ?75bNU%R9$zE0GM>5Ul==-RRcs2Vh@o88H0Zu{8QI+Yi?D14^l@#KxZVE?zk9kBO zWJJfi?xJFQx6u9jO4#!r-gLRP9yqHyLqn$+T&s_&u+Df-KJoJkGddeq*{%hxnPON> zl0f`HwFS*w!X;1U{j=>GaOs>~P&j5eJPBJ0%JXMoxok2$cJ4i$QXnkwN!bR%KaW9E zpC7K3^1@EFw`lhKHj&w%KrP?sv7v0us+27e25#qh*IyUv*=fQj!Hh>2zrNe zfy#7&h{`j3xWWg|y)oiq@A(rqQ=Zu_{TLoNSdbjIEV}%GJiU5E1SS|8z{sbwcwe;u ze*N?2?>uevR^UuHE9(yaeoL8Ii^dYu^BMG0-EQ>TzXZMxIWzf7|IoOkT3}Nh$$Q}_ zJiFW$^saMsWvu|yCE`hb{VF!itOo6-n3AP19@o6g#U0*Hsd14lN{{5yEBn)k_bMqc z)8ls&I1|<&=PS*)FpSp9eD{Fsu<-oBFk7k>X}b6f#$kLYR(t;;%M-N_zc0t0!gS)) z8cO_C_wa6m#qfD!0Bdo}k4lLrkdgp-ygoV)hi<)~C0rOTIp>FM)taE~d<>8F$Z*+z zx5BwMiz?YBqp!;M1w(j-^slBRj%km6kc&&x(mfs5| z_tmgyqaw_%)x=nf*EC!b$tRa$i%9k%_Rfoe=%WwGlc-7fd3HLpVO$}Vyi$rH>jsGY zaYZ&Yei11Q2@4!l*u6Hf7`&k!61g0!qM^-OHny;EL81Gqs{5h&}Py_GR^u6T|RFKEiiKck@45b z@P~0|AXCD3R4>ulzmDJ?8&67?h!FW|Pb}0s2-lUzfXA#*yf?uL0)y7VTh@f!OSFXV z_PlGM_dcG#t4{S?Gug=SDM)lC16yiEihcKB=-VeG`n@#M!_RMZ|@Q=|#Ur zdg}H$&{(*GiH^NXM`Cxull)8MhwDi+l~N%Szj5?IR4m(ZJ^`2rb5JDDsmTWic1`UJ zR7$qTW$})%^H(FY!PtN;&Qm4V%gwN;Ulu1vP9n*NmI85qMo+XZgZdN!M2eO$qwm#W z&Fg&gJ~a(o8LvjKBz8e830PbVLa4X^i{;kS@$#zZ zb0r!a_9TMp&G+;qdcu`a-91z|`JYeW^pXmk6Z(*}g>0fV);BTv;TWO?rF_@@Cd$3|jouB( zFyJ*vf2q6S+|RlK;(Hp^jNf8si>2WFfAcu6F*eZr{5 z{SF=vQWNMa&*xsz2C_4{50s|Ypv^ZsZgfE?4w?-Uf1Pgp9VbA?;(SzUzl~kK;(}+V z@1yDQ_gMP737X9UvGH&duI6{Oc3pgg$3C=ScZ!feIyE0}J$p=_e{>N@PV7aQ;MeGL z><5gAp38l|Yt2R7kLJB!8Cb?==EK)hkf+O%8(Wrg|18`A6Mo^RhbeS0{xi>P*W>C( zYH`~#3Ly3#Q=-!Op43?GK<;?%mG>EZ@HY*-gU7+7rTp*2xfd|>?lJBUa{vs=AR}=~ z_(nU8k>7cl_h(+l`XP?rF{R9}IsOoQ${TgOuU+x#KPK0;pRjN)JY;aT9&Bq4J6?)aQ%hxrSz{ouh__rOKfp zI}OuqmvIZaKA_RhUiw(w(W39ZJGNGM3ZDEJMLAD#&U|hTh%KH^C4VMy&jxvy?a&@n zUN4WYv*rq-stZWR9p2Tb-;1vlFVhau7^vg125ybsLL#4tEY#5zwpnZ z4VfyKk4^!0B? zZkB!kiPZ!y!8#X9*Di$7V?j7?d_1|-4g!n7m3ZKp0e5LlJ{Yc);;gy#Xl{3cOLiY7 zoef`5Si&0;O!T=!TscNJYYX}_YjJ0>F$(mS3f3$V;rz!Hp??S?5Ee*)thE8USk~cq zF)5TN72{gVDxmg#DZRO-5}xTB3O*Nh+ip(hC(dA$!a zUFtFIr5GwHsbYIwA=u^SVkI*|7r1H&MlFuP=X3HX?W0Z;56v52R^S>!~N%})S~4v8RvJI#P15A`W?m=UkbA5S~?D& zRL&6i|F>4~Gb@(%-8T_fP6xQA9LnE6{NRVAJw#YdrfW-bct6Tra_GJTD(bqhS*4a_ z$w&s?eQXDL&t(J$gTt{Lzwi#f`*^D3EIE;X8+CVg(-HUQ{GQ(uQY<{5ZfoJ+vu*%$ ztRevq)kmrE=NAsvk#*rpA`NlTroNYm*jh*S2;SSRM zF^nu-H5MaN8_A?6{7#`?D15E2BC$smx%b`07@Ay3Uf4RK@x^Qm*(ApOH{X_s$mGFE zoq9-Lp8-GO_wfIxXM!CaT4pz}Tudv|x>6ayEuiMxnTCN`pIv>SXWy9O~kuYsIUFiMQE zq4{1RWX(Aa3~t}2JsBz_as3yX6kSF&Hb=4pPUY}ra2{PZ@d?}slM~qP3&sBUIJ9r^ z#kANnxWMx%+<2M=?l&0RyqMoNewskawFlYRwS^d z>ntq%XWlk@RpVa~-4x1r+kK&eicoCk{8`~mugVSnUgZ12pHS|8FOsq-#=yabHZ6@| z5-hWc!R_NPteKA>^$(rKPv&+82gB4c7cCB*+$8YjrV45{H`2jvi(t0X4Lp;!9PO=Z z*_R_L;na+(il>_wf>h>iGGpsEdSyustkqy%>vckk(dW3L2Rcvu>^A35-E>PfnC@nI}m5QxS$B4GYtA6WY= zf@F9@rADH#wZ|K=+#gJQct^LpB=ou}<1Ie>7gE?lohM5PE`GWM@-KOIe*}d)iM#3O zoFU@9uLz1x=#!qxI;wW_CmPTDg<)F%@UNdTb7-0?4DV?pMqkbFv)~o;pH&}?RX7Q! z26f0~=PGc$^OO87c}br84w1ye|H#Z8p=iiw4@V{(g6Vf&K>hJ17+d*@EK|=T`uv>o z^yUpHIub?=6c)2-_WQA==n5w4T&Ym=6-B4&_x%2Y5myjmP8O7?0Sx(*A>NPGwM+#3 zQj+1~o?NuijAVNvR^qO&Lu}{s>x5N)!?sMxVH-0{aecNq9qu&6`Cp7_N?tym`Zk1% z)s(sR>@JiOe?{&2fk4xnP;$?RXTcsyq|JVpardE1C|97*1qMfRWWE7B&G|?RTT-A! zw+Q;aZ-$8VH^ALn5BmCy92pZ3F$cm55q1 zxb6W1E=#VH*eyzUPaVX(o`qb6?qu%bx(~Q-+E~s-VKvtkV$8jr zUBK>_(1pU`GjM$7byz)_cQI?3Cb0-YgO`ABIL`=f z5E8t}c!t8tF%<&GN?dRHkCw$M!{==NH(0w79QNgc&Q?WGsu?GkRGmt;i2etm1qF~Z zX(tTuyDHzu$aCwqnsDBFc3gcn!cGMj2zK09PI^8~gEMjHHDx=8K8ZAI>_Y@cTF-gI;vD%sG(zYz#W`hNNHketBO?0||RH z#2TG`M(NNp&|9nqk_R$jG+du5#NH*^vn@HL{3keN%tCI#+Chw~dxx$La@?5gDCldu zTX8yt@1!l7i8E*KCXxF+*q6M|bKLL|OtLnC>*t0@lgk0n zC%7Vx;8k-PW=(Xa&IdVic1bt3-zb2Y%2`a@yCLSW#30T-beh|~;}I6VvLHw6zAz7q zT&Z482&B9{57WJ6@j>M}+P~s5jW_?zTB%H{V27OH*Ww;(RehC22aChSz0ZmDLJgj8 z!t<1N#q*pdZT!Ta%^i+uMEC0xI*Gr>%o=+V0*Xc9V^bi#?YWoVXHw^W9ACgaUS`C} zA9=&)Ok2T3zmF*$yNA~9Jq{~*W?ZC8AL2C`On)rRS;|l1eob)TVy7?RYzB&Ppk@Om zch{n1+Ym0=tVd+|oWm@pikj3Ib4ROgz}hC>jg^?q{CFlI;O|z{{+c8a^XnmREBB)Q z-82loX$U&C+FW>oBmW)eU7GDHaq7G{w6YI{^q!YgBc+EVu8D+zpI>O#k`!_xX%F#L zT!p`)cvoI|8?;x2@y@AUB3Bm=mFXflSA7o6J{jNaB}!HtWKza{pH&D>cwo>=dqX6n|R^E()mbK4Itld4wC00c%kSjd|nxjDGMU$ zvK$+TKI3Nba;y*)4f#gmY+s^7)jHxlk_pQ*!>HuBCce+^1-^zVOrIp=^4)^C6tx-;`n)KSe}~< z%$cqD{JR+p*Vx1UDrpQ4>7+kh`yqevH{PKbNIdGpz)#qWb*`9;2ivPb_w6c_lJ=s8 zooS%Vf94T#-gv*{Dfw?mimNi~!CQsjz)P}{_U0w9vxVl;b+_$t-Lx?@d>qe^R7%03 zo4fF1usYNm&cnm{=h)+U*Wk*}TbOobD=Ve@1RBOTW6RPfbobNUq~%uy{+%^St<#K% zf&4eTerX)1Y`7PujZ?t!|IT7UT{fJr=|$65SFnLC0rhNSdf~hb?K->&JSU3bqpl-V zIV}{fE5u;A=PW9b>qrh7q+sUc9-P?V0{UP7knALXdgo&_q`bL`i*9{K)y@E-BqWV@ z(+`n_3ra!Lwu@#>PDXP<2MMl?#>cjo>$9hD;949Z#|rWEo-Oz!_%XEl z&O<@Q7;qe0LdMKU!CCdvc&MutD)eh{)#fl56Z;$V-~<+b$R$mwp+NIhxSpnGP`3OH zxQva&iJ!ON8;$=UGSd&!ubnNAJCFsx*)hmI%x9AZ%E7X76WRo{u{Sa@V768?UiB}< zmFDGeM(q|&j*DZ<>-la=RwEWG6jSTJBk+FMAA{ZQlpC~0!rv1+@OaM*Fr6e1M>WdP z*FhbMuY3UUYsPp)Oa==5`(Y(J4io>xzP|{+)N>wq&$zT_U%E@+tISa2c<52d!y_P*QTd{7EA_$;pOL^ga~5P+vY^x#EFj-MB7$7N=s zSokOthP%A6U64ZtE*HYF-ZS|6!T}snl7Q>JYPcg*gUq#Dz;1V7#j(+?PI$ zqQ^t{U6*Jm`5S@xj!`6?&y+XRTw{+3HcH8jT2s+gcHo!!3-?;2;A$OXm>hipon2gL z?)y9tdlG@c!k+kAKOCg@c(J=L%aVr&E|Yx&GHm{LCy>75Ll>9IbHnny|F|=lTdZ<~ zY~De67Nj=(nAL-?N29pkCq41$lr<175>Hh4%(8MM5H`k)GDSb1f#o(~R8XxT20iC2 z^dwCAUAs3_c<*#hx?2*`kFaQWy_bIZ=8RvLZluzScau}c`O9}$C;3yHNHrFn1DRX? z(El+7JYW03H<^bhaqS>n&7VV`CrFc1;WoSv_G!wPuSmkJ&a6S1l>18#3PL&Z1W#E-n!El+19eC=LGXI!dg+Rgd9@FF9k zIA4h!E0CmHUn{|y+;Vz)NDuE86+v`Eg1MdE99X;UHSLeJMBSB%`0vszxLaSy^A+Dh z0@DF*mA?^u590a>4{)Ko1iJ584`01(aNmm0RQH!5hA3^o(WT$P&r?}olbl8S9go0* zq-64+m>(Wb{Dm#sw4iET8xDBoGkNN=f}W{2Ky?3O{GF}D(i7>pY3C%m^!@}8IDRG@ zE%%VBe0!qS6VGm$a1FgK4&$6>{y1r46uHg3A`uJMk(WP%(e2nc-0=4VQH*>}K1!-% z&1nJ34V&U7$trZ;b&dSHYR4I!{)O`whS6O+qEIP3f&6+F#hg3o!1sL4W6jqv)Ok*5 z(ZL1~=W~kN+H(bq`!B)dP!7~(KUie!7l)}G&1l>hhd)=QlbWi@+@uW?1hsdL(4kr7 zn9(wydwJEE(^1#No&LcHI)CxBR~UXasYk7eM%<4*PjFY$dBmF!co)f6Vxn{jH`WVK zJ1Z9|3l)iG#zH2>qYeb6ezCE$8_{GME`^a{YL*rE7i{}Pf&hI>b zu6#>sIRkjNQQYG2dL>wL*$bu}aE0}WeNcC!4F24eCmU|?OksWw?vT76HCJyY5xi4d zTW%g_P&tEZketqy+10Z?_sl^z%MF4wG)wg>g}H+Q6RvD+FCJR|lr&^L!=m{4FsNEX zENu;eYrM-2;&c4kIT7whULYT=H8EF80)mG<;H4m%&b<1Se6+44zm8lYeG6Cd^CxZc zBW4z!IG)C5nkyi(EDbh_@eD4l25_<-Cn)jhgopbiV3BnyZPAp073E6YkWVWLdrjeN zrxjzaxu?a_$Glfn)|V;%k^m>G5=d;d1$yi%gJ51f8rfW6{;4e)CQAeml_Sfbb!IXS zZSJAB4D7I1Ly`2(%7zm&-=RcTFfN{Y9dEpgWu_SN`AgSI2szeI>|dQB12>f|JRWby zeV59~^||d-TK6oneNh#kH+QjPKfA%hDYe8=I0OS9rJ}>*OniTA9Cv5YHCD4_BTP^g zfr&O_u}aejbx&VL&-W#C!hwabL1i88SR4hbU5dfPr2(otI*5e_zwUGN20T&~#l{yyd-ce(*}TQzZ>n;$cMlKna^ac0X9$xd26vj=_hh>*UQFU7C3G z6kRb>fe|?!MAOFoB2|_bU{d%Q2-<%N9%`H+YMdT$q#l|q3(1e39n`)y5p=(bfYp}U zY%}k))GNP0mf6U{`&p|&Nb3%?1l+)akqhKr<#-gIT!)FWo*=$vKa9Do1oC(7(Qxk- zRAaSaZtfxA&FGMvkVY1J9{}cD0O=2!OxT-?;NzSBVAnx@4%qvOv8q+33S(x#zN|Ll zL7uWV!}Zv@j662zzB$MguOu!edFZmX!=nFDGPcUE=Y&@;;{H6+;}TY?u!lcgz%)lu z&V9mdn0s?Jtk#}a(V|w2>XODdd$A8bootT6Uk3=S>>}Tq&oDOe4^XD$E*5ovh0m)p zp)r|3>07N-=CUQm_X=@gjdRd+*<(6#RUO$T8cKw%6(3lD$O9NR(sdRpR+!1{zPb;iJR`)@kW|6meY3d6@je zy#W$}BY#)X-sKzV2Zs+9qL!^}W!3-&I7-8qlh^2`Dp#yF5`m!Y@#OZdG?Kd^f-2;N zQaPyrI;CebiK^d)yGS^?e%K9-uQTy>KohO_<^-=*?-4J4iQ#_37;fJ2=X_@M4Ch=H z<(Y`pXn4{c58FmCen)iSl-g6|9`_KfUw?UL%36q6y$tI;X5k8t9L!GoN?O%xE2^cp z(7Ul2p!F#L3#c+TEqN;-^}x>dBj_y8af>; zpm?qstgDnG<6CXv?eeG3p?XcdY=`L3!+mT18X-GlLHOHRQu~d){~E2J`2;&?`56@X^*J%#|=e z+2w@0;k2DQQT!5TL=`cgXUAjugb;FVem}^kiQt*^HPmDWf1hfc!8KcK=5z*Y;AD*> zte>3&E`{~mDu@WW160YqBR48`wBKVUn`SZW z1Rd<7<>8$OErS4>CQ&PmK! zK{7czD}(Ih-HheteAn3E8J#4xfYtuv$+J>llC94!l8IY(m#4mIq!rh<;(_PpL}9la z9NfABN|txhMIyYHuu_=kU}}Nm$xozLPOrjq_doJ2q=zJ%dZ6i{hp=MRB*Dv~o1na9 z8k20e3HArL!@mhX8Hr+Rn!3B1L_3T!8*ZJZO|K7;O+H=FZT<-YC0rovPAWO5YKg^j z73nyeFjhXUrp#&QD{^eS6wENoC4fA`$>B1ox7-8=l~zKZehE1ANecEhw8Htccl4a7 zHTm<>7&CJgP{(Fhro8h!t8FO94Q(yN#i?qXZ1rqh8nqhRXU@dc9!})W@H*I-A4D!6 zSOo9e%OQU-4koNJBl~YV!12G6$bSw3oa)`js#i^e`iZ&~HuEpg8m62UNhiU+Bzy2t zbRq)FOHklC3K1@P5N~?~Wz;TH{JEK|zq^Cx%nc>I+J3}%tQaRhV24Y#2h;vMaX22i zmz1U5VlGe6A&W(X8230SJR6ry26Wa}B;TI_)nguF>STuZly4#3$B!}@9gFFM^B2kS zdEzi@8$$#AEtnA#4rSzJ@%n-aHq-koJK<&^^Rcg%%>EHbUIxBnr-=76E1Yd1QeifG+ZaxkPhS4RL8#6el??ovRca=U#5N=i=7y55 zO(8Lia+4J9S=7nSsPMPwv}ftny36En!d!Z@CWa~AyPY}oY?N*$K z-hqJHo1l-ofhBeoz2nQ#J~RBcX1s~tKt$4HQvFQ zGO1Ww+du~-`-tAot!$zvkcD=O$)Ayai!*oh`5j$(Iw?62oXZq2plgHYf~_jc&>;F8wAmO+((bO2GA6pbr5zq zo7H+R1NDU}AmKfo{LWE^g{HwwYsMk6_v}(ux8OTjxY?R6cKlco!uK%GJh(>Ur~w*e z%%&~Fscg*ak0c9Pp}t=4nuWGY#!Ruvs zHED=;b}YeJ>f`DDtIO#Qt#Z0k*OOd)l*4pxd&$U8J%D*-EAX+J6gsKIVgc`3+x2(? zTJar(9SfeY4~m7bWnnljd-JEFH1jTm=y&3W{KqtA##Qom`v`Sk_KD10l?LCo*`wo@ z-Herf4vd@lk!Q(rOj7!N$bR5M_r#@Ib|_=u!FKZJ;XE+^ z!85bGT=7ML23~IYPWiN(xzHWJt~9DE7xJ&B8{52zOq>|)Q5B(uPZyKVSHu{l!F%LF zVF-x*`A3r0ZpYxYYw7W3QJB`HfIg}V@Nma!s;Os+DftZL++NcQ-#cl3jtyg`)y*^+ zNuuh08BpG-3-?y~68*s<(lUQ4TW0*6_8kmoFvAamQ&zx@@pUBsvjln9eu+6{a}+kr zQ-c>-o=-6R|544(STD&I-;@Mgi>p7wzzMhe|1wt+{-dv95hYZtL*-Z&m#6{A&hZwuTcA7du$8 z+ym;jD+;zWyoW%=ZEW_k`EXr?T6pe|!VA~)`0r;2eIdSwnjU*e{<6zJrSLS~JF^(C zl~hyBl0CF}HiLZ@J&fb*8DzG6PQ?`YRqXP*N^&S|j36xY0i3)MK(|hK%f3^(%H+8I zve2C(4d3kFL)zFm`0mx9MNOwUU9sUa(Ri|qY_fSs?s`%pUttUulJg-dQxIiO@Txjf2t@{^kQ(ZmzBTQ z!ESnVmh9gXNydNC3-Wvr`d$n9DEfh^KElw{M#=Z7q(4JMr{i_J|QS$=LeGSBKdJ!BvGn(Esljy)fGqetE72>A7ccLYOSh_JKF?+O@e03cqNp6>j?3M!ZU&~t9)$I(u9x+s7tu*bu z;LiG<^u-6;7Gbkf8nfN}I-ZbH!E4r;82ZK%+NVw7&(3jlqSanZoE*jOH3`LvDFl~h z%RrD(38vczV~m@dd9ZN{`aGMCLtA=Ks4fE4hiq`ltYmCwv@5E9{)2N_zo6&h33#IO z1hZc8Nrt{UUhvy4Sb4e(PAESF-O_%zKl>@{_iu+jl{_?Ut;7Yc{V3Nmi7fHxsaUpU zn5qvsVfB38DP9wclT=5Uo4sZ9{muc#)=yY)pzszfANdIqD`IihkqC7Bmx&o;Kk~1w zo<8ps=H@up!Q00s0uA@^f}%GAV4E#2=-rwEHMb%`_(c+Q9@z&w9ip&Ncq`d1z7IEs zMnHO6BD@SNrmxKy^1XQmB-Uv_qnL!CY<1-Sb7zQ&`$RTx9Ksp#19+flHwK4IZKtoMmAhdfa zezkcGBS#0o#17$8&lNUjxi&s3JcZ-a+EAldm77qJiVq@6a2NmE-1~efx7I3&oD@9| z3E}JDb;SfuR?L7KX=uh={qw}{j}jd3oPci5E%-d-H;Rp)#)Z~4pxUM$6d06X;{g&5&Fsp_UO3xzKc-!f=H^T}$y%yQW37rZ=dbqz6mG=8vl?$` z_OyUIO~GK3G7oB-BWYUY1!%c9+dRe50VU5*Ky#j})v9$1xE*P5Q&o;tOkaW<&wRm! zeq*__)|c6}J3{H+8$R@=upRAh9$>3P{E1(zGfq+%ixQXY@yxP6*#3y`7L=D_U0NYN z8@mGcx|-w4C2=Hz1fkHTO02a@MGWR}wYUn3R`6XpBT0eg$Zt?n{f3X{2jC2)DV$wd zJzCso!J&H+T&b@%o4x-g-MsM{J$XM0oiE(Q7atzrfBK{Jy|@Xnd>jkQs{?7&x>9)Y zA7ag94>VbP3p3B%!C}X8d>$&wxxMAPSCwAa)Dw#$Id3t~KMB^fm3S^3z{cwy{&^xI>|sk{?~HEL1#^_mQK_lJR?QPfPZY3LSAko*bPZe4?C z2L(j^WC$qO-hvBytx(+t@M+)|ye&8;p`0$Fj#33vwV=>P5FZKc5odW&0i01=Y9uQ6oP+C*MjUSS-$@p!3=ll^NyD- zI6-ecPMr`6O9T&rbqT@)pX5p6>>!+|kbJcTz~9zemlTg1n+94=4OffI*y*n4q^v zguTnp*9{+lam!RN_h*Qg&_A*#DU(dyycvV^oapVyYVsq+9J;<9B(Lc&QYxBL;nA-F zQmw||ui%R3r!=F?4i!!`x{#f1oCF#|mx%n)4pKEUlej9}rR(^9`9*pRKlv!pLTx?| zNuP&O^Cog;M>o*JKda1GK9c+)z79{~X((kTkot#Zcs3Kb18Edjxh_MUpK9#*Dciw9 zA(SR%J44#?`EX-69o>TV;J$*(_=o51eyCSKlOtkal)jHT@R|G3?Yr1X#ba@$whIV# zje~`rf5`OeSup9tJm6Cpo?$&iCtE5)EZoD*{fn4)Jo~J42H#av=^(Yw?O|_&9kEMN z0;4bw4BoU0^zMbj%xmf7oJB5ispc8`Mg0A#bu-SIWCk0QCc-1G2*uWV;Ec&ObkOw+ zu~gm8s%3}})xU-$D_5BIh#aJoZ&uP_eQTQeemiyjR85Y$jpsd?+tJjGp!Ers?Jp0o zkX){g+a0d6R(c=E&NANRCV!B`{O&QA8zs2KN)+n<@;TjIE>mf#+j!Bb7L}vs@EKqvHF58Np{q6!IOQZQ zvOEpV9S#t`)P;nf=2@HLn~82}D0?DU0mejX;G>wY^nKqtp2al|LTvc^T-Fpp_Mse@ z)H(+PRhP4myXQcBs}>b!Wx%)25e6Cx=5RQo6O>0teU*I=RRhQ)i@Ch_Kms*acCrHysrdWoY}@ zjgaRrA_$zL!ri&jsV5e4vPKJvWBbxK5+v)#8~jITM)s$R3{5 zzoTJ(1Ef>kjNje=&5BpZ!P5Wy0Q3W}$yFG~1W!cmRg3V|zc=Nx-<*UR(YK75UKrSJ zd(WH-a-*dycEPB_0{Fesi~FmOyq*D`Wa+Ok@U6F27iOv7L9>l767gMrD7RPs%RBRfj0Z)4z$y%$+sPcLUdk-g| zGD$>tS4(`WYRjw=<1_xhKN!Kfr94}879LTUORX3YoMk|8FPMU$%^POk$;HDPlQ1Om z3W{jhf%dbl=(qd~rCH^Sbdd}mXcWWKv)!35AERj3>6!3zUpMiz>Z1B5`Ap?2fw+<_ z^g#0;;&?Tg`Y9i^@YS40+pBI6mp|gBMXmb{Gp!=lgP+b zTbv@PX!T@Cwe z!$EVZ4eYF#NfLgExR{PPEH%EvPBE5i!Oq9St&VeQbpcwZ6h^} zR*)XmM|bW|qcU&Z!SYH78^LFEAu_m48o2) zqow61y%+At?o)^%5ke;9K*4<6^=n=#NRpkI$jCbPbwnCQ z32AAm(4@XnT7KvE2Y9`_p4ao7bKlqX`Mlq@1@RDYEDcUX|01r(Z1K>P@v!UCcCukr z1c-f3u5qr)qg`c%I4W`?S~XrKx>L$YK(G+fCwSm#c0Bg}mV;y69%S?}FPL_2JNx>< zd$NDhE8^r5fhXHP;O^a`_y%5L#Lzl4>ledy3q~;`3ufUR4;kXOG79^;bOn#hzd@HF z-{m|-nJZ!g0>K7PJox$mb6)l&|9>rne<>1>xipkn!u#TPd@rHRnTc@Lun_;$E+XST zyoCMXfjmQF4<_ie!4-{Fc-QMHk&bjELakjWVk?f)9#Mk0H!--&L7GwE&lzSfc%PBD zEPWkO_5yPZC3L<$xXiIxmbdTCs=3iBH2Y+cC69%>>S- zWMJ$4dS;P%WX-RhB2-u<3N3>y_+L&%;rP>3;zJ?$MefBbQ&uph8U|1@yBpP8+{jmL zN2++^2tJ;`&(ZrIl2v7osKK_kboH2@%n$iz*tb{@U}Yc78p_3ng^6?#(+OukC6Y7A zJIK0S%c01f(p%eYdFF-~`B-ob*FBN}%Zg@j+`zlE)~)AvxmrSb3o82B@V;8 z6R5xJE(zakI8Tj@^WlLrKYLF5O3HOY z;owOTu4m;gSZ3V;^7U_V>KFs~EO8SSdq-jZ2U&9Kun5NA6|!-)J7@O3c8zqiNZqW!+) z-?$OHXZwe|(ftCxZeF-OI)_>)KPOw}%W(%8Kbr0N1~f8fa-F}!*z20ILM>ZSPVLbm zlGR@VNB>;F+n$#6m&OF)zp%?_8=p+>uF40Wq9#0fAdAgsm%$~iC>+wBj#h7PlBFT8 z`0Y|W`{#x&$gcZ|gV%Rsv3no<*w5nJDVIs=KXrIm)r!x}j*{`48=+)R35}m&&Zs_E z3zq9rP$ANcyyf$hK|91a=kdR-=1iR+RBgYF`9tsMQRNGC3+_OuD_WidT!fFrU!>3I&wly9u zc$~l>o>_VPod9%_$8eW+XHkp#$7HOa4gdK3U{pURfWC(vF=cPu^!L;D}S5XOW~fmC3cV`cK>is&{H`5cBCH|#d{64t7sF=pQ3S|C}*TmSjAN!1K$W^!Z zz#eo*`$HQ19kzh)E0|!+_Up`cCpEaH?1b~TSzvl*F2J(XD5;N?jaY38dyuunbfAz2wKgqqVMHpzo`;$G$_!n&)Q(g-@|t;$s=F4c7U6c1n2U{6*4Y)@jLT3II&0vPBtdN zobLvlvyUTmt0@bgm$s7#k!v_{_%TXIM3Z^T`iOIKG!3jB%Sv+#Y4=!Jl316|D!gAr zO&2M%aTl7gaZMTS9G2limVP6O zhL;Kzq4!W4G!I&1^`aygGtm`-xKzfL4nc?Ead@`c2=f-&0UXu=-R33m)u{l3c09tZ zNyqWtnX3>QxB{-9^MGS^;-GbvKiu)GP8sK|WY7KtfNRHy=AIfDu;K9SV8{ImxU>#65)v07dS5F4l+6W!SuvT zI5Tq_ba-ar-OWF7*iMVP7g0jrq*RetMOp0lE`En)mq{N_?ZZvHKhw_{ZbZRh>WAN$4r7XDU#5a@(ga+oWPqa^91IfC81;d zE10o0n8YXML4EKzp`V>2^i5s{eomusT+k@`>Emr|{4*8#eiANJ+e~ugcrL6B=j;PehDSYsEvVmvs_1elftj@lUCFN)lBXcN(Xbxlw8M+LQ+k}Beog=I#Z%LbaF_p2g0?A$(CZn;MN-fAk zsGS3WfltXp8!@^b>g0JJDwX(7&`TY7 zFfId<&Yl9P{TXobD9;wq{6s2rb8Fh&SqitE(IRv$yF#oFBDF41?;;g6Y+MUFVGV3* zRfC8G#4G1{mt(|5OwnEdU&r~v(x)>a&v!XoS4$&_FhFFJr$Aq+53FBfNSS-u&^WXU z`i#3^H0ARM&(4Eda~;HrcQ7kve5U5BOUYn;Kk+QQOm3L31D7=!Ab7qSuvnMithJ!h zoeyJW3ZSmAglc)Z(&UMh>KoUA&7)UjhomV)zs^R#P(>VlVzIz~Wf;z%d>S*7CAmdc zM7aLz>bQz$aCJu%VR5(wcPO!+rv7N5*Z^Vx23dt=V5eF}%xKXIAyExc)04GY^imc;?2J49V!qeQD9~=bebU>$OJ2>zx>o*V|9Jc}B=PL%x&C{Yh}}@m&0o-vSp6UQ*x4WTL+@5HGrmaC+$v@#nau!p4q<|dw`Vi(z7GZoi4GJc_#1V9*&MZup&n|DBg+%c2j%-4!43 zWI;T7Em=v;_Vi+Hn~XKHvslf_U{EWeA zsGDc~IO7jR8IaDI1|BgwD4kkMidD)HF6^NlWsM+`EX{>&{zW?!{-bibmoTDHjyvbf zKMMm*Na@zY@b2quLRvjul-BtOvnS#72DAG%XPfpqRjPg$^q`(Tbz^?M=s8q z3WH_MG}Eqt*!)(Yf)RN(SELU0x{SG`iIX_fXj6J*VxYmR68p{Z6TeOoWeX9{KELC8Jm^co;SopmJYYYQG9 zDg|XyfsM=C>5&8GoPJvdHsxQWM&d$zsKK8(q7x<&lBn~c$u8Zg;7 z7fn9#S%L`}yjQ`NL}}fGYqlfMlAr~mZ$yQA_q-uhbt+uwZ#OF86kp@XvpPP;Y#>C% z3wC+lh3tO>p2fIA_J2XVLz8FYYV3uTZ>wR!1_%7CWsRB=Jv`S!8I1Mxg(Kv=K$UmL zzYWd|0*z3u4 zFpl?uKN~z*`EeKM9@dnZICz5=x{MdjIqN`X&w5R5m%ZYA1>1P;ycg=JPUc*0EN3nx zD{&z!Wl{DTzxN$hfcdw1cJ2Bx`00T$91AGost!$q!qQ_9b9xAyUi8tGrI%=`s55yV zmjnrQ)iiazBRT1Q9u}^=#}x2+6}^r?EZuxH^%nI#g-@M?zzq9wUHvU$K{9JwU`3mNu zX+7G${fO(v-6iwB*0Os8e{pGlF4Ma|jTz_a(cBu5Kw9{kXJpL2!9Dk_hT!8;!c8H+ zxOC;$^hj7Km8+4$u*qV=!DTbi8@|Cu>HF*f4LvH%pH+Ln@Gf-HA{tsK6h5gQ%kK-0 zlB|k+?!Vc9tL%gE=cZgPAY&p_z*QJMk#L{6f5emb`FVbyhmJj#^kzUXpMO;3v|4L%pU}r{$GKfKydNeco6fr^E%>)QjRqCY z1-n`gE_}5WtH?;nGv^3) z9S-J#m^5-JJSa1d!xC#nAtINcfqe-{-J@hze-YN9>&XJ;Gz z@NFuiTf$o$x2I4UTUXXOXg?@eod?OuDRA$jGIwI5A)b18nw|^Eg{1Ny>|uunI@jU? z*P=8By5hsFUY@!I(=P_V2|*YSU5gSfSDg-(2UWQSnuCW$2`-3l!uG~?u5Ri@xOVg; zIdR_}8``BH;LAryD^;qM;$1Hu-umTgPX2#wPBrf zM#d=OnIp|@{C=06cgUK4@4p~8`C6WJoU06Lgn_hp>kH_-aSx|vE5fD)62ciVn|NmY zIlK|B4r^}(;c~i}RKI9p5^Z|uxh)#-P(BByWk$k}^A@<_**D(j+W;o&^SSgc7xeup z12dP~aoCX*xNk$&M6&NZb60*p$lZ5>yS-09O}dgxtcV4(5Xw%<4iu`4E`j$oeBSs# z09od@4_S`}co`Cjhf@d0Ph&awwaJicIO+lg*>RZhdOG$@X{DEHm8kfd94_oa8EH*> z1vU<8oK!|MDXvanBM+bAEV`5NpY<)YRTm?lIX@z;HUV7+bnz0u|1v!@huFFp3KuCL zY~uTI-+b%HSJO89HDM>}UA)DnoIXZ|_wFP9ysKGso)XS!)d4N9Ej<69yR4n5&mc6oPcT6$H5N%EN zO6M(-e=(h`KUs|zpT-~#?8aS6^YEhHDTtO2XYTi`!F3bX!-LQ#XrSecM;y!GzC|Kx zMDhOGxv%ijmX~;=D+qh$6frwixRU1)(THnjfvAZ;weU(npEsTCtPAM?b8ge!7qZFJ zhB@%ZiO^YBObM7siUpNtToWdYa-3E>ti&D=CSJ~nnYuI(cXG#4gGvG8^ znH}S!aP&?u>>S<%+T|h$PHaugIiG)zD6y! z?C<27~7Ev-ifoZC|iOp?9vCD2V$Iod@i=PZe%XpvY`K*){tN6_B7~j2zou8j>n~% z$g*ihs4#5>j&|v#V#bGIs*@eEQYqx`lrGp9`~(cURD@sIUwCeP8QB)9O+LIHgD(zO z(1dekU=YD#W#2ZkHYfrv7Vp8%F;4Jl=TB%|tixR$V~VT9#zOFxT~2GKboP5X!sm5W&(-0!34aGr4 z_%t(->K-*;DZsn(I_v<*a1 ztYID3I;x#cTJ4P6#bfb5u_yGsWD}8G;l_>*bzr|;l;E~&@OR3nK$sl;nH@iK4U{=> zjI3A*j>_$(M^zIcYHSRD&rrnq+OMhW3JFr+XvlW1TM2F*HLzx(G}dhkfZ>}3xFy9O zZ#+H8xD?-{3b%96D|H^1G201*vo^Atvp(RLePY~)vKl_qltG$|B&oa!&ptbB%$Up{ z4Wch<2{ajzXRSrdjLqsu4d&6SEt_$t*a#K1EU;==%;K^Y5f~Zp91c3iB@Qg8nh*U~r8t#3lwY zt$x!uJ1GGiTeW%znF%7h#)ErM942JO|R>GPy*U2rbacFo=sm4#z4_|Pjg>rir zGwU}QfkbKmX@D2R#ZQ!o-T%ghZSZ4BHwI|xPkI56S2sV?O zO>{D^q2@G-#v2f0TnJ7WZ%Fl(=E2IyA=t(5$V}_b<0{VwsG^sEn?C8|)cGeZ-4KBxwu@2iOEddk(SR=03joJE@w`pNFH6kKif zh1u?COcp2_(qkVwQQ>?OZ5D}x+MsZBkmFg{Y!Kq-d^76 z!{5~`uis==3@yOr8*}KT`uXsw)EAe|JAiDQ1UQbB6Fv;FC5bXK&`W&+w|~J?GG6To zJu5j=*lo6#zA*ZCS~=rZ2Wmo z_b09(gnCgbsibp_7EAKG28U?IMu%@kAJ^S<075Su$T?ShYoF+EV3FB zt`ETxu9^%nsi-)qoNhlc8)mz2!Cwtm;eh&X{BolO*S&m&76-nN;4xy{=mKRtCLy4? zuo|Du-Oi9b*6`|H5xFUu4h})xxZq7RIH;b2tM>WeF=dEG+sfjbRmE^H$xW1e=RG{WS5)EE)5^nWWGbef;L&!K8^bR*o|@- zRPwDqv_4-z2h^`~S_5<8&WJtMeQ*<2s9eNPDmyX$s|0EdWKi$b3Di4t9QI%1vye}F zsq&sFLhq=vkgAv}c=TL~Thx06Y#rqw?!R$3%O?`P9h}A;G?C@__L=Z;?ReqyT>&K8 zuMUpvQ{znb-o(4Vs;KOHDemSCV|0(+Puri~CvAx#7|!R~ekkcz7VG|cCc>sTdn_MQv{ zsRIiz>E#VtwfQpMPLCzN`7QL4-5{)2R&oP$*7l=;VjQ^6?7?YYgZa*!8TV158RtYvz#@J&x9GnUIDK3zJ5JRc_I*>w zSh;OvdYvfKHPT4bb}ocijS0YRJAu9T8Jys;3%*Xhfd8VWqU)EgL;VDmubJ zt1cIO=@nd?yB^;hH^7ySSFwmw7iQmA#9vdc17qNdy8IY!p0pLt628T;&*afVy$pPm z9;59m4?Zu^gX_(cs1H9w{cNTV&%`G}^`x!fX}bv9E}ekGFSZf=$lchJ!teW{-8hLy zw&c8?jWSYM2|KICY+Ncm$}7U;mV^p|Kc53af9$Va6*1}+p<%u ze$v^Dlu$|eCy9F{&ei#A!jlwNSbtXzxQbR9t_8U5WB^w0$e=gtv&i@D{#3z^Wz4?* zS5w%t9heTDSLbpHhsE`{Lvz!~38^_;d8H(GCvhR{w$h>xWm4Fe7FuxXwLcu4{E}E? z&%)~CF-)TCYJ6&^L~o9{LKhy)!%(pVtW=rGy_)$A2QNEv0^w}P`SzY@>lHHYYl~|J zW1=BaXpP15#Dur_{mw3lOq!9$b0l^K5Z!OrVe`XsnDFKcj<#8iE|(3tr9C&fQ=*3f zJT_uiWQ(9`-AkJKF$_m}cGDe)w9spTJ{&k?go8CQ*sa&He@@W25q>n&-L zttC$FeL`9`DnZ8eff{djo-Gm*%>>%0@_yVVc%u=;6g66-dH;F{`cX;F=Dx%`uhv0q za4B`oFh>;4fu5D2;8`4nPK&R=UmL_xX>D|CUMWmeOrzz67x3&BRakEjK;wd^pflg= zJMv~5&D|)*EoqtqgR$;Zz2O!PWG#laU4Be`fhzs+Vl9~e7KOSs-l#P_7~b4SCqB`t zXwY$h&n%vUX^}Z-z1aoaX7>`APnYoPV}A(QC?xq(XYq8r7`PrOC0C#c*UH8cnP34e z@c2hjW-ZwiCV{%lEn>6P07S=xGG5OY;;XhkNP3YfaA-?L@t(y{IaU%EEj)ltDrxY2 zC7-AGaR7xOS@1yREm>3V$UY9QGdxO#PwA#;PX1{wB=On?jGFc{|P1-%fR9bC(z?VHssCK zMeDnr82)<^GIEtslPC$Udagb1pL$&HU5~j<$vMvVU-WYTE zX5$0;T5?=jf(=L-o6(x_Zn#EK4c^l_L1#o5)-Oti;+*kN@h1Vtt_X!KYm(s0onGAj z=nqLtkN_%s4yVZH!R{+j=r~OAkdq;Ne3wrYH^pH>Zyr3}YervwzJ%v@$wT(xJ6Lio z9JT(~L$_5vMD0@NoWp}b!X|}I`gs~t+?GP6zdTL6e~#|S2!Y;xjYPWp1D`Ewp@sTV zJX?AyYI{^c{@e&62)~V=hFk%|Uc+zeGuXjC1dXPrWW@;;%uM|QHs@tvN$q9KdA%E& ze3NKX@fCD`un+q0wo{9D4j8W=Q)4`D96H|mK-V|30NWz4YmpFl1-e3Wx(fykY{bBh z*>H4$5|R2;jQi(!f=0+*47c{hg&~{acBvL*Dn7-W#XO(x=nK|EWf4qvFooCGdEZt> zIiHogN`DLA97sfp*wJx2`7~5?ABOW zRR6RWjkd_sZ>#j7tp;h6*KsNrzY(2%uVQ$9I#rS%$DW^R1=-0q%-{dK(XT%l9v*T< zL*-H6BUVLaYFFV+o-et<(1G_*Fr4Steu%vI1%4;LrLmzI%xz%<%(D+=%L!a|8sd^F_6rMumvTaR48 zQ4@loX#HZeCae%rrX<$8^UFlj;uBZmXq`dqWp~*Q(;yd9h#_ z83}&7(r8t$31@O@4qjNj5m(>}FgERn(<%|n-_l^7$#n&PL_Ni)?P|CnwvZ&WN($|a zmtg4IP*k!E#m6}hXm;lcc3g`MDsGpG8@UE&fDss=Bo)l~}ImhThj#szCgX%Ek}QYEN8sg;>- zRLRbIUVswt6*uB8zh z#DsCy=@2E6$PTEkg^(G+B*>}f1s zXL}^lY0~T8yp!P@@h}a+Gb$>$Jz^J85K$pB$4!8jGi6}RWCwVCcoH~woMzgOUSh?` z74p|wfo@egOrAxpqQ}}-vJXG&VWiL&mz;dTdtsI`WB>NB-U_pYZ9O)^@Sja!yP=)B znvTM)f_><2HH}SQag1#|I03f&vmmGL$f9IrC!L5_7}uT|;Pu@UIzLFmp`Gi--Q4Ml5m~gCX%ghR`QhWO z|EShB3E|jQMd6I6HE{OW6Y}}WTgu+>W^c-e!UVIYjN!T#de1+T&RB2}gocIi>$3p_ z?9zp`3u6eT1fr3FI$4(zhF7YRSl@adI_;Sa+RTY&9QSwBL@o&5&kM zLah#`@r(gIC>rX3?SB@7>z$cc<^oK;m?{SIJ$9}1o4DA^2%iSElC;V-RQ~6B`ddB# za~@V8YkhRSTTI^)Ekod&(iy`FYG^lhRElAVT9YQnC8DC@WvyI zm|ExIlSx@%dWgb8^KyMnoq^sgS($|liLAz%; z`EG88%htZ8ZP!y-u^qE8K3PeqUaufbiU@mnU3hRF5kWSn6@Z5VF z`>t~a^m!>ky{`llV)&ZIc{|ft=hSeQsSXpmXbx$5`Ha@={Y-x@F2ut4CHU=&DMWPb zXLowVP>;zSY?pjDJv2WYwPz~gZ23*}9OMB?)X2s@Af^DK&KTNR(FP3)wl+eBfrVh{85t~ms3l* z#Nb>#^4;$fNk%#0!l%Qq@bD_s?QP-jLt@-Hd275T^APVi7vf0QDJcGAiEP7qc)a=u z$t&n$1)3}H+<+`IW#FHsN^Kgg3Wz^9fuKzb_yC*UlFU&O?zv`%K zBT+6qh_WSNB;3EdT6|eHwQQO#Y%;C|@tQW`c>WkqO(~=>08TIJCr`pJN3v0X(botXw5VXCIm2=w*0li&U&Y~9{#bie-5QpPy5Ym%CieaDyfdPsDr+v)*mj6uPS*jRBPfqmnKE45<|69QPz}#sc7hvDC+Fsb;HuU4$ez>3sX=ED zsoGvhe17qs;rKgv^J^P)W@rmWns3ugpAyoSI)jtS=|m=eKOWh9fzCfTLeeiQQ`sj^ zNX}!P{a~9(%VuqZE!Ac;L}M0CUB8{5QH&M(_>2~=$k|FqPaI}{m3$&2@~81mR~t^2 zK7h*{68JpQDR%o+3)Fsg927PFu+!eAP{z}Z2<6UU#D~xH&5RkSH!}e?t51UF*$TXS zC<(9Kx(OS!BJtFke8`TO480B~*aM;sBJIil8@w=72!n2P9qw;G#4>hvBe{KKPh~J4eD``FL3}+fxPRYMa1+ z&PI5`@h=%4s!2cg4#Lb?`52jNgtHSllAbdGT@>3Pt|b_46y)%1eFVNrI7HU(c4UL! zuB7WF`MuXZ0nXsui@XiGbY-C9DgJl0`G+89Fc3x5 z#L-=D4K_86$JMd((TmTEOnR{#|31G2rw$d9KMpk*$?(pX7~T=JI0oOJ|0S?)aD%@e z9C1L^1cN~hY8K>lY zCUM$L`dv){8p4xs;<=@Gpl}X0ElXoEXC32N&b_>UjL&awmqv^XjR>!kPH!N)J)-vt-3HvT(o2FgY;#DxUC}jzXOP z^h=o!`=*>CI*Dia8E+K6+Tn*6RVSbtpIt0Jq$Tusxd73Z)zIfw26;PCjjqm@fN?J? zNODmyWk2QAc=>0seKr=b<$F3B3lFd@_m-0dB}ZuF{A5y)sKf8no9OhlV^Qi%Cb@Ou zEd3TDj-8p8>EZTF%&2eST?oI}Az}~r7p{VF&RP)sI}&9x9-)++5BikKfye#luxLXl zMBZ5py^@xY-yOk!?vokos5!LP$=J$LdktN;dlqWXxImr4x8UPVh2+tV$slzsmiYV5 zz?_e1#6@u$?wIP19gQi~Gf&B3*DqB(7ox|!YkIa*x zL}u4_Z8$BE2YZo9>dW7&@&~2iUFABoaW-S0=SiYbS|Gk@{lX-*c+uRwx>&zGhNhnR zO>SterOP~Ocn@~{@nNcNG1y_2Ex)-$F;GZ9}~Z$U>D zD}E<-vijuJW};WU56||FhqG~tP=4$MTOyexeDXeDczKqmP-S?PFwr_$I9mFE(8W0h z*S_>Xr3{KfsYS4DB#SQZT>`7cqflk!2-;~&;qw9sj9suw@VV*>jcDcb=xtNE(5x`* zxqp!s+VL!$_il9Ozh5MzfOjAEXOSI~`hm<)6INKYz>WAW*6iF3>UZWWMw{Kh*gsBq zQqr3KT4sSS7e8P>c&ZUo*8)s?(24_{7jS-l1>-7{^!A3+HG35Pk$vA= znBiZ|WYaRv?85 zOO6fB6~r_o*$cu=N8waAhRB2D}wOP%*(8~>3%xHOBXI5H^y1wv+<+dCi;6v z5Lw6falOkwQAIpJ9wiZ#9E=i9h`1~~95F|D&!-3m|FaYx z*H46^yAoJBqD)ubnLtvm4x;AZZM-@40cuGp3HQk#6aF~2PN@Gyh!(H(!PVp)Be&oQ zUY*#9{|4{k+`JLmy7-#F;lVJSrvYUBwoLN3dVn=d2q#+42WcuJiqE=Z85>1isv>Ny z$@?Lobd53EZAfK8uMW}^q7xyTJ55T4_pwn1M#z~xB)BdakL?k`>^Gt~uXQO*95YAg zI;{+p`x@9x%_KUb>IFOTY#Q40T>Z9Fo%m*tCS7>BksV?t@Ova1Drr(f?l_HsgZyvL z+Q>z?>-2QtC7TQMxw9(PNonH34o_BbO)oLs5lHr)-AZ=c$|19*g z&XV*g2k~@SHP$be#Lq*Iu-(#?lcq|X#`Q0xqE4FXv9rnC1(Qhjur5x`YG=Yu{=($6 z*El=w6Kb_*Rara{APiaJwZ0zoap!ld>q_xi(;%#!6v z^jPuP7xX*#2$PN^k=j5nF6zuX{I_c? z=jibT)!oIQ)^{6S8nGX$&$0Y0&jYM>gu;j`pUJHog@5;Y<84g|6mhtXPv-~X{WU!p zSZa?WkDBmQSOdGvE&{%P(!x^}3$V&>uXu$0FCNvL^^B)q`ZTrw(Yz~S zrS2Z=6&ZssU%V&frUvkp=i#M4cuSifPQnMK`KZ`B3XK&q@nY0JaLe!IGZ+s^g4t(A z&Q^@S^S@;NHmkw5ft8qDxR_)FPUTisb~5R|bve^6!2UJSDDi3;>!M$deH;9^))aI+lpY=pZ|hg+pk*JNh(AP=~u( z;JZkaOFp-rOFV4IO@H2kyYvmXfk?0qkgkEo+F&mN(!oH0p!zK+B zgAT)v%_`JHOAn^)Oe4AZ77+4y9horZGN~H*3d0w6P-}M|^7G<)n5FcQ->JF~zx`1# z#d#Ihw@v`BvpZ<@kr*O1y^g-|OoWoT6>#HiAyx|vxmQU#-0U%YucylZqj!44-bp{e z)$|kjr*4MT1I^&~hv!aKX~5rkL$rwJdbK!u(2dg~QOQ=Fg!ii8e`lr?EPUyML&VtarMM)ndo+Uio6? zqTT;EJI}Zt->?t2Bn_1`WE6!+Lp0QVoxe0BX^JSjQiOyuBBf1B(H_!NDwX~k_jTSf zOUM=>8Htn02>NoE5If73LCNV3xkJl!k+YANHXY*%# znyn4KGmipEl@Pov4kzUm%lJ-YJsVMX1SC#I;^QJeG>KEk%#ih9(PMzZ%AwRE<0y$d zzKA?@bOh_TwU81#ADzaI$H3?FvEgYy?rY}vZ1vOe7k{UvN}p3*twZqfs5-hnVdpGl|G#StBIt@`b#05i%S3s8=fwL9yL?xvUi+XvM{)jQe<%h%HskN9TT#QRP z8c=(E4%t#sO?nO%!Od;5+*$F7+}t!d?$}BJ{MkQ9hDZSsJM{ny`S<+0q`6$1Oeng4 zNx*{IZfvuW<|1w6xu=_}@$1bvuq}Q;oHTproP>!O(p!jv(Ow+yrHyTh7n!1@HDtxv zxp1;~JYD-O6qa2R6U^f~f&&hcf;U_q<}P=@{^>T@5_Al8l`CNTaxKa|0Lc6@hel<` z^Z2)XTw$|;Q;@abet%zv^1a8Xzui;R+B$)2H`C*6BDQe5gd9OPH&#$+Dk7Nwu$Dca zc@-Ol$8g$H=b~e2C$>k)2~0jM7qs>jF&?*b!S{?XbhHhyc~x@o8Ode&tVNO<=#TCpQxC{-bFR(C!W*#|7_a9DyqY)e! zb=+Y?&KzSrJC9Q5Q3-+F1vPMsTZYrkQ@J@u+Hm2mNl<01#QqqlA%8=9=-BMX>^%pbSA0(2r$9D;L=o5R3-#`BHSHv58)Ic>imB@eRz%$hh4?np@ zf@k~Slsmv<=DT2K^E@(nuRt(M>nGUWslvcxhA?BM8w4k5kyrZ+K<#lfa*RIf&Sc72yEkJ*I3c}fmc3T^VJRmdNQbN$QYcx zdq3p;cY&IA9!IbLoG@3szQqPAR&<)&`f?P*^qkSq-H#;6 zDd1_HS9qpb84epCfZY>HU>TUOjxQcGg^QjB$+TpUn7#|cs&Fkd+xZ*+`Bx6reYY!d@3 zq}_3`@m;2tAfZ1upc%iXh>r0F9v^`>c_zHqCm*;o*$q?vX5bU2={UMLg3q1pAx3^k zOg~k_?3x}pc%TGWn@A34e&L7BcyaswY{N3iCe3Y_Z1XJdp2hPz6wHLDF zi`NS93(6rOdYjQ)<1a~wKaPz$L*(6;(@;>_g9pfc>`IlPYun@Krp*nwBXK!Mcyz#4 zv19OaY8hQRD+{+uMbmGOCy^LWHw<1)pzM7W>{wWYq5H1V^vUu9wTNgw4^|DYcC2Q= zXqc5!a3scd;WY1i6kglUF{{Eo;r>7oIQZNL&7NO0=dCriZrBW0c;~I;%bjS()9Rjc zVPwW!OB%O~gO77#V3$caXJAvsX`ft+t+S-Lsq)`x>@8mq)#$_rhSxFe_YnUqJBNGK zJ+bYh12s4LMneARfog3BNfx)FJsJ*pY{4!%<+TmX;JwZXIak=SVQubo*D20DxD4x; zT9YUd0j>UJM$PuAa$C~ux#l=Kv^W@s?T-UU|D^5kBWOAv`D#Sd_Z)<~&1QJ!gbqBO z0LZIymnNhsKBb>>0Cm;&A;pbFwZ3Eq*?x4n}+r#oqxB{r*7H-p>G4MOlIE zHFv7@=?jVl2XQ+JoH_F!l4!PmHrU;tz`2IUV&9!9X!AOYV1f&%?++z~i+@1h$5av; zmxHo5T+RN@HV{PE67+wYh1cujX;Ioly!6c(espew1D{7|_UzmA92C7Y7yrm^q=f|{xX+IF0dLlV1MO!} zU$P0Gy(__cGH+q;=f#3>tqZW?ygU9kg=eR~*8^v}9K2e;6*Ujm!&Y{-kIWW6cl z@sZC?yC|UEd;_zl$xlh$#u5^qX@JX*n)2SNNunw3y~wg*e$0&Gs#8# zF8xpi`5LCoJ*}4#1RNHkdw+gH2SZb`dWj>CzMV6VTVF9 zcz>RKDC}Ao1D8MRLfzml7_vS_Z)r<%hjxgQeEBFme7%uM{|+W8ITxAIq83zjjDdix z7IY9Z7F4}_%g(92PSk2cz8-lKA zCGURa?=rUN;7|+iwb%bgbC>PJUg6zX`Smb0yuJ)C9J-4^KT>g>MK@hov>EsR-b^bE zqzNJAv}UjjGk;frm!A!-IM{++8!|A==PBN>yMzDjw?oZ{3v{!61nNaD64c9m1(o%75Ka-ERjQEJ5S0xWH9#UlsD;&aO<%$ENq za|IXqNZPz)SS9R4U;1ieo|g+)-H(PCu7vJ4i-r?PQ8?@*hidtrbp7S20<2Kvvdk7h zwC*$V!tOo|{5(ioR!1@G+kMS$G5=B7UI|?BaX!9$*$d(CK9Kd^BPbe@$9OMS<%-uv zVqRSWtnuW%%Bg91Bm6p;@ow%q4@b;hwjYH0N}mf1d3dCKJFHn(p0VL|HfZob^a5pTKHV)SlyZ>I$B9{V;(m4gy z>9zi=ZXFn550rf%o4p(S(p5=CJvAEW2HWfATxwTb>?qaxbE) z?kTt`t&J3Hwn5$I2PnIZcTu-aB93OBctl|V{K$Ps{X$2Xf-?R+67`YkshWV=eu?xx z@uTOo(}?c8HYjWTP4|15<9`#KnJViQuqm*d7|wZ1zfM2O{)qmJmjfpXbc&SVcf(sa zRd^VB46AX6p9y@O!!T-#Pq8syLJ>ks@!eh=3W^94{*5BkX??Ad~ zJB3}^697AR#NxEv`E$ zUf(FL)lQbZk~2Mdxe=E3Ho)+#74YlNMN%hv0+I_33ApKM^h-}Fb9I|PI*H0-=79@X zu;e^`QF=i;&-p=%{$pmz?0mGTqRc(92E03$=Sr`e54QcmSks>i4Q^4~<)TxN9sQen zavVAL<}w7!ucXp#@^IvnGMwG)N|P^jgMZRjx_hD&bqR@v$8C4H0L^Tya~;Jm781DY zKo-uKX9{JGb>vS%3h0VtH)(ErMN;3f)MWQYs5pKOg*!Zm$tEAtA7oADTc|S2euxs! zsyz0yaym6m6~pbF|AEZjd}tggM6a!x(6Vmu?&OC;?){u6T-fblTxFV0 z_AA#iGP5+X<--TuBykEZ4UUCb;S(W2vlh1Zo8Yj~GSDMDA92qiv!n|$h^@qV%Y%1mpyF-_zGm2Gbp()>!+xE^8uFA-Q|C!yeWhjL`{!a$> zi>A|;>k=97v?e-1Yc1rK=zwnbB>GrkA!8IH2~)Fpo|>Htcs_pyXU&em!s!#h-!6$R z?BV-TDUNs~HW7zDZH4v?m5~0P=jdtIvf<=4u97;>NYq%-Z?_M#wKG1^fg2)txh9`{ zf4-d^7#fGWpO%2X{Z;b#@di5n*ccp|po8CyCLy=s4?W>A9)=TnA8LI9Twc5jewHz0 z(lc%1Hen+QZ%sk>&S2EfA# zdi^?cqt=N+s~4KyzC#zgKP7FNR~a$6IZTh(5!AkSip2ij$^2_w#AKiEgZW*Outlwr z9I{b>34xlhbuf?K{Ig@x34Ojt_^*q4T=~ezUTX%CpF3f1|8&-%<_b9{XGHwF9C_~? z??#Wwqn%!5%$Q%7Az-X0QEFyLn~MmuFtvl((KrPRo7{OuXBy(Ye|WYemwGuqA_Jdz zSDlP3SGKPJ+=CAQ`7%f*FRfukznTz@(0cmYz#03zuH(1y1$Zmalg5u?|dayy}3R}U?F%pG` zISE-;ZfZ<5Cwj;X^L&M1@Ja*7-Fb#gToG4#A{C?WUd5W7^?01@#gmes%zEnh?1tbd zo}c24F-4LjG++z<^I8MXz19$!geCa(uPc_U$fTR@DTBeAD&z?f=r5$g`|s>w^87wJxBEO4B≀UVJ`2-R#y3r?9XD~QC9G{9?f@($pkvmL4?fpU&<=tTa zz750U_@nIh{ugY_%^tFB*9r;U%{*95OJnsDylIvjy8*vTU`+7Lun|!Lh29+@8gkILqrk`10v#GG6FB$-SM0 zc8~bKZ^)zmEn`rvlAjx{^|5)sUjZn;C$E3a#o9j>FjuyYv7Y;kMvZH39K5T9cMfi6 zVjt(@_Djbx%6l!XZT(KAbE?TWStn{Ja+HK5iSe!y6>zwmfg#Fu7@g*Z)(x3vt%rJG zX!-)KTAI($@a~t9M^Q|`B?s*64Q2&7VxT7{1ueGonQWC}#z{vBj+|1%qenzJ;Z-GY z@csaJrn->Ax|{!F2@rG58;bxqFDOsGqYgHa!`Ny0;lu*rFle>yI(jT z?OrON$Z8*~c&Ujim6Kp>mq2j*Z@nDok8uc(OZH)ff8 zHen>g;Y`~$%(h+%Z_Ezh>{>f=`p{Tz)uk??zm}tVhhGAe#={Y@spR4HJ)BGxd$BA^-9i&TCui*KvT=HZ%1=OCuB6q9L(3Go%^qv1sa2cs(G^8d#@5g-TYxBf4 zA-1?6G@Xe`T8^4&uQ0vcok}U6B=a@%iLOjKd0qRRT7}x8_u^`3FB%{}!^>&viq)Kn zz6On+8OCm2RE{5S>EMO^(x|g64xQvrlfC7?$(_zdoPYlmx58?W`F{aO4pOI?yuUNW3287qf;E{FP=;=jj z*lsD0kqc+CZ|id*al!7v5_=q>uIdF!<0LGWpz}#xd@8Q0y9si(H<7(k2Ri4MB|3 zBrmkBtzirvHPTV{o1|Z7E{La{!>=1^(7P>+mcL2Dw80E$_j(2&4(M|Ibzeb-XPu7w z62?i6L}SY;MO>Y|oozTAN3+e3VxZ`KVtQy6mdo5E`pqRIDsLC#doYA~b#^7(zH0%$ zUv)rpekGH65VR*d~*@|(EEq>$GGn{a1wGachKf@N39N}E+qPUSItyE~C4jRveB1DkaNk7* zABBF9fGaA8E$IoM#>2JeSsAi-%q9Ey#Ffh2DzQI26} z1y-3^nkw%~*Q@M6;Ieg{$5|*ykCf?}^(Dh*foXUygDz5y)<<{x^3x7wzQ3JlCC7Dd!1=5f%v={y|?r>hpkz4=Ch&wg^Aosl} zf~H6DwAnKqj88G5(8L(x{puXt+Gs}q z$Z`Hm)|rGuLq#WBXq3$QT9%TjX;Hib;sG`5|4wqIW)ZEH|LEzX(%i9nKTab(mFnJf zB6CMo!6e}z)^4r@)!1rq5X)qu3obFXe&w+Ep*(u}%wl)$*$OKqe>SCtf2Q0$EoeLw z-ZW#Kia;i)lZ@UtiYrr&!9V*q?1QQ-yf34Sn(n6=QyD$H(f^)kh8d91d(Ci@&xgK> z?}L>&iCo{P9Tye0kux&SU>Zu|$@xVS$fAN<_<`qx^uFqbEOkc+D$AtVf(g{MSPxIv zUtlyZoTj3zFt84%$n85bLAk94TWh^xVk#k4;hQOnNWox2C~OPV1Gx#Ytl~F6yzqd- zS6Ru#YfKzT@)Abtu|jBm=OZ|7*oX4dN6=p7KB>FlN7m|Z06JM7(zg!dxR>v6#*R*G zhBzwRngG$8mB4bR6gTtDY(ZGp5WEA1+Z1iiiCv3UB6 z(1N^rAMDHj0iQ$LKr+sW*e;dhx?Ll{sx6BKznl*pC4oe~YA)tfjWD)6o6ax!6gz3G z5CjG~kZ)9&Te>5M+7)l4vP(uVM8T8v9+04SrMKdg+9LR|Fz@k-m(%ZF5V`sFSgRW_(*JVk7*L?9RtG6 zf9aU!V`xzJheVEyWnKfnbQrltMz;PY=_h)za(EV}IBXy&-{X#j%VKfn7)^{{d=ZoE zFVGs}CG4haHuU^=TT;k8gbVXBaI58Ex=LX!b3|qt6*=byEw9ta$QDn$k)#frB2Gd> zWgdC^%!0iypMlHSv1mC^Ku?P(K!J)OCjK#oh4EM5h6u%eg#)l_%mF&?tt<>$jDscB zKCpd#1Zr>>h}qRnsx#{dH@0j87jx2-n>Q^3ZEV+pg@pv8)_R1zeZB*%vJ4=}K@~Un zzlAE3CbF=zlq5*#@b|)-;PrN%6EHpljAF-ut?FkOnym!_mmmyD>V}z}>mX80jf(DX zAs3c3p~>>e=yl_)BDtUV-y@dP)uFpOK9cD3nFDRo*MNa zYxu4Oj80`k=)_4lARbNrOHyZZ#8U9-A`1u%mW3ayhA0hcBI7$J!&%3N%;j||;4CCb z?0sU`&Nvmp#3jGTC%*ry7?^BU>@UQ+&9kIFmupDbl_Us}6vEcUC&A)d4Ycj{#T^PJ zpg))hMI@1EOyh9H%5@}$(k{@)i zYhEG& za*^-x(U~|LmAb(=A54UjPe}65zHZ6@Z3vY9L-|>a?)XoUbWPuY>yF>WNU>P-=vUxU z`kpbzSGe#wMkJlz2nyD{reWP0T=~2va@u8$nc9t5oTtO_&lY3sO)A1^QEs5G=1QiH z>_fq{QD*Y;K{&hd1$Dk7M?baqlIuEzc)J`!bbU&qid-@2<3hNpfb`i1CD0n01ioH2 zo_^}G#N{f#C~Eq#dWkBy!A_LhaMBYa zoF-w7fdsB(g#}9&3V{UA0^Fo7;HutU#xvKiwyU3P4I^&3ol49vG^$+0r z_%;l;%!Kf|Tkyi^FR4DKN?#77lfBN}bdRGLtdc)Q2R8ZR{8zO&sum5O{5`-T7w$tPfShkF72|drz6$HquOw8)m?v;7}+GI81%~s>z~^ zTypTFGuta|0=jAZy}^|VQas5Q2dv(ZsOWq;bbK|jII)v-oAQpK-YF29twwiMJfcQ0 zmB3Ie%Ui&)BxzkwWeC$K>=PGu(Hej+W0KOE-JxG4Hw` zVy$Q?J5FvcS|2UM!xPPj)k$6W_$!l0{fhyR@rBm>1K4@o7)B#`7P@%~+4aEZy)Q|id^C)A!}4#?G-0AEjBCxp9zA)e$a%xAp0|jW z9C60dzb*8Et3S5t-KUS=1>?rgUG$KhJ>aCtAm?UC&b#=U&2U*kOYViCSBD%{s9KVT z&imQhp0aG^C4IX8p%oT8{H4=wJJGIJtH8eb47sn;#Li01p#`#r=P;FL4kyO-jOcodU?->&BeBc!5zom&ZKVlgdm@8AG&mH6VSP3aq~?%m(ea zN|oZ4klM*M^vP0y)g^IY;mGf?e4Su0s~jSFi^y-;i{!R{Kg+6z;ptrM)t?{6k za_Bglzp0DudDMYZ=E=dou*pPD;}WjgRtZ6nO(Pw|xHW#IW^;yfaN;W|sH+O6jm=}o zo#SQLD3r|RU^<9h_(tAb-3mS`f#6zr96f#kq<5XJWS(HJZ|TuRRP zq_f?1`Lt`tQ!>x*0Q;x*3JreS09~sb;7vZy3K3oeA5#m-Apifqt@M*D6x|Fx3&UVb z`4?icBZ|cC{!4=Ia2OlT`@7Mo3d;4dpjoZk$R?Zqp}%b=O`TTUoCsquZPw>EG?Y#o*q1T@vw zZ)Vq&P&#$04p`g1q7E;fk)EkZq^rA@ng8TF_H`J;MF&ZEe}5aR>JbJrXUn5c>K)_~ z{4qC5fIIp8yjIp)@^)7?8{_Cn-KQ^v@s0hg{j&o0;H@k5SvA(w4YP66i)!X#?^(vf z<5H9NDp#yIb(W2BQfH&|Dye~f7OqR%iQOyaGGi<)@yX_SxN_W?rt!+#(bvBj?TxgF zSwt>6)N*!{bgfXb&l za@qPV@%Md6o<1ESE3K5EuB?#pek21om=|o@$`ss~_y9^@r9#!grOZ$Z2fzOHkw?FF zlD;VsO+hp7H%(vE$*4WnC2{I9sQXY8)jE_wPbL!z($*mf(c%ni$8Zbo9pnO!*p z3i5LJ46R5xM@)n^Kzp$~oa^T>^YC0QZti;&&izXyi~8wy$B86ft^`}JDsh%I#@z0{ zbWCkh;zqs3aH)c85amX!0HdxZXAyo{2Rx{=%K5P24c2Pu_X&q^eD6IM7)}q?~NYm=_G4{n3zC%pMDCN4sIx^dR=^8zY_(!=IVr zB^ZmdKiStxiuf&|fd(X|!Y0o|_Ul6xh_Nyt)$-|7ZpF=}s?l7ccUA>;#T2+1J*M1= zTxZVt{UWH5xWRgese-%M50HJO2G0*wz_`+xf)%6BVATVjE7P|DjCN$f^DqHCtG`Do zZy#mvxjligWAn(4o4P#P)eA}&>%jWc|B;s)Wl{d16A5o+;Kj%@;`qY|bUhS_dA~i% zZkGi0BcUYjBi}bz7R_~i^WapRk8)4^O=!vDSknArJQxYC!st9b5XrWN2)+;0_Ur_N zIG%xNOd0z)av^g>x|fRB&%>_r`Iwfjj6N3pY-p(iKmKWe(j;Fpap(}ddluewbG8{h zLyxm-o@C&qqr5X{$PWu8RMDeV4k|xtQC;RW-O)Ek@I3c5jCZ~Py-}}eU~MpZ^zK26 zExgA%@s291&8Y_6QA1^3U2@zwE&YD97 z)xU`U?Hh16wUBywMBt&dW3gB0JabXajyv`!f;0S{$X)$8L-1p#1tB%@aN+VXl4c)6 z19Dzc3!XVD2s=-%9#>`t-AX}qbQY1x2B>dJX54~RprALLot)5uL-|%*&Zj=i%(HvL-XDM@vr@0%tNl`?IVf3O`~cG(lXtVN97 zUzvc#3$<2bH{TR5xq}cB<7zKssF9c zxLi8HD*JbmP5D-&i_N0mT|4oum>X^7g7@W!hI zaz4(7gTnkBw%%4~{bWw|-u^^&z8OL8*Jzl0R)X8tr-ngS6;SR@Bde>j5GOC%hdWC+ z>~p=s94v03u`&*r=r@Lx+3V2PtB+!F2mjeuodq86o#|WOIvkoIjN8WZ+|tcyxMKJv z^**-+&zjF9`&<5yzHznazAA`vXA_!A4O`IpY#QE;*#KEEj0WumkTD?*P8r2u^$p(7 zs5_g6-8ZAdJLc2EdF5owlXUtyYzrCByFqQgiQxy+?ex5`3&d|yq3pe7pdS_kfei<- zX-7Z#%fx~8K{?Pl_6B8B;z>cj7Z$L2{JC%twzOTRk7EI^o;Zz%3o|fbLorn8H-N#* zi`07{0F6=#88_FfSeg8Rv^~{lYt*V6W^Qb!wy%IC`2}QS+eF+s;VO~J4`Q~4u7mQg z+hKlpDOEfxid%lpCAqbexR`^}@T2`VzGHNq`Wl|2FKxr{>Pc^GekYI9Qf$nobf3pt z8-?(Rx;I!@f5hexCkW>|1%Fmm@|>>@CXVOx>E3oFJ|aqFIXV$f{7%d*R4-)N5%HBri`+u^aqK?I-sI>1i*{hO@B}#&dI_@R_rl{qO;($PGwa(d;O+g*^wpn< zATA>f!V@Nf+4Oy|>Dpm%Sm=dg$2X9R?(?a4)G=(EhA`CI4v!s?vfxn2g+Yl|<*GQsML!izgf(cx?4Y~)1NVvv)6rexc-CYJ& zL6<2l5QnQux#W>V0)CFZ50(>5VV6}BTeq&0keP{K7UEB`UKK&V>|!D`kdGaqfn?32 zI+@SV(sf_XnC4(oJuCJMRTu53l_F6W79?PbBV-$EVOb*n( z+zGZ1m*KXvx^&`lKi-MxLbTO$P}kZNg+uN$!slOtsiitA?nn7-krrxdf2RB7*EGh; zhC`)LBUXxMF+b;Rg_t~da(w6@8BV)FPSou|mrXC(ec!II9*6f55py@%w(lZ1r{|G# zryYpOj?1LBbv?1Q3V`*u9>KgbONo!cNz|7QK-oXa$SkoVCR_T5XV6Y;d03BG}u%T==QG3vYO>zyzazG*3Mj@8vd;OVd`9CoWD{d+IWV ziTgt6RD`vn*_cOfh*5EX>XkH_fN_pI2k56=w9 zWE@}KfWr}bX#bx!T$v!vi0Mw@tdGaw{xV8q!gj)=y$$%YmS+O}2&FP}oItX*5Dv+f z@C@l>yuEZ5`)^MmJb11HrMa$9mhm6G9)1p$#q9|9=OT*#Ww9W7Dcm@k3MIbZNn*PW zYSUW$JZMTYCMtpH_N&-;hM@DDJyN=+qb zljgI-X@}`hj4QGBv?BISN;t{(J%s9u2`UeyK&HP3+p+rwe3&_xdo0$+43Bw6V!f;I z;A|Gpj+w|aJ@-+mlnL-^P9*)eGnIxm7{cLBIVibyhT6UuW#(VXLt1K&jWZOviQlJi z*YsvH#wVA6jr3*sbiD^nue_sSS7w9UQzvZVI|?5zsnPX1HP~-?4DQcT6-+Q6qCF~l zT*Izby6kBZI_*kmsulW8y9;OGrFX%s?ZiRSUS@}Drr)NbF2~@{=R7P|<@Y(;x1hxn zZLVb90klq6TQuMN6)j0fWdjH0kvYdR^+J;2dN~6{%v*Y8ivnFf9EdWfZ!u4->KO^i zYG8({(R=EA^5}T5S?`BQFrK*sCwlA!x7OW7rRVR-{qiojIurxqbs_Y2k);W$~GW5)s0SHJ_X)Y0*RTc$Vu(Ji~;iov3|oPo(~&@9o$EH zJ;xXB#pr>H?@5w0XCZF$jf9H(QUc!%yRl-h9x7)z2@Yip;Z`+8PMUv~9;iBrrGJ*u z;L%h_{_KU?E>nqRp%VCjqPXpQES{Zu zfn-%ipmEVfdhx^)GQD3OVoP1%&x70WHu)t=UY`u2e=C}9tq$Nbce+q*$w0U8b+T>0 z9=J-`Ld%MDG#bVoN_#hDciG*XAeQquZcA5|^4li|r(n3+8l<_32#QD5Y^ zv9im#I`0&GE#^#IhovETfjjwHRDvrPOLO6&cIfJ9&h^c{g35~-?vNJm?`WxJbJt1H z=U9V(?^<$$0%PuJgc^70)C6w4SOl)WI+o_PM6u3159*i7TH1Zz`KkJzYC==Dl$k!jRy>REFfBg``K>`6!G+OJ3RGC81{zP zLPD`LTyL4oZO(4QeL<3(fl@Ua?l%XF&68mJ<}uXVX9x=vE}*f1&n~|b6GUC@g~eA= zz$!46Nc4-roUxPf!Pw{Y`9*%N50l}$Am!xeZ$&KQ`TgPTVrW{n1-n+?w^?kDX)PB>hj8a^3;63i0H;40x_{1n3RUmsN9HLts&){|I zduY160P{CSLZ6}{$S^E(_JI#dD#p|4t>IXGWD@-N*uhjqi?CF4OoO@y=i zyN>&5x{n(;@CLPbuh%2p5?o-~0XKtI2-bG`3G`Gpx9y>CJk4COM9aKe`+e z^!~t@&sX4+oH_pc#&d))+xIlltZ z2kmjwsy%r6vN5fnR!=)VSwcn+YYLy$;aaI06cP;x*d#`VBlTf+P8ZuE#1X@n>VosO z{Qu~kxuDr@7cEq+VHC9&n!QTc#*7nOBYA17S?|~Pi3QKT`MK?L)7ScPY_`loSr1F@ z=e0O~Pb7pF{Kmk!Eia(1Xb}oMnI~A7{hlNi&1MTXTY-C?Ij1#zRin zTAUy7w5>7Jd{F>3`L7yo-1tc%X7rKuL6#6} zna&lKr-C#qDJbclOM1G-!-*X{_bICZTq~2wPZ1kXcGfn_CMEoH+J`7TT!{-dnUmc+ zcar5r4{2h%0$#Yaj1KU3#Pu;oJg-3pj(y<=7U?)r(v=J*zwT0%*~zQ~-#vTX+e8)z zuO(kLXh89%xpd%>2v$fQrL*!IIiq{!+$!1QT){MPX2fVVK1!-!--2*iA7qAGl6 zZc<(okuTd#?s!NEGGq0*b4FHR{f8kxlXEx`BSp@3+-7d;lz;R=dL*ACu!B^uZD`sw znVgpq6BOHhr#HUr0reDHpcXZZ@2Cac`u3I;pJPVoI1Sh~ z^(-j!VmbGKw`5FL>liTl4SWcM69`pI3dS09MZsOZT#`-2kCJPLfSeOX8|37~Mg{<+MwJu11Xe6R0 z*=X$_0P$W`j8M0$jJewnR zlHfs>c^j6C{k4x7MkbYEIy1?D?q4?L;yf;>{0tu1UO{FmnR1c0-MK|}rXc@qfVuR} zn(>|B2nGC%KTS?RitMYHEuhW~9}4Cc|4iV$GU=cycbS>I=rc3!$}IFbv5{f<9z$Kd zIBb;_r!HMJ)OYm|KAf0?-!nwG1>47RM^IhC?^)8W#qT!dJ^ZULcG)#kaLRGgdbPoyksqTpVdnAPn%(_(GwE6 zWfj{?_drTZB>us_I7@aW`KKulpI5wMx`Rq_Yv(yCTB(FB&YJYbjK3Irr~toM&4!MN zR_rc0VO;p=FP`LIkHL`$A|?3-s`J7Se^g`Fwoj<;@{A_hS~BJ*mQ&%ZIatTfja_y} z;b_|c24B8{YD-_3qd2{`Pj zB7z0aiU0UW5Sr(VpRQ>$&GAMc;WGg&66&zK=q;Mb&*L6Fd;*_^$8jIEJJDi>J<8>; z!-op8IBDYp%p57;b{EXx+Wn1bjG;4GF|eIFhrA%=$F9-()-}X`YYJtS$-yL^!KqOm zLBgi&VRw%CMgED8fhI17DvIYq8Wd zWFq%2#LjHE<_!5!mrZsg3z(ebuBOrHQ`jMc*R;WWfM~ZEKx&066%I;fM82k?*E1JV z33tf33KomUsnaTxOq^~nDadv-r-D&OGz`rl)3hsLq*0F==jW51^6#ne>Hsq9DnMw z+&_~hj2d%NJ3r%~2k&50zfKLMl=0b{8N`*HL?*WjlUbJpT!F1A*vt-rH#rVq#1rqF zlFO-cun61}nFetOZ9sm3C0m@Z0kRK_Fe+pAvO(?jSXSPT3U=N+OD3NlRK80+SJgmf zLI)l>qCl^E%z%-R5#ALc$vsfa;@mzQ;92p3=+quc_&ZCqtC)yK)25qUMjyJ-pLc;9 zPU4>aS3$yeOrbL;CE#Yush}s6h@oHl>3{PqQFB@s-hV32mHQRqg*a2Vu6Y>7-!#Hg zZUK;pefXl7(s}A>Xmr}KNOvJJXY@;$7NPVGDAtqh>8-5b6-zV(U4h^5|zG{P)dsm z*+io35mG`*3(vW)yM)pnT4sgzPE+G|et-Gbi|0J&zOU=^dB5{t-VyS5+pvDiGw#lb zC|F>88RoWRg7F_2pbia~=D0?@g44yfix5?{O;|~09s26Y!SqLuX=m6BjAaEF`bQZ> z;}da5Ebu1r4CyGP*r&lElcUE$A3 z$+KZO+wpYU6832FP|+oQ8T33gh1(YxC3Kf!@I$*N_6qmXm+u9(%t*n(YwU~f>l~Ty z=VNFte$Jn@N}=zj%dyxdIq2V^g(Cu=Qg@*%K63sioN|8!n>w=v?`F2qNl_H* z-hX&&Pd1jB|H7FS<+$YLCwwI>$2>-tli-aG}kg%3GJUuhIN1Q!D;UUOcLhlx6h6eKDYAlr6yU} zQ|-lfd+nIRRAuJ7rU^#IJ;RwV)!>lnI2c>|lGysxaJBEnxXrK#7tcF~Z)dgQG&4Uu z?xf3pH2=hcm?(GU_gOXz&H>a_~fn!E>nAmwRZ@zYPBs_MraXMY!ggG7B(SgXhx& z(cpbNI<0>ybeA4tU#%QldwKz@S}(%}zmaDXEvj+k&vx#!)DQHC*v^Ms4q{8UZoJ&iI&4-`mc$nKwa=wZ7krzE9mjYlZ}5XS7o9iPF## z)kvaef59-bkMu#f;8r#&#`Q}U;x^WdJ@a$0eXb4ra8-*PJ{J%BcWSV#vFWIpcn&in zi}0`HV|2ZyhwI(Xz=6HNn412Yls`zMBPwqZe`>}zEw&YneV8apD~=>4yegY;{W>+T z8&4BdRM7jvIDYEl!`Qeony(2P%w{EMgSzA{QYd#8PPau7%`I7QY@rrQ|1k!GFHd1# zZ36LnTqL%UALwp0m<6OnVtYz2Si30UsUPZia$Rl^^ZSty>(bF#MM;nchfm};Aey%Hk4aHN_38gU5EJ{Ga_O@hB7t_q_Ud%@J< zC1jwml{885_-peYcqy$+D@UBek)pxm#-b%Kw?-beA8f@h72|M1pfx@cd;w#}XVOEN zk$Ce%3$2n>6I^%~kiHQ*y9)j2wsVr_vMzjn)r|MeKMI;g$ub!lKBGt9|rmfTC!>v_LIZA4HyA zXH-in0e^huv43Nk;6qME()X6$bacn@%M_S;z%P^>RZCxrTWHqpYZw8}K(@AVwk_T` zdUZDq7>J=;{kP!01%6N#=#TYNi}87vm+h52V{T5;F#7nT16ehrjNR7o!dU4u@HaOW z#ly2;j;oUB{gw_q?l%!nSu6(4sln9tk20$?xIx<1Tfm>-8w)kXXZwOO!ua#(FT4;WPa0R5fy$l+ z?tJtS+hO5yn6{?|KkT`L4|;x}g=i$Z`csh3>r%;py#fslot- zj+MjF;c7AJ`dY$o`6D=RFPUAiUcnV=NnrlmDY&NQEi7(51xq@HiafdqZOA^3+@C z*xtB|mwkuejFV5P%@3i6JZT%$$W0}@!z}nXcQEYt+d@919;8yYIJWi9Gx|eXl3g(3 zscA2;d#_}uMd%pnXD}7#K9UpeTSJKa)jpi$KLfohf1k;*+^9Pbw}!hwsf`?Y z_U;Hj0(~*zQ#F%wenAeK?FAX%OqgKWLD&0#=L+7Pz`uu2;rOG6>9{Im)-KykKW7zk zDbHO5$GI6cD&|9%OfU{=)U#q3GuAY zEH#7o$=Os#$PI)Hp2z~fPKWZaZ=AhgV%Dc)pnv{O5~25j$keTaze;Lxw8c?-K91+J!qR+pr|h6#wR5q_sC5 z;sVV8VwQCqBBg`j&RjXTOdPl;YY*Ug+e28YVvQ!a6`S4m+JcT1eRi&hPCb{$)%G}p zZ0-hZ%Nor520{dW!bH@tGQphir^G2R1e2`V#hXTtA-C=hfi+j%&_pGdKJUxLBb5f! z@%nkNaF0f6TtSQumy%68i;2Y8gIHT^fkRb-z`^wkd`|I#qE#2^5iKoTuv!btU!Ota zJUf&#wIYY)k|ASVp12`uBK%GK#D6>Hh}oM1c%y`Ur2l#f?z;1WJiKy?PCt@KQno9@ zqQ$%L;XMb~IINGl>{1r(UA-MtK#U&Eb_Z0HIg1_mQw&2#Y z&<05(1$aApr8q>GXI~p(2&+wEX_fg5e!69o?cH$;NOOKB@t=@Do?R&5G-k>3EB%+y zxG#rkZHgJL=w1aG!;jHR8*HiHm^gTxv22cOcOPdPE5Y!gcev2e@wnqx8ul)1 z~a!>vNbj7mc3pv&549 z`w0KYn&{Q-WLdK}sG`wDT@ z`-%^ps!&1tDe*IX1E#O@=%cj{_*s=>S#-e>*md)-t;?6Aa78|fG%-ECT|ti}zbGds z3%`+}9~J06KO4|VdP6$(H1LGONAA9hEPv;h9?LtY2Qs^KNb-NK#IMJkMtBwy?KW-Z ze%ukA^^>T{3288Sy^8Y>+rsZL%Muu~!JL}KX!u}b0zOjByu0RJh?ai$+fu< zU>V9uTPJcujpV7VQ3^coA5Znd7E=Y~T#`S!f@a2DfuhwP=y~}Wf>Q1X_0#OY2Paxl z|3x+nhm0~SajKx^!R?_e)-J96d{*-XqYxcC#Tb}9iOn=U=OM* zT8qrj?!${Mi&=|j8_q~{7x(RVK*tb$+&iI+EF2XJQm7*MfDT~9-voT3q)y)?&xA$U zYw@CxbNF*e3I9w|#5m<^G}dwq4w1e@N6j1sKkN2jM}->9Ry!(q{0(F z+n*#haaJk17kWTIW;edck0g`E%?Fu1W8vNV#W;MwC$W>Rz$kQpr#}Q=y_qy5J4D0X z$07J9M}_GbC8JlR9@PGv~jRL*Sn00%2!(pZn`!OME2DAg$&boj5ia|1Gj2ipvbi zh~K4LkH;t!Qe$L6=xH#xn!}`5Ph>jz@5B~w+z9u@4=1_j(E%ZgGc9Bpb!@PQaIMuW zN!5<&a`TzC=S(IMkcRS(r%2n(3Me%junBtkf;$`*1W)ul;OLM~q`=;etUL3Nj?qa% zmL!XFPuzwLIwz>Kfgw0;DkQn(1;n&t4%)9jPHqdCV^6-DJXUdmwJMJEV5>2im1Yh5 zPMyMIzdhLECM6c(mxg*xd)fC}B35?KhGo8afNLj7!S+N;T$)hy{aAkPp3sF#m)BnfL7j>9bY?t8W#sV5W}XX5Yoz znRa1<@*P?dbv6kMzaCY*uO|3grkN94i(ndXy^<{c zAawh5B8YER5?Aj3mt<7#1DC=3kxY)p%U(j(BxN(1aCxXG%(9n8Rg{tqI|oSfs|=ie zc00@N;%VgB`7H0kdo1&Kj-k6;g=|U?{5DSE)XLtI?2$R3C(GdSd(Qb-lWndI4wTl9T3qv?Gs(YSIUORrC1Uyd=( z^bugBfa)#U#+?W;mYq>c+(QvZvajx*8Cs&bf$p|a!JNIeA{HwCBZJwwrj*{mv36$=j^ zLpD#9{9L3eHcHjSRK-cCqEZeaQX^o=4kgs`*9R*bo-+(5uvW-5U)=75&41p}mM&+K zoTx`uHNNCR?;oMgLrz1^+S|l5Z3txi9EzjoHn6oT4uay2d>Egjh-uqg=w3b=6w(I6 zz$8w<0sp-m;v`*sVJs+t?+EMyPq))Wo$50zT2YBLU2DeYmg=0Gn+2O=*7|;k9 z`Y?tn>`EZ>EM0KE(|IgW+J}GE-KSGd`ZI~DOWei{`Xa40@3=u_1w?tn2)yeR%kHN6 zF@qQp#y1s_!t7<@Cy(Q3&Kkn0>BmA|@pVY~JdqiOUcr>#bHMxV2wMNf2)whN*}N*N z6<7d0IBnSe4uV6YV{p8%4};&8m^0cjvBI6uM#Ew56G&t(#qY%`Y|X`4ppk2e{;wwC?u|=9KjM`* zKL&|p$Vf<&OhkyFZq#Aoy3}y%J2yDCC^H8n7h9;Ca z;W3wZK5}>zf8oPQx zoercNr{$kdl7=HnR6AWA6YuBHhr%4Zx;K)~_mgCCG6J(m{Eb&$NIA(aPx8KEIxQ_Q zgM!JU8Qrabnsb)1pRqpp_o_3t{xo2aISY;ALh$yV30U?xh#A>VX2zFK)8GoB&-ge3 z!ejEuhwQa<`9c{SwXq4G{MyZjJ3Sz~w;Bu2UN-vGsCf$4im*MFY4Qgs0a?$E&0)Oz?c@CHYWXt9ko@+O*%TI}`WLzEjAiHeir z$muFIw#!HI0eSN%eV`Mo?8JQ)MMK^pM&OeT3UZ~|_xTEY@4@8I*wXLvTk9RK}` zBSit7q<6v&Vx1RlTbyMwZeGP7@{Ouy!QN`mYinggnBqOB5qQE!ktA zO)z}-9PXG-AT8=%0UleGKuLKmXL0B~c1gu^&rQ~%*lROs-z16Nl~T|WnQ!a(dIf7O zX~1cJwm|#r2>PzdT=42XCV$;m5-*=1^j#dwUs`q&bNYF#&v*w$LlBfN{{oW<-PmXQ z2KO2|;%nJP>{oDOx0jl;xpfO@$@5(p%wn*$NQG6d^g;V`gq|{xV-Le;FkjD>_(tLhL#ecAIPYl)V9Sv^cwJzOxujq*O53|DO*Ml(Dl1F`qZ2r{`cNqTW9o7uL20r(z(C+gFe0H~p zDW-Zbg9F2v#urCC8L$fU&M0Ea{Z=w^ss>na9K`>N;6}NGWA&s!DsxN75FI=wR=j2g z_P?#*lgcQvZ=wOL)-8b}#ecXrk;ULOF$j-IUc$=bNf@12#{S&f#};`u2>;Kykn{c% zZ(5X%Vd?3l%wr;i4!J;l$7{og^85JwS|f~CNW_)PcVVTP6T7823JfoGVC7+1Y)CFd zDXB+ve25ud4=Ta0s>7L*;vsxG$_j2RScIn~6+j~riNu2}E`4z`4eO6!w~~gk;C2hv z92JG4MMJS6RtGyG!OfX!$_Jr)4G8|tB9fY;zg4;X55i*2l&^`MFHr14&TLF+QDXOUPdOl|EPREHS zs_>Yt3(hiLC+v)5+02U3><0e_+xErNfj!l5)ggd9jJ^#apBKY|SK$yfH5O>T0{r@( zPxU5#;3o~U!1jeJ&|pU@hP$eOc1w!T3yj5Z*|D@VB7jb`)1jASd#IpC!x!&49I^cb z)%qu79lrrNeJq~O+$cp8_Q;{fx_s{PAZx1kHizvijbl?Dc2oO(#h46HuxGR|3!QkB z{VW>5ZNnp&p+^Q2FBa~RlK)WRd=?GfZo=_JOJK#^FZ9LJsbsT51djBz;Y>Ejb_W=e?Cu%xP1cN=y6U4&k|)USe}^)r3e3NMI820 z-FoO6ZInMwo;eT0TVC<-QhkW1WZWv^C|}2IXfGmD6C;?!&OWfdHWWfj-SK+ZTQc6~ zI=!j58^6q@oWqY180#L#o(-SELhOt2{mcTGFK}XRR!*ZY?Sk-)qbI(Li=iRcgpS^_ z4mxyJ3#Q+Zz~@HqY4*8St%4n|o{-7GZ8TU~2e%zjpi9IFMBOH0k^c1z;i zYMyfAVr;2;*IqLD(M%GTX$s5Mc=8T=MaXtL%#psJ zB4kz#+I&ezZW_C+^cHp5LVzFZZ8i6`(I8O5y+%*ziK6ki^VWVkbe=5A{#K`-i3*-M zumUeHiJ^PAdw6A330Yq8m{?d$!LJ8DVB>@36c>2Ffjfst=$IJLX?jY`(uL=3k34;M zpptsr5IUH*#-VFjiTK6k0Cqpn4|f_U(>3H6h71}>`Ly}A-ua#MrQ!~$L!hFlI>!6%w);+}oBt*e^glTGR1 znXbb9dlT%~JOrYjwa`uDi^#W?L3m@y8{B=@iTakQ^R*kKiS)|F+~e8pWMo;a1-y2@3h(b#@*dV|$#R7xRaxJr%q`JFLi%J=G`| zGz3n~-iS9m+^}o&HL!~=p*u$z;a+c9@EWEgF!ZXpr(zG-`LKzaCfO02o6WXLHd)-G z!O=L$V*X{+_pPBmJ6&;2vK5uwmPWC9 zA2&z07T2E8g{>8CFne+U^qe1#QA$PVbi|Z-_aCPk>BhLVwvalVEue|QyCSzq8a3}+ z;TE2I$*oypiIzcc=#u5X={DyQ>J(^1zI{rg!QBRE9&;bh{J6&t2wdwaXCt7Y;WPQ; zlmgZUB~&~76F=;BBFGLm$MM;^7`ax7MhDM_?-M-9vB8t*iH%BFn`XkbgeTC@*8Qa6 zLLDc0PzX>g-HukLgQ2%L^{w?wSz zTZhK+=_oPUiT!v`%guL7B@x;tKz^2T)&Y*>Z~IjKnr95xRb5Ulej0(rS6zfN`%5m; z&=kM^o{V~47If3nZlWhQo);_Ik?Dd@Hhq#Gz2rR%$3z=L#M#Lhw)rPr@~#*BbG)hb z$?sgwlR?m~6;B$K_4u%(W@x@+vaR{~57v^R?aCFE}eKL-CYt9-V2INmg$(z*Ezs_;fcB$$8`k zQoFt34-LgD+hmD`RSdll7R5>GZp5SV%HZ2=iA9eWz|#8DSQK)TGtg{+ zWriZ6m^2JjmtNu0oz>y7o+BiTu%hFfWMIdh;h=qEA@|7KEUdUc7Ki@)OZs-I z!Qr2j4Cr0eUpIHWU@(=?wom7V|%@L6Nx0bx_7Va&2CQ$s^ z8_Q;R0=s2E8ZJ&Jfg&kx=I|$^L}4hbTxbrm@&+tlHJeIo(SU&oeh~1Xliu7Yiz{h2 zDOoZc7T?t7pX|~jQJ&c(;6eep*ZGj!QWHr2`3JGh%Y4X!DZ@Z6P?lN0QOCKjj^jo0 z24)N1)6U%{7@IYg-5EWFpMPT<%Eo15_n;PX&%cgKEQsej6<_1*N9iEN+la17%}1hm zfJhIGfI!hil&#>{%X>{+=YC&kiWm!$KA~{q<9u?`IacW62^;~TXMAxDM{Bpbi=Wj9 zd7g5PWh6whj?+39cV>C0#kivq8r>T$pd0ISco4{Yo5%S~G%w(|? zgk?qImEXNWMj#u1yc&<6)7+_L%K>7orGUB;)!e>J6B=Xu7F%8Ja21~EcsFz-79AZy zopbWZ@DdxMElW^axc?5mww9m#tAzZ!H%Vacq@mP7HTL^T10EFmWyLjV)W~_9DEdhp z;7oV;Y~W3ZXD1okaF4SrGQ)l655uqkC)zN)T$uge71D}h&`Y`*mUUetEA5oY+$j$^ zqpCJ2GLAUlSeo-QmQrIk;~LcWoE)g8gt zw=0o%du5>PyabFAzO$z7aV4IU1h;vM;2Mp4P9G04#rEI(@xy`~*pf7a%(2%3JIO^D z@F);>uRBJjdAuakd?e|rYx@XOhiNBlw^{n>pP>I=D%CK3)xw zXR4d^;HtkB?b#TFUgj3~BioM^y`Bb(`zF8>cNdUuna8g$U4S=dl;a-LWq774jc&`i z%b&8%rWQr%=pLO2vsUo9sZ<5CRd0eWswEtWn^OYS;DNR zU`@I$F(_)MjkW1wkHZCYN81XTy`q2^HvENkGmgTEgs^(3N@;_+a=bRDZh!XOtJBk)#IB9ym3T^&0%wyJsrTfQN)G&J$N*`#O{kS^1|}nfGf|xw zb1ogFlCQ5onYktj85xR!A!>ZIgAaT*`^wkZ=))EV3G(iKiqPE`_V1~AIJbTZ+t&2} zqi))h=x@i-bI)MfF!eUQJsMh=dmqZ-!BExBH<3^)V&?w8(hM) zw6(bVL=v;Tp~KvlA~E}tOWvnng}Pf`$dhNCT4dI=aceZeg}o zqDL~v@VM=l4+}PXiCLEgwn0+@S4>Sw=eJcvx1>~X5#Avd^ct1eu$=gJ%i`3bim+uu zCjV{B34t$X3AszGp;Em8BBesGO#iI z%m>+r>-p2xb~t}g8|KG-M&D=qv2tBFIX;f321PP3BryzDZCc0g{*wys_lEG_gG%Ym zHRaS?Zwfhh_i_FByk|7)iapyt#|PIHjRPCs1CU<-gP6?HC5Aco$>E_%xT&KU4{uI{ z?`yYl7S^LlDtyEVA4+kP(|WT0!Fy=fP=#XO5v1qsBXapz0W4pV3@!cbU=(GDxn{Mv zKUNYO4Wlt;({`Nqb0@Bg(?%)dXZY{sHtyF+4d$a!OG{6`hRZvQN!oE+W}qSuVWEk5 zaMvEFczK=5I%Q&Tu+X{v{Et@+jwEUMV`1o_a`b1H@ge=hIY&yO*Bzg+~AFI3`5 zk26>}Nsa3j^7Zk{|B0<{%EIFHd#Q}`X{v7_hPgL}!l62CX0X_r4ZeC0hn!nUM#n^= zUHWoJ3kx^8mdte+`zbfbHmKiir_9ES(fhq5Bm|8PKe7gl^@XlXy0v1J2X&FD5VJEMyq z^}7xY7w;D5yII1q(|zzgMjMI`h+)y7k)pa!ck#xPCHUn1YFss9I_c{BM)zos2h~jz z;o0axqQq84hRV!m3vceg>?(C!ptlUXua;p&KsynyQiQ0n`$2X61bi5tN$}QCj9jh; zXG^ZawL#<9#F8^KrpK1}kG+ABs>#Ir_&Ppxatv0-Ah&*tC*G+{!!N!U`PfH!FwaRF zOyb9Zyy6A$&sGq9?2rbYm6D@2Zz-IYKo1Km(6Kwm4~$Q-{h8`YgXwR)df=?kgL{Ox zvXxm$LLvHU%wc0@kGv2qmb}n1F%#~z+`#q;GI(ui1jzg?!IWyDx3bI<9O-scR%)TW zH%`-M9cMB2+-XuSQw75nmxJ_eGg5q636JxJC^aefoQ>7p(wh; z+M4_s9ff^M%NXBL3#*n6!X_0j_|hZIHVS#`ld^xv_oFM=&c;T84`7Ou@{2h0vrov7 zA-^zq+DO{sRSeaAkGQ`MgVET(hJ(}T;-EYBbeKLPuR?8Lo$fo@uC)n=I(5+GRnkl` zVka9Fz86ENuoIYb2&XS8#mTgtPFUkc&%8Ve+Wo$CpG^ep3h;s-9#Lf7^g`S&`+&Ri z>M?yXZU@SQM?unzLhMrCk2f6!r~01@xIB3qeKRy20+n1y>68*E`uU7b{5OnL-y#Bk z#e-$U}T2CZYy*Pg-0wxiHAGl+azBZZ@-4xokH2^_hllpaZy2Av(Ipr(7C6qgNP@ES7+ zFuDl#XIwa~F9oDIK?YuwCqYu*aY){f4wEN;=f4kF+I~NzLzRaqF{7PEC?D*Hru_*J zXD}Gw`Q&o1zr~3ktE;j;*v-^Kwt(x(EI6C*O_s#`qU~cBck!M9El zIO&!GBLj-R%5p9lU%B%~f54ep>xt>PN&E}nW*A~U98{l2 z3;A(RVkqRdmr9jkeETz7kuAmAy%iWgM4C?MdBNQ>OeS+$wQ%PWPWYah1({V&5Ib2L z&5{CZlj^-8!ug)Sy)}g^Pd4EopK_|)*+})Y;y`>%8M1D^B1Y3RVOZ*D@}~6>)vdLG z$F>5K)2@hyoE(8Q{~GD@)~jUS&Qkg!#*L_)E>MZJu=$F zin*O+iT@Ow@$0>}l=41;R&(M&cPUOzU2jayw|W2&4e zs$3q-=48@;>rM+fDOd6|wSs)ie?@e1JmK8z5+eC>GJDG$u`N>-Wyb6$8+Gj9Zq7z9 z8#^0T%(Q}STDhb+Jc*FbKo})-cZZS^lGU0>Mo2uNsb%J{!f7P)dN~q?Ts;WlN2|FB z-&*kIav>)*W18qcm@PUheoK=;0v*r%q6=v(HMy(prdjn7oYA52V5uuG zRLu?$;+GCb<|>KC2|Ytc$&=Xo&zeb^JtPwG((obj0%0kOXiUxXPc zvvwv@%M@^W|0?1=+<+@xdkVBZ3v7SEt?TD>!uF^vQXRMHxaLX}@fbcGrOsvY*IXw; zhdH&4+~G};FLj>;7mtj@2bm{OTfRo@Z!r~A z?uM~_?}oDfXSUvF^>Me;-{3*9I#U(;M@6&h!Rk*Ye)WwbKFu59_BL0RddCP&2g<2M zyE59yc|hInY2ed$l%wVuWSVB3IOfxITpQ`Yo!b({&AC!btuCymS``ZLb9%0LM1~hv zY&x3Gmh_?PuFk=YFN&yrzZ`h(x=gRIGxWuOFUXfK2gq-cAr{$AWXHvoF!8$vc0RRa z1vkgx!@+ZTIj6;V`=m6#>%%P!I{ATHwt?|WydrUq(+S)&CWn)^Tnw`&gu~eK@vQac zOnmj_B2j;@4dHSh`Gj$Rqe|1L%P2EUj4q&UAN}A!XA|8M9f^y`4SZNy%x$jP0O5Uc zXj`rX_qN^w`#dE!?@S1`v=3ri#wm!rPrVRkl+RE;+)`k)w@^o?NH%LmHAzsZpttIW zGT*QU;-=_BecSS=@=F_P_SFv8>Uv_v@2hmujVkfUt(W;NV^hhGjf25wR65SJw}35T zEx0%QDL*Z6FF8yjVQu3#+N~DH?J9aLlO$NM%8#FJj294A+@Jy&Rl0!E1^~iFpOgRXHUkf`jOI7kUznkcP zJIFs`VGq(-JqhBpEqwXVXL&oO+#mzBY}#Jn>FE zy~GT;-_|&JB~RAN3?YW5=1jI}HM4p12Pf745{G9sk^CxEUXGK2tu|*UMlGW{J@(|l zi3ajwT?JZQTZJ}1zu0^kvkoi|C4Z*f{Pz#)7d{=vGza>wah2FO)dqHmTZv};Xn_f$0(VXnk;%vAQ!}3gvgXAIaz($7 z4m;z4TeAb`s;4hGt=-OWM8*>;eqJVrSDK1bU9?g5++4EOP6yUo#pB1E>lm5E(Vs!m z^-@&_p{HXax@q|l4RZ~g{4^IG!kQ^5^n$zUOSokVYWVsDE!)m1A#kPNjN~oPr8n<> z;*9UKP>prxIq$&nc+DS4ue>AlTOFg;qA$v^*Y}ZVW79?^Nq-{ST5pk8y~=Fgy)v9GUqD^Imk6D2 z6_nHHCQ62F)CDv)~RY`&QZZ9{Bl zTuYQT6yxqIKd80g8RB_(7#^2$;1#5`NXgY2o83K==&=#qaPx8n4tUO@x|b_xBHbiv zQ~v>fltiL==ij25+#psw(*f_PEn((iF09fmlqDP;0nbiefW)iH*yw1;N~BM*=;akS zGe-Clx@`>MPt?F&U{`lf8I3n$qseTunXq@u7?3>rnfSgnVUM5AWBLv6S?o~`;tFRV zZM}l)B!XGVq1CAV^E~Zn)F5?%rsSztCh;>NBurcLa=oFqU03oIq0^&V#CX#x&9XB?<2u zE%YA?Y3}ME=)5R}V|SHs4;`wB`JDoyn!<3#yG~B$N<6I-WDu*ps^5+! zSi7;Mv!u!Dk9*P9H0MshsR`O$hY|;7Zh^^LLHSs;kh)VY+X;iRUXniV~FVYiWq3hh$2t_Y-A74&0&wzmDr_Ws}Kv zPOG4U4y#~Se+>M$Rf@V6$6__~(wn~%nAX$;=68D*{+()om4o#$^TrzR&r)W$YG+~S zo*F8e5r-?zD`D975IS+YBKr}ynVk$$XRancNM+syqFEy3sXu=siEpaGbxQ$licf%i znZ?9LV2yg-F{h@l--AW}AW_QEi6rB$1ZUxUwf^X^5g7Dn5w>RP;jTh~F=uOw7xvts zsuAa*rzsPQA0*-!`jXz)l*Dzy_k^}Uium45P+_PFjXx=myM({J!k!Kc?RUcBQx&u` zQx@CXHAKH;RzcmOIM6)Ni*w!WnERtsFu1@?bmHAbs9bm&Kdtu_S#9wXd9NBRk{FUn z!ye9KyUKTR5iR2&<8PuU!ps0hsu;nl;kqzVVlgBMeu0wnGE7yw0{s`a(_YalGKcFV zMRVU0UEj$#v)GDSKRE#d%W7@+9{!AjI&wtqDHBAe7O9F}-!H|6S+B%my+g3S-&FL# z%mvnDX2B5s*`iZF;-JDm3s$NKm3bo8q5oI)F(TzecGU<#ISunp4f9G`4wyaXz zFu?(veMhkUU)At$!&~m|jsUorr_DT6eOQaBD!2)|N{Q|^JiN)2Wi~IOetRFm6h}*7 zQ9of=x*g7aet_J5Y7Uaxc@SDw3BS*)h>S*CisFAV2!1Wi^woc&T}?bKOpE}>=l2NR z|6T9`ea9^VVBnaG4}F&r3`Nf0;gOY&Xxx=M(-Z$ydZbHihWoQ?%=9rEW!K_(MJ$HKWRCz=5ko^dyl}abAGW z9(~2%enAA6arpO{4D~(M13h_$=uAXlbHr81T{mLgPfNB=*r)UHHPkw+n4GgwW~S$( zQ0sR9h`b)~QuA#%eckayuOkY2-kb!9FB>p$~1;nZPAKbRy1S z3b>{|LOA;=5h7awd6rjT+<^jU`+{IBWsgqZMu6|B4E*@60rE?FdFM&_7^-bUmUlfN zb~_^Rc#jkd_fJFb)Dleg8pUcB9>n2h-Q*)NW|{@z`1N=)4C_Jq>8d3ZD$jrn&%vTZ z`%F+tGiO5*wehH-DQvYIkKcJMtR7}4bP(R6kK+F*IuE~`-Zzf7)7GRV?W{^F={fiH zBn?D~h{!HmNLEFAN&^k6XlRfkqRxHYGNX{libOKMR*_`=&hIaHUa#k=bKmE>KA-nn zf-M-=iy6690s}6bjhZ@$c^$jOH9RQ*qrs+}fBFFYTC9TQnLxdBzR*K88CYX9fyuV1 z;SZ(lICo<_>b?o1UiYTpIgRtM{(K2aq}#KT!W^N22(vJ$CFtKI$7)RGFrSR;L^Dqh zBj7QAHtsg<92||!FN(?Q9pkxLdm#^eVJa>iyodxU?WXfqS>uP0AlO#oAvV0b2A77p z3QX(&$la_@s4;pN)Fv(_|Hg{B0|So)HvAzvSt*7a`+%oAS|*W|TI=|aW2{L?b)T?5 z8Aa#&k0kS@^XZ(O!cNNFjXwW*l;&oX&~-`mRE4ex1PY zmRybco%gUOMiP~~kAjVsn4|{2pz6jzv@4>S*$09;LZAFY&MmZ0+ecrHnTny;qshTh zE9syk@l~p{l#$N z6rx8D^P8V41%bvmPEwZ)O|drKG1KQ`>V?VbW#Xx7i9M zlB_|e#15mh58%k>d(j;>P;s{$_dTVL%xZ4I+pk_yx!-Y6to@A6>p`NudMde;t%XnT zrjg_+OUaHHVb|upj4#$xgL%yr)U2h5D7!Bubxt|lx}E3fs5Nhhs#zD+^FPDI`;~Jp z?yIogV>dl^Hk*K_6kgNLqfNM!Wa5RLh?)T58nd=4;jS__#}?o3URtzq=*d@@++ zGTroXfVK@iO_N#^gq-#V{)G4$IRa(e_7CO6x9m9CP-nxbq?MAcqGcrQ%mse!%`~w6 zYQ#((Qs}gY{W$e|G5LH>8q)HkFhQvl8$2Va!m|T7&%zeoFMmw(gpQ8ylS@RZ#E|#{ zp_zAu;n%}N@?yFbvsdiK*(W6EYo8JLH9M1DIy;T-95PE%aH^AVN-k>3;0ohMJLAG29dHhII+;UY+{VUAjoSc;?v3LaA zQn>TZHtc39Wt7EHw0(s!uP3_;rlIx8^b|&!`KB(RZQyUM$-8nYo(Z6jpLRB3j zv(A&90zcqG*iJfAfswUehp_<*d3e~(s57f4AO8YTKQ@)CURO+s zY$QMHnKny%E#eEbmjk=F1813MfJRve*2pBYQ3uo+HT;KkO&S`lx8+m?28~?HShBcO z75>d0$1NT$jTW_oA%FWG+9d46ow~K)@G?JGWWAJQ+-UBe&?$7R+Dl`a9Ju=UFz%55 zDV|y?(&2SCxebQTiO#mMH1oUh{y+ zkO;Y7FBAP;(cJGzS@gzpWf-@xfp)r9)9wSEL}$1@ndm)EzdcT}Mrz?pj|SqrVm~>!Tm{^IP9oZxg5UbXak?ot zhCVV&!jf;c_({6)RjXPgCJ_8>mpT<{1kI{AWU2(%b6?~?! zh#2X|&>XHnIOpByoi%1`jU+sNlQ>clwwr8AQEmGU74GL0A}(qyx3kC8K}x?r}j0{`$)BUk$csGp7~|zkDO?6d zP#Gq$cq>fk+DAJnvAjUzgU@5I?|GD;XNvMKFW{3APq0%V0h>|4~`9~g)ILEw6 zG;?Mw?&aI)1`iHN(Kax8VMBgZI-o(56+B*ejyT>+g#3kqoA6I94HUS$K|#LQygG|p zzB!E>dZ~kUj`gL5`>M%d)of&2gp62`3ohnmK~~KehL0D)X4XscKaGbsJB=XmyTFLA zbz>D--t2dvH;Gp6B6E$jAakraK5g27b$PPP`C?Lih52*7+^LzEXS92~4n%B3#G=XFBIC{i6)+GWzI zcLbLF+6$~~V+QJr)Nt6HSi1SxF#0Uok^UH~NY07`zuDN^n9AKie85kFld$F zwR^2Xmb!G-sj>p%V&qCC<2v}3GtuR;-)z{p_&u6;pUOe}K?x{l0_{J|X*XOfHY-t0Nt^TWTx@w0;8(z>%WhJ@%@ z`Emas=l1A}aDc)4I51_?Xv`Ku3+=vZaY z6uRP4C$92WdKJK;BAG6E+)kfw&n4^M2cXJZU36CSqOYHSqj#O>@#2?-BGiK^wL|u!NeDFqBfZrx%CS((9Msa#GL3;Hib+$R6QC z46Yo+W;H{&USKQCx0A^HxEu7y*V)wVivw{zycPCIM{^!WO~DS=z;@ka5*}YdqleAI zISOlqjKMFFbwmrbIjB#cSY4-ww8BLNK3mz<&mwkRm}i>4wt$blv_4WYkUab~l^J@M zlBq_=uodpn-%CCbyLC@!%<*^_JI)&9zJ+iimz$V=qzogD&I6xbb8%1q2=Si2aQAoy9ni$1R=+K6)S>v$kbq!VO zAA`ZaFVcUp!JuI)0d6-(vF`H$C@ZqS6TC9MSb6{shYiJMhf*->&j$ECgxzu-CH9+< z2d{$?h==ld*m37OtnT_vb&7qUajk5P1eMiw6!owLJMS1$itPwG8*Y?!f9`6C1l_^J#|qFoyHbpt)VRRMK9q> z=Xt_Z&kOk7;RTLO-H#92RA}F(mE?lEHR?(V9&nTOSUpEs@Hm$9a_6+!12!M)zTKgY zGvnz%WhU;mKFtT6%p!_zBIpT8AonGUA#~79^3}5t7AG0O4{L0N>o>@P2!zKRn+&$QP`w&-%v0T&=b(-Mv5cre*A{z0LR2mFL$p(Q5ntlTJ{>|n~ zmWqf1e-LI=`_+Zd`b*jIFF3q85CU(Ehpl_vsodYWFg-p2^4o*NVP2cW36bgGR2D06 zIyd5{i*0z|w=`4gtU=|)Bhc=V5y=W#0Gk!}5UJ}caoLh!Oo;-^Mpnqt~i* zWBw@|uW%au$E_quaZ#M+v&m@M(SfvTE*-rm8(lUEJqF98IAHEV^cJSVjyaut^WU3P z<#I9Xksl9S)B(6&?Tp@UgnP%he1SPqNLs%i#M!}v*?>zuPL@`tkr6T=S-O~By?p`P zzYAQ`CCxZzs}8oS?j$?hJfOBK02)IJKxH77+cs(hykDrw?wqy2MRL=?>(L9U`m2@P zv#=58jb8(ow~xaMhs;P>swouPD1g!CXBR+1mudI=;c@>A^y(`o$H!OT>gz!`_JBX} zb5KLG&jBFOy^V>srJ&V1M+{R~iM_j)aFZ7(vi`W=XslEt9`q#pLa>aa6CS0QU!tMUQ@2At!nn=YEvLyVl!j-aCQ&SFw)h5>>QmZ-H<_ zIr_}m176Gx<}5dzr@u{mh)mx~NHqCD{XNd1o=PHlF4o1KdR^|_&W%`T`iIXo9>pGx ze2h)AV3J`eq?~%SFF;Wprlt87Nse4Of}Uu)(q4q2|MGA#Xkyf9Jg+bn!*&2<*o;Y(y`Oz}Rv-1G-L3jlI{FUCEO zIlA?P0<&>5VC+yU8E&3Jo{y1bYt3a?f6^f2TpTcNQZq`=3Blt(!?3&KJQW>WM_mW& zi?uF|X36=j=(g(-o-p1H!wdajzS}{Jv#A61Wx@`5_gg+<*B}`Fpn~^si$g)v!d9M8 zraX58a(8#rHzsbmddV&MrU*jR=Vz}{1A7%|rrAhgN*c;bOJUX`(75$ds zeIG4xXT=E`WBylQ!Ii*2$CX5CW@EkGR#!+@xQZ|TM3CbHci`#VYHU3hPKFp~!h<8% z;fI_H7~bmw{%aC+$H|K)P1_+(+2|?m2$x3hKdHFsegK}<2qf7<*V2Cx9;hi$Q0-;s zkyipI!FsB|1o&1-(D5)n7;y@9~+XQ;3@WdV<6*dc9aVc(pMC4m~8SM4#h8ykk3zcdki zr-~-d%B&}3C1$)@PF#K)vm0@dsKV0W&izPu{oEd!jh8{knmG9MXBOylIheEh09q}K zg7tkppp<$OTS8>m-w0_k%v=||hVLMkL#6RsQ7+xnu@U10PJ)?vA!*Bs`%51D^0lJ&-vr0N&9l%~)Gz9&SWn^w9?;3%pCc1{&C+A)Q6qBsh6{fI)tl4i8O`-dMj-yTO>*P(Tz4%TQ%i6>ue z2F}w_a5Ms0(C!5B)?-=uWChOALQ|NLJ%U1wkxVB25dHG52||6EFvos8**(jWKfTj| z&D@)fCohFUU}`b7k4;6>`*+b`n+B^5Gl9ikO3)KO2sRZzz^7J^u*XRjA5Wc(x5i$k zg=PNKtgsg9?p!BLJq-H~N{Pd_mV#!mCRFdJrCW68(vJo*aK+V=G^Q)FgrWKHAzEMH z9Sp4>Yj~dYIQ~Khe zmuE-Q&Xv>1-s#Uc%sa@rNEj3As(xa(F@!AAu*ZMSKDcbiae8puar$_QEL;2{j8&iD z(LK8tqE5WRoPJNXbW}WwRtsH~k%w@8f++@t24Y`w4o>v2!u8+moF6W3fKkyFB+fgm-n0C`o2CA?o&I;?pFXFA%mz?dbKrk%|<1RVH`@Q)zC%49xuux0!}Y+Hx2{k z5X7^MQRwhK2dY{>qDo>mK6rAIv?@-5F+0PA=ZGU&w^oT+cg}*hx(o0!FTyPvPvHt$ zV(z{g_`Pl$X$WbjYnR=}-EAjugNrHAS?r05pSDojs!HB}urn;0*bOVa-jR$y9ca66 zH)>wu@bO4l@%wl+=-YvCK(7$PFKh>;(qb}5)e)Ojl*C~wr)fn?19|dHf_)fBVk@|* zEar5B@V=u66CYXO(}(pqv^5Zy%lgn4Y#Q2Drr}zxLaxw!4!2uY8%v`-IMwLySaiOY z%#}Nd$!Sqw`=EiY-bKj|s~B=zaz8ooO+oB1?hdBKCE$8@H*^wuI(ua`+4bZ3{C2A= z;61GlL$|!aZ-Y+Y=3~7;hG@{Gj|E2g7dhBJqLI6^^dYC0xEorvn#hSv7my9v33apf zqsO&ItlFD{eeFGXJ8cCsQ-6sw|I=nBb@Q;T#f(*nwqlP}2QcTg&=zEdSw~%9%xFmx zwLwN)Fl9N~rFPT9&wrxcx_G>{h+~mUYw1m^y;R%hGvD%jFf(m>4O@1r!}Y%J#NnGM zh8n!2;<<;Zqgf0cv-KppC93hW5`1AsNhPc}D8nB3d2`oH!oYs{B48eVFxYef44r35 z78|;tdV{d%=$B&mm(7RGvnFG5as})^6UtfqwZ>=}g2yKdJsE*bGV9g_^y8KAN}3US zd~zjDl~|6+Zdz>G$2=-|ISKlob>Mt^hSQabpy{;`PyO?cm{i8&vhw}t_Dl?}M?cUf zpG|S(H+ht>l}G>4!&uukL((kXP0G1?yy)IcCD#4Jm#->OvD<}(y?#So<91`~8FO^6 zxJGJ+ej$px4r6C$J@Kzujl?q!zc|Wsa|16!)7TR*d-xPE@AwF*gSFYIpNfpz;f#m! z3&D4dHH(z|i1UV=M0K5+aJPIDwfHAwClgkKlHE?+kQ)b2xUqP}Oz6b#4TW0`bD((S zJ;)1^VOdAqSWtW>Ca%mT!G=R{?E8~MQU5ZWxp^4ISj!8Z5dF0g zr_o_aBJ0{j1UG#F916Zp#w`~Zrdkr>t2Ykeo?+p*{qHWCJ8CE|p`HiV91p;peQGQq zR71RkeZf&B)+}{UBsfmY!~E@sv83YyZe1Zo^i{jaJ*#Z|{r44yySuTus|FxWZXXsF zTe2Z*Gho@i{oq;R1L2lLJafodERiY3QQyU=ZFvQUrrpDx=cL(+aZOkmo{n1gO2Fvt zF)-I06PirURowv@+DKCM#M<(Km>qD?-(iF6O7>o-J+2Qig z@!;z87mZXN5G6?i=AHN(k6fL|%AU#5_?}n%nbu@le9D|@&3X$`qaX6$hL_Q`pN(;* zKmmI`e;xVtUPqu_MlyOW0+qTfz@>F4C-wI|*`j|HtXCX^jlTN0e0?d&T@rwDm1jtg zygY2)u@4Uf3FnY2k5hzf%ghQ5agx9WaNT#ChV_3TC$5{o*6E5gYC-3OnaPf%>X#(Z z?~zBd*&+C(E)VsmPZj50R~Daaxr}pY3P=oU!jSL@ETz;3ze`uZ`l?2lzgfr#tsh5M zRCeP{yDNBkT`w8AR9d{*(gC*ZQDRbMl604c7JHf_i#p+>+4a}nG*@~z9o}w_OVhjQ z`re25ajF!%^HZ2f=FekV1+Vb4=?x-p)kkkv&%ojhBiQivOfJYTx&Dq?GxZ#JN`|bi z!KpJv+=nagsI6Q(H992)#<%6fhh+wf^~1fskBOk9xEz#=zmNpy zJ7nXt57ckrCGu&F0qOmef>Rc1La+A)*i14p*3A!JmmJ1!7{U_ckHY$*VpJG80k<_v zGS&1J=gB%IEUqyV><#XL$&VId=wyyI?+#IOX+7AnB#sVTPQtwMHep|^i0w|N=&Q47aZib{q8%LA_5_xOFu! z?d}m)Qz~$>6a}|q$_a={I|feU-;?2u({NeiFQORqh@46CrkldziK2ppxH()BzPo;> zVwFh{vegmJJ}smNZcGFduOJ-ZEO0!HXQEE5ki(QHCkf?1p1o-2(u2lbzgi6m()^E>jz$8p2q)) z*@IsPUBiA2O;}ZANY`b5z{qM}x}Uryql)b4=4WHro&S#E+?D6B+SC?fa<%FD152?f zt^@r7Z0Q&6o4BzRe{0T8L$hD@w41u z^g8cBGY9mr?dDqyQVpP2%DPGCY$GC}bDj*-DaPj0XK}IqY&`XHF8?|%UDU8Pj0{?B z!@~U@(<>LR;KMC3;JIM}y8Zl3@IwOi>sdp_{qiR79wqSyrdTpP+dX)_egJiJm*9CD zN3^YF)LS$fOWOdKUOYkD-K@yK9v6g&0n&4-ne(nrhYeR8NXf`7Vr2H7T-Y}ShWJF& zyz#54oO`9lInlFqBRB>w)_OuG$P=St#G!N|MqI*rq2&Y@2SN__=ce zv$DuwDYuTH-{VZOVy!G}TeKU~^A8~%7>k-6!{M3BnR-KyXPo-sQhMLLfaEkbg5fa< zPP_gH1i#3qX4!v9?CKCMeyT1I`&Y!_r!aet??9t=MYiC=Ui@rZft`=@NtI0ue7$~@ zoSnFYT5Z%}?u8$yQ-C#iF3`Y4yJ$#Lk;G#QrEtT_W_aVb6TQo$aF2=s>B9z#@p}U{UH(g5|89gtS3eLBI5My0mTY$5WD#6k1cRz85ca>NyX0SB zXh|*au#4wY&u_>2_BJ$jlYmb_soaQNLGWRE9UneQhyS5)0GFFtRXKJxe%Fn3FI{*#e`Bb6yOZ~?=!{Bwg3%rfJ1w;0QdICO>CpjuTWRqGNM?HPf%#%`dGDbG`NSD!%Eu*u((aYq^Om&zK|j=9{HH}vJ72zIpf-(S19N2N$rLu zV`AhCvd6m*)@9!0Jyv-`_L19|)BBGeDNVo+gT8=|p%%2N2cqtYO72ol57$zrz*ag3 z;9QFW=WS90^km>$oV(tS8eM%tE8B{Avr((?$)CI2ik>@kLGA`fa5xB-nY(FKu%pn` zK18PcY{OY*O}N5h1P(R*i-v{L?9|d{80y{(D++}N-9#Cj{5J#CZ}i{+v}YeqUBbN9 zi7dl>2u8G5!oZq20^eo^alibXPN+DJiekYz5~oPE1{Khn#AsMC-%E7f{s4L{9D})@ z=SfX{4$L3^7}ZQ?(ZMoKBwAmSF4IkbarIfyy5<#GlYSc(pPdOZ#jj%;joIhrVtA%jiElQZq!!b4 zxlNv<;ih~XxC`^GxL@CK-692Sb4wvpR?I}TI|ATfOLKjm`F4y>vjm5K0+(ObgoLjz zpj!rALe?^ll?l1=mA{wZ)`eE6p_mEbLk0dr)KOF{Ure5ymEt#Uvc)l0b<}W4HcSl= z=CM~!(dOtKT>F(SqV@9{#N2%io6o%Ap3JhqW%62LoBmBiFQSQF?mLPeXE_#iKOH=J zd$^as3USwY!SS2MQ>T3kant8pX9-Ko-1Q3PvW4d>qXA79vJ6n3?KLI zVdGApgpYaGaroX{&Ju}ZnE&TEZ1-7#pImgsC3_uVgynH8s1$mv8AbHW;Ugp|QG$)o zc4zr_M}Y5+2k3jD8qSQqAh3F+nC{(p`swyl;#ho+W``7E!2Efb8Ewvdw@M-1Rm9EF z;z2Zi6-+<)9y5-agO*_fEULXlICR4FJ~4h?evZu=?II2^>>`^AdFS41*Ff|!8j3@)Sqk>{zl!{KGRgPFv6AkthWVDtU}hM-5}oMGl|@zJH+4RGP!KjNUHu8 zP=^Vscx2Rb*fgY?m)Lok7^I#AzQzUDUOq`)S5j*2m4)SB89F;Eu_vC!Am8#5*{DM9 zztz?lHsA&6%@;{7XAK2Hc)R+fEUqVq@u}cTv%TAml`3CJ{qH(KESIwg#R_ANc={nrg_)LipAga$~$LNg6Gv4i7TatM&%Jt^__x?I1pi=tc~k#O?*` zg}M**;5c+XJ8fLdT-xg~c1?TT%=bb+vug;wHslCQ98!#P0-w-7pA9gwLKl<+AK{GE z=4|`G40zG5$`PdNdJgBvr(HvGVXK)Ev~$r$cVQVeZy;ZMJ#aVrHzP zhepQ)$3kxmc3qT3NT|XSrC%6dHI$_vet?G8U2yykeK=UxO=X9c3X9|+^>15*yuZ8{ z)7fqQU8>+#DsB_}8I|~IW+(q+*<tdSsk=JDyPOM}2l2wZ`hvH1i8QtXjwC=3V8_ zeL08A)k@JxZX9fhjUo@cpVP6kqUw*2K29U1By%&RisG`#f60`9sg6y!JJ z(Y$M*A90Obv8}-)GZ%x3VJLBYc@fVnNx-p_f=PO)G8B&=0+N6C!&$>=WJYTaX-|oR z!a#wQz1SARO;1xk_6azJzH+{CZWl>ho=&$;u;I+^U#GvGFm8dgtwwOPy8bYhx zq~Y-SL~NO8#TU!Hg7}Sc7-8%Ht^v+C>q`Nh(=1D;1sageFBjH-&OE@$^&UVsrPD+@ zYCPtorC>tQRXp1zyx;7PqU8@Pqb#6SU`^Ivy9#kO-Z<_MPj~ssk@vBaA?lal3EOvpjGDX%9t9qtE<(p> z!VDkYauuVCJDi+h| zjwUynDW#6!*F@82*kiFMj=s(6hI1(lN)ja(7p}CZs`#AE?jA%iapK z$p2j3MB}Z{@Z)`b>Zjq@JXV4(Q<#iP?98zFUMntW9ZC92pOeM~9pqN6CTdp~(D-E% z@Mw1>L~i*HjY=|UcTo;>Djp}znIox&bOgSA(ulFjr|2jN57-dCPqbD42H9C+13q$T zaI0VsZv3S~4lht7!*y4p$+~QOJ#dp5sT8oQ<2=PXUB_}uq@sy;T`B$9HW3q650Jpv z9MGc5IIGBj4C;PNzkc~d7f2sO*V%G-Ku;Yny{zPCa}NYhLL`3GJ3!YAJ`cWR5S^Pn zSLoQAL&i)cXs=Bo1{>UP&W$sqMq@Euw78M8__-Rrb3n|H%Utr)HRNWzSj?**UDl4 z!sGN|u>?6=myW5O`E0`8L^3wV3l{laU^`<~us1^a;?&KR?6%1>_!OYRW+n=5KffxH zy{-hUqB?%6W%>77ER(GlMh>#GDG36RB@pS-m5NP>$4r% z->gHx26C}xPM$Dlm0(xY!*Or!LwuFf4HgI92wv|()J=1Q808O;wI~O6glt7q|Dz}^ zdk6o8?4?VJoA^1&rKn*Yg|@d}(iQ$ZZL{8sMg7O2aK=Buw^EEw#Y*B`)eEWGx>z{A zbQOGw561FI3iQW%H^`heUGP5dN5$W#;Xz<5NbR#^B}{{PH;gCCF8(8fBc(0|968FI z;|)a(iN}$5cZDHsKe!Qp&*0j#S?vABRJb}x4>F%*L6xiz>7qvLzU*76NZ=b})o3fdhwD2Id zESbPQ#Yf_)ycF0RR*2=2+AP4|7W}$@kTurBnMfl6POjZUj=EI{8Gi%fTYiO$3pve- z!|roe&3WP|;mI!NnBf@%VQ2N999!*rncgz`LR=~Uujs{q@sVUqXcw`og)*$>Mj!qC zSp-w=aPTfp=wvC}M7y8A;7a{)k&TKLeBxuVC`O*?4PC}`46WdTTL(7IB>t4-nwV3)QY|(y#2Feh&*N1-4fGX1(4A`=+4R@b;B+rfBf9fJVqPfy zGqVk*-m_*`GcVAXNPD(&#B}D}5lag%Cqq+qDJe4dgWq~xAR!S*l;IMV+J!T#;9Y{J z^XG)s7YBB+N*sT=5(h1uvnZ4pFeEQF3zN@CYO5pIqg0(<2|WX(c#@Qrj~b7F6^e!H#s!E-R& z3@9Pf41>k$L*G!F3`I5wSF$L-BTyQiM?bqfLg_)KP}H)F2CUM@v!TyO{;e zw%CsCAKNS%F~$gK)L8aA^bDQx@-}*@d_(=TK>C0rKvrBXrn?bbriS`u189J_$HQ&S+Y}LHZwerq}}(jo*xS%6_3H z=f&j~w1S=eOI$vt92{GM*sI3{WZSQkSde@O;tUh;TZcQwo$(6Z{&1lQ6u>i8MezwXJo-;b&FJJts3}G2TGPyzRru&8N883XM?ocOU=W zrjQ12dQRqU5ZL977w~M24)i~=0EICbWRlto+TQ)1dpl()d4D{S>uOCQGP@$dFh(2X z%S&O?k`yxecRsl&pGz9N|B*iTDsD+l3d}H)#?SVWg`C1{;{9+8Y#&_%XE&lKGc=bBRX>B8L-b+m z%OuF|+zTd3Yv8|5BeA`u9kktZg26jeU}eWTjGrwq!-Ec!4-M0yN8$z>C_Bnb_toNL z+i|pPXeB0&E5|IY(-18<1*f!526amr*7#->a~i1v4Jiky>x(|R5-*wv7OP@d=XMHGqGZBsy+fNT&zKog~iFkT&DlM{p3_B#JdNhmthgv0*&^QFHO#dq&ar-$^t;mDgnn5bjH zHg;9wL(yTfR4tFP)2{rD8!Nc4(>{?j=PNYf{UzdHuFeXriqUSxJt!Iemn5mw@%wWQ zVUOe)+I86mM|XH(+uu3#QOR*oZ)m0-5~D%!4$}E4XXwB~Wwy$8H=Ee?15e7i)5#9z zP;^EH)V!8qM|(KC`THq*`E5KZuHx9Y!?T(F(N}b$Mjm`h8O|JyAJbc}8)*0^N%Zg< z#OA~^y6bH}eY;=*zVjQy5_fpAe7kj^zHk<8eIN%F=X5|zN(stM4#0`Rxol=FVVhqx z;vqW&(tmg|cxnfeoK+R%O-2I$Y4AcEtapywn7)}mHfkFgJw*whPFV^oLyyCno+x_r z=w9YmN|?bef0&onO%unKbI*#@nJ7OVU4&W0!plNu>*;%X&iXM;oAD4I3>(MB?+gaZ zCydS?^oDMqqy>h*lVDh!Hf?-o2#0UqB_sA~5(BwKw6A2WkY~*%r;JaLE74=%W>dZ> zaJG=oeE5qjY|%!ErWkHoMULQ;`NBVQ9D>1Q1~t6UT6bP8o6pYAp=o_a;?dk@GAAA& zuIVtGzpw~258TD|O$UkAEumjGy-N^TTd~{3n+cKBdg*{{qB4XL-E-&hfa73PSCeJq zWV?t%?*W)H=>Qy!HwKl(_qepsy+oFeeN~ERmKCC?Tmo6+F$=auyQ>&{C z|Aq&_!W+AwR6h~6jy?r0F@oEM*+6gSei#~3MLt;tb3@K>I3x2qZQI)knyH@DXZJ)= z>w|+ZEjxqJx@NxP;{h1$Sy8t&R2OwdD&oeGC48uhJuL4!PK~<9a~5)8a3XgV6u-A6 z*M1F9QCcv@$*#qf3#9OVMml6lCejPhC%KV3cN2NNGq~d0X&iRt7T$Ip0F0M?E2>t9)pki6z|q)lAMuzo$A@yHQvYf?jO`+%)aRittprF+!1a zbxq0I zOi7>zO3oErZ>D!i>X%0F8{!FVLvF*hmr*p|{Wgsf?IHUXA4XY~L!5MN4LyBeDCWd1 z#y>JykmV=G68EIRk_I0CD0N^~Up?ogAg1+yYOykJ2(>;jhwQu^2U_8Oak=?4Ts&IH z$xqKf`D4OaINu(;Bd%adhda4vkPo59YtYB50(-oxac99MocC=hT`spDBgf4GsRgQ- z7X6(>ZwzCqdOO%WFL#hCj|cP34@huh9hs!Dhz;|b#Li6;7@%7%0hN#QKabskM(KB_^4WS(1gNHrMcn z(F7=Na)%?P*$_Lbg?}GEmKsRDB5u4DxjObPsZjex7X0zSy+3r=tlU#r_FZ5Mo5gZ5 zDP|xgIKes)!nY7_1F zF5H7;+=y4t5qj&Z3XA*fiJ$ikAx&Gje;VBZ*XjdHeGnM4{MW;(WDLXaCp-RoSI|IJZA62Ql&h)eS;&X+&G6E z+bRXGt!_Yi4BKT3f(> z%O2Vn=7fVXT1m8gI;*cbg3ne?MC0kp*_Wk0*xGdwlXtI#oHcQj^=h&1x-711PdKc6 zTTHgp94GD9cj3{3e%!XFoX>ceMkK#V@I3)D(L``pWlI~1$z4YXWM4^b52YT<^0_&A zMRb?`B!QP?!}f1X$ELASWa+mQ`l_@X?R=*}*-TUF-ER*83D@ZEyaJfh=1*RTVzE?J z4=nFl)8BF<@s{TY=(({4zVH0UU4AKx^j$7Z)>=+NY-PoLOVYqU{}m{!DT`nI*#S$Y z^uspINi4=ho@q<=VBWp|xVmc}iDLQ*_3*|u}a*y587Po9-QyGJgqt`Ra>_eWs7sT|H4YDf-7OcEHi12iPs zjf|J+qn7Jmh`c6gR(Aj=*q?Ir>Kf9ZM0^_ZR4%lDq}1A`y`k;`Vs1+h*cx@|asIspfXPsa{o zIp`!B9$P|UHjjsKlLvu8^Ba(+6CkSV1)1oPNp&`+lOK1LVBYO4_H=0ps#&?QB@xkZ zX4efAXU4;ijQ8BuEuYA)hJMcCZZ_EJhGVFuIXnB`aQb!IIy!;B0IoCAV6MXr+WY7i zDJYX6pF};xo^8S-5lO^i+a`YMp_^nvPYgJKCA!=8BMsZa_R8NzC7TO)@v$d0y|9Km z+FneZPM6}aAS?D`s3Lnm{RUCm7);xF7ww)TTw0maJe^OL#qOm)KNpemc3H9A18LUvM(}5japTtdedfbMRuWB%`Czdj z2PUToK7k5p-1S!t$Id@R>Xxsi+R}^oxW#RB{XajNxppK5nJTFJz?#KB3 z-ZtU?<~&Uq(8bEPXE0|loSJ=G{k-?P8ya%GTmq3rM_cW+tfxb$4o?VkO6Jt18Lym)9@{Eq4?q@1^E2K9JWWu ziuc`Hj+WjkY_VuF`#fL^**A3AP5W}TrD8kTZgv!&9x8(|H?}$JIIQ6Mstw_J-XZk) z?*hJ%H>+R&<*vxU^e36n;!7TwIKigZf$;LkaPg0E@5!91kNkfKSe!B$SAEWef69e0 zcv>7+mOGKk-kyso>8d38(|p`MN=>*|slx0D6EQTIr_Tkp@Yhat5+3=5O0#pMzb21g zd_9|NG`U4e_b%al{PM}}Ucuk1HW3yT|Kzrf%;uW+so?g9KB${_nmE~0?5KMSJ-@coV_C@<^Ro#vua?8NdHz^_uY!CZG7|E%{*mQb{g|IU6SmEd5%%gS z{MpEL;I~}{I;X^L63@iG=UaiaRG~50Bf{VPe`d zT5huvK2ND4!J~)K)DMQx-C_rB36-kx^PSkfzaA+QaYs{_uMK@aOZK zbMA9ppU?Y!M&JwZce`kB+aK)xvyl`O943B(E8^##RVbPJj%-+R9OwMa<`%YICd!Kr z5UmI!IK0se@_rIt_3;BbY4j=j+(+OJXC#p^uf@pruWGnT*bNpL^kCT2t5mgYCtZ2= z8)^%k-*rN!FL!o2yz||PzeZ2TmPaBo>4OW-*S|nsr!5d!=xVUEmC6wAJqUI|4P;xm zGJX2j56`@7v8KD7VBJN5{aP4|xxdeml%h3ot!D}J37PN6^yQ+tCMG2K0|zVFE|M)L ziKA05@UYsCgy$x6-7VW8{&GE-_m*%#PUqCqo?`tDy9gJ48lX1D#UK>6Gbb zpvFQ9G9O0ppLDvhZRZir=GX;N`Yst0eQO|Tav@#%YXGA?%Wa--oR2cH?{Ht>4dGl+ zM`zbMv1wpP_3avPZt*JoEoqO1-8V>I;&5CxkSs8qG+Ei(lNfbXpR0~cL)mp>;my${ z=uzH@3Pt-cs13j(As5=;9!9@^X{hk*4w2gV8Q)%xBtgIBS+M^FlyS=vbKDihz<|#b5vEI?TnlcMSCZ6ITHVH@eEx& zup}^@mpI zy5&GV*-T^;t`^Yv+Bq28RY|>`J)>v$Ok`pD9@OBix@eg}CMwt{!g1RqirOYvBKk}7 z+YgZ5byl$3CLeEm*dcvC5kpe)@!W+9*qf70B@RRZ7c4jj58j2* zwVQpRdY}hg#!f=hT{-l2{%qRx>IX~@k;kvSag=7wfd%gQsI`w!W2bBMSkF=J1WBTF z&P}|p(17G=H-6Vu$Z}~+m+a|DXy_Vx9^M0bG zF^;hbI)rH`fb&grwsgpP=p1(#=6m_Vz3P6rn0EtI)kd(5;);yFESO$b{D+qf_fze0 z5;#v)xWC;0kN+aDTE?HJcvmU=e4~~wV#hEc3sQcv&8(-zn?gKx#L!wOj$XgGs4+{K{{z>ewc>=7K7P6V8 z%|zzLd{FW$#LH&`S^A=3wC-Xi?s0#Bw546(lxdh3z}s+jAZ}zGl!w!)iV}V>?~t9mlOW)=h&xjHA)s z;pD5EKZbq|!J-d0NrCtueEi-V`>(~IQQlZ|l@7q@9)0?L?^5(R--DA^rwR-8n#}dnw{hgRy9uPEj)4nmo>5&PVyz;&QoM4A|L+ z`x9-MzWxfR9-(QI`7Q-F1*Fh8$8~5k;}1RiN)zIw)!CJ8Pq`bzhqKP~n;6kx#NrE1 zaD}opm>fNv{q*V}v1dbA+sgCkKXpIMTD=)psrR9(^EX;8l}3-8yos$Bhl|$w-4#9s zr&fYDZu(Y$(JCeUkr+d&Qz*=kW{yXt)8gQ;dJ*(JG~oXK{D^t1Om(M9)|(Q*L4%L|Uz>a`@i zv)0;nttnh}Hlw24NFrJ7Ohz>X@!XacGIv!rIeC2owc8a;J)({gk3D<%pD*mlcmpfK zetpLyCA#Rkeu42oQSPc9qLGxApxw5b(p`(3*iCNmP z+_yu>-!w0Lcs!GY<|~kLPbK#BcLI8CE~D+Sf{$9@`=qV3r&T+1$t3$9T-}e$0-oIV9jut@C(YJe(U9A7iul%Mv(V7zTF5;}MM1 z$g>zL>MUN!XALOB*Q=_uQ~EJ+xgP46Y3Ce z0}sCF;;%9wE$SDE%nzPLtaDOlQyXrKp3bDFe&DJl-qIbq*)*y8JFOWvmP^+# zCU1{rlO350alSm_i4=RB`REIs@hOWcUz1_lC)L19E30bu@27M~qBUB5-NSB{r=s1U z&}I93hs()~!qrh$pwMOqdPQBdEGPpN_B}(-)C(A;vkR)N#8`EWkj+z2=jUB4$M4Z) z6q8b6!KQRvyqmC@63^+I76lMjSwP8x3!KD@d8q0YfrZoJ`18k}QbUhEvi{Il@b1|K z7gEl{VS~dYLH`VAHR=Jm*TacgDj2=XIkPVNJo4l5OPY23B6j~>0GoPEnQC_qp3WAL zrDI$1k7*3(PD>*TE2H@@^Bc%4Arl@S-OQio#zOxOQ>q#>4`RZXfa>^-@b%3M?#AsB zx@@ltNDf*F?B*$Gq_h>jA4;Ka5SD5rg1@tEV+K$^XcMc1YBso4o*JhLjDCWqW2 zcWqTbCO!-0k2i6XO8n`L%*$j~S{>>7u$e4d>rGN_sdLEmgQvNE8FTj^_3<#%EX8J4O9 zm4R}&a7HvNP8vllPo!5hs6C+zW?ma2@JX5nbGUTh+x&%jzFb9#G(UZ7HmzNz2|rK#;`*lq zVDg8XoVu3^ysUDfJ8X)$$;yp99b<=g>zv6N^J;L9I1YoiPm|syH{srpRN}NIoSfTx zkF4`wPj)p&0Lm;OOOGppA_im9Q*kude-P@D*Ymbg`-tfYC$!$}$^Qd8w7Ivr>cf&Q z&g5hWiXDGPW6BG-Nay?fl()8U;L;yVvy)?^RVFj*S8G{;*-3mq^a!o*-N(N6OR>`f zfvEDr2V0ub$@q6q$j9$)=%?a@o!|gzO6SR)@?M+8&R*El90w_r+R2vfme%t!o2Vx@ zmrSnM!yhm##{My3#MJ#ScPL{DlUP)R?VlX@*Dnv^nJa#%{2>)yf43oD_T^I-qv5#J zRtzqS8Bxtc7DAV>0;hb^V*x_`e&gxIaOmRdswsU7Y2}BV%p*>j)$H%Ixdat(;1faQ?ga+a_K?3@jJw!|vhRN$C5N zV4hGyrVlIxhjCr#>}QC+rTQT4{+vr27t71*JSWzz*6i+4eRgj99auFdi<<2`Nl#9D z!5{cw%qy`jMQX=OnQqUYG%R6VvE_{PhCFN1H=IKCFXWIVgpr&c3Fe$}uFi@-uz$ z`4v&l6PUZH;e28FFT7S7k6Cy0@sE}fK52`_ADPE-@wpJ>m}KlC?BFL zcZVD79>QD<3o(4tE@u4Gf>bn?(#5O7D;q=lsdS(Wj+kCeTuZkxr!*nMqq3Zt=JwJz zr#A5ug&E6-_g*;kx;K6@5QFH#k(|niOft67mWbK>!EcNI;7YIS)acJ!Oui6-OUj3^ zyJ>Y0roKS5SD_Yjx}0I3yA?U+QNi62-^E(2cQ6sLV^f73(aGn=q^Q6Ezf6y2LFtV+ zFpTw5vmkzN7~le(F;Om>dl93w3wa; z(|j>1_x8BWw2x7ouR{SjwnGsI`Y$23{5gs6IbbzQ;8&cnt0k)1#`Kzko8V=M0M6f| zYKK`3?G!k!T%B;Q-L64KOgoL1(;dmyE&;c)h|+Iz!{GPvZ$xzk!~Y&0rODz3Q2JpD z);MZFxo#dk$>K^z9ifv;m0+4iCP?eJ!K;tANz8-8+{1~A zP^G?tlpSz^JxbGPja&jM^`GMvXEkBWEl-HMlS7^?ih+8C-)LQbi>h*G$>07wcI;0v zL`x1A1>RR<;yvH-M%7O)(>9@M;YE4U)1<*~X`O+NCZT-LoG0{~kSB85IFs(^uBKnj z#97MdHd-7=-!^A9Y$Yhph=U>wewN<=zI{4F{;Pw7Zq6W z+i=`H{~@otyq{F=tl+l$E8*cEqXEu;ru)*ySFP!hg(GF{llf;=1n*}D_3139c^`D)te+!gTD8QyN0#)u@W=;`#yLNyvRi}m z(Cgjl?4OXl@QWGuD7D+%=PkwpZ{Pe}}+$$Ni9V>L8j0wNTWEr`@?Ca!h_0 zp0VtpW9wA7if1}dl6;D$zwyAEWF@Y@-bnUI{~&eFkH`l85&csb3ZXDab|sA^N-I*U zo~2vaTwSbz|11mnV5{k*%=rs;ZqH_`rOnv08)Ec`OgY(dBZ^zRvyD34*h({BzJ}z_ zYLNZlBz<-B8eOs{0fJvg!K7Dp+|zI)+&cnz>|xI)6!E|3*HdJ&VXQ1-QHMA!(5qMy@~8#>=7I zRB~P;=k#JHTOgi}<3Wj?8YUrN6+EC-<_d{tzQVU>bFEux|O%`n?e8l zS4v+zVJLQVEo+^A97{$A@a5}@>7zgkP>(TyY+p;5_%@uYJJrso2uy|djYTxuS_#%v z`H`$!-*D={L7Kd}lrIt2q4zXxN$z7iBKvy}5yca@_t+ZN_1(pmbJp0B7>PzkSIC@* zd_2w7@a)zXax2w@w2Ybp=Uo?(`+YA+_ST8;XxePw&i}Vaa#Xr^2;b;eMT~a$iQKYPj;g zu^yzjD2kqbmV&a62;H6M#@A9`3oNAL|V}xf*L-BnwK1Pq)$zA7~_sPM0)fx0q zzk+D!N?TaZx1@;*C3w_^~H|e*ZTg7HJEd=DX$CJ|~tMRYwXu z5epcjoesCcra}GpUxI&aGTaWcgmamv!QLPYYm9HuQvpLEZ1!aAkPpB)(NSE!q5@X$ zO8^DB4IayHfV+*Q#8qk;^}N_l-mMw|zd|N~)EpE1Ht_~k*g6+=

p?(1lhFxCgVJ zEvI+ejUZmNi-cZXh*9qq@#R%_dhTr*xiv=|HfQ8v#aj#b@cJS7R$fZ3hlrumf1k

>SYhbV^CUXK5pH_AP>rk4iM_5Skvuz@7R?F=)ks%3 z7o&zgFIwo?hhreUs*aPa>js0@Q#qME)zrJ&0Oi-$;e_TLI6C129~&6SdE0FhSU<;j z&x8*0?yR@K7t_OOUP3PAvl|&6>dE(1MiXmyfkDgNsxmuaQWet?Ln@VmVf~JsLGVG`>AYWHWi){TCK}>uu zG!1oz6VgLOrAdpR(pLrO=;2VTs*LxSsG?_12p{xO89(M4u~%8i%>4(jCpRJ?;Lc{$ zpPI}0jnCs=hDLLZQt39qry6K=*CR6T%@8;msD%rg=kZsg(z(w*d1$sbg3dV=4v+2r zkzJ?d;QUet(XGtucygi?t2tYS$*F$$H)cPbslFZ~_g$h_e2U?sz;OI~FBQ~o-=h|t zQ;3B6DLQcfH1&Hi0xc3}P;no98ob?^XwOZs8L%4zj~XLscjf{do}~@GWeaUq+Ue~V<=a8MNVo*3j6RbZrQ^mI%n}dGIFXZ zPMs1fc++o@sH_0UVHG6g?+H3|mEc=7^JjAIM+L7y1kZ1Ag&E5gaZ;cr`j0V!ffm6L zx6GE)u=Hl)^LeVPI0cUNMpCn^i#V@qA9?z6JY4>Gk~=x?FQ@NO&g26P+5Uv>%+RYG z$psVmy-EePUfala(Mc>aDIG_hab`+ef*6hr<80L(pxh${K8W^^zIWjw&D#rLuaZ-h z#rYhv^IS1`7jlDJk#P;)Kjh%>?@W5IEf?hkx>VFKQ(){1)6{K?$Ytti)$ zfTcGBp!53}RP&pQGE z?uRGqTv%EBD7L6cgc`mZaNxr__OEII>}VUotQw?Qo}CPsgwJPTnMaB9)GivIRnNW# z8G#U!C3_WuA^rLnGK7!7iQ)31)9z^)e=8ehwly(jr4*r)<$#g(-4H2ousj<-lXZpl zl~G0J@Ish<)t?=OSKe1pqoKq2jLO>#RPU0UlooQS#hXp)a)UQB)o}HZa5}{L2-C4h zBV)}C@!oW4k)FpafqS%;jd^TB4a=-xh&K^k)*T0SL*+=SVK#GmpMt{z_mC|UPr#`O zMi{%_OPDP!qP7cj;H3L@Ui)qe+h$tN_tXXu-MP=WDWlz1z&ID$|$CAQ5#QQ8_qt(0W@44 zAVZ!dGZ`}jGF@|!JbQYINvCDNBK1U&?ot!A?zV-{VM^4)-cn$T@5B4%v#8yuZ{QPD z432@-WYqMlXy;Z!9=Q3DR<$wg-n@0#WO0?G1ZlF0-a@R<{Z;kRrUYNv$lwmELUQ1( zB+(R~!FIcLkpn_ba{MDA8aY!17Mkd=#JVq3>3bfgOw6WheeKx?pYv>a-Oy^yjdQVW zTn7K(oEp0xIU3)ucuiNPeaDvpIuNzyF?sAiM({}(;>V9$=*8s;Y+zR}G#;qnYYpcx z#fPaZtdyaa&lfuHPaCQDcYy3qlNTKvrAr>3(zS`%`j>~yX^7|gy+W-3o-`fN66tjGh?`UbXspiXcEe}sMWJViPGGT60Slc@T9J$52& z7grX!n5z-GbDMOhz$BYBqkhvHu5Uzf-<4VO~M*w^Glr7W~J zMlkV=9mFqj1nFGqz+SjTBW!En&L-Kjw$wX#`K2Q{erOYBh;L#qe#g_xd#x%Q zxp4%q1Sv4v{Hy4>=@9$z?iS|UlL5`&?cm*gf!2lBke!B-?DF9}IJl#WoR@3D`*p(X z)L)z(iE?G|`U(Eq_W|EJrtnIR{q(2v3I5y00&dMpS&^?^0$fi_!Q@bT(%>FT;%7w& z?3OSnDUv2qN{XTrF-``IjR~c-^c!();^_qcaRoBy-#}^`%oC{{khl!Su2OhL{ zp>b(XFe2a=x3g4-I7Hg>mST?7QP&#$;{#za&B7=5H?U=a9D5%nu$J#lg4=y(2&Y=Z z8Cxfi4I@*l#^3hBNe8XTvF#i@k4PY;l6};kEG9b^cUNU^-N)5NufRBN0UKE@$#R7~ zx7fl+tk2IPC6&Wz@a$6X^Ev>X|1Bfi!HO=&gnnh^{AN;$-3v_83X?i|Fd_7}J&-gqa^3vIGj&<4K(OiGIe z>uv*R|6zp}cfCco5rf>EHR7Db^RqTA=LY%heUNmzKIR{Pl!Ay`{oH~FjyTE42Kuy{ zxfN5tk{NcIVCp49^VbSImcE6Uu~3?Qa*M&mJ}W`*(}u{8-uJRA?#izm#Xy&e*1~Q?A{J8x#BSW;X^sa z7kXTDO|4D-uMpf!d+^1@V*F($!SeUr#U}M|v{0G^_Ft^QZSrtB{PqOgv$l^eRFhx^ z#_Qncv3=muJ_)1W4DkQDX0T!Rr(j^YIDEToE3$t$3|_(?8C(!@PAn|lQ&5KQVLVZr1p^{IFX}66dS$^X+$i~mY6^^C+ zs@bJ9V9^TrJw#XFCFsJ5(P?-T%B&Jr-{)Clulgmx2v{o5ES9^#??@eQS>UP`If9{~o zy^(lXZwBV~?n0e^N7!>qWjcOJ4f=FcVSarTmZi_;W-Jc{ru+mQtJ+9c$7%>TZYRnb zdyW;)l?K?MiI10Nz@`dA99DA*B32E;?D-Mk`8fc@rOM&!!d|K~?I+RuSAb;?*Rbgu zTp{t=5_TzEiF`j6&REs48(0Fl+x?C8+nqC!x@Nd+4 zH3rR7-LQUv4tsfSHd9(Y$muR>#nG?2a6tDQY?e0@Jw4%!R_|)blRB%xoy+pXwEi8G27~h^!3&Rz@F^gzJ61=Aj-Y>{yn(AGo$j6(r7$I5x z*Zmq*)}9AHSPt6T#M2GW266ho8)(>e2z-BrLwofTjJZPyrV)I9a09#gBMJsHLa2$t zeKOQN5^AMh;nf{yvBIdGTs0S%irf5YxvnO_9Ffqyt)=TN3B)JPhwaH)qFevsDK)x= z@w?NBlz9nUP!w`O1|n*g?Mh^aO@WPZnh@4D9_L6!!Qyof*`40g-2D#%KWW_ocDP(t zBw@TylyP+vrWMb^;{H4Mq1l{egyvWI=&i-3)w3bbArQu2{cBTat`C2|RFlwqGgz#S z0_<%)fcwkukPU+RNMVfw9+Dr*Ha`nua~@hiNq`cq^gbt8>3&rvo!Y=1^U*=KzO~pd zFz=@a`{MBp6QQuC5JQCd^3jbG==^uSl>S^oi-i2!)ba`R%%WP9HNSvmbra|%rBF6z zsGi7b@DP-vH~HP8A%S$OB!(90TcNjJ*ol=AeY|yG(bbt0e6b;g?ULtY0^Xv(L zT?tt;i=y5mHEb?9hDT4=;MfWduSJ#c9FNenhl!NwwWs8I32y|#x@PPPxacb|Y`wG{$CZUmDK*-I^&GqIp%G<`61 zHdcKf1#%gbyw&G%WioAGGlmm!8>-v)|}4=^gtp>1-bHvk*8^>tB=d z1@2fcwTtVhOg~de)I_pWa&SFYv0j@$8YsyvJouZ)HH^aK1=={f5P88b1spM0Hy%qJ!bxsa73kU?q5bO|$Oda& zTup~EuMCRMiciy}D=Nu4Hz{oS^Bsdi6;W@*RJQbAGmg2UgKf9N>D*mA!9L_L)`rK; zaF;kqQ>1-@J4doESjtDKv1pZ+`>)#p7Snr7D=Oj&o7{A3$$?6sISE3jTJevFo{+!7{V3_&azFF`s^vz zp~c!#bRX{y+YOe0Zp$S|;hMooU&y;}nLwOWmlIW=R-AQx8ZMvlgGRnHm zx{Lfs^wl_+wo&j{JyL+mH^U&%%M?;Z18p-lBvU3y;{u~R9P@q{3~o;&w=Ue{jOKi| zxpX^lT z3o{xe*o(ItaMnH%G)kIqeuKA6hbqU$D24z;Fa@mEKy0uvx)j>Xg-$7h5bc;vv+9X@DI7g zM+M)-3p^Dg#u{8I=)M0^aHgXj9;y&L#cw*%{8>6ad8339rEBRbzjyS?^wBJ%uZutL zqQvI!eT54L&vA=Y7qP&{KI~qVCHw2&g0r?g=j!*bf)|bHxb<~0m%T;_6zP5bT(^-Z z_;(~cPB39T2{(z$cr7eGa2?e~--O$^ zwQYr0<{hwO=|p_5J{j^JwP0u2TXg6?h%4idz~I*_7*iFT1NM1BtxOgo5uPs{`s zb|}^^HAhDYL$=|R!1z_ZKo;Ni5}g|U5NhWcp{np(>MdW7!E0UE--UmKoSPZj`urLu z*Y?wInl{8X$(<1r!Q*_$Zv^`Gf* zFY?mBH#3eTCmVC-A(t?Ijj_m|>WDtpCvm&N>gnGZA~1M94?NOSnH4L5t)q0=As>Me zt=0~~14nUthz>jQ#{ndJEJbBgrjze$>S5@AJF2EeS+P}PKH}VSw_&|=1=z?{p_0Z% zJiX+K-~?F58s=|AGt0HOX^{dGn;g%7RcNa68lxkMJaLg4N2~)w(`@p4ss>l9>jG=W z6ygxUX}M%fBel@l1Sh^o;KMv;Dm5yQxD>BsLuIOP)}K#A;f^Hpx7&}k>L*FfEJGGr zbfu<`SH>(dgH9!&~kCfmDl!VqIt90Cu?Ai{)|u>j=MM9a zk0M(<)ZoGqA!j&s2a}j5FdQsYQF)UJs~5O*BcrWJxQ;1IFm59AcDCV@=NV8foNd&D z`*26)ZdN%gn@!9b0?W@@!-2_XNyOrM;`^zX#?sB4QBp6Fe)W#u_|%ZR5*)Qj+BX?j zrcPgkt)|mdEGTboN#pZG`1eFCj@4_0X{)-(u#US#X;Lo<+%T6mSWjZBJti^9VaBk| zp7L9|FT6lYc>yzeh;#2bsEAxcl0xHe~SZbMd+)*OAz%UXnNs}6kMx`!?pF~IL` zZ{=1^$m6u1=kT>3w_@|GI_~xnj`U}0VTW}m)jFvI-f3OLdLR{5$T3V{*NFU_6Wre0 z1*A-E7HPD6$L*Z+lyt>M*bEH!L`S`9;xnb4-t@G@-o|?DjCF&-1HY-!+AVN=(id=i zZGgpg0n9yoCA9qN#nA)Z{KCuc@U)7+@)^=aH=}~6I@ud?zG|b%8d`X=4 zXydSs!)*M-am>WZn)vILRr-I*CWhHlasKOh!v4UCnVmCW7WZ#}Mmb^qm9rtL!CvIF zT}ouupg@SQXIS&B1kDr!U_yy64)|N3&Zt~cj3eN7jxl}mC4n4hkKqja;>m(9F{C|s z4U1BG#b(YsK~t3Dc=Lc|WQnE-_J&`zF+3nfGdWwzx%!bye$(h2gLpckvyh18h7qk& zJ!)Z5Bk)-dW9K|ck>7>K)TFV2ejaWD+7>Gy%_)O4K3@#7(i&iQ@FF(M3joh?<{;~7 zf$!?Vh1^9hYQA`dkvGRf{~;}88=A?B?=F0^Y(BkcD$7=EwxbT(HeAU5l@N4X9>!^W zBO~jrNp5u(5igggeC!fxrs_@?MGnI)ZZpYkpIP90+m^l=w4#!W)A@192AF8oO&5sG zWXl|8eU&L)@L zrahMnV83ZFjSD#ogVG~O!2TSP9=o49#GHb0s}JFekRA+Ob%duI3TZ~!R9w1G2_`); zBI|cqV|&G3a9k1!wL`CxV@u;;(iedp@p&h?b!j9X>T(C&J@3iLpSk>olNw}Gtq#{d zp`T<($kN(-hwxX@2Dr7toU9T*#rvkb#SOFGlRKV1kaD7r+xAe7h0NBY(8KRQ!tKq^uoXMZgW?bKc;rF;q*+l1KN&l=j#@ULuc19tX+NzB0K&A zuNjBn-hwdB`D-uupgoKkp@RDC}>>^g{}>xQvQK9RWUdMKWH z^qS249zg!&5FX9 zEOv~}beqcm!=Log`{OiXNh|N}v=zE_!s#U^A69c+gH0cDhc7o0c9!PLc)6F8aQX9C zNcZt1sVkF&2=w+k9TI5RaKZmPaI0zPba_h15mx!kS2}IB^JR2sO9g2TaupP zn3B1$Njs2R>zswRM@oyLl(k`5`DIun;YIv~`O(wBBjlU93FciH3K7B|=B-Cp>n^&Z zARZN;jDT8~>16cS61pWpg1v9_py8MHL&2aFY$!^H*2lx}+rv^Q3NZv9**`Syh!)M- zy$%y&ws4R1<_j*matM-Dg|I#ch}v2S^DT^sqR$?#$wZZ`tVyLm6Dav@phC@q_Hrf8 zkE_(43w_kB(NtAzDW)$B!}*G?*#GMijErfgT7`?DIX4AWCH>$;bu@kWpG(#0qMQ5} zVb)!-4iB_5qFXl42YpXC2W(_3fA<~Q%E3e_|4;K_eu{F0Jc2J+FH4*D8qwg&*;Q^%NWcHdqu(+b zh+WB2YCl$%ezaUiYf58qu+3A*42aR@&2Dt;ifhDSbRyLZYaoS^hiE~>FS=lN6xH~1 zfp+`k6Pw}B$aiTouxJV(J3S`jg{emP^3HgYY2Sy9Z*~#i#s?G+b4>c9IxAGtqCLNl z+mP#>G(2`Z>{}8|rd>M@r0x#69d!x+UW|dqv%|qx*b6>yH6ltq%Z2Ck6&%-L0{hKl zh^^Q|vejii|1-cBqrW%^%&OCn;QnebljmW#gBtAInMV_>HqlneHOyzYGCW)1 z&)q#Nj$>EbLxfBSbxn-t50xfx2ji!~o+3@KOV`G(Ys+}=$GOBxn6s~W){MWUZqdgx zt`G&eR9bs!E9Mn6km2MB**;keNBw+G#=5SiG*K1?e-%O7T}6I`u-BMYKODYtY z&7tk5ddT{N@2WDJ?{eC06KQ=)0_t=xW!71aC{qBir&X65_T?fg^o#@fKklgbc0bwJ z@tcd!cqsIgGq|^1SzvfLlpeQ^r;VboWaz?7-16iDPF_3#76>`OZYeVw6{QR!VK?~^ zBgwC~Z-}d@z;!;~L3b)nf^F$nXy?vqA%mEX)~{~h&BkE*ctZhs_s#_#u91U{bDOG4 zpO{e7g;AVt=|kqZ+zk>?0;WF5CmDnDIjxi&a$>?Ovb|NwrfseSJl~jBaa>Is*4Q7# zCRZ6)7UzgM7YFF)q+FU(w*uRB6CmXGRpQ!LN2cfA<=y7mQ-zoNapLZCC?o#`Pf2OA z&YJTW8gd91|5yvj6>Irh&(H8q|8W8rM#x;x8AIx86N#+tCCWW3<+9c+5Li0ztIqrR z3Co>>q~pjN-oYmw#T_GZ8(ht+7c~xY8>2z{x=smuAjX%9DmsolF z?;UzVC5rplG#hGSKJzM1&7kR7Bncjugr83cj_R0iq(_thXDi#mSiX*m6}!Tjpcb+# ztB;Hvug7g!y9C{a3$B8k19bY5LvSVN78&`VgwvE;U!_&}AL*E+$;Q}LVnetesT`4G zb1E!`)DIP|i-X3bvhoOhYQ73Xesp2h$9E*ueJ*ZOu*7GB(cF>E8zIQ`I?27Xic8f^ z!QTR-rSe-bCTo78cRr{ysa$`!d?*2fp@NSWI1;Jfj#HGdrER}d$bYA&QF*m0$Q~X^ zj)}_ohtCYjqbJw!(A8MnBk(Xi+>=S`Jz>_UkU;OMEX3^tk$6PtuCK{c#k^MwfD@aA zUA>zzcT6~yExJ$3de71LHA9%#&0=yXD~VWth!gzDY4GmSM*2m_Q|xRVM;AAqr5yHeQF@;x$Eo-!H()G9rkwTuRMq6wv0U z67j83!7~kKx%m^ndz|+108+=Y4dv<16AV zz607urV$vXijz7g6SFl{L`mla-G5$+EuVV2ikvgxeS5>m*JNpK%IY{;mm&kZ??!R~ zeds z;}+HbXn<)g6?vP$hH_<=^>Qr^G1h`I&uo$)lTBXhS+duL9N&EPA!ixBkCvPM99|W)T6&fJxz-~i*8eV>g=k;2!<9C`mZS7(@8AE&Ps#ma3yMIyA&s3RfwQ!1b5v`5>tu z^!-sus6K2=ygY6BkHzz8he{H07e!Z{jOa(r?G&G;7J%-LT+;3nK^|P%MfVyPR^1n_ zi`BPc;o%PvGb&D|gKND=$M8+Z96`cyv{#L4IDH06NW3$X7$CY15fwzp*A^m#QAJ?^o{{_E;vQJejh@QsxJCG zv5CA^+fJf|{+vN?VAZn^!g+h;XwiykCs}rnz#l3d2J7QH$f3p(qI)-!FFaC3q;?5C z;%oYtshC9fCsxp6^ZivrMt$MV?mR*lU6TRQI+OQ4p-#qmey0P%IdadbrL5gL0S|P& zAtl2JnWz?xu8T(D)=i6Wy?7^;zn4l57*V`1>=bc$HL+@s_zGGhokA1WwUYf>pKK)3 z=7P8OAG-La0hT2GBs2RhNs*-t=^d>i__j~bMd>N@+pA#QZJq;d?vgmNtpl%C)`9Fw zbMo~~D!;BG3~y*YK&v8gwt8Lz@pbq`=b4+)-bJw>xjF>Y#mu3m(r zFNWJr*Gb2lKV0LPVp?4`6x!@QQzPqi?#NzyI@8&aduulzekWxRTbIQc_4YDn+GGlU zx7dQ2gcVEqRgM*Iy=3yMzgDxRH*q(P?txalnZ$FM1$vsxv2&}xP&&g2?)9gmo=lv; z0`248Dh+8b+C~GBNMuAITSa9g;e$xBg=bX>yx}56~2ISyipfsNCSp_nxoAKo}d8V;4z3SHW1hi;P zrptw$z_*MuR7_V9y9OVViPV(dYBnHlLB^0L_l-|laG5^6pHA9`2GT5QOgE0xXDhYe z<5TyKRNnppnK$(~ckT6NoNa7`^9#E9Thq_e-}aK|n|GGHIy(eYZ+XGI4Qbe7QikM! z2=ASq2W}?mb{e)TLA_@;v`rf-JUv&#{;!*1nZq*tj>=40<_sSFc?hQe3E>*f2H2S$ z${{}=Xrg3&Jf?2mB6!H|Kpi)f-4uSKV~iy5-mlqsbWjgV&6gu<>OzM(`(df4z*G0q z5PI8Y@NmdPyd_$I3po|=I=mMf3)+OaES4CQoyV%`Qh^~kkxlnL{tZF_p z-Y5LK_C<6{?tU`0dMb&P=%x-ObD3(uRZLgn;o8E>yp#16I;!z7K6c8c;Ymq&Ahihm zlFpFGhZ)dk5(W1nC&9$NY|#27DP*2s(8qs=;lVYFtFM{#;aVp(=9WE*(KiKns-};0 zHk6T@O~0w;1>Wx724@=ZcMUhPOPgp2eXq?~qwE@7|8OKqmLw-C*f|aShNDYo(VKB) zAkI=t%ixrFAn z5$g>a%s`1_e&w$*Y(qBP`spBfu2oG<0_4yjyN|Az^@Q{${=jpM|KZeYR=|b%g7N-U zLSOzKW`}HN_DiK%D>o9hNiJn0TD933A4dvP7LuuMVZ77(D74w|8Jm_Fq13ZkwCLyn zzwFZ(NO}x({+c16RGx{Zg&lM`mx``$EZForSLy6cw=rWcp{BkFB{!S z9@ZV;A2ppNvxVQ6{F^*^acDlCTqK4wv%1I);|ZYeEk=_vP7(h;j%2HPV8z08YVfU# ze<5Z{vUL&HDNyogOd@0%yysU-RNJ6<$yDQR89$ z!4h@e^I<9-*D#D$OGSfmXd1jOTZvyfzY@PCLttZg5e`09BSV&D0r$g>eQ^xoJ?bjq z*dAjT@H~qT;2U?pcO*U>Ziqg=pTUXgWjKDikb5_~2NK!?_`u^Y>Nq6etq5nZnWcn_ zDn@~rdoNl%k%HY@r=dvs2np=&CdolNxxC>fETE|cgLVxOMFn2Rdmi^u|HXS8^c{oa zmL%~eX2)Q_`WKgc&yyBE@Bv~Xi7qoc=ydx<=&0#Vu;LdTvF9{R3OGn=Ho4KCmAZ6W zaV5q_b&-2J2k4Ih1(;e9Pu@t(u=~_>h1`)i0>z4MWNh&tqVcGkOM0w; z5B=*%=dx5g#n-u1{zn|^T;b2c`Xt%DLTPrV(wMogYk@qW11sCqgbNodVpZ=@qBTz) zhMKQ~V3$mK$x04>*P5V1Lo}XP+)j)+fgg6Dh#py>E&7@(i5ltexX8mZ$(r?dc$Yt! zWb7Sx+@W^?CcJE==QrBZ%ab#>fNl}1S#<$t+(^d2z&8B({TOPMZ2}+p>FltIIJ54U zMKo3%rVoNtC`p_J3Lz$J@~BEwE4xV><*TaeYL&6#_ZD~>v!A{gKboo-8o|oB;!&>T{~%aFb|= zMT5S@B+zJ7rdP7Al7>y2@%a)F9c|Vl<2f^0m?`(|Ncwf`wfsKQ{?IRr&{3u@GSJ_Tfyi#sW97P2K7+xAlFY! zfN7TxVaKww_;5%hO%2#e+fH~B#kmu>rmafwrArz#v;*MNBz0K1P#!lY+rt>gB140$TS4lCB8a@$N)`e#9pw`;KJ1y1llZ#y|6@cQPe zW|Mrais}Qmep0Q|>zG2cJu_Q44E{@Pr4dw>7JDd@7`;uv+d-KJSxHzy2BMf?QwZs(Gc|nDu)kzGAyoBxfLpew1A-H+? zaoTBR4Tqi{z^h;Pb2sN6<=x-TB0DnM$;L8k$P{);DjI*tEB)H4bEp52OHsD4aLb#mJc_RwDUY)9Q#e^g$FU;{h_#4 z;4tnp+e$Xo-sS&HEWi~7_i5t_Q>IsG!P=B_S;4F^ICJAAVYjQqwZ{A6IZ-3sZ52&C zZ~P(8dM)9P?>Ta|FbB?ioC5w7rInGRu=z+9Uogv=RZa_J(koBnoXFeMFlZs@8BKy? zr9$pHBmnn{AHliVD{$c|4Uwyyn`o@dUn;7%2dBF6@X?`1$QjqzUDgqyZsmDg7FvP_ z-3#$v=Sflo9I}3+7F0$1vR9)wvIQg%$M0JVS|9^ol2_u@&TCZZ)l4F*nJUbU1@z{Pd|oYt z=c+O~DR@X@asMUSK6g1`rz*G!s>sH z#bx6>Sy+z?jczUDKMWY~m5+)A9&ijU{#=FG#DJyw4P$#w&c&^!f2rNnx!mb=fmzi1 zfm2gn!To;O$WOR9l;ws_#^n-<1TUz;z(g~6G$|M~-&NA`4k_$zn25pDS!KHT?wN>v3k|1h9zEgfEN4>F0cFU4?10U?PD1C~D`e02 zF}$q1B93YgrSb9=+}|yYJojH8SgT!zDq;V5w(SMU@zg}`>_n3O`68cJAcKe6*O1lR zMXdE%iT-jTAev*?&pl?Wcg;hb^Ew}1-uQ`Dzec06yBELmYzcpg;}c<8>U8x}?Co5abYqZ9e?Z^JNZw1#Mtgd+U8cv9E}2h;Am z>L_0BM15=W>?&I*^%~xYpNi&)j{g}B*(qb-_S~0vuy8DyW7>;yKbN7!(owkNi9H$% zWr(2*2o3o0i*%+xgUpQ+sha6qu5d{MsBdu=Sp6IcRM-a6*NgG_a31C_RwmUus3vX zb+TCuR!B^xmOAn{?fD{9ZX6Adgsi}ullO?6%ry$|nPh)%pu>f1;v0cY96VH(s_UKu zr?L51{?P|BGv4xzef6aGgD!qgtRs*c4K8c^V95_Ln)uEIN7`4>(`rpLBX%85yH!Eq zq%n;2`%L>ZW2nlW>wM`%#BZ1Ppzr=@6iaEO?|+ZLFL$arZmASpRr3HY*AT5_lmt#- z0Imv_z_x#<;Ye9JvCk`o{7d6uk-N}^mbwZ1-38{~^5gt8?LwI8*a|!SX2ZE%hsfWX zvaq)N6uQm7%&mJ{NY@Nz^Yf%vqE?n6{5aAb>=LZW^|7xq7DOUXqW<^>k-g}Q zlP=paY3VjxBbN%(;^tt!;Q=9gdY4Z9n2f1Xi*a@I4Nx+cqg!jIi1d54AVc034Bd5^ zhWCAV9drzquiswXxYPlI%Zt%?=`q|H>5lt%6`MJ#)3CaXd<#8R@QQM z83Qq#EH|Hpj4@#2L>X+w%I~N@d^eip$xy_D63PMjD!GtChtQ&pH=nV{?RlVal?JmBxXEQk<1t9#aG(dzV28;P*2 z>eeM*RQ^>a{X1m=M?Q@q1M(Iu);FIw>FJ=m*Iwt^#Fj$Mq4y-@+(rESXBBSPe~X^; zJcbXozu}=B%h;f11h0BK7j=^#ae2`X$qN00+*N)!?9l#AW{73dHIAjI*_%M2NR3t~ zZRbLc??dCav(c>WBKIjVk=OiFNY}2r2M%W8WMpdqdIjC2gBQF7e!nUG?w&|nJ{qC? zFn3t}CyZ;=nJakAvRLG)HSC=1DRyJ6FMDjZjBY;BO^%xEA)jTkNd1IFUSIh#rr`H?UmtQ?71bBBH*Jk3ijEb~KyrPBo!pqO53y$|CBex)E)b zPbQ1c-hespU=F1^j2(v>5r617Q9*Z|Jb}CO zatU+1j6Y3M;CaJjc%G@oTf04?`AKQ`-~3kY!`w$C#NsSY`1=G!ZhjEq(ZUs37{TrY zJ6iATjXv?tFkEg3{pQs~nQRfRFOq@i_&V-de*i8xxC@;6z2U$lf#WxMHQ7Jo7OE9A zVULirOS~TrSKTKOuj(**pg4%mI`NGJB<_IW=Qd!om>wLrR3u`X_hOV?2|9&75**G$ z$lc?@3^idSU2*#@S8`NvWuzTqBgOid!=(XUsy~6wC=DdkQUMxiGfk?OXEw`nF}7nj z=3E=YrdmGc&$|Xd?RSAyP>~6y&(nxg_(@zkDD z5itI)ns8S!!nzC3$P-yDcqAEvBPK;upLsJ#jn3u}ixr39XjWcFvEkrQG%A#(C$@KUQ-n^>_`N`mv z8`WRef!k$%6S3EGA!yim*igCvk4Fx}qo=1b$LEVtS#Pdrrh*8^*9)xCfpu^yLxmI? znn9+(Q*RjKh}l8YiDY;s>A>G4XZ$iQ$a5auJ3O1*n!Az6NHTQrJK1ta z0e5bC1h-TZ@TPM(Xp0Z0Qdw#IpNA7!sr@(_x@8Yc)gF)MAKxK2J}!mh_jbe3oGSV; z)CK=V*dZQ!3m^RM^Z>#l_EsfuKr+PIS@L?1^*jYuR{aeUS z>0VAtHJ$(Y?=O4nA;lgD+<=QiN-?x*4xjS*8)=(3iQX08&oA)&%vm2ji<9P0WiH?U z!w=p^;Pc~WoSA+T^~#hH{n7|$-5Rk>YS=CA2-nN>^PE}S);xaw)1jpFm=T=+dIy&X z4!rg6%%RX=H`h=j%WHD&T$zasGre3#rW_W7<#%<#;7vZ6INcb__|-TzC>PVyLm>KJ zE**BV#%}i6ETNO3j)ylLB?mhbU`Sp*<=2nJ-8x?UvN=gprmV%qF$b{5UP z;{bi3w&3;pc@>WbdAlApcxq|}W{3M|aJCeaRK1P0V^ayewviu9Q-S)*znuK8W87g! zSs|w|n&>^=iPjx!Y+wGkN7;iic+t+YmKIAkbMFbZOJGbWKYK;B1M|3RSMH!Qk!0I` z-^7kd{^+KD9`4Lv0SdduvZZ4m&<_K-B$mlw-Dg?a9BIjH()1Y!OwJDj9^9_^!9;!c zV$hC~M9=yj-Z|+9go+J8*R2O>Y>2?QS31Vu|KtPNPiD~KW0SBpZ!C2kZ${ob8?y3C zDMDw}1f;)Z(+AH_(mxL#kyh;m%%{k9lD z7tX$p=;7Kv-h_^QFG;VJlW;dlAs8YKd%d*qT*?jlod{jAzS-QYeWCbsN+O?gzn^=N zl}g5nN1$6m9Q8ckLq;y}<2_op<2dGy%g+kmziv6!5myLLCmv$$cIWZ2z~PM$cypVj zMCic3L64wt-1StsWYhb z)flvWc98TxIYaMWY9f=DL!;{%yzGa3ZBPL zr`aN&iYtb4H>I8Fvya*IYF#sl%P8Y#`mLj*W_6J8nFiJ7uNz76*+lx@as={|-^JQXA5s^RQ}S zjE8XJ(51$2O3A4@JresnhkQi?vT;Tztz5s4jP2r)d`PidaBvSSy`>1Ud#2ER?}VswDdAi$5Qd*u;lFK1An8aMf(%0zOA{ ztIBW5VSe}wds=Rpz(2Znj06=~lE0T<)3|^~)U0r#T_xo6K`AO^%Iv>nuwyzYeBzEW z^eeaU+EDn;&EW#OipbA*rtm%`2=qhlkOi@wBrmLkR!4P^?V9%F)luQiQQc3Ce;I+t zkqrLC{uktPVK*5SGMjD|W|Wf4OR0azX_8g^jPKlM1oo2@X;=WEuli?^huQN4KV>=R z{$w5MNEdMDJExFCGPb0-JC^S2*vy5z`pNyAr%ZknawH&5jC@Yfr(*RjeCoDHyM&6X z{Lm6Na%sCfy?W^#=eNR#tUr)I=e!ggI>c5Y5ZPDS=MQI5@qkgIw0{pmz^!!HC<3 zxatYp;Ais@?%w5zIN{=ZK1S%Zt-3m%eCo;|mu^HTAR5g z^EdJnazgpUzfyFGz`;}8uSD&;AJF4k;_TJhk64&{osK&x{C$TQk#nY>h3t0?*I=`n z+)|Z*Pku5qY{y94{bfB}?t1{uZbXtn_jP2>xJEMcU>uSDJrYcu2#H^@673#kl80OZ zar7FAIdF>5|vRq{DJI zyRgB>F!)y) zgsxpOB);qry{a)5Hz?&mvGFw4_jD-u?Oliawj06p(H&evup)aEQH0M{g$upStGHNm z2JP2OC$oza_$$9JlNg8Vb`$R{;J!D`AkAyWLhmqTlsPEH?x)+b>r4DW(96 zbm~B5yd>mnhGT)vIBueFui55(j7)lBiDeTMpn1a^96xd-{ugGBFRaEhy{+n~eq$JO zunn_&t&|RK4fCpBU#lPyfvRkS!+2)4Wg9VdwG)_^n_<0-jb< zBI(Wd$(iU5y6nqk+*0Do$`7leQ=KI}+3v!xM#F=rYsAz&X4t8r{ zlsH9=!?`%ASp!VE|B{=|-elT|GFsICf^Jw+MDD8(rzy!p_}`y9tFQYh!SS0LxMsBy zI(z#EPP%Or9>})m=1j}s%+9>yE$&3Y8Ox94U(ZJ}Y|#u%oarlY^&P-)(h_=&y~X&u zmDDaloZY-%$hsfIv4;^sIQGOA>g8ijW|SsloY7?RY?=|CZ2v%Ig71=bZ52>ZZbhEo zodmT(yKTh-Y&pHhvJk#JmJU(vz<>3MqI;YBu0v}%OZ4E$)`3W!%?TSgJ1WcfZ9VEDVekotN*;_-z^uu zvvU_$>nth8HvRp$Vu%L@Jk?=8wr8Qg_e(lk)>2d$ycS2$@NQwf*!Aao? z_uPFfyc*ggILn5?8i7M$IZfbqxPB%7NyhR<9(lNQM>l8l!x`L3n>n?SW%M`McZb-r_w6Be3UqI`TlQz^+|d&`zF6)zQmF){^?l;5QHZgBW(Z#$Rx zpYSb@F2pqx>q+4QalZfFc&6+8maon6dIJw#?@jZqU^rwY}+Nl!XiT)O9!U zoLWUr<`0qwB`-b_-+u=kuv!fGY( zRLD8vE;)b@)+hLlJRsyu9gGxi$8GEqXL0Ikb=LML^s>?@NG~V!lY=?#_)t*Y*tigl z?yf~~%V+#muVjn|IXIq_MVP}|h`Y9iMMkQy()<6R$>$4nN>nqsGNl_b0=nrVScb-@ zbzu3ar_^FoB5lpLWnYxysIsy(T$1p_(25ZB2_Mb{Yud0)zNNf(W-GCqk&%Ak?;LGUx58IQaAw5hTNcD{n5?U9x1QTdzRfDo4^7)CgU1>)C7dDOl#>ioKtY z(2$sTy7{gFvp;kacls~IRLM~Md*=IyalQ26CvcSdxpUS)2|ts+wnoM@;xL#BH1osVj3$tI+eVC zu!pb_lhI1W3zhc!SI2bh!&xQg(6`1E-h8oU?psd_UdazsEHj_W^woy`(=k}NmuEpm zg>2Zv3@oM>sM0!NclvQ<)#d~F?1_&ob5_gX{+0+V<5_j2^vPWEH~$nIU6G5gw&dWa zN)vpZmku891=m1TDOOB9ig};!;$)3pJe@Za9Q=2256{K3A^!4g-xWJ@->aC!)Qy0R z&kR_{-k;bRw-ZG#BYCBbcgZKMdGy7N4LEw*DA3&{vfDiLmu>MzBUYugh-z=}m* zVn;gThx*T;n(9?tihnM=P;>*c@`s2ts(R5tD!DoPl)c&GRzFGBK9Hn@ZYf=kOy+TL~|x3DRl}XfRv2)oO=} zB(DF0k*%66{$T<(8~0&-*Ex_@A0SF5I&k35JkD}$49oo$ibv;*sD*6_8r{wz-&foP z<@^X5cKHLHIoW{tz4fQRJYHePo*p`H#t{q})q@85DO_&Jcbrs{joXG8!|PkK*^@20 zWV@#U819l4_{ghhhP)=pZ7m?Vt4>kvZO5^o-ASarBo1#?n6MUsw{UHLFdXo;hNTTw za5ln#3CCh`uKGN&Prk^now$n**+`h4{Z95vRp1y*7dkXX5;UU3lX3-;aQonVa%J&V z`m@&t=SnJaC;xunbtkx@cR67R5s^@AJq!+*d*Pa1edyHNO~&L^a-Lg;!N-|7)S6jSz%28;SUy5^1y>>Lood2az$CWPUa zGJSfhHIsI0Df2t`j00V@1!!9{50qp#0PhlyQn?@bCm}(^Ize2NIZ)1@*j9n^%M(#4 zuMA%gzg9i}j1~9mv<19)J%{YvnvU9wuM4i0!^r(LgQNMfqQE;xIn{z{T#k2zgq;G& zJQrqhCKYIplg%HT9|Lq)xqKY519!L`TOHTdq7O{}$W<`R)h!UA>l8 zCMTkEaTP24`ItRYaAU)}tf1(m1)R&A&#J{g;o%2aP+DOHA!kdN=yV9F4{U4oCTZ6v0KRucG({!M|TCHa%mN=ovkK%Ewc)CRk({Z-J`K? zMkVJjl?o3oG=fWxGv6!0@b}^qr0Vut613Qt`h19n#nXJac{`3%hqGc;dNTJp@9Z$} z3pvQNdKeBcIb;~f55LgL-*ucsU)p}g!J02urf8joxr6D%d3-A-+L%E9neo*38$)}=YRDKH!nWHF71d2l$BB;r zNWkz!lB(wg<-aJoR562gZWm)W*DEr4O;2XC^d4<(zK!$8rsBr9{p5$)1!A&=a9tOs zV$5ScUu)B`TU3myxWW>P9^9eM+Jx{AT)9}2`lPdjx zHe|2gN^)7!5;flGV9*dkM`(MqzOh&xOw5 zu;ao=5~6n%3MR>rkuvvaL;DgM)|Y{MeWS?Wlk2c3Y%;vEodCyPHX!W8s-ffW)5mus z@nL8(H#8!Nx;TYmfVu@%3aoOS-8I~ice1c<(IFTb%82=)o%B&pE78vogO1cHSeY%4 z*|kUE`o=G0SbYnOzZ=;xrAa)bSAwcxdEID8_85%o*qooWwyf>!C+(p{r%@4 z(R-ame9!d}hY?%pjUEG$#o0%AecU=GK5{wCRf@t7>cw>a=|OCnAh?Yi71^3(OLp;1 z7h1{vMMrBX+$nU3PK-9g%H`?c8FhlJORK|>u@$gIq>sx*Ww`fzH6^mwxUCN=!F%B% zj4yU0OFdqZkK(o>@zwpr{`(SGe!hiXTAzUlK=X5vG9Gsm7@{q~vWI+yYq`rC=`_dnIr zXA?okMGyKMKGOQ0SaR5+nkv5WK=(I`VR=wFsc)H#$M*Wr1(9b-aQ-~_@AN>@`VAvt>U?SJj?l!B3+7|jA#wiE&b{=XK@S1j9GYt< zCsI>=NX5pdke`y}p?;b%;y^9g{9Ji{)v=7l=Vho@DR|TGl zJKcO7dEY>1-YmF^>)8)$1@`&VH`ZMdj_w;gA?}eT)CC8Uewht8H0&-ZzdDaZoGiks zO&3X)%W1l!IfYp@iI~^k%k*bMIDKFy3!`qRlDh)~d~7>l+V&Cnb=_N@szg!wz8pU4 zdJnym;>Jh%^>O>e5Py~DP?v+VVSZK;%KcG*XD!ZfLwJ8ob@|Dg&)m-zE3c;2p1Zg> zWfPL4u^+RRRDih2R`5(a4z~NJu`CmP){*duPIDQ`7l+*;J44RX&cLzod2=dlz8Q(% zLag!M*8pf+WW`DcS7K^L1?(G4po(!E-4dsSZ;z$nv*R22{GZ8?_E-fz=nWC=xbTW> z96gzOtrK$5BQ8SQ>jt70n1SOT#xPm$AK1thqq4IpKl@HN-STc8&NJUG5MB&0wq+H0 z6XZop_X#tA-fq@8^*H--@(kq%Gw6A#yVSR2Kc~C0k5(f;xyC_?s%Cf*`Qk)BI82UWp;cv!KF{L;*(7CPnB zIPe5r)w_2ey@<_3zRIJ_*(x7UN)$nTZ5L?X!mzWb=dgzPK4yK_Cc7(W8S z$M}$80zYHQ;RSG6#+wGe)dF>`F5)ZXpR|9gikg3>6w`1)f?Rty%IN)1FU;11H1@pKz6XV(^OizqQ6J z%bjpobp%uSFdT2(Si{DjI3oP5Mq;3YJ+<0@3f8ha;IVTwcub37*M4xle&wq~wQqE{dApT-8O zcU+;%VYE;9Lf_V%Leas?+zTPIQ0-Q;Anhjc{jMBL*j=fWOLI zEO(rRGwo-P?SVO1dBhKLzDtqLj7QEJ8A)Wai zkotHm8aggRf7PM*OYSML9B`ypzV^|Q5e_hi#6jbkY9WVfM#~!7xKm=O=&KZmYpPb- z^5w^IilZS!*C;~p-V5B)j%j@GVPlxt)xz~xB~yvjspQ?h;oz7m#b%Y9t!^woj5l8| z;JphjL9uYlf0pS^l&`(S(Wd{9+z91MW`7{tjl-ec{y4N($l)mSOXTg7LL8Boiy;%5 z@zmr}Vko$MR!-fC>%IGlLr)|z-ZTyCgwEuv_$JcCP68x2VWK*v0jK1WgL5w|gaiOzcT*1A97v@v$Fd?J71Ct&qpk3k>JpV}+1~R6>{2Phb%ADxz z_Y(Nv^-RGPEd>sJz17M;<4Izu8hM&qO3n|Qg_WN>Nn6M?qPnOL7hN*Op83)wc;RfS zT6i9|yC@OATpRqX^$n^r%HYen3AF$9SE4&9ka!8M`4)?BxH#1a-|R2PitY<2Nnc#u zeqIJlcZadNmnN}Q(*kkm&onf5dPu>lm&7{uak09p5YsP$N2b2?dd@1izeWk3s~17d z^Je5x4mZ1Y552q12X~KM T!!X~AI;6`n*e@rR~IDLXf zm;$)WvZ1NTtJz&CSC*@h3FAsG!tTqaT*i=Nn4h@`G{BUAXk%Bh;BEyvw97qnBh2`uwb=r-LGd|6en( zUbwY?8+MqTt@dLU5od^)u=_u4xpfdU;H^wOs z7A~9xRqgAcZ|6J~9(oaHgeJkxbPk_QtfyldGw3$Ky|Deq6l#BcGc5rQZCb}e@3$Xh znCEP;ZgItRkA*(-_jfRP({SN;{0~t$c8lEYX{9!?tI^`09EQF6L~i}Ggc(V$q+Rn4 zDs1~ie&vdbT$40K`MX-VBu`mzcfZKZA>L@b$)ewAlgCG}o2NXU% z$F`@s7$fA^i*|-{#$OqPMpi&`%?KJ^S%fcRC$P_x#|pa36!NxXG&Q&*$A7O3Cfki0i5G^q?kuHm-FE0>W&n{)XtH{|sbKcd8-=ks8Z!^$RHWG;*JyA)Y`zu`Zx5wX}I7CZXlyBf5NpNMzF$=XkHZZ3h}?| z0$Zz??g-Wcy)oh3yb+&>$)EN7Y*8ywO`lBKVz$xniu9@nr?TnNT0@cL`$#H&$%HLD zIR^FqXv1DM0^D*gllg|aRI%F^$+quQyx;)Tu}WM}dK!~d9<#833pmNFmZT?>uyM{V>T*1?AwRqb!n>4O+tFC)#fk_rY zAXD_nZt;&huq)IP-YJ{lbgrBCNxZ~C#YFIRt^gg)k?g_>f0Pq4i{7J0;QGnJcV2T7 zH19eES^uI?`^jqhpUpS=e)K&c-^b$2ZW&zUb`G=O6aat36m>F+xyf}SVe*D$c!&Ap zVlodS8!}YCngYpYJeS%xEo!2HXg!eYjm0{3Ws(|LiY~>*tmAJix~%AjwI1m-u}}q*1ulX> zT*Aa<;oPuW&mei}8f-`z;J>VXjV&urk%*BxG)p5Hm46?>XTP3Nx6B{a)n3XBkjN$Dh7VqKPF0%X1jLu?*j*=Hrq>nb_nY35piupvKz(i*HS% z&+Z$6-;`&#+h-DcWW9yic`pQy2w$jNKN?gvE@N>Ey3i-95ewAx@pX*>88LA-uk6rl z`zmGu`PeJ?SWdmfSs&*nSPudAZRno9O?DO+Da-;A5WWWL%VIUPr0rLWPEPQyXae39f|E7*vG>1@4y>{JjHbWN(@Lv8{=|_0TzK6Al_xAl-`_)6 zWWfp&>ky7#1{H)E(WvUKbv67bjwabE-|^!~AExP(&HnkU#tk#{Sf}4efw^r6)q^hl zm*OX=D(%TCM(DCUpW!gYv>%g39%m`f2Z^5Jcp-1KAOF=16IEj@+L-I|D4)E7R>TGvg z3E6+*5`Sh%4uD-LNM|IXcIq{}b47=(8a<0Y)ZU1Ss+%B6JDr*-jAh%r4>F&6S(sfv zl}+-Uh3~d!B5$;dS?}3J?@t;g%6u`1^J7M`cRr`dl%oRe*-f0O3cb;`n-uJ`*1@-t zy0Bg7B23zR0SuPS0mBS=_RM!Nmh}qEJ&8ERPg7&n`F8kv^)WJP>1beaA8479HT&(D zK~;*z;{Hq5*|*Gi_8?^+KHuTPUL9B<`nYC}NNM#Bj32TclwzW3jA03C2)7#V840MD zW`{m$JHbBiD6A=biOSms;Imu{9GEU^9@HlelHq^l!CT7@#T{$W`t3caR*3-l^^@5n zjc&Yi%ockUX0bA3q3`k`fp%G)qskWL@K19yiL?sD{hpWkRkq@yd2`&@=o4p|l!(VA z!%UchNg1`dI18To`ZMV_3Amg9E3>+RLw=~RE8U71J=9Osn^7-PYgV2R? z%mcwSemPrZUWHP(o}qJrGBZfILN^SY0Oi3cEazl7UbuK4Z=N(J`Hym7x#>roHYtr6 zY&*-ot=C{Oepj*fO0!Lf%mseuOFK4hzAQ6uJ%^89rIJyL+wh^e7*=0D02g$3k`W)| zao#;uQFrec(eH6*VBnjakQufivhReMBIYB`pEaBmOo)V%zBthFp9UJ|WO0qTE!(qV z1Tz*KI!E2k!z}T1w3sv*_i1Ucox3Dp*f`<0-sT;4kDA7gRUgI5kXbDEsv}4&*}`i1 z%Y27RGPXWg12+zxfHS9t;37q1xLh-g-Qs0oa#c7cR13bH9WQbIe{L-9c?gWLd4Q=K zU!h#;Xz&WOB}%7u($GaG!DGDySEbxVOI@$S&00q)DdemkUbs(RPWnWfn!oV>b9Sa- zHN8^#(Qb?lAvnHA}Y9K=z5hK0aT_1~2R{SHg4LzchNnnBRizU7$ z(?pio?P^7}@i?#FA@0~9Ut>;l+}<<)y5nrW zcP86!qruWIeS?3CVh^kke|$dGuVom5={*iS2ENC1aj$u$e6AfvU_7Kupz5*f&HQ52Nhm$4tLZ)n};_AHKFv-XO&UMi` z{Mpa|6BAR}HMa!r)X7D7ph`oWai-Fp^@h}Ks!!)%RYBC22`v8VPIkRopB3nAXZJpJ zQoE(RW36jFDSh@Mzp~>fIM0Lp>Q1=X&l;ltd9#T&Qz*u)j6#~6XB1Iz z_3jwx{a6V7l|R8>r(5iU!Di;Fy^kJU_JgDy?XbOLGHn`Y0`HU0ao6(ZQSsjvYHwB* zOa@=Wh7Ff^XU(VlHbXTU_&$SPTMrdpX0_qZ0eP%*r3B1O-*B^v`c@BJ|C!sQnZV1~ z-iG2k^=$LE09>h?%WBR>V;(=7IrKGwV>4z`L(E}R^Q~iXJH~)Y+hu0ss!j+TTtPZnzT_7DoA$UY#E=W`P|Z0LWUDnvJFN^d zbsxaftWN-)gW0HOs?2v*CATo~IDDuN;5~a4vglJ;*f;$s9(WW*ddJMj!eemXaEx8c`OIoR_^@R!MP};nulye`Z9&E8H5J*l@tX&m(eMTG6g_7f zMRk=zc8fetekP%Q_Az)*Gl{HXMzN=#M30W>9o7)t9fP|xnb)_sVE$$vd>{6j{}g?j zt=-cE5zCzEP0dJDAqzuL2e{l#hQ}IsS78FNfyyEY9Fn5bOyKry<>KIp0lKUQdsbsc`vI5dfVJLUsSUpzbc4RrZa;rAcYROFN@ZkuG6la@iyf)`ugOd05Y0ZE zM`}HC!X=&c)Gf-bu4moD1nmG?8#a^DJ2Jp&?P-3E$oYSyBG0;Iec*{n2#zqE#|({6 zlfmsrD4s}>U5*vE?W=em>0!wf46Sjfj>uwFDyFWJB52YnoVpTP@Kiw+s?3=~pJonX zTjfvEdCkLYkGvNZJSe9{Z$3!?s6=MUHB6~_G(GppgVUUJ&a>7;n-5pG!- zjaglG?6rRwSEp=EFy$%bo$cf)`7Ld~Z_m;it(fx;P0`!FW1e{S+83i- zy@idAUN(y@@4C;Sh3&Q-h^Krh*Qd^@nh}D_2_)#VVJhEo% z5)7c1_EcQ9)C6_Kd;j;U^U_gTg{(CGBmd23AxszV<6{+_AoOkz|83}f)O~QA77RN^ zIZ9oybwPhl|5gT^aW!K6`xxkWWy;#zI@#_A0vv1KOAAzsQMjK&3s1#!=YLP9Ri^C} z@9>=zd{bfJ>x=L$D}xO>aD+Q`+=yilPh;11?5Cn$%0gpEBhB-U<*#kMZ3qX7;A6w}Ty8{$xYwl~DfB!V9}$As`#*h%WhZlG2PaTqgFUOh_Z@rZjHgwvZ=+GpPrNt3NTN`c#Qo@docj8Igo1%v zAtTa-f=88;g}OQZ(isTZb<5e=6i4cj-;4eD`Jml@8y|2$g9Z6VVTEit^b`H+dBv`j z6!`qt}yzEZTQ2HTCo+&jmZcYPG&l@}v$gEI7{oevIOk zZC-M+Gall-wjZcF@h}udjl(N0E7`Q?^XZD7ENxDY#(}wMxGFppy?=~mJKORkGkU2C z-$pNDE5)wF1hu2AV#7qC$?B=pK2+Cn$?d=79v?1x9%|S`Suwxv5WFqZ}v{i8ZxW%@dQ z0o+l02hoia`F61@I@$a(`V4R&mmV7&y7e?!Xe^=u_A&5r^#pNmu?0@fv85X?e(-{L zO)ON-MElM8czUiJJ-&LGJKg?VWL@v2UJ;BfKfMReOwI+;n+nCES6=UNB=xsk=y2Y& zm+*VHw&O}mL&p_=r-4^_7C1a;#($!F*2y-FbvP%pVjXdAIsP6seBLc)e+IDa!^A*GhIciti_8ZeM0p3#8H$@g)eO%(6g5)Z8V z2-j5U#J4CnQ(N$Vv}<1vQ-3RF0}fw-D_3&4L0#S8bA1rh?%R#c(XJR3+lv*S&49T< zO2V$d5Lok1oX=_qY-Gd@GI0tN&sj>)_NN0hPfU^Q{+3C}4$81Enqw83_Ds7uLOc(0 zBv%6fy?|Hn@v?w{W{qS%(uUtuTuSI z$5=7ZgNWobfW5-8fM9_lU7|wfs^)T+~Y2P_Ls!W zfYUaLF)m_$!4uKx$s96TxrsH86z5NCFVNnsD7fh}0ZvUigj<`$PVJ=}v@fngVd6#B zU{^)cex=chjS1jzTFm%Zc+liqdudJnJUVj5oHh3vO*R|$()Po*P&H7KeYuvvm1{3z zog#aZ?5DDy(;V5~Ny~7RtQM`ZSU__Wn;~F*F8unk8OmijcE-#Z583+?<&Go6O^fl+ zra*d_SdV`f#?gp3vTStB6smY6Lmt*u^s1jWvo1Y@2W2bJ_3QvLpIS<@0HmO5>}y` z%JSl0Fqg>|aO#}MrcrwZy)*8?PuRosoF=eqF~eEtp7r!yqlzk(M$=)hfz+h%3a;CE zQJlp&Jd^?n_jY~%L`6v0!n2+D&dl6nb$M0Tyh~^*N57FO$!vIx($cGYqeP%7Q zkQbl3U9X_1_j~Sw*i*Os+?&>KkT7LyQ{10YhtuaMGw1YZvdD_3xXWYc`29%S7F>wk zvvjCeg9CmJmSycGDtz#oYvfS7hDIFd&5oGs;(XcO==Cd#-H{Gqnl^eE5>&$ZYv(eV zM;+Y2Kf@`h%tv@R%|b}~xR8BYlg3;WXON`#YWkpZfo%(r#kn)X*jU3d^mgb=H>dO> zE%$KfQ?;C1u24aF$})6Cqn2x1aFR|sexi(AU*18(6QUBPz?8Hn(DrgSo~kzF*JjPc zlxHIZy__}l_tJLu*;d)i z35n{9Vc0NRQu-7NmA-+b7u&@*8t$V#KAez}R6^s|Zsv7kck(r2rarBs8(oiCGOY(o z;MJs)IN-=N3|oGks_nw59{TZ5w)U-BQ|9mm2o2% zqOqPOIBd?P#OG^@1UMS_~4cP`O$L7LR z<1j2{Zg6{n0@_z7kxlG8NIIh-@laEyhCYv=IgKEB^9T;bE~4D+Nci~Yq{MAQAzc3P zgtuC=pZ7G@pr&zW@riFBYV3K4vWf?}uJ=lK|IIn6tNT*!p{lFHkwjglHAtHY(+|Uc z5jONB_$A+Td@V)J@5CdaeQ`u~I24Rd0-3hy+ytAGeAn{*P&iQqt%kVpI-k?=GX(FeA#(e#F0^O*s7ODE`sQLMV9L2L=_sahS!Lp!Saktk9Oj zr#FpYbMp|s`eF-rSx&+g+n*5q6DRnPO;@?4c{e!w?n~T#hqpLCeGBT3vLc87)}qIQ zJiOxN$ye45qn6ip6g*&yLBY#WW>JPKk3Q9st zv@1LIVg%f9v*khaH`gagMRWu%qIz-lj_&;dGd4k-y$J;>$O<>*qqzz%HTYpL9EL4< zDS7zafDN1b7j|{rhP}04u&;YF_T=lsc#HcI%@8raU4N8{nm^F{qXTHu1`D!kox@F7 zna(wh&&Hkn4~J?&AKc#9!M#Cx)VT5zhQVnZvO5n}D%{~F^YVP7b}r6cHw2x`+8tCT zra(Iv1p_-`KyAZFJ|pcb|D$aO{xc7gKE0R;DthMJvEY7mloNAIANJCqE$86j{fltE zQWcC0mQwxjwPa>6nzF<*3%!FS{Ked}+#rubywdL@IKzSy9Uo(aw(HY`3#kS0Y*rxF zi1&dbF&U^R^MErq^ng(4snb>@o9r+X1 ztIcCa;~Q|~y@cC`-7_IqE(tp_xSeN(| zFHUQqejlpXYV8jG&|Y)0wAT<)3{p_%X9Xy(wPY*0&cMI$b6j*+D_)o~2EX(!M{h9C|2?u8J9ryFDw|DM=D$#!CdpcrC|uc@u=sAIkZIlk=&xejPQ2nc&SuC;9V7 ztGVW{Qdm0Gl|CB}qd&{;!TM2&@Z+sDewkt_ESfIoSouwk+#FSesOeR}|LG4t3eT`> z{0lTGiKVAID`}i&7+Bn&!R{+J@t&Tez^Er1%2V`()fewlx~vU+|2P2tO%cZRQF$oL}ubPd}b*x43=l4H+LiTJJf+cbW5s#-fo7w>-G4n(`zLT zH_LHN%`jG$x{FKyC#Cb3e&FwvCN4wUgT1J`N&{;JQVVwyUhhr7x7}hVQ1l+Vc#MZt z5n={!UWBl?`z+2_B7>{iUt-w)m>1NV=R5>1iqrd%unfHtxy>8XR z;>vUI@yY|R7v6)T>21Ex2;|G#N8<7{GxkPwc|4v%9(2xd35M(>n!Grf zp4cEN<1%J??<9)}(Qtg)-GF9Cmf*3LKx+Fp4Zk^$z)=}(aO`J3sSLS;c+!_9Px>JV zd(*-#(+Pt>qaCz#@oeV8OK5({P3}5+RtKE8Pkuk<$&KJQ6}(#->K z$Av3+n=e6$)_n{*QwiOlEBS-}U1Sq9KcdrORXV!oAI{#>i)HCN<5pEQa~%h6fa|+b z$(c3txWs9R(6D|G=_Z$YHd*rWhaxJ*$78mgGq`sQfx&x%;r)0&@<~XhnKPD9=!K2+Y4$<7{%sYl zI=CKw|N4RF#*PzyjNL-RP1Z=XDZ{6FGb_d#mXLo5uzZ4yH9bZ&KlU zk#)H+l2-QdrfJNT8fP5=@AgV=(%v+@WS&5ux71UpIVXtpAeL@_6?z`@Wyd{-VAS(+ z9Q(xyFKpTj{hLGhDz)LzJu3>Y_Kbt=bLZd#KjeJd_Bqge7s-%wWt7QzLj5ucFG9j7 zl)a{D4YMfj^>&IMl86yY{OFlyH4Z60RHYLg!e+~T=>CQc2<0Uqm7 zMNXVIZ7$?8OkcpPO;4~iQ=Xcjk>~IGvSMv}=G#`tmRy=C^^v}(M5Q~JJU^bkZr5O? zHw!5$(3}rC8xIrePH_bTYxon^{o!&_3p_RdN>II&78zDbXU}WlzK$xb)_Hw{Gv6{s zcy>I8e_@xz8Z1jN*5m^>@5Wi)VoV9Y=j1YIaj=C!3!ZZ0{wQOU`f0vzM}ZVel(EOQ zlRu*_`ZM1yV(}BUF~2H#wnFUj={@=fT+vL2e2=Ys-_9Ox3JT~ zUz+1F!A6ZQUl50LP^U@i;2xn=rCd^_yW6DjsZ1pSUP*9wItEV|U2XiXmeX(?Nb{X+py{(5&&BN&86{?5`8k^}_aBJ!S9Ce(D7Aq7 z*;F{2Si!&U{eWH+Ug0;}j>gI{W6AEjHQ4?)nBVZ@5?7?&$zSfslQax|C3V!&q}DA9 z_(AcF88T?C~;jHnAJhAID+8(rijdG@yg8f?>737FUv!!siVc$xrR& z$dzxp%D+kfDXm^H5kJ&y;GQO&MwiNdXkR%6R2}?Tb!WV(sWEnc< zYE4G9NjUOQ2A}0~p3~T?O80(!g5xu^n2M^KR6RqDUAQ}irI$9q(D+MWwKkT0pFNxu z&HCfa)&a0=_BP3iRi1dj)(S5hXtJe^mmy(72<^DNk2CW)%Y`uUrY9=%Xl zy4@FAwQcBP_A|6xB~O>F|6uEUZ8r18CSD=r7SUFof_MMeOr#2JCSogFus5-m=mBiVVdf{)bWP3d4UvG^T`eFQn7Jp3XU5j_yW4MtMqS^Ro8l38< zxA=B)w`6=p3>3IVLqO7KST(wm=hj&>arXv8&I!_KNm9(NoyE@%83DoJ3Z(O;a-nV@?tnl^vfK;1$B z+|~XM^7=c_^k*{gkN2SGF6*RYzxHsa+>Pkd+-Wpa&5Sf2o`5B1`(brmAuiil&S$-M z!g(JW@$63}QWgKBA9l#`)vE0%j@DV_;Y0ALaSnQ|H$v-Y3B2|7zmkvLeaJVao35+= zrkbQC^r%gQdO0ba{&SZ+-R@wlzAq;GwUD9vwpfZT>Ty;R=)8IcU0d1 z`ZNk$%eWHwDe@nlt;?b$=}}baR-pk7joe1PDcqqYhPY|?0x_e12}ab^(2TREpl3)q zXy-qdJa*d#qxnX#6glsc{gh!%(k}EoGnhO*!r?@%J{#Jq&x8pBsB_b6QsgHJiW|Cl zt2i5Y<2n@@5;#t7N;Im^RDzz48psjZQpYt1kgQf8rubUL@m9AYy}<{(+Ri)NvDb#M zH~A^qZ|R^d3mehFsGRG63i-zud`Pypz;78+i}s!kC}&rLu=x>YAx zE^u?(P&x!{eQisP<yGkEL2v6nT&@bu+@3Nr$Peu!REV%oRRajbcq#p1{X^ z3-(ZC?JF%=kH?{iz6?183j2aUY&t^N5P6oh!5kEw8liSae>Q6DT$Zis&Kp*z!Bmx_ zOt-b1GClXftI6N!#mi7g2@RH3-dF)|e`qoDRsSUa+&7_x@g=Tu`Xpgj%wDE%a*p!F zIp~X0C7f9LAGiFNywKMF0Iu@w4Yy|;^_IsXuZdT z+^;VX&TST-d;9v3aZ?LtF3#Ym=^ciLeP^=+lOICO6-RMh^PJ1@PUq;dl)YS*!3V-r zn51dR>T=ydx9lYhJJ!#!K1j)NTIXiwcwVFE`ZE=MC#UJ^%i#UGe9S4qYy$EYZj78T} zA1-h47~ya~(QkRY4+eC_Vo<6Y%#(WyzaFKaB@Cxi>-wtAzxCCqtyW^mu zEXdxMi=G?4VW#?PyvT{!`m7Ap8Bm5_Q!{ag&Tl$(MT>&+^@Mk)DlzrMOxBX(j5hx^ z;>zRl@Y?Sb{+`I;iyevZa%K`+p|c3z?^1U(sZHR#msE1*fy%6P!!6WqccMGLizKtA ztHGZQD`8Mplw_((EA0O`3hPGQgU_=JK|M->(W}nk&6;JrNsBN2*l>{_HsCMjmZd>) zgeS8u*^k~6x8T=Man2S=kl-!L?scY1Pfa3>k!{4IvFbF_K!>g!A4d}1S~9Nug&oda z=yM{SUv5-D)_Kc_Wt#~XB5y;w{9JZW<(Tw%$3;BnlEeE<5kL2XU8wM%K-NC)1bu6y zItwo1m#F>x+Spd=iC;&mL%VpTZ=bPAna z15nlQ)S+NIp{f0Kn9j@aLDe?+IQ=4g{I~_wB6hR!z8+*VLPGhw-hpb^BxcZE2#HR) zlqdTS660n_a@hb}GqxE1t*GHY9&m%UuCuuR>Pdd6m~WZArG%fovK*_<-sBJ73xQ*W zov5_t|S(LIurQKqx%FHmLQ zA%`fE9lqgO_vy+x)x~hR8&*hw3w!QQi>d=)H*cB`R_YEb5`$uOCf% zyPu4oSPH*)7~!UZ1zcNOCF@*05W=Sp0S_4^+6;qm-h}TIxvqlVeNW&If4s(fzL>_x z-52?tvO$pYt{Yt2ZOM1(N$8X2%zn%}f=hL!d}H8H(lc2CeV->lN>E=iSFy(JMRH)5 za1z!FX4PQl3Sya+RV+Bc?XAl}hs$EWTx2c%eUZdF$JC(4w0=;peFRS)j^Z6sda?Q5 zx50SD8eFe=jOw#g*cAf{HgND~d{c3V=TEBOZo?ohT~?3vmOI4!*4DA7xvOyUOiz}n zJOL+e`ok4H9!7nY$~k>mYuMqM3};?-(pR%3cwp5_erc#IH`3rN_PzZce=nYgE(ZqF zpy5dv+_;%OCC#RY?Gfm?wh#}zRAA8`M$^^HMmWACfPef-7u<&4#_+Ndv`%)R(GwPP z{rYc0SHDaAj{c|lvPrwd>o*ctzda0VluChmx9~6I7x1|zZM^=kSnxe+O}o1%(BS7z zc%-yTVir6E-JSPSj^NFIyz&w|#vbLgE)0VyP6?=Hr%ZazQFy@n5bytd4{Y9Z9(7?E zE)xG^KYpZfs(UT)iTQM1^5HJne-DQhrFGmd-G{^*4q^TFGM3||L8QkqQ)gEi807+Y zHu*sG#cB{o9sC!4Ls*kNnY%l%9+sQ-WuF&sV1oh;k+~F0x4yl|-S}@cr#Edq`+Il* z(~p_QN6Zj7s*1p2cA{%2N1k<_O@dv`%W#cv+cNb9x~zQeA^81Nn+sLSM##`3iNp|A{!M|ID@!4(F0y*l zy9Kasr~zGBa~8_}=FpVrlYGpTK$vo}0^%~?!H>UsT>HUfEGzKDvx_s?txtm-i`W40 z&VR;#@HB^e8fcPA8XiNQzW+xFyfycZ$E zYo4`$7l*G(FKak(DtA@6(wZLrXxSM2fCssN&@3)%OA$Kle1#jdv+>rL%iOU08FZlG zE;eO90*mk-tWLqF7tO3mr=Ebbc}Lng`JXMGS7`y z`ZE=m*be1dUQa~T3wK0jge|^~-HQR%GcaRN0>A6}1&qIx$#wNofR2g_{1?mT=vLK< zo30*`e%`7^bsJxx#peE^+h+u{+damO6&bAV>2j9ov57^^8;QM#dyD+oWw1Y6iQXk1 z#E58R7IWN_&1jcphuiMMuVT?xxAGkMNL)C7e|hqVw+0z2Mb>Jui+QH>a4Vj#!^fLv zf~HHb&MA9c9$a5=DPk<6=q55M@Iji9?3xNHCYlaAP;N&70oq1vMw49jlu zOIvf`LC9B5VX{4be&^gJDKzGkz$KM9P?f zK1(J*&zzHztoTiB5uD(Y-+<>oDK9z651vMdKG`RaAOh z1fyz#;je`;zx3HSVPkq0i<{6yV?`d+Kx(A0=|ulh>{!m?UTjI@R+{kPF5ciSf}OaA zXw7NICr?d5T70%T@TVTrx0nd?Wkmm5q%yDVl8x&VTX0oy9!<78O}&HS`PJ{$xvqQl zlGs-&6!7M}v>Z!D6y%2t)6+@4!3z z2H&b*hz&c=fXT#E_&c~q8njgvosy?g*lT~b)5n7>=T9N|jVH+RXE+TC)THa(*cs>6T@#T0YD@Q5}Zcev&FK?w0N{utTj=b?`Pb zlZ9=yMs2TUICVP0IOEZzc6=K(CKpIGf^XnQ<4H_uw+?US?ZHX+#L&Lo`M9z91y-Lq z#dXJh#-$zec&p|jXg?E6hu0=h@cxk)Cf|n{O&G!&ZtJjh$K;^=MLRV~ycp=&@&2(! z{Goll&_~CW4n)L&XZtSL5pI1F|=v)I5nFF@FS znSVG|5uR?20Q3E4#B%zy=h_v_U@i810o=XMc)?HBhSqO^$BD!#%YC%6 zqM2?;;?O@~AzQarnRTlVV(P*RJn+|(x@=0pQ_~UoGkWAvDM!vnCyV>|K8(=jT&U(pSg4ZK_p)_x}!|+-k zw$rbHx1F0#M~^sDuOEvjedJGeSBE)v?22HS|I}%1(>(GQ&$Rzcn1G4x2hsIWF23^> z=Y{bD@Q=fI%-Xn$8&e)l=RYUop@PF`?$F5s4-X-LMGaqrwzoN16G z8TR8b=&n2l-8$*8+fwu=>=?$ZA7p{1>l#|nlfc_d)u*1gKJ4~qYPJyrw#vNP1N@(0R1P$rW_ZGg`&;GMtW80cdxy*nz6&z$yD+A?St zKVNl##BSLqzS=32oR@?F)regEwV&a_>k+Ic#sCY>Kcs^$XDR*Nar}_76*d*rfXeYT6_$c@%zgUrpH^lNh_!z@`S2K=`9Y z4mW-)3cOVTHBHRK>`Q0i#^=j$CEzqAmgVEpO~)ut&H=U;>T(pB(w*-&gN9!rnf-_kA5G0a@CKkYpiN;_tKC3O{V z(%ZD0^86Ipa?@!TUZ=tDt~rVdIU;9fks5aOl=iq2aS-6tFf2YsF01 zKp6vg(5ttg6WWF|gEKIrp$u0~I3%@F3WoMeb=(V?^)&QO0~FSr1#AzOI;@k>+sR^H zJTQow{hfro)8Dx#p^NF&F2n&--b>51^vPg!81M2kn3X-shL!(KLQ{2yNnh+>d&WO0 zKZp2DavWjUI0~@2z&fkPias7I_Q&%B=6ti{^c9I1NL3ob-E3LH$GSgf}Z zvz?d2g^axg&;HY4J1Q4*!ykS}v%Q`mZyrWwT88w^awwaV;vsr!m&3jH*%05cfm3d6 z<@$ZD!~v#fDW~NMr*QuhJTC4n+_P2|!atY@cEMXD7xqm;o3fd>@5DXWG1it?MK`@{ zox!SC4q%CD%3S;8Ebx0{27SF(vRUzCxda1kK63gB)NPo>2b9H9uTdVj^>7_7S**=A zH>NSK83)LA_IWn5tD0F=_GYaKEnHGb9-f%Bk}5r)b5*rhLFV03xcE_l`r3>ZhD?-$ z15GvD^H)P@e}N73Ng2d83O$m32I7dUq=ul3m#UT&+$Mch%UG7Xga zb`58}CZDI-EkRK5Km;#N+DUJXvS{G>W{_U(W_EK2Gp|eInb&^@(50{opZDJktHXwp zvHenB;c^M@Dw7ROOItbG9?5#e90X7Q67Fcd*kpX0uXdxzy3{e_sz`^b8n5&HW0 zq0Oh|obiKpd|fh>%$$bMsFIF&9^GSWZftbixn;EDz?bok z-ESJj*~JGsuav<*%Dar)?4L^PbyvZ(H7?*$KL*piH_~9mDXh$J7Vjo8$JZIX9aRTh zVE?8c;jU4| zlP!(>vK>3L_lS&{DR8hO2``P<4L9_6VV2Q!>hVgZb%v_K54+=7_WB7%6-F|%kQY?E zW)#`=h^*3Wg|K&y1AO+c|fr$g3I3Jg5eb{p(XMaY%snIi_iOrpI;5v%rS=fv&5K{of>y~tSTrS za)EAN751Ogd^W~3f$ep9K_(wsX+ZpC&T`I0&q(|bLH_0hJa+2Kp6?8+-?*#}c>jNFdTyA;{e z!s{^lBB61TJv~q|0H2IFbblU?3S%Q^yU}kfZYTh&?@A)W&=>7<4@1XrMOL|0-mgY+*OVjKWrt zL1uJkFZ482NfcYMuwrQn%-=PL_0;BbYMtBp$U~!9m1-t*PHTkrul>2>-n-zu^K>|t zmWn}Lb0A{POozn&xA|^ob(p8s0m*rGbbr7ATKY^Gug$o?DfBgWbS;+`B90&D9T&;L z^;Oy+@w8??S8d?>Ee(b1KM&B&F~@ny;7u6SSV|vmC)3j|Msgj4C~#0D9~9<^exkE{ zhSCk3zBLhouZ}|9Fx&rZ(lM=eo<+2MLy z`0H*n?t(@mCJanj$9|?Ju$0ZW zAi%_iUPsDuc1}IK;+sq8{vZhz25x5s-)^z$37yLzawE3jihmLE^PS1({R$+n!F1=`eNmaM^L`|m*i5J&LqNQSNEhv+~~Fm6X9s*Fj1yiGh$tE2p}FR`-2aqPtdd zz;%?cEUg!*K75F~POYI+n~{@imBNQzYQpHbCtr2;o8SG9Ki+i|1 z5=EL=t<5*TJIH#h?{dMX4MD*(4qt}6`tA#Aa0F2uYh%=~{q1>vzWrwQL6nfhEUah}UqcoHB_1 z?J%I{j<47U*eQ0Cl39LgAGZ6G7mO?EHem^&*Q!q=UF*t>@ zUdf)R`>~*&T;@LEDy>YMDZ~!`O3rKRL1XAPkqtHp3v;dbE8ELx#Ls%>Hq$~FAGC

iPTGf{)~Z^kSOyY8HL$6HZ)JLy3x-8`T=xMh={r2np`vc-NylvpC&` zs$|R}ml7HJb!04jmg!-vM6!cGAZFi6MV}tW;j8A@yT*_iEY7COCED@N z(MQy2NSCYqSHZhP9m)Op7c_hzo9c`eX1(epI587Nw*6%=X2($!3A~8gdvxd}YL5_OP2|^P)05#Q82;xTP8nPa3zvlB zlf>s_`r2F6)VBtEJG7y9atubSN~i3GKh#6!7+C__AUsllfB3yX$<@o`>4OSlnHVkb zP;8}I19_x)NiDh3vKjS?zo49>8Jz4IqAQ&E=h6074CL=wpNJMuZf0& zY3`8oWGmSauElJ%(BvY_zEipQeIR5tj^PO4;Ij(o6jGt(W6zU{$Ez?ohG)+#>Z8dP zBe;s}V(zSICHZGRlSR{YF=&Gt+Zt^`ihZ)+;@)yb=0*>ZGW&x2KVHT`>l)H%Z~-Dj zrgBF$EIDcW6U6-A0NEwAn_N})gr7PIYz99I{$D5)jsNL^cGV<;&&AHKwPY*`WgK@sN?Ryr?5DkK=g%e10{S#)`R)_}J zSyHRN>NHf$f^i6s#6s#w9|)ghtoaVrc%@L<{=0$DgxT6DN+j7>Lo<_fEG8X;ELtf z6=3e1H`FXBocU;ML|Tn^;(g8SXQPF8UG0-`ULfp$F^*i(u3nF{Uf!BBj-?26II-I#gR z+(Jy(#zBiq0XZdQM(vL`;Q@Vf&Uw##Txnd)_?Cv!!u3(K@wpzRZ}zEvton6wayAXb1b2yNDi$oo@Cx=+$6uIye31-1ytwYLFlTcI74(S z+0^IVSl*rinFo+GS3K&MGg(r1*HuynJu+juqTI~ zEBVC3-CuP?C&~nMX0(wC2@D1X1S9QjmIbfEq<3P3ydlL%9Q}FTJ11 zoZL);Q>3A3=qHx$}Z*pS3Bdg8drVuq;j#0|o^o>gQonyLBf|;WESG zNEbAp*2&hb^MVI+Eg)6l4^d3YK^cDmwC_Fw2mk8e-anIpVpQ3<)CZr zNy5BN;GOZaaKNvMh=$yNg0v^Fq9_}18rP@crMabi58V(Pcvew`SU4(9Jk2V3JMmsaKGQwR1ri5* z!RYS=I@7NfV&5b~;=)*J!JnVMbiGIAh6vbuYzvb;dJ#MRWMNs-d{j6+hWqgL0It@M z;rA1d@uy}E8UKKiFmGMK$mHXoUPa;CMh7?-GzT=-Y{E}-4e;pMa=I->jZ0ZR3tJVY z;KJ4Gnbt8k;L^EN*b+UT>ut5=Ru{i!Yrc7+g2-a{GeL+<=}KUi4&A1;;RW>Kyc z!`BRaD%%56YnqvYZYM1EGRMv3ZKU?L6DtUCCgX#R@W$0zx(=T6oG3Ar?NP=ipBnP~ zKq9_#n}wxT`HW%ARXU|ul9rO0yytBRr`Vp@cyj75qku;FXO zu+?c5&HC-dtP@*|KQpRuk5M1EFKH5J&&;P^8W{|#tKT|^&gh?0jvk4g9@A9lvu?~Rl1 z6{E}K7BY4@zvo>bjmKUt}3OcR)S>EzK? z0UohBPdsi#P}7ZxP%M6vZuK#NrJ>rORW}8t92Vit`zMG(ei8X&C=QF>y22;r|HwAe zsZiwqpJ2mkX>RO=RCp!W&%YKwXc-8Ct>Ky^Yg8Bfj;$oMdy4R+6^mX8&Zsx>CE2?M zxw#Q{Q03lzsMO>bDgp$lA73r6hlLxh_t7M{hC{w2u(?ji0&bfs44~_I=d3CI*%fG4N1~cfH%r;4bZQ z01%6cxATgwu?fPY(8I^YJL!RxF0O+9`N_VG|j5dk_XZY>56`72IL*05q#* zx!JlqxqhLiG@f@fSAGlPdP5Iz56|S#mMsfe5zTv;`spaWFI~?1-xEfOxQnd7)Q~gu z3go1=#-QGbi2q>G5D6~ zw0ZJ#p=xqJcP95P@EV?7(?hgHeqrNTZO(DgKXzczHPRjFKuXrmCbuR|fk+V-s%gbL z-fRuI*26O~$95*Qb}A&npR(EZ?WMd=(-}WCbnsAQTMT$8M&DPKfqZ1D-H>W7a{fBp zwHcG>@*nkR;{K2;2);p&|0jai!~KOlBB1rtnpCjE2?m}n)k2< zqrqAiOeFvky%XTqxH~Y{wu0HyB#9Eg7GmbKg2r8ag}8c~9e43e70%xb{GMVIGu|Cx z{;u|e=NWUU9lu*qs+NLdtq#;j)Cv!G*>a_258rd7D~po_D)SP_ z<-T#GL$wAft{Jm`p2To@E0@qe)sndT(il8$P(vplF^0#p7#MZ6Bkz~hQD{Fwt?y<+ z`LhLhz;6m?bmTG_s_bX4v?i1I^9wl(-KA8AXONwBEWp_jd?xSg0I9mCN_9FXL09E& zs7|{~CqAshkKPNoC%!Ip!U|F5hyDRrCq6{0=I(;)KTeV2q|c2tJxNqQ^Euf&K@)A) zkHN;<^I4ZsKQ`jdSw?v`&%zLjf?@;0nLSUz^_$Po2U6PDujCF3Iy^{cg$%}A2;|nU zj>M3(O#c1vCn;T**e?gS5xph-M6*^N4(s#2-?pFRb0F`db@w28`6-Sw6K-iHzDd_@vGr*1`Q@pD+U`Xlq#%>X?Qy{B6@@s1f|SE>`C z&RjIRMbd1Zkcr<;Q-u#-@K?q~TtD9$BLY@p%EAzu+>wO>CWckGa|R3UT64F*W(t<@ zZjiPZNvy;vcp&K<{@FEwDl^f{Eq!nF=Q)*AqxpW`Rqm3LsReAc-NxK{?ntP}6o?-m z$&_r6#w1BaDC*N-gnpf(yBx~k_^dZ1CejTQy4QiXLoD;EZ-mTTqDT7ebh(w~kFc%r zOXDY=mk=Ui2Lo@1$(dXoX7a&YT)lP=&&KM)61!Z?xRT3BE+MQSW*KBW_oQ)xZ^ZWU zTbfg-!F}CP%O1a)PD+kzH+FO`CL@vl@Cl2^g_0H{alTXLGE7y#h0TilBaQ7=YsJ#4z zIDSlkAJPd-`ntuyE%8L<)jZkS#r7G=@I#iES#JdNZ1z4DB4Qy<>8 zUxNGW%P9<&k-K8I$j!>>;HG~9mT63fA*VI)WmJNjBCk(QRSUrJw|V0o%p9DEPI zOmc+!8@62T=DoCAvySzi%lj3Rj(}XWD)VT^7M#f*K}p9yRA|RqbZog!PMsVw*WX}bvA7(5*!#eb4gc`gjl1~4)tN-SPJ%h^Nm!DbhO>6|;?2ejv|{faVszy@ zJv@yc^IZJRsA?K>b_-HSia4t>Tu1i$*v z=8pfc0_lw=f@`wZNyU?HTqLv#7adtnBJ=>1W7)>ECrEE}|HQcL31Bn+Ho5w`jP}Mi z;MxIA&O=H8A6Op2tkTK+nd=m1{-~Wk@DgWd=qoahU7t`>k81SzX+mRDZE#TTJRy@y zNbS)EycUv=ez#+&)@@7btN561v9TrEDT(;${S>2V)ME7pm(jH2a(vmbJY08X{k(pU`DwURexzD+xrJ^a4h)~*^+Q0kz zAAGL!ob!IaUNI~`@hjcWlM>DvSdfX8NfrIKt70(*{glSGsS?V1{^q=fbry`%0 z{pb1(Ear*bp39~9WYBBi)AGPwMGNrP0YQOOpNe1~eu~3q!&z(dd-~{`P;vGE}o*p5bJ= zX;drxYVPFz3ko8GOG?7`@a4E@k{X5fE5W%{7x+`f2dQp+6w`TW&ZaO`dQzlEyY7yq zzz`>zcX&QGR_77l`dk5)jdvD&HG0K8;zh{w%!J{R?_hlH0Fx>*#kJZlxG^#UjMm$; zhudE;y{cKXabgT_G2axj+^18V-+mmJZ$k6GpT|61@vR&-PH6Z%Q3z2tqut3Xana~@ zcorj&MF~O|J3)^6CdA(hY4@gsc>SjjZMpc7B1BI& zg-w>2A6miKw;O!-pT4;Ez7O8}Qvgl&1uX7&Hrw}9LVrUguqMe0pVdav^jGPi;+ufF z{S!d;Sswp5)s5|ON@3MeDdcxN4paXt0oSfNwEq2?dpf`t*VS)kQ(xHN`m#8F+0sjN zEI^$?AJpNV2bVEuY6<&u^FD-GirK(p6NNpkQmCc99wIiq;Oz3!_)l4{VZljH2v#d$ z-(D|;)Qwxk8<86Kx$^@38F7zgtsH^}4yADl${Xyn&RL+w_LF#a&}2;JFGIgR+bI5& z60NkfrEw0!nBn#ws2Asj;-V$cH0C5Nl~gj9pM}U?6|vEoV<2LOF7-GXVXew}Ry839 zPTy^3$|F7SlglZHs7yq!a7B#B9W43*uH(HnKb*UH8!T8^CGzZYD9Cq-$agMd^N-14 zOm!Y7d8;7UKaOGp%wt%>snzWG*Iwpi*3D)V-r!8Nt;k(Iims|2ps=OIpfg4tWIV?+ z`>vG?9NFNBgs;3eTq3v%2E7!b}`X-iw!F_lIF9K0%hc85Pr`BeD0kIX1p4WJ%3G=~|#V zE)$vEjv2AsZ1L9kZmuPzOeml?PwnaAc@4q^)nMTp4F+*X`JD1GxUJ78u6fXD9QslP zA551O?8ZK!e*YfOSx$q3J@tjnqb#ZYZ(rfm(SE|($ zM&Q_A;IO~PgNzQ&p^4!);P$&_ym;S_@;B~9|CwX3Wb|A(>!ih$jvXiaO_@w~u@|J? zAI2tJHx%Aw%YohPtMqFZ2gUcNvDkAWb35o`#hD{k^uxnO%&<(Ng}JVLxp*}DnOO^3 zQLkC_21d>5A8F9BC*=86?CZ{r6}E>6bWOVwoE~pLU;B9&`RzRSDdaqNOXTt&-h)_t z2q^ZRxe$SRWCTqWIkvNfg(rGA?JfUcP1OZXJues(u3cd|EooH!UX#w8yTNkL&!RCE zBBNg@gy=g5aYe#6dVFP&Q2IEF>$_@QdFE3?d^YqpPKZ0o`YhSTTd5KoVl^1$-YlSr z&Zl5cbs}wC7z;0CWQ6b?@_4hLfMh%83J0PO!Sv@DaPoB|cVAN-kK74h@ttad@+8p# z>8l6RXQ`2mm@An!Qw5zL$}y$3=M{Cf2bd6ntXG;#jn!GIVlJ3KLn-jonT`c~Yk%k)s=Hi`)eV|n9Hb3Lkw~GEH1Gy<8 zY{kYx^yFsF=L=7*6j{kiG-pK&b+77#WanInR9(+6mFa_*MP|~Zrh1UOkipjYUZAKE z=IH7(&!HEhF<^@&vi6}=GXEXe+Fa%wj;RQGHJj>W))J>qj~_2`SFZeJ zq4}qU2WK8p``o=)c3>%{dao266f2o-j@Ye`9OtKxUPq>e$`~s-ji>J$2v=tr2*cW> z(dCQC({m`HRfk`|;lL;iRxG2_tLKo4(klD|GayDw8RuLVy%XO~vYwUV9_YQu4v+HW zr>TyFpEnBF>#q>AG~eg75Kv(E39`$L;5-~ z&pu`W7@NOf|*%|Av-e^qcuRw{eh?oG5}5JcLvGw0&|Y*X

Otc=1dmqudjOhK;pe^Df_ca>IFUG(e0whPXMY9p_GKaLQmZ-XS}-Js;wi+K4 zc%sB-G&*;$R7r%Xo zEhBd;QuR`!zcy0Xyfcg@3_L>%J_q8CO)_|hn=DK|XHPdn267{w9l}2jwe-XDCyjpa z0LR&!=U*!i5OQ(`(=I6kfonTXr`EJkeAYp7Pl&@@ot0SAzdv^=FoNuoMx#fgDt(V$ z4NpFq@U@{!C^qOhztXD+cd2|~4aMD{vLTheD?~u@&kdB`FdfWxM#82UjbOe|WFvmp zWxBgy$ySE^eNe34D;}I z#6!2&2=1yO5NmEs#nZ;)mG1>u^)DaeD$+^j+z!$aS!N>?2hhTPYGQxW7Y-TtLBpq1 zZb8*%klJS{#Lwvi`I1OJJ#PqZS3JNV>LPqRI}dH8-ii5zne?{4i)s9-1RtRumSh2X z9koKawJXtRekbeP_6mbzE}_%8+n8HF8@F7`!9(4ZY*;`m+jK>V7I+V4+8J-4>{L6` zFifKOrTbXm_(tS&-@+)bUN)dm0e9u^M&qDH=4esJWPbGmZmEX;uQH*qZiJZaS%%4&ZAMgD@JU5!{!$pv%J(doyI?^_A0#<~mp z!k*CYE%WGSp(Uw>`>_{GUyxZ*e_^#`06h~k-5>gu)78VwNts!-J_9U*tJ zKK(RlV|jYlm~4^{o^suXkrq<4ZJ8>5O&QMXyf_CFvNK^_VGHw1S;&8$)yOYay(D=w zKw$P(eMs8nFI)I9f-X;pkR+Gs3(4EX@5hQ^c<);r*AqDmf82ivl`9YNQd|quE8Psg zd~fhFp^jwW>xufC2IEw20PM<3C8_1!_S35+^w#f&WWm4`F>h0dCf@7tU`-&ZuiHRH z&luAy-cIhhu2`Vj4^`g8u8!g zUxj=-ez9x&Ias;z2=w2nL_KrdN#53ryV*B}Hgy+pGZ$KtvYG-qi`{c!oHPx8_ky3< z5e{#KF`)K%G=656(QJozTQ ze37Y_&D?wk(AmqE!FQDdtQkw>4qMvpLAp+Z*Ic9%J(6m2iR3{ADz+D z5A9pW;K>kuZd9ln4Y?ghPkohf%*Y4~`KQZ2i@XN2KTXH)PeuQU!YA~W?Tnm|l4)B>5 zLtwh=a`gL9%BM8SV4hnX8?4_-ZRR=q?{Md#U#jZb02`3EfTz8-#Zbimw8 zfplqf5-iN0i-wvKjJ6az?c%$~K5)E{zS@lPeJpW#{04Fwd76B+8Ys59jd}0Z5VYov zXQ6gqSasJ0I&g3jSEHkfXH_hz>%$}#mubW1hZ|8`#8*iYZYi&H^@4z1=fPg&teX3p zK+B2I=sHqYP`Dot?DS!3xt7JeK044>?=sSq+$VQuOKfOfhxHR=F+RD3dVMshyheli zt2yEA^~%(b)KS-OGRv3y%<>FN=+vtw=9v=A&C86&_CTw|e~B63V~*r%%^TXtPh&x%?!jI*R<;ACo`WVn*5zI?3)bhxZwDFv**qFI+@nJ7Or< zvXwhC=b!k!5txCcK%bi&ss5}J8zXWp`wcc=U*DPt*7H}0eA##$(xe5B0|!&tuWQuN zy_H6K9D;n^P*@&*n;os3PoXV&P{%!I+V_Ha88L%lpleEB=G>y6B5PsRd^emcl>sVw z>(M(-4=1iMq6*VX@b_#L7$)n|$no`T>F+@_JY0eOE4Tv+N;x32CJ@~Aci}RR9Bh++ zNm*YESiB9aRRn(NisIOExSMX~DaFR@z!BwuAg7rl`(Rz-7f%h8#x=0{%azIWS}e>-5FP4=vuXS0Ja+GW zJ0)g6qO2G0^ttE}$SrS%_=3ZHt#=5w=_s=JKO1Ps%p+hXRmQJbpGBPfLTVlx#??&d zq{&&aly%z*S1&n01C@*DM@Im@TYm-iesQUYJh6qw4GH5f4oQcANd@Tj{yU_L|0}Dm zs3C0MiJyzw%G);-L&uL7w0hejx*o3uKH|ORLf$usez=E~t@}aR))90x?=z%g&$iF>bp9hs zO70PvOP_g-5P4LXvmE*#ea8Hk+fn%FDHQv13~pK{Pi3DRnfS;Br+w-~L*GM1(hQg~ zMVzx+PScOxA7rZki}4oCWO-RuXt*x!RYgzDj;B&WlGG(snqf&>#wwFc>NBP`M!tOG z`FLUewVV8uhx3I|ag&7MI-x96M}j)$Pr>?x4SkI5j~Ay})9><|?Cv{bS{;)>7N;%g z*^F86mv7+K?p`L`yIzd7H*;93#bR0}yM|gWrU=XbOP18tALX4_bwTTciGoodQ)mk8 z1e;PNOyBU6{mPSwj`(@-;OG>_ahfER-$*?k4yd$Z3_DkSiH*K}jc{l?H>PC2{XZyxTAb`-*67f?ct8{H8ze~WJqg6mmJ*&{_OTG^;7oE`p) z{##ZpW=n_REQ@fEiO!&9*H6Q-P-#pv??+d5PoRwU)2wY*wxIj86b>$V&c7_0gF8=) z8Sw$ZSjmgCho?0xv_8$wRcsY=bf)Otz7GbDn!(-Pw+LEy4)|Q^|E%Id@EC zF2$X(#Jxk7ig#f>p+)!yw;x*K??Y7(ls-}TE_43>IU#Uv&S!7F8)46B7oM#$#TM8r-3CbVynXMYKlXq@wlSHkv zrppEkNOhz)JsYS%S1-R{dq0}6L$3y+cU2<{`}&H~_FQ1vF|OG8%2K%KvsievOI^6E z9mJ3M`;@vZUa-->gSkbHde}Fp5{ye^@#m#0)RcLP)^~iMh1Ex)=-6?1eL>)DBy;(q zj6%@aQGs6q|8dICjxw8)3SN29oqZ-ww-MkBQvD+pOx zI_&VG2W*a;G)V&0A-zr?cRw&^3LW9Jt1Ad5X3fIJABB*#uN7)~`qIX8DRf}JJXEJg zagp=#SevXGSGN8b7^Jqe6)P zFBsWY2E7}nf=R0nvL&1O*`=u{WpN!F507WId*|Ugb!FOCC`CW=tbawS^8D3M4Lc5w&%pa;RJPPes@y>sOpo#&)oDy$Ze@;&*P1!)F zJI=19#^e3+vv8-@7)~Bi7MA`RMajB((6{P7)&44C2_oyWS*{mytrKxUiyBTI5XpzS zsbXH^2P$}4LMP0>CB#BI+5eVq$jk&ls_}TH|-t&`~4KQAl!tNU7rK1j&bl? zGLnVlh0+bv1mVbzeAaEKN%f1A$k!_$7Nx!b`Px2oWy(Hy*AWJ8bH{Q2-LrzBvcafw z`v`UKXlJum+6z^C|3i~Pk(l%G|@%7%yg@UyekJ zkb%^*vCAu~!oVlE!l4x{(z%FQvFZ zz%2b&FvVNUYKeEGyMINN;rvIOMeh+d?M)$V)$U>^+fOZchC;oST1v@0jzLGZ^_1rYM&PQT=Mzl*lgT=52rn20UP<^;_vx#B_KWv;wbA z0+cV1zEV#UrZNSz~5MlC+w(iuJSCO$w@&BAy6uLL)1z>;`9DZe zT*Hp+*@k1=Vu8u0lHBrS`Z27V3-~6gKWt~?X=Q7PT>t*0qBTgEcdoAxXDvth?~^2} zm3E+?JF&y5m`1lYu!m)jSpM&eaChi^ zimCRY=)lp;a7j753B1TEs*|WV!V(8M&B4*KC!lJygntwt#)g;pu-~6m=+c-L);2H{ zu1rgY`|hHzXvGq?Zb1?~OZmt_TN+cr(~`u}(-r2Do&4n7ATDB~1=}e*6elQapkYoX zT^k^a^Va9G{;G=bYRq+Lx~By@K5oD>o3w-lcg!nZDuvK`{}8H7(sy|Ba2785u#rps zw2f@G$kO({eX;CzHM`xGNADI(3kyCkz_Uw=xT*mvcq%XihX%%QRp-md{P{$x3?2X$ z(wcZnbt6t65)UQ!HlWADR&3Qa!sfd%=w0-<;?3bY=>IF7+84D!$og&g>5~j)zAmI7 z`8t{+&RE|Y^!ORUBL(SU_AJSB4BQzp3XgpYw!b&>F&jNL1NtWP6Lcoq;Z?nhv`={v z_G(tKaV;tWd)tq0AG!-4CM2O>$v^S$OPkhG9TfK*So|Di_}0=y zo2zAT1nr??9>=J1iYk3c&qD8sUPAD0f8K2mFPS~*KTb}*n|o+u&upq+GtC9!?xeE~ zq+NbPnaI{_4~@Y?Ijx{7a%vaGB!OAw36|OCA=`3u6W`aIW+(Lbd!w-{}YMqqD@KSn*spp#B(xWR79LV^W?Gx9ywlx-nL?Yhp)%T~<8kIP4;%;Wg`pDdP6G-O^S)|`}o zIXtP^LXR`kF|#p+r2VEdzixq_XVpzR--U|~{e05MX}e>8IcW6a>~?sd_39etCC`=0&Mo^uHPtw@X8vbUYHSQEnu*R&17?Y-n?{>66ne#8)V=|bo z*J-iIcOBWq-Yb%WIq57?$p~9I#aps=B)r{{0_R@L#oF<9bj2>4{fcSGF|G4E?r$X$2uPue31k7(ayDv|Hm zg6JzS{ai6QZF@=;o4t9tN!H9tN9+bPA3=Tb+x*|yIRswgXw8SG?Bz!@P8c>D_dA+m zYe71#xOSV|m#l@VD;jtew}_qsQ@A0c19M*%;y%w7{!)>Imzp<^mG{@Bg%ia0xMhh9tit?lz6i@S$3xKrNBrn(?%*)(IGGsgvom%E!t*bF^!c+Eru=$H z+827^x3vq@9eFOi%I_2o1^l5eiF3tWX*89mYO+aT!E8zGId<}*6~>6O=YP?o=wOo! z%FnE&;QNCqZ&_&9SZfCpaGz+?)uTop~CH7;?cUHC4k;b`pG0pEl zGlDJ2<6I9e_|Zi_HZBuV;xF=D7mZO`UeiIYPb{BY;UOrDloGCsoznq7#|n0bpHPXI zTY2y-(gRM;1fcD$kA;aM~+O`1Bb2cet7 zcov(x3j55K5=ye)(>Kq1)VaBnvc(-hR7L`~M(Q}3i+8!)4YpJ@K0x?&X$K6)FI?^3 zB?8MCfqR`@(L^x@%lDsw;d{ezijK4Wvb9FSxIzBHSD`s)2fQ!(1b$}XUTFVN{8;#af8iY>GD_EBWN=x9^rB#P zFt`vRv!meO#qol^emdF9t`%>p#UxZs5jJgYqH@J?!rzm#gfFQk!kDXXIKKc*ieEbf zPkIbPlg%@@;>L4mb)|(8cSO)Q$N{IdeBowIG}Sc2KW5ob@g|tfbG${a|3YzC=3bnC zPW@>5nw%I(j6|$8`52cp^T6bNxkBvo44oyl@82E=&>Kd#%uA>H!B-&tsSP#$;&8X`geOiB_bSdWtvjw-eI_^b}P!*Y{&Yp zNG{>vAaqw6D|#*bC7;AzjbnTm@A1AGB($6XoYS>0_Oge^BPcfPl$fX43vTNtbG_E5;Zo2(vh(p|HD|>8t;T6)xHpQs z=6?s8hW$sGHi~du(SX0voGacI{rE;@X>vLc$R)V6z|yI`{P*vrux#6D(%2(1)YPPK z#+h7vy68RR*T|D!M?DkegN&PK!18My!FkFh&aB}cNqwD7%P?AUXwyPaDNlz@%1;9*&O%@KqFO<5t0VQ9+yz(Wh4JB| zr-H`a!*uP?e&%eJS}r@^p1Rk*<$IOQ;peo0_+Z9sR`q8w30`V~NmC1KKPUF@ulI%X zf8UdPR4|2|7n#~UhJvp|nX07BVePWnc`GNaxoPo{?f2fk2rZcndQ1#w%?EBwX*lu2n%xxg%{+-A!-A#vE9=Ys* zqzy*hzE0-d1MnZ^fWh4JbWJspR&=GZ)iXzf|1!kp6~44-mp`{u!A0~|RaOi+Wkti3 zl<4i^MZzRZ67V`fWb0{{9VyHg2T>*Nj+Q!c~-BEbfli zokfGQ32fD|r_6QEJdA6ogSYQrb2g^+xS>xG(;B5m_n%3GL@5_REA9Tci05 z*>tYaV~9|kqbvMHOF`D6M0hu52^9A^4oe{s6V8>;okI$gGAK*(^7G9K(-SI!#wc%^ z*{UYESqu|qoK&JsqjR|D=ZnR+NE`RQv;iVoXHj-u2wR^~$z%`zhOfr=xhu2h;JwZM z{7JJlRJqxdEw&%YmRyf!)2jBPp%kb zU85-eHaD=_4?>u)lm)^#hl;IJ7gKKRV`yHp4Yg`ocsUyl;ru;KeBYG-Lz5es?SaV{ zmAn})38un;!y6#yk}LPZsvTV7$IzB#m#8r)hFeto7z7QGX%=u4N^A6~!lVfz(sJRG z%mcK4Wem&z#Iw3D33woW0hPN(aH`hPZx2i>rtehPaZ5=AdYeP$cTfXe>mLG{-DbY0AO zPLR&yE+tm8yM>EePlkuZYA@^%;KhoSjOZY724rlMoVDGfw z!GB*|V76%#e7z|8tvxh^_-(ZnpAviE!H@n>C9;tgn~$bVohR6~x6S-hO~TMQ)i@<= zJ>R!$6+3Fy5BF`?W#OS|eEq#%2)3;u`ynCxFmp4i&l-qfWk*TV@wUeIHTH|2Ss z%j-d9NEDa3@+Me?xQjE7qwsaLj8LUJpLbU_XC8@jVdJO{xTU7Y&dyWEr=ZuXa7KZ>FvBO9g;?C;~By5vPy%3O3g-G_aA=|kVAU&FOk$?$w*1?MqB zT{xfdnsWLmzzvK2!eGNJL1~W?HU3E_rOg#&eJTx{0I6xyYtC2_%y0WHdc{Iz&}sAn zbboq|HrP+2evY$n*|s=V5qyRMHW}cD#v*bw8^t6+QC!r_LF9J*C%Es@0h6+3X3#N? zev6%1kKy87{9X%u|26_w`&bD5E^MJquT#-B-;Vqn#IEW;(Qy^z3rpPOgm-&>kx5K8 z4L8)l;EX6LycS2hBL`th=5TWLeg(}-#kpksK)n0x0o!@D2B$SGKx4^!GIiQcXI4z- zh5f2ROTY8fEre44b*->|*aeLLw-_U&#m;h~G#)aOP+aE$X0=34xG1YFO!BDZ?;D4s za?J!^ci6?#zlLJGfyslyY;UHCg~Jlzj|<^GhmU^)CK=#Qommsy_b9QdWb6Z(8p5Hbs%Fz6L?Ia>z_yT|1T ziM`iYxxuLl-6!G=ak)2Ue~QAeTaGN!#F*w>9S*i~E8%CKZZ>wN4|ZlASn%1FJ%P-jKJBIV0c${5TKO{+5 zIt%I&JuyZ-2xcvR3&X-*(5L(N=uYin+AqG<78%yk+KPiXShok9(vM=m2oNUVc)zUj5_jxle(vaYRl@mj;z^&Tx>0^v3gH6XS{^n(jCw; z`6^!hPnzzh$~ef&PQ&{z^{`~65Bd3aR|NdG6ICBxK|8VgXJp%lzMCxtEj35l>k48A zuZ4X{mI!n1SBe~jHXM`EOsn!;@cDmJ;I>l}XT)bSc3TzK_4`57ednU5`xlP??>PQR z(dSjf{odjPBT|1eo;;#?_^N*n>Q>EUO5szWxXy^QR7_ZUpjE~3ZLevKSFMTf1DBgHYe*K_ zIcLxYyEuN&PXpGfY$&X2tRwI57g_k)cDD226S$}{6Y5uu5d4e{*$IUdJTm$WlkYsn z4m8fUm&#Q@{gIRL!*L$|#GYaAO$Oix4QrTa8?heyI}uvwq1|Ccg#g!ZWnlR$6F2rG#?Z2t8xo&{ktF z+Lo|NF#A1On09LqbI;#Tl8|uX_w%eJRgF?cWI^ViaFCyF%?#E(q2&Q%gwB_KFn-uX z;VO%w6`>_8b+f4u;hIf#Su5aI6+&FF9ov@kn5>RfQN6n}T>P+s&JVkc2RBZ|PwVDV zk>)Pkp^_t_W>fIx?8i9ik(KZy7lc{S#w^#bl{J*yh6JN6Z2!=+tlC{un7C{(#;zd(kAGnmKBqIsHXl;awU&NFy6-FXcDc$Lgv&A!XeOxCBQ ze>-7u?@0Wcs7M`w3EWOqb?nxaK^v=`7%F!F5AW6!KB&*5S%U}BczH!mXQLX_hU-wZ zqC1%mRutw02V;rzWlrm1Dhybzhc8$8!|#uZboYHN$qrk>HW|kfH@1ZJst%_$P2))Q z&|2oV-i4hNO7VH>6|}K_#PvEGVPcg4Z+kw_<%{X;oJR}H*p@((((DMV6om;}@~Fse zX~n1*SZVkYYYUna)Gnl8wsjzy=;EodA>ol1=fn$ z;jP;qvV(h5U_sLiPH)*1G4m>gKFf;Oh7&eSE=m?1KN#YOlva?>8H24N2hjdV7yEG1 zo2p`!skp&c7$jzFRCTlgGtW`}p`}o^BTRUG&;yM3*$G`n{Me zm|lq%p5Slla8VRih|cAa&5MN3ef7w-z8>QA>UgzRIxyP+3|oc5zei za_v2myt+ZV#4KB%jlo>|m=VITrxGD-X*jx`@xyaRU5I=98ji$k@UOZFE^l+(lJ)#J5p%*SU{@+JNW;c*W2&V(iaAn*u&x* z1Bu>kVs>+vuwU)-NF!69CLLc*=3ft!MY@=c?S279YHR6J#}ydn`W^axy2BOq%w>r^ z4d`qA3*?geko>Ceu;+&bc7HXY+wTYQ4ShvC;q56RCukX_J6FN;-UPbjwULUor!Slt8pWoi^xs$!|LG((B`4hpco=y>cD+55kR+${6<%F*(iSQ#Q3<_PgV!_7> zj9FdBw9Dnuzi}y_)3J>=)vV@Tmw)6RN}Y!!>m!(7vL3%(a>LBK_jyafkR~e+!;v4x zk?Y24dV92q`H0>Qde{IjlI1}!*d7WEW9f?1RK6hH6Q=fm$)ZzpX+io3p{(;NteIUz zTi%}+PNrwl-Qs`Dph1yVu8d&8Ay?tAMFV%Pj~qtCI^)aIXPm-%MRHpB0X#Py#8(Dd zLS~RRt+F>q!v|edeItmA8)OR~d=^1T*Jh?aQ-%dDPk?iReBBBm(d;7nEQ}!G;1|3SxDIDdFu^-(EivNbTQ)>7 zhWQS84?zn%=>0Y~Hp%xOb17IQ4BZ`0nx*~dP|a!jGtq?KcRy?Zg0 z4xb68?MJV(cC#kF>Az;yb6<@%<6GXeWfHa+cT^lX%QG)iIefl7QZTwamb$u>P(jS? zEu0a~N)o;C>Z8GQ&Pwc|zKVgB=M$NpIfCNBX!dk?5WmUr04KNc3=8Iu@o=>gI$l4a z_w$PB!FZ8T*7E>X!$tV8P^TR>f!aAIGZeGAQ@SJ4zThoK$?ISn2WC{LGUs zoOz=yoT*z2>y*k!uBn<$*2zFMJxlzwI0N!!pYhIT{aMrEELQnI^n|~)fEibeVevsn z{A-#?C4CC0D?N^yD~q@zZ~e&bUmJ^(s$_2ag80c7mebn4V>#nzU2K6;BGpVsRzJjm zy;pd`&SxZH%%gI4MBSFY{9S-$5kpA1ZYlaqh!*0^ZG^XfHH1Na7Win`Wq$wuI{NZ- z8m4`LfXgm(3xZer=QfaE_FLZD)5UQp>Wxu_`skm$c zC8;)ZuL5+L*G5$qdEXmH87Pp8$8qwvEw;O_l27x?q=ZSk!!Ub)0A~Nt5ZV{k!i}YN zLe4WS8gXGW1Y!wQi(J5#DNXF1+eCPN@Hi{J<<318=jr`h`wL6+cY$4zEkq8zOLAM^ zz-{d>v@7EpNyq&K9seA@zyBvT$K#UZMEWl3`8r?d8D38@mxHO#1a;AgxRL*F$wW45 zToC78QNwp^n8VhMOlLy|Y@yVB8p42LaW8bEkR_&7V&D1mv8cazua>rj`HiBFdD4^e zcjs2Xldp?tvh)_bYW4(9j87w_Z;R;lzh4rysxZ*meF$dFRzgLWRC+kp1~+vDapuwH3ZbW4w5`q_oxwJshO zw;rM6J`*saz?scWY$Vt5Z4k7;ib5vn2vxVU7_HHx8;&Wg`(znRJ+m3|wS%BeC7M;< zn20a$7Q()S$KY{o5URF= znxpLaRt<<>;zD*uN79m^A*^89b#^VR0dAhx#1A-yS!KV4B{S-1{0h-!z2PhRwI(9( zK7?i*P^0Hv&)J`Yc{FyME8Bu1N9aX3b{EZLJ*%Xkc*i{&^D_;+FZ6}4i6Rd?V<+Sv z09^Q56>U#8GIMQ5EX6dK-RA&19(0Vgi%##uw^b>rccxHbR!lN8N06!Y5L7C72%}mv zX?a8m`}ispC%TUjio?D^LzoiwaT|vfN&^JCr6nxCn-7voz0gE0rEqhh4A5N5g}IEZVx8ohnlop6<~U_T4w6i5jyj&Q2_4vp*+Gj_t7r zPnXve8Yy;)73HX(*zX%QvIZh{dI>i|4e-qBEAVaaS{m=rAFoVNrp2ji=+6W#di+`B z`hM6#o<@2YaWMrRTfLGD@OOtrE;s1E={tmbYSthdum%mTDNy<7w=h06+J4bTOSI{Q32UmQB1$t3MCJlc$dIL$CdYy_a*K;B^#5m6y^w(-neA zd7PlQZz9SpOJrUBzS7H57xq{-fb03>1Rq?c(!TXO;A}w-o7irK=00uQkCE?~`p8+h zA|zEZDotH@#cu?MsRJGEUfzq^PNQgbc0FY{bmLy%VZ62PWcV;;40PRTz{i7(P_HAO z^0%Fk%e+)dwu7mEIQrOq>FdB~g!omz*h~@IH)p7wtI^l=eUw)?63Sro&t7I)#&facK2Caba z@LTIV#2jeiuMD|Pb>;_|?pI5cF&>Co2F)~Wk)jYY^)YL19LYt6m2-JtwxZ2^O(FY_ zfso+PRq?}c5_A~+fxyElti^9341Q~d($=Bew#_ND%qE69+_zT5EOo@0U*H&6= zc@3TS##2wnC)zMI1eD^xvr&nCXktq%=?(E6=fl zH@+aNz!>sP>4IbX7jiSAi?OorWXNb0xlB=WsAzo$yZ&wj)jcS~wPrHt9FT^iGq$qa zl52wThk_fdba`hMId$UtO}q#Ecj`CcK68_2=nm zrV@Gj4yME7egkLH1;tP6$;CB`9tD-dqA+i)pW=kB`F?n$N(p-?Vfg3{e z*urTyAbiwTT=e@bllGzt*@{^>+IT5u)mP%lfG$3;{}>@sb~SiKPsTfgcHq^MauD)A ziq1O_%kK^2Rz_vh&_YH;Map{4jZztrq^VuB6e&a{g{*`kR5n=^T9W77r=*AymGvcQ zDP^R+`#rzE-#_2?d7pFc`?@}#FnxHq(;Fpvmyv0Wu9kYg$55jPQQH4&GhCOP262AU z3l*R;%vb@%LG3^dPVmOXY?0?|6zG=9!6Xd z{7`NY{PoJei5Xx>LnyysTChsw4?pXHogH-pb458NA%{k=xMygA0ySU6x^ET3on?1Lk*# zw(M4D7JW-Af3%|dF`+l(_JT%^?8GEDT^8&+fo*qdBIis!skNRKbhs}@4}tM%;NLBH zt-kScEoUGvj3dF)MRe+e3uHX4rddt0xb}h;(Q*{Qw^PT%g6c0AI7^R=i>cynK3hpH zXJ+wfUOX&WU&}vC{Y46!>IJ{I5o}&;OnFg7yx-YR`^%bXiA*HUs*Pnkl~U1G%>pp- zIJdK33QU_v;+pDEE=X=K7KjdKB|~b=ed%!aMWO&{*A_O@;~C%4t47rprxTTw8KB-O z_&h2LZGH)?Gz)WLw(>o_f9NrhH`+vBZOf!D-=D|aal_&3vq*lS zfgqh$H7*qY_gjn5>ZtW_=!uKXZ5UYJYPq zJ$4o2?bN;CtZ@vtxs78Qx`(hqOBvllQZfsf-_z z!jc?r?vGiZFk&OOU+}uQhm3^Gq|3yqB#B)4Jq0F7oyMZEaaa{qPCqX7M!V!&=-;k{ zYsM+E+ke6#?@%hfRsDr?9dsd^|I4*5SVw+5kA^I*NV;!U2*j&*gWOdQpt<9?>?I?a zhlg_m1^@;3U;dWL^qY$bC9PTKmvQ4q8>pBh96J=(32be4TA%F8v<6{$&_ z-h@zY1^H(wR}zW8D@XUDjlN+-Os#TL$RZ-~kDlX!ewKT#YT4nyM>vgp98 zn9w$zP-QEWYqrDgX@fMUE0w$5ya*sy5_V~CCUt%Ju(0b48GUydmIW1ad#;M0@AMCt zpEQPze6}2~Tzlaa^a^wsDq;#MVc2kG0@E$F=u9iNt!$)2wjQvEis&k_%+VsV+w|spF;CZ3Q#`um@{+vLw=}@q1z5wQOpf<09&kIV`Ui^1T6v1+J}?a4uPa z%=Ii;t6ECT=LX{E)-ky4Z4&Ozp2Zux#u4*lo^WOQbo#-}9inGH1pBm4ba8)#;2(7Z zt;yNEsBs9bQV~4lo9n>n{!IGh`7>&rXpKH!W|7h}BCK!s8j$Y`A{*aF(9H3du`WBA z!gW(-A4~0>76JHt*qw!C89~1;EMbcod!qK^2n&EnPg&i1f3H)cCP!})e%>x52gjqDP6HV?_9ROeo5jxfw$W|r`IMG9V}HM( za%edQCugRUN2OBGynZfmk)4QP`%h!Ds-0!@$_9L0AoNi5ku00;gq`#m-)ojec8;<^ z@4`N6@=xF+v_0jXWMva8TOSzk7ThcOVR%flhhLQYk$5-S2n#q_XfcyO^{oa_-rj^C zhPiU51lDz$W&}RE`jwaT61uz_M6ecr(-96am~EI%e;+8~yblTt*oES3h0ZF>)melu zH=dyu{=PIU@fmUV6=z?sm}0n+FG)#1N)~u9I^uF1d0H#a+CE(*%e7{}l*@+H<^D2i z?oh^ES)NVO?0?dsu$$!YZDyvwL=3xc_a|OWg+lq+9eAK zlg+qc)C<@xC7fF%Zjt~ge@-s3!CWz5Dm8gW>8ZGxB<-IK)$U?AK1hNb$`7R~eLhTn z!4sU*C_%ffX3>*p{*iC41XI;Lah!oIF0HpfSIKZLTttS38w%dU*6%ob&QGE+CyL&* z8wFy{h2(|nOpv!Vphiz(Vcl$F_%NM=ErX@hENec7YU#kmg@9tqV?k~H7rHaz8@_DX z#U;Ewhf3!u^^g95PrWW+@z>|1c33@?NNf{2b>B%{HOJaED!}}YepFRzBdIZQW?Gsl z=p;LUpLf5cYuz6b@#RzLsH^GJ{ox)xuDSg`zg%FN!fntJu7TYlVeic=Id1lb-%*1pvX zXG*7`3#WlwRtkM07b}_|kwTk{=h4}@UHD|tiISo%vXBj^&+ABH9wo&8n+07pGCv(4fFVMKf8%ss5IDn%91^E zmeJ<5KXKoUWQ^J?YFYK)Y4XqiIGq+Vkvr&Yh$m~?L8UzpvpEU)=9xp5g-l?@#*c|t zsvNI$bc-;^3?R1?N>MG%4<1fh1d1BdX;bJ#tjnDNJz8sswUoZFt92xGh0&N;aFL2u zO@Og>vv^;Z1LUPcKPeD0R%6qQ_>TH@q<7dDTCim*UO$x2S5~o}bNRf!4_4<0 z?H1;>_fJ#1E2k>6z8k}q(OOXTTa!edt0f-y<`IuWNXlIaxA~O?jE^vd;jU)ncUBM$ z%rB;+f8QtXoI1&`inUN@B!~4WshoH5Qp@#o6!82W1MuH`80((;a-Mf5z^d&F@Xix0 z@^!5t+uQm8J0>KMU5!8J5cCQjl^N9e(+sd`l7&EHbqor)4~E)r(JkAJ=^i->R+cHm zP$HXO#Wth-`%;`f_c;DMS|sr2@8dPe9&Uy@@-3Ht@*|F`pyA2moYiavE=HS`(x1XYvezn*dZF%s-vM zbuArKF2!w`(vYtv#+vnH+4tEzXo!^H>uZ^$u?FbrHc@soktaG;8RYms53(a{IQV{D z#HGBGXS*Y9(e6kl_hEcInyr0H-z@%tR$Dr`@*qi&o+in*Ik%9W@L}|8pB0;XW54(07onoB z9)95ZxSGl%@JjJCp51RlW+us!JN}WpjB5v#@0SyJ)oHY&#tzKo-03jWv$#}MmHGPb z!Hm}*@x{M5dj5+iojCO?+Ai>c?5Fd|;d#OA(>sF8HfTZE`ze?=7rO;n12EziE63I_fp&-VH7~PIQU#nK!G=C7QhaOn=*hvE?o5@4kN1|7} z95X0VgsQR-7=G+C=arLBYG?*&viZtIOd3Id)$D=HmqF-#w^+#iCU7%TAE5roF_sg5 z#S!rYIkfAa$3p9B@rUtVe06#uvH2QGlpl*Rv#Z_$oHh>Y#8=|Y(8D+_*ckRn=u=Od zx%AG)`}~u23TR<%jpY@2#7jGjI!>C2vmWf?zfbBRxkHNu=BpUtrm3);fCPBbxxO}oBdrboX}%HI_j)pbrx^s(S?ln{2RLzCIC@}rjUSPH(`wo^$R zPx1!k;Bj;on`YEOo?SNN?j?IeftM00WGmhX+;f49{QHwUE%7A{Z@*)*$^#nwAclx~ zWz&ao7hu+gRWRAh02&3aM~8F~94os+3g_ow)9ON^tv?UP%v&kUGRNY!foHTu;Mn`F zNu%a!f=BhpS@O>F9cepvgKXB&Vx#_(WbbbIv8A_U@sh2u|86Z3T!N|)vAR|8&D$ z6{jR{>q5>l><;Ocd_vq;I6}j=Y_dMC3d0KWN#EdmlDlgX^WzHx(}3E|wEGl|^cy$aB%u^$$gsNkOD^JG?DFlU!r%{xL7F8DK?hSGF4{&EC8 zqIiog_uGRZHPz&?r33rETptHCaxo-3lt-=%OH4$GQvG{;tn;2TZrF=|Jd<%%j4?SY zu;V7Sh!MSRp;J_Nf+QbRu~cxgAfuD7kxkdF>C#T2b3VsVm}#EpHqD@%(qo`|Lep{0 z$~tUZ=>%_^jp6tY5B_p_I2sA(g+o$fh>`I5xs4C#j5;rjxYWzpM2rHRE1OY_kEYYD zL+Dsn2^i?ygo^?lEG14Q!>40rFvC|ACQlsB`RV?`*oSlQvbhLXvG@y-j9AH?KPLxK zTRm{?k8pnVic4Jhn@nuYGNGzsi%ChsY*Mpl69hPvz{U+@5#FA_hdYWeLg-t_t-fCw z@i~H~P1nO2(W8V6vjO~)`9&7%xq?OT1orhr8~L|>1}vEMlguBlNCS4sqs_+EbcEUv z->xjrEiZWmMW2t5V2$Sj|0Y6UHFl8^UP5lTvz{cKdQW@|C0U4dBb8aR5}quY1qU1> zaGf-ADJI=?iOxoNmE;3?P7ctwxt3FIe@S{YzYyL(h4j3*#%nxyMn;U|K(9=dJ{i9P zw%5GlD;Au?qwj)f_tFF`QSze1It%u@MWLeWU8>oY2UY|AIAc>h(=F>nOL<|Bl#>c) z!e0}c=pD3LDju=31S4fmpxB2E+?@S_Pqe`Y8#C{NuhQT=kRkW=Nk5ceB1kq=xsVpBmUjx>JH_Q-HYCk(F$6a7^%qow;kaOUW>CZ z2PM{2CC&cRuv3=xo2IhA{nv=G=y>oHiNUUo8hET>0?CMx!%#X5oFyl5 z^9tsXWD8mFP3yOv7hgjtVj|;>V+TfOFRru`h4ZgZ>85^&c!3W$C zL(u(5?V>N@r_d%CecA*%EE>7CF%=*&c^|b|841!<0Xxq(lja*^$z(i9X11Qg1;5S- zd$o<&y4Rev@Vhwm&rj(Er7J{b#5I1$$Sa_9LIHGUTY}b8eX{D98}EG8n5{lKk-kV; z#GKo&KPo-bh%q#}~i@>3_h4?5p76;@QEN6JEr$_-Ia z_4Ja(H9CbQ9jV0SSM+G>^(LBf4h8? ztr46FZ87A{0;CRZlAv(kjF4*=iLTBI5-xL6V8Gm`+3NA+_qISjWQim_oOYT{`$*_i zXE{0}J(Quqn@kchf}Pe{tg3T5c{2AHPJ4D1Cg%P}VuxPx5pPVu)jLICkTqJC95moc zPi3=o8A%4Q@S2^t-Fuc`)`q|oMLi&=4_Z3=+e|h zjm^K)34e^qiGwT9yCsO0jgz3$Y& z{Ubd!bu6q1{maRe=HR)kOdxME$j#K-w*gvj1M5#9`?t>6wh3T)p08$n0AU zg$6m?$JI4(Nn0GBcfTjz0}512*r89JmB-8INW$InVHlR~j0PX#ET6PJr7a_>_-oU` zuu9FF={7v2Bi6nqCyU~^U8mydjfYC`_J_Jv?p{0U>+=<#yXWJtcf-N>2~V!g97DrC zMS=fN9a$KBiLm3bV0|bGj~zGUH`K?H7oSrFUz!%r+x((WJ3~M^uOGu+PX=dOZ?Y)n zIesi0%jRT8W7^MV^5mZ#e>0Ylef};`byDyqt{;ODdi!8YO9^SZ)JT%*+{lvK>o61y z(dm5+*S(_^-J2eA^(C?(9wUWQ1E&3&XSfvkiQE?$ zsbd^nq3ixdxV%r;C5eV`sZj+a=1eB3n=J5=O|7Vm!F)`8cpUW|)!3i*U@$P12l1BC znBwsU8?}s4G);{ty`F{J62t^v+HU?q`XS8T=FTtI45n%M&JZ2a#6OW0L5XMyyg8$k zZgm=gnpqjJ`i;7$z`x$W`8;B?a=vOt&{54RH77_V1zH?Byde8w^oz0U4N zz}(sNl=o?nza$0PW-8>UR5>15?1Ka6HQ0dPX*TwPHZyxXjb5tGAX_{%(J#oA8cli+ zPqB`KXrzN|a%-h>;4ETos7ARYF>D!o1vFo05TdI^3x|8+9n)Pn`tCS9yh3;uFWR7c zKqr4&BAs{JUP`t#j|6eO8T4%H5j;Cz1Uwy!$?WsHh~!=gGE+vKU*}>bxW*1K?EqTPmYt5DVI1aQ-3gY8X#f0#!&3p4mT=p66-r>EzLiS zgPIwmV8Ig`Y##B4noqe%VZakhedHj!Ya>s5646Uma9MSQ6LI-p7Q??K;F0~uP+NBz z+z}gzQTkJ{qvst-oo7cEjE}=SKO6RGZXhct6c~}e+9>aU;d(mLGGWtEvAund1po<^O zAcI=5U^B;@_{~qjjblBTGj{a$f&#;Ave+HYqEKwO=cMTWCWan&ff+j-`Zmw_!hf)JemLGz{Amg^AIP zY;_`sI$8txEi#D})*M9hu2^bkU$h4zvopg&orN_{%901ICIlu_^X2*Rhj-GOd;C z{=NtIhWWBkwSDw>O(xd#JA9n<|VL zthED&qW{dyXN`mLv+TIkJ&sJjcNqKRZv(Rry~5Y0GpP0K&E&n>Y5H&dNP0O&g^d4w z%5t>fY|cC)4NV>9V*Cq9^n22UbGKzc`_geN`0OgSCi$wR<{VXQ5Eag8`4Z^(@;kaI zo`4$7I6Af^0gE=TV7qU7!b0u2xR7Wvc%DMTv4}W1Eg{`Q!RQk+ir;hMC2`S?q;~mf z~fVR-9B%xRRi0~sxl^`a>PiuTd$3-ua*$&7H7^tE`?l>3g;?~T`bqH%)_ixHt_Y^ zKYGt#ISG7a#rb=RVgH6aa!SzH$j}8pEjJW zw2?+D155IufF~N~ZjqVJ*)%GDfQ&2`Ty6OvFgEX!?4ORfMd0tX)~u)1)wWDWMj9Ws z>a)d3KjNI%;d&A^3q0vnaC^SLx~i0qn`&7x5b3Pdlyj2!g$#s{X#OkVXqsXe5`+GaKLWzkkRdZ8*oSOSaJK8eLzm&dCS8#l!l^w0`|bS{E;di^s>oBHw6Ya2lMkkV3&p_L$5$Hy5hu=@N@s`6;vO2>6YQmR8pT03p%^iula`II7wghom zn}8Yn6Y2P9q#IRlA)k*`tN-Gj4b*)cokk4akFw)hZuXaeHb&T3?eK1E(kM#jcE3EEcV~dpleP}fUX!3 zVk_fn`RVvrIQ?CbbxVX3op4`H>rW9*==a1zmvA_{<`v&`S{kA@2(zf6aQwO~mQ2X# zBImNU6RCE6IJ2_^Mwb>4!^Usq=LkKzeBCZA|9P88-Wtu{*L`lfrt%qX|U&W6xA(WpU zOYDX3wLWYqMy;O?dIyrZO_x4!+Mmwh>_j_yLAj9svRoG9{;5FrZDUyFgS4I3hAqQe zsC4NtxcyO^e%)_Mje@tardmVxz4r;;|C&#n?g#MIPO}8}<7sj*vw=H%OP?4&97gG! zwXA=z7guaNKrZfmf`==WFlpE?&V9u`H1b_Z6hDO_oOsF=oQWb-ZyUJn2%*{98K@e0 zfSa8YLkCAW;D0H1@PNR7Tyjts@{&wYXHO!)rSN5pWGUZ0;BrDt__*kwj-wy%W zs-U-P5m}WMiaOzDXnH0PZM(9m@!!KVT}=tRt|(Ha8wsSwYdt)gd50Zjt;q^D3ev5<$iX zhS21%Gx_YNRT#JRBgP*b5PGun;p+Lv^!{H{VQ#-0?%qBjFbCeik4p-fRBPCl_0;(r}PneeUfUzP%z=6<6K zyShng=|T9|u7;-$$df+@?qQ7EFf!eFDtWTYjozEH0(Ok+CVjtsICgJ47I$}JZPZen z;CB_1-D>b{lnCrst0I#h??+XK&8T5Kg)S{P0-^H@Nmq6jozcjV-qH$U_UNpohy4gx zw`xABgy!QN?~$m{cN7#8(@?o@9U2>Jvil!T6a9gFQl5O4`c2dZW0@pQrNn@Vreva= zmCVzp?`$k=IsX)Fx9q_n z9XNdbGU_qZOl>Z7)77_~Kz&Y~xx&ns#K^0T?AMV&IX_FtKJ?gHOmxw>#ud z>`|inj*uO(CFmsik&m9@j3UD9MEPeo7v*P+))R$0{k1+?u|x^1TzMGC*oF0nd};nZ zdn;R$Z-O6v9Lap^Kz9n+`d|+W^75w=GyasxJO&%N`jvh#?OF&4Jgo`VM=Vg0HsGzt z^Ksc5J!lSZCIf?a@pa5$s=D7Bvo(GA4VO>iML#{%vRFp%7{^1;nQL6&pdOfDo#R%oEam7`}oIq)vzo<$a4i|&{)w-BKmV1z5Bg`+WShf7d}V$&{Z3;#_|IO zhfw-(Owojz^`*|C)->L$an}3t` zhfZ+9WEV7C$%Lk|Y!LsQfbTSo@w%)rJMnrlioA?PQ@fiexmjG;RsSZccXojljpP-& zE)%kBE1Z%QAv2S^xHb2@Vdqf;_;2eBAyb=58hcpfjOw@4!SN@zDj|zz+>&PUaw3rU zwH(6iZt%PHBv`aCOH381;8zuz;>w5_Y`fF~HhHcVYYj>S{~A9yzh?@qq{ksVtA-o1 zpbrn%KBt`$YP49a0*t;7N3!-6oz!YTJ|)UQz~c;TjeB$hp2gAL2}u}yHHEYdp9gAf zv#Ckt63px0%awbS@W~A>BuqU34llllE+LvI;&qGGTvKJ;4@8*O+B0NPg)#}aRfsMJ zv&gFmC8jJ_ipIBv%+?`A{)^=cVwft&&SaR86&r$R?IUIOJ$Vcsd2s|(CmkVX?RwN( zKb$L78w=~TI*4ds9tnMs%^wggpurO&_^X*xklGSVr%HaTJh#dRd?J66-O9BvbJIai z#bG-~l_iNzS1wrPJtdocyYQ~J9Hx)HWO-Bf8TnSd1{DlS>5Y}6snR($ba8CJi<5G2 z|Jjq^w0tousk)0c(aos7@&!&@FOS;PYM#*QXm~G-uU+>XH z&G^M^N&GGt=2<}Is`qnVZd0&%Y(C#&EDz5zEa02(DjN3lkEPOBTl7nDM|m?*h+OB+ z?(!Pcpwo)@C1j&x(@N$pcbGj;jJr{M=`dcEIE3XduHliCQa*T%GA!>>gu~x5Ebn`j z-C%beQT0v)XEU+P;?q@GTGBoh6(^d&)vQDu)m|&?P~Ver{=)9NcP+-)2cXAoG1j;) z6W0tTV$Yki*l~3%j(?qxlO5Mn9t$j=##(}SXaQ%xvz-Jre&BB%nU8U6wqd$fFlW2n zi*l>);5!|CcG|)MmQ=n$t%xC&p?Ke?2{lh)lAjh6vzJV5t z?~ob&TX3nbB!urQ#*)}R@_yti{@$PzW*A3v4&L2d<@1NItIZsdzd(k0kiz7nIPbLuZf~CtyJr7GX{L)&yXTOLxii7gH5^BEm=T%J_4Lfz+c;`a z$RSogBgy$EQD5Bwq?Vm0C%>Mf9!j!AvP0mJ>xT&YZx4FlPzpIRRg8`KF3#@0GNwCZ zUr_74#e#Fo0ix^gack#EzzYi{UQM=@c;?%|zl5{A=p7w+#E(U%>&I}+oS8!J+Y~KU z&LOa854r|!AvPwl+_9Ki%bP2g(ydkki)KnIhNiwj)5k-Vjhd%;^PQK8`|lXgojeU} zFNM?L<`223UHhR_Au6e(M zWHpb0Qz|oH`wbf`msrMxND)Tu1P6J&ieQ~phWq5%WEclUWPL#w&uUaQx6PRJPiGe7aSh+!=iE z-**1i0Yw}PYa-(nY+=;tII>5s0K6BQ;$Okv2uH_al)yF1k?5t31z9-STFCF%awMqY zEmtlx5@(A$6JOagxHzrBGIsY=dVBnO%)D5QO=4=?OQ&Z<|B(p)Nag@xZy!Uk9jdP;cEa4;g*DOR$IcNa zt5|Nw$tP4|c@c{n{e@Wl_J_cG+XUCUJ6E|AVVmL?Z0oW@E5$8P^xsS>ZEgW=esM5& z)?C~IUfk=xX|QB(De7b&A#A}#lzBFVIX-%UZ94hLDV`@QB{Jw&=Qs4;tz6EJd%#Oe zZn3oe9>LwQwuFnPj>DAQnVhrGvwmD=#Qrj&yBk38o4{Z#*|UalAG^tE`&ODh*#*^q z30Y_9Oj04Y2!7u>hI2(ispLt3!QVy1XpA|n2v)?yoHIUmF$FiHR-!3o3u2BlAlES! zliN~1KdIY65a;1AFK+wKu_$sbrHT$6*8#ptFnrlAO{k3P% zXmuf~nWo_DAHwryaSi-CuYmbvLwZFdn)BBhiD5bNI9l!m{`X@$3;rc|5t2@zihMJ_ zc{#wp(~IDxdI6RRJmwv1%c%dgSSs~oB`u78NnZ-x41d9Sw{A-^RQ&?7r@tSMXNj}# zx=V1a=t!#~eg;$t4tJ6FF>qnyEP*R$Xf?2;81FhR6vv%i&g?$!#3VUyu*tK>Nt;jbcIVnj7iWQ=w~Jc#-k*hX)oLX5xi23& zo+BapM%>oCGqhe)2Tj}kgxUNs5|E|)# z6K%Y4n6w{%0Q>9*NYqDTh!WMrD+|)7ScomR{qt(lViL`r7(54)`fpK>_ldMf`x!0v zOW`^jk%k3C;#lo_5cqipR_WB=`rjz7rePO3yUd4#4XdHSOXT6|tpd!o zUQGiJ%!WlmUh2+TVWuiS9j!OwcQ&BFA>PSf39*Bbv!nRK}>8 zjdXGJdHV9kYto>Quo~AO(0q z3bzyqyNa6YoMdr2gpKx~_jDBCmPaPBSC|G0JLi%mb#sU!yN2sW*`vAIQ~F~;Fxwhz z!hUbKLpF}crQ!pJI5Y1;jH;I;Cj;M6*Cbbr%=*OTLK+=q8H`ihpVHaenyFIw643Hj zV8dOkNLW@VnXPt>JhfM%W@@H5RCgK6jNOTC<$m%br;BbSjhJ@5mcQmSAL0yq`Dp7+ zq$MYkaKjA2vCje0eCLsZN!nzj_%Sj^V2XL!jAn`tbh-VDa(Ttb+jyed6N>*IQchoj z%8@_t_0J*8t3Ex{QSd?9kMYKu-0!F>zYy(D_M@_d5;m>xC)S_Z=-v}fMCxWXRlF>V z?RzTmO$o!@uhd|e_gJV5h$npp(<|p`&LU@*PKOl%PH1}4o=JyhvK4vhB&Bl>*`V3Z z1>}fBi$ou}KT3@4ezcICeHw$45>#Qf*EYDE`i>r0`I6jlDn!^51Y7%)y6Cpl~+Rb0mx(Uv)v5+ZyiX^ZvR5r#K zP8IqR@%jm%5@rXYudk4-#o2VTawgFcvOe)`<5<7dBrN`sOZs$*=uR>WcaF>e%jpFC zymjE}yBv~za1Hx>@hv)^)x=#o;biCN2!7|E1oHDh1t_0LU}-_c{N#}$u%>o0G1yYd zwH_Rc>nk{nUnPxGymBkE2IJ{jmB}#JJ_)w8_oCU~wWNGQ53w!HAYZnt;;#os$@wdB z{F?(d@KMMbi0f6bdr!_`0GEd)J9@ar?x{R{UIxaE@!XH&f9d@3Wpot(0s<9uKwsRQ z8*y+e+;g%bBc!V_!|e^7)l^IOjXwk}<#DiZ!)iA5NDRj01v2-Rq^BhwY7S1qSjZ5&ZuZ$*9!^F6!Slb}xEQ98QB!}n8u z!adZN^sQe&eim(mzEMx1n;JwR9mc_j-`9R zH%OS-N?Wjse~BUi7Dm&E{!nPB?$d1pFj1yUf zR;D3T;Z!!h*PF(gJX>gO*I2L-_80{TT5v{uBRp90-BRBrf_t0xldsm>!FSyZ1G$3fuYXm?x0Jrt{xBD_sK(4h5>esno8B}gM#f6Y3xYXN+y{g#)b>|h@!Ro@wQ_mxphB@ z-1Kt8&Sow0d6*Y384yVgw{rM=u{C)WrpSt#vheh`RO&fQo>i?&MXgjn4Ss@NnRLp~=B9;SN^oLL}zYa_-_2=>;Lf<@aYfoi6DM8tQ(W z#2pT2mg-Q<&ZI>fgIVdNNrYw{ArqSviS}JzI;qPQZH}KqB;L3l9R9g;IHqQ!y)HY|7yKkJn*c-?(Zt~kYUiT)XIsoWki22IJz zq3dMv(z#T<=NcKIYeN4Wp3b~RIU)aT23h3s1!|Ty(}N+RWYk}`ika1y;K_wZ($anu z>t%OC$c;4es4f!~?l+J><981oSZep9P zj{R#ku`g*~P@^f6wmr_JK?<^1bhyWIL%AnT=vj+5UY3%T`wwHtx)+!~Z7GSV1`ZV>MZ{)~gmv$-v;j##?pI&S-F zD0s|!Ndw&hCW7022Zcc{JOtx`nR0v=FAf>L+et;zZqH5ZlV; z(1f@D(Z;p`RIz_V_ZcXVg(@qcxjdfSnDu~6?s!Qu=UR}Q_^tR`zlS82eKXh2>>&;H z({XF@G`8r(B0N$!5yGL!kI%#liH&?v)R_d-g8qDay}qf*hJQNw?~?+2eap5u5P``LS~dtKLg z9-w!=gwnup6Wpqji-E(u`0aa!<8p&4!DZZs7U?lqd%2&q3)l8k^9WK=VrL_N@Ppu~ z9)aD;Ikg81ba1xI2kQ@7Z|LAn!JZs2jc%J(LN#AqC8;K}aZoA|?RWpi-W?{)BSMO$ z82yK)89gM)BA5Jq;7_*C_=a$;fNZ^X0w(GTnY(o|T$G8__+-V1RmK`j2^xM)kXB#&Rv@kj*i0q z#hc*qqQmejX8`9tw57jGy}4--4s=flkW@P3G4xSaD z`#GEwSrAE!=8BLg2Q{`TpqzFtvV@StcNkKqi1(MJ@ZM_D?6OliDHHFcL7vCRk)j&B z>iU!J()vaGe}{72hsMJ4S2yYUDtX#j7EC0c7^CjOpZM`{7k)iCnog(=7M@`r@znE$ z#I7=xtGq7`YG*5EK9XxAO83uO_ni0Ucb%MqlJk@3`S%0VedQ>a;CGx}-K>n;ruLDd z(Iv#tB2GAOPUDW-DpT3n^T_X+hq+~Q4-u=j&Ct~t3`g!~py#=_Bz)ULJa2a!Z)lvL z!-Xl4@!nqCubW4J0d>XAV7N zmP9|;#FI4dWD?(~NH+BSM%T4dY(ygdl9&)zuu+=^dy*138SBX~VfzdGvg$k0Jfn!` zj*Miq+8?>onOyPxb$Ft0ion0Jq3UZ#GUb(?ROX~5oS5)}(x_uJdO(Bxuu)?^Z@kFF zh1(I2QElX6==5QBb+}9i2S3<5)90(Bh?-oYX9wpuZo!rZtmeEgT(r`4BjK z8%f3E-;;b3M@$JG!VL;7@k>FF&L0^op#OW&nS-fdA{Ps2`!+L~$^4D2ZZC@(Yf8U3@v&Nv0rX$Xg zG==Hb*HK4CpStT0kqO-|c{RxhOfc3!QHJC-C2 zcg1TXBj_Zz!}LSZS{PLqPuir4v9?}~epM^O(uy{IhhMeO%>z19-~^3n-%UcVI3cU5 z#EEyOfM3WL65v)1N=x&gd~+InvRs7K0UhMG%>Z}a)<&Sn_|!D$Zi98Yn^1nvaGO;T zGuQ@S3H)>0gA3AJ23+toGI607^Img^|Fl*Lx|1U4lp_NqX5l|lHYS-ob@mr}+(~qs zg%Ulyu$SD6-a&4)Dp9pxitxC5BTamL0^*|7XoC6y7`*p@Owz7Exku$}xQHhFh?9lo zX-~<(zM1Ur3t9ZR>aumn`DlE4As&r=!Z0l>faFYW#u1>#DuldF{tr7+Qgn{4xz@y4 zs=JeL13erNSqZrT$ylxUlyfWgz*|=C)~`F}5PRuE@LnT?|LELGeIkWk;2RBMDx60v zEaDhH#fW>PIT~WBCE@8eM||8;N|XOxq<=cT;l~}})J#Q@#3i1m-#5C!h7Jw13tmYr zqRdIn;-#>wHwhgMkHz}ew{Y>wX!`P65;5K-cw^Q$LSn@-zWVNBoHNLw;o&KqqpJ(~ z!WBW=4Mk88NY^{sJn&hiEE{KI0=Ha^ucHzC}h^-wm;*iKn87W2(x20(lWQJDADW6Jj%uJU*H(b&)doUnpsNS zLLXYWNu=TEfWwgQS;gHDlf}^M8d$ummD87&rsu3W;mk6O!@b9^Tr2lx$>ou-fDqx2*LS*Zy(= zT>j!9cyip?#HS;O#q(r3OU@X}p9=Tg)2l!+Ar=eIHDdKYZ<;gy6*_Kf!JQFrNw>i6 z|GH*d&B^=`WR|;vkfrs&-bEGI;@nLJ@0F3*3w4+ybO)OjO2E>L0=ST^$zGUFfQQ3! zIG4dp{9_hR@=D**C_8bwc<>Dj{Z|DUIFqQpl7Md)vasX6C3-HA$8UZ~7+S4E;@Ae7G9EkGmlbo^cIhv~@|=%U}Z$*mLP$z=yq zRBcFy7G;5T-LxC*QPW1jIs=Lt*5hL%pW0KOW602iJihgf61sK%rkA;Bvb?8(F1%+) z|4J!fr}RFa9<-sS*P7#`mtTmr>_XhX{tYp_@6S1if8>^$@iae1pUtQ*ptG7x5Kef( zs~x($tFsp_Pb}vZMfQ)_f{Gz~gA3bj`5 z6xb!-h{;$}j9D3pre_}V>4v%BuD%qnn50`z_AIL1{eC{)Gge_8>zh$B;Tk8F`jy;X z`JTk{%VG2-JNhm$g58?vhFeAk(6U@@Jh4Gs=&Fst@$9acz>363Z4eXZYIOk z#WnaT-V9PRI9_MGGcl^HAopPhCe#km!vEet*5U#}{yd?ndI9LOxrm@qvq|E*$#BR~3r7_;kf@3*n*X+#95Z`{(v~r3D?6F<=cOU}=uPWizGk?Wn-4jk z%~3?d7(DZ;Xwl5Y=ys-v{B4z@ev7=R&ZQ?@*`tf4s030h5b%9a(xcj0_gP?n;q>qXgXPgl%tRHLjXk97<8kedbj zxl2>_6OEsRBy(Fn_t$1Mu-&cP@;`@3g>E4=SYVC|cV%)?!<=E%TEQohl7iL`zfhC) z7hy!60;*lyDx4*zWA#KWc+r){>U9U8q4YQy*2%MTJ2RMBusxFu^n{zUfUP(niZ9AC z@Q+mjeHnQa8)u!guGkO;R|3r7gN`{goiFE4dG9CZX1*hDLRX-Ko*i`&<}Tx%&ap8a z@(@_4!FCy)WY2EPf&QX(G(xKcC;dw$XD3a;4A)vPqw)}3bsx+}nbQl^Dp0)cG*S5` zFv!*<(m1nZ9?r(o?!89jiB4zDYGXUJ*IW-)@;!7{i8i~Pk%@aE`>`okgEjxHfCJhB z`ykUvI0sq}J9{3G86QuhpTK4;w)u=`mFmPA#aZ@N6_#$Y3lnW(NpY*K^(ULL zU@6u^W-Jx88Rt02U3Id;Cv_9>cwIFtQFW(9N^#`*lwQH0yvMA;-_+|8#r!~3;tW6E`s(s7bi)r!>F9(#_9=AL8D1JSJX{#4YtGMq)L_b@9lZCF`; z6OUvumIYz3e${m{x9b{y3Hw2=M0vpUouyRDS)b^{TG2%cWyq-#5Vh$bdD%I1*#hDE zDBL5k4JM)gt^m|BKfz2UmcbguY@B>{1AMz!f@wcQ@x$#fRDFAx%-Lwcf>%!`-l@`c zwz`UR!#i81FU-6*whNukJ)cQ-PBb*GE=RAlZS0RpH?rV~bg(3r4_X|{5{#^Ll3Sm^D@ zIq}2~JY^?8ij(Wnu{>{3KvNrDkRj#$v}$(> zd)AfA$@Pk}8I2jNTTkddIP}59gcky9z7#%460&SrFq`yE3Uen4oh{*~CG`-B`Mcn{ zwh+=EA%jbYpSBL2{fm4~PDGDWIYiSp27dN6(%yVC5|lU%PWVfrndwA!(kcP7n=@Ha zPZn!Ws;BDt*HP}TKYGVIu?uIKaqL9goS&H{Wa zwS=*Oc63fxXOUlB1*hgw^4ek%w7y9sy~8Z=y0C`&RNzWZyiB4F;lbQesnu+)&W zgX%UXQs1o~$e;e5L~((zr|XNeu^iBp7Uki&@&nlP&X>$slZj&XR%~{w z5e~E3gMS8dNXyg~8nHtRU9^PG^w>I*`Yej>+~&YC4?iHMWNjhf$5J>zKhe^Anp{)h zOR`-1Hi?y949iDm;Hud%{NYv0Nu20Q8mDZ>TtoJw^11+M^SOafR^`&^o63+{x}n?D zdt{fxLLh&XXi)qCvbkj&cn14(^$h{kaGp>Hb)8K=f4)ZJ1qSxzQfF*ov2ZbQG&n!z ziRIo#+Me`^RQ{2L7^{5pt}2Q?aO@#kaFP?bWCCZGcaYE3f9Tw%4eU--J-3N}OEjG` zN#2J{GGnNMA9^!~jtjLR&CA5tGo3mIS6;xEcQ?W6;+I^7aKBXdJdUkr4&vM=H%aC_7u26&fDgN^ zSnPYjzjCk+qV=_#3nN3^$EI|6{R`OBK2(K+#g(|~d zk}0)%sI4DJU2GGGX5cH5Ry>0f=Dp;#xCu$C8a^{(gn#XcwbSAJ5C^}^L}2_lGwUs| zkL+&he+{^N zv<&OLK0pEvy{;|(&mY40eYjFFo%c(hO>L9)aI?8D`{!qbV!c81oLw{YzLJD$-z@5t zIRf8I5Mll`1LUxQ7^qnZJj#k0IHAyt=5oDM?VKv>d(uEByjFy>ztZ@chz01{;{o{a z8@3r}I+$hl&+=D(oL|0)MoZF@YipOK$uX2^O#~7ss`( zz^cEOF=wL{JEbSAxo5ql<6Z&TXZM<)EPjiY9gjwffQMAv&zFvUB@0(mhJjC|4Y`%! z2-k*>0^J4T#Cq;8Zfc}1KjE^xwQ)@=xv|d@7M2|%KR@3g=E1&nyVNJHY^6E7c1sn0 z|4~7Mm48sTQjXQBWWwPOA}sE^DLvcYKVYgUUuerFuC;w`vza}!9ovY#Cc!( zxY?8RPA!LkdCs^;$Ye-8_$;`GQ*dmD7Pn!_3sg;b$pUv@<3mPXkPe?;D91y2rBSG+#1%rY7u~A5A?JE|D9^leZPTz#=SwSH7_jt96EQnUmpym@Zg_ zCsUlCP35|ekizACBrxQZz<3Hn<@Y_9v)z+@DC?t3KJ8{Y$5P=`+jp4%(F$4`m%@f_ zf%QP1faq5hn+Cn@VDdnV6<2@7*=>%nT|&ubnw2UuoU-0}guo|g_uaz^8+FjO;4e5G z`c8Z9|D<0J2cqU^Z{~C0FrD121`1g@X#LU*l~4afcZW>OO^=3T#YU{mR;BT){?NQ7 z>HJ~g`61nuY~6nB80z^kRN9bA8)N6AtfJtPYC1}tYh&4=&0;oV^IFKp&30HcUKii3 z%b=%qb6MziN7i$75+?p@=FfbQVzDphlFo(GVR_1Y61DmPvSSuxq4FkT|7N+6&HaJJ z>PM_Eh(09m?JIHcj^Lh%)gim5NI~_>Da_d005(l>!{DFyNzQ^&s$V9L%}qJ*C-)sm zP%6S=QqDD}G?Q&lghj(q89HjIF5bPK0v~1!hi+|ofE}H5`lodyB0CdDT^51h!=kWS zZ6o$(>k`#VA)w@?if`khz+XxU17^Gydsc*3K*bwehZlWug1kTF* zk1Nd={L4<3Ed9%NTKh|uhQ|#fH{=~5EPf*PJWRl+b@RZ!O^5m}-h~z#UU*FMBC-7^ zB#q8!VV!#tu~pm8{>mMMZ4>1oad8S0vC3o9MvJl=2W7c3A=h{CZw~X4@TJ_90(y6+ zI^0y;3+nFq?BkViuE4K=<}KR>IvQ>CC;N}?zK{-b*W$>n@ZDtohbDT+%7(0+@DGDu z+j7Ty;>pa353rek$oAB}w!ZYa1rG(ks(J7$3tHz)CG!#_`0({H+>=ogslT?U(91~V z3Lku6j|Y0`zCSK-ZhwXWQ zvjTah@O3|qY6uqC7qhUPE@mU=*szeML3-s%JWT3+37;13;wl5{K`!hWi94JpINp2t z+1f_jyUYaI_U}C!Xl!C8yDpM}!$b7q{`F9jE`~nej9A=_d^B6Pl9oBTU|X~g+*cUn zjx5)OV&@=I=kvOH;^Ir#zP=O<)+iB? z)UiNjG52(&Dw?K7(~=$Xu<^e*TqWm=2VaY0NM9RmyH*ETf}=j-r8q12r@#z1&!D>+ z1s6cuNZ5Tl2&}{sNmpk!*?aaemt^6Bb3Sc=R|{2%{Yq8ZQ?Jd`Zrr7I7f;ZKgXwJI zDnmA=VLB=A+>d|cM&ao?Pl!6XAB=8_fp(Z29S|*|`PS=c_^>oyr-YEk+eMhFRfEek z6YzAm0ee)v0k{6RP8?)@aK=U>@yUPYbajd`M3p{6`_53T2uerauM2S8&D*%}AhH9W z{0JGA1>c4~lF$n~$yLR}czp6#^5SP0v}BnhKO+l0mm5N@{0zFG;3%!j51=XznI!DR zcPwhYj;p0dIqTy53XDEbUz!T#A*7^f(Wc6)VwxgA>f$Y0{n%YJX7Z-sQov}2_ zdk^iaxl1p1r_p&MRp9NGpJd|%d$Q)&ds=%Ziq!83pm}yGp!1;=C9erSgf_ra~G_4hvgxTy>I++zG< z`U;n{9E8}KHe&qA9z;w%VdS~7@ao@Org62DI9z!~)|aW$Hot$wCGRI)-oAi5w?9WK zSDD!)c#E^hkYxBX*B9>3TO`b5-{HRNFKD;JO0uNdnw(wJ4Cg(hvBJd(f?{movAztu zcW(hiUsK}$jN+K?n;NpyJegNLF9D4s#cA=nRTw$nj1|oBhg1J#t?zpus`+jxg+F9% zanLRYb0T-JFpnl&(!T?>?j-P?SLh%Pj|uZ#4N#~bOu+B&O*_jXK>Aw z6#l865t#nz6&$<|{@)#R*nKP3?G!REbsDTA?JtJ<$wI|AQPMWzAg&X?Oh@`WAz%Mq zCP(ItfElHYczr@C76xx%kItT^cL(aM+oFpEUcqtV93FzxuZ`i?+P&mXdaHnCUky%r zF%K6XzE9iN#!w?EHC#g<(HUF2N%Kr|mg{N*>VC7yxd3+znIOzGL?2+VY8C&I+dwBx zK11)k(Zpxu9j5&yY}|%u{5?1i{o5_+$aACc(fB?ZrLRuz3h$vNazEA!>(A#BQ~4zK zUh+ii9-eKtWL@!Ej8iD&W9sa{tA31)=qSOnxosSq^T&*>(qBT>U5VlHaT-pKh`<+e zo4FT@-LUV^B;K}l1O#M!q-O*AsMp$^Z0O4@CVeZEQ*5zG+`P@be9iumc;Q|;3AR}P9|NMO+^Zna zHhF;4hL_T-)`AoCX*rql3&{SrPT*{Qm}{OWMTb=#qi>_aP_6Jd^@}`C?H!Nur`nBy z`ZbVCor#>U&k$MI+(T--^+{m1BwKMOo*0>4hi8+Qftc(UoRWQmw`_~Xd((EHd*o8= zmmdyQYH^UiQ3*D+J*8vr)R1i~5k8FahrL1J;Q#JAK5d*t*xw<#E$cJ&AKOBwy!OSF z(_J~ocz1GtvKaB(<%1`xoNFx$^N5WDqq_YibjwI<+B(e(F-{`qTlkQnSEugfR)0 zZpM8hhfsM!Jl4MvI7zFUaNueVy?a`j-SioY&T~CTzuk7yx;C71`6$YzY&c2Sr(itv zY$PldDh#2^n|V6#9o?bngF6^GoT$-M9YY)7mKFhMW$a*svZRiBSXqGkJ*HwjwGpZReL&Uq$b(ZfwJwYitFRigYUxo&dM zC7#-Klyb)XBQbfU5jz}tp7sP}@`sK^gWG9w_k^wl^T87gHnIpPRk zacBpBqrQLUp#6P(C@{=yw-%8@t~c@Q$9(*gi1?>V3qzuBvQa|+Gx*(i5^}m9i-j!L zq=F`5dHp=^ma2_$2YNBk)|p>>PamfDl@tBd3MkXO5sy~I!FAUGenY7uGd{K-sm}}$ zs^L{ByIv6cwwZ4QF(JHEeg^$j|GG#In)v`Mf)FbS0<2o)(TJ zdN-W$g(2m?E1N))S~9&XFN@EMHQ-HI7kM6YfUNr1#*4ob16h@H67HQYI11OISI!vR zF=HjV`ek68+X#dWd337nX%e!%!FtOGNA7o`EQI{`n*_emfU=}`t04WU_=^j`o40xh zx2PO+RUOH>VF&1pbL*M3uM>0JX$&7E#z0$o1WKfj1LyP}Qe=0C{#YQ7I(OcoyW9w{ zy#A2&%BZ2Qs?SVwQx&pp{AluSdJR!?72u0oKGx<<-$>8>u!mtkQb@Ms2fQjBkCVTyNEVoF&rNiX$c7o9BpCOQJg>I0JY&0frnu@GtE6S;-(1F~IFjhT?zT1^R zuE#v0;d39}-yge$5+{UU~kE~rceya7YImImtnGPqm z92U4!b8w2gIeVg?L_f|ui5q5DV?*;r@_4ocHn#q=zWzZS-484Pw;851NIw!gbew2? ztUKNxrpM;j$-v$6sf1e*hF3S-qFZezlNIu#xYn78)Mkbj-dWp7F0aUgqaLR~JV6eZ z#@&To!pvu|$_26VG?u?R%DmLXaIQ)v8TBCt1dY2*+|4s!>1f8R|D8g0gK$o7k0Dn5 z3>UhN(}=?@eWLq}LzRf#V4}PZ4v$iT<-0%9H|v7$t0uJh72dx_&uIV!$l4O|&;L^pFU=yTE%9PzqXQ!T+l)TV;cHhqDycpP&*Bk8Zd zi*S$H5>E8g2q;Svr(xb%cx{d=NJ@x+efM?Ta$cBU{&>%&SxyGclJS@$bSNUSR)WUz zN4TZ?DO&TRxg$CGWUQ_Zrh25I zCCVKam?$4qV2-fPymeELoF3u984X^eo8uxOWaLGHj(w<=mxu?HE#T(&|5$0|GF*C3 z8cYi}(V!d$f=^W~R}MZ;8BU~yUbpnAa$>Yd7CYOg@XGg+aZ+oC zaDKS1HuL2@6pdK~2A4O$_xBcL?fJi;Jv7AM%GyNV$@?(pS5l~C*GyV<>~a6ai)6%` zmw4$|C#nkF2;CkL2vqmQvhkkujJh|B)jiMGB&K4RUN&?u%@WSqywN&XomE_##PTlp zkg{HYXKFh8yWsezY~Oxc7ApDd}Nw<(Enze`UHZ=w47a%6h)x`r4I_g`v$Nr=anZt1>v{Rd`IBw`aZWX|0JWji?AL7awEY}ipzwlY ze_KexZ7B?R^@nK2I#A$&Jkt=GWu=x5$=`69jyJj&?Zc0Czzjn3V&?ssL?D!{$9=_?=}wS&#>bHck?j* zkWxhDszdmEWi=g_phYXq6jA#|8Tol94Zf7SvAnLWpqSzgYpiV1WX2JCdvZN095LFzV)8Jij2KjHbCt6A62%c>pyd&gs_s5&F?oal(;ZOs<-$-#u#B5MFcn0L`!`bts z0-_e4MB29-!Lwl>QA$%BhubZ|r#z#p!WMu;t~)kLFUF`@ef+Gd7o=hC2r^r_41AB< zf>?<>aEsch@!tx1$i{~K6cJ|+1~u6q$v3!q$N?kDw6QwImjukRph=VTz;^#>l=&`A zQ|mU8=^0am%y%+rd)i7@%`Yc6W$&V|bs5QQcc#HEU#&B*OF-=C)p&Tw6XqUGg?%IH zNV$pxe9BOU7u&N*$AWV-HOL+fwpqdBJ{2M_a7yZ~n+e?}Ey7LK0*gyAyrtk6e5|QJ zH6Dr(x+@fpeh`{$WrF`(s)B1gC-e=aM>E4&(_m%Y2^e?P7*+O3V4$$4>v<8+EZnT< zxn06~KQ)bNoC_0}s{!b&VT6Vv+sW;asZ77*4o*n2$N0q~!TX{ld$ZC2RT6V)*!#QW zx%d;_bB+p=a0`UEZ~9>V$OHr4X5y_mC&>AU(e!zk2kE{ha9HxxxhUm8cBH5mEH2Ea zn^(0{Wxq5bW?i&FWx3bCCSSj{v(NlW8Uk;_$IpaW=j7_D)N9y%blm!N@A*+0Y z;8N0Q>{yWo4V!(SU^dT3?!QixWt^d8QxcWcyG+wo2H-)lTu8WQ4%5YRX@Bko^1Gvm zv&g7~$jbe6Qj7~+Uax^>ZgObUyd2cKjo`Po2f6-70`|6*lLdq8N$uV&YI4q)1-~1{ zyeH4+m`4ley|0x&a;+E_b{Yx49tF;??_su;C~KW_6CMh+6cK@`ap_kVj`>gACbP*1 z#P>{KjpMS&#^_uet27r^ca`%E>3X0%I|1u6T$yL?7Cfh1N8X=q;sUEPP+T0?uzpWE z>t6xgc;z8^@%svyW>n9|j2g)g?~;l zAnp-HbL@I>=7Y&h?_o1J*D6A)q@rM3eFC`L)x_+v?ex2|0|qQe!1rDv{6m2ybfcq> zZ=KXeRM!PjJz?+qxg#4-3?IwhI!?f8p=YR{lpY<=2@DMJT9Py=iD{mj54U$F!pity zOeQyIok=f$%|-_FQ3fRbBp|Q;73YL3tL?KH4u&0KWXelBnz3C3bVsRR=57U?)^HMn zio$5aZ9^*e!2-oH+;L1+Hoa1F6VY7-uX_~J7q^7H$O%0bU@l4Hl)9<>nr!|`c^tHj z4!~G}-=%X&1YQOefcM$s0vkackGq&~dY^O9>&+N^^GW!OB$p6b+2GoxEn1vR&}w*j zznNYf_(HA2ht)p+n~i(_T_G|BaWF~r%*>^85;6SV7#MfYn8j+(VIM*}_t_R(J`>C7%NccMU2p*XBo4$xn#l5Kk^!}`H z)bd&(u!WRi(Otr1e`S)zwM)3S3syt!1wGPvauUvYYXb&)i}@SFrs3H?oA7MP6;gBi z7m_UaCXJ*?y9(xtO4_HO>{wBNDve z5?^u^+h&Hqlz-MV;Z7;JS`dkLQ$3i>xjdHodIZxmxPZ5>zM~x%ro$~V6%Q|GjBAa; z`8k$od0(9G);@}NRL;`#A$l;X+f}%~eVf_2UV})TG=#EHD|oUciC>OKG3&urmME~+ zhmK~GjM39z$2J!%Hn_yOZ5U0YkNLyZ8U{`m+TqL&QH!lG({_bi;i^`a@hF z7q%Y4%F6rn{hPITbN)Zvy3Z0f2<)Nr+;zNcTSXR6QWjigawOfso;(SiOx&!JA!D*M zYB~L&Z%>Y34Go24doALf6Mx8~{>S{cAaOW+)q;LlHj;=+0(bPmbZ{%}BNrlyNnxyT z?vQPbxxGzn%;F-takM2|O_0EJejhbno_CVkFcGS_~Cruz#JP^kTUPdo255C-N!5^Wb^g>W6 zemP!5d%yM56XQ+5De@FAUeJhYd&gsGl&QcD{7U}YRDoIh$Fq+MKahc+-E@OOFkh=` z&t0Ror0VZnIGVB^ziM{VTO-^E2VLaMk#X?+tP)N?TO;%o45_-G;2+#1hwWrPc^|eG zJ1>|s#i@)c3;dS@n|nBorKK<{Ruf+vh=Tv{Ch|_Zm)uG(p$aFL(ATZ|$&G;)+A;MB z+&Jb0SC`zu3sTeJNA)ARQ0Sz3^qJ6a-rH+`>Nio1zpuDB!Ktg)CI#cZRYK)$C(=GC zgP0axqYcwM$isE6MZU|Nqe)l7K_L3rbFRTF?+;cGhk|pZ*>(FGUbXxGdnC`Zf z0viu88Y!_EI=KHxOl~}sJ{7Y6wLy5${2*~qSAiu*P4P;}Qf7bW63ssTfv$+OWgEWe zg5~5WZpYs^u-f{Ej4R#^+OaFi{^)j+mNb=yMElTLsZ~VQS+K0Wzl;5CzOZvZ4!j1V zh*^3JFTHLS3D=gPXB&>;;lL8QH@lKlzxlBmDqmUDA@Jy0Nf52IKqp|Xlzz1_1HTa zzI)!n%79?Jn-$4lO#hGm)efWy8QIvt-=>$&`> zwZbg;IHt_G!Oa$P=hYu((=Y9Bseyec#+P|w>Q7silGRO;KJ*ay^_)yEeN0Aq%7Vn- zE;1}}8~dji03%yGsDeu|b^DQvVRAd@+s#L5zw>hvs681&rmjcR&wz0g%dH*e zrOS$sh@F|pv$#j=&VvryW6|39G zm%md%cVjBXeIA25S4gvY&r``1wTEPXV+VP%Oq8AF57YMhg3nCco^Af|nkM&`a%^@3 zKW{yu^zvQOdFU!CHBmY#zzDpK-Q>GvB)NW>%OoQHUQLdWpESVzaQQ`72lvyD~pKph(q-EbyMQ*V|Mx&qPY&6D_oI!q=)9F_TahpMigHkz?MH!6XJ<}I7?9tWbfQ0eA?-SlMT#EB0{6Sja3z%$;<}wo3K=~9$d|Dh!-1Fbjb1O~}hm$j` zcAqf7FJH@mvlsyh9^s@RVL1u<5D)z?Ea9XuES@{Rl0H^9q+awOxNaR3&NxnB?1wbW zuNnuD?r&+*J{ux=lq2)r3+J6C5lp%J6k5*?V#}s>;)uMNpyb(LovOVBz3+8Wm1P>V zv*s+6hS~^c{g%w;T_J|e{=t3f;Azx-d(ta(8D?a0czs+87qeGj4VH|do?89HvU3+- za{CHz>oSbL(N~FKPe%(GsL5>E$8~hZ<;UchcLJCS9e~tMK);V0*{k&Tv^mcfuU|QZ zo@2xqGm=k_YB$WW&%A9}J7VKzCY z_8h+@A_XUo6ysmq>mXgSdgN1^1sfVaoAt`78=sW;cqR@9l-}9Fm9TR&FF&lQmC4LpDcAeLxrw-&HD$JY0r972D##FC@P$tx}t?r zl3uLi#~RWdwF*sTl%YjM4u-;_xd-8T^vd2>bovp0>^eOUD(C=JZ7$}YNGQQjl|8m< z>mpn2z;xG%vf?Zg!n{RmSLV&e2(#s+dxZsG`B9PiyVvtUJH3g8Q8|`w8o^YaA3{6l zM;NvBQtkUI^Ko0l3i6g^NRTA)T4V0^Ye|zR}RJGwCOJX zt%nJG){6v-`^Gjq^Jm-0&r!k6GG~bI&lyCb(}-=o`w+D&YB1lW2iBL=^Jg}1B(6nA z$=mm3Xfw42%}tlV#0-BZ-SvZ5y$#3Hk6kcYVJbb?B8d|%wlni{bMfm%Q|P_qhF=%_ z!*?>~uw#lLI6RsLk|&jZ;T~y{jG!HxOLy`o|%stAHo|VVIYlMQ=(zpb9#e&hFvDlhqB0Kr6HOivyjD#pXXj*&8G8Sbdb|~WSB%Q!eX&l zxOcLXluRw=Wb1fxX1=F&leQdem0UyLnL1DDDM!TKLwQwt`c&GJxg6^0?cC(!MqE(_I=Vd)0Z$aX8?giFI}-Ht58 zk0W#O@`DSw$|4dwTddGicLDjXLfHQ0-Kcr`AnJVjO0SS%?7@^a^z~ZAFCxE=TDs4t)2tcNAS|I>a@=y{0*W8U+8qO#`)35eKDdY|vboywEJO-~lr z>nPzHodm)C_KLr-Sb{Z+DY56;1}t=^2)i^m9-<2)I4_U2lm#C{=Lr>97!}LHlxKjR zQv~-c7je_CHK?O@jvhHQk*L)f(@NPzctc4WdV-#j(6mZ$U)n>XC(mM1W>v7ZXFIr@ zaD=-L+sMM;eDd|Jtl&P_Lp8pfr%?^VAXCX5t}B{izRe^~)OIboeO0ygw&27uO7-N5 ztBd(mzZ&|r=mkHkXA-#YU4+SnWh7+SMKoLJj*H*wpxcW{P?j4&KluK`&{yC2TZVhl z&)_*%rQr$_HVb*r`IFIa@H!5-M&Mvt1D3`t!FzXt*pX?Mt^d5YCkunz5Kbq;E5pwO z)fbVZQ3*7zeJoCFF~**Ri(H^l27Dg1oV<%GN81@$WP!qDSfknr23ncSST!0e9v!D< z$uV@}xz#W^t%q!AjwI@W`-&~`XT1Mqd~R%x*&0n$&c+Uw?&zhW&(v!2bUV2*kp}RP z>9XLp^*HDEHzMQeI`i)6Vq)MgVSPoW988yuW*wWtF|lTt&~r)R$KUb>|27BQS~v?2 zh`Xa`?lJVaJ_QfPn!*Wz4`fzn#a+v7;$r?T#Q|@M%Xk&^8LLCO&1RBArZL>q{5@p+ zo^~3#@EOuEsVIHLi%HvTg+HU5!Fy#Z_#|4O*^Oy1euXUFT;hx-P9@+keU`RV63cDdroo1 zI~U`Kf=l$z;@ecra0698S;?<(6l0HyuUnrsX1M%LGVZz{LNAWXBlTU2=|zoSLKk>| zKc1gM$Lda`>DxY$8Mgu;f9V_e{39Md<{iiBPfhW*?{pe;I~L*K9o(`>nPs&65&r%h zyo=Vw_LK}vZi^x5)6JnrU}vql^A?+QcYyMZR^E94x38REqozX&zNCTu%nt7L zhYK`5HivvOdJxCwWe(g(HNXZF8@lKEe7fMs zfAG&Vhl&m+a{u|}voAS^*w6Zhr0$0;daTRC&bDIuDA9?ec<<(x$?e8&>3Vz;dX`>X z6pLFEwWwj{Av$TH1_Zo|CZoqHfkdATLi{~6E)ZueAvW}8d#QPDxB>;Y*@h6$s9rQ*MmJ@_m)N~DC3&P z-L!U!z%5B#jj691x%93kt9F@2OccH&zNgFalX*F5w0TV9=RbuZxiM_!;0bDYY!Res zX+c29CA_0xB0R@e@j1e7X-jcAf9u9lV&AL*hkBJ*wMY}$I?W$NEK39r^(M^Bjbv7_ zCOE9Dg%*sX{IBn+;Ai%V*mX@O8%B!5K=yX3nz4#AI@nHEJlX&^mfyk))jMFiu_XJM z_8NQpUm@o?o^G^DqD|ESs7WP&9Gwj|6T|6!izpNc%S3;>pR}U;rS**!%EBJ%5Ip|3 z3qmeTgj2=c^gv}J{U1eV8j#cXwc#`g-ifA!$-1l{pa=nU$2%s3K|7 zj7lZRkUD#9L_$c(^eZ7mh=lStz32Vjm!5jg*?X<~zOMYjLM&09jOrl|gna#FzF{bV zoF61KtFw}56`8@5Gq!x(rOgnw)*P0OD#43|-{|g%o^WHV2G>`l4igoY(wNL)8yRZRItBfVMu4?v56y3Dq;?6hAX7COlD`DfSH{Npzf=D>k2n?jK+=Xf zDny~&Lo>^F;1k(MCRA2&&362U$EVvda@F#N{+|T*pm8}hMc>6S}@Zv1Ei|3P-yG?+0rIM^o zH_7>nfw*?beljsf3fwIO7Sr7{Vi9$P+JC2EF=!TmkW;F2NT#V zmw2qU?IGjtjJRBOTMk#$yywJ1Ch)^wc2TL?3?%!-*!%KtY{6?wE6`6`fm~*n$gnmBiRvmsLUl#n*P#EjV!+JQWEux6skOJpRC55EaK(6%rN|) z8~HKHol~4A%I013U_aFF(C7axgKD2xVpL`gW%WOyY_bGtZ8Kof3-;iYJ0`GQCz(EZ zI*hNwky{eH6;+ki@DE>`G5rlfr^VQkmG`yM?V7n*hPrs}LL7u|{z^>Cg$~uCadg;8 zkIglD$nCAMfH~zV@UHqkcP~iD6F8*9r`;{|QL#4bczYFRy!k~o$()5?odjHUB?sCc zzoh$wOlxgq5!L8;3qI z)BD+MmlCpA=ufo7%_WL8SICz%RW{<=QucAV0t@>5QD}#|v6H_N`7pr+JW}x@X*?Ti zeQZvd&}SKick8vF{jDK>SQA42L>j=UGet0U_L_#tJ0DZi>PuM4ZjkRSTGYNN4$B9J zad^}X(pfQz^&8!z@0KoVNa%^>Z>Y}apVJZ6Mjx(_-1b!(nM7kvX`ZH2sa?U#&bz_w(s&mp(}u`xA&=`D zBF|o1jADKUN^GgrWVjU>$mP$=z``~cRy9$Ud~dNeyQHgOB9i;-rR zRlef5-ye|6GzaTP`yu131ZOZkmqw^_7?YC7xsUdP8>u5PL_Y%0HVWRGPYP%}M+A1b zt26uJWl;1~85dlPt52EUBQTXU$dgSCG_}(K92DHi)pwV%Jtd87J)~>B(JmL?oKPh@ zMr|h6rHfH2`YkuPQW1&neX`Uh)k=5uag>dBzJa-0S1d#?FFki_TeZylo1hrH{E2#vkamwSe70QB?fgQ<@pDzoBM{0+_cs z(T?r|)^m?%a%=s@1AQus?=lr=#PY<3QLkr_-!mj(MpQnZD)d8&D$k-!lsHvZ*o;;^ zLeFqf78?CHfrs4tIZ1hUDy>JMsrI)+Nx&VeD7 zOuE{~7A_fSLt4W~^q0_wLsstAS*Nmz#gjN_a!TbLRJIUt{T`C+QcoOgk5a!4E*P;h z22Fq1q4}Ycxc{jfi&HSB)p1+NS3@n(-9T~iwzu3!?RMhb4`jN}Omge`cKG`0J&`%S zjB|JHBDMoRdFT9Oe7jDKd*&|6jx?CUh15Gxd3iEj2X;6w`~~s0AGB^CK1tTzDxy>K z?cuER89F~_Jne5iM#?_?rd5$bCnVJv%}2))FPR3CIywtiZd0*xA44qLmLxDf)6@%pt2R9bQg?lg*lg7CS|ux|n7?NRs$j^Uz>H%Q=o zS)r3Z19Pku@u_g1d~&8353c-6PTcDyGmcDwQHHb7{PPtW@LL(ZR7KgH^^y=YQU!e` z8IqE5J}@h%4(-Q?L9ow1`n_W%CS4Q?MUaCNbn6LxjJVO`vv}nNpY~uJtQv&Wa1c$Gj zA+5{&;C5aaZIK!c8&@4A`h%-Tabg-DCS;J34@DA{dnWK&?*MmkC>B4+WP+x4A%8)o z0aq_4Y?Dzu;`wyUj95&v8&ipd(MZTaZFb?g3f9%Gzy%wPVCnB{vQ_RNKiw%E&xsgf zUzI)vop!{uT6ugI>_sgDir7qpvltSZLLHpk@Oq&H(;RSO$IkTQ-h=~WX7?N7F#RV< zF9`yJ(hJ-!^Ag6s=+V-t!`y@y->`lD2AU`TnY3g}5$U;gXl*r~>AqeI_k=#J)`3gN zbtvJw1v+%M^I7~Vwgl!2JKC&y0eGTx6u&>Mh8uIJh>ng*gTyHU+vIUPsK40@Gy9b= zv3MiZ>=b2LO+~12O%?g^s^BCq1EE>+(5#&;}|Jo6jdZ@$fG&}0NJ|8=_m*E3P9yU$o`H|uqNQ!w7D9devOH)gsB{&r3 zZ@!QGCoibp=1OXhT&0svguwNHP5f*4!pm>)pyz%ovcvu3*fJeAyw%=G{!LH93z{dn zJUesp>Y^deZ7!h+Lyu7WSsC8ml!>$To?@)n>-sUnFCco)Ou8>Wi0lZ}p^aG@Y{98G z)SqPt`#!xQvaiOmCDBgo;mG@t)+=Mvx$1xTzF-Q-`y%g?lED+jMy|Ix80>mgpx0st z9IZc39=`cO*2n(Bd2>rJV?-{FZMI_%GfhcvMLjv?nFi8|p`apt94!vK!7q0T@qyAp zR@kjjzt$=Vld?*2)@28Np7%GRpqEcdPUd5V&TSkz$I$bx85UH_!vvAJsDH|YNcANU zlXW`8s4x@5*L!308aEm>s|2E=mvD3R6zLv$0#Y51xhohpQ-pT+blF91SANB% z3@-OjJh^^s22n2Y!rdJ^NT>vbfrAxrf3F&SSU8TQO}LKn4vVmHvmvwvJ?2eE2t3j& z&&g2ZGIX$7f$N0ltf6HgRsT8)mKlhW?xla}4KFt)v1<`&o2bEFZO+7BTPazzCYnF( zD>d7)c|YW7-XoHO_i$XrrH1&o`gmS?E3ug(!5-Xxg*TFBurYnHj8{}-Qkj?7c5Q*l z{M?oPv>Z>>_->lv?*|DZ6(J=e5+}^e5V|cHVDBahS3jy!jRP{6n5v0gp|jayi%XO* zwFc`RW&C%k6#c$6(j2?>tnXG4PPZ*3LpKgmZJ`%B%iWQg&0TIVV3Bwp< ztjFD|T*vi&eFBGaDEih4++%^~t{FI*^(JPpvrF&MUd|P=W5U<2)Ux zcuDpvKH-lP*YkfLi?FAsw3*Y(R~Qc`U@+?~Ii=$vum@sboG-m(7_t~j9PUzW844q4thb?CF4^~5K zMHqZLuMRt%ZP;_&R(vCJi8SP_ViG$hu`_ow&`A52Fq3>qv9gm-5S&I!xn3|yyp2+_ z5~x!ZM&5oxdg4YrZ>1avUz~lh^1ljB**_K)CKiyFi_cS&51*;SpMLn=IvK9ikHWw| zhsfpkW0}Cytwaf_}A)rn-WoG4HN1#QQgpquQ?_@uUMb4(37B`LDcFO$Ckf zO(U8kqtJJDGPmTVGMS|w%VqzT=EUy{Ow#GonR%uNJ!C4%_dSwlW9)@pa@ce@=Qy9j z+DP6kLmr&X)tS>)31&7a8uzzU(9p|UAWSildf!>jhE&ARZlHwpbz88>#krssavsM| zD22Taepq8y0JmS)ab6xP*vsIFY|-Pt{^k*m zI`)%$IzAWe_WFbT$q8^;=pzKLFl5G&0=waYCy~1yk2mtgm`cV+Qq`P9>rem0>Xvsj zwvw=Dv&o=nKON?-I?gs0h2oFqT=)?G8;?fa#4pKR^v3!TY}3^&TD)s46zj;shx{b+ zpTiWvKYmH@mKwsDyj-p~=n=|IRff|?0wMWLB3+xVEA#~y;?AhYSn%x`nKf0KbfoO# zR#lpV)_xoE{8JZ}eGszt5tD*-fCi_B4?gdyQm{naS6*xC%MZHy9)Tj{bf03okyN z23lv{S^HhA!YgyAY4)pCmo%qXzyKFoR@!5@T)B*GB^KWOqe@lofl@KvllYa;}PudLp4;=5`n*i zjhwcPaK4pvWOA0Z5SCOy*^*Lv)o3l8{5S*d{LhGA*?$#WwUJcT{NeN(Uy)s{(o{5D zi_UYW#A?oM_^D8QV= z>%>LxCh-@YN8{YD)1^^&$dQXb$(3#9T$AQEY!iD*9`3(qotqmAl7AZMoNe_ueX$iJ zk1FKnddy@B9Y^TxWi~d2x0N_jZcX1VlYu0mGo+dJ3(7w2=Q1~LMs}AG6^m>>eegL= zDse;kGzVrk?;%w#mLWH1>97my=VNB67$f2d0t-?S)JC)jjzu>z7=UPWb_lob3&Ck! zlB}c53SZVMv($uz7V~g zt;K?4bsu&Vr@_vL|M(I1n^ zWc`~Omyf~<1ylGu{yw?-QU*%~V>q?8dzkq?h{b6iWb1WW@SkZe*`<1iUMoHYT5gi) zB0WUKX6KO&+qXkj`E9z3yk}KU%sTE`-^`hs{g3VVvp&q*Wz5}*={*TCyKS=(?&*vW=J4PN{xCeIIj~^iNBkQlM`b(i;HXvOg!9g6av=IER+rr8 zM*1}nW#Fr6{%o*`i@h1MN{;hGz*mrP1vRxxwxXw3rmyrh=Zmmlm(XHmQ!Z< zOK{p3JI;o*8PYadXDnIMUnN+=M0gpEwN&x)7}hmegeeO1?P}Fx9GCARd_THL`M(0p zv$jY15?!*Xg(qTvO6mD&Dro6BjkKAEp|64>iv5ZrV@yU0z5OM)JI|Rf`zSaB$14hs zz5?!Ph;2M^siBO(Gl))Ad$i z*u3F1JyynWH=Rr+OmA5?4nM<+O>>xi*+&eEOT|CSKJX>apOD{{ zX{cbM2ui-qka^-g8GTcdB>vGvm@$!f?62iF)Ewmk?j&KhOab?!HD_xwbz?yVz=YbJ0tIXZCh-X{Kv>tp_J@hMU=PXsM5RN{!^ z^|<=(TwML5pRBWV*5JB%tPCH)gEm zVC5_?(4U=z8CxUyJByw*#AIpGD*Ns5D^w5k6E?$&3qzzRatmD}sti8!qIk8|Q7|Or zfkWq?CeowKX!v*)V)JV}Pp3bpE9TEc3H3q2M^OoyX0gm*=0r5hm4k66XYg?|ppxLD z3%k|MwHwrP%fB3;fn9ZY%6UDRG&P@`HWP*WmNSX0+z;w;Xd>_0oeoVe6d+jdF&acU zVRU~Vb&_qRO~SsLn`MKorX$JwYp(2_jSJecyasr&mv zTF|)=<7)x@-L+xMvLb$$dleTUX@y76KjTl2U4v8qSA_5Q7v!>rBld~LqMqgx&NqCB zU#))_=Ol;W3kOk~Rre23vq(LTv9k!RW zl9vmYgR9sT?)KXUBu_VyhPR$Xn~ooJxv)FZd3lc<9$SEM4ub1S=wzMA-%L#Qlu*+T zU(xK!2RwSy9nXDxg4aLK7kHcxY37=kB_Xhj*phjRb~DE-Q{edJ>vWZUEAm16*h^7!+${LKErg!S z7WKW{`ipa^&a!EA&1e}&6Sy*J5j^*HwLVwI&0&6hH^{WFZ^@{WB{)yl7JFB#LvqhM z`r~*nH*Ea6;rkp7)#c=vRoXio?Jkb?4_@Q9Zaq*RmxKj!BUzWgrb}9? z2wwCi##~&$u6;Uzk8c#=^^yX5;rxBF`qhPctqopqRq8u;HbVo~ueT)9^QPjq`sc)G zxhI4;WD)6-Az^21fWfZ0>|^a+w)1N_9e76G%+58yPgj+^D+Pv^cm$ct~kMCR)akQ_O4{3@bLY>X6{m&tB!(R|=a zm4Z-NX(m>6t)xl*QW*Ml2`aW2K_A<~@+^z!q?_Gbo6>0dbkbv5Fkpzz^(h#*tCMWl zuMfTg2i0(^HVwL^iSa*;;Krt2JXZgcTpB%sWgKoKCZjilq-GpanqUDWNx= ztg&~C0X`qL0hg0^=+i)dYl}_VEZXomhFbX$w;W2lT7Bw|2<-Nq(?YP5gz(pcbRnb6 zMp&X-)zy6RBQxj32tEC0B>Yq+8M*Qt=_xwy&ZUFd=P;x76TbG(#LMoDoZ8DUoWG`o&)BC;rAD-2Y_)Syb)^5bU z2w?x@{CWE`7Gy}>j$9CW9QQKr@QD+uiN2&Sx?D-Z+fNU(O?x&_O%L0K-_kDF5|~db zj#^;)@q0vhwJ}pQI>8-Z9z!aUv&p3SS#%k6bLR`^p;q;Q`UUx4IR*71oH@c2ymn0n zm(yN2`H4PsRgEAma@~~o?xPbtH?rl)YUIYPRE%-{Otz%#V+ZFgfP40Ygtc=dc!vsn z(8|Kdra53==ScMD3tj4UTUon;3VW$94V(G}7Qm-A-pRs;Zk%AkJw0N`sARfu&lpKP zJuN^S6>!Sk1UByEJHA8eB{3UjB&T>A>Cq79^TO^U8^@*vf6FacL3BRjr;J2tR4pV>VT@1A+bJTNR=KRrN84& zatf;pXu9)?roFXTx2PX3*G5?Q>~o>|PE3DyrI;Kpu0s?*Rx(*8UpBR0;(6)MC0rl~q`Kt>&B z=fuF+$KPPojnS+#!;x3~o<a^uYtf=~A)l_%5gW&@sEc(c!ZPSQ}jCR;nK0JykWB zdeH$E{`eMk`CLUsbR5{?8SM>Aj~sxbCblHRY&Qg_wQ=8N)!}I477YKs1&%ySz`)@) ztXrpnJz8^Nf#CXnYW#?MdCh{eZjA?@FZo1oy%r1q8Ht0XugO0l>s6O)g;y(Y@i`Or zqotjK&AzTt=(tuLOe&t>e-jk2QzRJV+q>zx%U*nYkl+gZFcumv{G{oa#;=tyv+k>2 zLsWO1r_-jU<772wa`2-Ob^l(Bsb;~ro->B;)!D>q<3ecqWde?EUewWT1F8zY`GJcE zX;0rR(yx~UPNT-r;BB4s(YXThEFuNpcYBkzM+Ic+EGO(^#;EI|%UnK9WA9C7lc{;J zhtf*5y}hjgWX>CtLOmA5DQm4=-W4uA-LJ@z~{w7wx^uwQdmtgq&M5v4jhKUB<)Ke)6BjuJ-IZb== zTCtX>d>gcWf2Rr0l|DfKakH4M>~~^XW=K!`aRGg6D>&r$fTZ+KhKWn9iR`%DbU^td zsbwpv%awS1>=%Z!ONL3F{!MyT`Y@!HPNBDIg0XmlmoTT>3&lorL9&RC%+29$Dn^mnDlPmk?K1xJLU%YfN)1}B$AIBKcUZJZ6Mt-f&3$h% zV`{ZYtljn(CMHNQ*)DxzydWOVrN5zR9m?4Mvy^7}dgJv&^T9>MfW28+jpoB^N!Gk# zn7nWf?0;QNPG;=Emt-t7PAuk@6`B&Uo*!u052);$`+EEv>6TGcKGvOUWnqVF;@8Tc|1*(n?*-Q z&j)%-0<){{ke@}PuDYPRSYtJK2dsnQ`UHHkt(!|pisp`C5_?#vV@;m>1Jq zNkrINyfe!bV@G9i0ihewR5po|JS_<)gm*?+^kllrXeP9DiL%Khoiy%OI<+=5q>Teh z$T@{6#BRuwR<}9fxKT^MMCv}#c^}AE9$dmNlBmF)9%UHwZYq-~YlFro(E?X-9!P~3 zb2pv7)2n!iEE_Y8Xn&uB$G@K;O+i;l@P}(;<`f6I>d-MNdKbaM=Lb}sxyxVY>hRhh zPb^}a`KL9iv`TI>Kqx-fE=qGc76#kRi8(RCOrv_HSx%i@`l^+?q_f_-r8T z(6uFpC!E9DoB|T!Uqt@hJAyX5{*it8*TKBY6}LvILC4KnlB$=4{s$M(=*JdtK|Y#} z-#s4hJ-CU7%@pvleixbdrWkh(x!}2R$#^SL1S%ehKro)*d-Gj{%yct(v|qS`?-U&H z0vw_0mMuN4{)Uv?@r3(1elS|(9D(oJr1_o|?Wq@oXR;N1`~NaY(W)nqZYFR;LanI( z=t>&&tPLNkPG?V3?THCK=GG3rBhGI3anSZI#;UxauB`&Uuk;z}>0jfN){lle|3YDf zXD+xr(4hws8cF_nM|=~#1>eL-f$8|YFl%-N&OObj^DIkv>t={=HI7n)!dmpTjlqPA zrrga9d+9*!M3~d>0T)ikao-<`gTn3!aBq$;20l=P-$BK6PntW3s9&U?tj=OZN!%jmSM*PvHMvo$BDl<7l7(@L$lXy^vTJ{z=0h zEJUC1UF_T47HU?q8lw(1Q_HG3*ln|!s`!ip7c0y951Wq?6P5eivB%#Up7|w$>nK}T zr5XwC7Pe&WpG*?u-$7Mc_4pZiljw|rRIdJ{9v-{)k2iTI4{FZtuypl0Qr3T!&uY^} z&omE|{{(cr;|o6Xc|J{>?@VNVwGgkJH-$U6&>cP|xGSZ$(Z`?M;qzlDJf+*p&uHEP zPc6#GudAMP^ymZp496?n_szlVom)3|wcUo6jeUeRk;zzbxR(SBIicjgam2IwFZS&} z%Q9&OuC1#^gCd?@sp}*YQf~9n|0Z%f7Fghn8x!fW=U2&^E*-Y7(2}jcGY7Z36q7mw zQCt+T7(`#hWAF%5Osp6K^7>JvNV5d5EE$3K&xY}pmM-uoX)mW%#L?7+8}UF}9sTrY z1u>k}O=^Cf#3JcVe$9W&Nkok{edN%KMC2$joh1*iXe}4NVmGc2T35en-8cN-g**(? zeNDr=_oG{%BDy^`5x8GDT+7#7dS%d=XfB_H+ip+7XQO(E$xKJksJslH`_|L{)HH}< z?lgXb?LW@DK!)xY&a0jkH~4YR+qos&W)i&ZD5l@2CCBB0=>(&b+}(LoFhut`r@m<= z1UxduokG`Y7c6EO1C8jvse(VX^*K3|Z4Bc?&+yK>nxTKTBU!)W0@X6oV`gW%s9t~% zdN)+i10STQ&dyfy@PZT@ktR3@?0hlnc>>Wb;Q5_qxvGVMPPLdHvBZsB;aYL11Ll$$W?&=stoWWpL(DZ$924cN3GfR?&>L;Z;+dZxl0 zFB}ri1-m5ZlYJ`;5Kk;oE5G4pIY{(Y4VcgDSMC|YKA~_4XJ;{ex@iw4||gSIlPo@77zQ5jX+X zNKA&WKSNmWyuaMZ(8Hjf6V)(THXlFGa5{Oc0Sv6?$g*V(@GDRm)N!Fw*^?^<@j*dWNfJySOF&!QS)sr%*a~AjZX?i$@BH3)5VDL zo)0J+dl0DcD)J)F7=4y%V8nnk>>v3T7SlM=Und7+u8p9@PrMk_Ujn_h*07P+q1D@F zvC|2cQQXZL{hLRCRe&#Cvaq0kfT z$DNAO@MwJ*u{QCgIVgu6Khx>W!FsZ&|2bLL-9jC%&W_tH!`AkC z{(G1+^q0k=`Lf%%QkbR6ZwZ0>+KuF(?gD|cI*I(X7y}WGk0_x?Lj*PgFZ!F7vJiL* zi%41XOA>xi0~7u}CEp(!v*v@=*l92xOH$MEXLA@RtRE0M*t;-9Ne>UTIN`1hyYY=t z4PARPi(9X7#OmFhvz+fz3Dj6#0q?U-pv3DnnSB2YvPD|tt)3**J1m16Pd=gf0uSnW zu|=3q<&&C+wX|T`n1vG ze^&|jna@-?x|ObYZHeB@pM~t!#D=0}=x87XA!a_nx+=L5&WE_V+5d>{pc-|aKSX>L zx5Mr$yI}DBfR*F|B{aOJ4xN7lFI>4Km>KB8`7jH%CVLE;$5+#8Q-Qf-b{Y$3lp+rJ zk=P({G__H~FMbxJcj;y8oV-Nqqa+d2SE;g8w_~_-`VhTV8UTJiFGt8|0n zNUCYJ6bkG53U?X+Sdg&BTVz=Crv)L7c>gipiXJHCoY3~9ZeOY9B zcLlkim_W9@K8L+y?qYg#A~ooqix>=lJ8RIi$`&s zn<|3NWHO|%hQ2tZiCS|!$xS^`e830M+~0qhzXvZc(-NS1wa}~El}3Ui3W>3aIO|fq zhIbdDLGa*6{h{QM}w3~94jV9W{wrX6w5$7 zb*G(wzvCm+{LdL1tJK*=cNsD#(U*)KPNqTYy6N8Ia**lrlb^>mkmIMXlk8l;C6JojJK6fH69^BUIvhj7@*cmPG z#WpU$H*IS8({wDfyctRBqzXvdWJwq!m%+QL<x$`hqpH^LOU>1=0N{2)0O z&&7l`e41xT67QKq$<3XlLM)Otr%!|R*i2)F&)}Xg4g3+RL5}>%CFa#0U>5h2Z27_H zLmv&|(CZ3+4kx3v=yuNkLn~>juOb--lQ2DLCR-r(2KCX8C><0B?EsFfwmeMamcJx^ zU%HWwH)5FqZ}FDqa&E?V4$rHWK~K0dogZ_Q{-_sRcxeF`VDXYZ%aNeZrr)Epj&)RA_5waldrH=2 z#6x&%gwS71pb}dbLlp7BPy3GZ3RC`(@WJIk`~}y@=_vYQ-z%bRSVA^?%G=x)iR2Pz z9wgSEKGS)f4ui?5{Dr{xsQNRPrA!N>J1~8YL zzpKE}{v|PB_6?uj+2X*XiTpL8aFZ-4q$>Zujdr7P)w2w) zyW|uf(p7*{wB4}bxDOjxBxh#oz-O50!q(WIywd$asubso#tvuD z=Xp0#Klg`he<=lvkB5+ixhZU=Za3Oi7~|ZA-PoT$fo*m?%l;Ez z4Sy`I@S8UzVrPvr=4>Aazm7czZ}0mQ%l+BPWzOiVo=u%oimB5LD`Gsjgm~3*@cUdD zakVJGYS(EXxoZje(Pj^;TwR5k*bXw#5XEKQK1rgr3j}|!8YpkOPtHHEfXO|txylts zDX6-ErTj=%z3>#dFvcHF2=jlr)|ogqRRNSis!g7Tj032h_fmG=JSdl7eV))COnwL7>!kQa+!CQ^ih`hkmJ`K9UOdj)$w$Nn7cYMmGWd71k;eY$*Gd$Bf z08^T>P$_pE^Gr*iF7Y{3YMm!ZUYJMZOm)bVUNP(vor+;;^{6zs77uRxL?hRhQ=3L* zA}{ig+;EWqv8|q1Vb=mGd)iQ^z6TGvsF=? z@H5`x1+5)qw~_~F?>ENa2S=&9_)J{)8fn)cIrCd zpmmkb4|D+QyOZ$3;*HcZrUUbR4^Z)@8CRr))+8xteeq zpG@h(1N(F!;ZG|THATU5=?Z?O!xe!y`X1jloaEf9`tY^6Gprk!Nyf-Zunk^gSbTvi zR5a)c`;+H{DIO=**E*3T(_dakf8N5h}k2Fo(`PFh}GANojOr?>+3< zKCj2rbyP7%SbQLAv5EBUNpN>UqzEYM)DH+ z%5+U#COp{;7;x<;o|-L=3$&k*+{Xv#=wpdgAvc9u7yRRWq_5B|1q~SbsR;+tDGAo$Q%y{J3CXW{)DKk6@e6eHJBiF6;CdU#i4QquCH#GQ#m*peojk7 z?KO$?l;jiQEOc9UCjPX}^4v#!=YOLe3W2n3u@x!(lMA6`qe#g7r@Xk}R7#b1#Rw-q z)L2kRbN*G((!+u1Bdtrltro%!fzNr(Dh^}TTVcWIV{q?VA&sot&EE+LhLGpip-?jc z6hBUcO~skSZSGjk{7D(~pSw(!fjO+mZl~%AbJ;;*c4Bd43ib*du(gee^h0$uXL@*m zhHKapC67|#b6x?oKgPldzj9~;|0g6C$G;!bpuIVRh`U)awn>_=(5@$WvmW4GO;4(S)e;(o zSyPhG8?95gfChJf;Ez}IY-zU8jf=qwA7yrVK^(qv55vB`{jl2H5Iwedv+8zdxLmA; z!&K;h{<}b?o%%tN);*xmXN5xqkLh-;G-;^`uF5${27Oha z`B-j4h?*YV_;MqpzVaYfSQYF`iAA;90i6AYOR(&ZJU;$XM%G`7fKB(daK1uD`OJ+j zSYN#azpimX0G*lQM9WN(4ZCmulk_f|Y^mdC!9?V;~1Z=?%^)XG&8Z=0MfC5uowV67Iqxx+?rWuygJ;N>SQ~b8iz!K7Uw~3|yV;n%HRy6)n++$Il5i0(x^-GJ zSER;pUr-&oh6htihjnbnjs(=U%%jTMvD|om9o|3S1-iBI*!}zhzbiEa{hz$TzpsZ# zskfHRvUEx}CB}|ERK~@Vgm*IE zj=LJi1XR77sEp9xuvLrkDXW)|cuY`{9MR@Y+K2C66hk7kt^!eZa z={_-!Lo6%S#K%hlKyCE!2(Hx_2e{fCN2*Fcko3V2s- zMftC4ERVm3>z?}J;Vv5@F=WYkd_N2ey4KL8k$_*i^RQ^p4xECnkn&4yWK7{Dq7l4= z)(d{@4Rhi#^oSyp<6>yUjYZTTp@=K{@c`0aOr7+V3K00~2Atd}{7JT3X6 z2}8|eSlp&Sy5^@iYn-s1?d|NPj(ZM*rF|CtIavWNt(Z%vx7XlwAx}jX7{UC!96qr1 zIXe31kZgruT)TKbo4e5o*EGvRdUZYd$Xj4!^axV^={^_u+aKDFoP%m*6ZqFDgT_fu zaf|suuGxa{#fi48>1--)3hJe8dpa=OXd&zT*GLzQZX|m?h{MgJTflPED7I^k1a5au zB;_;5vD!p$>>QXy%66BL9piJj+n-Ovw^4fJ{Y(e&Y}$aCoBL>=<_`MJVk5OR>%@P1 zb-?e=bD|@7+iOkU(^R3SGi=4M^_DzJuTsR4waZB?bD~xYCqm@jHma#N7K)-QXp*fy znhCpvH#SQ6y2^*__+m@g6Py{svBfP3f8I8p92%D zy+S77LrZlwym&fxF7}2d>m;&JyONkF^m47$lFaJ(M3yn<5%qo+z$`D0Wzs8a$=uFs z^kDlT`0jm;Y~Zt~S?G6OtU#4ae}52tg&o0dn`oTTVa&LOZX9e%p%Ie8d(|)j3O&BT zY|TpC=rm0KyXsI5!j;vGPb z;8IlTY{Y)!=Gpx%n`xcnHjHTx2D7azxHBhik$a93;1rflRd?mU%j^j3ddFyF+bnuF z^g1@Z8g2a|IgGA8w3fuz_F}<}EK)et5^km(CqrBXIkEpRy)*HLkcpSYi(5GuzB@?j zOdH_cRZsZ%@iRY3Bu(fs&Y%DzLUNn}$d)=4y@0@$;jvB)vFH;OB+2@f&WT zMAS*0j(_I1)kDnrI*DoFz({iNwR|JTp6o)>>>-7hU?t@}74G&Ji&3-ZXa9}NJehOfA?Y|CR3Ui5uM=(6>oln)w+F>&t zrO9S{$j#+4aChe@`yu=|6nVEn&R_sYuf2!QTjiMeg)LMoOP>EeSAmk~TKeNeBC+0l zh?yRnh$aRpRxh4qlk-*v+=nk!M7^vKej6^}I#p7uCJ6_VYk%zVwTC0l*}NDJejczI zQ8r<{`gh{f99;~2=moLsqlx;DQoMDiiks?_j|*v^4h2{pLCfhimaf{jx~=_FB2R{Lo? zed`*H%|~TOGVcId_p6c-N#IXnCL7ax8bTb`@CNOacjDW4abmaf0W*D?IV5`lF6cDn z${Nq&?w#&Tm8LYdzX{{EluF<(UT&v+(#Q#j8V`4GbbpA&NGlS{lY+p1Blp(H6mLjTtt;ObI3pBZufR_t6 zywX%jOAlGV8r50UA|nPLfEJaTP|Vcq=6TgyCXTAaBpA8$P7(4zi0v0Z%{_TF@$hA&2lc-1&4cGv*P zVf{4Vj5&I97x6*zSa!eFf8?)02)rDOX44H1VQT+0a&XO9tI~HWbaR>le151zr#U3y zho^wO6EtClejQUUQO!h`uEFIwA{f+h480^~z}73_7?$!1YxbUo*UR?d7==t+ErUEa z>m)YT%){9FLK?ZP13&zfr!hl9%q)Ii=UViTjT=}5v!r{_xOF^NF!o1PWR5i?sGQ*4 z_A98I#db`xZA7E?2zuj9Db_orP=}Wqq)mSl-dbKlgHE~-GkF0WwAn>M7VpFzI{f~) zZwmj;nFa@k?BLzq1Q6Q%>ZSH@L(Y>5JR zR~UlJh%^15t^j?VOSt@}5Ad%(&+;z}1{rTJ=>7SG3ME9~_ewoZVdEMy`&BBa&y6H2 zre7fU&jvx|^;VL1cn_EiEGE*L4C|RO2NkDC;V4s$<%`s~tuOr-K&diSIaay zW3Y+6ry~r?Zxe~Qe#Rm!=S{ss#DoENtCTN+c5}(sNSVNR24Z{W*M_VI+bXv6G%;;Uy1t zfM27Wm?+PAK03vfK>{4RG9Zl?MSgA zsi!WGc%fq$YWIt^cV3F$3*51oW^`1?%07T#E_ki@4goi_kL+KUCQCvH=E$z!!zvn5=lJrZ!J98f11u| zkH%zLKsVhD!__K_u+|`j_FMnNi#f-+9G<<3kENKBkTkA-b2AxDOeRMYqe0povGVk3 z-1;q<&X^UC`RYZuieFFJj@gCnLJOH>J!P)OB1ArSC}_BjYQ)NkQWV` zA@Te}tE7oDkkYD6Cz ztPRmi@SA=)co#k;M8e|7t#Ei6?`ZQ_M$T5`VD(KGT;2PPN-h0KO|?JL;M#8d7_}3G zBQKIxy(_rd_$-udQKeHORZ!M-8EfS;j*9QeAzJa?T*s~qDp{oq-=0|DQRR!U-KY+X z8Z`OATM#YCeTxnQ_qYX`p7e2xe5IAkO4!gKNl48J`pxM9aW4;Mo-=wNl)0ZO{WlJs zrzO&n1SL*KR+$9J9>Kvr8?OJ}Y?%3E4HPmS+=#t9uKRe5T_fWG@5J|ml=Vgwjg7~p z)p=yp|1j}cy^j5OeJ3VYq=3&bfywF-Jc~h+yR!TRit0_rj2jE!@4Fqe^<)?Iojie) zwLQS9#8+d?fgUQVFALAK*J9S|OpJUXhB3mjg6Ka{u;TuIV9N7q%i70q8;T;RX~S`* zE?bKD3rDl7Cb!UIeYa8U(=U3`<~_Y*Z_1S)wZEM?cweF%RJpVe$E8Y~i|z$3gXf4C z{t1GQ&*b5D#33}ew3v#;$V1NFHqcFK!IzUyF*#R$(}!MJm@}^hBYGcU;)i7VGeHt0 z3!dVXc5`Cj0(9xbc1H5mJ5)4#Ow*$m;K}!Ze*+wds%tyrlVwfb)w4@4rJ{h3S*7JB73@Vm7&3`~a5O zUBu5>n@H_$-Ku9LUy0EeA7WbYn;8F2q%U6TVYB#G_Kc}4&uUr%YWdw{ZjUi)GzXEA zCwu6w|I}gMYCm?9`3;a{CR;XJs+SsF^u^R37%lmvK%V&5ZZd-Hb1&BQ*Xc=qSo5*^egpa%mR}}WZS3OvkFE?S5xIf5nkd4r8-CG?BeAaBz~??n!i6rPuUX)}d6M?+}0dmy>4V~SD-tU<`=A$8SK#yu}$xQ_|? z#O?>*Wsx=FhEEMM&rTUqvxO2gu22;Az2#6hcr&+a!zrqM%@UgV_4S55N$~RcYwAOm zv8zj5>A)99sxeW8yJdQXWLEH8%@s0ySBT78;=O`%jyTMB_~v3)W*GCZ=MbsV^@s4e zp?IsN7dIEa!t;Hp*m}AHhf>}%PSIAV;C%r0_fH_@xPa)$xT5D~0ez_IOgoC6W6Eqh z9P^(K=R9pW=f}If#4DQUaNGo3w>AvE@%*KcgJRsO->X3P_#5)Y&sQ*%pUtS}h{BM` zdm3zU83Xw2M`JbL{ZZM%WzNaRuNcE6&l$k+&H8Lbh8r~5g_D@{K=y`XJSp7!2p_cs zz>ypstM>jJ@JioBm5W?3J~$WnJtOq+UFw0OPLRK0z%5#|6$>`a#x7~WfdlS;o(Qg5stG-JeFX)4H)gIWu zoWTPzc{qAa8Qh;r^K*`y;Bw$3G}_8h-HWC${c#-qaYP@6FA1~j)&J7=;VJmgHyHm- zkRU`sfwS1&NjwS?+4^IX*$I)#P(M!uwR!K*wy=BTae_63Z2p6l2Yg9=yByxLe^9B( z@6jGbB;X~*Aa3h&bEvY@rMesR8RKu}cy05yss%1Z7F5# zALrolX_{nuVFJuO=!s`{AEG8&3MleWnw`D+ER1;NP`?yMniaDIWaIr2E?!52*#D?v z!3}cY-T`>fcZP0wC=VyUZ=pKd6F_F7F3k5VXY}uB;=OWrver`?Zr!;}=gBOhW!I{x zuZ9}eWG@VR+nqq4jOT1Cr8%(;H*v?20@^!mGLRjmaI$bWnR@Isk&vmN6Z@w_cn8l3 z81&>6)ibU9VrHSS*dTGc_`)(k#s-#5zs>lp72D|MeB0S@ri>RL?@ z)P(U)ylFz*g$xtg&O0h@>~e!(t1s|A#|qe|o#hAf^1=HdM^eR z+MeaoU*AE$GxyoV*O%${we#TQ>>R>&t!A|(f3m^ux%5MaA~dZJgJYzCj=U?TZ?_ns zQL`o&=8*$$w+u4>X|Lg~D>dVetZ=aYmWrFlrO~;(e_*GlIeoOHij+iML0Q?;oK&_K zm3@&4{m)(S?7r!EUDO|?3?_2tpB{t!u^++m&sc$aNgTMxf^b1XerL_SCEgkNu^ zKy#WcpL?((QK}PhMHq`qo<>#L46PwaKh4NGXB7-BSHbM1r_nOqm`x2YV*3B^>ytKZ z;`-hjqWVYqGuW`I**4J(BfEyH6FJ7*^r!@HeN{YT<4De&?;%Im*we3<<;nQLAUc{5 zhO+!U%bA#i#=l>5?mcB18Jj|m2S>wCBT3ldsS4*7-$LUKM{=dq6du(tCmVhR@Y!8E zVmWI*IN4mnV^-^M#Y!dIHSH5x^v7enVjK2-x=CL6nv$^DCfp3|EKXv*3OBq;9lO%f zK{{3u%Jph-me@w1BD_hY!0n;~djs>(=Q--waxRGG!{*39kkvmo6^9P&5%;~s%H z*Lk#(IXb5hS6La-Xz2o?&B`*5_bHMZzlCtIqJ+wLDL~$xmqa)0Imv!CofEg}CM)h- z;h76-$gFiMVQ$ShUUB)3t~<6Iwe4FN8S@-&yDQIYc%gwEHa0{v^)i;c)`J($;WRVq z6}eySZPoo(7maT5e3VDyv7O`SsZKwP;7oqzOQ@C#)11kJ7s+V{3p@?!3p$67LYg0FrQ30PS`T5;DMcEa^^+v84`Nzcw?OKQxmX_h3@c}R!I>QtFTT4^ zc4f;!>S-NL(h_m0(@3FsFPyiq7x3Kvp*=Fsr>f@JoXV zaC`x&Q}lstkH^3Y?gA#Bbd%j|--X8JFrlZZXn!khW}v|VE*6TdQrS{W&m zy0yA^dgV)U-1Z6Y(&Dov(aYh{xor0Gb`>tErI7}X6XK>PS7Su2IA=Psfh;po!H=i3 zxO;Kguu(J#p6F|`%_UyMRVWAVPglapRoYy(>=+!$NhI0p_QRA8o+D7P4C7|5K*xb_ zcw3x7UQ1j>4YAAkO>Qxh`$iXUt6#HvP^1XM`@A^W;7GdA<0h$lZ;t26HMqBd1z0~l zyQ<0A5cNJ@!pU1}aPxtgpkx|F9ESNEMeHj&c=H?0ob5?kI%IhMdmi)n#YR{sO4vEp zk#zmwc0B#{As+rYfIbTGC|GqHZbn^&t#PJUBceqmPL$A@3a0dWMlVLW1(2zqHsjQP z>X2&E1RA+LY(sH4Ixcl3;-@E)cc<#BOdWIaSg<9hea(jxtb30}F$y?Y=`61JnN6&3 zPXPhH-rRoKko-v9&*@CcLg6+6Z)_>qTwuf4!v-xM&g=e?Hq447ZOfgtg^1~-b@ zLwuI{4;S1)ur{LqR+1$YX3jFO~NM2q0gZ>gcyCC)$Obqej zn$xaumFq&-t@`_Dx%)=u(G7E$+b9pY8tH71`z3t$+W>|elSucW-L$;d1#9MokU26V z^xroNoVs@^TBxf->5~)ikE>%J>{!p1X%v(69}mf9-yD*elek=jIKVWtz-=U0F*~Hlz_<_u2U3 z`!VR5pGFK|#EJ}#<@W!)z>e|$OO5>tFfl}u6B1g4Cn}<`seC*a>g>riOA6z~(nb#QwoyY^(}s?ljO2${2PXc3S2cq-3Bs3c60MR5-l;#TI&^r#Ltazq-Peaw@fnY3mRVHWE>u}7xW}17p zi%!7HT-dZW3|y|l?O!9!ee#-%QKb%K|5ZcSy>B}f^o-Dvz64A(ZbRFeVhnhlfgigi z@pZH`ZvNPZ>c`IESw~0Cm**f{nU_gfQHERZJcsLCwwl%M_*i8)X9}v{j-tzLtN7<% zAR6aR#C?YTR^e8%T<=^lE~Mc;O)PA}aYA`)&D3}{e*PI4e`APv)i!}6?1J;#=Rmi@ z2vL$S1~WYYiGO(&v==QUCX$h`%l;_!tUeFP3v{q#R2nw^P=}!}8?biF9L(>tA;=kl z(zuPx-bN+s18I!gTOCcyD07peDs-ocS&g7iWNO*-@l z#hW&w`MZhSMeAnh&$tH%EpuR5;S_XfdISqBH&dBCWAKiy7#ykmO2(>wAk+D7d&bxm zcx#WeAVz%*oL2gU77Mz`W_LaMF;R;SkFFudrM9xSpC6$qLNf){jW_X-bS|yCdd4bn z|8o-TriP2%VwlnLQmZdF!{N_1NkQbAv4WOs3~I#&!qf3f1Px7RXtt6(l`6Ex+|Qe^ z^Wxnqon~X0-Z%_H+k^#|Is)mRdLyFha0?FAtCP#)D0#9hls^~y3QWx`;-xN4OTv-7C(y9rc94{9E{M*ns;U9|LuJ`Tgs&VYD0HLERL?;aI_bLGGh3RRyY5Fnc#k zi=LOkP-&E4B*H|HsyYco)umuc_jyREoB?B062K&&4qNRH(M^**Kv|~?vimr<#dcZa!1*HlI(V9^j(rbCD-;FNrbN)+cZOY`v=yrcKjV(7B-*f3 zUr;;mfIy?SlnQ%9pl^5&zUjP;_s6fmkX3i^P{LO7P$iT06f~h@;RQUt=n8wUz=m7( zpbMo&{bUZ>z64tM%=aZ$2J>I{ zTb+`e1l_JBg3R36WRvL>^7w-;W_c!{`^Wj{<0Q;g6iIQLHG+}4rqjADL)5_W8ZdK7cm~(KlAV1g$K6~_2yPzuUzaWPqj~%E~%`2Sy*#dnojD?h=a~S{P45nXX z0bHEZ3ZEA3$1~zlgq55iP&1}b4LlQ-JY$bfn*v#r!eDmuDcE)hnc0gyNO{~5{8MU+ zmhtz9X+#%2a$5qg#gvk_M^+-e9Ebx)RKXjML%^JZh_rO{N&mKBfCWew@xgh z;^J#~YgD88TPgo||Q z`Sb3zD5HPW%G=-|k?=W3%D;^lT%M#Zh&D_Ct=0kfJ7xmk^-MiLQmE$2?=oYCin+$cKVdQ=BTxNT9ES)l+?F)=^wBdmy?21}Rt%Lnx z{S6jW-)I;6E%(_hbjlGvx)K&(*&cTHZ1U2|8V@bPt+ zLMmwWj2k#ZA83h!Ebbo6B^eu)nUWtOoKE*Ppj!-~easrBwMUUDQh!J{-wR;B>3HLb zv96#vv>1C96%&b4J+@Gg$GiC+;;WfX+}iF-q`S@xG}m!d_vu_38R|t>^6&4h(Uaj_ z5$`*z9FHz%!_mI(AWo?oM5`2eP>Z^NYPK!-Pe&QvoKd2s&2ro_{liSPb{C!bZ8o*l zmd1uHGf`JM5vNDmg81#(#P@YGeK~N8$cI~l&FC+Bxc3M~Wosj8{D;q;uEafxJHch! zL^3&b0bzM|_xOaTczlZ&hLre1siG7Y^FjsV-{sQ3ucvcri%-DK+j4?QylW|GODMg+ z*b1)2%)y+GeBLVG6r3zmNh_lRL7A;ouyh}G`z@f^9~4lqQU@dG)sebJR-Maqa@> z>KJ9lZ!ScYJF1`CFtM=Jns% zC3ixI!pCs-`qxy{(lo_EBW_|-i5b4dbcc~G+s3bzj~^2! z#?P;iIwuX3&|5*;;sdOPtkkWpSU6aA9}R=*;7h<(3Def^z2s*@60u+&(|n!R^ycv! zR;S`RN$MCSR6S(`x5G@}_!eE<=r@UbvaK33OI}trrY7KID{V5B&)B`r)?#?i3tBFm z#(YsaPaFSSM3-tIF7|#n^~^9My^Z0t2}6++5qko^V(;^;8_)nbEvL00JfW-RW#QE$~= z)5iGf{bBd~<2hJ=FOldHBaq@|!MBudHb=_|Kb(4jnkw^POwJw>c4iFoU$zP+(MX!V zP68JW-G;v89e7;E3_58zqhIb)E%)?nA1l2lmaBZv)ov2sHNyOHn`I@~XFLoC;2dzWvt|D^WC4<%5xr90W zIEs6(7{t0vJxwYle$Y0pMKEE(KN?@K5J&aBvGjpB{E_>P^)VW_x3h?cdOPu9Sho;lW6$+1ZtW!6(lqi1tY(&@;uREtpo4o)DeF_GxF$DC^ABCVOgaWUU??UNoG56zfXm8n+k7YMeHM@EP9S! zyzGg`KMqr=ckPVQ>ei~WHI=mX`8<5nJ4h}ZoC^zOwcu{S`l=a!gXq)>X|8_RWD??Z zjda<)BSG%h;QCko-K&ueWwW1>Blc4PL*7!G@z=5IQadTKj3d%oXK4Dl>zJaXPitqE zLV2qKbq`)ywQQ^xx-;c;%-@}?*JMAuuHR1Z#2%6K>|8-49CTsp<9A*dnqR?e+m?a6ieL z!eVxHrD)ZonkYu%pf`H6vec`{o|(RRALJ&!M4jJJ=*bi$14^S zi_$w}^MIOubPiOMXdBVIl1f}hhNPRSX6 zon!dl=h(*IqPSASiR@6dLnW?;8IhDHTfCd7Utm5?oVK5;yvT!NSAbq`_`uJ$`q=Ap zZZLvieK;Z=Po6KFMULs$Q=C5q^<4B}&1P%5dO|uoKcIk$pZmzzf2m}ebQQRn{@W+v0tw6UbF;j;+L@*-_K1Y zvQD=_d+IEry2cDVjtUD74m#j#z4Q1^&p_WSQDnps@_CNw2E!$^x6ue64;^HB zdv?-&&e5o7@`J85yhP{Eu4Z2Q9>A*xzpXTv`ZHU?u2u~eYohH3F>d16EqG3LH=#&is zovlsm!i@9eb-XB+o4z0?>c5eLNwFm1k2hMG9VCSOWLK7r2irrcn4OaZ!5zWyK}!SV zWbe=lTUodi@tPg*=|Sh-lQ_$vf;4)_(6n@V!J(3xJAT2_&_vVZ z+%PdC1zaaxAsr39H0SO$HtkItIrQy6rZHrURg>HyFg$mizLkqWS#~?L$;nV{VJW`L z_JN$wo&x>TebJ%V6E4>OWX}vIQvVc5Vlkthc-5zq`n3{tNvJ!-S@O=)f0m$jbTz3l zvICQzEGU+e=ML?Bg$Ekm5XnUuX#Z~m6VHr;%VKBQ^$JTtPMXhU7@cG*c>Y_&6cIM; z*jK7Nbd^oF5{B9(kJz%w{dC@xFuXoV6w}H=5>1@nP;Ld?-&m-Xd_$SxRajWa#a<=A9%jfb-FiV9lSj+h4Wqeh=^f4%AEK?Zd~1r zLT^^F>d{xJM41Tp=cg~7GP0E|diR@X$~?i#^9!iA^Jl!JcaqK$POo~pNCNH;)sYbA z?NBwa0)qA~g73u($%N>Sw@P4)Fr;H_*n6>pmZ%P(!ja#c~hy^nVf?Eb{(drirzNmtn? zvoAq>k_C8cyWniscpC6dgz;Y>0q0YPsEO(gblY4)r$qduJFlofwM8>EyCe)=U)<@} zrt7%=TsEB^Ta9`$;>`DPLhxr42>Zeci3{l(i8XI&n1X~5O$?KJYU=||A8H{^^(f*lW#a^4-}d^z_ja5a+ZA;ud)@fmE0~DZPcW*7HEqJ2%KIkH=(O@KKzMZ_d z^ouFY+RUs;NyGtV9hf8BMdJdZ@Y=g`^mgZ1uDDx_t1CT+3UN0`u38D%IAao-SKh}8 zZ9aw*d@kdvv}iiRw1L_CI0M#ZbEGmThWO5C#Q!Fa<-O}+q{Lwx^f?6jCn7mBtprOT^H{6lGO^t=;I z8h&n7;M$DacFAGq<<&Ul%NAN&%lBFqNiua8-C%pDG}p5(7mu)SNcx2vG*165t;+yb z+&v$_Z2&?p+{d_SYMg&Xx0Ueg-EI#j~^{=x819aq&?>3;$9Q zwBH6^jXy~eMGHv!%05!I$cTBm?;@1jxM8naD4Hm5;d5X0JZrQWi~rpM1kiSKJSVN%f!)8{sB^-TNpB-l?W{~hw(jIA({|3sg-ql05wH3{w_PQDWI!w2zme zQ+LWTQ|wbI-;_tET$&3JNEUS&(rc{+Qs z8K1Y9f?mbnna0_5Wb$8cZu%2n+~U6t#3ys4ZmbaYEk41w*#1>L&>*B`sFvgv@$S0r zN`e>Dan*z-W)nZB^>{yq(HNB?K5h%~@?K58Q(6k?=JDk4Rle(@CBWTn6FCR1queq3 z7eq1cF6(_N5E4CW(P*qA=jJ?CqGn<2YN0#ALjUt|a~2U55|WS*`oDh03o$}IyEkmv zx__gQug_i|BNH=Yi#aA1b4?v=Erh)Pmp{V)a{&|Ih+X3UywLxA{(l!>_5ZklImUC$ z9c<_RZ!aJZ{$qsfzS1)iCft6lB{*)f6gNlfHQhD%0DWi9=H7VAa{no7aa8UAx;9kf zgqq1mG-Y?p+j#P3^@(J`}3OxKaOsJ|87jf z(JNNye>suQrdq=vnOu6IOr3kP`x)*FkmDTqwO+kpDL3LJ#VI|RffoEMcZ_Yyg@U;D4D?Wt z=7u(f2=)nkK!)ENQvYHtL_F(&hBYg}dHi9Ru~USay73*lX&q$9%TREPXPN7U4?&ip z47k-ez-3oUa(}3TaM_~Jx^FR<*d3=~dq>H6hXH&iqJsr0QE*V+lsccwA`4I7BWb;s zm=_8>n=Xnh!>f!on~2Lky5UvHP#V8J05%=CkV^#&*3Bo|`Aevy{|11<^?EK#U7Ml>cE>IeH6~i~?!j zUImC)ew8g8P=Kb@>geDthn4qCnJxLxz(npRgf_Z?_FNBGBl`hzE<^!ma}w#1h4j*# zWw3h7ApGI^EExxPQjg^6^t-7WN7~+i)1Xd?I(qZ7QzH0TgR=2J3=A zD!)?}_f97yWlbYYC^<;Bs;sxP7VRd#Ei`ijZ#pZI~SCG+#z}9Rj_KF z1-sUI4|Bq$50n{D*lKit3RXiRCzGCEliGFi^Y>4c-DwPC9ZxJgkLwAVC{{amYWud zkn>I^R3+oAAarY?VDP$wpg?RESZ3G>-gXuVZYR8=y1IJcq;3X{ZdT;3PB}*M-{UXm z(`3xe1yx7dB~hP06MHg$DXii1ghm^Ahm?OFR>z*hGFxM!Jy?JuD(mpR8qeV3pV5y_ zRg&%Ix**(q67S6Dq}ppU(MzfXkDGa7fgk@~7@5Q#|2~%s{iP^4x?nsk%az8Wgt>Tk zr3PJeTS4GCJPs)+)E+B(p&=Zj!Id~^7%G3?&qwN=%`o^ zKl!<-#*8nd^UN!v-;qPN4$k9lK61w7Q;9VDcraGVt8hV+zryvgHE_|&P9SBY2@)a_ z0_jROL4cVmyd1s3%D?F#5q+af@Pak8L&k~R`11!^lV=D5UZ;VRurFwxRT5-u&xg$- z6~sesC0=oy4i6WU!Qot`&&K>`aC zTfuGTV1aP|ST5?^3HF@-ccQd-3)ULSkr~_xCQqyd^X=xq%b9ApMdmP*I%+@zUDNrQ z5(5EB6>#rY2Wx2DMieh~k&MGa#N?H%;DU4-u49(rWTTn9OJXK16cr#w)xbWTjp)*y zLnj_OfFHWfVeR$=9Oq}xQN`I@*0V9(OYSZWI-Q9H@s()(=RGuOCxLwaFDN*A19U<( z1^MYp`1dN`pL`)Ks9VRO&aGDV`rKJy=f-urD`87X_-vs@ws#NxR9_{P8faPy82st=`?(o!v zt$ltlH}4`m;WKe(U)kVMpUtrCUNOW-_(8tEs$f@BB_q9&XL?WDhTWMv$(tXhtgu!s zP8NSi!(WABOVDX(SrfsemJ5@YykpYfkGx?1>ahY#lLHtVGZwp+e4+m$l*zB`S&$@S z!s+c1R|n|`gwUc)m2ZjXeZv->29tyP61Jr;sy{X)TyTQTH;Zx+(2 zfi#Uvq0g2+rsg*kxxZRdF|5=W52&W$T>e=+(6kY%zAS~^?wxRON;|2sZ3L&Dk7Sbk zExO=DB&1#ShCQ#>(x(HbA!b|^hFFS0N4u9mb})!@oBSCAN~`fgk{NfkcRIIM*oj-n zXQ>v4pJb+#-G;e0{7F*%5Lo7Ygyrjw3TCf~ghw~R1wOYvf%x|f*xA?$@sp?FvhFXW z*mgFUrt)2dp=|t|n~h=3Cn2@6gNzN>LZi3&lj^V8guAE-+7V~zxj_XSYH-ILk1I%X zs2DkEAcHIJS_!UaUWJmQm#C_skLb79gYerFcwTW2Vt;hAzr_r(Jo*_lu4$*t_)O9- zo(Nsr5+UHzO3?d#9RV<>!?fyf=<8Nwej`z^ze68aHlg&a%QQkH zkH#1W(4hB{pis9AMx#_=)edvGlQy1szB$0O7MX!_!AkU9I*7Sxk<5|j1>hdrLd87? z@XzPRba3`^!I}Abg0JDXz|ZawyePJSJs*||HmF{tr`3{JGtmM1*QtPMx@L-5$7Qi| zdp0Bf+K@Y8lLhnQj)Kv(N6d}TK@#Gqi5+EwRDRlb*0Ib5KHjh9=ao{Z*EowzT&V!- zT;j;?Z3XbRClQWnj{)tRt8gkW9lpPNKs@+!x2?;Jq4M?sq?RYbDvx^jG=qbj`BkjT zg8=rXp*Q%S?u5$$A~a*~Bywu;U(&Oc&+8>yVTt!vSez&WUfHI&Cg}+L;#|q}%rdxi zQIcB~Kbu;6OJVBiKpc-AblH3#TyygzH!hC9ciJwH{(u8in)g=MU2H~W{T%o`|0yKy z&IAi?5tLZ8g3Y%#;Mq2oirDH<{ncj~i7FS;UbYE-3{4Sif1b}Ryz~@@{rJwji~>G+ zx}59ou0a2p(*!TBGAJ8WidR2rV8h}|e8IXMZthay?1z*&m)$QYbLJs*8a*LDpD1w& z3liwx_s`*>`C}0GRi)?6UeeU?Tr&L6fgYChhc^|xljYwla`=fgjbqF?*HKZJmTJzs zB7*3YaUIxkw1QsAoP@g`+6lyOFBCxCM1iG5CirP?6Rb^NAyBNC$f^B0;v#0Ditu#Rhtl97su` zKB1OS)7A~CLWf!TGcO@%(hoRS_noBXyTj>uBkX$TpRCS7I~c!}&)xX7H$#rj8-R&zl9&D(|T;w+#xnZo+FXXW+snUD&tuG`uW7kNbN(Y2!-|2rNs4 zB2W=H6sg0@omnt$XcH{|_Jqvq9HxHD5@5USS@fLv0qy5kpn}32Zpj3S@&0Bscx*mt z+xeA7blZdOAt6}%WC;GW9Rbm=iEy{Zn(C5A#Q$L#6_e0GryG%6{G=dmnWr(QbB%CL z+dVi%6>HA@dyrM-Vsp~kv=B8tGU=S*!}QU}R=|2+PTnz$E}4{GMcmb3mPdG{WM_b7 zkliT|ox;yP`X;h@-=twqku5p&Y62BFoJH%}ELyCZi1A`OP^o$wjq{AaXpuZ{F`WV< zt73T%A@8g^H4prCh+y|!4Z86A4{CnDlODepg3q7(4Ac%g3&H(`Y0lX>wh2(+>5 z>_<%LKg{)(JVqTcVQ#uxH64C3NUym+N9N2JPIypHpw(<4klAg80}lh?;YS65gpC7O zP7K41ui~k0ekL3~-%LYo_R*;?bOfqqL3pnr1Y36V&cAnBaC5^7uD!ySv)#&4g&Vg} zXOS^L702 zScxL!cKZ{zH$UNEZaTDU%){;z?PNifm|(wzzM#@ii@T#dg736ah)<0Nyp{S)mHZ@; z*%ihGOI6?xm2sHk{~B9y2*XP%VBe`FRsr-C`By9hwkp?%TEmMcg_)m#Y<83)L;_z%jTL7RvE6vMn32V7vfEx5tsWHZ?@;!3&I32_WYJ z9O<ygP=u4_x@L~$@HNX|!8>rjFq1sJwG^H`bc8*z&G4VOp1^GK z0fBIL7)%~(5A!Vej@!9nuufYJMSX+hhU62Fn=J=5QzoKIMg*+vvcm3rtz>(P82PWK z2D@HSOkrB!i0cQq75532%;*3~tA5&2*-OVB9VhS#7ZJQF_kkduE|<6 zynjQQF8gRj&bjbsedQeO=f4l9OgW8eW56_EA`vZ3V8?JbW6n74{g#5TeAx%$Ihg|FRG3t1e7`Xzn3# zu`+lm^c3{jyO6>_O}ti70vFakfVDx#L1oiPShwR0JeT3Riq+RMy-ufTaD+9vmD@(- ztna{sXFFkjLK%e?!xLHd z8s(GpJ*nLPJD7g-j6f@{KcT>WAK7nd&N#1`&Ezc}famw4VC}4}%(AmGOba=QigXaW zM0HW=+b5FOw+J@hn9S@}F$CqZN052@6iE$<$NdAhL38&KSYFx+32ZW)-?jr6OiiV~ zi-nnrzj06|&E;87Pi2y=mtk3P1RSsT;N3scPF5~h&Pu6_qpwE>Zrt2gLzYRh(q*UV z_tMp9^X)BGZ|%SU)St>}62c|%C-VbnJei*(a2ceL77+%eqkB2>mA?~OadAsNpT;9g<#&6ftYe6;` zF}5gv8Pe7Vu81RfEK&?cKQQ5M*`YXrUY;F7j<*Bb= z?Ym#l7E+8}9p}K!KNM7(r!+9F3 z=G}vF} z!mN472DW&m5Hq4#2NKJS`MXbT=LHF9z@*H4dT(nOZ~Jm3ilH88O`f8Rw+N1JWW-xDZ-aTaZf5t&2 zHReh44+z(ffXyK%Q19s*GDAp)4P$jts5Y3d?{|+b5B^3KxjDRdffw%D;g81Aessd8 zFnX-f0`|L~K;PIRvc#?!uk1Sszl5`?Uri^S+FORBjqk8^Sugf-{bR;H8;C@lFgB-j z@kAE>!l|R1$)Ma+Ty@hB4%-$%!RZOicD*1xhE_Pmzy!Ta*FwNZ9W33>_1%53g9}kC z)D8s0-JC6O&0mFxOHE;X)5_s+$6|af>B@#Z89|f(M$p*akt&Nd;XB5h+2jm(-{C3m z%Z72v|2z?7$EVc9R*69Sy)|^ud=tO^`+HP6Z-o0jELnYt`M72cmB4&F9FKL*$z8V2ZlNskSE>WS=yG}`p^4(?u;Lp<}3W0sT~*`ykW zySg6Z_Pgg$+A$ohY=^LZ#!DQym1x@k=D=J`~ z%mGmGLs)S+6e9g4;KlE&@X@jkc$cQZb;9+=E_6ea#D415P)!;Ki?O_UGTSS7fm{zR zqQ^>(VhPvX`S8GJ!alqPsU4hV<6?+juft&^;wDsoDa7b+n@}-qIaCksfWKlr@LX~e zOp}|;=zB&nw>f@wlTrn_=e~}3VR%UE@&S$A(yU~kUyU3bjw|9 zOq{)+U+*gqWdh+u;06y5dR(AI!a+30za0zod(nK@mwxY-WF1mWaE!~$ow65%)SsJC z=6yV7TC6~^i2`hN;vc$3G7!xw+Oac!3j5{nVpI{o4A-p7_}x-v)V$y~Zr7{B#Nl~N z$iXR$%y0?xl_!8F@qw*Zqxk2Ir@_q=1u#+PH0ev#eEosj;!6fU*aqg|8=M(=w<b# zd)D!L$2NlYeJy5+P8;|9>wv_A9H=bhX!IjPU~YW`T)F#g$UA8`U9+1$(mck`94?_A zvn=4~1ZSo==m~^GrGUXm0V{|WVYfs+x|}>#6BH^*HaENT(yy98{D0Qe zt?fJsFOJ5^BDqw|O$biiONWm)!btf3FW9AXgxy*)1LwQ{#HNvw>a!~paPJyZ6xwtM zo4aM1tT6_}qbhj)({cz^2B@5Vs@i^hKJ!QB6mv~?8PR&a9YrV3WBm4(FntgF>A%|| z7V7F3LLzXYvAIi8NuQN`=^{{>NM|KsdII@4{DdPw=(QK_>IpX6C*5IoJ~z z2cp*QOlfo;%;r;gSIGdXs~%_pOB)OS++a=pptsi-UrS7)DF}qQ{m_WfvI> zGkR8;#6i##9c*IRHLNNtd{P^q=N9AOLc+Q(A4fekMR?#i11~?ni6)y@@Tzu(VBE@c z*vRRhB~??&)$WgA(*7C)TmXdUKct2=SJAs7gSVya1exbA#oF?^EKN-Z@N7;7mjO*c z{WMGZ{hAi;KQ;mr4K1L--W8;G#nF!L35TrccD|M? z8s1!w7N<90=9z!=$CnY9x^D-j?0(C8pZc}K+O|wta8UGY@9AjU;Gil)g@YNNgB69dN7xTJ_%OA?+S1AJ!^Oq;0F%{ zlt4*4j0{*lgWW;g4#00CVnD0ql&n3Br;G!mY{-Ezv>t)1vKah!+6UtyUd+IgBG5g# z3Os)cGaY&IpuDySWF`ecOI{t8Oj`s$UazC24c$CRlQ>Y!-vcd$Nz81u47eYo10L5| zVyrZknf2`<`PcFf!~(BC^c5f2c|8b(TUIj(+l0|_{2;4*?g6?d|K@EmH^7BgKV#s* zVk#fhK|+1*!LkN5=73iwbQZ_(*ez|K_Awt*R7P-y@Kj+c2j@}D2u>C}neH<=@&X~K5OC-XxO7NUZqIh%bl2J>3?qkx7eC|ed$vH3`M zn;ho5{r-m=HWs1lj`euw?R1EZSVChxtg$J+0(>&WSo^vGbP1frI@D4wH>JtWC1LnF zY!TR3q+`OK95R?gpk3fEsj#Yqi}z!|!}mW_(b9p)tRG}ZG8*O|HpC!xLwunfLWB=Y zqI(5nKxIZUPe;}W4_9x4`uD9Q#n2lU>rOy>{Y>s~D<&J~NpM-ORGJZY2=8q8NB-Ma zgqd?nQTdfI!+n5k2SJ9hd>_O8VAf=bm|4L)pV7``X@)r8lLkAdb1HjV$d>=ee=)U~ zjey4iQe8OKnArqYU$zd}L_IvP{1@R3O5xTO=TN9V*`lVTgoyC`aq!e1e(-q*dT4sY~=x*{TqhuD)YEJR5y^v2VhjR7S5}`f_c67iNG{jj5*?F zIra7%a#vz8)sUKkZ$E28TWbc0PM!>I8^SHlcGki`XFi$uXdVo{4aQ9GQy3K}0>U>+ zh}eU3;ITv;Dy{Z|Vmimaxl==9;?jv!6%Vhzn+;1MXToquIy_0{eD?DDa5@u5>}{R# z@Y)&}>?$P(Wo)5&z73hOy%~JvYGH64Vt-l#no))_F7x2rhj3=0)_nZoCV@_ILAW;c zCQs=#k_zX`sJyX}&JPl0{(V(vl--13y@CSMth$kXBD|Yz{WgYcR{g;ouCML=19`S` zjw`D(XA@(SlLAWNjl@GNfzR7t1af(PoURhXMjSnaXH*)X+#&~3``__Ohd$y0-B^f6 zimgimNxXG4Dm8ro&4Q)OtMvzQ?Cos!HD8-O{6U|UzqEjbxz|xjP!}@B=fFAc_N)>R z%1eBpGwj&OqCXZ?Nk^AJKU0 z0$1lvVd&djF#0P8U+danDa1jVs33Ffr6b7J?&UpN9nRaNE>5c}7xUI=I$)-MB`{06 z>53zdXirE2-s0Ur&46pf)TN5(@K5uvj4E3?+?>Gv*t>@3>Jo(?vg2XiV$Bj}QwQGd$wi=XEFKQb6Ghcu1|0X` zCu!8%2OXT}Xz>my@Nweu(40Q7Dz^*PDKxexcfvRt1)iF9-A4$Us`vM)*jo&hB8iU zS#T7;K3pI|4vhUl822_tNX|5!Ow8MM4(~(}}-iP~E%)WabGl%B_3gK%FCv zoNosk+kW#B8B>sab(a{MQyO$a5kJDyA1}N`S@vGENmkS==jggg^E5{PdpUy^H{fNi<1lSu~ulySaW#m1Ofl_iBdr$t~^PfeP z+!kR{hd!(R^&9Zxx59?=a(JXb4Azam!b6UqVU$Sy3h6l97>qmoZMg~-?2ty&y z5puE?GcC>2K=@Dw<1@aOS+<}Y?wibKO8*8kBS}u+cfAni#iWA3n^a!6#X0aP4r8I< z8`z9VV4-j;ufJ&m!?Y}5rZ{K9*7`)UEb<4`J}QU4-i#~9sXNjF-> z6OkSX#z3MEW**rGyWDcYdU`bI$J;_#ZVcBW{E$5Ry?|MKTptgwmSc$t$DU9f$E$`e z)arc>4!GP!!4vB-z&@C)@s+PB=hy((C(l7A4=t)QxsBSH*7G*SH)6j9*A;U-#L^^w zE{5hxW3JhBR1=IQJ5H&f<&!fsOc0pqYNw%9Hw{NjH{mk*+tY0&h!49Q(+yRB!J8i#{|#T zr&1fWac<@}`XnO-Q=_7>YLJK1%1S}}r5N+n>oSIXsGxPix8Ub-!rU`tp)TVc@rsRr zTO&N?QTK27n_$J<>^@4QoASW3KbpU8?>lhpf53YI7R$1b>2$y;d?1AJ>k zJUe+cMk9~?YCnw6ceqkR9eZrL+5<)-(c}VYCAaxV)(5)5C*_00?O!$t-?SBaJ1>LR zeHkWkJQhOtJHvK?ISf6UiboE6;;uCpY2=1wIC*q6Rk)jqPdSd+&%{B(;eKIbcnw{} znK-03b9X#-dv^Fm7d#uaX3fn6*h}wjV9QE3P;i`sFZ>oW7S?N-i?2q(-TymO2qwb6 z>jzL?upMuH$Rf%ImvDNZCUi&$GOotAVam;6s+Z15e|?o;Sl56>LBd#Gn#)#{bdX=~ zR+9M(a*1+w7~fOlFPXOL3?0;-4Fgw|F>tIEA51iY*Yo>{&gL_?tuYToPrm1!*S|n- zdCr9!%BzU6S0D)-drvw(*3-Kmrm+Vlt2nP7*YSCN0xMR}-CI4HXx5GoXkcc7L!Syj z{cR3^(hpLGOfF!-hfGV&{5aJ3bP*SR48o=H!!S~9!nAPb;`g~z!R=-c z)U9^luge((+p0!*bz6;@nEV%xbkE?ux78wACj;@5x+gvu4W@M>a_E}y5t0w=g3GPh zc;Y0Vn)OU1>;L|R^?$i@`c{!;@WZvBmdayv|7!&uv0CU;Fk*7;W!UzIQ3l`%O!4bT%>6*af>1Z&}lg9$pAM~ZpZ!YN~n188*fI_RI=#8Om?BW8uIJb zLeWDf&>gx>Rv&hx5-C!+IhMM{-`Pr;B>8%`v zGi4IYRv#ItG_S$V719_R;!Rdg9U!G|5Ay_PmcjJtlbQNAiC|6kKx55ZMn1=eG1V-^ z18XkPwaKbzs}PUAM~>l?ch^zr-&2mYbshYb9$Ws(=;7PNcU1r99s%RuPeaufO>!}? z0L?abfZWhE2pInlx^wT)B$Y?}*{?pqH{K=Mw{j&~?u($zNI4NY&;)wcg5-c9^3EUJ z2p-3C$*z;pyd#+tU|D)Ki5`<78RPpgs`m`u?T^I&1O=<>{hq=cPP-m5%BvP%qQb5( zbRxpXtH_uBSMXHb9%CanVqyLv5dPT@W0$r-VATfb{%%6=>`7>o z)W-q;G@h0IA`;b?h>B(t*?saW@tUC*yg$*8KR3=}%M#|GsQe5#@SvS4cjnNIytg=M zi8Ctta14bRnZprVlnCUgL7*59cf)tL!2MHX7rtKiUAJLvHgXKEUo4wL_^NEFPRX5 zHg~@B0)%{VDD4HVdhLhv+v4f20ZF**;tHc3p1jF9O6VP9LX>0I@lTGn@;)?4*fnmf_*{FED&v5o z&ui$*KZ{8E!wxv`aVbx`pa@z)2ozUbCraO+LMC_TyLI9h|8bZG`FuVWl5B>_o8+A& zd94`!vdSfopX`GN?czB5$Y%T=Tm~M^o#c|L8(MDQSjampsp4yQT)wiEoNlSbx~`eb z)SHp8cV{(?Dm)Ek8%3DIXK%u}u}s)mWsGii+MqXcJDD)h2t&(rEbFzsf%&x=Hxzfn zoTL^A{K?IM=TAXo!);V+egGW{B$=x8rKnVQk(h;qfa_ZWT;_is)t=tvsg)RD@oA1h z6USq1|9s?h=_4@Ip8$s=CGoXx33;A+9^L+2BN1--tJb`st$O%Na>_GwhDNPwu^N!rv;5(4n{z7MjX4t}klg zr|A%BE}KWEFHPXrgl?caQX^4nmn}++Ux6UE%_uwifCdy#;qP6%2)2#w!7U|C7LOmX zWTphiC(c$u`HS8tqN_lvrT-(s%xscjxrMHKo=y&au*CV>mt&7!7U^p5M%gui!AV1H+evlSI{ z86o#zpnmVk#9&c+?>1rjVltSkUd>cI_$84qMq2xc%aMwW(etinfLYIin{+SlSZ5=b zeQe}1JF1X!#2jZFeNXN!s3vx0VsPL2F<2#r!_wg=(Av|AJ3pKOukd8}l&eMNCYi(1 zyVDtk=Z5I8%#NMdmxb!J4{>n9GtAZL!Rj}6af0nilub_~Uh(O~_w^ibFP;o8Z>GS3 zq#ihYdkNY#U4=RI>zJF-B23xjNcwnN99AwnPFDA8;B1F^?5k&6=qD8+w#U91uWYD8 zr-{+nCtXvM_V^x#sBVHaT{g&iWRZ7uA>_+aA9VlPh>OFMutjn)DxJs$$1tuJ+%l33 zYOI8>v5V2y$eStj-T-gv4|yBAm;SewVNYK$gF{{)dFwZbv3tZkS>2T)tcH>= zjpF!riOWkc(L@*pv^AMYOIC4Oml`qRd>L|?0T8JnKz@GlpoiK%<0Ji25-rW~>IXm2 z-}?vQ=zfl?l$ndU&fVbb{U2Gh=?N*C8AO_Y3qYU_0q3G=pu5tNgm(Gye&)82P>vr& zcAkPsYu`X!wG>RW9t2~RsqooOlYKPrIokXggXs^a!S5t#%W%DRv@Nd!5?%uQ2U!@M zW&nqZgqZsYiID&77=6Abk+Hu14L`ryfy2A=;qQeAB60T^(P}S(3nKrx{O<=W6g0=G z{@>`(yq>lD(u=mi(=fsLN6k%@6xgh!ZE45tNC(!blgpBl%;Ex$4ZV}wFZLLL*3=x7 z47dsdeltk3=Zl)#x|TS1N;3U5K1|k2b?}rIj^m5hGZEY+u=L(Ed~z@vrpgE~NsBj= zp1N?j7ZpTbTXgI*6n zRLvX3pHgkh<>p|Y0 z>AIBvCmVho(WQUBe?fokXViVND^@6*qIjAUt<18*ru-dz$)HnsZto;kV6G8UYa$Hc zB02D+M}wIay&alQe8KV)qrB)>#q_g%Le0QTN$_tBw`{eGCdz#gyf3X6_|HQ`aNVa6 zRM--T+vURH>YKT^$}9zlx&kvpXCYnQ{GOObWPwE3G@K_OOuNOE*ySTVbWfui*m#Mf zk9I!3vyx%*GF6{? z$d;e$sPr{0@R=1x6n`%wyTkI}vZFC0zdwWo7)`^u>n`HAISME`nd8aEUgGrQa$J+X z9^ZPE@~jVh#V;nID7mW^(;}ppE=?25axXpHozq7j|6T#TMZhgK}82_IgRR5*aD6x!;qT8;(6VR2OT4F2TvK2= zflViy+1Ld#{6@EVM5m{djBhv(GelP7`MGaMx)V2pTr`D8P{hwqnLq}+cXByfIreCf z9MpNthxnSkM50WE%t;)9nSEL?ygLvI2sbnQUJJ0thH7clV1$$-mH!ce6Z$!p!EH;T zZ7TtR`BT|a6)hGlJHW%{EdSrOPjtTGM`9ay6V|+pfuzP*sP%eIw0fS=0p~WHz*6hZNW3R4n#1q!+S(aNYUnz<$pwVg$9!G8t_Ke;>CdLGv~dK=FN zkD$`Styq7+iLUn-V8){^Q7h{xy!$r<%V%<3+UX*6*<4Rl*73r}?JCgg`)JkI;O1-^{_EF}T^!BBtl29=Eag1`aS#!AE ze+W%eDIgbob9nq~BUq|A4{z*L1MT26x?)liHt&{(Pcyus>9Qv3#GOM%dLDRr*%3eK za^B>SYAhE}z-N&!sMuLOoW-6Y_MgjWmfB3d`7$HYJw4O%{EQ3u&O3!}m2SuAbS@vh zy9;d%B}jEZIVrjmO%v0SaGPQo9zQYzw%H1hN1Ly49Wzl_-Ls1kXS}fWKf-IH;}==5#%Ols;U8@yqhDpf086?5qZO^jaMjq!z&JDK{W-`W!5F zd_fOzng2ldEjYE)818!3QGZ=o_MudMjZ20t9QXS`&pyorjfN)67hF!@-Y{bAYzbU_ zV+!8VmSkQ2W}}~%KWHq8g3WTL@UB5QmI&U2ySGA7rLC9phq-=U>3E*eR89vGbH{eL z2(XKQh$P4Nzqp?cim9<)7P&ZO(*-K>TZHlWH4I{UvH0wpINXxW1JxYvcRP*nbe3Cy zduA&h+#t*9f0JYWDh@-R+CNz-K5nq(DF zDe#4I-8d4HXA7vi8n1Lytm1p(53Z9h_32BhP{v?;J7P>_1+2Pf2=>!A@v^H9l4RK^ zj+^@sjb>dZ!}V$OzQqAZzZ}b3ZSWQ>hZ#(h*$LYVI*34+EVbI70*T?_Fm>q^`0>n$ zy?%KBGE$}B<>4^cGED{>e;Ke{iG=xdFBBS#+(9a8B9Tr>!)e|Au(a|PKaZY<%{_*& zefnI|y}Xy?Kg%NHOLUlpf&cJ_;!A2NO_&DtsSJGjN}f4{kwBUA#!i*D_S$G6=nWY?i`B6r0aB$u?4 z>cK^%pADVOfDNi^IKNzoEtW1 z`fnnT3}tY#n9g}K^HI#s0t^b2AQT=$c77#R=8!!0T|k2h|Am0 zqWq9M^rsjQQ~jGnsQeoozGXlTj`d--@lLcH$icN)yQ%n;A?l)Chygmi97p>se%lej zUpM1D4SGF+thh0UChbq+U7vj(^U`(L13RZc38$maNI8Svf8LSJe}_O~ju>+~?JGEy zoq^v?JEm(&~;2F9(T%!)@B);E%=odogak#Ep4zUS)SRss2^-cUV@32 zJri%|4o`%nn9-^t%NcGnz?Id5m>u$r#LM4MyiT0i6Ss^kT^EK8a~8qQ7Z2$AHJrgS zyN`(fQ|H-!r({(nr~Swq(^-q1Vd{bJ&|dln7sv$TYtCEsW0^hdJSzm@QF$n~bRrwL zM;_m_}5kr4dY z?~dx(61e6=FSUb@^vF;py-NS__MkLwBbu;1-JEZ%D@Pml#^HBe&MPzX9(?({4psUJ z(B?)F`X%j0$Mthrr5kooxaI&Tj7??GF&$>7oh4!e+u+Qdes~pCiN#wVldD>9`7x3$ zL|1hUCf#hJ-HT-)pyeBB{LshGQD~#V&FA6M;xuSkmV(n#xsIdP!{qyQ7x+EMv5-=h zk^6aPX}}3DR3FZuJM*J(Fs2&+_1waXeq3ioQ$A73*8`;$H;7=n5;MqUkHg>Hq2pm8 z`0CnTUgZUO$Sy2|(6M0b8d^YB2lYYazH-dnwFN9+{)DMBBVoR~FEduJ$JnL_G6rR9 zIlp*6q#n&gv78oS{=DMtHu?Bh@>Va`k6uw+*hXtPN(Xb>8 z{Y!p9+2RM};#UdA?tlSYX+4Pne=biQ{-|XS?RB70e83H0EVbJB>MeWL>$ljPp9Mf4zWq#Mtn@41z z@tzUH=T zZ#TP%apnH5v-hMRQdyZX^;?D8kF?@1J!fqAx)vWZMf~jDA8D~6r%Cr`^IR7A(EE3+ z8QB_h^7V8w`YWh2KE+!Yp^&M}r3aKbO zbQTNZL{Vc&C2vRA0GX6EmmDqFh6l`NLpARqN$Ht_7wcMx)AT7&;S+&dUYOyk-WYzJ z>^6SU=`^xoWEyKTKMybei=8bH}cC{CPDg`H^-l}#;l8H&@Ox-_}#xw zb3KpXjnq4QY5AvgvqL6q&oD(0Ud9x-Y-IY@w!r0+Js{;6$h=tf1E^Fx8qEmA9^*2y zA-EH@NDPWLCs5F6Mc%U~cqVuen#AbivhA}`_mUQI4wS|3L*cNoL6*NsEC7D(l3|t! zo8yxE{&3yBhaNMjCuiFCV)A`kD6CC}#T!b%YyV;RX(j=a6(3Ly-ddP6V3Vwyz<}Q@#n>HzU}e}jMn-cWVu5uh&L?dy!un% z$=}C7zE@f%2gva(dj61aE8fALCqLlyo&}(L^B?Z|nu1^1G;C45iCsHuG4i$&`@Ss> z%@u7Se5WnAxt*esA@b<*%nR->wqmQ)NO-JrE( zzuZ1hY)XS~C&fwVdmj{ORzvyQ7kN%*I=I?X7kX{aL9;_Bnshc(to@5;u5)^O+~wwvxK!P3P)4NSnf^#h3=1ej5d1$@%x2C%zmG#jO_bS@K#Vn4xmAP?Tny}W-eI9 zeLo}}m(Uj5tLPFrpXN<6MZLOqGWAITl(|>&C(S+ue^_Z|hhhpWJ{H1uXz}T<{O5Sb z^A?<()lc^-x1!3?)ohjDb5!wEXQyOzV{SJOa&s@^fs^;}Q+OacUCW{8PkiP*w*3$F z4;Ya{OAMJQUP35%p%qpy5N7TcB!S4~@BDZzBSy|_lqjsWh4iOfAKOztT;M!akGGEC zsO1%8oxT#Ef^FodiXf}YszUVm3z+^}mo%sECoNWw(Yv{T9+u=d1BX5Et;j5(8w+^b z_l9D}`t|hpmlJT(?<6>Fya8d$@A2Ax5JamyxDkFGwk-{YnICWPUR;<1)uoa2#`Z*x z>1PU;CaRFMQ&DKGB@5xLyRdCx0_v)_(5@fq_$BBJnqT*zFDRn={)Oz6u9sxH<_cC~ zzZp#{bVpt7DX4j=0LAVf1=aV$B*TrMXOA%^J4KRAZeHvUx`Z7Zd#b%!8UIPe&_7ao z6yN02AB(fGB+ZE3yZ;5sFZgJYk+74*$4p@r$K0@N36B@_trz@^rqG>*GI;BC z0*a6RMUg*e=!ASb)D;lM=~X2>vwAt^7w-XG>l8zbt9OB3ur=(vt_OFuXENuC(%?+k zC6r@!k+ZwRU?hsuvddP&z?rE`#;7#&$n7Or*&)gN!E)w%k`(h$M~P`IR|1j!?Qm#j zE|KA~o-Jqf0M=e-qJJ4O!(%@nTlyrcVGiLQ9N$P;!{`7t$(>-c7#x?DGBEJi%H1-4eKTJkT&&ZWUzB<+ei`*!exUsS@HKMiDB;&j~Nc>;{q<-(0D zC!8UlV7Yl*246DcRBEX!9dGwV=CK6hbJ3L_(m4y+GfS{y!+F$uVuj*!ia^so5Kd77 z%&T)`f6;g1`)Dax*$mUc?Hh2@c~f}n`IF}$o{m!rH6gfp8%RIUz~w!KP{L1uCKY20 zl72=t^U_EJ*X>dQJY11%Km;$c0PQY5tTyY1%Hh=*`X&x0yn6~2;)^~r?O{y4+48CN;>;(CKny0!5Jirn(YDPIKGBo`rge2#^s{@h%ssfnG1 z^T{^f0H};lfeTrt%$q_5=JBoLa7C;aB;#Jvp9chCx9Bdob7nc57Yc(Aql>`wW)t~Z zN6TGi?yaa^}t*(>H7^2jS`?c=OfG%$Y%c3Yz4k>3LLo5MZO-q z!asTV4e9JLCdEy&8HGKjuw!;|XhYS3uI`4RsUWGeS8W-^uyk)Nu5`1q|O?ri=}c4cm)wky=Y)%+BG z;ML;65n)zuPAcuIiKmQw3h7*V(=u$612I)b=*cR*J=o+aA+53D;?C zY$48IPazqb#{3L+fxIy-Nc$Z@wm;j%lM+75Pnl3fAJ%-sK7KlF>289syK*=^J|K&~bvtetpo zmLZv5Euitj8MdVAF>6-z!@NjwW`4~nM(faGBD3WU)pg9`nCug<>I`eCUDFOm;c?)$ z%?nBvo`FAALDaW&Asyjn*5paaMC|!;Jk`~~YjXx_(YjrEd8S1*H7 z4yoJ)}44^hvjMYYG&O`12t_zEJwW|gRRdr$x>bxQs){29F-zhTTLkFHdmkgH= zaJ$Zv7wN%<8oI$z5m!CbV5Kj0@qM<-WB2P2xK>?B#xJLlyI+H$ZMQv~*|LS|Z8?hC z?vAu;u^R3aNT#2-y#A7~+dR$m6?9)fG)!RSSX1XtTq|$Iz6p|GdrC!_L%iF3V{b9W zaitpD65kCEFWJEjzwhX&5>4)QHW2c_nzYum6E|lYMu_wJHG53L)m=hJ4SGp*y(B&t z5M^SjHCemuzc5f-2S5Dn#C6Zj*$2hJc%|+(ydP8GLx(uamUzO~9?s9#inMc23VCIs zfesB{P|t&7jlR-Fm8FNUey|dR)(BbdJf4VZkxkTevk7cbMd)2t(O_X^U_i32l{uC4TWF@(0JZc`*rJe0op9)t;g~*D<6s zbP#3tc%u8sbMSFy13x;skt$~BGrqUV(N|^w9t2q7t<_o7d+92W?bYBsFX3SB%Jpjg zl%{{K>qEu5P$&?nuBn^k%zUf}V8_!ov12hD*I&F7S970BcY*_V&o~6l9DmKcXprM8 zDr4G(G-Pu0*j)j8LGHaSG73+@nBIZ>L4Ts|)k!ufo~Iq2eZ0+wi>QiRCZ1Xn4lnJE zVbHG$7E~<-%}7C_xu}fq`S~=tIYSi}&WvNC{w{&gJ_8{8zCrXBJ|ntGgn3pN&zRqr zWIX1j5dKSh*mi6jaw4*r17dk}j@}v4JAD%A$CVo8{viipczMo@cuqNZWOzZ8{WRa$YG?*j2dBnss>cZp2MU+M%X;> zEvh{JMzxpvqxEEa+-e`O^d=At;4uk{~L}>&c?%{0_?w+Lxe=V z<*B6J!_59fER@fI5%cq~G;bfhtgK1xE#JTr?U$foJr9#ADzN&N8Roj~!W4ykTqARY z>nW39gsPI6^>-IDRvPOWzTmqJ*XTXvV*XPd zHB6QgW4G}$amZ2+tISVQ-;$sFKOm3o$5QZ>)plGm#tP71=gDc!ig?a97hS>XKb$TQ> zLk!GL*>U;$>9{CN1SYkuhjpLkar+KCw5(Nu=-KrUp_WdDzQw`bj$+7C_zf$x-;m9Z z+Q=SHO4 z2%GN-k;pl-kZ3mG(=R@l`rHrQC)AMr*^}^Wa07Z|R?~z1Z?H6ABfbCTG;v(ALwAT@=n`RBz5^)4m)=p^LNG6})>;rFxcUviK>QRH-ri6I*eI@oHH4IhS4W zX$Ggyc(B?tXR|@OEZAYaNOB{n7k(*u@JCoL8{3`Pq%)=?`px^#3s0%iM*ur6$YdVd+GhrR3cPNm}_UG}|*fCUEFU;CJzrgM= z@nW0ekDyuNPV)BPbNWZ@4yv}xRK0k0p(Y_<4!3(Ngw87~A%zaYGrcx)jI#>0u;KRGjX{2k^3H59lAkxf0V)aj#+H+q}A-g!bH+*;ehp=w!U@zC|+2a ziOd;SZdWbO$S?GzaXuR0u$SXy{`v|l$7VAB9g|@M{rAAUA5Tah6=zo~w@|enWBTFb zLdI;%5F8LS0X2z!&>a2`%JieCq5W)j!+TEmXpF++sR}G#iOVS&d*ZWl&0voS^w34k+8m$E1fFG$@~0L4 zq`|&izN!5pR;)^8t@l_GpN>cp9d-@1%M0me-$7E@cB9&`UWVI;Ht?6(za`;o4}pto z2x%77ri@b!e3PDsJAQJX!}kPGdw&7?o_T<8g9a0HGX(xv-6n0R+j&QCT!CFT4e9r; zP%yM!RsF5jkU1T|RZD(#T$%9|o^QjQZ!3tp;w0dNSZU%Df)nCg|T&WW4 z)N=VYbG!(or znk*enx52)9!yr1Wz_jjBfY{#qpmEoR$yBIe&h5AZDi=1Ak4G3R*>D9Ds1A-SttF1z zZo<~>+MLIymS1)Ce-xefBUj%a$FpTcAu1v&iiU)8&+ACC(m)w$sVJ2XEiK8;jy`b~feBSJJ93EVA8aGH)V6Me!OxNm0d*re9*#ex>pNM1UiD8pu z8yj9OI8JrN@WYfJpuX}ExgQ;my7w2*7m8AXcS0A9c1pA2ZWBBtQU#;5IC6Wp9j`0Q z1@ezAV$#Eq9M?)ATRR>@$&=L>?&Bkx`ON|9CLYH!!~1yiqzX1QUqX>d4Yca&Q=N+y zL^`}0D?iv$8AWYso_d0Jm?X|0pYa#hc6XuDwFv&im3oY+pF}l(--NV?5I8++AK9TC zLwET*QZ=)yxYS9=16auN^M428`DvM=r?-2sZiWWd`{p8V^Mcd;R{)_hqqzfy9RDIx z8KaaoL*S?(qRr8Zuy3&g6LjTltzRN+x*<&$UK>fXn}K=1i+3iE5*ZG6gP!L~(4h7c z$F69>V}nxchx!RPujCJkvy0JiWIow^FHFcSw&8czeb}P765EdH!0!FiAnso@Tx<6N zjX_hI+EfNPKYBTxE5XoqKLZudo`x#pAT~}vgzcUGR+RtG6Rs2naF%(uLBaS8J6xSg zVD%s*<=BDFoy-@eq_qy&_w|@r{gXWI9nHU9AI|IE+=p4Q3cUHQ#PameP@H8tiC>v>8zYt< zgI#iJ^lMWWk#9WAy^;7qZclsx%l~Txm->3{kDVgZTsH}pN|g|cDLOE%?>*Tw<&Eg4 zhd*vum4P!KMzDXj6L6?;6MA|mkh!IsuwmyyeDVAmC+l((-}FQg7%sS{tQVV4wU-E=)(;b^nywkIH>xv zo=ip2wt1)EYoIhlPCrE=Hy?(Y^-&;GBL;SIcP$fi#G&=IJ8m<}W}b70lZQ#MV5_p1 z6>G_0&M0l1ASFrL#H z%5mZwUkuv5k)Jy+kKg05hIgqBLT4BQ9eu~JY3VM!Yi7Yq){0SnaRuD{Z4UPP?h3OP zExNBojLtZH6OTNu7Um>s@NwfAGV@;qe*W`~RiCqlQWcJ|{bty=^NZ-~yqo3K#)8Y? z!5sQ^@@Q(LwFhQi%g4P2vQ*cv8B_u&|8TA*|D^f>o;C`{i&eE4x$iTI9MUmsPb-E< z#t|=53(+T`8~5qMXKdYU#zxNRWpk$X6Yc5%<{8u;nJz7}&7!u`LV5G+@ zHgCZ&iXCoL=J!YJ_@0ceN&C1~Z%6zv^bRNM6oSu$Z|4Z7F&OktgPif+f~rgrXWt3K zlfgr<|M_8-|Ls4^vNu=DLzTBOzdTboVLy_;m$nH@qqA7S{bBsBth+R*uGLxc$|0LwvP>msBd*{XyK$^uAqF0HjiRDq=a}`#!{keK z5!-t2GAtTA4910D$!XC}ytTg-yd-TwZOSERP&A-By{@5}iKmd^3V?Tmu{fb@D}Tmi zB3{ssV!_q|rzJ?>5+~0GDT`~O$)|te!u$sKbV8auS!oFJ2^U3sGiqULaXt9f2)sjW zNf;YA1|HiiguIgm_^MQn9^bDB<#jsz-sDQ~oS;eh)$;7L+6Hzqei;9HN+HUr2eQe} zgztpHP3%~&$8R?oMQgv`ffs^D(OG{q*vj;g2N`=gSx$pxr>Yh}?gy`%c0?0&*cN=(sN>!OcnzlU-^xwyqD^}6>S@xKiosN+j zh3GJB1__aJXWyS+bIIO5{Ldks=q307odu?0$GDmNfwY7C<3ueO zH|T)wt+B$r{~42}pYWrwtI-J?&(D$5<`-*9@cTBjv1^h%FUYV(6Zf0b8)v1#qwWI) zO$-Mc4For#LzPg_h>zG&95?Mi`S#U$uq@VzU3}Ois@q`9%auf7`rKy>hI^s!{G&KJ zTY>LAq=RRQKj1H04S)ke4$E2$cee|9_4GD;{Vxqdm_Gc_UW<2jm!Mg2I`4nj9A_w8 zW2!!TS-e+-;E$7q;eKjx;P@4I*e`St)l;BTU`_PzdrUN93SsQ3Rn#nTko{Pb%nEu= zaAh4Mc(|6HD z;4^a;D_O`vg6B+t#!zgb0&~M+B(J(qhAv7x1%Z}-L2HQxKDCO)l4WJE=5H+Qi&254 zxgDg)dVwhM;dv6R-;N<;m2vD_DL(4pEN1-YJd~~&Fyri~58FY0F-=M0u96&@nNTs^MM z=wQq5&%#4B!=NrI5A-fakfH^7Y-ySfwYZuJ@~u{?lp&vEpiy)zX=+HKfq9$ zL*iqflK4_HIK8}3^z1+xObONz7`W+VkHF+gz92ZAEtkL;$7uLbG8G&IPtNp5G3Z~m zR@9~Y8jJiGDfT%J(t;n^?4>w$$j$}Tlv0@WVKLv#`H1|+5@KsBFpgIQ;c*8MQM8@{ zA3B`LZC3>xGDjB0AMD~9rw*_OUaPTlRw~@GpFzynW~d%=4&QA(M=rmg0Y#1pP+W5z zW(J=kZ(4SJHb%18*2Bf(08h0;AS8^15XmDMEgQod9E5y^=uKWa~0^<)#fmE zLl{}@TaHI_EkzI_xX_F?ay1igK;~9)yf*0!TX}Ucs@e(s@L92V!{j%fTiML+6ntjx zLmxBWCQpLPj^n5cab#|)V7uFLj1}JO72Q$k#g-X{{LHhTM2}`pz~}IT)M;76?q!A$ zee@gY4m--N*f5XUZz+JhvIe5pa04V4{({>BJNe5sH8?Fq#3@g^D$F_lupxe%kZ)1O zN&PX9bLtt$_ADS%s#7UvX%5mC3Pq)X5;S;>3hB=`;g>y{$2VCc&a6-2RdW(J?XW+V zmsa&Lzlz^%;@s093f@Ql-AX0vavriS`ATxH?JQB^3kmoBVoI=eNZ`qpd z)_7=XI%scshgO|(sC3^Q|C-j~x~YP*CT|$cveXo9S+tlis~C1>*B=%%?l#HRy957> z((%8}1W5cb498ryz-@~J&dlz0Se{FU&k2=FWP^@*Brn@pC$IVCY_F7-C;U4o^G*5)186^fHqFc62fMyGEHs9$Z*oC`q^KHG;A49he=w02WSM3p76lpROgLJJEkRbFqz5N0RS7?g9rD`;DUNGF20a1vg1S`o1 z#VDOo{LHl~{O_Jyczt6ahU)CXUzPrl`v2V1vlmQxt+30KXJm}cWcDn+lC1KRq~32= zz-&2B{PaN&Z>;b!HZ35*rgIv74Dr))7=3|!(<(C$o z#iU_~p=B3H(+v-{Qez>%=*w`Jc1IpG@3+F|(ixz(<^~qyAp5Xh4{9!l!?g4ltaxG} zk#abNTkf1E`r`Hwx~Y(CKmLa}Pr3+q#BFKO$C12)s2RIIh(XYgIPSCHDyd;t_{CeQ zv38d(Z+Pq#y87&3YghWi@9Q7oN%c{d+c}a1@83yO7cVDbkH(_UgmdufuCHk8<~jKG zOEJiYK7{ojr+`nL5-;Zdhs^NGB*j)8Ox8FF-EpYkV%f)^OFKx8FVbgXPZjaw{dDeV z!&kOypcWf{%_VP_D#DinB{11Lj5^fn(}j`#P@^1-%L0U6rnDvd@9t}I*zUQ&7xBjL zT2srpEd zXU+uC#so18N*hjJ*V@AU>_F^kQe<|i2e8Vaj+tF=gzPgdV02|ZOgbtJLw;!TJ0mUm zd>JuZbYGP>zvzz|Egt-D(4{QU?)X7lXV=kI5)5Zq- zRWW0)4-9CwlcyR&S699Udx}o7W{G%sC;tTswTm$DxCYN2H^TV4dvJomLTE@-hCw4w z++JV_Q-wSB%(tUyPwY@!=hX)hIzC`jS`MoNV<9`#0gCs|fb}~SdBbb_pgHnA@GTnT zlihG&A1=ezj8~Ys$`E5DhtiA_KjG5L1LWY)TIl!6hir$HXd&=VcBeci3%?O~fM z%JTuA#dzyhU!40`7L&Hh5`{yqT%)ZdFZIHJuTXf89~1P1tg$}ts=pCs1itCgGI^T% z%?JvUt4Z|RWKrbELp+SD2h$Ke(0Te=RN1%?>L%GR%^oSf=wLTXAMz0v8FUJv{Y|2R z7+p4TXgd?l-HyQ<7NckXDZITV0bL3&<754FUaaOUOc|OBNyR_NlZ%_#P(wvDx2`V# zmVX9o*$_S@V>6pqtk1`Oo(KTmWxO;{0o-H9>PY|PB@hNm&r}E5;{N=`Cl8=`0iuY=sz|T;|~wA z`XXb{7+Fe+n$;|qs8pcKH;VqxIv7Z)@oxF@_(pdhR4>}V`6(rdluqhnS>6D<~xx*7`Qvldu>=7E93BSs$UQ^A}#JHYV##&a%C!cLbi|e?s*%^!e7oU=ll2R-QpUq?qSl4sxaiY z7jHHq7W@TEVz0|+6D&?@2X=#>_!w;Ybds3{5^xsgf^R0IL)f0N zA}itk=^DEjB`06Qxoscdk=AK=(PN6=+YaE%oQv>sZ4K;`8U{0jo%XSn>$$b+^GR>S z1ZMONn6-x&9_VdAe_`gj$!c6kw{#sDIz5&?4Vz_4{{De;Q zklBY_m|0Ib<`%8Ty69;_-|Qa_Y#75|ww#Z{R?1_ez7tfxcj2bDdyCrrhk)y6V|qRC zHU!B2g`JbD*eQ7j?*6Fp^!JD1e8r`8c&cz11_(SrRTFW(P9>AfIOm3IOrDAE?@D1u zzHGwl(N18${F3OCMK*sTu?k7+c685kCG|;bcyX62ywQpwg1=KJuOtki6;5Y}|Ep0{ ze9c>t^`(c{7;=TPdYnw+3q0Xj@W2%h65!62`qz@N5qhg~0sLY0pJ z|GT=3y?vz(Stb+N$c@#QGQ*wMH2FnVIxi8N;M2+Zfv4#0+{nD+;!tjy7H;k?hbu3g z(C3LbInw={JR6e(Mn5{izcmrKm_@8#a6DYdwWCd9TdDKfRnVGej{l8yfbNx>S@`Nw z;y7_Lu309@XSgkdUoTJKgVop2;JPCQv@hX)P1J;)?zN;iawWMjZx|L7$79C4c5c3V z1@6Cenzg+XgJri~oJ$JV=$^ zNSZq^fjaJ*i|e)y=l$-?z!x3%s5Fp)Il6C!U1AvO88+j$CttXLuSLwUemc9(KEPM| z$HaHR6Sx!;4e9lT@YhEL)JtYs&I>(`VIPZx^Y$S)I3T*827okDjljGr)K?u}twc}S$ zPsDVA3q7E?3Kw4<$!{{N;4>Ep45PuDT-U01+@hg}nD66ufk9tJ_PG?{(8ohi_GFi6 z+s^l--@cIiF=@jK->sS9({hqJE)K?+Rl(-mUbZZ&pFQR#faKL=(A!?cIx3WLxWYK{ z?6ozqe!iRiuylfS;r-L{Jf7&@Q-V*;0uwn?nO@T_61lpFk$t;TL@$nHp-!q3I4%4{ z%Gas!6TRn?lQW7~6*-94n6jK#ow#hh*8C=bcUU3^vdCJmXa}@ErcMEA)x&vhdw&<15MdgqyPkV=? z!J+bND8?n=-e6t0_@kXX=v1Uj9vb4>i)&D}a5-N)SLlb=eM4(!b4(dyM15AvQ-_>~ zQ-P}ulJG#*v)dW#&RvOJc@s326T!XMN5!{vrU12sihSj9JBfCmR zv*-fAHL{jmWzl%@Y1J`icg$SyQv%MQyF_h^^T?O6i(!*tCCbc6Mir+ph>*JjjwUbQ zXoEW_2)w16Dj&=I6|$k_Y6|({R%VK39~0aZPkL~7bRXKzImKrVrm$h3qHtlHkY$Z%Wi_)4 z(Cpj-wz_RN?l4J(rA?XSpV(iV@?k2>dniGV%qDn8(Tq#pH3}zl0#B@}QkB*f z;AFH6Qr#t~OV@N7w!#6FY9~Pc)k82oNKx>cK4Pmq69k`!BHg+^0mc`G&z|$781_#% zfS2Qh$FN2Vd$()xCEf8jM6sV_%lR_#E?NHPxJEWLI2^X?Si!V_ukfwZom9t{v47Q( zLMP}7d2*u*2kl326UEx$rRPDaqa;HoUo#V(#CK5RtU&p{!dcLv%Cfnl1@f-1yWF)!|MJQN!n2R{)f-a5c;n8a$=NqI6D_#%b zGgWW0?P+x?FED#9r_(t-?yzl{B5HVx ziQHsNiOtn0lnHt!_^u1c9sU+L&aM-!+<6CfDRcs{7diud61+!MvXEuogbVH#k;=d# zbXVbVxfr2y)njO{oB}msTWGJUAKm}yBK@h6Oh&!>1P4^^fLG-cQSgWBqPoyHNZ#v2 z*R?m1gHj6={v?UkdJ{Qwahd*dCpYN1KF)d$nlBY-VDeG8?K%T0bsS+`*$s%9H;EWIyWzdfk+_;y zC8OL%!&04tV9@=Yi1!UJ3)>Ow=2|h7N{E7!cVnU8MJi0(9}ToSirq4g1ouagre8f{tYaGGa2ppbLKj}F?<$Dvw6gJ z9{I@i?wbY&MBaF1*JPaJe-cJs7=ZczJW#n$3Qbdf5?xt;^7q$ioTBc?Z=SjmbJnzp zCN+%Vo9FyO&#!0LxwVraV`v(2*nJpmCa(gskOz=~@37;k1>dnt0^BZ}u4O&prYXg=b;$+I6tH#Rg8wUjs6BA>Hq&yF>RvmYA!Q8{xE#ZM%oO?V{c z8UIG*ut~heeMk5v>|M^k^v5d`e&d~^6Zp8A9X$6eo>FQ>PZzx+TJLv3>dN(GzNREC z<=)}PLcs^xF@~3P|5^T3AsTJb5nsNXi#^LE;GMoEM1RVrq0t zyd;$iZGv3`hakV_A*AdGLz83TSTa5rbAC`rjj+T>u|hQHGp3SaY2ekqUidAJ2d(?c zFz%KjPO|jiB`xbQ_rI&;`H@zfvQEf#9HsoX*8lL%-auTiWCY%~48=QR8GHBmJvq2R zoj1C!0Gkibf%sM%P}#o+^ctH<8_wfjo2cXV*Zbh{nP#&1`+Xd&iWl8(y1~}f%mkH| z93huQI5CkLe)Lr%BC8m9)ISfeNhv~>svNNmeooTt4d}-&z3A{n$d>;yBzvy4ve{Wl z$Xmza%xDR$`uYm9H>8uSbuG}~RtbeS%g7$r9yaT(L3wnY3^pzJ!#yxq4P|Y^=!p`+ zjq}DBmZp`!s~g{ltKn|^ym1t-S$qhV9twkpDf8KXc{4#LcO_`{z9H`tKH!HX#^7Kv z479)7Fuhr)L3PtXqDe;bek;uRy;jfRw*MfzSyhSaCPZSHoIV{?)Th;dBVo+WFxV3} zo9aK5p#{TY!Tac8*0J|0vol?Uq%#6$oNnUMPfml~iVI-eu3(Ufi$&3&W!%%41u(hH z6@IMT4Qh90L-&(MqFSLl=pUCWI{$GK-+5s(-fSup&btb@r^1#`ip|3#E;TqfaW8Lq z&|D-R^qRcsm*S_mJi_ek7R#zW1*#jQFB0dYEnhs+gimjOLGTYpYAWO>N}MA>ebY&y z$5Ri6-+q8?dl4LHdP9P5r9eF=M@+1==w}~s$j?H)_KY20zQ+tBvZc{KTAlx)>&TZ^ z72&RRTKGL+Kdv9!E_A%)(5{=o(7pqhd}TFJiV7#8!@^*+<#S?p>nV1+PvU1@$w!GT z8_}({04M!%5bkhVe5&dcYI&s$)~hD7t}juf#Na&i>Y2kgzk}p+$Z8BcnvQw0Z}FJP zGpuSB$FNxw@bBjWyvLrPWZo3mvTrIJsF8w~ZLN;DTcefU~X>n z;p4Xhps)G|L~a}K(7I|E^ik#~zm^8AP4RGQLn8fAmI>wiw}Mo=3d=8jL%u3Ua^Kfo zCOezmXsauS&)s=+uCaxKYX+E=^a4=d9|g{{{ZL(W9KFp?6QgIF$fSZ=u6yhQOcw4Q zY6~AQbtj#2pO{s!_(%;TO-~^kI%e|hv?VY8&>v%*uVSO#XuilemhH@mf=vy@_+j~F zSkkr#;#yisa##)yzb(a|+_{BCp4|muz2Z1)gaV%S)#ev?4JY9u;f`d(0Sim!+G|UQ zjmHZ5Ji;BOo?TDYebK;83$&2Wtb)ZeRQMmE%SqCaI5IXnflL>=n=7ARLRB)KRt-Ig zKNpJOn7`6o{PQxlYT~9P;N!HcmIF5pAlu%q-8T z@l)m>z+IorQS#Unex=F@7$si}O6`6id*>%JpZg4yg&dAkt`l6HM`@z_bY`r0jzx;= zb6Fajf%|nBE;uTX4Kv!gA-pR)B4m5=Y!7f#_nyX$+a}`0H=ek@?-@J&#Q~ShS0cve z-Jm9GEl#nD7x><1QCDUc@9TFL^hecGiwE0i_4l#VH|Z&H*q}%aCdbm+vTSH*n*ph+ zX=K&AGr&S)*vr^z2yGqCgU@idq`U!k7@s4bp$VG01^$ZMd^+-+4a%!Mfp;+xa1SF` zahD=Bdt*!wo|2^c#&5Z^fwQ4H`U9xD3jTqKax_D&i=6wN2Z@c(;f3cQ6lp8c?Oms^ z!!`nYi)LcP${hArir|Z-S8=D9zzZ&p<7)rzC8k@A*nc*|;qlM{FphWPf3`ivNZaS- zGUvvk>&5qMXJ@eHghdV!9L7qM-BJvNKJ;N^}+_;O()xJ`I~wmtFq zHgqUGYB(8gG*^S$fnOr2$Ja<@sT4mDEzgHvxJhg;Dv4D5mob~P931OZ#@#*H#2P=} z0`06hL|(oRj&3W01*eVK#I23ENF!TR5jd9wWb70T7ddk|H!MM`hjNSjE{Sv&`+-VB zJE#3!0_@p2nAP%U)CzZFKKvOwk{*wC< z=AKZXryrifc8Aw+CDj=YelcaC&FW5yJ0x0g;P32*YpL&{4&WAp5!; zhQ%1Oz=g@QJ}{e{@Tmlgr7m>oJrxu)UxeF?o`7zl5q`E39AmouIA`KRoRcdBC;!Is zyLM>vL#mxIRBj;yISohf|(WBjC`AdHc+35as zSX4L*&Os|~y|$dS9=?ksLJPpiKwSum?;%|OX?T+_xQ#XH;g{`m_;Oy1>T??*R8pRP zTTuvCA7sO}nHtnSx*W~EF{1am7rJ_-sK%KjSa7)sBl-yOl8$39pA@rO)1R?N!knO? z*^+z~p7$T?7ouIoNIb~}ivnuygVe5YPT_hN+~ZD_9nw9@KH2MVkF_2E7B7eG70#r# z!3nbOJ%?NEZKP6jE{HWvK#3|zw(VXQ1V*gDTa_)M%rQD_o%m?l>E;5N<`=kkcVk&w z#Bnr;IKeijwL#WyH+rE~79^Iq3U}c%ma%)Eg8l+2s$KDiELwjIB-8F#)(boAd8xuo zE-@ZFYnFkvzwn*#bVJY0O2|E(ipFhfpnF8%&$vbq`eB){cYlo1%M?&0B!sEToPc(L z&qywmf%w;IVF$VvSDqb8FlLBwxG#^}K_3`flUGVvsjcM2G1&=^Gx^3@6 zV|s~1T2;b7MOVoDbPJZhm!TGGlp$$-I4WeUA-g7;psZ_xNK`po*f*a<@1uuU>C$Il zc9OAwN=lG=SCvkDFkUo|>_dl#mJsJO41@2E6&26WLvx{{fBr%@oL_eWnkp2)MU=wQ zZcK+`b1h-fm@#1Ocp2(0I>I{)!q=aUU%g7CL2gI zwQWa9l?$bR46^Xvi%e!49>ktJc}s3>Fo!R-TD0fHFo7SLL35fNsJEsY9T1$kS~CQX zKtv-vdixbTng__|J%J#%A%mHFO-7$r97>cF5!dOu{OyKAcx2H;&>b2~I%i+UEuZhB zsIZ9N{y>iF)|mv?wxtVm;u4rTbOTCHHRVbja`Cf66$!N}#gal5K5{|_Uap^quMTe_ zM@*IBhMhQR-PVk;exJ}eQ0U7!UB{%QUy1DATrNRb5ia{#^Vh6SvSm8{;Qv)!bf9$; zgohu%ZM&tw{*nm%G!*Dop_6edCLXiYlc2xt4VYyq@#hyO!Qi=XhIqz%$Fd9)Bk-@!6a9^z3y<7Ru@83BAoG+N?$Ed|c%|G# z&!&xn_SMTU;72@m&z;2ET`qwB{`pi=lW{=<3GBkKNp$_3!@`c_2+q297=MK$I<31z zevX`irqk5f5PFl0J-Cxo7IJj%M~`BEcp8)*%fRW)x^&dh6tEeuMWtna!tP&^H2sJV z-P*AT9^KK0!B-cF}5*g^1G5y&Y zU}mgM|9Q-0UYDX+W``~;O{|6z-iD4_Ajse2W6Ma-+1-|k8_yH(WA9m~_eie2<0n4!InPYu^T}<~ zTG+lqni>}fJH;1$ka1m)>NXvO?4E5T#lxFxy89Bg?x66?=PLYiJ_fg;wrqPnidgmU>AnVR|BXTmRs^>g<3Xc*sTG)`fh*d9*&#p0=pvL7+()oZq&MiJe&k zT&*IFOi-lr<4oWTq|v^KHwjQE##Z8Xgz?Ft?02bM?ET>GiYV{f>E} zhrgT3=F~O7t&xi%_hLSX1^yuhU+3^aydvm(6yWoHv*1_uc{1cU!Kz0_aJwZ>^i#45 zUgL4xo_rI&MulMRwzd4PY2X~E@lPd&baY3YQbjO9TPEY z@qi9hb$6iU^ZSTJc^heNv!~Ls+i8HMJm2dkhW5*6!d2s;n9x5Jx_-cW6!OaeGD6+bQX3Yn&<+d}nj*kLX79|RixKqA1Sd3nHodDC? zZ{X{!3%K;&7UE;K38Hn|*j=S!j-R`TSIL*AH0T2 zU^Qn0QJ2-BZ(9j#(ssqV?-|S>DFV$!OF%cY;6?a+OgI=yD_%7TW_ZOv`eqyAAKEL8KZ#mqj3@5zV!M+clCa`5=L~iqU zz*MDtrXwH6;O6=FY1Mg*lmv(90;gD{dteEPjDc{H$~$6A7?~TLu_!J z$zN{4^%@Zt{I<*%KM2Zw<;?X=4x94CiD*m@gQRc%xOS%&#^jwv+m(6b;`t8IZ!K|V z6z)O%r$wVWE+ld9%*p2VIneQ9GnDL{1)`!HV(MB-4jtA<)d%V5;Rk%B*HWI}=?rS` zZi1oo6_S1M6s(MyNO!8-0Gqb8SpQ@g+R8^`mzuztbL=5*4GrA1HAzr-V)DY4L4ESp)@?UcpkRwe@@glG?3*}LrCj&!AX1Y3_3^sWQW(~5~`ApK6bfSQ#+P^ z9V$(0rkuiJ6K{y~uY`(_MN~IxKGb&wfXT~n))n@bteJWV!?WUX^uR;%>`)Sb8Gx@| z3~o7l5e(|fi0qh)@SswH4rbay&et@M)D!ZCIVlj-CU`%d3iGvohIz{cSJ3`4cE|56 z2HpP-+Oh{BTs?{#Id>W-q40*wh;qac#}<>yU1Fpm`49`}H^P4_yy3ZjCTJyHV>Zb@ zK=Pd!TeDP)?;4kdGL{WEqC%V>C+|v{r|83|&LKGc*8!0`_ntWhAEGiTN+df{7x}OW zxGi%Wd6ncs44iIa%TY%f`AnBi+BSl^AMv2G6vgNRdV*G_Y@j9&Lg3A(w{WSjhp8W6 zl>6enc^HacfaUa}bhGMbUxeu~uuhWl|g8H{6`o6zIA60=8>}QRQbnc;34V zEk2KDTTDg>F2pD0Tjob`DTTX5nqNln3D=L{{1XSD$9X+8`ksU6$VTq-z&tYPUIz=- z8-et68kTfyW9x5)(z`G35hH~wxH$Pdc*9ll`OI28f9yD!G*OtXO}zk0IgdbUg}Sf< zv7y7tS3&THFpjOfH}k|NYrTwU%nqNU_gw!*@_*Sd`7#y$plhbBANAMFUE9D?xZx zDpRzH!@96<#JlejsXlH%rzrWc<(FecF6&H$-FGznmEqyJu{}FdY>x9kF2t~{op|7O z0Q$}xjw!n3%(gO7IIHIh+1l$|oQpX;&=9y{vus%LjPC@xY+z&8coGzIimXZ#vF}4v zu=u?-f8>J{e%Dn5#pAK8&o+U`?kuvLs`ecA&B!M=06Ezl_$$HnG8z! zTa$X>|MX;2A!@%6=MDGX5!m)ZZw8DZO1zdt4t#~^oL(r~RY2c1%z`D?58$-6^`M$r z&AsY*grjtwp@r9=%DcK@nd?8;zEU6hO3Go$k?mYtr7p^aFB82j)Cc=DXV|0n6Zp(a z4)`Xl5vRUW;xW1r^8=bN@vt2K0)BFvYHr|YB@KSG?H|nEFUx0LUd`8rC(;C`4Epbn zJLvG;#JEHXRy$-siDnF)7MMppOaH>}Xg~UCK|U>6wi)I}eGoZzq{5c;Czi$Pmh`Q9 zA$+Vbpsj}8Y=*`~Ouj!7=g(EgUSU@H)n6L#eE7sx1f}Bh3~BJ{G$RcG5$u;!Ith1Y zkTz)qui|wUo0IRcn`u|^ey$^Kb?Iii@0-x&&1>jtFK4>+c_X~?oJYg{9j3C{!>uMy z;AoTl1sdq>L|bLT=qBGZR6c&Cy#B7hnK>_y>7MbdcvcRGO)Y>4{kI`dejK)c8;Qq0 z^o{qB>gvr7$p%i2QHPmQ2o#>x@Z zafuKR<_(#Pj&bXQW{V>4211uu0E+3|rr*@-sn0q`8k^|}jvF%}X^0=Zbly$ZsE1Pv z#iL*uc^3B7pMcD=wsQJNioYLu6z69~v+_`0ByH1;`}t}-`aq1W{4oX`3?4v4t*kKb zTmkQdyGm>89~{AZafy3U(MWY4+i@olRfDHPWDaJY_l1AoAob^HYUdWk z$3Jet5KmWZelLc;b&WV;17h@aGyd;tZ#?q5N3^Uj4F8@`hev6O%*t4VhU;Ci{;nAI z#*4$7)JtscQ)zy9MH1t}J(*gispX9LOs>6!p}8nW^e`*3~> zHdS1=e0S{>xO~#a?%c&BS2PJ`Z%kyC8*`vV&J^1L|i5iQ5by>rogR2BBj8jBKPH%Os^CWNP1L0#{$a`%im z=oK)6&if=}lT}~gWM^-jbs-83CR(9`%xHdvUIG3rcqQb3&A@lzS`zivNAOK((?8$L z@bV`;j8PtgGCzdB(aTqH(>>2!-Cu!L7RFC!-{1o2xiI6|T5t8-# z%U%8)hKYiTcKD2wP~qhW0a@C7Z-NKzG2M+4KlcM2dxLw@!=rSg1Rqot&O&50`Axye zxI*F&i|UJI`2y3#^4CGIeiMi;4;SN#g{O#T?+3`>B%vVcI(wY9P&Bc44>}!e#4m5o zFt?MftaPLomm6q^sxn46{QWA?;g*?nrw4-E1b=cmREAcEokzD-()_#2n}yE6UcCEx z1fO`tp1;-Ui?_r>EKlf%kPl{4F{A%0Ya4$J-x^Dj7u^S;)8jo{;cMZUpAOys+K&!a zO@w<9QFvm-6?|1U0Zl8**x!80o0s0jej{bnG`Pd9`?eaYTh?Om^h4;7_5z)e;jc(1 zT;(Ie!(R{K<{U%*N4-BE{ZoQ2&H`VEoBxED}$_9jxZqZse^ni%ii!$!3H#Xze#yli}n>}hg?*QeV! zd=N&WcPo*Ku3kvql8sePp=c=|%KS2;aaK_UWN*EUS2HE~`sHD~o>mt&{WykBofG&u z^$Ykf+m3=u>JY4y9YqVSNWl4Hhv31(Q}B2AA$&Y(F>X2-MdT8k`tE_|OBRETOe$nPQbm%!m%I_qk>1k6ci!Hf+6vt=WAv3rodsf#i!xFc@gXtin@B=lC2^riU!E9>^il z11DkKTXpa}kp}aI7?7oEj>Y+3*pS#-WL5vKx@vf5Q z#jQ(Wk?BS#TisFq+e%)TkqFEN>ma<{BycQ(%gD$5Z%NdFRM41djOhuxdE>6*kTUy_ zz!rKf>>^50Yeg7-v%d%hF48b(f1BvYMQf;Hiq!LaC>+mL!xUE&{^_@+WY*LjaBr_* zobePmR!So7*>^{*+t`IR_YUCQ({D)2@e@!lA4i0zkaOLB0?fxb;knGM z{IUz{`NL^XaC1xx9`Co{k2k6C8;#9ic+xqjYj8yiD>t&QOO7_DtHJl(C!ku%6zQ}N zq$ypNpKfA@*DC}^bYK^Gpxq+erH4V)yrmeIuMZ!EzVj5FP^j532YvGzMAywV@pIi; zjJoHK&Uah6`Q9ZA#|^_xX4Rs7GSXOn(+C@iJ>hD{SSr=;MT|N|@@uXRsTy$;0$*MS zJ$-x3ubV?J&E85Y3un@O){gZ2x?*;@BOYAtmO_T726P1lLU%z4F;h>4#A;udv{wrR~q*=FZn()bbH-a#LXNs3CdQszhG|UW8?`$%6Os zPsse++HgPN9%xxzC3kA1s77lf`UVyY-%V~|txaJ>>LA}ktVzZN@xhpP?;9~INx?nu z?m?0b&u_ZA2&>F1(d6(C9HA463iIuWq=6*YrX|Cj_V(bJD|5-Gw|X#kI?spiiNOOh zf%NC88({Z+7no@3^8L{+kmz|2v-RTW6_0yxXYtIk&pjOM(FF*a6b=q^iU<_*_u;88 z;G<9oD~+}Z&Id%HZl(g2P>6!3X-4pBwi3kaUWRU|wd|0K6!3d6y5MponPa^aW)=m& zB)u(yq%AT`!zI~j|GYHTadj8^t$2pfN`J9WSsB~3^oXR^Q9QIyjkT}I29Y{>ChX9_ z@mhU&=4BW9A2nhgzfHKC_EOx5N=^7yZ44=VhBBCh5pCn~==1YFKYO_ce?>>p+7V9R ztvm|o>`el*Nz>rHnH-M!*#L7sEr(ThChY#I2ym6rhfmIa(Cohh&J|vUPXmjgIY^gg z1gwUzb<3dmXeD_+(+!ngodhK=5CV(@M0HLEOfPu?0k(?NGHfFVrfH*+lQh-!$wO)| zo*jMvozxvT#`hThVbJAdbg`)=7CK7*=kb#N;>wuoyAu#$F#})DD8vk&ed43|1E`q| zp4A!<{@(r;u2)7f$gd_jAH~V)>6tL%pEW&j5y+Fqc7&- z&3$KBlPQ0`a8E#XB$aGd%)-nvS*~H;EzaFi8)dB4LpmLegAaANmj^a;<4&)`bNBD# z+9 zacLzJc@+WkytCN0%Q`qg5DWQc?^%6?3~O5Oo@JeHCiz1)@ck(7{T}*DzUYjjN~Z37 z76x%<(N6SF(q-Y>GigWoRveq7!i|4WjUT&OzuY>EV2 zxgbxK{tUvV)w5`YK?Zf_d8HPyMIaj7Ch(E07sOu-V%MMlffJ`h!R%Zi?_t*h-^fB( z@0TvfcG*Cez6hpY{c-@Ck3q?b(=fd<9Lnw~(?&~Kx_KlWtL<=6+IT8 z5u9n_{guXY)Op-!^bXvD!EIs~QD03wM%=)?2b-{|U=F<4ppOH$0@&=W)dC4+Y3}vs zRoLrkN^dXN09waigTA%}L|VEN(_e!^!MSQ2j4UIM8vY7|hr2-bhY~fJD+EKMJ5XQy z%S!BW9NFvjmPC1tg;8FmaN0^p97XL(`khe%TL>ZFj%5oTKi&eHPu@qr4qMdJIL$Ur zNn-bO=R^LxXP|%k3VL00!=L>TWIQTiWp6M}sPd&66%|zKp(=ea6+yRV9#?zgF!c~7 zP_k(%eu$dHrE*n_p1v>;p{wX-EVos-wDTf`ACm`PJ7#%%(7_4+uxynvy(i*fL)an}9k34>bHr<_mHK>PI z+H#=aeILwH%iyuma@J8>%odkw!VcO>I)cxULoQnpjK6~HP%I3c4P+wqdBo`ZDt3LX zDM?kc!LXooa@ILfpwV=mxo&l%GI#CZ61_yng>Xy3mSi)O=#^sWGS>JdZz}W;R}dTiPSbANgZI4G3nQj_!z(ZT zcQ&q_T@U6RG#+|b{WTKuKk{x3wQv~Hyd#*pF&D+jT~tk7NxVM!;@u@jgk<;{-rBwz z^F1cBgI7+Ycv2o~9uX(FZop^GbOLbG9Nvv7`x-6!o)c< zxMMk+xSl1?!RdV;lR7j47kj_J>k|esL~0yMS^AnR8&idb@3r|p!#%b_w}PcMBh$o> zY-YSEr@QJWTew7p>+Ms-`9Cic8KoJh9#<_$P9|VwbU;`Z76wbhKd=^|2(__`reDu} zg;>?y5Vh?X4SQAy&wbmW{Q6dwV_PaPm;VHAGjiGVRoVDYbqP#n56P%U*PvI&0QSEA z30JRgB73YWdCt{Q+{ynRyik^b%KAxI#H0mblag^ zWdwUSSzFL~YBL`Fz`$$s1FSf!$#r(^Ci~8ckn6)i%y-Uww7FY{q1%nQr2a^%acDJ- zhDK2Nld@SoqTedAiUiiLpFU2A~#!) z!NEnrxVT%LQ;cng5glUS7+MFRe)Vwwtbm%YEup(cM}ndq5pJt^BiL>;4J>o($wcq> z;Bapa_GkRSS(k3(!)<<0lRlSQXr#q?_nm^t-me74Gb&kf(rK$xJ8RHij1vw#2x3J? z#(-<^59T7J!{(dv{C+32O#Ln#=`~3ATyKS*l_}5Yy^Ng+u>^5G4jPxc@IuT0NO;N{B9{&UpT1{{+ z&7EGXS%|qd?@--O6>UGqU{K=U3Qc^gl<7;tx8F0&uydH9F%&)q#k&ep__V;3)Q!Tcf`GP7se0TPc3U}Vyit|kP1R`y5@L*p)^yIfO z>sPYD>#sY*9Vd2c}HUXlc_z<1Dk zb$~72v>bQVkEgP0wV}U8gS+EzFQl4t*hTHjq$cm1Rr{t&yp*<%o3u)qliwDGb5&R3 znr1~#u~nX1w6ImU1cw>yor}WO={VnOt+3)&v1O622=!1Og~o9*R#H{tL1%TKrPszK zR*TN4)2>DN*y3G^J2-KCJ(nw~ zgfDxS+)@E0S;` zQL~0f!L0YB{M>tdaOoZ9#oD8DrzNbG7lHRy;Y_g2h12|34HGrTeb7@=;;BCc-0UQsztY1CzsZvGOs zFCHSN&+zwz!U*y~AWmNncMt`WWrC^Z{!q0w6#DC1(XUwvSG)aWDZ_s3;wXF6Zxtzj z`{{1E!YUCQyK@vK1wF+PPu{_m(M!0$<@ww_VJuWx`-9l-Ch!bN73!|B1nQR#4}OM{ zUw5}~-AAMFRP7CRC63Q0``o|D-INKQWQT`lxdut`AY_^l~s1YrqXTJomF>I&KLc#zn`T!q}}1f*jilc4c`Q zOdK8oH|Lte=)KjL@$d{;^~4;f*tfHdJmG!v%>~?dYD9L7p3aqQRe(!kA(*&tJI8LD za<>bc!Qp)>#HnYofOC&vgVqB$mXiyvvyah~bK#J1rUm6IlR@=%Fz%hU1M25{(5L5W z;Jl>**r7WyheH(+x*`8#>&3EK+@OXI4?}Xb`ZeYQp_3X?_{%kia#?R4X zgr)}^@kX}{{xRHzxsjXkbk%WI+_6a@o01RD;+*LYnd{`Uw-M#xl_dL7F1YXv|63tJ zg4WZQQGZB?i$hl7)zy4ItXPEmxAGQutSuP2Lq-s@76!{&o)C-P1tg&>Uf5rD5Zm}U z>nOa0EweO;^>isJH}5gnTpx|&Zq5Tw-OV^_e>NOB@NH!;y7yd$AONRKUevXpWR4qBP-emQF~DuE(&Q! z)!KRJ8q|duPW;?V*$VaDT9~pR0l%&0XKE*dv3?*LCDvy!-RCCUjaIoeRiRj2>V(QNT@br!Gu&tufogvO-!vD4>s%Q+^x2U{^z-=w!!j~{%s5)P z^&NG55|7FfVsV_GHw@bsX zS@skj0DW=sE-3C*1@-fL!FK9NKIiHHU%EySG`a#x-SvViuX~8;9(hpfy#>!5w!;3@ z=cLVfB3OzRLZ@RB&ma@UM1eS$mN-jT^~#$l56xmHb@Va&MKlDAiNL3g$-*It3HU0k zLeQAN4D~-^VZho53`ducUa)9sH^|X6WqpT4aQPd1T#_4!KiBx+V1Eir+;@&l zbvy!Bo{Q6})IrEN)<=}a`;epl)l4&DIa#?^o_l3Af_txa5~H4obH~)QxHpwg;Eu6B zxAL;5;Cx{O%shRC&2&Af z>tnI!3xsE#wvjcDg8=Tjp$oWoVJS7I0cXSeH4)WZB!x1QGy<}$4^V4iNm0!M`g5ejOQoN6h+hdE3Q{Lh% zwgN39&ohN%5_oKbF$7#+1DEbUBWK-egsF}H(5U<>`UfbI;ZeLg*LMjVpJ)fHO9ZP$ za^a!mesX2XH>lfj9S`P4^LHg9ND4~Cd4nfmKUc$MHyuXZ^U|DT*eEvUwH6E*50+Ps zC_=fCCZ_Dx$@iV#lkpA*$RD*B7BJx;Y5p^dbAIz(&>0!PvkK;c_@x%BZg)d?_C*?N zJfDJ5sT}4%R^(iLjakGHf~VjHXh9zdx>JKoHF*DPs|g)=c2fB2rUs`|&?FfDYZVxL zm=E&2mvXIXJ|u6QL65H6&ehi|)9m`Wu>SF6xTFzfk6Y?BtmUA1QA zBh65dcnlJLM1V`sN2|5-&ae%|a@^g*XjVA*qkN9l94Iuo0Zlry!Fvwhsrk1zW<; z%RB8s<9{!k6^?_3Y&yi=QAY7$gy;{Gxi6Cgv8qsw1TC?^Ow}>m>*rnEQ47>;lHCg{xH3!&aiVXU|b#4Rob2baybGOn0}+UAj63bRP#(q`;eGU3MF zyU#p6q{6$@Bo=tzidcwm0Ga9{5~3T}jzXjFi&_35bBvDM4ig)b@WiE?q$;K!;wFVbx9?WA>SGk)8f59L2i-WuNC%I; zlLO(jbW&C$i>p3{SC0;^PPph(TD06)46-@k>&9552^I^#S<^qa+xCYapAHpDETTA3$C2y zzgtteOHSWl_t$2@<;K}CX=EwMGp%NF`{$9amYrDbUQ86_P54~N9P)ZS&$!7C;Cq@H z+~b6w%wpDDT+%xSo(9YS$y815o%ZhXPm>bZEvLiOGj}tF{?Nwl7O_^n|1z=k+ff*^ z;xK#^DHE*OW{7^3GW4`#8l75|UViv`4^b=+ByTsXpvtCWti#=>yr=sttXEl%GNmHK z#n zPV;#umnsb5+eF*d9dXNAV{Yg_ptt(Nq2`VZmHqO#{M3pxe4(%pb<&pNDT`dJn-M|! zx+Zh;>eI+_k2&a59*&iNx8SDlN4d|U(y*BKu1-{+LW1NG!oME@o0kssZ4u8~zHE+< z)>R_S`Ge~||3j_jQ)Jx(AyZ{WNG>qK(;m12J=b8yIJ%z!ces$nCjF_Q;eIe0z)4El}c)yeuP|zINeVrBt$? zcM*N9dPHQxl;~=QKrmAK2DpM>Uo9tdL;~iBEP>YGO?y}45Ybl60KzO9albw%RdKo z<00jt7FiKL0t#&};k<1taNiU$ZmH^DJg8sF`8F%!9nTg#EBcX$wI5~zkrU*1d>cs* zv}KAsLuYFFVjSum1HS4eNDgXnwi6Rzj-?r_a*}~V$ILO+@hXJQOrug3&&l5*8JHPR zh%J1l%FyQ~q^l$E2KcL5r^67EYTR`bc7JBeT*hBgP36Kz{T~@a)isb&+3jXUYPu{mnDP zq-*5FPjPNscnB`~9$Y^1P(2!oMzW)qu)jDB_wpl-NFR;;`N{4ASVraZL(b zVnQ8S*ZZ+CVPdeqFax}jI`@ z{vQI!tB9S?}^u7%M``*E({O32~=z0M#r8dF_|`>b=x*4Nt3R-rLdNIvL5a~nVWGz^=&q~8H4|fDfAkD`@LfDqgdfHDgGMA$ z#tBvqso|bz1)TTp44hMn0x_dyIM~i}Uci8DRjCm~Jb#Zb&o1R`7xoLEHhr_&-^hWU zof=N)FhZvoaWFlZ1gOHr>lksa%Ko;(K`(PLFVs?QNFU;z{<1 z#FNGsuC(R;UV3QuWwzYOjp$p>q}#WqQt{le0(!EN{PtOazlIm#?cYzq*=iZps%*f@ zG&SKu*L)JS=rSsAt-!H&__^Eh`*3K%0NFA&lvSwdW8DN(Vf?~wsD2eBlr-A{AvNt* zCv8=rJ&5P5i!Emdx87w=MfcfC6FcUdD#rboZGz&4m+^j68kjpdvD7uSAYWEWT6Y!0 zis28$%)5`pOw(oKR+?xJI?Rld)nVBdItR=khrktx;UxTHa0>VK9EIPnHetbh5zgf# z&v(3b3N#FTxY)Q6cs2J%dBs|3?1?QQakPZAKE8_!>w|A6>dL}09aGr)dItHkR~Lhn zo5|I2Rj4384z9av2$E;=XZr&Ucx78ZjNg0+6^DAUOnViWeK1Ca&=2H^bQDG`kQT1| z{EpeI+D4N9t)(06RzPc_8$?EeEGE;E%$%Wd~ zBv|2<4L5d~p`!H?v`;&Oo(K4B#g`S_N@%+Cy?2cS1=&m>iBU;;0y6puPuJpl2QLR|| z#{$0>MKPzo+bmW@l&2rG2&4aX1N$2VJn|)Jhh-#fA4Up24?j%}k_3ZyL}O zN$rrn%$U4z=@S|!WK*BKGCJ+BA2&NTlC!%m$vLf(!q02&;?jpMTw}o`sC3X_aiNl2 ziS}=d+_RrFXzPKe)NG({mY{~SBFuW0$tIimqVlwLxNF=`7?@B(THrd*!`%-eK7T}1WtKmniLwp2jy|N2y{u{{^ zw+7?giid2~H#f}I`idHR8b~`om+oq;0k*Fii|>%~e}k^{pVoEQGe(|gBK(Hw6^t#v z5QWF)MUq9IcEDO+dm8&!z~8?jVf|M=$Kfon6uDSIRxg=?PXZV=PB=ww$oInsvWYZh zJj1d-i|9Db3J__S27eBIg!54sA-w+()jqrsQU(r@67}a~M|vP=IpqmjpB6)R${JWz z-Y2xG8;3;k1RV9;4k1<7;8m4Bm+RDxMKSx)Eog#pNtcc=r??AS7pFn-S3}lSIf`@# zN3i#4cCgVp5mct>fsMB=l^m90QCJTvf6S%r9%@uIeiBnv{m%CR72xQ+O2{s0hB3!? z!>Etm@byw!`JeBEd`)O0!I6A+u4Na`83}{hi}}zv*Z~$p)^s>L1b*(4Cr#;_@N8!c z3u^9V$~({Cnb~IK+_8tm*@w`wHM6Mvv~sw#NP%wKtc#DUvoS{y#Bp==;mfMiEWpnY zZG{4I%uw)UGYj^r%HS=}ljO&{5nTD{5P05j z0UMuBU`mg^Vqe5aZuNW@oI5KTc6eFQPaV~erE?AzD?cE9O|!AaFd2MQILYC%l8HYpsqg_C;4 zxDs!K%8nOSzJ)U}?9EEzD&8Y#jSgT}?oEO9Wm53(S1+l%pn@-2BUsU3JZkul<$|=` zaJqvc?4K|iWzMM5Q`#?Js*N4&{_q!kFHHmWTm?Z{(s|Nq83c32T5wo+vB2KZC2V* z59NbdSR7(ZV8VVPX5_)7-L;5_mFZKo;a}zB?JE$z?5C2*Ev+S8x++jUGkS#KXhhw9OqV;sB?(AhxI%eXx@EiEJ!Hnz9*h3EQ zxqx*cHv~Qlno5@+IR&;-F~aiU6f%%L3h&u`C9-<FPZeT2cA0A!N>1Q=+jwY?BDe&3`~EC&UqF1lp3JCXax?u z=oj3QG{m%d*Mut>Ljvv9Neo~pN07vx+6*`5Rcq1@XuEU`3#>`HFMzY+svZDR+2U*5@% zPU|EKvP#L3jar=9_BNvaU>)c+tq0qgKsy4 ztoE|1e6$yO{U(tU%0I}tPHk@P>22Ju?|g53OtR23@QiTV&QH91Iu*n{!q8bH3-xYv z3ie!U0SkvlW~5(Fz9urZbN^vtGwlp&48~#Hv3jsCQKrZEo!7J+Mna$jbgU=06W#VL%vHc@>;rFc)UJdJ=2U+ z;#_gFpD8qqvVd>TtoXT#9@ks24qsi$;>;Y4@ISrL@FIO8$t?*5>4%0i{{1ApV0e#7 z<*or{!82&WopD~>C6xKG0cNinf{O+`v;7m#U8su#n!h3js2IM;JExtCf##_yT4Y4E1i*vD}5hfWWpx&{5ukBtBzx`Vm5;J2+nJ!F1Oe( zmhIdrM~gh9c<*c>ai1u|_1I71oHzb0mp-XLeak)QI+Jbu4vEmp502d6@M?70c$Gwo zjHfSyV`1haGp^pbmUt>R3pcN+71lf0z?E7*q|+{tiX*Ey;gMOwijo)5<1~(VzTCq$ zj7o>lNgJ&sHdsQ=$^Bq-pajAf4#FPI30!&OGv*Xm0Z)F5K#zkgB*Zkq@F^=UYqTiL zud^V#D@CD5g!j`Q;Rl0SBdJKaF6ibKqw9rJ7;Y1T-qw>ixBmSg`oSM0lZ?1_A4Ame zd?+|v)h9es#xsoWd$Yd}R^SA_6YdgYgHrY~Z0&Fj3p^r)nY~^3qWmP@c_YbfUXzUN zd!@M6wWfG){A3Jzt&RSN_ONx2TrffVE$;m&#Ab!0a`IvklY40o(G61MUXm(kq?}-@ z&2oi>VLvdq>jMjZ91TNv+&GhnK=gcFzsi3v zf@ruan#T-wJ!5tyb8%_GVXXb6!42~F+(@0XtR#9db55GaC1j*=V+_N|wB!5PHjyyY zFS^1GMe5`4Q!TK4!U(E*xRXUslcQ3K-sJL(QM5_+G`V1OmfWnWXUF+{wAzJPaM@ri zXn4MYpgLVWO5E-Uen##%0{rWmi(hO6G`vd&C-!H!9MK zVF4L`briSmsuZVvc^bBPXhY#T56G+&q3<6k!+XtHWU_3!&9NUB8LQ4Z?{TGf3>stgNtCR7_0e5WJ zHl1WGrC@hu7hS$9oi$HVMpegBGAsTJXg>C#At=kmo9E!8@)s~Ce3mfQdQecV!E<&* zHp9%crSypUJrFmOqSiBViRJbWP%SYVojs33$QE6w*>w!g3X{pAopETryO?&Z9!tYB zuRw}XsIckH4;T!Hp}~8*vHh+P*Bv{IYfn_Mm)4Al?fHk#O!7f5!-|gGcun|dY&8~q z*oY+$Wa+oe<|b)%B6op9_42V?<pslRgkXF zGX#n~>E?;f5FGyB3-0Dw78?mG4hJiej$g_?|dFgADP~$Bv&o&$lcYQZD|0 zUK4ZbAo(1+Rf2`SEu+|?fg)j-^Z*WfRHAoz78Hz1#-U~7nBDn77IpXo>zI|r@<)%s z(X!Iqy;UQq;^jc<=QV z?{HF~2nN4)}!+Moa?tU_M9G zyNc$;{}U!CjNm>u@z0OaN%qJr606*K7DZ?t&_fa+wdDlQ&dP^@(WeCCex5}!(U+|7 zS`J(st4H=^9A#_oC6tSs8&dniM)KeMUoiiSI*Gd^iy?c1LG5WKW~l99r@H>JpmXz3 zBVL01_jLp(-=~LR%H>wWOU@H7Pa!Iq1R_kjjT^dF!*Hz*X6anPqA}s5G$j|Rq|d;R z))}6|#sn)QLfOLwYcMXSkGNe`fXc!*tZG&yTCORBu$Fxw2)}?vJ__)v^9Cx4O28E7 z&!E$i4PmqUaR0w-TyUkE{WebFzYB6)`L1Q0p4mJ;D;(J{KU>_dzvl8ef;CiWg| z?y7}ik_}6qh|wEk*HA4dRS;Wrhlp)WWXAbNga=l45?5#^Nxav0p*X^bgRj_J!z7`j z#7M4p{F~dISt?*L!wh2f9f5t_b)Zq5#Z>%v0v)vC)Mc_!{F){?@0urEe)JLU>zT|+ z9gwH9FQ}5A^PezVjbE%URE+BV_W;6jPLhNrvMA2;(`JNFlD6hNxHX+38ap09n@F#~ z`EeY)wVN%R>Rd=5D>e{fIc13}D@x7_~hexpaN-RU`1@ouwaOYr zLpNk%gxzRKQ}ODp$LMzaB(8#m__+TSHkgQT zJ$)mfnD>@#GhU41A0sf+Ziv0VJq;T|Twzw*Ng`1p%B|jC#zHbS(!~01n7n8onq1?Z zyJ{7%H%Ir`0SstC0S>@s*ZsE?c({bL&Q~2z?4R;d$3PP9Lu^+M(Y*hCsl(3t@ zeOBrti|h*pcrF$Py-(o0PGuVTFs`?aIr@8HVQdibteLt#IPa1_4tk#+SOY zF?O{Mx4cr7J^sP_K4i-yENz~xxss1itvy36W01jfJr%bYY$z(F9xep)68FA}_b`gtd zxnRGjo*9Juk)W;8T$Wgn)rci7RJI|I9p6)q@~x5>A025Gm)b&n$19Q`@VM$mekDc2&WRt%)h&+&-woH{F|ACNIL{bh_AzEO|2Y;PN)c>3=Z?T=;Zh1oc}miYEMx05{WGB71Hw%{(kk(?`prsrWtC;Nu7}BWJ*>JS&0R&{dYH zo`uJIHSqDc$LveAKMs_X@=lpsCz`Llg9 z*Do)@cD$2jzgIYtsHx>d^>PxNt(zlQ){}(+k7sa`-H!+omK+t{e(TA3mo&2GX@yw4 zeI(C64QA529B6LVQv5eEkz5;-4AWGLQ7%gfH~;(!6NJI|{VyY*i*;D?^i)B8l^UyP z=l8m|vM@m+U10hm9{yhAGr~p=s3uX%vumahtCqzO{h}0X`UeHClSZPK^CUduas{m- z{n5wJf&G1Mjz>b1af(ziN+xCD9#t!B9||IqB7B4a4o?tF^61xy!_=p{19sQT(G_dE z!DD8!pxSI3X8$&$_x4|awxM0r`qop}-kJ-&;i0&!;UZ?GT*4RDGe}3(0AtOsal)~; z#Q4()x)=iK=OQQCJnlOb?6ar$+|1}}EoVp*C{s49NB2+4roOAS=+Mk3xLx*~Ku_`{ zJcy_Tv7jgL>z)gXuRJK&qZ|gIHfxygZ%bU5dVmdfMOl29GQ?UsQXw{JkoEc`GL5L8 zY(!`e%(V`HwF!JSq{)iB9$o^1`7-$S>27e1zX6Lno#|dbJ1QIB2W~GE=mAl28h+c0 zYki>2$xnPIH2f0F0t1(kVdeLTb4GGq*OSpy?K0cEf)oB6^$NqzPT`z7()c-DACVCs z19YwwyKvzt=5ANQQMr#{=9Q!H)ciQp%g$gcWtKwUnj1taQ4$X<-^8EYa@et*f7#!N z>3kM_H8*cV5?1DBm;d-Z9n)n`;+YqrcreqNG(5>9VLU%yYSVi%Qm&OnW%B28-#Wa0 zxzg&v>A~{-##*>#c@NWe>nD4!o`4(Qj+5Q1XGr%;P$KGpm1p~ zn65O$(=OAQjZzs)II@Pg9lilRAwS^Erw|+)-fuNdaXV4Y$zcii&B(0ExkUD;H<_b< z6Wr>o1c!9pnClfjC;HR`*xv+Lb>=MWbZ`KcPGQd;MPS`Qz~vOc%(weV#M6s>)-MMO zuZ_gdVLPvBlp{mAa05hd7t1$)PY}+)}L3AL8?nG31pbLUpzbp5;%v zk3TM=dE0No@#W$0D5g-5{QU-OlTo0%gW6ev-vc7YAqk`#y>n5!Sf!omGW2O}nkoOWDT#jp^RzxDM1(CEatODqExB=13XXriwX4$ z;Lth`bpQE_Ewm11xsEl&b2yIF)K%fs6A|3(RS(hIlJ7(0PvC~qvbc@Cx8dnUK;`$GlrJls(8>u*G#PHF9b%J!0Qt)N7VNClL{GF(Ycb>*zQO{

fuQ!RG|9ec0)xya=GdZ~lrrg}3<+y3N8aK*)7@aB| zh{R%dfpt%iZjdI`@L&u4q560)`~6|bk=Wa)$c=p9H_hWsqx8I&9&kY%sRgr3(#b(uPK7 zJb8Hw+a&*ZfS(?*?70)?{k~;+H!zIG0dlXSmTtMpY%}3yy>}zX7QEtU8 z=43b>A1p}0uu;FrpYL+^TE`LYPurJs`0AANo;adB)uZfVdoMRCia;E^qrcAv;WF43Ew+l zivB%%>vI)3sWSyPHh$x0M_S>-O>>zo+zi^j`yj43Rlk8EwnYFma1FbKssS<5C^LUnpoxqcugDB^&;M+q2_T`K&r^mpe?#jZ)!^ zxeT`t+6`ON&U4QIW7w1;3F}1V`J;>FnQgyPQEvAZ#?zq_j_1~!pAPip{k-l9QFTvX zdPW;)+IHXy2Kgcy=Dgn|hntXy;pGemiLw ziCtid!QEwKb*~XbNhE-cOCPQaIt4{p++2!GV+!V%;j2J@{Jukj|J#|zJ2LYEU7;h+ z-c@!4on^1+=TD}@eC>7_XcHCH|hkw|66d1{A^d{T;XXQF>uC9*k=_7Y}T^?m*f|nGdW4W2Rw6=yZ*PYLJrB=hn z8!d3s#DWQmP-MJ0Uzmxn5Zl@!3)4QoM}f{8beZV{R%g2@R7p#~sukY6+?E_rN?JhH z=QcrB(F6GXI~7|S-w_e}M>JsTVQ$Vi6W3@|b5wUza-i@KSs1?ydwxEHC~LjBhpzCshnJs)^8&n5`96cC*gB;JZJs5tY?;=In^ujJ+Dk1ucf%v;u~y z4y<`P0G%pRnA?h~3~f_{;VTIc?<)*uaY3+l=5$OMy+E%w-3FI#u9tNG1Io>q%r0A} z0gVS#FzbsIim#uJJ6hW@ta23**s`Arw>`vVH*49_Gty}ADViQ@Eu{Bq)>4jz!xIm|HW?9CTyp{Zw#)&U;WFsi43vE<1Nugj zAc*&tHd%%8OB2TV*6ZtOKX-QSoty0^~=f(QNl+IU9Y)(~@pj+-vkjxKOvBZS z-%{F`7OMh&A(kjrtI&*B|KkHQ5R}Bu=5>>l}_NvH)j%wP)BMn zIGOFpISp}VFVo7$0q}R*TFCU}F(I!Xz^&coaMP2IiJnquw$>CAr^~bA?_Tp~sP|Di zUxtCnnQ+VhG%R|o3av9|G41L>%-Ewi^GB!m&_~^mpeJ_&Y?>%ZjI$NMtty*lbDq-3 z3}uk)R%Km&N3iKO5AlMw5SiAW3Ka`?qHx1%&lnqu97gjVT(K))_0H2 z;`XQ&cR7TloNZ5Ni>G4Vaxb;gCm&~9xJ_Gi(8L8UkiF;p8hPBJXvB6cIY?YqC+WTCjJMZ<7 zc0GI03{3^iEK$~Xr44&z#S3`u*aCu&bs3fH=b)@Ng8nsQG)K#iUaQmP4TM#pw89{D zv)KsF4aP9Lird}hnBe~GHP}#Y4d-~pv|(2gj%I#@h*uY2t7kPxG^@ec@w2?uWBMTA z_?P6}afEx^E=!^~21JI`anB(MxLEiU(|t9_`@fbrn7smC_#OQ6}zvEOTa; zKRGk&ImB1Khima6@K5O^^Y!jt#tFTcvxYKE{!IZ!w<(c*BmLiwK;D*{5I?I9(*3#2>4PK8-is$;<}-j&tpWIcrH#JmxDPIWa=>s^ zDZU|_QGS;Y)AGrNeey5^Q#YJugtvZ1y>nZc9*u{v^`;?_-6z1j{7}RE)trYJ(><8a z#sgGPFAtRuP6vm<#cZ*{Njl@vEA)7>g4++ok>e5*A-FsNqGCHp)us=$WqBdmY!zT~ z^7FuRuQVn(7h_1~L{zVx$voVy#Yi0(h5jec!Gm{&UM%aulP6bW_l6|cQSgUOuNCB6 ze{1oWEZ37vs-Yw6TQOxpJ*sZf!4HX7=zk*u*fhxxcgn`$>L;hre_W6C(v^hh&;Bq) zDFwu@nnCSdFOZ|v)P7SRztU8JrQEZfkx*twzQ3hfgWQf`ESWhF;my32?gccEpnkQE zjNIyF%;CUba?+|2{``3dox$1UO*jB-mZZChG-F>F2Qq?=%&B{OaGrE6CIzns++mH1 zOP*7M<%Xd6?J>Ofi^7?{&h)79Nk-+X5>vSMF$9-*z#-|IP=4$tod35OG9qtM#o6^_ z;F%Q^7M=zzsZw}Q*T)O84uMyDbLr>h@~nh)5{*Bf4--y>FiuaS8L638%)9|TrowPJ z<0zcPd}D7CSD8vsJ$ee($b2U1wwLkDyHwy$uxHt~|4``q2{dv8!NyZzE;h&h>0|QNi_t><=iu&+Zq3RAIZeh47ej+MkSgvY5f!h z5cSzbwY%@4Q|}19SD8XpS8XGq6{}FRPzZ-6Eo8d=r!Z!lq?zhPf{eh~6nJm33?@E^ zgcifK^je8I_A3@cUQ{b>OqmD)Q3dq5da3#KZgG^Ewv!*P9*uFA9I&J393DBf9PgU| zLf=~~zU_o<_v%oyrXS5hRrvDa`K?zWYon=Ab3tIHnzu^Ce2 z*S^nm?RRZnM8OcX+j<+%f2p9O-xoowQWkGY=@ZiaXPFrp7>28dq*(R-s}R3RjP=?P zLti=+qh;7OjO?nx5s@YMZn-U*-YLar+I@t-;~ah5FAh@82_)8XFXn9=1dsfD#xAsy zH+z#Zq%CMCS0iRHnN1;#jq7YC@zz5AW|5nGbLCOkmwpa3c4{(t$Bv-GeFlq`*U)5l zE4aMz0QQOeqx<{PAo#`!{P*-ZiYz%{z6dM%!I5ggA|+ZRljMY>lAU)zr~)!X6u2d;F8mC7OUrm|3PHF;-iN*!4#tIY)7Y-Rf~;g&KV6#cfgijq z>554avfwM>ti7ez8+#4WvjRg!GjKxEur47Yvtk``86|iA!kOrMIL>sYX_%@W&40E&B zr5j$8zCWEOi z?S3TWW1#(7FUEM?fe`>tD#QjVfkt(+3N&V&y29~*J>!V=!f^d z9`r}vW&W$t^Wf{#0lRz4VQj59#&5dbxTQE2>^&whKb*~&rK3eKsI;FKwON4wC@=|C ze=G&|-FM9UHVegD)3GGg4=vqz>~b*&l#Xr&u?CKvGxUkJFYbjGN!Lj_dkO=?_@&uEL=2;NvjeW+^p;vGDlm!7=JJu2U*@n!kI7)2-3oej zb1ue5O=6eT>#;S(l2rb4I3$j!5p3%ramI_VEO;JngIBucBPkV4jnGd!k?*Ydtt7TS4u;JL*4+rAM4OpR14+c8r%o-_}5Q@G%^2UpWWy z*1uuxZDa7YXd-3r|07-t74h(WAx1GH1Lk>|v184#FzbpmzMu>7men0FXcS}?9tmdj zKkG3?-0M==TniU?3XIJh>pPLR$mKkihvXp(@riwc!D6)>@I=q@+38C9yO$c3EMl(>zQ#>cvTKHU=egKmT`_1qC zkz{wx_QYRpRj6ID5hoqHjE_&`#H>#Ys=I^-m| z;Ze!Ay!zhUx;O;~*2|&y%`{lRF(SV3`Wv2oen!4mchU_d3>>++20XLpW1x==JHKKY z8~sB94nJK<6wfSYHzl>Q-f_QB@FK%r3aRIZOmGEe(sXuc(L2;P|AGdI%B;t^n^ddb zg5Ii{!X_Cl!@LM}HmFMp{dHt`bL{QuiG7EWeYhKaw!Fk6doyY0<0_c0@rl|8yd{E` ze7J2ti~Tk75c_+?;il3(vr*j~z5q|ayiP3*JWue+k*&%&C&>(cpgVtNrwvvX$l+Kx z51wajM4ruhJUJ-|+gIt4c_C|HYvO!*-rz0GT~Le_HM&?w)ELjU1Ujs0iYlMv=#!V1 z@raQjn{6A-ZuDD2+dpfFMb^$8|bg;{k>qg(HqWNn?Nx4A<$ZCr~ zmxl_QRTqxW4d;;`la6EcybKiDSAo&Brr28(i>DqwqC)9&@avpHbRKtNr)o2}taS{3 zg_^^jCyl(D^4Ixyl@l;aVgt%N4neoHU3j!G5@&{U{5P75AS}!~Aa@WuFIACSafEI= zvIkB(t%PnrLp1$1fV&1&;b83|{B>A_NV&S9jMxY->`NhG_g|&i+hrj&t`X&}c+7@r zIn;QnC~P_9$-l<+30^4WSK`OXr^pF`tU%)o=26^zhoZec@F?E~*Kxf+<`pai4nI?M^XIlJ6!sa-^k{{o& zNpDI+CVw2)&<%`FpDrVQSQ*_CSz>e0hp~3x_{Ba?K>uSqZPR;h=E1IoZ7CzfPD}{m z(<|}9PD4z;c^emPS&s4p;_R;=QMUMpI_uc{2(t3Nk_Pb!nEK)m+1$~Gk=sJ4@s`g} zWqOM)ZEc3G-*WsJb0T0>vl`pN@q5nS`w#1;gyM@$_N?pQWXQbx0SpAKv9+XtZq<52 zmVJ|?a?jS1D5L4TB7Kgt^W{D*2|mG3p6`X0ozt1`n%nZJxs1v& zLuTRmjr{t|V7z11f-6c#aKkaq3y?stz-5%Ed?;&_KUs-7g52KEUM)SkJ4*=imO(aUpFs;?5J`>n76kg?IUreIAhzUN892L~#0)!ZUgk!7JkCpBk+% z$hnwsRE^YR)e4Om!<`)aYHT4H-gZey!;kR_IQM}W$eB9v?(Ah4@@)to3Mt};J2UBvDIGN7 z`#BP#aSN}WRcC}PPQu9AGjOZ(1-)~k9gP2tLG!stkQVyHKYRTmulc?wP>27>)g{|m zz2HGy9G{9E$Fy*oq9%!V;9M-B^6bjm*TntN|IbPVAR`+?d#tkH!s-#Y>l_3_zaNuT zNh^uY?q+_%(+s-LVGDca$ztYSMi<;Kyh&6hhCxV;AT#jjHhy=@pfasD$r^44l3N&z z0`ey?)oCjG{YV4#{T&T6xIJ#QEq_i>3_{Yw=AVePn4Oo|3=`w_dclaxeo(zU#La06Rwf6hy2fN^eN|-98Q>s6SBK7 zz-|u`-(eJ2HKs2XXt2|_^Vs>_K$Zl>fzeiNtou<62fhdL8^)v<9@B|^Qcc(%a~QWR zPr|2Sk$m1FBeuPV!Gg8w?2_^V^o!QOw)fq5uUm@UeC{~AWuhC$gIYl5oVUSaOHIk= zP7%iN!B+Oy;XL#bT!LQ)j?tt?c|0YvPAFV9nLQ|y2@kHlA==&ZA<-uuw%`0fYpulC zyp1bpZQfEm_TRYK(QlLRR*D;*secUXc5-KxP7CJ!&mtJ(Cv)FDNv1+en00eLN=()k zbKTr`=A|?8!6y?zF!clGyzS>%{rv;oYi{uRtH1Hp#rlZcl3bKY*J9?ay#%*SUXnj4 z8Qi|?Cv+{IL=)qlpqGv_6Vaal4rZ(2kmZ?1`sxEbdRD+w;yf|w7B;B;UnyhY>po|8? z%MW(Jwc57mxJZosojJyi>!dJSm7W6IQHH8z2DotFJXTRIjHxPRSqI+>&>j+v&D-U8 zKN`j0$31sc*umn}@~@;jF<{`CUY zG+ePY<~r9;%4_`hc^A*XQVyg(I-+q!D|zYul&H!s<{OG~UP%EjT(@iroBYU@VW(?w zp6?wnxo;_KS`$E>^gmMRXc;E&%X*liG0ZQRe25$?ISJAy<8h7U0etIyAB7U0(tESC z;I)-B@4*oXT=L`!y>Dm^(bn@he$fa-4KdDUQ_|C)MryKo=KuLHN6~_+55@7s+*?X5N{EA5w(Sj9Fblf+!CJ5cP%uk~gs& zE+00-L+{?;o;hE5Lsvi2c~kDwiP`)eRKJ{HUu1ftdOJ@DMNn3>a1 z0oNOinU=(4Cg?S?VbE0#TcnAl9>ZiY4R??j*=GkyrfnpTf8->y3;&iIk7;B*}=wx+`bnOAi(*p>^#BbZZ*8lzfYJIrMOCI6@(g0K>OH3 zsJiVdEWDOSt3$=mrK*;s{F~0q{rwFNy)A{Viw}u%Sr+798sG~^i$llpO|0F6cntd> z!-$#HE0V*LAny8i8ZW%j<7MC3z>hzd1J8EF z5|!jCBAIuKp2^l{!$j>`q0o6IDg`LAR}Z+t_4G;zdlC-6-|r>5PwHT|z8}i+4id@j zpXl!Co^;C@5hiKM8F2FcMQ0_-vZjtVV3mj%d`Ox=%6`gV%bFc1Z%~PBKM=qHI!aS< zk>VcUjSaBAulIn6m=lZ(KOr5T8^PO@yWbssWWH`aLS{ebmK)9k7tynL`^8jvF7AY) zBIVHa;Q*OBONI2TNr&dddM+zs!PXZTqdVsopV=3JL@New#m#0z?CbH;8P^hDy}pAg!zi)cr29b7-m#8a=J_OxL(Hh*lSKqKEI&97mSf_`u8cL?#gU5 z)nIl6u;jc&1JzynnS_@v!x`f>!0H8IL2n|j`|5E>Op+lRg=gTg&*8X1Z57Fm`#{B9 zkK(GsE2)%}4g7mH%KJ&;;Lx>h`q^PAR_Uk$buos$PZX(_k3FWTYOx2b!@;Xcoyjya z#*rypb~WJ*Y%(oig&iv)Bx^T1Ejv%oS(n14R#Womw=i2)FpbqNXUU&)ui-~{GmiY3 zisBcoac1BsmdOjTVowgTUrRT!jg0Nz!bJihPb1+dyb0`g1Bk>~%7`i)eO7caG$%Urkjw@Wf}P;>=&nO1^! zYZW>APlv3um&LdrVwWXn@9 zw~~S84UZvw)qeKUP9^LKcflWCLnyJe7~dIIH9DtUhIQ$eA+Gh&EE(qMU%NuHL>H3z z9|bXg&PA$PIgwf3;tq)yeDH2=k@=wle~{+*txIc6P}_70?n{j_w;8rHHysV4;Xz?^ zY|$*%=J`C-w$WmCak;u6zc5lSK0wSJPr$BVBeu>F?D7W zx3ov3=FxSyII)bUp{PeJxV_f=cT3Q!{2bZ7A`;7@CD^Y`TrOySDU>?BN8vI%`gi3G zock@Ew1wQm&O5s7skvELv?_oI1b;wvkDILQ(MfFBG>(V*t_IVN8$q$@OZ;b$i)Tzv zkkr0p6soM|S=a9}zv{n=d>FU}f5+?DAKO>puIObTq!Y|YHou3)L&cCZK8VKW-Rb$C z2{<{+2aG+l$x5qx5I8vw&&pY2h~P;yiM_*bm3N{ATKmbjXAjwZjubZxh|s;WZQ1WJ z$D!qBEGiz#24^mJVR7LRu5MQ*W0y{Ylw~T6yNj_U>ESTVZx`>o#3~ZIr4HOiF2eQl zhluMpgQb}gz6c40;>b>V{kJg8)Lh0C9oJ-h#1fe&VlF(-fS<5@_Ce;4uRmrP*3$73 z4{R)&LIt!BVPJ(Uv-ZCjns-bN?MmeNxhVrwVrda^ahGMCp4h`WeM=mdS&Q_CDKF@u z91+~{51d1z*azQ7(dhUfmY3V%yDZKRUm(b4emugyaea#qw>995+h1@Ru0ehGHsU$1 z0hVEE@cNb?yttQ!&3kRp%0UhjuFAo>Q(x)K;N#SD{vBSHp*P2tE#mEPS7(J%1=yI) ztz>T}4|UGwV*kDv6zGn?$P4$W+Ft`YD}?L8yy6&P|LmcU)IzkL80>S2hQ+(vi4y0a zC_nuZ3U^2`KQqB89BvWaPM#w zwc79kf}1|W+rd1>5LNh=`wi&;iyh@`sMd7@x!(eiM0@tO!2J8hr#?xsR<8%59 ztUm6B69+tzutS((^plpRFJ-}YI(6kGL#Mw4IH@ON%g8d~+x-jUwNnc_bl!h@) z8q6~1K%Us$D7>5an%qw-A~Dl0Vw(MN8n3L#CbwKACm^1yEgwnbeMNjz(8kGcbgAR6 zLB5F2Q7l|1%y#WJq7etGF-V!g;MY2^%eNF9%G3C33^Z`mw}l)ok)+yfC9tzu1y361 zgXF~y1SOeN;8}eocHHmc>U2YF^EuACcywXB^-VHwi7pI_3NWXSN-&9W z(#-3KH}ubrmE`Qy%j9ff5HWQaE_=LEoKv9 z=A+-12>kL$A9|03V_?}bG!ts4{c@AhcBuoDa_5EVGw;%DbyfD-iIe2;87WX?{77b@ zE10jVC5|uU*&QR^?A%idth$XPdWdp+{4Z(nxFZvCR=k26$Ymdk{b0peCs=rQA6Z^D zg;`k;%{?P;qKd{u62x@|7SG)W2XYU=8Yfqp>7homW^93DiYs8Sb2fZ4?uN=8l>pCy?;V$OPo)Qp{PKH;3CeZehb5T4hqCYMi$7o#2 zyj9&r|3-2h;ews8?gb&2K3JlL^bYoJ`w;0|Ich$ywzct)RTaN0_Z3R_ijmL1S3rFy zw>LUSq0`rer@gFzjCSm0vjq6i$~+|Xso&w4(p5McbPNteN8pO>LS(V)dP1y>8e}Sj z;dQn!HmQBWu&Ojn6_j8vbJ=v69eix*pT#ca|A%rG=B)bG>%{F^0B<1A9$Fq>M+t#! zs&J!-eVMb2IsUW-hTfb7%ozc_u%PXjAq()ilp~wJj`JkSw&1=T5iE3>f@cJ6 z@alF$w00eXdJP-abD#xxM%}=3snb|7nEx42#W~^(d8** zw!k(j-($j)yZQt|+ZVto=R(|Gw;5$RjcKlUJbe&ifF)P|!nfWqXj3Zdi^`lw#pr@3Sl)|;Pg|PdO>-Z%AJM|^yqhFzcoU5{ zk%zyReuM1$cTphw1pcWF1D-%4?|Aw&m}L+IlCrh*wDNkayA}a!A9v&ZNK5$ScNM?# z!y6vH9%Ul4O8F1MPMEGDE*__oy z`CnJ!gDvLZ5PlL;^Sg;7HW=9arFt`$k&y@0JaL0~5YE$qMW0sT zzrzD0c=tpY+8d57|9Y{qaD@DR90s$8qsW^(<9PLMF*#!XRu;B4Jj!~`f*uF8NT zZhH8j*c9E?#j{?=XW?Xy+5biJ9BkBb1KcdZTuYpVQj_&z?z$#w=PkrYx{8v9Y4UJo z{wLU*qrhh57B^G`e4}R^c_iYy9ria*V4|PQWOuedfgfs0jJW4?o?7G}T(L@qfIBrb zb50X^_<9oDeB471^xY(mJ@c9Q9wxYN6(u4qSqTA>Erm@qjs6@zU^0p)#BbSq;MWf)HA@jZU3+5yV2y zlGI{2>N>jJm1NJXbzx1f?|=w_VE*BuOelI{$gWF!g*QAV;CGRYB;0?5K8yKEV`m2A zWrcP0h;jwl@M#)z`ma2**W)aVy5{5lj7V1gD?v?5F`V{Z9;fuV)8DrKG*Bf6A1Q3c z4CM)I<3(pOPn^PqCDw3yf;en&^~bOvH?o{w#(louvHso#65MYF3X;b$K<*40-AJZp zhnw-n9e?D_D8;WO=`hhe5w-^{fV$$9=uwcwwuf#+hc#vRPjd)AyDs51OC%8aLos;7 zXEk1*)kjn_kATV$aL?-V{Q0M^qMzJdw22PF@X1-UVvieL^0JQvON9nWRkH1*KVLV2hG`}d+O$74E-g|mj~c=JkTb<_jcs4xzyI==M2Yab0&oeA~}44{VJNZN12 z!RMe_C`pci10Rtl+_hwn1R+MwRma|M$KZ?26>yw67uD+oVRL{r&hv7_b;JPQoqmE{ zE>;-1U!FbI8;(XldF-Ccf~<@67y4hMKCD}Flw^tdk(b$!HRk2syv zD1tAes`!?>9eA5GkFg6UEQkG5J7{qC0xW@55XC-*2AyVVUfY2G(sM}YaXH3)UKz^z+|{e!bj|*svUW_$?gok`8b)K>u1ZF)^MH1`Snz(#SJ$n z>9g)_Sy;J7hg-D8V(bqCn7>hqSmYLC⪙(d@j#q1!v-sz}qxX^f-d|A=Sg1h^x) zjm@zTXPsYn@ofSZ!a9|Wa9l$YPrZJFC-+NZeN#BxeKLo4;&2|+ZM_9AUUC`1RypQP zY6iHD^g_x!F=jmJKCIL`3636>Wd5B!_}8Ko&oA5#!kX34`CA#2=ihUgky3-Am!kop&q@Ube7w|7PgyW;o?_6OH%{bF{7EM`W?E$x?ZosL0zd@?NU^nktB zKjGrcDl%uyKCCWq#DW#v1)j^DogSvhKXL)C?CXHFA)D!+1YOV^eo5AtsbP@EP5x|~ z(>V95B#1nhVba9DlJFJ>jycKaypcIXWzJOA^qK^_10&fsohcwvGy+>VuW9bnQ(TA8 ziM`<%2q*2i`E*+-eLHW27N)zyDaNkR?{@>N-2H}l51WHopae9P|0JhMSh(Q&AMod# zq;tH4nL353Ae1G}oamPa>l8USZ2yf)_!LvoIc`@aF3K^P89`){?oBwUbCl>*)ewcK zx8z~x5t3|sgPb0fqSY@1@buth)=&R9I%TcHLo-d_dyfs&TeRcBr@v75dmox0$CdEx zhKr+aAa|#ZdM(ez53g2XBMhL(v2z%@gX_U3cr^-+ecEKRz=URa*=3hu;zt!!r;boee&HBB(EygqH>0@UO*g zC9aP%aDj6a9+>xtE|cEDAA0SJeHL->N~;mBzdA>TG?Y+wmOf~Hi3KfQ1Vp=T!r9i} zAv2a^Jv@9s5@z*4-mh*L=zIsao;fqE>-IqKRxYy=F^GRUB-li?XPDewjDsTUv0A2| zl+>G&J(9u@w89GSxtY_ND~51K>?{Y)z!j8VdV=C9hP4b>mHeYofiQt=## zSA{b9d8n2O{v0P}Z@pmSEmz(lT_-e<+yK)ST|q%^4s)P;0+e4lPqmE#h}71HaGQG# zCJJi8gd6HONb0dU=P7#VYU092+-K;M5!6imh3dBV(e3bk8qgbx$zTtYN=_onb*ArM zTuwg550mX5_2EF{3I64|vvATDHQZ`DNQ)_hmAZjAdB$Io?L7w`M}}jcFvo-b6phtE z$LKFXSv>S#6o09l82_5FG`LMEYS7$y0L}L;XLiU5FcP^L@Z4?)-rBEcW~8Okpz-JQ zSdKE$C^rJ%qd(|Z%}V?le;##BGf}4^l)b0o$_fq@k-M>P!KF_Wv`w{GErk(t?_K`9 z3yUqVAYL1K-pQ~Zl=N64vlH7ipMb`HDvZT^Rc3TcBlNj{WY5j0XU}mn5`i!&97B$C zuAK@U(tVIV(F%_H-=%Ixe-ZIWXK+|tO;^RMqigP2c$2LU`FH*!t@l$n!P zx#K5lO^78MFT}w2^STh~+CrDWS^U`LL07Kv!{bj9d3L@x8W+8rkNw%V(Yn|c(r%f- z?v45M-jX`1?0A!$RNTZ|3%&@|#bWgLA3o8j(u6HF`Q)mbj0b_0`!etJhOanfJh~d`%zoa=E)QA5IL+WY$hw20OlX@!f8&ht5DA9Bt8O zJYSxNcVL6_Quq%A2+xMaP(>5|VWeWRY+YM4BtWK+K z2T4Pc3JS@jp;P88NH4Nv{{7^38Xd|oMOTd7rZXF3W;xPZb^dI3%qw2E#UM;_zRojP zJehHrJ{9k8Dx;{%a0*CgBvX@e?(h_eMV)!2ABqq9)1_6(ldbOmdcPsFg~GS+U92P-gtl;?d)3=%^W$jRM-*l?_YbXX)e z`Uz;0AYL5x-~F7_sLzI?*ig*8{}vp3^GLDjPx9>SJCf#d917uVVTA?wmvIeGTlMk&cVG9NQ!tGs^dKL%}63MCn8tihU3wUrwzda~&;- z1Lu1QnjsCZJJ!-qacALGn+&XzmSF_so{$H(?0{{$f>XyYu~qJ=D0p)xW^*heHpLAJ zi&M=kD;}Wq30L;_@MaVk>7xnfPtYwW2_rFk;DnHHQ}}Z=ydYix(;btc(17EPrLyq( zXdG?k@+_lc5AkPH0m}bQh0(MPtm#G-_F#7##vhyoW6lq$Z@4ve=sik{ODc)MA4AB~ zS780Nh9l3j7-#o$pTmJ~urgDX?GSi~kEEMPoYn!hdG!L6bSWpd|I=WV&&|N42hQ+t z#RS@5tITSewn3g&J=kZkbY77(NKJiDUW{en+e?0+pSYGaYLCH9)1Q!W_eofx_Y{`~ z34`SuE=yoB0XzaH@}Isc<=8E=P}jDFn!FRo&db#te^3SGR}RC-7a8WmL@RWe5Zahq z83p5qqu}b1PH6lT#6;6w_~q&t{tJF@w#7>cJ!=clUhN{$xo3{$T?+7@AkY(mTd`!d zBf9N9gFCq%sf}I)Fs9Aq<_bb4i-dsCHB-F4)(To}yJ)3<5gF6UCW~Ay@pF5mF|_&! zdQ@w&htEV~%hE4g{_h@EH+r#dmE!Cz%c)=;R7B=@erkLqn)#Sov7cRf9T@CnaBTQT9P13j5ur7)t^Y=rXF}UUI4;bs_fMI1MEnlID5Fk8g;$y@OKwXXA*NM zshM*ZS9^TnAGUo1;kRFs1fh?lQ-iQ4w7CpOiw{^!orgCA3`6Y}fW@)DkmW7Hs>ff( z`7K!(AM=a&N>AHyQGRgZY=tw5hV@2cG)BB6w=*%w#`%MYUhYa&s_IJbEDt zbKXbNSMRRj1D7PCuXqQQqTbULe~_|=eHGDbC0lwDHp|M&e|ISqN`Tuf2XxtcF9w$KI7AI0MgFE-{*baA2DWh7+ zUOXJXgb^Nn265lGY-(*ZH2%@x*-YC4QJ$Y*)_@`yDy_lLObxbhgFAceuoS&)IKxu- z*9h5Yo`;VT9njUzg)NXP#}LIfawLeGi+ngvme`*H{p~AQCyveDGct(T*8A}8xGrWM zn~Gjr{P6sqQG6E2^@eORAmG15co`(l*w)CfQnQWdtW}cu>)A8J=T{^IpOPC3S5U`In&7$YDh^+;XEimJptivk ze9^7JHeJxer!*H^bxzW?C0+ai@%Qwcu{t=svS!=F#MyTr{*dz?JI)1#`-xy6ltsIig_G~5?+I_?BS=r5#Xr8bhwl_M zoy#TYGjlel!i4@LsOVK<_Dt=lvXtwt`m#5dM0aoD`hMl~*wF=Oo^6c3b(El}CpmK+UFAK{tX&w8?%f)fTY3w-JyXB{iNB1-Qv}TBRF=mL8YvyR9 z5ndrp(n0EdBuHH-p#FP;Sbf0=T$&n34^QMgAgbHo*+>^I_$19r@=u}d%@6#6xbyJ& z*)86TiQOd0i{te-2$6(I@+59`Kj#FDC3g!K;WyKlbhzpy8f|+^PN)aNn%qTbw)P=F z8o;<=HA*@RL)j-8gj0*yspH*npt6s;W||O_EIB;=W-`q4Y=!Qai|FNPPRmNnplew% ze^WNn$h%AFxbIoo8Y2yp2KT_*>F-Gn|1rn}bkKhnDlqaF#bmhysJwL&>wP~P&u-Ty zt2a-f2R#=yXge3dwl9fxU6d+5RZRm4*mGa%G3@NUmgxGWKDUyHfTaYKbMo8umfMtSH;J-TVQ{n44%4h0UhL0NKfT65M0#) z9#{_M5p!|!_jYQyyMgQ<2*R@kgDCQM3Pg*}!dIpMVd@g3v*!$yd<=ry#lCFpyVLBg zrXH$fR!DYQR?zsQLR^sl6zU#05&19kz<9?Ayl~|d%=H#V?>RX%#pWsvHx!{tl{2Aq zM-1<^I_DW!Qp5l6uL>F3)kAt`aJyxF1@LZ}z_ibkV>fYK^=m1~H28@KmO1o)*%!tdYA~WQ6^1ed z$jNhJ&^A0mbiPajxvVs>P~8lhJr={$kQ#_`b!1KiPhzzXBqMW8AA^59pwVrI+1^@t zwn0~!8QG8p&T_9v7o$ZiYU+ta?-_b$cn^D2wiaWpBFXaWmNEWCyocV#kMiwI(bYM&Sy)-+<3!-AJO^z z6}WgOxN3Ar2G2Bdez0ezFnstN@G9$UjL$VtnfTqD>o^S#G+u?54<<1B$9CiX5Q^P_ z6WBw~lUW@BQMOef71?}OoOA0o$3;2BpH|&%^)1#4?le52>pw>EdVijVS);wAWla~p zJQ~D`?uelK=RBiM{=yjWL?4p9+u+mKPPo0M5Qg?n0=p1VG}+LBii)jx>x&_5O7RD~ zZ*#F_mMA_s`3#Nv66r%U;OQUNqZ|L2;MQDu=BD5zE@v?n#n+oLe%_Cumpun_6i>pc z2TIK7`bij=o`biSZzF@^xo|@B28zy%gruF~I9>88ezqSt>9ScGg+zL0q|G-L3IDk2J?O~_)oJCBpdUHl3Opac>0+H-CafJ zKeZ=srpU5s8<&CAnH1VHdK3d1UQyopYPz&C1j8(6;*6#oJjF9a=Mg2?68j8}TaDma zyLN83K8nA(uCuWo4{=pY4z9Ri$__R5@Gq}1WE6fFbIjxYkZGihpUrnL3*O{1_{|d6 z*Stc1dqoVoEW(!OWYR;Yh3P?i4JN(gCDhLHgVbCL@GPB5wx@_=cCj-FDxXW`bpaMd zx#L&edU`IX7TJa=Xl(op2cPQG4Yxyi19EM2V09?(i2ZfCLO!2tRCQ-YR+huX`>IS) z;ymm#lVs!K7qi_pUDz$Ni0!O2Wc?MSShI^Ikgci!`MIt1iB${9vz3EMGi}KE{S&CW zw>22eaYp;tNHPYW`C`+xv6SyltTQEGxbGLadFVO$V_||lM?JAL5AmjEK3!>Q0tzKD z{8!!Dj2(Lq8)yq$h^WMpNryqZYYXgi(;?2w7m~fJzTq$D$?Pi2NYqTL!e4JKS=X3G zRIt*<8LE#^J?#un)ixA=V~Jn0mtI?QbmW2YxxU>YtQBT4NiK(b&n z=yq3t>)I(e@h``*z9s|3TXV?2X9M(!@?uEX_80yvt%ljbmtoqAi4eJUE+ahWG;>Bx z#cI-#H(cu~?Xjvl+Sn)ZrtwB9=ej zf$BzWP?|8x+x*0p#w@;NvpwoFto|VdD@v>|BPFEL*f9n)ua`jhSAA~IT!d9QgS>Q` zhx{|1t9Tn{my<8&1(>CKqaoD|SkYc1Fz8wWyCYfPgumoW4#%J1X7|ohQ_1oE4Af5l zf!EOopUXN0ZFZrHi49oEdX z!O^yzn10j}-Es@iRJE?^txhi9-YtmnZQcC!lP2@-$g#AXD-xY1)0mf5dQ1Z?$6a^E zdA4B*bj37N7+X^dp`N8sW{|-sKC)rvEb@SoSw(O=aRX6Q=jMB3BM{*A6>Kk;!|Pqr zFlQo#ef%_j$-!aFD1O5;KbsB;p9NU+LtGZe?iYVhDW1Q3%XjG3S_qG(M{^8nfADu| zgxUidjB48f*q3;ozso}ckGHBbE_*qT?`A>z=jlXHG0r1pNrk+u9aV5&@gTfUp2jBS zox|BlF$Q<-8KpNnmq1u83o}mE zL)qd{QWvEUqLEf~mQf<_Z0|5_F~0_hrb^7EO*K$?`~^>M5`|-4VT@nza>iBuAJ+?y zBc(Nq!5%-5n8AzvJ9u3X2*VHT|IsSgl*{2O$rups^>kZG^2V7#ZzWiCB#L=aww zoult4x6))&C#kVtHa4@a0s@TCgiz|vgp!;w6X==4VhSsQZ0nmz@VzP zf6l@ToimuA8wrl94M@O~JhIbA5@%klBdh8xnDrl}n8A)-t`AxUQuG`7@=}PQhwnl2 z?lkmN4=1G-2e}z&esc=olZEH%v+u{S{8$QI~M^zZQ%>a!y6r#@d zIOImyC|ytpD{f4KhKw8}HB0GcdndS;?2Q7f7uS3GLCYpy!W+-uQ2x4u(3CX}B0bUA zRL7&@GHpa<{zl>>>P8}C^U(FD3%ZZYCymEDY}RLZ)6gyRPWN5`XnLnx@%NtC#|+*7TT{VW&y< zBzBRVq4V&1H^tz)z2u_nOi;FZ%ipe(UbXJa3mE*m4~h#S$@aKtxM&;;{L$&mniIQu zVU_-5>*a3R;~QaP5Y>#`n`61$nGr0WoCbxTFMv|XZoa;yHNFv&M!y0z#wBAmW0)z+ zY*~GTnQDEEnXPz+Q4bAA*AG!(aCIr#DBr*rX`x8Y-=y0LI3~kl7cwI4!{6oRMuIO3 zlX#g6`0HKAq|qrg>cT9_jA?|T&jJF-)6hB-og>Ml6Q}1lRBF%=K9wQxz3=~>H%U@ zcMQApE}?V(BoO?5k#60l1jdcON&c%lyuW+r;YBHTs9*a4Y^{%B*~i9JfdYzv9TOW}WG~ibnG4$WNZ-iAv=~ogFUJFWY@e;L~^VE++@Re^)}Do+U)y8;MF>wgIWnpxpx*; z^6r4zY*kQsdlDTu=e&K8E~qI7@DEGe;m^Hto@Zb(hbh{`IZ*5`k@pQ7U?}Q7Ue?vB z^3Yid1|)>kb$lTKhw|}Vf(f{_ilf<&A62D|qHK2KN_N_sC5+hNNz99H%U~2#D^p0__Y2!{#^MIeb#$HP{IZh@@WeBD)OB=*%?9escMK`+6Glx3*f4Z z3fQdUI(kEsF(%m`>;t2yuxJ^_yYoSb6Un4vr59=|0Dkq5go9-rq+2xww0)y^I}{hQ zqv3bS5dQ^8B`5I5D`K&FtphRaeaVX>i`m&Z1N1;uE;fD}r{1=WRJ-yTjj1M>Gix_a z@Rgy5o)_~U)+AENrYfjTJ_}t3rm!z(+p*&OJ}TsX4*VX!MA;oxs9!si4S9JSFSe$^ zjlvt`YR*Ia{^S<=3a80>u&3<0Fwjnn zU47BD>O+zM{M_G9^A5d4zXmR!-LD6Z6UL#nunDBQK7i0BedbJB2OhRKiG4!)IFajh z{ytbr7XOujleQywK9{2BIt|v>%#Pj5&HZ*9Y{IJ8v*d2f9oP^3 z&?NaB4?f)wO$}|NdDlClJ7WdZDh`0LZ3CD;ivlQaw27h4I zo-nc}Faa$`{^F7H6|6chlpYDYiC?C#XJc>q(C?Zyc+)MsYHL6)j=UPdkpAbeQt=Y( z9})(ql@_dd-7K0qc?wt;|K;y|e*)7c$pLAZ!&;p_!*F!4`cSMyV*Qlqp6FI(D7e6@)5ZRZjNzA(ueEg=7Zq95V zDf*@`s-8z9Z$9LSul8WCebr#qhhD)q!(b?_+QPfDE}E>fTZs++A&^va87-`UQe+@?dQLaxob2&HDZG$NX&co{OTn;$x4>oL+U@t1lGgA|m zFmg{Pfg0Z!x;SRfmX6hI&&e{lxK$r&7O$+@Qc%mi6Ires@{x$?1|Z+!0Q*qC^MfBsT1#C$h~-sEmrn|U9!-buof9roC`)|Z;DdCya*n@*><1vMKx$O;Df3Gz)LVc!dmJ`rSBaC~Km7M;%f3U?xNsaA)$slB&SC zjeN@-Mf{}b3E!m;5s`cm^5Li-OfL}N_6NH`3C-AoCpJXO;wo&JWQ^=14f^x&Z#c=# z15Tfr&VJ&0viqL>r7x6qaJxh@_Gb>#rwbOae|~S`>k5_7ib-K~>TdzoZDkF&UzXx# zoNMuSnL2APtj=BJ7n2g&##e~Y#o5bN@eB@`fYHTAwAkb!;fr5KlglE!gwUmAl8rCE z49>>AQs!_qrUR@^PXQyS0k^XSn5SPQ7?1cgFul%o&5~81JKT`2sWJv$z!|dKvjNtd zI8#63O-|T_@!EW2=*h$>%>8?^@JAvBdF_(Wv9O*De11$uWAzzF-4bkEqek~cq|$}s znYen>Hq3R8$JqKrVz!HONt}I*a;ul4lIRxv_~jzu#t@)B&j8C-werk1PG)zHK4!lf zv_SouZ+LN=JZ1!Tk#Uu8c+#|$HY~FP`2{tk+iwhC&l6`w3@_t?BWdi%#X5j|9{ezHF5A`kpgMsvUnStq-ZpA5ABmDq2bD;s=4b^>E3T5 zsIhGZ9DKWkcHHp785?%sXz3~V6MqSsRrm%Xu1=*A# z2M!&U4!k5ekbJNSmp-y&hc-BY;!Yhb4bfr_>kiNhULx$}0}tWilmNKa_>reQkb{EZ zC#VXSvDmms0Ds?&hK(&^V6S9~^<_DbyI&1Vo?62U`%HMKxeTJm#G&|6G+cuNJlpt_ z&?YekD$d)n>hfeZHbNYAhkGHk^(aPd8CbBtCM#xhKL9ASO3~YlU`15aQF)M|pVWRFdawo3>_bKXvup9U6Mfyl? z+*Q2x{SGS6L)zTj&!ZKZtgo6K6utRFwjX>!+>#bDymyMsmiKC`dg8^FzP&ys3mOE6LJV!9b z@SVBP`Sv4BRTm-ka@tHyU^Tv;=tkKMx;Vgf@oboKl-l(MZ-{JyDO-h^fWX7VI@KC^ z^H+k$^$N0on-070K?t@jQ)drMc!aWdD5}f`cJ)C^WNA|mO?P1+`G8|F%#Cy-(NJP*g##dYxqF;UGDVe+>ch6NqsYf!L{%#2G(+Y@AS_f=9 zl+5Le6hWanh_5+c7P8)Epw)|E+|KQdzsDzn(e%v_Rr-Zq*(}d=)G9ISb<-fpFB$J# z>ZD8bp5wWqTsnR0OkPz$3)!%y1T{M?FlgQ;5L`D#h|wnagSot;w+CRKaw^0<{)L^c z1pL96(e5xpXDJ5HqY*k?tsw%_i&(`)r}@W(7xP2hW1um_mu;*3M-8T4!4+C=Xyy8y z)@aA0fT}G_kFCJbjZ=9$TR7M0$RhmgFGSuYe1K7hDa<>M_4rCI31v6+!JOFFpis0Q z9?O-J(3}qX+qfL8e9Hi57}AVGuH;gp5VN-Q2W-?4fH31&GOQQ|LD|-L#$fF+=kP(g-IIzXq-st?RhGcGTA!?_Rsa?+$_Q2dX=wqJ^|K9F}l(oCy_|`1mwolTy z?~e)`yD!OH(`=@PE!5~C4HxussV5iCOVI9%zO37*H{kV>Z9)udCYMxD!k{E@w&fA`)eLrfS>!IV8*$3IfH(r3yW_wVdI0%xKHIUTx1mk;4$yd(vBi3(1 zi>D=X{u((l6nK^@n%a|&{|r&DmZ;{((2vRK!&io}L}S z74svhta}vhO*P=yJMPfqI1v}7e#hHM{#XzgN%t)uvd(#Wl`k8124X~&7_%=+VbF@p zfPB-zaivkx>@&hwnt6>p&pJ$f;|J+@LN0l*ZUx?G$)Y)2PqlCAKeBq;E0O`OP;}N1 z%3O9p!n8D$T%L-T59M+^B@am9`Jt3Y7-qb0hjis)5EV7W$`4!c!_SYDRp=#|=?^i5 zxS__0Fy5{fQ&{r&Bw7!oVcYM0FkE;L`uT#a#Y+pA+wKgGv69$&uahk6Fo8eva^O9C z9$sFM%b!ttjX!1T3wo`p6XkS&V0wEam0$S)rg$XKfw`x^X3P_MEqrMChf|g1J)Pt> zUk8#c3-IS6HB94N+i!ngB$s?js~#qDJIKu zdXphp<2ve2h{XY`3Xs_I84Rpk$(De>bkFG+y30D5`fLznTwiH|E_ZKAF8WOJUK+#1 z8HI%S=8@mU^Wj&2I&=m+0>9>bXdgTQd;K$s+4gesL+J$tz>|%WHZA48p&!9%aSNe1AY2G|# z&Xx7)Cb+zPMmNn}%Opr^F(fXRUsiGpr@A{(1)E;H@=%?<^Y=N|k($89aBs`CZ(X5! z&l;MiZ3n%(*2A}m$?)xzHLhBpPXq$wV9SzGF!sAmgiU!6IzN|pMt7WicdDg*M`F3I z>~nncJqf40Jq0PhzCqwTOJ;ilcMb#>{ud3Q|8syXNK#cmO#wa|Kf{k)o{UoqW@Gb% zT>83UB0Mwd=j|Cfk7nJ+(Iw0oZ{FC2@5aJ-lb84qws1FuR&jira(!Ihw-qCvw(-6Q z+kmNpAv=9y0bOcaL~rFd5y{J|(9K+pA3tD-uV8zCUIHH$JCgn3DTZq{bSFzdegqNm+DDRvK z?eiyt>*ZW3@KB4?$gg6L_P61f_DOPgVJ$DpzK?E{t>9_2ap!mc1lXs2f;C*FfF3Rn zIM>G&o1N1X*u=jn*m0tWY>bGsBPMRG^vmojaMP&aZB{!=Yvt4U#b;M@^RDOkF(QJ9x%d*V4+~Mr zcnZ3@gcI@jP~0`ZCo6yF)2iprwI*;Phe(w(XND`^=AHBDh84+&UF}6&;OX@unEi;ek%)<5p+=5VTUBbR$ihhw>S z@!^<#)y)aqTx&-!T_$ggs(PU|!D(@@NahQMX=OuD_cM6(`3_{NdvHv}X?V&*7>+N= zg)N$*@CR6ag=sUN&6i|%svIKO`_p+ZCdT3E=>52xhG9-gDD`l;#b0QmgFiG*kkr5P z$x^*aYF%H8(MN<>aj|gzI_F(5yCexb@h}`OtEGy+s>!y4;^^zb-BBl902v<#SU@81 z&a@azwmM07M@~iI!$qhup&H-({tXBA1;Xhpc_{xfiFu^;9nLuKVRTC3X{C1`9!e5s z#h--ntbKFoo`qVB4__PJg)GC>XA+#JXY!IgYaFkDUMEN$ezJ`JuO+zMIcrf_(`8^h;Y!dUfm_`3c! z_^g^prt|JWmZKU>bB*P@>bbDjKHS8%garEhj|@Dtc|bGfSl^DExq zqZOAh!K47*Kg>qGZ+-NQ_-E{K5Q8ntk}(rrz={Y5P??$mK6~!Ne-FA~hg%<##4bc(L8H#<6F?6-qZIvC_>E=yT{3 z-7fEpZ@%58!p`AXH>Z!J>+IontSV?)kPXqo=jf;H3vv3y7~ZB;kBH5?Fh031hz(Q3 zVPK*JI9Lm_GjAu7m3g7)bR!HzxH;ds)$d4^q6ZGt+k*C=S1?%DLiA=g)ABW!NN|24 zEc3)W^)?ag-Zp<5hY7wEwILP^F~@Bt`1PM}6dmeXGMVSH>;4|!vj5N@VVPOEZr z@mpKaYi$&|y~?4Tx1Yh)OEai~^dNO_ECY+#?mQ_G5sazKCQm-wuxq^U;qX`)nl!1g zKl;sBi#AbKi}O-MUlD?BdyJv*Z5$br%cBc}B1ox%Chl(x!y}~+$XeA;l=&A^~MGNQ2%>ut(fB zu)0UC;bsw2yzaA(irW1{VXMdJcJ>BdZ?6C&(>Ju`3_~R%lR1}W5DqQ6kC&=GP%(aN z<H-H9a7;9o58aQ6wB{vk!<8^!uvaAe4*NpyoRBJusiewq5psm3ICI}hAWl{N_B}~!IA~M9-V32?)e(l_!VK2bUl1Kx{^_;+6uTz63p#Gz;?wY zc=`DkF;~6CJMc|`NnN}f+Tvf(#G@aO=WPPc(%io6hA5Jq060QeCR(9SLsw^9;1x!t_Wbe*D`$J7lg{~@9Fpy zEhgM~820u(B#QGrA$R6gF#Sm&Z#UE;mIkpQ)|lkk4t zbz1GIQq}0S4&E;Dz(enyY4P<0%+P;?pE|m!k!n3X;Dq#j!a_Ek`GPB$N3^->H>DZJ zG3Cy46kGHUHkKzsggfweE)9frF&S{~Whn2^+z4h)>>J1ra>WgP-YC1XnYaH*HQD_p z3KqXR3tq|#No7Di&f0DQ&0J?X>-iP3qskInFZiSIb$1ji4O+@ zgznU*iJE~RrY6Knf0l<*e?><7u`IK^ONl90sfHkNX@-rV5Xv!-C6vXPr|XNLGcp?D zW3GazESEEsSqoL#Cd^xZRo=+*PHMf?7`6x7!F9I}{5vaRphL$SWEFR^Q?(*k`58UD zg}E20$H5ScZ?Ir3w;QorLXKj3$pMasUO=-gjd4X+A9}{;B9HT~O|YyYCQ8%smeo7H z$eS$KVwH-IWjtWblm&E#Tr6r9_~sN3 zo(hApQ)^HmsT~Im@++-HO^E)yg{UsA55ET0v2AS`KI+_oyQXi#SKF)Lvb+J4-=V=} zO0Dszr3YLU%7&MX0kHOkAfpkGOD$rBAV;tVo~e4nL)%&SP^%2fIj()kcsA8IXNA>r zzN}zpBFJP%6T7y<_=V&5x74ITlur@-lCgkj{xA{WlTO_h=FqOB&-8&H$6yNWCNJ|d z=whc9n(8Bt%|>(JMWZUS7CL zByisb_~x(1OzuAhJCxdB@~W?Vqe>m-$?tR+`S%9Cx@&kerSoSQUvkv27V zFD8D@p4gD}1?F{LC7~BKVuy4HrhIe2u{B;O$7Nd&?tOrtN37Ah-V&S4CPVP^XP6gx z3ZHxHvfHgU^EBTdhDUlb%%2w?u>Ok-Rpc#!Eu4oW;<6y?^JNmg?K*_F=W64+vz0{a zs5#e({Y@fFBuH)T9MUCV!d#F~z{5kiz}qg#7%UFvFFmw_3?-Cd$mV4D;P)8|UpmnJ zmJ6UIdJ@y(tcI*|Bwz84D7;$r5S%~Q;5JJ=lwFeot;+h0>W>Q`frj8JKUR6U0J+?n z2;Q?2f{_)*Fu5%Zm$Am!-?$$I_5ewFtcD?H)kweUC9?b3WpJj-uwvUwIwMgPhYVye zV&D|ci~fLO!q%ub^j{obp|q{+^hQgW9Yv5H{hXH6y)UXg)cT|$uExKDE2iQ4j5P9tL|Im zb$SfF6q<`KmwI9C0xymiSIA=UWcGQ1EzX_f&W238&T8hYW%sh>@Qm}NCC&N>zJ`wQ z>d+%{cXb@m$fNjtr#!pnN*pn_Sr0aE^swpc2)>Em0_O&;iS~;=GH=Zv`21Im`FzEe z8B^q5YYQEkw17JcPMFRZy;#ZQ3e06*^>AJ372~{rVG?L}N0hd`NQ9dqy(rrf4^v;? zA+;VM{NoEeaj3b8#;gy)HB25@FNvnxx{Gl0?^!phexPGxrEv9$Mk=li2P_SD9G%fa{;96-a_vbs!H4mgy zHAL`>bEPRh`Q=%lgOYy3f{R*+&WLk)ZV$U^zvTk_#xYH_`XtW;99@P24HMZrpTqFJ=5zE5GVROC^s~B(yuAFekdoh#P%3s!G>~a%^;#KAZFE7#p~42|WHR1W&w9^2eW0l=>$Lt(re!Z|iBUfBz9il*Ym9)=hf! zdoVM+dnau@D9hL&SYN+}#fH z5AJ~e4MXP5l1iA<9L{-NJIHkp73P7VCi8LY38<1hhK(P>U}ji18JM^VW<1nl6!kyQ zRbJasMCnJ>vgYTsd^8N3qpH!T`X4&)MYzk&VA{%SV5nMvec0v)dMkp-^$S|?QOtvR z(xZYd(1@u;CX|Eh;C+!(P^OScHO!Y&%|685q6_HN(}2k@!`ahICbLeLR#EGD&X^v9 z>;k!5_TX9_xJ8SJd%ZmMoh8JoKm>1pxF0;rT8Q@pUSXY?B3t5n3ElT+fbB>(3TytO z(o+t?-R1YGbGJBMCiWci^|a_F_5^8o`T(y>JcdcpoD00~7i3&D1HQz3=5$y#F`jQg z!(3Tj=lTq=+O5R&538`H7tT<_b0c`(+mP)}L7erUH(E%AQP=V`DvwdP%cz|{y?7Q* zc1tLq!VW`-fb$C>OEl?I(_?sCaXoWAb_aXmOA+h5r2xf7 zG{_yn2-YdahusZ9_uBdFyyLSJf_ zu+lwz^l$%6&M3_ReJ;n=a+1qvB&RW>{y%ZP&Scyg{2fqvA{-oaW`lZ-Se@V<^u~eP zVE!hP3cbBeeof}Q@O5@j_pX3HK6xL17D4R3IT7bxHeiHcEg(Ojwm(s$CDS?yiZ)6HBY`mA5{-f3SjRz8=EP^9Fb+E)v9Q8hD3w(!tC$4(pVwq2&308?pey=jvd(WOpw`??ip7053Z#VPR~4>G z!;A(EvYr1CWNI{+kAAa2@{b^#-4aLlz57e2xNdm^B*Cl+cnp7ZpV7-5Jp z3(NfGf^%Fk+0))k$;Cr>&)1wBDARynimf!a;2_Nvm8XU+W#pinCT$BSMz3$X@b|$e zny%@{XV*{1^+JMl#h(PyDxi$+Hy6{jdB)J`a}k&jXSB~sLA&W;pcVTRUfmRA=GZ1t z7qcAbNjb+<)z*@yg0^I;csu#GB!@iSz6Bgb?=v9qn?}Ds%zLCtX>KK#!3_REN(&!B zz@#H|P`;EuYB&oW`?9faza{?tAiy@J>#>gwoAFOz0l!`=p7fcpz&=uEGmWAc&oVvc z%i%;Oe@P<@eF|sh8~HNNTux9MN3K_R`45S$Bk;N8Fz<6}4NkCD0sCWW=x4tkXZ$fI z;wL6>Zc9bh)?qW4w=;CZ*7ancLp>ZEvSEU+-RF8@HDs8~;koQcN9BTa@@Pq5m2FW0 zA=~63zc!g>iSDO<0XzA)@CICXr$H{6o`vFoUT$V_6WUvwA#T@l(k^+GXIPVrSAI#b z`r6B3r~fgs-(?%hx)tH%3)z(m3aW_4fjPJ*a~rsrbmBU(NSv(ZPE?f{;F=Afb?_p; zX+VIH>aiiK>=I!pDFc*#O@JvAwnF&Rg~Z)OiRyj-3A;a^B!){{z$)Sc2)3RhqEm}V zzi~Wl`S1ZY{8FPM{>Ny?FV3xU=pkHqEegdR-sEgiCwbvv2kFaq@GQ+|GymmJ#y@Ip zs2s5zF1&FA=@xBlTlWTLN3`HjB|fFpn}k=#9S1 z{HdpxGgnd#nQC4OzA>7>F8DT|?Rmcu$;n}SeqlS?FL{MtesU4&UWG%GM+bSacoLJ) zSPY8tQmpsBM|dTWhx0>{s2j(tjom*^et4TP2kW@M6H*WHXmTWk+I8@n4d0OQz4~O{ zY%$pXOP_cDoGCkDHeILg0`4bX2~J@*ZAd46MqS{7j0zvu0AiE-L3nH?vypqI&*v>*42_hTXXE{(IX?@|ujnE17K+FykJ%he ziwEtXuS8iU0{73ZgN~UOz_G6m>g~_M-c1vjotnkuZJPn-i5MpTwnW2`ZMpn!+A~N_ zcqFkMl!oVb*MYpzJJPj(4Jf*nP`UV9{N*M}Dco;UrXQ;ZlBj|+rt}l z`%O;?c2cK(*NMAZ8P~z(_7J10l>fPs%1FKEA52Jv6Qw#t{e~t9=`+GQaZT{pegbXX zI8S|46#cM3leKKRL3g>o$Gd55_=FSPt~XA(>2u|6%4Y}5A+>QZ|V<#p2W&N_35tmeAR`rjbqvn_FW zJ`VSuEoSEYv}Ei#4)*Z7jp(pIpBV@b1KU?`N&MH>bo1#r9Jh}JwZjU`elsa1lOaL_rso!GM?Arg((M@8cA0bDMDQXyu2M>_>K?%a#V0{tt$>%F#lZ&ZMtJM=)x zKM6a2E+eF?06B0OY8`bz9d6#V*-jX*)tUrXy?k^eYv7c?E(ni%3s+_>!?Hnb zGH1mGu=+kkj{82RoiBT+@ADfp)6JZoyI&1QrPe`MQYb`SbAxOB-NgRtRx;2x0!@F; zFz1|N8HU@>ExP%KFjcW|Q#lDM*Tv!2SuddS-%F6@hcUiy=R)&q?yi}$p7~_SFirhW z@b}hql=uHqb!3kN$?02+Vx3d5W|ILUz1xI&%0_|Tpeo8vTE{p_UBd8|qx=a=3}O3v z0Z5Z{11+UVaOqVfE{p1=JvuXS@g-$c-aQ-2j*BwiyK9-`)j2f1FahuUIE@=OCc?#& zp3LuYH89BKc&~vsP1AQo#6yWK_xc}2=NXXW+lKKdm4@~} zNP|==iBivX-b#xiNg)zK6ovG+lBPD=CGDM}RFaA`MjCxfyp<^n+}-A=I7e0?}WI#8lUa zxl!%M=y*gxC)X8!uecJAt-g+4&!(_#Cy%n8bAQ3HdzmmSaEZu8iQ%4kL7*Z&o%!=` z9~3tK0KFXu&5pmJv+SIi<;^ziu{}b!3Qu8@OAqqp_qAbBNhZY7AT->HZecppx>l5(WWfk1N`6EdhbpUr`jwxE$i*>m%kguagIG3xL2FCKQS;u3y zwIbG6owQlhLOuj|Y=UV~Rk6)9}oRXuj9k|Ij98 zBVD<=i*5;8LM1lY;PdVUu&ewd28amrPS-c$!IPuZXnHJTIJ=J5<-^TY#ZyV|_QQD1 z&lQ}QKy+rg?$;|BJbqmlO=7blS*UzS7-?qv>#M87#D4 z3zK%wV(;Z~9DBDh{_kc-=JvpKazHc^`ZON%j%uES)QS`2SIst3GtzFRGJYBVT$)2F zHtRr|`Et$){S$-lwcuTe2Y592D=o3?fEf8Co-iVAS=h}n`?FyC=QPyG<#Guz5g=AH z0m%F?-nWhEuwZH@q`VYmbaJG@s5yxDTQ!?5`ZWu+LfVkDDC6tkbjUht4U4%PXZi6- z%<&0e`51-FtXLhv6%EI(X` z@=z)kWTNuuiOKuuWF1Fn3acbpk!Ebi`YHI|L3N1LQh-HEK2!Q6n(W@P7%z;8aZk`4 zr1paedF8H(`(g_kuIH9wQwMinsFuZY^8m>FuO1d2I0IuR8ZqRIBF~D;^l#14riP;C zd?}+>xI>6@u_;_c|5KVIxVIhcW__pW#hcLM=p{P!B!h#CgjrLk)1*%_k_PHthG#i^ z__fXne^_ruE#7zBd7&IEE?R^3`U$MLe>s&X+RqEOO2w~_bIIJ8^EMCuJ|K{#?c4)K=pizmLO zwztICwH-0^bgKkZIB~OnAz_>+TSI>zNTYTp#k5y=2gwRkW;`-pP_-X^AjmF2y$RVc zb)Nx~cQOSP@rp#9eA z5IbctYh(J4_!oGRag%3MBBB%(N6(Uihv&dr*%!)p>0zzIQ%oz=LXSVc@azf)yk~9C zo?LnepM0!^ee-(J@;M)KobHjd?B^JNCjmeCZ^2+L*AX})9}6C>!Rjs^B=e$>PSj%S zrscrQ>HGL%=`LJ1ClzP1KZZ^vZD$5;zDoXiJHpHUKRwg)AX z<=L|)dAKoMfXm1=!>pt6U_W;{^Ic(EL*)@?=r>YeB-1Htyw?$cH;Jus55bStg_@&wDiG3dhLxn!Ht&CIL88%_Fsfu%AY}7?jhv`D=@|q`^b&L=pZ>A4WjR!ZPqYXM)qXH~fro zcSydI046g!=&v;n7&mMI-?%Q;#8aQ~SN(n*^R35S>C0i;CMsFI zv!Q-d3YMs>Wz}B<;m84NIPhl^JN@x-n%lepJq(Rm5tU9>;8zrhc2vV>sV{KGx;Pxv ztRs4#_Q7Sb9r&n4gx%Vf!qi%DzaPw_Rz-+yOC%U7)(?JCcfdrp21O(z*a7c1$a^Ul(hSM(qTj%_qssubD8oOH|_oP0)PKYHj zgMzr$P?b&KzVA-CEZAUChB+&qky!mU=y}lqHp=xx_LeyjI?8!EX3FA1nLE5;n;=pv z*a3KG0c5`8d3}4MLJdS39L4O9S7~jO}Z=_)S$x~E~`-hCwbTHQ8@iyk=5v%V{ z;gP2;ZO(X%qS}H??QaFL!A^~Hw#riJ%b&1r-EWu~D@5xi#KYH*v)QL9bMVfTAusk?oW{+~<5RHJg9CKKj{REdyd<!FS6nTw}6=?p;x9ru1Jw*>KI8nhoqe`Ov(-68*#mpyZQ1E_A(9|i_72$$Im#P zH9|DpazJCT1DouUOB&yO#a6{4R7h24jLOa-yjlpka^7@tuK*tH&}vA2b`SS|6E({^ zr41w1QZVJ}Ow!U>fELqlVBLKO#**ve={CK`jw#9X?PNxHbb~PKus#+~IUOP|t6QNcQI+v1kYa9q z9f3t(CNW7x2jKXc67;>aoV+NoW;_=6Lf8aRP!-q!|K9sEsVXaB_s(-Bd zL#W0I9$Pe;LX|c~Li#U3ZboBBkN)QVhS^zc?{BWd=lO;xl&Zn9j$m9jdJ!soXRyvr z$-I)^_h?2bpV~YY4(Qo*tRgWbp7(UcCB(+IK|m1l1xFW?A2lVebdcYQ}Z5&Zua_|qt*kpv;;|n`6KEns(|Jbhsc$Cu6)%K>)BnB zk*w^|ne3dY?YQXvSzM#<&2BiI!ZE5@eCxP{?^#ks_4Qfw)^lL{&T>gT`LobqHrjBz z+ZtPJMd9d`TEL25=-<5vU*$Z&v1xDVqosd{p^!JsvHi>+c51@wxvjWaiJL8ihw}cg zYIvz^5tHGr!+ai;VAj+>hB4)(Op@t$c=0n5uR$%5nqH4ZYC(83HwR;NTTx3s43#UC z+3Kt6bZzey+I#UKKCx0~llFW;)%@p}`FbHL`-fq#c?b#!>tm70a$^7b8}a967mv)E z$iphG6CkM%7F+De;g?6z=4US9jLO70{3G>UJd7LOp9ftl3%EQsPS&(&!l&?m#NEmg zR-W-emkL*WE@w^C@BHSkf0l}AvKNTET?iOG(*x6x{b&0mbPG z{E)SptgrbT-uN^dX0h23VyB|cJZTXlVspJg@n9v~y3|51KapYJAqyt@NoY1ziavO2 z!?{}a!J>&GP_s)F*Us5NWFDuOnOWW^s!uq6A(tQe{rEdQHQAruZFHe?G*js$56=A{ zR|l;tqe;j+ZSdsISlqJ>1PBTdZ;hL<`rdo`cK; z*`&VW760Jfcp{f5_s(m5?BwI3J3TL(c#<^Wq`rQLh%S&+omH@}A z$e^DhJ-~MDLRj{|5d4p(qjl3B*4v?j9UC%1E34UbMAU(P-aUs_%!nobdN{tR$!GMI zoeY0A*F)|0xoo-jVzzhb7JMrA30F$(0#zfotVXovn_#z|7JnJ?_8{J`em9Q3Fw#{U&7rcWn zw<~d(*FSi8BN!TIz63*$bZEQS55*6)8S|L=sE|~^dk{4SZt|(b&s3JVBYuulcfSS= z?{co!cM8?rHF51x9u>cMheRG+$@GeI{_)Hp&ixZ^Cd*}kBaId?^VBSu(cKBmtV`>d zg-(WyYgH)oro)PxHF5sjSS#$@6^w%1Irivq7yr;pB{s7b=tbEPh?mVHmPua3=bjlV zQ60iu7RSsRJ>*?iH^ka=9H*N}uw`@;$&y_~JWoC(pMr#-Y#@ZVFoEQwXcKsH>rHQp zBOBWOnhI&~fKIqd|69`rtw|hb_0?&(cU2Dbr4Hc>$b+|Mhsf;Lo_M*^o8O@>3`?#) zBGZ<5^Ja0g=qrQ4pl)r3&^Zeq3$BKOfH10m=o)@b(qUfTjewhi^_Zl*92z1yt+$Rn zsfg8OQ^o|?%DHlEYU3gd)i47i9KZl(A_Q!&=KBg3(!R$PJldlV`b}2kfd3NqT4yoF z{rQ4T6ZF{B*Iie8hHqj?rzp%HF-M)%6F`{T8$=yU8nQI5(li@}sn!9eSuKL!!1X7W zJPLvpDN4+=M$VV?gN1%QMVQ&yMGIc%(UwqU_Bi)OKmSTFdo}ky21ul%Zc84bxDp$o z(nsWvJma}-Z^GGcCXs9N&8h32o$xJu5c?v-apJK!Xg!#SykSwioWaL5nO|UOu84d0 zn-lleUedpR7Ccj)!Bag@jr~Ulxjn}S<7^h-t$;E78@8H_^IpMZhz0Q{ibOzjHKY z;saWN;S^A)Rlp~< zlX>F(OJVZh8rXh+9l4me8?5*L;kUSchV?%q;0rgOesa%{oi}p_Yx!LkYtQY6SZzBt zcTHdNtE20X0a2rhgdJiYG{)KAoA z(&e55lW>+kOb}=MFBmanqK^2R>$W~|PUM%bO~GB#T2x865hDj;$ixIUFshD)!L1tb zukSej6<>!f9$8Dji4k6j_;MPa@d||_rnCCWXR*0I82&7|4iWn-Xj{=Q^qWAK-zAQ4 zjlDq5I(w26cH4+Z{&wzrBbYJ#h0c(YfT_Dr(GpV$oW3QV^YYxphl%cNxbQprZ_OPd zc0Z9pWmD;Q zp^LQX#vO9-2alxsm_eP)4Ai^PjoFZhi+%`@zyCPqfN2(j@P15rdz4Dv$|f%!EW&^# zjaXz*0Q;Y2k@<7B(9`ZUq*`E{_J;J6XVF)AsiWud(^EbTVg%u0M-ADK*2#D+f5v5H1$Svn>=j%(#~I~) z1k5}xBvJG733z*8Ik<#t!}f8GsnTf$Pj*W%g)ZO8Ig*15CpKaDOJB_JddS_mUcnr* zOi&d*4-?7=k^Wl8v2j+RHkY|@9dO2p`5!3$e2eq4^J$v04V$jV#~9a}boIK+n0pUt z-}rO*pni^DJYOE`N>Ugbkz82Du@Z&%mGYIix9$t?$H9k5XPpdW*d*C3oPOGiXucoe zk)RqP+;f)69Y4hk)Lw&DO9kdvn>ynkpa3FeQtX5k8ED${5x;-o*2=;8c;xzde$>+r z;wb1u3PSxrR_PGxtmHEBE%A8uxFu3|Yn;dBWc)s~;RhxkYjJ>h7|XHgU+d}mdla6~ za+0L0!*20yz}l8Jy4doM8T7XSeQZVM+drcYMx*?O*O%F(X=dzJ_hoGB%xV-qb&k0G=A=3{r;{6Q=4+SHO~s}be;t_u1`{`bpah-X)+Np zi!oGRhp$%uoBw@P4jC9ygY{CEL0~I@mC7|JBxAdkVf9+q6SdPtj0C@F6`;NLa^}K4f-)d(6Ra?ELmj-my0ym z&wW;K;Bq01Naa9kpB!z}UyXt;6y~~hlg7O*sQ%Ozf=*eGs)Gke+Omr@-mx8BX)GEI zJSJD=Ccz7>4BpD=l`vYl0v@hwZ@6?hn`d`704sXd^4cOj_&2ZRk&#$0)GRb2IrGnP zeUeR>y(1T;!iV7Vvk_ut8V9B)=HL_~F0;=u#-oBbo^;?tc&QrL)Jj=PfGZkNGccMm3;KSPgK^O(3kUtH+(lRC_@r;crtvFPnpqNj2S^x{mw zGw~Ia+$sTME=O}acMBc0DuHLarjfMRS7chmQ+#Q(6Zj7j;mIajT%zNGCUZ@o;PF}z zy^x41DHGYfGu7BD*4t@^q^p_V*0pT=Ofe?xtS4P|C>g>-4Irq0F)7n6B@u15jIY@c zt=wNoY_v0A?wJbm@m3=^mM&%d8WuC+N%72Ct!d0&y_L+W`JqgId?imxEd%Ag*V6ZY zbCKab17j_7x_*xfv$>zkgX*7w16BH@RxAX3-=;#!f>`ifv=X+-?tp>axp*QvmbmyS z;3~Bjyw>@Cw6p9KUqZx~T-5g@t;K7(bhseJNyqxFQ+%j)peWN!rjCU5F`nTChiPKGQrd#6*WrVxFx0 zN2+@fBBk{it;hh3ih6*2kud(2)eYby%`u3%{Y1{A1#6~mr{Qg@VAr@4*&?eC%PbeL zB_{eza#tZOM~nF3mF_ zD|;6*wR8t_V1p8KL0A~x&THiLb=`&rP=ny?i*&bze1o^cCUAPt2}|5rkZ#lhvq&N4 zhyP0u?v#On$@)wUjDp+w=X?Xrne?61OE@<_1@a4x>AU?)_!EyyFoDu4%&j^4$e-$q zbF6nT_Fr_E^G?c4TcUi#G$uTtZO2By}cVQJvj^FKkFgJZ!Q0JK@S>vghOD#MKU*DiS_$7#+ULH!jCbV zaYN}P-hs*~oNsy>>nQes>a9ry#WXC)cw{#3Rtgac z%cq{QXZi7w@tC)4F+K9gjJ8Yb;8Qho)-1Lg1y&ry#mgkY?B8y9B|DQ{*L(=8zU~7L z4QX=o`T)Ou#cntzTnwB1eqhDhZRj3g#rhte$1zQ_xt!+|h!-!WnxmTRPwfe4En$aS zLRC=6^dEL9aGurD=^V%AHvF7q$+&IFp?6ejL4NZZ(0aK6)Pkon#f}JREm=MdNpFh2-gTxc9AW6j+muccM~opK&K%;qPMu@|3_t z>;rTbQ}Tw(BJW(pV@?Y+g1ANmYAMf!EB~btrJd4vwDTUswEl$5>!vXE$0=Tkb6kV# zxyzstwu?BKFX6ijD>Js@e{oj1Akmv3K_)FRhGUDgFsxw@D^)DQ)~cxD_%BmpIKBxd zMxQ}VrKON;*#d#whPMNFkBdPsX&ni8 z(MZqkk0voHix?~0^Wd}SB<|2m#f#G1d}Zt>8f(X()y30PMc4+)Qu81#F#s)JW(Y%Iy9YH*jl$r$%5Y=T0dN_y!=v*p z(PPnh{%y&L+;QF&!+J-k^acx@>=%aZcWa26NIp@|bLNYh?Z@!QdhnN{q=cJA*JxeE zsirs4f=2VM&#%GMU6OF^OA@;({1w&2+32)i1!tIc)0_|MSQDBF%k8cc|F3qiP#S9HUp3O0ibw{eYdj3oD+=5(P0|rYxlpV%<3I8QnxR15PvI zQ>5|8IVa5iP=PO0?NPWd8_5k8#NymxTXZv2t$$7Y3p>beE^qq2KpX1~ityPVNAT&B zVMZO+GxcUp7;D%8s}GDrSW^%$(!dF4H4njz*$H&{tXcSDD2C1szDsRH?~#APPAK@@ zn`rhKfaBBc#MSmH4K(*ht(i)w{aBgZ?l;8iKDY;uTCE|sChL;~^K+!%%?pHVFT>!0 zk0^g9i~3uw>vK!>t9-xIqI>kNzaf%x%DI z#eOgvyns9S>9osOhVk8B567Ap!>=2EsKB#u@+w)3p(87pvJ%c|C#%Rb9y?Dy&W>bd zdP0GVCUQOSCiIKZ1pA^ye$3?a;1(vtx{fQc-75ufqs}4TL1r^OF!+nt6d6zS;T-&K zXIPzz8a%BdNE1gqv2f`!w(xo`TzeqMnjV`(^tSHe4b5M`iVy0r4x!eNGG>p?%T|%U z@AP1kLosEpjNz57PjUJ?0cdoNq$^JCCC#m#Wb67=I{e209?ZPOxde{$Qd`Z)x77#C z+C}^6<^BrX_e_P@<*$I1oIljMSq}QbHbQQ=Jm|eKVFb4%!~fah3NcR9BYYS5YWvfB zs?NAQ-38ls`~@fJ0-|)9^CY}{M+`Ct`I=EX@RGF@$aJlOr8mp)u*eO1({l|Q!LMX> zvc~AEZRb$yb8alcw2r;gQdVjPdj!`rv60JY~!9cKcJbp1ukK zQ~$#w&5CSN=U+6X^Lbl#S~Gdug)vBaDf>*@34<=ILbdWnteV7Q&kQ@W>waG%TN{7T zKJH8$DWySXzUrdAZ;}Cr|3b^m6nN;loXpra1(VbT`2W-vG9v9$@MuXBWPTfll}01@ zvULgyb^74Brd!+{H-*%BsDXP2*MaZ2k0RP%NXWu8d?lENw{5QRjlF*p?c8JNel3K4 z=D3E8uK}uFDMpP>ABtn(RUPD=v!-#Za}T4fBp&Lw=R^T+Ve02KY1)3v2=$V8_s*>piK~0Ht)hbs=zC`g8 zGX`GOQ}EV~dX8(Lz*^Z)VpsbcFo~fJNIHMfr%`9=nNLEXeJUE1jBoSC>jugB4L#&T zf)z;rQH28s9@13{bIke^gP=tz2>U?~PdE6(Ihhr3@Zz8_$4o0vX(uno~-Ne|txyO+9ChIw_pko?mXWe&dC z!nvI5iOPQ)c$;Uqqs05gxbK!IYqb6pG|(_2q!mih;> zRPS-KHm>~nSQ=kSR>I4p&GdOn7R-~VM&AB<{&uA@^8Q8(H16*Hi)Mu|D(6oY z*n|(3uvJs8)%Z0^NF?94zRKvVDCt_g#nuN9g$8Kg>RwQ#5zrC`S#y5$BxOOf% z+cuLuZvGzc?V-GmjZwrbD~$@fM$&)_vdnt}8|*(*M)~)bU?ImXP;RWEXZGHrGkJM5 z-}6609Xjd#-w$Y~;wo(TpaBU#Ift`x2DoO_lKpezNgvk*k6At$+I;*`|Ibeh;~!@? zE;-GL-HAjMa+O-WJviyMVJF<>^1@N^QPe1EAtsGPunYFYF|8irOn|>U6U<1$g~K9D@^`L#IV!}g z>=eS`xG7A{acwO6e1R1)IS;0%RG0|eJGeS)0xQdP!@^6>VPuUe8+CXIIFx8H>u*eE z90lh>d-f)rc6T#O=uVW#&k4=D3n`2M0+>eK}Rz-iRJjQM9wO87;%(QF2>2 zX1H8O$N96VW|0tFO*;VFhYipvFcDunBdp$jnyx&#k8d1MNH^Muvz?O;;jD*=#JkxC zOtWpNV5=PaBs-G*m;M#@6iT3~YZ5-y_(01XrZA7r=umq04EzdzL1L5>==p|Tl0Uf& zgy#7H`KtjZrj=12y=G`v{)CF|YB=H5d-~+DI@{kAO}3XzhVNPJBsHXfNFCQea(^|b zzON$jK8DgmuwQW& zYNbSRP$~?5e^KVAR8xAgyb^{EhG1^!U34pYfWt`*P%@NCeu+4tUG4+_vHvSra_jlyrfX1m$ptEML-3Q+cG$i) z3oI;CX|BE&>-xxo6#nrxGtLUdOK~CG8)rQ4Y~Lr`ST~2Z8k&R8x!!||2do`t(PCvVdRZF;m3))zNnh|Ld=e;uTGn@Y|E=tY%TWyd?3suK z>jFTP;|_KHmSJr99M>#TiP^1c2Ng1a+PX1V^H~pUguo_X-Q$@T~3 zpeTXfft?WPO)9%7ep)go>UFJO66dU0m5#Tu|=$|CY`doffsPGBvs?ZrK= z8=y(-JcbVJCUR>x!HEltuuoS6vt6s9=ZzPrq|fJHd6=*8fKrBxKV2k#pGj9|n=t&%*4e z7|<(mA{V-LVZ)j!tl5FR@J>R8w7#_9-Zc}z`3uK=w0;D&d(wgLl?AINM1ySd2E3}1 zjPg^f7`dS^jx**7QZCowzRpwrmgZtbaXvVS}9B3IJ ziMz95_qMIf8ttjfA|p-azp)gk_kIA*2N%&l%0+njfhv=;T%7vGGy?vO_#{~<8*+mK)iJnwn>Co%7huXP3ODQAc*<)JU^SNq z9#h$Y&h^8%R&@{at!oGr^M%>JoknIFI)i2k*=6+c_yp2c69^@F`It2##Vo&3$^1Cy z1%eLaC|Vo>D|SAG6+VyPPWBaOPAG>O;S64G!Vp{xI>`JLTgup7RAAi~yhJJY7j)m8 z3C#DyH>Y}dS4KFIRx={_#uY6?SlgkkK^~QQz)bN z9v2FDpqa==X!UQTCj2?Pa;YD<(=?I%T>25_FPBB(hbAad8G|PltFktCMxf)$PQJwF zI-cw&L6~pXN!AEw!s%t3a7U9V^E1_uQLNNsYVa_4g?+>xXK7?Ix&Hk`AUpEW$A}NEWPukMlSmJFIcpO+6E`G0xHk zuUl|$r=p?ob9O85jRzkubSSeKz1>)G+!ouMhRBwS3GlP?7|fde6aL&=&3+_J_;m9J zys0t-kE(3YIJ+Gk&$qyZ>O$VM5Fbdbb0xe9+|QO(+!^!}9FEjvyu|*(O35hfSQ3L7 z2Ml?;gX6$C?>se$o5Cx)5dx+Tl*+D}%${5)%+CMijk`*B;X+9_oDdPpN;clbrEg}? z9+k)RW}_&&aoIu?sE0Fms^KHA2g+7;&}%L4pd|h@slBO&+95O86O)uNdhRPs+o*{e z-D2RT#3R|sSs3kq51($}?iSCdF)oiOZhKgYug`Lv@o^K52a0@^?@@5^ayK%DZmeT7 z$MUFJjYq#rQYZ6X-b2GF(=)w}wVw3vA}uuhBluXl1j$sT#eMf(jL4_1Pe zL7(Zy)GYWA=94Ud%sULjn$&Qj3Q7T$kt$R-fAdAD_-+ zcgJl8Ax}r@TWAF`R0M{X?Si;~1n^a9K&ZNmO_O)hgU`~@xvd30XC*+KU?MsSy~35J ze!_1zTXMj`j>N0aWODc-fkgy9FXs4}ip=k| zL41#!R@g2oj}6;z6Wi;jVef-3{Bl(do*0gi-)3$gvZ4a&jmt>KyktyJy2f+)+yKnR z>!=bZ0siJoq4d!oI%AMyCEp7q4!8ZF!ovXuwLa2ivu9D4;CsYe+=MmVnheE(CwND0 zCJ@T&2Y>An^b_E$NIsLui3^W-Y3;7auV2kE9F2IcDPd4#5=4DeZ_>Tn3!w00IBEPm z48OEkL|m zjHxGb=oCDSRK12Nc$|vwaTOZ)T&4L3i-=~q8`K_OgD)?(^Dn0VqWk7nl3)ANA*GVS zJEj(HJ*?u~RiBCUEPY1Bon!kOcH!??7Oc9CEk25sMj?k5^xk|Ii>lpd?C&4=>)H;Q zqA~~jjgFz-fATb`bS7Fq*v|I1YO=>;UT|H%c-n6&#+1DH3(D;ZB-BZe$z6Aeq)*)l z6DLcO{i|eYVid=LOfV#!W)47gZ&Q_jApW-(x~G-W&F|mf z^Yc7rn)OSN%P_^?mBqyAsRVe8jKB=zcc2(y&;GSA!bjr2h*9Ygv#+TNtQ)jI+HoQH zIFdq?KXsFiNndDro+`}O62fPCJ84^#I~pj3bD7Nr_@jORkHkj8j3*vgEgsChAN!!% zH)+fkZ{b{L6N&%L_3+_X0v!7KiG)O(5FZ^q_KR+?K2N(2iGoAwG~A_aff+ZpvnyRJ zS)cPU_+G@7PB<|K0@So{PU;6VlUvNj=6A!f-c9g* zNyHbQAHL6%7O|seUV5>HCYNDLNfbPPa0^nb{)0JXn(%e8H)G1>1BUZspfdFwsHjH4 z4Mj1gX>gpJ@IJyw&FX@|f=%GRvXpqAP+$Z)&k;ch3!Gyq16Riq$)@3NWG)tB0k;>w zr>8LU#RV=el>@~sx+GL!E_$n8NB#Hqd~HP=k}BA1w#~c~qB`p!>W2sO$$lZ2>m7k_ zALqggMv|>v{SyO2?7(pR0_eIfg01GU@NaE0ImNS!b5$7?&TqiaSz93oKe zjX55=`~v(a=S9DCfGiGMPY>OfX17F}06*zDZ19nyPBY|j{+6}4bK^#$xTh7fF7t8H zi740^91PDbrTJUyD(Nq~gEar^36xvwiS?cK*f!A(j^6l4*tIW7;Gtj?n7N-6&Z#p~ zDr(^w8R_F%OhVxg>sirH+Qp3uf_e?J4C|4xG~>-V9?i&5@`XtLy-;L|=g>yKYEpv(u za@~MF?{?9>@sIgtc0tI5Eylosw^+4H8;zH)L5IDiu;J4(c8ixIyLa9Pyywhu`4WceOqIPir2TmA^gYzbvx3o7lZgP4CV9A8xyLfk!2e!~=e_a_Zkj)~&_ z)NFhpx}qUP`Zs^I(`lS!^Oj0`&1J^)6KL3A0)J=qRlCiZo~{6HCI$$nElMhvOpUxwCd@o~*mA9G$6k5+~Nkp{L**qEveeJ@Rj$ zpI#6NTsBO$+(@9?xt^5zksgi*<;MPqzmK|SXX56C+xVGo$MHyG9veKqmermj%d9Hi z3yNPHF;*;<>#s~WpYl+#Y%U4l_^1-rx4@_KFD>Wx)@yY} zI2I?sc0HcJmg%p8zcTBw+(w03h;sMf-&5$T;Y(y%eiX(AkMl&9O~;eZvvI5a30xic z9yb1Q#3=p}VqP+xCRRw1xWw}yAgsv#H`N~Vb)xb2f8FG!k-Dy$I`N zoZDkB#?`w6`bGC^EOf`T9M6zS9dLsyB}jT{8%O{7g{EAG3x|uzd}ZcXHtQPbb{3#Pv}lq;X(gFddhlf-Oeh za9S`A)!kgFZDTgAozp@UA8?%vnLu82>JZ1h+=$E(8Djcvm@iwTgZ$z=(~Y+m;oRNH z*cW^dA4PIUWP~a<-qGX9QQxen>{1pOn$dZF{d7; zz?oMcV3}ABERUQGN3Px_WaKNCVYkJ@N8+GZ*__$`AsY@}5QhARMQ}R!JoZT?!;DLA zJT;pz&SmMs%2&hq~h;F*~A2n zGhrJ-{pvfYp!+qVTdoI15m~fY^dLPJtc&MVmH^j@!lcnlUr_JIWi%8VfJFODxcM=TMrb>;Gi7o>HSZdcC{^S| z7Rtl3h!E5>%fQvSZmik+LF#(shs zotlxr4;i5Eqs7_s_qVZWdN6KPTMA2`=fKxq*M^fHW2pL)cB1sH9^_>Cc&}p~`NH|M zo)2>0hfzNqqmTxi z%t(L$JvI`PoEewy2P1`^y~kg}q0q{Yt;3PuF+je-fp71yJu?>v~hWjpUd zv>4m(~>%&a8#Y z2>rHw2IYts-ZTFfvjB@V$w-6*(PqfTJ76r9G>AU;o) zzU@SCMrC%2b01ynOc2~!u)~gX37Oj9%&Mz6+;@WKZDWV(_CcsAnSgT>y;;ez)9lYC z9zFg#iR*kwv%@PtVHf9ui}}N4?Kt*NboWw-QHq75Tt?@xQW(rjoku%btB6j19|`)K z4zhFq^3-iwVe3>MPpefKOBq7o?X0z=YRNu zK6@}%^a2LFpTPA3dokC<6wBK6asFqd>-NmzJVD&wAYTDzT-}OY!8>qdmLyu$+`>yG z8AvNPf#8D8@WXL{S2lem{!!OqGql=ZK*yY)FqBGXsA-_!HwAXdyBOA2ES7k#c7f{I zrt~E3p%+fRCnq(s8LJ24%<64#@!q<9T#@1pCU&0(a@ZI0U;cpbTjNB#@B^uybQJiF zqmcP;l=rl%k*I!!s&l(=*{4jbZq1YE-(K>4W4gfKgo;1cCS+eb2+Y9)D-rR z;}BN(#FN)i<))hckLh#;F?vB1@t%rQB^l=*!p8$bEdJHuS2 zF1#8V-`?RJ&~?WGnYmb3{hIb&%fRz;F_>~zpUq+X@lTH)@?XzlmPpNjXtt54_R26f zWD2m!#hA5mI>9c`j${o->zRiFtH_E?H>mTrCRi*uk#U?W$*j6@9p1<-2IbZ;YV*5> z@OAUZx>=Wr4l9WlXX?PpMTPW?y$3A3Y|pFFmSTP0jo^lBPx!C$m&5SELqvFQHk|AN z=(t_QpR_(7oTt`+#os5?{DTTs2;?J3lyLnzL1rfthu;Q1;;@MY4XU~VRXzCEx!{VD?l@<-bVW#%{A-%vP^;qBG<- zk$>NHVMfVaGU1LO&h`#x8!eQXjPxS7eq;~ax@ib54EEU9Y^I?T@7t?Z z=xJZU>oA+llv=7VeZzM_h`a^YoEBKWPy+;yIRbzFKcXAGlqoG1W1@8BSr?Ha>{K78 zS5F^?8%>(zmUID(J1?MO;nBqDZa-9ui8I06Z_EFA&6@1`%0@9hv51N&rqT)6w{*ZvHi&z99`WD zi?}^$P^)%{)Z!Je@pm!A`TV0JHW#q?Lp0qYQ4et+qv%9~E>f@R12wWvkax}*wyL;8 z!i`{>^j8FbxR(>SJxpGhYBD~N45Rlw4POim(JgHv=yPx}Zhf+p9{m}Op|}q+o&3=3 z&Ll2hVvcWe`}h{!w^9CSDCdn&CsqIEVCUmp-sjOXaAETfP+2cY+D*=4{gfK|;PpIQ zGB6+3RIW!ikvDYa17(o${vSo>{g36_#&L-xG7}jQQQ1_4`#er6NlH=~r6moCq+J@Q z?3s~GNF_u>T<39>28klHwb8C`X|Lyc{s2FC-S>U2|fvhTV{e}FJs5)H5x2_rt~O(b95BFmFswdm}{VUk=VyQ7u`uKrH=*2A9| zY`sIo9tuq5-2$iOQVm+DS%}8yJHo#>f`3Q3@N-?0uv{&Sj5q#_9cSC{miZrudS?Qe z!8I7PXB^#}#i+*|Z>sPh8!QbDz`6ysblzG6HcTcNvRVyKxiG~QlFl7)BBfLqGc%6up-+5P~I zY+NV2uP)-XFI|AjeKx$pU>AGrl#W3)As8$-jK9=TCu9&d5nXvVw&%hq?5^K~{91AP zzI!g}ALLl`$1SkpjUIJLtc3+5HStB=V|0*y$6joW6J~%hIK=t@oa?s1Gn+fuF$clO z7$o?gyIk1E+1kYaksQdE`$Ef51^%*f2OMd1rEHue?C+g}dozoO-`+@ofrC&y`V=hF zD}*Hp)@%rMq8cYG@pWGvYX50rv9VF?W4Jg!EM5^6Wwc1_UMoKN>=&Hv{FU{sT8%LU zZn)@4BEJ7So2y&%9p8$7g@ppgOKjtMJXJOmH(f~uixf}Me$6@*zpc+t>{!6R^;F?C zqzbTVPm--!(=YaKSPV%134+b^8;JLKU}X1U++Oqpee&1ALB%kTS{a3>wk+l68EB&S zqjT(N_8j!m7V^){Md+084msW*iHVjY*|}GZJbqq}3e5%Rs_`5hJ|AbMoBK(A_fMSv zBp1_!&bWnpHFJ+MzzVpAmt+-~j4|Sj)ZxO~c!WI}e-}Ti&cocFHF!*sWi%z_E>tb15MLhriQT5lQDtBpUJTWSZ)@(u zrARSWoEr1mxC4nWC#af#RL2|)%_-43+JNon`yJKmFvEQV* zN|??@ItcuVFbVp=%@T%2t`Qk@oj~=fG{(IUfg|sLgL4;Sh@LcAnbnOBSG0K58$#Bj zSjfdpJ0`UB%NY$T&g8#Z=&K5Ynn{OoT%2%6xTeFG9+82zTgJ2@ z;x(}v%qF!H<^u5?LcMZjX?gocNVx9;KkHSoqNf3uzc_)jR!%1^-SGlLcph&abPPmC z>R?x!G&QzN1Y?Wi#Q(`g=y(#`_~@24d=S`C#^(93oVx}RNt7LW{2E`(Ux7EIw{Yu3 zk!1Cm=d8pm0!6kKST-gSaxW#~mTe-pZjOSopL#Tin1mhQL!2FRp+sV?E}PReHDB%lz_|!Pth**3vRiT zhSTDILG!jmR^hgVCYKlC0@YUB+#vJ{PXTF*;8Xk9Q7WOK7SV7KN&&%=1KE!cg~^g6s0WJGv8QE=e=y>;OaO{N~uhz}s z$&PN+*-rVrcAWzI=QUbpb%F_>#ci!S%dBVZCj+zGh{WFn0xEgbZiy)_swVs)fx}q1 z$&ap-ub?L$6oBGYTiRY7%d{T)!q|zU=?uM%bffi1`t_IqhO%!WH3tS*Yy3I#VP!eY z*)4-RYz&YS&f)*I$>YuwlX>Tz*=Uk_1kKNw;xM%;{1_CCz_;MkiyD|CO2Fwk-fUoP zCSDCVMqcQYLc)eD_EBdlooi7Eoc?hf_r5@6nOp(+hjmHA$3LilasUlO`f;XA22TE# zh$mky=Yr16fL47on*a7N_&RVHC(W4TxPFxUE`q(Os&u3D26{*N6qE93BFPn&)bqp$ za`Et5IPmB@kuuT;^}-U+XuZQ8mz=|l7b;Y$u$+wByQnc&`sk3q;BrCs~>xwJD_@pDOI25K86dKMlzLuI@ug=?z%5hzbL_P+AhiO{icM8 zf&*uxs*>=MKPWo5ERc;p+S*_^;yG~*aYDD)UzoMr3vZg8f@7I6kP=`+(ihF7qe~BA zRzNi;64*uYfj#7oQwYpncoxj}Uj&r}^-Okoqs@}snOIIHg8lR^czIOB%~QI|w6a!U zy{3ZD|4G7{uTMm_xgtv2o569(1M)FcoHs0K!)n(gNNkjZp#t2$szk`~ZjOb$JyP^) z;7H_8?j*M*GQqknpIr7=!E@8Y;P|E*2>a|QbX>bQ$E`!*-JOqMk*x&w_Z9fx*(Us? zTrE6w!3rHcbV=~9_bhPddYmdY1~XKV%yL#Ii*M<|0!JBe7&@8uEGdI^HuJIn(|+(O z-we+uUxVW@)!_196ZXYi#*|qzQOzIE-EQ%CDebb;{ZNe=>AQ!!4X!frFZ7vBhiyj;l2(z9Fm#G_1 zJU^N&ZQqD<;~#NRb`GdhtH=MJebne_$dY`jVdVKY(4E_i&t-G?>B}_f9>&|eeyxK3 zx)(_Qy(ZD%p=vJ7SqU1RSkWif?PwUkfr}e@NK~DqPRGP6(94pU*rs}dgzf#pq-W2; z;8uUmsQVcq(@Wv;s|K)ZECJ6y+sLr|lVH7S2VDEGlyz(pG2WGsoQtxMYIvBxSUd|V zUwq(Jjk17!E0Q3gT#Ti9JY}vUm1*zbG3ZSSfOofeB4@P?lE1sby9Fii)O#(RIzPMN z*QQFgHAf2`SMT8KT*DlN^*v)=9h4ZHfC<7oaJ%>j`a%2^JZk#@A9pMf z?!{$jmKg>P@3nEgW)*wq{+oquu!kj++sOI>4SJ-e1Y{#lG27dztnO(tie5DfeDoG5 z|MixH&ll#KD-M&ML(+JfZUfhWSP;nQXk6@p${H7#cCFBrF1p7w)C}2f*Jl_Y`wDC8 zGw`jw4{Vs-izn`VN2!`zG^*1F-Gj&Qy(jffb6d%5NxKo@NY#&+NG+b;5qRcPEJb z%0jW6e&%^-oZv|HKno#Z;~Lb z34FI(8u#aX706vf+KKyK>eB?nA9o<7E`}AzWuz&PR`w*U%@BPcdIzq67U(G z3*6s*qr$;>YA`8$^%A^>y5gP{$eQ0)W5n11~Dsv+I7j_e0Z3$x42^L@^ zI5BUW9!V4b^rF4eIaE+zX*+djB#7OW7d+>wyz$5L@I=iE3>y47uLY%Se8euW-)+K| z)oOvx(@L0jK8?soIpCL4VYZ$ijo)h!UuB#IA!Dc#D#Ne%6Hn6;?!x_t!Q}hId~TfC z5;!0@b~I046gkO=@OgnFxQw5UB2R59aV}USVO%ft*YB}L>rm$UeFproKZa%%j_i`0 zF;yuT4WG&nz|W$C{H+J?uzQ;hwM%!WSEqD|E+-u&zE2Lb(c2BdK}}%$%}*z}-vl1Y z-|1vy@d;>pqe=s0J@Ct-mpExy09KX*rlg=Kp|=*|18OnaPnA!#*(tnvUl16FL;3oT zD`C<0>D0QV8DF@3V`nC;A{)czLYScjHtb&p%f+>6&Y@rs*Lq0);%EHwUm2ul0gMx! zXTb%x$Y4kjL^WO}<4+)LjGG8@S{ZN|R zY>3^n3GTd>;nOBnKsWp%u3w$$%xPZaY(*YvTXLH%n0J7y)K8(Cbk6fbS`YHy{~g3K ziTODC5x|d86X-O}WVS;q61*oy!m6{=!78kRd|EibHTIO?$QN_5^5{r@_{~sUd^>@> z5XsXiYtvb1q8fA_Hp3@HUopb!gfMG1=Qp;V#k4Pr@b341w%;I#9p>fWaDR!NZO9aO zfA%6dR~iEmdv8M2f$aKO3TkpyqIn-`!T5$U{Wj-0I<;pB z?l1`=iul0FH=6TKp7)5J?hue|9RYjpkH$|5y8I1Gfm59@3x4Vd>!wFD2_K~mwr0Ap zn=q_cHXkpk9EHek4*mvZ@F#^cq;i}c+~2o|y=#*O!+|`a!=J%xN2*c#y&25$IFA=~ zYm`X>G_~*m%w9PXz4Sn#5tD?os@8DIc$nzJ(59 zzlcF`dE=p)OZaHlKK`DwyTE)pfb+~bJQyo<9!*U7N7HI?R@hhkJ68(d9d9BpD^kh% z|Nge^AQ3JTGk zqyJb^%=M;-m-9r#_9Kn0~uD08Wt ze!LMaaHZdf(n6A9?OqQsIyfBL#P6Wv7~sb4uSwC3Gm=1fN1-X{($sfv8NAWGNPL}5(O7RW);B)1d$>V?St^ewi;hX* z*(Ph&xu+S;CjLZEaYbGs)STb+<^<{5;6VP((hci!F;mFfKSZ-}k1BBU_ zQDGEqTtC2m#Y*Flncj9k$3^1znZsahNHQv!Tkv&HogmUK1{xHelIYQ{U@j084W0(z z+R*W6qx24g|CsZc3wTzp;tIyP0>8r~3c}qgn9aoZgwM{SVk@M`nb|@PYk@1|EteC8 zPc|SI_SWDU$q%5vexhK_6=7wljF z!$kg)uvbp|{sN4py^(xQM+-Aa?ruyBffH@qv@^reNqalAosox*!&BJg93caL$PhJD z1@FaGO**Ff8;CAm!Kz))MAPP<5Up^Ggqb=fK&-qepIro|@0MZe>EYP3D-4~6Td^1N z<8kWOOYG>O-`uI~32gB8I8jBXaQ2xr4b4}_K*5MpaC=Jxv}`;8M>Qsi_Fl2YGT(5z zWU&>zEISABi#u6MYlrR1-8V_x)B&7+c`R+)IgkC=CC(n*zKU8-e)QGfD5|XLkHZce zWna(vvX{bs^YpbeU#cfTt*_{!%Jti<#A_GoZy(ApDy_!_eoOK9e_pQ@cps_#6D#oMhEwO$*)TmO4Ti_b;F{qMM8?Ss zcD<0NgKrnWv!DL_D%&o6yYeb?)QrQT=+)>waVT$=>q9=rZfFcjngEx_#$zH~0Ct0^ zaACeXoxbq~%($Z_y1#riPCk?`vMf7{L+(h!^ARKL6mopfH*gpU#bBz)Plqo5B$j>n zpU|6}2@{Gg5{}K_b}ak@E_;eWHaJwYDzvk4Qb8nEIxrS|MH|x}WkZhmU!wHXiv?R~ z@tJm4krY>e_6#+ANb`vDlCMmPY5zS}Bg3-rsF!yOHY)U^aI&xPTHr9H< z9ZhMbZBxdUYb|bcUv1A%-KWlfd;0=EzqBBWE-%DYPxL`8^gFlQgo0Ys0{m$p%P%@y zimSSZ<3iayD$~A~-rZ`)CTnL4xy>Sc{YhX1L>|XREk*wJ?k+ewS%MaySENBgM=k1I zBzL(-g67x>Gf}A@oXJudQ0N-Rn{l=H3$G;R

0TB<9DmK7`Rg&f){1L`Jz1Z?T*6N)~DHTx`X|kD6j~m@5A9fOG@3G zg*?GuR{B}^E=EowAMX#NIwlX`d0!zIHiUuvruD3-@+8ETE5o$XD_A(u4#)0p#*^Y( zh-S1b`zYkBcTZ8GzYPLuQtC-8H@r`b3Z&`EYXNlAsvB&1>StEBQy2U?RG`Dvl(xSa z&%en$0;&#XIAHox^i}^WaoOC3uZk6D@5nlQk=ey5zfp!Ac6Cho=3Y#n76!*;b@6?Q z47>@?Cv!q-Kso3wBslc46X~b*2mGhPpiS~zxjzXzGT#2E`}SHhS52G~&MN3M3n z!29NPOn4jr0+@1GQt>PBnWy5a{}rk@6yi4k1Bp9(*1t~9TI?i}gq)uS7o z!Xas&kKh>yfibCC)NA@>cr(<4-kmp&21wom^W!zl=lD$U9^ng4Crs!PPk$V%8V}pM zmGGMNY`nO2Gpy`vz^0eS@owK&?u_Sa?8{h8a;t@zt;v0ot^W#^RIfnW=5oBp|3tTN zeUMb}LZh{v>`-1A>lJ>7DH1m!D_R^r#qH*wy(q(-(xo`!$pk)q?H4{LK%M{Z^C>cE z&w8PkJcknwtJu27o~pc$#Q1N&=%P_EG|R^a;zy-`a^eygb>4&~Jng5iORJ&lc_49* z?0|4RNBZ~4GwQj`mX6fZqE&(4AcPL5t1@h8*ynKiqIn&@SwEbAn|g%&o9_a?|1tzW z>>BR;jrTa*yn~HQJVQzXRNzA0dDO}u!tXQJ1$Dm%sJVI}=qjkg(R(ti#!Zr2K3GHs&?eClDjOY6kKNq|HcA`$x0lL@>B+HR{L_~_nq|pUts~j* zo#(Kp`#qCT*#R>4ns8R#fo|ESKm+C+2ZuRt;1SmY!T084*VlTOc9??@pAb4Y`zb76 zR3UgBoFH}oRq&V|#YvsYAz|Absh(&om9_o_CofqO&P|`Lb$kS4Yc8-&qobg>?GnlV zc?r2<1O8O3EblLu4N|{`Q1uzUU=QC|;Mzl?B^^e%>TfN67$S6yRvjkw5`{$Vh5-hR z8j96z>+#~{N`NElp!-iQ3|?CV4PPrjqBNIo9^C`GzSV%*{+aM&q=P`||c<>ym+v{NS^pV8&#WxTZLV9ynBP_h$#^%RXkq-O0=yhAj z_gwad8G9n(_twRH=;HFmitATJSn-ybrk^AUHp2U*LnJ&_RO81dMxozj8D=ljh}XM^ z@=xn7lO{W4w-wfsd#fyj_lhv$^2458RcHnC>R$3RssJ><{wFLCxVb zym{_GPC)^g+?fYCGc3S)#yb4IABe)g zA?8qZsNK%zr8sXNH4%UQEo6E(%+VuvD;iJH;_rU4!SP+!Nd0GZ656Io=IKRqdRl5W zOXuA|{d0HmLjPMB@Av_%4t--^ZAQahzlo5Z`kZ{M+<{pZX)MgMhr3}~$hud|WpjTT zq3i8ZIBzu?SndQ^XWIwUG!KAuF2`(kO%r7;bc2E9QFh!C9Sncm#C&%(v;EzLf`|A6 z*|Gfz8%$Kfn@{6O+0D|1hha6e<&G1T`RN9$q9f^oZ|&57%`=Ghyvb6keDGldhZp}k zFgFhkZiZ`#=;y#SqIhQ}nuU0<4Oge&yjdcEMe!irrUlh**Bj>9trQiA1_aLhYUuEY zXO9+LBToG@aMHtdI7@Yq6<&*go9klGH7{CpXypaMEqKPh2>uTXec_GwW-j^11pn{+ zQFPp$pCI#LEgZJYA~*WGVRrEXI=4`fdKsOBKN|JWe0dS||1*IuH#_bu2e-82>tlj~*mD%;v}F|hxM@m4nT={^(?CdLwA4k`wL{SHl^&8TWp^ zO78aSvoXE$@ayV0EM6H32JhaJL8%)==D8_1)AcQ}+qZ#NbV18Ja_UJP?r$wSm*D}mof;9|^a)cbyr7_3$ke6OqF>E#~6*&iqW zdW8I;;vW`gGnu!G{3UQErTC-QS9AVzqIqjKWxjXO4W>NhG*iB1jGdXMS=y49WU*5P ziO`=9RprUB%P|qUbk$kvnrzI;;J6V2SA2R)yeRk5R#vuuxoG~Qc7DCiGpwHfkJWdK zLCxF?EH^!tH>!V+RS&%Jmd6vpjVEM_49x^rtt%~A>kg}9Whuvd!M5BNqMLGw&=x00 zSH|wa)wi-SW9&pWf>Xq_`Q2!Au7pMX8^n@R+o9J~QaEcTk~voeOr-N|HhWnqQ`G;D zE$>R;4(_?cTGp+?9{wIYwB1kM3cYD{zskvPpx@UI+- z2E-dys~>`6>Pf6@y%k8itbw%#Q*ig9P?60TXL8SQ0GwCd6WJU|f>{^`qZizO+a1rj z(JmqI_iGO7|1pQaQdx|snGZ4>nxSHoaOVk5h44IQcyMqrJ=EKF4hJ=3jDMGOr}!3jq5-yA&oPh zvkU|~pWW(!u{8eOa#$ih61uF$a0{yav5~vXn4ckjJJ5^9F<;2L34Q`EbpPM&WbH3k3VnDUGCYT2K} zyWE)-n^BbI2TQG2fXtiQ*nhSdzwG-$3WCI7Zq+i-G%|-*!%~C}#1v9(^A4S=n_*!5 zU(vQ1Cj{o{E%rgGmRnJ?kg8JQj`q(C?sx>ijWa#q?Ikb}5(J#n*mjJsJA?aI?FPA* zrP$@!&vt#hW*egKjq&1pMeFCS6iqqUjk^oB@mreCg4VY;NUrCB!{(E=eBo9F22oL>~2JLcB~CM@DUhg)cRs>v%2sq$w_R z!AhC)He(#~pbid{PKV&)o0t{z0#^<5#jCxlcunNUe|lzw1|w78iPtpPCiE13OruOx zUB&j=n3UPm2b!MJ}rmXt>KeF@jn&NeqRayR<*MY zt|LXI)tVq*bDCWn8qRfvh(t=gQSfuCupa!o2x$U;b5{cga-Y1(y_XH}nYSsjsid(Ocedb(}f*@0;-cq1OXC z0mG@;`v;iRGlEYrIZMoh_gRIF97<(K@*DF0klju(ELIVqVdgTyJ1y`HysPn}uQDx; zl7|-_A%uMlrmL49rcH%CIP-J`_!QTYqRraeoc1X0y6FknE^{9BhABY#e}Z3aPcEzM zkHIBZSL6HRuh~CIZ8o<=8ZBmhC9*LYpb#iNK!g`6|-5U4DR5-mpb&*yP`Q`!xTTMD?OMuH#i zF_GUc-Zx}~{ z9BjD_k>|+~GbapNWkJC4x=1yokW{|vBmN&QvWSJP>|4J+@l&tI{pOP7h>Zka^6VO0 z=XDIehi1c%G;?@grUnnse1gdIqi8wFi41NGgIgxX;M;XVaC|QzjWkH)YPA{1^+#g+ zR!>YnJd{6uvk9yWtx*_Q;bC@wl(~%M+#H7TmtU;qy*vxK!<+w#8WiVZR{02+(Q_M& zHfS`aUm4Cj&v}Pku3_ho++T-Z`epdrH(GJzf`zzsqcS{eIzcXt{wr#$ZD$f0U9dFK z917-jk;#vbLzL`MXgRhW6l#Tcp}Kd%9DN5`t-sEqmmeT+rtXH=ONx+?(njj0m60jd zzwzPIT<{V4uSXvMMz*(KZ=K14mo?gqkZ!cdm8=RC~1wF&ymOQ7e&C|nRT zh94F53)|kC;O(JOa3ReA?zLS5DF-Lft-}h8-+2l@>~F(lS9knp8v!c|jriNYLzu$g zUGO!lgSrcoAv5Cz3z(n;-cSFrMy+x9$M_Y=j5#9u5cd)%NVK8F{0HDR!V|a8Jps@8 z3lM+w1THv|$CC#_CqiW`{J%q77r716PNWfiTXFg_`5bhVwSdC4G!pk%nOAOC6F8!y znYKbQo*$JD&+4RUX~$i`ZRQHW?=9)=NJ*CIDUBU{7diKDt)gsaeg2QXHJ{S3i}aLK z!pePt@UqVv47eaTbgvW*w(;=h&NKL>8wqiy!Mxco1BCa1%{zL<>5w%6hsmpmNUD21tF=h>~d=kV}CGrl=c2_I!j z@MX`Rvi?vp&SKA1A|LvfmE|;H@^W$h`Fbh5*`0~X^P=!_?k#fqi6Y*Kr!>hy=rff+ z!10ef;m-UcIAz*Hvb=c$nq{q{V?BSsBjW(}L8MO%uSW{5(J+!8d;vOdG(tt1F{}F0 zPn6DTLe|~e0#By|_N&~tdzV-Vh4~fCJ!>jy(E2Aj=k=8>^>ZTax0FHY#sc^^tdp&u zIf8l_e#f5$!}+P|i5Muglslf;fJwoGzbw@%8tYodKGz(ETrYQMS~wijMrLDWcM?k+ z`-#*i9f6P|-t0@W2*(|J{d+Ad39lp-T|lzX@U9xZR)rqf(;)wo?h^~1}$UH6aKR;%t<)g$d;XE zSL)9|*ak0%z8r~`)P}3!P4Prj?;OTI5$-7Wc~FpI2+d=Uuzg0&EXQ&I z_GzYqZo6>CnQ{jD_9XK17qf9w{4FBk>(v8um2F0%Ov=wyBC=xM*$@H)Add@=Vqan;A-vH}o`C*E%ul|wU$oz-L(JE&Km)1H7 z9X%sjVbMTMA4CvmVg9Dn`;wUpv!A)D_FQ4jI@ml!h2FHi24+QbsQuYHuqn12UMzb6 zVjDK|R;EdO@#_(M=eennu{DN`{P%$MDk{SEq;hVOc`!+--oPA-i#T1WJvcp39)30{ zz}Cjp#>?uGILa^wSIw7$Cdn9XOywx>|7`)+I|c5I^Qrq|$QZJyg zh~g`0p_97j9jNOk@(VQMz_c;blGwT~_ za{EEr;TLhqQzg@Cr{Tfc*WhsEA$j!mHE26d#K_E4wyjuT_mxh9pX;o7bBz%YbkK_9 z49!K6JvU%yN(PBB6r2|o!EM@1sO>*bdOsZzGDkk}HeZwd>2{{S zU;2{U2Zek8wLG@dbtoR27E3m6-p#KwY)2i9G*A=oCbtGkL9rnU3>8iZS;Q55Xvzfs zm-sd8TxiWFgjT?*8AT2H(HiI!V1*qz`^ZQIOHO0zUVdP?5C7Ej1FYLInKh<9;#{jA zl6SsRv~~YOYz&wVal2i~&AS&xtFAx8$Z$l+OyUtJBrYw-!@h(o~Lb#PV2C zn*zV%%LcCQZYfsAT*Pd_$(M5J3H-2|2bb$F!7$_7%;%;jKjh112>#~8x*SLICt_9k zvB}5rTY)BT`8kOn2|CGAPaLHog*fW8@u`!fP>P;cz zHXsQ#P{e(mdAR5KRaBm^4~(RDaEd{4{FXj>DEIf^RCkWyEjN#bJtNg%&&zCBI=dJ` zgx`iw%qUzwVK&y@%qO?K@4(lvk&xi~i0P@0f_0{od0UzFG_d6>R5fP6SMUe>$x>9( zC>z5}Bw}IHnM%8{{IF-OKi_jb(Rr{J>ae z4^+0VxEI5pe)ro*NAN9B-$N6N>JY_Iy zJ&Ez!2K<=xxA6O?2UvCgXnfN(9Qr?h2T%17TD|uH$amR@lBOTVMQ6OBHFyX3+!)3m zvh9Sl@eVNBuMBl3WHwCB{s!%mMJz8{2CswEoDdCoed!Q7MgG**w!he2-uypQhb~8?lwq1KnR{W5F>e?uB z{j4R;-(P_JJ1cP8heJ4QcL}b%Z3+L<-m?MsMf}-->vo;yY4Bu^Ei^i}L0i!XS{f3@ zcgtSKoZ~?(G9n7+t84I+(slXJ1}ieVIv#FLyv6k`yGz2-^I6(#AH2Kzx#;0Jp*KCt z9F{edg5{wpbajyf-ovN-Ph|>Yvlrod$zIW?k5#ZrSsUi9z5%=QlJIExM4Ws58MFy5 zQwN8e=9{$F<~@)e&HVESmnbf69%u>loI*3?V@ynV`Y*Tf>TZ?(HiM! z?2sQIPaVhOifm74A66)wH%(!?b0_F$-xsZ0pAAR$9R-bfih`SG9~R>xev;!+Y`wk^ z_mB;##xi0i>i_W3gh>85Y9nrXUw!mWXsK>{O^0+Xh23tLEkQsFX zZ_#lJWJd&ncl0}wYqf&^rd5h5a}sd(P=8qX^A&eRA_^m(f5$xX5+TV8Hi{47Hz>;S z!=)}malSvzY|)}Vi`1w}TMyi|@}qSD%2d160IzlBV?+y^to~T`o)A9dXB>pP8@4Y5XaJ=`?;Z0ZjxzB>oM4C0|v`zvdhvI(0la@ zS8}@>?w_2@M_()h<;ns^4fl%7Dl^D*aeusHycQ$9htpcS|5(}C!-y3Ie5$58U-{gR z{wh$XX~hR1ewZ}-A$AZg*Y?BH!Kdtm;xaa~-wFIwig2PzKeYD9kclUZc*zSQ(y~U5 zJP$u5lAM#tLW4C4x^05EZh_qu2tu~bos4uC3IqI7NX{-GgSQ$n+4mXwV{#RWY@(=Q z#R5K7J^?qm`ay%{7g(h12JXJ*C_ooLA#@eGC)ePH3(lOLPc~l5jK$Qx!+3$S8e H!7Ic7`icFUJnj`tKn8?(PQn=RcT*#4&bz_zt|eaWUgphm*EC zhF?R|uz5EjQ?1iL|7-ycI<3A39k^7jE4B4}-TKg^GPD zaNRlt_J@q2L$*|-!ljj{`(&ozS5L+6_Er2izrDmaW)vG^C1AYXFGstJ}p)7BFyVgCAOq5}L0|PTyt`r<-i{qj{GDI2?=>zg%V|875xj2Kw#icG#D^fZ zt_OzPC;+hpe`ajs$~?^XL2aN4H?~4pf7hF$>&LHTob^sX_YtDSb9G^Ogfc(MV;pXI zE@UJgO(wOyG1z*#5=Q=Y0LzR+uvN8D^!k}Nhz8$~60u@PxGXF1BQAmI-wNDV)dY7_ zn_%W?9-5*TQ+8!9`yspq+9X?8ml@6hI_SgGqO+N@@{j~68b(>+)4P|yNL{M8&20>xB|wi zUm><>Kb;E+`pb6=Y)I?}nXA%ZWtYO@K1>jm{j%qa6;Ba^^;O86{;_kV z^D%E(6h7R&12@-Xu&a&5_@aWs4XXrvG-^2;-FyPqmLEZX6ML2;)4@*l>CzK(62N0- z7g(S72AMkv?1+XZ#zW%AL-vOoK6c z#?V&Q1m`^R$WzrD?D#-4d;leW^Lb~{qnRSmcppZx-@hmOGD|=r{V~kKU-9 zw}=w6CE$I)LG&8$!=*@^7rj+h!Q>DF-Uf3q%Rdu368^xk$2> z5e=1843d)N1pA2$uNiNQ!*+FJYjXhC?ja5P_rAm{m50dm88;bw63nhGH{g4IEnzKz zop|iABOle7f=-vFP)D`~+TY}%<&-qAPdER%wwnI4X229KTbsHxDM2`a@?NeZl?Di1FCOoYp) zLt)A3$Ecp;0r%I05ZO_MEW<*BzOmd#SDai(8xG%tmrcfW$*U&%%xN3-5B~sd{0Wk{ zHwLonLdl~uY2Z?I3}%{DL)9%+8sB~&Cb+26=2uH-i^m=OW~@&vKuL6H#&_7dT7r)J z;6ib2Fj#Ai!M~ycoY6lEOm?ZE(Z;pBn$k*$G_(>pGHKB6(FO&RwquV)GAx}ZfeW6> zL-+{P21)K7YFv&NWfy=MG=Mo39n_RVE5pXe`au@ekP^s|I{u zA`6XqhH#_8mLnGygKEoM>T?dMTBth}3C!?PcVQi=45mZkBdF2FGLQ~%q?xsgcp;|E z|IjtYfb?n{`$(F>`jo~y;ID6iKT24De)nr_F#huqdM|~(p#?At- zibCdgzXn%~w}4i2S!^(kM2o+ti1(37;vJjG=IlR;UsAu}-tGCsBg2G$aWxII=Zb^2 z@Rkv|D;(}<>w@Z$O(gN>L>#w!C4|je&LocP!`S4dm}D4*uQc*;=l*A?c)AY4ovy){ zoGv2Sk^*y_OF<>b3AVd zHMfeP9*5!T)b%MNP1^}$Mdaq=+<=-ubf_kn=7p=5KX?$d4jjVv5i?O;*BShd1wxFyu$~3Rke5nhKMfmr`X_Pk}&AYtX2nVG5(P(EX zTKX5@<$E?bQL`SEPuHM2Go?pQ&w&f~hoQ61Ad%l;3oXxttn=Fe=GxJNExQ`gWs;D+ zdiNLYb~T`;q9=Rwx6w{TJO8h`km#J2AR7%!&G?++`*&AWdy)#^U< z9{T~?QtuPXw+m77_z*sGdI-k4m}B*1DSYS}gFEHY(JfWPo%o(j++I)R{O6rw&8qRD zQlhEVdWaKQzu$5;8k3h{*e+TV8|NZpi{ z&_kf(B6Oneeq`G;O4v(Nby7K@hz+S5NtKi1=$3P0q5r{v%^5wy1skgbUs=;inrgAZol*0wRc^b7`r>PPwFc?K}C zMH07kDuZriHMp*I&eO(xCPIF3)Yrxtj>NNdO@!2Zl$pf`rM=FUm)8(02wl( zaF5HYs-6>u9*yeAxB_UdqZg3D2wZkHweUc_TgpDng-718MgHi#CMey zpNTw)p<2=Kq%;NO>vqD=KjNIjq&WV5WymyN*VBJq6dN~sqfcOu;N`e>dhk>fa!zBp zd{uW?8gLYTD@T$0{c+sY)i+svU^*%Ax1(MYC(%Q1_T$y%M=ukmLd>|&?1w}t+|tOm*67XLZnO{z@Jq9F4J>?HvCZn*XMh%Wr-TJdyauG zk5-^`^+vEbpIsGZKaO6VJ%t!nP@K~A2t%8Ku=#N~Mj0QWQ=D(pJ9hI>B~pd>bR8Eg z+eKm5vQy-|>mj%m_XmFM|AJy`yD&Ue9?liT(r*L7ShH{g`bGxhgNdHp@t0cM?nr5F zue}@$T}tAX{I+NA6>KMYl2<4bzwHQ~Sro>(UrpNtLu;6~} zI|X+;o0&J}P?M@-0*}TBY}g@fJ7O}=et96Q)S;qyP;D-&e>sa8T-(H?Z(X93N{evw z7)SJW&cNy4D=vxVc;x&8(@bjbcgqP4ElA981GOD~LoS^I~n>$bOa{lg*X zKjw@@&597`RRYFU6<~g}7IQihiSQx6Tm2(OZ#W*Jd!?qbHQPPN1k*ffchiHc4VmY5P9VP5MMqN$203MqT9mJXdG9Ia~Ei!+*}1}G@5sI zoQ)!VuN26JrD8VjtDVS)a1k`T90W5|N@2l>6c`;I1&2|eXzGsVZuUz+OwuO&{xgM^ zR!YDi&rw!%4n(U43M18sbWAKNrT<3^uUEj*k6G|OPmL4(R0Hov zM&owP#Z>q=3j*Ew8+d4o;QJpnYB<)L^)8ypIj||*y`z>m)BZ2*D&^-*)0g0KSS+#ptAf*R9ID#GO+&%tVCae0=7yGK!KMEa;kc&@ zOP+p;{SYi*AEZ6mv1`V#tg`}gR34K(#yvDy&coKL^%ctAwgc;dDYSJ(K78}oOTCJ_ zajfD1F51>7{3@qGkDoGR>Wc=*(}ZFiTlF8xEOCJL8KU42Ux^hEPYOd;SDE}<152kH zVRv638RmJzP2G+3`b-%Xo~y)$Peo(!LjiA6kHv|;+H6_&J6iT|F83S*1fymuaOMLS ziDbP4TPSr>xIdE5ILa#EN$oVe_16(s8avSml8?yJ4TR$uce~UKSj%4~D6*ls$GBfGCfQy3+(BNA( z{6F*j#>ojzR2R_h=ZKrSR?|b9({ZJ88(bR@;-&a@)K(tJ{Qpd4W0lSE%$y())+S-_ zgL{~%c8VH@oR@vIO_u${oT7uR`cS zQIwI40P|HkT>O_Q!1Zyx(=g}iYIXX`E;%5791h(h7}u@@OSSBlpk3~RI1jIy>+d4V(M4; zWjRK0%_I{Y9^c@96HXF|v;E>9+;JQmacKKhFd>6iA0;Y(2eRq&p<7{ z8)-*Bw3iAa_&eP*ZUaPQj^x&9T?d)2y|Caz1hj5lOhogVU`r2$=_zvDWOEbFdP6?p zdX7V7v&CBxUiFR>BiU)I0| zqZ+XIZbPZKG>hDHT{tef2t;*u!`sq2o=+1`OD``&)h+|fU$PKNr-ur|cYnpvby;NX zeF6SycLB+J1UJ7rLcFIUxzOju3Vlr2s2vd)9grh*n?HsfNj|^~l{_JBmOn`DpM?-{ z0fv2CA+`Jqbh|cSnr0?O-0h|D6(8{2nfWZkr5V%tJ+k7_COW$7Eg5?t9)b={<`Uzs zLwce$(|*HH_YnUL=kGE(jtb21UM;pQNk+?mUDSYgeLd{ZV}3uUW2L$e?7C@$3ry2V zY?D8FkJ?49EL0>HwT{54&;*{LJ3xLe=n#DA-9i?RSPOppe~>W#U0A`-z_~RetHiJK zbM35QBKd4T#?FX@=lVN@cYo(n2m1&izHmi%yYb*JF_Sg+n6bl+{O#YMi?;g})2{vH zIGm}#X;m3;-*2Xo*5_m3-qSmwYEe*w14+TtL8VCE8#NgT8HkLLFScYSR0j;)9H@!~c<#ZPJF zY+X{yz#F|&A?k3R}i>3Z{0!S3*K4ELIV<1WVIK#M+#cKMU?o%-0pcX`^4zmXj}Qrw`G z7_7e?gIf0jvHiHMzGxUEBQV#xt?jtGaBp=Slwhhk{M0bO*ukR{l9{sTr1 zE~K=&O1NN$C8V2QM<4#X`}H|TcZ|0q`}CHfo<^eJqFXgkXKlgry$OPT@dnWTtISnz zFMwys6o&cC>-%vJ$i@-j{0&Z>)a^VBd*;;(>*`JknkR3^kL9H>Va##rU(yE8N~%ET zcnSoa+`yKFEX4C-aj@fR6xbSs;4Qh)>`A8!CY>1%aod+;$9-{jEa(kPm)rzPmSlp9 z4Zj06UI2AdL+Px=-jHbQk0(y+2sDGA!-_||+q^0n*F5tE_F*lKRgQln?X6yNOhn`(lK~`U13xV^Gkn8F7FmH7<7ux+C#EeEG_hSU7QXeZwI^Pai zx3h`ZeIygT0-=X@6KjiR;k(AIM6<9BTjU3-k|cu!fd|cK>6R3#$9I6et21!^B{^EP zNk|lRo9WU%Wn6YVml|})(EaC+Qa{;j;<98sZRN7?B8w&}wyg5boMccBpTaGT4TgxB z;cW3^H{#RhMwDVx@tq9++jc7o=YJEY3C^7)(djk0GwB$*@bh>ydV!zeW$;?8%Z;1w z0|WM{u=%4u@KkSVv1$^R>f?`DBaNX#^99VEDFTjZcZuRtf8p6$o`D%vP9}_dL#lKQ zp?;zv^l`bQLO2tqdaH81v9^#JSw<>TkHf<{J5K)XFwC;B=eCG?!m0gp$)UUnfRaU2 z>D?=V$0#q-zxO(w+)_`hdtK?$!O>WB;T@W;8pqz;yNGd@RLD84FQ})xhS@y3h7b5Y zX7EB$>{y_J$C|&Adr#IN*G^#lB00)k_oJdYyr0qe7E)0Gxhhh^_m;O}s`x!LB0hqS zKdG2%=>ktGJh*>N`of7ms_a47FnpT3p6loe;E2{#x^JEt?8z}8N)u9vt=cT`T-^bU zH-~VveH+zVJcyS5quA2bMy%(LCd-s|M_HjY9htrXgQfT```{Q{clsOdx6}FJGkL%;+#-@8qV}O152${xbfdTF<^TU){ozSA66=%!=1S}^SKyX z<=-rP+TIJjmycrF{I~RULo5t*Hwn#nFX}6q*SMwX1Z}nx!JYqAqH&7}lsbqK4}1tQ zf3D$&suhs?ej{;Mm4$ViUcjOz1F+uFOzvzL2ThZ+$>ha4I698O6Foh$%V`!l(!qN} z+xp0hvxm_qL=D0g7{l=U+1%pIhv7lQK}eg^Nc)bgfXYKcfx6rQ`X;1=9)6igw};5! z73s-1KT94@yp$2JsI!o9y^TcO?-M-z;)oT$mHE6`B8} z65m}hYW-!H9VvxV-ah4-GUsX7gKM~`;1rbKA5ZRY+=_F4I^)G3x)?D$2m1LLv}Nr9 zVskehUD`XytZ`Gw_5nBHu>uC}|GDDI_@8{2$qCBsq@b)wlKUdqjV=m6#soUS%R?rh zd6j2kJbgn#zsEJu)o9Ku^7o15jpO=~Os!h1*z(utFK9MaRlTJ$)X3~MPHjKNm znq3~ggIo5ig57T`Dk1TR{L4=l7|*D{OLagV7VzGOSR;tqQz$ghxhf3xk*CR5B z{BDyqUSN`*!^4q|h zop`#xzLA93uYd*9V!+I77H$&;K#X!BELiyu&uof>&kmEgp3#FmjV#6XP4Y?XyIzFp z8W!Lv^Gvv6r3V^33BhMlV(d+p3QoVe5I$M(9nP+3L169~nwGU6?}Zv+%NG;8ArcE^ zyuZ%Fcn&PyHQj=eOc<+7%y7_*g4=j_gm(LSD` zS;2w_R#Cz+GnLWGyg|sd)#BBAN}#r>x=OXs9P6*TF_9|?RQXO04&E-LTfa_%8KZf( z($^ha=L~rit`;W^7e3IvF8u5t-{gqx6O%eVUVxf0xJS9;_UT8-8l`oRE6MM~dptok z^aALbwF>553PR=Fbc|e1=(@x#Tp9O`&UGq=;ME(sl%iIUihoBwr3$!LH%+<9&tBYF z1%z+eEj+vB05sjyg{|L0fMyormsC-9q1Ozv$N9sk&o@ZLh*r8gd>uS$$%oDVEV;5> zqiI6WUhtbMh7(UAa}_g_;oWf4ESnWBh}HSRE~5U%Yq z=4uuO!;YYIV)$kp%~u>Pe7V_@E#ty4bG8~x3{8a3NBQs2_;v8!Fo(Q-zXfCm`0nY+ zVS%gX1L61T9pq(T7LwJ$*t#&9s3M=?nS2>MpHLXf&sCP%d*FDn0@tY2Bs5$%65XDZ z(Z{;U3yB> z;EHHa$nU26-n=HK1I*#|$XV!RJC5&9sSBFj_u}=xBiUcCLHxQ-iM>?~7an&>7XG)Q z34QrY^ZxWwXtvORkILQX{)uPV-LT=kO;YUFo-5Ea?>jZyXM#(TohTJ#kb$4hu){}` zMZ7aZ?-z5Z!jI`7x1^tJ+1f_7t9r2U`Krw7us9oaEDHZx+`+I>Q<(Oiw-Bf|i=&sM zIK$YJB

wKy#Tbr_(hVUwfrMmUl0m{!SFUa}}Ar>0Eqf{TxP)N{6hX@mTHfkVw9j z;DWC&G)jG*&zKTK88fg1r2 zVf6HPQm3^G2ZPUpbXpkhQm~@|<;yYb7tiv9DEOc(&duZZ&k2W~K>Dm|vTa!toSYWT zrT4z(Zh7rR|4~&q;`ASSS73t?&3PE1RgW^-1TX^S$Z6b=^?dX3Sk{*I^ge z>>__Am~wYgfJ-|40J6r%<58;x7&2lV;WSyI>rbwyp=;`R_nk-*YVW z-^rvycHqMOZX~MPRrtbb1H4|wnk^PDk)Xst1T^p%~&MC}^NqB3V>It;~)%V6$ zwK1~MfmN$dK&`18tuhVfj5e>~Z0`2Mgv4^AU76dB>wu9*m-W$9l6(;U`K;DohYB=?&K<}q5cT)5bh&P8} ziH;$9e%}Bi+a`myp0BW7_9^jgt%670l>`laX)wRHa4M|E!9hD_(i(!ZmbamNizMre zF~B4D5>Q|DBpyC}5*vRG;eqEZ#MwL@d}AZYzNg+;e9eJ+M+VR<`-HqF_6AKlZI4Hv z93)!HFOd>E-huZ@#U^LM->N6~7lE|L4~WY#K=&2TgsTrM=YF|<0NJXZD(>}qlzCEw z*ZD4p!v0P4XqF@yQCUM2e4f%r#iB6wi2|-KvLi44{i2zE>9j7&60E)hY!|Ktli4ct zQ++FS*g6EcdA)GfO_FsuHzUbTLcbAbu+98F#!t3q_tV=!&c6^UHM;5C*$tI*E{qX! z^1PQMHxo7Lg5cq;Y7#Q6&kV&HNn?8ij0ue6HY}2aC+5xqOY;?I>1a&WYmVh26#H@M zm@Tw#3IBG@QpO<%RoI?ti1|U`G<}H^+#6FPd>59Ao7BFb|70U(uJa%Fsdx#}i*Lx2 z4h|ppS7CTpB8~f3fy`nps#?auYSB>Wx)cKXQNL|2>&<~y{VEbTc0H2`Jb{m-LxrRA zC*U*ZlaNz*7Cz2sLGRsmcz45Hc$Co#UX#v4(yck%`dw|r@wYrzH?@F>>D7Zmg%^rU zkOG!}On7MXJ9>VL7#qFUmi>CD%H6inC5vwlz}G%$(41llzT3tKZq8HVxbJqv^J)Ty zIccNZ;TPEb=nXX8IR$3@Z)m*66WavY_1r0g`P_#Go1rbK2q*1ahy&%a*lLjsqW5jc zq_^S>-gbfI!2--zQGm*#O@ft!gScaQ6g4W0$M(&0(DJzn^V>5l*oKlYu_p@mXkLQL zD!gmcB^{ENZ^q)Zk)(!q4J?*y<+I@q+~1FvIStcJvL&dNKHRUuO?$8pmz_r%05`%l`YiDI)YkLY+1{sV660x z0(Fy$O3$NHP~)@@hKMy0sRcaWe$5-`7cpXEYd6wOKbFDH(C5VEo(5dlJw>QAMH<|= zTD-SMj_tHCV}B&w*v@6U@ycsO*b_GnTi*1OSQ!^gye-czc<*MlYUb?K^iizP1E8+` zE)+<=7e16%qEmC9kckQHI6Qw6C$n`TMov*h)gyz%Xn_J($>(83Mg2j)bPJt&I~+-- zfUV~D2n+T)5dJNW1Nl|26#B5=KD0oggX$=vjl(A_2iUCVdD9YZT{`8@rbB@k=%YR#sD$&00L+~Qrh?9FQ&Lrgd%n}*PnLdlcRbk7ZCRi4BU&sUL zZSMTevsZ9fL77EOdjV~4<4NtbhxEZ3o@ZiigWak;gRy!Q`DiOflQ-rU?py8ZE}x!y>q zqLBl|3RfUxTO?MiErBe4FP0eVcXMWm@+pEeo4Tpp{l#}#G#0kHfsj^EZ9BxXX%qkB~+AcxC*P}S;dKq3?Ad3Cd-?s8-;2rU@Olh4GpASJbG08=Ybs(+U@6$uu3(>P?Ahb%D(p1cL)*k{ zXw#b@`0~+(D~}iB#9FUFb%q|!pW%dkvV(AU(E+?7C&o6uP~aplmSIqOC3^inLwseA z;!+drRu7OBD) zC*6UMGaXolj1_qu+zE8n5tM>9ka1W@S1C{8JcY8{1O+CX5FE>UTe9JW$zAAfOvdz! zi$QUjH0&;#%p?ZpkiF3wLa9@#n73@b?fM%v#Bb>-(pOzAoON(4Y*oDje)12Xzw15O zz2Y%}+-6wC&B0(kt9-gF8$V>&F;6=UFpzOzXX0bno$ew~Ud-n>6682tF-usyvshs2 z?Zk;l#Dn>Eb@cIV#zrYw7E0q;U;j;(w0RhWE)K1L_!>!e6 zxJV~)wc~`Bp2hUqlN*C4^-50beOvYJT!}0ecQ#@tv4F_(Bp>XIl-C37`*J8ZU zC@U74&JBW%^bANYn+44$%XmkFK254$&8+jEk;_pGPp&>6*y@%=O$P5{olb-xb>4jbj%9&Q z?JlZ12cYL&1A6w(grRm7+!}D7E_Ba@ReYDq z({%yM+*yHZKSd#Vnn%RuNpYi;6JfcQ8(J?26CB#k|1NF?2syl*J^m*Or7c1X`XFy( zd%BoBp7DhYk}aq-^Bri`MM3B*6RNNI9n<#v3eDFAaw7*CdH&yF`0k@Z2e(9l(WV*P zS8WsSSh5kfvgfZLkMi!DxuRV4r}5Ac!g0UJDWcvVh`Q7AFlV(2XVJTah&KkXN1d~| zlvjmh+Sql%yS!7QKFEPCpEwr71kdurrVnL1yYRHE2g*fskmj?J+{2`& zFn9M&ay94}P6T7K=;rGx4aweSyr+JmHh` zgW#z!35wt4;r5VNBCq7g1bp8IgF}SeyH%-zHLy@X$ zIQQNjvdE|!Lf-lB8GPV&hzk2MSXkX3E9+gw0*+9ezHSEYGV3P+ zH|7XO^r*4OY$a)22SQ~Df7=`;n3#k9i1L(5q1;!uIWfSIn#Qpa2P?ULunlCAWW%r&@MLl~q zG$IbeGCrgA-x{HOn=5AD7!9ol{cw}WKJvEKj{R#sz@EQ0XUX5DfiLe+Tya~DJHIOt zZxxjZUtK5?T7(zVOMBd4$M$ZT$vY!YwXLbu86ur*i0_##Lo^Th?R-Gm*KL+*Nfw<~GB-CqLzz&lf7&d;5)6b64Jj`K<+{e~NG**{#Et$)Y%LXIi4|EVQ||7?VABTvx{PXeH3Xc?$_@coyt4ftg(70i{2 zg~x;PT=p{xZzGgA^>HFtG06-Eug&I8ypV!M10DDyQwT2-p25?dQFtP1Elc+KT2(di z8fH)3NjIGCgqM0!;Ps66Qi(kRiOQ$wI%y@FC^K)HeQ|e-7?U93Z1l zT?DUy2-t5FfI1y*q`%k>eC4aqGd}}n+wJ2Xoy&yp9=jmeXcLap^rfnYElAn)Yl3dI zY}lF_0i%iKC|z)tt{fo`AA3i`W9LfL_K?Ebf6cfmRFr&7HD~F2N8|5vJ8{If z+3-&9AQ%SnDJd;TAK>*f25G`3-UD3DGNVd=4XfpGAw=d zR2EQM1n;^&L(PecLd)5f>_oObn{++^2EGee?c-ONFr@>dH|Jtz{%10Hh{3b1Pidip z6-{vw<9ExGxEtIAPUfa1M20=IZA*Ux^>s_A?XP1fI<%aP=(%|yNE}0 zS7EN~U;KV}2M^z{cu!M*5yCg>~6BTqbanGDIt7lyOhH)#!a?ZRCgzFJ8)>!d>b zvPZDOR*rjSG7R14rgGnslhJ*6CYaXDgOti-co(!z2=Q_3_QeCF{P#tS$oK-YQZrDF zpHCxC6jQzUjhy({D50jt3sPmY1J@g_fbzXk&|-FrjI-ND#h%6pTjzcy zQ?kbzIT22r-*E*lbOJZA!*o}V0eR>$f_DG>PQM*HfJtu)=+>mI)KE@~ot(D;?7ZEG zwwweOES$mS^_;-8+l8QeNFCyHoIrD?k6`r{8BqT;Oa*}(1S?j3!8qk?+LU~f&wTm8 zBdG#m+4@w$HlHK#{@O&?{7i+@l==#{Cp3{KF7L46<#MPVOra^s>RhM(2=1X#l%Q6s z9t=gv4)%+V6{rL*1UXmo7sm25cAf3i;(kADctnu7Rf4KWy8Rr^J zz*Vbkm|XKT$h6{j^Y#j`?MV?0I$H|T*Ye5F(~>APSjg`Nv&ip+7GjxYPqG^Rf5|b;ib$2J0UbDclcZFEXzxn#~77__Nee8Xl3m5B;(Y$MB=qfjg z+}m{?BKUoW_Q^pa?<^r)CLJ!^>(ow`EbAv#o>PSDPl_|qtU#7AFq&OI&vWH?$9Owf zv8n`5?hWs3x!;k6e>d?AqJTCOlQ{>U(zS^gi^JUIsbp7rIp6!{IYgFO#6P8#HfODb zL=9uyF*}Fcm^+EA>^==Y+v@R-M>+m=mtf=0{(#)7j}R^StGvE91~ta(;h8RfSUz$b zm(nE1ie>b$!|@b0Tjf^W{CJSQ>`13duD76ln?9SK5sSCokKpI5}kv8va&}9WkAQBSOAWquXYfp;5u#Mw0N!vX9uZ^A(1{Qq=n0N!1SWchWVE zIA`S=^l*KKiNPUIYVQhb+8iML!)Mr`yp+zc6la^;3|Pt9D7@JQm;~lxlo9?iXy}Oz8stj1R+Y&`z9Y%+DqcBP|PH^WO&!=4R z3vAcOZ~;Xlpts+ci?)3ZCDqeVdz&2RJn}!Vem(-tjGV|C**+*cAi?F@7Q*zd8rU96 zz_U$~@0YeiuAe-YC^(8`m#Z-MZ5V2;Zoo2~aC{Z`4Cg;H}hsTBoqtj;uEy355%y!P;k#WM}BH z3<-W-oux<|rk9YnE+2&xZ#C1$sr-Gav7VSZ7YY8fR|y+#2>EdSH}YkY7ERjL-2CuTtQ zwk+!WsfK!dV8r+01cBj@8BOvi#P?_FS=(bh<{u`7KZUo2pM{O&cZ4#l+iAt^emIV! zKR3{j#m~tfZ+#HCIEmA`*$gl1=8<-%K@z7D0{tgtaq-(ys(4QyU;J>!Cm)Rk`rmbM z^7CS>=$0mu5RPtM(ZZBaQwWMJ5-9ndhZM81+^+y9R+2M~MICELyJP1ChocVQ9G=DW zv$s}Yb8Q6CeIgB4ej31gS%1v=+yM47%-CuDlUTkY6Fbg6q=O;VSaNAHwhmd~)g`}d z{sh?L0#7w`@|necoUo(Y_J70^HxFTZgcF|G&);Tly&@{(uVUz_XlmTI9cT3@3uoxr zqi^(L;ZC1Uy1Dl)G*yQX@uIJ^ZL}^V)lP)yGDDQE&!_X>XJWRq7agTOELiCO8Z-75 z;CIJYCYbl zYlZ7En;_$MHqeJtLAG8E;*M%Ct>_l~%fHe3{<*RPa?@#u(-v$y?}~?H*5KDAd4jc) zdN`AJvO9@9g@uPTxJRq+(PH&nh-~CDYR!_|^wJ)vI@d=;OpGu|r3~dN#xr-ncvf;n zf#qv-VoGX1joCaO`{iojNtXdPIO{f~#n{kiGX@}jC>A2$$#H!%rHIMb6uv(#CYY?4 z07Ni@D&*@z?|Th4%gGcwpSz&sPa!#6=}iW+j|m#&e&YF0{rrY!9qyi@CA_3oi#{G7 zQKse)vlP_h;3F5De*6qEI2cV`SH=s{)-PjLeXn7EuO^>+>JWB*FeSrYIdp%g9t#X} zWH0T1(V3N{wn=N1ptD7jdoV%g>eHXV@#=5hUvCGG{b$mWS8?#qb19rE$cD2s zpH=m@dUEQnBbb%bVvGnV2br+T;2q}%7yj5(Uim4)r8QiDaybq1c@w`g*jS8A)))ui zDe8wzVU3n1sCUp6I~q@d%tuM$SXYnB`UnlHQ6ML@j)9x{9MJT-Nl$KG4k7I$;l=($ zu$Z-l=VVMJJBQ_HMNKq3jow7v#6<-E3iWV+yG92kgrNDTjV66s%(I2l>AlVi0;}aI zR5sY3G=G(ajhQa6?eto@;Qc%DT;gOVo_;+E`|=QMI|n-Hc?^`+f>e&^67TX5Dp z0o4mAdE;C~@7tu41Ls!24?iP_O-+ZJ#Gc=mHyam37*NOL#4gBnvhU)dds!AFaN!kZ_*7;*GO|#9QqM91);aq-i zcQqSMT#yFxwT4dkx|!xS{^Ff9D^PyUB+lrHBHOo}pBXIVS;3|SI3n;I+8VBguDd@V zJ2;zsyR{TNo;@euC0xmd=rroO=^FkhvSedIm2qBPkWgPl9I81>Ceql5^<_E2omXN} z(>|T{*+;-~-SfER_azKptuM%5K3y1@_7U@igtcVxa}m!scw{?(XI8zdAH7Chuvl(@NS(@rafh-X$Y2yS zJM*l$|DpvG_luMKac06%Blg0DXYt@@@eI`bW!di=Ga{x3-tAuXeF|e}P7{0x) zvuQ*%eE#P=DB@vM_-e^n&isjyi^gF3Hg~}onVAC1nov5Rr^NQ2mFE`g)WLXveb9AD zMAgX$QSPHOSF+rWy#C!v>{YD6qO((Q@5p4FVOoP4{-OAI+hw}3s*reykAm?{W#r+@ ze6T9-hCO`8EpNwLk{srWcB<2%NoxTd-dN7&v{bQec~w+l!+mUu5@rA0k!6x858ztU zBv@kFBD}x-4%B}75Av3!uoE@bIGyKRtyay2ypS`zpJNA%sVgCPQU|8}oe$OPRA8@6 zI&8PK#hd$wi1;?%H=w8vK4F<8`%5x)n9uu{V_Wc&VLX(U-hrXSOm!BvoSP{S9kRkhic=vx)sb}zbn#BpUqOij zz@-vXspapb`1!XI&0JXmVlQvN*NWq|6X{Lzt67VSQ#u8Y_1y(4dRKt^g^5B- z6xSR&s0upa1B| zFvc0)eYAeam}&YvCl_v82*#Q#u%8d7vhhO^w0Cv~&i<%^r;q&tebxK)SCs)Sn=A=o znJ>t(NtbAaGg7_$4P@YvA{a@C(#`wwFnsnz>*9MpGMpsGZoic-yF8&3xT-4xG zcr|2={tgpW`5Wv+ErjJAB+Pm$?v!6dKKZ9&#%6goXKn`YT6^l~M|(1#VMU0?Nh&LqLa@ zAR*8JSGZfE?&Z%_<1Xm%{D&GWco)HYYfj>|e`*YU2GRTIFY-1oj)YzcCZ~2!gi9tf z=<|l9uw|+yj5w>yefKtk2}xz(q!J7RN5(+#=|3<g4Bf=u4Bw| z_;qA2f0KxWNmm5iyZB)kUp@;p-*@0dE|7hza%8Tnf8eQdPrg?x1YNt=a6iBo#->}L zy}@Hx7hu7a7u%4TU4@V;@)r)ySLbdeB*G@!m+()=n$^pHgooX-oUrIP`Ff7`L#88E z<{Usr>1KLmcOogKqp-PkJU+f@$_f(2x%FQ)aLmdM>USm^8jN^Hl*%dSl8WV0a$7-~ zbHyUDgS1j&33t6#n!67+DQZFWoHq3sD zi`(kywH8H8eq0Xo!zZF_jya@#66a>mjmD+R-#!P}S<$fD@5%2fPf&Th1;cJP;~4RiXrWd{Y@76%l$!u&z6+$c zS4gAvq!aL`^98Zh&|qrQhj8iaLCk$~3I7;Q0K?hlWWN14PSJk@G|u}9>aF)-b=qdg zb)U-~{r-l}3NBN1tyt^{$|dPZJMdF^GU3K|qnuj|wb(j`RQ)!Ge7kBqz4#bK6cz5)Vc?^qLialwsNcrFr5QDgou!N3d#5Al%CC z2IuMrLeY)osM1hHe;dc6YU2r^H{T(Yl(2#mqdyAF6i0L7KSkKMWs8NgeNK=g>KkDC zlxyg6yMzvBF2>P&tKs&UCVJK^3OD|o2X&!ZTxPLAxNuNJu-qgHO#>Qh2ZzO>-7gUBol=gnGoU;SlCj3gOwPGNDm{ z7vzuEg=b+|xJ2tPw22$S3`-UA@q0RCYhDMXwb`U;TN8QX(#?`2>o^x05eI+Nw?Yb}IP6ak%2R3nbRPB)ffL1yeLN zx$pJy(6PFJNLExpL-Ap{&kF(GJ_HAuDKwxujSf6pVbd__9?_w}7_6zvJ>3-xD{q^_ zQ+IW4dio{!I#q#db{Ao{8^sx`iN~g6ap;#(gwjc~nC_fFtcZ8RIg)qj*H4RZto0Lt zqn1AI+Svlr_T<8C`&{VE`3<)|a1bxK4KkF+!ffdP*tXpt2KQ}3iQ*41hbtB=mm0%j z(!P+N?+j?{byqOimc^btDx(K1`iP4CM0C_0kFNg&SoWZfuKF~Q=G^*9`@*l`$QEOi zyyqggx37}-jfXOsW;aMs{|DXy)?D8&8#uGTfm@`M$US??Gl80;+0REK*#>7DX82!_ zaN!$AxPG`x=iR7S;X|k2+*~% zsV8rKC^0z4ig4cENOm#izHVlH+`~Y+sdVnW{)i^&@2M<&$v5wSHtkribs#fF> z$tr*FEdG!D!vAr0-hVa!|Nn1Ew6!D+A{D8GQs;4hR8*1~k@d2&M@U9Q3X!C=OGDZe zWu!Wf`{QX2z{RKpTEC-&*jn|&^hP%IFI|dU9VEOe{~g@kFF*jnhVgk zqyTsNOu-3yOG(F^#h6>D#aHgH!k=CHAXWXgbiWXW#jAefWD7sO%G8-F`Cmn^rPi2G ztmU}4c?KK%sS@OM6cEx+7$Z3u78nu@Ms(R zRk|OZX;m>@o&C5~QVvF9jUhKC2KJdH!P+)s+}0im{kzuT!}Dgmf6;mNC-<=QusosKB?Bcrp$W-<5 zKjDtZidNw!;=FQ`LZ!6N`5SDk(MGJcRFk$9>hJ|(CbU)VIlC`oiw>?E@MU)a(cFAb zGUU$)vT9bh=*fMC^QRxj0q5L#`2|(ng~>QStQvE5|4P65=wr*-e9|lXGrOX&n|KUg zgAdOAU_q;mCF7r6BY9k&h6oFx$uk-DoxM!T7iGb{^Wv`a6gAN)drkB^C8GW*V{CZ8 z1Tr4Z;UP0E=>Lz?Q~B+7|}KgHoU`_ocz_*07d_O^}fHfCO(J&YoP?9V)R{(WwX7DKA-8>sqMYDzZLg zUohkPCNxm&ga3FJc{Ou9REc+T*&NOmUiE{iZ+}Cx&Kro|{y~ybln(`o)i8A3cgVP1 zN|Nq-i!;B6Fh11=UtUQ8_h>1*JscWhhq0bbI&Avm5E7{d z@OQ#9Cim$Gd%C`!jl24ksWb$$t5@H!g^eTNS7J0yjO3_$WEUSaH5!^kSHje1yFof! z+_A}DgB!++bJcxXxH0at!su1uMS+u7pA_NP*BNW>>=eoJ&* zh9fr2#F|wy_^#w3)-B$LA?x;vmNP}3arhB z;HQgovB+d7OwdhdM}|Kqu6J}u+PMfQXz?TN0dpy>xk&x~jFK+Rm!V#w)6heq0q#YH zqgQI6v?n}`COjiF;n5J7|MMAy8!V)jZrQMRKn0}zol0{$!eH&OCU9_Zqdzy+!93;X znAUB|l`qTD@c{ww^rxrf%by`|&QY81+4l)vJh?@bjNI^=-ejB^bY7aa;S`J+qf6f1 zx>4SBJrugU717&%6+RNV%7^^jNx|EH(htI6$A|kg#Myp2_V_JF3ukj){3IW1%jfaD z)B}7%v1iu8z5HhICUx$xn+i ziT9!^qT6v7k1ulI(g6&U+~l~#t_WYs-NIR;dckC;&2)dJ==R<70ssw9s5}MpUA>K&V0XEK$;7If$r9%#-s?=i8^me!OsAReL{6+KeM=kMzZwZ=sF>D+% zj{Nnt!0!c>=s0g08{VLgL0Pk5QGQKDU1ukIM2s+A&zF?sYNPK_74Eh)h+RFpNAk#_ z5gp%H@DI8-6-jg7w0y#B#VghV9&p3YIpY z@#Ub?n&yGb;mB*XWUWw6fO;h62sY#Ln z4eaLw_fGbslZG47CmS=cW|9GaFewPV!y}xMtS)2d;gRTTc0qKY#Nj?48+P!k7g2V( zApJJ-JIlXfh=XUw0PE6aGk%!x2PR>-y?QA&UDIXGZw8~$Z3UQQa++oPULY$yB0+of zXOR){pF~UcHg^0vQMqk-1H5t?Mq}PlaN9E;AovHfmaQV$o{GGrZzFo|)P$p^T4d+3 zI*^tgCBerPXl~$Dm_24cQCpadyA@;{=j?rfkv6`t^RXMqo?A?J9-2yDi2NoKmkHEC z?h>6~nL&M(-oool4H};Fmi+wR49v!kj^DhW9{HyMpMUYnm~WOar9P6Ry&VXh%IRc= zT^SZ-{Lh2*8Z7_wl762f*5U&mK~`%NR=#b=5l6e(s|P-q7xz|j%BCA5m99Xoo0aHe zJxjQ?F8Bov0J#b)ke}NRD&#v!uTj5P!{iL89d-lrysIm|WOFdTSwIH1`$2Ep|6rhU z5=MnEwDg}vUp5?t?n$!r(eEMjsL^t`BRU$5SRmJ1d5nMZGT=sIpW?L3rhHS_Qr^G) z9=ug4gE0X;n44b!%~mf_gFj)C@?RvRcQ-!1qX(t$3(?};KKQYs1r;4)$z2mWh$*NA z?V46N^yxY5n6rzjKkH2c{#3*IE{c^lY8g9JD@Knpe{5`5d_0Z-J ziuqWULOaQ$)|D((=`fBy_zKm-%$W6vJnQNOfkkg{ zUsk@DKQiHiUf#sUi8I-WuMrp+ABsamo7q(w2p2Y(Q^9=-(W(7GW`@icgN2Xb#IIjq z7lmN`!2qsa7zxrJS~M!npR5zP#0BgWJbQhV-Tol!_pz~Z%LmG@-)7}8kEg0 z;RpGB;MS}~pWn@3i`VPplmAxXs@q|hXkbnHI4Y9JxNhihuY+qVrbFr}Q@Y4_F%7&h zfVTAv#_G;DWLl2}=&HU)`NAf)ZpH-|Bxl5rxSO)iMw!g0p+A?MJAxNJTY-H_BiO3W zofy+Mk7)>bENA)~l6Wf-ubD;g^qX$Hd;AD)JFpo?i8Z^AatYSlIgGLjdvK!WDBS;| zN}BWj3fuA_k;xo;j8!A7dGPF)#P?3F(>e#S&pK)*KhVAk6*7y+h2K+|#V`{z_AO@z z-Mw)6>TfKqE)I|0&u3>XRnVaA6Wa}%IOBH|Mtv~g*{=t1*`IG%>%5tMWfr0a%^$NIowTvkViPhkpt*O3nxcy28gGXCzc#Gcr^R`TsG5wAGMaPQajW85^>RH%=8}#q{ z0%`&caP*!G{qkuAJ*pE<7hla5_w$W-(J&)C8!#3Q zI>}3M&*Z%2Xc$u;3!d{2z$%+N(t?mJQvH||GNAAjZ1h_R4j2qd@6?HLNDmqJuncuW zH(>lnQ*yVr4)lu1M7c~KbRlwRpQ%rVG_aG?G+(Y>j} zm5)Y9(i@{lg}EKMY1M=~(__(HKOV;@XR>yWIP|`u&P}t;xtiBP^7P&a$VrLC^M=}( zvBd{=P6muyBE^k|(_wbnE^L-*LhB_G`efHHFe`VFzTIgKT@BZn_WJoKHOyk}$ug)o z>nzaddeIrwhH^dP`7}r!zhB5g-*1Vi^jqwMWG#kbo6)HY;L+oO) zb`#rMPk-EHp;HaO^7%ISd(@BplQ~B2T@tyNLl5B8U@g?EOu#L(n%VI2e;|!%OZF7n zgPcXaQ@)7J0=+qDU>#xYZRdIQJWRWX%K4x?!*3`A|O(_&8O6g)T`3O=vPnM|xK zKd0|Qdd2O4$mW;CdBazDJAEn?oZ3kjq`sDz+&TqvT5qAWQW=c3?}1x(2)C#`fGtfr z@S?2|ugza1b_mMxy51&eA2tn4ZW+<174Kn?S0rQ{tc0chbD{5=k0kr}A+Ud34C+%d z;E4H7a0xVK^*gQj(>~7J)_gEmuXE(*3-q~tPBMR(D*6>y@4_b$0r1P{1dBAP#uLrz zFh+kJ%vZ2O7F|HHcgzL#0ll%t;Ww)u8Y{v!BrIU?HmGP+;f2dSqC&?Wh_RE0t@C4{ zr)`V$eEdk_5}ZeF&y8k=mW9k|g&T`(EoX;TWwO3`vG}gnJL&h28rc5!Ka$mLz^Z1e z5|>k{sBtSCs=j36g^VHO=)51W_~A?_@2`gk4#?2U-Nj-a@jUqvoKCLpYKH3vbD?g# z8Qwj69MmGk-dg=F2(BJYuf}ZyH?9jsi(~O)a}t}WQAh4%q(Doz> zXy^5ZQwvGu($JsqclBVrv$YXpGeke`*X^*@rH1|1$-;MoU%`g+}Vdmab#Qxgs zN1NP(qQftaOv}7a${GTsH#_#?=g7D0$RN?dX#If9|J+7OPa2S%Q*+?Dr4DXRChV=*5k zXADxH{J#`vH?1SB%k}W3Odr8u}7ez#g8zy3Qh`f*EQ6($Qqoa7|OY87=RoF1mXz<}m}4Nc)(pp!MRUWONHnUXjfuVC#VftPLj#2N(5mRwi+_OnK zdYwNQuDJ#+!$T@3Oi~3$(RF0n`3cSv4 zfXv9pxM$Q9vPNW^syj6k^58w{PCHFjPEJGLsl#x(-z5AJwGQ?qxRc%Yv%$M78EDoi zNF7*41~qP$?==qZ@j%5?FT9YT z#p@OH@muI($t91W?76uDS8VtIWnLU^uGA(uIR!BL>^2ON83uPem(fqT_aNK`FiRqV z_B&-HZCN7pt{mz#_h?`8#?q19O6n$M?UTu`;q54=Tf@i4x1$4}k6p1FkuThiiVY`F zzcYd7j$V)BY!g`jc|-8*TxGs)=RT2-AB>MqJK(cVWsr6Gm89it1{}4UNz`OqdDsa( z@-NpPH_85Hsm(fKM$i(Mj#i`_gIO9#@rj(iX-fYIw~5a~7j~^^0Y-j_!-g%@ zSk)3nX7=08=GvLjcZr2izu|#n)v_mI9H5*i+7x2(;5r;wKNF9ryg_pB4MvTa&(}69 zaJPjwa9YnKI&7IOcS>)NwC*opDxDE%-VqFaHOt6LuPv;@xR+#buoB(x`WhBxD&sA- zo$S-`1}~f0_$kg7>fg$7OJ7$AD2W9{ZUTn;?vTi?JeGQ30;b&24ZRhY(*HtQ$+4_CsPQfwYric(l_+_%O{_*E4=cJYJPIr= zdP7Hs16f%+O`0I~ryl#?fs9%~67_+TSx5ha2PX~S@{|=2X|2p1;;XP||5lvtyimMz zi~C}A2e{h3Dn>;X*vSbo*!|ypqT2T>8K{#D(SAozaNLVO`!qR)teB2%ugl?hs+pLB z>V?~P*kVin8#w00WDN0hlfLM1r=l+hS8jSrhI*W3h33zsep{F0x67|d;G0GmzF7?x zk9o)v?iS!>-Tl&rkW=7!@E8fN8VPm7avVz@-No^K#;|bIbe6Ebla&7GmdtuKhaAaI zVB|e$#=%~GqK0k9%J>7#1ZdWCl z>1Ra$av3}6QbKxr+Vi_D8(>}UBGNti8=LD=3pYBnX<*4Tu3Id}#}&Jw-HCJT%d+>Z z%YGyM{?r67Y`F*<>lNsv$C;9~wYs!5)B%%i#)|XKR&rD!0Z%wTBc2~@fSpv9)a34k z>UCLYdo_u?Q1^x9G#cHz4)BTHzeshuxK|P6DDJTW{tNP9Uw-TJGGfCQI$lE4FePb3 zh9%!-{}JaIJ8`#1O%kbYI8nWKP|~(z79_?rOYS#@U{ZZBH2JF2g~uboz;Y#ie*D-m z^UpKP=>3ySJ~IsruZ5;f-k_KeN)%FGkT2dx(NN8Un|E1p zAwbMSzqo>tCxg(2KgHdPU$Bal0sM&dLO5o;666D=s8M>8{E6$s1NOv{~XjSO&*UI^gAy0aSY90(j_Ap88-DZrc2TOm(`&Qi}Jo zZ3)NmWnwyvNWTi(?WSPky@eR0&S9Be93-UZ@V09uSoP@{+xnv~W@lKJ(Vn9S_*0&-z9=t9JckzV(BIgBA3X3Tf6UvFS4~{?a&J7S)32r ze~O_B{(|9j9X{k$Ci|S1$#T3Fv$!q=>U>WE&u@Q(H;Xs11rJ1yOnx?#Ntp;QOSef( zCZA!EVF2yln&C)dD9n44$TB{Mf%?Wg=q-`w@7mtugv|-~EBhl}GmXc|g`zjwUWZ$U z9VHsyD-oTI$c_b5@an4Sc}OjzDnrFOIMGfOG;m`$D<1+)qlT9k1f~^ z3Z@D)y7)-ty=E(v>36YGR?|un`RBCM5=YXiGA;Hu3mM!~fQvu48YQf$87_8ZaxBCq zFJ-4of06uCc~BpbLhfrAqIY|UWN6(axN^>w4%F6!*{@6Cw8L7+j}n>3FT(hUpi^LK zn?PPeIC+1>0fx3sm&k|8(#&D~x%%u0B3HE>4?Zm+=MGq*)?q2y1)4*d$8xZ{rywOJ zv&eR(Pm)uGW-wFf3e!2V0rYFf!)fUP=vs4tU2_WLO^-&RE7Glre9^Sy!uiCR^Q5&58F4X|7 z<5INUn}qXg@9;NCsZb@OPLDRKBJ<89?Vfj0-8xqKBJw`z?py~?R+)lts1Ep_xXM?E z`)OGlZ&Tx2=c!S^7-~1mh<|Izqb}1gQFh#m8kK9&e|2??S_`;-;YWz7SdaZ4zpkt? zx*@u3wP4MyyJXH|Malj5);#J#43T6tv3dJ6;H3Rv{2gG7(WlPh;havVo!i5~|KTCL z=<}E9A2LAQ>25Sy*MU0s^dji2#CG~Vk#5~^TH-nQoYR#7Cy36(QtZ9qFnO$U3L240 zV#bU9ka=71{PY&g8Qm#8{bD%uyz>`x^CP9Elh@!J$FFRkatLf|bRj3@oMHS2WtyDZ z4m2(aG9iRUrb>KA344iarAy^y7fSGC*iyb989~JwcNDSLzERa{pt8-#sPO zAp)r6tfT>hPD!rD7D>{s{RjE-2jSP!^W0ovI9}Ma8Haxx2veQyu-^~~Yxq@ziH(#u zRQKhtv{s8Qj7a|B-cUZcyB*oj7}jBtOkS*21Hn-rPGvYi)07$TT&jSwlU<PeRR zQGpojf5d(b9LT#aDnjCn2V}G6E9M*0Om5H1g*C*Po=@&a+mEb<_e0N<6&gDkxT`{A zwa7@)NMc(yU4|Qpn(W~9lTiJl{oh8Hnx6)4$ZcUqsr&PS^ddb7+s=-2MZ!_Z)^Zs=S|`Z0#9P@mSWPNGaf4U zs>7sV14w{%7F%f$!w$_(XR4jym|Q-Nofq%gUOJbVwu29io~DP#`?=!gEzhMfnIZh3 zS2XtSV}?eC&mnPO1F5>MK;KPF2HlTW$?%pI*kKqAbu&95G{F!jcHM>M?t%PJQ7_1K;}#l0!$BG5*Xe`h zzL`)tgTPzmDBMDn2)dV$S^df(*>^aai)Zw@bZge~!5f{U{z)e+TmT_2ZDEi_J+|zc z#54f3j>{ogb^u<F3M?!uDAc-w!0JGrg#ZQPxXhAUC#WST@@eC^*MA*l>`(lmHPH>XPrxY z5WkOxxXJ}sabgVAhxfsUU1`km%L>@HU@6$Uo6?6qsx((s4F;Zw0p~YqWa!ZEENpWh z>TpeD=-oEJm-0mFwe=jy`!5Rq8{{J?o>C2e=N7|Ir3|=mZY&s2JOLetl8D8OBW!_* zBP#2zC*g8$*rw;BV5i(|5PlNe^KUJFufU4m-pOp`-EBBGRvQNnOroYEUV(;SL?<0` zgs99Z^zqG1`ec|E_w9F|7#rH5^`Rm7xG0Lf5?LL>spGgKXf?!3pOUdl9>ZJ_&U-}X z4;iq>Q(`0T96$P83SUi1h)=>0Z0(f@6}yzUc}y0b3HIcnm$qQ!o5T2b-!VLu(wD}K z&H?NG_gVkZq1X|)7orwfz~37$S)IQewTQb#^jkey;Us;j;TJm?FZOf#J=ez8&lb|H z+VfB;OqL&<6TvnqyMfdG$z&8B!)@0uw%O-vppw-&QcX$E{i%LKk;b&Hrc zld{7`&&mBuflem9HPExM9rHKrViO)*L+uYiI52!0&U~JYIyaQ~%lT?##=uahHtfaP z4q38S7sVOwvu38ZXEehK75YbI6hy18#*_CC!-`*n>7y4@V8+&y&~!qZZUrwey;TO! z)-Rx<+Yd)LI*Iw7ZP+EZ3sWjqkh`)XbI5uTOS_XMy)|PP>|ldo>g9cySg;W-gLPrU z{iE=6x2niWaEGsoMPT{J2)4^xU{FggI24=0Z69^mU6EZmbImZ3*TZ3Z;4An?YCxwg z9kv@@hpJQSVBnz?*mEYEB^(yWLku*2LpTy#Q72X z@n(%bQ@W@CXOCBr<&)ntVaQwb?*D*|h)sa9xL&uBevapV1&+Ty!!bx+43iooYsyO=TiN+i%S`Nv-aS>#eQ4Ja0?84 zq)T1R&FO$DUm9pO4%T08gVK;obj{m|v{he&-y8Y_Y7P%!{^jN{P*`|y$d+gX*~I5Nv)I4oL>2rbuFGo z=3GsOy4EOO=@822FMEQTkxD$7*J6CzVZQrp3A`EHfD?M@g8Q*>R1KBqQk4s&InD}p z4KBfs(`J&qDfw9XI8rJ-7s{5KY2mj8d0_SW3eldJ}RcC9(Lpo6;~14TzIZhUfWZaARjbR2i1UoHv|e2c^}O{p#h2 zR#PUMv?v_D{%456elv0BJOz@UWlGhG-+^PT5+C%p7g}YWK`-%H3(p>kk_vZzA$Jv& zx(%hT@{-x5!&PK9cLAkEU94);B&K=G4HVN8Fk9st?mei-W#z+h%FR)9`%0-Kk>#?t z>kVk;_{G$DKpjl)=^*OH=B#mjGHbKCgfYR5;JWb~sPwOa3T1m*54B9O+ZOt0XiIz_ zZ-zUCbugi?5ok{CkCzJFg$W>fn>eMtoTtVe!cNhy36LHG}Seev~^k0#J(pXu0)@`$aKLk=SIOMH3OLV zB$Dh%YlSh-{*wCvS|F`BPD(ae!@;4R_~4mXKN-Bnyhlw;UfzfNyQ#7+{>{Q$s^Mhq1#u6@e2=7NPBrVg<%I`Iboks8XVA>>37aqa(e={1@sImYC%N$_ zaaoByPP~-}dE#AKam6H1{dAD5j|?SY5rU+CUnC@5FO!V^^@p5!>m!l$4Hx}K^H}+x z7v%4^IZms^e7CXfc!+h|ht2=Kv6x{qnXAirvhb@s_?gA9)Jx|{{;~Va)hq@piodWQ zNy>0w!EBMalRy{zD}y(~)M$6#Kd{E{7?tOK^xnW>eBQQ=IP2nLoP48@M7D=xpDWie zz3?`~H#oq`RX6dG!(&N_%zm~ZeK=mv_y&nrvOzDU1~#wX0XzCS(*Unyw4Gd7-JZO>K3Z>$kHJ6&6+3O znR*I;X}rT58{;rlF$g^dij?r8Ytor^VPxHi*DS8HFU=}bBYB2<;n%1)tZDx|k>}1q ze#T{3&@u!hv>){wJDr8S2dMBo$!^VzhA&x*>9^0DA@|o6;uR{s?`{3q)GAlFF~9_j z{5mn|k_=&|HBqKCQA%7UV#TxDXgH?;`sokms(shNzWM&5qsts@$~~de?>2j*+aKq8 z$K&Oe^WbE<2R?6YgIN1?)Ea8Y^=&SpYsT_Q<$u5Msh=6|Ej>+!_qrt3S@s~ax`8AjRkT38k=S(O_}k4&mQTh{^ZX!cSQUQZ!_fa) zHTkor9l{qmQHybw^ohGW86)yXoL8Jy+@P3@FW*ICXX+ljuB(FID%Y`!&Tz0Y z`vIF$m83~83(4eruH^R5H6&V2>^I4(W6e85q9@re$#py=?KH|_MS41DP}f9k4Wi+c z?Gw21jfiLcG8V8$)DGSHYkDpTHwv}%kQ&4YQvCk z%Y*WrqW|o@2lfov$dZn9LD7q&kfNeU`O^RB{)u+<>)I=%ud6)$Z_Zk}rbeCG3{;{~ zce}`ZuZxwNM}L=O5e<6gp(mLqk)!S2SESdENZE*w7w~CZGCI4xVjr)K#}$DJaMj5e z{w&%81*@;(#?wb(dW0rhaRAwYz666hJ~F*Us&GJNH5+q%7Py99M3-~#urz2p`%=3V zBJ97zR>e0^mn~%;&9)w>D18Knva{Sd-Q10zPPwTb9j*u}BA9D*X zHxFS?Dn-f7yLh(hZ(s6!`v|D=QRXp|yK$ypA*PkS$E_p|N@PYtjddxEu}Fc0#Aj@k z@>tem(Vr<5T_X?H^~2y0S83l=RZQJ{hCH9uo3>YtqNl}LO4r4h-Z7)p^IHO3>6D=< zj~;{diG%QQZgOSG>|g9#&xXqN-#oc*Q!A7&s|BT9msmhT6zslH1(P%PQvaRi{A=4u z=I|2Wgr*sONx4J3T_1qy<3Qg0@dYkzorFJnmSez`8GOmzg?wY5w@_PphMf3x3Rdm1 zhaO8mVmB&QYLfmB+5)@aRpSBNI;)8BuemrNDHVggtoVr0q0~3RlYd%afW;s3aPj-q z^yRcbNV$(#c7783^FxnJIOxx&OwoW5Vy^LK;{c-Tq5&BV8Stsb5nc4BpvD6$i7UvV z+V*sGBhQ%-mCPJs40wt5G#)2OMY{H{2U+ruehKI??=%`K z_vMkEO{i-l&!@Zol9U={f{yv}Z4GGk~f-c5Bk zJmyBF*8J+bRqVgAl{{Og9dGS>h?#G<@)Iv6^Zza-;Et`;U<5AO4uTkBcl%v)t_ z!!MB?B(4KDkCf$$M6S!HdjnuuXAf~wdha(2KEL)&F}vBDwVXs6A8&U%kUbJs`) zNlr3}qduFrXas8gFyiCJD5HgG02?%V6yKv&Ny6x4nDwML{p660Sr#I<8GmLK01k6nk{v7p~dT%vy7$-81Sj^B5sGOKzCK5DSWwmCL<^+$np z!)ymS*JwD+X$u1NM}6oS!}ZeMy?WEfzsk{g&oW$M{=;eh6+=*;Acu2@3D^`J0y?oy zATRR`za8ogj{TVAk+u^w8%)M|Z6BDqm=C!fH4BfVMdRoaO?+x-fkO^WV_HY9LX^&C z*m6yle)v9?F0S;2{9#kjOy-}YCE}H&`uGztNbCmnC(G$t8)YQ9&C(DdYC)>rKt^{(MdCg<@OV^ zEp@hBE%zOc{*won11#Z{d>9stx5Lnj8InI!T)=F@FBGcd2s?2Mmwy?H>ohlk>GGL8 zp=vwxaHW9WDp#yW_a@akh zwJ{Fx^=arM7)pnIbiz>o47Me06os(?z1gC3wjlL!z_b( zvTWvm7`aUgjRIYm;jA9?AJ>mRJ-eRkmMx{;X1(beTSoHMKLy=csU;%L1y0n=WAp68CqTXh-Sslq2G5J)B4~{E+evJ%~TF? zk!;3C>0aPtHKX{h1xtkE-#E`WRza_q9;PvVSLl+NQ>b;=VBuAqx=`yABAX% zKV6+?LaA~+ETP~$(A3KLU_?Aki~3K9SnMR+JZ&ai@2KNb>h@#ns9NaVu1jsw2hgs4 zrL^CHJJh4sZW`}mMz5?IBpg-mD+ubwLR`7AaB?g*4MB9#pO*=*5iUcuEf%%x}^4Px?4tx%3RP zAKKCLyT;KTx{kWd@&m8$?flqzSK-vf65$4!CG@KH;KH$NXdE`5->+KFFZO^S*Z7jh zrCSKf+kJ$6yHbTRu1-~^fz;y zQ*Gp(vHTab6sB?irmNmNtZA1x~@g@vb9(y7Du(Uf;NB0FD4 z$ZDD-ggAN#bNqb-g*177==KtPoil+i8gYx)1u6*Rx>SS@m)wP(pR0sG?dQ}d+nx4Y zxJx6?XVQ)VdxS9K)54wbFk#t=*}}NjDf|x(6bg4(307BCg_w|j&J{M+&KsvZr2Dq& zIZrWq16`ww=;4LKoc}#jfyamT)BEYs^u(XNs4{LR?KZ#7r|2pOC2u7{y6p+DO*;S$ z>o(EZlQ%-U=6i%)%7WI`(ZVMUPoaZQq2cd6dS!s6bD!-$srCB(FwyE4-=Q;>FZb`} z%^L>_WrsDmL4>t%r^r+Y$m-!rBX3iO&sp?kb_QLyH=h2xF_(??7Wa&1`oWjJBe~bX zM*iV(D<3au=OYdN@MmZK(I={I&O-yIIUA?s@>x|(_gVm>E?-hHYpOjNzctJa<3zk>%0^&@im_U?W{=qwO= zj<@m=A$GzBR}gG8JcO{>69k7#O9cOfu{$2W=Y<#*8M_p?HePeHUc3DvD$s;vxz*X5oen&$=*qo{>}{k) z%L(uFRfUmi`hw5jZd`O(!MP{=JiT5p+&TSDC!JGJMs>X3P>Z^HYNmFVmUc3_VwW$y z=KGM!t@=RC-Rh~6={Z_t^oM>)8BXOaN@(8ho3vudNIJi8FfHEgz}1sK@WBcn`JnV( z!tJ7$r0MDc?ziEV*c#N~KV3)h`RO)vNO29M%C)e2H%AL@L4$-4r>Vl3&x?dUYWMif z>EHQ}DYLov>0(}RM&7w9IG%b8t)K=Yb7^hfQ9fNsmwUZAz%DeHbGiN+R5SMibhMbz z9N)>jTk(LftuRm6Idp+AVc;&o-bqQg7CA|1QCumw3i*8Jv$?ePy%(szn@As4Y@|cA z_VL~$)OoA*3qEON9v{CintsoXpjTQu=rT`P=Xoic_^}y%`1)b(eDb>z9uxKu%wEJ$ z>De%9X!copYi$*nS*Y`FO&@f#s^)Tk5TlNLC%c~bQMr*iD5sr31Ga7E7dH3d)4v11 zanF`V+g_y)vrf^%JMVF6ig*?!Y!(g_FA<(pj1cPj+0vMbG%L}TZG$4BV%rioZK3+8lF(4IjxXq%BCMCh3w4jegxVoN!l{9mCB->;Y{=2g zTt)XU^Vk$it1f<_+y1mtac75~8(u(Pj7y{L+8^lX4N+9aPs_O|($+cL&)vCMeh*De zjS&pBf`rs3MuPkLp~6Iur|2^GJCo^`ty1zhpE${M`s_ zo4jDTODCvbPp60Horf7kYIM?^!Q?|gE%nOzMXwzzpk8yc=-AuNqHC^`*EYT2`O z`4k7dJj;mwS+R=>rUUq#H@1Ss@E`o4O)HmL`txHcBVg}Q3mVW@o_;4WIA(h`^&fGP z=DED0eov=Dl(RRTtyV#|SY*=Ib3O$_z?rVN?YDw0AiC)kc5E|OJfWBLv zMwQ(z(Yp(8Q$1}}s(W`V-MQc@Q|@QSFBB=m`zB*rw|yKSz&-J1o&ds6;v?+@fB zy@Vg%ztNBJBb+~^A9Q{_W{C5^@-p#vXe|Baa)D}WSw;N}){^lX%-jnuH&J#&N; zU1q{D?+L={WDQ})j?F^!nw3KNh4sRXVW||2{b30AlI1mi$ZW5WKOs>lciknP zvG;h_u4vAM>)fYllIYbP!uJUz!d%X$m8o>W?~^nv>oh&vQcGL^iSMibOFCNn3_jPoOHJgi z@{XiG)bhhix+6!4J}^80p;5>bFZ%HCk&AiDg#J9wXDc!NJ&i85?}E1>3#s3ac*zlq zoBX<#B|kqm7n#*QJ~pif|G64)ZWO~a$NwKg*Bwq}8^&c5MabT>(2ydW_kOh`5-Fk( zmA*o0X-LQ}G8#mZRJ3Ijao+nTq>M@_4WqPZ@1*g)f1f|jxvq0P&vW0uDThYmA}RM# zJ{8EGqRZNgY1P#Im^ttaN6$Nh0b>9c%upuFgR97-U!9aa=FrRohNO_{Micj`QMZc< zMV-o|0o(5=cCZgk-_57nZqJzJji=zD{TfIQo(e~j?U>hk^^94b7R<}{;N(1|sqf-i z)O?(YuO6)CK0G=AcjNhY|KekOk>Ckxx@M@KIs$gZyntVA)zBC89ZYsMz?~<>O!7Dh z!KQj;`fh7LW)I@&+>f<%Fi?kLi+WM)jy^doY{%n=V&U>#amd=%%M|R1hV%1dVeHO4 z_%A{z@L3WLGyNs#c7igUWu_2qFd*~fN}RpTk}_KjNXog8&c9ntx?>XP?yq?24qHyj z@5hk0<~Tb0QkyJaRbj~RC-`^!xuE2=I(!lAhOCDyJu>@nPt-S-Ao_suW=6NGCYjXa}WXYvCr}E$nwklbcnv zoiRRZ&x{wVWZF1Q!Ip#)*tlN>+zKCqxk)X|#&q!TxW^ptu4JaoJk9T7TJ%I|4yC^} zrkU%o?vD+2e@e&ipHJg^nPnsus!v0m zlgZX+?ZZqglpW_3+3Jh3c-a|?mB{;SHeoBw00s-{w2)^;kax_}}DP2A(x zJE5|y4c4j~vr{(wg<^3Rn4y0VTz_Z6K+Fb676y@|KK_u%jNr5Wqm~qLCGN`9K#>&oz8u#)>A|pbfPbwAusO1Z*KkW>UdS>| zV(V-!t$PfYnNZH%Kk$rM7nT6Zn8R3UFW|yHoJJAl(R6n;-34%2 zts3(Oql<01wo&Z@{cjc!FlY&1!}xBdx@Nqv=nLxm&BpF;(;&ZK2=G@L6n$=j?76?W z6T))TQ7XWf=8|03S9fk#^hoae(M+6~;ed~w9N0?r5o}FoDSTg|!y2@Nf$>RSW^ic% zvrD{;^UiwD)-_jqz{3x+f4cyfO= zI6IWU03WqCef?IL)jEzX_}c-WC)UGNHAS{s>;~)_Oa{j@WgvS%8&>|47ab&xCYO`8zl&9;R!e%jsn840^FQm!h5| z!l{-CU>o5LZ$6)ZrqO_(wzs24Y(5S<#**0aMU*plGCe(IP5v>zQRVg=2wrdszJ1yU z0nxml>+fuM9k3RT7#P7v%NsDWOM&e=q0O#2q{jB`lwl`duV+3@+zrjP&Cu}noWm>O zFvh97)5f%7Og)f;j`0D+{7xeMNIgnhGmqv4S7O?^TbOi56%C!N(1)Et1MW|7UBn~& z6L<;_F53tixxT28deY%AFTZaodkgvhwZli-FR)Bw5b`wFfzzwqux-geod~&{!CNG zr*95i9QB?_SzsVI7Br6})IMV28UAxCJ(^x5U@{h_g47oSXpLPAU6BXqnBGXbX>LKy z^Z^AoE`!~lFP!LuCm@AoknP^WSXCSsXf|vDmlZ3)GW3bixuYBRiK(NWlQ_9;yUx6= z`^U{lxGD%Yvfz&2lt=UJ>zDzBCERFN{t#&8L(rNoVl0kY#(NGla-q!QIt(l=1>RZ@ zIn(Q~u5>2%RmuvYTN`j@S03L1vlslVl-O%GEZLmTOGD5*1=~CEz3g(o$V+bE5qppaeN-qW6Zo~&P4ot17$KV_&YosMn<;4 z)rH9dR{sWPIF!i=G{u-KZSo6Z}_FfNrk3Qv;9{6bPb zavxW{s71w%LR6nCN-LLI(eDXc>2r@f8Cs~&*i=n+`>U65ZPrT&;`@5G+KgjvkFEpB zMiX|`Xitzma#<)Ax&uu12_W8S0q1u93GA?vVtr&svJq##LhsD^ptY-tNnIxcUtXML z7WM4o^vt}tk+Cni2~Tq!tUhjnIE(G@seprtMapnmaWwt*X~S1SclzQ8kPmBLqkahl4!My?T2U^X%Dn*y4B30yh1_~Mnz8>edd$?JrquiR%Wo`nBz>t3!?nT&hZ5m= z>2xaYKTFkn0x0yTFG&o@QOd9bWrxMlryc=Kww+a*^XV=lCoc|Fe@E4RgNK4!?Mv}y zSw2(iI}Rpq4S@G!_;>6_K3ov@1h!L)OSmu=A6>}dVwWsLi-tNh4H97zyS8B9R}ly2 zCE4IAaR*)`Oa=Xl6c7zcf_+8*m7GXC8Wh_T>Z2*|=6n^y`J#d8wc* zt^~TG1z_6cioruRl$`PgPj}tKgD=7`h}aYbFMk>{H+>iI80af_TW-qw$==|+GnAOOucFbY`jUf~4hvDMQb8#9 zz+O~tEZoQ#LuzepIMN2ONc{!A#ReSwwhhIc7T`OPAxuv#fSak>tc`m;NS)XU$={_w zmT`uJY(Lnt>N&$s*obR6Td?9*gIs79w=>U&i95d=BbUlJh~@9&qB2@>?Il~xwm*w6 z7aKFGn=L4NoDO;Y9YMFd6sYR1ByApTL1t3R>7nvOR6lHnUwr=I6qgq`{7Z~Nr|pF% z@jh^HEQ8>C=RutDMxF6X$SD3D%9aYT=i$nl^}**kDzD&VS^+EHoyFWsZ_w)N4@{rY zhk4O|!C+)L6jr7|Sa&@96urhQXe$<+8l?&P*>>=~xEuVtWT3Vn47v;s!K8REG6YA; z5oM`&j0<`BzTykz>S3?N3HW?VlnuNm%DSnqWcQ@RvWFxj*})~#F|zLl9`*Uf>$cWm z&h9bnYv&XgYY+v~ov%Y$ZVgbyKCu2Igz%~S{Whrpsx}#K$n3wITdOPl@NMBbx*fUiHzZ-t>Ly5yehQ6x z9ORlzfhn#^Xi->=!CbDu)Y~6R4o89h-s_;qWq@$o10G2*6eiQ#%e6LIW_Z^D9Q|8T2_2kp2P zL35KQ(#8NOSbn1c-paN^h4UqF`1=&I-99q&M@cYIRrz>K|1mghGl%mgl}yNw0!IHp zKFsXT1m#n;@Y2j4iYn?jwSXVoqnz6{hOe_hCHyHQq_x01^G-0yHD#MFx>Fb(pg)ri zlGgTOTAATRZ3b~9d1*EsHjE*KB~t7Xn@k9vCc^HXuEEwd#!!@D2GwOBAUE(Q#mp?` zXjv0XE&B%QxE7`_JVy6dX40KYp;V`kN-0I@%wUcJZn?>KFHM$(BM}DBo1p^(>(V(5 z4|R0fDmOi;aUv zxm8S5#%f0XUl#N2D+T#(UXO9v~`(8&5SXK8ms(BfsxT(}%c z_`;rc<@nNdSvA`J>>1AsMnG9KpFKKm8JDG=#of(VhW8Cp@y~)xG$}7(6MH2Vfem2Oo>;l>2S0Vr9X(&r=W|kLE1G`#LoZ#id9LvuTyzPC7 zE$zSY#uQ~bK319xEI-4U#@k@?xCoXGE`%x%2@-fNq%eimK*;5Sv5-Q6k6R<+GxzMXx= zyibv3U1lk<4o`5m%!r_CRiBS zf#kJ|lYBa|;U|%HunuG5Gn; z1Fq)kZmuSwS$KA{0}O3HL09I4&>0ood$N|J2!0^GW9E@$O?7{w)&$o_OKik~~f4N4xv`~+jXwWb~~%TJ*@ zqnh!?X))^Fr%c;*4N0QElTqpqfzQ`WVD6e!n9|kE%nIHoq`n3&yU+&2#WLWjqA@%? zxd^sJy?|SJd~WsDc*xIs57`I9;ZwRXSzdIdd4{bRIHwUQ)tg~Y@_a+sKGOW7O&(?* z^gG;>TCe$2|0Nw7QcJ-3@(zq)m?Bl3*^Pf1_k+}$2Bz`&wwkrp=Y;r848q2L;l0Mn zkT|x5iLN$g&TV(V8;hpma^Ze#b{s*`TMei%^EujFxr$?KWmvVITKL#<6#gik1^x8_ z6h8a~S+fzeTxKb`XRB~QqFHcZ^$@If_5!D?D@pc|4r$(xrJ8O@ig`4HxvHN4i_H>Y z&vXUQY5xP&!@BH>ZHjFA=`rj_Uw>>~d!4J`#-ZYV5A;(Ja>E`|anH&K{9gYNPh=f} za{SIrxNQLY7Z$=|FEuAzcF?#-nFbyikw~x>Ie!{MTem1e!~B&@@Fs0W zCBhB(t}nrQ?j^!$5p3USjxiDGG|glIJ&5z6vwQsMLYO$a>8c28VxiAY&r)FR#qD5& ztUkOVCwTZOPN;f$Je`V_qo@W$N-X`4_Vty6`k`T_+4#P&{D&wb&8(!oZxl#(=V}c3 z7RK%F^I>}KZsuy24RQVA$GC4DD`575$>3&pkg09bVg~TK;L#l+W1q<2+S(<6@|(oGu~aw-!eHRngLcPY>h*dE(U#@=hvsq@wXB%{rp1Aqwwg{H zH6hFYM$nvr6rA#O1$U&{05kLEq13}Q`0$_#>BvaaqJPHp)8GewS@{_%Hx9tdAZgaK z^BWjCtFhPr=(GEVWm$_kd9eLsD5sS9o9A;ALFH}*C%>>7#-G=Kqq+0Jf#=QtZQ}5q zk|#BrPoogeS#)TcJk^|1r~Ep1%FP`?5l;n+lM|0?=eQU9v&3O zK-*1ic=&e}__fZXix0FYp*n{8`m9LA;0ouj7LRGOGr4^QvN&I*k16O0W8*GMSJ&u8ni^QWqXbvYr_W=9s+4S6iT{<%VvV}e*_?~&tYnZZ z8_Z7$lRm{eL_W~K6Gamlfu1WxY2P5~w90M1=Wj-(b z#|b?K@tJHq7C9*3k=wJm!`o(K(ZF1cc2_`8?FG!q)2ZCV&gr;ahQ}?29)qk;DvZ^< z2QuEuuqN9CCU+adpUG$7$DVRbJ*iFe+?LUW`}&;0vQV;ry_&YAMv#a6LR$E(83Qg( zq#Mi|{F0Z1PCHV_Y)vUu23(-u_jSqoXdal~Q(|44_%qkF0IWOb;h4%EY)(zUB}|m? zDUYR@c0J?XW=;~6RTX2`k!YB-YBOx}XoI;j0D9|9VGolA{%sjgh24>q@G_M~>`SJx=^B)B zD^|EYcP;L+lZSn2+u-R>{*6Cf4OH@Ct0$pX&UsRUIfoxyHpN%Aqxr*Z!UP|(a( z)Sldnzy1W{Qi-uNd5a|Jd(~pEzm@RndwHB4nE`Eo3pj1(r*JT`2cBK*fV|r!V459@ zH{S2V_*ZgxbmAHA*12~Oc-E3VAFj`C&Kk>_H($bg2{l+0Y>v~K)ak-jf{wPE5Mx@z z*F4Vw|F5dF&*}`Srf}TeiLOjWX(}~#EGC=HXQLG>0h z+>!GJ-^H9q@6ukpF0^#lf{O>%z#KOhNODu*XA}$Q{Wbya$8UutS`MJQFcB8} zzh%DkE`@_S+0fstAb^l`nEr|PCNFem+(TC}vdOm@qe^>>D(>ccZsf6d$Q(9#90Jh2 z0#_w_AlRu1>_?@;6{|gLB)3(>7!>2rTk(^`eh9sx^n|JR(7C`<`R5xi-S>uenxA_ zb>6Enp7B~!QF|m{lE84kDM-4>((1+)RDEe|7#Ea18RS2%J^ znl818(D0>^bk1-By;fUEfp*dqZTXkeZCU}>Um3CN-HVL4csP7tu^KiA)Ir%g2_~&x z2_uwu!8$87IMlTq;s#2=BXukzyX-IX6g(I<#g{wqV>$C?O(MGMPR6(E$D@?{d0eU` zL#lO4sr=Ro+T^209_@-$5%m%$pL>JheJk+J87(?BUz$prRB2S>2oO`BJ zn^uJ9C-(ApGN6NcESF}unTP5c;Ln;msEka(X4@pZW35Uq&T6eZFq4k|WzrNY!R+)R>ispD&Q!)z#uQPS(hvuZ zi(fIP++a!{DiNyYV_o5AT`On!D(f>R9pTWW~}Q3vBO<3 zUiJ*k8_s0H5>GN?bY2Tvba_49hl8XqeSiku$57yzSbA0dlo{yHf}pwP(Dd#r^X0)F z>h856iKm%?bnj(m5DTGco9uKe2KmZOYw)8Az98CM^}1FaoW~xaJRG#WW zf|Bs}<3HSoTgi;HcOK5=buJsqH7FcZDDRRwP2D_-`|R2Ts!x-_CgB`2!*n*--5SRX zcZS057kO|zW3N!oWI33vI06oFJ3+qEhZcG-r+2@$()sQfoT~MbyYti$9F4y+cfZ`K zZ88qxoWg$y-87S#beq{ISG)+GMis-v&(X~3pXXr0zj;h~(R;|;^Z+#ft%c?OVzj2z z94{?ts`U!Ug2tZ}sFkWfGhf}nL)Ka}Y7*Z;`_T|9&98GFc6;#pG)uVHbOuaMo~y~0 zTLf00u0r&NZrD2b8@l2;!Np+)$ZYY)9gB6ipwc#GgPA=(*XYDW7A5#)X%lW7+D+?r zG^6s^2k3llB5vvILpv=NH+#v@sJ>BjxOEk`IzXQ5PItk`-IAC(?wr8?{Se+7pT)H- zZN#;^f~fJzc3PG98%xhlrgH5I_-^eMF4(>Sty=bAWq~_elB~|Up1TFnBXwCBgRQXa zdAlI8Zv=NqqJZ(8c#)a)Wfl|oq>r<0iosR6{&4kNDeSI045QskK&j+AtX<0QFRfN= zTKFR{vbMqSBrUz;VxwNHgmUw+?bZVyJ360fV_S?(#C=~TK=jAU(cV* zeczad33GFC#yBO?jl7H@zwPO+v>1urIg2W;K!@b!(Xu>us*RgXrTsVYNccVM`DRSE zycaqm@;GxrR+rBV)D@_PX?{ZF7a9&!d^SEgn#GJfGQC?hMjixv=e* zDZFcpgQf%fKw{1%u$nsoR)p$-on0|{_k=U~tT!s84{^VZO({@zJK9F4Fx%d$a&{vB z!BN-iU~^c7eRT8;>9g*^XyF}D%e@50I@SlbNg!dED+(4w!tun!E05#T|OA zjHR28;7z$=Vb6;o5OF^UR*ye1LjxW#d}|+En)ig!_L&90HY>u<$fttPMg`EYYhfmx zQev6xB|t0mz(Gd=ejS;Ct>(Ojc%kO??bx{bg8$%i+T@v($f7L zK0o#re=Q%v8HWo|OTG`D3o5~-UY2do6Jw`_#KAZ37b%}0@}ycw=vuFak)6=O3d$+3}#MA?;>2f%Ykg?*jY3gxFo*oLodlslq`;?2*{ z)c5r?-=F94Uf+f-68B)|mrj`ePy>W})M1KEAS1VMKlm>$p~dpyblBa4?xt>_?jmVk zV_}aqgBNR4hU$cllHxQb>q1BpLq@rR$Cit=}w9tDsTv-{tGvz4PAi zBok&l`dRRBWquIYVopR*6pvUl`#CnUEe~rbx!Jx@cv5PGnoOd z!~0=*P7dhqx&_O+CD`P7`H+3)8k{w)1~K0*xb0K}OZQ%bGqa`G=@aIIKs*&bi%p~N zUp(kh%qOhZ`-A(YorWT#eCT?X2J|W&d`9l0E3cDzEoC&dJLeOlIHb(5e@ry^7xs?bjbXC8b5U^ z;C&A}(r`G5|K@k|9+5(U?Sv$_U>*s_9(ciK$8w0j@sycu#ODkTsp0y(23V`(25SxV z;GHqR>O12dO0LI4=$lobYM8`4IIqm-o@>DSbXQnsZ%hVNqwsL?dpvgFD$bU!V^-G4 zLd33*aIN?=;8Q<{4pxN0pzAQ-RRY*278vzthw!8K5$qeKj_ZcS@Ys}k6q$P(L&|>( zeJ9^zL|gYUNv9b&H>rVHFP4K&K_{^B*&sG|Zlndt5%6EfBB(MkgC7fL!R@yWSZQ~G z+4|@xYWYm1T_cyne*bHXmJxqOy(*Y3i3WIh+n{ho0H3*W%ZroJmlyuFKf~Q0tq8lT z7BhtlSAwiyJxs}ngT;9+uY2{@@R?@$#!s2Jd49A> z&W{c(Gp7}jX5_TVAETb9CKSkE~D=Jv%9pfMZt?yLrZ&3rg>{T6pc?hD@= zl8Jk&+k}=|wAk@`CE31dyarzBF1@=?bUregiWTZe@mrJdkot3GAg-LreYXm;6Ow2V z@mfm#SQ6QFf$RjI7`q>>O!vSqftuGDn5dt}_a4STNkZak2EoT zN)gQLFEjaG&_}h_&f_pvw~qT3UIjhls_nuz4_=K+tq^*rgo1u$B?RWL1JkqR%v!NV zX8hS`2$6aLKW%P6$d-NFo5p7tu&@fN`$y2_r;F(0$|@{LKZ%(w%gN(=3I?r4EHB-G zrknQS^srUDK4lj@lwL~$w(V=&yFeS( zu*J-{@p@o&BMvSXtU%Fk^Mohcdj*SH7IBL!M{^oNTy2bbsUO2~l2Ts9< z_wMjGrkH8gn2xDlmm&LSIe1w8f?8f%GS}J|yXy7PQECFFyzAwR#-+i*!X|iR{}FCF z4Z@vW$Aq&lnc6cqaLn(;EQGvW{_3=6UDN+(S_$CB+r~fG4&;| za?TCt^Bd1@jFe_qXd2L()oSEBt`>(=mC0(2C_2SFz^t?HaJa^Tyd|5NbyF;W39@4r z-)Ip2HJ-}4X-!~fbO5`^<}p|f=`!kVkx*WL8WNNDf@l6nS~_|KO|PFv?W6sujrLxx z+HJvvdql&cjGYdlpLSy3l2bUa;RS}r{Kco|N|{^p`WVNQQFJBOpZh$u1RGthzSduEv8Zm1uXk5hLw~u+{zno;kjN>Jw~enPdpv690)GLE%EGGD{h&{E5Yj4>2$<1o;)`PlJ~ZeG{AQnO}wB3B#{F>$}aj58~wd41%}JKnb&J#;nm45a1ga%V+Mc0&`MQyjgmI|!)6S# z@t6Q=3d^9^l7sf=W^8h&1Y46T2?pye;BMbryw>AL4(^$Bt+g7rEkJlW5CYe(jDf{= znvnLI6BLTYF*^^RMccG(_~p!R&Ou570+)-!sQqdX#`{JrFTH?|b2_1GNdsJtNr81) zd7##K7p8ov26N+n+{@31J61m8Ey9EhbZSV-8$+`W{}}p9LqEb+F^{Z`d~WCI9`rrcU)1+?xM@d45zA z8}q^htGwpnOqH+balnymoafU9@wK#dj|q-V-ht{bw&N^(%-HfcZk~^&AZ=I@Tg6`B zh10h5d0Q~02oKQ0FaaIjIR>9Bkm5eC2Cv#uj%} z&$0p#N##Jiq#m9d3gOI$i4-wfi{6>%;+l?Cn3&oCsd1xOWBxs|EO3NNTh>8@xgM+= zvI5KVpaz2l&c*I_sN z-vA%l2YW>fn7xnYV(9NU9K58%$X@lK?E?{{?w?5}7U@(`eHUc*w}I%6T4)i7v7`Ot zpf>O~#C`68iC-Qu-!@p2-R-e7G(3qU(t2^((N6f-&HFtIpFxIiJR~gL2JE2%IQStB z9&GG^i{sS5_UHp)+4~K+W0N;#YM(~+9#4umE1;`~4Cuwj7Ot)J2{ZBTXvVvLG^_|W zgXB&Zc;OxeJ6IKXEV&&8Oc`$7Sj&WV26BCR(cEU^L`Gz;8RRat!h8A^Lh;4K7#+{$ z(t-=|fLk(cGB`o~t!JswyMxw0-c6ZbeCcbs5`7)1POXCq0!^t@Zt>1H+$E)B*e=Cj z1UCx%ty8eArIq`0_yE31IEp4iJhyMd>mrY}2=iK(;j~qsF@LQGCam*Di3`=3J^dkm z&y2%;+(D?)-U?0c%ix)c56FlefLXIU!R%=dY-=!rdFxV0Vb&IsT^mn*fq-=zE~D{G zKb-U>N7!?GCR$xir*W5$(U!5r)KNNvlC{!lxF?d7SC6AZV%B85?JTUD_8eyaehK%3 z6XDUYJc&5`!l@Hq;!l&E6riBTcC3ro}}A#VYjrPU2p%7dI$ z;yt0u^<`XO`Up%ImVwFUhj?sn3~Ooc1J4#4(xf6QEbI56(V|ME%IgweE!&N$ON6NS z&WICfs}rhqCo`$Vk3h@D0a)J;5UulsNgeMFyYsFxdnD$O$l=3$>P!wj3p1b-$Nn%{ zBA=k2l@WS`v~dnfanN^5m31jO0p5dmKy0)KJ9oYk6FqB{ebCAkV6E{CtG-&(!%Z&q z%WgCUe|3lCT62&&XvPL*4#9)Jag6Age@rhvx%Xy}hHnfmT&gTEYR!UW z5tG61E}v5~UloQ=y@Cg>J+Rt#AEWz59d3~%H?%BS@K-SjL<$t((4!dibb zlM6bU=&^Uxrm|Z0lC0>pQS7uyKVjTtGghVa0(=a50ap}tAvaQcn8uidNU{IvwZfLIa%|o2UKq1LmHjt(33^Abfa$-V!GY`9kTD#I zV_6y6@pleYr+!CvC;~pnN5KA)V(1te$tI0?jHJ9#7-OZv`M+u5E=}f8cC9mAZVID1 z&ncwbvzzuUT}=t=Oi0?&p6V|SFkjLpz^yeq7@t}-kXb6Ahl2s+mJ>;n&j!)_5q)?( zNQ{ySVC{Ii8UHz;*tygq?YAf{_DBe7I^Xwgr62Hmzy=sQ zE0NKj=>uBFU%&@HMY`mwN^L1c7;$hRug~-+?^{7M-OYh&QRH?H7K# zH?}s_Muq*oyMk#m6^GT!W8uaMU%~UdL(HNT5~w%tBt+lngOs>t*cvjKdd~Thf0rAD zr1HKzIaAWfai=|BCy-v$PP#KO4h3VM;8!jdUmHvCy%24XcGZ|&IY*5B7pKJfZs8NE zb!^##ywAZQ;2s1Xtz>SS)WONw-`rOYX&}xlS5d$m>x(gVEf_wsGsGEu_1GXBD2~+5PlExl}E6T&$PgAnG8H2 zT7&D?yum~Kxq8{nfa*Vcm_Ah-PMhw8AgyLtDsIKjm(yZD?dXT!pLt(ctTU;s8&8|m zCen&`-@)%_GRPlMh4nZ0!o)GURQcDPvb(eB&I4Hr$%%!Q(|r&>IR$);8eptdDTw$+ zfXl5|m{jBf6^};a#pm;d>f!O2B;$cMr7u#5eKuX$mQOe#ovcNVQ{{}=WL_dehi9mg zK1*h&FV zWp?CuBW(Dk1qq{{ zaGR==I8i1ajg}VSxR^n_Gs~1#+*$&jsw>OyVj6au-!3VE($g}V%URkAq z6??jHNcT2Z@v?}UP~utpRrLe2>g#-dzs%x72XwLH>IaTJ|BB06IvHE_7By z`hWvhs<`WVcIeu;69Ew8|TelGk8eJps2Q(>mADPWWW zY>Ja55%1A-E7Ok7ie;0Dfg)ZFJBn*=R0x)p$2ldzK1(9N)-QV4ymi}B>S8o z8sG0pKS+fp@n`d>jVvAW)x+NDcBJ=Fg!G4`u*0p4)3{;CDu{mt(YxtTnwJmb>v^vC za;*b8=TBA-;{r&)DzSpnDM~v@C$=wNJrEtN^m*W8vp6 zPVn(bsPI7B8tjNG5=?lf3C9miffN4gp+6!OQk%~~cYQJ(&#s3Nf5h3rKtGbxmZaau ztw_hQAL~ZZ?pKjQr(ZO=wfl-(>iGW9le&k181Cy23|zDTeg=F;qw7mjSY@gwMYn~8q0%A^t8 zfb){1XrJIXNi`m&!=+bAJ7Xko+aztCfb7ES3b2Z!NU_)}ZUtO?YI_s5-K zP0;{ciRberG{>+N^A3{jiA?%hWla&QM$iNm7b>~oPSe$O$Y6v%MJ+2}3f__+EnN@J zeFz8F-LJVTUv|Q+4JSciX(3!yh{FXx=L#lTN24Li;eY&F;jXb4j+G_D%tuY|Ku%F8 z{85KyyZmvQ<6$i2{gOsEL&@o7J{40A4chXFuv-mbhkYExPwEo}1to)OKmzPhIs&Ch zrtqwz5bnhs2ibgQIJx);oI8F6K9^{*KlA%xw_gkVcya*v+ZU1|q@g#!7%Zke$L4b< z5C^-_Jh2&9cFjbUb0Ju)eFv+zCSc*Ym4a~xA~?H7ADD6H1+>k%fSJ`Rac6iax~|Hr zu~yAv%5ttVlaK3wS5OI4>1n`9Z?R_|c1f^pdXHe#ayLjbSPZv2vY5FE#ayr7C>q@F zPooD7sAtV1zIs}b&ika0isD5&<#q_|RSwb8Icq8GRy@shH>KzEO-Z=dpDvcw(QC~l z`u#?lEfLdYBh2o>BPTKTb8{cmS09Ii9qPOf%@8bm6<~T`ArsIU3$u6nLiGMj{I+N@ zhD|f%td*Aw< zHPV@Wi}^ZLg?*#-67n18LR@e*98qFnl1mW$*DuCOFANc6Prd@0nLim}W*Tg@m8V^c zEGbRuIc_)Ki#ZSZe2oYxfxPk|{N;HNLK~(qTDF?d(9HAZLt8jW&Wk(IUWi87ft++} z3L{c@05$9tsKfpg`t%oYg2C}Zo8o(nx#tAXk+uNc52?bBt*(4t)-LAgj{S^5k~@mz zTt=;j0dE3Xf$U{bYv9Kqjq?)_Z{dStAvb)(_!D_AujW?ERN>6gr@8% ztihNQba%&oIv~T+SC7Tiv}rHIh+jm3N(nkt#tBB}C&N3<&Cr#!4Yt261+Uw3w86a} zmu+8%OO-cblG-BBP}~8jXKEm6!g0_(wF1;X9)V}qPQ&<%RdDg}3{smPNIB|}G(9AM z>U6SE!Q?NxDAr;7t4v(ECzkK)l7&e<|AA>th0yI%IeaR73I$Q}Y+jNayX%S*NiCX5 zXXi4ccfJmrf?65Fz$`eiXb(&=F@-CywV62@nv@WH7jv^lk*$v-UFCCZ_Rg~+>94$~ z#@dvW9Cp&m;{EiaRe`$eSK!A_$I$uSNHY0ehlNu6v9e5wlw%Un)8{nq?0Sy}Eg#|F z_mO1KSc5yPk3!8)3C>Mz4l`l*BWRjq$k?)Tn12D8DEIok;KF?q{Q9|FP=C{!`FIb|>~s-+ zTCPLB7Vl81Ubpt)ab4zGpCa={JDJb>5oc%blV{(j=RoJUi?C*#9=m#j7`xj9!BqAp z(|>U$SO)ble}a;kYkcpvo{Fy!mGKgW~Cn+@uiy_eH+Q@K1g5( z=bS>_19t2-<61V|aycXi{sgHZEgV1d8-MoMEU~`vA;H15Onf4+SbWTx7f$`4KKtdt z8rlr9qtx}-Nt-vYI+8)`nyrV0^EHo;Q4(SM>J{Q0cF6x-6~>zQd|;hr3t2Dq->lPg zBR-)tns=SJo)8sJG)`We7BEt*5b}?A z?Be*l&FbQsTx0RcfUBULv{?L2WfTAP-%*g8dXxFAXvA0Bh{QWgu87;^o`}o(SBp)L zY!T0lR1z=zwM}d(A1}V6zl~oVFU>xmDki&)lz6jE0qkFR$S%_g%Ko>-g-sJ5U<2MB5qc-fS?d-R{Bd$7A7}dj z&k8-#xP_q(n*S|wn6*sG;ii8$J2U1!Kg>Ce&oeOMm)v~HUz->wo|+dUmQ%_S&u!Wu zP7UzpZ^WG7?|L2OpN(FR?tTC9{rZ-CXoV90b$u`9-f$GyL(ADMWi^nYGMqgnyvM_y zSn@8p>-m~FZ*azWNe9KDdJf5}h(qYGnGSz0X*ztpG1;M^bD~4%@v#mi@PNHyf$aIT zbL^OX4(vI(b?n(HffdC}4fe zWF2-?OFHZ-E){mDGx%vi4Sc0yH$Q60F8-wVEGpG8L6|`elE~YxSRnjQ4j%l0&wj7t z_lHzLW58Ti=9vb>M8yf5=xJ=VL=I@QjpTA4ikRHpi5Mxe1##3jFo|_w*S=L}U))s| zPmQt^PZaL3sz%@VLOFjteqkKXE)X&~F#-J2@=Ivd^#jZHY~rmSjO3&K4TJg)fri0#KV0b z2)i&j;@F##4!ZV39U>*A96AlIvo${+3SQQUZ1nLy`8Oc!MJC+DiR!a4a-#>I9{bvk{k$usT(w!;{$o>0)-$>tVb`$_)H=audoa z=klxT_1U2h&De~^{Zt7YaM^hg@AZN3NeaRI;Q1H)+ry{%y?;;f+tX5cmpD(pr?Z*$ z@3_pi2cBn_6r{5(6UOgLI?69q`oK3kr1Jec)7c(PjvZ=6+3)+J*%zV0-Fog9_R=n% zjTj%xzR%R+wdT*^^QJH4CwEG)egn(txjz; zqiYUy$Vw~}fTP0a|B^pC-9{jg{{QetB_tdi{^vU!ZhV^~fKMg<=ktI14mSTg1J%-I znz_wPr|JJs4AkA3t;EH)j1;9@BFVPr>FJW6G;QxcDoqv8#@Y^_K5^x;jgOIR!%IYV z(^1F~oUIpH72u5WQR2FKIn(uX0`7NQiK8d1#QdiRs7~HZYVm0WNlY?@e3dlf*Z+nX zUM?nAKc|pkcgpNcOZ3SpBW>z=?5w@rPZiuBP4JArhz6-nLZ2DNILk(wONgHZnX9CM zZRKEw`&qKn%mN-IHj;hizqs!~aj2f~2#urDFv3U)O>WHQzKxnkesn592V(<;W1Ar? zbO(fWdO^{&G2G6Ez33}H38!jq#O82{TSIQ)p)h?qX>byWcUJ?+u4Z_vF0hyOeS-KG zr=dOFOLTs4H2Qr0jbpPqFy!2Kf#=+WBEf4i>~ef(UpD9sTh+2O2wgfx4*Tbuo~Q=UD1?-bl9c|$|S zWnc_v#D8xa#XnCp^DujH`4~zTY*bLyG@s@Vzi?KWM3MSFjID6iH zoKjXnU${uaa_#kS@lGswm#@e4IAd&E>Vy}vK8jBCKB8~mkD)0BvEFkbp_Z># zk`|5m#JWY`V1;EtbBz(Wem+9gZ*C?(L=0S9cZ0&x4o1T&2?Dx&VZq{`)OP12!oS^x zGBuX?z-u}T?5?K~x7XoW8c&nwZozWLd8mUwYWxAXMLe5zlTuO{9-D;Sxk(2UXh&P6GTssovX|@ zFC`wTgG{40Pp6DO#2lQXK_0J@#pl-w>C4zHP8gTwpB#80$~ z9)46so_A=F+k;WGbIeXIudjrgR1!=l+Gs$;C|P=NbS!^-FByd~W zX~i;Qn5Gj(|I11fHFOzKx6|ghCsq=wen=79jps@7ELnWICYwC{EJ0kg&(XIwLompp zj@+HsK{Cg#rNOB^+~X6KoRR1#Bjwb{jLTWbysa=IR>^{^d5Jr5v^|5XTwC#Qd<-c# z-vpB!hqG^@ed)2E-N-a)@H-6dldnrhur_nm*&q^x_br?8_vL(?RF?`H8WzEwIx}it zyANL`7%}I42gqV}KAer4O2_-AU{lv_8g}LcxnVF73Ljd-yqu}DxL*l<4j7`2Dk;jlo^#Uc!tAv)Nj|Qm9-nWC+Y%DzhCH!&L%pX;v;OlO*~U6IxTgFO+l$?K=Zi~mb>}At zHFjq;MdM&Wk1U@lJDQgn?nN4N>fyupLoi1s)ZX|3i~Z3b=*%alNofBaT3mdV8vj`Y z?&gc&Ut}xey*dliEhyfZkx08jJ3&oS4Egi%Tj-3LF_O!v0<>8QpvFM~an)f-h0-v3GP9*Fmvwu?K;a}i1e9|lQ z5dOQ0DL+kNV)h?6ZSMtNo{z@ZP!|0&M`FmAmn3ZU1yU)mfK$GXCAEues25*OTo+29 zpI#Wzjhl~4Pvp?7Ddn6|+HN}DV8DJ|kq1rI?4WZrP4M}R@pOUUr1A7=Csx_-Y3QM| zbdmQDx^K@$JkxrcMqKpd-b%2zWydmf9JAFf{i_)^FE*$B5BzX?lmk~08%Mv~K1Ht! zF1*3gTxvhZ3N1wCG=7?h?*8$YUcJAY9{xTZ%_a{;n^;47vZ|CStu#ejH+5>+DR^-f zYT}U{QB?nr1o`jj5WM7Ki90JJ>DlVB`1ajsq$)+!sAmWi{pW#S*Bqw$m!;9Ppn zTObXN1HVOn4L8WvP4lSbXH{q!TS#_ZyUbiwRRa>H3`)rj-04?mi1YfN_J8yb!!$)r zd>x>OJvAXXuqcDR5wrBij$fRfb`nupBqMMcj)Jq$lQ>HK06ax;g?r|VJQV>`xV+hWwzSj^Q=65OPMBe!K8?od|7+`c-p zz$(6S!EiZZ=e`a$NX~*Tfw#vv=Y!IQeXu;@7YSZ|2GV9vB4y3GWTY3*DFr=XGW(8D zm0h~HKybap>{r1ZZs)Mk*#;vHFxcvBif7lY!+CeiQSC-S<*L+?L|5qrd2A*oRr>DW z@AcXK+0%V6I3^ifY+N8Vf+1V;RpD-X5Ii32W`?M_l8e8t(~oAM=;ESDQ%VWFbyb>< z*fa?jnFZq^ktRBry`s-%e4%dZm*SpoMRb+EO7{OLrq(;Zlm3q$plw^uNa-6xw_Gmt zq^rT!?aN`owhhczbBvCe zafdl1TR`n!mf-4VB~-6uDd^{q0-NgNpp?EE(l;hi?a-I#HoOpXso!AS`rg6|_TL_IKOUbc;spzWx4j1Mx z#*X?Ntd2eeU28bNsdX?OQ5lMw zJ%zYn%o?1d*U6baSPPT?ro-t$fh(FKCYu-9VBx1q3^?}+XOGy5XFj|mt8U$c=>K`dR!8WB0L4h!bpD1Rc-$9CL5f!(S|(sz6uYw7_qu*yP;Xi3-skfF#o%*;LMib zI}ewlXy7)NzTp}qE;L}Z{`A3PJ)XRp|ASuDypPoaN1$-S2HdXejHj0E1ljQCu>IF( zSaf>{6z;!FELLAai%1Q=>EbW+8@2#X>T6o6vN%f z74X)`AHRP!z)UZ1%=jHgU+T%=*eF^2L`z6WUJ%$%T?mgBm%(bSLgvpkLmaLmIs!`6)!C30&a9b4 zFC?s~XWU18z#SGd`C&gs@gv)t=>4^gptxrUo3ZU7F!#d+_fZS;>F60u{#k}jla}H} zwu62+m_j20eaX2tKTtbz1?ZINa9?)}1S>StUq*-Nol+@Y@w^m2`AIRIyLk)fUi<`2 zo4$abXFez%u_kxSi;;Jb<8v1OK|z9tg~y+f%jZiWe&lF23YHCQNn=-CvwVmA{4J*-CwMc(4uoD@KA<(G)lx z6c76wMuAURH~H0Kiea(JsAF*$QygpPx0B!L=@>84Ab6U$?K}v?c?rZ$*-yI7KQId( z#9(lHJQ^%5LiHdwT6-{)3{1QUyLu|1bW<3Vl~-{(GYYX{l`Oxj>oN|wHxjGT{a|7E z0Zvr>g8;~6%=^=D<5pSTYGMt_YNS$WiTA+w7eZmt6^N8or^m&M@KD4Jv>IxF^2ram z2Tuba;;=C8LeEDSQVucxMhC&e5{@>NbbEHjya;s6doGG$o>SBYpTeH2RG^U2?a1w zZ6Pi7S%uoAx1sk<0j_npgr5%j!*9c>r0$nBF8Ue6ol-anlYblIb6SMuO4{tV`C|mn z^Ddl=)J|ukAziZODm-C#;N@CZO#SExFB)dze;zt`@wh#v4NWGK;*W6C3q~U28d`Z# zF@;m>jKck@*SHm1w{TYS)2ScXNn=!$sqJ4={2=p)p10z-hH2O6(TOYQ>_f4{V5JoI z%1D|H54OkOfBczc+5fn=I;Y63-l5zyi9{OsNO=C{MAMJ+bBX?=L(FYsJ+K}sB7Sxu zbb|b9()INXGpooEp6u)<=XWoKZ%TWKNfQghG@p|_JwiXB>M~hSVhVRR`-!YNSg3f` zLga@J2Z?tRVS&vIkdbX8zZDc4lT;R|;`U@+EM>oF4i zSF&^|$1GN(^e6ekjZx93+ujzFH|67TPyY-~VO1jI?DUm(>3kO5418>V-ufnYXg{T6 zy}#0_9oER5-;E!ij>Me9_vpY~6Kucri<(@0L?b@m;C$^~QudZVHg8tM3N0;KFi{aN z?S4hew2SD3G1E|n`%0}GCZf7&2=%f_=Imu0vFDr@O^A@kJvZ*r?OXHd9_ufRq17So zQHdgaUtvf;^^c{_w4cTqo+RT$-GTeZ?j-+mG&H7zi7lS0+XCQj;575%UI7g^c$mRVDgh?AHmuxdU+gPv@m^=%&b zDtsMiZZXA=VH?T(?iHY#b{<`ym}6d`U`AiqdXyRnui`0B`Kb(K~0A*u6VNrV}8Jt~2qDzd3-OC7U z+^51;o5kXhIHZ~(2IPstB)C`^3JW`BdDpF>(9lpudbmSqy;la-MIA=l-WYfxr33qO zoq2iHJZRla2rjvb;X`zwNZJWvu4mxL&1)fE(gR#-QZZtl5m*a5zn3ma0ds@Ff|un~ zKK(I#w>(VtJGYU>>l5&a-(`B$R}o$or&Dbqe{WnOk27amk%R5og7dB!omNTnaaO@h zD{cnUJ+j!*Ey1t92EV%O>FI1SdM#xi$^(mb3R{hI}2dXt$IpEXdY$AonJJp#uT8epioIo9A4NK-kBOT43r zzm+%C1?S<9DPpENMHxoxc;P0do$`y@A^Uz6ovtzur(PcpH_;fLevQXRCOXW~H6^fr ztu*i0?!$<}q9OLcR63`ikLI+W!uq8#(9sf1-m83}Hc6q}-+p`Oh`EafhIwGVYcp&Nd5XT)4!IajT?D12*8y@`Lb1TZ!Z?dX6h$3Hf9_o%`F{ zO+P$JfS8_rj3~4orD7y;{pC_P62lOKZJ+R8T>==lW{{GiXsrM7nsiAu5z8byJm$F# zTz0R(33rA-;WkM)Thd1p_Zx#n$TFx-$!At9DMaPtC*jD=N5t**SRB3cv&c!-pMIHr z7Ir3<(cS~G_}%>x@g02^q>dZo^28bF`PGhTi%O%(>$1T5_H|MGoozV0o`+AYHf=GRAfigQ3xkcZDfN zyNFjZBbbQ_0EJs`!{6KokQp%mC*9ql^m+x^6r_NWm*3;_*SFBGLz+LixdvBQsNo2l z0Sg`!L$k+YC@|@R3qPL0>sECbK02AWbqh|@ljm`L3&q$-b-vuI88;l?j)4zLNbC|v zSZg57UOqkuI>p1;>dIueY?2Hf59Z*mJ9p4VQ<#yqX!D-ue`DQJ4}5c7m*%LJz@y(% z?A?!wtW@7Oh_tJPK$kCMfu!KjcDjy@<2&%-!S}d(c^{6t;Dw5{ubE>P#=*oT&Tv{@ zfj!IJfH7w);KH$0Tz}1GM#xG1JirdOS>Ct}BUy%VW zs%s!M+MSDe^_ix$Sz&K^JU$)jf?wA=qjVof+SgVv>dWVVhxct_7F$H_F1SifLjTb3 zP3^Q}vo$k6#fLjy;EkSUVRXc>D(WOSL;Jq4WTvz>*PQX43)y;6@>X)|4xt;{qlsFGtRZqgRF$JFR!J(X;o zjC!|+;eIo3Y@Rj?$4#1sQtJ%S*X1UY@vDnoJ~Rv81p4F2Das;Av!~qhuiEt0t)@{km?ogA$^13yO=k!#=eVB>#DV8^O~SSyh}iJA+Zzg!^vbPkztjALBt z!^vBXi`=rhd@euFR5WAe8ESbgkBaWOP?z<9eM{D2mh5->yweiH(u--mNjsNKF4EP9 z?D1P#82wo{nrs+(pQ_+Z&NMQfX6xP}8j|ao9o~c7JKJc|E%+H1eY9ZKci&--^W9|m z^o>yEx&+D(e8l2x3tSew7FR0m#P@w$nJo*CK}vKz)TuYYih_AiuV}!TK6^mbX1}H5 zZZ*)!>z~uqEu~Cn$tl7a#*hsv@-W?D98nyTPnGowsN2XyoO=5zCZ+Ddgjjw2Fte7p zW;#L5?<&ygO@>@Y27(gx;QG;d-3_C7&umE-7bEl;k> zS03l>xljFeiKyhU|B!EgPWv2xa1Wk1!L4t*$@J<3`&FCG33WXUtwHDoth7 zFZ?Sh5B8zMPaPpge&45G%;czo8>KIk$K$NX&$P)*24&2Qu&-wV^=LI@)NXc@0;%mH zIrn-Za0%hrw#}fq<|@%|nnzM!T%roaOR4@v89a7jC+)qLOK$IrWva_JlD7D2CTqxZ z+UU_sPj2u*G)l&tjz&6V$6S<}K7{gjpOD!zuECfkso+EWL7a4gBuB>4kh>Xn+QXl6 zW=q`hQ=%4;DUG2%=iV~CCud`Dj1d+)Y2gmz(oz}qhP{N7wV zuxSWR-rdDbShx~81}mXAMV5)ZTt_cuM}zu;&txD&3Yt?M*x&4$k0de?2ZjY9&X&bz zyC%`ocuwA8T#+YCMm=X@(hbsoZVpIVdpFQ`%MgekHiF&tY8&gKJd<5-KL9@AwZwhw4czSN#?O&j z&9`@{@{w=C5#=fTSfk1+OS-dOp&IO|mxbi@t9n|IkcPGYN%IBc|KQEEefV*tG&Rf^ z0~z)LpZM2$81^L;B2%QGXI+bJz035QQSkiDe5y-JUt%c++LF~xtY*CClsoRYss0eliaQzdsM3l z!(sI{DE=o;)#~a=t78^?Fh30sYp254FJdaa$OeyIID$Fr@^JnlLlk?N6X~GU@FyV^ z9Ij_VK9z-hyU8>lG?s2IOUH^8=dj($0;8^`GwO#&L4~F&lzS(`G~KOmVBA#EaMNpa zc+zTY-#87cXHLQ7i`(epv_Kd!>L)21(ZTHPsKTF*g5caeEBNF44kK?!v0rlYx!Ab@ zw7C2x7ZtQlL^ex7SdJdAF!T`wKYk3-6DRYgv17pK1A(_4->LuP`}Quxl-SKNqx+8A zG5NknD0f8}I6lo@^Q;HV$v=q;6=DR})Oqf3Mgbnt7UmR&zO?45A?5lL=+@xbG?UIj|l%0qxn zBQx01fZ76EMfG|p(>%f+Uli?zlESH&Q{Ta5r4!JQ&Bw86PEh@K3@^Xn6O3K70=`&S z@Ga-HSicKPQA2eP2{JN-ka{D$^6?L-{1y5$PD{uFJzpZ>*N@|xRni(&MU3tHvf^`L{nrbFne#Z7L?K zPJYFa`)c9l;9WBNphH{UN<$$NkC_K^Oh2k&{YhN;MSuu`t3XlVL z=W%dip$K1}_J;W62gGj1eT)y=0E-)!!HYk^C^IpJk;_W3%URV*UXN|UjSm{&;FCeo zjNOh{x>brOaeh=X?+J<8t_BM%C!lAfIkX$<(5)Lz;vGX*(ei%_L1I@nIG9+$5F2fn zw!o8f`?VDZ)|@~`!H3%_zKXd%DU{~kB5SVP6V8rXu*yk}eR#J7iXEazs`P1G`S&l9 zAy3f%q9K3o*h$PeIfC%s=Fm{0#ICR!&0d((3k7Ypuv)mA_XW&H>Lbr5DLe21lZNn0 z?OAx~Z4!BXMqoj+Cak`s4eR-_5A=?>z(Bq$WvBMxM?Z66Z(WX$Ju3J}y1lVZ$ds?V zC&8xMHN!&goiM+D0$(lvBeOOxK<6{3aaCX=4#d2`pW_ifZ`x+R`r2}cbaDrYAr~M{ z{Q^{t{nOZN1s$995v(%7RS%T znVVMN%xqbvd%f^J^}PtSf&)qK^)Z;0@EPodoc86uA~IKacF*ltN>${ov1i5>yyRVl zvyaxHZty#_{kIK!+S9N^){$#9$>d_Md?vDAUPEW>EBFxK2Mf4=U}bU$j=8-e&KJ%y zY={z`H;u*XCqLjgzeh+6KH-wkOqBD_Kz+46^c7b?GSCVR=n7ntsxy#Lbq!)~b;F;+ z3^0vaNG2|9ChJxY!PclFRCb&g|0JJ8?QJT2;QZl8)uYgPiZUwA2qn{tMnHSVLXgj+ za9ZFhXBRAo`FB*|`?*k2=kQOYugeELEgR^$8PmC{ZpCgXl)i26o?8`eZOU#_aNrc)lG@}j$hKpb3h_>7&`tH5-* z3(j?(fl*EC;dPf5IO#Q`PoV}(^H5?gdu$;|XVl>5hfi23_ZRG(roq1I5+jSBLG~UqOa$8XOPao~qe<53{DCB;x#JTTB(C}eVV5<5QJ6m1BnoNZB8&B|~ zV=>uXlMQ=BI`~qtgnm;P4=?XX!e8|ke4S%Qc1zud-ix*iJy(^;d3~ z?-Nc7&3}L>{3DvByI@|V1`PRkge>-0jr)RC>c=a$Enz}W^p3?@mm3B>{Vbt z5C4It5qG)g2TsC&Uk@RtbCI4FX~H>~I5^UnjJ4{&(f^<#n;u)tO#NGmilQEnHeHAj zN_)`DPm$g6H47K4HNt55d!RhWkT*6Mgs7z>XutjyoKbQGj_Rr6?|u__?O2aF??mvo z>?lo-yMzIQ!gSGe9hNP5%t__GvQ0V~S2@jWAzii15^If$Vd%K+^!^TCbo*@yy>;H$ z8WamAzbonPQIXWY%N#uOjf>G>)QS!pvD-6Dl^o00!bH1V8k=^z037VLFMz0Dz_T$Hk7?4}UecyW$ zran6Zn*N8VPNO3!>aOPM5_{;k6Ra=~8BOXG73u8mAGo=%noGZ(2dattNKn^4oMF{1 zII~O0X1iZ>g!O*NTa`~Y8I}`+3QD)WT8{h~KYDypJH2t(7)QZhx_hv5(|@p0u^zr39YEz3d7zlm0hT-UcxK{ZFxSojIqPTGyIaUmO}|7| zcqzbt6;*h^FAd`Dnn}udXgm?+WNs9t6$+59sKmoJWjxT4KymaKslA676P!@fXd zoiy&=R!1GO`i1zd0csr{!tM~gKys@=boBQLQaF4#9+xq&4L1qK8;U}Xuq=u`S|Cl8 zyk9~2n=9z|rh=KKRRZ%LKta(no ztS>Q{C-%_!r#PBEE`gcRnN8{r)=*OdluePN6}>6+MZr^QXwgQFxUVO(tH+XpktO*-)!pKTDX$YJ8BxP9@cHnHd{a43IK@@4D)A2~sP-b~Mtx%1 zwp(*iDNh(1=>+%^X9o8p4g&KlQuM*w1r(~%NlIJ?iT}!SJ>}csFaLwIs&c7j`EKUX~H+28mc7GLa7L5-yct! ze9l~SM)eNU`06-K*Xl>#lsLRPzm`iIyNWtgIK!XT=`a{%LbaNc$u~E7fK|cdU$hMv z82Ccq{X2|(fv^+4FB-1V9njUUMGYdKax-7Jl328ZfV(=>>SY;sX4P4NuEyG z%$&IHiubL$Y2v=y^vf0od&LE~Mt2|Hyp(~V$`NQn_v4ro2T9|LooI2U4nLcGrE^~f zSB}jYCo*yigd>tuVC9%N;;x>?Z7R`$uaZtMO7a98cybwj%+`a?i*J!JWy9fjr5rq3 zSC(x zw1Y12R--05vGlK*Iv17U%`HC`&xns5B_%_rpv=c&>Zs<%Jc^$IeUb_gu=yWhHI6X6 zkiYU9F&ad!Ct;iQcT#ZQoMV$y$VyH}kbL~2q3_J-(7rs8NuU%NGON#C@VD9@Y2oR0 zkuq*7u*SgLTh#4bDN$;fhFhB^!3Nw9JEj4{kLi+4lNXZljgI!`H(w?XwL<8Mm>tBpKbJVo8b@}x3%u_M zQZRi>6lZWy198eU&@mZ7;>UI{vX-mJon>8>Z_<64FIQgD+%B=mW$gvR1&+dJ%RbUW zcTSVE!f15s46>h_U&USiV1<94o~M!3S#;aULh?Lq8CEy_rfwG)$ecS4*M?nUD)Bi} z;TA{mg#q>2)J;armmm{ z`ERM@24yG>^&?Lo|Kt3cevl1~a`vi1&9+<+*u3u4qweFoXjO0#r?zA{YR$02nZFF_ zqtbkO&&&ju285IMBYo+PLl(k4QyoWbk0;EQZ}j&nS5%#$1ZT-pvR>dYRvUYBgNA`% z{vwZQ%Ft&XZwMd(g)Xq8xSUi4pS7Q<%0R09ap5yL6BxZ$p8CB!OFe|TN$yGrBsNN7 zZd47!xH`f$VK4hw*(#WM4?oH)BeSE=8ku$=>o*v-;OQ|)%*UMwTP|g-QH%}!KwIj&CxI^}GC=XVrE1AjP{Oq@!4kO)5W)Oo9 z-o*B69B0wFjd<4oVweN*MABp*b+`G-bm&;pGY51T3+;PM>CPCcpt_obHrrDFJlBk@hokw87@+e3a85!8`3?))v)gFTRJ;XjsztAp*OmB(#JVpX^7|(9ydr6%~NP~q6zmZFq$r{n}i=1`QX_(52$ROKXqKAM^_dHaBWAo&}D8R zqQ}us=waW{bm47NEa+ZK>h8*;{>W%%kxT$nCcl?Vm{mcU7JaInQ^xGB{6J0S%0j%~ zN^&~p7ghchOVszOk<^V($|2G=){W0dWX1;Oti(`e>4mXSX`4aL z*X5B$4+F@k9m1?OzCd&=WWY4^C<0S3fxSFbv$j3TxYrZ@?dr+1o?}o5T z{I|e;+cF5ZET{B99Nzo-6IBu<_}-CY_@ItNRM0z$OQPkWS79$)hF9=D_8Y{h*1?$} z*P%qMo;4kt_uN3(I_d#0!Jjvk6Zf_Ei{)M2D&I363Whd0iq(cpJ&lxu4vM#ExYTiF~~ zZ*diVzI21yHK|0^cqtU`+E4B`=%Ieb6f8S-5cS2gP`yqEwf1Gw=*y9~Xzp#svU4W* z6(oYl=O7s{hzHB=&qSeU8QimprG@w0=*V~1Xqf&bJOd`^5HX%kxIByLT%E(MTU^NO zJ+mGL8i(ScflzW~N+OY|K2F|9m_f}CJCJMDg4+_p9;ohfPV?*^DkHyuG_$hg`rQnw z?|Gey%-g8m`8bm0F$AOEDbZXz7R#)NWNLefd-idMTlDZllO< z7f+ISO#|koB*Rw?WAL^+YP()g__Z+O zYgnPCIz$7#w&R?7dyHOef%Reodhe)z1KO`hc9 z-UDD!WG7|>t_vn&?RzH{|?P9t( zWC3{Y_`@vZ1IgM?TVT@cT`&+DOM2%Hhxk7yiTQ_J^W)6dE5--F+eEM}zA<(Uhw{!qVt6Y!q>P3Eb2EBV}z zM1Kt?(c>c$$>Aqiq*=I+U4Bex$?l18zx4$ROc9B(elel{QHzYc06ews8gMX`1h-AzJ*nX*&M0-x+PTXlA_r)F%c5pxG$x(&u zbs8kA_AMD4CIha8BRKQs4UEZ04N@6@m3cO3MJ^5QBp*9Q(8SDg&aUwrH~Y?2x^>Yp zdURJZ-9etyraQ&VnDT#gOF}k%pefYp&lVwf)D#mZl+s9RTU;n@gz-71w8f*A`#oY; z<$(SI)cHFG^=9eOo*l}VrKEvpLe`H)uCq`aQSgv?A{O;y?^NJ!qPqu3hQk z1oCH{Hl$Uyb9T8C!FyO9Nqihm4&Co0ZQY$r=1pP$!{n^JxW14q&cANoyK^2$3g>v# zr3Rwr%fL6~^Q53Bk=%O!A5ppeort1DEuyzFe$h=i3N&t&Fc0kIxvubw^hJXyDYN**1qZ~_ zoy{f8j#=aFeLp^=myQ|Zg4coce^GQE{#3qi98VH55-KC35Rs&$IM02Zl+yT$N<&N1 z(vVcNBP5%aWTZtl+0Juahe~Bu(v%dHw6(Y2^ZN(9yd2MY?)x*|AK7OLxJq?4ZG#+C z-TsR0eRvA`oa@2R%Re#UYZgqqwVvvXJem5O6yZ6Er)4JBaG3HD;B{xygw7>2F6}it zv_uiaI%atHa1r`!%VX0ugbfm%k0`xQ)UUfkQtLhs;y(7t+*}XtzuO{|u zTM{)IhJ*XBKWwvAHkH0pgav#c(;dgN`HouXzWF%s>}?G@x0bTPw!KuCeGVeHLN?Q; zhHe~4hSDef*vOU7NxeM>_GF$#%hGs#^yvj00mtp+`;5g+oAHvsi> zYvBFkHhSywSkk>%Q*7<_hipe&VPkgG!-9(S^i1v|3;Ad$y`Eu-)$&l7|4)T^+FZcjl~X`{Ng36RI1Gy=msxb)RJ^+}9Q@vwkp3!tKC(xS zg8PjG4VwvgV$>-5cJ?!TuDFE%O+QYD=bwT#JG`-E^F9g)4TC|uhGYK^_582aYh2r- zaQX|J+?*7h0BAwJYQk^&rN~7SN1S~s5Vg)XMO-Qkcj^8;=i{GFY% z6TvZ|GpwC|7UGZJq0Jq?fcb`z*&2Pa>-!gUW5m+o2~liAwmZaa?=LlWegty;H1V!# z8M-U0v#~|LS?TU%m>E(_`d+9-pMqAFfV_WgbNnCI5~%z^tBP zI-&QOX+OC~)0L9RdG;_G7O;yY&6T0PyYrcT+6z#EF7&Wc01uTHP`084_i}g0r)rwm z)btWYe%&gqo1_KvwwS`k^>WgbuU1e~GzA_)Fx^?DA?P6r5c??s)ZdwsuH!CtI_xiw z8FY|mZyvFG&TiCFWr5~vk5e6|469QL@Z5_D6r1`PE2=`UX`LQU>gW*y>TSVR-elh%+>hPb62?FiG!JG(Fm6AA8h|jpai_oJHn2auLaW> z8_H~rVli1W;l`qU+-!CLG;Leh0qg5b#zBi!<+S3nfmgV?GsY~S%o-NdAIHl% zni9vAA`5Pgz){uwZ=)p8?rMc+3vjI~zao}W3BUyBPA@6E7mEGp^F+sAH zIr=YU*5j@*ZL=M`i_H`K)!@brN;ry#E*xQgMiWr#RlsRK=+A8S_MdAFtN;WZ2lAA{E zw^mn9_ErOp+5P$T-N#TQN`q;4KeM-U+R0bYQj&jv=g-Ff2JS@~<;xjJUzLA>*I9Xd z-TdP?wK@y0jnESE&M&#Lu1x3-nMX}SSK_%WS6C&wP9B$^!GISpuytFy=wn_XdzZGH z4#^k5A|DxPJm&_%8+O2$+F06dyq0@==r8M6lZ{?^S9tSZyV&aG^O@h4hjiDc70NW! zAu7pE>Kyt5uH7$Sely#-yzPhBQiHSPRvH4Ydw0OBMFo^I{XA_Pm&MAQ6JXGFJMw)g z@a9tu=|PhgML!uYUON9FDp+#Tx`^jsXgW#MXZT$Fab_&!eLV|}+UJe0%P=O>#^gF5-0e4^+WR9rcV10w=xg>dejcyA}`9NUI^E2OxY#!$)QK9tcmMChW~ za$8k~3KMiyoJZ1V)bL%2-@i>lKRqWHeOps>wa-m%3=3m7FVBS5`om!N_?x6)ai0Cf z+P*Lbt64|iZtmQzMu~n?G8;Mc7OERZFnz!M7+ITx+L0-ekwvlSvaON%dCg(1k?U!v znln`_PsE=f&o;&W#@er2aH(z`_scbo4cBaEir2!h|KX#!c18$unc0hPRVJ}&IX!R$ zMSf^G;WaHoLI2kS=ZnI+f{F>Z+^Rorxp|u1g;m0{cOOmPrSpNQPLR29IAk6rZr?vM z^fr|t%?H)&c<^zYIC?++nm>(3=w|bM6K}F3Wp5-mCXb_cc2nqj%oL%k^N`6KYEk+E zYf<;0ome{Z8N2AU#ooG&V@2X1w&#Z$_iukO>(O+QG#+rp+o?0aKKB`0GQb>$lnMQ^ zbs>^k1)hJdvWV>vo~u0bzF?8<0XC;wmn18M{G|myu-(`!Tw>Tc6W>~xZGi&ix#r`?mXoAJD`pcaD18;qB15$y@=h>BK2Qh!U`OB*wLtTbKRkA+Vo$5> zVA4Q&I3m{%>{o7sZ5?4Oc>O~B@GwHsV{b`cEaJ$$e=owwHm+}!&^>ywl>W(`rjl>r z#LErh%4e95LtYv>P|EEA23R)yykW4Ld3ZxdB zDcW#p59kTs9sk{33+L}mg_9H)DM9MoFK9UoBR5N9o$%FLX$z{Tolun0Q_nSbYi_xF6(Z|@@(JmduL z*6l`~I&wnSXBUM;*i+&lN19w2M6;&kv%qW>;7j9Rh0YSFiA{pTwd2`&htw z{s?6rZbTNH$T^+XhmVb;*{;Bm(7Z4PcLaA~p^#+`l4S1&nMNb-VcbpMDovA0(i*KQ&F5-mA-#Hb_ z32f-dkr?C?i<1>xXmj~0`o6`463ez@$+d47ZJx)*3px2M3+wogD;P6M)8wplDzNR@ z9?-Gx5xw%N#h`WL>EH8Rf__+trVEzgzH1k`Q(raUu-9ewti_OSr=NoW#ZDZe-;cf= zt)PJyOHh6JRUCHP2`-(V#Acjqun$O?B1HX8JT;MaU+Bgpr=7Sq zv;fo98kv*)axfBd8F$M2~D2wlb5^qgmwLVs^w(r9XW7sXW9 zeiTjGc@~|HH?wh34fynYsc(sTWNuOWWzw{8$<`xty{+YN+G! ze)@4i&~KC@u~B(GCJP*lAftG;t@<#G7#;)PCl&!`Tns}Gw!l`kM0os-hhvRj*yi!U z+<(jV()6HJbn)sS>Fmrbam#kv$%e38sc8 zp}X`I%=o(v%mc51TZuBbu015_{9A&HSCyhBBDq{zOhK1Js8Q%M#Eq1t!8DazCk5d2 zoyK_cuVCA%7_{wh^pWifwxk__U$?!a% zBX2DF@9;dFIq((E3@*esN?S2+$pU&e{Uy5j7~-HUR&?>j75sF>t2=7HdVXQ{l_O@h*N*@4x2kL z>_`*apSzEy7%NNje$~)nc|){VQvjRq7O}TqD_QxpGfXM7fk`)xhNYLsvVMxq)OKzg zX~Y%L=#Z6oF+@&0LSA0{YT70c?U=#_&5WWA=I`jl(P7dxa#@sKIUQb4G8Q*~*({Eh z&4K6Y*U0Cfm-NUf3#niapxg33VsWdscw2{(SUMmaTm&}ktpv0h6hlCR*r zOHX{XZiv|L;Re*a`Iu%;eL(~MYo}Rr`m;IzO5nnWv#`hS5|ru@Kcsausbxiy&bBi0 zl8s^?OHHBls=&l}zlW&`UVDx>5Y>jXpqE)Ie`}H=^P0Gj3pjWcb>~+ylu?JSD{I)3 z+edLR84(Wfr<*F**}1U-V=3}I`!T*9eda&LG*4@+c)Sz4E{*}WMayy1S&nP3e~Q~y z^kD_h^g!v94C`;NjHm4S(x;vXHokQ^EQci|JdBLkNEY+Aa0q?cV1dEk2_fdHS)8!9gwXz{)n+>A+zw626TM&I%risfX!^ojx4<&n z-_ug0$d(#yM^Vt5ek55IMD8a{>59<`I$Iw|xBs|Pr}hk*_^3fNbNEKQAwQay z9-GL%{g#0lSA6l}T4QGaR>axpS_m8A@7eyRaNlXdgsS9!b8 zpPfqBU2H%jgihG%gjqCnvMfDVrcLs{>ndZ6H{$PpuOT5ePUt%a(dZ+C#51k?v;EgT zQNlG{xPE^J*&2JW>DMMg#g~gXL;V$-dv}BA;?I}ttKljpiAdq!JE^nUX*M|FsRygA zawa$7JnfP8gDtjSKnlI-AidcEPLcQ6x)ZDU`47_hECp+Brg1pSi4i(t^Ph9F+Amqq z{lU~B^+ivNft8WIRPFBq){c4H?`I-d`nCiX{y0V3_}AQrXF`rMN`e;?CXwgQx1t{3 zF|evVnoX!n7kuFY|l@Ab!*F}I&+x`)gy)|t(&*PctTNq_WV35n zSnxx<<#~w3b~}Q5M~nT`TSeef?130I7&kYcfWY{_%K}QQ*z+Sk{!6S ze=yGbVj<+Gj?%{TQV9Aa54syC(OKbLHzD7QzhW~58@L{cSJ6y}Q|=WhupN4z0}s3ds5 zxLNd1?<$;)jiAPgV$zpiDemKGMknDFP2D>moDPKZgMR#@Ps@J8$<=FV{D?K!doT*t zn3+i3jYMMgIXu}-Hi5G;@1Z?KRqFks0o-RN(>gL>`_|2ej3zUw-b*8}a#0e_H(vy2 z*$rOb{fE+xv3#$+51bH6THAsvXwMfxFWK-7hld@eZN@VAy<;7XyS5Ut^22b?!!Xhb zOoqM#Eg2mTgp?sNU=e?tnpdeojNefn`c+X0*%t@$1)7e~*ZCNo=uQ+kGT%X7yoPchi6-7mhIu#6&}ora zyf-D7#ZH+iwe6D%-wSKFO$v=HVc0q-eP$}vKXMYr50LO%wdT>(6@wt`jv|Q+$1#W4 zz3}S8c$l_GLD=wUfoq3yz*4)8mcPmrx{YI5iciT=V{qPM5Nxz_ z2D#@oP-ra&>!KLDD;+^47e4UgcF9oh{BjgWZ>Qy!%Csf;7QWrth6lDd()5|z=w0z# z()Jck9xM}ygb(WhZ$tUGU$xrW_zKFWRvJjLbw0{U)NL*w!kDCdg1s0&Wt1nx2N{Na}5ZArQqe4A<))o~g_@cpd+9Qz2A%v{9PNr)YUgMJW zzO?JW9VT`d!&IU9nt9LGht~WhwvBU}H8^}SE(CZ)axfps9HbKis zXTiU?$rM9s+3II8OmUbkr#PSlKis@1;N)~+lH?#ICDSK4l3XaE$M7LM<=hV8R(LYch z688v zdIFdny2Zkej1~O*Pi(825}a6kj^!>L1M5C$g85(rX!w4PJqXWW^5==D3(@dfGR^&B{SFD;G*RolX_lo zUsevnjfr}wa43a0^DN}7COg37U^AvM(UBc()?#x?BUn_m3OyZrj$Iqo!nxiIt7m!l<;19u#tDr+>Gp2EWWw%2Y2^ zxz!5GLn=$@!&hyHxU>%Y(lhpZs{vMTu_uh12QG;oINJOYyW_lr(dOk%e_PQeQug{Yg;4~NSS z$L8x+oL+b>4&7h?W9R;dc;gDTMM(M5-a>x-+jBVUs39bCX} z6B+~t5Mt8Fc}72AUd3P7+n1?q)X)~DmOYB`rshy9dys7wf8gfA5{s+i__q{Ib3X+5Mk$fBLUy=Wd>3;m(s_@#b{wT9gKoUkM}ZOQ3+Q0e&Q-} zRJd|)Z*V>Ghy53JjA?mq;71(_U{ys*;CQxyBa~s}yn$Czn@0NDyUFy>U2X?^iI*NcLz7LW zAR0E1-rl@}hYQwAF23$d(JvD*VLp(|e@!gRBry8Q{k8%uqK zOZ?uV>CiOHd-w@oPfx|c%S@?dM3wMeJ(Q+iawdq=Chgho^kTqW_No1&NOL4d20BJ$ zR1<`^0@n!)GcTw=;zTa9mQd!7Gx(&h8_5T+LErRqtjBOYZqxI@OAn8t!;CMy>)DxX zfSo06R`bMZ{$_mC$S|Decz|M3g=}fc9gNu8fWa#-q3PYBEPQ7w+y2=E%qk|DI#jV#6 zWDo~qzvMuA*f!Dm{CSk#dV(4pI!QBEMw%NYboS;PBPHDnob{y}Y%O;I$`oG1w6!wg zq;nM@^Y#kF)CRKmt9MfW^S;DS{YH|r1$0}QLE{Tc$WeC!Nj|S)dgpF2v&}c*O!P-c zY3YQOUh!b%-~<~!jllBmB@}#PJryhP)P7+VX%vJ}yUiL>l+2`(CzDyU=1ErhIRYHU zMuD5l9*DmY3g>SG3L2Ou3`#pEx+(gAY1QFmzj7jVJo`k3LN}gGN~CkPPcc*ISs1<1 zg$?;USf#3qwTqO+&&(>p>thLAw7HG{ookN4y%bk~jth0%}?&)Me- z39vWhA4F#~gJY4rxV7;FoXOOJ3y0s97e%6@A5xgR^kV< z-)!}7VUJ<)WUx2b2gYT(0CNM_ikV92^{ERhFU`VYkqw3WWMMAtz~G)xUN<6=tvy%I z-+pSqei~R~r%?j`F0F=Xw{OEsa`)I{1tGQy=u>49lNJN*yXW& zrLqtHc6!FeKR$`^^2@m|dCSq}5~IzBud(=8d2m(ui`K*QQOoTLPV5{{pJoQ3zvwdT z+ER>T7A3IvZ)90|&l46DeB<%>>o!5%R zYhv*0tMhCef17X4&f{{_i`cju6G^q_4ayuc5^Zaj<8B_g$0uA{2j2w0ZHE%EprRu5 z959y!c^XkfvH=_@+=D--EXC=n0x!(9onNh7&C3nG&y?Z?7XHl`{-6I{&RNJx_Fp{! z?!`D#-gkdEY-vI(rF-Gcs~CJWKe94x>o-wOd^M|`pvQF1#EH^#AFvbitL<}EPo=j# zV$r|4Lu}RV3Fv8FfCH5*00!jYkv&>;uk{Xh``a=0P+-tCB;JF0=C@h#xE);AKOrCK zWJ{{&&QexKD+RdmbgZHV-y9C3>sMmg>R%#Qu+9P=zlwx61G8Xdi7z-8XF*olALdbV zg(dvA82+Rsv*+j3N%2VlMLSmE`5vI>JFb#NZ!YCrjUq3zO|((yp*Huugt9M=v9||; zn9bPZ(6VR*ynI;!8)p@R^TRm7FVF+~0tc9VSsmg^llcAP^(lAy0lKR-jj9FyuF}2n zWcW6jY!S%6@;aT62N9TQl-E6Qnx)Qp1i}F zQe0V|bVa^x-xP|=7toGL=tt*Ax)FO;I%8ihF#=>XIavKHjGIFy&e4TrRO`Piwjfca*^5Sb2 z>@7s5Cn|zr-3O*KD-YM)8&3Ugl*#(ULwuGyf_%@~kYk4)-M?l+uNI6Vv*NKh*5Ve+ zvNJHFGW-j5Dk(@Z*)^bjRx;-QRW-m$bV=!TTbZ zzjZ%E?lZ6a)_RQkBwA8BmyJJdSHZV{P_AFkG4vcG1Cw?8GSjM0qFwq^QEir(Wt~Xm zl%Ff$^^4=^*r-4}bku@;wKZTvm+7#x5Eyejf7;o@AUC$?LSz zN3&s4r+;55bo5^)`=6dTTz9D0MzjMarZ!MqlDTxPT7s}o@Q1BXJPJK8F2cNngJGoi zP)hw4MJZ+tG;m)pTUB;WV4+n*@efDvZW=@?@5|`ihYr%M9D`TxPKT88E@-jY0R3E? zXymRN^mF)e8sf4DeVxN#SZF9*nxg<-fqqQun~)(IA49W`yYrPfifpk%7hCA5%q6^* zhdQJGur$nrsWtYZjZ%~3(TN;1J8)RCtKb{|wZUGZEO(7{J&1=(TALx|@?0wAvgy9S zHq|pWrsd1T5UkV;OBYQ5`!A{_zh^%E74pnObtcf`x3l5c`fz4wR?AfTePx$2O-UJI zY2U{yyf{aV1+ z(&}TwIsNlLaNUn0T7FrTo>?u#k|`z>x<=^0z0XIPzZLupuTN~ks2T9ZubcfjV+%4a zwxDy#6q?uOuwe<7eC37N>`dEah&P-A{()N1U(&_9e>%wq+WS*b(`DQz;&HGb#?iYoPRv`ajiP39t;QdH^bmYN;;T6QG$mDZ!*=8fvl<5ok=RD z!lT3kl@omPkuSZ-@~dSaUw{ua3`&QV&;ecrN&ej5^*LlDNwf^u_KOKxT6k)@`Jl4fqLeDD^PC0N! z;;Yum_uh+V8{C)hPD9IZ!mLD6-m(@Q9OvSW)ID@~iX-v6H&CZl7dv)wJe!!Q#59*2 zWAV0XC_Sfzf7cnpel2}=AuR;Wqqg&uqymEnzF}XCYx&!MQ{aT0G8;1EDK|OfnIuZ5 z2ET1i<2B~H;p>_|eAmy@ICcCgY>Zz>Uuyfotim;P+gy`=SA6Efe(nQ{DSCKsh&IHx zWiZoMo?yvb_-XZ_Y>-Yg8?t#detdpi@-tGb z_WQigQC)PJ{2tBlI+iSStek4fVEFRKO!mWEC^|oq9S@(zK0EA%_gzsC9`l+lHVJ}+ z^SW&FBQZNT#0s~Bqr}7U3cBRX!76zkWk%^zm{lz4^&ZBwtQ0C%4<}k>Kv9a0bbD

UZE5{1E}^(RgeXY!kOn`uR-T2_+dt{84<;uFl;n}z==rD2~-s$?;KG#!@Jq^Er< zdBqv=q63a?Y*79Ih)OVGVUq@fhU+x=s5t>UZ2x0>6+_teVj*w->>nFmKAsEvVnIF^ zrFgzgg)9X>BPZ>Ur1bAiObyYY>4|9=xojBq9-fKwRFjzB$};Y6$84CZC1C}RjzCFl zC~S{RW5&~mg4(Gy!e+@fcr$mngjb&qdjC3+%m0oqPWGn@GAUR$@fkW9s#09gHF6j> zjc&{fqcfrAv_xYOO-YPklYfi_#ilxTBG3V5j7f&%F(Gi#ss##<9D}}ptl?T*F3ji~ z13e1|uvG5yF(=bs<1Ix?CadQlhk!V^QeJprH zf0c-;qpy>5J4dxvQB-sAAno4sMwH#vW`EU6%HFu`g)EhGpz-{y;Nzf>__Twu8Hun= z<$X2PG^5~i5~}MhLM$D^&0s{5sM(xd@naw&6_Q{AI+u(3*Fe`V$P}5 zleO&G%f_^>1;;tvj5oO@Nmd<>8+*5MlN_G&`JFMi|GX(yZTH79-tv<2o&9miSzkIF zupL`to$j$ccV!W~v|A3=`~PKW-8N9O#uCa;J!CUH`anNp zH6XqJYZJra)StcVaNH}lz$F(ChqiHNi?6bEiXH+h^$yx_F?eYDM(X8k=%wJ`oeF5- z7tiR$H6j-(@o}doE+eUA^ezhB*@b)jL-Dw`8AVV3ipNZ{aMYN?k^w4TaPI|M=3$@% z`;Wch^+ap%==x~pSM4ZtE`~zRPz_jkx|r#<+k#+YG1<66rr@gufiK+v_cgOe=Pg06 zt`AI#p2)w-E9IXyOlL(i=d!W87o*t&L(H@=XAeG%#TtRlZJis!X5LL_ow5mN`zIQk zzQy8HTOCsKPQttgYjL)y8#%Q!Opb8IRHe^+?FUodd5ajcGIO~j_IeWIFgfaYu@Voh z;n?4=PVD}>L7bn#YktUGGk%uYYS@s_hKs+?gLU2Uw7TFAMkGYRPo3qII4_8F%C2Kd zMlYPb){9psnlAanR!1A$GXfy2&ed>1)3T`Gz)a3`#sO#U^^K?UOd~=QA z=t9=66@o2412M3q5l^*y;G5ejP*RvoevvO(_>WRn*JVS->>%zx83K)w0i-|m47q(u zW*U;qWXk*@bBe(KmfMcAkM*Fza$o2tR+7k-3L2klCb-7dGSTTzBdfzcD8v zN@ogL*?#mc2Zi9Q9hfDW61G`d1pphjb4IP`pGR2oUWAhNom0eDj)0IJ5 zn?m1WXT#ZnKUn;sVj4RAI7k}avdGCi+dJwYRL);PGMfFs_D3<7;u#sKN%f@Qp4IG) zQ4G{%@1XD2#^~qa4>uY$sZ{A4#;5lYUtFwAV~i^)d%{z`=yU+Qm=XxTzUku1;7rI3 zYK8Z!?MTjAgY-Qn!Go;VT+2*TxKVx(zY47Kdact!HblbT@MFMwY{=_l1vn-jWo8E! z()z1|01Y*8iH#ja{0gU0sm|2L)q%FwEkd`(Kdf!^QQYKS0AEiIg>lC<*+_?2jGt=^ z`AOBdwp0P;Pv6O0Ux!g*EQ3?B!LWbmWHkLx6S9KLA-QideH!qZtK1(1@Kc?q>>xh97E$Xbp+N#G2C(ylim$A?$duq`Ahnf zVZx_F;OjOT{RfYto@I?7jLBoiNpHe|pMcwEWpv|e9-NFyMdRg{ZU`cIjS^jbv$|~GJiTSbc`R;gBPV>RF z%Z)K1eS*p zZkO?cm=1k>6am4vjUbBa%a`?i%<1p!U>dKcz+>|^QQpl~?(nc}YDW>;@d)Q$-f>|V%~5kv^L&}nH@V&*I^`ADd)yHrjMlfo-CBRV~d|ZrSpZ(E&Kt06+CVj zCiHjYVV6ZHE^Fgp)|5}^4%#qK`zcG@u!3t|lFKH&d&e|BhpOavGTDKJ^8n>uu(q??ApYAE?rHB|`+}_` zd9)#uEn7MpZ%j?YwMB)T=!Y(rJ}l%zY+qw!zB;_z+Q5%4KOx$jD36+h+nC#fX>{jw z6U(st&h+YsQSsL$v^YeK6wIvA>FIGSbW;L!QQ==#U&Z1-^>}LfJzN)*hFakh>EZWv zc%yy@X8G8$sI5yW_{}Zu^m=XTiu;cSXj%#WQM;tAFq1z#ypEj{HU)--9Hizerf~Jw zTd3a^2|p9u=~2*V=^j=`j-L-hxxp6@Yd6EzOew3?x=j1^c>;k$8}Gy-?)NzkFdvXE^H;m%1IZt zYf;O~qtK|>Pkd-a4V){Rg`I02k&<0MsYan1uQTZq{HX5_1sPH7ZOT_kXt)cmnP|;^ zyFcJ7oz9Y|Mv?aI4}%rEYoPdY0r<>WN89eT(1D!;q_fk<^GA+c0_9{UT9unY z7$;!k8}+5LyGKe@yQ1h|f|+>U{ypNPhnb)x=IPO_gVIYr^3pux4KT&mK|FQwII$#P z1zu4XvSf>lq;I>HQ$H_n)_gq+6eit)i@Oux<=1;S!ulOeRsTp3t40&v*TLUArsAFD zTOn*=4(Y0Vq1cW<@|(H^M(T$Fm+An+=lDn*SO38JKkI1r1sPtoA&~9S3}EMWJz+y; zrd3L&T9M6nWaiVavb(z?IiC$8w!SpTeg_*tRlCDb?`=D)6#C>pDxUIQdmQnzix!Vn zy71_SDR@sap>D@MG$l@h>ETO>-Fgq~9$3+JaTl)8N3n-7Y)XIbiy78}9aCri15zGKO z;8@w)?9-7cwDor{_WZua|CFp@@=M#;6H^Zy_38j!(>{r#uNSZrj+Z(9kt+Vx`XKNk z1U(8hS=I=lr@s;?vPjCZDo??+w9mZN>kTx$XEJzN*0U3nim|twgXWu0*xIq$7_Z_F zR4NA(KFTxm2Yu<@xL|rb=L8n|*0G8u+01v1uH>*)6&B39hQW$^=)iVCgJ0K<&S`pB z_<98$td7B(hrcns&kGb!w@Fq^;hU& z!kH-hnp@}KRI?d#wd&yuT)c3wVGWy8G=uvYtxGL?%_!gc0R7#hMGHersn4pL=vSUf zTZQeB)tzIo@LoFhZ;0oYdmh84(z|Td^w;e5X;qlsQOB;paCkge4Dv=F*zF%d5N`So z*N04k-Q7P~PkabfPCP;8(HRt;myIT&0(0zx9ZXI`NO}1SYLi3Z)`(ZkalAFZWm`E_ zUv8vr5Bf?A<|NaW6)JSrR}~(pRDq9^qIluz3aGHsg7ry%SwX=pI#4{B{wsMvQ@>uo zCyP(fZG)rSSXo{Bwz&<7Kz4ad&4DHz!`oh0R+$T!JR z+=$uSQPl!IW5HPPh)ZO@P9I`@Ld5LudIzqeX*Byia}c#`S;rlDJb@xgoAIX5d-giV zi{xYlt*M_jOf2fh%{nCH7ynw~!tENA==lS?F1^C}dinVEv~bt(a{-Hi+n_^nJ}ja@ zuzKb~O&PCo&1VUPwqGQf#tXD+jtrgEf5_Djy#aSN6vDI#Rlv6=L)Q9C*4nn6W(hu# z_Ut5@_CAT~T`o|Wvxr$)Msa6GB>{&s;re?u82o)1Hk6&it)Fk=W~T+Ha&9h5SIz?) zb8GNfJr0%!DbcA@I_qIz74Qb(}~=Fd*F(z5$spVGS=|p50kAv#I?Jtac9rWW}EaAa9DFW zpK5ZM-MLZ0iY$ap7jIe03(;YE!hI?(f11Q{t~u7mg!9SQ4Y)o_OVDdk9Y*TpOD3op z^A$RIoMOpHd@)&{qBAtnX!LOrS8#|q?0bu2jAQut{9&BEhbh~Cbpl~sBwm`e1=6k0 zvyH!(iHaSh_U0bD-}9abN{t$z;^9Fn3p>SyVgJEuU&AZ3+LC9Qc|EqV?iGn z)R`sEo}X$`iA zAZs0PX%seyj`o2?!tb-@8PDwP1YWM-GiiEl!C#fmT&uPSE`J8{R=>(`{In2t_q3yv z$dV#oTH=(>YFzbu1fBadi-w&t#H@YYtY2#!{@JBR0Y7!=a@SGR+&C4D7kYDd>%75l zuQe?4%;$4sg!k<6dDxhr!CY(?f&1WSSp952nCNZir%e@}hqW*HEh-G$Vo%|S+0phE zCGVMZk~wI{8grwii^(nU8ZH$wTuY-bGLx}|s8X#>Va~7d(VHr~8ET22&(yK%m-As( z^KkZWX#-neo5e<$G~t*wTY*Uzj$id%sG{Yg{l=w#MY&uDEBq+~N3#?l^p`QDoJwWd zOEXyES#uf@pFt}wbdl(7Fm)|n#u5aEuIA+d;`T?H;H z&0Ku0E&#S}{zNUVE2Q7HEtJl-dq)~s=IF6!y7=pRPjSR(UGaz?EpR=jkaEY2lR9== zN(VdWNcEf@aH+6e(zH)b9PnqbI8HeXuxu~7$1JAa8Bb`O&tpm!dCD;U50F z1X`Z;h4nA9S%Pf_?_e$TZU-CD@b*9<6Hcd|M7iOXT2H)qbdQ_GSUJn;9MrIas`yWFI*sSOT#)}t{M?a?FS}|Q*yG9>RpWxWUyBEVT zZjOCgbuTN$q|xIXFTG!1)Ap}&B~?t^P6hKM;GpRYXuf|OoJZaf+tCOnX8BTFmZX4= z9LG+4+B$p@s85UkG!rWi1-Lr-AQ+yF;yPfxWZm?s(7q;)YAP0E)%GM@D0B?>Ma{*@ zKGsCxsT`c*H^P5w-Qk#!0E)Ucz$TA})Z#)h9+w>_u-^Tf^Vx7b@jn^#!POE-H5-EY z!IOB>v%Z7>0R|Ee#iQ0cF+oDVmO!!ZE4ssWRL|BT&7Z#lGpfsD?}+o<_dbKd7&YQB zpO3%7#RM*$Is&OHvVwv`S!fl;<;=1Sc~ihFjUAZezcRLeln z_xCaS>^hD^+*x%x-3n^kf5Dqh1759U7g+ZUF$QaM(OL2f-m23TygxNw(CuA8Q?K;f z`u}(f+w7%yABSamE1zA5PePi+VAm-OSLs4Y7{#L03>0rVi=kP2h>^}|NL~3H-iHbC z{MPY*rhON1q;Jd{LKA zpU0SjjtZ9*RVgB4SVGn1&(bBY?Wy!t58Pz;mhS$#50Ch)B}acx;y>heCZT#y!Qfjk z^qgYhKM7N2T-|io*Q-F{cb`Q|jY;@S%@N<&gkil-5B0ux9F+uR*xGv-F9eyBuLgU_ z<6HL7`DYsJ8Et_&xdgbuyax?(4Ki^yrMm~36C;gAkK;_z75Yv+k*Q#!U^S)|0vPBIlCP{%vavHuzs)&4+U{~b3K;pS8hiufhuh|i{#U(eDXr>B-D?j08fhJp%&MH=!t4j z{}vAW?ej>7R}UH1T7h9jY4qLYNX%+|jfd(IFi<27A618A)e>cFnzf8Jy*a_`e>g(? zhPiXav0IJl6&U!^0B5=EWnteDcogUl>2_tdBlV^zU%3iR4`rZ#v@v%_xQc--r*To- z6MDS(8ZB##q|MSh;kf-Z^5B;PoN^O|FP{0J6PLk!J2^36gw7#{ZP$&Mlv4WayIJ zxe?@ODtFhNag|Dyr?bIQ7r;g5JXvr?fI3ypxW}cF+N~IX;tC_4VUHNDdL$yy`0x@Z z1S!Ev&rVR6F@@~BhveD23A9kI0RM>9W7OOg7${Z?-!8lH@)yh{=ck(p(l*-)KHbq` zWWR3Uxjxn7%~!R+A@TJBg`r`5VR;wUwW;xxbN&+j>R^0aV~bbx#?cy)es){#5i+(R zls!d{K=H0|SU!}A?zUHmbG{mQ>7HO3az#k5s|gMV%*4UG6u48tLePdQd}Vfmz~AjP zZdq`P^!?W7?K&vRd%xZUcRS4xoQWAH7?lVj(q5duyg-5XWyB3m4T}ks?`jHco?WGJ zsR%1}eggO8&!j2kDqXRs8vBgTU`p*rCjDz992R!~RU0moSdhkAR5suv8j3}F!O);7 z!h5Oe4NBMVqs-fx0*_Ba^uPa9dG#6Eyi-%(uq*y77L=_J5|~zK(9eBJybX`u!1Nq5IAc)xWM zeeITsGZ%@X!ah&<5qu6V`|B_(WJ6H*eGD#l4`#)Wu+VT>h?nIpPRs;GDDjWtbO}p= zy?Gn{Hd_KRhYfgMT$p{CEBCeAh&WiQ3KEuW7yKNZCz$x83h#dU4R54@S0ro9)Bh^Q z6Y7;<9hW}F4bC7KS2$fTKV1xmIFH-23BN(@`WGy&@RZXGv*f5-|f`xG9zg!^+cO>FaT5rWDWZvMce<>u%gJ zjV1HvZGvSky_|FA5p-p7{6?pG`tn9DuI69Cs`zCX-8PO~-+vvfZm)+IkkwaGY~k#&cr z+`UdZw4Gtl@){BO5J4uG$I=zmhq1nF3C{UG7t?Okf^w=Y&mwU>Sf0I$P0_0a)(H~? zlaR8o^UNg$4(hIglf}bKU=*SvvI`>Ws z&HlI@%w9W!3YTAebvqg_3Hf6;qs{YCT0=e_krmXb=rix!uYuHt_xQNzHWbd>N>f6X z(NkYr=r6${xck=#N@K(XEx~d;^OyUuuz4XmoVMYGg!u|&Tu&0g&tiDI`vp6@d;|Sc zl8mi&ZN#l87w&ow;>kZ=T(0#LB>(l}cB?*|%e4{b|0sw4{#>%gXD@^**I~NBD2BBP z@jT2cVBMECyzx1qditeY(kr@*{g@dH8fQz;**FC!)oSoQn>n$iIj&fH`#1C8mJ50@ znXq#Tw}TXUg7fo&K_&Pfje71z=e5j%0@IafHLsXD#@O==**Sv1!f@QPG#6emFF^DB zJ@%>cc!9E}C%C5vVB-Qk=<@|$n^uJ2{*)Z1?G z9cLzQFrkX~uF_>K-)Q8&hj4bj9)6Ya!@{5Pa5Sz5uZKV9ylZYSHfV`XGu!!2>ap~~ z%I!<7D*pF`}*ev%&_jxci^MwxK|m(bAj4b!>!4%_uEiyZjA8~E!umXV=7`KlIx z?GaNj(L|2EV?Ntzo4uiH=0s9aks9`fe;mBEiy-Dkl|)iMjN;Eb%&&WU=_H|*@M-sN zHoI{;o!jt;>#=O6arXE5W#d!Gw`sZoMG4{EZRO(8ND&3=uue z&HL^0Aao&TPl^4(R{T4H!Ape6D}8JHutgqb8JN+f2e!bL%Xj&YFYF;tpB=}N3Vlf7 z_yt~)aqMgj20l*OOx4E^k*D)6vyY~E+e(x?#4DKw{Qo-l*z&h60l6es{{BC9G@{D^ zHwNj0mg992bNeI-9BZOSr5cE1zZ@y$Sox3t>>}%owP-}9J(@Oa^Q}Je$j+Q)FlTWQ z>C;-t9P!X)FJ4aKGAiNJr)wiPeMurw*9W=3v-vo0`aM!QZ3Dk*T`J#jOD5SlB#H*L zNqFz{YKTdgiIuL_gmsUgZ?jrRe1i_e{23wdH~E2^nJiiB(8jvo*-1R(2WVuqifyxK zFY_|?A{$j~06YAIV9o;`drAHkJ9~UESY193>pAyW?}Qq5x^F%%8dD(ZmiZw1<|?~a zDg>tqU&Xhc!}#&aFWSERJF`>On>Y<;Gji5SFv{;G?+*;qCiOd%-%&y{Bs6W!-<=|P z(~C*{yz@kR)D&yBWl(P&AJ9CsoF0yofw>uJT;Hydxpsao)g3!bDrFy&KR@coE-Oje zsda`PToi9}e6}h^yqgcdn*NY82`fpnwI;stkH$wD?nJ5XFnCyt!a6i2t+_X;wvIZ9 zc>aJ?mb|9rwW8pf6V7_($)f3*COY5oIq@tL1wW-)I%)eH{MVmCB0s6%kas1&Nqq~m z-b)H!Dw~1mxg>I|)xfsfs*M_5Dq^gimyzVy59G`VO`H3v9BZd&gj^6z;&xfbXoqn* zy?u>iRxXYuuQzc_-Qh=!?n+;LNea<_u`{-uHinjm!(_n$SyXV_f}iZp;84XJbV$>I z>5kt>`;`!uzt0;CjE9KDz!xIDzz^&VlwhFO63ptOskZYqRx9)+JE5N?gV#>7^Iic-?q%|NM=2lmdlLs6?D5)A2 z+s^^BW@D&4XQz#aZI25ecnKC%1mR_JX@)a{Sq2J?*g4Xm_&5f`_iJ7htTZYTXHht5EJ60k6%+R zF`JJMk<_p`7;b4qHM8T%UO^)jh%KhcK4Z+m71>OYKoMO=9xTQ8d; zZ~i9eO0p&T<%ak#v4%PNK7r_*tS8YsIM>gMeK?+D8m2~C6RWg1>bl(yg_fzKZJ#lg zX)GWM)QsqQld0&bFld_^<3rAGPr$J2&1}liUnK9#N80IAOq@3wVo|3wbSJ7|`m$(J z|D&E5hUn30pA6uLZyV7{PJpgy8YJfCMP`N3E_h(`f@s$!^B0V>fJA30NRN6-jO$-8 z=d;Wp$lC@Mn}w1ONg0&CP6W>Li46ho=IL|1;1m2*l^i3#R^_l-s zx|7!Y{YD?NNyv9zLj!!?)47Jycq@4`O83XZ_hx4h8tNe{^txGLcbvX}eAmxr6~9iV3%14K&&*b`<4!AWI{Ak# z*W5x)S8@#VvjMo;E1X>weTWvaOSo>J995g#NmYAIahInPzH*#pdtJy9UrxD3hPO|k z5|ImW)T@$Gt3#|re<@uub0UoF^kXGk$86p}(iO#S`Wsq!H`;`eBF5=#Kdq~1i9*M6=B(WR%h}vaUNUw>bb*IH3t@1syh`ZBS z^IVBp{7JHX`DwCr-WfXTZa|EV&&By+H%V5H7)Wx?wZkhpDDc5w^wX#3WMpX{J!pK8 z4j62pc0H=J#@_~4mnmcBmJpi7w9vXPZ5nmNig~o1hdl)ybi$c+#C(M&T|*Svg=$VT z_;@$-{-F-NC@@B6xhW8^JDTpUQHM~@AxajC(y}|Vm?d+hVXbsN-Ff;F`PHz8Ra{<4 zo@9Ms4ZkLlidY9CZt#p=uwP4tW~zY86>f(u(Bk|F@{rY5z}InDM-z9((4T)|h+Li} zo7!B-DEt#8Tm2@JvY|{ee{v#8a5p1oE1uK-KO$&E5@`M^6ZmHFj%FtI&_&x-ar^^K znCoJK4+-ZVRait9K2w34#Zy3Evy1pYor}Y>)=(F(7#QC6n#>hi#GTt9oVPif5&57G zR`K!RTH4L#v@XLCwU0FE%6KpgI7^>DxruAG^-%wfIdq79Y+H5mH#3%(N{qN!$SK&t z)D0FvcySluXrK^b)<*Y#UQgCdeoVs5F0hTjp!e|-{{2ZxB-nHt!lO$#k~otkf7$Pq(dV-YMSLV!W2WS&@;X<45NoS}Y9Ba0M{|4zX?TYQL)m+vtvW`Z|Xk*Re7i{vw=X8-) zISC7^Wfu-~P>0n%WVyHoNU*(f1I6a^I<2KH86;Wu}SC1ovX8L(?3R2$w%Sr z+1aJEPq&vDS^UoS&(D6=mFpwzTGLC{_Xl$vpmbXPbCelmj-oGr0}l47)5Yfg)SmOB zjNNEvHMuwU(%1?*+93%^zNNI%O@e$@{g2$=8A$e<{bT%I4KcHKdXps=_Q9d|&CF1A zC0V<}A9h&Hgx_*bd~X*n^Zh6jCi&>WoB{!xSfWGJ;5OMxcT)PLosRK-vJUM5^xWY% zvdbcx)_C1vt;eeAeGYs&ReLqH?XRfhIG!;CuGds0fnW^v8}0@Xz`7i zDA7E2SLhnJZ(xWYtVHpCjU`yxw$M|ZLF{7Raq#C(IWd!oA}52Dux<;b=Fx3rsGREv zy)wdEVbwHdPANRzYDp^t-00@I3hEI5p1gi=m2G>iN}8v7a~(ZO9+j!!Y(H+#{WyZw z6{ye;`V7bh$eD_kZ~JlyK%Rw`;g-%R|=}lkxAC%`{0oox0zANav6Mvc+rmzyD5JmHntAaPIU!#+12^v z)le&qeqlk3Ccr2iXgj}s8 zk)1x&!KRcAu3e1DQyfusku8?qRz%y?n`xNG7vdJ5K@X27kn~0mvPu#lRcC-~x^D|b z&&EOBb7U;J&h(d@V+9;e!L;-?9X_$45VAD| zmhIa@;{1^&U$@3-&ni&d>L2!}i3uvTvaw{vcgpX#A%EU9b336Dc(#=Rm75`ueQgVj z6CMwTbv_WeZT29fk`AG%Zp2e!KAlyoi_O`muutd$I^Md4DoS~{lH)z@lFwoApeMv$ zMQ|%>fc)Ba2%1$3`j>)0%}ziRzEn|{=DYawYd=PFkf@j>0Uni#!Y5L9Fs?0)WbSnY z=e=%lQAUKfuR@GxM`!R@qYBv7eiD9ObH8&SvsHuFVD3dM*fqx z7@)QqKmO(V8zK|f)baMLNNo_b%;H=S=NzCPH-j9{i7YSmg_A>EcQ>5j=l^rYPQTeq zy|5U1I%wd&$!EEYQ5C)ONEDZ6x072j65wOa<%t$b!vs|!vQhL6k!&g=3a?}7-oR!m z$$h_H!au2+p9k{4KVzT$si*Tj6xr0qa@KNiJh+!UC0dUQ$fDV5V6ZfbmWx`^JaaK7 zCc}x&D-B^P12?gq*A+12W-j&A)Ph?Yy?hV(hjgNp8Od-iBe$lvGtwMqM<5bKbH^9( zwXMQ!H>oPX<4x0`erO!1Hu>T2q?af!_6uD<4%26jGQ7UqCOo0nqP)#nn`qN=ErIlc zMS_(}o?~#02W-816MRod@b*P#flTBt@&Z`8|AYcw-1G@wo#@39j|7gPs}F~>27qjy zzzg~k3EnRl6jMmU+8@REc+d@vU!LI$mAz)ZTwDbAGfknI`wVx;c*}Z(G}0S+=V^LH z2W{)T%I>RD#RF#HxckE*M#Azlw~rL#S^l&I7kr2PJR`xmvzeGAv=0)esPR4pKZP?< zSOK3uBMsw7b_ZMK~V+l$VngkN$w`<%6*G zT?iHXAVt9<=I5#f$++E*c&}XAYHafAT+leCul!~1$lZrivx1J@xHTRj&41^Sa<`!4qn48 zss2=IhBvgnb%n)c0ybanD*U%619-EPh<0!b$Q7T)*XL%UQhOVdJfWMWt|($QugK?e zB0I@29)b-&=L@cM*J4q>G4JYzEF7KOk0-7jh23GgjMb77+M0fs`2PG1#*LaxVQn>g z%!G5B8#~eiqtPVd!2o2MSi^CjIQ(i`#y0es!@K??xJ6(}H?(Ub%60Txiq2UHQjQtjy`zLL#-vhX?W%zqPhPjdH?Mu$f@0Btc()r%SX4#+!7no zYU9ToUHJ#(13!?2dZf>%hOlq^%gD;RJ^X=x0NTeBk@xl$I`oPY&*w+kgo<1&HkBiC zu`c*Pgv%y`h2qcWhV117g|K74F}0hgk7wE_X&3oSzDqR|lgV@8=8Z)VzmRix*Y6`& z?4w}FECSUYIK#OxBfOJl3|>AS3A zimDDQAWpO5@Ms*z05v(md>vehH&v5}!@D%9c|{t|%Wyuu>C1_gj4N$X&7cSQkJ*o- zqhzm~6#b*44Z=;cvFCj(=FQ83oR$dGIFSY`CUK6uKntorJ^~bXG~xk~hv4#o%fJK} zV6q~|G_6sHRbtsSl~xG5;L5n?3@GAyqWx@+4e0ISaQHO7r$QcEI`v zih^mQ4PY?pgA*lQk-p>S!6wy2Ko6(EqK|wWSbvE)?JWTbzLFr`O&x}%jzZuQF?M2C zFqk}=hv#lghV*P>XNJCH1OtjXlhx><1 z!8eXilw0w{g=(~>?04J*bxbVmU*wBw(@+Hy;pgp_xb;{G zI9yStKg>7bA5mRi+S~+e;+T*3-?`w2oatcESqlY;jG$ra9Wp(HrJiyX=(zYa ziD)PXpTilbX&wXC(PydQo)rH22l?=hmxA7AE7 zeV1oLLr@0Y;zZ!aqNSL6@-OxO%jIvnr(^Je`JH({>WxA~WVt>@3u6$tCYjbyG*Z94hK^6m)df(y00J zw6|szjwUaHrdQoGV)P!ZN$w$U7x$7H*)=HNnnA}8yK~ts6I6Vq2KM)+fknw@+OoBn zXqFtHo6ZgJN9t};A5SmHy3o$Nk(FE>MJV!sfnmhd`AnaCqwB`CAu!lkEjk! zVY5r_(`_lzWXX*IB2!^c+gHn@ZDBsi66d_AM>R?6cssts)p0m)D3bbad`bM48DM!v zE)&VkO*}rH1=}gv=&wBy6Wvr{h1>^P5urvyHI1pxJW0x*>P-`!9mvp+YWDLnRYoXu zA-ZWE0I8Zy#4Y~>-VD-%#fGU=>0c@7oW&Buv4wC?A%q0biC|{Ahf&(~n8+E&&sFok zO;+}M!Yu3@3*pbz#%h&#Kg5=v9;L;b4q)Fi#$!!yXE`c%}+A~IL zXZhgtnOv?d!kgGV8zYtZVz4dGfIfaRksLTULVIO7hf%3L5sC{VcYe+TQ)e~SEX$0^ zFVVwdD|uWn{E;@ubdhPV!uVef-h>OETj(vtVvdWl244nq@3o4FMA=pY8e?y94l@RB zZq-LNO&@R1H$tBjpjwWJuy$t<)0Q1gdwv#Eg%4{{b5kgO^@(PTW^W*Cw06TL`QFPm948&fT@G$$dbH7QfDrNZpZR4z*q=9?y8`fEd?g9lnlzQ z0=eW&w!VBZRTwD4Zz^SUoAPPecW)AT;PQ%23@D|Ffegl(XrZ<53|co9!j|~^Ghs%T zXkxMy4ND7Qm+sCX@nLe*!8(dat}!OD<2Wbg)UUK9Vk5#Nd5lT*qw?iIYCgL#!Z7Z?M2;>hI}Bl_uL)8YYbM)JkSh#Gl+*UPiZSmC@f_ zhIr(c0yQjsPi}GVgS1L3xZNU6-ifXtUS4LnWyU)?)!80zc$(tDE@7N=#1xi|3xJaD zBA7FBmXtInlZ_qwk^eIm-#*pFXD@xZ978+V?EQ;yR6_W)M*~())CawQS-3~r3!mK_ zqBW@t@n!rscAeFBGPPk2z$0${Ud_!^W*9Pg>ffnzSu$0(LiTv147xiTl80kSb3P4`l|gyMoAleKGc@w8 z2ljg{BehbiVWvkq5c4R|=eotyP1+fY`UrF~sYTu069qqoj$q)D3nXcJ6FimU@s>ww z@_t?~gE{vdFrajTVC$Uaf^G>5K?l=-wa=2sxtAt9$H$v^e~J(AYGNATPu*IaY+xwp zj@u^aao;C!3VMiYOVy#lU6Hq=Z5OZ8b|KHa?FSUptf4k@B?Zp%R)XQNc>*IwOd#Wx z&qh@}2Nj8#y!SUI^S*VI!+pxlh+|D~C`(w7(Jd;7bc@C;_0Ke=E}e|qp8FMbY_5$ z%4{r>_b0LIX2IXu7G}r7K{D;HF@|nVp{M?nz_s@~=&86?@>Z5R=e0)U!Ww08lfQ}n z60;yve=@WiYG6R*17^diHgYA5hgEiiu;SMv`fA=i;y5xwmI^(j`n!LyK9|Sgq3$lq zp4o+)UUNN{CGIrwuQ&C*;Evilw;>N4V4g@WGrzfo?sNC#1$5RQ|g zI2+n_`eFHqFVdViazmmW^&KtnurCYm-xlHPk`d-*$T9M^J(zE!(gfpj6WC-K3)na$ z3F1u`*xkF*Am{TCm9XO3=KOUa(>niBho9M4x@$iMtk5CZ(wvWLixA3%9e{1CH^RiO zsW52wh7NSxfNcdmH2>HtcB9K)IMpwpKW~H+Hnx!rjD8`xw@OeYq?PPUPQnz4`&fT? z4aYmG2iN{^SpDTF+^FiI&ksIfnrmJ##<3Bs(1I>%6zxH3rPblsYd&-2ObdJafH{48 zFbTU=GSO?Wkao_Arnhq^;B^ea-GFquq9F?IPQm2?Hgrfx6ur%F)6-L=@xmNR&u*AU z_guL{`!c-A)ol{cbJ!H}{zQRKBm;H#&oO62%t3DKH1uXHA`2Q+;pqo+P|cqLvo2qS zg|AbYWd;ryJ?%QF9VsXB-6kBT#sW`u$m52QO%M|iPTe@3$-t3EOz~`UTD~?OpZ`u~ zz4~{PtbG@W)wWek!++eYCN-RjFSH_^*{j%X+Z8asYCB=0Zj#o5C&Xv+clOJ-vveWH z$Z*^}od^m1Y2>e|w3rcs<3FTHK#K=;zp6`hnq5FI-h=qP;AYy9KWME(APK#DiM}e7 zrb`8v=|kH{sO3d-|EqqovJ!#B;v&)zzjYLEh?9v&ZZP$|Qe=s5HFb+#K(l1tu}1a< zB#rq8iB9L~3$5Upss<2@cVs0i%5W&Wf`n!1XF!9hgYX7ivSbXfAtp z@(DV9;d!!^S3vWd_TsusMclZ?jjoEl%0}9W!DvtoQO-L8))PWWVmG&+96JfAzb}(q zr+Xx_q>VM;2rt=gl6d*ZbGjnxJ=JpCi<9lna~=V4JnPG`%LXUVi4nEr^80Aov0*>8 ztU1mW%X)zC$_wOdh%THv+(hd2rbBgKGT6V)fs@!k4oW4^sAXX^ZN>!}>d;A5?!TuW z_Jy` z6O2q2VIMvyB`SkytmE35Fs}6tS*v3SLxCaWTMXAhU!G6SJ0*j|g{S0D#s%p8_mN~S z_)AV+lAtx=#kN-K95B>tIjT0~;C7!PoTq;oGyNU$x6E4nUg?PfDG_{~wF{iPgNS~~ zIam|62Y72wK+liMaNx=%keIs}?&V~_FIiLKTlkWBJ@*lHjoyuAzb$Z8iW6GjZbZAz zDm?op3r{AT#?8u{i^1B0F7P&ho@6g5c_9pecAd~CSPM}v8^|t`6A%$UA4HxdlaZhj zI&aBMwsCPXo%exC@A+D^~|abx-k&DnjVjj}j>4_<*V{=w>DMI^yO( zs;JjmL+>uv!eKRS^m+4|#%9*i)wbKPmCFU_ad*B)^WM?MRhOvn+ZOWIN)px0<8bSx zf6RBwFVxR0nm+v9MjoG5;yz>5*x%gmc~JHY**dJY;5P)g|o;*i`xj4hW#Wmf_@Y35@q+IJH+JT!qhA=l}qapSmo*HhNLPX+Epzaj^& zK41@c4U$(%uJmIukLqqZMr$ud(`thRoaCiNS{IKa+jj53jKe;(>D(8(KD&?Z>eazX zcaeH$$l}lIuV}lGCGL3pnySy8i2r1N(!ZPeR8ey){byQE2QS~FdK)tNzfEVe9xf;7 zUe2Gq?8^x8+9-n7itA|8Tr-AUWDO+Zqiwr(AL$ZugQ;_~Z6C|M&j=?~BARf2V+mF~Yk=j#M{)hJBudo=c^C!XN??chWLW$*iDR%dkN%EI3(s72qa=A@(S*50lVp%f?%(& zFrVuPwEum;{#jQ6qK6*isVY%aIkuG4PT5VC_pgM{KX>6!${RYX+ngHsC%|*F0%ZP3 zW9`-$Fv;rWc6SRPUw#L5l;hY+aes+%Rtd%$odmlt)4|E_JT-izfgK7Cw9MOt%q%*D zpBz@ezrQY^)m24g`%JMSPtSI<`+uO^8c2q{vhaq^CJ21F8ZY{*!j?C>;KlKN&-*5W z#|41suxeBvu!2d7%VFq-u%P$KL2%qH%p0m0MK2v^TZQkOPxILnR5ibW646pHEuoTs zkc~zw_6`h+-e&FQHeu*{ZQdpMB6P1d$C_46UU6L__8gX=oog320IEot_z@%wCE6FdDm?{eyy14lG%E;LMsOSiMpcl%zr*D?{8 z8;pleS1EM7It{LI@1MZ*tK<;PWSph+ak7dmh###ZS>*+6Lbey)RaSx7%eRvg0TC#+ z@+4iRv;p)i^yssmb@6dSl0 z+s&%4_rNRN&*AqgG2Y1qg;?^|inz~WX+*UKO~`Me3g7f`(U1|d#YzViH-0ALXHSLl zEqCb$*9-7!i8y<4-~`#T7^-J@U&LBFF86R;6+RpqFNif(q(vVeLC5{&_+Bs({B)xr z=9mFS-wA;rz8tgi$rszoJp^~uM&nq=GtSN3kAu!4xHKjc0t1SO@T6MU%Q+YF4(@{n z+cYwLG>&-<4VW!*2(4wZ(d*uI6iZzWOFZ(aXC&vr*(bp}^DG4d#dB@<>A#@a&wrDO zGrH9MqKx3}c4@)&tP-{)Ns6ghmq4sGCWB#8Cd{F5 zRE}q4&qJEOA-8{b0jZXykf=YAO3OSX=L*yL-8u=ZxZ+=WcfGHzv8xEK*nJLN3LDu& z2|9F_K$7QW{+QORvZsgN`oMv#EnGjB3p5v-&@CKSD5%vQjMSpg&~5--EZ-A{ zFb7`0$1hqWQHV+PRU}|dG8P4#A$9U=Fz_D7ecdm@8_O%ipBCx@ZTT8_X;nowDYsAw z?I=jxnnzX!J!1+TGSS|I&n#Bp7NJYt!+5Py7>g1Ygnpg|3wPh5Ti%$EoUETj?CTo* zeQz(k&G9E!o_Ld?s4)B+KS~+7a1v#D6Bt=(9LKrITGY=oUUH`NMY|K3dw)C{9L)j! z!C|trKod*OMB%TrTf|jl1r~E%6nPI(JX$o7s%I2~*rFxq>G*}F>wl(34`jh7xR!Vv zH)96;R#Q>V=RNm`fPT8!Kr%jU1j4-+4~34gIH2;J#sXcuTd}X)OwwXssaLIN4Cuu!ucJUbUPYjU0zS;1(WFOfh6+uMo zol$(r989uVL3%!Q+n#gr0PJ2&OzM7ene64*Iz0r%e)Q7E-w~E>ngDMm>Y{gsHH<(1 znbqjwoIy>eK}zQsn`d(jEH#%fHtEmEZrwasU3`apw=*TufiKvA;djJKO$v@zn?m2N zUF5!cBr|1cC^=)uF}lllMEkc6#`pEuHJU6UP^qlF#QaGqutYRN;pa&Y5?Pc!3JNuCdMjbNnu2(Dia$A^EX3Zlj1u`!_>3)kG@~F>49c$h!KtbY2)g)$%y{>Jj`vc-$eH=L>yZim=aB+4em#fSr!$Ds{LOg$ zzcf^B3a8Brmy^?5fIKv-Ca0vE$bnW-IC`Uqopk;risbCW_&rhVz1C8g{p%*EU-(A; ziDlwmT_eHfOD%Mtn=DUzf(*}jq=HB+P!Tk)F%i^E{>DH5)}Hs&QikWYRSSuTv*5|P z7kHP+2agsyIR*MU;f9GrHP1_8eZG85VLnrb}`FQF2RjT;9 z3f_7t@j?xQpFQl%9|D%V!C z-W{hw_RoLRZQ?TM78Qmsu^U)Dp>i6(WD$5-U8FneP06bHHsHkdNtM^z;`;zgyf*2x zb;5rh4BL5;*3JwehmSm^M+3Lxl+mU1ZkQ=4lNm=lPiUcde=#v!5X1;N52H!5H!1%S zLgt*lNO%L%XxEcWi~Q5bh4UTsLuU%vey)qkNF64s1B&GJlMLHGg;_+h_Py=uLKl`6 zwKMv$T#rudGWp$bkKLon@j2aEscTOPsrxPqJG8&iH8V8We?IF-02gp6ocfxzvJ9dU z!yU}{T4_kzu!8lvdXW5y6{Q(L&l%COT1H8nrRSe!ka=>~$ZoH3Fn#O^HMl+zWB&6b z*Gn#E3z7&v6B3}4r$83D@LxE&o(k15|FXNHjN7!aD z3$j2?jXpSJhu2q`l630IDwOQPVy_{#bEuc&yx$@N+jH12VOqF&>?66Mev%65n$oAA zR)XX*J{!KXhMDN$243zJ?3w42;h&B&S*&&dJlcQL*K;y$4`1&jGv%tO9IKCB$9-T# zdnqhi$=%gVr=h6SWS-}=$pSn1IFR352ewxk!M6$ZaAo#Ksx&8-Y}436KqZrYp0NqO zX(nM?;REc+k>ovJGnv|4=Ggx~RDVZB5r{aiT(mG36efSJwo=bR85b33;B*0dcows#SV8-4^+Npo5|Hvko-K7q=d zllWWM4b;EtVUEZ-yzH&c>wfV8w`ekWZp5nLOBNCw@}!kDT^I5HQ00QJ7J!(1g%kQj!H5&i;t1pG4 z;;Aq;ZB&@ZEni`ieaxAPB(15w(O-zaL>l2jXddYogyY#nc_259QRGz=DQ-JKmsr;_ ziFwU%{H-2(s`;baR8ijE=m4~xc!0L0nZoi3vZ(Si4AymPk_ow&VTS$;th+Zrl+EPH zkK>&raQXoAVVw(JtNH~7Pm~ZpNkP%68q^r7B_Sj4!TZT`)cvYSEyRz3?O+Yb<`^q_ z8*h+Rk5q}g?^@LJ3`IQdMp8$Y(W4$7bo~Q=R%osZRQX$Bvt=R{d~abb+)6P!K@)~& zD9|0#CqR>e43qwCF~puR0zJPq5csfw+PsXUO7R?%VPrQGzn~63o=B&OISTmq^flUX z@Um^~k4z9&mLr>*qu}JQIP~^Tgtv|}A(VRq$mTvFND*MP}uc!b7gp-PU;;R;ykDJCPw0qUZu-S3m!8;x)EWcscIdpNXSa zG%zHq5m$U#iE~GTajmN^o^0=@eBnz(*@R;#h(?G zu#xF%c)%(Mwf3gq$iwFtUSEnLRh9TE0O_3TBA9n{K4fSvfZ4}wVdJtJARMs?)V%3cLO{4E*n?tCZqVd2RLWZ7u;a<3d?mTpwSzStMDTZKB{SfoJ~7y zY<>+z8~Z`DxCP4n)!~-J5@Iqth6Z--Mf25t=sCV0J3idU&l^3^%=856t{9^+yD~_c zP6;^wHxK&v-vh(7B~bd|5>z}`2lAJX5DglP_q4Ck_&`sb?0g309z>w`>eWcQZ{X&k zLh?s=COzku26^pVk9Bz(=x48kdMRgEYOF{`E(&4s(@m7#%AsR<3+U>Ff&U}!J%gfb zp0#l@NR}WVNkmYRB4Ky$ncXD_2%@MMFcSBT6$vH|CLrFTQysCf9mcDJv}p5UrkRmFyx+?e^=zLbYHk7 zw3hRcQB5z@N#w@fz9?G!B8L{H-{uVyV zGi>gn%hPE=*#=tbA)Biz=E0iQZOXCQtSwq!TO#H+a^?JfBTc`a!Q|@gd%$^fS0FM? z;?lRRVz@glb#NxWcc&k!w1}c2yEwBI4vWU>dWfS*Cug$fWO~cdY>uqSE1{wO1JT?b zNxE=sA!q8#PGL}Q8h52bDLvuX4$<7~F!39K8=bW6qG;;6ja-q$hzJ>+6vLsbIeF0~ zq6GiF!UuEBptxLs7>ln*AIiVQDZiM+wRtj3->^8uRrq{~d$sW({mn0hp4cnrboIj` zQ3?^`Sjx4z&5KM$m%W(uGQoQK{@SzL6~~s-**mT1n_osa<6&-`1I_WGkFIOz9Q%7* z)1B4y@3qzJi3NW|%Y+P0vs9co2k5o<{#%1Stf5cWPATFZvdgBY$TrYrJ$_ujRo{do zyjQfD%VW{Ij3TaHRVT+`wIZE5-JH87N}QLJ8z#!_U(C@yc15^bStQDjZV)D~Un2~= zvx2iPIF)<$g%@|Iu$=B~RS?zk9Jq5;oVW`ns*6VT7I8F2ySdYzCv!?hlj&2&C%L^D z(xNC86Rv8J6}NtmE9dh`Gj97d4X)w*xpbk1H}{O(D)>Gwg}!OwO|PC}$ayAtPW&bm z#~m5H$5AoQrGKzi~+O{aBN zr#fxWddJPnFs99Z5PExQ3tdSk(y`UKoS>r{xErVQxXHW1xu=UdIG0VO>4Y^yboQA> z+A*t!t@J`%m#f!vU zJ;MdeS6oX+HEz7k8PU#Dk@Q^e+w^*F4$X5O!|KOYa({M7JFU}x%$1T6h)RAxpl{}% zr&oz4abJdDbZwtJ*GMR?L3H^hd(DAh(btd5L~^&B=*gSc(9<5O(w0*bxPSAnaAo4E zII0Ko>7giN+K{gy3h55#s&AvXl*SuQrS=BS*}Mt#3GeB&@pfz4tAWee-RQ>sv(=vK zsgo(v@oW^mO)R4AMsLu;Vm!|QGs0zo4DMQ9p=eTSCVgf9N_uyb3b%Z0I+s%$$gNM= z%*ox5O6!K!al@c3*Bg0pyN$$otw4i*;r2$Pb7rGRG{u9yuq{)}skV&HUN28iTY8b+ zQb|OiqpEcA&LqyIjOCnf)=Riyajsm!Z5!_1E=!Ta;XT}~PCvw4OWEA2&wyKa{WHf? z)=Z4Ms24T`T8c7r_33l|)f~;398Sx#UXi==aiqm{H0E#{#29gj zpWnrPa5a~g>&4wyyM{Y?)l*Cfj<8FQi)%LbeC8b3VnHLLADo(}%Y=&W?}`fjENQkPSbx^+1ZGgfkV1(Dn^(O$0C_Ep>j$tRq^ zxPEpp|A=VV@(FY?Q>D%3tJ5D2Ho@;##I<0P!-VV`ZnUi6nJAE=+zp(4D=7Sgald~Jw zdglX<`{WJmf@!azpPL5l`_zLr&3-Se-Bdz9p1(wNu`gNF);ojV8tq9dbO8F#o-C1R zp$~m?gN|@uI9(Vb6~{GcOW$c5w+x0rm3I}WWhh@P1T*Bmb&u-+{_+H?Ay!V>JeJ96x zRinr;NxmlveV4Pp644t;pb%>Bg}4=@f~g7Ecq!Pk$)7 zm>(&+<+YYB>4>6(YaY?`nxSl2%?Lx!V>7t_ndi1JHInmGMPlUxi zXE;o`I*!w)OPqOKZQS*xJ)GFgXzrE^9qjey54k#C{#@h2=bTD0_aj#(gL7UdlB+u3 zg|o&*lg>*}6E!}@?1hrzdRbzu%eb97ZS!4;zM_Tckm=(0-Wv++YbA2D$DDdm$6Fuz z`Py!g!MXs>>gWLyHSmW$OV)-X?}Iq^y3TM^%P{9e&udOkUO8tSUx6D`k;=Je5y+uE zUvWwo%W*_`%eVvDo$P5Elj$)rejuu+RT#g9uzh52u`f(A;ygKR$ML`NT(ozamdGqw zn*Q1XIFCcpI5LY1M1In@gnqdbIk@qzh%X+;hC~+;oc2ld03U#!n5Iq^!^ zcVU3D$Xj0Fq$n;%qE$lTpSH+bT1G-bK|%cWfBkeMphVz~^&y-0Y@D?ra97|gdx~Yt z_~(!1;mVX)_3wU?|0tl}Cwp2_EbgE7Unv0oZv`kjJ4XjQ2U`zU#=lhH|1XJ5uxHxYv24X^%=+)D@gLgnV9&6%r5OL)_B%K_%x19ujs5>K{V$eJ z*)zr6@Zau#35gh|f5tvEwBmwz;!FI~{;RQP{y$^yz_fRmZ6{s>_W#|n|7XRB#X0{Y zKcno3-@lg6VA?wTOZord{ByKpQ1-L!{{8&>kL5FMnJlsK|GVqo(T-`ywEs8r??1QS zk@cVWzyAUG)IWy*FZREK?Q9mymSQsgjsN*i>=z5BY?+kZKaT%@*M7>DvUT`Z`_F&q zfBxwLJ39x)KQX|+@_+vW{{J5x9GUjw4fy|m|L;%pm;h>PD1_G+3d3EWkAFM?j-Nmw3GlMWGLV zQ&EYFQPxKb82b)^-u62n|8^5>zJ3-yte1jc1NTvP7TO^N{tZ^iiRs8H+7-W^z6_Ud zW25M=e$?bs3@lPi1T*%Jfh6f;;76h@u)pR)om_8+l`UlOQ)6vxS09CQq|UMSdAVYr z%eDC83?bI^m&e^)w-FwxHHWjWXM(Mbi{bU2 z1R&bo0_>voK(whc+;VS-Z{96Wo&A$ZnZ&68jb#Y9dbXfFYO*AGPzN52m`5h`SF)ZI zSz?Aq3(kZWLGc!Cc=VPOH0FB3j?sB=g{3=)J}C_}m*fk6cb=wl=?^ID}-6Ie| zW4z#MKE&r!A(3AN>;LFrrKM)bv(N!|UnzvwS%=_6?QxJ;GY$-YRe%ZHec(x@slZT0 z7W4}>!5)DCG`d=X1srL3{b4Z*{0MN3cOfvn?gP@*C!z|DG?l%NmogWgNSg(FOxA;0&W><6?I%@Hcmqgu*DWUbro|N#MyEh4H=N!fn(^h~bD|X4U4V{fo#b(m^P9JTKYBE(XXu84Ii4JW>_-V z_fiYKy$<1{FVo>Sy$fKY$5H&I=_j?<-xdb>D?>pQk807hM~zEUz#fh~xUH5Cg5*Cj zd)3W>g}yQge|{e`)>#8|=mpSk)Dq9(6X?0b2({SY8n~+d2jsnZN5$m(!UGpInVRQv zD8FzaO0${>-Ml}7CkAogi17x=hW zpd0%GU~#uDk=3_FDqMXqFTnu3G15h$Rh5*%-#K_n`Z!bn#$~3)dlA!Qcpd*~;%I~6 zl2*s}T~Q5r8VbxC%0GGZ@1J&ZJS%34Q@La5y~?_JH?#U|?Hv1j_DyEg%@xf0sD7s4 z(e(z=p+2T^QUO!t*BOT2A2zSsIlN)YWgC9k`)s~xT^2tG5Pp_^D_?)z6<&dUN&U4% zz`JJ^=x|zgHPig;dxlAL6(i?eEVK8eHgik1DX%p+rRGW5fx2(`Piq&H?PtgpeP_%X z@NtknoayNO)5=8ff0cXA^mf4jkU>pWhv~cPlgbmIoN2ON{D&Hpexb)|ThLs2Jc4*L|OXwKXJ;{(Q`?KL;Tyb4pLREyr;FJ4eFN#4M!5R>P-_H8k8qV)IKZp=4Z~n z#b+1mGsj9a7#r*T7=`QpI&O#@a){T+WxSlI#YlLe%G9rj=XY&kGI;m*@Lh^Z9dnj1 z<*x^ZOuWT|x7XxUeN?FqGcS(A*lS#8-O%Z82S^#Rtc ze3^5`js?ef{B{?6ewu_bv#jGUU?j3^D@krZD?2jWo<_nqBAdT#s>! zcYx1v+2T-n`~lN`d7_<;?n36*@*W;O^^S4aa2`K~I>6Uuui$?zR%dFweZqUWM6==1 zU1xsh#@GCwz16(ZV4Lb?t!rnU=w856xngWTF#Af~lFK_B?$;8rfqU1O z$^B!^OlLokn3RiLnNt#)2l01PK(e#GF2z$g+JKDgrOj2hqPCitx$*8K&*}>p{eL0(F zzxtQm=eHt9i)Tk0k^^tov%SQxa1s@aH3l?)?Gan%>Q$}|d!;3rGxo0KpL$ft*RhCW zeAnL1yPk4_XKW_Jzf)J@7-8jB|K;1T;CF}_IyP4pE!LaDpBCZEH`|Q{B`}4`G3owIIg?x%)h<)i=)ZXosKEb^BO#F z?5*uMJ6*szyMZsgW0FHvrWJ3=Ur9c&^XDf9eQpS9_vh;j_w(aJF4z6KUc(5GJksEi zpIMKtQS~;v=G8BdU(I|lHpZ`hXv~}Rcsg_OM&W#j6MO=yo#JboHnL^)J!7O1I(g3o0Xxx){~ zMaOigS(iVvb{fs4$kKY$cKs7l?K{pwc9Bp!3eX;jqbTXB6YOD^fyJvI z1L@;5U>Dz|&j0*M4XxZu^)sfyskLjU zN9J$IB1O}3NM+wf;=PzbN}lH7m#@_DzMDBLbX^;jr*xu}`XUB?`B(7!sS=#{je+$e z6Gj^8D!r8qo@XV8wuqJpv^iWZS{wbQ^$H8JC{dE~=aw$C65)I9ACxIG!am=8g0%D7Ofgu%4nCsKwgZ42h*jo+yJv_y- ztWAfOii<(sxksSiyW)Wffjq1W^@=yrwy zmoO6GCg2Nf67ulHRX4zi4hOWFa`hHi6@Cqy z%e|p`4FJAeU5kX)m(g!U3!F@QQF+P`>vcDQyql3A%_56>vcH}Bz*ZLU)5lmJ#yHR? z#|eKrP>nwH9$_6<*TEkG51^r%F%&0Yk;qMH*mq?iw2NfmvK?2^hUgwJ|4cugbMPYA z(>;N0?qY&RzHbJF$L-*7ia%xDdy5(!c!T#Geu>;J9>x!H<=`*VX|T2G6$pC%otS4$ zA)Q~=3M0L9h{8;3;p%yn!lBa)tm&kK{8h3@R@N&<3%{Q-rzgUg*B&M=WYOO->3{kv}G-gi?wiyH{C~h=q}a zx^WIQ_Vo)o_bLf8H)>$sN&%T9UqNbo4an_aNunO6M?kWH7!_ z9x|6@<9&T|vCE`%T)Z)#Jji}X9)0Y@Dwk%02M%NWm9cY4%}6HYUb`3j?W)1+I4IblL9Yr04$WeZm+l@Rv%*TB6$6%Tq8 zqJ4$GfC^U%BjFKz{>Ck;YKtyP-Ej{sTNy$jZyj{%DG#h+UnaNie#UP-*Fb~U*-)@h z9*nQT!2gyxVbx6_dkbH}{i~{}bq(g&>p&vjdjAO$GeHVRma7Xt&Tl6XZGl8@+8Xkw z#t~OcyFqmNQ^+f=+l1Poil;7-hg+q-gDsKYuy&w~@D{NU9%lPep^7rNe)$TJGAjWW z?}f1UST}z6upi$#dW)(sGl8sBKDw6`NA8(O2-jcn6{FQkf$pau*lsr;50!@D-YZW8 zLEtOMQ#cA$es#izvocWc-7mo|vmBHYxsv1^%qB^}mBe9PJz7025r(&|f=2aO)ah#` zz-GJ_>#hs|y0SsI;l&S@r+gwxP0Jw@uXdqxg$gX~?s()UqlV4f6#yM}270v3ARgO( zu-cv3c*Ns6+Gjcfn(}R+W7;$!@tzI^Vf|pqeQSJtYoFkQOg^=8mmi?7z2F(}ow3jE z$wKQ1hC-(+YC?x%Um+!bliIUa68BSg(EI&cgmXvZg-U^2gkw=5prrBwv^MdF!_!?z z{mdm~+q5}C@86%1HES_RuQa2$!>UM6D+13{ev5M-R|?`Mp2pNoIs6Fbk_^`*%+)o9 z+IN)U{P)+uBR75cGBS;tTKZ0~YVI2<-)$>$I^0KIjz^=EFGq046btD5B7~Y+U?iM! zB!?WNONgT7U)(%rKiN2LN@l(~icjBJP4*0&CIv}HaIN-v^sA~F7oMguNk0f$bLW5` z*Btqd**t3Zkqqka0w;Jx_Zv|E@Pkw?ScYG0Gk{rd87NuK0)Ag?1k;l9faTVOyr45l z;NnV%@xAFd(LI;))Lw+2Ts=VENt_@Hg9?D>@#$pg@Ivy=D4ATn^#`ZD_)a;j(PHd0 zct-tVW)LNvMZ{-97I~-n13jCYOKvK+;XQ3Uo(6n}FXpOYDPBKi^DvLPpYsG=e?LV~ zUUnKiuaQ8Hhtv2D;c?h5r;}QHfW}!1cH<|>mT(V90EX9wz+$6?Aa1rRE>SZk-#9hc z?uazLcQ%c>v(*7!u2>A4Y`@1Tk}PwoO16gjR2@LElx)H6-><=?!59|PI{`^egsiIi zZ-VcgX?WLgF1mA38(-Wm2Ok(M!w=^rqMBe z=`8$Qpah@8WAKdMFm(F19-TgBgC%~&p-!_H<}L43?0-`N&-$x_{zNsRJw~g@W3xz< zTi%T-Gml}94+bQ9f<9rF-4w?ht6`IYF?+(p5_O&t2~ScbAsMo)kG@xM<=k?7 zKI}Wb!F>uq#|c;0 zd|hxJJ~6rs2kRz7s_rn-?TEk|%M^&?6vV1rlPoyhwh654Ri%D8UdEn*r*PyMKYT7j zmuiVs1#Ze4D8KWQ1&*H2nfn!$sY7fX!H8!Wi-WT%>Cjl}-Ent(P$v)#w2lL7RZAc< zzYc8}`~x=IZeaX2%>rwCt>DT1JNMD)o~Gb&EnyR^^V}wH76ib##6p4_CDdsQh$uW?(|wPnB&Om>?mcOdsW7oJFABE?x89k)V+aM z8l9z@g1B&Mh&q)nEejt`Spf!ZRH$zwqrhHc9Q=}+2aKP*1SBsW*2(I?fb)f*A<-WN z?jHhz3;nFuGSgwx+E{2IEQQ;&-$Lr|7udPtCe_-vk=3hzlV9z8L{J>sgB&zgA@3p| zZ2D;pW|g1E@&&qRs`D@~RGo@U<~fix>euml(=v=?c>Fb0fVFprmcVx)67O{@2W}hl zfJ1d7T9qXY?(Vim2JCWjM(P*VkI2D(dn`%JjV)NSQ3mF2QAPEK4l|aRWuq>C2U5Fg z9xmFbj^%|H$fXTW@VU_8bajVoB7Rb&!H`61?`S!(#(-u38d z!)InPU!G-k>;(0K9;EvHmy^0-oRH3Q4} zn_|f=RpjP{>15iqLP`)OODg;)5xoslh=a*Z%nG%n7Jdmuq%fAG_Z>yUmM^J=1CAu= zTQ@$lei!=OaT$HFcSfZ*X0j%HmB&ZA?|~N z!s3d6c<1e2lD}>eQ3!a5?pt}_ZTdWLx%wpd=wE`{zN+D8QL7QH8;#PAe`LkE-4!gg zy-c3m+e36)wiDeaj!e_X($rdqMAltiH#Vfp@$2kJtZe-Nd&OM?)SH(euCS6Q=EM-R zb~pJs-hi~yP9Sz?0&Xn6h4Z6NV=;vXE;(;a-s#T3$71f|d+KH|T;nQO&@YPu+ow{V zK1ZlequHSBmk;4?lM(Dc^qk3CdXM>d*AjtUsV#5Sh!^p8E2NAzECY2Of#i4leqvp# zL9DX6sr~z#(F!#=9HnuM>al4=rp|F-q+x_S+%-b-(y!v^AC=4wpIKKBhM!@nY)Z~F^PSPpH7&AtJBX4T;j^H z?DB6o!hR|)vu?!q_nyM%J^5r&oi$E)~aZ^4nU(>+Xg?*<}f`Sabw@IoAued{xJhw{&sKq1!60{Df2S)odZv%%(Yhb0vod*BwHgh-&M5G*6v~mGv>lpTTtfApVY9qvPoP%XmQscC zdepPi%2f1gTay-CIX=*BvKLE_`%ycsY0yk_eoP4e{!6bC8&@iIwl5jjIn9Fh74j#=QTC$zLcF z1BBc%Tq;pYP3>x7zVtiKQj!^C8SHJM7VPdtyV|3umbgb~zQGs9;J|V`QPfUdbIql$ zvE?xOoI_n+YJvSs)L_6t8m@f<;A3-nSipNjwf*Qu`@L0JfkqkRgnS)-QyW9c%wVwR zw3xBI{Z_IIlsB=a2+uON=VUXh@}&TyMUFZbEKfy!3&6Lu-wL!p6bs7Ema)2}-H_@h zX;h*z%y(G15m4Ya)_ksyg8A_*=E6|W^HW|JQCdSLEaKw8Qzl@#w;p!$&KC^H%2Fjo z^31m%qQF__mx2~i46=8Bi^dCFh+9-2zN)WFx-D1X_N&48NO&S(%2we&_tU7?^=(wX ze-7CAW)f~qv?PxbKQZrH_);cre8HY0`+%Bb8f^M@7;XtT118>UK?O6mgC9R9;PIAh zv2R(6Q%w>mrQZh;uy7)puM6PDT_JGxO&MTT6AOGNT7h+UoTx*o-U84)BruCIL6zQ0 z{L|~1=uY_$#yI~l5XCxCD~mkv+JbU$3D#134`;KaI@750pK_@S-ZQCAi(C{jXDPbb z-w&G7bkMv{ast8N8rF8@RD#3oNU$W6tQ+XWl&K@`;-3JsUoQa3$E0xVs3bO1P6i+A zF>km~DhLfsFt!IZH{@TRGKLyPTehiG|8#?=&aY=#MPPQD4E0EHK|$1>j>vZa)} z8C2iaS>VX^VQP)13Xur+#=N;d@lIDQmXfO`YuEg8mWI;^Op;H5)fRhU(ViEuZ^r|8 zs%!{Wl}}@l)=(Cs!~*v?R3cuH5mgqE2twRsVBf7DEQa$zkPzey<)#vlJ2U|X_`F4) z(xsG@wg%|Lm>Lfwti=abqaUUB9Ln7F1dm>?pniP1gCv{ZQDKKCW1lltVD+-&f>%CU z1ur|Nvc4K>p&wsPqN8Uz(C%11rT%&|UR~NsoqqHLz4>yR<#_!CdR7t!+Ws5{nopC# z;WL^*+rJc9+9W9DknNgVs;10Kj91I=2R@WM}hxaye)wx5EL+?B(4RCXJ@ZuMFR0xKfo>OBF0lM?|0KR|S2m9O1 zAWs-MWJzBecov{7JjYNMF0`2}T;h(%r{ZFA|7g1#5&S#m1D;v32C?M2sF53**wiDA zHCIcE*k26BS((Xb;nD$ewf!@($b3S+d&iL8sBFSJv5ZWTRwAt)^_1iCGt`L9aol|Q zDEVEJNT%;{2jQ!AK+stukUvn&T3H6s&vYqDLhrd?{@4n1c9Mr+o8l9}ZTr2f{7wC= zDGi3gWsYXTGv}p+`cvgeNf@8hO!XzdFTNp%vyFwyA-73x(jvltn@M(fG05oM8N%96 zazeefkGTI=E9#8@OjR2OQrUy?xT7IOxYW3 zx>ZRS6g^Y;e9%egb)pPQp3^1*KV9N`aRaG3s7*R-?a7w|Tr2@qNyUcK_=a0E{uv$1 zDpgk@k8F!cSfHWs=yPLXn3=xtg5DQWr_LeQ&VR<`jV#=pJ%Y3Mc%Xg9s*%CxbyT!C zPrkt=j*PXYld0CX`IDoUq5Z}>)V5_4gf^A;$j_j1VsStd14Dbf*4Pg)Y@zUWp@(or zc^^^SUc-_twg*F}j6v%_9lqAwjdZKx@VtNpZBD^DsN~Hg4D=g`3u=k*=n3JS`=Z zR6K2C*%%z4Pp7ix zYZwvD=6%FR36M|n4_R?>+i+26C7OXcsg5I_)Z(kk_|*+(a#D2>Iq0MUy~jQAgSm28 zqcsk@Se&O+%@<=ZkW9T^f1U8Oj*%1Qb=YcxJo8DsgCJp;rViQI3g=yP6P{-S;n=BB zY%DvOc*m{8cV|lrD_K{_<0BPVVipuG?wlx$iA=!y}uJ0%gEwg8zZO> z1KD8dj4HhE*&fm=ABz7(jiO*t3GQqu#!HS^W7UJ={cf@%SQjQw^{dsPg)yg)e7FSp z+%HKWYN9%sT`OMIX2}sqm7D=9;PgOP<3Y_0LCh%1QU zS=HNsm#H59+T4uWb_T+>+90SUj!_xs%E|B_E-7?M#DC*`F{9E5Jo8GU4lXSgszG*F zEIymj3XT?MAT;fUV0pS69zVnbykGO6XHx=6x>HCRZjS)>npcRuKzyEhXN5dG#CwOC zim-ia3Hjb|89U|YQs-26BIYs*&s%&In?@}M<+9(wtD(8Xe|RC)a6TMsSY_a!_DW>) zy0^Gg%*Q{bXixfOJcvVe7N&F?aGP^5HQaFkUNH88B_F>6t-vi{OJ5gqiMAqj6KwDy z%?iA#CLb_Y$54ic?2)BH6seq=4~B;8aJa<@5YA!XK7(cW$^1Xqc}PeqTrnOs$ivtC zdzjmoC-74uDKKs75LixiVk^IN{A`X1&eXDmEoWu$sS|PdOvM8{)Syi5X*0md(p;cE zql&Z(USikh>EQ7q%zvu30?+iR5iISJ!ioR{T89GTQ#vtu_kDJ<^t=}D#+aPlY|Epvq-gvGrn9BPw7WQ;mOW48tl;pO7=}C zY?m6CJTr(YeJ+FV2AK1swCCX3zCzr0_oSeQ$HN;(f1!aLS>W!1(|GlbOf1lzf=*qV z0va|wW_g;~k=S%)(x_nw(GEjo zMrXTWHIChW@frq0Fu!fH=u5PkDW_+jUE%$w8$ zu20QG8zC{TXRF8lwcd0-ruFzhWb4iH~Te?+u_GA`6xc>X8jyMOejhHO$VMiw~}02{cxuz^Tn@ zLu!r&TR7VnXX`O zn-=wcw=HFM?>d^RRmZ9>wx#Z|=L;%IrSVItyJ*g#0cMXspblLPK*>5RR>QYMb{h574bU5=s%}M~lJHBY-^aK2FHMXGUeL z0iHh|ibIx$;rgN(>=B+mJ9W7#dtzA~<)_yNpYCD7Uw(Q(=DHQ!Gi*YxPv`_G)Nj0x~M1s`$v-KhSv8tJl?C1=A7z`9<6BVOmr(yO12 ze>v`ho@*?ioyBt4{Q%+HYd2t*3p$|GWe{~ce`o%+y2lFgxJA_(4+8hjDtuk^g0)C( z1E3b}64b6_v&NH+SQV*tj8}H!aDMuGVD^1F{4(z)I=`nEe}&tz#O@^Gcl;)nlvT!D zJsYeY@(Jl0A4k8h7J)E@!-7K-=aNgmM&ZsKSD?~JFT9hO0Bf$g;;s!7;E8V+skF*Y zWNhVtrgn0u6rnq|T{}cYr48UopU#oPrOCL%Z722CXMkMtVUg20r*L=q7*>>?KqfSM zQ3p+8Sayy60$VRxXy#xAJUUq=x0Zm@4_OnzBn}o=!6fW8E^M89 z6WKGxd4ARR`r-YkbQo883I_JfAjn!4+C`9tBhVSd-r6le7ONHJ-3m} z3OYkxTFoU(*Cc|}Yo*}t3>ET2BtFlq5#qwTko+8W;qD`<3CYhN;jmu4fk=zX<(B+2rPm9H#RXYxpX> z8pPE-#0zfk$9^-#_rQ74p#2`eVdc?id?w|XQP4nwVP zLfBzM!|{0uSXoCAhu0WlKi?|W@jJHg)Wk{{mLZM1HXo#HV&yZ$P|L`DQpI}cC#CMq|k9op@x^gfQdkO_#FouKjOF>}YG?G`O2j5Q$g2mar*e<#V z2DcZ%m)m2YTw*Y`=`5orF6M!O`?4_edOddDvzvt7(q?CxxxkI>Db&D@K4hoiKq(~^ z@yn_VLB`2?icUI-HumO$uAGY$r~4V@-k2bm>XC_lUt5C{4wMpG{We_X^aHu7CV-QJ z#>ADJ1l{Y80n7J$h`do1`Fc~AynV+)kCr{c^6$JcB|QoX3a`U8^R1}KgFOugUihH` zhdaEs!FzZ_Djq)&?~RXLr7(Lq0B>2^M7?)W0Kr!AAkKFtF%L)}!TNsq;^D1$?|?Vr zwM(P?Hx}^F^Bl5aFi8ylOlF_z<-z8qS@3hsM40$kiL~3hlhn9)viZeF{PWv!JlS7? ztQ$HF3*(+685eoTo!$bT&v}QsqI|$ugfDWC+0Xi#b{b4mpGFI5ENZ5(XP{ zvSH&)AyA070oOn;mf#zK`3((}qx5M!)SZDU`aS_Ip#dKCnS&(DJK#sk0B(qP$7woD z%IKmv$^N4}>NMTTYSv!{pXRAS`#H^6&FUu(79`>yi^Q>6$X=YdYAYoy(1YPCC18*E zzM>#B$CH)?maIs-U2U>D{M| z!>qpGw!B30x-^oonG|`q?H1VkDFP(+H(=hv60~YemQ;pBk&M^l$bG3jX6{r4 zW3qgZ`gbA;bMD3C<6l91l^ie*H36e@J-`L=z&~RzXKh%24Q$z&#Xl~Q3RaXLDAQFB z7oLoV)O!V_H{&QZcVsgP=`j_C$=eHAdsh&xjxv1o4z6YHAcB!*~UGM^WyHwJeXY z(0b*3I#9RSh`;fb0R)IeG>c=~KJczj39479!$7#w^-%~m^yo5K?EoO$y2sE-AE ziRMPQVe4ry{C+i+%`}F_1IZw$_9xKCW`woO8gKri2z{CXI5egK3bWYwfcY6g(ZnS5 zr7;3(s;e4JRInR$Nl}NDIpBD;WVO?JWH67Vj68ez-vnh_LKKwDNvDpXZ7D2BjQ7DrO4&j|2#wCUpyr zpD)E3k2~?R+oeG6_in-7dk+Opjae-93tBjXbrUgPU@E3m1Ku4s0Zl9ia@4n!TM#q;# zZ(@?Ly9K05T^6gp9Y&AJ3em&b+gRelASZm*W zV)=a&Xw5$aRGwE+F>yj{_n;L=y4}Lh4b#D#-867Lcnbf0kP6yQ*Z{K+mjO$O3yRk6 zqj(h;aiNbgxUg@QV7<*rK(~D4>txj-MRJ@g1Ae5wei^kz8Hv!xA%FQm3{y2>jD-P!eTx!Fn%y@U&}`6&zwI z7;Lg*t(=#NG7P>__+upQpd@jG!U?2zBvb6meaOc?eNb#}1T4pL!P}`nsBKRDD0IOJ zAXhPntFSs5k}O0Scl7ZCm1fpu;zuHRbBKwYHQ6{S2UV&v#P3#d#9-Pda#nS!aKgl; zLj41dyCTe8A43=GSn+mdRX_N+s~o%vqez*# zo}l$zwqRxIbHQ4NdaOSLSnK`MsXHT!AoeuFH7jLdd|(1FS8F7iTX*1lDrc#3>s5GW zu(CkMB?P2wQ9*L8`*EwA3OF&EEvP8#<~!LT^lXv=@clm&ooPIj-y6p55+b|oYe=M+ zdCobsQ_9k+kT&`yOQl5$W#1{X6e$!*B3b4+=Zr|&CD5{T*w9=86n9}D%lz^50y z=$>>567krS5$ld-e%Gk6YlZGW-|d-TvTKke2#e7Re|1s+7b%o)H3_eNH34-i3{lgB zDjKWLz~7$;Q72OiI(3B^N%5SIb{{bzGY!|!@hflO0wXQ_VLul?30VbJ=U0;8iw5x5 z3T?uwx%FkHzlQ{evKMTCn>4A%vej`Hn55gZ~ z(eM;b28ZfyaLx$!K^3OQ;OFckJV<9u(_q|KuV)hk`^!W*2=9-|B zgWW(Uzz00nX#hoas<>eKeUJNfq^%0wbwiN z!L}YbmG?un02#WjT$AR^mjn_1I!xh!KHZzT8a}js5Av^n7FqG2pCr8~AR>!&;mVAESnt3!-n#jxvEQUCTUKHYs-kR> z;_Nv%FQb|Yy>Eg&$(P7(m7{F>qbBgSP6kTly2HOG1UlJpl;rBnry|>p;OYzwdOOn| z?K?RgnDSrXi1L5f`f-)>Px6pt|Mg2KC%M_VOaNpD)K12iWeV@Kz1*(;f5u{ zc$^o8(^s4zNz+A$%;#d*QkV}l0jGJZ?6tj($2g#%RgoIhL>>qb@U7NgbX4shn# zQQX9T&yMNuL09LQL#Kj$FfRtGTG7kGfe%#F@%>Hv6#k=po>t-z~ zHF}ZG(4K*~ZS%<6O&e&RbS(T-BkM?J@_h z{&|vAO}~kPdbMm1E-Oa5>RL!j`!gK;E`rkbC4h);N8pJ0YWmLKmaIH{izGOmM%N>y z(Y#w;@Hy9?L?smicY&#Rt$q%$npHvGZJN)dzek}rj*I9Dg?8}Zf*Oi6D*|S0EV|IdqTeq*LgqvVVMAST@7Ni<;Ol%kJG&Po z*dNAruSJnw2_ZvK_lei}5EAD%1{ZnuVvhM~vwABeB^QncgdJtUbC3hwVzLQV#oNshuN;^a>8in<%5L|Ba$@;S(MsDkizOVQN< zXGzb)R@ndK6#g#tqHe^nkO3EGGfOs$**>s)fFhsD*#10y8>YNI4rqi8nANt{_Qx72 z+jAqsNJ2IfrCFAdD%Y8G&6*SN^|}Ah>2OhWa6>osv2vq1=>z1;rytCTPZ3NWUdXHa z%)*+cjqsOp4O=W$jE>1$qKGBOQD<}wxg6UI-FB}gmgc5tPgy4MPgW%<8!%0s`5&pP zoFq!|j_^W-DcWl{Nt#8L@XuKa^QAuBq1*p1#go+vR4!Q)Yn~`ajoWXbr%HR!)ab>w z9}1jp={jxOwzNuwnR)17Y7V@a9|%uxiH0vi4M|{92RXEL2a*5q5S$W@M*(y0@?5gT zY1H%s&_L)k7*1G1%`46j(PNKc?5ZcEziS4*)p7^RHD4jKjI(LEab%{HvPM4SkHGhH1Mj5>HRTYmDQHgE7uM(`;?lYrPUV~Nhbx45tX29%? zWyIe}L(4IZ=Y?}2n0N^kfDq|DCs4!prD&0m8Qkf#3#Od4 zf!<#>VKhsM7FgfMp6bOUZ2oGZH~N)q{(h3U@U_TT#ZuCmxE=7E&#*&2ibQm$2}#Qo z?2P*4N&1EDT-qT8)`*LJ36|io1>+#(kA8Uw?FZQy6G%%TeK7#tQX8D z;)76FqaO6GZzFCuh2XuL{WxTfHu*jz!mbc|K$M@|CYt^Iz`QmR7m3Rf#k_OCI@=oR zN;`u0E1G#K#gOMSXEk%brwnf_GzQ)VNp**|Si-jz`C!$3b!z)K3p`z6CYT9Q!E6&x zP!x2B$#E_rxBKI9+%^+(_s3UY-}9CvpB*P7-V)SO?lW6Zj>xQ=A%gG7geuBfutwnJ zG_G=ClRjJ_vgJ}lso^saYP`y-hewgWc?ZcRRSt2E+C)-=`k1(it@ypX8k7F+Gybh- zK?;(k;ckJp$@d)s8cE~ezmMAp>^&jyMO*Pl@`n6gIK~EDzlBFn)v_mNO5;V`qd4Z) zE%xM7EvC<=42K!qCYdk2iM-wwvNNO&7w8$#zcXZM@~tF1Dmdg>c6=tdJ7WOfZ0#Y7 z3#ZcDv>A-!P&jx{R|VQH8IjrcpGZ`|Fp17z28>@tLOZn}$h&0528l@Hl@-#U! z-^b8>KmADk)2G~q33F;9nnR{!y&`S`FV`ZeoV=9blV~w#{PAZ35o(8Q-F6AKNw=F7 z)$$<-J|x!@M@XpD0vO@bhZ8p z_;Z&&Q|mU941bd*$%}Up;jkg{WZVR7TVTrsE-}D)2Ume+6(M-6JPefJV%~o5VDNhD zcjm@kIV55k0}bys!?MfiWOu3@jVk#^dhIYi{PHPiUvnICc(-axU0Z8mmK>|~fD2Kr zCU{$zOJZMd1y`5l!jm?BSZKL4vuwg2&x}k5(dr{i$P7DIVzH;6~Dm;@@TWJs8 z9x)&r4gKhm@NC8?)rav5KMiIH@)`H$SkRY;bg9QrCHCRL>qI4MD|sYIZJcI)0kIp8 zg3ZDb)cc{}{;FvRy<@7(edpWBjt2MBy!Qtgtyia6bHjU7FGiA$7*--1X124xMn>w^ z`)hDltyl~!K4_qXmqXy=q%ih9kxF9ZoXNs@54ge8*O0t(zp=*=8+>1_1*Z(nrBU5^ zr2c3d{HbqGvb5%q2J0lK?)(ZI`F)QUmA?p|TK5bVm0cu?O$AKaqeo29%|Ar=wH496 z+5z4@xJfP-3&HtgE5Hx09W4K79_bXkwVPG>bUdkuJhBWS2Rnb07h%G%vGN73Iyk{Z z>9%6rK}p89ak4(nkZ!koNY48lr#d^Gp(P#&C&{@(1ysK5I3Yw2y)0=Uk zMH_8Ym_p<96zPr`qI9-?JLx{>K`I<%_?u$WutT2=wVtz&u4^zQXPv9brKtb7IYCRn zr1V+lMZ_a!ue}DEdqCg?O%s9Pe=~6%y@fkZIgsx0CJ;XK1*?8!;P3U3B*G}1+)n(& z#LJ7pmJcPcdv!h0G<2tV^XlmK%{(goU<-b`e3ZRfXh=60HRBiRH?U&GR%-C~AX$8M zl1Ltt=DRD&lc^%%By>BVv#LtT$s?BJz4cGx*Rzn3soBpM9hC&XzCGsc`_j%9EZTvS z&s`xstDC{RNd-Do(TcSX#Ng2h3_@3IhAN*n(_?d`sPc>&T+F@7C^Xptw(CCYDv|{T z3;pTvCrk2dJ|Pz*#ObW3{^aNsfqt4dl`rhL5-)k8PmH!+B8T3l63eg=GVi4v)yl;9 zT!{|jFl{Y7Ka)?KYjQ~x_(~&}7vk^A>O}wMTKvIFz=pKkB_}sW61|0|$)xCP8tU+x zjK`s6sXN)VR?j^f}uMqL>J~Fx_j>x9!ldt6j#-x43H*I5qSIY#6bNfZ= zlJ3$OK3(9}HAQl7ZXPXF?Blx6UkNUJy9}pT{DR`1i6k-kFH412L$*PI*E<(FJg+v^fkE5{6cq5%{}G1}mJTOHewYs51EM4KS-BOe44;gL>;oR9yI3`7e@$qjPaO4>h#-&EjYi5(JoGAZ>JUFsB_y% zrtkssVV^bK_HijKMq{*8+XZq%WvKU+0pOE-8g7x%W}3}YSx-(b`Jl6$c5P~=$#05? z_la~e(W(YBO0~ha+$3PKqlC5}TSN--R}uW^1?}B@0#A9tW$VW7l8%o1=*Xrtk`r13>(Cu7g7dnE9HqOI3QQ##Xhgy=8Iq*V{g z;g{(;D6k+9?VQsEhYaG$$J=>Sv$Byco7GG2Io_h;5*JC5^lrSzF@bDWTFYJ5cn&^# zyj-xmen93qz2Mnr=9B4P&azMI0*RB85}1~z0%iVcK{2UnaG@dz`r84*T|FE5?{R@Y zyQN6VPFJ%1O&yu}ev~A;E719`y!p#6*z>D)E#z(kR~sAisuQXLcjBRsPxuj)S!2VTc!U5zRf&DpU-k4E!ib#amxr4Dd>Ug zIriwfTri2eE{n1!-oOX5^GTkJ5G?MtfK|J6k-bM16uo(mN+n*To4+T6SL@7C&tGNq z*?T**&Ur(&);%U4un^6S4#&yvS$wI`)qIb~$7%Hj3uF=yffNmv+$=MrE(Nl+J$UmY(+|`wlMNgp6Jwr(SETgf1f83J{E(fUrB_TF+bo;ivp0OTL%-*7}6tM@8J4C3AD}Q zGXK>TH@?-51$@bA;qYSHD6izx53G7pl3!N*n`ZQj^C!;O^Ltb6`D-(;2>JzF=sB;+IgVQGsw4B= zjgaM`1^m)o_O$rh1*rZ;jCL8^gRFd+35If=7w9 zW=#0yGYGvX97fDvdy<^PdL-EM8~HL3Lls2-l6yb0$-^EyJiF*1Oh5gP_=r4bpPPr$ zl@1}KZ`X9L+9DGayDOD;9qA;_xie_c^aN6MQ;(VdxevOT=VM>hIn>%ciS+3T(|&#~ zZVcMX{&RmpHYl>x&Ax$75$@+`B}5ZH?WKfih@-ygmgL3F6YydvOMc}MSSFE=d{+IV zc1Js@+v25k=_MO9`(_Q$1OUDX3&UAjRz$MApX5$<(8dy3dhv1`S@?IDUN_ct!`{~*li|L)p!{F@G7c_ZXhX4C|5xsOJ5~?-7 z#V@O>VE@%@9DdT4bd88WL1DVh_C|(=-}ytPtlC0TUT>pSZK`ytrl8xc@-aN;xs<#; z_M1)N&1Md>yO|T|f>|j45>0Y2=Rax^r-u~nsA!-e$~o%H53mp6|2lSF#m==DmK3X&030}GLmJF?4AQGGFoZYSE4di3!ECTx{BK$0}CtR zQ#nZfn90$XCk|7`=i2n+lMFguB1$<@L*&w5H{#m61Y4&oBPsY57p~q++7lW2;;a(L zwxal1>K_!HoU!di0t>tgxlJXWRKg1qQ#B~(!lb$0xQ{Jn-q`9CB0>`X7= zg~S2$WdzT>p~+P6zhJJi5S`|&P1lLOB+WU|^hr?#ZQ7JUcYgqI?Sm9rEWDLAG_EBV z7Yl)L`jD*-nn%#LT>RAa29UX9PxbhM&um`>J8z{z#Rx^D`%Mh6%XEl+ zmkxgRX#r74h$OG>E70zV0%q)k997cjg-FMfYNpB1K39hDtg7Kk1uN+Lxs+YMZUKGJ zyMv0zn^SNqgi62f!}s(o=tYBAT)9DvT;$mi#hOLn-^51JE29kD-ntX_e0gY5`-J`W z%Zcc~diL3ceQf1?dE|RA3cIa!BV=(k-YDtLqYuu)A0tz#`PE@^6{!%Byc@)6?N6ds ztVxa=MN!Y{Ogh-UhHi2AO4hOq=#)3hsppn_D6vS13a?Ltx@&*p*y%FY;X|E zFEoZCGnbR@kQHRQL@b>O_{_=DkK_xUPt6ZakbC;eDXiZ`cjlYW$AbOu8$r%_fl4j8 z`3Dkt?n*%xDU8)Sxffn;Xv4?f9V9Yi>tM{>r{ItG2&X-ul)U*SU~i@WVh8gO@Ook) zX)>!PP$ZcBGF^x^Us-}TI(AZRZ(nN7jipu|$zXX<3fX>N0jkd9!84b2iHn>jGv1{L z=jSS+PczmaBP&E^{CkYQO``zcJW0l8 z%yPO8{Npx|^tw?pcVH?ykanGE`JD#DQ;c~b%lmP{ga_R7&4*lFuS=Fp8Nsq`55SXq zyWymORS*kIgLQGK@EYeNbk|u4(^XD_o9}K4xTUv13dRy&<`Lu13j%W@F2GOiOW|j) zne0`mZDi3q4I*J4j6PQGMlKv{l%i{gMh#Y>BT7@@3!~XEYg-?T4=4rk@sVI|#CNtk zp`2-+y#&rY=0$Q{y6`UkRJ5N+BiY~c(U`3*dd5Z~rIB=W=EpTSBgp|CwX^^>N~!Ru zku|alH-xX;uE48T9Em_-z|`OP?AL z)9G9&;T6Zwh>!TitC`H1@8P)odOJ2vbYvE5En%O3U(PzFIB>5Yqg)3sJw`6}4YQ@f zj-_`t@b287fc-!HV9&2W&~aiE98=8#BmUhuW~D1|{cs!)-;aX(1as_xB@4-G|7!Ak z{sOqt@((!UO2NZzE8xGcJ@Ap6IQV*PKZ;oHfo{rAK{7p?(EWo8VL?zg^e+zKxu#cQ zl}_u#zq4KQS1FXS8s=;**_cw71f49+b> zOZM$Yp?Q{Iqm3!-{Cyv0)w}j%h;;C2ms<|zs7FD0;7%3I+1 z!Y42+eiWXojs~&v2}s1d6i6=+BY!*tLI21zX2sqpM$O6#ZoQfXcCQUX8(x;7H|4z` z-%S?^lgmJA${M)ufE%=EG$ZZZAM4I))xfUIQUQ;C96rCe30%>(C$#h^-e@zAUGe(} z*X`D6rdZ}6G&ySsa|ZH3;EsRb^ymU`#eWwd`GUSNc@|z9mLz?*N?^o|YD3;+4wazW-uC`=l~RkRJorotnWcr#PTvIts*p zT?VJB5}C~-SK-dUb{Nqc#k=&k9lbCI=#*0$)IPTlFSU0e^J;F8#X)*#$82R(Yr?{I zwO)|p!p&rv5 zIQ4!y2>22O{x&@ZCYJ-@rWs;T>uW6Jq^80hmd(&eB?<4fngutW`~vERgkW5xC>#jn zL60W|;Cy&Kt6VJ1EBUXI?VAw`pDF5~{a^fmcFS$JsoxjD+!*NNlZ`ATuc0s=hD%zN zP#t>+-PijFi!(gYLG3;0AukLDO%>$pJ5At@Vh?6^a4FQfZHfYq)WExW;!Mj8%IGZD z#^XNXxarDI{OI8YmUWm*lpjfxk-3<(T5opshr@jc@(}U!}i+XU-AP1XI zjU_eXXUV{r44J(96W^nSOsF7#dCz7TO&q(}o+h@G$hZT3HY1REO98| z`i1Ipue~3_R;5Yg%jZXU?HOg<{y7UgF#W)^{)^^LD(2PsRoehZtvXQX6iNQcdw>8V z5mwXL9cYC`vw_$1*i(5^aOdI~WZ6(RzM0H&-?=Yl?0n-{i%`U-SB@|~GX;LyHc@ze z@+}VddW-U%QmFg+YxJ9C67}$3O=TzM)1#`Nak_R5Q|>hbX9&7B_jSZG-<_p#WJ@O= zQ$5LKAIadQg}9Sbi}W$)$!wf+rvhwAy}&L*%W>$@uSAS{2RAQlB2^W^;M)^7lEQBX zzAhW#-w|h$)}aBT*4fc$A2k}5^np|ccHn0rK6p=v68SPPijOze;W>#xfK$+l6I`vx zBAK&LIw=F3S}=nLt=1Ir{F(rS|DpBV5%~c`+!Hw84{f|gQN^3)~T-`_`nhgym_k_v~+uli#ul# z&~1*_$hct~%pn_XRN+hG27GGbJZ`nH#U*c*Fe6~&Go{iPXS=bbQWNX-0;`2cmxcr_9+cUYCwb@}uy>34v4syCw?8FA*=u(Y?K5u4s{?VbT)5Yi)!QLS*#Rtk(E(V*r za&fxjCEVV42-bz03v$0VaKM+fL?mpGaKlak)j7%_<+=$gl;#5rR{3I0(R4vBdm1$* zMI?apnJ`_8!Nc}gaI)+;Ec`nce*B?<$IEr`563G+6!# zdSup<5J)|h@qsK2vT?r_v95Z?PI5lu@Z^y?$@o+}mAjov$k))5&6rl(rIH(a-0(BS z)1-0HGKhi_>JnT&Gp%$!$O(837QT!Gj*kCu=_M(W>m-d!r@jU}Rb@QK>N*(i*i7Uu zWn#HQUwB<=DP)^#5B7U)jg6+?A_eP=Xu6-LfQNoW?)sQf{pWW`$=XF!SYit;urDJI zIIZMKYcpy1>q3&=ePfOxMWi$oNv=P4q>b6rXsq&5@<3&f_lXUu4!`2Yo2Jym8F=Qw znOk?8lPi9a{lBMrB2t-|JC|c)ci*(``4LCnfRQqHjYCdtX#Y|h<;0VC)>#RgjrNFJ zE%BSveYo02Y=t&QF~^c;-;q`)|9pTa)RoKdr0cl)8LjL=KLZ?CRK>I}ZDxP5`Wzwk zAKbF1`#4L7b+{*xJYGHH;M{a4cdqP-QoN+Ul=sM`6K~1R!|#rq#V+gSa4n)2)x^$K zw9(VN#>oyZO%op+=aGFt-qYIzM7k=z|(pV%9;6oOU;y? z4(4 z;hIjga=IJS>&7&qxPSiSa2+*gal1=d;NI zT-g|xYrIp>RJbR`$Jkoy436~z4_;c38rJ6RWED;eG1s5nr ziq_?ihtyHm25XZ`O5A}Fe;yj2w7zxvj!pNw*PQ#i8*1hFGgQC`x-#5bs`8u?9^n2b z?PhZrctm)T(dszKcU!;2BN2nW%i zjtd7u*k6;Kb?svpxv6CjInCuo?6uVfyp!$=tzVd11Bvsuct@M#xEB>qFbgi^REuWc zw!U-XfsNfaowe3FI*l!;p2~zL#qt*X7lDh4 zjd>Ow_BLzGrm@@GGzscC&8?Ub%Wk@(04{b8);YSax6r9znI5hw@O#EFmD$2^4OOrfK$&wlFOjvqzi}q|7%z&p zKU`@$DrsoDQ9%|hS8N0UeqFF!`!4ML`kI;H_Yr95hk}D=BoN%t0}X6upo>k(_=RBy zQEnWB+uAhf^LJiwAh8>k$Ou@QJU_(LWy1UVL*%y44I-QKldR4*0nq0>{vN`@S&qx# ziZTfIiQWZ=4y2&Ir={p()f&`e??>|b>&chAwYcDC4by(x5>6A10)fY~;0L{(q($Kx zTxN8ZJ>R8IrkuLMUY-)bK3edLb8758@Au#1;N7NrvU15K5`Ts=iA*&Zd{zcR((3W= zP9KnYF&@Z&Uj+MQrnB7YUTC$&9H_2OhiR|$(T;~PsPFX&wBD--9dL`I^Ia0D%%R`p z&-MAV;MO|&o|co7`c`0hg9&l=_l7SIbb<>@t%0nM8q{665b~SjV0eKdT4ELjj%X;r zy7DMedno|fc$A{>o@n&>F^@>TIYv$&tAI?&V$^fN1#JpQLN}z;QBhh9x!+U^7oHQa z_j%9ohOw<^a~Hu0-pXXgxoNO5%n01>4JK`Rc38sc1Ds&`VSD=ll&mF(zWe*4Jry!Y zcEv_??@>Q&Q;R~KFU~?utHW3^br?s~--A*5OW=Tb7u+6S2U3o^!hG?wKr`|$k#Ki} z!WX2urBYF(Xt&^I#lBh?1Aj+qLXXF)cz!&AL&H*N zjjMV~Y*dE@sXyW5Sr2$`P8|FiafCRV2tf_?c=FiyBDim@3VMbdq13Jfv}aQu z+O4ICB24lj*EMSh2f1WYC@DH-P*W)A!6%tck1FM%m#h%*( zv8?3?05f~>{^jS$=JtC~r(-uODe(CIna+gWQ&z%xt1a=HM~Nuuzb`O&*)4LyN&;4T zC&F_Ldy(0T4`@D|fl9Sb!k=s#obGInw!d>n%^YK7o*-fSAo@OTr|O{;BV)8^ zwkO)ZmkR@=dw}av1oz$$q75+!KCxd&Kc4G_v#sZ%AIrsTql{yi=uMB{y+VJ~nNfhE z-AfS4-HAdUe1;lRb)bHv5`6e@Iyp070qV}S11m`{diH1;S?HAkS8d-0s^=@i`k_6r zWn>KsUm}KMs!dRiwG#2YxB(^ZutwT<@|dqZ7U-aSC~Ex^hxQuT*!ul-u~6Cg@0Q?c51YEf(mg@=V~RsQ@@{-@#H=71>GYBB!^n z@nk^`oOo%1)IXlZs^#;@qjO`h{Ma<~B=HaYJ@*FO`y&CC9T#-8eyD(po;SeofjjV@ zs4+@#DkO`(xspF!H}O^RC-C8+*_0#hMQ4gMS-gYh?*ZQ_9B$ zW*^DV>8iGJJ!{b2RU3%ZAy+^>^2wSsWigHJs=t14rwDByFWwo0L!C24R4d5)j%aD~Ad6ZqFO z32xajSs22Iu%)6O!QBFXtP-Y0qPwWHVIhf5X9xA zqg*RrG(M<>X1s|)caM0}nTxXMk3afU(8CF}6gL1kq=b&m-vZZuse-2^98l5(7f-6Y z6MXGEylDEI^qI;s`nMVZZ__jAvPT(G!9H$b@Ox-T<(;SdhW@m%(eas_{y{v+TC-U*U^-J1SYEL(A@p zqG>yW==-J$R@SlynyuMKG_I?V;a$mO+m>XqzR)Lhi{=`pq zE5bRO=HgL@CRXT^IvN-B34EGf0oBIM;ek;($k(fejjSa4*XV%iTc3iT>r&x^wldf? zI1K+ud()M&J@5rkM_W_@YI$sccBQR{<*J5o`&c(#kQIRsq~-zl^%*$VbOP+_ddXG3 zHV;`={KHo?FJTFh>F}7*3cAqVkY03g$3ekd6dJk`iNu^jgEf{^W?&udx}!= zzs#4$Zt_A>*%?6XXDKw6NPx@7E|bR}Qqb*{Lbg+m3b?f^J+OPI1%^#n(D$at>6W~o z)JpCG_4T|$FTqU$jpyY2yxBB$ML7dE za1lH90sOw(2kd-4VUJ((wD3m?*s^&AdG&P{Yg}ClAGUCj&y)!$w3bKiDMi5U!w6X~ zT|iB~_|dsd1z^_r8FXWdH_})Xi82~Lk#)}!$XcIZr1khG89g*foKQ5~X_Sc$o9{;3 zd|TkPhFk1;;deMx^cog#4*@<`M&Yui75JB=DfE%Ag@fW1wD#Ou$Ss%xjX#RP6sz4_ zPS8`Pi@%zqf8YUoCEf(|dtU`N%bakWq$K(KCy#UV^I7O9Ax>5$3%a6G`dNeYYasiC z4n3fw3nMDbApdPPH~X0>u6B$DC+3O3`CiMgXq_7yV!aMKYa~NO@k&^4^?}LV^n=9e z1(C7K@4>a1_n4CFeo$pH0m_@@Fhx@0pzO_Qp6S+4jKA?WmKU&@QN7&4StW14G1%Q_ zBeKq!(X+VE(^#{e9r$sZZM*NnsPt4Z9Y-3W-?6#K*6acN)AL z40Kk^fR(EEV7Rs-SR?s}-R4J0Tl*}!ZF4GplhH>aGd5C-vTSnU?Eq-`FbAH5c`zrV zkPVEBWG3!=Gs5DEL_YW?H>Xwt-lo2w*xn8%4|J2rT5+hMp#xecmVg%q4bb{wGz%N+ z@fqD`aLZIPXd8N#9QM_NttEqSs*fw)_FMy&-(s-Q4m;StGzsUSAYM}gWKA>#*&eBK zM&<54-rGuH(5;#Tc4fPP_$(FpO1la;{>sGfhEHP0wBu~$coVPgjUrxf$Pg&~vtUu2 zA+#xdL*@zIU1{$mkih8zm)AZ4N7PTDBe(5ISxXHEHC@IUmowR0!wW%Ik2MgkO@K7F zoV!ii5UiKm2aK~10m~p+P-4P?K7u}I?$s8!;Dk9m?)iq-yxoeCES?X`w>OZIGmF5{ zeiLgVQvV0L;oaG&=d6TDpl^yqAXJ>Bxq z`L!8Saaxg_jF?L#M5jPyD_1y81wnf75Y`iwRQTko8yJ;QX#WJR~xoUC=hbZ4*5V z*BnqoOZ3d~lu}Vxlr0P3&0g^2-7!Y(`$<-};u5Fw#&d8c<0}{#=;RJB2u4KB1PK+s zV@D)J@wuSC?7`8U>_5jLsIF3qi_SQZfxu6!x04V^%r*w>M{$UzD}nBocd*~U1>&k} zyzhHwlJ{5h@GfpY2zy?|6T4l;{1EgM#BJZk*?-{@2+&x8J$6vsKW&_;8^J7RZ43+xlYimHROded< zKLwf=Vy0ui1342m4~lDc5j964B6-UdoVj&|evix|MW)qM-0K&;Y;3|GJDNh{|Ebe* z6&-j;+yPBl8O5)kW=mDMnWXaDbVMaikYzGbOz*8atE!)j%2tv{!@uOY{w?m*F>Ml?7fB=B4p6p4ivGP_PuJH<@zZu)qmx_D zqUO#H6eePgh?e>7jIzC&L%v=Fv?yUm?#W6D+e7^uc|KfNjga5_3@{GCc1u z+!1;MY7A;nw%HLMJRV4-t0dw4I7#4qL>XNUeG8?ctI*5Y@`#gbfi&GE(Xw;RICFR> z?S0CUuVJ&vhOJv6quv2e{?_LIb9X|m+jW78kw?dZE6MVLXolMq3P1nXzfv%pdMUj7V4(gQ;+_v+%sitb?{oo{R}$FF}LL$x=b0|B3BL7C@Up;zfi0Sl}N5AJ_XzaBkBS6bg?GRtL= zWY0D9M&8uc=u-e%94BL|@njGQ|9t?z@B0sZ*nbs?G~~i{E&v29I{`yC{m1g1rgC>L zZUi_X3$|#^Ko5^L3K%<6G*A@)kK8c@KjFI52<=9W8k*74IPEH&@!EO;&s}<3y z1=SO&SZ>cD+-m-qcdjvmRi9g8oz+e{QJC%yCaZBNH}pv zUwD(FSJ^Ee1-g*$#EJwq;^>ST-XHD6>fFLH1y2MJO04J!5e;bkl?kf9q092TET{|YpeTn;sr=_yo(ep$D+<03+KG0oTT50THN=w^K?8FK*)?bVpvhNn;IDMMt{Gy+E$%pH zQuz=r(zSsP87KVYNIULr-2ww_RIzHwG+V_#tC6O-7SDWh8Id@pksjC%i4%Mp1I>?q5;&;It(VRGsM;VM;RRaj;nPckCX6w5NWn9g9p~v*c^E5 zO~&U=V^*}>09)hjNqk^6$5hybJ9=y#DjvhQ?`nvZzTK8)jV>0%HRqy+YvQ-OM zHgk2v}a{I3*Lvghj`1bB$P~MRS_4iU_ z{#l-@;4vK|Ng2S5=EI^KRk}D|5*W-=WVf)_uz657bNKO5oGPqQMXR>rfQ^}~!k9ms zASI2}oIbKw)?Nn}yvmWgQV##YsyD3Z-!8UHI-U*Y>aZ7AUIPM>!j%fW4DCxVfa%qi z+*zwUiJlybtdzpRBj;hf_fH{SzwSK$!Eh;Xt>|Kk6p~oMm1Zpd4Ws$n)YwRmnWQyP z9{rVQAWBL*NwtF`qmxv^dh3bWCtD05h4%e?V^tZj&N>?`@>5{ewKt=IO^V3o-%9p* z>lCDRY8Y>PE5^?JaF#)tj%e$o1+dk|5iPu}!>1 zmQ)qgI=TUvpPY<5RpxRne<#r6APt2#zUPWE7qF~(A-tn)L}F!+z)A%>e#F33Zos*t zY|mqN>?G*~cBUMG(rHJ?8^m%I`Dq}zAdkDnF@aI8p9H%aipjApy7*D<0O}fEg&WGqk_$}Us@*Nv^{SIcW)``BT)w@H;m=Vl;XRWUqz*B{<5D8<#e%KVa+8Cd>L zI9QbOg_*q11kdHk;-26|@OO9z=u*jIWA%m`=Sn}s_ez3r_^N|UfLjP|t*QeP>qc0k zKSt>Eq9#N|tjA{krx$FGrU}jBnvp$T-)V@v0i8Yh5%N47 z0}pphWB$FMpyJnI{)mewjQW)eE`jCflUNo?h>=E?{pUf4(tMoVC_+@{-X*pdgigiu z5u$y@hPoP7V!Qe_^ykq*R#r0tUOjx3oOwSFw%k04{*H>-2bFFD@9N``i11gu-M5Y| z|6o90eUL##o@#LA$R#E?YdTDMyaDyRxP*+SUPPkueXK-=9~1H9EU$FKD%kKy5)}+C zM6WcmzzJ0rE^wDX%igUg1K~pFRqcJ&;kzfj4Ag1nruQW1LlwCZsf7P6P$5*9oiT0H z7%u$(d!|vrF#453zsWCDHg3rvUni{J=mmOI^~kz;)5$(5K3SKsfi}#Kp|gYz%!<9< z?6ym5xD7%_@w=~6QRkg57&7x344I>ZOxJElhZgKZ8%+XG*Qs;RsBIH0yweXi{VYe- z=5x{Ni&xm~mV3aEhz;S`T&5-EVMz9eBwAviiWg_^WOJCcO#h*`!oT<3NGsrk@VT^s zIVlrRDqR=Wl7HaJ%1AU+;bR|XDTb!~GekY5bI_C0eIOw8GfcEfK@LjR=!LHEOmaZY z{=d~ire|LTcl#145|nY7#y&ZPYYa~!owNFamhCmPTWtyYFKz^P_*%o+gL}agjYY^x z^8zx-HnM-%P)S_b+rqhP6HswX2}-8?#_yLELKW$M@XLxnFg&3Yu9EF%%1+ndvk$XK zu&@T$+$@1+FFye<-7m(@eT%T^x<}A1J{R|gX(3KV1DsSW3tp$xlLF1%aDm|(aGx<_ zLgimHijCLN+>e1YOj(=8&X-{C5B*>q)@Xx>^+IRY&LF6MX*Zh`EX=0-9s=79r{b#M zL84x!N85rDY5vU*_*uaTX5O87Y+a|0aDGiVHx+1Pzj-+VPGxIBoNOjG+$>^m<*IA1 z_t4C~-ElINa4ZHLTWz4Ct`Q8Gmd(}PqzQfue?dg?JsZ5bb3VmV5<~)4Qrx67dXxZQXByXRx zQD4ybFpi2ktw&*XUi8na1LXA1%_Qt)FD_p#kIIw9(Ei{6?$h)U&P78J?v&~0;N9?Q zbW0)(^(?6dulk1J-tjT`ZOBmgzDts+1#j8=Ij&$zn>PAyMl(|M?t)HF|A8=*Hn>M+ zJ3aFximF{&NjJ{>PJ7DhsKvR(N73R4W&4?Vm&vk;rLc6~X6UFQiNsgA(TP9G^mgzO z^1Nmt2}N%B%l6C6hvB`uXn;qA-TY>pbVI3 zabOC52EKeLz%RV~p}SrqHj!(@H&Qkd?$Ue2@_ig$`Qan?z4Ti!tf@#Ad0Zoj^#a)GC#@x3PxZ)s%RZuOnveT;>XLO1EDktl1J4@Avx|<$kf6wQP`BIx z4sX?`+?&GulNk(|bWhYiDa#X%UJ9oT%3H|$BZIgh@hd*6w3@1145drwacB@{F36Iu z;3p)X07I7-km+q*%sHxSlE0TI71Gt3h{vXoNRJX z^E%lZZbQ;fKLQiJb>jehF9#Z(5DKPa=4Yd*aN!y}VP)@WA?l|}fud`NXcUL-)2&n_)$Z!busxyMR zhNCcEO@lwUp%!ON^&xgu)nt`fC-d~cI7x9zrk{(YX-kbB+ShN-wHwbPIO!?Tm(-vu zpakMP(1ERPlLXt^9muDECh|vcffyFcfQJ(+f$H@$jHI~*ntRXSg70h2pOZA@82$NFsO?`-<=t)#7 zD+?|#O~n4&`{0gk{`h*3ICSdV#Fdnp2DdluBTEM!gPbXakbm|jG*=3TAFMOrcR@8g zlz9`Xif#e5;#-N&6=Sea=w+z3nG7==`+#2B3iM&gZMaqG7`(3CLRzz4LetI9u->kH ze6>hf^c8r+Nl(h4g7_S`eZ4O{XhI<`EDAPSyaRf-V#qPmQQV)h1&ZpQCuN!y#3oNu z==#tgOWI7}bT>5+_v{_eT6Y|FTpEH-<5i^6P@Gsy-wO1W^h4gqBT&md1R5SFhACS* z!RHSp&|_pbZ+K-O*Ja)r%VXO`}@!Zn6jD$8*@3-QdGF+ zzs0cJF@0RJ|1=&sI+IbmB9LW4wQs@!TllM zuyJEKIhB+L1KL&CU!l{8?zjvbw z0i1my6aI56h5Ix0kzL?t=rp+nPAI!!UxU>^r{)bf_9_ZTJ3e9$&)We-GvuJ!7XjRo za~kFeW`d{}pTXUO`vCYZoaP@r#r{(M1&X#f3M?Y_#XbSxreLFykmK)>v5 zplDni>I+%1W1DT^lYOy3J!%59K8}U23rE2<_gL;K8BV%NdR$>B#{FTWg3Fw7~K&W;I z4fZgk-rk<{PC*3yPxlYC2V#O7cbCD~!yJ<1u0stDUW8p=j=+sRk({BsO~h{QWU_71 zak?!>7B%_Kg|O!~WWgz9_1qg}#!W>GlS%`A)X|0ei)sB-4b+!dMxJXl5fm|%u9!9t z%}}W$TX-$xwBtg!s-}sE&mJW!?(C&+=0wmJmlu=1xYuBZuufQ?>qQ=PErb6){$QH# zSP^BDbolrGc~1HMP-I~}9@~A8&XliVtu7uX&!0hhd`=?lU2qOmH44Cm7!~yIyDjW# z3cQANO0d1}50)hvt0lzmLA8M&0Qmw^!TwRHBnoHQKdm^Om*FoY& zQb@F)GyYiclBDC?=+nq3dT1^5Q*Bv07Mi)dilI ze}~>(f#^&cA3Txg*c-FfhzR!9c>8ZoQQ^B1s&axK|M#3 zNa*K!Dt#~>sm+o^Rl{xcZf3Sn>Cp`>!mP>ewp=7RvcB>P0y=2t?Z!6ILkw-&MC9p9AaV`k5wW|3yDban`wEcyk15!Eo+3KkTm?PCC(-+_RuJZr@`?vv_~ZbW zW_9TJwGMY4yA6!IdPqiJJ>)hR@*~1?u!8t`uH3OY?(Mtx$kT*%bnJ2~)taw@dgctn zCZSWN;GzVc(0oKb?tevlql|8+K;CnK!gYND8qg~IQ$1%Hh-1pd)h0{i@_f`h$o z$ZlCY+T^YSUiLmE=5s&e;7w}qsmeBLxGDg)9B+W0-xJXD05`bwM-o|3(G9ot6cRC| zt8~Yu4agx*9sS3@&XhmePqe(Yb2h&woUIC4#5!G*$IirfOJqJ;w*M$H(3V6e4`spw zjsC>$`)vH$IhrUYonwasw~)S*@s$6v3{2no3-)))LT#(%1e8=q`Bs8|vI;NbjgU2_J|C zx;@e)K<6Eaev|__0p4_$-wmlXv~g?~&;`56 z&;KOQixUrsSAr4x7jA@_uN9(pm;bPZgBk5q{Yi{a98!J07nMUZ#8`QuBNgUIZ?FwY zB)kBE4{b1{C55~`v>Qd#o<R`fXh&d$X%wJ`o z!n+rs&NHoj#co`s2d3Ot1=hn3WN(Ne8z*vxyS*lz&;KoL_iN#q#-RrlEMHlX^QdP= z^j9o8hu(q_^(rwxl7ztIM2j{ z_4}V~;EstCf9A`l{L{Vdmuu?@$3aD$o2OXHJ-sp7Zm!IEZjoZ9opH-CPV?`%z(AAB zs_m}ePQIRr*X}D~uTG)7xes>ollIm&-j2`Y-`e1cHDX?K|Fo}YkejoVIed)Q@Nnxx z&Jl$GW<~#My8xaRXZxH@`2LE^%*M-ixdVBk_=jN*zpcob|4dmEYyGQbMdW$>#>Veh zS7Hq-X8W3#HmRKDc0J|BCP;&hgPGj(o02(q_eYm~KW=Q9tj>5pEwWvGC5}Bfdm)pp*vcJW*~Yis+r&A#Hl9CxJqjOD^5>5FXEAxH zfoyx_Nxt5>`;9-}%@lHim26gl8o&S0b|Gu5&$^lpH-^2f#tZa=@rLP$Z}wJ}Yi1y6 z`?e{IFJ7>ld+~uZCvh;2KSys7i495N%3Ph#Th4pTPYX2S1TJ^rEVVp}tv|YPs(e2; zhPKCm`@MrKkeSL6EsR%bDG%=zGnPEMwAm-B}#0b%5mklCGib;dvVs5 zpPcI96a2U`31(b1xN(_DzTF43j##`)wCUSs*?6O`0-NijjzAF;Kco0 zPb`jJWoAl!wEHoZjSKA;@Lj%Y@H(%|X7@+F;ZAea;k}n^VaB6RvDfA9G)!H-60iHF z%}2}!j{Ef&_|KI{c5aR(KgU25m^V)c8}<}4cYb9`Q_vLAgOGtT4)*Dd<7-I)G3TM}K&5$~AIqe;`5vXwV^h9cfJ4t8bS z6v1aZUWNzn6L$(WZRR!3JkxhM!d#EpGfSg!+w=Cu7g?gbjBTYhY4@`l4_!Dz{>)Y4 z$Nu}s_qIsn4wQ6pcP-LkzXTeQDK$nOIt z_-5uJBH6YBch7&1b$1oxJ;D=M@T4Q8Y1Jk5<2caPOUzdlDQ$X$dH`{+0yQSv%&_rJon3Z{eo8NI;bNc)}{i@x=-AL z!#@0xxKQ4x{u=T%IFdZ9P$Iv3M5)x+aZ&PFnQPDJyE_bxY;Pwa0`nST$U5|RW#%9ZvzE7y&?(7{kp-=m@i6}FD#)4#_mz= z-RU%Wr->kVmAAmQdx79CTt)XSR-rlPd&#YnyQuYvL~=4;4my4KM;3kHvm$L}M8|L~ zaeEcaR;`=D{<)%w?091EtK=jqaMU7$(<6yU`c5{q-<+DRn?qB<1pc|!0_>|{uu`fH z)ATonoVWXdmtM9dRSz_%_f9+bW8EKI-*|}KU2+zCPLg3BMJJNZz!gN{hd&OL86h|H zTxs0>RQhh|C3a7SEPvMX9Ik!3oM5LyG*#MAMtd&KB5Udl@cl=kIIC`f7eA zH)5uNop)ElM>j^u)H&u<^}tUODfHELwkE(8k8^3Ea|2CGn@bn{ilKYINYT0356N1q zPP%c@Vu4bMpJ4y{+w^77RjSz7LKR;qkkh8AWZ7USxtW>-as=Zzsp$h>ZM7y9iIu04 zU$m$rM;85l97WVU^P!ksGn_7*@s^m?PGnV1k{Kh@n5u~ysAPDS{%6~R8`nR9axasx zS^G-N)6*qM>lEn16)$M;*a`ag?IAKyUr6G{0Uf;5LMr<=(Ho+7sH@35x@E$W7;J1@nPh;uT%Td%$;vl_oC<*;*bU=UQkD&M8_uD&L_}lNPon(LA-WcQ? zx+A&mBIuH=GIK&O#L4^F!0#M*1RJ&Iq5QD~H0MAU9*CPv9z5R!{$08b9b2-W67Kl#^F8#^X2Nu^_A;c{R7*H%I|78`c)rsj>O^x=f~jw zL@xQc_9dATya^54zD9$G)#1uJeq=%_h&;OU7J6l>q2ZVJ;Oc1vy^U5v`4_Sf=61u2 z&y&ExtvXOqE`j8of(<^=vO-qmJ80@ZjOM3zqkxy|;PWm$R5dkG==p6R$8EM_@7-1C z?bWleb!QBbpW9|>x|Bc(C;OLZrC60DWn)V(b^a z0@9mXv9_5$m;KfZRrd|7CIPb9moz%`mcUD!Sb+?xa>mQADDJ_R;I(Akli>-kVsC& zi{K289VqVYDF|I9P(kE4nd|qEe75+;+go{oByU)XL>jVDu96vaaZG}5{JGHmXb@?s z3Pw8j6wp_%G|UZM0BUzCf;`_=Xz?Tft#d4avxZ;5z?*$g)qNgX`m7gZ$Myg!I~Uk~ zEFza~J-|gOZ*ZWnHt7D@o;;OIz38r`=4mYXn~|G@4bN?o310A4YRLXV|a(Ck7;s3@{dXt=#IT*+?S<+UIPrwdbYi zkhc8`VO{l0oHF?UOc1prv66t!Z~To~Yuyp2@hRGIOdXkeUPhYp|03tQ0`^5r5?c6!~iaDQRh={TEiDMHf7fuKNk(cZ#u}78GT_)@>y@Vseh~I$DWl++17~ zu$WvG`WHXieg{`>6CmA!g;^p(rlaUDywoiUe;9vdEZ*D(75+Z(zp;2u;LBcK>*mvV z?T^1gH~J3bT~mubeK^m36`~~Ezg{sSxeln=Z!pmF43g9^M}O82 zBBi%Us3*4qR(#0Balc3Lg|Va1{IeQtERlj&%GAN*H@fV|b4B){>vizTE(+y7QME6- zNtvOSFSvCg^MUn&NysMG5u5ESf~#a?ptWQl^th!0mzDIwHnq9vvsoP)_I?Z-dWXO| z2_<-;{v*02WLP6)?P;j$bddOjfqEG&5G-!QzdnUQyZ?saJU1<&^XfD*%m0W3u%6NH z2?FCvhH&YlFk$Y@MZ8hG1l>8ZA3KDO5|5A`vXeVRJ>#YclH0@usiJ$Sqk;&!_T(ZN zRliBx7iG=HM{VG{A*B@Sw*=n(eM9@3MkNDPWvK=O`ixGD4}IXaL= zdF3KBLbaT3Js(ez)+=IB(SS66hN9smPtl*1GU&YQ48(pFj=9{8`W-nP>U=%lO zAj$vL;@w-a;nLuT56Lq`4rxc!;_vg}`V0Qkdr@ zi>{sUMukuT4ag@VlSh$wfEAET$!xkt>O0C7FGX(qufo>ICUUb^i=6YSCATWNsYH;Q z;BQzUov$n+2uN8Yc&sfikl~mL_%db!<>O|8(~24b<;fcaTklv3_Q-yPedns+`le7) zP_IXWzrV&V@DtUTsD)K^aU^G;9j6@DV|xDj!UelD(Lz5b!7Vyd;NWd1;D2a=vvXdd zUuYcFEt03-t$vc6-4XQiy%7SwY@wWOu_WW2hwz-04;Hz60Rt)4c+-e1PAU+^RjWSG zqgFTR!6=rly7mvsZ0{zn#tc>3zJnH{c>1^GAE{X3O@lKDZ1iX+Hu6VkBE3j2lP21x zw}Q@&c#VUqmr(I5cBp>oENFc4BJ7x=Ls|0`xKF+sO8r_uO9LlEr@N!z-h*Q9`tS$b zx|`GS9`#bXVai^j^70G3FtdnM2Z^GHpJm8jF_M}_%@oY;nMP|Um)gIvrs?~akeCyb z1Y0*VwUDygu6e|<|)xpfh)alQ`>)qil@tM8$w!dX{~ z&UdKx#dJZlhOEH2W;vBOk^mSJdBMtlHwb?Y&}FY!n&iQf(UC%;=o3yh-fY7r!?UsY zCl7MEF^rrFdW5=FHzRWWJKVQUpMI$x!wY9#ho{sad|-8*@f%x9$`8IGNwJc2E80o+ zK9!-RRiX6bXeIgDxK@zU&!hXUms0mtt@Ht#P7ZDnKC?%=;5|cga>&LMHI!1p1i3JUk5@NfPGY zZpx*d$4=4epY6n@u$Gnun^FbCCb~KG23>a5kfz_AL>A6J4!IFrRFfJHd74pZOgIN~ z$-;uxO3cKsveF5!sepQYnbv#j^@gjFK_?xTbzNDHa2x(Ocnu-PFm`5Aq zw-=zb2M(jX(OLNV_9a}VD@*l5_fdIvmSE(-W%_%*0?G+ahT*02@X7Qb`n2i_ zQII#zVABFcNdpEEF(kvcCV%5{#r1~E{+lTzgl z0MWI=`S(-gwBx4!F;acMg;-uVC-grGYvCsAN$!)wWabYm z$eENydv0V4?;~C`F}7N`*Il6I`5d~g>MZd7*TKx1*~iXoaDfVwZc%ZkYI=3^R`fOV zEot2pM0SWj!k^Qc!V)AuhpdZ z=V{_ywwHe0uS@^CrAlAUl_1V1YeCL&KGv1lPt^lXP|iwcQt_mg9`tde#do4mko75W zc#Uu$So#;Vikt^$lOBO|_f23R?GV!XHk&N&3q&q~vD{PXZ#nlG&$BVUd+4*(W^l2I zJ6cwlLw>v}Am27wz|H1*G;rP|WZ%_^$`g!fOIs>tceX=ATR+&gD-Sx?U4kYvc2W20 zBpTT>LGMbBQ0oc{YHV?d9G_zc6uEZbx>NwH_0~X{dnTjn=N!@R#5y`%=nf;nH1>Jf+c2qA$!h z>;1Emk<3q}S~)e;Uu84M_FqRUw{9dp#m8{RmV;RB^*Qi!c@Lo)4{6AiajG~&gFO_U zOiIL$($FL6Oz40lJg@p0aNpU1o;77`9hQMUGN0JX{3Xo%kajY*YMej5x0!LaX3+iR zBD~C=V{qVs2mD}Rg@<>=1J0+@w8tn&=tkC}E#(`*l8JDTIEN2)TqAMN>^>}3mWd-B zj^Lg4RqT#$CrOfV3wBAbX;4mohpVJ?@YwYs{3XyD@(W_&IW;woiE}D*bcQUvm6gU; zX2dsMN@(L<`Vx!9m#$%|mi&W*tDuO}(h zN`m{`WNa1PM`Qiu1*&Ql;I;fAvh?Bz*{G&QuDG1SW^0#|h?D|u&OT>m;h(Eu-&+A` z*}faZ&&~vug>HORiFy!cpHI)&+$PiOion5)b;O3_%Us=(1ipB`VUBIy1B#1+sNN+- zdTU)E+0wcJlrA_2b0=zny4nbNCc2nD{(O$Vd_bCY{FO%zV z(cp8EGGGAJ=gp^f=VOV#yA*rdaVbuESVgjSrLd1rE5PY~@9;?Ncc!jE$WjThn4sE>+kizk>XBm!=)_Oxej) zmxi_P0^1)ZfSppsq+4bLuOEyhZzsQJPr2&@zgk!RLw+f}ymdaEU*Aoxu1ce(JM;t- zwXHbkWhH&^I91rsEav}uv>qn@oC1FzHK5+Je8K0%kV7zP-XqjIAewPBOy#=zpwp_9#DM(n})SlELnMU@VV zna;-Tu)fKZrf4Fn8>oaI#MF`u?sB~9m?U^nmJ0?K{lNu&>caK%BerGg_@M4pFp#s2 zOnyHTFaB9g_uXkFe>C_wTg4ofxO0f_ydc#9t4j;MX^6 z@r_6ca-;nUY&q}-E;U;T)@)Tn`%LAbvxGPruUZIW?2EwpXENyS0cE(;<0x#M;X(o# z84{8XaBO@gX*e?pJ&n2v9Ybv3rtQM}XXz4>pe7CfXT*@IxazT+zfbBh*C(aMoXI;MyMp3oIA1PC~bq(XBobFS7$ig*l+oR&VrL-Iz2c z1(D$1&oIbW8`duU&Np0q1hejr;>Sl6w>0z@Mh%0{M7ZlOxMjz!d>QMGa8w@=0w)j`9i}G=jf9Ecqy)glw z+E9b{=eO~rKUK4vjic?3&*nq>QE$+k8V~^Nfq5qrksf;mj-W<$T$Wn37v`^6Qv})%^MmH_`(y3pJ1`9KP<4bMRqFU zC;;Aqin|iwXT!%ZXGsVQW_jS&JUJ*xT1@uu$izccBiPdV6Z0m_lBjw=g`F4F(By`0 z{DhnVpS}meq=O}-$q126&+@S+{SS1Wn*oxX)M0z96zjJ83-dv$0-kAiK?62x(4R9t zNKM8FEEH&fx%)ZbO1K0vlvn`M%sqh4={uZ;h)$rH-vK{oh@kUf*I}-~B}QaB!wx#u z!&A2(vYq!U_5?x=LI->d=R6LIv~Mg56paj0a(S? z!5ayqpo-CjKRQ;ho)^ADv8X-po6RL4wapz4NNa)51A0hO{ysbw*g;M|<%0$Hh5JmG zE?k#i3mOg{fNjOgfm%^4iaELy)pMqz@}W+y%%0OMuR_2oggu8po1XKByu9J;zYk5r^g*-1Bk8jX(r32c#uZrY){Q)UY#L(x=t9TY7z%;2DD%Lz6NEdeo&-OO4WC8Bvg11ne# zuy;)hfinLzYv5c44veOdu*1SWzF8)Ya2UkaDaWDPv}kl&&I`%T?}Qn)No;{(4!-3T z2FzXah?qt+3iun8szG3o#@1rszb?`K?T_M1? zridbPSp{lUEro=qh%cjAcy+^6a>+4<@vgo^cyE&dr;kAqv;n&6i@?I?;$&?@Je=6x z40i1G#h<>e#otf4@VyOhkiB8Mv3KiP%udlJ-gq33AGm{O_{h?z1y-Qfe-_)n{2RI# z;|mQ;&aul*3;DdMQ&E-9Qs%~~$tbz=4nDZFnJvg3Vjf5sW8ckVcKZu#klLzy&}7*X zAR8A5xAG~lxu!#=-PXayWe4#W_>OO!x0Nu@dhyzCChShx!=U}4o{(XQ;jUDYV7(q( zN=F(c7*h3TVOk-i=;9winyCyW98qSCsEg;S=($|{;uycN%zVaP;|}`>~7a5XXU1l zhtIWvZ{`LZ`tu?C1V-~+_ zGnWeZ_|N_6jjy;a95477uedwJ$OgHB^m0j3 znZ1$dXBe?3%MPQQBsI8xg%c5!I6xztQ|Q{FIL0}84c$5vfCN!1uyL@Z{gI{eNJCeY z$q`ku&p5FfeU6#Lzy4g&{zA)Zw#6xe9d`^TyJla*L$g|$ZWA7vQJG0C7AsEc0>-=rP%plWKfY3o`NC@;Qgg#dVX+8) zx%vxCwvA$L7-Hf1>ntuYe#3~r5-_sr_GFQ27+Ybt7+h_71m%Pd&&g@gY{F~d3}&1Y zGE-T`G#&elGlZOuTmN=uLq$Fw?{elkl$SNExH=!^q}F3jXESemSSZo{B#k)+63}j| zH7?XF#D6ZfG0CG#(857^SY%bhJrSRc6K|h|t~qo0(ryc3&N+SdjK~e{2FqeHYve1* zoJhs1I9GtmmZOmNmIKR@GbHQyQkZqml;^l+CeZ%3ioKe*4Gdyel5xJ1Y>q4?cTJ3# zZ?|T_i!J|g|K!J#L!b!{3j4jXcy!y~BJ?DF-3FdvJ31IHSgvFmh}615|pH zYkpM9K0M|ox@xcyU}sZ2r=B6;Fo!)5qCq+hHpARc2E@Nx&Hk|kpFh>%G~T*06iWNbRv;GVtM-)RX>_MFUamioc{+^EFt=T2gt zublxFw{GLfod3d2JZB6G5;DLSqf}1ZMP2Z`|2+ss(Z@TDS+5$uLhS8S~Bh3+gv zq{4)c)7zU6N9f&1OrHwQJynMP&h|A%D%G>no+aq_IUaOS)d5fRPVxJVra+0jF@6gZ zP0a1Tus>hiW>(E!g}Td*a(w1*X*{*ll8%O+2P&z;jJ;D^=+e#3H2;DUP0uSNyg3cv zlF$bV^|)w@Vibh?7BUncyV)&YEH+=E--S6C~#6N|pHBL)GTG=1$f8W^NY z=Q~@&Rloi+VQ(hD?bMgV`dl!c?O4sUg=^6ny;&f}>k9Ck^pvu94XK58GiAEeVeVBQ zCiZ(eUA{JtZ<6qx{kFydzmU|SYrW=D^8OYX;-oVC+^xjMLYs6S=p+@AhspAio!BKs zpSn93ku9&!GEV32QmdjSnuUc~nJ%KJSmzO6u&t<&VBjl5{>ehEx4hF1lOZ{ zsVAqBOdl1YJD=85^RRk4cKZ)a7~(?LCN=8+NsRv1m@Vl0x?W&3EGAGue~9GmDmwqv zVr7CGsxW`5pm4!|bgHi#ba#rU4-Z|Ug|(W3q`GeUa~~4yN-?Hj z*E-5;EW#%6CXE}^6-e(S^yua?Sa`UFyjDoTi#lDfczr*aFFF&p4P**?K+<&a%NsQM zqaIQ;_oX+20`S)UmBcNb4`n|_z`Bg>B>hAS?1C@w*y}v1D8{3sabkjdd15qchLEGa zXU3>^^s3QbvAZr-39iK^Z*aX zYP{y(3~1znhot@7ZX(Wip~v?cBc^^o8G}%8tXEdxu6LWhvgEL|NF4mUt18Uz+KKj^ zaYwQ54#1o027+)0O~H$NQ^A0Ab>oJj^|YY0oo?UyjNCanozBdVrj{X7$)-EY!LXJr zblQ*v*FGu7`bEca`Rt!m^Fllo*(+qo=^x^;?;5w_h6foBHWD%i2iZE=Pnh@l1X<~N zn>NPG1fe(HQMtv_nH4uqP;#(NP?b|B$Q~#Wm}a~n%7^pl)|nb~o^~ZImz+n`6^zk3 zCn-T|gP6ce8%K`?cwIYNhDMbigQ3rU_$DLwbNj7FuL zpziHQ=|#1VFpqy6#_=2|vnr7`N~h9)5qg5nGY6^t&*w1fj3#;}tO*ACi&b~ zp*y%^>!vedLML<8z5V#GrV)8DUd|niG+`oJwy>Rko7nM)CE#c65$2f4MmFh6D>k2_ zLE0B{xk29xnSa9vnTx^gY=e&}c4L;}4Wp)&up3E{8#MbyDhBCKA~spjXQU1&Iuec zoJVGf-E17$Qp2joA7mfgTg3d0t;CsqVWi|vDF4r>7%9I}O7c^GH5^p!=Xh>D(`cG! z%ji#u08`h!;4ixThZV$#uu(rYGwqu)IoTE4Sgzg(HnmIO8vc=F&MZ&m`h9lgE1ZAX3st@=8oA-8ja%gKa*h>*1xe!tCV4F7FM!P z?dv$hliWaL*By5E6H#_f2p@mwi6@h%&ESvE6>H$PUgEU5e&YPMb}?h;8pXgFr!M~# zUB#$cz2|?FT0J+fUy0u}$3f^JN*av|z4aORH26r-J$T$hT(pT8%ZDFAb5H2y%-kc5&Ht@r24a-C(t*{uHZ78OeD!;7 zLYO-?-m{E36^1+rIxtGpwuJ+|GP*8zSwMCg=%V#XFq!(9tO=mW3v1T2`H(}|| zUi>MtS=_yeU%8fFQh7~3?ZK;M3;7Qfw=uMDlv$Dqxwn1par4hOvo&`2*vAr>Bei2C zU+R!AZ@q#&quAHX=;fZ~Du3n_WvJ4=N*@07shdwP_(Bem4-A#J@>gz zr4o`PGei>^nW<2cG?YXN?Tjc%iuBy)oJZspAyQTp5!sRyGT-~J{(L^4o^fB}cYVK( z+@zMdoM>QyXhEGdudUiDey6)pw9`X4f?Th}FZkpr9zA*ny3}Uo;NCrzJEmjDtqvIA zmi=_Q5d6bQ{AaPAL#&Yxq`3_6$3$;X!=jmvvDZ`_C4a*m{l5>V#_uXfsgQkIUeirt z(^ANZmUg~6s+C;RbU-Us1<{Y0A5rY=M`-$H3$o*92tG6HDqIqJ>}4S{978ikJL*oJ z;W$6slDP=?=MSD)fW~!^W2V(2$2mb7jwx-&s2;zD{y4Q9ZUi~P*IAo{T>J#Qy5kT| zF&%;n9rj^p=%y=Oyea8#g|bD8;7`Lu0|%`5DBX0jYso^;?Hhwq>h1YD76w* zzb+O{8UL5RcRZZ*4ZB0^cRz*)#~+|oCdt(P-ENw2!~whZ_+q`hG;CifjXUN2@RzWD z@Va1xjqjPkfureEZ|MoLWUX*_qC1Tw`fdP$nS=vc?Hx@cJsk_vDfoTc23P8VlC4tI zX?bJKF&kykA$Lu3aI6g2^gKpc@n$@4*-Cz2ETfApv;^nhOc0Tg@N!oo$@DQHdaA2Y z#<(eLyXI&*>=30j2lEBSXC;}1o^U!#pFsQMdZ-Uo#7@4k^s(^6c#-?bzY_Lnu@|<` z*sa=ZYW*Tw z44FP7o?CXAZrpR6bpF20zh7yCO>>HH#j#vGe#s7AoE*u)%o%+3DJRJBQ^q4US>XNA z`}iHZ;y6#04Ai}QD)CDEt?pz`7%dTLofogUnvPECd^!Ri`LnvjXo_RmLB z$umS%&(AXX^2w#DXz-QZ$(w4Mfc`jD>faeiUWzi1R4U^B zTr@&6H-1EV9XlOgwqra;`4e83EJY5+PQ#;MfUYKk;p62gC}T_^*>UzG%4xN!X*@rT z_B!@ZaqAI&%)UijaFQZ>8e@e$?Pud+&7t5~u!;Qe&?B8Hq~`I-AX26zf-kzCkjWA$ z2pnhvr2UnwTcbn!l!?IYSwR23yUCR*By(A(S92ZziHJgS8ah^EKo`b&Fq`Rej_H4^ zY2T?-(eh|9Z*;p1s_PN^Q8SLF#SF#2r_N#tKX0;`4<0=2>g$`xhoo|3&Ynzo9pTomOFV9v3O>z+>9i2{ZU) ze6M0Q6?xpH8@Fs=!+$)bEs4tPZ&o;7TN1@ghHhl>eG^%o%XPBBQb0{i0(Rqh9#;LP z=9t{7?RfjyK~Np71Ebt{Y6(Aig^w!q>!S@!_r)bTS4xJujHz25O%mSkBKYxKtqz+nh@>S)+0leHm2eBjt{VEzCY?Nyak3Tpmg_a2X7=vy3 zw5QSz@`^tY|5lAnr}<%L|3*4^`T=HGSWmZ(UJMK0`;g7qZ^^U7@$A4vC6Mvm z%>I0wizf$YF{21O);Px>lwApLbmR`TwZBLnOnbsr#_vQuj%s+d;wG@3cbupPX48ku z8i>qYfzzkl0fjNI=+iNSSj~t*R7g#ci`(qjllhxj?2;~e#WsOe&-u(E59&%(!^AA# ztc0#|JIa@ARV7}pKkB>XBOCHqVBp&I(#dd_D|N1dK_SaX?SKX)FLR)9RxodP zx)}MdErRO_k2$0B<@6;e zqDKlo8 ziZHw^f|Z|L$Y#S{s18Kr+D`y@JGta`@if8>G9W4MyFFp=l0r z=%KMCyt=N9jiZd<#lqdN>-jGfP^gMWUG|Wq51A{uI$2fHbf}a@4Jjut91l^Q^D->d zvy9maZrjCvM^NvxQ*_~C4CnVHvO%64z0jCP^FR61Wz8MLKO`IWnG2k~r$2a=uLzD$ z*a3Tc*5gfn-_Rn(gXnvNCTd!CNc8)}U7oom;K4ynppultUiWWfwLe>!`HkDqzpDq? zP$RBqv^;wkl1^4Xtfr4@hA^pD@$8GsY)1FXu$^t|h1<3kVv;+WG+*~2(UpTqe&BSr zCb1H#jy{8XZ>Hcq^1>~~YX>ya^ennL>n2h0D+iUdWZp7eEL`q73wNofn8)>EcDdmk z%j{Z3=Iuuit=wf`n^mb zu!_+$-gfxdhS~JgvrJYwQ;jFa&VVst zOW4A1EBIlP%h6AeBcVS=vEU4C`mpf?$y8QgmBp9&jW!vy+b^0`w#;YS10J$r3-7_9 z$J+Q7$%o;YT69D&(6GEVl&LJtl9LB=mk0V%E+zW zWJvoaFqT&P;iIvA+;FPKj+^O9jIR%sc$la`de&8v()|kD6_4|O_k1JMW@ggi;qO_@ zfjZW*>or?!TMr-KHy~ZLNVwm`5t;qjsP>yVnyade0_;8#sb4mrJWv2;|E9Cx-BTow z6bwl0CSCk*-zR8sZxzR$ISb;>qwwd19PQ~a#PHQ13pB4PTDO5mzGcA1A;yAM5B9?-%UCu?P|_eojY*Dxulmlpxi0C=T5pN^Yec zX5Xvtpf^k2a&sQY(xe0t`WIymTg!LD0d;FS=l&GBW4tXJ=D$^xF=ZT#y(74rwktr6 z)CAg898A}4SnDb)1AHBIf zZQVm&jvGaP+j@iT(_%ygKb*mbBzDJ2pDnf5qNzKiahQGw(&`z*5~AIh)Rqj&kLUp- z=}UZqg%a!e=D>VzmcZ_v^KinpOx*bU2CjB$hcMAMcyTS4OK+OUw8Ju(+4Oj7c1RQd zs(OZg-=EL0$1AKJxS#!yv5~BNGapB%)R2;0eKbhO+THo+OB+XiLq!5%QbnnN&L|iT zdMnKNF^Iz9`?m1-ekJ0}UsJP-2U$yUGxC`IfJ@n|4XR=tkeTI2&vcdYa$4!wYtl9F zpB%^r@J#s8+(Zfr_v5p}Fnu4P$-I}8u=O3Q>8mM)yzje1_~DBzwsY1T&RFh0th8e^ zS&pN@(aDX?YCDN{`|iSyBa~3`-d#eDH3QTX^q_r}3ryl)BF!h$V0Uf?y{LHrI^Hit zU!Z`#3>$`rX(mJW3`Km^K=2(NZ9?G(#=^hg$D&-rApD_iGM(XGPnB|xP}kj#%s*HO zF5fW39TQ|h?(%v5UE(?z`fCpgc$U94pM5FK$P4M76D?%UVjZV%ZurI%o>h zy>Gx4gnfrQriq}DiyfUZy3rnQeI|O~0^>*f(fJ<7p>?w@N_l8ZO@>})Z$!KCI>%j5 z@-L1YnOFm@=eL8+!Z!5dUm1Lu`UCZbiQwE-HSF7Nz^?UbF!Sc)=lKbOO_1d)k+M=!QkP{-r+upler4ByRYzX8W>DkK&W^T2ZAJz?tA(O`sE)}v9%t4 zQ}+Sn`kQENZ#X-3Yq(_8QW?s39R!OxR!n)=Wws>i2-8_Ih&?MbfT5FjnIXwyieAtL}>y`{lpXeo-1x5ITtM{rsM;EJm{lKf;}(lL1_ zw{;*CPwG4YvxDtu;D$Wrv)@T#7BfVW?Y^AR(OGnP{#mrN_aGZT7l~a)H?uEhf|E6a za@R+M(2>uw*f)L!J(Um*eqYphU)SkmsJ<@IR#F7VmX*9>@-tC((+Vz5HA6h>iE6l9Bijs)^CQL6P-|)- z66x>ae3Jg~X`w6F&cBtkaeFx0qP?2UnYoLc8=uU7X|cxQ{cB*F{c@=7Jd^t7%673=-P-}+vY%f5f$(ZM0)gqBdtEgH|oUEYgb zN_23}0%evy_ycY?^20aR#N)OeDT(O(WcVjN0BZ%#n3J&+wsBsCtIDMW9+U})t2RO{ z9W2S^7D%q`{=g=jnFY3zu~3=4gSqU?h48~t^x(h-VkrL!Ze~1%y{qo9_u6%I^zaI% z;B$w~9GlFlpDNMwgHq^YV>QW{Ei$yVLlZ@(TfoCb@^DelgG~M1Ltn=z)6aA7gVkSM zNZNZ5xdqw5gygm4k^2vRgm4S+-##c0@aA6y*=}q@eGDhhp>YI51HO%{o+WY14{0+W1BsX2CK@@9> zo`qc)WPL$wcG8hqEiHxNX(yT3L5@%B8&CY74ilI#!hOn%0c0?%0Z6MeoWq--dB=0u zviS|PMCf3z1JU&B`83AW^pf75mtb|RpXFWIK>rOtgQvVWindWF(bBxxX#W;dsxY9* z)P@^iq#K1-F3ZJ1CyTMsu6Hc3Wg+UmH5*5s`v~it4#VyFb?CvBDKIHa8yYvxrDRed ze%G3e4nfIvL z_^*iUM?2_s%}_K%*#TbeKE}5z{p5T3t8BfhCsF$CO!Nm*@P7+_!KN??npJ%P=6YJQ z*}f&r0TO+VNP|AzrFoh4w8rrcEh{G`Hj~)i3{#7@VDkGH-dKgfJgiesKxuyfu_g zSQ$(nf9*pbzZRig+yn@19mjbswBas2$QFqncmS&RA;fJmTK6s$=KmRj19igifmjPR z!7zdS)X!&Cvl40lg=W6%!%n*WPY~=A`qhGGp)_D!B+|ZjpPW2hiPmr#_Et|UVT}D5 zZt&MbV4kLgb-Zej&9oaZ?a~QwGh0C0Gek*6WxXT+OJosy4Nu?-9 z-(_^vS^=+*7v`sLC*ePZeekGgF!P)ez;uM3blF5zY%By8pR0$!@P*x|aa<9cZf!%8 z#y&?;*GA!ngkVUnv!VyPMzDdYxzxti+&-pDU!*jB51-mKpXi6Z6P0O=Gl# ze8-(Qy7Fr}{ds0EQi(nVyDUZE?OOq1mlh`HcEc+=Jy=vZHv_roq zOu{k4ALBW`UHD<>JF=_j443=4gKmk_!S9-1;+2l2cx*dIG}ex#Gat{Q;^8Ohs+A@z zDzS^2Z74(KCT8%hO%>aW45jWdviNNQ1D~g%WK&cP`pvn}F0lbE%o>Vr{GLHyPr@wY z_hTwQeiifRwZ|)WJ3AKkAxG&P#J0XRW1Yn(m|BT2+kIm}LL&xosjhM8sP-~4?Y=jR zy>SKU*9JqW?^(z(w4m$q_tMrk_TXPLMLc$?CGu04LO-dw(CXKJssA_+x*~!Dy{+{S zyDtT8S&_)TX=k%5TGN1$Q&fI z*4x6&*ya{0{nSsSZI;qzcWZJ?LjxOL{SO4UJknII?=Ja4Hq)wwtC^hQd8#>T zoMcA$2uW;xKO3EJ1U*5&iPTyi`NgELVa49;?#;<`?8eLJwQ3`N>ab^1K3adGg1@*KvQEi2=>d-sUidL3q% zcaZtnwyq(fnMbU<)q~C+V zJCt$SmS3dEb2uHj`Zu|B`x(z`o3PGofzSKEfk@W`@yg)Eul75_zU_)0`GI<>Ot6ffQ4++vpN`q-`1nqs+ zhFYDsi|i+AlX!U>6w8uOGkRq?Xend}a>W_yklh)rLl^J8mwl5uf{5*gWY_U)pqMD@#jdRlKitg~;N!yrt$8Mru8ZUZq$l9TS7~!&Eo^)N8 z0ZX!K<87;sf?in)*|PW~TDyKVc6X#)%!$eDK(i6c80pIPJ~6~ivrF;Xl%w!0UPm%~ z;uy)5zgV*B2!iHcXV9D7<5<%=H(=iT+6OKZ1>_UdU6n$hAi)2aQ#g(kQ;`S-Kpod z9@tF^*5%t5zd0?wm}$z{U4Ozoi!`N~LC?fip@Oma^)@UMX@{lbo}z7sPH|0R-Nla# zBC*Hucx=ZRQ{&Ic^ox}O+tGiPHmFXKBt9^e_^H@Pv}-ooN%dwaL^X@f~@a_y_3-zP(i8wjU`y;SI*s z)A0+G#d}JR5WD6$AR||^6{mkw(vrv29pf0UhgtTFIF{z~nr7PCvD8KWMAKZz{bvQh z6!S`&>Mhh}0^`=^=RRiLQcgXM4Pm046&pq0QybwnIdA3()H3la-uEsGDrK<6KpbqT zkAT@9UJI<$da`lpTWG#{f%89mpU!!UXmUg{>@Yb4*_TG+2g~$v(Kn7=zTZVJZ|Wlx z_z3o)z!9Yy4;2yByrMaI@-LSS-53bFs8Vr1~y8r%GRL>Kj@z@)lZ;HAg2 zbviZlg4l+?I(s=$?W%>$g>Llp>`m;j)h{}$c^vzwBoB?@8cJ8pbq{| zke6FFSZuHc=~NNEt=EVG(+@dJ@x4S0T%_3F!Mmuks|?Qb_Mv0#?m`AleOt+asN>^+rf-=&6*DX!UIrkzzRMuemu%5vqBrzUg4*W$>HYpYq8X`56R~1 z9dy)~I@oQu0^U2Vf{Bg7{_b)o>VI?>R8I<7x}c-1$1R;HM`qI+E|h-skPy>r+N|MZ z0$IJ4g3*akcsU?PB&UKP@}mw^4M}9N!2*BjsTS*(p95nHYGLrR#rS#FYhJuz5mxwY zK(E|L5xlJ7beL{8wJX{~rB;sz{=+82SJ3ovrnrL@G8N6gl~O&{}5;hYCko+P&;G znf7=S5r1prHm&qVwWb`^to5hQVk*(b*=|&K_8@Mi)@tBIh$Ow2fuGU!#6>rcOw2YT zpC^1^onK{HvB7-iTJ6ikqFeNtz>Izugjnb}YwEeZ5L}0^W|k|0S$qCWYI&d-q_16~ zSGMm0t0_io&Dx3huZagNdZKZby8-C8F1`5R1`dZ;4}&+kOr)b4N>>XYf6oO9G=#%}65V<#CkMj7p$Fa|bT zC~?{i0qE*+X|nHbB>mW#1Yd4UgXF8~)Y#1&->a-8+#WRoj(&*WF^&8jmdeMk;mEe9 zrPM$>1NEhtLUhCzlx$`U3x6QQot%i1iaKb|=}4xYIgZVJb(M1a_fXU|jRoB{ruQW( z)VDg6|Kt)NnxL--K^ne%^F??1xKxk*ol;He#u>5XXB?9nEytF}B!hkV2>!*Me584* zfgTVXTNAYgYB^FSJnFnBOUjVBLO`(BX2H?}?q3}9y9sCV;Bd&TP&UODos&P-qyD)&MZ`_Q$HXw4z<-r?*jnQ;e9jw1yVP6A@5Q<_{{39k zb94(h@8|{Hal+rVk79Wl33XG>hQrYV;-f9c zp)a8dw47tn{P*+Wz@}^%bce&Ko22ml8J$o(e<;?B-9m#aYU$2R%_MoLKKoG}#%AkQ zkvmDd*~e2UY*g_-UfHz~-o`AUX5Wml`|Ld0<6(dXdv?-SK@+L@B6&2i#|T*-`VXB* zzJm-DvmHYAYI3g@3_}A!Qcw%#@S#`7fbp;s)Sy!ya-ycfjFU2=%C8O zMccjp6`hig2` zoOpzL^}GYzMw??V=k;{>7eBV!I+P@<-lZD?_M*y@Q*m*#Dr`S_9!**>6}7*wMGL&{ z(4eYF)N}nfM@qBMmTm;kPnqI^@Eo+^T@rY#31@*bZCTP2JE~B+2tZpAYRo=Tr|+rU zsMP`>cI`d>tBfiwTBreSd2`UR4fUwp;SZ{}i$iz$_pm}F+{rD%_&z$sUKa(k)R7_h zSkxTEKBcoi&lW*ca1o3eQ^Q52BvZw!i)f?SCX`T;4KwZT0e8=sjT{yXKb|VHA!8g^ zI2)ii)*e{bbBLLnjBA{a^W|zOeBHT|B*|hnZdf0R?`?LYQo$6qn?=D)fx{Eqq(bM6 zA3-Pe9wmLRFntu&M$MKO;MV<8%yHcn2-|jyTIreNO8H=cyWd5R*xzP{Udyr>bEn}0 zUqVr1bpSO(=_Ks^AeP*8ou=5l{nsBr7Q3f$ zo1{yS*Iy-e^G+qXoSM$*L|0IQ`rD#wfgZ$}o55_?MzS6sH+;I~A1RkIBiFju(W|lt z*({GBR?{eW?R?D1My0c)r^W@gBqY=2g0Dnctc!p5j0EFeM^ZWMC{%yl4Wr`X@X()s zaM>sx$M5yVQBy5&m0W|6!;k}&z{y}fcoGx7pVL!Av?!;v95Q) zI4vU$zmDXYw1p{C`7a+r^Oq5e$x<-hWFM{bdTD>-S`iF4xQTrP=f}0r0^_FP08?HQ z%Kmy}gWQyOtadYy{pNQO>zq=qDcl|XY1~J$)PK{47b5oJ#!b=d|=hCm0w>HSBJ%-iN&XDb}m!r z_%E_JBc=`>&R>G9JHK#s#rrvhKrgN}F^)wj@8Fbe%ZP@CCb_&e5kjobbE-P$$no7O z?A4{q+*$AS%qcVwM#v@NoU*qxVl4u_&S`kz7Y#m*>#k|9C4AiRGqFG z4c&Lf6~^{+bdN&-T@qvt-#+TWU%_i73-k`Es9`7*yndXTF>6^{zi89 zu7bq9Zcr}Dz~(;7;eJISol+i7&I~RgTSLrn*6_h>=6E0Ar$u7*wj-=PQs8JkX=3|s zTjHXSeE6`KK=|VJ>J?w=CJfdLBmDUj#Arq-d(-uWHl|||2m%*x>k+3kx zegM4%&tatUL@L^1%-na?iDOckbnnkah(^;{_I(5FqsmtUi4bOY)t^nwxgxSurgT z*?Aw_B0ta?(SH9@$%+8S~%e<>Z73T&0WPw9= z=fNT8v3Mi0W#Nv;gq@K$v(?`%IX)``Jq7$1zlnWS(j=ELXQ5g{VeBiofuDepkSz!Vfe@svK%EyRpg>26f{v zijrQ=z;#YC;H4_`ZeQ(LE6v6gD+#{+=>xcpz>rtljb(yAvyrncu--xy?8_&ClZC0I z@w^yXAP?Uj;UV!|mBy5;K;q=^kezI>pd(*)u~%NElF=ReAa#!ms}EAaE3g|3&rJa0 zI12?g>0`MMhaYXuAS;qyvQfA5@kfnhc&(9&P5dNu%TEoe&KEP1Z%6-CZb_o7r^a-2hYdEG~<#C8j2o!Lqnbgx^bhw8DyV;jTE!=#}jm=MHRn4*h#BZ1K5V>B(}&v3-xZli1S-@C9mHc6u4rO z@%wu%@aLa7Ho5d1$FyuG>|r#yjAd|Jc@adqR?=vnT}=6UI!l#)L6>grr2)-jkb4Ng z+x`-2dd&;V)bs#&7s8Mf6KHy`Ao=H#z-DVD5w9gSVCp6ig`>AR?mD64XuGBhUwe2Q zq%LVmvRU{_4>KSEB^}=KN`{F8rt*=g#3QZmBK5xf;_5&oS z#6;op{9Lr9%=Cf|=KQseZvU1rVwv8NMB0qH|OkagDb_1wv_SoeL2~IyppSET3CNr|>8<~T2!-e%sW11=|v{?sU+YVBduNEX? zp*}g%ei}X+Uxk|Gedz96O5Lsxht@@HAaimINM-LqZ6~I3HalC9QKK9gUhhFWl?-8g z#y2n$yks$dJ`v}ew?ISK)0vQ3(3pW$>ZLiFNNn*j zO7GAW`|O#2{w-MfV-S#SdQ3@wC}}(K2t8?>$$o6UO*?ihWu5UaX~s`IvUQ#s+wYdi zJlZ-@P=yIL-Q@RyS{8Ysa?yu0bRgv6F6)~v@KY{KwB6`#& zk_^th4L9sHNlSMQZ|A+3FUTn&NgYF3@PIi@mG|d<)d%oT!)!siD1!xd9%K6gwzBVb zHuS)y9W?IJf3&=b(UFSVIAdp3feM;Q+*L}cmSi3ch%mz8=0{=X$sDScXh8!y$MCDe zWzmoG<`8$}BszA;5RJ)ZWX8=wY`n00PNmoA)#3{*%IPB6x%Ms%Up5sUpKK|uBM(_x1&^V|O96#G!V8tF%7 z(9iqr@L#LJNX~=N*AF|Xe8?FhUO1lp%ynWT@76)}aXV_XQ4aU2&myCLIkB0VkLg0G z`SjwNKZyS}hN-vv(P5h7>4`H%bbIArsJ?%d_DacMPBfJqU06h6$XYf}m=(Tm_Q$Wk zOhq*#Fj9I(k#WFmU`KW7B~^EN>TnC6*IvgJh#yn`E2p^gudb1Si3^#zk0)J_kpb)W zS>j+L5uGCN5`ESSe7iSZ>``MeRKC}M(&2N6=DR`sgEvNaW>YmDYmq}7YZt(TL6-FP zIA|pS3M?%h_OuWoz(?(^aCc`F7ys zXiB|}cfp9Sap3o&ir&jHgYwA9$n?Qws(C2jg1F@`AC3f`dGTmy@^OVD*Lrb~@-_PD zB9H%Mox~10RpQIp>JTX(i^VY>FjIQs&o?d=Y?`os7~XEU+C z)?%Svzbh`(8iv|R)xkN=0Hlk3sPZJjH~7t>q4L$VCU_8ZMlccixL(Lc=W^mTXXvm5 zH&A!K51xq@g6DHP(lzgc&)o;`9NEox5twSJ2hDMj|()TdnUKm z%8Kq@e}!$VS^!-pCSZ&2u}D7#V=jCKv|oQ;8$0r<>7($u9X}DTVD{r>pf~M5diPtV=sE2>IyN+9)cM&WfuEhJ6XyD3) z+IT686kE=cfgfRMbm;FtoVfcbcv_mV?Lh~?dFL@WwdFS?RZge*>3TSG)fZ?!y^Oq? z{a)1kAWl4D$5f=-ITzR89pk8_A3_En?j;six0Bgj=lSl-icmI4U{RgvCVH*1_?pr` zIC91vj5&%oe#;;u9ORK-VmzO7It(s-cSH{M`@!~*4oUB| z;1;*-CsOeWu+y;`q<4;IMc!vn--%PuV;=;$U$%nAo(XtouLAZluC39UJ050CF~YMh zq@u&;o>eQnNknfKz9PTzVQ@WM3OPR#@b?|x@Yl@UjyL2ikBtE@+p39Qe5}JWy*BaRo#r??70n^T4i=;8in&m5TZ*;>dZLSgT0~7@5V>u+7+xPeIh9=6?WsEG26#C73C(UTjnl!qlISD=apbjd7 zcOZ*gTk5kdl`gw{p6R~*4m<9j;?m!Q(8~LRk=5!BSeH2it6!ZD(=OMM!L8m9JA|7WcDa6XLtC$NIHTj4ncr6@m9 z1Dm}3horKaQHT6(baR+Go;xufvKHJR>fsM4H#G(Rx@2*h9c%EW!u_CIYKl#UzQOZ` zZx)@@pU3Opy1`XD*P-ZcKXTGLLu5D32#>t0jzhZj#aYfjVX6N<^pgz1w?vx}O$rfj zY>kA+S#~&ppUCsgKcU;|IO+G>%a5LQ3i{JJVeI`gsCUg9bS+qheve3j?(&W36@5w8 zj-QWTM4rSA=~cWzY$+;qb^`-}&oL-X3!QIy3@_9D@l?fg;IMN9{xVe?Em%IDPZd0< z(~D&xMwl;+h`NO0A~NW@O)vS8_s7xtKOA_x5rf+Ig|MjklX&4Zb+lPU3(hVd%KbgF zmZ%0~kwsbiYs|lD;pWTQWZgI=?o!1o(G;WSFg5|9uHds=8|~vSJvxZQN0Ld?-a%~r zvw!gU?Ov#O9}PywjNxD1WVoL-1brWT0qOEO;-B6@{E}_DkHzKBb@6E(A5b$|Ow9uBh_z2U zU{LVD)30yBH|D1EHg7d(jfNEJJCFfs$5zm#YAR4NXB9N9x`JdE$l@7tTF^XvF5b5( z5|6&Dj?JA*@yJsPadfOa4qdqu2XUGe~4a}GFc$Oas|d>hI-V@aQ%pNlHrseqM~ zFV}eAl|&8}Tr|()#lGtj_$8`6ob%)$QZ?)wpJ(xf&tK6EVb>z@$}S5oKE?rWk1@qD zBmKBL+g3oN(G*aWO^1@l2VuDx0;9G0kaM9Mu1-FUXTE3#Sy72V*!T&%yzdf+A`Luq z2gc#G^0@Z5J4huZ;kg%FVf6?j9KBB-AE@0XQe1b8)9F@4=}CGtWK9`)Dbvgao~Q!- ztxEWtYATj}lt(NF9i-|0>;_J&jVb7rn*lTPwzEA!6*~J@4fBjl= zo0#C4f35KWekt}3{S5TK^-z9qBHrI)jTO)A$MNOCct!U_==vcKC8>9i#eEH;a+h3!$?uc}@<9r19h9uOT!u3AgO4bl7ox5iK5{gU0^k(Jx0k&{WSy`M$c? z@y$})aQ`;Gnz09WYl`vf!zEa?ybFfr?1yXx37mcqM-Kh;;%xUw#NNAJb2%o5VZHu6 z?$P1|IJ^bnb+>Q9w#_BL1-W8{9fjf<)8=qn&Pv0?^?jVLQ6ch}I{;jw2d;RbgKtcf z!AqyU;(k4th^Nd-z-I!pu=|wx=-@v@I-`{F67BuStM-_9f0!!dq~GO_41I%|MOiq` zz63?-B=Hr|6i!!uMWY|cgF{<3I)7*mWcFL3n86Q0Kk*36j~s%xUVj2A>s%p0cvfy$ ze+N`esf$}s?w2-5(YKhEO_;wAL>Xf zfgkSx-@pDt)K`r#6y#^LbU4lpX9F#S8!v3g{7k=h;qPfQ=^I?CJ zsQn;$R`JOi7x3q!P|J(%cr7Pg=kb{`s_rnFQYB)SF9E;AbfT+#Bc#zIt zd}&ZJaX#JzRuzKlY<(g9JjWGoe5xezWz+H0lkZ60WCB|S?%6tZd;IXdJY;X#hEt`y z!RvJk72&&(6?2`F>aOJi8&AXVZJBV!#{*|83ZEjKTKeSQB|6$`nt0vb2>$qaJ=hs$ z&3CVSgVg=U;kZX%_^RG3#No5xNW0}J>`I5C%`gr(1rCQ!x0R5z{{vfTnXKE2InKZpW3y8$}NeJ+idrhx6Y zv{K1>UtklD(D5NVas9n2{N(K+Jaw`^ic<0=k9@80s=4NPvi>=!_O2BaMhnnwIdj;Y zHyX0j&VZA3KYnEX0grwt1=1rD;N%b+oH2Yp(^je{zm6L~=1yboo_7sNw>g3>On-w- zTQ0pCZU#Su=SI|JCvd;6?pS+1PIx96;$x3@BL`a%cd#~%>4Wy(}@`;&|DOH!ar6gttYfWpuj@L`*9+!!>0?m& z8D?911|+yD!{*eZL}O3~Hr~HLEbHHpXLr2dab*QD$#Mj#&xP={HJfz%Z70$yx7l@e zJlt-sL=M(!k>ZVAcqO=<2-es@?vppn?9j`2pIRpdZ&E&a<1r&xHB@^G5lEWqw ziT5cVcJfkb=F8Su{7mBsNBxpT+OcW#J<<-bO%1`7=M^=z2J$ zSVm{s`p{R zdNX5&0GH(-Vk&;jgeR4c$ad|ExG&=aUMpzkzAc}@qc{7(vqet8t)m3H^|S=P=kpF!Sn0?v;MY&@IvTgE$LoJHn zq?!f$^Y4S}%3A2X;v%T3*Ao;5yoC*?+yv4m)(J|!M+#OyI)JOpY}k&aF1YiL0aM

i$QB|Gl5{uSbygG=pfQE`of=vC#3>47Leo z(%=tYalFzG^3^n(>@wYq)qzp0Q133>xp^C;A6P;20ug?D&5e9C_TnfS4d1TJg0L^n zu=T-JM&9HOhDAn`{Ea=Bv41qqmQQ5OpX~yp`7*#;e1fR|mrNlf=}K%NafFRrLZ9Db8bg6IECrO?lvtiX!K3>Y(D8D0ur;0n{Z; zz|Cs~lx>&=$FVqW@lT>Xu?7BWDGO!>7z=ck=?dQc zRuo8j)`8}55s_Zz09nFd*!QyvCM2l~KE@mXm2p<&?u7+RzUoGD(=?Gp1*{<9#{V%s zSH=LbwZ{l~Sv+g;hB=@)iSi@Dme13kF=-AguL{j1x*%M_;&d)S}#XZq@?`0q!sfxkqanINfXAH5Y zX&ydwe8}v{EJT$AU0i$o4W5Tq_NTK3RKB$W!|_V6x62p&<6pvzr{CdEV;NjzBiN!n zTC{(}Wwe#L1bORRA+cmW9M^h5>dQL$|1|e;oaO;WBHNq}9KD4%Opn1!)&X9v$|ZBn zQo-e!Er@2NgYSr1uzPg~KCS2@>blb8+e8&QcAGuDs9b@}!${WL#e){uEhAXzjPtc7Q9BF6TXVx}Rk%0F^ z@@=Xe6K%w2Kem?fo6hTSto2t+a*hhy8GsO5Vm zXOLfCxO|YTkYgnM$3ME$hUcvL6emmhh!PJd+Zybbre{~_vq`O+@yYlN45KRrw@rkM z_J3>1{ctNBVafUGBVRD3l1pI6p*ECzYJ*8zhM8qIuai}K0Sls2i0=Aua_O5lR@ym> zpIa)5=jN2%1-i*~CeS4<#Ql*3o&Wa` z+8s&aPwYNPH#GK9ygESLw;L0u;6MhNE}=cMkv#qTjDPp89iIQPot|BwLQ9lZ(v9m? z$k@rVSW&}f;#yS19tcVlEeLsze!B`uO?fMGQ}-qwbE<%L3u(dg6WySBcQpL;UI5V< z^P#FtlO`)@(EKNdahy1iEj2P>4i1c9a{ZO4wnQsG6WuhdQgfIhmdKb(nu= zLWNMnt%MxaQ=vLdA#_LlV!V3%DANpI(F7OL(fbYPU;8NRa34WC&1I>$e-JN@naeRL z4zu0w``DpBvuVgtPmTo;&-Jaiyy)d2@@k`qe=Dqu|M902t{At4d+%B>*O#uv_X$67 z&U7uzo`0J>7JSFdmyN7KWVoM24NgNb; zBvitL{+Vtlt_KZq*7SeG@A-J}nWsmlRH}(5IUc8hxhx4sG7%Nco?*dA9?yPv(WGXjXIN0b>6?F z`7+Cyzx{J@A(s=3o!rH)PfBLXm=#Q-j3tgbJetaO&!e8}B-jG~YW`a{KBjw3rK{>b zV7q1q-PwGdG{#BOZI4Z<>Q8xrAH+ey+8wam>O1waJ4w5rwb2&kJiNaofjwhvNGtNJ zi00n=%nxTVW1tm526{P%%3WD@SHA{z7~hQV?s{Os`ePhtx{36j+`yb1R6z%cFp*Ef z6*RN{!Jiy^6@Rj36e3G#y7Cn&!Jef~DScQQs7LV11w0!lhi@OxrH{^zp#wRC*i++3 z(qqTd$$z%fwM#G3ub17h>Bts(Y~6ioo%flRYd*&_4H0+_(?xp+P73#05Ab6HukqF# zSx)uAkG9;@p&FY^F|u!-P;JZt67=ngNXRcl7Pm0rb8Ar6eIkxq{(xNVZQ^fm6_d)a zd}6ic29vqZ40@+)!5td|+`p9blH0XmQRNiUoBEP57yHn1OHUZDxELD!_2I@~g6Qm# zHXIq9kH^%xcjEU3yy&4#_1{gvMXMZep#C7rvX%Uo7dP-Hc%(4NcSe$h4VhScb3M=3 zYZ!aCjHRWXU+}?z6dm#h`rm)5bbi7Bp3qIgMSh>zi0lsbo$5~Z+rty=ZZjX-OUK)V zri%lF|B8Z$*(7yhG{TV4kLH|0bChw*fo{|o?qbCo&tpJ<5p`B~0_&r}@aM@y7+74$ z#ORJCGxjAi!G6WIO}l(VQ|4HTBzTG_YN!=CB`sh-uKg%{+tANHu(yzTdZv^S?D$jV znO!0huQ`Z~+jXe!<8kzGoHlqAuq5!+3&zgL5{{-yV^mlKPB_1g45ydbw!S%xA#Drb z@2vTtX5$KG)21;0{!At1<7Z==P?Dy(AE2s(W|RoC=<}~vsm9ff)HBr&YusD#!rHqy z$Bf54w-(dI5@B?TryJ@@_+hpFQIreZO6;}{GShow@#HI62wf@(GDCet?fy@`n?(YD zgY}t8n>+)$rplEDWDL_+_20-8f9}1gdV&}*204w7+ zlHIemfk~1KMZvHv-8j}r@pVJ(HC`g5Z}qO*pCXoQ8to^S%Mq}})^!Wj!+ z$C6L@oN*qvlHc$nk`%1AWq1 zvv^uWji{nS2k~PXNOtldnU!=N>U~78()9rNYL3M2W<#pER~svGcJlLGWN@EgAv5=0 zA~S)niXEp;kZgk}^4uXC<%c}jkrBDLd_*JbJ7PWKa7BW>k}`ry_}^s{)_!90+Ygd| zzNYZ9Op+dVAQ)1!jC$Itz+Y!0_&fZLtgJ6)UHe8cdlKqM*x%DwSK~>B_s?M;kCuRp zHP4CHsXZW^qDULN=HRMvokU}e1XFm-g8$!*PW~wQi^Acj??uLLZEVlu0$ymDJ{!59 zgZ&b}p1;~whMzVmo}Uw=#@_-Dc=1UynG4O?yoAlBtX+kG@o;`$HSad^55;=$pymP3 zP`HDydfJjNu}GTlvd542?_d%CyGA-+S2m6BDu0weQ^}kE%i*J_uC|sr?Rc0cI2tEX z;s(hrKhQ??b4^vl1zq-+{miNu&2~|9ZUSRnF@?84yRFLLbuhnea0=gL?%b+~k=o4M z5odYTtG4r|YRm8?1qr_zS(!G9PE>Rw-^jV7ou2NR()!!GAqpf(ckHg=&jl3%3UL^Y*OML5U@5e9V1j z`}67~(ddnPtDHjX`EMEmnCicB@GeHh z@ON}`nchu}qCh7fc58#7uyNHQQDC-%NURvZv$cK6L1o=VbqUi&M;VH|d5@Rn?f4a6Q3T&(swf+dbz^9(6+$HQJ27Uu&Ik|L@7%xQ-zImTg4l%wwj;0qLZH>v4EMmL?qm^tDE&LZWqapoXXd?&=FY{E)m{) zp~Evu8nk8i>9Fe4Mpd0Ekr1T?YT&XPCcOAdQ~9%-F0kz2Y2Lf^R?)F97kB}8qePO+ zN<|ZAxrln6ich zo|57;{@3~|qA$Ulc`vl8Y(wqBSSgt!HcKm0g}?fWMOsgZP<=r%|Ekg$tj*Q^mRex3LDeqv*uLB?O+wfowlEpZ0CF^!L_T>tQF`&=H_CXAF~cA0(S)!-kV zXXimtOhAaxc+}NOt&eN@3zw&g4pn*CN-g?bX`Qx3czJU{@DpV*2P0MLY*B^{A(*k$iH-u`1OR4k0O@z1TfGkS4#f9 zJArQ332V%`MIzlYQBXJ%B|qlik=g@vi{m&N(8RGV=B&iW!G2`a>u{!{KMdmiwV-vu zIGC#SocJ#*B-Kv##9>YWoo#rIp6WLweI9?=O&7 zv>K@yjAB$>zlaQtk}9(GNu5{??lgSvB&8b zGx-x1+pt^K$pfM(?%9*Y@1`}- zYd3UAeB?cLrija3oo=F|Pub9#=*iSnisSU*JnHc67Ju0G15Q4jPRv&Q$K2jhOrI=~ z7pweaX+i0Adhn$_F@Cy{88K*sD}u(-!UxYV;;;k88fUX@i&Aml%t&g(vCI^Lr18cS zk?4SifWMer=SMG3q8W?6V%j-JI_l0!qRrf8eLPIrIE_pyadH98G0&v`%d{y!NSKy2Z;X~}vY$K+9#~A88 z$B?$E%FyV;Z6g2fwd|CiMa)LCQ2JQyIG%O|S}rpQ!*yPwd{+tH?T@B{aseG!`Wg=y z=V18idiHzOVfLI<*CehK3Zab!riH`?_4)an;i7|@-_a{?uTZWr*opAfFd?I#=z?b7*i1@sG zqAB$cf3A9lXM>-zU8lG{T5JiEvEeFy-5^Pwt{);lDsSOQyU{eRs|y!-uAxU7HDQZr z4m^%n2`7mU9hZ|uPv4QGFJhzUCRuqpH+w(c{z7p1>0ZpfQ%FkREW}$2w~-L5MqBP8 zW9@!rTC-{dRvy-;uQQFQ%C3X>-P?@bH=0gcU)3;<=S*nc_+oa~;|ijrvWKpDHi5p# z+l_@4{xto4yYPit8L^DsiKiZ^&`0Tcq7PgqsRX>}n`tAc!_pD-_Xc~MJnjQ4F=I2S z->{1OHa|}?WRY+gF{!Ec4C%!f>jSf8Nq#fx>Vkb`*ES+!~w=W-y zUw>_94{txnJf2g_+RWZWPmeXlc}u3yS>Aahx+Q@$s3qdLu~F2@eG_?akV%^q=i=h# zC?Yi>jYNbN61g>Dcxv`_oIg%Md@cVh?GsL>ia1JGsq8Dt8E?$`p3(qcxg31DLWg$m zSWdP5kp>!v(y9ednP(cynd3hvGfh9fpjpT>n%ZRxU+bRnGFACUkh#Cc%4Soj}Y&?HdcJC^1H3?_}$p* zd!JpAXu?R?37N<3Rm_IPD)?>Fa&S;hh8U&4*cFgMex$dt>B%!;U_Q5F;_`cavoC@1 z`dny}zJPa*$cr8SsEaMbiYVDOLNFM6lB{jEXS}mR;LVq6W|xmLoxd;`HRcSm9ghE) z$H$M5WwQs!``O3I79STf-29dJO;%;b+#V9i*e)ZB-z;ECmLzeE%^Bp@8YQCWKZLh? z#aO9)nW*nQK+ELzQH*1#M^F*|UYkXm<}ad}i$)2mYZ8fYtpW3*JsG6Vlw+v5G;{V) zKXhMwLe``nrA|Mkm^DY++2tNP_*+XCkwn2$(W)kA(JPf&IOQ6L0)0O`>s>8!J=V$! zKHS0Kic2W?JrWbhQ>iNr>*oEn^9aoM9F{9!Liy`~(;NBffL z=!j<=I+#hVE`D+X@ASx7a*6_}uin|#-*(a2mMdbywnwRe!xgt)UyJ$CvbeW29CFu zv5HyQK8q*9US8d6Cwl5>5*>Blj;&g^m&6_lCNu0fUtGPs&|fhQzjj12Y{Oypv1~p6 zr+ypJu979sczW1()dibZyyCJrooL=Fz`rFD;%ys^Xjr)p%Gzjy-RBjcu2u!xqw28c zTY~8GDQas|D$8G>u^Cia=R+UwH-B%-K303(2X^C+S44TP2FBH1W!^vh#rS@(fmOe2 zNX1|axskgD9&1{#NtzX8RqtXFG$WXFBKKsWA$)=c5Wf8nPgold~Y4e+Y)|Erp(I z$tduf%zlir!jAeYxbD;?Z1rA7Yzn35!A-sVE_-c|E@{PYQuX-uJ;BaqSLV3>FQ(!2 zQ}%mv81v#0*H!m9j#_aW@s)NVeoT#^Rp&GCoAFHayu^74&9Ac49Svy7u`Ya(sl$xE z#+|2}rNF9th~w@Y#FHsL{GfsZxaR9Da;RkiS$yIRzxV-{v03?m<3K1fw&PC_^?L`< z!tWqzmR{nlGrgp1P6^qbdIc8tPK36H??wK@E<|#f20eRjAqhAi#~evLz_vdBfH5@| zbW36p`wM2G%BRDumS+Kx_6(-YbPhHpry}pyC~``?oZNVFgDkzX1_$?7aL!&i_>`;% zTe^_(zE?o@shs4!j_N1o)kdJ8UWR--6XLzv2*yTQVenrYsyX=o}TY zYi~E=oV1=ybM|lk&ru@c*=mJuivt1m1=kq9KTGhzYw4$TVS8-4YSjpvP%Xpu|}VNqvFg7Y@l5cbM&h& z=qFrekA5n~+{<GIGZ>77BWdY5HE9ks0qg%aE|Lo zjycmtJ}mx+9exSucitWIEDac*|0Q5<+<JKN*(rYFqnAx_%)c?nBvJHeCM^osl!u0gh? z&tyi8pNFSKjZF63Ow66SkxEAg(K^)-dU>L*xML1W1=B`T$qFkP+h0y!Z+uJ5bgxk5 z?rrqSoEFyd{Y3ir`BG}T+malTR}&`{Jf#y-_R;R8*=(IhGYN^QB-cDs*t;u3Nvf!v z*puBvd6P7AZlAH}Y@>*|{-cOYm%fPZidlI6fg5AWegZ2Gi? zj+rry=DydX+pDLerSn><{HX)$eD>q~ujjaIdot(fBeZ{vHRD*_Oguerk=o_yB%kK7 z;rrUz``bd%Kbm6#isGqboHMOH*vhZ&%Vn=j)5VtBPwd0jBiOB)68txom+%i8 z&Bq&OYKeWuNNVcMd8Eg5vN|hovp;LpaZPU${cl1#jn}B4Uy_^A(nFng>M~fnxrBZ3 z#E=}D;z){0Bxq%WD@%fkQS){c9d)Rh+U{|oOzfem!`8>J!TlWPCN-w__3WwIlxSK$ z=>&=Rn?%YcaUGScFs zy*>22lM?j}nn2e{$iss9ja75a#;{tmwb(o1<+!=>2YykZ=vFC3&b$O{4~QbAyQ=B& zBtG@8orj{SqnPy{-?O`wkesUDjJu~F!=1kk@di93L)-M}mf2cVC(vDFV5<)W;=iOW zO%26C(&Ta9MKtxiNcQbBBx_#86SseHbWOg!IGTT-ewT`?2G2%-0LAnwRTr`pf+*84zo6|8pCz@!l|IE90aWhqPN~Kpd-6)nW!o3x`_|;^Qb|~ zOX{ag=-MGm4E>yg2QOSgy(DD}y%dEqmsT*Rx0mCiwE!2*my^HqeTS-qp`^tB z{ua%bmXVN>v3mR3pbg>cECN=CuC|zAX~mmiJA<*coNLdM@cIAytsZP- zJvNxr)emm6CfDzged{`q@By#sO{KN*(X`^-d)lWzR^0#MBt0k{NrSEq(@5J&dVa}S zx+pS>`wmwXKS`P)-u>BFto&PAJZ17v4%3)Vl_q_pdNNJa`N0vo@yazS_i-h?Recs0 z^i5(+ivsYH`Ul4Lh7N=epM$j;`p`QqlE`aaM_>L%9K3#$t|{i8pIAu?QHb7)cTij3 z)%2t1Tl{-|Ayvf#)c)o&I&iIylB1{Ti5U-YoyK`uzTb@=m-3?bIy334xYhLVhvl@Y zE1L=*Eu^oeC()Ve7wP$}t@uolfw4z9?zQm&SW_bdKepI_*~IOT?B@q>3lNebDj?_6 zTNt_NDHKjjfN0}HXxd&3L+ zpwJ`%nmJxz^HhIgI&;h555*tR8_oQyGK zQrr*Y-e(7h^O6+emNJOSe-&U+Z7g2@G6OmqYvA}|Zr70*373}c2kT{4;L>`6UEdr) z2Aq1y?SHOh{U14g=Ys>7WUEat_W4jPt*zAm?k~nGlsk96Sco=vf8si;2x`Ae84l0L z!nq1vI2RmY;8F;j`F9`IVgYOpJOamg{UmwRM@HQ*89Im3VEv(G!22r!uZ(_>n2oF9 z@Y74gBVPw}Qx&2Acr?soi=owA1cB9?xQ;EyRPxh%RT=fw`A*x`IBn02?L>Nv@Y;28pD%IIm|AFON4LT0~`Na24e#=;f~Ht zD61_6Et7L_K%<-dY<|JqOyP3ShQG<*&}6t;W&#U^n()c`1TK?FAcc<(kwq8Nh>n^* zah!9XZ1^5VZoEq%-6N9W)z~SpF*O~|asTrJy>(FXW(;VT3Sb)=fWzXMu=1D&G{4Iu ztFLpp8umG(+7QaTp0ydfJYoT>xD0RDen@I2iU!iq=oeI)1^l|Hn+Di4vyqhQvJhp_E%HUyqI23O;jgMO`u#5VSj9!V#Vxt9)1 z>n#YqcmuxGHL;xFiA0umi>47jQuul#ILOa|dCD;`Mn4!_vaU1omR7L%PBA1~j1ZhD z>wwRfvw*k$BD9I>p}2G}B=kN6sp*yQBGdxJqDGS1ca;DYL-w_X8$Y91llm-X1j@40#Nd6}UD`tT|G7uY!z9G_$DF6N)-vLm>bl~Titp4WQdcb4 zb%OdNEumi*PNp^|^s#GR1<@FrO+=nPMD3zEVW+E6_tPWji)%@!R9Sek@zvk2W_ZlZ$7~=qS(8^vKzh^sLenDs@DHZr)srBPv&M zxt_)JmAoyUbdRPwD~Hg`zJ(l(eM9v28o+jy4$iqT%na8Z6Pa{&hzzd&z^@wf@W$C8 z)O40(ctS69pL>=y7_9|s>0Fo}+zewjyo8XzL0I)(RxoI2D!6gz4$KI70QxDXA$!Gb zSia>9WN76>eig$V&9O`GSObX|l~lOJIJaIn4gLA3I%j;h=sM)~X!E zUiosIee@#UA|<4HjUtu5X+pAEro)!&YG9Yz!i4>O#Mpb@X85{3WNYAArbuHNzu>$U zyeymmt#7j-{mxVJJ5nEX)1%4acpbKVPaBzfwVQ01{faO>Om6JBMNV(M#5~C?Wwb_Q z5grksY3)m#;3b66{ZDHho#UFcOPb*>4pt~KA<(H6LhXz0Q>IwV82UKkg;;4KyPy&w3+w7en(|N zM|uRL-Q%*pH+I0+n=hGc&6Tiet(Xvn8ito%fTbBSbcXW*T77K+Kym%9s~Z*rGd$ifo;aqFz6CD z=d&3Q%{oUo5Z(%C*yaLK&(z{*`n=iIRA-j+(fxN?N?l=mxDmw2WT% zw4=it3aIfSb1E&;p~+hMT#k4j-!v~0t9mZtg_rAyZ^{<(sG<^mRQ+h$mhIGM=0D;1 zt8&mY{W#+_Hx|Oy$3bdQHiSIZhIgx?;b8Me@;7e=1Zi;BZuEALU$vd2-Z7z)B2#)O zdMOs!*+PTGBl7X{c36M!K`}9V-Yx5`iZD}qYy?H-ve!YVJn0%fphscTZZrrCyE~(VhWG5Xe|9~!c zWbky!Nmjl~1rFCAfN3q~$vatN(3o-zYP~+d)rWmBG_n_L*9+k{ch0a+5W#wV&hnY( z1Jhg`A;iTTdYs0=>yR2)cTGX?=<`k3)pQr;CU-;8rf@JCbptdmT?5I*&q<)mXy&{) z25v=4fEu^AS~@vL(@h83z)wcmlXB;*ureV9yTlyB3LWpnA*^&hF_7C#!|8N}Y*?@OPg%%zvM z%%$ed0xA)FlUC41R8gwJ>}4TT;f)>@pO%1)C+`y{YaQ5WKMAJ1I?r5j?j~-xnpx{f ztjJSj07F)jU~#@HM48l)^;(m_xmzD*Ja+(A`50vO*@5Ma%P{ueWiZ(h49+)>5QP_x zz^*QYHLdB8p=J-eD|W-Dn#)9YlO_qtn@4`T=d(`fS8(tr*Fz7;rAq`w^y%o6^m5Wf zT3=6T)h0Wd*|(Nna%!aGR{!s8nnA73RMQcE*3j^>VA>;hl5>-D_sv{=`rh3Rr$j_E zyywa^WT`&W7NkT+e3qb|oxN=8PAf8Nbi;KM&x4&4H&X?$CDR4*aq<5ClZF zL7n^^u(dk@GVA)_tZ_R$H6AHQl{yb|w%Woe&O-*%t3lhJ<9{qm1eddCz-Z(J?pn5i zOJ5v(;kZemnTGI8Lmj469EAG&FWBsT`{Ch{i4@PM5m{Rg(0x>ghpuLmC8@XZ%v3q( z-*AOwmmP&?#>Fu9*Ktr-oDJb?6QM%b2JUSk zm-`I9E;1EFR!R%htWUx6leJK;XACjdbRp!SEz{cnig|bAFJ>Oqqo(I$SntmX!c}=j zRA>KpY~7?wM+RANv;PIitBphO*or~28f?=TE3(R?o>d%ty; zsqs5&*^tRP#vWtmopJ;j)m5S~Vd+qr^^$Yhs1dW27%*4=L3S9=BeDJzqYumk-^eW> z70?4vr4z?#~}#?6)PU*covOP0FO{ zuKoLHO2c}ZDknu>>CC5VuI1tv#mD&g{vbxUWZ-!FJ50beP4*hS$Ui79V{3=MVV`yZ zTK)1R;-EidxO@Z*jo*U@C49)+PZWQrwBmqsHv299J9a$2j8C`hKn2y=^s~)W*5KJ^ zj1??Lp9f;ZMpyc+Ta#KPJV03JL*sj9(DdVzY1^nUHta?;eSc;i=f=^ajVgOkxJMDK zixaTe-hqx?c^1>YeZY8`E<7_LoLv=q3)5W7+0C|Ubhmysd7-Y24{nFi2x3CB&V^Gq z%a!zWxB~r?xd07LZ$|Gko_Kl7awh1d1o`tTj0qKeXXiaJqEpX!pvJol)R|<9&u&KX zqz7`T@eD0`x~&r1l^>z#$tSX&JjH?5-DL2`ZIa|vLdJeO%iQE2BOCI05SbzXnQwJO z!JNR+tQL{>ogwzZ14}Ydqm}W$9zs5w_K=svk~mG3p*Irt5bd5ye0<)J>`;+_rmfRp zQ=%rU+%yS7Uk8CL*C)N284IR;qhYx6G&H^TUmeN z>B|W6CnPez1*@mZB3ujs-;Jg4Z&w)ww%J*Tk z@_vsA{d1{J)H42$L0SCudo}G{`v>34tb`e}s`*;HGq`!cf@)e%q-~8~@X6^Wycai% z>a_2nO}97GdDbbks|ksH+fq{6DI^K|Y{6bf9?YKafEQ^N;CI9e3OCw8cvBF};cthg zuj!CjoDS*D@esdxH)%3AWX=y}vlp)OaoLw^__Rx%#uZIu)z0KG=fT} zg%A92+X5nCorNAdSV_@dp(-;@VaZ!RkVg6WCI`?M*Hu!Q`1HK1UEYPCY*P76T zmN)F4m)CLAy1DF&tvqaAorn(%teK9w(Ns-upz3627{p`+JnHNMgWcHdea-$6DB&LkEwaxM#c;eiH7$vIlKUzN-hDJK0Ti zu2dO`rxsSdo=wI#eq&-LmXXc*2iRjAPq%rJ9A;?5k<$Sm$mX*V%o&qZxF4JV3szf# z&7o34v@IYw$qkkr_)coujfr$cBWd$;A-ylm$Ws1wV(9#k{dzwUp3COIGtVoqr(qwM zcb|hxQ*yy&-6i;HWDIPoJY@Qo5XB%Hls=kCM%;hNr2XTO&O`m|c>SxKw_OdrpLU~9 zS`2TA7oo;00_fgX_fR3XlC5~~g7pg0!8uFqXqS8ist@bo>(s|6TUEiUba0>sv97d4 zq6g=U-HSSRY(+zM^O#l2DzvZSK5j7IK{dk?sN)?kYG(cm3%2<4Z{1TMIT}|nDiBGb z(gZg0)k*T!_AUwAdyTJ|JBc2Z9Z7GlF`#;tFInDl3%V}QlV<*jpl*_mG}QhKI`}5j zZAy-GX8US99BEFs#}<=cp6MTn>lgkN_?5qV$SGpa0Z!5=c>AQkn4=u2-!53HCeMIv+H&8~j zhRMoy#EP9W=zkSjG+HT?t=04=FW)E=^D22d;If}e*iNJ{_cHpN{er&_a;?Sp_wc62 zeD-pzC$B62A{ILpU~|M7=JVtoFK-%~~{`Y5LY|4bUD zctPVwJ_K{Vl&Xs(!RS{n>G*6!I#v(}k>5rFj{hO1x7NbaTTN_VL;!nzPo7A}%L7+@ z(WX~@oAArMAX?vEjFly$=?FPzE^}{1_dJ|Q2OCbY;!I=NisrxMXhTLUQ-98qZ2ta?v<$BY=Nty!bblc>qG}A^ z(M@J=lY`)5b#UIg2@d>}Co6>Y#L0AkS(MR=SBmuLRDBhC^YJ|Tdbg0>q!Wy%>~o2^ zQz5%2Sf6dYvyQ1X-9q!HFQu=IQ}K6GB@_NCfqWBP#!^KkI`hwY?3Y|aM`awL7cDQ* zg^jDxTX{7-((?yBQm3$b+Lu`i!8Ky~)dLc)oP^mCrBDNh8NFW%$?3R>;9-3TYQ|K+ zyhp{b?r<)6jrj-nGGD{AbFbig#8$X*URp42<#Nb;F-^uwUV$!U$2lR{& z!gy|0EEBE>wWpLo{#!I`u-FG9$s?vSJ03=*E`e8j8_3LQF~s?q2~2o60UDcsk@@`z z*tp;jj(kn=@y}s=vSup%H}5~p@fk#|>$229VK$w2RF0azi>DT!R?x4BgUFQX&`(Xu zbolr*dZG9chOBd;mxL?n6t!(!-zbZ|H_yd^&RE7@=_sSqEd}zm@tBi822PfR!|zHv zSn+NY99pOi`_mi9qdx;AwW)?xsTc{?Q@2CUWg%47ynZCI0j3+6mO4X3`; zLy>PW#6~(ncpQOuYh5Ad;vJ%4&h?msuZisIl+bIW0S!EuNk{1?(Ui)qbUQJl+9q>p zW!(W z(IRVfcU8kS2qF6oGRdP0X<$M^z^+me;>Dx5Z{ZKbZ2m0hK6r*WzqTS-Z=W)`3v|Fl z*%&@;(Ex$uMle)NgH7*tLG6$|DAp11otY08BoZMcItI2rnFg2kXTh1w&DcFXoblRk zLE?3)ar!J@n&K@>3wMJQ`)G=b!~$bX@-u zw~?cmtT!XbW}RECO3hsK4%Xv(A*!N*?g(ajmLkmoTax%;I$7d29w&KUA3|K6(hk8w4V7iMK z2E8Qz$Iy8PQuT*nJcMLtB-sg((!zP)n@TcL$Z9VQ4W*r?y=RC}Mnp73;-2rTtfu6r zo%WQ{&``Ae&foW+d+t5wd*1OppGVzf9hRPW3YxhG;rpHtZsEiI@GoEzuAejoU;M0v z&#`X=wv{6!JRZ$0Sgeg7nr*O0$lpC94z@83{109YAB7rEk3)XSF#MHs8j~lC#c$%< zoX~W`xn*upxc344cjOv7{eA(YZdXU8_%E=3kQ+R&%z_bb?!gkzdWgFk2@75=z&rok zaL?lL*q#`JF+y+td<%!`HbsKRz^yiKW-s8M#H}Gw>Hso-UO=9kqv*Yp5y_S{v8V3` z)4tQn^r7J-yD;Sy-%cI2_uCUlCiy#ykG(^V#d5W;LY|PGPAqkI39~M_S+sfXT+(xC zXNMFaN^EH>|Dd+*cV%Id{YC+M)$yA3q72&J%jC&&?m?6qf~UempS-2 zux*=Ov9EQdY{jw?US^CruQGTljQp1hhjJ=;yZhRhkh=qu402H?>I_CUwcxzk_jpcI zUUW3j84cB)gl*jyen$C3IUg1i#uy zJ^DK68&h1b$7Z1=J&~?skvntQZohl1v#S%<9=-s4>_BeSfG2Qn@*a%4<%>O)s#w=_ z0wwBP(cdi(-+yq%1=S9)$ln7VB`jht(V-yKSPR$dH!vv`F?f!3#GHY1IEj`gpzWuL zxr>E*#5@+eymw*ps1x|>XEe^6djl;-tibH1TwJ3aiADKWATlTtTVyM_P1k<2-K|M% zw)I1{{!|IQ)e0c>Ne`%RLpQZg(W#a0AA`8_DI^VB45wNmVd4C}sHXjwk6G~n6c#pd zI=dT~9>lP2pYf!B$C5-kuB5TUjuou`z{Z3Pp@BypvWfYw8GWJ<(sEl#&wr-~8krah-VsFUBPwXcWx-FFy@O0W_R{53 z)5wguQ+AXeH7H23&8h_~E-R8g40a{KyhJ(!R?*)_mcl(Zf{xkC(4gYw6#Zcg{b&xM zS%)MjT6qL*xqO5DbvnV;x5O~>sSg<`FtL2~cF1-+A}|k#iSH@U^fWIzR6K#cmdv4` z1|{NK!Uzrzr&OVD(-zL&H$N=Mam*OfA8f&P#!2&z&I71E`Za5Mx{K{Gea6(XCJG@X zR|SPu|J0KqxaH5FPk!b@PpA5^C4y85KBN1M8h^vTI>CXJ5VNa}?;G}|kk9(=PW+v4$5KBSEO zSSDzqdd{+`ZG+%=?Ff{gnZXWr41{mC3e2f~7pzuV!{TgI$kXX1)sI-#M;re!Gh&<#-ZuEs7Iw#whj2q1@Ts59X zuUf^e*OjEXZ&zSm*KT1~lMZb!)zPX=@cZ8l<_uTnaTS5}6xXjruD9*LTG^iKTiDD_ zN4|q9o87qT_zzqCD-XCoO6Dxp%bBKoFNN5{@ub;3m>$$Eh14k%u&_@~tn%58a`zv` z1M_0=l;L04mQaBL3LRtS@HnqX8*N_+^Zxi7Ft%g|sMuYG+yDK6h_lgX(vgQ}lzDWs z(iDx8(G^*~7vrX5CveBo6dW90j7g(zqMr2%{G1p9o6ZQ#rvOX3cWWWd)VaiRudQJk z#}e2gLv7YtUI-IM?MJUe-%%;K3}-&LhHLk{#uM>nXmsKj%C{fF_2&ZdkKtOBliY@{ z?y2Fs%dODfEQUv-si=KuHx9qpiYtd*#V>8M@c67CTtY%XlUp=olW{`ux2)pJ8d zeV*nbg~9N4#Ok2R^i*XW3X^p0g7=v!`0i^ zpVjd!)^H_%u~W=RXqd8O|>@T>R~6B>Vld92PB?r=>519JqnP zF7=!TUHLPH`ua!Eq0KYN;)OEm9+IOGFZIxAdIGc4XaS8c$)FPPoeOT2N3p=zlu?_C zH#5f}>DR&#-^@aqabOi^eA^R3w@TBpRsRW`gh6y5%8|ZZuVv5cIMRGCMKcypgSozg;APP- z&fv2TbJ%IYrEND5a=u@HrfmRLO-si7drz@(>_Y4uw;WgR%SYLHE@(Jz5z|hvqup#h zg|-e3_-Ac@68V?os&i)H;46|VGnkC|HsgIDiH3jlV3E7m^S1~?w4UQK2j}ucL z3yi?AxcG%V)@+=D7l-8HJcm&j_N^M0nTw(D%RXT@#A=GCr-Ikb67KUCWt=gv08W&P z_@=;t%u~n}8Zjsw4u(y(6>!|p<03CM*n5>N%IxA&?pm{&358HuKLB>8MskVk{ovE? zsW586OR$KpW=a}o*k#>&VBaiFFY7$n)n6Ckm)a1}U;BjNd`UVoX)LMj-^++lc&AI# zs>DfTmb*jH&i!Ru1CZ1&gp+IQA$oSRii-T7kytf`bY-(?C)-DJFC8V(n9Fq7&wwyH zg!tWa$yisOxd^)3&5IY)aO(o5ao!2l$Hk#Vc{_e=d5X&F2k`rq6R4uphMpB$F?iuR zblY+kmO0&FS?2?o;>mj~LotpWc_dG-4At1kUu860n1k^frqP;HS4i>ROsY&uB_p+| zRFE%8D-#~DQCGLq_;*&sTuWKfopWNt3#-}Dl|eP9;w0$D*b#Jo`VVH&aFovNtf%VH zqbcB1IB6!oX8Ru(Q~k^g-pAj+=90}#ehk~sx+Isd%oF1&8pWZHpWx^~ljt z-Gi)q#1xob+{5-edost84)*guGbq0`oBpZDQv4X^J$GHHhR)735$vaqqF%-t<)P1`Oys+~nrVIH)1dOfw<>C@va3u%{3 z6gi!nN-aiP=zhdSsIQ8FAm2_nZKZ)SE_b2uO%XQzyN;^{g$n=49*j7+4a*lLq1??6 zxJ2bNmRSD6h#p}_vQtsC#k2`q>#H$PE&#KF*W;zefy`tM&vu69z}UfW5aWn z>HFD6dD#(mYEC-q&rYJ(A_qF^lT6>uLP@>vFMWtRP4AAJAuac6T3Pji0!+H;L&`&{ z2s%#AyK-nq)=`D@6}& z=Y^V}S*{4bMtNXfYc_7Ze*xE+_~LHg5{z4(i?OZ({dde*g5!@ z+Xn$}d|*%KC1+vwALCdomcZPYX2yve_?(S>P56ucnIH?JYi!|-RTd7@*nmNON_fD} zPhgNuMsA2UrgVnj8D}j#@O6ys+dGHgMz$rt@!K9jOS*}kJ<+2#Yqe-~rEiT$a~d_3 zuV5>L`D0Ajf9%mDQ~r1P3eede&C+Hi(M#RwRJgRp)+_8GQ)-Z-YE5IBzP*&yNDblr zCJA1sVFpb1qOh0xTxNT_)14iq5A5_$C-$BO(X&yqwEWpS_VLhlrciCeEM(S0$(|bV zoBw9Bh6@|mmOrZ8)kt<14d;56jD{tqOB#vzoB!s<3HY`j^*HavHL zuIgr+A2)TOea=1p$Gl9gP2Ja>CHW*e8{2H9PB)M5>eY0}5g8;fyUdo)hiS&kvYIQcziO4lN{E~+()Vk9>%Ba5x0gdJT-|51b(fY+aYe^K*!(}t8R~_eg2X3PT=S#RTa`*r6{tzkV&P34lu9D z<0-($Met@<)7H4roc9n_>aJfzHV$Jbm>@3+UlNFQ6%OdW;1o9`Y9(B~Q6}W4?1z)HrJ+4f zsEfCZq<2C-)}K33tmA>8XL|F3o!%0G8dLJn$;boqQ@ru*w+;Nn$wId39!soPwMftk z{lME}{qW|JGCXl-E%y4^;z8RqEO^}w-7VH^j)5|`o{oVCk&tDhHjTOr-m}NX=OHXv zf_e`Lo{r*als0fAbF?4H7bJz^=;5L8((^P-GaHWQ$rd+sd10Gm5Ei&bV2AB#Jk=|( z?Dn3*_pJ^1-*w@8907-o({o~4ns~?(86{$j>bB(XujW9A7nkOP(|}Pv`I@~{ML1BPpRNT zU8P39QVn_2z}q#IMi<$d!Mdy=xSJiVH)1VsXR)kdM$9x&U{CTH;CR#!zBfwIxQ{;U zjdVNz^3GwbYm>(!*E8sr9f7@*w6SXHSR9!iftq8yVDrs(JS!R^8g}vr&YV0zG~tQ5 zs2rNH+D%GyP&`C5H02L=DNPfZ?U4};eQPQj*YX_4bqQL|`Y7Brb_)czt-((-r{Hys z(I{=^fv3i{vwZ_xs6%HMRqI($!GNP!TiA}zWYf@P=`)ns+JsL_gF*iIYZyKD54ZW= zQ0A+y#zg0nh27Fj7XQkf?WnC~hv0Tiw~Z$K_%{r9EqEY)9IS!mg_+<#s)#?Sc^5WC z{9{o!FR?W55ZJSJIT$(%JN=L_<}ebgZV+;dF0kDd z8(El#GG&>qrw*Z>?c4jA8CFE|6Avc9L-#oB|NRhOI;$f+F+^pQLec(w{qdxkMA z+q)8aDndEWvFF$(ks>$+&!pl7x-`?EgDE7{G1qlItV2zi%z9&}I5pfR%cl@l>3Kk@ za~gJE66Mx(EiHq#r$SbzdpdI;|@%3;~jSTyl@_dMX}JtJXSLGjd;F`CSFi0=e>@YVYtfy zY=~Tkep0pAv*|t#eDoDhKQ$FOZ|}v#LsFv087=7NUVyt58nLJMC%!TshjsY}QFe3| zj+nL*qjzq{YhlH>D98n8wobu${%^rI{U>xB?qHU)|FKaqkuWOh7Toi&#-_pN!FTaZ z5VxEVU)3~bzwQ3Apodj(JnRPKw^!Bt2-3!TSC}YChHsF#DQu+dP_WsM%MS_|gc@rT*kaf5F#gwL z@qZ7znCE&s_T{jUJD4|siN<7ra`RNyUp0+Pa#dO8C{OkwD}q@ryAJxQZS3Hz)6D$b zEb()VCoF!I84drhoIO2ojqNeN$NXHTl4XJ6;qNib)o)$8|YV|8;#!A$L>FO zp-RQYl#v}nuX_`SFHxcZk7l-hQWU?h@ssVqpEr2h8K+q9=9%nRbrXJEv%?`ZWn7ArJ`{W@hv7CBH1^mjSo=$rZ8u+f%Y%#KesU{_@epkHGS{z@;xwp1mY^5HW6t9*vKuYcll zTYr2avkK3@cV~&Shr`O>b<9NV43oW7g8(R*$2Tp5arMBFIsj%P$m-z@NA3Vee*79B&;C4xvxje$!=4+%?;_dCUcC)g&v} zq8o|pR!ZQ4TNe0G#skOYq;VS``Qk#m`M7=HK}@u~3ni$6y0%lXW~C{cbhQWMR7^2M zX9-TNn1%kO=Rl4XgX!z(DAV7J@7j*xdaI)loV*zWlP}@p@2a@7TLI6%(qk(In7}KS zxtMiE2OlUMWM2mLf{9KWwkoA#Vw)$v4Z96j<5!74yxt7EL{ z%Fu*=8uY?e%kUpJv``m6y>rLnqf#h0=Qf{sJQ*}A&hr_g?O;=JoUO%6YcMSwE{dD@ zAI3%A$LkO7VBX9ujFUczU z__sRC*78X=#9#MjT#p%ey-Z>of2q=|iVrM#{|Jg-ra&J%Pt&@y18Lv{Z|ZungX*$R z(M5GXYHrM>{V5sb<`hEq7xd`U*Uc0mb(>9XFr$xZW$gBXJRxr))#4>aT(EiF_pfW^d- zsD%yny3fqwLf9XZSZn?-^QSC zN({b!;)gTLK8Zh`8U(Y~A#VR7CDawCV3kmB`sfJxe7`~y=fX`+wz=^Ltz+_7l>Kql}^VR7rNMIDqR}29a)I$9!Lvc%N24}RD=arX8 zW9l_UjC^jw&zNzOMO;XSGmBrd16gXqj46{vsqPR@TPOI!Cls>0Yokf?jRK^7$fi|S z7w{_wpSOLajEg(O z9;Qt~*E!GO{Er>n{hKxTdYTO`Tu}{e*$#MVY9u-+37#dNxhN;&hLHn;Fyj0He04<+ zcaO;8|M*ski!8jE?f51ZSUih`CEVoS{1UYA)5FL+`~vm&GBP}wO`+yKblXBn=#e(i zQ*{}#TJnxL-tdIel0HyswuGdDo7s~ui`cetmzijoAujLO$;G~w#eC~%9FU%Y?AcbF zy^I76W;6OdUxN`9Nw{$q;LszRg|I(qvMLHAg{Ty2`_xQ_OAnEak26gc2}~Z7Zq`_t z!v4ioklv(LwqxyTit$UJkh`bI-co{^xw$mr#6Ie?&Y{Gf%~bhRpC((_(^u_K8YdIY zo+(|hT{7$~b6c*>6x^n=s2~aU_W3KeI<1U}H*RHnLMGD0oL4OAb^u&8(!{u#5#aS( z6Gy#u!D_i^+*W6cC9wlAR67+qEah?6^y^r*uNlV&yhhu5Z_$3@9Xt{dj+IAWW6SOk zbWfasXIrKeD!s9{>yZJZsTCI%#mitJc79=?@Hn2@&|m<+o>dNyPv|`jbr-=QWZ$MV> zY_ZQd2tK}cggKp7crZd4e@&i=iWf5xQvZXSchaC>& zcMN9Nc%VeUOO~gc1;!PhSoXBDFfr;4;L>_Fc&sO_&W$A*ooGrMJBEH9R-;3E($bgumR1%c3KCIO_9b+tRKS_X_o4mi8K}#7EX?*gAUHzs z&aPX=+#Zjm^q>3L`JqkFd{`ScnrUNM_fS-~-i_%eP4M5-&1e@}2}@PVL5u$Z_1{9# zZRaRFr)rJneyJEf@`NyJiAVkW+4y*MK5o9qqd5B*YS>=Et##`$Nzxyu48H`GEla^l z)1Lo$P;5In?K&rljf4JQ2cT`fA@@7+A*VMm2oz)bx%m(LVENP*&dPW!%bC#3_Ury+ zi}M!Qw&px$!`i%<@{&eYb@D2!@&3rBr@Xf4p#gVwk$D1)Kz?qwBZqP9@ITX$IZnk6TG13&`yo{O( zoatQO1ls=ODci6qfNd3?tq1jG)Y8|hX)_=wwoReJ1lLBeI!uch7-%?*JG+C=1Hd=CSQEHzBKa z1alB{9X=CGP@5Zxx!rp(>~TE0`tLyB0teMK`j@qP|>E@K|9jSs;w<7Ln;;TB)};|MFtHx~T18r;OjL3HTt zcXnAt3#!^{U{%s{kn|h~(~2Ln?~)uVsyoQ$pU-8(Z~KVvs9)wz$g7czg%4fiJ6PVg zYIb^r1kDvMpa=RZDD&G!dVhAFZE#*FD-bfwH|V6Ww3rY!C83z!c1)&m>A|$+To|2} zlB(^z8BL$;R?(26B>FJ+6wNm}K|j)VQFBiz{mU{Z&+F!7xa>Ory3T@Z&H*J}EEHc3 zccPIOPtowwR0{cek`C;TBtGW`*@mB@hM>U|AN3h_&vnNXeRd(cMVF#)8jW_Rh~QE+{5(idkB0lbh`13xPg*^Aei+yoCLnq(Lz%VUSmPR6Hd& zlehCrgTs=tjK)2(y)I;J9gp74`|vKbsJfWKS7Z>kB8y7%Z&DXarqAgH)Nl#OVdpP) zYS~_LHmhWTUMzf%QG;t<*5KW^P@I(f7@7yhpxOs5 zY*3kv#-65V__7(!Ub+mwF1&}()(Chw_AAuRJ_43&9B}#0c=0vG64+U~0{X4hQ8qAx zRc<2)UP64hPZlg_5;zaGUi{?nY4pf+9T~wlCX??+19E56wQFlA^VLdP zD85SBiyLYyV25fno~m(0U9Vo~$UTHxuS!7L_Tj9taWbC}e4n?ue-)MpbE9YVA-L{B zGqgRr0g`(U@YhxTu*9l^aCT@k==nv#;tNS|@{bhOL`YMLk^^0`8%W3d;^_6SG+JL6 zPrs+6QdvM0NjEx?i8fMXzr{MXm!me_#7)R+JM{Y?N{xz71oA!@DkBB08D$L@l zvd{Cca#ERf;6c`?IvM&r2E!@GZA@*II(Z}kU0K_}_NMvL-id$N>xn8fqoINwt6WcG zyiBQ5@F@24-`R>OPg%pCQZBq<7`*NfvfF9}?RQQppO6{C?z)t*tun)Cg`*19ZHi!; z5og$#AKU0y<}v0~XhuaQu4Hc-MWA|yVoPO6dyoWK_l=}s|He>4uP>c&+(!+665*=7 z1^?`?yuc-tBYQKNnljqC#(pohit>3 zq+_UN=!sU_H(~MXcr?}wM(56Kl($bq#*SiLjX&xf8G%bS9l`lc-Z($m6JnASxSl~X zVC>31h<=bM-uUAm6b@F#akq}ZQcAdRimqMM15Y{gJ>8tak666%-H_+3-MCdrMaN#2k6>>5$T2b()x?aQMo)V%;VUD%r!fkZVzksKT4RHFLEuiA7As&@G1=}+R zi0*tkgJ0{lMOT@TD8fx$G~D+$27J~Pz3@^KjrNfiO*1zWg{~ef$`*e2h4ErMHnA2D z-K>B$8YN(7oep(&L$M%x7@pV9f#-7+x%dn_PV3AlW|gVSntode*DMP`0uLh1O3)GO zXut;tMXDZphqG{FtR?O^(~ZBtMom}dKE_zk_t0jRJNpyU-fG>&0Q_*?dXFnx+55tF0iIs z4nC~ul_qNDdE?UAZm4xv3wzvxFz9<5h<7c;E`3A1-x-Xi{bi6Qd6^X^>+nh|gkUGZ z-!45u4qZ^{rpf{E33-`0%5+=f1Q^EEA4RTN*J%whtZ%*)c!A z)!^k%(Kx**20vTQf^T_RSo!8RNGJ)}EDLAD(|PGoUOEo`$v1F`Rq;$i{}$8gk)(=g zDG+`(o7GH8XB%hp+|KSN;^6W~R&h~|jhu9Yd2ap2cEpK+TRhix;piEdZLNkIvujYT zMGeh1%}07(4eS3Z(ut}9nDFm9=j^r}C({^gwHSl{4ekWv*XF1rPArDaRUYtfr5hxc?&tJU(pHg z|50!Gbz*O-$VU2JeS@9YY|b{eonyzcoUd|&NmFY*MB%7zy1XOADW%n#yT_v)kGX$RCvGQx=h`|$cG z4(rw*f`2cXV4<%X%Bn196}GYLJAaOgI3zGCoZ9(;%_Xov`UEf9W{8t@e}P5o5HxXG zi5?5A(eJ?oG+Y(H6&=dvZdo|88^eyU4PAw7Ti*^g_QM8F()c#_VD@Qt_ESCVa67^6 zFztnNu~#`3eE{s9g`)S;5m?qR7+udqU|DY=4t`LFrU#zloz{yO`pOnvPL`mZy5NCs znS%~*O7ZXSSi~h>n6J_W7cWQPh9|KwuJezLTf=P@6>P+0tCX48aW;G(5z7~LXY=Fo zuX8~y3rQ|xBLxcjhE66ywfhyRXrLi1e!rH*uMK98mpZad-BxJ3p&AtGwDE7IF{jW{ zA+DP85Z<12p=aCL*&{y{`s8gwsoN)TQ`j&TeAtt@X9@0$-J1Y5>A)dpft|OSgN64$ zLF0CFaLEyv1o^dW6*q*64YlaoyrDFt#GX>!O8F=K+5De9?Of}`Cv4vbGx$Nn_`Sc) zvF5lpUfK2+u5Q2H)<=jkDb=>Vy0+CL-`1ao-^mi}EtuF#l>$4}? z?KlGp-N(T~H30vv{|G9(-@{WF0r5?9xZOV`p+#;xQ?DzjetJ2VO)(w_$Cv%)A5?@e z_X9)GI^YQZQ%0EzURY4$AQetcZ6XX`vxoP-+{M3sEp5AcWdTcZj%Ra|G|0(pA`LmS zg`YFrikbgZWf<~pefU}W{-v1sia@e_UQ88-!s)fjUuK+TPhB6j zk}fC5w6u*V`La9>O!>%07Zs3UL?oSiC!*fl7wN9yI@%w!mU`Fa@jrJhp_LhhWD@(0 z%uEN?UUQD4`=5ga4I-m^o5mAJIMF3XfiHM6jK(D&rsmf^Y`RVum_KiV^N;nhU$|#B zu?x83zya)&94K1gs3CIQqa!MK@d>BB`ip@VltiWHmUoowJyjw(Q02KmKA(?0yXS zeE};3ChDHRSe$cZJ}x<$j(=C(!bjTb`19xhTzsvhDy?KB>b`#oP7;g6x3uiA&@BOF zA7!HE_1Rb{^j%pXg=KRGqKA_qn!i`T|C)sy>3B(e|4j`C5B&kLgTit1_)L5)>|c&q zpT$K{a-zDUF?h$c6BAYs6IsA=^m&)s%+;}5j)7uYSV+P^Tqh)aQPyy^*Kp-Oy zh1RkvsO+6i1J-1~!Ye`#E`1z@#6N7k-j5%;F5v#8JJ_F>k2e!8qKfz`O1-^_ZLzbJLHE$x|Oe8rM9OL{M9#|N^BwMp#zn;CpTNj|$> zn9ruv+_W8jO^0_%mV^2Y+N2zpPD>(+>CONNlDwizF~@{CMr|ikIcQG_9~J3(n;}!& zmclN2=#zz00_8uCp&K!0N%wk@z>pSEmgyrtFxrc0f1AaP)EP`0uP%m?!WrDw0q(5M z(1g8{3!?`n7PL`WM0>qv(V01swBqR+@@kq%dRYT#4EGERGt>-EvMsSHW2e~MxZ*2UzKkU3hC?v!mW$lY3=-RX*{`med_Hm{Ho)q>kUIWi^ zZIfzQYSt%qDRLVd#HG;Kw-Gc_U|lv%4W&(2Uh&?m2N1vUIV%V}$bY%l3`2Fb*u3+q z>`84qG>FtuG&uz8oEk7MO<+ujGO>R_H2P0}fF175s9zx^5+6N=0qYvky8J2XckAKP zi-@~^e}_*E-yrkqG#qbx8N#nV;yR+IW3-B z^ko)*#!?NG{fy}46Lsi#t^o2e(d;mbV)9ZWNeWD|kcxwDb1A%a4z?=am(znSHu zR5tKQER%8d<;!GcsGwLI9%$&I(`##-@=OcTp6jBnM+EMBe-J;<@)2fXF&G*ulom5X z@lilHYWZX$K0b_+W~mtU+XWwe3CFwX8}M3eENT{>Dx) zn`6LPbvS_1K@H3)O~bEWLhy8>8~(|)f#b7tK|WXlmmZL%3C~Pv?6#-O_PB|S!Sys| z(q>QlLQSaW-Z*}m^guTAnjb&jC4hJTlOgz9tyxOhIe1uH0gbUzkU7{0(rZef(_#=F z-u8eqD!&E>Tc&bnPIcSlKX2sI-TJwRA48yhsS#d0?GKL?li^if9_Vf{g+Rwrm@Rz@ zHojHGEy}*gwVsCwYPtNXS&QJ-bO-z+Fl_ptB1TurVUzHzPYyP~%C6P$sQo{P`?LVI zP1MB^Y7_9q_XuQ{E~CPbdc3b1gtJyGMdfiP@pwlih6^4&@AJ*L!`=_&Oje-Yl}#uu zFgH#IAb@xgw_!yhKQC$;bBh?pm6wkvFR$ZV@bz%2o^pUDTwhJ<2Yl%J?QC|HjiBFw ziZuI1Br6{~gtmJwqia{<*>gLAfhM~Mdd4n+bJKPU9^+XMTBV5>O@7-p%#Y?4Z}kSd zvn@<&^kHscl9(5pgzyh99Dyg}T`;`o8Ql96f@7Tr3LaxY1OMR$tT<}NN=!cSw~v{V zO0K|0{rX0Hd)s$@Spd%lA2`G>Fg0a=?jGd^hc1PQ6Q<$FHJ#vYxt^`9uj7>+BKU6; z1(tJf0Bb8SfYF*-+`07!SVrtw?pbm**Pmk~%ts${p;O1Rk9&=1S_RKSy^Uy4w%wq!;L`$eAQzC?qXbA-R{+i2V?fbm38X$)g_IoY**cdlHenqaw?OZ_muaT(8jgv$km>lRYaXfnBZ?(PE%%_py2yMsYv+?Id-&= z#M3HrEr_P=u@h;q%X<3WE?*lm%(Rv!_vZv!zl4Z6;G~xfcnCI zGtxw0*{qWkE%U6$>UuSi)RsJa81xI@-|E58gsV74v=ZN!zQ)Ve+4%Iu0p7XwJYCM6 zNk?S=FrE6h)Ob0Ljy(7wWYQ^8@uw8Bk#H2e!{ux{erBon3rOqEVKy@U4E>qlO#0j$ zdhc&Uq97eu`Pd0`x8=a5BwOsMyAQw8cXDz|ZZliU0o15;n$8m?0wHrx@M-Akx3d>^R!?)miL-lRcd=iA#WlTUKD?9F8uO3{Mmtd4~*Z9CbbtCs>?C z8>?2d%)Np#L*L_4*CSZ-ybOOx$%>3FHKXLWv-r#4Et*FBL0|bd_{Vk>I(!|2rvw() zUsDGhkz>y!E}n%~Q_k8tCdaUUCxm*p{0%#ndy_ByagPrjb`c5$zkX-3nEyC278mXq zPF75nmT$~r#m{w^k-!7;YYQSAcY^jR8`Ft73S_4m%Hl4W)11C(WUq6_dc1;?@V@&D zS>CGX8h;e-s~=)snu7OIM}#h^1vpzcKVqAbQD=7my3XH$r@ANLrcFw`*M_Cy7cslx z=Jv_#NqHw%bFYdf4LC_!KGKxFIfdd!&!%-_;%LYWF?lzCg`Ij@_|J3*E_0d%V}HEi zH%A6zhVvRWH~kEsmOa5XZ`%wE9#Hv!VA4aLE9e)ZhuQ}qm>2Vn1--J`kQ}L8V4}SJ;MZHl0*hBxo zCjBDx*2#8Y!PGFfn5n3Yq%#+?X`S~|?rY!&&Q7m~ z@Ah8>WT1v){C&{z+;;FNj)$z!Qh0Oy8+YgNVpu5kf|;CE#aRP7#4vdvJ?_lpAN1>z zR@G1;d!mW8JB}o?$7AS*%Ve6jppl9WM^cyGbn4reD)=gMnY=6s+7?S%AXhHFao&L6 zd}|D~CF&B+JIUgWcF`44KKqeW1^eW`Fv~($YH^wXS&|dMC-5|@AJE5Ij0VBwAUU}F zd!)GJ+Fm%gtQG$A&x6j|3!qbJfEn^f@KMHBbh*@k6E;bU9$G%aAoGiO@~x2bTQE@M zAXA5S5s0aarot-4R=D`eoEurc(?6wR#mDw5+1W zd(WxMcZ@KHFJK>p+(j+z7#bkqW8E7Q#%A%w@FVO7KeFZ~(`oJFwv3d8=67d!-$Uzc zy^0NZ-vRd+ndFdn$ZGmt6hTEYvNUa}J1k1wjz<=CL;vq3=$smZ^}zvnXxkA`)p`N- zt*h~&?s#mNHU;B+qw&Z3IyBtX3HILVSeNofJjg|cHU}<&KfX)wu6HQzf4mjFWrpDP zoOk^69i8B<`hxSgYtQ`t(<1B3-`V5iWo*szt8A!>peMN;OL>|h^fsiNWF2-=zs^?{ z*=Iuk+8Wr~A#PAHY$DDTJPSLHeucaHT_NOHmXHN>ksW>11V=8PWzWr*@i#ud0L9LJ z+biZfZ4HWM(({R8Hve)oOHbKmGox-VxD}N`U$7n;J{-n>S@aLyhstCBr1wmsEC7_= zTY_P~AxtUR#?dHgDm?yz^OmiU{(KN#oNkKs{@=jeH4g@b=W{14 z_VQR6#>_Hq@Tu zs|AltjKU3n|6%r$R>W&+=(0rU59az}zf(G1E3ih5m~$YxdN`WBUJWy}g4xFnu`F|4 zF>e^Ch{q$`(Dtq&dT%J=Y)c%Vs6mrR)`q5y&1ZK8USzLxW^MPQ3+8Q%N1Tj*Pt zTCJEz+A zvBqDvS<0CB5rOpnPp;ruJx2~FA5uZ~Te>r}g=VT0kx_LF?HuJy;}!||V5Xa>Q1&nz zwV)1V-6OG2e>C=LN}`fw3EX-271SohVx>w7o9=LutsamIhBmWsPe&#;+?xeAM5A!k z4o{pU7W@mcBiLN2SN!|I8)5F|6fB&74hr(7bAfmMz@GEB*rf$wOcXeh4hPDkbXqFJ zF8TzafhBBkR|04y{1WS(GQzn5i?DCiEQCHsNI4@959@u%Vb(T!DR?fMg?DYi`yX7J zz`)5^7((4;2`uoWBQ{QI7xsxG5N;jkI&FkZkfJTv)fjr(!+XA>QOXE1}1@* zy(4tY&j;ONW%f-}DX!oijQQ*_?(RK9NyeJy`TF!L}f%uLt0uI(o!0#ulhZ|Km6}JuXFC_ zy06dY^ZsDbhBW+iyAY4m$YMyg5MTDD2>+3GA6gEc!-h0-zLfiVerMl5ZibTO^BP~E zc#I@p_30P%T-b>=Z5^oj`Wdc~=Hq6q7iivEjl&|Ac(~CDeYwu$Z~tQP&Ml7Fx7`>Q zOEO%)aymVt^P0vV2%#R!#qg2ePpT`h9E)aUvzqHu>7LPb$PD?@`&&eDB1RZKlVM4D z$$5CYauM9Vc@U1=jDdBBQ(&2FI{fU*fg8v2Xo2x#8uf7{(XC%X8^3mw7a_-~lgbI^ z=kEjTw(-l2H(H-CJueLylrbh&_4`=slgf}>xtZ*Lzn9i(J|ZU9eRXrWs5P2==1f2x zjH{qHo%e7pR{zsRos}-QZK^b`)zzlYy0#c@J!#2m@Z;zYXM!Hb4Dh_hG(=#s01}1f=KWPpEy2lCk}~KaB*T76xSVzh%GKrrrC2})s$)O!0FtKtWNXW~> z;UsgIF1!WiY0ZPaTpN&XF@tYhR%OFaGgx($1;_9v=((8#{I~@03L}tr&I7t*Pr=u; zHhAk_2(|||!ibM1^i3~><^faa6OIO(t6mU%E0DWCMnd3d4wS^N2ai^D=-79byj)}j zi6OcW8}^mN_EwYAo^qV0Igma0;S9ZAUc_2;ay^4#nwTpTKwJ95NlMf#^le>?P&pSz z_=4DknsnMMaVm4)mo8iJfaV-Ii`6U3(aksle|XtqL4hGM_LpezX%{NZak(yc&c%wb z74%w(5MA|M*>F6C(;za+$l>WJOsh;j<3H#?|NAWnl`GCe-14)WZ+(CS+lzzHLoawN z`-3#Z=71QdOR>sRcpd*f!s{L4yc?yLKp~(IGQM7fX^}xND!m+tWjm-XPX+#yRH&*j zBD<`X!>ZCiB5l5xP2X{e*<&yp9xRTBh_M#pC_jgsFi9hJ!$r{MD(eZssc<1vZ3@g;23iCtJ z>`FQYd1T|z*=Y2*UX0fpSxlI?0c`@#Vp_c;F360aCKeT*=k(xQ^*4Bo#{^I=NtWVgM+l`kFOY%dL z^03cJf-jJLA3c{}#89OU)ZUwc2^li{+Twcrv@8{0E_cWCSp~TCV+j&%NxswX->Civ zuzfS9`}c1{n^{ljvwcHU_D>N`QBlMFC&F>J9@3p>!cqI&J1T@4So^jVTb?;%yHXS` zXxWH`9J}N6lKt2+rJGJMF2>9-9*UQ8IhEJTaQ4FK;JoZ1dB5}oiA$DaJQMuM-FM|= z>U$OXXjqbs|JO!dE!jdM>fR8Cp9~~BL=pExuUS*mI%>J*23;iy^mAC>}jn#$=~X`^*wE`sMoAWF z@B2fHo3;>pjwKVwb0p^S=HyG&Z_;n5PfkrRg4H3B5EbnQ+fQACUX`1WU6RT9Y)?W_ zVj1|Yw}pK!nz(E^zbg@??12r*ZK?%#ushLWSnoMWZ-%f#4DFf2*uZ{`1{)CG6`{MRu z7n)^ahr11*)AET5_{G!%H?`GKCC*sTF`1?JMBgxi$#>Yq#k=VHHL0wW{MEX;;Vznt zP4v&5In>Q|9Ui+l0ppj3<2l10G}xw_GJ*qi)5Qbq72XZvIb_0h5?v;p$7IQjGs(m+ z@d5Q5XPGDNBKS|+lWKi#rbgVIBmBR96dIa@7q|N1Y$p%2HdevA?`)`bml8g#SHUSR z_1L;R6&tfA;Ed7@D4;CDb!)D~qpKIYsHuJeZQ@%xoRQY^j{@dlns7&NA7R7Jryz2pU-0%vc z)E{do*RzX$J06C6OkMCg$GM!aiDOlTEhSg}Y;3e%6$S6MvgCSqGD*1qnFtXlrrTc; z9{t_{vB`G8A8LespQ7N}-&Cl7x)UC{H^5%8c8IT@NB2YYyVLM_wGvP1 zI{_`%1c>wM1VtBaj+`yVBuJ*yvcQAn%9Tti6fej5bxcXb6Elvp`-LeyG{)}mzCp4% z?_9i|D+DWdlKxCrF#Kl+^^rwH{n{0#?3bHi^A3Ho&-OcAYW|0{!D`wxHHK|G_KKWQ z$){UXO)>gUJe_{M4y*FM)1Q+#4xC;J9^JG8@17M!E#c|R+uL(N^U-E#DvyUyVb0H3 zPz&=WCV>YxKP+)t2;2Wnq|Z5XSxeyoB2#gTe7LX=#6>qk&T}`oV;%?^WfSNpb)?28 zM9@QgEuKg?N2j0$y_>y+^VZc+Y5ouTzVjTFNhrrMEXH7sF{)GOj2%8L)LKCaB`V}_ z;agvvx2+1Vdhf^jY*93bSV#;O%3eEY0vzpZA*XNbhJT!I;YMx;@ z!S!-XgkSxZ^P@IkH&; zDum9$y>|%jHHtxGZZ;fj%Y+42OJVofW>~d78Y&ak0-4VFnne6S#I^)3cdmqEoHz5N zQVF!ygu>7oA$WDNtzoO<1?IiQd*+*@6ZO2hmCoM`sm*ZW$!&5rq{Bc}LXWIHbc6U*q>=a6WH^1si|l>!g>>#Q zAcD^xGg{L;m|HuVn5&1mPDeCoeDm}a6?AN1O=@2=f11k3J*oh&_cbt%*#pa+Woa>Y)FQWU!C*1KFLdAS7*YCW}_KHfAhX2KHBN&JKY(w={bT=p9y zO$%NTH&0q@z~t;u?<_c2@o=7M&^!AfjyrndD9m|%?9vSe6=*a!sJCx5Ht`4uFp4s`L$3;Hmm{n+;9kPb%gZL z9>RVdC)fKnLV(S0BI#%X4f;Odwo;H3Tvh^)5M`jZOUS2LSD3dYk>rbO6pU#lfYeYm zu-50`)SVNcGf|N@@>YOn+)x3ofs(xHJJ%sL@+Fb!w*dQHkC_=3ryC~$hiJz&52EG$&afICfF z;owGRunIW=CsY2ww0m;A$m1+bRzD8C!u{lLMHF;>4Tn<~*TcRKS@4F|LtRJ$sIQyH zlj3W8E35bn@dymmf#2PDw6(N3^rd|M)(cpuxiIDNVD{V zfPOA7#k#_-UNdNQI!t;U0KP5C9h2EJNiV63Th?I+P#FOJbojU;$*Sa`AUnc|=GBeZ4GBN|fah_Tc4 ziT1~2^4nj7MsPjkD>gP$nsc6(*sY{W`AKZNa0Zn+lFsU48V!}}Bc}ET$oxo6h{>G+ z6D_I?_P^<8zfZMfg1zkMqqlmfhY|)kqIMZAqbNd<3OmL+lgCGfRx0kAY-`%_UCK{y9aH= zZLKv#ADaNuoQF^3`+8DvJA(-eHDU%-lWA4gSH|E+F4;M9lqPR}&&)w7()-<;^n`If zurEEN%ee^l*u}zG_jvHN(kC(^{IVZC6X%S-=Y)6MRQ`qNi z?^xr-cgX^F0(^YC5ln~!xE~ONnLm3;(WOdOa?TAlEK`h_mW;E`=UV8sfi&vK`q13! zdWPu~BXcb{Z>4S>8SUnD0DUh~>uyRzQ;)EuwP2O^j~b=kzD|+QNS6 z72rC&k1=vKx5zV-vt+>8nI=hw(yyU~^vliz)OV^FE8A{QUzj)3c&GBlzg;7YeohLx z;%x&{xa{?+`y1)wlRe~)_I4gD0qvoSphWTZNn zm}hTco9`;n(YO_K<@-E#^;an>xyk`(Xcoz3q~M>}CAx-VKTZ0D>@gWFlJ%gKiJNtc z9=dpsey__kJR(;}owk+Gec#Nnw=$J-Xf@g@5W#-QtR%lQXJgW>M~3mMB}s5VKbyoT z!iLx8%vv8YeB-)}h^tkw+7cpQyF8wRUbF|L2WOx&(htmxdcepn8rJR#fh$*C;J-Fe zc%*-d$Z$KK+)J(O{r+SUo3W9oHVu>X%Qr~n9z_s0KLlNKPQuS=*T9!Q4P-l(z}38J z&R4t?{6a!N>OCc~DrVqsRzaLUgfVT`9}%X|r}4(Ye&&$!c|(E35~3=Y&iLiMFi3f3 zN=}KDlYmdoQ1-_Ll7hV;e#8y7?<$746azTmyA%w|wc$#<4Rk*(gp&#!Kgr=MQAl%z zKaEkaE~<^>%B_Yc?K*I5_e=;(o(A>q~dgWgPPQH5x zZ^`e+tg#R@dzXS0>Bq4~dnX><9EFOl#}L1*zzr$`^!2_)?3LjOU~HxY>-lMgN`A&d9ggTO^= zxIVEAW*+f|Jx^wE??(9V43~|HgJcn;h!!RF63;3&54J>v9kn*XHDZR zd#BDbWv{?9$s3Rm=trdbM9EIyA@byn15r~fB=#}=G^;0z?D%ulP(o`Po>~7F)%ykb zHKzaJ>WV>pUNec$f1}M$^1g-SzYlneQNibzxIeGI2EFY_)YfZ}?P?8%l51NXO(2MKP<@BF4?#wBk3VX&kL-}Vn_+&L7?t1r=)^#Qzn&gg&{$JUc zJzJ=n+Bd@~fo-JBcZdmu!%SwfG+uGsOn8^lAmrslUb5L!I9l)+TD3D^=usw|j*Ej^ zo9`3d+AI=mYsc!E+cGbfpQS%{wlN#-&tsA@WRXnjr#+7aan=VLT=M(?ReZb$WUq!` zE9ZfOZh8K~PoHsgYC7@`yuvT@M$nSqhO;*op<1H{o*CPUbN9x9a8xWfJ%0wmXN7oG zwehfHum-wbZwG?|1rWGmD)h~n&GF#nVEsUh;obf=`mscdecZGeJdTD?;d?D~cHb&6 z%W5(Fr0In^)8ldcg#xzy@S_&?O1Mb&0|~r%lk^?QgY!9S$-{5&$$5hWVx}trx@&LJ zTYFW|`;0EW*fd7d3|!D%k(-}7h0*)7B(9DTLfb9>sMmrpoWQ=Pjke+F-?If(1ZfqhM^pv_0fq?ye1(@WXNV=C=i69~i*egp(jF+D^LP zMmBDEQ^@>zRyq8_Lm4pDdb{u1MR?h=b)9S;opv#GuNeh-^qrq_1ApPywgUH1Vb~cD5$~+pq#GB~5_Y+@or{NxDAukm=krm%Ct!o25U}y#aUV&eJQ|hW*hbeaTN| z+Qev5;PQ;r|I;DEb~@Cq`y{z|WdgL#E+Q4Vk{pL%B}~uv1rrBB23WqY&$zZ=-tfFrJZNTMxJ$m5tCA#UvAZwecPfu{1l!pld7-^iw z_T393lP6WPQQ5L=z4Ck*OQI}4u49U&MINtpflG*Kmo>iM+OD`W3i+Yet@seD;| zv2|l+Z4_F*lcKkLikY@@M`nAw95gt7C8IavNw)1b=0U#}MYfqSpMRS;4Oo-h9}U#i zbtdN_SON1!D(IcAR(fZRGTLfl#wUGPpv_KE`qIL zhsXk_6j&WQ$mz2!aFWZr9^Nt)tW%#8X@i+yui^r$+Gl`AmN`rj=4P>?%fu_+oIE+T zlBf(Vpt}y4AgmF?m3|8`v;HB5c5&_K_L*q&VDW)xS`84kdp4XYqzogHiPm&N89^HrQTGn!XVPEOTt52zgUo#3j zv#5V_F@JgF0{+IIs{CUgRrzPCxsK@jGx=-Nl==CYKhSr4G5>(bCwzJD1b&F>rqdlZ z=uiDyO3`*Fv%N|Hm7Y4&*1!?s^nl^S4UYM)c>rcS$cKjg=itcSaVWd&2|JIO!sCCF zAx?h_l}Wb2tQU6pA~+6>RF2^HtGO5``3ec(ad-d?~df* z489D1;ayq2MT;^YcuM@^{&V;Zp0oIEy}!7-SSMyLTZDn@qsf_;ugsJLfSMI5Fz{qO zbm!^7@hjy}=b_B=G#i4ipWnb;-%+Uf{1c{->yU4w$pByi-+lzOs19(otD|lQ@5pT-A13amD+j%+?CcK&2 zs=Rc=J78#D3-bAyFynI;$1c_ctC}v5pLY{pwoc%ko-Dz;?(iCnuMESSt^!bYn+@ix z_Cngbbm-Zu215n+i8ZJ8ZZx@uuLA;5`MeT8CTaoy#RMmQ^uRW*-qeEIhVKc&m&ke;V(Ro7v}A^&WE0ibKq&T z3AU+b5rv8M?3PGj_}nlBPY(3aKT)gbobx<#s(BfPwkP3Ufm*zOx(7qorQ+*l>Zp4^ zjx6*YCrOG`V5@5b(SJfANMvo@4Qn(-*A&K8}*7e&h1r^ZC2n=J8L(%kYnV z@5G|_fq2$uDSF9h!Ow~iNUY6-wgWSH+xq)KJ%1W6C;BzK->bsQ@|Wk$lAp?3d}tBR zZ7>JsUw;k7GY3ic%~l#YZ!7MQdV^{0QhXiB20SR5fzix2v|IlaKX=R0?VRW9*((v4 z?2Vv2PaAez+6WuYmXJn=D2!EKfL9Jb#qAAc=xZW@UwB-Gbn;xVF*1T}pEY26hd*rG zV-DqeE<&pNMTklcg;1+#_$a6b-Dh&BA;{sYng;wJ9)eQbPPf@04DVZf#TU*-{M=}H zzSLU{zPf-lzczml|F(!Xf8_fLzT9?W{_Z!W?0-H6WdE%1WcGD16nyxKwhy>ADB)TX zRv&=ujwG^ZJ`WaGP6wMZ4KT8r#Vl!?Kr$6ZiBkGM5-KFrxOw7#%*H!vT%WTBSby*} zEb+X+F|&GEYbRBFSzS-d7D+N%cCIk6(j6R21Hhf@f=v7#3nTj5iI&7JnB%e=qFW5W zt40mxT@9cIq9bWPJ5J4#j99-UKia^g)5l!aV@k$lYNF?au^JlqL~Rm2?z6?!zMGNJ z-h;zZ(U_Pjgql4D_)^{)U+CPx#_!j0Wpov~9! z2v2@MbbZ%@WRWb~uI@6peSA78m63omZQ+mZ_>NjgLocg$fneV=y_8b z*A33Y%V!s(O3ywVPd$TeRaLl9A|4-v|HbzgE?|4#K0Nq>;D355I0$^y3w6O4hf**| zDg}=`AEht0`(r_ZDoQV&iHY*%RPB&D-gRVfzs6&feVmKS4hZtEi~L2;lX;kH^c0us z3Gge<&*sO>c!^t{oB6ZKmYnS+>r4bfBt!f<2UZn(g!Rg zH7w-C9g^gIE4cvEWZ%OnT*xc!-NZZ8x0rYQkS_1_dU2k^J1yQXtz4M0O`mt-vp&!N z^i#-QqsOa^TgW?^@f318m+@fCh}Y=0gV%O(1&?2_m?wN{3GeY=D_*|nAS4=1;aT)6 z@T$x`+0SE17&l`NrV~vzvR; zzW_7jQXr?Xj0A}KK)?50vLxs{JvXF>HQt6*g9#f@hQ0Ff9rKTxTJW%Z}A-$CO!M%bx?!B2+AD5ffqBJeVcSD6ZquvYU)H!~D*B+dxl!+J53G>6>Ji&{T@^Dt=OH?0h zMe_@S{3C1sqO!Ose=o-)3DS9iAk-Qoeiwx_V4_KK1cF)^yFI8L?vR&qXt zJG8GWl_pG4LeYCuiKl}cwb;$+zNe-@|H2d!Y~lf(K_i6f$5Nkb^>prFJM+1E8vDg1 z5ASf=?tgU;Y3k@iOr$$$DpWA}$G+3AG3CUCW|PJ7N5JA9_k*yyNhZ@UIMKEb-aL#1 zlW9f}cSr){Pr5^$`83eua%G-|M-6tTN5b%F5vW-}ncFd?WY!HI$d_(_U$0}qK;{ea zoURG-S@BG$#RyH_uZ^juJ8|mUNX)pKg<(C(c*pH3Dt*X9<-mEU$NNsZ*-UbnV=ZXc zW>fxzA#&~1C@W^B32!g8k=~|in6>30G&oDa3}LPZg3IP^(9MFg(>B3_b@A{+*aEOUClI!^QH!gt)cp`?D0 z4u9+9p8W+@c-BMk6`RJp;yHmQL+*n17X@D5EE(SSs+GJOx2N+0f;D)yqw{&Wr|0kl zl80f_pd?TF?s?em{sl@UzJRoiB=1DK5^s*ee%?+KTi%Yfj=ViQL*9S4*7HnJhZmu? zpSM)!D`a#Ko)%A=H}{eEaQy&C+BjPE!_C;4`>i|1RAfZ=wUG*1I}tZ=~J>#7;wxB!wLbBDgWIL_7; zag2@A6Yx@oDBfa)>A&Ou*hkD>l-zlb9x}4V&{~E%C5hqd3|oBvvyAz@N|MQ5qYniw zlc6yCEWLfvoz>Xp4_5WgAf!73t}RVtyaQy25x0Y?inWJUmrIbMeh?JnW5F}x929Tw z0Cy!lc-~h4RxW>`_@^73@8ZMurQEq;wE|N6AH%~ltsq|909Tf3fJy63a%5#FFvW-A zF2`~DV3z?+Y6I|A;1eWW`vNX2FN4fj8RR}}hQ+^H;c<=+NRDyy#+P0aGhhcj2d}^a z&g(I2VG(TaD~7tO`OtM(8JxI|A>Yf*B=_TTl%!60s(&Z3SPVEz+JO8cIA_=%r@hL#9DPS zDTp%wX4W|Q^kW$u2nt~zCrQv&s}XANw;27UjIiliGPc{j#`Bfear#3~taO}*Z*T8M zf5Q!U@?2J z@0Y)zn>JWciv#7SA``Y})S;LSwD>aGj@%@z9-HTqff`lF<8w z4E2?g(fMJt_ewumyhoQl6iB3x5*FZ~o;F%`&1aIwiYUptNhLOfk$=RWw8<@ni)w@9 z)Y)5&=Fa;da?NKVzCr=agk-_zRv396>f2beV?*Pt+y+#IukbbC3_M+TupXmd4fxX=s2CyE{_}LMbate zf9TW?W3+X21tj zYcWVNZzi|q>;W&;0(d1_2|^x%yiYw%u&Fp1YSUw((KHH%u4qEV+cC1uYdZZ9QBN0) z1=71ev)T2o4_SSODn=;7fl+SULAL4*(Pk+_V!bn-^i*@*%eCdqg8oBkbp&4f6K++z6GcGJn8(ik!6Jzeoup56QVCwcr*5q2z?Olml8cBOqP?QT8E zmZe;xvUc61nJ)m(=5btfW-k7cBH}EnufVBNFAbYn*(#e|&X||y{8U|V8-v|#BR8OQ5tQ)m#(M6MhXLMfFbUdmM zPeYaF(wNY2)KqJscHb7@tVgTqqB3(-O$n!G41d$poWJvnbQJ0yNJ6W=)%Zy+40{(I z#vkIVF;72|c9mPQRnpVR5{FuL*^~{yo_gEp4NcM^*)6S7$CRS}TnR0w4%6>dW&7}otvDy$Rz8z1V zOPG*~7i#4BwmPQk++?h?wxpjTuhF|&7wPZMTC}15G`(T+gYJ?IM)C1Ys2R-p?yZjE zd3kTt_L-0Db_I@|z7d7e9Vk6sLH&9sV#3lbgo=8T%`FY!QGXqp)yp7Qw+Ln^Hgk8L zUN~N=%XPq}!0W6E7>X){zpn#fT0s(6RlEY3@@CKw@d7E|cp&^Z2)vaKrzO5RRgl+u zGotl-KY6BjhU{=^CqL|}>HDl?62H2MOsfr`bL?lJgw9(=Gv+W6xY|k9Us-DCvZbBq zXzwyKfhp|aclibe2cM8mjonPdxB?vX8DT8%&Lkl#ZxG#88W@zEN*{i6pu4^*5K$V% z9+ezFLd-~rfj{tmlN zOc2VZ6*nHY^d_68m5_th2{1a>2ZqOtVZKW^OcgYN&9a4L?(sYl@MteM_z6JAl5ODC zp}~1n8i~|OZ+PzS1|_saW- zVR9s-wby~gpG8nHzYfObB4E*YKG8n%k@RdA0tv}XfZq{d;dBN9^|Ro+y%6u)ZWk__ zp8)3@D`9TtJIJY&stN)+L8O^UGU z-AD3fM>`4H|AlmWq>*Wn;Sl{Y4W_+41L3ik;freotmAT7Ds?kqYmO*cz4Z{o=w7C` z7q8{|+Hz>Uqcz_9$Kd8;8?gH4D(q`hMoB+;EI1p74zCoExv7G~3AG#>;S_#3w-gU} zc_XR1N2Oob&{7dy_N$~=#IDT@xb!IrUojKj+?9ux(mHk(*G*HmP7UXOHkiJ=5*6v4J+`-_&eI!I-TC(&Vj@CpTpC!Uf5bZ z0PE(R1x?$NpuSlL-X6<IOe4#4-|LDI#N^c&mVKPr{GJMy#3nrGIW}7a{aSX0YMEen! zZF#>B&mD`!noCmnsOc@^vimC$d+z`}UU^Wg77TYqCE@lDKlp2}4$p#L}zJVZ}`fhFfx zxo`#cc4tA}OCK0&WZ_foMnV(KV9s9)`fjk0t+1=5Yv<45K35b@_sQe388i3=i-h^- z4I6RNld}kQ-_ati2_@J5z%nk+emqYb_js$)A6rGRY$BJxHhIn((N_&|RXyy1Xy*pa z>5EDKnz{&4k;n1Gv-_!}*4Ch*16mxbHKa2>VRYGJNm54`uz0f&q*c!M87 zXS*-FdYT8Tt#aUNUm9p1Y=tviXGO&O7_j^)1U3F&N%>WPAwdHY7;%@pf31K!Y)+uu z*il?_DIQOFa@_&8^U&t)QIvkY6tj$`;JKNd_@pob)n4)O{_j%=+Q%^3G9RPowqnP0 zo4{j?pk>F+35VPqS|;(rep;Y5uV?j9l!2 zBju;?yWwiAaN{yX8ACW*u?F4W+o8_yL_Fruf|E8p!MTC8+`YDe>tQOuKg&3-fT$X( zKADX3Zth0!1T}mzZib^1IDI5Mmz^u(iTmaUp>Ihbs!b?G$zEqn2`fMyw+cMj5QWF3 zLNGo)5#?)((EhU#j+i^*OZx>x%H=i_IzOp+z#7Y8O>BU zu8r$S)J<2xqUJ1=OG`%HHzO3&lEwc2 zgniz|ip!Q`O$cR6HUt_Nz%b^Bju64kjH3Gci ze^Ysi8?Hh8w{j5nNd?olRiG@-s0sC>%Lm;^Ib^uRP&3moQLX&fExK(U)exJdIlN=9Ervsa%nZ}URFm+E}J%2Fl% zs&BS@Tl?jF89zP#*#S%btgWtm-dTpfY%bwz&6DNd7W#w-zl-qc1bKee<#v3q@iyMD z-h#Quy;%*vyHxp=K3ebV$Fp-*;a*=ye11+22M3}IRnIh0t*%)ZZX1egcAdvH^CR$b z(Jg$C?uss#&fs=|QcRO}M*997CiPk3()&m8YvLN5tiAy~t$Wz%&Dpz%qm<0&Hn@M4CGAI@#N?XC}wLiojZvo5dK@f{cgOrO!P@j>~ zxU6{##xlq8vRfQ3S{jO4{-e*Wx<6i5PuLnr`(lAvn<%Dmk9SvNOAg zL-h_SI`Xsez%zTeYAOsGG7~^!j{xl8ydAdR3OKHEB8(14K&a9HSgh}c2cHCZwucbz z?`i^#o&Fqup^|7YeynTJ3d13{a-y~VFiBn)#5{QLivCGo2x^bdH*QeSC)drm`%Q>B z4*hyd<&O$r|GKYqw?hXt7v*}Fd@`A(T9?_npTDW>m*XZ1A{KLhYtU7LCaCeq5f`6SL`VN!xZOLG`bBbQ zi4)K}s!0YD4NZyOawkgD<)Cls17@qBe}hf0F)Xh;MV5ZoW55oD^B<-)2r7?}eb~y%-8N%{0u0IO`gLUDUvm=oc?mvuk%zYtuZfoH9Q5LkldlIP zVaefj;6MdHwfh%I_Rl0Gfy*IqyAGU{+{fNq{EN+;8B4dU5yp4>F3>`sdUi$ZDLUg` z5>*vliL?C5h(c*0(O!`c)&)Ym&Q)88o(acj@{We;<-6eTqO%}ACkHxggg`M?rSb25 z8{%_s8qvz>O|KqZW{AjsW6nb?JUE9Jj@qQUb|2={t`wdZaPZ(;wE5!Y`&vM#8 zEXr|b){QP@>{)T14s%(oNt3qXpEsAON;HoQT^t~rPGy0?cp8|SMMIOK2K3HOID_($YX9FE+0<5+eNUsJu1II_k-r+w?!rVERu^1s7Ydr5#7gsK&utq_i)EOiMR{DHb;% zzUnU+f1l19Fc9JG-yy*h6rav}H$#}0@$3Po2Xw+$zi9ZjBpPmAsDTF(D`DuI6L=rp z4X!%VV05<)8Fy$Qi+E{lH?x=)CLTr;>AjfSww!y{%q0WXvD7Kdnf|*TNz#{&5#v`T z(ENlU6PXld`alkm78RtI)CZW?4QttTqgUywLmN@#6%Wt#r;$zJ!(=+Ihs0X(V8AGk z=&Pm?>CVfH^4$`8-@l1{v{Iic`%j>*a<^z)x)z2d57U0F2kgpJe=1y+O>G=?u{T*5 z>k`}P-a>i&D9^EMei-A{q#N)l_blj0*23-&J`iB51Ltl%BeS_Y;_9GUat@k_v5Ew} zJ6)MtE3ZVK4~l4bauIVa$e5^J$RUkT!XAu!&PvNIVY;uoQt>WB@)j-dR1CKpUoZn- z*Eiw4cU;!Q(i^w6q@a9|Cx(oy!}x#d7<6zA{_A$Zp;kTg?Iq|nEJ?EB?y@H*f1zoL zr|G7DBlMomEWEt`JqMUzOSE>A@_ZEuM(*5(Gy^Y&h-mGL^tji$B zFIkZP#Aed%%SGyc?&sz!j>-0R#xm+M_JmxK4}s*g>A=5c4DNEO@Q5!CwNZ26@z^b< zZq;mhZYfJ=ulYp}I*8-Eu^pH*YJ_vH%|eOa_UK$ufRdA^p`WW0epyq1afc((hGUQ1 zRPLi+qjsa2{CsNG$L(IY{WN=16+bp6vU6NsGv-?)(bk>g4W=H(uq+omUdO|N;Ik-P zuZQ1!=i;`N8}ML~I)?gr;$h1Xs;F|0?dzXUcl^@8o>?j=&s&Vudjv3NB#mxUy})@< zOQ@=in3Kdzp^M0KwF>Gb zuS8t9-R|AF{p8YQM<}{>5JVE!K~w5kSb5hDZQnUk38A|YS9xml| z?G@vltx5pn4V>;e#RV|xE~H<+2m1?N!)5I$GGf72jZ@LEIWXen{H zqez+_2*FSLN@~*-PLKRqgfHE;Q7_FRs*>17*X^~$?x*jljnP(|uOP-W&Y4kfnlcyX zY8TLTFB_?J`DYpu?T)jtYY}JYQ_S}7Hx+jL{{*SmfkH+eI z|NkRG6AERB451=3_qq0MNTq=$sYJ6hsgNdVFjEK_3P~A?43X*H*S;H+QjrFlCrX+O z%^J|}yuNGw{`q~^`mXhTz2EEixz{?2+gjG%XAjreXYc3rc%Gj~J)65gd%^{1EP4c1 z4~u(QxR)f(ki?BGK0{}c^uP~&|=&7t< zZZT6f>Js0VDy~ewK#zUVlw>;wN;3)LPiQ<b`=V)Q6l84~19Ih+T*P8aGA8gS5v$cfU`sAJ5}i-vGO9>})JpQ>t&og& zv?4A%r}&xbN>Ckt53-gnCp#*{#AWtAvLdRGc<;VI+I=I)=sZuN6lKf57yW?O+HK&M zJOVduyTq+(z6yp1V&F=}2e2qy2Eh$}4@&ip?JS{CL1$YiM?<0!!254u-W4YB;yNwSNELCb^XFp zy;)3m;6dCoITq(Q48(1z^0es72s)=JNNth<{7ki^ts>BffUtO`P*uT6`m7 zlsGKQT)cjtqPS**qIl)hRz~K2VkUJX#dRTz#j~|t#LGVFh-GDj;$2cx#j++oVpB&4 z@p}0oV)ucg#Pf~C%s==cn>o{rjh0Kn6F-k|+td}vv&b0OqQdXVRx)nPZ{BaRWFbj@ zp+)-dhLDlYTI9fk$M7mSo~*4hCmHoSpt@ff?!r0j-=n}XC-lO4GK5`jNM%Rmbj4$) z7>GxtjTPUpUo0MAr29$O6v$) zWOfTj-AWMNl1&x%e%v4|*?CC#ZNhw^)IbfPvHEH9E-;6b{27NXJg@J;Elb?wxR$(^ zMq(=wLf*D_gX5ALq%d4e=1&YDdZy8I+mdinS?UEgkx^vX^1(zrv;d51BB1qh4(w^| zgXVFTaQ2-x&di<%4=hK(EZIXaz9NDgjOqrbKnfmjV`28JRpe9o8wl>IAfxpZglBuh ziHv^)soC_7^c;>PA9Y;`8N30!XM8~O8sn-p>BCq`d=ne7{t7eN-h|e-d|7isHZxL` zfl~{{lEI73$OMUAP|~m^VvjOtO;ICu^E>F0k==qGkAb-C=~&V~N}J5<+QGHkniHFt zVWek{DtY+X6pqAo>F-)M+-dq7H3zTdUIbKO=Yk`sGtdC9 zZVkoLm+xZy07G_9g|JUUquBDL%Iu|%r>M;&0M5n5aQ&C#!Fi|*`IPAcs(Kfo_sK(u zTKI-D-IfNWbEU!Qgff?&Y{KMTUZWd6x?_vzD;_>AU@?3S{?Fh3klrQP7HsxjMAJEWK*_pU?1I*O7Z_>q)qm+u+xz%`o;=1teJ~ zW7lNfw{Xm1G*i%EhcgE=Et?~({#6?DyD%I_H;!Z>2Zh-4Fhn%$y(wJ1UI?es6ma0- zg}k464o#K?o`Il^_MQ*9?GDlC;Pe;=X^&?~!CTpa>xuV=Edh{YnS>9HT*%Q|O;N-MC3Xk!{RW zW%4Pd>_M_C8?+$_&yCmQ9QtD5Qga_HFJC|uqzAB+=4e4w>UGicn^Q?*gc&JwKLS+; zr0{C;9t_;M1>&FCfoexPJ!)(~cAA8fi?-Xz=c>Y0Iz}BJHLJDA#c3K)seKBTyKE$eFOQN54-&}Fu2iyX&T=wd zY7lWP6_dssY0~N*O`^shB9YTJk~ih1q}Mr=Q!+ILjeT1{s2fJA`2TsC`4papu7!0{ zgGJqP@}#jh3)_D2Ywp}K{8f7c^^ASFSz8AX$sM0z-(eT>am6Td`IIvm>Yzn#)61}d zdXeRxTZmYB6ISS#12gc%3?olzIW|@dK{&Q79={ zDMu#%qtuZ2clNGlnh#wMq2wSNN(73@_lj{+4L@#3ixjl#$?$m>w(O7g1*+BB$?1*UE*RJ}jV&5> zl&Kr!;=|Y#taa}}wj@i4i=3QUXyrI|^6)0sAk1N6m2*sSY${WB&t?`?Wo-H%XC}w% z|E3)XW*$YgOsHDHHpQM|l~X6OnL&@y?~1)Zr)@qfTXvdlSsu^)kB`Dn1%a&bgA_}W zZ^xbO_N>j>o&A;gfa;$YLiNK}sMwi<7rsZZz5&0e?++6^sjtlh@S1blK zZ9=Q^BlM$WJ*aKm3EGSF+2E{0@ag(1{5d$B8+%EX?Yuq$FZn1!#Ax1I$+-(Z-aN#* z@5f>J+c#L2o4_Kz-$%`Hfh=Eo8vE`vkcFzYq2KbkY>tOCdrZ?{(YR+IRinada3#p9 z0}-GW)&fJFqu|m~WfFZw2{VccaA09Hj2Tsb?Pe?v`W@6Fb0oNsd-J1o2U|#umT#fc_RBCo zr!M$;I0Rlq@P5*}wvb^b)JVq>O=5jmjmlK&LvK98+ud!{bT_ZF%8DZyM~i5dg)CF) z9mHmgJAo1R8gbI^7Bu};FFL!`M=)GY9bzBx-t|Y$anD+`nZa-mW}a~i@0bn2G>1C4 zIP^4*$YYSja}aL*`9&+t+o4f^49Z?sC7TMYLB3X-8Nck{*V*N`h2L(f>EWFO|8AiH zAa^rIZNAYh?8Apz4}H4 z?DLn$m($*2&(#PxKS~u6lHxg~wmfXHy2Kgy)M3`kF=T~#7@2iCo2UmSlk~aaB;#-q zacetHYzCD9eQrk{OO7PRey%0@hCEj{^FG&cAe?KOdLO-0Pot@V0-|}h9-#UjqDtBYbO@3_Yr>Sh*TVp1;u8xwSsw`p9H+(5Ik)2-e!3-CTX2n)JbH8{ZaBH@)5L_ zOd$`m#9W!i9-6;A5R7!Tf$C{T==5@d!y8U>9e34WcRKbi)q^{&U%b4%A?qw)kMqaec^e`}&$h7z-IoWUmB zjb`Zy63ofJ8&@r~z_I6V(zU1kp}TDcjw|kmXlmXDcw%c>)i14<)U+ zJ7I%D4h&yjf$fiq@$=xB_@>;K`y&-cw>cPd7b42&KCdu3<3^ewd5JS!Fw#dL9(WJe zM9zYxYLn4wT%*V?2=Qg+S-}!1UxE6BAe?8koVjf&zykq!aL?XA5WLeG@0QtM)B4NY zG>N^iW~m}s5I%yO+MWpqDiXlp$!qj=J%x2Ps;nR@gY8OR#5NwvWiM$W8)4zaX6g-Q z=Okyap;NM1>b`7d5VnKKS|#xBZoH=SnJ>RDcwF?(${9m1Rp77|J1{J%gRLzGA^(g! zTzXQ3!P6qq=amv<$)2af%Z$Nz#5<~9nGLc!@f@eyjgEO$_<{HGl)f<&G)^7iY~H&< z@`NI)Fmkx)ySW4UEb>8%xDGtZ=b(<6lz={o@o3qpfJH{TxbqLRIhFDI;bfJN``T{K z{dPJF0RbM^|9K3IotMO6{~@s676;j3?=ZlbqO8Mw_H&pDi<;3$^QK3ly;BKvgsq2A zgTt^&Weez=zT+-PC%_}T2D`Oaf_>Rkl4+<<&SuHLl8!k39Ay{=zmS1;`%F>A&0Tc2 zhYQxY>##G&fJMJu%C3@uJgd7KKlwdFX9Ep>%{q#?I3|G)?{9GDo+asATLL=Trex0E zaPsU(EV=qPmPpUaCRg4^k_-IVvZ}rg(f+WI{Pa7{{e0~V^Oi5g(`M3a!uCLrZJQ1< zi)WKm%L<^OFXo`r%(+2_~%-iFaIAu-=MASZ@--7R-%fYafkaQkVICl?IMI?l5J3 zK9TI<%l8mA?GeN_OT(bR&4Sr~BuMk9a~S6{ zy&DGlo#%FnH{gXApJ9Sm7(B8ZLgJq~(8{{CkTt3TZwi_yoQzwd$V-x(O; zv4-uwT#i27d3Z78KANs{!j0{|cxS#cI6dhQaIb(3xn_<*PGR^yy_cS{kH(li3HVSk zotlnY;NabOo^qaLb4q-i*I8<;67Yy^@-aR`iF)U zzJiNgyvE|gL9+I?8mW)_4H?s&p`mdTxV;wOw8uZWuN#kp!c1p;v0x1E2jR~4+fHW^ zkG>)FyvJ&(KQv<2DBM?2kDf~h;I!Lo>1!)3OuMZGmbVAfw*597otF&zJ{E!Zw|1~U zVF0VoMuS%9MB1cb$Xylx#7X-Hu=01=XvNp>V(bvM&cm1eelE?XMoGb-;s?OY_HlIS zTYTZN9afz^0#Xw%g6-ZOZZDA|BVBhB1^>6Ot|5+pCVLal#pU4hd=(tnElUa|5@Bxq z9eSd|5!Dt-u@7(G;_>e2YHPnaT(|alc)XF%dz96O<-Ff(!RTGY|JXO!Qh6Jc`E%^P z#Bo%+MS>01P{p31`J#(j?yRr(Hs>BZ7Agu~fO}sC;=4X@JAH`!`4LQ(yctUxi_M9n z-X9oJ7D-$j?~*l%B_z?Ng1kE+NmiEbCK-SE^ZlA)68~sAu_S(Vt%sh9tyE`z3k z#hwq)x=n%mp{&F|!$z=0r3Y}YQ5{A!oyUsEbX2*^uYtNv*sn}o9A`V0b!?u+J}-NT z28QkQ#WLP6cBMNEnY(}sh>#JDey8seGTq+9%Xkv8kk4P*4lC!Fc9 zkxuNs4;wB?!s3H!?gWRR+=#oJ*_~pLd^Lwemkfl^DRYR!+EViQ>TUAH z?=vVSo0Bcu>&TwnSII@~T2iy-7U4z)vpczNZ1mk@oYv|)HbaLr0srsZlk^%GyEGlF zS0=L0_p)*2frD6CtOdqZLx`Q)E%H<=niLx45#67gh@Bkoscq5^iFMLscyb8%@%w0l z+6!UaykpdT43Yq0484^gTjJfR5>}4-TE<$)i=iQObaa*yd;6$n-;)E z6r5(ZI|EqyHbigR2%Fd&GQ>q=Czs%}fyC<#;PW7XGz>aRvg1X(M&uHCS(QSre~2Qx z*S#kb3gm>no0Nsz9ChLB+pa?PceL=nCJ52QbaY z3tzTwCxIQNWS`1CnD1nTAqQS!XC}jqvrX`z`V>49n1=yIJ6TL~5sucgWyzKY*;Lzl zHf&lGOPid+gxeOniaTceHH--$e9@74miZZQkrtlNw^kzcuWPo&_>El)aj zqI#9;M#iloe)uiBK(OY72!17G!RB*XMD|S#bl)8Yvz?vb+9oN%o%IROnJ>jwUr)sE zv%iVH8%Z<0+gEY>)FtfFRbBRU$2Gb}?>r5$@Zzd2&7qk)?1`)09Wt-Bh@5fgB`55j zk@Nifh`sMcRbiug!7EbRisk(6j9pUOtgpnCd*R8 z2shD&On>P~h7?#4u7{8!Q5f+&piJ(Y&mf1-EF_P9G!Q*;Bl&pdKChE}Mx4|69O_T0 zB%?K+oGA__`#B5pcHsmFGYE#31_JZc6o??^EjQ$W3f9hV1f8W?^x*7NocVPR^xO@F zA2#2?)uoEtVk1q|x9}{LcbSl@zRsp!AWimn^+1xPBv~!^1&5o|iRG~$aIVN7Cja>f z<86Dv@N+NhyJABeqcd^fS#zj56hUVTts(f=HXLkv8%;bCphH{*=?k*ZMrPFjCE#{SC#IqaEly;}pi; zY@~)~Z$eD{eY!(sB7WL?9^}>yC$se0=%%?PqN#U^aPTKfnx7I)>ZFQbcXkYSwEGY@ z=$Mec@f=2)$uaoYBt>pV=h4d{E?kS*J3(K{RqWlpjaBs@V&lUz*rzQESf7I?OVvNh z3WNQb?PfzZ=5R3kVwlJlPn^Ok`P{wVnWvfdpo^>lp5mn4am?p>1fJo#@LS16nzvpC z!#3WN_l@=hRYp_cs^$e%yT@4R-F&-4ZUYKv!`W;TMGTgg$Aic33Bq2l#xY*}`B+^FcE^ooL(_kuCFh0@ z>Q*z_FoU)4$-z}zE4KEjGp`WBv&Sh;`X$diy_{IN$71$D`3&xFT}Quvx`x&Sq!MU=B~?SwQ+O2AaJ7b9xX+0tv;^mNf(9M3a1W!U4lyW7B zzS%-@{oE9ypkqyRQ|n;H!x*^wEf)IhufrW>d)jbh1w5+iz(`#=JTfd7qIV?V@vl1U z?8gbHH)JO3T=@j|E;+#d)C4p3!;u9XIfm2sm*S$hKvw=pi9LH0#A=5fW9-cVcA_VW zt@*GKuEBAhM3v@kRZZLs}Pkl2TLNM`7DkAiNg3xyKh~ z7udn8&+BN&NNpPQSJ|pl98Q++JOsUWGhy?~DbUf)doOCgfY}PEaJca}cRIa^TD?i; zy)r`B&q2{F^uQ42ro9hKUimTE-~KG)u@eiHaAtjT2eI!xAGluKaU36uiT{^F;Oe=Z zT#ugy8%8I?{LVpS{R18HsY{oH9ykeVneDK6gEbty(u{@8Gg)CyD9gT?&8FNq$wpmW z#Gd@x&z5|)V9jG3nS5U}H|S;@R-eg%%ZX6sv)b zK?&Vzj(iNBjJncJC_PPq*h-$mUZuBOx9LIdn^+aA^1E^Cv5TOlv6m+GM=-hK2$m2T z#0szGvyD-InWo7j_R#bhv-8VmUXjJ@hNVCIK7J3=F@KL8Q9Ds47}sl3N|?eT!i_j}=Q_Y$%!$B8r~KZKsQo!qFd->|Io z3U~g{0P^L{F0T8Nm{atQ;XY^OLoZx{%*|%pF@Z04>-%XmwPp`goP4wdvP59YL?UGTQivY3=k$|Z)9b>xO@Fv(`U@LJ7UHRsVVBMH(|IGprUjV5M~jflo1A+gx)MikRW!^_fdbi3vR@SOh~v>JAk z^;_FPS>By?T{9*w+V-S9xtAMXGK$Zmb0Yc`B6!#&O+LzZ(O=9L_1100Xj3WnT;7t+ z6uQ&WnA_Zym62RYgagGRg?Oq)fz&w+K_`jX?9zxG_;qAG_vEJ}Tc@qcstt|UyI@cD z_RTsr%YPoDzk*m@?`}5e%}AUuF$)%6{fdX=l`t@x15bs0G-l&E!SDzbk~Lb1%o^wp zJNQ15c)zBiS-*lf{e&a(d(Qdc=yJPTWNTL?5x`j+jeF`)s-^v zYV;#Hn-&qFFp!KtVog@mY$4Va(FBJZLs;fO&>tyJq)ZdQ?(9u!ZyU<>re)K=2NSt5 zDZ1=UjS|ZCX+gi<6rNk;1Cs=YFi%ny{p)SGt#-1klDmMf45s4twKD`QMi()6Ql=o+ zFdjX}|AJEzWBB(dSNNs@TkwvB3Ob3slIjz;K^e-0dliAnkD+?wNf;1NjL|Wo{>a?=fR( zGbItv>9Z!AI6Sa*KGV4T6;H{}$8F=xG3&1$8{1aO^!nGaJ$#(lTrJD`H7ByQ9s^b8vQ2wWV2R3Pkw_dwQmrPF8ogw&m^K`41Z$B*6Hl^c)(g1lmI@y< zhH@7UzoEyu4vhcG=UF~Eg@>o);QCSXS;A~(Cb7bXhWe?&U5|Fb@OfYGsNw-!{Qft7 zN{(lhe{y_zO6+Eu|m5W@!Gd51$VTs4I(cDvm=R$5}Z^QYv`m`2WJbyr! z*obF=7m?dDbec72@3K4*<%*Ry3oNl#e@rXO7y!;J&jsGGCcVZR0r_ zZq3*_K#px`D8k#Lws8T&>*$lx@AyXg1cn$dMe4nQB!|!7^C8pt^M+zBIwFbFxhP5A zS_YBk1Mi_*(GN}*FAzMI8AdKF2X4TMCr~t~8;Wmy1`pA8s6ID@Tnj5BnKXciyiUR{ zvX?Y zg!jnghzY_AhR#AeztzG7k)H75+*QJ8yBWd&&RIA>=po$IF+)hp{DuDOHVcQxE)bq? zUm*OwYNb%`mxpk#%N}8>2zUI90+sHbHd5*ErhZ;CPJQ1E-byKD(q4} zMP9%l;kc?zykGVoY8O??Qcgs&rFrhGW#(vc_2p^ej86~PtK1&;?53vpV^j(&)sSW@ zULVE1Up}GMDS#0lW6@xKJ{z$#hrRh~#SSiAjcId2*|v$kxS?8JkUoRMyslWf?3fda z|M3f##m;A3N(37jkjC^^m$8(>V)kucI`1P(**5RfEZZ)TU9?PNYdfa1;V%N1>dqUu z_o*`b;+@J|P2^Z-pc``|=W*TeAk3Qe2*H-epm5N==!6Adq z5AtI*e(Tx2!TmV0E`T-G@Mm$_s`PhDJ*}$Pj%RgD;nBiLY-Q&Qatd*22);4{^F(0y^yoMvuqR%(de(CZ(kcK1cc!LD)z% z{OLj)ro7}PdY|J)=*1xOj>fiJe>U{t8;m|^#l}>p;m>s!aP53IX8woat6XWM5u@0; zp$Tkx^E@U!Jf7*PY+)KLA?(2TG?sbg75<*Rk7peiv%H;#?7Wu}@5Ouze^*Pv%mu1o z`2B=v|H@3!_7&A|yH1z&NPI?D@nly2W*ZK$`btX!ZqoacC5S_I6P#RM1v^XANs?78 z2|LqF#w9Hy{nrPRFCR;Y{XO1akgX-e8K-*|EI#?e=pq3%EHWwZvg9mQ#jvbe2EVKR3Y~G(Z&Dr{w*y9 z7Urh^L-%iEF0i)w5AOf}9ccfw!L!=Bs(-%!bN!#X!Qy}K2F~2tMqugWX!3vC4gZDq z=d3I(t@*P5C&nLVZfRjD;0OP|AAkSp{NpUm%sK1-nek^~%HO}a<$rMh|JnSvFcnyv z^7HTCxBq`-{99O=S(x$-@jo&CEv#)!1^=rF{?+*ZkK@0EjfEv=Y58x*-+!n51$^sU zT3P>#_WuvZpTLF_aQ{4j|NZ>`51&640?vlxO#f~C`_KEIGZUCu|Lges-|2tO#@dYg zm(QR7F#cL_0xN;}zaIbpar;}EnOgA?;NPyl{tM+>TbptmKL!3({Q0l+zoof_mHGdQ zf&bU*kN;BnR@SC`75@K@KW3bTxs{crIY0UTZT$If-2Xq-u(r1NSMyKnXB?#8UyV=9 zN)g=>XmG3xwHq~s3p%etWAwACb)qg{*vRL2a(@_;@wm@2$~W?i{}uF6ficeZtKdG# z`_jN|4{1SYqs_6wk@PtZ2Io7BRz5Sq(dvG9w{55>V)m!%)jkAW8y(om*C8m}WzB}Q zbi$$3U7}8{WiajOm+HS~SBZuVIY^`XXNw9Ke57_cz1-eT3;djwgSV=AF6@z3!H=iY zd5?KjkUcn)UO6@#4@{kbcD+k!(4Qf=<6H#p&40z^<=cUUZx`I_uD~T~$5HY-|DJF3 z3gn@)VV{c!tDX<{`#dL@OuXK{JwTnrxEw1AQqf# z{UCGZ1?uAUShRM-YQUT37`!Zj?v>6)%llDOYkUsMR7{4ueXj(^oE^YQBL|c|@i}7+ ztFhsJ9?rX5%(i=;WO_EoS-@;1Jy<6+_AGu&wnC#p1fJzjniEm%H3p8M%_mCr|u;O5MIR_*T*Ni(#?)zN*X z+@Tkn(X-uvNGQ!EJBol*dfkA9bLNugSKotj&ncFCbuQD%9zfLcK8SXI8wl>F9I$8g zBn*9~gZ*C?WBj5Wv~-#yOmsX4YhF*RzAUB96@^KN`ku|;=KqNnT;jbDnoh=x^!|RN zbtbYRH`fC=Hh&@w(Y(XOdQA~clpRAwA+o3`$Pj#AxgPgfhrx#4aKWoqPLSStfpc`# z5xsq@E;x$K^x~#zw0x!xK2nfFQJ5AuRx+5WwGrIVmQGor47>L4h0aBZHr9FVRBMt2 zy3N{#dAw&skX{w$I7*33r^|Cal2Y`X^eSq*Z7dX(3BkFh3@109z+YO!Sy4|d&6Y{T zSr2xwlUKTw>7C_t!LP+N3`J_+cr>H>u9_nv*!3)<4cy6D_mDWcJ z+85jd&9Eh~cF1F!#RoD40ykCCcvq44yIW3%SA7P*^C{4hcO51!9!xg;nn$K=KTiU` z3dlpjBN)AO9w}YMGhA~nz*db?;o-(2p`GBA(0I-U4*aSF>i15H9Q?hAce9v0n4d}p z{=7{K8XLJtlLqjAG=gltl?g#cFSr2>gWy_XC=qFSk*|AHh+5lBqEE}o)Yp6Y^D{#t zmtG30y2yR5j_2Okw%Q0~X3+Q@b0ETgB^jE3j(VHL&?)OoI1?k8^0pI81Q~Rb=xMq= zu{fkbUB%IIn=Y|WSjE)MM6qAKs7j1Y5D5j_u&iif7Qx0p@72Wq1alI%w0a73|3uFB9_k}58Y2dfxjdhmZ@`F ztj=?)^NhHtQa^aH-WkvD+E2eL4nvn5{!Gx|A_{^=uuFgYs8#AkYB2l=chPDz>|Lb> zu30BVU$kSuZ%H~F{J0NpIqro(8E2U0z7{5^TfzgyUV;6bQ-byCN%Z7$bKG2cqpEI~ z9==*N4sXXxLfQroSa{bJZ>=25=T?TmOXFVXZI>o-Fc=0(i3IltR&tN0dI)r!6R~i~ zNm`VsMmFiCLuOYDycs4b9GCxxJm)!Wr;W$j1a-&?5(NRA`|<&hJ~kAz#{t}4+CVMt zHXt5uM}NJWT;GUt?&FD6E>TtuAC^!qrrrYR`9GYWMi-c6c*C(7I<#X?GF>-5ms4A# zj6JrqM12Rt1t&_6LEAciIMsn%#&UOJe5ejaa9^R~lsY$7N0)m(_9(Z*Oaf~PMxo%* z4_aGo3c-bp`|-t<`&(fM+qaa$_hURqujd2^(&oVFzBq2~9VtHFa|$`$-3MO>Y2o(6 zPIx#Y*=BFZIr>$UVC}EW(ODMV-0DPiZn*P(E_zWMbgV0YA>JPKQh^8Os|YOfL4J&U46<{2!|?v9(%X$` zxGsAcU2Z3$a-%;|`I~Dn?}V7UxVeq`pYRvd)#zdT7<=xl@I1G$b|;K))&zr?XWT&3 zb2ua5I6Tja2P3mN_%S{U-W@nV&E$^JUwMi2zHcl_-!^51tJ-l=;bLrGJ(VfxD6rPC z>LMxM^Rz<#5a%?_3DSa2aZk2C#Ip;Sjpf+$cvD=5lG|Shl5U>oCR`dU8g0-@hj;%F zxp?Q(NduRP+?Q($s_nxC+Ut8c_9v68HG73YbPV@pg9)Z(%Tcb)oBFH2#o~bD)T!E+ zuGo_#h%*ewnvbb8c@7Bs30= zgXdDcFyrET);#+-^Xsx=RYmK$vhO|A!OTIV-n0T%8tuo^F;#fu`40BLXA_g-`Bbw6 zZ@8x6TcEr*7CxqAa|ss);rh}mG^E)WBfnpQAeYG?YrbFf+GQ=Ibmh~^pU2rR(Ia&2 z+9hfkEr$y0ytwsx<}lcKuYm0<ao^q-u=_0|*v-ke1n95;j9jg$yje6|wemXN zRJ|g)Is2q&!&V(Z>R(S>8>Yls^b^tX>qU|K!*8f9%jX*j^U?9980S5)W-l8g*y5FW zg3ErfTtkgBR18SqGCZx=>!;eHp-O&Gy32-^&YQ>CC>ns=Q+3Rs0irDZH#Bu@13r;V zrCKAEp;^iqbHEa|>wl+V#`5&@+%X_e&EO2T8%!7Tg5F1xxTe*gkCDpY`t2R3{;m`& zX3oc92exye6Q@A}j;Y+RvHHZ~_84N{6NVGwZb5;0KZq;% zpF4s`sgAW!#pMN85Yj;lBeufW_lo4+5osc8-w7(ldx*j{3-Y7*KCydhNM81jg3o7? zU|HV~vZr5?l(c?@2?2wd@3%5)Yj}ne*4J>IB`)wRdscPMV;ds1Ifr|GS%d2`mBXph z$GAJorhv<31;NPu33RvB@9F^a5N_LRDZ22=L=dV!q}wBS-RF33?n?0(n199*YCk5E zlGFCW;~^`AQOd5uFz0r{2Ix*(M@9+WLgLppIJRO8x$V;m zeapl!#aSq7(+xz)<#q5PN6eXBXaG00li;+kj$5!i1-et-3KSdfa8GNrXq%vpUVpDd z3eVPXm$Y5T<-WtT>`FdVYI3CIumsF`e~3S~Q|I1p<-N4M1=uu63Kl2uxg*Jbp!Caw zeC4%pXym{%T}_GFw2SnLoen%G2?g+`TwwWKnBH`n=}?o0r!5pr;w&r_OZ z6ayEs@1on#cQjQYm3!AN%}n3aQlo1+*wU#BZ4I9MdLvTsNl7Gf-Qx#^QwEX&4q<}H z6YZ+@FEfPU0iT(c*-Lh-=_LF1awsz{pU8X`2-)r*)6m3)&to?hqtg9yCaAo}vvV8R zBtEy)p3j$$sn=w=2L>?fc{TW;)EcH|E8tXaGhSAZ;Jr#}Y3swYD871^cVfuL_m>`u zMkK|C9@Dn*4_Zw_65Y^fW)(d%s-0GS-Y!Vw=3%L}EKAt*7RA9& z(Co1Wp4u^-)##tX82w3XL){$KeS9*PuOGoxZkq=x>637Aas%CPvkO1ZK8)UO=eefU zVourSG@2}xp~=|?=qSb4B0I8+d+{Se^hk3dHrb&dH?W#&Pi^K-F2Bj0-fjX@Jw))N z$_viw#&gr`BSh~yU(?SkC!zR^91ojx`Q*M+c@21if^@h@!EMC+#BS9rwJeDKXu457ce%CbU6PY1adwJm(Wl`M_9#y<*J>n8^E)qq>~@+Jp&||}`ob0s zt!AYSimW8ki0v+Xijo_ZnN9a$oH*+t6~>!lz0o1OeEc}h4B3c+?$e^VAI03?spD-X z?%RsRgU7Pw*D@IL$pM=RCFmJmrziPk3Ei9dR>H`F6EdYA9EE{JDT!38jw`W!T?1TXt~& zURLoeid8$rGs~m0*l}|vQ#`((XW`DH@inDZw>Jjj(Lb{w?wvm5dwk?V?S|3~p6^6w zk}?DhvnR5%sfxJn`yBSswgdmBwxPK~Git6%XY*DbVJ1uZ*~D>q=m+IGNq0EA$6b`%p~2!h!kK#TFx=Mjk#^~X(|vWn(EXeYNWQFre%okvuBwU|8Qf%f z+HzQ9B1vqG`f=xj%j|FHRu(dQDziKz$#fPvvyTU2Sr{?Kx*^lp-6&q8)4YV8^E+} z`(OXkJvoX@BV`!%@bBhast$5CkKC3pMhV~L1Qk>6@YVt9l?-iQs4I#RxR*}9xR%HGl7xA9liQ;LZX<`js2i86} z8#|U((v#kAF<@FTbIpE-v6~Kx-?S%**DuZ#x4-VA_u8sq=*AQXm;7G+^qVi5U$RHb z6+iHqNs~Z)KNf-b!h?P*usSw|y?H)}OS?W5jJ3z2WWXU%T{DPejm!h55A$%|zU8>_ zS{x^~4#UaTeYj@FRkY3#;4LL*X0`XWXzrzb_-?8#e!tp_&uiW@)9%-7`MP85W_tyw zs}129wC8D@*8x;A_hNS6m$Ch&JX7fJI?PpBhAXmXu~_-9I8gcywN)I!)^BdZ`?X8h z>BFJmJM<@~F=9MS4_%L^{WY;{!W7neR|PWSSKyAI*%&BWg5_^2QD%nOnE0|De%QgaL0pKxA3J!l{s*I>qOxb0$s#S=IuiKD2qWdt=CwTIW@ z>;$vojVSN%i>v$Xi1CK!G2QPfPJJ>IC2Y%aSbH%x zLeBK%4Q_G$UOJ><6)vhBh4c7)lNY6~kn8OU+jlPj8KajJkLf(o!lYX+?OW(|LCje!Q+nY7kwA1v0s05YS-L9fYq z$k-u44%_+AQ-;>`bXG6bI`v(!YuaIc&+9Vn@v_7R#t*7DKNi3~x206%^_E`Io`ojm z1@z)aDOzr%L=7%30ILr&h6j-&I(SM-|#^u7`tn=5ysvgV5hjimd$YOXl#5ipPi6Kzw2Xm@Ie< z%{x=8-Tf8t$AQWCE@Kc(x>!V~Hc*>zbs2bN?+Nphj&QHm4xbcQmM!+hdi?AhPjKpr~%4 z6<4w>j@x3q1*aJCxqOm$Av}b@AGb!%wAxKHSErhO9WO0NsfnV2?x*P3EwZ9N>Pl?Q zhWEJ1Mh9On_QTbq?xMmo33fw@aYq{GaO+wIied|&q0E_!7%L~inCc-6DjUb{8{WoA zDi=U&H-{rm7_t-99_a3J65mg-rmKsea-+{HL&C=M@J??sX&(0m_VOO;)%RtH3Ow|bkV8mYM-20^?g>X557 zH@^8j^_l4oinr~#Q$tlzRQZ}`B3Q#=gF+Y>eB<9=+u_1Tey)9QrU|{Zbcoa- zZo`EUICkMPS{+u4i=Ccu*P{DHH=1t#KNOvLAQj&m#x2Q~HL|u?ix4H|yyv!1B3dOX zT9mXGrM*NF*+U3fD1mbay2ynvA|^!&iMOC z9kL71!=KmB#6^;$qlicoou7=;ax>mY}!d zI(v4AH!$s;1&K%pKgw{%u1oHb^8xeerP^?KUNi=7EO92fhfUC;S20k$CK^_a`~c*g z4%s<@M8dg;(r1Y3?py;YZo@FQ zfOwlMTD3ILYT)BVlrk-zbUH3Tiop`t{-!+kGTMVR$Eo4s+HBHfA{1V&y32fAD8{8^ z++hpnn4{d{d{o~#3G_ox!FG|3vW0_|XhT*QskL`OZF@N354TeE`OsmgDf!1#LI8Ob zx`xUdX_APe&lm-5T_$&rm+;W`YG^35#uejgp}6QNY&Mi_q*@vimwEpnN;(L1M{l5Qz5~J+L=mm{rU1s#9Y}tD19WLvG6`3%(=9{2 z{8Bi96x|4B7FyS{bCbf*?ws#1!%Lc4Czr!~l?;fhe+GM{X5d40yI|tfC^9zk8@S)9 z2CK^fXt!NmS^ShJ$nQ19W`WlsxF-(1czhgUs~5rPn=;UOt`8Nqy&zfURm8`<4PNcw zxyOHIP#2?os;zfScsMkaX!Q*+NNzSf-A}>d-BOsGnFA*_1Va0+G%#-z0J#uCT#Ki{ zqGD~xng(2=VLchUV-T44-7x#O5B#tl1i#Bip=O~ZB!82K|F+JD2+avpbloUhPaR40 zecmm6Y$`XsrG<7a(;)|9PNO|)EmYC92BqJTqId7VBC|5A;I7U(HqxmY#=H~ZT)u7K zS~^3ykr#?c!{|7?u2TRs<{mt;nohq(NYTz5caqUojihUBX?)2ZbR#>Ixmz?7*|;1< zB5t{8>-bMJ((D5DjSYeFOBdlp@-pE&-rcyH-_49DmqX`M4V3-d2EC9ArOV69iSuJI zYIj%*UU&J>rMpMz2S-D)NlKDt%ZHJ zrF5?O3)(rq76OYcvG(gHye{=Kv9x@G#tLSlfO&@~_xU@!^7$@ubKf`iXRbYN7ntC7 z@rB@W71Mv&PYKzj2Zg`(K=zV1~<8fyDEZ&I@LRvwX;6a&6T%EbK_njzicP@o?gu? zb)QU6?5?F3Emwl3e>u@BtVAL2&%w%Dg+f0oZ!$DinRZD?S_l&2lp&^=_!XZ zT4SMjX*o(>BnD^NqS4(#E%f8{0kSi=it+Jhkm_bf2xz=d{hq6E!=HL-;dy;G5?8YHr=6d1-A1f6}wSE`ErNUYG{=BOY zZM+iA%-aJ-zf`%oo8y@Sy1PlgK#~zusnXI(ljt1(7^*jaBNJbL4$Z07MA>cTkhJ;+ zN$ja)b`&3E#@Vc=-QPFx=fO;7n*IpN9x|cbl~OeF&~M`_CBKFJDkQ2;fhtFFjHL1`H2HEEns7oK ze+!BOZDL2O4?cxOW^Obr_9R!e=M<`VVhoDH`}EU|%^>VigMD@7;HDL1{c@EonK_M* zhWq%^jBQTb-fyPd)g%$&nZh_4VX4LBCM3YrG%fVsuF24%CQ1As8nWs;OHjr>V-gl2 zLiP0rP*F_Gy@uM zh*RgRWh`tbqFVJO%-Z&D^!czUV|{ryEy-_WCOvk6WfePN{Ec(0$=fokUr#mY;zRuW z!=~6eRZfomxTOIc_Pdd#hhs^t#u|9KG!bflu(0UTeR%bHClh;akWRARN>3ejqi0=S zlhoNp5c>Wi@lqY7mlm95_B>dNbhBTP@T6PZ)#azSFM8S3J9RpFk?%@3IY@AajY1e< ze+ZdR5^0Qb2ia_*P8A2GxJJ7?IQOUtx?Z$GKF>jrUgE@&vV)xUrztQ=^%7y$WeeB0 zHj)+EXXy7$Vf5UpEu`3rcREZJ=Te&{auWv6aZ~tn(a!cfWPN5V*?WYcVasbt-7#bI zZmkFm`;C+xo^y{FtiDMvTwB3KMPK2DZ_MPZ^}CquXB?Rf#YJRe*m-V6u?0173lLJ? zhhXQW!~Ike<)Yo@qEFVIG)>Z!R&15#B>QU7)t+R+$%%qYp98XAq)Z2-0$|j6Chb@q zM@&w!Bx<@mXEtUTRr&e@jn!VyO*;RH(^@tKWNyzF`ox`Q4&E=PZ=_anQ3e~hi8@m$ zGsBQNugtbCF=sdf+pnZ<=mDt~O3)yU9(GmMS89~jLi)qs&@bWcoR@(l9NK0Akp`#P z_A8S)t>{}s=}@xOU;iwo(r_*v)NP^TwMvOlGv4~fwm}kmD1hu-HhSy!Z*lgLUJ$dq zP>L!H-$$+doYPSupGmuYl_?hc!Tt7n&pqGpm6KmSirmU|xwhD7?z!?1JM?7O`qI5S z#Qc&Ux7j6~ew%!d4f`ohZ>()c|MDC`-5)XVVRIAdjEjgclHdOQmsKJ z-4UaK3Vr08Z8y;zFHg>NYEW10WGdV4MWbFCQv3QfbcU)2W8bqGDVQvQeBHUos!@_W zsT;%DU~9Bg{F#+9vxky-Nn}&%bz*XQfUIQ8nd!_qvL|mGomli4t)#=~*)tvHWkMvG zRr89mpfX6kybay_(ub;lCs|J(H;L8Q?}J|b%4DN28qlm1Wv(e;9_OdElnz|fL^bAR zq^(LHc~`S!yW?Ua9;3p$P=8SOU9!~A!;?_;^GJs3``e2=ue6-jRRI$F~-37hD*!pSy0DtD@# zP5o>F>#ejPYB-L3ZCyp2a-V>)Ac4*+52dTb@6zKreMDDZ6y@-HJ^xZmQpR(wJ}f;! z4PWI@vzrYxCF(WIIOD*Xv0y>V%b9W=shGg`kF!)pZDkt^OBtzX9)xEgUP_Mxu|*I0;x}# zj^8;y0JeQR4R@SK$1RMZ-(F}#)s9BeGCz>8eJT75<+XM0P$jGWw?lY-`wM=i@PHWH zJxzoMb}?ERo9M+RWg_~sk!<^9LBn3%LScNr^z6$ebVJBoYLRr4oEjKm3$`W-v(f{> zrX?EIw5X6x5{F2-{9?Kz^&fc=?#t?KIw{&m;{W=99j#8PvQ%7iKQ zWu%IEzut{xazo?|8pu8Y8YK+m?2HHbA05`W%`T(Vzv4t}@^AK(j4QDzTZ(GYQ}ygXN6o!>eI4`fG|o zS=jp}G?%oJ=B9GGdh{!rEh>S+9qN!vo+VE2N{1_-#UQUpo8JGtiOMcsNLmzaP}7hk zcAH8cz~Cw1XU%lfr{PZbE6xNeKBGfedvzp)@TN1}ERA z<6cLhCHmc@&YhCOft?`7vzJ{xg30K=d~)}bB&-83bRqC~Y4DVLY+~^l(3+AD;d_0+ zU{^D`baM_g$Sk9cQ=(B$$pj?Q^;Q__dkO0AM8MN8PH_6v2$)qOxcR7narx80Op6zR zzK3GMY~eRFfCf?H@Jb}x9|8^6`asYS7f8bQ3^<1iDxPs@4%IR)v-_Fb$WU92<{*%8+Mz()Mv%lJa z=sXqX#d1gZQSc31Dh83O@c5EcO;wPYqyT^GgCJ)(jaB;MPhJJ` zbBbJ7G;-6xYOKs^_8>oR3h0pGe$RbOPE1T94-418D%n|B!!w^}vx~#`jesLXbMYsi zSbUG?U3JAihoF#aSf+CXTgjh>qr6`$fG)y&U(bSxs$TFgZ59r^Z3(a5H-dZj3-GN{ z;&QGF>D78AjH7Si6<0;Z5_?d4y-ajP_Z_VC~t8#AfoMS~;u)u*e-(3I-jtl7h_Altd@t1JTTLB;C zXM|c+9CzV0pAPK zMzy{wI5}x0PO5lD1*_$;e(`p^=4vM!Qk@OuwZ>2RXJyi zj?Paf6@kIRyt%d1&QAg21npET+MT|d9Edf7ByspmeMnO7XXEa@f-44|=vJaKSERER z#+i)Aj`=bq%;p@C$69pPIz+=?Tx1%r6_OhX{siUkqhm-ebxTkp4&V0>Ikt!{5yM~eb|ROM zFpJuh?_pK*&!Ni-`|!st5%{>zUc9n@B0LQ5fiIR*z^;EPr$3FQcA;U^Zq^(0I_EIW zua2T$T`4D2TFW)*E~Va4HeAX0qx5w7I(j%Jh@M)1l|H)4dwrD7fOJ_HEOZ7qVDgVW z+~~qR{x+Vwr|D@O{6vYK{+&lI1?nR5pMdu@zh-Q;W677z6S$0N-dx$dOyqG*2fAZE zkz9{=FlTlqXlxSDD4im9yiBV#Exf}D)|>(3XLaNQwNO%s0r?&$)zOjSDmND_R#W(~Rf)Zl@|IBfL$9%+ra##G%M zPd4}|a66W)M#cVr9qU&dKTMo{~jN()152>NY z4)Pvy(`h(z@h>RZngn$ugvR?~E_$&DC+b?s=h+IF`T6y%#Wi(ifX^8e^6bl$sGVG{ zk0MwPNpRP$^PT^{<)nW(pV>;1<7!e>=<1VWxgWxroVwUr8k<^yerLT#jXdjZLn8xj zo1KZi5u(Sh8gk(VJo_6hqx;o{>D2r>`iSScSDYy2PP8L#h2j~Rttm6@LK_T7x`^+eqI#)nq#3I-QZ%@+qWF1`= z;zpcL+~Mv`Ddy52)Y2)z6Cr+KD4+30R4zM;tF|lWem-d7hR^>;pJhzp{^hx$+fEPB zggRdotT6?>bFhNd9jd5ZSqYZ7@O-$pdr@y<6?(U;ohttfCb#?|Xk=*)9T@N+W!2lD zBO#gE4?U*iHd@m?wV%;WpQ*&pp^Nn{PeKJte4+E|Tkg8xmgcHq4(2M2* z4bfm^2KVbt2$$Pv#PdydQ1xf#GaM{BaJ_Vs z>vf8z?j=S)t+~60(r9A#X?mnnjI-@d;8Z?_aDK0OuRtszuuqM8K0VB~mn750Jt0(g zt2lkMSsiM=`orj36{^x7MMod)qAlqgIHSKQXv?Yb!r?3-X*+(Hjy-dM&i-M;ZAea} zYhCrZewvibCD_Eg+HdU4A(>%p)Hx?XtGdfw)3G-m~+b0+;Jbq~rJlff0J zeK46=BOzJquSSn4d(sDGJCNk-B+AU@7HHgIpy=j zby*FGU6q8yQft``)rqX#7pUl%S}8Y;HGI> zbGrlYQOhO=E>5qGJG-!(CfG#L29q#4Q9Ylk*=MlbfB1XWPXdjdD9@o_0e2?2m2CTJ zgYphJvHp+q=}4zLsV~T(oof|ngWLeM?BH=XuIJ#5k{nDg2%&Pt&*__8OX<+#K;n`1 zjvOem;Yf&(TNpcz+c>1ho%27-NzRSrJ@R$jPKi|RXrMQJxIKb=?}=nqT$AA{db4Rw z&qWenu$U(MW8}PXG8bx60xnlRqVS>|6m)bxC%nFc+HK3CVE=^NcP^t7%I;9-lcwxg z#V~f%aULwEX0S?aJKA@94Y_VU#75#jgha-%FVuoiUab$bx#xkK<7}du)rFRc%!T$K z6Y4m}fW9nV0(afWgPw|Jc20~8&k)z<>VFt;UvDdMH+3Va%-bEAq3 zEFTJI%`#c9qaozaY-9T8P#E66)Ci2OW|C`~me?sX6Wi&$#p@<%i91 z=!hBq>i+}=WM**RCo$Z*NEz<1^CkH5B?0Lw7?IW2Bsi15;&{QQV>o%E0z~YTfZ}8o zdc%^b{RG_8%I%Q5+2NO=;RK zM`}M&gI2Fp!V}xC!mUgb&|h(byuP@s>|cjATAghMecLfo37A6$6vFY2Qr`96dI8#Y z#*?VLL+nMB?G1ql-)VK_1joNk-Linr$QcX}i? zc%z0J`17UgcNgSuql}`?5Mkc426mH3rBGJRg}nV2fWQ3i1uuBTa?wKAy^3c)^3R`4 zo!!iX@wKR$_YOq4iwVBZ7$azkZ^P#8|Da>}eq76U#I_ndr#0J`P@R{WIJ;H}jKgf< z^FvoSlVc8LUk%~GGBKE{Xvig>xJ~y8WND4K4tiB`4Sm?3f)@JsB1W!)$SPSQtx_?X zI8;n5-@am6%|nHf|9sH|%_qpJZ!#%!oPr(&5;CVW6AfHQg?#aQ?Bh?z(D>npXy}zR z8*1qT0WPv|QzRF4o+^eZ{;eozRRSyxzeJXrE739b?sV72gK*_iA3TUO!}sUTp%;sQeR6G3)wb-M`ZQrCz zo3nMexvP$GiuV6NZ6e>fvi!=18|m;lyZPL5VI-%*XE!hC@%vonFpaOBkM}G%4eqiB z&di)(Ipws^=@|Y@rSLS-M%ob+X7ySsfecySBC|~vQ$|6Q zEB_v<`*@C>5H3mIJ?lp7DGm7S*N+re$Z#** zS$cQrWE5pzPpd(N_MNy(zHwn>u|@;&R5zu^{5Qg|m_EAP9!{s}sZsN}86-z{C0&!I z3?1|NJ&2|${FYw~SL=OIJ)aMpsZw8YO+-KU1x4@iBCFr%;1F-xh2Wqcc zkpG@8rg4T5{-AsTo+k{WfgjhHaGal>uux<->*nX`k+n~UJuYH2K*UCiwDj79ObQRL0zIpn`L zu{`_Hnd8D<5RYr~Xp_AIU85+*C2h2)%1``h_rQOsRxgn5=zhxv*{3tl)@@+x3x=(V z7doQeTct>+u7%m#`31CJ`qNbnvAjDho~R5jWiRC3M^B2y@HodyaM>sixZA!+LPQt7 zuL=Noe-U5o55TYcwPE{Qb!PGRe&%O7K~hH~aLcuPn6`Ep?Xo{b-Hw!i{<*7IZ(R#k zHvf+Q&iRVX-!J4))kMzivkaYjDF?;NP9S!Tr>Ik>9^*61%sRh62*+NTi)|gBK#%<$ zn6Z8jsC!A^rxoe=_w(}*pP9tIzivTef;JI7heEV{@eL;FttVCZG?_lD$!BUjUgCe5 zb+|>Y4X>5cf*G&xAd!Aqw7N+Vo1{A;iP+D=1;_>$?ql#s!ckaQGJ(#0GKTDH=tike zuj8_)gE;#`EUe3U3`w2xc=`%^+w2~ID(2FtepL`Lr-@u(O^5%Z(~%(hs0 zT)BWPN8aK{on7?E#S~f=|D9Dvu_z=p3+7+r?;rK&%+8BV%wEv|#5P@^@%1X)P;NDS zaO?&-m79$;&KyFYtk$AqzEhCi=wzq}U4kF2{0qC(KA^e|eGs~v&=fvi*#7W2c1lqY zD145@eM(VKIC%*AAJ+*@S^ePshs zk=A{}@%{z(rDtHy!tJNm?vf$g z@^OdWXEhK$y#R_g#}UCH5g5vr1ffI*I=;h_o&7Z%<&}Jf&E8SSQ%itT$}|PHYJTC2 zl}GU1XTO+x0q>FRjy@>=u7GBGn9}iN3xeZ16w2LMi4!5n``StJoumw60)4eTbY~ z?L_}H8gn0ACt+<|$X^>o8OOlq~V_&B5o(mThswK7@LRza($BS1i4!nT`=38ln%`$!#HXBw%j6o>a zly3RgMI+DGv+G)yk<`9gA}%wHv*4Y(c1J9zxs3?*lbAtc_)eGWO?PrmZxW1^ZUV!8 zBiN+Zf;Ksn;hp`R_T4PkGQ|73pad!f`9!Pp?;57(+|#3$S2_}oJ%!>aSCskKU?~&k4H#z zW%2tt@dLNertXEH9sneH;$7OC_n19(BMIIa8KbqQ!VZG1=9IMsJQk6vT zw>l~8`MV8Hn6uzeG>p{l@tME)Lsb5rCbC*54Mqp|ac3Ve+^oKGx+_+T+)^*WW1jP{U7R^CT;>bM@{^!Y^*{Dlz*y>7F@hB>E3xgVSZp709Y^qCsLR=& z_`Q4pwHRAW{~g5C|Cuk{Dp5{{1YXuJ`-RAZ&-z>}y9y2Ru4HqfH=TasFlB0H@O`Eo zc>OPmxMx3b{-B$n@R__Iq{&Qh>#(Xopdlr2>5gFT>zqX=j;51CmtF9(-(&Hc7b@65 z$%t-z?+*Fxw?K9BIvN}iMh_huV#M6e@h^-S(B_;Czvhad-9tt2*5?VFT3o`oGkho5 zKoxu3Qz98slR;8nl~EY#rNYaiq+sE7G~VSiFizjmk}!E{9rd>KdYCA_sqP5|y?sbQ zdAKa!D}wDk&w%x}2H277N=r+&(AFgrgo0#c;{FNII`$gLh zX)94U^qnbt*Fdi6hp^>6_t@HHN`M{fAhbS(-K^M&nlFdbyBiE>(cAGn`?rmG*S8za zbk@OcWByrtI0Uk_vS5py9B4RB6grqk63f5}u%NLpSMe})_ijf{2z@qA>_n9+HY58JGx?w=>^0Q}#?S%8ZBJ9|JIt6JS_$*{j;`14wa}galr?Z` zCBr*gn17xO_FI|&=7ABcL+l-9lR+Wb{W6vbXr5p__(&Q_W=w`NZ^nasKrylxD&fAE zAb220(BZd}$UCu0Br^1fs*yREd2=2&c!}Z(dHz(?@+nz6kVpSm@;!+}UA$I??{XTa zF#{Xa@SQ^ga8R%e1|%<`(iP9hi{?7?()%hiQ+R>(4;)9W|*f#uZ#AUZdk6!TcAB327&5cU?p5h&^>_KTWm5 zQqV37B_iXOPkV;K{{MR^A>%=hxI96x&ZNMh_zBRgJ%N_iM?!Y~Vsb~Jn-p_d z#HTKd9GYna8?MMxZ~cYj<=|;#xj31O)VGx_du)jGbQgoZumrS!w7^uQ>o{a7?|Rz! zk=Y~}f{lhFal(WItUF(xUh?v1wuFqwXIoQ=@tjh!xxto^RP6$8+d^XNa+iMGodA|R zd$!{3JiLG3dvarokdb}dVV$VFoNkP_L`BtK!RpRZ=rFM+jkCoW#Y@$w%U~l6&2vJP zTN#|WvKHJ1?!)OdN%WCz1KILk2W}dN(yo__!J_3l@8tc<;?`-jY_Bc)XGT#%z5|5i zorkD5<8bSUDP}5V@Rngm^5A<0c@e4!a^50b@`oCFeD4oP^m7H5%6ssyDMvq`Hz()?8qs6%8eiD-s41R})nKb^LX!f>LIO+QYe6%$1{?{w85}t=|zPkjq zeq-U_s0Uv2?lNefI)jeyn_{KWn1-BcXQ2Fra`0zM8gLK{HoQLUaH|9UqcR)IUl8CW zqgMFKK?&A#MK9TUniD?WFoy4?yR(B!-+^DfJx<#rjz#lq@!aX|_}E??mWi>3$2>+dZD%=%%m2~b}M3yJn;O-{A zOOm$+pH+1P>(_6=r|v%Q*4qbbEh4PnmwKWMohi7-DGc6JY^AFYT9Snu%}~9O6Z09jF^BOQDxgPO&8*Y|2;x=Q~5b)G#QYsE~&vQ7T@-H%q-GtLox>edBGj{xZZ zegqr;RFfY*)B!7J(k8enW+=hA3 z*17?9{hR}*rO(5Q?UP}blr%o_TLVvyB+x5J=V!(G+<3DNIGuBrD(z8%a~55A_L2iw zRq8Tmo>Rd+_txO)%jNMimnq=%Gz|8{+=PSMI?=Q}A+TLb3d9}ap^DZ)$jXN#VzVXI zKQJ5bGl;>Lw*3OHgK}8nhbgx3jc0y$zahsDiNdWN_R#6)1C1+JVPQiHTg-FWXFP16 zo9iZ{CC;hnLfcs=woOOEcouDXW=NBTdSLk>2QJRPhBh$GP*|A{U7to_{2o!9@@yQI zOKhjbGfd06wUXHKd%Pq5;Tqa}Pl?)fTp|mjbLr!#LuBaWB>3mOjvDD{b8m{}saSvx zHvJidmv-;NDb-sbU(^Ph`7FRDicLseTn}5TtjCe%0pPa#D!Lf{i2V65OzL;H(Y>$3 zxDB5U3b)1AkSX&uVRx52T<@h|yw8aSNVkB^d?6V#U`PF2Q;7637xKJK%UVKqp|EvT zEzCZ91^)gzgR<_YF)Pl@qkWA7)`b}l(8lAPkd+bzN46J0;(|^jsqhev#*ZN1rAByN z%^xrpmBE&}Cn4%@8grpGk5TkkKnow-L+LdhFz#G9F&^^>Q9&h$oc{$ocQ1h5>RV{O z-7ScJ^pGwqGbVXHA>?1;4Rnu5BDqXFa!?Imx>lPBPmEV$HS-hU$GPR$Av+G1^I@2& zHCb%1%LJ-@_c!^cxtZqahaxk>SyX+F3fzsqZ0%8H3%|4z}@~mV0P&%EVZ46MZT@Y^h+bG2>(qovKB*@ z&n&v!!V2_yXTdIkH8J%agKuRTg6H=zaGN#^j{P<0Pfa|^`^Li5BZ-hOo&R5m{X-tU zo1ncW1e7Z6K>2SFR807YnxCAcclkW_gQ#-aASJ`_gL3BTmJN`){4N>Vqm0zmJdwM< z1Nc8wgZ}x+bm{~Xu6lwsJ>BIBSNf!RFXL1i!*i>T&RK?%m+{=_D;J?PayhIP`weFA zf5BixCF2n=4fdDtS(ziNX@BPn*x~e?RDK%^(rp=RLFF+h7_x@?)1hebIG%sD_aUlW zIvracKLB&h1O&W)60CqDC;Wk@_JiE|VCZ>hMHjDlNIGt4!;!)ccyi+}|Gj5SDVdn&K_?x{p%aAr>7hGu+(kB;OMjNg=@~8u!=QIiS9Jqo=kQs& zM&6~{X2S)okf+)!)A^2H9H)Fx7FWBd;j!MTSU>(6p7u=-m${q=PqYoZ`TIDv%MCuC zl*d+<=g<$|JwgZPuSnzf3Ss5j$)(AEzcJj$WVF*tl8%wjM{|E}Vs@1MMrBiMLDFn1 z+WEs5+{DLFR;!0~`8XGSRhGoUUx9ezr?0SGF&1yQ_YTM0xr~(-^dJox{#r@xCxvDb zoK&$gw`=cns?>i1g`Kbim(!QPH^7g3|8tZiPhCjlMYh3n>+#rv@DBCYnt06avyd;i zg+|ywES4FLYg4b{{)}1FOkoWDo;QvGK}CRbca@`cPe;iVVLkFb zQ-V}odFNKsG+6JN59Fc@&XG<6iVQdlg%~uoLk4%2s?{z`u*1Ud}aE&#{Z-j5fl6&iDt8iM9x-d*fm||}cc^NpD7=5=2q6y!;P_z;P&F(DjYMxOHNhP}N_!6$S`^9n`zAGCCY<)*weDML* z%#%>lQ-N-|#32_8NffU518ic?ATImmwW6E(b3|Bwm2<3ErL8q*QITW`O2RR+a0@*6aYP(JvCAPrLy#p|L=WN=R^N5lI2HdoF zQ;_K{MG%D-@Mzmbradj0SvUP7*t^cbeZLg(Hihe;bN>XU zxQid#euw$L8gTPWS;4=YI{d7CD&DBr4}XrAz(c^&kYq7q>{_#tI4%q zbuwr;%x;a*!3kc9IQJ@pbH65m-{>?5NUB2*PR_JaE|!Or^=4@5qaDlHoY_WV$cJcT)Mw@?|Zp(WHqjSc=u7NtxKRJewxzkZo zvoj3oGT7&)BD_l+A>u;-?eE)=NJ%FO`ni~FIWn0(-4eh*jS`?l*9|;ZC!iT&m)Trv zf7Wcw0$Rm9$Se z>r}xl&lm5vEWoYO2^c#x;Z(UMywjivE19O@9TO{XrbrY1xH^Pv|Mr3DT%^i2#U2&- z< z`c8p@il;!=Du&c?HjiOv6{b&SLxMNUR-v6F+$%4tdj_FpGfa3^~4JNnnKX=1H6vu^!tV=*Ouf z7VllN4>z6&fE9u<_)df*ZudL~H)SuwY?l;p(cS~=)<*JqcMYtV_68*Pg@Eb)IB@=v zg?u*Z;%c>%*ihP@1e8i)w^AWGGOEDLTO7l4rabYhPp7fRMGbuJf-P2cryx8VjlSFs zV>d=!;ly}P%~2J9F65&IRjDbXo_o!qF<2Z_$|Km6hl^>|^$OPI@h2;}n(H9&$ZY5Ce)l1@L&yg9hn9izoN6(_;*>$GQP`~yH6>JZp z3km^cNv`BB*^cEBTNrMyt`oIyzC~u`owxQpFrK~L+eE$Itmm#Dzre1$Q_qAPP9YW- zhv{C^hf z;?qoeH+?G|857LK4aRUvM!@Y}x{7|zG~i0E7;rl_%;J4o5!7Gi0#%DhN9P6igj&k> zr9(4qII(a;u3>l&+x$k2`)e%58J{$wk9`l*`|y$EGlA%X^a!=AlH#0wKG1?zKO}zs z3`w1`omwrOEzr?M1XtVF#_6Aq5Lg9AqY>hgh`;RB3oz6&X7@n<`C7rxVT;(sW@M z?eI+`lXagHMT_Um_rH_L>fl*jfxiSU0T||^ZO5;@p_$e-`Dl|yx)`O5J(`P-e!ZZG?a1a26wnn zHxV}~e&p8HG-2!3;nH@W&tSRD1`mok=UW%|K!wakSod@ce)*Ft=zE&u#Fs^Q^^^`& zMabcrwj}PeeG>QNSs8>aQ01Mrc#CImk;(R9B>s0>pWUq9NgcJNHmx_OOG2!S!F@#< z>R*`%1tV``tG6~AS+^PNO5<<`Ca^otLfIJW2Y5f?G3NE<$chi$#HwBVAF~>SK2vVN z=gfa#^ydy8`!E^5&F_a#zTTuS#xH{1yT%K;PT6#p-F#*s??VNPL9{Z&OXM&=7b+*O zCDYRr;FDf1AzF^C*<~fWxjmVy?q9Gs71DN^-HuqUx)_R(lPwl7(6$+oAcf+ zW{_vJv3VDcu*I_(`?MvNefaQ)j+g!+i|iA@;LH@VmE5OhhSxBA5u^I|L}wJ0r8Uo- z;LfE=__Z<>`>|d$3U^@V)ffy2;vq@A_D|DL#GOsyNa7EG>sW#}x}9;HAVc1)_a?3L zGq{~<=g7K6&Gg#B6>zxB3Fo}(7G{nc&b_}8PhWKyvE!$F*@P5zwr%z%mXzng2Dpia z$q&`+{hnG@BzkB&Hr&JNq7ppmu?trRyrB-kYsso#WzvN5mCSQk43=Kp%Xw$r6%Mb= zCtJSt$1hvzF*L0M=U)_g7%R5pC9NxHym=!1TU?N4LH-sNuI`C}P z63Nt@aD1?10On?h{uMp&DMQmq+RtI^&7WPu#K=DQ^Ya@z{Y@^tHv0?d zGvgA~t~75M_UI$uh=)t<39rOkF1i$Yy!2jN5(|Ku#mkLI*7%w}vup|`XuZX_I zdNq>f9E|6ygWyciAy_Z3qp1Y!jkH=fmv9ILt-{O7rK+ZYt-l`eU_J^>b zU<2mcwx4y~o6NteQRQQt|G=`oDp-GMF$QN@f~sF5nwBZBNl=NScj&R$KxJklufrbc zw_umF57V8lz?3I9Qw7HaJh^2PCZ~$o`ZtYuu-{(1GUSlxe@qc{)80#SOZu|q9xCjs zZ6(URyC|vuaRhfK{-Yz7&B6MDYxIOP1+Mo=!sehkP%vo+#7wY)?`K?4L42N1svO4D zTQc!fbsq>9vylVpQ`nCML2SesKW^C74zm1IFW#idNPCM>Sz>c~-xUcOy{nyx3IcE1K8nS5J#3MSD=sY10&n~^g)v4x?Gx`1LhIQ#e z+vykc&_~p`3c&Nj%xaMOtA$kknSLO*e%Io6^ zX2+XMjwR-eUey1S1x~1!6S|zsh?7JGjm*&4Y8A`Vt$LKc~1yb=@sQ+z1S?^GGKv$PNj^|2RsUgnV14C3e+R=$@c&Nsh-q$ErWEn09(3^$uSMbG1iG zLfVR`n*SpVoRZ6C2Q}gNAII_4f&|8Y*vJYd2D1xcCAe*jFEH;ezLzQT(i&= z*1OlRo6~IB`-oy3X*iYBxY0poIE})=A6Bw+SKi~I(~HQd#A;l${WCSXHW^0SZpFDH z?6~j#dxZz5&vT=!I=PdzK#O#28H$|b>SHe0;eCRgsU5>QMy?|Reum4{Tl{h!}L00;^T_G{&|@G_X3_)eNGqjLtNQdh86=6 zxW{^6oT4ChcjNI^=Ot{DCgJ`qd)cz2(X4CYRMd@-V?py%NWvz4TvD&YE*@fdaL95n zD2#-hno=;=zY7yK1fcVnYE)4kj7x*BNOlKhb3gpYz`LnD88u=M(W?1C!j9f$k+b7i z&*+bs6FEz|S}6g?s9(dL^Tsqceu|(~e1#4uyNo5*+tE657W7y6h_8F!;@&9*`1rLB z`Ly7ju;6?f;b&mFy!~TQZ5;Lr*N(@5q@C=_dL+x6!$7 z9@MgPWVupfl0?sc8r|(ah-|d-XD~;Zx%n*+^G4&j`*%j**^X{ZJ@o~La{BDTcLBX^ z;^M>&Sd-C?1EyD_ws!>XIJb#BIah+UNpYCy`w}gDUf?Q)65NN& zaMM?H8aCbl8plfDOM5GZt;xl~LoVX|fz#NocMW*#slQ}KfGO*1Z;HkPR>9m4+3aie zJf@M8j~mXdLb;*y@lR~GO;fNBh92!g`-#?Y(r+(WrKEppAMo3~o#b^-=k1dY!>uzWXkR^;4_R%=UvOxF`$3Mp&)OQu@;o9kaYhLy zU1CRZRyw^=_5)7WoA7_HR%7flBR>A^W%%<)AJXn;^Y)Tn(3JZ}QVUMnR4Lkv{1qvg zazva9ILx3meJZG1K%}5|=D85rG?<f+2Vj|A*pK@E4NLeR4t#3DpOo}PWk2NwS2SJph>sb>nA`Rf4*wG+R8p&tKg&r;}Y zF{L5m-ukP_LjKvYr!XQ}muyTA6ImW{5V%QnVkTfIndG-5L! zv#*D2oBdO=S~ZOHQw|gPQWtT+o^E{gO$FzqZQ##^n(#_~%EFpsf8gP|lkmPxo}V{2 zk$m)DP8vnts$4`eNfF==RQj|9G6LBZ+`n*3c8Q?;;h2+AUFF(lW zZ*g$N%m!Yox)48SeY*IIA!+zL7S0yw36k*XFhMHMYlXIwAK#m#aN!(o63?0~|Gc4V z#(b{p)FD{`|`I`0yZz-LZg$LBt==dB`#@{(`ApfP*|f8g+A zTJ`BPgub1MZu5V_a?fJ0(D=h$T_yU3EqJKBJDjf+9QcKOXYxlUPvuoh9+Ah2av+pX z#_3bk`HfKu{6r%G`c18Xz=CVytj8N2Pwx{mx&V`IPvIZS?B~~RdjboJwu9^Jxn#Ed zJ8o5Bb2U$L7aYdO-sPG}fNADqxKbpvFI+7gCU52j>tKr+8BT%GV49@S&;eh{7nm8n@ zvS^Y$^^jG79fmi^AzDW6>g>RR3E_Oa3-J0O+Wfx5AGwOIqqz507_1-U$p5e?<$EOl zysl9Oe>UVafBWM;&{OMxkp4^Hzu>#VFh@!b-9CpZb7sJzZ|8~Qiz9G3p+i{fDN7nR zJr`7LtoZD4k&tXLj^vG;L3~5}z+vGJkbNF(^Q+$z?q%Z^srTDNvcXW1{0;EE%eTqWG@)gy(&)NlzU+MW-LP!ET#Js2<;pvm$?jdPguU zw~~dg-ACZX0zJMWSP^Arv~G=VFHte%JTXImRS)nX`n zLW@0e84cTOL}#>@JnXn2zVlrUiSJfz9oK|cM?c5)YeJ!2+|MSR z^TFS*`@mQId!+K%Jn-&&ij3s9)BA(VY`iY}F~gf>EI`(zYSy>_CbOaujqXO`=Hs7G zIUUhza1^TNy=23ak!h9PX9EpN*kF#GG%SwjB^ma_aZ)*Oz- zc=aUou^a|(9m|A(ag%ARmJypbwFmFqFlMXF^w8fmgcauJF#UqjX!Jsd8dYtTUL32- z8m3KV8T$^ihAvBC(v^$cQ>*9n)R_jfYwN{HCf;PGLnH2U*B}+$4jAUV0@Y)KgzoiY zh~jlQ<`Wgo^oH-l2Ok%*4a=KEe(51}syBhUmtJ_le>Ki<3g(P!W;5#%JMoc*3RcMz z<~?E*i(KqdHGWNW)%0mQs|<9suvGNXYo@BQlESM@;qfBoR|3u$HJ^R3Ta4;oA~85dpo$jDa7$Vw-twqLpWkco>q$-aeb_J-l~#vZ z6)VY#Kd({7$g9dQ)~~9)eoEE6e)+g_fC^hDyN>k~Ux3r~^Dzz?Q(aU^T)U8a`ylgj-Gk4(B z{hr{ojB$deqcy6e=ioG>893y<9JA_4V=@&JSxfUQrr6emGv?05is4<`(C0&ITA%F_3wAUzN6ZdXh_D^w@vR=0zC^NO_Y17>)*E_jos@p9+0FH@F(8+w-4UiFhf^E3eL~^EL=sOn z2+XVB^I2=>DJ9^S~>iA@_!VX*WR_Pv9+ z_vB;X_+m_c@C}PE&BjpcN$iEeXVK%LKoiszffgoEA)s2!kSu`DU-RhAE$L)x)FqTJ ze=Io9%A~Dd4^eOJ)%1x|5y?tShN6W!7`OU3Sgoxi&An6c@7yA2HcVsBbIjO=LE>Ci zZZdP8^91kKnqu{{5_-=mNPISUp!|w_VlufLY@-_J)TE1Cj+Fv8XvGkSu*spTrY*rc z%RK0WocW}#Y!adgyhL%#NEIDf)^0@a=s1dh8RrDj&m^lrms?rGOM zy70$0V(H$OdW5{e*NZr|Zr>e@j&i{tGaAXbekttqmcxCJ2r|GtM)S&?B0Tp4w@;Lgs)EL7QpR=y0kdw9XIru_;AZ^*Joq=A{9DgM@Q=5+Xnry0Ncu{ObNtaoH=c{0s)~=@cd&C7 zOK{!!A=u-@h}FC-D2?X9YyBzes&Iu=^`8c3bH0&=ltS{WK7yXLJBZhhFGD_dKHDx6 z%r@|?2{4=_6HGKV@Z>DDnel+_^<_8l?`k1{q@r9Kus8wBiH-%o8 zh}=@?Z46#oO_N*K(<5O+!1r=Ga}T|QWO**>@v9+@>znYPT?kWnnM-8U(m>`)3%%f> zK_hBxF(!1q;C z)`6lgW*u{vv1E!5j$!`nD7;`jfY#OH zzUMxwmcBs!`y#JDFox;HC$jen3G_>rk@WbBr)2TV0GPBdojM;|Csg?KWeeN1;iI7( zGaY5ZzBr?(&$bbC7uD`HZ{S}#OU`1MH zQZSpL%%}Mo@+X#B!VXY_^bjR9uGub;4v+`MWA)swiXxoxr~n;JC&R?@YEaF4OlQv@ z$m_pcK;r+drcRCn;q)jYkggfWg)faEp3P^7cJe}EddQAWuDOXbOcYsw>R@(m%vCyg z@enNhY7F<%RmJb#L7bg4;nmUEz~6F0kLFjXe4+w6E+*KNy{HjvT4zpLpz;#y5;&v%C{>?@|BJY?FVyckjpQK ziNJ_;(_@gi0*JU3QokH;re2L&nk&8dV>@X9aXj*PcTV|Tm#?LPByLTsjEV{e2GwbN;q~myg%Xk*} z+lM)hPq5xzSwZGV)ac#?HlX%>GxSaz3_V8gNt^0iFx#aKnKm-8wb~tQ-;DyT;M+KR z+A%bLSdUe=`oii_5*X@#65_8b;*R7=G-5_NXL*zB!%htpbAFHXL zxp6l8@pl#u@p(&z_0>nGn+TU@C#TF2hxsHzvn9s|sbHx2E>hNW& z9*n686s8VZ4Gkj)6LnYdvu`~{BJ+XY8g9u?Khq1Y9?xKP(={kOSVwoL86)*-f zgKVq{zryrAT>E6oZ?KZ(>r-Bkm(|N@;m1ajR1*%mt0qA3PaXcGyf;W~<#6ocXJqBG z>yq*x5=p4$a3VA4HR%Xf7CTxk;C%cHIBkDP2POD(s#>cd{?r;UBKL`DCC~r#iROu> z0-wHTFYG;d0d)R_!TVMx5`bmggQ6Fb5t2Jl+IomDjQ)=#_=Q1a)GYq+VF%uzH;mMz zSn}>S-hh*~B@WurL#vnfWxFfp@ehopJS$MK?dptxp%MN0{DyEaS6+nX^F#RNZdYE5 zB_i~Bg}aoq@!n~kA0R)2Kj^8%XDDySsYQd(A&AoNK_hV2MMJLCoA7%qflSru!%uWk zI-z6K6cck6xx|YjN1bw!ZllwOzXq8xA@13Ue6Q6#)=v-yr?+ za5Hb|b(q)wTnW=Oz3AEWMtE-i1UjSQxkG4!I~A+&u-!;n5wH%!zut$aa$o4wTPgN5 zDYq(J9Sx6tqYu8`5M~fH{)cBWuT&!X4emeV+D{~si1T?|+I|l@z3*PwZy5`J&6MFl zyC0O2MFdiENa4#s>HTlzk{wk8dB^z?yxFY*{KmQy&}Y_9T5B99=gll_TlEWsg$RhJ!N<|&8hFB*lD?iJ9VCHhPy z(3`pU8N-YwDOEkuy21L(-eB-9lMPy%%0hn4WN9g}*dCWv$>rDLwep?J_Kzv63OkNF zDmZ3B<}%+w3z^m#3B9@0NAksQ16lmeABsc6fYQA|80rD}BTLqu)+eVMI*<`c%Kc zYblzbA>Rk|#z*6vVdH7huu2L>59psQ`gq_&0{$*F!-|%}q9@>{(CMv(&!6vOPqY(R z{`(4MHt8pB-Z7bt@-Ja=S7cG)?*}^2rWun%H$f{^L+`46g51wP1n(>H>)#LIQ@dBP z-rbpabXqDI;Fn5oHKzzq6+F3fYsRs0@1xoFtr;xcG!_+x|K@^)FQi%Y6OGhcLBl@! zkjmOS*syF0OETKY249_vkNv_iRZ*4YOtli$#05}I`DsGL{$AqV5+G5`E22}rji8F# z3hC$tPRuBynDxH(VfOzWWkQlNtNJ&eZF)6y|vQCiv94J>?JJglEs&k+Ht4vJ8Zw!mmPRC7{`vDM2{Y_!OKl!*m6Z( z7V4tI*4)yj=lyr!+=kzHX;puepBaa<#ryBne#Lb15{`SX)Q^kIE+ZyJQ;2uZXnMG? zT5y-dq2}ixv=2LAvsZp2^%2(KB-dk@Imw2`qmPZ#yAqgoO?0MxUH}hV9#Aj{MYj=$ z(AnRVJuOROY1EP3rO`|w{wV87Ucx+t0c_yV!R$PIM2+{s=sQP(F%4JIZ_p${z1yX+ zVQ$mS_1FSFl6F?Xac4BRW)X1?Dvz6gBTb+v)yTZnc#7ZQBmc zvwzX8Z%2~8$|fv+mD0gHp9VHy1<%!l9eSH?Ke6Pob zzMhOB)4GU#vOf3d+<2BD?g!mBUB(QvO*~ZR!}w48Z5k5nVT5+c{~1l>=dF0$<#n4> zzWzox&U%ZvwzHU$hxiU|3ga=a7ys_iq_b2~aQ=ZabbTcuIV(NUx+fgV_vg^4pqq5a z%X!#0Vi5Dan}McvE;N2vF1-|2Pn|LWuK!6Ve`Dw2+kS(&dW(bbYH=$`oqv&JP8kSS z|JuMIcSA6_=MKw~Z;>Z1=5Q{dkUPKC73ZF?!Ci0DFh0Hx=YI5I;f9mg9ooO@MvH9K z{Bg+6g}d=C=#5Eh_M*Yj0NArs3Ozm#KrY;gJx-pCi^e~rWdSQO>6aNAeD)!Sv@+0p z(0F1)a)ei2BY}3A;K)H+!E@kY>Kk!FaOXYA!UX}Gyh9eWHLAlA&5>+au^f95OX-Uf zc^LSs9R2UqkebX!^iOORM)|mtSz*aA>`*2A=xX-t4Q_2uB!lYw8anBx60g>bTG z9zWu}IUQ;B2%GI&=vEQ(ZZX`9{EC&(K5pT}`kN+%wl)cc>&`*$npm!R-#L2y>_Iy7 zcmz53>OXobA%^CTkVS*jrNV>$t@O6{VcKj}04HDX#N&D8Xu5eex|uwLm@Hk`?z|QQ zD-Ynjk(p?|C9h-?&r3dUVra@wXf`36Cd6V1nr+*fGM6o~_Ckoj{JTLis11 zGNXug*H+Q%V=s`7OS{mrK%U<3&LFkDo?@?g5~N#Nq2AeZ*w?5K4!xa#jenx?O5|!b zed@F-mnU4Eh29xU=L)q_7li78vG|>FGj~H&MX3y+) z!kNwzqR~Gd{b#0<{317yD^=%n6t1Ioa|XHR{0d)vzD4DI3utW6HXP{Fh1VK-*f2TS zDpw^x_9J>2TbVeNe-K$G$klt%KT)HJ1yl-42VRz#7T44FFQVAKHR@I8?Uz+$j~z#o zzPF(IrT=h)Z-4eA|1XX=7{Q#$5Y{}{h0PFM9QRKMblxpbZ0frnPvI0iYE;B}trD4X zW+`m1AHt4?GWK}yB33)w5>tz=)1r48tWYHwLz4!PhuK=}Ol1@EyzEl-V4njk9}|j> z;0LZJ&Oq48_waG?D+nJumXNCf)cM6(9B^|i>-blQD-~DZl(6SC+~g77U$>pzxxSHJ z|F#0ORt<*j|NMnfp<9LM4h{OYu0MJ-9_K8{D!fek;LzKp^!LICoT1MwT%Br-y-$sC z-0wq>Hs%7j+n2)FISg`t4};`TJMjz}%`R~^DV z3wufH&*KSxIL4Z6GeB970$`i(aA+{o-+bxssbaq`R%M_xAW7~cW_$W@i5kL z4Ethb#O_*Y!NP!7q_^lD@d&PiiT{lF#}VGVhPy3B6pd!_-9wqz%*CAd)K1CP22S$T zsEpc6l)3fJ?x=cI9=G0Jk3$wem6~tZ4L6mS!;;N%u)|!D75K;F<|IFGDYL}*%86{C zCSrDwxVIdcMJFH4=T@#kvTg7w;_CDO*3EC_D^KYv6V5Jy1!T&A*U8A|&s%#PjJ7MD~?Ddb$nZca1;7 z*XP=Rxp4+@p0J#3rn-=Ft^=QV?7_IF8FcW(>EuF}xGy@KKyOV77g?fu!sN^!+~F&2 zLgi>z+PO=C*Xqfz393CnVa_FNY&t-9ny$h<0l)dwlu}+pDT)7hbR<9cQ4k-u(1Bl5 z^@7-X`$2Y>AC%wyk30-*L#<20X;9cQGNMif*Z7SRR*jlSdo*-}0pfgC?LaJfe4&9< z?0*jjSAHe^r!|pJ!-_=z^GWEe*WmwL5%b59{g~4%f%eV1D7+`;IP!8N?kO?G?^Bv! z&5A29kL@A{A95rvJcXWnNQI|&+l6xDdt}(|%b1XVTp}|thi>TO4j=WZxUA%Pg4}~f zVbY5~V3)j$x02TJvO{lyVfS}va98A=$!fOfD$lCoMBZQ24Op~%KflA|B(DQ?^i`lO z%=PJj$xvU$Dxlj`Kt-`iqD>I{dB}g4_sR#GHWBl*%SR!L_$uWz!zaw z`aPU~{we-aF#A z4WIWoa{XNFb1Ijbb&SV!YX!ROdn_D1rvg#8_rmLCEs(u-2A*&|OtdGo)ANRfT$V={ zwZ8Bd_r2UfbIm?;cXgHs3;%6G1-ZHCd?_5NuSNoHK18-WcY~Q}v+(OwODqmdXM<3YavKH4xr?HoB%bDNF zy)3h(jQM0{Gp)6n*zGTe=QdbVIcGEI?GW<=FRCPyCy2~Kox#|g@ro`ka-=Ie8U4EZ z5)S-3Pbk?frD;WTP_49+9R0BfO#6y6RX;<{I&`&Uc-MYfW6{I?3)%&v!#3a)GK!sV z_F_uLr&&eyT_^tz^zj(9#!JmXlQOelhG>=yQYr<}EFZE<;8|sDz z;{#QB@Gp*pUDFG2ahxf;bio+)lqqb?(%`S2-9$zgY++MEDLFA-zYJeY=+|Ys2GvwZ%J+^+z@vubJ0=jCms8JCldV>YJ1m-7|m@&euW-v?)8BCGXlYX zS*$SI_+Dkwy{))oa3WdlI*nX;XoJ;_x482jC(&r_U9>S>MMw9ONJ>N8IOTo*l4Hv4 z+!-ZTqO{Hr%6@+nioR)(1#&v{_C0NaS##h;SrkXdYeS33Ty`=umaM<1K##p?=lsge zXq%!54j;?Wz8c|BHF0sJL7Wq0zOSYxMvC~zW0No_b`@vZqk(x@sxUm>l(~v`{FUEK zSj}KvjI&CIpTi{(aJQDG#rq2PolgoSO0nc_d@YSqn2j1IE`bc`ekeZg z2H&43KzNWkHfSZ|;jn%*p|hTjUm|kAgEDD%_a4FqZiX!-*Fa(KX1t@a1G7yc(dg)D za^}4b#QxYVeX?aI&3|}Es!}b7_dc9};#V(01?y&ixMU6>Y(6)aS zE}FcOUzs+5uW~#H%XL)v)2BZQue5cD`<{c)wDm9?(YOoqhW??^dZ$pQUJds)IpKf* zj*!RejoFX>rtI6Z-x#l7M#o<*r8<8eV~uPZ%36Mgslzz%7r9F@heSu7_iFIndPBIc z@KvhrW{dT~azuWkF?h=U;f!@A6K|KpsJuZ7ohrCWy6`s~Xn77MniKh@bLX(@!bqke zdmnABx09`lzk-M2Fy3?RcUUy^HT2Yu=3k%bLzb;KhLhWrI;SHyGXCU845WQ+~KlWJ`7QL#mW6kk^EQcMZR?G2AzuyWcuwiRx{cYzjF^oLp1l%6T<;_y^# z9LKR>t#0zoHx>Q*7^#^=9elWQ*Fs-4Gi zpWHx*@OUk#$%^g=Hz&p{Z_}qA^g#E`H1Q4)0jclt7j_Pu2wYft|EfxS(yrG)DZv`b~ChQO6rvXrsj%J|1Povz5fN z!5ir+Jq{ZBfp9P-jUGxJ$#xt$gQ?Rs@Z+B+FgDq>D(vT;D#gvgRjk(w@{VTGStTQA zAPZyz!#?6j0~r?lw2h^ux3koOB6ek7D<}Ej0Hc&QL55+A^j%;$xAMCx#J$x=>5m`6 zn8`+H89oZO>{-gD%sS4V#oYk!JNE?hR9%|&Z9Ed~4oCDYUPS^}rXjI)=&<_-QQDqz0qw`fP zZ(JRVKK}>TG>>Jy;i*Jeth^ZvTNCZ1a}0l(%H;DEmRQ0sdQ_G!w&JlWg0 zA;XXw+Z`(o`IH^|5 zAZEa}hKabKw1!UHV9b9HamMpR5Y&1vkE|(7D zx0}cC@H;~c_0{3LW4A)a<_GZNM<~DX&UxOdE0;I$y+%C_q+-eQ0eIbU8K>J#Sl)}Cz zqWijJDeK!B$_!eT2|M?^BvjHtz6_~Dd&B4WS*HNbTn?rlw|Y38xuVn4^CXEI6aZ)M zc#yxMS84i&J1~6N8Bki$MAZ$+3wO_f=;L`LJ!&|V z-MAvMl3n>foA&c3$}aNkV-vruJ_GJN4d+db``i9WUM4*r0pQHGa_r7zRKjhXAM_>F$hZgilwapixvAYfy z-M&EUXb(L#N}rYI=`n401_Q+tb@Ub^{-C(O$mFc?#g`$_M%CH)d+sd$!&H{462l@_ zPi1*m^02;bDqB#l%$)x|#A8O|_&0a0`G0}p`>9*J_h1_zUfjwr?EHm6EkA`eZC%K8 zzK?$@=D__$I{dJi?!4-Va{iWCDZjkmdN|dcLw7d^i2LwEWcS=TL{J}yWhdnsw{s8R zCRtvk{yfZkmjfHx>#5dFdH&}LQ$FhT6^M-ePVzcSK~CcZ^vzd7PyHvbP&J;){JV>% z((}RLpcnq#yN*qclRz)Y#$H>F#h%dQKdk`CLk~}!(K?AgSoc~I^!EYRaPlyn_f8en z4S>&acnaH3w!Qpz)klj&|dJIoLno55xNtJ&a^G?)#p4s zkQeV8Hs;jK)Q}n*b#S)EHN>b^ti9~FrUSP21BFAkiQB6Kkh^vlNpHGN6qH8~Yczq84!qU|CwD6fehDT~-&s>J8wYvsB43deS(@# zZP?Z27VKE!agYQ`=qI1%xzUz-C&rPu z*4x4+S4}d^XFtg>iDdEO&gkBwb9pzo!t_6#=D!BF4mY;;fIdY_!o$+@^koCcbKtxJ^*TCyia? z`2z`bZZjvtju%slEy`H``2=ceY!W*;w>d}kcwxqq-ATE|gp} zATb|hh2wexvCps;{?41s-1!tX?Pd**PYyuefG^me^kwtLu41QsquH{-HjFuybFZGA z5Nw?k;kI)ryEZDwP6*A{_lP1Eq?c5&yd@&o%=#4-n=2!}*@;fkOHss2D2 z>h?;HSa=xFx>2J^t(TiHwX#xLR;NvEPc4R9&l5;6@Gc=B# zgavn^NC6ip=G%^dOVvKff{~VjLFavPKKBd`+-`^X#2dG~)MkcPG+C#mJ~J6CI#>;K zS(TRpR;rh>ihFe|bpA6ICT1iqOOj}Zy#zpHYZbj-gteNd@r0q+)x7A28UIev%iRz0 zT#E*SiXm*dP8TV%Uq)WdI0AD_&J%OJ!`wN`V7k40DP2396LUyDMfv)~?2Uhht3)gkNlag&GxaUg#7FLAsAuwqi3sJUZHZN{s zm-2Xa(tId;`)>&9N{aE>^FnlAdQdvEYXeE!D*>Z7s#InJAUE1p^kF@qL86QH$VU%$ zRiy;4-!Ft?%}kQ9No+Arv_$p$`!GPRn~Q#I0|N)-OXC$?i2qUXl=2`5R&B2%s>9Rp z_%{irdX-?}AZPGR*W-J{SygeSC3bwx<|uCkL$mEMtY0lv$~D2dQ%&T`j0iGrrwsAl z;!i)l?-rajM>7wPy_j9_$mXBiPP)W99>Ud!;ig&raQBJ5B=uJsIUpUyea8W0QtBDH z!u2M~d^IPB=k(AsN6u2WBzWRE3` zK?|>$*ym_J%-*U7LHjeYz_c8*OHLxpio@X*|4`~Op4+`u7Ry$Zqu;!;6=hcZ=$`LnLGE#QaeW2A_|%#@XGe1*%hTT!mNklwuH zfD0_Pa;x_mkf9#KaBtEpeDJIy{rmw?+X zWT^F%Vw%-aO+OrPqns+YFGqaiam zL47M-_A^n4%;}M+8rxwbrs4q~Gb~h_h0BVc;s&iTU@1Ny&Od6V`}t%}JHLQ#af!!6 zpQAw5S_eX#EXl>{JQ}+#i>Tc$LY1)DL_Rr|%4OZ91y4ku?c;CwLY#vg`DcY8&haF` zw}L)dB>K6%|44-OQ>pFn2lSiIBuv;5PV)ZrWj``@fbSiBP`sW-2T2vUtxIl6Wy3Dg ze$9Q@y&9N)c=m~9f773Vz2 zlT)IbCSw7+7!=FKU<4Egs7Mbyl;@_cSA*J{SKzw4DjIJTe|L{9oAq3cZFiYWJq?xl zHKVoo{LEyb=hj@5J@#x>Kc+Kn_JRZ3zjB)CCuDq7xkhGx|sTDMx4zcINR ze$F;zMd2mrP;wR*-JMR@J5PS>(Sdw?N*?~WUWa80U(t2SKJZWeB2jqOO9stJCqZ9s z)6^~{)SbhK)>jic{jwHhS=Nyc_DRsBb{K3&#$a)iH(fA363X+x3O@@65#^!)`p7LC zZlzp+N1d0*>_MgE>&!mrC1%}!>^KOilefd*gAI5m{4|L@ITW>=E>fR}t2oKw4E|V} zOG=H7L4x^YIN3jjdW46<{3Eh3;%O1B?o5JNOX?tROa?Jo5C?0-jM0$9#UKRg!6PvT z_T|$gn0}h0TE17|)bL5XoSqD(&rQVFL*-cgRbPl#jfJVzlVL&C_N$qj#mGGx;U8c^eLJ&4In&{!8@0QCLg0AR9;0dzeL!m!6Rr&j@C2 zd#zYo)CGJaGO8C>^=B82%F$bKG$wqPlI*TmXg*$*JQ=tZj;g&S0aGr}MVuAe7%&nB z$vP8=aRxVs{HKJs>j^>w%DZ!zQb2%Rs-I^DWBgi#X!k^+@k$#Pp3bB!G9C<+b@+`^Wws0c(0Mlh;osA1&~K$HtMFgJ zTw0asu4yVv>E?LcS9X_*C7b zE(^qc-SAQ&^8-azBhDP&?4!XqMsvfySxYoL%0@NXwn$T^^`ZLpH@U>L{w#V=0Gr|v zMwc6Hq1jhQz{Xk=X_Q-?^MznHFmf-hei_z0YJTu{@+YW(NJWm|S~2taW{y0YmF3TB z2b;32(s)VwJ&ray{-Jj9G18JQD|Atonbho{3_T<=N<53z@wcfWeiVYL z2NpF;G@0lHpXMY?IU>hhI^hkBPp5k?B#7_BUP;XHHINYfk)(Ec;+FX$Z#ydu-p|t@ zJ622Rn0GZ)=drjOxprL`2HVNEU)J#BbuG=@Kp)?-yH52I(R!Dk_!l*nY#0dA_eeQGC{cUOaxrblt-gIi#1-2mwQeH}vz$D?eW zG23qRTy(W$L&?c+SmLaLC%h+v`lni|Ueb@9ZU0U5-!)TRA9-A*znVYaWW?7sej#hN z6_9I-6!_StBH$+N5FU8AqsfRkdL+)G+Qs`Dxl(?GTiHDhtzwM%c@?i=qFS`JnH3y;Rb7SO} z=~4CQD(>x_O`;R}Fpfxi41OcE`S2sHuU?z>7~apip5jfL|JthaIu)=Hb>%;zh}_D z*HqBA$bxfvA^gf=ZelNPCux$M0*_6z>BsrI$Qic5_{=72y zWcNXRQX>s)J54R_?t*sJ=VangQ`nMR1y;SgVPxQ6DA0|k*RJmuv@0FyZiPlL*f9g} zxdQOL(L`H3+uR=Kjg{#^G=6g+yfZGN&RgQpo zQskh|aXDV0Xc6zHF`W;wPv*bL74hkB`t$viD&gg}Hb`At1FJ;Fe8lpNP^Z(EtRD~y zzixy>F4qCNPptXF=H|Ro);7{TCl$^r?}q4LS(hh!y7*>;68=_FIcdOyAS)E{?+$lB z`nq2*gAC@^#LnR5Y~p#Fr-}UOea&#eWH>+3V-KIwc@5O9E&2I_b$P$)W|;g~ozLqV z!55T$6FcbX#D4o%;=N8^m|{E-?4J5V`?zixx2F`69Bp})bGceJX%AcoQQ>!apN1SE zhBQ1LiPK>u)L}Y+D#oq>wpEE>r#iqA{TJuvo;}Mf)8%|mClDc1@WepqxqHJ zDtH^ZkndNd%I^tpBhStoi2K6zSpM;=i@WuA7d7FD%f|vWAY#_(fS)$JEbfJwGjw>2 z+0HvZdc=piJmG8CE4%C*wTg_w>Cml_0FC|)G-G%X4CxMl+Iv0Z>y{(%B%=rnTs>e- zjWMJTx(nAu*Pq=cZT{GoFnD+(nl`D*a@6)MH|mum$qA{FzRP}0i)XEdsKW0sCw2`e z+&={yL+W9!j0LYc;SM}GCwl9T+w<`U!6nSu)@A9xRW5VwP2oYvL`Yt*%G=+#0>2_x z5KHwmQ0u%3i}IcTPEUfPAA|YrvWfgNLw8=7Qw_VUZ23Rz2i((&B=UuLn4F^RN>3zeREtu&64N>|LY^-`XQV#_21a=(vmMhson+qcVQV`}usjOS_Rp_N>vuLVzF8_Pr6n-ZELu4+*k?Aez z(Acd3Cab1#LuXf$^IEfEe^@F^G*#mjZzO^clt7;=mC(dpHIf~Va=>?48JXc($tm2J zipeGJLiAe)#F9pG*=`3t^yEF=(lVdeKlVexIS{?pnF-_5EaB!SH(Ib^JTGOYVEJPZ zzRtcxc0JX{l)eXPZp&cw?Wv)GZmaRjv^>!GR6^1g&WEwXy?Dpr%KWT31V*Qv5`upm z5LWK55`4v4OGcU+er(=Ld!PNHZ*0cX8k-gP=lO9ic7u8Eg^GMb%{kZd0&X<5hoZCeV6=D?jnYVY*6yv2=7;}rA2xh$?^Dy zG<2ksIGZ*m^uRhYaL9G)x_K(EuQH6^Ho~1PU%r|zO*Dg%UsmB8r|TjgVmDlkBh=}{ zP{3|IxYM-@XYO1Ag;V0V+X?BE`o`gm<%(Q*Zl0Jik^}bX4K2NBidNtE((j)QU|FaF zp~j)mHaUfuepQEi`vkb{)5AHve206C#b=D}6!LsJPv0B5Q)}IN+OZ7ikbesF^qi&K zqm4`HR{gC-Gc!5wsr=XVTkb z_-u{oP#Mx@N5I_Z${!5s70`Ze#OGIo4t~4`og^ zV0e0e-1cM-ChproeI{kVqhK|*Z2VgCxh)9Frz)Y<6Y;Dx`vi`9DklxNqDUWxgwl0; zHCa`c8(FX*TN0_7&Gz++XBzhY?C9V1>|;(EQ(iZK9r~__VR;d#J}ZE!?GstU*F|n) z&sDV4+{Y4Kq}Ucc32ZBlv7ox+EXM67uAe&sth#rC*7p2yU$WCC`$rl5V-P%Mb?r#CW`@T<2H?h_ekN|p`Oc-{`K zBd?iVa~x1JCSz#L-kF!VU7KEzviE`PXT<;u{(r*xW4AU$$tve0o!LO%HRhq0UH;Ub|LYpTS(rnLt>A#xj}roD8z z%1TDRXx0oo^O4PIXl9a%kz&u0;@2(1*wgo}Qi~tn_*LdBJ|EnLS^a(qH=IXcWkeMv zc1Cz_Nhv*W(ie5s^`XQ1?na|fWu~=aHw~SClpKuuiE(;E$&bVsYVx2L9NQ~l*n$>L ze)lLu62;9~djJET8?&854qq_mG;MwTP{aOEo& zTpo!rYV|LLsy#L3MopY#WjXlbw_I4}%SYJoi_ETcu zZ-=vo`U{T~;^E(RZRtscImp}0#CNT$DJv+YO1lfH+q3VWzqu?s!~d#|R1?|rb~mtS zMl(HnbUev=EaC9PYC3O9DCT{*4vX#G;hl#Zs8#8pXQ}8(A^MPDa29$#81iZ3_wcUI zy!bGXrg7Tl-7r8Q0ozsB*^^EYkjp(5>mx1P14Lvh^SxZ0MitjL@sMyc8wj{7 zwO zd9CD>^BElKdXL6P>#I+-h*^D$P;#wOh5!9w8h>r*OIX=1GH!+_!^y(|gw*5+-6^u9 zrDlXwFf79*8~@PoYpQ(LHFtO^_Q=;JX3)^HO_G@-O}Rvy-?UlfIF9IR3+r_C;EReZ zUe8p-cP~@8#o$dIE5=~@izKN^)F)2kkUcg9FdVz3fM`Z%(4KAO9WMFJQ=D31ZbB{5WkNku=7(fxza34BPQ8leq@n25BH-A{T!s*N=nEK+sUxi zxg88Xbyln1(4dy>o{;D%_NrTM(A(8*(k{JD@M;CPT>FjMEa-(&zYTDGobbp@Cl_fNzov7-X9)5;iRnz2$Xt_hW!44v$ zYKYTQ8@LPgj)JxOK}h=(10G$*5O5$>^a1zMBGVZVl{gRjDmT$PD$ejuEd(qIbs@!| z5A4qJhL~gv{^cBJK0C#V&l%Vs9L>w=uhau{_3lQ_`RY#Y_pV%StwK4?x^P?A6}2A* zbGGc!7&W%>a68d?KMPZ5ufoU|>-nk(gf(tWgnM;S82CwvCFS_B({)LL>wizV^9PTL zcTW{~N&1mUokcKLtA%#n%cWnZ#*&4u59og$N6<2LCk}5N$gj(@;C+1dLY>8(V2lv*k9 zZNp39?jA$1k~88wXAS3EHK&v0vLt*^cNyz`Q=GrVn0&M!hCeq6wCSiZ^ijSBGJUFG zrpWEPyQ+nhM_AV^Va>LqIIV0Hsx9j!{_Ff< zsH-X6eZ`nxVfLC^;}?y;bQIY}T~GFR7h{tXfXczv_tP-(LXL^lH9`#bps`q8VWu3UE%^DZX`9)d91ah?gH*7oG zmy56Il_*~G6y6=zqiwZ$+?Z{DxJ|#*p||EHcRTzb3rQ2R?Qg2VBu>IEF5=jVx8hw{ z+|6XP>=OnJoPqkk&H0HBLrBfWAo{zv7=BCa!L2_b-?I|HGj%Un?DvW`sPqV1-q^C{ zERkU$&Ri0%PC(Dg`7l~A=9lV+lW+f0(Bqy2R;v%=kF2oa&uytd$9L8!#C;?O*1O`# zxgY7&|9*4ltp?z6&uOsh*9(!yB(fv)eh_7g1gd49Kn)rTh;5}eys#N99jIBtosZ9< zZpJV1`@nBFGt>(EYab%M`wmL#b&6^6kfXHFx`c#1G^g{Ai9IRnF|0`=fbIPf%$IpB z<6k)g^skQKAGv!$R{t70@!|)Xa6X9+@yjPN-v)9X6Z;95o?WY6o1n)kR(wU%C-=yg z8$;12>LiVNK7{lu{YYfBLipHOUC?T##lB3uha3LQV4i2E!MlDd=*Pa5l7~lQ(Ej#y ze7^Au4*%XPqG7dtG1Z*+j$41r zi7KCzC7}!dz_{B*R6EXDh%2i=Bdtz&k+%{y4EMp@tp8Bqb0U1HsiO6}W1+-x8y=px zh8$ODArFU-#aEj?pilc9T>8M8Mbsy-F#)D*lQ<{n)z?3<^)eO~vixvAz{#kH{k3n?~Z-&Sl_xO_iOw{1lIznoNdn z;m~Zs9dLa1irPAfj8%CVGVpBr|Ngm?6w_oi2{8yDYrS?~q*T z+((xGQGjEjuW{5gc^qC6fJAL1-+!!U|I zCQ@8%pCtG`0NQ?kF3s~?fb;BIvFmjvc3-svb+QwlP8rCbTOpoJ%FO7Dr|sMYlENJ` z`Yc?)ZhGmu5u0f6QSvf)J6@`YBiq9YP?%?eCV|C}75GYcW^4wFrtg!E{i({A%Bk@c zhvXzby$G45ae(e!E8!O^8O`KJl{Q+fao zh(AAJtpju1HJBMbet_k*JS^@|p&$OWh}V?8*g8Oq*IkoDdI!7GS2kyGtm=5|ziSr8 zjf{e;U+lrb`VmgPvIC!WoW+g>soeeX=SfYZweU+V4y_&+fmiN9c>FyC)`)I_Whwb| z!)ODvPC7!(7fqyVQp6pXo*f?3e@!KBJyhfIeN@m_V-p;bF~5JKbJV6G^rH6r>Z7W= zh2Mu%$+-i~T=?@Y`ncqc@VR~p8S8P6;O4_%$jk6$)4x&o#T|n0VSCmcq0Bx+jc4Yk z_DT%M7;;`8C_I$~tHQsM)WZp+V*VPO9@vi!e=vf*+clOAITp;)p6q9#6Ys-vy)7{7 z*-(7`!va5Ty@LPdymFqYwv!4Q-RRicLxiZT{X&RE5ofo(gnsV79!LQpQRE9_P^k9{SmT>Fl2rLxO(`x%=*~6jY-Ll~cR~k2r z?)hU&+~4S638>=kPbcui)XDIsuLbinEy6u}4cO^F)@*sD0rtpCFgQMfiIf9qDwXGD zN8O_lHx}U8bJIo9>t^y>KAA`jrDSNxXIee?E*YZ#oSx~jVKNdQ@-^TJ=QL86x_ZT7 z*tJ9)l=qYV)}KhSjJ4>+t8>Nq(IR5CSc{fUFeh97*^}etYw19{7Wx=}W0PbLo4Bcy z&B!{+`s5sCf5Sep)#3JRS>!v`V(^J6Wl3p>cNu84CX!G`BuiIaBZ~{-;qSXa__I-s z?HfIeS$qzY%&@yHbY_^KV`3!sw2HaG3yL@^{kcRq(N4P5a-^B-ROvmr3Ze4wQPMD_ zo7?$tGda0)4N(#^nRf&JQ9&{h5;U)nU*g;N)SkB|Ni7%1wA1>Q&Cdf5=Lz|-+8Pc?yJ-h75ES%)nijC*cATCo_lYEvKrkR4Obt(67 zMFX)zXg z)h$Bck`1I#d|f_`ir{Y63%+}J7}QspbuXE~a!kdZ*B3L^#^i-7))&!g%T3NvwnKPr zYXPc5I>~XH4#?~&CZh*tN{;L4W8;q^_I8mTYYKWUI_6rrJys&aWZ5lJ6d)DOELOzb zLFTydfE^j=n9S)0uEsW%@eo$GhwR){N30SPu}|M{T&%+5Y12Pk|23xAesc+|J8~J! ziuY2J*^R;vhg`V5T9ba+KaL%i9Zrw!KLawB!$8Gj1|HA7Lr#uuMu&bn==N<7{rB5S zdhzQKR<*-|f97Yxo-F-{w_<-%b*U5X>T<)XQcYp#G%p-rGYK}e1!M574p;sAM!%=)YZA^sDY5(bsL@?*2ramaGeTcX>4T zQ{+8%1(MP0iy->>3Np1$7c09)LY##c8xu2x9*|C?0qWzJvxNZ*w|gcXu2P7j$9$(g zRr@eS^9XzLLaD}m{zs-iqK7-Ve-nxC*9T_VRg%Ml(qZgIN8weII+MNR1C>S!_<3C^ zcSSFT%(9)Vb&a=oBo>j)itrC#W+zckeTF`puKe|7l6}8sQMZ<-y7+2>*r%tuN zz=Kbzg}Q`4^8O#x^rqqBg46W*bxYFg@RpwKH&v?VX;IzwRfi02`c-YvKZa~;nL=*4 zN#J?FLC$UOL2}vU4ow%8g6WIakjtMhk)tnNM6kOOZ+PbmwPZ1aOt7~DvjvVyVIuY~<`e8_&RYGX6UPh^88{6VE@&7}70D5h5P8S@*=F~r^h z&t03$;(mX_xu=py=Y&9HS-@BaHV* zMwPZ!<{&$ionJJL1@!l3OHU4F&pKpS^Wj0v;pKHYGFkL79P_4AiJGJVSYJ2MVQL1QeR3<8Bz}Ll2HT-Vr5D~$ zd?~d)ah57}SHkMRa5Qzx;Kr1c(uM1G!u}d-xbrL+94~$lR@CcZpAoAFKedu-slTGJ z{xbNm$p!q*CEx}~z<~RUQNDc*R*CyrURIZRe-}YL1N0z{|3udqwBqlozu2%Rq^3`u zOHH+(NzF;i)i~^H0D3Ll582?wu8eSDvlX+@O=}<1dFsV7%}rUn_i~hO`G)ttQ`|5+ z0T(a51DWP-ymi+P)cZS(9jXpz_qHpt{7rpWK1SvGyKBv^?R*vrGfmy5WES(kBpaVt!k`+&x?Y@*=d&*okz zV9QK5GV3P}tTxbu1!s7&T46Bzu6`07xCrci6$+j6eTB0+kBDjHO)|yp4BR(dO|lb> zn9tm~IBkv*i_r`~!+LEtduj}`^;1V)eiO>O^k<*i@6rC(i?Muz3tX`~0e|#yfj(cqsk zyPxs{Q*Ir`rLT9P%YPo^q2qei|6c(k?Yr2;;0{z4?~g239;(%JV3|t>rdeCipRw!M zfGdO9J~I<`WcyS6)qEdEMn_`C6KfWHUYoTHI*hryoLP8&C=M29y=gJ^_@%{x-P)0h z_kO>^=WfZU9;pJ??@c33m)k(z%O2*azD9#(F}R?*OQ;vKLq9)1l8jkAjhitd5uU4F zrRKCk^k2Gg)4y0_^?_LYJp+NilfnLLE(Djog*O4i z>6xcjMBl>_vV4>d`z3mj7t_z&-=Ck*xvK)#&KXPQKlG)G^lYV(-&NT&E**uyDX4eQ zTlB6^U@2RgME2wtYW_NfUdomuJ35WAW_=sg`S_i)-XO!@=@PLTm*_-M$*Ji_25+)?G&kny`kdUHXAz>|WAa?^|e9{wX2Oqmy_~*@IrTLAWt( zEY+zpB|plBk-K>l@Kf+B@!n)DvT-_{7sTm8tcefFy0HT-zfZ-kv?gk0Q-)R7I24w> zrm4Lx^y2n^@M}T=L}{GDZtXtoko_n&a_?yv(H2hTXotgae=~G_Iusv-rK02*rH|br z1wA+U>kVV_cGH8+F0KYEOfUHwBx%oJb zj_;1g-M`L| z2M2!>-=DJhbNF^U#Nf}Szrv>|JFt(N0ym6g zX}Mzsx1#Ai8EqaeSpIjL>u-i!ebGa@p!yBXcFmw>A4L~cR3Vn+%*RWQ#T~|ov()vj z$OiVJ!j%a{Tuoptlt!A+-a`?R`aWZ6*|EK7)EIzs60Xv_QaQ1AvmXYES?4*2i{O@P zC>$EANq)Mnz{92iv>(Y3wpHd(gZ)Qn>7>EZsF(e)LC2on4YY#%V=slrO&29oR*Zpe z@s7Oc_9=XRC>Ra|A()+bKzhHMV480QY+KnxCL}qNhwsd=Vq-fyt~M53oaKf0a<%wT zt%r_H-i02IFJbwjdlI9BMSSTt@%}3Iqlb_Ud{p#)ek4ofos=%KvG4CP%hrReXVM?y zdixahxc`kF{;>ek{x!p?@tGiB;Kxr^|3{XHL%V(_jQFDIIrQd-G3H-W$ws>~|l`HG;AJ zQ8=uZ6hQJecRcz0BQ8(&6w+)o@V#CDOidn)$y=*&rEnKkp6LR8xfk$thzkGHco-ih zyOej*S`JShNFid=CDJD%mUehZaAHC`!mIxHOA?O=tRg<| ztKt5&KwkAk0xEu41SU_mfcu2!w9#Ax>+Je4i&zPoiR^;Q$}e%#w%f#QQg(H%ha&L~ z*$ifJN#v%x4tyNB1Qf!@@y(m{Vb#GEEI4)mS2^=E7~ZtSXDRzp+^FGyoBPv*8xho% z8;>vhjAfTs{f7^xI>Y4O_aJEDC>X5%AFc212X~4)X-7p7P5Ey+uHKYRUpUA^=-qR4 ze6HtC>pWjp-);fC|BT7qd;HcZ_g z#}97m1dH@A;eo+`VuKQ)r&k8Q1}9VRM^DMK(kCK^xsBd_Wk`Dqeh`60vz&}r)-ciq zm*0OTJ+VR^lEwX&VT^3eD!po!e4~VInoqOOaKTG7!t`_-KuhFe{BDKCR zvUv9A;0ZC$Ij}{{h0K!0vGotAlKA`3|} zBa(BuBr@@{dd2tD9;R-SflaPftbSDh3v)ll#=Pr~^Vwt?cw{OWyJa2oPOZg_n;l5* z!`*bMQxJ9NQ$}m*f0G;2O=;pjTfA;qfw*BI?9-YD62!*&H&Bj$KkE*ae93QVz*l+**;+;Ov}uKMYXM9KThl~8WoXK?$?AfORI=( zIU}Q{J;n0Lw=hOSi|LwAV&5}GR#)OKxa<;yF9v^+Y_QXXgS(3{{cIJDPKZP6vP>~& zUdsKukp>#iv2 z7n;ZvzdhKuv>SsWbFnC`oZQrrCl7K%1y%olX#TM;p4-n0xwTih4Z;pw{b?e-SsqWf zENT+uFPY<#9sca#iowhyGaip}E%>{c$A-$gv{20ubs7XL3_pn4PP>F7EwZR_cm&t| zp_x@!-(WxF&1%9*ZevKTA=;Wr@XNNrP<^Kqd|R~P{xf@)^e>pn{5>GD*QY?=Pjy88 z&mc1QTPl9MXF*RF>ESzk!MXe_gtDF2sm|mbq<>`z4gPSF1|%)ODbr@)%&)aLD`OAW z*XxKhrCW#M9CKmFdSfo_b2e39t%`fb`{6YaglDwSRIutu$LHh!;2!-G`0?E`)GDqL z{YwvVo^>-j78%L1em@X8u1U%MrPW;Kp7YfA>oOesZw|W8x`&guhztrne{y{FaB=Qq zfXVZpNPn-^!X~#nga>!jIHQJFUYgNQ4IgOd;wyB}kG?q1bRXSrNF{x~>5=YgeV?SuyWEUu7M}G(SX5zs=CxHV3<= zDzPneI`dYJV6#0_(EPXs?LQ+KlcU>lmX{HgO&$mtQPD88-IZ=|FD79#OiA)@1JbcD z6DF-+1j}ys&}$17=tqNK_^(a;yIZDV`_zAQgo77|AxC}`h_e_+6~QS zd!V822&OfDpt+YR9eeC97t*B&SLXS`pXyPdY9j;S{TaHaYVaf5Pw#JV?_UTQ?9DUIWEn#S`TC&l~Ef~!#DlZ=H;D{<@-H|UJX zlms+o(|-pPP{G!a#Q)R6^L~A4RAoBVQ#(%D@7coaz=qi0TH1+TZLJjzZ0g z%pSVF<_TWYa;NrM&UnQ)6$?jX;hJ7Ih`MyE7UjVc|HEksB}k>eUrJRHdlw zXGV@IPZe6cr*p0m9K5% zPEds5um6zhz-Qn!;4tm@;U{KCzH-`*4T6bJ6!%wl9eF)H6DubgqLID=JU&?=&SOO< zd_x>e^gIb;w2PtV#s|7gbsV>Nk?6K|8_t(!R-?xFm)ygPtKia^!K`hA3i2}#Q-!eI zY|UL0*7>{$JA%znFiC-9HdW;2%suFAbcJdZ-Xf)IBUr~AC6==53rcA&r1g(*W&`Gd ze`qcFH=5CbLkIDGn~b2$U@iC2H;J}Jyd^X2_Q0VSe|VR0jHW1$<_n7n|0B2nd*fqq zS4A1R`8l(*%1Z2HLMDyBRD^BTCiwlOGj4&ua3iM!oX^X{nFQe1CPnc7yxpi;fz_+bQuwhO=&T0Q2peu**kxp)W`KX7aQrtu2 z+&V?dcT6H>v46<7ryn6^<7oEYyZ|rl>7|eE$FbZpF%KAWj6blQ!;GA6TI;s}V(*s= zt4+&=6_W}iZ+HKI71PFxUhO!3aeW@m%tJC^bstm7EcGR3>ji$%!jsKYQC%m>c4azLE8uLx7IxzYHelWz-?gd_okp_=v-W~eg-Pe z_rmjw_Q29rS!Cajd>~u8(M8bzjO$AR_9 zA40>!VlLW=uGdM~gK z&U_t={x+AKf5o418apyh@~Zt5Y%p~pIdkpFihYULJS3l9vcF93mK~rLF)xMn2h8bA zVJC^K$fBderqlfOQ!uY(Cx+{fz*7efk>ZL5{AM|umN;wkA0%S`Mns;ys&W#0bu&1# zXOGc3br9rF8%-l(_RuWFOm?%?kaa|Mv8g|Ala*(V3Ql*8=zmVfaU#E*?$1t!BUWl` zY*{v5l^w}sJ|{Vwok*hf3rFHxZ&m);@5yj>-4fcfw}%vYO@QcmYEa^-g30$j(xMxM zg5}O`sx;^$y~GqCPqo(h`R(D60<;PN4a30W^NJCCQ6N3&iu}~n0j*)F1oWEeGg`Q+6SJt?&9>Dw&1|M z-@)){4ruk5;pKTs;-*_0Wprft0b94DhieY83^T*|%9CmIaCyjV(xS8{4OcpR;{wWe zQ*Kck9r7WAOnxio3*_=!LIr4r%Af6!FAI)@bwIahZTh|E&D$Fb~+84*G-}h=JvQW zPW0Qw@52RsuLv$xo-{2@mpytV2*!%)B;laAV;}2CN@C){^}I4aZ+9*}j8bNE>>Fuv zi#yxDZzfwXwGQ-+jhMMcJUf(A3yap+!0zTlsAj2!w+nU<{njT$p|u(s@+!${GdtL< zxDv+Q7a6xM6JY&$7k0b)G+y=|h&f$j@XoXM7=HObB&y3W`ieV#? zDGu()Z^V^XD(Qq2$+d;fuON`D6X=OwyClf`V>+AX+j!5$j6QwHDM$bbb;#QkmeVZ8rlKd7Jgrtd6w z(B-3su;&T$(SGFzp(gwtyy~bSNd@xkZD1{qOfSKJdw+0C{sAI?e>H8-v!XlBM8dDO zIq*61I~uHz*y0w(}M(soYCz=Lqyg zpNT?zfS7?yABkOAiRc+%$MVeCN|ziq1=f9vpY-gu=WpJ#-yBfk{=Tb>f- zbLM1S`dD)6jXHPx{0FI9%K@P}qAyF>70Q;%O4&zoDPH`(w1+fRr``qDjV#(cx#SQao^n7F_L2KLSaS6Ts(1o65R5=N<)2gq3~fU#MU_S4F+G}U2mJPrSJ=F zHK1JnC@GEJ_m6he9iXvur;29+(XIJ)0ogk)j-JTq$IdLCk5Tsw=$!?YlB{wyVc7l* zXkC#Fjqi7INBu3h)=h6Yslrum;rdu^(eEf!a|mL_K6*6qy&ObzH^4%}k6c4~4#;;` zz~LELY*6|WTyk0b>H|Q9IYaD^3MdP@gTl#_ep-0Gfm9xs8 zcy`9s3*)N{F(NU76=hR)?tvG1HLwg`9lgwr8lO-1OHK7ue`CY$ydIuY+D zf5w45YAjGi9{BQh(H*o8S6nc{;18?VYknDv3(3Z*mlQB=##NdTdkzh!Y2)P;#xS_x zj>Islm}z;Ku?dBSys_?a=2^3w{hs#)cPCcD-Cw2nuu`5)g(JALXcOJ|dInnl=L?>H z?!l;;4?*YEr)qL}D5g(LVv?na)LZ8~8s9mJf;i9l&I*a7a0D}YeH~Kshx3xv3?4tr zL4Nuh(CE{j|B|QxU4QQieUHb{eI^l@8mGvropXtn;s+tk{VT3)(1ca}r}C$U&*n@1 z>jiXsE2PR@qUYzeadxT*^Oa`e_g;?8_$Z6`fLFA9uP|&KB+~=CFKS zj-5YQFLvy%L2cA(oMoxQw%zW}4*RB9g{}~9+-PYrT8J4(V{U4g_WFn0Fvw)`i zHwgZvija{#A3xsDz^>XNZf(UM680#Z?#$ayg*7WUwecs&ANK@$tLqS*b4ZmnO&)|x zgO8F%VGyS>^aP#w!V>-WC1Ht09`W502s0{ggWd02bibei%O014dgO9c8DYt~CVe5V z4xYsB8v$rl_m#6)m4GirPaeK>WW7(Lm}yTI^Ld)jTpm_HSIJ)**18(2#MfG^oK8}! z2a^RVUAXRuCbPm&zWp)O<2=HC`>U_`~Un_mjXo z*^0N%A4H2MnZu_BFMPYl4t)NNA#djamHS*t#^eqpPlA15=4X3aSyh3Sm+gp-k1;q# zXL9@x8O*cYM1{F_7#vn2>~CqtR@+V*lsT2md{RS-n#1WL(>%=XYKAUFbw0FXEJ*5? z(FxW_dOnwf*P06$6PkyiL5VDOYzp{%DFE<10A|&q+owXHF@nhc(p2O_z4ZAo?Ra{j z?0*cMcOaEt7{?>YmX(#!Kqy4H?|F`rwiG`t35C*Ll2Rn260)MGP)en#GVXhxSCJ6W z(4I7mhLrZg@BRDw>$>;8=Q+>!`}qt8G$!yPY#udycpmmHZN`0fBhcqfe_Ww{9veQX z3kzD)iT4t3A{FCo!HJf1QdulLHo}ayM}H-ky2Z?2a|~$62arU=ryzGXoJ#tp!L%(e z!Qb#P%nO!fd&K^2Yt=UR+VYkqah;NXdl&Hq?gDOgjK#%^P1)aN&5|)sJjv45d`2E` zrsf`t>3J1@YU8?#ojLpl(#2V`e6I@YFFKaBauYylvj)a`G?2z29^`RRU#i`84O%mA z^OHZ6f@kn=7&cUiDz{wVf82-?XP4(-G3gI)WIsV#vaBCW)ZQ(+?QU`grV}MihVNOU zRWZDXd5YGXDoIjv7fzg-3pY3RAwlaD@$2(F*jRgyALVt1?R=(0F6X46 z%(+d*AGd)!P2;iDaX#;F^pvQ~JPf7g6R3XhJ}O_GM}5uB=`s6VG)rY2D7XxDdfGO~ zDY{wHY3XN;Qu%orRIxRW-qUR0Wy@!iW4hvd)ODR?4D*EwbvLr;zZ@_VT_N){)ycG@ zAJEVCCI+676EqrB>Fh~K^ilsU^tJANxMjMOt;lsE@#(+V%-e{^Z&b1KjU~`3??P>h z>RI0JQ0bNn3S>cB5AP=&A--1g(7(9|`ze%yX@i0jKw&f&J;9#!p>!5C)V6@{3dyn620acFOJ`1#9JK>@IJ3Hg_i+x~D^z zY*7;uydHDgRU={4*lse+Lx9@&b8My(1EYcr80>P7wRsODv9c7CZmh+$-VP8}-RGQL zXR*GDigbeM2XZuH1~xCXphiP?vL6>YICi=j(jJ@BJdmTkqXgz!#yfg4xJ7uk!G}W`3TyXSVekfM&Q(+%X#WO5ljl+bJM;Z1U~gPw8kx= z>XXD-+gaob{(45Z_9B>JI-d52zDz3h#pdA@WhRK6{I4doe} zm9S?XF<|uX2poOs#|?HH$L)Bt0i^lzklr_z==Avw=70J^&c6fO;%pJ{lgO$pgMBN{EHVOU~irK&Y3*l$@f9%G}{lxP}A9^;=j+!bv zL(c0`QdTdE1|CnK;>I7&-_?3$DRws<7W8Fv;&ebm6t_ z>WKKkn}PYxrueDp6%5LXAfn_3zWvybPja%*{lR(cw>%7|HB|D!+6mwtFbS34W^wUH zZDFC|BDfGEg3&*pVLdq;_<~~_*@S%+(yY~?i0w`JSpmkGufR`kzqC^O{(5- zfR)~hC0Fc^@o%z6Kul8;ccI^XUT3q7^nm8?lKYxkj(=~MgKh3%xZlCZ*N;b_pREVn z)wl#<-g97;$bBzc@|Cr%OJV-ww%`MWLuAuuOAMX5hgnSnuuGW@KD&Kjl#4g13~2%% z^9`V%J{J9sU2zU^j+EfUZeTKmI2YYxGi#NPdJ42O`@%#o21t#L3 z@xzit3`&|0aQZq6{Tz#dY1{ZuFi$e!-4t%C*IwS${<(CCic}IPcBp-y$?&UFj+1n= zA2{#nXnt~GBX2^haZ1=}_JWUtt$lBRgGDS+clW`tcOyud{(K>6NCocRuosotTNXCl z6CKV1`;fN~Ro4!}Oi3mx|4U$|yOo6WNruAoI4LS*3`CPt)`Cxj2g~f1Cmv6n@Z;CV zg`-N95Kh=tI@+?AD@)J~Yv zV@ZILykLHy7Z)t=#FaY-fiU(yjGEOSnz{nuUbBL5wnI@!_P1%L8Mdr9gxUNfSUkTCu6d7VcDGbZb6W$L_nnK}pjJy5RJDopyhzJGOlIg#~e4zn&<$Rucjb9?k1HD(S&h5 zHn?+6DeJjgfiJhVvrVeHkdu|bZq5t9%P+FId7UA|&~d9UwSOn6?em_Wh{0^& z@V*$Hat_TG9mA}npYYwj`FJ5S84?ywhXF;Oxvmf10;{RPt20utkFOftazSKx%1Oa& zw}((-JXdJE^a>5$>?eI5c;JQsmmzule(0=srrW-YOa%1>ylbEweYS5ZU5J_ZD&1T- z)oX*F&nl7piszuko3Oap@51MItA*V3KN$786WV1psPE3PH2W%#(-%F!nZ1evv9cCI zq9zFjRi}lInS0pngBMBY0XLGGpe5#KM5k4j0ll+kBUM*iPd&yx0tZ@1oF?z$o2d%S zUde%>%?CI(H<1RvvBs5a`=f!JEACStO$??e!@cdPyo<+oIJ03lwX1QZ*9P5$e#%Dt zqk>>Q@tGHlt&~u;xqE1~+eA7i>M`B)FoM3;-vJ{+tm(+wAK)2sNZ6C2BRF0S5b_Uw zCc{;M9gb50}23{$kH*yv-wUrS}enkKz6spj~>t*nK?ihMdyA`6U_d{Uo80gFjWCyPF zp@zFA)EnQRSr+XpzwbwUqvS zcN5L#uMwV0jD)w@$8q1@AU6NvK*9O`RQ%dDfIPl$K+jE{NjoNu#k;?z(!_yugF7tt@k3U0Q%Q!s#xsn^F7(y3@4Jh3kw45s& zn!#%Z>A_K}3Q}yO53|fGNN4|(Ag5OZ3q}#4+1y3A899-K`1!(&)sv`8x;5P}s6WUr zt0Rr`L}$(FF*I^QDlFDlV)|ZlrSCQYt(Ucf@NRB|)A!0mG}0pj52{j=ee#_ay^h{GzMu?ezBy z@}I)KQ%B&?7=JL^NwpUYJvXOcX3b5MNfmnOV)o%FB{kP8s;7Nz?$9S*~5vwZ0w>( zeiij<7OIW zi95kfbM;}Y^G+DGK7^jp=pysojbUZ;HNrQ`Qz@R|*3*?x{Av&c{qZ8tT2Dgo5+?>) z60*Kdm-|s&4Az7$J##r|YRB%%BUX@Qsxutn|eo!HQQ zf%VwM3XO9Nh483U_9NDgtQPrWPdn!`!=)~0cesH?WyNBd)@F2@>w?wWg300m^5k`I zG|}_3fmhCP5L2_8m7OoefPKx}zV|kwH_e}scdF=L8OK^5>(U!q)|4AtLPpNE7n7Mn2Sy{q+ORbDJ0=7pA~5C(g%wVZ30gN7jR8> z1fxUkIC1Siod1ys@bD&K$3DYHlbi5&d@j0OUPX+iA3-j6Coa%RA|raMaai3dykvd> zB{5a(;bAvcaI=t9*k9mJEn31`eW_&J-=}=Ti>KrS>XXHXb0zCX%%!W%b?8(6GK4I) z1(%r{xk`TvIA<||xIZW$_YKdIAisga%|3ECWLXMktEQ4CC$Dk*13mIFYv zb%|8pF5*632xR&Wk&v#-lj9|`#0;tguGSXd_urTCTE_>L6TeRqsF_Pr!)_4gsiQDy zwmnXMe~l?F(uDoR7szqBC8YG_H@5L<4~u=HhvnDW;GDJ(yth2TKKVw1d%sv{QdgVENx?@}yLe=viBXiB~)6XSI$*bi2|teLu+SlS)3Vme4;J9O;Oyi6SA+pEL-i zgHO+eQV9*zlpuFH)$TsbGrlcs+A!oE4M>nLkQb-F_nvS*M;|MIM6x14*sd>z~<%~ z{Hwp$B)>I|@!{^p#QxU^*1V76T^l_-3Y7lXSWhojyrlXa;lw7=Te{skRI;@16I?Hz zX>1)&!@=6o^v~LjRQb*ddf@C7+|(x?FB+Gk=8}obdq_4P6lTbR?j?{gdoyVKSqp~- znBjSgZ_NL%HY9B64}*#-r2o=gOB3f85s$GMWZcZxY?8=6XcGGwDRwXTTHM=3ineW+l*9b2D$dC7&+{BedxDWcqETJ{{?2j5+aR#P0h5;mfJp ze8_J*=wDF*mV3Ow`AspucV_^v;QSLVJ*tF}{bz#Na82UYF_0fKc?zGKyGYFc{N+?j zdPwLJ@x9%WL#!&!kO`ALx$u`c{P9^8{M~7r$lRopF!JDWsMS8g4WbiBwt5q(^Gqf) zY+Au@{1WP~br#fLD^myOFHkc2DqLMv26ZUkO6w^kY!ELj6qIh&0H|K#onOUxcX)b?AV0{J#g~x%MiY822 z;w^T(FTqgfqfi_m?m+X$OMd?gficG9x|Ka1C;;dVnfajBciiLny%t&~~iUT9X&Q+FKl*}A@XyC{9+yaa=xFli zQ8Y74KZBPSYN4`<3`Uv!z^~ou_`0Qv8Te-7wN<7#R&y8We@_D2&+A~u%oxmdmzBEi zv@HFVeA;nF$Pjkfc?TD3zKn#b{*|noxeOM1+=gFgrW0T9MpBmgp1EggLl`E)>pKcK zKC^%iN-N`3tLwqn&4En#-IqK2t*kWj&JRxa;aL7bQVd@cQO_5j9!k=!KvKC>oz*Fd z9afc>#C4G!w~rby~2zv+Evd|%j8MOk`tw&{^4Z#8f8+u zw+Lkt$}nNwW$vtHA+ggsOm@2{k|`ZxXG?Cb=w6$}h5eEfnUNyvIkaMtr1j4(IT`D7tmx#-y#-tJF=BCj>~0jScYBzk0Hy zZVE3*2=hx@h9iszVCh0nNoDeRFu!^SvJ&?bx9%!7Nl%Fr8n^J*hq{(b`KntQ^CM0g zP_4y$hQx9cvdJF)#yL4j*S0odz`Z6qEO<+2l~hDw4dOK%M#){%6f)=-9s> zoBvw|C+hb==uFMhYd`$CiQ|8;;qpK5$1Z)$`6=eS7AXqL?=+Ie$aXTyyMp0(W%fwo z3qQ;+k@%(#B}xlqxh^#=VWvYk(R3TkGVVF!r^AB;*9Bj( zxulJ4j+c@LFAn1GwP%I*GM+-gYynRP+kwMrKaf3i7u{!^=2X)D!jF6v=nklXo!!Ie z=L9P{N_i9NB^+m)b0@Q#RXOabqYBus7k73m(%9kzXO`|V4lhqT%%(g(z*Z{zknfi| zSaoL@xix<`=s0DS#2GqZsYwm~Xe)wVm1~^WtYSWX#|L)XD+1H@P9Py85Ino5(TztX zbbqlSbuHaP*G|!-tw$PhNCk&A4s~$g#V$PO=!-l3ci`-6x7eEKC&a7rgJfRe3W=J% z7P^{?nd72D+@;?aKSbN2rE@Jx`&g3nt{reXF&={GNjyBWg>4ae(7$EXS>J_a#KpCS zojiDtf2QrqO&w=|5B|uqrQ1!2&4WX@MdXxeKbnW{z7^r$5r;{l?=)6@+l0)||4g*4 zw@CD(Ccs5CEAm`-IY#Q(LfERW@I9;o){4dBuEwqvamXQR^pp{9!lVi;HF6*M?(}?I^M;(;PzUb4B-{ z46fys;aKlwSiH6ZMjif&r!pltbfyzb8k7Thu`)>Hb4c~dPSR)nUHCFL5tcfM8Pvjh z9N83wA0Au*Z-)dDa&U@}S#}iXj`br(ODL~f5{rAvt=k*bvu;Qj{3=NE*;Gcc8ryZ{{(ob zSj!B4_2Kn~pGF;vVA6Ifj~o@}y@58Da6*d}Dts257P-4|&g&i${m6i5nNP#A8)=N! z6{9A$RYKYA3xcivdG>4BbC6M&p~X`wNy{rYk-wr$w$yh?AaMtH%`691sBptAh z6tQ{x4KUuLFD%gA4Y&8c<4cZyO?8I7pGu4>yAG{Epn)1Zca31qpuL@f`x?z+4Uf>7sBFmrEkw}qC>vUTi->vPz zvYq?6_cDp>K!_ZAdL74L-vlvhHb59~OPgI&xJ$YZJ4l=VM3eEQYsuvi{rQDl4$SXV zhU|WhY@W^ue0V?=Pd-hI2<#Vp41KnWLXQ%00Z2BYT74{TrNcnIvA#=ma5 zOR@*vA$l^qv73uRrMoYs(W%Mko%6PI&m{>q537bqhnJ zX?SkE8CuTYMaaw(lF(xyZ2g?b>)S??6+K(h`Qybi;zWZ;^_Lpb8}F_!jo1!x|2hVz^W92waJ`@h7(vTyfLB`r`GJ$1kE zW3U05&VE5wJg*=ca*N5C{Sg=upC)~ztOJpMm%`m)+MJxo%)jgG09T4-SolYG$W+*Y z`YVRxxZFVyc~*eO{0MluEuXADl1c*1cM+*&CUXj1B6%@kG5hm7hWmVIDb$Y0gN0f9 z+1M-TFw%4#%)XSuYpHpo>;AXWh~@Pdq<0?QZ~npZ+vcOm`xqnYf0|7`BBvKrg*W4DO6=>vnupF+hs*ZF71 z)p!S^B(hpJ725t2`A2V?A=rO2I~F44)@`;Rt|>+s-cMdKZsG%Ws<|)UrB{g(H6`IJ z9e@)rxsf8Ji}>Fl7E&)^_d%d%I&7mM(8&ii1D!Y};Kv z`caf%?h?g=)A5nUWqcW>0h34yH*&le+jHwOi@tr5 z%swyj?6PuTp5{X=+{Dk>AxrMN#wIVkT7vSK4`7rhkm8f42dZW9d+OFk)4ciU%x+jo$kPn z?2{LsRCe$URz`S_Jm!XH_T$y&bKDi#XvcZp;h?Pkn@qMp$;Nri;zROpG5%x{DV*kp zBMa?t;BFiK=D=wXI&=s)F0X{0m5NwzkxCk>{K0A74t~z#g;-y90asmHj_xCyw zP(HPg4`23_6kgp*>JO;U|8zdG_Nsm)ru{1mzmi1O%<*9lI8&iW%)h-~nTnkow7|ey zfyCK_v%IXa@K&pq*txIgR&TWch008>YHtvv?apKC&S$a>>!0E7A9Lv3Y`mQZ z(46dry(_wS#o7?w&1n%nSgF&P1WD=8~{9w^&#`+72zU z3yJk}TNeG+46}m=qJ_(0G;&Wzs|5~%-cc`M)72e#Hee1$Ot?x4zEFYq7zxXZicz;{ z593`bnfcYlB;xOX?ECv&yv4!M?2^?;>?fzl|LT;%0@2+nllltlPRu9kJSITXnUAd2 zN&(?g1T!;eCX%z`h{~|jBJ4~Qv-C}aRK9}XkU4(1%c0&EADBjSahc~($kDFvi@j(VRiQf!H{%zU8 zwB$Xc?mE|5wFW1+d5bQYRo|I(f+@>YkKp=$PU1o@9}+HjL<`5y-@>1F_mSt54w2sa zeQeoQTYRN-o5h>eB3XNaX&u=^6zp{HmEJz#GJjS$n!HBH*%!=dMjpjk(*UQpUBiPh z7IzOg65UHz9Qyt5hr|}B=S_8?1ZMOC_J+MD5gBk zK!rz!*ft~!QyyN%Eb}~O6#IciE(WxzUC6slA-qF^GEoj)4?pkMp;o4nkQDw|Iyid+ zI2q1Gy_NsiOHmcITKsKoy7da(YWoWM3e|YRK;)&>Yva*#jZ%G`WjJl6lquL*(YDYI zAg4#t3;r4G^6J|d;5kL8Kc*|3v^KSg9)Z^}WUqRV&E|nWq0O13Z z*lyX;r0kLm4iWb_R^#p10pg2}{?@{&1OEw8g(1T0@qNlRJggAsJdtxaxtZ&e`jVT} zdKnAoBjUJoC#L>=irP;eV%F6T^!mI;kkhAvt#kmR(>G)2hInCR&N9>*l7q5Cmf@eN zec+4IBX-sB2L3%SZqH0&S^SbKqGQEah&h=igeD&r>WpXOm$bL|DuxJ~;yc+Rg$Py> zQG$gx<4UIMlIT%E5>miK|=L6nX;-L$Fj65*U>OL72nia2B2u2eI$$ry}FWN_griC)ggCAb8G>N9R9oLQ&}+L0Ln!%wm`Vzv1^Vc$+>CVmE9S zd(!tI^wkq+H!;9*t6WLIikp(a6MDk$>~$EwNLE-_@}4Xiwin{igQA-g&fZcg_J36b z@y3B&O>5`#l2ys_%F)7^!7Bv2@?#R83|(<=vq;E(dO!%UlVu^lHe=7BT-k!_Io^G_ixIf#C13}N*+o--=4zww?o+J2*z!G zVnoh)R+4~W>shy98EblXoc;Gr4T`6Ke)I z;D5ya+%VYwNCu)O-Xt04v!rJZM6$o@19)!QQxeR#v#wf1RrZO$?(>z*9-j>N$#77} zRzt~T(HC^{9&=ke2CozyWf1uj+Y7XWgKNEoh!L;3sLhi}<>S39!|t&}g7sq*YK<{- zvx%TI(Md2^t0EA;P|$g*Lymd$XJ@>MNh&uEe00XasDCl+!yo*uwklH|sCaoKPoyFYcA* zmg~XHyE0^qkrJSSgz@DX;Pbc~5D}174L}C)_tCh(f?;41+ z=DCo&EgkIIo=4KjcB2_q?}X(6UqlAWaiUb3!}8^FnXY{~``Rs!(n2rRm}P_EA+c;j ztR^bv_akQlj`Cgyr^BFsTfyz2D{D)cQEJhliw2vXNCI4Dv9-Y}Wc}S?xM{P%!WB5& z=4nnAKRU_B4*Jf99-Yr_li{fP$Pc5G=3vLRKKNH}k|ge33Yo8($qkh{;n+%bGG*rw zu=l4pwcLiyRyXGO1Ow=;@8V`(9Ltoyb+YM0hN8ec$<1x2pwiqI#!Xns2CY2|tIu2~ z0slq7>ZlV?8*zaRY(9nOYQ|#44juY8brHR^1!=ruF|koKqRV5<=-lE~a!6q{*~Yc7 zPn(t4NX1^5#_gtgTr93zsv_iu$O&2LNmTRAIT}0u0IgY+z{Y18pzGsu)|!2REZv|1 zmBq$z;MQEIHZuUfZ}~tMAgXyz0+W?lU_9p>*zSmd?&-21O`XHKGL8{^ahaX{cperS zjiNP)XXwgZk6Cw!Eo#zU=Kp*o8jOF+q~h)%(IK zBOiPKHD+`G_g9TJ9q&VxR;RFC0~H`hp$mH*V*H zMfiOTbGM*oH%3$SkIB$&y#coPQpi^APifoZbK^1}ajTqJ26ypQXTGzUI}U>0ls4GUe+LI`JNkOP=tr4rLU+uaKqIbBpGz+PncF6ibpdJLKQrk~uH|K@RfU$2r*>M^1^JC;LO>3sb5u#$G{Go(j{ zz9mQBCDEFrM<`#CNu_H2pzge6@U@~D8= z&Lhxg=1MkE+|LwQ+l}`98$t>id~j^J0!1+cbn;j{2-}Xqs0A@lxn2(T$7fESGGnCRDV~ff;(TG$Pa*pB&zSc9}}xQa=)o z$TIGNTMm>T@?et1+T;h#f^4EFG#OUlrA=$F^xSAZ?cQ8=E`L5Z2IJw%qr+Uv$akDh zaV|N~l184m@o+}#CSIR0lO62;fSu_a&kC%5a-$q_;CPRZBtNkXJNNiu|3V+wTeJ%| zg}%lACaE&*ZPUShL=fB(eXMDXp9wc*Bde1$1G{N{P~>V!{i-ss&Mt{>Fh~N&`vqj3 zTNd7Z?=7r%or`OZr^1?OW4h5#o-NBZVfJ#T$;3xm*!%DqS!g|+bQPS!0^?-fEA}0v z>@3F{FXDvTzbpmaU9<3;=?rYCJxPWg9gZ(gXEEPjvDodV$Zo#dh9kvu(yslhF(N|D zVU@+<#>p$#L5ou69lS>DZ+3yxtMbx)uJhT?Xl3DAS0Vf2ew*00Dq_P0Ph1f{4L`_x zpjt`_zqv%@8W5drY52uj9qeZ~3&+w-g>S#r zg{l@!yuSDxHXhr{+}BLTM~kbu=fB+{T1bT-%7x^$G z+&ZR1H>hkB9A-Sgdqt}3%(k~I2BxqS%cG)a<1Z^c8OnMRPw{m(GjPk`$EY~%1Tw!0 z^!QkTcVrKse&iiK{XrcUTA~JTYTQb?4=iA7uH>`GVht!sm7w1B7IwndQIL#J5H4%4 z7uqtjp}_tVx$$o(4r-gq_CCtx^m?UGDLV+9_V|&335C3F>I24l)N$3g3i2js9y8jV z4^RE?L$zBId7yfT*_F@Xq~k<~{QE?{DdQq*(@K>7ZI2?-_b0ige@f7*uZn$+^%1uF z7~ommn|Rn&TgWat#`?Y(!+eJvVVQb6@Tb2&`Yu0>D;#3bt1AyDPl-dWAr8&+VzA$} zzQU+rSt!IDe4uGAJdwyjdT|mlS`p1`CZB@7=a<3_(TO<3e;#|G8pcKp_{P51h(6aR zfuM9ngN84yK#lF+@Id}I%v?7T4r&+?qi^*P&^uJ>az>wdCB-x4cnh|Bvp+xe=@9AC z-JHm5KLelU_Js>~-LQ6gIr@Hfhliep%xr-vndb*=w@x=XVex>ZJRgHiZQf+I|9-Gt zdhe>m4Im`d1m(JalkW@NMO9 zueydE5B_04k7LB}Sss}Z*8zShf$)`%2W^qZ)#}Cb$$8h=!Sc=Ik5Lwv9&!vui~N=h zLy_YUY(Tg9?}UA`@3HQ=+?aIX8g#6Y!_mW|(dPLY?96))27jN3?k086IHy$NIW3=@ zlph4gBs|t7(FNNAj(O;!CmE>0I)0(?SVbiwf5uJ7)ZeBb z{TeEsn{|zaWU4O=`LK$8_^w47tB*mr$1BKuAa=6G*yDLmWw>!n9y~plk*GyZWY9e~ zj2tl-wz5MIu78pVMT42ToG;GbTukaU4d{(wsi6MyDny7qoyTsyyqnSyZsos)P?!?Q z>F-*~Iy^fiq$&^c)+*58AWf<%dbp~e+S1#vazOWY2}FD;qw_cHqR9_-L(%9cwCjG3 zAI8WN{gwIbNVXezHyVKPT~E5ZXdxZMz2*FFHgbi#dN@5g0H#b|1coNr96xFc?5%kM zM>neB!y$v1PH;Ujdm0OI+(p!_yuuWctVv7Y703verJWu9Y1-=;u|Ji@UhcCa{!tf+ z@edO;4&H;&ZIQ%TIUT;bFwnm=6NXKD#g;Dp1C>wDkSX6ePza4+L*u0U825i%{mZow z9oYhd6F-69oc%azS{j#qJr$mJpNDg0kKlP$Gm-0?fW*=R)ORUR9sUI|2pf(YUYufv z@kl%-Hc3+l_v6Oed*LGaWTO7t3U?K}Vdqy5hRWO|Qo8S=VPXv7WapUD>FZ`U)T zJz+1qo!!ea?HY-+yo%%8^C9(E9Ep5>6CC`aDc5?8R=fCe7gWn&%gN~W?k9&a#U>f%B&WiaPbZ+T{vhf4TtebjFMxHHd2qVI zmyGt^!^TYiOboNKiK$uu7`A0YQuAVN@xo~Kr}qK+s>R~972)`H_!WM2m>C|99K&5R zRiYW6=5Q08#ohMWXi$72-dQh{rRxf$EXm1^T!?LjrE&w{r2j1Fo7YMz;$K1Y&8u)b zAp`QF)wsf7yMH+F*)!lx!L+e^tEq+O|SMrS^j%4-sJ_C?*_mraYsIMS11200X zQv>X|wF#1$K3ue$0b@IAxhWs6lg~e7@#mrO_`LQJ|IfjVZY#I~!-HRgo8Ky$G1Zjr z%=V#dXB1uIDh>kQoT&NVEspsUV%dcJM#6vJAmr~kk52>svUi#XNVRh-XJqd&+USQq zYBZh&`^XP`UZDXFxby>uUK=mC&E11DpBW2=5d#Iip(T89$uUl=DhCscM?x30a;dx{ ztb4zN=K?>nof$jvIn^e=YUhER>m3+8QNzpteakks~%y0mubR~v(AFXQ%#|R-@tknyu_gW zH*oE1Mrx{+Ansr=NhJF~csCm7F6+b({C)UdX-C%%8b)&xTG?iinezAbU0iEvk44M~ zr*`~fc{0O<30FMG&6`7n>&mhM47i9}RygDS>JGBAD;*6(-{A5F|sdtBDj)8D-Lp>Q;o=yg+tM3hyxbiThGRw55Wb) z{=@V`>o9tS1cl@8nAz=6EYbf~x^>kAJo(ZR9V<+T(^@ebd}I`SxOz+yw$TbaX2p_{ zv}vZfQ9*{}~w|4gO}R=$DW z95ap|s^#?eyt>n*t=A~I&EV?8Xt0}bmharAPmP~#r42pH>9xro^hwk^S~OHjZJ$i1 zhc8Yg`}NhZ+pCPkq(PBBZ{n!g0&DuGz>!^j@PNeMybJb0W02jlVK;Xg!X~30Xz2xw5?+&vF40c|=3BWWZ>E06PsXNS=?i5C6#Y%@k~Ah1YQmY&<(mzq~EfXrnjZ1boYFnHv5P=97j z6B0rg(hg2llPVU4mSt^L(OuRh&N`}=89 z-?B9N-S;9jc(;en)k}c|zLQB#`f2t&tpuKU>(FVh_L8QMFwm(w1S!k(Y2o%vdN%MV z-9G3&?C3Iu3vmGQ_hb0V0DY=bXh`S0+|GPcPs4xS*)Y9FAO5CV(L1yX6m|!|A^$iU z#Vw~}KOuB}Jwe4T0zL4W((BVS>Eo4Ju-7z{y!!75-cEmvRbzAUdTb+kE!fiN$acxw z^hYAk%#2D_)WC;+`Ve4z56(Z7r6(S1;^g%g=;4YQDlt-|_vCiN)znw)XV^@pt}~mB z3-9E^=WK-wDag*hnF<@!g)X9@$eTVw%BkZ z%73;H(t@k;;z4b4z1@tRe0iCLIBrCh6|Y&O^%$J;UcAHcH-`wrshGI<9aX(}nl>*C zqBFi`LHlwuI@RnrxNLYxIE$Ag$nhMUIGV*?|5G9IR_$EkZ7UiZ{fL=rL<$xAwD9q} zDX7#N4gqT`G4iMd-uki&gSY(Sk3F-(rOyZ8?HEP=qRqGxwOi?&=I?rVC24{^yE1Xr zCnwOVSj?XJw(#i(b5Z|tKjHZ&H=KOC1xt;^F3@`y!Jp0~|1o9uBxfL={d*V-zS;>v zpOy;SH)>IXig9#@^K)qK^O@S5m(qpv7gFtTJKD9r6)v7Lrm7nKokGRlx>ui9utVtw zZpp~UN-?Yd@{k2TuwfxHI6jOl3%iRyJ6EIs&L0?i_!^Ll&%y4JKvW(MrN`?n=^p7Q zO8ed78_i1?H4R}bdpRr`F2j@!KeCCpx8dS+1DrF{j+?sq84HY@!9Kk0OQ+7>i%0tG z#O|8E>_Y=jVkUQzy!xTA(rGlAtg9xx9px+RgW6wc}4io@QH=(+Mp(RBh~0{^FEBWXzvo zxYc|KN%gP?i4P-&E!q6wVI4#`{D|dfi5;;GM!3>=2dgr-p}Eh_z|zq75a#ay{rwsw z-7jyDu6N2<8YF>9s$q0RZUb|F6^jLf#qNJvE7`25iOp*?1sSI)G|}@bVBKiCc8(^^ zm?!QpM{lRx8#O*?_B@g}QOa)o&)QoDWz~i6zjT8NNOzZjps+s+6a+gEJFq}7u`w~| zRs<~Uc-;6Ja2OS@JG#^M^R8{`x72Qc)Lq$bL zN9F(ZW3CdS5<79|h>=qg{NiJiV*Q#0_y;u)Bse&zZA7q2|Np(8|K~*iGmaOua4vY*!i0i;QzmI{)VM?!gR2mcz54klpENItf4kS zW0$!|9VGP+>Mccw{WjuYPYW?Px}nH&Hx*@FOvEp1OEKq!xo96{CK|P}7DvX~${sp1 z@!^JzFq>#8-2X8bBj=ckexoeJ2tNy9y1__9<{66Q1Vf=aPES-EYbewb8VTo4rb1b1 zDo%}%y;l_$;#ioSczeh~{N8UT{4ZOG1BJF?d|x|JqGl^@mRgIZ->rmon!UKW+Dhcj zwiYF)EJR&`wMgh;BVy`oM1h5s(4A>7w&j?KobAS9*QAD`T@NEs9$+HM?>7`33mS+o z?HdWh{6?bTmWI;f&qO}6O+@WIQ{nftk=VQ6Q1pmwD11#Dh`Yt|yWI?>@1&mSu*5*z znP4c?;^p6;8HkP2H~ExV1F^rgzG$Wx2zM_-QTVcvIF{T{3_9CL__S^)CTKPi#h=W? zrRC-#xXxU3>TD(Tek{e^Y;%#~WG0qpnF-AZQ&DUqIa+oN#F-i+v7moLVZ7Z$7>62) zmq86hd-cZR+W-rZ)xccrdFdduZaIjfOYOzDvreM-Drb>B#zTzFknz|)Z_!+Mi`tK# z!t|1tSdio{)Q32Ueoftk!9fqvq^Fl?chW_;9k3O%?pllNleVH=2M3Yz-a!=l1+vAx zsr0b%NAg6_wP?kdQ(-unwPE@eA#oXhj1ro$+0ze;O$sj2Zv0NO=i3WsYV6$DSnf@$ zGp-1l1Olbq?p?kPf8Uw_^A0R%n`qkfYm{r-ws%k{!xK_aG{> z{aC8zjhm4hKF8hIa=;s(NBNmUn zS<%s+e>H4~`0UIG6Cakw`s42##H`N&oE{TI&fx$$nEEkgQV?k;gBW&PVflqn083o@4klxkKc4AL$y)1*L2oxy-q>Kc+lF1Cwv6cI zK${)z?0nONOC_FkcXh$C(1AqBO?%qJ5nBU4d|n6PIV_N}ukK7dUsn%cWydb4kC7+&{mlJyyw{+f|gR|7H6g4PntjTP!kC6Z^Yph{?@V#EwEu5mr=B zdTcez)2q1QT}A%0Di-~C&)#`uWSD(H_vA-<=)7lO+5_U}+~rEnb5id<stQwP!W0$RK?NydRDKkC30d7 z$+N2XuhU0rHQuqs;T`*OUJ@w##tdDhb85>fyc*ZAckEC0m;5DXfbHMO zI%3aXZSiEWzIc2 zPS&&4T}@Pc|Hg-vzv*&bO&n3t68AiH#Kd*lV&66$VbD@XEOyipr<-YseqNg5N1&?E z_tg?1W!mC(2TgIgLQ8DjtSiP2(-(_+=!g^l>5Dn94MoNdJrSxHirJ@i#r76bS9^z+g{TDurv2>MUdp$lKx%8c)MBfOeGYTZ#{{- z*O#!j{ZNY=%B1sySZ6dC%DWNqJesWctvOWGk%oU-v7=`s`)7r6?ccV{Ob+Ej64X2o zB0&w}^+QleiQq$COFDH5!k}{t7W9F8DxI-U4q@!pcDVP7LT_^vFO8zG`PqRRHm!;3 z-;(oZdqS@ms`e!EU_&=n>U5&!Lo{1{b))_J_5?qJtKD0&QQp%|pTc-~y$2o#`!gV` z4`uiI;~3r*t&QzD^0GCX=69sckM@i|62^zv5bW|JiOh}SgkK-5r}tpSjDeJY>d9xn z;cS@EgQW0CR5E)~9n}vXyS`kS9!1|(9hrV7ikQ8jY<6nS{qjJD7Awq_d*(V{FC5gu zsqWi_IjMv3cOAz1q-ah(>jb`${O8u0x-H>sX&S+mR$T~t)tU`!LTNrZ5F72Ll>ZE6 zUs(X&nN4Wj+n=sJZPBUg&7x(!>2bXy|9M4{K2xFBq83sMx+7(mB1klCk80bNY%UOZ z^cPHxXdy*1I`ZXiCz?#{&vmbXX#e)|)99o~r)$s*X^v}g2KO663ne6^Ag-=ZjD9O)ZW=amH z=BJ}lpU#;E$yvv4G{Z*^GB9AZYqru8+;*t4S8UTyt3y zSiowt3~rB;8p7?fse6(^N{2!`dSs#(F^3IrXYh|r2J6jc5Y}WC9lEDbJ}HfHcjY** zWN@b}lj_^4%+*che#}f(ny1rhODe67W^m?g8oj?}(5gim;cBTU`6*oaoy?5UnIsve zpyQj)?4y}Hu*s%>O&T5DQ+R5W!lVW1xFkzox~~#zN=oJImRXo}NhSSCCYvHtiEE!u zxBl6%|Il-Xdu!JwZ!t(>LPpjACCR0Ma92{ z0_|#Ae)~++lq$M;{^9SFPwY>4&EJgs?2dX){I?2;?K5$@G1^$CSGc&e=arxgV-uZ)u#(%ihT}7;HUrp7x z9~joEiO_SZ!t$$%I8^nA)S^0`zNsa6$q%+3{!H}4N}AsN#a)9cx~5c;kz0pSQG>am zs#x?+Mfks~XJ&pq(_g8GkJHq}6=OAFci<=P_iE9L{mGo|sv;yzQ#_rfDh!UPh_QCH zoLBvh82OE@FMjdy=TE-QR1xlOe>pczL;SVT6}#H&i0p={B7Vsq3OlHX?_cX^8uyI} zM=DSreBwjL-wZhUoqKD)VJ_FJbgPC~zfxUTOxF~PT{Xpx2f89{^F7L^+{L8Xb$%57 z#}CuLFGSwJ-STM9$EWaa0ZJ<7l=N8j{BQ0Fu(Izw0fLlQ|Tq_ z46bl;&OHWfx<&ZeJ0#qGg#FBij7+)4$72@=9C4I~%g@j*@&qh9!m}GEXu0_;$1YxC z;Mx=9eK^IsfCC&fJWj{S$59)2l<+UdIH7-m5C0uwe8?%bjXi;4c!Fp0^FtqgIQ>PxwUWT-l%T^C44r z-KF!idz2>JqSWdx9md=ve(b-T-u)lnf8W4t{#~v|KP0m2Jxng#BFOk2Ga6rG`_sEL zU37={`VYz2FZD4@?s8UKN2lZ(aYJwMZ^d=e*GNWE=bPNHyu|MNml-tn4#`^gC_Z+N z+imXSnfZXT?#~!h`Uumx_qiJL2&dsu^ek<|!iVkIy1g4NIvv@zIFzD$t!Q*El!{kj zl$C|^v>BZ1@5$C9O&Mk3NA{Kg87GGlJD@q|J_+{3)N{ zCp|0tvHR@FcC{uvwP?bFb~0uj=giQb?(DzdL(fDvVs6;WK5jRzOo|Cd3GnAeishe9cqy-jn=Y6ndD7KT#`P9KlzI5D+9r&(U)u0nt2Jq#J2D|8l36w#$eP`r1#|i^IK2;* zA36~|Q|h6QiN<+nYi9NCNcY&*7`*Pl?-tP%HSEvD-=m1K>dqn4D0D}3rdqWpM*9Y0 zku{j)s{Z(1=*ixxL$JLg^*I-H;j5v<6uo2VIb$T#JYu;7s0d zdX*1lNSA)hZxcsW$~a=~4xu)A1dFe9V#0-fWZmvchc$ia@0iH&i0)(t4JIh7Kbu3M zNKWZb56Nlh)ORqxoc$PS0e5V$Fs64DKUN76xEj%i~2HtKv!Na?1+Va zFIvv&M%J4sK5dM~GC}6NLEV`(Kx(4)j$+uhwiM6rN9F7&rnKl!bG2Ta>)a3H&;%@3 z_aNnPcicYq<)mdST^ps+=ypHiF7@Kdq%K_7iNegdJ^LR;GeOS(oKqO1R)*83NfeF` zrN8>*jx3lsgi^044%hWyjYBl=b9!)lWg@!oWWS(ycjAWiB{O{@6Uqj&PrI-5-5to+ zCou%SAHb-#y-8?2j4%29xY>LVuGRgxFesKE>4_{-deW}GKeva+qA7->GkFk)=5|L< zErDOP{n#VdJ8EMBY7sG%Oc_K~Z66kDtIE7dhsDP8aSHY!~Hoa1`IJH-k^``n5Z&Byt8ZDxRt8#0K9@hdidA>l0Q4_S7byJhL z*E$TY(U;F`RX(j#Mtz)M`^?dn$g9;2-%Jmv;{C23xyib$Y4pKL9xf*tkzm%zQDs)<{!Qu-l4C+v?>~5t-zm96GD6dp{ zPN`Dfz51%0+@^y1g8yjG9_oxLnY~Jg;LkDOj+;zRGHoOnKE4J zs;=nsR8h(wE6(OOm5d+vl~QBeNg5ODO0vh zt5gn3{M$6EN_k)XK}p{6PDxu-p{$$qRq<>5L8%ESQxhQZpjX{GnIJ>hUFCH7x?Sm#8r)n{#UYm3;9hTkI=Ft)z1~_REF12mW_5Y)E+NnaY zL%lL|++W2p?5i@>^R-go{6YECwnFLHvO+of?yGYB_9vyi|7&ICfVYa(?)OT|%|Dg6 z=~5f#L7j5^l?r)nRayS;7bR3vm4ywflwSrvl*KRqD96gaDz~+MD8U9_6)oRiihIsa zrPs~h%8edX$`sE(iu!?1ifjER#dxp=kCy#Vy!8Jl;eTqB)|XYq*-90mJwrtV`PY+u z<_|NSYv^)MMI4X&gGQ-}2>$k)XxHE1U(KBhYU1c=4bf+Ys<4}_Cbk{Z6lEEjqE9b1 z5vZpo3Wc z9s5sdh$Hfxd-OEL*Aew}_*Bgxy=ty~|4Q)MFBGX&vmxU<`=@@xbI*ICnv_w!?+aUN ztFUhL9qZaU-hNUQ!?($LQ%_Bi`$t>E4p9}&)~JcS`WnL3R83qnQWtafX^5PE)I^kx znizUXORVXrA;xNHi!-s>!s)5LP;JvdwCU1F6x22l?Z@g#ok1NDv{*xQx}hpYH26;C z$XdoetRg$Jiq=a%vt`z2tO7ovomIoNDKZv!)R362nkX2dB}Pt_HJ^JLVw>dB{#>Ui z_ODYDAt!23yRIg>?UieHsFqLd>rl_G!aMda9cQVF15azo_*BcTXDULqQe8|~qa|96 z))jpfJ@NIEju`n@SES4{5Y?6X!oObjd;92$2n@wDLp|{+M@x+FsU!LuXp5zPHN`Av zRndQ`mKZTsM?|(&5eW@6#JBNkVt20^2DJLas>we&>@tN{(_=|38B3qgsl4tyR`zHm zkZ3!KkT#QPkv;~obu4iP<9IY8j_kQ3sjD8r#*=Z34;w|e_E36uji-CzNE$|rWN?#M zy1nX2VQfE!RQD$?CXNApMl-;CG(SBj@v6%h`MVX*j2&^Doe<0M^`jXYpFmmtD70e| zs4$L4Wx_!AnGa#sjsCpZGnDQZhI9YeKxB71ZvzH0W@`*u(mfrBC4y z@@xjs;rC$j`gA4FyC(yU2Xk>{Uy61O6tae<*rY!`L1$Le>KYuzYkZ-%kud^KAkJ_Cpx5vIpbbWi6>9o+Vv}GP!;b zZ6jnH@U91o_s7uWYfm~n?}y|k-IP4zDt6?@&9=D`%6MN!h|K-n9#%%bn5r!6xX}{fw?O`S~(>IYGI}M15HKD@D zl9e;fIk4A)b1xh7w6X!~S~q5bia9Ywro4G(&xG4{G#zZi*`YRU_i^M!7i$a#*x>fe zg5NO?^dD=*uRUg1-*je6Ln|D$E%~;|fu^~(_}{c-iWBNCF4!+}CDh!O zq0KF*8f(e!UDAWLr!C`KIDZTxp3Q>R(~wRq9K5vSpbK9-7wZ``F_-&W5gj_IRvt zVCfl0`pvUPW1^f_a|en{Z86^Oz{96b-0kEf>q#CgJm<*W9j??2cSB>UgS^LVdEdiM z#w2#sjB;V+7cVj#UHI41i#sb^@ITN5z9*VaKyxNa?Ck6VlXDs6;Mu9+y!B*l(QybxO%T`33vlscXcH+3y zfJ>h2CjLHj6XSn5izA~Qg!ezL;+vC$cqy^1Y>mbu>``NJai)oIkp1_ziH${tak8DvCu)BQg;;} zB|fTd;VlNGG!=n+0>qWph=wB&i4tq?aWOzFNe&b@>wJa7WDhZ}wU=BUPvN$}RUF&u zA?#+lii^HYL{4Kbap8-faeQ;}&p09an*@ue{enf`d(A{#TodtKUK8^| z57Dca)RFRc6F)XOh|u9~!t;Wo*fzyQwEg5PdL}rF4h!9c$$B3#LM>R%y}8)-v6%?H z)m+?as|ce((&KShg!n6c-`plfh{-7dV)DV}Vp&^1u{zpM1iteYDK{H3d7=?bUTX6; z+W=*-HhLFTSs>%p3XSi|?YJss*|#6ctdHN7`l%NRUXd!i{P|1CUt6wZ%>Ja@ z{QX_&>sF~;?DAe2SyQe|pZr0oJ@iTW<@#7@Vft2SHM>lySubPd=^vFR^L{A3tm~Az zIu&N8)GI9xsY-uPb!=W};vOaY)h$$c5UIhE4S$qoYwIPi>AQ04d%0q6{8MQ+^R?8x z`KXM&@<9gyNV&?xw@zMXF*r}B%Nd=FU(;eO__b+}_%s$sD z`_jKDebdxvxkjCL5WQ=+8n=(x0gA%;2PN|kL;*IuSlp34w%F&}= zm2uarm3=cbc;8fm7tJ)}Uav)?N)6Vu)aCRjeN3MJR)(FaQaZY-(!{@38FaZ?SyNo2 z?CA1c89n@$lJ{4I%S&q&gNbtecdL@UtyY;fOjABPG}soXBb$y@7;U17iJCrN3)HBw z*X7qnO`=C=AQos~oUg^~N$NB^uFkY>YA6lWNW3Sn-PON}bzVb0X6iC7PJ^7~>Lkt8 z6m?%z#SYomwO?C$!t1Gs%3>9vzUVKitLnM%RZBS8YKd#Jqz~M0O)*7NUwQ=@h!e$z z!f%|J_)=gZW@t4MiQZPy|dg64xXq|2<85#)lh=XU@jLXOq52PB9ewB!<;7Sx>BA zqb&j=v_z+oIzrsh5FHIP#Y%H6(SC-eINw5!Ek{QLj@1+<#+t&~T}S#uXp2+d)kTA^ zs$yrgnrJ*-OAOqsBT5eGh}fC(+=05{?LK)9TWt{>rYg>)Yl<<~v_M)6GguU1KgZHW`ZxBdo>e+ZN*aCUfC7)LLwtXfB4& zw-kpxti@Y>C-G#Fxj2*1KxiB`7HeYk#P#isL}avq=sH7RH2POpR7DvG{b@QP_lT~T z7oaWLDEcDnc0+No%ur~n8HvtPBiVMbk#PLaQt0G42*b8cLZi|_Y_xP1`i72TkfV)Q z+tW;>wl)!^8pSlYI-j>~3+WS4KoW&Ks-H(vk7BZVtt5KuLQ+&$aZ7E1)Vo~F+>3>@ z%*$nwRyMJ3vnbJ)*h^hDhf-5{&^eV(?%6!pltJO_EQZ^qGF~l*^>Z@lxFVn6efe}4 zk;)DIH12t&(9bc2+UwIXd^m~LmFYNzW|Gz^jcPKOC*S)n&gJgTES6j4<2qsv=M3`D zDNE;CQYMFYrZHwrHoL1*i0C(k<6AQM{xyev5m`hRO5A8z29Fk{v$-gT<16#9RLerE zqr_??F15aQ4%^LA763T)ncG&f^y1Sv#MF&ld7{YZ1qa zim;XQ-|?)7l}qwCx3-Xw{1S|w=W?;#BKkKfq~Y&Ex!)FGJboVfnZ>9SF2MHFLM$82 zMZJgQ=ASL3>P8X&4JqMO?|Bs8$ma5g9KL%NVisP=wuuX9-e)0uYUfdAUBYzZVge%< zK&N7~YfC9Qxs4`NR+&E(~RYb0Fp-oH!-D&GGDdPF)+V{| zPefC?rutE6Vv@_^uJa)5bxJKOKZ$T@a1p0_bGy%f^>p1crOFD>INQt^KGo^re2Q zAGNB!%)jMHqHrWZ)s7Yu-Lc&2MnJZk)ChK_PmnW5UdX!jHz)S^d6R2lkKRXTvMzda zC&-Pqd5+}kx$w~1iK#(;oIDIl!=uT)Lxiho-2eamMbF%IUdiw-$z$I8dXB12q2)4e56QO?m zZP9|;gkYHwAcMCI6@2|2LN`B!%Tcn9 z9@m=slrW}@YeBwF7^%a;xsVdfvW6`fcOeL4mk{g&0_kfOO6s$wxDODRx+=sQgpl`8 zV&geMeCrs1?Zg0j|4^9b-jWV&0{Ahi8J1(3v-wc~`q`drkk@g`Pk&qkeOR~2i^XM4 z8I%{w42?jdhq~j@ycx;w{Bcie#jp|WxD(x)%f4-qwg+T&2*dD?!h?P-`QZ>q(P%mD z;%0m{3Snd87ED+XPEGqZ_{mzk!JQ7YKii2{9iyZsYBaeMx?#4kE59weV0pPcD;q|W zdb$J6ln%`9)QPbUZHZaaiu=vm;@lyUgB=wl%!)<#5Eh6?UiXc_CAhuRHjO4JxEl@Z z+S9-Lx&C+ankuu>_MB&6%VYM!Kx+e_tQL8vQWZv}}z>zg~O{5Tp(-gsRGUD0Ds6AgBq(U)8s9ol*5vtU&tGMDPJ!^{vHTU}bHXtAn+ z)Jp!OiR*Y(Hia9|Cq@^WC5H6S*XQA74a`~_Fd$BsJ%7~r5~@SuM{UAh=+Z`>V`*Dm z0%h#jc!>(VQZ?jrL4~Y_s&qM^#`pf3L=4g+;)XVR)@Z^+ZJKqeS00Sh;_-eRxnFAG zaYLUCmyPheqs0$xBR<|T;FGlhm#X!ct)@exQ3j0Ip~K1uUFOy6GDd12J*v`Sgr1h< zdHhzaZv0YCnpZ1l8vItCHmp&usA{6=slu@zzm(YADrNbGTIH5ktup_YI)|F6^Ph$x zeZ!21zpTwfIrm-vYEu157ZVKmcwdc+$$AWXU8lGm)y8PK7RKkkDKFP+QEj7z!%j77 zyQ@kpRZU`wdbFLWOZ-tCQWt5n`;{K0YPxt9Xfwl4lbIj%(e-M~j8HuuOHZ)%j~fsz zV?4bUs_e2+VfsEzsqZJ(X1pHlTN$y~SDiiyx`c;o6S_j)5G}&vh}qPQ%Sm0yJlO~B zi^Cb#I|;X-k(j@k&FvcrT$R|&i^|Rto}1 zl#)Q7Oj)~$AA#=J$($HIjn7tNSly>Pt9A~gyF)Au-}K_hhb|1b(jDX680>q<$pHkNQ!r-<|BRJ~$gB@M%*bqn`I;bXix1xeO+|^I%>cie_ET5PXz@Xl(As z_lm(7h7O?VlF|HkAfBWTgQ%Q02>Z3e@%x&9)vrW6whv(FvYa$8bUVn=L++z#{kFv~lPrd1SrV zH9H#bDbZwH97z1~9u(yD;&V<6!%7EpK0SeM3;Gl4*@K`T$bqaL9Mnx9 zrZkDQ>ysJ#VG0u>M=-v02xb8>sH{$8&8IjXu1TQKcmn##!|56_nUnq#xi@em4_c1n zanD&O`eXQgd>manBr?=%7~Xe>vU&R`8di;`ZP08UTg+rsvl(35J(D*3XH#M`UiKf4 zV%?H>zIPmi`G#1mQYO;o@OWuUe@S9|{B+!^M-w7@ zCyL@Gu(NzDj!P!d;=?#b>kgz%)+nw;$XesFAyn#)qOo}Z}-hF8s zIgF3WI97)Z=Rglx2UH!#gqlPKN==X_;e%;2GM1Th2D6~=7+eOA;_>=1Z0R|i`r24N zyT;JtW`AD0L^Iu>A4dKXU)wl@q}3yt`KLFjs&R~vHN^Gu7z5YITHDbC{x%rNn)->v zE*{ORq$F&fj$w+`XgWm?KPaNIkK^~p>0EC*kynqWNUUrEU0%jB z?PnZwR*&M($1z;Wl>9cY(L_0n<;BUNte4)LzkZG6N$7Ycl8%@HDQM@RWddYT^s7W5j#A|~&9(flbD%{o zIz<^$Z+0$i`sQ=xZVpr4&LHVYI-kkL-5`s&9|WU9c#2xZ8)^hMcyAy{3mO>m0Ggqy2pmE5A11F zVTwm!L##T;{?Lt$Su1rPsX3&Nt$8t`Qcp1|5FUg62!jsk~-}6osI)*p?@0?!>jR z#xK#1_wTGIigl!?xtzyJd;VOo#$ z>t<}ZYrr}`Q`BCWuxX_}hxeIcqvOb?kLC>dWX=jVNBKKuhjWDu4`y3%`neUI8pt|y zGkd;7J9E&*kt=mB43xFqzA}GVKEy$CUhO$J%Yh`B%bcyUpw}2LN}@g4-Nl*C|5)<) zlNEaom~j80IY)+=q1D2Q)l+O3mu7`)vIRGrIPkuq1Aittk{sdA4pUD$cl1N6w>SCS zBv<~1BeZfPc9i6|NL`c7Z{0}R;Khozp`48Lrcbde*U!13TG))Lv`_|T2XpCkF#jd` z(cRgb%eN&@=Rq^6vk<|i?!kE8ZHh{aCyRFa((ik))G-g@()&nSm^R~^c}pg47opv{_z>#uMtGzT z-8(ghON_M{eNq+D+tjThBzF^}Kwt zmb%5s#`wc97wVnlc{$WM!Cdt28N2^xrh<_sewDzs!Y}Ps)Zm%Y+ z(HiXYR?~0y61JUL%jev66pmU&>!a&XzrUW{f7Vd`YqRY6*vO9coAA22g>_ALupx6J zj}B}m)%_n{uUXH=9V;;3E9*O-mQuW~l$|$M(m{JUFN4=n>9dV@!CO)9u%Ce(W?jZX z{>C38>C#EgJUc}A*FzL;I7(E}c0MoPLW8=kJnFa^?Wvn+baw-)v)7SRx1Q*a>&Tq3 z4$mHI3Hq{u5qVps=fYP0ll~PqS8d~h!w&vwv_;-8J85!xI~{_z@GxdO^OJU?8o8S@ z7q(yWb~<0$j^&+wyx|}ZwjW`0z<$Ed?Pq$o1L$@?z^CP7=n;`f+q+|^ zIhM%VIg@F&J(&rqvw6NU1&gRG_Q~h&w#NDVmz~S7svOoCX7j#R4ln2D(0fZ7eII0# zGBu5a$&z0!d-Y#BXQDrEHox!ZFfcNoiZ${%=bFr}wy9W`WRdJWPwJy)OU_va%dTee z;=gqE^v}dbBbVi2S@`{viIaT}8QPgl377og!(#DQHoe2z|MlV=8@Q)grL zZ6>v?QZW5BovA~obE;?xW0xfpy&#!Y57N2!cQ(5UC!up}63d^@;KHbE(xY<8)XqV> zYbu8G({S&eNx{He4v&x=_<32}(433AMh+Hkb18V8%i#4HX#L72z%Yj}hZIh%nNH-d zu`INm%z^I{aO^vp8t(+gx0{Y`$u#D=r0{-aIz#mG*%f~->jO*ed{cCy-8wQgCr(ipT>j5X~en8b+sGMqUvc}yDsy^fRWjcZ%yVGdDErpwn(;2-h6Bij5KeCnl`qC-1 z$)1X@+qCg)?u?%#eIzrf+%THs_eQgcQ8@3N%25ZY-9K#-!zz;ba(EIyb0<+H^(?|i zrLgo^60h_oV_Gzw89&ER*Es)QJ&yUX=ITb)rCg(F`kHYU9mJZ41aAFQ0UFXoexJt3y@Kb5DTVi;- zUMT}Is}=1PHOlyH-;~JC70N`(eR?SIF8kJBl-AupC{rb$>*$Q%is7-}NZc!%BE+P%HsW>75>&KUo*5> zJEcrHn)y}nZtz~&Vf0y9KKzZ6mGe|Fnfypu=lEFJaN(x%e$#zr=+(Q5)%gcXRrhC# z`k(iT@7{8yS6rPkzgCqychuN+qDIjTeXDG>mAtc{d&>N{yNdt0`-)!SJ;m$oEv3KN zb?K4(pR)JZV`cD#TgsYO4;0P0ca^P|ZYX(i&y|J$JXdCSeWK{Qyiqi6eO9hCsZ?$+ zl9=Lv3PmjXq>OIyQt52+NLd#9SlM2FSyA=4uQclQNI8=JT-hD_PH~v}URibNgYwJf zky1JSp)&i?Tg7Mjb0t0dixU5|Oli~UgYrP~)aH-=q7;3oQevKdRPrXjRvxr`t6b~z zQkih(nR0IP3#Hcnv9jj%OXZ0A9c4kz3uW=EPs(MR3dN%BOXb}Dr%Lzs50w`WUMY+=KoOg&%Ra89DlDc z@U3#;{RgFu`X42~?yGWbV7bycqEcxq_msiA7m{gH%$&kPvNn~_Y5XDr{w$_h;zH8D zF2dKV6wR-N%$ZljZ;uk(yDetj$ztk#i%>r_ANPWJyp38w*3cyww=ZFF*9GJ;ms`a} zls+rOsK;D}&0oa)WyKtMQo@c4g@k*|CH+u7Yq}Ls>{U#^_H%I^IG5y!g_y1^AW!Cw z+dk*9q;DSmpXRYkHH!(CvN7qD!3W1oIz7x`($j3VS!Ix(m_^*%Jan(;W74sJa)}%D zznY7>Q#O+q=HapzwFN9KStfI)Mcgx3!N`O~j3`~iJI_*n zt1TkYdLgB*^SLS4ZCpVCr=@pUT$>V1M(3kU%OP#XJiZ++r00YM`k3=*MU$~58CXpqldi>e>ALl_tTZ| zFv+|5&lsIX)|@|Q#TP?6z82XMQSOX=T?5&ZYs{Sz6Lw9MTBwuEP<9$JP3CvErZi%1 zT|?G;HIR9*F0Y#C;%{ZZoSTLm2$SClHDKvb16dm}Vv1;l-c5b}wJ}8E#muso{G3Oo zT-<2H?oCEKnqC78%>y2(U?3vdxk%-WOQ3= zULP{Ycex=6?`7_~+?Y*S4cY$Ol7_;P;j0~JzQmf|GiC1NXhG*)4Jp<)B00(w)fL9* zpEakcn z&ivchirG!=32SZ5wXmjgkMP0H)QiY@Zltz$=B>Rwug*BKtBnimE8JO}<4R_%JxARg zm^aOZz%CxtcXy}S!I|C$&dlFx!yS8D`TVn!99w&Wf4NcWWX+6U@_aq5@IPh4em@6U zPqN_KB`aBHb>jYDXPlN=(vpT#miE_R{+Ai0*AZn&Oyp(@Ia-W$BR_Q46WGhU1_ zc4F{mSAI`(W7#fm2A=R@$5mI}UUTEyHxIn0dolW^#QSPwkI90j3~l8~*b_TOX1VZT zoWvSWIpKNRlbVr|+ttp4lNa3Coa2JiEhmZ($)5Hkcg!crarrvav&cpA%UxtoffZMN zSkQ8>1=j{Q=FT@;wvKXOri%m5{>Q@&ia zm-q4{FG91t*=r>=;s5cbgM~i}mp7rdz>W3Z&CogQ&9{s|W?X7YyS)y~?QYFb6-!PC zTRJ|lVcIr%t@Isa?~4@%O|5x!$dXKf`vEp zlG4;bvPxDmQt>;#zdw3CkLvTeN1yw?&+9mk<9(_i|Cl^f9?4;9i5f<@kw())I*Xth zR47+3=PHQ|RaqSAQbd3JMED$2#%pUiYpUKQ>Dm{DF^NM@_02;1?q{DAn;cT z*>hBJ!B-8FJY_M?QyB^yr=pkozXGqPp!(rd+|Hi@D^W#Eh*gBefD(R+(KR}xh6HOB zj8v0`+qy|ey(JH?2nGBYuLPUbQaGwS5n7{V=$OY+W^5efh~QU94?f|VBtLlBz;uDXVT>Uu98FQ9chf&Err1|^0-k+`}R~GsSXM# zjaR`iUj;l%RKetrQ{k{t4Qiyj=&n~ql!|cpZ9u-LYu*dfcO4k*v>pT+p#bZOR6CRZ*@w#swEwy2H247b03)uwQNql7u#+ znld{JcDO>>b}NiO?tovaJNXM7@YKi!`i|7o>2W~sbT9napC%a&c zHSISYL&q9V2z+-#jqGL|$=HZ#9iAwod{TU=GmesmK9l^mzrNaI#^8GNY;wS?G-o)x zvV;D6C#>6MOV4>vtnAqU(Pi$Kt54aaefF5R$Q?JoQ4UDNhx$>Tn7+XoYF?zVZ?lJ5 zf+OO_uSeGjAAB6=f)}fuaOIFY&W!Rws2lli*Z5-eTX)D#r?p3uE;OK`@OMcd)N%n4)Z@A@ldGyMHGA&Ab)k34UD({ab0 z`_8C!+<<|3+wgT$KYr@`gz%ZSnBV#lZ%%Yluk|hUb-&?YXcxxc_<|7f@hfzE#psgn zsK3>RqPO26tMUciBH!`&>o-gZ>OtlE@5mha1MV08U`=@++IN4!7K;zKXY>KfMLH37 z;RCz|JK$jb0-ISYJ}i9=pYb2C?bKToNW4Ye%O2b+`ieJKzrZA;7cz}MvG>+5OtKt= z@Uua5k0p-6ubxQ7rNe`HBPRSSrM`?fya3bP#Wc4eD>ZSh64|qp^$A4M9C@lPh^rp9v3jc!4&fmykzi@QS0RG+oi@^Ww`?zoL zlm3ox*M49G8^ZC#zmRYrz$@Lqm?|L1H+&z0=Aa-CzxxC44TQOi-Y{M&CdlI!3-Nk- z56g3ZAnnu#AFVzl{rrQgGk!u^Pk^{J!aQ3|gr8YSbEf=-`K$fIxUAG@Ui4rDF~G-g z^>A_ibLSX->FY?oGp~{H<|}c8NqMO9nCMvkK%?xqqvB-DBspH zoR3ZW%MPRp!Y@z|)(eHuv3)pBw20!Vp&-J@^C;LSjE=wocIL?+W~cp)UG?Z;J{yOa z{}>?zTok}8Z4nGVCXCb_!{9PS6fAu>svZmC`{ciD-ratd|95~zTJ|%)*F&uRWFPaR ztW!b9FIKi}h`CApVrDX5nUL`in-)_ zx3zI34k?Vs(}`oCwOJfi!zIxpDovXDL=;aGgKo_Tc%_fR{>x(!sWb)?j#K^VDTdZ` zG3Xo}iNom=p(!yIW5dOe&@mRtEn`tXFdSQ5N25D^3~UC*LH<9gJw3-_b$Ab2Lb~%{*oZp0SzgY;|nkkzk`I$v@_Onaz1I*F(Co3H( zfVn3{VY+mXEj%iMgTsW;xm^fW(}yGA;}DzcGaO54yP=W7n438a_fCnxV?{j8`iX?| z#b^v63RC<@TN8+b&rKlbC88JwZ zqxqDi=RK~AfY+fo7%d9G2cJkpq=dq9R220WLm?~`iQ5}PXi`H6W>_g7e=Qx1#$@kcEcz7!^(8xyk{FJw#2gOYg%}(yZ*| zQ)%9A8s_(>QT??S`ALT`?cfnuHy%UaU5PWA?O>SD_urel?h~_9b~L>+sBvGUKYRAor;S#t&W~ zLBAEfya~Qmlu2LQflle?@D*vry$7!`J^ndqrp?%KqZta{p2N_t4iSz|p**P?daY%! zKl~JyA4~CjO))BWK7!(wO1vql!hxbjoSE4O%d$quoNk2r_h%Tfwgt7dZ8#A79@}od zf>Xl_*oVAE^o9;}A8R1Lcr|49mZR9T41Z7#AGJ!T$(3TzvK-5L$`GCY6jyl}?&lF;PMiCS$OCg$Dgt{g5=(^hi z3%>@)&3%n} zGXFF|;QcEk&wNcj$ad&_d<)m`PjC|Yier*LaPQd%c)PtqhYaQK>fYn<#h18a^Au?f zHb|xn*Y+{%v3X_yx~e^}DAXM(V<~qe>4wQSJrF(F2etd{u%mE2zGPY>Y=tWZm0e&v z+!@zKI3j(eEyktT!?fH6N9lWS^~mQyb;fR2XJlDzL3tY0W!rb)*=JwMKT|e*^jefy zxZzp87b4}!FCpLy*~4CFI^%<&9v?Utldoc619oVx#<^2#z(1{pcjJ0EnmD1Q!WNY- z4tVg&1$M6;u=Rj`Mv^m?!H5m-!C(T^%ys*;i(3t z8g($r4&j&9VdO+R_|103c*?DHj&a4i*)G^{%NuXYcHm-704m=GVIuukO)LyW;TsWi z(ixqq9$>c4=yh~JvG!V6X4t}yIiul_C06tbput@bcUJvn&J%wzrXfJtu;1*8`(O5R zb1%zl{mj-T3Q(;#z#<+Eu*;%0haSu5Hj?=&uM~CC*R?ck^!bf z-a{|7-z@6C0d}#vhiNVO$)2STv8p&h#Jv-N_?f@#Qj8$}J{QHmg(F~8I2;|}MFzoFg0l6I`aW7dML8hX}z9|g%ga6pecok?FsKcW~7295`LGG6tq{1{|_(lZ< zzZBtOM0!q*3TBi3H(86;ooi?&g_}G|#wg)yq7wCA74fe?7PeN(s92sb<>6#D5ry^=C@Yo6-r=O>luF}Ts01?h$-v+$>2b1hn4KvP(~0!jZE7%8 zQHHaD9R4$sf#ysF%snlQO-rQV(Wi{ZpOvYHr3M)Rb=bMdK{G=Q3m2(i=#~=Z=PE-! zL=8g~iZF9k!N*($=)X{g*)>&IW$9q-tLYfpKOK(yv|wC28HfIx3|^s*z8ZCmZPtLv zd`+lEsbEQ)7W!u?qxFp%bjW`ZRXhb|rp8HINZ351RwZ7%@#2G3TUkh_Zk;uc%^ciz*I%QNja9bxe(-tYDZvYI?Nc6QzxO zOI1XEia^En9S~Ly#MO=DqnsFn_7A>Do=&;n3xTLo3c-=)FvwD`w&CRt70u@Lvd(t402~hAv-=C zVdKNFQDHlBwc=p5FA^TJqi|?<5Z3fZU?25?jhf@}jJCJ8It~N09u`Rt!)(zQ97~LV zvoqDFSy9B=Bd=Ih6s^UmSM3yr;B#T9k_o}0RZ*}x6ARs(Fnp$(>$f)bmbb(sDMm z#2Ha_-l*$$!rOLFoTQwzh_E-FjtPQ?%?^C=+76nQfwO5*cyVwCM)62&^4p1`VKKx> z3WLIEe`Muv$Gm?#v1%v`Ig3IuV?{91rbl4u;|M5z4#LjD0K9qYhs6scps+I-QO%Kf zq#g~)`7t;u69?(Hpb0at=-v3V5fCHri#=K%Rz#cd#Vbv4Z`b3n;T z8?-TRIP{a=vYl!NLs!^YTBDIN*MDrNW_Y$9w@9lQXKsTDb2p>4%>@g~N#`IgpXXu= z*ciCssDv#(`+7q9q7OWyt?<*v0zcF@;DLk_=F@%_l{lh&mp7KTyJJP)7Fb`k!|0u^ zu;_Qi!wOd%hbKBaw?S{jR{U-CB9@vT64E_z=BYJ$cdmoPY?^Vg#v5y|xxh%)1p=F$ zp}5HoGtaxjD8&(9E#1(x&kkFxZIGXC4M96Q$Rw|WV97f8CfY;dtOIdmoMAkf>a1jM zXx#F`)CL#ynz-W94i~8YbO-zGjSH>5D5~(ovCmtvTR#By5pMW;$`7eM+wnd)0{WR@ z#M=u+_^&W%r$=Lo_fBYvM-!uO2Tb0P&sTCg6dwBFXnZgNqN8vp-yf<&zK9Eq#x%Mn z=V%5&Ff1A>o>a4STH}9Bq^6nUdh<%yU$MdR7;9L*Tm`|Oj`&dM1+htP z*gbO%rXIIMcTHJJ^uEE6PIQl4e$Tv+wY zL*?X!X#Bbi$E{~$;X|M)))Yg(jNux~VWhki!l&jVS9LM!k{024h&8Me7h_V|0z~av zLOGMg=&E0ck}hIlIW0w&(*isQFoVUI#jw1#2+kvyV1O7t4wg!CQIC%-= zcF#v*;tEJdE{5*+c}P`SL78)^-fsn{Q{2Z>%lmoi=d z8BN8CW_2W;QiOMmCiGo3@MeWNF8^pSm57cCbk=Oi=@(~PGeg?ecEBPky?#t^5Rb+AH`W3%8aSRWz2?qx-+ zq8z;@Js;2gnF290ZL|fMDZv!`5clVs!lQaV&h6q*eLWWnF61}-X++FlJ*-vNh5063C@IZGq|-bEn9Rdf z2_qZ|F(eKb<

X<2_~n&4@AVqCv;os*Wc~Q}N@nI_%YS=(A)1^La*4%Qb*vr_IWyj{k5U2KOJLAbg^T>WQg@BV98S*s2?+e>F8M)y%I2( zZVu~0QwaW>4PI@G8A9#&va6N+tIZISe*=@To#?B5kI}B(SRVHRy8IQ^A8bKbbq$t% zZ^Su;=a{;%5o^_|ajN_Yjx2lzU4v)v%6o#5zZ+1|(2SkBEy#OQhfcd@_-i*KyyGb* z7d^%D4GkEmquJ52<%s`W1^WY4XunniS=kbJoU6cr!$pX5E63@z&oDiq5!Gi}ka)cX z$3AvKWJWinE#JXl;02PewBu=WGe$|*A^h_bD3v`!@2@Ag9an=M?K;S0)d^b`w*0j(8pEnAJ%6*$CmXdMLkX#u@2moRw~a=Gis~iF`nZ(QC-mc2Nec8(G9q zy!-6~u|s>X>su$fd|qMo&KFQ_XoO>8GuB>h$Ia_?*p$+SVS29-^7lQ4Zge2n?Hv~V z?4WsD#KJuF2^X#U@!hHy?>F?KZo(IwTJjC^t$T4lYY+`R-4HkVikPZ57!m&lo9*6V z<;55HX!{l!+Apy|gfx)ecKFTsfUnQq;I2v!6xerUkq-Sku@{R9X!bzdcSKtZP@m{8 ztmOZo>(Xz;<^IA-+MXTz3xC%i*xv93DZ{?w!G+IQvF`&O&v}cP$LW~#yO6a%pBZ-F zVgFq77_{=)rlwnLYubHwBIy?Ut9qLSHr-?_y@1tBy1~?2Z?cPzuP~jP+3fO!%WT8e z3+$)udFDR+9NTsK3iFWq|uCwk#cbW6j zt4wnCZMN`b0XwxjF>zI?gLOlMqShf>cl z_rIr^{?!cDv^bMlZ9T{0HlJr3i%zh4&1acf;6>(dmca@JF0nev3|2ojgB|>p$=Y#) zg-^M}X1%`7)`;b@Rc|k{88Ufn&4e2)`Hn;9+3W)&~9m|SZvlNGzqCU|AD#rBt(!^3NA-IXjRFg=&$uFoU>+I1!-b&cI? zxxLH%V6 zq~+o;r!tZJmI;*ICGVS0ES$f`V%?80S|3D1$UFk4??qvPA2D^ccHsOQf1Ga E68 z7`Fw(V_gjHhy)?gJqlB)rbTTeKD>y>y{*wW5*Cgv#i3A{9|i;80Mv}z3i*+nAidik zZ>S~@7TW@qY2Mh~w+YYfVqmf{3VO@JQI-`9eX3<_ZpWZXg4P|I!!dL;3e%2<;_%KG zn5qY&?YuuMQi71NeG44+Z6~hFR%F-&Bi=It6WzB%`$!xnZI6WR;12Rl?m*z-0LU!b zN_tczeA@%a%O3`x)Ij7NjK$&^@wmS{8fP{|;n&4@`1M4iqdyYA?RMgp1=Z?L;xN<^ zOZ9Ie4wDzO{xJEfQetu1W)BXIO2xLQ16V&}KPEg#LBqi`u)CDY{*nsw(W&U#n27Z} z4hP+0G3{6)#*9zG>%3$vxU>iBUG`wguM|j??uGrn1331Z@-|_6VJ-Y0^(>B}$C8-9 zl<#;y^*E+PAB0ieF?^Rlg!*5H;I(QG0zMx?kz^VwO-WCtEPwT|!_ez7z}H2Z=vgui z=~w8vyLlQ!XK16gi#+E!s+f_dg?Ca4NEtH)gIyXl(`_2Q8EIqN3q$ne7-IJ^ssY4R zQLL*8r_qx!g7OZ>>{MZvr3&#!Q;@e<6OtJ!Xqht^*;l3@fxInw+f~rMRUJc5r=aqm z2IQ)yVT9@wOuaM}%Bt${{!Mj6yB=vCz<$aRWY3|ss-6yBE|`wCBib07tAydXh73n9d}PC z;y7hQE__l%^jSsJepCd{CC|-G88lhRBPLb`j)s#U@mU(d!sH21nv9~8N(gwZgzL|g zaE1H?vrkXP*-SMwODW?`{51SHG!qV5I*{c0#D$`oKu8xq)ihDiB!j?0MX-F*LC7C) zs)MwLQsVrZXcBj72EKmQg#ti#AvIVTPs6r<({U?64_m}_;B*mqlM0MU zFhjy~j%8WKSRZbHWpfQ-{m2+gG#Tmp4C}{FN0Wjc8a=7bOVdSswHC6bXkj?%Iy=Tp z!_9Z}^=LZz2Q-N3sfC$N>aeoZKxMTW66wpF^paiVVe+~(9rcuPU~bbHdd>Jy znvq~P6R+Ru!HVj*;Ck||l6mF^jDwVMx742rSQ%nylrHoF3=vaHya6#I zm{Ue$hVD%0CmB!M-6+o-ZjSi0)Lumo7E5rA>q1H%&ayqYPD>E*xL#Al}Rn zv7Z@FV>TrBnc#5QY`mv=67Oc4LAh~0 zMzqoli4b6WuL%@>FTm7r6V!Op42w6@VLILj(r0w=LEjMeFZFQzk`d;nnnFEmE@>9C zk-p9Z#{W2)4;mt)W+t3fiJ5Cd9RHtG6R+0CoY0vFq}s2-)|gle+p+8^c`$|rBYOB2 z2zN%oK_Ld_`H9G-oZX6>yAj`(ga_Sm_-9KR!yx6}=I@5zi9KlBkc_+HiMUrD2M_A? zEs##cjFKd*YuSg-kN-oY!a+!~eF(SN2Wj7A_>>US!!i};bPmGa?+{MA{)bCq2OvTI zN`=5Z)KibcOUlhXG2R71-9+S{NP=@)GM0zz#BXPseRedM_|{uVi{61bQ?}xYa0qGl zVfeBx7;8?1Vdt7CEX|E1R&NAkJcDrLa~K3ih9WM?539Fr$NA7mtaup+2TOmtHa##n z%o+1^-O+Q@3Ei)qV3FW~!*Sd2#%BwvI)iX#33*FIx8v#Y?J&GRIokEX_+_~rGScM3 zXr$-P-58X+L}Hgv7|l?LA#YDK!o=vAc_|95Cw5}=j3kI70qXtgFnlKd~ZK z{95FfSt9to4SqkhfZSGFD8H~o!WvtY39m!n@-@gOAI{Atdo;atz~8FXs1dV4wt^Kp z{LJBFvJ`=@7ZA743=i~8@mk6h7sO|yM0+k?c+SCQYh&`v&>FpH7LK(pM#CGLWe{MF zjpHq`I@$r52klUq=ty<31y=Z2BSdf&y!)3T#9}ecFEc!3o~4$YwWexQe3E51VzDlSZ%!&nlel2p0x~xmn`6} zvl_3zTHr&61v(8Z(B`rlKMPkND{nb@P*=df*&OXLOHufD2^@B;L}K((oRwIP8=CeQ z?YRmovsYpKLTil4UV~W5yL(dq>)Qfnv_G>){zDfmRdT}}Wp{MeZGu$RCN%%qgnT7` z%>B0s?WEamGunnsWe=PrUCM5l2fmoP;4JA>j!T`1S>cK^4o(OixdFoz*JHY`9qGl% zFmp*n^7Y*i-%5S@fe6aG?nHY^66UWf75^(%*B91>MPET+m*xw|y@1fl8ie1?MA_<9cDM%fjLUUhI zp-7tS$DkA}&ZT_GLdyF3rDAzc3|z|-@PcM5RDK9XCi%_8T7n?u9EgT4%6h~HB4gN2 zyjdHK`vdU^Hra^~x;I@-+ku?orL^raoE-v2{XztX_Umki*h>_r=n5OO6UDdIMzRj#m6m) zxSJ4#1IuWR#i3+~6z)UO&cmqMm5jBPMS6 zh?3s>VEu07)KVQfFBQ$}_hJ;~d&b%&!}fq03NK1Q@{$6|h_!Q^YMO%5GaW`zh#i~mwwM%6|2t4 z;@V?nSpHL@8QPO!L$z7y2?d-YhLAzM62800p)5!m%f;nk=q`g=+ew&39^x%WWYGMP z>Q1V~CODFBlsuOUPRgP5pcGtxC}D1;G}@;qV(}woh}FuVn!Y6b<;V{%4Y`F<&`6ZV zGwR)Z4Oc+#Qzi0pDIj~)1Y|6mglSev5Y(6i)e;G`E6E~st|V}D9QLdb$7A`4SaeB} zbbe`kI5!b%Z%%^lW*OWumWEyMM2wwJzTPJ~=zgb)6+QB>Od-FfvI=p|Wue(hp7;_) zwB1z3;w%|#NTY0SjxrLHq>;%bG0uJxQgS9>){${I?<|hd$0w4$GYJz`%G3RKBG!zR zz_@@3aEOsYsF)N?)=i|jJBq|#Q$SsU0z%HIQl5Acviqg+Xs0~-Uz3M*j|?2T#=)SA z{NwjDpfPJIMDo;d;PNyikUw;=N(qWa(m3p-fK@Xjq3tP+VT1DcN3~aBGWF#wsU|-| z-szH0EYYl&ZEXC&guGhWT$4`Lz3&rKis@w21)eaAmrYD#?=yDDq?V29u4aE6+t~QR zHuih*3wC$^Yt~oL#xkVdFw4u&nU6~+JM-fcOEY=T-ca$g>&**xG2|svOZdcos`s)T zCLdVssL!l`vPNUJzGJy=-7IaOlf}q(GQWgAHp}xXYY_X)JPls4LkS)1!j%_ncvBM# z$f{?>*PgQ_8^2&awcX5hZg-vYAk6QMr zzKM;SP{r=qRIw(z8dkCE6I-g<#(HBv5f*bot z19-oX_m5*hhBz(?=#X_@4OUbdIAxvgvWZ*aiRJ%&B33%@{qzA|wB@n+t|m z+smJ9aHxwZPkPTD6K8wauF3clLfWJ)%>pYhLd`i{Y_pq=HHA9ZF+l4<(qlvYr$cy# zAuc-XrO=Yy0fdin?1(kOm6b-ANOKV~qV>=jI}4)1hA4f@p($vJNDX~BPSrXDy99oH_aV5)*ViZW;|*ghGzbrcbnsz_XP zd1SS!KrlxWem3%Oq4{<(>Kw|mH+T=_*Eq*|CD#eO=j+0Z(3P>I%f+SoH}DvD(& zE3u2_zGW%mPm&78X3}+>tc-uFRdFLq6RV~eK;S&^gScAbK59YQ)Ce}i3^4IOBg)0h zKt#6|7E9@)%}j${Ylx&E9k`yH3CnGI*s3=do0^v5tJF-~cbJF0eRCmkh30N-Wf(b} z&e=LMyqh;0d)Lo_W1b1-dU5KPF2{~_7D(%|fY~-PbQ>*&`gTM3)R@qW)Cqik9nIg7 z8NrKmhH*6=AwEx2ke~4u;NqJE`0uK}$lCjl?yW=c?;1e*EdefZ^bh#W-%zy_;dU*5 z;Ar;?f9CYzp9^_tmy^%-`ga_C-jAGxgJ?AV2dh2-o^(ruk1hR!fHGlz%lI2UFBjm0 zK7)|1A40+RKQLS^$fKYB#{B#Av*X{9y_vkmw!a|xVF*3G0({n>2pld~KenpAY7u>t`9zug%n4tX)=O4U7 zPx*Ireff^_O}}Yw+yD;G`HNdFgD_nF7sJwr@mp2H_~MQce8kfceA5aszVh>M{>Eh_ z_jVi3gO`f%tKWtA@FfELVx9mGUQDy*Y=-fXiXuFB{%~$7H=3^$5alUN#MB%Z&4;xO z>$J&kNXHSmcSI&&(3Qoejfmq%K+9KTQhY(+pKZ3828^)c! zjpY5pqxj$cF?^e^IDa)tl7~E#;*J^$+<%o4-+4rdPYF@sQ$-bd&;>%WoooKsmLCG!7Xi3)ZwD@= zlMiyU2PC>ZVSa8is$`u}+_DZYk{vLW`nNYciGOrsBNps-LGYdpSY*2a&JWjPf}AI^ z=F)qJ_`|Hv3xBAl_1#Hptp!^!CB+xlYPX|PVjC_k^Mrk&H^xnH#PS6$I5FsekZ+sd zSRDYn-cZV>ZNj22zUa2~p!K&qGUs_C%F_|!r+Fan*$(39dBQQp2gze;_Q(@D#x_rA z!wV<=+l04H-c(2W;QN-<)a!M`A{Ac@)wz@AxDJ=F26wuhk^jmTwaeEd)!GwtjGW1< z<%hRc0Z>v2#kgehCcF+LKbZ%XXRXDHceYTxXN%iY?D6mF259VEPxXQm?)+Si_Fb#d zbKU~k)vMsR!y3zL?O=3|m{NOfpf-9ftY6xq@xU6)-f9Q;h3gQ0cq5X#)PcV()z^_I+U|M=dlsDMtR}4`6j4ayQ5G4C_Yf`u*NG5eI0voyOrj;UEYm%`ukC< zvL7c$ro&TlAC~>xi+Z~>zm&fZzmNbu;s5`A3AoppfIx*r>NirR zTq*$%Vkr+Vo`A1vyAXdW8B)vtLwj60t~wrvop>hluAjiB+~fGwn}NKX(}=G)hI?x> zuwZXGoUb3m=&J`|f_*6WBAxNn5bC%N3rN+_C+UA7-fq zqD(syzxGEVVSqC9o>7oYi^NYBOYgZ08V-rL`CvC@MDIlUrC12wh(XKTaF{+P&r4b` zl&T{zq_+!Cr1#}w*Nk~AKMGX{9OnjU(gbp zSbTdCh23i+i3cAKWAd2G*vFv5Bo-6RTi{<9I3t&hVcJ@3d(+x%LL7EI zPQdb+T%~@uD#uJ3~?t5>4!<@z($M34r++u@q z+}>^i@0K1o*~Wij!E-fH>CKgk&-;&`2_BtI)P`^PU1(yWq9Qx zDc&}2JXcQ~$4gtr@a-4G`ICdeKljn|@XtiPQ+*Q8bCTn)s^s~-+e`(H@#VJjqfF}<$*=p>#IFp0-$NpsVyvh;kH=T;-= z7@^2R9ux2R%|u>OKatxtNb>q*NuKdZii;M@^7sTf-jO5AcM$jf>QM!L3%d-Pz`Kre< z{G^s7zdKWsfA5v&KI0Vm;%~(1hZ=V^QRQ})>U?9q3ZLSo%til`<#ogC_=($#`22yn z{A}86K89=ar}kQWxv>r}EYaWsVjA29^8Ecdn$cY($GyUpc>egweB6Fr9;~9n&vt9@ zy#iXivv&s9`l!X*CQjpWqo#1J`)Yj8TMZuXt4-%imp{C!%lmZoczXPF-f&_D@A{8( zkF_TJu{-CI*JkoZ#H$y%V9YgdQ5z#vhuf=Y^5NsOc}|2TSBlW$io109@qN?yDcY{> zEFG>{K7+sX)8SX`b@;=>Iy^zvh>MvU@cqMd`RhSIEGx)L?O&-vx#kE~DxV6$Wu62;+4@}eGc11KZyHK6$uh-xz zpEY^l$SJ&ET7zF|*5b41ch&Z5a-%B-e3F_G|8ZE4NB%J4y26Wih4XwqxY>fQ6x_h= zer@Da8eDi#jRRkM#G6-2ZsJLIU3l7iUtaXimv6k{&9_K+^O9&+{#DzCe;u^p-rp?v zBTW;2LBxo6*%vGA(M%+n!E-x6v`I`?W{Q1u1Jh{+{zuaoi9|ZgI zWwZVGlT>$Z_aunhM~3piUje-GKqx5UkL+xiFfBG!UKhg7|I+L;!nV5X|0!EHL z4YAZyP+XaTF&(FAM(IUp$eqE8rgZ%4Jc$Vz$51JL3Q6oVF?Ww+1+Kv0au)SUFX8sJ z90c?g;7LIqVyy4tgv~7&HDzPbf{Up7lY!8AC$ZJ+82NQ+?rG>={2h}Ft@32ZS|#B7 zsJ%$9JBVJzGpJ8JiMJkxn9J* zhHJ#uyM{RvZV?mg7A~C3Lge{tSh4mRTCK8Cn{yYl9^SyA6IXDiBpVhPmtncn`BVnY++vJqE=AO{B0L{nin^!@m^D1aPscKd zDV5`%Z3)s;DriPZDY_~v@jUYhB8J!Fg=jrGmeir__*0Z6JwccKb37}kgUg6|Y~R&H zJ*ozL(r?H1^mq7C)q%YFPW*9yjun$z;YB`}&j%W?=*&|%XH~++{xQNn-p9{_MRV ztwDDmotypT*i~7LA?F$>uX&32Nj3PRQ;mo_Wl+CZN!my?ymYD%lm7@+?$zYqFM-#E z2MF=2#F>Ipd><;u$LaU?vHSO9=v^s>uwM~!=9gg1tH&^&_!yB|_wZKxG2VIJ z#_`sCyeqg1H}!nVd*4NlR{=Z~?!&pN5UVHN!R)Jdu;p?d9{A+orT2B{9lwc3IXU=! z;5G{A^Q{nh9b%$6Nb9{0xieYVX?K%6sM+LMxK7-!D>y*c#FG}PX(O&188u z?;V;~o(<Ps~J%G;EyXX?V2ZNZ0lwT{v&DDj(vUz~Yh+vzznq(k>5240F*+|6gjC zji&?I*f${uLsqwt(sKjXr)OhJ%UwE8cM!Kc7frkJ@Fp@35yrQWOWRh@&qr+W2MA|( zAV=s8Y(~F^M{x(I*_o^&w^1Z>(z`#3L~Q9!lCre)ccST=|uJBkyq7 zrUf}$D8uLf44Nw|!3BzNY(x>}(|yJ`hVDsMAE7X{0Cqyebxg5%#>@Lz8$np~RkXV@!LR5rosc?dC)PQMle z4m9D?ntB8;X~s-}CWtMshgek|dfwI|d_psl%$w0sR*&IBO?Z5#9!1aVu(0+SDpa3B zPv!{%g6Udrt3>vKI^3iC&)z#P@m#wT6SKRa{Hq7Q_H|)<^a~V>Z-G4BoAQI(P!vaV zj^9^Me)$R3cve8Ys~meK6=O?9DNOxp@kg&2^%+g*p868bT5m9L_boCFIN`l zf>%dgpt7_RKkLaWNc$@C{T1Bjy}+m5PCUtYO|!&aqSCJoOU^&Xmg-hyMb<+ivj*I( z34?p9alp17()2f4+fohT*a|!vEQ73hH9T#s@#FAvR4iQv^976WLCzeq&X!pA!Wv4l zwvZoZMT{BJE{?Cm+R3(PSmucRr!27Y&jze`;RHjywb&MBi%X@p*iIQawXfDV(X|fB z8Z@V^a}D-IY#>hHM!b|<58KTSxO~nD<16TAqnxm8xCerQ-Ed6X8|Rj|La}or0x#Jj zR+eV0cH2N%(Vpfl+oIy1IeMg4pzqQOl;x9GnzpYx&Kws~RuDgSCHXR};i^b8);_Ex zR>UG4Uc3r7PuW7p!3O8DY#_XG1#)GVqcVFXR5fiON#62~gcW#|V}&N?HCTSf2I>o! z<6-=2gt=Q_RizzXmblWa6DtH+TjN2WHSXB1hMVg$6bUXT@5@Sj8D)Wu$%`P;V2$-h zJu&|22IP*RJfxsAO5p&RnJ$Dwar2eGODIpQpR zPjH6^UPrjV#@QFcvmIcuX(cKX>@n)9IaCx^;B|#1&Uf0Q*x3e2Q=M_Hm>3xm>xhx$ z0t*jkuv;$JD&q;^FjoY;cg9y?5BLd@eze2~(?V(O8sdy)UhAcyNDUcS23v*wM^9U8O>-a zW5U;-uwRAMEIRTL(;i#EoUfI#+u_x0VD~c?Q(wn6$~UrV@oFX`SHV<+%h>Lvb*yhy zE!(`IiIu6>lfRUDDca2}y!{#bs$9)3R}#mqxS5qxFC@wN3tOx4f*HU}DbrWU&bn4G^#`>~M6#T1xKqi_g*{{eR(F`jsfX;>?mMic<`yeayv;Vv&SB#A z4;fqv*tO$z)K7ZKmd7@tc3}q~*kYlAt=I*MpmdiCc7VC<7DW){jPLwBKabbNwUG_WwI*}U^}b__ z=f24)vYOflSI~y>C3IzU9=YzjMyuYG&@p~4q<)Qz3`;0uY##M*&7{gZ0%+uwxXGA4Q6&I3Cb~(wPDWvN!>ZomEBQ<}mqU{;Q)HJA=#F*>kGoYRY z8MQQGQa#iBF|PzGlS$e{gUb;B|B+6 zij>BIJF+O?tmpDG^4#lDMEB3~=zpC1e9_Xlyz?(98Y|*>mI^{gc4wY^H&`F&hI3lo z5ggP9yB@J0_-c1FQV*<|!v2~g?%3Sy3L{M)9DBlZEu3kHd@Ue_1nB*=#t|7y44q<( z)o(2EWRp3DbH8cxSPNJgn_>2T?h}sW?4}p{;8g@<);J^4%^9EjI&lw%dt7BA?0X1M z8fuBov4n>nmbjv1je__7j+($87@`%M;um37EhPvf5E5*xbg@!s0{vyv0ozD_+`MD*wjyC7@xDmE-Kk#;@1yWbp;MX$t*xVHHZ?gyc zH$?A$G{Ft2{WLWIVEF&wQ4*^8b6wAu{>K^XoIU7R_NASz%^A7 z5$<-}%i?**Ar_G1Tuco2631zdLRVIPw+SpSo3T&a6pJ&gaNpblg10qN|5zbHkLL`P&7o+=Gm)XA zu{PQm7gdJi!9XL}TAAVgT|3yg5LEBjLUE)6Dpc&4jm!@ESi;A1VL0(D3}<~q(IdnM z_qsV?a*8b?{am2e;tG!ijxgc(-SGp1c%K{yr~3c7>>UyQ#(~#IfrvErM8q0L^xy9Q z$+5Bc;~NIW3G8A&-~;F39w`6pja31Tc>C3hv$=l!-pg5gP5!qAjzvMXKh7WJT$#QX z^!7Vr+HqGbyXAs0|GY4OzyD%~BYH&%uqb!LS98wOsJbxw+7|wEJ#b0N0Y3TGsETpJ z1ubt>-t+<0c_ZenJ1V#Mz^GdUj4%0P37?0Xw~2V^>5ZOc9$2jI30*^fAb2dMDEPqo zp)1yI^}vkj-gq^E^IPZaaPB`J)LMDr+7uV&88~v*#R)a}9tdsnMcg6Ilo99k2Rb9{ zm_75KJlTQkg*Ed$5S8uvANa)i z!m*nVX3zFVs9g~ESp5<6?|&IeZ-@$Bh>drKT7);0)&`;_E)d$u{!p|T3n%??*eN{@ zC0G2hB|j9Q%v7koTu+(CcPU`RJvwvqKH0|IBURsLKboRm%`h2O8 z=6Y69nrRJnNS9KUSVRwG`29QZCYA5MMh6%NRyA47yh*c^ z8Yr=|o^}l>ry-IOO4?jSfBW90`j|U(_}6XDd)y$)+-e#YSwrbps_Aw2d(?FM2IW-U zB4dq5w8!=~DcCiVWyL+pcDq5RZrvfBDK#{i*C=xa-=ZbFMi|n)f%@h&kk6%3I%H8o z2P-N_d29nM7*$J;?%$;G1-B{kDX)#>>#4lrHpTRNM0>CDy6SEnO~|hxZKYeZGx8>7 zth`Oy(pB`;uAJm%RnYVGmDFo~9sPB@OY-v{ljANvrep7uo9#X7AJ9Yx%bw6l-xfL* z|B!C9wa|5wCnP*-CfCu==*6Q)^riJVy}kLI9`5~*HVu15EeGDy(krj2l77*Fr9bHK zR^~8$_(JdRy{G%jcO!PtP9*Yt-e2zJP(o;oCTxO>B!}t zY<7>u>_=FFsmEsHmeL|T3tWZ`>$oSeaVdT#FU0MrWl)P>4#$`T$oyd6hs_!^O4i{x z_XA(WZibrOHr|_UKyFA1Ms~3aM3dbhzY=joZZkZcH{v++#&vpag?7Sv1lO-)kJmcb z+9knw%3AymOU9zib=R2|)?agT@(%6GpPmW_#$q|(1?Z?VTnOMpgmcqa_M??bZhF6cGwLOW+frcBrmnTh+cD&-(TR_^Dy+u{TSxf-96P@SCu#|3M#cp>L|N)mDVaWYD!c`jT(5wrg#A+RtJ zEj;`Ec)?=!LMGvk#cDkI#4eWIeEvL7!d%Zpyq~y&dFP3stWOtS0hCx5$5MtA&ci7otO!=Txl&d*RA0^2y=hKxkvVQ4Za^vW>(`G zTsyM{l{tyHI&A|~_c1>yGZo+BHX-Iz5(aNxi#i|nN@^yfy6-w%Hc!P#?KLRVNyMbb ztFVXjMiUHIBY@A@B@0(0xBF_;lqNyhW)siyY{Z1m{M`68WEE~kP7ym(x^ID#cPb43 zY(QOlDs#*e*$I`3{eQR@T(=6domV4jRU(?A*I@dA71(iX9V{pCIi1Ftsl;TQ_>&0N zk=$F9=Dy<2bvVg9_?o&zJmLOjG|$nlO7Hb{9yX&lz@=$D zX1~~pf-fns>azu@CTU2^+66depx9^!I?^_y+x$(aaH@krVig?j)XxsHJu`?aHv42XBm#}DnW4abqrWogp<;R zSn{bH`wU7kg|0ELsfg#JiZI@!9Cj;pp7hPegM=~2LQ+}BXBwH!M6S8-SUI{xh7 zpSxMaIqvIxonL`|c`lsS=41DatIYqpf*O}x*qzVEgQV-Id|L#SM@7u2sDYka8CLhK z#+^5nXgpDda;ai=U*<5!wScq6*KzAhJ{Hfq4!!mKa}5Q|O)5l1S24Wxim`?nQBS@U zAZ=4Euf@vnjvphfUp4xztB3KSD$L5LMwD?a29=j1=>2t=^ejQI6W3s8eVwz+B~Ym= zz{C7}{8>|i!D9=!R+J0beL2{B@Cq7-7h=?bVm@AlcyPQN!@2fSK$ZBlyB@3Ko}$VA zK7L)l50iV$(^}kw<*m1IxTX>_?w9cTs}QjhOR#5BJ@&bCecH8w`wun9<^R@isD_Kt z4d@QLg@PxyQ8D=z9G7#wXIVY&XjU@E^9Inr0)H1(LSb(y7S|O*b7l!zUsqxy>hSza z8M;lV$13gt*zU~c^;Ir@1)k@u{bjHn4i+ZaIFxn~1=c4qv@8dHE;(4c{yYxezJNs; zmocL6H7vWB2cw2d*s<{x#wlM#fO{^C(%DHd`UI8-9K$B=lXz+z#DL?+_&#z3liCl% zdd&$~habhmydyYp{W#9eI11lq+%vjx9t#$oL&4ayI6wIUR=ZzB<*u_h_2VoS4!Vfn z5f||+A{!O@=dk$F8C=t6mmpJ`4Nb1GnhngxuveFm^X3%3@b`yQb6s2hJl37Q z1oh%8I5P7x^1`k$w>OWu4cVyS8uLo|%Q$`MGIlCoghx;=blR_C=$&kgTAPCt^O;E& zco{dF&tuZRTuAm^LBp&(bmm-x`S|M?YE%F*J`ZkB^DrkPA7z)go*sJ{bG33%`!*YO z7CFf6dl6wl{QHnCfM!-HUhiNoT5<{c+$ca(k5aT+mEyqY0_LEVU{FaRatD|~tCSF~ zW{o@bc9=R;f=!o2^uJ((=SR(vkRo8vIzZ~ZDcqe*aYxDoefRLZ;~@vUNwGu38!L>+ zGUi!-OZ>bvn)5Vf*m#04r^}cb)8@<-X78py&zJ5rMeUB!=sA@gookrNpfMENds(9A zJOdPm86#$(Em9IB*rQ;EwQly9ac?x*3P-?xqYzep!F3`$U%}t!XvPe63%>UVsPAcs@8eDIdAljjsF>hw&1fWbx5hOq zg7y&!mNKts3eQvjH`Ew~&PLcbYXqz}j>MsU>|&ZlkY3B|fe+^F6|zFw7TyOR=h@ZO z=7@+lLd;wdzw87Y7~_cMB1_0r8ABt^lv(w|5m+}I*6Yo%W|9S*Cyd1DJHv4!a0LE! z3_*9Ex2?%H!TIH;P%;+K{?-~B`j}(NeC8p%CRiRYfoqu=PJ~X9pa(~x?okBzL-ksUg*GHgDJrw7DjD@^D&#Ju)MvwlXxOzVVbJj*8;c+nRABVt$ zGdz-8?ikFqPnlQ`#AJIwzMmg2z&j*vFyNje0V<=%O?fEa;+V9 z9d*XB0&j%YxF9Uf0kid8@Xx{)ul(FFi}P{2_Sm3uuz(%otNiwblf? zVn<@8wgKW=N5Dm&*%ij7$UR|<*{XzOISJx_o1l9?Ygjb`NvSq4-tCN*U-rl^b3n^8 zN4R`v?#4++)NmalV5&Q$C);9Sr4@?b5wgzn?`eoRLYJ7MG+zRXrKV7Bvc}u(0^gEx3b0DdlEHr_@b*!~f5lg~k) zNj_-N}$}X7xopXq8gv{vzh94%Rezsz- zvOCY_1#+Izjr*|dsa@pEE+i3?{T(np+8dIHV6MUW4q%%(c^#Sih@OrBYV!qkq^$=_H%{0g2G+W4CU{CH< zGg$RAMb!&}%peg73vIEY&;bQ!-M9zq0()ko{99y>lGy?-dP`tdWrBZ(rpP#D$!u!@ z|Be8X5KEY6O5oean0?2FQ0;Gu0eY4=S8ELQEykGa%lA!_Q843k=)gl0yoniwf=i?D zbg>z>oaDOF4GETH6IMCfp*PQJ|GRIG!YpgpI!dtPvpw3^`{3#uU(E001)m=d%4_ICZz{PW(JXhe&*;C&CKX66!L>E{U_;NnY2W6JbJWLMZ zI#Vb%?+Zm1&!`%m2}8QlYjRI|LLoixQ`-I)v~%e*cELQPHqK=n?e>^r+TPK;$`)GD z=OvwAa+hS;QzEbQl;*#0rtZt`(W}SRv~y5B`Te^|jfNGJCjW%Wa%(B*WDR*VG}3h6 zQo8EWKu-EKbfln)?1tT?v2k}u;rBy&k$eCDI!WxGKBPM>A8FRdhh%yEC5_wigu<>h z()!EQw0dtTy?j(oI8sMFm))R?UvJRdceV7OwUHthJfWMCJG6JfElLS&qFX;}$oy;* zwO_eMW4APs-hs!IB;Q08T3q-4TSQ{vO&ap2icB8gre%ko(9h$yDWdubwL~`4;5pA| zt@}5+Iq4&<==+_b=e?)VKVOhdz*l;{;x$DjHGJ$Xb8G7ssn&Lf)ju!TJP{YQTc?vefR`(%;J8IAwfHq_T%=4U1k; zV2{`29ru=wmOi6nov)~Aqq*ojeYALa^-%HXs=?y%VFqIFLxaU44m-pt8;Emm4-->T zhKXgm!^EdWBgEm;&BRBGjl`%HL7e4aEjm;R;;5UV__Ekuv~95xQ{LE$r+$+d5p5#A z^b$p*AvWSmLrZbtR4XyF(L%KCZz=AYY$Yb8I*J*OYQKn#96LUghlB9@*WE}jn(1#z9>sEVXQ>_B}3wLV~M!+ zt0<}lO2jvh?8S@=w&E6T7cpy(qxg7_yBHAUEZ+O>_5y~jO=m}=UbAvY`CLn6zd{p?h!=a&4PHR*-l)Q zZZ96aYAwG0X(f&}wG{()Scp@c%*2TaJP+4pBIf59iDIIKm}P4%{wTK*4J;+%;{suC{cC$5;IBoHT94-6|(!?%TEnK{+ z$@SoFNI0yG!Y(Z+AJf69BrSBy(LsibHr(E+V*>YRHfyQk;F)fCJ53qu-zsC?K^3Hg zsN&LKWehu@g7NRA(Z@_41GcDRSBeZ)JypaDD^+YZk>_=$2G+LA!LU#cZgJ{xy`jc4 zM4HGiSAo?>B^=?Nl{i2iCUJ5oEm6Xp5weh7pa7pl6@-;3KfbHCmsf|{}0W(Et z-B5&(tcb`IO*EFWza~-@lNYMP=9?xC+*QMWCp9s=zcy;0YGB$nW|F^Bg-WLyOjFgM zWx$WQL@=`P&P_LbIq3zvQa6OJ_eG0|D?8XYi(WZ~ogn^*lJUWRi^d>>#-S@AlDQ0%@%>6P zTHg6!J+JfEEBj-i{5V|SXOE34UidI~3~D?*(R{=M!BL#Up6Z9#P2R}-m6>5%fYj$&%Fni%0VEM}yg0wvbalgT6lOKBlIHk8vAZ9p ziNWZ4=Er-Yafsg&0FN;Nu*-@-_eN$JF84=MZY26AOhsdREb?~;z)u*5ao=?A^T;OKUc9Eo37!7NwylwqkZdGVTrBfXQRl^IDZTIDgmS zK4-a~Pu_%}qAf5UmJ6ZlojW&2}{H*p2%ecA#C{ z1=T)31%VJbELv z(>KB9*H&!w-GsD%>miw#h^*nsyiZw;wEYR#%em}{a}#l%>yY=}a>o0{DyU%%oGsaH z&Y6Mr1*>o`W-U^rIe)_SLZ{=a;klS|;vPv@`aBtDmZjjb53?sGCG&c2J!-C{^4^SH z(EB&=9NPxy?M+3JJbS~2ZilPEW}G^=nOzlWhzQ?_NrpR^ua}PWs4XaS+lGSHbWHoQ z9lf`0hl3-#tCiV%rmzk@eyztEcKxZJOvRYHi8y$f+4euzGedj>J~(cI>BMb#e}6L` zG(1$6N(+_SUBJyR-MVfl*RiHvMK|&Uu0la%y!Jv-hlx(cOyM*7k(lG z)PUD9J3l;1mfN$FKpuL zcf`5?_5yg~Xo?TUopQsYM^3ojpV#0$9k`$70)`yTbglA6oi{;nqB7=w*0djgL3BCh_$r@q$`U-oGsJ z#ojI-r18Gor_>oS=a@_QmB3>s=)s=b)$1&AVva5LL=ocp8^iIT3-|dXn50Gsbuq=Z z7nW%J7o3`7dwzmn{O zpjQxr1@V5&Mf63?(=pJ#I0j43je*PG36MTM7Ph~*uNo1Gxt&3nwlxT+SMk1&IT)ud zj7LM?P<(NkfP?>{pr0O!0c|0;z&+dMUSpu&;RN$)KiEF@!l9EMXx!+={47sAxyQ%Z zGXT35%3X}K;kbd zjZ{Rh^HRv})k+gQq?j%DhC-izrb!n+Q(f>klHK!>LJj{TvrNucPy0qWhCgUs`Y%$q z;u#2j?>uhyj}}M&rRj<9Nat}o?G5}xJ*?#5z34Xuk5#~jNuA`nT8X)c{9gB73El}x z7`a@TYdBg6AJ+qiWxK;%ULE<>J@K@i`Pdx_*tbFfpI6KC{kwzQwn;<0sffD)9dz8X zldc8)rmNx1<$nK{jC%bb1D>a-mF*(?yzkUQ?hj2K^_!M^eWsC$-$`ys7qzbZNtdpC zpvAtQDW{P0-AyfYGq8mgNFLJ97x!r|&rHcCwNTdiPh>UwHF;Qmp`Z~TDZlJJ8S`vQ z--^E!tIh23sxG>A?X!$9)jq z*$o3ES}<6v1)mq1&>5|cyQbZ7x11UdOT;*1F2m-aD8%5#D?`j z!PB1j+NTf9RQh4VkRHrt(#GHMJ>d682VWa?nE#{8Gwi*YJ<}UU4)s8*doL9F^hAWE z4*Q^cVC(O0>~>-%gI7<~{pk%y>E1YFtc4?%J+Zc~H=bYYhl68=pnp(*I7tt{r%HVo z4C#mDdR=T@t&cvE0oY?P0C|%JAob)xJm0Q|(?)$!GC>ywU45}}d0z}0JP-x{4DeWa z5PQQ1;$tbh=4AA6K-vJ6vO_V(VmLC4^bvBH9bYsUGv)L#Ies{Es&&vkeh{+F^s(BK z85XvBIM~(?|FIX(xBoC)7_ZM92|XPD(ig4WdSc_59`JnD16g*xU?tNF_HX*()EFI{ ztmua~mh7_IFaY_5dI$*DL)yfFy!O?@kgtOfv6laTV?UhkqlcK_-aLyk00}vLQB>6v zzpm)ua3)=)#pY#)wkd>jrx{vXR~08|#}WB1MmIyj`3{)Ls&NbLd|Rai`^ z^QvjysYVL+ucy0j8|aW&O9RY`C}?;QeQ7DAu>Y=8SJgEt?{$qHw({2l%gCTOpUUoW zhuN);?iUskow`gvwTs9|qmX3Ym6DCiHPV=wPd_{gNT(y8OwCGpA6!XlH)`pEQ3cK2 zRYAw5)Kas0HTh}P(W|&p+Ht*%Y;;OU+NGQ<9Bb)@VijGms;2oxb(C6HOIu2p9W zHJ+)bQR^${Y(3AKE?7L3? zeJ>#22ldoXuZVma%IW!pLOOk*o~(3l)8pxPX}d-vrGBm=Poo-|*`u74@0OFGRzqIQ zcIfx;5tSw1rOiA`I60z;+)mw~?T4%AY+XIw_;ZuqJibBl_ioYfJ@u3kQb#$S-N~dh+!=C752Ns=8bf7Zy-T9C1-Dt z>E0R|bLu`_UVWe1Hs7YD+g{KNo-wrO`+yE!xJ7R?o2aw6j@-?w>EHBfdeVNK7LF{S zDS0{6yzDaFo|i{gQAjJ!l+k>i`Aqk0potf5QsfJJoO%Q_aP3{u%MynIMzJeqI1VxE zEZkfdIZeZn?KBv-qent1V+bzj55l^MBayz<97|hF*dIQcIja_^%HWScQz*6?WARiY zT+CyR?~Bn$8O)A*o@H{)9}e7R_avY9-M;8!=7C`-Yi9=F1|y_BVppLs8lI<_mv_k& zUR>KQYqUTPk1+O%vVc<#*GjLLW3rZj-C8D?e#QjJ{~1Eh#{`BexE{wnhuXt{m8%U# z*%7lPx!x>iiXU8Kvyw=7EoOogHAl1pcwB7?sdNkUZ#6<&?MT?pAI;pSQP}w19Pz)5 zu_S6FGm=cXKVpcf%vH&pW(dO^Q`9vL#^GFjG`N^!Y@Hd#GJkDp#Awuanc*1E#tj}Y z0t%HQaq^QszQ?l*vInyeW7$=?dkFW>jquaV2q`bj5UD>LdhEU&zmktzrv-WqGXYtd zb8pxj>OuDCaIwXS46Z9{irAQHj)UDS&}~2WM1sws^~(r{epy2|+8#bHnd|Tp>{fNe z?n*m!9CpG67aN36cf!Fb_SmImgKGIM^!}44eSuxrY4e(su=uGQM6W27Dx3{6f=*@h-Z5%h|?V9M5FyOqHtbL%*m8t zo`aP5CPiBOr28ASS--IEwv@Q5rvtt8&4WhWbR6|VmJjcn@ROzIDe3IB|}R$rJ`{SM!be??eJ z8`s`Da9HUtvU_&nvZ|E$QA1WNYLOPZ#&zNBkWREMkP=&KJE30n7t8#l#hHVclTiE% zS`i=d&hjIoY=59>avL1Y{@~=1Kaig!C9ZUo5$8Sm3*{|Ud>eyjZZ3zR?8K| zJSd6dbfsMAm;T-=Dnf&-&P4Vq?WiffPia2YRrnv5+^im^A;#IW(*MDwZYVt%QH_-(0%xO|(2IPho>QGKM2XzI~hd>^bO#yj*8 zbzcn-NA?&d4(n?u-qai-c4iC`)#QhZM~p^^PH#txXI)Ih&}d`P+|N+-s2(lqOPPw- zh8v1Ho45{kXsGC>!=5x|5S?10jTxsk5bvbHa}ZiEYE^^HSb4~Zs_1n=6-$!j@Z*jm z-Z#r5^Smso2XM_{n=IyjR!3EvJc+MmI!=Whw~El_+1_LKRgl56vA+)VWc{WlWLh;~9i@e~-KxkJ zmGSYBGVb0`!M#9Dgp60h22WL_t<%J-m;Cj=T-(alKzXVrmNsf~?m!Fs_;`Mk(?Yth z4z!mjVBt(EM? zXHKTF5t8D~5HrUT#ysyk+0YO(H*pV4#~dMw2FU5KKqfO^4f2O!Y#Yz*<}hb!!XRkL z>SD0ZaORU4;eDn7EEESobJ{5A{m{n-eSN$c%r4=?p?EW42txOcz$07sk@OypaqK_o zK4b*mwhe`aPapWJ=|U-g2ozWLhs@0WxaQIedx{34UZpoqCh5ZVSYOmx_l4421Dp&R zghSOsk#}PlhWIcmy}JQy@=VZcvne8`n<9I*A;#wpf{cb9bk_|)SC~G#VY#OGsSi6g zMjXN-R#!dmh3D1KraDm2D7YOg!-#o$+XYMf%=p6sNhZ(+gcf#OsHzb+xEO)FU zJm1?OZl^o`izoEEC_wj~0j8>|;h2sZR`1b-ahMuPa<$p9r4FgkZU~80gO0Zn{$|PG zLa8P@%ep~%sxt0bs=+8y8J{03BYL|W8kfpq`i=L|T zC|IWqXHlJ5XsTGcLJ=8r6ftv&9Bj_F)99#9x-z?i-p-PPbgBY8Ua`aIh63KZ{v|EF z4%+eIE4}B({TeHUW-~c_`YnUbFEU7wmPcfw63mLFv3s2~*CUnD!&@CMmDI7UQ5$a; zYI0^#8>ZdVpzul#jvZ?7j#kGL0~KhSt71x)7IxQY;*g9wRxi@Pv&q`%AE$yVtumA7zzC6@< zFPGxi4Ht?PI0M~DeI2@JZ=Ni3))ioVnfK9LZ_ho!o2IR_eMl$8f0u*l zk~SI~t%wy*enV zgjs3Bw^#AvFze& zD8G>lUHv>v)hvg}i)siVl^CU0gTY-DnDG53EIF&UF}esY3k#4rvj|TYT*2tnY=kvm z!uRHLQ15pc4!v{G5OfJM-{#=6$91$g=c01OW%gKeepBTtnx*paY-|ejaAz z7GMVFo-@8)f$Hl#^gEV~K}zS)vFRix9LU1!Z^uyVc?|E&`9tRrV%Hr5?Y@MqMcH^h z58+MsCX7{>p7?HcmbE2vQd?F0{1I9d%p7o zKBS&N#*1wB^VRh&ab_E{uMxx#+nt2o5Rq2(cHuk`EC|8yBiQr8eFDMYqb30_AR@MDxB z#kvZ6X4IjdM>(Hww=hPz2EIJ6X5vx@#W7W=mMUXkV+|$_EQ4P`2~x9ebH1|@Bh$*^ zyPs!%c9lVTa1pa6`E~G2TS0sc3Nv^%sJsL(#!KTpJISJtcG400NN#eRTe0{}A@lxG zMuZGbsVJf-Nfsv`$RbWb0sVK&qG9ttvfkE3V|%i5bk#eGKK7OFIDDp2H(t_x`R~*i z)=nvHZ%EGgCtYb`uk1A`933Ese!sty?q@}e+%Apm;T<&STr1r^@r5?`{YIr#zbIL? zjqd-E#lxsD6WKcTp2YL09Ml;t^W|w@Vwm$Fa96RBr z)P0~^s$Z%1kJl7=<_8J!?Xrh2(m;Ou7 z<$lw6UO(;LBMsU6f5>2x6ka+>!GFC197b}@hO<(4r%9uy@PTe#;bUvvMwTV*WR&xp zdY|qf`;y;uLhlEyefNvT-uX@sT-s?7W|=O`(hbP>P!ZIxc-9bH1;(oRxxs z$#0r_^*5_d2{fL16tzaZyh-6lG9M6nJ!bW;B`h-T}Qw7&C zmqr0yA=ouM0_P5n$Ay52JUbl4^`uDrPMU<8e$27;48g%`A$V*&5u47$!ow{Z`F>IC z+~6E(^mIg}L}SCsSj_A(2|F9v8>JtN<5m$!xE_X>>To1~3By4X&Z6i<<6{BmPtU}_ z^2=m+eU8NgeqH09MsXg4bDBnzp!;VW9uEme-u*z%Yz3hBzhLI-hhUrE1hm(%gX%#v zLTX}g_DdY}nVB?Dej#S5&w{e`JVbn)gY<{9P_4WGA9ddBWfJ6n#$bg#d%fl_!%Kt3DDF28>S^yPpE0T>?^faZSzXle+Dy-XP1sEmiskx)eD zhT_g@e*OCXXlBlH+N4mVZX1jH9l;oLG6YN6+dkMX7{7BOV9)2kYpZZfE3g-8YKuE%aq4qiyC$@#c+GYZ}4suO< zPYjxRj>lCR%QNah>@M@evmfC&q7aU?rePRSHx9EVjKPfWfoS5~?5t@)=x_|gwb{Xt zoimyJo>MW~lR4KyB$%gx*wJaDhd;BqDQ#dxhAA|qOg3x?A0*OZBVb8Nk z&gnt;czryEay`Cwdl)+J%z)DGI8fnQiX^BX>VkjPysoLl7LfS2>gO!1FgvKpQi;`|<0k zzsBeA@`(sp9gfGt#$wCRAnb6Rh~nK-kgYcbHeIn8dx6iHF(DZKHx|prPr=FUQ!%|Z z7VYT zXY9?_F_YOk0qtj(;n$=%_*Bna;hFveyPd!MFpZ)`S(`p(6d;yCPV z;_Hx)ML~D=){Y2;)#^}WH-;d$nt%60#^ae)AZ)jc!NWs*UUSde%X19o4G2SFyB{`5 z1!3pCP&lUW^XSXhbjNN^bu}?XL`|fMGW|oAV5zNywU5%G*D-qeg0$DYyu&8_? zN`r7S+*y>!QB7$!uoa}oKx|^ z1Hl)CKi%+%;EEUT-SOZ*o)Ksb#9UQ> zID2~FY>gj|=#4?`j8NqE@r6=?FB&g%FHFY?r?p)XJI@}6m905vVTYYOdw85PuZ3lFgtqR48B@yVyb9BPyNeFlriis2X-rpREAx1v9e9jw#m4eXYawunc z!Vpq61t}+^p&TB7gUYG)X!F-O0LCt9-tvAxI}3SYSg7&rzOef;3>rKmKtzq6iH zwUpHVKK#98<)oyPl=yG|&!65>R#H)OqhqHnoMJpVYJQZlg_()O(n4ZtYUb)FNQM3X z{c-CZE^$0{UGm|wDP4(I7nbWx5Kffu7XH(?V%ewltYnHyf09)LX_=j%mr*=}_wgj@ z*T~?QOf$87sTI1$y3vsbTWC$;TH&U`IpI@foe=R-O2|1XP4W2?Dct6eE(N2+sH0Sj~!`}!P zGG9{5J7%Bq`q*3210m|p?19}b8Bp+A_&LRkrZ;#BF)OzS+43#IEr|j>e)?5#Y*nV8 zx26e^y(5HC39p4Y1DOjw{3fYQO{N+xBO2%tOI~~Y$am;hi)&}!O6py{3JL0`$ZS?B zsjgT>x5WK|iJbv?ABmx;nq1+%^(~>cN4uamI8Auc^;zhs38o7!x;WXs1A`9!fP?cu zx*76GXf0?IJ{Nk@kz23mE@$`$ynR81CQ_KqJ+U90fd`f9$UMOt2vz1>T=V>y}}d({b271_eNmoF@SRmVsk=jRCXG^SWk zN3P_NN}^z)e%c~*rJkf+&Ry6amrV}WyR&;^6yAFVu_sg;n?GdIe3e0B+QG?UN%bOe z>J?>)d16mWQ_3RS(w&x)xRaK9k4On8rRr$xsAFXPF`cB2R+=lC)>_=Eu9x%+J|o#X zD@CxDMbVJLFTE8R#+g=HQS+2sfH=l)Zm!zm)-3Iggz8)4AlIBbH*S65v`Q^0n z{3=@hZJTA=%wkE^_Y=a%sV0_Pmu_3yd{z>YE}W8-7>^M?tSuILo!lUdtNt$J-yT3K zUhL(Gh0{VnL%s0j?N*DVx|5Oy(<8!(`3{!WCVM0*X~%@CA5AQc?T1QIH_Z_yB(wfXRa8Up4A{;7*IKA0@d*gRsJ`L{L)HqFz>msQI!zmBc&Jj;uJ!So~3l zmi6W<{wV!hx|P0nHzr-v@4^%P9n?C_l;l6%5Ni8f7V4}o3JYG|6m})C##_#ju2$O9 z`46#p>beHytEOV|%=5@HKEWRIB%HEcEqI5x3E4H<1+R4vgs=9$h5jvil+d9;2UKqf zb1Y^EOYY8wx@Hvi$62F~yrJY&&u5maN-s;&gEgpi!cO6pnu##TrW+}GSkTnlfwZN0 z8ufoMnl!Gq3l>9ah0{&4l(<2e`v2!5gl*Ykp?5glVpz{s$;qF&mcNdiv{*JNngZ_| zQNX~5g510xlst9m6rRY zic7Dgh{Y+Z#JyMFNaEdJStiz+NM1Fbr-&)VG-ydC-F>ctw%fAUvicPzX!W2bISuL_ z*`3Z!n@bx;ZKsM4i)p8B01bLOnHF@Oq;(}{1+6)Egz%J4g4&ll%ZmvYEJE~JEN&Fv z6r3W}>D!bp(zWG?1 zHj;b8aY~(@LE&ULX+ee#85)C@}@ z+k}Txoy4|ZgzOUukzH_UPw(~93+%6-(;d|-L zya+*iY%j~=Z?;rrw8aD zWcLBuaweS~*vet|jVrXXQkNdKg;EdcH^QK&s|4#){-l1>n0j`4)21W7lxLVo3NEP> zG0Kqrw#+DqU4VPe^O?gQj#45BWFE(+VsUnFgz+FEVIxO`px&D zMF*TI_s~Eu?~>2Gno$31Po+3dy~gOII@- zX?@#GdYP9_Te>W0j`9J)bW6Fg+F6yBjLav)!yjnFpqK1r?1|MbT>A*POS@j}5od3` zAU2;a5R-~N3%a*cgzQOuh0{@eso7nD>yFBlbB5=wxb_?VmuE5hnvzCLp0IQDU?C`C z3^}>XrkD{csLbRwot}GuHoRR)A55nT1p_sNU(Z#A@n2$vA*rV%TH&UWdyjO;c5R9f zw%J5j{eBT;YAmAZiq@3A_npPHm~EE-)_;;bSw4@}KfWmx?HVn-tEds&U3N&e_4+C? zSeq!S^_(TD?D7&NetW6Y@ig_>b%i4CjH2zQdXf5HQ!2Vxe8JrhXE=y5dSe-EArmA2(cPI@>kkH-L$#myW zo*;csP6&Jf3-f=yDe-zg)($>efr#SfhB_YsBl+VB~?7R!z5P^j|++MyqD zQqZ5?o>Ad%+%>sDfEHhoKL`0So}_=lpYru8F88>wv7V5559iPjtmW_9ni_$jY9PzjafpRe@A5f^!k7{M?r zbv+(@UyK@4ZwzL#?AhpIHhqddNn8!Yw(ph@gey3mF^2I5HCz|nrIlVz^w86jmNiw- zf^8+F?OI6g(Q9dzw=w^AQJ%}nzK84>Yu>?du23iJ45Z?CzIqf zS1RAa=sQbfi{=bq12c2jAYD}~TxN^8ITP?9P(#Q~l+)Z#yJ+$B!xXr=n@yNCoY{@; zkN>`EqQF)aKHt?*-LIM)msHX(uR2;aW-Hx3KABDhokn@=HUyqb!{eCSwBXhsYEm79 z=+0oag`)=o^aE5r29Ui`IqE^m1^gMYp>k7m4E zO=Eq+=#N$r)j!T8C&WJOr~FXCpr74vD%hQr5BORPS@2s>V;D^4J{qJx>z|85I+{}q}pQiu40EE^e&*1-t5%Jp|b1DF0O?&=c>~&^$zlqx=G60O6f&|EEPQ{W`6b~i91}OA9)9e zwg06BBg7b)8whVlH+*cjM&6R)Fg`p4ad!q|-CjN99DGAdueZ|63z=l?U`kiFEXRT! zGce*#I9}Wx53BB(IQx1Eu0Otyha;XL=8GIZe0(`Zc8&v^C5Czq*cF2TY|~jA_C;L> zW?u+T_NED?#KOCwaVQN_G^aDh&*4&?jgb2@@z!n%^h$K#zUD8j>9P`i?Ch|+Ew3vY zF=#tfWnQ7+qapA0Jwy%fooMWAW6}y0Bfoty20wd)WaU5D7T1Yzhf5e9TZ-ISV-Zv* zfeVEyFzEk@{5=Zj&HU}u_f{(PefpV>>90i0sAx>yHx7B*6!?rm)_nDEH(@8Ri5H(t z5IS%caOs9OH2t9%k|n-yESpaK$DX9e52cVHzCmyEhTuWaHR!gdak`MkJ%ZQprWp;? z%tJ6hV1T^7sEdK#QlOkhDt#}Bcf(&&-6Id`f5sVV^F~1?G#Gtt?15#52+$e=U*lAA z(Um4M%Oj*4E{Wc=7b#V_mxd>eg!IEWOl)_+O81|%^v6itt9^$tPsSp2xF!x>AB_A7 z8z9}W4?%@HvGU|KbiZ!Hx1ZI}aV!_#tUWBgyWpTWZeoA3GioE}v;`RU;sCb0$76Yh zE)Ez7Ee2stym;$nt}Ww^kE*9ByJZ1g-ROoxhJ*0eRTX8Mgw9{-mx9OY2o@jM!xnW$ zGp~+QY^>)^w$9Iw$xCf#ev>Tl+%gEYI+NkxeV-I#<>C4BGf9STv-?@UQ1p7lBbI7@ znPxhiBkysZC`mj(ikocEW!;~9U!Nnk@p2U3^n&=FR2L#jo}*;P5j5!!LY&`t_*A$+ z{<{Pu^xG-u@N|JC^PPND`(WBNNhlWvQ1^CcT(3*SsMxPm|7#4yS%dMlF@Q#|38Zp^ zF_aQAm&PkjqmWd8YCbQb9dcvo@66fwIZ^`m8ZGhp%sLp`cG9w&XX%?wDdn{|p(A(* z68ucict#iDJNhGZnm#P7BvHKB13%3wg=p1dTCXz&hLQ!akiPF zuMnwn3kIftne~7#)TW(>;>s4pnQg+A*CVk;$UANtJ)3`gHiM7ZEy5<>nQ(kwiq-mx zLKo3AESA{|`-p#{PQ`GN$uLE2iWM&r`X&@d2lAtz43X`-8p9tJVZ!PH>^9wlU*Q|E zJD?b+{#K!TS`#KLsm7d)Z}5A164o-=NLj9mxWZ)W^EMm{yLRDGQX*!Rmy$y45ad4E z1eYIXym+80UlsNdXWGhnt;0bsow=IlR9s*oo?~gLUO#-7)TUoUwdnTiZ#4ehTe@K2 zhVyBjST}SePgy^Nmt+fB>#LIR=~qgPQ-)(e*F4g_ewual^`lE0)5!6?DZMNIO67~y zaALI*9AET>yQ4bpj;N+A%VsM3Gz0sM`rzBNYUcMklUC>TM`}02<@~--iT}cq2HG*N z=p&d^WP^V?f%M_1A8Q!;hFw|SOrNLSz(>Is;qCk#|1?gczxXY77~IC6VZP$M$3w)r zUxUSGQcS2m)SvED-lp!Si_y6v86zjXhT@AX&_`|T(|$pf4vW#XB?6ia_5vd=4BHwe z!EA#$+Wn^Baqw~6w3Ol(B}T*0-5qAO4tN^bho0=|X5mW<*`&)+Si9994~6Z+&*8dQ z+T28~qjM-!dF#kIN9{x#W2NYu$x-_4wwYxbJY!#8*OTk{d?vp{7teAgi!FT|#h28E ziLbxjg0KY_@J;&_w5LeIELIC4f1P2kwj66zlaRXWFBu*i0?ART*f6|`t}Q5HG2=VQ z-)t#)N35YbK&JLP5VLBPnV*GA8*!WDyMIfkDFG{EuBy@nx%|WyoXa9V~o(7C!qDam)Q6 zoFDH&WRM~9W(mIbIU{)FJtN-bJ`w8<%s^W4fB5r!Fk7;0CtX&QgSU+=&l7xUUdhw= z%9f4Pn=H(4{S~?9r~af=?Mg+5TBs*n7E@A`utMq$1^+E(TjpuA=pq%-$k&7TolE`s z@d&}oVkilXkA094FO8;Yx%fLc1rzFWkiH}m1|``D_4z#&oUs-gPDDVr z@(oD`m6GiFRpj_nL>qN)u+gjLkYQs0n}I^y5i9Ezh;4=83WhnEiq-uf9Ys4XQBgdO%N7YpLw{clxvXH0?Qfh+0ZAu_q}7 z>D&BaXmSin(>7vcbOcVw$BB(>7K_hJohi(D;&( zY{H@<<|c2${6|`0ewzyB*jfpiTPB%*&7-VQh2*x{nDPyj$u6mvT`bUnxkVcVv^-|L zE55UerVUKyVHHbru@DCgwh(LV(--TumqV14j)QHncr@oD`ED`Bh`y7tsGkm&)qbPN zQuTCU>p0aWUG@z%uH~uD`y3KF6z#p>wqz{a^ljl8D*{gFLhK=dwz^Ix z@^fPrercj0A#--NcNYIQIfB=@>f_SUZTPo64gXBOL28_kS9HFF1O;hWElh@L-X1Kq zt3&)n518MLChyqKlr=<|Tz?g?_Ul2&m>vx0c5gT|kH#Mb4#^g4bQOp2g^^DDs+c%90`W83v8>6nIBzddknofUf2yt(pHYoR0h zA@aJN=*u+|#JcApMRg1rY?Gt-8G1~@@)L{F<}~u&V|sQs73TKx= zv}NFGp@$tyPl?uwtwl2P9?_eD39Kjmh3M(4g*0cIJdM^iVcGZN(K+0fmVfSKw$qXj zn=l-w7nab=nlpHLU^~*Ba)qL4d479=G}lk~3X7sW;`q_|;~ld-hpOzkLA3KU4ABs*!4{T!9yR zp=mjn8cp}pvaJ4)Xw#yb=}j!*>|yoW8EGmK)3U(+xJVfL(iM46dK=um1ddHmQ+ zN#RNK?C4uM-&w^p9d|MR`T(*C6^rH#;Z(6%NBm^MTVC3BlpAQyhA^t4aF#3PR5-(C zC`W#RCf>f0$0*cOu^USHa&W(lX`x&81 z!ynZF!nq>vgg0rPEUaxbxl0@&l?5I=@`BJ3(h|gv3jK|1btSlMosdx<7K@hr%{VTx z0uuce)5c~UdcMY&9NNAyua}vk$dmWjn76}uakStOy&#Fe!Ab1jAw7z`z6dKKovHGC zB@LBx!j0gSc<_87hA!1Ypl$^2KPy6KOg3n@4O&8tDQtx%J3VJR`E1Fg+DUIHbYK@< zZ*W4+)%Cc%=mwsuD08=BPwrUk!cU(QbG*|Ka({y`DzYDLq+3zHi!s!1s0t+a3HPIn zJN;+XKsM?&kct-Wjxt-W^l3HUP03_9ubvjZmnWmD{#5y_ipn!J5#f|aHParky3Ko8 z!G?N--S~pxT6-|<`EQzdA)f{xGsNw!PpE9te-u0TC(D0-f)1{iheL`J6rcJ)dToEa ze|wJ71#WL7*?o!% z-=lOr{3z)kyiI?S--_aHuN2K6HObD;_Y#F)zD}ll2EylE302>}$R5gi3Xm2nv7*qW zbI4UutSn?evP%77u4aNq<8RQb!v3(8m=Cqy#h9g-k9kLQ5a~V^O%r_)m~9CI|144~ zRwV7C8q~Nt2Y(#gF}77%(3y)#-u1R-e(Wn+jNPHxu`0f%ojqA}6l@4>^ zyYA3Witd~&qa}vwoY%yoE=Y`Z%kIPW-UaB*-3_aT?X2H`fz;(AL$&!G=yT_gkf~}w zYpX66N;4Fi??&;iW@go=h_r_9qWO86xU3^W&y`Uyzwv_&R|m5+n$7lxB?x)co!s_T z9G_{X0aa%!Opi^%lMxr`u9|#^Cc0o{D|f)l$_dw zWmoKwy7LE<>p4r)oI|kiiY_!KC?Yk}1;4*E&~7%7W^7i+726^{|3(F`bWqd1* zL&Ue8I5zej_I~|<`IkT>+jeIU;=@N|(5dc5mQwPB zeO)g`+8-618@!(kYZds0-g{_iE`rv{OLTsx1V&X^q22favlBF`8%otwXrsg*T+`*B zg^ZieJSVOwu&o1?oAJFcSR5;`9-dtA5r-?^#=szfi;{l@6W^br_J$TJwH}OqdcOoM zL7m6FdkpXYX0Q{nsm$QR1`3Jo2eZRh>48@zb+jttt&TmZeo&(=CF_y7R=7+0h9mm3 zGrcl^2)b^Mvm4a;!U*kUea-J}iWqHH5Qm z|3l(qE$G(9B^VpA6_Gb`G23=4LN4}2$>)hE9~F#k5Bp=!#J?1BP6My=g4uLw6BsYb zMwb3TdLP!G&-W1cp+zFT_Ou1Rn>3LBsR(9KB1c#pe+#)24op1$G!t-(v@s+X`@Pno zL}nv~F4UvcPnN7SPKBPA>GCzHw@@JfkE}ndqi}i}w(D&|$G8|#%?ly=u#-?+X2I`_ zuoS%IfjoB1X{K23LiyJW@VZxpe*QGZtJeqMopzT_t{aPI72{#K)0dCl)5H~ImBg~6 z@{p}H1CLG>pn7x-lGArW@?9)ETa95c$QH{(3h>+iISqf>LH>PhDQD#|w!!iOr3_6& zxV0&b5Z=Eh?PkGvn;c>W7L#GZGFI)c2d(ShXj{)m^1SWK(#||2y5I`$6Em30X(Fkf z=k(L)0_o@pZ|9p5Xj&}ES1N~asgFzfD4CHw>DGMS_m0r{_~jRm_-!WMtvW%hSZ2U3 zo3+!r`?H04wvKAe#=)5K8qtLF&8w5UMnro*T7NrJ!lvuDFe@4ZSeCD#dH&Xpl=} zD#^!8fb^X2B!7Aqv??qi?lafUVvs4T4t`8lyFZhh>q$}wdqBOCx^%8kTj(DMz^-a3 zGOW;`$$G_P^Von~pNr_&l00hA+(rkLk5hbbIvKS&qub5}t4=55VaG%G{&1u*%XT8V zHk6jIs}%8bC`w3 z)#SQJo9h3|pn}uN2w#_jVzr}qIyMQj{t6tg?pB_q@tOaBHgb94Ok*AVD1VL&ddw)#_iY-;R7;_OTvK642e9i9b-QoOa zpGp{ICsV&EoL+sm#e1M{uL_fT4nyPkMXb}9gyyykvY0Uy zUJfs)%xy2VEgv9`RO8}|d5gs5NB5$+F^TqN1PZp964INz2R*BvBXrjfnpQL&j^+g1H0BurIdJp7-3&@Zm00)~Djrm7QcI_J`#L zExNjDo=8+NPi#6fT->w4LcHe1ZWNj+@Fiw;ynNPn7NR_jj=dB^@szW`Tp5ImtD{Au zR}wW%{z4uRzOXuXludarB6AqS;bPR#8J(c{-WY2TMS;Y+n zdfhPvsiN;>G+7sA-TR$sfA_asS7)_9K_q1BKcb``pjXOKDTBaN{S67A|=%kIu|XIpK0Mbq~uv$Oj{ zSgf@N3(xEjjT_KJCk_w8GWmOm=}e%G%DHq(Y)4~+Gb(3of7)g-nk$atkc^pYqmX;kfK=><(!JSh=wgx+Z8}jw`wmDW)w_)qS>0!~ z!9ys2v5>KFG38Ndrd*@Bnhv~~LaJF8$$NGfU0GPes+X=1xpi!(4Z&*k^v1k{Gz@+D>3p}3GNTd!tto@tiyI4ji75}ou`Ug z#ifuEIOb0j_fY1NrBt3OMNxkSW8c_z@>rrq5!G>&Ba{Vrk2nHfZ&~(u^dy>7{gK)< z%b1Vow@4 zQhxCeLHqUaIu?n+2&qm0`b+zGv^sZm?Zn zlj;4OH>6*6o2p9gkp7cvRJ+~B&UubE%h+T~J$dV)Uo{YYPny!fZRTuA%MSYZPS~X! zK1^vdmEhF(CYimm75T>uV+tq6u<;u0>`{Ray}L00 z`~b?BI0hf8PGGCM1kWBRM~iFM&$;8nlP4HLE30*A(pXd4VFWBciZK z8R{$$-iTJYqVvDk*y>B|7im=wq^nWK*-zu^cGrLPC%NfTlp8BUY12Q$;-m_XwUH4r zsB_qp1$pd~oCj$K9>f}Lc`p6Pi1Q0GggG$)Bd*Uzz^uVo>`_X4rmtjbQyb`6>v1x+ ze8|cigV?Q$YgpVoXZjlKf`Mz5@H{_{GKQC6{kGvkCUO!VH{-m5;3;Wa`DlN^5LrI1ebMD6w^^nUZ<)6&VMINZk_F@kHfWbjgSzpICn z!W~##9tfkNujH_9x1bZ2QJd#mvR{@-eUBZ%k9%p5zhr{Dn!A|S_>H28S=mg_e-qg+ zA4@;dEJ*9}Xl%V>As`KUNTNEM;_tm6%aUsPX>ga!mmQ^Osr~e}RhO%M-iXkidGv@+ z!wcyyHuZ0zsOe2NhtT`1uP(4!gDbGSb~Cmu@WYAxyJU82EHqWt;lpY>SbX>^===9s z&1-MWUw9UqB`%`&wI1pkP0&8v4)=ejqhOI2nw}gcsRs%;X4y)a!dI?eN*np!I!w0* zKcuw9mYDhVH~Hxp!aiv%Rn0d+#Pj6{y>pN01#x<8R7QeDmC*-js-`bHbFXZC91-|Gp7pgxuOW09vC-+uO1g(fhX^7Cl^G(ce_mdX5^F@dj zm?ILY2E0mHIODeH<7bo?qB?h>;%70*G-#rGM;0!tjX>?DB4pQY!>DU(Q4=r2sgDd>N6KFCP-H zir*DjTU=I>RliqQ;~_zl-Zj$Dcn`dPIhdP{bLM$p zOu3?2UwY?NO5Z-IA+Es_1D^#z)4~*s4rU`mZ#UL2O~GlVWW;O z(wA<5Y~n3ev3VI)dh}32Seane|Ah%MpRjOZDfaaUJHoED==i0=*GSs)98)`9p)`V0 zOc&9xbIQo?PC{AHF${myh>TNqc;TOqzdzEkYs(1Ud1*f%G&r4Kdb$N;K261Z>+_VZ zpNRlrmUR75jw$AwFe&CP;-p(3cUK>a117`hbPkpsPb3r4qFs|-F|C9LG~k>E3_L92VAv{=f59>p1aA==0*Zt~9Yj*ylr~88NA#WU;{P32Yr}itBc5gpD zo_>=|($7-$=QGr-YJ(2LD7*>Xi^M(h0*};(WZGlt*tRS>4|STIdy&ogrioj*MZz0D zjr@AY5#9&W&kwSw+x&rM?7Tp;rcA*tv+b0tpGk|ZSfV8T5|56-#rT-h}ZUs`ogTyPyjmS^F&>{x7HYK=u5F}VMS z5Z_N4nb)7tfF5aVh?2pd#G4rRQOG4X1!Atv2zX`gqdvc0vJTH8Cbh4CEgyBB3NOc^ zbwC7^$A#1T=qPggd5QiFDPYrvG_ZdQ4zZ!5ve>=oXuHx!t67+He=@h8&J+sW>CqY; z@sztt;$I;$;!209$aK99uXzGfSMXwvQ=A45oiF5Ce~%P}{@@Z`!lP14<#$B+EJ7IlYOLj&?Hhv4vE!@C+>PM#D|R z92+m$W7EJXaE!Z80Xw{StyBb${Hn>P|CfUidg;)=y97zQ9*Hyt1yjK4@#wlG%gs+3 z^9LdSu*2^&hAcAU7c<=X?a4~mwml4!inHL+F6d9YE-@Fw9?@B6Po^XuHyt7c7N zV+oi`nNqd81v_)q6z!pN@GmG5 zOC$E7Rw)SfyJT?jNddl0oQlCF*U9hi3F@%*!<@_*c&ZHJ-ECU@u*xUY2s`;~(b<@Vth_E% zwA`^!;EQHp#e--_-csXl%w)Ktz%kyZ6eC{s&RLw^C@=1L?ums%J)yea7^$lhs7B&7 zYqm`gIqj6g$_*8ixc(6J_6YuV&32}AA&j-DBwQ3 z(q}8{wfQch-_G=F-+fjTlE~up@&wPWFB00+G34ztTCDqqrWUl4nvpWf&mX3%Myh1w zS3rk6TSSNFUA3#zzRpH{USwB3+J+s^eagktwhBHfFTN(}Ad&=c4qg3ChiAvbcSR6B z4V6RsMl-mSb(3{+6v6uy-A@juTT@b5d3qHj+g73VxC-vmPx7qYNh=+P(fp?VsDJd9 z>83GJlyfGsTGA225^+kql2&9Y(&MyxQOk*~XfjE`f&vv3q^%>D>C*J3^1kT3qZ5TL z{>px}JFz!hOyqNdT|BvoP1kb9!h@OQE2BWu1BT*ihcTw=bkfw@68y{hedubn#1oe| zv_2b!1+vN*(&bGy!~5a0+*q7*o6S4!yYsFVJI*R5@l^v|`GOCQ{Io_V-kH>6*Vi<( z@4v@_M-5|F*L@PT-+XL$!^yyQi2k>c`dvr({K)xS{g5eNSg(U8yLZrn#%Ox z#DdwksPgeGdUNj(UO7cUyG+<^tH<)Rg0=ixVg#@E(c%|J-9h4iN6?nC3j3no(mhiN zD4u=54xf)@nRN-YNzsi*wMy}dQIByTnQ_(3Fuv+}IgdPek50Rfp`0v#+T}e6^3$%+ z+3DL!>24d#@7+tLCoB=ac`e$rp3#&eqbN0ZA)9?~E6d2aMDxQ2F{_5%4DSo*>-0^q zZO?|~%Be7smZQQ31#F$5z*DE)pcUWZsWoLg(GG9)>JP`F|AMjRl`enqPgcnLH6njT zo%q1ve6dC72Jv&FY&`li3#--@;Mq!n&rl%2I|~(g(ORJ^XQ?pW$m)kCQ~{H0qrdPSVf@H6EwhH`iUiU!!*Gx!_>8z* zsF)AM%d+u&zVu}b{}7G8J3pc9w=ho#GvHe5ZM3pV5lR{3!HwkD(7F>e&N>HAh3=N{ zvTl^rtU#ebI>^yR=rL7+`NQAn8Wn{v@xF*MwWiuj1~k!eGd0Cl;ll_)Hw|I5`oT!>$gJ+3j0G+=WUyr%_3RlY=qYN@o-iE!tdEkBYcfr0 z+r#c09tR2fP47kzBF$s9bogB#IIZ`>`Ru-=K4dd%U$TRZ{o;%{rfO*Gy-m>%CAj1b zA6z}CgdW?C_&CW4+vb{J$JP|F^X6c&^G*S;CiMfE(K6_7^n~qn-wuAV4nNbLV!hyS zo)L2%elxXrgo`uUqUK}Ogz<>FzD@A8hv4Q8J=}TGhA~RJvC?Q4TFREw)1L{X6A(|0 z3%4R)O#wrielzD~b7+K>Db4O^p`l+YMU{^w*(hsk_N7OPhflapkvPa!PF^d#+56Gq z2P;^HhM4I|xv}Qmhea}b6~t{Xz6*S#mwf7n*H{qx7;`rkMbgPX?| zBj=?a;$KO^zi&11hFr?qHJ&aNZK5;b-Yi#qfd%bIBoEOOXpDP<`9=fz#~t!Ktn0DR zTYLp8&5A|sGYUnHv7=d`?|IlCUW`}gR^z0cB!AK7hJUSP^d@p4?=N`)b-hDT_BW6Y z%ub{x%l-7>qda?TFKzeT;*6*^Xggo%x{vpaDdG?Q%!b+Xeh_WgP6n~yzYJx$dfp*y zAw&K^O_x8vqs48Oc_Z*=70t99M=cToShjbNphqfGjq6O|&7^}{Uu)@}#&>d+JxWc! z;Z#^}$Gg2Nu{u!_f;og+?NH|ixldu6y95`{FB38ZKG?G&3!dhKpl5iD_TGAd*I!c6 z|D!xszE9y&0^hq}>Jy9#_uxKC19|cEKiGR}D;u45n9a!=$M(6oVewLVoYPLF$A9xl zJ~o&o8Es*ro2j@f=ZXQX%P66|oF-Sx)0RRHmMRBw6MP=AndPEqZpUat+fJId#GX_f zK9NbrbJi^%%3`-#;#Z^%y?Q!UBxzqj!>aO0!c!S*JzA0d-wF6WZ-MsJXtC6Nckwer zBk`h>L*WMnbgv$P(j$8jZKeg2b6TYG?J(Ae${=xW6Y9C(Yjdt5zb%Or8e@euFBM@i zY#%Kdr_P7vDf1f-yJ0A^ky2MLpoBj69iM;Gs>-SV)r@jRj&ZgjPWDT8Y zx365J1jCo`O+-fV{GLv{@S)(!k8JT}J-nQ;iv~NKU^`~~ zBm=)L3iZpNO^E`pB}0!}cfY{;-V|u&3I5nymY8s21$q|QqrdRR_xo2xaa0{i*54^uoek4?)tE3%k)Mf8UWcgnhl z2$U5vUZ+RV^>hm92ZlXN7cwQ%o4S#5`R0Dn9N#_1m9qZ&t*gA=oUkNEE4X4pX57@Q4y)1ULD)adn&it2+=Zjag!1U!4>N}|jukwl!T(Aboxv6-x zc|Z099*4P&7q(XJM&|Qh@XvG*`yZGp{%^oc@l$a-9rRRXZXz9;D0tBGtzVMb=G~-G zHh?s({i$@xPV)J64#AhtqPZ&wE}yKBX|opErgdm7zeyXd#2D9DhArFPQtpIU3RA0P z=61W;_9j_2ws@OpNOuMGIX)62TUKJi<%txyH=AC{v{3d+kw~@CLUiA?i|xJ_j|_z% zVb_#Q?t!PU->{OrCO6sbm7FL}XCuW9XZwgRmT$l^wc}WG>n;YwXHakKHjky$hZS?@fM9-)crz1?(cTOFnPn9`!zTJ$LL2$QaKB-O7z z)b(&DJv28W^`F`lw@{V8{`(7G&Ir24t_>u%zCxe-%cJbadw2z3L9BKbHqVen-)JM$ zUS5Y3wNmn4Ih$=uT+7DXe1<(17JTA{34CU(9zO32gi**f$d9jJ-zQ(S`+ytlOTjaG zp4&m;GX3DTRq%WiZKM_zLGKM+1?Pb}{K8W&p4XdCU-QS~kL4cB8EU|<2%SnneU$m# zgd8N!YQz0-p+i38AUrqyf!OCG-mezu$||?XDSjn=y?2$ggA@_7PYL&;`$9fj9tG>C zV#>_>JT%}mf03NdHH5cN#4Qs(!N!WuX*QtqC3euexD(or*=QDKjHl@rP&y|PV~;LE zVq83|C;Ic$nd5kIz;M1Ia3!hlyG+BwvaqW?oQ9utBCmyqXi?W?7CCVT3)$4kJ|4`& zv>OSiC=ooi%4=Y!Jr8EJ78s%7i_MLtq8mcCStZULw=OIZJ5)r7%iZUQ zgZ!7Ui|QB2{M~=>ojitQUsO_Um@fvV&*xhPFXcv#G5mVwZkiZ!o$Z|6m+4JX#Dr^` zDMlN(-3EywlA6Qz+@xx6ry-643kogLWKl7x{kN&rWz)QO&yp zH1}^4)k)|;Q!}0I-+G%0Cfh@i3HSX_klM(1boJ93+Fvk|lBS(Sa$Kjt**MBo{_Wzs zo#cgl!a-(jFr2Q5*P?CLah!>0!fVTM*lf82dTVmAc)uhcJ9z`ntou!YT8ojCPypjr zA@3ZLL(7{VvOkuOL=UKnl!W|yt4}{{ESJGv6&YN(b(ymFTk=`!PYeC>OYvZkJ))|_ zkj?c%VFibE=p;mF2SR`Ed{`Q+#FcXkVEx4slB&`$d##J{$5r_;*#p9xY#nCJxQpH9 z)mUS)9^1NA5Vy%6o*PTyaDAdM->ik*v*qwC9W0zVf~GKZIBF*Dg53Bz+_Bh#a)D9p za&b0)**u@WEEBX6frd%KpQDo3~)R;0@23AowfHJy4Tsh5`NR@zULsYD~=W?OlI?DWybZ)f=e) zG&xj`+evEg4RFXS4H@43(CIOoRj)e0?pCko9$ilS#>)a6v9CbsZwnrfJeDimeaQ|4 zH?fT6f|l>o0Vj>y$mnpy@7m4OS8^YF*WyUyw~XShlWL)AJpqen$?(7PCUc*``(YKf zf!eDiMAbd3>G9G%qJ`Ew*!@kXXu;RTcsRW;HAlFU@tTde(SIAv<{KlTutJo(AQDfv z+=SQoAd-K#oRsvhQc{mC9MoQt(&s_+y4wUACh1sv-vcuf8u59mHa3+T(Bx^!?0xx9 z_H3bWX7rV%se}HqvHoLFsGu+I?bH)5|E4b9yLKAAE|nMFTJz|`#`CC6IEziy8zDdE zJre8EA^%DEJ~#KHaNA(queyRh%6sws<6JzOt_(X(;mk36NP3SFk-vKYT3ynxd_fW> zhe^<0u@g2s2)^qF5v)vEhJMDXVfQ#2mi|?WE?o4#@O<{`d7W|e7hJ2#jn zPTNAETN(tOHU zAI)nx$tesLv7GA|xBMpVyCvgw@B(^N?o3L><>c(!f;Z2`@z6I3{K3~#DD0Dj!U-Nup9K{xVI>CLx8~H{szgRuKZWkWT?lf!(txraa=FGNbaSFr~1MC5n)f`Lb|3P z$qwrupy``^$nRhRTh&#CMj2V&805&$3u~%b2{H8VMFvfLdxbQcfV&P^nCl`;0J2Ir z+K`01=F6}&bfW; zF^Hnh2%5VrkG0g`dLngxG-5M3TOFiaqi&iaHHR|%H&T9QHRV0Cq~rq$)FNa|SDZKF zEdq}xz}KC7$yPx@q6RmP37qwe&7^uGiPZjB)7w4SG~TXYmMVuj0|LEas(RWgf ze$Ud?Q>Y{V0*TigM{7TGc%9tI&P;8i!Y8+BxB3LSsd<+E^Vvj`s$A%&g}{I43W3ix zXPUN1N$^<=!FOjZ>{@3pWO9OVwLJ;6%nq*`HE_to8Xea(#Sdd`#0O>V#Z9Ub*m5?W z?j4PlBq{KQ9tT*OuVw8hz1h0l2G9!!s^ zf##&u%(t0SN=+F_#m|DKi6W}cRZ*{83uPRU7v4TXwlyb!blN+pEJP2YJK@w7dX4C- zAqK5FLi0L5kioS~8Zo7aW-Htyi>Cu&9k^4-QtKdQY!VH5<_o!f>+$Y$9A?ZrLGqcE z^l`E;jhTO!$;CWiI@|lw#aBwq=!*%(Z(T`-Ps?C>E)$0!WNYmYQF`ZB`tPeOru`@6 zZ3XsztIsvezc~Ze$2}4LWG*f~)}jHk_1S|}= z4e}#@O&u><`R*wT_^^%1PtsspmT9tc+LxKzEImIu9$rhX}`5#6Yu|} zFkyBMeIS8M*_||b-VAaG(xHsrSU&F3fBe9t1>EB3UCQ|UhUS^wA@5%3-+2Hg{rXyzqY0d(OTPa;H_}4TKHn#8foYazex*D zn*sWMc%I{r)vX!geSNaUxgP1_&Mv|0^fwCL2FEbbd@qR}DM0J_Iy?|M)fd?9;`8fj z`M*9H)bwK*wdGu9&Hv`%ow^$ySgB&xQg8gI6#553K4E9`H70hM%L?3LnWOsy5sxWj ze-BKe%S*noqNo^}R=b;wMedZmxQ9LW=QPjtrs(lLMUtAFL%AtA=($~l?{7wMkLJTr zHWBj^EfqW?MN`PX$cpDZHW&NdU5A^g^I;@-*rF{No(npjWuw6Lc-Fwq^gBe`ziFX$ z?riM87YDh@L|ia47S7@=uzsH|=&sUODV%qo1`Wf07ZLTo{>Mn=Eqia5hN!Gfu#UEpye?p-yosUMTXn}oqIZ3{;E_ZbK}Q;S)ZS5esiH0+loQ@W`c ze78m8@gYsV;m2S;@3D~mvdZA$$_IJkt3&*Td=Uk`N=9MxG5k$VL;cUQSl4+GQ5FBA z=)A+R@ZK*)`Tgtaa_ODtJkNc9KHZVvWrb0T<1l^A54v$;1*#HM`F&MS@|-P?HU|#N6N2}7 zOOc;Ct-(vR9V8PyMAv?N8*3Wc4@W z#$uHHB{FaBOW%%d=CgLj^Bx)T+^9kap36oHXJsmG>fI#EO$TXRoC@SbHs;ou`IuwA zAAP(32hUTEXtLUbOM^W5^XLA25nIm}$4tS?Tnns@8wun4i_u|xle%pU9kF+a6YgiOMUq=@qz0X!g9#xNISCO_5%@v0l3QZ!4+)d`3;C7pbgUve?acQm367hDDx(`R#C=GM|I-cLw7vEk(J$ z4|?~D#z5&gWNj%xe&RJ2I(0jBndX3m5k+i4>>82%mBqxwXm;0b8Ef>IL1|i9l3Qyu zn9Yko3iz*>rT4lane{MFa(mZAF{cZsXZ{cAz4(5;R^p7R9V75jZUmnmG?bqeIi}f< z21)Dc=SxpDt(49nR}|ffz`wm>6P{+q1|RK8oyW7uze*X}lVvdDhB}NNzsIm?Z!mjj z1+15*3O1P(i~qZeoIzC-e5H_%KJ5bSpK4U8u#?hktI6v|D{ZWqi?~6%VEE@P-d#1| zX%~C*NrI7~>7s-UBk$25xP4(u;J8199XM}-5IYb`g|)zf3W0> zy(jRcydOUs?~JFKE@-ZD6YbRH- zC#RnA(|JdD&wsP{tXeg6*{DiN+PA2B%?4hbYs%X<)}nvtRm#2~B?aC7673;dVZL38 zpE^>MTu{NMSANv9)0nVd-Wy>6}|DLLtjU3@qMtktWU&6_K z;WQX8#J(HO=+?Fl<#RM}+e7S6&CZis$8@&0X^@q1SSB4mS4bxQtrS+j6@eOtSaLj< zrlmckm~~auX1fHDkrOG}^DbL8Dvla^=3t}2Jgn{!*)(EXh zqHVp`A_vUPo0q7L_CK-P3lRQ^i^#bc|xufLcP+axr2&?Xm zme3IhP1uX$BaP7W<3lRx6G^4+6L^&HNb2g%;62ux3#Ps?mu+sv#29zM=r-qlf2#6p zTUub@a|!2s4qck=J1gZZYm_ZZvt2K|;?p_k+Hsl%=axlut- zc%FjL$LlcSQ7?3Vpn_Ajb7{Ibi?lC3k7iE~W@qqDTQ6aNi-ERttLa$AXEOvH3Lj+xw75JU-5QseWZoCXN=&hx4?c>JHmsyMb0$ z+#tiru~a>C3|*hJgI3pwT$9XD6k9$L#>_)(K#0hrt z%6HcIYB%jmT8gKO<@vZ`k!#L3C99P)>3dBe^^e?xft@RmFr`0syltgz@&(LACyou8 z6e>MyyIq&o=))e)Ms zbiUYu6|vjTUrNsDD2vR(Q}nk8f}@kz-KzM~-YcW1IcXi$Uiwa#%tb%*u|G=Y+$Q~R z9;|SfImtP7uvO+O81-&pPuIU7mv=rQBhZt^?C(UVbO@i9CrZE1hjE|EXnr+6fVz0hv85(3r@hQf1xO9dyGT_(ScKYVb$|J=vI)Z;vAVY&)9P7)is~O?JAbo=vMsV29K1lG(FBx?cN?*7r`v zoz-=CD{aM-+Kn{MVKBA*N}~}w#knd;4x%oAx?%q)^YIXh9%Rm(FHS{?qdIc$MbVpI zQusG6r_79}bm^)qzYsc%?>yUzHoXvZe%Olo)!NAVol7#e)kr;H33J|9LgPhlWc=h< zru=OSeHn6>E?92FCZ{CUbmtZuH^!fKOJCBDX$%9>!Z6E#i9f*ttCkp}<+K8S@uV03 zIH4JbBTDJhk*_pYMuIr~K2YAK0}au07^rCvJsih3$paj-m-6!lf?53h2sbQk!q6UK z1{FS#7jJK3>Gj)*>UPkHX)WwdVGRwmT#MCG9UKVM#^pEu@GWY9&p(smAk!yP9lzq&lHkmZ!>*fH4a%T zXJYt@1!U5$j=6O&Yu&X$Ssg`fB+*h8ttju91`M2or ztFv%87>%nnt1(SvY4u(`!{SjNA>FfyW;=;4gy^X~JfVXZg4fj3VH<8`_GS;ijbesF zI#|q!-SpG-AldD^PJS{Si0Dw_E!7g9dy7$o{bs6K{EWK9DYCvgC9LS)Ia-ud#7sVI zr0X*@QCL+)qgOeg%y0oNb!kGTk_<1;RpUjj1L%ow5N&rigZ?E@Yse)UVrUE8564l* zEbOg(grOEq^lHW{Dn8#pGm?5>+~O55H%UX+DFIlimjh$-BB($4DX~(2!n$b|)2Leu zF{48;WF+>Oq*hJg4l-C2B#Sq7<@{h^itsoG@V_>}IAN)Xh@Fhy2*2jVq4!Z!Dt@*; z3*J-P5qs|!+*{1CZ-^Ymj;Nu@vQsflvL18Ru0qAMP}())8WY1z_VGdyb>ttXJh^nb zJFA}DI}6yFZ^c&b(ZhK`lPcGee~D`qntaiyZ%FYj!evxY*Ng*nE8m&!6wacsH>%Vx zXDq$?HA1+dx^l_yY*dz1@)(8Fd|Tl*-WG2{8-J@&iSAgo=Kc&!iJpSV!#3k>@s*LT zGyb!RdbgY?S(7@+f`jp@DMxM``V?WY{!R^jhsD#krp(L2j0~7B>_TD^vuh z$cYTLx3h)=6G+kZFAa0KM!(k&qcyXBP<81|vepA1V z{ho07SEr%8V<&c;Uk`J^ta$lGhuatw;968V_l`Tt=j;)8m?uWCm(jxa;aXVpAVBgt z*4E17x4PBU&kOkG-Gg{wS7qMx<`_9%GAGShdze{IJ$$+Ql)lFK)0)~dDEsdi();Iw zsqSZ6xAvs=FOi~W-brR>XxEd zqX;#N`|^k-@38Ssv0za2=4WFbqogqsrIsJ5X|xONJF=0~tTHIy=MyU~9miCEzK5iR zu;*GE{*6_})oWsYDi|vhl%mPEVHR$p7SZ`qB(FY$2@-j}SLFB4?=|FBxexKqa{w04 zHo;6Qdo&DEK&_K0E}UNjogaO1rY;^stukPi6GHL%o=`lj#n()B<{AGM@w$u7Jm`Zt z-nPq7`kBRaXSfZ*7RM6~L{YQ9Ci4F6!$UoBpV3$12knmWh>UxD%Z?@dfaDA!)Z9e| z-4eSDQZc{(I_%o0f)8~*SQm8?VGd!49lJ%aP6i=-;4&un<|57gG7bNC3B6x=96$S? zA!+ypbS>G80hcdP!~qS`eb-6nr@CXq5)B^H_7PRqa%i44on3D-LSlj!Tn-Tdr_D!8sXVs6(MRICzH_g-XClk@( z9`I`zl3weOj zrffXI-Y>7DSwF`k{ERZV`8`^2C!NM6iC)XmiEyjgK`ZnGpQt7bBgGl|hmtWKOc9-) z!*Qg*XJFGtV~nvpf(X$gxjEk#TJp_Ir%3@bv$i6qxF?!h=OQxo1781CMpm3EN^IA| z@^ltuS^7~#qde5^yrGbVfml(Y%y+Md#+%EYxbBmPYMEQ~!Sb0^jqw;-Ebq!D^y$tt zbgwf-)pS~3|Bro6aba)shNA0N9cSyFF z`qxU3Cl`oDq(!iR02? zBzX;u=?3Tvv6nYL5!>xIg*bBkUQ3$raFzHb7%Evss2*wf|s7swt?HE>))$D zCAggIHrZkSwLUnbJr#*Lhfv}5gGc7eORe`SO0|ma)8FI!sdPgY&7XURVpa{olgMe9 zHSa#NwvT6-4hLDb9izaff0%OH5IQ$UFii9~J(?Cy@lC7Pq)D?``js)vtJefdUgjfM z-|^&|eVI1Zn$u&W*KBpc2aKIuhsg)-LTPCke#FS|`s#IDZ?mG*V_GK4K51&lJLEa8`zuYXFQ|&4a4!D zYycLhEW?hS9teK2AKncyh}nJ&;rdy4yDWs>j}m*Lx35W~A&33g{+(S|=T8wax$tr< zfb)wx_#kr{!Ke2k=~I~C3>`zBOb+zU=Of&thW==T(Db->RI6skv(tMD7vu0{v6WmMJpmA1^Ehw$B&@SORROietHt0#~062T9@;xRwd z{)xLfYtR5=9ePzAP5WxEvT8>)_Hn*F3o4(+k9;c^^WtDso(iXozkQJ&z7|sr4Vgk{ zoupw{J!bE;h*K@o`@Q_^xZg-0oH!wkP_-X_+Tnul%BSXO}~xAr+&J3T|!F5O%Py zAuSY-+q7oEmN}Y2m1o1~OmsIorn3U4Z$G6fD3Dw9A}T+mP9sMzqjgF@$%ZySf29h( zt_h^H3Z6 zK8-=fjtcgxRGzIm`-h534w2p(6ABW%)Eyfe>Glt6sBIA0=!`bP=m0XiQpB~q%Xykf zA@@#si6?HaF!JRM6sT^akW^oq(Z`37?-Awp(!G2pL&f8sd@z3Wz|2H6= zyX0)|Nv4LZMfELXdcbe947!kUg{|`%OZ!BkxtmNV+gFo;Eopni%!J`PNiO}6NPXiqr7g4G@gbh-{7ZBver~=f zIECvl&Eo(n)*OWWsUIKr_G1yDVCrkW7KkOQ=<0F*N_w&?3tPq%O`bcD=+oY^E$16+~i%7S7(iA|EXvA_Ro-Sk(2+`4@|CWcKq4a=kxc zBep>zABnF!%f$148)>!6_=)RBc$QNoT>ix)Z#6^2M?F-9&&Kmd_sJnwJdZzPX;$%F z+Haf%Ujqk>h?a%rCL10tTvav)#`BX2k1^=mAM82aoi8bfpytVGlps1Ah050jANwYX zz3P#OA~L+_O5w5RY5$HKu5zn@U!LU2gJ=N18d`(Cs}}K5j^3g}Cwe%RO!Pwr!y|Mq zhK%h7lh7NG`I3a5v*Qtym4!2w9dJv$3wK?0Zrs}sV>}}m#_ko_lPC)QafM6=48r19 z7VvM?LVBiP)%HjszvQJNk1Lqi#}n~q(@G4wCeEn~GAXe^4yzt{BR)^eLuVhx=}$|f z{F|@z+Mqz`w7jP@Hp@$r;x-7q6Hj8m9s};aych4(UCq-H0(j=fMcj&QM4(>>9eXCa z>s2|zYZ#9pHD`paDHWV9tH$ygG?5)$?(DmFX{Fba{Bl7c@%58Zp5v=dc<5c;(Yawvh zzL*~FzeJfG5`3CzgT+oy5qB*ecaELK%yEU-`Ee_rWF?^PNh}%;ufU>MJNCXj1;=J(0Y;T@|@Zhb;{FZ>T@70B0c?zHTcoTnmD*-)aF5*a4EMA!9 z;cnV@WY3V{#`0cdwatvao>V8(IkEI{!fWz-BhKfxW+?I8Lza3&nQ4z$K27!v-;6AN ze6k;YE{w(GQ^ms3bAeo@trC*yt}DEaY0^!l2PnIXsVsyT)$?8)b` zmIZuY^nG%jzkzlssZvibUu>Bp`joqi(Qii>8>rlu)tVe;!|mkpI^reG{d$jH3=D>B zvJ0vwB;do0O4cz!mjz8Y%r1R=L&KI$!7IhdxL*|q`%&H)wks6pOuHk`P8H^A`{-PM z9g;H$C8gd8w9EH3)d`1^nsQfOu03D6CC^Q|&&@;nd|Wn_E6S|2E9Q;q zk723dH1&is>Ic6iU)vUPs60oTQkG)Py7wYGbby7_kE7f^?PT{t@Q}XwQ+VGTn(p=& zOE$a&lMls-z|nl_Pi203lq}cX<}bC_6e|6q?I*pxUF<}4x^YDXU4C<{6Y9!#Bec6O zE^DTVKC3vl&OArzQx99M_}eH6a@J#q(uzqTxi9*5osA5cy>w`mj_8qy{jIejroR{u z2fhRM&+d|_f9=Bb&#IFD>r7Y;@xYWCQLmF2Qsd%D^vI@&PCZ|Mc~Ae(^%I%b)&d$k zE|7)i&tMzfj*?BSF}YlP&#XL;BDXme+U_aPepx56U}ISIJPo$5w{RVYNO;V6;Vn$u zPWK}BQhDSbGSK)*14qrJpG7IO=eR!fYfhu??_Yc^3C2VDLa{fiMP}b^h#R>L*7w&7 zKTsX*>845_<_@4GUmw!v(ub5=m`gi5U!jkfvut`6f}9bZl&X9jGT!Il_fwWn8UF)I zuT~;udIF8oZD4k(dNfp3yp~4AAyN{K^tIk>PW~Lp`q_+nbm`3;tOw9rg$R1yTtI&} zOA#dYdktM2`MSK3+{np-7k7TAq!tMZ8Zuy=c~JCwPvTxp73~T5hQ~{OVCEhd^qe&j z)!`5EORWZ;#~)$CU}Jt@)dC*Y`z%zh97LFV8zp^~(8K1-R8^*i;0xy{bxtiyKi{7{ z8de86rL$Ogh48PBIkt^)!tbl$xYsfc#m^#f)IS494HEGs=Pphh6f>jLqwL=FS@dVj z9V+k*faQuBXpZj3&oADAr{%6VskaxuO-{fqeK&@Tyg-H`19ZHr1i!&|va7 zyoWlEMbhG$XH?i|fiCYRBE?}j)g>rljqW)3_SV9a*XC#%*$rB0Cd_D3Je#|Y)maI+_(g$p-pZ6fZv zHc+gmEcEk>>DLhnO^qLe=p%7-BandlXlbB4Mn$3$iJ>keP8DT6f1nJ4$?3_gRg{-G0!J zNFVC{;5B`?KA0{nPo#~a|Dd|ki@fB<&;%1p$~=6D;*$G7wR<~7UmSy)wW6PtQ$oM4x1#Jx$W*dN}PxAnS#)B#K2UMTud#~snJJew|`cupGH4b*<|77bK6j@+_= zJfk~7qVylSU(&&ZQGsA=A6nIaV$?NP{E?4_2k_%g0@rjfk* zEJ`|cgKV28W9fD+1TPAvUjA?BMy7b)O)sFj`~-9>0#MvsiC2fRpcnoOLoX=s@rN~O ziNQJc=ipq5k!aDs#yX~NbCzWb=J43C(eLhNjWGe%NJ<%iVS?!uvMPj9oaeyx zY9ykcD~kXBE~?u-i)_NK)5vy5!M>i2XZusxmGdbqWr8p553-@B4Js%P9ff-6sxxwvNacvqw*+7bU~hq*>=obe{6LWekZq1`WCkdy?k>h;y*c34C;Yf zm&HDQZy1%1I|`NJV%d!pV{~;eaS2d{JPmlVI7tV*) zGr~C`i=mH1?|#}4bltrcgPX!(etHXhUGBo@unt%7?21hx({RjeFP3juj^2|dL)Xe$ z%x}Mvk4z3JK2N4dF3Ty(gnKU9%uhHE!F$<>D5w`K%ayNafcyshEh)r2f}! zNqC9(^x)TzoWpk*p z`szja=`tQx>R-wGUNT$PIzdu$Ed!Pz0l2Mjo^G2M;p>u7STVs0o32m68m*%kxVRX% z&#%NanQfRCw;qSDRMO(JfwXUh_^$T5jsu>zASspMU60O`);g_|emM{zH9wV48h3*z zVO297K5on(d@$qzUa~y;faoghm?Um@%B6KRy`QI`1rGn*CRO$TE~uvX=P%y(i78)iNc zTC?1s@N$m07p)-~J!5L8N%VZ#STb4PO#NgG&~GPU?aPy7rd3Txb_!0zf(TsySBRG# zB0I8bD%LCiC)oeGsH+%_SQ#_iba2JY7tzQSte^F3TG_mFFKLL{9$2fEOT>tj*6jaG z1uOgUCo_lhW&J1c!>zI~@B59svusd#^%Gq+oQdcC{GhsL6vt*49xuGr%7Ykp2oK=f zx+idh^-)YkVn#14<^h$ckwo{HurCgGN+2jo`{_Cvj7Q@pyh&UCd@x zko{hf%Xh1SLDYGq%$`N@=Bbo@W&_5a{7vU$uhQn^M_6s=%C%x<^Z7?c@&fIzRu?~{ zj@-VnkePg$Ku=g8O%4oMpK zSr@hoKXd+4z6tFSn4W0dLi1^xt*dWe$JMXCRz*`qF z@8krO1y8{9DS23ZwE+FRN74JdB-)zO4V}Jn)UV8#^!87Lm9{fJj1NPk>?|0WTI0&4 zdGN|J!;X(_R)0GZB!?|d(}`XI%>8^fwyGnad0tr#>l>OFJn;{`Tp2{G%GZ(RSwp%$ zunf~rAI2-2FLb&>S33XZ45`nlrBdhdDYWdrbF`(;FnpH8V!2=?6)rG`hgt+yd*`5Q zVlcjyrDOcv5m@PSkoNK)$f}eq%_ksSgi2or7cD+sLrK49edo;AW>gMgEM&6>;`>G_eF54AbeUR(}d>^r!a= zO=yhi7?Re{q8E;K_^{{#*6I!7$E9<)mzh0JpE;E;d0@z|ZjU>Kvn%Snfo-xY#NCxd>zVL>vZ_8QC)HG=Q+BNKao~-T`u@gHpqOj2#5QG@$~5) zyxVXIpV66$cKbM#>Fmc}u`l;Bc}c%bG%%>458ll^Df~oRNNcMJx=){iB_c=tpedgH zlxda(?`vlB=2`Ijp+mWi$bL?{|AAU33YBfOcnm-H5*`K(sP71%>o!4nljRB3ykThU z>5Cyt@8hVmI&aNb%ukG1$}fm}MCPJTk{y=1BvTto`#*)i@%BOVDLMd;s8DJ@?Loe~ z_tKk3%bCPgnGzkx(!|wLN_v<<=a0m}uSU3)1VinCrXhCCmM30vll}{K5bs%i$*O&J ztmS`M-BW!az!fSOSxM6Dozxn<$ zm5qK&`MLApcH|ahU&q68RT$P+xM5*rAiC`c#OzHwg#%?SyrTEw?bti)_UM^RS9UFH zQ#Roy$`iQz5AX$oweKc#mMuQLhH;VE>M~jI^LHJ!uMYKN(KK0)1yvH3VE(7Y2CJ;dEP9P+~3_z`$mqwT3PNN*3JtP)e ze_0_|Z9Eh|trI)tc-pXkD|zJ}6n76>%rg_OY4iE~=UO{HyYF`HG}V+}7T%;&6Q=Uk z-K})bX(Y7EoKePPup-=->=%Zz-PK}uc}*WtVm4~JUW!DkW;!aB!>trstm-J%40d@FqAI2MmE zMfAWd!RmS-UdwwhW~!a^o2`ZP`IoWMAII~cr<;ao!&Kq6U4^BZOEA~&rG$LcJWNuGFL9zY>^CsU5rZtPj6uqsJ{C%JJXDE);4nN49NKXr*5^ zh5Q>wTE7NTgqLt*Z(Jqv^+|O8*L1R9{f@TG7WueUG1PWFT_U)K68pus(0Wpa$9gOB zi7FT285xP^wqXz?J?ZO;e$v?zBdPIPCyX{SM{iX}G#Z*?^tl|8mpe=0?;oRlM=_?# z1>$L2I`fHKB1xI1NpT~(BiO$$was;>k+;R`yxS=95uI^uYhB^k`$2s~=3O(`A6K=$ zQ1k+j`buY*;UdlaWJvLMnyA(31!)=dL2+Pz`1wAk^7XBxA36_fE5{?a>r|Nd8Dr2n zYYaNH8`E!Ir@HmM5E8r#>FwsUDRKunTE3?x!j%zLDm-=9O)xn-QE<=ONZLyq>DKmG z@Yo7Ye;l#r=^*}6c@ajNr!wC@!Mtjy53h*c#OIk6BWUgmTH9|ZUR;)^6t7vNd7uld zcEzxqFnPLsRlN5e_Q!1be9(%Iu&T>rQQ4L>ZHqZZtr&<4qdvfGt&|5_TkthA#_*6n zj=ZB@>}_6`L+f`1I%fpYL;Za;;l(1FdoGn8NDQ#oC``fG>e;kGu&kJCE7cEhfu8(%_vO&aw@c9v_}At=rHa?K=Hn z-6Gy|4+KMRogMnVKZE3(!=*XKmeQ4;Bcz(cbZ6CHiUG(s){wm3t-{7A&Zl2G!h@RHNnHlpVX@9`mUhLRRb>~jb2Gcq!Y(~%MDP_> zG^>vKH)~^(>|m7Wy+rdu4c_S14#%OP-1GN*zT?1TK6=h(>F}fw=@6HF((20BZ2fdk z>dZJwB`+1Q=JzC6T;B(6i$l0jcmzjxrK4N?0J8QF{Blb_sHoMF*WJmm*73%guutSR z$ONCo+~9eZ$cvxLJP zr%Sfk*D}3%U9ek{L+VuCdf)MKGNcF(l`u zg0~M%*~lz^c4+A^dKMKcxnb4HE`AkPf;;4HKiUoNm0UF2_v70?+_2>F)nUYT7M=yALl@y{JdFsdMeDY}veq)O-wR--bZr_i> z&P$b!>!h&SsRmSJXhc2l)vz}?ku2z34F(@z++yG?esSpn94${mU8X)?bZ|5kABv!2 zStrVOyYcFD5Pa_?k)2Nqee7*4yoOgL?mZT==~koZ)UJbgWGC+PJB~`a>S{rE z$qMS6cuk^ttv4-}jH2a|Wa8ELaBRy)#Qf_|(=NM<-GdPZrCW)+YA6kTc9N=X6`FP?jzqy>sl>~OlhiA-kj(F1rtcIkS=}(34c6~qmu&`NR}WuIXi7ktmJHW@ zz6tTde|E)kfV6swzI0d*UFl8*PpG~7i~aMAxcUfl8W28&9?7nuG5SW>TVYLWzfUDY zZy(9^GhbO>?`1UmPOQYqawr=ovTOsVB+#=0_Uo|Df z3f}fBwL=s)A)Pt-Iwk?J9bBPFtusz-;J)h%7pdF_W@CyL;wB+KsvH`4Ec6Oiya7XIfJ z(g1OfHH?m-MOw;y<115sZQKO@%G(?#ZMWi~QYs47e~o-mRBdH^{2{w?JeysZ_LKEk zWWv(NXi|;EI+`Q$ORi52a#nB13vFvK*+L(U(bs67N;kxZX!AJJkI;8K15>>UdKFee zh1-77f`CLCQk}{Qj~!wMdw(Y5!F5#YnM+?*YqBZ_Kgr?X?@YJ-7J2Od&W1WATXp&g zuO0h_cY-nD>o5ULPei7vX(g_16+PSPpBQtwCx7yz1ajy1!2CasipYu3^!Fv!!-6F* znaRf`YxAo2|KV6rDKFghgg*~|%d>QQiTmd<@=iF3#-ouyGURc50)#6aKYKL$I%OmWx1jFp} zDdkEd9{gO5nG?pKZhIvP}t_oRGis5ye~zqF60Od~=pV zL*oI`%Sl9WpWZ;AGJL9y`0bm6_*WkjF8frlmwoOESKA~0Me!k@QFxM1J$aMM^~i=w znyHuphtS;Vu{d-67 zKs!jEI;@^hLgYVsd@z=lPwEM$3zejB>OG=6QnBvsGMp+IfM1XI<3AZoo;GhU`(~3x z_l!p)K0XxhT}I*X+(_Kq-ADs$ljv>0c2fV8Oc}j{X{W^~jDK29r*6csmPtpL%;|3Y z^qlqBCtSFt51sgBi&p%|EX2iO4@e^RWICcBdq<@gUbue}Gg}Ab%t++#8iM%D1#Ud_ z^LM(pw%+>o&>Tae|XML@?!g zO(N5Zqipq0Ys#*4A;o18g1J1NBGlE$V4Mj*boMPsFvhQL|AC4Bl_7WhQzXxMh0C6g z(epqx5;Z0xE^Y==ou=)nD68^{LS?cOQxbg+nBb+DnOkzSQI09p=0#g%*{aq_mSQlyNW~eJlM@ z)PFD3@1bwG|vx(B{Rgl-~-iGD)B{1=im{(3NH_aAo=lB!MyxV zLc2m)D&I z-#`PI%h%Gw4ZW!R$_(<(Ie?be;&JPC1zE!;(YQWaX+zyh+H*MsJL3wlwXhv6|Mleh z3;Ob24Gw(WQ{v^#!}uZlFEEUcLHf97!ue5zJpcJ*6{JTsgT2Tm)<|%Ej)>nG4=tBD zEcfPH!P>cvPcu5u@Nqovci4xw?6yUpSA(FquNP7VO+)=}1N3_RlOAfu!pB;$M|ZVj zptlCE-yY7N?#SeZp0!*OoyY4w?dSR$DhRSYjM=Wc@L<*{q!>L!<*PE}H?07>qJX`o z+vuK#$j-R*#TOlS9G>(GckUg9zlp$l7yZOznVYz?QVoiKZ;(-80Jd6{V~2kioYWEA z+-*aJujC8KT<-@RZwq)qoU;__+0DFb5~Zobaj(J?mk-TD$X_$=_&9@y^^}wDk++13 zVk2d%ji-;MBYBI4BmY*lf{)28;CF|A;1fG!rQQq1BFWYeIz45OJHRD^7(`}jlOWg7 zVO2DLJel?|#fnJ_S!_!<6?7i}*S1A?J@*t$odcoJriy$WMLbmvbupOV?!*@*Z|4!A3wY&)%arS2K*jx}_QjcQIFbXLE*R^{0^4^bCqD4}taM-=t-AoVJkzRwoyrBA^ZpAAYh^M^@3A zj$rz}YY-*){$U1(6G=a7Cnn0u@^h0N`9r~NZ5lV3?(A+LiQr{fb(|#6Y3g{Z)r$-* z&yzdSU>&+%nxy9@Jr%H6YEb)_)CTqwJr)DVCJg5v|5)=>o&J1k-7Q`=D}w)zqce}D z@@vDeh>(m)ROUocl(V0EAC(jlkp|6^Mnx(`i3VetLL^d@R5Z{$sf2{0QIk@LUxQ}S zr26*y{NUo&W)T~{3GwGX$A zH2KuIgJH8I6)r(b5k57T?w5-`()7pF*GQg*-s_61bFN@?lM@Y*+S8ROVdOl-#`fZ_ z3#|THDqR~{gz=x=;@rPCkiL0{Y!hQ%I;J}}wBpiD>wKkB`y^?vb&)hRc{$rd;6)Tq6c(ZPyF% zB1{h^!=7QtI2F9MKhNSaWd(mEf;8WGGR)|}$;M$kztxEci~fZBxez$+6yLwjN_0BS z;$CBe`N9dy@u#=w`m2h}d!{*_-yej2s|BO{^b(w?`h$<#C-5z6{@`2v2QkOI50l4% zkXS~+MeQ`4TW`SWV+`W&WKq9hDTQjDB{TWKc%H#{&z;)**+y;NX5IzyHJb2f`$VVQ z)c9hZKbSe^5uQ%o%G`!*Vo5j0Q^%GW)J^L)TYTLFx&iU5A$GeM#2zG*B}J5cw;Mv2 z4nXwdpG^1BI@asvc=jv!1fT4d&BL}v^0drCBFhw7A=eG=-_9X(Zyd}*z43e6TO2jK zjfC!zn7g`2dsc(mD# z!o)2<=z-Z_s#=tRyi8?2`+X8WH7rbc9~VKzIshJx4W7u_>qR#&fsPxmCmo$0XdIP*Jlxx1k`JrmUpDtvYFYxqqnMz@sSG+A^KQmx(#-s1%f ze{=!9Mdx7X@&-j$N?^C96x*ho@UNE*xa{3G(CG3O`R^Oi)vy9x&K8ne!C&&LG=Te! zJ^X?CDDGjd#UHe4h(gy5+TSaa8h;Go=T0c{tiRu}H&v7Sl!z?oCtIAp_6@JAGofJW z2a6a+&kt>3GJ6%NZ|y~F?~wv&btF#p*^8u_Burnq0E^zpihFM(vMgo!r5n0jv#L8c z3VIF2y*m+{Bc7LW56FMk1)N=_2h92@?th)|$0ei|q{8@XC|7QZ;XSOPxaY^~_~2ZK zGj$4RiYq6Rn&-4#HIiC~52Kp{PqNhyOIX>FFzoo(4Ha7s(~$pWlH1+Ilse@W>8CuV zWewRRyW$MJ3*C#hOOuf|{S^he&LhdN4b-$PjCMC@;bR|dgx^!gRk6n#(X@+>j<2Dt zb7KGAUsm*WDv;8xE1w+w3Wa|U;MLGsQrm|krM~~Hq}v`w;YE*RtQq~DR8CA|<kGQ`UuKn%*BeXW z#|qfU9~YRe%QY&_^QW?PF4XnZBa%1spvDv{>U152zj1!J{gdO@F;|Qjx{{WLH!#OX zF|^@V3u`EirZ=DBBudAX*&9nsa&uOf{`b3!lnrU&ZP|9vt{RWfDu3bj8-U(bjr7^+ zE)}RhfywX`T#Y+JRi7SF!=q~Ym;aZ1MB`&sa zDm8zyGiuF9_P+`5fhUn$b%_ouhq5nA-?3=L%NV=q7QP0TKrs`mV}lDz$l3_cXp>VGO@yG?`b6v-|7UOJOc) zpw@yTlHYbE65SyyS=xq6wue(3BpX*2NdB!c#oa&QnAP8oU+j5>71*dzUHENse0~EL zwx7g|r!suH?g^rTGh`un-HM&DW2S<>>nI3dG+LWEIZ)O$R z4<#W}G}yE2yV=w^k^Jide=d%k{QA0I6smAYc)1ElvFS2d{98_Wbqn$I;(ydtaT1Om z7Was;4!C+&2^~|4NV!0t`z}`JS@XN{`*UPDxBi8Ns{auC@c>ix)?jZ&9%G5~me32q zdP}T(Ok1A~fy4J|(rP(O*9{$M%!nYm;gdt1HQA&+rU$wn{7ofaMpI5nCe0dhfPy1R zne(D**68(!^}3UbdBN*3)xsPrMR%a*nYp;S^buvaCZp1K8(i)k!+2$TX4F27wa(=m=9$Xy zsCZ5?jzh3F#Y99b+k{O?jl5YoExeMA#fnWVO}8ugy!^!SgV)h;|3hT)rI;FfypVJm zGl_B|<>AynLDJ(=w8T}ljj0qbp&kJn$#rTIb-O+SjoUmB_jCzT7Q|xKqj;Pj9f(02 zPH`ordOl}-6W3SHpnhAokn$Bngb%c&%5YoK?D2vQHeIBT&7Ub{vn4KTikZyd3;41m z6lcW@@qa4i{3ix;5QjB)YgV?)ai23ln1rp6PGI$7Aw5aYq*p z{{!M4<%nPLBe8$1DZa|7;9JQ&!CU`JhdpdauIe<4bU4bq?~bM0Gd<|dq*WsI)H{Uernj^+aXDFKYM^Y*Wx90Um6lW~)AuC7 z+dFs{C(8!$J%>CXmlTd;NA6+t3IF7iUnU z*oPUaZb!$SF5FBzjd?6pqqTy4Z0bH0(y~h7!RW?+1`EcF(;=Fxbe$$oYKE>+8Lr!X zC8NeI_#pT+V{6}$-!^R;{-lBBeN+`$*>ALNk}<*-IiufXKb(n?=WDzd@f@9x*r1}z zF^TaN?+kGIuDs~TzbEAm1v+>kpXpBBNjhKGB1LqUS5Ll)`7IH=zfLRP6)z`Ed#b?~ zH3ZXjN)B%HF%?52yB~{!FlXfWSvaM zs&1z#BDfcZ+0DW9+Fzt}*BBKS1@nFF18&-9jI`r!lyqmsTqGv7Q%b!Iy*etxUwV8& zi{^VcI(zWS8kPL&TFwP`{c_k ze$3)CfAr)w;=C|3XaIM)7mcF6;bfn@McOf8nY3q?wewJqV;nNdr-u(?;t~ZNVjx zBR*izAFEF;AIes#NEW}&WTmSU;W*z&c;kf6d$9?E23|tdIxYVF{uVOUx5mRM;aGNN z9O=!jqQ~0=&(CBcUTeqVzl?n7<)xsb^c1FsWuV9`gwkePh+b4GR-dn?UR67g`&*IQ z^q2VJn$oK(Aji+*;Q_-#*U8PmMwx@m)X}Uqy%2 zXFZNiI7Al`YU!l20q!j|z{ilT*qj}J`Qo%kvqCTHbHjqVXTLn?M9R3!Ci+h8} zv83I_pc^feU)B#^i8&&RBUMv(o# z%j8qFjlKTfizz9$F`WZr@ZQB8nvb=yaFFoo3@t!G`AU2`97*wKj!5>nT@*aWIP^ZW z9EERfP(0u}rn)>u!_iyfIWrCx=U&jiovvi^?;*)%6;OGbDb$3=FfrAF{m|G#yTay5 zgT}f^9fpsPs;shrY;`VaEqOv$Udr=5dWt+Q?hkq#)RVf^s7a+q>-hb~@9e9Xr8B=_ zxK~!ou3di3Y*SQdNjC}pTW^TPBi&$XBC>noqLaLS1~sdtl2d*o<;b6-Z(Aa0!o35O zw_gGAMU$yCY9lSw-9;U{AJ96%QLUM$f^LR`d58Zw(U%fzK6%codUoTpRbsGA%S?LT zqn|YMZg;8Un^N4iy@t7GD`D0o2i^ZHQJo_^?&1CME=X{2FWsZqD$%Ps^^M#$2hf@= zt#tX^bDBLp6xB43{WszTTl#xDJs9f(gWzD%ojWDCi7{-@u4>8oAyrJTZWD%1N2b046Ax#^G4Y8r9X`7A!>%Y`i)D;Cc?^DrPJA5;apr7X)*_q}q*tD(- zAL#TF^&?Gi=}Hnb|CC`?jSAnh$$~E~apob>V!rn(2HF#Hu;Y6!$z3?eUfzmlCs$gK z&%`=02>y<2H<%d>$K}b-85X8;r`SK~jD``dID9WA!{J6>H=DvQm1p`!Yq* zEa8G53iue|#XYd8>Pd{fzRc3zs^V|3ROBu*VC#L1+C&#lY0gDbepM?99zPHecn>Cn zp0Ih&$LQgv38>SQ!@kuYM3nkGscSx_B^yF;t2iB=icf8=yjIeV%`xcch{P<%olvy9 z4$C4vOzCEb*Ndz$cC-avEd4^=;+yGfdnV;~-Gr6_qOa0ojE;}}d4q2bchd^v1~zl~ zT_;=KQqzZ8n_R`YG7hVU_k~&F05}@y;!R#T*{5qlwlo!E%m=YxJ1bHuu*5^vkJuA= z6VjslsJWGc(WmyHE+P`mff*86zi`?zc?v??-oi1|m^(OW^V8E`GK;vrV!soGdB<*2 zy3tb_Z6F6%!!J}Xogn-@_INgJB_2K-!Dntf#UK36g;BXXo_aZ=8XHlNv=57Z-hr}8 z5e*k^fy?fZSl7)KW%fEaFhJyX_!bsR=e=Qkx{Tw^_VlC}*xk8U8x(h7z4K}as12$_) zdG{+$+~Ics-)pp)htKfk6%Q`+b}cRG%dsv}saq7;FIj*;Ha9SHlkiKu+eWMWOvrw` z5+Abs37%DyVo^fXuwS>5NYOkUDskPg-0VNf)#-wzM*XVdsA%OxbgkBayNJWDbh!PJtA$U8We{)uea zNjc#qQS8b3-*IPWZ;IKBU?SZtO~-P-PQ=VqhiYG6+&bWopC%EsXm1Op^%#f9fjeyL zEHv1C!6?$J)Z_KVOS#gwG_G{@H|=@0TkyM7`9_5>GL7j@SHgVQ?u{$S&vztQo9-g@ z2@?DenO;ZdF^FFB2prb@n%H~A&FjM-Ha|g;vjpBtBgn|EgN))m5HW2$E3Z^%q4VRI zTjN?ht$DmJ;Hb#7^Bz;9&dc5ov9j#n~i7Nk?yFn&3-|GQvzstw> z>n*)B$X$A)Rm_qW%3)rcF75R$r2Rh=sW!HdGRlMaMTc#?Iz5ChG+01>*V<`KtrtRl zU$HkB?bCirN%J3)(B`clj5R9*bfziCL0C zzsIs2GfQdFrjNAOv7Qtz=1MY_zPEk$CXBra3g$KoNAN)pPT;9U3h%XL9KZAJG}deC zOO1!RN>^*TNWX=YvC*zun8lwmRy?ef{azD8wheOV^KUs0Y~GJY_Tfls{7T1*9k4EZ z735nq=(U_3eYxKcD>Fus?9ODt67!^r2S@4ho1f%wy%{kdRVeC~C#~LfTevy8^O7#( z`MHNW{L9%xv~QgqO6#)lw&gZEo~wn%svv0h|4Z|q+rz(f6%^~&V)MlJbazBN6`S?o z%kO&dt@psYjwzuZX;DZXrq1i8r=f*DFu`oGoEf=_(>;x2jfwBUo_6zPpK_eX}X*ho?f_!;lKCbxp@QJj%neE z;Ek;ul8XiAmQ-Tbg>0(&Q9#Q|Nl1M>``&Xi^_oo>+|L1Z)5LT5bqoD>;t>^VpQYP^ zGvqSc0RN6m!51~b)H!HNFDFIQ!K?^&^m{g&G1Gt^zHGylt@-fon}xQE8&KCX1DjK_ zv19&mm}w>9(le2bAH0pV@19EMT)Lvk>Y&uzYQOZ>zf5UM|1b)jE?jMOi)nY~V94}3 z3>~JxPhC#szK7QHqa71@V0a7duAhT5Lk}V$b_RAy4?w>*53AQ`Q)qxM)qj!4$iL?( z^RgnGUua-*714_s7s#nj7Ma7kNwO6G49lLM$Trq3pj|&jZz$Z6Mt69#q;VfuFRyK^ z)WQIn)8sIra~M+Uf8zNfSzbQkBLaN-N`3Y@OUoBWOIu|dD4`o+?4uosI24aHIk8xl zCvt)B(x_0{4R7|(K=#N)gv}MR%b5mZUf7FvG^Eg<0psair#jGc$)j}bE2KX_5w~0jdul!L0^4C| z6oSpqB{;j4)9e2Rlm31Mdg?ctTiwv+KMdM&%sCPri9Yz$*8{5#Psh8xyI|YWihl!w zaq4Fl3L?(qM@$gy+P8}`Vh_^my3zQc!ls+%@`8x{OHF5}SG%$e5S zU%_TL-e>yOe6F-?=>p$Z2e2vpcn{Q)f@P@^&}i$_@j4o8U5UJ49wV(o7x^F{ktiwtARdK?QsMp znyFARP=#Wf8j9;O$f`Dx()N$VnaOH6-_}CW-5oG&UKh-&-%a+%^f6f04CSr0aM4Ub znc&2%x?_O%iTg2U#BZuT?!*c|>=R5SYqBa+qPiXrX_-qay(myb=n6~tc5$JL5#2EP zLI9HAbkdS$ZThuk4!5qb3Z+&@&p%gjZu zGo^(_pRAxFi)j?AsY7e#m9wQrCumHGsB&(UXSZ@|SZ?k{wqKNl}vyb;S=vZ z0jKU;AkE2)9scs!wjf9iC*D>P4?0Q1rinYm;Y0M?Ka9Hk%ORuoeWYVOiT)lv2_3!` zxsq8}qwt1)zuwQB({HlN(FdhvE*qtr&yuwE#~zGYvlVIIhl}^C_#ZBR3$;7__}J;c zc(k;dzYNIYWh2)yi{snrljxmk-af{X2gI|k%trEU%V2I`cnbkRiGuxJK##}2BBc@P z2yj%Pvc{XVz2Y8qZx6%n!M|v*)+>7D`JC$B1ks`%2iU#;&e5&v6V$Ww994huf=7)N zEqcw_5~Rh(>)!I7AMD{eu$d+|R?~6>bn?VXj!!m#cOk$mgq=!8oWvAUMYdhGF5DK<>RzgV%mc#~(v){;X~WFYRm2L#8I6%pnwE zO5?D&{x92kcY#FFK0p#$GL3&FDNnQ4<2wiUM(dqUDmi+TK1bDH>hKy^9e9mW8*M(Y zpbsD3s>!crA7{NZW>L(Ddr5i?PL#U><8VAKCABBw!DC}C*_oipr1^;_q$KPxj? z$~I$ee(wsj-Va1!@d}YA-6;5gTVU$Sa47#BO`lXr2{~qXRViG%L$}cV8aa5i>qA30 z<3COlyXTOtSRs>%35kLM)MGKXSD3{IDd|a8&)XuIJ#?V(ASzKx=Q}nxU@V)x!iZLC zF*g3}2Z=%FD;lHpm+kqtiOtxaL&9P~s|`!3>PjIS`m~gV%oMDYPuetPMyupqf-5u0 z?}2a8X|%I+9mROfVcwZOlE+DTwo4Q4&^s?#EZf)><5w+2we3jktsIAdAv2{$qJ9_B z+Ee;TWK1o54d63b5l8)xQU2%GtbAW2>vyPzPxZOZz3-jjZ>$5wd6gk zdU1&+Pd!afer}~_liX?lK|4}Vv7rqM21Dt1cLYoeq+NHD(dU>a2I)^k=Eq#V>EUK> zd40HG^Sq&Vmkv?o$sD?~HytuTQxJLQ0R<&xP>q@btnR#^ySbyt_kbz;dLUm?VxtD@ zx{2buJQ4l|&)FsoOLpA6oh5CS!?b+^(Uz?ySg1XqwDC8&M86=zZyZNN_pZyjNaonD z7plu%lU7?gd8`em_SkgBGRlWtpB+!Jjy zTDsdFN&RNx)yCOae*G9;d2%t+A45K`J84m2B)#3+lOB}2pr%j>(WX}T@MAwlZ{LV( z4+r63M+o=sy@^-phw{AU7xl6q-7`G^0eeyuC`Qx|i7!~P+3o6LEDA^5n-gSk)7 zcf@qAus_?4>2a(tbp2i7)kRIPO1e_bz;CSgz;&cF@Pc4)?ZPvIR6L$~4K|@+7`Aj1 zYSmoXlCQCn*v2}Ej=Lj#M-9Yei9VbH?~>ELaB`SnOI@SKpm|7NES{Q10%eIUs&UZ$ z@s`}~_hf5iOqkQGaHe&15bd@QoQ1oUZ0lPim@HcgMW=5x!Q2Aw$1AA6ZX_lAn}aUh zKhl}=`{>o8+34w!gY?8Y9O2&5lbz1e9tKQ0#PTj#grrd5-oa#^7K@R_K3FgBj>T%P zkYl_Liv5=2wUR3)PY;LJolSTZ*g|t|2|rq$3@$4k#%Gz!NSNJ(1(mnyLCSSHQGSxT z_g_Fc-Ye;q;9q?(9D;T8W+VIMW<>g0(a7~dq@&v`?$-ruYS#&p8{J>90R{=JnB{|9+J` zSC`=h)%nJAu6(sIYxRQNz48S6{pzpcj;rfjE5t4nK6i0oDp}4y7@dc-W(Q{2PtLaeRArwz!npc z!B$eCn=0j!%HDe=y3wm_8AAE)ebc#u?gH-iO$x)6wN!d@3)Noh%eBYnB7BK4{9Sc; zm7XcLI%vWlZ(=Za9ZJV~$x_R+Rf6Br2me-;kcM9bcI8X)WRVIc+}wuSC;P)~_zU`M z6winK*~oePAa1Od$`@S?;1SG$-(invrmGRHEACF0Jde^C!OiVr@>*nQWw_-QSM=(< zLluRqd7pn~JoV0J+)|x{VZE^90&Xltq%F~e}HNlUz9tbJhLfHnl z*#12Y%;nj2(&#PzHy3kh;n`Z&Z0aX*_|j?Hd#gJYt@4!YestA#JPXIZ6{BD$+d{W9 zo8fn2E7F@)khLubWsa#ht*?a(i;lthNh~hLP65=eK|ZPhYVxxDSz3y8esz@8=D5E! zXXZ7k-%`%ZwDsAyJmDLc$->}xU;HWfhYNW-u-33Uv;|xKn94SI{|kV`IszB`Gcj*f z1hRMDL|8yHy#E|V53!%@S*3y9{Z(<#yay}kw_lRqBVIE9zj#Q!6L5Z10@^niIqAKZFD3uQBH&iUXoeFA358DpTD z8Q!_{6vQrd$jS{sv!CcH9ZP3pzvxh@aQEHR426x$e3Vz(!*1Ac6s|SI+mp4ZpCmZz z!&33)ZZln9uY$b3CJ635lCoU|2i-V=*@cAA(+Aniptv`C`MDeat1I$3pVAPs{V?KB zufgvh>gdxkiA%$#@J?$tzRvUpDybCa0msppwHhXE)#R2Z7_jaAxaFSNeA1P0{wuB< zr+`NY(^25uHC4Dll(RINc7qs+BV!kH#S15RlV>5?4sOA)_H}UD6%VVxb66-ni|XVg#QnaB znWLKE5U_{*(p%}eiwUyNda{;?B}^%|8_ivkEZhWJaIs?%?4~8MuF50XdG)UBM|mwB zJHLm#i+cctpaR>`9O?CJWKchp!sk@|jp!!B+#r)xg_`Rel{c(DoR z?($>#A-PT5E@%;#7(e4Ss#?-N#|BBQ*&qz4H%Cj6A3k(@L(fZnXiHigeGc1=fp^@o ze~*yzlmb$iP(*e2^jJd65oQ(}OAXIHP;;XZk~S$re`ysv z^`}H4Gj__g?P&4_wW%BI_+#C>c2UiP9Riv4zY!gAcd+gfZ$k*uBD zi;!4G^I;4}3kO7w?j-Gzi-dFV5)`xsNy>*zX7jfu(t$OlG=Gx|1t(smT=Qs&r&+P2 zVQx>dcT>c;c`amoYzJ*RbD3^y_9WM5l`Pj+g5`O-_|p?1ClGL2k-kfEsWLAfBTCM*p|ZL3dDjhe zDr@mBsm}aVLpqIa7x{{H)}qHd7D`XYquZ6`aPgQ%&u8S2*8?#tbr(0ob23)!DWPVYD=cE%Rm|-_fQOfVM!%{i$ugI@!YwX^iIFljTaRJO=d7T0Qw&hj zRfC&ub>JQsBdNlAD@p1D(WPt`j-7obdNQIjEApfYGR`UqBIAD+3X z3d=u;9J{3v9^0F5x#mp*_8>x%XXGtjS^BJw3vG~w)5%I14Y5FTqx?ANz zPS^U`4*0!@^}3-bNv&IAdn4S0T$)FV&oGQe3Z9Sd*v-gkD@3YG4XU!5+2~%kCGOrw z*++$H%rG)Sx^ooeDB02{t1c|rHG@fY`_lbq8)?$^qa=H-7Z>e%d|!PTZ7T(L!FM z?!edLy_7nx6;t0QAyg^{n+#6N{wZVBWsx&F5l**4D(Lrx9aQ>74O1p5;z-XfST*e( zwOrOhIyZw`a~!SVan!g;305A_d_IfG4EyUl`26#~;&T@llX-Hc#ovf;&Q0rpmdiex< zb^~N}#}UH@zMx;0k)*cq47TSyMBry5zEK%4Ts>UO2_7WZr_tA6LPioVc z3snul$Z3s`c0I97>bZNebgr!ongYgPyo%@xdelOF*bUgk-9lhQFTSv*9K%gSH@Q6y zio=)VN$3#FpP7Tu;H!egc^hNH89c=qZlP5s2Dc5v%N$#*&z6T{ogAenTG7q5^XY(- z;I&LhMBl5qsG|&)`r|$u^>RH${uqGtkMg*&zL9#r+eL$myTNwU9VAXYiD!oD+#C~m z={i&C&748f^!kC)l7s5}!cj9Gw~X;owmT?p!Djj}shkC^))txB{y5sM1nK>0Si5r| zf`SJNF7S2e91-v88|(j{741908arMt#H0d!xG#K0f8&qQ!kyw>E9UP9{S7eZwFh#V zbrH8=G`fG;fQPI75wXb&L5p|cL^Fq1w@Xy!QAzS=$Iwu#9qc9>PI9@Qm`Zpu#ozn{ zn-y*FQdZ&$&o09K*943m(@NQy0eobPFW)qk_~?^=vG$S-Z=3!FRnaaquzNdQvNwR| zpN9yj?aJSqX>$(+M_7N;fbP&5+9NWdy%s-*?^PS@u8kG$@j0;XHy*pD&7v>;j1hZi z2cm2ua5P{l8m?prZ{!75Qzk=kVHvcj(T^hC?a8i71=&l^Kw-)$tZaIWNt-L-ZCZqF zI@t&cS45=--w~$Ku;VKwQUbFFBekc1m-UDcoB&idxk&#&TwW0rGj_0emmvt>q}~@SJA!K<#esy0I& za?c&e8HMAq9HZtwvSN;?iu0dL(J1y_Uo!?F$?6cTwSPiB=HDpR)P-V#Cet|?L;4aO zi{)oRaCNnqcWnMl>wl`kskfG36>8FSm3-Pd@*OFLCE(qLTbS)~1M8Nwv5Z~g=|R9r z$~Y^|aEX(shvhS}?K1;y;(f5G?G&bqIqu5Wm-r5A1!)4g<4uJbrcG?1(!0u7DQ1AC zs(L7Rcn~EUc8U3!4`x1B$A~L^@Lx)Qe3@~YW%>Ei1h@Azaik|UluyR=gWC|hAd`ho zsAE}uN7L}QPh|WvlSc7Xl#?7nZ$Bnb9~*U~h#bJb-C9VzHv@D0i{<;JGR;D z;^B2S2ff#A#s;JRaAMzUY?&R6iSoUrt0y{1dn!(on#BY$y-B-h$-1dXnmmzuxeuT@ zWdmt>*>+ew-YTBw(@+@v-1aO#!cOJ{(#a?(`9$c`?ezy);nUl!%UL5jax|S>)N07< z;&+-`{+LEzI4jv<{Zi7j+>%yK6x|D-IPCYE4*zY{6#ip8wEAShO;1H~Elr(84FNf2 zXs|&J*IA%KJdI9Jra9aFDfrZWnv(N{-qcxP*Em;9M566EaX)BYdxd%i=P+M$CyJT% zgGvmhN^G3d*`!CYpfnu$NBr&b*nasuVQbJqjE=neizuRn}OjTGoY}0 z1+GP3!lYjEyuaadn)*kZW=21!b4lQPJI%SC^gc#giSxRgFLGnAz-s+}BG)T?#EuMw zy>{bR&VJ;tGQ-fI&AfN*Wv)By4eyohz#o21<6X=vcvis}D%zz&*YyTPZ-eQUl&y6kPX)G;;yN^cfWWTBg+);bgNZ~AbzlRLxu#`i3fa;QW#DxBeR@DY!lCBI!x=kBs0CPn+?d8&*NhAdpAwE)^~u zHJa(aS=xUvJCM zSUjFP+L`en&3e2pQo`Q+bXe@w=I+WGI3u!@zDtz&Exl9Nmw5_iheQYH*B02fjKk{N zPpF`A9QV!L#5*mP@Qm`q^lsEd+%CTkXQyUzRXhsStFAn0j2G5+rr~>36PESyMU3#w zy}h)Rjr`3e>mC}>NXb=1v}U8VxD#hQmAUG(g|sUzoc`LE(9MbrY&fwU>gDOE*NcN~ zS$|xQi=`v)u5+i-Z2rbHoV(7Nh2t-a$TDISCB@x`WnLJnm4t_P#0T6H8GMJb?-*w; zI^ZS=uub2NW9+P8C)~qG^IK4=(t(ZSpm_FgrLpO9_^-?inJbMa#g#KYhW#1$>>E# zAJ1T4N3UfE7n!5@mOUo-@j^`xOQyETn*BN4z>+VkQy@*H!}G%^*+L8I=Emsh=nbDG z_sQUm2d&wV$Kpgz^;Ciyx@shl>gHCgc%{wTZ`*RqcrEGnC7-$b(Hb7L&9y{gu zn6Z8MXtBewpS*@n+pQ3O`&D=`^)B71wPKck3SpA{3g275qx0-TS~*(>LmoPazgq$m zw~6%MD_t5h<`U{Ad7x&f;HmQnERWcYo+EzlX~4 zgD@;8n+_LGrrSR)k+4@CJCuG>)gR8Z^{?6beK;y9xYvwwPr!bG`xgYpFb=0M{lKA(H>Ut(~tH!cR@M>61b`~;?r zD~I<}N9=YVBtFkHR;V8$(XFhc$F4i+*vwJP;XoJR9;>B7u>)V(q`-Af7*P4Ro|ya9 z7KSA+X`Sr~B-Y%-g(AWE`gnmf7FE)S-j4V&+ZRd+%i!1;K%FtgG}^q5ss!6(tWPAV zgeV~{OpRGCIK=L468FfwuXtFfz}L@ugjkVj?bG-Rq-n(Ws1K#cbKB{>O*eSo`ayvT zopdfz0g=kBH1&n(Q5}y!)bHoG=&r>L?alZMe+6z5ZNTkgT)5Xv!PIe#A(^X(Y0cXX z7N9+hmVenn`Q^QlyF?CMx4opSUrO}2B%ea8pV9=u;8?rf1ZQsbz#=7CT=`W`=gTk9 zH!`6mX?A2HA4BEuFTs}0!Ni!k=r=;hW(B+0@N6m71V+);kVJalN zda_~9seYXg*(x8Ubv<(^dkiB>)oY>8KYQW(!2RfKPQi!6=aJK* zFTL2WkMsmnmG0CQnX9Xf^q;k09QRVE9Q8AF!&UgGpY%tEfgWN8E1)25IknwBLg~pn z=zN1YP1vOghw%)5Ml-USWK2^=_|W-_`LwU~J*_xiBX&GsK02x_CBc>|w~M^|!C+h! zzL+r=x|8$vJG3XdA4aEKr7Qc5uvN@xLq98H=M@E*ojpP?3aiL1;0|e4R8d=ap_qN{ zz>{s!IMOc5tfTgl#e4B{H?@V|M+=2GinXVfF}WE-$aC6d`WO>SJ^}H9HT9YJ<;uzdtj0r~hItQ%=jQ^-epO9FCkyZUp{Hd0Etf`BKBpH& z3HW{V1P&CR#Z-d~7B_1wZJwP<)sL>xSnLH(m5 z=bnym3UbE1Z<8>mb_ux;%%Yj6BnTbMnZ>GXn%N+7WAf*rD8EcFeKIJoOo4aJX@{kU zIoA+;xgT%);QcEtEP45c`i1s^-|zvLtm@2qoQb9hE0WQk6^PR{hp=2ppD*{d=XIZ* z`B7&lZg4b`c|NhF$+8`EJpC)XRkuMfhtp{FoQE{Db2L0(PDa>GFDzegh5Nzf_|-}k|1``xrlOO4(y zG(}HuZ`|oW5XzG`QgQt(tT&CpfYr&;#_n-am){Z6UICtzky?^U&AY?KbWd!~U+fBBsqUF}O6nfIyXdm1TRyG19cBc1|2$aB-P!Y>n7i3Qt}*>~J$(#3XU^fr+>u^?6(Aj|&nTWsX4 zBIb2*IQ1>wNL}ygQ^j3v1YFC<(?0p|P4I(++tS6t8*JBtp7dqgIMH=bA>C;utYzL5 z!6w|rmgFkY$x3zVzvHp!9{wfim3`#=Lk$L*PMEar8J%)i3MhR?=KIT}xxg9^N^9xq zt$gaVe#G{F7)|$1ZJ>dprikAwd~6f$@#(YT`LwaB{Mw(hbSXoeV?TJ)-XKdp;p=$5 zU1lVIm!Z!0_ijPa>%EBGo*=SGet2Ny3C|CG5vW!{d6!ZtQ&Y$lrU@T$qAa>y?TQAd zz;sVKnQneU`j)XM%qlR@qhoIJ1>&+6<@SQJaW-wu)btEfx& zyETj3;xCNb)PSQ$*22sp55ISfz~AT~9D0||9mnqB3nWSW{*o<}a&`b!yfvo6Y1xvG z)5kEA>57!Bww8x2_<&^97MeQ!EYm)kBWZ2B!hY05k?Y_bDmkSh<}welCvOuDFJDR~ zr;KppgEy?N^uTHND%xrXnDEyaXDao^C_B$wI$QH0=Cp9;8*&}E{> z9T0}D{)RZ+wug%Pbg~l)QSd41Ag^V8Q8(rkU2t%wxL@7KNLvnrEo>-p{&NbbwZov+ zG~u|cL7tWl4Ug^11|Bh|Nj`5`s^tw*y{Cig((PzFum%rTex$kP3+avD3Q$um6b)41 zEq0TKHeA7lmoj|+Jb!#H6>QDSF`_SK4eg{MknYaLDy<9R-WiGE_3JV8{2Iy*@6TKY z&gIf3HJ<4oh(}JJ5V-9xdig!a9OLfH(Org){g0yaaLDm{!?;3QX-j=071Gdp&wZU2 z5lt%D8QFwv$_x#fQpsqekTj4|l&p{l(GZf6RYvyicYc3@Jn!?IbKlqX`3R0{_dS6W z?Nvel9A#m5KN%4Php_El6W$koqSf|l*q`%~mdx1#``2;U*O-WrGxKP^<`68lpMh1| z4Y5?QgpR#yV9TypvV-r>NH%(oW&Vb8n4qYFLwyx-B(0Dnz30=r5G~p|upfVUUY~mz zY4I-0_v8D<`M74_h6{T(QT2>P3{DE4^x?;)Z)wK9VPf1$pJo`T?kI5=If__&RQ0tQ@`r9NJe*c{$52w92=Kq9Z z`h<{czY!Q7EcW#M1|Xtt4w!Esbd}~|`syw;V&`p^eWQ{h{OzSqzfGmdCp4w+TdnC@ zvm71##Xb?W3KOPB->6Fu*#1K^ieP@Yc^bDXKwyv zSDo9~r7hYN?x{?=%O!9P4HMmi-I58Wp0upTa5nqO0QSQ67GGFV&PS+a^R0#!Fb;IV zg20go6*(*C1{GA;X#u|lQ)1H`3iUImqhc?XG`Bm3PL~sIs_vK;@Rt;ueo*F6KQfZ> zp`1I#q`UAPd#XP~aOM}#_=mBA#qm)1s0L#^ra|FT0j*pl*nDBaJ-2WeY%8sCeL^MO z6)coV%f3;UFWrQn=LOCBdYFDaI8Q0w^N6Vg(2}>=c=k8}M)T&Q(q<4oy&Hy}%@&yI zwheEFmcaGxE##}2<4Lg-tSZ)F!l5iCxt7K}U;ktBqa9$Qkb((+9*9}tZL~T3MaQ{nX&( z^@a?034T?eA)*sRudBTQH}b4_)5V#*{MCHO=+Djtlq;pbHK1-oKDN!o;qa>x_=^IBy^e-XgnCXsp0hW>PNQk!6lJQ&@KT=4?LUI>OjDs>wv}gY-(t5m|nwI6VSAFB~@k%7#JlTfs z)~5XB#^;!jDvy%$2e7x?22Wa=Y0e5WRF(~(!|rq0^M>8jcIPSUswPK1|F*KFeV?$D z!Winb|Ad$ks}Z`eiq!q4pyt+X3Jm3Rd2TPB^7uXGj^ByO@B}P(O~$6$WEiT%O1iBT z{PItM)MCJR%+KjuS8!sJA4$397k|G0djbz{75jqAakwY?zNcJHQGQ`8Rc_T2%2V+pZc)$G;-}lPV zs2A)U=sQx!qlUrjMpMg@!O-V=H6r^mkxNH?*LUc4@)EwZ(=XPq%mnyJMP*Pg(R zm9;qO(4CJ`SLJRkAE`}78$*)@!0dIKeQrsx{ULo~r&j&u%>#6#2c*{0sYkw$bhYq) zpPWkLMDD!y=66^KN2ih2LcT;gmhana!;J=IQR|!>^#85M|743Du6!h!Uw$jxDHj-Z zo@VcS&r)ZH0-|n=g-*P1A^Y^ASsUk&%;w`1{~?nNOVg)v7zqcL|)Fm`A6Cd>RznjpT}S4z|A+O@MZ z{o_R3aehxYTSEOWR#1VtpCrtpL$Y&VGpqR^oTW1vx7M`fror`eB)O1=JExGdf{SpB zt;NCAZJ6=R3oaT%vG}Szre8F{pvQd?Q6r1FhT7b$Dhu2;rhEx_}NzBf^T_DqZ>2n zQd)2RT>UiyAD>0i2|2+pQRAWMCcMh4l||IPWBYP{urzs91nD%B>i0PM>EuQqLrU1$ z=Ti1;;6|po^qwU0o+1m+yocWVGSTi61`n6PJae4^mu^$wozIJr5i59p;+qj+7>-2` zIChzK$Cbz`I;9xRHa5qwoSk}HFR=+jCX_>?vYM-}Jj8wdBY6EjpX`+uu8WcWat)LA_7P6r7Fno;lvCE-kGRlq85Ye62rL>W?XliNns?Dfx_P(> zMsERDr5WR@MjH$J{*_rdJZ7<0k(jX08|n`h;C<+AdN(wUG~q^5*VR$WkLlDk$DC}A z=;B)47c$NMM$wuP1ct$}5dkLW-^nOVUJssqM=t^DGzE#A>7kwzNypfsa>Cxz>avCw^ zKiWKN7i+pX54kOZ(RNfYmChN7d4Ct^kYIy4ekP^pYv(?I@z0gE=JUVnAoEeq$T# zr(ke<80Nf4!6wf|e3!L`U*<#Bahkw z4>%rFp%G@P>~Urn`%um0WHD+vDgC)3_8?DLlg&(audSI<0@C^66WjRzsVkp6y|L}2 z9!Af}r#U7~%y^;^QvM6XhM%=`B#@wEH4{12JE_Lqn8taTL#wd|{~&x_)5lzeifPmv; zI`#y8`L>m^uH2v(MUUBUgIG#`7Kw!?vS{Mo@$AXcNY=;A9S=rKqCI__P_$zwOlO*7 z@{en@!7B`2;T2GEevJ1+S5o6wd$dmr!lJo;(&`C5(v+DtQk7-il5CM@wlix%&l^wZ z$(SjmbZIwv)EtM+`*E1M>N>4mJD&z!Kf{_G8O!)Q62)&Nc%1x{+GlrC_^wpMdtAd- z`Doz`pG8F;c9fQ_MCI!Q7p5YXbQhT6;nzOMDLhOw-i1<)%^jxlP?shJXt8*gZDf4* z7Ab3lQsTWGY|5P$s&Ji3B3A|d-Cwa{NE0`*NaH1c4{;^oysUT~g9}rAaj=&+bjF%M zdqX#*I=R!`6{|7Iyb|>rPUBSH?J#(148vd1bYPRES@ z*h%rOv?H=g97$t;Dp7_{E)^SVa<%EFVU`&Twa+`SCvqLMKdNDUz6|dhQik%~62UXp zgISmdc6a(?aqn`bzI!z7%5lW-Q*Th4q{Qd-zlk18E{nap$SXSy=dxebVAZw~Ta(V< zj?Xrn44;J6V-`ahhO}O{mTkO{OWW^k5`O6fo?E|&%O^gkr!IxmxAPr&ob*HDx5L=l zl!L6$qo^1hj&8>5&_ADHW`h-t=sT5tJiL(>R9$6j7L8&Df83z=zg(%^Zw_r66NT7q zGf@5N3F$B0FEaB*c=VwLI$ksJV5$`7;HyK8d2XtN!rDNiZrC!f+QC@)tuO8lVR-su z2X+_@grkZK{_GMQxP|?Ze^D1Bw)drO5m^$GioNWRfir$oPNMD(Hznt?j??)ErQ|lb zh@3wMle~%&D@xkV@&t!kE3+Rod%P0t2QS<_e4bLJ`lMO=3QGb`W4h%|yr1eW?frzL z8IP2uzAxpm^i(L+3ua-m^JyIV;S1B%V_~B4i>+VzlHOj6MTti!ntCk6f1ho!aOW;m zSj_$Bldjq>)w)_OdvI$nMIRTH{*D;Gp-La#6HU;`W1MeyLK z?n$jgXO9m0G&R$UxjHb?y358T*3x}>Ed*S(ML>)2S0DZ<_~FNCY^xgC95_tZ%jZzd z+#WRK!hHHJ*r9u>&ckcg4(u}9g99f|qUpSFNQpgoH|Ga*CNGz6-j?ufx&8RO`d@I| z@5(0+8q0eX>hgudR?|i2saW)%$d)L-l8ii1BN-hKP5)*tmi#*q$9$H}r9#6?_9id& zXqo7i4AV_zk|XD6`oB3iBbYxU(qa%bR)gOYdDEkFEJ*3iKFZnn3dMJO@a)rXG5Wk2 zb*lBIMFaBK(mx&KF+@f9EQ5qYz!WX>!tnliDmJn*1o^A-?xsFG=zAjGt~`%PqiW#& zNtIoyvt(BJiRAv}7&Qdlq|~)qFciDE0f`6TA6$WH%}P8mB%V*NyuxoA$l~J5A-MLw zEB9XXh|Kmcz^!g;v8?p9pw4z4V~~ncqs4UOU5mHvOPzMf)PB(M))Ag+To>` z1Eu@@aWwBLKBqNe`1?aBUs%odcub@ovE?N5_di~K_$oS8PLj3pJ?dN8LT1bR;lsaZ zsK&=&)Yug4OHjoAB_pwH&m7dHWb@w(yE~_PPo7 zSO*UFzYmw3*~s*jl8)+>$`Yb>lKZF`k|yas)D+X7PEogy(xc08nqde zQZx6oPu=%kvTv6n4QcPgUr!Z&Nxh{AY&%E`R(BWd(m|NG*MWYmmly6xFX6!42@ScE z7+HG>Zmt7RpS=WME(aiaSQLJDrr@*w5$L+lq&~4e@JTAcrjM2OH?3mXB8!2vM4XdM zkINxRH4b^H#x%!Up0>{1ND;R+n2O3tmV7Rep7z~EzvXR_W_uj>-kMWY)kE?ewi?^R zn#p$QO;Y%-gK3|`>3PmVs+ajhx-qA*{$hq;q0HlFPYZ{d$U7XV@y3g+YILhgNB-0{OYWQ^A2=IpX4Oza$`4@g%KTi`q^e{|(SEq&NP*OMX`_{Hc=6qF2m!@{s z)ke6UXI$jH^p0}nH52*XL27)rfdfBi>A=e{OZX6*NzyzL{tGhkVZw6c9125FksP)s z<&ukmD!SF3gR)&4?k(%bhg`JfZdq?Ip=l?av_G?g_IxsG^uyH?s^rz@G#OrZ!O%L9 z^R_<2{KGG^aF~Md`D)G)QqH_bv$m7d>QfL+5iLEw} zx3S`zvj5}BBf^lpcrCn5wV{7>19h)Hikszm2%Xx99YxlXFdX1Cn5Al#L zDqA2me7YSmlRIhkDF2q4$-HN^%s18xZJs{# zuTxh%A0d=?s+E$`?WlF^6S^BF!*4->Un@?BqfG~@zq+94jytOzZX`WgK253|?IfLj zB!?-ST1~TRLa<6O8^#2-k(=Tp(UmDf{@pX!Kg%3po~3Z^7KOk_DI^IQbnS2}oiQ!I zw`rT0QpH!c|Ir6B-WM!9;2SWY`6b@IuC7M=vY`tw+XI z!QC}H0f$B5kgQbU>n=O-E7$vp>_r0Yd6`cdi)G=Yn#Kb1BPG9bJ~G}n92ryXc|iAj zG+^Eh8XPeIFNdsy&tPvFf4ZKc&ppQ13zLz!a33vqa>pq7PTF~HE}5*>;U5-?o~PG; zxRV(w>GIT@8Ty#f!_B(rb*YI8?@R;poF#mF|5(wq!)%xK04&*I2(K+G>B`NU5~;mbF``&+gli7plu{1npIh5LsP*7qBeW$}H+tHObkG15}ocr;B>m3+vugy1y z%h15c-*otG9gT7ipzoR5tX-{y&7bj=sx34z?70R$T355%b%r!ZxNDSm%|?T^4(@7p z(ihFS!gcnEW-LEUy0!*_AD1nhIzOo}@Hsa9e1=^S9cXImia!enV$zZMQ2&}sMjmbC zt*HauUapvzpigSTMafke7_ziG5Q=!T(Oxdy0(y=fyh`1j-5+T7+q;O&bqx+qUtx6)H64m zu6N9$iu3a+(P}u1kB%V^{cJXB)FUR4V5E7%m(zVs-9i^!DCcMDrf}u&E*6 z>aEE8-Vt*{lmEEs4}=s0BpK_mo>EQVGlablgSFf%O7kA zjq_b`<8msw#AzUAwll`*$CBTeA1q*SI;~7RL@9CdWOZ{PX^dM)rjyEP_Rg-@Fm66g z>G@bv_jf4MJE4!0VH4roClwQY+}R_IZ8SQ$h0aLA?9)#jX1;SvX@$v!1_(!HpJ`QFvEZT>3BoTU9Uqx>fQXnfCx zi@f=mP9T1@BF4ICVN#D&>ORL`@>i{zd1q`A&U!ssFwKSfs8*1{pQmiz<{hG!(oO^B z8{*e+aR+s2B>z=ih4)}8Y;=0?-`&4r8Sliek4NZFfuHb)4kX7BS;)LP6N+vc(0KO| zuhu=pig`ED#4?26&2iSI7@=iEluU%q5be*k>nYyiC*ri_@-$Ky~`RXy)R~WL!v-MU-+5OSKLYIAz!(@ zlT`nT9vROhpp`f>a(n5Ki3i3vO6TF5J zoS1Y9k|l)*H1(nh7j?;Lof?_#+eWwEO<_O0eoF$LrqJY%=c&+A4{hSknR_{x-oJ^V zi(}%M$yz6-xoHba*;C1WDz9Qg2R>k4N!=Lhzn*zdj$<0CGSc7n&qe;?5;tl&fViYR z$dNySGmo=qnwS}y1SF8sNgcsjO`sgxiF9Mx7Z!F~OMEva6mV7pw;$!wMfn{1U?ZHn z@AB!%ucvfp;3~3KlhRF?169HRTA4+-ZV;*|tc`>w~{C zDRk&y7G;gtNvS6evdF4!tbT3+Yp#tW*TtY42NdYbrYV$|FW4^w;^^$J5OD?%C!g*$ zlsdlzDl5v-`+#6kY!K&i%L#b-$quQ40g^nZfSSuj(H4_Dw4S~U&yrSjk&)-Q_OGEk z<}60e7!Fp_A4Wc6#;|IFWR2o$iTWyY*5+g>QT}&S;;a_L_KvwP2_JWf&6zk5Pn+-2 zj<}H+9ub8*_g&zVoQD~OyTqB&iq2n}LKME8WPBP?a@!E2Eq+NRv2*0!zGldW`dRUn_3ikr)02PBH0E1wJYp5$Hgw!b zcykV!&|U9YWV3k=ZRxciGTD8R;b=r}Htpd)PY&`_>2iK-ahkaAUzL>lcCgZ2Nr--# ziusc?acI_Z_G)A(Q>@KneUBFqn{bw?dx<`^`2H0L27B()8kAbB##X(v?B&ctdzq4C z%&^WO)B5df#vnB;&z**Y^T*@e_o0aJ>`71ccF^%jy%BZQ9}QlE5%x$AOK%^bN$;Mr ztF^tbc+nk-y0wRH92z9emhqL&ZCWagd*ng>RUP*KO~{w1I3yzdlE`6?HRh|_Z=u0k zonN1!#Z|KIlI*1{%H7{Ws?3uft}LLv6A#mw6<*+x?wFsl79Q_sk=3_Wx{wo!9jz;A zrP=_=(vPloCbkXi)qZtyimN3hOAT58Mlwkw$$6TU{nfK)?Q}l4OS-Q=i%0lQwXx&b zpNCzcFtrC8rRB{gjkJ|2b@Z3Y3^A6*7lvWp46)xhBlI?8yd85 zkA%Edv4{^TEJXbuxw+QU!xtXNNlFtt?5;=;1CAQqr=vF#FyZY_3|*Rpi~owC``-ez zWjLWjV_uXkXmwGsc9>m+yx>)>YV zho{VqEvuFl9E3q^>Em=ZAU>GP*DpiSv|iY{)E8Ve8E(qc>X-Ysxk`iOKQ^qBdoFoM@(_Azd#q_OKcOgXdxaoKWWW?6tU zaW$}VKZ&-nhE!eBo8;xLP|Sy4?18~0NsW($xrv!*s@+X`zG(nG`LBScdMlD-nIrq_ z;72QwEosx*Y5#TN2*HmN-kLQ5^kQ}`JCtX|Zb)WIc!RgdmcC`zE5EQwzPDjMEgV6I zmf^GFHl&E(aGG8!^1O z$DgvT6ws$rxB-q|r=Xxh+Nm5u+CJ|o{J~7DQ}`%xpQcD7-MufrFIb; z-b&(=3M9d8iS#jLs$|2s!4i{*6#G#*qp9X@2-}yfKu4e7WgAwWku2zUTH>`uiI>(I z;?Wu|I{6D|!JS*oto=LX{Y{X(og60_shlra_*SriP@J{c1v@`6jsa!g1?o8O&m8%ho2x%#_qJ)^Jf%GIH|~r zqmv}jj||vtH5DdZ|5vg_vxW30ZKMkS23EBzfIcqWCs8kSwNqZ!Xm7Eh#{SLHV*3!X zfz=CFEV!kD&~JsbHO7V-X7(WG{hIhYrj(NMMiQTyj{OdrFiJF}1vLjH$tQQPpSiOo z%bPOon~sc=DETF@-t*`4BhkLxP)9JcYg4GqHgA12krrzjuSfr#vU;k&s7*B5aJ zY4t`-t~s^KcTo6lbL8B8%ETd-<|^$ao%sb+we}JH7WXsRgjKYu%P?BL*@WDMzp>lW zQlxyIic57zp(&hT2a6BkRQ42XeBv*9?m4Kci$}%#QPBINi?}LJ$W@F+pR$KkR`wlh z={($S#$&+2<@}$!7tg95&4;(_k_0a}E=d?z#!6chaBclZ>UdL1U5&4>q_MXozCPF3 z<9UB*qx@cy54_73KV3)%G!2EBxq+6BTFb^OKV;pCF0&_-MJ|g6v)k9Z(ydbklKrzB z+4Z`)Y}5ETG_h?9)7m+W1;2Gd#VIS?bu@xap9q*4BqH_C6zsXBBYjx@o6l9P<7GS4 zF-+MC-TqF;-5ENVD%dW=UK0v-X>bL91wK5s2j8HQN3`rBjaKJG4G6-DGBoy0L zxxj3i9Nzy?#OdX~$@hu^E*{IJp0)<~>o5ojN&jfoW6+i%5$s#QCR{5YheX{8*kkgO z%<^{A#AbauB$rC%*HbAqd^fH9poIr@GWhg7o5FuCq+@C)SJtmPkhBEU)3>dvqMJUH zTEfHW%=J^0>y$vs(}Qtlk3K9f8q;TcQ@${FJnxt0#;5jwjVWu|k#z7jO#Sq5DR(Dc z8K&Z&n2EEy!_Z+i4`p+6A?;rPn~kohDt*OYv|i`NvnKNPCmV42{cFJzFy(cEu~9yL z5S(Z_>HK@gUVoXyY{d0okneQ#5jpT4w(Du76-SqfB?uF{tYG07N@(gM@vru06^X#w z`-U`mUA1K6bW@Cu9*JjfR5AH%5{mk5B-u!jF&7zv4P(8LcJm+os!Yf3Et|0OhH$mK z4~Aq|HSPO*oEhcrknS$rE`8K(nsnC4d-U(JB_h+LXf+qkGtqSlzjX-lMgK71xiy!J zbmGUSTe0QiP1uLAUQGA&Jha}Efl7!a1&aPot>`P-tV_VPUz^$Ly-`egQdc_r!H4O) zOqcln8N|Mhvc!19Z={mjLKHcV+$WC|juPQd{JT(ctL-!Et)z_KZX#E1v(6yF$ zO8z+(Ct_#fXyZW4KkI`0*FNxFHv_uAJuvKK9gS0TpuZQo;BIpKx zYOhG?YaTrwKAsM&ct#Oh1M&08CX6jwkBblbqAkD&zlxR!$KEtFt@X#hfdprth1lBS zAl`{5s5r|My(0g^y}M#>xy1w>b0?zlrYp=G3sLptAvRX4@C8lD*y)=Nxx)eQ)`_H- z{+G$d<2Ie@kmHFR#R%-T4=JYIc;lb`ypNU!AC{TU+&iY>=&A3p>>>Jv-+xfyFyGCQ2 zZ%+kA0GohDk7O?h8_!Pt!*v(8L`#$UIpVtBXweE9FBlsqRO! zdwdtY*O`dFGz+=Gr_lS`W0ChO5zKeMRrJBzJC0zjfQ@$;N~K>aM@hpW52NdsXq$sQ z{8}|J!XcWDw5U_~^DDR-`3+ybcIUah&C%n(a=NJM$IcprVE^%O46oUQheukeHfT4x z_B{pqmq|gJjx&V`BiWtqmZVePE;*p0%KWQu(u;x9VUcQvp0%@Sgi#&^e0W7>*E8*9 zHF8*DtuGBTZa}X?$1qf>fznG`QLlKDGCft9l99Qzu1lYivXP=@NYCTZ&~4 zmW-_@V$FAHVDxrOE7!&(1;MPY?T6BXe%Np}3PG>(L{w@iZhG&++B7YU+2V{4v8yc9#&Cb(GOGodn8WyH@OM?V(UCyvA{TvEE}c)g){r zGd`cr^v@yLJIW~37Bhn+XSh$AhPAyPAiXq1xPSDa-M#`2&Y`e47>2COAJCj}7Pa?< zAIL+T5%%uJ*+=KlVjf7=Rk18qBbYLc+HmcH3ip})9P1v6F6FD0kmr{WZrxx%_2MHs zHz^CUTH~mn=^rLrxu1%I-%+sjZ6>$WU$U)CjcZ!i@WxpV+~?(JI%e%egW4~UTiR|^ zEJ((XiS>w1Y(-UhAO3ZOA(!pa2dBlmVU$!biYxEXmBCwRabO~I>-&o?SgIp&jSIR9 zU+0n~4y2l#Mfb#3(L=A0dr$^ zULKuDeJaD~)B!7z&o$zTHx2l`u4X*>PAz3Bhmyfb8zf>O-96mMKCdfa9=H9)^BX{i znLjlJ4#l0)VBFEm!Y_9lSP%arIJ}*7Me!5$)qlaJJji4XtUulRu#NJnMaH_t7wfk) z(jOZw8o8hW`NeG*x=e$wz5a{#R>@+-GGLiT0v6pp4w*Ms#rLP09B0_#Q(xg7IGiQ@ zu=j}c?#?`^T-H4n`cRP!o;o{p|yCQ&EY*ZMFPKJE`0AMXmE z4UcK~8AWXKm&JzhZS?k_Ih!@8m7VdN$8wVUk^e0jiX2c#O*XyY<(hz*W8UJFleP3i zwVkx;vAuNOUj@9A$)E(Qbn?G(9JK+$eS0MY6@pJvAvo}hlf=x*=cr&C#=_Sn0^jl& zrD``*?3g!{k$s2sL_ab>`K@4Y?WP?IEy?EDcXsRVdn$YHf_XEKq4VKMG&mN+Y*0PA z**`%}#zXL)2k`E16}9OLX8=~xr~V1_RdXawP*kFll@dH{t$1 z%5N)Yrz5l25NQkRb017w386Ju}gRPjiy@bqXcR`mP|HwGl4ZSaS zK~Vf48ojDW(q&U9B`ldLSXbYeTW&n7AE(Cjf9%B3jps#x{-1xcp=<%w)gQl`ijy-=0AQX`iTFLpV_U3n*$_v1D&QYx*X3Zfr}p%}NXU``(bsf9z*<`5~-ygCSj&)KR=nF`H?6f!$TV$fgbRW}C(~NUUsSNq_zc z>hAuP=^W_76z+G!kKY5CZ2dd?uiXpyfj=4C>r4=Dx2a&OuPD)$G?B@eqA!@zJFu>6 zI-YggLzR16gwh}i`$aFVcGoh|U8*J10UGRsaQG-^oTLFi?WpbCda7?-LZ@sdGxIp% zQh1bB{}$h?k_Y3%JTpV zHKa^*CEw}4m~*)d`|tHQN$k@y=2|U!IALetI{PmwBdn=n=N7uel8R;fjc+_`1j_Ud#asFX!u#3SV!5{pR`hxWkvkMQC z7#flE2}+sI;q6ZRvBxvI-EXwu;cmk1_0C{nZYcb=72`f?YM@F7|chbFfW7hdYlc^qAOh0!g zF}v1}?0sk*^LR7aJ}{%fesD=27I*dtnu}9$D>?(#4KloSgF4?iVIE(aYa_ihbDY%o zc!+dRlr`y#{N+oB1k8OL&-!iXLw-%!WVh`!CaPp(i_u-wi#=DB)PI52L$&f9Q*R;OSlo#s*Y^QRvRi$ zF~DQdMbn=5j*>h2;iXCfJ6mi=o8m5z?u?&g>8pO1~SrzDrXT%JWwULK&;@(mIdMY zL>K#mCbHd6+F0_>uB6cAJ^Q5knho-J!v-bwqo{0qN*iZN569^<_nH~3|HXV3SGt%o zj;*CanNs$8h7GwG8Pcu%LY5Wa%FgVJU~gL6*mk9N)Or6ReN}tIJdamM=6_$xwsq}b z)BR0Jaj-rmtFNM<)2k`zTL=ZUuc1kI7w}^f7x1Le6S{XLuW+$tQ_{J6pCe}mqWVY+xBlfcLBQuHXCDETPBe}EY ztbJFFSem9^EaqI_nW@HAvgx?bB*w)||J)RQ*>OA{?lXw{?24dw9!sc;=_=aa=Qp$K zC}IBNZZLQW=g6!WT0Y`68?wLF-szVc8@H#G?I?W7vRA!j%eA!Vnb|$&_*#yFCo0gr z`?0KPlss*FdW#j--C*N$kFc_q8*GjJHahL5N@nLvS@YMvRQP!W1x9$%$d!rq_qP;F zR=C8lgaXd8o}Fa(PxquXlaI2X33Y5n`XGwmyPAxyt|v3Y0aTQ@mYpiAWfR~0X3O3= zQ9$>1Y}`XxvN%!5Y=4J(*7bTctxF2qG`NxVpDLVy(<6lcB#i1TCcyT_P;8Le zV~BMimCx)aGRn%-?fn{wV=sUE_aClG&U=M3UHx1MZCxWdFj<@5uQTNnhtK0Z{8aH* z*;VkdT*bFuU-ZBfuy)ry`Yv|#Bj(6*zs6zQ#vuvmXTmZ2$|9T>-zuNymRNDu2|JG- zrOi8b(I$ffw5>-DEtcM)*8>e;Q96u=sTuL^N19B{D=J_ z({ep*Hw;%7!oANc;RO3iPhGWf(QX6^`y}B09W7p<@E^Wss-siv{onuDO{RBT>DhyB z?7G%8dY!F;KRu5kzRiZbL_c_U+Z@^yq9mBpJ(xa%?61s7Mf+f}tBeqtt$%VnuuPSE z)z3m{z!YiFx%JXH59djBM~QDyP6;;O`UhgHk_jH9REvNRo^f(2t;urn&g>!`{V-C@8&6%s&yb}5 zKMc`(fYd8TaneK$G9Si};e`UpMptz{=x=`>)j6CiWjTO}43}iB8Mc(i(DN}zDKl*b zJzLO@f!8PSyX$xG$na%8^#GzDxu9`iH`IO^f&7h8SQqDyo}XM1@K1`RyByHXaxkq` zE}@`V3h>{00<&KF@|=%H7M+V&EJC#GZG#1Fi7sghLltcCQ)Y)hmM=mW>d(M)}l z9X4x(Aw9ny&EaWaJG0@?)~DUsCbtQiVqKmYiT6qUA7ebXQ}i@gt7Fc-y|@Rx70|B!@0BxwBy+p zvgoRcW&Q4=PAeK^BD1;kcm$P>FhOX&CdNMKLuu%4{EYolWUlFQaBlD41i&;p19@L3lV3B!g6*j9uF+TmY#=La8J>fDL9GP zm0xJ|KH=P`O~uMmv5$)&#N1Ru&(u-a^w$dom&I;(Q7N6s)5WK)Gcfy+c=lR8Q`nwn zd|GafwbB(Z+@6kqQ=G{{CxX@KTCgF84=5f0WJE z_KPI`yoecG52S1F_t3XFWs<}Z+N>;XsC~oF8c0r>@Y9u|FJsjYHA<1_?KA>r7begt z(N*!BodLP&Em-*HxbS%dV9>?G(5l#vLW^^#{@cVp-kHs!*0qT|-Uag3HALPAA1vK5 z7p=vyD1A5yN9BboF5?p=-^!&OP9G80bQK??jzKr_3_0#G$FbgtI4T%j>W@~^1vzta zbl5}N)|+90R}!xM_>D_iQ+S(u3@?b>12r{s?iS+6uMIv-w(7!DChvopfn9KO=2NQn z?Fqlp!ky&S1)ELQQxB^~%G|Dx+2RhOA-DoY{ui*U-JAzDg`iz78kfA+Vdb0!ln}UI z@LoR9cDXd9IJQIn?+=k#7(>w?bLs7Y{wO|UhJMjK1QYEyt;=u1B)M|LYldUu>Nk8; zovbwOYENmmYwoyGAQ&=}N^m1K9>)KEpmVq>Kj%@1U*}WN5Ofyh15C00j6FJnv+=h} z6?wXZkaFFw7gBQz^q`E>WghkQFt&V4p*MU zV{1wnDGSad^W03EcbQ7^W_q)!c~WK+HXeOS-XSwsiFZ@B;FG_s;@xd!_|-Uh>CCN0 z(!^Wyk6avpN!PxFRYKzduck^jJZft6;8l%#Vgo0{lY2pHv0jqjM(EIUQ~Ba zk>>r(A}5Pil>DTQ9u!=lD_82M@%v4R4(bm3(Di8Of0lCkszR<<15byi*dWT#< zpy+rxh-b~|`5xi6n}G|8fw&x`2p9Fy2>7LdFB6RDve{vhnf#Evw&);FTOX{<8RzG? zqtG#o9owZ#T7h%u=(8wzRK#GF@JC1etZOyshMZQT#TW!zk6*jsp9y zlGv>5kaWwgV>8u`GTW+YtkpV=<^C`e`>RY!KH|+nimiFvvBA8rMSuSE%~ev&`a(Vv zx(i0>E_f`-!D!`3I7F|9a^-5A_&gs!=R07WLmy-eyH49*X7I%(J9x2zBX`!UMZym` zKHlypB1ONiXHO|cM7^g{`G4%f`6YBYNH`O(6Y_1MX{mh^Yj~H5gqTz4`Qs2`Ow%d5 z>H=-9Nhj~g&QOkiK?9!ZQ%#`Yan*}1^}OY{`s@n%X6_-+&hHdDdn?X$Fp=LpPq{NT zV~_BoCJuFj)mCNf>V1jYpC^&L-Y}`xNh|4^H9FEF1y`K32!ZPgH#}-7Ln4XH=YR8Y zJ^v@xJ?qJ}T+I05-XRk2l-CmdqJ_-n?hfkq>>@iDVZ<)yzePxw8+as>im$raY<25# z_CLqwbnM3{uK%t8<0$ulv2^4%tMCEGL@lZC@Rz0Ywbi6 z8V$`V85*cW^B~-FU*C6ke?I5kd2{yX?7i0dJ9-t8E>1dgKW8NRh|T zGZO+nKUpdvw}zUz(@y1lc|&dQ>7^F=U8UNR&r{OJ^1!{ryv$VXa^PHCnfwP*0yvZF4*CfJF`qD}#C26aIrwHT~A^p@Axnm`TtdQeZhl&HQ?KVaA34=z@T z0>$wl5bqrh1~!F*(UEQ7hR1O*4U)m$m;cKY`UwP$eu1IVa`168kFh7_0GvGcQ+iRF z)V~w!s8QK2(6*`!95W3B7Q%+GFV6;Ua5IM&-kbnZJ6b@B=UremITtiG2?1S$VXFUD zJ;kX`q14VBQ33)_f!&+K;KBPe;Mfz-p{ZKVs8l~^P5BJ<@vsFDSK0>@vx=#AJiRZF z(F1mNcfeH6eE5E)06f&P2E_i#2LJK0C~uo3fQvNGpR_9hTy0f1!wq|>S3T>24v#hU z_jLofzV#D0@aQ5?O%j5Ff3)Dz=et0>ttL?P{X=be>q%9Vs8F6~c2Mo_<-u@~2v7U- z1p>ar5RPjZEF zPg)8*m*)T(-hFF`=J^_UdF9X}slp!_oSzzJ zsW^k3ly68ORTBK3GOM|1_Wr-QX2Yze8Gpuqa8gPf_5{+v(1wE9cAX%d$5H;)YJs+H zTZNw7f#{4AFXL2y6fDceAXfM*=gRxjoYk)VocByWIKQL-bUuy+$~;ZXEnEnU<_3dU zhNlagexW$;7lG&Z)=}rAyQpdhWss{qLw&ul3oN;DllrL<#*w}Kj^paEow5_I2FE3` zz^u)O!04+(oX1UVly+DM2z|E6%>2VAPL(-Lg&*C`Ir&YH`Y-G#6&v14WzBJ?ge)IY z{!br)?SCHvpPE*%Vnv`C^=t_>Ct3)6eALrFzJGP*95%hpd7Dx}{ot|O)3vWs zj)6iJ!DlHL6ut{+pS}sq zqB=qR#pAqtKcDJW(WTU`jB#|&RdXuiAMw2852&&DDayeojv5|%LJ5qV=Tvrvn0+>E zfGV5)pzhWlFq!m@cdp3eah5FrKS=xN17Knl~udFF=>sTs;pP{5~pF)EV zb|E)mW8`w=E;V!OJ5~Am1SmSI2=!L+n6C*}fpqC^_|Sd{jJNp&9!hPY%$)vk-f=Xj zO<%TC8z%xdt9zH5wJ4kdXZ~SO?%Dw==H8>IMK>sE(;~`a>3Pcf-yl^sR}~09;PD`K zM*;J_Q9vVpJ*PB-pR+G2kK>XP45TiEf|}I{phH&)ym=rD{DkCzrS?KfUNn|El;#Jf zz7N5Xopmtl_8I8LW4V}=SApU`rNC;c2Yl(i119vZf&K1(DbXDRl)7sHxUZQE7GISG z$JRtsxmU%ZYMLnI`}6=*91{d*c{yFPbD`Az`wM{l#!%2);0wfc4p32a9JMUOhMH_j z<+1O=0GL?=a_usy1w4<-zh)y~0T~duE*v|;$MZ7)>L**T=rZRp1mjme+fWX&;X%T6NC#J9)j7MeR&MOR?5Kk z2DS7mueX(RhMG210D+^LVAcBqP?|7pmffXDJ+b~uSv&Cb=O+ijv&UsT4L}UUEQzI* z14=kg1|Ue`ivkK>=3tJCA!kM}!c6i+tC`UIIUuQ>A8>VzLDO~tAdxafx!;oo*Fq9u zRM087Wn&$T_gx8KF96Hl8h{!ROG;_lf~xJ~`N)H4kbG?|=wH7AJRULxL6;VRV*fCp zvHmO2uU-PbbGAa0SK46EypX!yssrX&2vg2Y%cz;&c*?}l$n4-68&lKCd1l<`2Jk|+ z0SNUJga2OSQ$pYoB{#VU$N)tk$79>qx}Tx?`<1}oARd=#Tp#GWEu;Qv?xUp6Zl|{R z4}sL{eDHm&2>eu&1)fT#f$E?8!G48O5d0+rs6HqMzNWq)w`3>KQ`iOsms_G`Cp=J9 z=rOd~tD1UW-c7BQZl#JZ+&6ok=)hS*1yS>Ii@>tomqF9@alq3sfm<<`YX7r^>O=>? z$Jv?SPR2=av229%M6Q}s4P2hWPUoDe%H+6x z|H%2WB!TifTSEO#p(x>f1Du!jmYh6O#JRLg4fyYK1b57HLHyZFup_PlME&Xo#X?f3 zB2ft)6`q55fMgJ{ir3AR5(q$t2eoDukAvyD6l{M3!C@~a(7I?I2nehLpS5^ikEDyB z<{~fSJC;QGrOpO18+xgWk{75Vo=*H(em93A?>VyXk5Rt`PE%I(O_bpSB_OZa2rRj` z!Mw%ac>9=KN;9RE+8?hBW<{k_arTNJR&+neh>!v4oZQl_VVC27b|5hW@ zrN%r=3x~r42(6&_HvRvuGxxS29MlT7oo*tD--=1Y0yXmU%jst4+jY2Uj3INclPQH=~oDkEfR>l6!q?8686mx40_c8)cfLd)rAS&@`M#9y*KuP{+_I_#3a&C?ygn zN64xlQ_R9kv)O0w70?BPC~`NhlEik_!tn-Y)OYMW#uY92NWCPi-7<@K{VE2Jem}&H z9fGiLy(bxD_mN#uCn1Q@M}}2HSVd0@`MBS~?*9gvMZrP!#ii!F#Lz|aMiFC zEK)hk_-P)2BVT=yg2wua}SsV3CMR_5Q-CPw6yBQkS0Cr>Nmi2nTJ zWRrajF516{v`i1<$K($@^1_=acBhh4`?ATV?+3X5eh=chl9KKUTc&a&Qz6Sg?8)$cwM z^T)@?MVpP-;phl4S+7X^L$b(GLv7Y)*(F?P8G?fg61bLU{J@hV1&pIg6!@`AhinXa z1Z+Zl0VgS*>i4ok(pfK{ap@t_deevN6sHJp-^_zojVH)|VYkV?QUUT&*O0kmdJKt* zZi9&l&O~JWKQeYska+!fg5A5|I++pZ#Md~&WF2;3F3i15{`$$FtTj#e-bE*3Lw+$g z`@@=##fl(5m0GZ8Iu|cp9fM5{+2Z<eP z=^$-s4vgiorKIM>6P#Fn5&B=)fSfy)p?dXJIQm2h+YKs{;xR?`sC^GsyI6pi{#gaq zjb)HWt@?zSFUdx}T!X)_&SwO?F}AwuLSBY!hfA}*Vk4VS9Q09^_Gvgm=Amp7b>Ep- zIljfTrySC}XGN#JehrZfFth(^ijS#fL3}uhbe(>HOv3KrSu*3z37+x{ymo~V4ZVy{ z$0rl-JDX5n;ySYIT?{Ta^oQ>0OK|GCbo{|>KF7K829*?%K_-8ekQEa>NMnmT9&0~~ z8U(b^N8>^`@K=|PtI0+GI%VkrBT;1b8}T-cG_xsq3o`DB=B(0QO%!}yq2bjc=%C?e z^eFQ*G+w9#s|yjg_iz)+{BaM>O%+B>UO_Ne-;Ef`+QJJme2nSK8?ela1DciQqMa@C znJORz{5<0Di((7ruv-~Rr-#i9b$63C)7| zy#Hdx_pLHD?J)$bzVm~dd*(v3y1hug>mKUNT19(2>mml{lJU~%G6th+x}TFs_oNea zY`B%z4Ob5Tp8o^)zVIZ)LFW-_l81$s*Wk?Vd~`>%9e$hsj?TK#c#p2R+18#jIN{A6 z9M^Lh>MqA5eNHk`kF`Sg*6MSc_jhp$)nd8prOdHquOfWo8iV%eq$1(&XeK;E1L*L$ z>QQ3`xc=FBaAou@{<9*8?AVzNQy18hZ=>H}W`6@p(cTMJY`c%G?;IrIbLYeI!1YW< z>?&w{rHNdV%z;-YQ<)>(jp%ikE@@Yv!e_pg!q*)aXcSYn9be%6Ka| zu_X|;Yw)sO^-9Ft;S8D=Hi)8NHhCo6j1#O*GkJoI%&zGqc%h}0*YU8J`PpWM`_y&= zLv9atCt4_s7Ds~r-GZ&QvD}#pCtzr+BszNqkoWVmm>F^lKdlzS%Vz}e?jwcRdtfb& z*t!_kOpGJfdmP$Z{wvzBunV3UDQT#zSkL?oizT9MZ%M~FEz-O!v-)T}}b?l+QkT_Tdw#$oLy`U@Q%q_B0Uws52-#shvyrtgHOd5!-rE3z~fcY%Db-g{(=|OriCAI*S=`# z=Xe)xok@m%qwm1paI-?IF$v`RUX!1D)Qx-*2&WIln19taWr3TLp?N*MBA&cgRUZ-Z?= z2hq6BDd=IP4R*+029J`@Aej_Hl3^POlu`xh&b&zYf!n!*@QIkanqZOnCzwAjy@|o9}(J%MF&{Qv$jZcT~ z%r_B@CB4L}ubRw~k6})&TSD3IJB-fNHe;dQL3k~SOCA~;;-ScNsQ2GJ=!`WJk9sDnBlUj?A8f*HOrv zJs(=|dxv`FOM5ij?>>Ms+xSpfj5Zm!>LIT_q!DYOXi^@aOeV!LaH&WNPTfmEDcKy@ zmfOYMQCG`Ne|7-b1=sIfA?}d^f_i$IX208I=9r?OTjh?5| z34t<%cbgyWfqGuhqyl#xtQA#<|gNwFG_+pr{pyzBEqdY#4a(Q6@eT1*Ll*}R4L zh3TOnr{ln_{x1~!BupkZj}f!X2;wt*9A7F3B`s5n00-_y&iwkMLVGa`kvs^w@Wh=Agf2d!e@F6Zj;74;G5efKzJ)asRe|xNs_r7C2T+!xg%;>%%#@<0?42w+1f}J%qh|rP0a2P_x)l1#)=#1eA)CM0I>} z=*XY}d8ac4m#i;jF8QY-y+8Bd)=7eI`ksKd{5{YuYa^t!eKWZ@)kLI#6baospZcqR zgACl*Pr4R|;ZoUXtYq^MZ##Sfhy!JEL(-Y36%#Dwu0qZoScPXsG&wO3Uc=6tKauA+EHdFsfCEWyD_^K@C%15`U7PY?;#J?V&Gmg{IR8V>*wh$CqH~ z7%o{5n?oi(FM=&%t;A?z0QOw@4*hSez*0K%89P^9M#Q_3Y5wa$=#B&)b80m;VZIhJ z%_8I-t{^jscey4dwrs?bJ~DlXA!Ey>@y>l`IAy%q;^_~$Xy~mWoER~Kn`QZcphy$5 zZ)F@z?D|a_=EbqK|7daRawV9vspn;89XR>81ILkDBz1K`=a&ne9bz6>LT&j?chhDRs09 zKZng_#>~a*1y|2VdB^qUAYH>vbU*fusy`-z)JGmKhll<9q z1bck2MUGNd#Lqnmf4-fF`cD)<`W`>B{AWg}3Q^WXP@Y}o_6dX;S3tASZK%wx6o2>G zhsCt`(V5pZB=g8x;uq12-7oPGzk&%860(RKo_K-Z=&6#YyR6C6Jn2Re@{Ljy^<)HU zEAiGpPqD3q7;=9R4st(LVQZy2YUi^ga;ppxKIzit-cnKIm2nNOvd>|>4JNr`OgA&9 zdJ1lj=3{P1=fEw~cDPuAceebrnRth4kgprolQv#guSb751Ycg^{R8L8BA*a8)u){b zx)gy|mD>=vC;;JP-2Ze5pd5U&quUTX#;@|SsgNV(GiLq%V9htwqy6bi%3#W7}?M1fZN>d z2#?`K+^2!gK4L*KLgzIj8W@ag$Y_@Rv zi6ahWQhyH!k&n44B;B}>6b948L~RyoxO9!Q zOo6VDvru^A3=XOgLewG#n{emR{!$reR4oSmShNS`%&W!4cB#z4ZTfi2;1am~_YiQL zD+D8^@{kv^5XySVke!S^{kTMk-acl6=6nu=a*Y-sdy*orm1w*wYd$GCk2xNf3h;)K zlbDPuFkTS{DE7_@l#{jr>P%+A$L^8v`NR}gaQPB+S-uHcZ>wPDa4X?S@|VjEZ2?*4 z09#86p_HLV@Xrlt;+k?6hYpGI&UV+(+7M@A?3w}xB7RX}hgM<#t%h)8d<2|K-+}H% z??E`O0DG?4haP2%((;#EVb-g9rubP3nV`+_mV?I-zeX|Y+*v?JjmXg7y6RB9ts?fj z|By=Q(MDHVWgs`P1rIf)V&hyIO}>?-^R~_+8V+};7x&}Pcy$t4*&vRzeF6IKMi@{& z`_ZgT_AwsKtwb@6S*U@LL5;UF(H-JX3NJ*G^7e3eQ9_*P?P-GL&oC}}LX+syPS6&+ z4(%9NN;hWFXmuqA9jU0r3->A^`S}d_BNB$1Pj5gk(o0cz&0KmeBYLU`J`)JSsvr%mtBZod%Uqd*PunSp+V@PBw);pqmc`i zBYl|xc>ZW3+M&V4j#6=W-*%d~oiL`Qu1nB+mZ>252dl7`wl?y5Qi$Ttd68> z@CLJQcjE4(g~WW{Q6jonn+fV(N(i?NKA6}#T}YJ)X~4m@Op?CO2^IXC54zetP_ex{iE0+(%_=`Zhs?)t z$5RC|htmk#PQ1a<;TzEfhfiqmi6Q-C^9JslVkMA8j-U}~XRNSGofKLa<1t+qu=8^q z6f-MVzIIU=ZD?mbvU&m_mGE(qo7ac#w###3s1vT`$ODu~wGV9hq*t*6J?@#Dw z_T<;Xj)+#Yz(@w~GZkg3&f0*16mR&f@HG&uScuNqjyM0SIl?`-`!CK+Qec`^g>b7C zyP!=^9mafFOihOdkroyutNzVmew!(vhYB4$Muom{>ev?iHYFtcYC(7%#u9_ zj*~IP9{PI~)4EU%y$TrObSZvnRykNl&v3bDIY>uaSH1)G zkA%#MJfbK+g?Pl?{DjwjK7w7Eyzr*uZ}9kmAntL&Vz~CduUJ--2D&aaOAo9MpSKr2FMZL;TpkKy=bg*&xWY4&mkt`Qs%EzFVL>3sp!+E9{lUUT5@q| zBbhR^BO}vmQGr1ntT$Yaa|9(wAzKSI1#jZEIyWqxED8>FYBSe%Ze_R5RKgc})5Iqu ziJoZ|rMIJCHnV;m+qdTcG2gFDCEb-lqQjDm$EP-WheMXV&HO^!_ALT+dfm-r6Bbu&HRmSw!Y&u^en&xY^MyFOkCf?c2(Dj@HysAX- z(X*T2_;YtgrBWGnet(ACWKA%-{Tfdk4Imi-9z?~e7#0_aB9E)0u-3_w)CR3Yb$>6x z!c9XQwIYW5L4Sf<>Rf}zgHAD(QzP8x`#-^mPb_>J{FSn~2JmvByU=w}JcwR%ohir@ zB!_%|;sxS>exmY*e0lN}THB}5>N1V=q{LBp_3d9~ar1vbif=wt8$U`9uN9|bLOdCl z_FNWL*000rv`(oWV6XYS`JBnx)*edL29mT z95DF6<4348Q9Eb*G=E=IOG+Yokag1*^wGz*$sO1+FF)(x3peMX=F(xaIZ=LGRig{A z`)wR3Ulc@T(*-cHy&t{(Q3Z`<=i}U?c?=`53U?gXN8hp5A$k|I$hnSIR7W*q!OOf} z#=zagDR3xDuekWn$7x(#i12N&v9O~3>)kk%O>#7o_$dn_;d^)$)lm9DMN}} z->?L}%F)HgU;G1-!CG)-*e`s-AQ?X`x4_>g{K);P98$JC9jg^-A-iv8=vbK$Ngd3? z3m<-Lez%K83Klu2Z=nm}sSda+MVQ?26sNnUUNNS-ZjuJx=5~kGKgK-G3a8nRW4#hr zLcQv2u07ODRPT)tX>pcV{yMdMrupB$c}a>9@iy zaL2)iz>y?!y+x`myshZPMXxiVc$7 zaLg_VcuUnDfy1u6zE@$gu~Cv(GyFvKm^=9<--GoXg3nxldHVF+5N)z!H zzR0jE1tCEd5<^X5neJNXcHk*-5I?~#6TFW*4SIRLv;%0Ziy8Slc^-b#QDn!KM8V0% z5qP-P0nS^#mdR{*nkr^93nRN1@PRi zLaZV$N7g@)W2*w|2)9>=<*N7Lm(O`HHMv~)M)f=yVr~+ppC#DUr-ck!NV3mo?!leo zSFqH&ARyvU%`Ce85@aa~A|st*;@jRtn*8+e*5N+VDDjrq#n|K5_XP=K_y?^0~+Oo2+PbUZd5hx}Cle{yr zYcj*+1h4@$8^Pj|0-PVEg6!vrli+6+*ygG@`?x^?QW1%a|4V<=tbdZcmlpvxX%6f% zw^}Uj+=;(=g^*l@9MXEW6>~2)F&Sl+3|D>x%braq8{20=L#;}9`M>k{VL8TZw=VHG zm5(&zpODoHW+N%-6dai<&sCg7bGzP|g9@Im<&-jo$DJBjwKx0O#H~tfga8F`Vh~v& zZH68EmZSYQ7Z8ORE(v~Pj}Aus#s*(C$%oM~e8Wu-&xwq}Ti3hbz4c8vePD$AO}&E; zx|pN#ZLdjWRSfa*XvS;{B=bFEi24~h6nuRhhCbfpnUet=_}YhUwhOVnRS9GyN`_F5 zDR?Hum|Rc3!fZso?2=2FL?H4u3~=kjy_tc~**%Q}EIfoH{qCZ~dG07O^h49*R4X3e zQjZiw<-w>vx> z1~-$G=9E!#q_VrUOWp~;w9@pfzuPi$dX@; zgdfVG9xOM6tq)BRqZ+eJWP=ktyqN{+^Ocz})9;MW-!d#(7l*y;_mVlcH>1x^Q7}MW z1BvB6Vmh;TGS@;?$grC&59Pu^Y za~2|&MBHmb8TmE@9bX#(2tN}RuF92YIHhRL~mNKCvO{dQ8HK4qDR8a#{f z{OCn^xk)s$ChQ8MdBg?R9nK@Yq8BkHs=({p5ObzL91S(>AX8#8?6zSm%K4-rQray^ z;+r+`z7x5mdc^>))M=;g&HhY1n?8$j9{B=W%RIF1cP9Q7|CCu>B|u~{`S4u{7XRHm z2Ol`3L0bInVJ=sV=()^hyMG;lZ#Q!2t4SV+S=z`u|Aa%E{jXq)xDAYSG{FA)ckydp zE?0EHGV*BuYBDGHFm%e2CF7Gj$ThYJuJ(|oSL$AajptQ43O8Nwn1VSfGRq+!9V~FM z`~=)T%EBLE^5oyb0ixn83EP=%;}?Eq#`z4Y?ng`?MU=eftFMKK_F$ zfF5MtBQ<=lsDqRodW1wWGh2~xJw6jjsTszT`yT&GZbb&I z{}@$V3)&MGlEQtu$kIsyS=J}xR;>&8y=4RTva82z&uYee!UDGsT;{&%&%&0j4v6pl zG(M&dL7{9dT1{2c8uo#lrkZeW!xu;V0R}Qh)1ETNF8h)IEQwCnCX$8|VszOJWv+;v z0NEA02D$NH;(pmP2Yr)Wi0ZWG5qar8xQ@St>}$N+SSGXr?|*JdtZtW)qddQ6P`?QK zL72x}>5sv3$7f*~*uhP3O@J)_A;iag!mB^|v#&8EFP5tk`$HZ0)q=ffjsgc=&>Mk+ ze)fprI|`GWUNZd)ju2mizl_ieMO2h_ob;@^%6wjB;o4k$ z`1@1L=vv{M-p(W>tO@Tk`bK;!j{|=HH|P@IT%?)cft@A(fVfdN=ydfz#z;;DJuzv) ztEw(iyG7e!bcP9&J$;rr;bMfc#2nDC_c8d%5;gQVI1=058i4-B35@kaGqU>OY;O1C zeE3@Q9}Hcng&t?0z^d~fFfIQaa94XQOdZ@yKF%?Mk35e;0mpRYcCr_2muug(76es};h{iN;(93&)?7P>O=f-r;rD6KFE17!zW0|r zcjl0e&~DS%|U}bLu^Vv@rb*wpy zZatpFd*hUGgr*wlI&1=_<+?-SG?pI;4)GSzPwQp z9FJg`&l!qnnUocNcX2JVBYYv&kcfsmJw%E6BS{jqdlm@3UC#{ubi>Ln>zHdNLZNqo z1e>9Klj*G0hNrh2CO^*Tf%8Wm;>$xT*(shD=%1VdcSznqz59DGICvSp%r3z88lyP$ z7KCe~oPk7A5fl2B*8y3-h?w}ifVBZmq$v9&&b_h&c;rPl<*vO(Y+LN1n@JLk&l(|( zR=2SCo=hC8ERVOW=b(Lkap}mbMt0&ln`X&Wy(C| zG@{viY}iMqj*&h~2YkFB4DzK9fv-NgFk>VhXN_#YO>3NR(`^QS>ZpdHmsSvm7!77~ z-53nsa{|*ks@Pgq9m%#9z_${4WdFXUBq;nAaqQp0)o9)gYLkUg=Q39!qM86@jEnH> zg*M2XTmz?*oMDPBfWo`h5}VTbjK1g|cKgjp?v^AgWM?xBS5qg*``s^?|6(N3^yz6V z^dOK7t~-ZSd8~s(3pwcY?h8IQ?nz`DzF{d@UzB!Emh_a=FwyH0kSG5!a&(gm)X8*5 z8Czy!!=9B``}6}mzV0qjp1uPo$tAMb;1Q|iXtIkx^0dR3N1Kf93GGx#1DDHlM!Q_SnNUoE--UmD#7Im-8-;HqT&WtA@zH?IWs-e`vB zPr2YbH%`Ks16DxmZz@rK5kpR{{sWRN&SCV>nyA-gLh)WfdP4Xi(fr~=7Kkg5uXc&> zd`%<##nZe0j1u@;bqQ?$)eCoe6%dwsL}oqFhH8&namlF?()=b58Gqe?41!P^almrB9!+W-)!{_n&S8#m+qjNuuYstf z50&0^hPXz*g%-T+*H#4!_^PCgcBl=Meh~G#DeF}I=Hp#d!ah2DY z!z7msuiAw+EZ1K1H@B$tdH19!lRdkFuzV!yhOaQgObBs7PmE z&#+63%Dq5h=BLN5X#573=(`DMX$Y8wa-yUn$_1@l=)=S+ zzX38_XFNYI2-jv8GL!ie#MEe<%yx1nAiSQM{cr=*UFnbB7Y1QIwPoz#soCr<6O8Bg zNRft`W^!q*3!Jy=ED7G(%tU#}5vR2WQRAom_)E(c{5idhd?*rSKfH2ero7}pISjyU zcY0y-R0b4FxCYGCcag%gG3?dC4J`jlZ)9$;5`C%eq!tI%pno%6cvGSvd7B}Mw&}@( zKOU;6_){J@uDBK3c0UGFqV5Em8*n$T$j4icm2gJ_bn&+9uYv2@V^HU%94Q$az`lR7 zXzjbDyczik5LpH|cKx1Enb)lH&`}+kJ9E)}+ZmiBww4Gl*o|gcXQ0e*7xH7Q26+^1 zp)G^;VAA2UsIkQx6$Di?e0$>oD?J~5ukYje35C&yL-*0kQ@dbw-C1ssb|V~5a7CsiQAD5)HAV6KH>ZxM(kmWUV5^}0py`heBkraF51mj) zOyeY6te3~Ney)siQuye(EJES}x^&y9F*27FL}R8Yh_<$*9WF<}XIeKvzDXXr{cs!} z@~fe5rRUL3zt56C?o|j!>_nNS3+Uad?~&u6G_5H#3wbQ$bpdY61y5AeX}4D^kfAg0 znO<-UDY?hf(9ek8Wvz~{E?q;9)ymS>?@OcVzLj)@SsPkWYKT%}%HT%f+ULaG4TZw>bCn&uy%=!$m=)=PrdP%D){lKdPCViyn_|6UVb8CHiSCl?=T}d3aD4{xS-G8leX|-Bg;b!amT>fOa{-$BXCcg8dyCdJ5e0TC zvhc?tHxyE{gU>0%~0ys5^W8c$v&hx;j{g9u?u0 zpOU3`f4UUAKqL~3>Ayxkdlu4~yjhgq@p<&`75b>Qkhi|>Nr9XcUa#G?-y}=X&^$V) zlb-)>FB+V40sY#$ianV36|3H;|CMup{tXQckp6XJB zkrt`gFZdD+It)$MbPR&Ra&v?Aj-}EjpTSU?&VW_YjvGwqDXL#Gs^zQL4G}2 zDEnOzGuo1iTJsLmkKD6(o}n$YS@r_@%7btuVQETNeRHE9cW-2(1_aTjF)@sWXQ(3{ zD@p%`31HHj~W29V4G2JYc`S3_>vvB!B{)|FmI*)MHh zV8YEzk}btWD}LT(H6qH{)rB$aL5uxt+x8MRWJ?A6T;Tv38Xm)OhdrUhiqq))^95++ zCC{rkUlVt|5Q0APF3bmx7uvo*9;$Y|WE8n7sMpK^uHE*NI&fzU#)zJW`@NroFT-=; z6>UrWVn;kaHpc`-Ee?lyRZFqk-ubji`bT8X--If^KPPQ1bJ6&nclha_$Kci>XS8c^ z9Ox#myv|Bf5&6r=rD}YNvg+ zfM&yDxK&L7`FqME>v{1gRBMv$`Dk$^$xr#p!h@}hlIbN_`=SKSq`iY08)qSzy?iim z;5^(Goq*+4ZIJL#21@x*iVU8(pf}fZ@Sml}!Tnl4Z1}s9+%1`PCFX84TfTT16W-=y zX7aBEf9?gRpnDC;6F;@Qre07gD|vPpa_<%N?}iAOR-T8lF`wUN!T?<8L7ry$5(v5 z;hdd3UuC>3JT0jW5fKB+7HgCLOwYjaLKbi27p84S4e9(Nds)9tlh9-*AF8Szf+yW# zkd*E*)H!vDQUCWCY8F2y)!;VS=&FS7oVrdNAC{4pR*F2Cpvao0VkE8UjxNp;MmqOD zLnjOQD_vRFSmhK=cFTk^+Bu%vc={rRbG!4{A0Mu>b61`ssc97IVQ;}dOdheEQ~x5c-(X4AhZX2P+jqp~{)nNXcp?^Z8c{@m~UArA0oLaQohTeyWTS z2){w=xj&`L_V1uyi%{h4MlM)i7lzApv$$mj_GoROE*)(19;XHS?6uXN9KbX`ut7pYT?{E?l|T zp1ra@1-0hpz#^9`)W4HM8;Qb{@9AjS@=CI& zUw|Gh_aqT_wXuU#GotMiQzuP)Ds7wK^}y%d6Xg$eY*DSZzMZtDHsJ zpQhneU2$Z!+d*)#?g~g2T*EZ@EMyvkmZBBQzCrUS1?b#ijw8GuqyJ&(Jlv^#!#GYd z6PYPQ31yTN=Y8(CkR-||w5X)Cw1*~z?8=rEAzMZ&=Y8&1i&92QMWvxsXehsy_V4`% z&bh90uIsti_xt(k9h?WZ?~TC!+%MA3SG6#&`6aoj6@sEtLq6BUi_Vsm;?m!~An}c1 zWI1i18r_JIoC|JBILAzw@|wvEJV}R5cY3b@-^6;<{ z^O-3T%P-#aNZ*z@5~-{nddYrU$g$O( zE$p5Nu}oo-7BrhlLP3!&&Nr9lrOg#kr_T)z6}i#45|vN9v(y z$^RK41xJ@lp`E!uWZHF##&sMrQeKKm#y->rp zy+?_B`!crSPcK{x*5dc)3 z%wPsqjNX9rI_}dd7gzFev>X%;PKSiE`LKvvf$d*Qsl|E3+J_40Vz3iemss#C_a$N6 zoHSOiAe&ucJsHkV&*Elm^F?O321x9Z#MgVn$*Zr2Y04r+Y~|tbCLe~*Nf!_OR1b7h&*L5haWRlnjhlzMWopu%J+yJVEN>1 zC{S<}c*vvK;?shMvEK}o`qK0$$H*p%d}~ zNr$eQpmBTz8pd1LfZe^o%3cR1vz$uUeC5>MjwUws%lVw+!xQWR@jkxO*CE*-pc44?cIxUq)+NsVNdWhMrzYTZqw~rYI!e=+I@_`VN;R6vsn>SZ|d;x z`;PLp)^lMpX`o-fh^c}1XUrJB1fRd02Jqu28K@ipb*pq(R~19XbsN&;abfgFu^u<@ z#t~Kt^=;~{9dypde7^tSMSkYF4N&xL2UMs1A@d_Ya7}COGQA@%5T$wjwEMd;u6ezR zZ0t&*J?9$H=)`P3@(%~%dsh*@pBxVNe=mcaZ$hriq9xE8?*nP8p2EY8Q8-#0h~ZPS z$iIc9L@P`N4L%(KQ{U}a|2PH9o_oWo>GJgBGCOitVW}ueUxCiC?Bt3RE69u|OE{}n z>Ll*nArdDtCO-~Kqn&3YclP5j>bYe!PMTd!>q||j>Q{SWCMqHqw--|(M};i3eoE&3 z4TF`RjzarqcRDP7E{>p^sCQ2_xO#tt$#2!MU}hXC)Ot($jO)b4!ftGKj~w$+^jOH9 z?S`G(50SYl>+!4Sedot=lV$6Qepr{ zz3WLe7LfS5KO8gl2%Y`u2l>gG!jy5VAli5uDBe^fTd)5|^$%ReX*W+2wMC7B2C+?~ z_>-qs=Jbd=4o0BIf_OZelf~%$@Dwkcxso()Nu*st2eD1M7n2?s@U6mrGwqc*$hGCt zikKPHwZjnm!*a(aq$f@(_P7C6VYgL55}HwhWn9Le0hi^AKpBP z-*aXtQQKoo7Vq=tr2f-~V}E~(mlubUMIE{z^9;z)lsn|hSuJw&q8gk}t!Fw5>uJ%6 zU<|yqt+rNo4cRy$iXQX6NTf@5pm zHwHM*T@OYE0RDONnkZ&(qkH?l;KvFUXNHzi{d=4^;iDOB+;)_%j1gG*`3u4Gb`6)B zuRuEv2jDF!dv27J0`1l~LPj)iM2kojdULZMJSo}(pOz_tpx42E$@!>a5)Zw}uef7+ zAt3I#OI#jQ!h8jLkUI%@$0Py0+Z*xr?tkPz&+q7Ne*uRoI?#_2#dL1vVd@ZRgA#L+ z!RMK<<9a5-{5)Y_7!`tVc1e)U&IieXEhCZC9>Ap!y3wPCqG#qd##g_Wta&8p<9D>v zj$tcky>mJ_rk^0*wdWi&%WXNK+h5Z`>Vq@&iqDtL(u9y+JzTV{m8;E=hJ%YH13JdQ z(+QHoEPswx?S8==PB13nCWE3)TUok`|G~}oO@W>pBU#^pW3Vc)mK%M%jbtx+N4pd^ z!G%#$B&frZF$B1h<~d%#c%6q>Bj=vFY%fzK55G&-5ZJLH?LqPMH{k1 zyN<%d+4UsxLLV*MITYq6W{B1dIZOg1WQqTd7jUs48Xiv3roI~tV1=|QHvMiExi*c) zD3eff>F!jzF|C)56Q-)Y-dDuO!$TQEGb;@A`APJ08p*Z4shF7b4-dMgvT4sv>DzOK z_;sQxpwCK}=orH}|M!7=Xp#Vx0})VK-yt55^QKj*@mO6f0}_7@p!9QIXrk0uC&j0v zJi3sTyZ#XBpTz^5ILdZxPvU-_D#RSEK#U4^roCQk(Z3>!!k`9zk>lu6zd}3`{g5P; ze0kYlg+(dxL|>!@XQDaQ z0FSYI-`*k`!6JMp_}ij-0@)_jmG~oI1CE$#i3TYH5EZ$WJKs2pY*l?v1GcT831j2P zp9U9*y|o2@ey>OMaVb>UniKE$eorf=n-RG+V<5H7fiT;iQ}@Rz@MrKFGhKHW-0xgL z&izgz0}~bSyiz|+8Qn}T2A?5Pg9X%0R+WaX5b;4Kx6mu|1D&zah3St|XC@v!Mx`|* zS(SYm?DnrYY^~CE#`I9Q*m1*Y>UP8cS0xK~s=e8$z0CzK)pVm(MmfHYo(xxRY^8%b zN2#yypD3Al3*I>cn7g+D?%%i#DFqK8*SJA!*5gYPTU%&pnlknU_|UW=-DG#3C(b#g z%$^&2g#CU!lzCtL9D)U(+OM}I7-E(WFGU;3(`$=qWzI_yAGCuxx<3h{H*t6^*b5hS z-Q?zk#nA!Anz`^xmanpD#1NNz^h;(GEL)*2+-W|8&wf>m^iUvY7EfiT|6YXtyT*f) zdm^lgv%x6&Bs9HPgZ{Q2)Jnku=6x>*qi^bB?aqrd(#4z@uZqK!w|3yu(_@j(n}TtB zc5&bDt|gBSUBrzq%vjI<(`=XXNjlzCgLfS-=9=;%SeqxWNQYeoqfj)Tc2Dl0MZZ*G z!mlB`cgqC+T8KJckF>{Z+cR|Z8-_eGED&G1bB?=LlFHe8WwC#Q3fZ|S+gZE35%{UE zo!VFwa+}f)vCCrf*;DH%IT5rA-4Z*{_mBkKx#7V|=mfILf8|*hTQBJPr$w1uQ^=o7 z$#sJv7;*m@vm){b-LuCT_PPhb-Qcn4;Ma_o&M)GEJKu2y6-BtttOf7>GzX8kr=(-( z6dJ7&Cmw!GA8zl=5PHo)GPLgnDDP;t3IDp8baAM8(^nx5XK1wgIw5THvY65>ue{%ihCEZ*E@M<1M6*)BV_S5Aj^*iOft|6Ri# z4T3GPX6$y!<>X_cB>gxzUPm#da>oyFt%sp%()pYb1V zlid4ScbkzI(lAG~dRQuTP?O*sx6ULoYyV>|y%-d?$a&#hw@?(K&|o0=6-@d#7f-r{ z!wBvIG`!dVUGr=)nw|)5W$@XjLeEW+;%ALBOywsHO5+^8bez?5l?Kdcq)lJ0k;emaX!HC#QL|Kn z-D>a1F~e9~dDjKwn{2UYLnlUC$D>n&7p|$gN7YVHqWRI1{+?>Ut=K7tcica6H?Oh8 z`}I;Xa)cR#o|`3jJfm=_)pXF=KZU&gyaa2joKTW^1m?bua9A=1CTdHd;`(&%)Z*Q^ zH(ZstZx~?y%HHRSGd9stkU{sG?&2o9Dx+_ABV8oh$2H#{&DGR?px#?W+^xVQ8nrzJ z7Ucwhbi^Y(tMP-jj{8gJ_@#2=x3^-YpARltA5YHq8dFu3ERy|50Y4@C&{4V`)OG(t z=25mQSASktY%sGD=E*7}>#4@Xo7uwy*_Cj6wG8Gz&BGs`zR)mw6qmgHLQKC8VFNbL zf;m$HK-IYyCQW~a$@>O@mHq}Fn-cMz?H2LZBt<&yiiOzMMw8}*imB(4GIIRTEXwSx zq>kwUH1XkgqN^H%OP0+B?^Ze9&G;X-EzsmA*q3nu$CV*;i9NH19>XnfzS8xTmdqB9 zdg|=nONR?lbo%Bo;?-SJT(iFcwY{>G|Bx5R2c2=^@2YShGvXd-+*iTSB3a(W(Td;o zs~H`35%9ce2!CwU_+R}V_@t#C@7%JWKR0z!>(BYb_I4D`7!$#j7Y^qiXk_Cs=Q83p zG>DnfVuc4%_u{-6R-`rGhUu2oV85=36$s$7*n2%TsHN=9y;80PgD>aB`#-jDAv%j{ z4GL1NzS#SS>CBxpW??#gC;6E6eF%g4mRs~>Qyu(}?5zE9vQW?qJFquxnxLj|fViL3 zLv~ImzDyfKi|3Y*C%Q+mb=7Hz!3%6IoMnecN@9t52Ks90V9xkQv`Y5@38;_6_Q*)8 z`jQoCYkRHtp!WmlfOdrAaD9&?HL3!O7BsM0B8`Crg<%M2ycGD{U(^C_^yxpG4 zkM4z`R~jJSM-de@FN>YVY=@Q;)8NaaZK7#sMzAjxl$;Iy{3^ zSDF)B!if%!oQ`>AV{xr!1Lv_Yj<&wuNB{Yc;12YL)A}Jp$b8j$fsQzg8Vfr9d%5vU z$3uj{rc}^h;{}xcpuWQtM_zr}P0)T+u70 z!cHF#+s>h-3+0HF@j2WQyF;kOPqAj#Xy7~KFngvXRJ1PV2QuPsHEPHAKS(iwBXJ6-T z?bikCg>OmeuPr#9Nd)U?e~@tx$HA_VxJYLt`MYm5zJ5J|wLhT3T5dLC1FxvlkDvF5 zGU;J>(wRV}h4&CC<`S8Ga)5q)l!*swcH!G!3M@5?0^p=?9-8wCEGGXVziUIp zCz`@otA8KBFJ%QJ-8e=00e{pPHxBZ=6vX?N-KVDqm%wJfY|+(C2l2~k7A3f&Y))?` zjEeOKzD@^^XjU+(`B`+GpAwUPqnqB&ybC(d?vmLSTBx4dK>c6FW7L-&^oB({S|Di9pH6oU;Mx?&(3Gi{p{TGJ?+q4` zck&)sbY=~%nGW z2jOgP!z4o!_H@xhQZ{iFL{&xO*N9f+v_` zI;jV1&*u5k@cK~F+n!8=h8~35{`TN~N8qJY*`R5j7!i{$7|Io98EH*CJ^jg5Cz`Tq5>#PCBf z=t>M>ADnUosS2RwSNBlA`g&5iAd0T~FQ5BtZoxG;htL4Q8+?#a!j6gi=#1Y>sY+xK zbsFfV5JK{rc2P;o3Fw|pFtW*r-L*lE z{hfJ;FEHk5^`s~)J8TPW3n{bVPc=zaJqufQ_*2Ht0e;*o5&agte$Q%;pk?z!d^L9z zs+I_I)U&IoyWNrwM4AY9s5+ERuoMk=B%#*se0-z$lN>CZzAZwp|+bj zZ28{G7=@fCC+8-DV}J#3^J)NFk6y#D$E*2&RD*AK55rrdgwKe`r+4m`ldZQh_{K(e z-a#dd_gHR9gSU*wBQXQy3uu$IPU>*TrwU7^oTq7qI(T)Fn9DjZ#C(PnFgN91ijHjG zi)tCAG|(%8**R2rKT2}R-0`{0-v<$NVPr79*<448H}WKAnJlX0v{9QhJCHfB3G0Mj zGibLXy|7&d|NS>a{QIks=%N=#Z_&wUQ&@%TOMX)sGvRF(fOhm}2&(OE#nB(*=%~XA zuZRq9~DlgHL2p*1Sxvq%U`lO|2fnu0sM-+2$$s&i2s&}IOWR$ zxD%5EGeR?Hfb?zAwWK?Ep!YOhmC#|1i@O=0uYGj!=+jK>%8uGoOV2Pa2VZ00Y9;=H z@^30jHo(^UjU+fh(1_y;X7%f5u>mYXKl3|b1x9vAV= z&q-kSD4*7CSaSZW?Ih3(m<=b3Hp4CDr_9@+4kj*b8Gbv{Dz@0KPK-kn!DFZmJ?5ss zJgGH-e;Uoq{&jIysf+Dd=WQ||6*!z4g==#j4M?-XH_`@qHbKeOLrj6v)U~PI%+tPBJz68Ijjti}puqV4{m1)72qIJ9Z_& zcyo0!bBCCms|#WlyYl$XKnwMX_Cvzqd@xL~1hYtgNK{UT`m{~3Z;vcI7%ZgHx{;j5 zo)s)JSVyw{C=%zhN%Z=GA@tiy3wmU1JcUM0y3o>y-pUE1&&Mp}*+acR{G@2rpgxtE z--PjBzc3fGyhTnq{#a@Bn#PS4VWE98KVpg!+^rpjl}E2p8^ahnC;Jr9^&SR_8)SLs zBR6SE+LIcyZh+BN5oEIT18%aF96MFjoDQVOGRc>m_$|g2xWWGlZne6jcsL$<#V1kbhZ?NVhTq{?hce>~DVeJ9f?=*~$$jAZxOSeFN!$@}g zg_H$19j1uY?&cg_$3=CZxN#~tP5Tq)Txf>=&xmm6H@K!dK25MibHMUR$i8(ju zjSxMUR{xJk*{`N0)?qX*ndcO0)?;Bz6vKO&5&vCDuxXqa{3me`(P06KM{?vy zP8L^^oeU|9fox1aPh4j85jCwfWWPxq*}8pzlQ?prR%z8pkgqody}>=$Ult2Nf{y9v zqA0QCh%$1kIgk5(MH81?X@-4f!(g8wh?V_$uE@z6)Z1-g~bI~W}xM+ z49UNyFuQqsMt4^XXsrE4DsKick-LnDTG;^InJCYmy|fS1me)em^v9qgt;}l8IsoB6 z;zem2lnJKR(EZ;tz~8wOw(nm8hiXDtY?XpD{Ao@%Opnd`FpCcDcfc3rU!c>#8hq3g z!F92MxIa3Pak+h#ZcY0EA0%aAhMU0Dx*J;S;$6TUw_PD}usDvF!_B}h^{G(TnqalA z1>1DRS8SxPmQ2j*q)(R@;p37l^4h-~wB<(A`7_PP&!d?j*U-kOT?nJGLOvKsSTU@F zAB>ut3KjOvv?h@#0u?->{piJ@uu_+oj-JL^w6N&XX5r zX3XviJN)!PU(nPXr!Frva8pSl(?>Vb{2wXI;XR|_^jjJ7W745{Ov7C#qHjPTXNdNu$N|sx!Bn{1$uxn)sb%^MzU8efS`s<`>xcPvi zwF+unWTQUXz7RuDx0qg>CCi?5>xAsjo5+mKZ#ePEC8TFkA3SMZ&IFuWN@45^Q1w60 zC0PTs-9kEUwk@&fzetvhmIUql=a?5cMa+xy46~71KxJ|Ev@z;w;<5VW^lgeKY@Jm> z4t%@H5CV@fHhJ~)aE_`Hj~_-Z|yvbLHuWbyQpbAc%Atqr?=!)$i4E3hfO z?u^~mF@P_Psce2Rtmqzs$`PvYD@}r(uoKa*u$!4~j%C z%i~~2V=m0@NT%ET9r5s{2yW{5W?^Twmb^LO0?KQ4k}XR^h!SL*yTlJ zv*>(F5?DA!CeD#(9h=6npI6$kYsb7|9i3wd?sg}GK}k3*D-2#8C}x87Y{~kBIgohP zm0b6|LMnm;#^Of@@D9$ko;BtOQ}{EBd$>y-W|Z)xe5(OGarjBrcg9h-^qbU=A7F;8 z=%Ss4KH{JpDN>W2O8je)?4OfC?>LZaoCE_m-TIY%ZBnt0X4B(@9UBC5Uz>i5q{dr8kCX zvnx=JlgL-(wy7VceqEL1NNXn1duk4M$@s~7C4=Qg5)i5RYwlK2CA@UVtZ-veM4GXqX|Vh=U|`9 z1M=#r6x;FSAyYeZ7z>A+7~e;#)b?E?c>sR>KFC7$VltffM>wgeojx!uJ`2GN%YdPIF0r{Q_q4#Ad-`6+xtrJ3)Jy zD+a}1!^!e?__FjCtWHw~m)@nYWNJFhNZbUHlB0c&_Gna9H5QJvZ4KU`C0{=HK z44=792HC0i$*xWJ$hRO1xU!{#j%r`SZ84#cGjk+3`o4nOk}~jo*a{LhKY+Xof69#! z_^vm`PbC?bko*h1GgZ59C>1MhdF323E9%G;Gu3MOt3YjBhRd$hl-TJknyIVjTz1>{6^Fp7!={yUxcN6xxxfXl**bmtBFo(qD%D~?( znyf^E6Z1hrmy76Z;Hp}dlkhR^BuLwnkeekiAhv*%8Bs)j^GPVs_lB!wvqMi&lNateDAn8h^xe9uSo+7(03T2END6*i@%F$eGV5%YtEWOM#jn7AYX zwEQMQUd&W*`t1h_&aUjz;?XRLbb{|P6N#Om|1TuYSh1&(3{$lszv_lD@aG+@-kDCm zj+(^Qtxkd=*=nr)5DrF%6vA!0k!1XMb+XWR3LT7>hN|z{5Sr6LUaHl@6L8)jOZ9L2Gj562+%WfX#H*LDota*W3D;UN_DTdQ z$*scRvc1%|Y$9G!(S|X;W60K5v3TZUH1jiSD^bdqV6QJ&O4GD<;B434H@n$cYhKNu=*#v6SL`qCDyl+49bhJQn=z zGp=qSBgZeJZ7Rc1tffaX^L45D3>FGcPJ%l!qu2{o4(J;<$TarO5Fh&ffJuKogdPlQ zBvF=1^oa9&(R4K-Pw(I@>uQTUG2610YRyb!auYStRMHmX)r`R5RsaLbr0~yeF><+H zD4zU_mKXMrDFS0UfIOlaQrDPWXIr>degS_p z(wdhAG^58|xWNR$sSEeHM~;r7&|!I0N2ZZ5BbT!U&7!(8X_;8ho$g#f|IR)wuJ@JU+WG>hU!(%K2s?)g zl8X&(mDKc&5j83$wFl?famksA%;^1vXz4H>D@;AXrbmO-0!#L8gqX@Z8*@q9{)z0( zmO{Y~f&07GPxMKgitRg;=wd~0?#Ay}^y(SK-!QDeFO@5Z`shBUacV4?w=4^DJipZlT+UlZ>WpQWDU&s+yu(l8ly6vM?5E*FHI$7dWdcN-p0vgJ%&!bQdj z8^udQPGh^Y132A$O7`r0LAWO-v|rk8suAZ)$2r=HKHK~xPBSHV{&NN%tk>s1I1Vta zzAm)4h!aU@MB>d56?AX8gg51!NLf@5-LD{MK$phhqUV`7OWpv_evDvB;~&xF;9jQs zw4jSl87=fubtGtERBd;@3X{9Q2!mgHaGZ9z=;syJ+ICdt{?)kAl9eCm&dy6*a%CMI zVz!^&Oxr8ItXp8U?D=%jGW|7#Y&D0hNyh}mDay0roncjGa>?~eKHl7Y6N`vRP*ACtA+<7tH8CTCdR&E#lDtRL2 zs^c)<@CAFJ9n$`KVAT;TsC5d z)Kx(LAyazcSF6}s)F4W`_eHeT$c${dF`cF`N}|hlRGcGZS0s&(g}4?1#fe$X``4^3EyFC=NqzNsy(~qQ#6fu6hLI%s=;r- z5llLh$c*DVXyT>ec)D*`?ZB;tknHoFQ~Dtb4n{fLLZ|WU=U8B3{TTMjt9W=Xu+XXw zEyJAmBVo#hElfgF0bOYLnOxd-67i23J9WW$;`lU|%nmOlb9Zaux9dfe5qf*y`Eg9@ z7Xwax$2P#h^NdrVG9E8m0@d?c!FzcM(;Rr6!=n%B$Fp*zLX^U}Z3)DXvfYffNCFMA zDD&aBEOq^500jvLA@JopVs*$6jY52h!o+(_*1iUsH8hSa6#e01*3D(bKFX~0a(VVc z+Z#G+ogUeBZVF*_LaCCQ46UdSqZVg>(@R|@boQ!k7?}QjELn5xI#=(p24dO)?teK3WB3Nt zR*r_0PlrLxY6O_&6uJ){bpb$zT@QjOX-E~rQER8 z3?fMc-m8oPJo)9vyh_?6Fq9=lQ%oO_6$!3nmV}*k&xlEAN2}2?<(aj#!T`~FUPoVr zU%;_PBf&mhm@TZL85f6cbXtFo6t9tj1?LJ_<)}I+tImL~CtC=bH{t5rOX;FXQLOoG zBeo$s5z+0PZ24kmri7CLjZepkcY7~%sfqCVDk<<&E@I|1pRHB! z4W;pS-H7Tufe&b^1SRViFvVU8+>W!=XtynqJc*tpy6oLXe1EcZ>JLwP^4LbcTi1&2 ze6*6aQGWpv4fEKY2?=cV94F9d(`P@2pTu{)g#B)K6tA7p$Gy`F&^zb=9#B2NrS?0M z@3ED1yq66m+C`B9_O-~eq)f^U=*>LT34VnZJx1VpLlYI$TT!?7GgVL9AmjlQV^8g9 za#eRC%~MRq`gt)pHqMTZSg?U!@T}o{{A_TS+B`HkIu`166PX|4WLo#PjoYnx3fdC6(hnW+P zCeruMwNTBkh-$>2qV&*9+&z6I)Txi=dB+Nze6NvNwagU^E1I~a7aXx$D~NWd&ZQX& zjnrkfB<_g4ib~G?T)t5unkaelXRNKD_jez;p}q(oZ5l%1cQNS2dy{)PecYw2cv=!F zhZ=HM>B^tgSka)yr=4FwcdSsM<#SbupJF&TeYisHHb|1PTTb+63(~<2SLu7V1kAAz zc5g$!)4Lx)%FjxDX5kF7IcWVIudv~Kc2KtGRM`5D&RA?3^tYQ zq?xxHBR{K1YS2njZZvKK)qi#G5e0_K|pI9 zUAM=aUz;|7N3X2~g`u+WT<$UX^Zp9veLIN1KE$Jp!AYuq;<=#V+=hQDGx=wJKSkOy z0Au3g@Z-}v^w0{PTkNqPVz0^2oFiUn`9+PnER{#Rf_13;YhOAndmKFPn$7JwTts6J z8;CB*DbrrTHy&Ek!Ht$Vg9#(+$)%GM==D@TG}=CuCP;J>`GGq4+&vVu)&--%?o^t- z_8RW*EEP46vH@G2S(qIdk7G|u^642m^O-Z{$+5EF`YDPQy$e)js+dNvqBN>B=e)5FsUV)B57eSU;Q&2Q+8yc$A~AW z{5VzosLcbt6)xb+)EhK0U;l~ceI2*f8;eXz zhLhBeU$lGcE!>fIiF;Pd(S^Jx{&-dg_H)+5i*KW`Qs=7Z+1Rt^H#`#KrQOxI;?Hxs ze(oB2e#CjAxu%IbIkyZi-%n!<;#P1uyg39O7T6~#HE`pIJmk3VuJx+AKv%sg;gq)% ze%?h{es#t~T)zG=rw~?y<2~j1VIQioLEZ2CbS9HHyNMMS5@6SfFh z9fq?6aXN0nZ886GZyp|}r7z!8v3dY@pRXa~b|mpXzAeLDV_P^=13jFuej!diaTqm1 z4)T3$82+tL!&^hD`4u?!8PpT|>|-4%$$^ zKk1BX)mF0p?|LG)hlejNzbS3aLDk5ixcApr*gLrt;%8ZdypW@@>)m30ty&>nqH%{A z>fcWcLYI*pQy!pY-!y)*-wb}7xgR6zG9H@_&4i_05m?_?LiID8k=x(G^(O5jMIPg+ z|Kgd%R`C?~tVx5J`0E5MdKW9Muy&)?z0ZmF^ca#o#q|89eMxmpdfRF4B@#^bMjTdC^D5bRs~i@e_x$&3rk5ZUVq zAr=pBGN+i+VCR-(T0aYpAtbfWE`TpFV2&Q zt4m=~p$$r}Jb|u0G5Dn1M11+k9}?>CL$0RF;FT3E6s#1{=X-?ph94aDi;Cr5+vSmP znL+xnx{FS=A1B&$^aS;NaGXYc>7mPqxe}%O8o26iHeY%ro!5M~gI5uLn~ABO)aB<> zQuX>LGg>>1nD-s#b~I`*U#B#Q3x^mnR+Cr2}nwPJp3r^NE9<1s%xCCUb9 zUop&uJNoq2Zy9p5at^)u@)R>|Ni0YmQe<1t)Ps9-DANWsVtE67?4*uy&|Enymsbj*q2#yo9;4w-i0TZ4mC}Ipm_QG0D&< z7H7%Fz`LJ38r({TUaeG`-(^P))f>384_(n`)oUVW8i(uj28fkU6TM#%sX>`AS5GqDhTq7JmieUZ?lf`t%tNAW#~+jR-B-w8`!)FA!y{au3FWe) z!ok&T8{ICs1Z?_`(RC8~`1Hsf(s1*-DBvx_$kgnCZ5w|wk8c$b&i*Y~9FRmS$Lom6^2Vn`vuMCXHXVi}{>vO?3h+NX)B^ zu*dxn(fRU%_(Xe>p4t`M_xK@9oXtQv>3Oazs_7ESybuyY7!lsES@)ME5GqY z5-)!#6gMUvp~tr|C>!KP=F<86hIRcY|FwZy?)yS2<~$>Z)qHW%<>QS0mU66-Hbz|y z1>SP_3;f&p9WR8X;E0`f1#a?VyeN|=XgHMd#E?ngQ0u|R)Q#o+-)Qq4;a>FFBU8GY z^%Z3w8;*T<2Pg{JLP~AHIQC5qcdayx`{d&+1fb{Ra%f(~7?e`k#0^_@jjn!Bi^bc0`KRq- zJ~l*(9U`zhvzG2cpDzl$_nj^(m$pFEhud&)S|WG;RyHwcG{QFvB>AJ4)p>2dWuRwt z5&pZnoi4t#1-1kjL6f~6i8$*^zGkJ+KiOZ!<~@$MsIi&xk#vUytpO&o?JnnkCYMR= zRRFsQE`0vRXnw;Rbv~u<3C?X67zMLGgBf=M9tpYb3LERm!Rgmv(xQHtv8#zrzZikb zWhJmnY{Fk#sDciE^l^oeH1F;aMy~H04d%agk+bX`l9TpU)M>9xXN;Kwm5&t3*VFm< zY3L??rhPuvrJuD*nPmz&t&?D{h7Z0gRpLe3>-e#|=Rnqt1p1XVkFPwx_bq@EXy28M$|LA}#%VlDf4G;=Zc` z)H(AYtau{}*Qc1k$fBdfNc|zB@~%wDYsd)l9iC$EXas&ATX)XBYIKeUTWaFsxf zD`#lD(DQ1|3WLcNsqoESf~;!&}DDxn1MGRRej-#Y26t9ne#k}Sfkaxa=n=LK*o*ny^NGmD*YksFk8x;Nc;RTP-ZrYi{3h#&djw!{5_giKem@Y zu|PzFKi%cDd%ly&sjIF1-G+z;&m}Mq8)k4rR(>a?t6Z@3(KxKDD_}03GsNDTWz5^a zJXGrmN4O`&28th%^jR^eSlow0p3OzYWdfgj<88Vq;vCiK|4w|XYN1R08PM9K+s$t$e2E9ma0*8sedl2{F5}n5#j$%#{fl z46R;--=kRk6YB^wmqx>*!ilhGa0qSuHOTF~{|w90#^6!p$&s)yke#88na#WCixQUi z%<#au=i?ZKYhSQca}bL{v#IrZIeyD1W&XH(Jm+FihVPrjIKN~FmA!qZy3Vz4>0 zr0WcGWco1v+#?Na8hD9wO){|0{3+4YY9~$C&e7fiN&Zt*9ND!!mj;?Apmy^dUdDSr zIpaH=JnAb&Y5(7J__x!jIbDZWlpT!)l@Xj+>p1OhK7k}Lld&>hh+DqZFm9g44Bt2l zEE~O%*qotjTK;23oqI+ebt_OCxsm8sq{eR)M{uY9&Le-v_K{@U545|c8gI6~z$Ge! zMB-95Q68hlYhQiHwRSC|3m5iai+yXb+h8;HD&a`Qj>=r&hP2-N89@Fw6PgudT_RUx?)G4~)Jf^BF zPYleAS4O|$rUAGG;yfz`Qt?m6*=wUqwVai&iiCy)sT zGVrZ#Cuw(2wfd8y!M?i|1tyF9sLGLd^su8y)HnViGvnkrv3b)e@IN?(pV4BA+mEZ@ z=MhPq-NZ0@ZnlBo(Y(Nq$-GZO9u|Y=A$R`M#9)5xVpMARSA0KtC9=&t>2VrOM{U4kQKs7WaX3p+)v|v z0LU^uOJ+Upul@WW2HQVQ;>+C^)K<9`)J*`>C`o=8Z6e;Ls#1+9t#h0?ZpXnSjJ)W zpVb1(bHC!_<+?nhT5NrI+$84P;dyjPe+6^3te5zQ8S{J86~%SUv&g?CENXme!W7Bf zbk1;FTIr{bxxv-oV`v1OG4aeqwdeS4gurasqeY*k|EALx?BJVoDn&0wI`dG^!p04Lfkz(l%Uj$&P>aV+~A>zX(P0_lV?e4LVR& zjk+$D8cId3KT;H-!b zT#i$Ep_;p-BS+pDYr>L}q0oD36jOdZOYm*8IO{FYkJpCc^0%wtkXi`s@J@%gi^ExW z&sFT(f-_hdT_p0%w8!^OcOiY5H~g0-=!Rpp1y0Rqx(#Zmc6lJrew#x-%{0JY7u$)= z*C=AKErPh6naJ9N5_nn{1%*kq)Z~N;`Bdi&+2L#9=clQV(fNz{xIt0K_6*0i*;3r^ zg5xmlMiv~pmJL6zSYe^`PIOtk9}`XVVBZK+%1w#}w}pC)_wFW6K01sR1lLpZ5kCC4 zXKzs}VIKXeltor}sl(TVRaA5MHKP5kh)cDPfXxS@N#F}H)6{*2ZP6%ZeJSVl_;<3tP z@6@f>q4t~VoXDp6i3V)ini~)vw2@T}=HaEU6U5)V4;!1}puM}2`S4#U+~BkXo1P|- zEA=6PQ--pgN5er=)FbQ`^{CX6WFn=b2V9pF<|(d*__$p(<cODE@dXoO*H;G-#5EA}39X4(n3r+hq1aFT&v3nRx!XDN!Eg6p3>R5)F8+z#W zN7C?W#CM#sMjh=`MzRxCH9^65Id}Kwda^zvAFKK$ar|jR*7egSxYn%6Cg|l5<n*D9dKBa318c%dG*$->GoTQDl{1g(R~@H*B44%j+?+W6=2+ohe9Rh@#|tZ$@8 zO}NJwZ-$!aeCv0nmbgmLck64|!m>yQ=3@0-VljUhJo}%cGmop`egAM&+JsgVEuvke z?aXseDq18F*_SMZ?DVllq7)^Ov}zMYD%v~q+#_3wlBANX5k=N0rHJ2rfB&C<&g;y~ zdFH;~@9VmL0VXEOu|BRpsC)f(=y6Se2$Qw2K10|c^k3x`=xXv-A;EOA*$Cb~>KJ6^ zULmJvD$+MZ4fZTHCNJ;n;mZhLR%V|SyS!~J>*a6(A8Fd5<%}M@e>F>&9Bh?OI<#y5xaaVHs*XYn2tS3e)`8VyI!b(B=1Yu=1I`C?h2jMUr*Af4PifO z$qJc}+cdB#l$%j8488RJk-n)c{x`T!U^VMtWm+A(r7M@6u9Ly~&KVTPcPPM2;XZLU ze?*)J<=Q7LCVcT4?yZO?ho8z3lYNHlI{g+>X)ZyH`c$DMrI_a4{K&8`;`trXtGST^ z_x6zEP`v+Dm5pLSyR z=_p)nc7U7aYsSA)x=A^Ktx>hQS5O)(Sc8bGkNJS{#O1Y=8DA#K4aNSoa_ArnNz26V@Z&B+`&2I zvO^SR?(sv{*sDT+(FXaI|6#DZ7J2h%l6b4@erCX0!8+*MfBgQn-Kag{8V(%%isBd} z>L2MwJ(7Be!pK-SHEJlkTE?E-@(iYTA}Q%q;<5I11)ltnPm*g=$_~U7m6h9QQq4z2 zw7KIjZ7Q5Zi@wi8H~UV~RyUSSxP70ATX%{1G~oy_esKX0XlG&&#K0ln2(IheRHiX* zG+DX-D6N~j-P-f^WYMF*H{@IHdYrX)6q#k7g6n_Q;QMz&_$|$EaZU6rw7#fBw9}oy ztfvwF9?eIqmJZUK<;^Ecl;e(rlX&T>4g8eEWIBKO5$?r4J@~2a#$Fg^kL3buo-b5J z{WY=tmZzt9Q&A1tE{H_y8ixF`7)@{QFyc=ZOu?B8-j?0%E+eZqErsF@9#CALW8#`T3c)l2ClhgOUP|I_37hJIK6M>~gqFTQ0C_zffa4?ba=;t0_Q z%Ma*xGYw^yMKPw!dhqCABA;FQ40WIW<$QLTQwg~y%J^r{p<0r>$5a#Ey`Vw#Lp6j^ z%E^a^Za1j$HWduYsuWFqU4;U^l-V$TH%woBmw(~#60;7z;>BBA$+s2tjGg04W__R= zPDY`(`%)3wM`>b{gTT)a&Zn&%KyIq2^LM5x;9%Yj{FPyjUq6}haxXp5*`bvvD5&xY z0+ZU~>StzNU9ioOZQrpwXngUzC0U|%st@r=s5*Q7fWN@a--!2huHwP_1c$2Kz`3P! zu--6;_zP~)b&Gv*w$Bj$*Ow%cy4DCyhO1CN6M;3ocs8jmSU|Qvyh1IOKPOK`D{S4) z%|NG|k&I2tUApSuRv_*x_wV-&5Z_ItkBT(FbF(x3HEjj(UW9eOC<8N} zzb6{}7T_(l*?$_L(5UenZtZJ>)E^7VQcnJa%Eu?+`jt&WA5E3plFX4UBY#up)O?a7 z^p_&%u~1yBDnxTK;HbF;yW{M5CU?pJQ>i1t_HEUKvpZK1&D)Z~dC`~Li62E04}_xR zpH1Qd>C^bk$_cA3xr5Wyu^4bzh7A&&WwNh+!mMYRkT{{5y6-jOy5t)nUIWRM^X{;P znzNJM_7m9}3I;vv2%CPJ_G{iiio>Wu;9u;R-A`NC@z}L;JaP4ECC)7m;rYrk^3t`J zc#K|cJGSp2`Y5b~DNe$>Dm_9+Z@dE+Ev4C_qw3T^^qHwYtI0c!Y9(jPt--oSo_&y; zNIZuHQ#NM@iP=mA7J?%C;f@qxic0Xc!Zo<4R*&5wdm*Ih0hqX|u+0`ju_<30B$5Ky zs72av?D~6p%^-*Qxn~^75@x15US8xLb?1@jvS^Y#T9M&}+|thd$H2u_nX@fzfVE$p z=ox#2m>Vu2{yrCLPT8;w{m8U{6us_ah{ZE?*eA2&;rp{R;xej~4jVvPLM=fqI!2_{ ze-@126v8Rz2(s|uL-6LlLb*-?TuKo$2a|Ql>6H(~m%{V7&8zyzZ9_vanwQCxDnz4K zXdJyB9)P~j6fx}!hyC7OFyVD7w``->D zMklLn{2x(DxlaN^p2MbT22fx7h=^+6(V&e6tVQ1ldgM+rBn3vm$uUQWC}BTSzedQJ zn;m4wCp;w1*NsWnQd@ApB@13hpOBfd25=(i5jpbqCv7`>mdam=MDJ8>)-6_+JvE!B z-DiVDYtjQSe5x~kJ*)t$)Q=LW4+mjT(w6eVZrogMo#=o{Ha-?+?Vs`3 zL9WW4KpW{J)Ms%%owH90?$yOow-=El@0$|ry>fv$@pdBCKT@NYYPaDSI|4zff%J7i z8n^tNEM4s%2nLrV*ao|H`qoeZ7q(r+iAu+qBbo}}7%B_ZW=Eksh-KHaR`i*yiBKRI zj`t$=vHSFQIWmsg2zj&PCUYVC zULcv0IfHzvae$@`s<@jIU?xIG_vw32P(QRA7jL%YPe$D%-&zVtA*~mkzwsEtJNJRo z_OYz4*-NjnBMp9oHO*XKfkQL8Vh~nV`#C3fTO!;#P9-0patAOh;Q7=?n zDOqi+?O4a?6)D59kLTf);VnF{@M+mtqg~9w$L9o|=<~V{+cL!N? zXc1hSpvG=fe8zmsI#2FYwNiI=C#K_wmsqLFky*N!5PtGD_&uwb1e$cf>t3XrUzgI> zsHviPUtiFFof%Y_X{P5hRfRd@9x!eF3)VB87@ud!VmYcuO`lYd*G~@6X2~$x&E&(~ zL-Rp@ND1?CrwQ}?fgMVU21GSiQbexxZ^%8@7Gdt(Nq>HqgQm>Ya9Z#?fBJdW)+FyX zQD5IB+T;F}KGzGRr`)38%8DG28WsRxoMk&cekmHRiUli$T$-~fM657O3l~rCfITZ- zgGrN;sDE+@87BE$BmiAScU{6r=9Ew3m9yh)*#IMS6M4~<@4Upn>p47g))lpm%M;ay z3Pk;d5q`Ic0g-DN9Z?lTwa7Hwwp@hi0m3wt;12Mn%8qzPR zkwj5h*^?(FWh*O#navrCxb-oe@cfwyvDnzHU(br9J|eE$ySfwE{6PDvf#Zu$kccucFiL zf>$A89_W4$6K>^6&eY72?z)mF8aH(UH+9ZMuKn45x?ue;QuVul`yT%eLZ`TslGGY9 zt7;E@kgW!K9V6L)hYd*>zlbn?T9EqXFq3;(o?nh4R9K2wYwksSq;`=pwbRIibCTdH z&!ftaVHo0^#ORgSV#~(?F3Li1$Uj^q`rh)yhSX@5y8K$tu$QC74zW#)&d+(QL1_Y1 zuv)P0+e~iY=5lL=z>~P+*nikU&M;H&7vfU4W^BK{0ku95fy<4SN=~Rs=4Tz+PSq(Fq?w&QyXuu)vAUF%4mT?YFX9(;V0zM~AXfojb^f4q5iB$3`|Yy`J1& zUkjm^G>PAaTBtpE11=aPfbqN>)~~yUbq{Q3uUc(~dyD%>&u*k$EnQ5K$}raOz-_p% z@Plzk&jr3>E79L9#g^9Fvg=engVMG@*qu!1_h+}5pCcpSOYcsu{;UTXUoK)2e;&qy z2hpOgSNq{raR08Xej6MA+nVJX_L5;`P2}e9v0!Cn z%#P7q1PjWYAZX1_(3bSacUiV@VZae)$PyL1hHj zT>ux$Oz^3h3@f9oSgWz$#G|rYuzBJRGWEhlNa-|Z%g=Ackv^Q3tF&r)om1MtIzNH=&`{DJ)e-QX_Ih%jg5v~qr$g`izh(p?Sh-kNE zkIYPCek;{;JwyYdf9O$z?rze-{S>axkL)9_R91iVJhtfVG&q#v2kqHrZ2Qi5a9;X^ zdM!;9O;{{fVx*PNwU67woH@Gz(3_G~jz?jS;ddcJo(0!$p9P-^V>a{QWVYe%X$TD4 z0K>fd0LT6&<=K0&Sh@i`wp`^*Zw|n#(^~BI$2VzkkOx~a@gkXvl*XJN1q(g$1Rv3C z;zx$Vo9Ak*#=lhd;!0lhcVINru(zMt<)jD|Y0(h8s0UJYwb;FWx$M>tXIRq=c{q?e zKvqj6z=FXO1dEQ7;d>HDaOVd&Vv;~gN@JK_u_ya4UX7J|dk{uB^pHuBzA!jl7Vclk zK(Vs}9DOhzBu9Bb-UE5|Rq713_e_QDWZA{^&Z`bCJVcjQf8TQhus;Ug_k zkU@9H0o0iF1poSP!r?nF;maO15W-koeV?Rw(uE8>^-WBBye{M9=~jI99^pEv{XqK7 zl_8v~4kgho#UnBN-6iJt zvj)xIxqzqBCUCJAJP3UXysF?5n$tK%7-7H1lxA7p=K2_Z^!k&y=R+#4$W1|R<7^oF z^1bMd>{&D_E~D!k6>-m~pJYi#KdQ?N#R*kc@p5Dd9kRTiJ8k8Qeene}u4oRiTW5-i z@jAFLL=q)*#z6BT3!FI1hM(8#NMxSxBYBmYcyZ)Cy8h=Z9KJ9L(X9b@9Ze_3<3uQ7 znnUmTt-wzb86rI+5srQOlN|bbN#Kk`;#Sb7k$uis=y-#E3JaksGE=z0HebeaoH0{= zuaNP@<5YH@1YS@-K+@EY&?(QAg`K++c&6!t_Q4(;cSwtV9aLm?1T;{q)P6kmLk@qY zN71GWTWOoG0oPL*ff+-Fqs)F+uz#(J(&zHZzRC#wj-lQ}ZH>ULD4EDqFMrM8MLDdz z{*jrx(ifIrOGM3yr}1q^7pbnXB8O#l$hzS#$dyHV$$hOd`Zvc2o(nn8Awx%FU4lIQ zcC`fmlqb+^16QuS*c+lljRd~KBW~SFed7A3h&(u~z=vADr}vFu8RU{&szU zY12wbXV-lywh!Q*E!&6#j4RT!Vr+M1FfQ;ti4mR!OB&sfj{itqwM}!iH_>7bel!EB0Y`k^tdReV9NA~$G5#QEPK-(X^N;xWR#{!BiVzZGxc zzEXpzFp#nPgcYVIn8u(OI=jybRE1sLu_HzFN$fBtysnB~-*c3D^aR7m%V*)-$}BkC zCS=)||K=lD^;L4W-z(cx22F;AZMW{AC-$b`Wx;3s z%dZos-WPa8LFrV~k;N4aP9yPGB|&!KHhLvK1x$R8qR!ty$eHc{2e#dVUf(60tfL52 z7cC|E`!nf|b1_05+l3Dry++(qIfANQaf6=v5!ff+NGs3XfYDRF(1RXEpt$oL)z6oM z%Nj^OUzosrja~+|$#aCPjv=X*(-!@h&4>%e4$SQIg?6h1}%!`7ZMd=Ml_zUn;Z0@@HI zN5$c{nK6)4c!`T!<%|mVB3$F4NpW-u(k3Y6wC$i(LZ4u zZBQx5FUI1WO|jIdr-xaw{<3I= z6eUA-5^>9#x$yno7-qmdf~pHnf?l^F7&Tgp4{sZaISL7+;fyZK8s0&^X8F;V5<%qG zmwC{A-;~<@UBMOgz9#{p7r8tKCwz2O0=0fOaq|w$pf7|>&A;E_P?PNg`lEHo&eAfj z)GHn<{Iki5y*aeY*MSclwE+hmlEiDyXklQO3>bdQqnASNi)TNcijC>lsF&Ys%(k%R zzp4o=2g!p_zfzNBuE&v?n@^E0qj-2U$pV}uUeSR6#Avb76vy|OVb3!?%==*mXXs1X z;yRHUf0&MEt}n+W%gpgs&UET1(@YObAzDUeFzUlL;8@2(?A&64_of-~2QDe_Q(gsO z>+3zZyX!c!viZ4?$8SOB*DLw_xoJ3;TgI!KSm4RJ3aTpbSL_u)zB_-8(49 z8@?D}o4fr!zuvxt|G@;~^9G@_Rq>isjV%#>Z+T38pPoeLvKsDwX$kIF&0cza=$l+St(mhJb5ISl_nBa-x8SSx_)gv4eZ*_l z`>52u(|ob;S1~QllbgLdlAbop7JSd~bi|vRxTx_led7^957gI-{yljl(mHUC@P{?| zxx1d=Z!a5ks5^^cGb4E7^Os0Q=|G2Q4eAbbQXS!WS#lwawr$KnZkG*TTRMyHKeit4 zV^pbi5zvcJ^^pJ}XnTX~h$4By> z|0eMxA}^z_hcYHFc}y9@XE^i0Ec7#8$AuT($L#@o>EOxvgxzS10n;M*h!$VCkZr|t z*S3j|@8363W8;Pc?A5C-lfd?n# zlGd{4dPb>5xKbZB(Z1fK$W>c?b^zc)xnJUz#pRZK%`U3GzD_K%*K>4#hW zDlx^u2i3OiL(&`0+?%0^>O$V$qV7A|UHd~HxjFK0uIuqX?w4Xp!AyF7g$n;*s4o=J$=uczn9~mK$ZY~-I!$l}cpjKO!#Pl(v_9sLj##ZMF1jkPK#@L>F6JUO8gqT|!~x1Wo#Dk&C^ zn5tp2(q_K-<9wljhixs+*@Y? zOT`;l%LaM!r9_9c3Hwm3ID657ZSnMb_Y$(XdJ!&ouvX-`mXjB z?J$^;1do@E;Fz*SpI|cn;A;a zTnrQ~T67rVHfD+H9yXHdeUiZcy@_vKJYmNEFm}`$S$3JsCg$A6Q{b|_3P-OIvT^_F z+2Q9hSnmQ39CvAO2VPl=)1^&l)PcG9TYoe=-Z7l*d)EiuVPpox`nX|127XtD+idyM39cd9RJ&`7R~?qyCcK{Ymug z`@_V2ale>VW{E603HM_y$ie5^*%>aac;(_Oe5+8yY&;!*3I-DVDUPA{nGUI73RL69hC!vt7cD%5%#$X0p^v#=OplOuX*R1gx|s z<|8$!uA)4YUj0jBgC}8*axGq;e2reBPF!4P4(qW<@Kv1s4zHIQQ2EYG=JwW3Tz7O0 zyMJ^$i=h;jnjJ44le_|cHN?R4bOpNO&>uiMYfuQre~zXbH+qx28#hFb zl^&c#VVX#+kU^%imqlOS$HIKaS2Vpi0Ny+1ai#*ZLBT&6&y|+pm4QzDpmnWm>ZiH1 zH|{r>nsc^(pOKz18-hC{R)JjAVWRKfWb<~G1TJwZ0G0f1*tV>aJk0(BFOF%!ii;&& z-BBy*jM|*M>LG6Qg}0pgs3)LV8}EXtncEyx7>n zb~Q-ZX(U?O{kgV_OSOK%#}Hl&d%9{o-`FBl2j*T*ovM45d-gXy8dK+eM= z5WOmeUWc_U7M@ar=r=Ec6kfsFASW($ha+%Dl-U)N#}NNP@S>KcgY>gqu*hOo z596+dWXcS0%1#wrI|>)M8#1BP?@2DxCvbDCRP~tiZ=$(!zo+0+PH9xnYGS{jg^3=e z4^}th!0UY*bEkPYjOyM`y@vdtHoCfLVv$4`duE$FV-20%hpMthhZe%+4_c#`DO@72}fZ1 z$wa*MK8tL8X+g&{+ru#{4k~i5GO5!|a98j=th}hmx9(0B2j}>6O@)#$UdSxgE^-6O zbEo;TW9t0slRGh3&PLoj*Gt@AK9ej;ODp}On@-d-GofAa2#KNAq_pQc?kXO_i!!JU}>xHjPo zoos7IIvU2XMvFE>!zV-V4?!|BVhfxP8;5;%!a3;bHoWZ9Knvf=LHt7{dOpV%-#s2f z-X=N2(dHw__U%Knt-V;WI+0)htA-vbyFyh$hk#bvX=-uh4y5_LAxq9r$Cl3=^YEgW z>gKlMOd$(axZe{WO68EU>Lf6n7X}R;?)1SdWpVX|9-3}*9D+wZVjBC;5s|7YoOx=2 z&l2US!IBD)5qQSaecC`LDHVVJ8_B!ohJfz!c{uVnNA1FDi6~+QeevX}R1OvF?3ov7jkZQ^7;f*hOo2=*3D;|+}oYjLv{e+EqA zCwlxRaM5og%Iu_HYt2aPhPMRec90A{gZn$fk*kgsyyTV0Pzudl%RltiaJ);|ME`D0|$@FN&va#-+)*wgy_(Qus}V;koP{z`T<`y#0X-Q$`3GU!5%1q&9FR?J7A$3|dog zzRNF6_q~UG%a0LwFM<`qmvu`e+!Wif1$2_v2jxzKO?jhPZZ8@4 zHFgTT+L}{PKdYRSU%NolFI+~AhgEcI%yCloR~J`Z&Zjp`CUV~$lOU#M7?td`r;V|b zVA;)HGPcD88SCXFe#{_wRX!UO)Mat#^&^nM?Zsns;*F4>7?8Fd zq$B*8<|mO*T;s$@C?}B-yUx+zEwR}4`5mo#QixNV{22AC)@W#b0>7@`L^uB#q`z+F zkaN>inB=?<^j@nN?+f`N(VH%~Wv#;I+&9JmKS_{NAB}uLJfnQ(HaYa^7&+9ChGK(l z)Gl@`F4}a7&NeDff~`5E|Pf0+_5b=>j%7?rqg zN0TpbFK3dPUP$?O#h_jo zK?78r;3AQNyuuOeZz$=C_zLQ#|R-(8<_2 zqY7KT>Vx&G1aV_W3>oyvLS0ycn{qYzbE#E$#Vnj|9d{Szd3cu%U-pphW){(d5`W7^ zEx#tb^IwIKYp&w;qCO^BSJ<65%aW4KCFJfU5u`MYATPqg1eS~m_PHjBl03U@zU`UF z*m|B6nb%dI&vH+}i>Y4rVYD@$zx1n^1Z&aAq7<0^DiZ9Y6~*T&rQw7|2lTr~m5CzX z;B>tTDEGGk-RSXnTDu6`kKY#mJg3au9X5pZ5LoedGuw#ef!oAz*F~y6x176GRZ179 z8Hx;lMp4f#Ui{e$vW$Ls8)@EAN2WXbPtg~A(uL!zaQgSh++tI~F=o-ycpyFxWj`Eb zikDragPa;VY2IY=+it-3(aog2v5`2{Or|S$&7#2#X}!4934Mc9N=a`Ov; z`~#lMpEKFm6zvZ(&w0{&#sjRPYhmrhl}ug17kZoBi(N3gVyM6n4e%p(j<-chW@-dhx=ZsmvJr587bi0LB}dyf zyW{4&Mr2BUCZl8HLK5T*gg&h&2_BWjD61~T^M9mhW|lnM@Lb9m+|LqoWEYY5)wI;u0Q4iCiS({_OmeLTxeV@ruPhq) z-=fOqYTOy4N=D605uXN~BY}-O5$i@%X*+eQ=B-jDA^%WtNFA}wt7gbqhitgLI+dAp zRf@J=NhCvh9&rD4*U()qM!5dmYr1b#3l#mg2>!VY*xYqHj;c|iD8ElEHrF%g;1%WcP5 z9;Z!P*O9~D47dvQ4$e4DnVbp;VuChm+RB+{iRSeeFfx}n!X77K4-%(nQ&92%CkSra z%Dd_`%he3OuNj1$8SdhJHjkO=5!bnMv#qiGe;(_==P*m~PCt8<%G~=S%O@#JrGbZ6 zal3`y)JLr(Jh)_N)rmB$%u{}c$ zMPBAQo}lpG#8Z{m2K>a!+lfuhVa#-Yh;?s5QT69vdh^&wT%_I)4?Vsy`U{@Jcg-xY zJ39h%^vm!=hB{k%DF>Yow4wTj0mNxy?76PV>m}#Yt#@`~v%_695gGB>Ys2u;?PCJ= zUT{k+`ia+Vg6Y?1uj${xP_kpE6txr>&%SrH*~1Ad!Fl|B`gOK3rWUUz=Z1QbnrZ#8 zFgS->55Z$^W zFcR07g5Uf+5*K-t4(&Gv#fL3;XPzv-c0dD+tkdbh5M`WwNg1qcTWRs=W6U#oT{tWm zj7iEX$O?m8k}JN+^sMpWu1)dB30aaTd%_9|u3o3>&cwr>RkqMHvWdnG2>duxd$deB zESj&B303EZ(28Co;&DA5=RF^Z+g~`SrwByMYY5;}7ClH(^D% z5pj)MLrpIT8AG{T@x#bOaF?74sRc$%RnI7r{bLvuX~!{!GGXvyr*K~1GoJh&uT9Ch z%Vl4NMUeV1C%6-T1j`n*psZ^GItn|5$W(u9&k%Sle!uCC)a`WL``2Jm{EKAXeqg)M z+<@wSnG0oCh2J*yF%B3~Vmwn4BzLW*%V#B%V{s&hjJkjh+&Vu`^Jq z{sMDPU|L2CY=}LrGnjoPKGgBc4Mvy5fzmS%@Ds1YNoT7_iK`Mho*N9;quxTs6-`m4 zOe<=ge+i$pj729RR^!CIVYtjio^y!~gO5|J`GFz26vV#S|K)Y{)JPGRIdz(y&zGF42*n4BXDAP?I2yKTSKy;74^f&#;=YHv9k` zHEo!cqKSzsOVJ=rQP`ITfxTf5{a&Zb?cjgWfPqbnCUX?ZHW`URmUz&&Wk=j zwFR7GW;RS1lw~3_l_7b0HkIMOpo*RiuK!bqa~8Y9(*Z?ve40;7qgTUzhX_vV=?!{m zkq?hZ*vN*gxR7w$>Y%6umlIBh6~Pl4cfQIkCU}O zMmtVEBNI0ovAvSjq}s=g>kX>K;;q*ZPZ{uuR`R^lS6y70VvPrFXL1F}d%4f$v#{EI zI6ER;hC9A!0J~O~;mcx6tnyfn9R|g?DEd8l+k27T?0m><+R{aJ%hX_XuriMDo(!|s zXo25!V_t3KcwVyjh`6`lB9iIR^g)axFRyP4!e|#)4Y>s2&dF#icBNsGQ6#L{5);a@ za6?cRW{FjxOeO(i2Zfoe>O4Hz`vU zeDY6WkL7ido2K`k-Wsf-A6}2bzs^~ZrF$Dg$4}-zPs)Tx)P>LSa>bbO%Q5m)Fn#*{ z0*pAV&svrm;`8_^AlXn2Vy~q%NzYpx;U5Yu^+BZeuh3PVcMSIAfUqZh1O;Uk!ptiS z#_hRGr>RW>#m6%VYQN&_@}yBtS)1ARj__`eb5U6-l>~1(M_jiswBrQ>+nbuP+&_sv z4t5|?Y8$X1vIv&x7vc}I9n7NxzY#A>!bg7>I-*^~PJ3F(xOWf1;k%cX&75!vW49^s zyY*fXdnOHgMke8cS4rexMLKt#eMhu%r2ur!aTOP)!DC^kXFYZX%uzFm0NHUb~y$I&}ed)plmh7#V! z-h4$6cRa>WgJedBok|NeVqx^s;p9Q@Y)E>{lK1x#$)iDMAu~M!+HPbs`<%X_+0%O> zSqoKeRDvU?7Um33d=%-?YQZJb`Wl~kDdCIB#?0y6BAhBVB^e$nyq}ahTUo5be)`!0 zTwN7rdVF+JVcXJNtYbf*YcIlzxsvS(@p0A*^)|72)3S$Y`&lJ4f$w^qDVK<{-_S+D&wvBT+k9VCX%bgg*T|<@66w;u%hWu7Z8K@vtua z8Mc0q;w!d2CG{`M!K3RtcYAXO^JEvtbk3a&PbL_!%LbB}U{y&h{&^FXBU6bKjt9|p zG2FZ|$i#oOgu9nd!_L35#PnAHmiJ|1O@JmS)Mt<%ixv1q{~i%3Prk7> z{97m<`(_bQ2-(iMXY9oM*SbRg?Km7#`bhXuhe0PS8%A44GF~~)LB?bh8|ecc7L z?vXBEVwudW!I#2b;{h>=UjY?kjrn^iH(;@<7RfvFo61((;EJzTxD_1^Ws}XcVeGpC zFfkiRtiA!q_Fks2R}S8dzh=N(ts-xCpll83&h=cB)u`;kP)07 zTt`I~4pE5WUI%25qDu~B(vBB&uJlss-u(=B7Z##b@D=iMUY;;(d&Ffwenjt$v4c_v z19AP@p`ui+1!TshWAwkRZmegx0jR%ZX^#F4+c7IgF>CD0=yd5sDxa;!2Dz`GZ|>}6 znujZ~p>aE~FpZ;0-@lM|IzeQelRugJEL?1}I+2;*xQG)k9?Cmqj>3GGY4F7NG;F&x zl<`&nNlt%HB))D}iP^I^@KARmb=zY^|2td&=JT#Hp2KyyT|aWUb*=h*eAYdDEbQmY zR5>t>=w)mc##7s=Zv_^(6DWpVfF6}LnD^{0cP?cUS^fN%SnN{9{d#te@t##itUSHw zU^*{sgeYFFUWfbo+bHv&23s5&L(MG4@(2DCxK&d3>Eadh@H#M!0(HuWI)Z5op{ol&-PkC?EZVs z*cItQ_Q&lw{9FX?EIvV1ZdTd!}C*Q>7lRZxKB3(YSNo1 zv68D4C?3-0rO&cN!8Xgb(6ofU{v|8c-#(87yc#-JBE7VQSA^Ky<#<)vs4zPG(80myd=4|x_l#qnsBZ>KbPC^va zLt&fZVR(6C4(xh2j(w+81m-&;aK3XeHNQ=%0W2JYm0 z`b)aQ{ugp?iP(B|K5;v?6~``+;GDu=(Tx$uQRQhRHeMQw%ZJ8bQ)n?G+oXWXJ(74R zD}!pfuOjy{Bk7S{Cq;V5Gy4+mQ%f^Ho1v2g*ZbC~bZMX)cOWf+&QjQq9fuP*mzF58 zq1H!1pgmeZA-7ZY%f(b>Obj*Ll7b$_ee_`EY8a-=le)>PxHTL@x6d0(cB}M~pJPG< z$IL18VbDl&R$3&*W7~WraNwQ{LT(~-qaX72Q^W}INNq(7NH~VGia3hL7Y*>a`)UnaMNCm zp%&hbsD3^S;}+*|o>?2p7R*z?uGwlB@-$Jr$mb;uYK|wH{EoxlwOT0qoPn!Fy)>u3 zmh9@YrK{|9P!DBk{HihZ+w?eEwPy&7Te^(7($6*Q~GeTrVNLdx4fn#5lGIO8@p^SFnW=XN6@*Ew`0*ATTFwn9yY!Vo=}MfU$K$Cl31 zu}#%JIKxaMpS>^R(Qluax3QP0By-lGr|<+Hnn)riZs-v$i_^?ApGczeG>Z4;%0bw; zFAhnq(1lzifbHJx!6#ms;}+*R#PDeV`8s<6mS4M=h{U$xn+DPN-K8rys5u8EIo)QQ z!_v{#$cv=oMGgC2R)Nm$xKBbSjmVkO`?y}u2wS~jm>>CETxwg3j+>U?O*PUuYQaqA zw1p=BkK|qISXu(JPj`^&rms_D6Um%{JC|^b?sAB{zYY#9UVyttI`OyM+f2g|9zJoo zn8+2e_xPxT#4b+?|5H|B1{HP?(|2E(JNecq(AbC+iEC12x6Yz-dnAbdOC=bpnuq&^ zN*ML7>ExQ{MY6y+0^k33mAsGkB=G|xiLm z(0ROZ$z0sA$bmeqi9vhDkK%#r+j(Tm!am|ZjPG9lX9 z_y@~gG8ob!YFn(yi?L2ZD=^IGyd389-UE<#FCWLs9mcwTE?AptpuDyha{}gck`svL zZ!L?*kGiCZcL3`p{F;rc`rSA;w`{|mYl_Kqo)5VazlC^9<`F8Ule6bv8IF57hlFkJ z$8|O7$Z0i^L+7a9f=NXqXqa_(p#*;e}pTT6El)1Yu1H7QJv3TcuYqnG5bP74-T zhM76S+N7uXCAQEmr5?E6!`T`}NMPe$=C=r9T)kgIS<$8^PveE6nyMJffbdixTGqKGqf>9+!PgwS9(lRGp9jgKDZ)U)>y*b;3J`+m8j=Z(M zf|Wx`SwHJ#qWbj*Z2o%;bc-~QPjE2~o<5(b?)k=XvNY!JjoQ!OQ}Z7Z($|7_CwZ6_ zR3wc;(j>e1Ch~fcfsTnTB2^oD!EB-fd#Hy|CrdK8SvoEtb*vbUN-u!lY<5d-`!IRA zw4A61{UzHorjy2e9bOt6;z&F_n_g9_Nk5h8#{L<9$RM9adJ}8$v8N}oO=vAySZfcG zC!`>M-f>=0+v25LGLFG~#X_djQimK~89{!&9Y19BP{qRXEw^N2dfL_ z5sk{FSi`gjz4;xF--H+9r>@Z)y})r^x6Cm0OT8Nj2x02RpBCO9M`z-5Z98cQU|G|D zJWVw>ZzI#3kr2rI;=0x%eTw{_?+_DpMZK&HMgC-)P$vYc$5)yib`FTzs zN@_C6Sg<6SNb4co=6ozGt4b1jZWGfc2eQkp40-QwAp4$p6Nis8$o!94M0@f(6S>!$ zs5=mR<&HY}xv!cTOl8kaJYoF7z8Oo$Y7wzFxhVL>VXA6d5M$?&M5cM!@y$=JVRMr+ z`9ivj;l+ssAcns3?kvjYNcyZNiajD`3w&G5Rwz~Bw}O7;*SBVnt+FJa9hJy;qKiwu zc4K=7K=-^akgLldq1xFhD5Q1VjHfEkY-p0@zuPZQxLgsmvTi#3*Pe_wOsX^YR6DSI zOa~rZ^Ogt=q{9!%98y0ml?e)yBB#&k;hZHenbRAWqPOf^Lj3lBXji8x;eKC>9|j-b zltzT}OeeI-lZAgN=id`BFn9@0iCrMmj_e}#nf`beCx>hKbt_pBJ0DwF#N(BBZ1K*E ztPkjW7WHCI1{QcO%WwRaL~gHMfL7m=hS67kIAl>9+NiPu9dL>fS*%msnyI*&iau!kTgue zDq>D%276Z#-C6NuzvV2hn~oj6-uH)O{9}V(xhJ78_A-&9R-wB3I^cfeqGxA}@yJLQ z7%$eM^Co+tQ)(8}&Q$}nb_+chEy2d+v@m@W0EpmW+1>x{pzmW55KkFggJP2aj=HmWmQuO_7PxMT61cl$iP;z-S zoOz;x$|D{l%Tug2?ARP!-^!kAr%vNDT~T0Y{u)FbE<(tPaK>J8B^hbE&E$5CabR~b z@n0_hhC}MepnVf_NOd0m$rZxnc{&KbWjje%y-9;f6^staLGpqZW*JN&wD)|#J3)=o z*{=_7_UoW3Y8Q2J#teFyoGe7deZ-Dd|3ODzEOZp_M{=S`Ah+ESrv8nf#nOD#KQkE= z#iGcFW+WMZlf!b#B;e(`P@c!HZE)>zATV}`ASofptMv@yW*9SQwbV`I%w_?LH0HtW zpOG;CqbiyD^OV{u98I>Fh=A=S3jID%jZe-DqxR*faH7R%Xnm}TQj2%;1U$rHv6DW^ zTo{g&?`(oFmfdzlsS?fYsX$bDIXZq_9~LgCNBR5B83mgnl9X!5ukLtFDbUyPI-zUK z;gLhU=Fbx7IQZk-u3wyyID0xzDFnJAO*sY3aoox}%Dt{PF`IAhf?M7yA#+Ayzb97{yJA-dat$(v;`R>8>JU$WzjsC9>=;9n`-gER3GN&e8h(`3&>m1XV@Yy z953k^f_p5NQ;>CWPUvQn$~R6(ZMut@dTcJNTpS6jvbfyKVH%9~T0=6a9mVrCJB1FE zw^2?kZ!z_^HhDe2gxSlP!xSC0<`kSUB`3or`KQK{>A%Z-fa|>vcoJD?-Lo`Q6%&VF z1ewv_Vm`sVo&TYY-*VuMO9pg)xl6uIlO`Q?=4i!>6clV=0eT!my6X95vff^RBv@M@ zS&`5sYcCg5Q_HRK44;*Vd8~lGr0P;v4g>ipXWs7VjJm16M8LLR02Wfn8t^B^iv55XpH6Wn;mU&FnjMDB^}$n{njp zGS>RE5U2|c6jxUe`r5LP6n&bR)jkJJsLMdQ{$kQS%C^_#j>6U03_O-`kSeoF!PZCB zpsl5wBBCKk)7_N{UVj5O}LGIa&mMt__!lDvoytgWR29!ujr8GL->NG0UOyAZNq5$th1 z&$aN5N7?7Xk;F@7bZl8XQ>L|^b$CDHkEqPZ)omuS!0hjuq<&!$m zu`(qsYLUd;Sc>#4NQDaJMCc!rBzl82j8bJ9x@akjoiQ|eK%XcTpzoQ(!7T#rb?Q?m>!xH4sg~h4u|SV6FIL0 z>JVLvHubzEor+21zW_o0>O6m*t<5_|XX!O?*euTYC*@L0@+XtM+?ix;J(`#yRU#S3J;d?w(U)DzZ z5;EcMiZdu#vXt_VuOWTC4ph&U8}Ln*2k-G=Hf?8_s{3} zT#yUPe+%YRP%O{#9@}&B3cy>ADnfmZ64PGJu1lAPc;ft{_^`kC;Gvkvi*%jap zt%9MD5E)4LjTD+}OTvmSdzf4e4jJ=(%>DyU7LOHkO(*(f)m6fPz=;CCQR=hgcrDbq*t`}OxTL^9!ia_RwImrM9EftynIIbEPCNv-bkZwef{b z)Fd1^Awy?7yv2H5lT5t-FgZJ{%)?`tx}bCdXWX_&M+{y-tHwHLu1Z83zM7zjcRIv# zk0D;JnugpC7NTXh4IyH@i%M}(LMxmT(GjWlc;3NsbXQ&yZY0iU_nJHsI~;~>bagnz zre8RhKlZ^Xjw9?_8-QAQTxPp_DV!*rMvJvifS0NdeXO<^u2yu>mxhPwlqpYIhW65DLT#6A`P@MFCp$pcg$aE1eJGXjq6mJSQh}#K%INfWNw9RVfU{0BX!4c@TvW>C98vnj2o=@iK@S0>_p6@TW-^O?(_Dd+Jd{v+)>5c! zDuvqkBY(li!NWpPs=nVJq34(J_VdPc->!Le%Rcw0ivsRNu52t{vB8?%o038vX4!Nv{Vjj}j!K zCsb&U%p#QhOo|%+tcGSSJ_5az2Sn(%4eYFF^)#A%;$fq1Z%R~+hYrth zLMllu@QdZfjh2+a1GxZ7bDb@IrwCh+imkeDE5TueFbUo+D3(vKf{9 zxasKcZUfqMQ#)EDlnZxCcG6|yaWqW#V6HqEhQ>50I!MuubLOZBY7gHAGqju#9kdr- zNv6S!ybV;!m5*pwybw5>ui;$_eox(xD`7O=UxMm6uYkKJ61JWbrjU#i1472g$wq_z z()0la(%e|z*%r>yxx-YE$|(N0yA*?};>2;IbQAD^S$ns?F}iCtiBjp&6^0?UT~J4;u23e(GV zo5|n@Q zBamSfN?U8LflChB^zHr(lv$s`wMtFoeq5u5$geo||0Ivsy5|X~-kM2Y`ONZ=*v#*& zN3t-LXA7=Z4=|s;%tQK(k5NgvDa{``24Ywq`j_s)oU?*-VDdxIFqlEPWSOIe3(YVh z{1FiW30kZ5FWP!yI(m}Rgq^FUVN*~VHdQ@hx-UH({KSkQNmU*n5R`+eZRf0p zN#o>u$RNW())U~B4=X<|r3==^K;+gSdR3o1E%C}7tj}YVF3Wa*wvVG3o>fryNQu6C zHv>+6k;C?BN0833Q1mbyQRty zcWw}3jH*#bg#b30|6m26W!Qc(y$?7fH+ zuFb>kai-)^d>^{D^C|i~>l*wVvxjU8FX%Xv!0-puIZbCZiNl@QB=o)}u3pNn0Rd6; zV$V8|4=G^QnA_orsr{ta(GF@-uTjOJ&$%}jV_sH!Gz#Wrfz_e6AQyWaM)b1am%&}o zeGx~^6%U6LbxB&oq7T%M9L6deI+5*?OU$g_bAkG*K?0{DnfVg0P=iM)T6N!t{-PpE zXFf=TO%hEg{+kf4S@wbXxH_Dk&EmRvv+p8fp#-MAc!ZaeHOy1AlE8CxlW^r;8J6i8 ziSGr?0ILtj@jK6Ju=|cBEM^j6^~E~eo&6chIxN8cDWTYJyC0gP{}77^GdSv48`4UN zf!l&fI9;U*8S#{9A0-aT`EipPyPgPL9gm?f*@G->(;@q7JJ96Nz>uq!vg)qj}mH6g7Vfeg#8Svh* zJ$*?-c>Y)$TUvYa3RtYlwm&LZCjKU__YtIL`uV_LNe<{_WP-=kCde%mCI|9@@vx&9 znmo3F?cw>e-^XICEZ_~fD=c`uM^3@kC6%DRR0kVpr=pH?gJ{XY-JH~c2I?tqCUwqw z4B73wilj=fp`m7f_)xkDj2L^c@+o4!dc(O-aY)#G42VS|v*u_CGOn>est2}U`9mBy z6U9ZDVJl(C^*V&DEvJqhwm|&r0@T9BVYq+K5Bda!$xomAFh?2D1wNBtT$_mV5}MIk zmOc{|-4{Ri#p8G|t1FId!r;fR(feXMIG{emY>=SQug&+X&F>4TeL{(^k8z$2U5 zuzdmi9QjVo{9p{fbT_fw26LFbsG4>ST^WunQ2_IV2h7%)lH}UU4%Dof3H>S29C3@oWOHjV(keZH zX8FzoTYE7qef|nFY^;k_?(vcS1qCGRyq9TSAwh3-D`7JLv0$*=ndpVfgHY}OTKM%A z>N&WCveuhJoD0=if4E?x<^p4D-*YG0Xuw62! z%s2xYF(R-leol3TivdI}N9g@D!pyu^gdScS<| zpR!rh_8V3({qw@=RQqN)G*m$X%;Sj9c}e9=Z6Dwz<8adH05M8DPy7r!VDS?} zu>6sLRuux$H@l4WK92G}EvkgtgQ_^^l@zWm66Nq$y`Wl2D*9;=4YRpVc_~WXc+Cp| zTK5!&+HcF~)i3Tt#RCh<_SH-X`P2v%12Qli76NtGW`OatF5LUd3@Q97N4DZ}%!>EG zb{b^KhMD@5M0_(2)CnSiBa*bJ+(x*wO$N!1a*=DmbsRlVz$7d>jo;6#M^E&d%#4+f zV|~vM6f-*oWbalG)$6NxoA2C5W8vv!)kiy27Q2|-?qc)dkj3=o)ZmHN51{geI(jT2 z!1TQT%LIS2L*b)w%!}?Dl$7H+SmoHocpH}CXB$1haAh6z1b#x5xSH+y90QBMRaE=z z)l{aI0G&O@m2BRiizR#IsN2uJksZf{aeAR0o;I@;&zxLOWk+RVZgd5A!>e%KhiB>J z!Sp~%(lN)(-ieFk=06Abt!lLF<8j`_K~wsqmJ_|n@**1UdknRSA3%NhAeV3T9g@bf z=rYC}a}QM$$%FZ5QPrr~aBw;V$InFWVL4d-&3Q&*VrnSQdo0S{tM3gd(L1DW({UrWA)xdt@1*4;&L;s2INkReF065e z6AnBi{xlcuzHNi1V$Zmp{H1gUE#7MvJT9W>=pR-WL1@?J`GvW?K%nYUyX{KGm z>t@X-Nk`|Cwm~`c_vCD(r{aTth8gjbbY|00a!=8fkV#}&=!CK!?}smUgt2hFGFItZ z#r)lS3HMYAXa^clKmwkm&t zx(t4`BLM$=m``mS6~~e$SJ}XwR$h5&WnD;>z>Ag z;>HQ|@e7+hV`o@nW8zHpp;UNX(}XilH6eqtC$Qbii*#r3@Y{1@L`%||eA^#_UKh<} zzVuatpLQbXbV#raC_|#$Ck+oD7{XZ>K6aEnK)p;(Cn*h5{LX)hR4Z*wkNqh@bG|Z6 zYuqa;n0pq-y>&o$Wlp2&+CuX2bsjM@52Sv%2;p+oNPOjZ2lDE?g~Ad&$#S)BIH2nc z4iYngssB>_utN|(tKLtp4a#A)d5hUKB$oWLAH(8u;z&E`65Kf&2Le6^0iUge6zg(E zv&RJ&7+Vp))kTc>{Wenj$cUIW2BWCuD>x}ub>Wfn~_fnJ_SzJ{TOP}NgCHo(`1l+KT=cngnJK% zIf279wE2b-dVA=6dNgJ!>Z(zv%FYYGzb#`xr9OpD#Y`BF4~N9YG2!g_5_;NX z6)1`*z;09wvh2B|(IbvJ`;=v+C?u1qk2R#Xr2u_s^1x#93CuLBPfTI7Bza!Eii+~G zrJ2AANNg|Q`&dMfm>UCR>l{pW%sN8L7+b^LBTGy9g#KOZ`Akm=>FQN({D|`xM)dE3r{RPO&DgyeVKlalQqz)e&9Qk@$THpwfuG{)ba$8&}z1a7*1D5716F{1=QDWM2fDOk{dha zmf70%(q+q2=**Yt=r|P)FPBI|DK>$ZI#&_bXdSxLREImukC4;RZyRBm378U+;M+fMC-()7 z;fHx2PcL>CQmEC(nH4&8(6T$^mwP0q`lcQ>Nm@m0lEj#ZFk|HT+X#>xAE~gPJL9wp z#;?E=|2)4Dhka_tV;<88HTx3Ivp9*iYQ3j?7&*GSemhZn$8rtt8!*?u1v2Y3Tgc4$ zCggYb0cu&+GW>9F3Z|)>ctceR{+aWTt7~q_aC1$Nk);KUEX*V?Y&~#^up9HncLUzh z#fplLinH$5$5`FHfg2{WiuIRVLLR~jNbvn^wB~&-y(_?i9{emww@K|{Y#V4C3QO18P(VI+3kqtw>KN3Z%j?Jar0Rg>S8OnKC4E<{dw(>oVmPVCx z?6PJNh5mXpnibBZndS2O_8Z~intM#u9RXtDk_-yfH^__B-;D6X0p^)^CAsPRhWgK2 z5ce6gU7ygsjH#D6-!Ebt^?Kb7bTr?GN)clF7E&!}`k(}vdQ?h&dQ^hw!|UjY!7$Tg zmBDgAr}6*(WdHs`2g%9&5bSGlfXYAC#8f4*&YTbdJpb@j6!&&9^u24MQbreX^x~x{ zkvjv-meyFj{8Tj9ojXH2{JKUz6;NkqY4NzD?h0yO^BA2juE!e3;uyn$6?igc5Wi~p zj!zngqidT=ped2;8DNnAZl*&ZFu0;U9SRKps-8_c8-41Z>2kSC^L9$p=JD8Z9 z-^_cpybdjri{b9!shPD8v1`tgCKU0y#%xM&9_)|Rqv!gb2B&IUc%oQ_deYMH3XyF3 z@<1u=Y~KQmqaBiVaE8GX3Q)dFA8X320H+3jJaK*{Dr(zEPq~~2!<*Y6Bx*IB7jooy z>8>R=)*3OpjdhU`o{bi?KVV$L;_*%A^I-6792SkMqb(^TXw`}!_^@US^;~7(=L&7G zRJ5QPhhnJM2_M!ZV2O|HOhh$X{XuHLm6qEkf!ge@W5sqaZb+af;r7du;s-3NajQIR zEWXMtv1_B`P5P;7u_(Nzuo6lq0@3w{VQj~O3#*M8sC!Y2Ey`yylXl|R2JIm$T4oTn zhI(WY9|_&g=ipf9GU(_CLoym$P{ZF`_;IikWo*XHz)Gv#|~LviFC$FLU9? zxD;G`y%uixm4LDS9@rH2hRLQRVZzH2hK`O_k7$d~^X~W5SG3|WKmNwjTiKA9xK&!T|qqjY|i~RS*Q(xLg7-98P=oZ)veQ}~N zz3w{GAY=}Y-*mzFqQ!8ta53`P8G}?`+-1F;mN>lf0JYQa0(_+(B3YLdChA&WmWaprW_rW-N)>?eNZ$B(AVKduh&G7K?dUWc&5&4w88+S)w_7DTGSLa{aakpCLWfwH{pMh!q zKX6oLGVkO2N4WLSOy=!pJ(SP;!}*|I$!M=PpdUKCq46%U^Om7vy^q*?!)gc@ z)P~GaSu99J;4k?XQFq}9xco8|!gV8|rUZkg8 z3)^urgtQuBw}ZV@qwjgBn$Zk5!;Zn7KaJFBncv9%(i0@Lql|r<5TpA8eju*ecUUm0 zMq80ExUuvKe7akTimUH&4qWwtuU^LRpW+t!-||EFipyr?(s>%n!g#Eo&V=sr6Q@-i zuAqJ2gs7Wuh0G=;Y@w|n3GPpS2MhBDKxIb@_&B7&mby)#FRTQAew~9CCJyvK@q2J1 zf6=Y3SumIFAIg7Cz=y8Yz^isA&=nEO2x<$XU63)ZJU%p>a`#gH6K7DZqIr4B37hzmk@tc-Vuam z55Tz2Dti8T1z5j#F710?i*7RzWFq$HGb$c4kwp1+NE(iz79>Z5{m;uN%1i>Lmfir) z#%`SBK9fG`JOzH=m1zkv4ju4e5gq1khXx{qnDdhZXu(=@_@Q(eERMd0+H!!)Q)3*P zJ^PVMs{p)NF9zQ7_Q=v>3>n_Oh2jS*KxkhWT9Y7#7lec}QwI0Y-sdT}w>TQz3Ni=( z21okI!LP8P$OifU6#~T>%IHh05x4t{3kt5zr@F$6VAFMr+Nx_f1TGe?*ikgROhGFH&U-t<((hQ zHVtjV5#%N+NPdYWD-1Daxfo<;I9J}vfUpd8Moq8*1S~JphUQOEZt_2FlW;!Dd~*bp z-4?;)e0y*(sfJymR~hMlb8*M)$4E_oIz3k4L8oQ2JxjTFyab~fWae0e8s~_TY^g`s zwdpIKZePK?3edyvU3`(;{St8b(}vzt-Be^<20F7q0FBPcg(C&WK|!Jq<{wzid$u)| zX$&voOqSk*(_bU8g_nWZRH+_WU3Ce`M6ILJQlel(Ml(patpQK268Y89&bgnU!!npv zz>M81udO*^_PzKKO$^u4a}CmHo6Ak0x;~MPisIuvdkIIeS%b=Ws>=J%Oo$4JZpH>j z`?%gWUr|?T592BAOL)ZJh1`2p!CdY8kGJln5t{YseYL{ku%%Y6^HIC439&wSmb9sE zLz9gU&Ei9+q51b!=s5OlvxChTz{zZwWp#~qpRPe0G(TX@U3MUr=QU6+cQxa&dx0+Z0}E%0J{14bdbJJq>P#4NMrgms#TE6lb_?%kmlGS)a(qUhMIA1I!xR0aCrX zw8et?z)Lfvsu$)F;UAjh{jCpFe(yq5+AGQv&^Do>UDng#OU}Zf73xsmxeAC}Z`=ZTrHZ+J0XaI^uqEN`JgO`aZu zL(KY)_gLhM8O;6>jI1_E^E3nJEc=7r(<8UCAZ@)Xqv!pu+H_eknpafRr zuxzG0$7!bs=gkvISlVMjOlQYIf=@M*nB2oGc~prvifhBxz79s^<#&$qr99|Ul!eE> z$}F(|2e~?6NBh_3&?W_uux(W>2vzT*V~qEL{-QYABEX&gx^)a9{+-~)G?j9W=IsS} z%^N6Se-z9a&Bhe`B>(qDzV z80vz5&bQI`sjom)JQ~8iq1et4yV_yzw1It)&j$H?E;|^JJI7g*aO5&pbNU zeK9RQkd5~_Nzk5Nn)KWkQyfus20o<5AUCgMPVb~FyMMi=+OCvSI|>82gE<#D@1t&; znVY?4c1pTKUB+p=MI#q|vDE%23>k-n>5l9dPlP7q!i1l5#f|p#K%S z()DvNG%wR4dB4LsTY3_hM~Qm$mv7#vXR`&XV@@NWzI%WEgcT2fW&<8?A}!0NVs} z^hvFrN?Z=mF_cNBbGvb~vo^}UQg3?0NgscT5yG>MVYGU6HWJyc4Ck|YkZdMJ_n0SA zb2YV@9z8e4^~p2xfKxy|N4mn$p`WODi8vXbt;e(Cmr<9xm!jVi#%6o|XJ(ELbF8LHAoI*9?qfucGTgv z)uS+3tVt#;w&3SuawNoaH?`$Q9nyB+0xP;};IOwm?QukeAH}-}Q?A0e`*jXwy=EQf zo1+^?zj-Hm?{S@)-eSkBkpDz!UzLDg#EO_OskBt^8Q8H(fc|E{&O9etz_dM`cAa_* zC61=pcV!20ix9?<|HZ?ImJ;o}>jFLB+Z3c-pMcX!8MP3GMiMSb?FF#0fzsVG0p?Ct!B z_HHwPG06pJ7MpMApZgcBdTmYAZq9)c_ci#2(>)&hk56j`xzHi2v}u#kjf|$oO1!~3 z239Lp;3uAwM0DsR{lhQ{>Q+Y3zwYYO7w`Syt$neZb>n2Q?r=rcAC^Mr=;zYY8vYW; zoF0@M`ycautuZvaC8C)Y2k>6OOixGu}m^i`IE_r$v3ZQ%WZ;7$^bL4-s3w2E` zLCKH3(5A=?MylKt*FPSio?eW_k-iPYX2(C0RX#};O+O7ETb`rC(#dfDnmW=`YD987 zd`Qo zo6ur^KG^T*C8#F4c zX(TEA?V!(l2_a{equm@^ly#>BZ$Ed2Y}v7vFWi_-`g4-NxJ?ZIbiTqlrI(DKFVNyo zzA`|S6E4L4;&T$3CCi^(I7p^SKQrGu=aG9ttt6^BiOhCe%0K>j6>nLi02F1r zb2=*9NMKJNsSALDe(r7X9 zOj||Hd5qw_bJNMfocrYVrcvTFqlPSht51}YXOQ^Q7daO83y}m@9sIpck+Fk+$eRl% z%oaR9O#&L4!A?3D}9n0uM8eKVbC_!bcBKW*flLJIu!*>9Gw zmP+P_n1Q_S2C7SVFaCDpE-5@^PL{kjBm9RP5HnSy9~5Y!i-%`}$Z$Uh>KA}590srN zD{;7c0&LiG9UZZA1NX$2B#)g>ZCU&kMRE=B*W77DWbGjD*|}CmY@ZB}`@2~e%O5J7 zxrnz9gkz^oTX4Bz5{lfih>SfIg0n+*xNfOEY<6A^4fS*Qk6-E$FS`Xyu+&jh|9BI9MNKw0k^^Viyn-ds8|4y`a-cg0zRXG9_(s@IFe5~ zptk!nDIE{Tg}aXtX*h=8tuCc%xhHYZj`_^iwiVyq(vHe!;EP0K(-0jX^$~-1( zZu#@MhGxu%?bAt|oC-O0@+i^z(oGCsn(}XcJH~f&jo^2?Sny?rMM?XhIN#)RB40Uu z1NxYG3r{<@gp9o8qKON=Y}ciWIrqR2alLhcckU{_sl1$|c;xVAU1>+x*thEwCf;bq z!)RE&<0+0j9>U&hEQz0n0hxK$j{5KHGwQ)>ZIa}A8ElSygGK%YNcUG6n)p%9`XdxU zX+&YE<_<&h#$p0Xr%LjzGZ*qTA4uTa9cR$HU@P<_V6j;NOjP#VMP>>{`24AHa{gvBvCf<%>c=ngmJI0d*1u_{Y?IW< zEVH$EwYx0&wsMlWy>&M;JIsey~T($I+e zbX+MvOeO!b#b<{;Bdj4y=CbE14Uex>`=c<%;?x|{s~3j*OK*`+pB3@q)cx3VAe*=b z>M)uLDv(j3NZbo{Vejlprf*pF}iT=Os~sOR90I=NM7`?8mQG z3gnLlKf^2ZQdwr!CL;THHjcjIj#n15`>W1J=2G4=P_f^LKad7a*gRP@bmBNu&K1B{ zQ(LI?`m@Y>&M9&0O4#7R6=ycNfmH1G_zHS)yoRmHnBzLnW(U((4v4*=K zX0q?02PPKSKwN+<{&0!XGXIZUII|c9zjFtNh^wS}hZpV%--?%v{-Tr&lG!`o1c|Y{ zKvKra@W!ruyg9H6kKC!@$+X5Hi|p+z8(fj>ijT(M!sT%F+bp7}{)OE7&_ekWvIB`)MZQ6}~@ zNyNvB6L_sZN~pT!z35TeeA1F7#$VKMmyB&ybh_j^MZp8= zUBx`crQQ{4%MMdBewi{wMdG|e7xeM6Z*oYYOp}q0Wx2@F75q52Q~ZndGW;$#5x(Q; zQqm?TNCK})li-G3cyGuP)OSG)9;{O&8jlOfz{OA4>7O1UZPDcP1`n!LO&jDEw4ecJ zeR3g}?SW6m=5yiS^OMz0=Tvo^{sWv^G33tcFQwL{hn&$3F%sqwh+wu=MR&9bM0r4W`;P1G& zXdax(zQHtDwld4ESi#z&rh64%aB)_g-b*m$**vl-8jVy`akcWxxR zsxFBfxkCK4s0B5jyeFD;!_hGgc=!IlY~r6`L}pu)=mJWb zZ@dl#7f4jf8a#U>4E?-q$ve#TKqTG^ zRwLaoKB|K?=q$uHuHI+!mny?_nO7KWddQRKP3WhhJWbv|9TwPafwc*$SZ(oz+NB;H zsQy<8u8?s>auv5|iIEj~_)HB~^RwUZUwX{czoQwSUTJ3UiCH9aYzLb2F^Qd&99*~H z{Q-8FawrzJZDCdW8(81dVrX?s1=jUTfMO3h?#to660dq{I+ONrsHZo-|d%JMMeD12YY7N&P z=7(IwWznqJXNl)@J$zu2CSH1J9!3izg^ruEMTR>jlCby5boqx>G`hJO!i|rhDxN_% z(=!u$7N^0{t?QxD=ooo=ZyNoZ8bwnYPC>-FC)i&t3l|iew3kT!23vy1Gs=Clne3eH zcx_q?7XDKv5_9HbeRC`Lbs(3D88iav9)R_WkK*T+K_L3BO3!Qmrk3U=uxUve-slnl z#Lb7uFNtB-oI8V)t5m`D^FMqyUV)RFu#>L5P)>gyo=*z-E>5mYF`0kTkfQCjXx+Ve zG)YQ{oJgr9B@ebB-O>^oV_*yYk8G*ot?@9;B7`vCb;J_K)r8{M-OD9*f$5_$^PS>)c>v|b=Xixua&%`Res~)njvCBbgz===Hqlp&T~9JbvHfx z=NcCMSb;d*rNmSD0?Je0g=hW#&HlQ(m7WV#gSK^=XhicGWW7-Zuge$~J8Xc^vHRfo zf0OVipV#>7@e~-+E#Mu4Eoh{@fPTJMAh_P+3hJZfn7EWAIzGRfdjH!6#!3#jXG|gX z+?>UBjSPuy+*e??-Ps6!|E(ad6))@F-%O{5RTl(L#;n4@PkYFGmlf>~r7>LlQ17+aWFO-?DP(lbtvu;x5>AORJi__!ACM~HCn zQX$>ns!7M&^wBXXBh-)9QJDZa#xLgqNP8iA>324Nwxki)#Bh=HGELrHF@Z5xkwn!Q zaqO@Aqv$lX4aEELeSv1E60=<16)vtMFx#q%xuxTewcnQ@x0{k6J#>J5Q7R2CwsV=Q zp6NI?$Q-cu8-7OJOKs9MX+-c?V(OO!XNwG&;9(8WKk3g|##0#5OvBBw@pMdV5j#%o z82xP3P8A|QP`{6BAfxLzb}r0>?3<4cK~>=&A7k?L<1RO)#)wJ3c{`XqRU$pI%~H=~ZE z@l1l2z-}tO;|+aW(jxFG?ZZyf(rCppaj4mLiiuN622uSvvZ94|3~tM0ayO~q|MXW; zuhIspHLnQ-VXCZa$8@snk{d}bdn9_Ke*xxL+tLs(U#{6KhepJn#{P#s;|c9A1XI;> z*>YVoG9hyz?9R!gl7|C{>7{DXr+qr~>0mG|tBFH5$D|<%kAEn|U?R_v&cWMrR3J;3 zhc707rqZ%PZ>Cz+nKi|G47u`DJ`Ie2eakxlXlWHX#2k zChYNoDDs872V2C98T-yWX4Y^6%@D@W^b2N?@o|JK>u2$=w$*gi)$ugtRvq-Ha zTjbTx^Q6D;4yp5Zp_`r_qqDAGg+(NYx=yGfzjYPaB^4)Vgwh1KQk4f5f3Pt0mYOI% zKo=@wZQ)J~Er=lCCx(k^P z=l>C7A6stwtWvrz*@6ha*pPJBxd5*fL7xi zioEB7t*tS&x4Z|MajPNXp+2**Z8ZG1Q-|^89RY@RUqnhank2b<39bD)9sLftL-Y+E zKz8f_y6hkk-KmrS_k=yzb?7P*m;&xRAC1?X(+AIc+lg~5!OEL@MIU1`u=^1pTk2zA zWvT+xQ)&>wD|aaw5mx`{7Nze?NdlpO4@T)91ijT0lA)a?x!4eW0_{NmThZ3f)(J z1X-{UUW<0|{+2+n{jwV^%#FfJrLNN@TTJmz1y>Y(K%ePgmolrpH3eILOkj>`JcP)K z!))Muf5vc9D072LVM6^3pvAC5bZOi+khk{)6(s|3yY>fVYbLf-!o#cJ+ReZX{j0*M~CS~8=i<$?X0dMAk0JRi$LY6o=s-T6P zqgr77isR5db1IyczYc{72u}~&ir1^&1-ad>?EQ*LnCTb8Y}}O#<*C#0a-KUhdQ21! zjrqvhSewEoH)ANhar>4IwEjQ7Pmzm>zd<$H zAajb&M=#!;GM-gHcp&pr5+nI`>ckAE;=iT z!B@4+=j?+=X`@kMZx7gV=G>aWMDwh5X(v_~g7ekV%pU=e!+c z1;5i1vHzs#>-SY?;ptI;lFLNnOI?`AQa5PR{)|6r@V&tH-$G?wMLIq%g#3z(!3~NT z_?5a0v(zCBCTZ9(t?Ldm;}$g#3+Gm(F?v6Gr92M&%JM-z@g$zBpn>B;#Hf{wm}uk8 zcG6?q#8!e03Fz4lJBr`qj8CB?DAkU6c(jE%;k1pJzGE$Ow89k<19l+KI|FEF{Rncu zaThGFNQ3cO1LmzkKKP&cfd3e|a0bW4n9-?DG_H6rh&$M^$zo$gTWLIQ%!~#3`3Y1p zx`Ny(bVTV3)tFs7+sSwNT()_|HNu@$V#bd+Fv68`jIaI<@RN697tc~7mhw{c(}YY` z;jRm8I3EblcetXF(RW39JlpZVb{+cdS0e6BDBFvuXpi#zu zpe9j_SLGf38=)&xj@o}1cAS_xb!LQH0OI9RP#OKV>d(i zUO)$SpP5c~#ER(H+sd>lua#bo5M#7#LgV`?dwD5B! zo@Q+$y0rBnwKwUb4MswmA81G&Hcg|``(?o(dk#Ic{xDmf;DSvI3-HWSAK70A`R=DD z4-!A&0j{00*gHjvdzL@QJ~0g^{V5VKYWOKh8m?g{-B?2Vku`jCIsw!7Pl2dC=RwuP z4Pv@YxGOgUf%|6&?nNGSlx{Q4=?KJWm)GHMHk%;H)f{#hX+y^@8SEC)%zFV+=_Mx> z`t`Fho}F}_e=Z&)Lucdg!ymusc2)vXGoKQvvgb5o{wH?Z8eOQ-`a%|L9!>v?_(ftZ zg=AfnD!#OG1rqysk$zcbiQCS~G4DtlmhF@vvhO!S#h?MLPr3(X!5NU_s=zkn&!>(f zvGn;$S=wnH5y;Fu6sIRR8cjFZ(Im-&C9nT<0{S9T0854)6iS!T8$|0u@ zp!a9oN#5IfIQ!Qa&gl>ork^7EoVJJCAxlX5TQk7pq_A^q5`Gmk3p<$Jv0oEy#J+Z3 zO*um;Op*i1xD`#P?9ghe<)F(I$1I_){h=Uv>?QOM4~l}OYXK6(GY6Lu++3GNetX>G zSyOWK^?e1d{ADJ+HA5cv{HF?UWz=AN^I|;Z;5pXh;v;hBw}!xfraHIE_ZdBX$C4c0 z<_LrGYVhQ<6L;_&rJLk?$ihBlR8@lM#R&&V{OO-q&-E-Gwn}8JiY&0Rln}7W1%aD_ zfZqDaJ67ZW2t7>@edb`qy%Pixr$jlLH|`_e&~;7pLSGA+MrMKSkI~SlFphSn71Ou- zW`j{z1UqSS1DiQ@zg^~{0vhAWXXBSfb5C5B(Bh2WSTy;V@Z<+8Y86{R7Ov1He`_$j zQW7D>R0r}<@)?c%GM!tyA)Tv8zYU)DnRMX4bSen+M+U7a^zMmrq~<5hrf!nr?mbw| zNoJK|o7dy$T45Glc3~N{Q^=z2sf9Fq`vMwyu$W5CFrm_|2~>E2!hsY^X!{#Qzd0ft z{b~;09{!r({aujil0fDqE}$xR4-w0qjv%jEN~M$rkZXX ziQH&!Z6-aM_yA7{-cV=Vw;bmRl&H?eC>p@>4$w3iYMp%@O-a_o%MSbpi}!C4^`xY- zf?Z$mBU4k9_a_{zmX={0ti(87mIIgaNcg%l7iV6W4(Z9`QE6N+pWQCP)3k%Z$LBE~ zu9Cnd$2`%?d|T$qNHo*+DHE6dh=g6c`MHVGWgPs*5o(98;FXK}pkZb!$Sui$k|aB3 zahE@w|EobS#VEqb+rIcStAN)s8cgETD&llxCX}cT;n^>J;L~+CJh*QJyT2B)odKD+ zNOe$j_`)#zadWXC#QZo~ys;BBg2MI>W z<8RtkaBrd_D2DHa99|ZFU6h5brl``IXYu&kvo5?xr<9ei_P}08z2Qk?AhZ3e4OE|$ z0kLBdpl&z`|KdBa4ND$_;fOVk_A-IZa<{SQpf&$K83hqXOF(&=InArT$;N;tm_#d+ zpFz1u@w^4NmxVFArl>J_n_OVY3Q4THzXpOsj7eB-720CSkhVBi1fr(`x8M-MIYhup zu!1G+Coyf!g9A5DV2l6$(nH^DQHJ;oIOQsbufDv)_C9MTE$LHWjZhNmkLFYB(zPU7 z=M8CUU&~H-9}j<4N5Fwk$El7pNxc3Bg zw8P7`ct+gmHH_3t(MX;Xus+`lS{K*@IyC^rqi4hPsb=)gma%-sWeyXMZbDgoI=J0$ z0jUjfh9`3ep{LLg`c5t)Mbkrg&v!7K-z>*C7>%QoRwdBs2^)w{=w4`C|9~vWsD)_* z#h_AE3~s(j=#%XM@ZZsa$D6KZ41Xs>92nvF+cmH&S`H@dtwzdLDm))&9cWc%GGfZj zcvGJ|NY-D3x&Uu_#Mlel%zF-trC#IS{f(se(hP{@B(beLKLb1U7KW@v@TO)SnYgu( zSk$cnC2L~{(|QHn4u{ZY1q)>M%!J9Cf|y*VSmbDR2|Ky`1nG5J0=a`z@Iv=jtcsQ| zv2S+)=bOiwYOTqz#j%-0SF%_tfuB=N`^d&#*orF;?LhmA<-n@H9G^Bn00y=eV50Du z_~B>xPhtXWQ$4?qJ4rpAqPkNI+wkACy~Nfd$(n8GZL`ShT+p z_j1cHw8Rmsv8iOvA8F`a^$~TwY=efHQ$%h_DLZaxGDsHBVb&W+GqbfTATv~s=P>*r zK}Wyhrv1YU3asAjgf(&pc5Bw+5 zWfGnKO_JW(IYez1G*S19Td>`+v2Zs0H9b39n^;DrVnM1R7NTFGXx`&nc*mHV^1~E; z%D9FXUJk_C7A3UpZyMb`li~hLo6Eg+%%&s1RjBOb81(P{LvV|DkL2x|?8bc2pk2lv z(7N(m*nvQJ+=>iUSc9qj;KIkAEqB2*k=_V0;(W*% zq(U-9o4$W!(~3@VNkwAZ+)sPav6oBOxP6Wg5<3g7ZRun+O2nwL_)2Q^ilKWq5&BM9 znPuBAQ;)Sn)LFKXlHv+7c>5kzN_FJUcCUrK8p?E5*mvA~at&Ttrv&cHYT2GoXW>tl zI?QrQ0rAmy@VXU$1cf};?cBy@(YJyzsEQ3G!!y0v)T=-%Y^M^HdNUdsdx*RU(TA+O zJSq$6#Q1q3{Z!)3{j!zd-p{QiUsv_Aqb?Q0I4uSoy$1p4U(l{Ey5zWls*xRAgF8`b1e>t#o8f8%?d(pqeYc)9spD z>8=BFX-oDgb~3~f+r#Dpy(Q~u$Q~^^dt(SDa|6KQYc)MIHb?keI-E8uCDU8~_-|Dj zN%VuN1XtU&Y2hFhwN9NvmP?r8HU%M`fFrq`{}tm`6Hap%o=xE{)izVdfX%e!q7>eB z2`y^3g4(f(i(fEC&8?$&VML*{5jlTqX+j&-V5&ZUxT>i4&?W(G?e{QC*UrF}QUB3t2cMyor!NKGJIC6$eaH}Q`<-Dw({2`s0xqH539a^nzPs&*_8b;^ zY(H<8X;%f}yAR?Ye>H>=e&)=L4j+6b?Je2TnoLzpexl0N#W+{5h`D@p2GMoag{HX( zgcHW#)~;pjOrBX%w`dOxn@8HW-+BRS#KoEEqxZ36^K0;Rg(7U2UyVI8W-*>kPQ>)p zPgMQJnGG>~ikF99Msj|HOq(zn+I{~ZcD(@@9J3)g<1XT!Z$1$9t{fHE#0yUGef_~; zOvkX%g4Ri5OztfekXu%cyha~@_a9W)OF2z~_t#(IH%nS@(7krSiog;)!R~{=dwru| z?3>MKPQMZ~9I*p090$s~G9mQjM^TXM31}_9hf`B7;YWJ{F!LxHh3nP{(&w)y=XajN zaTWK#dgCA1qNf2te;x~#=$o?Bwh!W<)y+6}YZ;sG1JJIWkF4a* zfi{{AS>@@1BSxn9`TLi+J1dGU+}Q~`T&nS2wNM!dD{YNGapVs2=Cp zXlwStBQ-zRY@|ayL-?GlP=Q(Ds={1OvLG`?`f-a;2cObzMCDdC@VQ?O@+X@ivC=#s z!Iikw!UV}Z?W@a`T!##`_3X9ZU4?mBO7<41Pedc~zj?Q!4xFfA1#$D&fzqWz_9@CM zap7ngP_r?@ORyPs{rwIK$x|?BGR5Qm$N^vd0+*>xh_!!(FYz6J<6qJ+Z=xo>yJ;@= zJDXtd5gU)k-!20yuO1v&XUi0e8^DB7%k6Qw6;o$61Kyo!6Anem!$5YbU`5?xk!*$s zwYn5b3iIm)+z=x=w0@IN5T0s3k)|^en-P<7s{)&cs0kZA_S6+gxCk#?tHN&=8Zi5f z7Qu}!afka;OPRU5Y)EiRC@7p*iSuKBvws4-VWRdtaw_);wpG{yma`qP#1bj&=6akS z$ufgSr*efAlJ`aLmoKxQ!_BoDqj8=|JiM53bueN+{Mdn~a%1qBPouD;mjRC0ttsj@ zKWm@UE6&8+ZWDAKUPmUrux1PgRe46>Zjs&g@o4U6Jrb}t6b**SF#0c}U`=c>YkonS z?yFmmts^Dz9>;ae`b`#CrL_o2ziMWa>Wxvtk>l7jJO$a`btY49SqSd`#qe6Vp3yM= zfm;h-qwW|FTsiqGm@B2?%LNHI$^VNe{H1H%-E|A_%}1*sbe&Lef8kP8-g8AXvds^@ z=~*t=yRU`BCf$XNU(O1~={n=~z&4z@)QeVZl|zf3KS9oYM*MwC0x#!zAY$1uIL7xT zwB)H%#gGzG`*#kE+b2nVYbL_?S3hv_Mp-s|yg#aXK8p%#jtd`{wFuu2Y$0V%U$N;X z0khc4jY-u%g-eb%@ve+3;Dc5RX8H;Fj?fnNu33)A$oV&s33a97_HFp)W`aXy{s<$s zr-96|O7iZ$4E|O<84~onNVTc}6!JXqnY@>5@H+#+DqmT0RD3f&mM%@E;uIL&b%->6 zSW9*(7eM2&L4lzE0n!fX#t+wK66M`H@r`NiFmit^bM(15V=(B=ytIgBe!LXmZGHwM zd+R%_7dMtQKItWjFnEl^4A#T(BXjLx*aWrz?k8XP4xPlcU@U!QKJ{H$iFZ1rkobqI znbZt_xZrMuubW1(H9DPoUIeMZaqi2$n-c+_f1Z^sJcl zU%AZg+utBcLJ_{UmE#T_J7(eFH>_KS@xmLwajB0#^NE?k7=1CId<_JKYWRDOzY6z$ zw>0VfcZ0;w{mnB4Cy@<7Vfb_d2OBRKL8t!%^7ig=Y`yg_ONLTVjf5j4#!o_lam9i~ zid$*OVQ>2MlPd3CoC;GPJtN`QyYY{5ekZ6a4&uM81Gm*KT!h?iZd%9@?EN~apG5+bqju<=_63+V^)g<$>Y~th$0%B) ztreUR{l;4#^4ZIigN#~KfQVGvqv$D80y%_PxlD1Ye2vdVOEa+9=hOVXZWMb*S%yj9 zyq0F=xxoo>58^U*D+-)E0V4RhTi`M`#-pDy66X)#sLMx~{ZEB3(4~qc?b#R#B>}S>;caoC(z7bQ)ChqO5gsu$^Xv-$c#j1s>Il$)P*PQ zoyF|QfPW%UZ0?2$QbPKw?hq3)-IH0Ao&*VQ4fL|a6v3w_cX8KH9l(_bct?;6D6hYa z#n)7kz$4mB*?Lv(c;itTZaxJMx%Kf*lPtQ|>jOQ%SB@OJmWcnW`vKc+67Z4YU!scD z_2~45SjxTirk^jgQmd2F_G^Dk5FRYi;`GT``-;mS+1A>5blbdBthZ#EZU5G#-0Tcx zbY#jAGS%=OTP>D=T(d@#<6?{1=^a*Ng~Q=GdAY-^_M9)m?94zx!>Y;T@1=O#2TdF7 zSL~D|FEqE}`S610w%_u8!`qg7mWpx)WxOIWJ-<>Kr78Gg!cg9jO{emvJQlCU_J}zV- zfE7HJ)nzXxI|$0BNMXIi18mj)chtVImKOeZh=>_B;C-793(kJF#GkPogk%}B{u-D^5ug0 zYa|5k0vp+Xi?g`q8h!4Aflhtjn;hi3{S=*P#nFFavuXJ?%za$B1Zxd7(irW7+$_b> z_5VgMN3NEET-MEn@a>T`Srw5)Crw+a zW8VQz`TY3$oM(i%2dLn3nPm3q=X}8x4@I&vy#TEk(h)fSjjuMJzd*1znP>E$zb#Tf z?t@R;%L&|+X0rAkH%L?BdiKkT&-DD2I`Vf{J?W`Tp)pg^Xi>5YV$7V_pb!@}v0jX5 zs%8lt%#*Qph&{4UpNa)h;hfI7^_;??aQfSzihh6bNwlX`OnBen7Gg9_g!=i(qFa8d z^xC~R^fE04e+)CjgZHn{!0N>`;3Ch1OKY@y$9Lg6cOa6!dm`E!@Ovacso%07B8|%ZxS5GBjJvQ`Y@*V1lW2uVaC~mn~B@BGmVTSCjlNtEk1HbsSntjv*m~!Mz$(pHATKxt>7HY-;b4ZnvR?*i zw%^T^)d@)DoKp5?Yk*zO-|vLG?u7H?UF<&ZOQ#Q|%5i(rQPJeT)8XU0T(rLK1H7ED z5`B9!iJZ0V5b2FtNF}S+kzTC}{BH6}(*6G){L4yqRn!}S_N1xo<$be+N}Cw0z%y(r zzji`Z?-FkB)k@MP5x{;mN+%`LF0j)XU3NlXC#W3a_hc*=pvTpr#Lvi*kxfo#RP&4K z))_IZNMO!P);(iatGY#W&Urbz$6`C$_BaSuO??3&hCQguA(CBoIGzc4(aXw>dJBY7>BMqOqnx5$qL4AcYdOAr%bj3Cj z=c?pD(SQ*rQ7s0Y#m;zx`9?IUU6nZJEo7!8N;8=!?jYw@7m??}#b`tCdzSmiBf~s& zN&Jq#WRFszV69oIVE+qkfkD?7YBKnXXg_!?TEs}Oxp)u;Y7Qmu$9U|MC94gtq0I7$ykuMt;aIMcLvInhO1na~~ zuaOMYt2 z;`X?$V20Jde}3?!^%)~b~Pv-dmtS-DE|pI9f6>-K>uq1sfL_dTq<&gbC+$9M+L4eE9I zHl5iTPQ5Q*rw;pHLTHr)QI~EaZI^e__S5v$w(y2)o>d4Z@46CvZp1yJ9h0xQq>LWZ9rJsau{eUGof=5G z{3L~EQS#16VVWNfJ#~#Xof2mjr>HPVy*U6A|AT++0>*w$F4SL;#Rfs`BqUs!L>Jq@ zcXeJsn#}t%qu$|5&n)3&S|1U9e~!RUo1tw>Kyuz`=w!b_&n$gtDfcBIK^IBXH5XE| zq7!S^Oh?bWZ^M+G{&?l6qo~SsB{s;-hmZW;?fx9zbvim5RGZ4!1>e7e!r=jM_YMce z*2~0u;4zB6F@_GjaHPBVuA7H*>ZKopsIAx0!R!W- zlBkF}e5TT!M(2>8vmR`YPRAE#Y!*dZ*pY@rA$C001IIje!AFmm==+x${BOIKy6?M< zvkr`c^m4+^Ul58fZR>$WGFpuEU>tiR;}^+WrvyjqC2+8I3?5!GiPim*DH!XWgU7w7 zM7E1<1+y>Sg81h#(A{Q;d)xHrrH2v3r@jFnqA{Y%O;3eW19?~Jzh6Xbzp7=3e!*<=K1P?$y)wv}e323TEO`X8zg^~;Z+Fq2)6ryq zIm7OH8VjkhZTvGm2CE*fU=8?uNwfC?+?gN!Xj-IQY`PdDP0S8w6tb-fg9~b&wEy5$Q69pb) z|Bz{t9WZs>4szl}5G?h)02kWLna%bX=djW^jrTFlycJ0rtb?)5g#dCtY7e~0Fn|?D z66|{yE5Y^4_t{B&x6i|1JlACY1~-%)ApR=bNO~Lc zw>wO|$Fo3v#NgWGL7a2BZUm#zNX}A9gmMVeT$$#LE7Q1+Q9Pk~c}O1lnQAsClkAIC5Qx z^=D~z9i|&gwt-KpAIv@!MK>k#^N6=uT*MSFJj4Agta6+W&dut~+7H2k+gjStci)aq zGj(E@etXYm2lU}bJKR{;l2|nUUkn^4@!%}>oNV|$D7fOp2<>vtVc#$Wd%PzhNt04~ z%rl6rh#o*UT;(8pVJc3VtVI7awdGdGWs&r*uOQ!0jMe{g_@&b{lEM2|M-&ur?Qsfz z_YT3F4d2MLTXA5xCJH2dkHFA@&v^fsxtwX!3UYPlV@RJpij>K=vezWHk$;M9s8?T| zc_}jjKciMq%c^YRlwu50V|&PCuMI5h97fwTL!mi+9%Ke|3y-uAGS`10DZlhpq-RyZ zn(+>ST{@@P@1aX@!~o%sKZ>Drl@6XePa7{Napb%g$D`4A+u-TLBC^5L zn#^Q>v(iBi==G3K{7yKZo&A^1zVr{Fmur?{`OEoaWHrziAvXl(7bnB`hJEzS!e&;` zeUf^=dXFAGdWDzut%TV-j-lmeP9x3o9&D#n1sZ}is8ai*;Na>{~6{0Go407a|H=d$~ zSiLnux=2QfPRnW~jeM`MI%Y4oszU;n?g*t#R`aV3)jG^FUq7ado;bOR!!RD_mJheW+bfa9zi^F(zo{!gdXz4Vo&(J z)^e>~gW_E(@lWlN-+r_$Nwd0;^%4niQ20ENrghLuucV5dCS{YMwwG=wb z$b$U)uFi}}JVEY43(WPh0@-$XvPZWF7XK)LXVWH%H0~^c>i$w3=5~?{g<6oJr`q^K z%{yB5{ScR}5zij>cfyL(i!m{ifIRiPZ2z7tko9I@y82@Hc5e>7h61tVc?y4SdcpMU zlh9sXjuxFff}^da@W~`~=wA;Vg z{~|6j_ZIAABt5E_T?>Hw{PYo#oN}53mNoJWF-1Bqd>i@rd;{6NR12R87lRol-RSIA zCssvSj;uG9#Yg5AvbAQzc$29iZg4h%BI&DGW>y(~eo`0BS2iUr`3Yp`Qw4c=)QI?i z4RtsvBFlRC{euP5!T0KMj8D%2i^40IS(HW=75Cv`zgFV$@(qdN`Qh?Y=aLf1_gKEv zkC9lLiraZFirw-JG_GkAmC@aggQv`OG}w^65lvZZN6pgsvu{6U;xm81kz?|p)lA5Nfw%Z^-F>po z;yiVK>I5@mF2SiQYvIJ=E#G~XbtF} zZGhYtOcVITZV}DzKTqba-4Dlp@Idj`#jvBNl4^4^@b_&E=(c?$*zB|iliMHJTW2Q{ zPUaAwqe(I9``f5Y(rEf#Y9$_UUkFNO?O=`uu)A_JJK8FN2A$;C)q66T$F?6pReKBe zJjEfIrZaG;YYp_6sN&7F*3`IJ8EdarBXL<`@H4I$JKvDT|H5iVhtFEX2V2ZO+n{MgcEjkk^O7RP{qUJ*jV;8G@jG| z+g?c~@b4FX7w85Yi5kT|xwnaJckMxrm7YvVvo!N(bU71bu!$NJPKHSC94k|42?x)3 zVd?&J@XLjN)&(6zmJ?!Wd;3h#y!ZWUz$E;%_ZIeaFrn4q^>9gk9FuV>f|=J5 zipsiFne+p>SZBE#nX#f9k2zg}HS5&4KMykLk~dGWgFJ_=+MPjA+ktb!qhXS`EaV-V zN3|q-;cnm{ESqtHIWwUi7i_m6(Y~+HQPVOu!cB-29-J0b<09rD!pxp+CQNhJRV@2e z2ZkQ*goZud#9+jYCMm1o@=cXc5VsrRy%XVM{viGQS&NRp5(__hx1!qdB9gsxE;0H= z=z?#v;F-%oaFKk%iobB6&z^M9L&n>Y#Z7m5dHfh!Q1*mQ`h10s-<3!|7&lUJj}RiI zI}sTUjL=6&j6B$-hYNMZd5?2BJL?=dAIQ0rc%7@n-m>#xkKWfjHL%}^nk{ZTGBhM79Ki1WG}6;*82kA747^r9TsH&chx8rGMyHPPOtVcwy3aTquiZBhO1pC5xpO*h^*BdrH!6bE)t5jt;%z_vdrc>213qeD{qQm+K1L zkH_Zl!$yLi$;@Rd)?`6sjT}u{I1RF&@!VnOi+FIgGdA_Jh1J>7G%@8dt%|I}8(kCe zx*PuZ-%U#{(%?Tx**Txg5a;i&H)r6q9E9gB+JF__JY-jGszsx|xG_qfEWzavYd=YV)sH+Br5S@MAbn}$T%Z&5&l7VV}sO?SX|$}l_p+8X_soC7NYB+pdVNLUJ(wOsIyVFN62!8AUJ9? z6Z>xYLGgQ6TKaD+-6*KV%V$PlGuIXLB#osFXY#SmgF2oimBlu6)KUMo>7=ypJiXt& zhu$@({ElrGd%wny%k2!NtF|kV@-lT0xirvQK~LF~wfmU9KT#ktv=Sno4FP zmmi{4UCMPSOYg(}AZ>Ex@ffx*X%SXXQ4`6#6|sihja>P&CeA5yC)6)5XP1ht!hsWX zk@N5g+_L8nQLt4fpEd+SS>PI$bKJqaQvHVIjPFxjnLfevITs+->nrK}{e!sGcyT8d z6wuB|gLGj_3v8P-pW1&tiPV;DVV>tl|4O?^ZaC zwz8wIy|ZzO#!fP+dm;Ulc$;3+n$KFEH)U#N%VCL~0^ZkK4Z_9u$ftY^XJ7j=LxzKx zo!^Y}(`S%1)tB&O+$DN;ei!}Mqf0i-JtJJM!Dq#{-oqo!V<2IW3SXYSgs*4Spytt8 z_~ai+5(|1@m69COEkBFNTdoE+-LJ6R)F#2%Mi%D0T?_XVB^cY=tC>E%yC{igR+x=` zfhUSfGX`NLL}LF2ws_4hn0&pI?q_Ae?&Juz^BM)l3$)nU=FfPX;V;3)#q-(DR~cxL z!wDSfQbe5p?&p6htH{|?V_m$p zlwKE1{gR5C8g8?}t_lnZ?1Hn)t%+A*4NVMi;cl;<$knYsjU7J-U~uaard?AO{=T1$ zjd#dH;Q1;xV98D#Q`;(9(PD~`aUip>`zq|vuflAc7HAFdjD)?$colz#{E}0LyIV6* z*0qUD>-YrZ5@kii{w9;c$*0+;Q#rW7vtTEUIAVk9JzNJ?BDP;5NY7m#W|^1+Ewwty zUNgKyid{!>O|Rv_@b4FFb}A8zC&)pNAAuuoC`xJ20PRG3wK?cllH96oodpzRCi(1OE_ zSe=`XU{xSQuAB&-qB&g1MO#+lz(OML@)UYEbiuuwi|9Q&btp_83)%a9=v&1?eDB99 zR_lx+Gk>oIKC&tnYVQ$vB6|;CNOOU0vVADX4^f#PdzhcM4VnK@bRLday>A#NJ6p)e zEMz1h?|GhsG*otIh)RQ|wE0$|5F$cEN(xEJOy2X{M?^A83n?wK5}Ha%^gF+Q;eF4z z&;4B2=QDXQn^3EBkkm;&Z(g+q+Bd@DHFo*T$Bcv6(tj!ZVB_$I*~7fp5p|rTPt)^% zN#G|k9(XEpjiX`HjpJI`t%{>&v`W3&_jo7AerJtCe5;$cTwI<|ISz>bM*(hH?WEn4(H-{`8`~I zu?KBmWrti>{l#(f&%#9w3+%932lLLGGPS#VaO0>Uz4oOWnYK3#5529%KerB8ef;!~ zYBBc1&37)NeNGfoe|4YV@WYryoL8rFy~1#f)GCev+>JGlnS|SHs-86=d_3p*iE|bW8dI*|xcO{;^ z^9R1dv%v>n9HBcS0_i77rx;>?ja;xfiC*)@@cPFgAn8Ai*}O%V&NfKInxDnl;n|yL zi|+||NuxHbm%fZkV-<1auOw7g(?Z!^Gvj+ZdXWQ0F~mji5sq7DMVIR(!SxRz2<=CF z%Zz4naH|tKacUtGtzd(PGiY4fpn?O;jVbG)AjGUWQ%=`zgxzI>xcruCyeecEi^Yd-lr-<#tsM^clYO{kp@ULw&47VOE|4>-|g7OgEd zM*X)i4@v?bfLz!R#eXM(mH(b1QqJmReWoqH^2;Ukp`?SX9gIN3QUcWUY4&jZbU8jL z-cH53?I0VjAHhLGuc?WnChVCY&e@xjin;GXe5~OXqOShL;;QY`=OY(Uov{QR7@LXe z+ScHojvYAFB@yi}IgUO>%EJ6qafJM*Q$=Ad#AjkQFG@!SPy4%o-rqb8_ljB3PF5fA z3vME7GU>{x%l7x<#<(+23^+n5KAu)!}n*+ zV7H75qPnY7@UkfsR66I8gR8XA2YVmNq9_`bFxAk0HJzvtX_D-iNWB(&%A+Nkap}65 zV9+K`BpfBtqg|&d=k7i(ORNV)W^lU6}gOH#1x6#&KGW4~#HZ=Au#MYj&#B13NET(Km$zET} zngm3$0Slg>;nf+u>A{nf$r(d(LFfRo`n?;!*%uAZmn>n_$2*YG_myaNHG|SG`r?L| zDb&N;b!c~>Agnn2nTgBqgxm#bV6Hs_eaX0m7>?g~%`F23ajrEs=RaKet_B@(KOhkzXpB_ zd!W3%a!_-13fg?w5NBp-(I>u3)1q(k@uf|ESpGy9q`s~t8^7H^>QARK%bEw!ex*+I zI7SOphK=zTsR!7j#f`0xs^qRUC495@GR5t?>HHo(?s&^Rf1Zk=MH5nt@84M1y>c~O zc*BYA@w1@iC3d2~xRWSU*PuedSA{y%zmz-{q1K@?-0+x zHz;Ux7{0ia`%nI=(83c3ftfJJmrvX!wR#A@S!=|uyfz66{nBWQaxXI9xE#w)*$AlT zG^knK<}IBgh#LOR$GTZZF_N7}KQ>dLl}Z=U2S#(PI%+I<6JJuW?vgvC_V^Os$vL|j z-Mk^{47!h2|FL4fONWz7fOXW~{vK>t70q>Jtl@}xI}!HgL$z@&9@uR{fA>PPpZWkL z?jwq(ng}qCT7KxI?R%bgRuNv5_7yJP>j%5KG-Ff4jR|ziqfN}M^aQAsy9|?KiOhvX{%E!n z!N}GK+a5cHN@`N!?@3j-`o|N`_6j0-Y2{4a%-#5O$~U6Nr{QJUcf2shj{Zk!(-Fp& z1Z|4rJ>*xAEW0uEeZeuVXRsBelx?64?9KT#SwmKmYWDD3?mnu!pveDoa4A+J^HE|! z4^upoJLC9wig|gxp7UV1&}FUyIFVV49}J$yhq&3=rq+W<{^b0pR-aM(wiG;5y^NQN z)}wGc325$BM0RO;DE*}+N{E?_75C&}@1u9I*_IFVk=YOEfve7Bx=Io;bXtuS?%Gnj zB%;xIM_aP@?>Pn)v_Z{WeH>e&Pp^~FW6S4i;Kz#Iq+rWKY<=5__G|c#3-|Wm1O;DU zGz!u3g-)bsr5t#^3SrFjG*Ga~CZ5X+dDAPsaUxQ}Nzr3y z-t;5T9QBV9sGN^|=1-$%*j+_e4jYisnmf3CPaDpu{!4~V^y2*^UHILWH2OcY4OY+B z%F`ZxS-3j<5zl>vFI+j?!nxNrfxCJs{P0{yiVo|dXBAq^@;6fC*(P^ll^R1fmL8(| z3#_QywNLS?i39Yzm%0^8_SD}#^|Ti)JXFv5R5Wo;s019Sz5qcrbLjqgrTFTrIx^TN z0?PMnA<1ktZ6j9=I)Ov@*t;9hQ4t3ZW{TZjI3tiea1FsixxQ#u(Q4}Eu{fTvW}KC8 z?LCsHtW1nfCo^Tru5@N&2~h99@=vuy;u5oeux3R!Jghds?%X|hqQn>18*yj8u^Q~B zH~Osa3KLpREDwX=P2#%59Hb^blgso=a(__@72^Gr+8%1p_=MOJL4hA=%91(MPLVLY zWZ*p`$<5!KNP|2hy3DtPwY=(#6ndl0UFx}eKgV+s!44(TIN_)!iS*W>y}P2X zrl&#pQRIZxGtnCKJM=VtNMb=nTcSJ^yGuac+_&R|0k!yH-`!(d4foXq{PP~miQ!#pa3dgp(#`Mc{ z@(Za5?)>)z>G1!DJ6;@zDS8sw;h_kFu6KO^e`0u^YE#(Q+yzVA*GPD>iA4p=T5fQ9%fpaY@ zPQe{BHlyVc$53&+BsOo#B+tXjs7Ks6acrXj?egLtelZw9ZhyRh#;-^7cW4_D_T^Is zyXNz>xZcWkA7hl}b`80js$t7VXP7sIAF1y*lF^@(Lj;^S2F^TDe8TJup6`1TpLMjv zx{_}Z|6nCwHgGojf1kE$%OH9vokZ>!TVm5OUo>~-W2Dz|k({wDVw#O(h(<{wd6=;Y zo&PRI{R&y(mqYN6Ly%~KBI(1{#Y`$qa{R5I1eS3lr!VD6=cqv4a_^PTQi!m z72Ps^NDXmrR*`M@NUkx*0g}AS2+38U;})Cng5GTMu0)?{IJA;yuwxiq9&AR9P9>C= z1LwCqFq2m1-g6fM2xC6{p(6Xkb>{WuC}fOZqwN~s@C)akXvL`mc$fA9Y&vTe#x^|U zwf<-!=@rYaXZk=s>(8V-Xg$b-}gc$rNP>*xDmv ze!p)r3F+k=-HUZe=i%c>kK@F&f2gw>>(HiJ8du>;Qzf(k_!VBg)6fq07<5oZ1uy*; zjvgP}h+p{GVW~Y^@D0C(_)j7CyS8*AarZ{loSBKQ&Z)sJ`Hgt2YaCrNE1{IG&YjQm zyT~|hH^mK{<7=C*1U`9N1h2|G$@L`~$=Jtvc@4Y`#9wGz6US}goa(h+AMou=BhKY6i&t-~L5pm1@sVfKaP;Zh+{@G%J%XQT z(FQ^E`(goJUEYT+>1h55pL^7=&)+EHgM`xfT258|a7FXD6AzhxHHZv{;Uwb#B=O}O z8vkzvrp~0JsGGacG|@HW7w4k;B$!DVGGnOV;&*EKS7GY<@fzld@;)T{;tH8l;7VFW z!jM3EC2DN9q_@qzVO1S>27NAGju@R7@@AF|f7HsJ)N2(nrn%P0Rl1kDkW+$!J8MzK zO#$rO6pnnvBKfxib|We!1)W)2%ZSTM;jF_-xaN5@#}|!2Gv2#XAJ&=j%f8F==Pc;Q z@wwf&qGmmDdZoqGpInDtw~KN8=x8cZq?GE4dQXY>@4(%5VOXd+2{&_|?LGG6Xy<*C z`RBNqIl8EudizBR`KvXc*L2d_xsvQ}Ae2;R2l}h&L|k6J!yTjP zSazg?eB87Oz0Ij%glB!{m>CMxXwO;V=KGcBcG(>pEKMhR2{w%NwjN|xn@1d7N+>;! zqmdTh%~Tb|@h0aEqa)WGDYK^O)W2n%@AIStdUN#&Kc({y_Q{IDlk2VN)jM{g^lDKU zndl}woo%4=!5vFI8G$UG1RN6OTq!-VMEI{A|7&t0B_m=8Ta#X)#dDhY3g-REqJ3FY%ccMf%&?XI3sZdl?m>0)D4X z9&Rnrg!UJD5K&`~Z`*9bzn-qcl^fh(pZ5du!uc$@H0>b1mJ|)8T_O0}NN%+la$Q#OzF~lAk zF72mQL}0S&z+AlP!hU?~tS7!0@{?T3ISvC$HP}Vgm%*U&0BQa!NL~x%le@iq>dls0 zAgH_;4jsDzYg^0k?bf?Q?vM=avAUf(JWGVxp5D(>-oKbQ9s5iK`^D*=;oZcB6pp6VSTvq6Zhr~4l#}5?z2VIQ<)BOrcw+aE>VNr z5*0k;F`KR%b;kB{!ojfn5i>3QDX-%0RJ^M=m>dy50phJ8kQgzI^|q5{RU{iox?LQk z-><_54rmiiO+UEv=ogN$Hf4H*rm$I0e-ML`5#%o>f|!Ko{2PZFn4h1Mn2?BCs;e)7 zsaPG&j$QU-&F7wAuk}457gaZ+l3*X+<$XH%j+8;Ik(ZD`}bW$a86Wxw5(g&b^*RKkSsDcVDi<9UT{O*}*P6Urw2By&*d+r-l)AumQ#EC zE){g)$Lrj7GF z#fD+WZ5bX}H7IB7 zFgKjYH(eu9sm)+Hpo&e{n~aGk+}CKHj1VvMcPc+IWQdYls)5QSv$0-pJl0>lo4M;`0Iyc)LRn)J z`nr*if<(CMJ_od2LG9P0F2H=Vl$Tr5;J2P z>^kK~4p^r!n~x^&-4DvpcJ)s187!-C6r*j59+{I?$U4=3RT zN;7ywSBnH4&O>X$>d?cD2F#Cu1?ZAu1XAQYk5tV=Ca!T6d+5RC(u(ArNv)aVB< zRQ+Lu>31@R&Mj%s!ha8~hF)N0B27a7*(2*pGZOG~75Z~bhXh`nhxh7EM>QS+Fy~Ds zX7Vz?HgF@JX03`E`G$IxeAX@s=kn3pP1;P5CL^7qC zycJ6(27Z{z3SYo{R(efR`aWROqk$xJoj(1#cNLl}oQ~Y4n}ds*2vXbenxFjMtYUWF zcBHpJnd+&F#S8T6k)^<6>bjUNF@Gk(PdR#oY5o>VjyDbR#6|O|6@tdp$6#aVe{TVh zumF6OhH*+|Ek7!3fSJYh(4>l;$luCcRt-bhV8AAjrE_|adXO^m-LnL(Ba)2KYMctrS`*y z(wMt~b^lV&6gZ}kIUlaz_O>fH_=z?xcKSHQ@KzDwwRWW2zZ

z8G_EM2pZtaBim^5-zIqi6scJH{9GjtfW(Fv& z4QGB$NkC`j)bR$^ej(2!*T9rxFL>L=)o|wcQH4l*dkw5%^z44g%jWSKc^`M@z7rr`ppt&w7RN{LHDn8(X)i%n3 z|MSp)C`{%O&v1Vxy*TtGeSMcTcMXdn3o$uLKtYC>mllu&UOYZIC<|pna=7xV3J!T_ z&pdZcqQZPt$v);R3ENYIjb~I~+xGS3*5eFxRqG_F{V0t!DmPQ3Hv(zL!Xy;9poX|O zzeTrp^GR0uc4Rwy2^kx%!Sm;q!Qw$RyeCnVwm2R^hj8Fq`N!Ak)BAU!;By1StMUfw z*%ME_uQx;)0>0#ag#xb4lmN-cMnp`WW0~wSqunDF=~VIo{hQ^3t{hRP=A2i8UB9$( z_u4tiPn;^Zlm6TB48Ky7d2K*kb06&Y3Wb8&c z1n(ov;dU(sPrFb3in~pVY`RG&S2yE?zZ>!C+8##0Cj%R(A7Ugswlc0)O0cA97;fj7 z5$n4$5mt<+-X4h}@k4vjnOh3TX&@a>ZK`MXB_LorRM58Ga(=vh6JMu%oz>c3>&dmN zk5RzfgLtdgMsi188BZQXo)>Mr%REu0@GRe~PoL@+u3J>lKcOQJibPWWd<(T zc$=s9%#g`;S$H;mtVl)*rRZ}n9FW&KdDh2R z7LRF3p}v7vjIFgU@|KZ=`iUh}$4y(dM(7kjVox5W``!e1-bmoMsWw|zYhX``OTA|0T%4S$F(rviosG5>E4+TN>*w`3_&E3h2==2#V))C@dVzXCstI>|_Vy-|^S zJ`bN$%b}EoGPvuxgSuY73GcI!qJN}U;%bcmX5P?29Y|Lx8Bqwu``cQo~lt$z1|!W zA1pyn(naYg&g&`p#E$8=aKmN^8g#wdGU}6XIf^Je3*FyE==E|gw9?LB9N*y#I$rh> zWwx$@i578GTBnGemT{aXst4tKNGA90Kh*E?%k-Ze9oTVLl3jE}hH?lUK&~?az-TPMy4+E&`ygRYP=QM)1D(b7|#2QdE!FIbQ2{3dk1bF^h*c zklPb$fQiUMQ_?+oBYXt-NNdFr?vW&yQ;Fqiw+h}SLzFN(74ORk=bfJ|iay8pp{FZU zu%)3Ko@+h_CD|q6f%-VK^6L~-KX)~zcZOgi;W&IWPMs)hX(n?XtAT2J3yJvG$HRX< zOG94_D=}*zK=b4l1wM@ zfdO?4n`2{0S3@osCx6Fbt%cM*-VJ{41{b(l^NreHbrJ%9T!d`32IfwF6mL`f8K!l- zk{Nz$#!pz#0HoBG5?SyV1>1XI6?r$}hXCz0l%Uh&#L%;lAGfD&NyqzQOPQY6`%w1n zT~O87##|DfL|^}t!8wbhabp&tz8ATW^i*Nkbzucn9A3mc>51nB?p@3E&{IkO^&Hd} zd;`x|6^ReZ%kfWb_oS+3>HP;_Ld%R@zp#buo&Sva z$pdH|KLnqTM8e~Pv%oWMfOMCCLV-GW(a=s4)?n}$NdGs5KWi)-U3j$ye_DK&Q3+az zk9C~Dw(2*L#uOun>1=_G*TY~f?-0`1aE~brD5r8S#N#D{rdBugmvi5cq4?LdRY)D# zpRibR0$CKQ3o}K%h7{geG+r{INrR< z6Mpl{VO@6xIlpv1oZl!6N#okA+IR_n3{Hd3iEz@!|H?U0%3!1`k@<4rHRO4$fS=-a z?D<$VcIUoP&Z%w(hj*)yyVraf#~MR4SzwH(Ta0pimDOay?s7_2+Y)W5u7mYDW1#$* zhOw;~Y^3aKcIOiXR(DVerF^`LHXW2f<8QX0sciu;V^tJfD;I>r6-n@Hb^*ubbY%*@ z&gZR;9pVLW+fZviA@KLMhJdHeFjf5nv%M-GSFss*QZoZz6lYQRM@0}eT>^XRCD?!4 z+ELwq?R-_6b4V~mgx#5cfy5;b^HMG2VfhgR--CWI3O`H8*#}+lIp#JcuM$R{M5K_o z+|{r@>>XnncoNoV?qKcqG1MOA8xVeV7vJZfBt&&{oFOd{IP^;jrY0M6c{peG7Vi}v z=s%Cnet87#|Fr>b+JL2_b@5ZXDDv;GICa3Zmmg9-4+TFZq|R6*#L|`9#Fy3zJ{rCpDG%{PQ#t zS)s`maBlirQjth{nDZh^oTZ{;dDPCuy^y-Nk*uV0;bFEW@9r`UP&u4PItPkSrsr+) zL*^umf7wdD>PnK!n@*E8jnj#v7V&23A}8PVFnK3~y|_h@ox$bb z?*s%hasQqZp{rrU)?WmcaD0Ed;3xn6c28tv5)92}rm;#lRj|8(In--Qv+)z+CLJ3vnsYWKWQb%hqART~X*wlmK?>`U9TDy3C=e@=SoL5BQztGU+c%h{mxDIJC^3 zS0kzoFG_jfYg!C1RY*=o#khQPuW@b4u%-re|D63Hk zH-yhZ!?=baXKn)(QUuS0wAmR}^I6?w9;?{l&yHI9fp<$PQ(GaxYNW3M=K4F1PF7DT zIjN&hecF)xxdH^*rjWjlry$YMKpirwMBXBH=uPWYWHBIv9JESMz4j)~7$k^1$Tm3Y zKEN1Fq%*%_qsh_qNGR2t!iFrBWiS1yh2Vl&Bx%+~X3?vQFr{HFD;E0&-Ku>CH>Uzz z*}n%#j?5r_4x%KR>lF9zZpTH#_n5H^XPoZRj`xiElPM}|sjja!c!Q)m>bia$^B!n` zLe?WNk!nJZkK1F}rYI&_qKd4LDS!==+u6-xCZs!WEq3bnpfj$WL!ODDw55DUvPRFNWM~0*SJi_` zzb$l1Ji@WBBI&O_h^KRYf@vDb1lydMQ1?EbC~^!P%IztPNaZmq2fx6Af7=0FGvck= zIm$c^oQ^xM=36<0x8CQlVW$#1pd|_gY3I@W`5tJ;6FZ!CIs%J|a^KV5J-j9x6M7EEQRedgEJ zTt>&D(sAf7DYA8q5DC`Khb}R3^83Ydc$D#)G%eXnz|{(>#d_h7X*ZQ~R*EuLIE?T2 z+~mc7*a_An7VuN)WO|Mg*8+8R zThdyTvGWSlbw{C&WG#}BBSimrB1X^S+)BIM1>j%q8zywv3~Aom0pfwF@bCT`RQ1!8 z{rsZ^zP8__Lb#ds-qmcp{fZ&8>cUE<;cyj|wo9M-rMHB6;V(!AKk36ig%j)|UInXh zu$GGMb|M8F2TCrwi~98NCl={>PhNhh!|(Gh;~8hdVfBDK&o+CIb6bmo$IC{%)uIpS zeZ7kJs$?;m94ks|>L_E}D~QJWir{6a4%`lnqYfmC)7LHzqTbF>epAkMW}&qY85&#) z?dR)Y`?4$Wd0;C3xxNl%ow!E!r`ZynqoL5qok8@?dxPWFFQaE3+epuEccaHc)Hz8y3h{k2++8q{5= z-j5FndQrqriT9u+pVZ)$@qf@PNk_(qV?p-WUuD{B2l+lX%E-YhOR>0$EP3vrM+WWo z;|<0S7|(JHAI3SKQ^N>#Pq~}#%+KOK9rVHrRvh7%H7vnbG^`O%x&)0VspCzh=Sf}? zO9iZ}p#q+WkpI%2^XpevkOOM>Nb)Lf_cW!MA5*&qerC?2{`a*1$6!H0yHO9HiM z&EZ^J^+bvMaCpJ6Z+|1xR)0!BLkJJqv?F6caOsL1;Rk7f&XECX$E9ah^8y~DRrbs9 z*2p|E0WXPil?i(8^p798qK<#-;Y%#4a036VHpFIYr$GG^MZ9azF=BiDG?xE$pAo5i z2{#>&!o~jvi2p=3I<;M%`E)IjS=RP|*}E>S;-TdZ^s?hFsxK3wXYY{3e`jV=U4Qe@ zWa&aI5Oj}>Z5qKlPmJRi|2&ADMgUgy+(8!CaRv0YQ@D-Vj#IykFu zD%HrGP7h;oqDvItlk*)t+u?)8zT`16iKmE#mk0+X)8NlmAD(aSI*mNg3rC+<){cyE%?EO# zMg>+R_fmC|eLO#(UwGXS4tz^yOtc`n^R7 z^>;|a%IT44?uq@Vm#yTF)QaQqVrdY3E{<<=zWHq7 zOse-4$8?o_jKz|pASkzjw{!DW6cOXj471npPA4<`MrR-CurkLBECYGR{nDwni7zCW zeS#bI9YYI>qIt=>cOkMwmi@-F=6aTw$nS~A%!PeAyzMy-WL>N!X&g(S%7>$H&3-97 zu|bzw`kd>eewD>azD2z2^9b44)rf@$XmYeel-vncMTdH}<6Z$B=G8e9q!Lz7-8=Jx z*|Oy@BfRGumJ**ud*&L_PYxF264way)zThXF1X;co0g!N+%A0AsR*9=pOaMZ!<$&> zydSpVdiKTW1sZH}faYmYSk)yOSN{Eoln;L-J921{+MY)xr5|P9MkY|b>u%uCy=zHl zQz`Xl?Pl7LDouB7JZkm$#2gY*cnV*&e8;Tj`6DwU8hdlzg50gq_@#deKB$y)w9NPvhW3a_; zwDYqb-0WJ4_b417Upb#!h=3p%#_Z)-as}9IBZB-#=Rp2R7E`y>3Tu8zfJG595G(PA z`Db|&o&K&)CPpG*MPnM=+jNA?6TL|+1AUmeY4^y?o^vGDPnD7H5~(pP8vLOse(J#IEM7yDX@7kp47^vkf<#SNch+S z=z4hq;(t5>Vd(&}m7jw>-fbgEwNXUtwFkogk|3h<3oPwa0xt^#m}C2sz}S7>(xf<4 z`S32$l{H31+snzg>It-^YcAWg&=mf135r&x|B`M=IX=w2WufEz5#ZhawKT8JNh+j1_pD5kWE|`$V|K=Z0}vT zd0;kH-&Re^%%s7rPnFym^@Xs_96zS84&3zp;8|%882yugz`l3ntVlOCuWd7&bL4V& z`Pt0OZe617s{<{`55cFugfD!7V_tWnEgM@cEB-`DZsrkT0q4NSJ3cq%jxK z9Q|2fsK^kzWGJD?m4NRAPkR~NB6 zmrL3C?I3LYHbKJJ3&i;Fepnc028Ho^A<*YE3O^tUv1PHWiNIR2bU2-FWW$GYzi{~K z_7_U}c_eCjJ5|&A25>lTnh1QKsEN2@j1%iXdh?wB9Ak30g$)k=_FcY@Mb?ZMm9 zlgwwdB$!CAJ}Py#7{`&0MaQe^sFCb=m^HW<0$#tTE+t!2GYa03PdjyBM7^I}U33bf zYu~^^Pd%h9I58!>jXI1h@4DjXSCX8-&+NCrygVyhkEXut7tZa=sl zdOn>3`rLE!{$4iuz}*X0;+-Vy#URWrpAWV3_+-Nu7ohuXNcKDgc*K(LPhAJ6V?IC*w*AkCsZqT|#h85h-@!uEUBWmiU$baDvCOv2=Dn2_8#%4R=qLv!; z%K*b&eAyom_<#=AcROy&NK-SQ`*AL_KVK4um_^~+ zxfkK5vjg~4*x?eV4RCqVjfh>1<*&SEiN^O@@xEnqnTc(ov`~`5F*+XthyTSu2DP;3Yjb>)*(wRSK{J~xf-&K8t_Cf)?-E|B^ zpCl09%LhpgZwXJ~&pgVj^cciE)`S3`>u75@BECL;XxPD@lFO?`~=Ahg;RWY_s)JK6)*-&KN*lU!c@ zpAa42aF*kK=|k3v1UP!P7qXew3egi;Fc~#SN>r+#J31fi?A1B$sW5J|S^-B3T8W?J zW%B2n2}!8g&!iP>Ccj+oBEKD}=y&)tV#4kJmv+W538w?_qHeB>`}7L&Dt^c>F>NL` zyFWntnW?PWCK=pd_nH_*h=Gf%0n^;Aiam0q_VJ0qzf*;X;dCEj zAkzgST8ik0R~55fX(dQp;=`qR_wbCMWytf~1Z6nB46jr(TAat@t9RpfbwNyNoGs*1g|IN(72>yrG5cdK^DE2waJcAiz&nF#@;S-_ z4S)WN%@kd6OXhcCBk#{!v`>~-()bbOiFBi*ld;T^=7&W5k0{t_za*_Umf}xyN*Qez zH9Yk42)SA43p)Ef(a{|vsP@ijyv3p%n_as}9scy3nf>x2nRq`&?uT7~miUi|m~SGZ zMoRQ9_XkwV?{~y#<1ut(u@}dW_>3ElUME-kzo5OhS^VwRb^LpM8n!H!ViP;l2&ekJ$_AOw(fKAm)VTp&e)3!6P&@mxS#y;ibpaHDe&&%5oXWAH;i4bD=6GD zBr8?>$!<>nkS}otb>j6fapWDdJhhO~?VF06_18k!;08E3nkaj=-N2?W(QNhJz$;T^1>M~o|50qU^9x_*X9T#K7lI!n3<7Uyyj1=Z+2vfp@gF}CpQhEMRIdsmWi}(7 z$j7k5e+$gp*M)!YV#(DGRdz5)j=wouiY=9x$@)%-LG%=1@Gh{z*D}pPPU;!;jpOdh ze7ui3q>b6vI|L!TJc>7sHjJSs z>V94@$5u_`I3r3n7xCx179#g6k>f=iK%SzOB)Kw#skJx;m$Q!(9lN)T!}j;EsX?Da zPo6;!9_7Ny>k{xl=_%=&aT+zuDuuu)4sbd7I()q?!&q4;!s5PsM);;ayXn0pJgsv^ zebIyDgX&)NqREBtxtx&ZL2vLM5rZIeJBD-K@SToElc7bSIK}2ZFReENF|(BL-PQBq z=A;=@Lgz8dRfX79BVp{%#3JwsH~=cGoPWc6f@pg^Kn;`yu9$@JAe@M}&7 z5(%Bf3Jhi5u4eoQbxjV0xOs7p8z$XfR8Yf8dGyk?6ow@4!z<%M@Y-uVy0B0M7u-=r z3Ea2t%0V-nA=b)t3co^g<{ZSi1^ciTTw>Q8y9@eHW{{%H`zS5=BxbhT;2oPoh@sZH(M|M>zVm&;%aG_b7UJ5S z@zwp$K<$1wR%5xd_>ir5<*72{z&TXj^rSP5&L>IyoiJiBafjGPoWf#0VQl%jFk(FY z54ChhCjKv^gZMkXMxGNgu;q~)rLcx&UQX`D1r6ft%vdF8T4X>>lEa}SKaE}Odls1C z({Le}hcyo*!ND|6%~qI$q2@0+e<%a@JmI5@m6pgOZWCH0@(Q;njS)kGyX3R}D=zya zM>lV_fZ{(hnBK)P&{gFLUZ47i^Fhu-R5?Pm`ubc*`N?B3Ip?Yv+2KL*|W!15me4`jh@yp^;h1ImkQ$S z5}E7d->?rWEqjX8PmN{-F7AWw;pdQV6+-=aE`YCTJ&W zkJNPw3p|e}#pE|8!4;E*VDvt)T{%wvqO;e^eCQ^#gHVeLSnu>7cR&>#-=(W;Hk8X9Z#EGk*;={+&`TU z={$3i5%!fVCKRV_Q!5kUjufK)nfpIvkTa! zvvTZ6L_NH@C4d!VA5rf&J|G_VCD=+-NbHK&v-B<__Mt#D346AIHK>}#n!VS2=B8OURG(1{Ps%oP0NjNNtzfoDVZQ|6k3s1g8&>3EaN--b~9r- zKZ%DQmnrq`CwhwIWLf`Bu-I~s7|lv0@3$q9vdvZGjG!HFd;_P$2<>4{#tyO1b1QGGGk zWnxcWb)ANW`@L{3d^P!%Aq8Jvya4Te@u0YH{5I)OC0AoiVQO{unfA@x1$z|zzLB+n*r8KK9d^3*4! zv?Cd;=bh*0E=|MUn)xvE=|g-^a35C6PeLgoDSU$^mhh*;o))D}a6^@WwUmr8ofF`W zou(O~sF~^f5%)#Zd-n`#ijF3{89c?T6VSxtTeA4uoGZ!V2^FfUUKI=3mY}Pj_9B^U zE2x(<)v2u+f2mW|s~GXSO_XGrD6&+P=dOmiI46?AQ{!@(<0_(bqfS1~iM6B`>gce3 z6>s6}hg-xo;VyplYAb#?p@V;F+2L(&UDVd#dK4ZaL0??cOiO!5k!0I4JgVnOmW3+Q zs;&p|(6H_Q7&`N?nz}HGmr|NFs5EMjM$=dKp1n_lh-9dwfmCK0nq(}YLM0_C%_FI3 zBI=&~o+d(sLK0Cjgi47hgue6ddG7PvKhCiCUhi7Jg%59a0$=;fxW#-LQ~az3%z0vX z`yYE4iLgh{Uk{>mr6_C@;DDu*p%sb5 zHlE-cZjKVvyvzndV|c^S+wX9u&ts7T!g@G5dw?Zu`lvu!to?ms^qk zWrRilrqO@DTF|?G?4V~KH9{iirlW#?M`0oJA6>S5F@0_tjc2zg(ifKOq7`GW(z~Ab z;h8Jqu*$n>SSZmA3kgZ0`lv-%p!^@2X$53~;SRxzkqdBjhaU=dl*2D(+0yr~C|CNs zEn_6yb`!^HZ9HV-LVN6RgrTd;sngdis2QQoSbY0&5?!xCwTEQT{{Ix|sgdIJRIem( zm&zhRwZW(-e<>J+DDsnDy+=-G9^x-+W8vhYb2$6{FL)Pbk17~z+_hi^eOB)nb$U}V zb#3AoPUIU~9y)Di8D+E=w`}UgV_u<_46mX zbAKfsxM>U)n~Qi&cL%UTI+wq2kQTHx3ebpgAQNEkNA;_`Lh0+Lg4=a@_7=z7!VX^` zp(7ij8|08-b3Mt4UV>VsvT=fH1T}n@^O9&?!{$fB0FwAnYWp9y`({GMjZE1fZBte} zK$LR1x|XcAECO#AG1R#7GLb%#4Nqsy22$h(OJA(VjuXo$J-cmu`+v77yIYAwamo-h z-j84dysz*^HeVu>9y%k^a2S15;rh@vO+$a4yrH_Ck0aZPsW_+qFZ1d2Pi9)G5FNUC zHTY#$;7cyh_R7dpV!-jin6$SJZdkG3MuBSkp< zZ5|jH9AkH|mh9TGk8I8JZ=@+goXy`g2HTG9g67?P{A0u)pGY`F1-!LKzgiNo><$rX zU2imnZzLj~<1>=SKMkL!V*&rmMz*WY8ny)r*Bm>jgGF}fvY)7UR2l1rP0!duk8%q> z5oXP|J7NxD6LPr0AP*!v5L*}c7yWZ>W}aoJvU&S0*?lt4h{;A(MxtAa<6{({Xo(Ia z_((AZwy!+!g|Yj&@$_nppoleRY)XA zcPJEA=JR3lXCwAn!gf|r)x=_DM{wZtVD>z3rsU0fJTN;Mo;Mk>%jZY1?OWb}v7Q-w zCr%RdK4ikq8~*5@Y&O~GewKXdm<4as^I<%Gt6p34%?C(JW`1jH%a z#I9!FAYaoEjz1-AuTVM|Iu)^%vo-K&!b19zwGh2@RV>)VePN20RKVcwvm{LSH}mSR z5+zfYO9i)HL`5NhJA#Ipe)U*ND$NX!dqfc9$=&d>I0~MGGmv(287%!RhPUQL!nYbs zoO4u(X6Q~l;PH%kd5Ggm**T#lyBfiN^GsIhOAK2tGzees=0Hfq0<6H6A0>?Om&NIMu&zMqpAgw z95eI){!5!MHs6h4kNaa-E*8w+9}-LIv+ke^zcF?1>Nj{5?@n#Yr_h%tDOghWFr@iM zF*X-wva2AM?Z5XEd0%n?JC05B@NqH{H5F!cr~Lrq;xP8YvqDhX|C)Sj(uU6u)L~9n zGdUEi3ft#%`Nz}=vfNey#ZAk>WwQ+F8xC&xfq^=Hb?vKSbi_dFK_KdNO&6( z9)81r^SXgw_Tw3IS1k#pOMJj<+Puih3UR@}vc&(Ud`qXcJNZ#{Q`6CXaTS{PZ7HQ@ zsZ-@6V}&+%8KPfYH?c&?5fb3Kj~DU&GIi4K1@+*MI5}`i4nHbdC5YdYjytCX;)iJ* zN5JGAv$AZupjE{e{(hg$tg|uz?RCfCXF(#Pvuy}fKe&k}K`hUvCjl)#XwR_fC(*9m z!B`oR@QIc2f=)*Y#kz_@NVFU_d729Ul!x#E+60YN8CwYcyry<+@a1hxD&pA$n^Vp& zt&pU{UcB`3Aru2bw5!V)Zn4#&0}ZDk?udad^xeZMhwh@z;y%1ZrJRm?bXMTA)&X~B z+ak-Zb9|K!SNuwPNmZBcYRZWxNm7JG@WSPY5}g-~4W-Rc?T6F&&Py45^pG2U;+-M- z*0UEW`G`Xg9z`OSVz?qi3|~Kb1E&hh(hqc}({t7i;Afs(ZpX))uI7FlYM#$u(f=N8 zic=;A{e`4s{dD^9zF?ln=~>l9D#mExo#%MXg<|^Qwhi>71?5!KJO$*`euk%`ycDYE}_mZDUoWXt+Cp;COG)00XUyq;_qkqZJ zMuuN%Ux&7by5S+~Z)jq-KUVsbf;PYRrjn)g@YwDX)cmAmY9js=b3inoe=YGmem6^m z?z^~?p6y*ul0^!5`)?cJx|oyL$!{E=8I8ePw-ht62Ud}p{(Ss)!%uG2KiTc$vVU-qYiL~^O+$+!6i zo1>7-A9=KV-dk$--RUIyLL)Qv-~eUtGmY$i&!8&D74*gp*0h&(1LYpF1kvwfQGJ6I zbw}(nWkE+EudI)JYfT}1{$(!}eDf=xm--+7MRW>f6K0E6b9wNYZ!_S>`+U-TBnsc0 zWj(AAbfO)ppf!S@Q03Z0zsna{vL3_0-BuOt4++Hw%Tz$9V85Xap z5B%R;e%>6r45gryvoiR3z+5~S(ujUc*x>QhTe#bG1D5~#9lxD62@CmjR4*D+1% z>j3I8mm)0@{#ZpLpQjj{igvGBhyNBu(gG>JOS^e5M#qExAM4r>!Qt zr@tm^3Lb#f9v!rMO)5|2mo?1qv!fj5Iia!&3F7DHL}j|?5_!Q%aLKxaWl!XhE6QIP z#m>J>U9lE;2V|1Wm6!Mt^-k3t1A3IbCX33xY+x3gx{J!n6<{}cM4VsMKaNjX@Jw$bE(qEG$j1tjbMJ$VNzDG77Mpd zVu}96b|~?h%FS3;q+#w(PMpIvffB@E z&k|<)TpQfC{vIXE^(e}G$tHa&myycCT9i`#m6@gIfyad$@y2-%(Zhjn{DN{P@@e`# z=2hll()sWv#{yhLxwe?1wy=6;WPG*_VB2CQ^JoZ`6|urA{hRpyejQ}*$Zh0(P7N8&$i`0x zOPKclEFwlu~Ss`H9~SKJ2^cN2Lkr?r{V z!ylNSb2ISK=6B4)Lq=4SLj&nK5Q99!qd2C|2IROh09H*bf%*fANTzxt8ab~hKwu(A2~a3gmFL+#Ar>je|q>=fOboBoRDb#lGA64ppfyrH&n`LPjA?#C=i< z=$SqTolZ}7#p4C6#;I7eE<*uMR-Y!%X&Q~cNaBZWo=+SGF0kH6nSJc>gncVFhs)IR zq4(HSR(8~uy!e;ENb??1|9!N@CcHe9TJRDH-c3deN2lOHDGK?XnL=#BxDMxa38c$F zomy^DDcE6kky>%Z9hJ0P6XdH;VzOZ)NDKs`E|bqBTF8$Hey_`E#JW_c6oZ>CWZ~`G z(_qucStwX5!nXoa|k7!?UW&b7Gu)8k|LihS8 zIPvT)L}Xd8V=niRcSjc0KPrvZsv3c7ZYN3VWZ_1xs<5@4}| z3SRW%BP^T03wmxZMSnyMsj}Qm2(Rl!e<~x1>unvlG_(*pbW|9-!btXbz-~5ndo+90 zOA87|xE|!IK4i+L5%MMPjv$0iV+Q4nV0cv^n)Q9Jz-RIbtTagqGxiB6(B24*8BT^X zKMkSuZ4FM`A;zxbalRY%UF4bOWpW@o3@+J?^Vy&l^6|^ zUVD;Rw`)0i-kyVk4K?w*>~pZw#1p3E=&%;f1WwiRSdWFatcs^P`F^AFC&Rs~%>ypR}VHG#G3anv+8hwNGM8riywbIi?r7?EhfnGF);)Du6Hvu+6Kx9=eb z9UddM10XO`kw&Ja4+Z<*w^bXJn$ud`xpm#pTJZ0AOhWW)k;n{FlwE#}Y2G5>k4pPe zf69O3q&HSf{QVsaa_fe7p31CZ$`PI_C!*kiXs zh+N1m${iEq_>MUO#Fve3uV*q7rv&iK49@7-5{*@AdO^Hnv&m( z?3uNb>ae@%dh+5*47RZDKn>EGB#?I#XLWc9K02pSmi)_Tu16A{do7815;RH@p4RhS zYtP~<>1(KeMc44;4Ri2Lw_d)EVhi~dXN+g>KaLY;{v9WrrzNp2jT zO6zzm$Ln!D@_KTZ@@iLv5nE$A@c1%1+(m-kPW7Oq*RojW>@kkNd=wY%JWX=$ufx4X zb8xp>5w%TdI*mpT3yd{BQDx!v&>ON1KUl{{N4i?bw4-YLs{^r&^{T5>zMdKStt<{^ z!p8W*+;lSIhbSGEv;i~pRjgHdoLS*Tq0W#U^x|YK`h(31`s-wW9Q?+Ry3%m~1$>rZ zmfb{Xj+{AoWwc?n?7C{+xn&rBxnt4xLcx$t4vG&xh+VInp|?#!csN{`&X&-^n@tjE z5f69%M%My<+?_pi`IJ=?Jok4;U&J-cP_v99UVqwR7SB$uL71-bO=bqd)1*JsozJBc_-wc!~}Jbc{A z7dP`H!Cxi=*L@dI?Llq)gArS)6)raTv~V+SNqUDfy)hjvYC}AZ=#U1Tel+m$9M^qO z%=gOshi~Qe;)hPFD3!m*L1?7`zV;@ap7&)str30%(|^~|+ija^zTyer`CPVwOTb^Ers$?#&Qv+mwy% zWyDB)!UIa^dIQq?U4vf~4WI{{-$UVh8}4~mhb%>1317b($=}aod@YNyiogsDOO5k= zALT%7UI0|;&!r~~65Nw>0^e_sLDDK_a9~{y&YmVjZ`it$n9a$nmOBQR(%MI}kH+zx zm~+&P$G@28AsQrc{5`V!*@gpaL@bqsBk3=nUZY2j-_ZV1OJ3jp$3!@K39Vn+jyAbr zx==Ek{>pcumzD>EapM_WcxpOU^4o=@MpQ9-a)P#|CKr+|t zsgSL`#HwW}V$Cg?LeMV35JOCVI-j4&v0a;!@2QK7wOJ%`G<;&?{JTkKD^kj29Pm`fGacc?Jnkz^~E?Lc1+uzc?aeprNP zF`j06tQT)On{{UjyD=n@8PB-_V@*FOsdduuN#qCo3KV5aTdd6EDhu#oxdcSN(1I6b z#vC__!87C$==L3l@TdazSgJmLbmI)Hv04Nv=fa^jD;$?d7?A5>DsZ4}4tdmD$G^i} zUnz8w#2C^zqsfQ2sLm0O*KCE5!Zsor|BJfd8_(Ror3f)Imq5h;$NO@=hj#UAqC3|U zsjGVwm^ByQ0h~=Ba^aKlY1qsz+4X{m{~d%YCFkM(_L)!+@gJlpTQRcN8>y($rwq(; zB{|pnDd)pysjuG`fphx;jz^XWA2sJ=&nYyiwCn}Fj!;%oF`JmT27~SsacbcjMGMo9 z;q0TqSLC+{VYY5^#4fc@;ZIQjf6L2E_CZ$~-0e4`ZXTl9AoDbKyf~BnBrt?~CSGuz zhsaAKJ@$iQDQFLuL!shgj%_E76AvUq&%nOw?`l2*V}s|&yibB0d&Fr({3I-{s>68N z1%j$o10#{Y0?w%AQTa_VXmu3F|9>{0>>l?;eVP*RYqlhaHVi=D+emg|7h=TMHZxcB zUmzz_e=`4dEOo#48iW%mw%_s#a%hZZ`C+naNSmAbMVmO{}eEw^x_5Jhv%qz`d2I zz;XqSUl9f|4p~T`tqopDtC;wI{gD2181zb%1jcf4P&~Dk>^Za^$0xAtLi#TIBJ&*Q zGfjrFenZ&m9S?MQCFlLNXJ6Oc25Faz>~eM{JN0TIv0dWJw*DCcx3(YE3;G@q(zg>G zd}qd*PLULZOcEgL5iwL6Z^4?&JCoB#SxB6mL-OWUGSMZ?<_Yt|NZrO4Bs)3TBwAinjYglPwjzzDhGm=VZosncGvwDHUpS!V{QLr5+awtCf&Gz*Ox`0 zd<8dHEt$vC`_{6*|2qyA$5s-*c0RePcan_cxl=Z2TY)jprP9)Lu~t?O_|!^4&qY6w zU-$@CYQI6XN<6su?g5lF8AI#wEY{fVDcQCwfx30j0^5HCa3_WdBeL6%jQI0XNt-ef*k`?1vC#q8lflUPB~BK{@0AUx?@ zH>`*Sa5-B5fhXn*G)yX}byoW!4qROddz4SZ)6SymcQ(sNuYC>L9JZO{2$ZP9 z4zbW?ZUXa`@>uXc&#?@a;RxkM&`ep319#p*%7qcUzZXi7%hM&;ZN+IyUr>fbwPlzO zvDqNM;}$BIvl$uA8llL-MdS^WQ?*!nJ=M8Qz;!ZRrmB2KDB%;^1U)LUU@=Xam|nNR zHA9V*+Oily&aWrfv#lI&mRU&tuJ{5Q#Y{jsSZSCy{4l7mBc4j|dTo5-r_ zWX}6{A8xr7fozd78gTwY1!p-hf07en*FYP*pV&flC%TYTE62aL8OGiI>WCRl<6V*O zMu#TWV)=s_h?*>f_Xjwj=E#K%t-p&srfUaYhH8*3o{rm(N}y6jC-N!^;g_B)`TV?) zNh~^x&b*n8GrE$Q`XW1aR;vvg5Lg7F4h)i0Qo{Jq22wBBiPEcd@ss*aIMiu_*Q6&h zM!V8UnT0#Kd#{W0s!wL!?w(;(4SX1*+z@1&c@Z_IOo!3~QG}J!gh1LCUq98(G3Ztx z^Q&FH{-F|yNt8j-J$XXdW5 zhx>6I*sCoO{>_epEpPMjhK^a>)!{PKNmy)4Tj{W{J9^banp&ueWY%$*g zjm|BHG}UMb%HKo+HZzc0WWk2p>f^ri?zsKz0Z4jCk+xkncrcHm?&R%6@_p|q4Zpe2 ze(5@Cnx921Or(LI+K7LZ1p+!O#O2m<(Qw%>!PR_csxI7#{Os4FzO(Cyq{T+$yzV&6 ze*cd&&I*Q}#$~MTQ(yL+nkkItO@YwLaAx|~pJ-ib7S>r%Pu4H{js8~2;FSo0s&_Q@ zZxxU~=bgy(;T&Q~JtET=*3a|u<0kZ5|yAAixONB=Z|w73i0!G-^lSkI~w-QV{TNOAkqg< zG4l0h=>5qWrtpXintd&iv6FqpWJVlA|3z=%4{drP@UqmTB@e&AUr!6sekIQ+l{?mG zLzx0GQmDXBXV#!9hjAvTnn7{b4N-Id3+hEuHJ+362Yl*gqF*UDY!09)~XkQ zd+}>rf|PMfxC9FQb)A26RTZ%hm<;0EIo^7(8jE5NqJw?!_($vOVRV%T;pM+#^u7Jz zeJtm1Yi@$7ecNG!Nfun3f02kUR0Bt!`6PMSJb06x2px?PXwk4I&X~If&N!VRzs5zd z*WVmUT3e6P_nMGZ-EFiZo=2|!eMQasz7WeeBZw3^22u%;$h&8?X*jXA?UKB^R{@iSA_KSIc;{|Aakd@nk7e z`S3ANJ(X+&H@Bp%x5DSYA|%b~2QvEP%akfRQb#_>v-%nR#OHz=5>|+2Q?wsJdtN=X zPXsV$`}b4-{aV3vKP{j({~ks8KS%Jp11a!owmx(Vd$T=m?dW8dHa$f}8poJeL5#o* z@@ItNhdbV5ui+TDlMus38%eP7WIo=dq5{2()!88p9U_(AN~Yhu2WkOkps?f;T6z34 zBpGd{uBK*VbS)MqQJL^sb}G9qToEd#l?$j-Q>!&h&oeu0=73;rJ{5cWCE4=q4zc(% z49{))*wItnaOr~xTOxzm%s340GoP^mNy_Y7-*|pZT)d!SQz_bgY9IJi-$XgylC1aq zO(6H=4bxPa0<~L}NV5iksSQMl-{}+n1IX|%Gf@SRKM~88V-xa~h?uU>fDkj)p zGXkNWizIKK@HcJV3*PyP?)eK z-k+whm*!4lhlT&cJe6JGTNcYsagU@<-wVRiM$VD01E;9t(;o6;S4J@ls;y!2HQ}0b z=S*vkj5))RCo$~oJbiX!nh<|cTN-tIZ3{KJG?rWsXvC#-5VrZDg|_r_osROinCQFX zWP8sK=x%=jV+AK5%kv&tzuub-UNaA6s%~M8j_)C5TeAeaG9FOEoEE;&)Q4cD|DY}| z9o?8KMyh7tB9HQ#$Vo<$Bz~C8POq9=bNXmB^R{_9lBPB=(O06$nlg2G^I;0l9wswl zVoES1f0rsA*a)dUJUG7U3iDd$KcumAkbE9Fi;hgc17EIa;UBw=aq{cs1hR*}xg>dg;QtF)iQDSi=X2I=9+Qh{us^HsS0-vy!~EQ|NJP6k*%%ucSKMyU^O zV00C?z={{Y87s$FYXAK8DE84yenin6Xgpw#4(nua4sS9T!XO8ody;gx%Km7~eD~?4CoQEsDz>%_% z^s)FUB2qoRdw>d0Ye(1nJCI~Kg%+>ps?J;W@VCDf7$Rp8*@w1hs9B83j5$Q&=j=tI zQyP)8oISa7_%8a=xd*SWoP}4Mki@>qFA2pYQQHe!s(!||3#!+Namd9LNJi!)S>U7& zkCqs*#mOf~Z}((~_+Y{uHL$=Qqc&){j1!hV&avn=omvD^OQnche;A5+$8-S7>;qK3`OTAbkBUO6=KU@ufM%IUgRxr#kL_T=QXEq z{xl*ERpaQrvn>t^P$nH0lZoCOu7BdmP29C+GPZC3&J3L>q82z8Ax-KbF8!2@sWBaV z>6{1*9~`IpzElbJ-)u$=2b9Rt=?!FoVm8vtvLxRhnUdtMvAFkM1ts}S6faN*?Eg;{ z7w1jF-ak$<%3Dt3LJ_XZy;~0VXoM37m%r%O!9#RMdl>y=r38IbFppHLTqmj*vXH^w zB-9&O$fV{fA?-_v#OhKBzx1MjT2b|nqU}YQ(t)AsqcF)XF>zJl?$*C8wSXhLo3XDYSv0@XKDK*EeUPDsZzoGM-c!DoDk=O;5* zx`r>&%>xE;ud1M>rvri&bJp+!(jXt@m19MZzP3CMEEGB&}hwsh;kVfo!R*4Hb2~x zTnA3>Um2eruSn~HN1!<=nOI$u1|#ndRIuA(I*;D?NOH5+6g20e7Mc3{EBW*L7>V3=oTz4u@SkN( zhjprwutTv5b$Ep%tB*OTZxdtjGW7|ybZZ9_x#kcy>j^*#tvp0Me~(>*wdiv#-t=4~ zMpjRnjGdM!f`3atdL8tE=x4~0zv5A(#Bw!r9Diq)jS>_%-h!wgTO`J5zv5+!aosmb z!fVW-Z05+4>asNOe)XQxH*I5rXDh%Q_haaXzBskXau>7fejPM6Uk1!+$_KYzqdfdJ zz~F!nDAa3_J)RY4&?Afy#ffNLu`YI9(na1Js-Qd_{!z1&|B|rhdr^I}Je=IWjA@-1 z#ry+0^pp?d*y+GFd_|@Qb&Kpo&)*GI|EH#euD(4)>09+vOXeLX$t6=@%Esl$;TlDB zvdFnw>&ASbu!%=-p)k{KvMX4*|>yw@2r6YGm0N$fnt-jM+HmoxF^ zd$Wj`ZU!-ywWEY2=TJM>O+`0F=kSHk*PYxZXL31k>ug`B^EK-5W-ucX;5(0CY4 z!s7Ee7LyivS=qqv|BypiSz9P;n2L9NT0$HSOAFjvV@UagB6wHGAO)!?w9M}i`9&@= zi;iXU#Sfd{*t#t0#@HP4H@zGMP7ILiww}=0-2vNLQ%LPGDN=Sx2Xx+ekTQ;?+H3a~ z_;a+`W&Jzh$&_g*(Ci6PufLA6#OIM0*`JW$lPB53&!U(=&yeNqTq^T{BHg>_3p$cu zgI9ie%+GyMkMx(8kio6jn5`*4xDF;n3mFl>x5DA-pK$E5_9NM5C4y(1j7N*A&mq;# zhLCjQA*uV3iY}NwVPeg$ph@FT$&U$LXkT&^T4K(U<+|UIa-}GI?au%o?w-eqU!eE_ z8g%tU3Toa~$;8e*h07Ni;U?!M)OhSY<1e2B-xo#W#v_V&cU2bJ9eW*@21(*0wG!ma zQww-<&ILL=Y+y>l7It#pGormahSJP=CUX9ElS;o!p{A~wXL>~{kUb$JP2a|;EU z;^**1!^6a}QU+%?#)Dr*89EtRK|C~Xb4+V=jOAhOf$Ue2O=*=v$!&Zm`)V2>N~&LVqTsO=`Q*m!@41_v{2J zuDMEvk_Sm+6CaWK^VIZ%Y9Rla>nn@M;{X0{m^mHlftO!?%mltDfz=|itnvNns9g0P zRcW`3^|$Gw=F4gm{k(Qql{l9@bz?Pq;rk9&XL>U1)t7|LT|Tfq{1beeEk)~Hxm|W!^5l!5m|Or5XDCl5( zOvbNy62B@UUr=tPP4;rvYZiRQv0n~TW9DigQf7kLx;Q9!A%-6Z18%yR2oAg)5_?bu zZWwC8vGd|k9v6p_Jbm$3-z)syEh%Jju^ZP_mmp|(a|8cge2tVC*^!=!yU-r6OAwG8 zK{n4iNlt`SU~2I$vP^FY?CX*M_COnX@-dFwtB=C9^>gXcvTRs+m0Oc*PoobmEG2m+ zj-;h@!X@8E^vg>Zuc-+~d+$e+40Z<`QJx?M#WS$V#>Z5@rayKL7GZ_1WwEsj93f=3 zEd6DcfZDCF4Ax@_B6&WJ)WkPYd5-F|)M-t4!>U5sk94+9oX6J4aNm~PX!K}KD`{Qd z4hf-ySn7c^z4lxmzGmbEry3IBRAD;xNYIIYi96w-=c(BEb}@MWB}l?a4ZpPr;X0UZ zBNN#NNY!#C=Pl7D-+96CPUQk6^17SkpL@#qz2Uq`?)rFZ>}@0xx0UvlQi7S)g-p5c zDRf!Bhn$bt!7kJsBw@T;XiM~A=oQZc@t;cYG*@3`(nouw5>^@Rc!-g6Dj-Puj%tTw+P!EUJ-aau$p*->1?DC5sA#XcjVo>8Q{ZdYw_KYRQvbsq<`)nJlR7MTCYVx-V8&SmpqTG zm(3(!E32TSm{9&3xNpqL65`5vgEy9QKDwG(YN5wN>WFs|vB(-j)~<=v9iOe#+XW?1 zHt0<~nr%*|U3iF2^QJ*4lMi-d(_vHn1?Ep`O||QY0@|}C3wH2kf%D&Lke{JVdPC)q zyI?v_-TF~*$4d=_IhGw~yd`%cGRXVAnIu$y1zK3bCo8hU@kI1$^5%LP`DiP{qM<*G z$E+L)ptW~GF1`{ z#B0+ha`@^T^hMc_a701#&SD2m2;lDyL60kbzek`=pU4AI-jpA63EL|J%$QI zZ&14r-z1e9g+xNU6yo%@vaIg{*3L~6Ru3kjRGUT^p0^a1`*eVZ_-dH170NeSV+e}_ zI87`+4zwP5LqPTe@LWF^YPLm#P=q$yBIXHQ4Lww|ksop0Wy-E9-w44$kx<@wj4uOS zs9fqe@#p+i63fIvzr2iWjGczlYllfqIf7_~jo`dDoosUYL+%?FW2)SQd?|edM}m(d zcfBik!0Z658lOxhw1^?!{#8hHl?Had5eGpW`(@pkNV08}Au2f|h5`}}V}I#A*!e~| zNNPEf8E>q?K#9+~ZwR4YNNB*(GxtFwHWJo*jgZxqrfl}>B#UW6!PL<7XQb?tDhiJ0 z_V_ks=s=G;yZ^v*V2<7=iNAh;>;*F}f4iH_8d*dMEr{kLXQo`kgg zY<^tiVzlUQImun2kDsrajrM$f%`|-@3p@ux26^bsqf&l+;hw_-v>-; z!X?T#R2Qn0s>!&uBU%$ZlSo$kfwhKYmT4j+YeMym@dc02v)cD$cupO=C^eHUn-6T1r7t$v7zCk0^6*nj z8z#eLFerF#X*;K5oJcPbl z>*DK``NYn82TJj+!jC5Z#G$97@Xp{?%$u_VyTtmE6_?Y|?M<7Yfh6J;T4s30vP!h- zEa$;2aKICuos6iJB>kS#Qf1@wk>~FDxOKZVxpPQ?9{ack8>EQSfx(m5&r3X6JSfI0 zc+SE_S*u9x4Of_*x12s2^Z<#B-KRpnyW&GpT!)hRD-tAc1MV-i8K>rh=tGe>c2jwY z6gsk*vDXpT`4&5nzF-^rWYUhCrzq1?>rdgqLSJN-A&%e4CX-exA<*!$#d*>H&}E6U zsH)14Uh&z4ZZe92=&4ndG{@g}7g>d*7uRCz#UWTYVkP>wTnf*5ZUDusCYx1!17coX zMNOSi`1s`8M5fyWT`P@)9qa4~pU10S_VOXzku$??FZo0ydO9pmo&hCe>tSS_EV63( z!npN`;R8A;biQU8{;ye;<{zctw#Z*81P=#Dx z%hSy>xK7dYJ4&At%t~}E=f9I zo)rDYZW%rCBAUM2l!}XgIpgo1dSu_(4R{l~A2%Plj-$r4=;E65B%)yp$%!(Aq1sY2|YSO-tCgkV>V8MAtH0p*+DPJ&-Da9FdKx*nN_+?Gnf z$iwB39{vy#Pw=q+I%WJOCJgTzD#nSITX4hKVrJa30!gadVl~kR*v0z=PJE$DW){vO z>Gob^AGf}dUoEKm97g1q0dD)Af$w^4p(d zNvND;A`*w$-_=PWV2U!ea3$!40)fo6Wz9udGMc9?U5UDPBOjUmmA#AcU8OTsz zaxWZ#5pbj*$e*AO?3=(xB+XIpUUQUH#=wA!3(owZfNj)jiO$7Wg16}o~V_=}yYnsI{Eak5Y(7M++7EUKJel^=OabKiqyS7-xp+VsSH0-=EmT zi?`oJBxi^bA8jY_`PD^l|Cpyh3OC?SNIjZ#K?3I9 zoJsuBpP=N&j_7$$E0eNnCwgM5g~UY4z{Y(e;wu$WBVFn!zV$SYm%7B5Kb(X%|Bk}H zW#{2P>?elWy8)j&x*G{DZKoc5)J39e&!b2ENz91Fdq(DG4wEV$$umtbKugT(k&Bl$ zGm*0cH>#%N1M{Sbjo2w1U@D*rr5i}uXkN9^P99E}R*eETuE9zY!cddu1t0CjAXu3p zt(QJB_f*BGX$D`2WP&HWRn(z6SM%|;A|ZIu{)fC7jX=e{h3M+<4x&2g5i#R0C9xmQ zqb*6bDC%Vmx%PA`e&ZobDQ$~j3Uz}S#p+u~`(F^A;`ND?q|Sn_fG&^eL|IY9>)>+eEXv34D$ z{y+n6yb6XkMi0dD&Cs5sY2>UE4^I8ff%A0@WR5tG1onu*!8KuUFz*d1#&VE>@SI^5KA1kAyao@;ioT-s{^~DS0d2J|d1c1nGk`yAf+2vgKRP zJ3(I3s@R47L&6g5n7y~6h*nWJoI7!m>ifPF`S~|e5h9W}%FYP(ntPEULZcnYkU5i`uk&2&`n_L*h12GhV>u1iF|vUnZdxDj%imsB^6O zf5fk=7M<*H!Cz=&xQsn)vL7veh)uO_2)tpR+R!r|OORuz4&~l39^VAsNnI zk0OPXp%T#?rFoVn3K0@2Aw$Ys=1QDrJxa+?p_C|TP|@7KS%Zpqzn|Z4@3}5t&gIj& z&e?lEYyE!rJ!jb&R@k$YxoxQ|IC^Qbe}A{l)E27R($4u2lYel zg(++}bvy_&=ffOrX?XTz9BWN|gH`dN)s{)t_D^1k6Hx~_Hs$>X*jCmF4|dBlOX-WC z*6>L1ui}=Vd+%9nEgU%7VO&LGQG z0XqULZ6;6RtR~qB>^Rry0+B`kzt_Y25!|OxH?;kDgZ_F{f(1b}_*Fj_ypP=%zM2_a zb8W^9IJV6g`?7q54ZEBhE3I{Bba$p939JY z=3xvW+q3|yHC)<$!&tY=`!SuQUI#BjKdKZN8;jJ za!jy05zO!KyB#79fw{fh)Y=Mk-crt4%n*kiUn1zZHal+Xv4`kA`6IoWvJbOs&S34K zYqYFvB0L*!1~DrusDsf27Q5pJ_ipV%>a3hbM`q`72TmPg6C!S6p~hx#$>h13o9t*W zzmpI)X%qOlg_F>*C&Az1VWr+yq3ysTwleS|EAK31T{>|f6%&oAQRf66WzSJ>{~y+* z7|7~&d(}3D@2u65+*^BCJfhazMXq+_&K5TN#Bq4xd>ytpiIBktWiT6506qH>jK6<` zzH|2G`K88O#|v*Z(pa12$hmQ+R$oMyMN?Vv99Kwd6tGaM0%mwv98|qMU_@G(U1;}P z;m_=5>OH5`@Cc~X7 z`0b;B`5WY+<(kzn-s>QnvpXC{_~W%E6UMWjI;8t5UBZXxfkEl%;%z+ZVFKj%= z6OT-oMyG8NFim<;phW+03#kHFT6*9WOP=AiKp#xDy|LSLONL2SZ(%Po(4$tw(L1!!igd|PH57l=!ye&8xUstCkVg|tOr z$UPo6gN;lw!8Mp;2XUd-&sHXLK52`^?!;vIh- zw&>z!wk2vCTJ^~=mod_0?gIx5izwqb>+=m`DJK_j+qb{L>m#-uJa&)puV{<#Te z+x4Q$9ZNP)aFJ=rO=P#;{6oVIAJ{2;hxr@Nuyv9~biLdsDlcrSam`j`M~~*P`Gb*8(KF*a zMkTr7>79|-Rqlj=zCYpBYD=`4H6#8}5s|x?f6oYfTjDssCAk>TQi8ljg*nVFWF9;7xj03{nICv|>8e3NMPwrF zE5C{2TNblRN`5F}^dD6_Ig+`q{lbZHI(XghlQ6MK79xd*u{lu{?{Ai2al9M*+p`RO zCLK#Rh=sF1r4=k-nN01peTUfdj%d`3s)ajBlVF0j7K_?b%x>`a;9JQHnf;+S`fSM% z_1|!Z+NtMZ$ct!NwYZbY>4_4%e~nP%8iBu-e84aId)c3iCj3$Ml?z&C#TqwB@iXE< zCMYWBpKE1V*o7>hFB9O9PzJ6UPGKgEdH8q27yPvF7=GI`ENqf~i9R(}EW2Zv&XlU5 z6ZboUviM{iDqk+-=6caF*~7SMrU8BLr^~#fs@aCo?QHU~w`^I%brca9#X0@X5H{G2 z#gT?Fs^!2hk-1QUY^{#|tm-pb8-o*&`{2x>OC&^Y%+sF#O7DGd`H7Q&v1J9FV zF>A61JJ*oRO}9;;Z!blI=SFqraP|-reVWBiUA%_g(T6$F(0GWesN~kyeHD&h^ohFO z7e&WEJo9>oEjS6cvKX;rEN%Y}_LJ;lDmRnxzqtFbJm3m0{4<6bEtv;W8oDfdWtL$4 z&_!rUXPB3Hh5I&s9^^i@fuD}wR762r0fyYdmaLh zcy>gfY%~T)^2cqvBL79Co>2i4${1JPW8Ynk_p@$qgM`kcvKIagr^?`{c zR5J2s6gq#LjN<+l7@RnssPC1)^15NJ`jbBHY4AbUW(ipA=LWI6)X0A2YMi&}8TF`` zfZs#YIinr?S@)VI3sIQIGuuDFr0`AL#=t!Ibo3$|y$txEOc9eqxARBF^}>=6Jz;h) z$9}7(vA3DcjK)r+3f4(<+Lt#V<}ydvr$LFM z2X3tDG`c-cmKA>gN>VmmW=4q>?CrH|I1~DfMpcQ@gIwvxM|1qlok3Dw24O5_$f;^3n;ce8zQlI!uF%e=kGAk13Lz zmy1ZPhAC~2YaB(iVJ$53w z-0=!@6!J*NtL<=GdjqVG>*Nw9Dc4>}x2i2`^JDb~6WMj!c`UJGCcc~%3p4JFBKZ{@ zp;tuhefB&9=~^Luy=-5N)$xDyo99yq3QB=#`G+8H#WlL_b~y8MTZb|etw|4xu&b53 z_*{7=y3-^W4!6LNioLMbse;H1Rd8E2EM)FkNt}Z$N520xuutGwFn4)BsM=b4*1cms zyPJ`W|5n6;woVkp=U#(L&zF+cnc84H(jC?&)(BN>Wyt=cSKyoTPE?E?&GQJHm|&$8 z8n@gfYiGI=;fVRdM5#{L@w}7M*`tX)yhm!cY9&*)EMbAac#rPR<*+PqEU~(8Owu+# zwX3?~4t{l)FtZ{a0^`n+k&=7|cR&(8cD~~({R&~$?;P@9e*qD{bB458>yXnW4JiFk zoFz9_0q5|!YaoplC3l1pjdAQ z(cV`M>cz6SZEXnN4fi8LgE97puBh7g%yJ+fw7+wiRXI@g-yDJS@d8xO&4=azLkK;> z^Ke`K3&|2uC(a`OVW;0p?%rb}C_T+ThZHTwU9U#LiJhlWyuuu|8K@H9 zWr>`b6~fPsTS8$;0;ibP0DD(e!I1xXyU46}@Kzxbx40XTvisU3ci7neTX#0Gebfhk zbyqQ0(JN@RejP^nd6D_|Mv)l(L!^I_lb!r7Yx4B9C26quQN#CaLFPpfm_9ZU?i$MG z>ar@R?DqoL+jE8pa|fY%(P+4`=%-M!&zCH4@1;=|8EocBF*Yw8@M5zl{x?DdPb{^k5w1Bf>e>@bPTs-UZ&e}L2iwThCFW$XHHiqMQ&{NI z29~^G6cG`9OM=?(k{juU!yM?`sZ#mNn zp`FihEqPOnn#XTrO_La_pHR1X(p={9tMu3B34%lu$QYFwL zHWf;2L*P<&J+ceWJ`{t*SIu-*K^6vmjUWj_kHBPcA?I{Hk^0=9Mf@)R2TLa;(#IoD z(ADLuXm!s;&Q*RCIsd#G?l}G8+9&4$^L~r(>s{ET7p=4&qsjA0#!P3m3cD2a-R^Su zVWEU*5u`fa<2DN42qs*pv(2)Y0gZLKHM3{juGzEvHwJ!BWKU=0vgJ3#$k^OVq~DGs zU6WqG-P^k%X;eS1c(96z$KPY8;Skf?l*T)MQppaJF+}J3X|8cwvA}1a3dwJqfNM4j zIoSpi%LYG}v!SIT z!s#N_cv3upUEpWf^LusKvo>$qR8|g7sv7og1JmqxD~S_VgEUqWmdob79>9m5$!zPC zZq}tA0`tqoiP08IGEzGft2Y0|qQ9QdwC*P;dYwgMm037pLL(Gz{z|=1n-j5tRpjq2 zbK3U7hwYknkyX4_tDUCwh!Lk!vU~9=lHVLjl&%g6UOpU2a?Q)Zec>V+xz!as;@kyF z*WTfj4f+DwaRzUW&8K3`Jvdrjmeoo7f_2Mevi9|6CL+3%85m5){6*uKwD)58a7L3$ z?kVAH&` zs$f@XrciH$9F6$$sb3Bp7p?r=BY=f`XJAGu=#<98>Ye)V`qM;gcZ;M`2mQgOv4yJqn!pWw;GN+= zwql(8Qgq|{%i{PMlV>M!PF+!4;Z_ay`NwR#XEP^~zDesL+T$vROBUkCu!9&pI|{80 z*Rhr?ccvmy!)%N*$MdWndo`WZF`O2d!PgP^}qqT_dMg2QZCR9 z)rxp_##)WA?qeSd>W!#fsFYnhGHZS9k4?#-eJg@)yi<%Bf!>hy$d>JI znZ$1IiGYXN4+VSne1a>(2ITDPOb+d&aBAN=GG?L%%lzfU9r4`7;J;!|?(xGyu07WfgRFd5wf+G(V&n^RMg)QP zWrXWvO1Xa~S86n`9>BRt1zg*CX%KHZg7;GST&v@4YTmGpD^%7b_eXEH+p%~)=d!{A zA6a(tXVEoqAybd_O^(O2j}qAS`nULV(+K|iAI_YPpNHPs1ZYew0x$Ky@S#$X$h+Kv zC$skxSJ}CouhCaH_tyr#{-;5r-nqe#`x%_JPMKg}lNyeDe->+H8*!$MHa?9D#;BEJ z*hW~5WAyBCf^0NRbTq)a<2O+KsV|NmD^7+cNPx{p};@ zu`?YsW2X|`%x5OePZ_hhtwGQ+kN=1H&yJu_QeMjpoQQv0@I*Rga-r zu{^52i-1_EL^$zWlGNK}(n}3jndjY&Ot`v3m=P|{%6!|wLQC);tU(H!% zsRq@q{DAMDnsfRWVzBdfv0c@u161}&0gU+?CG08`$H0Bw(bI>`v>|2=iXD+4GTV>h33hyNE?6d9&-B z4S0ISW2%)53*)=0GjGjh&L0-AF$p=G!JGmty3&lnPI?>6 zHRmIOgKNZ~Skn&1XsAKX${0>$Y#;qt+eeL69?~@nZcuJv6`po%MYhWr{Bkc&zm zmngy2l6ZBKC0*-oai^%-KzcMSR4j&BnU=*6UwY=xsH??(8DMfSer2LCK` zee(;Laqcm&e+9z7_kRfU6im3(^1W4oIoojS0eze()Mk!TUr@!|PHxKl3y^QxS@R@5 zf?9Sp3v-v87fN**bB#qdsHpY=kE%hJ%pEiQbV+%aD$cEPW4AaF6&*FwL zC(t%;GhPaBpgYcg!$}F&*tT#NyOh_4-Tkq2Wy5*uxV4$8Dp}ec7M)6uMLeR%J(tsI z_qsqwvz)%ZuEu)q9;fP0BQX6;5sk5!!)^1|gh%m{SR3y*FxqFvLb`L|YETaRxGKR- zJRFO^)qU{F*0(4qO{<<)@68^z|G;J5txU1^3F~6ZS!L*5`l&ICZrNu*{_zaNHBCiq zTYdn3J~KlopYDOD?(D*Xz>#D@ei?leF&E1IYI0Yl{%}3vFQ`>o4RzD$pq%=R8ol9* z)WKIBXO9&l_8uPCdjAZ2v|%2mbw#0g+9r=dx$(Tv_nY9=tT76eYJh zqTF36dal8geK}RYj(l!s=46;1dG~~kE!b2mcVuSm0jYX+;Q2YM4I5jTi;n#XnXFNSEsf>dyYb%MCG5n}N_0Cq zipBnSjgxa|=gbzAA=C&_w6_ow2KsrnR~}~~x0uQgMdQ|KJA@O1_tHg`$GB%{?U*f0 zq84i((D_rNV5I&_zRz`;re{w_*)&x_Zki64)O4F(7?lBA ze@8+~cr>{1*MTwS2can522xM#gnpx5_)_6T#5P4h%nLQ4=*0)z(p+sCV|)OY{N=fm zUtb7rhW*8{y8L`~>REWp-QjoieJq!gu2srSXD%kkS@_1?&^fYzD(eC3v2|u`6TTz0 z7=zZ@JMoG_Jh6$=BySb8V3~an7f`tjBzHWbZZ~GI`Aa0(-}KS!(u9p{L1Z(wM0w-U zf{`rS<~t3HG{fdIv-sZEeYE_;7vIBTsGfTb=TtBno4oUC&SxWa_uNJwdB$)XW6J5N z1{G|&HJzI-Nvq(+OR#6YuJlKBlhf61@BaLN7chl*lKVZb@MCH`}0%W zp%4aL96t-@v!Mz@-8idDj9DpKvk~8Uwtn6maB%sCOB}`7!#8_jin}#>S{@hPUU(Jy zS09E`Lx_`#nz7FB9-XRl2%lWqgVWyWV|835_Eadcj4eBv2cLiXkihZin+CQ&V=)`J z@5F9r?Z*Rr9|+oKkR_v5(d_lvoXqe`Jnm*k?CK0jPR4C6v*7}cDZ7gE_+5_eSZB7a zy&U`G`?$Ps3_hF81~DsfB71c$nICl!lt%*h=NiwZ%xs|3*31WynJik&3Io?7!2Ms3 zz}b^67%uq`dic5agu<2d(~U-GJ}-iyE|Iurn+iC5nFghAOyJ47D#(xS;`C%k!yeV6 z^sdEqG%~k_gn2%w=@tXulNDfXX&}}&@O&`L#S@L6xvj3Jg(fZwIp=yQ@)<>r7D45R zF@l0chSck!It{OO!z}~BXrcr3cG?f@UAU6HuMUS9G9F}+awi;jY60mA8+5Gc=Vn<> z#DNQ&$i^kSqeessZk2XS&i%VEzjBz$oasld&7AGLdlF(NIRZ$C5^_bFELOe>S(_KY z{Hh1snDP+#xYh+4%+v7fz&Ug+EyeCyLppYcJ^qO@g}LLagv+|U;7EuA_agENw?~tJ z`S71j^F`U@lk24R7uwA<6(dN4lmx!hc)vM}J{YwH`O^wFW zBBkj0Mo7=Pb_#@O2v+;Y2_Gdc=2k4W6x5AaE9S zws7}r0h|q-MN>Oc={&zEP-~t^Hr!4DbxB)z6Y>uFw(X${^6jx>@k5$=Pe{eNAmI@i z3qfq|Ct;?~auoS$hBue{v0H9otow-+bG0^L(KgJ_&^uI-qj53FL1dN0nvC_;qO$RHo{K=~8QA zY^%n}KjT>j$IVf@?DS`*Ft^?35&H(Re9Km!&`TAi)QeN|tfam@`p z{!D`E$MaoRQ88j!D9$U;4KVUo0fcV9!0i}Bfo72(G+*z=I*;{Oy|f5_m>=YFTKn;R zj29f8TF%Wwo`n==L7WBmK($~cC;M#{%!zzmv*u3@>^=9ire(xift-~#(^F7p!Xi@^ zezS{u>mlkM8VfqXM+7TnbMS$SYwa#+>)LOh#@2dE#ds2CSXUV)#saW*{Sh3OQXyRLpbHi&53+q?8)2H*X11}~fD=#X#M0NPSaaYK z7OznU+lyuRaA_7UYqsSc)Ya25d6)TlcQ2R}iIbM3BpTg4jcZsQ#u0}II%A3%hVGw> z5*OXUV|xJ&`g@6Htp1BJTa;OSj3L-0Oy_P_wQ$)>O@t=*CX?%@{DmokT@bO`66BZG z;62}X-f>+DT4@Ubt<{L_v{<-VHOx78sbH1wW8Ahug$XvvwplHj(nn#HW;J90zu8G@>bK~+L!rT;W zV*;qz@IqmbTOEeYGvaDJGcfCIHr8}!K+Zl{^6F6*XB=$GM9N;%tkcru`%f7x`!ocf zrc~iBc{k=gpZ6`^=g&s1tI<l4<@}U_hIaZzI9eE2kydAkX!!nkfyAf|`qzL?0xS&_i9>EUnXvoc%BI$qj zqf$XU&hqpX>P|-*@IwW5NS1-pKQ;FLyBs^D#ee>TuOX^!4DlJd3fY^ka|MiI-I!(c z%%(bs6Ayv5SM8X*On`n9l-Y9gR93;~366!oAd8NFfHpe?(xp=b4F%`%bM*j?2ymi< zEB&BNst2DmzlON$G0-eKKu=%)&RJeeU=P}PPxa?P`0_}a$er3nx(5Cud15wXa>z|c zYAuBI5)>m+qtG>24P4gBf?HS)UD2fg?>t7p%2Q*=tmqYJH*q{u|1JWnrmVt~9S1Pb za1^g83n1-dY%zV?1$ywzNt`+1A9q$Mil{xC!cDh|#1HFY=+Tt5PL2Lxv0ya0^q`g-e%X!9teEzAl|svosi5>{9~r(e0{b4XfW-wOL^!jG zKGOa{e?0Ue{Q(1Dt^ZEI@wv_~sb-`wAd*O?p5&_kZDT&85SOg+XEr6tFs!A9DeNrn zUw<606=Y&%!#LtRIfn4UZ<6(S5e|AKvEgaUxpg9?!jlQAOycJfvLxaaiu7|#QbGwU zt%h-?(GpU2c>rjh7oU%f;6#iXIh~Lpq16Fj*nZWEOPAaSxo?Dm!J5-ppIwQkqKe_k zD?h@9A8{dn+u&8O5)1lo4XnJNMo#r^AU@-#!g&ufmiuNEU9|Xt?VLS*FeG!F{8u@j z4A?nC$2NPqF5?Te`Z>(cs*ee!J!Hs>hEz86cpsEsd7DC?cQW%?k8a(x~4QQCW~${vqDPIU?#?r)`;+8y|z_XZuIyo#GRk>Hm9Zo=qCyJ@|N1^!;OhizB2 z;XA}BC?>7R?z1TRHOT@r#3H%ru4_>2yeY^Gn~}RpE}Tb^9%h?IQD>Ezbf$To-ADIZ zT;KvdSP*WC>Z#J`d2EqTeoO&+Haw(*?=}lE{^;V(--T#&d^O6CHK56X%1~w2$!jU7~DWbvbRA(?M?*cEMeP_0ZsPlV;pq zByf4Di07Ru?Hrw4spowUexEm+v^2e^Kl}>tU2nf&M(|$pGSPyhccg*uOD~Mva~XaQ zAOzU?laIce$(>CLnFfE|oG+Ip_-Z3YlO!_nbk2Z4=~E2&7>{G;kJTViw8E-aC(x#D z7d>?^ozB7b7=!Pq3zI6?Tomz=qqkjVN*^i&U!~!`rAGQdBn9&b4Dl=yrI471v zR_q0Ey!($ky7MS*9-U8T`aTy#{2QirlM&y|U`|%Q&mbGrXJDV63R@ZJ z%tC+4!EbAOdSZDjSJJJF*P~5Apj(Zu5vF)gJ`E1dp9oL>U4p(F8thYn0(l+SPE#%v zpxv%f_-XD0a><^M#-cfB;(Y>(u6?c%Ojp9>`OoRMDj7`vxrggIe;Za-6yTm!E3iM* z9Ea?l(;R^V+x%k#?OV7IM%dev6!C5Naa1>RALGE<9y+p;i@kKofdym+mSX-+8F*8@ z2?nR?k(jLqA%J(qF%MIsZ+0Kv-3`ONtGaMaUW6d^`COjYX^q$B7zoE`c3_sB3#y-Y zfC6nzk~FRcx13IbD()oude+hU!65X?8bxy^DKhJNS>CnP2u_b=z~hPy>5M%BA{(Az z*wJ=eJn0Zyuad*`15G1JS z5Tx=;3Y;eBZ7nnSAAz5UYLhNSr(2c9a zSxx;OCX(gIoR2SMrcw_WxAid9U#LnBl%L?%bcnO29C2oPW(0Fwoy=N$2XVAw0CTyt ztv11B3hVhbiLF##z~uSa%(b^$*-w+XY^=u{ye9LK=Y|}IR_miQK<%tBcTkqdNsK0X za|yY3Y&(9ak_R7tr}EJ-neWj`qm_dbJL2*Lk}Z0<6&A|G)u;`8*4N`XPMl@NB!aE7 z1xo(h%A|KSFx_Kn?8c)!dhE0wwA$PeEaMr^ew!q~#443{yZhkwz47#_$W6Q&QG;(6 zNTQKb5FEYRh05kGaP;7FSR9y-N@Z2Jk)KDccc%j?KxAeY}Bg;}U zXTP6bAd64Gg40R9P9eNLo!HpV<{hClnxvC?bb@#YQ#l-_hwIA_CjoW3*#C5z+Gsy!X+40LhL z^ALR1Y|05gCfGfnaD)sd>k?PDT)ZzG39Hr~CQCG*5rbVBU^l-D{EG)@^noB+D|(fl zJf*|#SGlmC7mJxqpEkP;lkxiXDL|zRnfMNCZrOG*Vpt)HW9oRH%Gg6_Gr0?=?p9#Y zk-RH)lOBZ4w?osnVd(ih3o;s|S#g;+Mx7O~ft+JxS+_k%xGiN~o?BR9t2ud`tqOf# zOj!Et6X?u)iBImofIEcqAf~aG-%pPxO%|y%KctRet+`w`h1?b#e24n zo}=Qj1$eCf5j=PD7k(NygvNZPWll8j`}4{M?nopY3f~1~t#M?6_93?JLqB$fi@^lG zM>f0a0%u+IkNX;%fO=0v$he9U@^HEnLDK`!yTzE(X*!C7{BQQ$l`t0IZOC%tx1-%q z5x!MfP6z)JVY}s%(00pzxXafBo(){UBFz-^xG%{zYd)h>V{}M^ zkHE&rM*g03mFgBoaX|+J(^l^%~Cje zVl<4)G-rQb+p+O^C3HcUA0(a$Lhq7Kbi>EhSTJ%Ldvh!g}I*th~x7$cx1f^SFuV4Zi>s0r_R-6l==ud z(8AAt9|?tYmKA63(u}^XqNHVR0*a+BoiIK)2QGw~+BztYMZ3{q%*Zi} ziO-f~3a7qN^$u}ZI^F<8745hax-Il|=yE~u{HU67#jW_yY!c%nrqYU#Za{|%wzoGs zf`Zsx7+5FGHq8kj4`x`f<$Zs+G&v!6@|GMo)GWbSDony30HZ%W#uoJm4BRE8X?2(r2XWlX)^@h;#5fF2LU_qr;N#7 zIL}OSV!21AKk4}|H?ZgR6qdB-F<#y?lU?FngeEIH>F#DXuB5h+d*2{Ok47h8+vp}v zuzx8LT6`d58p61HJQw7@kyFW?{R+(Vx+Rsq%J(spuH)ah*Sx=LD;vFoqJ_#)+^>~Q z-g%^uWRgynZZIZKwYJg(`{O85z|V6GoawE`99-L&4e9d(K&U$jAKiF}$41BCt&#U} zMsEccv>H;K`V^*BJPo`ioMAnqb!%r#*v?X;9Ju=h8$f4FfJws}3vjFMx{(o)Q_Tm8Ue+<NENyIum=&`fYg*9LV=Ecx~`Lu!pfz1%h7?fPE9873jhGLeG{Z{x7w0R$3 z6k6t%z>2gH%t$t!7#t9R2KP$Ll;}d6C)POXi3Pbl(Vf-g$5j(@}J&|0<^ zf^LE!cwP;TImxjFMbh-Y&DlUsxMINEVeWjnKKQo&g5kaI;o4mpuIh~xJNm;Lj?9k0 z?G^&|d6z#HrX9e~qa)$*>u_OTt`x|d)ZWV1{8nm*BP^{02R!L$9M?ML`rD zZWLqwlarwE;7hvstpz@*yNSb3H;g!Z7$2wor7e|8Si4#Q$nN)c*(Q_W zY@Z5BOlzPz1urpN`yqEswU-NWdJcl^5%_Iw96s(Zg3mw2h-Y3Xd7smYqa_uXN#`YW ztSvx?lM~oWMMKn{osV5{;IYCpO>q2T%c4Od+*&Qdn;|8vlv z9+USK6uOqs(9gx-8n=M&85EK|9g*z)h^_3ow=LUvS)GY*-+}Ayw$jo^&%vl)S6DV= z00!A{U^Tgp^Yj}5AHB}dfl~?G3x%cFUo{1v9)Cdh!%3Vxu%eo)gD|xsVylj*1;stMM8^?8Vuf+1P;{= zf)CQ`iQZQyc29CLOPXGR?u#Ah1n(1^*YC5CQm0J64}8WILOnKMxgUN-w1V4wPc+ns z!kbPiY~=a7)PCh1e73m)+cWNA`M1q*`r9GiWw-XDjO$>JT*f8x~3HUJ3 zm8EU_2vVQ#;gm5^F#i4%tnbZa4V}e!d+KB+V;2uOibhPOLyM(f7Gf^XlWg^zPEPo& zWAAm(u$X33R`FGiz2Bk28g3Wim%D|yslbF?yVZ`md1Jw2TO!2zjRZ~UX9C%4&fK{Z zhuOCH@oZ{)4#>MtW4D_h!#~|(pvKE^#bztKMoZAnd=uuhxR5jI>p`(6M)=Hi6sYv5 zkfmC?Vdy{`C++AD13WA47sq?-v};k2*DO#jH^jVp4Gfn`rz-~aaJ0@*jIF%}J_Y>D z>W?cHc&M@Kemsxl$7*zwIg8IHO+mdJ1r*3;;M+eu5AU8XQ)^Mj>4nGeov1x55VYd7 zNyF%?c@U==O=6ohj$_i29b|Q{I*Q~AK<4N-Zr1u$U@4I0{X)BWr(z^#&)J6IAB}33 zm7Ed&D$fA1a}&ARBLm#pXF}S`KLeALAnxS+GdTTQ8~1LH8?&(*&DOn~z`Wvo*k7e{ z5Wn>ST3O$xlfE9pPk%m8SK|trmykuLMaI+JX2v|XG@JHv>oL5d1V>Fxf`QW2sJ%~) zoZWjvsLyBVx)i427WGBcYx+jqT`5Wa*o{ZY)mg&6b7lB>hAW0nD2AFtn)KdsTi#iq zhPnRVIF}p2FxFZCA4jJ$>0jY&!NVN>JeYxoF0Tdarsy#X!Fg=`H=l`~48ec%+v%I& z2#8Dig7t5*vFVc;%Qrd#KjKrl2Y=qfX~n7VLEa78`TLy*bB5Hs48eritHIZH9(URP zjNRY{RqSrJHWWeE(L}gfy%+x5AA`rl z-q3SJ3%FR-FE#r)Iqcr}jc#eNz?Y3(^p5{6t~KO0Wf5{<-S-jgR_ix-tUVTw9L=ET9pb6e zp?Pp++%RVs*2_J7;sm;(=4|tjDvPoC56ey~vb{s8H1F#v>S@|8SoV_d>`8lJ_t`Y` z{C5zm@693lYc9a+%M_lLOaPlNUtp#6Yq&f0F+9=8r^Svju-zz;UJE}d_>h@LjkieC zOP~LviVuvK1-}QW9oNYf+&)U)w7$k0hcCc)_dC#AWI*Z)CX(5bu{iyMA+epCA=tR` z2F{+jnH5g2K^>NaGJL)%(R-j~#wcYx94AH8xc$`7Ux?Sv%_M`1PvUgrtyt1>1F!D& zL(xwHe01+Um(2#bf#gWe>g_3rdU~D<3_8bMoasdB2W2r@CJ{~VuSctA<1x5kyvC)Wao@z|B1K=st&Khao_84^3iNuwQ~uLK5|1~+A7Me2w99X zMD@Yk*ovj?+6mQdi>c#+Nch$Ln5O5u&}fSoF5gB3Plx+)$K1!^riOb$bh(5tgYIE% z>RW8fyNz=X?7mS>CmgkX5BF58 z1TuAP{)eLT4#)ca;<#B>BqIuul?ust?)#>QzS1sQn%bqbcPgupt<02FNRkoHea@pO zQbY<3X$tM3CC%UG_m{u8E*CEMIp_U;y;#$Z9Hvoc%-juh#bwb-ptL>?dJjDVtq(EK zdhd&n!8WI1Dleg5%x(B8yAb8m6LHPdoshV!7{0Xbrn{C;AuvJ-+M|UV_JmEK?H>f+ zT&(d)&Rh5;cOAS8r{VjnmMFVe9hNn!;P* zr5us8?I*E9WUi|10=)2*hVAFZu&igYWYfA)82Ar0^*xL?o@3zHO zuPt;(ZUZbTv_Vz1MBEzj367c%$GF`Je5+9*)KMQexhV$hyBCqOg8Rlkw2wM8I1@?r zGLjT|M11JLQ#iiQ1L~HK1-4fcBue#Ab+Q}g4fTWrZRuqE&RUw56--1Il(|onF8uha zFABJ?M>_+8=#{dgP_sRW_sAQfVILTZ;-OmJW#0hxx)f!9YO@^NT+KBDDtu##Ng+Epcfzem0!SlNf z(I_p4e&L#U%X5%Ky{--&y4R!2y&zULM@*XzuVHW9BH?F*F79t$A>=bAqQ*=)rZRLh zjaxexr`64YNUKO1=DLJ}r4|eJ$%i{p33woMJ}MT3(TyLxS)QW4$RcPqz{yn;@!iVx*h(#hP} zW=478L1`+o*bJwCHIwOmuF$*V2z@Nf`ZG4U!Tos-^iO|Pq1zz^W12IF_IN*bVwNGB z-yM(ho#x?5>)&w5PuQzErNXptx1b|rJGY4=Ow<{}mYm$me1@0MNoix*zp`mi*b@ii zXV-y#synsODkGH-H^6^^bKtUCJa12X2n~sW)WG%%J#hXsiSR!wWN{cgjx(Zb3QcHu zSv_rO-NV+Fw3qLgBY{!X@i0q|;9Dbu>!egPvg0&~a}hH62+EE`kg4&aZ6J_)QuV zKkWtirCH#8XDF_?oDGZURHk^&n5D>%hbMm(X-RGlVH#DGo10RdrfxD{+)6eyInrI} zjeJk`ET-l&lNlJe2z)#_s>D)c1W*+YSkY{Hye%-ZRyeVZPT<(6)BRyY0!VzhKkvk2(PBMc0 zO=oy$|JfLDSBZ%yPMCvttN{JgyYP`Rat;roiYywq4WGIauz zp16>uI4*%Rp7X`O2b!VN-$D4TYY=8V10I<@6i=Uw;c=&0_z*c$*d~8hd^{(Dzugzf zhdBVWsTq;F{3&#C!UTHhiNDCoLIpCnFM^Jo{;;%2AIe))dBnN`KDqKKUpT5i7Dy*R zrP^i)>F*($`pZES5@!ukw*-a30)fwArhzRtH=(}9-m?4*6}%gF42<##$n_4RD`&e> zkK^g&;g?}-ib4|~xtF2dt1NC>Vhe$_`{}!!7qDr19I@}I5w-OOLfa_?s;?$+DF53J z9Rt?V-ABCWm^-UTPm3XYk=a2rr5eP(g$!csUW%8*48b2R=9t*$P5q6v@JqjRabD^! znC&@}zqBn7Z%Bx?Jw5UbkGpl8w^}X-=V{WIaWWY9Z+3?Uxl^=6>md{@azPjQp)kcZ zk6XlChp8LXP-o@=e$Xlx_S})e7lmm!G<+UvNh>;h>y)OJ2HJRFJEsHXjG2kHDgO2x z0P^#UNuG}rKWu%IUd}RS68%kV>bPC>oMlbOE09?;{#R{?b1Ke$&KSUwZCMqG)+;7I#fCq7FemqJ6{8m8p5U zK#EZh4Hx!X6;twPYpe#hewM`5N1VaT-L&6^Rg{zP)Ib#TNeuu$bS>SBM^^laC$4T?oWPYMY zhfGMkCSGrLlh_*w-MhoFB=qTdI=KBEUT_#AiB&?Jx+M8k% z{TLj4XArUQ3gwl5(?mbqd%^nWEtumR&PTgQG3(4tq-N=3KE=F}%>Oo*CoIl^855SG z?coa2dh$6v|1}qU1_^J3rLXzwfnGuu{4tp>67qEq{*gW7?P%@m4k${o0G>7s9+asg za|$3Q_Ed3E?*cC0c!z8-)T4&Ir=Y7Zl{ZM0g2qxCQ1LB-2m2q=uZP{)E!A_7u>2dW zRJ$p@6J-XXFMF9oy&ki#GAMtyi^8qlQ9S44Nj`J7pf;Sn z?LoGq3Gzm<&|&PRA4^)e z-Q-^*Et=hyiif_4VS80ltEeZZZrsiGV)}d-$@FrQ&(n%ZZC#u(Hw9Od&He?9y-qBee?qo#OHVCVdQ8(bFYbh-#wxsHd zGwD1<1r|B4gv4y0PK;{n$->NjOtO9_jeF6^yT1j%(`HRDulvqL$#;$3Y@QyP{~TcN%i0*|JX7dd)nKm@%N`7+IN{7t(c8B|4kDp!i#V+oX3wT2eG@O zkzBi}L3OTZ!KGWN5PdA4&YtebnyQBI`fIH?By%@z`k~AI##hmj)r0YOwKe=5xDs>| z9jV@|bK>4Tqj|y1ul&{>Gpcbq9?02dX!WWht)n_b6$NYI?8Q|yu>2gC?HkQ+B#ncO zKC5|%?K;{hN#VOj?Sv?um15K$hbMjf1y9pa8aU}PPsDOGODIO1QVClGzU%W$XAGW* zq_0*6Zd`gmsx}YhP6~JVvXFG%^1g+Kn@EHEHA5<|=Sic+X@j@@c``mUid&Q#aCaX1i&-L%8n+t4@5(YyDI*|#7aXdkZtxE= z&q3%PfYKbnr=(^^9Bv-wGrbIPzc51=#l^sLXCF|z>`irE2h#T20aPKgg+4o70pr8u zsNb(ve$1wlxGh0YP_O~%v2O)teGz=0RLGrkUkg1)3Uzu`_HD)sDkOVee$o1F(Xct|7TJKcfZPtXL!rA-?FPwyz4UF+jJZE)CHj8 z?L2589;lvt)8W#_^;mR07Awz9M&+hJjB)9Pl-%=lq(VMlw0pRC13e_xt9=a9vTs24 zsz z-W4aXhYw7smB2R2xShcj2BwgiN;l}NKcRHe0ZsB&w4bk^Hi)EucZQ=j%5c#5KmMV@ zNhEK%hI(q7k%o*5T%{lg!ev&`DM5{*x9(o#W5GE3`J6Azbli**Up1JKdH`MLZ3XYo zv*;d&OaLnxS~^`FUv3J3|7-%u>6FJ%e)%~0RY_p~<=On#<2U5|t~aD>Y8m;t;TyNg z?SQW###p;Ycptugz#VSA;p#W1ki{N))Is64_?xFH+)LX*+@>V+Q+lm&`cB8UFJM4|j z19&5c$p?&J`)y6SCpntMN?Wjqba^&faW<6Iz5#K~4fvMQ3@TrO>C?06;9%_!ZHKSY zJFy;gcli*Oymcvb?Hx`gm7f$X8GewI#NH_1bE%dqoYrN7-8G?nN6f`Q&G#D_xx3;8FhNLk@YUOqGL{_h;yE*;gd;zIQB_4vg&xYeQP%JRmi7l zN9NKiX|v$`^guH5z$skLV=-~UVcJ||Da^-l;%l#)xKVxtUKAx#(~uL~DQCC9$vF-+ zVM*L!y*5Z2N@(;YH}PGUWco|xBE_( z{Mb@Jz00rhg8%m8{Z%7yT0LckEl;V>h&NF0Y>Y22NgPr_|a7k7uX8n zf8YP;idn+jHpP!e)fqrPjUW8#u?u{aAd~x@!eHj6e0G`% zuIW3tPIn4TIP(_{FH^v)!feu!g8Z?9(36fDj1G(Hq4)0}m~*Cp|LS}q>U{E>zpK~) zIwc!me5NgU9?ylce+%H#;~f6(+C0+KI*iItM^W@t;V!=-9DLeu!kU=-ke{lHYvy%8 z+4n|RS+WG3CalK+PAgFF#!5c+-F|kYwHk6Rry#Os?5s=1nzw7veR%=S+{RSbJ3`)FN+M}2 z^?rX$?>h!=ZAq|iOa?ewEQ4X~@}f^$CW;LcRuN_*!zY*a!r`bCjPj7+{z?%VdZiNC zq!myr(+=}*2>IOX6%hKv9y;iIKKJS#OsyV{*VR9fel^F)k%alUamiiy?2UY8P7vHx z7kp<{8zA?;kIGkh0V!^IV4b%WoO?e*(wjjTHq{Y3w%h}z|CkOF?oF7V zOzLfB(XnH%gIjML*B^3{)->wgs-iH{r)B4z2&8tX}Jo!j+H{}Y(HGzHVkX(c46tB323z2lKOty4{w@UxXP;2 zKpS=u(cnEazv&1Kw8`Uh=Pwia1+9fEjoLU9_3&!AaGx4~7GsKqce0WizUr~X8z<7i z_n90sZ*WIhv!nP=x&VJ>Sm3Svzc5De4c+Chf-etP3(jnRczx$2IBm-2MouN{)LMYz zqR%iRR1tmDRB=OT42J4DWBV^-Try@GVx=FZ|Cf(JeluZj${)I|#F5@joyLwm*QMfP zX%Ja<98{DUjp!di@90^w)B9TZEYAmATU!y7CmBQ48)fh?HswhhyZNp$TSWH%!uiKb zi)hQ}+2Fi!K8kb8$Yoa*OwLNfw=aiczloQ??70aJygdw#yRE~ZoDNa_*zH1>T9p@E zTaWv%GpznOlh5}4AhPT41wW+6;1YkKZ#l?;x@Z|-RH)FccyNrm9eV`K0EKpI34f$( zMD~P5&`PSws?E~*8`WCa`>_#@q$@z4_APPU(`npU74Y4|I((V83#WhSDid8&C04f# z$!xt8nxY|%tL-j=fAkpP?fVvjyx&4!fIavOuLa#MCz$?D3Pp-fpusE`ZY>gW*z+bK z*|-?5Hicq9y$Nbgb;jtf(U`W(4Ycibgj_@)g!cu*$M<_-MZgIRvfqXV%b(G;8B*9z zdtu(vc2Kmof$B6nNN}0NMXfi;D2E$7JTDbwFJ)kbMlyr$SL%uion&Aq`eXObcGmwp9pq4Vhdu0l8_@CYw$4(AQVpJCACd!pLk z&-t)bQE=tIr!e5TDcl6`O*fXE=Pctu^CV6 zsU=qm8i=mohxavW;O1?wx%Q~v-0E>6R~wYU=Pl^tHm?GJDr9l(slUYSN1FJf=Y3r3 zj|qIvs^Tt2MnYeE04;WOM0a5vY4Q+y(`_}Fl$L`&ky?1^hAeY&T#S-eBSAmdgSU>n zUS2oY05u1Oipt#niZzn-=)-{Dd~%)%M7(EY-~3g);MfMfdW;ril|_iWHy#D3DS{yT z(>QX!Gpzk;L|w^V^r(=-QJ>rBz|wo1U?Qr8c;YnIAaK}_11~g&V8|vA|NSf6cD>7Z zzuBoIP3Mrne={Z`*+O_RKLB4jPQt8yr%|?i0xmnW4Ue@gNBQb)xNdDa{Onf3%gsi3 z^zSIt^eKk@t{(XDiviFau;snnn824g4j4m4weHx~;Dnbu>JKus{L!rPn7|FwT z6hr!sbS%_%!mg0Z{I7SQSaqu+2CI3YW$8%NIFSL?qC=4VD}`3Hn9)9mF;H#kgtEWJ z;qMa<;p?tT5MM4Ok-s)w@b4w#-GC6(Epdlc{jJ#TSbxY_9fFChnqL1i06MN-6ZWa2 zaH&Eibv#$omBWSH;`n*huLhN`+^$Oulx%U+;o1=&;R(rxfH5nY?J_A=A z)e~h+F+-h$$KbH9HQrO$h{fs~@k_E5o>@PFb$)$_OZPp2)ebkI;KN?b@{2*&Ja?h z9iPt^c2&_U;4*u_{|I@p4Sa~C(Mwj+tT`EimMUV~z3sSp#6P^8HdZo5_Yl^U7m-^B zc2cF)GkEp9b$Fto5-UGt@zm&vSZS|9X0~tVfs4|3udF;wc>f5>uJ*_7m#X-%YC5!d zbdgKeJ3wA%B$j)|qO6e?EnX4?#_Jj}<3E9?{8UH88fD3^BubWlbL7tpPSU$aE|HTm ze$*$anw)t$u>9dTJy_6@fV17SaAm?U$%^53@#&Y#@a*+X(6E$3)13}D_4IS@7ie8x zqAf!@iNLSf;(}J&W}>e!uV3DK1-rH-W8WkXJheU=Teeu@M*ZEGHC6`q{^thoRw_t7 zl|PV7I4@lhKXlWD8#4x;rO$Uu<3QOaO+Hk$AM08 z$kc)w9ePevXJ%1I7)b_9B78`6G7Xw9hjT~ELSxnw2=<7DA5nZ*df{R5M3;BG^WQ6U6J~9-G&6~EPZC`J9u42$ zd!x>(If8q{R`Req5%sQpf!&?vT;+)?9CYWf>PI`~UY>(RVg7XMw~2H`K?prGcrmqf zb|N=UOUMD)GH6--20lF8#z&TfU{FOj7NnlQxid0xQP63)JMAdy7^_S6O^uewoph7j zh--tv-fEJs#y*lEhZjiZrCH&WX9Z+lW*enjJ78kYTOM0?7;|Rb0JEZ8s**_@ z{8QDTz%^hl>J5^{;r{R^D$JNkW5{S z&2eU$v}6^xkev1zEqNMn9?uCLxC;ii;DF#~s|`0sCcFW5jZ6@kHfxFd^7}yx9f&Un z$-@BF&b89B$dbAKRL9^N{M_r0zYnF6H_;Jf&ifAnMcV=82a zuG(N<^I6Qf^BzB+UxHWqOmM~2laQ=%6k8z@%^sxUjT?K!>&5w?ADIG~CGqer{S5jZ zPDYzc383UY8xFqShV>pzkdYZiqZ)RBX%VC5G10Ux{s}jEoCVUpYv6eFF@ABr9Q^gl z;WM5bhIuiK)bm>sr9&1`lN~q2w=#jxeqbs&``Q3q&h8{h;Rz)ClQkC4_yc}}M==H0 z%^0$s(s6&TIxPCR8MK-!%5~fy5Xr3sSUCO`Y!V!bW6buW*8CsPJv0G~YSU=nF+ZAo zpz!=Kb;GsY!oDe zCvC&Dnkn#U>QmCsa29pjsYK&Lt?Arwp)0$89vm<`DBig}%$~JqVdKb5&^@~g&J^jO z+X@9*|NJbEtT;-ZveAO;dlZ{o)1TB$TZo;{JRoFT7LN!X50N&ikfl%vpGR4Oj7u~> zK5RNXdul)jwQJzRA?3JQcR1<@Gso~1a`^0H3v7zLhck?aNzzJA;h_>K9`gGj4a)r~ z8vV;w6#vwiYkWYF@q@pluzx6ho0(1;N*0j6qD-rpGaT@C~y2 z>~B$GWFB2vU(J75%P^-@gjiiAH+>qHNPd1-U-MO@xb0X+<4`EZFWz5i(#tvWZ*;o6=cAsIJ; zoK;Haoj(qfgNy9MuOOW)?KLdViL1fsa$)%Q+EB9kN;Gj56dNnD=Ft2}n{lgf8;E_j z8n!>NLW^Uwv0T>;XRjHCl7W%<5Mt^3Aqx;wrYn&-~O911QbyhGy02xM9*x z9JDG7M-FbJ?>%g==(&_6Y(A4%=|@X8?YbuEZ9gk%xUgJeQ~DmhlzZc=lw6!Us2Y#t zx#Qp?_ql0(8=gG7T9TIWR_Lt7+jr=$=byi{lRxHaXm;9B;`w9={xa()x#=U^oiw|I z9;|Rnntv46`j+A#ixkLxpMp89RiJ7pEeY;9^B9V_o4 z>hta;<920TX}F8-&mTt?oGK@)rQXsOy|u71#|tkRY=?cz z71-W|+U&Y>0uOl8PMu#YpxvEi{CBt|lv-<%Jr#ERV2a>!v5gkJ(FiS{KgEmuU1msk z&2J%Y<3r%ks33CIX%c)?^>@fC98E+0kHGV!FXGq>IW%ZT0>`R8a;fDg54($^`??3| zL3MdPN!hOUj)+pAuniKqdOn7ZZy1Y+D;EzD!}LZ=hQH(j6`ql z;^mt@h!gVfF{>R%SyXaA$HBV2OyYZ+^*mRlmxjvF^3Mu<;$uVcN<{n%^RME!?niFX~m z3C}x1;p&($p(lA1a--5ta~Z z4VXIJ0(!nDL1M-vmT}{K`K3q4n0L@Y`^7t+u;K>--w|V3_NBpO+t>HJ_oY3b*FKQN zW$MxMGD_4tRhk`ICV1%@X7j)9cc{TQ4eTs7gu2y*FlKHNJ^x0A-F+d$ zi1zER&GtNTg3Qy0sp^G9*1c6sWsa+nzTv|$)lEvc8!lyJoD7wD8$|9ueMZB)Oi)hf z2M-lLp@$u$@bAz7t{?Z6{&=xZ)RSsR)+Bh*gD&PU{YerZv~e$qkUm9V>3+K5lqQWf zI!QZE1d9zLB3Z`=eSuH0TW}&wq^GVd7x-dB!Lc%sI5(UZHLQKeTYZfM*Z4WQsH=oq zC7h@8w1c7Ty&ihxJ)_by0!aF>IQnpEide6>fNL*5Ozu1w$*LNXSz7o13xLM-M~)KB z>%1u5GBg0(HofBu6WpLsb0~9I5=Fny>xEvE0Qj4%DyYLwLVw$x)Z##@xL<`Zn_Ud9lbp3{xq_Xfh z@xH80dBP`gKlja~JhUHLU400N$2nZQ%@{9zYJV)hjlMCEkjr*&MOWnwggn6=s#{aX z=e0%@&`bgsU89xX zYQR453Z$s3;ND$D)U`2}%1@3X^4X@W+9H}7&-zd7dipi(P@G9aqk8Ednn=(5oleE` z#iE|W95cJXv|sXn_o z=rMaX@VSHIha|DJjyiSMkq47YCTzh;MK;PT1ZHFIN4>KMF^x&5i#mkaUe1RVZFM7y?wZnvk7h9M za1r}HKAPqTGrf<`Xlf9%m0vzuOk}DS(tk%^@FO!1z^&IO2`rna{5b_W%57oG?+Jj{u8?)X4@u#EpqCpsDSba2yiNp@ z^(ygvI;(~E`@f0xZw3*ypflXy?iTWT&H!*tHs{KA<7xKh0JieKSG?wHgTpw7A*|t7 zJ+0bP4yg{~@I>e%adP=~Dto<*zS1^lkIaW~;~*z?;jt}yFZ>UTIMqO7N(A=D>hTl; zGGK8j(!qa*(h*Mr=_uJ|k#xlnY84+(L|uh^vB_u*`*<0O6*kb0NDa`hl&4+$oLG5p z4ZV`ShMsXK=A_q|`UaliN*_%5*N@%2%jBTl5Um${`UoxPw^ZmF-B^yD6~!dM(}qRW z7_j>l1zcs>2AKELht|9FXWB1oX^ZX-h;UYBAzt|;_LBz4`h4NrUfkoZ+rE&RA{A!2 z_8I*?4f*slC&VF1UDPHr5W27WFe3$N7JAB$lJn!Kjh-HE{n8GnPR^lU?;Phpf7Q~l zN?+;w(>BavsgO@n74FZeE>P1&X>R*8x@BAy*_u$r7i$U|=?xCJ;*lFoQOYOFdX}(B z*%f5Di$6W=WxyUSnMTWp&f>Onr;_vUazJ(HDbg&wFFPJBM5nyz(7j*yA34Vw-DQB9 zs_N1Kr-SG{=X^T;$tpHOw2rGRSE3rRrnKAbIizK&VD0+@IBoNWZ#UEh?;jO zya@@V2Y%XNs7X9R6Hg4L`Iy$UH_*fuA#e7k%GukZ>6>2qS5 zz3w0PH(m>2VlQr#l>?_2JK(`5H{g0zuK13}KX6`PiIX2F;E9ZXa9&#u3p525?+8t{ z=btIPGg*f&oGZm2Mdm}{BQHMqaTl!LkO|}N{S^Ahi^#z5!f`S{UNX@76XuxeV@=w1 z(ETWM?;8%nqP$adc7qI-2={c)HVsKydV)YWPLNFVTqB9m-683GGD4zTaRU5**h0c7 zQ|O7lEbu7O_^U`~*nMOJYYE+lzv49n4d1$@Tx==!oOUgH+TIb z7d1;!?pq+vv}nh(=_Zh)9Y|Xj6+^1qFDf4}1^<>?P%_Ip>x#WW83wZXp-_bIwCDTHui1K2cT{ zs9Z!`mc=nJc0k*k!|+w@Ak;Cx!~LG0;d-V)Vn3yPd-##rDT=!X1XYw1(o# z$5v!tS~84E+l<3Ii^=j=dE{fHEe6Z<;DQK>cXG42%8(^6(kBL_^?G1UvJ9)3vjyw} zu5jOAUG9=6g+8^iP`X_V-+tVL>aakj5b&O7cejE`O*_s%XM+Ax`sB+`6>QG4CE+&x za9`IEh@BIS3p87SM3-_G`~AG=fINB_?#HMuW!!#bGR!HBC(;F}_(1DCMrT~bM%8{e zGdPMCDW0QOTcn|VYYI-)+Aqw=)mT3E3T9pzjw933;rE8?sC=#q#c3WmPF)R`O#cCC z%ITtaP4Vz+{&&G1C_1&%J(cf%GdlG4&i72kQCbl(7$Vo#xs&2 z+fEnCQpfRFyHb9_G#sA1J`7%)N8r?Bk8n)YD_&+Vjn7Ro@P1Meeo9osXNzMYV~Y%o zv+&^wii*6gD+n*XcEVXMYp~<+Fi41h21?g!_{;yIF?W&zF6%oClY5k~?Qc6AsaXPD zqJB76YZrQ`HN(W&#h!7mfe$V0#OQpJqrC>kiO|?Mq?9C^d{5 zz_4O&FEuN^3cspG(v-9)nwpi5M$4LTo}M<=*zJaYXGfs>$X-~sR~`?(NQWgO>}h(( z01ySeBd!VE(7uV_&FHJx@NX@gvAYb1kCcGT*X?YPLOD3A&gVV>9U_m0Oi)k>#G_J` z(EYIvPi4lT^^~JDr)ae(tm?h!vv6a#TJ%huUjB-oJAR7?2>wCylpq{ibrjz17BYOk zOR;_hho`?!UvyU{JQsQnq5l;w?Z4f zX%}`so(p-olLrP!jfKr_mefBwlr?N!2t_@~*r>J>xBs@F3)+8hU*iIrbl!tj&wWC& z-xxYvueB9<8(AS-xJ=sOqiOr4aWKm0CCPuZQ|ufeqAy-=gm4_mRlL@cxUpl&WZeK* zvr%A+svhGXM)kvl$PTjBA`qBnvFMP~W-2<35+=+*uFragG?S~Ck;p9P$5soO@B<>=ssN_)yU3Kj$DLC9pdf9TxU01Rrq0jegZ&LrUr8EY#$4e}qdxLW zf(Lkoy1+eIF@WYOhT)?5Le^oI2oBwihBdN*n3wz$=H^+j6^Fy<+=K7vg*_)|{O?zy zNwAo723TX4j~Q%yRX{YN){@nq()cvpb#P7K@Lle%CmH4|;9;+vxW2X)I-W&f(cLn9 z9$^j#lwU&3a%(7WIsrZ#kWOB_1hx}-vHiZ|IAHp4RvsM1tDU>J<}yw6+%gB|)%sA| z&c(3m`B6~rvcj3yd*H&#VUjG}6x8#62)$48VCPN`af{V0-XOfsj=V00puNUi;gb*K z&mN6(pX%V_Nn>dDnTB^1oap=Ms*-=hdr)@EMT}Y5L93<=#EwfsM$-de)WCc=`{)Sp z3VF1;S&a+AwdwZo9q_s7F+BIL7qzRqL$BT^*s-~r|9N8wKMl&kPs<$IzlBhxes@Gi zn{GHPS~?Wg?^UA1R=VLd#djzQ?Z)t7U+DQW@etr{#~a^yfxMm}_Nk;3h3_9>>u+Uz zY!HM@(;xg~Juv*kYj`$c1Pz=wM55)8EIB!;QS#rHNs{_WQIf%7QIaL=PV$dK?t{s_ zA_#slfOT44e#eB?D$X>Pzjpoe&l`(3t_(UGe`F4v88zzM}nj+xxX`vsy<2CkK z%tfP)dfv2_(ySR1;jQf;RKD#(a(;gmN6!#$Wm22?yXk)P)0-5kR;o(3vL7(7a1b^Y z*iq$Tb^bDX1ijxGOdX%P(YcbtU>_0+6Rj?gWvc>VsbUWND{#WGS*}#mat8A~{e-T4 z84g~xW{_<(1;RSgVce_P)bipN(H_w}a&T)1sT*tw9?|Mxf1w)k;=Mr1d83f|97;U| zr_LsW7BJV|42Bc?;TpID6c0>y~#lS1f@aO@$CQJ`by3HkO)fcIQ z#UR%AQw?GinnCMWJ)d)M9M~l7rtG3SJAQVquupi+Zd{9DI&TEWgY0d}oRw*M{b;r* zH-TL_vz0yC@rk9bGIuPpp5Qom`*6pNTaK}3CyMFKn**>s){A4LK0nV7 zd=;^)SDWbqa+Q1uR;7=wucSYH7!+Ah3Cn> z+5mdUqnViW$GK6Fd{+c35cV9~CtTy%(u+Vh?-T!bFM@|jhRW>;WF^`=naV^X8vNK8QZk1#J^e9s#%UGu@%0~` z_`(jC&QZtul%1?+eHW`T7kX_zzsOaKXY_A!9KDy#P}|QCjjaJLrkFzLNGs6Z-beTR zcuF7q35Sx2N_6lwRl4K08T(bWoV^ctN)HAFu_@(v zQ3sO6=bURfB@e(m^$sW-ork-^eJC>`m&y6PW&MuFGRqV-cBM~;y*M+OD{tLFURAm< zX>Awy@MfCucP`v#RLdQH8kf)=8SB{S&9AAm&qF#hOP$)qXYviz`gG;PG4x{0AckXS z3+#f6{QK-LJVQ+x&fbw`12;d0ID`8nv)@VTsjwMJ7KgDv6(iW-n}k|B%P_mE&4f?A z!n^Cv6Nm33h9E)K(0y|XO-ubu>}LqK3mVUdI){U8Dk<*n!7MAbL~9^h`J$7W5JgI!oueT?y!oy{@$l|}EGsTM z&XOPWvcQOejx9NDEVIRv&EKQ}n;owST@*90{5=?qw@X9q-7V0a_LTh)%e-bkpPmHcFN_Yo(n)xDBXK{| z4;D7KL*12AP~8?oWDT-;L4Xm2**_pPe`2})4m;TWI-VyAA4@XYZCh(P`GWjNzQk-g zG3Vo8e0LC(?#$txmH8yW|1pGR824lKhWXHxE8nXvghIG`s^K?++KHBFhxXE-w=#ck2DeW5w-P1d0qUI46cVRcX zHeQGBJ0dV3CIdC9OK0brZ?MUFa~!=#^>@4wx1ULTZo#Ut-$bwYJZhb0!EWq#qA6dE zSx`S?cz*H=2=;Wqn|I;E_(KqSI}Dav=R;szG5lUuLzg`(Vx^I%>HCy!(W;y&#JN$>doxJTMu5Xd!3F|PGB-kk#ugb z8_e)NN()X6q*`kAymXBMta-4HCf>5+V>ifACG!o?XMF>BN&AG|L|J|z&fI0)o6$%+QmN?YO|)@i)h)GQ>-cJDqEp`kJ>#`V!QQk34V|mu-jb? zcV0I`BrhW-*RSxf*7Kk!dja0JRKqysN@$Y#%k(@3Ikv2w-v1& z?`btdX~&lPbY{4`<1h2+k_S$Uilb$LkiHF@@_`E6 zfKM79;tGRrWOVfnQf&1Ir9W1v{BJRqI%R-Uwh#5&nv7)o&@bk*UB(6T4^(_uyjhX%|wB=xhj2`w$kQ5OEqGxYNdUXZk z^9s>T;Q^?eje-qQ#xQxCH~4G*hiP3mSm(;X1>Ip&S>y*b zK`s5N(BXL>tgl;(B*Ks!Ca-QTJvY|?;M)Y)9DAe`pjV4Wl z@WyqCd7nQ9-;(9Tr)(^E4w=gyNqqmf^+zN|_A&gHsftI99$@RO-t6#>NOtpcy4d9G zLi-G;gAM=m>AX|rWYM5l^!T|0{nmA`Pic9gZ>k1j%C+)>W^>j$GgvH=-hIkDc7g6> z8DX=`J@ziIAN#c~f(;BaAvgC)jE?3la67c0=i9#5!$ZaBkp%<1dGsMVWC_Y zEAiLH-F5%LEc9b#HiyU&QHj6m;Fu|O(Ony(Q76L`7IYn!?!>0_#@-GHj&1=%hcgux zM@q;oKM&&n{0%#hb(0NC@q#mHX>fSf2^KKD5uZ%j#&>#&q~|5GeTAvL!jkRgt=v%D!q(csq4X+XJlN@$=NQ}+Yek@`CKK5*e0k8CnKusfE&?*~? z?KgTthQ2p$iiic9#yULqyIc6}S46f4C$TM+HDr!O91+D7a^b0zC;FfPia%CLoW$D> z9z>os)$JiWZs-$vzrGUx_z3at@qpR*k0dM3tRkE3=aBOCb4dSBiKNef1Tr)<4FaVO zwQ-<3Jk-o1my(XdMAJ-E51K||eQjXeBZmH~7;An?G<&Z@le%Z2iSKZX2))jfG;1+p;3;yaZ*B!G-vDC{^uaR^ zf?;C42h1RAP&%Ppkz)K2y+Mga;9N0L2x zU&(r(mt^g-+u}_ZMZP$7DS1&jj7}P*Lg%+8Lmm4=q7hluzPSHct zPmw6&X@q(feF#~93EQJJ_`SIY$8Xrs0{wiD7j@e76zII<*gmR-J@>9og_e zu`fSZro~-*zC`1@2T^mBG>_|(K|Cw9aOUk?Hhb}D;kTwess74I@4!iL>J5Q8fiYyO z#xUOKa2s>H`}0ek7x9UbEzaNWhcj2^z~U=r*i+~U1t!IUTk;rQ_URc6G%ccQJ=ZwrK6x!g@0CMLBF%r z-#cFzJAOU%S|#fcT$e`ND%1tT;m_ESwF4d0OOI9<4m*O&V|QW8usbCBh!wxL-+_sO zPR04-i`ZI&FixWT@u|a1fY{VP=<^n66AQ&rS67SIiWTUCsb`KkwcRv+at8iCbIA|2Sc&-duUe53&X%E?`J`36g z3=+aWdf<3RHxZX)iVuEl0^|9+QEj1y#G5t8UB#!@G^kXN!Kczrkn8Nqa3K`6m4#HtV75; zHy;AEf7v_zY`}MFa`0b!3;Z{s8w0=jvUaCa%+X4VGQ7fW#^9*y9~oIZ2rPY4*oh=FP$*D_lrZTm z|1w%QlCclkJQexb5N`;((nyuuBIuuNbLtkL=jd~{7$Xmw@aCBX(DL^y2DEgON868* z_A`MzFII*g7}5&Av?^S}b(*T#Er zblND1DcZt>y(=(Cavn6yyTFR)dtl$o_n7C^39R#mrnH}q#^AZ_WYv_#E=&h_M(rcd$t)BQy8vp1}M9}H<31-N9$2d2JF zna3CC(upOi4u{PTk=C?+e7J=-U)OsQ+Yuu}@63zmWj}^-hYdZslJ+;UZ=E~sYdefp z{!`$ZK1X=bs>u!)8$V*hahu9kt(j2de+YwPYNQ;$1722G!E$qCU~>0;@k;bMj4_et zg*#SC^H(MCa*-H2b2C`*SHh>Si{^Zq2ls18=5Jvn2ft7nc_WdR2TbO>Z5Q&D{SJe_ zSi`)}-@|Z^^*kWJnERf|!UbPd=@y&i^uX(hSkiMeZ-3~CE2Y_stL;_xrC#b@DEeXb z@+3U{U_XAc*oOB9j{&38`(cdccs7#417RuO}IETRX;=N&d+L zR*`5fc~4$m-HO%s_v5#Iy1ZC@BwKsx4l9|O$HsaLq3Yk2v6t2j@Vzo4rs$a7g%Q6kQ&xJ&GFY~Av_+gx&RkNAY?3jTzk&-VnI~q?KbYticeW>=( zr!#j}U|f$!sJll->Yyb%{2e7v^CwNhO-JRpzmX1Kwki+lslC+fy$|n_or2~Sdtjnl zB98yHgdZJs8qb*a7pAOQF%MHTW#`~y$s0M3> z$a9~Wz^ks1yi;dU4fm5d>V6Pw2+~`~%-fXZG2V~mvR~D+6GDsh9KTt#ClsD+Q zdNlWlPbPzGrc38iG%i_u4KI9h#+<}@f}KP7s>grHj5*OhepIR6#B870TO9 z6u&DCfPvGTVTp`7U$~I)->w$i;r)AYRGb>Qy5TzH4m=9E4i!vY%B`P&_>JUhTi`;A z8)9GUrFeE3fdy;#LE(zCOs6JH*pOpDqL!``d|R%O(0UVbOS&9eU3QDyGny>U*fSbS z;`igP>Q=Ge*Bv+y1!mc}gFUVb;j5fF@uQ zI0j@V8|r)Os#bD6=AnYWHvids6C3#5gb!4z#S0C|>~_vB9O?I#jJ1=Y)AsJcxu>oZTf>MFK*-yJwDFe<=7nisIyuCmqGVBRhc%Ne z{Kb4vJZB^B?_j-3g3)?TfAQh0Cv5pMMJa!5h?|US2qYRYM_*}2a`X->*rLVvKJAZv z1S7igKW*;Q{1ii$|Hkd+jdOJCK<(X!FtHSZ@ANgVr9U^i}IO%$q%#52{dSKdBt5MmJ-xRf~ka*mxqV$tMd0VzGI)o4*2!f zJKTJt0Gj@c;J53|xRbQA4^YtITFF20gR>qTF))jlE0#)5)E_u9p#dive}u?$lwQ9y zO<49*o--dKeq)s#jIQ2+3#6j&5cO)IWUZIOqI-7aa_ct;@mNORS8j%Gm(%Rw6&t!~ z(E{{88c#pBxX@M|5%xT`!|j#v#3G~u#*S10ucCP{Ga(Q*wk45s^S8m?)gxi&v~<#A z{4sc&qAzu**R$|(FCqTyMYgB?50p7bozlX+sG9PUxeY913%3%<$JY$Av(-@Lv>`pL zI1p#s>fxI9lQ640j2OnnkgOGcLd6jq=D0={Z$BL&d{{pS!WzcGyU0S)z5Y*mjD<1u zsttwy)82}oBKxE2goR?)HEr^sqEc*-=D^RFsnZ-qJ-T<;0r7&@^Kb(YRi=YUeWV2uMcJ=X*Z9gFDC9j44D%vNksO?B8*F71*fMpe1yX%-?|4@Y$U z>BQqxVQX)Fx^{6m-kz#J7hJjyLz7(S%A6ArWg!E`vfo+y3rE&N>7J-6&24YvE;y4u z+~L>RR&p_-2TY7AW$q82lOBLCnR~mRmDL})Z2%AfdP7~`pyi`-F0|4 zx`v(dU&TZ_Q_>VsiC_P0XO*R~=#x|>%?H{=rH$Y)VqzruxM418X4W(BTUJaj)kHk1 zp-6w;3Ze0Rvgw#za*#2!p6OmYhPs_AiOq#*_&9wZyDBlN)NFU-e==%hruRw+$Wx`7 zJ5{OAe|Ez1L+xxtLoWL^T$NP)xyam?2BJmqQ@r?SgOFGi!D`n=GSkh;m?3qZU+Dxx zANhwcKzSK{v)2`W(02GU>?+tVs$=o8vBIgXx7f(OSD5#zdnC`nh1G7hWZ%D-!Aqk7 zOvQ}~2lkI-HHKLjw(Aj=3|J@pSdk>TEsL>-jll=EC~=X;mE_27=pD)z*;1OgjU*_)yq9B?!feg8~_o9}m%?54xQs6&y= zLOzUc+-py-drYCCA%nm*eVDSN6|`&>X!R6#u*!{xtCyC-7@mi(>(!7yzEobp

&l$kUc!+YS>F4c2NnuSEMLu%E|Kyg+Vy>DN{b1ec%4PmFI2as7ELrL)MD{M!IqL4bKmDJUwK-q*vu*hgAw>xIb$zth_a@(HYE0yP03P!L6 zX0BxT?Dge4r~!2=wV-$Ef~Z_<5vZRV0?QqA>5wIQbRjpO<11Wf2ptd0pG2^=hes3r z%-1k_aJ%>*;wWs=oJXPqWN4O2C>->Nhw#OEu(Wd{F^B%3?GuOXelxJ7sgUha>q&z6 zbl5s{8caVwjq%0M&h z&KpuKHQ;XVbk=L==&5;WtY1++8S;KOlfrr!!qs3<`!Kkm z5(2^1pM(+TWSDDhCtEB(o2?0S#(8H-pwFy47&3V~1pRm_2IagJ=MRc@_*u}&a!eXw zR$n=CP0N{+TaT5Zo~`&#RXWkV|&;?@@qX(gjvtS{V@_@h%Trr@asdsq{VB7XTJq`mq! z61~PAuQpYQk5?F>!kMXfKj{pcbauM9U0-rDwmUO5Y5&2BhU2;;2c_BIOQGS+3HI`2 zWQFB%HGDVeINQA^ll*!9hn+m>%Y>TGEcT8(?YKCMj`)xd)d|npl~r2sdFwCi>m^WnRS zFRXgL144?_;c)d=LQsc!R9$3?6=bN2&RZO19S3st>Gs#nhBKSpinJ|zI~ivzPc_?Y zu|9o=blrWilbTbzhm@)u06f9VJ1DZVj3UfkwV+= zJgE6C?XaH<__UCqjD9Gdx}ZmVd-S4XLL=Ggo*!ZUup4N;eu1ETB!wt0-ii9N9*83> z)WB=CHjTNJ1O@kNAv9FVTs23rO7m(mK;@Q?-;8!*p=z%)?q8W;7q8EOL$7(59*LKgQh;L$NIDEIjmYHh7S zE$OTXs;w24m%I@RzwHz%RrKJ~fZoh%j1Q`Idw|CE2%^Ue*qcaDwu6>B@7PRF3vta2}he!xYi?>{SiaR?ME-5bJH2{t~x_Y+ctq-7o$tY zFd8vO5r5t9rsMLIXsmMpRo|@5WIJO?(g#Y&ZfbuzhAue& z5r3~smGULQ^!gymmh97qR~JTutaUJSoyv#cJ^qqA_YBl*%VT!JAjjMJqiMzxJ>IeA z2kv}1h~(%@qeHuI;)PWzeEho(>@0i=y^Rge)cCv*rw6_#r9{!u*gS)JoY)PecGk36v&&xb>?b3ZD0|XRdRz3whq1@Ov#S-961LNK3tr%Cx9{w#k`n6bUt}A; zCqSj4#CJ}UElHY2S0k- zM2$!6Ife2=6uEltVeAaLN)Ai=n@fl9l4lrI_rv1>J*cnVEwTt9qUW z3#T@c%G_2=G;$?M@iS=9@E~N$+gOHlhv{+Tm{8tW4+ozq@cSiQ81Tx6#73mhZIYk7 zOx6T?F0aR1!;hoI%1>Zx=S8z4ze4VmTx@Rc4f=z3I^<4W$9!B=`BaTyUijP^Zpr3T zgY72ZERLY&<3>>HdkOTDTsicjpUAB;`HIZ`2f?9D>Hr=ppj#}L&>j;?5lI2wef)-e z?)4VVUF`k60rw7F$A)><0zDN9R~;u~ z$~GlOz3Sg|^)Vw_{AreOr^kCSw=&Uga@B8k^sfR=*)vRX8x+Cy=??U7h)8rNY~Uj& z7YWvXV(>+F1zcX&1V{Y!sY!r24R|VbEA}j4R{3*il*t@=t7kIpTq=1U|E_6#7IC2V za~gVmRHa372T}c#2I+j#muk3%L#KZwb#j#FrE@*X)8fMT>XZ_p=0$~g>EUK@o^)M& z4t6}RUXzG%*@AJ|Q?~Pk0Zs_LS~2yHD_J7#p_azl5}DVpgq||{#XBx@*ggY!^w{r4 z9M-i925)`nKzCp2@%c484H!?;WgVbC)f+Bc-cDr=R=}(7WH9qx3YiU_-0b-n;i>gK z)^tLI%L@jGhU2qw&DRWi<+=l{chaWy2cxL@DrK%`C(lElCt=k{$pgwnlD*G?G1E)KS43$_j}v#g#>`u{UR>w|K9z`9PRRyTk|U6jtZ3UeIE6QTc3 zb7-|$3HOGi&{4_L`IjFJg67||aIC$Al=XP${pt6t>m0}McX!zFuzcd4 z+yXalZKTIbj-ZbHI83w~z?z3G#GHi$7n$^dNGD}peyD-2*NB0HQXt9)B)*BX%lz5s z$Htk-a`hP*V!VC|6h4~>rz8h>gw_wXVu~F4E0oi7lTMP(mrC$hel=Eo^56w~1pJ%7 z67}7hF>CA|49b2gIZ%MV3h59gx7;PW4LqrbqCI_LW$Wm_tP5`6%iyWCpRum?HXI8m z2HSdnhzWHe3u^`1C3v9n&|5@ig9dhci(r+gjBm0p;P3(7aBaUInO(Pwc|1FdCG%x? zoVN}y^-W^~ekD5$cS@z7uG_%iww{!oHF3N?vVgvc+zQ9e$%2dPM9D4j3NwxDFfSvK ztFQBmE&}8Ya4M4bwcx-sTeWn8e5ba&dXf3gZxQNtjTF%&uf}c@!~tkbec)^ zw5y1%MGyY&Se_W&i;CN}T*kk3`tY~op&)L#D1>)k1)Fyk^k~Rm-27Yv(vDcdSL>Z{ z?88s!$~p+|lmFuC@e$Bt1xrmhIE2(bjI*`KJO*ofN>xdpCBji9ov(=PTxyjt4mtEp%<@ zLm!_hp&1HS=sL+y{EVqM_PEqZ)$i%jv$rCs^Q9PiXtFv#)|DaggJhVnpbDMK$Mcy9 zlj$H$W4b~86nY&=BWpvpa`|yHK}#jh{?N=cx@mAC@t5sOzg;?nqlPz>nFlJ;jwgyV zGa-RA2yXEE$3oh7xCPH@^uXmF?(A=mAT;TB5I02PmixV_XDSGap!%!HEHLdEV#D8mAd^2s_?Niz!y;s(!BR1 z8V!BL#2G56I;{?l9H@lwoBlK=p%M}z^0C1~krv2OM~%2?jw5DyIF1EgsX<{lTyC!XDgW*HHFp1IsC##CxZH1#6{s^wZtRbmPoKIxp=E6-)2X-d%&K z!>~jsqdV|jWIS86p3?>r3(JnGu{PD-RFHRoek+Z5yyqsav?L0Qg;8L*Vd(W_xK}(0dz)3@mRYl;9OFxzwe=tP1ZTnISYLKC)0nF7E~OT? zn&}ijcZj)v1T~{Z@UI!&xHR`M^Xf|CE6(+#F5i3d%~`InXsR*C9!^|Iz7yRZi`e)o z6tmnO!pGy=NbJ#`bfUz?{4^*XZJl)BwO~v=s$FSj?8SA*2KCf`T zs~5)2T_9*|$#vM_>j;%?-*AfUeijt@iwE}~#n<+*mwHn%V7lr(+}JdLybDp6oYs2y zwn~j(kQlfh%fq0}MsiK<`i-t9mxGLVC;Rt(H0j&#G~P;1<9nAc!tm@V#Nw|O9!xlhLF&A`gFGPO)}y9R{}GV z(f-aRVaR|dqJ!ghYzb_I)*;R~Dny36_-#Q)sUuKc{1tBYR^}6KEI`9O+vpDK4pH{~ z2=qHR5Jx{?|yK`v>V6D>_R718)2tcl(5s@0~@s?=mhm&5SG^x5j(dtcczq2d=ovs{9zdmo86I-dle&Pi=w-eOtTdV`J`cBGk0R>C`Cs(e-)-Yj zrg|i7l6D*howj_4mm69gH|KF0(P%Ki0sdV129|GQ1f&T>VQODuo7RWAPTK^oMKeg- z^b6?x-iHNVG2p=t`uuuJ6vMSc=o)oG%=_30Jr|FrZO2PMJ6+0zYZA+{rMEtI37M%mq)LX-e`jo=#Zll)a8^4JO5jsUv3ZQ zVUumCb!vW&IToA!`5=0E|1OAP(=~DA}`c}IX^YmQt zaElsOr;}-M)iMZPEDy!sXVQ~nCepa>FfA<+X8zI2U3|4Y9}Oep*y`C`V$Tt~(J)Kkdmfj< z+*EzB>qQdWi#CA&WPXahYM-$4XHT-fIs33@@dMYckvFnC@7xm%~MV!QPZiubfnaWJa{4p zFWAuCQc>YY4CHvA;Ue;Uk1BsgIsdvj70*@PB>OIyLGY47_SQ|6?l&-k zzl(=rzU_I^a_t1X3*Nz9FWYg263N-Mt()l2UWl8cpOMyV1wP`lpRcOsT2}q&LkQ;C1`TwDc_;LU)sya6mCWde85m?} z(lbWRwB~LO+w^A}Q@N@JRg;bRMf11#!#IJ)gs*|e^S{9RWiF0~75>qa`Fr5q)-u@p zQE66=x3mpSz!uK{)_UdFZZQHVq3SaMl!v1mS8InqM)3Pz{UpdT3 zH{qkw=dg83W_)^~BW*74gfHo8eDfv+RM~S8PoC`}4OV*K-QmXm419yBG0(+l(u%R? z_ru2}L;3ze5;NVjA6wy7gIQhkaH(k#WY?yn*Zn%YG}H}$KR(R96)CV+@4|%qU*7n5 zyu>6KV*sl+se)4|W!_~HXYct^X7g!{U?W{e<*`mkf+bn5!4w=wJ zDt`Ul&gR{8WLchx;P)&BBV$XLNAX^X1F1$PRsDcl4TCXp{3X0J!hruLxu${~cY^&i zX+Lpd3cvB*l>04CkarbXcUvrmZE>(%tb z2SvJBrw$FKwc@3q?_>`1f!&q5mHnUet^82>heynmzRy1dk{094buuE+v`UXm4*kZi zPVb9&EK1C!*U2(9LN%XmgF|T&W7xDFCY)XlBNAo#wLj?cNYVyezx!bJFL zuR@0e#(~aJ1-h@gzqsp2fqOB7gi zmlA*2ti^jJIAQ%N6FPZDPv}(i=8DsvVc&6ybcJj)j`qI7&z31y#yr%moQ8LKvX3s` zy)K__|EfhduCEh9q~E#!s6Du1Zalv1^@J(36|iqh`r^*k@jRf>08_;0IOxXNgwmXR-f=dt$wT9O`Et z#GtOXsN5aEf*JzI@6n4fE;1Ga3|?UOVt2B-tq*%SE{#UZ=`g3-znDI-0OOZf)7xLO z(c1Ap?rm;~_4;10vX40r)Nf}X>;~0iVen=t;UnTU^5Xr9)b!zQs-`gr>Z){j;8AZ< ztyUl`7~+nbqt;=@@JzPX#fCVU+3-zEKa*>{+AwX!Ote@x41VaO5#@qJ*3n-J3dR9k z-*ktaT;2)IF>_#g$asG3MybQmL>0c|y}&JZafnMsxcpb0?@CJLH3_Sz)oLrcB7F+v z6}pMGTW>h*U+jp#Om^b~e>a@mJ%jn2dk3#?k8s55u|C`CQk3G#5@i$FC1fc=)6vlTCMIe?S*sA7P(UJj`7-2H(e=M~5$f3%ieVT^oO%Wo`{M zUwmjlVv5vz(cvE@PE50b1wZ%Akhfm0qJ6TYUe>q}9=^d9PaWS)pDH+jTz{z>I?Djh z9J(a2K%BV8`x&I+cQ9|7poMc%Q|OqWdvGM*o_~*dOaj`UV$R;H%;orR@u2m$3iAnr z>DOHAb(4JS?4l5m)>=iyx~j;fJ;ry?FE?(aqQ^ymN}e4BtYu|K$bZ(*)c) zP)TqJxIz|HY$xoAKFHY)$ERQG*sq=0QtrJSYn%3wqdB_lfnGB)o_&xF$#%dI1#ejK zdM)g4Xi7acR+1CDOvTlg5?QatEckS=A2Yr>h@Gx9gbyRT$^En^tgiStR|+%c4~E-w z)WU=LcGABn~$OF*%^9xjl-`1Nrt|6~)$<>pr6&|r7I z>x>r`nT{os>?h&inf*vzpci?(#R{M4o@X<5k75hdo{_Pwamd4zh)#;+G5S#?c_4#O zVc#JkBkj5PCLobKOdE?ccr06JCx?f;3*ljm#N?}A#T-T|Ky%m)v0wkrMoo!>2S|Qc_yjYE0&hyy6COeFi7#wClJi+VgT(+-t0N-?O8OY4ChFpokG|lfbW~ww|r}hwX z&w41lt> zp;xoYNV2z~#Fs9_-FaiNB{v?`{M)cz;vj8QJdfHpiv-(i8`&3f7Pd*>l7^^ZV(X3u z=v3BYPb=zi@-73ozw{m3`!R@Rh9-e$xgL#G8wKO)&*MDb-dGS|hFbS?F-lbAdS#Ac z=Whq{g=M1runVwh5FmcsmGa?m)Ax4K7 zSG9rVADblR(?Rmc=l~mCy$dGYDuN?fe$+mA0loCM25#mBkgm!*Alu9xf^Rj6T^a~0 zZ7sm?xGi4jvyshza1qX*+=?CX6WNA8+FU>APlKYT{nW^d9tq ziumoYza$IxS4@DH)~E5@{)@cAe+|ts?xI@#4$}FF>3Hi=A&p-GBy_qwgrB=2@%gQB z>bNW7>whL>rpp=(N!tQbl2Xz3r5xY;I}~z;pLWogug5$iI#JarkFGtpTkL)54gA@V z0_}}A$%|fnX~gwm80fCWlFTh=oW2Ol{_PgkjnmjN`9?4#!|34BtAgL)B^6&C)8QnJ zl-QlxD7Q|Y$cI;nn+Kg??vf`ZFWr)!Jn%`cd4&bY- zghiU()ZeX#Xs}kDyt;W5t?pi=MLvzR?|^tZae1eh6)sDiXWyl!ljI%M=n=~1X;YSS zij3M@M;3026sXi^Y#P=F&fMJ1RPW_O;f5lTBkW|;Hy(zKQ-g@}Sbgqf^pk0IX+!9I z6RLhWitg&ta-0)iO$%xZVDjtr(EPAd{I8@I*T{_E*F!s4>yG`z;9hTf!?%*Q#eX9k zW*uhdx{_4vxkHLswJ=C|6b<}!fgG5nfh&8rVw!_J&v1Q;4&HquDgxP6%u*Xlu9=Kx;-CgcZW{0y+lK{tt2~s)v!RPDKsP^o@|ot z^=p04Nd2UHXa=7x zT}c`!K&X>4^tq)7_IcJYbykGXbaWJ$xA&)VcT(X=O&4id`d)naAr5Y8DI$5W2kLEq z3Wd8jVY$r$a&G@e;yl$I*N=Wjug7|k!EJ-ttEqe7Zb=R~2K~uiu>|g}u7+Ku`7lLt zcgg?VA&!~#sloxhHMKL>N@-P#KZTu#86KzZmRj)$fG4dRa3&amL)Y#4KX4f%olEvn`8Sp*v z64;mIlFcPmr1$)j4r?5n@$Uv#SU!Q!*>E4uOtBuMEf|p@@i8x|;l8nhv7(0!8PuZ;(pR*TaPP-#kFms5h#WgMIkdnKT*;)cy@HUyGk1r?Omyw=FrW5tNTH@OM zP>|nph2$+*2yQ!V$?XGOqUw`uQhz6p#mpWLEyoXnPkf8mW|0S>J2PowcP{m+FTj3Z zPe54UO90c2utHstDjD2|fsTQ2_6m|2))F^%>usXu)d_arePP_$i{!YHE6IJ_Oe$?n z#6KFph){PP{eSLZI{r5tMnrm2@3l#^Nn-27uDFA;H9T;1&e(JQl{CZ?tgb`3(e8U->kSx$I(-j&W1vf$8-yR1dZ1w7V~#A3ZaiLLr! zV!@VypuaDWPWYkEy;7smF(S~xdUh=RHOGYJ+|MSyX8P2}_z)cB@#qrg&TKn0aAuee z-kYE6P%UNMZJR2@!!wO(jOoydPjCnJ*8YLb9~DsX?}LgUx#Prc&L#T3sEX>I8HRGa z1S7A>z?{Fc+3Q18N*b&pGbHBfo6@mj`-dYHonB33%5y8sn6ic>ciW(aj|_Z`DPdNU zpHV-@g2`1G(R)Mx37U2Dn2Rzx{C*gXXHDBgoMccjrQVY2`&;m^ap&-1RuegSHKih^ zuL<0BG@yM}-++yZ{l%&2Hz4ujI!JgPM9Q4{fxYG>s-!PNZSp`V*E<@7P!w8{_1 zrR{=o*Q7kwFd4+P(i>@c4?40u4S(a`^7}JwK(!s=@RYZ-_Qy`@b~q6yjZJ4Crrd_k z?R88;aXrdw94236PQx*s^U&T$k*=1x1=rH_g|`<5v1f0s(c*SKG!1Ge{g)L97bneP zCjH*wEzM|jbFE{;{fyyY{93%1nM!wg5!$L5L{A)8NPqOd>M&(RJ5G$)0-bH{ST|cF zZnA5@>h)Zhs8;m4!jAf8VNXI|1`Ed z;5$5#ZpW6^9?OZdwTN|&6rHcA0P9y)GR5wP%SlI4&i_9qMJN@#>B6 zxMMH9`9&R!Zr&r$CrYDRJtLn==YnGJ0-_YG4gZWA$w|iufKi3e*lkRgCU_IAfA-{5 zOB>tu@iW`GRu4bC795%3Vm@&}xn%h`9lrU81KhV$hS1An>3GPX)mNrKjJh1|oG}Cw zJUSpxMw)gC*;!rl0(c*B8vYG)`7(3#tn<-!Xx z?Re1IXp%Hjn^w(Hf-Ze2zW8M-pFUlMKmM0aCqI&*<9CO`=8KP^afyyd+vzJFs=bNv z-c|5$bu?`_tqcVY2WWQh7j}z0B#YkPlN{S|2Bt(Ff?`8`Hh|f&v-UakQs-{?b0VIc z@?3~%pNDhBwvoK9Yz5!D$^z4FI)j$YQd)682bQiJP7AY*_|ctvXpFWV^n2q5KV&ac zNs$lrHaUw|wk+fy@?`npF9!U;q*gA{Qz{P_X^h_c-;=qz7N|ZQ=>F@JE}3=-9RcVe znKSURE*-_Iz6y63QkXr=0=B=dDtx#d|)_&yBJKH*#_&+r!`~FhuzP=3q ztlmeg_i_BMrvTiwqhYgW36n3(z|rDgxG?A{QPLqpj?>}Ti;I;Kc0%f^iAflM72M$8p9!G z**@xhR7zm@J(Wy+W<`rn7So$|SJI%z%`oHV0N6J`kFRq|r$ru$;}N*9R-noG`pTG!)lISm9dVAv9!CJ6tb5KxLBC z=;_95uv9$*icFK(jAyw5x6ZrNd14zx?AGJPeKnHby8u<+Zli-n8M92Mg{0g6F8ftz z3OD>>U~|@Jm~L)>`(&30UB}0`e@`W7Ow5C#dDBr9WC_()!`mKtsQ%a!bi!}KdgW6X z7qwCp5i5(TnG*D<_(t~V>(N(Qt3kz}M3N)$ALo0=@hS1I@m%I7aPhAeIQG><;q#fcn9;YQ@%+8jQNsSc6nUFF!BrVs zw#F%+Ne&IiS=;}kK6``6&ekC4e@Bn_Z+?JZ1;*jO*mTR_IrC`XF#{Z&X+UMY-s0~k zr*MUo4R)BnAQqeFgXiz7EXXzle@CdGbbTI`ePu74Aq~*RJr+M5RzS%{ZN6;lAd!nx zg=CGuQPO-qiToU(i)XvsQDwU_b2A{cHZhafe)$Vg;#k;l|0>h5G3IqC$;>}AfF7IT zi8H>;q3bE38{0YiZ1s%jp6O5JEsw)&cOz8MYD9S=iwEyTh@xbU5=R#)c1iec#+M2) zY0DAV(dkqc{#uW1UM@vQ=Lm_j9VGC{M z(B-lMJE>|F?(+OcWdo#7IaD?p%am0& zp_h0!8a*5UxlbO8yoEQYQLXkcVYfP*kMBcAqu7^3@Jmizhei z>VcMZubI8YCwMkHj+&&Fi)Ls>(%z|#G%x%)*eE*F;YI>;Uge!AHbIu}o-M-Q)$8c0 z<8Ej%V;H@Z_y{IV-z<4pe~eVxm4ge6f$x*Fv7|$j1&(ZIkN-_3qZ0~At%*GrUN%PU zEk}fR{cj+ZJ(j*u5_}{EyXdv_iDdYcZRq+e0_Fu35g!}DL%z05aJ`*}rJI)GeM|;< zhY+wy8bmkQcv0;M-Zb`KFOmL!4xZ_br#5-B#8_UaF%JpGOYPe{FUMoKR@(WXL z715!|;dIvBZa8qNOgK~LaPuu{+$dokuid;0w2u9PKs8mmU3EH*ndnH1H+r(66QgNK z_BEKd2ElLow@#-k`%IuA;WWF4X1KK@M>5v7 z8Tah-qbnb65c0o`tTgv3mQQv<#rP#`j(;z@+=h zHXl=Dt?2UTPU3%V23~yV4_jgs;OzcOBUG!imWMJ9A^ zlqt=0>n8U3Q%U!aWx{zn*y`nWZ4z9)h4gzYF#X;S;0gUjaFE@Hxv$m{*QAs1EFPd@ zbRAm$OQ3fLOX0|yz4VwyBGq3I$$xv?=7Igg_-xZdu-{^W_?*v0db_Vs`0h?q#cT3n znc`t$-@Y$&fBYfweA(@5MVQj z{Zd&DwL8fx(;C%n`DY=Psz!_!nfJgCkN84E(nP0el=YL$niUicboJmyM(q|1L6OxOWuFBTs*N3J>Q?kvu_tbazHM8OMVT- zomRA7<`QZK=hCt1>9l8Ysz`Nh4z1g(D$4)eLdRT8p{vHt=2mC2*|5;rxM}!pw#ikU z$xk>%m-ItgcP^4PNjIX>iAr?O%_04}&oie~H6EDrmdsX6$FNNb{JO^y(E;6mq*Z4; z?zyjtv0*g`t48w?bJo#sZkg2IbQbOQ38CJ3i}=FzKXJ*YP%yDLK?kJXfrp1`=vrG( z+@z$%Y-$GMO-*NZU|t&Dym^&ve`d`NH_Dq z=;)yHR{KLZmref7&-``5UTtf*wsaQNYuYTZP;=SD+9qyXwU3V z=_2))yZEz7FEGw0oy?5tW&K+xK#$CQ_T^0z4cbF#Xs;9Xee#S>k935!kF2?Bt7-Xx z-3H~$QBognM>J|Nj0noy@QoKIw!mdRWRGX z<6)@WV@!`+z(?-w=2n|eGQ$fQxF|aSN1j-YK1H=Gw)iru%o1EIZ_RMVC{-4`P=sxj z>O>>iSaj>sHTFi?kbQq3iypBb*oe!aq{rWeq<>?Qps{Nq@z^SIDn1xTWkvBEqZa)3 zT8dd|jKC4uTP3R|uH@^_-{a567GcX(1*%YU2Mtz_L~EjmKkKY`$7pqGsMm~FQ#gi{ zjG`Ke3D~@65+8c}Jb&%ij}7!Mg|ym%V5r-Ti9?Do`cwf#%>4weMl$em{XnRcQH7h^ z2GL)KAF-v4GUBIAa^myNl&Yng@w&ST)W09c_B*chNOTKb{COh{pDCOz{iLw;xCVSG z91hctZ6mAVYAs_c)Ult>AJSbT@P-8DU)yqRwlLrt26zwVE7n-^UqO@U()p1#xH6`g{%Whl z`)d|r+dT_RQOP1-Ln%ZbkAe0{f6&Nu8awnRjDG8Lpknu7eDAYZdVK$MI0!>AdYK#x zxABHV(JM(j{;~X9KLAI`irL9qlr?6tYae3Jl&fmL8S`>3jU) z(bml((SyUR{#`Abp1B#5@7{$A-)D=L(En(yxf)$vJW2ffdpSl-Z$wMY48H6_GZxQC zr6Xzpp9_7YL*h@kL~!4>KGlQC5xsCW-x(y5aAIO})G{t>+#J+xBHFKURx`}gA}7cAhBkVAdA@fJkM>(ZaS)%4!D&HTiK0*uoS z#16QLm0Rp+osuKMl}7$VI+4rsUG!~d5}r1y!CN-6{FtZ}wchUG50YnMUF;2!`1>b_ z{M-vC1`X!c2kJ4t?KE*P7Px?~n_0-?R@T3C2Ac9)=qPgj;Y^S(#=qGmZqfNWe*lT z_JKrxcej<#gBOx$;f?$BCkM92Emg8q zHITg%CJLC6#S}9_N#noMqMrtw)h@9o)2m8|?P1}bbo&&W(JOR(lFcPgl)cbU>j`=J zHVxj4ddSQLzG9ZqS-~;4RJ8ZJ1oTHq)5sqREarqG*q=#)>yPh}Rl|nUF9rpW-XqKj z#~;S0b4r-)m{X!>uS9gnoxRlX!bW;FAei1iAx#fidD9@5>-gm;1)0JGa#Nho!Mxt8$2!lk|@R_m)j99iBhOGgzPcgmJ zwq}NqgEVF#A^I>lcRZFo@+7^t#^5cviF}u&9kcSfz;o|hkY5!B_s-9?8moAPX=`*! z_8l7vF&`;OIU)tC{z^edwl=sl=D>&1N-%SU8yjg5B$|0-6tzhV#{E4DHT=MNxk z@<&*v(!z2QhDlx}5b$Q_Am-dlV7>xd=logb{4<$3-;fqJ?rEfhE?S6dUnq$0uC^4f zd-jnQSnd+l@4v-94ZO$#qx*|q4$dGgb_SA%1%a^Ez>XgOzLj2SZO5xJ#l*9#3FOZT z+<~+O&@;Xl2AH1&%`|oHnd!%+jJ2s;+gVZdtP>zUqD05V=z^lM3Z30^2R6(5Np`Cd z=+n+9%ZnA>`c|3XEU#^#Jvs$!{!YXtagl6cDOgt6M`3v37szRu3-q1^EHI8`F&~b> zio|^aw`&~C6W08O{g)uQxejg&S&vg5*o3S};vE1*1U z2gHZJEPMORl#K0I%~BF&m|bNm4A+#S>65y!@!>n{vyQ_}p{F-1a{3!FS{+BK_bKDS@8RK{MoP!~aBWL5~p`UxOXj8*CR&rL1TXWZ= z^DWQPsRK)hYJ09|k;zBm{re*A70J<`vF_yaVp-arJ%t+_c_A9^IE#L%|Hy6|y0EAl z1F3A(0)F4knU8UZgGt}kN-TbiL$zji$@U{kEcC#7yq&(7rWK{qtri+IsQNN})~}Kl zRv)3WU&d31=Zk5(z&>zYRYH9i7Q*lV9qzn5n;bcC4YmzD#7f*W@KcF7*jj6gzG)jX zhpYl}vt=1rP6`C27JZ4KYC7I@n2Q6?2b0yYUtok?BEiuE3EgxQ#(TR_54Epg{a^w; z`NxRXpB_yEcUgjSRv8iSQtZRns}S!$lGHwxg0dnFVYV9wMYfmW#cyeRC#6BHeI~LM zqlE5x)k_vRuE+@cS$6q4t79(u={)WBXatShu zE~2+WD>MGQ7M*{+C5t4#t#XxiL*o4$_Pvr3*>7)1>PTH2Wfjb{LoSjkh3}#vMxLy1 z|4kUOf#OdhO>0f}cO z+?|jQUAFJw%(!G!2r}g<%lgwD)27qRv{iJr+$&h`6hkkcj;8gdSupgVAO6|<1hb`9 zi>yu!;wPK3C7xMUL}s!ks|}8XC0~m~?pJ;?Ew>G#!VrO5x~xB27JCsYE(AfI#TXvD zd@*^{8_wS4&t=LHlgms`DUeGYv&bUL3s%RKO4y9WL$N@AkL3B|2Da{2r$nVZpe)j* zmF4w&AsImf*I`~!t+dN?)&Qw-|(pGD9h`TSk z?Ggm*mklIO0w02Z@f}H%hX-zWI-b29dzO6L-%e^(9?`baI{K*N0NrNr4%ZY&Fx%)p z+fcIOxF9gijVj!U9* z!z3r)M*CONDbBKHBJIU;~Su{9|&?`F+(qYG+g39#KblRVGyleUnzKO@u!gVI3 zWX@YGiM<60#RV*7&H)%%HVh3(e;li7gL;Ds$OdaIxN1-Yv;CFD1NTyKSADcNtbUKU z-_JlgF{KuqoP@hWTa4hxYGMHw`_ujneT?Xw6c|>|$=kXXa%pW$S^SXkaN+7tW@Bna zpY51Re``>w5)u8?x(o&t2H?MkL*az=FbsU|1;5Nj z(8+ER+JE~ia!WlMbSpxMV!AXrw(ll_l{$5urox+-J%mcjb{OrTOxHU)(n|_a^o`4L za69_|PIfBO9qXzf_TU59q?kw^8Hd4zA<9H3EW^P0li}S;bvWm_OZa96^3}IYX;+^k zd24%}T)Ms)Lk2F#ueXIhwB2HQY1csreSR3uWPSlJ+Y=}<+(&hb%UHXiA85sfK;O*^ z&}O}w$l5Eym_3t__pb*!@dT_LQUS?qE@ZTxhNcpAHqUigS>#b`y8qrNy3yT^HoqQ1 zbAzquq0`dPyTS@)eW(z8(sQVyBpGUsN8tPeg6FAxIGyUYjP~pNOjnz~p{}O8sPk@F zTtwjvQu9FG*gQwtX_I+QCu6>(BL>+(_Y*e@Hz72`%rKCZbd z>0dhwe(u-?6TJ(8&bVEsFlI8Gzae;`{VD|S_i^emtN|a&8^X@$GJ4kW0sZ$VjsB{= zD6GkSuskS~-{0y-=RYo_UNV!ZDwtBcE$7LKhk}>W=CEjdzfVM#JdxxLy2ECB)IxA( zHmtlmhYpcDO4*)V$oW2&UR{(6AAb6=X+GwVcAyWAIGunuKc|C`L=qKzXK353g5%~2 z>@1^TXX z9Mwl0zh_8(L^}%+K6$!A)s{|D_oruX$I;cJPEm~=F?5U24|sOYifXOeNC&M_6!%^i z)FtA9RLUt93NAU}+s=G=va1Y_Yn(*$`NP0)lPU@Pt4J4)dJh^mN=Q(C17yVovE`R7 zs8WX|HNR|0S0*IU`kJA1Zv17CKW$B|_PWtE3DRQwHeaeLn?_aBdci*-j@%gbniLzC zQ%e;u`f1ZsHsWcsWLNJGa9makWMwss^V&)qm=XPMG#y{9%c0*Vo6r@f=TMgI4d%Xg z;pf8RY{;!y^y8ao@LrG$rk>~F_ErV@UxqUzZmonY!Lu$Mb`j#F@@dMXX|&V-FZh4n z2ltGALHLKERKNTe=)7(rvwl1yPQweqci%!`ugwLA+1FXmw{mK0BPVLPW=Sg_E~X#a zfHw95p3ZY+7e-9M?6m*jdfxxp?+K!eo>5fJaSna)MDY7+3;Xr7P|z{21U~8+i3ugzEVu?hh1bAg4=VzQ9H1}ojdIChkU4b2b@c2&T}r?sfHH4kdkri$kGSELJ;Eg#615vX~2Or#@~fM8Zb%6bGRqPSf& ze*Y%=M=OOkt+$|WZ8GS)_Y%7HjU_c2IG=b$goAq92D)={D0g^mME@3kBUQmzJ7wu$hxOO>y186foQ%;+SSXlSa=6xP=z#AnJB^!gc1zMrXp z+d5s$W8E^C<(32ot?t3LD}{7!Nf_Gd4Ws8Wf04Aw-rT8hAnLTfg)6n2sBOe}deE(c zY)C7I&$s(gw|R@fdEr#L;ofFA*kukmw~Ik_v=21DH~<>E>tJ0(y~J5-Cb+hKV*UR- zm7F*m!`>}>E2>L&P^0z$B=^Ms2cd3b+ zUz`G){dR)iCXB{RdP%?RPXj||AtMlY1I`4dQ&HqG)|jD)@_X+oy~Y)PgC|~ak2{kzAXiG6q+Tke!3Bb(R$z$Is)c?Pls)tRgxbQL)f%eiEv525O!|43Wg6G ziHC6$6Y#twMl}Ft#y$E7(r3fiddJv{F}azourwgb zVrEx@_m?hMdA16#{Y@i{v)6(7FENezbcwd7E7N`<{&b|z8#vimNWe`cml2Ur9f>2hcOPRJgOdfXubYAX&SB(vzt~R=0>)+9%Sz zMz-Y94IePOwh+Wxf55+KHXI8-409XjQ~HaZXx2N91&*64wlMP-)a!xbb!*;; ze>i)J-+d_(>q}LG;h82lzNG_BAJ36Y3cEvJZSN$-&ll1)C7SesdNVyOzZCp;D$?;? z8!>I~4T#+0NZo^efOD8G)!Hw^r6wuUHn}~J)ldYRBa30pA5R)MTAya@m`cL){*v*y z1oiYUqg}uve)ZWo+@x1c)=yjj6P=EegyC0M_)uvmSQ~`96~|(*`djixW;Rs3>A-=X zM7X@Cl4zJqk-vBDv)txnNm*G6)NWHEbL>A8-RhB&l|{FR>i2%4D{taJ!*2-udOlS0 zy{}F5VU;eGnJ@`*@(+`Go(-u(w&KWt0a!5psl?{va~Ql=mTvlR7!LclgR8*Bn;&S; z`rfDzw>=B+mQ)85T@NU`*%c}Z^^GEB_bjQsgDb=s{KH5MSCXEv9<4r0V5hAL3u!ok zF)cmVrm&UI@@vPT)yd4oZY8_n^n(>JkInw z%XsNj_UoiJH7G3=eGtX)z-|fuFUo|gs2*iiAJ+2TQ#$2Sy1V$bHye1n-*ytn+m53DJz#z% zt)QuJLEs~}^SYNSxX&_YlsT7$es-6kp-)EW15Mz!g&f0mojF`%Z4g!4>PwQ|x${+) zSMhfL9KLekG_E>oCfD|MBU|^X;huqO@N+*s{L%Icbqh|Qc#s;6I`SOEX+v;g;(wxb zbBDtA`T$y2D>&esldO)H49_l5Ke|ka!`NL5+@K%*MTdh2RjX9!5|D1QBCD=qH zSM8`p%LK{Q2vg`)Sx6lU*N|BQw!)LvXlQgy!YSR=?EapeJk=+emkPOu>xaHI>ef08j_1K71m23+%23I=G)^J~h(z~^)jpCZoXFiX1Jvf>E$ zDpI34?$V^uCmqKmHe&HP1J>t}3WEnkN-l8&_QIe)n6Ai^LT^`fKb}tzf3y1m0YAfCoEF_)7y- zzGzJdukGp2V*}*q?DU7QV)q5M{O&KoKXn|>mOPMjjEQ0X1B^(P$uW5U*Nf_#ZlZQ^ zGVs_wn`orzus9q=uM7_mWjlppTlZh$x5k}&Q^A5QZQ5aahT`7Ab6%eimg0s z&YhN(;*teJ`5UJk9Bkr)W%I9#eCq_x@0(uaUV8AvuMmGYM4(4WAIcAWiFR+qd_`OX znzpEuqpDKa&-FFw-Y&YxL^7?8x~M)GtY zQ#Q8$bTlnDVgp+R*RA0u=$(~BUOz*tqn&%nktL&{#J87(sHi~T0)NT9hX%xUqa7(9 z9}S(;3P_~g6tV|&$R3@cR>l1$kVT2NB~=qv1Dlk;52Bxa$kk*O`k94F^s3%t zWKJZfb&tuK$aT0$Cx-_wX~&h(k41z1BdyBMRgtmtb4Y%XwCJY1N13&5H#>af3TwG8 z<~FfyxF@9(~X+LwpHj_`N#p)xL08Dv5%wpkSC+)j64|#$_R&lw|e1$pA*kmp3j%K z7DBn&BXZeQnJTyn9OZeDBp)fR?f(*uUQVP1-J>ZKTGQ6_=fq&KJD0psn5Pa^_MELvZK>X*!)8pkm8wpjMKD^UC6} zHY)`SCqzm%zsbZkY78H43CE=xmH26>3+j(3f!)6!2={@l+&upcKj&r37ko`8-!^ID zwTU-LPP#VV7!|?S-BW_Z0 zJn~&YdB^5k{Nj2?KKn%|eB78%g5&n!X}MoaR4&iYq>QKXh2PO{_b@(j@O89H?azPo z3OVzWze)V+7I^x2B&3u1d_ey|{@~CM?sk3vMo4VveS?MYVY@8fwPYo1-twGTORc4C zE7h3Iws3qkHvmM3WN4JO!0Ws!M>|&y5XEf_lT2q{U{mxx(Dw2H!|D&9@Fz!9o^^>F zFA>Uk2j8*cBROP=XgGweF~GGNugTH@st|g51Q{=-4kb2Wtiwr-t#GQsLz%z8v-Bi6 zAJUIVHnfP8_h}NR0(;$olcXtNm*{SNKkBWV5A&k#LF=^3_~~8- z9(QkGPHY=)vJE8*1Ad6s*qkBV?E|18!PPm4;ahKoM^C3OAL1Bs*YEm?L!Ut=})JYj6an?DB=g%6ZVmciB+VaVsp+m4uPMh1^cnlftcd=CF z6)PBQ#7m~m;K2`F!SL^TSm*T|PaM>O6}znQT=0ITJG=<}^Dc@C1^<9<^YYT?pT>e` z7bnA8=8+4MrDeA~1!q>`VeCwu%~vT;$B-nF)rC=h}pX{iTT=AHaNo)PV08E)eFue z9k0U+3PSL)Su&=iu9djyxR4VwX48BeJ>FKcx_^Vf zRdpN4_r#M4QM=JY@{QE#FQeUU`^bcYdhkFs1nT%OiIc@jycE|1F*RDeWy2&c)tOmB(8{pN9Ao4Lj1rI$P}xFQmIGAY*?HOqq6o>)3`#PU>{B zCuSGe;cGFt{=;a#pr!+Liqr9&+gc)>_mo|%x@LLev8u@5b~-fr8Gzxs{#*pw^sw0q z=3kV}QcI%Qb}tZm3?e?uVJdXiOL0#FL;N(&3C7w^X#PABukC-yhOLPs#kJ0$ zd^JKeByk}>)G6jEhRxh8Ie{PUNZ=+pM{(2GA9#MqAt+h#6s~X4fVAKJxb7i&2;Ej8 z+I4LZd^7tAdB#ytloSU2>(!uX@@dPig;t{feoRKSK0Pk;)*mbWHVJ#!McCx!4o(mR zhw^=y(~R?E-QyQ*j!`cuuxz9yS7hL4{tcdfb~c|YqsLW)xK{BfHU>rsp2uj(*m-$ORiOs6uAdTkAc+;E6zmH@KXvZ;*y}Ze1AbhF$y;jQ7-Du+|$j;O6^)KgFupxOwKN=y&+6bu zq1W5$X+k&Ni$GOvk&t87LI>3a0w4PoUEXsXMvluQ%052auPh1oe3jZ|Lp|IPz?aMuWQEVD<9z-eVeMY3$4`ZHv00h&a9P!aeR-fmuBWMsk`KNXc{*QWS7W?c z@v0Z>)R!K1bMrDHb;ld#oy^9N{4K1s(1btSl7ZG<+88>&7z>;9Aa#BZ8+Pd(S@Oh( z7L>jv)b$bw4HF3R4aI4jh0ffz7)~C$C&WJU@+xq?zZy;9Ne- z*eSe`)0r@1_7hmKmEB2(Hrks9}$+PG0Z~#pSUT&%?{MtXNL+ zgx;Ih$7`bDHK(yEc@H~bArH;xl31qkBj(aEk#u>S7d4FYgwpaLTod3)Z{@_% z$y%@Ykg8PPV&%#k8R062J7ra{RXpi9f4n7qlc~X$diuVb3umk33@O2=M zZM=u!{X=m5;byibz=9-o_Q9EZr-;wAg*fneDce{wjrl*$r1`rpQC8>)zk^cv9lZct zz4s#Q|FaT{*UUuUza7}P_$|A7Em!Cf?1t4hIJ_Eu3ijQd4#Q7>g-EFl=x=(Lu~S3& z)gC+A=VwJTe=Mihub0xIyLo~`Hx?S)@?gTxzr;a07H?@)phnGgT$5o&yYj5r?m8Qy zJznSmJyXYveWS6yp%~5@UZDNUmV@>jMbYF>>f9hw4>@EtEt2C4YK=gN3+^Y5T{ zgzlZ!mL~G3s{x9F0U8ti@$GI`^sKHb8y=#>d*v5Vy}*}HrPqZ!3@^jvd$Dwqk}938 zsRV8pK9Kt7wxSce|FLstikR!rgObYkg8$h@pN;s@i-`-n@tK7wFPlCA8ftxkXx$N| zT|Z1RH%s&Nhl9|eGLQtkNg>3n5HIYGAl@2-IXZ8pRy8-Ua*s7w8&Ad?w`1t2@WXV` z+rxD2=@ay6%xW5LLuo*9rKrSt9^A{+r{|1?aPx9?{&cl2+;(P=bF6@hn=e3Zk1E%l zGLt{Cb;bCC4>&XY9Go93#d}N=Y1mv#7)S%ynkmOoUudLvR*a%Un7|a;Kaz9jr=oO) zYF0YU3+EkigqWpayz1&5mVVb9VnRzu^8Fi-Zxmm4{!a+V&v1b`jw3P3HJ6MYw#~9c zMhUD#7SKtW?yx&C7j~rI#>CZ;ShMLJH83~C%L$`+x>YVHem12#-AQ!C?p*BBYA5AW z7f^*a@)(`lgKC`%NJW`B63a0Z<)S2!+n%%ccVba1(Hu`N$RL{ixABc7v1CZTG)CSq z;oDX3!SYM_bXiXZt&}?gN#Dcq|LgYeaC%-K9&u+S6IP&w{m8 z7AmMmf$bb2uWQkWt$SX}t z6S&8_yiq;6ophNUVSk>df!zAB@O{u_+-N@wVh3%c@ARd`;VMzoxj>7LklzgCND}gAeilZ1ds1 zjRVf`k&qEZy8NPa0bVN%;TT7`aD-~#&g&BQx ziK5ci-!R2Bh?H(hWaE53(usZcbgR+?dZcQuz?=|V;5>$$jQ)sML>cflNmD#SL6`oL zosQjYe?(`D7?C#Fz!j8bdGOC9=-+P=%b0r@vRC{?KDhy%_Sa)viZ6&g@3B4G|FN4x z*RiD}voz?IpVic?aS)}cM96?)R&x?$Nt>NJURl_WF0b*2*-j%sY#&y7vHEsd_JW;I zFl;Uj`E!*`xNA%Pwnae4N*(A{7!P9)@29)%wo>lgkNDiWM~0anAs-*Eqk-ejLR7E` z-1|`_I=*5P8*pMcni#x*l_z3^O!hQUlt({4y;*_(=R1-I>c7DIe$6;}*ib%ky%Dam zZNLvnFVHg~0UPVn_(o@4=tzDCYFW9g=(j9Ye1DA8YzQUGrkap3`NznK(h|vQ-7aR6 zC862c7IgLH5FB7F14&W=$_K+s1R1GX^w#b~0_B zKbWt3uSk#Ny3?w)^1^+29$Yqj3Jn)e80SnbyPR=D?gISVW z)Wgda9c@!YvwOd?gU&^wN~;&-V!t1FsHuexSrA5FRK;V@tXxs#m3WprSDj?aghPsd zDZg*Jj3;dy3m5)ck`a@m=ufrZkmRFEHx!taC#QsEwkYB6O~YvII(^uGU?6Q( z7%A40nZ6}e4~(0}{}lRT z#4`8Th#Y)1?zb-1(z-E6>GND3&JgDTvR#$ z$8-+nI3)@PY!=w!iLb$Gq%u{St1fn3pU<8pC7@L8Ba{uljn^us6(6|9NnT1JM~#gRpFHR7`8!g0#yy4dT^ zMi}{bHU2psB4#tyV9e~F$D3S)viU z9aW4S%K7Aj1H-F@8aVT)184_%S`FUwMSM*51) z?wN?ooj-_%jzg(_>pwQvW+i^vI1i4mLo_*-B63%}$$nl|;O@Rh!7x-{){IKnvs0$b z#oHN<9h-qNQtz44k{)(TLxc=4mHra$@~K@;f7jGyt?Z?lfciVMa|NfqP~gTfBctOXFO#8 z9lih~1vl>Lo;k$B{SWH5D)VtKhLAz=QheR`gJ7aHo?mmzLxm69P;;aL{0NuD#E#vp zNxTZy*<5Gh7JIP1B#BS8iiqi`C|v1p$S<7>0jXXG;f#6$B1cYxU&U6qG`kaBvJAm1 z{~9@QIUR87f2=a%13E8>Wi{4e#PT+>ipW+{owXXjR4`%w^Bl7RJ)w5BAL&f?K*uf# z(7B-`cxKk{v9sn{jaZQeHFx{brbmHPZ@fxwN?d{KLsnu`aTptN*M&W(-NJ~14F7iG z1&%#87i)$k;TjT+HW!DWY|( zC(KNz3b)ihHFu$Nln4Lw{52WjYC}w=ACgZ7Q83FmPV}EvHp?#F52GJaupID^oth)h zZ*Dq`Ujpu;#Y;`zD__GF2Tp^ng3G~l(^=8U`7vz0jwIeyj6#>MZ*lv5WzgG@BJLP# z#GC!&$o;-5xUqi#=x4tsJ9kY$+edrwQ;I8?8>frsZ8d_%B`?5MJ%n6XUtTujV-^eu zT1bDrzDlfA&EfI+DWva3643~Zu1KqDBSwlctT|9ee6}JJvlUcX-JV~fgj-w4rlu9l zLoc6|hEHS`p$}olraNp zLoS{Y_oS(cJ)+m+Pk~{YDL)K?hmVJiCYA7xm7=eEKc?0W;~LAZVs4kF`1!&T43All z+P|j5nweQ-&<=a%{4G1R2szk&Gx=mW6}~pJRos74 zi!Bzu6J~q%q8$t(vo{$-Ousg@{hk5eE}aug-aCSJ$!Bo$8ZV4oACLRg^zi9W9cEiA z3llBFY04-?+#xj`1~<5|E&UhZvBgU9Zr!V7`^RTwZJ-HX9213->+Z3*brvvRPUxdu ze*sUZxe%F`E39U`R^__rgQrH9iPO}-qM97TJ!fy=-anN%!^4=b%YBJAjvC>iRdw*K z?+IS{F^wlb+9>QoAC~X-U&AsKAL99*<;3!xEE->H!>WzOOg1tDLP!;qzI;puF7V|E zYc)}NfI74GSHf4HGFVNp8ART?C-$%^gn2c-n5yoAa(=_{#mPkY(>jt+B2%4fhq8!z2JoMXNR8sKgl01rEi91X!A77#un_r4GCf&si zLg&H4(Th>qeJVdTbR6Fw7>R>cjN#jMNW%jY8=Nwy4()PZb&uE<@1Z8R^9WgPGjyfCXlyqeqz#Pm?Qdy_B?qM zG!wN}wz5R&baJX$k-rgUciwX)>7~4A*nd`TVNy71ncxAEqe~?}kga;PWc;{N{-9_xnQ;9y*eL3NvSKYc6jmiNI6XD18L|5C)%PpJ z)Mt{i$%8@D%o^`(UCND~X`$X~6Y8745{7m^XBL9z>d?iz7?J9Tm+M^7^1@?w;n!9) zDHi6*PlY+j-l=rjbD`THwhxv~?W^z=?E&9DDKO1l&Yi_x^yNB6Lw*;*@;mavp8Grg z$Nu70nL;{3ssVUL4Xx?7wRZ2@WUZ54WvxFd*E&1@Ijw#%pPbhesKN9f8e%?H3mdLx zTfU6)w_BFc+JR)Qbgc%8tqKAFKrbavu+>0vy* zozt5myQrV*EijhNq6dZ+p}y}DUi`<9a)odzU7$d33~i*+Uw_gbk+${p&|7rZp9A#h zYw6OYC-<1%!nr{-~l+!;a{^#MsE|5PHa&E_b>L$E@UO z{V!8$u9S?^xFr1)fV7aNTCF!9BDl`Wp?dZZka;zZyY8Kjr-eR9`$<_;@y=)Zz(Lab z=Zj02ah$SQp%YW{qZ&S8Wy-7bH?H2*Jj-mW) z^I4W?uLbVrw}_3U8l1l4!`iP#;E4$jnWv8hXen3Iz_t?j#LhtR*aP&Wqo%cu!eu(( z?N6w4J4vF<#^LGXMWR^!mHfd9O_1qVx4uvf)K_hj=vBaOF7Eon%>y0jh#hn3)a^f@ zs&Xz7CA>7WyYMT1X2({yu!!rA#TB1TU}cOMU}(PR{io)NimKJHIB7bZTOLm2$5-Rp z5s{)%`omzjL|*xE{bRU)wE`U8_Xc~0zNb++hsCYk9jtAWI`5hHh%9vrqEs!5&inb9 zoOZt~{yJqCk&yoZ+8>2G_Ld>kJaI#5{GteA`s`?h{@n5CXZ#heoc#$tVhW2EwvfU= z3A$vXNyQc}c!HWYkqo(8B9)aTRyk55VQ1tJQd`@}I{9F1jne?XwTF1zG#@S#8OfI% zyur7m4~K@U+xZc#5hB@BTVd7nJQ(2`#^=5rPkqNBr=7)5EOTJ?Wlttir&Ye#?kW4W^(>saC}bxVJR?Vc7tp9*y|nnb3@H!$ zVD(nWVD;ZxNjo&}(|M0R&}Hj`ptWM3@!n(9)kAx5BCZz?)uJc>T+ypLa>C-JP4r_tO*83*h+ zA@=S)$(#y|=;j5{)McW4Mf3KvL|Od`X%PIV+rtLX%M&j{$a#dWpjwgKfhrb$O9e8r z;z?h#BAI+=C2@6`fGM3X*^Zue^s*?$ZP{@+Y;*>^3Y!b5G4|AFMI-1G9K@B$3S@HX zZxV4raP?`P+M>quK`zg?+k0WV7Hdy;X73 zbS3`#DS>KlIbPRPrt<~YYkc)~dQNyp(I_>lDs-Gg**v1NW?ceX=gH!PPlh<-VgmHe zHKSiXZlsadb}ko{AfZPTm2y9g$H&nmg0J@ z{z9&PCQ(S&qK$J)pf*Vv?{9IT>Aq)3@RJZyy>*%BJr%>l8*|ZC9D={BQ*hj}Ik+Nx zJt&NNMgH6FO^C>kdL1cexx06=Vq^U#um1RToIf7 zT8=igcC^$#mwK-nEGk!0CdrrVxLHaPNZq(kLvO64zL~@5xOAY3+H&-4);oG@u%vb3 zh(XrNBQ+qeHd(CN?2VUHd<3qKpj(zNbbZ_hFTW|-kYNIwt*yD^Dw^JJd7UseS{3#Z%IRT>(dt+acZ`@r-X3aze!i6&#~-Cvw=<^Fn+uQQ1s82(3%ty`1@SrKsp*<( z_Vj|7o(@QZz$@2aT=gnAyzvYc1~gi(n7o@5=naC^sgK}G=V*{_+$JuYkc@H>>a;R6 zgenNVn_uS$-YaPlRt1=YW=t$-XXvugJz*GndK`XNNfQ71(#E>f4pLH%O|-^<#gtJ&zq0{gH0kwko=#4=LoS!jPQ3ir045A!u>=3GzATopyH-cO(| z%l?wH+e_J`#R6}xy9^(FD#Fn}28rF$Y}rql6R7m73VRnTA@LjqZBpN@nCt+M`EUsw zhs=Nlva`V>V-z2p;KCEP4i)1ZLn!7UEazMZaa7jB#KkeVdeLIMujK@DcbmhAwmtad z!Ah8rAttBx}&y#p7~X=PT&Kk7AD05me^C(r;N@I zl(bH19cR7XaXHDhNul3r{Hd;$D!n;cml`Hm!}l$pSoDM|_%><}&iLC$l>Vwi>D>)b z`=A7CU5{g{yNBRidq~rorqYl|Q~LPia%$mmm-K}>(I%nSsLu*vh}uh>U-TVIZinC} z@73%-qRvw`%!bJZ9VmA#6B9Bhyn!*NC+uWvd}rd;A<4 zl_t~ub>r#s^EoW*u{|oKnbKW}IiS4gHE0wrrfJ=NFn6a4sHTmhs~>Nm3z{mR_PPUk zw|)#epIr|cJ$WQ8*AlvmuE46CO}Hf>0)7k{%C+GOdwt_D9&eIiv$`$lEdTph-5!L} zS2S_-f2O$S?mBu<;yJB-6Ao93N<~2ySIEeqE-;RN#tOaW@CWH@`SmX~SZQiRkB=S= z&Go5Jzp;wx#eO97t}4QWOu z)V|;)&hZW*V^3d#s4&6nxoexKtvH75QTsyfJXw#vE9#ka*+^7#T7iF$4@W&oJ!&v* zGuwIb8nHbq%U0ldGV91I(W+M|tUNh}Wahd-RqT4YxOp{6UGtaJ=u2akbOLLVH^=gr zCMLgJh96T_fjOOhVyAl<>~*~sbL?3JMKu=SW7)y-RHRtRyeg6^+=up72p*G*9Z

zD>vj)w$Xe_#Hc#Li~ z97jD@B*Q(i7R}sgjdN!wvFZzUP$Xnz54KIl(9|cyL3%SuJ@SE2GdK8NIS_*<>+{V` zw=kfv7oJ`F4sON2Sed&Z7dNjYX@5eY=GSYYU10|+#IM+mwi%dt!2sqQ@PxsWkHf@| z)dF8$Nezb%qL%`$KwPf{ZjF0h@#A6&xusnNf6uoI_XNRj6=aV78_r?aN*iW*S&2-P zd@pJ|Gmjoot%kfip74B{1M0TSBKB1>&=j)~I%FI`bKW1g8!!Qm`g`L1-3{cbbPUP< z=T8DOd&%QZwm7x4oOrj7qv>@GgHb2!xyIER#N(}^UEAgc^+lDunYiFLm`Tr|yKdxl66RH!D1ZdorSudf0^s*1o60431oVp5-3zClKTcnE8Gl5<96c@qG)P|SyBdg zSUVn!?E9fFF@;O$^%#1P@&nTpPcODZQVLP79?;lz#zSyMr*|=_TUq=|+n# zuEp|{1Ncf{D8z;9slM4mTIhA0`25O3*N|9rW7=rc>d3dgBK+*JDZY&N)` z_y}u8TG9&v12AiBjrfG|UGZ|KX*^10BepL6C*%}Qz|FM1q&-dx`mJZcVZD5^d%HWm zQq?Fn)@+96AaC>yG3S&2CXkfqlNeIi4B_AGnDp<9#C+drSa9MHv3{7xt|S~px9jso z9!qMm$GIP^JaS;_glKd!8Aem<=it4)2bpQiG_mLSJ~p~i1^kzwJ|Os7N@zp9-9@1dj7PdX3-1rEKiBMLvx$i)0cMcVaR zo7n02frH{&OxaL{KL?~h>k1j(J9h>P(UPS43V!U9t2x}fJ`=sJn^>QH^^vNGK2WiE zsP!M?YWnr|u*xZqH`1BkM$s2F>sYwy0yucu0havU!?wMOrCqH%fZP$Gf8Z)~YLMj5 zFjE||A(h4j&xc2cK7ibA7YJ|a$8$33+;w3gDvq?He|DF_*kUc}_PPnXgCAgvWe2&| zb{S2dKSzcB1bo$Y4@=%XXAM!e;QlZrNU5L9g;pXw{9`u%@NEPR_%8ug{Vj&FH-Cuj zl8sYj(?|~z=bou$52Da&A7G6-B zC7Q5I34exnk^efXiRVCBAsZtm6*h<2@$T(>P%uMrjVhUxbcQTenQs;KV=44C8;YLl z7mJ%*65m-O@ajLa#OGFGqoH@BUvdtoeXG2i@16{(-WK>K3xSo`)02v+D*|}R=w5E8Tg<;vEqXHIFn!&531nxMpx>e_ zNIJEapDKNe-wsTp{m$-W{IUemcEeU&s@{xdL+9fMz6l2xCBTUBS4i*~2FBC#*`)V! zboLnmGZLv|?L&X`^#6+BV9Z4?`An-$$xV3R&Tzc^W)8bJ?-+ZnJerfeiQID32j*0J4wnR%(KtOPI;~9| zwwb+U%6SgtO3zeW*sMm!n6H9o%06_(ivp<8HNwYhGuUju1;qaIb@6w#f9UhL6V7^V zr#mjMqD7-rAkbInfmmEdBD!DT#r5B@?&Ug>%$*W2zMcrr&bh;yf3ra$!xGJ2h)95- zhUgiTPct1y*Hal16Y?>Waw-`E;3^ET!^k{0B9$5_7Gb3P0#-%O2PhLQ=N=JGp+ zYq_#Q72JJcPq}X&6rVV6HSf~{l9c|seD99I?4h**WHp3@#jPNtdbr}zRY@$8YanVV z`=G06E4*8Dn~c5E!~&M>B&K94S+%1Pw#QEZyT9GAy-I}Rf)t=kxkda?!1A`3*<-<{ z7qG)_$-r(KSQiuuYhUQ#P|+yppAZj;>MC?hN-6nrpbOoe{bMg8 zj4@(&7`nG-;+)T0MW&@P{FA|4z9e%7&K)XrwFt8{x3fy@&4MP@;QAM9H%aqr>pC!2 z&58c~Qb!%)MKruwidspi(|I3GL1#%N$Ue=&_|QZ=HoKWu|19Jy;(~E$*9`J|+<$0x z8@R;GC9FKijP3n;8VX-{kvVFsh?>I$(fH0=%sF0g4A~l>a*Hh+tdxXf8k337`%Adk zeIDj!_KL$}l1Q}6Xke2>G=IWcQAbKO+zOJWYn(QKnS}~O4q1(_W^40Ep%d=8-DA9! zr9t1Q-4`3L+XNHWeS|C9&Qhaig;c*R4r<3pk?8suJh@G-Nw_bG~LSFWF-53 z=BmhS#XM+x`41kg+(ZVpgbCS^a&fk&Ge(AYg7ZIFI=pKYY>at@vuByLDh9W6%!X_u-{gR0=pALMt&VbN9)LlXgv~Q_C?$?Ne!Du=5zf*5myxUe7VBBp~*dt+8dje8^$c>E|)j* zQPcDJXlp59=IKq=G~dMK?rL1=jwx@S;e}?a0%4J1C4_zOXT6G7acO5U%1)RKwHM&Fg?TYtjfyg@8PGfM33t}A{KWsdLH&O(VXX>5Rsx79#lmSp>50;rUAz?+V8 zkW?PevMR6Q=C`9Ty!8(tBOb7#=!ds4LW8=G&CAZR^FAitF*j1 zs8arOFxE_-W<7tm0vvI^K}OmR$4i$+uvNJ$;7O(nsKn~+~IZZ-eJ(&G-0GpZV-M+pwi6MjU7 zh>3gPP&62s0?#jWfS!g8ZV-Cn$2uT6HEs$xHzkmVIaQz)IuR;2e`Yh5j6(U+ogxqW zcJz1i!w!cOB))%@@4wj3{6Y%@7bfVOTvxKdYDOyBbG7u50@&$*$dF#P@m`) z%d@99I>g?m9>ecR6(SdY4hl4S*s2r>&|31ANR}$YgcXE@=saaRrrfl8aX6V=aXl}% zU(&?iI@R#qI0KAw)&O6%a<=2+1>*Km5{67n#QQI1!&ohb!?#4UbAChMd-h0_5*$gf zHh!?z<%T#&$Q=Z^HnW~BQ%R*%omJYqMznr-6{EY0(4b*B{ii2E55`V|HtDVK_V!{f z*Q&(}la`T54X>cy?m5HBrCTJ6}>Q zuLSMI4~;YMeVrOB@Y;(?6W_6s_AVgzQgDs7$`aYcLUI4+I(#c-FLpA@frCQULMc;$ zM;_7PYo&GhzUmqja_3NG6#=Kiv&dOVC7g0ZA4_(xW?uc>WUP82bGRx+8!uhMyYGkd z62JQ-`BW{ukkN;(O_#}mJ&W0Rm6>eX{*i(w(};`-qGG3BS%|~|VBOm!mJLrsCxwHe zpn}0vESXKZ3<){9JFPqAK(@Z%M-N9>#w6G{MEj-(nl<+qmJ? zD|8b0l$kc2@blmpn!zhYUQKsJCYtZCH`W2`<_F{OT{oHWU|DF@nlH*Z)Wfvy#9Iz| za}FDX`{uEj60~uN!Fx9^Voc39v>?{1&SM@Bs+w$Z`NV$L7Xs`(|hU)7GcM*N3yZCyk|B3~r!?Sla=k8$Yo z0Kg3e5K$9W#^P7Py|^xN=k;b*An|}*j){XA7q*Gdn*^fk$&sQj3I@zv?gBntmV{c* zitwgVI{cOPz*RRB@ny6jzr7|9BltHQ5%-l@l$heg1S4+reH=fidk{NWSw%>r782_R z_)Pg1PHZp4cAp6G)J?f~?@B*I4!=3 zU0zaL!}J-{rmm$g7G~p|DR=O_+6t(vQRFkd*DxxhgJsG0V6DbTJltZB%Z2=c>n~(U zjx!+r+nG}FZN^SLnt;70V#ty+H_6#Kr_iU$3ikR2iYtW9oC8ha=xeD8qfabm?FCA^Gyt6Lje~02*JKI|guHlT(j_q&j~u@S zvr>#iZRW15eAYj37(gJt#2Wop2SFIw0SN(Hm}j&#(@S_N&OB>_A1=A!{pvA-fASzs z|LD#ma_-{Gz)AQb#Q_8RuaK+HhR}bxCt=4T87i(;hk~$`7_=`>l=8`%?hwV3)CJdB zw%cg#nLR}xxs$r@!WrwlChD^jSfjcO$DLXSO7~Q7{Y6a-zMBeG z#tQiBYZi{5^a#>)H0bXq-q@a2fqZZ_&d4jY%AI~y)N^?V{#${>q-X|NP;jh5deSyj zb02{g@&&lI<$wL#sa#{_I$nM4i+F>Jz@;jO!o2r~xs}meeknYhe?6Zf+7OmZBu>?^ zHOCe3lEPs2%v=pLZ016=esVaJJXU{28nu*azV=U;>|sf#r6#eoJzFdEpU0#2dJ$adRbVHn z4#clOE6p$O6tQ5L0EkR>BytOWqEMENcT45@-%E4wTTB(3Y-Ypeeid9=zX{uOz=F*W z7zvHJ!)U=y1vFb;$r}4h@SChG|I-^R4j55Nq9qrSl8-GSGxzDl>C^(yd#@&H>b0a( zlDyzTass*@mP8OGV)yn(tSKwPGS1MR)yuXk? z)yJa7PsHoD*MObl|^<-yF7QgBGkS8O|gpDe;^Cc8M=3Y{vudj^pHz&9G0>mxi=D@!5~I zaF40y@wnVz-sR`YWDOEPEay+3)Vu)WnbUFprwp7|n!_}2Zh^wR7eGq>J)4%BfalDv z68+3Bn0UEPOwW;uCr8Wh(aDkA?7JxzHI0Jr)VfY6dLRr!(}GC5$SZI0zQ??0&82w`GwAjW^)K#J*In5qf9V91q)?&h_sa!NA8oxG=tZ!Y8S+>m_$p@lr9!2U8F#45K#dR2#BVmak+fq< z;xF>f>{+l#@IDJHL;p)=({f27#IdMx@Gi7R9fuFsHNo$R z;BxbjqKb;o$fs`$!N}n_us@%$W%fdT^TPoM2s{Ea)>N?nb|{J8I!{77rypY7u2`tw z^PK!`*#_CI!aZ@+X;GigVbO#?2GlILoUEF!BOaNa1G`^G!{;WTZ-yO(rUqTB z@h#`A>bD8^#(*Yf8!}65A*VuGgZ7f1JWrT-Xc&FJBn{pdro!tNv*2a%b%B3W2ATbb z;82Tj4i5N%7dPsP#xGnAH_Q9roy28!r>+ieHy4tbxtl<3)qg~J(NOAaFr6BO+@|MC z23zM`I!9|7GT?x!9mbAm73o!HN;(8}I{wD!kC>Kz|P&m~0Ai=rYrPa}?gsEPum zp~~<#EfW@(^+DpI04hCgE!|x`fDYX~8jZrFFci=|r2!IPt7V9mZDICaBY?B;b6Jb#DNe<#~% zZ;>>0y808O<+NExy9{g|<^#LOpM#{3ix8!53)25>#g5fcO#bUF7${c)ak*-+;b0mJ zS*=OddY>d!zwVGtpVeufX#)Kiag^qUjs*?f<*?6Ef)9B(my3Fz;WSxyQRUzY@YKsL zi;{^EY5sHIzt=3M*KKFg6Yhy5`ptOJmiLAbbX}W^U1!1=u|DMm2jS)P4@BF|03Lkw z!3Rep;JB3?Tsvb0o|WH-ZOgMO83{~%Gfz3(xA-5?T z^6DDc38$yPbU-D#q(1JTkBU8Md$J=q&YujA zj3>jScLGa2KE7g>i2@v$auG_L8ewPrW)`z`DC~SSoOWFM26f63)Uj2P_B`G}>-Xu= z@j6Yg-!u)hgWZ76A#6+f5ZZfRj#dOdgt9B%Xqm1F51hSNX;BUgh+9Pr|Hq6(JOrt< zGWOLmhjhe5FpKObFn)_8crL6VzmAuR(@l2}NtuVDpy+?FC}9pZ_^Qw|`G>%G;Az;> z<49AV9wh5aH^Jc^BeHsxK1_*N!tAvzNcO>S*q4*RdMo7NnyxlD6bsoL=`+xK^E3Fn zI@4c8ui);85s)+C6uGg}oUmK1AiFaPT)J9`pX_qcC+|{r<#9S3zoSkM>*>Bu) zJOB!gDA2y~ONE^uP$L@|I(Eo%n7UDl6cnr{!73|=x@xawt$i&0m>y3j?kTmh|J_${ z?579lRgI(%Qbj9~<_WQCy9_PTJPFAg zZGg<1N~c;|)6W-eK;KIm((g}$IbAxmAmk0v@ES`4r6Q=Y&12C`n~9*J5d!5&?yNUe z3X-0ikzETfi?&Yx4vT|=A;9Vc=GQ(D$L_EPlLyzJ)#f1hv={>u*yxn+dx?CVws4P6 z1rOhHLh&6K|9w<3vBUxvx%T2KD=|zF{zhWdY;s6D0m#UEqEGc1%)q!+T`?3&Hhrr(uZ_4M4!Z6{Hmuc#<>!3K$@mWP zIXX@>BPWV<4|qm)^<5{8=80syRUkVi^^P2!BTGLv4yN^kFTv)#6EIJ-6n*ZT5$jL= zEZR5Lfwdp`3$xbc5FIlH`~Ds!p_CQQ~o=N`l<%ii=;`rz`7k@It<^83v}=I70JA-grC6=iR-4L?9cFbWQnK1 ztevqy--g?GOj?@O3NEFV+$7kb{jOqdlHdjTatqRiI>C|;lJqY7!A9O$1oJ}_pdR(O zhq4F1)+$4yZMCTFKvP;WZw#&Wjs|DBUlq|)bg?A$hj>O=1P;p0Eg$jTPn4A;2ZJ;W z;beV1j5kz-&b6{ovF`@_6}m2ZqNA;pgDT0Z8)}3^I+EM>8%TMr3EBHy3s^!fG!1xx zuJ&X2zWHGowdf*@oUyf7g9sUHadlBEA5hNveIz2OOJc*ibL6SRLQGJpR#y1Hpcaa~;E;41`#wJ^>oS;oK zCduRUb<+4q(M-JWhdmbUJ1UCOh=)C@i@4|I7vfN9NLl1*^#K<*867@tth&dLnwU+Yoa0jH8c?U8slZc-A%KI$NC|3h(n&P-n?! zHgnTVD#0`G%(0p9JajB;*^!L;u@gj&?dOU8u#s%Cc@f@ow1%f zMnv`FXp@{KlR4haww&DtjxQIG+=nwDBu@>pZdVeQi-RF`Dg(z>6MAm!Q2t)ghp&hZ z=ZXhv_@Td_K_~41ou8m4%$IWT>E#^ZoIZt5x?v8-(iCu|(^uYn?HzZXeI1>(=hDH$ z)?&~iMRw*>9ZogPL0uPp?r=H-x2nd9{WONq>WHR%w>!AGIe0u86JK{f9n3Izgh!0khz;mgXspknFEWU}C?cL8` zSZGx)r`;HFstQleJOa6&+;Qutd-(0|Q0~{^fzx|b(D;E7`#bXk4m+QT<=)=R$-16} zw2q?BN9NF)Cj_hwmhf2#lezQD(UmdkxsVzjPF*cW2p;c;;Fj_iM(v(JT`%pT+wKcJ z3d^*C8D8VshH91T*XxUv2CYQ9UQ;?LT${FTmf%z3G`Pq55jcOlGp!bSE$n1vxS1y9 z+rn-6@J?COo?C_*g^Or!bOtQH7sy|QZ01e=()@thPU7OY4%*DF!lr*N#G_#a?pAVu zmGYOc7!bH zC!-g4tZQd3TW^rb$8yPz5nA|eKr+_5J7M{$Dg2emIW!UST$_hnWAM8fhx^O(32O=Z z8bsig3&t$rgC5FTF5vU^!ZBk@D8B9eN6v(b*-|T0!O^M2t?RnM%`2SQ>HUO3eO7$p zI$s{9azp5LcgApGuT_-j4vBJ<9r$tJhL>}+3RS8@iIO)BCKw?4&F3*7$SfPi|dIw}1SL&Tsossp}A*vUC$qx#^GdboyYy19xcfOvAhj=SW*b1wQDW zOqbm?q(=f(=)0BkdG>*gIDcRy)Ts)&@#2Ma<;y~<^iG1CJyGH(YLzOtI}M>rXQ|N% z({DoXEJ^B=^sIbl#A@)ElPmfx=!ngpys_%pXvRZa>UrL~7u@)KJ`r84hQCz}VzV87 zQ2*mCtX`WB?|wz2?w${LDy0?;<^W%|bpy|dH^ZZ*cSP&m|ANu`DL7s45D8hF36Vy} zMYE46(u)V~knFQ9n6^Gt$oYSRmf#6=oc~L>_gsQlyza+W>chFg-Y@J#tq8c`D6qDW zr;bz4KwF4CPHdXY{mVU2`@1q^fAWVOpH955y-V21i4cpu$>oue;8J~_R8|+G%1=j( zE6>CDjS+07RTxO$I|qq1U-_eJ=lERNjXZKy7t7nR4GRBlq6-|R(8{Y4U@<8fmsD5q z-io7;`Z5X+%uB_xD@vegT*@|#Is)0z(Wszgk4Mj!v0W;SOvk9RHWrgI-#L zX+`XC8WtakyFVwg=hedO=|%<)YP&|(d31r1t0ljx9?A_2l`A#e*76kZr7+2AI-EXm zg!vW>!#lTb@I7OS_yc!)3@F-<5B;w*t%HBqu!r>(i>p2`iL?=Pak?p*owXso&6+gz z@demuvBIY_-pFo3!10T-6l5CO96KI7+ZhkMkiv zbul;H7QywmR`cw>RA4J^alL;7;rfJp zvRZlq)v=fYzmE;4>zr=F`gh7$mwA9^Xxpk$^vk&?WR!CQUR6}*p>JD=I6Rrgzca*m*SomwxE`N+ z{U)Do+fO=1V%rlLAaqfW1EfH>&FCR~U zjw6Z?)bj)j9enVQvLb}4c#+Aot}=VsT=*}$67Ihio>9+xqQ3N%qCp+f)(`!sSZlBR zL${0@X8m7LE^R34292RvWcrv9B>37@GJNV5)J*RZ?~~huULFTT_hZV?8cm3a?Ll^} zPL&2!>w&_~dI(-Jin`_b@q8syes{?xUY)2|Ir(A^e2`V9h89^o{H6ha(q}5Vz5WsC zNhA>a({gl0;v`hsA1dxVe1j;<%8}u2@vOvUC)@j4$g&9MiW&B0_+3&HX2+F5){?=mc=mdy*J`5hdzW09=orgb`-y6qeB!o~h zlLpDGpth!l29s&NMyF7q4<_c{qEmi;CWup z?Rn0*uFvQFW}nt)Ac1?i=uwt0Y;|mBPUZywHF*P&U(IaS0;$L=0&OvI)t5v@6 z?#GSQa(Mg12WZ!w3lOIufXx(a(1w95loPB5`Ahh2>hel>_1`EG|B1jxdj$&ZdH~&q zYstwA6(G3rDLdo04mR->2h#&$Soi5mI8&wx0g|5=qoGzX>F+{UlIy{a5r^v!C1{ky zZS*EvmF6Ea!k6n?aON?-8x-`N_wByN^82#D{?BZzaJmJ~-IBmp9_d4W%0lva(_Qvi zMKRUbHbCB)1fqyoA9$LU2OFOn<9>UL^!Dp>gBL(Uk0%+*+ujG0MJ)y&@X;Sj9W$M z%?DdZhsGt={~(JBcS<7lf*nZVa2Pt<_YHJE#X;b2$}WhW3ZXMonAaM*{F${8tly`@ z)Lp3{JH3Dgh3>$0swOo4%5tvca5dMz+lc%UR_B78W+RnRHAt68g;QqHa9m0aN37g| z4+nFoge0P|{Wn17AV9=wp4pWqMXNeKpsf?M;oBo)+SJm>cwV@TQco1H5`IgG zmMwhV)!Q3_x9}MqQVJVNyI{6&2R)V)Lx+C$^K3RgKl`GQeQmh}Y+9egoZK)ddv_WR zxut<%njjvJG=r4cMw%Y>jcw!oJtz4&v2cnPt=?M1xEwi!Qn!f# zx}VOfs25NtiChr%*o8lRQU>MY-f+=a1b6#aW4(l{L@0Ry-rqa|dF{Ex=+r~lR(u!l zD7(j&t#U>lk~T1UBo~a6e0b-SC<uzLC))~0MjK|S{w=T9YQ7k}q3 zPU}QDg}F?^-(>i=HVIw2vJ<4-HF(~k2Th+HK*w9!_`T6dbXVOUVv2&0^@g3uhUX&8 zyo9zXc!<0ogre!^ZX4f|6=q!*QFBjMZp7 zIDWDwG7nRkzZ%i##_cdrC-aG^Llx58yMQ){h;#4XbF@2Lmws$bqOWf^knPIyIPN;{ zuW^-xu(?w(NqY@#<67{2BpI$){{+=vy^ykQ2qkapC4q;+P(b8vXv|$lPV|Zrj$_ck z-|ys(TnQMJ&R{z&P2lFGYG(faQe@&5OOIqELsmMlWBrS0Cw_=Z*9kz~vkTyR_85F$ zbDu<-)iR+zx%8-N8Bq^Yr=I*#;F@|0oSeO#Dd=R$Vm=deL;4&!$}`74>dl9DNm^*T zVTVykg9ZvvkixCo){&!^-jP*WvuI@H1;##m0c;q#1frHzD5|iStXUU~zMLK;IcXuV z^=c=I(U zW48#@%J}!kX8;Zt332Ca_HyR+v2f+{Ugnx(9{KliDwO5ML;iawq~PjL&fQT&g@qx+ z=Vmo|w|72KwGX7b?kmypm!FJmZ@5#pO?~8_#~Iehz=$d?eNrLWy9X*4hl7%BFu01! z!?C-L;PO%$ttej(%Zgs0mw!~bAKJCl=5I6YNytX0mzdCj=haj(1Nd?OI=R|2j90_9QqYJ`Z9a)i7zkvDE&4C>4w; zBQ6>r$*rkA^zPC-U_R{#baP{1vVI|6DyD!HG7G@_k_~hk+QXsLe^9wQ7}Cvt!X5O4 zshPQ)=UWJ4{gW!NEy)S2xk$QReKD?34q_TL@?l?GJ^u{Hp`5JOs9s7OQtsM<@4~OB zJAWq<5!irU^(&*@;pHeh!W?zXzlM%2nT12%JOV%IK6EbO1v7up8co?dP{}4$6Wfo@ zbVr|*9jee_L~rEl zK!AC+Lj6cJ>p7B)y+oOkjqo<8j_TM+lP6>INKB#* z-M-=jjc=buGX-nm+lpE`FLMyuH;$5d+Dh;z(v!<7=_2cnyU@h*XTV&zrfS81%Crkd z)7x?%Vajhi{M>N~(UvPl=GMH^t;>Q4RM$Z5>ifv#TL{|N`W(ZMB=iZLNso*&Og_9>7Y?(go4H{ zAo%c>xCr`?+qRYLH{VWrR>gvw<7>x3TPax-lZU#$igKoJd?=o~63knNAhBX9`Q~(r zMxL1hB?kgvEKVKUuFb}kj-I&EZv%WBctAwvEy5!TUd+3j;@o*VMQ-DDMb0vX!DcsD z?Ed5<-27z2q0u3#{D8oas~T~a6@dCx=fG?y3xnAcncNo-Xzb`q)UoV0qgNr$yu$HW zRg{$r#O73@mhHJr)(RQ6@+b@D{#Hf;y_yg|6bp|-j??YaCArg&0%_b$D>&ooh9lSz zJbbPJZm7(lL!xc4pr8u-E{}oa%xJ8l_7kjMRfCd&Gh5~CK^JOVpha6M;r8|ja9z6r zEbT2&Ua~pN?wo)#4Hw`=%m(==+ z`13QG7Pf`%i@1gqrvW@P37ae_4^b8AW6-9}bg5hx^&Yy&btqu?s33wPKiPo} zo(SJscjFYT5Zk7ze zqFob+)qfT!<%lXMJ2!%9wm(dqrvP~=`RLDy38-jw8jTulqpwy~qZuxCuz5_LhOLyK zNA(?OS?(nI(tAHlXbGbq7p%AH~SCuu`B{)_(Y%7%)Y`;h7P6f$~LpA^p6L~6?(8Ht|I1;<^X z&{cQ^n$AooZ3?xt!A6t1JW(YIhGA8q>c;F?kOnFz@;vL-3p6BJ;L;H}zBjN1?R<8I zj7=yaQxc>=arM~ZvV?QQQF{t;AGfRWG?(8AH^|IE^;vXSO&jqO^5d*QmFXn zez3G6(C~B;H*bnEl~2nh$LxO~msU4g*SL%n_8ul9*_M#9b__cID&sgC8B+Q(0*u9r zm?5X7^!vu6pg#9AJWoCh^Y(lsiaR%<-ri;Ci^z5?(Q1vaN8W>MF>8d1m6<0I8ALHO z(J1_O9eQ)Y2$p>eAnIa~=*7h;T)NU}DjJaj#wjsO^6+`|yWwQjh=>dAs<}oBZEuic zf#1mA2UT>zqeDnk_bRkJ2&S)GI~X6UDM)aJ02*BsKoo)&P}?Or>}64RxM<@Bv-myW z`7NgGr?=CI#^3=;Gewy1H}&DK;YO+)&-YCm(vWM6E?nC50%(Z@T-~#q9CL5ATfpXrG1dSn8rTB9oBA$4(UeZYEkTyPGtxv`1c&m5cY!RDkb&Y0yC{;H;D} z**fM)j_E8$=X|_rhq4Hk-(iN|W=_VT=8<&IcbUak3SEgxwGO=_ZU(ZKCFp0(FtmjG zL&U1?(Dn0MsiPf3eyqHKW{m~1z2*uaVR?u(`z(q>Z>iw2n{#22tsiJE%wUyPZ^y|p z(zq_x59S9?qWk=f(U;fBRTEQnK=yYHyc2r|C33Z}@0lZgF!um$nlg(D9uKQ}{w|Sc z2kl`W6d4*?%50`y4^64r3J*HlTa5eLejn6tT?fV7GMZcz&I7L}-N3yr>6?!3x}(&LUD6BEyA-y3&krMQ($|BxsLz1o=Jt zP@2YH@Hbk9elFCfw;M{yKI>t~-8c5wlfY`3CR1v1o;ZcZA}yJ7Wai0dM)L(uGsg~= zvihdn;um+lAnolc?6Y?|x+qM^c9cgOI#p4aUN9}ME@W!*1sJEuCZ=$}oSMxM!P7pb z8HI1XMbBln8A(kXWFCv?;eOXbJks|a{K|WfO>PO@y+9StFMYwY*Ia3fOcxxD4Z}q- zk=V4v3OPMGi>6Hzg6x8AU@dKmlZ@Bnp?T+-k=q4y-|0$f@ntFJ`gkurqn3t#zL*Fz zrdA`|V}b7%4D;ESOgJ>&P99m^XCh6Ts9Nj^dh3=hZK%m4&+Vsh!(ub44%rb zjbsMeSn7)&PkKwQ6(e@}oL+h_CKU!R_2Adu^2QDM2{?FKK3yCp$l-x3Dn*~s1D7hG zD9925MIWF^l_T`!K5rU$nkAQ?^pYJnt?2N92dwdh5b$y^!ZY)A`LCx@)H7nt9$hq@ zyR9k7ea<{fqcs%q^XiHCiFZX+p;;mN-JD8n`l8{jKqLwd&H<_TY53+06>P0jL$t!< z$P-QPTnS?#{qVT5U%~W%48j@I`igpOUhXKV< zEK_dFbCvbL{hldjF;#%hU1*3J^BdU742v=s^s}!IxY3~N()9I`Iy%2B3SFsgg`oF2 zxWhY^$^}e@ltsJf7n?=U7Ai&OTZh8ZUoU8Wnl-xS6HUg&Ui(Qtt z1>?30@!Y4;_Yj~k!T4pv8si16<@nx|DOe_a5M^w%0@Vvo*(d+a!Fp@Uh{MHQ(0qwq zyhYawY**D%zc07xns+(K|LIQH(~OAC??wo^J_kG0O)wsHj>UyDGSGzhYvA+Kgxt4& zM!#*O@1YDC@!vkRUD z!{ka#XD7A7p1a*NC;k85p?Rozbr+PZ1lTRygtQby@vfCd zdIj)~p-1eQ*}Rv_D*>-QaSz$oH9=)|5><3=fEDKrU~|9$9JKf@wMw`SDG4*NQL+qv z-e&^`KK_6ZYDg!#s&UPE@95@eUsQSc4}3i3Kvg&wa_*}=sD?+to-!fcQK|#N<8Q%9 z=qUvFF93(q$?(r75{4H}r2z}t$h<~1i0Ck*OJ~`Fko^?M{iTIpzAPp`RU@H_%YxUF zm7r~w0=>3B3SQ})#gU#BxcE{Htz0og*9iEa)0)w2m)vF8bu9<>8W|A*YeBs9YY7bT z{IYq)PSpIjAU^bR8qG?N2g9B?{MD%i$2O0_j#zc*WKW_S+rEL-B_Xb5)(leiFpM>j zvB7nEBG~)hRqC3^pB;P?sJL)3T~7RHo6t+6+|R-A)G@npYg352eIn(OpA57x82EnMod=6x(_=51^;Q|Z-rU68Ob`Z^@8*BxcPTY; z;b_Eq0SeV^hpQ`i7SZA3#Oxx45;(?;+$@KX4xsltIC@h*jLqkFzy=C@uKJNS@xYp} zB{d&>p9~|t{M)oaRsj~oKd;(xwS@dw6M};?&qIzxF*$Nsj9hwYhcfN7Xi?J+{Kmr- zznNPIa}D2sL1jIo`uRBe&$$3?>sW;^?AIaFf7tW#=AY!!Z+U14SU?K)<)XmH;k3SS zAx;?+<7Xg+5Igk~WMvHD*Q+Mrik13M=_*7M4m!btw;Fh`!WHt~$-{v|Sru=>qiBuR zR+wF+1zYYoDjrBPY#`?E1o=6QijV*FnYWoIq>}GEMliT5%ROTjeP= z+W4Ay`b5!t*&|5ybSsIO)I<7hmGCo-)p$XeHdake0-M`boZc%$O!(-{_LWbeU(XfZ zu8u>AR2UDQmLambi>Szt2PEx-369;RW?c7~@BLK8vN;}48R30Lu+O|9c%lL0Ogx#aq~Vh9~ng2u{x zP&G;-IHMXI&hj0GFRAn|=_BO^x6qc~7G&#{Mc8SpjPWzMcEgrA59o~Rr{FP@i}N&o z!soaK`uUm?gstT>RD7mS<8LHA_s9lIY_r4G>xIb5HBI#8wG{H%Z4tfOBZ;q!<$%n> zMRbKu8C|(Gk|u4`pbr(6kf>or>e*8V4#FE(NuaH3 zzfxC=4*JZ@0r`gr!$0F$Sld$<=H{C~(5Dbc5Bv|rk39tQiRbYegJv{Rv;~!Fhf_lh z#j3PChFP6sNd1<#FcpQ{Xq#>WQw-TgmS-a21y%%Z?Jla%v)@j)l(45pq&Szli=38TB|2%cR*!p;z2;VQYxZj5>^y80bjxV9KsV?nxHx*P0M z`*HUcX}nah4V~DsniwvePBmv0z_)%cyzJCGO=xG*(h+1<9L>@Pj=IVCh4BEc#0YzxPrm zHX30x>53{Ym|}zv^exB6@d&=yeaC(x*C6Ji3$6Ssi`>~@?7}9%aOWws+s%b`=U+sf zvI6*|={#&-a|t!Y{s(9J_2Jv8JFxCV8mfv@Hnw;@&p0t}01wqBpgPCNxUhXK`l4_U z%LkkUODQoHX-^uaJ8=f)@4;{I5o-2djJ?hMu=vLF z*x>OPP1^T?_Ku4}<iZCg;Q=}EFDISP7=Fy9^bz(+RjC%z-^$?OZC zky%wEQDp@g-FH!7VPuJ>DL!U*i6tYa3F2gJ6vlTP-Jn+F8LVgtgJs7Ku>FTc;P1Gi zQE`zhCs-IwgC7S$qmTeuoIU~K9F@^=9iG(?5e#j?KCs(T4*Zg4LJII)zjxnh#W6uF z9#jew1+Jq2o*65Xq(b9_&!I=NLr_52ez>fqLW`2WBc0%*aJ;MvzBFuS{1^9DSsmx^ zck6_?j8nl-xu_cQN8I7P>Mi1X@H@Pj(ZDu8E+l5P2_R1}U1$54WR!lPNAkMSEsHaF z+Wwu`FYXA=isRTjwpp;>gy2Epo7mQU3|{ROz}IBI!o?{pl(@yMIVUloJKSlJDB%n zy;0j0AzGdziM=B!Y&WieT?>m+YHg}ii< z!d2SG;lhI)_){kcv1Sior%@4;)AWT&9@#{fo4!InDs|xRmr=5#r2;A4yaOMPHb6n> zO-L{L1uG^gLAQh>ZWg|PVZm2g{5TqKdyUOBZRVOe5G$yNSJp7h>tTX?W@7 zYFHm_2$@ZpXsENAZiE^%_Am@jtue)`M&@B(vyafSIT^p4^$gd~3g!EFYj9hV2VVKI zlXNb;4PG)!p+R*qmXSG$50+DSviLHcfAA%&x(@W`uq;kqe}{K`=|bSrW%P=r7|zVr z!^4qt!BPJYJN77(Cf?~lPNU%vaUzaHhEyOtPajsxZ3DOWn{W#MI_jt_KJn2C{|v~% zeIt8d)T{yuj6+b&v~H6R4^xYu_jdK$6jPII?yomdk4++Zrp; z$T5G2`tbRiLFc?VLLvCo5fMn`76rBAGu3F8YmTl8vQkpgx zZ=VE3B9p+oaRO%JJwPcyAIrTUl@F@hn1j8TDi6mHkHiq9m7@$DUo-gGojo9)Hsa>`LP4M& z9bBdfcGq~GU`H!cSw4x2Ir0}3Xmn7K2YfF3csmK*RYm2mNOHRQQkkI;^wswLmqzq$lZYb=A1h!DE*aXC>w5r)g%bRaqS6nz9@IDf%;&=CJgmu3nf|DZc) za&#j3tG$zqm23mesamkC6jyl-5^z~v4}xnQU`tOJ{_B~_dm8-kjtObhga2$pHiW|E z*%CPFoG&cc8;GouKOrHXpX8Rm1#I*_2|l@ExND~l7C54fnw{l2<+J&y*7hLCKGlP_ zFpb?e&^+;rNN|k;p8H!3GR!2nqR%J5YUUrKQy{2dpVZrvH?6V1~z|-fiEd0xa`VBsQc83P8(;_7uEu3%5hiN zFw+?BnoqzlL#=RZYa^^a6#>3$w?Oz6bG&<K+}FKF#dzwF!m%fTrN}Q_X%$Aa1LrV+YDYj-zC*Pm^f5DM&pVF zMlQcikczc3UVnWDZoT*d+J_I~&htGGI5QL1uP~$=^oBrRDi>!7wUa%9!NgJg4Cy&G z4cC6YXmp@{G598UVh3M&uqlk=?;qv#rQTZHe=3G%E2P3g@c=G%jXCLc@u81p=OMkP zH)*r428}+e%4JFIfT}QUx>oE4ev%M?4ph7#`VRuRkHR9vEoC;j`Y4+!J`^V(AEqK_ zgIx4!?G9?eX3&I6KbV|88)w~OS>erUwCB)9a2JT9O65;!0;5X%cB#XGZ%HWm{u#)h zlmu6MHW00J1x()l0duVx(7L7_^-Uf`*VZ4U&ngz9pG1_*;rG%GBa`?{c>&Ryoec%w z5!xyj4VR?zAlbkbK2M7WCzr{%mCxbllY?ZIzAPO4R?qzRTax|>tVZS|ybE?>3_KdY z$YxtQ@$*qvsB!nhvld6u)MFApu(T>3cPu?^yn5zxW66Z&#Oh&Fe#{=F)mFUHOKb z@wmn2UC4qx;RV#{_!+iL+lPGFbON!t6L9CPVvrkq4~7aMblaWV@NaZEiCvyY&*G10 zK0XGrC*B}Ci+_;yy?`wdcPF~7$6$y4R8Z@hhMqfUkcegZNbj^19n-l;oQrg*Pn|IR zbWn}k`zQ%f!RO5P?Vi|Ffkkgi1-N9a!X?!g5+cYuWOl`~wbv;bZqVZ19Sx?63Ib^3 zs{O<;MuXg0x{?~Z3^L8n+{o!;f2e`E9r0T=&U#(TMO);B!Tq%YJyJcJ65(W|`eZ+i ze|QL%l(dqzlfjUfA3@SKUM0(4O{RY4XUXKXn2BDb3oC|{n3;11pmS6anmp#=EC19; z14KdTsrRV!o;XalkAPW)Aw1{g01o&e10TJYLjD#5n02}tUZ~WfUvaGzdKb~hTe`_Y zv=>TuCE_xHKe)MlE&l6JiyE>lX!V1i^tHYn>-3X6A|)$fUZPS@F~d(soQ>GrkX_WPut) zSGnS9%^t9NScQYQI()sn1lCCkRlnPLkLSdnf}1mR=#_`vpr1buZ!W(Kk13 z_ZPx%=Um2G-$ZbjgBF+GsSIX=bBwm`TEtcMjr0D@Y&czV0Zzoeh34T@qly7H2r4$f zpGM`d!Gpt4vOyAGz7pN+s(lZpFaRa3>(yEwsmH!AzGjQC$U4pO6-di^V;R|TsGH-Cs&cC5qa zK7E1hCRX_PxCAtu@1cbot{~gb4Pesl#3i+?;tI!2*|VW@KvzkDo{kWpdLAozT#-0e zxI2ni2q};QZ~1#@W)ZEahz8qrb->o9K}1d{TD5OCm<42F^G#Rq^xZpg&qE2^`{fKM zWEEqpQZKkBvW9z5x15v6|HkwMFQqqIR?z5aA$XfxAa05i$75T)AoTNk7!RXpzC}Ee zvb~NBlPw@p)PgRk@rLG1AueRZl(ygCoz35iiI4Lmc1_|G?$Kv)=0yA!e00_+oFTcN zru!ZSKi_P0CoB@1MH0B=(*~Qrc96a>3Fbk!AZVYu3`5)NvB}p|ysKd%e*d!wR=h36 zPU~fiZI&n-cgXP^{^5IEpN2IUlyJ06LI_H_=ECL6b!g4Wt8}-_WYF)gq{7ux;NRW@ zFtcSUHrQj1w-1G)`MeWUzwQvw%_8{Hehu9Eb0Pot&e1VaNUm4O!=z)Au!Nrw?68Od zUBM^BZ<8BNzvhQ0yw%6n0hdAH_6E4}UpKNisE($rl_%$#huNQ6jqIDVEg(^_mtK0Z zjkp8~a09BdX!S%5$oVbHWn51r`;y+%U%evSf%)P@@{$v9s}_)~z`1mAr9WBNKaU<; zg!nlus;}7_n{BVWbtdwyCCd3NTx1GfX(aXkeC4lczm@1(svMa{Ern_x`ffo zK|B}n^BZ_GJ^>#p&ZpV6cHBeVcGBFOhb~5la9dJ`$<}NWNSNusshv^f^nUzc%2iCL zyviKN9}0(?)4QN;*9M{{unNy$=<#CRgs_05XO2+B0 zX)8IOb2apkhCKc1=|>`(w5fCIM*8R8G~$^slS<7pU?xN&cHz=Y)DokHV~%*@6%Aha zMA20ov~4!FpXJET7Am0Ab=Pom>r}aQDUnoRB!j&d(Mp1E8^GPMIsA7@9JmVIAya=% zV1?5asN|>5uyN!7F4b1VN5#Vd9xQ=>>u#|d_O>Elr#z62ZlOLs;O+OdU+o`Kw2X!AOWvWuH~$eC?@6F8{|34FOyO+qzNO#RAA`4nmF#bS zU#`%78Ry+LlM@rWPiyfRTwW(_EF*Xg-~a5unO&J#Jr#G+858qKv)oJM=o<{#UJoGu zmo&CA7Gjn!afZ$07PVo$aoA=_{PC~{xOZQos@EqNbnJZ2B{L@r?@Qx?tqDsgwv<`jiv*P!-*Er45 zym2ey{wT5LOyDPO66XiPs=! zaV=M+wVIP!ei7UMAy~}57)+d(ao>1Vf@gsSWZpVX-)+jIxhgr-_vTr;{28Sd9$H*; z!D6bpU=itkG!5^*CPlwnX>q1e`?x*tACln4tEkIL5cT<*^VASeFg!39!oOZZR;Fh3 zL*52b^?nr!%sCRxP+Aauq6Lrb| z$j&QNBjOY0g0-YMlzLhsX?JC;a`!w{KWt8GH->_3rapSqIf1K8A^xqjs>A5__EaCeZd}iy|vE8h|lnjX7lt`{$E&^_Na(TXWN4_PpgVb6^EbGwA^r&t#)I13y4@dU%pJUw4~&66)VdAgisI7RFzs z3wxtk9k&_SW$`bb0T)YstG*M>@TK(Umm4TPKZD99Cvtq;f?NN+239uyhqg2oBZ2&T zpr-u>4SFsjx?{m8^~D1ARiic7oRUZCw%s6Iyc?AbvSd#3I67B$ju`kQz@CzKQ1NL3 zF^pYgBrfyD1ZWeHTw;53ge?TuxJ0?Zx!Rc}p@q zehGP>o&)Vxl`!XRCOKYrilz*VAW2hiSn}Bs@{?;pO=%-1@-%}Jc>D;B1?!QoSA&>} zicYYj6|{T$DL!`+Ku=|FqNFcJRE)O#WPYQ61dMwa|ry z;>_l?H5HO;_}>uTnd{zo8EtJdM;|Vv(T8)x$rF=MGCwY{^36rc2Ea1fX)I5pY}b>8 z=~l%5-&6?MV1PcVx059C7<9wSm{#k(BzZIKsL1Sk;`vjQF7+-Wsyi0rH|Mj6UDUXd zhVDh;6(vKv-=>4d>qF%4;RDcmB#F(+nh9I7L%@P-0!E_{U#IW!r@}ToJ)J+Nc=zG6 zV(;-CBFJ`j-=K}tV`=BryYS843Lj0a2gR>q#OIp_r0HnGotqf@uXU$U6(6WY-fiSL zS(kJTH=?JpE8utfNpMOOg$4RbR5rgH*-n*Vj;N2I=yiXnl}`_yt-B6F*2=KP@B-EF z+4VW0#c(eC4r_JMfOaY?(Zj*>Y1MImn46afhwPs-dvXsWs|;y6c*_cY7bwEW_D57@ zXC0#bm~v}=FwvLvK+s?aBv-r8%zue=X08KU@5kqY+P9I!`4bsm*n#w03L$P!77hBI zLx;QO6JIBD=2Uq&Tl2dDy7wE|sB1TzqyNtqzyTsz}RZyI$K$E4O z89ib6S@olJ^xs_r{LiO=R8J}fH~Dnz_97L(ubzdqV(pn$zpGSApih8dS6J z3~1G5fpNhZc4);v7*5Fo-2*&RZ)!j3dGHeS4fo=fesio-E{YRo6oLGmAl&b%fQuVi z!F25$WPfNj?;$=3Q7V(jDTAq0FXJ=v6G)Fk0+mA(z zbiiqZ&x?|)hsD`OzRY@d?_~Dm)F$F$n@msFHKCP8syKF*9gNMaL$^vJp`t_`m;XzL zdb*n^jmUFf&rc>I|F%a$N<5&MqM>MV|CK-<8i-~ z*z%hECt7EO@01#)loMDo+yu^+QsDl(8;#rCq+?}Q=$o>C=&7DCH$APCPbO(_x6G(BDMK<7)VUpdf6|D@c8rxmIkYa%hYv#a@Sl?yvPqQYY|a}qc5t*>3EP{+(5^%7oX@XwT;+><;DSHWD62t6esvBz zHC-4E<>azi5=ZFoIkV~V%|*2MTn9=Sd{1RNp3vjEj%-kF44pCa7CY@>4M|otqG9!e z?56h^g&WL<>PxEBF6J!SX)KSw%TEF4@9V*SqzIYXif|8&HbK>HW#*&qXqBVSHQMMO z%&l1G%)J}UK_(SyDEoN@ivIkIohZQXtKvSAE7=h=;7SpaEU08!!e5~Gf(2mhmyhl# z<)VlqjnvQb7G8LJH}P7*pQ{A{Cj56FdD|55%n}1ScAC$|Og|5C#>-hl zi3}!<&-+{$dPT1P{YLxSyJ1fow8g@H^Ym8N)NA>Edh2|o< zMIwi;o+!_)?rK2BKc(T_;|LNlsv6ow#Q;>*df*&8i*gqWe8`j;b6zzAKswUwe-GOXHY|Y3FdLw+OUd zX{QJOl+ojVCUeIdHL0hVEcb_JN<^*df`gl0!~NB7XqcoW7vTCH7oE!ozh6=`=)xJJ z>ST5F@=QH(URX;v9};I?8D6PcR?EMp0Y}Ef=^~Z9QbNN|6Q4eWd^WhncBdNv3pb+qup2%SFZ09M)&jj|ujUp$qwpEXoD|vCp1C+Re zK?yDqE>ZK?UiPiBG{;HZ+Q-W{nMG|^}A65N=R9JevQ9Lp>fMi-C$VK=|~f&!eb(XX47;q<({tlr`#Fj%?|ItOkr z_0*JGxc)V5jSNF~6U=#jc0FBJWz4;)&85op(VJEJ|2kRM*4fG>f^(Z)vW?`dwXFxZ!@=H+ZL`XLCCl%OcPEyMu5qS zco3IzW_&DM7%`14IEiwgQ2vgn({8fzdK}$m;K?au|D|etE=uT;3THbx6RE)-Vt4-^ ziRJfL%l!grb+;>UqOptjANUTDsj>9K{)ya=_6eMLu_3thh;tuJy||t3iPY@d7OGb` zoi^V{=l6VX$fg3GY4+Zk=nS7B{mbI%Ygaw0-MHydt4D` zTJ(w8@!~hpjy=n%P38S(Sxb;Y+G6(h-Wbk$W;WH)aHE=$t?;50+-;m=i^$EYd&d1))TCuR@-1Z+t=KO5}*HOeYpodD}= zE+Q|!>!<>ENufs_tq*RkI`APD*A~nItGcH!$HW24=C=^G?<=&I=^Del5Y%KS z!wo-rNWY)HLWUA%ke4HQM0i0gt=p!=jTX5=#kNrR-Yx)9{%!2FtGP&0V2o*brppGL z^Jco&hJetdK=x&q7|aUtBOhx_iS^oYs>|h~XMbH#lSeQ->$w7!%Q$dqNnvcQ_3&(| zyV$8Go(P5)Q{SRs>cMx+htv0yc`IKMy`Mt(mh=O#v@W4oD+n%Fds6?W)o?9&KUxN} z;b+Pnl&erLpYGTf_vn z(bV({#QNC9`zI^)$z*C20>mGJh}pmf3>_QWCiId$rt| z+m4)&a~-G*N6^E1`dqtzF#XSeCf?W@4%e(1Bd6QySSK|euK(9f{HwQ70behu3buvy zLM8Asx|;3S6^%Yd3K;8{r2?XxK*DG?`uM;Ry(wHmQ)0Am=ICj3!nPISZgqpDa~l-y z%z~~7GRB^6LGZ@Qj?8QN%V;~Vq>kQkbWKGA8dtDn^d|pA$0xDGPOFHzjr&onnV(VQ zkT7veEGD{l)VQxYN4eLWK3x{|j_Tj7qMOc_(#$EHAevy#tUcRCa{|+8hU<;W;9FOS zokJH{{bfFPG?C92xI~Zui*$Nv+i9q|ew+Mhr=VYsAv{EzV_%H`NXOG5HyzGP=K!m! z+k?J7{mi?bFEi(gZqdv4MY+4R@nOEh@>FOpH+n|Sc(7ZHn>cF%C)rHV z1FKKSp)8+~`yCD!>2E49C`U`1rC@_%3X)52qOZSPB6MdwrdMuoHpxoW+gj4OuV;!l zx3UDh(DfXMO^u>58K=0*zofX#zEN6#zZ{kR*htHJE4l24cI5VCf6!=+p^Xl~%&CuI z)IBT@bI-TJN~aL$RohRUmVM)WyH_BBjiHD6y-AZJp9?>5x$2sTC5xvv7}@Sqr3Q2R zz)__GzPc5mw7H)2#V0|!MCS|94Z4hCoLk5SF9q1_D}s+pHNj=GWbE$}PEvP>V|-8v z_ZGiqmA+&#TWtQIO}b~HK)@Iph5$s=l90|kOlxYMvg}HAR8et0DVii;;~6UKvI+YB#ew7I6tZ=?Fqwbk7CLlxfl*D@GQ_K6LG(rXInO^PC-thDzYPzk?cJ}QyGPCzx($WJg=V7xzBZdKJWLV!EkHIE7oM?$o4%| zW1|OcAW9L^^xv7j^ufx3LeST-)T~(v%36R*hl*LXJVgO0F@QWN2yDp*S@eg3_RbL#_pd?w@lP{V7z8<1o zt1#GBffo)Ch{-=+7V|E@Jb1$)(4A&Un?~40-BHH1h4Dn~%p_*>$`)6PIZUU>oji4GBX9U&#f`pn;)~CjVCA?NrUr-7C41xP zE7fPLPDh2k*<2_o`&kE?@671XkqVF{GYju`CSu*>3rtgAooa2;9|zXN~;nV|dqo4Xw=+S(;`Y5a$?+?=H;+ zI>L;84ITg~8S81pA9v!p%>^G??2{0+YotU!7&RW}NzP>669#o2#&J7k=ut6GeCJme ze3?by>r!=eYp4-~xcwNoQ=5K}Z-m=P>3BBisJ+xkbsYWZALRF6kKy5;LI36s(La8K zDXoaXz}{+bDab{`KR)m`Erxk{JY=svO+$xCBWWLzyV_^(aC&D!D*gN7>b@H<<{tnJ-cmsTE@j%kC3>lQ=S%sa$r{twYTQHY)E)rn@uQP|@g42$*W zppH=v9$Bu#iyo8Q6$>P#BWj7(NmVRuA4wPZ9>oif zf5Cc{S|K#{I4NAE3e(4$kdVog{T;Y~Z7FHS@h`s;)rumP`aJ?%mhNWV{qB%($7fL(zV#%{h{}I=#MP!7Q1#f>U5D)c#w6o zxlNfk_r!OhcJE8FcSx*cvVo59?VJp~pWz0}Qfk}$W7>Q<%l%Cb^SrKayOC$jDkr;Q|a{1IB0J^24zLE(0^_` zcHgUmhoXDaf1N&95Bf&7?AeX`mKMNKGXYb4%(3REoKy0aDD z7-xm;4c5YlMW0Bys}aQ98cgzNBvz$KBd&CXZBG`7x1=bZ_d5zg!) zYsLM_`BA7j+Jeenc7)fO%R#ZANV0LS41M@+5enwg%-Fh}Y_?m)YUn#QZ|YWZb;SkJ z&m^ATn=MT{@6ChTFN^6Imc>?=9>K+j=A*prRhYQQge=pEkn9>}OmCivz_8XNn$_t- z{-w#2nQDPd(OC3TUm$QKvx)fhdSmcf2a;%5Eo|-+?a=r97krnR3UXea5KZ?;D)lJ& z#HM4(24kLRGK9xm>BdRpE#TCx)iia@99+CJ3tnx}#SeFi=$vX-(ODFQBO6A8W_=!v zd7{RDyeSoC7jA_~OHP9Kgj+0XnGu=;RUZD;3EFD%&_(YpgkMSJCjzb$;gucu zm>8q-3=KRs^*V&EEGI)_^r_qF3q-!JCiL^742I1FZO1m)G`NT;Mn%)n<5!V6F&l}u zv>sbfvWuFmIZp$m3ZVOhE$s{wv(O_~(_mR~RuW{0Eh2-(^X_5cSAf2-y1|tDr>dhU zL_>Fl+ay&>m0z{41I11N%mhHo3fE7;AA>m z;cvl?y10|e&poOC>(@{pU`%TheZVw#4r)Z`!Ngc|G(XZ#!e1-nD#rv#{8l%?Kf;|^ zzkLecBA4cG=TMrYX^dNp4?zFt?trf#=#^ps zJqK@7^O@G&YI3o@A09g7jY9^yNm?#0#5G?+(P6+eaC?4_+11X1jvzg#NSO|to?E`+EHD;%OXQ?Y(B{bow>(uKBjc=@3TVkshRZa;caYhqyY|^@tV!? zIz_e4YtvqhHMD_@q+Z)faP}}AGz;GY!_}3z&IL`TD1Db?%5=k_41KyRsRC?*o;t+Z z-G`M&2+2HBOI8~`V-p8kVr^zLJ@?`vzMdpcXRAM#6uurVj9G3ZGDm?YJaUDCv^heB z(?D{3X&vsIr${^co(G4WqO)gfJ1mX9OwP>F!Stpa=D((lP3s?qv064f;FSswYkDHI zHV<+9G}J@fMJsdP{8qZjX^7*i@qUmRdk3P1xY16nN1#6_fqJ~&PBh+pE6(;L?0U!NMXpl3Vz`ACH-`$$t0@2_N+J|lYX&%luA>mau@f{r#+rI+Sz zgy)_KWUiYmkLd1jVEd-fa@`Pq$fgBzf=`O``2&Pqw1?e)ufeUYa=dARG57W;}44^L7N=;u`@UDN7h;NTr!XpR4szS z6ADm$<0<+aGUP!Gap2dqM06*2kYUchCF2j+v%DP}(Ku!=bU6_^uKfZYY7ECEHkGW@ zXg6E_FCIo(s^aGaalfrHl^T4Yl*?7J6^r|V>m5^^)y!D;^KD=|E(Wg%=h)<;LRK~M zJ)YSUkB{7*k-%PkT%ju1U{MEmp)p`jF6FV}e3*Y1Dzen-eoNxHNaU@^){V#3{ zeZo7m>?_;`Ht;R)-s3x)&*ZE0XVJTR12la;qmp+Ojvp@t>Y_^~e8&uY=sA;nrCfo! z?;lxV*$vPR*PxwEqOT~l64svC#lxL;@SJ-gWMw7bJG=RqGqDABI8{rQ85d)7wluzq z+6LQqDbkE5=6vajKX^c7_AO5J$0L1ON!XPpRB&*>lV>iX6?4SNlUI;YK{5EUX9qj? zWe=`wm!i#$izTa8h+geM*2HQVrODmf9rmtv<7<{{@=L=vLg(EaxTopGBL8j#50!Ut z@bev9I#!PV=oEP{rc0Hn>KY!tw_f$*1)5Z1?DNJa}ye&$-)-gTE}n3UTf%clidIpV8-cySrea z^$_~eX${%a9m@}nJHjJZ`Ej3EX$-NdhVVH)xUWN&c6(nIhCDCD)y7{4n9X5z$0lLT z?y15_#h=3A$>liEP=+6QIuu`B`AgQwNCO1l!pfL@;fawE`=V&g8!MOdPIW`teEALv zr2}~Bz9{A#VFICTh;FlVx$gpP()FWWI20nwJSivA5P&y=YIT|L$L7O z?j`xs&lgTh1@WDyFX9&6N0NcHL2&a?QQ6#oC(xQmaq_AaD~86Ar~7BH30kwD++i>^ z@;pLS-#gGXe|m(`E8^I^{93{2$Z|~8Fvp;Cr_kf=Zk!$T0Jm<6C9>~#KwygKZB=vP z{dha;pIb&Q#r-3{wmZOrqdqXhkVtAQ>&s1CHjxP=PVilj~WZw}a@k&JCzoQ-EG~yoAwR_Sl(O&W0;WLGEw~ua6&(Q@1@s zJJ-o1WauukU+WX5or^n)0e^8-V~kNBv*GrrsRz1%;4JeRrMA~>b)WGdZh?05BX z)@b~hIsY=J8Vhbi_2IE>waEI6)HjElvJQNVNf6GQBE`F}29!_T9L(azri!~rDN>~0 zjENun;-chOydT$>nm5?cL#>D>pXcJeL-FvizLY%GF`z9|ve`ZFJJ{xR8TXIB&fu*% zSwHw4q2dnAT%0wwj@9IT$G&0Sr4$&jupbRM(}xA#G2*?s?f7kR36uNfO6m*6SyQwl z4S9Hpxj&YHIG5dU(XJnvYUBzVozuwLqz~AC;dXBL{4RDqnaXO%yd_@?XOm0!ORy;; z2Ia>d#>1iJ?9&u@Vxz}X3V0eVis*2!&n+;=~7qzP!$dRy|MCld+L=IIxENnP$=*rR1g zUu%i`ZnG5Sm9{eQNN!1sIv7fi1t4 zc%nrw`kqOaED4^5RrB)US9b-3-ATu39v1w*%U8Jg)fO_0=J7CdW3GG2id$WF;znc! zsLszXuTZLoY>j%*@;eN|QYUiPDwiE!Ya|Q|lIF)gykYkzTqGx=Y*~r+UWtds1*|D3 zg|Tk+80kEhOb?_iz*GVEnOgEA?_OefvI)QRa}oKwVUaL1a|kPYV9A(gB^enyj@=Xg zj(Hb9vwuP2e)TsKd$JxF)1}9o&nFRwj!EEg_AJ|ZK?Cd_RV16@a zA-3K>j_%7Ba=F6>75*DP@K*yx-gjqs`T5WB5D@bXHosqv`Qbx>JXPc0#l2a0{Y~=n zxB{IwAsP6_J~T9-23s}9vb8;R_+_yJkZdPhwDS@gXm+yG%WgwSlD3Sce zZkqFrS$@!7eVY>dad7^{QKmmTk~77fl8v`t!hqF@m{YJHEp~YD@`}Bvu9uHyi5Be7 z@n(3k!kQmZUBWl#Yx63Zqqskz&hA0J*m*2o3J(^!@DKAh;FR_WaAM&a+`UT=UtjKv z&f_QY!x^fme0VTSR7!`X>+X{k9&s32DM!oaJCWdO9X58Uw_tqzBU@5vK^n4m!=m^$ zu(7Q`_5X57{9X<2t8f~}ESL|s29Jh|*EYj{1+^GBu@(nZUS?LS+hCD&4U0Z~RZ!Y6 z6^^#3ko#-oFh|XmJ(#~xcwiHN$^TVLq?%R84J9fW^VAWBmBr$rDO1UR5YNooX0VR* zdqifl9f>@@4F4^f0o(2F5$&xf1)?6szFFOr7(X>-4Pku++bO?@Lv}nC=#R#(X*&tC z8wXDkpTf06ccA}!gvrr*=tKt*2A}w?W=4gjjw(2NBuyaFMCbM2{)t{u4sY9 z*?1ybzYC_kl0fP~MKbI+V%pX$sGDvKEAr-ojf@n@+1|raKUKrD;hv5qnl_Hklm4US zzjYj4R%cMlm7G1yiyL+R^SRTYBBjS}UwuLZAs;f|`y;gs6f`x~qOV>!lNpoGbRPU- zXJ4Gc1xK9FEQPZT)?YC9L~i-j%?DVyQyv&sYvM(vN%+F;G)Xb-2eZC>h47%g_~ThK zPIz^gU)?<0u~aMB@%_tK$9+#$JL-7$P(e}8aS^#nEVtZ-9r>p~xuzeP)18ZtW*x;P zVY(1~WT`NqY%JdQX-Btfk!UsPB(7b&ofiIVg|_rCs;kl>x&T+u%yJ$2jP779^A}Q^ z>!Z^~^;kKuQlaCZLJYLG;!(Y5vo_>j-WyWj3Z}W#{Ni(#dFO^GK7ob^PG!iAM9!HLGfuqSUxU%65Vwk3d6!wBlVAct1>4CQCuE)ZnH zCliO{66Sfd(xLm-&9b&$cWkvfff*f|F#coq3o))C)_4(BWUD&bh3cS1b8qa;(3u6j~Nxbc1are7mc4GKnm{{l!YGZ~_e_jvs zE|=l(!MzeTVmjT_a*JupNTFzdMU)rMQHx}@dCxQqC_cvG7oKDxxytm}AHZW3SJ~$B zlW1jm2ZIdEvBajA?0esr*I&uTcv%-TWO8^UvXcCW8pS5uItfwUcgYTJfm2?ZknkHR`KDMj2fKU4v4bc`CSkiS1Avv1$!j^W+J6(lrCpTvZ`v-3fC1X(9Gl4ApaoJCp=jF}00Z!cmPEXmFtY;GCbUx+pYJZY z+^3OzJ#mYeO7DZ5rE1j5eK&k7b%lEh68gR82V5RFnWm=MVwvk|vLtCP8+&a$OZhy2 z7tCk?m-HU|vRR3LcgTdJ3OVJfCSJt<*I;PdIEKAlF9l~%72-}j#F8>sEI5*czS=U7 zXYiQ8e-g>_WF~Cu(?zU9x5Ee9!A%XmMs40D>j6xqT<^oHq_%A*)wFG*pEIBrByQcE%_n)r%YD+KNB=_X~@27qW z*+Q_-eO9!po|L$!mCx*Z%pp{zf}~maW7_*7d}tIY>2mx+u0?VpoSQ?g1y>UHfriYZ z$DB;FISrA5A>E<1lWOcvfPoG}=$&yExJ6k9?o?Ue(^2vGy3GK`85cODFC9%T_WKV5 z*C>FnT?3Tgoh1B+10At#89C7rCrQuhWSh1eXXi|J;D7Fku<3LtyqA^%vso4F-;fiM zo~T#M_HmvtYOF1+dbEHwUMgdk-%iJTR}VJIEey)vJTqys%)@c~&CM^G1{Fjn5q(PkoP4eq$WQ^gm4u z)Rf?HWIwQ;;7Ds-MHU;X@JDmA@t;M816z5VMNF(B8ZRx`uU<>CYSJwDq*Mut(>B2x z?YofuG9TD_2~&`BM2kdq0&EQf=*Qy8`}30nPf1Or4E}4(Rb(4 zYWHL?-5ui?>6Yvm6;$D93Hu%Ib;de|YR;fL=GwE1u}XAJu?h^GWX*PsDP~tnB5=-C zcW{*Uk*tQr`d@yUrDF9x1Jds3w1O1!}CQV z1FBRHoHCj~wwS}&nTW+s^I%`%0C@3XFgQ0ULbJ#*8B-tyeeKd=Q7eUnISwScTMp`i z-jNvV!x%QJ6*lX7kta9P+2zt2c-~&@XX1Wztp|X8?E=UOI6h@XUu8*VcQ;Y+Gy&Y|^~a_TUtUo{C7I@V{9$`uxP?&kyYSmfTPn(TmT z@indf>qNu#mciA$lbe{xPNjf(Vu5VEDdgxeV-g~#KCT~ z4jhKk{O z&f^$~8yP38Dw8&Uz2KeG@rXaWd4L&5V$bZo`c(QjRYG_OG(S40@#vxOv8+`&! z*gEq?6|;HX#vP~{c9T{3O~Ro3VLaMXj$a$phbyj4Ws%>05W~n87&+xU7>aH_Jmd;Z z_YL`!U=_Tv{yNz(tcZ8sTEP7bx|!4rf4Fg`2|aJUhUfR<*khI3c-tTmuQwbg?Tuv+ zy4MhHy4)cdz3P1RQ$mZI&*0i}X-ruDTVfFw2bSWzvtfc0ACbC?dxd^L*ieb4i z?3S-CH6u!LXF)N>h?^vMU{^^YRQ{@!xMuz=*IM=y%a;yfK^}F?V*W#kk3k*gi<#9+ zmOiMVE+-7U5ds6(IAAk7jML6mLdq>|JW_s~1uawL4Qoa2nLAkV}w$ZKadgU{=(I&y>zaJK96hZgJ0)n!=#dG+&NGdfBzg!HF#0m=W{{RyPGA9&SKGqaLU< z-X%V(kB6n#ip1=eD*5{I4U27C0lKfpq2|Vo(5lfX?43WGxQYD(?QaJ_@$pq0wCNc9 z`0mSh`ddkyA3udP@~N1!KF?$Zzq6ztUlU%7 zvMN<+OB62AOu2I_WxCUyy$ z{G)Le*IUmqqxm@--0u*%u;MX`h<}H5Z%%RBGh?`H!b*I<$(+5_J`{5`st2?$FC6DsI-#4aXEkwHcWu6C6&-NcLs{1 zRPeg<7tH;lLHlJD1do~vqos!u84n68^wgn8Wj0PWm!=6GL)nyyY`mN2PV4Qy;m3{N zV8xMqe7n~jX8w#s?HmRA-@)F4fKd`0)$c=y) z2NjnT78{=@dB47#mF>1?Yn+0J$&d}qHdqN+%Uy}-tTSMHtC@JOyCgZ=+e&=5Rgc3pL+If znJ4JYib9&8hDI~Ig>j+2xb@o!thG7IKL5GEMxD1t-To?Q-_^-hM)#>0)iw{SCmIq3 zr)wgM;5=&Yd<`+HJ8;B!7w#0kkV|Rb!Vf7@V65#<*Lj(ODA6NtSHv=FkL9HFM55@} zECAWZb7+cI0vs5r4dZ;8+5B8UUxU9QPj3WQ{;ACkm!88tHUha~b`9UEeH8r^<`9xS zlqzpmg}Z@si4eCK&HMip*3m%THmMXQK5b-y@|Piheh3cUFq#ZgTP#@^(uS@kUnOxB zwi=swhho0qH+ zeHZfNqxgM^ICqT<_;;FoX;}@I=Z%-Fj9!kNSrc$}UOF51!vS9Ry92W|?~%Q7uF%@B z9bNt?@Y?SKD?Hjp@XWwY@?LorACN3QFO1TJeQNsLQ+GK3Yf#BQ-I~dDZQeKZhtq7CWrzjtAqnS#(f zWFP+a+)hKUT;iS^7Vzc$DKGr0%O?pgSg$F%k5&xgn;+ETG^bN!>!Kc^)Aa@v{fnW$ z+XKa{_J5>chzn-ubg{nn{UCGwPS7t|#vKkh@hV$0*!=Y@G;R3LL1#uEy0&ZSALc#to69yiA( zg$<;3U=!GUXhLQ39fH?w1)1zmq|R$LzV;D+KdUrpN}L<0{P`>Sy;%wBzYP=f3p$|h zlP~Be>aypy)~r6WfBC!nv8+VwMEw}j#J)u-^NAJVc;HGj)Fm8;zziGw?LbhuK}Ir6 zr-^7U+Gl?;R0juoy&`oR&%z!1p=^g`7Wr!=gN}pq@tEWt^HLLCQ8P?Y$9oRGpLUJ8 z9UYD-W#-^{&;pewOeJ;mlm5@hrrz4q*eIDvlAZCFh_kH==;V~Z237-V@6KXdRx1Yo zKdljX5!VizPDj+5(O15X#7=iR+bVeSFSCx2svyz*BYg%|-$?@5-^b|k_6o{d(r8kC zA(cHaoLV}s<@bISu)nS0xc&1!_IjcQ6cIT{xX~l1Z+;^=&=kO@ukOX;O*_hsUZ}w2 zqCZTg-I%?$O2t}JIV@}$M(ZFF%+nX}Fn>$TA5n+Ue2LvJDViTqpVO))R+G!LQlU19CjBU?n$_ znJs39-%8q+`3Q|4uVZ1JGTh(s1urf4g^-_1;mF2(5Z3#{-1jm3nWhC_^YAm7cJV&k zaM$8r^as3q=}pRr>qd~8&wnmHc?qZ$dG{IrL~ ze42vpf97!G$TW7VWh@(X{ScfomZO&E0zi0>gaZZ)hGi!OeB5;%9vy0fDcgpDeS00U z%z3=D;unj=bcmM8q(9D%qPoi;GWX;#@E@nZ4*s1;x(7sy&dA-^bKxjpq!%b!Ws%Y9 zA}gfk6(o(GPNGMCVY6bgaQdKDv^|u~UKEJEI)z!n?KSU(NM#GC(Mu(cN%5pxD+%K# z%2TgsFE+SnD4ve-z<{?aal@fpxVmf}CLNMOpQugcb5!oLj+x5%b@?1@RAEA8oD6H4 zdyJ6l;e5*UWOgG&gYA9MPE^|tU`U`6*8K6~*8Un8en=O`Hnp&fKVpWgr;J4(Fywp3 z?nf2z9&$}iMUt>k4hwD@k>EH@$x#hMA?;l<%gG!Ke^wpDu4k?2~0oleB2A2 z3onuXYP7LHsZ99dyczb+SEobzhQ8d!}EcMQLAURV4SiUL^%NEM;isNh1eq1}G z1@)sAvr^E$LkSlwHs_OeCGuq|@#r%-98S+X!F-QcvO&-0;;d?0@^FDaxioKsWMi*3 zzWMlw1s@vA90%3`X)pj&*>NQ6cD(2roQ29S#&YXTm6Gi*0?3d(^Ddv%|-X@C5EtAI;3(Rqc`YfE1 zy8~r(%=s;uOf;Qo46{>*uyDT;cDX>GwB@x*$hNJlGrqe>i>>NNtEe_pb~0cI-Y*?9!bvawh>%6#-jgt6PO~cOY{0DK%;gB z**W?G=^f_IbyH5_GMi-d2pB@QjZnb1TQ!BmfhQ$@G!P<(UIR79o5bv_KmIDn!tQzL z7!rLRc4RGqQl%?wZ}9*+zrcm&8xAM9d?DD6bCP_npU>`{naIpHhLb=@l2lkrIh_19 z6n+o(A_WV7lMfN2aQnpd<+;iCMAp3(Jr}x>B~^Z7*XQ&~=I6TOy=k++VJ!!xkbRgp z$d>%)n@uuj*MLotDz3TT4G(iyg8r~-VgCL<5*eNay6ghDowx*=n--J$`bw+}b;R~P zN=$#;C>(FH5i1?+@k41FtPr`saUV)xywWw2bS7o3gUYgo1I>z#(GvS4_ z0ghBXB9z&Trh8sogNK{9NNU!Mr$fy-TWWn8!|on|oy~DbJ}%`)hK)s^*T*qiMvngN zpdyz}89x@5!SOw}d^q_{2HBFbhKKjD;PtS&yj*WD6ZX6zN zZy|~bVYC+hf$((&^_%nsW||pFvKJR(&z2|HF8^o!WO0xNoUj@KdMXoOLctFDW@!gZs*eFtOpDSc zGBCJ72VN|*qH7D+m-qhqBgi?PfffrnguEc$E}e%HeB~&Hif3(aw?oV7Nw~!OeA!UT zoy0(7PQ5TX4#WE!Q0LNv*gV4u^}Qd%!q+>Ar_N6@{-z&OU9z1I`SusztIp>;GzW19 zDo>Rj6tf3rsj%+FdT4Fb<)=q2Vr^Y|)Mv60JlQMFEtK+D<4<*Tkn6%v{Xbyry1_iZ z!Jc0OO`b`+ie?A6GNh$$BF_8^c9EaPg z4|C^}e|dc$1wL_s0JaSupexCjTYi09o?O(<`u@C%T}y^x{)v_LTN0(nyN9kg=U^i5 z>mSQy8+37AsVk4#c@dSh%~7T|l@E4&jKTZ0Xr$p4>ajeA4&KtvH@YiUJQMGUUzdl% z^V^L$RK9`VfBXS^e@=&s&WGs&^E8^#XA+d?N?`i0c66R&#Y2Ca@MC|+QCTlHG<{Tq zKTr4PK8mS4yLKS|C%gh%1y6c(g)SQI2WCt!U(3Ilcx<*$@y^VMgiF#qXiL09Z7H>Ul- z6ZZ|k#(x(*@Y9x-ddeVUCzwG{1RZ{SAI;MZVaVZiXqp zGPR3u7@|^fXz3!pP|WKn|51THiNl1&1MP*EynXmByWE=dA}coZ`W%|Aj;Qz_Q`cEfA2FNGhr?@30ElH|``j+fV;N2fq_ zDwz{O=Y3c#cBKdM(`93M^80Cg!{ZH5e?x(vk%~as^nI+$sDxPuUW8%`JE$C20XISl z;l!78jC!9(zL(w<-Ov7T)AbrWk(Lvg96r1~p^z+C%WzSJGXMT@gQVtc5H6~_g&xY-{J;*>>Z1Vv2BuGjX|`weHI;7W=p$+i_qju51zlYn#Y;U;;9jpZ&LGBc37^r>OX7Kr<|D-qna0Z=YEpGTiN!nUW_@ni{m8p^dp%K6IsZro)5Cdo^M zVUS@rlf>f$mORz~8w&7(dYvJPBJxt+ZB`%sbO7wpl(zz?o zkVy4RXsx#c?>BMKC*?}nk%6Ph;OJ#I@7-ejN@n5l?b+Bewhq^Yb)(tV#XRu%PHv@; z$`kv3!&j3g@*9B6{Q5czepY2EFI}~TkFJQp5b<}dFSQF?>SocudSmHh>03~%><#6+tfBErJ5}}V zrU#cCr>6?$(|xrDm?ZrI&mTX^+FB=YzODskSVocZ{)_O$Cv{9)l+83d+%PuuEGx)~ zC!-#ngF(OElZj~@`L|W*?59V+nWW>L#1KBvC6dX!4o926G4Ni07~U^-;&j?4Sg^C7 zMDAB-(R6j*B4JGEAmmlT(R`z z2C}a>LUedGz>l=10&v(Gg z`DW!|^B!aQ^2J2i+K!rBG!Z6UVvv{N0lyCi@TRp6yyJl$_e>f^6^oDI5OrSuBWD+G zSos;Ni%W21;Y(PTIhGzVDTl6-M(iKEhi7i7;zK%QL^t+AIQ&E)lc#hE89iUb9rPf! zsbD+hhaG9kAxBW2F;bAb@SpzBLNniZ!< zM@nj$7!IlkST+SREss+9y#=Dbs0t>Zi@|Hf(y*kEa^HT3x&KacesW4XrmegH>%4T~ z-_RaRX;$PGyXQ$doV01t!E%_FSnqH*=?5wtd(A5LHNh$05@BC#G%k-?3Oxa>tSq$_ z*I8bNd+%1#=|NH?X@4!>>2sBj96cJ34~S-G8_%-1bIc&^F& ze5~O2&9ZUvD$xn^;w{rp^@C4iqw)93a5q4m}k}$D1FfqIsQ#E36k? z&vQ7h5Z&l5nsh-&2v+lSQtZ4^GR#Z^pFYY&AB`w5>8=7py>{W^^A+qwH)og3?08Y{ zQ0T~=4J}`rNtub5rPg1^O1Ijx-xl?__<#%Ct_Z``a_RV{u@#4&b!XaV3&aj@l4SqK z1omvV7C7vRCSlMIB!V}pIN7j?zjVkP!5jk2`q4#&87ygvEx9o30G6(DhT&Q!<(;dO zASYlf_1(4&RX?iosS`C=&A2NP9yFA^o1ld!m%L_{gFmp^Wh3EMT`bMfUrD`1$NA}_ z4rnw`S<>y{$TmGd>}!4ycPXgz+uELZrRy?$yETOUy_qR-n{I)vd1G+k`=yJep{CSC^e-A+ImaxYY{iV&@92MfA6~mVSLCSoxKe!7fk^p{m!a`m@2Q%UvWupDmZdHu zG5+!(nd9 zpNZ#c&!*>+@1K52o?jL7n0;Jv!Hr9-_tPYPRaQ7dH#5X2&z@@tR6AoRW2+53bl^q;AMiYH$_ai(gOI; z=oLy9-V)+HG+Cv=JZMitoVL6QbFYm;xr|uJ;4!yJLtY=asc+9GA4FWN;eh4kjo5Uh z6g*Oj=?mXfXqnZ>+=|k1uD_k2m>`2C%!eL0*#SO9x`>-q!0>k)8DO5r>K@F5gHuGO z@kA|{bg6?azBmluj-Cb!UUZ=6&HZ?4Mh;hOS%GVlMNYJ?1~FD%N^4Sb>CKWUT>s{4 zto3u_Z-cTxw!b`#8+VB$H=FWW?HlNQdmMY0IF0m+kHg2CM)6q1sXSS7Oq@?>a>w}& zc)oo*9&l_&r4(&0t0ZPLhs4@%4zZ;vW(TQ9;4vb%+R z8sBSU2j}i#ms2ajH9de{-0w$|EOg<#%|2!kX^ve>Z^MTJy0kWGHEc}V!xA(yg+trN zI(W#=EHB-&fTgT$Wm%6J(AuJlP1_~|I|eYA?|p@QwMxYP+Li3pLQB+I+(N!DsbMch z~cTX9ASdGm|3$UI4119u)$l6g+!>es)$lORv)*RYX z-ny+zcsr&A6RS?(95+jvu;DPA=``jaMb_=sSsiG)V<6n>l!b%wR=h5%o&CChg0y!B zU|ygcC@Srs&s~ zUG7prcpr=+4*M<%JGJ7uYQioaJiv;#DBndDnZxDVO6ADSbo_j*3mo$4| zFrJKC{S?J`8@~Ezgbu4$2$TC?z^29zxb9~u>(QUgThqT{?z&tIx!J%pa(78~=>33U z+mxa1g&(!wrq2_ks`$mVQv7IsI9vOrm|f~TUM`(2W(_P45x;47h07L)Bp?69!qow( zbg%4qytZZoA2Z`6e)%fT=S-i=9X`Cr0+ac;VoyJqvaFaKu`g#Y1CL<4-EkD$PNDuA zKdjhkBT+ULxhoz4+~PtTAE27f`)SMaw&3mTOOO`5IbkqNT2@XI^}ez1TAtkb>v#O@ zHxJ#ijQCLVVf6aw>>LY*Z3-WOr<#-FhMMPJy~ zT1bb^@aD%~zUS_48~7129$bWp7|{6>Zb&ZUvZ^<@`&J@;EmI`Fehy`&{gB5y~T@@WtEarM`o<=VY!G+ma`FY!lMt^0dS$*&t&VSX|oR_W?S ziCuwgp|UCDh~4b3SJWZmrYsww+m1Iaui>g)ui)|bE@H0hjlWMB@)+&@!s$2*dZlG9 zvGX2)5#7Uq*m^L`!LCQIS$fL#c0z{O;rTzsKX_^Z49*?z!*tdOe>F zkp`*5)#^ z7wzyodp#`W8(^h!H{ODkICqNVwLd%^9^Q=Po10&v(YFw2cYe(FH(W*UnqxeBZ78k{ zlc$G%8`2EtW>Ez{lgll#!r`)Zp?gpU#N2r=zWS%n288$stLNF{=G%9$bXOIRPjRNc z3tO?qM;rQon2vmp9fZCp5oCf~*`<15I>QGL>+thf_{fB(J+5X^trOs^pB?M;1Y-AD z1g(xV=$2A;dau5~Q!r_LKI&d$AHTkPUl-K{;*4Rw%F-Z z3LA3L+3f5}awR5PIu}!7bstN_w29xOIOhnMsdbe83|J{+^eZ-}_hNVR12VBJ-Rw1(x>&%rb!yn1YmU8xQt*p znybD{Ec1FO^_c9g;wh6KIDz?~-_$L1@m&L|yC@!rPhb98_Y^PaZ^@f_#}l>NF_8Gn z7<7A`VBcq^vp2~py!kSN_u>LtOXxEdo;Cl z*o$GYk>ae&reb=%Enjxsj!))I5-qp~RZNT*p6eVHyS|0uI#Ys;@1(5G9}TKz(}P~= zQv|;s>XILKcJkI{Q*!)$7{oQqfSRKtsCigM*{drPAV=cDxj6dJpuz*}r-v)fwPUbv zmIKI6k(auADlEETH&Hy@3O!OS;DmDn^KOd8$5F#UFV_UWSbryShjX#9S11#5T=Dp9 zPqKci9OGxB#Xg6kv8-hj&+m4GU&EG@|GauLn>ESo(~|;}PZQ8dZ$1W0dqqr|X7doG zQT$=uQ+7`|6|5t70KYqmzK9&hweK+Q<2!}dyN`tOlfj~F*-YptcYq6rvY5A06^3i) zV~Il@GzHxuH+~%?Gk$DC<6)b)g>pLIJh}~^!aG*G_`ay~rUV0;)aj_Y4Pg5wos3wz zi@fcbMple^hci_pxp+&7>;H;i38r_%s7R#vu`m`>|&oKZ2b~_>b7ev`7C=?EG)oMAcKH;>;s#bM@i(1 zgCz9BNO5_N0{f@_hBUY)1`b=9B%) zi;?#FV{^Ny6J&aIL;0+K@Ne2_a>(s4Xztj^SL>E=-+Wivx5q+>q1#ltc$Ws0+&oEM z_F;G|T@zw#PQvhh(^=o~ees^E9Lp^1MP=J|us-^q%G#G!Lr8TrJz^%k8*(iycz6m* zrua(TC4W4WT0%&z4V|*Um^%FDKm&G~VD_9e(MG=)v)MTWcNz@hmCp$uRAK~rMZZcz zmoDU%ha1>4^)%5+)tU15r*H?mCJs?}f>W*Bxn;98pY`|i*CJo4dd5?n#`c?E5& z98FtyZWrIsFi?`%lFw%JqSdQ@oSzd|B^+9yL3`}IDZB_74s$eI$?}9=I?(;~p&%Z}$u;YiPG6ckN=KyiK5e?pG7vgAKgu*$Q6lxrB~CIh6N3 z{*7&Tz@YcGq4dnIR6L)tR~UQy95}3-i6hsCVDEY3xQUB{ApODr-5>J#Q@D>`=gwGt&r54JJdz1cIvCYw_Cn zd%{TFgG^(|Mo`UuCS**p#T9*j3a{JKLB}*2lK&N8ZP0zTJ6F1=BiaOA8=adgWs0w3SpAnx9;&_rF6N1rma_kT|9Cmh*epYR~nzH(uxy{U0G zy{yoOx7J33X6I<4*lP?7U7G{Pr&~~;dSj{2JWpCNrFKxhwQ#^o9>1n!l-+Fc=3Up? zu;*ho?awu+G<(w4(*L|e`w*K?gmmYED9G6Uy8HSTJh>wJv?vx z5uQ69fG3)1^y-Yhv}e*-;{F*}Lw7nP7KE@-7hjNN2Sy1KKHe(*l^4b~?bMX|FYavn z_A)W8uLlhu#G&@XaoD_V5N2PtLzhA3+r6EP*&P!|!l@!A^Wq!3b4$R<^Q%DB zSRNYgrQ_i7Ls^l*aeV)AFYXFT!_e1PnC2^~lboqfH{FT?YsCO07KCqEEP3AUxq$b2 zeSC2G8GLm}VQ-e)Bzs)}YDX@?496rKE%(Xpt>-ZaZPTZ^e>c*<9S_06-4%LRgpy2# zcApzI(5|2>5LdYl*6xtUwnr2#em9W2LOQ;YGGBehIEt5buEPeOY_@LfO3`s^B2G)b z&%RvMf-R%uP}5}>-Et*}_KO4lX*uwqJ_j&#+W`D=qn^lzW`ejKaiHlj%-^iTuI6+= zRD?6#q}c#>r~StsJpUrBK2jt2%qeEUA68(G=~MYHjd^h3iYgJ7G?Dy(d zRN`qB2`>gupurs$v|9+L<{z!;jJ`==cW^aMmomLcHor)L)C1cr?~GIaG_u#PPcpAJ z{a|qAU*>c<7ejRL!-9-gc53rEWC;blg z;YHUk*iZHl zQcg&pwL1+uUllR^Y!@u4-OV-~TMfg;YSET8Bk8f6IV|JGM4aGTA$sL2k=)YX zr8&m}nahtoMDz1kHZarPuH^V-<~C&@v6nLwwSNYdDL7b{#dvzMl^KC>DYT8asn_7L zb*dLWWT8uSjN6*$Z)P59Mzm;Fd<5bMJTx``dwm~|a9 zZM<=pS}(9OZGcJJ4`I>F85rob7GIfbvvAoMxSA%-6Q3#2I*B3hQg*vg@u*xpDh>sm zhb~k_X} ze1J7RzO<);E)8pvW+lG@@%`u#+}1}yTzu>-q@S2V3u?5PpK-OA;!}dQH%r7hV=_nz z8bQ9;gr{#PVnDnWa;-EN5F~XJ*P95VM{dWJ#Y5=t;l}h@W-J7}J1e?88qV93J@`sX z6~0Fynz-@;^3hiwd++MYGP~vI5_5T~ZTnriKd0cI==GTL%NARTJ!sN_0{Zi_6Mehy zG<`ND+dgT0ru~;M3wV4>7jCiIgVP;P!?lU^@L$Gw`t|TG{^QU*+`HfonLKeH)hgaX zOWPCaQ0EW~jH`w*OEjsxk2S<;9)QLnt5EM*Gtv)Bx#$;-{j}WS$!7!FyQkFU{N#uu z0&d$X?Q|lK=h#rQ?_DhISq8~5Fc)V0SkBDn?gEcdQ^9p^8}kz@G0&@?%BBu7L6?WlSo1THxvlXP z*IBP;>ubhCQBPCY^kW2E*mn|#2t{O2+FtCb_E5BGxC2`fYs8#8=3Fh<3ui0j<6(5c z-dz!bg6CJZ*QkrQ`h62mIGijyao?TfogBv}AGP4KhAQ*l%d9Z)OMld`?I0dYu91Ur zQ=s0?h|aGqMvF(9xKch*ywTc+kM$nTUkIJ(r6zR-Zd@kYT~@GR(^oUsFS~G!ja~VZ z@(ez1eGE7CQK#PJ?x33D%aas#@r#E#(4+e>ULD_yPne}yK6}(3-koAvzW!tw-#uax zH&L2Ll%%X&cy21{ejCB>+Nki}E-S=gN#XFM#)EIK63}*-A`E&l0r(`zNffLhasMj0 z&f-Wu%D*R%ka%;$1PS&E=M+3Kg*9B*N&UB}fD2$~1>q2G!=vkbpJ z*z0x~HhxS)MPoOdaeNUeY|^6-ZhXXyGtyj_+`&2h@^R|<-I(%7lUM2ty(av7XFuLYa>YG0tHFy;bg9eyaJ=^C6eRSu;ahwiB<@8c#&=%kw`3&7 zs-+Sxc$`d3Li*uZk0f^eM-qNsm;wg1W=M4(v4fw7VtBGO?=+XPIX4&b>1EjN4n% zS{=>4Mji#Hy3u@itt$UgmC07Q$uX;#bfG)_D|xdniv93!2cIE#$Y$xhv1F5qU4X$@ z^cF^=dV3jGOJ30pQ!3fz^5n9epIS+j_Y27r{T&Z3p2~BY>v6E}115Xz9FAEygx@cv z82#P^!-v(D&5W|ZFMr)tAv=^iDU@^5gfR4bvyf|_SSQFxfB(ZbfM#E5zz@wY z*pc1|62YqquhRF<7u;B26)nqsuN|#oi8gSk%3q{QSc@fHFxXtmFC+7-u{aOK6-+WHz8ge zV8uPt!tnHr>qJ@VcKtZ0%hh)W@a=O3aDA6jT->Heev}_1F>kh$e?e-Pf5sFyN3}Bj zhK05f+owW_brPRacN&MZxnjnQE;dQ{BtD;33qE=$NiK_mSL>pn=t&ZKK?G5m-p-um z3h;-T1HMSOj56^PFsJchS$Wbaq0M&`~-Zw(*t~T(!*<*Z)dkwkJDTmZBGzi=Bko zd>GQxS6G$TRveytmiM6sTyy_FF{kA^iyZK$Y}NcMpy!8E*UfF*yy*}3Rh4{OBaQgl zsYAfZTATYs)?jy%6*grW;VQZRgs%r~!M{26(COBTpOJYaJ$r8P)JcQ6>)h8k=-6;} zbp3bimA~9@FeDrAIZDbCZUY8z%=CgqV|UYrO^`=%DrAMET7foP(J_D zFFxm?8()@F4EDE=qxPGTkYwf0d#WK%uR6u8I;Qi8Ki+WWKqGiu?ahxJzk>G)C-Fa~ zPw{@lZ8GGJ8O9y&B=i3|;+~UtaQhnt^7oSyERC{eCZXq~eC^8e!j+G>FskbR&!>66 zM{#p0i#90=xJM4f)l0vTpYI|e;=^3D9O_GCkKbeBk!4sGydOV`pXfHRf!>^MDp9u|A)16A7lTmgx%?P znj~nb1GvA2MC;w?-o1}MdT@eYo*T!z#uy62mp>3E8f0;UvuAOo-vF{#QH5=qY%7er z>jIPdIa3Wmhjt8i#%bk_bW_?MmMBPUkMs>NwvXiLyi)`7zHMc}zT+VCCuI?nRQTJh z4Y*FOgXwOaLqjDWbPHY-`cBManpeI+bayBXslGz*zSV}Zs+E!>&xw9>)};O0A4pC& zb?V%hMUpO#A zNo7B6U9*ELH7c;$4I?q&{%h!*E@OYbcsYG`?=)5VHW(fcwzD6zV~M?Ylbij~a247b zGlnhi9fn7rtHZ=?C1BpW0n{6R5Wfe{@bC1s(!00@Rk#;Te;@XRVLg7q^XN)AQe(=4 z?yiB3K~kFe83#{R3u=k%^w^U@{w#(+{cQ39zwlEx=i83UdUW{2V}ge!7WCf z3XUPXZfh~V`I9@d~&%EwTwf*zZb#(~0IHp5y6opbFhZyQ&noF%}q#QzS6dt-bo2v(nM&tV`{GQ|w z{Mo&jo?PWf!(C3%vKU+YjL1KBmDUGoxAqlueien_HUZGx z8Vj+05vRBJo6%)1sq~<-CLQjW$DbFgqR>m38@{)Kr9pBM%Tb1pi>im`vyxz5_*}fM zGl<&R_p;ymXAh0Aw(_pRWAG~ck$AyLmQO#T$e-P57KIab)NmBxRpHCP?Q0Cp z(7Zvdl*?)D0$e&#jU!+I%XPB5ciFd`@!IGPC3@63Y=;ntsZ0$`5j3~!I z+ghgJP(j8Ic_rL8oJUP>Jmy=T_he;nr_iX9OuF*yTRNzrhMrZQN9Pr`VDp$s++l~5 zHN5OcS0(+Tf9wP4JfmH>RDUt;XYawGe`B%ovV+(W9gN!f z--+`=RXVyW8wO$kUDr?#-P2QPedj0YIO;VNuO7w&rXS}|9gFeAl^vql@+g>7hC;`e#}%`^UR4(dwHau%^2iU;7$!XLQ1 zSaUX3xeR1Js}*>y{8gM$_MhFT7JqK~{w|*R+@pNCP9J(e)wHZ-&@S4!(Sy3h`0$%= zNAUL3kJzZmm2fmC89RrrzzSO{3^VhCA_Pt#bj6~0+B=%U9#{*#4goFn(xIPx_Ry)_ z?K)CC(_zsT{LIZYI`UB8TNM! zvS>_JKYmG6X0Dot;YOW4jT{(7(|c(R!zK&3#BPE5jDJ9Yc7U+SU1T@-8m^oKlpx`!&wUQ4&FmZKh9 zmeRSxX~-$R0DsM{QkBog=oZa!_5;T)MXi&qBrGLST=&BXL;PCAhQ?GfzHvH0= zi%sa7*nPCJ=^o6kKZwE0eldT!A~x&(AE@~2Lb7(rVvN#HytzU0Hf1=_5p@^n&GA2} z^TH>ze?uI|UkhaS+VpYOG*5a#!-ck~O{Ts&djw|YNq;%@r^DD-dNwzo?r*(^uT1CD zK)q1vmivYl4qj}ZaK*{~MWrsUZFk_iHBRtJZ`bf3D@9xzSqLAr_mF*e58(OzRy^$8 zLAIr97_Pc*LjO#3u>U4M3BuPOz~0ZbxK{2^ZZ$s|RF`Jb_GfxL#aD@Xtj?73pXmF~FlYKh$!Oi}%JVEQP|HvgHlU}#4b39SORG?NJoggqa(qp1uhX#)@KZ$LC2+-%gJ@p3 zhiT8-hG~kw;o^#bvhWX{+*U=MKf1{Isx5PP`He~Fv*jjxJf~PN$`Z^W;Urt$?E-}zE=2R^Gt7h&84NE$yxTzh5?f6U@|R)8IsOV%PW3vOZQiyX3h zp>!sv!boN9G_v!$CIqwukj>MM;Jk4OB+l6ypPW37FDj)eVLw^?G5i~P{GNoNi2=mc zCLKGoF5n0$BRtc^0G;A`@=C6W@}CnVw#R1D-Xaf1=cJyQ=2=jF+XL!KqL_~394z%Y zjuj?mykq%3eAtx4UC0-1o1noPI_hxB;jgUVWf7N8JI7~k(c|Nldf}$O19`kcKmM#b zpIGlP<-5Elp-YZDkL=_meNGs^zHJ`-ijQU<{bHES>=E#AoG%-`YEAi-9;xL`>%z+Y zl=aGUtzL7zTUzC_7lev;#u)LPY16sf33b-~CIYSB-eiBPx3PrFdN|x<94VU6#wI%V zKu1|iJW!d)A9KTUug&lH1nUF*Vqg<)bB^VYo?Eh)*%>I<>_;bweZ1dufM{Rq%KEO# z$5kO?p#5um_vJ+u=?&I>DqLLC;NYh zbI(}wVKS%Ko7c;+Dg7FJ@!EqOF<cE$JQ>*8mi&bn(@K_eL)6iI4i2#nX4<1T9s6GPVt{FLSZKFmiK+mGq-a))9J zFB#95D6Zr8xg)nY*N79=H{&y98=R@Ko(~yNfL5+LeCF~#<)ilvB>&{XiP)>6Z0*^x zMBaK2kez>-u5C{&4*p1LDoyw@)p@*QL6@A+HS0P+@O|-5*1F(P)hy4qCStRThln9bzUwxx@~S293ehI;;7Srxtv`qIL@h;lQ8UXqU@;d5niV5!IcQc)s>PzFpsdEZmdCf8~wB zBNhYrSkHdtXKzmBsovkPeq0U9mG@>AX`501^#ihHuRBIRR{#Zmn7vYU#Dd0ZyV*Js ze3`h3PwD>&&;I_1>igT!Hf1FLkvvpz+M^Bj@_R(s;l+Anda+&FT7rYc(bAsuF3wE| zV~w?US!zcBi0k`{2b^WGzV0}@9bpeWt|W>(7iE!Euni7rv=N0Ll%3HqA@k=i#2}pn z_Ote?#3yM+bIBXoHOd~mi*~WRjGL_1!?MwaLve^?mxemorbRVI0qH}I0?S6KL32J1#O2@8j9$4ft}@MQ5! zY~A1orIl-N)6Z1Q)!W7wd2isOUuUDv#5iFlOA=;(GKYXyZ*bs&W4QYHXA*zy7zoca zVDA%cY0&%EZ})`mW^fjU4%XWyy6uQfF#CiNWVtKS!)mRbV0cCNR202O2U@@SYKjXL|N!$*K-u`8I`rdKSa^zP z`RYZnbfy8WzVe>DE1v~_GL@w+g%gtY0HWcm4{m`O(t4DQRB^xXY%8iM8+TWH#^rjLmRnz{;UB%5cm^ zvv4{0eWw7C@{dUB14~Y?wsR7Dn~xKvy;X7s#@xk$G`Hi%3T6vrPM#*NJSa@%40{HWMda-`kChIJ~O2M^^NFB$NZ*9AQC>Ks08)OggN6^hL@ z8N6r74-A+!jGtATN9O!D5NvMVfq6=H?7*_GxGHfF|2FFxhAZ_XZ}05HRnk?a;-lqrxJ94S z*fu_gYGwC^ujI16e$UEuL)8D+{;>f>b@69Ih zcK1jaUGS29&Zr09U1m@-Hwt%aMzAqe6->$FIZTbv!-9%9c$?A-+&5w}k&M7Md5&W&b=A*)IU zoVHEIZiQQ{{-GxcNj7E2`mKY*rPADFo--c=S%4QDwC*Wn4v6D#puo(Am-nNE5;Po^l}Ej{V$sg8878i z-lW4^eTf%+DxKc?a*+%=ZAk+{;_0KTbJXop1vA_3K^B+(B~{xsvFXwa_Kz2o&Gpe2 zx2>~c)h<)Xf7NSX&%QyB)X<-vXm|)qik;}j);374u%Pc}N%?H?4P=g4k2gw-;ZNRB zh?8;_=4&XtTtAs+o8Dz5&O7nR^}eXJE{oU(t_FvqbC8~&BRSKqu}h}?pziS*GFID< zyg3&E19bGEpldSQQzQuPfm+a~|87FowL+ekh~CT1!~qJ8>{_A{8&{Xe;>J>p4_b~_ ztE1s!&1ksxp9KyK?aNOZd=Q_={UmE5=CVaCzu3+dx_Ge2j8$qKrGLDJ^0&iQvtyGg zVY;6J`JgyUyzcplG;V4W{ip0F9TRVm3gse}8`_N%k2VoEOBEP5{Hfj8T?6rVKq^&| zpGzO_{{wTqrZWr83q-rqijwOyCBOA0GU~rq!k@(##1U57VET0|ROX#zW;46lK%Y$1 z87+^mdyQfF@_I~d?=$gkiJ{oP<{7N|dkd|D5?G_XCA-8zFjk*%}amT{f! zU#|qaEw+$3xB6kmpLpcB`2Bt%V@~Zu%Lr3$`ZBz;&EX8@3K|FGW5xllW=1GONhLAjE$Rm9Q?x7XydYY zZ2FqRtoN9pp5wE!kLpvvy4ip}->XRa`^-bf<)Om0-?w0<(_W!wt{mKwGHzx$!%;da z6J&<2VH)+5Vey74*e&ZxcE(6Owu;TLKo6SJ&#`_MypQ^=%CtFh~bH!p=a-Qd@F5xi@|5DeVnv66u_~_sMf=!nCs| zk_Ac|*`km~$ZNG|ph`aboIaWKpSJ@36$~Pc)-%EDk1=kLxNw*1I$&Df2~cu72o`N3 zT<$-EmIZtijSDWouO45B?zu6bHt;&+3ljgR?6Ig@W(~JJi@^W;QWC%NI{eq*!^Ry5 zhUNd+k#kq-Ami>Da>Ke8&3UksR?Ye@+P*nSx+4v#a_cwtw_ROgXepD@tE%MeX$sD| z-a_`m6=ZWkfAn5dMC#2Lsnv=^?I&`ma_0f@cBo~q7aqrj@|FA~V0{*>mv97HBuE)kzC^(M~Q9Q-6%VMA^Barn{<2Mg=-_!@m#Du?!VDc+8Ve7 zY6hzd9rPBwGmZnLxIprLn+|c`<0tCYCc$WL6P&&75bK`hkN5p5F=N#=Qr{{HM;E?h z7vlr*^0H7^VJO;VO^qVAqJ|5ThiQTTjC5>py8_v9!&sA!v8~3ulkBVUGvcY;Dn5-L zM<@Qb32HQ_h@tiQ!oi{UNIdrx609-iupZdyoEyBgO2`@wQ8<$_4Y_T=q_yYpJeTgLz5zqR;mhVZk9861}el z)}{$y{Bbhc{z$X^t8_?MVmXrZ+hhsnFO?9Nlm5cm&|+Agz7ST%>?`xC-2{UgHEHt5 z5inuOENE7~2xEU=1N{p_sdjEO@w;;rx5RnSxF=a~=TU+9xo8!W$#NHL^`)7Z)LAi# zc>&+o^`!N+k}Id*J#u)ZHl(b7BpN-m#D5wYkFd^l!Go)_zYDTo0*#2S z!h=JPVu{&|(({9isJna~ga#?_6R+~w!7t-rnx`t=?Y9jAoub)By%bn5JXo-45-@(% z448Jp1IE|-lc_e_V6~bhKi^js&y~cn3!n^JU6aVf{r6$3PBxk6szi=XS->PZD*8%{ zoS3)r^ngyYSa#?cQM7FVCvhe`t&*kwSE3-?ma)~*v+0V>A=IjTs+6HMhdZ7tsGUs$ zWQ!)?r8k~ZDU<##`5Z9|^918OC)(iZNZ0LoDxRvt=-nLbh4`?f74NrnfBc@Lw%*%o(Rqqw@2A;r|uIt77^P@5S)m+rb zJ_`fS6iFG^hcHakz|1ZEx!LR{Sfc6<1L{Vho>vki{8>!`{0|ZvaVj?dlH5&w1`9`T zX>#j@{n32rG`7NLw79?Kw=g%zlsBl#!ds7MyL{It5Tn*l{HNr|gQtCgdZ}O4nD$zb zTRuy46<6Xd1$|mu<_y{`WiWlxSu+239vsLirw=yf!}{T|WT*+liQ!9FiNR;2cOr3q z)i0K=)hziCPh-K3XJEWfi^}AWp|s*Dwfg%4WuA?tquaKiPpRYyELg}|DwhaYmxdNE zj*+LM%!#MOe_CIu%2ac2lY`#MHP zL`yoF*-GopxW1bqr+Yp`7#=9gKd_&)&7LTJRq4xLH63Edbq?d1x#`q+=RSx@))Cx> z48RuyCo(M+DG{)5H*w3%Ba1ec!nlRgpxw|0PWBrPVeMgbN#}luuU|^87$(w?yk^lb zs3$)V-poLVXU#@O;L-C)dSU58bmf}-rtV~^Grtr6p6OvLRCGqRh8<$DfZyxhHri8Z)rg1Y3 z5BStJ4f#x2zVG~gNPj;OoY%%Mr|=&9i%&huR&=r2)rGJuuaflWG@%3k8$i{++OwjH z1m0_7PhK6$!8Bk3ST9;bubZAhwP{NH;phoCe}W-biFe_-MhE$~)f5|pU*e9Vvsn9* zSGJK~voX%J4~5rbXxG#f=A(FtWM9s;>vhW&osd zMuW;uqNTPGZ@s8NPb7rXag96p(Rr0%b~a1=+LA148(83_iBTj&-;vLasSp~Td?ibo zGDNk=lcL!hJ+2bCoB#At^AYogiah6}{*@=LHBQHgq;+T0%> zVXu7xpIxSfik6?b{|(6n=5&^vH~0ZjWw!9+@?SPoW*GTB=PP^dDlrdrOV|MQ-{AAt znOsv#gne6L>l3)QxgmJ1}nA68^SyCe=^#qhS&C zkf^s5AHTcK)+^=V5w|<+)HG?%F(*NA+%jC4Jy8$al7EuEJ( zfz!2}koDGwHV=`7j-$Qs^_^5|-SG&TzxzNO!OK^gwDiiKv|m`+u0=g|-*C zw?`P18lA>J4~9v6|6gp_<76CC|A&$I6tQS^Dju6~m^ds}wblHa&0@X_7&c-jdGFpX z7HeK)*{RjIx2V79dAyM28I+K_4i30lVn(~}ypI|Wtb~=mn=n1-lyKPTA!+gNgC;|I z(BbDi!Dhz+p<>5-T=BO|;vH@mxAt?z*v3Th({5iuzBC=nRbG&&b{VD?*@CB51c@-x z7oKgl<~aemq<6#=`d`E^lsivB>5?IIMyiPorN?;65+nQ=c~Pt{vLV0kE(NXZNODEm z-)rPtB*oMIBbIwFuwK?W9(MXv@tG2R$vMkiUb`J}BYg3&y z%IH@yT|BhwG?p)NXrJULa|0fJ&LHM_>02Xd_ zreA~|JVUt`RjX7-W6KKGHGzn;vZQ{}#pgKr&lJNp8IfDTYq@8(fYG^w+kFR#=d z3y054#$D-C(b&a?yKant)Sox)KE60Dx7Rr zo`g>h%%dt6el+3bW!zAn!>a=TzHWF(F2x*Y?S_RA^JOt6PgBDg!`$hHXRqLHv^2hL zYKE16OYNRKO(y5wHIU8bK`duy3h3vp033Y|gZd2SKIb3dqVw13d9x#Q_t#RmJHQUl zORm)cm4#x;(?KwAmIf663}Tzu0`|bn30JZ2WPI!c{9Qd1XH!kd;Qayfpzy61^|r`>A5EX&y23L4^l2}q+}XhrAGJWO zqQp`zkmK+AB=XX)663f$hpdwA0cmdD@MusGLii$79<`7xh-o|LW;qUwcH4nr*)y<> zmO`iR7EpFP0LE<%f=gq9$+ze|cywY3&N7dK#>__6Gsp#$_L{)Wwd47sKnL>0R6w)U z2}HGmK>&Kd=Y{F;JawR$WRuRCzIem6#FJ3%K8-ZRNO{03d%#Z0P-`arWZs8V>4251 z>Ba|y>YOQr4N{M3#jYmu_p0Plm(`$_LwsrI;@Q+hYZbMQ)8YL;#;_=l8uoXhBh@&W zOT`r$wDPeQ++Qqj|0s`Bq30dZ?_MpJ`yWN;;fVG3#c>(gQY5ofMv08@+;a&@GFnQa z(%wnmGTMcPj1n0kB_bM<_1tq&(vYS~!-|wNl#+&)-{+HGA)ICQ17v(W6jHO`rabqC+mi_=Mrh$=l|%ClNmJq zc{B|zT28MykLFTI@wlwf3|8l^BehW>_$_lbmSs)F%g3b1$r=9Q`>$NY9j%HOEO7h> zho-REmSgCVhaxKJI+SWns)J?$d46Ml9c+z{q3S z`o7?Fu}B1|8;-a*Z7S+(_5`R8aEXO#d4gjV?b_$q}VFIBdHg-IjKndQoTk zPe+em9x8_|zHwar@Kw~AJ01@ie}UC0AHYI)9Q`-sA_gA*1-F}&slv-FdcbKl_3fR3 zA_ZfNJbMVHd{v@4Y%7)1wC6>}hP?KU6Fwa~06scfkvn(R(F31mQ|s(-n%2q~ypZPn z>^ZcZX$3)B7ePtp2B=(OhVt=t)XS@uT31Qf6@9O!TN75pl3N|jBs`pEsrQN2`^9ma zRkd8r*MP@fD@M()i^!?CK{WNG9JQ1jNmpLG!QwKCK)HWAd+_EWtW@6$S3_rt>z`i6 zeUp9A+BBV&SMR6aKYpZTyDrnmZ7_X_x-znF3=8a32E34OmD%(1M- zs_4h!t)d>%q0%WhYKKAjs8Y7+rXn66Zj2p+E#bxtW9oW1mPSrr4POg=aD4DO(Yd@9 zTDVljE+k+!m9-fGNBRx;gq!pDRZ~?WKB7(?$KN3<)V+9O#70JTWs0lR5^>+T9+KlZ z9^5sXp|9ta_|VqLIO1u)Xwje^u~M}O{t8ASA1+T*-0m{lSNgD3N|Mf2TS0$nJ%d~K zCFtr5YdB*+1lFZ^vy~yo$hVub`4pKZJRcvCFEOzIkP=!i*31qTs!b5@iSBe-@F)DZk~b- z2Zuw0-(>up9!1+9bW_*$-ppyg4ENg@&41Z#C2O71=n68}j{VG|)rsabeCt;Jq~@n#Fgni0SGX7uyRn>1l>(~6X@^uUnO zc8j9AX#TMrG(JBUmS`S=X!(kpRn)0 zI$1}=S}@8UjU_uyv89h9MKZrC*~^x4k|tpc>3yrf8pl!7Yr$}%c?FG`H=mR>lrz`k zL+HEW6k2sCi0-`K4ns7p!N>O@#C2_?u{Y+@^>5UvtED`}79|q&zLfTbyoNU^8Z^`H z9Ba+cqkDgf%O;y@qExag{TaO!I!8ovyA5)JXXGf3v+WmGmbgRwwr=JeF3$!FjJwwM zL~iH)UF2DG42O!mj)3b}KsOR*J^xD^Cbo=O!!zPQtW!)&+dq9&LOC3jFuah_@ zF$h??>#b&0vtLNTa{rT%kRZUbLfsJX_d#h-Yt^jFtf>&^t#17h3sI zNoiY}yFeFynAhW=L_){D+(WyR7E{HRk?8Ha1lJ_Uad&qG98@bww;w#lN;H4s{Mb!6 zHK~vsja25&6MmCldCK&{?{uD)r^0u;>5JEzZAaBCx$??lTW;n&3{#_@5>JCz4FCO} zmi?`#z1pW}=HOav{Lsj&1n*j?dM^l;9e8{H0kt?MW4CUh5uIQ!!>$xH5c{PL^u+sO zy1rY7UdUKTM(HGiJbefy7s6oRXG0pU5RHR3o?{2f=b&BZ0`OLe#?40A)OhlKnqU=9 zaZf2*?;)b^PI^)09V003-wt}8u449E!LfX21Ml3I!JBj(xH?tikwvGmuq_HA6|Yg( z1yOb)y%@XR&DwS!g|kfl$w_cl3#W$%uAtrbmr{?uxm<0-b+jQ;JQcRnWA_%*31OFT zc=%&t93P2;SRJbUX@mol$H0a?YQkMyN=8WEf>jx+FwAcu)ftmacd86$Qr{i8=MOOs z{iV*!WPqhd&qjj?39uQEC>lFqBV1oEWEjreB=gPIkqJi(u|59sA0=GDPwX77UKYxgI_R53@>tIOk3EDEfj~;3&p^Y(=271n= z?IBsXLsWqmFO`d5Kll$-n!Q+&Z-scpsZ#ti?Iw{aenONEjDUxldi00GVCvD+MLxbC zLYy)+=;K|fbP+#F(>D$P)d#k?b3imS?KZ}9uS!VEB@g;@&t-D`Xe4AEzKj8e*Vy$l z74Uvq9Ew7pljIM%5j8G?^D3;VYRUcmkC zPwc|bbFk}`JT`X>#_wai@N{}Vn=UhIIhJlU`Us1^45DS9Q>o{jPat)EJt=UHVw>Nsh2bj=sKq@ATJq`@6n9ARi|X&- z$a5t+?jS6=DAvDb1MsrK!;KH>HIH7qJ77J|GIhr@%-o*2u z<2;9I9a&2(hXoV;XA+Qh(t_L1S`K^TXFy%=jdMr6@_P4af^IC{f>B~9C>Om~m#F02 zBFpq)LeFyvxvl3S@Q&?aRP1XQdRK5jydMFFzJ>s;_Jb?`1nv#3rTxhc^umfo@bn{v zAg>pYuTuztgVLyUm$2(Qzf;)9&p@rM!F>Yd7iAqCZVAen~QUyy7Wi-m{6vUt#WK=!dVrg`(UY8~l)vNBp&tM9!;? z`4!{ysD1AQ@@=UwaMK-F32A8ZPy?3>EaPi8gnVDTzOc8k6hhce60_wWd+s4gZHn|D zlp9l(qj#C^3PV0K?+liuNWO*l%Pq<@+0TbIC*;Aa zRtmFAf06RO!?5+rGl7pf2Fiz~!>mploNf6IZRQQ3`ElB?GP4YHwlu)=3`<(5vX36x zk_AB>4p6Z#4vLFiG5S*z{(G&2cY2dh?O`ih?kf+2K73>*(W9~Mai)0syX)fi5HXm? zorA^daX7a~MltW7lC{ZYZrb?64_-8rASdVfJB z_XQ|jxft=(5iA^HL$5Liy8D9`oLut=&wodlr>;WlgnfsnD-G%E8;2n~N}fCE*Rq(F zsjTISAC0y+3Qh61nPtE=COM!6n%1qR_hXzOOJH55H_gI9ZlB4orb_Is6FBM@7SU~` z|A|@bB>qPD4~ezsMteC76sfLHnvwGT~7u4BEPv_-?a+hewyd`GC#h zg{FsC;BY50{E|MNpZblZ)SqE09;Ib9dyoy)A5Dm!1-%=62!0iJ!+=)~&*^9WsB|&3qF1@DrIj{6AQ0q|I_i?8C+H z{CVq?`_Q~Qmn>x+bUO*4!@XQ^giMyW>}?EP_n(kqn?H(<6W*FL4jzWc0)KR<7|t7y zn9#rf!XW0eIejspkX$Iv#7!Q(7&!G6o|rRKJVA*Ft~GbkJ?pOMQF0Ghdk&^mlP<#z zqYJxOybG_M7=!NIA7tIND57QOMgme6V$X_J zDEA#t`=8pdWGy%rY>v(jk?IHWjULOwfQY%d`v=E|}- zDPJ1LU3-E5?iGL_Ifcb2}co!W~`rt})zFDs)p zidpbca|(RbPyy5UEL?SPA}?7P##b*Bn6TOjuz&s{8r68)HaMx#Hu9yi;Ow%XfA`IW zY&xD+rz?~><^&P7`XM;$>Is-uQVtT%$AFxQdEXM%{wgY$&SU+e_3GE*v1kj8bj-pM10UQ4>y=4K+BeF@;fGnoQu9j_m9!A z3*WKDF1-69RUREe|3>WN74tZ=ZyW=@L;b)?_YS$?VT9vlys)A#6MVb9==i-W=nsRn zWSNkqn09d?{{H$Ky)I~QkG<1*&LBl33%ybKUoFI)eT-4=+c4B&7XJB@1aIPx;bDt) zm>(9->fUAJ+7(`?(S8+vq*5H~tSFG{c88&NA zhw5*y$!M9pQrFtOY)$M~pexmIa`qXKW_l8{D9nZEbETsC^9buE?tw!FCa}vR9UM-| z;I50;$s5a)Fj-!Xs@BEBre0NEbz2rTHT)9C`Y#7l%NWR24uG{Ob?D_+2Ngp9D(cun zxbb{83+moP#~jR}Hm3w1WAR58Tq>L^#=~d;`~cenk4T)v0TN_v0atW>gWEeL9QY$@^7ZbMJYM%*jW$x@R7d0pNl`s4Z(JB=To=)Xn3sQ8PR zE-Se}cE<=kM|(?{*LGU8LPZs;_9)nj!*Zb~p&WM~X(lU*bD&F84)Przfv(~nFxSgt zJZ_-fhV(VG@9i-DaPSLQ?U4_Wa%S{JTju}$kMvHEHt4mTKn3F@c(5Mv^@Fcyw(t@eqGJj#R)vY&%PZM}l3`%47*7-ob%6Ps zqoZho_~^FVxc!P4v+{+!`=t`};}zt|Pi-C$t_AYthiLldg=o8N67Bh$&95E`;lp*k zd7sr7u9!ZIUTp~n8`&h?uyE50s9dQ2q6zZ9n z1;4W^(MZ^JHgTE@wmGM{*l`D%ob$m3w|Hy~t}1^`R95UTNguSFhN5!t zZcL4FBWW#{A@hD8=^MHhUeiLVQj9F$Lp zhR2!$YqU3vkwiV4^Ya(ee}A_C_Ztk4YtD;R<$uv9o+s(MBdXM2+a1&vh}qLiHBg~9 zoVps9!J=(5z}4sw^u#m~2hBm`z(^%}<-HFzX>g(QoF6Y;g!wwH4W*hW%1(OAx~IkzXQy@szb;BJqh|hW%wtV z6XLtG=CI!U4AJZrkuc+X2P|;<$-*K?;L$=Tl z-L_PMeYa$ym%s8R-4Q%yo# z{B5)Py-DB9WVYJ86xR>AOQv5pWD73{ZoLua%vW1U$Uz*3sGdNa8?%iC*lZ>JC5Om+ z84F%BA%?$hFXm65-{xLkZ+Ym`rTpNp)$~w(6um3DBFeQ6p>GTK(LRB#wq!>Q_*y;& zj}Z>^bVVFJI^zs{U%3sY$kBNx6-yaF30c@aAy zZ}qsyo^Pn!$s#7dBhy~gil^LMDEim-nKZ6G0Lzv<6`%Yi_|tCpktlHj8)6fSH80&o zIqGe&ZdW|SeUBs4ZSrAb)m!Xvo*}+nCDvMC?w)KWWEA4_}Ue9TVsBpvn{2e$fbn{O3dVyQ$R1XfHdlNRstS2i0*Sx6y_@MPC8-gzMrtsD+8RS z+~RNcu7oL?wlw0FG@ahU*%~b^*cFzD?N-9u(V-TXf4T$e3$jVwqzBMB?vKE^jfa$_ zVJtZ#j9q-IM3c9Ug5}w7*hEuzSX=V|l1CPkl(T!7^zPBPxlR)7TMDu0#UAX(8u8T; zGogRwEn?xOPpj`I!RXtwVZU`4F!bVxW?wN) zHg===l?rx?zYVn8a^eV`b-tH$&9tPsFRoKr)j9N@OcZvXD@W}Pbv|CmXzf$EMH0fg z;r5b9>YjH?V1U;$?KXjHC?!F2lOtgL_x)sU)?H$MKMqQS{T}zPRp@K-k2Tf_jK-yN zg}QG7S)(e3{qBX(6aEjKtp)E?b&*IeB?D?cN3+{~LGaGr93DtnLT#a^*jVT*Z_lWK zq8Jf%S`td%FFhd^)pcQ`{(0COoXeJaJr}ustPl@#V;a0&;W`VFNGdh8>UYEOupPKhe^g6 z)SPD0b} z#{MaGn47z_Ol?;k+BQeRd6OjAWgY<9UoFY+GmNYpt_=pSJ~031rEHRIBgQxB!oNGm zG5XwR;SJiv25guK^RmPw$9^F0Uv~mm6epBfC^v)F`Z$4a)P#3uq>3KNt049#GLuAe z_D@TP`uurG9M8y!;DZ^JU|MwBYK|{xB&T+|eEt2iq%yMzcKvN9vCa{Y*Cfl&9i7a< zX z(-5lg>zsJGSV}B&UIn&56*GFIgn30vAW_o^?w(C1dt~jA{6@{bW^kJsEK&nk;WU2;cK}^0@d`xUU*O$}SbN6nd6f%sy_b6>STF#v4gq z+A3}ypv!Tsh;QFy!yjJrpm!&^(C2QEtn;J}d)lIpFE&Wf;AMXBN=Aj|2ONfPE?1cP z=|Y_LC6YdMsivovY^T0k(}B1##zF6|mU0QM_1lIIQyy1gZX;sNlH^w(b5Po}04(MnBjs zj-dr4IB{WVb+;6b*4xL9pNR&UO_CzHgU0l%PXd*=6@k@77hpk)H~;u)F#bJF=&SR? zLHpZ8y4Yns-R6Fqx~SiP#K|AYR+VEgPIC({@!7~NKhEUpJ3io-@)*e0UPSgBO@Vp) zXY+$MQ^|j`W?{h2n6irE1nik6iCg}K^QzBfC~?VUI&kI&=onz4I8Y zPhEmt@$nc{r_JG0JRdkTia)=n!e1<1MMC#z)5b~fVPB#r-CFn#CaqTC3!2BlxrVV| z-%Fw8O*y1aLf9ni&|V)C0MU=+U{=^M@D=9nk^=|vYo3LCuHY_zW3~}X{_7$gV`h=? z$^pD)Q!VMe;lxh(^?_-n6@U2hANF_6;O}jf`KXBt@adye2pLlXGS8yP|2`P;1596GKHEu4mmZ)4`V#5zp`k=E z32XAL_8$oo*lcO_W)d4fn#*iN%?%e1PLw zuJIuQO|ve8zl{u8UF-sr*GJl3$o9d`&(heQHUriFlO{)>JrsH$Ht0Jv1HYC|MX!6M zMC##4lwWd@9BkOlmTa6v+|K z%ao_2PUnMny(9OHU1^$46m+TO(u1GmXkl>}=>C3*>jaiF)7cGShxS8k`e?rGVJSYi zY{moM>GLIDvjn=eCOdgy96b^KkE;KFm1ziC7ixr=~_3R4(fZdBX+s@67~!pt_a*+&_a} zcT|DA7G=Ae9T#Z9t{A)Y#j8CTUPHi>RwTI&Eb zO)bzi{twfbvj~)94Y149hy?`A!vExy`9{NYxIdY|?e0rt^5|6@x|VR*xQP2K^+%tO zO7Ya*0+0#50gA;5WKh8(8ge9z-hZ@@`)Ql7GXGJyy(pWd9Q@46R_$QDS8uTB-KMCa z-w)ldqsVRhVeIdrNg%6#1MTa6qwF_Z9(l|VkB@4A5>Y+mS(-uo6TzJq7{E1^s`=oa z7T$csi%;7$idU|hMUR^~lir7x+_*D@hKdcq{(Tl3xM(WZ^7Z3>jg>V0tB+m&i$1C= zWeYW%t@)IJ^*nW%z_zvSLhU>`bZ(u?rIN>stuGOnta24~b>i8B<0dQCbsp}#~ZEZ$fb-$@bl{^T)AF}uGuBWwYO`E-IoenmN|cU zwA&Xp`Je&3xHS%o&xOIAKr8t1+K_(iABV|<(&2#q4w2-)B!sUDWvgl>xb4dOq^*IH zUWXWVJ-#0u z7u?c`c}6+Z^O@k>eIUo3yGC-~P7S_4V+csTEk9*Q>;*1JVSF*pkXnLc_#-&JrU!K2)I&*34zS=% z8oDxjxQ4i7mZ!i@J^)|V>6%$_8L%)V1T^D={a_!_P_}9_}e+>CsMPF{+R?L;O zhrnySOxRpElg=`5EAyVZkp2UeQd`xBU@%^e8uFVE+F1lPE5AT|jVAiK%w<)75wgeJ zfK9U}fY~As_VU^`rr3QGcUU%|cJMOXE2!@8h4nJKakB*W%w?Y9ASOxIB;Z%h5X7wY+h7TxB9wbsdE_g?2pfiyY4ws)Ak5 ztguz577cCoV3g4-oTogB4K3&+yJlG+8*Yfub$}VI5^BHPUM-oZ9L+OVtl=+r1~7V{2qU)@;0E7X)L*2{qs9#6 zN{TDtn~fja-);%pxxhI;y$p5^9}9Xu1NfXZMm+Pdu%BLG#!l8Oz|`AQ`OIm@_>QKv z?3->5{bXNEKZ@@Qt*CHFTbIP8ky^9i~ma5qmDyi>aaJNQoNHnOB(G!0mP04?g*aCPT_ zyiNP0;IdT|6_h+CHM1z*G|6Yu_8M6GE{81AtwD`pmaKPI9J`%zk^HXLhV02_h|8y3 z@>y1jM8HHlDFC`gcA1w^GIS*B|M!SkDI{o9xRNXL}XBR%oL|paoA2 z7|mmYPQlLCqo}3%DA1lZhE<>Z1D;3Pptkc7^JzA!=%RkpHt_I%pjYY$570qtGg+K|?lk*mc$eI=&Vm_TH^F(4 zCKpY-#BZv9U@_Hpyz9q63_I}|7b$C3i!(m@ zp?QQaDhN(S5@kgTobsv6id+;6%-#4&*O8BZE|#@4B9T^s$o?49*`ub=z%>edPuU%= za$^dY-dqa1><*L5-tWlhQ55vrkK*a(-DK58CDzk7QK)YfK+ek=%#`1dmD`u|fU(FP z{1vjLjuYwTbt=5eB#kD1y+=2-CDAbbbQ){<3SLcF#hz|*q#x;b=z3yL%QO1enuYiI zxUlbh%B)ZH#6x*I3n@#xJ)Z@p{@Hcl^LY&%6J*+FymUDen5uAC60PDm+12EZQ=ayV zkI9vxtdkSO?DfGFgy6&9mZ)f=%9Vp3lHM#o+W&GXnfbsMdM21~9qYeb`RHC=X(Twk z;S~9Ib27@_6xgYA!(oT+acI1+4ZGtu&}bbskkC0|oBZ||wq})sQ}I0*RP~sr9(}+& z+-Jb?zdqn+Oz4l-7JRdOK2tm91KAT4@DX1Icb%iK&?XGJW30e&|4Y>Sx0tPpdJoCF zH&Ay=IlHhMa#Z_*9evX|lcrT4<~ezfxxCRO!FBtntm)!q_*xbR3CHZH*=aFYDF1^) zazl_?DnOF{ezw_56Yu<9gRdeI(KP5fiqAY?-}DX>nHBNiSmTJHZ~U=K`!LK>av_RC zgYd23!rd188@Fz_Uvfaj3(1uo7i{MJ*+F`hmv7ugD#A2y#()f=7MR0H5?eXP^6^n$6BmfNOd9uw`EsZ z!ISCC;ztIXwnK!@gXZJgAxBxKmMeMw;4v!@_!mbHspI0+r$}ZDC!?o-B?B_$dC007 z^j6mAPY!DFr=3yAElOag$_3G$7iw^N>R^)pBo7PZ^6}=NZZ<368zgPEriXurK;_7( z^xn@AbY}4>@a}&G8|nv=hOQ1+_A{8~xu=uY8ROX+69+nRR4i}$c7s2+zrwYLXYu3` zMeZrwd*9z|h5PPfgxV(8rfvQ~6mQLB9kt8xch`Ry7-B%y_nE`niyHLQw#T5npcGc6 zq>w9WV?h1kcKmVu1u@yHg5xr$~NG8!SVjEVGg!8S2B_R zDhT?KK_t!$AX^k9z#~W!U2lb<&xSgI1C}K+l%rsEp%v0r4Wq-J=0io-5n{E^iIp1L zBTE&NnA>OYMBzb-;}#sO<>!E#1V!-Y(_8|3pA%(Mkv&HkovHd&1XP zR~)tCGZ`mU0k^+&vOgXk=&L*h+y2Zaf9D;3Bbmr!>MhZ#PY8sZDlRu{fYf}?`mF|$MT05@F)p$rfPAgVZr#+X^W`% z>>|8rc2yL6=^3n^GXoB!8#C5y)eCzrzXgf=k!b6MzdWaE?#i=p6(gLumD3KBQ)4^!JafxZ`M z&?y#+`IN7DICg&o8H@(-RMHgfzpNt#|AvZBCN3v=`RNnd{**E)T8BoSWujNBpNq8g z*AuVEbWxMV1(+It8s^&1gB%vgGGr#$a4ir^Q;t~|S78m_F36?PLU;ir8YY(07&T4oM_vo0Gj?7>_Xr7rMq z$S+h?9fuxLfUnDB@qT7HOtMOc54Kxz$e}pgBVQoCGtL*XkNzXAD-0kr+8ic2%!kOF z#cZSHY_wiE2#-aML)$_1whzChlbSKrq_$`ila*P>wlAASy#4gC($J4Qh@UG8J~Ez9 zSEal}@jecf1l~Tk3ng_9U~#HBEMM2bU`Yo26}Teb$}bacrX{}mQwl|AGw}L9RdmfWZsSd zucyDEwXj~aY_cZZ*RKX=QfA( zMBbOQQQhqTE}3pwHg8!AOP4y#Chd^M3CY8#^CN-d^Yb`PUZ06kJFHOG`a7cjKbH8c z03vt}kmT#EIrRljF#9UxwvY`Hda4)Qm%`&I62#|;A}>3o!+SKY<5kxRCfcqJ!?u5b zyLFuy?Dhmddd4AZPlGy-V!WRwWZ;A|aP)r;ILu@pzpv&c`WbToFH~Q|mon<~QqDK{ z(0T`YK4#-mzZkTtPA8vE4re0oJ~psPo;vTb1EVid=)YY8O7a(DP`Xf;KO4y}948{P zohi&`Qw}VBtcfy#$-)dd7ng=l0=Ko3QNB)z4b-_R%HOviu4b82i;5ap={Jogb~M2g z`-vF2We=tt9S5GT>T%n#Kq22=YJ2cfBN*wXllgxNSWKTHpSR*TYP=Wt6?b~Lb73^s z%AQZGGrsa{zd_}bYP6{4Mu8#m{vs}RDnupq8oY8Y7~@ZHvER9Hve-n1bg0VmAmeiM zZwSXV*0m(n#s`l1W{FNX^kU}N$B-YD1aXfO#A9Y{AVaiEHyGMV9;QSfNqFLJ$2n922fLdK3wR3grb*2H??ve%bj zLuH8I(i%-0zUPzu1#?7yGOS5Sg7Jn!v@wl;le_Rx^)P4na~CnobaD>G6T zzXHgZ)p(+O0S@#M-bddmAnax_g#Wq(6*Ycz@cUT$WYIz#@wpT=Yj2W#%K^~4R6>+7 zDUDft6#Ui;vOplxy^~Gb_Y%=*;O0r+U<0`LjOeM zQM@pOH+3gNoBn1PAhnD%2Hb+HTeYzLPYn7W-3;M=TjBRGeLVKzIvBmw;?F-eZ68(bq@0F=9;iL*M!#%0#%awpLgJ}FS>QK(&joGc4AdrvVY-d=+ZUoGI#+U+pj!<7HF9M2Ehe8TOyT@}s7yHv1~wYBDp3;kHm70E z#E~3LtZ;?TLG;;rjJjBa(4KpaxP8PFzDub6Nmn=dy*mxU=^N5}Yd@@AH4o}e%Fv|4 z{p9S9Vk}MejKmVFx^P}tJ#-=~m7WIz&IItU@5Er^+>`|egcofY3GIDpZGQYZR5MQM= znRl$2&AYlT<3Ha?IL$hg4!Jpzo_2A?mu6#exVjndO3g!wz6+$x@DQm?^y0zo@A&TW zm%OLEhIfy0;_5QTd782a_w_nL(Bv3?ZA>A~l~m!!&6W6(`47RbPDeDd`59B&ehqxW zPJ{HytDviyCk{NP&MecXk-$rL*s>8O5NxSWyc7;XqW+SyClaANJpC(6(ryEV{e!VU zXD9Cbb`l2^y<+p{4dO;S<>=SVQ4l}Zp3TnvCf2y^!UwH*$9Auu&I-2J3VQ`pX@05| zt?gb0`K`}!hVD)n^*NHy89IO(OO``N@H(8oPK$rH`i@F#(z!$|z|++aZG#-=kgnWo z#85j^ys+XE%iL5AtLA+`+f66%YQ#dkHE=hGas2VGRncax`!6PcB?vFL{vn?2PSoUaBh=fipuaX)(Kos;p?dT% zy6WRdn)mKF9olTh4h-?b`_&0h;r|eNBnCp@!^zNPlnj0^FO$C~s=<4c5}iAJC@q{Z z1>|Jv#GT?hF!ydMSr)}fPoSOX)NB)+^q)MO8HmI%myujEcPOinCx3M!h|l9XSl0$#{r491G8?2hfc0%Calc3Q%jdkH)9E zLrh~n^yXd$sk;IES>QNs{2-E-98ctpX{OLTs6zB+uAS}rfSEAqdMhcim<*=QiZJKe zEGX?i2?nL5&}e>;4*K;HWS9>O5O(mYZ0wV(2dIv;64eWG zAb*bRqyoYSI$vjkM(iQFRy`Fm&kf}@!FuTCR>VK7Nfte-3CGfo9t?Ep6d(3e;2*Rr znc5avo}{}BcN%2j8ZHgTn>3u&UHe{!MNWSBXi-H~zk~>fLT>+sHB2e$VsY2^L4{2Lyh$7li!KJRg{l!W@bW<_x4RE2PWLdIf9JSt zi*$LT$4RblvVtcG3<+-yxpL#U6l!j4Nk^@5V?(@>`OY7vLLavs6V)WC(!mT!hue5- zi3={>D#`cToguc@L&&MB!w@%bH2N-H4z26EiPZ^#yJfhMwtR`A$K)q)r>|f6_6gox z>#Z7Jwzdp^Ep+6n8GiiO01IxOeHvqSf5Vc)I&{r}DRjqX0}ReofYGnBh^&si=($cE zUG86tBmEa``&(F(XewQM_^S9*>s>4| zF{J$Z6Hxrx0hSu$`TCe!@O$txGH$Imi^%(h7vDR>&#)ZwMr9myCwjuv7kYebmNM3^ zHQ*au#84mhS1f(O93@h^@HPgDXL~flh>_NG>5>5GI+F;umKU*sQ{~aqU^zAV^b*9v z%w{4e(E;yw3O?osqE#2lAn-zg?d-4q;*s-pu&Ox**7m8>&fcf+Sl}S7c|HKvt}%j^ zWM_CO_g&P~U`jgX#)!>5nu)4`6}udr`@RMU5o z)n*0xCwsvuLEuW}Zo=?16}me>iPlGCv6sdnWzIaA{rFNZ4vLR}<$p6F$afh0Ia&_G zE?$M%8;qcI=LCAH`vugO>*CM}T_+Mk@BB3OOA~Nl1J@@@c8B3a#qDi7O&_II*^E^{BC6p;sG_kj+ zL?4pUpp<4ML(xQ&hTr=BasE2zI@fj1zOMa;f3|h)z1Di3b>Hv%^~whWwl<<&TP~cz zx+DhA&7!*!_S1w232L~>Qm8H13Olqlh4r^IagWMl(eS4Sm>Fg&QoEEZ()uqH2lXXu zbww{kRc|<=xIL;OgIO|YR(c5{ris;h)<|Byd#|NdmF2$WmAawIs29sWZlKqCR4odZ zQ;W{LmsCm3Q&bt92roK=vEE1;r*CJy^^Iq$`Oo+woyY>=p}-1}S9^_U|I>9Sb+8fU zu(fWBmWi0x8Hv8xGi!IeVMK>B(}atUB%rEK6g~DW8M#&Y!iIjw+GiJbqDf6r&D-K_ zqAhC@L>D(Yi?aJ(&V*7r#zBGr*_wgo+^&1_99_8 zO>ZC0>v@mXY;U6BX9^y01Jtk=l54z`HGYAbwVnIsU}oSw;pM_|k==?t*nU=urg<~8 zf3q098hw}uk@?@c!m^R- z+(gaH2V55?BfR9ty4sy%1jg!ZLVJ~mBFURFurD|IlA9tI3zn8cZbv19&4{L~Ie_PUR$9cPbVJoRt#o&7mT-X{SUbMd@UQ(PA_%P7v}3(uL=y z$YFMMB37>!tMz@$X19L5fTq`#BF)8o;j+r9BHQVFOwn8*nv&iIiFu)-Vs_S3eCaOy z@0yV)b);J~?L(3%q&rfW9yDEKlG`k53_K!A>mCsuHf|6#s^@8i@AK zWP8!!i@|5#PN8PPWPA`mv3BEoSEwtWNWX=e!5{n0@a3nZC@0Jbhw?WFwSqU|GRe0! zh1P3D)+wVv=tPl6REOZMr5;^p^$hz9rigrlCS&c(|Aa=iuW|7%ae5+1g|11H!TlSS zqjmBavQ4yEl%2}vjRJt4yUg#txwWv|@Z!dkr54wbZ<+Jcu3wbKE$39+!O56;1h3DLOT7if|lv zz9{-qsp$5iwY3L~#)@_r<wyl!e#qc~t(+1w1$CfIaMed){0hTu3G1=%yst9Tv|dnd}o*FNt9p;ZIPuM53m! z?UPV?yRlG4-5+~*v2*^lGepM*GwA8gtJKo9n97Qk;L`rvsI%b}W_yK;s!GS#(Ci4`7b>xG$06KgL# z^b&an->g}PHZVbc8gN=QQ2gtBVLnPpmDd+D_ZYM$G`ua60I<9A;Y08 zzr^B3byCG4VtC(Bw9Rlejtm`v8(cM-c4sa zZx3-E+AZ3&F%sP*rlOTj6*&jOsC0g-<)CqleCK4D?)vE%j4q+{UYT z=Ds@pN@b{TtP*|T>qalv_>1c9&#pC+dMMi8*FZdMZ$Zx?Tbd+{W#>H~Fn2ziRj;d8 z5{2sZU>dHbQ+4%lIj$7BMXeAz4vvauIgH0~{6?G^@C<(I6I|8*xhBoKRpeZnCOqUM zPY3wM*#GfojhgpmEb(P?e%CyxXLUUOGTtbRdorVT^inceReN06xAF_rbj6T!Tru3} zb&l%w*HM>NPnz1%jtbhFs1NfQi{HD@7Y|R+eyx90%<{NJauBh2BVIxEJy}muU_Tb_ zP7{J!DrhMMi{2&epfmR>)t=NC5M4TO9IF=x(sp+%+MzO?@@sDrDY^}hEx#`O`_4@` z)U%uVJ&nQUO$<9@Hjk-R(WJUP+L zvH#mkL(D?V-PdF5w*8yUJl*%YoAJ3^b0RS3^3CVj5V1A?|6c#sKlsK zd;h;32k-y!ICwl}o{ic6*W(yxZG@{o#o?f48*Z31o5p@-=k&GmX^QS%sC={uf|R_$ zL^g>${gO)T-6oJt2j8+e(6OldPYE*vN{K~p5=q^gMgCY+z@n-Puz8mhj$W-q507*B zE4>{1zMdq-{-4Q3jUEsfA0zgfJh00G3|%@K%D-%Y=Vzuayn%Ixd*PU7 z`FIksSe`U=NfF6vAvyism2k)CK*w;Hp!J^`vtL?^1W6_k-z)pbbEhxl(Y)j6vNarE z8hhj2fMVS7xE$M^3em^#4J@;~1U{9Q;rGoIu*;7Do>c*Co_Jmmn&H8m=Z_;LPs5-x z=mwdrI!X$(N??a&Ivnm}>uWs*B+;gW%mrm=otcSm{I24b+D@!xI5o_mHH{(ghaZfm*z^3((`MYl z=BF#p*W$|Fd+2g$Gk&Foc$NPMeIH5Fae2D*r1Bh^M0?1R54(x4_d=o=C&VD5`#9%d zD{j0m5C3^-g2Afg5Yrk#ymm?xi<*-}NBc0U_++6NJDW9Su?9>X+zS&YiQ}XNnJ}Y& z4x}DefUY%d#5^txwr*@;_rzv+L`ol}j5V;=#}KLp#=^f6ZRj14z_eY==$N9`V7=o2*#0?y~Mt6Cy8i?Cyy3=g^RY9INNMKs`M%2bFLMhaZ^Rn z&{KwEc#F{!yUt9%nL%vwj*^$7X=Gnp8ocGlfNAR>h^3C_*}RqGkJ(?xH#nz&vouy> zQ|4|=9Xkg!^>~n9X9`#D_+jSkGL+4Jj^bHK(AWC_lB%!4L8FP7GIs{X?3<46p%M6` z{0u%1x{CLIrQwW=r*W(CMNBvTBuKrHFIcs9lz+Y<8xjXz!^Tztel6ipjl{WBr4r~; zkMlTNS(R3$E~9#_HaO121V4U}!0T^ANu}jMq9k{b3G2+|3P0%Z*YrCOlXtFg(e(%v z&%6Sr&fDM$IR@AEUxKCKFUa_fVle#aGr9YHGl{4;PTEy-V2GV-`fFGY&Tb>1zC4Gj z+^M7$VU={fkpc?-TY`>ByO8{2aI1zXZl1Rb6PvX_rZE~u?ESFaU@w?i`oi$-NN6{6 zC4V2AllC-g#$YfBj@aJ@^KYi;zH$v14sM1?OW6Hn?qx_nC5}mUlX2ne6L4jy6OLs! zz#+NCU^2@VxP=v9CUqM;<67X_^?A5|c#tfTF^2`pcfq986E08fA{Av^;;lUwI^Dvc zS>iR+D(wTWIqIl%V>On9FD5?c)JThO6EpCm8Ll^$Kubgf@D(mFsd3fJz2M_aTsBv5 zF#kSZVULgCQhp34aLo(uqn%gye_oC^ceymmdG)Q3^E=Xz(|1*od+)(xuE)}y{9`nn zUrP51Rv0c}T@?<`e^|m@x9Yi|WR)osHY-lhk}?4t&L}~a;3>Ir zNsgybZN+g>k>{k{J;tmX)+O=|G9>8bG`??{I^RiAm$xEt(9~+zLUWU*<-Ez?Tsf&B zXE-h`x!gJ5Q@9UA2E0GFuW}!JjyL^47fF2g)V zGlMTB{f@snE05oJOIFYrc%1*I<|hAP_bzUPj4a=xp2IbW9_D;S}A z#x(A6ZaY8JET6Od%0ZrAz*p{>?>&N=tN-{DI_vm)lKH&i%0!+-mpDJgCJcJcWzq`>Roux^ejMvlQv8&B5vQO!g?Gi) zo1LEK@gMK-S1Z^ee zOhJedQ}=W|fBoxa{J57&0>xE%9P!F(?xcY-?$49`yp3Oyxz41VTPrkT{9joz{i(;8 zm#t`i`E&~BtWCf9^F6lw!XJ|Sj^|-qHNAHJ(P(*r%pEg<;>kL0&thF(ij=!xK)!{S z{r)5`WrWMirH44DW88QtoI3t_-7x~ijRJnzY7M6Pf+2U?T;O%9M)7X%TFaBaSH!tl zKauZrN{7GcwK6~Sl@h=CSUoS)OOvlZKzU->R=jQ1a=hrg?Of3u2mX$!4g5yOR=&v# z9ZuBp3;c&+L7c}eQ~Ao5U-4ZZZsqs$&3PfRPx*Fw4ub3L3;1X6UgdjnW%%9`@AAqY zJm*Oy1oPJ|O*OrwN6pmFW}9t$Qp@}E?YsHezqOp%8@2hXdUW`e;<^0!-sT*|&&{04 ze#+bf{)K{n9j67aJn!+1eoWzYm6mX|b4+S9tp^|7YM3zi0eWPIZug_jb!HPSDv00>@37f}xslf$cX9f#AOz zJhQW5ytce7L3U9jD>Jy_HTpcOX^(itq09K<0w;dZn&mvdX(?Qf3R^Cp^P3ApJ^VW- zyag3?SNOvdRCz}X{dtd+SM$#*uMwPIyif3GrkX(C@i(VIp@^?ubAg`@ecW4)GF(lw zC|*#`Dkd}9hFLJDj_tkf{w?7UX+bbKsBHtsT^^Ab3peu5p&t(2;DLY7T~b>$1H{it!{23B z$h*g;;L|ifD(^QEg)5iQ@w+?zs29VEa(&ir+68l6wO~wM90=b{fZrEZ!`Sf*3>S%k z#Hb&cSnr8%t{P%z>sQ!)k%umsn)viVG0fhjh&Q^5A)6ZtN^%3>aI65tH>?3=k_%Qo z^FgBIHo+BA@Iy719ClJ7otA^d(<6bTIPE3__wz}=@=fN$?_lQ3wyp4Er8s=JCr75g zzCcFpt%#YG4C$F!jbax(Q8z>tUG(FyV5$@HZjHxc`>~jm_XIvCT!rMkIud{a*e-Yl;b>|z*nRlp_ShDN-{?pc|Fvk?@Z`Z;{ttOnZAqpi^*sP?! z45UobC);LqFdG9;~>E@EZoPShNy8MlG?_w#T$l>yc5!=&G<7xmPR#MP7=PKO&dko&wPKN-E8zg*09&@_S!KR;5aFjEFX8mc# z`8B)Ib&%lXq03O6ZUF=P_hqlm_~Ac0p}XCd{{+3CkQ~m@YcVe2pzdu6F{)-C*lnAwNjn&2+NHb~Xtd zO#p`)TPThlAXa~`K-phc`0FDM?M7U%O{^e?W;T(%n^(fR>Q1|{P`4?~rjtJvH)leC{h3sdxYLcl*$4Hp774)2l zA~g0j^IIntr+OLV%EOJY?TROCR~C^YagxOE^>TCwy#T=<*zCFU49tief$8HzVBjJH zbh#8b&YBKY*JY?h)D;ZA6NlY%dO)m5lf5pcgV(0hs8qTaBZ4&XgXS=KE$}95n!Yi2 zE%!rQ|2Oh6T8z}z9*0sehZdFv`u^rJxUl*HSzLL6=ms)dLp%cZO8bJ0vyc7>@89aD#^;YO-FBT{#cn@q*9f z`kMrNAF&lb@72MP-98v?F&USeCPDLA6Y$%hNaAiiVQx%)1ILPTU_$*8*!$cLCHJeM z(Z?aERGdYxHH4xo_3T(iQyuovTRX17tI%vXW~Yyfns}(79_*j-cSoaTlH;%{To5na*;S|gbst3!O z!x-))L6&`QA|1+$;D>K6iTXJKUW}dupPinJb$cqKPtwU-XDQIMlY=x50Yq)-Aa8aS zku~>>;HbMLd9?aEdAdUncCL&=o0>4>F3-T8tBRQIF$M$wOTwr63|cY97-AfU=Wb49 zJzmpjNG_M|({-Y$4>r=SN1N$ZEo~5I*Ovpvk#Km}ncU5`fnx@R5Y&?mhXO2+Syq7W zU&+DxfNXI8I}LZqMq>5RG?Y+gYt0hUxKPp(Ef0s`vcG*at+0m<1PxQy69%MKEt-6u zO+dVL0+rvYPOqo&Y4u@M(qpm}!gDj>PV*wL^^?HI++6INa1_p7F~_vBDinNp3T5@W zDA!lWI*_H%>R}B2v7JDl9`q%`$zx&5`s<(_A3}XP4^dl<5bAw%ES1ewqBm3ZX?ltp z%bxrP=Du?1plymuwd{< ze|ZY3tWn496F1@TUmjfFBaSyy*Wo&YO>kw^StuE2f@_yegdf9P4BMEDB33AqFFAm- z<|)#=BX7v1(O#mmWiI6Nv+TlbcUzj^|!OekvAR3J*|!D(U*n!yR+>30-g zD~+SRC&TGoUOgDwHN$#%4=S1yNXn~HCh4#o9N#Vt>NVQ1{)-;#n^{2jy*xly*k;q~ z6JwZl&$p7)5FPNGx1F}R1yGkov9x{433NLA1wFShR8ea(m0P}z=EXG8n96pzZm5W( zA6+ox5Rc^Yw8=}O2-1AvhQQ5q9eHAQi`?Gw5mE{?(RlI({I|k~lt(THHLXVI$x(sV zzIr2&`Ss#pqMTxE2T3mZqx1m&4#`!A7PhjGOqvK zOkI4N==82G8j*RD&OLgDN@-uCjyW|TA%78`O#1{MTf3PWZUo~;XOOScbx|zJ9s}l| zLF>_HWQXZ`kaoz2Z3g3z0gQdu17qs6u_-hZW!4G7Wri*` z40+&J4GF4bF`ibIOrYX^HJ~t=Wlub!cx2WVh+;2rwH?D`+ z@B)bBPQ(8mc>(Wu5hCc@Yo)Vo2yiz(Y>4eOA`l=^jDDhZ#(Epzan_}5(G@Y3oEYt zft>U@kl{?hon{EVIODGD&$U18ElJV12U`Rwt!^nX)tFmmh@1 z&br`KW5gFW}jTK1Zo zVI~cguQb4B#R+o#<`3r7=W;ULDv9}$pGo|ZG@0a9FNoXKLk28#Nx;6N&}f+kZFBly z^Ne+*_Ut+`%Wp0eRgHqdt!48z99OiDoNVrgJ8Jd5Km2JnPQo2p8W24lIJ}@Mq)By_JJk% z<$(om-Zu{qb2pGCEFlSM_efm`n>+pY4!=+N#QHC%lcROjWWz&GNYzLMwS{K*cWy2U z50=2sylJ?=Cm**Nr;`kW5b_~KoMlZa;;a9*h$MY%8WFs$efW4wRBlS5t%IPPHXy$)sOP2~w5=o<1%N(}`6=1yUjE<3`Q zkI{tAv(ZqXnF+VXrGPO{9Y$$2DezK+4~JfXw!S1c%q(JKnnBjr zKOz^&D>7oQ4ttlSV_WGV+LndFk7Ye!{376q~dQ4Sk5{{)^}^dkJzK|JAbWUXv`+T4x0d`KP-U} z{XIgQ9Gu|cgP-tiL;~$ky5OVL=OAy=AoNd?L20UrcPD;@zCWR`EOjh?$+E&l_uWvD zRKjLA5yZQ9!QX~1xH07=h~Gr~n4^wU*=|pw-Bx`4QWdf@-;lUrQ+OwKNU)rLQ*gbk zliBuoI@y2k5mVWrK>oAXMP9u+O5|5eBXd=5!hl#dBy`OKo4?M?p~{<#qNX0WJnt)T4@9`Vw1SwdUP&JJ?Is`OZh+~ZLa0%H z01G7mWb2-=#PRX)*AN-9C)F!7uGpggVTji zsM>f21%6@J@vt9StJ;9y_Y=}yc)_cyDsZNAJtUd#AVs$BOzhHgf`JK2kn_a=E?Q5A zM|&&rfx8v{3X;VAq4UvXvfNdl{3hSvOYoH zzRDoG|CAB$ZK0r5y9T`e8i9NAU5Gq+9NI)Pp|7Ta_)Tsm{lRsfi zrZ1LX)j-?Xjo`W`fj`qwS)X_C@i+ZL#1QUdR`OMYq*e_4+qJ2 z+Za$wa3%|aXAz@)J97C984d^>OSN4#dQ` z5S9TD4ChwafhvzntOK2x%XN#G-WomJeNzmzA{xOlfn7&al8C=@92q{{Nm}lY1zp+q zBHA7+11u&?H0q=n;g5r$<%$iLiCi1j9 ztcZV4x?XF5!>lsK?(_}8Hv1)l(@kOUH`*KQL$tu|(Ng$fWe)9j3t)4dKW>YhhU>e> z;D4MH;K$UEZC5496s21HmbDk9tC}i_h zy@zI_;%x>@4~oP1+qJ}EXbzPcF{j^^v}t(c0JUl9rpjMCXpwUiW6=76dET4BcyHUp z3vAxO{e0*a&n_p7e0u(jQ8a(aXy~64q?Dc&{CuXx%wG8ff2rKV^V{0lT-<4t9hZV# zJ~62KlY#ALmNY?amUkKj{yu+pmRa zvE^`2d^e{+a|PpzRtnRXdT)R6a$5W}F}zy3Wi5Ep=w~j1DIL`v+Lf>4X(OK7ra+He;Wx zfg>s7*xa=(+49$ztg-sZ+>Ts`&lDG7r27UubV`N}&y=8DmcP;P#STGm@lU?|_q+T_ z8jqOc-zu1g;Y*lB=H?`zXa}>*E1XeyW+OPhD}(=OZGoU`Z8J8r+8?R<98~aIPNZX1 ziB0WCrcnAWU735IJ}1G!oZe0%{3j66VHZ=dNIC`LJN&v?ZYFZa`{QHZu42+<%g9if zn`3CSlo$>C<%;H39#}nQE6SE~FnIr8ICl3Gr1b@odZl$_-HQ-Xwf0<_68={C!e{Sj?B@;MTDPwjBp+bh*h06IkT~o zlrLHeoH8BK*d|G2ERM1eElPuXc{IC$N9UyffU1UKkb>Lr?`0T$D;G>JeLqAm7wx1* zfxD^b%3fN&CmQ;Vcfqf?1F%KU%Ur)lW>({=I^^$sU*zMdk?(hPiSwihi6S765ItS7U&W0GGAAfgF)E28_fG-JkC>f4qKx3yE@ zzGNUI;UwyDMx37d{s#4r&nKHN>XEWde;5>nqC@Uh{QAKbYunkmBEtxj?uf-ZBnhsJ z4*=Ql2;d$Xz&tiXamnK@8cw%|YyYHRap_O;F#03P(nnYrS&e(5&!F4Ec=S5I5~s~M z3Y_ADpq9KBwx~YCPl+ukuTX;xZ3^_}_+N+$Ey(GJhB562;hLTs#P^@(@9O|TB=3Tt zF=;C9EZ4!TT}F7eRFTPkt;I;(zAWIo8DU6|G};>ffa6_~#4W^_)alxih(mp7C1}Q} z`So~7YL%dMrXJ(qmdym(0eop>b6OG{2$u_l+Y9ExWIJ`(q&pt^y2eA@>TVK7z2Kuo zGz4FY0NTZJYClHff_1xbVN@;ulSGjqJHJ&htjNckj5Zn@NTUiT0IWAV!Kruxsx5M4 ztKATjIENv&{;^^IqZ`W}hOe*;#ow_wQq z9sCeZB!-ovOxy)i;^B0Rtk=&X_rwnq<)_V{`|$(}XPtxt+Kt5RFPrcGaEr*f-z7&K zZj%U^F0$(AGR&^sjPvVfp;X>@(6cEf)9On}&PWVR=!~LaXT#~!L3WR7%)>B$4sJeg zNYyup)5S|4;!L9m>Z!em^6REk#mhI~Wp4pw^rXZ5B}!y=)>yLl-$HW2PnUf24k2$I zsF5Yh!@%>gHB?{aKv`ZW>|X?6q0mn*I+wxx{0dN5Pymw$jN!QIY_PQ8!(g>GcD$XA zuh&mP+2iav^*RxywZn1ggFyT^=#0M>v1}Q3-dFgK^_u^kjr$q`Fg|TFD#zL(yphEA zncXm!o%_r?Er#6J;wY3A$Kng0z~`?O{$u+JFXd8bRQgex{3C_#nez@>+vkDtmIuW6 z$5~j{7Y$9Gt6=4S$*48Xh-G4TlDn20;p2SP_tx-~WNL;G(F1^UA}Mt0k)?qL)amS2 zEjlZPy~cHOaeXiA6*%${7xk!9?S3(87jg>JLaxHIhZJ;F#!|5;bvk9=I4V;&55~|O zxMhA3{5tlb9zTY4`pmIwTi-Vk1dB$mJcv`or_kNU%m;~Xins?>t0j9UgW za{<25*W~qK4y?VL1p6Pk!nHfL%p&c4^4xeLZ1ny`X7uj>&Yl8rNbN_hxqTQ|_6j#d zKEz8MP59pCG%6ZY!X&*0w(d0sJ8axxUr;V=dvAyu@(W4r!4T%{<>P|!#@6t)KnqN| zjxmm^)@UyE9%_6plZ>Os;grJ+@aQ%mJ*xYtxZ6QWI=9es<@Q+G{ufs72! zA^oYhlvba61KX-5qnAu4RG0$QDO*Ix1uv!zv>jYJ2TAcVd9q-`cQm*+fxe#3I>`0K z=xlc08cZ%@YF`sx7;ZudHt*3Al|&Ef#8OSal{7=+E7VUdflHsHrveKX1l-{8^bcLYV5KLm#Y#|ir4Zs5cF=P*(u z4$U9mXZdWkFy`T5=%^niLhk%M7ADhhq2e>F>8VrCLL-3dz}P$o9_my&s^X`%|e*6TNm{?6iY|7@xO{q zxc1^)oOD7K_wcsh+Zj=4FX@dHYaB7oBODc;7h(V3GJLV@3Qk;_i>(LGqSLQ>yl|u* z=PX)+MrB*^PN+FPt9lJFMX$gH z78X)dEyRU?$nEVFjNhbHB+^V37KPc7T_sB7QI{GzJ^e)-zAh#<8-J0fcLGU&@(8i5 z3&G@bFHzrb8r|_GkEX|5ptCwkX??XLJ*Z|$hi-Cc!*+Fx71PFAs7Cit&u?1j@I)F{ zsgy#r_DWh@1JoeOn5KCu;AqA_$W{LaUh7J6!IvO>KhqL<(M!qOs*TL{Sy7Br^#~kj z&jjmES1=DNf~pcTxTD%k^ix8>HcSb`tFCXc z)XFxN5mi?(nwu+`p+Cm3`OXohV_-HjeM>u}rUAr0Dkq%10dS~ajfniOGW}{fbdhT` zby(#@plN4OS|8ZRAq6CjVj_6*Vq20*92Q2qM2R)MMA;8hAdh`tvWTS zoXZXP{lD*K4g*i_oujW5YpGLFBW?B6q3UI3w3+3m1&4N%dk1Gg*5qBVC*Ggdy|AbL zDr%JX{SwV+s-~td%BlPG7A#h~g|{5oyzbwLI6iy@s3e~g^l7cYc=n!IvEczQ``3e* z+e%XC)koqIHsYVHk{JBL7M|)v!MYw1b4uYW^LoQ?9RHhjjUF7tysT!y+Ta6>L97F5 zm-z(OUM)x7ybPQZ{E>w37-CGHHjxb5-SDjEFo+-4gF6nA=(SV~t!jGU*P9j$!*bL} zD?^{MD*}y_Ab!u@i~Mu{me5RX9qJ_g9CsPk5ZRRJ#IfKzb2M!TO_J;}jGaw&REr@J z_X~+kqAM(n>xcH31h$u_51T)_QI*s^^qE~GwP`;{XZ*K}9ymRlj*Hk&a?5>5`%xoO zu5up>o_)tXzvO7h<>#nc=z(K(ml@GB+PP%f{VHZM1Dw$T-x=KJn;`BmKDRqae)lA z{n~^d1%)`(JrQEOl;G^e4dnfVuQYncHR{rsK@TYU(M1O=>7)il`hIIKl~7W!2>z^O zvA^mT)nFQ_v3xCUF%aNnRVA#tFO54qA2Cm3N|-C>&oZJ9BJ>?d!j{E|>35o;Zr4gU znj-_ZPS!%#r+oOlvkjgKB(XB56*4r!K<3>RlrNOUSHsoNIZ=dl*2l0iz#fn5>Qk#k zBWl-ZM`yms#l>4Au>RJ5+~0i-ABd-+7YW8AgJQI%d<@k$9>%A4tk70W4X5rLhRD@@ zcyP1}W5?E@TV*;O)eEN2ycbaOssxLm9(RkWy*!KQj{BkWKX)*DtpEvDH<=2V$ISCH z4+P&+O5j|GGCVrwMPkpJf}=?t(HTBWOwP}zStWDnu8Xs2^T`Za{AVRy5~I!fZ7guD zqXD{1$pM=*4?2bAG9N#xM8lfPi1msA=6Rwc8D|+tw=7>wotDm|uCKG;fK&jesIP%D zQ^avWMFX5PNrNi42y*05G_htzX{+b2S_U`)V2lU*jWS&9zPJ z^P4s#Xv~KphdHoO-4yi3C&4eH-EhN&3tnSfQ8COHZSO3_>{iyhdiNpb&M3p^hBl1Y zSA^Ve2T-%?I(p3zV%XP4d{(%QzU^kHboexSFn$5R-WoFZ?=&)d;$@O~bSVk^HAsGR z_rvRl3m{WAo;Zt92)s~4G)()LUx$C9yLugdcV}50Dy+}+63b1zehs$QidpECKB1R# z8tJPYfpFH_2mZa$f?vuUI$Eqimoce*daO}hZ=V(&B6rK%R`Cyt3>Bs5Mk*Dr4poT`X|@98sOeE${l zjqSzt9&RAINgPtsU4hwmC$q;o72-FWVPumc_THDrWt@L7;XoEFSYr)RGp^8G=R@e+ zFrbx;Bkf%`i<;$3rg7VsFuD3N%(^8#{E3fs>A_BAnrtjj%SCGR!}bp-X4rw{CB}5V zRSiC7YYuOZ_~6QV2eeaH#7Mb?kWxM#rp?SJ@9sOpO7FQ~eEcOzh;Rgre{ad>wR6Zm z*9qjm33~gy}hA{&AZRFHHOv&I`m0r2mV!v#-=4LOib}C=2c*b;Qj?^i^%eK zbn!?7)oF}_p2TEwWL6z1<%n5WRJBlNgHrnWQ4F1UcsE`6cs>1)Erj(I7eQ+z8hUq$ z!93?HB=Djw@z{0{L$^;y4HX5JH@l9cEp=mjL=#AcaX7||gy8i|2R!rE7)qt=!S47( zmLYlxM4Q*b!RJ3n)yizR9)1)YMdiS>`cTedQ<{EqJhf=h#^NrB3MqY=2p zJrK9>7`#*Pj3lH?13jh-ejijPt)&kcaz~wHjp1VaIV*gs5rcY0_aLhy26UnjI^#2O z{hzIv5=GJ9o)3lU9LQ?3054}PviGJu`4(65LW6`6e}wg_&2JR#6rs|2Mff#l5YGen~4472v)24?744kO{1OrBUhVkSh+BU^f8 zFx~JdEMIC3v3Cbyi@hw)RM5lRg{2s~$_Gch>~Z(ecp5UloQijyrCqOe>745ZG6%J#_m%}DhQ={(el!V!yZgzM7zx~}Cjy#13->Iv!&j~Iar?<_ST)8Ej$QD8 z-4Ves{;)5;SU4HCYJGsKNqQh{zmR13-z0@cV&VO15wx+cAJgiwn17%X$VGX45%rh( zVL8AY`kg~2nL7b*{3h_Wz6Oq0<>71102y4P2J&H+Fx4%D)bKjUuDN$$q#zn1i~@oF zW&5Gq^l?J7Ena6cH-qfH^5X(cx{Ujg>02yC!p=Tpj!j>V5|S<$+`ba!)@Z>Z;S5;6 zhHYTQnF3_GL$coqa89~QE;wO}h%XHyf(^%{pae|drpP))9ndB;3 z65;hY^3KH@luii1nT(LC_;FBC`kDNB{FJ;;Wy zBXjZm4<^pZlsJ7Uf}WkuaI5MNc(OW^j|1D8fZRATs`!$e8}otuF;ND2`zYA2{2YGA zYoMM$1^oY2vHjwIWQXN?YFs*xK050{3vZ8w(Btn&OkO{kzjHq%+1w=aE3Ap>tX(iC zteRXKvjB{Qd8GB(b0RIT2R0A1VfT$HqS^L~?4D}_+-s)f>U#;Y<3$DWOu0ZBX2lVg zj$-DAFp#OR?qgD)i;;jmADH}{$t2(E0_Z+S1A*EQbbmjKriHI@SK2@9NCR@BP<&_+E+#ryHQfq79*hFpBku5*rGh59*}$3tYiLyS0-Z6B z;oYw;h<#TMVZ&i?VCoV0Pk0nYuTsn$Hyg`d+N15x8}Or~8s<;=5B`pi#`89OlzpOt z&RT4*wY-`v*|e9mFBU+c$2^vCvlSk?%)u;KO}tXChE3lr&Pr)ACZlf2RmT zXVpPJI~aDEdxBH%BKSNf6Ao`a0@vS!z?h@f@NbS8m;|zUyCW+>Ms*G(Ca}*Q#olBA z`yDUdc%FIfp$RV|^GVKL6Y|Vx6FkzA29HSIP!!pOVXsRfw%2(R`|yoKvBZHqU;CHnT&yEYZiJGepeT?}hy-aP zb4ap~gMydkr2cyfIaGKBi^gBT3X?Oq+M1o$e6Nc!J7!{eXaoLQk%9pSqtJybgs*w0 z;I1%(&GhcWAFTIio{1|C9$yASM~tERjvD+b$wj+YA?T9316QbZ;LU}XG15OD?e?hQ zM>cP(+BFU(I)~8wOgFyg-bVM5LKr6s2c2vuusVMNv%iI~bwhu2ZCeBmms^OT_;w-) z8iH>pPC;35I@~ucL0j8ojCKyerbS`snlTqElMV6r><8p&Oc42&53Pgpsz`K!bcyh1^ zp43UQeJwFee>({A7E@u~9&z}ubUIAe<6+MBsi^lt6UUbZkl@@3W_sKll5ZD|BW#`~ zerF8Ylsmx;u`tj{#be^J(K|vDO{3a8( zF3m*s`6m$853|oPl5UEO217%?l>2@?EYn>%e=UPJ-iN9cRD3eDdch7-5X#=ay=+P6)M=G@mN z&)=iSuHrW~bxR7f^EIqpTZ42r6(9b%i+vkvP<_JFC5><%O!hnPoRHeqUc{#IOVN0=2vX9pL4M&FF|X_luo6$qZF(_i%|@c1D9x9eSDRzF0>QaLn!5MZD2zZ+@Ql zT>g&4F?=p)fDxCP;X|Gd|M%lq{*1pIZ!YP~=Qi2!6ZCZWcZWVvd%6Ula50yhD&zQ! z(J{QD|8lajhPRhf(Hxo~VYkHc6}qz9TWP-rb zE4KyHNjQ6sft}Qiq}<**R2$&4wl@x1NTln~{J0b}ZlWP{=sU>w=cg zg7569BpkSRiK;jCQ@@=uKb!gUZrRiM4O&|K%-cKohERWgt4c4mo-L!3tK2Dg&RF^x zA;L4g*FeoT3jEg{XRDf>;6_~(T<~2Gn1M-#Bk=U+ zAFwpggGO|Xr&IFV>3N3^eLM6A(_~_B&0SS%m3LCs8hbi;K6@}9_%=N4PWm5e|x&iuWp>z{?*;p*h-+x=p7D@4JD* z?)oQ=)lntYMY(uS7tnwY#nZ}#XeD?HCl?H(&h@`>mc~xF{HQUn_<8sOiETW$> zmq^bupJW|#NG`CEnqQrvM(JX~H(gB8V*<&$cMApW38RVc75JwMb@^M0QvA;g%W30v z4|3GusWj{c=H)iy!VRa5)*s{UTJY8;Fy~tcQNnfjC3(NwzCZ z1f9TG@Zd(`Jxjs6cp(WZ0xU>-|8w@qC=$vKw1U>_LU!y)FspPQ46$owU|NR)awe{qM!=lD<#-!N`5FFib@mGm1yyrHU<>Tbl_>4R67c z%gylZ{9*8`eGG>CXW^(lR+tmN4VA;MpjLq}YaR%KxQ+)PSsVnL28DySo(44) ziy`~hMOYZIUOf9{w)p+pNnF(E!Q$|&AL2LB!q0z8IQ%pT?}mn$z#GaC-^o zb}5QGZEVTDx_3e2TMe9+H3g@>H$u}mGj#4ahBh~!v1`+Vz~zA$*SwsHl6t9_EF;_> zy}PmZR2-(3s8HVbet}PDjOus}2TZt$(*%~4{lG;up+tl_6PMwFB%zy%N6hV`X<`ph_HdU@*Ci-wLmaH-f>aCTN|M09vampgQCWoH=&`_~C}=J3AUP zRS&_JqI=N%_YW9p*RdNX6v5!PH~9WG2FnqSaAox=nE5zSG`~4rWKx^YNiTAS^V|R! zbwm{%g}apoX$$&s08VPT2?w`t!3|Ms>CQ}TmJq$4Wl6Zh+2~NrZ#<0U4~2Zuvny<2 z)J-l`m4`Az$arDPX4X=#1LT_a|e6>sgHvQbk44s_VHqYU7^TK@UzI%qM zr>p|gCKvQqa;3f&p`*=SjkX+^NWm!upg89O^zHiuOH?0;>QC(#oz9)inR-ZI>l-tC z(OQg0b_!?HqNjrY$`0RqXkoY4c)XQrjghjc?743~a}rBK=#g+ZKfD}5?|WdZ&@;Mi z%}u(uQ<*npb74+|HGq#6=G3Y(y&Vf#>{~e~AHSM)4%x?ok4=Ti`<1|Lyb|c%^98r? zLm+#o9BePjW7btg?$3NzwsWs9?EMr2`8DC7v|}{JT%OHp2f4B@L2Fr_{ZClFIG>Fw5W`lnijh|E>~f_$`L+_UG_n(QmL4H-XKf;X+T- zZJ78k4f>-*sBZojq_TFPlfD88J4_b!LdZfo-2rbq5lRV+P|52V1nviL^w?|^+UYUN zrxR1^9^kqSax^6IAUqW``O$l3V72Z9Fgf^`#jlD6QT9q$Yv2V77rUVMwSnSy?|sF& z=j568^B*uMV+5YJoP)>00oTmc#N*E;@bZrn%(qVmtR?qhWq>hW37L(HmK5Ocv41%C zPbZn3u^Q$FZ{S)LG`Zlm!Eh`50DK>i0iPb!LE+cs%puv3y$O2BOacp_Ohp4!&;Mo- z@_ML0K_5RJK%Dly4yG4LqMh0A`WX}_v#KoP;EGlEONm|uWzFFuwB&Wq>dw9{PD|;K74mp7l*~T z;3cOQc&;NA9sTd)EO(Bk&9?)sYZ|bxs|Ajfyo4QdyWrO@eH=R{49mnv@W#Wz^iEA^ ze%$q&iCq)0{iYFg?T>?9r%UkG-x!czaSlFscHvL$3J7YHMu*GSaeY-OBz3u4$TLi!%2xh(4prG`)%pKL_sGw-5I`EbZiNZbzXuo@#ooq8gEv3 zyBP*lJ%OTtFt9Nb*f|$X;YYa)?3m;Qp;{quqrVB_7kt8(fzNP{h6eF>9Z6GR5^mP; zL9cmgR5NTK`>^^r+iA-Sv&Sch9(@9LItv|_GKYois0QAsyN#=!NukyNY25Nd0asP% z1K#H0i_QUX-98+OWC~!FR|NWwcms-OM&lg*G@iKrk{!%SW3EQCL1{@A{@oFP6SWhu zR7aOBEMLY(J$%dDwT1KYiW2;k&O!0vY-m;e25%+TVS%y<2JQKdgLi$QX_?!Y*Xb&D zqB;})UCx11sSU8D)EPypzu7&@4NT^&;EUN6hv&6F<6g|B?7Q-)YpahI8#iI?i%3*I z>4eT*d$3lxBMa5#=+5!;^tN8`zp9)^6Tv(CEcQ1p9ce(7kuwB8Sq)rT*98^F&%m~S zyJ5$%n{36~>+E^%N0_kJMbNZ>&K*OxYqJ4gT?iA1F^ZLp|Sg**Ejz$Ty$W;Nx3 z-H)AEu}&Vs6WqZ4b{X_Y<-)+LSHb*uAN;D3gq_M_C=A|#HiEZ!jlTg5IUkEdbF{gP zVY1x0d3TxG)G(%!o5DKVKSTZjAusBhgF}m5X4ruAn1FG*B!9&4(kk?Db~j6dvSl$dt9*2k7iW6A}5!Nx2Fre<5Guk zs){_d48BW_jZ>j$gBqGjMBvyhv-y82Bls?m;yF;4+BhJB_sbM=sLEeS@2mmr(jUNbQy0S&?Q*aTu*IP04m^9u89yeTz`Pfc z^xk_L6n){~%=Lk|78>}qvh}?4uDd+bI)I7Y@?>rmNuDNd*1G#SYpeI_)*9=#;+>8= z=v!q#59;Nu$Ig_rex=gGw~QExe~K31LY;vWw`e=R(s(|v))pb~eTe^J^mA^zk7uBonM2kXLo<)t!hs5Guf{BJ*wy!#dZ%{;+3{W-z6T+HN8t4UkWoAi_4{rMF?*;tGJYWahVzt_^~ znRC&1k~1U^%Ev>#(d7E&Hbp%h#IIQ23AM#GDEoILSsss}REKPualM5u8qUU72j$45 zsfgz63FF%??Bt`B!}(h^``Cu@w^`x51gP(9B4^VR;;qkk@@ur*`4Z!Fru8m|h2^(0p^1`DYTwG&oebb- zdW;lus~526s62@;FQ*Xnqqtq8`OKN4NHjp|s%eAiL)=pO+muFOsnvYx-3We< zlnXyA`6KU_aFf5Wx|aWMU<^$BsRzw(vtjD0O?*qA0l%uNovI>2dF$>Ye39v0et5DY zzfN}`ul;5fia%ZvB1q z@3SAP>HPpNmYCtlv>v*xUPvo#6G`)`0`0#ca3x-Hbi*K?4h1eEM>%7Px;uj6EY}G7 zR363eTtW-&r&5Z^DOw@Cv)(*;PTqs0d0F{le#*Hxe#n^3e2%t^z>XfoA8s+=hmMx! zQzT^g_YR8u>joP>-b9gKq&0vKm0L=_so@m(;3Bp8i1^LxCh(fNBl!n6=JDPatofKZ z2E6|LcD`>#CI7TAng2cc1SZz&3i&@*?0Y#6n>Pb)eQ1tFlN9K$VFIqLtAPQ>hT-Ay zZ@}@JA`a5~LjPR4>HBhde&D2`bYt0e9B8}%J(aIgo6UF1z9Y?lyk*U&l}_Z>pPRy$ zWN*hore1Y@zyhS!47}13j*=Z;p-1l${C%|+#yn{UNvmEsc>N=6F%CoHc|n-w=#Nhi+-F<1 z_(Sf(a73{wZMC-{w}7eSld6V=e|Li1@nmRuHi)-N`$bXHI%p2WQ`N*HR3Mp0C3!+7 z%uZnMM%*R)&hhZ{!!l?X^AZ9kbi>K?eK5(x3KEq|@!2{#?B-6vS+5=-`65{Htqv3~ zxbmr5-u%QKXMTFIjmCwc3ivd$kizJK!1_+50_`t!viu{NiC)ngu8CBFPEgD4Sn51CpC3~; zov#VB<70&ke!)g5bm*4HnXiG*aUaF^bFzH-`ctGn>O4Kle@v3bwwP0W4}Z+nrc2My zQb*h!dT#oZPDKdp;z2fi+TE$VWpOB+w1@<$9CO(4<_Ws_mEh=E0+THIAI+LC!^>{@ zkDeSGfsuE+ncvh?%)~8$Le9ogv}-KIe7cXX+r?;f>n-%?_R{kuQasD}OZ~RIr&?iLhJSY8^$2*IlW=-Qx%zEX!*D`a`|}fhY_CvAX&M16C#8^I zI56Thu1@_A7b+@I@GBY8u9Bmx)xsU4*M-$Tr16gTRNhO>^C`|E{(LFV_WZK~gSXN^ z*&6)eufurbB|7|<4eR-BgFX55$qszI1k%OHU`n{ZwZ=L6mboN4O{6#FaC6P1DYOb!JL*msBkR?S3sFV z8*rfya7UjM9o4ul_))!3s(m`fFPe)*t^$*~+?aA?i0&wlrtkT~NV3&}N_wSGEnYbL zjpky+j*~cC={MGN%%aP^ZJgAdW)}aghiPs1r$TFgin|m|53koU|3ez^H2MYmQ!^Pu z*)`l*bPli1^T5p~jPaCWKgbQvgT#5V7&3G$IyjiaGd~Gf>XOH*=NaO59bHUqG{E&y zDj+F%i>`@6V4_VrlSrHeF{|`IafU74>fVgzdt(K*ei=?%@c@J4zT&_SX9Vr7nDlLn zXotKZdOn+i)h$jK-l2=yJu|U!_!4~mZa&6m3*6rjNhIk@SOMR54&M?O!{SZmz4Qr60`j z`RlD1nYtJ2G9&0rMgZLl2&VRl@lczd0*Zr=gNrbW7G~w+0liR+$-jn!7hFM`DKeB~ zu0ttL-01bi7&3HAr1EPQXuf9+ZA$q9;fXb19C`qvFYcuk&tu8*?@1c}Oaf|C457&W zDR>*p;jJuNe0g9pewp?FN6h#tpxws^-t{>27&i(Bi;m$+?=DD)?u6=uFYvBR8E<>+ zLZ|i}XyzJ5_di6E_KskZ?_Pi}-B#n1YC|lqSHqctU%x-=7)B2-#-R2qSmhCh8Fy@H z{}XRY@7Y0a*3V$g&1NvlEQ2{I>+rvaVYn!+7Qdbe#kn%+=vZ_a$1iAtfg=mSFE|)B zcpZk#BV`2*SrXw6M-@*v8gg?IEsqIez2B2q=e=*7VYm~TYOTYWpO2zx4382|?J#Tq z3~YGqL|cT8f{{BaX|hiQTsmxHsojO3u>k-_VBZJHJ@1mE~XG|_?!V`n6 z@uyrYdRc^Gmy0TfU1)}9Lwzve-(8elQj1k%%5Y8Fc`OaNfxChp_6}T&sv@a zBdH_sbKD=coj1hVJ)^PsMl@*5aKb5OQJ}Ef5-#(p+$*mPZe;u#QK^40`2CQ@YqPh5 zt8y*KgxrMVTY>~VG>Z9Ex3I5|qFBjGN!Tj23kLt30LJNuxE+60*qT?;Tvd0d==Z8D zQSn6^QPSR%T)KG%w=v@hH~K>*`?X>$bUU18KZG6ULrGOk^!f>ob$4O>{Bw{at^=Dl z5fHDw2NryM!#X4dzUPbQ@O!I*;Llu*y?ORn^{Eq%ww=N*S;1Fv#R%nn-EpAXOw_v} zgI5*5af|=>vA~6Q#ZNcZ!!*Z9=rS&b$;b4dKK>=LE{NH`kg|@pcVKi;U)CBEoA%0YOw(p7M#Wb zD;Sw_jmfyGunA`8;l;5;xL$LBac7pov+XNEF>gOhy_yHAXAi)`^J%PT=uueXk`G=& zf6j*q0%t|YlMInQf~+PU-2b}{r_36e&yH+3={+5rJA$yV;s*q{FgeeE3L8%Vm_^rFJE2ADpObBHfU+dw5kQ1FK)Ho!qEQ<*{IZ4o)_?4(wKp^Cv4`YGG8i;38Wt8uv5fO?nNEX0 zyZ!nF>-pWrly{9~-$RWs@UjhJ{}Jfau3{bf3UL2^KHH>z89z3u)1`qY(Yb0QcwP&J z9|QKY10Zkg*fS|2R@mW&ipm2n9p%xj^2NOnVcWV z9^EZq?b9x?zLTq`M+Vg|EQaqlr*Zc}pKz0t-8u1|QEbiR%WUdfS9aJ_jYgiCN{ZjU<4o;d zW;Js^TOHLdRuleT<{f;B_1*U{|N0i#s4pv=4})3tCQIBiDi(SNxU>4c5^NW;fRifo zLES_GlIz#PgT2B%{Qe4)5&G|>zH13ShBuJ;d>njE8p@ux&xIJ92+;ay#wM8daBgNv z>`Qkqx4UXJoAK`^8z$Sp%^RqRX*QPF`$iFc1_WT~zg#r%St)1);n>uWh`I(rpy@XW zQ3CPgx#ys~LK77(FUGyI%fL5bA+B~ zFbTvpU3;PCWfD9%Bm=XnPh!$|VV3xvh&PRNAYFer>sfh_ZJK!ls{3Uz#@rm=KOF^a zttGH|TqryrS|N71b4a{m+zKx8vpXy=e9E-=!*DM%9Iu&7!!UVGRDSk=xtaBFwG(!6 zyPS7nQNuSVpI8A2Vo7%C&=6KDQh*{8bFp!oo5-v97dNQM63nwzzzJ&Enad8``Htc2 z_vsogJU@}UohZZ7;sV(+OJ|t&`V!oIHUhj-2B25vbf}zB#$*B~VwsUNt~L4&uFgbH zJ>6(-pDV@ZbfTnFB`SyQ!mbBRjBl6%N#cLp9;L}*AF(ZWLFoII_B4Rs;#(|{)4{`^ zhTx)#2pBykmz_Nm50Ofmuv>Q*e6Cl)h<$?|;}ZFOVX4pC^23iVqX2;j_Sa zi|$&6=iL{e&x_&sd}jx9^LxN<-|i9Ec;n&A1raQhKL^X5_ksWRQrOlOjR#$xv$p4Y za6h{U=9OeX(LOPZai5Eiu7u(2x*@o=cL|J6x5nC@D&Z~T3RyCmFu!9Ybc*BPgy~kK z7Xt)6~4DS3_(w*^|ll&&o>Wlstu z&+)@E@222pGi5v_p2CJ3m2;+|ZZPu~JXijGkac1To;}&e7mF9 z%j4(8M`2@oDY%Xw3Jux?>~MuF40?Ehy^7udG6rv8`JrKW{)`mv>$(TrfRV87kt7R{ zHD=rI8?tM%j`%!A0!NSG@ao6qOkOLT4Oentoqw0HE0JBSCi5CN=G)LdIxKMqqo&STSP2&#iTp?~i-IN-ApwXW*Y`gi|u&!!#wsg`QJv!br=`R@YtD+2jyQ=oq@;^_ z{LYJfg5<@Ey>hKe^I|PWf8e5AG9L~hLL{vFw zxz(?$nxdcj=PX&CnAtwrAbJ#_Aab%wXKh+L#TRrWtWLd35?2=w6zkbmh%&u4iY9x9 zS@ro{v&skwwd^^vM$~CI-g2n>c+rIgQC9FO+0y=^zSXbZRIA7?KSGM8ofKs zsz<|xxyBoDW?vh`@&Z3aHIG>>@;Ple`O0GP!BkE}M?);netyl}8oyc`ptMbtacih} zz&#i7xT+w_8M#AQYDR`=mGpPZyUPvXios*f>8*_T`gtoZ$SzE5ki6Y;1%2o2(xt@9 zbdyE4FC4h}ITOYIHjNSORh=)A^{Ws|Rz+JXKZ_M#`=&1r^BpXn_o@jFhGnd>XQRg%Q! z98%|=U$hmQn2ljM8uDDPOQZPQr3%il*IaCArDwJAbAw2$_m#!ie`HxAWd2TihH*~z z2dqMEG?`4MJ7@2@%_=Egk%d$o!Oc^lBG8f@{T0WBpKD@V$y1T^im4)c7AqR!v{Gz-vt3lzxewBd_JQi) z1+ZH`fctRfsW@?Jzv$_qiJYTvrsXQdaEs=dfmYrTDk6KIQ=*b;8}6xpnbqtCqeb&o zLd3xy!sq9AMU?w^AE%c*j=R3yRD7%%1sOV8P}eb0ou9MVfZ;1NP^o z@bo&$UiJ>JmQ*1oiOGNU6Y;UMd^StN3?1Z)!6aS>-em0s zgLh_F@JAkpCwAi1>eY}jA{dUu#^YQ&G-dOoCYT?yKaEi`?M(7~MGF66G$#3z$ybDs0Ivq#sfnA4GgP~$TRL-kIf z=Zj=4d>1QtSUu6A@;q*u{EC*uIivdwfl7o9{=;Cchrf@mgta- zke?i$C-m(lU!{7Z?`%O%Gi$%B0=ItrMCYi_SeN2SVM$+T^OGmksQ-*kD^0>>lRo3F z7sL2p_GjsikT<(1Fvx1SP3XOI0VbRohf~ywAiUomT&z>U!ZRH+ZVA5TBvl#^CIcf5 zOTfa8TcGqaK*(#%MrW(BXm4?f{w#bUurr_H!&m-r=Ia{pN$vxi$M5*No^|}5rzQOL z>cvddaE5hk{sm(9eth741|O`_rg0cY-lowTjFtO=0S9i zaK#A~h46TTJl>rC3Wv`*${yY7X6uDs@+aaX{IB$z(AzQ`3!izy;4+1@d|^frcipL<#@hxPR#`9MQNJO)AHSBDbL%)gtfu3wBT%rXq8!muo z#XEL&Pz3w3{}6iDI+J6;Kf|A>Zig#?HyE5*F$Nw^D3 z@sW=^l?RTa3E{&@eVsR{++Irm{y5U@6EXC|b~iQlZ>I02xwJnwh}jpj#G?7>A}U2%(vEjH~cprFgCwDHbLH-qiS+-%+eY3txt(6KRXd@l0((Rmtb_d z5|#?u=&e7GsVn^|6?lcv+s2vH`p=o7H(LwL0D+U;Cvc}HF;W>OXz_C@kmo9hEjvt? z+k)s}`W~D6IMck%Q1I356a7l?36%MeZl&`~R%cVwG0@AI7KKXSm)oQ8@u^WLuJERMPE7j+Z}6)b`LJruZg}E74dzBG(BzU<)SufTc!Ad9 z?!@)DXK);T+1vygWffp{u^cuDKK;+f?t?>N83 zpEQ70HlD}ow$Eq}2ZY|Ytr)Mm7Z*PI5BI&1q_F|x$?kx$(7mFD59@~Fe92~+x*>_YHYbpCXB=(0 zwv}GE2GX{o7)sX-qtREhXjJwXnsI70iJxhbdfsX1lTL=#b-}QIRXpoo=>@sP(_!q| z6S%nfGb9R?*r0o`vf~5H%WMSsk0C_UX3?wnsM7{Vp>$zE znfRm2PUf*>DmV=`0G0K5u>4j&d|4TRcWMJreg0Hfwn>p2J>vs7%U!{y0&TY8P#HHl z`>?1`%bjT||KY}+AI9QF?i2aiMTpJ`y)P|Ncev}zmT(_;P8K&W*JhFb7O^$b_t}{} z2B2%W3m<*E4IRTxpi|>KELk&%%v)X9aY%);w<~aT@_M*;#e+S0xQw&gAxXzGBxvGplL1 z3)^Mmv8-V-I9*6%3+wDzLBj-)@UVmY^}>EFWU{~>67+&2ffQ{K4k^E+Ft)ak8+&FJ zWR_Jh>BUmmI>DaRU)e1VeQV0?T&x8#e;PQ&I(xCZTppX_B*UyX-Q&u2A~~D(P0+P< z7beCHf`Jbjxyc^c>{7Wli~A^?k%u^O^3plD&(RR4j%>!y!}MX;rG3z+vJR(b3E5&Z zH=Jjd&s46RX5%C#g4){$+|dSo+*YcL3zr{bc{;s*}v78R%9Jl*S(fW`6aUe4PCakr3j2Sxd>hj8J3wgn>&=0 z1wY%yQ-0S#-b3R$*JMUaHLXIpLkF;33oo&*qAOy*0Se5oTM@E1NzuI1r(kcICl>y8 zB$e$I%pj_b@&3T&^e3TSTPJodN~7{QQ=sj51}l>jxSJ^onDZ^3UWPB`C1$?^-NBz> z^R)nGd0_>Y-gHAOJxx#aETa%cf0}_J?~`0pO*}ihz!?<$6w$Kl5O-}8a{1HLxyP?J zFyq9-Y+j8FTd`Tt%|8w&7T&{}g8ewpnTN%BJ-j^zJ=G8@L>t^)~ZDk&njmle!w?@D(6krik<598KtZ4+zRPqccv-jxMZ zYhzA_11;DY#ZD(4;rg!DaTkIdxd9e|+==D0Kt=Hxmu}z6-Cm~2oJR?J{|h+_Sy<_Wp{^6$~~iKrktBY zS5v3*mC76W>vQ+=Z!UV1VS5jN&;B=x z3jYvQ|1qJ8LP_gO-g4F>;^eHm9fn}l*voKUbtk5{#PDC5Pw|^=FYpR4M&cmJBpfn( zHM;JW05m2#z>T# z{0GNa8&Ld;Lwvn+Iq&Y;#5cAJeDaVw+Fg5vbZ+I~kK@B|&9PwIub)L~n>*;ggO>a; z|4QD=@G0-Lv5VI_KcA{hGbme_oqo1&rD!`N+L0wq1(_lAIzFDX2R^2`t}^s4W+iq- zl;QX#Bgl>0NlNF7D5iQWnRUL$;{wm=(Yj39@VJVu<-8@|Y7HFmv;k(A>S1nf0+sL! zX|k3MExVC`5BfVWZB;8OS-wH9mumFl^;0y{U4q+IB;)R;Gx(-6lB^nhY0`)$+Q2Pf z3$-q=De2eQQE5AV|F2oRN8ch|<*^A?Y&SuT1Buw^VaK0t*XNh$4dC}43PKAZFZo*F zDbiaG6>`lmQ|AfBy5#VV_SJlFa5cX~wvq2X+`_AdHSqO`1MuJk!12Et@qEur-dy-y z_um-v`5lt{+&toMzX{_9uB+siSQqhX0V(|cWh*Fgzb&<%*iSz{w7@VyyZdUK2~)b0 z#6xVESWQkIjbz-ZuXzJqwVh9;AN}zAlyZzUy@DeH^sx5iWE7p8jmt(Tl57uKjI!-A$P#B4uhZ`)3r?}gF)M~f-eDU!C1|40YQ75NjJ75Qnu zOnKiJb(&tam@bVzM`QK~9=Fh=bWB-}uhCNCEx$|g+bh11{n)d3DqDxtP9LOa2QSfl zpF0%xtCh5?Q*iFm7faYQ2x5FIc`x2graT@x?wdAt5$`=>7WB(^f#MR8o%D^ z{2fD(oHmEmtr!QBOTV$*4h7r;LwWXLpA^aW{eks2cR{b@BKj6R1h!h76wfyp!=G$c zB{^Mfu0-#G(4)DJGg&u^Yxz$aJG$h=G3TaQMJ4@5XCrE{^G6`3*q%ex`wG!-zBF1n zt3%t~A0oFAA)>tB9S{*Unv*y+oLjIl66S{7;^vjl5f555mq}e+A-=Br%W{iE3&;m$ za;)N*IJ9pLW?pjUiru8dlQrt-%Qg?PQ!rq&Do0~>O&D{F{=sdKipQ2`W$e)FTy9UG z0<{f@gXBH4S;5wJTqbVAc?DNd_QO*g*LMJ;-Ik-W-6?MI`(#!=J&lvsjDg#kzd5z@ zW}I>F31J_;g_*qiEU;M18O_vV2QH*>zU#)&w_XqSw)&5_b?!(I?L5VHZ#%#aPf_M0 zCE`%hSeWB}hERiVC>E-vvy+V`!r7Mq&B?>r*4W1^C+4?UQan~{r=~{l_W$8rPA76s zuCMXdha=3!s*q#Hx`8hb6diSWBpS9_meK+>xeG(DaT<-XsHORYtNnbBd)WGdu3p=X z)3qc8PN)KJdA||wid(tkA9Z-?7wUBOhBYWj59G7uGjN6O0&(|lYyQ_DDN>pLopm`} z#3~_gRX)^Qlyzep6u0bRb#oVU_a$`b%A7g~dODpmqzmk8SUERZBA;bjTH;^9Bk)1&4VXA*7ppv`0)JR9)(bOAbHfNOm=3_Ex=C!n*CX7G%O#ksV-8KNp<>U{gJ5@0 zjU6phVvc{laU%^Tuw5$)xV;ufaNb`NxV$7&G%ZztEL;kp1>+c8>d}Y3MMt>GGxJ0m zi^iem9|=}5U?Qg}tm^vhP2jjH$0hzQfzbgvP$JXDo*Q<-`~r3E#$-dTX8Ib+?sh@f z@Hmm|lwovOTZxTXp~M~D?1%R2eIdpG;JIND-e2~Zsdy$b1+`7+*lCL!SD9gAZ4Z{! zRIz!f>JU+?j~zdz!MEZXHdx+(vL1edR~7{81IIvYVGgJJui8p6IYY={2ck}?32RR{ z43@GVx$hmvxMNohKq%qi?7rHHiv%z3IahgDk{H8{oo{8`>Gqx)1YU@aSuW!zn{jo= zWw->HY5ek)_tBxn2~G*@+)Jd)%}P_{rcHW-9v#t4u2jrE=hO=wWbM#dp2_@~5xz86 zf%%22xaT|c(8kS;_3jC>+8VSU6yvsVZmDCr?WJN^{6LxA_qxJu>EDAk3wAQqxCHSj z6IrY^_2+);t`=oY8U{J7nc{xWO0m~_J^q@IO%+wP!Uu;g43Y}Na+~anW?;uqy z{o>6X?;6IhX|4m?i-Q=H$AkI>9u{w(%Z|@3$8plvSkS3pG1jER%7bUPsV9^yEpL@k zgTi4lv06GrVwv|)r7&GuMBTZ?4)p;DUs z^qmg;r8^q9uW$9~X!%fDdYHE$g9`2tQ-m!vmxE8z3s zwIIJc9lkj|2l>5H_~6A1%xPGSlKV1nrHndlU6~D);t_b)Fa;lP`k$in@W=9d zM@3eYk|^PM?)#kUI*LM?3Q0;5($LV}RAz_!2N3r;6FPL=EwdG%eg?^HOgp!q!^ynj)l$!S=eW03TKsqu!Zl0N587!n&W$ZL*F|7^aM&CNJpXV z>pDz4(StY3Q}~uc;k-gnG;v%_@kskw{2t$kdBUB*akCfg@r*u+{*8oQ{z%5J z{evdCwOIK5Dz+^S$4&Z47`H-Tunb*Gx1@74>X|Iv;gg98&r|SwF{(`o( z9Tv`St>|@d44Zq-gI&Bx=%z6u@+ov033PAb4#wOPr$$PPtusaNV4DYB`a6Y`=9mLrVYFpMcUYOrY6x2U1{27N1svBsODV9hsOh-%&l zk4UGOdvjV88!pW~8LfajQuXjv_DZ~4CFFtZLTGK~Z0PDsgBcH2z&ia97#OQYA1)kC zQv|=hhTfJ9hWE_!G|LwaAavX?CXr zn2p?f^eU8M;v+gRxqK`%yZuM~Ml2*JET`~GQp@>_t4CpQb`IPf9|_~@^Qq^&WU84S zM$>o9M4uIa^AGTNWQZf4aT|k|BSxXhJ;AU3VHy7Z!{e^@b-2I{h1u8+lv*+ZZhcmQ z^g*(q@F50EKg_@c4T?oNz2IW@6Xv~^Li_F7@O%(J$B(5D@OT=1f2NXXzkg5UuIR%W zk3O2}t_$iM#pT1j^~B`^5<66D0zni1>f+B;{lAEu>&2t z1uvm&2ibgmGM)T0m=1Qcp}|_KsiL|%^$;>YD+Rv8m@GF~pK+D8w*RB89Ya8JUK$z< zIfuiB=iu8<0*}Tm1-?g?K;oq_%yac0ls)zhGj#IkmVFzkg6}mt``l4H;Ijt@#|2?T zemq@MT}0RKo=9!BhoNQDF03qHf;|esXjHHZpN2Z)q(wvUT=`dEjY_yE;4Ti_zmIEO z?%-NF9)6swqBClRhG6XkC_W#ImIhlf z#nKadR;J*m+d;Tyu^WEs(SrCh-PFQrI2evHgv-vOAWhByq<3hd{VQ4A=r90Y3MiXWICDo5G7FV)Mc5F5SvC~ceKDu+a)kaf?x7n- z%!cr79rQmTvv%~06{IUN@cF>Q#5oUWs^JCNSN)it-zE!dDisAkl@x4fbwtsa(YVq= z38#HC;Wz!g&nLROlkt0J;Ny}xxX^bIZvD@J)?e14O=?GJ)K*1|b~D13PbO$qH4?%M zKG0i<-{}sgQxH-V4EFI`;mghv_#QW3_#F&}X&rI+;9V5ndvpp5rrx0!b;GH~$20W& z`3PKR5rU&$#Nob?c^DLO22aamV4~YrG*C;!iIMgAW{q&JOiYLMa!p{lpq~VZBx#$3 z!0wplPcC>_((#iN=}ogR=&IiVe;+DKM@kfqW0HBO@xy$nRYPRNiVVkb?Jgevb^OE1sgyD}{XCZa?_?gMz!erQn(J zh8xpA(a~{!G&Soc_3S-Ni*!Axs=fonofABrPXpkfiwTZa)yJ&IEpQ~@I6S;_2v%h$ zLdsQNYHK^2u5JpZf31fDZ>$Lajeko=pVLLP+0w$?=n2?P+XT|v2-x*M6TE@jp&@hz zWE_--`JVUSh;at=P7VaQpXQJlp&{hJ<)J@A5}wye!nna|AiZTQ9E{e6?F*!#=&m$L zc{-ZZp9m)}qAJN|rAtJ8g9f>|DuX6fouyxUYG~+ZX)M@r50qE|C_fs^biGLc=MAl$kL#rM4^lkFW*+W$045h zeMu$ta*A-oGnTG9GLs%nPC--ENW6X~5YsCt9y3zIkcohm!nQT=hDEuAelS79dBzH`t@6s~L{{r-iexhvMj83sLR4Ew*$S zOC3 zW&flvZ@;AHJPy%2#+vAP=PS&*+62;fa`{g`pKzD#^!VPC0`PTO0ao@$Ku>NGnvMMi zYG=#fnfwNr*Z7kv%&(xA-K%M_?mA&k9zySZJ&xypZ^2H1t1i_*>GU;Cau1IJyw$sq9sX?J!}*BDkj4Zfg#wWIUZIN9AsmdH=F$j*v{B2 zSQZitr*EEsFbNG(Ib5FHb2lK1ebvAr`y+iD=?d+WmcX6{b;vnv0Vwp0-gvn{%wH=A zUvQi(n!kWl`*rf2uhhVBrzF@+90to;R?_>IjA^+rkNDT00(N6A1D-`#KP45Gt%!%> zJsFU`)&>LK@i<6cM)0pprQVn3(u3i4v~xRRyMzuZe^$Wr>)wFEw<4JCb`EqaCs0+n zGo-lfEHSJogptankRkaVx*N{J=3%i=>$wg__~qh(c{?yU-2)v<>zIX46q~KHlSSMJ zfhifG5Vl^pBbiIUtb-Nw$A1@vjQtCU`}-1R-4)}9#vmwPD1!J#8+cWbjJwhgVeaby zbiF0PE=65N)4oKkSWu4PEE1JYicl{O(fLs>WGZXmB)cF8jp0E%*8vn4YOq@Gk9fDS z1REc{gFMM}D4Zg6f!CKn>-Kmse$WQXQ+Wt@H3;6``bj&j<6&m$84#a31gZ!`^i_Ami1o*@ z8)$C&7MIt)75GqLLhr>B?B}nf%j{kd_wHsA z;#5y3M%bls{LeHZx4R7g-B5*#?`Feqsp*jQ<}Q-9 zvlt|wk7;!;Fx;~OCFfP+a-Y3;f0rAIq_$$IF2_Q4k6>Yn8muxT6DnrJztv$n(YTMmGHM<{5zM1W`S zD9b%{gDefJ-m`L@K)jb?i3?`##d8N7d^s@nT$^VQ2#sVYtc?6%u3~KiRtL)9Itg+oOb*ZuXlBr zcw3*5Xzzv7`9E;{7 z$`ts4)jga`bA&}9Gze~D8{WsakUQ{H1Ue7P_z#nvxGCzJdGnh=qLX1vEZZZ@c6)D& z0%l}bs627zzxVlxbDu<-H~L@TnzHrGrpkty&Y87<&#E!zK8#W4f(@g1wSs=JPy1bw zWNMh;)&7t3Ik3^f+++;*Iax#eT+3B#^tF;aeR)ItW?zc<`8|$27;MciN}nO_a(GMI zri~&hRWtaO7=5aDWe>kwbqv3b)4AN0htp0@I}FIp{bT=Wy2lCdSFS#kXGyH=vl|Mihe zEBtsB(C5a+xDk#BBI|(3>wqEH`ZAc6Ke{+aq)=E*5+@x6-NN zrb(UQ9%<=vpJvC9U2Oy6hlf6xtADvl>IKi>ju$t@@^wh8JmvYfiF3pqQ|HnbVZx02 zbh5>eQ-?|L3~8=u2#7O6!pP>cGDOn<9-r!xLC4lA^5%8f7WreplNmEDc$a*8{ zddp6hsP2{Kq_#y6<|Rw^)tKeA3KgFH$dL;H!x%kGXc^2OrhSTXj$HXsJsBn6HGjSDk^8Rn;a3$?K z`6)@WIo&U5qPlRwointJxRiY7pVUV28xP#zrsUibYbFil9qyjsa-QAc^>0t4<(m}v z8Dmw5-Ctwk=P-@aac&pwt108dy;McL-I|lWbd2Xswl3f{m5vfm%naqt=Y8Wu?3k!@ z`YYlc;K0rOvr+VV*=$~ZayECXxu28UBIY(dPBAa;O|e*Lnhn#lPxH@xw{RuSf)7c% zkBe8n$vqDevN){`Tyvg`*txlrz6xmMFWs%=PEBEC)5g+6;oH<3`yTp>8`rbv^G#5hDJEh zLk`tMW5y!<_%IE`&kxg$;Q~`k`;p*_wk6KLXHW~9Ao_C0Qh1_n&UUNLXWYpZ?2p?! zG*uqPE=e0OcbTbd)40_v!QG9av?;1SR!6lb)zG|HhE)j8fvXl{*utLU)XPE!29->M zxT}+J!BxTas&y7Zj!T0__8cmuJe;)jK7+=*Q*i##Zg}Bh3q@rm)MEW?TK2Asqy_2G zO2yk`S#To!a!iC8NWQvYD2x))eZMy@4sjW+-wie7b-AYGZS{id=9&< zAw^<04~4XdL0D)dO@Ag{As(%dsq_3l_)oB|wXEpJq`JxM;vioZld+9C2WCL@T)}rO z@GQ!UwJ^inA9g+6NA!Pg#hpIhcsAMrttV{+<&`#MX`2PP?t2;)v_f%-28Y=x^0+u} zBqVS$bmk{%R8le^B=pkK;8TKYwtHgpdBs2K&HHaUXYg=`4Bs0|y3>C&NV z1dd_oSa{`gnws@2hOybMAelP{^5>Mm=(?Y<$@FNyRagHAisIqW&?AkL5UmKT$hTv*K zqnpsjt$}S$Le8?@g9duZP@@qM^nLJhkPp2Bz49*Dmi8VN9J~tt!Ys}4#eF6lQNu=p}OoHC0aGqYF1+*P+Ge%8sLnNk4dvRKedhDyaND00S#>(c<+L z%wS_MwX7T@oVLR!!H*aGxfyKtCBwWiQ}NEDWvIQ)3uhR5Gw1Rntl7ATH8!|e8c*C~ znOYWQdDWVNMuI2A?sNv@;zZG5heGihiRoO2x*>klI)F<1JW=#VpI$$6o49lYlJ5bk zY|s-WcDm7w^=^q|VeLoQvX^IA(32n%7ju<#6dr@Ufs>eF&3?8r=rmiJ<^gZ67sI|^ zdqG3VfE_Fv!z8ELvOu9DsMB$lp31*Vqb*!e^1uPqdLJ+F>pl}nvo2y9uT17G3?dd2 z<7wTyRQl`NO72jd0vYsiG@lZ6g}nJNktjyQ@@aGjn9q0!c9}KclW>O4`SYD784iK7 z*C*k6A(yH714!yx1n0+gL!RmY zFu5n7{$&&?mQtWQ&y6M)=dMCY)&Ll&$HR^P=A+7{cuec9#~V-e$>;$&Qtu{5_AI-L z`+8oXjlfmjF!LPzE6oDi?0DFr_?qf#-J!cw%;2HAGwlnL6lRDvV6tW_dJSEN=MO~S z%)JR*)A$(v-70&s({3hS@Hv9R9}29k@6vqS>08|Cv7ww(moA&$x}CW_7O}2hMX<@Q z3LL)`LcHQn8r!G@C*8kOOO>;XyIjB~I^JQbrx&uJsR!9$;k)}69ut|P_Qdm*3t1ba z$rP^ZGs7GccKF14w7Hs$M!6ZN{w{{yX@AQuUbxS+I=g9$+iP0rWDMu7sgNN_dx_qU z2jt6VQ=%9(oGzY~L>JHBL?v=(!nyetu)ujbQ9SU3{LWlWjs&ft-#5IYFEeWBxQE|B z;*l)gdID&fE=%-|ZJ@=e_C)oXC+qnZ%{nVivtD%tT$j5FtJhsbq1DH}=1W-eyEQEf zwqJ(kw7)RsOA?wJ-Jse^Kj>MfS;8E95ovp^MNaWgc(H66mb`n9&do0{HE9&FKQNjM z7c#PIBqrjNtNQrNzyuTL45s}~C#gHNfqJcBpfRxe(DOb8Jr}$|>9vEIg!EH>*y?(6V(ciXILOHjBUm{to`vvwxXw(WghARtJ;mAcXJMe73D+bhg`@P z=1N~a9>tq|=BVG44LLQb&{47#iaJ&aKG`^M*G`1!ITJ9|X)K#~XeRR#xv^ojf|F5i zG1H9R%gPE>Sw*}D8!I<(O3$NXfBR{1|NbZ=UbdB{&*qW&bweL1kvn(<0 z-kb-z0a|d_VmmC`97f+s1=G;xdgdhih&j2Hv*~NsV98oX{1u>vJu)9yg!5oa^wP4_ zR(l2}yY%q+^~reP#x^#@CXUJHq_N(vba1IT2A0d0g2BWPCUTBvKLhu&m)DLmle##j z!NoA?v!j`v*pOY@KA0^y-NQsf{;`{HA2Jy=J)9=+5988g&^y(U$qT0@-JRiVR{As8 z^MYc|^EfmXx@@MQb+pg(|EShbw(xHZ8?yH>QylHaEN>oQI}4AoalY#4d)Nj?3-8Q3 zHR@>XD}mpwMhkQDKd8L>4@R#0f^QP8LhDz3`Zq>_Zus^Wqt5-oKaHGm^#JGGQMK$Ft+two_0+s^_tq_I6f-rQoKRxJVL>gBDL(%GQ@Mhb0_%YfB-^k_Sp^5eQLUAfd9(IL4eDXSf*w%$HFArAgYsLQf zsAJgIk%Aw=7~Sv9V*bIZEOAN^W?Mc0&EW6Q{IL;~6aDe?<|8N@dlgSQIps`wPA0QIdrWXkmaum% z`RvnNKejME7+>8A!bIWS?08C%E;(dC8@JSwyEzv~`#=M2yZMM}s0vQO-v!tfo`JVB zHL(Aju*Z|V2ctW+V8rEcYU}Do_YKs7rS4wzluW>$fkD{2d=Ng8=mO>0hL%ko`j&qR z)Gf1e7BH_X{_JziJSHmHiywc~pGnt>tg#K2mN_pih+pT43qDZu5?jkrLm7nN*ppi4$CS|$%- zH=mDYGyRaw&NF7v5`=#qC*rVm7tpRZ4{zTn!kR&sam!(8R`g1NJrg_?HX}w@8t+lH zTw!Ep+1%a=?|cogaLO1woT5sbUPjTQ&p*s*J-Yzgj*xrpVi(K!9qLENQ! zfG+I_p$bn1fe{!$=(S9kHlhXIrplv^%5ls(a}B)=y}+t84L;{ShWw(dSkmyh)=QZ|ASh97Wi;!1tMvbQ`Y+0r7xe8z6_4VVJr?C#xnCi-Q3`$^2?;^cF zO9^(}bAi42i`n8^_DtpU8rFZbSx96=V8yB!EE*{|KSW!=_uonQ_4o<4XkW)w!rxTh z{1nx_LvUPEHioWm1EZWaFuv9YO&1g4{Zc8svT!~=DSr;PiG;9AUyrY*q(j(lLt!6p zif0|P(A0Jmz8n#PoxdcoLojrio18=K%yg{3U4j8=lI%r8G_8A;L!U|2(H+X?kh#a= z8ILraH(3KKM^3^WI`i?-(8J{WHG3+nok#Cmjt5_9RnYn+@Lm?R;D|E|aaidN^t!N{ z4+oyVGVV367kU!s#J1th*~)C|d1EjMngXt0)!R}@sW4cpH!R2-lu==m? zD0no^x+r9>rKaJ}L(7+$RML}k4EM&8d^S`iSM-$X2or!WO2hgrQoL*Y-fJ)j(LD!~u zD2R%ucb$8v`TaI3p1Tr^xCam*Fo930UxTs(NHe23~E8En#{a@HHw%YOGp;q-|Quse1*qibffEujmUMdfbh(KH6%WW0n0 zkL>YrS0eP=)TiQxI!H|8TeOfkih zZZ(u#WQlg|d9aoL3X;|4 z!1t1HuoS7F=%FxLS=j}(Q)c3F|3x^oIu?BZ6@&j;J6$K56Pd{8t^vVRP=CTS>ke8&a+=SsVdTfjm8hjjOm z9YWuFDw7_xk?8OBroQiZA^s1XB>3Dxv7_B~tRM~8!^S35~(Y-9_xp#}|+%}}8%hY&lW*=#;e6ebUc^==XTe|Z;3R8bY3_E1f{ zAhe5rn$SxVEhI=+#b%z5kfnM=ojh1On@r3)K&{pd5G2`2H;;!QNvPUaZxoM|TFmgS{@ou9DEoycjCst$`~JT_Z=1b?6f2)Il@_ULOXW1RFahD92nH}5J$%E zB|4$ci3;6FUR$5l*@ zBR|(uYT6h~BA=Zm&880guf?4t!T2G>9&n|yo;76BZY3CRqQw8JE9HX*p3yk=o}YH2 z6xO^f#!=7KkPVx=>E_OQ=)Sa$dWhe_QOzQheJtW9uX{+RCs`vtJxOJ=rz5{thsFJp zBe9Y;ut&!X^%ohEJ)4_}!p5_7l42S+YP>Wt6?~^_{E_5T3zGUZ=mHHnt!rELM;{6c375fP630Bu~c4Z6-KPHW~a~Zq~gNI*fzJ7 zDF-Ff@Uv5Zwl+d*UJZ9{(oV9seiD6nTABaiK2}^m^)i>N;ZBB?1qyS$Gw3O|pEnNC zg!gTBxWa!P32Zw^TSCUczS%?Q%hVkH{J#i-9ztFuYc|YH(8ZhbN}x8>2TxfO_Hb`C zf9BP3s8zg%i>|LEuRb54Rzfc7Q{Oh4f47GET6^Qeae;hT`X-Xdje=6Fzerq&*f)O-Y|$X7>yuVo(FiBb6?1b#|Bi! zJd=()`G7_w&jgtw!AEf#V3T_&z>Xxy*R=+{J9)5lV=7!9*h7E4I#16?R8r1mC-w1_ zqvBB)NxhIwTt6)nzS$Q-b4d(bbch9ay9n5DXewm9s-H;a&JE83NepuC6Ku^{*Qvpv5 z4d>R6ZDCyMOraMrou_|u73qXu zF2tN3BzKl(>#(M?S`h&Q*Or6i0xY#2fNs%B7+T`J&fy-si|9ERjW9D3_H;Pt5v*m^mX8W_og zb%7?BT1yD-)DhVDb~1_shob5dz{d5C7~`5pm)4Ggmq&L4M}&D|b1*gv+<~hOVQ_J~ zkRP8H3p4lLpr_2FVe~mKcwhX3Z*#ax!j`@x&T^$>=6z!ab`@VflT5ajbU`bUx?8?m7iDyDi49ClhJQlI2wD?GXCJbrjL* zC?=vhRXWe%6_uNQ0QLt(z>R!G82<4e{n&n!+AK+-X|oIIreEi&dD0{pIc+OQyvc?? zuhMu;UQDJeox@uk-3a!5`ytJ5F;sqV1)pQ7aQ4%ApvMKb;?TbkB==Lu?h5av4UyF9 z-88!S`BaGQI}D#Ee1e%)+i8IA2r89%j;xt60z2mO_-wR=;65SLqCl5!K2<^1&m%PN z(G#+fq!XiO7Pz!B7Vi(rM!i`%a6jY+a1UOAN=GI5T~Wb3k4E8xKYvK=7A-o@<3I9q z^G~tf(pzLpLm}4@`=suk(yx zwefC{!4LF(&kWGe-U0s|K8>Ya&G=T-iMzJk1>G-aA?9=z48XehBZbfe;^}pXz>7lcr`@Rpv z8a(5tet*S#Ud|;gdq2`Wi?)L7;EnM6)+xBX@&a7yD0^w-xg zI(X|P+GW@eodud0C`(YT%N0z0Q$eMo70l--g65X-0P}``+inT0?i`9$Pe-6dp*}rU z6iH8NJf+1=ccEH2Q1HN3%QSIJP~;smiNh%!QE$Z(1RHm zz3n-tD4eIh-bo>DXI}CB?#eJzLxqU@O-0`BdvUS12Zm^$=Z%*33qG5XsOIs4n%Cw~ z57pD;&OAl_Se_mzWuK%we#nyP)@k%$lrp(_E<7a7%K=a@s7}{RWTlI%QsqO^Y{=1nZd#R(wiGvh(J>Vx^GDFGK z;c)7CJOBEH99Bx^Luu7_(2w{HV!;`kwKp1UGxgx|&rWWoz79>er9eBehQDQ9#K~u* zz~G7|Vl!EeO8r?28P|@`d#gW^w$o;~+v+Pl7t~8$Cm2G@%Bke|mMSvjMgb-s&_$Qj zhvY%A1qN0%f&aICWJ16p?0dHWCdnNo361)6nDYi=m2jDhks5|+=^x06SZ5M=LK~xW zl4wv#2*1~RJv{Yz%X!sJ8c;lsayG-m7(KG)R^Y?DqspRio_gR^sd!FL$ZA>Z`Qg5EIpkbnJ-XZLqII$hpJ(+BXKf6^*OnDTYSw4)Pfn+2 zE^Xv1)MR18!gSK?F^e3K(8rP+)8PYeK@|4&(EY{nyx$sOkNx%}4C#(G8aSx#8tPSF2|? zwV7JQ_h@Z`M(f$s+GHeeE^uE4qEtwxwJhBeHw*49+edTO#PN1!WAN0mbCBWa$X!n` z#|vvp;iKDHZe*IkFTeelY^?F2w&M>%(?~m*`=1mQt@;T4rz9bzEC4RKnc{-RN_y|V zLqz_i9QuYFCtokfQ_D4)_)lQrY#6zO&Z!BZi@k?a(a?9I*SVH>H+mtAZ`sIS54{ZQ z*JTo)*atj&u#(3AiYC*SP360WOT(((?eHM)692nm0lP*#S(BL#J$~yjHB1zf{5}z1 zpvl-%^%$Hq%$Ld-{Rj1yK19_ifFG8Tiqa{P@MY0^vUBzsjM01tEq}CV(6PCYvtk&{ zZd$;<7JPa(KAQB6iw5_m%>~4w050}wH2*p_kJdexBkzkfiQUf}sOBsVHr33O##rsc(%F z)rnd{Lkv&Ro(X<9;lg_Kic7+>+llz5^)&9!OTfW{9GUBc9jsucFH0C)$ab-_%vtcU z#2ci-bj#U5Vhv!Cg_+=vAuN8T89QNij4Cd01fvXdc-pcbW>1xasR`HVkv%8rV2@e! zc!3?AFuU;gK9f<^f*Cw79N3N;U=^+ za{=wX9mX~|#;}`z_p`9%EO>I!0zzi20=`t<^6gC}%T>kwENg@$J(l^HwzXKm$k9uf ziJ33koaD}i9s&4sP6gJe=aD3D9X7T}l}%sJhbwfA*z8w=o268WRX>toyT26U*@9Gj z5*@`{)%LMbEndvjrV0lCJq91MlOSWK91D}sWG{<#nVn@W6uniz2i^=TPCTc#&zr+3 zp_4P~*lj%Z@gKhSZpYon_A&2_qio5S7Xd%0ln4Lf1pm2mg#5hMegj_`0r{R-`-xk8zJc`_%;W zo&AvTK?XEGcOIR2;SANSGJ;mMLy$5=0z9J+(dWUNXt5O!dYlsc80$)ds0A4A@dkYr z1IX--#c$?C`0`{KiW8Md^(jBHt>-mAI!ccP*bibmAHT%GQ{8EgFngJ-sR=7YYRqKm zWHdF96LOztDEXUB%vS4C-|g10$ITb+cb@>OC^Nj`>VkpF>+tsaU2ImjCv&*Y81Ahm zdljVVm>Uylw3;-VI97sHH+{u3iv|gN9U%|8@)ZV*k7W7cQvnhV_fC zgY3~x;@fFLohJCvh~iMv<0%Ojk3~~U`I9(Y$O=|G8jDH7+;6?GTZo zdbs`q`15Hkd~D6vxrzfs^>(-{mN4 z;EjEHN1&}P0ZzWA(3$EBOWT74CRa3EofZdemluO%?+Ne>sl-`6@mP^J6|W_qz$N>N zvBB{HdI%2G6OOI;Iqf!%OtQddCw*`*JA&m4!!Yv834Ad-9c$XunEsxpXlp0%ljN1~ z+;S^)k21lA!!o$D@;eOntAK~LhsEdjkKUJF>_}aq9=`|4Rq0wUT zo@M--S+=}&P^M^`w=8+ppTx_iZQ#azb0)=)$M6p>cbXrL-OL#+e=5!o`zdZ0_?CRo zbN-XYY4Pv$GraUVMo;CGS}1JI=aq0+JlD@W)+`aM2qVm@b@vDVDxZB=s z7Dsb>#J^Toh)ptgSWK9*gI`mcM22tqWZ|{Fmb=?$#RvCxiB_Mh<0o2~Q6+nKaYA^M zz-VzM^SkuLiybGDHk&Gv?zKvER?3p|82XA+j~>oXHOL`ju7uO*LV=HYLzB)KWRtJr`WHjMuSg692vSgFyCwr04l;@i0Y8s-`-hj!W?s&2-HX zZ7r7ItelL*PQ|sPSI3`Zr4QwH)cNwy+AYah*#c4FQ7!Jku?(`O=aD#l)h%*md=*)q zUuV&&Zp-<(UgcFDy|L)-Q{aF9^5I2+PesYotHq`{<^1rbZ1X8Mba+hq%xT2V7nMGf z=N-0KaYa@6qWya!N&6*9+FI5v_WmR$Hplyj_4*4IYaTU=lCIj2UHfN>2M+!f54QPA z3jVASg_j-|Z~FrLkN&aZT@nFe!(mUz7W`%L`r&^;zqBit7#J5*zsQct7R930}qFEWbqcn}ww)UYs_%i%&@1$5GWmqHj@z;&n_e=L zpqamSiU%zfJeb4kiIUoJ^YPUOxxSOPxd5|K;(!On{KNrYveQ^W{76HKri_rMZyl6G z;?=)J4T%~2RN5&j*!P*+;xmNEKZ+CYR~sq*B56yd{;LrA<9|>#K-^oK*seC7d0%==ce`hkjHz( z7K>W##k;HrbIXiFL?_QqA&No~_@vdFSjv_6{h-32fCLke0J7v``tH^Axha~;>aWm)qSn-_EGNSj(l^L0B z#m4I|7BV~3^6sZ8miEu*Sl(0bqy=YZkz$cAJ#k5kRwz3OXYC2Z|A{u%ZwZ0XtLq@l z=?JX~sp4B_{No1_SD=h20e?OpK&_pPY+%_<;dl5FTQznP8#gdaaFgG}Ed#IEk%LcJ zS#Jw_nl=UJ#y3Eb;FEIw`vZTk&|$xujG0nGDrR4*0EJkGn)zb)a5ABPQpK{1KGC2-c0i4a&|IiD_xXuk6)s$O-#02hbZkJ z_~7#gG(P*WH|xUL?C@hud&E2%e%pj@-}{^HcX>fY6L0Z1D!cji`F~-1yFYxBXn?_w zBOza28g@>OhdG-zuycxY*w<&KEaPz;?HV}GgF*sLEXii68}pgfnF=Ohkw^>f11Pu;UM}-J-?8C8oV+9&!KM921no=65yo-)6yvT~XZn4qJ1xC#0!R+7qF-*sM8Vehd zN#BooOz&A6z;UG&oTVu*jFiA63FugvbC`=gx@ZTTI( zZezG+%ybfCpUPS9KETh|^n#8`6nZBw$KycuHA;Vv=J))Trv3%1VEF<=`1-6>c($K| zR_=ZH+PVW;_l08W&~ikn5*+F7LdMK~#fJ^e=6~1l?C~Qtw!A=&4Ii+l7t6eecwGj6 zYqKPqH7o%&?@6-oyE!zoy^_Bzl}d{olBj;vd77zMNe+H2Vo@(Iv7^7Punmh7*#7^* z*hB6Rn^vYm1#&tGd2yT{@p?Rab$1fm(J_NerK&Cft z8qG1AMAm-aOsX2za~|%E++iUvuQk1m#@*v!sFeZ)-~S9F4=aJkNC|Mb^psxxPn&E` zJwrwwzfU^GmynoEfrRfb#j^fJ?EZNXzfBP4rf)q7y!gmzO!Z-&Hx@G9X&#&4G>h59 z4`E|hDzke-ezM}F@|F>)nwFAwLCne~l5Kb$%Od5Df#DJ@2%S0=uHCL9*L>gd-;Vtf zKX3eqn}%LPoBOS3{dP9|OiH8CagHNpKJt5e#5~nTpKz9Wnh`=5Gc8r3CpAx2<*n;;JSYc^z2#>_X;<|E!>Zi3j1*y z?!a9tSK&(R4bW@Ogksg3nEvfJx;zlF7TPITVKN7kKTN}=e$wDN;RYRKlR{neE%CEt zFUY0723V4gyRxEjO5bXfG!3LK>%C}SxfwMYp9=vO;$d`l3OH%of%H8$*!gBSY@by_ z2bG_o;q_~&z7K`qo2HQOqX~M4meNVvSJReb%CxO46q5Xk>Auqe^n>oxxR-tn4{}?(CzZ%;xj%!FuQ(7w86s0=PeO>nxhC6qwz>KdK z1#UOsnx4DtU5~xv?k*?UD{=_WfjHES4aJwoPT;EFZ8)WK3r`vZU|C5PSTuS;tFRMv zB=OVrneXsPP$_y|9)S@yL02696SgphEVEY|Jq}7@-`4lgyx9X-%-LplmOcnN5NP^E>ykV4|z^juu@wWl;f8`j1V8oz>|RY zOR~}Y_z~3SU5Ry}cNrCq8S~`5EVE-`4gSbiLOB6-%-0mbflSto_EZ2x5B~-E)A3*v zHwVh~JIJA!-z2@Dhb(Cl!bS4E(Bjw%UZGqfdRUcQiXCBwP15nx%NP_n>xU~%8o}~F zD)5CQK#ETX3Uqhk;mjr+DIUcaQ$3h%+Jt#_JF)SQB;Jv$ftF)2m~Q5Yp62Ip-hm*J zVH!fNM3fMlT2H7vbc=XjUQ5P97UE|HF!zrpMqY3N6}v64dan}vH#!Rfg5Hr^HdVw+ zv>nB+c;fEM^Ko&s8w9ko4hfb$wER>Qb?LuE)4$o%Dy5w`ykalrc1vK(ooP60Ybs>x zxpeQ$uuITB&(_ZBmIiXK)lxWZ=5}NdYaSn-llgQj+{dpy_ z86fR>0Jib=FnfD$pyb2>%)YrFy?pf{E~%Yd8VDtW5sv8c%m{z4kVB{I_rW$~26%st zAWmYxQDRXClE!KjC=Mg>Y(GM>ViysnUU>fdHY{4U7aKpz;5msF*zQvf{K{v@LyuX+ z^DdCmyMZi6@+fSb+z1Pto?zln8mk{QJ4T2{78!%F*0Rrn~ zVWHIeTu(i&@SA!%LrnBIMY?z@R2dtnJ6jdw8b%P}EEV&APS5rBr=cLK%gPvrXu_#>ITMmuoC9vVS1@_+e z#6^m6sK!}MheSW&{YYW_Evk-P_gD|`$}Jf6$_wVYi(%TFT7;>OjDckv@d|MPn|m9; zINb=GOLD*|vlh;^&O=)^8@_1WG+1%J33S&OqWt7KdUe2@3VpVwI<*<-$sdOEKQ2La z)kwU2RGrFpzoJrte5k$D8>Ac@Np8@6EL{5zg-7*hPiH?k>$byQZ)x& zUSh^DBhQlA6Ull&=dZz0i71y5_cmlH%I-9>bjl11eMpaP?eckSYdR+1T5 z_bcJki`DpL~;WOoj$@a{Ju z?uFc5YTlPXW9}Wt2(zbbCP<(55)L`!VvF23c86oBj6+{mQO}AV`q@Ap=O^{S0ygvR z?_36w)v^$Ey0qN+#*M9|E66RM>3 zK(_l){FVA0w_pc$_#eVNS3VRMGQ{R5GUS)&ZIV-Ch#~B6QETc?2rLpr$<-t9Lb@Ne z2kgf>au*!a#OR2!3mqK&4|W&Gqxau~Sjc)3-nZ$Y{}>mIPpqe%ihgvNY#99L|nHn4&FmDu-2nQBY$(ZSn4@eRvZ`e>R5 zU$P}2v_yvH`x(#!v&`vX*p?;=8@(>EUFu~V2XhT_10_Ib zNQKn0cloU97mQ&KA9Lm0b>`*{SEhYWC?h%Xh3U*=XY3C@fsI-Yi1tW9!@Ll%wr9ZV z<6E}x@E>TbpN3zr*;4IR1$ut89$f>~(X)=tfclzY&Fl;k+qnuXcxE6D3DCm1GwJWS zv*_)PDP~&Tb>^{;D6@NO3eie2A?ZpoL{BXRw%n^E>8n+UQjIx;xy~b7`E3~9lk8{o zL>wv~ynz13)8KhJKPhGoqq{_6SUs#74D`{xOMYH-0Cjxvr| z3J}Lu!g-})%h+$vM3)IQmap|1-rJwTxx><^!n(_`tpQ_y97EMYZImbzr80Nwu=Y|Q z`q6ME?t(XGt%VS??cD>KS)EJg@P*UMV?%h@{|Dx_cB56>T1cr)f(c=7fEPu$B6Kx+ zB)x?HtfXjo+G1++cn)iWjKwkbJ6e6}8_ehvM$N`TELxDqIJ{_N3bpjn!FdvvwT(jITrTtB(jjJkhdE(> z+{Kq}qj)gtKTO*$2N7F%U_s+L2z@L;vMyRN@xFP?lvz1B)aXEN^mH=xUkY}~=i_ev zL{#CuX9$fv9IOA{aFjQ`!N^q>!2jg^^6qb&etEJCRXZnxOzG*2>yj1GiGsi#=CIx*EDgo zC7f}N_+4yn>A*a}YpAt z29aC`NsxV~4VAGGM7*egD7!^5={lE4`)&^s#;?rG_!z)!(ia2KUHq{2x(G1F33x93 z9zORELO(fk()UA-IinZPT#$*RzGgFMcQ52A=J@BhzB?$0cV?H-Y61NeyH+Sn9XiI7!atrXgnFw%)^r6wj1i0(ok}DM)uyDQw zTjHNW)2A^I4VAYip zK5!Y0va&E`O9$GzPNU1JPJz(jognTK2)*3JxL1IKmldrs)_*O8)NX}xmcjb(TQ7KN zu;+@`N029f7ctrYi_zEevY()pmPPamvS z-zTO;6dBaFf=HFIk;%o+cKiEg+?~W&~k7J4cGM2p- z9}h{h|AKLxHMy~7kcibfus!CVSiu#hOXX#$+2DC1JuD8lqzYjC#~V=NDFBIL*0Xsv zm!zGQgY50sV9&2FSp83wzJ9>ahnvQT%H!D}T&Vy>|FxjpMji?$rqhJ(YP>zvf`Zjw zv1qdZ{xVTQ{I&|uNy*Z`m)V~DgQYYq5`Ga4>V#pk=4vEHG9thUyN z%?*B_=KqCr&uSm@ZdDNRcv*oHcVA=l4SuTNABln=ZnC@547B*23#O+=V3sb+6ni2B z(#Bgro;Ma|hl}I*EM;7s%;BVLGAYN;i1M zOW@vSbCy{if~r%!MphMXIi0Bw7(=hSOv@W%vLr|yqBXXYg6J`l=4uB~m-yjEoh4ZH zMZnWU2D~Rvzy=3@5G&sT<}(sNM(!_kg$8i$i(ccHwqIwI4*20Mml*W7F2YZmVML>v zA4J5w;MkiL_{=5(>o^toJIjqRbR1=>-JQtKNEHaH^#R+!FJKW-Pinb(5V)@grX9+{ z>A}4Sw{RSNt{+xDO3C6j;kkOf1&*6*LB;vna$i|2|u!!TscE_bG3&|4x_=Q8& z9&c#X^dK3E`^a2N5rV3`;8*Dl3v!0wjiCWF?!E}y=LA8!wK?Vac9y>L3#FpZ%5bID zAZ|Id@ehL49sApO@{>2zo z)m3oG^4S>ld?jYZYk~xxhYQhpVDpiS(d9N66%>l;tRr@_jutZRW_XQ95n9g$!Btyv zoEYhZiu0oQS7$xG+_wld&u+ojH@%TZGn-zyeV%e&EvFuTc5@3RmvXbcRk$jJtP^Hk z9tlg(B(iP%hF*MW24@?<$mKfAxe!hz!uwjugmaaV^}JD|rO{uRqErF;`0*p0X1%=Y^kT6&-KE`1j)&bOCv(SGbUzS(jc*IhOslFCnrh=u}uC}(-g<-;&PrUe?`xzO*_lIm{VNEfImFl#6F zFrzU>WZNMxG8!sQoEAGVeioFRF!Cf#a*xRUL;B2ihutJoI*Vm+C*ihj;plF44d2h_ zBfk`MnF9|PX70sqnxJrtCMdhpgO^*#(l_PgIj_k}}fP zGUi;nc@)IL0UFkC2A}6QNKLRi+4tF<+)TBGdD}NZW8P9Q{ICrdH!&y`BaOqXcP0NK zFJl)hPBJBuXi1$veKqMym$|N?vRCw}?oDADcAy!?Qd7W|NrHuDQYcm^f&O7Uc(2!) zs@K|4A=}k7{^xIkUvfbvaR}surO0>g3sUSkLc*J$GSRg*BxLRwag7}(izcVYoQZLA z>Oabe=KbPWw(xTz%qkd@_@_)P8yS20RvL#gE8)+cODrp8p3xV#1)L699?n`eKlH{c z5%+XDv1j~bqH+H@6WxA=$(`B7DLwa`X^Jl)!Fwl(f?*LkEc=_>{N9hQu9fUvD*;z5 ze8_QmlgD{(Z_;d+1@y#5DH`o3P8*dIIDdXDV6$a!IPE+J zG~IAE^)391_7ivD(9L~dZ2g3+VDHplmcGD)hHWU2Xi8+1{mAC|nPh6pkFzUn4I?h) z!+gjB&5q+?S;ZdJP92}os*i}+VKRwE2pV17e zHS`(JzI}#T;w;0!<`o&5zn9p~NG388(T2NUUgyO3^l?hcj^d0(M=&5~H->Nd#@WW$ zGRu2*a-zbW@t&$YzV6S3p;a?UL5l=AQ#Qg(*IC4rjJPp6ixtTIuOskD^f6?e$^bQC zCpzS_l=|mt(C?@FNw(z>IcOjSTa%43AYdWR{V9%tZ+US4=eH2@@C(%QO(WJWKFm+> zAclsX;OyxEG`oRiUj=~R8*fNXMKC&0iYMih@yqoqXx`@r>B($|v2rb3GS~nM`~<=B zL@yD2^o-3776S9K3o;e_!9w*i>?=MAXKVGbzebYf^oikXxB0jLK#k%C3!c#SG4xO)M$T}-j@2p0=hy5P(4tN4LN;tvZ? z$Syh$*5%7VZcPcxGD(L;YQ-S-)B@h$nh95Y`-s;yEz;qk$<&ZNjPklUu=uYS_|@>i zl*V%Mv{Qft{PAZzWj8<#-9i3yX(t_TlL?HMkowKpWQVg5-r@ZPiG`n`_vAC&bGiiG zva@lfumQanJfBV*FrZR4O?a-X1(#lVf;qpt3Gpc)nqoX~tSyBkrH(K*)dD2%xeGY3 zto6kouEVGI-e5Q5CYu$m2eUJI7%UNwJb!(0;?OocYC8+p$ctm}{R=41IzcmLDq{C; z0en)@5ANY3z}ma;yQK!c7*WH{-H%AOKqk3Sa)hY42;k<^m9VZb3Wi==;71z`)JG|N zby5U9m%WFku>!c1+J*~l?_)fT$Lm@w=TaaOBBDQn@Xy<5d?y9pe96Iv*%!fP`Xp(d zD+cM4?;v=?BghHwgp8mdd_Qy^6KghNiqH(ard|Pw@%2D7t4UVrJ#wD8Pk6r1hpV3? zLGu_d++OH|mf`1cWc5D$(RqqU8S;|yMsAxCxd!EufbtU1=!3pfF&mt*!`D2 zR4Nn_%cua>E6_nkuAWDppG$Fd`5LVL5X-#wJIhJ_{f?6$;z_OsEGs$-rVQHRNXBUoP&Mb_DEgZ0zZVC4rs@KYARFWwK}-zgu^PF91WT@T6L*eXIo z&*Q~8JF#!J4R#y1gMz*KO%XSiaWTFj2xR#}VV^(CtvT<4{JBn;D8GTqHl3yB*PLj2(j!Q5+XAZu z^@)~s6RF6hz-4K~8Qo?2pbkp16WK9IlPbD&>a95)7s zpnG)&9@MkP#Kv;;zn+H!YS%F0_I9*hu@ATGbcK&r7hv+lJ2J|SA{V@f#?yoGC^=_7 z7PTc}?Js|99XkQ7U+rM^6Ve~n9RG3F%=9<8no6X>{Lk(}1 zil9kf3{2Bg!WV@am}-)V`Q9y{(ZP!?iL1S=)%Q=tUDxegwu9S0Uo{0 zCH>z%a{{z>q4I^_pvfqb3-Y{pGNKaZ>WM+a!B*V$(j>Lz6#j z$NU@CF3u#PN*CdsVimYuu4nqT>wxIRVyJKBvMl2Qly45l$k+MEk@$vPW30<%+>z7= z@W7`V)1iCr3e;Rz1?f}1?C+==ol1*qybMVgQ-K=tLk;KFim*d5CYmlbz% zPR|)-HVEz@r)>?e?c4(h^AUoL_qstVZUBOvhDgHANC?e(49m}ELz|!*RGjgK(^kT8 zu0R}%!n0tnnG|eWVU5L>r5LNxfFHK^L$9elo1?Ko{>Td;HvAWc^rvCg{h9cD<`FcI zUWr`|chIxg40C(RfO|U%p8D5=px7%|GbTd9&alqAMKxr!vzdCA7t=icyEOZL7+2R)UV`$)+PES=PQmLcibo5IW_MPLW_wD(quthK3w>5)q==G*Aq^F^t$soul zkHE6^$&{ZbgHHU3q`5ykVUv#$zPR%Nrh6X3;bwK*9HfK;DvF$i*|wae-+LM1_cKtV z^9OL>27-KlIuRe;4TF}aAnJ1r`QYwJN>al}EIWr6EpG<5yt{xO{m5D=E=Xjmz`0pF zP-T-D>a(m~;Z|Lm@ox_OEi6OhW9*m*kH3-)3J;)up9tO!T8@wXLNGMpFx3!RO*5Ri zbk$KEA{6(ClQk=bsZsn()>bSf7FU)Nx31TKe>)h{-6M>I`~pz)oiIA_b(qucW=#j) zSx~D(>*!>_L(*{Cp3F8-g8GdMxXCM2xtAA9a>E89nQbgrWU^fas`!^sJCjAU`S>iV zq^S%#gL>r5U>tk?;h{bC)2Om~JNmM2>$@#oc=384er{^OmMR~tUF3#eu3n*sC#~tR z!u53EP9~9b9VahW34+t|H}u5f*VH!mDYZ1;MAu}lp+yZ`%D2#&1fIDJSufJy>OwiH z^KmYvJ7&=$P6mDV`X24Mlto2@xLB>Y6UE$DqNkr2j@^ndI+1vVnI_lHC@W*-3CX!vIZe&%{1mv@yha=lqk6C6on60X0;+9<{gOLNcMB^#0PVK~s^A`*|w;tyd zy-{JTR=&n{LLYFDorkaZ;RAznli=%tF)&pXgxI1_Bw}q586GbHeb;xO8}|rG-o)T` zpFXtAdW(U6)0kO8iALHY26%gT0(Xk8v3Ph1}X-a9sKW6s8S3@r|*Gw z$ydNj-m;mSzwlhek_2m>h6M%ou;5h-wYb_qy*(b$J?rCXlZ^|#vdoTtN{!`2sF;z@ zFw4%Z<>}amEC~c7Er{ZA}^p$K3%S=kg{D52>Ii<~%R9qv`Wd~u`iiJj+ zFPd0S-7RL0c03bVCIT0o`iZiD2`n0zOI99oARk!|MbY;T^7nindHcYQoU#ZZ%$Xsw z?PmZAZXd*T@{KtE=N7heCJo^$S=OokQ+na_3+i^ZjYcvH>D5pX`s+Op^*b?xd$oUn zF1q-N&dCU&5^E07O4}`TaxEX4hO+1T8fC1mkzahG)X-@6!kJ9LP&c~uy5qFfj#w~$ z23jM1p>h8zSaCNH+rm6^uJ_f_$QKAyqPLxdUK%AvB8qW+s z9{v+_i|<(~D`i7Rm>DQsq=1h0V#t5)FzQ*wqeplYO7slj#DFK>Yw*V>>fPj2+88+? zx&#h?3C4@}uHx#4=MkzaX}Zc;TG)7y(#}uZ4xJ?KuHxO?-c47~?UEjL)eB!>xH`SovHDU&kt7%He&;AJ4}vG%arDYw^6$byXaBV)zqL(75DSy!^7ePFw!u`$H%pr=|i)!tVo8P3Nd$?P7Gr=QAw4z z_-tAV-VD&DddcIcNU|_1Iu2zGSHsoEakxY2JLVV_p(f*s2rT2R{|s%lm!y(!r|1%k zB5I|6nNB0usUGtQ{p?k!k@sfdaIca_;Nm7n>D7CsNdAnmL8Px=JAtE_|CngYoD8V~W`r_i)A3!m9l zU>VEkDecbyx7jI>wDc_ME=xeY7fz_^DaJiD_>-<;JCo-}ypX5w39O&mg`Q0sOVkCX zEn$SqxjDYJ)cR-(My+H!i6LvaiaKBD-;^Bcd&QBvoyu`f{x?iN7QDlvp$IfP&(4AG zCnBfa7prn@@zEw%4E$;Y*Q{=lnCGVtN74jj6C0i*qS z(YAt*ibMoruWJ!nXZ2F+W%cw)0GqLXGMk%_R85zJBW(&!qh_T$>6g$S2#+q%e+Smn z=lP*{ze$sdc(vg7MU$9VDM*C`qtPwF2M2iO(LaB0qrr)__~6wdlF#PzK6eUWLdjP+ z#_`8OF)7+ro6F`dH^R?n3>c3E!UoxsV9^xdX#S=%!Uez)(IyDL0Q>r+CIz&Sh} za+2HNYVRPC4w?s&;S1sBK0nZBc>#OQal!bME}1QGk_g^% zlDKdaQh0JUT)pCitM^CY<7K(Xxl@Cgo+X%ls{&Oo4uVQ}3b@;_o~V5y@FIUURLWIC zUEl}i?8OblF`|_8G(RUNSoe^Z#}m?gH44rShr#s3JW$h-gtrn>@W|a9CUxH7wkN{$ zk+KSviqXIYs+zcL&2mhbzY&^idGOky6r>J!VePsSxPIe4{F%tb)`?g6^{pt?JnH~= zn!F&%e;iI$<>I#c_4qq&3b#()!m+1Kc;Lb$4xVyE(UBNz?J2~E-iq|%oMqIb(S@!a zdJgwd2<6uD=njbQKioTe)@FA`m z>%|CjKiJya2j-OvFl=-M+^CL(MQifm-P0@J+*=DlT@$dL?U@#;=Hl_5bbR(i0M=-I zB{MQ*q4X)i^FH?IsNs$S)hsVKItP@v)v)G$B%XMa#j=kJ(3%Yu+>p3OTyF#t_ot%J zWx(d}Mq;4m<0}}?8-#$He5gwiewrA;8u28UX;w>`YsH|vY7I2XodhN8WUvZ+0kK`H zIVoMIm>1rMIDedYP`u0zRU0x`rubWsi~0;9g5z-gfj|BbZ$KNTIrQO@9P)707vfku zAG&!cE~r?JQ5!bkv$RWu(bx#W^=mQTxdvaU^y4v)S@e_CW}-p@m}&LB%;qU0mO;g` z#@X5ax%WVCtFNVJT{hB<)jHVyD$im2OlDOs`dGt$x zV=I$jC@{a7SyVvLHEJy`5%k7;Pg{Yteu0%z4#av*k_{V`Vewm0c+S-(`{&Ljvoxf^ zx3?J7^=d)$CA$j?N~UX@?ons{2h?+$5?X~^!O%_hxJmsF`C;_}`g4QO)A%_Q_%c{Q zs<2(!5^OD^q2*i?TwY`a7U6nOIlc<6k1mHthH=oEriEXhMu0+U9qcrE0TP4LN%2pA za$RZ<6qFBvtIP}7`{ox2&7sI)pP~4_ThXzg9~Lqr%-e@ZU(7r9&{ZC+}(T$QcQ%=c`O%QR3vb@y*>&pGDDT$@>u%B6xS}d#>@BB z;fk6N^7*j42%GzmRuY05RX6BYQ&+P6ODm8&{@AuC)yQi5J`P{l39`GsncUW9b5uPJ z)clz*y#4!+bopAt_5FvD%(F$4lo0IZzU6rOv;0w8Lu^oK(qgz6oQl!0(Wq=tj2HjjGny@T0(8y>V(G2b zWF$3@xzQK`CJ{2s;4V2L?8|`cS>K8K=H0x-@b6$V)+wDBJ$alAZ9fM0drEP`xnjO_iYe<+$3<_~P}hV@B*sAs4lY@V4<3DiXwlbjP`3b=*!*TTKA%pyvRKEZ zRwFE$cOI5^TSIYuHb;B!KO?&sZ!p@D2NlwpaOvSe{1MAF(!1i$EZ-mj$0b$}`;+@o zW!Egqbk-A{7sts$0W~OCKAXr_&O$-KGVH$K%j}*)aOBR%(tIVdXEB2|jdQ4s=^o}q zbP~C3rHU#~DOuaLp4^FN=XRWV3};y}iE5vYcLyl>Q~R4pvTkP!#9&SPE*Bu+vZd5{3Rr9Svg}> zmrK$Q`H~GjNaa;!xCgh+CEQ*QmVd+U-8K6-&o_KFloKvzs$mnet9&;Iu?&DIl~(vz z^BOmJ#ZooJNs=Y%Nv6*nfmmBb8l7uQS7o?Uv+xsSPjEN*eU_$5{4?M!J6K&eyo-vUP;c0+&UG$yWRG3%7s$w;0L^r9^)4tRSA!?2)9giL=MA$Z#kB z8jfAAC-aog47p_@^ynE2VkY869DeRwnymI}nfdiAS}pH>$0bo>@EUnt1>2^>Ia zSr-H+_^^2kXOQXghdh?k@L=UZ_^X&A=t;hXS_URpi6fga%%&aQ?U#zcsq~HiIy>%Q)1A`_72h+ zKS(xS3ll-Xu+IMk5%=^YJzXrb;L%g!@JSbJ^p3%Yu(O~Ue;Bx3-ta0iocva@fjLK< zp-T8TlUMedWGU1V=_R+w_uG5P24#0LDALZk94bOSERP}Mlbbk($^jhXd+!;${2J!0 zyf@QqeucTTrH`cZZ-qm_ZQ%1{C5|W<;u~F-cY4wU_K%B$aZ@Gf+!R658`y5D>^s6< z1ero{b8>EOAKB6VlsJ3y!qtsSV8gD{P_3K*Tl5BClQ+w_T4w}R0hy4~I0<=L7C3S5 z2A*I&N)=LxclU=k#0h>VI{k_9^4c}mQ`qxthrB^eJsybVulKuZajfO z$AXy=_c|iB?j}e#r*KYH%^*pI$)vQUhD1mu!#n3c@S4bA@@6fz)5hTYgI~$qecI4# zcM`;7pF-&M{b;5cga=FRK)%)oSi<&!7$aqtji`lBn`UCZuSjyBewh5?9VG&pVPyRkG4RFp z9Nj(5oV=k}Mt|);!uy;(PsWAdhxjyD@HY`gUf6@=dsUPVWgU>F%^2)hLT-{YvMNCb z91a#kxXV4zTwl-Tp|d#hBR$M3hasXVFNh`*dvKpx6l$KVU>tYcBo?d!;ML~K0w zyQ5t&7Wf#nqnlXHcn2C*k?ZwLzinYYeTLx74sI7cPp%6e_J<< zXF1^V)|F)aRyMcSwiN1@W|3@@ePr7>Lt59HlHzBwpl+ND`y*$=(uq(=5-0}8fn_Az z=Nh?d#S45=i{ZklXxO9u6dEsYgX{O70dsRc?q;1_>zlLS;}6zdd+eCka!%oHH2W4Q!VVu$b+oYSrBUUojBdy1V$m9aLh3o z3QSl&c(E(4^3Nns=e;8y9WP1c{xAq0WSu^BWx#RvLLc8o42t}UW_pXkmIpy%wjL<% zOoaZg<*;3G7=+}yNu}f%5xP#{!e2)cqBfhXH4-B>n|_crMK$Ezwki_gB0yw>SCLdc zZNl?QnAqBWXX-?R$@4eMAmHCo@SMr+qAyBfmzX%l)y>96{<*kos2_y%65-eR9rSI# z6E*(mPgAx@(4!Gr^z|mBFW(!3u$C+QS{wrW3Tw%Ps}U*T?;smYtHIu&2V5r|ac*Nf z_&JK>YBSd7^Fj|tO*f!sktf!v@F6W2g*8>vaOcTb_RPw~AE$=V?}#^CW?4hRnr;wR zw+7{1yb;;%=lKqnbNfUC>L$#fcJDB7#aG}^X(VbqQosxqz@_(m(6c27j?Qm@;$j}$ zp?DmV-4juBLmO_}{}<-&o`ZHWHaJCsuz)8GFMaPux$YFQ>zxrSnidQqdcN@Tj6WzY zxCMmGM=9L-3AY9$akua|xUl)LC7Ry2FY7(Aw@V||!MQ}xC>C^tdcdz&6tC6DV*<l>BFE%z5;y^H|( z{$MjieZ6Er;G5fEZ|ja$k>pv{WC@3lV!!ruv#$yMaZU@PfNeni|V z-NEFp9~@BYg^O!~m{I$8j8kAYbN^)&$m|$}`J2|@bJsl(v||r=dG3SnuULP8d@7#8 znKb=mKhgd?69(A+!Bvf;&?g}YH0Ck!lw1HePwKe@^pfIPuS;he!4c^lrAhkh0*h@a8vmOe6TE) zw5(e|O!a*kT@euur)t#5?|3caevluX*3QDQZRYr4-d9p4V+fm1c|%V|9Z^8VoTm6tjuOr z1|FX1UdHBL;_|W7LzPYx@WPbkJ`kUO8BX`5kWvc`5_wLQ1PUuc)dEk*w5)=X6m_U@ z35DC1dGJwgJ;8*vWL4Y)Qf#dT9w%2r)soFn9(fC%&8vqNGimgx+=aiSeQ<~=!>ZJI zp#C=m^5g~Z-ij8G!FNzTUj&U?mx0~daCoI(1!)87D5)HVw<;bXecBI+9AVUEv&OL< zkIC#;DdfiL`DAST2E_5-gxfQpLjCKjv@vHh75HjSQ?nG{3oBjDuM~&aIda6tM4qt_ zlx70wJb}A;*CA@!bvU*t1-@;L1ht;Ku)M1UKYx9U@nd)J&)E$$kI|+=duGuCZMOKM zXeI7AG6Rj+zS_I75BT_N5!(A62e&V#ko9a9SWoLgQJRgf4<=)BPC16Gd4*s8dxM34 z`yl>cCftsR104fC*twvY#LRp{3V(ZoR>)S^_TMgO=?EqKxvR*2BMvbLtR(+zj*zL7 z<|JX(7LW!$a6K*u`P@}-mF0ukP3yPvon4!h%F9g zY2Z2uF+6v1HKr|>#qS`7F}~C2bJ;q)bs`UCZ+*b?TAgU4l82P}0S5m*!NkpPP?IME z6P3Nh-aU@QOlR{6%dg_W)?+9*|1(+gE|GkaC?jSyjga*!8@5qLP-PiJ0I2s zw-AlPC8W;x6PRE0gmc?0VJh(zS)u=$=z2vHA;UR1Li*v9Y7Q)nuEm<+L>w(mM~$r$ z*s$dlo|b9BRRf#g%W6MJWL*W3;}Z1HA02w4NT25Uj^j;_$GEUB3zgYj$48M(cp!Br zrm>D`d)7;8qR7s^dLEN^Z=*@Tbr-T`p(a!`u$jl%b4j*Y6g>Ai10i(|@TyP}=1i6o zk)%P=vsDEiedmGXc3wEfzQ^x%cj7ng`8bAx@TR4mI6W*QsRn%9`0Cd*mu+O_ACD6JzJ|CJ$eHe*63Q|LkuqYxSU3XPvXp-uHdKuj}>V@Gz2Hsu8&Q z&mJL9$m`6$0_rGJqk(?kpkLTUO4VH?WvazL7W&(6(G_&I;~Kd(b+Z>!O>xM~4Ep~q zC7r)1G~wcNQf)rS(I6Yv@9S&U_{WM&lr+hW({! zmLP4@NQctdT?q8P&y6;^!A+l$&)(^65>I$CKrXeEG;dYRUf&^ zPU)Lqys-9eef@wbJ1OJF7+Lz#UzW&z5S0miz#rTin)x`3+`2r`V>l8vj|Rd#iueIx zE_{*pWIjm2n4k2^j6ZsD9KTU~4DUkrASd7e^o%ESHCqR>HJOgEz9VFM8cX1NAdm*` zx`8v5!n13m4n39;!8>*sHt|=mJV^K*c?FLjD$-ePS-z`HfnWXJgtuM-1m~HC|B5hSH9Q2P_n|2`XN*#L;V2ruD zzT#u*Hn!&C9hTqhPCNCU;m57dIP0{e!O$ z{)6L%T1W-PGvz~XnO~hHhV7h0s~-8%@TRFkE;0x%RR?h^rT~A&EI{+=%R&x>(bY?N zST!mIwTok*e0nz8CBmR&`xkD8@$7+NKKrEWht7w$*#dP(tll$0=o_?Qa%LN3e864r>HJQ|L$$RR#b=&uH3b8fpVw@2ghndjFZQab=RvC{^C?U3e5Y10X zAiZKqCF%nnt4=sXktrkNq8vP3LVo<|bb_S0LH*Yq_sj<0H(L?=WJq}6Lm zy0^60g3Np_z$1@~8L^j(${Eip>Wx@PNFGuYjzHzEEvbLxxkRZm9Bptw!{kkD)9RVb zQ~NplWcUSXcU9<}o{;OQ4`Rym=dtMSiEw@pN18%kbVGF=-HoeeOJC{1)JTomYew-S z$ByNgS3Vo>v}5a!G4F*5K6vwx|Fu2gSa zJ(|GI%I{%5-APO}shQ1lGr*yv&)CC%QqWs625Er;M|N-sD_JRpmm$-T8oLbn&N~sj zOK@f>`ZI}-gV?6_B+jTa*8cv=0~`~0GjU6L*#rAj7E+?iDlLWFxBq>p1oo$p?gA8^ zxebegQXvbM0q>59urXAFOGFKvO{x&L^(NkoRTIwo2HBS%y<%@w@Rscn&VXDOJFrPM z>#%5^6huD#IgL|7$IsJ?>yF)IKiYgc{X-&kj?1Ru2M%GqcN_xz>|uN{iPR3;(%KUz zkt!jH+++t_ShfI1WA;;*z?eDwwS^iJidp0%4Sd`%6!C&1_BTAqcx3>^I14^)!-dqa z>Hz76O@vl}8B)%8AiO$)x_Zl~fz{K$?x`q09*J+RMW~zUkA4;M@ZbE89Z%>*)5vQ$ zBbSWo@EUH|<4h(Jhp|aoZP>Hdh+YK*Q+&w^8nN&&&74?FXG0EhKW-(m$CFE#XK5>Y zKl&BxC_KfIpS;49f7;~vNtedH{K9k|YO`&!1pd zHiXd@n>cdG_orjC>$oxHzqs3PZ*bm|^BDhR4tM^bGpFGxtQnt~!M{-zjR`K8-Zc)B zua3b}_wQ`PZ84-D#o@DZEiN<-rSpS1S|sy~^PN4Ld%O67Sog$nEG~JD%sNN<9GQq8 z5STsIyP&z5(f2{>6k+@f?Y7z|6F{S)x2HLk%4X)&I|~z|6A-Jujolt_imm$cmW3?8 zPW|+2>7rgasrLBV|EstwKK(t(-uY+=9$cIy_*(-JSd>PG8?z|nZzA1E+dvgJ$I&lg zuJd`#6CBWg3xDZs{E6~oGFgdC<%j{B5IxlXXKz0iGFa%O_Vi$uts%wF)F=7Z*I1XQ zFGg=OhW59|?B3)I#%)ey+kO&ieV(ApS%ogno{i2Yme_nI0*7n6=)ea}{_A6H-b&&W z+^tQB7u>G-6C>$v?;`T&h`f$Iq(=tFX}ruf`uc;gMPU2wG_Als6D78GtTJ2at-#8J zJ%roA6jozx%=kNo$Q7@{kCtjUF&hMKH(_;G#&J+*)5E$^>PL)GgcXFQyuwFOUQ%mJa(D?z$3YIJ}2G%C#0p>g*r zAo;HptBZ6H0tBvE?s z46(|5ZT3)krd?})NAVZSt@hfo*G0p-r*Utum9PVQ;>D}oR@(ok;m%c;=-Bu6NV4>0 z9c;c-qg_hG8&Tt(PxkUzJ48XZ48&=A30$^ex@ZrxWLi6B;kefr+Zi9KL{b9Ip)fm3 zWO~_26g6y?y_Tn%{U(pET>AMd%)>X?CyEKw?A^$l!-3~*#%7Quy1ldW-q>%Ba%4lXzyck zfi?VjE&3W=$rKzbn7qetySDr5#SV!(xzhs*L@Jun;tOty_U{^-?NjR(uxny>M6_kw zkI56UIn(rDkzy~-`qsgIm?mOKQQELp?W#~2DTrH`TmKb|r# zgBb9C>+QzgZW6of;yBxw7ozcVWtddaTNZs`y4ck?+dg5;SSW`?h>lIzD89hUaDx>Z zMV7 zVgLJFGaE8ur0Dat-{Le&W5y}RiC&uaa0?|FyCV@H?z1W5vbOYbzc+ladva^0-DHg+ z_FETi7e8E*A{Im{EPML_Tj!-y?P@wA#PeS`i@g?yi9cZxYh9^eH%MNWlU5wc)ekHd z9Uk+DtsH(5yLQWn`e#XT775omlNr7u|EJeDPmNIa$oPt#(s&zAD(4%wWcn++ZDpzA zoVBCa8~rPG-O>HT?c1@{lXylS3Ss zcCVp0f5ttw!)7ST(_YyH<$e++`KJh8`P(A>iWHG&`bBnaeWb|a=>?|vLXF+aauU_ zuZR}=9*HL+kD3o`Pr?S=xNySmXoHCokK{rq$bPRBN;LgcqU%$+29;vk|DXyeGcv z=EsFQEWv;8lSNfSw~HI97jh3RR@gd>lNH6PI*9ifE*H)Ck;+;2OvBd?zwOq_l(Lp* z&g_rLY&%(J6E0%oANx%jhi!NK;YFJ!zZAb1+hzCg`cGs94&fK=!|A{j+df9cM)9&}LpU882 zoPDQ_1LL{Wkpc@}lrvXNY{}inpgIw*?rmi{tCRVn$+P+M>xS|duP(+WYYs+z6A?e+ zB(&kzivsw`rJ*xv24*EN3|%?~Svk3Q_8|dJ zKP*6HsU~F0y4mEc4=n1dUbljCw9RIx?~G-h$C=Ra zJI6MCy2Zsh7_mx^oy^v(h#A;kVTwNf{NV$m`AKBKFDpGvsuhC6&?kiM1QJa=`xUJ! zMUYe!(M(4xdX#2F8%;k_pTGixFQ_dKzj8QBcv}idWi$_~f$&eEtP_eq_iH zl>fp2(-hZzFERZJgD z|7I+Wn(9JFH(AnfT#9@Dk&tLXXOwG?AMo4j)O)8@7`+8`f6N`K|C zJmxB^QPI8|l;6ZRCD)HhKJJJnkCuAH~{yvqA$eU2>YA?3>5G zzn%i=m!S~U*pTX#;3TIm=58#Fv%kAA6Ey)vP_ih3V{|8q4*iGw+0ry@4$mw68qBwb z>+y7FCOq~FUO$hOSd$k|8@Gzd?%o&l*Z)UnmmZ=G>q4l1{~qpYwKkWQHi_$g@JQ$b zUZ&FCe5w_{q=fvBG;^BJrx#o`zrO9|lpYe7SoMi54$o)u#~RqJGezwBV?S0iWf9xe z(@*?r-CEIQ7b&sP{Z(x3u|M3A_iD`Pstl$lyRbl?9Cq?VCY;>@QMyVAf1a69(myRS z$R9|TWULUfrGd?gu*Pj;RhpY3FsFYOV)%$Fd{s;ZfB4!t{z~+2{_==u{@s5m{O(Ps zSa4rC8!-Qpz|2)s95NWmUm9o0_Z~6k>oy*xUB`~l#{qHl-|l$o@F6nE)+QZFfOT0A@(OP=euvPz z{h-1(>-N&)w=eA%?8vlVc{QH1dh7%{-`gxp<}{l)LxMtv9YDq76zmI9M#JBsSZr>M z5u$#OiH&Br&W>Q7nhcp+Z17cc2-M=0=*HzwIR4@sj(NOg4TV=&#pmnnuChBt=`l*K z@SxtEXK}dVASUw(m}{&Kn;vzNuNNbe*-?hO@f-UsNPyS`YE5bhuoax%xk| zf4A9?3qExnFKZJpMs*3?#)|N`GKI~*vYzSheSwW{RiX8H2u4-q;Pt};2ovYxT&)t; zw3@N=54t#Kr*s;2Y9qB5C6c3uEsONPNEb>TZ$#HpPSD=2VL}e`Io9oorq@e?Xp#3yO1{&Ge(%TeqspX12u5&wKNyFZ-M>nT0`!?FiHCJt{9*DM`KUv8+NYnA**r8WzqwN)7CN# zy7)(y%KW}@2e^p3_1F$Bt;5JSFpzA0);#YrxJS`A?s~}KSl##kvItU`@QXLgL$#(Q7Ly`reiv-xMM-t#k#bg}BTNwnTez6yk@fI=`aSbDg$`*!UekNLmlyKz?I&=r+7Eg0i>WB>5*{Ue zA)`Mje9@qX+-F-YEb+A_4_ARV*zy5;H{_6$lsB1q$br-Q^5^aE&R915Y3LWvEAVdS42_la&rc% zhMF+5VKU6=+Z5b;cn|Kk+YzlblZFfPeo?nJt@SxT$GIkYQ<+ZZuH@5+JyN{Nm_L+l z|CYKmh5q=M!(^&dAmj-Jmz&;E>>AUHK@Bn7_BA1pvYUm=-Q%HlV<*W@J4l`jSL0FV zIGQmxia2dm8a#OtZSGXz9mh{5e%%=wez%FXx)+lA{K0&6dL2Jem?f*nK43R%3$f#d zFrV6$O6R0?c$XhJ+`K#2aeNr)=iv!LMkt3aY7F6RrDq9@(r6kIJ&YeW*_7h?y(FEk zA^a_UU9$f0oBV%e@JrNdvHb5tXfiQ1T-!ovFQ(Gu{g0{pbP@X{@sJrb45O%JjH=^e z(X(bZE^JK13F*shYs?C|TiPXD(}QW4>tR~-+73)Po*n!<0U66zAo@rF?O6GbO6$uh z!G8x;elMoof9se-ge2{i>BiepVvOlirNsg(;HbJe?hn{YXN?chHa!V)Jmf&yMQ@r+QsoVR&FAWyh?9p69LJk)=;WSfPwGXW(g*5re7{1b}8aGZ>W0T_smfJC|;i0jHR{K)E51cEc#Yb;!ZD~`4sRrJ$q1> zOV~cB3QZeNk(=#5s^2q*pAsBT(p%Jd+i_d?R|Usu=5RTOF>;{}U!xjn{Kw4{X?>XZ z++0jLaFgwnt72FDQn|66Xbt^dmxV10%Xy5Zh{?3_`wqU(Dv5V|^n!nN)*mYB`)EMmI6m&xFWp=EfjV`k zv2WW>LcD%GDMbv%sJ;wbC~za=m)UHVqc=8bJVV@&$C##_PGZ0Q{70`;TF>`lqT+NC zzbrv}U?F<$yVLizFX3?}6Z3}+qOY%VQCK$u_aZMQ}-JD_-ZNHl4sgg$fKT z2_|J#8LY4H#@S;H*gbR>FCO=qiY5fpu9dM4J|Y)~Ky7u0Ltfhu)NTjY-d=XzMGMpZ zUSo3xuV)h_N|B|q2IAAYSpUPE?bfnpWBsexU41RusalEM<<%(ub&cP(Ad4S*IhuFr zCw!L;I%r7ceLAi2lLnjQ&|`r`@o0hsO$@F>VbdvO{@u>MFde~9>-|di?%u-fj_*(^ zt3W?PODL5CPNX}uczt|2t%kG4(ZW6S4zoL4$cWp^s@kCwMn z_VgOMDI7@66xU(L4r6LoXTt2|=hW+J6Q$G>Kx+&=AGco2SxENY76)@%K$z+`3<_qofC56xwtUv6hC%? zC$Cr_#b5eSjaT0r@izN9>;}p33A3M2@9;actW8(Q)8sLeeizwB?VQs<5Pe<3R= zYd%j=gMVbF$h$e7pu>rU^lNDx4c@qy-=DUFH+EmhPp(~$KcO7H9aM+$$bJ0tvm5xW z$v(V|t1SOB<|r+1h^2$slC;Js1tA9)BQj_V?^|Tbo9!FIKb1}9V=wsd=I=$kw~if| zEik7?zlTswS|yXra^o%pL~*+tib>7#Af1^PLwAKdcK0M7*15Hj+xm73-B><{bU%%! zoTr|sjuAMxv0s_8LMF8=+Dk_=Q)%>qTio$2=ea&ffrr!~u)Gqckcy-(^(~U40bX6m zTV4iTXD`+{3T)1z{cKWs9y$d6`uGxoq0_C+n*`sX=$B~}B>&L<-6I`tV^^j9)^RfY z_WI9cyS0+OHK(C@pm4TvN^t1im!O2~Ets8Zg`ibgIQ1_HdNNDUTs%YgyTP=#|9rX} z`Hru-SIocK9m1=Q5%F7nWcbiIpDAmKBGd4Bz%4pa&t=~^O&jF4(%zLLY5)FZe9|jF zey*JpUv)K$Lq``(h8tsQ^|1{uB7aH{~^HXv=H# z43?qbEyqwaVGTlKXW^l+o^ePmCzbXpYSQ=TKjnO(xJLyvv2i^0yfLJvelj%6sf@hc z^6Bj#VJ2_m29am2z8=*9^_jNxq@{*ukEvM0#l8NlA?U2-0;ESW=`Ldk@__eu5D6j4c@T-CPI1@ACAm;x}?r)#zcQ3Yi*bBK6`1p$D)ROSET^!S>DMJt=~&Pd|<8`I-2= ztO^=)=HYAY0u(oeqceXa>909PDx;23dHGw6t-6V&K0JNhWKGAmT2ueOJ~Sf6hr}LH zv}w^IveDXs{SxPJI3^yi^x`N!V?C`DvYT3ZL6kmY0&T2}pvEr&I4XGnnNEplZ{3e^ zg43sM@d22fk4C_~Bs9&bh1y1SnyD8{w|4I#aqbR!S9*lD`0l3Vsk^X8@g~gGZ$ovB zB)L|b(3yQB1h#4+jSr2aORD<_m64=7G@b5wCsJjnJsF=GLoR9K$wXrvP1c`FNk^v9 z4m$@F4%h(Ix51FR(TQnccX3iro~~aJuCeLr^z`*tfiF1@`_BoS>`6&jG$8@!hqvOz z?`mWx?IZpDF;v=jiPGn9AtmqK^nC7NI`?@A9gmMD^>Tq7uh>M}!&}KJ;wdThw&0Gf z934G5j9hlFqaE9wNTDK^u4>*wPgobkbDzM>?-dSmiCDVoHzH@2(~ZJfvJ9)Ea^E)) zzK&$lFp$!=2U4k$3rQtK)95au-BpDY>wTEgyW*&N;dVOI9!1W&sqoHNi%Zvb!S`Ab z@k8q9dv-D1*7e3m^)Pt)W#DgM5Qqo;VYvborj!YH3zwwp)3-$Qt z0X0bc+6&{2inKXhhK$R3`Xgk!%j_z#JmEfie!Rntvjy-HJlckNrgTlK3$o9e1wQOI zSpWTu6(7a)AZi3D?;A`3hi6i*@?2`28xL-E7TQbmP%uD^{*IbNgHQWXZNe&Aes~;> zxv`O2RJTxWU=Gc?d4ht|lQFjN0&b*NVfL^xgr)z(lc83$cSv&NmHb*Oozy9Elrb7C-1mq(%KM2abkf1zbT!RFG|zwls+UhTT|!* z7aH_p7HvMBO8;J8K#SigOqA$E#c`nb+H_nV9f@Cx9T*tapSoS8DZ*n8{kP7CWGZ6l z{ex1-R2@cl&nx&v7Q!rXFul5V4v*%ap&u73=-QDE@;oZ6i<<^e@Yg9cdz#=tJhccD zr>_ytnw4pZ?N=PSu0r4Z-I12*jUgtnxG**WiWLuVz3eF-9L#{6(9`_k_yUIt)}f=* z50my*AW3lDJh5_tx$X)mmL5Rj_(w3?S%%kZHd3hVR$AS+g3k5`_uY3_5qn$UzBKMe z?#K$P_Pmb9J+ZW3$XAxlETEXK9aM7u5bY@4NxtKpX~FkRG_=T_TE0iowZ3>VQQJ+u z>Fb4k;ATn{oB=D>%8;MEBN?3;O2wU<`C!#ae1Q2x-sPGK&kmXJ@p{s{W0@;W)DNV$ ztGy{rc$R68_98*YPAlgQBoC1xIk*p}Rh4V`4>jO>KJonHrnw}gOjN%zMDX&+lcBo= zeOdY(7EAV1+X!F!Ym`F%v6K0j*N*(lcNTnamIr12v!jz!c{=(gpX%K+=tf35rGNg6 zjg5^^U3Lj4C7o&I?il*CeHH!H+b{66Hjs@?G4)*k3=NfUaA|yxF%F#w9Ns=)C-zk^PhZ=^jAHvC(?0sNJkPo!L_z*nvx&d2Q4;17T9 zz?9_w1h;rM9L9OleyCD(kS!g~Y^12nQc@gWMVl`Uq?4cfllJQ0_-{xlneNUaU)jBM zcu)%e@IW^I<9a&3?|ue_R7a6sYZ}dSa-hU~FWN6;!F3$pP^NJkjqg?Fb&D;@c913o zpO{DIWi$8zOA`52(M$PA7d>)5sZL5ujH$K#FBIz}DX~?D4z!E-!g2>*^TjOwhS1;F z7VvpjyM1||1W&Si<3`~NqseMmH2+z1CjZoD8E?2{Dt%DQAj_kNNOg%Oxmdc;j+Vjn zE2x$_c0Q$+x>o9Yejj(5WvI+ems&!v(XqiY{Gz&N6squ?EdCq7e>ktlp9gOSh$##P_tq^fFCQKTSElfz&zIfy^YM>5uedl3a3?;&*)^&D*0%-DehQoB2{{ z*K*3=lSJc#1rAem0gW@LB>m^7C}{2oYIv?hoA^QGDrZSk49C&@7fVR)ktMx&W=L~i z@f4Hu5e{+kG)HMLZCT|`sZLewn~9+ zZH=cIVGTXzWIoNg)dQmw-!P>{lUACT@;d3iNHN2JUwun>O+nAeU3wsYIGuR4S-QN@ zcRPO2>Ze$1Ek##HDAB>oGSniLqSn)r6zR~0%pGN@Eoecui4ACL0frH3Rbsll6{irR+jjAW_q#gsI&(kR|$HUf< z-jz@awDhHl%8%GI?F*Wz3tgHu)Za>;xA2wXoxHn+YgFig-@8czDmL>uZsT}WjX+*j zREsUKtC9XG8ez{*u^00MUd91KWE;j{ezyzOesh99Uxc~UD#CjrO(9o+y_bdVwRR2e z&MSdd)Ix-4+TvArDY6|hke29+%MS&9;Mjgxb5n3zMc>AZ6VenE^#NC&JZ5^iZH)fe zVW;7DCY_eV3jY0KJB{AslE7b5pXv=W-+5RUmWGQ$W~A%Y8brTQM}6}f=IJR(cOLA6 zp1@D^zvl*%pVtHwZxd788iyGRO5uE2a0zv0p!t_MzL&4U#4HPek0m$&!{4w*2&dE3~;0JOz(EKl4B4ekvY@ZN2DzuR-mrt1zJI0cyi}TKI4kY3lI==9lrGhjY~o&c*1972=1-K)e~^+b{o6-coBP$pNvz&{kI~{34Zf5XpOQh zoKGKOuQiLA?Pd??Jdj0>$w(9%YQiM)H)~ot0BMr$_^9xYU2Qzg*g!p~p37thZ3G16 z@hIf4brtw_ru6l)vCt2*M7@?P?p1Eb5w#>7KQ|haWA`DcOAcuZ8j-%d7!nSlFsxdF z8{^~PP+b`fnb}BlB z?%MVLCSqSx3Om#Ajy26$OGyrabkb`+Z98F1U(~i^tmqg#l(U#~tC)pM)j?T!0+f^* znEcJUlpVMtplHqUzjd`!X9SF5plwZIaebpj86OHufo zSZFLWz=#o^$aX!6@J>TCl=$P~y-QH<7YaEiYY6l{n5B-uxS2kfwq6^)o6H0bq%Pzy z?PXp2cCiVe)~I%OWrMv$Y|4m1nBgS@?e{)HkMjXcZv19$BT}$-)o~2>zsb@AwzH{! zwlVoB>8xVOV;jgl>FhaE2a?e{K#lK6skVw;cer_H0~Ta}<$}_d)K>NG$(S!y+`~@bf2!6%9#f zblZe)J3KLDl<*lQ4}$(1Tg2^hz|VJu=#;ZWvx6x<9Q(yq%WI)FSQpmM79wfXD1=E` zW6xSs#7Pyfl8p=DJM%M>eRz^xKe+(u6V~I5aQ-JT;SEO7C+_V~OEzPh;EcSwlwCUg zlBLd?gfpF!ho& z{DgC%>03vle&b+P_PLc?zwrWB*?%-?s#;TrkX8TFHR^pS#8u({epl0nJ z9L_igBe|*M`t2^a<&waKyz+vZy*!_}H@s)ZMax)0=m2JB;>B(}b6}er`lBZE2GcJW z1Yv(RQ;~EcNxrb8v}0S?1PNQVc~ydRiW@md2)W_JV|&;?pGhq1 zu8^~REJ@|2YUFQ_McQZW*dD27wNwRGvrv=0vi!i}2g+hqvL(vy=-}{&!B}CYiZfas`1QDpZ5A9S za%cW=+ad(m6)j`kv$Sw`i5~KH4}<+*S$1G)2dDF9Hv89n7J3_MVZl9t^yh2XbhZN{ zPnb}J#sSW(_a}FF&1xoR;)gbtBskfkaCVp?JAPG#UDI2^T)zxt6Wj{8!lReDIc^hh zM@JPqR!tDHH-}k`c{scKFM>^ca+z~Z>d#J28P1O8>taHkJEHf!LF0#Nw$7x93DYa~ zpI|5YwmF~W8GL7Z<|r|>PDN%R-NY&9x4?XO6Fw*mA`PQ4?BHnwwrR;e)-LWx${{+m z`+{(G=D&vJ9iGZMW89d0#t}~8Y7aN0Nt;Oo&0xDWtFc?J7P6&3dBo?02n@xWFkY#K zXVY6)S)vRKHAGMuy$vZEk&x!Kna~91w)-z&C!#$ex7r(;K9lkA&s#2ZT^`p@R)Vem z$5V^@i4rNs-Abj#dp*r|>S09)IXhvgfbD zThO+DKQe$argh+yfQF!)l z3kEK|f#DsUxDb|soX1tTt921(;gS>+FoHgx4x+xwfwXn-cKYX@MGrHNV0b_+8s9y} zJ*`qaQNEAKGX$={=|7koXF(ESwiK&jLSGBx@m}zwm@hqq>FZD6#cl~&At6V;?Gvbf zh9C8BTqtC?+~7GnA6?O9xcvML@)hr)t7b70U!~%;;9;@+x*6}8aL%yi287`pQ7YXT^$T@#s|A0H8g*_9hw^`G1@4X??)Ua1voY@xd|?c|6LKlb zwN9W(cyA|#3v+bKY|@*0h2%pQlAd400TZFK zJ4c$^3;+9cpwvV?0e>(4U^^4ySS< z^CRDyE4V2JP-@mZn&g!RiFt?6THTMjg)^Qz5(4k1^$bjEGojJZfuB$GX{Grwlzpwm zkfBAmSGp131{dN@pEA|XjD+}dKT28j3wlr7@Jh249(}jry|)NG!o1n$(_pI8>_hm2 z<48}GrE|g>!ElrYm0Yl=vG>z4vMdrKGplg%U^UE}qmWj1LiqDZ90c-~+CbJ#N zCZ`d2?-Py;*ob7M9GoSA4SO{Z#d=F|V!H78O}~hZ!WmLRmNZFaUxUroCn$^i1lcEs zXrI^*8x!o1{$@G0H5DW7eldpMufZp~5`i-;MO(xzSl{1^!uF|=s_O!rXwO2VT?2%v zzwmxa)BF#@=RQ6ewM%=@yiGd_+lpi)-nPO7xm5IP*^JSRt59EFgelIA!hI23@g zepC1!?#D@~451@=ADjJR5xb%V{ZiZDK0%Yz)syl4%y$^kA*?%e1Qp&VFk0UeO9yb+ z*I5Q$z5>%1NK^GN4{BE2i=O3uxcc`uWP?7S-dvFiB4<&btrg_okAsnc2~N&A$=GdT zSptLW%Lq$Q-~wD8Zv~zDgOv33C`417sPX*?cHqTYM(Y#UN2@f%WqU*R>S2VO9)(NE zqcJ;cB6d#Ez%RW%)^VYeZBh+`YV!sxS@sJm*HT&7<-xd|)xxr*`dFN7H@i5?4n5&R z5!3sf#pPRLY|U(JvK)xkwQ1-YbBifFQNhFNHug6o6fp~eaNKq{95(c_c?UnT4Ciyq zE&Vpj$u!5x!f&joyPL86BsTBCBPJhG$QGZ_#8H=C7I=3*%S>owr~35p;ASzqXCHzu ze#)@!mdE*GXY3RDAyWMm(EMmHeypfNOl2br6=}lzo-R^$DB{I5F=~Z9<;LeV7~bfMIBRP>sX%*k438|!b`S^{fcGyPsN($S$<~YcMf7b-Z4me_nM8obP`8x zUa~e$2b=z`gv07^nrF3uMyD+$trST-S$2}$!c#W%_E6Y_THvVAlXMqetE!A4O>Fpu z6MDC>|MoMC)TlNEw*?#Vdkz-$O?>;=xusf^T2>LnjdG2 zJ_oV&cJAME1hh-dzep-j4!Sn9o-?B5Mt%t*;)XIo96IQ2i~Dsn|24=h?zf`dZe zF01w`nAVk##$7*-8DA#PJY*NbH zmIzL6WhJD2k`cbczu4UBST=WrBPNu%!zyey7KoE!nxcj`ixp7jz7W%{Br(U@r!0PR z7PA=rhy5O?j<2&qg@4}ta$WJ7V~lk6AIf*SuGN7w>QIl{0G)_VJei`1s;UsC^puR*r+XWY}$i5 zwtc#?;Quj#>7{12#qJ>cwc{^adpwsJHppR_t29&$heE7B09_%yZ2AucNdAz6T2T=T zUUH2|XMbfEBlWRa$xvA9H*s|dx7oAr8E{rC6}rpE@Iz+~1||%JmPZ_B#}u+RJ$Y<& z$xD_uu#ahIJrJSp{1b)ySM&4hw5_9eBOhz$LzWarz$Bh({{z6 z=MJd3EJ3b*hIH#qKkCf7NO3;p^lr$0GVIzw2M-LPVpSu0W5`qGAx3xXggIW4nEtb| z5zactQ0K76m^8NvXQn+xTw@e%l$}5?LfvT4$1HSaorL4|!$@^W#9WQp@OZid&COBt z=Jr-PzaxWmj;zA?#x2MY3+t$hgUEOqk!gQx8fLbL(*JFx*WIzSYF7iK7Vk%b-+5^C zP7vG%p0su5Y@z=bLK?+A)cPeNW!%ZXRB%;o z@+bd}IFgmvM4RPz(Cw9N@DklY$f#y0%xZyAcLQGMD$#V|nW&~!fiAmrgw(6k-(*dq zN{$ZL6ry(BURd}YhSue~XmmM+xPABVV}cnJ#xYpk9S3Eh13TT#h(^lW(!5E>$v1i% z^$kuS`-7|LZm}2Tis#dTe~)QOaSVMZJw>ANnzUnt2007a_ffvBaIbxdsV4v9wF`XO zdGCdJ-aWKeHK9Z5E()9)(Cj;tChQ$dNB^3V#^q!LmmfoH*d;9Y(x4&1>U414AiA=& zfp*nYP=v-Gy6|)ZW!Ps@Mwu{&{k8(!K}Xo1UybPnB{aZw8&#}Ipx^o}a4@=rvapM| zG(%cA^U|g8LPe4l_&LYzau9T>9FlL3l1foJ$<3@IYZi|Pf&1k&XE&~T?Z)KlG@Q~e zL09hq>>ZGV3F)Z_I({D2VejzzZ3|{B?nj#2l*!n_jAFwx5fCoSq=rM7~;r_aIn} zENzTQK86wF%_w4sImIN2DCe^|m2bDBm&so+bDF?QNvgriHA$$=Jb+nqQ=oGAA1V(& zM8@$JBuV);bmu1q; zHBt21aW`$5{*hLl$q?Aed#NGFi#BL;r0`ozPBp!df7c7$w@Q>ZOol>~`;&W+FqfIB zj8)bqXgfCp&o(~8#XoP+cjyP2*3G4bLiT9VomkSAy@Hdj_aLwuQQW>8O_AQh{?ZkG z?sI6hLL~LCiKmw5udwif0xgif1#SJ4SiU10dBW^HQvEsNdfo|VY*JKXpMec;_Mog` z9m?x>z(6t!Cl7fdQNo%oG-=U_xc>D1_aJna%|*k=b#UmaKyln7JTa|-@?Rmds}&B( zJ7*DY7)ei$2|lOr6ng%x8ty>}$Q*GE{j!wl?~wCwIQtqI?|jIsQlHLDvZE-OF#_LO zxX<|N(fw-`a6M5c^jojt;l>Y`qh(AFZVx2YXnE=o*aS}Y{i!KC7KR*GZ#t-44t+p~%FOQtZ98U&pr7I{vB_GU-u74e8{5kc#!Jtlh?<8`teUK?Jb z*>)AA9bQJ}Zdc$S68z!^N^!H0VcK^d11udee}&+!eq}+Dx+2n=m_wud5^0hA25M@O zpt)Ox>;19;HP{6p&T8hD6?SET9 z>&sGTqnyxT2polx}yZJxR4_l7&8opc^YsQ1FIut z`1L#5DcAfX{c4|2p)+F1W!)y4r))q=4+>5h1taP(sik+mXJ~ZmP0~XEMRkg4v)3wG z<(-PtyEb8U?=Bo!dzyN-ZKnx$V`z=TFx+eW!0N<(Y~jyvy0a*i_AOgamM+^V_|a0T zJv)uQxd|s6aXoa^NrHE@-A-p`$J6!e>*@a}IuA#x|2K*giBML9loZM=)P2sQln5;? zT1wg_Ee(+dLS;tswMu3p4VzL%rL9GghNjZi&``hk?>}&_dq4Mi&U4P|#IBcPRn=$c z<0!{td*&g(|1o5@C1P7rH@@R6&qZ~AkkS;1+M40$m0Sh&24{ABV8cfohVwq(Cs<a}ng$Do zZAl1^n>d!eT~>0*=0Mi`?8n_-2lJ_{X}Bk)k6Zmhu(n_p8`aL?L5WlOY0`RFo$J8mp^;_h8g?D04CmtVL2!-s8cw-a>WBacdCVZIT`>NpUV+$!WStMU-alEz{UY-3y zeBimb@$BZDu=Tvgaw?b2D}jF22PE5`!`(I7STdy_od0X4c!NIN8UGu6pbzg9_wv9` z!e1@5ix=Lj!}fa*VbNBM)^l^%G<_gHyJ;ilsXjvc`b6p5k5jK_6FA_^c((Q)#yckE zqQIaKJxTDvS1RLO(08)9d51k!Ee24Wmo?+P)qxhF@2d0bpaLt(K@cFn0`f8~N51GL0ZN&X>Vjtc$-HwM@DzfC@ zKMXsY!3Tz<@QBs>*uu?`W40KwOo%Zb`YeSi-#4_n^Cg*1_(%aWO|k!|smQ~&;k0oo zx=S5JY@R;$xi`@VlhK%#Fp%5A|6qPtHQH~hTg_RhYPI3bK?>AN9Jz{|rmHXMG6&dNodD9ZcEd`3Fo^4@HRTZQ+fI;shU0_MLE$ z4+$M1Cg>teEqhK6S2!i^XIfpT(ioD?dL^tl4~^US+3y70fV@1Mjh-FZ=>$xeGJpO zg6|s}@wMA;RAromY*RJr3#536%n|g-JB!eHF08o2iUX%=v3Y>dWnz_CaeH5GtCR2z zGd5YhfAZ+fONP&Ka7PV_hG`Ezpz|<4NGFr zz)5uW#fiRQSDo%0e0~gHok6_JS)Y^CJYd(d6n&TIqQ?3&zE{a}*N)RT5uwPdcU=&9 z!ZH*$cjv0#9|bq$D-`>_MM%IOagRA6y0@mYN*`IUU0K;>>_n7wJiqR=;J z9(5iDCL6Fhk?_@M6P^hFnZYbGz9px^Gk7q(CrLwV?PoID5x~t;Y+3IMbNTrVylYW7 zlbB&nmbvrao38v?WiZ=|`QnO_ittf&$KA6+d zJDWsGv90YZD5)l*ar6=7E^C6FS1WcT1&OXW9kfmwhEeIG@o8!jP`VD&zwF2N0aENe zy#(!}b1-343PN_IW4pyH%qkN*_RI(jd$u372?x1$_e9pzHRj%?3whE}2bRj*$r`my z+)e1Ls@AcrSu7qa$rk+9yEi{xFT-D>ocPkw!5rG}J)Ud6#+i2m_??fH)wQ9IIotOF zU%OV!$@zus(cZ|d*`lZ?!Gt4E^x-EbHMww>B2V-0&8-*Hp|wZodd>$h?nxHA75K91 zkG-5;`5F32qI+IEpX!zguV?!#ycn<-y_a<1bsK+U)4#FYaBl_wHBRE@yaS@UdJx<9 z?#IgY8mxBEhr+4q1FXts%^n?_ub%)DHvV$pXhvv#i#uZh`8#{%Z%3WE`?}bN~IjSPTuO!wO{Pf z_6^ftl_9`%Bp>_!5mtgdbGy7Bw^vqi&CpmT>mXh@UzbB}_vPL>Gx@~0Wd2*1%O5){ zxWdAQdnj7-t+je=FL*EmTq3b~O#}v;rQ>k>HT=^3h@erQFypkCw>#RP(R^IY-=~pn zate7H1xvZhcj2WKJ4zpUtAkzpSan(rvbwpaAJ2|rwso`Rp?OOT=Xjep4iRB!o|C`(;zkdJDb}6Lv0t$ueG7Ik@pZ&K;%Ahx?4;vH?rkSms#9dE?AE*jrdyAP+nH@28&iU!cR$>Z>_k8HK*Pq z-&2h*xfR3q;9G3o^937wuH>e&b=>q|8xQ@Y$b;F0kB1E4LdzbUHK`BV){J5ScH=V- z#Jv5-flnL`;(^vKdLkLYV+q0@}{!J<$hR)tm;4nRjqu$0tmvuYl@&(HrE@oz2(k^V7eI+~s^9K9SOf{!WjfK{&12}lBF3;&ckpH{w z#Pd74@vHB;JaOkp4)O|O*$Jz7((;vTbWru!LpW1s1Sb5+?iQe zsCN>lOig&kiC~u372f`@*HPY@f-4S%7~<>1GvpPyQEw%i4A0{G)2?ujdnb8j?Qqu9 zRc7rJd-k=r#k;h{$ej~`oX7(hJ~sk0juoP@c>%gad7yT+2UdS+hsC>Q^pzL)C%gOD z-AhX7BZ|Di`ZTm7E@JYOVkoyJVw7M{Wklb>qWvjYIW-yY2JFY>rjU7yV61dc^3|+dGS$_LUMDqR^}r4c?U9IzO=6Aw zafQBFnZs4yi{?(eE*T^LL-HwnJbKPcfx*%a%x&C`YbU0&Or>B|8)TAdP!ts%`AA{D z=gFwJl-~Xo9k*$JC8mX?lKAOXv`t5Z+`AUi`upc`QrjD9@=*v8J+H5Ng-RUc-bmKc zSxKF&8{G>WL4F=0<8Ep}eWtn5&+qeTq^}tiEKKmnC>g=cBYDeLBf+8V%FnH<@p@D| zov!pp$$uwkSwI~{_6nmS*TJ}iei+kT2}>=8v4LT<(67H^onHhFtIKiw@TsDtI0!GJ zX2SiaHcWgaZXq*seh&=n)FKA{KXH%a66?F#C5{FP*>(``xpu&Wa5 z!~k*68Hj#I4&nQPwb&r~_4Y^$o$PFXTx?H;^)*9uJMB#k4x=dUcM&D7nd~{P255 zl$}hS+vMoav&Xc1z+IFqdn$IGuafJk?<7mxlqlugLA)F%%OxArdAwI1y6n>CdWYWp zEV~U`JVNkHJQQ_?s^FuzD=vwCgsW!eIP-5Aet*>u?Mw)!8PO*#YB7?FW_%edXmnWNShU{KOV zYH(ai77O-}$~Qw89;-x`?ZRW`afS9Z9U~V>ImwPzqLM@n7*$Qg$HoXkYX&{~Do?8y z?xyFFk7(-93}i(upw_wi^!>>vGBeJ?q(Buo`>XMQAx?NR-vhmE`}4)ll~6ueN%gnK zK}HpR~VL{+B}X?^7d$%@tiWH?NnzLp*#rJ8zjJidlb&(VjG`6LAGP~)J7 z*C=go7KJO2)5yzZhWli9z2&`V%bOE*>lSe-kGj!bx-II zE?1Gg)-c{Hy0DNJ|kf0S!`|)??q81(sUK9@@i$R404~b!(%Ou z3)1Gc04;7DwiQkFg56>>1*_Zh+2?#1&s?9z_<4sn{J6xg<`=Sk;R5#&0}BdCTxf1P$B@d*KBcDs`IAKT73$C;WL}LKgpL7{)r?lX>NvnJ8L#2>tJ@ z!}}L!(IdYcHy7*`8I4%1?J4x!cQFXbE8roAPl~-bod>Q{vC<0Bv|4GYWu>!jDhJG% z$CI>1@<-#UuF(RV-ps`w8lUGnkxMJQG|kF=U+J;-Xn zH(9HPp676*doIqK9>oy7K32s~46K3_8-$1W7=|v7$L0aYad&VslA97ROXPwl#}Bfy z_8ww&=%I?$kUJ&(rX!0h`W)l8GQr$?!#*CUH5;nUP$LSGK z{88~14usdkqendKY&5Mxiduy)O~I;fRy6cKZh&#jHr&|Jz^Rwd@s~q2yi>b^ZA?z{ zX6-DV6+O~wiLtI#Qt-bu&%XuPmvz|ntVHDMcd^2=0FKd!;*PX^tQ64Jt=2EN#)mH*;_|YOXsnXu`O1IA?Y9%jiTSW> ze}U15n(8L_XR_d;-llkY%QWSi#~34Qp$at@N3#M8=0bCg7M zJ*{)({Jw!)heGCB?W?}Iu0dnddi`|*E5 zLu?c|=1tBdPO0++GxbVE-ZF{&p3mSAHOO?isv$mrnEa3TW5}e_HTu zH23lu&G*A&p?~QcwhnoSr&=HJgj>;j-(>^~pY_(uaR^^|0+rR_Sa)s){wo=axj(Pa z+_2rysEonEeKWBlb+yoKoFFSBxFn}ldCL3yST21E9hTi#b+5>Y#ood{g)oGbg~3bQ zZ=V{-@eXt09eORzFUPIIon;%bvU)uBO}Y!U4R7GmnT(FE@9{GH0$hjIU}#Y~iqwNa z%ObE*_*6`LR^svT0K}~n*~1=bFk7%01($?J*-aWU&(6`T|G-zXM)0lpa`fGO4~`Q< zFy>!8)(F4%Q-<%) zD?se=R~Y*w4HuG+^Vhrc`Ka?KR+Acsk#GAz#CtG&d@kDdcq3(BFfJSTaeei0wiSM; zUmK)2Y0@)1H~N4F@9sc5*cO8ggCW058Dlp*rM?61P>|vvp5|2u_oO>$b5X(d(Vysy zOa;y8GLWPEw0N>%4(fw;;Y;gESoW|**5VJizg>wlL@vxcz77#-@z{MV8krkocvh$h zUsy4S_x^i@=q*Yj-*OkXrj?^qekJTKuSR5UFrK|}$HcB(kTvxvk@+(^yZ;+)z4Hi1 z`&~v})oHZI)?&zeZ@8}TLe238Tue-YsnSj4hi*mZmu!4F>wxzp_LW`6{K!+}+_Z$= zqVN=IGCzH_{P+WmQTUAP zcxit7I}ZiZYmuyS4$Z}fM1JIeV6Hpi>iacFUSx@kZu}FC6rT10m^ErL;ssa!l(82k6ih@HrAhek+)Z#{pOD`7Gl(%(P^ZKsasgrqdeI5CGhXGD?n zvsHBP{xC3a(S=cHx~)P9mzj zNKfB-)9lgSq?BGksXuPwOr8{pSnNPV#Pto7y;neoFls-s8sM9K$ zZo3Q5i)K4ECA31uyC+*89md+-68Ox&LUy|N4J+0da>LLe?4=lrIKu!`hh|`t;I01( zxr*lbO-Q*IkF;aCcp^M^`I<>`zf9D;1(i0c2~ZkW}NVnw1J(?Ht>ROA}jVO zk4}zJhl^7*7Ts8h`~ME({V`8068YeJ!&kBT&2=KBFod3;9Y)7Cq=|mCoqYCfJQuZG z=ge6mC-PJZaW;C$84^p^(v`5GHJNh4`*B3;LM~sM#u>LJbDuR1Z2!=fe_jfOd2Jm| zHRYl(UX9OZiTm7ze~_$kLwUhRq>3C_*;#4W1`Wgg{R?m^SA(Z_4B>4zEZ8bii5fn4 zrC-W3oi@^D4GvMXfI1Yn5NHVm&n0oBps_iyE5(mdl%zqDPb;!_unIIO^SWXZ@r zO9x~|qS7z~!#(cei3q7)QW4y_SsfC$UmjG@J%lX2T%o(O6Dd3W1oc-iC&rd&|4pb zVds-zztsy7|I)Ek=L`yc!tihZ6+BRUf+EYum|WP2s&YB5d)*I5%k(iilki_`D!y*a zK+?0z_+)V$319m$$J%iF{zLfCw+&4-s$3=+MOvvN$Y%6vx|q2Xw-wH!ZO{v7{QZwE zb(2QXvbQuT<1%)CIfAiCg^=6$mL^mW!8sqG(|i;KjEW(fj-yn&S_fNg=3uC|D-Ld) zz>Nzhvd?54zO?THOh#z&gV2e5nELRK`8quQ#Z>P8@FexLJ4HH^&QiXEhzMssqd5!1 zX|DA~TDLNcj!0$F($^7ivpxdH8y7Lba}j07x>L$4Uz+hV7J_1d*KZ1;o8v>>J_b{- zhU4^0T>+<0twVm9gV-@-`1EJ-?1&z~as~eAe6*orMp?xE!JX;C(LUnBFm z>%-PPRj_k(4K{^U!gls>&>%Q$++F*}hmt+i$ zI*6hld3Y%uib*~pP(NcUdOAdp#kd|Y-82SIjKgqUZw#u3`eIexZyK)K9s4GBgX4%9 z_%dSxlub=wu=uF(?3U1ngJq;7Say?R0}&$l@UPluaK`$PT&)<$!+y%(VHYiU*BIl> zV;P|n`-}b{UyKa8P32Cbu_SQ?tdh#`KJ+&1mg?|7&so&IW-nc;-zYl2OxVy_+!fxL zvww37(HJcZ?pIAqp53N3BmPj|q1y0U8bmMRPt*AF>!kZ<5GvwT;N@Zt&C3UAL907C zmj%*M&kI~2I&G39yrlU!pl zRdiY-W3}k?6+Xo484EGi(FEfp!pA&77hh#sX^U$zUC7C&%J;h|X=XGUkBgg?Ml!

l!dB5;wlr1cJ1s53rU{Dy+C_Z^5(2{6MDO zJGyMt8I!KolHQD7yngZv_^O-mK*7cT2~958Yg0Qm0()C2T`Ng z@n7#V7;;MjV709XgNBG~(!HCwB4%Iedu}0c!Xmi4Md7i|Yjg-lrc&<~)J=GXbt}ZK z=DP#zCx!?PiaDfqPwDYJ!9ZkL97qHL<~iVkk1TW>6=C_Zne-=}ruOL%=>GGo)a&d7 zsGTx|tJMU&J(od$j8kcL;x$?#`g?a>f^jXfKrn$*v9tSRIJEpkSDAD=I5d=!7v@o5 ziv#Ko_`_h-G|aFXfStl8=(9W?3CFr(`{#75+S7s$87eRl{+w0g7vsW>8`SgVESNkq z6~00-`^-(DmbePKW7-E1mb0+^-yt!#70>Ts;yc+&AKC*?qFQ{%{#KLY(&5oGSSFcz z7)4O&PvQ3zT;VB4Z^BRHOfTzAj zhrIRLNJ-12O3R}puO2{M;}Y=StuV5+iNj8X5rQfIlM<8UVR&JTaG!6%wo;LUSP?`& zwgl0?t%oVGHJB!kb4P=tC%Qcefy%@*oV^~0wxpRz7Jbzg5?!1LKgl?(l?;|D;~3;n zGga)hCvQO9L=%*!U!ZhneVh=xo5zVN{5kbJ?z9EbXOTncVM18xYYgeZF;FO-32#eX zwC|sb4S{i3E_P&J^gE$u`(Lchmf?Fd6?k{?AlA9?i<*Bg$NBNu@ZG9~sFh}@{In0D zrEh4jI1^8M)*lMa?bwi1L%~jO;Q3q|4i7CbB7O|!mmZ{_Z)?cCMh&%%rPS7Xf@lKa zLatzsMFb#fh#EJV6j4~dA)NF+lB^_;j#!O_=6aD^9#=>b!?)DfR!Q083usBh6_Qya z++Zq^wC`^NWhD#FwNf8k|0OzeGaV6ipp4!)|0Z?YR$A2bmYP;-<7Y%)oNgb3#v{V9 z87gL(SqexG5_u-q9<1-6jGJP1I4DB&(4E@IZTVyB8`lNjo6V{1+GjfW{RD!)H&cv! zJN;H0D46u&nB}`07UMqQtYRc7tlEQ-q1CW{bCRBIE+@&*Pb7C%JR{^q4q4olrRiPq zU}~n|^pAzk#cT@jIZFZiw1msFlC+l<(dSJ=@SuMoT+a$7lRS7}mj>Er@SB#t))0;# zd49d*AI>c8z$eqW&~6zA(=&o4T>YC4$!x#}o0SNuZ^hmXn=mlPj|0tf=-jxyR54H! zcE%6M#>oKQ;}p?-L^obCR^*`mnekeQ*!9_6#TJp_Ggs`OF3l4WnX8F>`De8KTNPw~ z^u;QVPH6Y}NLqqTuCTrz)m+|5k0%*mVvQNF`zt;EdJ4631zTJ}yvIv)u)`<~gF?FkOmgdsPBjp{!SN{hcvAS-Y!Eua0rLa=5&FrWGM%oHcJyfBn!ii2 zYSAz^_rgrYo1~NaiTe2;#*uHW^!;iR-CI8usXJegcc7S;1)8F2^KaU7hmbaYFD%aW zz|b-q$VQCA<3IgydhtRuZr?6Az3-@CKnH1?4Tpc-e{|fsip<_j#g#>j`e9BYhqex5 z7X2k09EqFym+`%Ej>vBgMN-ZQ_$BWlr)W>Ap3w=;8HVVyVK3#1x%@Wce{|)>bo@H$ zj-2yhlr&rin_Z0X`|2m^EH6I)^WtzcU?K`S4#MEbb!^$v8I8)nDSfa5rn{&?Z9ecR zVkQFWtboYnf;&D|ybpUJMV#Gi{kp;b-EK;aSVg}texrYbwNSlc2NsEb`5Sp_Bw6l9 z#qiHaER7I(2_K}2|BaG-HT_KAOThzQQ>{z_ZT31&7y8|$FW5?zMdjq;{haz4*OTEm z;Xy5%k2Pu9g2UwrD~)h0etL&=V(!xQ1{0BIy-Gvmv@z?q44$1hOOZ+g(R1l2>=&Q) za6MIIIV{9ERU`cPy@I0VioB4ctT@ZH(=U-(X+G5*u@DSW*F20K@C+KlOTQqi6pvr{ z!7JYnKeDgD%zPr6yZok3hXe6N#Tu>+({OF~MBGz|ql{w;$Tj;QJSa)x^{m1R9e2{Z zA?B!6CE~kvHrgjxVAp#$BlqdG1^bNW6?V z+qEyeZ_UK`f=3kR5Jl<>-Dr{506M*(iA;Ogp`v#_of_AKZ)+dI*EhMAIWx{;AzsQ=NN!QTT(JQD-Eu2nh+EL~OEy$;oBVb`7F6|l&`L9^eeOzM=Q~oO zLRb75Vu(x4f!JP=$SdT+xNqhLPMR_UHc5SOZ*(M5oulz9|Cz{6-Nc-?n>f9nGdJ{m zfgKZjBKXK8_@)kpbeSpd`eV$4`l<33rL)wYGmUO2_a@KzE2&y-5^WxDOtUu_^3-{r z++){vc1XGlyDD)H^h-l!ohlNauE2{?ne=Y z1D^X$qK>mnt?T;Ju+%T~Lqi=KbmHi6x#0TE)k0$fLBpp@{~lT%zTFr{I;@XcSAjqpCC%zAeX~efkUzJ@tmpn5{S}Jd@MgC!@QD z5sJEsZoV`LD~rbCRQ*((tn{Ied%Z|)tT*kt*at(*?L_ulWJd;Qz{*k{3ztqsQ(Yn+ zKM`l`P6x1PYA6lt(hF`24$^zyV{n<04}bk))YqG%{?$4BZft_?qQ|&7_ai)nE39Xn zE2W*DN=dS=6xjHTR(U3ieCr`v6KzZ*7mXmlP=A^-RRRs=V63$0k8UsBQT$E@r}tIU zVr8J@q#D9>YN&GME;_t?GF`rCL!)F<$!GdEy5jMR*7R`4lrb_G(yfh-3ZKx&;wQBA z!~-gi{6`T_V(Cdt6}jKAG~UVx}3pZzU}6b+FN+^l4f43Rm@X7&+wr`U$M6M8x(JUgIAt# z)~Vmb%S&I-|ELOIT7CyJ6pvxo#xLyu?>aA;cbq#S{&D1$CXVhB$i=~7+;89l9%N?A z8Rru?Meq*>>aF5#S)X_WJrEtxA`Zw=l=|N|%muwlIQ3*9cU`xfb<*ibTUsTDX z9ZI-y;b&~>dleq{lHlPmgBP##-agW(!`QCSX>GCtZrR|-+v6D#;#^tW8yil1_OZ3^UMr5Xg^Whx*gH=|1UjP^)1=nj`bg&+IZ6`-$Vxlrp5jZH@%-zP3P19f zksMIGkLxp(+1+-y zG<{bScU{!P1`d~aWnQs3$wqMYel1oTFpM9IedH7OUL4TcjYqft#zosmF8Q~Vdp!P6 zFx1@m-+yi_tKz_&4~*rLR?H3B{rHVnEML4B&)u#%@g$zd2TF&qyHt&T2ZwXsgaA$( z;m_yhrEq{t93Qybk58z6;?qxmaqRa(mOtsnw=cQ!weMOS;TFV6A~c}x@5E#K&t;M| z=aSQwd?@_|G;WmOo8XIdL`~umg1=vG+?jJ02JjcbwA8KrfU=Y%wpkg@C%?F{>VjOp ztyjv6y9+OKWKTXLJVS=B8pOVH7$;>7=HEj*pzA0jomF1TH=T|#%hvN-ta8^2UJw{cK0u=q`c>Ih`X!KC*j| zC6~TD&2zi^b9!I|chfcz&UF>heeNncc}Z+~a-Cpsda~!Rmz=JV%Ly9=qkQWRmaBcj z&JKB;F#8iK#cs7_-&qVZoyH})y~JKgiRV48g5=RDcy7(cE4>+fM`0$Hz8J)nrW3V(`4_n&dm&#Sy)+d{ULf5r(SPrI|jgJr$n^S>d-c&1k} zOCu8L-_S#}`<;y_>Bw>4#67Rtu$iH zPI~otDS7U?EV0e2lN`I1Otb$bO9n62pqW?S(bdVfB$~g}$V5qv6r%@AwnUGS7??QF za}NVDP2EkgZr11*xs^)gZ6pT4Z>j62+Y$wr*_0=kL(i3l;LrMF60Lj3B$b1MC1+FL zN#5BTAh2vbjeD&?uI1OM)0A4tbC03)HC092v*l>d;2|_&dM-^*3zcMapF&|a-4QJ5 zO6#{@mGs&=p4_IHNtT*Bkbz4kd964l*?VLKtvlyWn|s7aHXGcPM0LMTUvnl=mUDN> z@Jml5zf%Jx9tv`FV%Y_9yFFDhFWiyD>rS>Gm=+&tl-xS=gbas1kX-z(Pjjqt$N2klJhbMI}EhxyvHpkIki=cVD2b!wTkU-p2c%V7sfe0!OU487^&Wf@v) z<499~9+12k=|bXlJE8DUXbm1lAdnZ}KjBo7j!& zyE#*o#&25v?VaRTlpo0j=F(lQ>yi;?f+VRfPwDW}(UNCC(7hs_a%vJsx!-waEV@4=*cGU#FKgRxPQLy8rUUtZA>L=i*-ysmmdHB~yciOL6`dn&PTW<0bZz8z&_+Wa~o7+HEt)b)7a;lQxo8bCkqx%`Hi(oka3C z$BYtndf;PFu4KWgEQzw&SQ=n2D>+~HPZIkx1OpzNr+Jrm(9-wm_&KzW#uR*zxYeyD zqqf5mWyL*YYxbS=MlYd)h`FTx_YO7XpQ0|kWl*NLP~`g;QbLC&-7eNcV9GSg-Sa>a zy6CCIq*+7Kr^_6P_xKjc(vzx^gXgtrg509tPCdKE;95^%|-Bvn#ga?~b#`WWgGAUr6FL6U?=aX%^ynG{D?&L>@UeD8FBmH*-dhM(J#q_z8aE1TPbbK>V&6@ zZAiJNF3sNLOec$HOM*gZ-@3M>q3A9-{~}Lf z)~<>DLw`!Vrkm0^)pClSbU>n--XPIEvzJoJ|4JUcO_t;ixGS-5@Rb;m0Fe*bVA~u+_$>620a9kDg-l%qqz3PC;2^#3#*$smm?+Nn?{g1gbu!~; zNhChh1>us%D8a^G1z(5d(Eh8*FZHzeNoWslwi?Gt;yhQs|B&!5d&9f<1LRMp3MN4X z9{np5?!Vh;eOU=z(T)9?a38z&o=3i~u5ecj;EEYh!Zo@b8q;R7@$EQto3RhC4xh)f zp}jC^st(lE_0a3PDeOM)Mbq8G@Hpjyv(4KvXnHsn*0~Bk^(LHI@|VuOn~h#EIhazg z8UE^nIXUwh5?DQDWbxgRTV=y5OcS|y(>^vI-D5=lPgzJ_1Lp^ok5OkZQry0;vUCb*!hrVTv(w&L}yA_V#++*@a1?jH6W$e&t63uWYf~^q^(mZ{*WuT*G^8lzh%Vz@j9tDNCB4PG zU1XG;56JV3XL@X}aTGZN)`~fHAkboqyGYT#a$I7YmpkH(fmQnBU&Py93ua{EU zap1xIIk5XU0YC5QFcdBtj(wVn@mFM(xiGrti1XT$sg0ubz zW*7V6_`$7spWrBV74FcxVT{k);^^{_NbGzR3xoYXaCyETHfQDG-8|8!lzW84U>k(E zDMKc8A`Yxx4fC7^B>QUd=#Cdy5-u35x4Q8C3ICAw@-HrTe~X@C{+kio8&f`X!k&<+ zuxZ?l+vgU+-0?3hzTF!?qqU$cmx^QCilOxJm3SZQgTG@m?iBgszWp@hcvR8M*q3y` zdpeTtE=G5+gLsm&0-H3vq3+|2&)OHUO#PnVi0QD^UNIZ5Hb8=|DXugMcbx7hT*#S& z9d5NaJhulc&0dDc4^=48-U`(?Tkf~Di{N>0aE82{oFe(cwP-bV%Wva2Sk z&2aR7vsbv*o?vrn2v+G$VW$s`+?cnEm(D&&$5&=hR~eD5xLghWv#)UFo-E7c+`<@_ zt0>}9jJnYqPc{uhztes2v1|t$U0K4rf3M~3e>8A@YcvH}-=&a`-LX^I6nAcD!*}0E z(W$hC&(H3-`|Km6&qR-2bd0zEE}@M}uF$Uu!YRD99`pOuEGph}oQysULBG-+!rP>UtLJ*7Tz@v68e3qLCEzdibUn&e z;fwgYL{*;uXqa%$-wurQi9}V7G7mU^3U#*W_>>!jQMdo%+Kvwh-yDxFFMIR290Qge zI*T2ht?+l15_07$t2AbpmcG2Hpo7eg8Q_790R{^Xwp&LJ z{K@%D^W5{m33uV6wE`*ry&+o4@N5-~|L^Z8Wt2CBv63#`JdEh)JH?sdE=({EQ(h<` zcyB1?%-e(dzkA{AZ2_~I4(gUDI0ajBshjA86dn>e?asH*bmSQOQ#cofuIF|~OXwMS zp=jhpoTxJuTsjTB2-3qj`MHo@S_JDc6JUD6ns4qK%**b!VwclPimi+x1N+nT#_}Qm zdEmot##dN3P6s_z8fdq_a9n2?BkZLzx}Vd4*AC%Q2)jUImuJ$;Mpr29m0(c6X*hYy zpZx4=D9SvOHl)9y&p|inm#HP*8U3Xhd8erBnp`sd@lP-YG6j~vQh7--#8_|l1gKm+X#%bD-eVJxuJ{B{! zIOMzwf%ZkgGTEJoa z{6MnPzDOSNmq4$t;9f;N#{7GO{kt9`UgUAdk`I{-3!!-%tL%()+g`}yS@yfhOT^B-TRz4RLO*3=X(nojr-DvwTi z8MH}TWbOxtKx$Tr#g7Zn{Z4T?4S`*b|!-1LWN93Tl#!9u?G(y+l`x&RC(UfVf>@>IjA}R#S-lSTyB-l z>VM91*Rdb?(8gr6AHIlG!3B%bkHyy0?*)tXGFIOo%_W`t^E{PszH@LeH@ffRYsRJA z)ISX-RUh%MlMJs`b>dyGZFxsO!MqROfCi`YNPBouaL8piaE~5;>lh#y^a||KU2xUJ z&vCae#G>}&IMh|04=+54tzN}gH|-*RK8c5#fL@klzrpvd5uA9~h4bF7WX(lW*ou?b zf656S9%jOd;k|`#V+ebFAH+MvIi*JAM15|}vS(od-{_FxVVnGTLtX%%nV*Nc z>Q?dD-^5whwsHB*b6je%pAUycqff*VT$}v}CY4DrFIVD$pG~=+_M|zXP@FuSqJ*vK8q<090x##1a zx&|-1WX#`chVX$I0o-Te0Y0I#l$B(HIb}!{Hhb(xp9QU=pOwj<*H*BzWjQ}t$*eWm znzy%};2~2FVcSau4%I)8%W0b2twoPB+M@XA;wZ>|$`EJY&sgi3fzI=4u%^+SLzb2! z>1KC+f7_5V+ghx3w6T^4qaYx(yn}Z!sXbGxz$U&j#W} zYJ9YYzu1E1cFyGDDf@XtWgzZ&?}c6YBgox5juw%}ca(GDkq2yf*nZ&#njXm=Cw#c` zoRO?!vW;iCJcRXgF`p_O#c$spgum|9iyCxr6 zHj2BdHskp|Q!c;Umz63SMJHR1`;_bP@jutmPvtavFZzhPN-xm3e;A+qX3QR4?Ks8X zi*H89^PIlIW&O1SwbMqidPxw%7QRG{v^zh^O2m=XTHF}Xk6o1BA+x}etq0BL?3n?4 z=+}R2?-(?@;U3FiW`nYVEe=xaV_JqXruue zF5N|~$99+@fW~<(pbM1_G~V3@XRQSP_>UzF28=|&ZNbQSaGTtX?dVZjA?;X7WV%ug zjhzI8X^Zeop1OyS8zyYx9fK&Zhe&T0+;XoPNQVx>36n8cF>EXrRb-+ub{QP3(lGgF zZv+nGaxm2fQ=NFkqR;l3y0==VKdb`qczVe4&Ua&Y`78)+6hq=tX7c(U|v+C~6rC*I6F_ zcW?hYo`UxVCVXm79Fj{nlC9G>`s@3Lu6Mde(S2W1iMaua&5KAQa6V4y^~bk}YAVe9 zMh#-`^q{mmp7xA`#i5Z20RswQ*-9cfU0mcs*1D|XU^mW-SGN^(P)pQKhXAe&P zO2&H8Pie5(f!XH{qjFve^iD^SQ+5Kmd1X@c$3b{4<`{dEMHfJ1y~jS=DcFz^v`}L> zP0LN9v+3a^d!~W<3_M2(jsvh!%pfCGR48TaFX}ElSdFEFP(0%hrd|+CsDL4OvTGj# zw?CnsnYXFQJBC(#wuk5hpw}Twgg&gsE2AHX?|T^Mdn)jT0u2u9)1Nn*hKRY0B1$kB zIl1Muv!;Pk?!?h;wQ5RKi6x6}*4V%GF+IxLiY1f((ffk?G_=eLzIK;seQ7Wn?Qh|s z)?u7%t*3wiY2+|0pGuYOVN$aIl9XE1MOz`}z8|(XuET4`0$iHZAEysAp_A_+@*Lqs z(%Z4r-_sm&E-P`U(F_>@W)x(l7(doZ~@t3O|zqQpg+w>tR-+V}>)BaKY zhl%*(ZHt<%5hM{z!3D}&>C1oj=y|+4ln;KU<<$?s1MXn+t|lx!tA@aJ2Pi>vFl_pr zrQ!wEG_5R}?&ewId8s-i)E#~1DBy#}54xHpi`n%txMr1rFRs?SG+xZ8YD1CVkb$bd z*|5!8F8cN^NKMm4&F0&5GVT;P-dV}kdp2{<=~;rms?7RDo3Pq&7s|(eCZohSI-^-g zj;{-8mF0QbzO#}_t{$U{SGUp6*+=O~zdZibD~yAU!ud<*wJ;IP#kOy~5OYwRZNHr+ zjpy&^9(5D0tq7z@7odyNGFBHmrDyw{xZ6BMs82je=1nQoXL~=C8^~foMFo{=#nYXI zyD8_CA32U3k0(p4pfmm!<#+mum=%H8{p2vz@1LihwE^UldYBy3M6T(_aq%A8M1w*< z(0$7k(v1(G(?#=mk);oJ6TPPR_6k1UvXoP0S8&R0OWrj7D~w09Ax5_yBWK-&yO<4p z7$J6Vtr8ZvZGOX+JRzbpSA`A1i(TCWtN19|KO8{Si4nZ`COC7WA=|wwB{S(FveXNr zd3xp;aa$RNe=(s~N;%FzVg#*CoV+KDcUBiSkNRo1v}5su+<+?uRIE>(LHcvOfv=AXHZMpL{d@9=6z@L zxU2k0&d95zr4IvW%O(f9z3l|`TNz0sZ-mlgR~!B;I7@pyaz($X-@El*?ew zta#G*pGV49tjIN|5i&Q-@Jme-=?8Ur`@Ah;uCNn3-@idw_7ntVEk*9KUA*|s9Il)P zUZLZHykp-f^UguCTHuH-_QE;R`vGNq&&HCj_pxq3Ir29)(GaUkbiaQ#S<5Yl>imtc z40Xj{i!-S7drDbh1M$AYg8jODz{ZlV2=LLwz8BS`P!UB5Uz724&sXXc`iO3t+(Fa` zG4t}Y#`)gL!XNXNu8JI1&Qy_in3PLDeEccZ!WaR2O30>H6?yIvos{eX8rby)J?_01 zTJJ|(rKA=vd=Qg za`2`@8ZP`L<~wrypJUsaP|R!nNB2g(qzn^3wz%TV$;(4{z~6UNJE(vn>Mzs#_jgEM zc`uodA5Lb~#iXVdOjYUQsI^K_@cUl~_Sp@z9lk*`KFL8h_cXa@rr`PV({Namh6Q_1 z(1lJ>^q?b|>V1}?=ZpmyuRa#Dg)1z2dMuLc7obWr8MUMCthNGY08uZDSgkR|cp!#Mp*E#p)M9uzeER)Ex>wI{bmtd^lcNhCb3!Yf% z!3H;G;D*&cY;O}Y>ZuA+kJlM|W6c+?9-u7!KDURo*sF!PBF(_~O$62+r4oeXId?|L4bc zA%Q&jPZ&E|o?_2A*}OIB3}0>=&O0JZxUG5+A0GAzWudR}slSr&#kwKm+Bn>=kw8(^ zj8|(+WBYJ1`!+wti*?I6``u->vDQX+kyrixM;kjHjuGB(Tf96Vd?8~NAZ6VGBzKz! zd$(y^bY>RM+B=aWob{!erK-}&#byX^R zo<75Mk5A+CnjGZ*jl$NZJqX+H1^4hUSnkr3%GL@G;As=-o*!=f{#hgsPVnW}r@MG! z)zHXgcHaJ9Ri)E|@B(I%d#;s$V`9O{^N0T`xZ>!21lq?VXzpbw3l`T@k=;|9vkCXg{ju=&4%E+Bfo(%aijML^q|GPP z^>0A?<;ytYsK}uPQ`qz3Fjk*$#Tyh`(YsY-VTW{JTe%4jd-ns$S?26vRe^o_pT*4o z4U{KdKvZleaTjhzY4;djpm?1}&MRkYzarkzc82%#Eav7p>u__BBT_Bq!1U-zJ{=vy zGr|&ulk7W&R%GMW-b>gJ{vIn%$nlA?J4m^ZfY`8O_^1A5pnh4E(NU!gpLWzOA}~w;h7x_f3&! zToSC{ccxgBXAWAt6We9I(YeP;_`9CwD)AoD&$-0&`ri_)r?a>kbQ^i{jKoLlFd=3F zHg5jSV_sfiWB+RYH@k(yXVh|iuX^^?{=k;0%F^7y_qf*eFYl4O=CTq6>GB{iP8fKU z;~Mt!zG@R z@(oQj>8ru&Qt5xP(obVQaro&re*OJ7JDkX4{g3IKb}NTlEK7NJ%LO+6Tf|x)RQNwP zLsosR%DqMxalhO=Hnq6PYkDe5bH1BMlk=^l%2hr2U$iCr^bpTv6+1rj*MMim4`;(( zTD(Kep5H`_V(n93cw2KL+x_Sw?bOs&xqw11*hr@z?j)7p`CcL%2qFyfcD`?96FHETJ7Cc!_X-_8GK;r&QQW>nwt#(tV4inHd4B z{x*QK_1EzWOOob>Nu_%-AieRyoBOGH^Wt0neCUQb-|n`MCvLXlW#>jn^G7d{J|E{P zwY^-+W0%+P@9=xzvo?`K14tqxAZRBGO$J$`*C|-{*+|ea-x<@>Wt$5~4 zAp>P}{pLj~tA0cx!y9?rcOFl>bR}>7&Noz*^n>15{H2mB)7hL+iylp4S=wE*D9CRw zy@)j;+joQb<>X&{JVT0JeO*uT$Ad^_NeIoza%NtMQ2O3?m=<5Orjuu!>16FHDwda{ z^6T^H#!*u$5@xS0fk%=%my=!KEMC14l?8DO z@|-{PZ8pbKTR)XoT`k5d{W+d@pkXF2hm?8#l@odWCF(rqy*n_#B@6|F{zJ8yCfGJh z0wr!_f$s8bJbbntjq;-Ka`jEL=S10F-46YIcA(Kd9v=P1;m-JElt|i!4eMf2wa*W) zrud_5M>=XA>OjFexA08JWgNdo1E0??hoOE=Jp8v9ui6{n4&EkIwerTzr4BfF+aJC7 z*$2^0QXfX}u3^p*P@x%S|v3y@jc(z46D@8<_6bh!?ljePSJUUy#I;Rl2x6Y6Pb5%E#~ZpV)cyH7<0m`hUik^SdHk8Ig?Q^J6i= zG6e-CmttwE7fNjaG~BB~_0mEV(&mPDrt#zUcg-NSl+DjN!_mud2Y&6b!Gnt&aJ|J4 zRO~K=Ia6J6+{dqZ%{d6Yi)5c3!3F{alfeDGWpYh`&jFv9$yv7b;gavOe0 zx4{dn2a`Ewk3S63P`qa*b`H3pIlGsAf1!?61b9yKG-|6K$Kvxb_{V}V?QZVF&4tWk zUwa(6g${H;#;*k^t)hAD{+EcsAqjs`YFbwm5{Og#3D_1*Tnrb1Z0R|e-Z=A*uI z6zWLCV!1iX53^mD(9=+?D73@R0$y0ux*eNhQZPBW2YHiQaZ18J6q;B8PoG=>7vcY4 z?@%EubkBks*C<#-)A4R5n;Ct|0mI7!5F&FEL@q|c-?V(_{dxgpQgdP5rngY>L=f{P zZh$XUvryo&Jm#i~;QG$}*cxtvhl^R)jF*C^ro>}-)gCMoKa9zJ;n-xvGWMZ0SQ^rX z79Zm5+Gg7Buj1p@Fm0e0W?e}LeY1@Fc z|7M`PMs8*SPcGIr7&LH9d`UDL>C1+@Z#2o zxTkLeK4X2BrHAD4@iBc&FBC-HyT7n=4q%9vJSy5u#h33`A2nqI%AeK3iiUOgXT$+V zc3g*Y1#{dLSq|ARB(eMIYy7#O5MRETK+Z|NbW_-kLcjZxf6h4=U6*NrnjT6*E3wJ%12v+GO&H581I$b#2P+#I;`wRrGF#o zS3)P6FOZ=GUs*b&CrGg|qBOo;jJ7s<(&5xVn!oN8EnVJ(tu{?K;olp~?PhEo&Hdz& zw1+IjmeON2M+(i|N@4>q&@xb%7R&R|$tgJFI|64?_#DTVZ zDWJ8N8II(#KKZeJj62bRt28E)JI3Pj*`=7Ka{(`oY@l5GXgZpfP7%HS6j$uSoa`}l z{|S$z{l)3j;`cLT{v(I`;Z%5pNBVhYjFoCZCqhig{1!)wYqUrybQF_f82d@{ zJu2Q8C%;Y+a+hYF>g1jDi9eQhc?6Mlc?!IMVp1)xbNc{+O$WB zw6^Xhb!SI9nz)b79Xv!C1tusdya}rn%<$O?bGkaRkK)bU={ekE{g~%?ucHo4heYW7 zBbMI^GoiT~1!%j+RQi#pN!L|#xfBp;0C$-_68<$bHQJnTFe#fUX zpW}kV4=}yz5qqX!JoQP-vAZM?4GPss=J76CeS87=Yqy};OF^o>Jf0fN=27|2WmLD* zk({=lz|srX(9@a2W{a%xTUNdL&Q=*`ZMHG8blV#{vlg4*RmUj~+{!C@syF3N`zud%G`&eGAt`!#? zxQmaszr&Ya;aKC6fM(nMFy{;N>4`*R!*ve~x*d%ZD{f%e^#-i2UV+1`^ZjU?9}1W2 z(&J)ps(tQBdU8QDvm=m(m-v#P4oBx#&LfM9D=AP_lKi9QQs00)U0fnXZ`p3~eT5>M z)y%?Et&aF{P6SS7%mvlexfsgzTNB;pue~9c55SgMzu4BL=WR?O&|q!pXw91fuoY8 z*ll;|(rS9y##}ed)%{i^5`#xC;L$70`?Xz|7JgQzP8lt-m_;;{!I5$M zd|I?ao87ZbY3ohKK8{;LHd8Fgkc+0jhmTT;#%b!)7oz@Te$tm<&nW5qBrEOvd^O?_k>9H`Z^36cL zS%J(^olohW1!R|7K|78=#jJImSRK`dzFP`$PgOWN%s7J*+cweC@cp!}&6HHnou~TB zb<~u=dKw8vyDC3R4^@h7Dl?o zF?2@Cj&eS4BkzSKbjpn})1|yAE7YCpx@i$>!a;g&@y>*^cr^QpxKpWw2|a(xzr!4UEl==Rc~fay}d!?5$s34;)%rnz7EyQUf@3)MOv!+3{4^$ zvF`2|a!TLu@0wv8&{Cy|$mUuHWa)(Y6xuR>CLM5M`Kllv>g-=jdfQc~^YaRtCoz>? zu^!d9ZQE&r>t0F>R3mHlymYNKm`+~ZOU`SzQLOJJ+O)ce`j+LA#=aMLXSoPD8+Bvc zlmWaGF^ybPzu-CM5;&#u6$=a|lI+GT^!dd2=qjn0bEO157mTNM>2g#x?k(D;ttC8NzoVEL|S+! zp4QI|rLJ+$sc~E|-QN<)yZ{U6%L!5X!U)B_tvKi0YT}5q;3M_kQMdF9_sXT2V&18R1D|^|QT>mS=oCVaa;X?t>meaPw zwsgsU3$>bMP@QxHZPoCm2JdLnQ9px+gv0SceKZbPSKwjUm)Lc&gyl_RabB$VcQLn-qJk7dCak2&I>UbT`)jQxo$!`x`&d0KpMQdO*G7`cu`K$sXFU1eSD{C~GfDLz>w*a?lVybsect$k zWphM%za9*bMnDj~3%*Q0J6@4Z$rc)$>rao^JSFs`H5G;G(fl7lbhzjZ{&>TDn%}$e zok=d4e`Mb`k=GQ>oLdWmCa~vm8M0;>g6+PqQD~+l8QccElky=dUR?EN0PJRXtGi`5~XjerAwz$5M>(?7Pg{Hn>6J<>%>%GuE~RTbhSx?rX5{L ziCe-EOWvaGm?Z67G=X0E&!@R(x6x0zes)%qe;)Ip1MTSz+MGXIKPe@*={#U@BsZV zJVfp+NsI+F!JjLie1auCO$kX@88`TG`*pDVa2pM)9499GzrbQTI$$ z3Rt*`;*zW>f7w4=D#KXO_Up){`y9rZj;D!A!)$N+4p({zlFMoKGl|T1!#pXi2Nu%9 zNwdfxFOGg}Jx95vL|-p)&?Jsyqs~$THt?0_mG_3Zo!f9jI(aK4K z9`ifU`h0&Hrzb~u`xI%J+DdYsWJ`aa8Pg<71FGPAkLE|NV!hBfD(e`)Rn;@;(<&os zaO5cFQY<~z4yR{FzoR5Wf?rZqr-YjwI60c1bw2gz_C0eNJQ_{|i_X&7K|fk*=}+@K zQph6hCYdemq~p!$))F;>J+9ai{O+Fc8R2!Z} zic{N3mM;S<^-EFj#BF?0q)jK3%;`+CF+I3#PeIqaa7!{{2%9J}9`q>lH%_IDh+9dZr#c42H&c~LLe)v&1X9vsHa~11Wd9G_(X|JVRY?}x)l8wK z-vkNmUSj!ARf-l8pet?@Xm}`*_6$eSU7r`^UZg}PBA3(Ew28Fg?08zy+l!@1B6PIg zl$7uBQB-t4?(jQ7HeY<%=N&}f9HVi>@*L{Suw}EJM&c&5(8BT`bj0~5j*p!}lD7wu z7g3Dwy>?^S0YuN?e-BR#aF8IpSuoyEfFN%fk)VW`xllzlA)vtQz(@uNRRFaQp8pbO8HNK>|)t7 zbcYJvlNm%AV_`bEo^_m0DUa?JhL{?85XD@-ECSS8|>%%HKr zHT1@R1^sYaNWx(e_#(;-AE_R~1Mmiy%9dkD(0J-;`GG#>kMP~Je@y`I_h+De@=YTx49hAJH5HQCfBuMM57y@v(s1gX`Zk5>BeQx;pudyeq4qtAUruCnhE3Mm( z(pctCq3Q|uby5^0uWN@>^}gKYadn)0#!kqe9nNj<8R1fQ24ce%OKj^*K`{FTQNvpy zEa4&d@)7=sn_Am&dZc&6Uw24GeO~`04zNFkaMYQ~hL6{Pbo~Qy(>14RwQ4eC1 zk3whV6_~G4iYFb65hhGWn)wh}V-21I!XrN!4Q9L^E?DyBOd6> z?|^Aj%wg9eYj~lWg0Bu%L!3=AWXhfAdS=*ySdAjM$0Wf5ZyOlgx&|udT*B3^eDH78 zJ8s`f5ePo@m;3kUGgq0(IHkSEvATIBS}gt!{P#Ya4?iyW!J&f7x{&$#Dr&w;`dVlU&;R|k?sV+PflLh`<6VO^20wdEqIP-o! zm>PKknhjVd{`wFnDX<5%6~M~S02urHo|?KDd3f;0@YDIFyYG)(9*R8qgka;^qn6bC_UqHzPtf`ttseF zKS0GX0N+Sj;=)f`aNf6noRL@r$A%u3Hh#&G(icLe3XXjzqK$Z)6UV3jqp2s z3v9oy0L5B5(7j0roP*99QO@$fS^03@-woC!^yb5NUm7vR23aSegFv;r`JX-h(1U$9S zUUWR%jhF;Fic?WFf%(HCOpt#@A$K|D5-0io0k=*;8M&KZCcpTP zvnt`kAf?xE!?6w$+5JN9{Tr}Mw#y*#`*k&*e zPpNhr_AL3|uEdJD^2T$To%4}fHp+Mln_yZY%ktXmWA2t_?ztf2yM>0r1T`1H8PD*b z&_5`fJIk>4nJii>t3v5*9Zv1kFl0+wL5p}A7xwr(dR(539>L9A@5|%(M>G#sX9pRk zi!Q{U7rw#LDGv?5Il1EX1YO2!vxb9rY|&11Hss76;5;vs;E~;oMflX0TNQR4b3N8z za9Jo9@O&9|{Y?Qrt18Z6n;gDPOa&gzf`x7GVC_ji*dFExi-$7t%Be7zQW$3V>-aRv zQZhxQ1D_3-25nnv=QK7{VvdGnN2-?99GopaC~BUc;~M$x=p-lXf3cEukX!EdNLM_!!L*BxLEVqxyy#Bek_-w2}BM+~!>6XS}{5eP}k z=QwQ%gx2G*WWtc)j7c+@mq!zB3&a>6?dD;B@kbEK&FxN8RWM zaG93QP0nA0Vtv~1F6+JFD#q0Q`~CwQ=}P4$ZQ6qGLdszIl|Ni04dMxYe>}0*l)Hp# zSf5x19&69R&H@eeXp4v6KP0&-<|h5t7Y#;jbKuFe5$Jz%1gz&rKsJ=;;Zruyi%JIg0Dk)0F8K5u zv>8`naYq~co_!gXcu0fuqFT;Kxd4A#>|uVeHC)2TD4eRE2VAH;=WeKpe}v}3;??Wm z-+x8uUn7Xu55*aVOXuRA!WL*du)**%dw(*Oc^FvXWjOd?5#D>d6n`$xMZXP(D3L9X zXLMPALVh#e)3d>d`aG=Exj@dFmFd3L8M>=)iW?$>P;+)R>tn`Z%z_A%-tCHGmzOb~ zLK#-tWa7`%Z&0;k3VOs0z_2Gfd#WT-cJ_8^7?Hx(N7kr5a1hlUKZBnU+iwQV#kE^! zvku90T)fB=jb{syRN4Yk^U@~+vu;Stc?c){e!%{b(`f1#h%1gXV|4l$BxMajYNjUI zozleYOat_}na-HYBFH!U0qFRM;={HEusgS z)fb?h4)QqJQj)O}4RBq#4+_8Fr=2O2DbJ7f70u_+*WLWIYjXm`>Ih?2lPnrvn}Z(z zq|o)C3YK3X{1Q@yZ>OeV#{Oook(9@z&AZV2X&3BmpN2Pk3ekJb911zJnhx^LQi9_O z+_^jm)wj6eyXX@9x@QfN!XVx&pN-Q}gz)#PboAPljnR1tSiSQqPI+ifaa(7RhIKX_ zZy5mFjeGF=ejS|sMhmT{jKj%0m!s1052(zZEi&|_C`NP|Z7-Zl+KiVrvzK*EKKX;5 zB6EM{sL{x--Q=R_Kzes2P-NL35DkpM{g0;Mi?NAVts{V0(|^OX+=X}vPh)k<6tqo~ zz^y|LXqw4~=aVFHcf)JOjSt1@4^y$yTZGNB)Ywkg95Yr_psSexwXL2)9{Tdskj9T2 z)~R5UpeG8;KZl%?8rY)Ddf6Ms@Q>+6)`xw9UounhdrT^BxOx^_u8FbtBp%;RtHEmF zX}EIf7R;K+_QkXP@upuK>x|vP_9y@0z7E!vZe^V3N@-jv>x#A?3s5%b75tR^0j3r# zH<7Q7ab1B}>=uev>ATU1Z#Oox9r6;Jg(&cDGnS7(kGhVtQ17TYHVLTUwFXmsQY3-K zZENv>V*!Y0KY$j-F8AAZ0Z#~RM#)E2=(D&0VO%S&ogz#IuB>j(k`p9MoNOY@s{Jc3?Dr8ArNg?S8JX?B`Vs^#5G@kK$Bz_HeC|O zb+05){gMIh&t8pBjuhk84e5CQFY`CA7sTIZ_u?T>Kinsniivt9SY;6h{Jj}4vil*} zemsuj^Pb=vwo5acH;%c2_1SmmGGYwUi?!n96yQ$$+k|l3x_1!g!9#9_X1c zCaO&%T#pvOswK=}`>y~$E@9rXv77jqF<536W})3;0V@1>76TvyQ>Uk)vwaOz7mG6g z%?uQHx)Ns(=wOTOLcF6^1rzz15ACHdzVcj&>DSNT{DY|&TB?agihIyB4scb`63qBJ z9&e}eQ8MEz;s+mWNo73JMT|e7Imp-+OX$^&L2%x$$~+nT*qQf>3k}!;)^DP@B#}HA zKN1Wt&5}Xrtpe^}&!O$@y|`Vu4|b%df!TaL+}$pT>YF;jr;gC?sWN`MkptYue$dFy z1i7GeNZj-ae0QC~fRIDDa#9XnSdk1S?Kk1Xv>>=p8U}sGBtQa;AtXly!*=_CVRI`i z^XJ3E7U}q@ssbi-iNpFuDo|#899EU8pqx)W%n;Yc%DNDUXT zd@hs}r@~l44dk32gZN@^94*Q~iv?%V^;i+}hTMRk?e_3ns-K%)RRC@p{?OtR20sd4 zz~X1D@1n9C-}SzN?&veDdtwO71#iLaf`8lu3uoNqw*{Tp96p=nmxt!GfxyBmP%~Z@ zle8Tm-{v2;_N^o6@5u$e*=xc5^kx(iXAE)Ii_rEs8+0T#;-=VqDDb_5kHV)zp8X1V zlCuGR3g3eLQvzYmpQUif=@xi-$l_k9IjGv$h!Hj8$-F?0Hf zk$7A>Gn!r7LTIma1fH41ga0Q3_yOO!$EQs}GlFrc54OR2=8v&m9Rw5n_kymBA#91P zf#DBgc>Ov1Ucc#JXW?A977&EDp80^&#^-SQm^a2rA)ftMi!BB-@kX&8!KgUPF zs_lQFlCiNQ0_q{Xp$bNnjigVc<%pDJz0cqC_Q1IUH7CWG~xEqe*YV?%wgYb!_(An$@f{RMw_t6%vU&9(U(kW0J zoeygjEaCQQb67`Tz};{gG+yIlXX^bB+tUt4zwbh<3S$jRg~6r_7cg4k2Mg|8fGXz{ zh}pLjWR-NW&*v(n2A9E_YZXwtqz3XX1jDtrui*JLcQCq~3I=D+LFQf;NIMw=jbdi# zvC0FAS7tz8*%Sz%EJ^ zZ?x`#udMeM5j_Yqo@kNblw#QUKoinRuS2odPT0&k#kw*P;87QkVf#P9qSJSQK0RZs zv|3oR+zYcKZ-H!}H4IAo!=inr7&kr}h7BG=>cL2OHgz$Gr+UDSf_3;wwi!hCet{>W z|G+UL60{Cn0j>4c%sUna(qn}XSK$bMTFN1&G!EiMuRzYrd$9fny9b2b!5qS2@kCGqul|0^w``CFGGSMb=Ej&j@to;OUl6eSpe+2 z${bw99%yLQ4hf1ML7&eE`o@2NZSDfF&uS7z-e`clRBH&@w*k~2PJ^pYgrL4r1iSAo z0fhkv*wZ8j>r**+A~7ANuiA#O&(6W9gD#HG8G^6zlW^MFUHGNJ0yGAX!@F1AV7N9O z9Hx1LTcJ24TyjB?s&e?5HVqQje23&E*WmDlvoMu21=lmuFfbegyBd^mMAILJ`1&C# zDG{drh=oaM*>FHs9EEuO@Og$duGy@IA1(>vcKdQj7dOUR7aE~Mv<}Wqjfai(cR(ep z9lo3P!LE!+xcE17yJ^S35yvy|HIuz(X|vG2leyJYdzq6#8)h{$z}s>HwHpGEc2o^! zOKbv(qDW}HCx-Jg{GjMT5lr~=in}p>C2nN?v*j0b;78LaSL&1oeaB_s=W-wLdn$~F zebiy?4K3DX$-v)NzJj9VE%?c1DVlR`LFN7?*2j;4++Tg%VoNvZEt`dDJM%#{Jr1sN zo*AVnVeqf9%V}x<<$Cn_}s0;imYoR-43WPGRt?Ebu94efI zRN5T<_?GVR0u1_Ps z!BOuMP+GeShpLO6;#2&1ddz{m~8>YL`v<;NvL=Z_kg zSDFfX{1?D9NeM0X-UW5bA8@|6j+_?zkW;cVeVB)IHc_3nS^mWIcZ^BOd{-Ug)yco+ zE=?bPM91>3kHFlFNII=1OP}Q5;-ZTqSaW$P8J!iTdv+hteprg;w~A2Z@&atxe3O=6%%GcWj~YLs z$@^_1z{~!{SWjF0D9v7ioDNK)|LUyBpYd|vy9hC-YAwxGX1-7R1X>XMgt^;ei7E`} z`q?&;3v;EwO_Ru5Af547!l^c;nrKPa1Y}MEM_d^=Zpyxtt?0LeAkkJE=LO= zOr*KtL-1NlYJ^{HmdJF%_a~5+q(>zrOX{W6n+x&(RqYoe&OBayC}g|d9)dde6UB-PX7dHQP}(5wYr zG~;$Hy$fZ181Nt|*9z>+yow5Y((z%~WZq}S(|tJ4k8HE}Xo=i=^f^<8@go<>Tkr^R z>GI_GFqu9dH>MLNEU)7;mG|dAEnYn9!HZ6=A|2N{Qi}~CRUt#>$dskoX9sb9!Gy=(Y7dSvOchccG~_& z`A#ZS+p9q$5ka(ZIEZ38*V1TJHkp6&BQwX9q<_AQW@+7~CHwU0+!k9}+q9UJG9)RM zHglyQRE#y zRZHlxk5qyMcoHdjq!3g>@3mPzG^HLLG|!{Bqy~8;zQGs6Eht{_6Bn^(YKw=*@kGaJ z^6)%Kq6g$@mI1pS515ncnFA-zXXlBZy~vr?FFaX<9!o% zY|eg*ZZEE`{4G`LUV4X`>#Nvp#*+-C5nFW%TdfK9!C2!rNN9%{V?HiFWeO0iS_Px@Sh~(9<_*5f}1_1 zsRvWh>Fua8R*F~D(r`U{mKc~!sMf8CpI7d}+Gl>)wJQ>DmW$yx>FIcE@?z%2n2cX@ zl3;M<65M~0bvl}wF$Oju8&-kF-LEinvnx0}ehnd74`9P+JB-N!{!$!(WkIfR-{UX1 ziSy%6?mi^3?x%^pC;kkw$C+U{xJ>;oY+U{b4>vZ$LIc2fDaNNfr2q|DFQ7If8kYaL z2w!WaqprYsRL`G=nzb`Ag}GGg|J}tyf|~TgXd&&6>xJLe84$Uu9~c}Q>~EAo;j4S# zx6KDzE5BoAgbR(EUX9tU>8Qqh6LPJ_&=kB2N}H8nv^)XB?UUK;FBDsK|H0OlH2BS2 z25S}5;AX`je9*cCr1nWQhCns zVDt&Nsl>xKGZE%_KaJTBV>>;?F*ArLF2 zBhfSG28PzFQD^B@7(AX1Qs)`-X-zC}Hg52+Q-=9NVjvZXZam$*70M;*;goJY?07f?&E9UP=GYJAwbc;t>^`*Y zG{M^Y{UG}52&~>Mgh84=;qE0d462gEZ+qsWK*(Bj(_Dii{Wf4Nv<@Z&`{6B>T-^MM zb%qbHxv*ju%?ZjUuikW;IA;aLKT@Csl?haRa|_iyGNaEaC&@795f#Yy)3)AnLf>(` z{@4;aG_Qnmp6(BJTuEVgH`- z+l*qS%Tw@L0rGL`$7Rh;ctD3`fGle0;FBooP7S59gc!2U+fPdWdC~D>OX=4N6V~Te zq(7HR$mQDw+WIn`Ws85KCv#F-^)QB;dk*c%2xT4fP;xw1L@A|bDMy959FkXkk;ZkIb*bseUpl4sO=C5I9RSXg1nEt1*b!X<#R+gD7HXlJ>r-Kch%i6T-Q_hw zZz-kzv1P9H74-G86CJ;Fk2amSN6)(|NWFO_9a@d_pw5iSGYlB-!ho_TGfweeQQnR{ ze7qCpFDdEZHmXhEM?zCK(8)`#^v9ExB3@|Ip=e>=4YU8bGk%buz)jMR%b>fvBWRgc z2KwDez~JkwYc+3-WEZ8=4zHW^Ugb0P*7f1io1ZZ9SvLLG6-9w3lj-ts3{9~JppW}l z-giWp_ljSHCvM(DAK&;;l(Hpp!&^ztel}fpH6cSqBf7RmhyLByr8!GP>0IU*p8GS5 zV~M}9#%Kx6dgMe-*MM>sFQA)i7tsC^OG@fYpdD@}>6i32y7fh!^&-T{$7?G6aZaTr zxsi1F*a7B{I8WhkBT4yGCQ0lnCi8|%wC#fveJ|Nf?o*D^GO-mjSma3WW!F#++jB@T zZp8HRxitOxa@tg6Oh*41kg0(gB{|Aa|KWFdYkC{Dq}!8kg(^)_9Y^8=%wX*G7t=n! z#S9BAQn>I3?MMhI%SSLA!a<#tF?3g05 zTU0b+4uMz)|iiB zK!8w4ipEB5mAcM4_vc>BF7h7`xye+UwR~@tz{`U6xLU zJ5N!>-0^fsNtiymJ;TgVB`Q5APT$g|QJ_a1-Hc8q$EHwf-g<`~R7KG^#-6n>dq@#- zWn{4+kw|(<;7(DTsMs(47xG(zXaBa;NkDPfwZ;qC>eAIl3MFJ znt9)lj6$YTYe6yAd6eN$@p!uF$4B*ByRfV^2e(u><6Y%BDBZRgPqG~JyJ_mQ{lg;4 zt~iN%&+yYV13w(ew!)k<$B*_Y%vINqRvS3_A+1BkYNx5E&=Wtl#G-Lg z2$ui4gm13LVA|?9{K7Ih-e3N}6}OEDb@}*s;VV4ola7^dgy>L@62-1pA@`L=m|p3O zNk@;ModAcyGdJP)4`;Aq+){RKwL(d+ML4c;0GHZ|(UYuJ_FQj8iOtb8t+kd!!kLF@ zmlT;QUB&4ObI`vp2gB#3W0UnTsvC@<+#klY@$15*#HG}BMuQSPXV8BqT&YO;D(>mJ zgGc+BS5=RXyk38Wj4O-r&@>@Du;)LdFLzKwF$`tL;xV;42^A$$u*F#(pTs|c4O7k0 zLn;enFZ$t?uZ*9k%=-VY?odlo0o5$Ki#9wV)}fH2b0XzvSaT1B7K>5JYAagxPm$z4 z`H)^|3ZBon#b&gp@xKs964ADzeuFd;p1KZWjwIq4!Bv>K(gE*JvcigywHTtbo5UBc zBk6lhq!2ucLBAGK&XF6K@jD5vtH0vE?jF4AqDa^093x&!BrX42gKvlKW6TR40o$$l zcudD_G7;F%d_1e}@=>JYe3t*V!3Sk}D7%?C?}|oofA0%?IP?~SclP3r!7e-ywVLr` zjOqBBy;LHYLCLeHvFlcz@+7S2x@I`K=gdX$6sDg}A!uH_8O3x~VzQ$KO3h+C^w;57 z&*z6zh6?ehL>_AI{f8fxXJGY#9~fwQ9cKsRvo84+40~IIR)V7R&8!{!K2)M~T_YM* zjiHQcDz1N?iO(lL#EWhB@QKn740>%q4N7ge{rOFH9*M_Mi6}IibP|7KCVnik#DvjH zs5SjMiu)cwfl>vWbG8zXJTOPeE0?g-$Q6%uC1c*{3wSU83fAlj$Ee8dm>JZ7RdN|9 zHmL;l#$UpGT^l^$ZHFOQi*V`PV6@dZjgRhq!1KoAY12dvnzPQB(hbuwY^D$`FkHdz z)i3aYLLqvrJb_-~pD;t`4bD`UK{M7WVT+eGdiO-5U>x%RNMFG{nGyI|VgK{nWvncmo9oS-tug3l=0eyQ~!wLR(uFG=8SDwa26NmUO~xOLCBTe z!^8>xxOv1JCnvF;Ka?Z?USm{fHb_>39ewAw4p^;DPy7?bzX^Pd>7!IorZmEpRuopkP{0a<*yLh+YN$wIlB zLXH^IoAP-iy1|Wxs&we=AxTP<+r$_<1L!+F4jl!5;E#7-sA%^(at+ELPhFOSy(dB< zj|E6*K_E^&;DTbSBT(haF*^Lxm6n$|QN^iMr21Wn#9!~G9coIXc3Ynm4l`fT>*=KU zVl`=YZ=}uu*17w8597Q3?sAD4URwPrvj1{oyYU z9^BpIbV2JI8+@Od{nRdgC|k!a{fNe@uzD) zj#B!A5OS(MKmoELRQ=b4&Lz9h`1?C(^!#u{t*pe#UtySODNC&zK7s!{$`14ZV#O=Nm9Qr3UJ+FF{QEB|<)&At{A>)bZp2Z5-%II{60F&3WwVXSpvc zSC;y`Whie?4Hf0K(SNLgt|o(w|GAKvv<5wQ^`QN{=eI3Ai{4m;;G}sD?hCxTV0n?R z&$6-f5_g8oE+pO3DB3AAKy)BMlHRN7P*;K;ReTyrXO@^!^=~8EV7`aKKI>7L(RBK` z$)C>mcoPdxBFD`k$l`AGBX;NbvszE4Cu682qmXtNroi2}75#Spz$1wWn&}!yHdc{z z?57PG4BkK&m#(MpV`}JELkTSp8YFspQ<{!A^Lu!!IX!yPg?EcOkj~j?BK}PtG^q?q zIT94Hs{ru0=_n~Z#{X=B0yYUYu*WMeHdH0*}(ZV7#vU`ui33M4<3!GAfihk8}0?DEsawyeZDaFZXAts%^y|8+qDct3_3MIwUK8 z0fxGtaDK@@>=wyU!j>2KZ9ST*k9~ry6aR*ZH{iKa925(aFrfQBoXS9CC%9i)OK$=eV4H%QT$r+Fj`8tSG>h$Nog{gC`<43%CJ8^0kx-Y zqip6$SRM{TtIH*PT5L!ow7H*TwKMrT7eFet1|4os@OA!J3XJ}OrvpY&OU*_4Q=du; zBX84`t$e0!n?@I3?xkWA?v1#68ViP>gO@}t?7v?{yxmuHs<`8_+(882FU5P_dzRhs z71}GM=-BIK3c^L2naoi8LeC=$`e+wslZ!dZoe2}*3Mp8=pNs5Zu zNm@G>(7VP&a$CBPCQB`%=l=7l>y9#MWvY?dH*@O0oYB30`lNi)m?*f25)wYs1g#UK zJGGih{98yW?mQXq@g+HHNAhb~O|K$t=;%=^T6Sp)eVW{eYq5EpX_f{VTX9jt-5N5x ztRPwv^dFxSKI84z{uDIkDo!a}hWOkhh#wH>#E`L6m}Nt&x1Ocw|1MMbUfv-bwvuMZ z7|}K(9a?F=oEF&dJ@|#bG&qL$g9l&1XU>x`RnVf1e5URHcmoaJvVdeJZlZ6|JLt~W zGmt$H1KED*NPL-#M5#*r-j;=?LuFW#`w#mYZsGdVlQcG&`}eM8la!sjD6e#=s4Qc! zsG(syd7btqJ9~F}|719QyQWTiDiuje=P>U(1(KF3cgC&MrNfh#(fAM|nanKeSC&C( zDLM4x3wIaBmC`i67az7b8V>#uaNNjQ6d$E&{N*9!VBU<(_oh^+YfNSHbjd^K5pvQV zWeX6L%@FeLDy-$Z$EhhdLQ~S#0 zG;Qc{WLDPV3h%9T@%8sslPmOfM>?tTUhpIJ*RYvbjOCg~Xp0!{t;O=z>fTVAIpH+z zeR71-Ec+-yy@fuH>?hh6(@gu1wNb&@deZ-9%-6S!%+_$<^BhU)2x`RitDNh=c>+00 z7m%^~I@%U8lpLpu==RbHI-C9y&Sw2;~|+!#2Y^xV_~A?sbjjj52A?ahOSK4|k$U zza1ZMx5LVUdlvndk|M3ArGsI+l)Q6v=*rk{xVp!Wg1@h!(vtQ3 zU9L*)QDdlUufRPfPp~HaCuG}t@ZCC@lpk})XigICe^Eg5LUxkx=95(2)JnrO8_9A; z13fXif{M@S2wYl))Sp&JdvzXxqf8+kED8HdBVaXD7xz|3fcP3qVNp1%%g6w;3x!yJ zJLt~ygOv3iNR6++p!KPcHV8(X%^5JBwNDxNkp0QXWBNv)(AIbr9eMX4nv}&xhZ?|S zmmGRd9%pxq-m;qEx$K?7S(L;)MeD=>yk0FqIgyVrNx`1lZ--#d#c5Ep55sAmO-;6) z4~5P781?QK%V$MQT=qQM;OvMV%cZ!Kz7_p;q~e+|lt$dGfXB5GW?@{;rhM{bvyEP| zlo~m>U6aAPv47Y%foIE2OIS;YJ6sLIQN&vy)_v<~PXFsv^=%-Xe&vtnR+sSOLo^({ z4#9Sk3iuZaPfW(6>t6tFKIo5h?Nn@y@qyRgN+jz}!`SXpp2vvbJ;A|L9*_&^phwuZ zelUzpN}0pd>&$Q1Y(%#j;L5Z)*n1@oA)BTm(e@}hHYjsbcSI^q|QW47YqgK7) zP~{y=$81IVUb~I71|ERX{wU;m)?(W4ZJ2s?F=uZn<9VwQ9Vy>WZ&b}lStSkE_S{Ox@kyda)E3jM#@LG@`a98Oo@hJynyt5l=jN1o7;kr5& zBg=G2JNgMMrm2$Hq|qqs9SI4Z1)DT!J1*#7L|JtZUVhmH_>9Czg*E6fl*3b<0k90u zW=1(Gn7#2Rl+P<+X52!|$S{T8Z3BF;J%PPRj#!qy4PT!h$IlnH@NC{9VE-t{s7!>b z_)x4mV9(b|a#%L!0&f23f@Exes=Zi%jQ(91Eq4i99?4K#*%WL|TnF*%=?JZLLE|Mg zT$J!Y$>kAPU@C)(8)I;`Y%%%_*FZZ6u${3SR?DuioOC_--uHw3nXOp--U}&44_NB) zu^6AL3z>jAyxdlTXPRLM>b`}gS=>$WMva_qPDHv&EhPAQcxF^O6U+;lAqrWB*IH_NI;ac}#vk^N zdH$$jqswwwW$h}oZSG^~>LYOH+IV{Nqy_!W+TeDnhUHsZ02yzYG(V@Xa*{xJct1F5 zat_t%KWHqzf!@oS7^TF0WvQG2E~8Hub*{nHQjWU(=LrSrp2Dl{wZd#&Z|0OTgpIr; z$IKUovXbp)EO(JL6Q3rcMZcU#TzVC~O>aTu4^lBFHuG5*nsE38PA41X1%u zA>DV5pr3zKNY%3yx|*GY1(sQY;d&>b?uVg}V6j5b(XkQ45_JTdqLth?IYrRw771Zf zD+Q1COu^7NPw>3U_rc3YlK4SoI;gn-b9076r?Q*bD<-g4Ro_|E<(tf0n|HxJhGR!z zEK(gbNn+ML7_1&a8d;%2)I}pf`-y?j@A)Kl_Gz=QQQ?i?d|Z~~b!(Bsxi6DmXA6~k zcM4w;;su-irv$f!jzVGcTH);A+1RGD6=9PMAuZ)CsA+`@fvbIm`4a{(*SHDH_SIMx zK5-I4&pu!`pYRT&eyUJ0YlE;T-dgB<09@Y{0OcMX{tlI6$$iHK9d94Oy`(_se)3#c zu(@35Yw5w$+HB4x4?>NfDtr1ROZf9eoVj?W3WvW&2&2TJgs#?Rw(QIaR+#mLC3}@( z?Z-x}@@<8ZiWHNpP7sD|D-cXKPo}^OJz_TmQau{bh9j+=zi>B+{2=9$zE(eCQE8E?L`C-L^>YcsptO8?uLmb%2`JRHj<|HA0iJI^#V=Y*VGB&%7 z!Q`W^xb-&(yJQWZ)}1NrZSWMPKbb_Kkk}m|)JjJS^>W?p&yZTCuhz-f zif7C={}bETbDmj>mk6hB%n=s-S1MFIj1l(FNE9+vJ%z6pl2Gg#i?Jr^xS5C z$;!$$2`|ph5vG1#BK(NXU}sxiGq1=!ta(O@5V<&3SfL#%=W1p|Aykqyn5I6TN)^VEG-97I&F|+2H4C1P za^H$24-e7&WN$jV*^@-k_pn3%Je~z~qi%yaxC#P)Q+MFdq)XJ*dy=$nw2-2#1=aN9 z`@jWxq`AhP4ELv!{PqHx>XSq-wd%;G>oJA@UW;}8%b+2OgGTygk{EWD21s3@@irqx zUL~@kO`Z}W@ho$^R`JEX#i{VfJC5bc&cJvZpBXw%!M)fQFFUUDS#B1&SsBy5?{<{$ z7D8{9{G$i+h=ACb$OHnP>2PfPfk(5-Pj9eyi9@ysY(UAYB@`63z-Eu!+4iDZ0d5jh&o zrQ2Hl=-0oU_+#&hkr#HsFMbitI5?W7xs0GN**8@DYnZEsd8y#(#FIv1XR5T7=qPYj-uu%08HY`)3sL2=UqihU4xg#yA{>6VTg#$@m zu^KBlfAFJC0hABj!M$tsGa6kyPJm(u((>(9z*ES@0%<&Szw~yp6g!hs>&B zF)@RqBw3P8H?DM$yZlg**h9Y0p;C>w&nf7qK1{T&T|-oGdAP{*?giYr5Qw+t7jW`J zBloWVrFWd=+~{pc)>kHw?(RGq)AWa~giaT!Y5IuXot30R-J58s+yQztp%lAXhmwlf z7@Baq2g59_sC2g+<;ma0FNF@Q@Eb%1WA5Ww+H=(R^m32w1d{jfgJzrqHOx$zd)ly&+f3_?h%W6W+_ZfMA8^@H#C{1H@(^{pWPJo=wKe3`;_^93_3K!{#Ov z&I_MQyIMSH;B9q^Rc*q~z%jJugArx^ol0*O`%&S~F?7B~k6iER(stRK*dxjPz84KB z=xi&sUpr5U^oLIEE~2@QBt-j#@gmXhPf$Iqz_V5YEphjwS#!&2;G0JJ_h1km;4BcQ zckZOLun@OrBw~$K6&{tuV%ni_eC3Y1B##wzVD>&byEvEJT3(~jc?g}FXv8_wMWj`j zM+RrSNTp?UiNCzxJ$R_X@DUr@=~y4eknS1-y$qngLHmOr0Gk& z(YIqIhIVX6K|~$m3Uz70#|6~B{R=E`19xu^B;{uXh|9bVlfN3&RUO71y*@a&G8o0D zZ$UaS4`*+N!fyK~>YFJcQV^Qyx_KESj60z+`x{hm1z>?}1_CDD#_ZU~lyE~#lv1Q5 z>K9$dGiNcoAdlxOo=zL^>tx7zJtBF zxB!y}pT-XJER2>6Mf&BxOgCj7$`zMmgq}WBj?2QbZa2PjmsL^p41AK=h`y%}m`P^< z%Wk~GcE34|um3$k4&UqZd@vXfe6Mo9$udYT3}OS8O+g)9XVs|++z&J!ST+I^3eH2X z=qk%}Hl^A%H(7>x5S!&F0k^D!xUgn3_J=DXGcAu@(y3%~p2xFBtD$fc9#hk1yPP^c^86FYZ5QKo;w)gp}Fi&y*Ougl(3~&I+>%Y9CB1=!u+Tf#+o_u zepo-u*n6L)Mt8BxY&l+LonfbBS0hKxpW)#J_^z7EU340>(#sePxl`d##n+Upa=62) z1r;{0xl6kro~9nfg^j~-_1H*Ue`AFk_5I-p?;hx_lPe>zk%a$+2a_DIXkr8ri`87oFDUN(pYTN$lUA<(Z8pdski1c zR9=9A>3pmuw;w|Tz57yX=@YA>zs$)C95FcV1WOEO;ES- zK1*|44Ji*_{0&XQbBQp#d8CABOAk!>;RefTgSm_DFWd7}9Wo=OG2VG398Su@$by>#&IZHZi#_23=riv1M7x8W0&d%psH#KMB+zT@osbEXG)^$S|Vd_R50seJCh0* z!@7xCc=tI1Oh(|p=Z9eK`jBjmWPQ9llr*Rxk}`**eJH_p_<2N{55p4!V_1Hg4-;!u zoS}cL`B)zdm{9?W9m&2!W0Cw)|WEZ&`_Qm}ycFJrT92yr12POFa z%>_-)gtZ5_i&`s#J6tP{;EZGd6!g!qeRY*A@{2z^^F@Rcv-#e)Pl@o%bug+7KC|iJ z_85D1H_VbxVbQq`Oo`}WTk4wFmx{}5?WTOz+fND;MsW`LU+!-5%wg|W9c4Rv#UZX* z$7&}6^E%XUC5Uqtq67H3o+1YC@n)Xa58&|R@9eKu6SLek1mg|LS;*l@xbo#RbG|A(Xk=O{5+a1nMmwRnikWZxf|pCI6Lnv z=t4Trh$nHENT3NtUEPYEV*}u9oeYHpRZQ}X!0Jblm^I&>O?k1FB?m{a$i3mDbm%vF z+uoyW{A7N9WP;jpQhaUL#A@vKvH43v+5VA#VIuV&MXtk0b=?6PQ|?KB&xg@WjWu|% zWe1|R&BbGnhkU<)JK0-mV7Gi13vD>aHa}g%wA}gM?erTj`8mw{;7_oLPl91d5X2^k zF;}ZDVW!RmCR==uZS*8|+BBK@{g}w~UHn-8haoI~cLIhF{)`9b?qSxz!?gdNKizCz zO^Xi>qdiZ=De!F#?u4n~)!_GR&<-(F8&+V4`WX5iV@@7z2IQyr5iY0Sf|ry?z48O( z$4im#pz|0K^%ljm#*vfT0=9dyHPdOE%~UdGz@B@R!~b{RMjgX0drcJWT7iv@CT!NZ zL2R(1F8egL3F>|k*rxp$=A79$Zf_EwPkd;5$raT3kD$+1&u~;S9u0dg!_TG|+nWbd zF+VquJu6Q4!qqWm#%w%j3`S1>-&EC;MT-5iX~4O9)^#V14bP8a5~F?DhtKC2WVqAM zVlGw8o=PWEH`2HfBgs2`A07T-MlwmO&@X5z&ee)g7dL@cxST~^^=7zt%FqHCbsEH( zXs0v`&?{pC+aC%zaqSMjU*l1-w+Vg!?PKwyqnXUg3U+5%0hYyFLf+Y0*hxQw&F@OQ ze~|#s?hcH07(!lZLC73qMg3a0klWiN?nf|1+OYmGc=m_Y<=SH1mPJslIEX~A`Iy;z z3gXsL&=w^#-M@is+YTxAso^Kq9%|!0!@n@o%g6ilN-XiN!NpZqvEk})QkM7xcSHUz zR?o)Q`xzK8D+O)q#Av~EF}l(#&?i=$C*Q|P9s>TMeeBe8Z|n{&fG0DUKF2c`19+M=naQUhBWrj zY_hNzO)84B*{zIVrlB9m=Dr*cl^lIkwaiBUySnU}gf8DRSjm>jC*$?(N<8D)l<}Lz z*uqSGHo!ra>EE1-<`K7WWN{??cwcpT%RGhvBerJ01fu-kuze~ELHtaicy=#rH&o-f z-!Z%`n~$iI`fxwk%x>O`W-p$cVGg6C;i`EP7Q^ykrd5uK-{|@9L_TRbFj;rIv~?J&aI$r8)uiy6e|kY2EEg$y!{GKt>bWV>lJvs9YZ5_Y0-k^PAtjm zvDvc8>#^8zJ%T5UMarXgmQ%VI^8{Z!N`DCB>_g}{It3L`)A(MaJHDKA!jzls2+53t z_ZK_Fe9A-(&j!oZZDsQW?mq0TV>SOhVfk0TF@r5e5H1~IMnfc_^=K0_IiJlmM_p#y zT9j}pbqf3ALwv6C!TP?DWF6581+(Fpf9@pjnbtF*I*$1}Pi5-|wXyAb>Db*{f&1Oj z5bycG)TdQ3_qt+sBW?_BQ{c|?O`Nfw!8t%w&9-!Ko`i=x%Un2%)fFeOfOun4|301y z1C&Vd#5Y#T86Ul)xcjMA8kZ(lumP{-Q9Ldg9*tLcwjvjG`)444wsH>GHRSkRVa4Gq znCS&wwkYBRPQ9)`gib7eJ8H4JjutH4WHWObIvyWlg7Coc1rj)GC-Ybco{tE|_NY!& zOVwbAf)a(;cEY;TmMXW6qGDYK1h{zNSkD-oI6Mq$_`z7!M67SsL3!&|_(cte%dL;B z|K?JbezBCvE#n-6g^QqNvlPZrHB5eeC$qRb1sgULv+_;GxHHKchhzDfQT#o2?5z#=clMu$&_Y+0r3Dg`V#Z*(W9LdYF>QRGL;`iimSc7EZ>qH+;@Kcm#(g z2O##?I!LVg#ny5MBboF==F`E*n<$0vS7$M6j~sYyv%y0bO>)nd#+SlL7`^r~jMVJW zRkV(hKrg!QrJydpZZ7bkmsKe@1L$JLgX| zDd4P`GG?xG!lMV%S+=)3+jV;zliX#AkY!KVu1WoA)E@!Cl1kLH;SA4b-CzdGR->s$ z1M@T%vlpF?%vEtZtKWZ`XdUftI9yG!7z7lp)^#yaDE=kshuE2g)224in zXX*dhup5c_?6F=CTf<#1jlYlY;=D#Qb50hg zS>hYmrcE(y=kZI-_}ETtyWxY=KKv{)Nf*;bh#`BE1uQ$K(?P|Dw66bJI&>ig`$ANC z=ZNPdRh0SOxDFCIlc49B1coIq##bc)-mUJ;hGp^DHiKo$s!_-cU0Rhjo5uMLfm+CD zl*Kw@SbY@R!FwV(k;Uw;h7zmMIL6l19%jVXIxD(C)w1&__iH+1=D*p}M`JLkCxyjc z`oT7QSc0oIOW6o{J}+1%vAr#$2~uVA8Is5N{YpM%O_5Hs)f8?_X4Z z)uMQfST<_n26kk74ja653R|=HBnxcc%jkbRciBkf%SGT8#V|jG0OtQHjDIioAZtZ3 z7Oly}x$ic#b@yKC){LTCQJ0vVc@|q-bC%saAC3t@X_z%Q5i+@6sM_}lIZZMYP$!3| zLnfH=)&kXIr_+j$-gLhaW` zZFt=>khFbA($95o*v>)nINiGz4sKVeIcpXvUARv363p30=dT358M}%jXyK5j2t1}jo4d1+UDAV6 zo*9_zCq_lwXBIJjE3MeQ2X|YnVA1LYS3eaPJqp` zw6rpqa6g;5t<%AZ!vgYL*I?wQ{`hid36w_f#u_<)yy>}wzO}coWX*BjCoF^G=o$>3 zw~{Ta31$i^H(5g9a%{d8f@StW$ZDu!ew#J0C~hP2G&zIqz-E5-ev#=Gj6ug=1t@DR z!^+IFkV@cdg}5TDt6jr9kA<@55~0lSv_59G`665<8M%iT*7hF*cPDeW1ioPowpHv^ zMKb(@1-3GIEqgoj5_9KW&r>r7(p)K7a#@uKuiqD7tYAcKS_5fDrx+dFAV)?E{g_W_ z2J2sRj?Lw>(dg_Q%yUlw(|mN4si*H{|1^%WirmvUpmh;#3HOnCx0n?U69OZ9a~1HI}~ZD?`iB|;qcpj0Cn30n$OoTd6yN*Xi5lL zibDD9GneN&hSJzUnxs&~Gp=K9!f#Oo>JRQmR^mBGXvd;!P9psGWnll+b679IT?jn8 zF=Rvv3{(ywEPfEVRZS)16kW;;Jx{|b(`ettb)+dTOSjVpQopXZ*fOvVFUEYs*|u)P zRIR1EU%z8jgd{0nR3VvM*8N-)Pzv zWJ1Z^i)sF9Me<%ZgdRQ`OG9@X(d9G^TFPCD+7GI5?`jMjul~b~=QZ%!^%Ty*N%Z~V zak{f0l=g|0W6=FL_-#pm^|{ly;&vLF#s)*<&?frsG>hcIxl85QDr)L6BmEN=bg0A) zsTDi$>-kjJ7<*FQh!-?|*8!TdViNTo=|#eo*Z4S|KmQ+A^S!hAaOk=Z7o9v*yf_4f zB5#Zt%f0lgjv|}S;d9n!qIbt(7?t#5M}`{xx|a_B8qR#z8%FJ7Paww~gJVh;($Y&3 zbm49$BqzMUFg#EQNV>CRDEVqe1ISU_Gf9&yQ3>zp@91G9JNsKqdEH ze?j006$)M3iYa66;F$~047z&dkNDwi+~@=B^0MOL%3**L(X4;hlL5 zzjy4yF^eR;%hy1%j1Atmwn8km8LE%vy+dJpb~@UmBI1TO znTiunV_>S2f~}DqC@IQ<#^MyHWIe@=juX&16$F_Z=dk|@XE%Q?guRUfMLyG{c%5j3 zWXn+Gp?HLUd5-TD_B5efL@ikd>8buc(u>$lUUvJ)*JvQg?*GiMaSH}Kc#4Vl6iD)P zIewQVQLBm(Z8kVVEq7AMEo}`sHu+M)Zet2O!TWLp823X@rr$e6)D@-0`*a8KLiG-k zIXh{Q**4n$S)9V!v}u1y2f{ZrLSNz{62952weP z*7PC%FEZRgF8az;y=e$tPdh}>ik5WWScz`FNrJXYDda7a5Xa9R^SUPS{+1qHd*VYU z3b&J-t~1@(ei}o&lkvzk0C@p@cylTRStgZm;CrTU>1TZ1u{{E^9~R>!ch^5y)P=o1W9h5MP1uc!18v%h zUnblYH18~=i*_SBA{*+jveEwhBr08B(URLGGtFlt@WYntJ6uP(T@fUy3I4=ckrSs5Z2?5W9OSp zR5^da=%DMQ7kG`9z2JSV8^dY5qz&D@{1jnDW9gvBSQ^(el(s!zMD7PN=uHB{ zTUSGA&E<9E;%JA*`62MTaRl$Wv#7^1k>+Pb(=>N}ADUI*)a*p`U$T?t&s|B&f9cZ$ z6C3h6U`-WcUgKseXF}|IOlcM_RKDDctiDLnt4%4~8=yqv3f^%Z-zn@?EayJ*U|i~| zL6>O&BrbDSV&o>o84n}P1@q|pZE>3SW+D|iC{u(7cXW(cPC8@lN%xZpHCJY@AK3C zB;nJI9Xz97Sk8G1yM19+vln|h_af;j=h^POgT4GbHfe?zI!4-I(YAT;tGI@3{~g9n ztL>0THzT*iUwCiw6MqjyV*ajNtjOe@9Q|8ZxVRGIR-Hj$>^Zz!>xanB-Dn8#qt-S4 zwCdmjN}6y5dj_VVrRFXcTUFwA=qLP`eGRYU_E4U+7ri?^lGb$w)2>yU$e!;fd);)R z!gm`f%zihCm)XE)pd~(%Hqt9qXhzag+&Ni<%^}IS_v8R1EaTx(e}Tqcc*JM19187s zjNZp@rxzthsmSCipBqj@FR4hKa5W^%C`T4Ff#!VWAjbd+^&ytsvQqxQcha>3H_41SI6yp)7bUzjxJO?I6oNll`$c(wm?E48_pn@>qENEA~gFAn4XEtX;np%jS;9 zNKb}W^Y&tnn*xH {5l846;W$dsE3wPT6Y8y!xk@1CMp`C_7*6W-D4raLtJ(h52+ z?m)e*&SZ66fgGm|qV5;NXy>v(`o1%e&QyEQnU001U8M%uPg3}{DwI0k1X4lsa@uc@ zLb>yT=~1B%SvB+Y-Ovd*-8u#;BaARqY(Car*n;@}JTqA)LwauAoHsch#S_h7+$YW% z)kOCF)hLm7x-<;C$Z>Z+k%U*t_Q2z*1R}3&#)j8JLexjxT6*TL;&EF{LY(62!^c=tSuUeD3T-GeH~+9$zhX+QLC zJBn+HE-?6%L6eMk)AS1mDD%=BeE1@dlDTFW)iZ^nrc5G-%hT!nkmUkAMw-@0A@&z^9p)5!*bSek-Zc&<% z4JT6zyewLT!-6(C_9vnCcLW}^ori;S9zwRBzW567k+S$x27Du?^sF?&sbA7 z|Arsod(9KYm!tiT8a8H*ryq*iRJXGR16{LlT%tccDY*l^xjYxS!3trm{9G|YoT7K0 zz(n(O1n@G^sO3_0yFiNnId%MQ1x}rjf)?*v1m5KSg*AMAlaz&d#@U$V9Sp%f9?yyv z;-TegtoXbKTbzdDM7caNMuZb=e93BsH&O9%>I{mc?dfajs&@uGuumn)Qh$=(XNfUn zjJmCc2rW{FT>mDvxb__Tsip?Qcj7RvmcnC)36z^Sf}|q`k>u3NNSc!d4SN?DJIPbT zC2?A|=_Pju4TI>ACJrAGP-OiAH+0XTd);|hD96L9Zas9B6mfq+2gdz?T?0e#p#iA;-R$*4w}_ANOJ=RJJZ z@9@TyyT>Tw*>W0{sZ5eFA1T)M4cR?TqA-oIz?x*{0Y2> z4nU9dPWTPbptrW_WZEZ9fuBRk@^Tmr@1DwCgxe5&(F=* zWzwOaabz3fPun~r$S%T#=6tr|F70aa?7c@>&F82!;WYW&OQwjrgY=l^yqrF{V`BXg zq(t7s0eeGIiYkUr=Sp&jm`1v35&!XX-SD( zeDBbi$NqHI!j?j+=kRMYf(qV`ro2b|*N$=$oqI|C>N#ZlSeKl~@T{tDHFRc;6rCR} zEmAsCMovc>QQ&wJXKodvdR8vV?gyjf%KrLq6eq9RqRY4iNBSFC*C_*cS9476c(sM4Pr`n2yq+-mhs)18;5k1z+qjO(Z(!E!e zFxAN6J@QldRT)adgR^PQe@-O7A(pOi`1Dfq#boY0NEBLqjwVmNOh2+O(?*|4nmRX* ztXpT&Ivaty45TRj^+d87XGAL{hf#v6B}HahP@lXWU9^^=J+JsVa%2xy2FB52$HSZx z#mGHoKHZibLo1!y`E1KOgEux2y*H!!is>Z(Y9bBh?p7r^6-qRHgN;=KXvfM>a()m` zk*#}ZC3o%2>OY!Vhc*D-B3g5AG|f+}LA&!XdeWgm#jc%@QOv{g=JPn5#67sKW9f3J z7}eN*foyJht=?rH{3#E3p}y=Z&Vhz9Z;|em!(ga?V6yGd8J= z#@_N!*pA(Tx~~7Q&}k+;x*<;NPdBscb3*$bKTP!KV(DGF@Q*RTQqy;cu$7=Ex_nLW z`6k@fcj4D4Gm3cVhbw(YP;uoRbVfeMX^}p4tui6y(jgdDH5RkteGs;&5Edyf5pi@h zi3Bmalc7xN?|B#2bPk@((8N3$Up$P`Mp5ZXOqhQRn`W6{{>2$s(fFwv+32Qm!l)&XspPW{2Q#M|LWSR}TG1mc}t9r`?X;6E0K z93yWO=#9ru$K4RWaTNRas-j`R7<`}ajpQjZbliUlHH2@adutz{>9I29ml~4&*)L4N z#1YX^y4+Vjo|^X9QT1DU8gP|+>b+lJW8EtZ;{1`!{Kb54;TWkL`NL?`Ec|rULv_Cm zn00U(7VTb%jNW&+xL=>_rq8AO)8D{@b5YJ(4W%#gyWxF{&o8yXIBpt*en%Y;thWx$ z;R8tdvpSuMT}u|%<>4&A zGzKDlVKC-Q;NP|wQ+ltyh~o5iQrO|^Q2+b_3mn&xV~-W8HZ6qva&LU>l;(_yvAArw z0g`>&;g{)zyCLB?_~ZOm~2R|2lY`NkE8)2PXVF z3eWS;`79TLeXCv~`(i(8$?aoz`JAm^Y=)rK-2GKC2WJ}g;=bi*`nf@qPJgqcdv$Vj z!(2o;yFBQV_GJWFHN(zdiyoJA4uf+bWY~F_&UZk4?S4#gyoK?zCthe^zp#U0`J6Zp%S?mQoaizKjJ*5NK50_K0@j_6QsNvhr-Y1SSKW* z-0&zKdRD;a!#~)|Gjxrc39(X59Qm~!%gZNY|9L;i@@FkNO&4neo!421c9E0bzBn&lph&wek2$LI0^Qyej^5`aLO)*M;Eyt5 z8iHZ1U`$cSK4`&hIR6^8#LQJj>n!=~y zPrNFOB*#L+R~zEDCgE1T8&bAL(6t!;f149ehZ8>1lm23&@4ao*;<<`k=I*AuhD&Ms z4P(0aj_01a2JuVPbPDp>PRoi9lN?`L%Jp=y*0wdcv}6eFkNJY|6I03PMk>YSCh^($ zHZ5z@;NF(unDcxR?{C>6rn*0d{aXU9$Z^zqy#p=0_no$R0#=xqV(4&fc$9pGb>VBw zm1)FVk8o14x6WAuKgq%J_`yUplT)Pb0d6TMzzN;=7$m71DE z`x@ivQg{{VsXb%)0i0j`Do0&&zl7!3VmorDPm*jrYpnV#G^Q$djbD$ zHqgP7Mf9oWCY|nfLHUv$2n_PUHr3S#Uo;<`@yF2b*a#91%%aSpb4ku17U3WHK48W< zOxr$!W~SZ3Y;_rGIMKyE>;h)V563A!6RnswgKYg~Q>oNR91(Lz!tyd`70V&4d;lt^ z4#2y_S8Vt+30x7^!XM)}*hFWdv!nr1Tlx1fEe`I5j_}{b`9%qpXyC5*b0&7^o_+*t zUAN-MgHZT?OTm`McVSf_hI3x(_@*Wxt79!~X>%p>r2n~l4^rsexm0`Y|2VtSc&ff| zZJuQcg{X)^kvM1XwN6oFC`wXEGBik|{*BT^GNlkgA@fj#GM)7+H4O_j#W|kF|tcuJ4g@?%v7O+|(jjR@z;d{TMr)og(11xPFmf zp6)SWmYd(_rW9*3zDuN;1va6K?A=o4{nHCfFGt{obfA(MI335>K3u>oOw(uD11B@j zOQqPPh+bx`oi^*yw1T;&pUE_sooB@AmN9o=DU;>k#;C(JuFUQ`T(eJ&+@=^mb_TPK zZM0g!h6|V?u{P$+Vd*f&V(=>4Y2V1kWrwg?W(G{~K@QW?u!!*pmu5t2zi|yBWSGG} zI?MyE9Md&4gZW)x#J&^ci0{5Ko9$kd%-DJ-GF<;^rd{Ham9jjUV9DE6B& z8ZMaGVWG~3sSPuB6$0#Bz_)F3oXnK8z2;tf*1`k`a@XZEuP~W6JsE{WNx_-~GvAkt zFpHvoGE?SuGb1iR%-e7;#@OTybKp=qYc-b0z6{7=&m}D2`nZmAyvGi4Gp|}PdFqRq z_RYb}o^5+LAOCoARwo5<-ne%#qd!I1wJ+W=W3DYscX%zM@lKGJJ?g}H!}H;c4K6!Th7i}@O>#;p97!#Q*?hV%E} zSx)jI8_tnG!5p>RP>y9_Bqt+0lC#D+l@rbhV}5G|Fcq1Cyu71aPW>>4Gt>AaXHm@= z&Y?>$IcpN@I2tJpoO>@`axQOr&3QSV!W=A2WMWt4F_(X=WM&Y_(I7AR{4- zQ8Ef+ZVpK@XU~swKP;ZabcW_KIuEOv?m=>!8tY`ZVlFTacNI#$E{l$2)s zP&{*QP9wAPNE?$@n85jR`Xa}o;4$aVxlnFWuNqg%YBRTRVkNUBkHctw)nfKk+HywE zyL0ktqc|MXWz5@tJ7%ESi`k>(%KmV*V*k}|VJ|=Z#W~%wmpe3l64&(Cbmr$w4aR+z zAP-^sJ!Xw|6O%V8%F5o}!r5Hp%_%qW;0VZ-Y@ph5w&vV+c9$&S98`MDc@a3mF>a6J zR2rP(oa?UT2n{Q;t<5Uz-zdx;HGaVS_&&n?vg}~?3aK(5Qk(@It~;5V2_o#j#xlX& ze2|&I0o*VxZAM6-37+<8fIB<#Gk3511Frs}=gigGN@ntxdZz2h3J%Oc9Fe1kIF;Ih zURWiWQ9Dq?l=oj@%)B2mey;UQY=;zE^ih(1?W@KXMEqe|gyh)btBCznKf$d!ZOAxD z>M||TU0mnlJKWPdBpI!z9Oj{eA+uAU!Adx8#d*`eoTK0!!dVm_$~0~YW5nN=GfE~_ z+=RE2xZxw!Tyy1crsicVb3&q$*?5x2X+&!{{dW^N9|Bf#-0O~T)YZ;#bcXgbr}bl* zL-C1>_Yy~rdgKmH+>x!EhD)nCQlFeS>7Tc9zHT_riM0&jc&D7>6e@1yc;1NNcpSUL z0TDTYK3t1AG)+!W?_uVghyr6Kxs;*GrnIkW~7;-0zb2=fEBN=+0vs>scmmQek{t)CS4Mcb|M&>Gv z$Db%B;qWvjz*UY3o*3d*8LZ>1+-t`1_s`?>1YBcUGAfu2pC2-BlCBFfn*~}DtEbFX zGX*v}YJx!?kYGKCMF5wEf|NpLr>2AY7}2WE8BV9Gl9B;M&%u_7nV@)Dn=Y8!L!xJ&L61+E&^%5) z++Ao!y-YUKzP)CYcf1!JNvfn()A!JgjSlpywHnR*A&(w0=g85B7nC(Nsg7?sM?0J5 z(LI=|I4adz(`6@k}_eQ3<{D6Vz(t4E?!fGF8)_O1;lIQinS?;Lr^TTE%}srnuIi z5Nb)Mto*@-JBXnJ8DCJ;U0-y&)*8JSe2rGGkAk!Nr_!ht7FHiL!0iEI^ke1&l5?_` zo{0h)wLYJk`~9J(zTKc2GMlK!>(A_(;LS9@P8m$yd7u_s2|MMZ7!BbQu%l%U9cgf( zd7^*e@Tn{qv9Ut!hqYkk3K5zf8%ST5rV<0YTISJ5e>U^rTN-G>r%f+}s`q=oh2LG< z=*@W()Gbw}x#(I82e^(9f!e;p9avT3jJPQ*}1bu00Xd;?i{3n-z?HK2V{uauOJZv}qA99Hw*_ zQJH_q%;%JSq-F6G8g`)!-Nenrs_zRuMt(%p34Kn$_U z`U6c~)u0o-4-$he)5hud>Cag;?49lfRR8uJ+B$8Bp1m)F4YL45;}*lMwKD93-AUx_ zoDljd(}DKvN{6geiS}5cvkEq2pBohdJ40(BPOvF6|JftBalJ?}0>W zc=(|}XF4AmyV|I+Q#D;7W`$I!1iZXi3i8Lcz{Yo4bjM`__G9g4x^!1Q87#O=UDVfA z2R)fu?e>|{|4bubO;#M;V!4*c%G#mNT(@S5GuV1&&w=TD!ajHCBV697w)0Wbp z-FyNYc|9&ee@TC)X)d`m-RmWI>Bb( zm`=n~b@AZ+bda{yrcbsA(XlmmxryILVPjkwQQmirOw$}e20vdw(UgDi_20p&9NR1? zdOHU^!zF1a++uVKKcJ2x1v*DjpC&Cfp(&9J37N%*0K0J-v`M;po0S(mCZbM5IbPI4 z?*QqNxB(673#pEc2WnFGqcbZVX-3l%+S$*i;=K<+OQr=b8J5te>n0(&5(VsOFM(#Q zEX7&dRe4)K=%9*^^RR#66ntUw4>EDx9!s4`rCBrWdCynw;k6jK@RS{H;PNH|`XhZl zesWTor?)Qze>fL{?FCs7Hx^FC9!|dW;`LhG^}rCxMVDefn^0cano{1BMs?mpuS-~` zIUG4Z5W-vH>hZqSt@w@ES9G9h2yM#hLm7G_XxH=+tkW=;m-nCx^G7K9;d=&I^SmM9 z3QJ!_K0&LSAA-KgF(kWdF3$d|4%r7z16xHIFRJ=4+L^o3YQegha( zcA;iDfltmt8?4f_20wF+#4GFTsg?RtoPJTAXTK#G+f75fvwP#QLZ2>g!fzcwSc+?$77-C|p7cX32Qf)Y&{CRLuJ{X(n%(mmRNcy*KTQk*pT7xVBywIeDytra>Ue6+HYi#H^`#xb?G@TCxY z>}&W1t^VzRqm?h>z1@}g(;qARfX4B9BEnG9fm7JTXc9Imcjo==Il~*?V9z_aXfGBm z+=Q<+kK@~MA$WM_8+<%hh^J~fhj*KI2P@?5LJQXk;cNHJsH{N|ai6$}i&!gyRPBH{8>0=zdurxp2h1*Y{12u7$?44$~#*t&pSFHi%&kv#p!}|yMvx!gX@Cl z46I_O-gtvEH(f#*SGl~3kLkGSwkqc6ZNP$jRh(qjhHrEilW!3Oi=?vHe1;nCsOiJ6 zr-xw4KOMOKqZKuIEeb9}^HIagEx5emINbM7#{M1AjOfQ%^t9=GBv$W?=LXzF?{|d3 z+;6&g(!K3Cs&ps5ka!h)*aqSX-+S0hssxS7EXEBB+M!|Db9_^#7O&s*2A`{1hNZ+_ zvJH({s8d9R=dGm4^Qt|Ar8Go&T|XA^R9)IJx3&e3T$_SLBWls)zVG;M%CbvBYEQP&ckT~Z^DFuIl}!k~nQVie>O4U! zPEN;ulkVfSFB#l9Z#kC#>w&i(l;H)B7eYdHCOyXJWA(=yAbe>z(qn8fe~uzvxArHN zs}Sco?|XwQqn==^4R3Jwi3t2qgTrf$yofipnetYR@*hF({ zWLXl2+V;<&^A?7IR)+=8!nzE9KPrZw9ah4VSKgyf>vHgX*&jG+&3wEv^&nR8%;#&u zZ2WJ`0M(d|BdfAVTs>rqc5W!+qrZ1xONA21JDKwwB>&-;W}{edZx0l^8G~-2BlgtS zpoKF(Vaad7SQDMYi|?or?doW((5%H14|<62=g!9oJu9fls^0)TU*WfLiqm-h{)9Z&&f-BWRDkqTrjT#sfR z*aY9d?u4d|Drnbe2-F0WqR@{fH2e{teiTWA!#1Db{^%mI)xVP%u8M|&kE(Q^fS200 zxB=|H@cFd|Qc;oBd6IVK6dHdWOPV}Gz`gx0_3+q->VAHM^;;;N_?|&-Z&e0YeQRoR z^8!T|Pe5|!Kcb!BijST2B8iT-LC$k2*|^z}B){}TjUgG-VC*r3=zW2cc^lBPE^|=$ z?Fc_7<-*%prJ#l9(GWR(+P!8PEXy;c(>BP`g!ev3m@EQ!IalafwTExLSrs{`$KY>w zkD*RgF*<2m4!dOUJ&4$;N&nl}!@eAP!R{Sb1%KhEWLev8^lpv7P)tvfMW^&AUvPF! znA~LxE#0Yab~3#`)Bz_=F0zleII`Wlfo<7c&TcboCGY3UQf9aew1m&lq(>WR%9Iqc z&UHT(*0@2OG>oaduor0vkfEh}^XSR4LdyTkW$g;V4e01Zdb4Z7z5-LORaEmUd9 zzW<1k_XKx#lNFplyb9J=tc4515|Dd&GdLc;3N8Ci5QTU#kkn3wW5N0S#a^$#uoNvh{DBD6R@5DQG}htUw57^#lh^2Bh@!&eYBrVTp^m8jOyv#f&qT`DqHg!&wgMGwP6;6ZIB zcuUT}3S%*7gUSMcR3|{r$D!%05L8vPLQ#SUycv8BnQul|^TS)=;GAcmzWX9*T&RIV zsEdd;{z962c^|dwqPeaJuLu|Kl#F_{uFc~R}9~Ccg3%q zN?FhL5a2&9hNTDo!(W^t*}tvtIBtTy-}dqb) zo5(78fp@-PBwQUlUv)Rjm0Y-ZjTD56($_BnD*5B5V7tx;i(4h=`%52SeP9;s>oF&3 zvnA2Aze4oZMMLaV(hqxntc1qrYa!KGlAp{2xs5pMq0=C4;(VJbTVXu-8Q?j6(Nz^`0<*qMhGOjwKuhUh~_VwdTVE=0V`2N+f zV{r%hEq?$Sd(idY|mPFFp`H(dv}uQ`5WQjW@{SZB}OEkhwz_2`Hx?GP!;UAYC_)%9%Q*Z z0#{uH`pZcU|2JiV{PlZ4R=)ORE{dAsb4La9Leu*K45yQCf5(spf6U<`@x9=`Sqq8c z2P|>3fhFqq$t_1C{PoNzx*e_0JTy^5FMaKC(Y?wjbe;Tb6Mwhb)IZ~^gsBWz;k zCHCveV0QYlS0r`J8H`0F=o#b&#?Eiq1(N$HH_{$;80PS7=+TpP$YD&1TE+fFyLt{okG?jn^zS1ww`TG$kKZHN zcJ)lk(s%3y6C;$7#X|8nbADqjU#q(~GBIql?-^R_ zwurO{a_NG`5iB{k9QqDTV_#@oLAk+y(965CsIF=h)V;ilH6o&rVsZjHF+pB{3|@EUCDIqb3!&C1w`s)~1Gtz#K@>1I=$>%0VK#g8KCo99qMsv}n*YB0jya2KN`J;^QB>$(oa9_}(;SSX@4c?DG6@S8N7)@1urfizM*kc_Yl| zeJ?!z%@v2CUsyc&7!KPMin)%*v5uKEKJ-%-b=fT7UwVBAuQbR;PCk3_uLqV`N}-P7 zPP%{-bB(axiu34pek)p&ERF{?ucI?XfI9R%arnDztfW`P+Q=)ibv~={3r$%(?0SqH z8E!`7C#CqM+K;&z#i97%OnY?dF-y>jsC%EeZO9RLk{| zm3k}+mz2bYmLh1+lSLShePE^YwxVdiS$J?rl<4mj#!7=;xLGhN*gfI^mu;KH~R{CSo#e&2?nWRc+A z)J5Xi>`&yD@|3-$8I4XkF2z%qr67ldJs^L3E6&IDWTV4pVmJQ=lJc2@1M-`3Z_hdW zEHV`PY3bk-{j1oUWp-#_v<{`1O)z|?gD5XR1uKlSpoYwRB%I$!o>$DqoAest^8q2O zxI7I1F!rWWGeiXAY=NA@B@Q?NVZrScX>>( z($Wd&r0XHvD;$V#JQTy`>v`DgtSDaPSb?lvipb+B8hC79IA~l;V&|zyqA1x^e0|IW z|L(}g#rwmsvRN<+zaZ#yeu8`@5MKN5DO!A04JJk2#dn4u zvcdzppm}>EHWs;zYR*OBzQY@_XOjWz_DL3J#4W_aYk63>M-~g47@(Qs<#?sXa}?L1 z0aAfSaPz4wHnB|#N+qta`(yvX4Ne#dl)ul84tDUL-ChAHHkdtrN|Of4NYIwG>HK8z zM!2%K1f?MvB+~8;OJ)dTN2$x`y|)N+b$=^zG_ynVN6N@QP6aw&91TVtLhwQCEAf+A z4Q2tUAal|Zj)Y0W2}usi`X?bBkwXyZw~=ITDI*5YolsPS5Y7EAL`9=kpg@}^q^+D` z%aV2@`gS%L^`;;lR)dY3Rm03bb`NF>v^Oa(OF3K>E2i?ZA3U{eWUmY_XXn3ZCcBPx zk~1fG$o|O_zI9g~)Lyy_qibLAUwCXI232ZY3)@VTt0{#)O|vGOjpaf4&qC(&wgE_T zo&<_o`ZVilCer+Q2=K>Alo|iQ_p#gwPf`u}qCJZEoBBT_U8snkx(1;;dZolIuOBH} zpGFx270gDC1(c+vvXdP;Q0`$d_C;?vIy_Y$znpHzsD{Pze?urr|6I!+|6v3vb6-O7 zs1?jwaRchvRC32Pjy16tWF!pcg2;tJvbz5g*q3V}|L>TX^}C^Tt5wh%(u68+tOLhM z64*xl9J|)*2iYYWMec1$VEIE=(T+P`KuFpFS^k|5yPQB+!%z(_xvL?71#2{L{9SX-n8&-0J)72PcG zDZ$LedPX%Ey<3OmU30;R{GA#{!i`wV-c*zOjm$u1H?Sl-$sdBn#G*5{{yP z%(=O>DmLhNRm{ySX7A4@X<9hUE|VclxDhH@zc}MPAyp28Qma|Hv!Pu5lks4D zd@7k*=S^P7JHhMyb?CIF3$mS-216H)p*MXWxD7Ow+&Z5CCsrk+IX(Fh^kEVl-n#_V zy|sZukniH3FBWL)T@$mHb!sTF`4;+rCD!e=?Qtt75FHlGZXi=kJ(Wu)ep z27Wr#7WL#C!#nd}sO@nD6@3>XBXgO5!*?Fc3lT@cZwtw|nG)Q2n}&S#<*2k<6^e@2 zgjXNO(Z7Od_@N?7Ej&9=kuL=mLH{$oKv2V!iP6A-39!@;!sY_6sgmd7G^24F7L}X> z9cTwiQJ#b~dc|-!Bhc9RRifU&a}XbR5PjcMfUdrKM8^Bu;gE_xofiEADocDIB)=Ix zTTe!}LT6&}b(@jhD+She`&1OXxB|8xyNPz3vZ2veG2Qh}6Ql%I`P)0*ppW0mNV=#t zexh?4h=?D|UzN+hsXPdfQHy>YOvI62H-nsd4lKO0l{yI!$D$iy5E0eN`etk-AIC*- zhq)hnDAth3cGjb()44GA)RzvJtgBKA4}-H7ot5vZ#F!Z=m(T)>X=qd7XZE}D2-0m8 zU`sBm!9}7SWnGa4^}W;Zufs~H>GoHA??(>GnK4SlcK1W9ZyUHoUc!H`V*JV26@zF4 ztnH0~nn)g-e(JsUe$k51{BMjPCp!aeCXtWwhsX!;on znNJtdo%2-bqB;8T(O;bUWw7Xb+F4p~<|})b^+%67f1}yILvilD2DI=+4tr2Ai%2SI zrN=~V(8%4bMEbv7kUD4vrEBv+(svV0cwj-B9!|yTK{v?{FJo9bB*078Jw^V<^5J5E zAFhtyhP{;gkgT&ibS*tW?fsWwJ?VHHYWbZwzFS8xPP#=;znDwmA_v4ibs!CgCe$b0 z%PMG}!zJBqDDYM~$r@6k(h@D`)~@#?Y-RzBPML<%U9G9#Q#ow5;Un92;wfSpVyI+9 z8Wl46jMDyuuyVU@qFr-m;OWZ@ap19OIM8VxTV37`S+{CQ!0|_d`9a zOd3B!5|Qs;$;pdq^xLg6wCM6@5Z|AGXt@>`?{%QZme!+l=Zn$&8$~3nZzntZo&M=^+ccqC@%6`zwy^OxSiz0rF0L6p7#L;;nHjJFhzFRVax_wR}ZTU~^ z!xkM}J~4vM%nm_Sdm2E~-VzRPpFtls4FfM?6&jne42N|LpurCXsC4l&IQdusO&N&e zzx&n?7s3vc1B;~TR!=>;@aI|f_Kk1s3<2aP9r^{*Hw43sxPvfwNFGU>_45<_hQKsj z8D7a6;wRs2Y5cb_GJVq+dYNX%e%NpWG`>2+{U|3AuDq4*f~%;z1mpIzsmLW|Io|AO zh!vKc=kxzQMZ>*^QADdNYu7v<1@qs)ERWOFbM-SgzV0&;vG&Jp37N3Hg$F`%Q>p3n z(`bf*I8y2E1zJ;1I9e=y%4C4am>S-~aU+>D8*P8pMbz9sp-HFa<3!_7oF?d*X4vh) zmL^k>%kfX>)-!WjS+xkv?X$SPlO>V(SO>CPJc2gsn;^ecJxKWbDAQ~DhjskmfHg)E zkcEXh7FvIqY!Dx-I;UDi3a8FT0e!nzW>F5men%y26q-ukTo%XAJgc$Dk9HFI)e(;> zpM;o%LC{Gmf{I(eSm)H!NbKVT5kEK&{-rsAqZ8qGH1a_sPMmfOShD}_?xRu?vGh94 zMNOs~S)tknBKsqn&Dk4=E~XxXTQzd%aA7m_2}hy!S(>0NEsdoQ4x@af7f9Pig}iz> z3AgQjjir0FK-bs-9oyoCE*a;s2ed6KeY2a$!H`+FZ2D3>94?OKPluu<{%^^KM>}BF zm?eE*FGKgb>CoVbVbXPF2A(@G9llQdVx2?+;BmPK`(82>V)eY~iErXCW!MG+s@hRm zVHLSOMHqh)O}ylZfkzVB|3YF48MN?()B zqsgFVJC81wOD5er;|23_U2@@}ClrXShs)tx(WHn|=#a`T_UL(eYLf7i3Ch?Ds%r#2 zX|p{M?zzT)AFdCL`X_w$c?Fb8}3&Bs_@bk&C*k=1AI3R0DkF;pwhl$N-c5nsE{ZWh!gzeyqAV0P< ztdae?p`L&HZxXp1p9*BZEpqdJh2DA&q8X|Ckwe{IcADcNI(#<*)2w@_pZp`EPSVis z`k5@7#)I$uI_$Z>Hb~iJIo?z+j$TAZg1rWZ%=oR!F5)MnzWx~4Qx`*urhSEa|5%i! za}H#`wXu&}jfk#GELwQc0{Lj`Ky}<}cB5cUxAv_X&h?y3=yP|}+E;*<5A{}=+BGs2 zACID>SJ%;znoiihEfZ4uN-B@6y^du4BKSqdmRM-H2fnJh3YQrCLZbD**<)QA^y2)t z$c!(}KC$(MAg2My_g{t%U2UfN>&NUS zSXxOF{8bFW!G9qb*qR`r_0Axkd|k2SZ!lYm;%Z1|lO);4jHmxX0xnPF#NtZH*m6u_>R)#IH8y ziQ5BKcu3%@n5_#wFJ|-YTy20ZWU%9vPV{A667G;3M=8U~*!Zk0>2Z6{R*09d1ECV^ zEEg+&(CsYtTGJ&mJUGD5?=MHu+lJWB+Ha7Wks?*pZ6RNNE8*I;7l@RMc~!ve1!(`0 z)5xdpEoc3J1DHP#2U)ptHsRz-Q1yrqVCO0{cKkb8@zy~wBX*!KH233~k5%x#rxw^> zFz=ZNO9#&{N&JsTm|1nbO!bDVsig|BzjlpvIqZ@GUC1+>IVjuVixcL-2d)ja2vNB=quBEXux81(o)4 za2&s6beuo)ci5-E=KJT#7ULkK5x0=`JM*xuIzneIRPh%kd%?B;ykYaVTGsLS6XY@Y zkZrL|L^IqX*n4F?Z0p+{NKVlMNqaP*{j20)pH>>;My11?0Wq{e#+d1rsA0E1N2s#> zGO|^Z!_4VfC_djC(!bcCYbr)4Qf>=v9eOA5E^x=!)-zP-sM-2Tf`5q_j8~*)$?$zbtQ1ywZYzG2*kga zpuO3xOeGAmgKO>zbZbA^=sBUxT)P<1)yCjwmp#{aQn9DU)`pFzAI|Ok# zrSQaA0Zn*DBhhu6apZ>Cc%A-I%&B>bJ{;A-Z4YLmoZZ2&tvLkiFH-}{&~)->ZWvr& zYDd3FIAOaGRXVqFAI-Y@2l3A=LUm8Q@G*27*?#myX<-%+BAQ0NiVgGQ7CnJ<<%^Ij z;0IJBieuBjZS;EfOY~Wg_k5uJKI?LTpzSRPbe%RJ>gmW@+sRT+?-}$lnTIa3H_-Xl zvT#oTsa}j-MHjy1!>YI(l4Lf6{B2zfKOfig^-h?QwBw58Sd#)XJQj}cTK-2lH=qts<%#CG8Y1s74*7c>0a?njl?dM;KJv8Gr2SAl0w zHFNUbJK_-bg$xI3;7i@{h*z=(cV98WN`j0$Qag+ir8AI_$rccOIE-G;HKAw2-=ZiF zE4D9H9BOY0bo2cIP}%c{?O(YLQJYJI|LF;`H{6XbB&-AV&p*-e+H7>`;{osq?}GdY z4xVZ_%AR&`AWH-+LU53UtpnxotLP1qT3$mwe4k2nf)#Pe^ENi=w}6AQ>Q&Vu^K^cZ z)@sPBIRgVlRqS})Csc6Gl$In7fL9(5rOmX!eIa_d^X5Xn(clBL)w;ovT% z#zyu#;Rki?%niLBs4E(TZ@XSIXP4We_4|LYjur}3y!<)&yYo7u74wz7^p29x%gc~| zA>ogtT!-xsF2LE77TD#`0Qi&)RaRK8Loatc29HlS(H3qS`e>m;Q14^(V^K7eZRDT| ze+6uRY7kr%CXmi|ah%RuiR|1g5i=hNaFH1}cg7klGgS%m?*`y_PgUA|%^j%;T-urz zAHuB`qnv$8HZ1s!R`Q*dI!**LJZ!|7~T5#+B$&3n6MmX5+Qn zZ{vwCg65UiKqhiN0H8OB5w{fxV-Vl1It%xq&;Y^KFF)As& zirQ;4ATHLL?l^ZBJzmjFUW_lL+5@6A$Ym;BlhOdks?HHRVOJ{kFOALTy3hxD8>o-i zFvNVC@J@?%W05XT9j1!SCqZkLQAnC~u@kEUCeh3pBRRpRRBm zgp2?A)5@Sn?1S*-sBWz|?0Osze~~hUm!l}FoA4bfr@?luS;!=z0h}VI(oMh8XxfNw z_3K1yI)$TyPon|6*7OvbUSUKNmrO+g_47dgfHY0>Orc{-Tj*ETi}qZMM=B|K^t9+T z*n2XZHcS?$qTFg4pBzp*?l@2l>D4%pEr4@(e!$rmMj&rLhHfd@!gHvk@AWRi68$Dt zQ`!rx_8p+QflhQW_Z7+4DkD=`vZ-NfAq}uAqI=SmsO#WD3i1VX=e@_!J&*U54 z!+V&PYzt%Gj7UMUBsA9#)3NWX=%PO>&;^Z9Izz3Caw`nzJcB0cW0wL6i#4kweRtA# zUH=%x?m%q2I1B}EScIZmL}=KPN%Y9cDeO~OZPb0>HGQu+g#E zvwGm-)*gs<1<*P47pfX$@Z8oa@+kN?m7Q&Yik8H|>b%`}rj3A`yV9QK?Te}M^pK^x ziIZ^t=m@b~G>uF&og~WkJSuuE1@09DgXXqT6k~T2PNz3QK*bVJxf)6*+COpUr18nB zsW(Z+8#5sNB4B2zh(v-n(xiuraBt&AI&6O*_BrlgneQ6(!{Zd{aVvw~QJ)2-9ePz) z6g%LMyEy&aUO;nB-=o(g_&{_Q)3fW0siW=_cm>y>@5)`UQM*pZKFU$WGh66?@{g&{ zWf=&m8bv`lF=Wt9i&jWi(fqJPx(B>yN8$uu$?g-(=@r!YJQrwKzMn=BS5jIc&R_ri zED{~hf--wo>hOvKyPdDl@m0drO)s0Mv+y)}{_;Ee)u9Sv1ix#g@F6PjKc|&}=lIJb z&ak*egF06D(5{0vRMz|m%}$w0%cd?uMq7mNyxR}Zo?Da9`pY#Cyu_BCPNekLOnK@) zDnZ@5!qB+z6m0xw4*fnPgNt_9!lfB?w4it`U32LSUFuc^`A2hUGgpWlY|ZC)^Mi?8 z!x^L)r$aTGrLlzQ5a0LNCxDh!!8007Rk?apacdrxKc-eyUed`u>vA8?y&4k4 zlVriK_CuhvT!)w>UMEyr6kxZjQ&DqHUA`q z#_OvTW2QrhiyW~%%qL%6RxnnJ`p8bfH$7$BXIQeuLSReKM5eqV+@T&Z0nUS<+@S}h%&?m{ zyHxuRF^SJ3+b_j~{Ml8IQga&$V!|QWP?Nl{3xmRs)5tmlfw!fH8mL-Hg5)}3$a!bY zUI^+V;&w*F?4=3p{Ii5T_2CN%p7R5`KJQ|yLyE|5O&z+TTi~nZH3{06{DY!GZ@B;Z z6$x+l;orUFOD104CfUkY$f%bzonVXDrL7`#!=y8?MnQ#E{uTz!mHPaFI8CV88Uf*( zPVxQkd5{mONgz`B8*&dAg5B+dQ2%@yXtzIRZrV~(ymcCM(}N+e%fpy3)@7w6GvZ=>cD7twNQjEZE>de zr4FN;rQSr#eiprc#|Z8!rxB0EL9k-*K1}jaqmMT{hp*la$j_h^g4qO;@H7Tis(t6* zO<&J1oE`^Djr<5N>of%We<6-7l)q0}iRjPrWj99NVKycOqdRNNKo83i`Jw$}@y{Uk zr+FKZUaCps);=QJV#dk&n}WS$GY5L-Y=qBlwQzizYh_GwJDgZ-Mi;kAg7dTGAYA_f zG+pPC(!l3@#mI8lVeuWxL$vAlY#!7WxT3K)!-Nx52GUQ%$XszdZtSNh2+#dRE>8N! z{IV_w8c;wkX`O+6c?V!nkSMLa;4By%&qbjTvte~yF+auNNX6!Rdq}CD8*I_6fFSQ= zvO&j`%qx#0N~4BkgGn86&evvLglfp#`7B>=^fdfR4uoKS1o2s~MII?XfCbVpv08N0Z2 zU7#b>m97%>g{I(dn8@za6N7cH+Q>Qwv&!%Vj!3M39hC3;M0Olkhlb(-zSD0j;uoI+ z_1C6?wSaeaxxJQXSqj0IDfMKs4^x>lJeM2|*}%_w@`iNm*beooyQx`FyTH>W6#6rQ zKwMlC48O*cn+sNv<~ipHvquOX^{wV#y7Q5QdzymA%w#ei;YldPG1wEcmDm=4 zBV*?)Vb_~M^63Xlbf1XA?6O%f`uz=wUo486kK~g|$8F?##8aYhObq(t3P`?25D|Bp z4UykF(A$R8ospVAI1~eK!?`^ z7Ti>{EUpSfG7!C zHw2urdA4-Rr?ohyL>fQ&PhK#?Qf6D8 z#~Nv3)uv@|e#HnoEI%8YmX{#+rCafJGMSATX@EW*dF(#l9$SQlv64pnDSCekL5Vs2 zxjYpnp4gJ?&NHBWHy!PK6$d8iZRo)jUF0U1N!e$-W^bknGUQ^e=BxLTa9a`&&KtS5}?0lBQ2IR z2JS!tTG4$Goyc~kyL%02-&`&-w7EcyF3d-(#aF@nX~pn8bp!PiNrHKHBV_!Y87-JB zNAu)zkhQ`Yy4m#-^e(-ET^<5FdLW8VtwiAYW0>9z6C{1UDWYU5A9S%i6dV>zZh!bO{v`(MOYDZ2SqM=jmj_XXAhel zq%9uxsP?Tp>{|4RQQXi(K7N=$dkbP=G%5(me9DB&52xVClG$|SrQ0a#!wCT|#u%$q zIgmr|GtiNZ$;im&8tlg`vT{F7m+HI)qpbn7;cYmWI4a{z?@UNEbz)a%70{C%M)>5A zD9yWcfv_=Y$nvxrwFu)NJx&R-_^HnP67NPg&VAVn4AuQT2|xT&!1~q-kgL%~A1q^WMmiTxlx(5mEA=7%PZ69=H^$kvDpWC~ z7rtKd!x2S=#4-i3%J3T9PyGOA)_;doPAtljFu~4Sm*4_l6ZFzUgMF?qMF08TK-2^&$))nPi}unQFC1W|EhfOIW4(r?n?5NY#4P*cjnj4mhk zmC!17j-clFvvM*WldJ`))9)Z+b_vbdxf^>L=-@R);h-`j9v7xf8sDA+wKzcNwd)WnmX8;Y(m|fVra$Vi;%T1k;*qp((J9j1e9B>jiEncS29_YUbaGM_^vC7`7Ibg3Lh^e*W4;L^4-` zzMZa5ElkdV#iwwH*>#jw@TO3crB9*f;axBcm!U7_X2QL<2VgDsr4c{Y(n)+dRIot; zzWU9h&rF^`qlzEJg+6RW!!*z;9s;&|7Jb(eMavh(LRG<5YQ8pwX1DF8YrafEMPWa| zC`O6QcT%U$X~XbpK!Vmq&1T=mPNLe9`)EU+0K3->VTT)&`1VQGRBKWWY=2b%rZ@@I zKeURfnJ5IabG@Xg zQ-Q>2DNq|>QK~MdK%aDWK>L&ocFpZ4WS2!DxY%8z_hYtEkscG;c29$@Y%!#}y>u}W zU`_U7-6&Ejk}CiDKxz<$_jB+$%^CqaAPJ z0P|S99(M{2KaRz5{uBB22|IXWXFonS_2#CjOSoijKMsDagG*LN@e?L5QMGS8cPd@T z)lVn#^qHPm_WTLoeX9|!ow$ZZ-+ZyBWieN{TZTFMKhXZDB%1Fs<8t-N{B*x?c$@A* z1qTg2eC$EAR=D*X?)eHbLf!r3d;;npnvOdZYH;lKiQFs zoBUF-dVc~_dR>dd-zQ?ix>QVgGL?I1IBO#Esdj>+dNcyIVd9x3cjlc*f8 z_XqsKYf(Acnm_Eghc_z?QMPgr7QPvUwR*{d*LwzU|8oW9r%*m;mJ2_*N0z(K+JL{J z75S#`llgncD$G+F!X1YzWA&?FxU%RzNG`7@O}d|emh0o98{Lqs?F>5~4-kJ&k{5RP zX(Yri88!Y><5xcx;?Cc{p;#>xY96~oV@Vx*CEG>Px{=u?HXa!%@EitbdL?wlod7y1|DU=S3qtA%8lN8+A$nQYLlPi({cCE)7;@bh#oKDs#s z*RvrfJVwW&-J)7?-u9nlS7{uzQ4*NI-7na*=0;I%w3t;JWQ)SY?}*BKdsrD5Bi?*J zk=(HQ4?Ik7!-GNb;)5d{$l!{U#@iRCi(e^y6*YQ25Yv4R#C9gpcI&bS5J~lgSTRWj zE($!V@9VCi}npt8})~Zh*%K<&HryoL&7{S2XL-5D^M&j{QpC^x4$EUZK zbE^->QTEd;lwU9kr;dq%;{*1xZ0Ff{=;dYXRzD8m^c0+aRFA`t$D!ZiYwW7ODlA-e z5K11sgmO>86CDLS)W(kAKjGEbaZ>RA+G$b!*-HFJh-WUQ zH||)LLnI>|$ht3-YCdj8pN@xQ^P&d$Bx_3cU(UwxUsclEehj}biXbL>zV+&&@+QNbn zs`Prtcwu2Kg8MTonAVlo5U%Efvc|88g&`I1z8=L+%4y>%KXt4eFcD^MVQf!;4Zdvu zOVo07=$nG4G&Xs*{jIOU`%jS}Ir!fQQSIeg@t~A}B>LDc=5?)yX@#z3&X;BI)`JAL zsj8efuChR>ZHL(JmU6uR)CW^zAHe8^vp_<&4d3jJWyXik;mi9jIJZ9(rhkq?1r-k{ z(bOj^uDl^uui8=CeL7CR^cBCI5;)_cF`~#}dUS)3oeNOz#`}w7a7fj5!Z!USk+JS% z^}>DNongnwvcqt6lpSt4ZiF5$roknxSk|z{1}{4<$9YREK*l4D8E44D-#!)8O(=j_ z16#x^_+jYKYyrPDZEUTM7ou>E9cxubpAEkB(69%vLvti;X54Ngsh6+z|QVF!4BUr#e}vpCgEm73P}a5*>N0$h5Mw_cLzjzgeSE$ad&^G}2Uic+8@d~Aud<_+&uqkzIdSZf z6*gu+kR;o`juAaEX(Zob<6y78EiT-+QykrY3-^pSqlYpRX_@H`yeZZvxmz;fa>W#! ze9VvsZQspmMj4P3<>naVSV)fhHi-wW&$myUEodeVY5eQBk=*O63^95$5vD&`DHzAj zLe=jsEbo5IUQE&CPu3)(Xp0l6x$G)vtO9#;!vwbE&tf2t1NgC{9o%e19$&xbvslXS zIl4>{%FM0D;t6cT$WN`PwZEK=>79lPAL`@oDH2piy%17zmU5fEd-h7J8pL~6#o&&v zM`E)aDV|aP2h%K()F}SLf2V*qEs$z#>^p^L)6_A;D}ba=n1xa<(;8*G?9nVK9xdiM zp|Sa0TL;jQdp_AJ)wG)c6fs*HDMJ>u0%jhp^e2MV+A1KjMp z1n<>xT)o+azm2c|qLJLuYb>VzQxN4Z(gc6U12}NP zc|4TY3Hp;Sp~s{+QCg27IVSinno|$J_KHyO%8zAt{uGGsR%noy+uKN_?>Ui1bT`w; z8-;(>3oyn}g~!V7!SN6E_{Qo@-1D<9#7EVL7g$SiK1zvaoajU+|IIl6;Ya-M=4-O} zu&QYNkhkKp+}m*C>U?(D$qVCyn$bR>fxTN@hN*RhD1SqmAGLMH-qlN(l#~a$PH97l z4c0vSRVp^kxQMHU1>r@{ld$orG*{Z)21$d?uod5fFi-0)N^LHHwW~9k`k|G0@%B|> zUF^ok_&vl-ndeL*b|gLto5Izsh4=s39(KH;ojpzX#WK?*K-Sg>`;HoLqdzM}A=lEe zZH6vx%_aEl$Zz}QrX8r5SAy@&k%Vk{43=hVG2_EO^ei>y->=5t>#M%FJ-!onx9)*`F`E2t!yJ%bdj)hz z8NO(oAU?g&5PM#}5H+qH%3oZ!_2i(N?(`t}+ z>!EnjrgivY*BIP)T7n!doQD%Wm7&_g2_(vE0nAx(5MK?B;}S`hP@XRqZ*+*jht;}# zS$u|?B1a{1O;yxSLzV-_nR zPdm$wBv**(Oc&G{RS&1s$D#YtUEp}M66U{NEK(D4Eg^bmiIVa@_&Y%#-tWD^+JDrN zy{@TXc`6#>wyRRkq?i`GkBg)u}&PcxAPm;bGc^U=`xX#)qWRQw8ABl%tG8P!` zun(30&TJwN@jsctL`wY*+;`EyeODHOvPz36$VuQwhB!fk;HCUCz?YmnY(YMbl7Jt1 z3P@|?5vJRdMYfwU5TO1(c*mJ(|n5c8naNYMUxL_g|?|S_h3g>L0gZ^58hwV;2@~j-Z_ZmiTn8vZB8|#q&I}X#%+k&Xc z2(yy2NXh&neo8-;Z5^scEuY!L`uB_A`e;Yu;+4rWMxG@{r`ggWBTvwug1>dVzt%bjHhWbQ?}u_w)aeeV{l4FE{9l^)oEymoqy!>&y5CF_Cmue zB&Vj;!K@EiuvhRz{aDdNc1;~gL}BOIxn~iuK*&laeqD##f6w8KmiZt*W+=U`HH5@A zj-nTOmZQF)Rd>&P4KdXx@X7ut@y@NPkn&QV?HZf}vmgkEx^~$I{Vo>$PAOnRx2rUbNrSjJ>G!touWqG z&2%BQYI=OC$F)ZL+j*p-cs9Rt>kdRqJ!e5nb+89>@n@un-JW`u1=uyRr*^3zbydi; z=q~_?ai2wg#%ZAR`y5#FllbR%cZ0%~TkKoDDm0DqgVj?PgU;B`Z0>~hP;kkVPFOyY z?)+zoHF?3PU%L;s7!2fJjw>)<=>$>dr7*lSz!6#kPJyDa0r!s70L9rq;KKa(;$KfNjUeqyD>FqRQQ4F~as7EVK~xahVmc zaYz|lHXls`1&;dZw+H!;sO9kPQ4FaSw9sMOQn0Rd2aKy9LHCPwXv(};7+~Z9C&EWk zcX16|C<}vU-&^g&iiJ)F7e)SNx+)$UrHb7*{kX|&q4m#=vl*t3xOdxroEi{7ACHL@ zo0p~7r&?WaXqU5w_bdBI?!_`u$`W$w z2iadH$A-OL580QD!Rluz7|m!PLyGUQ7jxz^o6b43BPpJ?`qjbT5`U5`quO{pD}p?w zbHUetI{8<7OKd*OUThc)kT#@+JhfI~-^c2JDoZ6-c6wpqvlJkI@32$(z9QulZtT$J zuj4mKgp$F^hgr=EOHrW3Ju++KP0`4Yt!#6=9{kvnjmc+o#d)!bB%{~@jC$im%Z5Dx z!^xTO>6;lHK*SKLn*(tn((>YTy!U}1y!UE)EKU5}(w^P%{3@OIV^GndQ{{6x0R zmE>RVctc385$V>BBFm%4LQQ`>rkvEHHB9Pm(W>5yXIbIJ!E4+!q*QaESdK%IG z;w75+DS^dHwvoN(BKMxPOF z9k-k~)f>TH#W9dA9}l->JBX#P5_1}LP+%L@lflazVC+^WSXi4*T4qZ@o`(yBFDU`@ zq>b=RZy$MYt^%)>?~`{4JITiTL*TG^3GJVr1e5-Tl9LgZOya``qL>^32dBBQ2g*&N zId4)xHS90hIQ5h`V$euj=uGx&%~g z1Umxfeg!jBh(gWAFrAG})bfkKX}FY33~4 zT7pQQNdkGvE96#PBz6afH2OaHNimNp3rObENCED>E020 zHMJ2^}i`=%n&2=2fEw1GEOyvr#AE znX(PKILd(Gn1STzkRBa znh6lolSQY`l7;6*jBGq$2v5hk!M?1Y7!nlGc)?NxW3?PfS8M`ln9@%in`KzP z^Bz&cjuz4DUtys3egfkk&Y*rz3+r~6!J@PevPL%ru#%OBInv*uckM88PneH0bK^Q~}RiZ$e&^LD_z7AUtFN2Kp65@r89@yZShcE3FlG^{uT%}xVe{YrdIdLk?d9t0Ol zvx#Iz6C3$xG_JX53r_?6$TG`ebj-k+G{du-+6_p?0nf+K?fY{ur}`o*UEeJ(IikbP zj7uXm*`w*FOAE=dW(D@kz5-Ss-GyUUEx?;O2gTv)6XBbj^|;2?3P>4uiCj?aC*QwZ zB$hL+;A2`3$k%&g3+;jT%@lVn>lb!beyrr{Y8-E@OqaLphR_k7WbMjkvTp5ilsfrK zWU)0B9|l^1+PZw~&#J-Ft5Klky8(_B)Yx}4*u$UA@5zBRUT|gPVT@}S*jVN%2^I2n z#3PKO>&qO7Gm3zu=aIy_GaWxQ=h(|>$AHO@4$+|fYjEh1Xr}sb8Qfj?iFEw9$DAEx z_`xAt;rYx&k{n!`BV>B!TEwO;Zwa79X31=hJEpc?mwxd zp_GBsx1S=(x}o^3RGaO*QpzMEXJhi5^O!n*AL&$>= zkDSoe(-wfEw+sVKv$4>(P6H#aE`SsRf3ojz8Z0>#0cU=Dg3c>7d|fkBys|cmR2J_M z70mv?KHnoqYo?&5Qx~4D^T1bE=Ypl@Nc5?m4KU9Qygw)5id-?O_w6OMwKFoji9@N^VVj|3EtX!#KKWtsyk~s?i41A8fw2DixG@GFM4o|v>MPmTsha%atQcZ+atO-$ zfw*LcIUQznh-?Y6L&FmVFyh)H7OVN3`J3&>-xK!Z`Q2arvf_4UKG!Kc^*8U zJF%|LIc)PoC03K^0#Ozdq0nGF%e*hfk|i4~mrMJ-`eGVk+CZKV8EtDn?h03v- z;wJ`TNdEbcRbK8O_eLq8MAjfa;{I8z6K4Kb(ZlhJl>`3h%*5T610d?`1+nZpEf~7A z6h1uqLFO8V!&WONm>LraE2XlC-@86^{qY0uXXc??SUJk0;DwiO!fy{#aMXdfMD<)N zo^3h8TrM6H=}-I2oJ&;r-24twq2NO#om^pHR|?Cx;41bx@5nRLoltts8;G9W0vqk` zqw}U(fRrXtyKol!1esxTe>&vFFTkTKN0TF+TVclg(GWAbkT>m*7P4nKTy5%Sme^nj zy;}gCv(Dkp;40!G63)W9EBN!P10T})8*R2s=VasYSJlnw%ov0k3TVS1+UsZ2sm%Z)AfaC zR)IEOS7ZqOeYRllCXe6gHoWF8=sPd_(DJAT?EV#pxBt!HC%YYSgC=L+%$Bn6o1#Qh zYxhF&d@GS>@;Un*|6=fW!Wur0*~BL`B#zJHNY)&8gAg z*Dg&ht#a6p<}$J>&m1<*Tfn^xZiq7r$76R+E?fD&35*xpHy+u$6C;gnP-$~AgBDeJ zI_$79Iz0JlH?-&^)BSHYoH$qrMCAqhlP+}W-isn8d8Qzhw1k#be*+cuILIj%!_jYd z;BV&%%gMptRalZA16t3O@J5&+ zwzqC3F9p5ha$5(hT@}nsPU@q|`bs=!odnh?r^)jAt)O@OCd9`n(O{J`B8ABD7%UdN z{D;4?nB7~5T3{G*zd*QGMZqV059wDaSmRp9wgxrn)6!R_-&xcuxfQ&GNTP;_)3 zgVieqFK2?i;DK$!IQQ+idx#8weZr7WJY$L8^@Bk(rbN7UtuMTgvq0OE8!-9Z5m*;F zg3p2r#8=J@Ma3+DoMbo-=B%jg@(pEgSIq&N;M@94OWp^g=q|G zQ}CL@d(tfToYjWelS9A5Sx{jaal3h+S)WJ;6R9V7E9eOLF8ae}O})noOqI}8PM0gq zsVBt?&w{I@9SvU>3&zDkm=vcZ`qC|Uh*#B;oj zYd=k=qkfkYsp*I5bGL!?zUery8hHSQJ)cgt1x*$C{LRdLh_V z9ibz}oM&?*_S3e%Zro~H0ofqyCwi3z@RgsopW>^*Mprn2tD`k!KQe+jw+2yD0|Hyd zUxlJ_GwPY3DaxuFNv0;<10BH`_wrf@$l9!i`CBfD{W`~hlUgp7o+o7Et$z{M&kN{l z&%yNQ$bt0t{0?ezu8C^S7t`Z4k3iE+L~Mf&;r78>>Bc37z&D0NgW6WwW?Kqt>o_Sj zeg-FI^^lfv*F|xM%Zb(644j~~SZvUgObaH@#m^y=Fs>{W%3mC3Q~x#)x1#Oz_3*`z zbco}L;hRC`ycYdAJA#H*yYLYb7cug45`Qv3ALecREUwvL427>Y@uad0US6!j8?CiL zZ}YTK!<*uvaPMO< z#*Dv?1xqsdip&4!S;=zEP)EGKC5nhw>B6Jc1Lz{lqbPMy5A6J2fpzsJqM~-6+`F_1 z>kD+n594B>y&@Ac@+9bv!Ywp#cstA5+C*BuT!8ZTm7vsb4^maNqU4o1&^9g!UVDZ% zxMjW}`U^kf=Y6KIdyX#8nU+n$^&RN~Ib|?>u}|cnJRI2TJZA7lmy8QZA^QW8@zP)u z9wXMld8cN;`PpisDN$`~)>jQ)T@=p5PjAu@3);BldR;0pM~^1ks4(XP?(Br%+xyxO zOeeWr#(vvYlrOU7Y8Pz4sO>8{*{(p{gX{rgLAPEG z$A{E`Z^>=&uqgrKn-!4v!31lQCKCCk`PeOc2N*41RnID7leDiZ9o!e|A~iFOQ+EPU=P0%b`&c;Hi6HjYEksTG1%64 z32MjOL^nY%?p?ZQ{o==Ck1BqR;)S7Lf0~0-s$s7Jn_chEk?F z`19ie%>Hl;cju48twL7gU)xo*u~>k!cg?`|k{PVeG6EG$hSQj9`VgTxfH^Jhz^Z}y zY~S2ae0{(SI`;l#ub+?MI*0eO$>UUb-iA-OUDlX;4cdQC{-^97i&UXshtRpXN; z*kgZ6!LxJUp`5H=G>851Im)Ci?ZKT3 zJ8{O(X?TBu61Pj~#SD+Dcs~C&{4AbG*X#Vms{v>6SYNvRo%6N0#{DIZtR2i3pZ(kz zv{#He+2e7vRRR9(+K&r}t3jD?SGVN<|Uto7(ZgRz%UQ6z_- zWoMGT#XBM7VFFj3UWB)OC-4u8M10|OY1mgK!i^^Kb7N5$X=QghmS5Y|oEqJQLWC zZ;DQd#UE4AtaK=ze0m!XU6!W9KS&9A){~eRs)F0Db`hnwXT?^12f*&6H1An%0~vMW z>EU&m+%i>`=dQQnCOzI{azF*yLuRvo%3b(%LnBU{{t7Sq>yyAM3FtGg9mh@E1sW^# z*qS2_#Jr~tuY^T#GkIn1cYPfse7MEzdX8eGz~@Zfuox2~KVrr%AGUk-2o$INCz{-v z$FBj!Q?54rn^O{=vPr}bLp6D!-%fBm;7y)9%EcFljP25+jx%3_heT7_mX*BH;3{|Q zvDe3syA)o=6h$@u;w8iE^8=Yp!2-PM{YV_9HxqlT+fXZcCzTT3~zsK#1SR!;weM4_-Kh&c0toDusHY`ZY)eg^^mEUcH0P>bKis4 zY#pZWH5tB6tisArXKt3;C-~3GakxSiSr#6JPgiXr%BPIDWr`&KGAj`^ycO{pjc4DE zByyc~w{f?SSA2HpDz>GV;r8}^aiel3-*Ms{%ZV(*u1`Dp4q@lT+v{Q1l&7T6HiqAf z?89;Q_Y)nr)1vZiMkqf?kKTOdPmc{Qf_7PD>hHUPruOGhw++fP#3_n8I5pC_Phx1| z8=x~q`>1QsWjfJth(pTm9dwwyE-bHjidGT1EMQ0?-PZA#>~+{lJp>)=Khr#{+I3gt zAgSko*K8fmWe#&V`zn+MEl8x@>*DBW|1jFWM%d|eCxgf5d^&PRI$S<|0p`t7ak!Jy zMt}CpIEdMR);Sv}Yq?tC$&CP$0t=e&Nl8fhaPYkIvYLiKZG$af>?-= zE`08ncWA9XM=cB-90nV{h4?TJP&TL3$j^oJr(OpOM6m zW6T{TD$|ul!-W}p_ocI7xn~rboSMQ)lb%AGj1-V;4TmnfjSi-!lN^#l^l0kE82T~L z3$*^rfZvMlOz!w~==jn~`*lv!(2N55sCg^BH_sl1pNoKj?bTFqO&r}UR;SBj4vM=< zm(!USrqYt-H^FRX7}l7@(r;eDv`lpuZTC4tm9KZ8-P4Ku#;+f6!hQ_685z(tpM!LZ zmYlMMH;9j{_Vn-#1bmPqze$L|t2 zZ_Xy7>t#%jCdfDBB&I`u+G2>m;!2-dX5+&hex!KcOd2pYorbwSA&0xIA#76vTJBg% zCF+Z?;gJR$@B9ji^Q!3X@MAQ0>T#;r)dyGX7SjmxG-C5K8NW*%#y8Q@JiTfc&Hw!g z23%C2<$Wb&-=i}y$W^$OJ)Lk`TbHI_1iag?LnDkDLHR;A?7RF0QeW*Q&H64lI@c3^ zxI{zX{&h6HJ(-Tnd`QR7ok{2Z*GHbOET@HQFM+cUWAl<}g96n9g&YOa~Ya=TXhQM6|9D_8!QlQzX;q z7>jngWs5G|oBo|Vd|&`=dVfhv)d8%UX-=oC7)O&7cG5=Mc~mNI4=AWBQI&rhw6Apx z&G9(~w=^oiJ!S$|e)A7>Z48HyAvGR%UP1k2-dQOhEMyT=mY!E+VZzOuUTAkYs zZ-PN(X8hE&6EH70A2wy|pMU0o?a4j< zDxiFTF139tO)DM?`PuLcI_O0S%y&>{I?0Ok+s$D#&BBLA1l2OEyAN|)U&7E;{ph!0 zIbyuPN%VNfN>$eKr~`xP>mY9L;*&<#NGa2y$&Bc)_Tdg!zXH3vg?ly$dY4%_S03m< zPrg0|feY$UWy1rkuYQR6vwg{7%Q(1MpeArGyK$dDpH8@yw2NepXu# z$NhN&y^prjzY8DW-E&vK)#5E!{=E(Uy}__xY&uryt>Mm!tEpbqN1}T(8ZuL(sFmw; z%nXd<+W%$q0kh_E{QNtJ=%uNA|smW?ssFRZzbHeNMbK`y@p5Gr^u-H%LuumgX#ms;Gd9> zBY&Pp)$%jUSF;#DudWhLsu@iU1P+X$(>PrH?*jZ>U4lyl|NNUfBXIBwp{ZSBAonk8 z#X-N9G0`6jPT~W2Ynn5cQ@TP7)xW@Cr9{;Btpf90!B_fOjmLeQ4;@8E#K(>#K=w*? zdidKk`eIcGn9S0l(IN4C@1qA0zio%0)21->de_E+2Y2CcKs2hhc*0!^d%j8kHmFpH z=;*E*a-%|@J`;FW8~0k#=hYIp_(D8CmOq_;e^tO&f3oD$1P`zO$;o8>CQ6?V7)IAD zzlc|BC3qVji}l4rx#Y*a@N(V<9CK(mANq77$j(|t!=p9nu?x}EHGeae3Y!Eo7dw!O z1by&K*5>NQk5GHeIoP`UF)B}xQ%`&`Zh~pJK3p0_&L5S+U0+l9e_!3`Xf0=$m)wm016*jr z<#(tscL1$yk>>mIH85n73#`c6h5w$e5szN=3?7?}qtP*%bZA8?{n`2nmxU%mPuU%up7hQLy-DR?T_8(WSZ9kDl?1<;% zM*m~417>m+ODCAZyYbaLj(HKoFeuWC9Q!XBomcjtlc}coX7V>Y`FRvg=bm(BRU&3~ zM}d3kFx>swn$OUkLH&ljg`KtgdG&Tp`s(>2wpqrEUechnCrW`t7G%M$^SyX`Tn*6) z-px1v9RulSz4IjRzZc8U#U4a@VIXr*g5-u%8FmT?C#y6|_MVH>_(%2=PVym6!SjO=)jUj(H?3(xm z=ZojzxVje9$-hTl#Ge)ZtH!k8dI?{2B!>4T$n$TIi>qcnVIEG&IKLqnW_Eg`#`V4A z>gL7dSLIN8xm_AdHR8Z-O9ajuE5{WC@9>6Dbz1a%r$~Q#8++Ha5$M(r?7jRls2&>x zvsRviqY74NS-uPB);1$N)#AKlv%vGxgz?{0hr)cOX* zqyy%kc?DbxQFZuX{8exONA@g-ty!gT!m0wC%Y&$`-vL(o#)?k=>x8mKTj&9gv0Uxs zey%uIo1YWQ<5+EJrlGV`;4NC>MA>b0izpR}6@*ybZwWlseUQr=jpPn17sKhxuVB`S zhs>!)f<`{=0`F}=42-9N`KeeqX_A7K3zHzwAhvPV{Q~x1(s*ph6ny;7$^6~o`TVYU z8mPY#d=9@HutGnd4-7lb#*Rv=KkR#hX>7IQOE1~;-yxIuU?bm@nn>+r858kZ(&@uor9_)c>!eXBQ!OCFD=WK=A!GJcNwCwx)vL;}2v z-%X>dJBi~19kkpzTF_rI!2SMFafIY?XxlUa)~g%xL&7;;+8W3Fw{;5a@D#Wz>x84t zO}LN4bn@ZB11#z9MfN2L+-~OLw&N<)=E!`$UV9|J_j4fq{^ldx(Re{#MH`bfI?6m_ z+8Oa!69>M>UYoxk;Rw$o!r^;fI#g(yg4ZKAuDMx}PEeJ?UmJhJt(W^*d7u=Y+H1>~ z72U@$mHklfAco4MwJ7kHj^r&)-_+?tHxE> z)+Jo~cxP_wGM~41%m5A3X^=AN4to-O5DQz*lc0&G$mY{Yu>SlEND*dKk5^LgVty)0 zI6IM%HMX$z!3PMQ{*Q$p=oJ;0K8BFasW_!-5kZF}sH6jV%A*lvO_nEqs$K}mbH8G+ znHK%AdpvR5u!FqFw&%H*%y>(!kSFach1f^dwD621F;mjz1Fz@s4+Yf_6S9j}E_qMt z`^M9QQY*+P=0G*&qi8pi=0z(~z-WaGSm-OlS*z`kS)RggY%%1O{hR3QKXP{EObB3&PI-)clJgIUL76uY@388 z`&)_m{oy3(xg&nBcLH6-P2_6LWSmpDfv$DEhM{jdLAqRt+K#j3j>Wq8b4W5=Tq7`= z3QQq4*^v?<-*$F^618|Yn@)hS^v(}$@FrUDe%F3}TUyU~C8O4gM;UUX);Fx^$$5A|G+!1Fkr%!Hx}G#G?|%(qGN1=y#hN^rpZUoEkU^_q^Lp zrPgHe=aW3)`R)1KefI!+@6M;Vy!bAxoSBI;8pC*}V=-7S@f3E4g&0yg5?@YK;lEaV zC8F0Ou;g4Hk3Zs!H{xu0?>8U*?dvOCTIB{kb7Ud1MV^QLd;~|gX>jjp%Se863BEpk zLGjP-(sqpHiidvRes1xBM~=43V_|;Wi9I*ZyGh#P(d} zvJDR}UI&K7&3G%b96J@fxW4}nOxjn1(dFl9(4jf7Z`NS?OCRWF7r|HT`I-bTy@JK9 zD0)z8#_wba8AnA=T;t)xUDbW~vc+bcEib`)bMoyKzX1FAE08P86u_^}V=VH$91jT@ zkEK`q_>!MHFx0&nVahn?*+u)t5mRWz`qR+9>K$&< zie}%1K$vz*5#IkxdF9O}jJi98%gz)tBasCTmZSXhE(v(EQklpaVc%0Z>29q+sVxt0 z7l-0}kAu)ZVLIBBOoH~<71&q0pXf=+kx%hQm_y|_d?#wfzShC8I)tLu@nWWSxQd&- zHs{5u4|(P=MZP0317*#fFnW0v6qxqn@Pk{~h?Vj9WyJ!Rsl5#R9*x3it0uNqA-G{o zsWJc0^cnuj66VZ33i#rw5%)Q8T+}jYF+XRm$y0At@WUSh@TtK?U>46n|2480PW#bW6vihg zrSY|m<6-A*A1>A6%$oylqldgT&vY@uyN5NI$iex(zQMTDv_QN<`l(9pHFIqDORc#KTS91o6*Cz0E`TN}EHRm6N`)hgO4PG#^ z5jXuk$ODN2S69s>7earqC%IO9MV=Y|_o)jTXK#f;gKUJXuM+4ji)T-lO@QvP;p9nv z5{6AZgc5n1aNB};d~jks59R_mX~pV_ov`tl ziD*;b2h_Ndimr>VH_j6{meM!R;k0{Ck-eKhkIZ|^M32+KH#~sY?P-FEYU4p`vlhyy zdV;>P4{4sFjgwSn<3i1Y>|o~{?1{0%181$d&dD-%BFuzO`7jzccwdBNL7l?h@hksg z?$6E3@33o<6JhfF0H}%f0%yZXT)t=p-&Sz~oBul|P7amDHFN^kixuX#eMdx5L8Ixn zmk#v5fH`ED_abUtIgbX1k0uS$gZT6CJ7h}00dgdNFf9FOM`}99;TR8H8nL<(qzcE8 zqQPh2;?_|39z6mkyM)2+e{mfYvBa!COcOZ(9UEE2cuK zj0Iahc^VjHO=SLCpW$MMQfyABg>xQ*@V@GLu-PnVtc7(V`cf05w^YHlOF!7TzayB} zgtc&{wx0aiHyTbTUl4YT*=RaR4Gzq46dhlqPud@x#`6njh~q4MiGj>ip0xWdiFm8P zI#hyar@tx^G-50JRrvVjiEO-aCoH@Aj7aX$~Y@Scju7lUh)|-;ny8!T-mpeG}k*o;Kh8B z^k5&8)4GW#UkntUDKhBXrvXbA&t`Ukr^uJqcCeprh%MXoK;hb9xUFeSli#f&+d~}D zO;jNGgXO88*K$Z(aEsmXp8!X?eu=k!uYeVn$8c7x0k+7M3Z1eG@MORq_$E1@C_I*e z<>RKo2NPS=YLLRKhxFl*`Airxy_BhDePB-;gbdcRQ&72WG8SF3#KlwJiltWdV7X`# zKBy~#(k>IQQ@TJTKWJg{{S&tGOK4Fjj>%DnmV^c;h95+C#qwbLXd6CF2JqWJL>=Lb;aaP08w*=9P0EY$XqE84Sw}-dr?2EOG^@ceCq@eqq}ic z=Npl2@Gz=BaVeV2@rJB#L$Tmp06eWq#msxbq%pr5uhSX0cKTe{NZ-TW@$q<4_d44? zN)GCcnuvW<29%qt(yYGiVxyVDu1!ziWoIlR!)$&KQSMB9@>dOJx-TWO-a9eni5YCn z?^J9Z*(Uar%Y%Zm^`g7IBD(kY8Rq0tgeQAX;iRx$;uAjr2FMwaSNF2;zIrSZ^@QV< z{;S|5UkEDcBav#!iFK=#VeI2tu%E4lflU?=Ae{qJzddnN=3u^AMS=E4-ys*&zmvfH zB$6~aL99K~40Nm(z^Ox7*b=@252je-K%b~yr z*llPAUU5Zi%JwGkT&|4;K^C|~VF&rqS4g_FJW*1}?M1zmkhGU5a}iUu3#gBT%&805%==WCw(vkYn{`FrY6R z+};XY=bv|o^1u}=?b8poJh;+6VVySAZ9Ui+CouHy=p?a`hlgO|ZwdA(Pll}z8jHtr z?y;U+5qj>=Wzs`Uu)B1hxcTQyA@k_YzxQXN>#_tkYKjlLq>~G;FK>Ydzq3GkgbuQQ zOGLG!AF{`v*VqT32wf#5xzFGV@yq8$WW#P7@##a-_*15o9eksW4hBwa&Epv&Jo|(x zKlWg!ocqLo2JW?Yo}Ep8*7~p$YC%Mf2lCfPW%)fDSzfQ!faZ~=Sl5yWOTHPSuHhp5 zI%5%&QFp~j4k{QJ|5tP~R|#$IQ9O{@jOOmI8+@u9h_5IG9rca*l)Bxx-rbD5r?s## zs$KQbifK&2qyRSu<&cZ&15thi!@b}9@xthF_*0z19@-25tL^*5D^9Os!>1^an61({ z#WDsbNDt;`s*A{C^?xYmW{uyTEg%x%r&*fEZxU%Rk!&}qXI^vt!MW-g8T$G*Gq@fq zGX0)KmQ8xYI_}TF?GsMI&6_~R-%S)h?TQvP?#N}P4hL9Id=%P0PR7kSGI-(7dm;Cp zk1mH?@POeU7N!4>^>0@}#f%I#yE}+|?(enF9V&z6UnC*z%MbCw3Atp=(oN_RehMqk z0UO{g?1l|IvH6)k1PGbH@RtWg4@85}E%;@loa=mX!25UX^?V5)`_vgzB(-smY6!kO zyizEbH2Z4+h#o!Z+JABX{VbuD)vUR((2vp ze+-?6Bb8qm$0f6pifjoD6*9s-&yhqsX`mzt?KGqzDVrpcO|qJjkd%9#qn)gN36-={ zl7>>Eq5R%I;qu=5p7T84@8@%}XAQYB>=m;Q7dW*|668$Dck#paP>eP+X5Ws-<5tOd z_~4;|_FD>=-v(LKiBJ}Y=o!F-2kRkVygu|v*y9&f9rCtsHZJ|opNtk5Mn$s~@#0~B z5?>f8PXAMj$NCpyREIp2w2r~bN$c^;nw|K~eX>>W^7uY##`yg5S(5XP;u42IffXH&du)ry#hgpznwl>fzwZ}UrN3ikuA>;-8cXcL<3+m- zE|wb~7W$ORQ_y>VHMYO>z@UuB;_1sIK`gxxB(o-C(OW5!61E(tn;sA)j*wx-e7@*# zvmR!zoye^6MvJ3YXrlOT1e!biV;z*r zivcZ%za~$zMOblU7pu@-jj44)e^7H3v7R=9Eo{nxbH>SHcHIs4MkqjngepF`x)*bV z8T;?IZA{|OHuhw4iKd~)WZcuKVt zGl=wJ@(mu$Tc-fciwB5LOW(n1=jK3k#Y@)rqmx`0Entrl-m>xtd2yAo5upQ2ffZOm ziR4=}IN8T+kM2XKNsa9M@+3U1mVr;K7GvSUHg0|#8(&iCGKc$)RN={Tk$_|H9DaW|l`LzJr&YM`4!Hdiu7kHDh?8A>xme;R44PXFTFr$p z{(Ar$FtiBr54FMGmQn~kmJ7{)gjwpl72rO(9OM4;z#YA-(f!I2d~{(EoI5K`+e}YF zt{G)L68-R1{{-UH<>+p@p6U!e1`G7v_^COAsm}vbdg+BbH#-~$tNea~@h*Gl5OVMx z9;4`$p^7wqpCPR#luDir5FBP}$mQcBv2N8ZG~GB+>|O22r=1Mv+iUM*Q_V^mb5x3c zENzD)(P5P4h2Yh@rR+|37PM_srN=%y&}r8W(#9SmK1g{WZ?ZGsOSBJ@-9j#PgV3eQ zd^)GRcUua*mJ>*GGLL|2Z4a0%m4|gJ8z61?1M$nNOIcRZRGPme3Ai_%nq=pZI zN5==#m32brny>-liS*#nXVCl12)->8 zQMn<~bpO;BAZyxBblgsXmhLQS{=yeO-t?jmlH_3AdO5If{Z88Q-&5t$axlI(8~&43 zq=&3(VD$cU@R%}?9vm(vx;NM3%^gAx_NO0yg5#JsvjJZ3OhmbpvuI3JB0c*r3|(sr z!SLfIdOxQEorY8r%ct|{xUGJ;;hhJhc~wHWPa6Hte;!?Vi-X0KN%X0#Wm)u?FR1mi z1!Db#zVZ7jbmRCtEMj3U929!&Q+?gbpSaI~7xffw+`I*sb`RiQoz8H<$rrq~2>ceA zei(LFhPw(qMc-}OV%?$>Ab#)&+TXg-#hsNnb;EQzSD~89_!NQWiYVCGrq6dNccYt$ zH<_=vi%Or~0$K7haC@#K{L457UY&!fYS|g^7{3?ej|=Cd71GrBTNoC(`$DkbtBxL) z39irbAk9XH-Vf7Z+sT{wyVh=8SISyv; zF@br@J<+zxk~!RGE1I+O50@qFyM*#nt9nO2=5 zH(F_`MvpF9Ch#ZR=!wWg_RIPjddnwZ>sN}W8;-ICQkHy~**`MO__HwQ=Ws@A7+1Ev z$Sx#jfY8_Gs|*xiV(@kzc`Y1&^%$Y6RVzO5kl;^JRQTCthGf%fDQ+kDo}TU4E_8l{ zoVr{pZrW?iizjU7kL9y*y_p_KTrS0}26tLG;3Zb9_6dhS90~`kb$Ry64jkE4L-HJj ze08%FzOS-j-sYWb#ed_8XI~aw6aQmkY#Qp1l zjyo@4(bFh&ju=bZTSl;F7Lm9}=8`xgHVdWiEdjMxV|j>kC|ntp4ZGi{!#NcXRFRK? zgInH%@8nWuzPA#T6slpPaVGTBQ7E_RHGZ|z<6%)fY;@HioIGp7-4q5KOWua zi#LWTm4Es$1sqh`Kt=r|l!i<2KK04`_VFAJg?ss+n{V)Hp$FFM?G|PUA$Z4U59`mJ zEbxX3&}mmMMm?6q;h&^2EJ=9;{8!M@Mzl( z%$t@=Qq0bw?7>W2vNxY>6=n&y`wy?6=j-h54bRcMR4q3l-~#M)57K^}=`W z0486o!L^bUSA}iGn>B@G^Vlb1i4lpoukaApDqhd?9(bVT^^Le$W;WOQ_m=&u`h~88 z+VQT*Gd7}n2vgf<3LhV-5}QRP{GZ!fAx9|BKkhjtzBbSoW~oQxfJ@^rpr{9fYNX)x z9!GLl=mun{&0?R=OQ6}Xa@^PLh!?-d5!EZ#m`%Mqe|h~3FSO|f|Ln;$!)G<^#-Z?K z7RR#-Iz-z&_QE)?P^`1;#n%eS+`DTbTwNDJ7j+MT9Y35Zc8e95_~;2f>De^#&!!3d z!Hz*_F6^@U?QQspFGtaN=+cU!FIlwWaSor5@eEGJ8B*ubZKBpOli;?EHO`xG0M^)s zRrLMxA>D^!`H=9j{KlNAbo zymB(dqaTiIo+n3sx(d(2!U|703I0{jntoWi0FE_YV|sJ{qNZ#rPKMJ=}EAudK{f6WS(T(kI)+WPpq9D!kZgu#f7wW;vYT5;-u|6EBqUz@%Do@ zY}J;ph?h{U7(Sr5Uy3X!teR|KsIxwY2 zgv^+^@sx@usipk);`49@?(-TKJMJ~-57t?4rir~<@cZ$3=;gnP`q#-@yNuN4E~hqC zs7pdg+kXh6TeN4rcYQ2&%Qnm(h4TiM>iSU<1&f5 zufBl+67S*ERd-r=W*R@=@Q9mruECjo(G)bQ`N3`rzO7FOZvMTS=)c&tR>%V^ILn&{+e5aTCmy%lQ&CoS1*SS=@y(kfXqlfBZ_w?A zq>3)OBCJU~@VO1G|MNy*4i=-ieu?EbYxj!wW;;I4XD+w5*vn7+ro48JFaOl*$hNpl zLCx2R#6o+I_BUgyVlH^6Hh&gah*3Q3;b2;`^-#r2y^DBfdlN3S zx8=D`VR-0TxaeK}0hlEj3Lg3^x!GJfzAQ?OskP6f>Bs-_JJpY|Azs^h+PP2;{StKc zQa}DX%e!J~>0#KdSc2TX!s zV0JIr^8P4Mw>gXApgnjuJ&p{r&S0M=`~vU2?(l0`6i$Bi7DlECzflts;LZm#G}Dpd z#}d$Sq$!X0Yl%d6xny+BSqEC4?Fi%5390&&v+inl5$<=e~XV%5%^4ORI6f>#r3ld+-*d zGd@BOyGSObr#3c+uK!e30}BqkgrsYv}`NrObIc(>#2&uGZ zV=D5nXcDwLjlQxiWJQ^k7$-3T4G(>WL+eHZmP^nEV@;~GFcmiZ{38}6+k)csi|qJ} z{k-C0gDt;d|!f2w#{KdWb!ju)@Ez1YasSas8ly zG}YFeniohxfzASY%xnN(Y_XLnXQV*l77>?oNuoy=TUdP-`Z3kpOSnRDBY17!0#TV_ z(dD8TvTCd$^$Y7FyQ;PDW6aSXd8z6cXat8mVmo2+oU4LIgmG0~57rr#b-wp^(q#36{h z*IC4FXq!>-<1=KPr5+}S4ucS{G}tdS1EyryvR%oRIR4WgSa^8~y7$aS*MFX*MCFjh z2am;YD8+=WPAegpGu;6_Y~C8{RLA;A1HtH)fVHXJSo4D z@`v@Fac7HXEywUhkJ!NXpIE!lL7nq{0JpIx+@{t>JVfATjURM{xUKCWDpCbZsyiM} zUC4ydv9qvkMU^<_%ORW<^%k%HM=&?b397>*=*k&~;g5kV)!7m`wXaS+h=8z z7mWJ^w_k~B3^>_H;mkk#;J~8qaBaj*h{-q$o=>vDZ3E!#%y^uyHRi=E`dwo-V0M-npc z4&XiuWT=yUF}hrP%EYV^etL+({kSo&@P1A1u6)jX=8fZ3@|*e4nRa{v*+<2N(dae> zc(Saz$Yw$;FSNVHm3o&k`@dnZVeJ)qAuyM=zcJ-IH}~>w7Kfo`?PAUkck=cXs{Dk@ zEB^h;K`!O=jX#U>!tuRd(RZsH_p6@A|IGHJ-*OJ(xjT4+sd3|Tl_Er+K`Lv>{Spk{Ncw#D?p= zsO36$Pw^`+Q+V{F^QhZ@hd(t{;3+08=%<}XZ#}q2JHnPud#;apDIk%{J~?wD}5H*=`O{WtK|ax5&{ zX#g*WDp1A~d4iOc;3^(aVfgVQU$s4p2cJ8|ho=rf-9mic6aeQQd4pxVC;cKH8yw^*WZ#(xJJ;V!Xqq8R6 zeC-t%ztgDb&oF1v>2h4IMMS6~Vs&;ueo5X4gWBZ5HB6e1^p%HBsjYOTaRHxn^cEb| z5?oLB4pWKo3;0c|J20}uhU<)x!GV{SGG3Ixd%qv#D-8GYFzbh0ZuEA(=eIIHIAj&i z`QU}QHT(HH*=78}f1hFH9%C-<=n&7^@r+01p5;1)({ON^9b5N391bU^@W^)(75~l4 z6By{{xkQ`}KP;=j?&b`su=O~@8~Yyc^hG=Q&+KU4Wqz7B);BWAQ9-LM5+Gm6j1JAvZcV!96wQI89{7&L1Y zRkV+$-zD8?*U_W=QPF*N&;2GTcTB|{rzx{kL-0*@M zrlw=vMOWPN_A@4$`O=@;DgWT`{|O#@;M2+b79;=g$C# zv*)Slq-xN(DPukE@DO@&P68=j9Ee`GMdZUcJ!&U16!PQ`;PgI4zN=|1UGaT7?9dIM z2G6sxRed|tzV??Das}*9WIWn$8Q(KDPgI~7K}Staf$}I1TG{xHJbSANnzL>|p!PwG zO8W=v#@NvLMQcDmVLZKgYz%)XZ$aieK4Q2s6;sVl!Ojl`{PB+ndTVbuTPd1JZ@6gj z6{6KJM(7j_35sH6x02{_cQ4wsa5t?mm?N;;ucPtJc=Bk+W>Kp`B~zQdp5Ji01#9Ie zgW=Dwu;z*pUvYLTdbPeI847N_af(HVH zd0ngmFSoR!t1P}jn&NN#T6h4e20q4@H)O>A{{kT4Xg*#_Fy!(suhIP3VK{Zknr`;8 z;T2zRlC1lZH1fNU&*~aSUsWma?kCA`v3D@9oG9dSQO{u$7cKo_;L4RxSlNHkF5vro!27Z{%jwN z{Njq^?qrjDHqN+Nu@-(9wz6qsXVHw7!}LI=8tp0l4U3HAgLIS-%+F;cF%dmdaLe$X+W=+R0V)txke!ngaR?dhM*!ex|bHzpY zw_XfKA9&MEe)m9G;{qFWUxJq2UJ3IjYoe~nYW^rh$fB-@fxwq}7<+ao_3VkES-S%; zqR=1g0;TBUG2Qrb#7{Q$i@+pyaHLhE1&Hl?DSF#tEcB?w^OjvI>|5D;qCNj1evdPQ z9cvH5f8p1y-fptSVTZECWj}799X&$ZcgcXyW`EeM7fTlz$WXPT_u!;m9(e8bV|`Zl z+4x&Zwi!YMyhcBT|=phZyZXg4V#-riY+xWKCkR@Lz z!N|F?(CO-kHYaxD!eOCw)9FkI6tcj#TOL{^O)bCz`I~G`!FNzNsY%zTWWlvC4XB(E z2n83`kmpJh=+v?I%C(dd>8Nc_@Qb<|?Q<*y6~*BI+x_S=Pr*BIcomas*1LStF_@LE zT119+Bx5z`)0*c8#kXaYua{yeDms1poW(hfu%s0GyNe zrkV~iJm5zZ40E4=H_gw%mcO>tRCEt)U!29$QMz4=kiIHSL4%{DOfrsmlgVGbJa`@o=?r_qQ!@BoXjK`e~cq9+zqM4M~O!8OrG{d zjvJKR#O*AKE?9R34r(psCKbK-bVW43(b$5p9qIId|7$lUn~d-d|gr5&f}ok{n>eZA0k)Nnx)2Pr=5UkzmZ*$Ai9 z?XW7ug$K^jS0&uT)? z7CwK2rr>Fdq#hB&Xzu+uKn{meEk7Dg3 zV)%S53L2Ge6HUjhd`ELJM#TJJTZ0nu!u zKWXN}k=7$U6)_`sBHB3~63^ULLL*LE()y@i`c1n78pCVImOKq+7-5a>CXa-f-%x($ zL?#-%P!!PCH$k^23!eN|W;vG!;p)xCeC0b|*wg00_YAb)`8A8FS;l49IMoch5BchBFOYlw)D#dBDn(@@dMB4wp7GDI%VPT~dS6FAv8=|L`XGLk_II#-%E&oj# ze^%m+Np4&(dMy4*d?5ODS#SnN0lx9uM5UAO3tp)bD3e-8*BwY9IYA1dhUIJVduTkX z%u(jmo`=byhZXR`vIQNVo8aO*@>I%A=o06Sz#l6XlgB{>@@s6VO3QyD4=a6Kj*|4r zssz5j{V@NPn~%df>S6YXb@ZoOH$BtdL~qp8qVIeKF1t&br^Gw*7m}0tt2y;p_~(Eq z@=zSUb2x{<;op!b~w-!|VeV|g#oS!#x!J+wenDV+GyS6m2l=uVmT}m|FEDE9L zHaX+pFh?4@;4nN$sxhlILH5_QZ?V7{Dz6{-W_ z$*Kyn%XuZ}ncXDzZw}y#xI5yONI>$y8gN?QXRQDnE#*=N@FSk1~3%&nG|Gf1)FLH^li1 zn_1!_8*r=Bg9oGH;kD#>^how#K4+xR$Ksw?yJrBN_LqYOUp3I!dK#Wjxdh@YAM#Cj zzLzZ&T+Z=&EdFuWlkQ?H&>H>Oa~Yd9*-JdyI2^xUIx4dM?;{vII{>i-i!rH3105K;Y04M;<{n6EMVMtHeW>_a{Plu-ug^jA~M4R>T+zY?+fBEcoNoi*N|gs6QMg& z5=7*#IATp3xzVLZJUlbR4bMV}nr)?Mm)l6lu3jW+2|NfY{X%~~WHM{X&A~#C$!zL~ zdhr0i$HJmv7>v$v1?`i%c-3+xm<%(8m6g93@69D=URs0cq*w5mi^W$GhLQaH{unTA zFH1bK1V-Ny_OZ@MEa}a1aNpk!chA0pe3K9|YfwByh8_be_jzRbbY~oHE(dF@*OK#p zLq+rb?8x~uABfS35opwwL~5@LqGc1!#2L|-;7zj?XbfBn?|!BeQ@R1t(x;R1yI06# z*%Rb-{av!#XcZ(6J4B`}w#ItdlWbv+t$1lpHs-9jOFnA4q1EueXj+^`^cBC5j==^b zaiSiXtv{Vf<#u3_R4B+?s24x&Yh@u*&XbMO#;AVk2YH=7f*i`RKs9AI?ECykY-627 zetX-)=FoidO*vffDg?sz^3`~>ynr#)~+RUE8OA8{2Qng(~9xuEnxM3$eI&d zSW#ZG*vo1>sC$nlvxKfsjZQln#1DsZg$DE}ybC)9Npa1%C#aBDB?=Y-Kt`wc7=T=k4?vYXVM;Np+0#|3 z(09Y3{PX5rIKO`mgqUxK&8CmZ;vJQ+F*cPM>v1-9`Uj@hwHs91i^z+r2l!}_kZToZ z5tp$C(Jnw>SY5Ma4@Y#cV+zBl_kX42O~^lT-TWij*e1>1Qxj-tiY6yt>5w~q(lFz~ zZx+633N{tSuvypTm|Q>_Y8`aO{ejP5;P*o2*r~vnwl0(UoeiSuD6;s$XBPkH88|$N z#@v|`Az|=)a?Lyoj!#?7Qe^_jbehL*cg}{Dshafj%QE!O{$1{{a5cZb^blyQZ(*Oe zOr*-G5n{8lUJ?^fLEQN%misLpKOISj?&@_=dsZ4>y&FJ%u4j>t{YPM8K!}jdkHy{L zFIc;8E$f!CpyRitikCQ@gP?I%Xxw(1P0!5~IJ|ZiDd$1{2%A zAs}}#9Hf+c;hLnpP?q!tjd!}@!TJT{N~088TOGzu`=^RbyfSfOY#wNQb>N3?C0RX) zy8-QR8I;Wyz}RnAboWYsc=k4ebeUPh&)Dy<^{*v=YyTC4@icp$+yoBus@TfSTZoCy z9CGANlqfH4KABf(1_Q;9S;V6zcGT-099Nlud#4{E1!Dr>zrd@cSzr-|omaVh&{3bR z_-KMI5%yO159Gu4<>fHy=T-QXK7jO?jzP2AS+GazAoG4PA6NV{6h)NG!T>W}<|=>1 z%H!!M@ynI3v2}h8>`vAY@ahVD&MQkOdJ+QjZichgGhfNp`&04AOIu{ zKSvm;?1`(u9*!@#P5!FoUH(=-38F5Iqv@B1K*B^DI`2v-N_##c7njE2-TFH4pRNv0 ze|A9QkW}{VP&!I!M6uF{8j-{<4QO7p336|!GjpOvO4kmD$*q}qeCiN-T6lg653Il; z6?#nhlp8n=TL=0VUcuBQIpB9xlJBBnuzQ*-`gc!;a@TqIW|=R@$6aCdc{|~qR6e=j zuZ>R9DsXk%aq`M}A1K?;6qvOcc;MK4xYGQEtREZ!$9qTedpG9N*qk>|`oRd#@Af4j zx3oaDa4N)K83NA|YhZ}hS$LXu17|v_V4__Fnw%6om{s}kPM(8_f2e5Jjc%BdlaAUc z!{Be82qhE*4yIu)N#>=fl&X(A?XK`BSD5m^(>)H#X??czdBvNbTZxIy@pnkd!nOixpe2sI=V-_3KEM-aGb;= z7`)$wsyG_a@XlK-c!NCETK^Dub{O6DpC?tkB0=|fKZgNHF4R6lsN*QuqCxd|di0M6 zy|8vG9nE*r8-pHRIyMVv>bw2)*o?pM>dYnD-}DFEP78Nj$D3ei-VFXJzac5S8XmrH z1m?VhzMbnt8*vC-rZ|-@T%%6C9+%U-C*0@+=V+RPkICm~2^w`y0y;~isjj^zO+2@c zp3cjIadUO}k{S={J@g$8kk+IARmpVDNKfka<33a@Yk<y8pY%TX{a6m7t`Fx{ft7HfVgR+>C^(hBHo&3|8{D<-6=*H#U|Z^( z!O>zgU6^74x9xV&#HeGiFMSfN2|P-D>Qkvqatr)6m7q!=GSOW*j!yDbrodmqjm`kt z)$R`sIV-47P!<@jEg_9ghv@0A)^w-$1)4c8iAL_Jr_pEU&`kl`XugxIb?$Xn+H7?m z;xzAp*12Bz+HXkZ{04xXVLBW=YYWBADzx2oHKxt9rWM&zAl;uwTuYa-FJm^+rkRte zv&S=1U>U@JrUcUxnbBn2gbiSxy9dr~Z()~}X3&I_A>~I(LO?x8#C^R~hxDsy2GuwfO9N+>Qj?AI=te1R z8e1QWjkAJjxZvXQo~kUivG+ym2n$*|aTl#?`b^V>{?hlA2Pl`dq;dxn=vPApYErCA zua5GiB|opg>eVq+t>X;+VR@5AWHf*{b_;d-89?BJayPDP0I_% zTTGs#O|#b|fZX`&P}VA<#WBt_L01>fR(n&Xyi4eD_kd;EDtg3o8uhPz3Z`SmK=CvQ znzBv!?z(QF&M$6&;-V_j92$-N=VCzS#vJ-~ggh81^1t!v^=2=w2CX7}_goCJ+(#(|~=}CVIA8$>ki|@^+`+}9J z%I1k8e(x-~wL*t_|5BrW)YZXx!2s&e^qt6MOsB8-2x_$`oc_#@qWRm-h(2m%gUadU z)aU47%<-|JP8aV%&Ab@8dVnMS5+w%DXZqBtUX~p$H5a%o)>LdHuy{hV>4M^1s`F2d zdKliPDz96a+Qk(3y~K#V|LKA!3Z_yNasYSkJ%sXz*)%#b4=vwyU+oI&+(24S|}A9CSy2C4H9c3o$(F(Y^~ZVh|K)i=(?J} ztzO0hvp0gTemtIj`<2*ERw*C7QG|a-d=nWAQlm!Nx_nir6NZIdWHlcKAu*_Cdv?DQ zCna_=ixHAArLKjk59&k!{CXS)9f%_g&$wtrGW&nabBD--7X@@iN zkClfOBQs(+F)v>mqrK zJvs?2Eq{yR*LZWO-_c~Vu?!xS;ABO3EUG2cvVrUUF|=8hwFIq#pBrED!Fwlivp?sU z?A{Zgp5wqj#hUQ3FD-d=h7Rb@ljq*WP0XouCQk{~6qe-I@U^D`t{wM;G)=#QVV4A- zwUr};Z@-Rnl(YH00dGW>I=7g^NF8h$P>-3jr@_#t$>?~u0zVF_MuVHzFf7lX$9~Dh zH8B~4jCA0VrMK9owdy=vD+ih6;sTc2^a|{&^-WAD77l{|Y%)_r(|eUnSkgdD=DD z_dAR!D+}+C+Ht(E@dcX|8N=W9W`HOmPW<2ZJ-Ba%JH~1J!m5lYEZXr9wjUKb)Yj@) zYi@u`X9?%};wn5PPv)1BU8v64yuz{+bHPH?I~x^OIZ>lRpZM z%TDHsF9L=0a||bcf{6e6@2nuRh)o`QjQQTw=iBnaF!HJ$%G8a+%HRrSsr46S4cgf! zIY%xvM*;$Bgno71HJ0>29iK+q!P3`v@x7-TuJKvVrqpZl6&CeOZoqx~btfOT4pHUr zCcI7fHPrKa4&3f9qLWk~QUe}G zU3LV~`k%Alsns!FGU<@`kKpx^7#vmpe#;T+ZqOnsc9=zP+U({{17}0!Vk5r#^l=EB zfmABAirOs}Q&MG5Wk>nYk5hsnqi`R+?R<=;S4PpE>ib~U{E3(!*+NgLcfiL@a;PPJ zj9$NINyo%)rv)|(V9=z$^g#AXswc08$Lh{N$<+~}-&4~d@~13atu&i%6D8sG>_oKa zD}c4RJ?Jvz2iZSfa0(4O2^-e{)vxV_QTB_NuXPlxI-Wth0@6tE>N5JQ(utby6r2i1 z?|}}Jrc28X(cxaPbnifK8u&$*_6u2sO?R(D1{1^X_j568kTb9;$H7D55PkkhiVm3i z5Bwqpub#O*GP9xl`(Ow3eXPz`|JX`@-Jb#03p-iSK}#rpdJ^irok4c?R2a1bp~N)~ za+71|5;lrTeQkxO8#_Q(@hE5(+y{$g)99IcKiE62S40|av)jV?s``K?)4cu)x)ctQ zh9X(IAou}!<6bYE`?dMpumrZ`n2zR8!+Io%yzp#5!y$uQ(n972Z*_MrjgtG&o65 ze|$x@ED8s%{2xpUx&ohvEk>ik*5q651A1~xE?vI-wb-(JAMN$*gXQy|!l3mtbv#d<6nQnjVOWS&`(n;}J z^iy#rHBK5pdyhxZ0qnTw)6T)x3w-y`2_uiuc&9UTLH-$7J1Us|d6-MbZ?&aW$L7*? zTHZ7*yhAK^F$Z7wkKmTeQ*gTBC3^COE3NFFPIaAX=p#MBSvWUZ*xMW-Is(IM){s;j z-F}8{HC&F*EGO`(N*73#btJEp(E^PJL+J0yZIt^w5+IR*^xn9B`X*pBo%4Jj9ep~5 zUe_>y5`n*H+NDF^RSLO~bPJSxDn~7nQy^t=EKMyvN(T$fVcV&(bkK1pTDZuF=3a@Q zj)%L*VgE67{oWX?-8z_BWNXmYkEf}zX)OKvy@u8f4W;WmE`#N^#k9#aw|w+bLwf6i z4I0uLq{Xrfa+~5QU3wl`U0j9kUND61pG&6=A0VC)7XZamU72FLxj!JEWakipQObUu z-o%EhMhNbQr=YN&(rnk!=u>FIj}4XQf7;?NDHU;uE_sKYv$de=#4(cmq?y+=iMi3V zU~Vs-jrm>u@Y*yNL*tD25~(ZV6BZhH<&E&0Q)5nBYR*E3%LTZ8+z%4QYvEI&gPP`O z&3`I1!1yicc<$w0{I2qW*tPzoGX^_TF3ioH9vy*8L>c9TOiEkT7JeeA8WRV4;*{0S zd{pswJa=O)w+Z?QtLAKkh;a&R-OT_n|GFC=q{{Q;)w>1uNxG0HSLPe-wZY?+zxY!1 zT9&b9GpsWkfNAj#d`!}P(4K#X^lW+!f6S$!cLSl9Eu-*r-ALTy=En1L2XoxN8#fMm z%c2&v!>SHWa_+}MU*bciIc@}xlPlzdt#7c|J>974R1XeEdRU_92+rPrANEXIDlim> z<8c>7E_Zt=N*~B28acml#H3zyahZU@E)AgTwwAU|J3?As3b~U$pf!apWRc-){%P3; z>T=K>x(~O&mkaN(Wn(|uJA3iMGeUo+I|}7{*Fj2K39J#iQdSE_vbL$l+`Dfg-Z5Rx zyMK-5=l5IFE|)yE@o^)aV)PSsdAY|YK5uIn_j5ah-(s@ao29{`&hwctNKOJAM|%+QLn|v8 z`*}ERw=(S=e2^9;eW6=Lk#s`aHdxu^5A6mvpsxH;ymk0h^o?RRQHob%w zo0gKD!CJg1i_ksmb>M(}0Pa@`0-2Le;;MO@uzcKL6pfsWwelgfc5NGTZC?bNH1#3EilspTIZVx5F2c zUGb4@Ix+u#iTBEG;;A0PIaM0Y<6p&)UrU>@Eh!oEFU=vpBdXY&f%nJ+v#*ez9S233 z^Qh|bK)QL+Sb9}849d-onAux-+_i?#c|Xe_Ht`bK8{xtV%C&gRlt!x)E{D)KEP&u3 zRs7PB2gX)KD820>tR9jD7xgc&n`L9f2f8jH(Y(#J^l5RMl19;^!|9L9p}6^XH>m85AtUD7 ziQHx+&;i13M>g&xAC~mNswt|DmM2_5^XPZzCx01j8}8s}M@6pfyqFawS&1F*o6>(z zZ_?G=m^&-T(2u$k`FQt@*!;W$=LWRn^K>CsVO0$+53no?l)x z2kT2z*;q{@mL<1|KeHdqHEeJ5M4M6YOlmn*OVgxTr#I7*4`aA)%X}UrCG=|k9N;@P zv_kdoaBBUi4Av$Np_ZGCN%1NTDrccV)5dD?xO(Je4)f`LtGh%wbOK%ROz1QFOrboV z@lLlYI_zi}&mA0%kN*8e@^u0qEO~|0Z*%55-uclwft8VAa3AJ%b>mmP6^A^d$IPw>B&W9i6z89G$5gqE%z26{%9;Pkbf zut4?=yt=rbUa3%{bvmy|U|Kvi5{Gc*!a4MkZxY=x{}7lg`M`@tXz>}IEBLTJWu8}% zEw)*F3eINd@YBn3ami9+*dU7LgXf62m4q`JLDO;icNe~=B7o}bF~M_&l3)-brunxn z(n~*MaL5>>aQ!}_{B_8)Un|yX%*D3s!}NB>Ur;UjZJA$X%*PZu@skf~P_$|tv$u7; z)YK78d)r;HRaOJ)Hp*J}UW;LqM+Hz@-NAH3zdyhDBbiI890Tb~Iw*DO7jAqIE;hPY zg3Ur^?ydVd{;6ae4zN_EnG}t0|=VxJjY{YrmVCF`LXW9uFfnq#9^dxFpmf{-zK0`B?DP4s9hIUM4Mk~1$z6$Fug~E6v z6>g|Qf{paNge_6q$*=BuvfTGByU^=H?#VWBQ|?GJZu)yTBh_yobvd8Tmd=9pH+lv6 z<`-erX^teBOG^;8 z9)h9CHw7KTSE1i>HnGio4h>z0xo0Qj@WzX;xcOiL!@c0IvE@kg5B-u;5o-#*nRT^8M-hCDxLDY zFKi?6`|<$SKK)Hq4`{$^*K43>K7u$ceasEGpCL2uZY0Tk7KdSB4QV)IOJu+7AXODg z#KUYNJuW(#7JeU~Me>C(c2^__ulxcljcod8_#%v4<^n%P^Et>JH&H_D7Ag5y0@j({ zu=ChP@A~~9U$*#b1Z$8(%?G;RmlP9TR zONi}^8uHI+2hj>}WzEis?9l2~Jes%?KAAZ%iCYiQbCC<&c9o+Q5(lXMkrW!=W=!fe zQb^s4Txgo9Np`F}iXPs)bK=Hr;hpA#{5&$3{%RSb#y_5b^q+LDNPIpFMNcF8tNH&= zG7nt7WrIrF6H*q4)aPCz_fg%NEPL)KIKa;u#@76!_nkk$hGqKXakU&&?U$I_-xVjH6r*|rr$WN0v+yy`B(V@0vpF;*62YY*V7)wNKx>jt>4 z90DJ{{`nxKMqC>X!Nag2xW;=vSFcncJHl6xt^qIhbG`+1J~~4NO$W!6HD3YAJhvzFmLD_@JcbXgRp6?548;516LhHu z__Kx$M8rULjSY_KUJL?{SW?@#jlMdVhIWeyS|!PV!MuWWmXW zc(*vlRqN ztYU^Y&+XcZPK$JrnTNAe(dlryIssc+lZaZ#HYi-A!AjnYqCdsnp=wADT|Uo;$#e#y ziLEJ7zF~)VB0Q+yn|k)wW<3i?Ji=zSUd7$#1?&k|i6f^j$EpWOJn1Sa^+Hv$YjXx~Gz8UR-u*^F zdH(`n*KsK(y|4r~?D&sbHf^CByB30EavFvtHgiss>rq2r3GR1`vaa1}Y{~d*@SoE+ zI(}g%{u^$_k#{ATk$5}2J2IAxp|2rnYC2}jG$STe5*VSTf`>o1q5UR1P-;>kxrxWX zEtjy5b@FJr!-QSUy@&U+#`4_Ri7Y7}$lCL{-1|qTpl|(ru5Ic?VbQar5J+MxT8BBuZ2_91;dk(kBOVrm?AY>SXgcKC2|V z8k39lLDIm6EO>8?x}SETi?b(N*tr^>Ww+44OH_DF$q3%$rwYD*RjrV7OJlxjiL7tz zG%iKq3Z5yMgodI%BsO21_^y6T-*S>ryN}OQ`DjcNz9q1)*;85O=^ae%m;%f7`3alH z&Sq1zJmEpiE%;I-O``Vab6y zMic9~Mr5PQA^I)doQ-S_z+N>E{AziZ8qLkdvj$U9SNt=s|FHtQcPYRfi|c&m{c+*H zsl)JkX#!^#Z-AldA`n!)zdZf-e_ZyrW5gxZ5Y|cWg;NJLNX@T7PHguk5;Arix9XEJ zO(9X7e%WuxcJX7P0VAMu$8zd3v6bhc>O<~{R-W1APvecUAhNU+GMi0V`(86@rErQ# zdp^ORvwy?us7Tu7lg6#9X`o?A`%vC%A`w}XBN!9g2p4K?KvMK0@6&w;^MsyQDflL= zzV3#-T^^k5(-i#kY93hZek=TFdYN;IYC)3w83%4ku)9ApV7J{n{5Lm%j*^=OT?J3r zY|Gs^@bny=o|r=2*WJUTDns~jC;|P}*s!-#uj8As`-S3i!FYRDD&#wO!?+QVL_6ym z+-cs4pH%d)C-DTEvC4rxtFmB=TfTwsk)LS%UkOGH+wgwP&ES`Ox!mg=zb0#3v2C9% z%zPJy-xlnGh<_PiGESV_&XHvvLtjzk{!PwI-j=%UUCgDw_r-YI`|wdK4n^EEpm%~S zDY-bE2#bc%!@d=N3X5^E!*m*RlVe+AHOZ5Yl{B?>IiB(QWxsMYa-o+^$k8na=o>k4 z$V_*Iq{kK{S$hO(T`?j@{U&i2_Z6|h;tIK_S8p%P@gl{K3j!EblM7Z`@_jZK2xEe_X@}6^Za{*Somw=1+HpkP>?gm z;rPeVWWegGaKct8q8=BBH3nruJ;y~DwCp*M-4F15MwDR3-p3G8Uks+QxzK%bFK4TH z6ApX}6c`tYk%DpA0F%y;*Svn)(Y1~||7JwgycUpekFD97`c)9uEK43P{|jb+CxH6q zBoe1Mj=b7+TBu@!Mr)bMs)!j4y+a^${#}42dc#PgFD#pjCHN|H(|V0 zC@KBXL@Z~mXR4heh*ggP*}Ef#>^%|4ol&pE2S4ga+JZ!qb+LoaC_G#Vwfs93yjf2Yei*Rx!Md<#oRH*4 zt8f?lPr||=B~s0=MT(Xh=;ihS55Fed4RNMtKLFxm&vNGkdhqmeFbO#|lFU_}Lt@uh zkfZ;?$&LILd~Tr1#wcyUj-8@}lf4dR+oqABk@Lx{@LG~{(R1g+Jz+KMvJlKw|+r%SS-1GY&5y$k_wIDn~@8Af;#DEaa67| z2pR*xV0$dgUeiTK4>7Qs>Q5dk@E*I}5k$qglD=PAh85w9xi`XCob7szxZX}A&A&a^ zu+DMRJRL{v)`gJYF4wuQSH9uXGC7Cu=gQcrhjPR~b-!@apUZ4tVKl>uJbJf#9>*OA90BY2Stm|P%dI4Ja}a)%UXJ# z^D{WhE@)jw@A3NVy-P9+$>DP_hexu~P-h}B(jG6%=wee`Fx#3n3ksA6aBcE;0Eq}< z?<~WX9;Nj5S(yq2ZJu#2bb+YY2N<*DC0Ee15&kYHrkZD+SZTNx6XdL5_D>t>o}n)0 zaaF5ge#A^xlKYc!8{1i2(1fwoZ2k*U$X@JD>;dMn}3VinkzHei?dkc*st8Wjw0+`YCuxAonaR_B9`6*x{Z7Oh5Yx9T>bt4=B!KSN7dzsloP4g9uTu$dFB&l+7--+cWh@ zp2;^e7hgVE4A-XcJj=a1m_FAkh`z7RO28ux@!_TV@(ANVg1b0_@Jbmi4E1UPH`>P zHN_U=t9VWM%xoN|@DS@}z5=7ArL=Bu8=a$Zls4#9H7M6&! z3mX#I4A)ES)Sp(A+3yPRoAg<5d=2;Ex+&8vPDEd+n|QaJ_cEOg#K_%s7_wHLJ&%^I zxV|BuhV~nCs^5%R=wt`BD7~C6KB-Pl%ueS0H6kq7X*YZLZ4G&nmttaZe|Hx6)@9$690mkksVi;;NSGO z_;kfK@^9CA(C^v8o@!5J=k4EeUNH_VaG;Ha+}X!IYCpuaGK{URX#r)evFs&(Pt^If zo&}zgW>2TTL}9cBYrbU9PR-xR_hQx9ixovISLr8ClDWphr#{A2zc;cx2VZt)Um~1Y zbq0MuMv^SEXjb-R6|s@kq1P_F=B^m^WA+gf;VXY9R=htPZl~SE%^`D`*XdJuLB$Un z*NC!zAB&hJzmF4~C$k+Lo1k0lIA2E>vbpn4vCuR2715gm*_}8Aa>Zy3v)rCrZMLnZ4I!6PmJN zCLZVBZS+9@<4P=4_8uo!egRN$N8n!m7k>U(z_pvcLFM&JnNJhX8QNRIwexElnjirU zV*}YxTrcDr4&tUYlS$yoAZ8=ZYlx3Iv;5in@q%X;Yg`b{Hq`epuMZ!X{S{TTvy5lb z7xu#p?P9Ptb_6f?V`Su*gLeCzCzAQOrPRPmk4skS;c6XBK({Xqa-7OJrE(pIO0#3c za?=blUGoq9HBk%-zX9aESLU`g@VvoCn;;|KQaJOT8AOXL6c{8fCp_Yj7^@s6+vf~I z9$!oDd+Gs=j`^@S?I(9E<7@ee2L*WVL@wv|pAMCrcbOQ}HIm`Od&v1u)5(r>zWy#y zC0}Q!Ql*+uPDb4t?ti)nZ~14EICqWv#Pe>g+akemnks0t`EuGm9QhFH4&KgTur~W5 z{nWD%>Z8R;n(901^FxX2%HeZpcQ(M`o@Bw?DLe<_l{^W_XHdKFDf#$RihM}8O+0xY z=F(XMQ1ZosMDH*L-+nXtBjFT$iC#?n`sa`<3+;rxvtB?--8>TEs6n#dw3FA%OdX;^ z_Yw)q2Qc_co473(C7avNz{A=+YQM7(a!!hIy4mp%Hs1%{PB{va&mY3`xBxP@ZkWCt zrwubzd?71p8FAenL%K{yksxJxhev6P9gMz)liAIxWcxWacyszNP^>1;!Zwk+0s|s4 z*gP?!&CJ`{`KLl4p@4~2+zu}c*6w!;;BU4L;V9|jx zvUZU_*wH`mM}Tk8$Ut8s1IiCIdpW10+3E$D^~ffvZ2auNxfr%m9-5xCx$ zM)fyNC054{k+ZyBFZa$R7`>X$o$`AMMLPLhKY`e>=F~dwT<|k!I3FcgbwPvcIbRHguAAVk zwK}B#aksbEzsybd6(RNwWninhjw@W5Nab|C;`n*r>6?yRnEFYZ1SnI8415S`k9k&k zaUe_%(ua$W5VZ<2eDO8xb>9iu+XSF1wv{w` z^5+YmG1c9Ckxcp`>TrsskQY(`WJJze^4e?;8C`Y^Zn^qH%z!(@gp7p)t+T0K&^>6% zy9NnX)nM6kn&?NC!`=!vRz6ceRYN~R`dkUf+Zx7ma);qilRas>c9IBxEJefESt!-& z4jWfXkcpB$+<(_f>8(`y|a`>_0=)hvO{+xi z!>S)PP*ORHthT+(sfVn_-0D>nM_9o*-_M{fn+gt^e1G2LFdWa>z%9D7lHX?$aM0&E zH!`UnTKsiMN6~UJN@5pit1cil2kOBv(ioJAEr^(_2CKFYC)rkspk00x)+u>l=5uKl z6@3>wZh2$by+Bkv7i7b?sglI^KVQ-{)1D|>sgQNc#7HO4A0<;lV6Mp`GW~=VS$QuM znh!;D?-Y)32I6^KQU8P6hR0pH=x{LGt zaQzARq!z;Fo<>j`-3?uCp=43wYIqfrg!(bv+@wdDnB4ONuGC2}8{0hS6Q;o>duI}H zOp{A4*-mTk?G$)%65tK%1ToECfG+#t#r;;8F>?l)bI_klk#GaU=mBAF#A1?vp3fKD zZAvh546zT+fYYz?Xpe9*skgkxZTgc4Uw1F#0>~^d-2Vs8`6s~0XQ@Pkb)daTpYYYg zN#xbB@x*36@7b??z&+|jfzz-n=qd9U+I~!r8s~DTrwuxUmb$*U$+aHXe&oIEK_dOt6} zR{z4DY&f3585r2GABW~**6VsMd^=w&NAvI6C3CjFa37+F~g?|8ff&rN|&m>S1VFR;I;$Pof*!%W0XC;|Sx80P&$8|>7;eQkc zlBDtWm(!e)f)U#L@x4o-C5yTKmZol=&mI^^;JTiLOyS^GCfqK|$~RiEeaHG}RQ^r+ z%tDV%u>6OIuRX%U6-i9EdqM?vNmZT4Va-wJgdAJ1tO;d*~8CKEZV=A zy|{RZO>$Ai(;M4SCMyF|ozt22(uFK$`y?nF(??gl{bL^*Itt?juFTLJ(A~(I_Jq8| zML&7}_Z~ymS)YSdCq2pivYU_~>|-4zquG=xIjDHzJYMcFWJ9e(tXMveW!^l?YOY!c z(+?!z^^2t#@X!asK3YI!s}Y$UeG|tx?q}n7hN5uKKd8&0)UANfzT*j8lhHj~YcRwz z#Qo6W*HJuTQ6!v`=7G5rZed7eBubPk(!J4V=+oLHx^AGDh2@!7w4^%Xj#)CSNqPZh zKibS}#=XVWU#GFdPjvV>ogY@#8nU?pA;^r=!m@WNQ1Rm*&G}h^c?y+GH|rE8WzNKR zAJ?*>r|r}s$_;nyID%V(qu4~g4>I#rFhiMawq5fKO59BqJ~-CNEt_zZmTHKzkZ)I+ z&9<@3{=#~;E8d;`Dr|*$^2*?y6Ui;Pzn4`#{EF3u5_r?@9-igv$rE{3@rATMJ~q?h z{7dVxOv4xErmbdky%IA^S7LgH35&Y1m#t3pN5@uMTz;|-C)^JM&l&ab zb)_-apmvFHbqjFqikao&&2La)*ps!WEM#?)kKpn$bNs>Uz2~msbqGh!;oSHW^s>_h zM&%c>CCk3jW~~57Zn}cyq3Uep;Vl?tWXU$HYsW^fuaMMv7SeBzW#@{$S%1GehL6%@ zXX0a-fA1oup7RAK|J_F#4$DDotO>gml8AbHYq*WRQ*efpIFzU^#LC_UxOwLq{2MV6 zJ?kg1JqrG~YSLp|$;Gnz3I*IUstT-AGU$}hVyaX@>FEp;P(EA_rAoWlTXAz%YhnR6 zsMe%DGfObjdnSrH3h~`+e8)~Vm2mK^1Nn)Pi%GAYm|{xCvAV^SaJVsc<(7jzvj^$tIm9#df&=F-<6&HiT!aUA@QVrS|KgJz)zT}d^VtVU- zCOzCg9ZGxT*{Bu1WY%;?cyM(GI*hj>@|!b3w}kgxtW+dY&FwTm>;a9cS0N>z!$Dv4 z81WlkN!PEfL-qCXP*YS1erob)Fjt4^JAH&NVIrh@iYJ*{v=G~P?`1RJtGxKO20RK} zsZ?YXEdF-Sex~F+C}2^nI=zIh6xG9pY07L2w;FCIro!X`Q_i5kl--wI!F#)RfwIvV z?oH8HT=Hi!v93)QxGk6lvWs>?nUXc?=Y*leuEUuBGZ%_H%&;sX0!oA`#*2!f*dd6!RB7h zD6AK3js1kTb8NWMOHugZ>Jrvux*s~SN_aov0b*J)mK>cY!xBm*d9C4DOh3ryAGif_ zr>wkaN1`T)-(5(v8^p=;dy2r^KjQ5hd0?Sy#E#mTkRX8ay-hSNi4!uKo!1D+q}45w$(BWfeDz~wD&fLJQ4Is^|qF2n4`%V@20 z2rHs*gIC@?NclMkzpqJR%V<|99x5o;QPm_acdT($XB=zM(}m3KzqqdV+N?4Z;k`r* z==Jxb&gVv|tTLT^>-YjAf+S#rh#%B{;_oj zxw=NEu6qXOYNx`5q(8zv1yPV>YemN1iXfA#CZfrrWIX4cO1;8ok&6x)oQaPa$UbSJ zd#9fjmUXq!ybTY^oxV3hNxcqNQ1Afld?tX8%s(z*U@_+(@RbI3&m{+Yi|7)aOE~|^ zDQ@J8G;Y6kD^!c{TEu|~tVvFnb&ZUnAHLQmDQqS+zO~#w*X#vDzbo$ZeUlc&9-hz!)ZQ$xv6K; zFn+;2F4=SoZRoSdkp}xP{l`q~_^U!#_AI=la}b_vbB2{edtekL+#kEqw^y8N0Tb?V>Z_%n892}m~m2@<;^W8u}2!1)lpt6vH@ZDEX|6Iqt~Fk z`gX)_6>KTuvm}N!Nr0Co3#c@~57s?+MSmDe+%>UB`7oP&%d$fK)D3Q%NhMfoJB}3{h-Sh3$VOF1LtgCPZWN13beHw z;p&Q0IOyoiey&T!vu~ONn-8AD4Wj}v!9^R#TtGChQ`YJ{nISdo;kqj8k`WBOw0 zA$Iu6F_3FI!mX-31NKVi;l{)~IOSo?aTYvB_0nkiWyTs-`%ttZl4s>c~98 z<)FUFksY5I%YJ3+ko|{z>@lCm`TpFGiqftD}nEAI2~Y5f3Kx=ieEH9F?*Vu3mn$hR}=G4iV@lhjPIXSew5&6|eUXtN1_8HC`u z@(8x{Ij=LF+K;1o-_E>kb4kgJ5yb6^8XNIX0s7b}SjwNxnzwf{kFiPA^nM4Z$%>F? z#u99;`&W$Hu@grYPG&tf_ONwyCM!5}2p0XdU`KWh(o7EtHa*6Paq41hRhk;UGSFbo zum9sjqu)ZVO&%O_TFPgMTo7(OZo+b|?x5SJr^B#uYWdp>c_}IJ8RexTg&tWz7^y^t-SXvWgDe+lwSf>j801s>FrUQ#jlEvMkw8 zk2G-mL3NV}?C|TeOH(Z1d6SNKO6&+4^~Ql$X(sJ97(*Ot9LQ3^LKbI|jN!9-ai6q2 z{gSRtEu9~8fioA7K{-{{x~^)L2Bp5{(YnM$|E7_#$=9f;^x zSMGCw4y%~F1mXtDAVT(jGg8U3jN zSLZK6fyNok?kI-=a}#{_gJbJA*Wmt>b+jsJ9jlhy2tTfwk0d`PEAyaJMa%%~c|ez3U-oUn5tY zngSIk6>(ONID8GyhMJAj@w>VTGe}&wSQlQg6K-P$4P`c`e zzrNL>{pGD_d@79XOZkuXOPisdrwo&j{2>hTe~Y_!KBrOArCdiIpEcV16kPk9pyzUu zy_EZOyf}XbjlNww=5vl%g+qV1WXdRB2(>ooaeHYyunUE6)73rm&F>1ZGs1b2+)@Y;r&@&XF1v z&JK!!tTKP5WxS6q|8f+wN@J;JLKu=)j?AV}ht0kuk9AEK_~$ic-^VDidA_0SeTH|M(jYpWWLIIIqBN+5&C%ePBBK zrX)+mZa<*clT4uHB7Xy#ap`9a)nPK%4{4?_u#>!mc+}@04Rw2@m z-(8IJ$IoP%lQL-Gb$zzDOq;W-Ekh&KX?ShV8I;rdza}UL1?DbTWm`cDG&FG5wRXCI z&q(VwkwA|pZ}CvGJW*cN2S>dwqkgs;@y@$|dV^2VO)(4tEhn>~3HIPW#s)KZ_SKC2 z7EG1bY;HX4j=>*eP{raM_8;Qcw&*aes1(PLlr#8AA`-j)NV78!CCQs#5%fOh!9u$U zIG5)cORSY(9SJw_=ij$f`9~S$w* z!>n@FGB4xF?92}dOg4TdFe>{=YomV(>c*zy@V*MP9u>*qUsalOwg5KXN5q4KJJ|Le zO}AR(@3Ah-+wLhEX@2Ix5+(G~#7pS0+loz_v4l+?C(fpM?Zkm%N*&}7=~!)?wr>O; z7iz%0!xLDx&N-}^K7`dj_1KS#8*w;gE1%7ygS*A_QT3V>JNfggU}``sY6$XJTk~OP z&bDR7Ml@dqvc`$Wj%W6d#Wr@(9RYA$)*yH7cu1dQg$x%o6tDi4%cY)aqdsv z;UftSx&4!cmT7(1aj+k58i=!28FMztBZ+ggeTZmyRyZvFAG;QrF3=TG#NhE_;B-os ztyka1a*Xy1c1j1b$EkPd@+HYQ)Ulp>@_HRfw{53w9 z3hotSk8ur-+V_+ij5~~uzx1JvS)#W{DV@Jb565o2fW1{-_^atOT~j#?6*sS>x=vcC zF~SWyB|c*Szt_U|NV3fvqHxZH4jL;I;Vkw)qksQd(Q$clsIA7cN$V|e(UsM7_sBYI z3i?gmtlYW06S}y3swVV@CZn}YitvQU2KMil4YRluzws=W6AhLlvK#W%C>RnpBsYn6ZBB4YXZBXD27MG%kYh2 zHg3J63rCb~X;?Js0id4ArZm@x|QDj%T-KkoT zDr_B=#&J)c;*Toc>(=0boV+LZ@XZ{&?0Oj6w-lqta36+;_@S9XIc53xXyAo9Uxk3A`rROgEc65}vv6 z6m!ig?e0ccVM|H}r*l(-uD#uWG4esEx9l8sytYYjEX^6$$68~$$3ZkK@27{}I^+4v zKZT1|Z^JP!!twK53pV4?9sFrNB=G&NjH4nu`LjnzTj*!1Qga>`TL>6oI_%fH+VUk2 z<1wlG6LuMIV8KeBLgyqcEOt!8WizINxYenW#S5YG${5^L zd5pf76X2G;nOL@_5MMP2aLkTu{9tz#Uj?tD*>fl@o2iCsuP#tr)4{VGqiB7qCfzhg zh#Su;q4o(M+;}aEc6Y?#OND#F!oaH-H#ru2P2ZwS^BG*M9ZBc;x(ROoIgg{ZO%Mh* zex)vn^4LB_1(wGev4UFz7?#vWgDPv#*y%i~*T$b(8y%Zfv-_h&3Bk=pdFw|UT zgAS^D(6V?rw*UIbCFF?X>YVQ=%8p>Qc^?``iep}j7&ZhwqOUSXaN)K4QBzKnZS|Gm z{%D>>p9pJQk-8WI`aaVLKU2(^qQ+M58_D9z9auKMP7_TV=z|kgsJhSwAK$G-Cpj*mzri%MG({ivT5 zoF7HJq>quRkv2qbcK~dfu8)>;FK{0HgD4qzAFQ{_LS5b%ZrgEb-m@)=-uZ^G58^pv zQYWyRaFc!CD8{C`dtutLWssWf2FvG9pt6th@nDK-`P}&*slSO4WHsyt`?)t^gmn=y z&VNVlt(;9N#$*XocXfdH=0zN?*JOtScMD6{WZ7cjz9crj)#fp&lUMBAYu>K!^t^g$)423MB&3bF!b1go76Pn zy24ZLq;ehR{M}9L$}W&SbM#1)<0_O|Wz5{~nsR--=dhmlHiga;(v55WBj3kRPGph+ z)#CEFq&IW8ZJyeE26qPM^Jf~am=%jY{4D>&zR?7?#FK*;uS4q0KCV4;9eJ>5HIaFf zOgy=7wBUIQ=b}9g&Sp74NMk*0d7CKMZdL>4rEAFFm3%H>zbOrzTuA+2Y=L`ECy_a; z{{g(d#i^ERuvJ847s?_5#N50Bv)(Cu`i)Qa;hz#Y^ts#SyEwSX|^G_J=B7t z`w9^2w;f!52VsA2CH?rXj;0(qfg!JUlGV<%76rH9kz zd6n>P&p81+FX4Ui4hUAf!q4&rP+%SkulKsJ&Vm%!@aGff9@xQ6 z@H&N`MBBLJ8);?fmTTeUNga&%s%hVm+=bcimY}%wB)Wv}2N~jSkoxh2n>{`bFPG`? zeks0xv0@e-lb?tQNg5z$od+rP2XWSfMIgRt6V}z++V@#ou(2_QsHUt(G%l5aE9{5% z zCkSKj&tUI7`h~|-CSbdnIQlH?1(OH%d>zzE2Y8)O(X{K_UXg>GQ&|kA$j-v>WDQEZ z!{D6aLT+7WfgtdOBE21DiDymoxFWpwnQ8o74pZ)jShX z(i}Y73rh88?SW?psi4-Q7CxPQ$+_2i)5u3DsMo2&DvAd9JclMQICzOpkRk9X(UL1j zcEvfHi@8*FQK9#aEKvEbL?=o5Zg^ zodg9bcgk+$8G~M(2pq1tLgcQh5vLcw_?5qfWqOM2Vrv{EKNRJ*9QB8gfhc&EJQJyl zHZgyD5*@rVxXk8RSbSH3tIgui0B0+5!A}=USDu8&+VjEF+W?;(7QpKyJ7GnIC{s~N z!yLDb@K)p%ci24$jBPc@)9{f7y*dZfbLjlgv@dZJI}Vmk3a6hEut>#03r zFvDmf6l&%QY!q!_`x;dm@4Aw6i&_d#*96jvsplvs^$<1_Nyyx?5H5Tj|8BA<(wI3JMQJ;I70(Zh5mf*2iw) z#pNbEx9kI#qiKha&c>pzpFf=X_&V-p(UbdswCRcQZzK_cP@XqE`Pi)&hwsg-{0@&a|7#jGSK?e zeB5XeF0fzM!+8`aaw0?&7nY?!^M9%Ewpt2*@7e&N`q|`Z`+c<7_yR|6r-B~G3^-Y_ zNpSeF2ngR2!S~7}xb|!h*z#vw^TYix=0!Fc@2f%oOAdld;U~fR^&I#xE*HDynqi$v z4h{-bF($JUKgQ1nulQgbIuOR0nst%p4@ClPi<9U}>(TJ?5O?I2Dz-M~lku}8Y0+(S zDtV&TGBXPm*bm6{LKZw9}G&{RK z72{r}acid;llWE(IN7qO#!otfj4K!>9^X>Yeg1Kw#6CCf=2a00PYxw5Uys4Gay#-+ zmg9SH)zCMs95lu1c)nx-2bT`f%%yk9p)gYlcmx#|UnEJZ>&TnMd8BvNO59TsQ|r!i z7w)bsfF}oyV7A?H&b{y*#MvuAqLr;Mqga6!%BVnT;8&2iSA>`1 z;^E2FEYO}e9UgocFrDydGQ4avp{I*-xiIe+WaIO_a5r`#z;%udWK(qx zSt>|zZ;$oBj2u3HU8=kQL26)S`r6~BVh0MADr=S-bL_5z=Z=6oO(4(G_wuISn5 z+!4WT$pd(IycV{-FrxOeOt5)IDB1DgCRo^hB5S+F!0r7_tdCR!u?!0mc+QYc7%u~k zTn?G?emWWPECaGumXMrYN#5Bd6z&dLQLD2h2}a(^g}ZL)#P{|CE@g5mO#BpP`rBtK zx6UFOe!gbh%aAI9v!tnhwGBubsKJuPG%iR?(ey^@6!N_&81(cdVBp_kkaj)+>KVu3 z-dPcc3_PAM!X4Uonn?5cr@!Y!FmTH;IFzM9r83s= zzLPwV=0B4YPpMHKyGitDe>#l!Dm5J|Xn>v=erJ&w4~@5C;ni|WxYH=jJH!^FRj>-I zx4z0VS)YL4M|(k-WHflorf?c_QsI5J27H*O0@rLO3pZ#L3l=?Vu4Q0;y651k2+GxXn8{NS$W8z$@VxXh%NbTsE|Dy`lTCVsruM8AJ*E zc8rFKjuH?n(SgaCV)VfYUs5LZ5S%ujgd6AeP&%7oRD-DC{+kWZtWgB70vw64WGoB@ z-62j__@7s=feq7&AU#owx)n#jBFzLc?pw3)z#?;=osj?s=lQ{plq9mYTnn78MuYT| z64=|S2%+^?U}M5u@b6J3&pW)hbLUlv--iX<8U1*4xt73*uUG-~lln++|3_G-+DxWe zgu(8zY@EJi5jSc3K`cnW0LHtNd0*xevahq3`@%D0H~+EVPAuk#xDwBgxLpM=zV>oS zL%ZOYY8?3L*~0tP>xtN|RA`^10xE-s#B25vs@6Rpb+@R|n3P}Sw#9L9>)1pJgCyvk z?$MxO>n6A}Wdgmol6Srt9s$v7t01Jjip)W(@X(t;9|Vpg2lF3;@oFoW_4OtBrNo~_yNkfZV>{6?O@dT&DKH6(q>39i(M_Hq zbZ(t2#$tiw_ZHYbq80X5@Vkh0G~M8!Otlq!cpsu6IIfvQrx}lhbxIFF zKB1NPTtZl}s2%tX39Z<(lx|CFBc-)_>F$|2AP`#91^nOkXIv$nym=H}s5&Qn_OVkq z-hMupAyotiQY3^Pu2MvGlM1b13#n?}ARJsZlaBhXMnlFtg+V^w9rP*~)~KH+J>?5P zCsYz{dj-OllU0y-Y8AX*_LJNkSPtg~qJ$R)h2(4favJella5=I1=BfOYUgVRVj0BIj14j;!`!I5jdkw}Ix7*5JAHx9Q|qWD&!V@AybDF%UbKHMt<5U81NAFE!I@{$#~*kN z5i@{B95<$`y^F|`Mdmc)q!#=%HH6p)>p=OK1QlD{#1$(FaZ@m%OLp+iq2_c5KC*(W z5g9=)8Xty#N~ooadc)xUZ;^5PHi zNm1bvGbafA`fZ`Y=NX({bDwkIpUb<;rRar|cCa;O0o7kTO#Y6ogpq&$fTtXQMNSs{ zt*?akTN7c=9|?LQQG#yj97CUbyW{IO{_sBi51L$BCA{Zg#SDPf#{Fmpdw#nSEOvHv_e12%M9?zMDPnLAxCO+%vH&3Kaug8mxm|lVI)hl=|$wX$$`;>ihyf{yiO0Gq8 zqeALRHhcC+wnB{0P#SbIjYu^X^;L}pw)U{s9KpiB&Zw&8!47Lk(zdcF_B5^xFAo^N z*iGR`FK6>pp({AyQxRLJJcMUGQaM>SA!~Zm&*^p*FzuY`KC$L~=5eCNh5FN8N6w{Dn#{++%o173*8zNYqsgSvs{>LU9&t=!m#o5Pijo9hY z$-0i1ut6<<);j($eAe2`&Yvn_H=Ep9*t~jtXSxi((K)Ol{~>!EqEq+WvWX2P6tTJL z3C#57n7ZHF5UV@xv)rAPoN{eEds2ChiK#Y%@VW?O7;Iq=cGq$zYVRIFWE;8HJGVGoi#Gd7DWJ^DZ*6j*wVQH#5tW>UrHH)~hPNz&JT`0tVTm4ziUOjd! zV*}l9pUSF_yk;qSUQB6kBgS?}3;zuKW3gYZvU@9w*e3Znn6NE_$vur@o8F7ovEh97 zde9x^b5dE0!2m1yzKR7~Sdi~qkF#ra2ohc2P-1-?yPGD*I*#68qiip-Y|jSf%d`1( zE-JG@yu=(FjA{w4*!v9A zKNrTtc3HEMd%!kn9bk24E-c(Jm&sf`i}BXcY?8MdGf4Wzk{=G?YOCW+CPlHXQTYZ6 z3=3IxXCX5e(`SsIZ7p*?&L(P~U?ID2qtB$bTuqQP9<-EYz9IfBTW>e(YdX&4H=Kh@ z85*@?4fAoPMj%uA@*ecx^SKPqG+Dkdk41~pW?d{KTz~g1Ggu;f+_Ca$=)1& z!4AvM#lx9PS$(7ilRTErW*Z)3?o;yF;HMa#TVKkhmxi!03aPl}+h?3D`5EN}M{(_2 zC02P~kG(#BjJ;3HXIU{7j7HY8@l{1^OW6P(kf?|H!P~6Fbpgu`t7mUNG@;2*0Vae9 zS?DcOG`e*hoo5{;fBh^tgU989L(0`cw^f?t-sT^IF^=1~N43G=26ckH@;xNRaLk^t4El&&W^~8v?8L8zNQaEgrL~2L{j1;VS;uQ_w}+r8MSmTDc^mUyLvW}RHjeE zh9e8W{8KY2oVMQ7Lnd7KZMrH+s6WEpn;?ai4UfrxM|H^J>JgQzZgMo@i6(%(%elKe|VV}E>tMbAc>rO;vw75)b!tf{_g6xU^nEy0(_9h7$xT%je9u>gL!T#e zmD)GBN1qdFQ{)YZ<1!FN-#;%yGZ8XzO{H)xl@l?e@p(isJ$VtYE_7JCq#<%Km|bTQ}ee^Bu2 z;3XnP+~Cu_Fu^Aq10wBtTyT86BkYN`A<=n;L_c{cw9afdJ(BX6Ov{oW3ol$0d^_zS zeA-aPJpg@SioOqdf6cA-XL1_}ovk74{IriVxs@ZFpnsA)!gs`V`ODhTud2wMzWpGw zrc`)+whHI;b&hbV$!a0@LJk6a(nyLy;Nzxws)z)k?Bx4Lk$+@e8q+$3T$Zs_z zpKb_2UMh+lwpm7eMptu_=3K9J+tezUHxNJyCoYAGegp1@^*1iY>8?@`I?gRdox8q0 zo~y736c%60B;hJ=gzF`y61!)qg8%MonaZ@t!1zUpLXC;JLZc%df;lzrq$&C`DO#|W zE4DT>UDRaEX?V)e4<~1Xj?E5X#NaDp^8Av(?{fu-EFKF!YMFwl)cXSY0XZo29wFSW zE5RMOHlLhu*CKKQYEW4G91cEe6@Fv@Ou|4994nOf80Y4>1q`3^JlYtH=;*YH2dCb%Ff?Xz~r6^3^b77u4W^K z_Eut~&I^9;&9fJb`~|PJUJ(5AevO}w_HkR*6TCM5EjRyJ0_@oR5=X7`WmMCI9X$C2 z1HV|(_hO1P$Knfo8$Xey%_u^}&!M%>TP3jhrXGYwmt6JI+A}sk)3VNSbVn#zk zysEa3pYcp}y$Otk~YWXSokgvQaa33X_{Ykp=3?^DgD@ydTpG-$x(8 zBo`UB#N`~G`VqmoCl%u1B3YKKJC1cm-NSX<8+`FGS1`Y=2USaLY2#C%H7E1u%4N)h z7d-!J8lPEc?FeG3o=U8G|65G_B+lAu#8}&8SysYV@LUHS*d%kr%s4l;pO(>+GsB^$ zEI1m{P+oIYK?%3eo0X6nfmf)H= z4(+x1ZpW-i%>CU}CfTxp`EQzq3%3i=w!8@i%^ygfvjVuOeIaqzRp`+MTdmLN-iBn0%VBTr$!-}4r7F0z#3EI`0AUwhgCrHJI1rdQPbFgdJz^mN0xQoO2U`VF5{5%NSrf$F;g#iKs>25yXGOp z+gu9%KI@F5R`uYoqr97bL@2p+JO(nFN8vO>AX?_daK)$s52SxZuY-g5kI&Tq$Y~d5 zT;w}_BbPJfRdsjbS*n(UdL@&W=+hmM67*}% ztoSPoY9nm^oIyo}=w0$Xje2Tx?|WD^eVW45dwZ|Xcn zt3{FM__LlEZjnKqXBAw*gc;0wU`1_1nFdq4q=x(Y_F+|218) zcrR6#;YtzK?pA_AUsnv|c~q%i_F&3MJ+{kdIkOTUMzb$-n8_;#c5?JN{`&k%G*#AN z@-=NtI9Sc++f3O-I@)vx1Ozr`gK;rBSN_J71faR=D!-YRap_8ROw6vBOcP|rPkZ_Khj z9AS$-FJMC#RoNlsAjnFS<8J=Aj~ut1sph>S;rG&rjGP3{yK@Pj=lnyWKNhVVvhX8Q zz_H_uSY6d|mM%Ml39~Gj@e)gR_TVx$v~nSvo5Q;!;-_#4rdBvzp%;69EXGE$ZD<|f z%5ELq#6}NhVC!-ryFdL7^X8e-B_6%_e4jVQ8_TebVUw`mNeeZ3cDw4%aCZCDQuc`2 zvaBzPtYp6-_bag)pXt0unOpzxN#92<^w4&;S+*13xtvDB7moO=x*U`La#(F$f~rYk ztSTgfqrFCgD&s{=XSXQZ+W>6pdp&NqUIDf?-sH^orJ}X|e_XmhhbNT!Q0#94H-7J0 zYP#eK5kEMYg1yK&w zxcPuR%Nak6FCC0oWSk%F@(V>(BNjB96CIgV!6ec*xTKP znL<&Prp}*h7N@c17YZ8g^Bt@U3z(dB0IYK`K=CJp8GOnCm9?@gCnE@_|I_6*7&hY9 z-eZ`x(FKo~-No?xvaDW~u+QC9xJfAuH(1O^z2p>3k*elaZ`&*^?j6PM>&Y_bz+@(R zf$yGguVJ~a!DQP38(eC}pFQZ z4L7!1vQWLL?Ay01DDnCe<_f=Jl6oi(O&kYCb~_=ebuZb_X4UfK{%Hlfwv9Sw#k2DCmCBlp2yP5`6kMlPq=;sS=?b1f*WlG zDE71+W8d`PobR19*>gS~KBC1AKAZP4 z?@$|pNqUR}o$2?JL(;J9NcI3aK9hn?d0P=BK?1l|`1o1ZT+JZU4uEjx?m z`_o}~ktO?8sLpg%SA$-=6r^}wBd_-fFz4)b5Ab^b8QRQsxKc3Px{=GuxB$zSY=xEei|`DnGF`i)bk~U2)TsO%IQgaFD`UP-zR;X4 zJTrtdKJK8iof=4RXc4`TeOnOuEfIF6|HR9uO7W)LXReTU;lQU9NZW=On>Selxqtyajhn3db<{%`9`* zWY$@@msR$T;qDfVfCaxym}qqZ&2sRjjf=*y!*>=_=^HEQyN+DyUtU(*l=YdvXTAa* z7Zli&r>kI7i~|cS%jLfQ`pNBleS}%`iwHDwk1&Tvqrlhf6nA$$|Gs5I@jlNDsh;?o zUiw}^w?^qe6CVw^a3}|V#dSi}zFugOFM`2SE%>q^1JpxqlMzXJWWuO^ocg30|IM0- znR1)q^FKb@!?{6X=MgB%wCCr0{Qa|*0QvA4EtSZniz9y1*$NRDsvU>cf&9H-;7csA z7#6Nsw-dJek7iHnqxd|qEbIQXozKQaQO|~WT+|;bd{}rM3i%yq$CzSJ*{MzC6{4w^ z+<3C>KM`^%GZbdcy@!js#OazP0#=#)n@UQ26TY&WKnr6%h*pb<;J)P;R(g9r+oTx5 zzR5_@V`nF`m9iFWtw}Pz^IwYNAGeUuzlQAmb$1l2`pWr^I7IUfnp35`5UjIpONz94qE3!eNm1QW+oxOwq8uC-2|J{)lmzdx2>arS-q z;LJ#B|Mn5=8$5!`8(f%0_(uFAJr738{^Fcm8TdHznXi{3sB@(eyO}RU)T?oSnA(38PVJRRZ`wn`Jna%tAR+J;KV#emc3 zMcm7LJ9=|#D81A$2o8=ukoA@Cq^7IV-yPYsNX7z`BZr~ZJqui`UFm@#N~^9m)A&{A zA!WfD9Ll>byuSNU?S*%JoZRMrsPj6GI}?`*!IR2};Y%BK+ASWx1d7ns4d+4Zj1&2K zu~JYvLyl%x@#pDedwBKLiTZWUfj9rT()E@I$97pkRcku=tn?m#m>q|={tmb$zZoaq z+s)qnlg4?!GjL#@Bc555#;N_f4Nj}hQl*Oj=(Ti&9?f#l+I9r}KffWH*h1{2rHIKZBkr8_7wK?lgWF7sOZae@V4d|_?jI7ugYgar21oW+T{rZ*;k>K?J@Yf zXDgUVc#!UY6!*E7fU|ZFQJuY&&MRz(&Llm^&J)E2X1npIvoUFq%!Rhra=P$GG_|>M zi{^_&gXD!Y&a$|SsNc+liZ2(rE)NryI-J1S?Cz`;*KmbOO?^1DV-$<=_huUJB3M;~ zDpQYqk5&zn%f$`~0fCbg92abmTLziaH1)W?O0L zb3?p7B+fRN&PQ%X4EItk5>!MLXxW!4DAMKx$4}fQR{Q<%cj&>|Jdab{n{O}SOl}hd zH;lu2>(6la!95JDVwj?@0*(c`@Z;t03%(2MYU|$Lh8w#Rx&A0$f!p<3uz%JA!XFzz zY1TssnRt%wPy^z5@LuiFgQ8fuu84n)b|^b06n1RgF36knLAb5M1;WbG`0PhFI(fF> zr^OOz{&NNtqZaiOjiv^*`{12eDNb!`p!U}a;X$V~ZUdiV7}5C~f4+z#d5zts zRU(nKqnrLi{R6(ZDA!78<2?@6Zx98dIu5i{3okg($%5#A!#F-sj2Vvbr6ntkIO=&1 zZa&yfFQ@$#s7}mV zpbH~F<<}`3)K7uo$`a_EHxD*$t;HKNHSzVG5bS?g1rC)NG}Nw4@Mm!ah{xujiL?uu zx5|l~OW4geu64zy!xA`3RSeHq&K52kYfn~f3#QlJFQKBsGR#%T&tXTx*gO6CiV|I^^!?Fhro%G>ceZ|Ihi^XHXE!Y9y7&jz={j?h{~pGt_omb6^-grOVIlR~ z917iiDd06Jn4CYxv#LKG;yE|vsMUENXHGf*3EyVJ_LKUM?zaaY8&;b(=8lDyAO)g( z{3tZlzo+B+yEyDJBJbAhg_M~M-03@WAos}+GHomGwh1kPltYHFByAZDualsyy&uTB zN2{3IQ-rRAktovjlSuw5ff#wqS}(K9TwGIet-IGtp}5OxXxlGMPS_UU%sJxJ-7Fsl z6h)~_H}98QsYO?)#E}`28@Uv6hO|a{aDKMF5at^PRq>BZJ@fWsWXmiFI5Uy<{W8IC z`W7InX~muJ;d6}Z+qjxo2Z(q|(MKyB_T(v{j#COZbuw_(Tav-@dnriLLVUNdlyoo+ zh^vniR3)B3r_&kOI3^5yY7OC*STN^jZh{%(g2CtOD&S`L(?88Zs4`}7*>f75eN>X9 zE^DI=w&P*4{7a0S`wnx*MMCP)6N2@c)#PK6B&>W{0f9#faYLPRxo#SZIw5PC7CU``(6- z6K>m)}C!%f!f;*AF;Be`9 z@Shw-SMw~jtDEP7Q(G1>7@-amw!Z*_xsC9m{RA;>T95xEZ)35|xY{jqR>JxZ2HZ}q zQnIb1gP1zDk)Qg{Ne`O`CP!BBYlb8Ua*KfNDazEd^?L2x8RO<&o_G!)i8`}hJ9V0r zGoHGSv7wu97Q^84bUfhgi@xKgk+p#{xE*PO+?*ZV4-dxW1yn3L&!>Oz@#Bny6J|#w+`y zK>n>4&#Rq-E8j{+0rFUwkJkV?)T$7CFcW z8cU7u&!Q7vIe|r7HFQthLAUNGf`w);xQp+1aoarlxvAOlWTEvZu&*9VQu8IDx_1{W zl<0*SUfCcwJd0aU*#Va&dB<DC61b-A?7S?WlE!2|db4_Wkq`__*yfXPh-a6C? zKTYGk)EDz%D-Gs^{8#46M3B@shuZFXT9Fq)Th)zWgLEDoDcTK5->Zn%w25eF7zhHcEXD5Ol41feNNmAx3pK#3tv$$D9Z_`6rUR71Wr{8a6=h&F%2& zP7-}vbC@$E$z-c|JQ?m4fvKlaa5FWF3op{94#OvevuD4D+L}mu;rLoq8gUY|G$v37 zKBs-Od)!(e5u{OVWKjVC8_sOJfzHP+MDwCEt zuLS4&depnWm@JdbfjGHzSoge~9-o;`@B2+Pi}pE!&l?}&;&G$tJQp?kJGB#zwP?|n z13fh3z!g+78)}!(7I_< zAe|0+e8E?HF$ahE%@#Qn`XH&=tQQ#~A(q?;VuBiJnO< z(z@u=H5F7DOQ;1oM-TAKkt)*-badHf*0=OFE>xUDXZp;eyTgU_62&d}IXm{CO?- zap3M!SUYD!C^ZU-f~9k((T8=_FgJJ-Aei%%2Iu9d%h6pTQ8p5;rigW?b7fjzckCndhfN_$tY~_o2RQu{r^w>(N zjczp;KV=VWK576r`_f3FLAkM^9vt0b&sM#m5WnO%KT}vjUuYy!k@;%0a>jCScd~?X2|vMCg<$yWd;?FL`lBc&$^uw3BdJuc{h8~ZQ zV*}4m(r-lr5L$kMKDCR3du#}}9C>=Q^cGY0nLsV?y~5LGCH!Z%B)cD@NQc8;z|gBm zmi1MemZW?0{!Jxn`spjrUks&9YAfM@s~oGeUqnsAZE1qbB|3X+0d4C10bdOw;JAJ# zy2r%Q*Uv(bRBfdpY$WySl`<>df04?3s^#n+=pZ+@7>es6=z6(KuKy`ZjB!4?etpo7TQz!0r!l1!E)xDreZPPM7&)Szc-D+aB>91 zGV{SaHw7N8(k4UQJYV%FrSIokF^5SZwDpBK_VRR={FnFp&6-Rdg{$a1;*Mo^ zo^gHW+o1PjH$GZ;mJ_v4#_Jnzqrz;Y*G~;|s^S-fU*bAx^P!P+Fn_w)m){liX-Wk+ zd3Hm*$pX4}`c5ix%N7#f1i&VdZ{$yms9BFm64i50rFV~&qvo-xv`a*n-g@wv<%|o$ zlYdiaL1Qc(v1A>!GoDGuuiM8SJ0_uhzCs-3Dp>R|RR8$h8-`G@rSo3Esi^n4YOZ@VA@Hz)#5>$Xdzv` zw1f2K*0I3Kg;eg|9y)8|6dG&VB5*qI0Bf@DP%XVe(*^pK7p`y77k_oO#{V`*G>Hod6oh^p@mK-?)GDs5kqj!jCe zq$&Yb(v0!-zIJX(mk&*u`3(QCK&th&f}ZL0r%I0|3r$p`sgX=NR@|J4g2g@P)-ws5 z3j?XGcrBgvs*ZNw(js1^A4rGwS{lE5A$NFdBRa3Y#+F?+rKX3J=qS&haCf_`FhqeZggbGTzX@t0qqElpby8~ zM$y5y_-GrF%PNm?_h$#;yc9Kf8Wc}v{k{g)b^)~YV4X-XPqtAqt~S!-+_G-%su&{_>BR{xu<9?=N<1Ay` zKOl#-VeJC%a8d9Ye+`O1IbzS0DfHj8YT=+R-xqY!r|qWc)G<1Uc83#R1ecg#3 za}QE)uAZxjxm@#(?>eYReF2w&Q*5wnGVRX0%#Lj?fy_Y2217d&(yQ62)F7CIE9F_F|SlHj_(#d9_h*M$i%VJ|EbpbNBCpr zB?XqTv=~iS1hc6fX_)$TJ`-z^U^f54@X&K#Tya(rd9W{*nQC(Df)m;IjuV38Psj1x z%2bS3=6N!&&*J2)Je0b`lNOx@IU~7RT(Vl7oq2c(=g>>6_WpZr%KORegOv;SThEj& zt>kx#^G7lr`5XBBvomyjkzx%3Yi9o09gkM_p_)}HvsoRFTa%R-;!dV3GYh@NLfCC^ z#>1}4tfJ2e{US#()$QN8D-D~3Jw;AzpJEgKy21D2KgZy>xjJm2aX%+c{T4_&Zsw1}xwGF;-&wUKIWs?$!)z4?#l}GjX&9n^7D;Cmd zz!?|4LNN;&T+`Qq0kt}$?ZjrTd$%mhI+@8%6h*Uy{j0&n+L&!EHrw8*{_>8xGF?yZHo?hg=|v&T&D9uzW?S+K@F9AzOYjkQh%aH~k`nCs2XVYmJqiONx-ew+TIP1Y4YLC8lF^!f8Fx#b zWjxGd#*tGoDu8DyCtbllnS6X@`VPbH^Bj|B@;E&^8+Y%0!qxpbfZIRa=D+XZF066F zy-7!emA}Tauk$XUdBGO4_-q9(Ru*Od9@sM17M`b^a1f1lIU)Th#@L&en3XESP){GM zbG7kTOaR-ewT5*_-NL=MJ8OqGFJPn2ufuU8`ReW6Y~GcV05jD)x%`k_xN6fxba-Zj zqh;rF-G-0J#^o2e#X5|rjZ?-fUwwhWw$E4>Kb83jZ)5rM(M+%DJ>9CH&6%{&p$0<3SDo-20cimo%1%UYy7?QaaGbZ>R9>o?5mpB!F#N zbe&af{|`g_b5Ibs5GU53X0B54Y;#5w!-P5HKYbk*tJ+Dvi)UbVG zzj*ZDT+aAM3zqeU;ErV};uo;C|4PT*Z;<|Jr-JQAZZ>OZyL@Fm+;w$F?RUq zNhB&}rlHv3Ib5(7rxbn(c;QDmb*BBseFjRu8}R+r(6YIX}`#pAU*169ZqZR z`!K2VcUkh-kPx^!YIt*x<;d?=D4$GKMYX zL;S4kESQB(p_^jX%2+YjR!=Hei%zR%pCw*iod!DJzUZp#+ZRVqC z;<-WakCdiQ9lznd7;(5MV#N9H(!q-*_iMq*v!;UQ^qxpQ$v(|4XZ`5NbSD^5^`j#2 zWevwHV`ISeWiW*3DnL6{K$3n2ob}D?ESp$l~LS)B`xt7l@ zgmugVod!9!dA10KKqG#b@)ft$XmKHz?MPnCV(hLTO*$a5cTmFcH7=V<+iMfB$~Nva<d*qRIa)!0Q)h zf%M)|=;yOQRk7x*Kgu0bd(~jYtxl#eT92tq3}S>yHXglm8qfMiar2rt!!zwyIC*Ll z{8(2&gIDjS-MSKV^`#QH|8ES9wN%GnVRq48R3_}G{+k^s{O;^w;pWIoGjM3Tbu@L@D=Lmc(ZA>jEp9; z$cQgbQFmwvAI{gN^V{Bmh23PfYP19MzS;u&`cJZ#-TQc-K@6_R&%>hrS9p&>US91^PdekW=V2m zt!sp9-ipf}VZP}3%7&$r;)v9k`g_1s9 zCS3-ad#yn|dz?ryN=f`njKB z@_#E)Z`VugzRPp$BCm3X(~oiP)fL?RPtR~~!47uiMk2m!Jtd^v2zI8qiv0?n&qn6x z<2I{x%>GOTcVX}soR(5yHIpBaxuN@*z<&*Qbdx*Py*7r+dvt)M?^k1=M~$s>>-)#H zT(E$;SIv0x(r2{Lt%RGY0VMB&CyMgteo(D58xhDeVdsQ0kpsDy65WPtM=64|k214b zF^9Q3+p?r9JfnBE6DIjShpgYrh*g;yiwQUdsjtnL^na;rrIadLJ9{Tve`OX=o99%{UgU7s;bZj3IL@@$YTlUK8_e1E~E6CQZs zx+kf9G=XKbjljn8P2{6rH16bow^dj7;_hcPY=QMncHFE157p|!JsT@F-DNFv{^o~E z&vl}at|}(a(_s0o$QCd8$V!jKvIPN4*|=m=uwZoeokQlzk2^?=u8}`YT_`Ah{_VB>|~2lQsT~UkR?JXZKUwEue4~R)so0oBt;~$ zBqg%lnYj_NgiuPf5tT@xl_=l&7w$P{?)+xn_j#zK{!K^}na(}q^ZaqX+}gE_$E?4u z1CtBDeq0Dg_n*yTxs^GfmARd6@Z%Bhn|buvDF@~@?}zD(Tfb;)ViDacCd=gFO_^R) z2JkX=(Tn}JSfA}mI@B)(Gv@Zuyu0t<{=_(XS9k}Ga`_HVfBLY4sTWyJwJXJ;y6|A+ z3AuXq6MHwN!bU5fqWB0gvpZW`pt}cAC)q_fG$x8!p1HyC2IJ7i`Yr5+S`mNT;S0Rn zCUJJ4+LTRIlo80MOb}G6b6k-vrmWg-B0H69fQ2fnNOMmv9y#vMG^E5>%$uzo>uQkh z-y9D*lf$v?vQ>1v#6zf*UP@nmI6!HuF}rkP0(|4LLw-$;#5X#SW-44_3vSN9ANvns zEz_H{y1s_qURO($icjJ(5>48$2dtR666?+282o?q=v%|5L~w9BYMG2!sR;(VabX z%f1{OWw(tFax<rTgNL2*#P9k z1iWEMlA6C;#mpv~!f5Rqt{Z-Vs)pE8@7P$j*LMRuuC2x1a*kGc zg(*yTzbLk}ucmp^TgY{n3nU@=61yM%1lg<4WhG&j>`mzaU3&5%oqnTA+% ziW72?tKV%LE^`eGpO1sv%R}gModxW1`Y82S@Q6D0nXzNB#z@G?nT*}p2jdTaEYaVEv?7b*JBpVld=v7H|N z?BwGQrY{qx;!xwAwDDLe?0#s1Zw?-Zcfw~_rcM&y=$0>gEhy!fx69a>O$vD7{kgQa zLy8qWj;8xZC??vN8#>AsEOlMVj4$^Eloo|Uqf;cDeLM!yZ3fui&mQp9 z6h5zUA<`%;XF{SMfUFF{X{u{*c+XKNA!gX1V;2s5vJ5TIyFq@nS+aD&HCnu(5r=j9 z17q8W=(OxSmrv$%;qsvqG! zQXPZoMx5TL_w*gG;n#(vQ_;xQGtJSjJxT>g^0X#lRXTwgEoBN*Vu75Fp;A zM6~SFC;Z%3g37wiBCS0^xYtn_mnK(WE8h=T_qqpMgs2ie85N9^Yt6vc;}Lu~IuFx2 z?yj=47?tL^;tr!2nBl(!tGMmJjT3&LPn(VLe!)7}e#{cLT)O~AcmJXrqtkHq{1e!0 z=~`m$)kpO{nDX~EJ>m8FuR)$|Q>fL^pGY<4De78t9hY(MR&^eCAEX(K51d_(`*(+8 z`{SAX-hd&HK0Aw=Hia?c+Avz?JRVOii)QV;N_c*&F8=N!iIE|t2gzYPZt6{tOPRn* z)(^lM%`r+2n9}|GmEg+r4%|9g1fMs!ngXH)3Pl6iRE-HF&fy8){5qHuwF7`eI& zz_eZ&XML2#Ti--LR#Fu19P35;MJZtHu$Y-eHo^CiVB9x70spBx1gE;KvF23+*lO_) z;vQTx`4^UmH-HyjeS8c4r7#zB>PP4s--8ya1%Z;aGua9=RQdJ zc@_J4=R$buEj;epDsblY!x7`@_z^d&ewx^Ut)EoF=2C|{p1A3b8|Kw3V#C^QXpkF+)u*2WUVJk$Z@UM=m;G_5btVpd zcpO@|EOhY7rFhvLDI(P}nf~qYq*Vc`=<&y9Ud7{yFsd+~<%CkcwU;IB8(2o$xB9}4 z$#LMNV1TSzBM>q%B^@Tx0?XGayj7a6aOO^1ap|#hr0M=Hba+V-)%nm&^4bIpTJ+F|O>+enS{k5|dmbH=SSWaTYaeK>QoucjUQ&l-SEjY@GJWl@Z<;qQ znf!2(#$7K}(GBGSkU5dYdlz6T2;J1c zBs9Q1QwbhA;M9vsOd^rk*4cura1nmZoKj?CX+-8|pFzL-C1GNvw4g~Y3L;u~aAcn( zbu#1l`=e*Ml<<9va2k_p87&qb)U*A$Fch_e@ug&^aEoWTAx*VBAE zg+v^mDqywVRCI*vp$JFvD^G_*>n9J$TGY#bK7K8=v$X-MTh1J>RKoPd!ZCF0lQ?Qr z@8FrrwQ!v2CnWxxJQVI3M2FM7$kBa!P|h1KRMnP&o|XUL#WmLOk9YED6XWtnV|(Fs z(i?h5;tPqJd={3p+SB8)B?u(r(L9r1XjiWdEZXXiEKhDht^1VGZ}n=@maviTS)GA0 z%(Y=z-Yh|bP61ChWE;Px=@3LM?nY{12pOra=6|^qLH?|W<#{y4(1dLxh#x42B=nu> z6VgsTnkazmH;z$}1n}va15Z_Jrr<_&J8#oGC75p=Pj?U36aDJ?qTv;}%*(qLt=iR( zWzZndZwz@@LsO^z!B%_ETP*OoOG!@zrKj_jD3{ zZnmtru(yt7Z&C-D^odY(TVG%}z~J*G&hK%mpAET zZ7D)}lNGqW_$-0F&Nwn*t~+prXgc&^D-?0=%mAk@)X9}CdNmIr3+l%|GnYI4a~Maj z$Q^`_PEGu{AH~GGSy*tyMuj9t#u7+V6>Lv?4*QjQVH?NdCTm3CPu)Xg`};hcOqJu? zzI%nv=jHHlS_;2Gw3TOk?HEh73+ByyB?pN;Cz*nNI&w{%2>Izof_o0l=v9F#PW0Y` zVy(MLv#1M|tq7sb4{M3uO9S*Ja~()}$$_Kp49N9}h31K`;l7e3T3av2J`~oWd(mlV zj+{LFT7uclt>;0yJrbmxgy7)=8~R&Q2NYWGfPI@MWV(Ok$IGrpjU}agKR-!YeCQ0B z82FW!(eswr8!8gRYzcU8tw5*D%S0DkY~Vp_4li^(LbpYINZ_axRo*C#=FDwDkN>K| z-18Fjzs(u2_jnMD|2&E+-!!7BXGFPm)sV7&8@Q=^p4OlOdU;hmn&_F%%U+~`3aup& zEl=cclFdZ(r-{JN+&C&79R+6LH;wPf#G(sX8FY4yJ1;JLBb9VlW}%8@rWda6rNaLS zk+1E|0sqgRtsuJc87Y4> zhnBxUY}syKno+lt=Y3!ob)0@28V^53v6nT#=SBtbZ zQujWdn5-OI^D+&sdF6oTN%j-tnJ-at#x(pjUYQwjDL^0RS<$e*G%9o> zn%rKi16e*^wAkx0DP4x(NcdqAc|Mwaw6Q{hGBemnV&L{oZcbAj<0~jA@ay(*XNmTc z5%1(B{&Uj+NEH@^`fOz+d-^_Hi{+up!xPBixN>6eu@S<4$-vCB7s;IT16)yHGViIm zG5oPp2j}Tlq^HiG>X}pcd}tk1%sLCZcS_S$JOX<{UqJZjSmM>vLYC)EgY}!0=u){# zm~`a@QOj{aXnGG)wbnD8c$oWrR&*7=op=+T&t45nZ{DB@59g4R$HZVD-5Z@a_?E4Pj6T9vzyWC*W;!_#(i zjP){@X;KVI_$Zun_dp@9CgB1R3>$w*LzMD#@BAl#xcK5^*Zr)effC2Bs?SpOFMO+TB z2dV$tK&J#%!(r{6c;7iCzU0Fcbcu6>E;%y+iu2-4`idf8aqE0~7eC^;zes_#cb=d{ zFQ;+60>K zp~~9gD$or{ak_HLN)&X$oM*dBlkLz8Aua8B5H+@zEVc-Nb0d9_ZFLJPw%@^NYIDsJ z+=Jl5oFB+MwI9~({R*xt{2^^!9Xu*@0A5}#D5Pi5O!LjSW1lPDI+_DboonfMj-&ke z&?VaGJpmG;IKF!6Y^1wUl6_2h$xc0JWjFOZ&|vEvUVeTG=*-Y2xh%9;>{>W26l#a1 zmP&YVz7a}ZB#9Ll=YZ4NMCek<<86C>iLNpUr3Ft6VdVEEXy7u*gV9pZ@^w7UNYcaI z>5^z>oELpJR8G4BZlREaZ$S~>0DsD4yy0dEXr(QKU3w!(Y=m=S73^V2YGJspU?)}# z+>doKn*h@6$+yxt8Z2{(tmx+0lAAl>-qmz8E_{%`=a(=(wdx&M3_r!!vrHTjGJeNjn_Si-GxL+sTyd z30Qr-BORE`^*d9)qSaIgmc5Yy`AZSJiz&MFM@KQ7OqxQK;@$y|TtmwOg?Jqcli?W& zgti08V1C*Vewrx5_F08sw|FL%TXYTBl}c28N}O7iEJc&n$nal#B|}}81;p<6#w55D znmgq1<*xa(EOP;wz4tR5!xO08zQ3lPB?B-ovkKZLOS1RNiqRoKIyI21CCEXTN2 zvZ>Z&gR{&9pxA|{Y8iJHxA}thzfb7;c;7~UTfvd$Kv|$eWO+S&a zVKuBmq+vsz40dgHprcx%tTWjk)?d$nh$3023kZcpd(VOO;7Pc{-Ji}}eT;Tj%|u$- zI&`c48e}^-lbNaw)11>cc^fqT@}ur&LdWvCyj|Z`^JmN&K$e!v;PUb)s9w^7a$0TK z<(5VC{mEvyFtPz}iaU(AahbL)9(@GPc0%VuKQ5bl8g5RgAcNO7pbL%%QAOZS^uj}j z4Myftd2wUfEY5ZOOS1TSbG+!3{%BIYsg{t6Xh?jxneG_>mQK%`LS@PtC@OeIgyPT8 zX$DEeZodhHJ>@}vj4)k$dJQ#vm5Y}DI7sbF9jHrHH#MHpNwu9-!Lj)Xia56o_ydL@ zD_%~GlGh<&;rIMJRHeWWY2voC^s+BoCZFxN9sA5 ztF#kxCJ&>CtqD~A_)+2-UB`Rpu0>@(TR`=_8C>4VolKBuAsGkUsrWk+>T7pn;Eb_oTgHB7U;$IxPLPd5dQh8rjHlwj0#`>x5H&seBXpSNxP0b+i+$VXvOep}P+j5z!svNKaZYD6ALb zJmn&^Usi`44G^IPFRke1$R#XoaUoyU^fg^8_KzQ!8Awe09@Ev2y1;0ByQvJv9&V7G z0!QcXBGYYG@O)J&_-6pPSH=dqWnTzc zx3_^R>;9$J{>`F~v-DZ-g{}O_rm~>vvXkvM(16TDd3GSvn=biRMgsJ2^Pj5JlSwOm z$(q6sye-SE(6r`b^qt~nT0UWfBK zB+B>9*JXw=%V_%NSu}rJ6rEOV2FG7UgOv6qXshU=sqWfTcBVM3J-7+x+n%R+@m%ji z`Wcl{{J``+4dr&5BClSN#0T^LPGmC1a3B%zBbx8DG0G()AM6S9npa$(0 zD6&mrE`l!!E6Ede8Tv}97-?|K znN>5m9&2D9YH=~4=?S*vwWbHTYMw`%0=O)Hj3jk06@{LY>*)E;R`exNi$ZZV`Ow`$ zZmtG$usf03c@GhJaWh`l+cN&Kg)Jo3;yzzgD+ad4r692(8TQ8VA$gIBh+zkJM|i%8 zo1GlVHX#}Q7(GqfcJXLrD>uW~hC+@;7_IxfhrW4RU^=AYMm1hplcjso$@mwlq|;sp zPhIncMs$AR?Te13rmfBBPEi%_rF0_I%i_}{F)h0CsS>MuB!jp4XYhUAiNmTN3m`L0 zj#QO?<#`5$fp_H&qGNLpMT&-k-yIBAwy#iB+b{HP<1oM6Vl{aAiSpyEb?Gh6FDIg$ zO}ZSv6RFV={{91-Vd=jz8nQ;8wioY&?@gwlyfTI39S4)4We5ZChXAb#gC2yxEx#mCpfS1f?l!X4;~T_)TXb4TSx zSBZ<~c>12)1Z|7=U^h{jK0OTZD&GMK-ZYX$Rqu$t(*soHvJjeF+^EiqQ!uhr8lHyD zB;ch&RphhL!5C|(yZVrxDlI_fIe(GYyys-$zOP^(ZbIGkEFd^G9!_z7kbWCA^l*j| z-XM^`-@M)x@1ExZcJ(sU>E>*b&~u5HWiBLc`RQo)t{|%016X?yx7Q=#R(GWea#!t2cnk z;|%0@Jb)gV>`!v?-OzLgGblU09n>0*qwHCOMC{0TD*K5?rhhPncb(aAY}P`M`aXq5 z^Ma`H%s~3?o+!^UDVJR3oGJ++`{BR#E-)*JCfeu5(Q{+z(E7W9WcTl zIJbHkf5+=la%m5P##S@Z(%Flu)2^b!Sr>WJ!ft}JrVr`)Hj^mSMDRC%u0t3-K=n+F z;dZpfwM@yE$m7jJr%6JD^mZ?VO9x5PeuU+8ttwEkyyDZ}IUZ7vL zO$2Hup|9^h!Q%I&{L6cFNTzfxZ(&v`+oci1ZYV5b5W-`{FL*3He-No|xWQ#9qF7sk zx*#oTlAzHpfW@5fVmAziX-A$c`)pK)+HOn`s4m&YCT>urYZs2t1I~PO`NVmq2@$M0 zNQybnm1keZMY7@ThgkyUxOR1cMEZ6G8$G~hJHA?y9p1qttNSL4(hS8uO|3LvB9Wb% z84SAS3)t#|uh>bkAeQa;g32sE%dGQ{qRQ{DnaggDB`CXwm7NvGHkJOMc=8SZPo6(b zyrfTMV|RjH{9bx->?-SbAEoK9y!g9=uh4(7t6=Px9{ZkA%CxKA(m?Yg?Dy80%+>ISx#Cgaam_3 zm^M+01?}pjT9025vzl47GFJuCHGjz9!lFPCbC=J^I6lA z*KCt`A5=vxqEF(_fJfyDrt8_c72JK4!8jZ>o3<~KzP$H=IZRQeuO~a9BQC{k zrSp3Dc1D-KH#>%kNM=w8aT~U=ej&TRYlxlrK20z>@RDWg+p*X89x$aDX1K+o2u`ne zV6P4zWmYC%X|TvjmcAvCdL%T^LdW0Wp{paXwzXrLR}ZkrZgcivI;IutZ_ypM7n5e; z?QAtagpPcDLw|gI!HTZ6fQQU57?yBu-LYRhE$uE+tft14Cj4b7-#@dTiYM7ce`7d! z7qJ5`qnTfgJ3BlnpFLc96wc?XGu1u&=$z5L?85RerdPh3b*~l?2<``h5~Q;ZD-X8s z^=2l{xfvfxEy5A&G+5Yw$5@*E5tevj5xe}a2W^Q|V`i%r1e(H8)aaxdcDD^Bf>19u zceIslm~xJpbNQJ&eG6Hqg`Obwbr#+3wT$U={7Sv$q5_=-@7Z{Ti2_GEYc|=`gjLO+ z#+>f&V_BnGY`I4lwc0*_R{mU2ynWJ9RwVhCvcy6$hj+wp|T1oqQXji2c0#%o+w zaS79l3}uO%uCS!0$Mj}o0@zeqv5%Mk!J__qEbk3wBcp4X_Vj0zSxsiSwm+$qh7B{a zm0)tqZqgas71%&s3iJK773)W|um(Rvc8YI7snA#6@ZNad#`1i?KOmcc5t85KnDmMMB zA}gNFNKIfq$vUhjxreH5R>50CH&YWULKnDIv9+InP``hVSiF)A!*fDdapMR}+LguR zEQ|%ea;@0xhI!20xr`d@9%J2qB-o> z-$JZd?{6N{Jf_W_^lf81if6K`aaI^Rx3Y6R$;{FrnhrD$v71MZ)0zchxG`l0J0Ln6 zCg+8L)ZjR*wUpx_oJut{-Vsf5sU(O@m0|P41iZ+=g*0w2$E-K`#<}UV@#pn-I0ozw zNN#8Vov-oaIOvkbW5?)~ z$wcxpNs9kXN*sq?_d%+Cfsng?3;pnC4)W#P>PI_5K_NRC@)qXNn803A+p>$k(4>4{ zeir=^{|wIQq?sH&C(O;6!EiSDI@~y34(~o^!nsv@iEv#lQY?K2Is3}_QI7i1p=5)L z<}rFj=PunKas@5vp~TWai>A*@g>kO}q5o4nDvvWEieh}a&};)8oOX`HEwKmrL_M10 zQw9UYL8hrMcA_C`X|Qvt}Clp{oaywetAeo)VopN&>2eJ)u0cpOQCFg5B=}aX?Az|Pr7FFcDgdE z04=aKgZUy?i0kL6wDI8{lH_m_9laq3PY%T)g@|kK0g~Dh;b! z0*bp+gXwzxSbBZ=e75_y2nl2Jp>5VMt$&pVtLI;araQ8l=&pyFURolP@OH>_j-yFB zdIY;G(jGel2v$47^Dj5UafeP*@i9&MRll5u)Z_qfwGaeZPv^VtRE9#y1;~1K80SaR zhMAmSWRpP{sh$@N0rNTUqf{%{%astzMl<9#ONh?S2_tU=LJ;iN#9QK+4c1GPARv|x z{T};i!?lm-<6&Ji_BM+C{MkiP9k+pMHs`hc6AqPi*@&O;lkAG0K&u2r^pzfG#|yXs zKV#Mrg`cG`J#_|J;O4`>rY{N!@`(^8dI)?ucaFx1^AJ@09ks1#hJv0gaA{K?YLkkf zdgAG1;NA(SlQ*E-mX48c$2?p|~c z_RN2SeElaw&wu_9>=8oe%`Hc>p3FAwj(3B_fmJOCi>4nsBIxjTCAMO71l+Y(o zE%WO=gZsW&Fs-+u`+ja`*f^z{-va1x%zDUFC#8v3#$-8J-;CH&P zKN-&&RmLIx5pcV9G0s_Z4V9rFmbtPG2Gu#AZ2UGFuR(xBEy2|iH|VkI6kPa6u&{qN zg$}d-Kv#Y>o0+A>Dq$-L5(>eC=`+|%*F-FKDHuA{1voNO0qW-2V26P%_{vIYq@Z*O zm-!yxEl8M&tBf+}kH;#+qB9!l$5`RvKc}H2XeVB9`Z0c8umo4#lEa+`m(rcKdyu%f z5q`CI2L5Ukg$F-7Fr7QywC7wb%M!a!^sd+=mGdy^_@Nzs<9- zO~*+xb|42TRAjM-@>(24e4ROqS~ z|G@Pvbk4j0qW;K)YEBW5q3%K$ELxA}&QXW-f@5G4?9X4*{S%h!$Kc6M2l2|~Q$fFR z6<#TK2uy&^{lIlPvBsi3Q#~eW$6+?ur$9+_;Ty-SUT~9ks%!>JqK- z3}Z{bMZ(_)3$fj{JJeeH4yYt$<9$|h@!AvDamb7;*co6AzCR|?HG)J);{5~joG{$0 zv;kfwZNUdMB+7w z!*Sg$%mQ4K5q>v?5{31|^~EDdpK3z2-~)XzDHN5Z+(GDl2wyLU>j#~l%FD{TOU1o| zag|Riyf5m7O>=s{ez<_fzDOqV!NFkdUqT=2^;4&Ut=Pv+9v64JsMY>D15 zx+8yr_KGO73O8ZyU*eBjak^PF9(~L=EKrL7tGsR3pwqoc=hvi^j79R*cgT2whm{($#^>Od=>=bBSz=+CHV~fb&)wI>XI`Oa8xxC~O)+y;wqDeMjf_}4RMkX@sO%O)D*(81L>qe}+NZ#9Fj!9?66d=Wgj z?)_>79nkF3!^?dp;859zupv+tYiPE^ofC3+XTlL|`!5@7q~3vW-2r^#r6^8NK8NQ< z?ZB(P7{ach2@tsK9B8%e#3ee}IHM&1f0kH?13Jv{(Cf=&%a5hSFF$sJv zED`y&T!DY?#vp6?5pBBh1>Ds(qwH^bc+{s4fvzVl>*vlm*Gd3O>*u@qD&uBjMeLT? z&Tn4!1j>#^!LgfZSfTPS7U>7vFgOY995Qel-wogYuaK;Bh(U(qZxyf2sf6iU_fz>! zBkbZ7g8Mhu;*Mh*@c1=O*mZ{&p0w@<&-}ng$Z>VU9wmivGH)NWe*6sP_Z{(~q4#jP zw;UF}JcM~eqS!IY09#G)#HiF44eLbl1IkiB#U&VjsLR9dUl8XQe*=E2FT$#6d+^J5 ze?hb(1KyMf!Sz@53oTtV$?P9yH2+OI5Qh#_H|;0*Dk)>xssmvg_vO;&b^m`|M88nn&}Sgxa1Kaagv|!puj20FP4L&Z3yB)2@J{s0VYS_+xKU*Jzn8>pRfwYQ<;WpW9_kp?#g zf*(!Yk9S`Z$J&19@Xdy9y!VkRR%{9|-F{CTCoUi2I0oCW+*J!4&9PH`&1T~CyGQYc zk!=wE)eP>UFe3z8=bL={U-du(+E#ecg02zwXxK&BQ8AN3m5pT*yWf1vKN>F~+R3m!&DV-+a_Jgz(mo4jnqRPi?Eaa>pDl3=92L5eOa zjHVg!Q?br)44%sifZ((ZW>6P`)gP?Kh4gP^Ez)1&&kpPrqmzc_(S-B@;5SQSJJWes#g|+ER^5ln z+}ZT7@H4U_;S=rjor^W@Tf_M?#&Fas25RCi;tL}#c#Ul^$IS}B7xXSr*)UC*ps)b1 zN_Y)dOe1lI-*3G2<1lWp3dQw%B(cUZ3tXgA4!?IxV~r*YY%=JMyB0^_?u=HfxaS{o z4WM|ZgMaa)bzjI0ovCJ5R6aqyz7$@%`YPVCrXKrGS&uhx{@@uLJ45ip8f`y08UB@c z(1BIQ@!UOgK;6b0a=OMqBufSVY>~ui7tQGPA_q|T>`%?w^zZ{`MJ%G#$Qx5NFq52S zgq_k%@U|s8@U-4t^yZrox@Y4=@^-^q9O!Zf58~f=@k9l-G-xw>^COH&uUQTw(;8s@ z6lJ*A^$EtAUxR;#7vRqa@A8&TxB$zaE8y9i`?30?NoIKXI$pZY5c`M(V8=`yJljbL z#%5^ZX=Tx{#CHN6*Cm5>Wv|hV#@X=TM>fU2CE)34i{&;i#)h#KNTFBV?2?g*S!(%j zy!k*p9SIVG`&Bh`$Ig7LyT$;fxOU+9F>Z}hIfdtc?7()v)}Wu?Phv6a1{}I70Nc)9 z1^zDE@L+-(K6FDKFNtx)KW*0I?KNlds}I~BU_cFj8;ZlK)Ej^9Xe0-kT-hyKIVzl) z2n!FF;V@TE?A(2Zx~`5zUCzO1OV>tx^v!(SZ!#5+?-J0N2LHkCR)5s=ErdGmw}ale zgLo!=jxVlWfKzSW;1N|B_DxR(M@%-r&);2#40}DCJ3gD*&6&d7XBaToqcW5}KS6$- zEaxrgIROg8{1Hta>a4Q=W28%HNelreAQwzkgJHo*^4} zuYzHY7%p7D0B>uWYNjJAhrg(CzIO|bbrvp-Q_aNi6&E?&V^l#WkNC0F#%gLGa2FDf zxbVIX7+^)8E6`E53$IKsgZEdw@vl@%Sg!K`_C5WG|Mhm^TL&e~G=1!F;l86-y|4ys zR;+?!bI0R?u9cwU9S0ABwds+She3CV5wp0(@gAh^k~WhX5b`#|4wk!Mnu|JBP`OMa z-#W1TgdBMLKm*6F<@Nwn5uQ0N=c}Do!R5IFrqeYt;CiAbl5nnsC9JX zlwOZJ2J7J7PzO~TY9`Dn1qD@B6R8CwkThWo4_D{llji%ep5``eIvj?h(}QsOt+V*r zOU~bbpE0i}VB#M!n|9|K-RfvVC3J$BuHj`Cez2P9NStG$<$(OrjRxG96l-%SHKx(a`9n?_GT$}Xv~v#U&(YvI!i&SmzGNd7djBNty*>|r zxjGHmZZ6#`0wX9K6JlkK#talf*{|k#>|)P27Wn=Q{aw0)67@WayRB(%fF=`9 zoXu2~;^5Z1&1_hDBlcG_XT3iqu=cju%=!LG)?#spHAwDa3YM7d{4UIXJG#)_v71=F z2VudQg>2f~GPEJkob5X<%%r>LFxB6G*(8e)T(#msv9V_*6{LuBoOW9_O&9nJPM)KN z$M#UIZ6#E_-VL8un#gXao?&HDc}%$XH-G#dM=Il}#|x7@fyAfkv4@v_vhFp;FlM_H zebD|zEDqMu$ijMRcefT|%2Wj3vIbd~4?-6TX0qEEZcN@Lj_#e=2fs_#Gml4%hIe^$ zd$?CH%kv>UZE}d3ztq6=pa~9tJAvu$<2rQdzu3Vuk-UlH{XrqZ0__SdMBTF+=(pYe z(Czh$1c}u#aZ$#sjIOYCsWI$~=0?Vkazh7)r!x3JrGYq63*R%=5@Hs&w%Z zuD*H@C@?VJPi7K-_xelAzCuAiUvJ-L$AvG z1xeKumSR^(m;I3@DW;fZzSd#AM|%-p^9@tAFk+rV>eOcMVQh8hEfE+zBNJUyP{N=V ziyNAc_4^pjTM-Vz<8n~4V*yyWjd9(8Af}{}O|1rN(f;P8EXgpLy=oq%Ol<=@Wz>r< zp4Dc?(l?oU9OtU|XTvtXm19=%E2#TcXV#)!MH9W1Nt89PXLBa7r4fzvv)Cg#<=FwY zXq+pJY}n8CiM!JIrvCJLO*ajnozA2`*)hFnE%IpgIl77at}g!8fPBSSxLqlfs;w9Y zN!PULqYejR`tm)@%}ZmBvYo6}sEjhiGU^*_o!-lG%yj7ZIKj)kThN9{djsPfh+Cbv!qv6maw z@YO0WtY}?JS9QqZ-c@oeuSu4c)9bant2a|m6!*}jI zx-e!!mOqFiMCy^<1MG>S=6hpc1=l6HdL2_p>EN*2IJL@P04sMk|fV$Vbk- zw7aFDIG+pmcWn#fnW8oHS6V&v{+mjhl^xi4y^l22p_4XhPeku8=uorW+mJn%J5hW6 zh-M@WgZJ#CtW{Q;TK^Y?NK+OFWhSEXwqq0)MzCE!vPf6=BN9-$4KDpXjMq4%l6SU- z*jU^ho)&I{ae;KRilmFZ!_qZWri@XOCG;PbrnkTuaM$oT5tk zvTXknHCDN46^lw$#V+2ytg`7YY5pKj7QWUaJoRDzy_kpm-yq6%d8@Eyn9s82h_OP) zO#1VLIp-!QCMQm8VxtYYXlw-GlZ{H0c>ScO?-%h-`K$!*KRWF4`XGL1_*p=c!{~wn zOK^wHeQz_jbbEXWwY|p#|ua*dB8F zT_8%nQ40FL`{~yYM`5#o`>i6YiL%@p+7Qx1)-0cd?G1_{|L-;$!Sx@eWQgME)FzHa zJH~q{%ykSu_n`|B(R3-C1${Rm{6pmxO8O9vWEC%ix=b zANA>Ur4pHv5ECX!m6R`Vdk$?jY2$fv)LDyGPu2igfji6`(jeI?3$WMOb2KST1JupE zK|>@P*1e0Ra^r0vBHtMmO*SXCdqSzKl{095-U|U+|L5$?`VH~F5DZ$ z>z5ATmQi6$JG2TPj@%~PknMtDjSuigpaHbXy^eli8aFvP%i6EVMon|Ktzj zdfhi!--=wmC8~xKsR|bOY_?!wG6J8U8i}h#c7lcdJIVU>U)Y>WhC=a);rK{o4o3z^ zV?T#aEO4MI>WqIPv_8{^Ydn#kQGZ6R8Tzr_Q8$I6!d=+4TA6JP_=J1E++~r2RWR>R z1sk!nFTWr0Q3yD-4c=)hR99BC30|WPGR^TWq`PS=?we2}S>MzLm0y1lc*taS%Qgr* zi&m0a_LwO)>tVCgNa2OTV-o6YLG*t2!-0w?*cEkm7X8ve%+e(Yj#>-x{T@qHSUiuZ zNBM9$|K0pfv@hOn+kmYD9r>;sH`z8}6y7}emdWH~i<$r3=oFKM<3;>fnb#}6qVo2AMCzQ zqVNq%DR zDsZ!*(=l$^V9DC_iTKQD5Nd34Vz(xgu-WZ5@OW&Su%o5~uGZO*tUlow7Jm$kyRC6k zr~)gh$(IPas(iuxe{5t=pvaT4#+xq;F?hjK;j8v-JQ0&7xOO>UaoZ0z;c_0Ev_~Bv;~Z=89gXW3fQ`ER)|Vi|@*d*ps6T?AL&C_@BQDuK)3jonL)Yciu2mP{1-!XYi>$g8HTDpP-JRQY9guh3=RGZFPdjYG$3i0SAG3R?? zH{Lbc%bY*ohtv<&BJ;SCHI|BK+t@J}nySDPD}S(&T86B){e$rG`yr*60o?Sg zA4|W`m$ZGngHuE-!~>60DEzU&3g=ehGr9=NSLNg6KqLIV{5bkVSHL#2e7M#67OnMG z;V*8pZF`Tg{wbsJ%f)l-=;L!@PM_l1Yk88JhR$NI;xu^_J3vS|z)3~OW%f*IJ<5$x zL#I<4(YwPOALq%iBe^%(ID22gr!_*Tn>P&&7FV%6ErM$+d?cS!j-%h@6x5!X08xw2 zG45)MFC`_ygrARDYy2%%BXfpj4APZ^FZm~!E!oMg>Wcm%D#QEfJZ4o-8(3v)2g_%V#~Qd}^KUI7;^`I)IAkU1J{itFxUWIO4+?^gRTbO%^%dH; zr;vy?UGAuT1htQru^0Wla9e3#)b;|Fa8etOf6tV3*gGTjjA!N6k5Oaa0j7EJ0{L+B z4S5)q%QGKb;vU{kuOIvLZ_P>N-mG!u#;UL$LTEuxDEzEmtL08T%Bl9J4w5D`~INKb{C-`R4 z=OQoj-jf;Z@X1H)zi~TA#*lNU`27?*j*DjNKT4>^*6}dTZ6dueIFqZrNn?J=%dn-N zGFzKA5_2sIafZG+^}VP?&%HUsuh_j~v}Og|`|=kq*P09SQ_96XXOCom+BvLSCGtL7 zblIRk!|=)iU*2J_R{h~qGTp0{1g)DY**4cnyh>#TpIbVfuULJAE_gM77Tq}kYD5pN zUW|c;Xf^WZY&3r9mGCWRs)@&u>EO6`HLOyw#r8YhxYn+muao@%RaBn#zZ8zj-I$w& z+fySiFSsN78DC61#Lo3Qh8Yohct!dZ3)b2QFH%>8sgvz6y8pW+|NZ0*E*(|{)vrg= zHHsi@w-j#+$_q z-m17Eq@5%ncmINu?;H{9O3rjyNe=bUSwjmV<_lMa4ww}q%NK51g|Ra`&~CmwI`^H) z22I)ldNP`z_OlO-PPr~92k0X?G8;Q}S8fm#t7fPA<6K;3rpg@(q*B zK*~S{LMtx9e?@ip&}kfhlV8G4zS#ud{%s^9^*%yscOeZvA)%YsJ|KE;=kdN<(v3u@Pxb7QCM@ET!rj2^E%4sn^3#(!kR;##zo)(n0KOm_!$t0xY zJoLC77wf3s@U6>z_-FJH!^#KnJzW>!ighB~dQ*b0rX8g*_Zwln`2EuIGmxa7!Z-XL z$$L}FNUZpu{&2qx$KMfs#}{=4Y)^o#<%;OB!jMhYJcrGN#>~_tm}|Pqa2T)*+WneA zyW|3ne&2x|XTI?nCkx1e;EDXJ?rgLj{20wA+k)doWxi_AKdf%q%H8A>d7X7EW+ayo z1!)=VSt*Mi*~6%F!(UJ|wV+0|+I+Zq6)v885p&0_MzihHu~D39s8LtJ>*|Nti#`fG z;Sb<|1tt)dUy7d+VldfhAy#!h#QO4lY}`MAr(2by=W$<5|I>#bo$LoO5BAUv-`3EE za&PS8n?V1MNshL&|l^J)ATweH`ZBUS>x|?(9 zUXj~wHe7*T99>EhERMj>{T?hM#GJZ)P#|9hi}lyZgTV7rBK@-zsE5RuW-92?i(;o!$)mhx*#B*z7m;zW7nS}+ic%K$_B8LJ|BR6ttV3F zFWywSxm5U8Fp|ojpGWtOOoqS9(&7H*_b|n<3t&vG-HmvWHB)~STvyKl!3c9NU0A^cvb{w z^qh&6V-na69YD*v)S+*j#Gm2x{_iC2)w ztJ+8ZJ77g6(%->X?mI9xjGnl6mUM0yPq!x$>MbT8&K{#LX1}3HU1c;&r!S5FIf~~lT?erle~9*D0oDg7)1Yl~^kkbQZQb1f zsGv_xBR0|h{*0utmFW=w>LHx74JBh)H3`i`_^NL~-^`AK5?vKqlj%=4|GNQ?LkSp8 zv8NM?j>GcQ*)%6|@Pvd(8`8u=1NzUo8eGaQLGt!9Nadu*uibEmYKVQFU5(zfss1Y+ zS~rn)6mEy?vQu!cVKYn6{|sp*|Iv-tuZew{V)|^>O=?%_Nq>FaL>D_PqFT@Osa~!n zz4cs+riRGSp(V0(nDtoN|Fj!j?>3lK$d3T!urUy+V8=*J7JdBT7FfQH#JK?*>DGzL zbV+qqS2V~hqpaL$7&M<~-dY1*_|e=}4bB=o?KQ_!GXOTV_yri#BB zK`(P33~D?9sD2k_7e~V!BTZ`m^AJo}dX|*_9RruW%5ZL94QerHD7}9^U1ZRgL6MP! zgy?OCiOP25lfNN#Qx=~glUCAw8+XtNU;ELVMcu6P(PZ#)ej(;W66p!C#yQXQ3rIa4 zK_4}j!m-R0IC;_#_O^aT$CL!PY`qhvN0pGt8K2>unlyZ`kwBZWF8yyx3^n<57YaW! zx^e9bP#|qkq826m+_@Uuq*GzZpLw)=gdXlZU_@4(RiqcjjE9rE>fqI!zVuDo1JSc% zLT}vaA%i#Eg?_`!C_KMHH_iZRgKgCowR_;o*=f}OK^Q%tC4N5Q1$nNTMjH)7*k-3d zes$UmuEmG*XDVHojt9dfdaO@E|=}tg+iHMujJ~zTvGAP5dIEa$Q=vA@x9VN zlrAvA$A;y?41+QFW1=+%W}e`yychG!!}nwM!x>n!>l7N?cEY2P^6a5^lu%iI1Vfge zA+B9J@babiY|nVqFL2GMEbP$xi49M>S-7fL+wb@X8&}}Tub;Yy#dV7OW{AkMKc>w$=j}mmZOH5E_c1>wIbP>E9)oV1 zi15ul=%JFxFUWfF%(wfv-i=~g$C3Sc#pGH1rv6lvc>X|-f*`h5u*S4L|L|JQ5H9PU zkMgRY$&oYPBz({XoU;EU>71#>HB9f|=CnPSn7tZR!Zi4|bRwunC)>8)4$hc_q+cTQUFPcwl zOmeaPuMzsqxCw6aOC(l5Z?QFRH}LhV?fHPQA4#7DFTu2T4c?eNj43H;aj#X%U=jS9 zwQuXo4<=qkC?6b) z1~Z#+qOLu+*}i~>ggJ7lE7ts3xhAg?*~w4tDwDg84CcK2inA0iKW(;u>?fIn;JL3smtNtWOLpzo)fbrt1$XwDyqpRkOd~<8lg59t*$&|1O8mcMyX1+UE$Ck_Z zS1O!z*ow^Qx)8oA#F*zLPN69(cTh$6!|aX?2j9-AtoPtE43YYb?}x^tXKMq?f4!fi zmoFpheu;U$Tgf=g_ZAaO60zKDCtJU92O8gN#*Z~FP+;rEuPoUq=y)fIGXf*glWnb@ za`YKGpX`F-bOXL|b_Ryk8S*me9q4_onb=pKM6>8Me9nItuyma?zuNyCOf#_Ndrz&# zdoPmsg13%P^G5@f-*#b-Z!DakFWKP@Qd~N!3+cqy6@PfWil59Cq`gknkb_*Wa6jFHcpIe%CgU-4EA7;u{m5HERoM zSL;C6g=Q>%J{*m`Zm^dgi!mj)kWJ7kmINIhB=J7F2e%%3#!}C#lHGx$F#AUUk%{of z-bx2%V<`GSbSm)Ck_TvLy-c{)`cmBKsg{E3fy8g5SLBFj4KSXgQxQPqAy($h+y z^N%_We!Kv;#@@z=hgqCH(#G`b5_Wg>3)`uUp%{PstR%ot^z;mMWK(r#F`Jb;aixhI zyl{QVf>!Kj`V}o0u)mWX3}1lvFDUb=2X(M;a6Z02a*Nz=Q6=A`r=aU@&FaalL~!^% zid|`GwR$qSBnC+#qPOW2G0*I<`TpVw`E>WO;C`o?bow5}!^w^K)9)ZR2sGy> zM(<|I)enW$!@Y2YVHwN!9EXqBC1Ub}v1sO_#H&}`Vd1TcXyYvhIdYB6?_Cm~V&4d> zYj=nqzW~(xcuRQrauTk&=EKfc?#3ef`S{TDmwc2Bqn0qw~= zupyPHIis*=&kP*+d=gm8mXm0sNY>wZ8GhTT3o@^Fz?R^>Y<{{1+to_((EFuK;ieYM z9q0=7Yw9u2?;ATmI>zqjlOwooOFny3Je2<`iNR^MpV68uL~i!%Dw0)Z8^2 zmj$Mx^X)<$-y;tN19q|-d%utq`VGu)_b^OP^Ch|MA0@|}zLJ>C2h2~UQc@sZ&zPi> zD8G6PdT1D8ld~ck4!y$0%KfMgmo>okI#1aHoqOms;{aM*Q^#p_D?!bvh_$P1g%Bw} zrgq>knOUfVt&hvFasE=!eCH>amak^Nrzl{^+Z6sF*Ap)93L|;V8R+hy0N+<$W~HBZ zGo2QHyxy9~+AanQFD6@%3%bUnN?i{%EZ2jsIG;E%{Wp79l*r!5sG|3vk<7NWo=sa? zjrPAbF!G|fS}I#d&{QiCT$3k3(6=#+i))V8qNS+q5y;l;+=7qWYgm!zH&9XUf_}NK zOg}inZdu(McKEfXWOD9ojQhJ9hYi{;S^PYQS!Rm!nYf8OS09JjpVLXs)egHpYB7Ra zafZ;@H=lj|Y7H7T3d(eG{~xwuUqw}gykmFp(s!+L$b+G!{ASso+# z`cjd6m^lGX-18Hy$+2o;_u(XAN;O;@LCIo$PKPQ0qv>UKRe>WMJ(?#?X zBEek<#S0$e$z0-a!9~lnyuiJ6qJ6--^%t!oieIZ(;rC|3vFP^Yu zJ1$bbkDJoHcueOJbnyL5?$jv4oV-Q+@mEv6O70|j8J=dlB{VgK7=W9_q z@+waFB?FN?$tdHg!G2h{i4GTMnlL7xowC{~_=`N^gSk)as?0Uet5p_acP^A1yYm9) z_lBeA>N|MFK-hXwL^2o~=7DE$1U zgs%0!;dSd_{F5Fp-7D5FF|_Ghu1rBZlJMG?2PO+&S?N`P;5czPUX&*CLuKcI`W zCMe^e`yOaqr$LqsQ04cH_OSa7su*%q3NHn!gWe%qOzU?`qLaLV6nPpzRPtV2{d7IL znb+abZy8LetHPA&+tI?+lZS8K&9!H{z>|_&?2@@MEDK&tWHb~o;*$rI#_uBEmm=wZ zyh(C(%_S@<_=ua%wzD&S6=bUTJ4D|fiN58=nC&})FDX{y*=43^w(SsFXBgoDi80m+ z8-#6oJ>ajv$?GCRNJ;%JBuuQp)*ExN)>q6ToeW~q+fDJy^iDEB;|F$3`%4OzrINln zr$H{e1_E}pLH4KbLZb9G42V998B&w6D7Fi_Nj$$E{t7!v4Vm1)eyqbF8$N}cqYGAV#FEt+Y@sj@3jg z6Z_Gw`15rusE!GyA2XKFc=@q*LGrC4Lt-qCyId}buXN@si;6K)zJpGBR*wIgXVS%? zePO}(K2Q{O4KI$&VN({Bf~)>gy5Wf(uR31MV`bI&te#ikdv^@p4!ciehdqQiO*J(8 zz;VUPaR7U^_pZpy4kS&_6`?NZzNEY}3$m_cvPUxqLZMX^Y*}9~nVezEr$4LbU)r=NaZZ3c z$&Z2F(BbkqH5mEnJ(fz9pol8xPDZE6``jIv;F$*MIYlBnW)ZVgzmMh(kA?LwlX#zr z|FFE^EO&`Z;yat#A@9pc9NuUIYi6C7SU<7h3j;6mdRI*-7~d&+n?whx!(y?9)(o3u zgRA1@qT!3qQPBHG_=gdGz<#yhMPVN1H z*?uv<;lvf-@0;zQr1S!3=WQY85)=OVM9=howQqD?fgCr4s&pZiT;B#K4wpfBNNtI=+7yFRYu!;rrns{L;Y9 zWY@H#>`L}u933^0&ONhVSaEnD8|0WIk`lf7iWS+=v1UK3J^Yo`{<;8z`3Tf&AWZ6@ zAs62DK(W#o9zuuU+>jtPN5h;)<(pE?8H3@pF6AmsH-!&R1M%U%7jVhh4;F09gW}VA z{AO1X>NF%^t#cFH3AjPzGt_v)>m9uE{0kP|cNd|0JHUAMKptnZlHae}2|ITkrAo(b znP%-@^iG}0KNyNmos2_dT>O9VsAUZ||C`K}P38Dp^8q|xxdPqlb+KI9m~VMd0O|2r zAa#8q9lPBS)3UYswyh&YhtEHv5%dY{?Em2HkQV&le}-NEEi&6j%(punB1OIaZen|i zUy{>_qu}OJ2~YX3fX@%QLB|K&!=Vms%+BP5=%arP5fL+ZT46n?%y|kQ`-fBd!h^ey z>&CgGyy&1sahNsi2Rpkjnx5aBidF6Xpf%lr8=O*r%FG{JL@dyI!%J*oP%D`>M4BzI z@r9S>>x4RjggVYXkBdp|##^<2q)q0TiUVwd*_OsNq zaXdJb@C(n@@pD~6vF*necKmBLo1bbxTh&&=@VU}8XSU9Pm@kuIg!^E;b1`&v2S?y-9*e1y%T{CQ|ardi|kC7d;pI;p3$KrHT+3~CMFvK+*WLLI9Mc4u^wLJ#bicAS>(tzo5X(Au2 z2i@G-&;}RN^tJlD!o(PjYl7KVBR}%XD;nSRbKvLFb@A$#CfrfKfyW)Q!gH;U(8qWc zVB9$v_2UTl3#}&I3#~wV@ICx)T5q@M>>f0RN!%eO2d7GUaP%%MT5W5Ncjd3Z*Y86> zOVv@VjQ|h)G8E>{-3`H6)2VNt06d;^59_ZR^FG>b0B6q&J4?+)w@e~gzFrS~6;9yu z-65dCM$toY{b|FxCNvE8@0}`Z&>zlc1}#83V>b8f-oyDuLq2M? zISQklxzjpRaVF7R67pMy%h}g3!=JCV<}#!L=hHxGZi3w>zs{U$O@Z3fwA9uB%v-Ei<;85ER=Is2=fLh-WaIOwbycl)fx zl|Gz9_vk#FwSeO>U(s10u6vnfTUnN*4;k)167~#h1j(9N)IRk+G5O>GvP~ZJnN|aF z3y($LMf>54dnK8W8o{2AE&}UMi=f#17?kuxLes$ryz1r;>U%=?yY-2D!TJ<%IQ|ze z^_*s@QPWU=ZZ+u&m!qU5i)6KDLA8z&O;rYHHf+aps@>p!GX@4-JXLjTZmH;9)`6dL ziLm{xBi(b!i8>{ef?mu7Y}_uutP2{rBv_VKxFnFl(gSL~b|X*D(c|;oufVmo1441@ zG;FI~K-;#(vkw(r(C~f+$Cr6R@MH@(Ynn_gH49+1izD`G=-`sxT;M|<1MX4hA9KYV zL7f5Gv>DRKU>TfQmd>AbPi5(K7BGrO3sS0{uyF+8Y(RTCPnX#-Zef ze-5lXwGi7f7?}4NMC)fYkkA|@!D59CZ+HAllzL(yxw8vCyTsG=THis&R-W#O?N1*p z3WxJi*6fMmTJTj4g_B?6;I#ZW-gV2AKHB;j=37|9(?6T3x?41zz3(?`kB`IHo9?{5 zu|GX@?f|OQ$>5fxdZy5p2^JA2u8w?6g}o_xKA9Y1eD(fr?V(|jht_EqHi(ryVm zI1~DN%;TZ!GjMD4M#(TkZSs7e7wx>Wkba(VAHxm~L4^mmhLRH?f0 z2)^_DLH6((hh0sF;Y_>^-LBRK@$4+z>exV|Ix`@a#*qu-@*#QbDfDmF0Ig#?*xpYy zWTT@SAu7Yc{hzDe{>?}okWHqlp`j#7ivA<&-qvFcN$9A7g0Jc3<@gys#!!{d%X zNXjstaZ`cV)|=s3lXuwa;Q{-GycQh#&!tBUuEFXblX01LEZgjCYiD0Qi4+FP)9?Rs zV6<;AN>_~FThm=&qtg@k_izl)PniYZ>X*}mdU;;tJR0)X>+zM9i@8&BCSF~r$xHnt z5WaXJ#0_bNzlnSKrbA+_?wldd{Q4Aq8ck59@ECveDHUvryhWF(Kh-<<5!^3rgbQn; z=-d%SxMZT3S3mATNlzH{k&BaLPpiXk6?(9@M-?r9d+{+rI?(Vc1DH!R9QS`h1}>83 z!yYSxZ;vh=f4`QcFAKm!E8Y0eWqSOP<7qtmV=;H1HwFxj$#4@fXPFWI8-I*B1l9ZW z!RgRe=z6As+lozXk^`+mC;2UFiMty_oW` z1Ey{fXA+VtxS}|7os~0*4{mCp74v-2WWZ7}C%2Vf7;zB3eX!=r-Q7%gv^;BEvXib) zJ`7LAo_s>}KU^{849WJCK)iARjQK8OH%b3K*6qoMzdZ}ss}ZrpV%b-a4m!k-RE@>d ze@A>M3iCJPtp03wV9H6#stB0rzWIQm)^EbLwnym*{-8|0_kmjvCHwHSFQb z@I5^LqXN5L>yJa1n^N7Plh}K}8f-#4B^?Vz*6*Ih++cJj_imoYN2rzYfxrF1dGJl7 z3tzM7jpp1hy&Ar3@`bK{`qY^X!H(?*@oh>PZyCRhpMT`T`%RMPzPylii?d_XmUqIH zs7nxV_bxnLh>GoN6xbROII0n_g!$oC$N1;-U#;o=1^3jF- zcfkcz5obtNeyPQu``bzDlqA~g89|K~x0A0kufpXT!Y8R~@Mk7Y7(CJmR}I?$$>Lmx z$^lav6?mICR65h^Rf;sPwic6evLU$4kFQE}U=?ywyoZF+F?Dj(McJEn{FoxnFi6uA z6(Pbr8PNy6uZ7Pr9L@7KJ3l6ad-|GoHe6YYt8wXb1E=B zLZ3$rCZJyEMuu2@msq)cWTQ8BqxP0@+%k0!n$lskZO#!;Qn##;x1NqRMkD#6L-BO( z^Skh5`5IU-(}eFIbrXz~-x3#*zgxdD0wPK&f0>X5K7lUS5LAx`(U>Q0=mHDncs{^u zDBUk5_QpFFN!DhVqGP&PXZc_bO+Q7)i(dpBw*SoQg$Y#GJr|A^MIw`t=O*Uo@m^Op zC|{YzhW|K(m+Y0e+ITts->x4R+B<+sS=@l(^UbJIa6i7oN?b#4Tth2<3f9)%fr1H9 zG)Bu9O&t5tf-ZM-w|q?QKd6G(sg9^sc!vC#)fcBNVDO*GeP&l|fRo~*Xi3pgmORlJ zo(=ziXB*OS&e#n!b@XRq;$e-iUI+3j*_-I@F-o|;Ih9)tW!A*^N=OnTGgwfb>?miek>;Ws?i|D@8 zff`H1|MLxXI(NrS+MA_8N`I)+$ewf>-XTZF|1-iOeNTS%a4HGvOreHO|5&z81KEDC z7>-LF7EE=PL(NQAvt(zKP2Y&B|&wu&@I+a!mpT8Dcgbc+E~s` zYz&0~4)*!`c9%jPTY!7oHR=7Avh;O= zShE-<<|77J+K&D+2URrdAZM~2Juvhm`BVNIU$qJ#f8CF|wl;uFnjD=uE1RyUbKu+F z&j;s~i;3pUd~q-T5z{M@G2hmUdX7#aW^wbx9`9A`SuvMH*p@@su$6S=#Jj{QJXy@V zXu{gNHn0Om&^fOZ@uN;B+^!CSwf>jk2%HdPE?*-gyQ?KWl|8;QH_TY85&rupC~btiv#Vh zo%E^YH8SJG1<`@Els-N7ijF0^w0=#z#5PxzI=ueL)RyYu#jOfdeZfpf(J%T7`Q zRdxEMK$=eMJ0DZpQpjeX*YE+7V3(3M-P~FyFbkZ-mhDiX)DyVtkTrAKOET9^%7 zn{7tqi%#RZZ_jb*JuwfN7>Uhx5v=jvC5T)dCAt;PNy5`T$U9|s`lV(L{dUJ5uEdTM z6pTvIQ6mwaxBq5?Ew6)hHBe^rL}W(qCdp6I*hr67HeT+Ku>Z3YH&k7+WICx23#+4lf-VImv|n z6FCMYmz#u4wE$2GxdH>LqUqD6o?NY21l9Ll{zLre}?pqIK4{anLB)Z^caU47H=b%Z%C%Xvyn`F;IU9x8NKRh{T zC_i;>E|)#G3gSJM8HL|PIR z@7ti#Pm1Sw4C3hz%7t~$gYezSMfk$=9)2jW$Cf}Bez-e zA6CJMzgyt;lxbu+X@;)AE|w{-%V}{&5Pv!xgUZxt%*tAlG9&^%HtaxOwUdzlcp9HL zK%K4HJ%Q^FjpLqM6mgV`I}ga;&BMBe@>w0T`IM~&B5 z#}u&5=`+B7QVf{N^e_d>(LDX654T9{0H?1PVDY45xWgkI zb*IsK*$doKMx4LdJo%Yc%V#DE5M9AW-j&{`!tDl!k8}@9=wOwuytg;l>tUK z_hXwrD)4V5_B_(TmtUVAg&XhPV$qdy_^bJ=ZP&BOc+s~7A6KjM8OP`1wok{<{-QGJ zGsqT~Saq}a3kLHpMHznk?|EE&={qhzATp}OdgRRPK9Xst18}tN3@BeFO$%3}=+<#! z{qy8OdQKCm^t{5tuFA3%$7kWM>Dqk%6m=Nd)+i~dIg87$n)8WQUx4PCKCnM{4A`9( zGihU#dDQC?6gwao-X|UF;{KQEiX#((&AH5(?QkeTi)u|*fK&6_pi6Ey*5>KLrAyyg z$&qSud6gVa&^O?b9*0oldmb(;ngwyzKZWFftN693a%iq!$Lf8XU}%{GiW+cynCwan zM?b-SgMy&!Y7D~21!Csp4%USB#WqKM`n^t#XT{xN>RypRed@71n?Q8mBnUGIBfocU zz{BfA58EC;ZegFn)|l9!YUD89Hti6JYe->7lHBk`_(_~FbT!o24CL4m%2i+4uyrGQ zSc1$!jC&?XZVWiUoOV~hu>9@9%M;;9!y~Yy+Y=s*6BCJXb0M--8jIYUvELE{T%(DS zwK}8N*X!HicxW)|T($&0>r3&Y-l;fjt08)dJl-#c`?y2DXe?#f6e)I_2~loh;ES|IsZ@`7|Olo6fTDZ<>6c48^^weRXTz^-_GJUH$iniF-t zrtu`oex8Us?#!VVM27zR6*BNMYZ~+&aGR(G`r&Mc9N6hmiUB5tBDX#XKYiki{uxN4 zEGx)y;{&MhyB(h!K4rb=$p^3hfLWXJ#C_XpmgO5wo+bHnyVfau_L$+M#35DUbL9vr zcRS1`XUPf5! z{RrPBXl~xYmSl34I@K0?s-(#Jl=tlEk#u&`{2ZSA)LMPJZW(&oT^EBaqw%6q4VG+6 zl&ojViS+GQ_MozdB$P$r?ICJ-e^@1+`HynGD2(i}T#L&@pIU?8Lu|X=&HN4*vLC8v zgaorNX0j+*@;<+xaDkVJJziBpNC5`Y($bp(ECAzX%Uz=dp@gJ%V&*4`>~I0N%N6u+8ZvxkJwr=k&eMT;5V$ zA*YF_mX9alx}V6YZxhhHz7x$f#IjXSE{?X6q7x4vfx&_YyqtYN^b$;DzXv_SY~^4S z(?d`>X((IrPpaA?-ihs)*+3fGqu5?cULlJrzocQ=n-DDy@CBVam?G(Or}lU&eA$M zan_&h?1xb_6Lsip`7%XnAm;FAJ@@4UZ3dIlsYa5>yHA+C#Q^5`uZ>w+ykmN^AClPo zOs1cHUC1&TgV8Fxu_{nYWKm?YN=(9R?=JM(tivOX%MoM`W7xlHcoOim%I@R~mc~zs zPJ1Ufo$;0R9(@WiMT7Y9*8+=+`YG6y$+NESE^KOh3a-u+u(|&(Y^^;5<7@g*`O*D> zH|=7k{yQ*P@)g@)0`EOBh&TC`kXpBL$?|n4*|F0vC6{Br6T$crvvGVRnEf%t3Gv_9 z8-sAj`gD%zK_ts47$K}ob_Bnxp4hUqkl2@8B^yE$Szv7e^ZGiMcn!EnG|x!k#*G7* z+Q&wrwRr{0J`u>yKXqZfHjeC2yA3L)T!DjXufgxXOg2-aiRB#L$E2!%RKFB^qZ!*@ z*quzaXCsTf$%;8_b+oib9^HPtqK<6&wXY7^9xW@F^eP_IN{P0Z`tmF zugE^@3=D1Nc}#Z0VsvGor+1pDDj1y}vPlK!{d*}^v} zlB9s8xT1IzhMri-F3o-}X?dP!C(~R~HS^eXTof3OiSPfh=hDhJ?MOc%%(q$6vosG! z9m>V1=N>HTp*o#8I}4H(hSJk}<51xm!T-kh;}1_out`^g1TFvRtRL5f3r#XOFO`i50kSL^4&=QtVgB$o<&S@(?{Hd4lXWo6M!GKen7+ej9@T+02e5_xpOQhu*L zqVK}XFgB?lbRJWz>HoNh2VQ=}eH>)zxBai-WNsSOTDcg`I1oC0P&RelF_E7ScSDb9 z|HKT>tm>ytQcOzGUv!)8;QiL+;Y-t_@VIRpzbK`}Z~a`x4{!Iyn`ds|g;RSmPHLyH zogU5Q@aQ+MU`T-vs=~3kSoH1f6~IqsWae_ObYIq_!`G~Y{To((fhTfF}Es+a2OY05Q>Q$_|aC?^3Ug-xmw;bz0cEl1gQ1bzlu71mpR+RFT zxV_x;j}`4ByMyM}pQVbmMl}0PDySK{lg`m1V}>7KN=6RM`K%XemQHx%eEN96+{@)~vV1p1SV zAs*A~lWEwXf%G9ANB`PVe(OO6`!0P{c-~r3t+RJnwa2uqDpdW47PA$Zg2PyR`kxcN z8h-%hhL50zf41O`EgwktmR({F=n0SC8qGV0p2f;jw(NAwUA)upHR`Ci*!Vv_V)>_^ zV%24fV4>3>hv^NxCv;1>6RGxv*nBTr(sfXl-+dU4FLr+;rAhuU;)A`AdU>U-X6s;f z{>u&aM5~y+3_buK=C{DC>U|J2VL8UCf5Px$(J631P14-YfXF>_MOoKJ5{C~HVR_6~ zk|niUaQa(7KDfCHPrEI7Ve1lHzX701bhB1X)`679T1myH`y_d)61=MQBl4R>7o*%( zbl0lJ;`gd}Np&?U+zLk3Hv_p%c06Wg%>&ooYs|Dq4+47U!~HjXQPtUy%@~{|?D%g2 zQ~st+9CsEoFyPiaVq1kq_lSyHnLC)T)p6mn+5vWZ8RuQT{cFzWuBr)m8d9Y`!;wX= zloA$x99p9!ZCrCCb_1VNG#36e-(|%slZ430_1Gu-CR&sp#m)c6(0TY%^~Q0$gzUXX zAtR$yxc5BILyHCx?LleMrm2ZSqJ(4=B?&3YzUTQqR|zc|BBS*yEtRM=wELaE;lA!U z=llJBKJT~aIoP_nF~#Vi(#F}}A$Ct5#2&Fmd0n+kU#^)M&Q4@DDz`+l_W$A2(w0er zlB4-GzlXy)_X*7G#4z@5)I->^t%APUcpN4jwP130U$EAVIm~jCj&$R(k00lv5DfwS9XJUT*w%FXf+=FK=L4?hch zqu_YjIgx38bi#E*Gw8B&Y5cR^UKYJ{x%A?HPEtFuFB{ybi zbx2C*DfmKQLO1==lMfN;h+SKKifvdcWnTF!8LZj}4Np+0s-op%$Kw1xjF&F{f!@AR zq>B%zg08j?_{9S9Kl)g5(u1#>n6ZwrOWW9t@KnhOkuLlFz6rLSRR^z{aCCN67Tvn? z6xv=n7`d!V#SlcG3>X(}?x+mmCE0%fD?t8^- zV{bE)!WG7U3gM<76pIZ{+2N|brDR36GWn^{2)}c8NCLx-@(Y43IeJjd)!t&JYwqfB zSDyu=T)*UbxUM{T$ko8#0@7ob!uuPa=}Mn@%;aMssMp_TBMziX?v0#;)XH&XzPcy+ ze0&n>$;rkqI|E%8j#}gzI_wu2qtOGBTAp2-vWOiIe^1cC`()tpJW{O@O-H@*mc-0B zT}eV_Lfw{q@JhcQZY3^67v9G6CS%>W)D3?D$IM_IT2{dnM#; z!x6IWa63u2A59jT8(?mE8Eb9NWEXD9unR}bAk9CC{+a$(l4lhw{`P1r*%f?BF`T4aTjPmkKPInK0y^j%(eu zKC-FR(lv(5p&#$-qpq4u*t+Hv7IRymeya?gDxC_WT))!fmc!zp4}^Y;T+FfI5BNo* zL#Wi$i|stq3p(S=aJOF<8awMXUB1NF_15xvt~vjWb1l54$`bTvv7PD$Ov_#@&M#S1 zc3hk?x@lh*EEwGhY4u68U93REV#hH5OI0j-;}DoJx1P3X2hg)q!@Wdyj{fAKuyFa%!Fq3Aj89hs@b8 zbTvnluO}|x%j?GApT8|p&z}nXG*%M^+D>P6-g&rfS~a;=cCh^RJ~tTE5`aHVKZvt! z7VYJ(OWKFDzz6|xLb?I z8Z?9RPRzs&7W|X{%5V~pL zIm~qWd**I4lg`q~;yY}V@hOwlXl=wJ*0-*KRWDX$fA(F5Q-5^{cX}BPo>h<2MEWHD z;T783c^R+BkHi*gZ$YZ!fNc&JqqvnTNWGi`Nvkdw{1E5x$(RXj+Tee{##Z7jKmFL= z(CsX`WjxCk^1}x|#=^26Htepo9$RT)hl4NNfi&EN<$MPbkKTFYNbW1Ts`?xJ7k-qc z?fM4q2WZ2WjwCEE^zCI@1@IqF6RwI_LCXu49FG&Weefc-2p^%X*LN18*;~OGtr1hBu;8G zxcO@P`OT9jLVH!6i&x{)iWcYg%3{A{a_i?dlF}iMwS-*M=~t#W>Rb)hldOV;yY-3I z!cf>U$b_9siUW1aNGaOgp?OEBvLyM%sMB&8y?7c+VNW(L5aP;U$;=LlUCe=~SydIm3=*ob>-5^&tn z474eymd2Hq;p~<9oc!_S{KiGQaZKn@81C_i8}^~6Dq^EAI=p-+F5LPS`K`soZ2K>! zwe>DTzcKrM8gmaXw@Qxc9!4+3JoHT633rE{#K-w_E+-63QEWmPn$Q-(M!M`}DQ*W? zT+jsRH*Hht==}N87d8saRIiPCyY7P#7nRt+MjPf|=SJ3-Yq)eL*h^xiv&3(`9KmXz z1@sAf3bki-OuBs%E4m`jZdwf|w_f^^Q&+9=*RjTU&D~_=Ip8KpQYjtdI8Brm)yVhJ zVwBlF7C(=(!>R`m*E(?|YTQ&TFtpp@XqyWCu`&pT4NVm9yO&PYu8YK5Hf*7dk=4RC zaScoob+FDI#%z(wEcmb263;C?1EzEQ#YH}6P*%)h+}_}hcieWQp7!hTf%WB(8t=rs z4P%+2uPjrl#O&#n&jKe(pN&~Rfk=Dj!lUuJY;e#-b}V@q`8Qad{JiVJdi1xm(^c_| zvDM^Cfg3s3T!k0QG{Q&Q8SF`}5pxncXY=BIO6G3lxgkqi`Pt#SByoDz=(sfkd)2@e z&n|rdp{pcVay5iBUP^-VnWOQK*KsUx=?OOU)?W5%>1`-z*$y^O5+$39CE|^Ri@5nA zyWwfsLxHKR0hY_&!ml^GfnA=1?Ow>Dz1d>u>k7dN*G}U}6HDmaBy|R-<#D{0A0Ava z0h76!B}c9gahz-%1m6PpDtTgX#x=Mpip&NG44 zs1m;jo(i9GRiRV3pJzF!5dVOBNRTMuRe^i)$-*Y|I$$C8dT|QZ`46WvWIV|i2M!i| zdBN@LJ}>dJR1Y6^7t`kUR5-%g|w*6iGL*ZL4R(xRi!ML4dXt! z;XO7prAg*`(gU~6q+DjmT(yt*KAZqt}iNkBz${P=2 z@zwgDT-BG6JbT(e=iJ*!ajO=#eftp?{gfwe(PPPvzxI`XcU={aQQjubSy~1gHhzS4 z36XHs8o?eT1L4+qA8iWkXVJR^H<_Kj^w4Z2_?6y-h^Pr2u}s7cNj_vqRSCvEhmcRD z9)9qm2i=rWklxrdl!azGvp<7$n0nD|JS-?0c?_&^?mj;iR4+DC{|Q5-qYH1bw68x{ z)$tz!ORA1l?{;S|^Ofix`LAfO;zY9M!5+{xJ}u;%j3M%9G`e2%gg-FYN8)&1A5wo_ z=Itg0KzH0^@OFZsop;i3 zkdRb|a`N8w1*y`CV@DLaq2J*Fe)LhWY~7F{UwvhW!oP#WOZG0-do_+oHh7W9d%tn+ z)MwamgKAawC|OZ>)qCeG&15i>Q;{qzy+@G1{F-=nq{;K6*yhSb=CUnJ;gvli&)x?JE zsWL&d^%^ZPy2i~=_r@!Aa>QFhD(KrgTD-2uMM*&DO>use0hHTD2!8E4s0-P_9PGRB zp;hCsR`y_AvNiYRj#2T~a@W(B%)L01w3H8$O#NXb3YVARk^;4uQHd|R zKK~%DHowBgd){DrHCLF-nBi;E!h_CZ{Y(4iPC*xU;Z~rrZ z?He38_vKS?nR7m`-7{aDGA|0UpATm33w>C*aPQmQDJy`~#0(8?&_Gdsu336q$O^kBr0x*r?+O1gC03Pm2Lw>*I$uKi`MG?+@c{ z?&@)o$?L0dcy?gaSbGOhJ737NIa=`R`cC-wV+iO+8;m`$p4AnHGoKTWS)Mt;?sbj$ z^0gSWBjzQQx`aXF7!|ysFqKOhF^9hq9WQYe?#fU52n*TDm}Oc%q&EqEOFn zalI4U{WO`~*m08GIeU~%Q24`iR7BD((+WhB|C8YYturOBZa>2BYb&vu;8E_eRl&7) zI=J@1epnJ!4Bjy^I9JHG-yJXj&iIZ7>s_rV|CKpzd7=l)L=M#CzahlLsg8QqH}h__ zz2#l$*0?tE2r6Amu)+m3B6=rFFpHCH-Y!%1<$x-$d0vfFTRldvuO%bdrjOWc@nXm= zaAwh4p5mDq*KpV6Hn?XSiN{~>;M>Z-anW0Ik+SU$Z2X~(1vqqNxU&E`0$5XufL5Ozx=`ZO)X-da4r1F-yfWp zMdC@He6Z5je!A$5z7eriJqhf9jQ|6GhYFf&+nqoSDBGl{Z_>P z@+xAwM+H`x)#H#SLFCD}GIV6~Ot`o-0o7b+6%D8%m08gm{0D`6fo^x79lzAZhTr;y zy(capKhlFq)7?)X@;?oOXBx8OJ|0;7b}#-gF#{~pQrLbk4e<7v%Z{8MSUKTWfvC)& zPSp8+BTn(W2W-C)$ygswLc(K-lO)#-=Cyublp(mw+!I}g%@Sskb${wkO1s16-J z&*8i&Rq$}$2kuu%CTtzP1?wFTAmuMS8E4jj4Yoy+MW^F%^IR*i@pE!rdQr}G@Iph^ zH-E14(e4qDrZ^UJejfp^C}1j1M^+rS58z_Ju6DbO1dYbqqh} z?%{m`4^Ed}Lya@?+4J{z=q261sP^|oT<36<6s*`!!lmkLctnhAsml^qxe?=Bcik^R z#a;jSETb#9GW`$fxcn4%Whl6|`PtAJW|!&1*e>LHRq#tTs6L598v#WdWb8`94 zb+Wh61MDLPk`>NfNnBkD|b^A*7`7BYqXXg}e;cf&I7VkypnVUa-ZIn0_8f?-nMLj~hRc zeJ#qwC1@u5W}AT}%@djZgP+uCwH03TbQ%^L-xGXT%C5eTN0EQ>ZDfUq0?`c!AXmrD zAg{bG<9!b%b#7Ft7%#nY^E_}Vk zQbIIcmrc=db(2?jT~}KqQGR@+YJPAs`YC&v({~oU zQGs2S?F9XGsVL%I7K;6G1%BIqA(m(LTwOa}6Kks_cUNDdo{9N*l}IZXu*#6i}7TOM7n9WJUKnkigbb>S{e}v zHurx}TWUbul$-=sm>k)~qB&bVz$K3UWa=|D#LV}@qB|c@(9n2%bxb}E&%TW}9`@jkqT%zQDF2Ue|9z4BpR#OvgH;U{Oc=W83)gjVYp@ z-9iuU?*Yk}PoN;0A zqw0xtL-$j0*>*j4y+0VL7Z%X*Wi=?kV+vfjrO3Vq2IAmeJC;}S0%YcSfSS?^y3f~x3Cfn zg-*7rsWN26&BYfl&%j|mKjHrJbM$^=JPbeP!LEl~gsPA3?EVpD*c%WYtF=(rbqf z!sD&NZ2#si4jdFsd!?VT|D*n@P^)HV>y}`Cr{K<{QBEY=pavbkejMgMH)qx{j%=+$ zGOi7qj?%X#a%&HVv-gRn>|@h6m|`XDMm9P^u*zo03-J`?HEb3Y>8Xm;D;klR&LR9| zMJs->_8heRcM9J!I7qg3g+o-nBlF&IoV^e-q+F;f@AC0#l~0Zl`Q5*snq0fZ$KTjM zzUc9_#DdQXpLmf-9YcW{698n|RS6OY-NiH+VX zlA9L~z?=ku=efxXCvVwKtq%J^+8+xz;4mEK%o4MY-a~Mp&pR3s|BtW9k;kWAd6D(2 z#U$Bx7=3+m7#>nOQ0OZBz)cE)?Du8ihOe0@aHluo(<7p2f6+KJy*L}qoGLh1+k^4b zo#!f(4U|L^Qk%tZ)?a3=t;5-@4TAePOplpN31!yb+@*2c5WI7sA#~lDgJL~RnEdoK z#;>Sg=5BE?=-eA_=|)?4Q{Rej*34#__Gg$#_8WPjc7}Y|kx2(E3j;ZsT59m&D?Iy@ z$cj&TvG%6(H23!cu)6OfdG=Lc{>(N&bNfp;ceh-)&@5!#@BZdh< z3u2!>Dl(T7MzB;fp8cCQpXsNqVV;rtY``%WW*c1w>b(nCLP!F4Sa}uS-%?6{cJ0SI zhn?ZCHOJ%h=lvo)bt2bfnJGDiV{zApXu7p$5>?0c_|fr5%IWP9-(I7OJF}bcn6DZ5 z#n^4Y$HhX%g<}wx9?b1~H;Q|9Tba$7_ZfSuJ;U{{En$c7fB5R8DfSvzL^Yn~;jKzO zWb`64@Vn#^gX|UsoGVu_v=iF@W6(i(_Az*X43ngt$D005n zQzz-P)^s)u8NiAJH&B`CLY9>>pAGou&N>|SvED7M^y{Jbw04_{r3-Lo6ehM5Tec(#GX znFKP;pwIACZnNm!fm)YcyQVs4dVWCF%dNma8^ewUId~DAPs5eVz--@N^l0RLDET^! z?Y|w*o%g>0Da|t=Q{@o-cV`REoTbN`rTB1r5<*dryBvsxhn*8zVdy#)awuPsynmug zwkIgFKcoIpy$(5cwnc<=ZKs0Go&{p=U?`i}^#G;?Hp3k&ZSK$QAif0IgV8iq{!rvh zG$lWsjvZZx=-en^(PQCo*A4u1ODx_pU=W%oE#>W+jKx|8muOKy8q53o1jg0hg%+WE zS|0w0bDs2z4skvMTBU*Ds(qK+PxdjDzCFzJ&oKtYPHf%3rA%R;GxLhIV?PUT!e8xJ zw(ycF;)xlaJtzX*HcJv4M-1bSMrKj9;m;-ZTI>1B56PTSmL=}z2H?#Gb<{&E3+`_I zQ&ro$m%d$h2tDO9>0Rm)XJhG->upIS^Pl70NgGhW zs7CZ=5XSqP)zP@6d+8E8fAmV{pt!r3M=lHgtGe5qj|}@ZliFeIxYDbHUb=MS|yX`n+rX|2#kXPKQA1? z7R&mw6+U)s%V;G$JH3NbwOt406AeJW(H}d%+r%C3c>`xgZ-w|RZg`f}8{BalCsxhnR)T}PvD^7u>2H1v7%NV4qcHyHmnj$NK)!pT zL}0lqx%6YxX>e%&3nRB`!t8%rG5q{Z8%`B~&x4O>e?T2BN^0h>tU)+J|1Z}z&Icbn zY6W`Mdg3=7LSwb&pm^D_7*v__RPy=_7C$pT#Y;ku@`sldLG!nFl8Z0~6l)&92G=EU zUO5#C-5!g(uPMPuM+tYp`j+J0zxR0Z26x<^brNMw7UAmjC_M11z^0HhhpgaWh)gen zfx5nUdvy@@*E7YoDu+@2Uk4=pvJmJhLozIA7#Zwz3a?}ZEaF`ut34maMn+tcY~S4{ zo>&_p`CS-5gI4aP+@*uib$>mKHm!lxh1&SV6(LjhxeI5XZ>HOSi@|E=J`N5J!~1v6 z#$h*~(lKEVQ1p{Zu6tY*H&17a_{d~!{5s+@6g;}e|M5G`ZOl)^MQ?bd^V|epwQa$s zi{x?F#2ox_;ds*UyO3@*tDxSozIb8pLvG$UaEU(PR=)c10+;O91?W=1QA`u>KvGf; z+wv_A+HVKac_ZFHv$Y4)x$Qt2Ut|8bP9{Cutj|{^9jaQDb%-0We;y8XDduLq?Qtn_ zzfSAayb*H-Dj-kTp>M}{=WQMW^VObh)njyjnK1Vz_EYtUulVJNdpLUTFjy({e7!#2 z;x^<>lzjQI9?~aG#>giXrkeHBK=8mTerdv@_Y-k{e;;-`@fhbf+z0z76WHGyDiF{t z%s)2v=y|C=JQ!X;o6?tKXA>{1J|T%(&O67cPuPQ{st3jDXD{)4L&u=r^0tbSh)5UM zmf|A6_8*S9OGve+;8!vq4`VLIvaqNamgN!vW~YWx*Vm`TiWx`QTeVxl{dzO29yb}) zZ)}ygyMN=xPS^}T-<*fLF&gam#2L&+>cXBVP7%J>&tRpIDzGeMFk~1f(DtI&SWh{O zh?F0S%4)T_-NjU*QgV*Q`BZJLviy(_VNu%PwJZ*qciW-DQzBSg{U5#d?X2)Ssk1M8 zROkRupk9ycX!n4xP_=0WEKPd|F$)LMbr+5Cu;I$gYwuYw{^I|8_%1c|){8zMYFVTIFx>CAp69Y55s==zrbG=|!Vc>o4kTPtK!ZI@`r1`o&!Cs0=(p`!!DA zy_Fo=-Us}|20YoL5j_vNAeL31DjqJU%cs7!fkkLGFDE!drHRv#8E4H++n0qVs63(h zp^tIW=$ZIb+(KkfzZZ!OkJHSj-Kwb!b2uT(2wX2I`*a) z2PdT9cQJw|enJG7l5q(??ps97G7{*zt;_HxKA-L!_MAFy`HA`$jjwcEGp};-qy0{; zGbchp$7q5qIlOs$D?TOZ2D74OTwb^wmTMfO-^`=w4WW-Qv2+}0H3~jbg_^2`PR{&% zFn}`at z0WU=9%QX1r54&iR>vX)M(U_I(lttHS9zs6*1CjG$=u{6|ym642{<~t03WrMJPGK^x z-<^zC#-FA8{o+vk0TpiX;W<^77IB=8hYY8Fth`ENx`8Aqu8-S`;*ivFKChTy$cN22 zD{(3~El!LbhP&j4VDoQV(V;V1H1XVY?3Cw%Z|__Mfm%UepP3=N1)L%2-~_ZhUvMEW z!)W^w3HOz{;_FwWwDf2tu6gzYhuJNlmjdg!VMg{SX<->2`Qs9uw>_4|tgGa%O+AO? zmMcL0_QR-Y@f!)d)roV5_ae0!Rn&3EcFLa_fj7p*f$6?MP^EYf2F5Dmz#Z4Hvf?bz z(R)HaFa60|nyAs}LL zW{K13ZEr=o?D_%zzUfkQZBl0CDCI;+qWBjs4?fD>?UQmr!BUu@8U#C*YLSnxqWGb` z*CoBTK1iDKwNS$njViy+XISUy;qv9zZ(hU(1y%X)KT@~K9JC>F14dR8nOXF7W`AAi z8h4z9k<*ogJIyo}xIL0y=q%?3zg|O&PNs@Ov?}?rl90;uI~mY3U?KcA17Z6(9?;Je zbO6WEdmm&_{_ZCz$m<}B^K2AddG&*|JYVjG^;!Xe}5m3PWUJB=Gt;t`NqofDACu*;EJUDOkb`>82EnJ5Z&5UolNg@=D;l>g zi?4kukEWy*()6QYFfk;bz3Una(`xK-ci(haxa}o)Djo)hXYuI9I2&5N^8vosU0;?MV) zPSVUge$RP?mQ52pVN>Q(8EaDrZ&||sh)5N8$}Hh`J)VJn8fVa&^ZTJ$G>DGUdnR7z zbDt~K&*Lkujz#^i4LE~&b^PArC)kmDn$l2TbLsCR?c88R4O}!c6`j2yBet{4M49o~ zka4!Ts%uvZ+Vo9YRXghfywBXvTfMagmtbE>uY0IuZKaCHV^sy@CJ$u=e^0^A*Jb$F zi&$D-A)Qf%Jm z%IS2N@P5mv_=^92$U1)+y0yQfDXlu9{nK(qbG%2nlw2{yUj>(6*l9!j;LJ?^t79De zcU0gscHZFIT&}~Om_cl~j4tF_K04wU~zTOBKL$Lsm*tKbOS zYB`W(+Rr3o->49+|MXDf?7OJcDw?{kIZo=Xsxt4uW7tyTGkv;K2@2P%3AfeLWKil{ zq?PPUMr=tZqxkRmN4Gcbv@XKwD>JZtaWdZRcac*I_QDflo*<=J?&8<0`XwvQzjp~y zx(KM;lU>pI%|>ao2=~)E=-MIh*)D$KTVGw_Bz}vL+UbdGe5Al+n!k*(9u2f+%xZk& z?om88K#j=DEW+Ez=(E`+iP8@bW=kDL`$!*~4q?_uC|r%W1GoGKV_$g{Xgn`s7yV~R zdW>9Y+9y|TR^KOdaPn$2;$0V)yV;!mcUg%YpLK-|wBC<*54%n47nLFpf%}uXwT2}s z=dv9OLgBMMrQfdlf|cPtPIdKo7oD;c z+XdVg1v9?ep_T^MXhDqU0QUI6K|EUV3FniPLE2L`!C0ABG+}HI8~OeS`OOI`~!ygT8*{l~aZ0AEiHlnZ(q8jQT zC44tD=WIu#i{C-wu*u9sS^{gXpM=ZZWdch)059IBNCtM);t7L{fhKn2_%)yCnzrpY z-{J%w=${A_HbGlr~_olZP64q_!DDjA|W*x`GH$e}q*@KGf-p|IFtWa5jPFv8i2Ib9(v zqeFq3tvE>+KN$+SXTI_=rfX3+%|g30wIMFEf@`&nf*WhiNqkEv@zX{$J;Vh|Oy&|F zw_^Mu^bn57EyqRzn`f$=;5f}nEJP zMZy5n9?LYOowvc zGXHYriOcA^_1N`hYx!MnK_|3ivV95n#kM zn!ji-a*LQrcMANHk>lhb*rFA+Bn_Z(jAs*`ll0=oJ~AQ{ix3Ej9y*okgq!`GbPEVh-=>IM6GwbsvEV67S*^vkzu z>ZUhkZOaTQ+BM`z)W>)-w8D~%k)25nj*KBkS4}1-Q%b-V4`Z3UDg3v>pG3GOlDKmh z$o*Krnulh=CeDMkkN*wk z-Um64NfC5(KqB=EdqhWHj^cN(Y~!u(VK(G}9WeJC%k9Weyi5M?~|R}hK|HzzONad^YZVi>;o9L?A}AC`;_pnC+CRJqOs z&`I16KZaQd?6-WbFIE@zE;_>N`74tV+rshT1G12Av!9;z_=bvY)^Wy{wNU*IUhMAZ z3?4q#5G9oAw{1~`+n-xt?$S7X=a@Xsx>A7!&?w&j{V^UFR)%&joO#-$6k+i7YN&NNz-G+X6zgT2OE-5^cEG>tYXI;E!jLH%=&FS;YOaY3lZz$n~#*ax8g?e zx2S32&l54(GH)+Qc)pbMU-qZ}7U{wEY0r7LPiNrr${6rXmtlQ33hDg5pLFZco4Eb! zNn|&IayIczf+mXX2m?h$|S z+%dc+<%DSezN?&)wKqC_Y{NDM@dt@N2v)aclh+eWqfk*iMO*6ToFIk{lB*xNZCM0y#55Dj69N4Pi%znaC z$(c%bs$b?pM@Q!I{d=SMB+W`*d(T@u$HI%82`(XF8-q~H4n2JM@LgQ{JB$4@jwd3= zapZE&No=Q<2L|Dlh$R?en}T;$v zE}3Ba9O0SKJBFD%Os7{*dC;1S3VbO2D4VgRnnmVjgI;oQmDa!t9P{uHT-8%!14}cR zhjIa%^u-HIXaC1;Y#%GFHkpg;HyuPnf7GMv|2-i0izMWp-Z0m!(s%qeio;or7i>x`o6`#vvtVzmjmggB^s5M z4rcrl%PD-?l)VzKlf#ko=|FPsjwU)bG?$Dt&|;IH8nW5G+DyKAo8(=GSyk5l9g;Uw z$FfSkkF|RzGEcj6pt!HK!$bBhjUp8Woowh-7b1BuIq)Q%;i-T$L z(d3u99T~HS-Eb%ejWU}+FDq!n-^qg4aNIz?J>EdH z#>Rx5JTAi?Jl(=dMx25pA1AO83uMW+t|WGplaZP=Nl;mRmR^|~&2LaFb=gx> zEcu7+Bq6g4$&T62@taw;rZXz6RzO>IuAK12%2IU4GvC#YB5p0~LR? z0oo#Xv(NuRiv-4Xy<7@5*ja<#7DN)*9YY3<#^^(5Aj{2e!G*^HaO=3kTwdj0TA6Qx zgQOQ>bX5~fxc&%tZ&*$8BgNR=Ujfs|!#DyPA%*Hs&~Seq^VHbIuG(bqRdW&BVWr86 zURr@)S~y--_6WOqZ{#mnDN+A714uynEm*U@8(NY~Nqc?_t9qrxYRB(l?kh^!hmA38 zrP3K;&NZb^ch^=$a8r=G$A56Er~=c0spJyx{ z;0wUrcR0mAqgd3QHkj$M7q;JdLx)auVE%1SV7E;y3^-VW3hukGVZIJ<^WPx+Yj-*} zaTlRVWDFOh8bC$hmb=e5M0d1qfvz9MeCdgI;Q8My+EQN*O&xXgkM=tHIWZ0^H2;JA zx%)sgJDTY|oCsG-OW?8GBbajG1wPxWj$JqB(8CXOalgwWYA=6`U+3r!UpG!+=yn>6 ztE}J+d<&rEteAb6@ej_Le?{jt7ZHVNNhHwzHyxle8;k^oz$~k2;*Vh#{PLt0WJOY0 zS4|+R)Qx1({_3P|ga+O{{}Jy$ZX3KVABqS2W|7-FgP6v)Z7e?RIvC1q!*YwG@X?M0 zE;Q?1#hIb*RfTVRU5auoAv$X;P0Tgp<4TiJo?II1bovQYr4a({l4+xi8B>{Yk^6Y| z8P}rZib&Eo-ZZiuC3kw^`@1s0sOb!Jj(bnzJJm^rwmf<1Gl5K!vqieI=5e+gukrJ2 zZ^Ep{hImfE0@U#12)yZD&Awd}dbI6fjHI7pCVO(&fNAcm^Kl#t_dfzI7BA7X89VUJ zZJXf4v2*y~TY|Uy--PlK4P2m>2#XD3s7-kSWSyaxnOU#PE#AKtPjhOfj>^Z`+CDpCG;=O_nm2^l zdT$nA`rIXIvtNeRcV8E9%->N_a3K`es|f6yf1GIXF-}7-nT8sC;)#9_B;?si;{)P_ z%xyGyy&TJ$zW;-?pjTqOJ%(u6XGb>e!Cff07$EM~%yMyBFp#@spoxOgbddFbV#`*%9x-^UTDYxVdW_@QZ zo-)#P8`m?dkq5Ashdg;wYDRpfv?GrR z;nSm7NS&XBZHCAbgS@lQQlX5W&d_F0f|oF@`3LB#KPosy_5cn~lu+AA!-$isDKl|> z%C#jofum$79;sj@oot{WZJG@1TI+o3o?1W`==&p=R|n8n->qCl>2NAHsfgN)nZvIN z*N68L2D5Z&7Hm_<1x*VXV)r5fXI4*P60tV=BfEmXv%ne7G{1y(5woFWvm=sSHbL}l z&J~fO%W&u5pE2Ni=`f4y9na1seZdEfqDk+j5k%?sYHsJRd+=s|GSI~d5N$e^RSh2k z-?!z0y*{B=I+U0PE`ghKQgO)chxBhjHV#-m9ZmH+haOtCu;9L2rk*awNy;Atwz>k` z^Nstbz7sw-uSV&?PxCAq45GfW$woZuK3Zh9q;hI#AMJq?I_M) zw2S|^z_=>(-w2Yp)QB{tc$2Tov|!hYB6#0C6H>~nu=I`-WNIbhfwHyO{?!a3{wZV? zcPg{CdBy5lM+{5c_mzR8_pJmq>iCa`nE$v7bu@EeH^i&AWr} zx<3(k=`uyodZd_;u?^^xuXceA6*a@~O{IJYnxue!ja0GoL<%MH)9@{bB}us4gp* zqkv@fyKz*RweZ~>NS=q=f!)|{bklz}T9rAF$^H&xsgZs}->?YFxK>CM+D^e`1s775 z^&1B^s7v1F$kWb}({$zbX+{ z`KgkMRTuD`TS2J*eL6p-b`VYeahms=Hwtqp@+eoV!s5s=iQJ)UG&E@@$zHV#3jMw4 z@!R+DVHpM2+Tn7pSFF0p8)1hyzkdc4O$Y$%uQKRC&H%~rq4UHuTu0KbMt=x>9KvUQ zYsNu$g$&Z>l_b<^0_`|&N{bG4QsJh6bGBE(i%ub58lR3v-mXPigQwBYAA4QCdOjFq znzOaC-9Q$0TdRnpCJra#P6e}y-%;$x+-2<6sDGf#M$q@O)WM}A9q(4v1#5#bc;uB? zEbhc5cql)Fz56tnT|S&HPT5%jGIR7uw_+SwcH51t-E)*v|EHN^CMz8e^72KG z$5zuJ?=TW&{ZG+(2h{lealEA=m1MMxmX?&(ea<--X;6Gigoc?-S(yr0^%*?k)Xx(fPGwrGMPMJi58o|*z-6J@h-XI@Lk@Luj#G8W zYTlsr6_e?}<(uTzbM=a&0aBc1E0CI9u7%Sd?m|2F^Uxfy6g4jjLUE!7NP7B3wr4z0 z=w#=|Wa|~<@0Vx8HKLCgkjZ^Vo~*<7410l9mjZugZy#z9grmVHX6O^gU~l}U3lGbf zgOeU2!vCbNS?r0Ng0e^f46ijpaRts)rB0dY%-n?!r{&PAE0yWg``gH$QdD+X>45O| zr3tK9`vrtf$v}%)6VY*zP^!)4P257b{Bgn->X@caltP1{yj3`KQzyW?>?#Z#kEJ7O z@5tfr=it&UCTQD(x#-D8bJWd*kwZ;^Aa3m{Y-2~z>ZKI<A(;G18nkhkzpxmAo)dAVA)MZfwJ5*G53O#95!-BmSwb<` zEtyQ>?Dyk@PkoGm))z)2K$fiu?ZJb8EU?oWZL-s`4`(cCWDM=jDir5}}OL8Yxku?>w z$j&pX;JOks==r@1+mt9Xm4htrXreLJ``jxq)t3RUc1&Qa9b$z~Z^yIi-(Mmfp(mjP z$D1g8AX1?nd6hm*=UzJ%VCBmOMsxVOV0VHRwk&QXrpJwF4e+P`^pc54-Y0xv%LGy# zvyb$Gg|ynEiB{$BrM6d&P`%){v)oG7_cbNgMN=TS-Sldn4e7PELRuT zWzNhWBCGZ)5Y>u>^yqg$f8}~I|LVp_j;T0J;vAXnE2D|k<`^iG>CaR?;&whazkri{ zb4jG134LGvhlKuiqxswal1_0%YWJLCJ%7%n#znnEMyvua9!!I>aTBrOvkLHZJOi|= zvP9uX5b5LCDO21siJ9nhx*}wdST5=j2o zb9mKAAI3-XVVv$6DrktO@sT<7$Tx9rr+$x&9ap8bQgfiyXfv)(3MD3)w=55CG_oiW zwdX16Ehm=@n^=uJMd;~yg6#KNOpGh?NyF}MVDGb^*f`!3*7#kcOic#yz4(ai5kEs5 z->X9Yc25%77i@mJN%glCA4QS90^UbkTLrjoruok zdWsE1`ubF8iZ3KjzIq33GmtzPsaa^Y&xNa!w9Xb^8tjYR)0g3x&wcyqjn@=aKVY zj*%7hW%zBkzR~)l zgkqxdLvfcRNc?a-GoV4f3gfHZp#K-sS|oiT9xiJNw?TreXVSXlM{R9U^-Ug z+ku5Q7a=#f1~k!n5&!0l^+NS3bcW$a}E3DsRj-B{g0@~$ku(1ev13vZiZ+16Z@%$0Y7zsqa2Dxb6g$ev3 zVFbFPc^{1z8Ic*)8$kJk0J3}sAC;VpM~^KHkep~9`Izd4t%`!s)h)AFh276sw~)PT zvSTG_u9c!67C2DX{MDd1_yQ@~P{(#CEJn`Stw>Bd2h6&oOXcr=CqEl?C>PHG5nB;_ z$T>2OZ<7@mN?#B}YK048m39zac?oi!NyECWS7=_eJb50fj5K=s;nGQ)&_dXYMW!7l zD@4+O_VZ)Rf4f7$oG3kxd1~`uJK^V6lD?J#l6q)18Xi(Y>&^yocQ#|X_;)d}o;^l3&lJM?q*MU?_0X%2 z;%M8&BIvotpYXn~BL%;_z?-UOT=ecIyT7NuEWhxv+3-Pb|1{|;`CAr%#isP51Iy+4 zj~!k4D}7e;KAn-ayn8~E)r+!0&q9}=7_SZJ!O&@o)P8?n+Hi?L=Tbk0W=ce>*qT(A zIiN6$8%PkK%MUKO0gt|QLOH7=(b?CrM91SWcjrh)XFu;^XQzn>UEWp+u0CMkRi9~? z^N0Wk=h^Ux;~%)@j3Uv~P(aV8PDHoZBIsBB6I$M^g|Dl25Rc|!T=zhf?a^Ao`cB^o z>&&G|e7*_z&mxeSFkP=K$=8Is=r<$9w^_*Ttr7Bd?VF!H+oCAK$isMj##87R0?3X3 zo?9A6JVAk9qtS%>Z@Ii{8|ck<2B&NUShDIgh?{#8G~IRqRdO?l%$pKC`|lb|zrBM) z0c+6k^0mm6bG#n`d}2B!1FLH0(eaZ;bgo)E$w{0<8gI-+JhgoAxwaAY274pgHOjnp z)Fx**CT7!WX*#iWI}^00goqm$kemmr@L{%&=d|%C<2p;380Pt5x7ZZCQRx;isJRc@ z9ArR>y*5fpS444=#YlW%3=x^%1K$lyfCb5mpo!^1XrP$~3s#>u)ydv}qw+;x`AjiC z9QAz)6)oeD=Wr^RzRH{?h+ijqmz{YztKKu3n|H8QFVeu03;}%Wyot2u?t^`a31zR1kxs@VWlv%+p z#v{ZxNdf7+H~{s%0-$K%eZhyXbyVS7J5f@450bV#q4i?tc{#tvNam`Y^nyY#&ENVH z8=TJ~qS324-_v{YOYRwa@$YL~mfOyj1YAW8_XMzj4J2p8HT&{&)Sls{4`YcrYhm?^Z)t{$m(H>1Hr57BO|5PX`6faiXdfiDd% zWT(CfIQ#9r;Pe-JaCP<;p&wkrlM0CvewLppJk)fS>ED;a%o>zNK7YL66`c_n;%JBz z=W`E+)g?U796w7BohIRbnJKXIPAA-$xeJPH>@3T_ID?*9pMzrKB z>H00HRP}v4UEEd=JbR|WRY`lv@E%Dx>(C@LeFUHj2c}w*iz&n{Bazg3nK<2P-wBpn z`_1uWpHiB~XF{@du-HunQZYK4ZLU88^6Vmcr7rWByOH|DaAH5WY6?ES~rqrCwJ-!70+bRS$-G!E}b?&^S)}G8Ubwgx6 zWQ_E=?A!7&* z?mlS#bVElO|K6=~qv`o{rNM1lymKPm-<=4Do}MQUc8eni(dR(7N}kM&{mENd6-Rs| z#pr|fKSbt1A5{DI80&8s2M!u`WLDuJ(Cm8$9-5y`wuF5nSEXg}J^xcQqp^`{Yy2Uv zbbo?WaG2z0$-{Fq&*Ot&8(G5p2Ua-lCpKXTgx4rXOI9q061tf%B6$t)BrT-rdo;Oy z+nSVoZzrIO>tr1%A?_bu)6uCv>DR_;s-q!-I=AjZS)CH-WyNCBG3gj1cp^zBSwCfj z|0D$SdM^^ieIK!abE+@*GsY`wFmKx^6!`Nxg&VWPVRm>Mh#kqrmplKEk1zb_Dvvc( zw=5XTk!$Rd#AP_*E%*QRk{fBC6hpQ}43LAZV)RrlCW2#xT1>5^vz)4Eaoz3;bAf5a z^23uV6jPk&uaRilHGK(n;;kc%VpH*vQdykxbuJD9OYyZIv+=F7Riroc9(IDf{v|z( zAF1}UOI?S^rol*Z*5MloQ&y#(Oayg*e^=<|xr|r0eKRw*qMb^5eIrii*3jk&r|3J0 z!}MBhD`?A%rJd$sxaq?v`LfRfzCSjVELK{97l(bJC)D$4xr71z8p0>-Z?nPfhIN+L z+C}Mqd)^Vf9k)pJ*b7qct3snnJgLkh)B8-Mf?ZXbB)c*AIe@C<~xnjv&zc3(x+H!|G4} z(8Tp!bT~PbYPNhNXVV$7`OI1vr_+JuUKy}0ZTG;o$A_32wYv;)u@Hu4uBBSJYV=pi zb;4atiILi3xYgE_Yzzn_v37r%xqlTnUgl2vSbmI3)EH6wmSW(1+mN`y1;q8Yj@eeZ zU8b9@XA3Sn=~#M=@?a(x;sH;C$;eq3l=1rxc@s-vTU8Gje(!=RQqQv9P8E!MumPBw zX+j>ybm7tmB{+kFIJzBiw?w8YOm0OIOLuO?HbHNN8MCGXxuc;>=MP;l7-vt;OqmN~ z)5qD|k{GC>ei`noHzS3k-IM8H z7>{O(>C;s;1b%Y5LY0j*IS;ZXXef=dxVo_wo?0#mi=zEu>%;RTa*t$1Q0^bHRzj{~ z_n!Uiq~SVtsg*ZA`7;I;d3X%dFR}&;XTJO>`uBMa2epR0-N;knYMO<3pA$I0m5&_z&p<;V}QB+q}&2}?P@sfeQ3-KDn#LTt21#xwKzRlsZWhhUx8;bjmes| zPw>aeEPTy>8}vp&;Etm{nd!0{hwSSdlzPx6DfeEM3 z%a0f&d9m=U$ zykkuAqkxEBDpCDrg$8pX;coq~v*IR&In+@-GAz zR8N~XyKNvV3&hx9YdPNSC1ie%zb4Wgej zpM0*_g5}~OXn3O=<(2%#&WW*t+8$Lw=C?*+If-DEBR1mXGf{9rF^vQd6XL!81blOe z%R*Z8ldS6n#J!>)DU|;~UMpttUtQ|~!CzzGy|g*tctQw?2q|Xu79}&cUdFI-ZQJmw z>Ic}u_Y|urW=eP~bcyy1P2T>BZp*UvQeLKR5XBd_(JYk#;%T##KHV=v%QyMb^Xs;t z{y{JF@|gl^@?Ag_5^us!11sP<nZ8Lc^rVW~Ei+UX} zmIssfp9Bf^YQUIb0SR0Tl9dIFDEAqY4eNnRJIk0Y-UHX9wSvNYSN1L|Excb<%j{E0 zV)i}MEMFYBqkOtnwfW`8QgF?!_0X^4D>x=o3z=U5V7-1T&?Ev7TKl2w$k1%)Ib038 zLxmW1T_8WAvf$5csZc$89N6@2goMoo|7P3;(wzpl$>$IoxOxcKmqfzOLjiE?ODmkO z;sNgz?ggxABD3JMfEDf*f@P_d?4Q(ho`IMxxa1pbv3*%2xU4BJ+`V;~AmvdixT4j_ ztl!wo6n~lnetq`l-8=q-{cL0)=#ji4Y`NISyIW>!t|`A9{M{uBa^gf#R@p~Tt1}z+ zDf_V==2^^MrM6wX~nV+fGZnXd2`0sfufZEpi^y+g>ab@YgLgVjJq?!_%4-(7Y5cr)yW?0RB-@A zh$DJD+69B6{=(BiDdd*59QF-P$IMq(%dMwG*n^Y`s|>cFM_->n6O|j#MCUGJ{yUGE zwI&yI4mPp}m0#dxlc$i{Z)@Q4&}5GBVZb=cw){INHeM9eoO%w`Rl9 z`GpXkosEz39l!=(Ct@7=9NIk3g0E_PaC=`SXsu`xg)|nHV^E#{=;4>ImLauTR$|Oh)3KUK=|z& zu<-Uts`8glO%&ts$)&!u;Jq2OPBo?)lO0G*fDUxGRe+7|?&xyH1(c+ogsQ)0GY`*v zWfxc%fU<5C@@mmcGQ` zb3k^rP6SmV?ZNvqURY;$AD(k23I)l$LMNaE-#WD%j;KY0#aeDSK2C{$ ztf&C_H%RfnIX030l#^J!=Nu_bUX3Qc=|X)UdQhL+CG?^x4K;ny<2(FSLSoA_(UHSP z(YnxPazH;4ey7Q_%|HRfi<}TN8YVHW>|8i@{UTZ(a}s^qqK}Sck3+W;yO2k!2YT%N z3Qd#$j((ls*q3Wt(fnE;ls5bezOgz4ui3aTmlkh9Ru!UXp~-0&aefK&F;1G8aQ(;A zydzb z3zb^}pyjhMhB`>1qGh7c5-QVuLQ`6Od=ibc&_;{Oqaj^*4#of8h}2B4z_(Wa~Wq!f2-)Lo; zIKS$PII9^mAC?Y@;1wbP=qI;Ze^Ocmw-Hs`W=U~d#4@xeSDW_E9%es^vLHIA7^N>) z;Wzw};|CnFqV|8?V8xYGZ0D{lEE~5IhOU<=EtOFAbg@)ioBLnPj$r(6F z>@&>!Z!(JQ9s=;v9P-6$CC#|7m?oDLunrRU*^k;6NoKBNS@4t2^1hy43&}(4aV%D% zkNZ<->%lX~(8(I|&B{5Z^mH_9oexr}F+@AJH;{lU#vqs4z>nAXjO_Cfb}Z~4-XXJv z`TA`-^7grgzTKEeC8j3fo^ciW&0PxkU9cb>+bhVMwJvb}_;2uE)?VP(_<~eje}zYF z6`4g~8ftmIi2PVT1V#jYaHc^#nz}=T?^~=7?fWheYp3~OmGCYuthhr?zHKFIABK<- z_W;n1D!{|%cVHyuqa)wNP{lNNx+*1{f(=*bgV1*%I$e@HwG<_*F2;df(HYp_NH?}t znEJ8k1?q~O+2r_xGBaNCg0XChVLMj{Q zkkLK!sJ3(h4gTXGv`Ci`hJR4A?7n^hxA`Z6ifLkGjg$c~oLVLDUjEZU+Om;MTDX#E zl&Zi|?^{%-?=}@{GoWhbQ<>1b>J%sAXW5Qux3)jp`S&F{$lpbSE4ArJ zfGqViScGd;F5v*%GGd{eO(aVaNlvUGd|^yjy5yFnk5L!n|7r($ynZdY>N&uicI@N2 z?=JMyy0!H0A#LR9#-J;;{#>5oCH4xRK)ZIwgM#^Qp`7C|_+x$pEZjE-Iegg*rGnDH zQQayMUnqjt&oIDa6DFhJ1%4>;%?4C&Rg4&)OlW+VV~PtTX@S-vGLOAZ15TKemzmQ* z*V~Dxym%`5H;@JYbe6(fwsy>xARbTQ%n>Zf?xU-u^2o1@XJAdrO_II24d6WmR+|L9z?4)XriEwayK7$4nrf%L`gr1$SXpnGoS(NDgw zX@}G``gP*~&a4aL9OnV-U#~E*>p(m84HZ&1C2cBKB2qEG)|~dp-ll#QcUkM1%YpN+ zW#C`QEt*_^iSCmOrKYd9Qs>mARN~ESdVIA8b825EQ0*8aPjr;P3@c-NvrHEoooQ#o zd^^cr`6<*c^ftXRw46RYIEB856GJKo9l?FO8$`tJ0bZD=N7hW9N#6ROhvyzC;MA3K zXwJcztl`DEysry?3uENHX+@ANEl^ue0{)v)@prX-1yi`7VnKT%E*Q$j$Gq?34cmou z8}Xqh4;Imd346iCfgAw8Y{XwK$W<(Gk*k=4n<+`yThVx!U%_s6s(8KZJSkB1!G}6A z@%?v^j=F@>5TQQj*G}f8lVAo6%doDUVZ_ny2ot%vm23^ss+f0hPQ~mwGb=9o7n1C> zc&7aem$5gEMzI{n`n22*Qgd>R&^#tcxF?0{Ox}(I2VF$Dp88E#9={eat1aPe&cQPF zl%d&ErRe6XmE^Ctvp`k0q}+7D8CJ&gEs2^xOs3fXBYF2D=!ueQy2ve&e%TaFS~}&0 z*;doZ^?~!k$}_KUb?tMkXB|LPUTD*l!gdW=Q&`!Bj zvVGPD_$zoVdgR^*#7iZ>j+qHe$TT@*!}XvoAHIhXbK_ue{5j|{FBoh|jI;FK5D6Y% zI0@kuQ`GS{Tc}<=k#pU|68`%*(xo2+&sTQ?0TT~by@e!Ldm(_&FEKHb zrl1o;Tt+=Ci+Hv-l1%TV$XMbLSiw5L^ivq@oF{-&+{9otlS-C7yA4BC7m^Xs#+;}L z0hyX%So5L;5xLq(mL5>0>aQGV;0AT7WRXDShMGwIuTUmGwhljcc4VIHl>^y|Z+H*x zjSJ80?&Te}zs@_I9!HX;Oes~(q@f{qq0S6n;(5i8IyJ@MJ6ul5K;9lmN!wGTlL2o^ zssQWFQ_)#y!sgClxmnJ0{PRz^P$gELs9u~$4+NZ`Z5j3CcpPHgKTV{DQ!HsnWDUI^ zvzDqqd_lwfeQ-<73Xo=!kKz3VK(NpnYrp!!`u>h112s>1&bbQE-a8hoyL%Att)Gu) zvBuD2Oas?-#xf_Ctp_P-u8>`kkJap>$qbHZcDl$M?@h`FElD@<&Dai@KQft0Sj?i5 z0}}Lv1;;1jUtlyu?lFlePN3`SBBEh*8L!>R%@-Z!A>|BHIBxfYEczxwnO0L0wzMAz z%M)2Y`y^t#YBk-hJwnuP)e>2g6EHGG5;>d>1~VQiL9FnO{Oi(YG(P^uPO+}^hE*ip z@Zlu~u(=3xDq^ADumV*NP#{U+5+tTq7kMc3;+59~V5#c?qAF7Xf}MU4{^}Usk6HeJ z4TvDc17|_QrM+P0MiF%Il{ypii$IUd$B}Md5k7Iih=%`}P3Kz%l9r-(mPe8gf&u?e zjMl`cna*meeUb`)ph!-lT4*m`-CnSMA`rQ3l?i9V~ z5RI0_Z>EZVR<6~di9EYfknLt@ondGkA%diF2y(M_k{LTu>MrLWlN z8bkW=g)XY}T8ZGEFXXkI22pNUg8avh)A|e_G9KFmdKYbDi##x>Y#Yb3_8&kA9i?PZ zyAN&o!D9lsnd0S|bZD^77(b7yLxuZ(vSM=KFDc^REISxOG}pT|5V6sotEW4X1jL3h*$?vA-+8P4diYE!oG ztPa$3nG;zwBcKRLzLrO$MFYs|i61}BcP`mbHJ_N@{*2L(4m#SpjGrzxi=Sqv#jF0~ zMHiVarlvn6kxXqeia0(F=LSU4({;De;ej^vef%`4jkiYAY>puHAKY($Nf>I`;DUUs z|0_QaOv_`cBA8^+M2?-gA69>0kdpINu;Y|2+Nc={ueFAv)oMFPlz2gHfG?YnTHs@p^UH+);I$f@juFa;NQFnP zxFEAVm(cV%>1f22qEFdZPfTt1d>m#V&z*88c+V0f@9+ezFRX-Bt|w^9`U^nK zupFgbej_~nB3bAL{P4}FFtk+N0Il!0K?(92^r*mys{TyE=R5S#!}K30RHhAH<@(1t z9cO6b;&`&&<1c>H{s%T4&_+R-QRQ-;>4I^TB5)7Lf+Z({nAzhtW{eSIYbijSE!5s52}?`O1IU@k%?=E z$mK~+bn!_i`u|^5Zw;c zqcbO(;vHYZh=i#=8$H^==!pMDOT@U$^TzpT*V`iWcTX*uaVU`ndj2G=hC3Y(aitYw z9JfLt9`4{8L3RqUsCQlmK5+ClUjEsGNIfZmE+-hW-Rd3lGc!jZv$msLWwIx)^FSRe z8+rj3nNbp+7C=?La$Q`JhqP;JCQ-F9WRjCUvsDghWFfiBIqi?((ax95V9qIK{gS_U zD9eD(h}uNwVH>&V?=K|YB-GpEAU*I^OZE)RFWmH;y3z<02V;(zY zS`vqQ!V>}9oq^-L-fO9Xjxv4Z^Kds_QKW_D#JZva!Eq#E_L+QGQ4HU#`%dmDJR^^S zcL@5U4};!sLbUD4L&oR`AEs-|qrA^waqON{?AobL4_%u>qcogpPr5lh{-jc90QT{Y z%J%S{^p}!D1K!|=N;=#+*O^?}w}ToR)zL-ESFmAYDrkG}7E}=^JM%d20*{U=y}UWyXU1|8_Fy$*>F)IgiTnfMFSOkV#wMUNeEr6yI8z~;$YR@3hg z{U&>d+zc?Iz3nzMWbI@6w~zahD+`-I+R^D1o)h2Gi9IRg_dpKkk-APt&)Lw3 zSC?^h#4Gkv^E2MIjVefUo*(qun1iqIWVsIDOd1a|X;b11+N?4MHmF9SP z(;?3rjOzP7bbfma$_2S-EFwX${=JM)3RL3kTL;jMkTqz=mOvnzAH*x~^~S^1zGOkj zX;h~qio9n}LC3eXqly`ykbI8`|Hkn!^!2?LT2iir-f4fsHh3dlBsc(!9~>!%k0%HZ z2V4>Ux1tO14}XEZ6RM$?aXS?K0)XqLFw4B!>)7P|8Wbvj4vq0;=+imSSPGEBcP3K>vrC z|FYDwlrXvC3q#MLz1Kci*=hx4Lqs@mvn#$Xl7l@RZGpjso7h}o5p(999||3Ag5fzA z(chpo(C_7UcwR#X-z%*!Q*kvYx1XGDvB7i~k?*etS-tM`>*9K_iXy}JZb zfA~VJ^rdLCK_@PF9?PDu;h3>745zIY0`CB>&)E~lQ97Q&LrWI ziDP7d>(#H2xdxlnyKry(C6X{0OmpoF;g0%em@F8;r$crj5s6o)kb%v6|tW)Sb-?yfpEN|2OlQN`aOXu=Kw$6gdCKU|~e@QX4o)7ylTVw~w4 ztp_w|340=aHg8XZ_50sqY1N@sq^AxAcIzPJ6CfrW<)M9OzrG zYT~VqsP({1gqZyDcRjl0JJwwk{HRNXZ-*X0_oV^QJjxK|C7wl*iQ?$&(ckc`eg`X^ zXF{2znWVeifn2sxB5QUofz1vga1ryA3wQ;xFW-E|6;)bjgkvv%2-Sf9xIWyOp&u|j zE^l!@{bVYs{s&mtVDw=kn z}rEF_LwQlfM82#{^ghOV*Ynre`}=NQh8r#y`5z7o{LfGc2o?zb1WIgZhe4v6zkFDcV|G|Y61R78^~$fXs8rXN+zdN5J@pUyyx?T2zTYe zvs%w!pjI>PP~c|ZmzRSzTddHfszb;#FcEg|4~F6#e?vC$0FEuZfft?r!(~K!nEaJJ zMCsQhxZ$A{U0f!I+OpHoHPxMHAbcS?XrKbOvUc=(??Ff;36YBs`eq;!>} zl^}m@40gS(lxk`V5tk1U07>}vq&>Q=@C)dc$XRNC zE#`@zj1W3!8iD)bpGjztH878NfoWb2z(%$JT)ufx_*ch+uJE`;_2lj82CM7H>7fR? znkz*{M{Bve;wW6|eIM!^IR z(N;7)b^u=9-hdX4D+$+}Pv)(ChnW-wM9y6zaK8H*VitQxpgS(ZW-3-Oj_ZcOnK@x3 z>GlF)@&2#Gx;O-ViU>y%D~uq=BqsJ;x@LK!0&gjaMx!zxP!xOs0uP9w&0<4fm|qL> zDp*T{S9i-!=x&8^!x|*-Xa##qrx?HcmP~xMtft;;X9yk7&J^zNt`)jHjDkke?oh%} z3}`J;1<8sbJa3Ihg7?>I@XU-j>h(N_4w{=%x9?d*P~Hshk|*R_q#Jpm91gvfXQJlq zr(u>|GOTUU0g>Fiog^W@Q`*5|8 zDEG5Trzx-7APRvtu?C4jdH2oFKdNK;LK6@kW_OoymF9Q}o zI}X3v{D(ZX3la7wNMZMEl+-W*p0YfE-8gkj^TI&%wq6poo6X}Zhc+Su?lr7r55f*2 zc({K73_R5eE;X2fRdSn{4)&w)u4)obODd4j|GflVLZ6Vjs|^2x2IQv%Tk-cZ&FGck zMd2fU7kgWE1;}iSLu#*=;>x2MaJ9OdFjk^eXa+9;(`!8VuyQu`ys#7hR9%LK`}U&x zJb)$~jzB4MgOReY0}7W)<;QFa!W2BRx_m34u6s9?1xe0E=# zaXfZc(5`C)=DDchcmFuYJsE&2qN~~JzwdZiS`9oG#XzC#ymC^zAOIK2T_jt%FrMA! zckp%b5YMTt0(l(PMB7c%k;EQ1ysdaD*|c~Ay8dVmy*kX2ITcCx?Hg-UB_WTbKN9eW ze~jcUx&u2KZQ$eUi7;ZyKTzy;2uWpn5-|;JBs*y}{PeFMN}o=_oy!6_pTl01R_Kp1 zjGv)`eaT#ZWhFXP5Q4T1+Vf&Wqj@h)6R?V&E$g4Ghu-d+j82}DhIfybvYNB};qgVr z;J2IxqJ+<}zlA6&I~pAwU5bX&C$UZ*i%1sNhdb%Bo7Xs~3=F!a!Q1m<@uXB8)Knvm zJU-UKo)1xs)8~ui!Q3hsyzVn7a^w1X3Jb`eFT0S>-F)bgd56a4+0Y$5NhEijs%548 z($eewkB!eR=zzNdP0+~T4CFd}0V?(_LHGLFkUH-L7)aCNIvTs7^G8i6x-Af1GQ{v{ zDaW??VTc}~Fyyw0!i^V_@wQhXa8_0V{A1jT-}!pN=~`{*6*t?;>39a+7RB({jFbFn z>L&b%d6G!{perL5KrrGTL8W$OFtp$ZQnBVE8z&2NMBV`tSQ%pAIR6;G44ya~62xsPDAm$p( zY~>fQcepd@^Ja0RUicUtPIiO(PZIIR1&Rbsxd{ZJi3`rkLPg$9d+k0hT8nX|i5*s?lO|K>KiK3s-|zC@r) z=dK{LO(tk6*B!Gtxssf)I*UAGl=w-SF3^N$1Jr)}1Yb=jBbi4t;3V^M-0sl_6sog< zmKC=jpEnI?L{lgtn}A{`%t6D2W2K2=I_3A&^UMwW_0cEJ;nTJL5!0mE3P-f(!qI>p z`0}bMzfvH=|C(RNDgMOamnb_}ni9uoBt^k!?P6Sg{x5m0>CbA9r@`$XCy*%faX2fz z8k&7jpv%h)saKK*)#|J#w{nwU0Lg%!rwB~4&PUOrz38(02J|Gm7A-pY9ZkFvi0->O zAyFMSC^!$lcR%l&I%U;6;RS9E~?9ee^e7g-_C2d~&XT^Zz7Glj%0y@Fo% zRHCS}w)}*MS^NdRIpBy`JUKYGk^Sf8NMvTUn=>D~Et=mImMW_|p{Jgg(HR^Ly4Ok4 zz0G>8**Q@<9F>8$+f~AF+t*Ou_AW|sRzQC99gt>TFPIS{K~AmRM$2lsbNQjDK+uEk!@2j>DUb^dlgRRgh-B}MAX{g*LYL6< zAotZPu=LtH_Sz1@%PQ%+q%! zU})VmbfEV%X^Wpp`L7TdiE#n!$^zi0gwSmhZ|JCB3@22G@Etl!(8Y86QA6o;7~(Mq zZ^fN}Zu|Z5^VSCX*Rz7wD>O5)PW6`YKMe5%_M%k$i}F!pq#P zNmrlBBO~Xg<6rsHh*Hi&B4RR^>i%cX{(5%>ZV?xuO{W5gAg%U*DTdu^7-o}+2raRhNjKflgpIfU0S6x`_`zd4 zxv!^7k!?1K!H!hQRT`NeGo%WGAT3>(;${|AV2dBZI$r;@*h(p0C{oM<+mLxR~4kzvDm?g-iuKm$ z!fSv!f1N?iiEptkc@5;N-+-NlR!q`WZ+2q0Gv0OjHF)JJgw8wugL`{k!>2>J@cO+j zsODb6EcR_6xT}~BqJHMw@&RUZP7%KNBM&y;i-FU-oFG1%U9x$;o<;kj>t^<+``B-H zUEo$fMeNyn1I_r8h1xf~MRR^Gh1Pf$QTF459r-tS(Mlac<#rMH=Cud&Foh3(kRI?) zJsq3LJp>Bwm*D7F5-Zz|$dx0}WRA=(_&{SF%vE>_Lkvnlbxk2~-sB|+Wu$Ok%0;1b z^-G4aihy$u{g1OZ4X5(`!v4uTmLZf14HS}jU+Xed64F5PsClA!9*`kJrVJq>DTzd( z*w?x?3ZazJOi?0Av%V@x_20j@|KmBH7tehh_lx~%U-#bkwAMP$^RvLNElL8{+7;1F zi)`$Syu<8HJTK|IQoZemZ6UXk-qakTB{K>cT)T&^F zoje?* zl>UmnM^~H{z|;G8(S3)IceV8kl2^tFcMlvx%O5NG62X0LzPbTN*4)C?J+rS_Sez*J zYrM}T1}BUAKwo($YS?wmcec~nG~Q0h!xKB|6!=`vZcP=QU%MH5&0KJm_e$>1Y(>5geME=kmWa-axyp}rSBfgL zHsJi-E?l#?m<>7BOcMh=Y@QriM}|3jvT-wi2tUePW0)3=gL*tTb)G@zl@1l&7s1a9B zoK1%IEGdl;5qfyP3^m%Jz-K33;A7+j=&{$Fg`C|>Zk-N<9TQXO^r_3J#=ey(-`YqA z?r);nLpHF;tmWiS@Km1r?;q4@IM4@khr)w9;%pN6iyO2 z^T3bFtmVK*KEmlJf30$x^QjN`b*aYEPo-#S^m~MX=UGv|y$oeve_$;iM$)2WJSJ`*rxZyu~7tFB7Vp4q&mjtKG*v!+C+*Fy3chh8r#FLH&Rs76*xEVXV8fyDp@( zI;FsN;L16^Zo~_b@#PQvz3iZ}(jT)#A8!pQE6O>}GY-z>ch{-#8;9TVLBmIi`WHEh z4jFEy)_JLX+}ARmd*=~fy7I9oZ(*Kj=trsJ)rHlUS&bS*hH6@J=o0%xYOaWq zyeve6SGd#ZQj^)4qmOXlharEYnJEfRbs^U)y|~gf&9d8mjUs!GH=>D;l0 zH}Iidog#Ui%c6O|97Vr8)w#sUdSaum2&W5o*p$nD79P`1WC3;OxWX7^p14$!_kLb2 z8ZAl{9d}qL8b9l5>CvNSg;#n{WA~(K*camjl46$cG<8n>hK~{%2c6|Q27YD2q77v~ zV|SK)?JnR~mR50{*p2+#!!sh)5@~w#oC&=gq{aWqPr{nLM%?C9m8f@YvZ!r#orrCS z!OQF2M8X4eM5?{N;ZoyLI)fWQl)7A5C1#84|0;++nlBV-H8+W#_GgKvm8_wY*6fCL ze=bqKK2=)%_9!vai{Lf}?`hfRe7g76L!MaI&&O+vS>jD^M7yJu%MObDL`s60qL{1u z%d(u7mep+aFRMGfl$I>VNdLx*0~@)zO0Z}{behOVXdwj{J zL8A2RJ))gw{6qtSTG8FM=X8pmglNCYY-(nBj{Y4TLwmhFxp{{!&xn!au1h3DRQCc8 zIo{9TDgaY+u@?C}*AxZiKch!n--6APg`#;>TjaZ50;|3*;SNn#smwnM+H>SG-WugD z&cV~7P10g!aQgkyedeQB>$V2l)UK7h>hjT*>mE>DaxbeA_hlL=$*;dbIJso>H76M@@yAc>^*YkWG z)g3DCE8qOhH3JOzuw_!B>ElcUn^#D)ITrUze}`L?a*ZkY`?8fN@V<_y(%+3Pm@Z4h z<+}MUCH#Ac-DSGoGUd-IakbWnc@W zbAsNBUkF2Qs$iXS2!0B;;D2{ViLNg{%w=R7ndeM7`CQEfeuO1&XkgtkdT)b zfBOGz+7fmWYj&^O?7e@(r1fj|u9;+QX=7&*Ha_r*iwh%X#G0 zB&-S!!J+Ang6l0)Y?sO|=915sa|IQ~hg}cnre_|pO>MV`XX91$e6KD%`E@V9{je7! zk50owyQi_$UnX*`mpggKqDZl3?=@HU-Ub;bYI%c1Fdx)DjvK09;X3Qec+Km_+%)V3 zpU^4R;dIX@A>j$IU`{#~u1BwZk*vmwH*0|q6|Lxoe8(8T+W|1#1I4LS>%#k2>ci{7rP72qrl-Pk6xQVgr`5V ziI1xJlyA-a(b8F5l0HS>C9iqj!{t0gsSZ7zrdKK8-H?RNTh2%w_yl=rdv4#UFCSFi_c}c{13cj;n!tg=0CAD z*(!*u|JllaJLU1(u}Qp^-AA1jov1o80*lMN`SZsP+*mS#n@26>v!fUB)eF;Edeag7 zpfrHp{)Rl`$x{rj_=UH7IqC)t!g+qN5HfiTw^Rwk{zD>%Bht560A_Ah=uO4j>jNXO@4@dDI)o#2z z;Vd_aSj;CS`itMa#o2?W+qwF?4_vQU%p3FG!!K-S*md+JYP@PdVZ?N9o@Nf|HSJ{l zBSWIu_nb-fE#c*&k-T`hHs7p#1JCUCr!onY>J6J~e;jFK6J17GW)`_Ub^I4cKiMfbsHl`6fx z@(9ex`AeWg5{y_s6o|EYKRW+|8xu9C*7tOBd`}j{*5!lDaZ@U_-kA=2aUXEyVEVHo zk6BMU!mh$}I7DiLhl^O9t& zG$aX+X28ziOmG<<2Hr(o!lij;VA^X%KZf0gqy};RXh}P4*;oZ7cW*(-D-qn(?t>*M zx$s}hC1@No6q=6AAyaBx=&R>luv$}(286Dr-@Vs>#5qH@$Ign}aV;V%r<3_eC{!JaRsbfjScY!|GCr2~JUF5eVWN)cofV<7n82f)rzbhV^C zP0n2i9k1%4p-PrE>n&nRW+~7-UykNl=u=7WdEmXC!L<#Vbp6DrWuct@ieUOAS- z+b>;kqDzeyN4QhTDsh&_ks!A0wi@hJS=Y#=)&_5Q3E=Uj@Xp+GhQ$K;|!j)3h z`xnRlV{m8Id1j?Q2Fe}+-O{uT zZVmnj5>CxvILBO^4PP!ip>+}--^hY6eI!+8&qUbH9Ii*IezfGiS28XDt$v!&i!U8%w_bL4S-UxZR zMUZgd33MKhVK;s+WDzlsuy*lL(oJ>RragNa|F2k%Z=E2= zq&|$no~s3{_xEC4*``FH}cU3z4*oN+B|6HJYK}IvFNC{=VUyLjTG*~ zo5Rys=)`bRTw%oLFNqh3t9<_0`4ZZ__vHy`s`%Qk3FibI;QgXf9ICP#hXl;w>7Rme z$-WY@zerNd2<*c>=N&n5*v)ev$8i0>jJvj7<2v8I3o_44fSX-`JUQBk7gRjLO^YV* z91SPDH+m^g9TkY%CQ6cn?(@((Yyp4kxSyw{T*qxuP8{Y><63JLLuvDSrt|4Ku^M|8 zg+~k7+VSQ%#`7L&{HO;%gIbCEtU%cNz=myCn}+**V|cT56i@%x!v{Vk^6>TIu$HhC zZox&oZSN^wHE|)gFNoxK=XUTt3R!&B`qkX#$x}XJ_HDju<~_c_Zw0IS+QAxD{2*sX z9b^M0X8h`FMXvWEl+S;99Ea`L!UN;ecuhzO-&dc`S5%JU1&)V!gD8^+AKc9)^Q&2& zUk!1+eVTRurYNte#FL^u`6?p^PFyeXN2%wzt+OgX?;3{mts>+XSGUFa%&Z>OUF)lOq5`U;3i)WT{@@&ExxZZh!ytf`w zYIXD;j<$7y;R|;V8^Koo@o)uq@^a-P<*IS-Pl6M+nqcv86WrW=iuga%C8=k;@#9X) zbLD+;QNkWBqYr#_Zy}m~sbC7Dtda$1- z?0v%D-8bfe4GYAao{8M`+;~1FeF~p**O!-G3xhVxN}+doiD2Mk7dTrf@{>d6@Upw7 zasIqk7WN+T5ofs3ek?50`$_6!bVY!)B2Q_=-!(_{c8e zn05pcWHk7v&Ouyex&Yt44ks?rv4Ya? z75y!gm&b zEv&T6L!0-G7h>~(9*=$1j$`L};k7Rt04!?2Z(ukkN)Yr}973Xg)G}jHESOkC;=aSJ zxM=^(ftm0bQgmFB0WF26RE~9e67h246@s z^c=1ggg91}&fg_I52BVsM6n?mGVnoo`r-iy%dLgNz!XrhoQyHjC$ZEqnl;`ShAlU} zAjxDrjHnyIi=Vg%+Kp=2!g_Z&CjFOedJ>MOVGu9eWP}?YrQ@d`zA&AKW8z&6T;h^R z+Rk*5<3HtCvGE=jX$scT)9hi4L=Y}Y^g`>~hIph&0=IS#h7~96*y5%#tjS(XhMNAz zKFW&U>l7#A0j+~@)J28JC3LTQ8N~O3P%)l>tQWrxDy)a#Sc+?*u&y%;09tOI!Oi%Ffx#@GV|a zTtQsl)Y#7P!1b{Pej5{u=l2&gyJrhP+My5Igk`unB^lpq4&$#2 zkyI`{iuT49$oBmd%zqG!7jI>vN8}frw6=sy^JNe?Ed{cZf?;PzJh8bq11f^`pzgzb zQa9`_(NOt@>N#pWD=J5*IUdNUWEViCIbhj!o@hAKu{A@_qJdW-j%+`P^iL{w<;@j5 z7%okk%!KH6&g?PyhQ z7T1L{s`s$X-F1R4m3u;A*b?$|(pq8oMis&qRgl`_ziqG92vF5E8}0VTu=vRd2pf7> zljS90c3(MLvo`^$VhmRO`-_7tWqH0&9y(fQ;8>v=Z(F9z&H9G%TFdSddF7vE#=ulq z(5ne&8gJm?9cS>=_;}3!Q-Obs=9jvp-Xfnz)sihkp0h0xKB#g)mD|l#&OxF0dV48<6x+X^BWjdL>(C9AxrZt_6xvs)e_bvec zsBa{6l|59tPY0N z9>a%a;(wh#@J?YFTVt9j=>NT6*#2ZMzA%`GzqIr)_HYCyTrJ1P(`4~jcMw{AI|n)Y z8{m%61Z+-E7W!|>6JHO}w#7XUm~!b|tY3Bwjp8b?WAj(c6W3Rl{jZUXXKDE1jT+y! zp+cN5Hxd2#PquCD4^+-xz?^0`qpI)@dJAR>)Yjh;+%mNXqhADW#GSywCD!bmOoq_A zW)Pli7TG35G+@iqGVGWskM}42z^!YmY!6TD7JBu+ChZZY$fHTY*w&hhO%?CZKkYJ3 zmUO}^LzM)nn}##zot0#pjsaw;sepcL5$n0?P575hF!X*c3ERS$;@LjT`XpC+`GF4@ z4BQ7Lx#{HR)W<|8QImZ1K0+#Q{1x8&EzSiWbpmgXiWinW%))2$BhkOCh0STvWCaVK zGiS>%Y*HV|3SN~7vd7fhlrOr^_Mgunlhw@dm)}S>^pyq#cZ=trm+zqd!XGSOlq$}H zyoT~(_Fet8{n%V`856xU_^qIETzN<&=1puP8VNaMe#;S|d#*auQnhD4W)|bMf$5}C z*9NSTHVI0?o?zgQH|XqU%(KDp_;@%8_Rt3o7`*xdxz zyQk3l>vO5PiV3~FXBfYp{Rc_NLmX+~gXuceEJr0maO=ZSvS#`hLGF_$?A{@1yuI)X z*;4z0c*Jt@pj5a{2%9_q-P3vG%C|iWAz~UcRW-) zkfJ$jf5FD4m+&*X0y>st3H;9*6HSTlV148inX*9q%XlPMfyF9z^olf}y@%n%S6A_Lk{v3&i)CBi z1cT$p8ZvLk3aBdkD;WO$nLuyER5E4oB)qaL7M`9NL|-QxKxL5%wCz(W>C==1weLb< z`8{W0)vy^8x5Eum4$rsSfc&-LFgPN~CjLPwK&OBl zTM-3DhbnN?;ASkb6xZdy8rj@uXE9V11ebLJp=*#GTueIzjW$9kJZD2>w@ATi$;)`) z?SQT4x&-2MIv#iNPo*-S4R98zgy$iHm|ogYgJJLG8G>5BkE_mT;N9cZr8aGKs(3{A3w)7L@|$M59mo=PS=JfFO&=cr#Ch@s6M>}G?Xt>$la%+5_0v`srBTsO!X z2LcL&e!rYx%JT!THG323Kk111Uwhc=)aAgerO<5jFBa&dE?BSo17-!T#(AgYc+3HD z{o^RXH{KW_OsSa#8Y>Tw@sl%1q)>^=*4meT)ey44czbA_avs*mue0^B4=c6q8OjR| zeZ)ZLn=I&%EpCu1fxaygz?7emk{>OQ=(K~yWI189)>{_j{{kKy8w$al$H@4?*`N$@ zpj#&3HCb;pNM?()6hMYZz=;nQbY}7oGOVpom}Po}+_zpr z&e^X<+v1J5Fi#76gt;s~FB4KH1j7Hai^a28v|FAGvYAJr)M5*a8FvP{Lx$k-X+6Tq z;9hiJcM3h^)A7vBbL`WnAT_`{BL@NjeZaIwe)5KOc*Ok$+goqN<&@{)6a|rMi!KA zxuy?retzjfwg8CXPvRT15?taILFhO}fLKF_UbqI&A1h(OSxNX`xC;#W z8bE#>h#}R}hKP@UG_zkF20Po_ply^k9bcXuxcFz6uWw$wH-;9peA+3bR>u1}@8sjJ8Jh^#)3DCf$8 zjC?!kGcO^-FIeLTLsk4$SOkJQ)5+%hA>g{j9B4xdX`U;9xU&i{<%l%cO8zYmn{}!Wg^!nJ;tanuJU!Y92G(oQVU^DyeEBA_SZ%tnew>JX^XnAu zc-zjLt0Uo~&nsA-Tm^pPq~O-WyC7S>)Yh{&i8u~Qg=SR`cw1o(8DAB}wLq(IlKfcM zy>2#{CFu-}28C>0UtpN#)==oM+Jm3w zTVek0tI%;OmfdLZhsf-B5}4LcW-m!cC5K2@vd{!SW_=QLo%Vw5F%)*s`YK2|6T>ue zZwfA{6*H;{k^fm~%@Dg4$|9?M5g}OMMD|8W)rnM+A^B`e#9Sa6XQmK7;sW zdf=4RMu2XwZBvY81uBC+lS>~YLGD+hP~ax6_1g7mW3&&|F93Q)O$vTr&KLIgm}06@ z9GGwY2_@CvLGr04HTDqDoOy>~bG$6|a;k%On}pyP{|VfUs$ux(TVT5@5LVnd4ZX)U zf=l-`NZarZf?ge__$HJR`F%8Si$AR`TS3jogwuJ^LG*HwAzc-!M31iR2Or5pbm7M1 zblYyRHZJ!!R7w$G(UM=?JOF7Wz$unt+!hYJdI+!|UQaVRVnp*pG1Nhg&lXyL< zJmVo4N*mLKMPimhrUi82OJR!SGIG8*l6|zEK`iJ=@@BEO;HJtv7zBeMf7MkX*Vq95 zlH;k$uxs$$^)9Y_SOk-g>p>izfN4pQnBtcXJ+cPS>bZekyKF>-HQ}41o{Z4Cv&QUtxxHq~Orqa9Hxy7oB!~#X{YXT z>NLs~a`z60j>RKNw|agc)4b3N5R7$kIzia6_w(_@Bsu@VWVrb!#+zuze{}*wT)+1Gj946kZp&?MDJ? z$t2%s7~ML`o)*rOrLn$J^h4o&pc9&)tvUmG?lBmgx`-Bx4xnvhI6b2n1s4i}OHCgC zCcT@TAi$v*hTK{YbIit(wRy#Y9Qg*)Ze9XQOjkgl_?pxj^sjWsrt7eLggPD3Hj*AV zM`_B9Ffw7=Rd#XEEc|&Z80xpj!E++k(EJI7W1=y1mBKpc$Wf&(K7(jS;1GIrTOLW6 zd!2ZH+C|I@9ZAZi5#&$A6}Y_c7&HZHg7<$Zuq~65uA7alWokT$bg;s{AX{oxD@!lr zji>6#Cz)QzC0HzGbiD|!A|oo!z#MV5Wivkn>MCd8#Ln+9wstS9dpwy=5wjwarJoV@ zW-a)ebdlLp1Hkp$c9_j-Vd)w|`(6(rjtf<=Hs}Kz|6(x7O1eWb7u2Bo<3e`HZ9E^m zOqn;NTJy~Z-=b0FQPL71*8eoB^A{K2Ch#(n$bC(C#9hZWbd>%yo z{{&ihE14C;9dS-*A!$1Hf$X^HOy<(^!w zU2hwd)ZB$7B15i<>7_+KE#Y^s*nhk&5Y%*7$7s9p`n_l<;d)d&_JRL@>DOQ0~l zoCOV0g{k4wVaB)JZ1l8a=>29D3gy-jn~hK5XoLzKJbNMQo?AjBLMEf}iJ@Q=Fb~Q_ zBgi6)7UCLxPB_Jg=^OWNqlb++iAG9}XFgd;4C*PMltiujgivUH|Ig-yU@_OKCdE zkSoV2bFQF?hc#SGkhfj=Vl7*g8j7d;m*Cv@XJB2|O>(b1gRIUdBepY+!b8mj!Oe_l z+?D%)1uslMt6RUZMXnBZ=QhC?*(WgXL%Xn|K8aO!ykUM0I{a{N2rdr>(3Z7;z55uN z9k)Q)&LliGub60*DnWYnPM9%ZKr&XFgYLM!WagjGAX}^g!L- zS^!R?r0kZykEfS*{)7WJ<%F~PR>AyY5mOm-fSSJ$!tPb^@cDuy?Pw2&b(e?Go{4us z3DfDzs^et7d^vTo-9W`=clvmmIQf4=7d((%OmF?up`%<<=_0$~%u%`l-6qe*e{Dt( z#Kp&LlyCE(a~8yGDuClGqEx|I9zR4S(JelvbZo+X&~H$o*{`MHPO2QnDu%-J<65+G z@k|v_{lJekQh-=oN6zCQlo^lCl z`(h(Xxyzzwiw%7@aFKqzGl~9lm!qCKw@B%Ui?C2`0(^X5Mkd^A!0R59=|q7tecb1Z z24bD!+e5SAN2NDR{9=O5+au`I z1uSiDH;Kt81N+{O@Zohh{cl7(jU3#Dk1CYtj%nKTM{*9yO4|pw_us)ko#$D&p&s}6 z(gSj#l?`eqR(BV`<q5RNeu%t{~1sg^cuWvN})Yu4r*lVK?yx4{w7P3N3K=GYvmzu z*>xn9nJG!-U+tp1Rz7AE(5(ovpA>BGKIP_Z0fS(-x9k(PuuIAW_m&x}Nt0AXV9WJ%>;k)t1d~&t|Z|!XoO7tQg7#&1x9=gJW#%P-``8iOuDohx= zQC#;7S_U77@4=l{A7jkDbnI_mL8RLfNpoF|QO2?*cb*wpnmOgCX$YxY;w3Y7O0jiWNAXnnj#hHhi^sc82 zlsQ(wyY)k1PWv;#pI#rdQwt(o=Mu&mt;2IyT~TX^GPjLrL7lcx(jnnUaDg_y=~{}f zcWIJaIdS0kHWKP876{Xxy1|2s)8OFERPs4j62H0jlumm!3?074l9U4lM3nc26nUiy zCq%?R=fo2@>6-yt8qxy)b?ky(91U788cEN|IQH;K9emc!h9eozm@A4KVo#zWD_V(; z8+rqDwVkl0I)?d$tHQ!p6!JCC!7}+^=r$@N`Rg_CV8%)IX2dM$b&JI-k@1*3^f57R z7qf^CRKs-1b7*q~@!IGC)~)OdU9GvG=#dO}9=^fAtO(3Zs4w+PIU)XiD{%2}MJ{(l zh|U2oU|zm5=66I3&ZVrvXz#OV_%MM?tairKm>isN|0--v%Ovw(n~|&u_XUHzIF?(- zflF~GM8$T(uwzXi>3)_hIM#v-r-!hnTn+eovX~_K7C`WRDKM3EhoEiN=rSk;Z}ZnI z(;vz9{z>@d={w;HU2oKR`-pjdzl!s&kHF0%Qs9C5Gw7!QDX3LZn$)l;0(S6Vr|9Q+QT^XmIfIWISx*rRps%um+<0)Cj9Uy2X+>@ z(a%2P>3drldhNa^X77E1|Bg=K_f{?7hL=_Ny8IeQyfcRm3Hk$iiuRZ>c`5|SIJ4~7 zAhK|L6s~G+#f=BV;jE<^T&1*BV0Uo?)Tdtu`J4z6ACN;n{VRY6o;~1u;w97u--SaN zCtz9UU)&tijDHee*HFkQc zc^ob|fHRvPp>y6jEcQ%9?MxN^c;sy1^e;L>efKy<0(E#L_TiAR2f6;2tvIl%98N!! zzofktblc}XpzI7KGT~0um__;i6VhoRdD$7q^MYNBf zf^BAF_}bsv{P!Vq+@Lg?PTUkg!vB|I=zEi>Dt{nlHACq!B_*gDX%E^023K(zJh-mQ ze^>}HvveoM%(q3F~j%Bbi=8FkYh#hpIP;^(r?7lLifg zhD{T3(fUSMH%6V9?>&SUBERAt{d8h32tb|SFL}6W&PnV}RE$ za^1p-WHg__H*cS@ap`~I-pDVow`4whZG9G_2Bt&!yeZhSPl0=1l;Tq^pNB0bw{X=O zWuDWh%sxMRi74{K;vW%k-k+fL=Q2=Gk>`IB=kTymo$OiVVE8vx18UX_@bPIMq3}V7 z;8?BLa&qq(S$%p4@!s+pt0PizXsR@Z&Isejgdt46UWWVN8yr053oCxJkR=$D;cVkf z7A^k;SD!S4&q;gX_~cgNs63c8HqRxhlP=@hW*a&}a}}L&<{k-6m4Y{^%KYNcCj9-b zpC8H2;orA5a;d-X`I^2vSi0mWfB&?DpBEm-*8h@l=Zjdpp1T@9X)x?wae?QyoaM@G z=efu3LjK{63FcZ>lYOU$z`LL$j82#f%QjgWPXxGJiJWlwN&)`11x$!U`nQRM^J(ApO*itT+wOyQJ{*w!4 zMDyL#BgB1n3#N6^4&LsQ1;M0e81`0`um7-uw<>m`)YdpIwJ(xAIBmoqC`#~<^1G$& zMdsXxM)B!MXZgYLrF^~F0{-%?3-~W97VBU(;_K&8{KuF^Y*hHe_5|wlAO8yY`TEsd z?a?3}A9_Vtbo@8QPTq#Z(oO7bjK*1w$vm_7-2a%(yjkijUvm5x-pr`tgA!AD(K1KD zAY3mv(ZmG3D_>w(r8~b{`jAfz$>S$`YWY>C2YgbTq)2gO6mPs!$D6h-KTB&&;uEV<3U&5;px+?dY27nJbK-`@O-HDhwN zTk(==64+K~K|*l?xEMs2J`6}?O`D(NR5Lqndvr7>57fC_lLz1RNsqr@y^6Qr=;muX zukn(iJAC;Q#^-t~^ZqS)SpUSHe@*GZXHuWJn!P>0Kd>D$H8*2p#8o(O{RD>T$MYA} zas1+DF(5@D`td02o)rp4}ZQ{MB`eBh-K0CKH68x`R)ux{JizN$HTl}Q z8K(8GhXKP>`obfGj&YNqN{{BzHRX%xkDsx$e6pAYvvoh+b95CoRyaY|I!V&c1rNy= zQ3cy!euo+Dnm}HbRFNMhN_5hSh4glyEv?ji3DM>I=+@4|G%?AL%Jq2DrBAyc|L#o~ z^DG}6o~pv6zT@{5u;g^%^go;NF;?^y0^xXp` zYPi^qR%A-kACs2TlxuI{@7rqdIl2nc!&B*|y&6<%pEK25VMZORJB3O1k4e;)4pJGG zOssn5Qkb7Z-_@nlZ^VpVPyI-yo;RWo+&W16NClemkka;;*)%9pntDALLU$S&(t>+K zX-n-?>eTWZB$C(CH%Sq&;LI_ye$5!vYJJ0k`plW^o1GX%uaE^LhoEzq9{sb~9&#qD zvh_yu;QpUi!swlg(MPunZPNubMZ=0_9kioUuZelnFYAb>+9csv%~$wCZ4%!7at<~Q zoq!X}Jh1ze1$qtYg~SD7rchiFM7q2ppH2O#>dl38e=?y)v)aImtfs$@?V{C8>}3_# z)30|2vDW|1YK1inz=F? zRBOgjSC`RX7hgqp|2Lklls*Ds|7p_O`k$dQ<0g!ra9r5CSP{>Je`aYf zt_Y7#-9e7`8&XZb9#CvQ05bK}P%$c=YKLv4IgMlJaIF(GQZ9zpJ=LRaQ^eXrr+B)) zES0i$SGr<|7R`BA4)yVVFv+nQO3L1{8#97vxY}=UYgkR&*0z9mTLyh^97T1aCs1Gg z#dOE~Ei}>kFWl8OrAZT8AzU8Ligp%;(Q`fsvwd3N!r{NfFZ>25EPD?b{)#ZfZ7iK% zvX7Re^uyh)t7zZoAiAu{lP(#&kY2hslG;2tOHb|yq^eHgbnQcRaSurZ&sKbgae1cT zVwMN4V-De%S=sFUtxQJu9UvXsj|!bM<*;mZ0`$A-fT`CuanCvePFK7IyY2^IY%z*$ zf00g%#*PJ#dFutbdv)l?`q@GkF}t*Jn;~h@JIMd>6vhqghmoVC==aA-Fwi#K_I{$hP&;Qdd}(qO!^7W_ z=W^q)V{soWER!V-Oa4F7zBI0;FKoM!QkpeM^B`mCtb6UX4HZ!+R7zx4f0;#Q(WH4E zmFCb)gF1U1nnlJ4Wk?D|l&NG&-t&Baf6tfa!})YRoU_+&uY0(z>%Q*GOh-ZoxqWO0 zQFho(h9c&Z<8eu3Xzx-+HRdAYB=jrJT{xeOxIK$ibeChBgZ1eAb=x?%>4yB$omJe} zCkfU|{0TX0d{>B}X|ku4lSzQgBj&?iH8S3NB76UA9ND)15VNR#EB~)<6i3zdShrAL z_QKLwR{!r^ven@pnXh@Cw9w@wlxt^HcE(Vne>yb6&x?l7kmbB&5LZxsfIOb7LpH8` zL;8Kk2)da={9UDpwCy8Op>9qzdNlYh>Vw?LRVncGOEK4ScnWu;d=1n7@DI1yOt>4S z?hw&XCuGU6GIdtXCB1HqM5;=XeJwVboU5{;C(`C~`9}NkoM-n)pQjQr<(i1xi^YUB zQlPT6;;gLI6jnND345Y*7H8!y#?Bp+VjCX}GV^X+CEeA*#3RA5TI#PXd3eT!Y@5?b ze*Ft1J$94GX5L{&F;j>g`&DuLr(=wjt$pUC+Ke|C_X{%Cfo%isa$NM6%?! z8c7P(Wp^mfC9}`#kYxfBZjV922T|(b($ZS7@y5|5#G56(aXrgXYFM6 zZxM1+h?{->(}s6V^(X1a9jMnnM{0X#7)`34imliBGcW7BN&5C}#3te<5$>2H8izTi zfA~M5IH#9v(H>yjeLgZWMyUjHC|No6F|+xNI&1zIIE%$L^zN2?(x4wz^<=_1;@Tm@ zCVY@%;Xy7@UtGe(Y#hf%{(H!5_s?R~n&)u3yJ|?xgQtv?bOb-;lme5ZFVrT#+rS=+ z_GJB|HnWTOda}z8`jI%ZC_;*Ul67H&L~q7qGRrHOx!9ggPR3c2Sr!#Uccm=FF{0h!ovbWQty? zk>$eKrXjTtWQMU7|Ke9sK2KAkdg;^vsxw7{{l|VJ64%cVuO(L{VLA-jYpK`us$G;lH(T~cjI?lPjtrY0_>1`=%?gFcPG*lc>Hb zB~k}}6TQ#DWX=h9aumj84i&3CPc?uC1#_=YU26#tq)`EvI2(=bwgDxFx3m5}OnUy0?z z1ZL^CEM{F!BeTG%2b2f>h{H1>&mG%H_M1wxe|ppKVee~<8b}kRor`FteHvLl&wxZO zx=to$&J^|+d&#PzU~?MYqJOcLk@zIa zNR>n|BL7J;uj3P#yhMNEeK>`w`eDaR(bXfMzm6=Kw~Pp8_A@h0m6#Q-OPOmyUi_k2 z3&_jpW@cdh7+(<;5WmAs~+ z|7HnFw}PNZQ=p^F3Qm5aRcg z9!>i~HwCM4V-pXMl|qfSQQJi3ztWG)_{OQ^?MqF8);2~E9etGE85I*8@evnf=PC&% zRY;P5yh5h*SS2%K(*X14c0IM}yg?lo7E`-#@&e6e+Vt}BIxbTB0&USsq^DQ2WUrVp z2{KJ(A|`b)j^CvO=~sGbkI`hosc%RilEjSjO(w5@v{NS0 zo@hKxpcB5!&?$H4lWnGA>@I~cp?1WD>vSGR@_vd7JfAC&6uv)KAFz&WckQH~XQVUY z<5kH5yAb;B@x0e#rz0TO(DVA7`E+F&1l+v#qiR5pk6J27r zhIB|oF+o2w$-P6wR}Q8@oy@y6-W!Pi+Nms0=0s0~@KG z>{9yXxDWke-AA2AKGLzFL7E|Qky;;eq;g-9h~&B=T;T4`Og!Gry`82jkY0Y8yPq^! zke@k$Zb_(SG`_8+5oyn;ebyMM-|>wJUvY;U=U_>Pwez`c8k*Hk_ez9*qG8p=|32c% z@5yxAhcafx_Sc-~csKqkuAeTLC`aZ#(jcczPSMnq$JBA%N9tN9BZyh8KqmTI3+%qw z3Vxsf?vLYG^|y!8g6rQU1m)ftbnmw%j8Rr3Hi}dvTQ{4Mn1`~U zc2hBxjT2OfhRf46rSF;3GL_82Fgr%{TO?gFa~D76%yJyG*O@tZSd?&bvzVcFRdOW! zG})U~OTP>%Gi4>EOxi&QT6jxBV7gFQV7W?JaP#gBni1MaQ?l>TcSD)fH|;as5gTxc*JuRv_}JchkcY|1(WCDKrx=iJsA zGA#dr2u?edLgu>MArn5wljEuuZ0f;jtlyQh?9+K6tp7?A_QdQ6vgKwgan5~C^1P?A z)q)!&WuiUttW;(-M6}sI!?Vc;>(}I4qatz2mlAT(k&Npy%vo66p(1M*ksXWd$Ue(G z#COLxk}M-jc25i!0oRaJ7qCzeDqcdD0MNnpAZ8@V+H!};m_ zrKHmFE&erQ8v7tti)~*gLk1Koh}Yh&G}h%Ar7=a!#^^9|)@2@zF1lpFi*z8`zeT#r?pYQz|-*l@)!&oQK1i6l(DL5?k~5o#+8 zS-NZi>%G34Wbc(`YlM8D(~QrA?^T4ol?~Yy4ZUR0djgp`ahNRc?IF`Q%w@D*v@uw+ zlx)oWMvh}`HpG7gTfXN6L31=1-zY=wsohq#cH3sE9;r_k2L0hgWL{C<_}$b{OUQ8! zE#hjMjMz)gIgIqs3u5+Lht+9a%>Ds8Hhj(y`REYF@+D%}E#;9UW8xap*bziZ>%7P& zn{#CRv_tHuY6|OLzm47MQjR?XfSAN8vRX6dv!?!+6IWo} z2PU#ho34`WBc4oQ;!8%=%#uXOsj#M3eP~lcH2E0gN1l}i6YJqi#5N?Ke7|T-RC25c zU%89v>svt_&Tc1)wJW)pO9{-Fg%_RN9z>OG63OQ)zC`ELdlF}`o4x;;u=BP*X7;48 zBy_0~yI+^EMYk=8K7W{*zHU9<|DPF=e4Ig}Iz`Cx4YzQ(Hz791b-2vSyT~cgWb!3z zfXRJW!Sy{^LN6SC#T61m`Y$b%q)rMaS9J$SY?>w6VWlWAc2yDNuCAqRM??i5j7*sn zn#k3~`0<;?^QnWxctKH}3#0Hil02Ca%>+-IFOUk?6098-C9TF^Ky_;ib3}gO`W+ZKDyK$LVWaYteS{BDIJ_-8;%K{)(LH7Dd6AIZbp&oSLBM)JPRQ zxecFiQWW&u|3PD4)zc7{VtV}z(2#jM>8$dXOvgJOou|taT$(*eV51}@Skf>@@YT79 zMl=B_-!RGyO4uyKp0Uw`n@v(L7Gz`yI%;yYbwbLs^VKHIE-*Bb-~il|YT0qy>_b76@Kc zn+kjuoAF(L#ZhzDMEZ7P8GpX58@;I!PXmRs%w}tBsQi+APFCy!b?6%}*gVS>H=A9h z89%Z)$9ZbBS3QSj|BRxQGc4)0kPK>dej4X;L`l%4CiK7AyqQ}&dX67#(o186by`;@ zA@hXX&h;Eg_R@hiu4s2F7vPynTJr}ju73E=Sq+>dS(|j&#+#pM+v+uBfnF<4Qn*Dv zEsv%vuO48C!7^^|?_!!eVo3T$W)S=1ujyB>ht#`9Oi(vVltu-L(hr7ORLOb+t-qB< zjq4+EzEvxE{q;QEykIiTdgn;hR_oHQ`JQy-QUDX#`j>N6DIf|iN`mz64zh1)J>Mg- zTd3)J!_VLGkoKL+qL}R3;IySi0;v# zOH8A&=%YZ+@m3n6r7A~b3op_#MuMEVwT!v-&`sdf_mIj+stIm&P8U@6NeKKwMsWFt zr{LTb9YIHZ4$0oA%S`^em(h;P$ARJ|q@^&J>MY>!!+=O~ID56=^M>gHmsj67pX8_1 z=|UHmezuDrAH>iSo#}$9_4l}NZj38>Eh(52rzxmAuOhH~G@tG)c}mOo|D?^w+p*1d zePJz0Dc@I;?AW14`kdUzxQs+nE_R98tF2EJ#6w9*X*08HemejAWjQ*o+C>m?w~^kk zZlT}mPLKtms$?S3BEhpaGMn4W7~MIlRQ`SwO?^FCuxv+T} zXCX;=wUK%R$Z#67UeIkPhADIKl;D)8lAzdh0fVH1`N?cP6RN36G#zvW(J4~}xtCQ0 zjiaNqhzllmd#^JF$ts-dg-jxP;2R@*^A4jwyM;;0?PM0Vhtt#7nnXTpa@@=UkI`4ZY zzIXdJ^JR7ocXv@5cjjF@_YZ%i4Qmz%{tHv1RTj~lWJEHR>d0Y^eo`TUWz(5^Y_~A& z4$)1Ocj>Y0NUHUtoa@|UM_xAVB@^-!NYwrlw6 zrY6jtfquU1u^w(|$$jCT6>ZM+-hSG5Fp9pIbdx^!O<|Ua-eK;aJ4Tb%yU-JMf%G1K zfblBZ%k6NxKsQTorCm!!1lNof(oLx%H1~}vwG|wuQmP;5wMO7n*Orq_1CzL~ye@9a zh$;OSzl9lD?#$sSkGc6Hn#|?PQ@B`bCAvt-h}b_VW{!S7$q1tdH!?v6yGDEPM{*Z& zTc-7M>e}CM_2Z=+Gu|IDClE8o?=pX{&p1-{po4LH?Mz!bW2t@NF|J^;6!qAUNu8BC z=%Vmkx>BT?s#>0;1qO9=Lil~UR8*H14@c6i&d=!47(R8l(@*_V?780D^#dFGeS zL#AML26eYI5+u_^ntvjI2EF`2%j4_mwDUuBPyS_UAFV{oC(8GK`6- zypDJGg)qi>ay-$Wxy;|O8JurzAgw5S%TFYQ-CcYhv7E`2Cqnl@)v z=Nd2LUiaVQ+DIoY@Hxi~Yc^I@8im7FqaN;Q-^<5)Z?7;Z%nOa8<8=o zAW~<|$vcyS%+d%S;`=+1DLZ6C;|{N*;#O~|#@ROdc^#jw3UMH{d!rb!volG|Jq_}; z}QerB1-TC%1wo-4<%n9&hue%V|I(KvnW->Mukh_W5+-!jZ znK!?j?jATnn_qq9ew?m@9o7GsjTzx2Qb~gU&Dfn9@5|?-Cy(i`L7{eOS~x6SsYCm6 zRH^dLQQGCH$itRECf{5n#|i}A^XrJ-cXUsr7Ohr>>Bk7c9HpFopafJho<8F7xN=aXQOiiy2GGp$B(*(bkAM(tha^ z!-~nWA0I2T2i3$_5}ZY&>`aKbv;tEnIzZF?O30jt+T`&nHFkk;&ZWFqitX;@Gym4) z(oW6o!r9~)a;*G1@mwQ<>zHz7R;+%OIjXIWBJ0E#!lQLGPr-ma+{=*zKPcHXU&w1-`A6ob#t?E>3u3Rn!W;CSGhLtj z$<8Z;-4gPLJZCX$nJ|exKKXPait*dOg&*)L_sUW+WfHmj zB%?DdP0#J_=fvMzve9Sk*~u~4MBzSWHz&+zPh0Ux=}#W<{2fVNq*XA%9^oWwU^#0f zu1}g&3fXy@YHZhq3uN@idt#6>jl{`0(1Hik>G?4!Qb)(L!6LKRj?QSJ8F-as*2uEc z)`_t#?uP7XA>M3meu@@;a3DYJg}R*YO$0ezAxn@Z0on0_nFl-QyPVmaPv9lajlY`m z77x%(zSZQVu$QRPn#H<))+Y`d&B*SxQKWj_6C!@_4_SC`5BYrX7l{ekE7bOkBMK73 zbgAD#;`fPgi}aM)_uKqQ)`Lv0RbvPK+S5dw9h{jLE&e3$LKU4qbq8&zWZ2X1CE261 zZD`S56>`O}f!4;|qZ%cA?!%c)xWQPO{N8O$qr+a1>XUcK=z1gSD{`B26=IHQo>z&u zYBNbWJ&#?oNnh9_>adxCo5{-sGibq+24?NsV9um7nMmo$vHl7h$YXk(9P1Ez+qZ>~ zWhY!1FQ>&MdfPVo^r1WHP58%X27l$=sjeeyN9sw#nj&U?Ry}$5LYlerK$#{&M0fQJlMT zB#~)yCI4eT+T-fJP-L~xdsO)O-|R>Kzy7B}5!syoga0Y7s2idyK7pFX2&|ht4BLFK zf}8D2ct=`5v@079J=}tR@7aJ}n@vNrzGJk=(f&iXQK8&LWTFv>TEvsk0h_6K{idlH+Na`yPIWBle+3E_G=RwcD{#U+3A)SY zU~zdXylD{Q|3VD#&#RjurE&@U6w`;14^u!tVggG0*~@$XxS416;Q`P5Kn;x6H-hqn z6llJ421>6NLTNz*h^jn-s*{;8sy7>?r9@%Dq;YV?Y&-;g0X(X?1nWuI;%JBI*vhN} z?HH>>CqG_7vF~1?j3|UdSO!muH^=urO~K+XmEmsN0ytK-4$QaN!{Ke0;O4gk(9ZCH z+8181CwURJ*=37E#;mcB_tD!mt!TCUQ}n%016IhV@k}=S=Ea||LkC|n=!U5V3i_IZPDt0H zO9Ky)seCESNsa~FvIE3knuF)7LFnCf1ERP}SZ5jzbz5YRQk)aI_cj=9{osL$&;5rQ z?e`(o;0;LQq82Lb8irfGouJv$2#$B{p`rRE1Y3&Z|IB3ZxR@?@y>JVd-`0eWcMx2- zr3eMLzreWO#}H~=3R7b?gKZuH+c+QSUhe~?K1<;8m#Nsw!2*XmV~ifCVWp;I$X}KX zOSAK#{oVl>n?~`ed}WX#w(kmYUb6*%=j2iIa_FqmVW5``J2kEu{Lf`%x!`84u_=NmHwZsk9eVhT;4nGB(aw&Ys zRUF$-lf|pcX5fYD)3NLvU0g6n7Hdgs;e>^1*d}cfeyAue^ig?*X7uskm%1Zt^;|Xe(^Jjx*e4@gm7|FLQ?EhEjHjUE za~rxu>fubCJUn%n2#HY>z;sXo9@{O0Uh6{;c6&4I^4bM2UKzu87bUnPVF`~u`hwF~ zF<9#agHn+_%*e0+gA<}q-XjB>*Xux4s3mAEdIIk3BS=!Y4`+uPVbdoy>{X_Lb)G2V zy!@X~SQ!a!KMLU1!3&UAo(-v<)yOAr7a9oOj>_)EqJ$|&P+!b3WUiHj7FV4|UKgLB z+l>?{ja)-wc{S*W(nL7FU>L29|A)pj5>bF-K8z1M587`FfG-;h1IOg?zLz@Km7g4826}m6-7~Kh##vAsH!@5g`!91fC^o~2AMN93F;`1$NSy2G;$&EnI z$Mlg^mnKq8%|tJ}uEWO07hw0k16*En@E|D&4SBvow{-iFx#MHRH^_or<_&@6wKEVqB@0Y9T!32V5V$hi5LWy= z4%^qo!-w2_a4t-Ok)abHzUTmaKD!#+JP$z21$Q{2?gA;*b6}~nEc!Zgkk`9z3+hl` zgT&wTHfi%UqFFelGf}#7ByKt!Tf<0ie^5K;q~@c<{^v z3L7WkD04adVcA4HLFXsvpU?(@_Id~`XW(A>CTMI<0NLfK;B+(xCp;gN+9Qb7PNd4paWa4BHgfjWOT|O?cUjk;;c>Jzk5F5x@;S) z9a)3cBNKGwxgwGqy~Ep4@c`ORK7o5X5nt*8+P$ z-0lU48&KFR;RBCP27&vraM-ZhAH@Dzqo4L~d1c@0c%47{c<$%Uz`p1VNZ(Wkb&B=S zDK-w@5T4g^H4V(1Q^02%$Aj!e9mt)v63|*7$Z;`5bFa@q-DlNNR+kAXxm^$5HkGh^ zavnG@iif?qx#))a9Q5C@ZeCk)7SH_QRirL{0OdtSpbf6C(R;u9Xrsb8v@tmhy+>I{ zgMS{nE2^R7SRtH(G8k9e4qI+MhSFgY5bzFx+VkDlZ6h-dqjwf9#<=n1X}-J@Dylg4}2IFn+okl!R%+f?)!| z9*5xO+-R85;s?`uU13qk2ugX}h0NI-Xxp0yXrXQoqO;_1(3l$jv{eC{T$0B1`&99f ztHwCiP8VT+^)TS2LzvVduy?O*=?H7U5Rm43b zQ?b|6Y4}OwWPIxT0hs6!48!8lu(Qw~+7&zC-i1$aX5AC$C=CS*QEl{3vyZpvz(iCx zTElyhnF*_Y2SStFddMAF1D(q|VfC$#pc*y+)e8M^Mp+uV+bv+$tkn>AR|8&U?m)9& zJVPh`2~f|r31G<|gZF+C_~Jw{{P(UTet)hTcrN|$?c`sWdh82ql^73A$@3sp!UfcH zXF$S|-AL@(S>zaBj`CYX!EkREwEijtj~hi0Ff|E^m7*cKBn-YJM}Uu66qxdNpqce{ z=#ESTdg0=QZYIZ~4v(|w%G@2uN;sxf_*WM1ESreWCCcFP&CO6@8wR6)h50Tb2ADnJ z(36n{XXpCB$)rPIlQkC}tiLbx8gGSxv>JG369Ee!%c0UA=13)aCK9_4gBqu#!Ma7I zFl}2a82vsE5g$TdwzY7PL0h3sXOzg;k%TVQG6Xq_zs6 zIr~0*3BLt>TkAo3L<{1(mqYX&0zH~O5c?Etv z%`+`%+{E!qFgKuYHMN%9b=fy*_ni7atB3QM~ z8T=L;fO^pa)R!T&)#g#-R3P_}mn zMq;I~43@Y>7qJ zcD_LKI|q@)Q(dT>_!Uy`e}{ukKfwFq7iiEMgu~HeP&57qg!+o$UfE2Pc~}Z&epG>n z`}CotU^i;hDM#0OD9mqIb6xTR)far_c@Tlwx?45Q9UY|JyF=bw$*x~_(&sU*f z7YF1z8j3VMj-wQt)nJNuKu7&z=;O5DaC;ZX?d*cH1y8{&H3Qy^n*wCkWbho(g^tMv z;B4aw`>Zv=ZlyZp|5JeRe`?{%)60v|1d`SCz%yUQ)QI=?A<#zYzqlJRq%o71*u`hWGXAaHCQJ$Q5yzIqn1M za%e-x&6QxoUsb5PL?BUnB{-Pog8k8Q&}xf=jD0Fl`9Tzh?n*$_T4mUi-6#Crxp4UF z7C5xR4Gi5EqvU5N(MLfb%K5JfP1xT6vlc&s>$4g`qxb?G+G`FAe(i#UKU-k`E(c-1 z!iWA-*C73UIh_3+0gvWoz{g2>aP46h>@+EXzwu$nTR#RJxE_vzOcGGRcw6+JhX>l+ zmWrN?JVXFCq5RT3n0)s!_LFc;%4E_rO&4+Oi>yZSyp(Ri$CydXh3*cr|8j#`; zP+f2kK4l+*>wnk5f70@>Qoa?P^r=N{mMy5r{0!QpHOdPp&_~)yX~<@h18O@o!V5my z#(V61n-^w!4%$AH!R@7`;Pvn_yslJ*sDziu>-lG7{H_3P=w1nrcWnkc%RR!owF)NL zT!ieM!W>Qcz&o1`>01^+BkK(Xf4m?vDi-=KzJ>FPMxiq07i?Jj6c*%V!K#H>a8kYq zuIvher4K%%U3HS7^J=7_Q5OlX37LHzl zfY>tdXg>iz+KKy zP2^@<0#m;}gyj+sVZza?u(Co0c~{Ipi@s?fUwd0rbNwP5yjBTQ3UWbX&lwOIQ$dGL zYoSeffTB_ykdC7~j8ASwQU@-e;_R2`*k~g(cs7Cm=O%Es=0M6V5xGX!qKnsrb(};F zIyvwe75D4I_6x=!)IGyrmqXy;91m8xArSK+365+Ef?kIx;kidc_pJ~x?{P+Z%OjDb zY%Utynva4P^zkD8JB2(N15vbFFv)b`xw~8S-@<@ z6^52y%PX?pTt1a+(PhvVhm$z*1(l$I}njKhyRYh;^h=gLe9^X(Y*#Ubo#|( zWc)%KWKPb3)>$*4>3SNva?c-4nRN&qoVgh76wLvbtI@sK%bWE zMH4r;ptQ165T_Rnjd^jf+4LZPp|h`(Xjoa=gIb>Hw%~dZVLP9Z^kc2pS-X=uzu*L_f)*xjPKe zv$c|F%fdHkkzNj}aZg6e3*!*aM+8}ZEac70U&l*I^W*K*2}1z~@=%pbD|%7ii`FNp z!huO5F!69dy2HDMrtEQpvQmH0TkHq&Hh$2PlmkmQWrB9YIXKhg3*`qZ;Lzn-7$_Bh z&yRC3vRWD*E;NGl))^quHVu}Q=bMQJ8R29rpfEhS`?^ngoks8qx-bj2zg1 zE)RMPGaGXr1;R#?OQ`*VIJoNnM!a+4 zFt}$J)D}x&zXnBoU_b@y>wScE2_=x95dw49uL1i98$jEC4cPf?g(R;K(6KVbdEv9L ztJ*A__EZP&*R#U*YIfN4@M5f{V~oRN&%n3-bl7f~3$0d}&@Zf)nCVt1pibCRE#HmG zCpg0B#}lw)Y6SGIh!o-wVxSS%hfKWwqE^YrNP4RiRL3!pKFJBjV#dLL-$b#G>lnBN z3hRrz9iToy@DgtaoV$7hUiHofkK}$tCsN;|W4k+0{^%`KF9X=afx$y_r(>O51>El$ zi#FWffYv4+K=sd0BmMKY(Q^JyB&t6jJ>PvEO}nFp3r^_ah-MA!-z&spx);Nn%WkmS zU_D6lFsxL50D3F`fNaJu@ROCm_b;pB#2@O|Yqva}byfmPhe=|?S=u=6!eo4HnB zL*wi`I5qs5cc_fzWqyd^tx3%0&E60X>q}GNc%TowX+c6vt{4{dMu1_AKh#x;LP;aS zF12>JPRQT&h)%~2yHnu2hW1P zHJS;Nq;5f$aG1zc@fm!XaT&e@X<@e)8aO*d1s^tf14T3Pp)j%t!X49K$zcx8Bs_wV zqpcA8{UMa+@*&0MK76~PinTUO#`+e9;VFX z!-LFgaI2;mlomXI#CI*=Vk!I|tJ7hDRUsG-7lJ}n1q3?gLB+zU5Ha!z`F*ZKyVp0N zkF$=#3>UlEJ#P@pA(Vu z+9dQvH42qn*T;VwhytY7C*O>>H}K@(Gv^J5clZ z1n5XT2L&_>S|c;T`$Q@DYE?s0d?}ndaTXk6J>amgE-S!i;OE&8$bJ3_*>)A63>QyS zS(Au3a~2|5gJwaef~cs#Tkmw{tq4cuRz2XmjRKv1(h%uJL6$Hl79Gfy0bU4EdgV;v~T z-wBy0zu{%xKFb?zImi2`eUg``2QYQJ95e(Bq4#q{VcJ$hFnp>9b|;PC#NSDvVOfY) zCHzGh>t#XZVG%OAafRn?Y>W0q*`f1pzGz%)3S!oYI0KWbO(CZEClXoQjuNK?pu@SLaIh}~HqAT*^W^8iZo^?-dfXJW zex*9fh;=|c@;l*3QxJ40N5SIsB`_i`%uOaaa6yP;N1O_V%ZH{z=XMv6RbpZEgfXls zA3z2Q{b+J@2Qu%tj(P&k;1q8+ob~dBWu5-8H=BVY2`eDFYc-S$>CrmcU36&Q%lgn2LK3F|8}SnJP&Z=TBFI6Vf;L`67PhPmes`(D^(a0B_y#gKi9o;6^p^rN|(fV{9NOk;*q;3zQcf-HYBN-{sPsl^N9m`R|>GSCL*KAaB#sZFhHG{uT zgfXgS2xWJS$IPe{R_zeM$=e=)Tjvbyvt0%wcgA-#-~Ts3d*NT; z0dP5R3a(g(3UTpgXxGJ?2(oXX#a*dzO6o4KeFAtY_Yf?@^{{lmE}pnu8TXXPW79p) zVI1#0yr!Sv@Q1GuE5zgjvwfho*$<-K+`#o;B>XbC1-==NApLb8bfy-;{FqR98g~YI zf*heGKNSvT$AX_(G;}Nog#-%&^xOLlPp3%|+3qq#Q>y2}M)R3)R$dc+%1guUt-@TC zr;N*QN?=~gbuc~L4SgN&;QO*SuyjQ$1RRV7m-=N8_P`!^TjbzbiYPucG7N76+u&SK zDvbR3jn02jg%g$PBe&MYeltpf3k?=Dh8Jv?c;r-jQ&^=WhXKk5@&7$Vu#(D$n#OpxnzkAT% zUy^VoRSwn(>tpTG4+wmFQBdM_v{mSlxolJcqZ~Q>Z^u7av3@4Ht!Rc$mcHiQ6H3Qt zUk^p6J}00Lxt{3EB0q#oLO@PC3|1M3g3M?*Oc2Ivm-rk=e3b;l>-n(oi7UibX~1qp z6UaT@kHSyO;@?v>@z)9Jc=<eUFg$xY1chotO0o$!3AHY7UyI@BHOlz$QhD63BZ()Ne*jNmOz7f< zA+PWr+^<)Ivk~J#`;9!9R{cetNBE#P@DgM`i{Tx!C*o20Jap*DNwnnBNyMmaLtVSx z^Q6MZctJPU@eC|f(QLbCFyq>7NI6>z?@Oz}-u5AM={7;#&#TbCJ_S0Z^>O(1nfOMZ zCcd~|2miFUhM3$@H1pklba`+yadKYJ$yR`WA{{qcMcoiqP5LP{b(UVpNmlA ze_v7EEmQnR%>sK{PsKWWHE_9%FXRs=LhrXcxN;-~{`7^S_EX+y#rWSm7a=~Br)`4E zW6f}XivbSbp@Q9a1;V0FxnQ^A9L${;0*=uIaFo{$QDLuOlzfA)EBLS@q6R9OF9>V5 zIPln(1Y=V!flf>b9Dm3ObGsc{5fUfV4OAfM^d?j-stzqB)=^I zsQ8u%T_(pNz)Rt)Nj$jZ5wCf9i4N8^F>-%C2b-;^<=2dCJF*G1IQP0e+lF6qiCzQC^hstDjw-X=Tl5zMXIp3^f(K?(_X@?yLRX`x@$QX@;P^?n%Efq8NfATE{wb@tJ4f9XcB@Ed^51^-}5v=lD=-VRS!r9qiiEOf2T zfFvnl49iYMwN@!;J-Z9t8>mF-604AHg9O?nJsWK~XNwA-i(@+*Ni6qD9cRZ(!k3kd z;B@#h=;j}WyzWrQ677dKLLMm1@Bvt^y$C*YbUqaRWkeYY;!Kg15Xv0j|Ts8@@ z{WKYvUsMLgw*;_vup8F4HbY;HHl8MJj5jA3;MqeeIPIl0E}pH7r>Du|PE}Pr>TZC2 zHZH(AwR3RO&?G!#SsT*)!y(%Z7tyvA&gg>h`{K{@@q0sO+@|l03#JIrW04YMdTqNHG2o9ZUdiEu_B!}wf;?_$ z7xu;f#e&F!WT=fR1#43wS20`;{U@>^G=f458?|=5jg<;!p)5`U5_-PPKdK?5HUjyK&Z38_1Jq~ZGpM*bEYT^AB10Z`$ z3ZtXF5L9&s#FAx!Es=$;0Yf+}+k;Gt;*qUyDhmCp3opJ3 zxqNv8NXwUlw9ZlRa()48=hcDqRu_;jnuI^LA)K|t65n_se0 z@i0T5q^?MD)5An39VGu%1{Ibh$B_j)#gn-<(pqlKO`3 zK2U>?_cyW-b}jS|h?h%W3HIAqJ~o?ZEuR2t0f) zg^}0xeSv(r3U}<2D$+g~V&3B3NSxaR!w;)+ z)yFhBi+^9>?OcbvQ~J1DFVO75=X5S7pA2uPVoQfWN|jEuGufByHS?+V=W()Hql&49 zo^WYAk1Iae*botd!RKcpeUc%^iIU6rw_?xS+l7nJ7|jjFXzom_JGaBmio5CTz}4(@ z=2A@hbAeV`+^5pXbn;FfNd{Jv((mW=@?0s^+EtVLomw(%eNFr4aJY4MGPEkvkoSH! zb`6}*4QgA#X?9-1?fEr^QwrG&|I%U@com~8t_0l|-JpzV-|0f03I_a9MU3WBG{~2r z>cLu6`$wYv{CjfGw!p^v!H9W14b$~`w0yF{0fcMuDQv)#vkF{2-;7ggAK~x%1}XA)=-<4W8Z?Y3rL|QO zc)BaCjjJb5Wdqc-cE|feR(O;h3N?{?n=n5SU$%!~gG3t-XTK!QTM5|{_R^8)42(_7 zfp26s@)mAK<;GUTC#!KAE~{{=2byqT+$1tinMXFq!zs5ogXaBx2G9I=n7jE78i!S3 zZlDiUn$M+cwuxl7Kb}-ueoO2s6q0pQ(xE+PbzkEzrF2V2+kr{Gg@cCDP66tXqFml6Y z$4JCDug9vWiTJZ&5F-DMgx!;I&<`Jrr9~Bxo$p9~H5te9!kM%% z;t~|4kI}XM5gI-yU~=gPvSXDwtMRH_&sp!WQ(pIs5-xCIWM;!ys6D)w(3jgfn8P_w}i#2gU$V>87( z6Saw4Cyb6X$JTkfBp%xLBxh5fNmd%E(duQ#(IeqMtg9=*u(#n@>)aLh7rJ9ccXxEn zG{zp0!|eK|6fre9$T<>+NAove#Emjt8vXAf$7=s;gRXOdOle9{<_PL9|1 z(CqP7>CO2@3bs9spv?`q6xNJY-S6Urhcb4Jwnf?{PXvg2fwfAzF!ps00uN_mty%)^ zt!_h3v{xyvLJ=;KU2Cb%(MVTavOMxQn z2$V10!146gxZ|@E_DMrw8~Ku!e+?na6IFCQ_9!hnUrjY(Pw8sZQ>bP=M#qYG==S+M z>KEv6s#Tr2TR$|ou9v?c{QFV7JtUy1>?&T}Ig91W`;fMGGv+M_#-!~#VA0D7D_=O{ zZK(wc4P7zk!eY8SI*_hLtfJK`^Qm}b862l1qrE%}QJWXy&7{|K;nx$==y*bs(^Zsv zelzCGKa86L%JAQrOk{6trka2CR9LW&LLv&Nr^N-h@2kNLgM)b2Z5sB}Z^aekHSkUd z!uyaB_&l`_Ds(Q=mVbrBJ^4*V<8--@Ynt3yV+}62<1_x{I%CP0F|b%Z9oxpZ!S>!X zSQ{O{)Jcir>=}mB&Htb$&i$HqG`W{|bhwcGKUg|glUtJ8hJZ;wQTs&HFyba+=*AHu zqcRR9hwqZdSqu1GnT|)xC*wwcZSKOck9c?YJ3el&2j3oy?Uoa9{SuF$i9CinG`ESw9F!fqH{8+y=w}!i5k+2ds7iNbF`?dpQYUQI@~)OeNKD1E_Wtb znS-+_9_1+EfzA(d{aj2hil5@R$$k9O5MYp#fNM!n@cvi|*Orq|lgknEQ{?FSKSQXy z$ei>_hEr1~?$ZW6uD(%?TQ|23amPb&_*w!cm877vdJ*yp_F`<+0UWW773)P5`u`C* zX19YV3qODW1w&(a3al2N#8St@sEdk$$D?#Q(d!dE%<7KVA3ZSApo09T^hEtFk^it8 zjYC1{)adwwc8h%X-Zm>ygLQ)}wvp;J#Lqt?g_KNJVfO4mD65P`2J4KV<1Z`Krzj%r5jC-U)H1VyyiZ&rSU2LL`zcH`PDjS*9dM8JN9C}~UR{sGzAFh(9hHDf{SHgY zUMrLE@|I-q>wZ*F84d5oQ1p5mi6Oz$Fl4APxABiDck)0F?(KAaZdqOf&8=2L(1uPp z+2;cV+MS{`&rD%u?Ew{W7t(Sm5-;}U!XO|SEmy7wgT$p@hqUEHaVV3Q7lv}b{J-Gt{} zQ}M!XA#6e?!qTB1LfnjysFI1XwrQ~WwGdUZR{ByI2JQ1}V3xKSpBDt-ho}#jZ8%Ho zFNV>IPbaB5_YZ!As&fPTsc|NCKTtkFihQ#$SgfB3Ns$bF`f0fpaiNUFryn*SBQKcBIvI-cf?>WWLThOnsY3*UD?$-1sDDh`jp z;a~o^{9BXTR{98Y{GXw0z+p^_b->NXCRo2g51%xBZg|J4$&zw3*f z$+a}uPTZ-CuEWN*qtG$mh7SweP<O6x6%#Wt0+9zm(K^ASBl}G)5ZlH0(CHiH5jNE&jq?F+os7q3^?6*M{f2-cr%3d-`UQ1_++Z}->1u%iYGVh=&Zd@o8>%kcH=F}%Eg65NUZ zu=3eGER)>AGrvkqyWUQ-N;R>rP!EgWe5X+l*0yH3FH?u%!dp^DuQth{Jvox~Uk*qrxc4+# zCk7(Gb`I>W`5~`GAGA6jY4xR;# z=;XB_82rx`o8nDyUw)jQ!6{Gctsz9Er{BqF0~n1{Xh zi!nWSFGj>J7kAdNIB_r^3;K(Ff7eXB9~6V>&he-fYu}a@34(nF!N$S~ZtI8Rj?NhD zcjd9D-W=CsJLB}9o^Z9AhkH7U5O!lbM*dt4<89w7o1;*ss%jmYaA`i- z4jx9K{U*_f+f^h@2}6GH672FB1J!VMkWx4{jm^UXQKxv;RD^pYvM{KBDn2P~Lg6up ze6SLiqp!-Hy4{YNyBE=98%A|!0%+B|B~)hXN5(HLF#Oaa{1tWF6Q}cGt91>XO5P%? z;XR6?@8Xp2L)@J30t3Fhg;MB6{1-kQlDiu*W$tEt*AK@~7Zoh8wZX;%#<2eQSUh_b zWZvTpnM^oK7dL#OU-N&6xsU@fM6CJdS(9+;R~<#XkD-rWB$Sugmu_F$MGyNNqP&t# zRAh0CPUJj9$D5~6HhKoZ?hba06wj*1ahRdG0J?U7(KyDGa7YF4Rf8VD6IWCy_^4=HfyM(?MM*Tt=xjeI~T!KGYkXIRwJ$D6waEB zrNX^RZ^oynK{!3uQ{E9z6ci|EtHH0JeXz`GrH+RW zNaJu3*;T~SMw3gFQT&(O&AVdT#7_9v7>JmZ9Z>4*hY-KMa9p2+{@DpAj}61Rb-l3T z`Uug(=!xGaCL>E`h%t}sF+Al19oYGS8rGa3T@N{a1RR2Hza0oL=z{J~2UA{-H)R_w zB+DUb^uXd6bS9jE>wN*O6ZfH`lQy@`#*phV%#eHBq{0R7Rl<|gy7>CtQuJPIfup4# zCJTNTSuDl(XBJrCbA*EJ`@&&GKU~&dg3|?!l=0dFP2#Lo*uxu7we7$noKTcK3gsS? z;Ogj4Yuay6ZkamHJnSs)Uo~Mf@D=Ikek9>%9S!aN7{^QAz~WB}9<)~DmFYqH&pMZm zpMOB_2HhZa(<)r=bs4pNY9V`dK-B*BA!>ChqQ%et{QGXWNnTR7)St9e{C!De>G~vg zC9lpBN?Q_2>`^T>&wYWnslRbfrolbP(BcZ0u0(ffBJQ_kLFecJ%pbZNPxys!v|fru zK7OzmbA@cSCQVDAHwx_haR2y5D5jwrHhWEGeqA#%-OZ7g&MA3@owhMS}ei6=9wU}#t z8S4K;&UoW|{1_XB$K`vF@GlJR|Ec4{E%BUP+Y_Z?O_;S^kF2J-(Tb*i^l-QnjWCeW zfZ0~mjY8;A{#mjfVuAB>xQ}hph6*ZG9 zWHzUW(!Gl4pUn@+R&@)C%97AH<#dw2Scmtwwqu!!I3Gp%Vuh9|)jlz#bt`{Lp5MGL zu`O(-(47MjCYgg3yCvv-q?z&(o2X&xBTCPGNtu4C=v8Wr&;gpz&lyX*r{&6*L1J5j;}^ zjia+AC!Nkn{D$3^tn%@s^*h%id+j&~emJ8ey3nXD-H2%rtiidDgX8!a;XqQ zDP5*e6&&f|3!=-M4P6~?OjqwIk$i9jDr4hCDri5ZU3J5ZS7rEBTa7cDuZtd;TAWb6 zO$oMn)Z^JzO8jt~J|0WQ)8YA;A?{D7?2SgFU0;mN@WQS%AB4Y{f#V&UabQy{OcM4Y zBWE7=tV@A~_`D9gb_8ngH)G=~I|x4pAyDp!ueOtM^oT;nN8tcu7;M9ryu~mTdzJdYxsu>9PbInj-Du2sMj9UAvslKi|A`P>5SiGP>MHcH4X3Z?`%=u~J|s_8qknx~l6$~Qa??>pd%Y&c_DCl7W)t-fiKp3{ zqGzZ5&P1 zO`^N|chdRRc=~B66L|q+a>=wIQwukm8JL9+*#$62Itafd3AnuHIhi#pq1!nv6n+0o zR=RCu<=Y*JPdlilx`*i1U5<9o9q_jlYsc6vXfaxWCdW7^iM6Cc+@F^R2V-(wig4^`K%ls#F|Na?Ta43Pkd@rSgX~opHYdLKd_ED|U7ScajO63-hDShmG zsIKzB`5vN|B(#xMw@sp5vHj@enXz=QI-G`Nj79bD;n>nT4y|__MNh;YER0RTnOAWl zKePfGJwx&MM+zo2CL!l?6c$DWVFN!^{0{nHT&^!1QkqHf_7vU8uBLXqN0f8?2Bo~$ zMBn~8;O}c<^sP*29!P^fGI0F#25|1VZVRnYewe z7d?;FqMD`Fq#SQbO7WY~PdtyNi8Z^&UBJ)Y5oMe^0;-2Xb@3pvNBTg$fA68uNhxIA zSVAXo8rRO%K@nYn9MMBzb43f^!(Fjc#-n?ZGB)Q#PNyh4%RU zcmSpy9*%GNd!TeG6U)-mkUSs*u~`-!p#7x0c^}nj?WCZcDP*6xnez2|lYN5^EnCx{Y&31?bh0!3{Mm(^ z4(3Qy4L86fF96za7s6?YHmpya#37sWFcJNBvqw~*Ze=mWBt%e)PYJ0i{HR3CWOzI5 z02Zr=y!_1Bq8EKQw5vCPX$8Y6cOi^YhM;xDe7M9-LEy#iNS{}OknMF)wm&P@l?=>u zKS;BxZqiiwDf0L3iXy!L^q3n7o!=pFD%^oywF{7bKNWj|60us$mnc;;gYwQ%7`1;i zPUT4uqIp>IZrL75h0%S9mcd_1v!5-!7=Dy)pSVC?y>_Pj!SGIYJ2pPd7rs5(5ONXyS3L zG3JY!VQ|g~%qqJ8qvH&IfA`@|{TsR+|A`*3&$PhwA-$TWih#i)n@auB(s?*yx^6>^ z*e~?2%0+mw$m2(bAvz`%`ESypvnU*+mE+)iA{$R^a4HMF9sjexB-u&)ERyQ`&*A1hn zYg!1so3w#)hIzqtiW8`2C>Bn(MYqt^_F^cR|L4TJtlh?FxtWpX={?jdJ zi<81EeHg{sdy)3N-ZcK5C6P{FHukDfXh!m<1gmF~PpwM~ENcm%^QkE;mScL~R>%V1om z56ky`5xK|={ts2LJ^X>>u6G34HLfN8&tbY!co6sfN^qoaF=`bhm_9cL52B8tm!trL z6bCT0|72p`FR8FW3mbi2()!y|NcG_=ik-TN>h^A;YU3mDog9ntTcYtsX$Uqc^@D6> ze>6At!&K^tUc3!e?a!BJ-~J=1Qg13Y4c%ur$Ru4ba z1|$3AP^gIExvA(Q{!Ue|mGHCO7?&!(khh^4VicP&e##`QYzN{>B%tpz zs9aC%{r}`pPe}_U*XJYpfC97bo<-!x(@3o7C+Z`Hu$LIab9rARUvv<6ynSGI(+H-g zl`-#K1*Hz$Nm?b*WL0!@2P<5GMDV!vi% z!K1OLEnbN}N9Mue_Au<-=0bKGqDlASK6-6cM+-xGA^L$knA=b=`##9bi9r0beNex> z7pkJK{ZoQ~2~%!k&_U7vn(z=R{<&1@A@;6MlF8YB3$4%Ii_zB(!)?GJTx%^snW_$A zI$Pm|^>p0sFYdWJRAF=U0{u=oN#Q#eh%=2Y8m@K4M};k>Sr35Y%XU--iMnFG$Y5sw zM93&@uI6B8F8HJ-S9AUcW?p)Uvy1+qaKR@y6nupDyfrA_JR7psQ5fX9K=cBa&{gB} zlyl}PnJ2W8dT0@iAAXVK;%jvKZR(*P4a+&gDQ}5%ZD0fXwwRn z#pEewUwHiRpt7VBlBrrRCD&&Rqc@Xem{#D8#rG$P=aoNh%~nv3=r^pd%cOgUqUiIf zv1lEWj+$ZdNZNY@gX&eeA_IL+f4Byhuu_G4kbMWKDlK>w_8f-dY#Am#-_=es*m+6N z>tGUyCTP z>>}N&Y@(Mwx9OS7NOYP%1uGI|xT)%Z(G54Tf5j_=RNNLBq5=#*De7-w0Z7ycLsZIq zIQ`v$x?dwPuW>Pw-;BoH7q=jta}yH75_lY0i5ux7DNSz&z5EkNZ||L^85dV#Qtu$F z=!imGTmYUm8gPPE5AMUcUfjP(JErpSaqzvlO!TXk;IutM!PF%5sOf^`yEWlbE7rWW&iL?XD4CBrL<1y0 z$Z7sfN^8-izdN*PT)WuAJ#?pe!4KfR|1CqTk?3Ve<)(EKi|h*RnG(GefC5V|Pv#f7Ix$Z7Tt`qn9-}~UMT^Afa;)d(n`{G3Z zF#H{nicZG+ak^(J;tRS#yQeLhSG(ZG&K}rXT14Gaipl179%b+X1-SG>tz-Xl zBn8pQND~U%vWPq$?gg_Ey_lm?;rU+-Ufqi&rH!ZP-K!h)dD4FJanKXFpJCAL<%;oR z9I#ugu`5QL2B#{=l_Umxixsd=S&ros7QpmgAfk?YiQ(RRkkmB^3&Q=y9Y8ke9aF*8 zl&%o3LBrqJ5De%u1qI5pF(POrwptd@{W}M#GVdrU`IpjPz9-o_MoM%%VkObL{UkXj zr;zKuRdi@W3D@>_ zqOxdHM-hCvBM9Fs^5^FlA$5|N<<(&Vr^Wpc6=#9qRhd}-OMp+JfJ3^OIPY^6C-a-| zR=-K)2rKX`?J~)(yrylZ)DiruGoIa$LEq5};fs0X4s^n%!G2Kr6ou5hB$O{ngTV$% zIMf2~zYav~WCtW^yp_~gn$g2T4vQOF*xe zY2u^ZFZzB&6KATGL>5NO`?zxr`q84E9+wa0Bn#{k^BbH8nc?CgQ#9q?r^H?M_z`Xg zC+P?@yzsyg-$1;7u@-+K+(e)IT9QO=C8JL}$$DTg^%VExPOC&MxcwI>Rhjc%9093x zHL)h_~PT!|+O^YLEfe~JcNM!fG^Tx!$e7C*iR5`D?V={s=L zWDO3j+lT;h*89^rR?L%f;rcw`ITeY7JGs`JdsRM|>z+N5>#IDJ<9+No!@4fqrC!$D z+*kd$h4;)k(^^rN3OtXetQ1rAPvXwfH#9t18;gzgu<*MLP6id>cTgqf_YwC~zeS&V zn;K_#TZ@aY(B(|+-oQ=&8}u5#B2}#CZz^x%gQ^90GD5;JTW8K|tt01@9}8_!gSe(1 zF3vV#P_lOAKCE!zJWZ@Pd#1td?46CNnftKVZ9j^KMWH|?5vTZ3@Lt~^H=HH-&jj37 zJy-6`Mvm*Y*Pc6SI1he)Q?Pip3pT&e#hrV*Fjbz3qv{0^;?5&wNmtJ7PjAk2W;f0* z&6s<=-<(@!;K<#3-Jc6JwdbrC$HBEz0n|kA=ii4_sMHvQDxDBqo+M_xJe5P@*_G2> zDr!EG4ouFzfTzZLux))X4jPZdt?uq<+|d;lj@_YEWQQjzy|8wp85(&TZlc$hNBA0TxCJS}{@!7FG_MVf!8O?A zn}p*bmMC-CE6!-;C|vv>-h90XrGQ@|XQsw=UZc*%`Lsgy^$_mk(qUYIod>sEA!a>J zkl@`7TkM+TfLM)a3>LL&tL%>m&TWC&t8=({@D_Z;p2$VaB|6KQa6XM0Xpg-Fhgpr7 zzWyTAvpCq7x}(&8JZ41r;PS7t)F{qZx+UVSNbK>h|LTJcPegq%WdPP@kH*n&8eCMR z0oN9$#r0@>j>(1L*kGNEpdAOW+d2V$6*`>GkWO5Nz7lse@j4P5%5cBB6zM4p&tA%L zOw==)k~7h`WjBn5#-P8&6zDgs!GSQ*`&@Mh$8TD2yIX8HgFI)hZyL|JUYU-9kYMa} zPKVxrbc`{mMR~(TtXy9UwyYdV_rHl*6SYuZav1-*?1#as9^zi&CLLc=KmjL%J@%kXg`E!HkI!c_yqsf_bmSTh_lxF-t{nT6>sTdSKPVqIs zL=igao%G7OnKG@+>wM4VJM3^`BHKLYm{8NM!Q$_$v)`rf1+8s&S-Iae7ILYFqTtC_ z7WZ=|8|M|t7EjKuEPGTaKj|084Q)9gyj2ZnGX4?UH+ZMwXKI2%N#?8Qr}>5TdSI%k z54Te&sc9dw38_^cq!cS;D$Lh<2gx}#Ytc%$(_O?kyu_;Yoarn72?><_a z8U6?r27g)4`W}AIEGN%mJFgnc7G5xxx%Ji(l&anc2dDLCIiFfw<5hmLu+k4~sQdsc zU)9XCuQ#!m^+t+=VS23d=rUNflQ=CbRyY z8O-$7HRhh>&$bPl!g}w%!rrXw%EW7nD*~t96O1nfE1I5LDx@K#@XF=+$eV(^w~Bag zv)VMpzgIgHTE7An=f4Lkr1s+!8Lxa4TFgPwX`iy9i_$gL@81qySuK-Y&+nmVv+k#u zwYr2&ebmOf#dlUbe{qyev(Z&7ztKf8;`VmMev>cE_1zohm^e;RS=P>+{59Fq32TJ! zZD-gtZZ{wAQN*qU8L{f;!)0w*4bq&C$JvJLJ#4{J1I4sXYKmn4dX_nI3#;W;u=H8m zgk_?o!8f7W^;G)<{^W%L!q_ogrM2m9iU_9}ioj7CibW5?Sr3mug=<`*;>3Y_tZk{0 z&=D9Z{JNs9=wR1bV^$%%Cx5~2_pKIgWt$1tk7_Gq#Ttr}qr+JL0%!K}od)Z1D@vHZ zYYEfoKS1$h)p^$B>*cz`t&lD4r>oFDnkx%EcT%XH-qlFOeY|mm3+6C4OE-T=U#MT4b6$d~<)QtTj?>{4He{FV1ALdSzz5D}gOK9M5jWqu#Z1sST!tw(HSiF}5i+R?K#ZI2ASYGF^s5`Tr?OF!@@BGJn)314K-16zNE)Mz3 zu3oBGbLbEI)q4mF_$1!h`R!TC-KTMI3jzF(=}Zn22%>)F7jE%sByl_w}3 ztWb8Vif(7CFVwMqtqWEakUGegDyi6r7 zjra6SuG+llA+w%ez_>G8`6cK7NarOxNDKGbv%S4~vAXkJ6f&DSrv2Yd`Ri4gtluF` zmf$`~(fXKI^zIVLqMBV5`X4#P0M=J=MSqB*>2fX0HEd^Fy&ZY!n@YJ?&o$CByPXxQ z#=9vz!aWt4$A>5$w9Ha;TiH$FabpE@y19)_yxGhK25l31WK^?px1HJ8PaSOZkr>wb zin~0!?>p8sNKMhWrj)fb%~eERS)i!5^;V2^Q(}cr&dBrM8VN^MWl2{W`pNolf5&$X zN@A@8BBVJUgZXeX?L2<}Yc5SRHe}2M#VPV~D!8LrRus2ATAGJJyL2zUBBO8U{ zUn6+g(ZfPni>q`+R-@2&w+6p>T_OL*{SK=hqNxb-vr~BIDA-@sbarrQ6FWWll>ElI zalG@7?R@Q^B&P9IiN$vg5pVI?C{(m>X6&~%yU`_?S)9Mka-()L?a7|9PQy)vzmxTt z<<&33>M;|f?mrCVv-^xuM9hd){PfzckX$zsMvbvuf+a9zGM8$JY|q3x5X4uH0fW|8G{Z2Q!rvaoe`C$}887BxqMb)j=dvU;Gp@e&yhswmPVhd32jH&_RYIk z%{^r{x?z{1Ha1SNXY3NivE5$6sjJ(WWJ`s-P%TmD>J-M(+M5MV=c6#v=Ba$5{#xlZ z`+VW|n|DIW;1*%%>vCaSes^Jpf0)o)KZE_9o5XZp8nLzw7lnzHA%Y(>6YQIoRW-ks zvDq3NGg!ZxDfKa9BXmQB$^f2yd=ku0Uv0-6RYO>0wv_k2!3jSC%mmH8o!I`R0fJ-B zA2w^JlETuzn$6kOB=;ITSm@}!PWUiVyuUu*MCg^3Rq1xFj5qIhg8%y2pZDo$%eMGd zvdV3RY~X-Y_R*o=idM}Xq>o=}Bm$`Zw!?WMx1 zRrS&fdYLTh)&|9*C}+ivc5B7+Idv>B&6%|vDHB4bYRaO|UaxffHmhoP&T4k^x1Qqk zunX+Ao{1vMXQ-m~`b5Qg-z~!XwY75Ae}dqksw$(~{ar7aba%a2_=GJ_GE?32k*78X`5c&d$JgB81&@*7PyzI%mWcQ#Y5bn>^b_mZ&? zQ4%F}uZxg%9C#$#>Td3qrgMhf_D*K02M_S+Zl(O{f@(hO++f#_nr=e%Be2t3x(j-K zsOZG2CGz(RLyOfFYFl+AS4})V;!2aWX2=T$qJThy0v=bFo*H= ztU&27Ti&yj54bs$4SafwS&S%Wck5hO+}awc+QjYhO^@^#7nsaH`=qCMw3laTX5XtG zhwHL&`MqRH1MbNi&69+?y%X51UX$gfJ-_g=2Ia1Ut39gT$PUWRtCq;J76r-VUq1_n zHb<}vmzso`jc@pyOJZbsOTFb1V-2DFc@y87;4GxJ_hSDJB(o6R4&m4Jll+^qQhw^% zNd8CfEz$>_#|TNo9EDNcQu%kWwtQmsXQ4}YJgYrZ#Xj1t;O|xzu~6^n!pl@Q*=%Dc z;l`qg%rEUCvwJ&|U7HilY|fRj@KwL%8*-x9pT7#hEg*s4P`lmLKBJE`DLdD7$dO^Z z@`-1>&B!(Kv*r~-Q+RKg(z2gb>6484d3>J0+lEUU!*}syVK1BQTP_T`tl%wQewMfY zs${_>%8HDy-I$lrD_+=lT~>4Qj?Bk0NeF+@hb>)~B#h)A$d~AyWS@)cg}2{bSjWH- ztU32A8|5>a-P>rOn4P4fIM}g-ty%P(y)G2Ln}3~HTW2rk5#7jMJ-aTyo-kC%tN6`c zO>TBo8y_Q^uDVe+-Kl}ijnZb#^DL#0Tki=Wb?aHu#hz@*lTqyYswAdXRKu)#zTyky zmt=D%EtmbbWUny0lObF5!k($@Y7oL2kIV8mGg*!QG1t#In}vBAg3!=>fT?#1W+T%) z*)GM}s)jLj!tSK+LWJc=*(U1(;lVp|HoM;vmT5nR*=zrEEx7eTD!eL@x?Jeb&u*K| zY&^q-`MURn^(Xf6BQFcWlGy|Jc=s3lGRYCCj-^zH^zf1O`|ypw`R)S0)ZIe1rK7iS zCNEb=cU;HL7CJM}W7DKuS$Ad>`%4~n>Oc8nwL(5)Yj4>lr!_LyF8;DtWgfDFZ=#v& z_9epCJ>mQoleNO;oK?cUl527$i(0`qu8`O7vy8Wp9pS%>@L)H5g5>if7xQj~`{eIc zA4+}S4v@b6!VAr7_6v8DQ-rLtM%mP@kuuGBF*2upFNMeoPd-HN7Vq_T7OVcGrnr?b zKrtoymUxDG@>f3I;^%B%FFU`^O!nxYMOEOHJ*-CgjIiy(1Xpf=F5fX@q4Zzq2dPr> zZ=uuXl|pcK4_*t|{G(m{W%snxWNEGG?8#AAMNBsn#X_%D{CH3M${TpbD~b1xPA~Ui zx##Y)Lzl113?>`9&A1yWEBfmq+@*~|`uK498Ph$oX!~-Ry1KdBHL*9HXn2_gZ~n-7 zn?q)!AI%3WP?hZ-TOxnF(ntPLwu%4VEl|F3c9>w=ZMblquFBR;za!(Posx~e86e|M z>A8KH_g3aHZ>TKJR##TTujJcny76^6nN>^2o3nSl!uYRCAIMy^d&x#@*B0h|GGV>H zGxoiB0=s6NFW;6%?99(lW|kZzI1JNes+*=)eyG*u>m4Tun%h$V8~=`4%N$gjWoYxUds?MnlN8C^>StFvna?-j*DyPKs7Gdlwb$R&W z`7GRW1KT$JMO9?~LDC2Lk{)P@#TF_I163}3}sDcrt?Dtm$LCPYq>O&z6>`e=3)N?a$&dFLwmIOWq-D+e_2H9cyiT?=Cb12-u< zRqk2!eduo4YegC#tNBrwc4IK}-PFvo#U5qEvjk@08_PN`lvO_Xahxd6&Z5y!WrsvUl(L^7Dqy6?#WXndDSFZ!i7D&sF5IuAOEIsp}%xr$x=e z{t@f=t}fAh`P2d-{7M$Tu>W+~7R&8y#ELfQ=FUd!?^QEl_*oOCo#Di~=P0pFUv~@5 zmfhtVXU|J5txNn_a$umBn6=M|m3XOSRY)gfuE zTrFk+cZ_ert{}Oq-+__Bw+DY*|J|Kf^)fq{_1|W}M$b&;qrFe^U%htm7tU4iS@RC? zGq*B+zR7w1Zd9S*R_@6y6Iz9hM|%n7tt+K6x3((tk5XAJ<78*M^^uJjY0j>iXfghm zwy>Z_2CvmelRcPg%6uX+*_#CyS2VM`}k~_kR+eZ=F5h# zT3^QAZk;Z;`+8OOG}~VF$fTso!S0gOe@}_9N63^{^;^UOw6$6H*9}s-<|_oxjN^|l zv5;MshcZpSXTs^SK}=tR!O8Lzrq zRkmbjR@G;t3c)A82o3Gcq!UN7C{+^H51JxW{!w8M2G3#E%~`_A6@R6Z7aH&jHSJl@gSo8# znLMU4B$}zt{2@r5Dzk}~%lN9jqvV+yrScK#c5Ll7HMTCGLa3WsCyX{I;`>kZ)AYJ3!AWg9gW1*{6 zSeSj4(B}78SoU3?*?AJbva>du_`R#Na3h2Ut7H7pp+T}ec^tnb{{-iK9j*&(!>fiL zv*zcS8Ohq`sme;H)eBFSrOTCv*~^~S6$vW6*9h51i(F;DodB#@x%%-9Ie?>|B;Pp1VidDYg_f8^TU6de<+i^nB zvRN(tF8#<~=xNCe=c=%kesiVQ5_`(l%#D(+tNS3&He4+jck>j_sXxMiD@DRjVXW{h zW(d1K)|+jM(qu<}dkVKNEf(6AjO6>{jBsxAOJU~1Xco3gT^{>QDm@y1L7sA9l6?2C zThjl-(0RvG{e^M75F&er$Osut-0yiV6@^ktk`|hzq=l$tW@jt1H)SNn{hsHNq(Lds z5^e1?P`{$_yZ_z)?)`pW=RW70&*%Ml9*vo=m}{MXXu-ZZ>|N-Z!{x6vW$>Z(nklMYGXwz z;COd6o!aq{&D+e<#gdQdQSMB#;(;`_{>~QEmQuof*4LE0)&m1iD3b5BzR;j2!{#P* zvEto<^!b%e=3a6mQ!nC0lS5+Zv0g{!;mIq^wnxcK_6Y+L^1Fi-*YalOve~RvK_zqg zU?#Js?<}eO5kq<&grn>8wJ87S6m?vkMoJATi1Y!$S}IjSR+K@RUHe$u04{Bu@qzB# z9*9Rit)o?vp5)N^tL!ClJMw(VLvsDuB_>tnAx&vE$5)ruk!r(nUZ$NosZ!QKb;(Ki zc0nfYp5BA^%(l{zYa=w(Ad}4ad5nDS7CfI%67K0wrth77U}V-8-ov~xD*3^X*`Aos zEPHsLGjp{Tc=8FH0VkGHSXQ^;6TaDXfec8SJH%!@`&`tCVun>e!MkD$pCnh57kA33Fo+wkF=k)~ua$ zY_U4VOApaOlSnFg4 zL9bcn(D|}=NR+)9EUKA`Lsbvi{jN)}D@BaI`oT9;+$PDdEmz`8h~{FvVI*eP7U3Sf zNPOWjN)`R>aqH_gT=dnNZ@ua@yX^COx>I&K&KpgoJ1R=(1KlYYRV9Txeb&K2qTHOGlIn*62OPAK$C1|@e}VAPIC{>FebzO_g_|JBeERKBcDwSVlu?&@^hc^v0_(Buc@a=~k`BRiGV$P$VoCmJ2*^AY& zXujE%dbd~5I!P_|!!mt~qeu0KmrEkDj_T+dXN=D-Ev0_)Tw1*QBAb10HqJ^9rz-PG zX4fanA@6wf)B z0o73@>3$8YQ|M>B_QVn^OifvrZ4+m0`QcTV_FTuzEeI`SLv%pS_ zCC5L0z+ks@Jdu%r{Vxu~=0FFs(V`9Rr6b$B!UHC_M}xL~H$6pk=;f|?M8~0;9#3ha zOClnnrY@diChkG5jD+Fk$4eRB)Er*H#htilw1BvA!{KkFH7IWj$6enIap3VHRDSgu z-&U=lf8M&`{c{h=lhbY-5mzby}2CzzEgTw`MaADUenIJ-=v3an`WcFzAY9M$Kr); zJiMo4i$(1MsPpbVocb&heJ`$~ONUpIz2Cl*VE-aw(|(D>HC$t7tWUwPAUV9;nVcLe!CbxTxU^vamt8l>x^eW#MKi9o&6Tist?k$AG3% zGIe-_7rj-NUGVTMW36F~mxJ}t@uV2vWMB%vxksB%FW#gFYuc$a+`##bWlXr+659Ie zg2kkzd929sT#QM1fxES&V71(8!18O%dAn4$@t!T)Bc%`7SGQVdt*{{bT=&qcGsWnd zR8_RN5>I*NuX);b$xK+=1|pL(iI~os4bSbjgL~3m{3~&TL>FXH=f77mAl3zkw0xMg z-L=d}ix20q;z{zR`U<}7n?+}htR=d~CgR>RpYX`e0}^5FZHEs=T0{@vp_>h2AsB6V8BquWI=3=z6e?(%>(SH0GOKG~oaK9z=z= z$ziDDR$|t6h~7Ia4U$V&z+HX{TWi)v{f>L0)|xr^?5hJ>KV6P5-jkJ`t4anVCN^tZAinNfa3 zcWMu<{l1k>x0nKAu1PfElqeJ%SpvJ|75MmDg6zkAjKU2AycrmaTbD#*XkrdZ9#_Hq z--XOptt@!HgM*hVKA>=~1y0<)2ns*2bZ^rnxU{kmmn~>S?Y=1Fd`rQ2fnR(&+YgUv ze}s(xD(P6^18mbW!dpCE;Nqan0$Vl$Iel7RG{HUIX` z&p6+tp0^=v9$m}Lp>{ilXwNKpFswTbKA{s}-i>6s$z?5(c1*y`EekQOD;5K4+t~HD z73lY}bi(+2L6c1%Fr+dUTRCq*=&%|{OkYIejd;XK^C)iIm(0txZ^gv9ak%E@P7JTm zfjRF^)5jfmnGb93V$u35D75oF`dB)#A>QZcsGK(%O3tB4_8v5CvJi1xzLGAp4JFe) zWWi#qL9D$siVYL5;$E*sxMvX;ON~uvT(1iN~&m7&kY*?C4{!`h{QuZ z;{2+sUAWO;4qkmLk7K)CsT=1dJ*GJdr@ntBwr}M5izNSJcNL#!?N=63)xNvX;}in> zgLA-pi7XnuaiL$LqVRD?Cw|KLf*-18@axNr=)DI>|1|8PykIMcO*s$0%qlV6K@x-( zhC!aoHhSyKQ8H+)1ZVYRX^rJFd^IG62EOC;&43<>&{;-nt|XA$1bNVsN#JP2+mMO8 zRP?cVO!MTX(S+&FG_N#ApaZLN0?l5rs;gGf%*Q)nkApn2@yVQWJtgwwV;XrkLmiG! z{Z0Np$fhHwREW%_iI{gojdET;rknotQ^nSas8oLjRfz@vSVA-Ee6YmWF+!*w| zUy_Nc#`NwoB@D63rl;$BC{yMRS8rFtmy$;$Kgo}J(*%5eYk*qn%5Y!1q_bNMUC7`U zacGgh#W^(|j(2Gk`fKmU;gUlrBzB1Y=G`VYJuB#hQ_7GYp#UDm%b@$vC+ZL%O{c3U zp~jLVTyub6YgsrxDx-AYSvjZ*RYZ$x-%+Y(0_AF4pieTCP%kA0)fJWbZmShYn(j;D z9l}*-i;m%g4aku?WZ!h|p&A)36X$Gs6H1TV^&_IPg~Xm2Akzby z$gIFIp0t4#ed0Qwmz!OLVQnpVK+u%&Mg2c&b!P!u_;)d4C3?g-tB@S_T+S%xOyhh$ zzZpNSJ;mOc*~eUUkzyiGY$TG#F65nF9cQ|qHp_hPWj&wNFj7_5EaRHg=n&;|A~Ym8 z*_OH>?F?{FK9$6-{liX6W610^C+V}tZS<{l27AVu!jma6ImxT0~cAHq&Cm@vnA`sh;Xnnb182j zJAdy58n?ZH%J?QSOZ^U6Udgb&sQy)xtvE1^o=%HIyH6ZcE6=CbxD}iXiLFhe9G0>A zw;UJG+JPZrvYhJZvpknsg6=;7$8a>$9e42`)9FWF)AsWzjQ+qB0cT-=HBjotC52`WnuesvMr&!gn-6tDmH+oTay)oyJ!Q_i6mF7rs)Q z$Q3#-&JF!Kh5IRaAI^VjO%8F&1!`0VdFIhY3=BO;)D=6cnGUgG>U9X8BnP687f5db%C?FHNE!_-He$JXz?%ll(qW;y)e zB2{dV#2Z^R`SAr+_%t&Kl{<7H|IHIJcFqvYSFxn^^ieXbkw|()AF=mC1~^;psiM}L zn@z8zPm?3-)#;iw@_6E6JbhXEm`=$%No}=cfYlb@8h%&7aZ(uLR@%*!+|FVjygSWC zT4&MvCV77E`>3X^i*tBaw*n~>sbo%Bd?8(*g^8tA3|`*13svPDaBz}1a(<;z-7RuV z>W7JF{`VbK@at!H^^cJx2TM43W0d+j?ZDu)%jmNphZe2x!9`nsVD$MYOb-4^{(O;w z!~GXH9_CB%9=%Fs7OLUy33-Ix4WO!J2KaURYs<+xvssNr-`T}6ALxdEXLu^F<52#o zKaneGCu+TN&=4oZH^04xfBtAMUh-TIX;&oS>UIl=*ZYfeQhwp$;>mnz#|dz8LKU<- zt$;UvL-fFe4p#Z_etIp`8|=?0P#^0uGF7hzS58*uXPl7c`;Y8|EC``iF~4a4h6GmH z_#jXHXun|2y)c^fUck6-9De?)WsFAQDQ5oa4fNLxIp|;43jK#-L9e`>>~>1xj9z#` zyUb;poQ^DH|NGj&B!2LKc*hOw{(GC*i5cOntAi8S4Z5_*CkzZMq)4n^0VTH!A*0?I zq`N~Prsg#iS7>vakB@)-;->%2GZ;_O1b-lmn&l3*NG z(YU|r4WoI{2j36LVfl$YoUX7%)YZNsc%X>93eu#_M@7jK z>9eGJl{*>MIYZ9Ho*`TB$H3iHFF`{s1Aa(EGuy8pWlq@y(Ov)T!o4#Ee*Toz;9F@% z?N9BY4@|e9ZMZIcZfSyNHhO5#eVv?5zf1bP3Sn<)H;kBn2iqE1?lTKfTqB~7PqU2;(275CfYfVDe%JglgZwzC=FQ==nhCzHP9t zM+wrOUS)q?E5##=)i7?oCk-r1Wm~70QTu?0kTAmGrkgG1DykmB)7tV_r6Gn6mLJ$4 z*DH+Z5FcXexhOYn9sZc^5fbsazjys%-sBfEiz+>tyT^l^&O_` zPuyk44@j_&FKexcOoUO<~ZH&9>j&_ySDAhbfFI*Ofg&{9^%IligSd(i|C^ib> zk$1poYCFl)SHwr!X-L=qWu9_NsNYB`@8Mbr?u;iBxaCuF1jGK`kaPOGKyt=KvTv;u zQ4m$3h)Y;bmMgRD@+UxtMsdf>w;q!*KT) zCeTWbxTi$0U#V>)3}lT zJZx8MpobD((%YBANORm<8u29=-7j9|e7M>`4*p0ZM}@p_)UJdzJdkdgwl|yJoV5$< zCUsHy#6vh(z%Vxt#ZhySsrb=rA(=LkMb6%sL*`rOvK!i-uy&^@cx5ev`se9zEci0< z{jdQlm%Eb286}t~;{ujbM_^jlT39~C1m38op!8fT@*%X3!I~CUp=LFw#l--5;g6~G zpIma{rY;#ed4Xi>=FpmLqR85?keq;n>xNM%&8mkWU$KdVXcnWDMF&FVkgUt+bI`c|8xs9ZSi)JS*s1 z+)8KsAYk1|=>esQ^toXHO->1Cgnt>}qmFUj-fa=A|Jhw|-~A80^vn&%Uln5YyFJwX zO((UAZ)#eQ7ELGj46?^n6VP>E0eL^$iKhD*V`hp7guZ{t(0Nz!^tu3co6>JO_iioc z+U5dQThWN4^hXQLbxdi<^#q)kah~d{l`)>sONZ11y73YX{<9OwxQzHhcaeLOtJiMy z&@x7@t}G+yG(^74wxkD___DWFZ6XS}i<(k$`)RD%efsW~z?bMdi>@RUD@?M;G1C?r z9DjhS4Du;^?-FZt_a>d4k%ZIsEhbKhKk=OXRV;hM;UD2l#4S^8NS)CJ=6mxz8gQ_f zntjf~Su+&*?usc)nb%6Z;4MMZ4sL_c@M}RhWHlj-Yl& z33$E)M$TFCXPSIPlcN!IVUii%=^J9_MhwuF_swjbvk-9;=ml|+4dkQNL6R8yiopq~ zg75l7OY>7m(ongD*HgmQHRz+y$|ARdD~QG~+j@sPmuh%fp!hCgrOGpH_*9 z;q!A@q;khF+i9D?t~94i-c=V2DKO@r?wCeJN-UB6_8iS?-myap?|EigDv)=lj!rUr z&iZYP$Gx#~?Bv;hS<8cNa8acUjc3o}zsNeq=b$0FDHY+oIz4#5E1PM_n!pCXsHOq` zP3CJ$xM6IOKV>u&`JXkF`R=E8;6D$6X0NrBDRC1+xAJ&A@X3!K{4J9|$H|vYiUk@k zYDSh#ZKho+8SD|`LRzzs!=E$Dl+PAR^8LpSGsWBY!|hc+iGJl{s_6U|XSVqACk8yg zDdDvU&;L^CLxp&5K4S8`3sg~L5?OrfA$gd;8@y^tFnFFVU$*ElKYQys^5u{$>I4O& z%CSgRYfb@sD`YD9URDh5r}9C5t|7beydhLt>%+~{(r}$OhkZXcfm84QgFXEIG4DqE zb)x<}8m^0ulY_skp``r=)Gr+*(fJyXDO*izU#-T%Z{x&m(HU?X4235~_9W6`KWRPZ zN7`={GxCi-Ou=7cZq+~oboJztm19D%B>W-~mHOSdZ4(#9M9WBFfR%u&dBjFD9h}l| z(Z;dzZKNhWg9-hiM*a>I!-hu+V4RzZ*Yw6Yg(p(T@dNX~|8F2DTj-HL4@KFf3+ySa zl!klXU5RDhd8jd7MYpIc)7uw%n1?2(S+|FFR9CPj>?iiqKkFS(`c4d8Vbn@a*~i1- zOUYnk-NUv>6wnkYF-$j6V_TO!pi@8HhV66Xxxz;7+$-xGVVQ*mIU%72>$R_e*b7bY z`na0>2s9wR6Wz&6SAVi^#Rh{iB=6OGm!h}W9mUHjLWBLoS{*RPu3sD(!yc-`SvxMtK3U-CN5;FJQdjp&N#po$EkT)fl+N{Z?lKn5p>)X~a_DGm)*=K}* zQ%h*bEoB@}`$^*jUaX7fBLV9p$MJH%NWT4gL#(EKCjO@P$>js#bRa|rIWw=}%{M~) z$b>Qa%(9L?{aQ`a(|tLzkF{x{{%iJKb0v*7c*6+$?4#z}pWx8FO#JXb6K~niCzJTK zoTrC_X!f@Z?5#3J<=InF=LE6?k-emxw-SHNI)+BudARL$GP=4=$863V)IeEWq|!rY zoEPqyp~Ue0)!m6Ll>xvX)sGkZtm9NX3?(G=o7M47D- zgx`LfT0QV$XS94Jw;DNUncvQ)G~Z^+o7R$uyE#O$bP+p0fKr7B%2;K~vW*2^%!Q^1 z`tR&>dboQz7VrE*4{pVY-@Ms$@T&=?Yu#gf$_kgUPJH)n`^YYfrBc1gx8PgwH zMC3^O*8_I^~$n!#Tmr z;aP$LDUtul+zQ%3zcfArwH7IE$yGCMSzrN-R8NCYfzSJF{vN1TDJ93a%W$aA2-c}T zYg&_Ec+tOChLmej&N2TZT^aoU6**;2Myi|gLY%jsN{xO?+WG|C8 z|03J?=^Q!GFcSo`3&@g3b0KL{J{&%rOLjVY(f_vClhkre__)gv-aed7Z=d|g&?o!Z zo=QJ*j?ygcBlkX z$$T+EC)^)CPQ;_KAEa%4~e@rfCd?bn@@zfrz-YAzuJ2; z+%%2+(EiqFdA5NSu3JI<93!aRtV$@YozA_xRE7KHeJOiJa~k_|`c_)}pD?u4mB94J z-mv)XEI8GoyTvK)8{#3N$zP0Y= zwH#?;-+ecvPqoyTb?jtF_uUMYw-r_w~jIwkG&eog9f}1Q3&~luA z)fWe)x8c7tVo;dpMz*f%BKJ$Zf&1PaPHFxn!@}8&i1`6#&FcregP$kE-zi6l!6Pwv zwAquaoUMYF?lzDeQxA}!O^Lk3JEJ7y$2dkMO%-G;GI7(6g92t`KP#tYLbawE@@?%} zae0p{HEnihmb+>*+R}H3)cF_Wp?EXX_tKcv{4>aO-_S#08%GpOl7neWH<6=Z$xtXD zh;1B`_*1Ob^0)3chww+Mpu*1%KEArmXz<3#zjiBD=dV4TeE&HaFI)+-ro~LlgAj7o z<1zT&i-WA6xkRK+1b?m))ODZhbX1VnaSm_=uMIWOHs>YGUOfSnMpWVRC2bg$NWuNq z0)%3)G6Lz_*Nf|e;gs-1=$UiHbeZhq=(FCn+y5VSHk13 zC3Glf21@OV#r_d5DjF3-cfJW=CQ0a{hPg5dW}ec#`ym zV_=#*PUBYc^rCk$=Cuk$)GdMjwejbO_vNrczEZr?H**>*ddDm;9A3d_{9MF7$BESG zXbMZPhHRbv`0t*1#(O+VfILj$J<8IY%AOK_u15cbE;z#c_?T#+h{D@9GUQv?J)@QFrqCkQBBz z%Ya|{A<$BC!3&xjP@Q*|UY41R$9HF;&eDAt_S6ogbX8FH!z5}mbs=oLlnJvp#IOhD zC1~UIV{GI)3mOs>MERCg_+iI86z!OZ7uTfXx?3j5bDD`FU6k6hm(if^Jvp?XpPBM+ zGYDDLu+LS(>5fJdfmh&TdBoa_r0?WF{-X}g7PXt?H19F9P`HxuoovqTip#e2&2pi6 z5gurGybQx~CsX@bJIT3!wWuQ3PR6|U(yEVMq{!kpx%lWCG5$Ci#MBOP1i4?jJuCqf zw#L$-3Q6*jUqOtf1veF(NF#|xF);NOheoy8;xqO1~^N8WR=i z-$%pLbh|3BBH4sz*-i_BjB%onGjCq`cBW#>XXa#!5WP9Ih&hoSM{3skkSl%_#M0B1 z%v1dXe_q!^R}4eUTM8LXhkDp!@RQKmm9$II4i**-kr{LHiT{L1HZ-!CI2k{pn}Wiy zddq+KFZBe8Uvh8uWLK0-HjQ9?0QE!tMb{+CDGI&S_Cr2Bw@!* zDJZ)Fke`_Y3KuAR&GLXb2P?=3i6iTMgovuJldK*bx-pwL1`t|hHgl5(%z!bZGrZAD6S7^$omreFwJrE{)k(R28N|kdCD|!);I%r1 zxMwFrn~fVd@e8R~k0I9F+D9Mt>hfnrsq?Ltl?(E={xC1IfK|L#MwSH6f#F47^!n$w zcx1*d;$^OnM&WvF`R^t0{(Uq2UBQCrspYieTQZTHN?`iy0y@`yKQ58)Mr9HQ!o?dQ zaLXUMd1D$$=n;d_6_=={=W2+VUP;0mU2w~S+suVU*6`EhE45Hx4y*GBqgWM#E5z2p zwdYr0+NN;0zCz%K2DrnL_43@3;i(W<;Rcd(ddLg%oP5=?1*i4TsnC1_n9`n$73qEG znzWt!;^%$Xw(bFpI{YA4e#BF-l)^ZJ_pnp37UnF=CAT@XjN{A$g8f*R?;s^hW?WW6 z-tugksFzQ)_kQKMi#MWUeIJ-Y6>Q3UO~n6XlIcOBDC+u>>TDQb<5a`&PrEOggyw_3 z7l+G=3d32?dZsx?l~;b{9bUTPil(lU_}%B!`CG-qNYeW>3=9!uls62Jf0+;HG+zh! zx_U8{`P@nE-#*0!etKwbw*{X|iekOWWQhM{LVoxCqd{IiG!;^CrQ2_gpT!y+ z_aWiIYV(=5ZQ}C!fNy+y++rxh~t)qLA<|t4*$oQC49BP6IAg+ zFZC+$qBXOxu+H5oM0wR65)`o>znd+k{nrzjo_AbOS#1M1oa*R$Yccw1>28dbzJshs zFTGS%!q)tep#C;C^hE0eMrJn5 z-~NgonLiW93ol?w@=J_dFogYnFv+ffxpr&nDZx$8W~iBeR3I*mRd%PBDT@h z;rD1s$r$rh(Drf9ib=XTjJw4Vsy9C2U1;VLro1J53X;1uO!m(x~7?=8F;^k^Z#&b{uyXj<{I9CTfMKkbRt2N$?t)MSs zJn*7I1olX2lXV+pK&S_^Y~PXT;ifHzWm z(SM;Vtl0dStaglN77ETA)hB1utvk+>-m{0X>(MDp*|G{(R;yvBpJUUA#(t9Rz6GN_ ztx*4A6huB(!7k5@G`d9aANM&<`cg#bbwPa>xg;bw&)4IO6%Pd21z{9kewtpIUBOt^ zX`oS-2*f>{4H3r@(YNg~%~chJkrp}hdGiWkP?$eBVHTcSm5H4xi|Bdy!3LTJQKwZ+ zbY=Z@IyAfxs*09@`K)5PWOo-;TJfINJYCD|I5vT5^u{#Bc;=8gDHSTREgE~j%CXhb zQOp6+ZA^=g6FwRAW=4#2$;hvM61a6KaK0a)vPUe*aJm^CzE?`GTn%SUE}o;$i_cKQ z|CWPlcsWmUz5T_jm*mI{acQbrmCj@prxLLbSIPL@32=Dv0CAoPmYHwgGvUFa(B9ZZ zqz6aXWj_O$@jM&mQvX_3?Lq{{a1D^wnjf5|9h=CDx@so0RUBtq##uIqmk|R|hMW{j zz6&TyOb>?RJ12ncOpHbZ(fDfNOXKodhu}QtX%vBv{`sAcG>eU~k zhkob4M|KP3bQ;6AY3lgba+Hzn?z5Clm4IJIMYwVc{9sAFJY5Bc=%anFNUmEB&f@N) z5B@%9hHu-zEs?*BdTTyJZQl!;%3&bU*+4pPo~Ozk$LJ4VVYqyzo2beoQMNOtaRN>L zlEf6|jplA5b+*v5{jM+acvDbby_sf2Mv}cU(Ihsgm|l=F7bhaR*^%CKucta5#um8zrtVBJypYWRf;J-h%KTOn-Zy8xWyG zYPi9nfWEO-$HToxK|wZ_)`#6DUEM9TC#cXudAlXOb6KD1ymOFw@03Uaqzp+*i7qKt z(xnSljZn}05S*angyy4RSSJ0C_KUbHX zKHVo#*9f~}ftAI~l`8DYuQj|wF>l!rsRy+C#20$`v@w07;6R_X7?33CKGt!D3A_87 z2IlG;vyrFFsdq*mof5khj6$c9!}`BylU6I^)wYi9nSX|$zZ2wFNP@G)EfT0?Nd5;XdjbkF0H51FmJrKJtLOIu{9Fm0!M1255I&%-b#$)0EFzmrIK^^$Xc z8=KY+Ev7fcTiEL0ljKQYCowU($gUF9e>v*y$LNb?3 zc4qJe-3K#gOamOeOJ-!NvVOv$G+bAoHoRR0QL$z);4u%5*O>5j7Cm6vo^B*RB@aN@ zR7YaMH-omc2%I)`8QMrmAbV*&v&X%Y{=3~kGNyKt#}5)x`Aj5dCMMi$QyhxaarrLFepjDExk(8i>86 zMQ1um+2<87*YG3+^#wv>Kpt}@y@!xDyLmCw>LBq^EWCGFNG5&p!3|YkNbt{t576n1LZ>bq2cbGOhf6pv>+E*cwT^|K3R|)eGL{}Fyy~_?JVH94^XErMRu9m^Cn%F4~(rKJF}*0E4dn%1c$%Z z;RO2-+(8m>!{KZ=^6@OmkC0@4yxxlQ4d=tRrv!!tdF=q{H=IuYVKQ79M4Kj`A=ln; znMc$3w5c?fF1aaBe-jhGkFNq6d0G8Kt z==lgObXyY1`-&OxQ>GZU?3zmG!>7DW!yAb6$|H2u-_7i{oCe;1>)+8k{2{ir&V$;E zMuF;zt8k{b0djsT^KHbOP{A~tM0gpYL9PjoRykvj*n4)<{AbLm&T5k3)rqWvJRUtb zz!sMGLTB9>IDfv7gk^rCUo&jjlc(C)Ro31#(N=}|5jKTM9Q7uj3%Ck2982?6orG2FPKiy(KtM8#!Tz#^&>{bB3xd~a+v)@Fw6j)Rr?CMfWH z#EiOYz@ig($Pp9kMr9!xy8pH(nQTJIQN{=t4G=scB2B)B?EyW{J{VRx4VGDE)baaV zy78h0RNm3%?;SD3ImcAUljZt2JtBo{Ryat^J)baP&+m|f=c9>I>1#}S){JE;Q!(n* zblyWPHZqPbiI{?+y48YW}716)-;m{?}hY1M>ExWm_(gYJV|QU zTv!}kN$$Ph%vdc+Xp(Vnfm>P8ki>Yizlwii7#5&Xr6Tsn#PB4q>GI-EpR$~v?#9;j zUSm!=Oo4`eZAO2$3Nv|VJJr~j%zt2^#J@fijV}MRFso`Ro*Y<2E5;@0zjuicP<$KS z$^rkF0f*n%I}bB=T9Ul)O0?(HR#N}EjXcknhyQlyfT0zaJw2(4j`jq^$P zZ&&EJEJH(Yo07v#@vzur4t_6ELT~moTQO2d#`%-D%)gBwck~gVU&^D+g|K(1g?t{0PY%?85 zmd#@A+b6QhW1;K>i;r|xoH!KKS(7K0T*AbQ(u-naY;yigazNOFvW+IBy1I{}yomwL z=hL7q4K%Ct2Y~V=tRNT_Zajz(ZtSgx=bXu3&X%=9rzW)C89@Exanyt zxOYeVyNf()epB%qZCfkHQw7o(GBVRzxmuoQaBetn;OC*TN^Fcj1 zma?LI!2NzEJlD*jFU*_B^==`k7BIsPixLFyQl48p!xBC}&4-cLMG&#BifliB4@?Kb zz<*92XV>Y=sGc(c2nO1sZta!!2Z7)h95;Tlmiv@yfAf*xjmv9G8~#0>Z$S_-PiCvv@w zc5}-FJxq4lXmT^dM7a7(UJ?i2YF7HbH!Wxr^vg(Z5-^H|aC3nnazf72xY|fIAjJ=c z7{PO{olms0-;rD8IYfT$Bj#U*9P4N+;9+*0BKqB-n6@B|K6=)QBiobNw$yHta{mI+ zb?qWY4mC1dr}^-HM+tlLT@0afR0Mb9OxDvio4kJ;4YxEeqp(LjR&KVyQKzNEEOf2q z%=4NU_x%qOsHsD)%4op8`X5h)+pLk=v``D*N%jxXI19Y`Vm1X6+7$SPd z62DJjD9O4(Wjr6?<~OQ*FD(&%)q^~ox5>(wH%F#$uzr33}iOmCD9vZ(cad2&YKCQjM((af?8rthb|PsTDlb+axXxOtRuWT zR?3r?j3D#IHQ>dBMa1=V1>AKkgICK^q2p}>tP!dPv(U+K?SL2Y8mncJQ-eXQy9-u% z<$<@nG)6|8qB{3)FgjVTM54TnJf8BBloeLPh0Qha!*l^_^r3^U6@5Y-JVM!3+b6)g zq+{e>sXAjcCc<>N<}+KiPhp0W+Bs4)UD+$9vss0)Z1}lGgG_pnPxt;e6E@vVh0OkD z==`>eQApy`m)Ez@g=1mRyP^yN&6=T0&@W~9HU}U76XYBxd%@=)NyK$mJ)1bsnb>`M z!-}NHGIl@wSpPGYQQeh zO{II^-sW8R9KjG3Ewb*aEDTw7f?35FsMJ5@y{Y&FjYF^DJmuqVt^(d`0pJ+gF|k_& z+M>=mvh>$xX7G2cWVd}1_>$lB;8+2@bRoU_}RRTI6j+5^tn0kX-g}-I(!hg17@IMR6_2|_JG@AlZo-( z4n}dX3%tJ#!%B+}kgDMU7yV=4d|n)giONFCSt*bdUj`XxEIB*W7xQ*4(x89)1!vM= zf$y81Pr7%%qscGMlR8r&vg$@IIis`{vKDAV;z$RW8yAonFL#of7CSQc_y%GWq6Gf3 zlSobDU2=U>C5f{3<~A=Cc<=Ho+)>Nqm7B-0JD<&>0$2vN`Xm4+Bm_U+I>mmj&0*A5 zXEx0qbRs_$^&oOvI`}w;lCji_%+hr}*s&puF;vO3v?($q%iYspfe8;jD!hUi%g4m_ z;&bx+QxwfuoJDO?Ke7&GIn1GR671K33^L8!iwutlF;W6A)zOK=ZfvY0F8U|P>6XDJ z)v6dUOe}>b2bH;J&*nqKKoESGk^&Bf$~4rWT#&_;Azxl@U^aBGWF!V3SUzo6B|U4d zGqGY3Z0={4n(^J4dB-kret8P+G43)%`-myWxxRz;yn99yqTZ0T{MF!N?F}wd^dW7x zF86%NOGv%`1iJ27kw3tqE@VE+@D(qxpE@jqv08cyXG?qM^bjF}0E3?Uib^{iJkswhN7 z8Z;`6h9;TGP{>$9hJ^Z~QS5iEEtRoEQY17_ie@w@oqfKZk7r+(4?gb8Ywxw5=l8pB zu<6xiMMSmP>L@H=Gf=W@uO6yic%4&xvJw~lE8q^F5a+Ehn+g8LpTM%o5ZYGEp|VdR zc}slN(B#iRsl!xyZ|hIH6Yfjs51%q9Irac%4ahLRjg^4BctJz$lI%2tvuZp9Nk8S6 zS8(Vz!OY3P3GcQD0iSinxUD81zxH+|H{>UQa?`Sw??zl_;6<{0 z;0EcOqQU8Z2&Pi0-{>K6ZPH9ml4oIO$$GDLk-hv~D3>HKLXT4>Ws5fnM&8j%S|BK8xvQ>q(A74npjjLHPCUYkK* z*?Hqd?c1QN+8&P`SqZ_4g(N@WKQhTuhVLnJfEK*e=ZlM8AuSn0sG`CUELkC9aQGL} z{ozU9Ep3BqO0M*hk|*`KSdJ<`ZY2g=8pvGVGho-el##PHrsVRw=?1YR|KM{|(7BKU!Pc0UE*ioeFfi5nJ zromcE=%Y}h8jsDR$TFk~)USsU4Wl@kIctz|%Oyd+Ie<59>3s+*kRrdVoXL|}uJGeU zBj{|l1dYB#+!;C@3Fi%yx!f^2?wU*sRoy6Y3ZWelnMi0}9;Xhgh?BlPe12CBk(>J9 zzaCxK@?t016Dmh<`4^Jq8tdt*rWl&|k_V%fBHU^FXr}D8C^>iW8@ab?A9bG|4Rs5D zpt3Q{gqnX$wl&aMqty`Mtnqa zB@_56l`cpthgOm*mwMyLdm;Q+Y&amHVP1EMCh6HB#^;vF^GDnoX_=Q`@9$>8rJfMXiUKBM z|AJ^_$eg83`#zA=CB{^-y@}gXdW!7Pi=qoeta-}u8|g`%0h;-`9t8(9QG6?fw#~Aq z+jq{RwWHPav3M38OtnBgW$E}y_%QZY(t>8cALPGvCG_`7W5PXMN6aQ}Ay&I9$xc%r zYW*miUesGjbDe#uQj--GnkL3?ESbh{?j5GyxuwW`_hPdC{u**q>n(S%!5-$w-R4F# zj&qV-f5FeOlE@gGC$9w0@X}q1+>UM91d1C`;#5VbkFY_{>(j{d%q8@lt`bSBm4RI6 z0iyR)p8xIC48Eg*2hGwIupPGDA+x6n*ltB@p>t0#OlY^I|G8WxdFKGIIWs9SzCVCLvbm=pV0Iy6gHK zke7c<2ijF=ld>oGg>gpC;(GYa5C*lwRVh>kp$!slOQ5M<*|b-Ukle)}s(WJ(nH z_=ob=ZrDSOc1qv&979_T-9^L ze&s}(ksw1Gf~9LR0<(zRa%)JMpiE1b1(W=5`NYsEgZB5;P|clL)VfBN(zqZ*{0SLJAEVinkS$zZ9;e#E@NQE&(L<$0vE zFhzkmggH2Ys0y%85BBtsiUm?sK2nDqk2fMu=XnqZ;ZzXz;M)z)Y=hR03c}AO#9!k+ zF%2sx5iJ+UgnEh+B2R(rdwQBG^Mg;n=6YIh z5+kQSq*dz>Z4?_OBA^YtV|=3XD3-)DZsQasj5yQYWb$pI7JbtwU@qN`B|j|-fxCK( zi2is;Ftl*$$(}U3Po18vNud{uKXQHEdA?bQX|hlU zbF&YSu>TA+`5NGp3;hNCLLEB;r&G9f7mE)U31e}d0e&pJ3;)q6K(?E&q4_iSbJNOt z@W3e%HsV+&4v`mO@9e*Wy~bO)K#6JMN!y91A%}Qq{S_iTGDOX5ekP3ZgYlxcn_v6-DL)M0e*lqe2 zc;2@vd~R$bq`kMt;XNX_Ao3q$zUC&%Ue}3)#k!&6QwW>o7|njaAI6TKvZfictmsJd zak{8ChVC(_C9zK~aMGK_S&NKM=vZwQZ^j{MY%dpstE+TzMp6Qvd^HlU+M$z z5;bsQIs@G~qM3}p8}h=Ad`^XM_|Wyt|rwdfKy7OKE?lYYT08xiodlLi%))1duv zGTUL-i>G;%;;q+qp`hX>I4|hFRGmLh-baQJzpzx0K3u}tAJN5<^aoNq*o^N!mtyB% zjmBO_JHfKu$kwFu80gk3U{aTib(~H^ME*MbZ}=aMZkA-f|Bk|6Cn@37HS*ZRrxgF} z7iO=x`r>SK39fB)MjplInZGxLS@TaTv5nRXt|0aWqZwvKBE$uZkE0Q+@ed*P-#1MG zt7Ikq&&M6RPIAEJfibvzXfMpJZiE{n25belfgRqe&z5YDgEM6_V9+On{EAq{;hkf`qHtaJyZTeG`(3tG13Im&8s8 zJN5{k83$nZOHXmjKn@mtSx!!e^T5AcjL2Pe;B;MYqI}^YJMsR<+^0w}JQS~wBMZNS zr}skkL$Vl~;g~LiRLh^^ zmYT?sq0Ua6ZTbq^?^T8Ih6ltUWDC7_PmbT4;!a|VuKr{SnV55jR-a1TZ zI#)zry|Jdw3k;z<+l<`Vb&R@PI!OPVoXP*2&ft9mZ@GkFH|Vq-0>9u+s`EXHb_A%< zjwKD9CW8%F?dO;-F%%o6+@|2_|(-oT=thtXo`yXD;mFd0ww2 zI!%-LrC*N1$mu*zdf*MyS^faewfPVFXDMMl-*jl2c?{pK6vdOym)f1rXaM*jhD}~c zaD_o?*m-R_Qa`DVS|<47!`@=Fro592$WJCEpOl!zlLcAlbqL?n$iiBG`Pe05GQ4^9 zo2(tqp;qo+xh&auEc(TrE}sJgzgFO2SuZrq--&-8T8qbrwWfB^ga2)+W~7I2eB(Vvw{G)K&eFLXa14-MbL<_3oBvKPT9 z*2tXBxGN+0hO8z%fivl`u}r(OlU|ak&xhbkrUHM^+>4kVI?BD#|IGPCiBjPM3qeZv zW{vJWCDvbS6w6;SXYA)^zEP(7w@HlV;AV*e`R9$gGx146*k8~ zh3eQ#Lx$sCs#1|1>&ewmz1*UoVXR(P06SN5Gdps3BK6yx!l_A2gfugAn5D_rC0W^9<6gSFFd?NDNJ|qEq+=)wCH<4%l zLjQhIPBc%N_x9Zg9x+y7m78DS3z|jM4esT3F}5+BtI_S6ccZbmHS;Ofk10&1mZ92BzS#I(4}x zN>ijp$(-a~&e&OkW+7d2YDOxfp=!qT{fs6%7QaMK5~R3GOY-oyS^Zch=o7y5V>9up z*@8vPBeCkjY6y&dh3va5@UHqeFvyIzNwZ3TT27ht395pr4LF)%u15O@A4oZs&Z>a>t(v=sp_C>QSc$H8#@HhTP>67^VD zL)IRMzyYh~;@vYoFseHYaFLQJl`fdi|86RSGx`1yFdhs2Z%whWPAQBsoA4@L1~EMm zME#Rh_@jY+q?esSkIMH`i?%&5Us#mxx^jt(XMZ8fa|C_6L#FWLOA)qge+|yfsjz&7 zA{Kk-4=UE}WNfnp>l74?4a5S;ac^TTtuBkZVC#cyZ?A-gFKJXF_#x5HIZlp9o8x7x zX5oohsX$((;WgW5;@swTNRhd~#r)m^ALfaIUWNkq&UQJSQgy%4T^Izqnxcnqz-{zCP;PC>@+CPohx zu=pVpI#2XGmQH@lovyqKOTX~o@Yz@J%l-w-+&33O5?^w+Eu+Ytp=B^Op@8e@k03j0 zT{(?bDQKPgK)^+OMCMc!6ZMb~*r_rGaldsj5}5$oX5ZtMcFyOI6+h$DSLa{rl zE{w(NgHmzL&UyH`&jflWvlTK5B*5tW4jO@Ppeh?x`m|NB+8yKu(IDj4x`w|knsfsW?yXTs;}K}C}?)z(~2wsfW8xP%xiZ{>o2 z70kqz^?YJ)zLfTfoFx317jWczA^9d&NskAdpmS#Gply*cpfJ-3&P{K%tNSyXIv{9J!QuRV?S{kVy5Ne|+M{tK|s(RRZ33B{W4I$*7<8`}Lm*7lR+B*BcN z4c4jVK(BT^G8w;&R}DYHHb-4)%%-FCv)U&b#+TuD#)J@;q7~RaK$*-{)D_HLOvqc2 z6}X(7q&`a)P%W8JxPeTdz}}ax|Jp|P?{0&dz(sIWC!St;w2C|ly~kzhUf@D~E))0A zGQ7he5q$k?VUxNsnv`cuUQH{f(X3t3InT3!( zAcC!i5Vvc63oO5XH>!Wz7S2+s) zl1+10O@xmzXW-!`TU;*4+D&&X@Is(%hQORV)x8 z-xW%Tw|*-OtW3xAB%UzqB4e=j&~x6CPwF(rvX{>Keulo%y-b&Uj3@ek1f7l0GsI|3 zCG0$5PX`aMRNs?_DsLPkb=;Sl)+bKz;$9Bf;+M(27o5u^<|$#rP2Esz8VQ}s1>CK} zZ6I9#hRoQpkh4|@fbWi(WLHQs)8*@gKFp0q4-bOw} z7E;v;2bvkx1-xt@d{1UN-twOTk!+BrbIyDtXY@r$*+emZhe{zm<+Yib3>Xr<$UywA z)(`a>grS$SLSX9Fxp4WH2KFcy#q9pO^zr9%Q2%B@Ci}{e$F|#Po=6^VNI?SXwX9$P z7fcGMJ1@9DjqDS%t?62kODDZLM}JJeLLFrmlBMp3)MLQ}I=JC5Z5DJvl#Ytgrzy^8 ziS|y${`^()`Liy+L2n7AB8SL!$6$PEbsuA=`iod(ClXl8($ZNGB;8Kn&zbB-7mbM0 zDBE;UoZmvOh{h7BCy&sPn|Y-7eGp_Wox>cM@*4iEI|?mw$2pg$17zL;N#c|K9xVL^ zp+L%5kY_ie>!fSwgqwXNrA7oRfda&jZh<>f9B}k59o)GqlG{}9hp1I$fUny)WJU2n z(QPg+f22x})}{;klrAv;gc4|%??ej9(-6Cc)D#((5So_*tM3ThL4Q@b*)@~#_DP+b zhqf`1P}_vpD@EdobKUU16fN9b*$eP>%q}i%8IfNQOV~qKu}`lHE`8Vyt_oR{%`wE` z71MF$nV8vdWt-*R8Jb}9+Ux8P?Fea=T$gZ5t-Q2c_NiZGc{d+hA_t_$>S?^W>gQ{UC7jTbkvK+wqkvg>+5}h4Pr=1M{8(=pHFksA zB^+R+fl5(HIF_h;i^Mv zI6VA6n5mwJ|F>r~r&4b?dsa{9AijWpfL5>Euon;IgnmHU~ODiqe(V@AlWnT)`D*cJ%ryPgDd6m4LRTLtsPQix2 z9Oj3oAbXxVz=?g|i_XTI!aBMDEA3Jwix$cv5j``G*Y8KxUsZt@%de2Ea&4T@9F2n# zMX`8=3;kT?j%PM#;K`Dcu*&Fra7t<64aOD0t{a!}d9NvW`h-!=YxFduVQ+*x`a5v? z_%kSOUym&$H1G=}1NP1l5jMTul2u%mi9XG!Mj0Puu}3%`roWD*1@^0$>|jfJwQmRU zuhhZWy9JNp^gA%wOc?i%$k5!6S7?&+8d&Kag%8e_Vrn+z;(Qe&&bIXc3LT8)Y(wOU za{6BU#AN^mUfvVz>gtg{&7qkKT{zqR+q{40+F`Rq5}tKgou|E~1lOx;!k5Y7Q2u&^ zYbr@6iTw^Rv2_8+tlvn=w-1BCCoArk%Uo;|`3;sAvUtPo?a->T3Ku#$;7R8B%#(v_ zAyHNt7v5cqdrz*%NA>)%?DSITn=XuEI3;{8_&w=pn1^535#DRrIxukl$y?O(9h;w` zSglmBm;JC2Mf_U;?~;DPoBc}w-?@)pT?O{YN`_T&Xyb*X?j{9A8N|hDBks9tf=xe6 zhf{UA=srIOl&7A@v$&Ii6i0oG!>Nay$idKB@J;_h`nINm>EX$^#M2kYOT6YJ z-fx9#r*5G@Az9MCWG%KiBu9tFf@%1J5N4IW0R+!E3a!RV>G2!l+`lcAutLE4coMV& zOZ3aITlJ=5*7~^Md7C5Xo*g5R&N(DbG>SL-?RR1~q(WmZ_|lfoA#izx4K0|tmHL$U z;$fF&@RM<3{Oa9sZ|;8Fyl)l0`#1%+kJ;ek%o6xJR)S12Tali!6vzL#lUR6qqW3l! z1*|(RaEOoDEq(uy=Uu|X!oOd^$nIU7$*!CDNQWUi>C+n4)1(zC{>mbj$3BwxG=Z$s z%3`Jt6{6(U+c>yu8QwQ%3+>|Kt*^s`RwRCP|--%MHO-Tux&rOFj-ZIESq#PsEL1SgPs0o$|E{ z=%%m!cvLGMi~F3#!!kOou1h^G^UcN)^3TbQN7bAd*X>9+swB8;ynaWwOgTgYj?RRdtr67J@GrNw zA{}n5-vYng_TbSIUU+drE9$N7tWkUz2vSF^=|7D%cEx{7=%d~8kdkjuhBF>LBI^ z)oD?C4V`g)1|8eq8U?{#{uc0QGw_>L~IjUZp1T_IbyXke>UNvi1_K+QU92=Z#C z5&@I>zcsSyhP&6P&DrDB?4k(Y;6VgEUNVE2otDQF3=NTgekSjA|5UziMK`s75KhMr z-?d>}^q8}DL%i8q>#4$%GI}flXxnClt<(kFxX4s|?(qeB>}xZv@R`8R)EHoD{@o$3 z_V3~>re8v(D4-)EFA0kplB>he6+h2b{$CTDZqp;+caP^j=#mwR*dX77hNPs-3koNnsy7 z^fivA-`60&d{hbAv5fGGl?jxeM)8Q!)w}tMy=|*i4+N5rU`s8{wJF{X|3O2bX4bnP_NF0qYwpx#Q9k@SU~d zq;hK}r~6BbJT7g5BK^hw({y-FlL?osIFY)_6_O6GdvLD5fka=GK%1uw zG2e+OT^GHVy39$VCqMa6;|E8nsGl0yEEkCr?j5AcVMW~4Ni|e!^d{BU@uh{uHAJ2m z!l9me&TXcEr@s6JS(y|DlBT`9ST-LS3J1|KsZ(_C*KV?Z?;SGGvWsfyU!o>;B~&}Q zh6IZ!b0uaAQTpyHwEf*C`tE=ZT{ho>Z}+EwwmxVe!ndo)MYupTr^JyLzp`xAPo$t( z&(;b&dkwS@U8kpy6w;&hJ4nQ4%ANVEjhASc^Y=dMp~eDO;n438x^qVhbRvNjItO`!13uws7NYZjVn%WKcQeBm!bmHhb`Yf%24z5k2>~A3& z)s;uZ8?xv}-zn5*fehc@$`qC@`i<^XB{GNauHx?Lg+YydJoJWbq?xLYG}%y?Zo4*< zCX`taBXXWjs#;1X&A-LHIdO*S3+CzX|29$K#}oO36UwN<_&ln*mP3K}U&8Z)$D!AK zJ&o=@O-A!sQsQ(A6f?WH&ni8jaxnnT&D}ynB=1wb-8J0D*Ny0({}z~%^M<@tJjXpa zq(_@;HPIi730NcKIhnsfi|ReP#x3*mBnM;!{N%4nw7bTHP8^q{+773PE&n7nTWdnQ zjU?#7CAYcm3mW`)XI{{ec@~^9u|FVR>2xs{_A5Q20+L}(UrmUgs{Lav(h12*oTl8tzYddTaJ`6hp^{EJJ zO?ii0vFf*U_?s!{LBFRw+z<=jn=YcWkw(mEu}ogU*AKAwKsf5C_Qd?PC8YSoGJ4It zfh?C;PURk4pxpvDZll-@s`={$O?tuO&I&U3tNJ9UYgs+IG4nim;jV_e?#m2|mJ5U?>g zYW_{YRMEcU4_TOmQZ5&i1%MwhO|PU17^ndE8r2iP{T9`OSWJX}*{r z#9pnZnz3zUlq(W=7bgg^$_uG(Sswjz_&0U9r@-Is{g2nim68c}+~M*5b1-`G7B|P= z9_ow5N!L_cE@a6daXWJ#Y|q}cSvG$cYVGy}nm2`RsZKr@4S?TTf>)Nm&j_^>w5rCw0uJ%Gfxl~GY4+ES|j%|L`?Vfs6T5$oHvdEd6OZ<8M?~+iO@)+HvBF`5Wna>w@Zzi8k zOVIpz0!O1IpJvp>#QyC zTjY?HLK0kB&=7IA7$XMV+FW!;0{8Xu0JT`^!`JWf<_qTAWVwwSak>5lo$rw6|20eC zC&~Epn`}(U2mhso4uR@`$(8#q6rgi``A{m_-n*`M_mTY4)beVHNIHLOjY zSA3#(l7Eop*ACG4sg&gMJjla`OYo01yKrue3FofA6@5^ug4{GU&d_Te*||myer0{zC zH-W6}^q|HUx~Y7i62AU)1*sDAA&Osf!EvV~ZK`!43gPd$M0F8?i%*Z?4pz{AXQxxC z<(-`H*Hl=LMBrMHI7zM-utaw(#zK2U$nz`riPpmgMl64UAp2)YC3b{RnR{DkVE<|) z)OP|ODG)^(t<%uK-!lnMz*3p9_%00F8c{qRNqdqmLx-&y3O5)b51)6?O{OWd$QR>9 z+Y<1lZWSzk@h35t62||{C`ZF{s(G6yb;1&fd(ijT7Z+M<5vlbLAl!T|z3zPn{prXi z${tf_q{MPTS4f~V^O55!PX!D50Y~V;)Ba?luP6MN6o$8=kFYPtnJa3$#Tm9;BmI9> zQOq_oraR**uVUsU{Ba25Nsg=Ws;ANT!iYKkvn>jH2IUd?vfmv3rY2yp-of*(t-;T{ z*I=>hB{1#7T)~{P($2(O0$+64hnc4yc*lwioVGO&PTYC@@`)f$|)W-{U0P*z8a-dhT=g{Vq4_w3} zP0C0etnGBlkLmJ;uhW5>A4!awN8u9 z@(m499kiqL-O6ixRM4XNX%@Y zPoKF_A<=Hzg&L|@@WF~VU3-n)bUI1tZ3c_|xX8`Mz-7IbARlID(t-V&+{O3r$hm7* zsnuk0k`-zTlj@7ehtO32bYpGoh*_f@KMN?{-#-8v-#J;jIrj2CPAcL9%r3S1FxfdX`AE*tkk!M zUS0i{uKiSirKX;vXVzZiSOHVOB$JSDOH zb0OTZl*oxWQjz253Hs4Sy8UlLqV09EqnaaqVdrRa@FFTB_l+c;Y$5!VFWjB3sl>lr zmh#mKNS)bRF4P)@nN0X%N>Xamh|7YHRP*B#TJ)R3v=BWSYF$Z=xS8>d z`WA8H0}mWL3md-Vzt0<3`_Y3yhJ9kng$DO-c%EJ&h{Tp!}*-0rO}p8$P<0{q;nHedy(;{KBBnrl>|r@c;a==Pw>>_Blz0J8X8|7NKf&nkizw<(5O)d zLT;*fdQ~M_aQZhlgIwSS&$&ZrmH|!cT}J7`3XWfKfbRTsn}k+P$2HxOcxkl;-1n+v z2gjmWT@gQ4oNQ!oeD!1Z35Bs1IjsV&QZWAKSB*O=jj2{`5VS0KgscsFd5=%}*uCAA z$eUoEK=mh`qGNZwsmAR?oTR5X{gQZ$*(h2Jp+^=&SL6tqe=&ire{_VcrE^*9`U{X% z>Wph&{(}uh;_Q_>%IwEC8ti68IriJ(zj$OrC7wbT(5lZ*>8g_f^lScPHZm(8Z!D?+ zqrkcB^m#kjsB}g^a1->p4|?OX*Te9*s2~@7+?E(zJPz+H8CGML5j&`E#3~4~pvPpx z$gAd9x-n9N&2LY}`oGO^ly5i08OdP1xPDmwVGDiPoJ=3I?4uiEwCSY6HMDP35tk_> zk$Sm5B=BrE@!8ohlgcx4c zWjV_I%_0|HXmf8}9gvNEgkAY=6|PWr3_=!OAbV4a;myE5Y_T^5Pd~Z?&)LAiVb^dR zrW*-`#vbh9?|Q8Me|K<@-Cn$9NZ_msJR$IAq+yK}xtykVj9@n(!Tu<+XH61s;-m4& z_`PcdPU@(|?W?=7az_ci%xs6n{YUT*`~CRB=q$XZ)RcXeA<2r%II=rhB(c2R8;HEJ zADcex$2)zZu;r#p;87=xest;7EJ$)jYM+zYK9NY)l(S)_p0_gj6V))O6~SBFF5{>b zQdlW}1bEFVSTiybMk=@96Bp8PuI5jS4?M;}MSZw5SC};k(PKqtEx~E-=GbwYFRoen zh*cE1%*xwuWY;KG> zfma2+?vp~tZoeRWNc!dSZD?Yq~gYfDBd_kWcx1&iRxoUihF2Qeva9_11^j zVA(kCT)K)%ntUg-N0(DaBQprUewOQ+mx*Y|B=$424CmBb!Z&Z7A|H27gx4)$ygiM! zRMDn{o7e7z6=M+T(-B2hQqNJ-LLf5D=A_R@jzqbQz-F%sI2pVN)&^;^m$a4GE7Ocw zz95smaFIH(6_??*J1D?L0}qs_VFOZs6iLcxDX5Isk@G86u$+k)e%+IZD<>_+nH7)m zvco*QY1%gOU)wS6&oos?-daWq5C4SSt0izivL-B_mBib-c?H^-x*Z~FLXnH`c5a$X zI-Fd&9-WPy3N1cbs7*)+KF$wBSA=drgn*ZL^t}fY)_s*E;D21-A9M0@rz$a<^&iQo znu^>-*OPKO1Gc+|VC9~1BE9$)X}!9J%U4~8ZoT#A&OSN7_$Kt@DAQV8KmQ4q89PJI z^a;~9d6)6-5gvNGsvCW4E~u$(TZ%Vy)#1vjojCrS5NS)@#Je0`%bVY)gl`qq0n@n| zNTLQ;I70LH#vgS3zroU5-7djI4!cmLfn?x;^62A#a=L+kPWpu_mh++;W#ungD9jX_=Zd%MV8%kXjuZR|Gw z9PR}g;XBFN*f3hq2|TC?9W9PbzvgYyUJ%TZ(kM1(r4##8NeK-;j^!rK4P=fz&PT=b z&+<(7Ca|MO16IWgxOoXmka6fDcjwn?D3c5Tsis&Ud(^12(01-gzCOOJ<;3|PFT!P~ zFWUXLxds)gSfHeLX4t}S3Ph-j;>9WqBvolKDT4cdiMtoK;MGa){C;UF`{yst<$mE= z0<3z^;b&lf)*b(MC<8b3igEjRz&VW zIum}PnQx5H+ve8{+;(I#U^b51bsl2w*Mre%6K=zjIJ8i{k@v_|2Bk228DEn&^d)5i zw>v~|j!~Tsp6{ClJ;^leIQk23i@%BM&+Q`n=S7lPf6k#Psi(mAr43K~N-xY2>gJ}p z*ph+lKyKNFS9q0y1RFdc!iK*JBN6YDpvh$qj`$%5Z}+8=;>J^G!lpm?;<6?j{m+tp z8zRc=E=gj{+IDkq`i8i7r@DAgk5F{K9y?dR#Dl(R7<^|JXq+TUXO1&L0FG-`@nU zNT7*}l*rb`Zv^$sq{^O)h<)95+V)(AzAJVT_;ZZ#e>4z%&lx2)wc_{1SWcm*Tcn>{kW~; zmB1+(?>w&E#=l5BzWvT-8Q<0!fAGXuhpfz^a!Vkf!hB>_R z;tkaGh#J?vB!j!QdpYE=*RTT(BBuMN8prV>HkwQH;UQN@Ou{Ir%yF3(DD9ic~-4 z6Y8@Law<-uMhzK~={AI94#{w{#`clAPlw3+&x@(moJ+($H;3-jDC7;uEyBm)FxsME zjdWJchWIbG)Krj(+87&7!=0~kao>I+CQFi2F|db7^)ztVA&kGbR4^zd8#bS^#lJ4Z z@a~)DlDAjw=CKWzBib3=>{-ak zoN}W_r%Thk(~Ifa^SxwGkt<1GVoBnr67nMAJ(9noO%COSQe%lbyw@+Z=@Wsh@>k_T z-t2|FJ zoI*306U@zbVrW{~5%kth5tcYyM-2wC=#<81#2(H-kAkC-McNdU9TSE+)5B5u{Q=b7 z^$d;cTtaRmH<0hixoAey4s?3(1B#yd5)H`k(4wehu(4hPXFiyN>R~-_ACv~?yTYJ2 z;XMktssn+xi{Z_dDM;6R*skbVEVHa*IU;AmkXPACv`u0@x_v#2x!3N37HxJwCY|}r z#`&9&*h3qX&$C8(iDoF`yajYWo&eK#{6xNWiy`+nKnR)xCe^PQdzH`3SdJKK{4o=j z%$5SP8~NyV!gC}!k)q9E<*4Ed27`pTVE92F_GDd0=JCUAdSOBZn9 z1ikCP_Ta60@UkFXV6;j3eU=aqu))HNYpJK;cYFbXKNeUWqAaB z3Xz2?$_mhRN)+S=L?QpuIMSW}72Rb4(zh6ceYhx?wmPA!gV!0G^4aK>juV6~m(J14caoZfB$6aB>dp7gE16vqiAhH9UT4{ z3c*j!!L-^Fl(HT};}b<3vu*`6RqsZU^E>UD%>;e&YH#qEvH{#f^=#>7gXsY@WpD;B8>Z9b6)xGxQfoDmo5x2Q7ydEV z^IzJzhb!*Xa;(GVGUf0DTB*?Js@u73a1Ba&>CMoG_hrn zv4~Pe=1%{iM`L-gU=9n#|5}jdg&H`$I1L8RO^3zq*N{W}8`P$y0p9hC;Dne8oRyS- zLeKSp8|`7wk_|A+K@k1Ys7L#s$1|mf2d|4IaD>Dptber-rk?MIC8K{}ws;z-OVmPD z|4G=i))$^CXuxXm|Imu-In2m`>5QqqIkU9263&Ku!y)NvWHn)k+3-7#*)!t@lklSn z?MqKU?d@quYaj*oKDh+pFWTU2XfO0%QNgFv74Q**bXdIn1n>^UgY}VzXwp4jm{lq8 z>3*#rICv*FM{9IRMc1{3O1K~HH8Du4gp z&c0>Ds`K9T^#m}A2H+a83g!;lkH|NZb0u24x!`% zL4LPxBkW^Wf$C2+=$=}Nf-mnwp=NVX^vOoFJhv95Pt8Y;<(e=SHHz3T*AY2#4&7bp zfqI@Pp_NrNu>lw1wEkqJeQW&M_ zcE~kc2kk!9&3LHXK&StfF)ve^<&j!MFR|8*dyK>-=fk78C{|7F)A<_y1wmC@ytVZND|#xA%@&vo7&xvlD6g9GO#>V z531fxf@B*$lhNJDbZFmXcQQ!Mqca&LnrK}$*BN8DAS#6}H zJyaS(L&-{rD66bUWQCT9kgSaNdH?N?#<%Bw?sKm5IoGv&2cC`&facG1XcR^w&aem< zm(54*ZY#K@YU4}eQWzXjghyvCISI_*>s>17mej$#XX5<*@*_B%G71%c^mw3-3O^k7 z3^ZT@KhtK$-%b4k&n1}<|Jz0CYWL{v!BNmXS;Ve{c!_Fag2*{a+S#n8sp?gDlqF-%4*%#*m*} zD>m&D^!aN71G&(W_muSKN1IzvH6)Tr_^%`LHIGPnRXDyRr$N~{9rC_MSm2R0th&Libx2lUL9_yGL?c5mn&wk6VvyjI+Y z;nQ`6t3D1tI0Y4$7#0yk!p+##nkD4OJ6YtdG&SW<{KVt>b{!DRa z%prE+3!Qarr~WSwvF5%XOuFqE8~JPiYu?|$M!O4t=f@$qD)o;p8i*sqNe0=?i%9$0 zCHBC71Rqf~nqSvY<3UePBHrB${{kB5@se3VNup`=@i?GU73ra_t z*t*flqRATDp;x#YiT{nmb?I9Clj(w%Zx1vpY>+Wt9x2|@^xv{3(TcfQ*1AJSS$9c| zfP5eEysTlG8348d{t1t0k$Xsg4G(~U3dl)rH@Gofxk+y9I-u`gK#eP#nvnyNjZJjtrxyEwicgp7jLJ>Bgx$h4U=zS++cNnAX|ao zsZ+zTMqL?p zH41wV$kLORH%$MbKDf?ecu%On*fRxq)$1qZ0v_V|Uh%w8@X+hlnZnQSTE^c#;E5b7x{d^sP3U*%2k-#(M8u9?qgkL*DB^P|}1b{~mv!ZD{_4{0CL=o2dG_aB>LWP$vkvs697l8HMm;mlzeS8u<9zE_R(EH@8g zZ@cg!e-O4ODr4Z_dnEPv9ob;XH0vMQmFz|mHkylmU#{jnem)^&+p)Yl5Hs6HJ8tyn!qi> z7U9L*Q`qHS3W;y~srR)EGDmph%}+P1Zb_%6`yJNiDJGP5S(c4gcNQI6mdw_h<&c+6 z6*=8qil}`dFk3SPr`9U-!uGvfg`V;6e$jYzybh0jdvW3O1x&xD&!<1O=eujV@%Zs8 zJaoxJdypoabj7j9{2AF7q~WvkFPuyp%bz?EW-?aKFr=RaA3nsH$L`y~w=G@H2WL&D zSH({B^>zeCzRY;_wvQc0r7)Az0a7;Y1-c3LZvco9xpWbCZ{K~w8w&>b60 z4tsB~=+1mr^Du^#g`TVSXFG*7O@X@8D|&e&4sS+^@rXIS$jo=)sxyV2qo^1KUt2I+ z{0ruX+@u9(t?}QdN+`}BiJN{apyQDQ-`ii=xS^kz4jaI#WpXKJ<|tYjZcSgW2a!yA z6rI@qi+Q9NLoLXa|l@7r=JLHl!FC!*I3*50_Qr&9;7+WHTSV z(QhDc`UBtF?cjRa6_d?3)6)gd(BI3QhZO|z6;iw5%cEe>ng<=n0xX*%$4$BoxZ^Yf zO!<$HQs{{8YF%6&ZAY&LttLu*i$R*kd_l^8yh*wUy3Oyf%DfSI{$`NmZ!to;9wi=o zu*J~|oAyh zf7qaTvha-jGskM55#)7%)3%e5)M37#-m1pY?BMh6QkMCliLM0akaz+NK~@$ExU`R^}%S#+8z z17&G}k0L93V8AA|4`wy0v#oQRdaRy~A8Rcp9co?KsbFH{hdn z*PS@}Xeon&ujC=WG0|%0*yUDVBpzGopUb341)7N2r;nX0w$eG( ze)QY@3{%fA$NJR_1MJOEX;upLIbnbk5PB<|K5G7?fWX!8eP)hR+y7AH+2>Rs<%1n9 zp@@^;hF|?^M1JREMKg21F`K3FIOmXvna>@CGd&*D|5(A`jwH5z3j;f~0Y#sLn(&Pz zELJzu>)~4|*E*7vDo>JBbSWJ*-+=1x0eCfbB}UF%h)NAz)U5tQi(*YN@KrC386lzo z?~62|vp;&2)S$5U7PQXx;D^a0NX;0G@aDXbDZOCj_M`F*k;~4$! zGHy0Tz~omFX4gc);rblX2#llr%atUwJGgZ+0NY;$;`)J3RIYxAp6VDBUGrey^twey z%m=f7Z@%Ev$WAzx_TpT?Q5yM_=$p3(%}Mx3H{Y(Ij<|VrYf3VnsL#Up-cWkAjp*wY zJG$o_OqO*!5$vLloYnzkxZw~6%$|;&Iwm9^FxTp3^E4}K*AG^G{(*=P+Xsu`kr-xK zjr6az=x1>f8t=ce@~sZ+(seD-xyy;TzaS04qt7Dv-zl7#T?j|vyr1qfqYJ(lSdFV4 zyD=dSe)IO>#j!;=q_Q6NljERZor`I|Khf}-NkTS-AQ(=J5ruaovrCsBQlj%dsM%h_ z&L7V)HHO1w(?s02cZACFzhv=73sQQ`G$Zm1-8mCV7h^owo~C@1l_ueV%L=f1C()3$ zYwX^oTdc2AxOW`O@b%eItX^hAS@O};e(D-+{wB|3oeX%s%}74b=Ma-w(H|ir=A!zG zBQ_*#1|PKx3pR1gs$qW04* z)^4xGxJW{a&y4>J{hk@zy>cQSU!cKFjk2*<=?J>!mf&2x6#j(kV&z^w{?kA<7*R?J9|fae=U zBqflXRZ|}}Gk??Y1Ziad6nfrm zhBzhkNaf3hli&E+?78DX(ThEutaDS0NUn=Sxzqo!*UBr|uW|`y6P$`9g;+dJiNw!< zb$Dpwi%vFd{S`5$0%b}nsjgXZCFhpG+<*jMB&{>Fz<_clQJ`syT&me5| zIqbKX208IrsN3a>Zf8}TF&==wzN#>|f116xohKSAPok`RZH(I2NJ;PSk*ItCeLJ8+ zV}E&&s-YZoUiOEbPYeAwejB#PM&Y{u5qJxGF6U0w;OtA`c~6YMq2a>UmL_3&iVL3I zUxOhh)tf(`w;vo0o(1Op{p~4 zGA}7pandigdx9<;!rf_T+93+vqKKnzt#sS;D-~XkhxA_odi)iNh?%{V8uNp;3>b!F ziA#vxUXBO-3-IBJ48P#t3N^1!@c8M-i;GxUz!;~vu8V%iuO_U0xM?n zc@-OxB0&k?Qs{Y9Ggan-f0-Ra(;8)wtNtJFoHP`dcg)2x$9ZUfGKvrW<-+;fVD41t z1?7_CWb1XF&b^k#^rYw1wc;jyU3LKLyx-IGB*E-kZ$G|o+k`#OJmBb~&evb%JZR7u zem%4c|6}-?qE#%oLGF}^|(5decq|?8Ft~k!c zZn4iKzV-{<+7ykbv94H?vKQBT&0yI+k?fk*ur&WQP#CX^J@!$guz_#2Gz;k?#-0YB!PO z;ZFJ>_!5rt?BP*V&o|fw^U(Zz2ypVoXjvKTbd=zu&Uy0q`FT7iI|bIB`6L8jv5c?H z)M>U8KTFQR`NcN+x#|r2Y&-%_E`(A{+;p~Qs2s-3I{?+L3>-MO23y+^O7B38Yto`c9ch^xk%QH@v@Yj32xbByMeBB8(IzCSmn|CSk2fwW; zcwY@$QE7{~UU^Q5jNi_;<*7#w@d7IszGIdJKi?owv{6C$aql87S{%dcMVUAx8%6Tz z0kphJ9O5$)vFZFDwE7BXHBpaiS9Gx8$#J6G13%cu2ALX zx${w&5~Ypj;`t~N+lb?535#}qq9ECFQeLrwzAsM2vQ>4+d-w_Wulgg(W+f2YM?2bu z>`k$7nyanBWXrN?#EkwZab-a5HPX3OO+nSGP@r)cTN7)sJl_kRgO#x-(4NBHRIv8z zhe+{ZD*cEN$Akg9@xSh3j8g*+3HMGy_DrZR`$3ZT8tCS$A+)dID9kivcsAs? z;{w1|$_nby&&a1c1qxPs(EKa|k;1GsD)tb4t#Goc?%M+A#EZCeRf)@54Cgh6e`AKB z7xqq`jm}Cx{L$NuzMbB*U4JYU-wLM1tuC0MrGayA_~iu@X%-s)$E#s2m5p( zchv`dBM&0TXC<;0{G>h0hGMIVD_#Dj#j>7QVrSJS+J5>P4YmCRE&0Eg?oy7}$GQ03 z)Cd#DF=#I-M%L37j7d=72bV^2*P>*8`J_D)c+q29zAwX?Kb|^{uQ@jmJzZOH-1iioJZ?tN+g^x{NN~@lMf|D%Y~COt z;$ec%=ZKO(+}01ph5S|gQ*j6%Fm5m3n)ViV7E18Q7&)F@<-oPIThM)UCz?Vg;BD(P zRJNC*>(WBtupSgT#ytrY;<|-4iUFj-NKiu1@loYQT&c-5;wVS&1*Ic<&)~l zkaSiP;U+sVcy$eV8#!_Nsb1V#X*0j#@}B4I_{@#l8~I0p&x(k;jChxG(B1w*sN2t= z^d<4GB{B5IOq}lhks(pwU@CZ$%j*0D-><)i5hToyengZ}s__GA?LA4I=Nsuz(MZ^5 z6k|_VKYo1k2LwJUg0gQ3l#YcUWT6ZGECTqA{(S1^1h|{}VWMj`|Mu#c07U88%#iUT zix1z~+bh#B^z~_6-6O?UhA8vlKSS}$ITsqA`}23TzOZyS1Zm|~d=HQ$+xKb=m+vr% z%A?#&l*bomXY;>KwJ20LipyIKd0zZk>VMar)*9TUFkMAheoLmnJRfQ*@fQ3L?eXE& zQplbt6L!hM(A#(kVlp{MRH!EP4;^fvk0cK{y${uI`eD$nZFGE998KDF0nL~1qeXND zeCZ(W7yAJpc09#oxwn}3Hxo(CE#x-tDfNxr1q%gP+!!Jb_n9q}psk8KqfOBDH3j!1 zUr~@r1|2XEG8MYyP#$aqHI2pSDeR;4nG$F>Qss))zY)+aWOzhg!K1M=q5VS}Gh0je zjMno!JMb`nUzv)8*g%Z_TZT}Dtr#feV7H&MgZz~aQf|LShc<@boZv5)GWHWbOs#_( zDdE4N1CTT=4`=OiXy6-78r;}N#f=GEt8orrZmGftBx-ZzoO~EXB_qymFGjzS=cg%) zj~pcp%u{(=_2P(Bvd1+6iI)DUMglOK5X2@|u-_+-a!g!)VKZ69QLZ;Q}RYC3sdkeXQcO4lYi(|+)b6CYtQ?MRGU<} zkt^p_lMC@=gA9EbBJ^7x!`Y0uy|^XyfdUt`icU;FfcCMcY2~+-*2VR%OylN$>j4XA zkKLs409&4<20IpFz zlso@^2TScXaw~CXPMgX_mYz-s7iJI^%gb=4qmOyH9V5jq5A16Ch3TEm_&8DUUf(z! z@nKu@&}R$1wb{q*ZX`#j#z+@*?#eyJ_pk3hLH1ghRwvcK_Q?>%NDU*1wca!0?+P zmu~XoV--Zqa+kl=y`ok&YuzjaWUaxIwDs8XX$kF`I-NydZ6qG{n~KA8siWSGLXQO! zt1_T{x(AqRWj6XUf{^@q8Wy`$kfp&*vf4F?8csBbyw+Q@+)1Tu=ejp^;l(60Dh0y8 zSb-nia2#692+2-$xE3`X|J5n8rJi?CbyA<7x;C1hDD>y68xQfByZ7*>&Eh=paDU!x z_7?~|Ll1+($<=NzB_*`b>*u{Bp>UmI{(Z8p-CJpGAXy|b7=8{Pc4y=FoEdO$c*rtO zcQ8X^6*7-pA=1jrwBFvE!wP)|^H1eEd|s4L^ZXLbJjCx=mnde?15u?m4avLG*nqA`mOnR>mByH2M5Z-DN_1f=zs=g7*I9o& zJylfg{fQ+Tgi-I(MpCo&6nrr{;BLDHu_o~}YfCZBzZOJiHtJwy4A47bBvzM=qQ}>= zX!OYn^74=7yHaELka|D9c3XdLFjbEG+YRJf{%k;8%UW39n1kWQ_efjK5F7o5>ZeAJ|z* z+q`rUw_6z=C;s4V>@z5a)3yQX_O`T3QyT{_#bTV0Q&nrRAKuFj!N8~-gKhuAIh#rt#)|PjbB$S3`xG{9 z#B@^qG6}7U(_qu&hshP0G`PS3Z|zRQK3sw9H=EJv`a^6>*=dOVljOEHhH&xOb1=+p zFFs`-M}VF+ZyIOBU3Ek}x-Opbhb+RV-wjCrzMov*k3zAd2uAiQm}uq%tG=g@51j!C z=8pe<)?s&Sv$dMuXDhca`>kC9$jhx=~;$5r>|NLnZzo9aLy)Lb0mu)A| z@zn?D-U>r_&X0qRvK*QJ9*XwODUdrpogJJi&l(U&uI2?4nE8PQA7#iezfIke!%-9( zfJ>g2sQh6(w%xmq7nw7WTxo?#fj`J)l`Z1aJ@I6xH`LNHY1uJx3+Y#f^^ZJhdXJZK@&OUaG|ft#n|%s)@o57tSgcn*mIdP_6Ed;pU3FJgM=1NaqPBh|;-@N~~1 zL`)vQ%~eM5hgL#XS!1ZcR_sH@uzHMMT}3}XMndLx2ChWLvtuAc_H^nTFMpG1de8(IxI7^>D=($m?z=R&mHpQ70I?d;K5988`le8yKt-#Q-tBn zOz1hBglD4+Qa_r*)O03psSz7$w%U=AHN<`xapmMl6%)JlrqD?V;XW@E&W7caZ zTaAN3StPY@bj7w;v-$kmFZ{65HokED8~k_TGre=oB)Rn>cGjSi)FdauRrff1r4>$| zs(oaW8N-5i&!@xl?^*Zu-V|-t^I4#MKVtvE=>5dPSOZoSzUYNk>lR+ zcD!e?9`{*&4TJBN(ZJmGEMTA^yv8a+sW$>2UnXE*i#ES?dky!UYQ=|5Qx^EV9Ng=l zgU3-)7+03ct_&JR+7HY{TPHP&E_~j>6YC~$Q;+ZHoan@S-xrG>kC@5!c$jfxD@m?= zI2}x86pIR3MOsdYSa{r@*Uu~CcU8{{djiwgS6d;2J8uskG0>a`8_Mw1YcC*5o`@*v z?-)37AaC{5##8xl3fOXp4z_2|eg9Ur>8KoRH2m?R`!f05uc66HU6EayMIP!unBvwv z+AJS}{FBPOshe~E3>$8qHIk?0sq$ThSMjr6fz$m|7#Exsjh|;vy_p6)*>?o@*{REq z*S_Lw`}Xl9p`Xxe8x7yiVHg~JRL~d#khfckn@ns%%)|)M_rN1;tnM>v(~A=I{5eLo z>l3i!&j;!&EoFJX6evm6B528tB3 zr%l$iWH!2;a<8;dk=`w&c?q6S+E;n~=rH~}d6 zq>yy>F8aQHFH=&=r6oU>k>sQ(>?;zqP|Hc6IB|YK(t$6YZcBmH(`n4>5L_GH&1@dr zrlW=z;TJQAY@gjE!-CnUYLmuBFC`dUQKp9r0;qLR71b{sflH3UJZ)43wBo5YiV+xte|^D;i~q+>cE- z{>K5f*Ls+qrUoQ+f3X3-N+^7G6fW%P!rP^B{KB9Z9$fH}#NB_QVB|f-tYPGHSkOWy z=pkyO1Qf*-d5vuncWwvin*63AWk>L}_c5aCr1-3HX;|Prk-gb?m;U)?VE^A{N?(zU z+75C4cD;cpebZKv+wB;ZSA2}_RJYSS3Bu>!dm)?q11lz&@l7wvDRbBbGA-+(@t3

BrPK|I8KEuZ%3E`CfaB#8h^bazVgeivu*h-aZND2PCJw)VBAvyBs)y~A1%K6aWaR$1@4`F#So0J=F({brWGC%m!%J^mk zSq^i;@1weG)ZwS>)+%LGhK6HYgQnooq$lvWhjF_563%ERBJ$;P^zTWAd&Ul2e7=*! z6B9A-w=B+xK{3WywV;4*AsOE%Sh~P|rI7@gCM(~S~BE#8j96=ldc7>agD zNmT7K#H+_gajaX0Z{2VMn!W7-&>e#)3oHl-_SXsk%wC^D=iub*&1#!&sfCW(vAXr5kM$ zU1t8Rhu9UT1dLKohK5%(YC|H)wIh*zNoZsr?jPiB!@9Vio4Czmg;wUVCz$P7+etNv zLwHKeZCL6}Le_uMB+akV{SDD*-N3Le$b+`qJY~hcgQf5j37lugI!={e_;!_7TzTCsUZ1oQGjtatUu*>~PwFCHlTCmiexFVwWcwI)>xww@Rl3gX4%_n0#nL?T zY$nE!zeXedHzH<46H4!#Vq5OMWv|99CI4JSt~YrYA2?Kx7ySE;e1Y?Is}tjE#{VHx zVJ`M)btV1vIYp_n*OBAkcB(obKs&!`;MbyPXvDl{)~CX#e2^71Pgv3#@8L-OT8JBo zQ!#YeLHLx{gRpVi2Go6w z#)9{E*^&2EtVZ9Aa(*wMuznIq5N5jf-c6x{0YPNCOcyQ}UHMZ#H~!jsD*v`>7)lR* zBIW(X^!L~?3Vg1OU~yrlTG|gWi_PGc;{@04KGBNX{iw~!8h6~5lEf-avX_}e`)*Ak zql!twU0sL!jcH^s^)N+N55_rx#ak^ONdDU!sKc_8G)pHzBv}Gu3xT80dWk>Q@8DTx zNy)sEx>M#r?~D{*)ijJ3UbW_9*jfBEQQ>bzOL=*^A~p&<$-wJpQPe1o#!;peHaSJK zwtFik23J6Pwj>XC$wE%e3rt>Ci)CXC*rknDv{+dV-=2x%{67w*>C> zBV+VfzP(;2D$4R3Xq_1m1=Y3b8X!IU| zUsXfVmknfcpo?X8BvNfiy)gS1YN=KIG5c&4zKZ9-0GCN>t{ly|HHNP99kkYpaQnna zOtYVh4iirVRUgH9>2ideX`#^%pA${_PCIMAP`Gjwt_psCmJ>Iy7m^900sowsVyFg8 zeGD+7+yN3@O4zc_2|+quShBaAhL5wr@{dOlSkq3Ez2nLAWeao43x&OF54O2$@?p+N zw0`d*d|m$$l?2Ew5L3OOI^#luOFQz1;GLuCgo^Q@$$m*0`5=$D{DrO=Z< z5f~zz_vl+Ekh@-j|629}_+5bk-%DU^`k0EOy@~J4$K7|~I2W@3QHKw}MmZ8Aro|v& zL>QCqnZ}0Pe#6Q#qxq8Mj7Qy3;x9@*(|65T2%6!K+VDoI`Y(oVT#v=0pYJIBq7&3> z*Q4pV7SY2zI;0gyUslf{z5nu9GLyry`B7AoJCVutzfW}mE3AKTk#+t%TkCH*awP5A z%}(DrN~=_)_ye0B_(WDiJxG_SI)7(2$8y=1R4LxNTIk)4QlKQg61m4$!{wA2Cf$@^ zqx-!QIVQO>){Jg)3q4Zp^j)u&1cV}hKWvm5_G)D9^95bBJpM! z-qz@`BPV}|+Vgj?VXJhw!MhfmesK|R4d)2y&$7$kFj54Gk%78;?y`ZIM*wo z=b8!j1vJr|;vwigX^3SN=G^*PHp=v;W5~B4Oq`LAIxG}>J4D=*L-VZuNUWF*|BecB z8fu7y_a!u{U7zd7^})@y1|b6LaH`dne|~Ut(uS zdPgW`o|uHB+fDR0ZX4a1{*@&4?D5U*nZQVW$G;!1(N8uV?QVPNoBKFgUcHo7FgV3r%gGv`NpQ8%Pj z=J?TG$#8l;MTN($RpKL}WrYlNPa2iJj=t+&V_V)mWuLXwnVribD*WjT%_S+=t62iC z1)1!rizF-Ae~z`yO`+HYkWI3&m8OXqjev|oq$|4qQ_swG?}Xe2*7 zry1$x{rQ>~Pa*cR40Ywnf<}1`9Rpv(_|G*8nP86AXWmene^zwz$_*>m_dH6GSIUMJ;or+57FYMA0&LerQ`j0qRAR z^TPtEY2omEIvT&Gsi9=xAj~?xh!h<6(qqXM(rZ+~q%T@HW^N%a8svVL zj_w?aC4C9B_h5hA{Az%_@Jx2-t2Ogc)}%MSE9hf*D;uglj&35G4fMZA>l@@C6Os>^ zro*TzNfmkoUwEIIi$^CGpukF!VYlSDGklT)Kp6d?LtO{xE*+jlyEH04N1>Qs3-O-%TsYHp?5wZl()8 zU>ZtRN7DKm#?X55onjBBQ|j+HGF&f%L!WBt&*q)js8@z5p*y&@wKQMgNBrib4jMEx zhKGzi%{OE!VPu8C1uUKhgI!CoHzx_t?wau1cS885_6KN`Hp1xaLR`);LdPImy!?BU zq`ukkBOa%@;*w3g^SF>-sx^&&{jAO<8kPC$O~k+c`j6*a`T}h&6E1zYjsG#2Y~!-u z&PHG8IigFpqrdxMVUH^mo$D0&rCFC?quvh}uLbex3H@wBty;(>A&h5hcJsrFl=;$M zbvXEM4}J`I!!OTQwmF-xY_n2*4zCK;;G;c+pT8{@g}Jr3eCe;iB)&(R!)Vk_D8`1H z+I+xSp?)wL$;%$!=B9o1T-7<3&wrrJ^=b~HfAMsrjX%P7Ykb0TQ3PZhZ?n9Z6S(z8 z@N>!@%@+qsbD8=;e$m00E6xalPQSzGt#ahYUl*XEcQ_9!9?TDfoWUUB{bac)iIxSS+>?}h<3b4JGVDKCm~KE;hcesmr^ zdk-OKb}4vK5%;$rZ8PoYS!Q79qdo2dWh5AoYX2yw{>3elX>(IYP_z#__wd6S!RN96sT|dm1K}M$RV>Qiw+wU7UZ2q<5xKj=-fD zx35IrcYDmaQHpefIF#hMA*$C42l^ky9zj#OeeO68FHVPF^>zfyw9`+kHTYA167zCR zP$H_LvJp1qfAt3L3k=?%x$=DZ<`ekXuNMtI{dvsVqu5>W5Pwyw(Bt=zB6gi2{ryGc zcQ1>IRiDw4>l2Z;GaA!d195bt3nri5Dr&ThVaMv<>X>*TxL%*XpCTCj`&z zvxRxu3|@F{KG(fFjW-6~qFo#MK`$W~p>-E9=jAK-W!*v1)!~pFGaS{jw)kDFfm6%O zVRvdMYGvQhhj?>5 zMHo!sBl>OT^G2ugpRvBeeIW2n3PVxzqLNk&GQl>@?Ko-|foRP<_{|h{tYtUC$1w>p z`QIUB{D8fB^_X?@XUuHICO&3vD7QEg&E2=^BC%8rD@B?ppERBP>x7wJ>Ue63yCmFc zRwKVl_^jsr^g zJM;?Oi;E|@n!^-xbqDE3?WB&vAc*NDBi*ezS`BpTBBEh=pF zf<-JK#t`dl1YX;5EPC7K(aXR-CbsTAvl*gG$L3TsW7kyrJoOZ$(of>=?M9s4rVQJW zhUmSoi3{-tSdp!Ox6PYSmDNS!PvS|ZHjefOX>spW{dv*_N$z}mAfMqefM?Z+^IZ$} zLe#qj#@n{z+nxj}zcB(@%LgI!au!8R52Jj|Oe*T=$J9=Hu|Z1XXz7CyP>fVSSb-|E zMyz3@CcI`=`(*f@e5aBv!xG) zIh@6jeJ!{naJO6j2pI@3npp9E1ClTAhYfL;s5bp4{iypuQGo|&&rx>@74|ph><7N;@OKVd=dP)6fd5owNkff!2Bw@ z;FZjFMmLLGwhUx;f5X|-n^MeW<`w3-F#_3BweZ6>gaS@H6^+sPV14b_W7f`$p<}}F zGiWTtb5=92>t6Kw)-XY%k>yc0&G?t-w zx0;E4P9uA*2W-Xt1ZMYE-0DTsURJ!$oGGpBW4rZt(0FZ2a_px@A`c}vpV7y{^JC%s zB^{fcTcJBtjqh0ZhZ<(sQ^uxzc5${FZU*OI*6kdeITi<_(G_@k`XS~Cd#nZ5woso` z6q|3Vj|uSyU_bvkJ~|A5XWdXNPTmf!O=}TSmmus^o(09xB2`D-t!23h^ewCe<;>`_GhVqy>p~ANe4%S=dbq zwdT0bGr_tivyca&%;#)9Jl^8cM>32X)1-6t)ae^Vq0^IH@D#n9fh6QPpZv1?8^ z3*R<`Jd3QUu8^=T-T`kv+oC|&C3>B(mR1BiQn{u#+fpCTBFv7mNC!zFGerk8UXDRs z!)z=Z7J{({T~TNwg#yhK#+(D0?fy!N6tbL}PaUMbQLjZNtsg~VdgiQrsV3%29-=v6 zS}2;L54md#1YT4cxAIq8&6{}0>aXNFHd)J=eW+Dpv9awe;YAC~Zgpb%*ymV~7sCz? z_GcX>4c5WQ8z^5ZiK2Y=)7n@PS!7WOK`5HW2Q)+M{b z^uSFzn5v4O3MO!xbCf+Wf`&t^l=mcfzl;C_%Gd6X%;*$Oj6b1jL9b%?5LTVqyEHI;?+q0N#dlUV$*1-{_ zDX4BV<`(PB`A#8=`TO;F?i7~Gt29%%!^li%w7UI|qVw>_`hBCg$jr`8A)`SH&vRer zLS?m=il&yLp`|^PLQ4vn(I7%0Z4C{TN?S>#?c3g^wENw^KY-UWKF>X_bI$u<&~P2( z7sX-Lhu*mUIS2v1pNY#JP-(gf5>gYe{@ef8?L=IUeM|O5n`!6qC~~+~2*u!|n5VlB z``WyOF!_yv3-s80Y#e!S=}e-y4@F;4aDS(^Jne=SXT5QTf7vN=ikO5eS=t;h(}5#$ zKcZ_v0$i6yW7rvE@a&nW_&pT2a?R0U{x1rsZXi3E6TIC19xax>ri0&>VV6l4KC^!? zA6Yz}QvyE2QFS|Jh$y_zG{PZIM=12|q^G$jMm8{P=8VDUjcS~buge{0e31U35*#e; zz$twP^FsHHsM@qmdR3!wq^~LGrKoYu;1^K4{sa@}KS0`!$#^+2ktbA+;k{qlv9;{i z9{8^lpGr64J2vC^FP5-%;3!V{a}g`$yk(2!d@OSNNUh~>@P157L}!;FW?vc7WX|`( ziQ)92Xa_~!OeBpllhEEx8{>vQq1upOuFJILB?muY>ebFH-b!tF`8W=l(Vh8-6JI>= z9E(T0vYoRn_s{-{39T>Fy~qcCAr4}m2!3eoy$r*AD)7$(VKbpGnOAl@K z@An-EHO6!{_n7j<=*`Ndwte}@n*prxv=yHmA@`!g*PwD^2L@aYB8t9Zbt(t-nQ+;vmXcs8PsPe{y?rb3ZdE3WhqM3R}TDPK_JR@ykRP-AA z?>n=~Sxt^;*Nx+}{CL|$KWKVpLNvTU&P9KsnK`1zo}2VQ(;E1K$^$Z0 zd0k|>WODUVoSd8@a>}lWfVy8;F7??B8;!YBpsiSWWF}eKcEYm!O!gkTkt>FdWQ#WW zV!O^661RTP*1kvZR6~Pzz1HBc;J&=_yd|duuf*SwxpbhzK-^xq4-p+zX^dev8dmj` z{+J}-(6&~*YqT@JHTywZ9*m-<^&V8y&V$k|`qG0ojNDfN2>F?SEpB&l;^Zmx8Enk6 zW<14=mLJ5TSGQ;}qi8l_BGoASV(=y%+`3at?=EhkmVOIpnAaas z`0@m;dm+#}t`jJ=PMB&3AibR)pV7I8f2Hy)I{u;X8W^Q)zo3G=o;Kjyn=`y~$upkc zW*jF^SSD=dSz^Gz4!j~G3q$|>g6R%*?(1!X9{qOW&YN@$+WL`pO!L9dpG7oTz6ZV# zgZI--B$w5-je-u{vBxNutdrw^r^o#~jks1I*F zD$r|`f|pq5^Qc*VsB6^&yM7*kh4Q#qTj5LzsheP#cUNS0PA0DxtB{rPlj7ej!PtPu z^fN6AO`m-y>$XO)^j4vr!*z-Nl#;>lP&TwmG?yEksb zJipoeTgZOq%CA&aV#`q@7xDDRC&_FziR~nXjO-#F#q8x7|M7E749lu|Q#MFsMi|T~04-Scy*_XtV zU9)MwJt>X+j8v~+S( zO`K4jTGNsa-bm)9YG@kR7Xj|A@XzNWo!_X%nZ^5%sx=1v3r!K0R!K%> zQ)t#vCHI?h7t5>{qom$|T`bFCDSPnB2Q=ZFm@+%KM) zrHuHvk6Nidq|nq->a^DakGfr=F$Jm=_g{)A?tPYGlRk<-JuC5FYc<|~;s)L=E`rJ~ z`Two2V(#DssGfg^XMwx1zEvm01TH|nc@Di=79}>wv!&u-Yf`&pL00h-u;KGj$~CwR5Mf@htphn-Gur0=&v zN#r()dwl>Y)6|fh-T^Z~2Q8eXn4a?@KJ- z7K6oAa(42i9DfJ?pmkcul&^Hp(dO%eBoFry63ypf^U<47v3mf&dFN@;IyLm0)lB*g zY@l)AAx16q=XF!8IBAX%Ke&7qRja4sX7yRB2)`)aXADD&n{!zjlQcsi^LbPn{u+XvHNoVW%3 zw&%g3$OEg=$I_zoZ|EPG2YFOLad9&~aqAzd3ST41xf&N1Y{ZV2jpQRUOK~8BZeq@C$14=etZ^-`(6+RJ5JJiUq|j% z-kLq8nDFDpJ=yQZQg&K9iECbD!M>;|Mz(8(y8r%>b(aMcac??}``(9M{p~|#C9+Pu z%UoDajfR*}1FO}RTr0DyO+Vg6f#NXE^^-%uq3dw2(;hLgx?1FS)TW*mmRvXN3)J)P z!+UT6oLZ^#^A$aL&9yq2rD}@IL{sTgmdpd&MrsywSNS~Di$}l6&y{V+Xev|MxoTON2`bSaWGD@8KrHw9HI;1tkO}PE~B6(}SM3DMJ<+|?Y z5gTBInyuI9W_btL%8c9~)AwSFn<}-Nk)xbDG+fzgUae4;>=1>cy3vId9eK>y5%Tqo zp>^l)DPt}EDcu%)QvNmfM$oCtI94!~J5TK`CX6Uk6xls@A2NEsc%M^AsWV@Yo^x|b zyBtN&Yktx2sC-en(@G5dK0)|AK1Iu=UKuQB1Z`&~BL71uwqCMD>hd`hTR%c1`JYp~ z*M6d~HjYyI9(W`kojV5CnMo8>UaL4qn`pBBZlU_tS@}Vy1UEJN^Pe`^yr-2VPumfQ zW*Zh!*7FBAT=Je?!CY*;pQdOzB}K8hjiz#^?i*!xFw^uIB z)x^V^>moYunZoe#A99%#O_~!kMM_O?di~=bIW6=?C!>};`-Bg)+~0|!Q6|U0hET-yazVLl=)xCSq7zM$A0?S)#4-AWhTT5%|T5#HV~R(zh* zSy9nRQ?%2#j2{^tB?m-7^RCyTb(8j}atf#7Kgo2ZOu(_U2)kmJ;(qH)3`$yqw0_!R z^MOcO9N>(Aoh#|}9XC{;&&6Z49MYedN;OkbY1XwQvB-ZSo!|bA^me48GWZAn6ln31 zySa3GkRkl5`#^JkOL)mFrP|w4Oe}0gy;`ea`kEeCG5i}ol$o(-z4H|1~ux}=z7+78s(CNg61#r!pD@yKiGkfFDnsi^8@#LdEk1O8&ZZCW5VnsnCsY- zPsW+?3bP}Wnb-#UXJ=#1kT@EyZVbO>jr1cyh2~HCNpBZKqhr=^^tcj@-&f)hw50EsuW*#>?9{c!t~(*!g@e+QoneKY)5M$SZf_Fo!EdIiDu;8AT$0)+QDSx zTv6cUDm+HU(8X)V#ll-UIDK#((%$u?w%%5t3QxR9dq=>$EehKlIaQV8<{G)>$}-Mc+1< z`>zxwU!t)q&KrsQZ}{r>Lja=wfF~=VGd8Ls}xr+hX)o3H`zO|29znlsM5-|sL>5%-+PZ>DTsDtZhuH#s=dgp}rFZ*`{4V&fsNGxXK@i5cIBqMpHKqyIll#Qk<{_{S`$9F78>#uN2&m+hQ|K2L zRNbz@?Qb8*{LBp4+_;AcbcvFN^Z{vxA<5*8(r4#D@#xSHJdh0cwl0D`r(C3tM?cb1 zlY@9YrZZbF{wD1Dy5i-J6KG+<)X7Rot9u_O-TAA<9I1)ck4uKatPXE4gmZDy8jeUP zrq_i@c=Y%-ULI)5y7%8nwuv?uHgUtsTpJ#%ww4Y5^T4_6@ATp4Van_$vlpvA3&pLY z)U?|NsyTXzI<-7bnbj(&nsbL*?buG8ynj*$>q~T~R}`fV%b?FEvWWVwrbqLP@VP>s zA+*}#z)ace=&J)^UP+3$@yH4Y!#ulDFdk@z+uJnJr_V*|GIS#b+U4WY&)-P&zeH=t znb3}N@&dP}HQ7}}kVe0clQw)yw3BcEr`q&xyl)6a<_P@owaPgfQYHhYrMw~OvZ5JVyJ}ngOubSXnw_TD= zG>?3pGBIrUI^0Z~0?Yj7(EgY~(G_2Xiqzfy6fC1ok9}xzu05rBs?wJ)tI1)o3CbnQ zL0NW-j)z|mPh=MGZox=W>;IDWhGo&D1M;qQWd-i^UXDYX2EnoWXtY={40}_35UhO! zvER;Ox#m4g)rleNt)>)SEvVWziFQ9sr2dP}(T)$BY1RQ_DmkSH!srGqU(rk5)ks;UgUe!{-gdflyEANxizO^vMNpED)b9 zjPYRcx9%(_6sk6Ga1 z+6C09C|R_vodV}VhE=E!=7|uR-)k+MkJ>Be8nu!yd=^3Rl0g>phOWGBf>5VBU`EK<_=G=!#t+P0_naVLAcibJc;CHp?V~Gi@n% ze}gz;{EO5l*ueg<4Z_@8NLH02lv2C+Y;_TTgD+vgyo;EAP6cNU?x9u1R%AWn7|kG*dr^7IBgoGCm6J z`!nnp>*8j|`Lsixn^jzz(XBQalG|8xrob#UL}I?eBN7S26Zz~N{ojP4}$ zhj$8O4DF6;tvu{-&6Lci2)r*oMPoKulEZ(u#o`gUuo|6(Yd))R&c8eC<+*p){RNb| zGB)lBQuE{z6!a=f`TM6Wo=cW;N(+06(FozFwSzf!v>o4EX2M^)o`hHukB1|9(zxA-NLD;;1Uk3+k`OxVfgY4X!}ggd$g=@a?CzN zeV>Q#3GK1>wE^x9`%KqrB}+*}^DQ~gJb-(6eRm+{UHC>mQkz+M%?X=M&%&*1E75eE zJ^n7h-MDCLa|~2?Y{qy)IS^0@6fx7r|hqoeyk@A8Zur{8#hT&ki3f;wmhfP z1!u@UQjhlx)aB_zpG$s=A-^x_YI=#bE0x~!azCy z>WrN6^4Xa&8LN^fOKX7wXIdu@Z>D&q(9eTi!S}qTa5m?H;`4J!k0d*<4m6g zoU^$-t3Ka>DR1@Jbx#LAB7F_!t$pM?M;k?o4qTxo`Qw%W=+kO0%!CJgYV=Vm{Z*Pl zBe8ydAq;C4lU^UDNngz*c<36v4m~N#MujPkA93Mur71trC0>@XhX0iraDdi%^fZg4 zJx;;mX;hL}d8@6c8FxrA{&1}_>T+j(vZ@3B-S!iMOtpDbi=MbQxj@d;RJi(l0$z5n z!5h!P=(eIMXI8((;D76JF1eao-2@sEop5>kTqMYhT;>pMp5tVV%bA8$c5FBH@EYVR zHe>Ur4e;pq1LOPou(oA)e5*Ic_s})S%~Rv@e|6aB@(J@dMBwU$K}cCAdE5CbxTEBN zoIK#mIrS?97n{MPdn2xRS>Z^~5Qsq*h?!FY%Ms2T(4h~z<(`2-G4sevo-B5yipI}@ zqHz3n@ypGQxAZmP%kxaR*v^VSsE_8MpOZOyO}bbTnnt(wouoZlS5Wrm6g@lWkG_$6 zF(>x}O7F|O|Le|Z@SK1%_lLpZ;xSyCqKXMJC%V{sJOAqa2~l@PVfyv!;^CbDC<@D< z`}{WIl`Xh_XAm2f_2u_ECOqQnU5r`!5?wuJ1FbHhc$pt21$2bPqC7+vnqxu1Dmb1` zheea4(2*G@+pz`s{$B#U4ecnxoNPs%p&Ca8-@wjgW_)JtO@ywtnm~P$vhqzZo`LrwB@L# z@3CLbV@7zS@~4RoG3#Kg zQ6e)bbD|X@+rFj1G*@nQF@n=>NnNSaHh%eg5AU{(=D?rFg{hncm%FCm!kxQNQQGq? zT?alN;>JT)8SslY6}XyMOn4x@0-eVTx6gO^#^%55QU8YX?{r4@oLW**RiV4#O|fm- zHgQyQjX29K*sr`8-LsZbYO1~H<+WS!XhW2+d+`^;WlsFli&znV)1C{iPDZS=JH7w4 z)&0pK2e+jaJ=_!1r_;W*0}s6GF?7H~Yzp?G`%iL|wzIo*aT&RR2Wws9 zE3fWz%#85})JehdQ`=yr_gfiyt6pJhJz9C=zwem&!GepQwc4{o=_Ek$>K0`~<^~N-1%%N{I_QWD&$()nxv?sK$TO0WQmilOxD`#{A zN@Bv$;$)=E6vZer9ZZz*G5TWc^EfnlpN6CTqcHovEw?WEhJq_V0`7br3C?NU-`A4zR;{LCl8+dD)(U%GeWG`NTk_>gWk}vjP+Q&? zdzMT?aO5L0+j zq~S3ozgSLsDVxZef8-klBSjTar+ z&%6p}g5F_%&lKDnu7dp$Pe}V{68#%;U05|&<*^?Q!1cf+yn6YGR8FT$hR!v~Y)FJB zr{H99I9fc~LUU4@Aw0_)XJ(9}j#SB zacFySi9F+L!dlawTqhU^_od;AXs;v;u{cF1EDgzTliXWMjJe~v^B6SI4y`__;o;S% zq`jdF84h?(I?dL@x4KSzpXW?kfqC>X-WV^o45m$u+eO{(p^{BK9o-&GK&zF1@m#6K zE}QBwB1j*NYkt%8_9nD{TpU;T4&wcLzT^G?V;-8T&jVck;_QJl_`a_QB`t>F=Wp4o z(<@Z|?qs$h-jT}w@6o=aPMr<(F>-Gg{B>T0r-nPJ>*=dBaMx&}ULAPOOg~QUZ_VGw zx8ss$T{&=EJMPz`HLjZtr1s@gM54tmY_T%rz#3IAl+I!XMrBx2II|c$=zK3j!rK~fSvL%y_@|>T%0}$c|RXh=U>^R zle`RK{X-n68jZtU?NCtB0)>Ay(OmkTCJmW^*%u7a?er47{c#sM`mt~;Rc9l8so@-S zgLRt#+W&nddM!#o!;LtOb6mh>0}Q#iB9B6>dZMB}2GQNuP(^t+${47Wpw%oCPHKZ< zja#&5&l+fTzeM%l-%!R&sa4qip`Y{h=tO8Lua4`-eTu%KG;J5XZQ2)wW!`w_D|wqn zUultg3I*G~#aUI!7PbwA+x`!vveg9hW=3PfR!5FoBs-?gm|%9wFre=a4AIz*C1bPU zctfAPnk1oe1XD?KJ^UxRsaj@X$Z*Pqc~f8fHb}#&7x_4{aV2aoS-`Jh2gY36j1IMt z^w`!BmD55|E5{MbfW(!g>hK*LK`Ec=(In#x#z-Ia z%D_4?!*)K==uTAGdmZIne@o%9!x5{!2!127vCLNkb>G}*%Ei^<)%nF@OPC$HZoP!_ zH6QTbSbMy8Dnap@&FC8ZMtX>3cJN?N^qOXmpub_b*J3BxsOO;j*>qHlh{i1y1-E`a zmnWDsXWcnj)J@Lg4SpnwHhWizljqhbi*0h07JA?VeR^>#S3}Ob+lT_2F$lf5mS#EW z@e++#czqp-wvF9+Zo5;6lC0WRg;C^fd_gQdtR>2V1JL7k1HG=xr1uRw`Na@tp7!J@ z#_Sr;&k_Y2XSd}3b01QaYcW~e*iN%Ah10?Qb4B&XeZ|>pr~~1W)*<%Tv7UXY8&7tT zci^>=q!-eOHoa@Wn(V2V>eLFo&lJ&@U{89d{2-Q|n-XigCR?-G4lKsGL)@j5~e4hr(%Vk#N|ZGD;n&cP~5g8@P$T zBn0z7FAFY8=*1bv8r<5U6z^sqr+f1ng;IUDFrM3o`>C3+hqa`^Z>z_rLG?Jg^a{E> z7{LJ(vSBpO6sO8p@WaQ!JgT`pn=V%6i0U`!=T!&W{%`PY?E<)s6qxeqITBtaWARS~ zY+R19{i7IO^=3J@IBVRG}?OCN~h3IpY=y?vFl?rssGYQI{zpOINi&-E;-9&*++RgXEJUs3iV6F3?kLS&O#;SgIv z=_#XdpzmK?8Zm`)t9NqstQK5ZFq;SL+`vt`tW|c*j#hTouT~ZftwO{3+vr>P4!0)m z!nUKGIVx%^-ziEIuLkRph_0m|eGT4y)S6=~JF#ECPh!QrIP$Wbjvy&7MaRpCc~A)hsHr4KSo9x!qhx>P08>Mn)YZq=99oo$Pd+^ev4 zb>csF%*ZhzizZJl#K*f*Lupe^Mn6}gbk9zjsIWxqmJ`wBf+|6;ZlZOi^ap*I&XbF;vCEQoNd1^38LS%V4uN)%l}j7+hJB1-Lx0>pBs#d!gOpo{ss>3wRl3ySaxkL_ufq#*oD`eQ0RR68%y?D7t9W($=A7%B*Xj6f2Ef-4FY0QW%;~RW2XdA75&d zsHAQQ#ZS`YPPf`~(9vET?VL}3$K=|w@vKsJ!3qw3AIZNnm<`iRIlF}xj~IU$0Y#fd z(`}YQb?InvIlu>D`}@Q3kpq6Vx`D^j9k}1wp1fmuIVPz{meqhcRPgu%yz4U&*D?lY z94B+k^R+xRZ8bL@(&J9Htyuj;6Mp*P41UVqiPYL?`JMMAi<}U}fsk^=pk|KzG{=%v zwZ7qo>jkpk?@CV;_r)mLzmf5!7gR0Q;cmPx4@*|(kxiv%{>3adx;dX;`1Ii|=}r0l zzH2bkti`sIa*X5RjNbQ6U~!kA-fAU24mynm(l`1%d^P4A$d_4)H_({$3fJ3TghRCz zSC^Qw-&Y+DtldfeTUF@em%da%xwyIJJ*wO5b5FvrzkT^jmG&Wv7JXAUz)jwM_ky!lbqJ`mR*7TFV%uyU+lqe1BsQkO_3q~DZ_@u z(3g?zxIU{Bci;XK%1*>)cYdPLWg-$rM2RN>3vjIII}B~7hDVEgW8t6mxY&<5^!HB~ z-rkA}KDCr~vjaLW3x{>j9$a*&6>p_|Sa)P8p0*f=0*yF)E|^C)hXX}3i;bcp!hv7! zor_TOSgO3QLn?Oaw&JZ{dVPqhi6R%!=It6B0r#&Y)6o@^Ox#_B%*Fn;Aj znBHtJY_d|6&WrAeuywf@>Qsufp?9$^ELO4$J-Bq;N*-@<5T}M7qbc=4q3VnaXOFb8@h4PM^ApbRhws7wdV4C3-0rF6HYGjMS^iFD7y8dQHNVAO$UutYIk(x z?1i$!euxh5*Y4(646i=|Ha0{B9?nzU&f2HnklMwSVnq5a6<4E%o z_EWp7$bY|;(qwj{-Q|-|zgmh;CX29o>t{OblqF{s*YWg4o^_|B@F{+rQv6{)6idNz=d`v zCEFz!&Aa!)i$USSI;2{BxhlB-rYY=oXgHtFe8lgs-)94ZQnnZ>dxU~|;m^`*Qg;c!q)%dJKFc-bfPxHS(q8pkW@2JqIAk`4RcP^E{L^l0jJg?5=4242yDLBt^R zuw2VQH#hK}^F8^Cfeop}+l$ZFtY~;z4c~uL!o5~3X7xS;DMF)wa@&8TyT?3GUO$<# zGs9@yr*c#ty9}ddVF>q`MUPJI7yb9{rq7byYT~(%7InX_T&1@S4GuNbDa&4Iw?Ear zVn?Fe>huZjE_dT;Sm!r1%b*CmyZ*tOruUHlHUkeE_KC@xt%dp1mC9d9@f>t{8^;DF z@P-I^wo%Dsi=-&dF-#THJufJ=tu;l-`dKi4x&r=MC5U|CfwK#xej_;@ou2H$~>HINT15rgo)2>F>f6VcKk^*n9I4&2-X7P-r?z*W^ODCtz5r%oJO!LaaqJibpom zw`-lTV{S_{@3S6le%bShQ*L}-W>o9SXOW89KVc#Jo92yb#-o+8yMD|;#7cjM?V=>Q z8h=b&8WK%Yq>rY&-id$5ZN>1iF?d<;jNoNWINR|!Ty(;)4*L@ zlV|r^TbORxfXYJ=SUyz?y)9JazHCREC-_mW=5wfcTtHdQ5}5un;N7w}HDH9)LRA$! zd7TZHeJ;oN+Cs^9G)8EN17_Y!#q0`uByZj&&K{ddUNNy!pOU_`A5r`wa2{7&Z_Nd3 ztDr8wgXteDp;0sx-HRh7qpAc3=bNxym)1P)b9dh6-I^yJ^yPqQli8<-FWu9=Mx`wT znk7ud!8e;x7Iz21FAVtQL|?ufuH>(d2kA!}=?7TS6k(x}+|DGA4GNF&^Oxo{6v$=kR{LmlPLjjb>UlXmPtA|J*u|=j`jvKldI+#($=axxw7UVy~EbWrTPg@Jl3b z)#qb70(gLk;YS`Xv1{oA48DCCYRkT2O8@2@c21kMR<6d%fD|}p9K!~AHdc$z#M|{M zd^G7VQSecEDEorGZ}k;FEL5m5#uj|h21~!j!|CY*nYUd>Ps_)VgQEr1t+Qz5sBHS? zufVLqqw#r+ih-`3RvuG8`0e(QQqGv&8 z+7Se-x8mzsVVn>;1O`_RA*!bpXM8K5v}XIsvaux`GTQNpo1Q#nRUdx0HH@#z+)lrK zXZYicmIx0iqaH`1Xx~C>9NXn7eeNriJ0&M^kZMal^}w7%qgJx#RmtoMXu`oUUufF4 z0Jxr0!syaBbQq}4IiG3~82%lrq=#kd;&~V{(UxM|9x5xU)kKg}HHX_A<-#c&Irh5& zytf!3Y^3DcWYtjo!%%u6Ggl)Ao@R5aNNBRW9>trTG5~E4wZ}kLpO^pZL@Im(LA^*x+8e%1+3{m7o&Tb!tGum zT|@|uIc>(N>0e~OkelRty7Sl-rL?UiQ`zWis*F3Uh687hh_}az74;MR_~b*0_UJD8 z+4JK$wtKkbQn%qTBOSRWMU_jRJV8Q4h}hvAAug}T7teM+!@-=R=-zr2&NkbDvD<>t z;iVb&tdBvl*D(Cm^o1BOLF$|HQKguL-SgI?3X zup-SK+aIpQ^pH@j=oF3YQJcir^s8dZG%qTj8Hfwhr;u9b7oxpRXK}4Zz1Z}h>~7ka zFTVT=QcCGkxzT*2;%mflx1v~M<>krgV#Jb%VqN_%u}`rZ?s5+~e`X`h)@*^L&u*;! zegc0RY^dMko-}5p1)Z)};ph3sFt*)f{M+#i>&IQerutpDx;2&_oKK>@u8B14faLx6 zD#EkW3GkI21NBEW_{%BjEib!D?;VXC=J)^P|qtAcGaOkrBp27XVsD|=O?D5qL)RU8Vsh#MU;@OG**cGYbZPXE=2(>tul zanDI)I441<4aUv#JNT&k87YRE{CLVhj(gLM+b)#+iMCg%Ri-! zI-t`H$;66@LD~I|XyUj9^D?s$S)C4@-GebYZ9bOFT!C%lP5A0QP1fK00qM`mMdBl8 zVNI))XH1viqleTKOBJMr-xu=KO>CKwTAm?Pt-ICK|`@FrqpHA zi@DDw!>|FKu^+I_Rh`qjT!-rGC>);SKyQ0J656>fXzqlKxcbEs9!^K7+X6L3JzZ-|d!grPKEn9X|mQ47g#E8Z2OMmGro6P#^{^`asx0TO*-J{RiBan>wY| z(?Vsm)Lj4NokW-3%Md-yowr3c<3Hs;u+F3t8V#P(;pA(SEAxB9`$pl|eP^6&cNnP& zs}S_G5WYTt5S4TsEpEhPspot&^z(&FQ(eGaMKMFU&#l~frF+8V&G1c1#(M9CQ0z-l ze4Wk8&mSHsL*?^#QvWRuc6$W7R*aaWpDpu-EzPQtwQ zFeYXM!su9AbWha6w904rFKPz@om%2|+Yx*=St;l3a?c9uht>(@lBXoKkz*6+dEY9f z`TZ7Ry2&Sm?*D|!22~z0u!Jj3Hu9rq)7f5gA~Y7j;Dj)deh;eR=D;!kN)+Vi9xGcl1@P|PW11` zv*Ui_`y*Qz%HD_Z>1U|_$61)89xyK@x(lfrnQemV8kYT&yUBc z7ke;6W_v%49LwkR&(j?1+vI)V4aFqbA;9y3m>n{kI!m^o?iNeLXtcxal!e$h=Oi`X zy&cW(Y5q1&#|%g0e-Ldjt{w|I$P9rgL!z6>moe#?tyUg#K+OOdVK(xco` zUO9gdze&Dja+5d`yP8rB_dqeRDTeG)(AH%}G%2(l-+$ed9SWim zT(X1y<<*MGAIlK9;+BYzYuDNSFhvM`zal5Ps%Bvrf=YVCHX{?1!|5j1%=TjKmc{!5M9qSEc7hT#`Y<#yCZ6Dhq zWs@$Ist*vy7vIE&kTbGFY%?w`e}!+old(TbfzAyIUOA^L4?Eq0t+Y14XQ&<8^r)vL z5#?05LYMmOP8OkC9d zC(OWRct`@;&e%!oTZVLma>ktggPPs^0^C}rXxdfTuC|D2o=YQL5YGWWuM zb3WWRX5n3%kz)VxXpvDGOKrT95%RAK`kOzcCh9%dEW(nnNCxkAjdV)U`yq65l0@L1 zU0nGqiD!6+uuEhZd)bHb>!(tCJE6;=e)pl2KE#@Kd02IDGxl~I5A9__>3zJN`|hvX zlz;MNZ_s07C`ZLp);eeUbFCF^iQFg}WB#BxPjZ+dyJ2M1M&5iii^qCyE!1s6FlyYQl4*=l#xWGa7L)QFQ1ymM{1!_`o?6)|P&) z(D~1C`DDKA2=8!`yOx%)QXRPcKIA@8;@D1?BUikpYlrfDolLhly7@G zC?0G!;m(28sO*@JSL^D@@8&A9EKHyqWzR7?AOq7n4uki~m)J8>_7)XKVMdd_EUV*r zSdcfzc?GkdvFxuJt;dBu55V{Eb~trejb%=je6*q(=QT5N=kZ!x()Y#5@y+1$?g$Jv z7viV%twmY4pqlN4N*zUCk?{LFJ&cc{mIq2iUhEJyN>%bfcysTzGx@;$UR>|)z`rz1 z(ck(TwdReq!LtLV%g<+QqKeEgw_%L}U0$vE2K;Ie54~^CQU9s%WlhN{(2_GBgZ5~D zY>$|Hev+86|GEgtR>$A=&uGXGRrL4Lmff!l$uH)ZDEI$MHce8f^4>D)Hsmr*x{^eG zQ>hoBZi0cXpUExX62m+ErQ~33R9ZFRFHw8pYBdwLOMj5AW`?kHeWvVWum_u} zui@RD?tJoD6~%lRE&`p@6fxKP;D^lqC3oFNO;uNr@szg0vin2vxPK1$$k+F=^GXq3 zJ(%w4WK&u00XlbdB)j>K=4ms>ux-p<%G-WUq!>GhS(VF#t3ylW%I=ZM!BmU#GbIT7 zFC3d*4n%yoH7-V~Av!b~2}klVsnaRAyKTU^Q3{yE9-@IWJHy<@i{?!A7x%-v(eECw zl!Z+bl&`C|D@D{Y#j?vmV&jFi9^!uFa&+x4JCu8OBkLS(@i#wT>DRFrY-?Td#LEXk1Kh=G z(+TeBWy9QkUs&SAl!>(QX1r+jO--~`>!(cjNmou>O|is*7IzkFzrir}B;3I5N*93WYM4g#E6yUQ?x%LQ*m} zijt&}Xf$UCg$zk0Br-HoNy-#LB1How8l=)p{q?=y>38_;>*}B*we9^r>$%r`|8BA^ z@W8jZzKnBfHuJ3QDZ3(}ku7*yzy=4Uv8&_5=xz3K+{`h6clKpd<^*}p*E$AMxZKWG z7e{J+c9=CD2xA@gq_EROSdLG`?YsmIu+sYCq%)17`)#YJZz`AZ+x?a5^g@*<6|8n+A-iqDc~-x} zo@F_I|AQHh>`n;}d>VcogSOhRXRaEtx2p}=6YqdEuo}bz>;9nd&v24%bOraAerW$z z3#&HiFh^J)7(37P(3?t7+)ggNARI+%Pjr~+H6kGL;wrQr;W60>M_`nHD*8-zq&W^9 zCPyL}wgp zf76QUZ8`oe$5}D>eFd`(A~=S-B*$WIpcDLi*;TQp*a*W!c5mG{8}gemEy!Pot9D+Z z;1ws?9FEC9%5{D!h!wH?XYX^FepSksBBuL~%{DbZ zx7qZ+A~H=ozT7nA!ba2ep@wvHp$aQ&be5ecpU8U37qIFrIjlvXJSFF4Qh)kpHrH_h zO0U0&vtQ=ZdY$L&L?@o6DVv=4AyCckj^%D)|>OfvfpailTt%$Y^AJe)d|*AXL%>u*k3@+?i4z$ zae_)CZCGQm3iid4_V7hF8g#aja_H&ZMxi6MD*mz1K;?P`1t5h#CE1@YO_t zqP+^SlzQ=9$1S{3bOrw%zXT)dJ)pQ>1zV&%@!<;2i5$V6lJ;2KB`ZZ(x$xbnpa&9*n<+3)sv=0G@1jnYSL|L1r+ zKmz+VH9~I0W5#vI@y!OnOa=zopZgkmoqON1kM!f5gG4D=5}b&g>HRfNrI8 zaKTENUWCR%*_~d0G5iFLVuR%JZR`L=QP~d~Cn_DR^_N z6TdDA!`5fHX#Cs~^=0p)&c!u2pzVV*qGC|)MJuSyx(eeJ+r_VPtZAV`7;GepH86XuNPtSJT;7Q7RF^B!W5f$6musp^0)9pRRJ$DPcdVH7N!IQWBnx;bbcSpq~AHrSY72ZGy=|0 zSziQ-Ps(9|QyeJyRzRA+CTfSzpcwObtVlkCzR4G13dbyK9hRi^Ax|KF*HP3;-i)eM z{fxDMEF#yJd%we+jk^WxjUT7TwZ{~G*Q%oH{zQD0wT(91xKC#30cb3s&OQ!NWrI%J z(RhL-=D7cbKN7QPv(OaO<2!|#nGWQ>N|P2E)Z^H@QWWk>W}3fE~fOO$7$ zw59^Kb!wYF<+={UtZZ1Xs2=8no&~6^c?yeu#A3uFbt-x@mEI`%Qje_&o#m~>nZm7b z*F^%m^>gu^V-K9nFT<$G1?c)_1@@?T!T0h7u=-aih#S2?@k%YyFWN<)8+Vdqvjd8a zCo`5qC*Xl|8FVbL<~kD>bGlg|ohqG(?DINQm@GjivB&W7HV4}OikshF-3Gy~V!TqJ zP8~faAUw+o{U<0<$jDVV;aS1>E?p1HWtKxy?`duZ--9-tn)p=l3LcWo#oNgnq3q%X z*zeoKTveBb#AEjuu8LC?R$$okzYyQD~W zjqZVcU@$uI72xS<0%&+B0xRYSqp0dRD4$rzoSJUNyX}4pdOuwU=LOp_p zNr@T!D3E3rrc3hPI!8fM#bMy*&S(EQtr+^a7tac2UwUZ~K90t1B8~Jh zY8$N=cET985Zl2NXHF@@#XsgU2J;z4K|&JyJszT=tRz||A#SNEXY%Z3pwi`&_@`9@ zUtYWj52DS15p*Hl_)d%{mqJtbmCPE)f4tol_e@T2UI@RI7%)!rD|y@G&f+vveF{1B z7=xSKXyx{^D1Q1Auj1G}>iu+#^fFcHcAFhWop_7&S>3po(~q}~?Z85&4*prh;AN3X zH1vHrE#59iwePm~goX|DF}?G8)Ix$ z<-x$x5p#v+qqF=8{Qbz70>6ynd472u;<`sloDbuFy+=@4r;6f#y5Qnns?4nZvslJA z7c<_3fobzaX5vH(Ca7nS*Hk`ELHcF%PhtcoKg@#C7J`E3cfddtX@4FcomdoyRhG`! zAXx>u&9NZ48K`ruStYhr?j zQ{4u%E&g;hI?gr&mZsN+X@{j zJfukbYtQ3@{YP>Cl6!bdHyevbIvH7^6<~eiCCVN=jsW^_J5UOZG&yFjr_ov= zoH}(gIgc!%jrA!Yq8y5I>z?C-{%O#++lv-wi*mdq2Pn&5iThN{@oU29hEY2cx_)~A@lHQ{*ItWWId}1E z&|>WDx{e0+y4W5Wj+Y${VNf!bzH|?J3r&WrG;=c zCmD^$GI7H_eY$qNga(}!kltt`a`zo7SG$V>Mtu16+!y#bZ6R*EXANoXE^z9z2b?_; zMAe79Df5~k=LJ6pvUZ1X@10!ine+zpS`yGF)fpCeNujX$A-p4W7q7d?(fLcG;OHZP zjxky|{YepCTNj2Ww`kC9x9wC>qe)WFXV4_CWXy7yj)Sg`aE@L)%G8A;kITl&?-+;f z4hfXh>;OhbngVW1lhRr_4CuZMj|23vLZy(l)J!7P?N{(`;|9!JFU54ni7~GvG|1{? zCS$bqFY|d$3Qm48pE};jl9v8&Om?0Lt9za>p8d=4uc$2T797T~pFmSSC1bkdVYFXX zg8Ti8Kqw>_{yUO{_a>{-#F#5NHQU;>*rdoUkd5gd|xgnbyqIfAywsvw3ts zzywA6Ey-e=8g)6J!HvPvw0QI?+*_u^cvUx{{I-es=&T29YbnFSS*d8fWd*k5MJR2& zjl4f)czjbf{0Pm!{mOpWdH(>UNSa_j$65IP%!ra6JJMCo(;l8Gi#NpONGEX==eCy6 z)9)|Q=Q|%XTsnb5j3pi~ZG)%*HSQjngvP?v7_Qz=K}oVCwzMBi&lltFu|>FFZx?Kr zv&ThQwJ80t7q2lJ*tjSe`Zetp@ZWlaR_Ct3o~{Y7+++^gtf|0};#hpSP!PQ>$C1x! zA!Twi%mfidbp5&q`A;@L?I|A29Zm&j)rs`>djSdmPNd^=iM7O&8`-mo)gOEdcLMFAOaE2X!{qkR94Z^9ue@}NycGqdDrwN9_zfK-Gs)oF zByyUZjUxYD#`Vg#uxxiEKJ7Y!R-Jq#cJ(|j`mh`9m-NP)SC2E9N38H}vW7hoO~6y|FLQ^2z+nlIW8bZ@bwvnUvB_DmE0)Cn$dd+o83}qirYT&tjR&|f-&TUUBPc(9)V*(DT*r? zP-^*2oa6f#bq0>$1?g&-TPq7+^1SgW*YExErw_+tc>_~UH==088w~D!PQ6SAT^oDO z+s5$0A|@WcPjli}#DZ+^OA)f2un{+Y{D8w-6YyE!Av6kTfs=<~uzGVhnZ29AK9BYz z|F>bV{cbi)E0V+mqsQ^twhtUKbw4o$v)Hkkmvk}c2O3vwAtg6ODok#G2ii06ZNWKQ z_2@LJRW0Jt8}9sjtB<2UH!!^1YEaPggW5J0;%(fS$#vV9RCxy~QcNQ&-4!$?Uj-g} z&R{%CHZgzglHkXAH8^W{4(;_O!{r!lFuL8&JgrK`XKM1qm!5?sH+eKYG!%ET#_&16 zmGNju!Kvdf$lk`5F@4ei!6Uku$eRJ-w-zu~e3#LJ%WX7XYNIJ{cTv8+B&03%hc6d6 z4PZk9SP2~k(Z6RgV3{0N?3>GuusBmB^c~7*Zh|y~FNPP_&XB;WM z=@8BBypF4DEh+JN017_W2Fr3c*cs&wH{Y9pjpcM!^Ii%~XfR-oyY?{Wj22^np;gQxnidp=DR#-mbevj4UkdjZA{Nz#YgA?e7bRhqP6-;8t0thSg zr_*}LFcg0c-I`9Il*3*WyNM{eZ4<2Lv;qBp(iHY89zqYx&>U&*?3j8P(60)0_>E}C z%ZV7gBNc;xeZX_GpTgMd4s^5QdS$=4l8ucKO?MTcw8Jy8G&vL>_=>@m^VXPr^(G@L z&Bu5@J&QWU`xqa50OpHIaQ36wm~gEN!ZSJMyW4osv#K3{NIz6+cMj$)&D zKaP#;AQP^SP`L3a6mYC@fiNAcZhMH2Ho21FRX5J$Ju)^BMYbwo=V}2{v2J7u955qa)WFH(u-W(hHz)qBpgv&MQ00)sJ_IOUOv*oGg5Ei)cq_x za_S0RFf4%5%WvsOAIIYnam3S)@6+d?YP`R07ev;71i}6~%#1O^&s`@`M3tedPh@Fg z(qt-55T*}nJ#c!W0Lj%wlGmFexsmSWButz26;*Ee z-~xP2T#JA)~h(8vyPUPJ*1z8$7$R}1{>b4MwgWb zabw>&h%XRBg^ig^S?U(Ndp;Bs4>#fi?~mBu@|xMQeK|gFyNsFvq0D>-H%3O&4z=$z zGivUMu*9bjJ?3SY6h`U7*iR|Wzr35$bkEa1BVTmfeiCx-tAUkCEMr%766Uh|nYb%? zI4G^d{x{@I8HvZ?bgMP!ZGDgYU-D44(jE0rTtdEF0XXL{g-yL7%DT3VW5PdMGL(A= zp4#qFVG!OPKLpScf$n`&1yU0}cH}vC!HFzwg=zg?7{M*T+8?;kF$X zFFk-}5@uww@*avV3xk^e>9p|XdaQjRL{Da4$F`Ro502}4sXXHk;_cID`<^@aO;8B8 z{cAzn>8a$I^A6uDoyK0p^JL=vg64k;MUAjav_VOW&Me%Er_zoy8#nYr;VOo;>Wx(w(S$I(7 z6rMY%3U?VwoMXwro zs|C*E8$S&!drvsnI0;K5wLz8_2akGyo<)7Z&bt@!bEF~O5|PJ$6*aJC!cz1Wea5r$ z76$tdGfmw%HtM755t?i8h82@AVBa&3aqsUxDP-6};oikryw3-)8GWTT^Ohe&A zdlT$*IR*d1HK5gDDpOfqZIZcgjQLdT#Z1~Jz__2}SOf{Z7*%!=<)&4D&DwRKl9K~! zKaRnX=oIerX~SF^J;h$ID50k7RiMUsx%#H^$ddDpc3xGX;^J@U^w}EQU!Eg5Ct*Cg zARS(Loy7BPh0L>UPZ-htXQ?RkHa?m*2gaQyGaHkRGKWpx@*aNYV=rfPqBwssEIn7m z?ZWq?Q1E{6`s_#dbSog?dn&Vk_y(RVxB^=wlbDG&5Bi%ua z{Ob=iLuw^OY%!!2X1&awTjFrX(3Mf&TE;xeF~J?$w=nq84^-MAkL#z{!UMj~sDDiY z%PyZ{29Mcd+L~clld_k2y)GS3o|{7%-Ql!r@Ctq|F~sl;U*^@+Mv(h;6uNw4m^+=z z@siGZc>JN9xnH=F-F`EN_T5Uy2U*9^?GS@~Hl5%h8H<&!ufeoR7BXGCn6(Lk_~ZB$ zOjSFKt5haK%gEUBY{;cX`~5U>6F0|B~lGPLPo9naP0FY z*edA)Uap(iCZj%ba=wBOq!rm+KUULIr&fq~uS_31_A@J0tl`Bk55~OgB!im$OrVGu znSKt#Nh=p&qc%DYXZu5)KnLNW8=ln8!P=XmJ*)7VXO zfYx3KWm7}1(z>KiSbm{_wLBHYUW<4_+7answNM{EpZy5wS!wkC(GDzqwT0`ZTS>tM zYe|md-Va}R4Zl`|!AbK(W<|jqR!20E=FB$6Qr%i!-+396-SS+$_hqf8V@r!5yd%n$z^Sgb+ z%!EI5WLYGQ{n(0*aiTbPt_FH|??mOq<+QzbKV6lwBp-{_IQP6PE|T@YvxWjFC@O|~ zEvKQ&PI(;6&A_~GGwJ#}F%pa3h(E&j<9@Gr9M*Qi70X-Evxz$^7z50jtc8+u1=(M& zq4eua9oT1INAvoLw8l^e^($tgeDZ8GDxHFlbNSe}S>mL0^$Y4QSD@6i`=Goq6TH?; zqKh{ZVd2OhsC(myyeCs2PB#{gemhGx#iRK3TRB#(w56T@ggD*O0!k{1(K)%6u{@mt zd3L`rWX%W`@t?v-EhqZ?w4LrP{D`&PDG*8MN$9VgBD@Io` zc=3cZ@izsNWJEb7y&r|V$yZ@U$ql@DB8-9_NYPy%VLGbqg{x+_qQKx|lnT~`X>I3V zhHeGk*=>RIv`;{EzdQP_w`EPd;@Nd^YP6;KHBGXMqq@|1jGOb5%dqKiOp8D=f3TR; z&WX_szIC)^fi&&YG=@Oa<51s_Pq64F^*`HB!v zbXkEQ3ZX9gBQA<-`b$hw=A8mYWH!CMCrEWw&Un7|11|G@h2xvs=tAc`Msz0NJ~KO< zr0-8!=3%7S)q(5VRM^d@_u!4tL9jocfNtT&$anoU1SVQjol7PZ96k!_TZCBN;~iAm zlaDShM<}-F1A0yKfY&1`tohDtQqDBPt!|pU2}bL9QQwU4b6OYX`N?AZyD(%w>Y}^H zX|ycu!O6N2Y}|G!c3`RkB}GmF({^Q^x@0M+E0w`+6ESXou?#mhTwo^%Tw@<>PNsFG z^T|c77TwB+!Suy03f#009d{hT6+w@%dWAal=FG;6JKb>p;ycWor59k1SpzfgLJeac z$#qtEim@fVO6+WVuBS?76y?PSVFhq!Z*T>SHV2|{`+w*v9YzTWmoV~C9Pa6}WA7%^ zlD%3Qo_OMh%BHIHP1>0jE*PTY3OssuLlJ*zrl3Q(JQpDQfUmASA|aKd?CJMUN&U1t z@0+X%^J^Bj1Gz(>q;rA-8`If?qRnt`?@SuqE5Iu3iN}b3LCTe$!1^=^!x@c0>~RjH zWu6k~Rhr6(O#g}vx@pYx;C@EBWewiXvg38f{`N zM@!(7iXZ$JvXoc&cL2JJ*V0OFL6Z3%g)wHuaK%7~cd_g>&*ID(AkT|TvsfD*g=|`M zvd+?Du;zm)Rqk!D3`tWK&7d-i|7%Z+{!$P|?@OS1H zoVQGdT2@Lihi84mipMR~HW)``j@+El&_~iKLyU8$6jZ;{z_{ob6Y~j2a3HXb&Yr%6 zZ3p@3&A>98VS#igxR<)elqmPoGfsyMqgztzFoWx<_f?K#9GRoc1-0F4G>8^nI>N@p>g|=wC%MF9gCQF5|!X=^D0AbpSpEuA`YtVyNxm zTrxTPiFyK(DDQPWPWj)=WO^B8Oq3$)p&YdNWk9uk>tW)5=V3#9B1k{_%)_VAjLg4% zDAKV6XE|=9<%apReEn&(&N8KP?ral`$~O@joeu&c-_bm&9i0#Uqww%7TBPm?e=KKE zRgx~1={4fJu?@(s)a1wun|sGjAY;rTsaO@c4sTBG8lq+IdLfTa{+s_ z_Yb>d_er*x@n$3MS+dC|7m;731c=u+b~GSPxsArqHSID7;d=6i)=%qUc9e=J&lmMse>1e3!8l=ex|L zmN8?R^m+l#x#z~Q_nc_xjX$kE-bxR@oFmJGeCA;iwmVNi18I?z2Ud6I{8QM%8T4m}oz*(^iXUB*cIpgTI!M()ngO*2jm zF}uuk;BVGf>?5AfokazF%G`cy8|kUI z(>ggL8W%Z?c|#Aeqh$+*jJzkMLmkvne2HG{tV8c37x8ta8d*wUZF&BIV^+ATsKWxR2Q)FQN7#lx~?aCR&s-4;fwsVy`VWvYSV zy!|8N!+4m6?m#ua3~dM928sR8X&HLSMUiB|GcNa^xE!UqYoNhgKEB?Z{w z)H0e;rUZw+_#xDY(=A_9Ocyu=L!$?9JpB;r^nQauUs($64MP9U$q<_q4a$BRG_YkC zZCVwH!e+bBfA>0SZU{%gjB8B6Bsnys81y;tkKWI?#Ii1_bR}dWw-4YW_dgZ1WReS} z%U;G+LOB?qHVHrd>%^h>Z2Yk68|v8|rQpOGBFci!d|%IQ-U2!p?DbDis!_ zkRndQ728Hz2Oq<6kpi3^k^v6mGjQqWBogT5V?|wF!+Wuhs6F{Tot(!{+pPRhSCF5Y zYK$?2(~oaVi@+%R30U)>2?C@&XrakJ{F5Aqkep9fHZH>(0{dw8;%_w5DhN&c#At3M zrx{cn0F{3-sBz&md|w|)AO1y9iOE_D;JAzV_xaeHmDcQ(@8VQ;{uS{V`OuTO{+Kl0 zihGm3;Nb#3RM@Uc&o(*JsDm*iH~)t5#e6h@*^0-8+-TKnUDE5b!rxPiP}c1jvO}Er z<>wIH^}a=6`pMMddlknG)zPVZHEsI9Pl-~wn0BX#N|c-F(A_YC_eba^*X>mA!N;-q z@6qt-I#N7#lIjiDl4RI%YV?>vb~h~9#`!_4)6$)+((rHi%c-te9~lhUauU0jUqS=T zRrrq6bQk-rrGxJdAuSdFjfLlVckb-wW#b(PxP2LZw2Z+XH}2;>;&w$##2BxLv-qsy zBU+i1fJM6wiE;f@7FJETs$75)uDPP$Y;kVy_m}jq0W0HHO)93wq@?CXZ2o%`tc=I* z6e~Q+{(u8PW^93sIji`&gx<}nq|uoRh~K6ek1eW26`NV~q3b5j3)2BSYR;S~;Rjh) zS4wSFC*uZT3V0d@W~Yx}Z|W!9uI|q3DHntmZtvCElaBI!0^E122>l*qG74+$arc+^ z=$@`d^RIBacDFw|I0&kh!ur7|GMc0FKPje8`ASrhO{nIa0kjf71#LL@)YhOD;OlCJ(R9#<{I zWiFX`E~beIc-anRtHPM=-)3U4hXoo}RicXaP8u^=Pof#-5Vbg)!7qW#`Q^8;{Z$H9 zgippz3e9-y#u50UvJpP%B1#9SFw0Ik^K=K3xGvCkSoS~yx<%FTsf8SV9yoYmH`47h_T~&sdn&Tuk&yQYB)*|CB zag2y*WD50f^Ey{IV8Ij}eAR!CS-4l2w52;?PYplL#^>b5`6`8~IPZAyBCP#jgZ`z5 zQ9ZPdBx6g-BwdvHy6)q@@qW0nR3Cem+fmfJBWU{|lul}7Wj&}_&AUKR9=4v% zNOQ#n3jDZKbtWF$=m7&KI#Fv}kzSjsELyMpv+nnF5k%lJ$+03uje-iZXTxUXOr0}w?^<(uN=(V`3nDh+CmrdU9st6G|d<1 z_7sl;kW%+ioJ1*&nO(rMaoZ@uA&zD%FGQ!I8z{w{XZ0rpoWrBb z!LFpWx&%9Jb}_?M^PJD zfnEqSV%J=Nl3@idv)WBcmap*WH$PO`KZ~-y@1?rv!vymZx_3@TS>V;dU3>T$MSmNdgAg1k-;- z$2c~E2&L#>gzgvHVaxpgaMiRPUU5Pg#JNPH?VEG3ZcPjmAa)p~*9p>^!w*PbvY2$< zm4U_YN)$f^Wazny#D#gZSFr$%mY3j%zd~exo$E&1aShiVJ&kaR+uIv2;WDZrY>VbH zdLh3YCsi&c`RBjUWxEJGJg10arwvGO)-sll0e;P|a5nl|ez{j>>Mf~aSL2Ra3+QnBhY%-%PQk;d9IU1O`96$pnWAK=JD&or9Vsi4W5myBWS-Vszf`D_}CV?R!Zm%pw;G$MA@PAxFDH~mPk@m6(33NzlH04#7xEX`Aw(n zNMeV)>Pau=4SBEIK-CIMxo2S@F6Wzv1@SlWhxB1ooMVT7mPCO_nkX|vLlXArS##&+ zHMny222_4}0_iy{!5^wZvBQb#;G$O|FYX&*o z^I#6VlHv9>%zC_XJ`1heJ5lkC4eUN11DaRPGnQnV!nz&ypue>NvfU~;Mgx~4lO4jpj2(qfpGXSY zt+=!|9}QcxP}w&VcV3-FLHxg=wcs}#cuh+n>ImGI31Y4;7-i;n_kguU z8S|y1jfp<5OwJ)wsr}pyoFe>{^7bjSFO~1p4ePrk*Zq<eI~~oi z#OV?aeD22JE7F)86pk%ichd(m7d9rjnyS~GLfwfEv79}EEFGgo2NWsD&j}N>I%)Hg zRO)N9r{z)qk>Yo5X1g6j zSw9t<)pxQP<`dcdgBGmty7f3%;E3mzX`<%FFs#m!!^F{ISgc!#lAU^vr z-J->R#HeH}5WO{y!$#gB=0NLIit2Afe?=qA*Iz-WzWM3vn+UrK~L4jE3HEZn^2nLX-_8}nxf3O z5a+Gu^2Tq5@#p@Rl;m=S{MOV$h>SI-s7X_Iw9sPD`Vb$SA@>G=KIusw!9lfKJuuF|SZGp} z=XpD_t`sZJ9LHN5MQDNhYPKd{o9$f1$L`3^$AD9Q_$KxM{yZ2??}n_Y++B(uu+1pH z@*ReDreMVMNPOlQiFXZC=yg#&7-lu0>iZQ`w{#zwH%($+<%_d-3b^}$pftu-ox^9j zvAA-%7Pb85?&J=~nX^4XFut9^)de$Aq9BmALK8^~#*^RQaQd;J2#*NbqHcE#iu{bn z%vN`B-sOoh(R$RrM+k2@C(-5NCQ{|tT77CPj&2K}@=t-d_NELK?QFoH14roOtzatI z??BQ|^HImlh_)pxq(+TPbm&j0Z8J8+nTj^xXKg}TM(SXkS&qSz2heTw8a($Z zLxaRav`1f^j{KZN#XGcV`Qv&j<9N)yLekXYJrQnO9;X$1{?YS9Nm5*JlvLDGDMC<~ zsxC&5@7QOW9OOi6t{kBqPjspO*fiSZVNdtjZ+JgJ5+u@UP$R#Rd?w1^N{wiEkzs|- zY5b&q_7EM-6{h9)1-Qb?&9slHr+VKw8vLa~m%C5m{(ZY4aW#W0o0029R;I!|+ zGBL61JGLFrB6r_Biq=0v`%)7rGfx1u6DCsC;uzA+Y-ip~<7Wg-8zC%D4mI!S(Yhj* zcCK!L^Ue9RKxG!&7Hoz3?Q_ZZiZp?V4#{yWg0GGCB(V1kJr2w#)hs6}K0h6cRdey- zon}yJSI2tZ9f(eU1*U^nXu|LL?7=yStk~Ws*xRN;Rjd#_roH&X!Va$}=AgKB259>j zlctCbeTaJuLL3jpTeN_VNPNU!mTJ_$ES18pN#UBNZFKaF5+TPTn{t!>2b`5S34B!+5bJG1 z?iLN0oO~Y*V#Khui_7dv3t(ClVql9t&Iw+F&-h%K<^*EIH$7)o+vVX>{RnIkcEEv? z)i_Bu4}X2XjV5PAFy2fa|2z~y@jHaYUX!_;>}>q6dl3*`POq5non ze0nSzpL__wN$#O|`K-z&jar0q>zIdEx zvjQU~IN_PUh|lj!po^v#_MAS65|VK!DzgqXj%A|nSOEGo%*QvzrsViMm3nU+C5HR6EG}12Fphm(A9s7@Jg8wDjrCHW&R8B*Ig|f z8I!_1tbkz408lF6GA=Ql@Fm>_;)A0h$ox2N;xu#(zZUegbfTFBQRrjgilg20@M_3H zJmy!4E>$D&LUI@nxBB^-QWCBjk?Sk(Qq!BlGVoegm`8O=_VAC~N z^wtE@a*UzbP#>4~4TA7mXH;H(4wJPfV9)+|9=$M-WrrjDx+bm?&o!0S5$;3 zNw(}VGD0Yu($J2U5>g38LPfSp2xUYz75dscGyLAaf8f^R-p}Vf&g-1#aUO%9d_{~Z zYNadfu9N$n33$;lm5$bme4M@&M-5S6-0sg_2X?~JHW441LlHJ&E5_bZh4v8(Jm6$x ziM&|B4F&FSOH1@@XVFl%M`U-u9cKCV7kaBFKJVYcp#zR{mzI-!I#KxU2HBwI^jKUD zP~`|uO@y7(#vsenzt<0z2N)Dl zcejJol=p+`{a2HrcPbUyzmTr2amU5fXH<5*J-(&n;Lo5bXqmPKedO)naUq{75-p|7R}j19jN+PaG5{T@tx% z(P8?oJIv-3!Ew5ngQ;l&vNSBN}k{KB7YyH$Gz-> zc#x|;|325AmMRQFam#4vzx_apT}~k9pg+Ep4ntPfOVamjqEj6n)3wNMNLi$dsMTAL zZ5EGolkxPy!vBl+k@Yy~d zSsxWKaDXyt7Dv+c$Vu$H=p5HA3*{8uv8>SDgPl}u`TM3IUZ5jpEbiy=LU9M^*w{SXLpl0v&zak9$qqw&)<`WgnFAQq=!86FSOrIKS72 z&)>~7x9vh6;PI6#5`amq-6+9ep0s6QApNy(!zqJj@R8&4~?ZPqMSIo(5H}HV& z3#8`h9x~&oBwA^>AIW|S(CDv=@sEP}%+vQ!74xlSsvAk(^#REe2gC8hBc8Ci1gi}% z;Cj3Vsb^Tw@oXcy*%%=_Q*HRXO)P?UkB1a+m`>R(qXzW^oSdtMXA7bz|NB5bo}!{ZUH46YtR4Pv0=><8Z7VN#PZ8`F!Sg|TJqYI zOTs5(^ksV(Tgp@mF*vA^q zjaaYUO5O*oxb9CkUYq|~iZu}4yk>8@P#Ft({!iNZFO!Ncy70`b_UzH352rUnY0hvn z_Wc;cojr$=cE?I;8~L7WPp*>MoGy_SWEe>*>a@_S3@)rBas7Wjgf16a%&OkmJ!p+gcyOEyzpo zZfyAGs|@t89gET3x6t4EY$~hhjEoDPXnVmKC~qmJIqyFqxO^K|BuO0o^#OKd8FRf2 z&?T)Er6!%Y{*A{ZOvW7P;t+M zFB!;nYge;*-cWqr`2(*%g<dh(dRq8ZG7)!j`(-V%FxTn${WTxAo=k=6 zxb~Q_SXc7%4|13vqY6bW4-8K22d5vKp_!0LX045sJGmS0ncJTSznd&_C0pU5naRd# z2l$FgJ}wl*VYbgCG->2Q!`>bTu8l)S4^945k;ADwlDS@ME1&JXi_>l;V_xtEzWczR zFPg{m!JX0Y8)Jl(;T|ylm&7%irNUR=m$gU5^KXrPe8$L$m4pX*l2ia$_%7s`_Tp_r zQ*rT39E~WsMC+A@)7`=w)bF|re>k>|d)?fPQ`vdwKj=LEoN&j$_vV=6;)2$xgHh^M z%u$Pm^Ux1FF<()KvzkKSJ+=eS7U$p@z3h1QR>8%(BsBD}^_XNf6E}T2^4}FAU*ggp zyX~UkGUpRLREkDgA8QU8oQXB7`k`QN6+DmHbDfoq@NFt`r&$3wX1k8&t4E{c)>!;f z7>ZcAf%vQ{q(vM=&vB02PUH#3cW;B0dMY?@s(`Ydf1^V$6fq)Bi(+(tX2R zX}p=pSX|)e5B-Wwnvr}e_rd6xvP56Ck47=HVKy$=d9%~&USX*K?diNP;p83Fi z7f)twcPGC6wlACiS_`FyrFiKjauxdwdEV%bSXtkUJ_X$o;WiY5ua=YNTx~48`~lt; z3vg)iBWdP@dFQiKRHsw{lMyTro^69LDP7BuI<%3t~mv$W+T#eZzdzB(PYycE5M9|GyW zlv3PR?$2Q>Uy}Fj9r(1}kbAGzCT>-s#s8jAtw98$=80K&LlT#2$f0w_1%<&Kd2UMw zqz~`OMPnDT%f3>qw=!exvBDn}GL)^Zg0J4Xf(+$GGP`yM^(p4OZFwhFsq+ziG~hAi z?J=PB7HYe6;l^iudBiF&uD1^4D~F7P=Dv))JdWeD6e)fAkc(Xzn*6lCIzp#wVZg*0 zSgE>{4_qE3*rN}iG(HxWlKS(i$tJvOb2hgYwvn$GG?!I%v_xJk1rcLS_=oRpJa$!N z!)52N)n_Ff$9xkU5_NumUxi->kL9aI39R#8_g7_q{p81@4=8fo&LiKH3HQh&ix%dAy@SZ=NBj@tj2o`@*m-E0*tS=|Wd=1ZEE%gRmaWwCopy4m?F=qCWo- zXSNDK0Do*f0Uv$iF!A3@teM>j6QVT{Gh+|3YD=k9IS!l5?jX1CB#hMF4*B zjn||3QM*XdXK+dQKUF!XHipBmx0Qm9{h_+D{rD+O!1+7cpihC)qHi-;ws8^HY4l`^ zYmT(`KSi9^H{wfPMHHWP0g)nCdVZM;#<#WNh`LT}=ii&!CoZS_Z>MRTqX*p7o+4Fl z#!1~)@z!`vd5oWjyduCK_dI`KVU#h)rJkUkGZtfHNeG@eMC0S2JoH|23_oXo#E*+g zyr#o*IF7!8dkSNDX37|z>2U!bAtMBPs1tM=B$UV7v4!w6B_;LXobUUjuxnbfw!DKo zi++P$LN_+DP?f(PS;>Q9Ch+rz9ps5UEaVG?zvQQ~g8W5CP0rY)4Hpwf&MJyw*R~t^ z>~ClJ_d)yQ-Hu+7uZXQi!Kp#K?z0lBRJfol_yYCaQ%zMT6nXcQgZNUm5~Gg|LH{WM zG*G2U($k&=<1YV5EvZRT)1NHP(uc6GC60Dg?L%6MKFwZ}h?AWXkf!>Qj(F@qq?#^@ ziq&}LYkS`3aExx2x$~}G8uAu-E^O_N!i~ioXv_sTMTzxpb{Ia4wnEx=Tc&Icxtl^S zo^U#i?bYk~SX2U&@cSH3~qq`ACHSCP9P*I~c&Hr)A6 zGp;<%NAZ#q(DzZ~mO+l_cW|H7r+BU87PXh>c}~Tkp{dfzj;5TyWv0j`8i~C$6wz7- zq`{*)IqWphV#jIvSdsfx@))yMu+bFcy4va7RxMcS)p;Iic<#pTn{u8xERFpWf1umW z*R;ueI~iUN!Gd05?(Q%_K6Rk8eB_RoXm_@T`!>Jfb%|vd^CFW@U#h3BUtPf}>WKGW zj2l(ODDSY1+rCKS$k#QT*CX&^UWnaav{j$=bhGVQ&`Bv{x2?3!JP+^NhDy&w2C4T#5BZwlOu>cskRLdsfTY&Dur2UrqlN=K zt?nQPx|y?^dnh|}w3eTl>>!_})R{l-cjr~DZtU$AO>Hl1K-e%pHrd#nza6wA*ScE{ z)8B5v+XYo9{$&T}VZX7mQ@mu@uB&YGmn_~hLq+~;$0&LA%QgHkPk~*{&XRdRp^S%b zm+kFdE2Y#Jl3qY>vhc1(dLLIl6n6yYzdfN|>$|aYzA|?-n!tsJT*W5)l}GmOAb;6I z9UAf}__J~+h3wT}v%j$jY~w%|QoG387A3RN_4^!~Y{fHX)Uxh>d-&wuUhH06h?Xmw z6kt9cyT_-(C!z~?JM)n0)=#7Uo2Oy(>Og)M;tK5_vnAh5IVGR$1cd@Wn2wC*(D+$A z%B~Kd#XRnk?sHJ^b%%Wm{^Fa;5I*tXHGa()%`3vYacQ+KzIq+R%W@gFSMSQUq6@1{ z&I-=c6FG3N!*DY4#JgWl$j5gPckI(v9_2Pt-W2l!zFQ`9OPe&#pC5yJf8Fsg?>?0} z%!KuPeLR0U0ts`gX#ctr$!2AaYuIFxvxbi9Z5R4b zF}E4v!z#zt@{j%*{Khbl`?Ooj*SnA9<_=fj@#GCQP1l6spf>n>Cy)LV29x9SrSMl- z#!XeDc!QV7Kz_-T6@}fBg$UhXSu=L=L%u z9p{=cEk(}FZ`9bd?hIKw^hD>6wkSQXLu<}_kbQo-Pr9hK3tJ_7^!Z?dO)(;8Khgom zH>jhV^Ap&mUBjyv;ex#-!=7=QaipXeFP6Gu($EUp)?z5jyEcq{4z%SwD+`od9nT4w z^_X*K7H$4LQ~2!%L(Q$0tY>J@l)hzjKT!`>1qS4tQ-ZbA1=nd}A+`DSOL9y*B@3+d zp*6YT(p)_m{ZR-Vomr+1p* z<-k-Lq_qI=eP>4fr?!y-yb|Cf z3t<0=E)>2y11lYuaboai*}TYp7}jGkpFVe(6r4w)MsFtyE_Q<2m*t52UP{L6@5_Qr zgJr5SI*?w?cp4HuRXQ?6MRG2Wf>z5(91JSMQ>8ilPfCa8wn&+C*jzq)!j9iGj>0tK z{?cPuOV1N#(eeFt+(pxtO;jgS+Zk0bJ!X$p)gF{=GMc9D{36}HWk4$}L|^jz>2gkj ze0_WfH0>7QbgniAYb5f(icpSK?8?<=y`^6&L$S(Kjc05bz$cVbxa(Jc4mgz$6DK2% zxa-BK;y(Ljgg(#m-$@6&MYC9(G8_Z>tH?$hNQ?%moe8`sg8mo&`c?VT5~UGG1H z36F3{p|r2!w6tQzLptlN%|EAK;L_#$F~0LyzH2d>e>e5vZK)0%^ri)c zckdxMDh_|U55&B;7wPxg6XfcWPuB5M3#hGiwmm_@K<|oRMLa|qWrgZM7B1%fr zu}^ORf6Gyj?`pil!M*kP$POdnT~x-Wd^MSLU51J0mT|aol9;6>^VY(BqNCeYaP*Dk zk!9oL*G}rn4Se_Uo{ho0yz^fyw+lv+$WUyaxS7i}FLJGsOuo7*Pp+mHB_(;BrEW$m zAk*!I4BrOIp4l0hrp#{Jf8*INW424p$NAM?>CSg`>G1nZw5(5uyU|Hf+a&Qi(NnfE zCYwgLE<%mBDdsu9lz4zG{+k_3QJTW1+_sLCvWyUNZzmf1UXW=B&0&nfL4LFIH8k6v z!K|b|Fl;n|x^_L?I8sa_osW{S&|Ej~8NhzSGRRoILcaCl6nic_DEVIy!%=6FrY zicYYfoAUvO90TdVsqtjlSd6Z=A~&$>Bt8p`(X%v#(qf(9>vEJLZtRyHZL5{~H;bNn zo#QAl2}PQ0yI`b={s<=<8t=FWj_L=YHN_nM3ipuK*#XferzEwV=2&kZf{fDR7+j!F33q->!uRHG9c? zBm~?+N^)-J;|>(Bhz*qk3F>+Mwy=6 zp?Nne_&D<4t(_4iuaPFa2*cfie3}&3LS~)DLv@}iFMJ!w5h1sEw0K{>TP`GxS-Eid zqrmMtgdtLCt)yt7OD_WAurIorH`q?$8&8Ko-$oCz-inldcMrMj%!21V8LGeKh}_gS zd{MZ>eXDDDQhFjKKCr~4Uu`-2_)PkxuLZ@SS14gv8TU)+D_{7nxBQ|5U{;-hlSNB8 z)akNp%>rZESLBOV>b-Hra44ndC`gx*+py*1QaD73Ja5c(bb1(%{zgsG;Kf4!QLp08 zF@Je_vn6l)F@cTR7~pC0=u*&O+l<_G!9j4M+4*+z2?Jf_Z# zHd)Bd5T3*CKk&vti8CiXf^)7R0-CqMrMm=g@oe5{`w<3BdvWMXXFhO8lb^SrhdnnZ zNmA-$46xmX%@J8}3ofJ2@zcaTQOvC_ZRH-%+p&&)u_QPnJbv#Ao_V^6-*mjli~Nr9 zQf+na^QMJOs`ceMf|Gv1vAulb0aN+VNxQh^d0XCYa}}#QIHGN|qTHv`47pcBfIKkM zA1}`j5Ps|t-07%3yEk;?ANL1AtNJZ+)E^+E&1>3DmN?^ENrP@wO7X7-V2kfpD7hJ+ z_C`-MBwv=PR{WBAF13J1swraMy5Q=Sbf_uSNe{B0Iyfb4Mfv;NbjR~1F70_Qb-8d~ z%0Kv-9$0y>VNE6aR9V8Twh+VBH5n&G_saRRR6S&D((&LrM!)C+pa6m>?P#7I%o^bDERbD_M^#u7|VB@7{dC$Cf`>yYoiZj{Gfo zEw570;3e;JG4I21I@ReI{W={+qff2pnJZSak5M{qt+2v|(dwjLmA+?-o@5jispVwV0?IiL|OObo4dhc(n^c3MSILZ}#Gj9EP(G zwor!>XDKjn0*`+;i<2xJFh>*;Tv+-Ocit=_@0YVM@jrJA|1gIx-(Ntnv#e04VGcbj zeHztTL$Sa1kg>i8Eavy41*iAOs;dI%jgk^DjDohgRiReq!*m9x}RSomW z)I*z3yd^qU8qqDOatyN)=2rj zLt4AN8~0O*;PC97ynT6BTGX|_v}RmAEzC8=$kO3lS(nXcQ&-58j~LMH=-H5^?x&=R zN8-78z>y&qI8q)7vDX5gw0C;5#vC{9|G?+n-#I z|27`Nr_5vgzTA?xh;EK+$G;)kwJ#K#ijX|xG{#MK;!AHl+4QnEa)K_AUY4AuzM77t z=^c1;&lz;q)>(G#(juNvIFp;3Joto`CyomLbMHsCJgQ!Bcn55DsB-?}@TzZ(w0?O{ zns_-~dN?Z<=hjbPy#-U*;vBPyrVktR%EqL^k*IrjltQN#Aarvh75-F4(4SSXtuSWq z*}lBg=sXsBzT=r=Te$phI3iru(K<^T6i!*kbN+VVpgDaxrbjkXQ`*D7qu}fuwGp1} za;e-%AKQE?aaDg5Kl1WGa$K~mPyeY>^t`*&dD3lsHGat(hqSQH?adt2=NmdcD3lCC zZ%ZNToq3YRCOk3UE42+7$rI=To?E!XKO&2}b|1}GtZ!rK{06L*BJp*xGETl9#4(3& zL3T8r`iz_>KYY$#e#O0yJhi8Se9V{cd@k3BAHGo*IgutwxyxgzT4+>W@fIjsyNI0i z6Q#a~yGVV^A|$09;V>6x)lH8dbF4rGBvv>yxr!@@p){owmiC6@I83kSo1d zCZneC!I)m9gTs;|#mubbfZ*#q-E#nIe|sR6_&Y#ncNmOUrbDjmj~Fr2*_zS=BjRoO z(s2h?{1n6f5qI$OwLk5u+0F6udvMGVMGW5cMoOPCoO&+lA$5K9f`dY;c<~e+_83sg z&#H@femU`vFq_XL0QbWvY`ZaDU7Wvh{Am}bRYMzmX6=|{VwS2le<+0>&Sc%J% z#hkfdD-xDka8!6#ZkN3Zr6WI*+nzDhFE~XSa#Nkn1Q)*R+a1Vo6+HKcHstvujs7`K zgpObYb$xbGx^$$4rZ4}5!O~wOwk+hirS)uktbqRv=u6|zIY_-`^u;>MQS@4AD_!!L zPGfT;*;8iCQPn=kES|*ubz?bZb{3BRUXOyQFG=YoLB3d-leZotot{(a?$!RBH?KV> zZ#setKey4517BqY>XFo6aHtNNR7>FwOQjS=!8&~?xXsOZG+tn%d`RV zZJ~zpaY5GXZz8V8*)t@@aM6jv{2^(wGWKJ z!E#+F%<9Z)x6SFe-dv=dN)*rYK#Z(BFSV>ahv%1Pi*tJvg4FiPdkuoztJ_%an<%^% zFO_)e#00^~4`El+C`7y|q_o2#%RVNP1M?5@^D}?hI5LrMs!xSi*$q@Cgmdi}Tka9p zijS6IWV!FD?5d_UD!jLoXRkVGzI-l%w%W^lK3>jM^kx-7UR zd9>Z#hl393$@RQipmOdp?=A_IS9$i6gGFvhFgkZD)FZk5c+To;#~srAalh?;HXWnE zGbai5g7Ec(e{WAW!*)m)hO4r&Y#yyKa+b;7naV5O`pLaBPqVp7F{KATr@*cjbY@5l zLe%tWsgI`Qu5uaHvmW4kWG@8!H1L-;rt%wE$62LUFuH3Gk_JtwlYG*0`Lo{@uG4PI zZ#_aeYv*2(^<73;HK+K#-4s3=SxFyekCQ(Cibd;C6V|tk$Nd@e`Mo2!!|pOl&-x9= z6b*iS@iZ`ZA+L+*!?paDHVeP)>^&#(qxpwac0UK7_Uvb)&(9={;1amszKoqi|B_Qw z2v$BQq4B?t;?CBONZezHHt)xy{{}6DggelH`~u4C84Zj0&$M{VT~sgnhi=u?Wak%3 ziZvPd;%bN1^~HEMXb6f8$H8i0BUxPTi%0+ZV&|rKDM&FE$8Op1yqRaIBs58^2Wkk; z>`1rm+QOp95uf&#(Wuw$xk4$J9ba{Ue#SxEI(!n(MP6)w9~EBEr48o?UXTvdW>B`^ zzQ0PXhw^f7KASX=?YrKh+6$8qa#n$z>w^&=`W5HST*3ao1JOr#$6tG%LEWt#sI|_d zPGgSIqPG)KUsQ`12m4^swbQi3+8$*acVXDLU|dk@#+Ic!1*co|d5O8`6vG18?K7jj z*J99nqv%ewoFbj-b`eXJB|cF7hgxn8NAryyh)CGVfn}}Z&Nmo~) zpxYnL$*SevqjqBExI9|a8iNj@4R|*zg{G}b7w7PkG^5x8)3S%t+FWn)ENCP5EIG#g z@93gY!wWa-12AcJ9Q6C%q}S7|@oH@({>})YsPhTvJnAWKwzeCkLmF?37}a)-Xh(9;9N z_hXDj>ylA+v_0AlHe*ic!D+7Bd4^cWlp6M-t@aC;jJM;S{}%8dhg@087bE-;TIJyM z$5^ps7J3IBmj2$JgDK&iP~=idK84z}+b@O6|82rGZ=vJ<+{Ygtcw?b;5P2J>qj*O) z_NyCFK=D6HPS3={Q9b36e;;$y%=hR!KZD(kUc=F+VPuX7^buVJ*|qBoB8J|*q& zn+(-;@faj}8s}x|^0W;)=zU=cuKiq%2RklPQ<&g$mu*DAgdTK%m=P_T+?J0@uQ~YT z5wgAh6K`G~hl1Z*I#&J}QzjkZhCxrH9Tv}pCK$;Pdv8`890Cv)WNt#tj{> z;bz7%{E?aR!vH&$WPNCo)h;UMwRGJe9KGW{LiwW5s$z}g0|GwFpSG!%7uUAX^<%eD z8QO}mon&|)@(hiZE1&$H-@YO-|u ztO0@^=|W58I^sU+u>Ff8=-FBHh}dV+tsyU^HZ|7Jh&+oPF-KuJ`YVPd-h*9z2nM!j z^TvO+tUm28R1Qq%@g^SJAxn=FMp@xzg9DaMR)JRHeZHZrD!)>%E00>E3D4X8(6se1 zz62g2?NzHO>&tF&pD@Nb*C$leT24MzavHnUPg)WFQubv{6Fr?(Lwo)R@83lqGD|y2 zj`M#>>%SbLRQ1Ocy)%Y_#Snd)&ll>kJD!XNRMF8h7ifUrPuh2OBu%-ggq~HEc>K%{ z`)3;?@%d`R{4J)Bolnu#UIj!hB3Jq!LuOTpce$#_3A+UM)LdLm8ikN;!uNe507`4` zN^4WQP`f+xNxrBTuJ>+0yPr=GIO8ZH+T4Jqb`2u;E<*OGLzvcfIKBwIDt}Q3c#1vJ zM@2BuI|=G;_Vd}QRD&Zf5Y*3VhXjpU&)!ZV%_f( z$j97O@l#|smCVl2fVHDw^mj6J{a2$}<)dU&GnO8;WRP9Yb*zxFk^9b{!T}>ji8;tt zGRV@Ur+b9IWc5l^-2Orr=KJ7^=qxeaHX3K+3h7K=EqLA?g6N;eskQMP^}Fc;&((#f z)K!50k}q_JyYsZx0epYTT?Aa{NGIM!QT^6+R9&^0e2=dp^}g+BiIxIBl-ocfB^{kk zP2jiwz4`FfL+FuINr@kIvE`UD);2ogX?hQYoHW7Z`KPJjc0c;(wVe(}dvjghVl?a3 z(wQT>pd!wrBdyod;{!GbJ}NxNFDD`JpC$R3IZKYO`#9Vl&KeroHsb@>tm$o8Y)f$_(9EG9U72J4mLu6l`h`r+pf19`z9V{Qf(cn2cxs_75 zvNskNik^ZORboFMPW$f;mlS1t1;<-~nk@QJ^GpTo>}`UrPuC#UcRtE4KBs;^%xJN( z51bEeMdund(cAY~rh0jwG{A2U{pl`d^ZhsRswwltOedeFd?+Aa6>D_uo=HQy4kJ6a zarD|>iTmlTt1P48Xn7wF6`v+p$QzN^PQXq zY=WoiHGI>!Ci|%NN4i#M0f&-%RM1s}DEOqzs#@H4M_s2U*iOn1pWsBlD)eyGr^&*1 zl7BY_X`McZZVq#}YPvwrBa>F%Pm%5X+C|!<*dAeX+^Ka%vNT|9I!fjbMPu@Qk#XvY zf)na+SOw_+AK+NqC z96Lx~zV_HB9&g=+Z{O5l>#h6oJ1z;biWKbo*bzH)i}-Qn7vx}@sJ27wmK+dw-K}X$ub2#zNZ-@|AZmz zM+oI~>dh@rw!*dT3^F;cPT}uv(uTS!yl?p;@);X3@|GVJzYOAnGH>4OyA+`rjLAp_6RYO4nEJSnr(HGzAZN?f(TABUd=U?)YPM`#HR-PfP@ z-q?J>0LrA1Gl7zqADExmvK9}PKSFEV`ASSfy=Uaj5v!OcSC#sdK(ViDIm4e53c z?7eyz?@zMh#7Gr7)7%AL*T%wBcAmCuIxltC8bVuk#PgghnLNrvi7gM##1efojNCmC z^0{4*vs{ew_pHD}E#W^Nx)iUnjw7!A zG`hO%r=_cg)0rwu8a?v?x_>jIHHpJzO`-#(()b94-ByA{$uv6NyGl0kjiF@tCIO31 z^`e?tyJSwWgZY5TC{A6z1DnkW#fy^Rc`BUuf1HD*<;wgm(w?&~iG3p`giSh|$$xs9 z$g}g5ad=BC+XqkNnO4bY(`gcJHYbw3=mk_vcNJ@ITRb|kp9jUAYaZ=WnJjmrF>=egbkfnNrr4RF?+#-eOgr1i%%U@7F%KR|XSSQO zr%_*d;u#D4xUmCuu0L`4TpRWZaNt*e&qF`t9A}pA=A3v%_Da{n91oEJ`|k%`8|nZn z+pZ{3)1^S2Dt_tMPd@p?O8G~{WT6EM*5q0R{%W2f@=!MDYW_#koz@-Cw$I@g{RRl$ z-2l96(cr1oe*8N3ITUYYNT+TJwr=HNGV4DN+O}2bY8b{k=H2DFFs82NDJdNDQ@zrA$!P4WWhIvFX7Y-4ZLVI#@P11=x<0f>JAE4RIoDFu2$zz z9|i25wp}_ic`iJDrl7NQl{&co!a8+VR-Douk>Ld}H`U;|-PTjw4>jI#A`^WV_rwOr z19bTCW1*?uM~cmMSzp;99L`zFxBg3};!*k3HRB5nD%m34DypET_pLDGw;^R%Yw(K; zu^fFN0*e~L&^<;0mR=!u5`1sJ%{kNxuJvs$1fzjF&k zZqX#PTVlwAW`*+4cmHtdyE=b$9>Q;hX8C7T4E|ayqB~1BAm2L+VfnA=+nv7TS@T_z znq;V0JdE2cZo;rbZmjY-7SHa#;yBSYeBr|c7|nl4sed*3PMhbLwMn4C!X~n7+H(xK zp(#2E7Gj6X5lX9`jtuK36qGm(?<*(s_1%eB{-c@l=eChYUz@{!?7ky<<{<9Uu|0>^ zMdH8{J)v_xM6Kf(ZeGD;F!qMj6q$I6-n6NL zOYf_q?`1hoq%4Ac#{g<;-7FdOp1F;{iZm=sOTxKiX8Ad zGGtR)$=k$?HDbahR7NzwEYwZz^G=sPUpb7%KXJUwE13_uTVm?ZJUrc-!tJy=b6#Fw zR&P?pGVkHCO80FX_`sDvwn-&-xfQxCZp%Z~wdJsRv7{ey5N4YWu=|{Ov|MD6$NsD5 zs&3{|?}QL+dXdb-6IWuuV_!ZPnZtQ^HEGYSC-^ful#@p?r>c*H^W^7}orz$V%R0cO zzJWZ{PEyaqLQkm~&iy^!v8QzeTlzWUO^I;c3C#Kwa4dT)D%^y_m3Zikj2V9ii0*?JT)8z3*unIc#1dyXc~e@p(Ka-n@D zlPgYN=P|u@@aNWz(!iFTa5x?>{4`n^xpV;E-gk<}k8_b?Zk0(Jb_?#$-G`VzjOaDVcy-KH9?)$CJ8o`6Uw4_4dx$bWTYQOst`xnPJ|8hittS>P8jowc zOYkIV8uwT44~sYB&~V~9@3F1J@rd=DzN`e{Wg1*_$cmfN6nS+*FSOol%S*qCZ0hhJ zI?(qOCA+<)qft)~w_Jf2JT^v6-Ez)e_lb3HG_!$2InEmX!2%w^)2@r&l#*09%~Qvg zX;bmUXd@Z}>d~iW42D0lV;kSId^vs~k4iZ#jJd-l({1VWy2^-uhI{cV_dL!X+=LH% z4bYm?j2d4Bc2V1ayQ;g$SLY#h8h8o@8-2O4ES*O@_uvH!-MFQBi74%f<&0<*`G%JM za*MW3@*2Nb)COF_v_Y9D9^WAM)jcUUTyrv{x z1>SA5kkV+1I&%veuu_i#{zcQySuK+5J7>&v>H?EV8t5@RiYCU)qg|@yl(Bk(WcyX3 z%77toE$x8I5%$!aHU<%OT3Fxsh^7ZD=i8!BdvwDf4EKxVVIxO!NU_Af#2h4iLIW+{ z=nh|%K77y~m{k~s5eb3lS>OwU)S>vW$P06KT4Cj+-$+S$FV&SR&~C-SH0$e0ym8-+ z=Qr=+)9yX!HEbWeoWPb(zSEiAF?2&PdTJ|Gc}Rs4f6m#8(eFF4Y*25`?iGfcngcNQ zzZJ0en~sCSJ|no{5AD_%fus?!^g^i|vl1Snd7GttZN4e$ok2?sJ7ou2i7$ zkt*7&7Lc3PAi)8cLy5oVkzc0hJ>9=ec!+y(fVm&HM7^Z}+1KdsxNEdbbd8XYRDy9bx-!bQ%oZ}xABWv4S-kmv_ z{R~ui=J?%sq&^=CQzl_XTn?RmH<7+Y3!kdyM!p^tkBgDvSXk(TReA^L>Gt3cvnUe zuFPw|WZ^LzR@atB1dOE%W;$40eh}^fMTow8MsVHN$X%bEaZVXdmE*Qt^G-9 zZVYCbx8TW)=deHenHqyHI%v2(ldXPqj5UN0RmpV-^{5v-hv1pipm~et-nqe>GKk|> z$H6kzT$Je~(slm}H1O;tJf4(>#8VR5>h+L&r4N@EZc>pi6|8W#qy6z@`v96~(-t1x z^q{R#MVmf%7fxE&6n3aTKSo|5p&bo-G8=`M4M}ZG+Pob-){W8m-)Kc+> zrn+`<$m%HgU>5Z>T)9LV|Kq*v?e?c^(x#0(H+B>2pS^=GhYD%!Jsp~pBecQADtvg*~DNMGP-2QzMIo<<@*Pz_uEm@9E}(=bHaYVO?K=*O!OO3R_)dHC6+utKBGR#?b zC^}}w7r)d~Ug9rV(c5BNEKT8Cm7(l4SAj1XpQa%lmyujc6-rhsc;nLV+-OenoXiH! z>D_`jpK#87Z^{XRoi_G%7;V|tA5P+4uD$XnU24-A+cXNvp#MR3?3Bdi%5_Nk^$+b2 zq(}jMjLFpV5zPLiQ`XV;vi$bLc-_4`e$CE;(_+jcofL()t_8gh_9P|oY<_ajlN~Oq zr=3;CeCwmcZWR}~eNiJ+7P^v+hwz$UCt7!juC3TK^j*1!Mt{_iEvv5LsGb+N-J0&a zWXF0s6Z4nKyPv1qg7q_O`9gy2`!d8N352s3J=JkAQ)){l!du3?L!H*35IJJBBR{wR}rk_Nmc3HDdgIJnh}qljT!Ft=-!l=kk4!}8I4VO~~+DTU4=!*CEv zW2Yi-m6&C?bVa{;n{i|Je<<5QJiu+4&=>_PJ5iG#-CM@}A8rC`RBfmj*bNNPTU!#=ATt5SBsx@r{myb75719kFepPhXx;=doditZm<3xYuFfLPyL7(nA z>|j*HujRqisptnCd2NaXclJ}CU&+$GY1gDVOHZQu%?|vMTsXCCKYbb2l~-xTbJ+F{ zR6i;PLwdCE&c8b7C%jELw1iiG7teJ~TLfBur5W41;=)WNZtx3b^N-H#H|>C6ObNbM zV-ohR7rl~y-eN+QAurmI!B*A^@@vzNk#*N5>i+F8&Zmr^mNW6PMsMG$?I0XWO)md}E#9H_nZu>`tZ}_|btM9NmtQik&z}|0pl5i-ohkB0s-i%=1e7 z(}7;LRA^*M$D^jA;>=EbN!|{uGvK#tHzBL{U<~oD!Z>lS+B81_8S9QpYo;o4+tXY5 z!KJUVardgEJFSUyM%9Xc7RAWVPW&m)9+E9(F8@K(L}7hcU{kQRaBt$iEg%$ z@KgB=qI%5}_dVnPadaO3SibKYC%Z(lDGHGW;<>NmylF{Pq@f}0G*qHA6@{`lNhPI7 zD1?NzhO|D?pdAfSRN6`OyS~4_z<52+ecji29>@EQ$FtUvpN|e%8o$O4r@5pX_EgY` ze8d8aED`A`kG*k6SqemUXvH*o-)9QmJkH}4nkbs!wTNEw^L@l!7Yw@Oj986SjFyYW zp(93!wdLH@O-?xWcRmtr&tb^YGN8nRdUO8XsQ0sJ@#0zZ+kAl7_+A~OG_~=z&tRIT zmI^VO=gHO^)AC;y^k|Z*_(k`%IJNT}dy+0e)Av`fuRJYu435Lt)D}>u2YQZ=XB+uB z(;3s7tzY&`IA=dmlD zlz2b%n0B^g%6b_Xq$eOO*bje8oKdZ#gaKv`Sl7B)R1`5=T=K<)hN|b1;>bLjTv^Tn zC+rjYA9&9+4YMU4fqMl+o@Og!8(3O&9GWyVsO`rJ;pS2=wyOnb9{mM_U{q=W*S)V4sZt(}$>$`DI-YW|J=PO=nct?NA)cDVLqTq?% zw6t5Fuk~*+wF`Hc-Kk;F2>OH-c@ybVl$S((bfDm2F_ATD9>l-2AE;Y!h7>+cWoADt zSk49wAv5F;i%k!ur1^uza^A1oc47`|TfbZK%h-l?s@s!a`4?!e$e}K`5mc_NgLxmc zsb+Bn)OP6c?rJzG_cx)$icFT?T!3j07Kl9dJh(n`_IUBo6(#X~zVCV$=PCgGO{@IV1nMybBq*Kp= z_1JgB0vBbLv!{cH;%2X%$oeCXr~?b2wyOi%+{RNtSvVb0)+1wm0sg6W+y$IXmg6T8 z8@Pcz+^&dsF~hLwY8Giu4 z(hv@XyRi|omoQ;KPt+_}L3)1~6&qbfPD&MI|G?bB*hj4xmdA}^p%YYczJ zm*RETEVkvVqVRj|9pQAb3Y%tO%W~!Rpx?t0bfS4EjXY!n&HNJ7cFNM`u-i~ERU!ZQ zaj-pnTrhg9Oa0};aQ4G6p}&JP25y%EZD)^Pbdn@ z4~`+50feF(iu7zcno9W=8)i) zNuvk$rdVTnI@@v-3)`fa+6ZIGff^&pZ@z2uYDpHX#tags=H265gbjEx?*09Kg`!c>WZf`>jJ97y;iVs#8m~*2 zT_@4ywqvAPw~hMboWaPCbr`H*PcPFR;ossY95x@0uDt1j-x5=(`t%3a)3H0Dj8sqB zqDQr-^)YW7l2z)!fZKNwEV~xpOP(v$-g#vb*=f0+#eS|(k=TY^`J*4~l z6n;!srytj{u;GIvIaeE#^P*pDU-L0)&)!G(-;SmcAEbCjV;Gjr>4DUDKiDpQf5{%T zfvvnBMib|F(uU{xRQulw*k_-D>`h&oKV>02M$bUE!53js=T*Vv4EH%Dd9zOSp(JOt zkop-7rPaUqen@{Sv-Dgw!rHbN4tlb5v7i{Y(tcB7Q3?4Cz68ZOQ#ximh>bq6h$U_P ziyE1~SQ0ay)}-ZgW@In%Pt9*q|7#*9Jy8+o&#a`NCw)ZS6~5x*n*&5|K9kvcMTs6> z+D?tbWo(B^o{KeFt>V!)QZ!_~JcbH!(yhDBT#-Oma=4aeE&U|G|8}|{-=1qX9;QW`S1PIiC7!C8~evZ!f5UUr2JOI zQinhmtL%oAd#wovCgDN%SIpm&NJ=MmQ{Tf9s=asYMAL=j05_`k{7Zw~lzmU7B>~07f8i%v$Ke}w%wXL+E=pnk4*K=lM9-iBj;*_~E z@7M+-=4J(CCsd)JdN`GD$|iYslZN)WNR2<&QdfvS4d17U?Bo^r8ax?Gciy7`s>|p{ z{VHVm$KmkWPIl9f_uB2dsbqkG`0^r&e%r=C-v17po8!Qis&iMJ=|vjIJBe*GR^n2~ zW~OIo$r5+ZAxG(>r0=wYWN$0tpL?Z{{8d|c&RM(F!{-TG2d_YrxRd_M`*Szo1T3t- zg{c0V55B{g`n$ZuljMt>5vhmoH@K%VXD{NP-es0k6-if%ckY*P7u&+Im}_awRWXi&M9X4zk+gX+@cXY>*&>(E z946gh>D$6s-un5toE|H-H%t~6{(eUd#n*6E<_R1AD-C6D`oTb#KOg^-u!~K~9AU9NL#WfB7!!)mK--exMY$|}SZ&0qR{t@z z1SvLdm^y~J@~-cMS}folzMRw@Xjt)$jR{xdEc~lv*u$GNkKae{s-9FmSQZbv`rv-Z z6ZAdX0t=a?Fmv}7N5!RzbrXk(a+gMu<9{HRHa?#KKSyzT_aECEFN2i<1yI@A%w~pLqVPlu1hqtHT6poCUj@cj#9@=q zAsD+9;?3KOuva&RM|L4K3$1jxPY`un=SM2Qwu-*`bF|n28WY z3A|%HW6~HJ8CZzPwsw%sScoeN+fnsoln`wgz%rJ1BQ(nE1G_wc2@$M2kI zI5k>q%HGw&N++yZ3(n4#SC{t{M-;~-e1Z3M(_-vC(MPq&db_{9xuK`WGyxVdfKcM(SG%ot`4o`v^^eQ}H8Z(9> zO~%m79~aP`aD>d|zEE(R7xh=Mgx60U*k6l;dB;HNV>E#ZHO?dZUvJK|Q-N2uJ}!j1 zQnh0%ai=h`C|OCbuh*E$q*i$C&1E6((HJn}H59*^vmSP3IB{S-NpvJQH0>!`{Zgo( zsyy#W-)FHyeemt6Dl>Z$#v+eSrf<&M1)0DnjQH+N*7vV)_o5uGnQuV(LI*rw@f~`H zb?EE+Jv=vc8^OxYapu})W?sr~^%dNWoM;Cn!$CA7j)5HXUr>3|IdlkxbEbJzM70$u8`egRVu=RQpR-nA&GM8>_eis~>5x zi5K>>m#**7D@a)|l={eK_1e#B6bo3Qsv_kDInn*|w}rPZeVBd=XLr&CHv69jEgYv! zgD4b@2cE-kts@DI`bgYLHs)fxqnr7PWK^r+n!b(=|LkT1L@#qWAr|M0yCmk zVpi8&d`ygjW4sk@D7(p~rF_Srecv#uhZ3FayOP>2J)^10(!zJSw@mI|CC(UXO1d|e zuu0n9A%y8K@o; zUiGR$ONk@pAG!-mk1u?dp+aNt@OP1XB^a4K#X|}2JNG|8=aEj+WOB&XMppcEI;UU3v{Hk?jvNnxCg(SQw~6JU|=3fJ^>r29X|$n(?iXLEmS`dth^m3HB@ zOb5$XZp5gY;V|7kfbYFHi@%YPD{f{N@q07^DcL8dKo%uqPeZb(VFFU?@?y z#k;dVsQp-!7(W`KLQ*bOn--I`!+fd=@MqCWJ>eiNPuAiB{u8JPo{`n!>#1zuc!!X>DceVW2AZz*vIGI*r`bQ_0zHB5Cs`g=X^&DJl zQ>C{)S!k|OL-VJdn9aKc!TyD~_%W39x9U-O4<)vF{dzp<3Z}`ZvM|hJoi$?NZk2T-bX1-{^<2yZYbn7hKH!fpK zJGI%Rxrdp2V;M^@48bJMgdJA+5!K29S>z28S7?nD-`$c14%Fhu6r%jrR2EQehG5PC zO*ZxC^J_(#*-wk4fA7bDH|KGDk{0Q&iACgUeL5=t443Tpa0Yce@H7Nzdgfc?DBIG=V3 z%cPFsMA=?kYOrJBCHt6K;Wf6C^T_7@No4s>rkG^X8+V?$2qlsI*~E)Is7)&l?c;Uu zdQ*qw`6o}7=2eGie&-aHhee;;!S|-uxd2cY`#TELaV?}O8ChX5tO^n{2hN(+)5&1TfJ$SvB?`qR9apqPs z%L~IHzFWEYQXQiQ_|O}}7*f2RNaa22`DZ8*{l9dP%?TO$?394U2r@=h3F`;M->33L>dbTRIGi%CQbe*=8W>UGE5y|ph!kLea!t9JvxNnK2dma05 z)9gR&3^d00!}@f}Fo#)hcBZ+-Dzv&yo{iqI7tNKf)a#cneY4DkLS6t3jn<*4&RZ0) z;4qzRy^mddZm+nu0NV;bpnVG8qr73vcgjqBeUrg6@Uv;3R|0-+Ux#lJ3C#1KVsv2$ zN%2{-v{^JHOLk&RT{9b$x(HKH>*A6V4Hv&iG-C8ja()qZ%sqo}40v?oFZY&PJwW z7)*ih_fX`CBF^1RAi5Y0wZ?gLQCcFJk1!TXB0eFTf5+6`rSsp%g`#hIk=n~JYUsA1 z+PXtn%Rj#?DWB!P+=&e-gRolhIFuaKg^t@B*j9-m<IUcp2SS%3hTFxq<`<<(c}E1G$@w$p^6N#t#&P)=qSbBh~;#{+zvMmgwZ$G zf3)W8cs!3i##!Yz*l*fS_sY4CeP$F4=B=eseqm%%moIuGI#H8i33Zx|qFDxZRI*MM z<3k@IasE#De!q>n8}HDttq%p2jiFor6zE20EnG6@B4$x4oD7xez=JgE^L7ue*bfjB zl~T!3Xh%SiD!lz=S%PwtFvn{GzmF-32iE(Oo^K{;7=4D7WFyvA<rKeeZ~Q`{>>7R7nyzGF-= zsV1ky}=B)qyG_>oDLHFh?q4SF(8Qcy= z^1~G9E}l(lZSzp^=o#y39mi^Gc>XQ96VET+VIEeFBvGuT@q6~Mdp48VYR*`xzcBzh z2S*|#&Iv;AcS3kSZQ8Kp9N!~XQEW#avBG;7DZ2x%k$-UJx-D&ft;gv?fnDA-bys>{Z3?FUnm&Pz9U%vZKf*2B~N^|N$f!85uq&98B!CLv%PoUqU>-44J~!WNe6kGn(rZk z$E-oj)%h5^Irty*=))N{wrPBZq)8XHT*uiU&c0DDL(`=!3hdOT`S(k(?q@vB++c%| zeF7yF180*1=Kv=SlEsVUTE4?jkleXtEaL}aSQEKc*)w$!q9Rp8f7Z!2Yf%$oUbzXbcww}l1qj#zebm^k0+ zl=!%MtvHSQi*EG&gr5J6p|x*i#r%g&H1Ogu>Um6^j5M}U=GZlqz&m|b=Z1<&=FjM+ z^I&#cVn(BD4pMr79l9r^O0stRu^AJ70RPVE(;w;Q2>i&|T1HbTDf756jB~%sYtPW? zk$-8!Wd-VL9gJ=3%p||9+Ju`gkJ}^&zU-~;Y2j#46L(BXiEHP}QQ_&y7<|VQf8W2y zR_7n^QyxKQ{>`Vq5m{8~KbMX_Fk$-)reOQm6iobnm1n)9g$Wg^!f;PVa+SNnG}UeC zp6~;413bvAOhPNSJEPfMnvH9^AtXA#?LI&aXU&17*AE)bLMKrfSX|6?-tU zV<3y$63C9~4yMADJo7DWB(_f+DE90XMh}0qknN#ml=k_#Fq6-e`ssQ?PuH5RZ&9Vm z{Cw$I_>qEaZE3+nE3!KlMWWqLQom^=u81ik*$R7%;Xe7D9>37D-*!>=ptV?cW0N?m z@jLvl=CBv~5}X>5fq))7CE?l@%tHAay@}b4LHIcbgJ@fVWps#njT z&oOp9XYY$Anu@r*@}?y9*I2>PISJO>L9qS8Q=C1th2m?&DNlAe6~9~wgMYt})^8A{ z#r46~vle8nl}UbfbK$<@1lG9pqR}O9QED=P>UCtP=aeP1C-xc6<$hv6vVSr!y+de- zREHAZk%`qx;=?}!M9pEV(LPC4sxm|Zq zuu>BX(a4!>+|`NQT*@P1$^}iqso|`qVLzelz%@QNn@3$*4db*E!+wVxub0A3VUJT%c=OJ zb`r|I+#MKOgpR@loLBUwfEL^C7$`ftv3^Tz~bNSeS54w3z*k zq`B*5?6!q42y}*4Vjuc%8s{jz$V2UK-u;m`M{C#~c=PU~%d$8@t2z?HCp#ly`VDl> z=KCqX9+bfEa8tGVLABM3(jSL&KJ*0`{@4WL4YC-J^OstD*NMlH^J&t&gEZLF6D(Fn zi0XcVqh4P0{=x+64wysdLdTQFaZS-IVJe+Iv;(<2zp@~OJJ_{2gkCPpqffmW$$L~H zUB7#Y7M_fUdPOWk&c+gUS(EBwEqa_cn>(n2QFA>9&FatDTUU8_ExyTQ<-8CyN}aY? z{2(T6UF`pCtoUQy2O54rnVdVm;>wq4sH$p$ZEy&kwKxkKu?;@mJw-hy8L?&I z5Gp%k$z8`V49VzUOIeOF#g>qOr(~nZbd-tK3#63eMO(4~*d(4*aVPmVL zN#U_M-JF)o9FJMRc+^?U%`Cv%Ek*qPvINKGY$cs9jTAKL0LhIWfk=CATH-R9-pBQ! zpS=5?@HGTOCdJ{s*(p5l9!Qt^9j6CjM`(*r8nLpGWL^6Q>t7_0>Y!DyRLRGsmJ-~m z*AtiCFc*h}XozY13b8OdkaNrj!foVbHhAb0c3EjOR`{(zzxpDah;N70*gmW)egG!c z$z$!8HLUOLKdf$-4~`7?N5-WT%;DLA@^p1}$XK2E^csuyH|N>x*h=^=$wu&M2kfpt z0)>J_n9Z7mm6lua$$KoV8S?^u57p>hZ{E3$74bN72vWqMn5T9a^8z)gr68PwWW71( z*o+qV_M_$lp-8k;gJFmv+*_Zq+6_{;H9JCRUE7Ji8|IVEs@*i>XBq|?gdw3h5_*?b z!A83jvZ5jS-LJ&w9T#AJ_cxm~z7jrLUCBtjANLR@GW*+{o%5>(j{AynZl5I??wLRz zCQikH;_ay1Hx!09chh*4-n8=p@3=~Zz%}m;QwTC=8S5FJiT!13ZY{>|;T^2~-wD>a z>K@bm8%JUDkJBmH^E9+k3T=0buzt4!XAjF!TIB{R8oL5A7TZzfb`C$bwPKp>MLx^5 zB(*`ic>gn)4VoB@9lL)cDEJ$zdf$R%-XEXVrvX+UCsEmywkI}%p_}2KYbxdf} z86>PQ4P<32~u>%p$WN>>U85}6dN}o6Lu>` zV|L3=_H~sp@_62=U|=9E%GgTL-@>qI&keTga0t}1ui^0GWawRw#-jg5z-r|lI0xjj z|DK;lUWOy(-pZrrjuIM^nh#-ZH?HnX!eC|t_dZ34yEvO_)LYq;oy)LCTLtel?qk4& zKIHu1I25?o%zXbnQax@-pAQ{DZ$C%0oj=OHS~#$cH~PUZ?<&Uq;-7`v1I55459qm( z8ofL@jP`7+!=ffnVPS3PvpMUo7e@$6ECRs}PXmSuLz zJF(q<4SuQi!o7B60-2sI z5von}AU~PU)NH&Y0efp~=BZ_}`*pn~lT)*Vze`pibjb)|w{JIF8-ED5uK5XOrmJw2 zv(sKb8Yfh1y<$iIRl{0gjg8kP8+_Z^mnH-f)6r4H?mu0eF?PT@zQKU`Bz=I+Bpr6= z?r-D`cA+_0%UDx#C)4NuS5w6(VM(wndpJ@bDKp9>GLJShtB9N2&0Aq@`7KBAD!0bF zqu!E>VY`^yu3IqHH>hMk(DR9A`xRn^h_NG4 z+qjjEyscnI)e}%Xpf?4k>}Th!_hapZV3If}vYEyHcr{XvcEm)mqu-6uJUtDImG-fp zUNYP*v4lE$WeW|Div{n8VQiX=if|~}06pTBDX`R-e}6B*_oqO^Y}6zrYaF<5A_DFH zFRd$2JQPM;_==LjaZJaulWlpJ2^XJL*yoza9?88ChLrR3X5tFUExJu7a?I&NfFOE) zu@pD%yCn(snho_xMe>|J8Usx~V#|#h+}^l`3A`^_d%Y0tRZrQ7+)-3}KboGMu%exd zp3{O~Rg~ssYV-ZzQ6VB}GU{~lB_*e91zqD?EVkl|@Lpl2VDWIbU|7DF_hZ7a-~R{H zdVUuEIn}bU!fdSK{my>vSI{$Ci5`uV6^vBAv)0blST$)6>A3F}tA{3t@+H@VZNUf8 zH7kRnd+Zh(e|JjS6aF%-TinroP?-XZ1L^zz!I0|F4!2$QwBe6@)y~GeW>GRmfiKW*WoyutuZ5+@CuKHc@LJ^-2+r9?>>;?e!%SzPMq> z=|o`*XBv#?IEaPca@g^cZA_thFY~K1V0$*k;`mZSL@XUi9jp6@Phy9O{!ynT3kv?Q z?RwwvDP)Au*b%|}Zk-cSdxPdr3eR}MoEE$_5!*;A0!RMpN>{$G5 zOuHOKsq0<|@|N+!&$la>WWqRRch-?Iw|=1xnOT&ywgWqR*|6(dq-f~CSrqhs9BIB& zrZTqfX;9^Bd76a-ly#wn2FxBL z=I$Mdje|I|v*ZXKC~QGq)>`~BGKFl^AksOkO$vKH;Xy+Yp2YR%87T)LUoIF?8}?z< za^Cg!8B2E^`cS%+GuxtK%T8;{pkm}MbR9N8^e{#IRCdLbuqm(`CSYmERGht+A^cld z0Mgk;O9vc+(i%T#56;F*J`?=)^&r-A2K0rwzu87DM%iKp-Q5w2aBo?15POs4svcCd zyfBPAVRx_3V+wArHB~<=&8x@ToKyxJHsrh3Z zg0f!V%Kj*1&x>JWp0}_mIoi0C6b^L{6-+++62;!lP}_2sC8zQC?s47hk$V!Rk8MMJ zRXlWCvM_1AHk%mvQc(ZYkF%-Qv%yAYEOSc~YWWTVlsA2RD>}#gxFEa)8lSBFU zJ(1MnGuZ05T0u!{g7Q&w`rP{f7KLcg%?B2=e(e|Rk5-`nCQaomFF~09?<2~0TjSxZ zeUcXCix5oC!%AT~cBGYJUA`L%`Ht6VwG;k?9fA%2Tlx>{6kNxZNe$yi5>l@{@IA?urM&zg2)3he>%aF{eDx2da$fq-b_=#7))^`V3GCqZzJkpr zd3b1OVZ3b_1jQqmwo9Nhwo@n~?K_*|y^xLgV2G~e=LSAba?o2dfIg5FA#l{oC zl@`ofbsrUK%Zk|{<3%6-x2N5-q_Ll#<3^{0pnuj`2+^Dg6{Bd3Sih0)mG!9B{uatl z+EBqF3zBO-iyYZLwAoCCdJew9T&9}A@xxvG_~^$L?LN#j@;%sYp$Gl#(}-6?CFpgq zi1Ms&(VH7@$+!Fg&uR=Ib-O>=H|o;#tFf=h%;#;dpSc57cKwaPOielCRA}sgoR2=yyVpdRK<}$h++9g~=Gp zd*;ubUo!Qd(zMclG>w0{Q0US3EHleM$H|3hvJ-!s5c!`DeY>|?h@Iz&j{D_^t1n>dZ-ya!(Isf_ z8%f?CKVYPG31(U4Y+Kwi+P|!j=38n|;bB|xrGbMudbigA z@GkD*99H`+6(PU>VdS?)_$m)2^X*58g+|f-SD&aYRbG5L%R@LjA%@*v`y0QGj}vx( zs$`{#f7p)SRcy|GhU}n@jsBvdsT~eeYRr>`!y-hbIF8!QnRVM{-}`ae~}Fty%4*;EMVqE{Q2$QglX4OXl~~P zd@TFL3SULzgoQa>Gki@wKc1Id-_$G=)}4aq&2m9)7U%W{&U1Os$b*p{p%U(ODw~&9-7PA7^8j$0BmeUnT67Ol1=qcH()KJb7eTh_~G< zM1!s=RO{vQ)Ov@!+lnHBpDNI+M#t?2H9y@xM*hw+tuZ4 zYi^6+GdrD0Jyyc$jXfwioBIKdm2=(o7d4@j+3t5Oc8snjVRCQ%8hRTssUngM7)yG#2?Z6I)~Y zQe~v2u!}R!U;3J%Dklz!BuAkSQ|Uqca5j1QeYTGOF8(~f^UeJbDU~-;BGW^*rXH_y z)XevG}5LVlU)?9K9YwjoKM^}AmxxnF;Y+35VoblmT<@{_mO`>a?@*XQ%&dd{!2 z&cdG&+@UkohDJ432>J#(tZ&RQ{FO=(KI(?BqRG`5&wZTp+)7xX2dFg98XbJ6YM+x! zK1pdpq4F6vQ~bylR&)w^%QxG&EpKJLyFSa;KIrg`3qKIM2b|GBr>9`zum-`!ux`DnxLxF#`I?G*_5qmH=nBAS>q z7rL1OjY=Z&d-qUc`79mRJVVJ~@l*CFdm`VJOre$hy~fyO4_#h=oSf=63-Z#NFnWy| zO%Q@fJk9-lDxD}hSuC`EFhYpi6k75snHpQ}QI8Bgv7xC!IIE<@20EQ)7vo<`9yZPr znwRW_@-vVy~eBdJDZH<{Y+#jU=}MW9vwch=w+qJ#9a;>=;!)rRJ-6AseE2Ym&PBVvohhVs?S;e zZr2XoZFks%W-F?@_!lmIgM}8$VXVqXbEqH@%Hbo3u zr{QP&NTFF@8h%4sDYf_j-ujd?_c9e|FLY<`tWL3X15+07a#hlOUlyl_-JprCk7-pO z1@ZFnWlZZ)9v*HE!{9A8xVlh{9<4cn5~hs(Z_CJj))>)Tg(Q9TGnw)1fn-m~82-Kj z1y2}iR4qyRT&kGbqHk*)e** z*Jp$)JuH~G$y_jM;rVj+^)!Cjd2-@e^uQ}`(BxmlZVyht_n8Vf9XxkD8auSor~6~qCjPLbABZP8u5mtdT`6H#`G zbZG4v$-CGDHh$DmOqPmZ8qKM&TQ!h|?dVI*KWgw<>L|W0nutYfWog!$zof@|&z{aw z*sf&G5-(M=4}X&=VDk@ZTfLL6`*F^*@ed|BGad4;-@#Wu3X5KGKH#8Kl%6W)9K$E5 z3z|nM!>lP*tboZoYieW~+;g`FMRxlk^7q6EmOXJM?IC}U^B{Be5_aZZEAyRi!mckr z#N?*fGM^0a8BP$syB)`->krUx_;+@uxkI?Dc2rn*sW09yjA7k-d$3REYjAm13YH(x zMDl}lHlKH!_NE0R*dhugd?zIEOlighAGWOP3H#=+iM}qoSg!JD7%kJr;3xz9RlI=0 zM{giCPk@wD95R|rQL|1Pf9Fl07k?+xl~F7CTt5|OX9VDYmos+6-DZJiT5Pdg084ql zm-Xy^F9a=mh>hnJDIg^Szx2`st!=v{xxS{tPis%M@{cAnm6|GeJZWG9SG2K$)+w0n z6N=(9bNM^-3>39(#^}vfkU#N@o!Hlf4O}W;_+k^idixf4yNXcLz8GJ>zGtUb+OoO# zMP{?7Q>Y$P$aWM;a9)0%U~|2eome2UUK=u)Aj)C>!}r30frIH|E~8o9rLgKvf`vh< zO{W*{G)Xh&;BZP(F>!`u!W0*rh`5D6mBlzaLY{_ngksX?N@jA$lLlb+nYh~iXBp5Ix7gewanv&sZFhLnKG^rhhe z^+<1wM^B}DtidD+K0cASDqqD?zdn-4RGpWU&Gr{!_C05{9-IYkF%S-=BjDd~pN(+d z%9c#n#s2OuWUWTY?DD`dFg$$>cl9=6i0T4Z{}Q47>o!w3_=UObIwf3aAI`oQHKBQv zH5`tAXO0^Z+4?OP*{1gM=>PgJTNj{&nW8&n_}#CgSe6x>Z)UgrhLEY!K=Rdifd-z( z@M@WY_U(xnc;G60Hhsp)j%(Z{mVHe;DcpJ%%KOtd zagi-zb*K3}Dfc|{3sr~wpPP?glNS_C`~T zP|_96FBC#~Yc56#9_!HF?PG0R`XGBzFVCNgCBC}Ysp`x^ z^4E^1@y8sgp+Wt+OeBAw;}(nBNZJw2kRm7;sS#N zv88k}DmQ&YOLHqKdOP6U>HbLk^@=&ptVQG1gA`G8?1a`kJVShW{DWzQ(gw$xL$16~(UnkyQ;C%27Pb!&d zLyvzQp{p+jlgGv_Ld0r&c*Hk><6{ItnxFwZqkvm^h$ZWQ@Jm+T);r5+a+(Vn?? z(8kY!Fh@7;6b+*t(xI$lnJ9eBtrD8;_}LV57wfLtz?*;9Qwn{dt-c3KGtH^))ED+( zOb@1xW}5Qk8g2QUL)TmCk-E)+)Q(rfv|5U6>z+|TRXk}sNYL|+EpBL(2!RFhOvT0u z<4?vQ_0o6>IhKqsW(!!AngUgvi$}InDgxG;l4orQh0oU|bBn{mAgx;EURcIE-m6S4JRwQ+E6VV#1k$RE#%SxK{y8*hB?y(`d=7RCNn=JTCsif<_(?S*uS?<$kSW&f< z-0SqIWd2RTAYiR9RaBxkya#G#VnBgUV}t>n25kFsA0(^w5jU?1N80ZeRwq%wrJZtY zZaagO!E&7YdW_^p&KFmW2o>MUfgQA}WP??eXy?{AvOIH|6gS3E&T$QR z-EpCh*RE3jmn3vvFeHuWJv2C68K-2pmpoR9YbyIuQThS)JEKjKah>Pp65^P;pFcv> zOQ|q^2ikqk3rA%BK<`N?MZQ`^Qrcde%cVqe3L)(CxCH*}P7&rz)q~55^^&mEC}HWX zEOymxDOIH|B)`qO*^BV`Wa)SqG0(=pFKRaX-LpTHIaJdm!I@vWHMj%7Z@Vp zGut7L1gVM)QcC36)08TtSDZy#>=SIYawccn5`NwqW4(57HegE-cIDW!MJER`?*ZF! z&9xHNiG6WRpF2==beWgMY02`;ufp^UJsjC)0HwA9x@@(F)&vfsoO2rVNntfx^z*Ux zDCxWOhj$T1{+q)-pHL=+IWa81Rf@fL?{z)Y{{YgoZLKmhNLATh5Eg3i)KhLd( zv(pq@teP!ukc$;{7Wj$%6#VJKSBb}bM8?XCzx%qHhtJGA$XMu4{xIbEMXZZEmJD!U(~za~<}w#8z8 z%Q?E=S?XN_Tmu1uuiysAxs9EE3_nx38pYZ3(R$u#y(#gyF`+K>9V_ zgsK+|MDWW6G|;aJmrIS=8@(d9NhM&c6Thpb$w5f_#`^Z)Ug1n{8qFO$twM|7+mcPC zq$(z?-9!GN%c#bD9MW^Si_5+Xo25IMr}aX%dvPWisyou!_RDZx;7!%8sVEvUkQq%= zp&qmZqsB@Licjt6#vm2CQs2x9cy_m`Y%=naJB2aZY)NHcJ$hKGFctj>lKxjj4Lx>2 zI$49hR&#!GO(hgYk0H&Z4Yb^r_n{{#Q|q?FD14uSR;!87>=z?BvvHr`a9NqAVF6t# z%*N{_>#+H0Bp#R@#n8)~3x@$s!RRR@gkBv{_ilR$VpR0$Z)WwYgpL4VAALF0r!x>?9japbc64Z7C8Nf?oEnx zXNxA2u9pb5HFa!5$#u%y3YB0-fs+g>M8hi z{S+)5A4*jxXYsQ)16^Bjj-(D2vuxWg?n}24>B85#nSrRS*Y>vH;# zpUd74U3h`@3e!uB@Jc+(IcJIEIfn}{dUpdovM3_GR(;w%%ZawOCc(Nn;{{(mV(B)^ zs;u`=;{J$9(+`#xdsHk%-$x|UA*Xd%FzF+=xn>?VcKsAgB1P_~ku3nny674k=*Nu& z{$D?zT2rQq+yt;laaK1AMYM;sR1#{2x zLSJL}mG+4gWp@{w+lIJ9ZUV0FoIMn&L$^UvU)4;Ln|pr?<1qsVr2I#>r2n z$NtM{Rn-yvyHAMz)6}Kh3oV>;wlb==Ho;ht0PZPsVu;(1MRBb&M(-FuW)kAlH9BK^hi-T?r}Aqpnr8f*?pj?Zi>O@e$u#)XzmrpDJ%6j`Lgk2Af<<&M7v{pDLxr8E$0 z-+aL}(w21MQ2`fZJq+H588^H|f<_2yvft+x8Z<~_ce4PK&e!8kkB?BiTpMSqPG@Z1 z28j0l3D2TZV9KMF_hEKt|1Xlbo1yW{EeGFVFc>0sfQZ%&tUgs8Xo>M z5|vH8Axh^0T$}SA&kL8scdJ4;#%6ghCRwA;`V*-8`87^s>=8TlZ_skcAN|r*p}8ZJ zyC%5>`{bixZRtu7(wYw+7Ingnv>s^h>4B6BQ}F7fWN1}efD`_VhyFkPa67XOj~M2m zYNs4IWM$))Q)R|->rG&6#YI@>@EJzO^f6BUXMDG$9=zRD@UU|#CRHDXTL-Q2C*!X5 z1vGOXe;MP!oi*rX6vPz-+GCn%6yDnP4s&&i(AFpn)$VP8ds9leus(LyZ7hJ`Ev~q< z*%x>JW>tgOlZb3jmdU)+JB}Bj&kSMQvgZ%%Zf``r6B0E4UlmLnoP^a;TX2`h16kg*j@d@q7~Jme%+7|s90yFzk;mm+EUw!pPcq9s;%)y;==@2DzH)sy zHP zFfP*+<{Mpfo$?J7`2Bj`f~m9fIbPWUd(HvIDF)-g3(5GcqYYZ>?6Ab@1A0t8&!y~& zfHS+gAXGRR_J;l8)-61NQjDXf!a9*ZW7eRJ)fJ5T5{fm!$(X*N2-8Mr3C@`QGoIAN zV;<|8ypl9W!>};skX%A4|2@OWjQv`!e1)Ej@4&f5di=k@QW{+S1;?caalf81Pf8oh z(9Qcw2gH*3pUUC9RVv}OXK`3Rq)hL0T0t^7nH$kl!~L)rgJD)>j4@lt6|)^gcTa{u zZT?iy`Qe6Y%?CIOc6T&Cya1OKjpmvwQaI61v4sIp>@#~oO13nq?gTv<*VWa zLP~Zp_O>YYOws4!0(apGNmY6<;UxR}AHyv^QSj_H%k}ozLPF-f)mQhLBQCSvfcqC*~4;dnx&jzjXB=U4#pWFPcg1c3P-YBqjHcGRZX)Z z*&^nF8JUjD#cvDJ*XzQ*zDc~7-D8q$3Z=_+vsgEo&4uFfAy~+dYI9Rac-cd|=cR!~ zQVXFx`wVyB$VpsLVt~otM&ge>c9n=5{04>MPvhcYR7)Y=_@vL=h^!Kx;F+|~y# zmPEkRr9wnjUSwSFgXVOZbN1^LI5xS{d$|O#J@kyL_Und%_s3zRRv6x!+6bV|LS3D!UxDo%!N>U8GZ+2Q#@am3tL*l=yy>ny<>Zbs$af1 zaL7e4IW!YYu3KVRtsR^xzQoPs`|)YkXmV^;q1(rVpq%8)EI+-jT5z6ohMOtp8IR>8JgXkS7(g2Fip_(1J+Y-U4u2Xf=AL-};)06z zLu2ng$nt!QU#{y=9LrgWYAzJ~w{Hfdt60-nwQ&6S@F`sD>*OwvjN?Y$6hkLtQJTGG z6!)`uC7k~JkK4U!9hdm27#HcZ;qi_AxTrZ8P1MYJ4TVko{N@?F;;Ah#IjEP-F5lp} zk}B%!siC8826Rws38mzBlZt;CT{tm?2A@43U(Mrm`z>=(c`QSRns$gyxQk;Ab-2K_ zD_~u+1~%_lhADYd(7Q1o7gr+JlU*)2a59Px+ecEoUM#}HINEgbJa+rGa-RFY7>&)a z2Z8iB+V@NjZw#j}&z~@VI*>WJqO8c$W;E_Ru0_0R90V#!qQ36{%C~)(Q`a?GnuK4&&$o^J0#*#%n7qD9rXP)M>iGmIMQwUUP$;lbb2A&RPi+*t0GJB zgx6f0^@ef1VjsfSW(QP$caI7d8`7{}0)6mb29G9O!pUz|(t-dd{Fpxry;ppJmWOj` z+5KvK-YzbFl^3c+_??kIcz42dPlP^rv1OgeZG z^%BNXT9qgEbw`kcaVsv+xrlw4S?IHd$g#Kv@?MBgwu2(++3LcW1&LhPyiNG{08DNi#D|CG;+r`s=pZr;YrLnUtm8>+4eG)-GYas2_bj?_sSy|3 zp2I%Y3oqE^fU<5mD8IE39rm8Ufoa8f^7>>n+jYMYli1ez^yXeF*8I_>-YM3&s^kTJe!hgZX=`$0%{1{K+{MxCNve}WACXO6i45JriOrhv4$I=xz3iAGs6=y7Em*0G*_r^OR; z7!t){_E~Oaz2AR~ZzQ^3o4KPW(cU$uDZ*BQ#wmV7ovv)WQ4)_9Yc(l7W<9>scEbCW zH*uPQEFN0C3y@X3`oR!k(Ez5?E<@MgKM_8iKlGx3?y)?9`q$eS_jV9> zb%o%rPn9UI*a$&(w{Ucb5*lnGTy%Ufag7fUGiEYYye#Qo9ZS>Rn1XFX5k}6^KB^BV~R;U&W}2){^Cl*D)?)l2&d-3NMkrvK@5AvOxVmSMJj;S1!Fz zohmhrXlKPa7+4UG5r49vHqQ<`M)ZJCmJjkzBWaw>RrvPb6ndI3NhgfhuGEY0Um}OC zTW#1bqYEWw*Wnhi7znX7pwG-tSLtC%N0UF{NRwy`zQobh&hh+XlSkw+IF~bFeeIo( zdhx@e2wJAMo1{x*sA}*CUDiHMO9Ury)1q*43rr+uwP?K2a01moH-WfCDZAUrVe-2d zTyybmh?I+_b?-(}<>4csAxWIX>k&}$A_z-wvn++02I+qP3VdH12`OsuWZX&H53?-4 zmpnh&@B@u_+KoSl7LecBIuy1E!J&1Ml>6Qt+uVa#@A4dk{U=Vs!`(E;=^!2PZA14} zRm{!0hHR@>(yunwb*xWEndiPZ_n|EAI6N2i51qj%EnhCjuTrotel|${+d*d*-pA6d z3mLz)m9`w8O;vm5LkHv69`;m#6VYnipN$uBgYRT2Fi^qmYwvT`#U-F&F&=03+G0iH zDtMwE2HO^%g0wRQc>aAB`Ca~oC(k-#Wr;6Zx7*-@SRa%ay9UQ381Ux5@=?r72bz|M z@KITTWU&4g+%!r7{prDID}0je5z_I#QY^IprzlVZPNqx@I$i z*DZI9;+Ea@W$@aw!A2^^>Jm$u*FG`F>5 zmzD!n?|h1jlw&|;V;Q&W=w4hnIumZ~N#`croPt}c>+#qAbkcK_p+_t8IYl1_L4?W> zst1NKeqS(6Q|lL8KJLn`j&wq&+Dtxv+gRSAC<0eKx&;SH0PYVRgnMg8qPR3dX5dIP zD>9`KNum4;?X!GD7zca8Ize!%8gB*_QYWv)i>GMwcJtZ@&RKPkfF!v9owjTX9}z#{mMTRH&1*!F_K4mMe*|gz#>tY!(FV6XS<$2PwhckkTI9rI4GiklR}Vr_Rb!yvJ=^RkRYTGT!i4h6i{%vkJb} zseyLLj^f|3@4|?Ih4dpP5O05+4LZr<6x*pn*S^%?@ykPk-TArP_XUbjvttaA-34q} zI39X$zXN0W74YV}B#ak|L5@A=yGQ~}E~i55rbRd>zX9TA=i=sUjy#!@;>Kq!GC6mP z>Sl@a?b?A9r>#uyWrcCMt2Y%r*COe4YB*GEP9tQi$>rAszR-4D~}@DGOm^JoxLb++dH%h-No%2 zx{pbW6JLGN3Ou$x!G8%6%!e_9BtK`9+n!u}>k%R7T4F%OlOjm8ligoUuVeY<+qic_ z8%VD1quEgtNi?_{oK-zZ*xHl2A4Fo$v&W>%vYlS)hjDtyC`vf>7x(zp!$;RC$Zy$8 zk0kffmeg#_E>6St52lc=tpRHS^+;rS1ddP&pb}m7bINq+<@eDr`>qsJGs4@mDbTbE*LpvPHa&o++tXm+Vm2=BWt?}j`>;y%242ai#et_sSVlhy zJz_ZWeqTgYW>&BrjqqDpE4rJ?qmiB?E%kbc&YdFYyI3DizK(=qw^O*xFbX#eoJPs_ zA*i|PG~V)>26n$c;`E0<;M*<2|AcdKc>XpTP&B5!I2~;DZ*Xs0eFfk2CBbd>Z0`1i zYGaq)y#%4bboa6pW8CkjA17~+hK?qWA6C$v!q4Qo<~+0gb*ZrvR$r519nyZ+&Bo{2-rm*}&CMW0MjMZBPumpC~FS!>`(0LIQ2YjH|+7lWZ zg5XrjQoMLllX}m_@(%X#yxsn5bhSs6Zu{!dz{+v7-m;sFRYKT~X+B;oxJj~8fxl|F zjt?7hp=sLhF)_&p$J8~@jQDHx`&|STwhTkep>Vj;)Gt_mUl_jXdg2C@WiDwW$Wpl>pxMIWGE-R#ai+bVH$`|P4Tr5y>c+Tm_#iO#H z5?wtyjy#SZqX`Sfk=9dTl9ISU!Y7_mg~0_{e#nZyvrCCjNIOS?wF0zOInSP}o2aeo z$}it`lRsR4f%o+M0B@4AuzXrI{@A<>N0m6>h){w{!e`*P^cOH|I1h0{65P5|kBlHq z0@N-a#~G*cAV;T*lXyBF4#(XWbT|hI9=(aglgHBGL&H{X$C+%@RWygKi<%+cFAxs) z#KFeG$9Qg@F6a2tkQ?6d5hfh|iS=1OP%9t>SNQT6tm}axX0n*SMhBd_{JFplQ8;P( z3k>%=heLt6_)p+O9z%0zrtEFju{0D2*UQ1`^4AdQI1Ouy)$pr_D+>B6amLfXxPGz_ zNemprU4F6n&qM{oPhCap==W$n_ak}@d%!;Y3HyF1Vtu+6&8cSjq3ux^xQ!*BO)hbV zjf1e-G!&m6o{E;Yu7k%%V?J${1z*fKC8O9&$U$ZHG~@MV{wIeJhwkwA0OY^ilOcg@P~>P zjNGp+fT{MN;<61Umd}PS??SjmVO`iVyqlbxW9i+u9|EBb6W~vb6ZS7K$1~P8l(hCJ zwQ}PbOWzB9c8{adj2N)-`i-+bA0XYTX?V{r9%szeAcL+Du-A$cRJQ)dH7<G#cFB!IcH*3%q8!LjYq& zRR(N9?c6+U6W@(@@jmP;$i?L3I;?!9$ecR`5aTnCj8};Bb0x+2a{pcsRWt-8>xnxsm6g6|}EImliLnqzvQN)EJ$G)%wCDvTXr*O4dT|jx=~RLkEmLHK6M8N8GcO zu5jMDom2gbAZ_~&u=G7HJSB{$T6W+p4NbaG^9@aE9^${vk8yERI+`=4X`%cDV|i~U zs?vqR%=BYmnO){i9_`<0pN6H9vl*%h{XYqFy_!o5^6oo zxkQG*X8CI9Y}W9{UgM8TZO!Xd`W$V@7ez zZ+<}99ZCk-*=xsP?wZmo{8~01SIqhk#%&NMji=MGx5EXp6C~-s&FWZpDFCW%4q@sm)&@PAKtfw|GKf4x;HGQ_nMBB|1uJpavrrr^}uD(cfsj| z0=)8bGG-4=07JhaF6Yig<2{jX#sfE(qPWO5*wGoqx!kIStA+Z4hF5vU1uye))@F0& z_%#IiA{&-BjmCK+K0?Q6V_f=+WyjytgHnwcKWx%K&+d()&<<;?COLLL9gFW78!4o6 z7`A6ifIQVlQsh%7ytm&cSr+0kA(V3^%oo zQH=g5lGhQ#Bi=J;v};hb1J!(HsGZ1f5G453hpf4jSD27!?B-_u;oAq{;C*_+_iKp8FLO#{xQU) zrt7%;?Oimm7+{XPX=M1o7abPx(C9E1w?A^n-}~BNzk)x39%EP47!qWUAwkzbYC6@K)GoQfqwqM*_I7zJ&L+MiXe74J&LdOjXF?z@qMiz(C z*mgS-l0S=`cN8(;aVfXJW}dN1T@*Yh9M3VG0r&Q}9Xi*Wat8Nqb3>(;{12yzd}`1+ zTGsa$p4AwFN7fzAF;#@Scr^m=hbfcQpI_t|?ZW@(tHfWQ%eZHrGwIEET}+I%q{V`K zjGJi#<@o^=v`U2jD|JDIBg|2+Hy4!DM{$~7S3oWMIVUZ5*;v#slviE;h%XZ_;L~rn zq40%8XwG_6>X+x>)s<m`%DS;Rt>_mtJZk#R~F{7-TCBa7x8^gIvz^T!LL8ZU~$tOP%E*;ag<7$axwBA9F%=H6t)Nmfz-gF?%{C3O`y@ua)exuky2~sh+4qg2>;K}J10w-?^ zZkEy!?pfV0(EJfalcS&D+fff7c10mNX&=M{u^_x-ItNp4%hO@9qg69X@z)kPk{pnx z#YUT{C-DtBFuwGlE90N;j9?6h3N-%A7QP(1=wTI^Ui=^!?Y{zCGH!ie)orW4G`yhvb=KFS5) zmz05TUosx>`d_eS)IrpZilt8>pJ*b>t~@-jA1kXGaNk1?fn(wy z5PW6c_Rd{oCm~L~nO(@;9^gjo-VKu-!eP&!lj!0(mR@+}hr~?n|Br6L;Ghf4!<_bb3=RwjZC1mk-XsIaRapk~m9gDt{9+ zxke)A$N1?>9}3!)o%wazlligP&#~5h7Isv*;3VI|)~$adr~Bo3bf*|V#${A(&c^S=VM+Fm5IK?m;y z_86DWz7AS5lDP|BZk+JQ$B?cn4qw;oK+kwn2srxyhU=r@siQ8|7CFN?dq0?e?;Ad| zSPMIr-!itW)n|;d6cS>7n#Y$vLrj4Z*YUiOJG0per}qp(@YHV3N~#WWPb|WvlRx14 zAp^RVBZKvxR>nV4o`G(AIr>IDLNzgYzU=x@-qT)~?|brFp!6mbB)$pAI$IZ>L}jw< zd^Y#aHIFhn3h2~XLpmztNk)I?lizz&EWG_1E5)^O5#y<-Z>_+K{-d~ef1-`MyeCsd z=~=WeS4FjjpScHrIqt;VFYtuT-dAfmh!Ld{;N|MI8v0TDcH zy8H=SR-V9RcPB#6z)MQAo=*FoPi6PFGqiY?5gqQ+p+XaV=B*!~vsaJcDgPg|-MNpJ zN3KOh`Fw6v=y~qSzMb6q|1P1v>R8OUy9r*_4xyN%95igV5L|UU!EduQikusYZt#V&a9C5{99qY=k=S;q4;z8Fl(&r0BdNDJvMoI-A2 zM#8+8CrK(L9FI3Baqp(HonHG8CpdJ>cykyE<}H#HR4(o^?h;zaFV)-4SBv!0+*`-+ zQp5u;!tA~AM41}!dmO=ia}+l2EsLXxELC20vRbgMb~3+NjOS-CUwZ>xhlo}o?#Q}b zf_WP@(@=O3rL%p>IN8lKbvoMt%%3b!ITZ-Ls|+Y%Lq6SG{DxZ~ItRBXn37V(6Ogo< z%4O8ppoOy?T)h1m97bNi4{txBeau6Y+gHsEm3Kf`rZz17u8BAH{H9H_Q!(A*9)!#< z#Kj_VVBDt4__n{ecjkAvy2wAqPvgaDF)vGV+|+1n$4jmr?qO3h;~ab37kC~~gB6nA z(C6t5UG6$`k^e!G5qWrIcnw$YZi8tPfps6Q(}A`vv_tVch-p@EvqN6M@QXyUeRPex zsj`;chxAA>Z4rGl7e=81C8(UAAUL9P1gejY;3H}cG0(+`j9hMjm(o7Qt(i+1tP@lc zH3dt*8-nZwQvt811IlB+!H$-{v46KGcS6*gDDL}l&AIB~ZD&~P2X z=eN((w~`Y2a>AcNA%=<`pTzjCGPt(QKv1(;T%dZk9+IRdaLsNm-0|lU`0(EzvRynt zzN#GWaUu-O`WMpR=tLatbcUS6H}G!*f78wDR!GxsVz1pN{18?F!8TLKP9v9sJs4-O z#0t&Au42*SwQ0(PCR%@SAvb}VriWs3He)l#%Rq$NUnmhgK(#MY zeAjAbkyv(@suv{F4)@*k@=zm9)t&)M#XA_6LyZK>Lh+1oB6ioyA(reVxud4=N&gl1 zVWv7?swF&2K4Cq-aCkJgAZiO(&uT|g%A3CaEt(csd4yNeIj|ZFvecG6=TNt<7!>j ziLVjm%it(~Wp5#y*H59TgK=cIEt{@i2*e&EQR)f~pxPTfym{?C-uCi5Y`T>|r;{Jj zx(_+DY+DO@9a6(!IUy7qo`NHG&m*7V9L_^GnaeWO!Z|xa@zlxydafQ#dd>GR>9Zz| zZ8D`0vlPtkiNvqFf@ zBpUMuY_B8rqK4o8+l5B1`+$Wbd<9l%j@AnG*aJa-KO5ih z(ITZb<~LkFjlzwz@j4fXKZIQH^Wg&0)OteGR@Bh#YX!8>?lkR^5zz6vUNA^?!l#0T z=P8t))CLb4fY#uU4blt+zO43Q^sLYZw_; z1@%2EU4)$aCOpRW*HTrr7&WRGcCWR^+_h%( zG2!6 z-(v{!hp<{^5bw(0X5TmUIB~oRjWMjjXBDqOGW9T3UR+A&0zabCRVjLCXvcEwad`3E zW{`}r#FBXrv32%i80X;(v$6)b-QhN5{jeRI)m-rd>$t63R*hSohPjzXo4Ma7H^Cyi z3v4HL8}E+{$75BS@U3PIEY;}2ar1*{M8go8_{5_|?`5d%Uxs77qflh$Rcx*)WuKFB zG`ac~dFOA?;9Lp6P6^?=TK^9D6N!PhTNOQRK_LHz`+C5!S;80 z-_*DU>GgsY@yqZ-`X$VkRYKta1CnhVN1wB&lUuhg{&9SSf8uMXZY=W@`-GFFO9zR+ zw}jxyS$HL4A$=acg9XnQQt7Pa?7aUM`)5Bx!{mjeaH0WK0>n{zR6T@wzkpy&hCQ!m z!-&jgTxc|jEKY>bY`a3}d30CsF2MnOqRr^G#b#=MqD|+8A8|wKN-+76Dw^Fh;J3c+ zrRK5>^oD){76C+coh+rGQ04 z%q6;d0C#`M!nC+-h>>ofEA=C3N%m;6{;WvLKXTL*s6#>BM`3nt3_N*Z1c#@61VffN z7uQa}$RkHE>5>s8q_|R~p&q_H@fh5_8aS^VvDo8hNe|WL)0Y>Q=~c2ny^(y4?Wbpw zn6we*a2?!`MI%l&&Bv$afvD$GN0&Bqk;>z0GIlcIRl-;D%MMNF-;3DrC(keBy_XyC z4xiT3+_<;6TeSmEcck%mTZ?$Bn{W8-17~sQ*?CwL7LKhi?dd(s3u-Mp2^$N-$Uu7x zm6&B=?ZRd*BjFNP9Y29f%2cJtLJ8;<$My!M#W?FjIG8S3B~aO`jCXUDa9iwr+_d^I zif$9d2WJN0%fC)M*8B$6JbMQ|DJBGMzj4EZMRYALl)`6N(x25&aCCwknR0HJW?9Di z8RGns84~pT4STQFo$ z@mp!T-~$w{Ri-n8N|cu`k8v5Bu|BOBkH4BjAC=ANi1R9XleC4D**pB?tZLHPR?hs= z^4NLl4$SE>q^-Z5iF+SRCI^z~*1!Xla=wA#{%`2hFA4rjB7BZWPm&`n1LUkgzBk7==6zND+ z;dv1SitCt7nsrk2>FF3uSWyQSmrU_Lk+axuMFly~9UEmkz;_&uDQ8;MZ?cR7!Liqxi5k01KpfG*3c zr>WZ`{$$R?>Lqk9JAr%-^pN=E<9Jgn3fsF<;E=`}Ed6#3>weB8hZ(t4{+|r5eeNAK z|Gkc9&oFo1D?d0`!oKtDf>5+`Jsx-$Pm9XMc<*hutKhjSw|rzYa1tPEabqb2XUVKT4G?oiTM z%Scxx6-`FY;WN)N&z^(^)=ZM*RV3W$;-{#>AYy+yz%*EBqj*8_Es!b(viX%1Rc@B!-W>P=aa&9VUS*805(r}%q&#lw_N!{m4WfJWOxC@uFBy8{ois6 zE@*(#j_J@By%bXhCqe#H89aQqAG=${aTj~X?%8dHy3Nf3!N4nl4%^>t)4ImZKW2=G+9m$PRAJzY(}H(wo+j-or`!D?Ut-rLPW zL+2{veW!%MJaHZjVL$iJ?=$$E&PRid&vE|=B#i-m>Nin^v5fokveXW5xtxc#L}&C_ zScgTcWpV5KUiAH?OZi|940yt|;8Ta0i2=5jB}4Ity-GdvaI%u5(- z;G=~%@fuAJI4R{d+$7!a-2F5|<8OMQ#%1ebxu3bVFm+}oEZ8ZFZ%Y^O!sqRI%?C=n zZqHrF>V1Lh)1RQ?;4ogRZz1{Yn-sC?qhQT~3Fx}>AtpAq!MLS4xYT+Q9Upy-+rhlq zIveU(p4|}sTt0(#w#Vt`)}u73SeDORXTguV7|7`qTY>g+Y5bfi4K4lw5HPp|hpH#g zLFO{Wyz@sqzi$C)8+y}?o^I~1_D4=e;Rx)nZH7NRDKNe97aYmS;a{J-%&$E?j$b?W zHe~GHfm^gi$fre_J7jv0(@1oH#P1QgVNvKn0h$YYN zb1r%N@Z>-o{rIiOKN)pQF!zc$6m;CgZzGgprj9FEE3l4N%`CDxugTjx`}1qxDpCHR zFL`<}5Ba@df&b1p?yKHh^v(Uqy)IUTd82;7?K~wIG#G)))-}SoX`*2L=06ns-Hofy zSkcd)BT0Le0{%TGjFv-k__J3I4Ynn56;(F`PA@9Cir?!YQ12N47Ng8~KW>Uk3mAsn zN9Eh<_;I%jeRE-5O&wG2-tYqMquO?S?>h&cqAfH@96~d_Dv(aEfSC`Y=Tyv3Z| z%c;oq2h6HCsNUHh-UXApio*8ZAv!WXxZ80ZniGKR0GK;eF5@@Zh2w!o221eE9 zqZ@Nnl&tlqpta$o;`oh@S-L^uBnO-+%|P2EU$qH z{!dXkk?o)k2a&&L78Pt2M$cEJ7*m}=_U!!Q5;z;gKE%Rre{(E#4xl+#Z&UANE4tk@ zl@3=WlTq?(+_WZuW{x>a!&zn|#Vw_A!UD=z#h7XbKI8kU1?czrH2(2hOeu{KH2Quu z7y8-&I+v>AFC%G;e>aPI)LE`~$wOMPS&kp;jQn7L7+i3hj$QfjWTF?qWi2j++u>Vj z)MUoX+06Vl?~dYvm5QYFDF?eVUxDhE8T3Z`EcH8@(XilNSZQykISo1;B1Ufp_D}tBzGlqZ?5XFUHcH$D*VCR6H(|N%sO^2k7K^xD*Qf5 zk}^jNQ**})+^Sv9-Rd?#$#}+(c^l8S#O>$9u8Y!s^9ztI9YIgtULfDy%p>5s2{Ble zl)C?6Ync?89$kie629PQ*CcAtmZvYtOXx~tBs~{?Lz`S9!DEsI^qY;s;h%v}zT`9{ z>W7lQh!jOj+=m?#+F`;Uk-A);tNvgvz;S^fVqHZe%J;!`1lc z-$k_54TK9UFB2c$2TLU8<3gS1q!w$>XJ#eR-kMN6>$`z$3uEZlijgo;t%;Lq-^odT zvEdh|-r^M$EO@Jfp%glDJx#39K<%+xILS_Fyd?7n!>UyYO1IIUN*6wQ(;x7WQNz?& z2Ymi&8lCH7&PT^jY-SOJdIz5a_wP6+>9w*fSQ1t=Pa&CbNp8uvg^fO) z^!?395=@_h>EJ)WuMjj3jSZwvnc`4{}dG2jQjJX=HBi#%;M1hAk;XIW9W5 zhk1JUPnnMO9Z#S&c@4+MY(Z9F<9s0mzgf0X^6R;9`msCdg%wc!dr3OB#g6V@pF^{B zrQnruF#NH93De4(@%q%+bmDI!o!A&g3u2x5*Q-zSx{s!yte_WfUSYjg;U^HG5QJl< zi_%Wr)wp!*V^r>p#Sz<<2zCx@OojmG$=s(Z9zPw+q6uF{ z@Nb+HdC8mi@pz~bELrKxeP`Z^Cv2A3*s%dao;BjetXcSc{%O?7ZJ`@1TUwI6m}D=f zk^1yA6#Qy2z4}rEJ`}+n?RX9euUFBkLTz5iOu$?H5~0;*lJv^81pMFbz}N-@EW3UQ z%KsCkjW0#$*U^>Oyto*Q&31Ah@3nBcZB1zQstNR6O}OYk%G5hE6y9VV6g>BlCk@tv z$HGE*Y%m{J=d5O|v9~ZKemj(}nu7_BNfvr&!ZDl z)#5OFZXB3xUO^rcE>mL(>#ms0N3CPk+!E0Mu25to@yo_gnr1yrS4f3z7c#)rV=hcg z52v(E>A1|I63m7CvGl40zTR2Qx&6z)vWgavT&fIHTIKP>e}`E8T!cDwqtKwt0D~Ea z`polC6!m<_d4(PmWK4D zv=~w5GHb=?+Yd25E}wA^Zy;ul!H$pR+`rUhT)0hz;EHu}J}M@B1kj5po2# zKh(hO1;%JI{S~UZU&S+q2B@?79bCBGD!8bx0-w)$q4zNv{GEOQ3Xd4m`S~xg>+(*_ zeb3sp;cU!RXmImgn)5rf&blzb(zHJ<@q)mHj8Z8yobD!sV zi&DzStfcZINo8b{mC)W(gG53`rG;onRI(CLC}dM9v!uj(y@&t&;W*0k-1l{z=lA>h zbQk;tky2F>+wBb(-WlRAw{0|*j;&(kU+_jykCzb8Rr^IOj*kn_O28NExN`P zF5SlMmbQi5u0Hs+S`8noq~iOSP3*gLhMrV}km8{*Jk&InBK zfa1c(@MCm=WoSf_Kg3_TY91tC!$ zPS!HkG91C;SC8S^nWbF8v+rDpdlMeak)khGWthvmgGMYq zLPAct){8uba*p9TVTn}d~=Btdm4!mvxOPAGYQLkN6_a^V-%a?3py8mQoQm~+!6X) z;J#9q#xh^EjX%)p*CphB@dIkUa|5Y!MWn7O#WxQeBwk;IZ!ft`x1LSH4WcE?OH+(? zeTQ+?pG~B4Vk=*Ex`2P;G>B_=xzJlZWy-I=jh*Z>x$Hm;meqqJ~D(Hi>R@sS852?rBtag!K)hrLkit zbKGfF;VtVflnYzS*|t>Sk_RkbzAPMDcNgPB_UvdF5~42uS15PG4CA;`*y(+YHs$5e z<28Mdd{-Ka-<`ylPG4!x_$K;sM~eD4D&hP+`6zZ?9Z&w7OdFGClI;<73RxyWDK_@> zHNA}G1V5u^&j7k#It5{(R=DDw4px&8PVh(o-E=#Ay;+9@x)tF5?c} zCfd3zmrmCGM}sj}aOc7rD86e*4 z#=nzAX{P@R$iG$qg=gz*!~oJ>`0 zPhMpjjCLc(QS$f)7{bnS0fqrZ&`L z(r-yz*g6ShKo3Klw{Y5+FG$__33ciwQg5ggJt~dj#9gM*fmfsX2H$e*xz|dfFhrLA zC2&J+I&pWHw|846rr*eTsi6l`f0?|n)Pxx z5H<_GJZi$-Lyw_yXzY26`IfCwQ#(d|#c-*NeOL2^+{B2u0u4hi9=LzFTzGOL@1+PT0t0@AT zd+E4kk^&~D=R(Q!gKXa@&WS6X=gPj{qe(NJc=w{I{MEb@v_g6hEj{DSJe9M#%GGml zwbWZ2YdMFTZf8OVZS?uP_EYpRt&TF3MX0;ZfOhqt$CF*5c-wy=TJ5Ys>!o&7G$e%k zM|+^uhwE@o)`~tmrQ*MbGr+;Efj{_cF26=ipMUl!5xFUgXxvdZdRn-NY$aN;WS|_r z-w`6edAq6PNI$k&-9`V}QhJ|ynxuDx;6zh3?$hzj(5@Jc--HS&d`<$ro|#Yf`=(O1 zwi4>D$fRq7o;dw-6BnJ54O894=?-JX&z|rHU;a>|>qre{&T*DI!38Ryv-4xc&`yw$cE7Shrx?FEKb``I{>XABG)$ z1~h%E6se3@PG?pxqk9Evw2R&OK~sR)jBBw&C=;fuMNyN`S~~GI5{o-ne^+uJ2@Sl* z{o3*P+q)h1-3x>@`=hwDrOROCO&$-bjO5OIKY-qTOKHxhQaCD@fij^_8NcWUs48=4 zw%`*K(0uOC6m80x7spt`COBB?ffKD)V{GFR>}wXsIU)6sdv_$rmM#Pp|M^hzTZU?? zPEc`<4daQm!Qr>daG{SUuFyG!Z%+x)xWrjFsnG-{u-~`HPZP4+B1}x$g(R)oKJx$KVTM?2pVu?4^*Jhx$ z$wLTWe%!6^{phLOWXifTmI_spX~|yJ?Glma=L&rynsJ}3yxGnw^(ZZ=Z^H9`T+!HP zI&2wb3mtm3n7c!Vg7%LknJO(1i!nff#Ub40KANAT;zn0rAE#ByY$iPQ40B1x;*XD! zG|BEgo5vf_ld3@S9kHHPXKFDYffxyxH*D(Z-*7YH6wZw@fgLv6xE)U>;g;eNWX&8D zGq!m#c4i&+Ww=nNL^Mpa{R2O zDWH;Er(j*dQbFULlQ>u83srY&@G(WE{L0J!aGpvoy$$<;6K?3yjI*vR`)5Fdg*(79 zKoZTEqjbUO+gw&nHz#(}0B)>)gq_oMkW4x5_)mX!A6-g9!WXFVog~Q&jG{>rY_1X6 zffow0@noq1s$W&2uX-ko=`evz4zHvqqMztN^LaYTGLFN?uA|9cbBqt0Mj-HRIi@L_GxpRSLm|vhTiIZ`yY!Y`p)c~yuqA6yW&77UiQ$y=B>VEf?do^F4 zg1b)8RzC?UJ6lYTYVOd(`9gT6#0PiPeMgZkWe}ZVjW>Qt^OtJBaTT|t@a&U7YMo<% zal%(Hq(1rvm)_Fx(QmIxB>S*DZth1YA837$f#Wpn_HqF@}Mo_ zjQI&Z%S&Q!bsc`qF$I%RZ{an&r#o~c!tp02q{Z^g!hh%Bfp-S#2KBU zB>SWj_Onc=+>-UQK}eIdg9_k3!A0&zRWM2zzJp?wJ!DZUO*=g2km0uP^d@eIUd-Fg zy<>b!-L1-`zgm`i+t7~R(yOV7WlBzd*otQ+R^XS#0cb0w4#%clhtK>5vYr}BanZ3j z;TvOnIdzeO^lg0geiLRj_)tl!fO5Wn!{Yc;_*^~?uR2~~yp+Ru^iLaIjB--lKCzTgWg!6HSNba*oyu@G|4cZ&*OUcFMI?I5Epv1nbFhgk! z*@!Jd;iB)5o5`{>+fnf5m^W5E zRm`oucpd$A*z-RgZpZcFJX$7LVwO}gDpyP;xSc?Wo22pn)34xO=1kfz645MV0+w8L zq$`03xn9N>p3eG&7H`w>@ap-jzj*>ZwvNFM<2x`l#RG>}=4ef)9-sC`mmcfxg-I=W z*p-k=m2c}g(ZXlUCm~C-v3Nvotf*{0`9Ln3|m>g;zHIB{4-CS=11*EVGCtmX9pNlp$46&9Ruen;*?!6g54+UQEn;wzbH4tGe+7N*XM`Q!+K=bY)XTU)pS)} zk+K7Cz};02P;);MRYzx$;)j3q$|IM`leP&Sc-ceFpW_1M&Qs_#S_GR0=LwoWNZ^z} z2^d~o&J`MO!pfc)E~R9l;MTrG+;%J<+IKs^y7Mv#2!}C~RAY{EcRAbA=SQ@8mfyd#sKihwrysqx2o#`c(*S zJ=Lj-5*=58z3klNc%g;)Gy>qX)HzG}1!}ZCKoNsq-Q~s=Z>O2hlF>i48A=Un5W-U- zD=S_g^Q|8~pYh~g;6cIFn~F5IAsUx!-Q|=S^YeyWJB$guAeimKQ=H%!{F`<|VD(=$ z3t_bWO-wBC!i_!(aBZY1wgz8?o%b>= zQ8op;b>Dz`j(Ad zE{$Bo#RxRXMfCqx0=d)n;n+2&!PbU(7sE#5qVw6PwyKahXk~Fj$zkq#X9KssDHO{$ z6~W6bHG+Mo4fy2tu@qdi6=a+O;9RW&=||0@-;GV6vuq1|a1R&oY=80f)hZlfZa}-2 z4)jliaWRXfnPWNwj3;apENz;~HCz`1u~*`P#j(bgt9U69M&$#%4q z*|5uU{QZkui2e`DcGVRYGy01V`g&u%|N- z(WN~Pm2pLkfV*NINv%JN(MR?XjK|${fs!z6xfbkYy(SUHmywli1EojqjDzAOhzc82!ZbIBiz5pP6sb5JZ-yiC<{`<`RG*N1$1^y3a5lD){S$XF{-EcLOlc9q~g zPMxE!?9rqbBsS%v=o0=qW3B#Y<4ki>tKpbRDaVI)VqlgWIDU-cw(%orxt1_J($mCf z@yC$dbcsu^n_#i8PzwSKURz$A-C}O~)O1Qha1o#5V8_>oRpGv86VN%`h8uR3g!GMp z0`=3kTLR#qZI-~?a+c+5(J-3Lx;|W<2l%#5 zrN;dma7~LXCtIFMs-f+4yvu_8A)LD>Vv1g>4%F;*h1*pgf{&ybgX{W3SS>n->Sqe^ zPx_9*-m*5_G9#bFSH*x`&NZm`y#mJ@U$gv_x=WBAaT&Y*xCs{6nsQ-p^SIlkFE|HX zTTW@b5(YRDGv}EQ1>QJ>Mub`v&cy$IBll4ikp^U>bbv)cmkf(t&8vL`PU8HYR z1$rOUQS#qgytqk(o}ADou?_tw^*oYWyX+*FBjgJ{*H&9>*>#Z%N>9fGxm}F)T7Yln zn^5N0rSpsMVz}O^M?zR4}z7p<>2TW4*lZCq4R7GdOeS( zUh<5Ci3 zS79;NGct`TmyV-1Ckz3%tbqqhh3V40L~{M)Lu1rUKlkMh~UG*IEbqLgMTwFgL1hh9kOBP zkX0q9bfO#r|0Ki3MUE)%BoB?=4q%fJ1uOlZUK_*M1zqF9*^pSpn`zw#3m zU(18s+8cO#C<;@bh~tCIR)n*G^k$s~|N6=d{_I<8dh(d3f*bSc#H&$w+&gs{mfuncFwc7L@<}gZ+L|7|71AD(o}4_~|@KwGl$| z_%ni-8~NPK*-l(;b1(e;CQeh!XW+XhGf`1n9tN*hVl(qxIj%3nQ;!C*@stjYmcGdt z?_q!~JY!5P;)|9r4rrbzEqb$y4z4-My}YU_Sa`NoAm!mnH-9#vH=F5|j2Xv$7Ba=h zn=Q!oHzQR_M>Ds@X&k&($*GIWKgLA_;+#vs$d%t)JOf;U1z5Cft@X0L8 zgeY+spQOYo&we2&viQm^pLEHxGpoQ-BiN0L&#&iZv)^UWf#dY+MiuUoj3w``blS1m z&@v=(Jtu3?4$abauvvH%o$?Uk`<6yPUzjy2+6q&%zba0YABoC7hjCk#u;6v#NC8YM z_uZ9H1ZpzL&iXuF7VVp4~JH>v%!UB63Wp@zc0w&`18B&bT z!% zn>k{iUSf`z$>i?DTt1gS(9dvX-okPxeKER6trmmiCmK#ouP(rF&NTG9B9A)vt-x-| z1Kh`~R9zo}EHp!~Wx-r8mPtI)x;XTd+)iU;?BQRW#_bnh|zFu4ai8>_HC zDS%u3B7)2On~0-cHp15mX)a1)k|5=>Hm<2$iOa_`Z*j~-@_MXC)pzSL`%D(Lk357O zqqXt-n{<50U&nmYV{o|ZDCaqR2*&^F;Z{5wf^jb`XoSuIx}=-SO;%z4w`ehP^SB{c z>n03)1J{CBi7sY}z2K@}Znezrxs2Oh8lmEE3Gn~45u$JQaBI4fpg@7cb56w|<~@h= z&Tyb-_2=+?!g?7~8uN?e-raA=@@wH#(S8o#KTeJI3Ldi#lHrDm5_JGzW7&vn=a`i)C?Bg(^cE*1Tm$FkuIAPgjj8i%aGz>4 z?jERuLl<=ELH2VDjQWCWvSz}i1Nk8Nx)zp6q*~hUn#;9)Jq#rZgMx60Btf#{Maw@s zf@%9AB7ea(yjOS@?0cjz=2awEiNwQY^-M@dGwAi{=bGLI3%)KW5p<<~#Tkl$XpqU+ zX0l;)fA(ke_m9Fi1y!&j<`Y*X84AG@kKqwJIa+yb0a>(*^0he=Xid=>ZUM{OU8p-m z>+aalnwb*39{-WXE5zaUvL_V3(UPA&;tyk2OY(|6%*W$2n;!gUK$DLN)1GmT__Mo$ z-fdmTFG~sG-SZ47XL<#d-)8Pfw>KDhPn@oQn}w%n7&a@U;_O?h^xbDP26!d1_vj(0 zajU|-_a*r5tSolf0vQYq;+2mRnY$?h6`lyuAE_RE8nB99O?ZsUZ?i&4b8~8)tRsz8L-ejbo3fkl z(!#SlDXH=o<5eSVddwKo2@1Glf;rj_sG`~TF(j<^kyc5Zq)we1c#^+}Q*a3>;~A>E za}9Som!f#pa^|4^f&G^6z_xc8PL1e=-eATFU3?vPzHgzc8ug^ldQK#r$h?k*RHt?s zPi#%cjqbN0wqqq!ZPCYr@0>xe?IT=isl^*R&vSLdW#G}Xg5IVikj=hfOx@x`?dKRH zVJYJ+#YK~=(ocMQObq2*{K+-#0Xj3E&ZPVV_8AP*5p6m%goi7$ zpmL=rhG?t9X4U%wyVwl+($R{ayX(MR+mCraQ}8MGo$en^7i?7bz`X1XnxI(9GIfVg zdEP0q4Y+_;T-fJS(g80{o=O2C%4GDQ8aE1wU`Vhln(Q(o)jSPcazletoOyzKI%Lf5 zP()wmtftf7<@kg~5$X>)D^R@n2k&{z#rXRQ=)YMGA5FEwUZp*F=V~q;xY7elDqewN z#UGS=QURAEkE4d{7kC|V0H^p*B9r$Pyp#5HetV}Vzo^@j(%+4wC$DwzzC}LnGo6aC z!W<7C6JXy8C0ZYNhOFI}lYT}d6)L;a0bK)%6nR76-~FRppZ)Zs^Cf0fJ_E0VUompb z2h?Lr&lab<7~Ja1b!`#mZvD04%iS*X9e%<5%3l#w^{h~Te(2(& z&RM@-McT2h{NMyJdKDqczxn-$@NO<$>$*edZ7fMUAc@>BOVdu~OWx?W8oplWh22~( zY?e$$`!;j@Q}hDFX)FKvi6{T!MGYNM*n}sqSmUIFouE)ow4x`Jr0U%8{ED?Sy!I4H znFW$gk3P$u){slHC662SlE3Rz40M#^cD(EAJBz4wXMo@=M4ES!D5Tmvmy z#knbMx40F;1sFWL0DswB#;X^cC^^@jTCXi)Z28A@$WV(Gcx}h$)8fb_@(?DzO9RWA zhp5&epz$*4_*Tl0S5OG%*9F<~CVrA+I@%aN>K(`D%OvoXu`NdHEC7SMlgVHU+X#suW`T$1*|K3PbL`)Lzg+-&l zRuqH#IyG^0T^IgoVV>~=!!YsXO^EsIMsxMHP^a_(;hMdPs+sJQ4NXCfwn>N;a+)Z94eRMfzCerJD53`%hN?Wq`keF=Gro4^581(rm#zg3mv6~;-Rn^0e>f7A&Q(ri*=y;0I--e~mVoWAQD)L#4|W+LK12#wH^;dB=w<9kxMDK`+Y2-9u^Rx5PbdCr`Bj{J5o< zBnl7E2lvhRq479nl?_wN#Tv?ST1_sG7f}1z^<*TvlV(n>r$_oqth<_rQ!bC;l&?o| z>lYQk$B>I4^zt0nn;=S2cT3QvMisr^Y{dGN{cuvFg>yZfir*KSVg0Ocu*=wisz#l~ zn`wp5|7wJLr9~h)iM} zhdBHb_Y#N3mZ4{J8jd@$32({YBlQJ!RR6??2A9>~IFZq0p(jGZ&(4tcpd{UDtiavx z%|W-eQqV9eNr3&|zuaGn#Zn$pwrt^~|DEQmeuwZPtCHza_)Y9Bngc6V zy@Fd^5wvY5?Ou_TnKKdpmCwg^OA%U7XpKWprsLvfLo!Ra zOkaK-r4e%8+$D#V@TIUF-t91lNqf?{l!7wu$~w6zDfNH&wpgBDV*U!pZ+ea|)70tk zVlf=A2Pijr9}G_jrE5cJSjlFBHls4=AZ6p&{oWYFJX}xs)7-ryje<)a4x~CX9>?6e z4&~elc>ePQ9L;8JvI$Li?cYrLQX`KMAtIdX11&hioPvAWg=i1sEycZKoUXVRY-bq9 z+|4>PQ_Ts(*QU}(_gYeoHD{dPMC>{D5e6$}@mhEH^0kMP`5Z+%@+)-387(?ECQk(N zhA%)#$!IpKU!y(fx7XLn3ixaCLHgx1%;BO@qB<6 z-1Q69!1nbJn0w{HBi3U%R{ahQ^8KiyUj#pJiTEq@5q{UwK}p>MII`ahMGd{-sbvQD z@99`LGygVP>|mbGEe7l>$_gsBO(L zE@8uKDEj7v+Od7$bw-Z$n93MuY#paoErpM~dx1ObK|w`~jkl}?TLKl(mh~UP_YQL< z+8OxhK@0XBv12T`Q4k!m2&ZrPjrGfuC?ZOmhJtoOkK6+s@7aZUi77a;V>gD(H=-5m zkK-QaDB6~`i(XAvrX9V}ocQA~>|36LVX30jws|MsT5<_;*URt`5;|1JzlM&}6G-{3 zCM~spjuS7h$Lo^4@S>1FD}4-ouYG`BF)3hc{swEN)H3ct2)=%^iF|r8=*6q;xNB?< zW5DO&qa`Y2_}^y~`KyGU<0>)tjt1Sd;+PA8Wl)&E$epnFpm9r4l@O_Z(a^RY9G^(KK#hD*g)F zjfU&k_x`RK-FTfwCuiR#yWK5ZcFI3a)@lN0AY2NKoHdT=j{!rq*?edH8NPA-89umo zGFY-_Lu$)m3{7eR%LAuz%lMB_c)tmjeNV%e@|lP|?WBLR1l0rFQQsyGpKFBSjZL>< z(Vp4lHj~2+)vw?>VLGn)SC5P5zsJ{kLOkqsCdGBIKIqxDme(yuQrK_Rp ziv>4#+8VkX>x|FGCc)T`W~6sOAIBf+=Ts9yD3OHe>bYZ}Q#qOUUO@bV)*>3I-cC)w zvPnwb5jUMK#W2I$FnN(UbPOa=@_9ol+kxnNYY0}lmve@Fv2aRk0m}R+fD~gZ?*H%G z^7Tm=?U)Y3x76{sMmsvi`Y~3|EU0E|@ysK0$nEwCv^v*}rcX4`c7Zr=yFrL=n0=9P zELP%U2*&MkV^HeV4a}|SzzGeCm>au}7Wa?j0~v4rsJ0Qk{i#F0lE%^Bj-`~(98C_x zZm?$YBn*`lLh!bO+BZsk%cBI+bWjAQSzm&B+b^S=-9{WGJUl`VN+ z^Tm$j(_lYuH8C@gRO*IfhR6Js|5@ zz-wNP#=WO3a9!d(Q0du2#x;o;%{UBwZYH$m?`G<4aidTRe>@?ch~3eFn9H&?>!q*I z*S}GutYL>|jU~CGwca?jO^;4QIg?#g1l~OP0^ju~V9S+c$Yvd^MeKQXVZapS{v2Z6 zvp<|jP%CVm`~>B@_CWCLP=Uv&VA8&P2c_J_(ZN(2$Evq-Z9ksFu>UvoFk1^3*WR#v zHPeJVSNr0<1N-3vO{c-`A1trBl=?L`(1FZhdOC;{I=l>ZD|6|G^G#Ah1HMn$mj9$7 z#b>{ir2DFAlp5<#t*Xv+v{Zq7jKt`L)?IXvXvZ~2g~|MVHww#()BE%iw1^ks$4QN& z@#X$_bifyNtgK*4>@Pv;*oSy)^*+igsiNAy^8CE7W*C#+#HnmuNDucXfqU9nsQFw5 zw+tny(V_)SCo%q2Krp4^7&Hp?!Mu|@k&|r%^?hdOofrnwW}kvnz0M@oTul+}!SvBU zpC%2|<4)Jf@Q9|7?fq?Z_W2r$$T^Aw5r<%zwXvYe_8lsy>EK=08{7|NmXmnAixb{Z zA?P-{L?cBf@fH4le0at!wCE5<548xc^UWsQUY<|Z4;E5%!$#8RpT)~IjpiThxlQK% zTj~DYYPcPGl-yoaVRd31+-;don>WuOvlgUDs_Sv2(r4I{c7PN4XHOf;*JHct4DOez z2fso4KL263ijUT1yFlGjbZ+t+)a)I=z^7|)THJHaCgmswMai+;>QZ{`znDhc+)tO5 z9m1f_EhKg6BlH?Xf{65eM1?-QaqT#MkseJ!ETcL2Cs(jD!;MnU_rjKW)&e)VDhm55 z%rBI@PpT_Z|2`&#C;3@*2{*9>kmc8KN)t>(N=|(g{GF5W;@BfJFlr&rU~}VN&DzlVFN3;X9H;1f7m|+K!oC}8p=%Sz zExm9CXYKtD;=TTIw+BnfH`j);*NnhZJAG-_A|YDKr9)(TB%K!LI-EopxF!X^CA3g+8$`GIeCNTJ?!bt-YYgcIw3!@?a?*sP(LehF*v$7B=G z&(w(AZZ4x^ySylU;t%w`)_{NQcZ1*XH}FlfLvigvymtN|HHpNKQs_VY@HGTSk{Q*_ z491gFZ0Nt>-?-Anll`tvkxl=2<^~>xD@^;q#QYw5j$TNiixFoW8-#PGv*_=eIBME? zh~&50k=kc58s9n!XX?)5gL8k;Dr-fWDm4!alODmXdvo!K@lo1*Mu?X(R;2FZsp#&1 z3;50$Xmm=V1)Gf-XE+)^h!5dz%?WU?@+9}*eje>kSWjbJzc9YTY+lAli0@}-H5WE# z*FC^EWtX;-&d>(ZpVf`)P0R3IRv8Wc$iR?k_psEX6w5BM8Pv82`0iyu+Xq)sSB@Q> zNHC$}uamH^APEd7-r&sgBq;7)16kWw)9T;ZXw{t}Sjjwu$@Wcf^w)Zve!Lg^<+CZp z`Ty?uIkaHK5L)k;L?41WsW9G(zNpN=l~*S4=UX%A*<(?B|G0=Oi{6p^B!4nhjKP}= zt>I(I9LoraaJukj76oSCfuc>5aM=ngdK`L?F4)V!I;BA@ZPwz82GY>*@)5E!GT>LP z(}B505j3XTm)9zNMUERyd8=k!-nd+f79OvoMVg}g?myo2Z=@s!#m2y@#qZE>+)9*$ z7D(H#gqc=WWTR9;7tG#KZ&d(2*F6M!+T#3wKN_gatQY@Oc4Mm80_x9-W4oNqeEQ5= zbfqZ?{Z+iFI)4rBRuI4=c6R$bTAW(Hyg|j}yTnQ7^0$wq@n1xnVZ)Sb_+Om`*=8Rl z+nC3+P3<5_SlpyD6?1rpFjM}GoB=N0QAg`q9+FJxEO6;s2*$fdllI*<)J@jF-;c5^ z9~GxzvyKE=4j;v|P#M1U8c#tAYEY9rnh)}^;>9n|;d`%Kq?k|z>d&^rMj1;O8xg>z z?V3#R(v6G0S%oWHcz)vP`&5#VME~9^GWXa3Zc}%r|CZU(rlV1$H+?c+ktEBHnwSDd zg!ORbz<%tDe}h4xBXD=fZZI1yQy+VJWQ%bBd_e6mL3vnxb9M(w6}?>PFeupW=? zYQiP6QgE*JD2)G4gew?z8g(}IBfb!&7)YQ9@W1E%zMZ2H(6!llA*)fW2onn+CyR={FE@(1?3?Vkg(^ z5epxMM+h={6>;Kw3zAZhq{M?pl=NDfJX$JIqjN2#F4v_R32Q<5YZsXI_&jVd`hYJF z22<>T>CBsMMaSyy(7_YY)G*zf5}VV=I&BPfD{D~Nf-k7Cbq~rLY-gRxchDHI7o}!N z)2W29)Fu>*PxB9QcN=aBmhV)+j_3&VWIQOV(f%-Y`a_DCbcpVMK1%xgRq*+~IvfQt z_^eHpvi6%$bCo_Ng&l$wFb~&9-Grv4~0{n-zs z=`@DjQ+A_8X)9L0t-}+-a^R_Q57*e_VP#MY7ZPqHxDa%T`-Bas@G=H1cDdsCAP<^l zIgR{RONznHBI1Dw8#rZ0FC_F<3+tO0d^j!<?Q-;Qmhp~C}KWvLM;?+ip@nau+L5qfkv`TUTIsCiN z<_80?+4UE8F^0~6TiJf=R155DTFddrKVVLDH_JVzpijqLTz#k?9W+w0c%~0blrsYRCcgQ` z_z$YrAu;j~^d$~*36F1a?a%YET)GvF7_VdN=TY=joXr8(jpl~u4dNyj3wnG;iA1eL z@b5fD5H@%NM?FmVF%fQjTiG$1qsy391wwde$c=k9J`&?%Soh~bIO71^r@u2j_&uJ6 zd}RH8keXHRf&1xwP6!k*7lCVx2${4SQunT85_55+t+6qj==@VC z?95nN(z4hirOmQA>$%Fcm%;ydIR5o~3bHFt(j4D5vaOrJ?n@s8Dg}=@k5eV_EibSzdoJz0qe#*l z3sHRjL~{Lp7G>|s!R*_NE6YvK&)~c2hu37M)gnjD4q`VpgUuo{`@G@4p*T z$ksvhAAN~>8J}*xk|P~T^n%1&!GdF*dIIUkXX$JKQc&Y__;1@0w7yaa>RdG!I>C}s z+|9{otPeIEKLLNwFu(71O}KPE31x@n7_09T{2eRcul)$&f7jabhi6Zw6D9j8I_Ut3 z3|$8F_XXcJYml?kEqngSjy( zYX!qU#c*=^RID$T!c80J;h0O_aP+Sp&VEvcdv3Sku(U4z5KYCL^YK*B7E6XcYq&;{ zSTGig#4%&0;_<^d@I&bp|CAB=8a<@qm5e=dN(VpRo=&4*vy~n&5}nLfZC5o$Rl(_sS=ADD(czUA$jS0m^%D z{2dXPVkblGp=U8##tVCAviDOI^RGxfh3vVz(6iwkOz$b@!a^NzW=|jr8|h=2W&)0S zE`w2<6&RyTm-#SHG4IV348A!B-#@vI|C&Rwupt+iDQgHbbps5z0*d zj803EaJPyG_J7v{?Qy!?vdjxu8}u7@WT&Fc*gJy0@}+`Pmy!fkHKiz{C;fR7)xgyQ4*A}B}wE)V~xJ*3{6JY-@p!nc%e4%+JMLtWQ4clsP zev<>a%Dh7&dXW&S|4DQG5>)a=`k&yDvfrO%6n5RvP5-TD=tSmrq z{G>n-RXad4MherhgA*Zf7Z1)^>3G&J9kX_o((wG*INjY0TOZA48IGG!)6>QouM-xe zmF>sBYJbp1>Ks4UB#Hm~wU3G>m*dr;ySQxQVytRpxf$C|diPn9eiZK@c_KV}e*;zE^Z6Iha zPXXVATKuBbKx${d(BRx1SYF}*O$|Jkd~uW@I64r1gvTJ%sB*b)v&q_FD}N=D^)Lq~ zl6&Vx96C^kL8Hgh62>X-Go4I#UB^@2@?yN9XNTEqM&X7fv!V4F<8(eh3HM*g2x8tk z;)~VZ)Lz#|w3aG7a=^#ys=-u&8CGnv?=0v9qLy(VN19KUb*lYo~7SKt}7Lli|>=@ zlIeU|)<3E&&_HR&P6)L0=Y(|ffZmVAF82OW3*Ca&V=XCE>n(~Ok)vgW-*DF98W6g+ z0%a_kp*Am`opl=VcWXVG=~|J)y1f|an+1!~c7ySw^W4MSP23IUSbo$s8C8-b$+grG z#WoE=sH6m0d#$49yXWAS6bFJII`ey;ZgN=$kn&WpyUd_t6dbBFG!>W~ks`dn-PFdl};IB*6&go$h`x8g0gWgpNvg9P>{f|JzTP zF0>hchKpd5wFc@*>7n%DDm>%=0isnWqCM+&e_OZ>r<*a~hL|+XH(XB{PBCo1bPKOf z-+~%ivW({~My{cvq%v+MPE=B&;};A_QvV)W7AjzET^O!3Pc%uXq$omBXRo!6iiAWeq(nbMNTHGpp`uw6 ziZVqgl_44=k)n(#Wh!Ne5+WHhi*JAbK-X1o?>_si^*r}IMH=2cm!k>!1`2NIZyUm-w^zR|wBsS8-0y6PRjPFPPr2Wa|=DC^*^`xA=5IiJBIgTv$PU zmHLr^PYxESt>#RU2{%TS}K z3ZK80GxgLFXfWM@PN%BTuCyq+CVc=}pF81_k#P{1B?pNoqG9I6)39aJ0eU1X=gc%U zTGloif~K6pv>|*3_V_)He>DKD#Ax<@z+B3etRu_jar7gp1&+;5#<)XK@L1&*9OwD5 zuG~HF_*VezS~XQL7&;RaJ`BgJFWfMA=yvog-HiF1cO>1Biy9SUY3z|TBECMWpvK4FmKRyMF8n$8C=qRob-=pJiA=U=2@it2uR*0mKZ!@c3zurE*< z_!RHut)T9uQM7u33>qnLwyO+B5OItpjBKlRN8v4WVy|q4b9D zt6~MtBlMqvz4EKD)~f_A7OuvSsvoH6ID#JJIFW^898NW`fQe`ayUTR2HqVLk{QFYg zo~!WKD2DeFSHYP*vsvIOXVl-)$(~!^WUufiswFLh;ZGIVl%u>e+b0BD4;bOX&#{y( zX5ziY3H*-y4gDm|=yAeG8t*lb`#z6s;>y(fQkXIFa)*`feFR z5tjT6ouG#!T%UmZ!5Ap5m@N9JHdDur!4$-d>EVDVtgLfnp4zr}ccdQR~>-+<_ z+IAjiZ}cF`!U0@M+XG)ZPlIt_9`g#579Aw^)8ONmu`4 za3Jd&W-at(i!fVt1#ES@3;h=c(5XAUnCy@YtEwF_;YBMuP-c#{-9X+B>XfXVf+^62 z)3#XgoR1qOe?JDcWfO3$lQF&AFo1O3FQIeCQrzKegMCI`!{(`bQ9(b3#=m?@Ez$kN zmEIn>cQC_osvBr!@?`vbOCL-=7J*xj1V)w|fb2_7P!q00KDxDdcG)ak*#8nB?L^DA zYJ7BA4wbhTvF{tT^J;8vHNoxr@-%DSZ=B%eNN2gKEAW=|(a618hGuoi#Sxgh4eyLYGH53%$@sY(0`eYc8#)mVQ0h z$J`T@_NGC;;~_ftC!B6f*@H7zsxs^3e7L%I3Z^>Qp_FC`d!zD=^{oE}W`X}S`iNk9@6ZXjGN&9%pe>mAK5VN2kS*OLo5vzyrEq^Fn<9YarhsV)}&ms8xX9ND6mO?uB z&T~HWEbN*xl6)VG-%}{z29LfR96B@OGantm1d?%EPMk(m5NwW!hlttS&t2(w}}z zxgdnSxB$^toaxBQXApH~FQCS192&42BD*)?x}H3&dpVM=;X5Y@ww*lPe z@Xv7W^1iG<+Z313mguW!_jWRN4%8JVx$LB!Gs`(M?lh`*ec(N5ZJNh--L+55>4niY z^i*F871?ue!U;1{OnAq;GUwUzO@8nn_)#a0;Czf*a4^^u_uDpM#Akv2RV>HaKiOcX zz8@s6E=7eSek|zgOuQblh_m^+A-p9U#`~VaF2%`orCWnGMk|wdS2r#<>%!hs^2|H; z2t+KNh6CKziXrDE#JmTmF?MSeuF`GC_nVY($FF4U(*1$gM`7`4BB9zaF0CU^% z)H2_c-0LUd(YD=$H|uGS!ETcFT?ZxBlA!6o2qind`S&bKpl0Q}c)9QtESr^x(!B+ky|ois0>0pqMn^g` z-G#QAKcda-BTc@oL`Izp=x|dL+3-EivyYSDmckI68KOj+7siN{y8>ypb!Tgl7)o=s7#YmrG-rf|4fQ_|<_m8;fY|TRm~=^*f~3 z=N}}>4x@Jx$yCMPkpsSz;|%i%dirOv@VRan9`LuIf>gkzS`WcfVJW-EIkqXrtvEU8 zC@SxIL)RBdi_=n{V}V?NU|(8*W_KUcCy$HpRwWjf1nk61f*ojDd=gf#l4Uz0tFhm# zew;1gf;&SdV1C>V9P%|EI|B}3=yX|n#@%;PZ>rE?xF%gV>O)aeHbGEn2UN9$<4{*? zQk(aW=gk-4r3xAJD!PpCr32~5-+tuuA)hwsU8G^k$_Vjg_<3Xyy4eBV8kK_Y2R^{x z&3~bC;BGiqU`A+sZE0yk)Zj1s1f4pA1Sva6v*L)9zl+mUVt*e{;0e7L}qS%vRAA zC$1boi_b~XGr1wyn*E5STAmO*&t7HGp@wYH;#WeA#T3D|XrmCJP$XntU5j2pFJY;t z8BTgL8n(P_huL43pux`j!lJY5glolTgx0Nn@tWH~FqKGTftjJyAtO(6oiEXF>0)MR zw_mWBsRGlS6hs+yOA6j9O}--uh8vw@=cg}VZ~WeaOJx|n)y$(oy1AUWHwVvLiG$;( zrCFYWf^fQ_({hcW8u)H*U>R09Y*)UzXtCB_{M&c0IO2;TbDiSHUIi^;LyVrlFx?Uq zrevYgknK2N+6NfcD+}H=>Ns0-4Q!8SfWkeqshFDZ&xcVsY}##HJ~Rk(+@)~q7{ta` zVHEzYjoN4D;+Du^Xx(56&pf8k(<{gE;NV(ZFnF4{x%>xp=jezFB}R*H3Npp?Z8c)~ zd_^?x-2>Ox-GGTM@+k9gJ$o~A0IaTjj8kN$!hN$;rW=)s3ImF0P(chS&gcZm?OJfm z#{yoBkfFJwrKonBGj%KMNAWMA=g30N2ySQA>vw~jWdo$=GzrbuV%Y3UQV<+AQk0Dv zL}dYg;pdG^y7EmKpEg~D;qGrlg}v3HNkW@=EwmYrr>WuXue{$KIe=VwM{&gd2kb>( zJ~L@fBz^CPH1nq%T3MZdne(2qao(yl|1)RHKWGK#gsCiB{{%a9^e^jO;bkTF$yJze zbDPy4<)!rW#2K_`bHaNma-^>FjIOPCMGKORp?X6A6#bb^9r>Zg?QlE)Z|O(kK^J z*f)$kERW&r$`sh_xk~sWiZG&JB&wI4WcRfc;aK7Io# zdu#BvZmp%Pma-MLXNstL`Q(>M{OEOz1M?FZPkopr4Az(^L`w?7C5{$%_68V$K>BuumU4YS)c zNZ_tXn}j`ZL`|A5t=`0p_b{kknTc1Ir(vij;he-++>j~@qaHnHuJjt-mF2+n54v!+ z;v;@G2xo_F7vR>2U+D6~p9!lzv0(llALv}iq8CbnpVl@|alFQUo;eAQH|yZt+~aIM zf7g*!y^WXk-L0C=97Mk(LG(zbiO#UZDo1+u3`p^p4>pTutbsOU~ z_e6Fqa{<`yU56)R&q0{fXJ&A25uEy{Cb*Ur@ow}bNKcCZlTImEyyY8Cm3%Kevipr& z%}_KNlYvh^m|NLTe#ibTlfvQip9yC*pP=8vo3z-1vp06DlDU>PB^N)3 zjA5L=&offLJ?Eh1^nA2^Hw6>c)Ipzsdc33sIj-DyP;ZkcDy@-SS zWeHenbPi8*&Q)*?(8c!mp!e9C_Gli%cY%3i6s;qMSILs*FH<(;kRD##JP&VI6*Jwj z%5+_3GF5KYfjRNt;p^)Scy8Jk*0FR2wA%B&%&RJ#QddYas}iYT#B$+C?+J*Pt3?l7 z%T6ub1iSLLfL)n9F6kc+&RG>WYDoZ&xu^w_3teGU-Fde2vOOLNOUHNX$6`W1ee&t_ z#jdUl%jU!DnCXLyyn)tNTw<9fUQQ*tbA1WudiTJoPdfBs_DCqy;0%$9{$yHJ41qou zaJRKFXR?ij-M*v9BccJHEc3u`GySOeV;rsZT*=m~Itu^Y(Z{!$!2PE#kXbVh&yFm> zj{QGaz2pY8d=vxtT5LAkEE&6=mkmQ zJCzw?de>t-{iue#qesxK8zHQ>X{x}qN3r0Kg9LAhE2wRKm@-F53MGLd5K({_U~>-6 zoMx2P=uKAN&ViD&4*Y9w#ufPnIOODfd_U|S`b7K`V)K*1_6*|GbH*@hS}QD1JB5Wm z&R7{}b1vPU+nf`fh%x`Q)1OO`yT9J!2_LH=8cu_4E}8S?(Cx^N`MVra{drRXWG#hf{r;1p_Kz^CbIYhITts z&OD4|zJoB}ZLHO(%)ii)$8(1lhC$F`J(}&ai|R9%LXW&QzV9A`=3f=*PT5JCSQkrg z2X3eO!Tr%ot8&Z*p5w$+2P)bbOj|ELpecP+L8v5eE-kUB zO`!6@H&A2JaN2fpIjuWrNN*2TlgzX&;)mcJ>{RLtJku|Rw0@MLrTJel))^1pk#ou3 zwHSZuOoYn#@Axpt2am{&V%zE?@p3~cX5^WIeRVgkso}Gbw{?g&&~9TGu9SL zA4A_^6|jDCC`84pV$px=*;g8bp`YX+mUonvZfGH$Tj{v$yB*F-y#_CQd9G!o1b!~L z3Z>hp!=5kuU}?oZe3bM7E$(ViZ3$-~S3iT8`*PIYWlwT_7s3JMS4{2TV0@M4MZNbr z>7s8rZ7H3^8TAiwqP8uS@^{sAor@%w8AwTaesuO%2I=WW(eOiuaN;B0Tb#EI2HWSL zBIgQb21Mf0AVWyc-XmC$=FVFISNF)#{TDU( zufGiaUZX1NRu|LSzdPadyHRLf6AHt1J=pJ0=E4E}cPwOID%EhHftJc&P@Q>&%ty?| zLv|6k^`|=Zt(Ks=u}4|VzQ-p~ zZ2MANRGT1-d$ABF@VUQR;tz1o5b)u%H+**!0*Z;@>}CHdSOkk{{vTaBF={VfE01H9 z=XjsQ>LcqPr-_fvdhuj(Jk5_RCe_hu)aGkU-@@GK%+fd1s%ZV4?L?%~{D$0;tz znD$qN;oWDOtfB+V@q_DB*sj*a6jEJ5g5Q@bHcz2D5lUj9Fo2XMc(Tn~xhHtnYrOLQ z4vtCr0m?=pY^&5RoHHv@EY&S0yEt=b8c+k>mPc8DmLyzUt3uY7rD#`h1RQvfj}8k} z#NQ=*M2+TZ&Jt;Y0M4iTD17IRPI(H7h$j8~XRPg?ANB^?Q&WRIcXZC8vW1OwfA3Yc zA!!u47hXZzH#0axL6?D^~B^-IgnhtyBlFY-`OzzSnw)=eyWHzSZJC{`ad1wJ7tkV!9%F^g8pP%GapF+58 z&AS8KFWVwdA)g0vhUpGS-gzBUG!Mge{yBYLn2kfU1d^IrLY+Y?$??K>oFr}(#_dXH zr6wvEd&GnqG#zNyo&RT$^yt3SZ;=0720@{1uwLICKKh)*8-6-;^_)9sPnM?+%^mc% z>K5!WIf!rH$&e_UgtDufa9&#qYb5uW-VUW^|UW!`|3W!qA&D;iFnV{BteK%C6l2ck<5-xy$0z zz$m;fF%#GHYJy)A&!28(*fg|^o&79FD@V?w9KDlZ7BrILOE*!&vt%~dX#$oO@Qkyu zG@E!Y8a9asVb8V*h+4mz`t%mkJoh5J>M)Svhndp>wH}zem}t|q$;2`z;r_92u}c3D zCJ%Uq-dg7L-;6elw;hiTYIo6DwVS(7V$e5Mn{Kta)6>Re$h5zOvHr^RH6@wW96d>f zuVcw}G4Cf8E8;e-D(v1@Bur|Yj32v9sp**_Yno98^QO6>cKmRVzr)bU=v>r-Z&vrCm(R0!X0V0YM90^BaUi`JzKA$4h8RLwq) zbtA@5d&+fi4VwjFX7xDikULYKtAj?CaMR(yf%~Grbr%|_X>f9z=vi}20Md*`Ggd81vphuV2f86g*CtLDHNDP?HCHEy2VGpysHiXbj~-7AW52O%l~6c0 zkLRhHb}{o7tXB5@w#9kM=b~_fb4s&48;xC)F`%QCmF{UQ2B*JEUaM*dbM}p z_R}wgqy$xRjov|z-3viPxjY5h#HVP`Q%UktZB{MY12HrB77f1K zPkb!wp`{&GgsM*LLBl2-aQzYf)24J32lQd~Y+Mm(L#cl86!#T~VJ;*iljpj_7%|H|)yqRp9X z?F&DQ_z_I|ziNv9x6SB_Nhl8b8%MWXW$D%2r8Hxlr8s7_g;=6Ko;-8UW8C@}G*Xr% zv+vn7D_emc67SZgo)R@*?4-oi13>-C81X`6DQcK2LiZeb&}@oD+s+X<{ZlHo%NWwK zKbi3R$0FhQHiCJR?m}NzdGY(=czU|D0!?n*WZSn!hy!ZV@!F zx}3@>fMi-27z3=wg0_tzWl|l*^n%vzr~5A^!>Jy^mFF!vuWv=Av*=>|oE`?pyVa zMjCBvAUdcki`SlU=b!uo*t$4_`$$O4`CCH1Q{$}q7#|fx$dy-bJOu_8> zcybsrkNMo<43iVWQ>eFu`|Q*ivq7@F0# zPFS}82);0Pp$NGo8q@Cz$;K|{zLFWVOj4a>wOuK_T$W~Z?W2%y=HjzavbeoyEd&(q z#TUJ~)O^Q`0z9_hK1tQFr;2}x1sVg7~`xNlqx zwq`CkaOG9di`#;jUV@)$2h)$CBB?5Wfl$sh`djA05^}DwJ1!>hi+hEf(;Lyz!-fWG zE~0}{g;>G&>+@T$<4YAe3_bE6{7_V2*2fHKwz|mo%tNR^LmF>$PS0_(8a%J?08dT- zfi`OK_|8}pUvmEN^)ZF;=b8uH;Acej8+B5<;D@encA}yCdhP=G0s|V%u%ve#+!#I- zJz6uUC2cz;8Jgpz3pFT)O3_L48JIFq77mtM3e|T%fWjjQtkPS+R_*P<>`akjm(3*8 z%KyN=s({UqI1N#U2chMa70^^_i4%Mne((svux%=||9}L|@iZjcWed@N-7T2-J{9B& z!eLsF4JCytig`=YDU2ymRmd8g${h%-%_+P34t=_?%njc3{`o=K4 zG`0aPw(UYSy?Q*$@2S;`Byg7WVjOW$mcmNfFk^ZEO0C+AkKK!rIK#)mR}UOwtkEXy z77h#igCD<2QJHECN@ab)lr978P)gvuu^ehUp+_;=Td-llFR*cW26CqwaKZQvl#coc z9*=s!Er4fWUKpS<23UEnFLYWlKXB&g$JgF8FPqxF-GIbpPcof~9J(g!j>apBA z6Zd~lg&a3+*dTmo4d;)chM_BsXv&~C-lO{W!wW+d8&Nar0$z(sqFZ(?WGUf9*z*wD z?M9&W#!DEkBEj({e&qFgJKEORP|U%7^gI6_b=)|Oo62l)M&f$#tMJ23uP0NAs@B?skn!NC~emk>$ zZ3W*y0ZEYLs-;{&4cjc+i{yvo7Fcjx^a&J`hN}BU0lH8vj!xQ(9qc|xOUcYIFF~9G# zHTQ}D*PW)WkAp9Y%yb3v(BI+m04hr5KY}K7`GZRZtUn5heQ+VD0N~IFWN7ZBq}xUd}cC zukHw%{foze?{;9r%TUfSyn&-6%F(5SdkF`Z;#8@3cvI1gehk_VYF9)w%5erah40w2 z){x%SuOo$VAMxRfblg2~5x@UyP*Ywq_RNmL$$JET50@gnmGdZMk}5?r1$rAU;(^Ro z-28#_2|mrFoKr(c|5XQmr7BrvIT29<0<#sKa5)Fjb7ImP^8s4kiFmyim%f! zhxc1jD=(nzHQvSIxvV=o>){xmjpt{@()i4wbjCChBQ^%2;$(Bm5Rc)Cc|ULj=i~n3 zo%!nf@#raUPQ3$n@R`JU{4vUdRw?hsImV9UpXWn`7W`cQK9anby&=24gK(Yb1&W;REmBE&vC)KemIi%2IQiL(D{h-XfeT<0*?bdbdn~2k6rj`+c6B! zkU)dIm!Kqg2N<^BLEmfj_^%=boZZ4%AFt17w0aU@lND9w=%GS5ceYKtfL3lZ=*Ihz zG|+i9)}Eb%XV)r|ih~KN44;5Ub4)k~y%$Db%p{wBOT>oUCsY{|jukc6@kNs|t@mT@`?TM>86VfL$JpGHI7Kj|xh@ivt!P8jH(F6z zbO5^Up|X7G9u@ma_eEfcrySXS$U$0`6i}-&lMnzXV4&jzyOR z$I#%FHct4~gQKEGlMeSp2V6gjdDJiIsd<(sX5s<9GXLDN$t9O? zPOJe2jW(q4FY{^c*mz8|IEfpg|HFIFM9#BUL9ca-(KGG`n%2)IeV1y~ygU$B96JdE zPV9jvwL~Tk0;NpY%=?Q^Fzv!Q9DpI{kQ&S9WdaSj(FYdz+R&aoQTVuKE2w4>9nY9d zqYaAC-^rft3_1oW>rI%)%*~j$cOlM+>x)roW@40XxOijYCeeRJF&-1rX>!mBsvkB0 zvZLZzsC^LU435XnyMFjE`v~s9d`O=wjiaQFpi;9BHeHzvN{bLaSLI+#+#FO%sE4l4 z>#%`y?*?RAp^w@zJSr^W`Qt%Y_bCM1*EVC__!;<4QwQtqPSD{)PiUr_9!g$ngMd{z z_~W%d)J!o3O`!zL3cWDTd?&R(k06<|1!Nu^Oc9F$aO3_bm@q?+ZXNrF343%fZ>=>d zne>Mx?~}=N*?m%b=}w6&GwABWlk}RqbplpYk>`1TQo7a+cIE<2U-}U5YMY}9?~woe z*^a)u;;G*o?l+mw`S?SIz|5UpaI4Ocp1V(?ksd!$`LYz6KOBfo)=w$t>04Zxun>+t zJ&*behv0-^)+pKFMlm~gwM--j+J zYmlAe5OnnF!_)^1C&^bHG_=QrG8c9N8Hq4w+!Z(;s6?gz#__(!U3R~?fOA!1YIIl3CzAJZfLq5O->IR4Q)^6tBZ3V0`Cc)TBNZ#_rLbNA3S{cYsLyQh~K zlEWw~aMcjW_`kz^9{LGt_U6H;vzMVPs|JnSvv7y(eC+WnroQ{~(MnetleU?Vdw3b| zC8~-qzn??B)^aF$y@&m$k^z!1s;lu<17L zZsly~HS)APEr~1&(n(8p0gAjcbK>a)bhtJQma#mn-*gnrPM>6j3H`6p7$=UB z4W!*Srqirvw&?9?ilH-?L+HZ?pug1vf1HygrT;$RrOdHtv80eb%x|JT*5B~k(MzyW zr4t4OFT&EYSlWMW6@7Zi|0e$PsmD7Pg%)|@OFwq!gOt+&K$KE7D4p+j~> zarCr6or0(OLg*_K_$XfoliHfF*GZCkn);JMNhEWcc!PbiQN!yy zQ1L`7_W#76DGqDE{OxU6SvL|@mg-WI(>%&H+KG>f^6+)|B+@aONe8P{siZBLc8rqa z-M>6cDUzcZX|yzJzfjpH4M}DTJ%4Zzehf0kOPW_ejpqp`y<3BG z0+TSFd+b)&zC!2HD3tA=i*b`K!-gHI*j0HBwa=X=sjU%I>N=My-fhEAHMhV##FqY( zOrjr^O}N&bv%A(eQ~rY&7~B4qR+vW9uE=)yWLAXAPZ^DIjwFS7Z`krf3V43xM+M63+)hm!`oZ-?9q9Lc6Q((;@ebt# z4C(!iJ+~IY5tVqnr|U=U#^EsHKLZ#Tq6llAPx!w(8~!vFG4UN^*uo$x;w=sda~6Hy z+$FCQMHpdjfWg5wq*G@}pIZD;&d(2*hfT*}oHLd`XBx?_FNP+5W?bj*=PPG#XOd4d z+4YCt@WVSJGVp7Lwjs{6$xl+e%X5B5`P?{gqA8|Cslk8+-1Xa1nW<4dADO5 zW%_-GN24kr(j);*Dq=x?qYCCH?!z*lAhMaZhu)luK)o3kF?8`5{4ul%@*EOji?KQE zp0kF%6glx{$8d4ZrbS}jkTkUC^PZesS8^Pk1;sC}fJL<#E&LD+?%mqB;bcBlg*xCR zom{%{D}^+>U*PPa+%wiGLEkMolgZ>UjxN%rq^Z@=q`ebD=AU5-C)~(EOr{&0bL9VR zlz3a?fp~nwQt^Fl9lTWL{ezw=yj(6pYpgd@Zn7!$^V)+4k9*Q`pYgP!x=VO8KOcKP z4X3v{lj%SoH!MCenf=-=$qc>r;q)x-6T1Ei!vid6tyTjb{5lZI<`cL?9AxwMO(Cl_ z2DEeMY3x&f8h7pGj=jm8S0+rRq5B8Z`Mvq*>~2qIBSxZ(M-4dL+=CuFL(rs01LrBd z6R%V>iSesK#78qPlH-F!GWANtnb$0^wPGo>!en?{r~`#-uY!7vKDygH5UdT1;M4MN z%(GSk7s?>Cx_)Gf3yrNtziSsZnR>CmPl}KY)Q6+~l@K2l1xeOp@ENYh?+(Fu_31#K zVJ=0P?g-xRIYJtHVrbBmeJFi=8J-q`vEqOiu31IcLI>f)Vk7$XY8Q=YxxssW1)Gm@aB^AHAXktFH6yu0W;0G$WbvmGZ-3d&z4k1Yexnx9{jm?;+J zjODB?8G3)Of(`{`(M&$q3%}M1POF?*;l|te$oCh{cN&W4y=(ESbPYaG{f!qk597SM zX*7%bl%($`0_#_W3)A_2{$(gs+N{9rS<$%osy=B>;GHJ!lY1GMC!AZ+ADr8Z;n1ib z_(U=Q!`;o1x#m&G)IHSfF@Zk$?!mStFHuO?PVxCAXkpJYKXtLF)agzgi$9~%)E>C- zJsQ-NEYWqcBgS4IiSGXQS?&!B_GDHK)?d{Df5Qm&qqrCk9esu+MdwlFL$1)e@U+l= zTa9I3Ervqr{*ZPcOc)0lw2^m`6k98?u=+1<^m#~6{N_^8(BJgJubbv9=RUmgJZpTo zSRC->l(;6XhB{XF&==V>YBXO!!@8U~b5fsv(LmUw;E08LR%7~|2)5*m9qcf0f%MO% zIBVu&?gEjawXPS051)Gly*5jB>VY9m$ZW)VF>zG{eaN2DmU^d_BI&_pWNO>V1MLmsFd}X-?tx( z{GN@kKVKl1=j#aXj}vkxodk^=CFqx;5AwO|@NpS`$H+Yo8{hGL&aoSu)m8{wgk?C9 zGg2mweabZNm}5@;A*_gyr1+MbFs-j5T{?9VUd36#+LSyP5`RG$Gj%Row0tX6Yc9hJ zqvgragzv&{Z%1A0Xk1-kKp}IQg}DzNvf3#*aHl#Ja~ihO0j)^tD&Xf8=Ub;_n&6?( zC}_K`M}4_pxVn5H%P7_Y-4t28J@5vsy~BBK7P2(xwH@wQ^AZ=0)}RjtCd@nN77GjN z2J>g~FfQ*k<~%+|i_^{u?^UBf^w^C3Uh6_#w;yV%2JwzLx*?G1*GkC25X6+ZE_1pPl*Ea8ARCEjY!DJ8H_yu-9unWz4w<2bOVO`jroOJDBq{ zmdv5SvyPxXhyl8 zr13-QTHL4l53~C`hcOnz;FUuVm{u)jwo|JB+cPn!DW60wYq4!LiOz-7q2Zb>7(G%& zt7-f$o!f_`uceUet;_iEpC+}LDp1zlP_(zLLXC}_)p%+FhAo#S<>jBD)@F+FxaZtJ>+{rh#I$*!a1=WZq%CSIpwyK04*pIza5Re#+6 zpAz>LreW-3QzEyyr2WuNoOOMbcyP^eI%&6?^a?*?YVQr0|8EBjU%nh&H>WU_@HJhgWtOjkXMBa*>-Lr-N&Yo)K>$VP7T9>BknS#8U8pVdKotP zjsVZd@A!lF(5{%msR9nC%XUqc)|oqnB+XQFt#xIdIN<>87% zWlD-^1N9-X_{V<%mb4Uto6vz@J!R-axC%ARIt|X3hNI;We!h?8p2){>q^3}dUSX{e zw_yQTs|ujD^#LTj2tuQaO&FZEjZ}EoCUn0#+i6&7Rpofps!7TKYSz}VU)f7=de|*+ zpP$Bd)CJ-3H_j+LT#k3v1YkdziY?`A&6A&!Xo;YC{|=1L45Tg2x7HekySw81K-2sYD}V$v*gygP5W zAi5l6VLFGwdtM(bd902OqaL8o;dy9dRE_S#*5JvTy8&#M@_y(vbU*f)<<=_T-7wBO zjU0<-{CZG#D(4Ga+yc5E=fUD1^{__0f|(S!;Gll|9LQM7rswZtf%Pru$@dabCo*U$ zn?qmzFm~>K7So6>U|*%vE#J%6Sq|>q$Ff(QhaxRM-0e0D`{!wk>xLVMr*9viER|BG z`{FT*ZU>0HH^o0+K9P)np!oKJfiPQbIDD+-o!Ckd9~%5dN4_7Pb^JN2xMYPnRwD`f z-egAiF5u3-+EaMvf4&+jGYB z?EW&*>D(i6t=(RsrXrZB84aegncN+>%#QO%9H^hg5%A{Une2%#F@B+M>QpJMz^g(~AncUrI&bc|#;$!XeVoUyUQa@Nso5q>& zo#J{S%QKg0PrHR*7S;*3PP`T>dPlRSpp7u``3$6yhhS)f0_dklL4(glxDxjfe+>SD zYcopt+2IYFC<$JH)c(P?A+Udy9j8!M-g>Qe+ z=dlhnDQJOzg&q#~^L^01&t4zDXs#9p?;FMOvlBGIf#Yh!)Nck;0E43Um z&kTdWmyQ@DY{qUkHOTt92$wKLN?xSKu4w&b)d6$hsZ$l(uyGUa-|-t?F0;VH36YrL z+<{ZKwZVLEf0SNv9)F9R5!JFBQ*YJak;oF$kJrr4P#O9G7Et}hR!rQ~UFyOB= z82!lrvAr+agxFK$7!5LZeu0XMd@&&ZBnEoi#o9s(+Ve`FTZ@mj}rZS(82dWiQtvYS$5yb@qe7Xc{En<7xz!eP^7_BNoB}T2>020UpEPjMAJuu zic+D{JdjEuvl5{)lrp4=REE-oA|gskgi4W0lOc`zUEhCy&sxtPkG1^mtmD4V+1K9h z{d)D}MdR--EsE0IPS)iP)EdqI{H$ZWCpcig z43@vI0NLY9DTCKk>#9x2XOus9Es();F`Nlt(hrve>7mQDrKmfiSO_}46n<^;$KEH` zXk)}JIy6+9eXR|GBflTuq~-}EoidyoV=b{m`T~}@q@fRIW1Tqmk-v8aU~kqn9F&uQ z6YtGOi>`-gmo|(B-JJq2@63VWTd%_E0S3Zwr3H|@Bbe{u+mMOSgKMb*0uEV#OIZoV zxoVQ*;abrR^>#sMRmEYFeYju$C;m=9h&7@J@VD{CghprF{MQ4Dvm>Cj#R-+R_M;gG zKhpW|fh^ById>kdz|)=APwxSvYWFgK#y~4=1@9QBRz;P&IKnj()HMU60;@$SpH)e{Tf-*}D@@WX!=; zr|01UlHYDd7x^k2kgHGHPw(-)ND1N8g+ih6 z-5T0^{}P(svldRLCE`kJ14NX^Em4W2k!4Hk{aev%E$`iWoraHdlfe4adem{}&zL(m z;p_a9?9j&JET{iWw*6u{Jw3>IrvKjgM%iRnKEcM(!Uj-k?KDyjO9w zU<_KUnaT5M?{K!E0xn#pM1MGkdWAgiL;Fspzqh7SoU=VO-dKpaFcJ!i4+-VroLhQ& z1D$r4BG(oVyj+(6QQ-#g@`^usUD|~ww8vxIGBp~wT@CY=xkFt_6b4nBqQA^}QoY2o7qbjOEhv!^J~-#vV{Kc9P@_&aOe$-6k{C$GKrd6uvxhnCK_rxj|dWcX_^ zOsZ-Tg7T`NYN-a@NqEQ4{td#C`nA~e)E`Q#lOXt#h*`cdW1dExB)l5QqWfMGE%KsO zmB(L2c++UJMT@YF%nzw96y7!0FZ^E7CVT`=cK7Q>(=kuWyk z7#KH)QBAHj+3H%6cIQzH%*+yOdRs)Y`?)hqG#ST6`H0^1sZz3eE1sCW567R>VLygV zB&&EsdV5A5-`~;$zX~VZ_sp8^UIOYXJ&T40YuWyo(d;nKL(VKj+`H=_u29#*eiQH0 z%b}dhvfdFNJ8H24zQ;cG<2iDhnMQ;9E|5dWLu|bsKt(MZ$a26V{$0(7O7DBHT(ugU zO^Q%#=!?6SxPnIZNt$ALCHSOKtSsR}nI6&QE=AFS(oP!pH;_6^ z#?f*2be7x^!~R}h&)T%32oprKbh$s;Du&W9^{M0-e-$?t)S&h8qd1!L<_lNLV(8YB za3R!Ebb6Hrs?E_6ehfV$TC=m9)ny%K1~%GkSL0AhmNi6Cbeiza&>o%z{}MvWk_2C} zA?w?>Ft43IQ*(#0*LEK$swIkY^ZS#<%?h}&>;_a!IY|5D&1iAVdAzSOjn;TQhnBnJ zg+|X+_}4cbUY7O~#id42_t+nt`Pl{*^{+tX)n*voQ^wsD{2j6OEEowqr3SGHVn}~CKHY>JH?(!g|m{^I!re}nssmw@QILHxPD9ow0D%k@v>{sxat9p zyQ55=$K@#Bpafn$9tb~^PeS3OE_kY}3wrY;A!LRw-&JnKQ%~lA^0WgOb$=@NyDdfS zB6-}fH5;v-XW`ZQ*QmAp2`YZeg1d(OS;d8sEaAlo+8|N{U;m-FyZts!U%n1?c|HDn z{CLRVKBmO!(#)eRicl56d@L-TaF|Aw zEF--)VmxX#id;PT`|#W~*uiK1eO~v3zQd|QmuD8f#VV*U{RfluH{&SRKD`1Zz#X3N*mMH^3&z9dEOJ_q1Ho$2sT(H2wM9tw&{{e*ra{^I1H?x-n~ zgR3;h(w)ajIN#t1cqp90f2G^;eaH~-mOh88-6Ao1(kXC05sWKtw+rz_8lsN%)9I;C zBGCxnDm1K34cFmN^S0^V4B-EZ+xT zJw*80HJGN&l%$}>NIaA^AH9Uxuy5*U=v`)vFGhS6y;XJ)DM%TSy3Ih$9O^B)F;Iq0 zvyx-5<`>y^D6nx4Qpuxd3%NZU%8uKqv$Y=#*uaiHx>}e*15#^n#LF~#-8qSpUW`Qt zgK&KF$OaQV@1o3_V6?t!jdOYj@_qJ&WT352+Z2{#)9%Cc#U>lycy+>H^?}rwc?C*; z#|a6OIXC1)Fr9X>C)IC6vp0-EjS=qP|6nRI)+UPE7iVe!X=dy0Ma`^~}T6se>UcC5_gsO{ZbwlM52n-1rU*vXw=}l6tQdH{HsFXM*;Ls7vIFd{B|>*-&lRwC7nS#iKw;3gEX^)_C=w+?eP z6ktq8vCuI$QV49@B=o#2!f$*(c<@~%q8V4<^0od{>nlr9<3`{$#s0V_+e&!n+YjAe znBcF}-B@jR5F_hfz*sdaRNPyM6-DJZWG_E=b?0Hr?GUoe4kPtP3M4yl5N#MW26C6A zgYVWP{uzH|R&U`VOs>*En?66BJc92aj%~%}G&34FQwI}0F5yw_9^rAs8+_fPKyBx8 zu=A`V|K0}+`ALbBo>_7p<+!n^a&m_dx+@hE>%!?@)B+m6%Z%j2l4w`-O=QUVMMIZy zZt<4&gmvYZS`@?U+Y|Wn^=}+Mw zi?Z*7%q~UhT=E9j-VDOX4L^jIUR~7WJ7qiSZDG>l53r>-5Wd+)!?W$)e6}h{3Y=lx zQFW5+7Z1drzv6|q<0a6%ITQ|VFBbe>-a^+m6UaMv8RwZ2Ne>>5AOCI>bvVx@AK5It z_w6-g?(a{Au1_#$@k1~kH5taej)TsDdeqkwD=aXL2ESP=LAhrl>>Tz2A7RNv+A3Oz;I&N&RzW8KimTp!zy)?xbcQq0@Zg=RMo!(#Jo!m;tU&8!{rg}hdI zTxZ&h4WVmcZ`yiL`l5|Cy3>W<-j^}&gCk|mU4W~&i*;r90<)quAMwA`SZrLBE;Jf_ zN9}K4U_!uXNEQvCp-_}>e2Z6i4->W4(VCNk?}$~@V3(g`;2YE)X(MUGVd5h zXGIHFcU&Tgob{}_@fa)9%E$LVwJAhoh?RU7!RvZFe)*UOJ(E)~|9r&il0WX*M z2;pnavApxi%oa6Szel!YUcZNE>kF)?-;RrRFGoj@9bi)EjUtB-e75Za?A)9Uvr{@? z`7=lSGNuiT4kW>7TM5`P{uAUKtrg9P`A_h-2*R2Pqd4zmIBM(;hT#f@ka@5VCQgdr z{oo+%{;&YboAk-%T{Mk)*n~X~jVbWUFzzR~f!2rk?x&g=PO<5Sk#ncgKnXokopv7C zoPo4PXElBkf(7`1Wb$ATZM*HyxpJ~x&}RzAe>Djn3nS@9#3E|FXpDP<11RyaK%)1t z7`9DUDWK%9$sr-?>-tYdGo*8cU`+qv&dV3I12Hk=JS}P`7yi`F&O; zPt7P)w<8SvG64&D2F~|&CxsRyQo;BWq>);Im!I##ml_t>&}BxScKf47qbF?8bVe(! zp$HHKCSUAHr#h5YUfaXDjAppnWfhhc>Eqgb&P@*)MZQlIY5&a(%w6HcxyrJ*+a5Tx zJRYx4_d#uOpAFEE&O}W7tAUH<_`LCPAzG_w zqAV}-*G4Ae=&TuNo^%m^^Yh{AS)1tX=@&TVcLQj8&BQy8HE7F?64dAZ?d9$S$$e>i$@Hup8U%4?_vpEO6HRgr5!+P@i%zDRO6;^00ICFEE_+vjgdA zkrHEvWSC(K&%*Dv!$5lxwC{YweU0&0_b3Ll6_(;TZE3pnU=?Sj`3k?YDue-DqhRk4 z6`HVk2w8pi#`4Wy;EjBau-DHLzVp7D@2y!de8POdVpXh}?1`Z_+Td}ih-TdPqj_(x z(1KZ4L2Yv@%sV_6P8n}SpHhO3xf zMeBFUv1!iA?D__Cie0NpT6fdXNX~>*hkZr4t-nC8sSka}-otQdf+W%y{QD^0nmNY0!7!->v$ z@H0Y=G}Z5t!+&|WSX705?fMjU?hSm-IF3fI{YdrWcTBd3q{^@+y7X~1nU;i6sLV;S z+ZrwW?XQVHwhSk!LnTyvVGfhndzg*usYa6@!^qzK7I#=)fvU86_}NyCt5;lwMJ*9{ z{FfoMJ?_DTTf<3QqDp&|&cmX|vG8AL3?B1w0nPkoyfS$!CDqE2JqAI}VR?LAE{pMx zb#cSS<9JtGB(Rk$@yD+z^x~kuXs*W!(b-Nv_`5ZPa}SNEBTbQBz1Rd=(M9lZ=}ep@ zQWpOCo`eF|{*+<83}yBn#F0T~DEH$$+LRnZ{r|njU8NGjdy_FZ0jzPJiW8nw7)mzV zxr^nyKwlL?NVnf!Dt$McuGV_ewe6L3yGsuJxLeetzcX3uwqvl=0bYMCrZ=MxU}c;k zI&1)n_nm*y`7nC&!;d}_iZsP9jJ^qM@*nBUK3<( z`HEwIPoUWI$8h5;V~QAAhgl;G%dk@t6XpLQb zuj$M>YkGEMHT^ko4kI^LK^fm?IB@I@7`_jnkydJK_LfCLwL>8OJZ40;Q&v-d`2##> zZAag${J{9lJ&akR%Kc-{Vb?E4f&H$~w(F0uxnUstzHB9%vLqh2{8c4IwPlohZUmrG&LRPkswe$$u;$lGu`_5te{WXxgP7wZd zl*9Dl<~ZKJ4}vy)#+*+Uu)oF^q5C&Jk61^OOEk%=ZvgvL5+(e0pANmj|GQ6fapsU1 zn3WNQs+Xn-w+<|X^?e(KM^i+!DgHgOFIS;hU5}zn6ESbs0d)O&9bd&%f;P>ff^7(c z&gA2BkF9h*IukbkbcbhMZ@}TRfK^wglXdZSs-CzPz4a&JwY>#&_{?LN^ko$7d-0Z* zE2aq#=f8u;FY@_}W-#cD4TgR`&NxN04V%&<(P!5MxbN5n2bN8t^RD$&FzE-{HJwMx z1itwD^$bNuM$x<+Mb_Xk3+JY_!uI{65$?IeDK{OVc0-BTR+DgHlz}?Lof?SyH}t@X z*&pH3)f=$MD;#Quqzjw3zJ$*ft3~JH?}-$a+rYgP3-aY2i4V5hu^;!z>@iP3(bCu8 zl)G4@8E{M_+Av$t`u!TRwL@T3ULDN!Q^eC3Pw@q>E_`<^P;~yWkC{SO=;phC!zRuMM%O3>v!e^CF`Wqf8g zk)6J;%$j4|s4~Bj^tXPcMxM=5-`EW&<|traUXRe!KMUh$UdF3Vp`1tHDX0bw6;-ru z5Zt;}@!h#zG;l7X=h{(_wqFI)R*t}$L`kwh56T^U8@Eo4#HI=1=#r{Sjn`MegVRI6 z>-83}t3$l>S6^uF?1z`j^{89z1{U+oLHsB~ioC}eEL(HoX6i-ixO9Y)=IE1cLya&^ zcORs2_M%Z-x{x0hEKJNkCH!X}Pf`E1k;arzy4aX2bUGQq5cU{ma<=m1f|=;r5)F4W zkAb69yU=i<2933kf!U<-g2$40QN5cA7FwutMpY-+&pbyvOy-cA#VAtz7z2BggFtEQ za567HkDJP}P@^AW+k=fHT3?7#jX8qZ`*k9ZBO;Muz$a1tb0z9^nv0B|Wy98|lGLeB zq*R(pcSlRoi?}l=(exFp=I_THe^pTDu{y-I@;s486gZ7}z|WlgTynh}&MVq*Pf9Bq zbsFHj!42^I!8BU;d=L6at-t}nFN83?dee@!0?|ETp{Ut4Ls0FRg=$J;8Qd7nZlB4Z z=}rphvONsaJ)?xPQ+tF}cctLGP$ftT#|0lzqu%RbJ%o5<&TnErQW@ZDFX3udq*Vz0mc8&m&tUz;I?EoZZze_(jA)gwb4@ zUO9&z`Sd5x@)hX)=ngJ@z5&l@T9K8tBL&ZWh8Ld)qpG+)(?klbEwM!Nq z?!JOMo;@Jzdkqdunu-tkxoLE^1Gc`(MCX5(aTL$s-{rOBK>UJ=Hf>n)>?F9pjD(Qj zi}17R1}$yKq6eoE_^g-GKmV76)$ zgcnu|MTra^eV+}b&!Z~T-Sey(AOd_csa}#wA&Zpri)&vrDMsR#f$K?g8}Y8 z&2u>a^sp_I&ldwegIm2d9u*6Cdpif|C=S)}gw2UCdIX;_O*##;o#(@#>}FK^dlE^M3gLQ#(YwtYORjXmd&x;yYC92|`ZvRe zHBaEN{cVsCwFuux2^6aj!8ToOXm4MOxeM2z+aPz05*6a~_7D{I`C^;=NzUtV!<_7; z*cW0)f6tiE)Y~eQ&Qm4{AyZ&azqhbxr!=OY`2>&F#zOokUA#BC69OvyVAVrqSTk@q zWDGACUZm^du}Q;l+tBlzanb}ArYK|Zcs;Z`TMHL6vmoVg7-Te`h5c8wV5v%ruvANi zc82kK<#h(W;=Gq}pZH$D>w9o2UIIV5eTNB`pMc4CNqV-m79(C(Ve}JINL!);S2ucs zatHrkNY?Y+v*YkS)s&W`h{(NZ0G+e zl7Zd{wh(O-3=5Rip@(PMo=h5z@qq#i?(Rp*k|p@dc@26^xDC$B;$gJgaahp$9^mj#>5`J8S?6^&gKN1?9KP<$jDI(F@biw8%+w_}}xi6*ao zCVzxmjSpapvmwZo1mRhgCD`x^Q2%EVe9W$ev@_|Te`PH0dHWYKa&Ev6c_-YoYX{0MI+oS?B+OX%^Gm!xUyfX2(EXmp=1if2XOk-wZp=JOJ6kTk8; zMS31Th~kyL!}3LE;Q8KfqLNK%xGAUr3uHXdQ|T5;UH^`=7k0xCGcOocQUjO6yWz$# zZ~VC45BJOY;?Ta0II^n?FMA)svL)_tL&ku@#_c4v%&A!B&vS7Vb{IZw6Ixbr1;e;$ z_;{-{=3Ta>2{U|Yz-Lul?7JGh0^(p;KRdb^yqkWM|3UfW>#%mqK+JIV6)uk)K(|&e zp|f&3FwbB-N;tFnSnNB;5yHxUNH(5+c{(A{A!e((Sj>E z3)XVX3kW`P2(qMG;dov=4z0F=wIki|NK61Wb&J8Jv;$W?lA>^)jcwC*p?n`JN{AT1 zz1x3qv!oZjn754-9mmm|>U^v}b`;(p;4UMl{&d(uf`T-cQjT8;Eh+M(#^_)&UNDtP z_I!b|sfSTvmojZ=_=msW-vRTx+PL+P242}~$@^zpaP_Buc)?;F&heAP?yJhM{J98w z@?BBus0e(_qcM>0)mnH|!3p62!pktue5%7xi$)4xu!kfcaH?ij9x5&PgFTB+;Jx|Q zD5oq#$?QD7{qP;rw3p(no|D42#$*|LJ4@0RwP!@-ZkI{25-h>_PjcO}JeAm9ucK(wgkSnCu`$>9qd`j9^iQr*l#}sn*!5u^sWCe#xj+jR_#TrS(EALl$DfQvL09N=bolZ379{96sCV` zLF=SITrjl*-dx{<%>~;acl}V3-DOY9cb8(@`39=Lt;1Hl1NL>!6;$J#x0(FRx4Yv4 z%EZN@->)ZdZKn*GUUena0w0<)&K2vLe&Ze;J-VsO`LLawTXf3<97zlByPW16k#`vB zuz(gl9z!=G!{F?!LY(EPO<9pIsMfDl_?q+_p3mXVLalUI{aziPu27)zM~ZmXa2>Xl z44{X@Lh;zx#dv!aVan%p2zx#U7b@SwUj>{`e8C0Ja<1+(Q4nl&sfI;5Hh6Ug!uG5e z@ZT&wOi|5+Pf>TU`_c@~*U!M+)n8GmsTFw>SG3H_1-{#xpsmp(F#9@$yZr)4Z}UX( zHMuG>}s1NdC;b4x7{TC~&eLtZx{Imxejuu5K$~?@dc6 zU{}G%KVX&@;o$LkEyg;~n9u<7rG-vWm8?T|zSAATm2wjrNYO zg_jYXg292$Lb2uqQFQAEvw3}9Fd_9Utiy9Kcf?9uk>N{!>k>&Mb)5d}eFANsNvQBn zhwQG{(UXA^tRm+nz0_?IK6G>nx>^&^JtGgB);ZFo2brYo$(=KI?&6lT5!~-S2r` z#rLB(2OnVJinDN4?*bYq7|?|1X4ooUjeUhzaivkNncO)9m+BlS9bQbot$f&L`IXFl zb3f=^UoKQzZ9wy!7~Ea)lRNWfkmUG9sFABmy+3!-l{belX8B(XQVXL;))A=mj5Cjl zR#I2Bt1x>{J~Yiq2T!>`Fi*~hKh{b}pAV3*zYld2?}3}T4Zbmyrs^ z1mC<27k4k9XY+TFasDdOk=q1G&rhPoo#AvLK^>JllW@bqjkq~;99eQ!b*#xz)<1uP zaOTHJxMih^qhqp!kn(&u_9qJS6N)Hym6%B)YQgl8^&3 zVCq#ZSbmrQD$av2^xa*dI6DKf_9bD8r!tt%8vrepr+_aGpjufhHuzn_n?LWtSYyCn zmvZ5nLzocU5-qr$`2qjh0^#EoYkc|pDNd1ahxF~fXy;swus;Q*qkGZkK@hbUKNF`$yhhtdDZ>-v;5HK4%gVWcv0DakLGK$4up8NE-97Vnus=LST(Jy^q})hB)DT7kuhSrEDixs(!iy6>anJ9iMG9{}_qg z>#DHL-W#jcOK{YUqj-a}h888~VR!jE?49rjH@1AnrxOlh)-(&;)-H(?e@W53v7Eaa zSkJv4cX3pM26oQo8JkU$$)si)zW=}-LEgX7`R8p+Q9Fr+PDAO7S0?s}gK%Hrf3)Z5 zDEfS{1m`bMr>vmqbbbDEsykqTb-jGAt;39h&U{9vQSwxEB?&d&1*7+D8S1+qf~ive zbaV1M5?6ZD(D$u)^Su^b*|>rZtaZl&mUghHbqk!ak$ zlhx8D?q422X^VE_m{r=a*0Bd5YzuvPyN(>p`MKA>hZt9Vf4qgmQ0^L#NaO~IBGx?tTrrxaEXCb=KBFuYG1-g?>4l( zX^&-=?s#isB^^9Go%yeLL;hjL2v&~xv-JWb{ioUS~K&xHJOaJ@AJz)Je-nU7bVoult%mr1Ln z7iT{0#kD^~*lgDg(%DZ zDW*=e_hkoJRcgSt8vfZlG#f9qYob)C0M4~4&>(#oRq765#Xl94{mmVDW)i6AcwX4E zVkCZD<$>qj4PdqAR(vF_hTn7M2*R8>P_QfmC7+pL!LL-*svC(m`xn8DOm);+XbeBj zeTBI1>x3H<6fwR0CI&0{V^M#ehq!HlQ_t*!vj*vqT09Lt-pv5pX+!ByIroKC=fU)5 zH4OY)1##Q9qy3X(_;u%66sBqlnZBVyZ|*EuR+5G?K?bBbt`MWcwxY7VG}`<1!^C$Q z*b%oC)poDN!!u9em80$$P5!v=sT3{M-;aw{F2_$kp}6a_0?IBo7ff!*p_NpB`uOG> zx@8>0BYD^H(H?Us@)%9Fi)T>$Wna7{_;nY=VM6$EA@P_us#_$W``rPs@njf0 zd#1tn&0;Ze(qAm>T#FUQGtk**B062R<=Hj|((;JI^k#V+H7y>0KKYDglaAus)12?M z>w|Fmc0XjMW_UKSAMR<)fZYwVDe}S_d?IrP)c(m5IPJwXt)szZ+;9|YPQfF*2e;w* z1;~>R!Q+oa7=AV%wUxTDbg>k{pHpb@s}a}Cxdod($AEp=a-f*^5IfEQE6l3krR#ir z@F@ zgO55g6#G3IW?fc6Q6Kk`Rc8w`tn%TUR6b6t8$q7imC5m&1U~3!g5%>$g!)&(LJdmr z%mF_oN+n=f+jeqNvgY%WZK4Lp6KFTT6NmE|CYu)v$wiz6F7F5Ot(>3Vb%&O04~MI( zPvXLlZtxQZgMrgu_`UNWVoMlawVaHz9IeTBZXo&G?S%z%x`g8oqao$wW;EnIc8A5> zohvgP>w?5|rz?^qtZQN6Nj3axeigH;Cvcxj5*)NL!^(&QX#bA8k;_gCD;G|{`O?2& z;=9dYDYhgfopzMJ{ZBO9`Z$hW<4p5*9i*APGdQ2E0C#h5!nVV0Lf{fPIO~y)5x%{+ zZ~Ar2-#?$+wcEhz;y7~rP=o8j^zlxY1Qqg~5SdOR==8b{SM)=1rrTE`C2p>eR@5Pk z_!ooGEz41Inh&(-Cz964N?K)^N&c5j@S3_cKIJt>&MPgf&7Of-dfUNgpgEfNuErr7 zpMtZC9>&hA#ch3maenJPEPTz~H5&g3m()KCH`n}!N$XBvl&K`${A+}{;|*ZmY#CXN zFDt)F6iZ6{?|0tp{!$VWva%BY$B(LnxrFPs)oa&zxf!o<-Qj91G6ge{86v_A85KdiOZ@Nm|8@;L6!QPO1`(OENt0-XjoJUpHGFcBq{S=GqkoYm z-v2#A6rtuVn%A5qYCmB@`Lhm@RQ6t)q&Zgn{j<858v?~gPd3tw_1CE(@(TG#nUK~s zW4foUO~o^gu+68#*!U1{_HRTei+UHqgasi?tI3vmbUHA%OS)`NaWUKBn8|V{TwtE1 z;cWZUX!hg8A-1@?hjP6}vs=p**%g^Ow&Bw&wxanngRnUe1Q}=LYU=ofQ zGy>cIG|-&XukW*pta15iwnDCvg>QMtI?F5APVWmeKQ)dX=tk1*#59z9 zpTzh4x8axFKmM1UBcm+!IqP=-X$nm}>{rwnAV)qiU-P(uZXW#S1 z>FTE;!>UIZDk%wG1xf7W(rxU5`Br8bnT6M96!M+GDExRxPK@cFm})~g8$I`tIQnI| zI3&JOyz|x==6ptlwR&l=OS73+s@6jMWyx%D&$KC`7?&hL@!w*>$612STkwHqPHiCn zx$1Q9(OH~TGnaHXoMG}U*O;YYCTki!8za^%28k01-1RI^+g3H=S$!?hkg7bgeSDm} zlY^*s#uF68x#+t0IPN@hiuwIaVUqn*S^7*L(@fBmIgENIsFK7S$!9SIIl7=emER(xC$ZMo#mr4jEPqbXLs{D zD3a%`M-Azs^l=>&{HC1GrN=OBtFcV#XFJtZmQza9C6emVVJ?MgY=P-SR=nDlnRHpR zv?2%Ae10v9h&E-9ol0qlSQ9V3rwb^ zhBv9`;B)rDqLIxU-^}9VK2UYoBzl$2@0VRL`>@83t=b*LzKpErtRT*9?^a<^HH%Dl zX+;`6Fi@S8_v#ks6d1u8iJve;NuKH4(qRiFBv`5PY6>W{r?k)YP+Ss>DYMTDv;WkZ z4Ty{Z%djV)nOBYLO04OH(JAa&nuzx^x1!6PX=L-^8qdjFp)WtTkN1tI4^o_c4Lln> zx)_(=Z2|ST&%*Lc{$M{mgy|2M&%94BX72)}Deh?!S$&dVPP4CIK=vSNn4&>_lKIqh zy^*SfND3Jv!J_sZqg0Be2cO*NlR^vm$<1ZD(F54F<0DyI;Vd?XY_NP+pYZeVZy}fh z@u=2&SfDThf87tBNWH3q{8Qgk)~5%wci%^L^}U{W;;H50s@xdid({cJ z#r;LN{u5awCXAiEi6XKVp)pj z*{DJmNb;K$S#7^CW?H7gw#Ce2_4@*uWBOZU?-a&fO7n+!i9C6afc+ws-cGdwF$=}C@b6ZK@t(1=bTgfvx!E}B= zJtcN3v8+^omKebKOEb?Bt_Y-G`RbU<-!1B;(O6-rNONRQ(~E&T@4GVs5z!L@%VY&c0nPQ`(;W+IfjS~0n%V=SXL zjqM#dn*Oeep&_%bW5V$Rm{0l?_+6gDy!~mZaWhSQIg073XOPk?f8N)+Kria0*@Di8 zG|Q_77v~Q~nVXqt)~-j-lV*#A*QVo z_LXzl;9D=K{!0}F-{p*Lp5t}Q3P!#5NS@(2iSzcoG<%eKQ?$2lhA8>(9$GK$M+q9I zsQpqG+h@|rdh4FCZJ5b>0cUB`obzP3R0i{R%*1(b6L8(Y+FjQX$q@1q9EZ z_~e@>t+vge=4XqT*2;BkRNo9XRXGCG9&xYX#oG}1Bpw6r$%^hTCbK0YgW2hbbIeVr znKelUuv2jxSa#7|cKe|O&CsbvIoCck_KpSXVteTC{2xC0q9ClhCMT$t&k+9gEHmAC zal4s=ag14%(@dN&mUGL^d%%5EoY*$_sQ83)x;W^5m1tkNwCK@RbMSl?$@IE6G4^IU z^VQua+T+_FgB)yNpwU3Fgin9*5qU}RlOJm4ZcYm3e-!$gcc^V5TLWKG9JY^Io*A=* z?Ah$m?*shZAtmg;(*EyNtteD zYg&J>ftIhCqg)AlWUY;jdrq2d$|^Nc*7quj^S93ZustaV-I<>@WTb0D3c~auG z=yYsfU?LcmuN1Bw-$E}2gixQ2DFzv<_Z1ylyR@R}i@gb~!tb0(Qx};=tA6CtJkDEk8|sD9@C9qt zmcj7vXU$etD4@E-FVP)?W$>*=jd>Ww(#J6aDP`hUw)mE!_=230c)~&wuW*_qP8~d2 zyl7M^Da}7Yu>;T0KZD68^>b%V8Z^1jaQpCA^j~8WDQxM-Jft&N#n}v&^6ebE(>9Vi ztiM3e%J;%IkYjgmYO&r4V#b_iu#nstEb+Q5yUfZ=zbKwDZGUZU8aw_N%w6To2R1`RCwJ6%+nyLS;cy{V?8vFP>lTEVOKsV-Z zq$S^;QdhsZ>`nIyCQP$n#$O{@lYTZsrEAQ)?;iKXK4nLLbh7x76@ujM3!sqq3`ips zt;ekdh2(O4Vs??HPi!W~3eJJIT+6+mEAeRfeYAfj#ywZ_059xAw^0+>1AqQMogd4Z zH2%?q1$NAG<6ib_GCwc0N0FXfGCe3-g$XYY;<6*swDQUa!L7wcG%qwlsEG?Q=~ljK z?00^ciT3PJ8WEmD)4Nm2aNJX|5qp0C)UlGE&VW+>}@ zQev_p{($hTH4?!mm1!-$&xU-v$!^uinX4)dGoLY1+1%lJ3hygM;(E>#HsAS&rN(rz zVLxTX=YFJ%0_$Z&1Bdc8^%pys!PICr7`8JP)e<&Q?Jf^v++;J`5?DFA%Eme*vydV= zab%~0cuC<;VU%C0C2jQre+Z)aoo3!Rr z;SpVO*Bn8yXGP5Fs5v{h&7RfGm1Nl`)Y-a)8f>YIm<^j_!7iM)V~4i1Vo~oBERSu% zRQn(r*%L`SEQ%>(-D7reNf+C-Ttcine-wqUJV~`ClI(l&Zgy$2F%y3*rsg6KIv=1) zpT|$)+0roD%JZMsf;S0XLtoN-A}EEva}Gc1^NunT+q%#3x1&tV=!Y~xA9|#247=J zPq#9s(F57?X(`nCB9^kRuBY<@57FxE>r^`=g|bp(X~u4K95&nxmlT(=N3;30hJ~?1 z)(-4TgbkCQF_9g0^k?nf0qpQ*Kh{!|%Vz&M$2)uR%t#NJM3yG2HXp~{xDFKOm+{); z$QbdSSAI*zNf_@wXSD;`~uM!YuAh4Vn@n=%kI82#!2+N-=WJ{h*XWq4gSm4|M z(eoGSqT{RY3(ZaEX^74&3i6bse>;A$#smGur&qsXajkW1@ZSV>Zmc`gUp*P_PF2D8 z$vSA&QbS?E5^UeQKcpTU$i&4vShR@_vsy5jec7kZ&Mzq@?bIk%uUx}1FS73glV_mW|!CtcA7ua+$65k;mRC3H&zpBbe6#Aag8vy zaurrOmq6N}`|zPuj_K4TQ(EpZO3$ld0k!r2A4TWkNY(qtaVsGysfdh(lzs0#=lOh& zh?G$@C`oB4Ez+>H?7dPpNg|P=5`{!5qDaYBsV}J{N-6O>zrW$!bMEuJ$Loz&M>}v( zn~!k|u0=EPQk+%%8nle#!S9SR+?cSY_V2EwI{8Z}3GZU$UssCS?O9m#CK$tHc4D>l zO1!CVLSEkwB6S6a$%?IQm}4l;R96Tx#n&64knM5*4L=WFt5B<$f9TIBs5Ejj;xqY-Xve4IR_+gzF-F=eJllk zl^PP3-9&V3EWmkC7(E@kU_)v#yLxtrvt%yM*AE1<-fH5^0l(M%wJs3H)&E z1U`#8i-r#cAS7Ww`7gtV%t`$Wqko&=%WxN{CC;P`=B%dw8|BjfWc&0kuR?C1iZ=Im z>K{Gv3RiBY)Gp32(*?V#-?BR!Z$Q)WHe)50%jD^$Fnj0su={Pl=nIktG|7lVf5T8b zA{~uU`tu<+L<9DYgb@4cTI_1c!(nbF4uAQ>EJ+(=%H}mO%SwWXmL(quzfpioqf3~u z2xVq!_97;rT%78a382O^?5W6QD{;n(W%zs7cPQ=4yW3P}b5*o~+9!hqZ)p(eVWM z60rudlNETIzg2S7n&R{-VmTPhG7Da{F2|PSIpk@t0d0IvhK#RW#QcftM^4;DT<2&6 zI;|f`!=FQ}zhsC<)-qUBXpFXVtKoF^eFzYG2RnZjqLEz$+IiSu(RD;6ugA0R>#*nb9qfIOf?H=L&wM+?RtTx?S4*dMghG;L$}Fmv{&yuz94~}X#tXV~EWtBZ zfN{9?7m?S5F|DGgvsn-yUCajY@5$V@h+o_h(;DtqmZQ60lcM-!=8!n;d`wm?!TE{x zIP_cxI?c^N(<>B4KCk8RQZCXVbPv~awgPTgNKomKAsRY8r@gD$&8F8$u=;u>G_|$E z1M3+WlTirODQ}=&cPsM9X(J-vZ5b?W9w}Qo;MfW=T^Y+ z?dql`ahpUT1C?L`>- zk6?(eFgB-t0-5NgL}U6m?UZ9lHt~hA`=*g(M($oBVtA3O{e4jH@%(#wLDVGtRC^4= zORvDhyNgUwRtckE$!a(gGzqX$BoEh@Q#u#QF}MC1+IcqO0371n)i0wit_&rk;Q?5Y za1M`ry@rJeUCiBsZ<#p5dM1>+h}<8K;x%L_&}AKp%s=hZO!zQtYRd@3V?Pw2Y0LoN#~tTn7=9=dSAIg zRf;+o8t8&|+I5J$+639xMbM>41CI#W0u5VGfaz!606<8IH$|L&G_ENHq%@*a` zhjBQXe{{(o!u>Kc5yYg=Kz7(hFibPXJiP(f`r$Hob1Oi0ipy^8xWY1<)iC^gDHNZP z1hva*WIpF3ky^78cut~NuU-OtbHgC~+7%d6tcHxd99SvCc2oEn-0Hste?KxLMsq(C zm0RawZGAmR+zbFU(K>QFX#ihlbfS)NA<8PR#g<7ce4c8BQcWYIe)ulQk-R{{cP_>= zBO{r^<*^)(Ck?*^p_= zuaX)eHzN95i|_?(;R^nh=Bb_*;vQjpRq7XX**;SoxGVib*(<-0F>S=ns{hgQmT!5Q zqw8opo)6Be6vePSPk0{@OY)`860^B~>3K3DnBr6kg*)n4{WKf@*xbg`UT<*Wc{x<$ zS3{izviPo|8ymkp#>uC3=rbpn*B-E%TAq2BYFI-t$?gjWmt^FB#cTsq3xpi<4Xkn)CXyBBcsbuSEu)!{VT0$jisiqguy(A(Su znJOB1w%QMdx>u6P&9msf5t?>B=SMb$t%pOw{qU0QsC!krfI&G!K5aQjU(jf!WoMgX zmAnf|=QyB!+h+LoF&W;BgaAF>M5Lll0mL=KKo1}8`1cY{S)2zS{q=Y@L=6|-4}p!w zJ`gMIP0GDXNdKaPu;0oYTFy&Axyd@Ty*>lmALoGAd=q@t<&3Kw_aGZA!Ul%`EPfe) z(F+-Hy{Q80DG9jm{+AaaDaP$TrbaFA?Eq7?_i)GiE4bRrBDqCy)?*9Yuokh-jD}A( zYv7fGJhrsWK)#xXP~)`?-;Bkhy-65`kDu22D$=HV@Tw;Fa^O#R93BU*3GAlLz7{+o z_Xhh`{=%V9JtB2*53&C~7sR)B(R;pe$)83m%AxrvWTl zh-BSWyl8S89iGHv*0z=UlP8V!dnRf9k9GTDbCNXNi;@NReZ|a9p*&^}-eBzdHqkTp zO;UY(B&di@)A5?eEDX>P$MrMYFsE{T7MuRIJ1$cY@U+`yfHj1CHzx zWS)x(G0*hJ@v~qC9?=azQ8@~;lrEt9q8R*XABaEae1bqWUwP$DIlIN~MHWdkQH^yJ z*|2av%5F5l=9!z3ugjh(cx%oqOyDuanJbyKlqthpv}48{Y>Bwx2HMT?7bSHolYYS~ zqI0YllaQpfNIt2ewf`dIm+wKVf>zYe9>HxZ*5Tfz?r7=A-o@1JVU=kDauYqUOxPPf z7PXS2ln%UF=b$I6vVpUP7r@=>IRX}iZ2HRA%bw{6a)3G2VlTr zm?R6gk$N>QB*&0hEZz1{wKtVEX(P$sdd- zF(GWW&@PRRPPj>TNXF8Je$u?s>N@V>OV!-&LK;l@CBRI}7-lEjU=;4fGv5+@nflj) zETPAStWVdZHhnpQem+a!kfS!JRGPt__qWOBPyJ++^(z)0w8TumEtnaH%!fjA#(EH$ z5~(&A`L7=gT)uw@2KD5JE)#86MCP9C_XzXkI#b&AWYOBI2qf( z^R*vTn>XV_-$qRA2}1HO9?xBm#@4UVIPA$XJA@Z-=gHmWrMSt$oECLhCV3RvT2o06 zcRyqt@20~a4MP#%QoPhGg|0`UIp-}Mbd@7dbB`w91ivXw)D-vw!vBIW;Z`c9MJA)# zl`R;yeh-em)WEkhuYkq#99r|S5vh5yj{Lq=MVdFr0nKy9sOVkjroRPiCFCJOSpc-z zJZRTnb6gS5qI+!sQ|rFMS?5ytwpy=P+~)NV``KYF1D73; z(=QFX-LJsBgeZu;6#+A(4};OwE8GYLagMlDC(94}2OWL$P)|%1^A9>ePsMpiHOPep zbx+`~s~gQ2!r6SXm&5t*bKhoVvWZk2A|TuEXMB^6{2+#_HsZ~&V1SD~pvw9P^Pfi1qQA>|N6*I7ZJmEX*)0PKcSXR?p5yRD zY9Ty)5de7b5^>i@OC0Drh3c$6 zwI%fo`0NS*y}_5DU>=E1YohSFLL}ZaT@R8GtH5%-9lZBUB`Pc@MoncKIXQ2{xtC^!OAstu?^k3*wm3M|YSDiOo!ci5tCE^DVuG)eW}%c#BSz;>@AmX^gb9 z3n3;KsZ|2`MDax;xMWAc@=rJ5dBrd0XI>Ap{%9v|Qf`qw#I9&~zNqSG|H=tRB1Y)O$E+wI6@J4#3l2 zJn(#W3QjZjgon|)A$pk(IG++@`}gV)Q)~nx^Y75gN2bF~+jr!gcrux493(&M6+k<^ zmM0$M$MZ}x;{C3dh2^6Kqiy}Jj(cL@%aze8p zD*O>7Cwbaj`U^yL z=nyz`lVSv)Ngcb z+!1IH8-~g=#jtPB4!HE7h4abcGB?QWDfjOCV3H4sj1B_Nunq33jYi%#l{Bx-f}3vfL= z3NO@T@v7u9hU#`^;x(d}TS<%Hp8XgJ!S94OrwDb|2V%?PP#hjz#nEqy_%&(u)lML* zw-$6ZEr)YY!pMUpE_|gdHOE+-0&6N+i2kxO$iv?^NDCD%z&yO zDU|2`3{SisgR+J_d>I#oJb^h7T~ZFAw}tR(6@{00D(Gk}fnKBkL4=+@c~YoEvSTz! ztLbW5{COztFe8KJYx}`lHL#iUPgRt6Y$OI=eGdk&ur08z>mI8S)}iF>s2d+3ax{{Z>K{u1mL*ttxD{E#W=6DzgCG zP3C1*V{-&< z_d^gnK%u#9FNw{_0x32}e=jzKsx-`?xxOu=t7JO_ewxQ}prycmLml~8SU~2T-b4;n z2ZOaO556B;0GE{a5d}7!pd;-tP9l??X zfS(?u(CQ)CRQwEzrwQTYK=&+|{dFFAoQZ`;zeeDs?hI73cYw5iXJF%vlQ2(b29%Fq zA$I3nNDe#5P?r!&hCCjSkW+WKlTjOak0g`0b`%G)95mt3qv@>9BMa(R6`^Cah01n3>lpND$pTqq> zWX6OJYOL|aOshD|^=^QhZ`JY4>>2plLzJxEa*yuijni?@Tjz5wccH@ zNJtI_F1&)rmn=}TR|N~qr7*sb#-A*!<3^ji!%Q#Y3Axk1vxGIiWC)C!fE|9{A!$vPo6X* zzkepr?5ZEnLiZ|niM}R0E?bEEVJrH}g~3(HSV(zN23p)D@a-XD@!~8@3g)F_sasY=V3id7iSDC(zID+O?0#BXu^FBB*&U)AyWw}3Aob?O ze>@Y%cf9#02FWg)0%#Nb3!xXQ$iq|{&{`&kUoUyU3RNEZhDD?CmTow;;4kcC9)N+g zA^o0Z9-g?r6zuMXz?QA`u(+JU)Cwb_>|;Rw1ox6djwj)XrX*}n7J}nuspR90%W&Z9 zH2j?!N415sJK4|q@L0_&5~OcM3g+m-ZvSZHKE8+V0!lHnFAoM*J%-yO>{d(97`zHl z#H2-wv2KY!3hgMv3nJHX^UZ58^wAO&BN>R$VD*L2L=5}F-tU*P-EVwDYKE6W<(dVw z=!gQH)U}&FbRm<-TeZ=q6?aHNk{yJJ*O3J*YsN<8JeI95!yf)Lv`5+}jub{MIDK=V%vzylZ_dzrn#5?7s0Z=TEl7jb~%xLB0*npFXG|nTDp078ws5X0F$lvA+1`U z92;db-N&ziuN(Vr*zEun^>>i{nWn>^UnP!l>|I>63r2?g;Y%_HGXIqk4f$v=?;8TA zR!Q0-XeL!N6->LN8$-{G5fYod9xC6L(9<0TXzgqrSfh8I?5;Tg3(kIKpK$^#YFGul zIvS*Pr05%Sx6;*)6lqYFCUz$)NSH+*Ij=U%L4RT0z?xFs@tgTzEu{skw}?VUh!XOL zRN)mNZ6=G|B$caaq!SthNce)|_*9`6|C)csTsbcuM>CH*bH6|Dx|b&O2717qBSlcs zJBhFD%$X!14@N6%9@&579`ATEPzQsokdj!4ioN@A*`;dEqU7Hk#cK{+ae;n1EA0@x zmal<@yLZ5f(>I{Gppor`R8f0W`bcEzN07Yx2yT8`iueC*$3gWCdjk(5}CEUXgS3x399 zuyAaYz>0}14ksWjMT*wnB^rVe(0IRt=e0G9@Q&m`w&-~x ztr$an+wq0I6fcSXa2W58@5fcQ${_aH9oSFZ1&g5jer89_9Zje`;`6F6B#W5PkU3)vNoceFL>JCc96-x?aIKE+|$xAqNfnKTV%x%%pM zrCib(9-7Y;dG?Hcn8EHX9{fU_QZpgVV;fXWB1&{7k)ixyQnTwa#&nspT`zCi;y;Em zP;2K|H#JaWQXSkX$%XVUlNkEsWG+Po%_Bjng5*@%Ak8%vCUzW6^5Nzk`attpdehP$ zROp@z-ba+e(~UGLomq;DPd$OShhO2Oh!9put^mi^98gvbB!_qw&k@SYlC9#*w$z>xkvRA~E{4+U7 zF6~ez;V=0>x4n(DM+oU1>*v?K@Jo@y3SF>!o;Nw26h+Q!Z^WCdhEVi(0qm6fNFDvx z%l3l<>0ns{2##C?`G;mRXLs$VE*mNEmUD8cUV1y}+q0HjEiHub6`E+DC60133*pv; zINZ7cVAp8|YPFpp`@9b3MJ!>mmmlGFKgi+!m%5ni(DIlHe#&lcP2a|wDJMcccDK@^ z8J*dHHmR7mmrTZ+!-i{e^jGsRu9u}O>=t`QN1t3kjv9NC?l-ZN zh6_b+Omm`(>VnC!ohkIydKoIK(}8FG#*Iq;n?zam3eop>>Ct1y9ZPOx8as`P@@nurJP>jmW9iiG=FY>(kQ;B{{H92%pADl9$Q+CJZ zQ!W+jxSukbU|-l6_$+G$tzRQ#%`F|6rgE3aKYveGm`SrPg#neVbr_D`lYyI;=YaVn z63u|iG_(c5mALbSl4S2;DPlOU)r$Q1m_&=2xP!;7<0x#s72VFtW3F#G92h+XRT{#u zX(_?U-}b0tBgCxvuZJ$LzE1~iO{FGy^XXs@r1|1&sZ#a5M9JU}z2)BmYJS~gqVjKm zJYajMZpmkfm`)IhNl|2!PF9kf?Si!1Ygu|lzp^_mV8^xX!LU#zMW;@Q``%`Bc)MvOoCWk<*-`@yNKydbCiwO#c9$; z*!sf_%TjIep5|8MKk))8f49NZ4gn16V{oyAE_P)|pfW}36+F_)+N;k~8dI2;s(a_zQ6@3Ry2H?g1P72iUMpa&R{_ZW-PU2vc8 zZanRN3@<7Kpm)-4+)=a-FTSkAE25=%=xim5m>&o0p`-BnlmQ6q^pm!zG$N~*MMhi_ zdAA?fal@=7xX!2cVPxYjtSseXUi)e0hM^y`>X{$oJ=jQOZsw9_S{Fvp&m2EoN{JZGY`&rWx|5%w^<%}1XH`< z0)sy3jOwdI?w_ZIbhIm%9ygUDdN0aIsH_Tj35T;>Tt9gBFa~D)i2{vIS1?b03bh7m zc>OPp%@6f4?Cclf6U8$7Jbhp^X&-rH(Lrnm+KJg`U3h8a3=8kD`_Y%yLb!9$$ zHmQeb#b|o*YGHC(xslFQ{|lC^A6&J27_{RzK&_!KoLtU>HJ0!3qQy(Bw0ehs`W3;| zB>=Sl1;d+<&ta;k1R|wacJ9vEIJ2Hc0VgfQ#MxNh&tS9BQS^+=hA^=(=*|BDN;&Vz zV23d@nYu%BTM3S3R^#v0moauXVYYC!8GA_%6KWgElx{iAI3EdM6yx>+M`J5kKXC!a z4gT~w#~^xia3&S7Qy0#NE`}ZN5gsYXGyaYW%(%;JChLY8GIpkTa%nEw%>PcF6`qCN z{0%UjZw9_ZK1}H@HILiu(cZE?Oy^4utNCBfhFrH}AbtBM96FW>215n#YC;7{ z)V9(-oBuox?Egx$_|6WeIZcT4=O(f zv-kW;^5E_rQgfk>n6Hgw`Jv%toeu@+l_99+ZG+om!p#0k$_hA%o?< zu=P?HqZAat7*zN(i+*o_xno=iXj}lhhEFi#r|g({w)V{BfIC#%;BIPVKnV5UdRBYb z_Y0CgiJ-+PZTu*wgd04hQM1budzRXvafSy5H??4-W(97DPeG5-3y|Vt2XMfH*vgv_ zrzRhgq5q7|PER0}+j;PlHyet}>cH`N6&zUk6ms60^E5xG>q@;{sw>{H5yqBZgy>DR zFjRY%+%d=}YrRL<-NfVgaoi6#xX0rgcC$@+uOVy`FC~fHKj3G<05lrkg)ldHGwQK0SO? zw+O!pOW_6mgZSW7DEi%I@ZgbfC_X<6o;--54wOx2&pb0A^pFSfzN-mO3e`by*+tMQ zDZrTFQ5BWgZ3L495Tw98J1Jp-w9`uC0Wc!oc%{-cJME$GA%_z|SDJC=vv!-eoz zpAg!Q-G=2GZ$R$!T1fo84ZSjlftP3lc5F^dUONyAOM-B+&H)}*%mqnxee&BnPj928 z5@+iS8SbgoZcyw{#B$IZq2Jvd5-VtS+gbodudjonVU}?B)gm~{p6e?-Z9s3S7V8I8 zaUFio;)s70CEk2jiPsAx^|I}BN3|$K^F=~WbqwSh%8@B{yW4P1D)eX0q|=a}G?V8< zfV&shPAXv@7DIwy7N`am!1D3yV7~Jae7QRhnaW9EEPoM^Fm}i1QXZ+&oQEe&C9!Ez z5P#jgK)Cg6hU>ctxe~tw=H#(@1eFbr?B>SIL#$t5CJoEd?1?jqv#c^5vQIXhWkg*h zTvtsZwV_jYVZ8@eamEbp*@8o`GMwc=FDW8kD`L=Vs20z~)Z^0$2Wp4s0&1bl1GBHVq_^_EoKNC*InrA`WwI`6DlS@U3rt^eKRS8DL!KT0$ za8s}zy6T=2)9V~~?4tyOF}1u*s)qg~okL*hbqYCuln$DQU!iY2rwCc zzvD-pKX`3jCGi|xN3wR@C0zLfoEXC-dS466r+Rw@x6e+-%WS^iapyX;5VgRb3J2uf z%|s==)7Zu8*qwT}u`~P*DnHM{?gc#N$pSX}Ft(Uk@FS19RHjKWg~jB$QVYE~{2}Px zvqt}zLwNHK7b|?)m=pzue%=yJ40?JjSU zkefp2)4ze-+4*e7;{sMqs_Va1n5Azs{TE~VFNHeR{DG2)vY_jZ3}K{%BJ<(d3Z~M- zotwnn%`>_cO|5wsg36RXMpWowZ$ci1rH7%V#t~Gn48X5F?)d8CTnxuPd_T7uFN9}e z;M5HAuXm2#w5-d#bwP==Gxe9A8quNGFFOPCBQ3y9AQHBzg|eN~NGz8*hqZ^YIPx+- z^#nfzaQfdJ#LsJK^a!p2iMK~!%e@$|aV&%rc`nTKiG7T<;wGk;DuyvLF~V0_OGSQP zgUOn!Fd=#l9{yDVU(_GNFVhRqF{Oycz7!UF6MRV7fOC2=ocY=X<;_o_dR8N>FlD`< zOA^fT?ZOP7brTvqqVUS^Ky-Q)h~ceUa7mdCXl{K@NmV~0cG^dI0ynDYJ>yAm#GrtD zjuxV)Cvz}oFcbJn?)~4d$2!fc$QdifX@Y<7L+EdeSoa;Nn>)!+3PUX3DG;Ls72XP4 zRi0+{ZQkpHBaoT@7S;xRg>}{is9bszO>$~bIv@nb%TLpb=H`&w>=`{`O&Yel)}ye# zHF`;_qGS6YxM_9i4Lj}P&3s|R`;Z(2<5!a)I^qCqP!M2rI{BFA3x$~u_F~1`PF$YbfT0!jNGo2&<@o~O zuPq9^#UDuZPA!&`G)UTo%^*=S7f*Za!!vBpc-QXd7(1oWBihJd2)u;Q8|Ixmt(yF7k9lvLs6PnV)*&E5?D7g>hYkT5PA zEI>rsYI9^d}_3^Y>psZ&fo?{=5alwLkKTJQ3{vXD!qes9Ou}) z``qoRl7uH|O5UwfCwyZ@m>GWxwfI zoWKC>6EI;d2H)i);Ypf5c*h37k~c@;GItSXnGXQx9N%MybZt^La<&cgBlmL|q-dQ7!3$?lwIdb(CBMSYyPK(DW)Xc~K%ZE6 zU4lo-_9**ViBYsogYZX0R<;PeCuJoD-+Xz`M;w)PQPm1r^1 zb+_sBTYu22`<9c>EDJY}%B4*-n`q&N55(0~f{Yhu(s^qx!z|HG*f3WVw>(}3O&h1- zl-UeMS)hv8nAK30Vl#-DupDs3E6G>SQj$)e=0$V2@$Or{;&$)63q4i&z+V*(C)$6a zkH~b!e1wnLZ9PW#67$KUuWKQ((jJyQ_XoGv)8JU&CER6MfHiLl&^s(0b?2VN@v&`a z|86DnmGWaV>x11coWQKz19*(hci0uIW!!oVn3W+srkd?kom{GqcZv>SJgcQu{T+o0 z)juEw#$2?S_l3Z1M3W zWekl`!iewJK{hV{e%xeP01n-xrGX!;mdpl$^$X~bOGoHkseU9@sE_;fZX|bYryuvQ zB73K{+yojGY=7haDrRbkpSj0ZgA*xsw2y`YMJlJ!bIyDP9;?sosvKv#Dgm@*#!Xt& z;vCC|7G(T<2hr;oydwwlHj`ZkEa=j9A^OUXILc~iBdM=&M(10msJv7Kr-Ida#SdCJ zxtYzpWaCTN&=ZLhdYe$Q)&c9Bj4=1}EcD;SV;Zznn80C4Mp$2ln8eN_8~9rlODttsYNMN?iCDi8v`D7nESJ-G`PFiK z-gW^he%pdCyKQ&driSaAcSG-+uaurCdmlT%X6pwW?t^5(VJuB$J>0?)_>~t8^4&YY zv0n(jZqNqt`L@t?Y%T~@%j2UvGw}EbA13E)#{*!4v+gP3(Fgl+gOffcvpe^5&xc|| z_Zd7Kya$z*Wh19I1Z6UJ;T)wwVyos!9OEyOhN_2{`k{{v1pdItYlpDz^a5-d`2-U$ z5QN;sK#%@Kwl3U4J-rq|=XdQQia&p1E9pa_nqoAVl*2D~o`B=gXY7013!heAh3@PU z=wvqv(@&(6xr*CKtw1E*|F)3owr1gn#xC41>wj^*eIvP_*t~04<8RPQU5Z(mBKlg} z`Wf|U&5Ytn2VVO2aQe=1S9(nIHmXi-LGcVrd>SrH?R(WqSM;iqxI7d?55`3HEVCxplR2X^pC^_Y#ydP=%(K!`#*qCp zajw8)@b1aN=HUYvG))(&`XoGA8iaiYhp}d&12rGlW6D@1HvRgK-nREM2j0%*jJUeN zRz5RuXqXPNvU247UR&Z6r$i#f{qYgo``k}+aq#9gPHgN7-qc|OYOmWuwA7ZviL|HS zQ<8&|vtrOCcq7`@=`zksl$h@#Kk(zud*s$}6=Ia;Ph5-CaQ}NHR2&dMqx+lCU!H?Y z_sXDvmLslh*I{$4zae>G8;Z>HMWsE#IK5W@8{NlYLs1aKH~hoG!+(*XUgPMAb9ixY z2qxLsq2u1Suq~*Z$m`C9K6fEfw*CSgb2yK@6tW>NM3X2vjtS8_Acy0eO6WWl3imIn zGGXT^=HQE^Oz6`q_~V=@ZX2>h!L4S{>fT0fTbHum|5UxG=}&nR0qc3I<*nFv&K9`} z>=sbtM{G%XjCC#9=o=IO#`C?&+0-b~{Adbyd$Zh^U!U;Yho4~L*$$?Ad|~EZH%{sO zO`H=Yxm=;!o1ys4Ot@e&LJ~f*J(3+QuyK(n9`D%*`5w9CzDYe1JmW_P^~BH_{GvqD z^)k$E9tUGVeJuPs9gZDY0R2)sA+YEv^q5QI;d}q!m!u^6T-%O&-D6NsnFB4Hz2IOr z0J8RcABlO+rF`* zRs~g2#WPHCL%>5&Ni&D^KL&8=R3;HUm`p~CE8+I&bbOO!!k)SRVat^zj6%qEX3m~f zu*`Y^beP>HV*64si}eXc9mjyLz?7z@IJDQ1UXIrBub>e;9}l!IL>Dz{lKZ!eR1PnP zhbQl&nL{_q)HmRvUuGDdy9GOi^>Doa+bdnOh(6Y_Mpx{p9Q@9HNumd`$+jn|p!BAd z93I|4*z+hnx#15{kMiJQ&Q)%om4&WmeXL%(0|b-A&~&aM%Tfp+ySk^y??4A& zN#Vd(83F1A)>ykdvB6c8p|m0|r-c!iIwMqAfGPeA!u;6g&mP z@>=M#@IO{ttO2!?`4E0F4Qy`&fk<2psg=A(HqNOd3yq`5sltA;TucQXxwgZf@DaFn zT>x{Pr{f1^DgJ0dTpgPZ1vdonMjC_VDL&-j{Svx-{T9kGheO79*ORY(41743fc##U zvE=*}e0F^vo_u10?^GRdLR$jd4o;Hg+}W`E6w6n#QwHTbh7e}X!9|+#s4F=O=f2X! zwYEw)u}c{@F4V&fAc7M6Uc&xTX(D~gn%uO!K@Y}lCPS0j#9YvT;0H(UXjv3D@b!G2 zu1gB&345{m#QbzJ6i0&_Rfm@K8l71>}* z7XC9Q|9nTmbZ;*FH;@ct-`w!M#%w|73=A+ei=C%AQ zj;B7&efYDEn>n@|rkaW&qplMk99%$VNA92xcV+RsE}p~um3hd?yo)E?V&Q1DGrWTR z;OBFeWuMQ6m9gz4KW>V6ej6l$dC~ydpXu&{yGeylCE2&U27b!Lf*vV?aJ^-miHW_u z<&SRbCLM{TzFKj}9(o>_FEYdKSN^Qe8i2W7hz2>Vhc~nZjo6_=nMRf~Cis=c+I%WB z>nL?f@H54pJ~6&z8Jh23fdjfV^yuy?${|sSei|);rH#ti)5q#m)@E>siHE}*lL6IU zfbC!wj8$jB9yX`{=F9^a{__sDC5L0}h6AXX0(fDD1AF6CSR1--CB#eNcGxBG`U7 z2Dmc<*8kYRi_IXsPmv3FUAo)gU4|Gqx3Ns>6B+3KFA%K~w&A1X0LYY*g{>ZyXW*Um*o>?M>p?8bHL%}{ppdaVA|01;=OLU?v9yqU8YW8aj)ACpv; zKl>2VCsT0zxF@doG!qh(bm5Ne8E`*Sgt#sr>6{!Cbs>1mcrm)YnS=2{nvmnK4vJ@t z;h;z_bbt8_lFOchQQ{YL?QUj$*-R|l9fSA7!!cOF4Zol}+6B7d>L3^7oQxrf_ukT{ zXI!M?CWXk>SXJ7^PMqE|a+$Y2dyG@0FpszEm@Eqa(8f2f%+aT~iy=du%+L7ejL*Oo z&Z$q~yh~*pcpoB-(XLw#%bJ96*@q0C(N(2`XWsd=0bvd?NGu zk}u=nzXW%pI?i0Y8b3csVfMG(U><+F%ZT3Q@n zXh^e?gQnX`7d9*K=BOHtYXG{DyPRz4u8fB_a|E zg~(7!WXh~0&7~sJq*0S(2%(HkREX-4D9IR-$dIA$JpZ+xZ@rK5uJ8SSt#y{wvbxT` z_r34?+{512H6j5E0(0=}>Km{>e#Hwpp@fhJYLz1_jKe@&BYn}R!89)%WC@InJnf$$J_AzNDxloI5Wbd z{P50~atxQfij!oL@zfV>#(Ts~df^x$lQr6k2{@!oT#r|inM)&3v>_0WCv8K=@K&6a zb{8uyC8O-Qr$EKmfX|gSkc@p#H*3A6D~0W}Q${fJW1S9NlRJue?DkL~^~;g!F;Jo| zoT?(<6?SlXc&Xs7^@H`3tYE!wWU_AtaS{+~JE@L}XH;;T0jipB0756hgf2~?dCG9| zpfikQHg3j{(79N~?cc*!q_HnNPO#dqkF$GswLrmoQ5+d5g`Q_z*|wKUS^E4;wk&h4 za8BYj;g6DTLCcR7SfRBD9cB$hUl$7q+Dnm-_vgZ^_2O*()0gOWI|1eL?XgnZ1VvnR z&_X4MjEHUK)0K1iBfD7ev3CbmjtDG$u@(_=8BJDjdS`tza!@273ooDEkAoKO7SPZ_ zjohY0&pO6&pe;7S2G1vuKQa*o!H2or+Gvb>P1uwa!hVY%#vb7Gz0c1|V`{IN5VqnV z%xcev#S@Oh=CUdH_9o)?bz?Cx)Eh3Z@&mDmQurRJz}>%1hkKE$A*g}lKm6VRf!Z5E zOU?`PUQNOBbvhVoF&LvZm}BAh$ru(p2l=15do49F@VuJE$Q*b?dF6XChh4VAl9-uf z-3)E!ui0`~IVlc4+;@bX;}fu}EFae&E=9K&%FO)sn~dA~7;s+c2g|kgfSsN%RF(D! znZAYek&B)5{GVfSh&m5XarpqTN&Del*>RX%*#lE}qo7)JDm^{8n5k?l!yQwuq4c#X zoV~vrPIYaA)lFgKqn9Ip^um?GK4%H)ghV?Ob8Fp)ywPAcd_43n_5xd_W_UI!fc&bf zqIqxqnDpe~U~ssDoY>Y%{Au13j?k*fV5fvgWafA3cydC9< zVrMtv_3>-5tho?=xm|}WV_Ce<)vh>Q35MqJRdBgsI*z)cgM|)5QQ^mMtf{v^!wl~3 z{D=oB^0Xo;?VHG%Q!Grngs^G87Z^{G$19>wAlh#^L>lA9fq|pC|^Ub9Rs}Y$S7eM@jLZ7N{)v4!4VK zFjm}&P5b4^2G*`&>%}S{z5N&*{j?K0Eb4CA9WvEDsxb0_kL`?H3Q$I2O)1cqKtwZ>IHd{7iyAB-`e%Up9_je z)Vu(4dx!&}x2Qr=+69vM@B+Da>=W?Kb#uwFAz! zACB=Y-4HZkBpPjsh2swUNK30YvGeq09?Nvl%F8mqZDuE#{yKou!&b&yX3F?@c`Muy z%AouvG43AeEnHrm1TzZnfjYO|ZK>Q(GG{F(c|VVn$3NBSr8S1M=Im5K(mpk0gXQs3 z-yq!Z>k@IN0Hp3+1CKaUcwoJoSo$4h+m=J)mOEf`h;(o9k&2g|N@Jv+e zr}9UNSqM1IR(eDypP4hLfLt?g14l(`Jj&JJn>r+upqhMGtssKJZQ~gjZo`Pn-(Y@K z3}afRy{AvjlIHZMhf;f$m4sD;`sgoLhBB^lkC{v5g5DTG{q`0z?6@qhIhTU9yLwS6(1s{c z(@19-r-wXD49`mCW5eV&{N6W&?wVdmZSl^clWe&@SS1bf%^OHkuPl3dt1c_Ke3gGy@X6OyDz&m3b1h_Pj7w)Sp?2?e6M~AH!g1o?&W5GXsAY|^>K^yVnQ;#7{mf3$#1tG!9? z^?V@|-h{3TEAil`fWLWrLH1}X$Kktz@IFrf?}{iG^JYGh?sKU9`8>`yi-WRYL z-xYFu9eN}sV)oDrcw*gKy2P}Z39gSJF-xX0dipn+aXXxeWM~ClH|sCqgOYAK{yYVn zDu;u|Gj9+UUxuIID^b{-fH%BW(u;Er(l=t`Nsrb(2tN}6*N>@kbv?ZT(<>{4J0qV7 z9)vz7=B^E-{M#NY6Is2e9MAQ7ld1?r6(Sp=n7L$1B8; z#LHr2rd<|E86ON&vL29Wd&IyFEihAVGrr(>XE*435-1u&{q+B$=y?T-W8za>|HqO#wLWro|x{fhFgD) zz$xYg+fxqV&|+`A=dlep#c0CDYu`xuh;C;6s2XOd(r8XEG=Rx6>R_gUCh6(kPcl;0 zW;}jadkV3hKlIRrK2QDZDubjRy zW5!f5Cu)MhxYr7>Lje|kkHZqlz4)y@3dd<#;ge$wmv7n*NuO3?^W=Fbu49dhiozk} zZ4#K?3xM0HZm>x*8rsVXpsryAX6e%S!I-<#|8N0M-a3U|Tpq&`zq2@ZULjZ0mxPP0 zMTkq%JZ4OA5pnZUVuDwyQjW>7%%$W)m>D${W(JQ0rL7vc+++~0su_x#EFXbY+z$v+ z`NCCTQb?t8bqMCaVA7c)VcO#i;qANQgN}+uIRN=(Nr-Jcf%4F#M9Qwna zFv|3<0$Ciy}CG#&-Fn6S?s9NsM{{7}@;Mipbb5x&$ z|L0gx$fV$r&ogp5cQ`2<;AKiAm-{=9`NdmE zAA56&o*ed+O0bL;o_LhNXwu*4$_K$=O|{%&CNI=FyY$jL6p>M)AxQ=3U4n#{1nC z=G}*RU=mSBmesX!x~i#UW1%Z~uj0s5*62a8=VsD3b2j<9Nf*lG?-EJ<^JI^WI0XL; zB;89^k>Xi>bdbeqx@LGfZLsM&?dI{6x}Ue6vU)a`3BT}#GLB27vcg9)Bd_qOPfR3T z(KZC;uPP#+t}bW3Gs%MWZxe-{C(Eb=juWg-J)F66NRzpEdMOL+%+~!XT>bnB z$~&Q$rlgFh0L|;nkQcp-=5$f|%x2kj4ZaO48gG^B|<+H6nG&C{S0JFmhD zw?nXqQG?j#cZ}8USjKe5TxM8ADRa60meA!_G?f!`1@vyqL79{e0)pWW@a3prG6PByV{0M@E2pwPT5RPUL{ZI z?J=T;+r1E8qw*PLc|Ydy+gj39dXp7yxX*gYSF>NLf); z@nyjhr#V8k=$*WVTq`u_U{R6FVP0dp5f#|I$nX;gSpRWo-qRHF34^H35)uF9{X0J+w?0Hu(tNaX}stLn1ItHSz*g%oU4GfNt0syxO7$3}EFtVR1(!5Dtl1?9Gz;;gzkd!u4xcXSR#$L@AC2e@{!y*RT7DhdO>!~&_&IKoe*C)8)mOf!(`QC zSam!Td)#iox3jfe&A}&#>iA|dd#9M7lkVX4wdKR)yPx1$=@4|B)&>@JF`(0V7)*wG z!SF;!c)W8f*x2gA^5hX<|5F2ah4TroZZ&giLLkSLBaSDWpFsSZT3CEY1Ru9>N6CjK z;4z|cx9 z{3&UVyEphkmyQO!Wk$k1rP(N#F#>bK-oQIKH(cE;g{~XY;IfSvrk#EYYRJ_Soe#$4 zraa`kcDd4WQ{tHD(o9U;J{7-l_Y$I)p7P&9K2z_j0fLiWqj34#%T&NbQCh*K zmOQU1A%-tl$V=2E=f$*Wt%nb3Nu_91(LaQ?U$f9>WiC$4jlu&9ys*RfGQ7Ps3U1m> zq*q6TL;3Uo$mh6l!p@u!3a%zFjxwhDX!B|q9_y|lS7&@sQlk_+79E2&S`+3L#ez83w)?6p z2wh-^Wrn8cyl*Owjts-}qy#j39g7Anq0H&*RJx#VKM`?`C5|)=pfSosWsypNaCLbozLW2$7lU0!B+sK&HzV0;`+oHo1qw%nf?X8V3t} z>*s}&IG(R@M(2sRj4!z+U&--9Dr3Hm3#?4hgg2w_;?wF=|A(twu`v>MU0ww$$x09u zWJzsQT}!!DpA!z}G!p8sgu-f%K)7<`F4pMX!Ffd&Frjw@)Sp#`k{BMe>}VzXZ@njC z6sL>!BncDeX5lcuSabkaR8F_S$*E2_Mt3r5YwP1yB@KMimx#V+Hlu9LY`i#$$4);j zU;&_{Dg*1k@5b|0Q}FR$sZe%=+mr5WAU@k}VRzF#ZZFQA*WML1E%p4t4x00d zy&XlvHrXp=*XcPhOv|1%-DA!g$eXhSZqmLkPR!kTwp4>AcfVsi2g^PqYOiQV>EO5M z65NKJ%OBy|z-lbFJdIyN4k4v_7(KZ2`b$Mo!fvgj)I084Ls0XJWOZvmg1Qup*W3&4 zUnnqM&uN6(Z$bHJUlcWW!qt_H!pO6;`TmCkgpYqoqi4efIPG~6+Ai$J8jZCm)9#O# zKlrm2QeLc)wgVfz=_Xmf2!={N4qYr!N;wm>xXj^p@U z8_Ca>aB^aUfJ9eiGl@3#%!dRWaB?XoTS8Mw#@KsMbdZy7U!npwj$a`xa0D91DB{N_ za;(YYA?yVaF?P5pa(hr$$m{in46juFj9rrj7r!eA7N}N%ZqhUOXw72Mhk87pdmWc9 zKZ8^29XOl5GDC0qGrfaP;gCU@NWIL#!pcNqUpE>q=t<)7>IUImI!|C?|B1KDy9x|t z4?ysAb@22s!=0N2=w9WHL-rcNn^jYx$03g6b9RK8CJRCBeg|nQ(PrYdQgrL#r_9#a zQn=7E6}o1Xk_F;d;C?Wt6Ms}4N@hr){`XL@uUSe|c|$RbyZ04bdlm|rujXL=vUPabFB+dcFoMQ~?ac1skH}c@qfC0m2WG3Q0h4RC zh6*vHDarET!nAc8mtmvbd614&*!BiQHH4{j@wL4e$&()-=ogvxhI!LL$ z2-^321rafM^xJL^>zqzQRf-s9+?9e0M@7Nm$_SX-ssox|W`gMBE95?R2AS(O4{nx- zzz*HxbmY*7LWPY-=-m|=r0&;W%qO;ta=o(zpN*44Gov;z&76QMew~D$br(RfBaS)3 z>G4gUb_#BFa(S&E9+DZ^66Djf&Cs5<9p=7@fVnxk+p_wj?COnL?@BON zI}LZGIAQC~64>rhLe>mTC3nhLsI-_W+`E1O)p)T(sGWI(e4jiT0`Ja;y_{Ckyyu&7 zp9rVb_3vRGWGk9455QnMbW5kria%wpp@ z0TynY&T*pb!O2RKq4=Eu)Q%u5y?q|$e%=GGhGoLnPj>j;a1j>F@WEx~-=R%T46j}s zgD0nX;?#SNoZiDcES;xMZaH5h(t}38Pm?rSY|J3icWg1kiZ$>IvTcQ8-fGd;r_8M`M(p8~bI02fOZh09$|bBDjxu0J2hHwexYnggmb-8~ zcV~vc>|y(`Rx1x`JRFL-gzjgrqIl1B9DABOcbvKhMwmxJrps1%7R~W8LkL_x zvJ4maM&g>xe2i$VMzgRMRQmoJoqT!d?jesI$sZv!uLKT;q`*Cce9-*xl%BSCIWuz1 z9MWuF3C-RzxJK6y8=`aIiSH^n&}fPJtQi8XVoLrKNKrmHUBg5wxq#KI|z&~pGO((X4IE_hi(fwzP;jh7!%DyznT8b z#1P z*b#P-Y`$_g^L5aBYMP3WNc7#ovrE)i$K(a<)^UN<eqV+8u@!pU&bhU$T&Y5)a(aEIw zD-RC5RVN*KCJYr5NSg2ibRXq%#=5(}N^UUqg-W90iU-jjzE*=zPYC2>dccLN%i++w zR47@kg>>OC)>Qi=j?`JWU&9){ly{Kn8zs?y!yWLzAn;tU1N76+!lQUiWc19j zxl<7vjKz>&&T(QGIYH5{V34_@!kv{Uf$f2PaDVGN98-7~gV`fkR{0U)D_=pvkS_4L z>d$eqTmkKwyYc2mO_(pcgOt0D2Ak(zOucsoljY-AeoB0#AT^NkoYZl?fj=4Bv zWC(6tJc{*QBEvR2b>g`$AsotD1#3bN!0C^c%v<3a+C|oadTmt?pKc$9w2hIl#VG=J zJ#fNFzhvACmYYJ?vCW0Gz~Oy z(3u7A^Dn{+`)}hl57=HJ zjnVlwY)P0Y%W>VaT~{Lo^@TSDcOEqhZ-#w`6>VymRWk--1GPY-u!S%SH%jUeG{`oHK#Ue*l(ZFUPqg5;qvw(jPu<}4_xI54ScO}L$D*pC4F>pE zF+E0#bn5sOg6ln{@P3^q+^QguVl)(s<`2U-m(e)=S~ju~r6}mALH(x@%$u!$QJXd{ zpb8WBKu`QG_+rTEffwiqY`X>t$L>rvO8u;l=d-3E!__w!zj;r)W~neICRve38p>cK zwSZnwopXkDB-8uis)pe2=1=0N%$32A`$QtT~4&1 zj6oT$syK^!4$rXY%3w;COX>!Me44qG)OmHo!gVEIBV=J})t%)X0u+ z1Y2Wf2CKehG|7DL9sYCXR!Y zA$;B-03(hEJZTC%YZQTB$8yO84SlGObA*MB5qPP`52rN;p^k3|c>Wv(N)Dso)228w z-lT@vw>FCS`|QR2-p<(GXow$kYA|m}5-N9CqJ>uq`7*+nC~i`OwW6kwwq^tDhqNrp|W8Y6#{MN0DxfLFG)8H8X zXwc(yTQ@TnRq@1i%t>&KuLP?_OMt*UOmEzc;u}(N*tM}3KhzUNoc3Z}RxK$}NFWai z50OnGBO$Cr6>dt%L-ey;w8%-s^88>7895!jY&PQ3;BA<@&z`^*nn_-Lf;nBjjhVLi zAT!A536pgulk_~SC)b;Xf^EhEL959$%Hz%(;fv3s2*x=vrgM%nJv#O5&JFL_6U(dF zrbc_JsJ%!yNynH;P5FTeGDmRvgRv+%Yrdd-rxv+>JBZA_{SL1U&B0l6-l!*;%>D>D zz$z|^U>{%dgSS5q!>P!>;P4(1)>=)Bb(dCSXM&Sp7;{dr>sE#^#X}s8>veH%#1tH= z^9T&)IKpgBe|JOcJ$$-Nl#QM{oLydg8uxG-UFR)J&~fQ6GI;rEqLX%&xv`y_f5z#6 ztF|Ewi=yZeQEDW6OES6WZH)oSvdA8n$B`)`sdU#?6P<~5JPn(JP|&*_Z03rf^*kHo zk$q?yl!{ufAETF7Gv0`Mk0-=T;FHy55>(zqTPE#=OG|dZ)8bSBw-Gq&t1d1T>fj3N zdtmOL04JAe0$x}TZ|_Dy#k%JpIGY6`emCKf;tR05mWJ~#XQOd=95#HI3q8hXz+$H| zelXaHGyClDop>iq1wQN$n?tmhpQg-jq=HKHZWz{fnpj`qc!@60#C?0XwR$Ya@8q%* zGXhfZ^^+h}U-uQfInMI8)0ab%-vTH*xB~QIj(pqz<2r*yifDMR3YOjX zAV#GZV%}L3r8#@a&NnxS_+B-f)2oQ1x+TzgC4~;Q)A3ueEqY0+p-S&OyvfyHNA6W; zG_H-NEAN()2cbS>eBydisBwclT+>T7kG@6@Z_6eg&K>lvcd4|+yi7*-YzUDtPA01E zl5jqLC;a*$f}y8P@a?kqbOd+r}6KCOHhAS~Q&`3D6SzPeO+J{=)Er~k1qw(!x3f~DsQ15U$)@$V9SLrZJ z9CiRDD^hTq$q@FhCJDCqU@sOe)`StCgfQPw6Abb-ncE4yw4E@F8nk*g85AuG!{(-w z8Be#9%{tGB*kW5qk^92m{cO3QzGVUpU$>%|d(2Lb%)94x)=y@u$-e z7@mBOEG!X&3zuC$!rcr~4Z6sJj7)sT)#^U5ZO0jQBgxQ7I&eD36ZCw-@D8rV^&h9< z>l3k%eLD!WKdpz04p#8$kSh1D0*rV$9M;`&hqDukxEh%Vl%+$lr#%ek_RYrSk7wZ% z8(aL8ew>NS>1OngH!~k{)TucCi&XbCEpqgW9n2_62GMn8pf|{dEam2d%R^_txcU0n zt@Rx;wT5#13MYZTz7uNo&2f@(94HM>0K>|B_~DewPrqHyn<-)}m~MX&_7w!d0!1zt zSLz+KAG;3)vSYC89iM#kPi3eBDWt%tk#x&&-^1n2ga^qa@!J(PpwF;qK|M%ar~zdg5i}OR2x<-q(2=tT8{6G+D93SLoNNzUaT2KSHimX@ZqIKM zj-=EJx6f9=^|yv%?+-O>%?X51RW%6C{Xs@OG{RgzQRLs0!oVr+IKD;&g>)_SJb4SM z8$~&Nr4G0odzRBaNya2@4J35lL>!N+lZR#-$dx;fm~ToO8SgJwm=9+?g!&V22`*Es zges!jxVq6jnC0n>Dy3UlQ+IoI!zp97;F&so@4h*GcC*~G>^*>(}<^=Erm7v2G5!| zL!o~@$iGd5X{xU9&LxRPKlC5Ad{5}^Cn)rTwElyr`A9o(OhH4x??yUE2$P10eMa8k?l;mRamvbv~ zuG>NAAF-Y);PfcmrkJ66CPC}fW3f1pJ4<{!880rJj~S^;p@!N|8m+j!alJdvD4CAW zj_YC>m)n`#T?6w*a9T)i%KSRp!+iVvp(crSO;91u@%fKyfZ#hpIMwtZUNtPlQHu;f z*+qlH+D+E}SVv~h3?UJd4-wm_5GZ5Y0<+kkzK zZ{YQ!6X+QkjT>c**)V?tc6_u38=~`)8Jn_%q3-TxOjE_sAX^6OW^#OP7nM+UjUt|? z;`q>|ZxS-~6uDr^>BOuoCY#0dVbj-INH%OjA(wkFC@u%ruG|NcMy!I}Vh2D~Hw8_% z1mV@FIBeo{=N;7&$phD1GSBG(C4WbYcIZx`Q&gu>tuxw%&znMpGkvmwdeR7WSH$p& z#XC0rK?8elSSj06GlfXy4r6)`70_)j&*JS--Z*#ELX77)6kWLU_nOxaiDUZ;w6Ny( z`AbxAOQ??Npdd}tQAG--q6hAR!0|JzXIo)rUKD$O*&$Z@d?cHr7(mZc&!y~Q)2L%- zHBi|@m8;Kq2@9Uw#>-NPn7<(omvzJ-+DGDOIvl@!GiM!?#jQyzZi7j%5216-^D4xNc`&_|Ma1>&8iJ`US z9;!m+mT-QF9X0pfcGx9v2@{rzK+54Qc;-$NMtsY}dt=_i4d1~Sxojw=WDg@Xi&RN^ zcs%(l%>k5j{o%r$OmOACFPBZ*p`~68|Kf7a54srQPYFvb+b)kndm8i9regBU3a0Zm z(lB`s$!p#LWK}ImS#+BTxEO?5!Xn)J_8E$7Ta4+K_i*Pw!KiHTp3J>G29BX3JinxY z;s-Tx*2^)dv%m`bKCQxshvSfNbp^JYLeO&eVa!O5g^#I^VdVp3JX9`8uH=}J=BDXv161Pi)z3^-(TbwwWZFt#AcdL}pcGWG!`L9)I9gWzN=7^5FMA@!qKk@A@V>b5T z6L@pE4N6XlV41WW`$X1`wYH69D-HCR?D+FS!|QyyE?^L9rVc_aWo4|H-ALrl|03H? zErR5{^JpShgHw%OP&G3iW{47{LL<~oLUk9GOC!tOL49By0o1WAb zN)G*y2Hofxm|(XQ=dI`J&A-l|_Rkekb@8@*FIoUSeP>|U_%h%vI*KNPF5}O`53q0R zQFLE%0!>UFV0r6dPG2bn=1VJKrK>oJdUS?9)_0jPm(a%C@k`ODJPNBWrIWyFAYTW4 zB{L2sfzZwj0&X3KP9_|vo@(wK{T#Hsii56)_u=~D58$x=3(+fAhmRH^Ff1bypIK#M z&9Ga@s|x^yg?SLdo%Jj>PllfDVIYi;g4>(JQAc439`v5c#lUW`_1_Qs7QBZpzm<4; zsv~j|6lgla2cDhT0b`v#K;{H8h7(WFSq|%I{_9~Ns&5B|v#a3H?S=6E*?ltBRTrY4 zufS=K6R>g3V^sH6MBC$z_{uyC9}GDO`mOCSZfys2Z8*cfn^0vkjb127Z*w7@>DA;_ zya9;DarJBdNuUvy2_pnMAZFVH=vvcEqIkZv%HY2!Lxrh~gx+_`d9fVR*Y%kg$m^41 zJ5MqSgFcZhH@}hSEhTXGi3nP37{sirm=0~uuV9KxJEL7-#Pr<1$(#)>ASTITR zCg0l-O}gJhk6tD$aNhw&fl{ciJ{gBay&<)o=g9q;NIs0e06#{XgWk!>w2S*D5Y1kpqAW#$){`BNQbJFB6~I~#l$0c>q=WV{M(nOXEHMmBUdv;DR?F`Hh-*w5WA z%w3#9@lM3><+(e1+S(j?>)(+rZDc zR*-e<6S?f?NH26wV~llj2!D?@%sb}@lV4}U$CxsjsS2kvY*fgmO~xR%MgrJYby=y-ru zE8yM!3~=n_p$>N*dZg40d;LbRO6oiLr;WaUXXaK8A~r*o&>?05!YKbFuiswBd?pXuH-}*UHcs7H?Kmo$ z+(g5|M_~K;JlI6Ef)(#HSrcGIv^F^t%lr}&WS>n|H`kD}PxXbwDMTPwvz5Q){A(Dw zQ4Z(SY2o^m$hH9jd<)xWJpvAGh|&Rei^z1FAg_Chpx>e&r%Z}87RU} z<2sZouE+1xMI2K^OqUMVF?B6hF&$Tm^z=!d#NR{)qF!?$GSQgLNnqGYhd$vRP-ke9 zRgB_wJDf0iDlVob;HkK4cxrDIF3-P&4q3UlwBsgT4@ts^g;Ff<`e4?F(;c(%KSItv zIZ4&+o5%F7yFpPee=rYY7m-aDX^0)90U?jgfbxjoxFp>1=1ePOj#LOwPL>d8yL}ga zy)gsnnR>X(nZ==D7htW9D|E+Zfpvl(JI`S|yTIFrbqMo>dpDjETVrj)F|OdK>lO&9 zx>&n2fykFCkqI9;-n`2@@YPE{{G~k?O=TwHYp)2*R1Cr)G7rI_?hL%(2Z59sm(Y}W z5#N<%;U@6Mkn#=aK&H#r$pA@5>BJG z2?jOjqLPRi?h1CqS$7rKr4v=yWJ_6A@bLo7Thu{xmPB!}{}mg?zC*{9YRs5q2c^5G z!t{-XF#Arc@VMPR;q5AMYOSgdr$?6o50)Hj>O$#;I5M&< zjKuQ~k&fweiE&yiT1)Oixe>Dv?07f<<9525dBV!jJf6xr{ zU~k|I!z`SX9FIoh5bTfn!{Lb&AS7@NYJ3hs^R@{5{KgP(JT<_f_q0%R-#T)I+jsBI z8_Ag0^6~055tPWf3Htm9OxPX4am6h`+p>|Eb$bl%W=-&v*EkG#q>s8BH*xD^XE=13 zfronzk|oNjgl{5B;uiUklz;##rTd_8(XeV^e54*!P0)t~j#obrve_TyN~T=3Vt(D_>QJjzWt$c)%Q$PxyAi2i&)?27|VB2#MSZT4tOM?beeC-}6EsLNpPi_|91B zX@bg~Ll6lcGZ*u*Yois`Zt+2rXb)_XT!kKavJgMzBdJ?4i+byMjHndOA*)x|kgjFo zkUvfVd^9eSp)=IcV3i7XI!WPO10ih@FGs&I)}Xm0DTui75FDSCfsnhSKlp)%s-9zU z;b&!bN5o+E!aWIA@8fi+du{-;Kkp>rQitK(#60k^4hQji6%^+*#qMpp4}&?4gMAzi zaE2kL9p}JlEQWo7skZ8PV)1tPTyy}ERdd0W$;Yh|j^kBf3LbkehjSYMV~1O#S(-R5 zydjGVetN^w)*Aw#&|e)}0I zVk)w1!*Y++zAH^uEZelqgc1;n#}m+mv6)Onod55BL`3|}|F*%?GwnF{Vv)a}|7HXE z%?2Tl&tn)0IU9ukH#Ug={c`nIi7Zj>#eYBl?d9wMFDHZ#l+9%3|K{bK%l>in-)1~2 zt33Dj?|(mkHbeeD`oVYQhK;M&t^NPuf1&i>Z>~#mcO1mM_3vl)4AH;;`F~?z!IqtI zHnT-U{(cnx{*QnC{AWZM`*^MS{jZ}xsh<#8zyWg4h=2X%Ar9|9|JsBv5HLJSz=Pj0 z&WUb3Epnaf9ParyhC=S3-!TY}fDowPF)~jY)QF7cFu3R67|~wh{V{+6q2PCn4_o`n z{~0I$#!!fm=#POy>~F^-IFUCnMzqG@0Wg41^=k_=w6tpAwkQ*dL>EdhFtX zeHvXW+0PC>kC6Tt|2b@a^GCGQZx})zFnl2Wd|E1<+c9u_#0-eR@R)x7Sh@S9l*l|T zxVYyxJ6_BDZU^TMZr0-c-nXH3ZQ1|KKmW#vHk0j-!7;^B{V+001N=p*I6JuK-x$i1 z2Ed?UfLQ-epN{<3c!?f7^xt-H;l<+^?x6pG`Ok^u-yn*T{Xn=pa+(^THGj4wegF_5 z5eT5aTmOqCL;71n2^k&_ncw^7&mi#wfzTA6fqtFypC6ci`$R=*083~Jg`}S)5h?NW z|A~@+gB%kk4vKJxlI_Rn1$GY|-$7V_zSqxipA(mQ}96vZuV`t{DA<0NJv5F!Ak z|Hu5#%~AP2fF&qk{-^om;D8`B#f8f6aq{ObR{YW55)@EEjwj}KkUu;2;6NbUTt7gt z{TW2@XFp4X+~Pzip!!+zUwUUC5a8C){Xzaq@BHd-3C$PsI9q=A$)6p2Z~zeQ%LM~i z^5^C#_w~1g`-*Z4*nScFXUE1500LZND7xQw=g&S-6p{aZJ`n;zj$!J5@yS3SNRXSV ze|PME@rh`EOOVeoLH{pdA3qQXVfX`VQU2V;%7X^51h{o<|K9mC$ie;~6rV?Pk@7!; zy|P$8OSn}Z!$a)XJAZa;`~V;VAuSjnV*ea$D&qYu5zq{e!}#T#I0F;2sJ z)W9F~_<=!aZmIOUPcBv796PX26y6VL3B&I{in}yEZ~SLR{O_ejwEVDsK7qfFo%(;! zj~;t+=%4NEzd;o4DGdlhnExT{qsNc<9Ynw*^xs?B-+hv^Loaz?pQub27)02AF+8qy z`_n)8WX8w=Eg9gG@us|Q?Z6-ka;gJ@aQD>xnxj1L=RfBr|8{KjUA5mqxa|(NS?NCj z|2&EnmZ=W_!U(wa@b8wiIk`;!=d|J9mc;x&!)}PNA!fifde3vfZiv7(tS&J&!mNvN zduIZd3^wloi%5)gQhTPQ9ecnEa$^``W(*wk1-5d~!({gC+RxYxnJ^oxAr@HrfO78> zKHx|vIHPM<94n3iI2>} zz7uan`RVDy_ufBy!bUoAE7qC$srl)X_dRg$eD=h9?|t~8hfd!+f8gO)pSktJtNTt@ zr|-L)$gQ&{J#TZiXWxluQt+CGAGr7Q$$KAu^=tOu@$lLC^z8iPnfo93r1|Wv?wvo( z9Cw^NJAdGzGq;|sPQU!@0}tJIa(?>NXCAIjzUIKKvnTJ{cj76z>BH5jSDcJ@(q;JxY)FoX_5CbGGj>Jx<;C;Mp_tle=!?^H%p*+kg1rf!TYP9`Dojc)mT} zxAeGUK6}CDY;@vTH|--o94E_e9y&eSzlwhC=+VQ+jvbgCDD($A`Zehf75c;T*^$lJ z(Z}3?ho^Ux`#U?flKI$Z?bv~%v-i9I)|Z3>^V$1LecwjC@q{7j4e&x#{6N&<$T6p$}Qb z&Dk8=KU$l8XyM-IxL+*yOA7ag&1WCJIUC<}HV5vw@o;{`<1ruQ;oL0DAK5X#ROXK= z%pX0Ueaz zl|7CgKHv-B<)z2HU61?hak}*2?dcVpv-@v)I1k)$)5H0|O8R4m4;($bHhZwpzp|r$ zNcvY5`d80q4{y$yI-i}9dcx!Rq&LjwFM0kGHlF0u zdA9U;&3yLBo3q#6^mHEF{dB%=RU6+z4;?x%`;@YQPwh7FX>Q<=LjLLV*=KCd>YJX< zgS(#2&s@npI(lGqbTs>{!u{DD_vgs{xrO`n^Vu6VXXkD@oP#@_&d+;1<|9n~`Gxrl zI_58w`HKqk7td$^ZgaM{>B%{`^6C7NRXvW34j(zd)A^;P$Cq_IzT6&PQF?skeD+nF zvqvA(W7X67)hm1KKQuagU~Tp_rN`HHJ-*H!UtfBB!+iFQo3rMor}N;=PvY28>HN-B%)F@`^{3)@`PO;QeD>W^N1jif z&hL52uzNbcxAgeF`Rw~QXInQtoolsIXt^qn18lo{yCX{zA*p7eD;f*vr9KUIcqDQ&R<&9Z2K`iRz01+xvB@x&VfVxeEhA_VwF5_w&i=5_|4~Q($I|~vq5sqQ?9Vo5?PG4hji>X^S26n|_ZV*v zf8o=)cRu?|sYg7YJe_~Fk;+@=UzZktGoStK=IqLeXYkSFm7jcO)rZ%=TixUE;SrC| z-YblB|7jKT8Xp(@T=<^~^MCG`|BKB3wJ`s; z`Rt!JXV*@=!>*xk{7C=rkHdTDAdlt$D7^o(aCEuXZ2*XdG!>oo7F8`Z|c0_X|3Lx zwAGytA38Kzt8O)(JHHKD*ZDS7J3kd!IzJ6g|o^rtg5(HN73xrcXzfrq4i|SMSJmvw9}io9^NDC%bwlf-CzSUgKAv!_~8l z=jXe6HmVIH5BZbcP(258W09Zo>YdR`<99)uSMSPovl?>U^~CS~AX`0`_^RGV_OBf{ zcI0sNZpQOo-yN;n>t0m%`X0#A@;%Y!)q8Q>toCu;^-9?F@wR#%;Z+?E9auZUr|0Uu zjpvTP1{Pt9!WadfD9k zvAOyXGFCM`bl}kb5!?MxLl0AYQlBb%VyW-?&=ibRUP@+I`U2F z<;HWzd(pa%_o3SHG_rL3B(!<;3a*>g{akmQ?v796)dK_{r{Deqylfscp8LHLt?Ty? zs{LMtEd5@MHm@G$x>-%R?s|Avenzimq^;`g+3sIkt19EU^BJ_R^BmRAXOX4zYtZJ^ zCv)AbUd#2S&Z}NFuOn@B=flTF2iB@jF`herDq7e1(@^dF2(onkbhLT(8C*B3n(MBY z&CQ?ZtIs53RntQU502IjuT`IAJU9Jpw65vrpxX3vk)`SD(dN|~xNcVGxZZSQZ@g?i zkKn3){4gKw-+!R`eB=2y(dr9OZFpb)xL|cEe@zu=_ALG~H>PwC1&3+kLx7ja8b+ccAEG@qhZC-s9*UjotuDf10yM806zMAl= zj)x8&ILenz)z=u$9lsW>>-cr3cKmu|>G%z3^XeP9ZdMJ~UB|oQ_k`-32(Ik6ws!Oo zKTyBfc<%QtXkEW=MYZ3zAxpn+N1IpQ!F97b&vn<+yYjb(>N`nW)!APlj?nqLjOWhZ zjn;Mk9#lJjFS2z0KD2rD{aiPzEv`3pUiG^90n%1?UOUFO8`TdQ&z*k=t?T^5sCNDl zWa<2)X!GjFxNcS#xbAx0-29tH_2Xo$YIt;!CRlg(q4WkIn0-wc;d#B z`5veGd5qHe7trR_FLK?iE^$5awtNu(qUWW) z)$gIrtKa9kSzYFO;%X&9e}MOK8XX$3r$02F8~qVl*XWN?ZS*I|(&$go=GC8Z z-K<)!`F)2yt@;h-&xx;U$(!C$zILeo!gy}^muOwfze2U;Un5J)zd@Tbaq|uw;%)9f8PDDRGg{Z}zo6RfzamSw|AscN{+a7$b&YF(^KQHh{yW}Pjd&UC z=L_oUe;DUw@br20KT+-U^p}JuyuSJuER+q@|Akum{cp5+^?$hX%jgqBID3MB-^%}x zRXqVSKc=T1IC_ZR)eekaeR}?ygVhuD-%i%U)A<9h;-5yUlQ;hLs(KR9N2)zs%Rh7F zQ1w>X`((WvPky9&N)FrU%>4AUx&`w?x6VB&&o&zLotBA3VbM zKh@hCFa6#Dbw|J3G5dZ`M=$-JfyS?&T>15r>)DN`K2p6C%DxA$Z}`E--Q?keNBtc8 zEb?i}*V5IqQEghjnXaCL*?&1*y)#DX`Yvev?#Y$kJ-I&kxZ64~I(UG0mFJSwb$B<_ z9Ua~s^LmH97^TB|pv|lIt^)=u8+SjK9BQHP1T5`uET!R9UTr}Uhi-aqjXq9n^%Xp zZdQl6p51u%Bh?Xz61UYg#CHm^Q_>t^+V zTpxd1{8Z%=T6GsmU4w|aqrp06-{3fUY4Aa4^XhJ{o7D$%J-hL4k5n&2d9!wPn65nf zFCxEddJk$j%lrk%hhX+!fP5%=X}W+s?I{Igj#C*Q;5 z)2(f#~lFseS4q^`rKq3-DL2&Fdja6r_NMgk^TShtFB*r(Ov)e&%gc$&v?W8yuSKM{Q1vGPQ&_DCVk)szwV3wV)TXY zmz#DUjqf{s@2S^Sj~b>;Uio9fsaKtOMRn@F>Z^%-UjAABz^fmaKXB@ylZ^5;rr$O0 zMtL$-c?)N%uO;va)z@*|$3Z_~9fa*tmV;=^M}3c*l)rZoJdRvo@Z+@tlo!-guXdcik9nJa^;WHr{<>@5Xy< zyywPyZS324-o|@xywAq-H{N&Sj*S=a#=n2#z{bIiwT(j?hc}LF9Njp!@qQcczj5cr z2W))c>9eQK=1*5QK1I!6H$7c_J)?Dh!IL9*oVxGSt8QKQQhful{PFYNhhH_VPMy|Ep{DMiTCS6aRQxrOfa+{{&hsbM#;TSImnm@>Xs7XLFvn>v8{u@aI+xbYsIm zpQAfBeA8?F7f5(;AHVeO`rdN)Ka>69V_NKdsb9*`oqJyWLjR=|d8_vP%Q<_5B+;x^vH~U)|d)@>cEnH*?<3 zJwN7^{aY&ry0Pcq&e5HFzUfu{J0v{#uY5oKyN_wH^F{r8Il6Ptt6$Q;zanqdo-gOT zoqK-F%lQvh40L19f0(1IzN!2pA`d>^cZ+|V%O3Mh;Gbl-C-^wu+x_XzgjE~*vz)N{ zJF#|W(;Hj(bKPgpobY4W>DSIHJ|g}Fu1BiBHg}kaK9*i z^asE0#UJ!8Cr-RIRWCoypUdXhU+;>&4Y9uTRqNyGZ-{P$6aKk&F46b5#pkD_?8|?L zU0utJ`oODSaq?BC=l4H+U-fr{PCS+Lm#m*SdHh3Odh#Rh`H&aBwEBD8 zXV2a--gwF_@BQN6IQd)e8-f@=?q7&MJG{rPzYu>x_ovnT1&V*1{KDhUz3Gp|cm3sf zF_>cAj?enOZyfsdnRrtj53cT5{?Ph`$6xTTcicV1zfu44m!fZ3fA;&7Tb}c0cR%<4 zVgKc?t^Pmu&wpqA7WJ?1a{muL_)XvQ|FZx5x6E%*|9_L`kH4knFP{HD_Rs$U@Rs%e zw|M@_Ut_;T{r^qwfB9?tx2S&`c@J=<5C-d63*%)75xah42sw@(BR2k8-54;x&oNKE zzN^H7D|NZC=nc}koZIEq#yaGBbmfFzQ|G`IIWHio3;0`n_HU3mo^H(AqUSk?HRzFi zIR{;xHRgm~qK@E7_~cCZwIjB=d7=gla5lL{K)?JgGNi(r{piE>Qj=#U9>Jsd+S!NR;7z6XIA08)$Zm$9U zbHEyF&*3*OC4kS^7`p}AV1&KrB04gWe-)oG`|G;^T<+I-?P1L|{V9Y6z61hz6yGIa z@7tK^JidmU0KYtSqz>OL#JrEU;U9uAvD@(-)ld8?_KWd5t9|9a0Njgw&gIMlr1L(9 zx`;fFy@o&1$KDfx&$-6v_HBqc<5BWzPVNI z>eebFFwY!y{mxqJ$CvklZmh)?>gI=KnYTetpx?Xk1e_zEf#(3s>-k#Z!hWLZ?3SDW zBDQ?aHUwG6=WJ7u^N{9c-Y!?309p`Zx!2AqryRy>>^vKGzL&<~^4UtnMF7U?SRls; z#Pu@67@3HPT#M~KYG5z>OqQj_=h%J`pK)IkEw}_0>f~=P%WHY>#Soqee67^vZxJ6H z58-aG7ic#-^6$LJ>keEcKaoEmF7J!U-3P9i4+hKf%v?hb$+t&BZ>gCO_x0_2#SrAz zAa+B{=S=QB-$d+As*k8~=JOzepuH^TwT*N} z&q9t!oqfesa0QG(h^6MzvdpUV%=ODT0pWOgZq)6U`Id#-xgvT(Heg7+j^)1e`SyB8BHN z0283!-sW370-j9+B7WrpBy{;hDmB)HC2^o`pREr1wVgi_o2(bE8tFZdGXbB00PJag zh~>GruEw4Kp9g1_XMr4nfbPt~T4$4|23ueN8Zhp3^J_3(vPTerT;A)Zvm4^pIYZB_ zKE;=3M)!VT+&zCJ zmVjqC1`{v^k^FjD_IVlUo`*oLxL)pYLN*}pi~I@kDe$@223tTL`4{U*u=jX5TRCcC zSI7w$%X>(x4y?IG+}F$+xPm^>j)l{o*&}d`T=yfF7?L{%zK^({1dp747Ptl64%|n8 z$M+0-D6b*|7!bPzxvw!WAYZiTL!|f8KABUA zT^mD}&pAT6kuR4%<3i51i<_5Mqy)&Bed@q@%+E;JENh!s`Z`yGoo7O~wg%=W5YbzZ zdlOIS*0}7Ic@laoc0f97L?&Q;0R3F%8I!X=g9beTmvMV3b6r9QEl4q3=hp6*e3710 z`=*j-Kj)Mu*TLBk%RESzd-i0K)Nrvdugt<_O*xodKvJ0 zW(~d;ct!!~UWV$RjfFq=TrYj5NIC7Bke-3koQQ0I^(h4$Wls}q&&*!x5oz4IfHco~ zYkR3Lt{0y@w66m@=cU4FT{Pa=@;kq)O{G_jG@fNdcg2oB5ki+!pL*_{u<51U&mZ;4 zjj!j#D!Tqu_HSMS+I2*-hQ#x;k}yC%bNJO=xj#8=D*LiW1FkQ1@_W9BEuZ&3rXr1H zKQaNAa!5{tv`3AM>Y%@m>Opk;4SM#aQfFTFQQs0vAc8!5D!k5}?TFB)&2mQl{Ta6C z#-do3{d%Sib^?>mr>$RK4XjshfpUm#uKc-2EPm%SW{&>00lA$mbZT z0oDl5P5XS8*P>6+m7Zl%g3w8O24Nl2{w;DRMoZ#g3KB8lYf)P+ZS&Rh{X$RZq6N-0 z2Hq3x3rKzHH8LRc>hy>$r?V*S)1QyJvxpI>K~$O##^~0_VNYfDqx*B)UJdAHagKq0 z(Bcn?d}QDIx}UiHL^RKZKGja04>{;^Rz9ynmD~I{|-CwJ2FLYs^@fJBCKgZGgyex@Fe9rDXV=x5r2tThb zz$M_lN#vTh_c-wma!-g|0fUR>_fbDj*1-Hr>mfX!+z}Xpgzf&!A6_huiA{`mGU0P}=a>Td z>gmo@`=Gdc7-JD%qJ9e?0j~#$LBGNv?UdL|q3Al_q`}JJ;V~Lq3Z{~(WZfDJB zgdS4aZ@}lW&luhOoRiAT#(OS%j+Zy~qQA!ozjNk2!j)@?M|^occ(3yiud(H}Z-16k z?1bN%5xQsCV7IBv;^0l6nW?Yg(&vcK~hw|62I z1Z!bDm1j$z^IQSGHYUsRUANDC9eI|Im)F9WSR%Fu{}wW0Zvpp~$dgA-_gn+>{aJ3_ z6g&HrZ!CRA_~jqtvnEpOtor4gfLcysmw~yL@cFtlM^0z)J<>dD7N8+7A?1=ckkcTw zt#zL@Sm5)0!uRPwy?YPDjag?+?#sLlzA30&BwTH1hBl*Vn5sygy)@~tdV$OaU1aJ=DRWJrM_83^x!r1`1-G}FG zeu_xoIk=|8+=sbMyiu2Jd^NElbv60`-Ta^rXe%Ri{W-T?p1rn+)x<_f{c`3!q#SK5 zK6#70246P2doLeyp#aCwffAhJ2_OxuN^^9cQ5FZ&NtMb3YR(7k3b?Y?kFPAy` z8g$RSMVe<$jv?)#ukMg_i$M2{6}lH-8N5 z>nw75Px`vti&~}onwSUNpM0*&#T*N4>#{$Uygt|dftoD4yogNX$t|DvIRg6xeD$)- zIzYeSTZ$I}J$gQ!R z!Q6lz(5sl-eM#u^9z9@~G zYv18|sc(?(ErLX>?U3gtKP=1hQX{v3?^*6O_ay8-#$W>W94|j()%7=U4VGngYn-74 zV=x6HkY}Qgxx@7`htK;Kn083SoX0b;ZU}k~=emCO#3N|Q7bEKB5XV~Is`t{l%+nU;k2`zR z#gq*(;eJFyA0Q)Ay+%I|0=hmqrl3U^&RwTcJ46=&Bx3SttM{?k>G^peW^Gh@KZw=9 zrObY0e+~`0oXUKjcfp-Po{ej+X|RQS%CIas17wXYSJLl7*EZHB;>ds;B89OCCcs+r z^!xmqryO^7!p?ibHmAY1u0=+qwJ!4)U<9lQvE*1&V`pxBqxF*0;J04xK-}fI#0XM( zukbv4W|KZ}72WfmB5Sbcc+ur-Kp>}0C4URs->dU^;rI3EEDieDTy#H+N9=&zB8@x0 zulq65K9lwG8Qnd+mi(*uTw`=|omE-KQh(>g?ztjIzO-KWhDhgJfQA~m12xt=+ZYUp zi7oUiNT1Iu=+0?vgYK*~@-i?tf<8w+W6vR9UiFr^asc{STk<^DDKR;;lc44@wmtkk z!n5MPH!1PyJ&sLXYG-#{xT$XHMot z%H{jfHu?~0-x1PfZ9pF*n}~kA=n-4)DKg=A4$$)D{5#}PDBcZ2M>;by>@hbi;bbpr{9w&zF9_+b5x8L)LC8RUR zQNx!P(pds9egQm0?jCp|y15O~KJI6T-aj{wVsByF!&%x?hwFedl-PE-Elc* zORo7L7Qb;(4IN;1AZ5FPu#d^RkZJrgE>d zPd`Y&{9c-4e$+1#9H2(mr}X)7spnY}2FP({kLYbIG9hygesdBq7C?=DJ&tWG0`tw$ z&Sy+Nbasq``DItRW0#&Ax6eD?8Nwevpk zWgY3WU_bp4T`up}0K`1>a7 za)iwJope9WVqNC59@s0YGp{)d&|**4%Y4=Z<06(k?Wj$S_2xukEj~Gh)T*0rUO*1J zJoj#{K4T41p0=}VZTMUycPs>{o3aBxy}>u8IS5p>;#l87~xkM zs|C9AT9a+&Y0y1WmvQX`+^h2{eeJpQJA12SY1g1h8tBZ|@6GcQ#_TC~3-ZFAfQW6} zy_jo1xwDPmxcvIm2kYgzZuDU*Ykc}^?5H0o6LyW95)0bE8uRQGyEeYJ~{1UZ|j9zHBgE`&UJt4e4XF;u;Vk&y1ZZag5H7weq{g+FdkBg*U0R9W08q? z3tajGaAxz2HOSb>9K+5&q|3bP`RLxKT#Fuz(OY|rKi^MegDtduF4Wu3S0l}pCogPs zGA~@}!k(FrcmUSdky(i~NO?SmgzTX`0(oBQ&GQU%KHL#|iXD*IM~*dZ=Z}dHux0{k z^n4#6T-t85ceVC#$c>>=k6WNxG! ziCE4hot(-F?JjoH(fy4(%A^>XTO^nn4sTAw!li0!IlnRlw4O59nUO`b$N;vb=VUjkBI z&&~6nV5?_dbXT2U%ezm|PwSYvAH2oi08bq&64axWk~4|_ZB-RMW@SMhmn{_L_&t~@6^ zEogviNKPOozy26Y%y`zZ2gDm>0FnGwA94zeC2aF+r03vyHl033*Qed;>!kf^bP++L z&c5|05P*6EwB3U~arfbKZH+7E5X)RKyK=t% z1RBsUc|t1t@;#u) z@-CLwK(0wQe*V_xVMnBMOvp1wpYK=O=$DYbKU@I~dN>Zy-FIG{e-E_{`VjvF4A4bD zcZQLA>gMGv68<4RbN$|}=WodIJ<RoCY%_AqaXekV27x71JY z`)nD%g491nzl@B?K#u3P$393`is%U_f$!4+>G@ihQu+V4Mq~in`0XvXePYB|*frcEDE?55U}rZJyG7=A}%#=bre@)8FUHqfaE@%n`Xj z+Rq*}wr8LnvCXs2+01Q$@fzO%*=U1aw{FtK&577KhMvgF_2_vK>+g*-N9xQKLtw5o z4e(j1!36ls$enXKY3&r~A0tDj+kXM9lTWz?zux?jfh`C8rw_kIlwd0C)6(x6F!4eeCNOzIPVzrd#TAazb59~>J#Fg z@fNa)h0}lck!Ogn1?EI@hxn8M+gJo6d=2m!Xo=a+`4fROU>}!ZE}*>5yN) z#0SL8i$D%#Kw2-au^dNFo!%gu7;>KXEx&)659BvCK+2cQV?}J|9(FRO!W)#41oM~W zbuvZ<;C*!lZF>)q`t5I>7_FCjd)DZ&llILs;M3ptw!b;XV+wiIND-H12Kx=b2-wqm ztuM#PbLHBU)qJUA;k0MI&q#H9c~8B+ExPE>-k$C|;ImK9uWX22KqjPlHM+=4Tf}nK zvD4*tf656`-q_i>53bb3dR}w$dl=IFsyC_J2j^_jwZ*WD=lNpEPxw76&&PiD*Op6r zij*Uw2c&b@!#w@wBv6B#gY@;(A_KMh6Sm8Fv@_IG@iqF$vFF4Z_=R>uP6C0P2&{1* z&Jlr}=IE10zq$IAL+w=RaxJz#_mykOaqY}cv7AAFmVMrs8hcErA6TEdz0)hH1A9i| z(RepQz+ZQMeKj$C=Gmh~Z;89is}qWdC7yLmS4?Gwgp83n?1io^mok7Rme-7S14iTp z`Z~9>`8>Of<$Pod7Wm!g7^z(Y_p9ym?6P)(E*f<6rs#Qg_UBw%*yas^eW%ELp2XDk z4UiEz1TFp=+4Cr^jj80zIl{N+c$wLA%=P#akRyo1|B%jVKYa_aUgolQexHrS=iVA> zgtf|T@?CP~ztfP1vN-M#E1CT<7rufgwGu7QMZuRv@J z{9ZP|vyI=M<>m~~t-A`YfD0f};|y|5l-OhZBHwFhlyb+;HzjWj%nkZu*`wz+1;(6b z8w|l`Gypu{J?_2x$6a$7$IxsSeBbgns9E`9lRkkHAse}i6w6ieQ@ z^}?^uoC{r^wGC*2InJ6=@kMg%AF$;K#LQQ^H|v~Lz8XExiEP0Z{sw=pN!>GqFA=Np z=UV&|d_H>(y32U%;Lmz>bG%1#c|I`}7YJe@flH?lV5kp z5w_2u{p3x^5%v(>eGRF3{>8%Wef0HbTn=SW;+vw6k=k`E@qmorJTYg;yyyeqx=Oyi zT|TQm-_GEE68;4;#}g90!B?aA{d~QR@$I2*5A}7V4g@ed9>N!2EWS1L0s1v^pMzW= z13r8CUBX^;In>>^@fvKQN2JU1$}*L`)sYx+XR#)Be)m70m6^#Ug!nN9$z| z$`-$J2m&zfeV%}14Yu{d6=U(q8-eF$FQs!ipZWy!=W1R8#)ihxm6;R0e=gkj6zCr# z8>F)>I=?&h((`HYX*-v*hz4JSbe;fGEP3iRy8Cy@nHO@j zUgm6&4Yp?`TJ(T4SHIHvQfKFV>Q7}pXA}uEz!~hPT!3C0ledl~H(<+U-StIWe*`rc zcKLYonduL)%+;6+Jd=oQ@jIvY-4$cuG2eNbRDAt?@@zcEgg*fF)|}2x*yb9G=nc}n z)Uo)LxrdTi0`|GIUUXxgvGr{ib3R|w_O>QOz-m(Ao)Q}aZRg2z$0FNKHc01~0QH*u zIu)P%5uY*7rbZv3n?LI8mfT#&tPwp4^30WIK(4y81oVLJ9DS_8?%|B~35mF8#cE^W z%|7ZLC2sGS3a>ogqw7BG2HU>Qsf=*sJrL6uyI8<)Ea)Sq-2ikHWCMQruV*~R2*h)uDxj%}|kY|lzxqNXJ$pqtxtG5sU- zh(FKL*<);bd5-$clQS>;Q*?P^>YgX`8kpz0{;b6E>;!6>RQTj>)tSwC>~C!gYV-*{ zZTAzAO$U8BmpW&$-kiL&Q#m_p^6Ls|e$Hhd&Jge=aQ*p6rOw)>i@D!CTVUZ&z}f+@ z&OG6>B)2&&Xox$Hx;|f9{j+7>kk|tFUa8*3@;ub*WuK9n8aT83&JmFbpS`V#O8oXt zpaEeCE*5N|%VBKq@$$JfqEGSrbJm~F>H*(na0@xk>3%QZzXZ&aqm6)=-&buT-TPHy zBRKYeNX{N&epbB=*vI~RiTN|QMsMM`g8!w&FCpdF#_!K>xoYB5{Pt0fLB!{CpMWzB zK)qgg%nM+OZwoXa|LnnMj(bVeJHO{*yg|3mkXVCkvDJ;sBLcoWM@${LG;b&{HgD7dky6GiNG&8znh4 zGS5Yw(pr1UnTWMu+Qqcp(}EoNM@Z}Cwzk1%zO$$|#+F5Q4tdRu_{}#rbaDBtF&2?E zQrkTDHKop(j9W9oZp;BKIEUTFk`r9$Io{RezWC&J4e&ct9g(T*$5=wwKLiQe^B8w} zz;1w2I9KkG3U7-~UVS;<&E8_yAjVST{1Ln7$oCD?y{68e;td{k{nqE2PHH=6=1xUs zE_7?ujrHf2>*<{r{s{7YL$)H8{J?VG4ZaXd%v$5FCYHX&%rVz9>(5Qz0PLSS+4php zo{>4{k#eZ(Gv8TT>_!kvK;3tLWf%!)IqDnmwT(f zvG|RLSopIK-dxl5RQEl@d3`o~_7c7!zA12?EquFb0o)eL0=vfcL|nH3LxZ@CWog_^dJKDje=L#nR)^b-2*aB%L%aqNxVHvrP2XdLuTA(^`cguH$9rXsy!t$Y zh;*(x7GLhc+yTFR8)B0VH9mWp8@n2P-scvnu57T?Be1V~Xwc=cu8qW*E8uh11jb+p z%(X5e?Pb0F{P~|iODxwjTQtrZ;}5ZL)a%`h?xFJ=^Yg#}xaS41MlP4nrO!_0!@mve zXT5W6flFWtnpEz0yDYCK`Q(?w&pua?+MdbaBDo-tFPY^8m_X1ycL*VaB5wk&fd(vqtBzUZ`{5jN`3$(n-qu%RJBvLVq%%ekK&=l9Kq61h>9Szl$qBLs zS#~+rZ-JKlF*0+a=W{RzIEVckWZNP4jPwElUw%!Z+P7bI>Ud(|8N%P7$Ig~ZeF%3z zx{M2-vleNen2^A{oQv+>tXDc~&*gKs*nt6h9SesuZjnC*4VVD=tha~#JdeJ{v&cGi zkvwNmTCcQjp`^|^b3M8;fiXVol})sx>t`6^3-}t5(6M7@TPv5(ME0Ri(5(;X`dpq< ze!f%jrN}(^taXklx?C-OeKiQ^xi_k|HIdw?1Vi+E{@SU`b{?JveT$fJ>$Pig8nD2p zO!&f1;FrsoE5>qG*X`8xB*!xh#PjNEavy!E?8WES9!g`iz;?Bf3y{Y%iK%eAx9sce znw&tLJR@SFF^2S6l2h&my~Qt=@qEtc6JoB3@ra*U&N69GuPS}k*sN_`}<2vd;*5}t^(g%#5MGQ^z)kYUk2vd$9l1i?dLA{u!TOt7l}Lj z9_qHSy~p0GtLXX`NP7)HORmfL#`@RG>!U&M-{VH;a+_;T9ZNs^rcS?#e+2j4_@?B( z6y49BiJDu`YvS&uC4Y);A7uaoB`YT8K)<@4&1}o%YsL2vXRuEK>VbS`s)4=Sr@yo9 zLvQdU>=0RCkt(I61K3vpP>`JOXvY#P0k2JY-<~0&c2OrOCJc-C`Tao zMGnAlz0@zz?LWnCsc+Hcir6k~{rOqMuN}0}m2z3HJ+KBn;g`SZ{MJTv=e%BveU6+q z;sVZRPOYtkFQL2i^&E2cac5Nb491|(sqqE)oX1+1T+V8~Xt2%6`_PANJXoWR9Ae9D ztgp*8U9SE_t~mj{1-U2t)$cs|avt(}9sbweoX$N43CQ8=?mUoRUjUw^xzqLXdUKxs zeL}u5u-_0FmCG{M2r0Lm_PGFz>38Ma^-}LVBOsp$ppJ#t^Nq{GRioGFBBHxjV?*tA zY@~J0>zqP=j&=T^-?)2Pmir=a1LV{%Z-}5<&YElPy0OocZajnLwjgsLT{)i0^U*}; zfOE>NoQ3fa zQ>hs&3r|4WD}vmQxV}hDL~pg#*Wm%)E9C?<`15PccnUeMUf$TVJf646oa7AAYh(kQ z-9Ab&?QD7NsZ5{-F5~9ulP{#gBWH^~Lb@+wEwG0i6ZFi1K3Xq74;rL=@ai#MIq%rps(arVjM1!p~ci71ZQci8-=35t$HPV@a zSTFuo-CU%7JQIC2h+v960im<4(~iKrp)oK9?rDTv0BdV7K(|(Degm|VGLZw6c|}5r zJm5?T+3(ewF}}PHB{Aa-FjsrhN%M?ZTO<3Nq4BuGhi={H)@4W(l9mT%TL#@R8{PYBjk#B`{jSlU0nY7e(CbvrVu&uUwm)|U z=*YEOBf7%qs8R_HZ_N<*U)NR0r-u&K4w*N{-yloNb=n+#{6Nz!<;W%1Dj%alO=QH^6!AZOnU; zkop69lS-ZS!ajM$a;7742B5|^KTCC%d-L@lQ>pd7S)cv*T%K1O%X#+C!w{eTgg*jj zNMJ;*`VhHup7YoNzrGO54DM0h23f0vA((X1x#W~DEhz)eH39PYObmf>x$P~V_rTdM zfxXAe>p|OCK8s~}J{QO zc#Jn-eFREptHBWD`$TU!A{YQUtj|8At4U=x=d5Fa{CS@B!tH#SkKEiJ+oi7GcoR8G z{mar%UVXU^pU-8&m+%eI1G;kr>>6o)Etn%B^NL)7_s{$sN6%6}kgp-ucD6oq15$ml zj)XJeYX!cDbO!r+R^~@^{nkW$#wWo1zCOpzCEg&dYhvl)dAXl~HTXS?NUX1Q58fa1 zU0KG`)8{6oa#kbju;nmEyGF0^xpF*}d+C`r=ruV5fi3LmJkIGptWUAzxTgRn#1poox$yj6MMVt~a_^ z`rL{BdUWHS)i&`3QeHXSubdaqucEieCL+-jQopYqYp38Ex^s-sBf9mM(VZoro1fpa zVwpugzo*HwcKQ}FExY@PUQ^qWA3@Ve_oC1JOo8WUuLbfdXwY*my74j6&qV#af!ezB z`SaQy{S4MEs7d(bdKq;Sa9{(luemSH3 z1ASl%-Sdk`c_v__9pT33Z2jxezUBuo=%l)SZEMEsWwvC!z|Q=ge~1(mC{r@sp#kM&EOhTy+r8o#hI8&RLdw9munfs6iy|zI>K^ z_9jUE4SF~ZU+24wY>1r)eg^k@QuB@b@9_3g_chbT^4@S2-JBpFy60&>rTg4MzlyXb zQIpWO(H}*Qz#6>nNp4rf=YEV2k@D2YOQ0steD`rTy4*3AbCjz=cOLtlLs}oyfiubD z{Q5l`YXiFPQN!cqePM#`?;Y|vgZIq3Nd6GpxINsj`*sHTK z9B`H{hFMJ7Y8-ec0)J%yX3I z60w^6n!FKqqfT%4uO0B)SN?Fk_~mm?jWPDC-J*~2Ia6NfeXaY7snlLV-+K`oSx0h~ zH~MbjYp|Ww8NBxkbZgC^F8>ap&6IUZ25y&bvVl$jkka)- zcITP`bHl~*?|9e9b)O-Yv-7zJ1`!NE!Y|Jh zTm>z@F}RId^UN87c3GZ>7t_Ntu*P1_9U@>g*bzI$av!V>-~v9MGh>%P#OFNPkzD)L z=-%V2>coZ@sinsGTF}Ij-x6=gbxqMjDto=u)ol~Ihq@M-fPB`=GeB>U{c;ZreBOsG zr1lU@(9MgnaOC@iKgWqhbeBCJ?fiaz@hq(K-Z}rH#9QqAe4;nR<#g$9BeBjuB4!`& znK{m5jt5>JkN5K${ug6!;~!!>zcXF~@?8M-l%tIWkK%g{HTpb1`z^?c zpdrU+!1KF}oGpA;h?`TB@7^Y0yYrn#cjgA$HAQOQN$tzvaQ|1Zt*Ijk*aOg_?;$54 zo%<^BfG(WZoL+aY3BPlV@Q=(*Wj6ORI9^`w+xX4%T+ZXagzY{j=nF6e4R!7z0B4K9 zJ&1gNu)RMK+gOCp{X4&B?-@+7t#zN)Ig{sT%@tssH4Aj%YOq_dh3)g>dgk%6Piy3r zC&Y3#HFD7T6R`mDGptTtM2?Y7#1E7$K4)}}06edVzb4P;z_XGg$A~vz0wQS1b6;D) z+;hNM=l59>&fcWLb9KG^4T^aqaDmv~i<~jYXF#oIu_sFVC2IvX;2dnfsR0Gr0`pGsk6~``rc;5Mr4r5_dNHPxX`M z>hE1l=&{@j?_G=D+esZ?w#myJ@I?GUu;l6Qc@uHa!+4vJ9nA6U5_XS>JvJ9#3wn+k zeSlwC$0F@DS}*s={LGI}DffbyXX5>6@G0H5b=qTf)(|Q3-cynK?Il-(?9a&ghCq%5 z(pmFdoJUKZHNBs#95Zh36nAhl-xQxc^|v6$@dfk&Qsnt@nWvp==%t>|hFD1G5rI^1 zKm^%dmKo)&@r|(c_kG9Mc|W0WM)LF-m$M-^^c?7(pFX)Cs^4S>Um+@ftMh%i=RvxN3av zZvvWx9znMT`09iU)Uk8+Ix?(xe)Jg8*^e?{i#8UYee=S$W@jvOr^4gjb3J+P*YnEf z(DPDiecF+qp8@h4ebn1i&VG(szczLHj0I$?WHqtW8q3c_Dm^_DW0$C_1-kFG{tT&& z;q!eah*bRhh>9BI+>3Uya{9XHYl41?B{#ef0_THg^561G>w;#?5!u7AfSlkFo*!GM|3A zusvh7{Ysz-EdSnfNdf_%2zdu2PKV~?==Gj|q0e>B9bQyO#jDKYnw1b(^PhclY% zGM4X~#>K85s`uVJ^T*WL)7PcE&LCgrO2xLPHA78X zZfCiIeihxG=E-A?y++6u-vHeeBRRl2bGL}QJjVh0oyW_6d&v9&`dA49Sge=&3HlUW zss9|h()0E`We;3Uq6hvVzJTstl`(ejRpryFb0^K(aywf0WC85ICEwX3qF?b-A3f4|EAOY(UgbMog~gkwGN|B zl6WKPG>;KsJ`b#`K|K+9CzRAH1$6~=1L`!B0d+0PY?Q$$6@-d9`4t4M474+mr+hkh zx<3>fjl2fAV3Np@PJL)^G#|~Dg0?$?`jRgMVSY|%YgDX>pjb1=k34az@}F6VQwlf( z^7Qj8i#P;5n;-<~M)VP=4=s-X%l84&X_lOVID=xzqdlmooBOFa1M$dFJr!-LNwpM2 zH0s0~P||*BUh_gd6FIuKCIw?vWYMNv`n=?z)S{2(q%)%LCo~JyA)USh&>f-t!KgPw zPK&<6SQF}n`=xVB!MH9M7mRV#gL-QUbd8ESF4}_u}SN-d;~og4+P0fv=z)XAt;yPD+S>D3_*Qq4%&B6 znq@}Hu|-fF+NS}1YC$f|n1xUg3WCm(_DQ+s*aEo-r3PXCOkl17VV)7C31LLgBB`ZV zT#9OmqkaU+*C+)N1@$qXD{=~g&YmRAL2)L86XsByf`~?)VpP=4zJfU7sE!70;!WsR zP*Sc2b@Gu-a~V;RpPn!I%urCG&pQIvm-~mg0=}` zD38vQYEcgLR8U4EPktjpqgZmCkTW8vj+w){Ci0;)An#s~qeV%}px}R01TL!M?vH8c zGoViO&9#v$+&ksd=h@6-4#iQ8!hQeL0<-P^qs&w58CweiatLy8`Vifm{8X!mKn9^qi3KZM=(DR zIya0{&^9WTI1R<2O+B3uG&{WmsE!GFs;fm%oDo5FX+Q zsvnFoAqbk2>MB?(B+c@isILZLqFnUlBPd=$j`CH6JA(StzD@p?vr|wSFo$v}HUmL5 zgAp2kOD*bWz85GdhW1t|y`V1Eqn=cU;#7ndLC@EiX4yvyN}7TAXhdP05%W}xBVRNk z2Vo%Kd=y7#LO6`jU>xEo&_}gAFwPc1&oT&kBf<+~ zDW2-nqVuI^Om(TZqToE}qgvSrilaW{3r1ujLJ+hJ{uV#kv=8$gK%RQhGYCc)739%J zHE5p{OJ_{F5jomh3W9p62+E;4PHC2zOepgaK?oDZlb`Mr%~yCfG&9YYf}nlmprlzf z2zP(WT~9XFZgK7)dOTqNDmM{yzOqi1hG&@(h47mRuq!X1%;Ha#b*P5aRz3}_Rt zAV+oS-D6ZNeBs`yFZrD?PDLLk&Vbf7_lCm0o`|d1nnUc`9A45JA!)CXQvUu17XB??{v%W3)G+D zDK`Tp&EbSxG=h52J`{{MA|HYvo#r#5kLEC-twlbl0OX^+kxzr5d?Ut~5Y$I0ur+8K zQ7??8o(A$Gs29ZtAxCuxAqW+@!~SR^f)T{g9N7r!LuX6-k48y-GLWZxMRQPH@{=TJ z(=2z2KHs!x4^-EHpk-1~M<4YzDF5kcLY?YS9`)C#miKtHXZG|RpeltzR}wai3y6x0h>us`|{v^UD7dFa`YZbBGTOMFP0 zvH|nh*jF^*wG7&*Ff>3G@=2{qMKF>5uKGmZ9Xavnk&&VA?bt#u> z(K(rGp`<<1tkjEUq?uLKa?Wa+<&5aA7uM3EZbUzU>JexbLNtQv(tL3U6Z)v0qFQE4 zNy8opx*ybomKOcl2-<@Nd4fB_TpuOP9gLt_3QDR)D+u)ve@hKcsE6i5|01Hy#<6y&Hc2k{Pc@@aD#!(+PPOSQjA$!pQ+?{GLEC_EN1N)oqb~&Y!lfz&d!k(Q8C3Kj zsFpb&IqE}qfP5;#q!jouMnO9W{S;5Ns2AD9(+m_(dSQPJE;=J8gb8^Kf^ujT)}T7H zPZLTFf^3q+(fkC;)1WjWZ$c;}RjffcRZu(4(w}P0rq6M{PRGoh_1;G7C$ z`CHDK<_tp49YM2E{b1CoH~DBr>Q6jUr#Zf+i&djVMPj^c}XjF_t zJ{Vy_-GChBX%Q;gwC6%ewP;rAM~n8DkD%G|k@F6)AUf0gb97oXa^zhPCk4d z|LbRzYMXmwtQJA_Rg_K$TB>R}8=8ski%GG_5VVb`s|XDNZQ`g0?SZ6%JjKx5?#P)? zH~Y{IMw@DrP0Oej=x9@{0ihItd{je07zz+VF$FyJR*A>?sHZ#H)Svokkf#|m2(>^` zd{WbA3DqQvpwL=6i`bh5byZyJJm~G?Sv> zzOcT6oI7&nb3>^`Uo?XDLuVU=dI-Wi7uKiRbYIAyiBM5j3;G&Sr#eQgO?ghJtEkfo zMm-2Yz0LOV$Ds!R8HIC7PcqnHfjjhILCr=YIEUM95Z zjJ3!uLQX+WgF4kejJ7-K#L@mTQSTC<(M(H68+|#5!o5&k%5g%DW}#eKG^46m_C@p1 zxlv8xXi;67*A{D<6w5QJfVw6CeHdp%PDMDO9)fWsDWB>Taw%w2e+_cM2%5u$oGsQ+ zFjtGZ0VSOYad9Yv5OMyPhmyVvQce!))K@`0A9do$rdZ-=(YYB_%X!jXi8moMisio1 zYy=bfLQvA4=`6AcsH+J2`AdDNAN`C@K^cvp`Kcx$5pDD55p@IBr#kd}OxP!ge8?$i zQx4T4Nj}QUK|KVe0YQ7yqD(7 zB-J3>EDLmsp;|_iPAJJwl57o16G{Vt>L?bDYOBam9!ZU2*}H<0Y_mkYFh}eEAI=^1 z!Z{5TUl2z*GP(tvgr`si~{chRU=J~MRRR0Q=jBA-(L1>@=SLAg|idW57|prWKjH8YT>b0WW@ z5Nql0jxjVh%}cdH&`&YuGeVAXiPs{hQ4lC8hvG?6tzhI#C@IH?(x_T=1L_n1To#LAnMxdY|+_&>06Imw|9ce{lruiS7~QQyp_p)#9V~EcKvSR747b z;>~jwoCVca73}lhma1YLLQznrS$b+v4?&&gr}HqNtwl-8iEI^tdp9YT8K|#$p;T1M zvxr9BgnCec)KCoC=4Xj2^`KdeC@IIhCH@<}QxN?f#m$|2AUltZ!R z9IQ(=@g|ZKi&8~UPvYs^$Y(@o(N+q429z2Vfs6cwwn;@DbJRi({d6Ax{h9t(e~hK` zA)EZ}n5QTfoooep%B301v!G6QjcgmVffc&(~H8GF!sb^t6@zj^j)PR!CECgeM z5h}&uBENxnggGzW;#-6<^qEgVnTc>m(A?D9Kz5pCHhNB(2O)(7Te} zUsT(OpgA-Milg^jG=k=IPq*}-x?QeX$CS#;`8`z$Q6J$yVZn}Y7*!SXil2VBfyelLY?}kC_@nR zPR~b4Ge#o}sM9{EFV$61(zzH_&41lTnuqF`&jfh|VN%dWjwIz$9xbC{>7k;J_C>vO zP&cR+J+pwPbE4l3RFo7$dkaQMXGD9TnwkQrm}^2%3@z$G^^6FLr9Kp^Al#`Ia%Nk# z#8Iq5z}hC18UpI7f*i(Ez7|2}Nt}Ur)e>u)Zh8K6rtS!;5sh&apNUWlph4+`&|q#> z8rDF#W3C24=T5VeUq!!tfaSi~q6|WgK>esb?Vs|guZl7QK|K}Bqx+G8px?Eq20gbB zgb95#ADu1z+@*8%LQt(>1nrTYyLW)4uNI{dZ58v3*q81WS`&;_b zK1nCgUQL*5LO;!xf}okH7xh$;FMKYE{%9j;k2I?h^W4)cISR&UP&cBaT*{%js2*`t zhd@hJ(2tJ~Qx6k@;tK01DxQCu<(#NL@j(SPNd-ZRVhn0QEv4YS zh?4eESeN!feHI~z4<;L9sGk$+G>3vZ^`TlC0_uflWd1CpZT6!yp`UWhu^6gFj(WNK zTlT4-q~}-Ii}p&;B7zW32s#I)unwI=8gdA_KZSki*%%2JM?F;p>9kbE(vKwdF03Eo zZ`ofoN^=kNQ9q4pIb+n!bDQ^~;0#!w_Ck9!>o}JIV`)#c6xC9n&b)BXRF8Tph(!OO zHg@(Gdp3Wq$-r<#P#YUtyzlmC{V&T+1DXL71`QuEWk`d;111ew0lx)0)*NAH_xeFjZmN&htpK_HDSxIeKdmhv zU01%<*oFCpJLtM!Ee~@(xWb?x!?afxFO!GAO<+S_wDEBqVvyf-8^BCs4L+khV%THq`(G$vgM1m%(s5Z-?JWF*p?-lyTZ$vt)KXqoM7#o_+?Sl8+x?Mi@t;gZki5uN4rH_NeF@7(JjpE^Vy zHTe+x>9Sg1YW@sZ_9c(^boimW0SB1A^aa+Wdq=IN*#;T770XND?tI4ja7jro9g z1M7dzvS(~r+0}>nq*RpHY9Af@zGX0m~= zy*I&_ldeADy{j?(Ye)9)@>_lI3I>zraQ8B)RBhMVdE$ zD*Vg0hPoZ+>2_2YAdhXcRCeeeuRGpr58LhY6v97k)DA28L3U^nEt|Z`NcUdp;P$Du z5Z$+b{fkld@`XOd@INfVjc$bZgd^g8p+8V8Cg6?Z4}h zvfrtPEM`@rZf-zj5RWzRc~os(v)JOS%q|1i$AxH}{_bP}JO;+R>*lk%-451gxg+E_ zRn&F6VF&(B?ZNq7OoPEGCD_d7+gbf-LihQ}Gq(CybMWgkQGfqWIcZ?ayFMST$20%# z!`b`cp7N~W$@~tv?T=a2HZa&>^?bL-m&eI;%#Q0=v-lM-&E=(72!AsJP zn6A(3+g5sYd7ReCYrQ_sHda4=%^H1t_(Y$fV6Pq6;)_q~+s$;@|GeaK?RLuUD^sK; z<-f7cEj1A4vs-rkH$&Rl&qXrnYUsyoz9PA>2hyC{=k#_)p9YPWw%7l@8K>i^KjgSQ zRbln(?Q-?c26^tW(sJ(=%XB3VCivw4?Wb>c`K-2bQc3OCzjNf1y{_u}H$EWkZW659 zv#_;pz$s_hy=)pA+iHfKw|H8EpQ*K^N8z`8E?u;jm0@$hH!4|f`)7-8;S3k)T*G4e zivMhNcNcl;Oz#(K!w(ISS8i!7@zTCN9^p-NeIHkrv*xAi%AT~XU)#%-g{=IbZ*ggx z&#<17uKg~1-Q(d2@@=0V?CM${dCKgv@`+E6S-_)}dgq%y(wiIynAmTJ#Fp&Qd3y)x zE;nB**<_X0ZrHp)zvJy}{dend`Tmn@I;CNlvc291uuJfcxx>UKWtXaB1+LF~* zH_Ro2-5Z>&zq{s-oV0d{&#LUJ@{#xm{qa&6+Q96uatp_eQi~UB>OY+1Cixt2*VlP> zOq&?7THB{bQ{Bd}FrV6XZRKT|0lJngVI+UHU6F!R%T?JGsl-e7S>ZO@sa;5_P4jw$ZXMjoxcpf^_RF z_(;pvNU*b*eC4vUzNhmFpQC<1d}_@(?34CkP<@~CFXXuDwvt2B#o7~9Vsuu^H)|iP z+Q4cbt^@fGo3P6SXXajE*UCkN>UjhxOB6FN!9TUKXQ zTE=GWekE^R`^9_QU>oVuuK_+0!|us*Z`P95HJ_-vpYl~V`^+}|js{0{*0slL|6DmI z|JpTAsywuk{&H4?ls=B@Bkq=h-d5$oqv&tBbHovS-J$vVAqHQ`wf7u-Q?Dj%f&vsSB%k5Z!hb=jhpAQ{I*V-^36(Gc6qI?OUNC4 zYMoj7{KHA ziWf)p`n+&`%*@U@r-O~yX#F}?yr>iN|Mka*t#g)&Uq7U6acG@xZCX?5&CbiZlJh;} zV|8q0b#P57{$X98L7~>V8~yg^{QjMgB9FSj`4aq5tOU?{z%jv*2=)yytiq z_B8E@)Y8i2Q+D4IU9u)lcX{7?`N)}AHb*m9%IaBHUaH3EdptJkZ^jszq0ClR`&k?L ze2*6TPg_k5d@9t_C2b$2_n9?D3T|~Pp)m_6UbgTW>-ShrY28diy=Nh^fM6%XhiKLYlY4uDzG8+h623VbM^ z4C8Nyi%N$k2-kk2#7r-LZd~HW+2SUAMa${%_Sjfx_$vg~Y+lF{v!?K!7bf#hZ#wcG zPbzahH4Scd>?3pw^^HGw|~BaSDaIc>w;bQ<5{+R z#8Xd}`n@shv^Gn&?wAPQtaiaq-wSYh=~&U|^I~!KeVoXfF@#NiRu(Ea)`OaF7lGH` zNf7vWGV~umUgV7EA-w%sh?L=n!1~<=Xg)s)G|vM1`_ zq7I+evm75-xhU^&Q76KD+X-V-E3xv!O4x0e0XKiIfo=hh;oOZZsOx?P_P+BGFLM0E z<0KHzX3vAGOU}TLYri1wryaCzs0Y78fpEm3CjW-8O)bk8zTj}|bq)Bl!U4i<=kds! zi@E#19lU%GiLcnB<*&yz;=h)47rEK3#GE@VL}joO?`~EU-W?l>_#ZQv@4oNs6zdH0 zZ_g4pn#>YU5BC%aWsbAA0S0z7yCVx~{*`4Hw)@{PA_a`zpqD%;4y zvX1k1u?KiZ$mAuzB(OSNs=%y;wPAXr@1lQ;Z({$ZTrsiICXhzXgL5g(;D~Q3f4p%I z-_<9LPZ+xjM)fy>_tQ`aV$H--e_!$FueUhbV+l-{x(gn#%MkNEUHr7(Cq7@^FYKRu z1pkR|py5FW{@Q(n7?3wYj4L%=1Z*?FF>MvNU(*+QjA_A7O=`wxX>~m8eq}bO%1L?h zn4R+5D+-(&f!`PQSAtQ7a^g+Aa|0T;E-wB}brdC{Dv5!g4E#u?CA?$t z3A|F^TefEL9yYSwDmKUN9JpG~gHx?1LBitRzC*J+`X+`1`o500V(Jt(w)tBl7UOi9 zJIvX|N7akv8!ub);)86tbk2tFEg#KqU5@2`TPN_9udPMK@<04sb1QM|j1w#Jv;^xM zH-NcqZz%p_>J(ONuOGMXgAl|%XBE0QhzSZFx-x`&`!yc9w zo(T@(SDOlAbaxZ8mbS2;2hXv>BR9xOVRD?fkBK*sHiQpt>2zz^u(l1 zaO+hFtT#m5`d3v{Jbj5jDgPMSwoHcLX)|HD#uMt`XRLI$)-cg4g1_iEpMUe3#osP& z55+!}g^F#PL&dVqMEXt#QT6pz&ccpC#K{!+Sv3{1W9o~lvG(F&>^)wLcjUQ=5>E{j zJUx7r?CB9IhnF8Bzx<$w>ep|wKCR2J+5ug}#k-fCuZtpl{YkethgbesRq%o*ey&w|4%=r79ZH zYrHdG^39g3ex-O9|8_j6YBxyS@rF6a*mKVpFX4OH%G^}76fZvg9&~M&4*3iBL3`(R zFtK`NeoA@?3DrYjX+VFNbtf5$96bcJ{gNSm;5y!Ue-zJnIgj6R{lnVTC;`LMv)RzAbT=5x!}o{R6YUvz2ujmdb7>76F@L zH~7!D2lxR0Og_B-X5NA==D$-Tc#+CA;N--M?DVXYtZY)8SP|Au?AupTEck8>)yjTm z8+#mMoonCb7l$9@HLf1zaR>b%*;|6@bKRkHQVC!GLBB-Ah{wXXAqmdST?+GFuYi*N zt$B1?NB*(GPpA@7o2PyTesH%NUsftobi5WOX8f8hq65}IG548ps_hhbykVKB`MsIA z9a>sU{5Ar@hc$wPDm|dIDMZZP+g>!wt1lY@-%6))Tl+Db&8^FO4Ia&ti?`?Fs_D7&fC>EY(?H%Q zrX}y6P!Uq2JYZ^jPq;KMo!2bBhiCI--oJlWxVrHQ8|rwReR$>}B1RV%$43?s)s_mr z*u|A!S?I&vayzo*yfyMa_t}!$XoXu3D9fic_T$r=t@QmBx5)SFxmmuuN;|TOc|&El znMI|8!5zf5tA65hfFcfMyYVwsO7UG;&tTK_-hBJa*8JDn!QA^~XYqKsPV5gYB|>K} zk~5dtvVQxmm`m;EB7Ly8*pO0He4OshAGG5UkZf7^3r6}{YiijNGD#ESo_^E7LzF@HvciS^s zTwUZV7EiYq-8a@{ho(J|T0OOsinjd6wx3A`w^U@;qCdEnh+V7RBe~Oa3s_$k0Myc}X z>-(hDUF*p5zD4q^z{XHvk2B2wh3CI97CbA*z%z{jF0(Omt+MfQza6dEhw-tXi*5${ zf;O|pp9I^s(1rEuvPeF5?-#T<{1KKII)Y7xKa8Fb!M0DX%x>3eA}?6DNBXnEO@{1n zcwI*U@3!k$ktdyv+>$@*VlB3&JlXsNsEZ2y-0KvH~{O>Fmek{8J z59qrAQhq#ONz)}Zr1&?vo&KF%+UhVIb({->+Pr1qmj=q$bJj?EKHt_oiLcJ*w|NLQ zwmvX6pqa>0i-`@=em-_ZSGfMk74E0)W+R@|=2f29@(o_+;dxhU=G1gctmBU|-Ve5CE;fC(vJg0^OpLTRHE8V6%yiRJz##ujygNru6 zkDeYd;v2{w*RDvv&+eCo9bW`C`!+yksh+Uu_#u|D!~t5)dd1v4d-GS@e}hf+DG+uq zfQ2Zotd`Lp-uPEzljrG}OZ^4Re$ofF`|x-6{qsxqeo6`XMz0t8xH)QwJ9{tuy zzaP5Ewl**1p}&))aluz5dshi=9(lwLv`l5SvoqPQ=U3RK@%>oHzfJP{v5xZgn|Ji5 zewXImiaPMvvd1Cvr>@rgHC2vrM%>pv+GK`r|*^a#v}!vmb1zrk9kPHe{r*z2v5uFUCRLCp$jjwYt&DQ{(2H`D z`RuPxjHRIJ*$pmi-<3Qp){4O1@JrT-IIGLJ|2D z8@^tuU$=eH zB5rDe*nDD*P!?Ll$KSWuy%W!vL+L)O!Kb2Z;G)y4Nu%9-=%4j`X6*!isPb=!>Eg(T zU9Hb=)^di7zV6@vCt0wl$S z;9Fam({m91y1syCb_(W4Q@Zo21PqW|#AyWaDft&+KO(|YhWsXLs#vI}DFZ)A^l z{b2g-9q4bKOY{8^z%)Y}a_GJA|9$5N1uiNh}FFE-%Y@c)s64u-B za@8O6qeb8Hc}~_MvVR))JD0*+C#CSmt6uWmpKZm|dhVib@=$SZ@&GX)u7^;p*U3K; zVr1u|ujH=2MPTI;CzyV@D3mSL2r9oT2Fq4OvsKnJ`MEa{eEFaA{Ek&F+viY%ZH;ur zd*jbedHJ)qdkk!M-~J+^S|_o7e`hiGgCuA6%Jh?8(AUEOg5}-gEL&-Ysnr@7&@AEQ)yv!pnjG zZBd55d()gBKDd(KP3jE$%WMSo@In}I-dfZy^_KsQf5WE*Rf0{8{;^-he=)-%8?ox2 zv*>=cu6S}|6t_(_@J9Dz`JwLCeE6^T@VLers95?FJGv$w7fgKXyTb`HyV@%gKH zUDGtaxzv2#%3mwd-1Xvi-nHRyzTr_Ium7nlAK0Q5pXu6_ulX~VJBIXRw>(X0{_+CO>~&(B`RF0FCz9|lXB{=ll&v2q)TfL^4j)ixUKq| zZ(miGCu%G4{;_p=|3xpj)2=)G!n#MigXTIP{^#4GkxWwMpZ|>4q;@VwWeZ)b>0!{$LSa-Z>D`ZMw79pSrR4u?GBku$c8TEn#C_ zLfOLZJK5?I5sX*pKt|S<{r}!m)HO(ceV>Il|zQx-p) zP*yCcR7%KwTJX^u`t#RqrtqsThOoMe`>>B2oLINI8u(SdD_FHihl=HNyh+iI;NfBg zm3wUEgWKNa%CfUOx4MCK*#C;%f8_)}a-tw@7AE&^kvg(MY2@S`|J%=ksWI#$yK|qVw3Jzv67Vz%UdtEWu@z1Vpj}}!EWnZ znD=TRoLd+Tm0v}{phY0G*II}xKe^CEtl}H}7V?opH}jWUbo@i(>HO*LP`>P4Ke7C& zB&40M_@UM(q>5F(OOKYWlM^PMV)HLvXKMzfv*~|5VBfuwOs(L~5~G&E#4{(svy>I@ z-RdKo8LUdzOlZD?UnfIyRSP1oe~8jv30& zpQ+Bl>ozPKvQ~~-$JmJV5809gh!n^!x)`k#(ByD*2>7#YfnY}(3lb}V9hHZ&1St$T^qPsfU3y#_&k zk##IR_Y5nxYCqpI_AMVBT1;e~y$LJ!7v=hmHF!XyCR}Z}9inpgur!YylHZXZQr69*B4Q9EB9aVquKTiYM&5&9U(cK6OR=0&4yE?(^=t=PG$ydoG zr?h;fc&r?M{Sp*y_#Pg_)ZxoN8^LkPGVo3Dgcr5Ec!k&9`J3qjxsO)EKOI>JA>kFl z$2owFx0}Uwp6JB7yn6}PeSSf_W((-1J`zpmeGm`wK8R919?6$qE|t11=p`jbKZkF% zOz=^z%&R^s2}M5jh6%32VEd@&Y;@joxp}$^t7bcsuP&CwTV5zG+6MW=)A(F}N#K7-F|zl@)ozl=w{I?i__oZ`E~qWM|VHC{3G z8Xx)lHqRYbMGxbIDNXr2b((naxS^AC~z(8f2aS~is0`3&MOyo1Cg?_e!|0pDlWWL>Y%gU8_u zpn>;vD3jU~>b&j(5C7JOS$1pS+2$o6R_C+bu?8qvY(6M2=ELbH&){O8Y-ka33_iyW z5Ov2-5Muo_k@9|?sML3rSUGZ?Sk$kR==E{1$iCQC%xD+D)^zF5R<`ogPr6p$}*}3h2^O$M4MZ}b%IW%b^8*6(L zmX*E*xn&1J(z-PMJtvdb&w0)#E*{ES&0NT?jd{oR%`e71M^xcSqw$>?QUhl0YtQ1( zII=7)t|kIS{^{9y;a+rjCP4p8cTSswYTCQrHL%sbX@3H$5S09VKF z?AJLr{v@;-FSan4AF?%psaig)Sfb${wrqgEyY|A{5>@%#`h)nulF@wPZzFfQ*Nf+# zTEds-#qbjI^7+avM=_wrF7CVk2>YZ>VVhkeS*0fj*vtG~Y}L~kw((9ANcH5P*xhGM zz6SElCu4b)R^54)^$4E&r8OUUA&}c-SA?oNUa?8<&$GVKF%X&N$vsEM^SD)iSlZJ( zmh$WzYtkc%&$e67o5@k!;n+jT%Xzs}f5LqBCas6i%CF zH|V2WXG?Wwnb*sQ;C;rLze~iwG5>18%h~nj{r6Ah6YB=Us|U?saBd3hDPqT;s`mU) zNmB{%}n#e)C2*KJ{fM z-a$n2j$;G(q!RY>=w21s>2+n;n593UQ89Nu6-*E?`hr;K`dF+SeO0uaQcb?Uue2QS zSI8qTzT?*4e)BR@^0+c?IHc7W4xa<3!KOMrc>ZO4uJ@Yrq3;wP^>Hw_z7flNcibmG zFXzFw^2hS}@%8xCvy3ZQow+xD9(G67<*VnW!?-dNdE|j)p8fO;f6&jspVeN$E03Ja zdo-&F&-~Uw4u1fbJKC^^%f00#PweEFuSJBLcX<(_uOR9UF9*NlI1K%?5;iTXC031X zDQ3pB7t2@Ph3!6@!9Hm!^coSy$L&bt_J50sKT;06xV}7;%I^tY{o2B{k^4a(QIqE# z)56}tUJzM&5|r!I9J;n@3wf_9!}TlA8Jqru^~tUdhIj=^RO|zP8peX=Q9Pu3je-13 z{5$Ey_Tu@swqnAe<{~e(sF+Zttl0d$g{bpD;eCtv@}E6c@SfMIv*9+eEYamFdvw8y zcU-N)%f>43sfQtS&PIqxd<_w&H^P~rQ^9z1G4zwRGTDC^`(q!$LVLw>m#=gAuC1ea z(=8>zchWu>RMLmTTx+g-X2Y`|yYXLlTEm-(C7{Hux2#bW;NMpM09%)PusmJEZy&Sc z&zctF1M;Ryhkp3Uz8@lGZ?8Gb=GqIkJf;$qxmW~@^D4sTyep8@wmK{=KN$bNi-~aj z+!CR>7gK0nPz^8H^ z8!J6w?%#K^?d`0gUG5e3I%XSt?p+o-Sxx348x1_+%2@uV*eA$e@)(|udkYiiCBlJ2 z8Xln!5w9ktl_u4NX?-V0{;!+|Fh~aUM9Y0w3Ys zg7+Rf3Ffypz|6BtAulh5w;ef&U-fCorwvVlleKn3{Pd!{Qq7jo|KCcuFw2^E+L9*+ zvw8A$YlZEQ4KQ%aIat&DHNdp~_(=p1u4q8~yno6QRpsu?ORIH@D**4vpt- z$I_Tf!hQDmr3PNFbAei!YoLGh6S$Gq2SPVbffUcb5Z`A7+t+3x+p>N!bG>;LY_48_ z&}Oe;&!w{b>dvD4iqVdD8B^9*TGPn4@yNcu)e_Euz4Lpp?$jBUH0#NSR>^`C`<|?L z@_u$E(1~5l&6i`UtYNQH*UP6JJIN28#X#M4kxDwW;>3SGg z*B;`R@LG;YFP1@(?Ez>$Loz%w`3%k3M%;$V}jV-VT)PV%TEw8^&03JZRZ&a zHbVD^9?)a%JxEEIz+NRaW?9lw+3mImuVdGWkD5`0f86IH+^l+w{H_rq^667KDN8H+ z&+P@BT4?yD>^ZQadOhZNxtw@9pp@9Prk+qlQ9f^VJ~X@i3%;yB$h$R3;Zm7Dd|^61 z^%l)|#;2vcdvZ0nbaRGWC1p0-`YAy4+&fe7T@k{$+Cq3dDvIb;I@$4biOjnZS zJ+3F$%&IRB86SfG#u>#I`5onxoOgiDfL9RE@jNu`FhRrx%@vyL8KS+nqxjUck2sgu zMQm+}e{V}&!8dy;{Pl(mv7qV^(OU@=UkCWhU#$}L?OTo2wQ?B+XRA$tI)1LuNj!vc zhcaM(^*-{5d1ZObq0TTgy9H~>M!}J9McCGME8x_9M>cC^rd}VkMDn`+Rn$D<=4&<2 z$9Hp{mi3raLhpAaT6%wZD!1LYk^9Zu!E;A+V%@r^vUSruc5U!a@jcnjcfln)--A=% z%Oh>vq_XSk$#=SW2)FdwqSw>=yiKP)u;Fn%c;|GPId|>Pw||-gzj_M3rp+(bvuGGQ z;&q%|7`|F`HXIi3%V&t2HAX^M*`;v1tOTu_Fi|OVikSXwp}1PN5x?2No&QU^2JIHk z=YcSY*WJAyDvS=6$9mnKrb%t?evA61C#a?XfUDJBwZO zs>aVmMZnz3v)SapGVE%HKvrLKSAOU@Nj}kN6Tk6s9hXLi@h`~>`PMr_dH?uKc=Njl zw_i4szjM3I(+=<8+U((cP_1v!?YM#6nzMi{9$XPVCjW#5Z{p$I!nv%WU47sC{k(j? zzVq-c(Q1tx^Ds%yZ(x+qDg(Jqeh(Nke>Jxj40T?OBMfuH+PLA3s7dV!C8zrQ4}Yhqw} z*E-C$ohLi@xCJ=(9}Y!(CcuYFe__;s4zf*&Wvo`zbyhQeF}pLMIkMLn2o5m8L%8__1rk#ihN7c{@;iNgE|n-wu~=Ox(qGOk4oB zH`wwnTd>n)wxnGZ!+{K1n9{x62kJ07bx zjN{p4ugH$bO!1uiy6z{Hj7lY>k{LqL($F44_9kRR$Y_um%1lNt4QWy-sc0x|sowL4 zzaO9Hoaa9Gef_?_?>XPWc=*mB%HhICM zhQsfQv5CCFhmFtRrC=zEKYonsgr-xz9}{@Xh?i!{JcDUx)}Z~25!`L|gQG9Jg6Mo= zv$nQ!<}>ds$X8m1ow=K!t1}0$%(z6G$DT9ChW*H2&N4EUd5<~wq?h!WTH{In7S60o zpP8iox$tc&8deJPps->d_BHD;&mIAp?|B|>8hMcAks;*VCqojxCJ0kBE~C1)5Oow| z^R3T3SU6)E6|Rf~@df;-6L|my9NR#_&joFc+{Cl`(%|7QkB_ch$JmR(*lC@DomUER zU|11MWM*M%!b~cz^@r?yUjeXJ8HdAynAxRPL}<}<7-93RsY~}D>Xige7trLccUsGR zF}|P6ogEGBH`zJYer2pO6vfL2&f-7gKD@DEA)b3Ghf5<=P^L=)zAXb#*pdx?c1rli zQ2}2y8l!CS0BLwu53VByxHDE5dPgRJZ`A_ik88m*Zx&NtW{~>LpGg;7OQL+y>Rh+A zyfoayg_iX5P2f ?=UZ5G-WChIuA=I>)1=v^gxEW^5R;mP5aM+Ok=ysYbGW^P$-LDam-iL4l|cd|;zXWpC-yGq)nBe~m2k>idFsfGL)$ zMWCgOG@jUa4!<;1)1atwT6a2+8eR^^Tt9!jZ0CfHvlFRiS~88^8%te~-iAJ_65 zpyVDdtaxw>qCXS7tK~x*u65Ao!^8CGyHT`x9f@1gGO+x99;xGF=P4sEVdLdc44JEf z)XN87*cOA^foSlT%ZBY$PRJ4%IQ!%+qPd z`p~qscJ!E#53D}UgXyZBC}p~kZaO_dTTqcZiPy=SoYmyyv;Cx7bP=5DP=bwS(h$2u z3v0ThFpqaWYRT*dX@6f-5qgD#dwTKu!33=MTZ^$??w}*C43)BOu>YDly)d|w<^*1* z`5o>s;Q5st-Bm$;#(cwl0?X)xp$paQ6e1csc}d4z1-N|cEag18O;aT#xeC$ku=nw5 z_;05@z=0av5T;2jRw&a&ygJbM@f+-KW#_I@_LO6{l@%qU-79e1nU6p6FQUe=ALz}T!1016eE$0}W@fh0cnG1}rYz}n|3u7O zt4Aa5n9*OdugH_JX!16(_+K zyIzpB)9TTwmXEH+V5)NIA85MgqlWYtF6Gyw(xXoF_K-XEa9E5HM_yyZcPV;XQH}Ob zdeB$=>!>Ir1$%GL1qBU549E$l`SXft)zBB3r(}koM;2gKl+ibZZzat3tQ6x4{y40_yR3lcZCdRa*4plGNwLg3=HoWqmNb?YMjxg zkH2@I!uRtq*&m8Q1q9`%ZO70VCkfA|aPU=r4Vz*fo(7iUU}p`gb&tT<>dSEDmML-`CZfr)UPkcG2jw2#|YM}iG<(HwWt)#M<2dn_X*SLA>mXFq;(i! zor^F1er_rC4)LaqAMJ404+ZGvyTUm=T8)mgCU8}PEgeW&P4_ny(0TLKxkl2pq5lpUPn2aqDYNPm(ra%uJmR)7kREm zG=mU@W1;;xG$gG z!!6p6!=)FXna7zdUEq$BJ(}Syox!W|B zaND;fa!=luCQIF_nKE%XlM}UQ>c_Q9w&(T}}aOQ`8G zW6E`!Lj}C}sABFR^5@WUCe)^vIpnQQ%{ry1X0a9Rt{nqz{B)XbA5KNI8u4{X7`FNf z;AYWmbUJenr+52O)9!t2zWfTFu9QOSN6EynY!g{uGr~wst)e=L`{x2Vi=kf4v$pCMhCD_# zZm9!tvj-QxDfoFf8UI~8k3Y_iVWN!+Cs4DPDY?!e;nP~#d_WRqr}kjKeiyCpct(fB z-_WS*>(J6-J%-wap~vpmXn0zfKK7nY4=VY9|40znnBRd^jap{dGnN#*QG`81S!lDR z5A~JhXhzB)R>u8d*C%0mHaCzc-t6K!9Ey*=r()VkO^^_}Np{q0!az|3 zB$wsEdfBINHMW9GG?bIYHbU@-!{GT(TQFbzBrfpxMzfe@=%l2B3W@*7iH92Cmr)F* z=S%T~#xwjfXiIxv7BWw-#4vo?l4R=aVImw@K&He?i8P4@ciUs2z90*V?}WnkgBM}d zkO?Zrt6|ZkDJpJBz?EtZB);+l$vG?t^#fJVvuF}RaW3BV7KfMn-;m7>vta2qAG{-c z7Mmn*V4}`WrhL&!W>1YMcxWin*Yb1erKcV=c)lVuzTpLTJ!=R(^aX}qiKE}fC0Mhx zlQ@0#0H>-bHZ%T&y}FXLKW;J00?%eyt=**0;4ewAb;jlV=~%R|7%%bnYBgE9h>$i?N$-&MYz?VT^aYCMPGv zp;tB%q+%z4H_HfT$a^DMWQ>DC8&E?z4qxQ^f!~~?uyx-HNdKNiWX6lg=GGAMZc8N2 z(1^w;b{`aZ?lb5MNMK{FFzT#1NVNlw)9SfVbi6hcbDGX!@v&ws(amA{#|D|cwRf4s z7cWUtwm;C@k6~c<2`cakM>mctlH1=mk?p&Vko|4qSgNxE-@eMhi04A&$JhCU zM3$1i`=OY(IRqs}Phn(OF_Bsn3iCAN(K=-XxqrluJTECFH@5CU^RQ&>4y?y?t2hqd;a}0KDC>3vR#2 z$FO@f=qUOUOKC0)^WFi&(^GJ{)rHOAQqWf83O;|zPm+~NnA(}wAbeE}eSA5J-nyf} z_3RBMBOCHbkuwj=L+v0JJ>O!&-cD-lm5fbBOHo(B0}o$bOLh;-5ShRjICW@>Ozd+) ztsF^e6TF7Z{38OHPD{|ta2YXM76t(hHMq~u5sZaC!!GxDy5>FG$C1#%N9%g&(JxZy z^ePK?#pY8nxn2laZ4b_jIZVHClYDv3_DL4JgJ!q6sHiWCAB(TS`g1-ozmFGw+8@Wc zRYGL>m^`d3Ve@YT7aWLhgoCHgLVUp?W|w#%=4wDOhiVSU3h&^VZf(3WHBJ9L(FFMJlXFH=S@ z*%!cQmw_ucgZ4xorE~Q8X;)z?Sl~Vb z2VLW!q&|?v5OP7ZM*_Fxe}U&&+Gy!tPyW(A(o(ez-@Q4BiiS?up)-!hcdx=dYc27n zr!`&L(n4mI>4BNdZ9Fh-4y@a~jXr&m!kxB9j3mc!nJwXI#DB_x951~{?CBB^EtY_Q z>aXD4%||t&j37|q32f2VrIqb>Vbb^_IQ|wvS&>xIBQk^eQND>R2(*K?T}lvr=mP4L zGdNmrhk|Pz@b20%5Z#@KUYCy11sdawEpvvvSkuQ@WFLcMv>jgrh0-H+?})-GH}c`F zBIpr5sQl@I&6;V{PFe`^U+My1!A$%ja15WCgyY-Dzj(s;I`icQk{x>SaImfcl0T=w z=H6-aTfjYR?>mD+xtCFS%ox?)I^)YqVHy{s3~StiV3~Uu&S?Bg?oJ!T=h~0y_P_fW zq0=*%iSkO4@>LBw+nqq;XD>$m$cCRniEwzXI&~(w__etV+tx8`=U_2ek{CmVS6E`q zzC7~$Tqaq(I~<;Qci~q56zozT!96bXF>25N$Nw{@0o#4)RdHo1>068%yXDcUF9>Zq zlIi((*EoAqk27Q_kyvFb5aqxWxVZZz3XKRNt2v>myx=mO$3c>!CPZ1b`9>qFU(>p;-LixB8}i}v5aCZRQPF@6GloYXQ%n=f2Yk{=04K$D6CHs_f z$m}cOq+~-q;R_alZQAUaz|{!kKEEKRFHtDmu?XI%OM}KvO;FotMV&vW(ZOm(YT0v$ zHvG1v4XakuS5>XJ;c*F`TW}Lgcas0k8H{+&HR%zBx0n zbiX4^XcCC=I|3Y;0V1T-O%{Gof$` z0QvUD7AE{k(aR+cpXHrJgQgE;Lc5bp9mpqblJ4*!dk)-r9t_7S-(l?6Qgrl7!8J(^ zFf&sUvd6_>pJ5nITVjd6^|II&-Gj<}w=uD?3{O;ifR^+}pf&LfjCUxq*Y`IpSt^HG z8%@As?lh2)*974>S7_W}4t4UraPxu!2KMhqxgHgIXx?Lx-|7n`r^<=-4pE4Z=p#8! zvM`&gg?39tu+MV>q-5KO^VDh5({hsZnA>CNU1?Mh;K2~nK3Lh<1sp^6{GWaT-xvBJ z=hiViVf`9%t3Sc?BrcX6F-IQPnV9T13oA8GVn|ai9=S7{P9F*TrpgS)nstrM+cO5z7l|;s@&JdUF zH8{0B7#9k~;!2-Ea%<^xa;f?cu?Z2zl_N6f!ZNS`OK&U_ix66siB4 zS@hNc5BP7sA>45OOC;V6vuwmmxIMWPKbZ%DeVPlTcCUo@V(+l6tQ${1s7BYXqA>nt z8qD&U4Iawpkw-Wj*JUT61K%ugKQ0HeAF03+;bxS)k&9ZNbMfr_PPpP+2*by(K!~v| zbm#7aj)x|A=f8ULt+b33th_+hEpGt%xC^j6?HOq4h~e}r(ik>ih8p#{__du6>m_f3 z?}lPH^*0Kf-iLyb^F>luRYbZjJ|LYk3|d_g$0Ff5_>o)4c1o_o0)@XYRJj{<3+&PI z%q9#!{t>D(1+gP<8CG>Z;G8&ehBJ|KfSF^=`d`!z!R^$uaQSEtRhnB(-;A`-XX<+x z`ImgmY_px5H9d(a_b?FsgdH(8%?lE?T0+f6D+qijglWm_8U2$I{@mG&DLI~)V`6}f zj|-V>&&$low2O?t`f;r2cEUq;0a%q34Yrd`aAnq0h~CbR z!E}&5avbaa=Hj^(HK@}*2Z~y?U}9l7_~hx}+ctSTbZDw~Lpunbr%A`jtj<<#-s$Y`?&m-J3%jU&+%;a*|Y2zn+u+_oQj9UMQohaF{r9 zUlNU08CWS8hyhAl@TT8M9BN7=>+VI6_9{Md`l1mG$urQ^9tk7L+d${cUdZmwgRn?- zB9vZ9J}OlaCD8(G^^3!X>I3MRZ4S>}GvKjZG3+*G4NIce$+WR2#BO6Ua`vxhX9x4( z(G3+MepiJ_=($IV4!$7z9G2}K_oh!`}RU*a}n@4Z@?y-ApGxw3I5%73LC=LqxRu7s3{c2c+Oiw`kpA0J7>1C z=l3Hh5hjBp1woudv!A8~)pt2X8>To;EnhgLyGJ=|Urd8aMKQuZmCsaf{z0BPB*GKt zYFN0QgIB95=J2k-_NjE{oO3@@@UfREKj}iHs%FtKBM};Q#DM9Gn&hmQ>SdyS$xvM> z3EC_03&%c7l7V?5WbLm~GHtUBE$XnK1`ZbVi}4X=iKH0QlUK@kjcK8v)ooBPIt-_k zZj$LLw$Qf28(f6GVD{iwyfMyC=PZ~*ZaIe#OJ{Xb?(IR<3JmBkUvYXP?HBC(Isz+0 zuY%j-cWAHi051(*Lu>CN%=^x8=F^IY%x{*RJr#TqqMpeU;h<+Grz#JdN=<)dYNMab z{KQH|&Gjr3^Unr7YRu4dvm9R3T}KjUUnB#W?^g<70-po3{hbc8 zy>KIQM2=NaHJl_B%t_LEzL1V&6wt)tv(*2IF4-fkL7aRinYOSjIQn4_zOvos(WikB zWuFEg&ToY>k3rIzwutEMSVb~>?m_srD2UTwIqIY>#HCD_1nNdG!fZczfm{}f@Pwe% z&(9>aayJABM1W~c8qS>-hNdqL&Sx$;XP8RSb=P&j}rR#Bt%BOB2qi(_hnO(kBC@Ath*3Z*P|3CzZ zGz%e{_iln?;%txpdk>Q_e2yf|h=<4T)0u<+UXtd79uP1-LoTHD!ujNGTq>bUntqMI zwoE?i{lpaR91zFnc*TQ7KXd;FoYeCzLAHO_Y08@Du(7WX{lacuh%)}xw-sUXGe29VkwIVnc zBTA+x<}>rZB@n4oG7xuS9`QSQ1mZ{IN#;2bu-9va!x`H+vk!|B8?!XB#Uqw!m#ZSO zzm~&1ANCn=GYEed$q_xDeyzkcb_ zR4)T!AoiEYNk_xmH+80>_XUW^5eKl&8Uc=vG?mhgrJa-VboVPodM9Wx{q(4Sd2}+0 zBn;X@rommvKG+Hgj0vvHsUa4vvBX(w2Z{O=PDG5CftVpX7xj6>6uC|k@1{T~Rjz>S zSF>=YMk;oce}NzEsu&Qu90vlA=y&XmU14^0WhDyw@zXzD0 zjw{Ul+*q>Mu9jpQEQ9i6?V#T(f{GE{%%R|;pgG3@cYoXu(YN*B76;((nQ6p5;34Co zxtSDtPLfGq19+Gri(gi=j738P$?8@j&mXu0Z|rZFPh(JiRyV^J?#HP8d(7Z1IaJSI zgo~By@$p0)S^Un5_}TCgJiG*MX&YewYIZiaVt^5upGNldYQenM#(2Gqi{B=4QN8{+ zv-O7^5exW790O*8^%FZdpQ4J_B_q(ZHxG~M_o7zcS;!Xt1!@1Hky$m5TyGJB{^)C9 zvt=>l#yX+O{4s1CZerT@YcVT zeleI`V?a7gx3c}d)ezSaM!r|pki@0GiOB0h*m%tVKP0E)WcYR9s#@aZ9eEf;L+N+U z3F;~uOJ9CzgTf=>$oupeCNBWa)}eWvyON5GPL&>+cGU(lTX=DF-2xO?`v5)l!|6ox zAmeYih~?ZfNxsO=guk*Mx_B9h7p-DWR zoq+RK*O9a(CrLLuOSSHmgQi8bki>qrmrW=^M|N&gn1*SNu9z~ChC63IMuh{b!B|}w zJ+20!YEwAp;md2JR-zp`dn5_J2p61RDB)P^Qk2gc#F*N{bYY()`2VN{`7%@7w7P;M zMJ|J<3DU@zu7O04#Sry23N&}ub1|)$%b%XdjZ6Cq{<5E-^#kh;VLMWn*uKBQi-Xvf z6%04|G~nSeSCCzp2G$E|p`r8^qzQ#Wg^Lo5vfh`e3SsoK7RJP1T6mz>5YtDL(Z+Tz z-dB>~`V{eVubuo%Ti4C!R<2g%8k|(%23;+rThC?DntLU54mTWj-{}I~i#5PsM$z48 zK2ED%fW|wY(9$Js^yTJGI_DlG(#=xjqJ1LSlT!(AuO-7Vt#hDjQHRI>CbGzE8lRi2ERTFOc@}+t9>1Ir_S$7M+SKkOy1Y zdtxzR21c1vzs@qVigG|*whV5}?}4UokyQLz1byF_L2buxLd=Q;h`e|Vw%eNF>pD|x zc;kuhmXy)0&(f%wW)e-4drU4E3&PX}Gq_;vNUi@Pls86!+J4?fSF3o@(*sB8jyjZK*_wLPw^|JT7Cw;kY8}KloX1aYEW1ey z@M~ixs&*vce@hZk=I9wL5ZQ>o8j^8g=tf$1(Tg5F=RqCX-eA?G4*ciVk1Hx7*v~se znO*VJ`sWJ*m%53==pEAh+5`5aG7zkz11byKV5BhzJig5Uv%U9GHR3VC?uQt9U7Y?f z;Gvd-Kk;6PHiYcagrmZ1z&>V>{y6xIhAe$VFX{+^RlzN?t5FvElmqF^TpyZMvWeEp zjX-B;1IRk0gOuwk?!vT%+$BRC?s)Pa&>I_s$kkJj`Q|>hjJ9A>{V-Zhc%k}#I=Dg4 z4-d|3rQMa))Yzt!D!u;($0O>1ZB@gr$B|S@$BmAji=vk}Dfn(60A;Uw;ZWp$x-#06 zCX215pO=SY()wVWkPpGhO)e;Cp^qKky)i;p0YuJf!nAR=ldQkIkFqA zG7GtZ4k_H$T`^q4ce$i*-4Sx8xQBRNK0+gw9iSn62WYg(a+D-%@ch=Lcs{=zXYp3x zW5Y(Yt%{;=_iT=`aLJMF0;rE$-hhs`&&~`G{;q%A{a^{pjtH@_U_ybX$?!j>zXnA@Dl^!B}u?{ z=>*g(c!TL&G1ON127ip+!(`Zfy6D#pnjT(Ga~}Mndv*-c)#J_d_6G+_H#5}ltQJ+Q zwS$v2o)B$k3af&oFg#-tR;s>+p{I`Y*Iz68%54D+v=_rohSTts!F?!B6M_J_x5R5y z5US_t;rB{YOa()BMq~z#E3T8JQA6aP52XtFCRAsU8eKM-0*5k>!1c?n&|U6~gR2*! zMD1c64LpQz3+yn;(E^V;3NpJc-Qx&UpXMZPizO$8TZv838B*mMO7)v}(uqVH8dw)d z_Ww>HV`X9F$(NgOOtlMQjq+h4%8lyGT1vaykw$6{gG<3x5URfj$9i%we`O4Q7d(zB za%|7%(*Pt@J%?gHef-g`k4hR6Xws}p*Xn4|FY2mv;+qV+XMF|ZVcjs}tr&KQ|AwsO zf6(0U6$G|jCl{48fvagkk}8!*Q*Id}mSqmdy*)w4!i3Ge<8d%77WG4d@R?i^OqiU2 zbDbd|U{;S69St5iYw^Ay^43)^WfLI zBwQyQin%{K@zRqRPWXbmR~R7w1|yNuN75$b6^NP~>}sNuT1z$^I}ZqBO!_i3v@?ei=! z95Ey$ZWha>iJJQ;^V>$%99c(B?M~#} zbi})2QrOmOPdA0z(3T#c+!7reZkU6*EdO%CV4PUEh>-z>JVxZ$d$z;dO`6=&NsH#mR>l^}7Mp4`0A1>v>Uwy;kpTJ%k^w&bUi`J(}fPqF6Zt6(y=Lr0|{K zwpv`wEyj&%MHqVE7xceq0R6FO*t12Hu3E`Y%g7fjnzsQX8G@e-G%-xBnaucDPkL%v zN&R&(oUoO|V_jdMRLF{J)kC(M+8eb+OO?)thC zwa@C~*{jM}>!nPebc#~X`d?_ey9)2#EW}$@ML5spKRjrD0fmclaHRAq$vIa_l-C?1 z<+YL2dB*wKszXWP441#&;cA6izfj-#=G;a7dW7RT^Q&g)=+(viekbw~F zzwrq;4C~(N$Tn?v!{j)3Wr@~k^Af}46r%f zGwEHNmhXe0@O=loH`z_z=n9fIgECCUw!e)3)&Y1D9Z!ar>|#zuY%)bX=003O&m7eR{;q%#0?KUyxYpv+GbsI+JneqpAMv zT4ri~7e{`EEaS9RpIC>@gx%#bjHTUrRF3u}w~i{4&uVHQlBSCDbNa|GmpaZ>BXLrX z0-WK_5~9AP8G1y&FjJd85&rGYB=+}dCbT@1v*}tH;}mJf+3;W#?pnCjv^%gnw%h`M-)-@fB^ClZ+oB3Df^e$cyd9j%k zv}VEOrKjFxm%b4#(hjD$4MEP|uGxjT+3G3R=q;M)A_pvpIv)|1i@WMn| zAi1|-ITlA$;)wz++#B@(+y}Jb-6kLMUzjRd%njn`mu)iDGKgm`J_===<#IVqyLJ%2 zBgf#b<45N6{ZuG={g*V3tR}*m&g5TtDan{t!B|J;Fqfp|iPuCaBd$}zX=gsN-KiKN zD%k?cTa)m#g*MpFoieR@wVHVJ{xUTWlZ5CSJZJ`SaGT4AMbTTxzPvi-J%2Zey*&V_ z*N4fmANwG%@B+zT*|(I%qj1dR67xLMiW6=h1|MhXpm=%?EGU#g+aOD@e#a08w;6cE zTpHqkRFI>-MXUpj2P=OIlgYFHIH7M3Z$HXiKIX;Ej;^3z5{v2L{PXmD zZx{Uf^Bb1CJp%qoFH+Y2g!xs+!+MRq=_lVLIy|$4s!R@}q{t84-TDtF?DRqM79ZUC zEDGl}N*UAO4a~)zU7UiSsaPo;ZwHp;?^ z>}>G0nThFOi(<+zptb!VjHIiT6+UBJBBvD>4P? zg2)g&5F3Ia#vV9r!F1^PCATT=WFeLPEK9d+UPL!ov;s*Vh3LKAAn-kooSC6YOucS0DOvmAa8d`+8EaueTPR7M zcaixn=g2sY-$k2=Y<$SN5u3cu;NfO><*P=1Oq7!Y$zb7((WK)^oykeEPnwf|TqK z2-=W_N>l1M#?u1Lm%E6I>l$*4-+`Q&JHQ-ovS54%A~;LGeWNoImASmX=W_EEDD!Xe zd(-2BZ%j?r97XLbuDG~84?{N|Ba5t(nW4wK$-cYy>AEE^XuWS4-rdmepiTu-YC zo2XJXo6$Y|0_j2Xuyh#dTiKQLfwdboUz&_*e#g)w(+>aT&A{lKd`KCZj?NL=U~MJ) z`*r@w9Q_c4b}QFl*AEF4Yw@P9<$|cCe=z;nQ4F?wL&3!QC}c;uLbYQNQR&YlH;+Cg zOC_hsy*Uk}{oi9WIM9Hv!_qPHQZq*Sj}oj)Hf);b~em^!+ubuQzU`+cxc_@+6c%_D`8~oWn$|U$~@%RM2lzcpvAQ(XpyrDMUbUx%Rk^)v+MxKWrcJwK5*tsf-y>+BoomWvx!GfEyEH z;MZsd5giNRk^5S>a=R8bRJ9ZL>PoVsu#hP8c)``+L?r(6-TN)S58i3=yeQ?kB12_mg2XFq@5E1?yo{R)Y*avtk8&u} z$&WG$yW#DL6ENHC5GX$E!5g;`#e!Y~(|Q(+&YXwBoC?rXDr2^ZPB4<2^T`Rm&q(4S9pHb^kHahL(0EY=8b3B? zM$0sr3pRP2z2nIkIEC0YqJn4dP2<{4N^obbkmug8*oYEs2hduEl@qYu-+zoeRiCRs z-`+hApOph(LEJvzZBif|fssu4M+1`aF&#wI?ZM}Z06OH&rX{wE=n`vd+L03pr|Z^$ zrnd$xT;7CcM+(rlrUG#;VQNjb9l0evttNR4A?8$j{8Gn~*m1}Xe);Jc#_E|uN{YxN5F(4h~utdG~qK$aHD@X`hY zAv$!l3zmjuLf_d`c>c&6S~|AE#hs0C*>Ni@x}XRAk5yoH4N&*Z5;VPBn2!EuNotDp ziRn-Vx!9Kv7cEbKi|Pp|vCPBaXQ?PRnTHSjPvE7e>+v;@6>d>^KJLiq7>FI+qp2{Sw!K+SOy+P51FaBT?F;3R~yR zB+X8LnQX~}3_S2752~h!1zCvOHk^h|(g?Gdo|djb316fxL*Ig0IKIl^09^zfg6^F8b}T9uB; z>sbfeTn>mU?j+hbMnUPwYP@;g8OQTCqWV=`RPJ1g>-cuz$sZR`WQ#hrn6{0+XT1t5 zkBV}>&FtqSJ7mGDTMDQ8bj&`5 zI{dem?V@zS4wWuQ70MwRk1N@CC-K3~N^6+>nFvno|G%b6mj2v~w7=Sv3WTWA(^sTu zgvBp>AIJ6p)}}yBeGJ=8&W087&!Ly?fRF8a1?PNz0$%c;JX}kwQJ&d zelasu>oQ^-${w=4iO;mCl9yX`-2$7G4Dfuh68^0jg^x~F7|cYXN?Rf-XB+@R_gqF`A@A>8i12E%<77-uzxZBdF;d|3j#3Y!k5w&7s(!H&-Ug7cLCdsV(e*J2q^UZ(@D^TR=qb9d?f!IJym&5tI{O2RJ>+r8t2p#*uEG+(M;QA@l9T&0)5L!@4-@MagyNSn@Rj9x z3~=AURoWKIJ$*Edd$G8m!>i(IVmoSJ@{Jo$-(AU}M;eN$R>%+dGItnexryPrg?+fp zY7h^5Uqr&$81`BY@ zU=jEm)PVh@5&JCtjBkwvsLEm`nz`jac;X_6|KjX$(fSi4NyCyEt%_rY%0%f6XB+C) z7EilXE%0H^UVMBf4+9>0Vr+9VCe#k1dCgU@om~ps3d^B3U5@)#Rg_!eB+Px7kPI)K zF2I$N9{60#e%Fgb!IRpMT?%*M@sUzUICvh?S%<9l`gdg7hEZaESq9=eP2i<=C9$xK zhp*w;;CLYxcs^P~&&&bxs7-(fO(jq}`EbhBI7|~CUZZ!GT%ZD2NEZsvp)p?+X~{HM zYB`<@v+A#ckzf(zr36CbLT|{7xJovRO40h7cNnfyg-zD?pnXLxtWW8KDNhM7^^*XR z!B0f2{0ugXZAWSO1z7zc7lf*wz>v>5*j@b_Rp0eteAZROQ9E!KlLbu?4c0-JiKi#n zo`R1T^7M%#vy>0*rFrnFU<`9?;dZ9JPoC-hPZ%be?vtznU6=?|r2)p%>6?OoI7`I} zj1TL>)$Bh+%Dx@krnSNIf>OBOJ)2IR6Qx;;zN1t`JM?W$gwNxNP}-G_-Mf=;$GijR zX4prSw>_pEV^8T|8!zt9Zi15ghv4T4Rq8|3Xh7X8Dk2dH&4-eqXU}YSk=YEIb@yT4 znRciev_&=jlek5!6r0TTX~gCER6ALp_Fr2Jx^tA_f!<0`3fhcH_H$9kPzD9oO_F;? zU8Heg2N_%vi!~S6Gll3jls6N=yhnea^n@VpAC#fiQDU@iCogq)-GtsTY}OHf21V>5 zaX2&#?>fg~T7DDGO}&M?@882N&4nc3=T*jYn;moR%|pT?+DNVr_7bZl3h+Wh1RM<( zvfU0F)T+`$ahC1xom_$LgL=63FzYh0m4_#j?c~DWBqAleh}x+LP^|@z(ABb>blAj^ zy*pNt9V8ytX++?YxCESkDU^Ini6bp6*Sjh51SQ-2=wSIeI{sZ7L@n1y_lIh0D zDm-ovgH@C z4eBF{@C*i|$6sAM7@zh2(1K5m)+qdG4lKsR` zSBqH1sF8VMe?heS6C`SV1?%;E+yeD4RJVPYhAY^SnU8jnKmBWoarh(jId&ai24$m7 zT@CoAU4RWNqpcd@j$MN5@B;hp6GgpyEbCo~#x0@jv(1vky_;gzT}WmO!l%;{ay)ce z=osc~eh;bZyWp^W4)ApS#MfdE@lIS1#xz*s{sY$Nf6x*C>o|$>rw-!Lxv}`hB@(vV zEP&5v>&W?DJxUstY2;2hdPGPN*KPj|7(4(A7HiYn33AjwOokd>6M8-=(*JtbbK>_c}5oaWz@{atnlPpT%aU>yYQ)X3+fY2A^+*!NWTu z_E>q54Tmq0l|DWouyZDeYRAHj>Lv)}J|VM{kAbl3dpNU0iV77XO_?X<1N@BRA^)aUWJpL@@Fy`IldJT$}^KOfP;j{dH& zv#1oxUj*V*$p`X&dYT-o(y9CFDU9j48>?p&po)_M+RD43*)Cr!-xUW7W`)6^_m=qW zE}%5=Jr26sjiZjgB=-fU$w2=Asj@6sChjjHJLcfw-E-+i##nk3>S%&a#<2^;4lbXdvVrYJtRCjyTVy z24z;r(}9aoX0O#Vs)TtsGfT9!CzQHuNQ; z6NPOF=(tuJ&pe66TPe=CLS)$t9daDYuAIXEjNfDERDctgOrYPTV)*mn1ikwcLl3Ot z>FJFL^jXQ2qVG~K~c6H zR1l=jjg4g>Wq&_NytEEn9s0pn|7e(Z2+}?};N$vC z^lPZ4`G%F`y5s|G`s#t2&R#eNk72};*WfaFh{&up6T!2`u(v#7MDSDaH5|=lrGq)Y zGLW;2sd(VgBfL1P6Z37icjjEw^@v!S6+!Jx!z)xbXK@Ear53^56;jH`kK9M&nN3v%| z(NpC@emUzR@3`}UH_g(3WeIX%eRK}XyKhH3uWqJsAB(ANl|6$-4 zO{#Ga+3E35Y2FNx^Bh=1%~7GkyfcAZg7{VrI^szF})YRvfO4A~M_VPKjCuZe#^32J?yy6Q5Fh<-1G6eU_*oqQhZ zEZ(xmH&TV4hr?)h4_)f9VL5v<@E1ss3xn~gGi@g5ZF{_i+Re2ko+Lg zBIsuk|FkHcchp@Mt{ZH_)J?O%C^CeF8mY2J9k0!!P(hOX3t`&D0m9qg3Bvl0Go-&s zfs}+JLX-JUs@7glYwD#;Tel5LrgsQRJ$kd~6jiRb#gpVEam-(w8|aVEfrFPr&6Ya3 z@T4ya{Ku(t%xAD1MwEWTu;NgfQ#c0C>`Gy2Z5hyLohIIOI$-H7jRQ2D*9W+J-EcZF^eOJoHKe}w6G`J|ny^Y|EREVz zgncUJLD;!T_+rj3q1ykN#k$@Z)MTYXLk>NWDBR1(KEiQGe3fc(YV0&>+@XgHyquus zT{)&UWJBj6T~_$u8tyo04aRPItYW$hf$?VHa?3*o1Aed=wE^Jg6Jb`TnaV7xr?MqV zrNU|TGAwFXj_*F}2!`T}LOS@l#QSBMa5(2CZ8ca-*SQZHY55%r_v?UUXO&r8UK?$A z*`Lm>xG&7v9tz9rSFz&TL6V5LH>kfgTrg`jV)fnP+aTmA`L6L3isx(cWl)dz{fD!J z>~A>!gfece%r(C=SRV6d?`93l4zU^d9LpBU0ZW>&gIssO$QN=YCm+8Het@!&oLinXK5fj!~G6TDrY;K1R zp7~FaNkb#xg32Std8(65 zrV!0SbT>9ZZhhOTN>)cp^>g-iy=Va9?%zkED@;VPT?TI{)x8OH~Fe@G&! zT6q0%FVyrZ24(q1HfWm#$sC%BE=3ceR3hMk1Ap1B?`F)+cPnjR~vw$+D{fU ze=DD6d{`90np z8wKimq`7Q2SQctPWUpqnU1ZYZkR)<+>_yhrm$96V zVWjPO3~)_gH_`)H=#n~y^LpXQmvNACB?Wdp3?)MMacx|XOUb{Gn{!>lSEKCY3U zF;F78L>UTM_8x1m&*EuoZMpv@#?95Vz+lsNR%lqne9BtDds8B4w)BC=_3ha2U<_`5 zWQ>Y>19_iG;*ZKYfyeax0kVzexNU$ehKas{+$KG`sijEYZjHgHKO!?v?3a(wKZ7k8 zf;;s;f%(!H)Ok4^$IHHfubY0+q#fDRw0s@uKfgvLSH#`ec~44`X@b|D3n6;Idv?`J zoK>|4;mbFJaD?MFHfX_T!KCH1U_Wmx1hH#u?Aqh(@g7|q7jpqj9;}0Lyb3B8-iNiG z_u!7W13Yu|JsKW}!T6&RxH(rW_b%AaW`{cSph4PvaK%g7nqiOEwCCW{$3t;ooD=Q* zE~RdHRa%_YU+jr`;1g(v-Kke;Q~W*}pgDz}=3j#r8GUG|Ph_!IKV#>}Obn1)jj9ih zLH4dC@O|4amfn8_-4f?E1CMs1*+Y8>pC}Jn{c@Pe))_o|t|C8prI{vVCyE_`ZB%w~ zBh5UxoQBu8;hEMsxUl#VEZ2<_-D$(Y^{@**`r(96r~HEIZLVziSP2ubAN%>D7#=J& zg3_mQuxem3MTQ5`cJ&n`F$kcy5KMBzyy)=1;V@kL1e;=&$_(#$flsx_SH4!j%zL|$ zZTT1)?9R!xeGeWVyc$3F&cSH|qxq}Y5PoaNHlDfe0{Rw|VVA-|RGZ!bBeTU#e(^7e z+^-H1RcG0Av!m>z!(O@q4Jf&ggfISVf-ajFxcw>`UL-exgV!W5zoH3ilKNnzvOFF( z+76P%CHTN)4?b_-jFn|E6mim&{_Hu64?_mS!C$KII9?W}Jt?KSg{jo@RvL{J#Qv#t zFIbJ(3>pbae5B!JY6{3DU6K8JP~ROj3LLRq%(d2B-%4v=&LRc10p#y&N9UIArs~^E zssD6!N;N-@>z5tC$Yai^9H@G zOw;MaR{N)zoY5)H&7Yy|oH6wL@mP8YPIRSzH&@i{g5G^Ep;i53GotVTAC#87->e@cT9P$&!feE|8(VNbZGl?JRkH;oVBQu zqUIj#|0)H2_LZP43F7Z@VWFgkQg-NlxdGKRk-J z)`W}BR2BZ@#9NAL&!qd&W2vD*i|oVyqvw_VU`13l)83N+Q*~Ej^SeO2581d;Pn=bp z`9|NTYVijNC$QZ@?8L}Ou&=0x-XyC>m$phsU(83>6|Loy>sIp3XA>|eIRKk7Q^k(s zPrA76HdVOzZetzk#?7aA>CR3xR4&Kw(hdqc@PqnHQsHyWM$(^%Z@6H` z5b7wqgrRvCaCq=@9Q91>JQsYY_HkDH_px$x_+5nh-wq;dy@ubi&SA`*YSc4JHNuv0Uyj+^dV$}oM z?^sUTw|7yaej^2$KOrafo$~)J<0^5BdGOkJ?qDqY1KT5Z;YU$d0wxfKPp2II-4YX`mku8BpYjm=Fg+9sKFfUN#2)`!KH2szKeOaHU+a1sE$a`W zqxfEzSKmopf45L-&I#(b2Ps=iWL^1-Or<;zURvtL`{+;SOUgpCX}*o z$I{^46UE((5#AlAhPflfw_*RI)a;x}9oG#h*jU{CeqTyoGzajAPE}45#VmoCrEUs{ zAf>P-GD_MD?}mMXEuWIWYRyubuC{>EFYTbm^^2rCYu%;464yus+WsN-#dl06&$!_+X!B_VIaTeHUeHFBD+A>`IG#pn3E#?L|hquk%!l&#z zZ|-Nm#%#a|MT=yoyZl;830F1B<2Q%uqoJ+@m!x{)*B%8ztWh2tSrZTGlMb^fcvi4! zF%YIMZSA0d<6(Z$_k zwD8#sL+mkrIHl(()6-u4sit!q#`XCra=uLIFto6bGnFCxtqlC?8Nj-qPGGvu_N+>M zIRErRj_b{Ap_xb4OX4OQSnTzEXtDcnu+Za86#Ir(na1g4Fe|GCuZ%gOZ{iwLsmf$W zt{i3!X&v;t;sW{4OQZ#wBfuF~vS{Pp!VwLB8X@i=2m4lG&}LaNmtKll!Lc|ZYBc+| zell}#(O?tvF2I}1Goi$`7x?tgp-{LlF1_9x1&fImTDVg zVPyw={JR~tJlGCF&M(0Jztc=UZUl?je~?D@Pa}o2UDPwIgRSP^-ER?#}Lk(_f8*0Sh`TJ}bEke@8?L z>te6qo{^Vu$Fe=RCh?znQN4mi@ZV8p`Ef-w^lcnXr)b(Vve9DA0I6hk^cfK(d5AWx zjG?*Zd#R_MEH2x73kLZDJp1~GuB-ed^+)ZbJn^oD%iX^gXXXvHu&KNyGF@KcLy`Y4 z%r9jX-G$6(<#Bc)R-0Vg`jGE4WqP^s8GAf^Ci~Fxm?=5+uaAC79E8QeZW4tG#+wMzXpy!4%cdYSf;W`|z)h_9~BwsRV4Hj8Pxfko0bAl(0djp$z~?GC&V9OQ;OF-=pzb{?l!{)+TklZyLoqXo z4rNPwY!?hNGuX&Ki-m^ohNKDJPVju2P4_a&XpP?(AqD@Q>5 zynW0lBa6Kga@oMW$zbnyin>=9Q2eW_G|d947K@7WJk(Jiy9iqc@{Y8%1F(rENV=XJ6u7SV0bLn*BZJ}-4 zI>;3n?CziJm_ex$9&Gic8#jhvm$fOD-x!7OuA6}2kISHNeKYp*yZrxi2(Q9YpbYroHC_1)J`k^uhPN_6D$sQ|S*y`k##Bdj?%m%r>jM3!~isbJJf`jO_u{D(%u z&@flLG-ovyu8+d`BEwR}K^FfTF$mYZvc#q@`fSJcE$nM-KCFvbB>cSJDtwT?53`oU zz@lFdAV((vd+pGLX9239QxJy7Pd^mih6J(t4`!G*EKqa>Eh1VP%Nt4(g%oR9)*Sbj zb==JtqLo*&C$DWF^0*XoOvB)mjt0G6kP7Ad7Guwq!4&Tya@-uX@sD9NHl2JcJWRdB z!usrjr8co(5W56=o1cZX+wQ@*IU2a>feE@?XkiUg=VGhB7im1|#U^~W5!T}aR-O?L z`D^!J#_J$T-LZ(B=qP5dE^LOI3sj+1%&>$S`%|UGb79ofTJ~d=C(eI+7f!n@#LxFc zbLzNR!q=1bETZHLl<6kGOCb&d+Rlnug+#O;5R7(RXYpTY1xr{{3SLL*@b>Cprgq~R zJJ3BGYvcytr=?>_@y9U!{zD)Ox^KPBO-%SHG=i?YT=>-PWInbX)e(>df5NsGL z;IfZW8rj*G!hSx)u+wD_vpNcIcN`J()Tf0JO*Inu86r5{pU9?$r-rdfclq={?NQGtOg1Jt^=+-zX zI~adT7@pY|9CcJ!M!O0-dbJPkI;Tpp-oq(qN(d!3c|vMa7f$m2NTxmRLA1$MSgGR* z_n&5fY>*o&n=hiDi@GF3_i2NZ#{!(tYdmdmv!vh&0Tg1Z4{a^#cm$<>QDg`Z+} z{OcYC+@UC*?@rZVo4y7=m!H8k6E89k!vtaQjM+l=Z!;!$8sOf3C%CQ@0{*+SS^U{) z>}u>lrs&Wmj3^0Z9h3D*`rHV#mo~GY?_Si)-Vt6*9m)*df5DXRI_zNzhvs26xFD`C z#u`jR=K+1;wc0oKrOgQTy;S0x?|enIJw>2h=?V8&4S{7pOCY&c8763Ju)ZzYY**qQ zyz|!q8|Ljqns|%Z?-|L~%s$E9bq7$Ai9Wu55(C=;|AXyYf>~4QR@O3MJNN9GM?b@R zqjA7!{8*)r9)?rUs;6U0l`=MK~GM2=Lk^dh} zir71d2KXp}j;Ahr+B5+2eoUaZGp6JA;RT>{vITlBc_=Jg87z6`|CY+WW#fD4C^S)@ zMG6M2+4L{*bh5fdXn|TfXNY}){{{lce`82vtUx~MxPa_XVXAal; zfnfo_m*4tS>DvSekLqFii!2zUpvVrbk77f+H=v1p0!;T9g2(4fgwqbifUk$3^21N? zzIh(|@p>}5aVZhX^yRQqYZbb7X)|*xGxkG0jFqRBv)g>5Fe223(UUruKXW`rRz%|d z;#in5OY|adcqitTf8ohRwrHy826FDT?7!X;=rbb{B;FgSRA_|fp9e!#PBt6wn!)BT zNCTt2iFn<35NKVFgh`PDDIq%yTk;-cNV>t~dpgT)eb4^QI0qBP zl)`K)Q(Uvt88Y{cV#CHKvXo9&vi%Z-zQP}P&L6^J?E+!g@$pRKUkbi0@rAEFdqab7 z6*!9SDvj#%==3}p+s;ei?`0kQF+3N{CjVwKV;FlqJ^}=lmFVgjilwX0vd(a4c4=)8 zv$Z-1?N{oA(Z8lMg$YCPx2-(B`fiW=Q1&B{F^ehC^Y{OJT&Kf$Ym9383ZSzy%y`n%HzgokA-OnH(^akjX73_SQ1l0@8anR!iR(q_1 z&3ZQz`1=Sp&&!N0@>>ck>j$E>+W{<{uFQu>%kW*0FSIy6W8SyNgC|aglM9AXRJuS} z`YzP9qa( zDE*TjrYrl?alcw$TaUXRI-vN25HPCg-2!~B;1=V_UZ0sZM z7=Ipyg564}Fx4LCo?8lH;}&oICuZJOX5!$oC#azFUG#~E)9)u$)DSkDI~uOWt*b}V zkFC9UKlc}ym}Nmn6$>eEM+(GO?g6K1AHX9&9!E_)gcE0{p^4iM7IkD9iyp5H`SBOo z2FooV7%W8(r(s~3|Njg~4E)ed!)oI_SmR}j(rc06_VyZ9WG)u}58dEUc%aDatAlw< zhNE>O~9=a6N+`7u*!gP6x7so6d{Q`xAoUl>gvg zNgsUj!GHIz3HEEy^YxhYg?A(J)NB-)@~IdUYnyzKvGY=ZP7uNs7Xw?wj#K)ga8XE{59P#So+$h{fAo zvCSq9FC0q3Shs=jx#tBa)DFOEgS(I_&I#Vh>T`BD5YxA>!)-IgtdOGEeLhnP78eF% zvDkY=pXb=Ju^U@@h;Bgt^T5Vm2ASp{@LwJWB`=-P`S)HS+3h$R5^sY>F%Q7J${r2q zEuP=01j9aOv+tuszk=s$Hl$V!8wU2EKo1ov+Ez*XZ(4D9>W}~Mu179-4~@n_KK5j=P6OB5j6tbF23gYJ5};h{JypZM=4%B<>wkp9o$`HP`= zMgB6LUi1$qsrI0teueB}%~F=MIGlz3y^CiO}v;7cdNzyz^{X- zS^rQlwfrfZYnac<_HWId>@n@pIavdkWq;633t3FyS|!){ybNI#P7oK`XL%(CM65jBgA^#iYYnJ$wX+ zA0&jmHUyioFZsymbZ*uskdGf=OJ{;)h`Wdk2Eh(X7bW8?h-*{54JJ?l4?PB z=4oNln?d}?&2oyA*-euce#U-@@%U(kCB~Lji`hLNnqbz49)42Bz%e2lGbx(>mW*JD zPKV*R=0=>FYs@(Eim|Ghl_iG}OPL&38c@eZqo3&@!LK0b##?9_0@nv{4^kC zULRFwHmy8emSYV78!mey7b#>Dn0)Y z%lp1v$1l$C={Aib`&A-WQj&%DUR{L68w=sVW`v4E=kclRa4HVHgAGgOQbdx- zXbu<8H2GL;`syI^JbTe}VKHCaxSC&958&R)I#9m(3XIdW!{j_0yd$QUBr?5l+Pn~G z^{rq72A8mpYrFXO5oh=&n_c`+!F#rUkO%B3oDTeB9M}b%fUMt_*xluWXn}kmx+C8t zz5(SiLpmR0`g)U$!BpN;D$89%4^SV8EOeX*WdBJ_pk($KZrSr4)mvVr3z4N@Cbx$T zc{vXv#nIFy)q%XZMVYq_v80ot$K$H4Dt-9~UQ5G?Fi|Fys+yK9Fzm1-#;N4Zr7n!Xvv5@NFX_`2D8| ze9hhzu!)QYnA8C?tSs?@%m*|a7%6h`98l{0c?-SEO<9Y1FxH` z#GyCe;*3*~Q2aL+K8t-hxAh+-yG+Nh{D@QVcS#`mI7HI&O&^F4^yNi5%ek3&4vF7x zyyN*7gPv}s`&T@0vg>Df+IIxHmnyIVxvPTC#_4Q*(k!m}o^dBFBi>(KLwqOd&>dw* z`o5%#SuFw>Yj_t{ygtB=?$n0s8ik;IwG`Ki{ZAwJgH)$(!RKt%<12fp@aFQT!sG9q zOk-6)IFhl7x~>hU`q}dII5-CnTCNj)^=mPCWjPb*DAOJ92|ZqJ!|tyMILfRJ7ru_i zqt%zO+4B}IuXzpaJO~CIzsN>b`@n&@M!01`DjwIpNfn`IsMqge8gnlW+kKDW>clv_ z959utt+Yw1IDp!fCQ!lAu~ay^7Q1?@L)#m9@RiZbd+FggY0mb`?iM@=zvgALwJy9JR+YM^HTh2l{S;BQz7U5-(3C#VF64!DZPhRTz$VXX#ogOFsqBYu0IJCGLd#@ZpV~pRE^0Hxkd_xp>U8FA*`07XkgFi|1Mz4a8;@{{W z@*I`~eTSW1tMJ&*v)EPL0Ev@_VIY3SyiPYPjde!7zk@I%V5X3K$yoBMI9+JIagJ+R zp5$k9k8%>Z0Nbs0vB#Gd3*pli(dZ}-+Pv%*ZK#-t#?E4b^WIuGQ#YB1ZF1mizAWLs zccuxkNq>cm3w7ovrUX;mlUUN7(?qg{ZE*W|Fx2@B!7EFaU~9TFTKpJ}rJs{wb3zpq zl#j%zt-E>YNlzZS(1)*jkd4c3Z^Q=QBz#oRLj62$)4s#c=)=_*9Ju!s#;*8?8xnHR zs!@Zy?uffoH!W5ke@K$+cTAv&(YXF-Fus;3(=z)nbou^E(Ho@6C$Fx;6Mth+@52CW zG!;FTxf5Z)kb!tcbg}V087#(s9%LpCrm7Wkl%Sq#~H$cp7F-T8UY2is+hmg6NEjB?-(G z`5Q695CtQ)+3W?Cc-*Ja;vVpv{VPy=<&9yl1zL0PEOoexUDj2n$n%|)oW+d$5z*h| z_qLPDL%-75$=X~zby-PjGP4h`V6DZfv@X8~jc}Yq_rBJ^Dc4Ns-+l}>WNf222P>LY zW=G-wqFJ$(BExhK_OQ4W)i#NDtI&0DsMkEm85Jzp3I&qof*_peo5}jHi!j!%1jOGj zIGAgUZ~OFt9<`!z)m-p;m*XWSwNH3u3{|z=fSx3Sl?){$uA>>|@q_lbM@q zHtos$i;v93{O2TPemHO;Ii|N_a#1MC6}X9v6$dyq$`RJ3<-n4LPcSEJ2^<*ug4M_7 zvO$NpQo({pI8i|sm15_>gnkp@Ld+#-+%pgUu5o9h6{5uN*=C&kb|4v3p?AYs8x&8mtEGl=E++aGp-sBHQSIK31jw;FaJs4+A1qvh4u493yvv|s-FZkb zT!~6Gitt1$m+6enWRmJP?DrN~_Ri#yaBONO1new;$d!Z9#AgCJJgX9Ur~lx-?@iog ztc|k0^Vy#9Q(#e%7N})Sfzbb^((ZLf(PxDUezH(vO=_AYY^TUUtXOgqX6UYg`#GYYaK{TV2Nn$Z8Cv4*#Fw;u z4`Ao`YEYkMg~m7Yq1)UNK40__cC4;v56j*`N8MCVI2FjMDg)VUU6EPc^qd`5>kk70 zE(z69E8(Na-CrSkTAanq^5jttI415FI9knw*Pk(eK9zuxpBjL-JqQ`8oYr z?!%tA>fj?VZR?Ez{^Fhd)j$eCRndi#R|^5u=&FaAv=BmY8_6%gyEyXQROGI zcbmXEgOy2mnubSb&&4}m_QK9bzHt1KE=Eg(*!w4=Sn$HVps`CTOj7s?-o?jJbHEWA z(!U42^AE*78H#i_GzZr?M`IuFK#Z8423uyUu#MHDDJlxdqeRT&$Nj>*pXs#)qC0+VI3#JSvedZd1lT&2VK6>< zbQ{JE3jro{=jF~2kocim>_U(-#kxtFLkpqz~V=HLgx582OcC~SO1 zn+_z>quZhb!u%KN%9f!{UM`N!sS&i#MGDzv$JqC2%a3iFS+Ra9aIua2|6S z0#uDLZKMS4gzND5z*`)0>pLDQ`iK|f=kOz)oVQOm;g*5@X}FaPJ%wL*_V6yU?XsZe z1%{L-c3|b4Q_=o}232?V#z8Aa;A^+Bxcy!rbsY$&`Ng7-;?@=X?p1}QTMAG^WMkDj zwd2eKsx;8vgAP5PMMYKL@LugrXfgN!R(raj^1x=i{>K*2y^+Twqxw)F$^VfN?^~qJn-4xC|Bi0Bn{gQi+{uA88%6hNYbKQLz6);4&O)BJ zmrQD33w(ST6_#J7Nz>Csrdt%9yE>JU7tN>Z>K{R2#AVR?&lKAipXDB}3i*a-x%`@+ z1)OYL1{qfmh&kL{RA?AVAN(R{X7dzW*rkW&X5#*LYX;XWI>f6MC-R?tk}&326xPqQ z#}}2>bbq@AS&bY^TW9G~+1WZQT>DH60_pLVSIS&^LWYMnj6@yxf%vQW0^I06N5d{9 z(3~MzG}u(+rfLquk)tE<)SY>JPWB9*l>l7ts}gN~(wmfye?TDX)BY&@E!LpzBA2T^=Q~=wi6W0k7ixbc{@nv_VCM7_IPLKnF(C79bibf!gLY3zt@LDy%ZaE{9el1w?i z_j5h1eOg6378GG+z3ApUyB}NH9Pp{qX51k=2UTbF;+CgB(JZBA+G22k#da%y^yGxC zW$>}A5R8J);@&J(dT#4T-O0Y}@4h2Ihw5QmSP4%3F@~J?{((n5zJq(74^Ft8gObM` z%)EUvn0IGGrs!W$ds>Uzx;VZR_gGP0GtpP}4n!Z>CiagQ){S2V7AvZ-sbM#bQi-9v zV-2Y8S`eko+eHc;bJ6gTHpTw>k0dciS!Vx0uuhwRt3^UXR_z@~)_x7%Z!Ti8l`7Q> zZD^}B5(kAS;--Hup=hEOygzvyMg$dsQ@>!$>f44^Z=&g3=TN+Uwgv~rtV2(UKc0`u zM_nF)P4ZWnX3rFmvAGC+dxfLx2BgMi3+ZBF2Rl*kjo&ka(J1*MtEeP6Qoe*4WCp|I zK93>hf-9^%+5pBsFCgCNL_e2D?BvE>n7X7ljXm`MzK{J6zl6BK@9Ms!c{PZlKSj{I zh;|6HNrsuH+}Q;Ap%{4R3D&4O)52*BXv~)@;@jdfwfKz37t^cpROc&%j(d3ic`(UteS^^BOYMut<%hK?MtQ_)WOd9 z4HP*yJt_FaEnN2E9m;uF;nB`cgs8nR)o(l2-3i2`j&QpBs(}8vd`0gkJ%z@#!rf8{|SBz9)i~Ud?0;>EQTD{6tmFpMLq}6$&GtS zZJIlaUArFcFSUfc^=bH1%nvGP=Hj;xdAQ=3*poT$F3$VE;_YO2+LXMATI3SpMDh}R zS2Y=314VAj8)q`n9*C{+S0y*EjAk)K--M!&YhYv^h+g~5QQ2iB+GZ!Aif<$uWvZiB zq&iJ3)1$|Jn;^7i2#yiwt*Ij$m`Ce**0hO1g4I&2{i#5or%oWtDN*QfBNxl}RY18+ z8*bcPjT(g;NGtTb*q^tck$bn}{E0S_Nckek!ZUe-;|lRRESrj_J3C>+)AQ6fSceNa zN<7ETmCbn37vAm4V`)P_LP_RzT)pZxW;;q*!|gCgsr$lK$jru=SUH*^a|Rzhh{RN9 zq{?Haw8B{d)Xx59HFxdUNLP7Md%lWTWFURgUyV5r2UFaodHBn&5O0h~!V~Yuqp zk>#r}b$2+<)Ja0^5nlL9Yd3aKxX6wyK$6=G;eTdBRR6uW{O}A?&QzpNA#HejK^n?t zDUi3h$dc}7AzfvlBh6ac$$c!8 zA?b8tLq9rqQxgxZOh%U#4cz}#3D4Smf`9*_NTXwVQ_nR+>A;>{{DGYxFZZ+NZ|j#) zwc>n=yb(;}kKbY@`k~@IGMNpx8A#Dwi5{JQf;Z3il3uF4#g|2<@U1f%&`EL_j}*p< zeXMD8Ox=R+WvJ7Mvv>Jsk4HQpt(04Li09yOFAA_0Xh(7&?|I*cmnV(mwmtS!=#eKh z`lu#9Cer{a2Cleh_7gN-SIoorr*gZ|d%1szDt#F(o)b=rv_oYEef1th$&>V`wpNkb z{Z=5Sl6*?rKZy1*PFIH7(cy!NbVBJbs(9YPHMjgkkFUu8Vh5?u<~DeFUj_RqEx{#L zu6#)B5We{EJsMfjLXzGGsrGaTogH(R8>yV;W^#wQO_mGKiyOfurvx7AKc25!@RPR1 zRa3TU8uwE(<#f`K_gK;iE;q#)`oS^ycc35MZ3;klGZQ?2?-uyww!+W;t5Hw-lE!6~ zlGgQfa<+R;E9!gjAB8_@6?ee-`XOkSz6B4Twv)Eb9xFAp&v-yH3v30x|DP;nNY-1&%8nGC7u$#6II99rL= zK!J9fD9vXx)rHR{FG&s$DahpIamV;e_W?Aet{)9{dW^9P`bbBSob;G&D=)jAg+UG} zcwuP|inG7NZH0^chi@!@G3_Gl{1Qi#3fIz(fJdlzdJe=|e?zOQzhO{LJbZcD2t9jd z63DNnf+-8=n1&}eZ=1p&SPbP`t*U8x`!!PinnOQldePilTd7~VJ*k=oqH%)Xg>G&VhS><#8=x!(Wh@iaGIqn z_O8=JFZbP4xZ8#n(`3p^&yn8xmMB#iyGuHMz+-TmJ`Z{*C&7lFZv4E%M7|}L^J1U< zbTd0h{0w%GiO3b)_{*M*UbxZrxCi|H<61sGJdRI@$Y-T_o0yYTjkt$uL6F%q>P6vet=)O zv-G01hg2hdwzOugG8Ox(ioNAVyj#_duLngkmsKXrD<_m)S~ZB}=jSqu8AGtW@H8Y3 zQpR}>!?7#QiESL&11Ps1CD92uCVdLZ1_z=?q%PJ3sA8UHB41w}!gt#_@`rQ(@<%s2 zdH#QuynDue9#@je(;q~04?8{Rtciet-=bTrTLaH_Uxc%2-LUSd8b4$^g4YCd9uV{v z7OWMsDBV4AYpxom4z7Y^h2!Al+XFrOIb(F%C2YI3mSTsLqt`=K%-XS&mOfZPiS2u7 zD_ww!r9VLPYaaZ&_y=!xR?yR1CrPdMcIu@%f%sqrI&1cly?NM!B`2;E?8}Z}jd2F< z`>+G+SPmV|>ZD`)jClQ)oorBCHEevYh_^?a$D}jX;?D65yr?tg@!sS4wn}6EuRRf; z3>uAk52xe#lyvf2+(dHIZqgQ;Of)+svcd-6L6gx>;QQWoxHl^dt>hYLsE6nSea3Zr}7fE6YEIr_)Y=J|i|lgL5+>Sc;qAH4D4q^FR2eg>|- zh^XYJh10qs@S#wJ%8@C|WSb#quhoGM_gax>O``UfzH};59jj^+p>3uDnp+&DThc&^ zoEb#p=U=1;ZrABX=UZ}5J4NprW9WQ?7j3JY$MsV_QQ7ZOx^_wr41)SV>X5Z;#WEE< z_4_YeY!$O|@0_vwrar{HZ(<8;Tkv((DmAnJvbIpcZ-ukHiXbUD<>(EOKrw)&e^lF16^;@Aqvb+0} zifumJ|6mC>mkfpQ5@$9>dXHVs(V?sK2iy0Y!sKi*o_kx9uMhb{-=CRdvGa_!7ur64#{oVEc(UbaqjSWm{-PW{=J@beZV{H4!BMke*Jg}Q|6bJ zU1QT0Yq5W4z1g=Kk>~t0gDONGgN-;NUpUefr@iGkmsB}Qhw%SVbRK>^{a+kUd!Sv? zPN7};+;h(RlgKO-No6D@GBP3}<4a0I(m=b2$|@;IDh=A9G{}z1$R@J+-QQmzk8bzA z-{-tu&*zi`4VEp|NK1<{=v&%SDl}4I>SmHm$+?}DPio-i&E3e=D7tdo{Vw{^G@PY9 zHf2paE8$H`WH7m)Hi8UsQbOCT05k zBGZISti8P!yNU|2OM=m92TS7i&LLBt6*+UxCt6wQLO+wr@uQ^*S~Oh7(uW7}TEl9z zlTSlulX#x>7DD2%fhH|eL;G<_@RK|#J4T;%Sj+Ik|2!%=!n=oS?&5FzcPL@3$x>g5 zu=x65zLykAA`3mqYM(vt%Dm0#j-1VTYQ@3D`5R%`rh{Ay?{{?27|D)|Eup;={Aiv- z8ulOcK*{BD7`1O9oy-)cPdtMP@20W8A7a_TQK2lSMuJ}R^UqG+Ngm$SNZwzLQ%&C# za^f@WxtG#-H>VxlG@U~DP?^3&n$Z3J*);Bg0qH59!nv0X=qKN8SFuI%ef;=-xHsrf8qPte_`_2a9k91w@YKK&e7ya^4x@R# zuk7)gwACnwCS0o_qf{Lx)RJawhX@4s9oDhV5ZmRn(f;gvkQjW(q6}Nuz2rI;a7dK;!{1}_I&0c;;u3yc z>wwae_{_to2{a~Ko!aWOXiri(yV87~rRWy0w;6{q&nFlcX_aEY0T=duk`$}i-9z^_ z3}A!rEq+Uvg(^ATZ0N^CX8g^Pb%`4JH$J4VLd{;&J3<>*F$ZgbFnp&|8F9il+ zfx}rG_NI-DB1&kLz&!a`Tg`LbRQj7ZQ|K0=~N^qMjy;_(W$kIbl;t$8!4|S^G5=8Xj_mumq0saIH9NO zacEC>hllT#*xgO#R8!JOok6-}yXrgMiL@gBY-#2k)JP?L@zinsE^cvpi&+&PFn=qs z22&F@Le7GnXpKj{MJz0Wzvpd@^7AO(ltJ&I>GHK62|7JL_Z7gIap*;L(=Y-V~sn=v+*X|z?b z7y7r^$Obi*6EK?P?9pV~k``bNk5=}W?2hS|?eXciIrz`j2^T!LffBjp*f3`pxy3Z2 z_24(W9X_74CwG7lzQUE$O%T^~5z2P^z(uhW;Nnw?U)!Fcm-`F+=vsi+DFcT+??4U3 z@33t1Xq@_XGH%V0LN%kM=xY5Nr%$a$t+CnoI^YB*{`)NO7a7g9{n6q|UJb*OnesR? z$p$0Rf6=iNdFDHEG>dfOIqp^r`uddchP)V=UKOR^*2AfPR~I?t^wX3pziIZDm8isL zM=~5^`F@QNmB|oo*%(B^)EIgv)YwP>r(Li z+dO8uZIC= z8CtPzG(8Tsrt`aR!c^pQbR|V_LN<}O>|8REyi8}t7UR|1T^LyT3lHB(qTg3bNp8k{ z+8P~%WA4V_?$PmBVY8J)^^TI|`ExWqmfyc8$*}vsrI}aseE8Fu09A9EKsqcAR~wp*^^-m~vR=5wm`qq!-yG2GZCKT&<-Pju<~fUe^#I3uH8ffUbeI_$B8Y5!css*JWV zfvYh-wVH;C`a$T^{2E^$lck&2_2`k)eZ0HA4fjs_j+IZ|(Z3a6Y1Ut&u-kAE>9f;*>GpX4j9kp)n|JT@Yx0l*0D#2$ydlxkgBtWj4*eGLjZex1^kUcfMB|hH|cR@a3cx zXmUxD)+%aK>k<_ju+@ck!AChZv=DAoNYc{%8Wb2ij#h_tv!UM4Y)|Y@mSk2fc=0Jw zP_yu^VExV~H0S>w?0GV5`Un-QJih`L4LGAyg$u=3c+<_b>uFzN5xotmq3y4`spaZS zfu5A2K&-S&oH<*X9Be)j3q0x`SiRvg<^-&$zp8~DA<^@(hU%VK8sZYQ&UqA(6E4484jis5fNKXkamI#G1iBMR^FP9C z{0z(c>p@hi}|n&IlS90bS0U| zaqNrkLMA&lnz_G|VC4r4SfRNin;w@!Tcj%}!|WaPpE(E@dXB)=V~t>5{u)!ye8$r< z6X^PzQm$%wFt@e0k!yP5iM|anIN1CG4~lDI$5(Gu4cv-Hm#WZ#(`xi~XgsCdj>R;e zTHYZlO_g2Sp|mg$rUpx*r0sMR(elB8$a>T+x(Snv%OR(4G+I8)!4K~*BbP9Y`mg14 zj$@B=ha%5&u@i+533FkkUmgfOoZ!izBHYr70E@GUAe@j5-4|uCzvdF&6@P-uRfkYv zhZuVntIST6jbk$2w_xvBRTM2X;b+qksQYp^jx5@OM*d>-v8o#jH{IvGE%kWHvJkWC zn)y9(KD?NB4mMso0^dHJ#0Z%}+$-@7OV#yoLH#P!4?T+i{pv)KnHKbgcjjeJNrHGc zP4HO|2s*7dA@R;{SaMPwPuN>gjMpSGh*?L=;wNI%J;aS$=Hs}XIrw#18?IO&M*bEF zm{^jF`y<;hQ+@>vUfEAptu^%YaudjvA?nFR^Y)>H@V%oR!53{jNpC7FLS^;{VD8SmWN+{KgLtt79{_$+$MEg zkl@*eaDl$xPttgx#(K9gcI#;$&R*Y#p_9Jg;d^R$<@rX;DCYl9J(qAnc{!HNQ>Gc^ zlFTVfmFX0lv(yS>EaAHmn&al7NY)~(i%jBYSC=p#x{wQPvI3>0mQYf*i{4+zqi1dp zsb!@DIzNuVfW_dkt6+EIhVG@uBPyhW3X(E9%@`M#m|v@$K!_20IUOV#W=HVO~UzJI>W**1W?_=?h^CB!fFU>MHogiD| zaQc*EK&IsqH16X!obojN2>smxlEYtw*>cY z*TKi*{z8Ou5w4jXiIV#rfhjdSR#!F=jjs*h_9Yh!t< zG7A4b!VR9;h^zNu+~@>+t(S#;d{-j6nD?sH68r3D!EU7nli~*+!@O6KWN(}%=Rb)w zxNAG*+-;yqqw=Ws{x(|O?*#in6p6@%UlBBRVz9qerqdyY==p#iaDohTtZ+;=da-Gh)G0 zmZQ=JRa$mKoK_sO#zlKXFj%P<9=R5E{hYe-=TV}=$iynY$P16I3E98(Z|6j(aiAvTz284A&b7v;Fij1 zsI{o)yMEcUYG(!wMkP}0ek56KO-lVE%{#t^;q6Q}PE2eCcXE#wDq5IeZP;`?U}{Fs z*6=mS0uA2HslyD`i!h}Fcgbk$KlJB!uZdCOcSA2r-ZOt&%$O)$Yq-;mDg<6S!=P5E;$xlEy=92!*RWV zFZPtK#Pa=WG-bUMMMwO_=XoNmIrBW_7M>!h&0}Fp>P;@b{|Gl(b0V7*Zo_u2w`N5b zZ{YmPr*Nxx7%rDtK-wizJiAwhcPk`QOxF&2YUf2a`_53@o_M;m#h(P8O58p^tF-ih zIQQoBK5Tn$jdKI#FvFeSrOPD4oC6s!cl8;1l&VI5NB_kQ$4u#Emw;@~aJ0Oml6BRV zuwd>q8~kI5gZ!Cdt=0yNU+@4U1?hb5&J3##rlQQ#Bls~d1*=upVoa|c_MVc*i9by7 z75c)hG3wwMUxtqs)?mL*BI=)ditpr46~t0O+U`D;N>;U{%Hc zJ4i8oF9~*2REDkf)JBgg-gn!)5aYQtyw2x}2Y05x!gsaw;puid9krNR!&7isPZ6$s zE>2#Os(9r=Gz?EU3<+JoaBHSF8rJQDb%SBpvgH{XA8*EMp0+el&ELh|yvLB~F?d$N z4uw_xj&z(F)3AC;DFIt(R(d-wNJ+=;rvG@C&sI)BvO~~mbVhK9pD7RHXSFZBPr-lt zO_@cD5j&b~&L%(p4)cb+gQq)R!oqfWTIMK1_j0Q-j{n`Yjgn+jemo)D+%gQT@xW2J zUic>BHgQ_Tl;WOA-!gX6h01g?t?QvbcK_j>EyM8N6dPRP{0iqR;LqEV{5R!OC-yzz z`+FN!;Y7z=oFo;8P#KR)Yu1r6|6Zut5kouAWy6Ib{!DlEBGwHHz(Yoa7u6&&U3D!T zb~m9I7d6r>l)%K-wNUr#6iD=?(=^2-s@2S*+)7Egf&P!fV1JSpOnoqxg*J(@ z{XvSX#CRV}5TGO&WiMsX2wPKrmm1=n{CbTWRniAy|0bQ ztM}2=wdQ2qe+6F{4nrk*Azu6B&!3wd+1+H0srHRyCTW^9?%!mZ`e7!m6M2L~-=pyz zEyJRVh0s%+2xr?fVQpC{_M9>0cT_?O%#g+M1;*$y(GN!+m8HK94>8|QgnZdtve7Z3 zt(j6(SVp+9PZL#VwZH=97V=JNq$j?mG<96Dxg4_Ifm>*GE&-xJ;gb^b7CC z8q?Z4suZCA-iFiAvGH9nSuiPz(emKwR4$=Nw0o{l!pT(V6D%uK)oKw0m+1?h^EWz(nwguGj~TZZ{0r>?%&3JF5b%h z(MsT;`#0r$JWt2B7t=ORN75bAr9DQk@&2_?`u2JsC2Ah0VP|sa^u=6i6UipU6G5nu zh4^cfHD)bt5hSbEaIdOEVC&X7%*xJ{DfBF1UOWq_F3^F-XicG%g0tE zPq9rk2emyOL1;x5480DA@aSGxl9&gcwl%O?ihmA|iIbcx?-^>f!SU~p@xE;-Q1!hH z2MS)pMAv>uc@czn{LQgW+nVp4cvE=sKC-g~5>2nf2Osm%m%o1>h?VBL>-+?Aqf-U5 zXV2%ppSUN;$vnW79O&ac9<`K{9!FB=+i*u~83skh;LZD&FnE3hE_=)SACBFo(Ho0M zx$*>s9XbTQG6C?Mci8p*+d^%8#v#DNi*}FbrvCRsWU!kr)N~2y_ZKsIt!P0{3ykR7 zVIwlZyg~Q_){47}fF1FkaLw`kWObY!03f`X3JYzEm99-zc$tBdg zoFhLPN$nDRzvrD4)mOiSta0z*+leADoqLe-vwbMxx-U7+ZO5JG^YD{f2sXG!P!i9g z468az?c2r3=Kc#j-N)xF(~sFSHKy4#IVaoXZ2f}yfmyi5xR~E*ondbSVp-$Zm8@#p zdCoC;JpAXc!h2B^+3pB5De9|bFciy+mT|+AKGU%Z0PLiBA zlNu{}xx-0C+^?WQ?%OLn@=7M~*oWMJYwdnFuA^8W6 zBAe$ou|V8S5a>gKa|NaX$-(^?!}qz5wDY~k3M1AZGmIHrts?OZ11cILL0=rdp!525 zG?-^c$=ehu|6DALJa!N!)EtBn$_B7nwvJQ3&oh*lP2yy`k8!^*?t^BD6)47MWw$6j zprEx6@R7JA-BGB)xHsU=<y)$oMJSIGEy3SQmHhbFO8&=JxA{oN_B^?nVcEgg%xN}04~%OQ&Te>Y`b z0XJq)4mV!iO7jgB@Komu_~{!4D^A}8#nr=5H+KZi)r!WwmgBH{!(ULs1iUEyAC^3B zz;TCnaYNh6xZk>ZAo@3%d$7xeJA7Oj-^Q7PslO6#m;QzNE7Ebv%LRDv%_F!e(*P~8 zHrSIZpvKhQBs;K=_MAIN9#fJz^=CrPD^vpnyob|MLLc>1O>mxV8qK$fqo9$~Nnb6P z>P`lb>TBKs{I(7km7Kz2f;IUbW`ZN?+ z@lOSszoHmZ!^YtDvKHt{;`zc_6Y;(03L4+xOAo?bsBgg}JZHNRZQ~|kke)u;xLD!f zq6_#e$d-%9d%$hhlED2&^=QfanN~ZgliH+XI555vXTmxxcs-rla!nnUTMS^_8856n zxt^ZiTfkC2DPobS98M17vjOf|+`5PazW=Zj9xS^8$us982utwyq$3zw@Dx|=-A8oU zo4Q9XrRU8Cv@TB`Q?7ET6?LDFTSkL*%w)LvQwEZUSKy-iXKT)FcEl+X`*N55-mk%!*Oi!`--_APKpOXj7Z!pWlR(-}0s%ANjt z1yZlD5Vn=hNAE~k@}4=J44#;h*TOC|N%#ttCXx8};#%sta@eNgvy`B1!y22}u5;u1)hU_bW3=5GkrlUnVc5oTe+pn%yD#iy6q97el6D zD;|%ZKoi3F@4f>+w+QE1IdS*k(`qxW@lhEk^?E0#?>>ucJrqdlr5vpv&%4!n2GM-@ z8M15Vncj!TqP}-7&I&(*X~8wv{AU3L-(JQ|-gBC3SQG<}OGK&UR~(b6iwNOpr`8f8$cCd(gHTENx9~G3V%u&M zK{ih!DOD+uemL%<Zt$u1Ye5<~I0MLG8Hv*|HGd^T8alJHQSlyEEWFg4NrD6|f$5%L6Y;djnm==R-Q z__$q6*l+Jdfdf$#610O}cI2SN#R)j`RTW5^%LxxoF&FwKP8Xg!I-Bi1?Zp)HTv_G5 zZ2a~m2HUonV9(4t3VD8$d^+-Idx;9$@Ysd+Jn+O5uj85Y(zi_Sy{Rz9NP+6EeZkMB zj+o2hnE#?McD2u({j*$3_2MR!JusUTXB>pKnRmIE*?oeawwp=ui5-QTe8+tshVghg zIpM_VlEQweX#BlkDK;)QN4*dM1&*13-iBu(Hd{>CCiaf?jK9bbaI`Ss*eGGb3Ep+p_|5YzWD+(qpKy1!Z<+`5t z>UceU1Fj^L$@!x-g9j+`P_b)Z=aHlmsU$7_Q3U23C^Zolw5SW$e{Lt=>M*+VNQ=Bgev;f4 z2ig{0i1N0>@%pV8h*B^CE8Xj)6+DX^?IkJoqlj?Nj;!nT zsDF7Yw)h3$sS@5NsdEOW^i9LFvuCr!YD*@4`VV>26Z(Bch|VfygzY}UUkEc{>_ zXKwqQ(`g=zwC6n0_&A!i=%sjV! zC%>8smbLU1GZ}ox>Tbru+e=G8!BP)2?oSc^?42OIy-7o;KXWBT+X)Hnb?BIkxlncF zG~q@=PN*K^02yPSaKU_UzLIy_U7M@Hj9SE4P^uotPT}Xx1%J4$(|&@ll@?67{#9Tr zF&%!I-{WSvBnieirg2^h?Of*1biu80BXGX6KKK@R16N}QT(>qH+%YJ4=4=Dcp2u-9 zS|VJdTMe$@GxZsZkMV_dZQg|^M~lqlP}ejQ^B+vd5!Q=gwT2lq|D6WQHxzKsoi*W^ z?FsHo&=jr|Z=qK7W*Y52gT&5nhNf2jEV5jUTl2Y;Tc&%IlXf1@?$7fzAk^UrlRu%=HYqDTT z@kzL_3AswkDz5%`p&(#)KHSzG11sMd!$N)*XLQS&~aXks_Q{kK(pXX|9|9GJ(>Qi ztir1?{-Dw`4@-(;@ZYX7?uW8HCbh1@tTVe|CC@W_Fx3lu=9P1DGqkvP?K*+_gi>%B zdl*MnK;CLfE6I}2cY*>2GOeTtKu zrbW{$F5>5>a=3J*n_zX@Zo$LpcWuIs)pG9Pa^R9HBN%^C6*V_x!0eWj;H(`89}ZS< z>OFUD9wxUzfBF%O@eSa6bdsIJL>!F>gjV&QNH9 zmv?7FZ0I!jaAr04>1Y66UDJhTo9j@#$OOXWMgp!iN^M z)xUxI?qukYe=bk~!B$myivqk;i`mvODFBtau?>d1+$YkTLhE?*_L- zaT^!-YZXk`cb1d(lmg%U5b9htfwGL&O@E4T8RYMBi6z>U(aMSZZBcf}g3FM?~ovISq4X5o}Q>NMf15l!^E4C8;-1NZJS++d@G&(1w( zLlMn9|3Mxdt7UP6(^rW2H5xzU%AnefhmbNxlAl$+;OGDC=sW8u{54w-x*4XRs+55r zpKQUb4~^(@S^*AqjevD#&aixPKJ>kgh6Q>A%2Se1WOpbE39FA*mPJDuFB49H__xrm&nac5{(1D z!}Y>*pna{L8V9b>Ox&zj1f(eB?x`o^sMpzjM1?jbKR^ zudS~aVI7Jc^y>XEv@e~7slFF@&qFv^u>jDd3*4Q$bn5#ONvHYWYdh~rH@D9gR6l+$ zsOj_~o4R0O%W7rz&P z_U`{6CvrJ=%`buLT(1JNk9MI#-dD_NPNrKcP3YM$0X@=^rr##Kd8TD4o*k=(nmKY* zGkz@9j5FM96tgwsj5H>L$vGES(3Taxsw6 z5DV2NGT6{)j~S-iO)IjBf%=F_%XATLhnO{3Ttx zopis-joz-~^D@8qp0Q6n4)-dfuyYqF=+_If9P5JRACBTxx0yK3*n!#4v0^!I)R=%q z;~cptImC0zUFCN}*yjb<9TkfEw7X^Par znV>}YuDAf&j^x6|(k3`4b_^D2e#4OL`&c||kh-^p^Bz!VZm!=+j?Y=0zLf#B^De*+ zol`J}^GEvxGi(Y{!|}?NIMt&SHmrLFr9;DMSN|RK%@QT`N@qIG+0dWLLfjr41499c zz_YZV@470LnadDo{urlD*5W-y=4fwM4EOpJ+0Bp+o~2$xO`Fs4&6^AS`Sv{CZ}@{V zgiCQ1tj9f)8=-aMA&3?k3(YM*$zCs&jA1KH>nVi55JT7~H45&}v1C5eWtrcjkCY~| zP2ioM!#R1~fsgf`Z0XfPw)C~EaHaPV7>KHdS)hUqYleY}!Vl24c?Qmri($5!B81z> z;{-KbZtUYO!5f45+|*D9v~9>n%cmt6C^CW@w8#;xDOKVmRUQePk}L#6x#EHo3vFQj zo#PzW_2(Z$K=(X|t(XPDR=r%L-6Z@Q zs>y#Bhp~=P!th z^fh}lc(4~8%xh77WjCh~=Kvm~D!?Rm5~}9fVA`q;m{XfUCa-2uu|q#D>;DaGU>AIT zkjrfDuu zxf8HV#EU!FV#?VatLHLCIC2e=(ztB)T#!xk;A{?D;3S(p1+`0_a*n^!KzI3Lj)~+z z@2W+ZonMD(mKrq1^dJS@ETDW6WCa>B>%*EE4dpY)|8Ly0Ws)%F>|yGgev$Idker#d!-Zf?Fty`m#oSAZLK4IY7wGNYhqpEb98>?JNmx0Wm1TL2&QpKvyHlb~45 z2?H~Aam$DU*ck4FTe#Dh5Tj1(o7C7j-l-&GYRs$+-r%f0OPYS(k80xMp;Y!R)Tw!( zT~Q>rYMd#zQ1TeZ_?&mn`b0|Xc}b;bec+I`5BxcF6#V>SxnIL(ao5rgaJ}La!Brz%dA_baX6JOX<{ zwin34o+AX_n|6@Sqe$}AdO)eh2he+UHio_!LAw>4DVWa+UF`>Ex$q%i$sE|fNfQ2+ z4x`NeuNbbIMH`)mqp4pr_@weYCUc-YXN&39d_J3XSB#3XU*PA4C%C%pHr0hWFw;~O z;p90*_-gD6Y#JC(3L}-M>D^5dRqUm!pJMUG;ASk~XU#MBNYR-ZD-u51jBi~{ar^XY zlr@*5&#xUuxGRYVBnc>b?>`cA~bBRh~pEoNDrF_S+ zwW?c~a&IR;S$M*|;_pr={Cg>u@2;G9I)_Gt@VnGc^YP4YN!%8tM{oF@uUGzL+o_-f+ z@O#h;!*t2>!Dz~l5s-Dm3HmKKK!+U-$v}4#H!WOa3qEr~wC_ANv$P)K8|s$d&lbcs+WJ;LvHEID4`3 z0@bbL8RM@XqMH9UyjL&`H(ulW3SW=lr?F8qcKTOb`=31JzL`Z+!Xjz&%4|BM^phMk zGts6>i|p<^rh}2Qi1qIv=j9J*K<7Ti9jait?V`fD@@^P1&IGm9mhoP^dI)nd$3S&= z^vm0VD#upQrPB-OqE;6vuJ}ayB26^DPnu;f%cn=zwWuKFJC@IH$AL=|@yaEA_|eb> zS$<#m-Lo4G+}A)D?FsXKX3>?ox2bD|EF1qP)aFL>A1?gyT)bVA%?8I_U@qSqm`K%s z)RtF{&cXxm%3>z&(TJrv-J_WJsSVT}$3HWYbI2tAG7Fp$$~wfP*?+g^^PCaDm{)07 zY!X4aqbn&*QHwbh?t?l$=Wq5&g=*b?)8fuI)Y@sljQIER;F*n-lNd^>_lmLXFeCTa zZVLKQfD`RZup=iD=2ev9xeZoia6OW$CiP)k;~KgWHIhj!y^aAJw&SGDvb5~s0KNZJ zL=U<(DR=8CRDb1&2krFm)%7CEkgcNeZmpDNXn}^=dKC8Z74;Y%rlmV_sj&S4jUK<4 zyq(2ZcDV(6=)Y}3YW{kG_?BoG{nwGK`Y+Qa-lz0><}BeptvN!A3-g7t+uCikyB`V0 z6eWO}XdG*E^k;GUMl5b12VG>2V1wWmzZ<$iqqs3_)Ny$>e~|~hQ+R>-WhwLzOSo}2 z5}^Jf?-RZ!!ZOQb*}-H(c49*!b>+p8`ouW8_s0-(`$wZzDZ^VL3vlMsr})|FASM*@PR=+> z@NZoSK_5?XuEW&;?e@d1{wQqsevGdrvv5BMxh<0txsGB{n5?>-n=+fh+Lu1?JD0=3 zs9ZezRGE}z7O=0%&MbA#6lN$(IBt1XtmHy7AYaR9$>1K!QJG$sNS)*REj4pZnaahpH&U zLh!|jc`VAqnXNqJ%T7;mfo7fqW>u&SO&?!C(C`7+{X?JU7Ijgr&TuA_8^g2>#`C+R zG59M*27j0;;RcBjC?@qC#;FH!l{dygQsO^ItK82SiT&bYdgGxw>p3SrS0B}kqcQDA z4}VAcgPvipu<%?Bjq|)n@sGM`0YCrKJwA>NofNV+;LG(4J%v5yn(*b=XEN;5W6t}` zm`6&Gb->U?eiy9Fz0;cwI=V|?%I6qJ(6z!h)2HK@a$B4dyhaeeSc;p}ro$~>QHqxj z3@4?D{8@jdv7kRDf$OOR*y!KM^75~-W5rk4`yXd)UVqrnU6z{%vnC87Db;T66rUcrH|gqkkI#o z%M$v*vbFXkRS-=5DhH{>*cPIy?r~vtUJ!D;h}~O#g1LS=&SH}^NJS_^dDmn}<9!l& zZ$3$FJ6_WI^Yg%Y#!)z{r;OrRN_?M=&oJ#DPZ1XO*5iLH5va6x2%3Tya*u;zxJRBV zxj*);G~S|zl&=1xLqCJ@U|b*On?>R7RA(rf{tX=0PR41s7I71KUfZ{or(wmFO!#!c z0F#ArSTg?`#4Z_zW@o#>=es+r6mw!@g6-L{U&ATi$&e1Okf(FYrsHVMW$2YP0%JNf zxU)O7;K_!tHBITp zAlsu5`1TqoJaoYw4+beh^A%N;)ziGKh7g}K9{iGCaylBv=oQ~VyBK$Yq~8sKi$gBT z{5_6R4^`P4XB~F(#%QLva55bZ^xz$${uH`cmHoCp z&D6g}KHEN0X6A95u!)BShY~Zm$4~b#rP^)myZClCH9-eYN=snjSpH7#9gOGoEy=s{ zG@U8dzF4hTG`0uOEI<+>gUL zCS(1L3Y#aViUl#sJRAP00}9T_;}+@Jm>gOz_&37=|J~#7<}3EnUd;oPyY38G={@5< zq!(~j=cS-8N*c-zNWt{Cm$1vRoO^3*3F{n|!imwTHWxk?aPy|;a-%(d305w@40~pe zfK>kd?ReLYhTjUK{rYzVPo^W+bZ8&U)*lpzM4g1)r6RCIw3`3-w!r2mfhbitio1BK zf%`0~1efxL!`AInxa6MQ*k2ifD#~Nvk zUqNg_6=%G@nEOy<2s1m*!(6|!V0Y>sK9)K_)f(S%>$_+$S$2n$9JmS_Wm^RHGFP~I zn?zW|R6yFRm)quW0tO?FLydnSyyV$8(F2#cHy)#z z^FVEt=)21o)EYLQguAy>*ZB2luW{k$2$OKbxkN0vsVQ)Xn8Cf+ ze_zn}vy=>fe4ss7q?pE=c8Y3a>?imXXcRX++>m^Z=#?hqzS zAI?cBxB?`s;oR?D;AU{IxW*f%Xwe*sdpcTilvFytIDUY>X7HUgE}4F<{3Q^%u@a2` z?dB4b>R~|oK5Cc93^YE;Bjp66GhlN#OH+2$}KAlRwC$^D- zd=x!rYTS#?2F~`4G+gZc%gK$m2hriPVD+svu>R3BIIi#&?7esnLcRcxPECLTqf>C= z+7lS8sza~(5bm$5CRl9M=lm+A=~tf>otrm~Ht;^aT(3V+;dBRT4Tqz_^CXmStHJIm zO^{-ejo&w|#D7)Vxb#Q@zIe#@YI-cVVfk;kn!R)3I9xQU_ge@ByMHqGkS->=+Z8NHm-LjLX$%5&6j z;+k3)8h1wq@4gbp6y4EyPi8dAdk9eRwFjCFO~CX-57gHENRmX_&stkR}$7V7oSrVzSH3Sc2O}kmC6i*_-C#C)f8>xKoPagT5e?t|QrY z51KCPME9aYxCx(Jx%i9X;56e8EIX`;&rJFAV1p;UHG7YC0^VIPI0Egv0yxzD!`Vll z!WSugW@Rz&pwrW4@@oHR_`ctC`mz9rlo&Q%TaPS1iHwzoaQIxFr#!)h&9RxlN-i6* zkc3aT<2&!E$krlbJ~OaG_b#cJSJ8$KH7Nfo2c^Sqql3dabZDr;qr-k+>&wOTBP5ar zk3J)>;oGTQiQf;`^Y^Iwcr>aVgHnx-nD^A28lH|O&w67jJ(&x|o8E&*6It%V8Mm0;po1P`+lVg2o!V3O)i6E7Omq6QB- zC+|mImlx5hkN(thPXhkiwg?u<)IgJ4ITeOf(W}LhY}NH%5Pc+uOSDXIyWeMellz0p z+oV~~H5a^Qk_- zpZMNv;6zMR?|`0@y13+j#Jzb`jc@q>o90RLAPE(XLVG{=eO-6bBq>RxNQ4xM$Q&Y? zNohcXN)(kbDhVl#Qc6lGbH+036B6-zzJH%}&RW0UIcuHP+JCiH&%O6*KhJ$%*ZX?C z-rBRr#gkdyUQH$&zX%t1@=oIqJMm_HIGO&yls?Y1rggV+;bcfFjD7eHeqVjh3@tyf zkMVLAd*@8x9nFzs>F^BlFSHul>@TCRr3@>Toj~-7B468D3}gIqAV)h4Ja5!P!}~Kh zH|7Jp{r&`sSNYJ|^tWV+u!8K!K9A=&zULcboy}k2ilNVFl0js(RPXs;|CdT+(O=@SF-&4jZ789fsuF zjr{rcHyn|;N&D?dtHNCi5{RG zD`dGJ+eZn-wE~3w1K|+d{~kJDL}8X!C|enJg}sWBvIyxxuj(-?}ZNGkk znVSMm9Z=v}W+Vv@rsxY-S!`yllcU+kydBJY>}4FK{uq;PnlQ1UGi3F8S^DI&8x@PR zBs%c{WJ4uA&|cOVJr{yHKGN5V*AbPz*IZWYWm+Jp;1Sge&ydBo>Udpx>?Pfdt3YgoS&8+1?7~^(iFr}i| zpxsjp%Wi7o=m2l#n|+k6%5Pv-M%@v%gwNsP?#$xSg^|=%`W)@PECL}FXCZRtc~Cjj z2F(k#=)UL+)J|;?h{`R7Qp*~;O~jqF>)a4F8AS{4B)t#>Cych3T}3U%+Iv`pYV}f? zvq`kRZ5(}`b`rI`Y*99G3JTtjBz}BWdGMGK7nSM7Y7BQWg>4zEHdk99$Sp88esM^s zz2r7_A8W$`8CiC@QH1slE~9Pc$@JER3D7cA9$Hr^QSU>CsVlurQ(L{jaJfBmy)&P^ zp?r?&NCTJ@dqP{0Dd-u*s_nSnYsTPS|=w5dE&D4<;91ba$O6FnWRAv z4veJTHFnhFSu8bJIvJ0hS&Z4w_v8M&AqZT;d+LK?kf$~t=)(#JXV26h8LAp z)#mr9ybFDCCFrcZ4eRo*kl$KKR5RTSx+KT4xP$<>9#RTJIR9^|oVDk2Z5mGG`-3 zEnrz0ndCsje&SP=N<`drnAuAQ_E0>E&CSxIvscv9mAe6=N+elD;7{D?_#dABod(*= zG9j$)Jf!F;SS(N*Y4NsWq=lKAH{9d>s#U|AK!@k-gjwl8_mC|2aot8bm>DZDP6 zIjNrzYa{x~Sclk8*MLWSR$#p5EYdkc3X1j^!V}rAbcA#~9sjPB)HVDN{yZqgW-gUr zYd?uGMdy9&@}wPX#n>(Eo_0HW6nEk-^9qcx?4xmhaIWkJn{x(DQF{sb}IVDzE+vUX+MqD?ihuRDB)73_^L&*fco4c7Ql?d^_18 zgFAdof;}83$_7Klm`?G2y4rFXWi`39WUVcfCz(P<<{x^w=>*ojjmGHYpWx9so{9z7 z6LY1p#w{m_GM?mdpnxN@cEYVoS72`OQ_$EL zfRa25Z`R5~xWz6X;aVooF3QI0tQhRvnt+cLlhC&EIk=901fv3mVffiER8SCMp#!{2 zD@d8Fnjy(~jGw_teYc@j(w|8E=zBEDNe;rNETO)8rwcU)N@0%lS!h`q2AQr-5I{#k z7Vo0}WRt`7S?m_5xor}t?@4Awf0LPcVOe*^#CLwOHg^hhaI#abwdf z?(5SFg86fnbGnX(t!Du}DGkaKWI$)! zBl^An2Q+EEgI5CskoR^Gq=Y2E>9$$0sE@Je6jP?{WyEX}qu@W^#e8j)?^WM_gDsQ) zq0~z;c1VMtxs*$W*x-wBV7L`3L{>wFd@0FYBTHjfmXb9=J>>bA13bgOj6WA3gvQ6A zcx3>FkMcmB5=BOMBmeJr;=?iw5 z^Lf{GDZ4PGik&%fovFx1@P4r(GU93%=dn5kR{87)D>Fyf_@oIgt!{$*tDk|Ry(jo3 zX+p^kO}fqUFs6PxhKgqjQR4IhQZwNs=XkG>%bar#e_Q0EpJx_&Mkc}b%Zor!`vF}o zb`7WT44~*^`KVX33$KLk#+^UrqvE=Y7%?{k4}=%NcB7nnVGl=)3_vHp*9&17#RoU{|>;=hE;gDL7X}Mj$?C;yUCfk zlc`9B2s|B}4ySz$U}iuA7;0}N3zc%Yyt+be@)~_waIlsf{ijHs_RI%?jx&@$sfL)b zyQp_t4$ape0e?Ik9f@P4bsE_8NhDA{81m@uu8 zV7Vy3=~@C0ya;i@PvAc`K2kyGVtu6`hqc7M5mC1C4+CL3ZX7dU;9>{Z=A? zMKYf_=d}r3Rks2Cx#B#%`741I{{2Fq{%nRzL8AC&{xpB4{G|z3U_;QM#w_wyJBa~a0Hb{ExH=xD0Nc=0T;vHN1^l~fGKM~96MUM|~ zTci|B^!mXh4SlEIz8<;+4_+GEL@&Z2^j(H-`&L7+kEFj zA7_vy^L}s#IAtO_W`G;F`x2LVJetf6RK##qHQcv=XKITNpmvc0M*k=KNxGm5w5HUdB;cB?XC7 zb)l#GG$d)LK}7xs`tGeZ*s1CAJa!3)$X>(uU}kggvJ{0Tf*9dyuX$|i4Ohmp7P4=8 zONm9H@+v{RBDE}TvpiY5zj zjR_a}R~(Pa>ENXwp%~^EM}vjRP!^*Dn_3YhdFJ^Pktk^LJ4=r|si78qIs6PAKac-K zkFVQrA!uT90PZJ?;xr8-Y&YA8P8&IJ457R9u78i+Qh>tuz6pTZ$#eKx&wGowduA=&@ zzG1*J5ys~A;T76Vm2N(xxPBEJdUFsGU$nqfp&9mBXkp49Q?xwlhpFx-a9(jMHmsLs z^P3KnCG%I4TdS55{rfj)bjMOs=rEts5r?@&dnAOp{r1A+Yk21H^S#hxy$s(4OVa>iIu<_j=;pZ*UTu;z(qM1DgF6lo9J&&U}D#ePb4u{d`4Ug!B ziZ{ft#g|^~7zx)33g{6nE71KRht<8CxSXgZWZL@%I`gI{J(2#K?pyc{>YPLAhAq?J zrKA9tWv6ok!VhG@>XWeQ&v*zNPJ=1yopHyA%VgB&JLGZgGZGb1OMAMEVD*H9ykD=I zTRMF^x6s~^=s!##`lm#x;*nnZNP37{JG+&{PC7*QX?B6Hoe0_%sv^JfAStdt=&~IC z+|=YscFx>SWNjDIQ7K2M=nWOHFp-9N+tv{`@15k*#?ds!bvJEoX`(UO&S1ZPHgT-s zJvP^MXy4;l?$g-WWW*w#Z+2)e7i{6dt&n{~a zZarKaz5_>l&G6uh6v5;BRM>cH7U@#G&E@D#C7Z6SA&tlRJ^#{5ko=GU!*% zes~9c>*vwCJ$dB&hEephAZ_q=;fgZdUD?zE?14C#c%FXe{Tv`%~nF&xAPEOw4K__NTlnnp3$X~W)Q~}9puz6 zp5Ip^NotMjIkEgB^jEhot-1VwwrtOV$$JK9g_b(}s67wN#8P-!(}fE;IhTAb<6yP# zI9$G0otayulYr~VBAX)PcZ+%XK}Xp(3M;Ashf-zq#vC{vU<#@<7a>#&jdR8NhW<}&_=_Zdg;6Idq8f} zLTsBQLfXdc6W;de6~;zgfTP{&sH>BKeTOd4hi4zrCL3vZc4P{*yr&3Jlcmuh;y5|d zo=mBLwYLg)Ul!7#bR`qat7Vy zB7l}$ewO6xSQLL4=WiG&%J?9l>Yg<~}hZO;ti-dnufl@w_5+ ziJLHKND`lxNYb$vo5=SUWfU^Tg7U1lv}FGUdi_oc-DUlgz8)08h{B~pwWFazk<4iB zk#{dQ;;uWbw~qpY6LNUBB?)DjDKnV#jBEU9#Q9TA&QCs%8hx~f)B7|~SEX3!RnSO| zo?Sy+lXzaIp)J2B5<%gW#hmg`3-|kB3fViZfV;A}OxSr@h0arVgKH-HL0M@4o=ts0 zQy%2NbM*x%(Dfp6oAwZ|fku+>Gn4u@j)0ef2XNSN7Kut+MMGOJz~Prx5VLv)n5J?d z@pma-lNW&%n)ktTts>psbC6CPLtvJWuWy9;Q`sQ{m|oBcq6?nG)XfjU;lWWXPAyl5PuYQluJe0EY#YIe{wt&?sU15>ytJp*9N_KjI z0h`P-5A@dwSa+#EBoyy};Hf8I9iPFFuTsYMy}To)xdXl*>VeW-3aE2nJSZ7XgW>HS z(BkU|v4la#Y)Lq9I~()v9mYxR$=I9H0@pO!;Q6)+s0b8;*^ZXrb0-GOHx`1^=Qvm~ zG8B#-Q^VdhQ?dH$bnLm<4l9Cg!mhXy$k4w=93-{Knk~wtMXiS(Tlk5u>+PivzUE{4 z+5*hD`4+ADJg(fSz2MRi21T+j;9Jv2*kM%C95q)^P2>%l{kJjHv+SQ_hU)#2~0RQ2w#OQu(|df zbl(XAOU7hXWd7SnFY3vM^#Rcz&p)chDx*qX2RxMKOM-I4;T@i^=6F;8c(av!s&Y4VoYA*KvI#wh$AjNf$*p|bEGVvi!vs#L;jd-|X? z^(xsd{h6pe9!&>yWbyJDIgFUd^Sr9{Ekex4S}bnSuozx)lKkx2M|R~a6B=m1o;PT+ zo=xg3>Ulg&NKb;llIieoZ6t=g2}f<4a9q3m93I)1hXdv%nA9CfzC|Yz@3Q%%tGxsl zJC~ufelb3I&trR@>f*D*vgkC|1;r$6uw2s4Rx>a2X zl~>AhyDfPpb)^x)&U3-s*R9fMHd8qLTeztrJlqL~j_oSahak&~Cd3{u9_whf{arY$El^iLw+Ar95v)JV&oh+HqC0#J3lsP3BUg*2(D% z7Qu&Tj8e9+s_h6FS=~$*rETKon{<;$$;n_TlVIMus*_7-HltU|?+9)FMbeM+GU1J% zG`FCBH8s()gZkz^bHANtoYlYSRDM^qa8u=6@=`*QnzT$M*Gj$Vj2~k_tn0Kea(x(` z(+=>(MVWKWI6}gTx51XddZDO{knU<54Q`o=Q(Hp#+2*2jcvND}{S8sZGXDH5YRI_p zml}mjh!|CPyhqr{=ar%x;^{!1u`r>1m{<(Q(mU#l1YZQ(NIdX#d(q%uYdw(@-KewOL3<#xTKb->i>2ZR*uu?K*2lB3oAeAll1uOZZG$>P z8{xMPuSsuHDXj@RZvMj7hb+FNNm~ZHx!kFC#4Z7O&bX2w@~JU5#l?z@HICv^FWSNp zeF3(w%Mnz+xWpyQizMHlWO7@7%fZa4V)$FQ*n9=4%jGf5%+RTG0j_<548d10$+zJvUHm&EYF!lP8@m#&$LdX$J{?tKYLu_wvIKWu6v>d zWUG+=XU>7ZSxKmWMUoD7yoAhlU9Nvo6PYD7fwF1FLazS+y*B~D>Ch8E$aNmLn&rO(xr`!y=Vcj}m zf}<4C)!_uukp=YDlN8ADk0i^hL@==X9hS^B5Y%r;5nh>dgkDoO7RnvILmSsUfxw(F zLD}ILF8<|c8***GEv)`f4JBV*KnUN1 zw%FlD?~a^9H|5K~f$OuWhrb!Mn65_OF3g47Z(oqP{Fl8ydO4hN;J_i!0p_*sgVFWl zpl*pGgv^Yk+xNF|0q*9Ui^_gLjTs>B=?#0=mx0IT128Gl4P1Y?v&Mr9SVgZB>%J7i zLOdeb#GiW@dy{~F6Qi(F^#D$@(1E1GJ+#vKE`7AN0IDYJ1$xXA9!^gJne+&l%(KJR ze42nOCy&GXmfCn-Nf`?I6yV3g+w>tncVtpi2tP9y!vmZ1u+?=g?EPa2MlQQyZGI#u z@Y(OX+nnG@Za>xgBm=nY1uTqELZd$ou$`}E-&|=68W{rEu&aR7o4IhcLwcP4qm9%k zYZ1*_r(@)14@Y)7f~v-tH| z87TN)ha8n$=xf)-+DR0*^V!G{xg78w41~k2s~}Z28qQm4z-)HE(w~oQ~WpU6obP%+Y2hdmU zIZCdnMakin^lW-0shxa|+{#{p-NRlO;XV%oGz#&so;z;vw8G_vqHsn333ZGQShI;45*I0 zL+{Vnhpv2JbozX-|EWw+~&Do+hcg=c^sB&yaK>>?yn^>T%ixXkL5N;( z64c&ogEPLyDEUMR&!@G)n~Hn1*5*02+Z;kg*YoV>gZog4x9AK;Rf2D9C{$S;gl%)~ zPGp35uTp;9W-0qAaJoUDD2>Ud8=Gu>oa3mcb4B5 z7cIw8r>!t2S_gW)TIfHX&3`Uk4eaK3(;LDo^!v|9)Qb1Sl6PKMX*mO3lP&PUE>$c# z@rXQFv5b0(=+kcjSBc}hm++@q)R6G&I+5cj}MgZEHp(KV?A?);AA z%(vTd-`0naLi+^lDZPtDUk%xWIvtBsI|UZ^9A{Z1?ayQJ9>-XBLINA8ki|nx9&3Z& zLRCLUOd4`I^E5^jMv7Wk-Wg%>{_IOOr==hNZanI`(;`6>D!|0s5?JA;q3-=PUp z+Z1eJQJIYM{zF4YZ)s z22w54NR#;=^65-BT{X>?tvb7q=RtX}F&49+y6+Ku`0ph&6iTzG3NaSPXPB0Rx}kN` zEL2-Q4fnV*QemM*3Zxpy$Gu%le{}`hvMih3?7mCvXIvvU9C-iX*&=fFC837=cYF8G zCf?5xj0^Tpz--GOFroh`By9KsM~Mi?tW|~2%|_t!Zy(V+OKIhh9NeCpM~vbp(p9x5 zNliv9`VHN}(yhIGp2r@~A5+3B%O;`RwyA7`*do@}>Bq_(8rjmTr`Y0zP`2k$KfJT^ z#}Ge$_@^tF>Darmy(^VW zqh?OOs82~27=BHG4ew;Za`s5@@d@JsUQx~^U6yoDo^6p9GR?we&uELHEeV3HMf-%Z zbHjvJwJxB`YW@u7uMC&u0V0&tAZM%)SmhU1c(R?b{A=t<+kf1iuA`)ExQX0&@fWw{ z4d8g6LOjrGLiNIL()m4xFgkH2E-F;Oq3xP@L*4*wwRr~W_94*swXnFUW^OUMSKs2* z(L!LMkEqxI-dT0ii2XAh%UV;_*i5lwpdO=!X^E!Tv3NVY;xh*K*9d9#>p7&n_AW8L z6-L>mZ&WUDFYHY=LFc4?gxKjGnyZ>DWf`>#LJU(=uRqsEOrnI>97;)4|;M66JI=Nz@-J zi_`{N3lS$*3y;ptup^wW9sD~DpM)yh)1NPe^60URGrHU64Q!{2PAbNdvjwNe(jV6IQ?OTVG|f(m?Hl!Mh`=b4vi zB+DP`%&h)Vuu*Q~ogK}v^rtqAl2e60xoYq)K?2&tKG0U1PCDhUoQ3b%U+k*(D;B7i zfZKW4=o*D=s493u?+-{qTYor}UCKLD!^Yt;sqyGAH;sNDu%MRdbLi1x8H<30Pgt;R zB{Q`7MDs43pcWcwG(D@G@9##TjcYbq9gN0DcTeNegZ21idZmwL+>Xi(_&w9e46nJioR?BU;i+yyZsk}&TNDc$D{}duaf%?%3ykFH|3@} z(9CExIQHlODYl+OSKYl$AFq<5N;M;BkyI!(xMEEz%vW&b9xh~2m?P%j(ZN^ypTj~NHmhGE#!Rh$p4Bo{6xPK!>{ zk6~Rbw6lV_hn!_U4pnlxH=c3(S2Ysppi_8nZ8$31D&utrK1Q%%F_bSehQ#>}NIq=9 zpus#$Zb=7K*<>*0*={21n8iEO$rd%2$5}*FPQ#bCc_!sTceI%&$&40c;OPcQ{L-HY zo*x#%QqdW()SvGqy=^A)5`4}5xf=E3Ewt{^j$}hq9t;+`!rIeDFqnUt?Uc%7Jrnn_ zRvBI$>*!72Sy@55eH!dprVZx?q#^XN9DJHbLEJd6n^Ig* zfc@c)I9Owj)7Gj|trk%_!rhQ|O;2UlV!Ti4lyHvu+a6;-^Ea{f7)XF_65%>|4)*a=0v{`Y_Q%h>+U8b)8t9?xG`pGl z4*VLX{O_*e|Fi3J0>uBL>+iGvpZC9Z?aA~1FW$dks(>)T|DV?v3Mlt~aD7<~cSta? zhra0xq3rP~@>Z*peE(8TQckSHRl5Rk0?#`7aPc}BwC^RxpE}9=bwhAs_XsSWFd5fK zHo@(AjcQ5!K3`Vv=g8Of{VW)!ALnmX@W_7#}mWdicIT#wY z41c%t*^%F0$oXD*dTUHEfkfV|5_Fz~i9CWe@rPho<|@>`;WKmt6(rzj0|^|KM|FPw zYfSD`JQ7kv+YZ)KrqxBa{qUkM^LA3h&7rh*^C_65{0c(H48c*SD(u|A^R)URvGVY3 zvNCxD&-=MUT$ePEO5Oo}>&gRiMX`~#J~>DC-u;i}Hk^Y8#&@A)xg=^x7lQlC5~yf9 z22&;~U>QF<(7ACWiYyhP!!J`*GH}Ju%n<%t6a@WZJNPp`-+Rq(qRkg&;U`&%Hq)QN zBy|yN-SPDJeCd`LePaJFk^xcdYZIhsAf9+{N)4#TaFPn{S*nGUQO=C zCKGKQBClWlhkWx7GvG{Y2w1A@CgUK3Y-{76 zm5pW4{^vZnJ=g~hbCbbJ?<9SQ}SKH`LmoHhPky#PDzjzWbNQH(LF zC4%-2a#QIeaWV+RbN{Wvych?xuUL#%p6TMwv%7JPvmS0aVUFj1E2Hm-_hfzf8&c9B zMGv#*kb6Q0XVnUD@x{-i^6o8CXZ(sRoMuS}E{vs{a?R+s*kRJ0_ld-}YtlPsZQp5ykBAhBQBP-iD*b*sHS)Yow!MBp%!;oOY5x3i?N&GmSa_7zYANC0HgOd1*20q= zp5W+u5+qL_gtzAT#O84^nLo3bC@$;;uZ_pS@zrhU>F4K?+`8!1<`(L%5XR5@CsSeZ z7HZU-OWMl65^Da9bnx?W`}H>g^_>8Uvn*kz**rLUeKb5xxk{Wu&Xa8qs)>U_G0|1N zLDF^Kkb6r%lWXqz#NpEz+RxVu7N6z&yp|C}qc4_xbw5gy{R)WUK}me=FNKT5Bv3WM z6{=2Hga3YG_z^Q7XLgw4Z~2+H^70Q7bhVPSzT`Qy3J-{9_71Yn`!aFqo`PP%`q)sc zi+|lVKQZ*C=SK_ok>KwU zWXLmzoJ$Rdhu2?2x_t`dwY5XYlXO@dbrz;;)$t7SII6QLgE|G3(q@b2)Zc=@&&Q$Y z^En#VPl!NUGXp%b!3n*_Iq+`KvAE;32AT|u<74Mee4eo#r9LF!$`|KQ{o)x6i9CrN zu7<4Q`zUrL+n7CnIU6NxSEKu+Kz!Kp88^6B;j))s@%_7Y9NM3SehbvupvpboC0U1e z|GSB21t!eBQJv}Sc4xsMv8Z(}3V;1Ri@Rqn#TUEhq0GX0ICot-%J-&VS86n(nj)UM zq=VOXs-Q=~5I*<+fNnB8XL9oZ95PeHVHahr$*YDDA$K5Gy%w}BYw^Bw9Ud@yhF&C! zITZV`>(j?G?|emEFDLB5a9(xoN*r9=Xw7vfi4aUhcmxq=3{Ovz1Z+MEf0|wYKQW-5O zC!)~pDQZR?L*IfM=%2Zjc^YZ55s${RAT0^3x;YUKA5y_g{WCZ?G6+3e_&JH@_vn1& z8|ojHWY?z{qVh3mj1?+|GgfY>}AfpRz;zFtRK(w-+>uh zpTM^L5@;ZJ58qtAVnn0{%YAeW2ikkUBd;EAd=Icl5Uj?YNfFrc^gAdfR>N}puMp)R&U=0SKxnTN9+ugRHjB2PE8WY(zaC-pA7pAG zQfzp#A5LzL!=?}`9DmOUrKe56VZyWYEiYodT^cS@%EKR{P0;_j39@-Y{P*HFe0)_0 zrf&M^vF$zFbsvcv*5|;ch*rF=HjLXG2T&lCMbo1ic>W!KPc+s<@pxHGPnd_hU$4OK z2eZ-P-$J~epMalsB%$HVXiV&Q09n_|;b&7J1lRxMy??Se>LXvG7kQ0wx_{7O(J=0) zn1`d^TjPav7v$3I*`W~@ENCiWo|Zhfq%{tWd0%glr4)*KsAIz7f3VfM4`n`&WB-1c zv%1(Bc;@H|tjXSne#@S~=lP#`ce^4=udaZ#2RcAKAP?&Ik6^x2pQHEnHe5eb8+9g) z#&^6oA+^d4A54tGA8{r)Dbt4cNcrHKZ>Laktv&iK<52C{DlGjzf{9#tgc=XuW7em2 z?AqJ)OibbuOYYwUU)%V@?#gzaM=41iPtT%SHKp{4!32EOq=X}LobanK2VJ!8qHu9C znw=VhmO;GlQ|>B`{k{)9&<|@CY(phgo~O@k!medmc&^R`N7{z*`GkuYxnKz{AMc8< z9?!%Tqiaw}=>#saEX8Mui;?#O;j4^vEL^F9d)n;rhN3^J>wSWljru&JP915~L%L=> zfA1f$4%F_I!krc+e6)m-2w^KwE^=_lakg+f#iTaE{Km%_J&I`DVpO$D&s{SHK5u7H{L5*So78b5uP!QSDCxKh&v zr5?>dEnhwd@aqyRe_sp*M>}vx;ys>Qb_L4|YvKNp5Abi(E4cnG494|FgPvF(Y!*~N zNF(o)TxfULSz~XaOVM zA!#Wngs3zB(_>Y_lcHbYD;tCJOPw%sj5n@={n+JNjQch;V{dpL9@sLPl`rRKEp+N& zPgyv`pKGDP&5DpcqleB|d>*Es$$|}+cy9*3z7Hequ_|INYCq&ZlS5muZu$w_DYhTa zjNgMsPcL9|yAAzw$B>qe@4r)tQ^v6SuNgj%SchA0Y{h`5zI1BMQmWa!j?Qs!qv?C1>4Dgzw5mZ!s|@GU z&Xgo7Ic*WjURZ(S@8{#DeaGo#<#=k5UrpCv6H*CzMt5ZEQe*$ZS{5-uvrGDzp?_O_v;`#I3DU0 zOX)!1J(}~<5~jRcOK)F{py$V}rTWslP)FVYT{oP>gB`P>tKJC`f z4&vOu7SGJb5?iuQxDRrVU1i-h+{5WuWAE5|58d z!R9Ghn7UPzozLdo+LER$W>f<8$$U-K8*D)&ZzZUfFNZAotDsty3{m@e4&q-ebiWu4 z+^bZmIB^#ARBNf8zZQ(<_cX>&r}HdtZwUFG&Sy2h(BQOVc6df24P(5|puuzxc%*0t+qz<4u8bqBTjvJrt+F9$x()1oGe~{4qTpXz z2)*T#3;_?-&@|8umbk=#gVQZY%Q*}cR1RNROu_pf#IWPN6|U8rjR#E@Ld`2bC}{S-WCD*CP`|3!i}JrYexF z)x|H_qrf3J1~%JQL)NqX;8}PQ&V_YCsKZpSJ!gxS`I+cdJ`M`8Q;EJ!Yt!yOAe`RDd2^gHbcgR63(e_cQ9SXW1v z@N9y!y-PuK^Fet3;UGLJZGeqFO3<+LFTL|N5!N)9)Aq{QP`alI*4=amz3OG~qI@*= zwWm;-pATr=kLjSCUk)y{(I9X32PWJ~rH2RK(0Ma$!6x=BwcfK3l8VJJ+U_LWI(QcD zZj8p2p3b}v`WQ5(WW$QG4e%>rAACQM4I^R#;iXeOS?{02pMe*Y{y{$s%B z;0-`7;VtODU4-K&$D^Y65$yDw33@3LAatA`1R8eG8~x@mREjvn^G^NGTEUEaWAGTC zPbgXB2c^+TVBQ!3*>8Qo?ZW{`-jPQGp1Q+GYXcOiUjoNbd5iTUpgCBDzLCFJmczoLz1}5wV2R^%Yoo7itYC8@Bn{`lqAsA9;)qrQ*M=-uV z7xujtf}(pI#Jo!a)n$nw94Cebp9>&WDG`ipkHcRVO=yc$gqLM{kXai5hbDPJkX$;H z@xA=F_BL=?zZiTy%c$H?Gkqm*%R9vV;X`gO6fWYk8Iz+Rbj>ALI^_nOzHkthjO&Jd zuP%Vg>PhG`au+@vjKtS{FX;O;2l%dPgmZecps_CtMr*EzmTo5iQUd;6nGoBJ*u_@h zKif>Cvw|UKVA%_*$ zEJ5t#5d6ER4GVu=hVgHTaML^~SXaHDUTvzUFd__E&lSPI@k*E+bR4QSzJz5wmv#4? z2~Zs20?!)_;JNI1Djxov4p=wRul7ZB^IbV;i!}wsbJyrZ91Rg~k3dV>33_~&56Jwx zNrRt7!0KNNm-YMxH;u`dp{#+Q`F-zzpA}fc2)>T(!+&lv5P#r1T^hF(ERENLap^G3 z>+6E+CMnSU_#pWH`v3zks$u4rXZZ8JBpz)Xpt-{HRAE z`%!SgP89xpnF7V$_hHq{BCz3k?~ckUsMYC>{ZUoBh+Xm!3a$m=;W^=G`!yXKZC8W1#|L`YfL}vL8C3Tu zV8+i<_!O@X2mUC)&^8}{dSCF$S^gKi2_N2cMa8iHfOkKo9bGT6M@5;U&x47z(h z@FlgFz6kMy&b4;1V)AnMcsZKqgp7koi^K2&%;0{&6$r>Rg%LUXVVt-(WO5-eC`<#b zuZQUD>$~Wxk@u)a&3JhJV>0+OZilt|6Y0lmnzZ8feR?qQ9&L^gfuc?V={In0#CkX2+d`tM;YzkE|4ke*|#q<#Rc= zZqe*I9k}q<7hAsA!1oy|@$wQ&I6XWIUg+C`fSLe%kqR=4y}(cZJ+-@R3cFm2dB(3L z{85U9(*@_iOVS*4p8ZGLxW%w=^dcyDpGV0BAq-!)1mCw)p+VFN@)Csbari2&>sE!3 zHFKboXC7tB@V&K5itw$!mo8a9AA+VULpi@Uw7eq(Rkjht=Inykmd8Oc*b7`Au7XA7 zo8E{5{D5O!zEJ z9sgO)*{=@KvqnR0_ed~tFo4%5#z5lRK+swf2wjh+!rQZ>!I~uF(a<%h)xHgbihLnM zCJ`>h?uBc51|VrM17; z1qaW4J@^)-3<``84p20xX=o zkIpUG21^dlp~9+o8g>3CO-qo*>JSzDw_p^eeUzq|A`i)`O`~X+vp(+<(4b;#eWFvvf@1W%(Xx`nV7tzBNG8@C1+@mkRAGR8Tnb3alTH!R@acp-0LI zj6ZDx6K_49XElL}G-}YqZ3eVpX*t|4`3iDg;m{Is4dz@!^lvh!HxgeG^Owfd!eSoW z{9hEEcRbeL8^$vd$;e8jp-3rdxv%@v(o#lfs#MaDqEOnUtg;Ca31yW?6cK3njbj>&x`@2Qc9-;M2%niVY-e4EY77jN8>4DpvcYL$v3;a^PO1_8&~RL<4s?WPoWIa>2TTJ zf!vNJ)0<`k>fh`^KGvnQeQXOkT{6YN0d-hk(GIVR!w?)*LFuA%_gu+Jiq=r1Nkv0Q zb(kfsDHwv>wGJ@vv75X~#?br7BFa3gK?~*|rF)_0=$XcE+UR-`i`Ed=Xh_t#iynOr ziKX8P_b7VI0czSi7~d|Rq58;!4=$WVCt9x8fR zOi#3b(8I(i8q?*M+0)v76kS+s)+IW{?DqscGOSjoWgACQ%>zA3`))%Q)7H>=(XABV zWk}DG22-Zx80uK)MbpJ>C;GAsEK~2$hv_Y}V@m}ot$GOalN(_E>kj&D%qJE7bTWGR zkv@!BLjTFzlFhrZr1|hL^&LEyY$|pl;MX;?5$1_>J#9bjQ|LlZyFN4<8#SD2uQr+~ zpPE9M9fZE$D~zAd;6h9R%Jvpwpk)+&m-C^=7t-m`+AwO7xk|04#Jxlz zo6_CN$xl4Hv}GTf{oX!|vfA3sj0<|w`##fYR__hucFGhv!Lv}UG!4NyGpKn|BuVA5 zG$v-N*iXwa#%Kt)ZHcBnKbF%RSE8TA1@t#z8M$?tOyBlvaLm*>@Eh?EpOtN>TgymN znV3cuOBHC_n@;3nHGvjP)}T|>Dx~(bEA5;di;n{@iaSdVI`?lillhfPkDR0F=#yah zXE`DD!W0} zmd0j$q?+g`w72&`+q;_->N6QJm8qn#DIDrz-gCkA1d2|qLG6e4sP&qJF(QX`Ai5{# zI8G6Hm}Mfj--G>^>tpQ?ZJ4I~!Tf#Rba_fSow<7)dg_s={2qL_wWyiPezixF-r&nE<1EY@fOKbQ2R$;}t}>bc1n zaM zOW#q+w&QU5@&l%=qv4}75^c>^i1&1c+J^yretB>c>8J839Izdn<^#}GIb-X@EW z-B3DsHx@p0p#6UTk`u=MPkw{GqYBBk(tIkeDqzr{iZIQE|#Hv~B5vrL(#qws;mU zj7h?!r(dvK@g`2Ei5&HZ#TabTho@Y8N%M9_8r zg@dru!WaEz1|Uk_k+-Ra!F<>@^xi&)gHIf#@`^gT>bwZ82Q1L5gYY5`QxIo_81fYJ z;8oUFa8mOVdHge&7ji_B_jP|7K_^DJ( zjh9=<#{C^_od1{>Zv&iPO~tAUU2t;6RQSe+LT~Lf(Q_G!iJi1K>X8X0-LXLcf4PyihwJNvC*sqT9t=rc+qbvHnBmcsKNL;;TrvSrW1nas9=`=CE;~% z6&&OfaCnyjk0hbVy!K|~ujIq`vN?8jE4k_Q$HSdAn7w!{Ub_q5kI49DcG&|KbI`T= zU3mX|(a#~cEMwH)klEqy6xDX022(#I=Efsx^d20{7|n|-zS8sAnG_gqg9U=2wlrBV zFwKXt{%05LZn8&S&MN5cY@xI36tSF*ur}r%Y$tc&Sppqa*1R#|EDnK_ zTOQ7YOo7J|SB!rUhLEtO@K;PG!})ixaiT3Q>o38A)sqluGl@fcT65wOIo>ox52<6i zL1WWXGFoMbl4gHe_2C(vS{IDrIjgbK&I!j$N3+U>RXi~+h&xOij66FJeDr@w$`5mC zf7u^;rT&**KiLc2{sB-u9*Yr~W3XjjG{&`T$F%+Pp;bQ~ua+u9d9U!Z#biL^{0W#B z7eRCNXRKNB2EG>kp>*UiHYwF3;{6p`;ZR4D^Gs0CbPV=Wqp?8fAJdHB# ziif_~d}knL%rJ+BH8iKFGK2b>UEgW5a6EW&JP&}|4*Ml9zZ)_NDC3fX9Jtu6x z@60pT_kcr*9~Q*zGn?r;4m&-1qWiX~sDCsMAwI4!S+^8sBLx}JN*(#4JO1HJ1zu#A z$NB~{U@Fcp@j6kk75>igi`8%?Hw+6pcv2Lt6q!eH=X+~`M6Ir9OBzl>kf8c`YvJJB zjosqMLvck{E*Y3z!x8XsuP;vowP=HBM;V6=+;d*CKH4>-5UIxB%U=>`XF|xxO?px1#wKrOy7>E z-f#d>K~4BPKO1jH-Gs~zNw6uD`K7ZlzO)^uE4wvuuy-0tMJ8osh$?#aT8YB69@w-` zo!g={u-PjCCsyWDiOo#BGq0vsQ9`@1U_|zjhrQsU#<&#W#hob3nZE^UBM(5YeKD#J zk3zyah^Yn;vBwv|)rl|(+@f|`hc)<{?2-aQ-mLXY}ON=M#AK=SBScW-$G}n=WPmGpS|xHHzALh45w=##OgCWP*5!|`c<3+GcJNE`t+p#Vy*FP%Q`abno8yF zDmcHhBR_v;!tsx6IAy|9$UR)d3+7Db4t4(cHP;_E8UpcF-Uk&CCfFHbfv!<|P`x?> zPtJ5>m#;>c{=x~(^S{xYY)f{sEaKe0k!-s=gWRjLDfvJe_4qrUa_=0bMK0ILY?TiE zXS#-Zresm|dR^&jjjZGr-BYqNl7srB+r(chs30$xJW@O;(xQyKM(9C*u_T_$GjO-> zVp6lKB3q%=eKeYf@$RWm`QDC|r6-^_r4Ch&mb`WG3+#S>1OJ}eveE4x@VPw`uS*wV z&8Dt&>os~_go|W`wnVeq0yxd<0vC}h`6hb7 z3ckeCtZ|c}E;vgMcE+N4(_t$3Swg$j@6*-Z50P7@AiX&zYC_j7Sf=V7sw{E8(ywzr2SG2eMK?J zR|(DVKX?BAVlQV_tfAaLL8RbxoE&@~k&@w4ivALYG#MSFKG#NWCqBOkv&4(^JOG1S28+AA+?qtr)tkk z8u~t!N&`#K+Ug4T0SC~>t(rESdO`IgnyFh;4_H6Yz{aG$^T@sRJhlB81sD5oGXs=r!$elj@q=nvVq=TOEmWz?4h(fvd7sie<+!o*m*pe4uqHkz=dtO>h5ennQ( zP2p}RgTJS25nWkPEZ2N?A=efLIbTA}qxNzCo82*#8luOUJ@vi=x*f7G6-FqZ*yS9=fKd+L*`pHUW ziwvbc@6@G~wBFKzd+|K8J%bf(b9lS_b3XUGo-?{W;K0yQR-RYQ-*TRE)yB z!4Ljz*h#W(QkEun)t0XGRgxa<|HeOVbdVMwtzfl}`+3LFbbj~4g=hWSz-PbO^Y_Px z*v}!6y?4%Feg8P=_(Feat$&d8WXvd5u+`&zKFqW4s_?LXTHLq)DBgXbk~@WWlHw0u z;L76(lHS2W$yGT{YTvVik9-`>7S;>+)GGsi5j>3Fe^uja`yzO3%{Jb2DT|%gRPoaL z*STYOD{ETK=Eg)1e&Ml}H}=+({w~*(PF8l7l1k;IMV$oatnewPeQIX=tHY#QK1z~@ z`0S3{^WeI{!&%j)l(*R);VqpLIO&I+bhCdqY0ps&Y1GtJy#9_ScYm~wM;tF<_?_W5 z$C7xmm9EtOOHTT_)?C_Nc!)pTOceWw8{cgjDs9fUl$68VC9RTOoVQANU1qG{_oA!$ zW9cgX6`sY{UT1Q}?hRbmxRrkF$N*Adb_u<7aCI33f{wANm!?7q1oYukF5U?(4@g zGGZN!i(*H`G(O^A&%G)OS!+-@zwUN|U**4J7uh84b^bX&k5-m?@9!eLRM(bf)i0E8 z1lmhZpO;Br?2LF{SbskH*@R8Kt@z-c^*lUe6;BYe_na3SS?+Iy@SKd4X5BQ9+E@0L zUO9G^EIOD;{X-{8pL!Zg-`)C3Ltl@S>>L~6vssqs=1=78o{;Xjn@g?ZJ*1x|3ew<^ zp;Eh#ige}38jjW8#a%~u3*MatFS)_2dSNKP*wtC8Jkdkivf5O#j_=E9slSo>TUW5` z{@_+aCyw`NK-fHM)+7_Y>e7=9YftkTx3her-&>x!#*mlWjpoyC@;p{=DW9lY$km7K z*)H`q7fQFdO6xq^z5K>2t-kZLcQ?7u0UsW0yqERQByzBi75AvIXWRFF{9mi+^ShVJ zyL%ny((!Zn&Yu~g@5F`GL`HGJ*1ml7qZTWhX|nR$2^{-(G+%z9FFC1;*KN|4mS_Lq zRLw{1?f;l(JY37;X$2d8-^|%Va`?=UmpprCA*Zi?&b6%#{K)eI!?P!^P*Y;BxytP5 z7r|GX&+|3mH=5?CC+YT9lUmc3OV@KO=N=EXRHTLN>-6Z2g^&f8~RIuJ(o+jN4rT8-@~Q2q#Ev+w~x21s^gHjCtNWq zlCMuqU~8**9+_~QpSd05cDadCmWqy4{BxspBX@|D7QayHw%SR$?%GXC`>H1GFHn`Z zTt!lO@{*SZDNE?}mVSw@x5o;q=`<$r4xm2lF7|L z?lHuTE$w#mlT|}mqwfH&d_0JI+iFYqHpokAdxl8?|4o(l9t)8ILQ5s}#)G`wem`$| zQOe(hPwj%C7t>!8Ry&{1Ys|yA^U-S@5n9h*SAXJVRXUPyX)HRX?eRQD?hp^pyunG^PV&rI+5E(~p8q&Di)V|Bv~uJlj_gy< zHO1vzW%inTEbJ=%`l2B9uN*Bk4W1-rdCrtJTDeQ13sy+u8+J-&O^!S})0hi%M)I6* znQW#L!z$V(tQ70QvKxFjN^39Ea7h|}d6cxq%UhCDG?Ic1XbI&b2B0NoS zvqL%7C5bKm8!3f`t4bpaCQIGYy!cVjAEYSzv-wX$?jJLlmwM>%Oj{pLe7>9i^RVZ< zEqc;{Bb_CUa$RZ9u9+M$C5C%EmiV0T(tB>QEf;x`3pUJ8V!Z<9enImfpS)s9rO-@zwoJ~%g5wU{rSh(`ij|+Lk zu|+)F%$-x^=J1EHh8&*)_PJ{$8IGDH#Vna5xw{xhhlZI+Z5CZ5la5zJXWnVPGCPg$ z&w9t5GGwH|y*f*y19YXgOSGiL!_B43o1J+2rwyFbSD!5(9p*WmQ#sZkMw|osNfTR1 z8oS+EN|{r^mP#i$zVZq`Y|xS{&!|fW7k881&mAiD-l8sQZ)~Y_m7xELhVNEhej~Dty%oDEPa){65z2w-m2xKqNV#sCq(x6ecZ+Kowtjhv`9X#FHRde7HHo|5sKxv&YX?Uz^JAsj z#T>1%oBJjDaH;nu_PQU(r{9XXTfQ3~F#_@RJ6zn!3+E#u){+aMzHeopAdBmAc zY(0J=N-IZWp zgL%c);oN7m8|(eof%~|$iWzEmzWt~f3R9oLo3**aMi*{-x{mkC?BlD;U*YZoG4~4m zfYI3s9IfALkkKk#JJAdW3dXSF4VdD`IzqMx~kq&L<; z+B#{x^r*8uZ#7lshAd6Kui}J3zoKyRY8d)C_vW61hVlN%!#SpA0e1-TWL?oGr{AlH zhd;Q))bR^npRFim_vTFmU;$cMN_oRZB`NoLSLwcSByXz<;4)6< zNnc0uia8QTf3o4@Pc}jInwXEM2f$ILgsa|M;h>tQJmR`Pc4aTY9lPCPMl}L1cNSt~ z*(6B%wwM>|iLN)-;(6m-KCiroFZ!+G`-XwoeS8bvmW1HwniwcV`k_?s4Cbm>bKvL6 zY&l{U56`;B+n&DU*4~X=S~85e!&DBIv*)mXb;2`s8@2A`81--!Pu&n8_?;{HrO@#s zT>itW!*bl7cMwtS(Qx%ViGa@QFx_t>o-SO2x_4c<@nRRjAY!)libI`Z1oDg@q3Y*7 ztoD3|bu~Bee)m8Q>}<&idh>YIBq#2F*NJUqVp+?|0af$2;`*BPtz z7|O~sXY;aA@oW)v2YC@4xwWz#mzI6xtR0$?NwK=rW)jFjTReHtvDM=4SB2GHRd8Fb z%;jpzQo=P^$v^rZ%NA&{=N|)}YB7Mrg+6s+dkDO9x1sZEYp6A*qpE%pJZi7=rSA8* z-m!uGXVs(cK^eA+`iM^B7V+>%doJmzD<=z;hecPRv9_w&)0u z?at}%_1V+3pXBdsA<0?UORtW{2;Ey5rR|1zzG$U*KSE1=7KzRAJ8)%c7^dtC!@MS6 zn3-ln<>5*6@)*ipBRg?OiV<&^G6U<>=VA2g9mw(O!k+upn71ZjSB(K5kd`q;eue?WCz}a*TI`W|IK*qe+k_>9e`4=Ehv2*fG*1e@%gF(!mP~j zrE9F{2HJtzg(Y}DX$n^MiJ@_Ag5h*T2fU#hRQij4nRHG3`!^B=!z@w0EeF@89Y=NL z9ei7Vj@o+KVdmPI)KIwx0~aj8;H+VyD=G~hTGPJu{)>LH1Z;s~kg^T#O`7DGqIAf9GZnPZGKUUvaXqKO^J66Do{E6AOod?P<-^&h2=C5@;rAxN!0jf> zYwr{xrtcASy0HZXL6O+%pM*b)3m`jTE_Zh_=1+B<*=~nD%dJ|#E0^8G=%pKx=WK>< z!f*1yYcaM}?m|m)7)+j?r4u7hkVEeldVl@A;EwfTMS~B>*g2WsIf$Ie@2!0DNhl7t z?ZhB;Wn}pDL$7UCc>d!LMMR9_#P3V^PVQl@J)p=6D}G?1!z1Cv4ZxdrOPo3Fi6@3r zp)qeFlA;OvvVx;CUWsR?kLB=hmdNP30;UhwW9t$FL@TOb$J1u|K4%_&kKBS+c|%}d z=>eSqVj~^54?KPqlpD5V<3m66PtL;9C4QLvZxjAq4#Ymu4Ryr70EgT=@>I=2B^Jeh=LS6ndoz5@zZMd8`nolxzy4;xJ8;lk)@jbNN?qNkgO$Z<}z0W^Sv(pYi zzBvpnPt~DutOK6(oI|twJfN)Hjdbcj1vQ3M>zU?xpX6s-`Ckw%NXr}33+i0QTU)max!Rout@CTjy z{C&MA-8@-LLBDTNPWgZI`IG|8`rM$}_1CfVZ#(IDzoy94)m&9x%&oTW{BqP&O7d$X z%~}=#V`~qwHkb9 z*fV^O>CbIR%edE|Mr^t{meZaTz z-r?HVKJ3@%#i0WQQ^9KhoWthgJT4%A{bdv-j)nUPKRie*!H6!w2)b8FZID5AgX0Q^1J9fiVYvGC>) zI6S^f$6^!d>V}z^5%htq)SaOd76YTRp7?A!pVAda(eBL&G_KnV`qT+fOn6R-zvXek zVgx4X_l3%{Xj&R_h2EY>r4gr{1YbHC3I@s;cD)leB`V`;|9)sO(nX-xLPY)=gn%tt zU>AaRLpbjC5wp|nAL&cyJG5f{8`@tl{Qa6QNY8OPSNF5$W7x||MVBbj>JxFFH>8x` z3%9P75fAQw-b4Ha`)xQToSBU8pT1DBt2qwH=0fgZKK&h*D|j%%0Lq~>+i#axpUE`k ziyFR8R!3!F8%4h@qK{|pk@fg&k`L?vJB1>u);Ga{*Jb$JLmyjgrok?m(0A!yat>UeHygQ z3C2ys@woK|nO4n3R`CV8rspZEF?jm(Q)I`G{nBZ4SH{5JfLf5*!xUKb?))tMy2iG7ty_UxSbzcnZTS=-V znH25PLSt9SiM7%R*#~=JLqq}n>wk~D5|Uy1`wS&-7*9KE`p`7{80d{Xg<|_|IF`2n z=L46(+1XbxJKqY1+!+cmx=uQ~8?gEFEMD<+E}QN;MTrT6DD71zYzPr^B;nt_87_-X zF1^s#S{@Hm4d76B5qCPs^Zr9-yhO|@6`v19#~X_HblVWuMCRLfU?EO)zKu_+n=vz8 zk+%lSMQo1nJJheBHEuU?xu^xsV*inSsEf-#S0c#xHJZAL{NBfCZjutjx%Vg^?zWG! z3XZZtl0WZr2;u4J(^$5ouVmz)FCBiOAvtGSNMET~lDet?C6dL=4eQzjA zw4jY0uD#{-1Q}`Qtw;ReWf=#&XyjjJ_t|_=vhY{POBcL~d4v$gg6-IH!2}+?t%NiBR&c`f1YWk z266e5*{uCb=$Th_Z{nFw&$xcz zV?HDD&3AkCl>EAOl=AhaNawZC>`Gps@Qq)_ z2+b;~ggr}3xPICxjvW)lodQm?oBmn;vMG+ud@Q8Lo5o66#sj4qz0>@qV+N-@i{$7V zrjlH$mgHEbA^jV+j8jD}(O{uIM+V>Io9l0Ig2HK5v^&9zckJQKcE>q;r6b#^+i>Q@ zS?nM*?*mK!a>LzwJUH2bcUgIJ(@{g#%g^P9C5L#~svth)RmxpYzvH7ByV-f|GB)b5 zT-^Idv2?wV_YO|sa`F9dXgkF+f700TT_EpEd&~ws9`LE17x?+{Dn9qKn0?n=<40;5 z(k(GhS%286p(K%9mqN9|$Ku20JPLg!2-tgpq-}yqBC9C~% z;HP>6xOd-R$>GjYY4r+QX-~PAl)czRiupWCn!L@4Bb)m13nvqPZtBfDR|(dUxPQIj zk&??%BPp>$Nvheff;W7!<1p*Ze9BIpPrvWNdzSU&h;czI``3xpXKdt8wQue%`nF}oy?2}c?k2I_KZC(b~#bvPZq)_fR{1sn4{ed@@ z-Q=@t`bd8^XiGZdQ@EG&re5Hdo7eei$a!AuI7|B4CP}rC{iTh=)=H%Z zZ6&Rt1EgGZ?*3&*u5eyl2~uG>t7VvdBL|CA zm$1RDV4g86iSv?A(%y*IG;i}L>}si|W{W7gyDf`y8ZN_8z5)h)lCjEM8KiKN)?KZq z!TPRjKRuOW@{-v|K?5HKO~u5k=1|&IONGlX(5vY(D81yx2E*6$uP^I(;-4bE>$09_ zz8=ow;@*+G6yT{ zccMAr7Piu9y!ibHUjBW!yw(p+cmIQSc2~~m-htctN!(|O3%f1%N7;BkM2g|=O4n0<4w)$M8i!=X^k`CN^kI<7E4fvaHPaG5+(2f=cIPg)OO^5ep(OS#0>IF1M zF@=`#6I!jW!ya3hCqB1guR$BxPJbq!ZadCyc`8!mlysh<+Edya(gzdA844cf6xerD z$EiMZ5I5f$kN&N|1P3?#yEXw4*uyJsX7Ow1QZ^frAU=~wo>5cIzRAk`txA>krg!I@ zyS|iP^^{(D$B?s{H~WZRpmXXWZ2AG*HOWQ3>21M6D5IE_ zFX;6FH%{pi$L_gNT&U2C6(%&7V;Y!wP`R4zl1E? zI_Q7yf#Y$vY4_<%l)w5l_5IsMZ!fmejmfv58KlZ`4sY?~PdObGJjI_&#G26!h3Be^ zC<{9Z7v0g=WR-@2mkaPt=K+fPmBRmiJwhiu!KQ1*toC<2CwD%~2X81zf3;+!juinE zf8i2M8TOl|g{q>%hu%J7k z+U`iNaKNqllla|m1j?J|3orE&7W-V3EGeFW#Wf*Tsg4Kk_ zW&>YFQ@ZFNws-L+jkO`P=4u=*^$Dk;_qrg;{17h7tKsOij(Bq~41*4-va)>=-TEWE zE<;|@vY$DEy^umf?<7!&RS?O)SwpcYz9J84h3hMIdAzC6T9#{IV(}1c8GHwOHD+Pt zQ&sxsu6k6j9II?@qZnX z&|sM&xE9AD)4he3D&40u(ejw7^Of8`OvjqCx!BP@5w&(c{|8@TQjbx1bNeq%X;emM z!GOAO+n;TGMzgt5CH3g6hRK@;3Qt%Tsjcv$@VkAWILI9G6@wAtE&8aRxlx?YSu%}J zA<54d2bRjD-%V2EHN4ogqD*f|&LA*bcdjAl-ytx)IzAWxkU*6G$ z$x(FE@;-eX>j7Qw40R-{k<_Vz3xzuA|26cFRoo;-4 zN|*16S;>+sRQD?r0q&yn)8{1KyXoLuStG8A|9LRx6AVd_FV$*!_b8Szw+ly9I z-p0$HX6P|Uli$aW!h(}05wi3ImTKkDRLAM$`Thn4w{#Rt!&d6)9!rheSHbA7Gjpv6 z#y3o$43}f%S0~XWkp;{9G?i*uMIWWu=WW(IjM!2@oA`==1VC7I=DN( zFYM)YpxZbLqbs@~F{uJ7){W%%I)>W1YhvR1>*Q9qpOp13(x08lBtLm4Eo)X3Jcem_ zUSo{+waQr2t(jUpsk%}pXe9zs&GMP=7X0RVi zvQ9yDE@OyJnaHe;hx?oEI29_37wetydEj1Tc$|ctqc+~Zio=8crC8#YM^C$5qvsl* z=zMh%1(jvc*;mD+vZ)*|n?EB^M<46%&%z|1;mEdrOjVZx=*)auymr}bcEfQO4vP$2 zq`58j9gm>-+OMgt#h11Q3kKEYD9o^mz}fB7S*@feTjs>^oY}Hi9`sLi$~IDRqYpxj zeQ_!}8li0|kYt7+eTn$IJmgqwUkZK?7{FzJk73|28>E&6A=t1685isD-L(=OG*z*6 z-V;hMX`yZ@+W622yBuRFckOO!pH>XjsCYCAjYl&;L0XdWk28Ngo@5a2M z1;4%V-z_^l)7yvvv$i5IKO5_tSy!q3h0^?uaPC(J7pN9a%#^A@^Y;-rWs{&4RVaI1~ox z30-!}ehllR$>VOw@s?Bd@LX7hga1yTV#$6S7`hfdt3ok+a2b;KhT+wxWw4$lkJU;6 zaQZbH6K_pH=bCJhdzvY<%_t1dJ_xC<6k9x^k!Tf$?N@|uCURAJ#kr`=EJw}qBG~(^ zL;vi42vwYfr@89rs5}HK%YKmMl2hoOz6a&W;kZBD5-}fjpqHox|5*>o66thT`8w_C zSO`UHf4FUq1#)zewBsuUbuOi~3ON{|Rfd9$BGiWE!TNgNU+gCvaBiUQvQ*$?PYurdVZ%47jdmU347)jqFQ*gTU0u5WKvaW+{?Hm8qZYtI6lY+eidOIH!Fv=aFji@~E0LVm(r z%=sMy|8*{Ks0zd11RESSK7=XmN6^JJ6qf~$Uh#gZ@UW|iv+_)F_cmqQ@TqJbk^$e* zm3aNB9KPeS5t;J{m7~MZ?N$!bOcUU%TL2pa8P2VE_mBmAh~ zPuRO*mdIi}StQT(D_^1d_+juiUu?M4jekY1LCp(m{K)Zx>X<+%_liWwe`#p=S4K76h3-f) zXe_zKq0U+S@L&yxh;zinhsP-NHX-d;Jw5e3PiK7!$W=FodM_M^{qZvV;6)m5vi`&h zXMXUUf&b`><2pLjW`-X_qewxen(`Im>5J=Sq`5wjzT3prfAiiEL z#~zPMc%31*Yoe1f#%2m<@2N+y#&E71HRH6F9-jo75nDFBefNsHBTb{m?aJ-R^vezSw1Yf z1=3oFpviGPf)DTF^A?Hx`)N9#xc!#wwndAKVIs`>hLX3!3rbMmjdhX1nA~Fv2Hx9+ z)2~bDulWv&^Dv>R>D_qGjx7FAU(KsZ2V&Ea0{q@B!*iEtiZjd%oZNB&uJ3PA^G74h zubqdeiazit))%uVGu*lUj}q43hTWH5e0q!=rY5(MyQ=t&d&bc{@!8%mYQYN^(QmS6 z88QkNVeu+8p%stAh6VTFTJ)Og-e=;;4IvXpiu3d|9y8SeeWu;T{FWeg{L_z>^@sANgu^`L zYCioOuno0n*5V|oG#L$N(Z6SZ zIQ(r+!R?A*E;L`E?=5wtG<62{*-yq)na9{)6htN$6;XCraJMfXqJBwfWOJ*D)aT~X zrZ72F9Ogy4~z@o&XJ zY%=|&_KfH3KWrvj zkDAAe0=jXD`Zv58FV=Ob8n;zFLHeq$d`xo;r*tS`Z_%H)ivzjb@f&yD`<#91{_(d9 zLA*nFpl5&G!k?{|u$NUF=Xy?KgS2&A^K~gF9CBk@TT@$APask^3(cDKJSY4WAJnPf%YBD)rm_d;kT=_Any~Dx zf&8k!A&*MB46W3o7{RqTIcFx12^Jk^3(;#SVp~m@#BH zXBa%gq*bzPzHcl4F8K*p6U&-56my2xmAg)+5l$XToi0%l{koL0 zh~Ss_yY&oSO_zuKg_=>f-v?~ zEUbJE!{?i-6z0}Zvg~%6lTQzmw5J(KFJtRCIjI9*$R_yM?j$-omk= zgcr4BbMm2W95!EcNZH&#x>+WSih}Sqc^{65{rXaIHTLX({6CJ)Je;fVYs2O_MIw|? zR7weFKkFzd%_SPXii$L8QfZ*Mj42r+v!o)DCJ99%DIy6KDiP73Nb}sg-{1ePF8OfI z-fKP2ePihn&iCT(%*=C6q_brrxxdMRu}NfR!5k$=xr zQeNdp6AW!=&~!t3w^E8aEe%O>u`I3US+ym<`ti)=9T-M3TG=g7L&GWx%=``GvTk@y zP@oUbo}hhk0k)gJ!0-5b2tTkBe}A;1SK|j#)>jbw+T7#a3y~Ct4WYL#gQ&h%mX0q{ zpk9t|vE=Pq>M69Q^Up-OQkF^uSJ%;+yc8-9%*D$;r5La_9om69)KS-qsq7^G0 zc~01Ymi-+=Mb^E^xYU^B!i{KFc7IY+Fd*NQMn0#gLZQA3xq5N#u*OfmVdc_NHq_M!=Zc1mqje?WFa&9VMz^V`?$^F`BEpQY?RIFoE7kO@icx1 zItY*8gLwSyFxxXo7arp;Fx9*1Y{Sbi_SV${YIBw$_lpnCq}VXAe;DiM)x_!pkFmQ3 zzu9@jvx)YFERr8GpPz|k-Qs@M>&HH3Qysu&9IJ$WuX{K>Bp1u`UofefX!hMyWSMto zGOu~}*v`rK*^5tkY|n*M&Tcr)`?h+_$;FYaALPO0-*TS6CXKo6tzgCt zXW81nH`vQNN7%_Jc}&gfH;YNnV)AmQSjc#HI9NuqQ_3G$`1(;;q-udg-FGY~>NJLV zKf(UH*P!HY3F}F^_~RpiyV|XMejUkddx9Cx_%Wv@Ddt>R%IstHaUe#Hwl=xqK%ZBv zHd7Ykk8Hs8$_?;2*B`DnKiOYbUuK=<&rXl(%_KEa;r5Pm^QCgwlTd$b{27b`rIxhxd!nQHN_b0oa zTEje?<5=;KuehUBj{C9|=zrhkc}=1!>iF=dh1C*Msh-*i{JU!p5YZjuy#ZW|+dUe^*mXTGB6mz&5e zq@8-{ahB`II(j(tGyUPaPqlF=JaZ2E60(T~(__&-qDZ98kz(tW0*d(@OiN{l(y-70 z)SG97lSJ1_T~S9mnM%$kQR9vLi?cq7}Ime9Yhg^<$ zv(Zz0>~vW)TKt0+Z8Q?STsmn{Wei<99!B=Zr;};G1NvY;pE52iqfPQ9{5-dqyoM=K zTJ0~2eHll4)I}Q55KA3JcI0((4F&yogVap?X!ovz#J1j~&@&m-_&taQ`79Qbhffl7 z`@JFA`3B;)__wtEi!bd=2^2l7cZ>d#BgJ7Mkua_>Z-L_H+ zv^hXG260w?@E%d^lfQVy$3k4w6)dXy&lk(uWJQ<$@>U1Lc5y&jrnq1xXSH^`g{<#c z?yfv0ZaU>ICM-}E4JVuwV<&rvBNc~vdh z=}M{^9cmmzU(Lr-nT#4m7d5~!R$8p@xkqK^Cri4aepC2XxdP}fg<)4VHR zaAtcW{_7nhdS#CmU-@W>YrA*Q;A{Vp?Q!k^n6#eERNcv|bP>&xctGBBPS8fT^PGA1 zpQyfAN;LY?&d2{({H*Io_Unzz?RTVRDka*4N0WF;zMs0rw z)72a6#a+J=#f2%dqGUZRJEiFS>6Mvv{;(_&nVN|mxslRd-sS}Ukws(HEtr6N7Lv;wKM1Q znu*!#%tWIKNpVw+z1YV43&Edxw&qG5H6`$j#k0#ac8H5O+uuU8Bx&(;-4pSi(k5~H zVq@{tF-Odsy9k!1-mv}b3B9wrxZO4spE6x17V>SH_-nx9}pND-E_Pr><=aM=7f3Q<*yjOOuv_WS>6jFLIs zIhRV&DwionX)@VZ%%Ux80_dyM4%C)^h3D)ysLhh44=ZIU^*m=IsF>i@;0f?prh(y# z2Gn}qi{7t3Ld6%Q(Ic)$tivI6UN*yZ**Tc1q6zcT447`={hcL$@Tzhs<*0STdy6Ew z&boyGuWRtqCKt)Ub~yQN9sG8P=-eBNm&XtBe6b()%56r%%iXBE&d;W&?xD@VfY|o_ zWK>^`)g_wL@3=N)g({PAML&A)Ye~gC_I$w)gbgF;U-K+1IJ*!Zw`|1XNE_NyH=Cq% zXVH(~{Z!_6h351D%V!{%u8-R*^| z3+G{hvONwb#M4`qlNA2<7wJ07)5MXo)RXuVzdsMf&GaS##b8jZsd-`SO+c8JoN!daM`V9IxU z0sXWncDOWI^p~W--Dgoas1AeA$Kb@{(G)p*C9N)SBUK46yp4(GyXPXz^UZ?SienhN zD3kBv-SOg^6ExQpz~r1Rm9F>*kAFt=qU%09D%#+hGm1{E4Iq1)6J#tG!5s?&5ye@S zV|1)xJnA)`#!J)UKfP&0wgGJvbg978nMOFLW1}cVfh(V4>7!%Vz_ZDu$irp`oM-9mO*{sG|c$q%Cr0qaQtjV@=3LrbGjMZyC>tu zXh+<%Gsj#{bv{d&0N1xc+?Bo;U)!ZfuDKH$tsb0h4GQ>nfb4}>Oyz9x5rdPlyz415 zXg5XO!=rI~)F33?+JwN4O$g31K~1uNePuS( zTtANtYWt9r3c(@47F$)qQM7b03T_kjM-M>2@y}@UC_w0z_qg{@pXC1yq}}&dliAp( zY*C68#@ri-`pr|3xJHY!&HABkiZq>Er$9xf#b|!#g_Y%Q@ZBE_`$=bMS!pEs*@sYY z-e7*uy$fY=4k_O%V~^IzAVPvc3|vOB{UT^VcL?3Lor^~&T>1W#XXiIe<_t3tzRM<} z#ZyGmstw38k3m~0DEG=#s+yQW^)1{nJbMYk?u^AQn@r3;^a#TwG%4rkQTTnRLAlQ_ z90|As#gXx_xYPuxU#n;~HTaxNKH4X5c*b|jq_Nhu)<1sShMR;`(hi?!uKads?Q z{(!MdgO%8qsvoTL+i*nh8_(=7PQs(Fc4(Nd$Aot(P=4Y_@iMZ5~fG2yE~ zxvi|!_!vq)Efrp!c4M|H`ZE>BXTtr81bEG^$AR|ELjTI%>`g_xFxN(gU4Lza(EaKB zZ*K_G^!>rE?aV{e%MQx|~N z#(7MGv*U+WMhm&~eb|?yWz69HYFv31FDQ>nXNlkUGL0c7IQj1oQ~sFEPS2RmI`iz< zlEa*dHv5m@Uzf@Dr1&vmW}Hwe^_K;vC9sq+R%}Az4Ai#hW069=pb)--$!9KO1AqMy zUd`uBEy)=;B6mngivYWqv5<|d9Ky<%hp>QzcJ`=RjcL{HU`De~u~m6WjBdrU;t$)| zx0Ol4-EY%zK*<2MgL^UM{=w{Mo)j9jl7vO8N3*$)bn&f0pWi z%$v!vcX#htdd)q+&Wk#%u^>lqa@fNTJ}6*^?pr6|~}fCI5HGf0!X` zmvUy;WW(9xC~KAxn#vyS^+dy;gUo2PJ*Et{Ca*SGLGoUMkZQl3owz@dDJNGljZ>V( znRivNnEsH}HBZKi>KwtN=?N=58Ooxc1q$VrGnwJIer)lh-E6OYGV}cE!L-kKAbqMn z!o!}jWipOTrsW;$H)JoXXwzn$M{cqTtplvH#-5GaHx6dGs#wuIR2ZB1kTqAyVa&E) zmNPDEGp93&?D-dOrq44sT^fs+Yx_vvS6I(B6h<-Y$RJ^x>phmCm&#~~F1xFKn&oC} zXDVD-44VHnl>y{_30X!D$$qo@l^)_jIstx34o> zc0@QXKZ$J~+rk!%lV&Lcr$K#L4+^YXSjgaL&NCX1kfE}y_DclIExN@nyQ{KQviniQ z9Z$yRCkX?K>{yM{E#Xv{HQTl-huNHqWY4nXS(w<$%))Be?E{wVifcSmTI4)A+JXu=g$hOK`s5%$Q4JE5k* z@rDn)GrCb3#P`9SU$NcnDfB0}A+>Hhw(s_Zh3P0*j}6B1kkkC^_6r-$e<5DA2Psj8 zs4~CEo;G;n!po0%|M()wRVUJ+duQ3ce+*Z5XR$a1FI@j_h2KZ~F=9?5JJ>J~F{Q`3 z7a))Q+pPt^iZ^VP&Qa`bo=r8Mb5QmE3{x5~3M*z0Kw6A|7p;d8=pD@7UJbu*|-|77^AR-c{`|O0faw>90iRf$SiPuML;UANM ze39>+l&;}<**<(f_8allCFsrB*G(N`N$c!xa&uonH6^FvGOdnnfg8RJv4qTvQ8>O| z7Se@tan1ZQdwPs8HXs4rK3zB;@f6orjHK?xed$f%S-8g6;;ft^-S2vVf`{Ab>MeO3 z<~^W86UtbH#30Hp8Aff!{-j|vg_8bkq3p*!kiu+iydZ(x_ zn(tQo)AT(r@G>J58~>@}_kcCrS9yim6#KBTyMo!h8%<212~g6DX6=1rVI&klH?0aI zWiGIJ{qmXIvkiD3VTiGFbP%=E8QHS&2uwHvP0qK4lZgFj5I`#u`ec5(?&Sz*+pGG-1+_CG-0p@c04d*TQ!q)ziv1-#|JgV11c-9b9 zIB~{TY%W@Vv|zgDT`bAlPLJ{=|dPdQ3v%;;<4;U74}Yx zL*q|5^626|j^=)_3j4$KhR#51S}jxXTY=rxuTegL(dCr`$wPJm4olyGeXn(J{ydIG z&$@w6-)c}Q=Z{yLZ$WL-d3-)L6r0Z-MB3pRmhw}EQV&&Np>_^t+>}98(r~=7zkxB@ znQYd4p2K7#k!+=bEj-WC-c-x%>~5gGUkUDr$FNgkJ}!>YLWOS_3OoFfJ@q)o-|b=R zcQFjF$w&LEedwFM2>loJhTf@5FtH9sVfr&f)mzirtNJutA(45my-E??{mEQeiaK6_ z!X>3?iOxTC`hJJQ2pw|VWk@4lTha4`{q(ZoD5*|fKreYNHreJmWhNHWnDp!-Lcm*y;1%N7V=j zjbo_eTz^_6Ie_Pme^9SW{ls1_^;B*=p5~MeA;r5OoBwvxAnwFTIlP*tS;>lPxqtEK z(wh{2yBz-3+~suF@_3l<-`|K zyt~wXj4TFp;Zb@y);ZNjxpeDwAmK4_)UZ)njZ}fF?355@@qQe>~JYT}Smz*o)_iGMh*+JYmI1z8mwM3ni zdU|m#hf++rZ!xfyEX<9?cAn=s94Aj-xQ}uN@AHMGpW{2;BDAUUE|i%qrF!!__SZ5@ z)JcFt=2;j8aFABaQVic=j-X;6Y#$wpP?bYypJGX)vS(1_{+Z;nEd;*X` zV_HcmEap5!{=$tU+t-^uG z{91;8%}A<#&F{(?yjOBI1+HQ?zS|_gjqhas=61 z($X`yC~HIo4xH$E)6MdL%;k|{VhZUHT9*bk| z0&%lE6q^Nm+*vG%#4(p(7H}4w9b2)6vpKdu`^4g^PBF_NoAG4v7+g$Sgq>Rd;hte5 zySF%t*)Fj~n)M$xNV=FUUc8HZ`Cio9vJG$EtIp;zLEJ5RP!Yl8!IHn-wpvm_R)#EkB*FaeIm``xie=ZXASb&B zGfr-Wl*cSAA1DJmeol~W|Hv|hPaWcA;)xH}*D?K^u^n?v<&0SB&wir>f zFuv4*&a}5e=5h?AHy*%a zw_X_G^P80`hU4i+a~zC)z~r>Uc{X?rCd#gdqtS4Tc9B6c&nY*ZABcB-3z_4(P-gu; z7T0>x;oTMkeTz5zt`f#h{qbkFb8aEs?HuInx1*-bfpe9452Nx7n<5{Ex*5H2L^Xo# zcsGLX9J>d5-_dY2nU3*?2cTen7rQ<6283J5kWs#knmM!adDStt;nN+K=AVMxe0xlF zZ)IEeRA7jKF335V9lNZ|`4&kiuKR_tnidVXf_M9KTcsyDciH zTR#NLP3FO5m>afzGsdhZc{non6HfRYLv;)1`j?G?vE?IXaUu{8y9&5Z<|H=1+=9Ti zZFrG64ph@4cheHKVa+2Btf!VwY(;5}cCQ_2r(dq}B^b76&q60Ctx)7&z&tfQlemiIGLiFDINVvNJMO)qAu*x6zewkpc+G?C^ zJ&uTi(+J@2=`{wNNmBnG8drzG)0}e_itI4r`xb0zibcPW*?7+LpzGA)kms=v&)Eie zV+W+ZUS)ZO&)IvQ#Yl+fxxjf}SVX2FdF6#*T#ph~73N~-=q!9t4#qP>BSg+V3hxU6 zU?Yu4=7S|k+{%N6wKEpmtl~=}BYdFC93Ww;U_sdzRITVLJ_B?~S^$!Y{ z55TjD7L@ko1itS%gQ^HA?){aZHP7uJReBzdyr=0Icu~R6);)z7 z)Au$e4xUfz{wmR}t158lFVA!2+<%yr0)*P{84cVoQKCg?3T)QTsH{pnZQB`Vn&j&)t#%zriax_?kZxvv7|y)(z%#4q^c zu1T44anS6N#hPeSEY$nM!au8#hpGjwulAl)(8$w;==g1hi)cAzCk5evF^;&XKf=T6(u(z@Mfyf=WtAN@t-cMZ~% zoI~Lcm1$Y!L-r-Cn0;M#iMh4~BQG!wzZ6Z#c4IU&O$%XqcqO$5T0r}203Ky%l1q#c zMxR!Lj1)s^S{}AG*TUl5bj($#W5p{J;Gy@0JzE;dX1&d2N9ZCOn5GQBn-b`hdCyjl z{ER!fmhh=aCcO%0E=fR#kA=pj|AyHZH=3-F$GXRz zWgR=$GS--iKa#)T;_``iMT7bN=sbRQw_wMD-slWh!(Rzk{8`+aHr~^s9hq~9D_n3S zZaqHpuEMMNGx^VKAjT(1K($*Nv(*Qo(WH>2)|DW!y9=Xx5zT-i46m7Ez@IQ|Fb>3~ zU0ZQyeI}0CMBxiRn_bAS#I#Ix?u@*~#71=svCbQb}2HUr8nMCM{aiL7v$XV$_x2)HhR3+|XY~ zJhf0lTyo|$E%@+{9LE&U(xzGDe8GzRg5+p<)UruH6u9TkZN{urOs4Z|hxo1un z$H^=d9m*DphH?K<y04}aqxHSWWH^%DOzONgs-a%uCXXwLr%5No<)#Qm8u zqR~f2Qx$j8hXR3=MoV+s1!(ENWPm(50&V7t{ zc%B9>yg_!vXX}e8nve{L7nSFw?n-5b<$`X3FL7>d2=~UwUm|8Ahq>%YX>G%GH zWN<){l)6<(UwB38e18{VkwvH2P*JC+w>X(|H!nK5iaUj6;&e?WMqU|7I#+Z^aq%dM z9ArvqT0^PtTN4U+{vU1m&yGUw&!W-}6EWo67@k?^CmvY6NSwM*5N)0riQRh*#Ep;i z#s6CRi7`PBxo5Kj$GdOyb4Q@)Te(s^7dl0(`E4(19vUa^ztcZVV;Ruvw-3sj-qH;LvrmUAvR5W zM5)Uj5u1`hdpOH8|4S783AUz-e4f|7&5F_|9^ zezQJljQfI#ZMm>XY72Tj_QoRXDBMvF!#@i< zTp8ZQc5q+TqU$=8Fu4OcL+-QaC!5*VV|vW9r2?^QWk`;*#q6HEgOU4tDBt$yJ;HZv z$-fR}KG`4ke8>H0zB^fjHA6mTG3+^$=TqlNHqggkurqie%={QGob&X?+(}0;gEQv0 zdbJ7x`=1N$Hiw1U1uL;7?hy2qOz`}_R5t%@G@H1&H}g`mW`l)2Y(<_5Q`>M4CC?uq z!zLIdU$!9P$8spi@bI!c zoP!>)rQbc7%y%DL-duyeJ}Xha$bj_J7t-f!Ng9+m6r@ziV6#e}lqfuiJ^17KV4qDEgL{kyH za3DrMIwl0?qzTy%7YhsSz7ozhrwY~EmJ0{w*^$CLfxH*)CC5kyoP2r+E-`sDD#_=egPHV#I^nxP z3Wi5{z(w*u2&*p%zpai7eZyjfYfA&*#XBMmcl|N=u^|nU7b(8_B_uBCQ*f{+{o1KQ z#~w>zmV7SLnHYkQVfQ$D>np z>aI=OzOTaR70>bF-Y3M5J0^U2aZT7M9VtYZZ6ouWlPIcp2i{H_L#sCTqbXKFcu~at zqSxHfONrs4;Rt53wNto1ze<>pW=V%QuYCi5*)OTRCH!%|D+J2=3l5Yh+`CsTBz7Jb z(x)l0@qNDwm#i)d$-!VzpQYI$pGx6KRHI-pd3g)px7A?uZg;$UZiB(y<|y0!j*Y1K$KU&mbhqhJ;k#}j#Z#6gEy)$` zEtO=8kuUTqKOw9yK7yZom$ZB((G&|`lts+J92aZO0TfvLhhvw)Ba)wuqyLho(v7|ndz zx5ECMO-xjFV;twATVs8kt%^xn@k#ngD zss0>DCzkD{UFtLFN+ZbBZQV%hTnw$bAa6*b3a`aE@*oZs!c409yZBR(x~v z4n>zfqvQco#a`~p;vWA#;+_RI)Z0UYQYvK$zZJ-g?{c&>im`5#4jG4cL393B~ z5AuJKLQy_h^W5$~xm8qYGKS)W;WYX~4`y<dxqhf9qAnZ5cOdo>Dzjgj&4H zEk|%>8ixN02N@>7F{lzAb}#T$CJPr;^J!dNDuut^K@Rg2XyqzC_cXi-nO!Qhvy8je zhQ7hIu7jKvwujDg=KIXTC$vSLbN65Je5jufXWs9oakn{+kncuP*N@=)3T?8|4<)@W z?(W#VhN>KlsnbHA!VC?`R$78qAC)5)iw699)P*Z&E}*kK5Xb&}q1#J}N%e3pl}y&7 zxhIV1`_Xi0yq!W{6volUR*~Xbwvna_XH%}SqJ?)J(OQc(`dku56G!~ytP;-3^|;4# zhs&vA=Wsg4U3yxw1*B77OiTzNmlgao^Pf+xzXcMcB*e`jAL)&3KDA!Vru?5NwAy7K z{VtqHC9{Xq*9n7Zr*{uoD7>O8NAHnpiYIADJJO8m-gL3Ofp-d;Fgf`O0=3%U^zJ&^ zf8K)JokE&7H;J}QJwPXgQM4s~C_StmL(^~0pgC_G$?Uy6$(uTHM-3yrStBV`YbN6JR-j7_pTcRe ztR}sk*`E$Xa9%afYE*oU!p8myh?YJGgB_o7#hhoBf4881`wz}1Ek$taL!M)3z?ubH zfpeSir637iNxGCa=P%sSc*kq~0CKC>p)0Lg^y{xY>Ad~{tE3TBKW;cmhFN3H8CR?w zavtIT%N%B5skI?}*0iR~%X&06=QB3mFNe|04vbsUh8tZ2sXTNqHNKsTh$)sZx^n@h zKGX4e_E5OE&BLyZeW-SU0_oo!L^Gn6;`>%h*oCi!L76V!mCZu>H+!7?K-?QNgidg; ztJlylT*#P+*LU1;Ny3a?EFDgEZ#YxY;|xA3zC_mkm-u@r3p=Ef;T6ewYFclx%kCCd zoK~TNzT5}u_8*S7#bTGcB6a=xgQR|KXd%vlkz9sjDcjI^rw66Cn-HDf1M7fn$R+ZA zcho`LYkq)usaKqhHH9L6qqg%wFxvF(8!)vp0fon}hcQh3+%-7wzg?!gz;Aovg5 zidPkj@!h!tUV~nv-tz}+R^5P~G|yIK9Y@~4PBhJWfU)X-crVle4aZ`T`6D^+T%XR(9!mNjHHig2L)D-r%&~cjck$}9>2!bEs6Lz~+bGejs(08~ z`x(nJV_@!Z1`n(Ep^HVxw7D5=-82P#rX0flmB;a5P8ky86lmMVulVo7N(ieoaV~-H z8aQug-p_hibaLh!`az?76~15`eA}Ghm*53YW$r!jGC+x+9b|6~!KvDSHXJB&v4A4$q?dxG7j zIXL6MGkMz&B1vf?ZWe51Z_Sr557mcEuIMp_I!jZzX8~r|R4N{MzTu`6;t8KbY?!*1F=$21gW}PUrdaYS5ZMte#xO zmb$4x`)M%?4m*PY*^6jNuSQLRCmKI-c6s3nlu8N_x<9RF(Iv??G+FI~=glraACq8h%kMT`7+B z-lRdNt~|uU9c>uD<|xuf^F7|hSj_THgoMvlXielhTR%_i<8y`Ma|1cYumzhPE+SSj z5H#T!29?aiQT2=HJog?wYlUNIDMuDhJRBzW2F=ec(47G49_8bbEl-n8a%lB63zNmXw!Uzs@+duuhqkuP8E2O z^&3y))M;JKA#}$bL5GYTp2Wt&=X4cl&v7(G>O$Jf3Wv^)M~>!X8tvbQq&9HZ(%6Y) zrYO*~{$_M*p$aXqmZKv^3MBG-Pf2qY&8lr7@OzUm)Z1pN-Q~)UZ({M zJm~89L*(QgMs;cP>09G^atT{cw?;ii^|=t}b>7F4`{oqI-05jSIIYO*B&qPLbgJ_< zc`R8$6SnOkx^S9|S4z^ki?=aml_CvY%$c6Uok%9!ncn%SP~);%GKii|=Fdjbjo!V; zS*kxRk-UwcLMKjb{){HIJlekGGUaSPMah4j(ffAJ%Afp*e4GZ;tf`)4>1|7ayffX? zqD1qwbog#y51oqSJ}=Ha$uxAMZDu?8ohy;*udm1cUnw}Xx)MEIJxH8kO5uLh*uu|; zLD!;ippmmMcHE-mA=k)Y=2c2`)TG^Wh~KHr=+WIM97G1fUCyE7J@?*}Eh3c%i^-!d zkgVz^)5(w>G|ciU>a!nUNy{xngny-elIhf7SF~l;v^sqj1+Ma<-U(~D8)^;Z{jyRD zBIoa8dZYV^EKc+huYSy->6c$pUHVs=->(nd((8fbRa1KJ&iyJyN~FwL6ZIqIMZ>f< z>U88BQ|^QumaZaJH%WFYWx3DxAM8me|(fO;RY3Ju+dVM;PoW3=Z&3+Yf zeXL1C$_9~q;X~XCNkCpu0iu&N$TO%n-QgLQvdclVru`1RzJH2#75qhI$q-VXqD>2v zCeT2^kV1x!rLI3G==IK{dFDkIEj?EEJvTRoP-s)y57xjy7GXaO~7TG2;7Kd$BX zf|967v`eE0QziHvNJf)x>^Gp?Xl;7k_ZyOPl&Mz|XS($M38zEKwC8khaz2nkaop4Q z7`MqFFpN45x=`*u7aB5hG@ZO{MfP*n((Kq{w6(yS+U$-~P4NX%^C;tewh}5@Cr?YN zHQRB&N%|aSzY)Fo-ulX!}EHQO?`Z!UFd!PoBW2HK+>=Niu^I+Qk zT$1AABx!r80eK~6Aw0DLAteR)#2M!Y*ISUr2TyVwBum9z{m8-oJyg}4Fi__rdUo(^ z%Gm?d5f?x!er=%P-|DDmMl&hy@1gRsJv8e_IYo77h({MWu(0#tY;4_SHm^ek`zQWj z13Kp8?9>gI6kLjU<=s5Dl!*K>O`Mza76ETZ;d_n&4(ZOp&Rv;M?W>O)Qf?SvHwNyP zcQWG-kJ)RRh0G|$i|y}iiNQS2C2DX7&%PUY_1|Xv%nX4IsgdugxzOf0Y!ibLwEiDO z=N*>g+s5&R_R`WqR90!A`#jIPQb}pqAqm+@WY2!0r6HrJq(W0WElMh@QYteQ8k7(s zX~=r7_YeO$9Ean1p8LM8^Zb549|seBet@~ZwQX^;!*uMtG8LcfG{=uQjnFrDE`!Q_rycB_&nQ| zvZwgaR)b5hdi`9q%^HRsQh(w4pIZ3Hm|1sx%yGrdeb|<3i9>TAK=t=spkSwjjcx02 z>8@g^s0JK6H3J3w{PkYoey? z7M$?$J&v8V3D>&)WCcSzs@PykKl>4L7}sLK!ru_sAcMP?X5pDi0n}b!jYsUx;P!J9 z@TvSD%w^i3zY}-iO~xdCnP`e7k^?OJ&GItbb{yUA3%0E}An0y}JEtIS7hA{9rt{FD z(H_5F4~B|__n_t(0atBv(dYCTJi)i1Mzl4Kx-$(G0|x+`_rN#tacDa815}xhLCbqn zQR`(FbmR%4E<1a~LJf2_T0*VuK^$a!xdquX7+ZA|ma_iN$^F|P8Dwzha1F%$o`ADu zQb6Oi8oFNxOx2x$5lOOmTS^cc7-!UHqBef;jl(~k0^}#@haSy8;px-`oSvK(%KRsZ zA^;#z&Sd@Vcu8e>(->9fkMIOWlu2=0M|kt04R$qLhfvA;(7>1z5E%nSb|J7gCKw)6Q&kh>`dgRa@*#FHl2Qa&HKcn{RtGK2-r_hIYh zOQ1C10-FyRpn)XIBD_+@!CzLW^(YbAS9xIY=td|HlfZeI??6Q96_~z@guy3*Smt34 z&0eW6D}s5l%O+suvcq_9UoEz0Nus)O5qQse3!7~vuv~jrV@n#PD&Y5G7cL(TX6@h!^QpoJI z1;yetyuMA9Y$hGSK*=m@&tZJdQ;%_{);z@X=9nQS#dxC4m=tm!>QI-iozH-#2V0=; zq%L}kegm@rfZq|#u*vlogtFh#$hly6)N6{{KMRu0)<@`l?l}n9Dxv?3-B`iSWc4#< zqE_}Iyvi81s{iCL*l`c)9hij;p~iURoI0w+okmBN8*25JrXQ!}@pc7B(@a-aVXs?nIY~`==5#~b zhtk(yg!#>O_%PxDeqAmMi)S*0X~KOlVy>xjxnJnMj~)5#W-O1Tia1thJ>$){;^A;fy8N>e z^)z--wQwYT%Un#=adkyYz zCYOYu(S9zyv@aue;iO9|F5!j$#^OTOP5kmT3~RNLusEug)vIbamyEmI!t^#S{G1Ut zh37fpc3-Y^<{8?hFqx!xgi%8HVSKzG2k#~Np=3OvydBG(hQ*@pjAl+hzni>Tl-I-xRVgg>xkpxmUOT^lMGB-F#y@MJMl^SS)Ba5hI^H9o4aWQ zV8({ZihD{iutJ=4Ke*skmY3DqIfI_IThhF?qqOEu968PvLX96{=sDg6{g0dg#q59J z;Q10%nx(Mpr!M}yz~P4VQTYF}q4e@eR6jNaL#CS3`9^d4y5JJ7cic=>V(Mg~e2HuC zisM28@;TpTLo%OgOR}P8DR@~2-d$}&Q74bm-*F*q#>KdXfx(z|>n?X)awfDl%mS%> z1t`*MKn|%aXD)t%`>h$x$tA^b_jL2Qs&m6!_YW;Nem;@AI<}MBC#VU_77cMX-(3N{ z#!@({(!{OT{lvv_j5%W*!j;@8zFi#q3zrIFjy1^vRA}WXSpfd(2;{t8xBCm z=e3+?U^bUD?+WKwt3y9eZX|=-*5t)HmMd?@K+p3KIQcZ4lWo7im0ir{UVQH6gm#;N z$Cy-zNoK6ydyeGxQ7 zc}Z#3mzjt2z8TVuzb8rU$$B!_HUT_>a=~)!KFDv*;zTBW;Z&=8xD%D#80|lmffpyy zm$mcIW8ehi8zkeZ=T+PrpD)~Uc*;eb@#eJt+s2(a;Lq*&$I+Vheyn4R|Nf!}+&lJ1 zGFkxKQ%_>Wy>pm%ZzaZv+Ttds11M5(2J6+?zS!(EUhFQwuKSO0V>U-$n7bwU@V?SdhHykIGbJ3(@q)z4$YUHsjkcywbruwp{l*D(D zfqo6u7Af+7MvQpf!Lw9l(?p-{Nbt#wr?IZcfhyY0)0Iqna=EaWj?|o>TVfL^_@yzG z33}0%rKbFibJo1qZBu@~M+>&4{K2hNTBNLAhQ43Fv&^L$+kcsm_@5A(SN)k{&wBCU zK1=z@_iXuR5u@nHLtEw`*-p1R#HiJr%`3-Rk!TDf13BJ(#;O(gS5e#gb6fB^qgBd!N{sbf+edB(&a;-kk=@j;$iKeZF*Od>L7%4O5(NFIpE$kzMNw z`fo-)*SMlWEYy0q5;!NrLPIJnsKSH<8tffiZUdouAK(3z>Y0K3j z@@Wv|X>v6VFVZFPNKXoQnM=K{Lu9O>&9{1*k)%jwbS zCqx#m{xsCs&Gtre^wEAg2~QN{@4PYKKa0EY7h5^%C~+l)%s8^&Qi*jzk|gM4NqaW^ ze@CZFu~&R)-P*}yB(;LNpZHN2^Te25HKpIxKt&O6g6=dBBM z53?Ol2ixP?hEvLAwj)~UO~Ye~smI2OdCUCBNHCEtye|??c|j@Jiu~JA{kVFy3EeMT zM2}`FQr#shifQ(w<KmOYOQz2kOKCVqgy*|0$j0M5eK5O8T_1Gl zbdmv;Y<8iqG4D|(z?}l@OUW+f3wG%1k>$;`^d$QR`Yo(QbN@FuNlu-fEtaGQ8#JlC z*`CZlMA44aG}?Yqk#??eB+ZFtRL>UpYMrw@-RrZY)HxDVb>=&! z320HJ^&dC(#LP3A3ZLV-zG-``%NjP+k?V^qsaBK zBoUTI^ZuTp7Pk}R@xF(YHNRo@%r2aFP?XferqIAJTMelCkxE=L)xCQ{Hd8*3TX8u~ z2>*xHj5yJ&6-{cX_sKCqo_{~ujV>?or-LVN(-oPo*zcx8D&Cglx%@s&5*#9<{vV`w zs|E+=PNx~tPGr1cGI?ybq#ZLiP~kjr3bvD^(oKd`YxSK(CEt>CwgBI|`5W$$b|umK z&XnsFOpbS>Y2LSZx}6wIYnnXi&tf3ky%N;6N0%;`gpovc3LShU$LFO<@n7a#pvS2L zWYK8AyNx=J(n-s4+c+canl^y~Tf|6eqbM16n&GIWZ=t^RBXbZ$QV9FM-e5Tb;awBx zmdbBb^vXgtwnv{HjTFs(GYjkl>9B7AdW25E!SDU(W#)sG+jilWjVgFze>zBRO@PMC zUep|Xg0l(_qWd^sYW==|ZjW}Lh3{>ta=9|qJ@}3~mXc^L$XNavRWNKJO;h~TsQ0xb zJ&PE^=`)XDqG|({-w9#Q;zO97=Y*eCSoi4YEquN#6>Y6lDVKF_4rM>YhL^?I)trSh zf33$#DP>CQ6Qm!yf@Iz*L^oppVqH%w$~v~;nq?=jcSkPPmK9+}TNBC*Fvf4}ADH&y z3v~KkfiEg+@wUMfoN?kgOy4GitNyk@Zg>W4*%(OUCOA+}pFTBDYet^UX4c)ljQn!I zQ8|^+lobz&`mtEzUyj$>r17td9A(*v&>52v%rRr$zuH%@r?e1a*gjg;Ta6CLsM8ac z6V*a>YP#HmHaBD_;ysYS%2tf;ID^I?s=+<99L)E;#fPi&vEz3brp*$k{5~z(enydA zJm|&^GyCya%WM4oZy_yk+>U1D7TDh%j)P0sJ$MxhHXkX$-mY_)C8CIJyHqK5-Y}Xo zP|5BrLMgKm=-wiTmnPQY;G%yJ8afF@v<=8JY0bTcR#|<0iQje(= z1vr_|!9W#?>imZZqGvHTegYYv_`=S`D>0+^A9VLaw*b#aE1Z zn*ZTo#eI~zd5ZM}Bk5D|I9i>ShlV<#_=WM$AMG=z3V$s+8$OkkB4$&>7nXniWkB`I zd}&9~JW6<{Or>`pF_&RA-r>qH?W_=~>>a>qK0@?0PJw3sRihW9uc1d@DK7gFf<0Zq z%qzZ~l!sVfMrRscs_>=peoj;oFp>0@%Fv4xS$dGrj$PKSxU^mo%lI!)urvt0wF9vF zR1`{-cVgtrew=O8iwDoLKHB!L7!li!CO(QN{&@t3Q}RJ>p$cuA`U?MUZ^b*Ui*b1m z^Ihns!@&ufjER3A${C9>-baD57YNe+kVdSMh@i&L-_XoKn=ZMy&_*9yI`<}k%DrxY zo&7qvF{Ttcs?71Kz6OF|8uVURh&yYJLv_h}2oq|+u{*kvvp? ztS``sPY&RCKc)f!-JH+^zd&sG!c= z88J7YGfWLLPOroR1FP^?BJ15T9@gKIWXvj0#>cwKII+nYW;G?iOjE`YleU3FDbum# zFUvDrn}80x+hJ%f;}PudfP$hQV7$i?m7TAHV&q5Q4Hal#O9yrz>0o|X3k*%jf&FuG z;nCiSR5SevN)?Zw{mv>_7xk1A56b~QAROlfKZ4m?|G;JMUdDOJhnY3iF#P=k{GM?I zZby~C>L+oiY@>+to2_W+$vP0!{mjMmSmVPdtQWO%72H0*6E?*tVEX5~u;Ji+D4tpk z20se9!KPUtG)(}NMify0?;5;Xr^CG5524eq2%fyEfRJ=k5IwOLo}1g@)3v|hH=8rr zElI;^UzJeCnE99EkD^&-3@rR|7RB{C@pZvvRG#+(JGgUjUPT=q#m>ghK{N31M@4+~ z&SVuwqn_W1c`ZDG{#UIEL@wLxtyo5y|&fCq_jFi(v##Yc`Sr?K4!t0!W3vJivxd`d2l~p5B$5wVbFmHXf*VL(e}4dWBViAx+n_2 zi#!C!<9XowY&WRLcfsG1YjC;!3*+&!&Z7?VKj|^1>e?IlPEC(YQYTS_6YKL$c?grn zT!M=W^w3)&5AIaFfps$%VAGf~=$@g7E(;z&L&XW`&O8tIB)-G5pxdCEc^b^0OQPbU zi?CYJ0`_aF(3YF0@SeR2ZXVyjq^-`-I&L={-yn<%%ZnlT!)1``&cosIc_^(i1s863 zgTI~xqQXcZPFRr+`kH&d(n=MkEEtC;KivV>pWh(0oiMm!DeA2VMZL;KC}Av$?082I zR{e~F1{ay5jCJoO<=}wVA@pba-6SJdS*&uj2cjE$eLA0tVfCu@K zn36gIcQ(esud8to$sA0#pI(ReUS6>B`B}VMszB>A!!hOc27Gk+C)l6rh7*V7aNe)4 zuqi?cUpK`=!>2xI8XN+r*$G(8-Zz|J71TZ-Lm5uvNbtTmZGNgkV?;7h$5n#_!>8i3 zFjEXT^b>ql1~9Q_6uHk)q#|Kq`cYhqNy@_*mr{Wps-pB&in-}lTch{t11Rj`fTq{v zsp?`4E}W2xugkQsMcf{P@2Fv*<{aE1(+}BM`Ji>|IKA)Q!uA;}sX17cMvlxTL+`hE zs9A*K2ZTtQ`S2bGzQ?F4aWb2wOewLSQ2p66{9y1NJNm?FqMi(GOOT@9jT8B(7A^j1 zvNpdU#}SWSS%q!))Uf~Zbv(E#AD>i)VAXAV%$&I%KL=Xl)^F>1wOzgOofmx?RJ z{pc+Dkxj-y5)<$xcb!n;b_bESR27?lUO|OA_Wh_~`I3f{sAuno@AuB4c9uzN&$XhB z0qpKKCy3sN&!e@|-(l~j5_FoGi?V*qTi7B(omCS^c3%LtOP zUi>@(jn!76nx!=s=WoW)7%}`$rx)uNxZq+f)&(_+$4|eQU(VYbzvVr|z%yOA=PcL=*Q3 zI4OED7WUVnj#C%zsA@*1m!UYRc^gi8b`1Bhe#8{%E_AiKg?4K4%ts^0`kQTxo$vwA zB|SnzqhZuZlb|CuKQL!bD;gPZB+oz-x>(hZK8`2J{lGDL+wD!l&Sp5dK^Bw5Hla&b zKMww6=gIpm*z@}b1}OC6A&qLB(RK|RYf}*Vn6u_k7>53hMuqNM_$#>uhjJd`n1{dc z^2R#WXMT*Clfv-m14)`&B0!Wpi0@X*P|RNy`dTAO1rjDSJxq~q=ysxv(h#=2e2I;+ zKUhv9AA_`OP;YBK9-kpWFMLvEZ}~qcYu((KLqU^f&Hsf;K$gI4*-Ua{`S5rxj#3f? z;BlS-^siXSQJN2!IeEgSpb_RiY=!Gy{&4fAD@4Q(p`pZ2Jl4Zep!8H2+OQv9ysdzj zyKca%Wx^o7w;HUI1mWvy2Pi!_6MUBK1T%|F7@1KF`r2Qh$9*&^8cCxipN$tA%CR^u z55wDnSVw;VGg6+hy@v^;NEm`&;Ra~l7YtjrdBV`hGtgTWhv{ptVa?febX-=7$uGC%)c}N4^^eY#*S9dX~_iF_F|9?l)|+5 z`IuOA0)I?eiwEy6z$>fIV^@+JXx>qTwB}}r?C=M_;#g>s-30emsKJT&i_mSy?%ijS z(Kqh{Hh-wZ^QE!ibWH?TMf!r$>eCSNM;})F8R2qt!S>X2U!vTqpFz!Yxq?=BIS#qaAZSpzz`6nA()H}gRrx*;R_u(~z zV>s8|5jV~|0D9Ih;aR9Cc7NXtxx41U%X`5f7j+ZA8nxkB>qjs~Fb2}QH^K0ORJe4i z0_yn?*l=D4RTiCqTZf%tkg@!P%9!)ibOnTbTLl)bHITh21TGC{LHkGyTq-*V(VsKm zb$1nvgT5d9(pc2D-iec8LvWnmx+W^y32PZ#nf0$^I!Ay_y$ zADZ;9gZFI>csAi3h-H++?2GxZaFrE|y|xRACxwEqgfm?GkqYTsb)Ygn7S@-%0Ht$I zpt!dMB%?y$;gAxQu>IA|npFqNjXmI)F&{*)i(<>VOc=d;DsJ!n2G$L8;p2bW;I}LcG;~a{@o*iS zk}$>6cX@DDc|9zCSqp+SZm^E8fLXIsV5RI@Y#kBD8I0SRChrVW$GiafqI@t(=Ha^W z0|>D!hp8Q2U_bN)mQVczU)J|x>2CpAr8l1Do;?7&-vofFSs0ka#No^8o!GiK4X5`q z{*2ZG+`Z8OH8o1%;r)%!RCWqBue=S-tJZ?LX*E=>AAvsRijCILMr*9eUUQAPXOq}@GnT@)J5j{5^DGCHjx(kQ;SslWILf#OBVjB#Zy!ZZ#^+)Un-vtq zZ^ja9U&<2ErJLKl$wp)y8Jo-|gU<`;i)0x!-zuW7t6$KJ^8 zJAOTok1bPApi8MP7DsB*e3=m}PL!f~qvw!nqB}jjrA(`YmeU;8^Zx8?OOp;%(RR-V zl%ZWrK4Td4I{Z(r$=36M?IvYQ#7?VXNl1s=jV+?;i zdezYma=d6FXAYu?@r2e8rK_$M> z)!`4Go6c{`*X8@pj;34t7L(HqHcL9bA4lDufd`nI%ahGeE;>KQnQkAkV$TN5a9D(~ zx}KOfdpu9Y3;5MTqxg&*Pl}p$gfzBJqu8DLRD4H^Ui$pRF4l$Fv__7mjf|nip1*K8 zFGaZyA5pNKIk~xQ6xQj$y4lQSQg9Eu1aio!Hw&{gAapY0@P>1VXYcopN=EyAiA zU3_cyf_&@B*qlC?90pA&jdfb$Y&y|+_Hw$HHl6gwt)OuEBlvb^5DxpO;y00E(j3Yq z&-1qA%Fg0G*~w`0hh=b-x1jHD9ZU&dg8o6pc$WE8vlZvybemSZ$nIBrmOjTglc_jz z^$hOVDue!Ag{X2f4COw>;I-3p$$EwY=?k>s>Y_dpI+RQ%#U0cuW=$ILlW13=8rl1I zp-$d&94Sj=893&_(p^HgCeEdTk>||Wn1zr2dxw&FAy|^{i}D42I9up4$}U)oryX|U z{+4nK6*U{I9w&lREeNClSN zSdR(gjnU`AdQ1dY9Ddq@AGc*7+Lz$o7CXAPL6D@ku-^EuCz!%Ar}dVB_`_P8_sRK*DNnI+UJW7(S2yKqYNF5?&H@Fk|d(^9HWygNug#bEpb0b>*+3CnR^}oE{epD z8Iq)VlFf2XsgviMm-rys8IvUjQ0+qr{FgGe5>48d&_ns~^PZ)%|;DrId!t63<~qPb3;{osIja6~U%R_IzOT zqa{bTGfv-DG&jBt4}FF4YC$vH+M>wZ1!Ji6p9*PwJBXQIW@5>bV#vs1^QG4ZQShi4 zt_FFuj~-_4`@gWrUk!&MEU-*TIp^xFiUui5B z&_K)BwK!{%Bpn#+#NywRX_=)n^&UUW&i`}p+u@xk6|w`rzrT&w1Nv}mO&h*bJw_Yd zm(UfJ5W41l0k6m!qOVT^-g=&n=?9z80NZhw!5SLv12pIOG5S<@4l6bmqn$%3%0*X_ zfMx+r`;$R8qmH2Y_;5UyRf{}oLr)TWK&vsvC)y$Mo# z0!eFJ7@eC}N7qiJ;VBKz zS9$=g#d@${wK4rSxF0KCJ7KzZ7QUQ1n)H`G!^-d}q@yx~9)|bOd}2NxStdZP&y{Gh z!dhzEd2%K#n!R}@${d)6xyzh4ZueV&d79MthO%L>MK zFG6E^2b7uX%h>+gXwbQec_>^dZsicmmZg(Xz9k)+p+QTxxlzL31lVGq43=#w_(t+L z9V^ZuUUD%FHbs*5&TH)6ah6VR%*5qlm$3P7BB~Ba;IqvFnDamqC#p=R0~_*i;sp_^ zU)lyH7B6AWXfd2;DuhF31u$t^KfL`k0vcyi;L{k!%JeBjm)SS*feG7X)V~Bt2V4BH zj%A>xa(HE>EFO$8L{q77#(@8crOvUqNm-l18>~tG=L{0MU5`#)Ay_J!jF~TUaI>u+ z79{%MV9GW0U!y>i^cvAXeGOJjJcj3T*I@SSaaelP77vSvV@G}mTx4f_*WnVVxIdM0 z>o}6ya*WnZo{tw>o`cL;0n97ei-mXm@Xp$3lqrnHZ97x({vK)49wkK?LyUjqTY^ek z7vi{6lIXom7=@emjw9pxe`p&mK)IXFVDgM?Tw$Jw8=nebHtC~x zs}8!jHGui*@32kt8f2dlp$8qTfBbngeT7uUguH}HZ64q#>C5QPW>Ce(j0<;-b(|p! zb27K0#J4=`pW=@#zb>HndShB{>OfvMjcMhcU~;Q2Cza|r%6ei-FHU)orMC^8S}sS8 z1uA6yQ=1e`u2Vrl3S|!8qZgVFpE{VV+bL!;@G(>Ihj@egA~ zSvPs*8#I4ihPT{T(PG0TWD&QR`uxi2W5zA|eZQ6FUks$JlOt%V-&wL4{hZ8>f1$Ut zJLp)P0I#p6#INSn`P+-$P=%c^f9K&3#=w-NpLaCqtg05-Kf6c6drJw{4N~<7ecpWB zZ2rPtcGkG8ON%r(vJ)6jZOVC+`M!dd9cCVo&TjJeW_$S2a{NiHY2^BII-T7pLAfU# z`7I0F__*Q4d|LK83es>PT~Al$HFTwd$$pf4bOC9rf5N4DVsvG266@6#;NA=6=(!~W zt!6IZ{~dDVZ>PKRigICOR2@t!_C=Cn*QEJce~OrdSgnsjH^ zYvv%3MX7F-MNHrjMqQH7Zn+D#%wAiSt5N<@wBQ9yHFSx5kmVvWp);}NZD<5=E9j?&Yn(rdiM(addaL z4Shbhn?#$gv{3SS>F0sGwqNh)0@?RWz3$-nRWjpz4ClL=$T?e3dOn;carrqPTbNReNCQJJ6Ipw6##SV2+->nSzF zgPz{3WqmaF{Zn^)%Hu^X&!vpWjU4fIi+xe#wE_@0(2+{Dd}s^n8_P6tCH=y}LF z`Z@6swS5euy~V+_Pbr($W$mEf<5$v>reNxH65-!$`b{skiSZ@zoA~QVC-__3aX!>m zk(49E$h}gT%)HCd;>bVDd;SsEi5$TRuhMW+KI6JxJwdn41L%Q$4D~gqkmMXm-t(&{ z?>tR|oaS=0Rm79F3GAXHo95Apbr)#nY1To}bfZfaHnig9OtM?NpG+%uP`ij4O{(~Y zorYs6Reb?t4(2yoK5Y4@3O)G+x@EE@i|9mjS3g%hr7Lz5#mq>-F&iv;b zU4F)EMgExPQ?}b^!f|uDu;Z5vrVby%g~pL+5VVG9;ntarAbQAw~=eW9QPd zc(b;g-mrT_G~Ok#V|6sO{~YPUBXWPdklYq8r+0~wbPsFrcz-PxyFNokyT?loN3s8F z4z6H6bAS6%eA+UI4VNEd^obS>a(j!PMI>nd`>{0AtwP)XJ4gCGS4o{Yy1dw&IILQV zW_N4Tr-=PD_-`#GYh}@vgaDlOb1zyiPr@zx#*xx!S-NV*ct`ef%(bOUYl`GaspK&3 zOy7(_c%9nj}h-yUb`0JHI+EXuwOiSq`)F z8oH*Y;Qi}Wc>L@ooWXJ+mBn>998rcoQf=5Xs7ZAftjKP%J>5F&O$BUztR^HyaY7U6 z2oxuSXC98_}D;hkgBDDSdM-UCe$#eIK3?ZRw_QujTn@*3DS>Xfu^=(5IKHccam- zRhZEbgvr~b`02soyxbx={*Q|Q|L9vA-6~M#?{pqQ74MViu{{t2gtM>{uVK%dBJ@x@ zPqX*M)7ZR6bnuriUbtb-X3liv`GHWLJ~aeLCSi0xH>o&@0=;dxY}}*9d5wF zK_v=YT!OMXw=q>~2!nP4`TS$NtWbGUeZ%;{4hi^gN;HnB-ocFJ)c!vg#18sn{N7!CZo?w$+hqT$_6l=XlXlMJzarwvRr63J3k32jwQ{D zw@_`7GJO!zBae6KxX!#3Cnx%1Y4JIHs#1XbXf`uD9gcoe7*CsVXHN^OP+kCY6#tn` z2Pe+NaaF!Jp~)0?If$TCyEJlP?0(UC5g(@Z;BJd&=qO&w9Iq;5=PO7Cr;V{zI0(bc zQ?PDv9C#_GfLTKmobahf6S1!t_^<%`dwNlk@c`H$in@{_nBTVs7hGh&C3VK3%}mCe zy*}ux8iIe>cYU4qUX=Ru4OgF#p(EXDbS7~c>OAa*H9b7;d>M~Qy|;1Pvv(LAuSQ}M zu8^@Z%e>asqtU57`0e5~=H7_KNY}GCuk|ikb z3q4LJyvC6Ap);h(y0szc))<7ZQD#&mKI^%R|Lv0^CDAtA@aGlY9ANxP$rTt}9*Glo zpFwFoNlLMaz*CYVsCYdPJ$`P-+4B2Pe$;IAnj?+o%JNivJQ78lBCvO`7MJIwVZrl2 zEPU`8Hys#5suyO`;Z5~;NYrqFR40KAakjXMBaIZ#&Bh7VksJUgjO25``^3(@{-eG6~ihWBG4C z*0;LE{J|w?{5%(Jn&ffK3-(+c&c(DG7px0aW#09zXlbg6QYsx_QvMe{s%T)2>qCrY zeM5id4iWpno)1R`pfKA5y$|oet*qz2FmWEssAS_(scV=yMS{+AuhII*B!bE*G)LtJ z?AqXhGXtlh)rVC4G1&==N3UX;gm5(Vlpx+!gqj1@aPl>0Jh~|hgYs<2rQD5T^fSq= z?+VU2(vAAlmFbDeG;H{?8*iAML<6fDhFkuORcx2*`aKW*3SCioOaN}KV*U?P8LV8W zj`B}$Kye=9x~}>ME;)_pta%r=`3X?oibMEJa1&0M7J_2zvo#)Ej~iSwv0=R;UOH!k zS_dbf+r$9mrGxS9m4{gRtOQ3D04cJ4;>N0c@_v1UqQVxC%syYb9Oy&!>;5CJk;U}$ zu`^?39;Y3*!fC3<8ETITr&-l?WMA-(Zg@-Zn=QxiXG64k-Rs1cTV5ok@H>?LC6lt} z-eR2l2lOlSDMjvzCySU8N*zjK%$wV^Vj9aJI|}gD4w1BXY!3zB7{&W4w~&7AQ+mHn zop0-1LQ&5nsj)YeX1rWYOLUy6=-M&TvYtZv-mBkc0mJlp!Wk^VR_$#jA7Cfo6CHxT4eZf39aGpQ2F33`dg7t>FMm8;WLKsvYf!1 zH!3sYt~|}2xS2dHq9{)^fx=%F(38=>>E>~9zA90i-*Cx^X5}xTWAcG?Y=QtU^Gu1~ z-fqG#XAGyi59g4;eK%SZ=1pk~lFxNnKu~qh3loOLM6pr;z4|ZKnJS%pan0lA7JGQWw0ViWglZ z5c`Wp`X%`f-<9}T+sE;5+AaCDT^9UQk=MA(O^rsXZOD1e0FJJ+rKUi85-4E%7%z@` zes8BMf2?^2gNgjE27P|&rU`t&=gIv2r3?5!)7ANIsj>WvRXY5S)`wUz`w`2G3esi; zTfSqLJ3mopGatRdf&bUIm_NB}6`vSI)Gq2o-M_a{-d9nYen^Z)XUoyN#R7crY6(7N zr4HYup-exdRN0Kvn0BqJL1CQ|v>9o@pAR0Qg7yS5zcY!hH5D)xR3)vwoKDGCKjE^d ze~3t+i&p?i{)T#M<2$PK0xZ=eg|jWnuDjxU(Df};1i(0TCynp~_# zfo?nKbMi3?G7YDapwm?Q_dI>y^#@yAl<0%60cA{;qUJrLsZ&yxX4VPQ@g-WcvssD$ zg@30%89!;|FG*h3M1maH-EY+|<_Kxt!R9Z))NXx}$`a4BtZ_6+NGH*P#+x+mRUc(u zy+bWkZCJQhiEiH2pp5ZLsH@eDcD~z73K7D*$1_QO$pIC<@G@fwT#6umPXdj1{Y&!x z?98gI#;=X(pfIfr>Kf>w`3lcXy~KY?!tuD)N1j0`wqv61{DmOM0C` zRIx~s@&t{^ZN3hrXpN>iH*KodcuS9i>#0Ct6fdMrl+ftaFE5KM4$sCRS!;M!}hKEN(V4IUBlL)l5$5aKunJN8V3qmxxQI}0x^l7Gk*#6)m)|4wtG z=T>qyqi%7sYb;sDIhf&UI5?L4|1t@LtvTCI#~Xo3UnVv!)E!bVEs51>@>Z&uwCxl-aYZ0 z(dsqaU;Q)O{1u^`tldB3ZA;pX6K0rkVjfcPN6?v@J$ipE#Zo;yyxx@i-O95 z&8TEig^AN%V{O0)cit?WTikdCTF0E@78;wuhMDW(=I1q>tG+R}^K&D2B=G^9^znt; z0%I`EDuO$_F7X4UV`JaiL!n;QXXaZt{U-&P>OP^D1oQ3NAEq z3hE18EIj#xPO_mJm1bOH*MjTbgOak)vd-ydslPAI$>OT{THq%xr#e~ z$rH%+Fvv^XgJb%?jL)vHmjxz7a|YH*#sRotz3cW&Us6VB2xlso0U6ir!oujR8Z_Qfucj3C4d|{@-^sUS5!dZQ_SPpP+4Q#1Vf zw*fY<=!I3RuDyUUq#WD5QGcyF{AyCdW`{T6bJ_tHul$7bSA4;R@#5H$Q^$2Wwm^i) zXiyq5fs&Y5m?@S9mvx4r==5)}({_U_ttha$D+2Scmf^~9vo9g%e{q6C%-0ObseSgk*y`IlQwrEo0hgnDD;pH(~NSSa5 zqSmTlMh_g^=d38Z8!=O z<@dw&)C$PY6vG208{yB~LjZCsap@mHoU~W~=XVT)+p!0L15y}kY6`b~vcURW0P20& z3}Lw&;a7Jrw2An^stfWM_ir0q(P@S2dEdA{E?yuJw;5h5H$v=+1u*j08J5mV2Y0hd z)`e!S3|~v!y~Y3^4jVwZ#cP-)FNM2al)=6e4){T2BGw+b2pR?ZVUeT->x?6QI;MaR zR;uD0iKSqT{{rHw$@R&`uBn2ufULr`K`2Mr$+AvbU$^u6=} z@pq%~>%9 z7v({e0b=BMAKd!rFdTYo46pnq;cClH=zOJuSqeE=FeQjNvGZ_6hZaUG-U}Stp*n9{ zk8)S+P(6#s`PZ*u(3T8T%Ml=-d@(v98ivY;*qowKhQE9i8c62(9F~ZN7SCe&US$0|N z-3dJLMUobuu)r}g%pY@DnELLgp`m6nO8&=~Znu=M&NmEyR~O@hA3nnIE*myrDW zo1oTH0kv*paYNH<2o^U-I~gaGJJ||BS*_r0dmK-;{h|o3Yoy(Bkht|zQ6sAt9E}rL zZfO>t88s8lCNe%nxi=>5Va}5pdwkyC3;}P|(J5V#O5)d}DdWcrtY;3sPFqqsGn0}Z zPi9WQ-84!qhMX&w5#M3Rm~vkD=e!`jczc(*c%OsDDhFH_Igi5k$CAAYkcr29a@u4~ zVlBHVjd@$}@_IBlUxbCMA731*OuMFjL;JEm;KuG`UK(X`nIeI;n$lSEvI-B+>%?&) zKQT(Vik`mr!Kj5*xNF@N)SN#6UxVdQIVusC7MtL@tm#-|&m8lD3vp5i;P#*r#y`^` zPsw$(;=yH#k72WPQ|9tiWt}ejDhPPg4^n3jTTy3D@hjBB8PGEy~A{SBnY2WOvhS@Qhewjf?tm*V{v~JDqnwwLoe7Y zShN6duVxu1ArX}QcpH^d|H4maD|CEpM*nncY3iIjT4Rt$+jlWfP5LNYpy`APSC-<0 zOTKvEiZeBbXyEM+mUwci3SDc9#GKFTF_QIZO1@g*j{s9VIm;UtO&J8m%j0pWzdF@B zn4r!IL+psn!Fv7+cuUA*pr-_mo@R*!{vJKl)B%_YzSbBAO1vv&uQB;uzjU96hzONjDmyubxpu!P_-DcsB z-8-0D?LOMCc!$PLpK#OXuZ#;Mh=*COF%B6cP7~OeLTjz zu|u`mhd8M!2T!~&z$cPvxWQ8%ONH%meNY`HFSWtg2W8k-AwiS*8E7q9f$bSn$#b|C zd{xGx-+@Kg=JgCJI(lHt?P~0F`3?$GCgH9?4f<1d2@D->!ah?TYkrn6&VnwzyR(Cg znuE|yY%|iCvGi`294^SV!b_$H@q%GKIV@*;o>lhrc##C|9eNADw5zasohowov(a~p zH?Cy7%{ga9F>!b^&Ml9pu$>F&yQ~A%D*gtsgL|;X_5gabF0D}kn~Og=fs+0+N$I*G zedv**=)3n}xv&5VE{Mc1=|$*N<;J|NqIBON4aaUP!=M$V6zph)xsyBLt#JUl9c#n? zHZuR2^;MJ-dWrj@#*olXN%}8K2gN3B!kpb27`s)Qj4#-b;mW^wt;Q3%CPh3oX914d z{~D$Et$5n-Jmw^CWgU(rO!!!jBS!CVFlQKj1pC?C&Vr^t4pn@$tVvEIyQg zI#YI{i%vR5o~S~FPfyVI^A&#os~95!aD{ z2Ulhy-Tez&-VejC1X)}g7tNgBMR@wzW9-@U7PZ9~$Jj}hv@Y$&wD~poS+t!oPWIt@ z=Iu=idWye_{qaebFIGHe9>4|})O?_T38$wbze%!V0hRsy@UpoZ`iA_!Gjku^<}Rb(32R8GFrN$* zPGY1`Df+iP!-6DdG!g&sqgpZOWPC%ECixp}zhQx4|r@JHvMEQI=Wym4j&_Gx;cD!X<>Q`uhU+i|?O zRhHUkJiy)7r&(Vs6jg6KG4B5^Jj}WkuBWaHD`DpVX5 zf@hs4qe-qf&aV;1GYY2oR4WIs+}Mbn4vAQ)=!r+z`*LPb5e_%TWAT|Icv>nCqe6b+ zNX-!pcRr1e+@50Jyg%qv>OrqAy3@05yT~Npo4wbZ@%r(D*gd}(FUUVdwT@(r>0~_6 z$~)-nn}R#HhU2_|HH_s}igPs9;!ouTC@0F{UtEU0Gt*(H0& zgRYJ2L<@OIS|+7Kg1jXe9=e80E*`)Of~?A*77)E6boYn^zRrI7XZqVVozNqS$UOy4DkF;yg$D0B9&Igml`&<(6xI%Jw^62So=Dpq1kBQl1sD8Tx9cOdN(KU{w^pJV8miOWP z%)7Yx&?HKFC`|)$!t`;;6tWqkMFJhnGZk8dWBaVrU~wS(T&kWxZUWOuz^DOrr61#D!7KPA^AP5;?)7WsPOMPAfMy>Q z$ZXATEVwL0Iku0m;^tY_&)<%+UFKx%KZDvv3(?;Z1DcqsOU=y55Ov9b@)}1_QfUaE z-Snd^;sS(D6Uckc6a3UNh?A#%#2RDPr+xkt>-Jqi?~g=xUQ5!yZQpUD!a*`M97Fq_ zm=TI;)9@X8dfUa=5nbOYMfNHEmcL6nj*Q&O&MO^4bMaunA{v(?LP-yv;oz<~THkF$ zPYXuSbLBWXI5>!xC%(e%FOK4j<~oG+71(@U0c%;|~efO@xspV|vtGNcZ zZF9wX?PyfK>W5z(7vM7c9r!tPAuV05Oj~NFkZ4&8PPd3e!wptgIAIUw-e|&q{0`K+ zZc0hF6sSEwj~t&R(N`@WDsM}tz3D9T=opKEwL4MosSr(iat=3*7ojP?r_2jipnxaQ&-;) z)J*6iEAs<%_MAVJX>6d{18%fAN0vP6a=G(z#av3X34DFQ!||p(_?7tu951+HN1GvY z8p`3E;WlpJfDnlK*n^s)4Op|;Kt&aSK*v!|;kXbaYrN-{_#fcjzSRZmMproY>J66@ zaE(h>%;Rp?KSG~d-|$92FIxUVA~70YT^sZ2i@U%&TqKceT^Gj zb&=DJ|HEAyHwu((`Z?~^3RoDE0S>M$uxhy>F1;X!g%{Y^^>Z~=Y9(NfdJi^i(V~9= zjof1YSKRhH-P}wE2{7X2M$JMvqa3A!$xrZCY;oE;UU@$xde(c{2B5LfcC~k$-V1>c^$3xzM zHE^{34+r7Ks{}B$R)()$onZGZ7b-mZVSv4l&)BX6 z76pW&U4IzgaUJJz|1alhv=+W>_Jv%}Y#5$p0a^AFz*cG&Y|XNScQ-$A<1XysL`N+H zb!Aa}l)W78OsE2>b2;3$k~;3FQ#BXX(7=uS{l$IP?t|5o$@v`!=Q3YB<3dMCg5r4& zg71cd_?!uFvdaO?q_;x$&j6^`S`JIL3!ri5N;a#^WIa!5434jZH1AZXJ#q^A({FP_ z(YLsY`V?-3<1udaU;(!-S_x{tJjPy|V>nP>hnIA|aP^N|p!i@RyqmNb>OS;wszSq4~CNw zuYXZxV*y88z+U$^_v~PIe+2|?{LU#K_lDb}u0YpWbC~1wo%!C3kZ7qZC_ zs^?_@e=87fH?N1Di!ylglpE)7bd=M*C;(3Ah*II;zvB&vhr&Ox}V-;6zjNAT~jO0;wh1!wJGVnv$rQn< zhm1k^*8b5V8T*FZ6_PSx3g>yaiLb zw9!Skiz8DpsGqb9s_s;A&Ltz9!vS9iP5hc3Qon7{{wL=-f(4)z9ccRt)J)IY71XWz^w4NGT>Fyq&QWf2FIQI%;akzOjVT zONxnZhSO~Iv*d7QE{)x=nl_4=)5}2{5^PDLLPvi(e`*H5_&nPcTZ-_%8|ISw&pC82 zXgX;|%Tqbea&XI(NZIHoWBW<)BeVY__i7bhy48U?(KdeB0DYi6<@yZu%5v7ty z5sGTCre8lxDD`?W)ostCR}oEg_}LBG>Q_g}^;^jLwFUWIuqH|2)712HAM3fLldEtM zY5vS155=97JNpwgd-TwK*}t^^r4f0}@}tl_D@f}?4CyMbWqVvN%K!eEy!Je!z5|ch zzkyxZ@~i3JZ%bM>%8RCy#L{lJWQyn=&o?N^@QJ^~_+O{y(r@i((#~5=gX`>ht=}4a zj^jjL=)5{z-E2=MhBe7NIEQ6@vT3PJ5os^lP5=3@@5{OzsuR{CRWW^%`^7SGB4hcq z7G=Km=x;J93Z*IIooK9;EghS>lZrLdNH;i<>hyQg>1!dBA00;{UDIi8KVxTzOrn!% zY4pvejPzTw$R%(Ssm%%@&va7~`*(s?MDC?HhXPunR7AQ$5%eHtF7>Wvob4MAsABnP z+Wk_Tw+A`?f$|uB+BKe!yXwZja+}Zpd8|U#+gT3w*%<1H-a)@K*-UUFW3)x6lJb1^ z81~9^zgmlb^Pd7=l%v8o%r)RWnpOF+5-R*Q)}_eMyF-q*57E)nQlxaE6YGk{lVE`p z|7O)1-gmJB|3=@8cRggy_m-RSHBDx;$=Hcvge=LbmgwVTX$oNsC$aJDu@;E&XEj9l zGiv|gw_u^70@B|Hbb)g zEJX=THuQXrDqTOux>$}Wl;K%R3Bw1-#o9yI9jdqomfu%(OYE)>GM5ZzbceY+>|%CpDvM$Jhywm*rU%-Bx%H;VB^t`qq= zk-~h-hI_;xK1^5RFVg5!MwHFGG}9j#(_%Ic)u}I`55F0^QY)J5bRuYytS@;s&7|p} zQ%UWCI7#N(lC9T73dlF3{ShhTrFWR-{LZE+5>b?Uz=8aP9LV4@dwnvP({9WfYL)h) zS4u<^*q?T!D~k#y7Si$~3AEm8h*n4bpjW?MlFXz9y!1Ige*L*sy#Ev@F35Q=CvaMY zyE^Y3bl#GK-@%1&cfS-W`hA7@>vEyw^b45W4lt|lHB9%EhVc24;L+a2R>=O%LpLVt6C)>paP{~hH%>-KTi6l*ci%nQ$L*onC(Vqx;0WXKbK#q|xU zg4@y*kSj}t^xR8Od{hr!cd|3kes#Drt`O`q?ci0Z3|MDo!N}nx*z)8HJ^k-0MZ_`J z#j)`aCy)=tw^X6cbSot9iHBdUdzgFs02i=vC3pVBAx`o{3H@C1nH=K+sGz+LO1yI6 z=Y->M{8$8h;?6-q@qREfiKM<63Di;*N7{-DxDK5J?#ul(-2JcOI3W0ilag)#|4$-x z^Z7+o(o!R(v^OZ2{0M(-kf%FBD`9oZQW&)+3$hmflg4%0QOHv6>vfk!- z8xD=;@5H2-n_%^M1N_ZMm$3n33o&wOZ`GDI3V!}D=@Q2cQzh!$@L zo9$8Hon(OZXD7oWzxnv-?RR+EAPYA`I^d=8WZ3*L5;jj2fv2n7;pXZM%*U<`nb{-o zaKIhDlwX9J414a}+Y8)ly`b6698HgN7O7 z;6aNL#9STZc2*wY`WhwRb%#6%_G|^~5G`m;bp^ADwNSIz1ssQ`LG@oXXl{PPeV%oT ztGu<5i){(#HrwSv{6H>*Zz_a_Bqs=5xDkNyxiXz};7y?es4Pi>@qM)*ecJ@`#yWt( zoGXx;JP#xucY@;89gx%_fHuRYVZ+Y-@I564mR_ubZkM}YnH+%`zt`aGqqUsd{4}nU zoe?uCesXEWDR4{C102mJpnEBY*?9suN?rkAMF!js9|PG2i(%>{7V~_Y2>#h_kg#(W z3`JDKl`lLz?q3V*4zzMy`yhPC7Q>;}(%9pE9(1DvK+Q%4P6rvoCe3p2`h1363!yN3 zwip=t&w&CrQ#kX?3EB@jLhEU^4-T3MJ%<$FSYi=3AS?ol9yN20se8GO3su}I#Rg7i zs0Iw(5H6J^!K=c#5OK#2ektFD8FzTd@=1jA>2+W_@i6OWG-3Y6kLc6X#rcXga=#45 zfN9(j@Y+5G$1VB*)5aVGfjA$KU>TgTlA3t+(`8h8>W$Td$KYm5FbqA?g6&K6@mGNz z&R#5t^@YpuhtnLkBZ|h_4n>&F{-3?aKXIk<_MkIe6Flq8pt^VwG|dkHv)!pMJ?jCc z%Kt*qG4Ju4%rd-pZZ!@?m0&_l6XJJ2wp2=kX5L*05wer(W3 z($Sp8PZ7<=(=Cl?-EkIs)Tf}{=-2S1YyxU$SkP8GS1SIIL95;Ss9r{gUo%ylA78W= z_o(c}I?GcyoH`MeMx-!MR|hL=(Or|9l_BjXH1fxu-Ctg=v#-@m;K1P>rJ)v*RASUSxBe z@XL(pC=pRW+Q%=_C&71gu(%P23t!{z1Jd;J@CIDSo~QA(lQ{UI7A>O&v1~wq-mOSL zx0s`7%yz&nelA$hcMa#i8AaEPLRjxK9{0*J4pZ$}JmJ-b30J0&y6{cbP56mg#$q&m zzYJ;i8pTv~Kr%*f75EIV$;)lU744%$7@IfNf z@Q$a@UEg8)>6!R08qoUpR4hx7!Ye_-Xe;G|I~54|874T#eN3|DEuza&JI&$Wi7son$RkQKq z!Zfs=Q-E8Sf5t;6?%)^J0a0Ai2bNEj5zaecMUWgm8EuGOekQm-><=7u8jFitmf?f0 zP}(prpH#lDBX!49cqMoYUD!5-p6y+aN6fb1rVA-pdA$$ghO`*}pLH=uR@3p@JE&b@ zJLzvy#_cS3r}<_9Zo27@%U36(7V{)24kpn^^e%DrgX8zjw}}CQ}>@)IFWH=ea5Ar*$Wxe zzwLr6W)(3nkt$v!6%?sdL_aBOD!8|u_P)!Y&n)Ytf4&pnYj(0u(`dS)G>KfxcnV?M zj-&Bv)V+@<`*A&(F(g2DyA`SE-)h=w7EIBTj*|FXDdu7#s#zaEz8#)WQF4XTV@#!o z{Z451c|9CGZH0ZKS#~M(Am`ycz)4%*z?IILn0jpl`pQ&j;E4xxY(I)_o?FmdX~fZ%xN~ghqo_3 zK-fo9P#@X>HEVCdVwFQ&`Q(cLT37HDW1|LoeZ*>?Q)n8Y36|a;U~a=894^}k+Hya^ zO8F@K7Zw1*l~XZ&Q5K{hDdprkN?BoZNT0h+#!sx7H9U zTDHS`%gOld`FHN^@8j^I<0gh*b^?<#qRdC!0-c(wpz_flr#)f3q?~-t|I>V!(UXC3 z{U5o)GZA1hbpqPH&W1-(58-vAI0lKH;vTw|z?5!Rmi3B+QFh^Y)ld&VZ$1Qq=1uTa z^eKK>lnQ*&49ti~gmF3ATu|`^&Ny8KUs|)wVxAj}tMdc^6^tk*Lf?r( zxF_h2Z(o*h`xDFH$kx5!YG=z$o-Yfpx&+WsQ3u|w&IGCCHMsTFde~t84g$K8A@)WZ zH`Xx{Y|ePY(i5zOh^@AoD*jZ2wslME&I8zTem=HcN+AfCbUQv z!QZ{2G`{j3n6I9WwlDhe<^IRe{P7Fw{QLoV|I)c5S{d-w_ym*(7{Yc}AzZAsAHBsh z;EI4M8ordoPn$zv86JcMTDu_nybLtddV%|xHy~c92VoI{XmT?cy~j?%e}`*v)4WMI zV55YG+ZFNIYzyRtvZ3~<9ByPDkc}RWP@t5LogsZtDeA(VI#9{Z$u*F5`Zl+(!V20} zx?iM1M-Z}~o zO-aG~yU)M^cFuVqvJ+z1&UmJ4G87*(#-qo(F>O{Qh-Hq!6a_mJ+aLq?5-&i=X2xV$ zXakE+6+(8?8O-_h8Wg+baN>K$0lito#3CB_EpicV-N??V8_&Vv&SFRwX75Mq92k7S z?!$D#K=<4UDC_+TG6Bh8sagzcz6oL7lQM4IjXJ2)6hiZieNa7S9cJ7cf&Ic6T-8Ty zP~ZCs&YaKVr2qcr{zQr5aUzBKRee6- z8}U7X9{ifEbNP^S_WUE(SDPIdq&Z%Z1X$DQl?ofgMSQf29vMzS4qCA87NualFQeB>(Odn7r^Z3T%<@`+9GTQN-&3o8R&g02_ zs{VG1{$0LJB8KMtt0v?P9eCcgu8w9;ZKdAA^Av1!h8z=)QGw1`GS^wmFU;`d|Gikm zU$nYN4$;R+WMer=)JE}-wng)NeGq9q5$o(2k zy;nzj)?TMyOZ+I}$0AyPHjWJbYoP}|O%yWjE`42)OK-N7QdZs#Ivz1Xn=K~rWAr5X zeA8>>OK-`~`V}p!o5pVfbH2k#pT}|!{`Qp(eA~GVywR%+-m$-sm*_3zZ?3VUIlFCX z_W>vRY{s5D^HGRIme7EHFsbhIrPLw$xZKmBJdW=tj&wKJWEDKEdP|ui|-t-_pF3f3P{9-@P}R_D#s78932?-QrvkcNx+R^)zMlX6t4@`n+ZF@Q;eM_Xm z8>yuKa}8biVMlY98dJ2E8QbwWQ{yr}-cHz=?=xJ%yNGs@*Yi$Vu5*W6W3%XacoW%} zZ)g5x0p9egAn$!ZhA)@V<`*8Z;wSx?$v?b2m+CKVA%iTAD7uP*SCo>ES|hDn9>_l^ z59fpAHuLN5w$Q2T_4Ho*HWke<=2wKe@R>W^c;T5-c&$b1{QYVTUP*=Z*d4RUr?7?` zEywX^#Afi?dnfasS)XjN9-IBX$RpRj&E&us!piO|D9gr7B1`m-I-6fxOIY>C|5ildskhB8QFoDze^3*CM`qJ z+#C=(6A!IlbwK(`HV7{unAIi$0x#0w&lG)lXe0y%!fhZD91VZg?uL`vNia}13ex|0 z!S_l*iu+HTV&ufhY|<#&efvCyu=^3yuXEsDo`Qigyd3=&Qmt3Mak0#R8i2~)x!thyG7QQCB!=g}K_A5M4F>a>8&!%*JHx2LdkuA$vU!Sw8hlk128l7| zAd;#NOVj^yPTR*rn>bP`n{nB&E_L8f7v`IiCuuhCj69SHsU9<6Q{^^TV|I&WtRCRU z>=JBy5(-J-M&Phf0tzLkV8vQF>>h4{Jd-l)7`qS04qS%M%@Js=V~68zad`N(Aq>sd zgTFl$oZ-h2aO^n&yL5Mf_4t)w6@COt&rgHJ{>$i_`CJO&gz0W;4i=Bj$B=i{81te>zX zvr8N@yob1B0o7cwz68FO%!Qjx3DA)04q@&Zu((zVcBB`>)j~O#oHrRFWz#U@)HEDu zKZLsCS~xT^2d&@T2S;%yu)G)x6SPC1UwJ*wJ^vFtUj$&Z@?o&|(1XGK+VDN^JH)h~ zg68PCFfB5bW%lf-)@B9CUk(8C%oSkpMIVHZY=TE$??Bv>U?{pB3V+3y!-xHfFv)%i z>MOB4)lGG*WIv0P@(1rT8$h(N59F-Bg2MMcE^p`(1iea!)2tIwA2gfsPS0_>9{u3v zO-q5p-E|QA#1u{mWI|%81=wzw4nIa>LBq!dRz6IC(82=v_>3{%&(}g*GyiZ2JML`X)6$peo$L!%zYZIKca)xJXUEr*`E_9_%g47-5VETF~^zP1wU-zUj zY{D_t>-Yj{D_28e>TW2X-~^}5HbCj(bujmmA#@!|g1Pgvp!}*P41SPC8N+ViB#uLy zts9u1UkMf`3ZOMt0s1!20GXGa+=FXtao_cP6w_n=@ka#^bteF(3pj(DtPD!X?*V)M z1aKBF@KD7xD)E(~qh^^6Bp-tCs*imT1 zQY4PAMDwmwd{!!Xh{lmu`g4k|e?-onH%RF4 zL%P0V6WuU&CdH&3ZK3q>}YZRU&($dGdxFn#jmJo$e%uGXVMOr3bGvY02e?mXT0%O{Bd*43V%PXR`5^!c*|eS1-dql86hTf<8n>hz}GuI&_cZ8?2p zOs42Dxx`OwrW3K>FtXzo{wr{$S=0Sl7r2ff;VY5C5>npQL_Q&nBxfTCAwrR7eG&FI~izgrx|S5rvLFO)tZR%^PXyw^8$1Fy~3X+R7TN%hu4wS zBci9H575ey&m?(SkzecfAHEmSq1bM3(yn(J)2a^yDKfox31llE1Po=40j zn}6}7{k)Mb>^Vq z%72YeptXC=siI&m^|_6vewXpYRS8gnRu|3RY|3A77Ug|^^kAW?0BwCHNoPA(Vb$_v zbP`I!{Nc;EUULYqaX)d+zEn!+y+snoTj<57`zW0L4W(rt;mw#&C^}t*P71QjbkKIX zC3BusKh#oCy(~FR+D~t%ohR|p##Dc%k{k*nXp!J3vN>!-x1Sl1LwqJR1e~IT>`kRut3p!-4Lw$@LrV3VM__ct}-)yC@w$r5H)=v`QLj3cw$FRTeJI=5i!0_24c`W5xO9M$>tdcky&r{{U_2jiCpXN2owWm@jja;%_~yq%A-GquUKbv|Dx* zgmX9u9- zsVQiew$|Fs#}cu zM3l&+`4IhhI}abXa1=3(Wli|BX88uJ$c z*>7O;;Wg1Hrx}H;_7kyHAgq-R90$%7s8I+7WjR-J@fqb zLrLmnGJ2s*`To*Wm#2c>jw_*JYZWw1I}Y!vTR>!LIb=FCf$_LDXkMBHBCr49&x9tN zE%yP3$IW2uX(3!1H4IbqEpXO3SG*H&4K@5uVcWS>jIPK*o1aRkWNM37O9#PapCc9f zpGC#ppKy<>3E7!Tlf@+wT2VWe%`BQQoiWW;vrci?1q;>@w4(9j-RWVs0e#F0q2Eg9 z=ulHBiA>BVp|!5$q>Z%oktN;JO`&7=cG15x+o>q-FE*HbMumELD!KOnL+-yvtv_|x zw55nLx-)5P?mm*46Hh}c>d7WJhTi=+Mvb@1sn9)`B9k`JjNuqsrWHYDO!WNT@hwLD zuwg8(d8AcfOlnS=q<=_=H;^CC4|t*lIGL%AC}~Q z*NcXRj#BylP!hCuqk-kIbkmY(NJWQ|+;YfsVF9UDwNWnf@=5uQqt$(k+2J;bc~|>U zR^u~v#zxRZCwCe%98J4s4Pc7+Fv<@PG5>QbeOK=y_ZjynV}}tGy&tVX@yyE?P@q5q@iF8xDwf(-%_84-K{Wc}66#Aa zpo0ZpX|&C4ihccq8s9Zz?32rQ@L4CGcC8_4*S+-p-VQqc^9&7*OQFaeAuJOwPH~I} zo}a=z(HEtucb+iC{SsyT!?koYg>_qRB5^O+9{UpW1<_qhJa-UB+ts1*u~SGX0wh-X z68SriF>q8S^;JiauTmLJyCz2YTO?@hhF)A1rA^TZa+HucgN~c#(bdZ?1diwP+ zy?nZv0zG1GwLE2=fYt@xFZv{VvKQtCQ>L&YLK@AD~Um`rgDzno#>_D^l`T zqE_d0GMeK{hctIm6*STjzn!$>V;p^)b{mQ=F<%ysdKF;pof3FzI0Ad-n_-{90}L2@0>6H6#;aS#vt5EQh9|DU zwF)i>DoyxtdT4c3RP=N`og-UJ2#{DatnMUm4*hR5c%LqxbA>D zzDbbAaGy$yVGHQizFz$Pnehlp)iJ+Y7k!ciaR0sSaJ2XY%&Abpk?fN=?P>ykXgz}} ze{W$$_IoH>x(^cKm`layCCvF54Br$^LDKbVaOpb?Mc_^UQr)nJ8kcS()CX#T4Fvfa41a^c) zr^sXYQaKl=%-n$%_ZQGVy)Kj;ti|~yRq)^StB}|H9PA_yqxeV$hQ=eF(W$}a=cVXS zb_C_;@$76Wg4<8dMyI)%^kGaYr6nAu4OOzJ>-Ge;?{`DfrABz+Hit2C>~+&H!-MzN z;s$|KG>twCOHPJ@&2(7UJ+FZ~|Iv?*ty%;hqkeAZ3{Z!$?k{3ea=VctN)arpL( zI}VGUf|MoG&?Y<`?2MGKXSyf`j<>~6O=IZGga>F|Z%?=T4Q z)`nHIt9S*RkgNmIksa`}bqAWPbD_fR9aznJ#`lG$Qvd5nit3C+3w1*rPG64U;wfwo z9S-;JJ%u^P)o3Z3AsCIGO3Qq7(B`Np#*VFk;DSBiIh+C=|D{2xwKPiFrorNSjc{^> zISN|kqf2uMmQ{+Q=)KFZIiVQ}tlD79mfH|2a|O(6r=h-L7>>Ukj&JHy@DWpVN6XKl zNlRa0Af}=6opE_fSFe6n}JMob-mp&L}lb9VyEkbBhhp z>_H<+d1YcDW6MvTT!K>D^Wp8ndidg%53yC6*z?{6BiJ>jJgpW#PP&b~DX-9tb@t0w z8DYN2cxu~9wVyH2?2k36no=evj3rAf4QI*HXJQh(0` z++A`8?dI3v%=?Y7%J>OLj`)DKeILZw6v6m!Ti~145Jqp%qutj9>BtRf3}p9FVYZq0 zJ@Y=+E4$&DC5`x1F71CDop)T1@At>4v_w%_L`$?pd0(${izE>ZN<&FSd}M@VZw<6X zlZMJnB2uEIL`Fs#QAQ|fh-8a?*YEo;5B-t*zOQqgb6(FU8_9hhHdXqZl^}!DLK&;_ zvBcZBD3~QY!qN^&+QVJKD}L3J&*BqwYKl9x8;GcJ_HxoYSBITqoHtEVgZ zbYfd9RrZupcfTSU{iK+xlST-&=jI5n4k`;@?rx%5;}{BG6v=n6`nX%6oz~C(|NHSD znzj5k-AlVoiMHj`EiNIfb@)dH`>h4b6W!GON?rJ7m%;m4C+PZ?qqJN7F243Oz(LlS zQahdLh38(18RA6`Bu;`xaCh^JbNI49L~!oER|qOd5Zn?6(`i*%N}8@r-Zu}y<;`|1 zpA&)JCD-WaxohNe<}Hnq3=zsT!v(t!p~CYODr8|JMiCWSod1xIpO)vb*S-Mj`B`H% zN}IBHZc0DAoIaP<(%^=4YUI1!@lz+#`j^{DU%Hpnum2%zxJ|_#V#3~^&9vpqH>#Ug zO9Nf2N$9*vPhF%)B~Fc&YHH9KPpoDkb8Mp&;*89gd= zi9)GT>jK#q@SX4~2_f*nBWiqAK^b1TxZ-~XjSU%C=;S1rotq)(mQNDel+9`Tb_<$3 zXC95Yx0}lHy@^)Mq%*hjVL$F6K8|QXVpj~N<=#QU%v{8*)e=7S3c`H_8^L4ERuc2_ zqiK4!^u0ZrvRgNkE!#z|H!eYf@}c3%bEm_t=|+(vWriEjs4Zz^vF>v06e9bASZ6+4AHkDLTq&6Ps*3KAwy zvJ|FxdkO(1L1Z?JI~DeN(Jq6h7;oN*3isF8e@{{ndpAgknk*^YT%Cj%jVipWi^bBp z;n<H{Yc{TBVFP8P$}WpYNEAUQ%P!OB&|-Qbb&*!Aiz1Jo{^XFqf{yHV zr@o#UV)V{*3@!L4 z34^Wg*aE%-Q`|9)q%9q2?zU*UvaSNJ*F6NaPbZmILn!mf7^;q&PK9IU;k~>PPTQYB z*xd6h=D-`KC^MPwMRUeNuN^$a9MR2LG~17UVWk5W;m6^{v~*|;IYh-!B*{>Tk^)`S zF{9t_V%WMFw^`Gh+sr+&9<3)WseGFQRU}-X^F`k&bL&aco#KyQdpBc9Z3@1NWk7Rb z5;|UTHc5;=O?q01HR)SuzyLMj(_K%}{H7*Iwm-oJQ6COxjv`;_GOV~*hnMqf(Y?!! z%=qnW1tn0)whw$?QTF+2QS9z z(372RxM*mHG%Gjs21b)^R0MSnK0~3t*U)dL82y?)jkYx^PS6U-VPe@y6h#G_B|Va6uVf~$J8`Bw{jb(-@#atMpd&T}91Gqyfd8Zt?$ z+y$XYU)D||We6Bms1b?#+0MVbkwvOox0j2;jTrCt3&Z+!%3**|K|P2 zI0QCN!ZPj;thn37Jbd$6=w=x_oH3ZL@9sy|hT1esVi}xUyIJ*Q&W9HDAgPr1{l{9+ zow3a@9zTlW>rLsF_hqCVu)+AP11L-}5@LO2n4$0nDRs$M-BN+TYtyOg>JXaQZ9w)0 zF;wZqcQ7AI@p@w!BKY6mTi1ed?E|T(gY(*L1zc#kfH#$?oM~PUU5{q$@8<{Sx?NBn z6^_3q+|wEK5^1?5a1o6l#Y*lSiir4^gr%93%}5K6u24cjf_aeQ1dQvb$5 z_iGXFeU+diFa~i;qoHeB14+A^c-nCvgY24M_P~SUF7F_ROTpw)y^@X>ETVgE3#fHx z5AtUB;LNHKl(NcW?#y9;7ID+Bei__8MusofVF-_{crg8VL&!e?)^ka-F#sbrEHD_9yxN+=mlt zL=)5WvCe!5!UO+9{KJtb`(TF&9lOzJbPTb+7ZJT>05yznM9Hb!JT0hBj^48HKA{BF zTLR~;e}t%?8vWwF2s@_&m?)OR!!jPXB8=c_!~I_>Gq5~bjs|l6v9E(dni5#9847ezzghnxB0`jp(zRB!#-afs5h;sKu4y{p~;+H**VF47Q;6;q9;( zDod{x%aKq29!mdum!8W#rPJHWp}F-M*n%poEiWVozMo=W6GHc%ou&49A=LLbh<5+D zM84WLsBe1(`3?I?Ev7Y8yP=p2w~5i5vUeyilB8v`>fkzMARY1+C-*6%sW^8mEv^!y z91T9BoAM0j=6u7bS~I~iorG!k1z{mSGb2x5L83wpzvs_|?{GUzE?EZ2fLr9+{~V91 zR8w2{GGT|SjZmwzo&TF~Sp5n>K+JmR`_F^yh%y1r7~7#YFC{~wLHB}3kWf1!)lKpeD0P~Rf47_Ikm@_C2^k=aD^#aYYeqX{#*pR7$&@oiffA0Z z(cP*+wB-9wj4RQj=WRXE@{uFaIp zo?BC%tTu)4OrU`l_ZhuSBun*`bh9{wro=p;tl~UUIGauuPmG29#$Oaau$%HNYw*6b zg*#~8qj|&)EL_lvC%xyOw`D3x$*7V4?y+=fKqXz#{zGx*k7(wq=Vbl*4ei8Bn)+Hn zD9;iX)NC?`}N`eEQ!_KIiNbNO~_zc~I zHXNUU;EpwTt+5IbZ5gyQ;u`JHzeH8C{0?U_9M$KH@uAEg59MZ{RMHYFxX*I!3Bcs5 zfO%PbM*LqdmQ@X)(~nFrSA8Qi^%ujr*9$=qRg6Me?^t?t#Si)at6-u;S`TWBH`O@4|I&v0f$(7Q6l``tAs-XWS)?#P%7ru|K zLF?wNq~yLx3QO`MEB64>Tf34r{`8^3gIRd7w*yUk<*6%hB`H=oQ`~=H)HTCWIJ`_r zNc2__3jO7vJ!TQw9TubX(*?Hi(LZLAl+X5EyNb)*OW6+7a5!z^bDIe_QRe#-yJDyD zyTEp+*@t86)C+j9p$J!fuVT`NO-POur?e%D@UDFuZdde!guy|~-f|Yh>uy5vULns( za4+0PLt0h22Tg7_utH}5O-UF`ooNHnd2KotS*PQN%yN7wPGC{(IcS(L9M7&PAt?i} zeSaJ`Q|FQ4$iWnvBaZ;d%?OG*#;&Twv82NtY>h%EYwdEzVxu7xQkRLHhYIMJtRoGq zxI>ZhJ78SW#he!?pyg&Sd%XNDoBASyuY{+qwM(5GV9jd9|c;Q?e|DTlFZ?9kuA2CLUELjAMp zs5-X-mR&|9%{kZM{C6edy`DL$gE3g=AR0DIgvKHdd~1k<)aYuw8xoCFv-faFl*2A7 z4T!y0SmG0 z#==&Rz;a_HPv#^C$%p&0!^K{K%ky9;Lztdstql6C5)xBrKj#scrS)C}$xX0Nwn_!sq zjLC5iW$vO|%+s=l>9$9(|8_rPo{M(kkKPz;;ViJFiD4MO{vZ}P@mZBhA0F&Ii{MBt z>L|)&&NuZye$Ux!#bCq?kjDOba^&vuk=4$)&vc~9M86O1XYaPQu>{E_OfQY^NOeT8 zOIA;q-C`q}u=W&m*W5yZ>S08=yg=BXr-+-Yf_NnrG)n$vG4GGyspWRW+r(qpo~Kac zp1Zb_+Vt1p43p0K!d{S@w{-EU;V_D?Usjn|Dp*D$jI?^*UiG0vcj!e;Md=;!GG$3GKc?qUR$UDJ8*jKAXvCvhrK z74v@$!^S22{;{W)jk+)cGBz{t?eS(*x){*oQVG)CU_h^2xO@4IKh*^9qsd*|QE040 zg9H6(1vF^VM@`BRccBvdLs*)?zoGT^bh73Kt(VE9Ee~H&OxAhsD;Ok9ye2OcPuM}q zzfaSTiqG`0Rfo2A&!^Otd+2@6V>~hKN8h_FsO#@nY>b{wkGm5n@MHp78x`n*RWS{= zil?0qYRPn&ny_qPAjyQUrU{$MC^T(6HhSh_2v1(*^%JAmOg(awTSX-SS1B6lWXQAo z4HBi8%loV&%$>+cbPZRZHsJcB0pvYwGp$i}yZk6B=n%e+{~##T{Orj?^>%EPfRBpyKT`@=M!I4SsdBBCMPIt{GF#3_cexA5D)W zBq-QEn%N&cTNoY(~{#RG&FS>_g1f?^!sNLo+7Wx{E|ib_&nzn0`i9;NWZ?`h7n z>saA=5o=oN`8~^$PShxonY;rn6pN$%!GnaJH*Mr^R|5a*vXrjrNeK!4$!x%48W#{t ziGLGl&yGwIGixW)m8}T+yq8SEU8!dFP4tYIOR7#+d3_WOouxHk>XW!*T#kmiP2s(G zWqQH+m_}>uX>R&-x<055UZ+=(^uI$iHJN)iwn)*81smw`Bws8VehJkbuB5`}ll?O# zXwd;H+EArMWtuh=d7bAQg{^e=d<$KbfMe{d*6j|-e z5UHO`6n(8#VX;xb&g@-4PH$9M;r^cS;XFkE3s`8 z<=9$@WEQQyl=VG)#E!0*!6LWpWku?e&^6OXrn4#wFU(>uzP-RqQ_$emPr*2UVBL!) zOgGbn-HmD%*{*65ok($D)7$4Gt&PWBw#VSMf-`3oyc9)GOA#eHXR!5CgV@m*+U))H z;cU)f8$SQY5RE(~!EB>vvt_Lt*kto@Y~AgC?E2BmBK52ww&CYq<~jZ;OQ|wuT46CP z>QDx=c6}o#Bc*kG)&e8ZaWH8K-Z1N3>fFaynA zQQoa}qDrX~q9en)L?*cr%zcnIzxXz@X|E7f*$!gb+a#Hntt6&N2$1*K zgLCumz_jrtuiZMKeXu`Ub9giBkry$k`GZ*c=q6F}IT37mCg7jG2itIbI#Ye}S~S?_ zuc*=BqiAQ?5E`^Zg6h9hMYVnS~ydW1t0h)ihfieI`AxA zv~3C4{fY+B_{U*vhTnbc2vDYDehRcTOAd3!$l&ZzeWZllW#2O%vnJ^{miP9&$UX9} z=<-fuc2)b8Xz$d8q7z|CY-qL^H6^dS`I*KMc zbAHl-ucD|45{$WDNuvf?rh zB!A@R*L)PIy%haP6JrlAyRv?c2ig0g5Z2wI4RH-a=JIw96CY^Aatw6fwqra7&R4{# zbY13F>cVtK+B4nw%WTugovc{HfGzo{%;x;+6n$;2V-@$iMJJDUh_u;o7Hn?F@|P8e z=DROoesl6gjoP0@z8P0Va>qBZp>fNZNk}Gh9OB1XI!CZiq{KXzk7qiqnWB!;VeD?X zF66>P@bc(ZsOr69JI5bqr|R!8jmZ0=sL;ovZHg1vRM%19v2sK&T!S5TjjZraGShjJ z&yHMDXO=Qn>|m+`JN#`r8(rkY%xMpk^0>?j6_+r3-$-U&o5~#v(V~YFv_${Dc8D%E zMzZT$=Cka^er)}!$D;n6d9i$mJBv`O5M8R#w+j& zw%`IAQDDt?I%08fYpW>Ho9BPGw2QKN{&K`nDfZH5qv(C|G?u8)#}>pCv-kyJY-YO> zlb$YO?c8-T?86kMGC!MbPiyoD4m&58>`vP1-Q|3$ojW(55ZYB;_KIi_2PE(&F6z=Y=rHx4^3{6|lXv z9!jylv3qkmls!V>&E4A#zoQWT^f8WkDNb%`Nl!E z&^ipA>SM@%yA0h};7!ALt+-HDp7bC0v9}bC2wf!%YieRH8U1lB=``DM|2a;pHQ@XC z9GuB0M9Q%-Y#e8$-|U_VCdy~)3*$hRy0BdyhV`2tamL_qNM2rv9|wbRe@`Z4c7U!W zWy9~{bX1S?hr~%2sPqd$^84ioDBTC8Dp1t8`zRhNLATptu+Z-}lc`$|mmWJ@I2!~U zGsgO)A*2?qPT%aMC_!|QU2x;sjbs6z%U98&@<5o>Dk5X^JC;^G8&hK{QK1`zDDwr- zJC=ym9)>XcV-0WB3E0|yFr+e#@bF+7qxUh&E zW{A_4qCFPe{daH;_6(d2g_Yxw;yM)g(+~R_74fq|llsjXNGINn#hIy^7=G&~^S3yK zBPE4sjNK3G6-Efzy${|p3n1AjLE%Gh!rI~oKa7yq-p1L&rV8~KzSr@enu z*hubNG8#V?#>f7#tn>2`c!2ZmHjhE`fBX1etp>VshLhfY?w1ub>H91(dVl^2)-+pE zdM)==*B9Zba1Dtuf7yQ3vn--2k3GAmM$HLh>ER+>w01th4YfwdN{iE{xmH-3Ckait zMaVF^$U5~r@Kl#DZskp+4ED$1OeKu|Rl}ZtyTlAQA0c^aDjQ*Phz+?>%-V}iv7JXB zvazFdF!^B)>T>K5cXc>Ctmi;%f&$XpC6H)X%UoW)Vjb_qP#WO_dA{pmcl{vrS9qdi zj5zJPUJS3&4UnEb7BTzn*lf=ZR_8hu%|>1*@ZN#USZ#Riuw{>yaNoS#UYJaKf;$%d zFuwCWli$GmA_tn;Rg-4+=f!ESM`w`wg#~D7;`y?9;W&4ym%TbAhrR** zVY0yhzIl33b6kUaG?-kPl_@RgF=T&svQftKS^mT<))mqp(rVJ^v(Q6Xl>#ZYjN)}n z0$6{}ioLuB=^Gu{>O6Vc(YlxP62?&dU2Wu5sPKJ{d>p(y8wX_a*%SGCcGleuH-fCt zE0YF=K^qa6W{AbDD~% zV2YR?RQD%e_^yU_dQl3X?#u4!3Tvla=qcAl6fv4XUNtmNT0b}j4FTxcO zle`h=bAi74KcH``hYHrk-XZLWP{a$$K=Yvv;U@p3i>tmd$V#{i3%gUm$!`q>p z_!!rm#3-aIhMX3~P;}IOKI1t-)wfQPQG6hY9iLD4Uq_PR;v$lGatggePeJKo8j|DM zuwBQ3emgheHiuvxO-v^1RUMSm(txve;`G#K0Qn^H87yb9yiVvtVAxXnF*S_OB0_0( z`!pyyZ9r9BA{2w}Q%z@o!D*2o>|GUunLLj%N^}&z*VZD*av=3u4W$DqM^LnL3!25} zA!YeGIJnQ`yC42AJbaLb8Qi4mjs#k}@-gQBQzvbU5_G;)q@eExlp|%tcTQ~Zt#31b z&Qfsl#$hzY6(hhZ4KW#p+;cyT@)w)aB=bDDrkf)yg-;<(XnsZkue#`$?6G`` z8j-W@ATfG6tV%N|efTAMFtdp^>MW+#@tbI>KIfVB`~dZTkMUDx((AY~_>3C^FPl@S zREimR|KPqyh2U>6&dXf^X^3 z$p|~@DEEfBZv@Vdo{P2ahLrYVF8RHkN__=M@b##HV0|0!0{7sh)C$#`yjXki8lVnYD$eZNH7yg9h9o zF^o*MjiX0yLol#G4zpb?@Y*r}UT-3CPJ-(fU}bI_wss*p-* z7A8wY10Gx9W;zc6vw}IteFJPbb6?4R2z4Yn(`ID{q^??y*mt2=wW$U{VQ(<>%Vf$O zIhaQM&l&`~UBRCtNAcgg88r6H3sm>@qoQIR>NBsw@5{$g@Q}|?Cx5~U zu|{lIco}=<>_c{~Eh=hXq1(?B>D9&voXMHr<`>}6`wp_ya3?bc2gck(?6jrOvKS9vH)*;wxs0x?$e^`tpP|kwkihx&D>r_I z#@1R?*FS;NYb%;_TSRw18&mj+_c)oJ2=|aB==b+D_MZ&JrR$+MbfXiw=g%Tk`ZMn< zS3u#5u12OilBLP8-i1k1NCQ`5AZ~d;ovTzGKFeJ~Wj)!hrw6@#Ok&EcBj*_1Uj6wfjF@ zd3y{beG6>&34A{I2D<0CV|~y_eBLaFD^EGSdi+3AvuZ=n;s#t@nhl=~8OV(sMHNdg zKu#qZmRiyD^UX9$avwrvZcZqDq60g7AJm8ZfUn6@jQm~0`)VU0;p%}wesS>O^Y4|% zs_^e;26m-~kjKKG*ufdvQ=ZBa=joDPF_Gt=ICz$3;=N5MPD#0tw?hNG#>8P~??^g+ zMcWWlLk?c|2vfJUW^c#3F!EwKn9QQ!XcpqTesfEjr=^^4h?|! z$R>ZaT3+DM$-t_8aO_yg!;}Qw6^aj7RZm` zS&&cYx%UBj=g;HSNjZAQ`JNUNt01d4nJ$VM(CcBH|MSqKh2%rt*9AGf%9OIU3v2m% zEphJ!Jl>pvzr|rxws7YXuN4N*CWQJPCzS}kQ+r#DE{WOF%~M7ck)}YWlltIV{|G5< zHTV!Sj85$6M`yP6Cu80p(@6?PLH&O8*C@km$s3sP@eCe1bBsA4arf=n0uXecHiG^lX3^H%aQ}Nxw zw0!kvi0{duJw^I-ddMJ}riaemGWPt~w0 z=lhU}wxue&b_`ftfRKp}G-S3m&Z`;X($i6}7B|7yrmt*Bi5}M7YT=%;LMZA@g{J&4 z(mo=E42An_%qKA{xN(TC+AgQYHv%=^3S;{Y4#JmDW$Z~x5$m0+&iq#HV~?Jv;z?3F zXZ_70)ssgk{;-swy-HkI|K$>!__&0%a))VmQv{nETE$tUC)kUcLD*-V#?o4vIJ@=y^|gsGJQE#09&yEmGXL%BOi|FGb`DJ?oMxtp=W&1{oNCkt^Z!hfE(Ahs2zEIs78f(gm-_C*^E|)mA2v8q9*Y8h$dEwp<=Wk&V`$KC05iK(~`XftW!6Kd3}s!r(6@5c*Yf`;a7~q z8Wp@o=+5E}lFw|NZVTIgbUp4e&LO(k4XZ(WsBVfA{Ho6~W$l4jp}|>(DG%6)js|8D z@f$CLx%St^i~K@PFe@wGAMvhd1~D=0%KTiWdyg~Rm*lf~FQ&5nH`7={`)F#OYeYZR zzJ)@0JqvXzV->42SoZQbOvz3~=N1cmS@aK`MfXw4GlutXjD^U^1UGHR;mqCXX#PG8 zE1C*WIY9>swo@r>@&gQBJ&t`?f1YJ&9)JecBJ^`A>~3V^hv8SImF|OIQU<7qvxS9#cwtf@@m@D(aoLA z${%suU5VUme=$*DFZ1r(f}U+r7~{VK61uslC^o^=TRYHE?T2LJcdU5$UGDF^$DSJy z+9z(~&+Y@P;m*M9slV71BRl+cw8o#-L+tW*AKvf2&XiWAv4C}JnM_0;M!wsM$U7Do zVJ!^{_L8N)+|PY`7TDKO$c}}XAo<%X_FzycOZ}e8if731JpO3xX;MUqNeEQ#zs9#S z#h9dF$BO={pgHRbTR2k(1B&fX{uxlt3B|kt>A1$f!xK#pAi`@uc9}Y1sdhIT^uZL@ z`;WskDw9djZepcAe-_6puaTeUnc~HNoic?QFVWwmzZb@-oXF(3L zE~@1j_Z?^sJb~xe7X3 z2*wk(dHOU=<9AHnV}|du;b>l;%H}30!C=&RtkjC+djum${&Oel>MY?t$%@u*euW5# z)0*L*nSNF)8@l@_3m9!m)x90K%^6I;FCD||yHD`2>H@EwbC^u@SyBw=f)6ZU!Y`mf+DJO6P27Cx?LpBlH~_-1QddK83j#$!>?vjs5t z2iavIxcW@M7R?OAmgKPwd!=D~3AoL@bBk&>W8&XEC_K+IIUnZ1Ih)@hMoHkvr~z2# zUcD-0GZi3Wo^TO^j)!sn?qOK-OxWl{DEBeZYE&aj9hmVCB&%)^-7c>8!$9cuc^y|uA7$qgqpsV+A z>&jg$y(3Fbf1ct~?q|HN8BV#j+@oe6M(3{{rLypH6z3P?T~aab72KxzJ%v=b^D#ZS zd>O+oaTei{Yp@v|Lc8nIs6;20H21sFVB=u&OA4g%BYsoHPbtBXPn>nG$p{a8q=jL} z#D%d%8F-LUiA9%-A$2$xfw4z0bVNC{+TCf5xi=YY}46|pu-)VWuWa$cL!p6Dr*96Fo!c5bAO%?Ie%k6>EqwvEgc zyXe8JKcsi+GhLi6CuI9h6qaQu2`nrPx9g6hP3J7WMeoH8v2$3>-^SMoFR5qn4YK3= zp9?+>7g}Cw3JtdUg5S1M40?PHcWl$J{Yf;f=qsZ-lYIJ;S`KUTqj+KR8^Pv}F*W5Y z_7CV!zw%{;vsV-Z&0t00U(EsYu^JF$Xct^Da8}e8I|DH8A-uPfMmI zU_#{;49~lc_otl%@!3m-qI72=Vr@8I919^y)o}7q9YdFD@KoMf|q0$ zuEg;yviNX5#~4QQ+X6^weGpyKi>Bjzet#hQD>8islH@fFx==QP>}9$4X{J1_`yfpc z?Y#H!Mvbx`NmAOw@zm{*M_(EDWaP`n$-d@;%*~l0_^ofwo1xp#|4{gc-+`1&c9S!q&sw z*~@>=oeqcaYu8iy`tUo=j_V{P<76u9zD&bguhX#Ld6fK(=#TtwspvvR&nOoP0csDZ>N37w|b{fgVlO?yIXZ;9QId4XD?r)SYXvXnPB^b%w75&?u!L?adkliXL)EF8FCp-tx!@fZ@ zGuDWl+MD^j@(Y?3o}*4RioTXCCi|g5wCjWnUEcBwZ#aM8?wflsc*&oEuBVu5n1TP& zufsp_3Q9b$<3v;nuJil*C_#*FZ~lh{L5}Y7pB=$7`!)N->5^Y0&IPn!rd5A32snWY zyg#@u{0_!%$wqx%9QMi=z)-6lQf}|y>F|=fhwbUNiw}JskxA_{y5KCMK*@Y&a_HM2 zVf_LbVSs3&;KxcZQ@j)}`Tp}{p#hcg^5mT@MzT7p!s?x3Lb8{Uux*kg_qj>Z*Snf@ zC;A!Q#}%M}b8&Ph7r-h09AZabLHDS;RP}7Iu=c$IXVjd>q0&2Ot^AHv^LNs$-axv0 zK8}7rOeEu!Bh*hmj6|*>Flt+mwgY^BQHF>0=(|ebFRnFqN#T`Tx>Ei@y!!BkQH>gwqu!CKeFEu1aEmC zY-6>MIlT^VzlS40{T^hu^ssG<&9I^FIl|^0WmV&Ev;N6LV7Bi$)@^$V&2`qaXa7nl z8B}1&Ild>D?2d$nJNU()g(4SC^w*h%kM3vT7oQEwtf!F4slr?C$4gN*fS34Lcyy+) z8ut$7_hT4(y!rm!!Pzj~6a?+7ibz;B5~Ig_#_cCR*p{td&}4W7S}#@c)F2!?za-*u zr5oJl9)h{wacuujm)Aowkejmt3pRTpQYsX)zvaL~Z6iX)F2Ucx)YD9CC@_#Cb_8a0bX zpIt*gB_FZ{pZL%FJ^_#Re`Sed2>Vtpz!dptDE&N-egS-LXUzW}*6!r}pEQ_09!N~V zfi@dE&>9I1xZYd`6-K!IO9PEP6FGm{4r`Vkg7WXnXxDnk`2^KWb;UC_Swaf~)yi3c zWH6LQ)nMV$Pdtpz-G_%rlu)3W#syUS@EZH>@rN}HmxUy+na7otV$|Csm~5}WRt>(){>@ZE zl)EDodA@3UX&cltCL;e}BJ%##Kwa|-%ei8OtHXD5r?ER8>KY>J^G+->nShfc-T|qT z$fI}}Hm0TH)kiUk`od?L_Z2zRH4-+}vyflpjPMI5Fna6+)cxh(bLlFqFWQ5i{&TVH z;sNYxe8i^3jli3DONiZ>gO$$KD5>Sn*=L22Yrl*l29TpL7tYSgs7boC~bsvIrq@ew1 zI&Pod4zXNInD<#@&G=ME6ivm_YoUly3&9b&0M2hpz~v#_>6^P9+P1EImp&7b)(2R> zuu67stOTy^IfI1qcert3ApNXks!-_UhrjmGnEE>r zr^4dd4)uICZrTy}d{_+^hjdIZdx$UghuF0Ixy*fcE89CDoedu1$Oet*Wd$=P<6kIu zLU!fxo$68Wtvicx3Uw%2eiAW3BdBm*7@e{hNgMT)IHPqS`TcdHOsn~{Zk9Q@t%`)= zn;eujy5U}06JNNH;u9qLz1MNvmctu}{plX;)S@i@f?j-+>A3~1o>(UkX6 zobDXWfzhpc(Qv;~TGCfSh0FKR4(BK`i;kr$ zik^sXn+EN_^YGMs169Ym(V1m#R3NP?O!!Y#(6a9@^u8TNii5gQai%|IEp0&4NEr&6 zSb>X2u2HXg3Y}YgnyyY-B)BA)2ntzF!l7O#k_vI8AxV~0^GThmU+Pd}u_~1;T1C%v zA}KXs4#ie!lkZiw@#Ninz}E2 zB&pmh+{=2Op1lIiE*ekvvo+`v&!XRnzKbJ<{5$fu9mW^=zI<;QKDkb#_rs>r^R1KU zYUMDxAg@HfYPxYk<1JOM`$+9NPpG;mkP5q;DY(>?EYEt-)4M997dMi0R;I(2pQC?D zd{J>efE4YV$;Q!!h5vsF=HI~7E3L4R&xCe{J0#!OWd-|NBXcgu>9A>d#eScWPX!e z_Z_DTg_o#!*#??dpH3@3PNBV=_wjQxzaRU3g7%^TRPXN%lu2R z=U@e*44X}XJgdHx^ET7-+j)LLg*(S3 zC~V6kB&2fwuGa@#``3vjTxUyYa_4FCFh5f%00C9as%l2`$wpJ<4&ovlW3Bx z1LYj3q=eurq_DD#3ib`B2lx1Q&O?R%^KXImjxNp$Ohb3VeF{5$nWo~@hT_!Xj*nnm{&cj->li0gH8^d}=P+*=mEwL7($={SocHank(k@LtDZOZ` zZpEniJd-)di&E@uNlA?}u=z#t$uQ0}X%nY~mYUSI`X@4E`!F-XpU#HZ(u-RwDQrSN zVUb4zStmD8b$14-@mXTu#J!XqYJgi}Hkj7B4ONy0;Qb&S2bSMKdsQ`BUVlO5a($i& zwx$a+7gOMT&RQ1s<4;~Cjy_VMdTAN@Z9bg>>UC)7I1$E$W}<7a64}P8Vs=_Iwnr3T zwC^pv&gE{r)z1;wIECu@Z1(i55b8WKi%M7VecOxSbSp%e%BQFkJ8DN>(Q&xzkq#Y| zI`q$pg~9Zrs0%OS08|UwBArjaU)PgW!bNwk*OG!%8O{6#7NR>oh}vVU>=5BCTKFQfzB8Dx_(4o;?37@LuYz7K_1mY;(6%?U7; zm!gjg6=`@h-<$V$!`x9x5SA4oFI<6|`rW3Kb*Ct8p*w8CUSO5MJG755M@eKHEN`8J zY3LfncE5xsz9U3>1xojn;lu81+&Ewf=jKMp?Qg*9{GHgLe-+W29$`g*9c1ga!c!#? zZ4V=`R8NI`77iniPyD?b^%#+fEzp7!7M(kTpYycn%xM4I!X-GpFWEAn-*ZGJNB4mUl$w>AJ zKO>6LRGQkQT^eSntdfQ#L{=1)Rmd!|^4;HGfY}59?Q?a!{R9I5dSPr?;vxf1oCn`PQ!bK zBIeu-ocJ^q_BY5DI_H=8=7%;CFMTeMNZ|x6l5AsJ0C>R$-XP$;4m2rcE&@+O%@Tm5@?z& zzxQ3`=bz=Zl#0T${R2BVNT|qbY7~qKTnr*tw_lNvE)Ukt?7II6`7 z`cQh3ehqv>Lx$+!eZgQ@F9!vGh@&-0w`huQIAyzL(fXoG)X#JS-3^MN%6(bnbFG68 z@k~`*nHgRbU*!AyGgLfl0RK&-QO#xEk6y*u1bl~U_24qCl<19x50dH9e=^)RDh2n1 zA-HxX1~#p`@g3UecwvV#&f74`=9REzzcp>sP~%Py3D)sXfqm6k!IHY4u(Pi-gzHja z!txP5-1{m`6{ZTbOg@3+)^xJPIS0_MS2ioXF`qqX?nT*#zlFE1$@KV$Hpz4>XIC<9 z=*6S)G`b_5Wx8uH)jzUq?3y>iHqO6P9aqHOeoCf;fw|;tHHwP&whEJ^lGsOwF!u3% zUzS^YKnPs&Nys$1A#`M^v4rnxY{r`sA@pLfaOvhN!NEL9u*m=J?C_V z9erzrf`7l+9iQ{;mk;0RJm14gH>t5*-2YZF{S51opUbv+@Eor}sqiS_mQXDJQCPps zf^~1zWirR_31=5XvC0Pl!jxfw)M7?L|7|@&zdC=x&ox(29Vf;3ubh3{r^Y&xt=S9L z45q4|L1CFZgHYg4Bi0OIHliHM-dV=9zId`5s247|8M1k^Bbfd_W#;*B1~Xr^pIs^Q zW#&UCvGC}=G(zHw;B+pR8Q4yzqUR&&V!A&o|Mx+-bU~i27~#wgJQ+pPItP)_;%}_P zyh>OnEg|eksu!NTj$}=9qgYMpMA|F;Tj)L9iKTKbSF+AOVfg9e!d})axLupVtYrPz z?}tvz)o47c2z@pF^2!#Wq~tOSS-VqMIKP5zdC(wqoRVP6uk>cu{|jPa`}$ z*3(Rw-rpNS^T7?mZPRZ;qs|GTD{h()C58x_ca30wdktZ)&jt!%^^e*5-eXz9g9bKQ z(^p7-Q6v0)^hR)%(O^6371^4%#lnyG5lp5$iS57iQ26xzve0p1Fq=PZ22;WeTo z#$VV__)LgSXce}8?_`gfCeZE~M%;&XM7XX$m`!N*5vI9HF&oWjAtu_HDV1ztKi%I8 zA8mKDwC=tnRw+|yA5GeP^{}u+>b!7r(=EX@SB=H4$`K}eitP1(!&HC!D3QZ6roSeU zIeBEzx?V%skO~#L>(R_AJeAnJ41bpKB8<(N>d3s!E(oud%xAlsp9>P*Q#s$NSy)@x zpY=?ZX6Is5gu%WZ%(kXC8$l()*J;E?S4gpK10#gQ`dXn#r5Ee_P+)78&1FL_pA#BQ zLj=R0A#7`q3^O?+u$1saf>+WQVSMmFmgk|uie-MWRodrRzt`NARUSo0^F1in;1oUB zzK)i9&7jy}>J*mqmbLxqMTvY%t~{ujjd?Pfu4XGy`)(yF)xX7-#ut&+8ZWYI2qMeu zqolDRfbw>`Q-98%vG|xqwb2Ji*LE8%KDUU%4>VH5IBApzerI3)rm(9w0-4{#5V~@p zoMdKwAi08S3Yu|&Lf%WV-T}&_RyC9U@D7so3p47tU`q}~sq{TSk2a37q;&@#Qrcz% z+V3ev>%s=pILk$-^w5Vc=MP_sD`Yz>FR}2QpM)=bH+Uyjj$BG}S%@KWS6D^OPqwzZn9Z7dot+!i#lm@(-SR>(-JfPfadUW|JSUUn z=kWI%ts!rzG+O>(IC;EkqzP%aXjN(wy`FxNbL9pgP*wt#gHO}Xj5Tz%RuKz>qgn4- zDRLfdMlF1&W@)#9t_0p91&dQOZvQ*#XJte+f%n+Wj2mpGVKrNrr%JYiHqp@1T3Yci zmS#UbL9RD5$zaDMGH?CEz7>z81B36#mcRFNh+s(9>PNRNFeW_wYIXger zoPM;8qY=e=G{~?Q-T4^Le%!y$&gw|hxQV)0mT!Rjo94ro=Tv6f+tPUZzf^usg0{9D zX9hLTS?ZWLgaw)%0-7T3^b8#U8 z3$)36;W7#zIg^}vBuTm1n>C*O&ffmfqY?5lWOYN98q0=}de(B9dt3!kZ#Ggd<-5#Z z_aRC4?Sn|NMO&4J(!u~E*2ei ziyi+&%xqiVv)xvj7Bq!4 zQ})q?yLsFtlSEzG{mA{64Xrm`N>_J3p@GWWvC@)*l_OTt%~@N>*vyL#om|7SSv;>) zFH73icPZEJDg8K9%h@+02<>C&w#6N$l9|Y2f81otJ#tyKXAo1H-^r|BXS26ULTMtO z*Jg|epfR^PS*5uV-JB~&H$zrZ_+(Yu+^kJcRW{L|o$0iF;t;AnUd86NtJ0Lg6X@lk zNPgbjMrRr>(3|bQXu+^By0FxPmR>srDee^-HX{v|X~l?KdKwnqC-9JG9UJ>t;QXCw zXwWnkOG;*mb($JtcaM$uE5urCp0ZLbm(~*fHYkaY{(gdwLK7-~-G;(v?y`Dz67QPw z@yRI*x}4{?q%af%`-kGk?Q`hUor?PH+;y!bh!LMO#DU{A#O$V57@ypQ<6~6B^&fuF zf-V^>UL=PJjiSZn?c!1GOE`A#FeF>5Aa!jmwq8hu#|}r>=9J>s$`WXW z7UNRTbzJ7Gx|hl2c;wL==U%=htNBuB|9T2N&72kDaT_n~$}!0#7R;&?g%Vbxlb@xy z{Gz(3b&azWltb|Dc?Kj$FNNc&37BR-3n@*x+{cgxoqjP$T9*XV19?cz&f#^dTxu2=_MZgnoP+8MOU!8UMeb`CY-Ll$_>()uK2h_;+qFL8!PJ$aRr@Yc z^De)y@O)%akS;9hc0#2t3)=lspmDO7_~Wd!=s2+pb2e{*zx!ScQ9Opt&iwBDVL0v- z^h02i4}uMRQ7W?u<3}ipv|Ux)BB>(AN~iJO`!*cPTaHGvQfLL8M4wT)sJ}H-e45fr zRJB$XtNnR4|F}28_6DJ0yB&Dt6z2UVq24b7y#otT^f(Q*lMiFS;4p-o@W*DYjTqtK ziUo$-@Ndy>41Ktj-+C5v_Vp_K>a`Ch3)ew4Z<6@;f}ZFhZz{$q2VvW?V7zbe!>g(j zd@hxQyL#7f!s#oH_WlQxz`o+gp*rH{>ssRX*zw}ODf4k^|7=|Q>4ww|iHO~jgqbFJ z=o7hAjP#o&s?JmwUCg3j^fDLIhejfHE%&kWPYb^h;HN=dg~_=Aq+g)S_jqS*4D3SxU0jJ#y^dQ*-wG zlfWF#JrsVw)M3kQHVA)dC`hA|&my6ynZku-=ml*yH;XnZ2Q}kn^uVn4qn})~B}7 zhEd-rKJP8NcUY0G`peMislhDFUX?95H;8TXp3HvQ5DPXQ$A(;($bF%c>7#`MRjt#e z)hZL&&RPd%>9Lu~*=(e_-Cb06^9C)vr%7E`KuL|Nbl(0jyE}U)o8W$t#ksr_%zYS3 z4}BxZg%2hBZLe6ZeFnS#z?yuj){}Fx8C@Cmjm7V`qdwaUn5wfT+h@I%ReDwmev{gT z-G3z5gO^1@pR-Zyhs{p5Hd~*4DH%zDoBc?CPcw5(lVuKJBiO82HwE3}<$~w0KSI*= zBw@-jZQ86RLj!N6ve`jWbRxi+m~Rc+A{vvYNi5B;`OG9aFRCso?c+58GgQPN=Hq z-duUk#0=njG4&ua_Oqb-Ng*_ve|FM$0@%f+KZP;xq$pm$4I_yude^dCDjLyK0g>O)UgFC)A# zBCDQd^u$+SIow0Fm#k31 zt)vS}tQyL0tqf;{tKYDaHYxJ?Rl-U#4Jd4~25tVpe=o*n6e~BGW@O8gcc&C*i9``? z4W#SwjclZHBl~t`I*Z%0g-p078H!i8wQ=o;IW!X-vA}i$x>Bd3Z6kNaEV)4K4w5M1ed4201~AiF zj8)1{Dc)ot9u3pNitY!r&?brIZg1wiOy0}r@P+k;6l{N)j}ajm_)q#IG^fsk-V<8{ zqz0f+b~G04Q-e*aI?7u$ka41i?C)`J9<2ODYm4Si9&!Ay)eEhG924B=cI z7oH39riSZ>$TXEd2UbK<|I9?HPdG?JjUs7Kx~BNgNkP=uVk<75kxb$9?oiTk-kGh6 zA=yR#bZ_`kiW)bLbL>^{%WXKaCtl^*@iuzC#Q@V|mGOj)z~D1Ou-DEWZH4;~Ar*j% zt1&28{~uyC%AmV^Huje(!HRnaioF(-w5uHjxpGf<>t60tKZryZKRB9~!(r`BIBvg= z!fzhbF#RaimOr7PtF58s%H zVx?OzvF(ZqYQx9lR`m$jmlTF>9dZH*$bymgWy=3Swj(qx{wJ3>DslBp$H0_v@wNsjY` z_YKP68IB6N_$-{p_qjqBXWplSaeXm!(qoLx`itLR?_uA(g>-Qs&whrwkm7eY%zXEQ zex9|%kVR){j(Z{X)$Wbe)t@+vLJG)j;qQSG8q>qyg+?jdx7H7S&fS!}W;rYu9wXCt zd~T5H&-1$KVi*ctx|$ z4ZzaPy>Wg1Xe5sGp;@WLRONb{beuEDHrRzE+h$T{`hN1X2q4XpDLk9S{bx2>81=yf zAD#Y@?JimPEi9#7Z_4O5?-WgPQb#WDl+7PQShH{+bh2ErYLgy6S21WL`;u{o89fhj zp`EvjiJfgB@xM0uYV?W*C6>_WoDx#s<3$_1rjY8jqqOQ^5fYTkpxpWgnb}5&ej^~# z#2#9+J89cY&U*XsohrwqqO>>}QFmkURP7d(4!lZ3_WY#=$Lln8K{Cyg^rOihfnm7oiNeFkCdf$W95!k7BSia5t|iIaefLe>voe&pA4GrTu&Fad()58!StVd zH!a?pMYD}Vx%)=}?*nt#Ks9A*=+U6@ zpF~=;$&S1Wn#k?FF1FRmpsDr>-F7UcDQhz5Q0_#EKlEI1Ic-DvDJE$DF^@Gr;yq1| zCOS}Wh6kT)acTEjdXlh-eou5FAO2kAS}Kx`Wb$mRbyEfvgs#VY2_`Q7DC)6re*kAvT`=s+pY zYn_ZC+v;=*s>|ei#2)J3e3s6S^Q1B1r>IFej8?zn`M{K5Y6kzloJb_YFST@1dp#v4 z5{*x@B>8b~Xzw}=TpOmya~4^&bZicdHwz;FdP7(yNnre%dMaNYhTHL*pc3l`vs0e% z&^ipAUfkEBG=hdljG?SC189gJ=R6-?Lnl)wP`9lH>CY}>lFbJ>t2h`PeFXF`nvX!X z6RJ7Kkoxfeihs0{LschvcJa*Kt8=WR-kgbie^RVbhvwjWP&)GnSN%M3w|X(MJN=Qr zat;;kOCZz46BM)-)cKzQ6~8HDUW$uRa5fS)Ict%+LzCRxf3WJe=NR>DVqvFTnEcT= zR(Y+PDNn6rUs|%*vC*yEqj8+(HNT@ar3%&++@F?vePj2mve=*Rnv^}XgIRi>BGLF6 zSscGXL;1{bO6Pkj8T^K}j{T3$=|3mkana<%c?SDUWXM~zpn@Shr#L-~20!Kgkw<#G zYuCYE>&sL8eoJ!WGgzyU{U~mF4qH*GO&9&L*ymeuXcE;B^kN+n-tlahqZt{mzQw$y z-N|l@DRsXYLJxxPayD`t?--t>dy_s;*E4la_imvv8(y;(Wi`5+c$KZ%Q%Xmk4X3d4 z!{~b(f4*LcK-Y=@$P`~9Q_JI|?RJz}dAENj=k+gWI!WE%l#p@b8twnHn>I=cbg7fy zAEHOoMD5YkT5E*R^`q$En9($%ppChh^KO`TFR~pR&R#aWXC;eoF~i$VY<*`cQ)s?R z7mWMUv$Pjf=Pb}V_d_&teL86mJj=$9dracSf2^%Ihn7Tt5}xSG!QZYg>XMRanw%#E zsSZW|KrhT{`b@8`WMS}>5SaU4Lec<5k~SPnyI*c0+oca_ucR89B8L;nWcu)9dD&+Ev{d<1U0RFQ3#G&GwhLfTRu`n#E`SKf|eS$Y7dq z+nqBKc~;LTKy22~6en(R5T8EGqi-rjG^#L*#zdTysHRSxKk5|nu*#0IT4W4(ei<1$KKT|@Br!!)?R8_hY* z^IyC-q86~6>`ms;`Ur2VesU0ZUdCf_|I0|~DM01buc#6aQ2w1{?sY39|C@Gv4n7yt z>lyS0tRdH1J4trLX_`N`fnL==q*`lfBvpriVnQ&yA|J0+AJL--dAz>S4=cx8QPP2x zRJ+EL`aZf#5ng@K{7MI#+Jg9Jaf%El#nYDdAb5SgjR%r^-WdIn()ta7*GYZcFEF94 zoFNfYSxm1NHjwgIW!y+t$B+$?)UPXz%2(7;%go)>mMDdFiWcyYY2cikuOvQwOgr3~ zSrwnrOG!IY#Z^;uDh@$j0B6AYji%lU%;|2+EV^}%(XuUW6jjnn&mZ)`+n#V3zCVhQ zCYh8U-9ni@()iL*LM0E<=~_($8FC(Der#`S_LM_R_Hxc^Vf=hpMdl?Qq_QQI_a48~ zlFiqVGVn6G4m^ik#V%SdX44$iJW@YhO`_3NvY%>?Q3125?e99WD!E8IB3r4aq>b+M z;_j@mhiTr=SQ^qDN3O*YWUwlWyB!*7=kZudJ{Coa-0i|aGql5gJ}s8^qWCF()bKu- zj87*}JO9mAS@zKMrL%A|=QvF{RX}Ne+-u~VM-n$5Q`XT3q#|Vm*#K?$c+N(S+A6Gy zkHgo8V5c3G)`Q%J`KdXzjkN8byyNs}OTo)y}=%OdVfyT|sqd!X5Xo2n! zTK+SbYJ5CtL0c~A*B&IJqv3SSB8cMnE{0+r(b$)Bq0KR+a>8hA|0oBYe+oEa`Gr2o z8KI(cD4feoFl+K_Qdq$MjL&1C_;)Ae@5`hUYa_{T@)dGb?jrYC&SUh^B)JdMD9|C3 zKIqKi&WvHS%g>zTls4nsj6$4q%)kcjFZ*+M1zvEc-|o^@8t8DAF1}NOfqyVG?Jwb# z*%L_etnJf{!89(mhm;Ku(wwYS)bYNCUYX_5)&72DquM|-#~s0=adAlYOoqPwO?qMV zm@eGxg;-tASMM_n6J-YBd*=j1309bxw*Y^RZ6F)#NQzQ$q;GC1w6^CW_X|CyMX4hY zXR`t=`ERlDVgPy>&4pU$DkOPK$L*g}P}gq|PB&ITq3Iktc13WXP7B(Pa{p`CS^Trt z7rUCrhzD}tqhnKd_N-FZK}AYo#%wN z|H4ug-dR~wN5)%A>B_ae;-UVN#KCio#OSm3qN;+n_-k1=-dLsJq2?Y;E8B;tZITFz zzez(9E|O-}B+>JQ67QQdpz~b~#a??%PwUgj=PJ+8R$d?r{(KZxc9U8{DKYP>6yZ>f zADZ5H8?hIzC(qL5N&Nkqo=?9ge5BV+-|3gB4Az8(Lr0!_BAu2Y;)9-86hA;5{;;pu zTA7A|*HM^+9r%$v3x#L&5yTymr@pnqeerdy79Um(-gjbV%8(0Fe< z22ERuyDuk-J913LUFTHA0cSF?X;BOYr3YcyCnG#uY>pE%hheIk5%0E{i_dQAiebS? zxEVEv&&maOFHfaUQKxB(^lf5KJISFUhH82pqhrZR z)&Ce=w;aRLwZ1^lJm|)*=e@1~JWox9o*i&yL_gg8`WjPj1FO=t<|B>8b%UPW^x@^$V!OSRp7CC5?lU1rL_7++&hZCEa zVwMr@os&-cOzerR9fF+8O6{?g${m33JST)?$s#Mr`Hph zVnv(q@^v4^v>eDe{S{SCe9p6zZ`ok|KWtfeD}61rVbawjSxU4Q^XU#|eA>nuPCjLU zvI1ZDM)V>a9Ou&_yyOhJfZw;@4guB9}V&))UM-=;p{ z8LZ>pMs{at5*u%%Mg1RyaqjjDW^Hwql@QjCm9RaVSFqvf(>N=$AFa61!nS2O(EtSx+BUg_{XTPo6~}&Hy9fSc zjr5KT`&?wJ8=Tk&1!HmyC}wKQHZh;|&sYF^#I!DPdlluq$z;!;dND-)WuIOQ#g_6!dRrSy_k;B4r`=svzowk+Yflv9k2eXOF{9bp zdp0cV>j7pND$@H4er%21TS0E-G-h?Hh`s8b&vra8VtEa*OhbPNP0dJUUNPmuyWU5H zS1mSd#++1EYx9t4?Ve9tJ_fTjnd!`a-yzca{X>YqxP>V?8I#k*aqO;Lxp3oq8+&iw ziDGe3+(JyJ`0xgWE)7yR_Q|K z)*;7e&@7fbA(*wq-(i7?`t)iZIS;4D*#3&z|Z%XGQwGX=A+|+x?-4`^X)cOYhn2 z*nbPzfRZw%W|7WXo~>Xzbq=$4iw?3E8QjCS@Gf(6tY&*mFR+4^i)46;=(}Sao3~P) zl>a_vsc)aMLn|gxYOFfxP8-MB;fbUft4+za3UuWmA@jN;v%58k#WxqQ5n)x#Kx!vz zkY395zPCDeKtumHGBv!~%YVlb)vnpCS7& zmsU4czUTt$^*4(q4PH!21IN-RMHBYTL4!$08#Aqrfjn0h#fH@KyvZC}_Tc+bHaz4n z^Xl-XZ}AmuXZj^pctNBpFEOt~;zrs<7@o!*ftK!%P?K#Zxk}B$k7{aSJXZ>il)9JM-)IPb8CR7~3 z{BGVq`I`)l3!~v6B|`suAoN13aN%h^1}dFLT9Gm~Sv6BrrxN;%%s@wb9yV%4gXsi9 zrj+l1&!+Is@H;&9>4IVJIyky0iZa)SitS6k!>mpYQLD{y_}XmlVT+_9t75vAUPPJa zeW3Ps8SZHX;eBl$eMmn`^7C%e#xgT8Thm@__#=obvMXtQLK(@WwNgN3HU+OgL&L7d zQ#*H`n<(eu^xmsz$PN%Y$M}n9V-AWgYxyocxRX|ouA!AZHRKiYh&1AB>5sYrW=$0E zyG;X^k?pNbOo#S*WjyHF?#x6q`SFiY0vpu>YidMUXIfi4cA+VZy#Df_WEi}il2a$ zU#xJdVFAXa09ilNsDN|$Cak(bv#%egnNv#XN8U}kv(`pjWxG&ZEhmb$pLamz1!uE8 z-jAK_*4Ws&m}jNAZ!+@&1ud>4Z=U)7d#N|KY9uMH&x^II-W~@B8ZOm zf+&3AUYk69jF_s2`MG*1<31?`sc!litB&?u?q1RzhyUVdVw!v-#1T*NVEa34T;GZV z{Xf9gy9s+8b;Y^AHN@OuBgF4tBPqryf;?m|(4VG%JXeUYtkF@-RFhfop`|2x{ z0$k93$O_B*a6dWk+K8RsDKhLVy=%(BWyRyD*jkPB>yPPQW;$7<*VC%xC|Y?UkoyPI zC_XBiQmK@b?ej@5=?~3V|BL2$|01h(jbyQkXa2Ps$>cI;?Icu_4ar#0$IB=4@8c&d)+$CDS=qPOJym0GU7~Wj@Mly?K zarzzi2L4N>kwvxCzc)YE4SLCWrWYvT|GcfbR4B=YW8?WEcxKK+`@m69o9cj$)0^q=l1AbT`{CkeOGEK=d>K{hNT6(_986z_Q&CPj&v#v> ztL^>7%dX#1w^d#&%o4Cnbpo7RhGJ~OWmCZ&I^@D=IHLL;6Pfq}X$qw*Ji})A^h|m|jC#2L((`vcZ9K zoc;IG7)vg3%jV`b`gCg&{)_0Mo*!>$w$FX4jrc&{qLi^ab{y=?cG9`KO7!U39om(| zp~uHP@k}!n2iwB%qxL_1Y>2>h-X*E+mVm{PE~?8Jitu}deT8Wxp5cIi4d=+YJcgDvW>V~TAD(GcqBgrog8LsjD{qK+ zuR+-Kz!=6$Tp*KZiamc8V3*u_EIqdn(G!7+bZe|O_rnX{o1_6h8Y!=jl-M+SqL)Hh zZJ`vK&`vzXiOK!EaddDRc1=#eujEr0=$J~XYmSo1>t*!tvnu}ZyFmW_<>h|l5` z`Z?_zU4O(o{}(ya=4dM|T^+$W5y@!$FB5yLv+3~jHkzZP3Hx5#$n8-ujkUi@jrvX0 z=b{FbSM|o;AAV$TeGSb@n?O~|eX;(>HMna0fI?9*Rd%)0q}ggv)k&t}z7aIM_jww5 zvoFm5dqrEeU8i-my#JAuPueS*iLQ84|MAPoS;`VCeS6`{yc=Y?_9Bh(cOu!Ti|8@m z`7W<|L!p(ws9;11ZM=Jk-uEr084b7TcAY9B-j~oq-Y**VV;Fd?0?reX@P2+ErI=4@b1oNyQA<4)5b8Z9%{M22`v}@(VVBPw5B$SGY&)P zdS@UtCp*yItRBbtD7BPt zPwL5K3Ctd22pzu&D$6P**}3UtCx4v6?2eqv)nLq!@LDR`Fc%qb1KMtj2^!4&Fn` zIvI4g-iIoAcXndJR^Ia;gfm6`V7AHx+QT=a|0M3bOtZu2rRAIl%y(@YE9uSNDfqm9 zDAqpK#;)L3WWDVpSugoa;WAqAd=Shwj6``;2==y0!O?0TO;349 z#uhE)^PxBPH<`d~M*$fO{zfM@5kP?Jt3WresI2i7|d~nxwF?t`EhtJCwBD8%ig0~v-UhyePo6t=4Q@bhR$!e6# zZO6q}-WfcRg)J69PkI7W?4UMX6Bg>Kua`HGyZ;TS?yW2j@80UeGKitj2IlRHkdWo_pi*X}+ z6y&ey;dGD+4*OS;=>hJ%`TCyThd7|;eFlcZ6RkZWe>dA;dI_IrE~%#C%b`^8sfubd zha%^6EoaV6MBoy2G`n-&d)QSv$9vVG+^Oj8lL_D9Z>Vo#7yX%hnU;Cnq}Pwn(Bvs8 z^hE|J&|8YaP=BoS*TIdLIY1zQ{G|@GOD(;=d`J**x8CZamC+6UR+GvE$ z=Ps8$;e584h7ED$FwFI({ad+D&M1OZlQ{?Zm_8&*)zCBB48ykCV&`LP%zDnBOV^s| zp!q%0w++I8lU`5|w!!3995wIc{H0@&lxiV^Re2pWc)tpG1rfDzlaXO$j!Ttm$^6A- zY8Y^q?m4%Ub)F&oT;5atkcD{NFd7|c%Bb%!MSJ`Z{?1dum5nn|Yd9arBFs>AU?OI% zvqY7_Tr9p9jl1ttk#yq#Hb>os#nC#nB)x{GMl?OV=R!-zFQ7kH|8W+G8VqxIM}WU? z$?-g0{TV{H(oUe=Mhlv}~V6 zx4|l>9cpL|Qt z=Jt>(&$do%8igmrdT943Z;YQ3i8sR!qOL#p-dL=~*GK*EM&l&Csef<&!(eGxHuDQNK=$)1T1z3Dsmcu$VTieo0aj8|Xi`W-=R? zfvC8A-j@zVAA?~0&RLH7y$o+y__WSewU@&fOe#>{_oN?6E7fvr-@N!=q!a9$l zUE&Cow&vpCtPnW89*v64e9UC1f?>7NxDnP(yK@;*Irns}(hthx|IVGp@<`qEkM|z= zF5AA5s^_(vtZgw+*N{5OoJRf7`Wi!qnAB&q73(&63okc6h zKqAi=eR*bE<@0Hb98`dt=Tq2GI zTteWfR%i_f!4}sjWNvMQ@!IonQ+|&n8=pdPa2s-eD2V||gT?5p{QvsoH7xA^5F2e& z5VB(#u5Q>2-H@G#+HCj((vrjG10Q(@A*4api!+)XPb9t-Ou9?0yAg9`^B`rl~m z+*V9~HvghLjSn>St1Iftli?KG?Q{h+Ot zeUTEbii0{lqwr-mhJRj#uOlk)bVLqJW4w_4EfiY1M__fo0y%4&(7Nq8wxvG6&-FK9 zDfI)R|3slIs|Fdr9->)N5_&d6Fy7@i^-8`BXWtt*qhF0~4+(K^qM?{}eWdt(oCcR>lhXn2?kb+<%(DGbLte5jXHpdea55ReJwwIJT3&<&bLtL z{RO)NUc+PJNb!;UWO1so9u%KXgS+H(e$Ujy&&`vuam{p$5?A6_wi6m94x!-rKKN!{ z#B<}T5Mu)Q=iCk5?f!6-|;LeewjaJSPCo(qz&(Z31Q_c?+!DrtD*Upk{Cg9WpB z7HLB^ol0$|n}16v&Fe3%-c(2K1?%yrJ^&LO+|lIwl*Fwaq`FlWa*F?`S05v|ESQXI z8Ef!#H2;q9Jok*I9++?JEzZ*FEgGj9CJW=hDyBIyT7+vs!C`E!~c@Qk|Gy`VvVj={Ir3B*b89QxNe5PYrp z=d>DjJY!Wl6wn&xg6`=4{v3dhY>Lx03_j2`)sHhSozVR|39XP>16&#!bkM;T=) zHIRG%3JKj`!IX`0E@B~GY@G#z_VxHY;wUz(&c@_heEzb`7rQwFn++I?290IBi@?9< z%MZh@`U7UWx1v{xoLKYSSPbwm5tWz{PrkLKV5;%AGX3C4)}ih13Tz*ie=mA&zB3u0Le|Il04^doG$=P9XjD0h%|@i=1C&(72uyHvRTVw#=c5{o=dx zw)7jgS6v7(uoOoZbRggEC+a$n(4P68^q;)YDZ)L;9`2`C$x)s~4irk$b4Vz9l7|l0Fi`er~Yk3!S9lNtEfqnCsf&YFbp?Z8g>cZBO=KDxeepg8Y`UO%~ zE_aI`R6uv*K6XMokWKYD&SuPr;GP&4?zOBU^$&&g%}5njHW^|1$XCp8rYaqLH;R(4 zDWYoeM)XJ?fKr|#)9iJXIUO%y|Kv6zRcbkwb0**ct59sJD1hVNVpy~%p~|fm=e1?T zm(%BxZ$Kb<)=!{=UfjX-%Z`-of@!q(aB`N7CYzFMdim`%a&_Lq`dDAFf8l;;oNfbN z{=$^EW09%30e*^y;oPx><+lg3>OaZMGq;`ZZlB=1q_H?8q8;~~dW$n`RK=gghe+A; zF!ddMjpnvrVK=#>P2z+O86RB->nG=Mn{yJMx!AI$FLyCTp9Hph)f#4cDw|y$eT8Ms zyv`iGF0$xTPuL!(aFVplqeh-1zUh2{sU^m-AjdSOm8D8?##QX?#w<2_VjGil`^e50 z4W-5>U&-Q=Ck_=jVP}aC{ZntCokem83aV$pUUyjcH5r<7?Ey{tdXGl7YoH-hi8fDX zB;IqR^%YOqgcZFBuQ>ywW*<9mvYlDZ+054c34(c-FT4h2Vtb<`4U=0;2bbH^Nu7T5 z+V3iR!?{a#V>;NY(=+MfQgxnP6p-yS4c>Np(6}X@^)F0i^W<~b!hb<5bap6{_LiVF z+YBW1zlQVCt=QbJliqK8N%mjlaH87+gNFSF%a`ZzXXiZnGA@aJZP-hFh8Hk1q_BBL zPuOAO8CVzN3auw2u~Uur5i^ph=JRDbr4zzt)*WTL)}3Lu;{Dliw*vO)?m?#Pe~YqA zmg2{ZD3o75%I2yM>VxDyzO z6r>?yVIDkRie@Jc5+{Dtp!F{}m$mRYYmdE81q<5gs_FZeljBo&AG(^TJbvewL?LD_F8zodyOlItj=V<*LQou<^j|D){8qpAAh zHf}7LGDb?IRAflFdpMdzg$$7pg=kVKMX4xr5sDP0LL{On+S|x+EHdn!j zZX(ECa~FigeE?HeNWr_w>HMtiJDYbHSvKBTU%@3>wgirNDly0#fQLidkYjk z-UBr(B+%?fW+0_c4Kxj_1Gg{A(AD}axGWTZMUF3d+~e)hoE z6Mq1CD{*+{{9HIXR~`yEADD2*8fJv@9eqs;_To;qS8|@Nddh(8+61t@jsz zhrYq!&J<^G5aj@^c{f3xPcL|{{t5g$Co6BARxDp>}6eSHz&uPi{u zx54Fy4Pa5s9=_X??;*O80h}~?1nsLWKzI{5?Ys`S@ii3K_)G*X z9ZImqObmKgbb{^=d~Vh3FxU{`4;FvT0$yQOAYPi+xi1*Oa$y8GYB&WhxHlQ_7z(iW zqZSyMJsz4n@Reg5n{t8V%+-v|m z`CgN$`BUJ8!{33wtPPNnhBv(+Tmk83{U+1r*IBYgkgfuS%=)L2w^P>%5h7bQPQp!7i z`$bUD@&_D?dk-ef;Qd%~Pe6uF7r44(1b9j3fN#pt;NzNNaNtn`e?Q&eyLcMGwJXO! z`alWL-JQcTkb*!@Q5Pu5iv;&FZ-GtBDzMRL2IO`rz#Z5BfxFQxC_WViHkJ1Q$={N& zV==!E^m1VD-NQWdFB%wSNI|2}CbA45XerfF_+B&}Q@*_$nyCLv8m#aW&7w zw`aj&nKsa)t_U;iR`ciXS+JgeSGM(6fCmQWK=GG`(CDZxOdCb;!_{cGrSlY&?W>37 z<1HX;*bledUI=x6--h#EJ%+VqlTnk65>%*Ggy)-VVaT7euyjW|{4_=gb>ol0bMeVg z?HvnG&=$C5mKgduRubi94Z?3zbVTwwFGeomVo>k>Fu2e2Al{wN1x3v_LFmg| zFz2TY%q@?CPxy81Y~nnm|HTDm$c3X-o`c}&x<_Eiy&FKk`6+nAGj7ASbb#a@BUr%+ zV0oGYT<&}bt~5UdQ*4`HNnkLXTXGU!UXlvm2Hb|9TCT%M5)Yy3qB}5S)^oT-LkhV` z8bQSe!SIUvT6l2V6<~CfKf9N`0sl(6;RHDr9{WvTcd;sJvYd;qpVvjJ|22WOJ`t$5 zybr#G2g5k?49M8$!xlAXxTRJXZa$k1l?+{ADw+nbR=GltUo+5$-Nr}-?M9L37eM7) zf0$L|2=D*Lb9%=;0O8AIpi9bZ*nfHpoOUE0KC!3)-%jR$ZOlC|#nk|P{_2c48j7}; z9fsQz)1f3i3r{hQu&?MS^hpnej~5!janG$GkG_B-snT!}GlSQPWH&V6AE4Nc#I&7D8NBacxK{mPzavDrf1Bpax+ zEDx$*{{S2HU%>M(E8$hYF-T-DhDw(%LK&`3C~}%3V(+E{v76_?vH6ccoueWm*5i>u zumF{^QngTvS0Z?{)4r;zVp zZQ%9qu*2=?}ZT~<)xj&q$egei$iibrnUcl`#&*96&NhmVW9BQarz`>R{$nbowEGZM{ ztz-wc%BR4_t+BA^Y#Xdwt_yQjq~Rt-J)R+c2h6tMJ9~oPgQ0V`K(2HRkhi}BVtF5M z*lY;9_;u2FZ3GOEu!ZiUKF~pK9(>2&e^}QXs(jiCGv951g*(&WYGW%Hb4LJo2P}e> z{H%Cjr2Yk1?&wXoDan=~lC^3VV10CSBy63>w zcmM=F7yx4Sk3a;^J}l1RKxTpv_U``<`)2%xr&O%qnz5$PjK8;6GT~5bT{Q2)-vhIc z$HB~>k?`QWD7b)sj%Tg83_kNbA;G={u$=c_=JkZYuT2c)A0t+^0=Y#Lu;os(4R}k`|nh;01#o2Ed}OKrlJ?4tNuE5iG-Y=!oia z^vvA_?eXdtLIxwF8@gakvHW+rbG=EiDtAa%mMrsv4nMC0j&;L(q~O z3lPb3Q}l~h1CxG7aQvAcXpr^=Pxx$E>B(?DV{idj)#iX?Dgw<055b1$ESUe;3%=lI z$=Y9e*1Bp1Z0a}(ZESL2w$n4%S)L8Y`2Q6Q2Wo=VFAagO{YSx~KXSl-n=yzFn~2uj zxdPACNFtjifWiuOP>5Uz(l?w6F6b=;SK{ z=Wec4r+||xe9u8o3*Wcf54{~^kf6{K4t~;xnH~o4)sz>4Mq>%!B;dO;Z*_tRLwwIy zr#H}-oDJH79l@vb?tE{76{^d#N7BEhA&K-T_~6MN_!ey9XJ=lZ|64dP-F*ryP1*}e zc)qaMjX3bWtqU~sy&qm5B%sTlCU9mqulE~j!kg~xe7_?HGQ?g0|Hq{QNAEhp=yXXC zJZ>@wKIzQQe^Y?M;l~1zDpAhdxJoA17M&~0K|@;1pB7g zp;t8%(2CU-$n!b?f_1aO&S#53LG=^Ri<`l{m+EkvoguIh?-DG-8^OuHW1!Z^D0uSS z7C!Ir1UJ0g!TpmMMAzqmw!D5&-Y){*B@Tedy9g|pWCFGv{U)f;m<9IxE&%o=&cJqG z0KfMy1f%w|!AC1K(AxQ1PoK?KC~Mw6t4qsrtqA`F|k0{dk{P> z=Y2a37-<}nKn1^}fSFk-kS)0g-tyfm!PEHJ)vOM%vEBgg>cH@yhAs3e-3kImy@9A@ zCkXYD2cr=_K=RdUu)otDgbgy_M#yTg7;gJBg z6Q_XcmX5%>uz4V5b{yzw$pL40&uCv}AY50v0mghk03Vk?_-ytjxG&cnxUj}R|3`;l z#M1_-?fWCxTKi3KG9VA$I8y}6o;`u-*c>_x>B7f1_-^}a$>561QLrp+4Y+XW5>yG` zXPXkmu#lex>K!=+>rd{2p}u2~e9j2m|GpS@HEn`3GdIBFRsIm7+t4XF1X}ZcAFPxM zZ*TX7ncr8zg0lj2vsD_Y$x0xlEqW+$Tsy3o_8E%JPlUO8d!Ucia(L8??@gDtfJb5= zRM(AwH)G@ADWM(oS4#yCRfEB+gAu^6BnoN%U59#SMxj?7k}x&n9Vq|M2|j5fFn>J) zYJvb@P+tPa{mF&zLe4>-z$57J3K@{5I%z+248K8=G}_O3AW+!^S$H3>S!Wx@@uim2$~5H$Jr5L%hpqb37@PD)Fm zf~*x#x^5l>U+my__gmoZa~3?fQ3?Kw6hn0}_8hS7$po8>0gE)|-P!B>*PA*9C!EDJU*H02M}P zBOSks@bQi#5PsPJ3ydA%zgg1Ie*ZM6(NhbYu0H{lzskVs&S}t8I05=3&4r{d7M{F0 z9~R%82(6~@9gp&dK>wpPAmds(1lmq;{r4UG>_i-PJSYRb1vQ|^Cl+2kkpj=Qxx&{5 zz0kn52r3txfm0V*qUUcO!(jh+FyUb$oV{uRjF8rYx2D$tHC`7m%t!!l@2-LaaqjS8 z@&c&eG6Svn(FQM$euYwn63AEVA@n%$2-+3fpcM{o=t55fI&#Sl)Hxx(3w12mI$i}0 zmgu8zH^!ii%?cs4o_RnyUmw(b^#YoKi$R&M4@e>bFrp#=4m14!{Fd)m-QWRC zUoHdByElTj>WjgDmPbI_X=nJ{V;MAg5eliHD~i~tf^JSwMrVhnL2X?=qii)6PU3U9 zVU;_9ijNyewdQkavYBuSuUX3Sex*`Ale{D+7N}Y9evr^eAhVy}N5>fg`b!&(@@wHL zgD|*dS~SdgW(Vg4_kl|nKZ5su&j9zc0Ho@>f#QlVaHw}BZ0wy0tzYTG)Piv6UE>8G zD8<9?i|xRT0N&GWp$wv`b^^WRByeHbV(@778<@7Y3szPW7;sY$)@%iEPLC|l6_|!J zbqC?g-d3n;9|)HyAAmh@4ZPu=2u{gu2ZM2&fZPs$^h`<%Z8V>S9tHjfXxev>QdI(E zx>tj8HNI1E%v{iA=nM{@umzI002tS{3jNSpgr-k^2T$kR2M^991C!Azpp#YxzsT?z zMzzy0e~Sp(eeU4TerKWBPbv7Jx*yE)DgvKZra{l|K9E%T!I;a&s7gl_^@#pK7l*Mj zyNwLh#TvxL{`c$RASop#CL<&Ee?F7MfSCWzfK8isuQv_!5A`=i78aIf$lP-79CK(f zXNk4B*y{iPN8LEhob2!riqDK-kBi4co7~0d!VV?!>G1e#^^2sS#wA%m_Pqq>4!Tfn zWqEpjLqb*duU|0eq95WqZkkE{w?=f-y&Onr3<+0PnnR;dgb&$d&Gi-@7d2nZCwKUE zfu$2yfWtFvNag-D@W4S2dL&<-Cft$-@^g)_uk2yAc2Kx$q(pI!>*`27aO` zF8pk0^W2K?xyxBA=~u+|=t|L$Cq*A$TH%dnSkWt!rNpD#0gJ0GtZYbX0rL+GlS4k^ zgr2ff@vm23(Bjqi*!?J}dx5t9d&arr$NTG5M?;&`%>p8n>h9{Wvf`~JuyDEtTMFmim1)einNAG3$mrhio%XvL7}4(!gg8*rDQx=CP=}u zuNkvaH}(jEx^pWE`SteDv8&|JTT}QZ?>$;orALD1U%pza(p6QIS_Y41ytCZSPY$o;q`v-@>2-pFaQ#|@=PMw~xc_gT}h zJ1R%em*#*z=a{h~r`f2x3}caEdDWO(a#(tMHyM6%tST|vg5>SLMnXREJeuZr=wb7A z{9*oyYZZV4bo!ah zE*2gaiavLc)8@CyKuK*C41WT(r%Mt)m6=d(DFFv;=ZG#o2Lg9<{xoIZzI&s<~A%BkVe`bF%Cp=E+R9U1hc%7NsUJYRr{)^P|SD-zFm6G@@2X09`Hih=_BzY!zve{K zURLl_;vZHv{fUO!@hT+%5 zr1&OPI^lZl-G~dmked%xCAYDq)w+0b%w-lIKU{H7NeiqfnuHj)d{L&OCia?WN9w04 zgVLLq(9Mwfta!CfmCl~2WM<%G95%OwXO`>GZP$`Xfv~Hh{nRQFn7xw*xFxXU+i^Vp z=6RGbm|GR+s7pGhDzWiHh33wFW0B#3CZv??1V7)OK{h)#lYo)yB3CP4G`_MN{#Y=h zs zUyc+E_nr_fN;0TAb3_t^T~M#`e|e%RGk6Z3F8#ft?fc&zGGl`i{QT8zWw{DVXl6^RuwKEXL{jn0fk5dP4 z#!8DWHxWx)ol2Iox-Iy4Y#ZNqP+>uw3xrC>8J0Hb_Hc*cd-8qoAe&b-fr8st$fc|g zmS@F(R_IN5FI+VJVCDBREsJdn9>Qa($-=~w*`lu7^1@@glB>AE55%KXjo$9wz!ros zmFrvzsw5f?TgdR;S{n@}&<4j5Feo-jbolX=sC@1cIcTzHrTCgDZH`lYt@rU&R>Xm}}gvt5vJAsw-72Rgh`;XN%^` ziL8Wu2J2G1o)xrwC*QQj()>LZtYwfP{ecdPj;>tE&b^!g-$(TUqy4r5w@r{cy_|C$lD40L^2bSy%JWE+2+%~VcsuH70!zFSg zYqPcRp3qv@RcysJIk9A+oHrQKo=nUpxRJLLrn09S3Mw`Xe*wQVu3jw&JTmG3y{_V$&QXgo?Hk#LG!Y!$lP`+?c3Nvwa%onWD0gg`V8+T@qLDtp7L?UiJ@dS4sjl#W?0e)66qhH?`8i2S_&|GC z#j3N~!i7^yE$S>os$5pRtt6f@LN75!aAjLO+YvWM&>o{A`2J>!NbsSC^sn+}Wy`8q zQQCabcy((qNw`Be_2A5k_M^;PwEm!l(-I4d*2*UqZw;=)B7Tl}V{-r~Ku=hwiC0C} zl4Q&sUN=>WjZPP>+pZ>@R&)ZcI;Tz=11<<=^hQaJc>5DWf+t=T% z4C>qg>aJ+9S*An6%k!>LEu9VYvEFM~byCR++?TUTcX-BxY-aJoWu1k}#kD67Q@yoaC*6FN{vcQrl#4kSxi%}Ckw00(7??a{COkgGD0g-8c~&!Y zz(Nlm#sB$OU!=yc|$BSDYOL1j@CfSWmE8kx4HOTn>a4`k;pWMI5YmP z+nD)bvW)wv3Uld`JQh#(#7V2Hk=t7q6+5Qm(cHz|JwOxD5L!3rcjW((~<^N?(`yN^S*7UG7lO{{yc zI(^>}g%|rQM!Ji9=sfa?Ui+>^_t$JE-z$=cZSovi*WB50LQL0X1 zj#T#J@gKG#^Ja5gW~u^(TlO+t$+#a4`*`!TBF|QTKK+GInvD4hW=Ie(DNIM zn6^e+USBsrFWFo8%nDb=^=vqkeE$o^hEGw&gNJCw(rWfn;cTlU6BVn^%l|UJPUxeK zL+kOz-(GmE?IPxJ+zvebbQD>2#SOoxlf>^U#qsmt41QR8lZkTcWOnSkfSa@|aAT+> zaZ31)u~f`t0@s!@S}|(0{p@7uaM~CRS>9#GXV1g>PW?n8;2&O`poAA)u4C)<2axD< zFcmB*q(9!(!p@^o(0qM5nVz_aRyd@g(KHc>5uC*(N>zAoo+HyCNT4=xKG;>u1jmhW z2jBC?F+ZPm;fI&!5&7%?VA0ELl&Y3NtIHFhTzn7~B`7j)Vt!yXn+));EQ+yuX~nG2 zna}u%<)WJ3=GfNBoEQtYvf_$^ELYXWdWNi^rlY>-=-S`-y^@&~cxq+UGIxfRQhp+_ zK5kDYO)+5=e=K7dr|FESz>U!@LX6teK`fgN*rxiL;#tK`azQvL@&TgaHkJDIl zjpuaLw*9ELfS*IR-y@}^^2~9qJGl3k78llc8&1y77Oi-mjD8Hc(?RhX{E$fF4x2Q3 zN4=V+l`3)CvW?6HjY*7{`Xr`h*H>nI=p$zGpFu`($9~35<_=RW_lQZ<`HE8irHZ;v zo}y)Ht0<})q){k_BYu~drM5?ylH4_n`o0>}xG4}{(qG3KuPs5B@($y(uODCyNik~O zm`esytFg>;1@yqn1OJJcj{k!3OnyZZUeBz+vxYu^fN(3!Hl4&9N~4%Z7B86zsWX|l zEg!K*Mm)`RAoO?E1hy|U5nT`TM7xfj!5znx@uuopqE9X}&nI(?TYV{$$d99+ZIoln zG#t>rd~<3XIYjNOhRN@1xy2SZLC%3s0ejXWq>AC_j9PT?dv{_Ml@%R>aD?~b&FvvcLd>^^37=Lw+k4K|APd36mS)q#W)rFGiqk)Ok%er!!N!# zSxt%2Hi~Crb~V!0w}dP`^bH$WJK&8M=41Ld1APszhHbIa@yBi6$fr-0W-ICtLH};H z{bPUS zRdQrf{Q~mwMFAXgX`*r+_Ym$a$8!|isL9>qc(gYU530G~2#F}1F^4j1QXXR*!24at zx3XsI^s$~C#H)NZFwb6g;*2wsm=ilokorVBw&&p$EHW{Gm4??)!;G0^c~~tB{~N@5 z$MfB<>tCX{-xcVdsFvv6SjPMwuESnA`>|WO29xV{knwKNWBRuzkv8?4l$b8(+M^0k z&bDA2ZX}6JV`6Zpk|p!$t|3#5mca!R4AFtt(QL|pr?INRKm6ozKeqVJv%ee6>Fc%> zx_0+_6tU|sd!Z&H$%_r@ly@NS}Z`PaoXvpxZmU%^J3yvX0nqvQ|OaQ><*~XGuR*lvcp^WqYV!q#1pXZ_G?&?HTC>>dcRIvUJ<*3c>o;og`QJBn}x4 z#gU%puv7O@^r1!)7bT9z+Z0Qg2|33YZ@IOM;hfDl!&C{mXVybG#R8_Jqm((btC%TM zC0M(=US%iKS`&Gy z^hA^TJUS)XxjvI^kUk8e*q^LbTO+%ZTP(WE?uI{~`cb8yi>U9BWHPJu9x;C-MUxt> zh-u16`Ya#O#`ZvDJ~*4zh`PpljdLUlaTr}Q)u%o$qe!30Kl1Fk3WWpjh|caZ8(uQd&jUr?)Dy=wiDq_VuLr3-I-YDg z`d)O?Ae=myuYyM&+LINRLP_#eW0Df`U8JzjmNaefL4%Jo=~LT#==bgmSly(v@VKWneqB*3y8dH1AopV|Cpm3K$-@GQ7N}tzGk@;=$4D;iN-f~=+nj*&QvDKo%+s;foT(`AZLk-f;*Y# z(3+uUBxw{#Cd9OolWv$E^6xpslyA~Ho;%L>hRI5A~OEj z5*jsUDd{uFA;C+Wpu1)uil2dY_g%J zu8GsF2li88_aBn!+C>0d#rm=aY{Y6qvPGoCdX%n2*3&6z&1y$;g}P*YBO-JEePGWn zOlMye9Aay$HAw6jdDbufut>K#0!6IPBfblck-4$r?6<)rBEP_v92+oW-*(SJQjR|C zsknSHTcTS8u5^-W`2lib`#d^v(Qej!{X?=VHG-|_cqw`}p#Uhih!NG%Ow{ah8f|{L zpDkN9hN@nwCt1xN)NxBX@_>h6o}M$R^AA7rr{MU4^6aX_IM0-yL>`-f==VN7GHguJrcuwUpgAjqVS-Nc^UBqL{Qd1b=u- zhI8WKht_DwHp~@)*!$2cUKS0zUS$1x)=^`!kRE)zl!lH!CYl!fn&kaXgda@afxibH zBMF&ya<{&Q=d5d^fBtQ735bHyVz^+dG|N4k%s$z-izuBw zOSk^rEP4`}z-PT%L3xz{)w_p?N$YGd;Z7@w|1g#OjSeLlkITt*Ga0f|*Ok@Qb7xck z)QhHd57IA^e&l!WT4dT8N~XZ$_+XJ7@{o6-lFy=u!-gUDYqcZ$SMmYbwq`8p&6nr4 zao|?+ck36s~SWLg1(>?Jr;Oq+bg;wWSAz%rJ}gzLgGE^49xRP z24X%F(awqkf|5B`VC%~z^q|HoG;It2oD1I3r=AJ;d|(A0F@8mWoHhM#bO#B%Ka<;W z*PRPGevCUXA)mX}5z94wpTf=2*Jh?TJ-`!w{-VxbP4L0M33yJh4?gOoOrDHhB|}pm zAd@+=+?r{bT%p4Uy2>vqYhElQ-A85Nt(QvZib)=I)(GZ|t5wL6lu)diE5rw!6HsM* zVbug{ZT9q&pDf<>l?E9$Qmr35=r#P8&Rgiqt(q0geRxzyvTt*6`&Aw6zoH1YeYu9% zXT#JNE!6VUEn4oolq?H53P*$Aq5n2k(L-M*a837ZvDxAdbk^Vst~L5k zu>AH-(nyur=TbsiGY~~z>@A>8OCA!N-WzO)r3`ZOh^J<4tHC()_t-fjgo<|PqV6tV z?rEhP(LGoN->#Cu+;~L<@>Y=W-wWB;k{bl;T_!@4Q;3=5ge(Rkkl05_Se-FM`W=LL z+`Vsjjg>xrdQzMG7a@T^o=)Vx*(~NXgD%ihyA?RC&Ly1bXb|^PZ5B1&Z;N9dq~Y{g zC%AlhJ5F7|aUlhBQ2E&LXv*~{)O}i#q|bbhq^-Z8a%BMxA0CB0)urf|O(hL>N}!pt z+4M*8XA*RO6RA#GiFIE*WnC=pz^QdZWK!Z6&UMaPPW(;`_fdYVrQGeCpv>W=aE~9a zfz-Ov<&#uU#k=Wb$F!e<{tc7y>Pz=IJ;?{$-q<>>oo77D3Z4)q;~ZHP?ZHK6i|Fsd zC7fGq0{!==5gvajgI&dfaHq?3@+EU8HTq)>CpP9F`wg2=z^-{*-#2+~!>VtzZuc3o z?V%HL=+H%9VpF(DgYn$7)4ALNV}@IDpWg$+2Jr9Cx%BoiT{?E)9Jy<_8CzZ4i7y(3 z;}3BU(eATCB&Bi&ZP^z?XJ%UC8{sSxn|K+k%FM&VxCPt2SKuoDYolu>EydTbA+$98 zHmrV{0YeNVxPzgK=_I&-rfQGja=RUn{iFLhY|mR-I8%X2X8KX@jJY&DuMgj@oKBB* zoj?^KPHb;_Et}3iiv#Xr+#fS}Y-*6seQF`xi??4m%>!j3zs>&0)$t(PDeFU9!(Px^ z_Y&y&s-^5a2WR~0ry+J*)PT0Wj>Gel+R(I}ujz%5mE1R36>iFtf8euQ9W+=qkzG7~ zAs22so=f^5jU5#wxd3k;+Au4I`oG%5jlZyio9Hlxt6AMo79?$_HVuKSw4EB4H#n8+ z*V5p|Xvo8r`(MDk=r!2twJY6}bAzr%RkUWB6CJUsAgiTL(>YFixD)Ep+$*6MS5oGW z3wBLGMFYinYi1PlX5bj(KJg;6V2L{3BG-imWIo_+tu@$xVbIKO4K2;6X1tEc46&@0qe3Dius%zN+dmyLd0IV`3p%@%TP6c@m9YuT!#e zNm*`Hw|u)**xtof<4QcO>`pjZ-5UJMT)Fg*2^no>7V>#^zqlo^AZsqt94AA}O2n|l z^f|bzGz>pG_7pe%o6lIt?Zhsj78srw#h&CQ9AGrjosvueonj_u5; zjp0o2+(>L-3F&@GSsHS79kzB=M8B6Np&7H%P_9`7ndxbx>wR^MHYSB5V_zqWUv zu8ZG*`>bfTqqbk5S3Mi0HfJCKj>H=xjqvUlW;j?U2^(kRvJuM^k+?w%YsNul%R*)5 zrnnjdxqQa?7iAKjQ)d2aIoz`62{KQUV18vtQ2jJ-?D#yEsZvbBI&2)aU8{+Ae);Id0@NNwOGI66ta89FrFil%p2+1Ouj@F(_t2d z$IcjrvZus|@}78@Z&}H6hkB5Mc?Y_#IF@|ui(pspVG$>pfu1gjLErE4U4(TDu`5%E zd$;!C;jBd1r6GZD>r~)!`}xd|nT1UEr5t8Sj22_D*qXT%Gln^`^CBgC+QH(^8 zft9dH*~(U`n9-hNjnuQ+1xy=6XaB@vX~Qh!wY3&S*vV0ueZ}mV?wdT%Lz%gKU7k@L z(P18oC8HH<#j)7jv$z>bL%R1S`ZYEST}g>zMgn&;H%HDe?Hb1+?5hOB|Hh%qlI!rM zzq|3Sx+}P`@&j9C*o%H^^ID948xz*w!>C2+S{>tkn!R=naF5k=JTt150ZO5a#wHi6 zFjTH&zrw36pU{llN@h&sdnT;xJ(K>YoF=7L zk&xN%k?ph`hH)g!)S*|*OjSQDKahzUrf4!3#>z0RRX+GyU;?UGx|B^**CWw^pU}N$ z_wa|y=FC#{9SmNjf~_YVWsWaj&txU%p-I&O;?T_DqP-P(dSxK~&m#v1`2JP2;WR)*Z6S@+^YUe}tgz+f??R#W<$eHj?Sm zDq_?QwBdZkBK&f43|_cW1z%sUg_`Tt@!^~DjPLLR{6emTIiuxF){8Ao5I`-kj3-O9hq~Nr!zT< zR{U8~1Xc~0@Vxjzo@V?2GtS5G=$RU9HLQsY4{4KYlKW9do*SB;{QzsnsWX(Oz$_yr z?Bd}Iec$zQBV*M$XR{>EaBnc3(c(z0CwkJuQ>QWG_oUIGmz!9H3&q$yRGg1O;afAQCYO!zk6YNJ zNC#&q?&QW>SaVrhbGfOhiCB857hf^VLtB1%lWiah7N<9{a>k3PP+p7cJTZoA+Sf#E zrKiFhn|Uwf=c#Dpr$^{{i2?q}7NXb6Q<=8HSxim*7laDavC`TYQ0}x4>HWSA*8Y>G zfgR4Y=-hFdnXOLcW?aRQ6OQBd+ym6ePmBLsGMZgif;aaEQ5DmPbXQasck0<~yg=&_ zw+u*fFW~~<^E(3R>^zLpW~I^oIXP6{Ul~bGkEFoFl3eZbB?{+iQ2Eb6_Rir5+P@7m z>xRR4&2I@4bg&lrm)1a!hY+PP1N5W39B0RiYX6m|;LK~m*z4;F_}J4KNAlNwM#3O^ zcdi7C(~K8+r6-AYJ-moTn!91v%Liol-8U%wj2O1%ujA}%5?n>U8%H0!rBhrU;|mG# zq}lJiC`>todATc+Dc^FQG3p#<9UQh{cp)0Sm^U6b84i=V(F?J9u`zP}nSrh(rlAIj zv1DoIEIe=dBxcRPMH&hJQnf!8)W)w7S;lnH=X0iya)#;qY(L%rWs=a?jF%9LUtdm%3Os zsz?#<`F@p@o7^HBmmHxNN1Et%JzcW7PKqv*uVwZ3j^{pz%4m;xK3RFjon+p<$F9#k z3I*#-Y4FiEG$Y-ID$Q2q)E~rgOFtxXMSH%8Ok!@+TjKsW*T9Mf%R6xfQ#$AkyAGVWW+Gi}brV}0y@_9{FC<%sZetH?BOpEV zmgqlbCp)_^i{{Q>zzLs*aNjyLaFksk`q*brZ%*6KskW+dk6th0+(PB?Cd1=o;f;+b z%1W7*<2JcL58`N0a3|ob&mO@$L*9Frs@8r z+??pM+-~Lux2aj0TWRr!&UaqIWj?)(yf6Glx8|th?urOJ%S3}&H8v9ajHJ+k)#cQ1 z-9Os>SDI@Cr|D}YA8Ku&LW6ta_bgWG_z8gPVzVZrYxXOZ_(4sgrRfeROK z`V1K?o7sx|kL2K~iaVHnUXz&7Gt;?4k}aa4={gpBo>t3}z&FJmj8*m!lO%FuvrTgPt(2qWr#Ow2Cs|k5HxVRl(b@Rc+4@+^)?NM}D zlf_P_27yRoR8SHu!#%TKRW@Z&9a_hwfBnS9>LHZhr@d zyR+e)-PzD~hCgTfewa2iexx%7C2_%#B{)}F4#!=a%E`umfIlbe(j|M&!juEv?Cw5W zI=TA@le5l=@%I_Rw)OI~FU5#D?hnS(LJo6g6}H^gRg}6NOJl7Lt7G2{N_aB5PxA&7 zsOKUbx_$T!igcHvXRKrr$EBelE_SFX%wP66Z?F(pkFG_ zal;cAaS1sBF30W?{v?-AE6ZC+TBIA6vUg?{DeE$Wwfl+A^%B;xNfB*X6N>H+ZzgsH z+2mNxBG#h!7Xq(W;nT}bQ?`Z-hKmVo3;dOviD>2|GOwG%`e4G z_1Wl;xjEgYoXa^X|nbnr?%Ii66& zX9Q1^W3@Y3?*wO(8W)D94-KN9oe6k`>QiLi*ntjIX2GXNbGVSNzTCvKBGU7u2-~gk z#_ z(m)q!X2=otRo^d?`8yPRf8{SSEt*MX9i~9N+XKQ&`Wrck`I9;MwVGT{{4CBk^($wq z-^s}w5;ApRdzmFJK1}h6VJc?xff`r2;;rUEbZ1;W+Z$#@4xF>)mL#;0?>6Gsv&$~h z*&Zzvol2mw{M_c}t7Awt)sgC|WRdzvP4?!6i)9_!*sG$L^R&+}aB1M#)IdH?W7@*Uh1{U|r_b?_WX}@onsrF;P%2A)ak@ zmE)Q()zS7)Th95*R<1ko9Cu|5U!k<#59YakB%dZR^yryrdM@N488hw*{g;1`CiYCF z*Npq9`I=r-eQ1ErKNrR|EqFx#o0h_@!KF}nbqV##)aP`>xaT=ch1V!%8{Bjc1Yk{d#17PM%!&)k<4_ zo+CpC+S%=Y|0B|>Wn|~O4amUA5_#(^LkGRp*@5;Wz-wnNRhWE}D%|d*Hd?JHC`20P zZyG}%e%i+=e@LU^)r+{nT~XNYWgf+mZ|Ex{dHiL66O{X6hn&x^#W|9_6wYnKJ8xZv zso}Q;>g9K<8qUqd8vMEHj(t8ll!}mVx)ohIo5H&4ZhH8m4%g9@Kz39G&?m7&tjdF2 zqLFn4tV^9ND&v{2yMk-E5()mfX_mYWzj~6RC6dB)&5dx~oKwWsMv|^gQWQb;F}$Z` zEcv1Pj?DK-=B~nJ+|}1%oaB%N3Fo<6)tO>s?Wu6)ztleZTq%J_yRD>Ut`|V(*sH9< zj;Hj6)pwL;u!T(CACH$i8)FmQbLd^JB)4sa8ID?!f_zRKvh3-4&R#z^f|9Jaa+URq zIPb6LxnBuoB(rIN-L_kW`dJp^U4J&wyX)(y=_>=6Db+7RV_yqK7G0+vEq}>6vv|0h z@uYW4F3|3sg_Jt=Ae{4tO;;#nkATHAb^8`-A^r@Vd6Wa^bPq$f6{p!_rvkaSsi~aB z*GMjGRsj*)DMY5D64dXnE`?^NNPVm>ZH;wc216@}Kq(&UoybIb2g+D-Bb{h>SdqB7 zPI!)b7~UV@&kc>;MXDNgsb}+Bxas|B_G#xb6y2guv-0X`<$_&;rI~fq#H^pbSyw}z zYQ^xVMN8_wE?pR&XTibi%3S!SMy~6#4br%OpUpbtfjp+xb2a8soL#aSm+?*(n_nJB zFZZn{u{+x+SG9)v&rPIRX9iK5r8_rKOpgsmU%~c?MR<^?vtXv<9>?@%@VQGU?pBwa0-8MOMouYbBNb}%aC?d5zb$J125JR zXT0rxTRIu}f#hqksM_=@YoH-RKKmc%qQehDZ!}w17$ZvlSiN?1{c9?jdDG z*U>NcLtvN+A#6#INr7@99diwa}GJ z+Yx~^O)x49l0{mtrKpCT6uw-ogx|K#z$K;aaNZ<}?bE(slm1v_eX)xiEx3tF&g~~& z9TI48;W08Ts~7ote1~k;S>$8jN0cmhy~FVqyZ`t4mmLMhRaW zznopO`7iNVv76p&T87ivSvW||2H&+ZM8)lYa7J}DmN)E1W-S?{xGaHwkeDx`n(D-+ z@*i0!FTf&uaonz6N1pm@#q@y^bI3KDqC-pZ>%4H3Bf9`Evg*WXVcJ;wr!*sQ9#1Mp z#ObvYi}9GgF7R$gHaXtA7Z=Vbz~ACjaGLrMF3}mqD?=;s%Ql9YFSUj>ndyW-Um8bS zbCYn}hc?`~dlg=u;ZGJvz9F{7Q_$>@O;y+Wmyk-C&2*jON|Lm4Kl`^ugn%_amzzIA zE)`j7#qi`E*e8uv>KBKgBp13YH57xe;7Isf3Chbj9b}zMhGRTB#Q4n=aM2M zrKCYgL#3q>?InAUY#}40tc349pCeI9%4n;Ek_t`wk@oNVC*0S4-FrUg^Sqzu>G+5~ zpO9u7wR7O-H#b!C?F{4fV*>JCTT&akN()!IPJ+wVCgW(^Wq6TIAr3F)NZRBi94b+d zgsMkTQHK_>f31a9>l`Ds^>Jk7)e9UystbKKUqsFe>Ee^~icm(M6B-=PLQfN4p{O;J zaJSe!bXiIi{pXg#vzwj3#N55b4DX5K8IIjV4^w)0bDabke9z~g(>G>e^@SryzF&pi zL#)U*%X*ZsFN%@-@f)$LGmt?;K6t&Hg>FSw^0rMtJeApt@!R_QNa~j;`?Ku?E-x0w z%C`)#%F?rB@55*Ong&Cx>VF=UPn2Ny=G{Pg{W^GQwgk@bz6MJV33|gKo%luhGr5c_ zlaa|eUHH}dAO4l0iCsdXn1I`Ati|;kIHu?VE-qlmq32FW{!9dyRA&qNV)o?wBL^<5 zLX#_gtOQ2+C5+#_bws^Lz%1OTM-mxh*tqQ~s$+s6=l3tHvZ4Zg+FOMceGIsy0YA>> z${e24(R+Mt|8G3`yw|uf&kL%RdpO_E*_c=~!NQnA)TFM8y+h^Ef*NZydB;hT^v9mL zaX*a_vYY~Y92IH1k{HpA(7@f=fh2CbEMt955lIajbL(d0ptz82=vav<)7EB>MOH)- zCw(7$;o4G?_(>hRhP;8xy9b%xEjH-mk3G2PYA-j|J`akH6k(B>nJ8bk9UVNf2~RQ- zW~Zcn#lyYDIA)dt6yQvVjlT^u?K_cmR0%468-aBtmZFuD|1jL7PmK4IG@|7ZggXV@ z(8d9?sZrfzShep9pk?pLM9<3*`7N51Jl}?yu^PPMMKZkTh=pIjwV~;W^LV1c^3CI(wdpkZ>o9W~Aaa)&vu?yX1OjJ6*lj5I~hu&Ofd zbkno*WKN|MSfn}QND)3U6ZrkG0)MQ{f$oQ~W!t{^*n*U{m~Iye!) zQkUTA*j6JN7v?R&r_bCZ%XfBiUw2z_Ef$S}9l`;et|?0l2lbKiJX^REB@eL;Ry0L_ z78UYsCtvSs)%NL}CWrNHnFSdW=*ZeK*l=JG9Ud)53Nz<%)?Z@C6oCgw@5>!7qt=W} z>zU8w+TSHB|6L$+J!L{2{Sgo~r zWTDAfh}&U-pIF&|=ev2Z;#Cwhx!FP4x<|NW%LnWzw-pjP)8QEZ0vub^z%*{Uf{^Y2 zRLwp}B01{A)3hP(M|N%y58 zrljE?UtP?SF56;6cQ|jMb8`>CfJP@?@lc=b&yIlaM-$-v69LoEB!ynati*?gmm|CC z3V3R|f;smg1|QpyOSg!;q9<40pxgFu0mr`Ob& z-K>I?y$;2i8tQbv@nxtQ_l0-j;l!}3glt~29&WmqaYi$jL$h8a8cA47^0Z!p)j2t~ zsqr{lxAu$A-vSgL-w!@V|dhR0Ikw03SA zJf;4m#l#-!o7@SrDHrQ6%m=ON9_UZZBJbBcLOY&(MAx-D@P_VLY~^<^@M%-| z@J*39^!Zvhg0-fzM$(a76%WwYZI!Ir*{#%Y`cZmtcoKWDbv5m^%A_LONAQL3wNzQY zfWl5OR`zHhD*caR_^*OFeB>XV6t))|N@a32g_Buf8!I+?TOckt+yZ`O56R>;4g~cn zVA-GsJf+JYAM-Q7)?02d22mC`d*TsL_OHZMKS$X|n&Y^m#vJqZcp&du2c$pd2b=pt zVQpeCR?JR^QgJK#BA^bdoj6Ta4n^X9Pm*e{{8mK4-*l0(*$2e!$pya^6N$T)IQ^t? z1!ieGpz^X!=+|=>lDYjCQ*=h4Et;=~51&60!>Rk}%=sF$5&L+DGj7&+5+6O zi2>mN9k5L5gw_@TCp@@_y;bhO{y29GKXk~#KI0SEk=w7aquUKsx~!AGFsnZi|c0HS5l!TW`mM;Yd82>%*gWTVLSZxm&Eb_hm@zC4-JI;i$q$6qcDB zfrYv81cs|oh`M0*+$vy!?bhNC^bfwUTi&X}MFi&I`igr5u;tCpZfqNAT;nFpxHVCowuT z9pr{A&>G=E^4dm;m``>kLrvFEitk)j)MY=+eIH0u;`8uYqchmIhk!~*50vVApyc(L zf-Z#)&Zys>4d`$up0m|x%xnWBf4POYYFqFn^Gs-i>{j~r=ULj+UX4HMdUEeOZXih? zH88kbhVQ*GV_R;IF)h0`qvEW62K`ivXl+C$x zz)EKQbi(_cz`wK2(86qW7$|<)2A}npvYpElan{l^IQC*Nl9*cnd&8dM)AA?SA^S6Q zZ<7~|H&vq7;%tb6@D#Gd-Gvua$YcBC1q`jr8_SbR-I$fKC)ooU7uaQfW<;~z7vHR! z$Y$>ehr2<;Fs2-aom8|~eEA-(3--m$InA8GDQ_lWhc3!+bEp4OFADDLfvlH-5cpi| zBS)My+4g}Gs5{<}xw>u^c8a~sG%_sFU6ai*N&5I|?i6&zu`nPahi&;y zT%y)?^xOIfJmVp(nSUEuNPobFy50C!i72}%KMh@@k5QnyBFb6p3rlq(VXeJ4Dm!o; z%lv%<%N6gVQ8#_aKMc^DW(#P}2R^9~!#i!`P{UJ0k}l|v%P^Y))i{g4EpHO8){tb) zyW`2@gcpRp`;`2VJy*NCUYp4?X=Tio%cJllcd{=_1P(6~A#c6^!qt@r;EQ2DGF)v# zd`I$0z=s9%Z!l>&42U;a+iRwjj>Uh-kG3qCMV@f)>@3?rGSS7fy##aHWo)f|?X z#jJmS6-mfGMh_onVy`cu%*`4rm=Smv8@p8?oBx(Fe3$n`x@jl&J8=$AmA#L@?h%9T zu7jk&>@%l0rUu6>Ml9d^x1&s@c_g|!okT=?AS20QJSW-<>v3;R&5`jQ0rU)pv`dw`*dNSnnGA zb5jbo{@c%#4aDJf(%HD_#%q-SUJ>U^?ZxW~$C)s{0z4tA8d7vc;Z$-U*7p#_y}D^U z@5rM_`?MI6ICUOL1W8~gcNZM^Jsf+^xP&!+72#YdFWmA+3(GI5MR}S{&|@dgVOyyiR3SyaFn^m=0^^b?V|_Lo16!IVU+1LxQ>(5G~rHZ7TGSrStFRjYI zHdhAtUTMVdxj)qKTv*#?U)HlBlnu*kWYr zS@{KK#0tL2HF9Ks>@j>!XOP6&%REV{C$6@~vijM3(P^xzpEReF@;v}-9O zGA$UVy>DjsJe`1-ww7Vtn5Ar6X*vFHzZ`qF9JBZ5Z6`0JE;H^IZ;{-$b4jAketdM1 z;GIUTL}z|qBpFO9Is8J7`n%@CHP34JI4d5-C2qnAa{l<6VJ3d<`T!N!44~{2g{b-O zEOximEp|cab8^wh5`N29kfMtN+`#5u{>;T#>qE*|pWR%Ntcm=6Y8&QGXgLHxb2YiPdcGM277)$;J*p zy0Q0)0enuO9{1iiVUPC)V$D}OVfChYI3v#hy^9nAWf5&YGuXykt2Ku1Yu4h)D^IY^ zXQSEL2~zB?w=Gy}K@#`v)+D@WpC!35Tb!Ld-jChhKEe;O22s&K0=P=cc-UvbgI6+4-AZ?OT#<^`92f@Sz1Q+X~#+8;4a3MzMWXw%{MM zh}36)!@H0FMT!Qw>pGmqDdL(=y6ZH{FoG4?ct4!+yB8gspGh{)DWIWgh6t% zBDrv19hdb;k^Ohl$Z8jNI4#Ifvz?lV@4s}zv~)fz_Tvx#=CZBiZN4aR%Q}st<|*MQ ze|I*cXg{H#1|L0V!K|22TpS>UC+(fd9`ZZGRaEXFvE!4;gtehqBkwg=ov;;`OILBp z-*WM!fmCe#K$>RPb%D%17rG_?FQ>P?9*<1h!7kpk2KgO1hyB#zP|+`MzNercIl}QL z5-t3~?Ny3nzSb~oaJ_&xS#5|PZ~BGS?hs?U%C@PCi~& z&pYh(T+m&mLhts^fFJ&rXz#WuNV#B~k-QrS6XXNgYH?LI;6M<*K7TvOl$&IjoGLbbFlYsnONf6HaN|rw;MpqUYA+=T)@+{90 z<<=*m@nj9iPzBz0nN~dZ0ka*~IpO|C#B&m#C_R!xieaUQ znH-H2{QJRlu{hmyteW}Sr3jk_s>$bUdHD9d01b(oBklHjqB&(1j;xEq_w#P@g}><| zLDdUcOs_J#dMnB9P?i`zJ%+5;9^~?ktGMYUp=epQ30f@^MJm^|pfgSr@ZpoCXwmRL zBAzJ&$7m*0Cm^o8VLKS#{S4!+4x}Ta64l+Ak5|9?i+n!aB(?E_P`$kXhW41_n9?fr zxL`9Fi%vo5zaJqZ_Xo_htah~4%n$kdd12KU4|u=H17gM3p+kO2FyMb1DRe#N=KhPs zDVf2r;a4oowA_PQ3P#DPWlhY|5g9h%<1*r_9K_X~jt1kR2CzG~4_>XU;O`Av!uUF; zqwE?7a{7k>=oqG;S@QO5b~?qQ8dg{`D;*gK`0UU3l9{xTAB-(2Av5pC!=a37c%7IE zV}?T{fes<{ygmd*m~SJPH9D(rp&cp?I7|5kRL-b4XEmGeObZ5KVpAC-EDcIn(h3oGy7IjOC9%XtUfKgS3f78@}7 z8}-0zqdT;y|3&4gHRwidD>6HmOx_0HN0Hto+}C1pGCMZ~zORb~)47o#xjK@uP02z} z&3=KSOKzaZ*;i2W z#B?t4n=*Nw*GDScWy$q}ndlqMB+8=`K<{+``S+rad_A?7QTyo!z3Wba?j9BNTuTbu zE@)?*W!u2)g#rwe6!QbO77?qiF=oPYOsQ$G{8_u6RRpRbH(fdm{H zX(I<>7vajBPyCqAiC{eK7AzL{`yAt!fX{`O{F4=iyswr!`Em{ZT*l5JWFNC0Z~GXB zo$l%5M@uk9+gRH1rHi?^z}nJrl?NBpHcX^LGI8h9)7ZH*S)etA@LHNbq6zh-n5Vy) zr@Aj1ONwZs?0q}&$3J@L*^d(PqF9-;@yjE-8GS01stpeYrNKSU4Q=0GkJSy-NpJQG zSbbd?crj7nXnlvPP13-6m%icDivTw0?m83|+om-i#dvCb|qG4ROc@lo~N`{_1CC;uX+Je1p zE4gl$2y`UYh)9+0M>cIPXt-bt>gQw`$4_dg;DIvJf4G%oI9Otbx9`YYw3WY2U5UT# zSqCHC8ce1;OyKr*wKIh)lJVY_k7(<-;NJM%g`MlWkhM)rV-$A{A*WY@xlV%Hv#A^iZwCtGi_J@pyRfdXUo{3FYs-RKZWk zNJo<^ws0?|*x>o2y6|H`FC^vs2FJxE{P#|zbX!-+U-Qt6CG4R$ce z$#Ze$w3XQZ>~<6+av!bwdLP|+u^akac9RD70{TzSz;c^S7&DykojV$E0O=MR<7|O$ zU8#60&gL$YuAar~TYQRC{no*$`YqhA9zULHyoqJd_X~{W z&M@v|-%kF$>t?8-Cl_^o=J-iFX0giqbh(}+Fv-QQ7p<#Uq%xqL7xzqk| zhpoLi_r+E;Lv4Re@63jI>6aT1Ob?#1noLYOP&e!TbU)|}<1Cq&|^8H~G@GQ3b0XGd6|s)MVM_S{Td^KUad z&3Pib<<}R^cjk7~Tv5#<0?wn#^AeJAdxmy=3?V0|c+St0h?HgXgeNW)c~m`jNlDYd-0+G{Glo|DnsO5$MX; zOO6#}eg6Unn$nZSSe@gZ1b>?2ijZ&B6lI`pOBJMtb8 z_;F=5k?BAzN+{SX@bMf+`-Pj3vTFgqKDw5lGUqx<9~wsMkCyY^%4RYvyVZCyb7Dd1 ztPE)QIC5hxZ}?q7d5o*8G*>2IMqf%53iNMhcCVQfQ|wX0JiV=m{AP)R@2ztnvs?=E zugOA{fFV$m86*0mX2hp+3mBc(YXUo)waXZc=hx1s(cHQZ#f1rF9GLcx;PwL)HJIL!t1oKuDjv}Z5Foqvl_SkzO_ z>31M8w2g&Z=e6PZvSjFa_z7lfsKW730q?1CohLZkLCJj)TxgqzFS=gjwfMJidI6qf zXP7AWwkVSHWYk(dW?o>4tIqg!^n5a{>jiVvAcRD$%Y;@_Em)vi%=J9yla$K~Eq81P zAu$W1xvOb%FvoNYu~;}2=M79Fna}o-Q|q$HPfKV1%D7SnAJXPFKRkf?TX>*Zu@DaC z8PKlR;dJOwD-FGrN^Gy?a*bgxxaT6y#3p+tBqv2e=+OYI`!eROKKPKw zZ5I5K?w6pymg$^$;SIcS;5az5GIZnl9B^tpg<4BvKrXNc9ygQ;7^X;23B)*exhH=? z^=u?85(vDdviGpw@YmOrwv;;ucl9c!4#bt_+h$G|4o1tWN z9;%rrPJ9*$gMCmZ?`aZ`pa&^ZR4}VQpOWJq;#2=;Xf>2Pq@H6 zPuRxI><;A84!$M*K@lXrt%WBPvJF>TsNml9HV|;f5&HgR(Q*y@KjYHX1^`hxvduKCf8sgPenZY zkvdpfOa}`VaW*rBP~8JzK{EVQnCl%#vU`WglXGSGd3*`Hvu_5Qh3ClyT`{WPpbd5Z z6v2I`fUQo820#5jmhl>MP>8+*h#$7+Dn({6W`k19L2ex}{alN4V*f*2 zx^Q0mzgSZIUI=8j4WVCypHPqF5;7&S3e7vZ8apRn;I=Q@2=6e41+_Usin`GU4bs!~ew_)<#O<=hS zA*V%CXiTyJe!r#}#jpH?u0@KXH2WE#WFt>qS1-ZpXP%NsF=?(jwTkqecf@<7%;}#H zS+=X89M#{-<;GQ{!Tz>xoeOVzUH@gKPO5NvUDwtLbuu{`b!xI5v@2+Up8qLdcVJl$ zbWa+AiK}_=#ityMdUV*cHtw|cH$Tg=f@RKxl6R5ItXbQ|f}{%DVB@ zn}2b4R~HWPoQ{}gae7v6Huu)9iuv!#OaAYP4G=!mizl_q<8!}F=wmxQ`Z*?u76v%s zeHU$!u7MGlJ(@|)#wrB<*J9cdu3Q%)a*?u&+vp~3anNv6gTaIT@Y2_i>b3pC|9mxR z@5gFdICqpb{?|&^9U26C-D7ZIdjZ#Fs>J@Dug&VGRs&i8AM9+qO1g^lXpmntxV4-i zl0=D4Id8@unlS+1GcG~&rA0Kqu?wV>Y^cl0D)2WGhM(*l?6^{rZWMMA_%|i#)`&R} zV_8D#cg28RMlU?JwxN%{>(SGbDsZ=FBKCcujhB}Uf|FJd)e%#m;gO-(tV;vtSD4Ym z3u2&o>N)bs-Jbe+m(l}Q{!nf8FJLUlim8shj7j;bG z&VhJNx5X94&%A?{@MwC!E{Z<4nh9BR-jjCWO2%*HDwKaL8U30PMoqhX==g@OQg)_*cZsWYa-zBwzALS!^(ZpcRsM@-PoWqNq0>a4n;8kM?^3(9pdnek+YN;i_W z8b;6S#c1IDNm$4^0q={NPdD_QCj7|9H11d@4R8RuPRT zc}o>)x09jZ zH8giYC^4Fwj2tG_BG0-(^eEy3SY17Y7VnD3&U(x1G^U-Z)A0?ile-to?3<&6R)(6w zRdN7E2lVNoxAt_=x?w235)L``=OB2kJUv{rf=<-S2j=ZPnsIC{{h)lACUj251APui z`soz>->#3a$vBycW&v$$dPaK3^=aQ?J^U~2I>dVU(ba1^$a&%AG(~X@M1_^XuI(0# zy?!`q)Y*v7u8<|dWjD#~I9ZtYW;<2sLv*I4B|W}g7+4O&Qby=_g5g1KSXPp z%d{=AllD&0u3NubtM0ara$V5I8Y!?5WkL+0o=qKY5Y(14FpsX$X|VWbOY>cd3tdKxp? zalCfVt7eNcUQfv9<|wAnUJA>^xZ?Dyi?Dgv4Aw<=Ik|K0BAOVggk}jCRrS;g@>4At zeZL@H_sv?rPU!nZY8JE&R=Iv*JS%{_OSr_yG2`&-{628fi^roDvgk#^ZM?;?3kyFl z!DmW~nL{ty@B{U1oci(^R?Pu;Vy=jnRtMu9i&atD8(lcukqR<}>R`CZhxn=%5z~dG z+@;pL#OjAAx!Yh4D{o(eYi?sG!BP$1c1(usqark{z#N*U)x*{6RWPtf39DRBg~^34 zU|;Jw%+nI3mmVKtf^RF+uSqj$-AZfvA^QO2`RZA^$w}cE_GzT^S1_}dJx{{l)ltV5 zJ34Ka7wze&AW~s1WT(SJlI)s7h7G748Y8uqk-p0IF%Hj8$`FAO6}HfLb*+?lY&Vj8h(H~*s1zaSd8F&`^v9me&m_3(pG zae8`Z5iMU)L3`apnD-)CC}>WEz}G&=8K!CqxD?r;z$p+ZCB z4#2WGeN0!QG{j8`MxF|jz$Rw_NCld~$CUlNgB|0rX8I`XZ(IQ)F;9r>eNkp!q7jY% zwuZiVI*S^-Sqv#t%&@=t8LXx4DbVM#n9hIeaI~^EG%rCo>q`&e|5t!YR=?tScBAN6 zVikNCkb@rmIvD=sURPXeR+sS8rtWX&cSiZ{ej+f5ac{gvK>B1D{hfN8Hr+k|<@46T zp;r<#X8AR=_L(Nu%#nw6Pi5$WY2tKNn+1I(k;kM_X+~3_opb7)M*Bx4>7M+{^v--K zUWegQ?&{Z4=(Oh0C6OnjwMT+&d$=9LJqdFD@@=Fye<9s8)(+324?)m-4Rp{upVzOV zf}L|LFD0lT(C%9ro76>*PkqZqC-k4;{& z3T5fYkSW3k(N~987V8;rUVHjm=KI6v+~Cv`Fg{fXcJ$vPHfJ?x;eTNq@4X-`>;6lU zmXw7WyRErRhu$z#{uD9q%pAE-U!uu-&wGqXVW(x{;R~d`RGno;reIKDNon_G0V``n z3MvWFcpFQ&#hv^EOVo(MBSn1gz-`90BnEum$PhW_^@Jz6l~IyNXA}J{q0{3Ps8vS~ zA9{L~6WRO#wj`;bN6Q~`%Z>Lk{eg}G-nNtUOB!Rz%L0G2Zz?x$ZwFc0B5Kvj6k=2V zN;EIQ6+JlP!SvrKu-u(3X5|p3%}TA>it55AvMVDtqme;35<5DNS6YrZ&#Wx8yd}&xwuvK;)D&UQ)Cn-TuM8@EI|X@B%7s2CO7og;b%x46NwE_Bh@i2Em9 zLrxtr!fPJHq7*YtkZJwPjAJGKQEGx!6%LV?f3mshOA>2!{WcPXrNR)rYAXG`FMw}( z_cOT{f1gv_@{v0sCxT|l?IEgj1x(EW3E)qz1(CM7B-7uMTwj|GsTYNc`{W6YwZyqJc1scPrBt9V1hf9d`V?Cl;r%Q&UPLWzq8>IBLuI7k5AL-6eU@J5K zLjkpm(8H;+?5;c8(KfYzmiPMeNcs9F{AC#&|Kwj6(8@hS_UnBDxA9J<{PtJw>4i_+ zOMe$u`sRB4^S`S^Joynxe;UZYy;lWkA2i1*cl>Z?+ywkZb2X4#J}^5t7=rJrV}nWY z+_Hvv`1)KFE(|{B=Os3g6N-0;*+?&UC2ow18E9iRRDG~?*Qn!LPut0VnW%;AhK`{? zZ)FIMs$t^Iy|C$=NcL;q4RBgs#%SAnkw%@Te4!^XWc1)HwDqDW;l^2RsP8QKabDoD zcAQ2Hy-tAUa(#HH!=nvJ`&pDG2{y;a$eX?K=$ZF)#vyGN+@z(?PTnGb51BJPJ~jN3Q_(1WR}gak{E!J({foly z6p?>%zsVL4NyR@>uWzh+W_5H3b04$20SLQgxz^g9h+qN zvt!oYtdWBlD>fF+I_<2&>ESz|b8H(fSkp*Wjn8M+I2rLwZ$H6Nv2A$J(FJzCTtipo zO48vTF}nW~LAxSn5N~Qk>W<5!dhG=AD6rMCM%D#Ct8c^$&u+%XPZjXgJ7uIJ3d&S zgp;0R;2->M7)bM{D!00sWqQhJv-LKfj?P`~jb;uGN;rvS4hw-dPnzRBDT26TuR-+h z7eU8J3t6Ua!MgSp!QU`3dRxN}CPwO@>219*;wZ?Ki5w<@S6j$@?;!S?r#Ne3agS}Y zNFg^Oo^!&{E!-9Fw{UHx4V)Gk1iPm?@K`h*j~qx~XAbO#?e+(8*7>XR+A7XZNj<@urgK>-^JpAA)dkJ+&VuluXm;OJH@3~_1+#k1EAEz7 z02dTdggkU5@%mL|M1&1Qnuh9l>S8GzTl^ODeB(A}q| z<0NUpeq-GXL5u`JtM%4l>7*jGV{R|=LHs61uf_7qOM>BN$`0T)xbf9Am1%+26#Dg! z;4Wx%i}&N*Y&>g{4cfja9qn6wn{+2X;!m8um%H+22J!r+$Hg^I<*)7!CEZ7?p?A3( zQrrV*-phMve}p9D(_2vHau)8bb!5j%rm_V^yIH(ak-OE)F_|;!V7Z{8VaMuNG%4K> z%^Wxf59a>FCvRx5S|tfQcA^r>YCpk6n&jY+qq6w*+fdHH-Hv4sW2{zv0NV##Aw|AI zc&BRyc3hOePBMvOZS^O!@c{+gq-JgY^_466+3%K9C3kD8a=i({{>X#FnNz5xcrr?C zj)vogqO97}W~9+HfrJbT-m$@KRzcPYDvtCKF8LwI=sV%aMMii?uaG%%z>T_##ZwjM zS>P^JLmF!fi29`!T-u=pZ2i(&w)pfDwo6?eg=ow`Zua(wCfH-gYu%{e?;MCQe#$rg zr^l?;;}HobG-(G z2HyO$6OSSDjq7RfPL_I#Rlx9LL1%0E6Qm|$$xE!21+nM$yfDf0aHYK)esgnR2(8Dr zb|j1{4T-=U9s&)FIFoOp;be8 zbO|4Kww!^^N!r-Y)t~H_n@gmu_3*xpQRD!!qG?OANZWBYoOULdS>AMoXP%kI6M8lQ zpNQIomHqwsJLFEV#&uxB~RV461T^u3X1xpFH0C?SjIFFHxL zd1S+x@H!HfaRYu2Ek>`7Zvqjm$IS7FN2qrs1F2@^kkKZ0hSfj9wNAI=+)K=vB4z{k zI!lhU2ra?(ktgx3#0;kWUL`71*v9*PEF7Y}hLOdnm!(xl9lxgvkxBV{&L<$0Ie2#h zDs8=v3?%kLzrr5SiHv}movEmN6^|)-{z!0~*TY1?{c(Gu2(^e>O>Tf3ep%|pd)X03 zCMj-#^H-F>`<@syuPA3`&sC;*6W!szmBGYGeLWWWdkOuCS%gkLs6ap8t)%UzZ-e8( zM##!4AiPT@FhfxqGLIObYHcGjOgnfV!2tL~}AUi(;p2m$s!i}w{s!^KgZO&u9eC%T;$F!4| zGBeo^VTQ0HM$h7dw-|bIPLtd9-*zOeK_F6Zq1fuE4@-GOTA%7rd(Mz{w-~ z*)@x&vD2s7VfMHvyX;08X)>}s!{xkSz z@`_tU4IuRJXF*R$xM0Ve5Bs7&z_da!%C2$4J(JA9IAk%{U-?5mm!{y?mlBa{QaoPN zVZqnk--ME}1mW=ov)^ws^a`}`@t@W_;n|1qTDNU@Xhk~Sb6%DedHo1KU*d?h|IHb(GFin~^TQf)Y_>gvM_7ycAD z_g~|ft#WM4OeJ>2pU1|o(t@PnY_j*?R(Rr7hTMO#Y|2*`w$<-EIwY3?nX?N(Eis-A z^`Fbi8(OlPtkdzO#}^^bAc4%#42JbfH0g(a6S^idkqxnFVo!ITWgp*5V8%yvkkUp& z+!OT{$Da#f_a~IFQBvQzU0(HEUconX;>&XQWKseS%Mw5pXW;hSVI1O~h>d!pS*wl7 ztk8}9tZttaegC}^7T5)_F?;REuI^f_^r(rvS-6S9yh_4vnwBY5odBF<20DxQ2Rn(2FOfnNTZicUV(1*z>vaNp5#ylH18-gvZ}IVtGN zv`)IpjDAWdiWTM1KFJlA<=kZjt&*%u`2eeER>U1NkKpDgTqhS?U*dA-e|W#OHGA!X zwxHjV;(O-d_{Fvi*7SNJ8~b-Ad;GBm={Ts1l0DDDcK>ATYB+-xQ&nR1T^2&q!3S{7 zN`%gzoeoXI$>_$i0$g`@KAw88lBb`%h_6-q6HR)5hm(JQ7TfM>Va0DHvoRM_*x8R> zbH04MpnfH(JE#G3+P8z0{9Wv( z{f5BCKKy3Fc6j!*8b>GB!O@&i5msUS)~E%OEe%7DcQiCrMb{n zLt@B$pFY$%F&`Cto{wEdWf$6$luo&HkT>ooD``mreIGIQP z-F!;i9;LwZ+Se#XFdHm-FcJS~m8J%<9pKz{7m8k0@fnw=NSeems~+7%e>KL*ispZ4 zaoQJ{{&qgQBJuz`x?hIBStXDu^F7O(u~v|;_@D`;Klum>9!>1dYmKb<0)002;!^%B zF%8S&yi6v`_Zizesgu>4+Rxtgvu5}G*~+%YZ^8H7Yp_Yg13Y-$guP_hfS(M!L_uy6 z@WLhs&kjw*dlhct!h9F{Q0*c;zTzPzhH`jN_j;|)OELJ1Q_(I#O8BDd#aQ$9L;QH) zJZsfcEI7h;!Ay^@+!Pg{OH~q1RK%#BWui#X`Z%##y20O7%U|RP-9ltB1i9` z4~KK{lC^@Ey#{$=_{)Iam7PNK0wrj|=s(gKVg{1&&x!k;HK6{}8+PPRhZ0#Y_S3ad z9IsiAjS3t9R`l}nOr+S}p}A17O9uv5^^>)~A0p$?Y1m%M4gV){gXpA6(!)#vldNEh z#pmzAZ%$l7yMEl`hplPgr{z}FxOZm)8z~BMr>n_H^HBOu^&)l1*-g3g(;(rM4{UoD zNBXYpM^_8l@VjZMtWW7hq!_meC##s@sTY*UlO@lHvTQ2XrD03;1ldid4^60b$b9Nh zAx!rQ9mIdekKph-cbO5(P)y$xlE%heOv=V`IIe=xTUAYP`tXJ`Y0Tv*W~_qh06EO| z^`JY!URdI<0=uqSu-7~xOzG%GSg>pjd_35Q^uk+Et@bHA4%1;_Wiqb#X@}O=TjR#v zCipR037vDEf?HW9aXprYm%Z{tSDsy993BbL8}YNKi)$gQNXtd@tux@i2r0Vf*Kuxj zO`2e5{t{P>Z3FYG=b2zPE%u>+4?LG@j^2Cf5Q{b`a5_&=rtxIFw0;|!AVtvQRpMB4 zY!HfmU54H&F}Nao6n}To!C!mqNb0BCwI*x3k$tNQr%>rb{ z*b*s%ec!TR5iSJOLXe&GvKH@ipUnRE#emf@eT5}klZe+`ahP482*F43UhPk0Ej71v`kHG5Y0X$Ow~ z+Cv_mn85q`dMTNyTnLNQ0&pWei}dYx!u*w4c!s<-`)RT}o=;AoyrWGF6X^(}Ub;{` zJRKKj&n62k&Jvs5?&RdbOenLhA@|HwSjz=P^h$&`T^7qw2f1vbRy2><^~DYyQhkbz z$23To_7miK(tv6@-i3fK=i$E`dGb6%g5CM}5staNi;C-Sr@d2c=r;!+u70KkcXvrA zqb5-XwF+Oc?!Y=$WVr^Y8cpPqBLp2Zfv!|y%Xe6nH3%D#6n5Hp7<*|v#;UOq0-mG~ zwDroNVRbHu1|9>AV=?59u_yhXqBD=C>ifg6F;nKu8Om6ap}KePeWXH^%+jE*MoB46 zie^GlW+56-#$*h4?|nomWiFMHkV>NljTGf~e*atRT6djw&)%QU`##T!3#-}1t3T1{ ze$6zyY#){T^qqJf(>uKp_ULII4xii zn!0r#RPhu+^NlQRSfT`^H3+pq_>6JfGmuK_2X&8 z->nR*|0ZDZ!**;2Cpg(vy(@){~0@jwq;Jn0(PMHYXlLO$0W zs9M1&YN=pRT-p?n>U0h^yR4@tgnaOq{3YmoeFcAo#vOtB{K<4%vKbP&Spai>*wR={ zF=~+UnAnu40+F#IAmT6&%qowBpCyjb0>4?%er^rY>$*wj>nnoaNq~JEyn~86exufN z-@{seNh)1y%>Lb_%~~26vnQWA(o)XXqUumZ_wSct6<2%E+fsR8NBA$CHd`5GX6919 z!7Gy1bBMHgkHE&dDa<96D4?U!DG*;`Pq($_P_34yByx8M{rxuwb;%t>Eo;xx5BVaj z^Fui{pzST#A^VlwZ}&30%x)Ml6Wt-vaV#~mwMLaMz{(|R**`t%Ssdyse8wF+3&^tyJed63UN#zP`9~GnSiNApN_xCvD zq$Ta=m^}6Qcj1!dzNmU5cOG}elt zaxSKwwp&R{Fax%xx#FA^Z^;k8bnXn8gvb8lGM$@VK-b?J$dfxs)H_L(o9(p|gRZsc z#mYUXb*mC9*qI8)FMcE=)=9A1;}SBdKZq^gEG0WFPoOn>t*Bmr2zkhTZ6WJqQP#H{ z=>Ag|{nva4xqa@SOw>(yjJc0nW<>D7!xQvC&InwN|A1$#JiBb?9wbsJ!tEH3K>l}S zaxT?|gxB4rB4T$)(&Ga()O`^z_$ixlKmrs+_gGMPC0rYr^Ua(36~n~dH#(aeq0RKY1V_;Hc>G3Ek?U8 ze5cCu_kg44h3UFf6J%0oLh5z9vCk86Aa+`g9b2@9{k_JI9Yy)XYIp{HvcH!`ns)&M zrUKS^8A7$)iEw58GT?vV;auBOeoW7$lW@pT5qvsggEf4!;Hp2D=eixf3$=6$1fg*r z{A`yw@QF(Sb_Vr3gN_B%Uy%21IY&Ec~EV5{;ND zaaTcs!0Ve8(q3|gT)f?co$M0W;7jVXq5|{YTWzC_Bg^o2-r^qQ&aZsCT26)>6n$znKgcTA{ zu8u!8PLtsay=jE6YM)m(vfAME0ge}7Zi1)Y;h3$V4?w7sCTyMijh}zxIsJ5Dh@4nw zK`xH3Lsipe5%<2gVC&XI>c6oX1^VyD4G}x2N=pqw=F2L>B2{RLza{+qDGmG@1azf$ zIlcVXk~NdEr>@u2$g&m@3&y6W?{_ppK{h8))fa|N(Zk~B< z0o7ulWQQPm&UEsjE(OP5Z=#LNL+U+zi)4GtGC3@-#+1wPr$7~YXlgCF^;>~3BL4V~ zWk-$l(GV(Sx&q$({fBRLMIPbgdt6@ChRr3J5^uQ6ItY z zisxYUjNU8$D+tk-A#=N9sdl|Om+LO%*c=T64SA53>Xb-_P-XN{Z3e zv=6XWJpw%3e*!#=?eWtIP#1Y6`0NHB+7*rWT4{}zxeNSwEf0QfxS@* z?r?a{%M=O}n3ZMDEvr}o%d*<$Ewr`Z`jFOwhnLJqtl)uu%-blXvI;m&EA zWq1*|IZVb^Q`d9+nrz^@@h~{Mw4*vCj;Wb*GXWp}+lD>AUZ-iQ*(CYXCBY*pDfC2A zm&{LDfg}D*MSB8=1xcQT0tYDzy0WeU>}vnRn|p1XB*$#R6;odDn?C{c?~D$~w%x=3 zDXfaF^q3^vH)>_i>Q5GpWH^*l*zhmEyEHtgk;r7`e2F0t zZOiY7J1{%nD#-fnd*Ivfz<~EQINL zXv8)RCOfKufS(KD#bv6%oGQUe5oeh7yBYciO2MzkHiMX{D%7sv5*djRrYauwr0iWB znLaHQmrU6~2mMCKs)BUlZ_Tl(TwZ~&zjx?*(K{r@C7djlI)Qr>a*^A{tLR-BA4Ymi zA_4V!R5+IKe(O3*+m^@Ey;lsWW=Roz5|u?Pxl9u5t6l8hmZ&zHnw&a>(B8%w~JK6MVyN-`ce`oyW_y}n}_KZwPSQml>-ut0w&2< zomsap28Jhef`ZNJNX;jex?HZO{hbYD$kCpx|8x_#ceUcCyA|-h=~dXRlLNMIwn3Xr znh_bEi66h)i=w6~Q?p!8@ahy4rkk-BjflTW?^^vKCx83UrvB&TM9>oA%W-+9?T@5?)^IF-q0L~hNe7NC zy$gq*KPE>O%ptW7VQ|>XnrzsYC-}P7geSSY0;af}frcWUaH+;#s`=&%SvclEfBocY zRrfZ~;`{~VUE491QTE_PFp=4e~v0A(XI?q@eq&}d3lr_5ehEd_&rCl zLmeW=OgJ;F4jQGNgQcU2_|5(#YOrf9{Zyd`j<32$t+xlzvWu1^MEWC^AfjaIjTG3< zghH>>3Op;yj@_v7jINZcB;PM=Aw?U_P`qOn?7!#4*7w$vzLrgBaak4l#@z|?lA558 z*)?wWRRpbOe+6qjxMxe>1(q@?#5*CM+Fd$#~UIlz~L-=apUa;uH7IX+b1=qF=5ti3ZljkPUw?Fq%=l7zc=iyT zx}-^mDi=bbnI(Aq4d+4^>V(H)IG&<;H4NDJm!IBjNkjj;0j6KMjw7xq!7Us4l&-R% z72XF(B%VY04^GolB|UVM5cgR}Br9nxeaBv&rV^ zoaJV*0o#Bf`$S7iStMJPMEO2Jf+ar5K~loMXC1Vw_EHv z?X|2?_AeF%H?Xr;w6X?xa_!>7Us%p4LPcj(fUh?1z>@Ge%oO1p^v~H$dP6#fTE=Z; ztiOq(i7l7-HVJd!&AfEDJQcut+mkR{eKw5^*JRz&#MtEEB(`|!N>=ek4&DDLmIj7T zW}o}JqB=*8=X;`&s`~JYLE!AmXgrg-I;%ze1}rtmqnl zTl}X(9`a&tVYM5x+0tTVD*EsfS|Nm(j5mDdT}d@=)^DK?zRsnNV;ksb?pqqt{EEx0 zPpef#3u^gx7PZS=0#TdG2dFG-$NZk@&dwT%XLbE&k-v%?*p9d;dUekXc1yD?>pw%3 z{VM0jDn2eoOM^9-LUj!~%<*H!KGe|akey^`=~r;crWarS;DH+44q zJAWtjmr!SYbQ`Epc_&T3{eosP8Z7-egEbbCXM!giGudZlm`cx2gbc1_Z^SHNStg1d z^a^90ZS&Z(J@d&#IF~VM+(jfR-NDr%6Z-FkFXnwOf+O1Q^mo%nq<`9+6?K)TwsM)Y z!tFYFU%H!h?+Ih$=I_EfZ^y7SZv(k|Nt4;N@CdU~CX-3e4Trr#MO5A(gDmaCFysnL z{bX;@m~>f^c`cic8`qHIA>Zjb`($=m{$+Ob?|V9=a~VtTj}pWbETQ-JZ(~#UCD6m> zW&HCxYGiG1A2f}&XWc~8*i4hNtbnJ^#xK}M`#GljZ8dqeN2-MW`=Z5q>UV%uZ`YG} zS!XJ$>dxwZtfaK59nO0*i7gA2qR7&h-LcCY?gxmy_evK@l^fCg+}mWwwl=c#PXN7n z&kI~gy#QaXyh8MUM^Y1^aoQhonKsT|$$r`+$9k<0Vq+8HkVwBU(6$nxoo>SH^!@kf zG0*Mn_}LTe!7CQ5>HJA#*j1U8n@q?@c#C{gsG)khq@dWL#kBE0*Vn7!e1kDE;H7~r z8Tva9MEr22t%_~r(|`f1{b_b>mZx5=+a=H13kuwwLvtpa*!LF9*6n2f+j){5WRuv- zK3pa<@fuQpa+Epn@()!Re+i8G2I=w$E9$FjL!Xt^A+_N+JpObk4f!VuukNv;-|hxe zHU1*DN>Ptv;=P0hJ@V{(EpvDcyha)=OPFK(Y}id7hUv@UahjNM5w&U=qx_J+BwF+X zsoLdDWK`lYWBMPvKwpGin7IvowD-hws++i+@&;OwR!7FydonrtkubofV6Juk@#@QQ z>b#Y&?=fpDA2SD)-Z9r=M3MQOJb1*s1io|cvKW^F~ z*gLRruEzLN{;K*!@?ypZoX~V!p#M38U%BKW&c2e1SMQ&}_mqsRaZA+%Igdxk+h$QD z@{$GCRa?;V&Cii;3`VjCUcw*C6d9%52{f@u63@M*C(u2ThxH3B@q%j%t|)5|Xo}>) zK5oBLdg(M3`g5@6P5RGNfLv%7P4^BfceJ2TYDkIl_JCP4Qvf$u!Ie2tZ z9!`53$R_+splRdYv_ftt*yWbXce^VX6Ecy^S(DRk_H~)c#H+0ZxmZ>dl*ElfiGeX^4%u zcZ}=`MQqc04Tg)g@++s73cBZs3T!pJ!4vg95V}PJ!^KK;a)S*0qYX$R-$k(e{XM}< z86)!X{cCt;eFofhW&k#O_<;ZR4FPf2cE0|`AN;FNJmwvDY^Vty>L88BUcwmJYglin zo1B|-p9Gw_Os2hwqfuXSKxdacS+aRLi4CbG4oDgP9Bl_59o7JMSp^(?J_CR3-iPNH znZgrq25^_-6uQnu3p9$dAhjc%gbd4(J^Ajax*`ObyqgcFSI9$y5pC!>TL_F=rV+!4 zDX3i~9p>pnfTtJ`ZGoP?LBP#TQcVWR8e_FmG_}ZpIEr%9S`<_Yg z$A?My;J?$rH+7QW6hp`r3qO2OHVxSp`GD7VlfmPW572Y*_eH3Fi5K=hr8HsOqRp=EbKl z_~RU3;AYeU!hXDh#&0kvT&@piuAWRc8i+*st--P<;W@yElQ!@312q zm~;|$@n|7M( zBn}&rUCH-OHxl+Gv*rj^6Wq(Sfivdr0uD;{aPP7*!Nfj8=6VB0ebdy?gENL)r>KhR z>>VYiuenpRk!Lt>-#t|QJ_SonngpBex1oi*epml_C=Slo7lU2I3=jQ@fyUx$@S-&5 z3J_wLQu-8KJ@W=6F;gKwNEofz&;@>^#?VX0dNEitmrB?#2S8?{9(QSqb`V$7M3~a}}47UPD@3Oo&ss8mJuV#tVwWz_XSZm=LiE z>6|@+4yzXlnoF$tjnR2{Q}_~K9OlIneb)g!drQfh%1E;J?n1nEz9f8|bA}jf`V7>E z)M%i|CEAy|g09#L1;;uy;7!>9=x1;Nw0Eq8M-HEWR}D_k(*bws>YL)!5&k77&w1lJ z*UZqvO;LQ&jG18L^$M&vKae`5jgvdamY^GkBAlc30^aSLgu~`1fT+2a;7VvedZ^Kg z>56Sc`kM&7@S%f*Yq-)0qcYm4SqiqB2*dcEDp+$fpS;z1OlaguvhmO#GG6FTZ(}1k z_tPGtt}l(3$i5LYSY(l(odsm@l^m+P*o|$R`{5mKW?=dDBDg#CvcR}=9~S;I$Zyh( z#+%Rhk>fu^sP5`N#8FO|s$ZRe^<9SOy^8|<{P#soJ;#mDO*x9h;+zC~Cs_jT(>tIb zUWgWiNFa+R`}i69$Klhf$?(UoTxclvh^*h;OvIeUsA||S|MiJLqMl?AB<)1_i8<%+ zhS+dE|IvJq+H(*qyQZS_hn(|$axy$H8S}LoZju%GE8#@>IG@C>p&f>s;nyu+$@$F& z!1m)6aO7wUa`?%E@;9VWK@^2U!>Yg{+Yu_aCV=;`>fnpY4rtJE8W#5iQ{{<7LY)7? zW@t$c+?;}3IIm{(5r0B$&B*wUX7Z~18Lndh)Of+7m^~tbWnQm9&6o-5xON$alrBfA z#TAhJnoe$1BG~bZ+lQ8>fdQ-sEv`pmNxNu#=iAmwKQCiwtl|%z%!{w1lIB&<^+(fv(Az2{(B%8FWg^}d6^Qhv)3ZgL#;Zcv% zScL1KO8vUedsu(8hBsXtL@w8c317eB8-Yn+*GW;@t%JcvxrQciNh)ek%tEvR& z^tCa}2mqp7hh@c<$)HhW6P$GX7vzRky^(>mJORk= zUIkATeHWNHRm}5@YlUGaRKWT&6UeGml81Nyqwj`uK-Iz|iNd;MknWzJhczk&{4@=pXL z8l?dnekv@OZ%lPPF4EWbav+o+4{{pbVYMS~@SwW_=O0J_GmVee3}3d!E$ahE#v^anLyU88qFJBkqBRpxbh9oZTnj zUpVZI2eLlk$tGWk)#7_l&zH~jsifeSJwF8g;z`7^?JzWVnnNx$?}z+5#;EmsE4198 z2_H&cCkw7Vf|sS;$i6?~Xy?v8EU+sCXQOTlQin7#4u2-dG+9Tk92FN#zWqhuuWbYU zax_tB*m)Rvz*Z1%*No2=>jIG`Z=htg5*7~~A+onF6Ir!%;#AD}aStg|!B`G->1>9J z^1Gq>w{ViFbV*P$z8LQ~pajxJwBS{p>lmw?1kE}PfOIbhF{>r8U8@T|;cbU66@Mj% z`lHCAJ=<~M%s6DTaEN#mnv(A0^U3j(Z}=_-<4*qFKQn=ah8OnIS^_H% zMqyIy03R()qi=Scflqgwmks128n z8j;4;Kf&IbDiYdV0{`@6z@Ix-!m`kD5RhtxEnn>>VP1hWP1J`T@T;ZLGoIle>e3|t zp=54` z>o~a<-sjFOjV)O&%v3`)1cjjS=e-98WK5j0Qt5>Bk!GxOvh$r6u!tB7EE#{^WGPbI>p`a zX!RpvS-6zU64?(9@9^Rul;a`cTkbHlI+%KUrbhN9P~AP;M{wmy-`!<=-2={@wWzfzF5dyvsf$u zAw>|xw1K1xhUkq*fuKU;A!Gtu(EZg1(C4HTNS5m=xR1G^<0g|)aUJJHYW@y`;t!%L z(ceI9xEP~bs>^`eGf9U-0t^;<5A_$s!hI)tQKOd=^Ph1f>PxqVu7RC38pfI^$@Mf` zpzn)5WV^$CAGy5hL1AXP#R3p?y9kauhJmJk4cJLz4HW4)iQdb9<{wZQ7mVFY2g4ti zp*MH3k(qESIAZ+>9=k9Gq|Z!)QoeD-w2yP6Up$6X#T6LzX*DC<)yJ4`+6V7jU9XD& zUI&-A%JMIET?LU<*O1kqJ@mcnNO;lkX>dtqiD5|G1Cd*r?$MVc}bl~ zTz>(&&Msss=6f^7Ir?-3Uze^m*^aVw&ch79lgLxD48?43gL_J}$RL`GPAf<=zw^Qw z*}Z8@c1bq~6N`hWWIj6c_&Pi>(+;Vxmtj6!lR@mNdU(!m4SdlPgxo+oDr)|STrWky zXeNy6c+{Zcf!k1}PLg>gftdILQz-s#Io$0#NE{z;B8BO?=<=TPXp_7fX<6b4=cb%O zJ65hnO^%%~|H&y>vEPe+@T*0$-z{ctJXZloP8IG{b0GJFzkq84?kMYtB-0eA3}OFS za&PW_@-Qt36&_xUjOzE{7NO4<)6B8@)@%z<^6LG}YH`fQ+;EZt^6BmS!- zW_fk!{Oc2ZovXrBESB3TJzNXi+#Z4ehl05SPVwmXEO(@MY!Z^B=C$NnwBJvg`uu!M4Yh691zMkp{)!Z==$nH?j7s5_ zwOVvn*GX{cXA0PmCWGbG!{KUPI95-)j3Z{o!b@MKVB3y1L9G5GzUd+bxZ~L-l5vkA z?`5wE9Lp*&R-ginO|3Oweuv;6qUu;`^9lNQpDc6eoHz_%kFy?AD(K>0p;QOVWA@C9 zK=<}IBIl@5GE|y^rYVUtQ}$1VA8Wk9)ciWoJ9L({b&6p9WS-F@Estp@&jj)BEkYlZ zM)*tJmQwvA&&h_k*=%*1GCP zx!jc`v+q;`Ef&s!+dZbhvqmq7S77o{aGBU3hCfYD3uQdyY;xa{C(SpP1J zWajRLpH1bEyo~|ste!!76#ej)JAc9Liz&#}fx_83(ny+S&`-f{1oP$wQZI!lc2>r6 zR*3VSI2VSq?On+%f3%4nhlglQNEOLR5aw-P!e_hVh1hMJqkhj4dH&45Jh1VyD6|vi zWGhwDm3@=g*($#nIC~Ng4mP4Yut~8gU!3$NY)p7V&DB6cqdL}KZu#JW~ROL_eX}l zh~*<~;6U}f2T1XnGW>aMA3yI}4Zpk%fQa#WI89uRJrh_$=tfs+80bYzw>Kas_z(Hd z-Ap$Ij?oX>RM?isJHeYX)2ZF2Xt<3hg|@6d43m`2_>O zs@28K4R3t~#y_9&?Uje{sh%LX@pvkm8kxwRQM6^1&#M#XuZr;1$Lk>U@H-rTO@Lkw zJb?laHAe36I}mqM3yEE=r^R*Y#3e)%JT!+$rr8GCj7=r~ZJ0z9J4_j!vE#^#90Cgi zufuK4El9CCiOL3Z-0y=5C@69Z%F{^0A(=^V083cMiQi{_KK_SheODua^Hk9qjw|WY zzKuLh^&rnpFmbUJCL2UL&_vfs zv}5OHw9#4w<~~>e70Q3|674R6OPjkGGLXRJdGnbIeswh?YEi(vy9hPCVQAfWEII2^ z09N|zz*_@|h~4%963fj=z}{MPaf&^Y-5tyHu7b?p4f0GD41n4RN$`oU6YRBBVh&{Z z)0M)@Q0a?Wa{KIIa;D}tc`^DJ8!NivSKf|fvz|3F z*##H+W{?371NbH86{uVnhL(-CBM-HeOmUM2Bk0`2oZN1Vq#DDJcUBgsD)VMJro) zJ+A<*nxo3xUq{g@H8I93IGs^=yBUe6^I>v0f}Z@FFoAQMbm3s2f}o3^{8RZerjYwV0*YGJnnG<@ZStETG4Nrl#w20*~v1& zV2%`Cw>*;*Q)I1AP;UySOHtONftgz0wMG4eFV4~+Cq4#ZUve&LxM_; zafhG47w{0*@4CIHLh#^mDE|^EBZimn<2xen!OW_CH13xiliInKagvyVb`9&IM;~s& z^-;&b8h>TjrWcH}rsfGmez=gx2lhnN?KARb4H@Ut<7lzsZF*s~Dm9q@me!-SC}90} z#H%)9?(W!tj%J-iGu>*z!s~iaM{G@ck=?N{=46`%tZJ4ep~cggvcIE(i&+h5d+EMxA^eA~#9oauPVaRnG-wI9AbxP@eK zjE;sfakzY`G5q)(LGz8HG`H7+zITcxG0!6ejUUIr%`jc)l?eIILX@C0H38j&n9uWG zNiOaRf%Bj3wV5>$XmJIR>w5Ot1%YtA~tRpaOR1(mxJ`&h3L0yWf;VqGC zctBw@Jbu{)%vrGxM_a0Z;Q!WxvK_)e^Ve~ha5o&*NwfivLps=FXae+SL<^pAnK*Id zQr@Px{nZM7w($0&ukgP;?PRM?0nz_^9X^WWq0yr`_`k@xK>offHnfa^-EXS-!Eb5> zhvuyZ*8OInR)#>O8y#>=UJtAE|HJE4D*zbH0$tJ@Vdc?YICIJ%{3P>|4Awd0dqp$g zpyYPE>|z5rIaJ0Ar|1vmqlz_S{*gl1GH%g z$5Rz~DzKZ@2%!BDpsVi#wvD|6_hjH z(lS;nK!f+7?v)gv(_IVFi)3Kq>hpBnBYj%I-2sBNuJX~YSe$+egQ#FD+?&b3KJGWD zOjDk3^~f1?&m0An_tG()sR%@ucJMlHC<*%KDZ^myZfjE?1a@;laS=wFlep32dUf(!b7h(53;@pqS7a79-3VvK8xHbe@qy3 z6`uif_*p>j>`W3Ass}bd$p!fRUZ~J<92B>|sZsiw3ZHuxLLr@Q{?f7iRm1ZoLBgaU zft>VNG9#Ju5$+M<^1{==b~P>1_3AQCXmEsU^11!Hmk0uV_0(c4l)87^!XB1N5OBMc zqVHFMNaH->fH%Q}HDQRC><51(MS`FQ27C@>M9Z}AkY5drcv4~`n7_3EK9gM!jko6V zQ)TS&GAHrg0~zU2llh-v+_| z%(!{*EDKolDw(JruR{vX^Qa@&?~dcgBX;X|Ecc)QW!|ep^6iEAQi%!`iJU>FjChll z+YgwW&Kkz`Squ}R8Nn!UUBJ@|-lKbgkE!54Ez+Zc(3W?{Y0r;yu76EyG9}pNfc3D%#EeL}`M}{@1z5de0c$TK%{okAs(S~ zXkYCg!H3_4f`R`kv9(t_a`63viiRUli_A}cVR{W<{~H3bFASlyUpdg86N#L^eV_?? zIfC>iPk6xL5?R^ZNtU+LdhzBP4)IZABES5;>1@%TFV(hi{RC zM>|?1XTv0Rh%qIfWMSXq?NH|3HS)_p0R@Cy0lk-c1z(?=;K8@^$e_%9qBf$AChH8$ zb^Jb`%a4yzyHp|8)OasGIGG2ZoUo;-8w_a^4=}5yoZ|X>S@h^_Bam`Vfo{^FWZddK zd9~P4kojXC*UchuY2#n;OHB$2x0f=VudNv|)qd3D@`AuoZ>qnunaqu8gEL=q_m~M+ z`uU75l}=VggW3Net7S~{rtYEk2S>3=%1;t?#gdL6?+4#HcS3)MD5Q0<3{^hk7;66@ z{xBv=#pa7LKKnb-qY8b-%IY$*+_)Ty|nIeCsZYY;48RGf`pm8|;CPg(N92csm+7H3_eG;-hZCEll_S z5;$?54zvV4wK%kUl`o@o zU?)2)Cd;nSA@u5wThMy9FDlBDK|(UF%!BqDC^cvvU0&xwW~nw2q4U$I&}cq0QzV?Z z%hO|WbAKYO?-&Ja5yHAtX0vaqSFtG;`BdBD3^e&UnU21?g`QnXhVMENAul z^1Z?&$7@H+`l$HbXx`YA zFc{nwiaHI$i9BRz_t-GNY@$ z86{lMWHR+Lm>VO`h#+HBDfbnX`yB+cIo`&nTp6UixfQ9W%c1_QuVD&Mq^CNzLW9vi zFtX!6P$s1XSC4S>sLfI2sLU;>a99l_C2K;7{;&BFxHBn@Q`TVY>{S=Xa96Pi%qTdS){g zWpkN#J3c`E#`Va@odxHPh#|d)8i=>H746vL&eF>5tijAwHsH1wtevVxPt0SlBKsBssu;MHDRgvUbIVX2Sf^5l#!kREIe&skxC@= zbf^VTX%jU#9f=Bi#8AL)9W+C(klc0qh9#!XLeVinD0|Z@WVXScd9ps9nG_d}b1!sW^`1=jJ`HLsczDN+a5QJ+C+fd+hpwG;k4_p= zVV3_E!!Kk80h^MI6F=A^Nny?#HRlTC%;?l@n8jhaU&-o4+N9l~8)loSv8o_hOQKI@I!Vl4X*yz(MQm)QqobKc zRN=P?qF*&u_`|n zq2JoGnC)Lv={_B9_9EO)CAV~fRkp+E>9_T0rBezSsw%>JIsX`aq=NE33iBWRE`^sp zIERJ)XL_BR2ln!&(u<-ZG^*tsb$!zWM_RlPY+k}S1{HB+$RmE-sd+d^U_cCY7BZ!a zzk+HzL*_$PAEPzY%|u*nC;R?Mf}$1eMCAzFxVz6NJQZwI-3uey zCo{@{!PMADl{H)MP1gv^L6HTV!^~lfcKXk*U9@s??NJxgTJ5ao?DT?{Y?XZ{t1*$x zp5ZyMc;q7c-CmC|%Y2I7o{8od8vEejggAP2iqD$yw^=I`@~yk~nOb)u?mLjXOuLRB zCrqjX96S(#9^7$9N%DF${0_(DIa39XyxqwzeZQM*Hr~&8(7%i*zk@lOcAv?)ImEdH z@1x+Cr_rtB^HEk*27YO$$=)b=OyV*$fam!`keA;BI^*7g->Y_`Znv+`_4GwN>jTmb z^uEJZ6sA)^?d=P`8F9pWMJBx1X%tj)qYmj<`3xl`)$0UENM9$Qg7~y8% zIbWZ7$?@ZFH67+RStpQ*87EMx_!@fro+(O_OsA4IgS4pFnK3rm2a~5+GLtlplDEq@ z(Z@=X%-PkWOnPh~V{k8&DV}$n9nyWq`6Ii~f2F2ud88R# z16#Fu97kj>%#oZ)zW*!XyGCq8dK3b~NqhT+5K zM^VX@DC-+%1Ffh3w6N|Nm*C|&H$ZFdeo&>5OH29Ri0wTa=ItMfw!{aa;V14$#zNWp zzQF*~7v9Qj9_uD8zZ*bXN*QT-H$eJ6Ok*+*Q1rlX3)|a1i`DTu#h%4Sb5sx5hse46h?1Yzgd2c?Lec|ihQDSr$V5;c4K=bG5r1t9_zi1vXsNoq{#wWi`cui%#o|cemL~Mv7)3u0mb0gCm$S<@-$Vz20<-S-S#sVdA08G-!L|iu z^pC+pCNablZQDDaIkj&kzUrq6m%!^JPR$67#z+WwJqCI(eQ@fytRsf%U0&J>}e z!ODzhb0#Y8uHriCTajNt26K6N9CKT%7JgfHiH!BgV|Ctlfo{76^C&f%IlAFJSu}SN zz5Cmr?y9PwS%2HWtHCaw;?D!{WPcOh7-R!9pM*ilZQO3cYBM~sH=G(l3=SXeh2`fq zkx6fg;6P~;tci_4V#141ugV;Btzt8?jBa2Som+x-%O#@FVQCb%53$0B>R~qLenvgQ zOht|{EF-Z*NywCR97}?y^DjY9&30sFE>5;yIfv%)8j!W>Q#iln9Q?Q09qoFj4*P{C z3v6^_NoGzZF&0^ma*NW@;i2_p{6RCymDxynF*i|dSvUG5FUh<)d4Zd$B%&6d*U-Z* z9G$$s5j`rOi-w)fAl0*RDA~0K-S>>9{-3KsP)LJdLFyK|M_mtGT2w2YSWFGl+)g*^<)*4fD7TV2WRJX zAJ?HJm$`f1T0?r^Cm-2fWZ}NjYp~5p7?*7mp;JDsA!quX(L_uva?AY#g;ySj3mqZ= zI1-Bb`mE8IxfwKcY)APEe96Ck0%E3h3#V*7iH8liOx>sm{8KrLR!u~KYc6(Zhx-du zyl4gz%s+&}(x#zf6TjixgW5FrSQg~p6r+`z|D)(k+^PE7Fq~QD5K>4*C{yC>wYE@7hKhcK5|R`eG%F%wB6CuPBxD{c!`bWI zrzAylQG_NcrD&izsc(OO!rAA#_Fn6KpXa{20!~?+9NNqPWeiH`!Yc>PUkv8WPr>(mT}E#CvNkL2L< z&GRH8-5%s+HZi`zchE6u8EDec#y`EeKKo7wI>cNKHw3LERz z0wxfDUpq<3IFAp{aHm(tifP4>E#w4?EVy^`V>h*QdYjtRhs5SZqE1KtDf{gB+mIcys72eqZlqqXm`nQAV*W0D~~*-Qe2uZXdK zBix`~Y%gqnE6G~dn9<7yh0w`q3kk@60Qvia;Ml$#RfhO8XihF#$M|BjERsw+XaJwr zHQ?RmPvG!M7VoJUAzm$OFl8RgSaIjEwkt6ZG)s&1xiUjgRJoruf>Aa%&KY**s}m>r z`TV^@E7^o9&Rs7piEKC7(2qoV=u8=kTsB)_wX`eImag$=5IS&UlV z$znD{1Z#Zay4t`o*01D&;-DSW&-@5QKeE^k2C<(N#Nm)oWa6bFPe@gI@TEA;k0jt$Z))bO|Xt*>va#Y%O&Px94sI~SNUk^ z?Ca>8+6p+Dxri=2cMN51ucG>E|c6~!F9E^gjs6bYH<1iztnuKm2s3On%Z;~I{)-dG*kozNF(0Y|J^s;>^zjVb( zI^@6_y6)Y4`qNhjyd!ZF>+mNN^~EF~ar6AT@ zOJF8AijtO(Pz=te{!XxLrsr=~(q2X|eNr9ia#CjBDkG}cGYu;J;=!>_3rqzp>-*y^ zURy8LSCuU&QD^U)r?RJf4DgRT>**}xllasvz*YMh z==$r*#$7xO?qec!#nA*jrrv{9TfO*^o63>cxf{GGqRyiEUx~3cQuCm)#Tg zBkrHO1Z}uDLA*zkq0Rd+Mc*r@?#BI79ad=y`7x&+No4-R^@%;ivXRy4twHRA4*SBk@+;ExL|8b~K=w`wt`l1*?-?*(c^ z(;E)`wZ$JTcc23;PN=rn9(VNLL{7J+(6KJ2w9M;a{41b@v~0dl1}Zn=mpM65Z&gP& zNHoIqgH<>y>j)#gSOvv3ax8II9(neMVQ#4&Aqq!0=H9P0kPvbj`DYd#ews3Lf6s)POjEY2c{!PJHX8kIc}R@kJ0owe`KX@lqfWN3#b12?Ve;xCW^Ok! zkv0yL%c@(PdoLX+w_G9T<*ay9Z%C0hi0cd$d!R|Pufn0(8IU|n23-9@No8;Xo|@0c zNkYXiXt58?yE%(D`^-P^so{f1;e8n2<;s2IkK=8!XVCgD&~&G&hy=bi244|ncJR&; z*0O31Tjei}|4lN)zPv4X|C!@x?wxMz(xOB^;-1ys94X{f^%z_irBbw-HYipW66-tr ziL~AZLXzb`|D_z9+oBDV->#!9U8XWK>~+}g`>L#!{Uun}aT%_CdIbfh?{SIOcKT1k zOB{NBCB=c}iCzDDYGIu!z3`|OZKW+l?>elFXC=B)DUVqE<^w?%GtBUk)$8zyf&cH~ zc#z(m0y`EbBhzaw7CWTF(H7K-e|MYHqZ)V7a9b4nro4a+R&QpDKGrZ&GtN-ik3uP% zeLlGOX#lJ!VM0k`05dkZ5rH!3_kQC>m*^Q zcIhRZ;dM}V;~_sfdOtcKc@7zRy5JlO4cPMN2!yE{;Ag+map~qPY!dqk{FZa9W7`#2 zMfV~Sx+ae`j#pvD1L4fCf>A2jU_ZL(cm|!>;En7{gt-}SAK9vM12_JD!LjvcVxLh% zykqBAyrR4kOS)XZE`|bZA7hS=K@VCrE(FafR`kvH+|Fx!7T)dmow8H#$Lg`WNcG_n z`ug#kv?4W3&+)oO3)u&sk)zcoW@4Lvspxfz03CR6h1mIN z;dRGT@D~10zGO!wv))A=y_zbIBQF|KqQ*C{&H*Xn;4&X{w%g*JZVtHhi5O0v=Y?O? zNz>cYI2O;|Ld@R&fac)y)WApy3)vCLcM@w~@qy!b!t&{=@?7^LGX zL88d%`y%>vl_u_01Ps;c=(Dj5TF;Xt7Th^~-_~j9m;Md@z`fIyUetZ&VYe(+Dd{Hc zmg9JXdjvM@*^IxbBx0RfZM-n}K2z+KMQD><%pw&7Jc-iA!J9I0Y55wG8hQwINZVk3 z!xMBuKN3HSZ9*?hkMQ3z2~9>9L>Rr{Hst@yh;|LHq3;AFp+!^fpxOVu#i~PU%==?O zILIjgU4HozhY4?`2c{}v*{LO5{%bxpm)mQ-vsHwmN8ibxpi_AK<9@pGSvULjgJ-Thv^|Rg&t7$oo2|;M&XEG) z=kl;|$7%5KE`SV~YcQ!)iG47+7>wp9;HGoZBx~jzHf)0$+cQgw^_xwxvd3H58NZLS z;+Ac!;6xBBzU~>DzV#KG(0PwlUvrdowCf?`KP<>fsfvb!;TG&R`*rNfQ~fZy4N=sf zBG#A~XHIqHV!gSDTqdqC(xHg^?e@UuM<2Y7A#0zyCv#T0LZ2p4Gb9**MD}w#wCdrol8{qHX8CpL( zk>amOE+wMf{V3$aZQ``>EP0*$l$pDIF&?*;z@Ll*Py|numtTGY#+_G#ftezTb(th^ zQvJ)0`HKp4GK+BfS+Wv3U9jGAHhZeY8K1ZdxPH@Se7;+Y{TmR%THaW~ z`b}O8&tvMzp3&vRH~uNSExQLByp#EJRv+MY57ES8ry^`Jt749y_rOn1mVoy;F>?LB z9M-BHpzH=tKndCcL09sLzI!0s;@AROk5)paMHn17ue&J@wJ8VD#6NB-K88Nsd=nOj7?TXjR2(w>G53;IVa%|lqhOdSW zvGzZ|ps@dx!NB=4`QE&pI+1*qp|btaA>b$K+s8w;e3q1 z=5YU|!7*nn|Ih{>84n=h21fwbY=*U&JHU8r6Fl9^K){lZ#PUrFt3T6`&0OQgUTGRc z>k}q1yBze;jI0XQsBDrzJ5fateOn)T3xwc7$S2hG6=9mou~hg?WhMU3;A!yW*zdXH zV4Wr-n4_sAXt*=TD$PrRGS(7${EQ(jW)>XxKMAYCHnLAPeW7A!m-93|mog$DVQeWo z6N*BX!y(yuY{eR3R)4N5*y!6pmw6mTEcX&NVh5D$`T!H%nGhqyaWvXTK<=0#`z`T2 z-cQ97@tTbwhAqFCPQ0DAoQ5!yzaqEDVMVsi^cVCD5B z*l}%`*VuZD(upg?{^3gW{T_X~c)vQ%bvGmBH-xa9eHGE2KOfeMY(%Tl0&!KzPc&tJ zFM3vdniy^CB26BqXiput`<=WR9!hYrsy7G8>k=>8tFeo&JYP>YTlvx|y5g2qKk{g+ z$Nxx9RAG)h<-?et?_Tek8>`2_Tx}iA$oS@of1PoRK-7?$Z#X z)!*3BGD7k&z3C;{es3Q5N!Y{DjjO>&ZIqhM7or)VOqxGohHKW2lX;Gh&_$OD@>si{ z8W%r^$2^WPFHZ-u9o)QR?(r%Rs@_6^CG*K7X9p7Pc#G;W|Hrcq^`$NeKf!;e=izaJzoVTuMK{e<9)tyY&Ih`i!OBzs* z?LPP#nFvD1uad0}Yw-6e%}lcNLA>D58G6<0q)3vSo^p(hUv`my7T`GJLhYnXkT!}MXSQ<`OE^eSVe|ks<*C1Fd^l9MFmYyHBdKYua!;5o|mdIuTTs3Q5i9tw3{VAQ_2 zlgRMvw0^RP<+R*~bVU>&TiaZ~^XHD?t8Xvii`tkzKP?u&ccWnOd>Q>CgIuy5?7UOpuHjM0yqrb3?0kl6Y_FiTn;syaC?6cYM;cAGFQ+AD+^6^7tEE@i zD$q|pOrcd<xj6FuQLOp<5F~IrgRHs7(Z}V#s25)EkYM*7U1E+ZLJA z)(&Af_Y8L~Oj`!u52k=v+jed~w1VTnJw$5j^$1VkMHO6LfIL}3+s=PT`$S%)L)Ac^qQWmi?ojj=@^J(7SC|XxT zjXrNNnNFVnggy}Tlb$M-X(qm^nb$9^iZmj<(7Dq+*wLf`mreAd&u1`dPi^x2bD`1qvB^d66Y z2)~i1N5*vN82PVwnb-*&)RT#0RGYEW-#Yv_eii-}{1Kfn{>NOoev;T+kU^Ku_F(BL ziS)jv&9t4DF1~W%8Xh|H7)O`(@ujL}a zJNQL3zMrQAV#Q>W^EIO8>q$R!dqK4HIlonU3=EzX!yj6?6IBnvQ41BZUve&&$MB`5 zc@7d)k7nwFnG_1@?I6cg=QEoto3X0@U-J3%3EJRk27O2=jIPUVfoaS2K+iq~tG-Ld z@3y&M?_W=_jptO5Ny{NaveCruWERYEam7<#F2KJngPBTRXu}sx z`8ZIcAAi|+lz#d8BvxJ%K?^Ph!)ucwqB!pzOxvEvluZ#ZofhY)rCb&^cDohY?{7w; zD^hTkTNbr{P!D#T^1x?W-Y_!+_poVQ4Y=FpLAKK?q^IFQ+a6WM=Z8+=r~B>``J0pA z@W3&0OS>E6Ph0RHmw5_2pAC=ZkAds&#UOOu7TgUT$%dI7@So9K@D+beHY=B+!_(S0 zHuY%|+mVGg_$H90Et)WAqXcZR&Ls=QSJ9O((#TK0bI{c(Pt4l_!Q$ghvd6a-&u{L> zW3#eo%?Dqwkl#Yu(w>`{f)jAYGuk7=n{L~#fouPL$6W&k_!>-x*Z#iPJpZkww zCm7)PpF2E*=+u1O@Xvae-k5TKu#G+Gd_7s(O{A$7nq@ zL-7`zepx~M>jKE=eujB|A(5JzAj!-<^zm0!dyAvHF54to>ihi=72m74~ z0nfD(P+ebu7v&h^{D(d0QcXXSnk5O}{>%g^k*Oq0eK)Ocv5c0>Hl^=>c7!aK?Re%$ zGBPQ^uz%(=kp>8*+j@TH;SyyUP@A;|@B$#Spj3%^_adN?^YA7(D$G3j8~f z^!mTuC{;U={<3>I*4Md<3~ap_y$Oo`;rtBiXn)5CUA(ZZTo1OLXAVruW8$)Qko@VZ zB=*8zQQ*_sBf`%NC({0blH4Ik*){(0%2z zX~rQFuiX|-cl?XP1qo4fa_42zm*7K&*XuBa22R90`X)c9b_PkQV9~h+>P#i?FrvAB zjM8uctk>PazWyBz^Nc5wTb9jj0zK9}?xh5oFSt(ine1Kp2=rq+xWo?nIarz*p#J{}l1gu|S@!ffaC9LPRq z2K2k-c*b9j+l}9%jn5+BOkX_XyKp`%S-b&$zs-cL#pV#%@uR8lNF^*+YXHx#<&dye zlS!MT469x0;ZgN6{Hsrwb@DsO#*HSEH_KfBo!UuIWDu$UGs5%n6owNnW=QDA7Oo)xh{ruuUo+Cp4x%JH^`8C3lO_ByQ`6-3}UClAJN-cD(JVo229wj zLzfq2o6i_NNnW&d^0nvZlaX>Qke94MvirTsu_9%-vpN)J{x=0{S8iv{FT!lcsTxLH z?idrDAPUvTb`YZg@x*qRw>z1zvf8ZdA8S@E z$eUr@3`wZ3fKraW4}~rl!S{d^zK=GwD>bWr;cJ&OyE|a@#-mz$sKlqw|Rji&F`CbID zwS+v|zKi8Wy+r-dVko2bHV*o+0Q{cLXZ=DI*@9cEn4V@kd}dc9Z1;T1yLZokd@Gqr zMC#_T;d}JhOMSiYwMYnmnf42!&$+R0m-mq-w_v!D{fk`n%|Ylte^jHSz)IeI!N^F3 zTjUsBVh+~k!=Ec_Kq@Yi)Tqaih3_U4{bE^mlAjcbOWg)Or*&akg%Y~Q-FuOLT$6tQ zhFtp{pd&BDUOud4+1RITnb~b+xgqQu9t#$w6+h2`$wU6&ou-Mczh6eGK9OXBq7nZ_ z?0uq!Y*9DWg7EV+(i?IF4=WwKy-oe}Q)h~DXIoN& zCi-Jm99<}xZ`ob8+j5_bv*nGUMV2?^H7!4enpxVGSkXcpV`1RSMLP7JEWRh~izk0g zKozqRX|R1qUzcv9k8c!ZE3YfDwCWso-l%}Qtdjum8W)ItTSse8l+qs#w$XD1Q`xbO zTxRuMN$QdG3hZ-1nq;_`(T{5s(cO!Sh;Y^*4kgBa_~VXb!2sRUDF+A3|vVKN*^bt+7mo8#{04Ixh3}rElDkww$)FgLd)~(DtZ@PWF~S$$Q1o zI~@f}CFfr>i2S5iQB%=r-}O}aO-qJNa>On>)#*9YU*WfPXK`zP8WOqHiXD5^>1j#L zc=lH%;@!N5ZW$4$r(9VS zQs>exhM{zEh%~L;ZHPthOW?m#dr2uj*0T2DPRrprTP(9mt+DM!jE{sn!-~%b=suBs z^h=$sRH=<0?%a5W2qmi#-6KosB{#Vmxs^Lz=DC2_V1k zWpkY+K3*nXj}5v_Xva4_G^JQZCo~dRD7qYG6z78hR))R#+IY5jANAqXT_R?y#@zm; z!>qQf!t>L`=%t3)xJa>xtv_{?)gaBRiMk<-*}B4D$sJhsOBR3UI+(Wl;vRW5=$S~M3i$d4_(~$g>v4$ z9{;ZRh=rn;(hI{c!MqnbU_411Zk=wSkG=jzKYCn7`^kL6Z_`R?asL#0&kQxTb@5lO zpOD5%)nB991jFcK@LA**DQbCYY_{dovr?A-#UDrFqym}l`^Plzy+%`)MJ#8ePS754 zE5UGlI>gJqW4x;`kk8LVK;c9zzsr9OI;I)Vygob$6}!bjimMU%@Yj#=e9}j)=t}1` zuIS?>|9ZmgSX0j&%sb5J%AH12=3ioh#p0Mm{S0!$<_C3oM3@nAZYQ%Ngc(@)gSjlq zgSYO*&?veFWH}3_Vo3rV5&lB-^Zmi#<44k3T~5Lajgi#$sU%;an=J3>V**Cz@fI)D zW}cScqbk;mL%qNP_W2)V3>V~);i;dg185o(pcTWPmej;+-6RC3KRzU$#pQ|*8(QL0#`oFT&g^T;W~}xr5TCG>$Ze7uD`5w4?Q9zPWZcQKJ&@1H z`CntEI!N>4JC0g(Xqi~_oO-|%aXa6aab1+f2@U$hd~bYNGo7-1AIo(AjwkPRUonl& z9$=ZKiQJ4b;7@!Z{QHzabh%E(R(C@pH@A|ZIChq|dIZ_>#uGXW7h?DQ5loImC0d!P zgwCCpLHkc>6E*)%JTCJBS-#PMjdxYZ>f$|&*M*j*-6o6S5;xnm8BOG&`<0BWXc#P{ zvzVM)5@^t6A5YHbKTwRcp%__jb3KC$GMmeEKMzzz9lzt4r)@?2&;9Pq_R&N9J@%pG zAH9(rS<=PWy{;mCdu@^Vls`c7{LdrG!?^nZ>1%W;1Ti@?^$pE2{BOA=&YM zHS^<)AM7014C^>o*`E#5sHXDwyq$+`SiE_ZioQO&#$@>v@DKL(^KXB3C$Uq~sJ+Mh zVP2&ktOp+|QpO#sZg;{fy8_U*Ttjr1@tB)86R6g3QD$V`BB-<%z-9g`rXq7Xqj2LM zQ@PRyFF0QTa?$qi>3SlmZ<~yLEIhzt?;dF4cr7mdfuw%sC;rLSW4zV#qByX`H0t-I zC^Tie5w$I=midzWh3pPlMJ6gv!GDgY;Mej1@;Lh$lO%nZsH7>8=XU<^c_}v|KXR1R zx}{-(u?X)%N*WSN>Lb_xO9bt@Q*f|#J!*ND&OA+g&+J^;fxdDXsvGBm$okFO(5vOW zyxU9m@b0aVrw;iqBZs;+HOc)+X!4z0#lJ?^FvBx*DEYYxX!+CKV7*F-$p2MEDTboV z@|<-R(+Yfv{cTCua66alXu6PV-8D=|iYdx#+5@4ptodWQ23XTD1ussMgQ$QM6tY@` z`BONEen>k~6PIN9w`%ej_0N+?$nsZIN6l&GEHe%{DTOe`dBV&}lPSsPGm+X}1@vAt z1lC4oGbdNgB01}v0bLsY4JUr5c}qlXqR zmqa1yKF|?Z2D<|`lIw0c+`WD>^~bl1GKqRbedk!uLXBczb|wP^MM~hQw~3i%nMf`_ zJq`;cZ}E*fEJ!|Fxcy-qHSS&@_v#)3G= zF&AfMpR%X+Xb3^cw~bV2_b<|Tawc+MVa$Xoy_ZG5lSLAm%mSC2J~sDK-pVY zXqsUL`=4%tk17|*jb?YU_gXO8+FbzIx|g9Y?=#u3Lx4;Jc9F~3{m6@BCCSD|6Yr1D z$i$mMyh82>j=La5?{o>|X4zNC@3;oSUU<@Ya93y3=d};WpN&c6Wc3u*{`mv)rE4kF zv~qKBhpAB0l!htSXk5r~ue*a7Ud}Gt#`J?E0zxu9;MB*ZMEJpXX86KS zrXnv1mEID8bA`d+|7H=9UpJjvm9va@N6#cD5`U=H*mCo*uMrmVu89_gGc3?@ zgGxwiC_<;r^w7Y?kUC^>L5j~RPB!S^|mg#-mY)Qa|>O&#nV-X3FjxV`8u`5PR}%MnEco*Cdn?t^?KI6XDsd5t1$I2I)s6iIux1IbB^uoNVKWh|Mw{J8vy9 zyb?{GF0Zr*`PfcXK5xfOVh>SmPY$l0y#XF(EXAWz^C&3dc6EDWp<(?77_O`1oeqxU zz7a!BN~=96*?A(kt4@We8LwckDGWiS&qMB=upH%kxAB=h5BNK6CP=~G10bFGm<;5J zGJ+Y={4n1m%=kxla(C5hNc|fP`YQ#9M3|wVlB-neS#9X#_{3{K0!nk_;hua5GfO4{ z{j9!BowZ0;sKEPsez|&&qGL{hr%Z3;na~aRFv|KV*Z(uQwNfX zuXzWVFs?(%f%_o-voYQ>CP`vHKVkez70Ip&W!gp81FzALgfk%mN?$LTIjVTXLT*7n z`S#=|+>$B6U-OFy+o(ViFDbzE+6I!f?+O3Vv}~lSB*h#IzDC{95hWg{!zk_W@6=r+ z%%9~p9q-|NK!?K4qSxEAnC|l{eLs!xn88B$S6j9XF7~ct7}foa2UH7REWu1B&KyiqG5I z$s+QQJna9(*W!2tR}T2`(`BFWVm7XWSZxtD((N?$^}b!qZXfKpt~4dKSu` zmIC`dIV8kwc{utxAG-OKV5SoVF1kM`y${bA>sh}ncHc?j=eH^|<@;3ldFj7+X5vqv zwwl|8>3*kP-^)PjN|RCh$tBDc%#f&oN~E%&22M{mfT7`KQ08SrJ*Y@wWLvEHZ$$Gz zGiZRMzSe^M(AH814$%(WJp4*y?#0q3>!AMVfrk$rOzg=pw zN?b}cd(DT3roybo@uhHJtrNxHz)~K$7%dkQ!@^}WPi>BorWm>uQ4m+ zqof96JEuXlRUp6FuZ+Lp#KI=8bB)wyt7Pu{W(t{FrcGrVQ=1lDzF|?kbwi`x_yfxS zZY&eL{}5^IR)K8;E!f%g4C%a3gwA^2VeY4TQ|Cg*DXV8RQ|>0rJbk^Ak@)RIzMY-| zvKc+hGKY0&S-$}?{ooBFp}XK?%pcM@`8c^hGzJURIWP6IKp4GZ!Fzn(hscQK@zqYI z5utC!)as~M#!y9_dw=#Y>%DW(o2oQCeDfpoPh^5he0i4Co5lFvkw@2poWQU<0sf|^q42e<;8FT0Iac9L>6M*8 zEu!kMEA$kbQ0L1*`@ zhJ)ANesEZZnWR;Xj%{_NCe84s?baSZ`YVIsrC9|sFH~f;yYktNz<97z zyo62nzF?o!M$@kqwCJaO@n~fB6UNlPo#bv>1S_tc9PC(6oe#T29qbLpT1zsS z7rW=+nSF2JSdu zS9GoLM7#xVoFWR_=tgFthB$V+-2zGu?@3_%6$mm3Li@&#;-~X!@Y?k%v}I8StXOAD zN&^>=In~F&#!VK^4MpMM<2h(|`y*=P&j=Ct#z5M(UUsYeW!A*(9EsfG2BQ~_gG9Xp zRvWm?v?s4)=D7aDZ({Z7DlSi5z4jPxl*4?Ffoa73z#4Ml97EpPdO^k0W5{Y(2@cjN zK*7JA@Xv5*_)z2s>&{BS;V-DD)J^cU9NUQJJ)SC7i% zwaLFINpu<)lI!lrsr(>SNX!@|b5oK62`_g%Oy zFoVn=yFjAKG1j$7Bkt=oAj(IFHEw^4#lxo411^4OwAUM#zLX^yo~!84+9wuIzl4y6 zUOuUPs)|2HE6{VR)#%0kEmXuUWo+`ypZs%O1JjTVd$GfwnVnn-$A3S9S$}KM^1ci7 z>e61i-h!ojR5GyH9Sa<%>BD3?iPP-BO5Cn^7^OUDVaiJt$i1!%uyYjyy)Um|wZMt3 z>H$rU)>(1w$|<;@ut$GL(l?yF-MTxvm1Vz{ev57 z!eIJR6y}Sy!-}6*iEg|ueRtzBx}3W^{Zcgn-BU}!!Q&hmJ1>prZmT16AN7-du}0ip zyBhXh{>uol>iD9b82%g>g(cMt=$seBcy~$|iuf5y3JUHq!3#qKI?p@UwI|2e9p}5r zl#&ZzbZ92IC#?)J6RW^Ap`08j-9$iM1rki6K&x9DEi&~WXx>hA_HZUd#hr$@Egzx% z;TUDp_nb2O*oKBeRFN{9!dutTKxXjDC?VE`>^f8rb}>x?$N0wrs|$QV#+*ucIB7Ds z+e*a~jGo{R<;5<0a+QSZjG!03N_gf$Mf|fTMzE_dSx`2RC^%@TflUC;0OKBj|)Pz?7IyuWo&;4|1x{I{I2 zP`w?*e`a*P$tg)3G1a?}ol_nJ1{`IpGc4JfHR7nHIi~{u2MvGEs6i{Rf}DUQSMMyQ2$BH zMkaH{1$ToU3f_A&0t4HVyn36WCe!OF(r|mV_N=FKsVi!1NMPR<(h7aXx}r zt*wG0gF}L~tF!q>+m|7Ksqd)tS|V$?e-^7vi?gEq1t@R7Iu*gOwX^_7K`ds86$SUr&qGmK$p+~C2&lSe`B?Rl`Vi(%h>Q(;RFi?Q$KABDx7 zw|pw=1K&4tbdjD;xZNMf`Y6|#S)CQ<75D1%9hFWp1>WV#DBJLQ7Q5tYC0P|50qn3cn%EJ-cS({&0sjt>dyJLf z-Gfi8{Ob>_`@&Gdbk>m|))3i#n?${~0223jLUp7E z3IBacuq`!Ku%AGM1o6&z?jk>ndhpEyT$XXxU9xubLo%0lp8VQt4+_)1q1#&n@FPtwzcToc`u*-H zdgCq0e7dVg-uB%jF5kD3-OBP{>ZlLPWhB6Q+b?QT=?voG+Dn%1mV>*QJWzBMC6VjZ z$)sa-MCpftu$b2T0(cG~aHlJlER&BRE@8Q3Xxen@p5Y+B zH|HF0_a9BPZ%-{mJ+FX+`92@mN!JK3t-Atc|p00Hgt47reY7 z`$%zCUej-ldMbX0FvOH*lUn;EqW3@<9uf}E~AL;hA45v>KDB(^Ys$cL^#l_#TlV!J1i$7YL|;B{5t+y4r- z_6>orq73^W3bVQQf5F7SZoY7prg^MrD#5FF;}n}-My5Fxt|bWbrYNkoShU1}8mnB% z%Q4=?^L8C%0&nf&i<&1HUgl>j`!g0z zxrDcj2FHUl!E@nFNGNqAnN88;)7+`>J0_mwj3%jJYGlijjm9mO@1U%!h)zNQUIw%9IxyG#DukyZ{b-mwQ&HR?a^RAET_mi>KuAK zVZrPPcSfnb_sM|_&NJeWV4n0Zn>Vp|7e7J>5o7uD)RV9U=##Q9*qcs;q_AQb@f#wh zZ=O)IG`=%YmqdxUSqHJ{w;?VJTHQ=2 zUTXsBuZLh$1IO8BYbb+D;6h0MJlN0{9Q*O0pt!SIaBx%dz1&;j{!^bvg@aM$DWw9DM_ z+!ukIJAD!@YF0w~=?fDlscAT}n8nMSY2y6C6vPI#NUWd`{exsYiP6OKK9!Q*_wv~L zcM+};i^2nU8nESQXL>Enp*Me-LJMcErn}n~&{KCF#^?UJ|2@KB8)eT zK~`8DwtA3qjwyIBgSz-K1IS3;0&YVT*n4_Dl<`xBZg&Mps_ib!`6h}F-8>7$NAqF% z*9Azue?AF$?*r}lZRoA+dXOV+5101rg+@j`r1Hrcun(>V`A<6d-qKnSI=PPuum1<0 ztapc=F&JC6E1|Q8x?%O5O32wyU|YBq{G+%AQRc5eQ`Qo!I^hGp@~)Ac9W&TegBhsC z(vh^KL1O*Ij8xriAl}w8*f1>&8Z*gM=Gj5MtM3b(|6dMPZxFEA@uvnRZ%V)$5+=a4 z!&Y!^)@){p`7XHN=M3_?J(zkVI7s<)jzG(~_TA*zZpnuU5g*Cf5zM@wj*c763Cwof+VEk3g?<@5G>vM9Q69D zqp&<9T==&QdX{*jt5JrOlRF?>@DN8P@-t4W*P-RA-iU`+(bqS=!p97jP(|*8uxC;b zZ8u$oa(@*N-QY{icK`GE{Z0j>ed!?b%P&IjX5Ix?=4rvIIg`+@>JywEFvY%gwRn!v z1UT|}6C9}ULD^%CuypHwOp6lEOK!B8gi+u(3gZOr1#)_QlgT{NctQiB@6xF4&{T$Uc?D6JU^d!hE6bF zay?;9l?U7^Rs)Ta^!d5CrQ~GD6cu@df0py9B2`l7$&bt7(1?j4Z-x1p)x=SnNf}_~ z=3S!NbW|V^Ealln%z+k#0qFAAB826yA+!7*d^Y0`BdWH6%NAaSL6z1>b@x)_zTzAK z>wW@lbe4Kbx!^rfm#NEj<4}J%lZaagBFm4{(8u;Neys475wHrPuC>lUJ3k2{kJSJj zaC-_ZorKYcbEW90eI+C0B0_M-EYkK}9Uh<$g1*@*Xm4vRTyb6nX$`3&UP~HvD?fn} z(fZAakljia)!$P`Iq*;Sp+X$QvrrYr5KA{yO|Bk&rv3agux9H3iRiR zQTi^T@W1ckuCA!^(-FW`b2r@iTc}Ce zNuWtvwVKVU_anCij9_M_Ec-lTkrt$LZ7#NV{0Bs0N9cnGM1gAMX0Ex}7a%*-0yULQ;dIPvDAq~C zPs&`7sKIj2F$Oes?FWN$O89+#4fw+J3Xxy=8eG`61AI~V1DgMf1VXP2Xyb(&$q(&X zIGV)+X7h6S>1F}oz3vqZZrOx(NX#QW_hrfF)kak0 zN;xnz90~84oA5n_r}6lJA$4-S+(xp>%ZpA9$FG*dbvhpE-4i~+Asp2$o(u@C1Muo+T>tl^3B zBK8h9yt^JluQ1z8!I|G^wYp)G5Umc+fA0k^?ssxiniS#t2Pdf?Hn~vgT^KBU zb`s3)m$1(DFSY3_`r3GG`HH5gLZhaa;|5K^(M#d|R2jJZ(_?ThtOu4If>7S!8l0gv zlPmI71I}5eM``cUW@(>jcF*F~?CR!@hL+r1>s)h%dT2BPeiuc6+Nz~^EF^LyI|;MAQ9 zbG~b%pixcqs85Ex-DV8$dMWe$?4jTu|L*_vgfAoJ!sEQxaNrCs?xY)13_*TeB&Q{7 z6W(}nlueayHR$Qi%q*&B$jf4>(FuN-D%v^K)$dl(>}H3o*%55W_=On}mVhEO_A3jC%r;0#?&Zf`S(`o6J+ zI-2Xn@pAUWI_IQ-imWp6UK9rlWWC{;lrCn6GtEew-=QT7$GA^c*ihfK-2re*W|M~3 zu-WU+vl|Zt(AN#5p-O)bzQg;=x}6qidRS-K)O}(Lg>HyYP3y9$?BG-IrRr8F=OoB& z30eWJPxk_u_6D5wK#7gpPea9T^6=lJCKUDifNy!G5|hzfpzw7Z`i(21?LDIY`C z6+&-;=G0?qL!K&(+8azo8>+&w#}FWO6W~`g#QpOknhE>X4*i#(1@0cFaGh5NezCBY zyonyc7ta+?Z!=uM>U1uImQR=;4{Wh|xIa*wj>6p~jeyyDABeuU!$HzHbZ>4Lr^rPd zOCP<1#nrmN$fY#0NcINVRWJ|N7~ZJw-Cfpj%siFbVQ+(D&a^O|7Z!thkGZVQ(FQU! ze?Dt&cMjUmUI)V5Mc|+KlVtnMXnuEL2H7gh;s@Uf@EgA4HuYZ$IFovbVBBqxIPYg%{IWuv=SI3$Kb=uZ-d)?9OB*Lh}Z8O z#((d0g3+a`fLCoa#fkdLZF^BhG>z|my`W%Ij2yX}G5nrQN>qJ+pT7h|`@Y++yfC<1A574%}f2^e84N7Js5c!i(<#%t-_XRqA;EnnhsF6bobB4FNUh3t(48 zgFKI~xTLs>-dpj7S5<|0lK&nNbD#gnOFvy!B-$J--TsDJ<#T|dOyu#(eYTjJ^%}f* zVPWIHV3KqEekEtaLI)NlyRcp(DeT>_gUpI~0i4${GF%y*6gd4(7jh-EVUu?Nly#Sd zDHD-E^UHR8bVDL=**u@jS^tva<~4y&TUJtV%$b$ZvSkcGpIn)Q#m+Zj&8xHE zN~{mZaUKYI7mi!2o-qktV#qF9nUt;)0MavpIfcR>c{gGQfvUnRIK4yycB=c5)_*5S zi`*8dG;szTZrhF}jQ(c>HegBNgtt ziLtrO0j*K*nLvAUa(A;o_v6$XN^QY@s@dxzNj_eLMMQ0h@WX7FsIm!$)g%FjW|~pX z)nZRBJHc*}c41aMjio+GX5xJ|`DEaiK9}C>4kJp`sBo~1Q_1hK{r7np75>Z_d|Ju@ z_V3^0WBlBMn`Q{)`K^Svk1d5910EJzzkz$KbLIn+KdVrV(^`xz2 zf-(Oyg1PsE;aveSEc53dHc|gfb#Ik~8%8%>_3au4kh5~2Pa@e!ZWD1sSR>G zHQDeNJ24MXz;nwi?iy1gxEeW-5T&=oIZc#=^y-tq$+vtSAQ4J3EL9TZi4!&%(F+SFhZ$0(Xw~1OYw}#PEDdcDL>Zt#5rQ21V$Sy{J z{ny0++H!=nath!{?+NPtYhze8W)7ZqzQhmuq=}Kg8*JTW$tKZf31>)%J)fq}kQHUT zdcUH^S6>pD-X%hm>&GhE?_V2kPCiT`RF^`%1zOaEkQL}*v_Sp0eT?twX#8PLI_ag= z8ZD9^aD5GuxUn;q?`0G$t`3Ja__W^%uH|K^Ltl@c|%OLLM&+(cg+v)scuF!FA2D9qtezL~3 zkvuzl5_kWLA}_kcNnM5+Y4?m_Hkh3z;TiJm;izbKs<#NcW-TX4UQ5B8{cDhsN*OL( zbeyQ&y9t|Jk3fG{fdAZ<$0U3S``_duqPj*0%QZDomzT6r)BD5W?CL3cKu#Hbj5rd2j7eH(NFKNBqC$%UV0u|Kj8IS}65?lZ$b~gjtfU7COuI^#t+)ur zGj7nGU2){hmuUF(?lApW>_1rj<2msiy-6v$B{Ltk`~~ORkMkVHB!P&}H|mszHJnhs z2F)H8gLkOSL~{wN0zP=R- z=GGS`{X+q*TvfqcaHEo$zF7!BY!BtlIRuSDw&Me=CXVZ0hr8c?Y z$jISb+%V>dTV~5)v2%L(=a@XcEc^o7@bhD(>+Rt(q{0a4e&v>RB{N>48>vA5^OQ>9 zW6Cl!jj5f>aN`FxxO3+`CZQiA$lh)xLUWyQt7SEJwx^zn&CfW?$NvHWGbh2l;nO7OMLy0{-UpYqr-NGbkYS5XF&@KD zxSP5Q@gd8N@UHPoUiohYY(FUnUW5x^D|iz(Y1I+M#z=7EdOg#eT8?)q#4`KcPGHra zN5NX{LnQerKUeH}32VzwGF9;`SYCRR@d=-Wr3$R^o!>X9p*(kvon59)-QeMLN;)A1jv+ZjwP_PW4mEPKRw^0{sY5}x4o&;zeHCXO$; z5^Utq#$0+iLKP`nP>&kU;hF#KBiF-B$-s^Td}HVYR?}HWZSoey-ES5#3IoHO!57NZ z(XlEr?=B+J>pH2nU{zwGn?}j-@AAvL)?!^w6_Y^pFkHQjDe%ms0%XqOj%hvObL$U| zQQQPAWox;a-fQTsl7b-IcrDcxXo%nVPGR*+p}4F|81Lir<`mV{nA3SNlxzKWD(R#j z_t^^t&U25)l#II+EOvj#l~%k0E?&JzZ?;%RAKv_d66s23cDo2Mp^J5Z#IOp!(9yxq z6N`}AzA!L;KbdlLHsa+u1zY?$qRvI)(|YUh9~RLOvU_N2xiIh zJIIRS#QwHCvCNMLo4dB~Iz=urhsONq&2m=w-QvZ}sqk#5SuIPmmrirouuAUT=GWMF z)=TWRUyr!HvF4dM-Qx$-1@}t55IBao(VD2y&^Wo9vDCSS zO)oe#lV2EIyoB)5?vv2E_3T601uU=2ko9+5%j(SWW-Yn;tWrZb>%?dD>?{;S`rDI< zpUwdy?5f3fh+iU2H3E1Bx09;SyN(4-6v)*fOOh6~lpMV-MKq!}z!32@WLJPDIL`M# zB&6o!MM;h1hZn_8q-9X=?B9cbUyhPf_!~%F90#{^8;QcRH}G%FUpQxZ9=xAfL@U0G zCN|$iShRc}^>_Lf2h#y!Q^j*ALmM?j+Lb{fe=ob8zFYThx{#r%Bq4 zr8v!O2+)fwpb-*-fA4lPE5>PXmn#8(@;S=|^Y@Uab;Hog@CxY;Th03ZxxkioJz?J+ zT18v%BH*=4J>2|UF=DovCO&ToVH9VvDpOG;yX6JRc)1w1257O@6O2jIk~|F8&&8@i zEsTLcIw?+iM?RHhFzXz(;KKv5P&aiG_wPR$SO_M=KBe!-2Nke z)`!{8&I?)LC1pfq;6C-)y^I{Z^ABu1f0A~+eFB&_^T1ZwTf|h~gzRjoBokU)$h6xK z&8;pa4*Z_sw^&Iq11V$vkN}Q2Nwd7e^NC)O1KFG?#=fhHBJ=l!6OkptY-s2+awf5t zC-Nx-o?dYumh?4H&o|wsGL`v!Qe8zTy#bK{VFNZSLX_P6tHetF7~{saq!2o~ifl?V zX0?~+6Ym9AiK@569E;vs?N8&Dn79ItNude4v8z z?BTAB0X(zfPQJg~9R^625r1wqZ(aBwZpPIzrq}WoWksLoZvGy|jEcuJ&LoifUgv_# zzAR-N`7i{-`!1xY=t!d$)0_i;{eMDXR9a&E?&Wbm=Xgb2Cw zzlTkj>S_^u>nEQRv#F2znHff{O^>He2-NUgSLrccyV^Nn9o5{15fglPd?vTy<`v4T z_b8Qbbc++(@46-q^4)Er(g=T0+0?N#)(&co2L)6@{gas!&^P^Qh^WeRx%?bmObGw_K91NY$iX z=7=P!@xteXQ^$_v(=&Q!k^vPzeDci|`tb5Yjl2gBaL)=Kx@=`P<@{O;o{hT5U4Eb) z52R00;#=bzu6~HbCl=|0SL=V#T{n%$LaAiZwkU%hie19kkFpf97oy(VtfWWzQZ11u zE9qPM#q`@PGl-Y=98QFJO(Qk07F^9v;Ft&RppTcF!H>+Baof-Xs$tGqEU#6-O)-Ch z*Q7axiIdh*OVLgNCpd(SBI`P&0#``$RRO(Tf>k#A^tq|Nf))Chuy zCABsmQs#2I@@ufNQE{VFR|H;kNe-mrb6kNKDeSWFB(ZT>%h8jJp?KZd%(yz{mQ~VR zsiM!+`M;mITf8$E@k1PFRV>dGotZ&xS4*KJc!P|PcN2CH{Q;>|s8CW;p;E0DFqam;5w8XJDlhUFop9O;lGd|Jdz%-fB8j2Ue>CUKLq^_F z7;o;1r}}=pwfXY#G$Z*mmg99>5?Ji#pZ}Csaf^eCnOCi5U5u%W@TXSZ zqxe%y=9)~}edA&#@XSYM$H`Qz6ktRz^fx0W7O5n_-GlcVf3XRAJC_CrW^o(S|8mRa z<`E5_ZQNu#MRH;`Vw+26(%3A4JG#Y(0s*f$@)2`EMdu6r)Aq0Rqr44FT9GggSdq#U zVk=BZmJ)XSBR5mf$2w4X0X5&Dm=pac5vMwQrktnhdD>R|o^DPe9?L4A-Y74_hKc*g zmj**B{kSu;^`ijAWVkWUU#>_YN=TS1oh@p9WaL>)=(K*}}}rK7ij1-J<@BTDuEBSY!-+6Sh+Nc8b6xxdD8AkOb;`+$mx07VtfP8#u7a6<;atW#sEQ zwCLOoc%Ew%*|D|{J4;@{^G1FGzv=>Nex zWbN+}epf>m7imsVJ`M-4-k~Zyb=80{+WJT)P!9*hDsz7vb0Fh}llV!n8JTgg87|gb z1^bWJ!)-McjBw&(c;ho7{p>!n(J_gP9xo%iLIapXd%Jo50hj4}TWi7EUtdA&2XUgZ z>?-yVy@>mVALDM?1b8gWhYBy}QaO17#Ig20o?emy9p!T1V0$ynf3=NxrN{QDy$Ty>dl#Vu#IUvF8!KUy(5{bL^>A|S`OQ@IWF3`$; z2sSOzKuzI+u;hp(ssmubsdRyqBU5zcL0AC zUqNg*tCk!ChE3uok;>d!OtjdOkYdVOikeP zwgN0zrc2@3eblp0UigUui*vGl3D~uo)J0z=j$!tMne4)LNRJpFzRvyiLyyVAl3 zGBD4*AI@*of>c!)rLblem0GuzSrL~3C0gnM&$tF!Y35OTR*1tbow@YA?j)!ZF#WtpC;MS&((2mcT@)W;^=ubel-cXF2=vJ*+7hXQ^~u$8F2R!g2k@slMg40@h68z zfJ)(Kjr8P^@WDY?5xxV5s%hfpf}4zMd=B?^qy~EOO$8Mz*`hTv{}Gpy2Z)9BD^glv zOB6I!iN=!w=Gs$hz+IJ(l|t>w?MGZn?Xv^5e>b2G8vlZ6mXmN>T`1nD_Xo5C;wJ{YR)JdI*RYQrjW+?Y458#AK%0H_T*NW|f2|uRsJsnwm;!iMW zedGcEZfnNDLUOQs$$C(A@g!amQi1)`rW)E@<-wbLH9nGS9i0C{5GIIyr*5=sfJOr` zxX`%*7I%t*m*tOk3Fh8!829Qmg3$X53En|O9?{av>5K{pHQTG$b# zN%Gw512D==1FYvu=Fes=pj5G#ofUGH(DQ$SK8sA!pQ+CpKe1zH4Rk{vPftqv)GGL> z<{#szB?7O#S_Yq|-NYyp!q55r6rDW;-_1{gbKmH|3coU@VQwj9YN^G@DIOwOlr1c+ z7l*gsWkC%c4pb)gkQ+Hj=bYbR-T7~T8@o{i>xbRPN>To>^!QFxJD`u&SuM+0ySCIrjkKRzZyX~0N1FKldt(woxo{bc)6j0514V;n<>9D?|gIYB$ zgkvwwf=`-Hz>B(jiAP*AN#JK-PGsj0shyu8xs}PRtU80s_s3wly_U>huW%e>TLuFk zWHS}18^9-pK3?Ygk6`1feCnvoA#x(R1GMkoj_=t$#ybyXv7emVS)q5i>|TlYM8j_z zF((zo{E88JyKW7Pp4Fy8f6RdVas)BM7bS0e0&XWjvj}%)g-|8 zGZ*2a-RbQ9nF;J8Hkw#}^}~7IPS|ktKFEC?Mie#7+0Km9Y~{dJAbxe6y8dk*9CMlv zEw|Qzf}dwGJ0b;4J>Jj{Cem@TWGF6h4hQ>n_rbFZ2FM#`lKF4$XSltQg)fHBfEU^+ zP$H3!s<%s{u3h>M$BMoM`UR-Sh(N-Mj>;uKony zyIP2TV& z`-0cab{WVqF>^X?E`}|nJa?Bv6^ljab4nn3xgrRBcz7NZ#$2O!8Tx?2t!C8G?x(~$ z<{QA)rPzLjFLm2$kZe?vWlIx^iRtfhY<=hp9{oEDJDpmH%zXdColc9{`)hwvchcQK z^6Ut(Frb|s_!G(QlZ#>_zNCY1QCC6#-Pg2YX99cv)&}Bxa5F1VAHZvm` zBdMFK-*CpiOj7gt{H;UnRn+O;HJq}BtyDDs3}LlT9G;wN!AF9BQ{1P|7(cOUCOumo z)=fPJCl(emH3p8@<1fKee??GNfeWgfCx;ZP)sg6eYw%Kc8uXu-z(!kQNWof~iA-_B zvGyI*w_U1W(OQb6PD9Mesv_dF3{!6d3&Gd!Nc>r)h@G!%%aU(4Y_f<5}6v)%E&Acb?;aklugQM{<+{v>m#!Fu=MEf#vN8wZ@Wd48u+9X}tp&)ossy-cTM+b^ zsD|&WSKupl8CZ9p3&cWsIN^;F+{{}%v+8Q@6SGPzI3xmB_VJ$^PWZyJ!OzHwm_VWzuE`qdo+B^L z-67h$=CR+{5*xWx28>*e#(!pYfS%hJG~FmC$DVxVg(}x_K3p>94OkD*XM0wayYd;lmZ&cM20WTB4sY!b!9$NC7!6SYFvs`| z^Woqd`q8XSxbpB~ko8gpT(;go{oAt~7;Kb=vxRPe=Wf32DltAMIa7=^`}`G5)xM(& zk4M3tx|sk;B`{51dKB`u$LB^H@%z>j_*V)4xS~~v`<7)f$CrQM_M0d%Yb)I;O&NJs zd_9+q%^D{Xr=vJig-@aER204Bi4xgUybL#=)gtD#9w6_GCA3qrfJ51b;h1R=e0WV3 zYHS&&=fAY!CJH?RZ!4_unohniV)hqiL5?W7)~tw~1!lmNelxMS=r9xYx`wfAGenOk zG~lmOcc8@aU(_kFHMl8Hfa){rA%Z)@SV7K0HX&Y*nLc})srZ=;23PjN{Rt<)oP_`y zFRF$)!<``ZT1ib&>U;Lk0{J%A5Tq>Bw|6zguz5j1(u=+o) zfn#m%v}D2m>l&6DspHL0)fy#_M$*O&7cJk&-muzpo^UQZ+yV8PzW^eM@XM}rW=Ne7 zmmi+UVmJwye96GpBXc1)*_wMRaR)Hu&vY&VZ*tzh1hQFGV0Ci}6;%}nl<4IsrE~}u zoqC9Y2eVb7MrrRp;*LP-c&?1F75@kts+8oD)g|hgx)(_mR>O}<1((tdg7u*ly&p4Uy z>(E@yduZWzG26n%Ta5WUBFNGBs2`j)iAwv+m(Ajl0Ko|^_$V?qZEeHJ}6#pomF3Ykt!CTssh!S|IP8OL>(aLh4P7#vp#qxvy4?tcv`%Y|^u z8#9t)n@u+IzJPni63DWCI|}9sA>bFs98tXw=0%opVV)_ zUF}}LqSZCHHkgL{6ptd=$|~rRL*U^A74V;aA}+B|glE0Jg8j9%Fxh1ev`>hF&8r0P zVe1xH{bnXw^+^-RRUj{gA^NHKIPMq)m=)*HEj51dO=MmAZrXXNAJtD>fPKw;M$RA6d*kmCD9?IkE3OtXNaW%%)_!Q#;%w+!n%0a!0O%v@;A+k6}G7%QG7<=!9NuHPoRQS z*^H3AMkd(&t`eCljHiu*)Y*x-Z?W|$VX}Q?HWWN~8cXV$q8;%mI8ZN~&~}BeR`CEE zztM>OT3gOed1ZkEp`~!{+ZpKUw)&(?&`-PZX_iS%G_n+sS9Y zXnINDI&@Z5j#wrfVA*tcD17cHxv5ivH8!Z>wZrMSBzG3su9l9YckUtaPj?aT%`xmL z`*Jp|#kfh}q85wdTp7)AUsn8{SW|O*9vgCX6}!*-Bg6s$M5@9WYMikE%|hoWf9`Cq zdglOpqUbSuuQ`R^pXATc90i(s#4}inzZ;-xI}LB#&VXsb8u&m<4E&ZNhGwk)PR-5# z#Jlug8g}tBVW|QSw%U0U(y?wK$5v$%gBK5g^cHWpTv{Go32J3hbDg1Nh84WJ&Yhh3 zrrflBT(BuXjAlPOs-r?ai*!ByLGILP0KMr!*7omDxWg!lRcn`quawfsLsyX|?eBbF zq?uXM;ij|fg~yOJ{}2P?19RDNRsOytd7fz+bTvqo|0X&$=Wp}MEWA`~p zvAVuj+3`I$iIu}F5Via*S=;>&4%%75f-mNv^X~w7_~9cM{$0Qv?usH;*B&DiJ(tPu z<{JEMY76KYa=^{e%~<;6VR(JkE3k3)Qf_{nCxVln=s}DrGF#}?_-~O3bz{RmkoR>d z&gHYOFR5mbiwWj%=wTArI@(WuF7tvL7M%l+7n(xP*i8KMQYPuzagwOC)xtY}g;Dsi z*|7IXP*eTV1x?HT`87#V06h8r823e=1jeCqY-fBr)HTy0U!{0(_n<2AF!CZboj<{m zk|{dQToT1p9fGTV&2PN?pF*R@kO5`Xx`^{FYo#^oB-}I@5Xgq;uY~&^i?ZH@lI#M5 zm&7&624?mwgfA?|!9SrkFrgraG_3Xk->35Isnjwe9Xtxv&AiZ=>nC9)cY<;Altwy1 z!SLs{_b}4$4;953#zubtbQDU(pQbb6V8am@+8hPZ+hS1oAemu2`pK!&S8*C^K- zKz>@Xwjt7UZH-5hsHlbKDVH;K%vu9)giYO_1~-!x1Ydu;B3)VERrH zC|k2=hjAftuiuUQHwB_AUw5I5<^Fsg#(DT7wFTT1Kg!w4XRI$Sslge7DWL1{T9OeW zis`C(*rjhfbJag5as)Cs(&JBOrquZ4fERbxfN zza-H+lgOO;gSFoufzDn&(BZ#QAmQN(q3}5Pa{LOcTF#%X%ZI}ERueGzh$}RFT18y4 z5~1!kIg~TljC-4xlNV~&!P`yUV8rDYSog0FtXcXIm{Rg+t8P1*8fyk;E*HR?+uy*i z=S1_HPCH?!iw+0 z!@(6qUcZ1`$cl!i3oMYrR2aIv<0$I1d5UIjRwQO4;~-9=767yd|QU9d*K?Qe7Bf<=I2za-Reo5QVf-U|avmB`S-FW{2F zb@+XqJqmn`kw-=%T6n$+$?r|Z#o{9PK`B4qQ6ornqo+Ze^B7)MS!Am5f8-8u=b1W+ zZtWscCN`{E$WF5TpA=IknS(3+&+;8Jtz>Gv8yKyXM?wDLtij!rB;$J?3~s*)Gy?b; z{J8HV^v@j-GSe8R>V&epTsM=u`Sx&qX$l;0UkbAw0jsld2h`D#hw3*Hp<JvdO*=he>_#<1($nEAqBlqXngn$Q{^>EytjNO4!nJ=#3?Tvb=-@Z zKe?E@C-YuI`4;s{Xx5#&qyDLE{l(Snhg-|oo!|4>v&)0nt=BfQp8dZY0ZWY#h$uE37?Fo@mdj+l)p90j}0jBZXFXAD`f1pUPBbmaYd=}+r z9JzmrT9)4d?WBa!?i1Qb!=Mzn9dUzO^cRq2;ritIo%v8{&TZBHDan|+`xeNX48x&=CeZfLMSM6&0KU5~ zkIi?jBCoTZaKOlIit1bk!~c8-7C~v??g?dZFX%G}YMRfSo*9QuMie9Sa91?Yp$Y%R zS22HI{Gk8(`mev`^k9XUl#10?iZng^v zf6rhKe45Q(TSvgolZjA${}nLuWH!`aTneMt{)Cq6m&0xRtShgq3F^P$JNaH2;?v== zWTERF&Yj}Dq<_v**5ro}JG8l-lyWza=>Qe1V7?0UY_9}zZDX8<&35eGnh3V>lN8H~ z>CCzIpz+r_G%q1ZUayR@j5cG_O$%(JQtOcvVo+Jpa z8KvRg)=1!qZgURry#;5uy(IxpE!fr%N<_c)2T1>~96Ie)BmV^Lab80LtbbGs^MjOt z>}wuw5`TuRiTb2s-gy+%q=VS3ok;8B58kr6R<6A!7t8AFvwm~tlk=q|%-TgNwtLq8MLCxD z(Aj)7+gBQ25T(4pcC6qAwKjGLU*30<(XlnQEz=l8mpL(5;Q@XwcZY3@M5=AIeX;EgP7X|#yoq{tEajV?v+^gj zqO9G+Mn2bQ8jPFS^fNn@-?hYGM92sz$6Jug8)X^a3$FANzn!r7_%STJd@eX%aUKFY{`AN5pL$c^!%wXI3lrP7&RS zZpCIpqm-<61SuM*=r0${NJO zdH*&ewN)cXRy@LX$Bs?5q45E>&)e33ZUt-fvNH#Urt|%S>mBKL;S-H_IyttEBAT}6 zR;t^c%lnA;gzvK4^mx2JC+-5dT5z8Pj?{gHR!(@>(mz9OM_M5=|I$Xzs$}x{Xc6S6aRvD- zEyu>Z9YHm%xyZ-<2I>r6PpUr2L*p6EXpm!$Za)e{dZiWQxoZdCiy@1So>%~b(Rt)< zI*CqSi$R9(OL3~pedf?{4pE-0hJn)2P*D6T319jWPv+hsx$)6NBGMg6xg7vEt=y2% z84ela?f~QUf3e`0w(TLVoNd#VEy%**2JSn(0vgP^17!vg8e0+x3f9zP59v>+ZRuUK zedz-vluQ#bt9bY@CKjD&`v-466GBcIhk%OL1ytIz8nr9yBgdoKWb%AI5!$6f9^06q z-a&2jEaw9Z{XN@um&bA2U++?FW#?(~^KPk7XE>JJ)vm;b;d7y6dKvB)or&s3M_}mW z4zyTxy{)*;IO0>!(alrskP~|g-d-e!(gRPC{x$1K%Q9g;S7IL8{B=L}(+?rr`uL8- z;7k$`n~d=bbF}UFeOP-|ADOPUKnq)g(Dl^$(0x%J&i%rHa|IdDq51;-*L@O=`rSa^ zy$?V&$ita4u0tzYz;@XjH?+KgLUZV|2w$~E*U$feU7K{kM(H@#>xTflxH=Bi3>`+b zR!7l+6Th*}lLWX%*opks+Jg$;=b~p%qhZ;BEL34m!@A%w#23<&Q%Zcd=3oSS=fx(* z-|ZcVUmihP*D0g289UIe8Z*@HILz-0?t@pNIdI#h&4@l_gEG^$!i&apFy~Jv6j5@A zjqYkl%dZwLGM0tkyEI9So*(L4b{;Lw$%jsN<6uh2ODe}-c2_JiO_tT7a;c}Qnz34`r%LsU=F7WV&R=seu9{K7bH z6dBo4WR$F5iir0)=e$a3NQF`=Nu;Hewo1b$lo1&j4HD9@-sjv$k!Xr`Aw?*qsh!5} z`4isjdan1JbKl?Z=ac6u!g@}YY!S{uN7E-j)x|OV^H~n;is>=t=I_Qbl0ry(*9f9A z=oZ~D*#y*1l>s?XjCbAm!7FM_6|G-W$-B50RPMg;$i_=-Yh7d|Q!R6FHFH>a46k@S zs`~Emf2^#-vf9GqjKGC4;pEJwaDG}_;YaltJZ12ItWQ($bl(}m%GS9L2e{e%9?{MIMs|!5GU;2d;M6mN!D?8VWVBQ&UY}=9u9V2I zSJpS_-IgQZyZ;e37_yMetBIqJct70ZHÄ!Htf_Ju zwuu6?-bx1>xbDP%+@nd)mW9N1&O`iX)D;@yQA1(ePuRFcn~pFr!1o7^Nlc!ol37VZ z#Fl$Lkk;CDIPk6?9jxHZY4?iAWTR*BW9uk5Rj*kspL>~|KU2zr!adn|i+cXZ&d(^x z`VjiCRm>+@exV^HL%`ms4kqy%q3>uX__^kj{A0nSxcVgA-(v=?MISixz~y+K;F}y} zX+=|PZF!9x30k1kie@P#vMZ(g*x-tj?18pab@crh);za_O}O&}(1({;rh6h-A5P~2 z?*2i+?{A}Y#8uxqn8gg)_P##hEOs~gwJ&}-vKzu6>$ zvfSDJ15ZF!whUffQlr&>6Pc0XP1d2Dxh4o%znZ}Mhh%O!{TD$TPjqI6>J9qlv{pmxQ z?8q+g|LV-z>-E5}Vk_nX)*=Iq>EsmZC4S#k$Y&EB_I+9?92)!1{!Y3TjLPYw2crpz zJgCO{)r+9*KQmBky8^3K0+IW3G4rxq1L_ANXz0-rh%fX8Z@U1YV_%G8dgkJbSex8k z8G$OBirMe;mziB{JClx(fmI1oc%8Bc4LLKLXgmGGp*dZ6eoqP7Tb{&+e{e@iMe%I* z>S9RUS_}&lF5w!BZz3J}lFGgZ%ju{mzba2mHV`_kTg94p){13I9K??&MKITE=^#;_ z%u;@bf~LGPZ#%*f%3F2FxAm)txal-$mwrro=W3C@=UF7)YCI0z5+`}QMTyS(-T<1` z4%qVEJ(TEp7f3ib%5mMj;tMYe~(#Hu|PR;EFnD z;)`y-$YB3JePA5v^FNF4t}z#%zBN?r2oFfgRz+0xW)=M4zrwZOm{x6frAst_@n46; z5y#ED$g;OJ#P`K4vGz$Taph+nvG+I~^5xhi{7+M!9Ml+$w{aqg{OGapL;D{-G9?7k znl)gT=50DlZVWMHE791yiG-G@lAw!%@1-e|+>8#yd zf`U7`VRw)!F`v7Vy%fzNlNKDnQ@j2^dciER!uAKA_UAD6yxB=#-A#kBQktyjoIe{I zlmkQ7Ze(`rOThQ^X1pmd6MvSrBu`6*65rD&p!Mkn{$ph(bUl~=hWj(ozusVp29_q; zqhI2%R}I+uku`QYw-_Xs=YU^IIzLWPPCR(xOQJjU0THX4ifY!yqTo$h)X+DCyt;js z{E{C`2HfA{jS9BVJ1GyFg%}X);x%|uurXPryPT~3y_+1GxtP2=S&mzp=W>Q|A1S?4 z#-#T0%=P9@R`g~Vueo6xr#|_)-K&$osgcuIDBl%TmATUaKSFNM-{ytC=`und#gIw9 zHl!$RqmUWbLr(wHxSL95cH6H-*w_@O>;a4Z)BXIP9+_rvWD{} zO^-R&ttT2gH3qbl&w$HrJ=~v{gjR*L(*cekw8tJNY#>Pao*7!@RnAZ1pVOu49{8hw z8ftEMBT>uF8N5^9-I?81| z^20@5N3g$;rC+2X%+vnq!K17?^y_FGB%XN2nf$peFs7e#n)9T{A8(GBt@9;y?fc=7 zO*z#qp2YOjGr_&-9<`NiqI=Y3aoe7M$ZnaENbhnz{dF%3tChu~wX;s*K?_{aLIqW3 z8(2aUD#qhnM+KO?;XAT6+6U(5@1b?fM&z3|6p2&=kxZF}$p6_&QTR0t`}y<+H&3@7 zzxvo9QvS39tXCbNfl^v@f>b3A-TDAkh3Mj98w>vM_w{s?TQ2JFjO2fMit#x8rC5cx z!Y2Zk^9$QexH6&h;hnsX-m$nV8kj#3FMMH+Cdo(g9(&SwE7y0N-k3A!?uSDDz~lmn zZf~2&kY=MD9xAw{Q3W47@*dr5T#fw-ld1Bv@Tw-6SUPyAJTg#XG`@CP)d~3m6cam0 z;@05|OJ0nI`#Wz*Ufg%2MW==doJe!%*gL~++UY5h9YNMOP6W7a zY4qN08oh!}Q`6s9c%u-Y*N-RgTQB{xpQB>L{ZiS%Uy~m#WNnAg{y!OX-uGf~`z6I> z97Zu?kJo7Lg)a1|Mi-k+dX9?Xf<+O50V4UfDt`Ted~S0gk9MZ-BH{g3Y!%0Sv(W(an)m~&nCi@6NB-=WrYHttDh>%Uqh>c z-r>Y)PS{e)nBWtq;mbEW!Kr7#hY)wxl$45B{HsSM_h!Qfn;^8U2 zGqIP819+DW5uL5g7oF+PBj#o^$c6WoNOp$`IySA3cF3H>+{AZ~G;ahos6RtLH=h8f z&KTC)*v@)h8jA01u@Iw*KO{6`N!5m5D^a`OYDn{bgfj}JkjT<*qT_AAo{NrRuT23s zjkggT6X%eziUM^k@+Hw7L&Pg~eym8ZS>Kk>D__whaDBdAZKpLdiG15NW?oU;8&Nv&!i-u8DS#C`h?w|i>A zC^^*bH_G6DzHK3!7T&{t9%Wd!avb@1%aFWU;DO&(V_GOPojPA^;UhkdV4K2QaqnM8 zwgK&DtM8j*ljXH`V|Q?36OCRn$ShZRcN1|Tv7kK}M?(vXvs|q)DW>T5i5BY#e$vF0N5_jsoJ-$+yfFItA z;MOf!i~FaK#VR*`&q#g^I7Jl2A_?$5;mf2AegSAN1@ZY1HpuBmwF zTw#80@Bs&DewSp*kAicH9q=^H7d>rj!AU2z(4ul%{OivGl>2lQo?`eLAAT$NdL1x1 zGV3(S&kHARh54ko+>VTTYfcuaHsE)+gna7mTli6kIsUd`Iy&BOfoz9_kQKuG4Xy4Z z6M7~hKDNOweAB4PLBmD}EWd}iUGHV>`}>~Di}ojz=LTP?-%9ZcD38C0o$#behm!pkit!Ryb1 zA$V&JiOh>4OA~qW^ur|lNF^9d?|elE9aTtmU?W7AK0!}k6+p=SQ?w)KTxIF?B?6CW z5Rr5K4jDHT*tKjOb~5aJRr``v>}%R}?0-V&dON>H}@G4h*R1IND(MRJFosOj|uFr;+AeoK-gx>GJ>OA{?=n3E#tT7ZxT zFUMYI9dX$>b7C0!Pcl5T0Xf%O;L)zDY4ZwiF!&aUOq;5xD98>5^P0H+%QJROCWoaC zs^CHfjphtHC%{19Imnx>#WbboV8!L9X^^WwzeDS;q|p4UJ(@-Jx1J9+m0h79{z36Nhjh}=>r!FX2^(U(?ZS(DGAZTA(pOS5;fE%zf?aA70d zI_ttVZ+3>35yK$2eI#4AZ!_whx*a~Q&Os(Y?(BF`6tmVlPA3fQg1z;95cI4~xQ8Ew z4?Du(<<&66Rb0ej@0T%lSC#*$5QpWa$j~{F_t3}HO(<{bUOL#+g?_D5r*bOJC_C2@ zj#th>;dA9+YUwx}y{G|eFEb=IzL5fFSR2fH!s(PJ^H`de9+q!$g~_H4ctE&!8U8b( zZZ*kpCT@UkdY#Cm15&VFOBgzHr-b(v3*I7~A5==uov+!u2z0*JFb{468}?Ea&Ddc{ zm31{M);}^4II|3`ahIo5zZFMwLHM`f9-O^d48@UC@j1cKlh;0!`nJT1R(;+taW7w8 zF=D-e{ropS?3X|Pi)B=|v#-s|*`)c=xFA6RW52tQeohLC-6W!Dp_`S>6svPlD@>R? zj`h^W!$H}Jtk(;(%{x=!-q?NY)TTM?d!P$Ref%AY11zAHucN!1CgVZ=3qZ;w8p-hi zz<#$==KO1|&!Wcse> zIMWfwu}@j$3?6>rQgy8H`Z)@?IrcW*DJg^}OZy?ZXBj*FE|%?~AK|=@Bi(y#BJ%Qh zg7wNxXxNb@=%H*47#T{P64JLM7zUo}qSLg5uK-sT@|88Qqd@B9nS$$L@!nUi#3rfzkbtXy@-Tb1hl zg-+B`=On-1b_!MQUyD=Q1M%bhS}Loi49c>@shQ#``13s)4bx%!fiK9Z^e8^`$q4C>yF_P=wPmft9mwGNVmjhj8S>s&hRcNa zaB@%#bSA~)f}t1Sla(`dXmVvUWg_A4XIY$97fG_-_}SfP%7gXr;G+%zQd{V+UE#PRoNC zF+5#4NQ6`4&vN%Xjj@cuO;PY%W4^`jCLK2k<738K(dBcuBtc0>EC0m(k(5KM{fJfn z?D{W+Gts}ZtZ$_;dwDquTwz&j$`H)N=`Isg3+EKL{IVw=jjRD#JkTS*jte$Ij&*%!g+MaPAyQ|*#k|9|*5}R&d&x&uR^)Cru*(C!U1!4-?q5qsw9Q*)??Kbt5Re*pEi8PU75N9YQ8Q zG?>rV^Gt8@EtaI`!^S1_Q4e+<46doLtrLAnv#>jJ^{YoC1fIaG=LcEeVpn!<^Hh>@ zISNmie;-EpsguZMeQ0b(EiGw4g_Agp5DNdIj{aCo;Bl+JubeWzCevYL)h!58MSp_8Q? z+R40e2UQoiW8x_{=rQAT7#Y~Xw38I7?7h{CZiovn z6TJ183JZBy2xTUWetq|WE>jNWdyd?oX0Pk;zc~ubVf7vuwXKz&IAp@4-EU!?Wi#-b z=6d8~;>E5Xa3FV<+3}taJZM}&0yIx@q+EIfnUK^6&^J+xx z^%^oTNuGF*FcP2e@*%1-ui!0;OL6|FF#2Gc4iqKFLv|;I0IxuJY^g>&ZPSQiR486h zw+4#e*%6~eT{0mo5^cLXlqvdlQ~$rA@Gdxv+qYm0N^ce6z-mwN{>edNBd@jM^Ftf? zAN+K@zke6nU$GQpct~8qcPamIsaF+N89jG0a31q5c35%K=#*V8-F_i<-95W2X zLGdRb_=XYxQpW*lPn(KDF4xh8mapg~55e2K{sP%+uPA=I@t&wbJAq!7GXS$W z7F6*~D7@Tz4jQH$;7=&ILdW$)*!#tfge-9;Ta#O`)wCSRdjAtpWB&ynzdMZ1w;E!r zJX`Et^_VMe@aCsP2h+-x7fJ7($K><60c?>lyRvb75%xJ|!Tqr@#3_YERC#keH2%EB zD=RNSGlQOEwD%a%Q!OSMV~fbhMdswHe@RDj>lQiTLtb$t zaz`@`Czsg7L+R{9;79Da`Xl;$tOT#y-pSw39s*?*-Z0PX#1kMH%swRJkI_=+BUc7Dp8yKPQYil6g6rxtQAP2ItnsNE?zusi#L5y47%y^D7^ZbW^KPWBkS{UvY)qe|bvT{J3Gpdu4ZUYaXk?{5Qqy?Q%tU z`MgH-YF#g16lZ#mSBR{O23ru)O&~bbjq2 zNyZe8ZQt;PPYczg15-8d(T{QXp_>XZzN10*7YYo!9ve8F9mUuuMG4p8&UKs4r5zhK zG4qo?4C$_9Q;!&9?*ct^adHlLq6TvDcLqUbnEcofT>a$fk?Kb=N2_)7|HAJ|kZdd7 zj}KbrSLr3ch9~~h*zyg}*~>@AnD=@yQ@X0h8xcF~Y^2>&K z<%RI#vxHlw>~TJ+?LYhSNxvm)JflJCK?J=6S~$JAklinB^;XIw4gKbz@R zzvH+@Ke)DjIJtT?3uo8ug~)&zjGHfJJ{qBH!;U1_wq~@zzRTdtoLzy3xspCVjQ^Z? z58wVdv6nV3tk%1co3U4f{#&Ak+(Hl0iCada{zFan28#;WoOzpAmEAk`=x7Nw*q;nq zBVPmYci~7@9<7(ziH9k5z>cwM?DOZXP}%tyJyJFx?OT?R3xmv=A=EG<>jrjXM-IQ| z%P`gg)0nirE!*f40Ij0YY*XxKus8h<+)_)HnRB0+zJAK?%iLjmn&07zuZOXm!^+Gh z<~}lB^9*fxwvhW{AZD(KhuARhAQtht2-2ed(rx9+__=op)|bz~@!4Ncr6`&G9(9=Y zUgFsrr)E5*j$xBg`fRX78h7t}C#kL}vP)}ro=7WhppW}``{ujz#A~s?*nIyo@zbMi zr1@DZS$5zfafoq66ZWXnk9lzgu1|2V1_#k#jB%q_ve0L$fKkU`PWVeWikSwG2oUe7#8g@pM@9v>dlNexS#d zD)HpE^5oFclW@E74K*4PPn%rq(Kt7AZqFk%l&LdCywh=pxX^5bxNY`kdams>c2a(c zS6Ug;l%rm-VfrDOFhxP!lRQ|wWWgWSEme@kpgF+f+2C*` z0u4|1BS$u-;CTUFkQu!MJB~47R|A9TxM5AiYPf8pDK|KpIjw)v2ez?M!a0Zg6^4g zKypc56%Ts48ejgE$#Wh7M1S!teD6OKGPhryl*a^+zW*kn-n6y2W4JP&`?(h9h1{YV zbHnMl@BTEwDhH>15L|`9cX8MtX}CTSZr*%xFjQ z`0YYiIKv!?TA$(bn{T7}ZrAJ!EW`LY2C#!Z8`!B5~a z?SM!9v)DxGSuFpF7dszl%5uy!8R^zwalf+28Z1xr!IZpkH^W!kmB^RdlhLx}(kx-h zK6WplovpLX#3feKNMo@Vvx)Fzd-CVArQfcC%pkzy8%BZ5Dk=QvmL;s#vBF6n#_aIM zBTQB17;B6vWPaaH37MuY$Ud#Wo@71CK^a* zsG?{u!?&GboR$-l&gejaHA|p<*&~QPGKlow%SQXa4hj^e({P<@V6#F3R+|;^^pqHA zbP#f8eh2wy_E}(UoeK~B6xi=`%dtXPX&M3o)M@5Ncfx zW2envSEP#A&Q%Spm#MNpYo5dE(bpjG{WH*A)s8I{2k-~CFkHK5ISw*hNjq_cEgkVI?Twp0!`+lG2i8k?Y;^PDvNx4#JUB` z&wBv#r}=Xeo3DY{=MUe5%tefthMKKu)}^c&&Kn6r>Ie~ZZI?Hm~QbQUQZ^%T51?LI|%F4172@!P1q45x*P(~kEKbB~G;$ccdFVt!VbF1@x;MAa_)a~(T)GZf|H%+R>^DZRg zUK3+X{>Z>p>d(KN7{Pp>1kia8W66dNFRI(21+O%e>BG^50{_DZN&UNmSAFzhlPxbo z?xm#NKm%q7b4Vxgdf_*r18;+Pq!hdKim7vijP3Q&0HZb)rr;L zs^G#7Pk|e{69?6m@wZyGu(zXcvD`ya?6rRhHh#LF_5aRi8=k7;9SRL-{;PA`@t$V% z#WfbYJ-dkh)=XoD+0R*HmTEQ8H$j%DhBG@XVtF^`BK6WGaB-%<2Rma%-;WDNZNsd~CeMtgd*om!B$N*h?d{alHk48hn#}wsRm) zr{@!=7iWoeW;xf~3xo!4AqQ&H@Q8;YY&MK#(XZx_>R@Gjkp07Ty|c)XoC-YC-2gWq zkj1Y57Ql=twp@UaO?j9%6T6trBR4ultT#-^3xqGggZFE&Z{y0?^})G#z^WK8_^2m( z`Q#WbR@;m2&h%ogR<-QOmxrWB?E+R6))WU7zM+5%F=VsR9P%eJhh)q=imzRGO;)I! zAxb7S?(*H9bSw!T*HiJ|Mn5j>CjNYI#70P7T&ntZ2x>~FVIjA&nHEV`Ue3-NowE>{lt$+mp%+R~0eWfIK1VGg_?D z>n@I3qU!>HLcC}!LS@P0fG4tw6x-@?2p9Id8BH^lkOg=G{(Ie0~y{nUF@<+$}`Stc~89$Z&J69=UA4 zh1}3Ri|5ao%&#-?tx8e8ggyM#h=J(@-aCH=F8FnnN^N&1zcd2KH90$?a$V?oJ-vnM z+%xIKM+<04m>V&)$3!tg1%33~0lpz|IA^VhjLVTCt9767ugWmiSUK5V#x7=}cDp?d z_&G-;Yw1T#EK`_$>N2*d#gRoj*Fn$EMEvT36mIJ(fP$!-w7y7+e$@DlE%9?|>1ssM z8^++o(du}p(I&)?pDh>u%> zSdF?rxX#gGPp5VRzfXp3U$vAym!3@qUp6A$5fky;x5vo=&py(VA4|&1x|!#V1orHC z46-2~xZ}@G@kwdxIsd6M+0nBT<(`7O)+=}YCSk>nVu)!O#=t~1$Ibc_9_Wc?Y_p4N2 z+O`Yl50ZlAW;@|hKqU92W!_C1E5n)RC@SybC-d z55q;THPB|A!ZHGnvaioPh~CL2+^O<|e?fo4peH8ic%Gl+_(DCp$iH2bGVKXJt6&hD zqVfe^>?mYQo&zEpAaS*vtRDot~$)Y>vo`Tv@rts{s5?=GG6^Gpyb|d%_ ze`8+`TfZucrTCjNkFS<^`29L~uto#C4U(wT*j&&KpT-`^bfFZH8*3A6skYM3VRXX` z7BFlx-g|sCk^dD>6!y#@3FFR+3eQ}$A3Oj1L|tBn#khTg8Wmp%5i+vV9K7k4Z{O{5 zKHUMsF(&xgY85j7uo1JkHy-vrHDq}{)_Bi5DYox(C+vG5jqdC;wJ$TZ;8Lbo@gwJk z+SxZv=R~?uq#!GrhL4EEL$8cyL$oq~eo*;a)N&a;#^uI4`e$oVd!qf}DeRd9?t2zwZ z>upKZwxPuQLpJUVAA{QNmf#1k%IUU>E~K*F3LmR;B{kpn;f}3a@dm*=RrF}Ga6ijs zYOQS$P&o;g`A_AYuZ?0i+jXIE=yPbPGOb?qSGt;=9$B3``!xOX#02mAPz+mzdxn#- zDy}zsM*G7=yuJ`dkz@B@YG_MU;bv951zFL$!IiA#x+z<*{XWZ^_KP~svBooMTWRPg zZPb~bj$MUwWma||{LQe2sZ8kImnXrI={EGu!sBRjXa;Kf7lGCkd_f03XpzOaC-ILF zrcmJ!ipedGn^DmtaMc)na9)k>C=1~({U>l(ewUz-f=_gSVY=TVM>yB@LEIHgKxzdk zH6BAYd<;Ql{8`@Y<^TejIM$h@&N4=Sgb>GO`e3C(Rl(tj#PXYwFo)hil_d#qSSc2M z^!L*4fQ59>r#QT65T zWF)vyevJg4N+X3$J>1qEU-_mtV-UY37hHT|;bWmJl>ZgFR@#@K#nc$3&VPiSKAr>4 zs`7a0a0_0g_&U}do(=;859oM&7hKB9=oG0>)cECe2qy0#=#P-2{K%+FRycGX(qUQo zV?btY2Y<;|55agX%o@94eycrO-x~q_TUNsUZf(4Ifdfu>m;!$0fi@{8;~3 zaJsV*?2~6>wJCR?`J);0o@)SGOU|OJ^cE)FC3s}g9L91MvG}`!3D@!=>h2Drhc)NJ zjxlng=<9}+C!2eC{=65|1C z=>0H7oU_1=F8O_ie=f_>rQ>Yi!qzXeP$?CHjuV(MZ~&yH8N=~s=J<7g01Z|0rrOj77JV1K<~CVG{3m|yA~&7FMvn%5|l!Zk8+u=J!UM?=%8nrq6G}>E0SjWd$+Uc{Bx-EIa?S6HXyQ=Y;?%w0T zMnn$cc75Iq<5ZHtN55C#8NLQn`&`)cP#r9tUWx`+U%+c)vOxLP2hbs(AalraiTaPZ z@c!#4wo6fo8)op28Vcp`I+y_KR}|8B?32NxUo$!X#KYqiT|XeicNQ4*sbi6p*+RGdlW|!h9GF<29X_c$ zl8hA2!ZP3O_{YdOk}|VOI!?%s9K9Tdy$&cqmZkxlQT`XYEq3xl%adWd$0ByeX%Hk7 z-Na3gj^RrvjepyM;ONlzXxexwJg;>rxG4^1redCKO35a_Q{Rxe2`c#N>63Wc;d{vZ z#%1bYk<8k@u3}zM{W$P_9JV%CMCN8MA?EY5N%hGsSp8Nyl=TIm(uJ$&g@R(Chku*O z&YH)*ZkmZb8*=cK>n7ZXKp(uZ#D-)y$`h@w-FQM=HI7PFB6U)ZC|gs98viP!x7F5G z-uW>~A}|*16KBt5b>n5=)#79L<~+e~RZ&m$j_ae`o`X zr(-xTFt!hdr;0IrM3L7cB)c!YfX{NK=;l&ke%#tk7o3y<=dgP0FggT37(Y_z z5_E!3KowuoeHzW0`ks3yLg0QtgBcpiL34jO-{m~fgojx6LOya27P33t!|5`CsnULVB>vTI4j;P} zLHc$pW_5AIK|>SiUebdVOXa}qV+Xfx!cKT3H5ikEmEbI;&%$TygTfhYbpBo`7O!;x zRJkY?Ss+8!aZey1+Xk`??PzC|-~gWUobFY-09*f@Vh(rJMVl|*M8C>PxUtK_Al3K) zamiaxj{6j2x1(=yf5TRA%J|0FAKXoMrVR$tQpLyroX1;Fzr-z2h>`>p9L2j7itx?E zWFk^Y!<>3E=-qC@35(6~twVl-M{x=M7`7RPUd@4)Ij5mUV87*!3Ws^iW>@XZC=m_4 z`3SAMJc_(g5#v+(WuU9H0DQGgNV`EJHtH*s1m^#hnBG4)k^Z_-85%c;zyIkj>>pag z6;GRq)vu@Et`L6Z&3^bs%@n>%&OlVFlZdsiK7f&n zvQgy3OlGxX3O*Mm#VmC%$u+@y&(B1EdXKk8?R5LG%R@)4xwd++0h?}wTX+x4c zUmu)$rhx1DuVDLk6+4%wO}Y(#g6&Xi8vf1@+-7e_9nu+G{;u16!p$Sd#IX=t{JM(d z!jDOQt0tgXevxRWl^0hL%Tw1?1=t9=U>DtcT-2vSTt@s0v^d=fZE%#~qt)Uh+fvf$ zz7?4;Uo!#5Hl61kr`f}^OG*%^UMrd0WiN2(wD4cM36gfTPaIxTkE+W`B`3P&antYX z5c}8w-M3a|HN!>BujCxpaB3dTUMCBS?@3XUP&a|cb`Nj(y-@gX+=Q;o+RVK;93e8C zI7zZ>-&hfoOOZUS*vNNHEJ7y7-r8F)D=X8i|4{knB!lus2A*sDVdg+LdUEU&cj3r$ zUb*%QHSgL%%^n^>K8v67jo}~ox6iKO-4^bW74s&e5EDz1St+mq6{FEghYf_!kLCJ| zq=~!8hn&%_r4bfS=%DBskQ3nu+k%I~r&rpb&=Q6Br0yhYOJ|Z3Zc@l2EDYa1xSY0( zS%BwOO`)&fAA$PTulT6s43<+mg@$>I!b5Jyl5_dzh|63h8lV)(&we)?Y3%RAuEh=5 zRNDg1ypX|Fb_UdBpb#3Ke6+99-d=TS-wa-qUjh@yNGyrP0@wNhRG>6!mfJ#m z%fslhbEW+9c4usSJOU;ySjX0t`capvawvWj&7NrQ|A%*5ACiFAOiKO`*?@*9aiCT%58s31@#3vcC?G(Uc8g z^ziI3bo6ow_jX_I=cq(T_&`Zjdy#_u8Ly z38tg?T;5z^sc31|8j;CNfBRn(K2Hq45mjEPew#>wUJ!n4FS#Y;G9I4S5>NU4k_@%Z zMv3#x=#W2}`0_tvawtoOUah%>AEthQjy<7B%{LX!tZv2^>ORtvvL*;`JPnS=2C(uX zAG|?@;LcNP`QV+~k<#3g*mZ{#$p{iS+Lon4O0yD`i;q&BjHiR$S}Ew8sefS>f+uN zpM-vF8m(A0jE$_HfD}G;bAMOFku`f;afOiKvz;XiuZ|QVk(?*urkNt%u$oH?(I5{t z-Nj@0SgiMW7rXHN1Ei;=vmU(@qPMAQxGPO-xGt?ah!Oa7r=)zz49R?SRCq^qT{@3? zmWDISyi2UzX9t_OGzb0sd06sLV>Z)TsZMsWtN8Yu^|W)2Ax^p}>{Y`TO3ZJ0!=LvT zpjY0Q#S8mf@z-j6WU?<2DU3(*LLUn$*Fk9MX*@SF6z>xpG0OkF!rl-|x-Ocqm`S

7l+lse1oFjUjRD%~!Lc$L22V@n8p(X##Cl583kr#QZi016w)O?O9 ziJ6*&FU=GJlfO%GDcK z9(&}zFN-b3qM%_zFL1vYm4C+=)dCF^;$n3)X~u>#KmW+~;%Uj2Omw_}FGAEnE@ z`^7bEy2>HemU9Js>;NVt8naoNig5Y#DO#tH!EN7>Ejf961HVhyd0gl|EpW24A=6+S zOFA3L+TJ~uwCbl}8`qIUX~cOx_}&Sc8NL!-S~P_84$Xl*VgKO{`Fga%!UYEi`}*bW zUATsK#2V3wu*gjjhoq|zw+o!WdDmhyCTX)@@=7eD)QYdke8MgD@1hJ3!D4~GKI&B} zUhEM^JYXytbqQ2(X+~}M*``R#ba5E&umEbXy9Bol+oZ{JbEh~f$iB8 zrnFm^J@Y(*_oe4@OS1a$>Fr&}cD^<_%cSv={WrKYhe}kJE>ktXaU}icrHqHKz6sq` zr}*e@HK;i{9VW-7p-NXCTl(okah%}u4c*Kh-x|qsm0IDVG3D0%F@#xT*Q50nt?2OZ zGw|LsNEAC&;KM16V@rkea9)DYH@oVMb+#D-rF& zijY(LRp1<`K||hITtD;$?i}My!Y268N%fQAw5#w9-q9C(Mk2g1Oc^O~g4~(Q`EYN+ z-*oQ)8Qd94*0;TZuG%zs8LbON6PKey(kCF&-W}Vfh{zz7lXzoRFYjvAL=CjhAeSaR zvVG+obg$AG&)71h>O}r@{66^=+T$^nR;K4jY_}gky@ugPu6`OUSJ;45C5Ne z8bLzs^2o5RZDbWVbNN3WP?y9eSgXp1HE@nLPW}etM*5Q9N&oPY z8;i;M6XCd1uAZN_Q5t54y@A8SHlb=E$DnC61Rr)yz%OrRiDvIQh%(nNLjGqK<39PR ztaSbvh@9?@4FWIF=;KT9{Bw=CG-{oQtJc6iS5Bk+5ejV0)upg+n*$6kwZdl{YpL8a z4!-Pbz?&9W5!GGW$wJ3JIQ7XWtnZ+Lg9p~ZnuTN8yc5T%I)M4-s2#UR&qhMG2ig>vT!_RB&Bk-KT5$&*w~^xSR}k}H}CFZB1LWj{MHb6!kl zPj*6idn<8_?0Fn$wjPD||3uyL(bRpzDem&y|Jc;w)-0>%Agb0cqS-Iba>oA?cji$w zMgQM#(xkbN=2S{FX}D+aeM2QOmCR%aDUl&kNK{Ir1|`j=NNJFUd-mQpNl0cY5+Wg@ zQBo1l`TqN?^*n3+e$Q{M=dN}CxNEi6KKHEFIp==%`~7-Fih|k;pWu1ZB*9STW>mC< zW1xJy3RlN1W{eB$al*WBydNs|*eA0B9$K6X+%j3Lv{0Y7&vCiH@1F@8d9n(AO@MY4eqm81zNo-1(h1lKw<6; z5WCX{cT^QJ_`D)Yqdu_GVJf-xrjnT52_oI`BJjFIm{o0FN#)cIF?Nn~v>+k3q^k2W zL`yeCqvtzWRMR0Uj7|Gam!5t`w_jC)dgrxR74O?@w}dD4-JeIlr)mg!iS2!cH7>_cFRw)(Hd5#gU6kB~n;;2}ud;kmgR7n22eE z%CGB*R^55j=PwOoUN^$-us_UmO+~PzZykHVW)3-v76wWYLCb`m}8uEJccUx1yu`pM21Pim2i z=-pKj)Nt(-I3U~(+jd9Vy}FT1%+{|XHY?wgE%pI|y*D%jh4H$0kzXistnoloO|8-0 z`?*lr*oe5@=_LEUpCsqz{sO~?3h~j9MB+(zL-%t{Sa#$Z9CxxE53x2x>)1Qg)A$(q zY%}J&|NDqXVq;j%iwJzKxJFzPKa-_9q{)wgIL-x6z=~0Is55H|@}`;S6AD98R&U_3 z!SnEi(qgjmpf2lwKAop(t;fjkkD-Uc&QcBS68fS|lG%N7mbK1x8NtQMd+a=CLlTpb z4|MbTiCW!z^6!r_-S^Xs98a4EC(RFlL-H)BQZWP?hA|ANO{$dZvbJ9Ea+%%opO9Y2 z5Ymz?OmFw-;$0UL1QX@L?W#5`1ATihll(tM(0^kvS}@s?WYcl+kgZ$4c58+K)%R z%m>xO$JqV5|5O^3gbNIwa6Ex4kBN`$8Z!CdKk_Tl0Uj!mLXk=)$SbXxWa;S%a#J_c zuSy-*^tKc`^`kxSVriP-^kAo*vm1~7tENo3^f2)-wI?HPC&=q9JLx*NDP-5?F!JR4 zDthw30#Z1m72jds^Dq=)f6m#>EOp!gWPg`}e@%Bs`rpDz#}Epz%) z_6u5-tfxmgDnD|pC zn63R5xZzYh4)R}1r|8Y6&ProM29M#jS#3;M@H4^R=pVN8#~UzHC#dpr)q?Cq4;u(| z{Bx-G`b{)$$!pFvGe{Ln#DzgN*;wQc=U#2I$L6S(ILL;g#jfGVQI!WLwdx`&<#Jkf2-7u;`sX>{k1C>qJVg*IF~1pMN^!CwC* zY=eO~#AkG9V!{#hv2zZxx15O>v=s>icVY3eNRG3>F>%`EsgL+OQYxZ^F6qsK%RO5` zZN+S|$Neu<=m98rFCj`G|ssYK3AeB#v9f| zPpk@Bb-xN`d(T5cb3R^pW-?juZUnhR=Mb`eE$QcGV3ux|(U`$<)Yq7e^iyge(SAUH z?{o>c4bC1nYg}x*T=$H)lD4jqz_Bo;l^ELsa9dYoJ z#U3cT&H^bM?16jI{E&vYB-kQv09V-k!Sd!+@QrH$$v$wUDoM@&72Z|FVrzdu*KAw3 zq`nh;-qaxoxjPIKd&=k;8B8;->!4E{Q>6V-C%iK{0?TLXA+Mjq@UY!BW~crK=5To? zd^NEU+uCU2d%`t5&8w%;MYne-&{LWJv8A7pD}Mo|&vha16FS+ezbs+Ttv#qHYy{4` zW(han+kqd&Po!ITZZPJKCU`D{WKsP@?Elud;%LKV!Kx{iz@V7VDvlZo>%3j)ftUqC z^+jg%Z=it8-tB^xDOob^(5Gt3>8tF_JZpZWk~RPPepP=!rwW8BI0e++40XOAD;WE(>vO|3a z?ae$*=O?|zio1nm+8-ObM)4^5v3(2t_e@B14+zi|m1!uf@DdHWa0F}%(!d7IH(^Hl zJ7_m74U)c1hpP<2q2aHS(DkVZy;5bv|2W)_z9kK!&ziT`2&3_|)Tovd_1!pb54@6C9}xyy2OsawvVE13v$8 z6)yRdPKK^1Qd_G;th>mKmtSlysG5DJYPo}yzT*#V1b=#N!EZ0bky z$tReCcUPFX`X1nwu##;u*B5?V?IU!);3C8)vhh4)OS-v`5G}V#Qu$bcinm0;139tO zVMK{e@C&1LBOBnYMjP7WUX0TdFW``Sg=FF*X)@urCynVlK@DES(8MFMtahO{Xj8vM z+=FhxAAM!G^vZp*b;fMm>ol3JKY0^3CHCsGTuv!!r zESk$6T-y#uusghS&7CB6Z4pST&Stlz`;g>ONz!h*l-?d!4=2BR0adtXz=}*edG3maA@NJAStFnRhH}}8;i3*Y5i;9FwLC&9XJCs zfeM-xDT|_gBXQPpj(2m~43|cJ0}&s?XvwoO`fFh-{RShcTXqXM5MfDYp0}n8Jmo;3 z?{hvAd|ZKJ zxxZ$%?bN_EE!#+BcnsVat`DiHFTLmUh8D^`rdx!Mpz(!`5G|I)SMTg)+^2>L?uS2O z{C7wZIhUz~?ua2FpFAMpR1# zZ#oz+2}e&mI}y%1n(=N7~<>LOE?U61F^ynrl0fitt$QIDRWyv3XIE_@x6$z3-xh%#T4&g0)Fs^Rj)z+Va0--w6C z#+%{xIF1J}R|8dr>XU?GTlgX3A-uzNB|q5Q1+#|yNVvp7e0APga(rqak@MY1;=8xN z*sWpg3@bTYy1;_W#V<)CE(fVMi$K!HPh4(2k=7=N)00dd?p>3Gjgy@1RE8qxhG$Et zg}5}GF)o9gFOtI*_U^by7(!!|gXx+z_i1ZX7mj`yN+f~|@S2EN&Motd4E%hGdwTZ5 zsI^;R{?2|-L_+W@UOsR3n+D!gO1c?)d;X%g=$%oUBH>o}Xjv zy~iOvw^1H%>ur?msg3&7Mxanlg4FxDgF>74yai6lv@Yg4srhl8TF6X?Q#p>ieRUbT zW=cMgUN;GD8=H&84qW7!;O$UhhaTMfmEamh797@=rYA+YnMltY{O@`wm}%Hej9nw3 zHaJb2z|JMy&2P+wgff|M$fu5-tRg_c z%@5Hc#k(k7_B+lRmq2^BbkViV=dekfJ9&j=!3N_tcu#f{Y+lsJya|n``;xhS?DRl- zPL%7A4J(pGhCcKKi9x(Z15}y%9sV6rLDG^jP+(;QBbM}VtY=YjJIfSw7Z8&1&Xah+ z({M~}Opwv4Kz%mfu#=mU3B|z(yJhGPD-8B0w?`3J`SCjVv@n-UyK{q#O*;TBj$I|q z&w`1X_hlIJlJlMj@<_{umtf2FG(1&a8%@uOhN&N0N#(3hSg>gbn3lPt;|lZ94!=75 zEK`E4*AIqAt~g_z-kSqOZ!aWOVr=e6;ujyj?bj`U-xKywiMiTI(=6s5yfk zym^rP%D4v^x4%%n#uSDIIm7fK1^SkKKJ%|O%-O{gV4+_*?zuOe*gGlW)!TGI*|q^&)8&%P zQtdQgf2fT8D7pmh+{iH$L+i-J9py|4$IK1hH3_$Dn1iv?IkGY24mNu+O8)tYQ)Ayr z^oUn7J$%cPCeB<=w`SYYqx3P>2(iVbi{7)9qn2!^zdFcToI-{gWQd0_0&f1!^#{#v z0@*-ztHr^NX$-ly{Ys!Eit_G{!eL7R> zsWm?Gd57IKi&J)2-<}eT{BlP%lMW({p<~EaJ_n~op8`+4)&dtDf1b33z)mz`1HO(6 zknGAbbTDNza&2=&XUtZjh-O=~2DajCtGT#jaS}MVkB2g<3t(NdIBLj1NMQkklqxjP zGE)x_?ik9fn(`eSWUY}PT@iZ7%aA-&01lL=z@w{zX-D4?!&XV6;=%5G%&%TYVE_8zBf4rfOQazcSIo=3%^PzAc<>cnW;35g?n5&1luhdNfgK6d*Ac zJC@G?iyI?R>*Q*9LN5^>iwOYxr7@uT+2l*-XJ}k^2a3ndgki3IXyNV@^!TVMsu>$4 z>!Y6I#Niq;OY%IB;W&g3#(LqU0(q1*b`f5@Bu0(T=i(RRli;`;LbBMiU$DPJ(e91$ zTXuwFd{`XRM0szPqQ9;CfkQ|u&R=|l%oe3y7riUC`HK46L2F+%CJC#Hl?XQ$56y`6g@F9TEf0 zl48-lS^2Q~$uyXv5)FI5XQPSxl#p^uBo=RKXN{do+51ZK$bwOAlA#h0a-=lz8hpP> zM`j7Rv``8ielH8J9LXk87lNVGEDK_HXDN&ZJ$G2UlORU~to6=y5*|snlf<|NTbjY}P6m z{wyEFoa2lGs$BP9_AyzUuZHeD&0+?`hsm3iCph!Q1*Cd(E;^;Bi1K~XiS+J$Ff@KW z>O#Y`bE=%Mc#pcUp=m2xAD>Rl(#}EB@EHF)ZpQ3={0x2;8HOK&rlPtp)!4;C5yd;l zqO6cR=oDX-4E%70azOVNj<2XKXK5LsIY zX|c5#dU1ahY6=U2cRwr9xV4Zft&RfE+PX+%!A)S)YXq6w_h9X>RWP}#oUZ<>1tnL! zhB1yTymZL`Esr~i*k*gC!MvKs9OSyGFG~g6Jm$g1N(Hi&^Gmh17lHKTBzWw@9JKb$ z9OSEV0Np64W8S1P@T*kgha1Qt&`cYI98r+mdQ9l}?-CJ$iesmtu*?kI3u-XJ} zecnLgULq7UXD(_m@k1|%I_M^ug|tvRjx7K92$m?1N0+2e!w)D9CX#+8_*^oq*&9wp zmu1njW|PRF`bwUI*EFPJR)r`3N+K^C%i#gXII`9_jx_fNVgAS>C_PCUr94@NL_cT~ z0pA@7G8Fk4p84pUsw6B1%{X$oHvSPA59=&mqxOy~Xx+k-fq`1kh+a?kK0s9ErifssQ3*#&d( z43SQ9<3a(>2g+#hnhkUuutur_gV3pA9U982WjEQ+0O_(uRzW$rm0?fv1%ocec7839 z0;yWAGs~fMFKjoU<4hE2`9*1J%JnbH7N(M|Ws*=i^aYR|aDWeIn^518BH9w6k4&BJ z!pA=kfz@4dbZA7Gw#JoGFV2r)$Sy{2`p2V=-qo~^4X36tzSLA?E^ti9XZKn^vy;Af z2L?{_qSL*#q0@yBwpTcT6rD=N`Hp4y<+NMGwqXJZOS_H(bZnW#=V8oWuVrvYr<(AK zR~@xo5=VwMOaLPB(`e%Ir{Ia~`bzsYC$>1C6D~5RutdWe^t>LSV>e&Z*k60shYE3$KePh=AndSN7`W5iT&`9)qSvXu$X*N^n%5~ ztx(f-BEbuDh((kUT2~bcm$hG~8S~<>tEBR9c{Km$>zjOD24yeE;_OI|o=X*T#I0)v* z9wQ&4*RZ|k#NhgvWGH)KJXEjZvIP12soTbd^xE(3G++B2>8P>-f!10?vz8;v|K7{A zjzq#qI-lX=nh3Zj{TW+zEt@ybHxls2S<>~PJ3v1L~Udx z>FPX=Q`@+c_ zhzi7;eXZ!6d!x9gdy-(OxByt)31%LaTQC=1C^BzIHgn?KI4UkT0PnWHh1Sat0uOG_ zsN3r=zTj<)IIG=p*MtIga=KkRZRqc%GMVr*k$jwaiG)nbWZhn5(ye#eiOdUEaK6eD%gU~%xyqDTf1(}O#_XiRvVZtq zp))+YqK;Yma53fb4^*MS3^1%ZHglB0PQJ0&ny-T%|7aug1$qRBg%i#DCG?@zAeXn= z!R|{uOGZyVCF>XLmbKy69~|+k2rYJl{T{)avJ$Xbh5NtS8Ub=wIsjK=O|opbfcra5ti519&}{Uk zmfQ51eHm`(FpSri{!36Dm?kK?dK~=LeO7hjpco$88pj(q7G@KeBwcmNF@tD; zVkXqvnR+F6vG>gWk-0-($(u2GE~A~uzIl8FY;0-+o9glgSX z;o2YNyq*V^F#o71_~Lql4LutNwS8sKzS&Pn*2g~hJ!A+^=WXDne+GWFTZ6-t zJ;~v#Ze&YcHX)a`5%-%5k<*h&9Orx|RZClgtd8qJ#V##mGjRp|y4!~CSCXJ5)!I&&77Gowo$`%}ar$vrDeNJAp)97yzK7DhFg%=!yp~V~)c2<h_5-Lp4}q~1JGi&&D*pDt1fmu%B>K$@?X$XtzIlEoavu%o)%o+O zw$@kNxy1#9oOeUsPlq7id^h=>G!A?(>-7@sKO<>4vup+NpZ%El zMw`;U0ckE{cO4ww%yqb)Si(2jZ{exs1+Y(U0&Jbx2X5Tvx`{_Kftk^4_{Yo*>7T6y z6U%?ItMOdiTR)q+%8HXat^w@!OgAQGv#VYHv+ibf`UIuNHtEP(Bm zb}A?9MN4*+*hx;5q9IGfY2}~g#PGftJ*2#mF880u&6L%M(}F;JFh39$1VVC>(<7Qu^wiuc@<8f3G1+cJyJJ6*P~ShKBKR-$e!rW{{keq1 zPw*x#@^|4tYd<_Tl0*BhI@1}N&16Wm30lX!rmvF%>9a^F;fwu$cwfBEvHwQb@>bTX z;MSgY0gX4nkAE8B?W>DeUPCZ1Wvwe+<-ZzgEsi4#SL?!*Ypcn=utI#QFo)h;VMI?y z{w90oi`bP4qb~9Gt2<2 zU0=>yaV3pCb~n%Nz?n#a!IyTsi}SwWcXBOokM9GRGj#Ks|dYtF$-t0)LavYE5c<9l*=6&sd$J zEI8FPlx+8SNmkyRhnE(9XQFMkZ{NxQw=jRZ8S4n8# zB?=|C=2YE(b(U8h@s*@MXeJJ8J@KdFT~OnNG_%65ioI-`i>vp#(TxYKX~A@s4EtEZ z4W93a)h1VRlDdL(yiDlJ#$veeZ^foU1=}izV4J>#!HTxmV^z=hcaWH;0aT-TB}Em6 zLX~Tx!rX>xsu5Jfq_E{|=#h425|<@v6ywv({x{^ox+f&O*$%9a<#Oj80($A+J=z&} zlMXcW!|0{S;C_xLePEs?ID$&R^*$cibT$@GpQ4F%A9Sy99W$eGASCF5(DbdPJUr}z|`{E8(} zb2P|%`*7Qf^Q1}e9!w)mF4NS<-|1zjA=Gt+!qpkW^zz_l!3C*GpfF)Bh;Zm+mszS3 zJ4qR;`OB8RQ_3TMi?hj^!8mFscAu)|tf8+$tZ0X!9y@UfrYXlMalD=b3hIs!wX#z1 z;9w!vUX)*nEyTseO7TuQjIdZN($^;}{9Dl1+)f70Celv6LsML==QH;*1qy4^RsiXpF|Bi zX~Q3I$!HevZ#JVe2vQZtWmK@KjNDW0z_IFic;RRi)wp||9t)76tBzEXl-trINHdfK z{CG)M?Yc|zZ#@LCh7lDvw@el?`@;v)K}-X6Dl?}7{B)`4f+ zY(TY73>aQn!fdLV1w4>D-n`9>`6pZkA79hPYiCnD`D+61nl%-Ao(_a7X63x&lRJn~`NbL~}N@!Q5hCijNBkQPM&l!5ZM2#+ZUkVcR z`k8fWpW|(h4id}w9i(4cgeWeb$Slm>3PcaffO`R|yqcBvc4xmD@s}^tbFZvcbo`0L0AHJy8TDaQGDf10B6 ztL7u67r|(2(N-e)NEChfIiBwx;{=W+aXi-3uSxWp)%^Umhxm6jviZuI7DzSF6@9}e z(6MWa$wmJHv~EE&Shg|>?Fi!7hk0MnrrFsj|7I()P#Z*H?XOtr3~s)3vXuN=d=&~? z>Y--V2V6HT9L=2E1|PH;AnTpR=x^3~bSu(?-p1YUtG)RAIfy^^2@5S3Si)ydTwzYC4|I39 z54Rnv055V~@P}kk`rEY$Z*IDYVl@AuRRcdz<*E*(q&bA{#x)}UVnzO{Lp7*N%mL|U zxU=yKJ6ZRT957%g%P;F)&UX&g;@|t(!b>0*V6C(JA0zgSQ8WU|sn9e$3q;r!!V=Sf-=J^XM1J(x5D^?HS)jUOVYmY4*+H8qvU z*}Z^PY7)@>+bK9LlU0TW`cS?iXy*+L(Jb>f1&<^ z>ulZTPz1{&k1_UqesAoB29E%5``KvbZP!oup?n5htZj}Sd7gzY?;VF@`vk1RHyJ8|IN$H5 zI9PsfJ@bxZ1X=Uc1?i&)q3SU$X71r2m^y9>l$(mM5dy7c=3} zN?Y_aeH(fpB}4NUAA_4MU%<&9<)_1({$MZwb)_;0nW`65;VIM4ejthjdq z{*mEv-lbJ&?>QYbP~8u0wHKkFI$I>UApwLhR)+s_&B;b?9`WkN2Jpmx5D)m2fyI9f zv5NRYay!I?_9Qsagn5Hxqmlv`@+yFB$DGmaWQ=;sd!S;C2_kg}nrhvEb}u24atPxU z72K~BI%?TC&GMs<`cp`V{}5YJuo=x|9MP_Z8OZsJ4BRlG98|y9$M`}`Xe_4)Qb#8O zYjF&us;2=9#Q;H$LKR4q5kq}78fY-z0O|N-KsF;9#-EmlO-pB^ONXaIk0n|-&nKSO zC_4_=$@_rkV;O=g4S&Gcc_VoJaW}X(BLYa586tNFL$taq8(zL}9rn~Ja5>d(JXJOo zZf@^~l1HW>u^c6$dc%XX$X!7XR$hPyRvm_GRX)LU@e<(W%=W6QcMgCjN6a~1FJbHF z|Ax-5xtXWrNigf&Ww^t<3(m{Ci4+b^Mx9kgSX*6(O^DaTDQC;Uj=^)x$D$Ez`&@5rT z=Fn%bbz&6YU$KPz8Lmv$>M&3yZU-yA9)>-?w?Pk|E*Nc`gKZZTGp+S2NtB;GeC_X! z*H;YUS=%`mUFdWwzpoaGe@=$A&vJMR!A%_U)sQvXz}TJpUdJknm_V7A8)3bD23*Tt zg0B{z0aGeofM4&GNq~8so!lZ{8!4YutrN{-p>xw**wb7_V5bPoQLd*e z6;IRp8B3sZ(Ol>is+0)vH z`(jaGQ*ax;Ho@TE*OSSMw;J@BP z_-KbTnx&nOTAUxCuU%tk&nI&<8nA#gv3j)peLeeg`zo*^;x=nuKAvoOf1L402kh#* zXM)*ZZbE%iE2@>#Lej43BQFU>_;%T9IBw=O^y$(as5LX1@iU>^esJK=n&A5^Bcu2|X7M}sxFPM_K+r7{ zL-f|=lF=p6P@yWB)Lg8A_Tvvj`&9s_S8XC9i~gdgp7Er2;Se#FoQkpo!iY*)I!TD0 z4)e5!xjXzIach5w9*YzbC8;VhQIF-{IJ|^!x#m2Ia=#9JbLYUq_X*(O7A3xZCt+=P zOMuI-mngRJB!78sD1VomEKM}5rIXH2r^>4o*@wah%zGOQT8S=l`JRhBoV$?BaVN5{ zuAI)h`-@1vn83F$zKISvwczXtN6}0p4gLf30i+%M7m8n-3b$~M(&b6oc)42(_L<7L zDLo#*ORA4xGp`oj7k`ew=7^*D`!1k0aX;+Pv>*`Wkp+i=BHViS9GMV!g(>-?f5lvwvb7qq@nVP#>mRwd`Acj*DYtHa2jTNIhx z8_&=F8j6gTYruT{>+r#l6PXsMjgFi-0ftV`p!jek9X+v^iawV|YYiT;(~D%tvrV<6 zz}TBnn0V8+vLc5!=cJy^q%UrEBMW>4cdbUq{L4o4;m~d7Kyeg0dL|s1Upb4UB|J&c z_-RZ~%to-*&JgXn5`)|O6IqUD!N|Y&p{2PGaizp>lJ=VG+uQJoThlDIX>BK?u}+KD z6(*2t5#dC|#GVd51Yqvsqj0tR4S3jpIk-2r9+pkhqDw?pf^u~aVzg5ps^x4Y3-sr} z=9osfOxhU#*{1;HFV(OSjxIF2egX<`s(=dG%Fs@J4be|L%=)i&B`a2nK&}2vxaw;T z9T}HK^}U}GyZ%M!p;$9BZ{1cj)2~|KKCgoL{h^)fC!e4#+Y%|GlS}jQOyK^f7j|O0TY$B2RvsF;wDg4~rC(4X%dsAGQ`+oZh$-YB<+ruF;b_rDnErercwJ0qB+^AzlAng%8M zi@@9)iZnBylCR2ctSUW9q*^wSiwz&i&Y6B>@J|bn8G6Nz&j|yuFBRE)UXy9J!V6#$ zxgU7{)*!}7b3rbz1r$*iBsVCA>|(aUvti;$-s%EOv3*X=-Cv|BW6&r$tH&)Y@}Kwr?HH{8PshOUY-i4MwD_+cgNR7z?EkF4W5@ZT z|CjY`cy>12Irwj^|3CBnZ5T^n_y2KyTPsW6|Ht!hJ(B_8zqkJX$oaRnwzRUb`EQ*6 z58e`VTkm|T`TRF&Iz5SI$t96fbLA-6qD4Bl9I*#5MZ$~AATxp-R1)!k*VD#_LD;Vp;-4CXZ!pOrbkT%3-K7 zGy-|g3PqKc3(+c~iVlq54XyjM;g{%Tq%p9X=<15oYZvEmd@RnnyJIdRn4^JIn$*#j zNDXxB^%JO-)c_mU*dvjjyV3pJ*+@OI0Z!X?0o=2cL!W2ghJ1G?_)?Sevb}9$hr2VF zW@iOr%H=hl7iW;Ho!S)rxIjWJ%E_tGEYh;HiYOo6Nt7xsgV7U#FyWR9H0#*}!zz_g z#h&TNdv_f)4H7~805!DvkTse+gX7blU5cFcGst6975y#TgbRKpL-)WE`1VX9S-xf% z`|tb&DwjRyeUo`hTho<<)*(*9qcO9DA^KB=)j=jgxznn`d0VOs(9z4VjTFMUWC zx1OW35_i%pf0oE+O`@$|rXcpE2omWY0cnTzk>u-jD4rfbW_Qft+Ojbmc;*N>rH$wb zm?F@Zi_!!t=)_3F~B3gf<3Kgv?c0p^nN|+B5DpbxI#6w3cqAie0|+{E#kf z^l|2xAJ<{P+7mD~sR?ZH(nIf@8T2jM4JB1%!*0!esBCm1EH+Xnf8kSbsKbkMGRe@5 zwx7sX_=L`S7)L{REUjy-rwQ$M>F$5|bjM>+;n^&4;fdGdgdI9lgzwc9g-d7Mqn4>x zC}$(1v!Fg56r`Dd1=@skHm^KHt`^LMvj;s<;xmP{3&gL{B0k0^3PlCOyiS)YyGXK6bwNfZ2 zXIutJq;V=4&2py;$Y=7S`2~5?SxmAPOhJhQ7a-~vgB{W>tU|04@u_Ua0YNiJO4A9l ztUirQ3Y4U4`=ZFDO>3!hUoRQRT|~udhRDqkN~Z3)K}2Q*!^s`#5=ty-{?yL5*(pPb7IsI#lj}1jYNbsj*ri3I8WU#U5mk%BMNx+JqbA z{fYsyN8uV)nrTdD-Ss3s(fLq*!&lm%rxrDr!q_5;0>3+Hy3C(vcXt z^w55)_;3+PD>WxQg{R3=F$>!IhVxqfm_=crFi**E!b z$)SSrbd8NPjn!RDJ0@3?VvPcDPkIzq+6N>5#Y1o;atG>(wnaMYd8lMrFO)EEgB!in z;f(X)#H(*UN?%{d2-N_cxOXXi;CX=56sA)(K`(KRdPy|j%h41Q3*uG`h+<4NJUY(| zh3Cvh!ya|;*6kOd^lTEW_Rv9BN+zSS;&kLGaTU#Z!{v8(De`CZCZoFSEabc~3z0u% z=q(kZ5%($-y#7AYk&j0X9DC{gEpHUKWH-4qc9TpjyiWF{rBfsMLGoD`PLFbYWBX!` z%lJK;yxp-Ii(bD5jqe>nTk|rJ-;Pn#)dcwy6K3*Pyk5xNF}(R_#vkO@eAv$a0Z#Bc zJ@@dho9*SNKUl^OzOk8qf0{era>^?HFrLF#5##aiM(gt5PSoLlOq#;?Z}^M+>qYoK zovM(TYXO=USA({Agri1z3g=t7phfFfA^!YL=zClrT=I4e3KLs~4jQ(jx}#(0vTZ%u zoD+leRC3XJxpdSdEJ5ktuA{?pV#wpwG<5C8BQSo-8e(E=N;%mnL5=z_xlszP{MZMa zri&uW+{194qC3o+IulwhYhrJg?InxOy@wT>EZ6H&`?S$4Hf!5zrT6CjMq8Oxv%^BeBSQ~ABp|N*UY`LYTBYJpaC1Wr=$9H zOu0~rwE+*Ya7agxU%FXfENUz03)dE;UD+Zq6x$|nIvXQU7)%x9*M$jYP4yMLY+54N zx__=<^vo%ZR=}?$>ubuYDKG55LEwN^R)9{wuZ`-^9uzbJ65T8TuuQqRize zoSQWrb5|!}pwLMidU+NjdbZ%?`*Jwx^-}y&=}EuXPQi0C&ftbuNw}Z88*uw`9Kk3T zefn--hZfgqUHl*ZlDUBwREyB!<^%k5>jCC$DZ%dLNAdW>Pq;4kHck(y!Fy7VkQ0q! zesm5hek?~{)i%sc{DN`q*YVZV9{jA7jHUJ|IJ#&7zNQA)?pjFuZZ5*E(&M;v&oI3> zKa1wC`$->L$I#XCQmCe_g03UWu`e`$l9YpV`z*5Vja<>n?+>OW*(tc` zqC5#&*v^|eZw`ok3}GI{jwhBec_cBS-Aq?Ek=*N6qer&eliyoQNaDjrnm+CyPqfp7 zy0nUs?Q=D7VD)@nbbB8?Y;%L^9|24)eL;Pjwqsm(3c8;6!h~RoRe!2Dm>(YnpEzcJ zR42-Dyi~XEqv-$W8IE;*M>YRi{5j@<2kwNSf;ixxdmgxC!3uUw`9^m2oLTH8v2m;m ztHmzAE6YyUBg9s|RcAjW^uVNf`$3{|7wEeulAP{p0#4po++D=qvi2F(aodVh`X{2@ z??gITyMcaa<7QHRB80UWz&o9#pyoZD&0J9dW=`!8s=~03KAW;hC!E+@GsD?}`mOBL zt1;~MYhG-w`TB&C2v57Ukzn)z`WdjRIx~y524f|tYDeJ(w zJr*BSXMag*v#Pyf?4t88VZ&e}9QjlNq1>)u=&d*mM{R@Ks}DiJUm;jK*9g{q3Zw-w z>Tssw9d$?zqch`+$$`z9F!@O*f9>lj5dF{qh8M4em9a%o;d>bdM^)HXzXR;jCVO_f z)j@W)(@Iurp$B{6ju*S&G{c^r_ZNgub%SbnKhz~T!rpI!VkY>jY$JH*tSP9uBqi9&b=&LjuM|Ms96?F`4nfj*J3+0V zlOW+&sKDrYsNiUMn84v+gg_xTR*>N7Cz#c2D$u)VCkV)N60lK{f|?vv!TQrr@bS%b zv{O$(&8Kl_#&vbeFV3#{Fi9SkPT2@cze|E|=TZLs&ZA7%{jW@5oent9I}g9)7eS>^ z1q5mVs8%(Tl56_lY7oe*%LBxgsWdW}?*9K-Agkh0ggtSSNo1 z|I?I1$>?R&KE{CmOTW8j{bLzeQL>4YmUQ!1^1{hUtp`MYn<{)RJW8iqedCx4S`f3J z+bz4CglT``VSMXExanUBMXQ$sn=A&;Vy}~d;|8Qf(*tg;y9A}S)?gy10pE>Jay+#s zBz&qWU7(*s)_k?$FD@-(His8br7Ab7EVYLgwuaGwEemjG<{et3J4(`=+i0u40?qib zmY(mdptpmEnW&aj-oX1w#MHZyU$aydUA43@Y_9=XCE`UFmo;-f84Idk`hbe6y=MZP z&oIlU0(L7_(xv&EXkn=mb+byMS9aNx#fxT=iY2%CKN%w^)E^>3%figGOrG&(ly@+e zUouGgJ{RU!{46GZm`4vzjU+88D$FgDLL%Asfp_hA1id4*1i$%dpiJ0&Y^z;>;r;gb z-LjM#9lb$cZ2mxhNAIC=^Ul*7{e@)8rh3wI=nLO6y^487-|}tRZqwwxizNOI!0&)# zye!kdd`;CH;(z)XNq^TvZX1M=k-#aCHd+Wq6VgEM>P^^a|CGGqvWp!%j7jt42r86q zP2G3`YIS!rX5TA9*ur^?TP$$PzG$4V+6&KI^v7|px!+A^6?cvlq8;=0(%HH0$an6g zI;Y&3tNn9m|9Mjs{3ya?TR@{uKGpfT4&5vFVeLkLswZU6-A|6vUF$i=l4cxs9=(sj z4+MBQ<0nqOD<;@>ezM?b-ank;+Kg$_$}s%nAXeq{qs0Bmg3srK1xr<`kWouV#s3uW zX>>9^z0Nt4TeVU21B3VY3+T)+Z|W*{0Mc$ zWGDgeIj5M;x(iI3JSF`x!O*`l13t>nfdWfzSFt{hK9J|`tt@S+^480AQm!Hns82#; zn;EFt)X!y*dw7X|c+YdX!`>EWPWQvPC*E*p zH$;iihsqR>0Z;bOiN2#B}Ukbtd zF<)*DSztbwdAxZ$v5~Ta$q{_AqvQ$a=jtU(-bTRRIgx~8=kQb4l=7K2f4Zr&hHlFm zqB8C=)cSNRUa7i<1*I48@L`J5A9kUW$a=ipW=&6BQNTGxeWbEh5~P-dL292SR5sT@ z%%|H7|9T31{hSGBx{_hV;~ikGWJIKX36WB!h5oq~4|857lVvCNLE(W3@O@bnImq!l z1iemRQNEY45}AZ2C*7shBYl{uPtp2ABVS!)3byUi!_^69w3`>mNUdrjA7*C}*A(5AC2HjzBqi(xj>#Da-dJ5zVbq96wvKu5z+OP=S#_~qLuUGQ9Cjne=a%8 zb>v^;wngpO8v6kC=Z+Wf$zS?WH3aqCC!lM+7`KOZAwE?dB)4=CoHp+y67N)DtJxu# zZx;;3btUk3MZ@g)Ens=HpRc=WjOhxN#lvePskE;$=YFeV9{rih-@Gmy z8y=ox3chco?&n*m!F)|(eK4LZEo|hw5GV5OjVA2>9>;VTZ-*ssO~E)M0b<%C!T#m~ zSU$B3mKCpnqn#OWeU$|$c6flvU^0Au;sdTZSumJd1VT1E_0@&|p{Y=!f+w;+uVLMjrO)WVz&ZNr;W=7c{)FS9&PN$;KAhvZ znuecNL#Gp;Xs+fT`d2mz-Ey3;a??lpu6;53=558AuCHkL7cfQD9?OO->HGbEsMly3 zMz$Qnjisrye7hB1BQmI6R7kt7^62#qi%3f7&8u4+%=gr=?!>g(M;^Ct0$(FA^glJ zWnA#1jS7BELD_MJIMOoNU7L*XXd3>9%dn*QHZLml6En8f4_^LQ1i{va zmpiKGJwajMm}>|F>Q8BgwNQIuZT+{0TUUPwA&3G6YQ1Hb$(5xqw)RO3_!ot|oD*1GpC z)e>L8{p}~vGifK?A#lO<%@eSqOaV=gq+$EzRp|YfMN_SE+I#+djcV>M`tHX8n&NE^ zJI3!L{R`vGIz6|MJ%{?qKf6=Z-03pS3X;X#YCBTpH$<0Qrc@$F7-yHApzF8v(nWF) zY4#Zf9G{VZ8xoT7zk3|5_O(BjMfEWYR)^D_$}>q~i7{v|S_4CpFUUQ&c!(j};Hk!4 z(k(dyy6-$Ez6)2ARi=mW7dKbgjaQF9( zyRne>J(3A*REJrGsbraiIreQaLjU)H7<={tYW|poN-2+U--1RwRi=w)H~3*h>>!=X zp9+&~EkRLL7ucA)ir|7jw=f~+ zKKfi7FHla}CXh>5DhQBSBWNjGB~Wl?1t)A(1PfwW!7*h;!TK+vf)3|qyq<6oueS`- z=2zJ~izj~Yt!F809}Nc6sErW5FNSQ^{Yz%O3?(@nKS*=#3Nm!k2(sVwGvE9^@k@Wt z!VmEqu&8Dm7Bwuylh2fKXM`Ij9Sy?iw_LHdSsYh)8PIh)>g3(JSn}xjYEt3mNIIAt zeoN3ev%@bn(UH3&T>E@Bj(i7t`QSXhp43lT_&1yDI81;o8*GWJ(Q*jNn+HjgWyq_9 zK%(Gj4jzxbkid<)a8YQCI7qpJBVPozKTC&8E1Do;-fL($^&SpNORy)~71@tJufy$L zS+)}gVcf4~*rjIf|T*K^z1n4P@$-652mG5R)1W@S@T< z6fdf#qXu?pHP4wyOwFWqb*uS4Qzw#Xei@|iVF9&YKaF@jEXMhc+1Mbu4pS0ug0?M! z>E1S&+~)x+21?-Ii!|^udP>%`PNx~pq3} ztyWL^$O&>P?=lHgo(YeS_mJoWYWhahj#S@JM2T10xO?Ol)o;ne1*SHrQ{aw4kE?Lf zjB4CDP=O_7o!Gz08?~DAaYjfTzSNJypj{fc(!-V>uIppAW^ulfRCAjAb`g>NW{Rfw zi^$Ie--TIP8>mLS3OOY-1LdXG(~6}sWWpjou~6U1dFj@Xh;dW-9z)G!QAZ49K?1A{ zS_>Z^UgGWp>wzAzfZv7N;IpM699OtSc3zteJt|+w>#0-V<99_^p?8B6-LmN{7@NFgvS!3F(Q1TV%unEd+WU@7(tAtpd=QXR-_DT#F6fhy z0tp!Ey-!qZr^AJwaHc89l=AMqr5^+K(W>$`vh>qzc+sl>?Ft9U&<{n3xv#{8SW~l$ z+G22b{1{b_l%@L|BS=`bF+AQAL!@^Y(62XVV%1~rbuMw2Mg&#hqU=z7@k<}yI(u-t zpbu22M493%t{)H-La#kyxT@|jvt?N@QQtHTT=($ERcQsdsUHFJPYB?pXfPB%SPNds z8L+xx55znS2IE2`LHyp7)VJn%#qE75p5GVxU2=ndxpVnzYbVDu?(ht z<3oklPdF++1~oI^!^z*`tWGKq{BVEPJZNHdI z)9N_}tlv(0yfKY`eX<&=XE|`0uylGORF>bBu#DN(+DLj>IT$+pml(}wpu<}drjI&M zrMX+_Y~yB{{8JuhuS=l+9Fs9&cPsimiN_xA0!*Bkfxb)rQo3m+is;{_hwT5OfBoOm zlTGR*IPV^M+#kkB_ep};^Q8rbbs_>86Fq^?<|zX6@@~ABSB1e-N^x&PGhP!aN;Xh)~yB&HT`hdsm5YpGX4h&)z@^|~* zp#@4CcprQxp!RppdF%9#q#u|_PhC>rXUxCHobK~*ai zsltwVRP@C&Gk=-Y^vN4_JZQ5Y^)=4njbABv>B1R2Sty5M!nycjQWDO*d>O~tYhcwn zz$^1o=##uGUii8mBFMT(0`n_qUV2OQS8H_=6YELtJ{{uk{Jw~9-8GYrT{Si{T}=6@ zl|giVfF)TQA5HS5fAKGg#qk%*RFn6Ea0* zLWjU``x#g}tOC#9_`>e1R$#8F17|p<;<-?NErjulz?&luLLIh+1-76Kc}}enuPptYDOaN-Ij)9+hTB2oFo=6^u{A6 z9niF3n7;REreEWhkVj$y-kX(WWWkDPay>{7#;MPOWp|KKW};2{cU z_NLHAuetmeuOE_Dr3~`!%0d`?pTrz>8zIk46yR?y3-6A3Lk73Mki2{nN_oFY=4&a? ze&P;Zr=v)ZQzq&46oaLX1I)!_7yiRXqOiTtlsp#vN8}_+2s7~u8S&_4h(@hxy_O_> zvW0Wu`DD^kqDUS3^srHjMIHBG9KTVPJ%3$=O^N>wO=kx|ZNp^tIr$D{TOUG0yfNEu zD9iQ+48qWxpHNm?25zbHtde;(+>gBqA|e{F(oY-{R z4kYjyiPah+*EiRb#t+jumUs$OTsr|zleR$nwN!YvB^j*#OM!fia#&K=4J`_V@X(&Z zkn>@1UY7jtb8EAPQhk%q!cgMNEBr@yZIOgf5 z;Eg7R_GrZ8vQl+?6*&nPX*FVkkPe2N&B1?{7vOLcV08B#Y%bzqf4D--ULGv1skI1lOwnqvJnB@cT9*KPl}gfyG+j&$oc7p}JrfattI)q=By_ z0V6Hf!1&2!`1J8Va0p!iPSV#vwQx6V-7Lv!=yk#UPYUd|sS{Y?mujrCLjw%9Jb+sb z@~l()N7&^#navtBU<00OvT@X$?SH5V6+Z%C{qIZgbzvUVB<_JD-#Bl0=u9|bIEQ$e zUSoK>`{xF59+BgDntMW@j!K1m%$_?B)M%!1c6TaPxf*kBToqhfF%$j5z|G zgCbnc+mUoXcP1)1ZA7lZtmf93Ex$Ed66&H3kOJ>4uCuBD+T-q!f)A1q_fQHpYWaco zjvR2HVUYf|3`Boj0u8%&;Cs0U;!LU`<3a>zT!@13%UaO!rHusT_?vCrIYRSQ>*#_@ zrzxwZNSAXiADg!f-FnQ1zck&Rh)Dmc@jRl85g948@U%9$^h=+fJk-kk+jz0Y_1_J8 zEm@3~?es(aPz{WGD2xRM^6=@pX#Au+4ZposMcGGh>D>E;IN_%_cAF;Rf0tbGkhcMj zIwxTB*F)IsxQ+6&uJAY9kcBU5tDy1EC2%j)25E zX9EkfPLrgPSTb#4G83A(i zIBBYCWk>##oI}*a7mXII}?kY$<_Cfi6rsrbvw|h3J*1#f;~S zB4T^anu=>!)5E{od10>%Vd{S?;N_QbpwhvD{W({7{OuTA@Kl0}-8;ZTA|E!bFo2Rx zvf#Ql4jwBO!ktgi(DkYqephaUvcL*Z>j(zP3FE*oPy-&^b%qIo39!Jhnc7El9r#6> zL}C`lGI+r;ZSz+1Pu+S+u2qImyP`}IdDaBN3oF34xeGo9Rm0_J`(aHV_pDaQfD6+a zq5X$E=sM=Z8W$VDdk%2QU=!zRp9|p{$6(8)(-1qD2-|0WhEb{UtTFffViqOP`8W}Z z{R?2dp(DxNp$-1NmZW)iJ9!=JM}O3Os%ca1tr@+ojper-P(CadHKI;oS7#e`*B-@_ zRmHgS#}$0#?~4XvCo$>CWnBLBF51cTqRbshfzG-FtUOzQ>Y**@xjzLPzwE^0I>+#Q zxd6rQwWF8=;HdDlTyPrHKK|FvMxr=K(@y#P12Eypz< zgOKM=@q%7CdQ2=pG(udnsshK2?L%=n0p^z{XuZIo%QvOG3zuzS*9*2;x_sz*7*K)pM_g#Kg;eCcuH~y^S>*)RV@hG|DJbhMt z8T-dsVY8?Z`u&)Oi+hga=64bZvIh8HZPUV+Tf^wdq6{kf@;=`@_7i`_?rpR$<~J2j zzC&N^Dq_@sPG|N`QNs$aW@=Gc#(U`ap18E85wjcG%(7%|6NB#%ppf`mzZTW5_k1sqBYP)cG{slCEv2&k)NHHnOXLC=mKY9d^&4AdLP_@-yh`BC@Uj0 z79>;s)7rG?=UztX-92iZ`Iy(rb)PiCMQSyC}CycD$1yTn!V2S?{kT1N>J?AEp zq96mPFD)YWd#6$1J)dZS@>Ep+A%;m`?9gm4gAF3 zeDG_&FN*LT@Z%OAR7+our5h{hSfUD^)HN{^*vrrd8x-;Ehd_EcgWHuX(#FF(eDPX_ z2^w0=!peawcv+Rj2rqriFNo&l3quCl)7*&DQw2 zF{fuPg$RyoM;cC&ZC}zz%&u7KZ)}W7-qUfo&JWYU6C>I$;9hPwUenu!FDshS;AuDK zLJ$!I`@Z6GDG#tI{1Iluc){5((*zw0lmxSu{J;@;jG^j;i>!D5Lx zY<&tdZzbUiF%R6P$9Zt1=b^N_KixlgnH~cX42eBS>+Lo1zoIs(Wh#Q(A05Qak>hb> z*E9NdVI8$wca*HQkiyteEqpmLKo_aaK}RJk{P#iu<5t#U^5GM>T9AO+N&5(lie{7kzdm_p1HxcCNL23BsngO}DHp6sFTPPCYcxYTcrTKsmOHS3n z{_;PN<0r*#DXxL0mMS>km<2oK84yen1LjnW*&m-ZM7Y+59Qm`AKAsuKkDdLQmn^8F z{oMOtE8gW80z>r73Ryfc;EDI^elU6!?mM z!TpTZZ`0Xn!jIu#wkGT9sm@-Ty_|hJZwedRYs}6plV!c@1<<>)7*gja4Op4CguQ0)7c9)B+34~PIJ@OGq?z7^ zm=&dv{5J%8N|uowF4AQBcSZE*JxdjiTGA!@lW{}OBWk_nDZQHbjb2IhqZai`(5qxR zBYa-ZjA!c1?Duz~rD|j4Q+;rA_E{a3Ai82*;atS!nG%!cRWGit@L{(+O`9$q|pM%*Tz<N^>9 zlTyX*Etjyp;Ti2$Q$o<~lo;#GMm{6Wk#3P4X&~23&)$_F1q- zI3L;_62MI{1FvsYK{kO9hr5|nUC_wXxvgUcafF{YIu1IXd%#o6F`_nC0hnP`Ncl;K z^7LS)MKH|2zs!}nC)7kGTtXOWP{B{Xgs9DCMe=nSkIZ^!!Rkn+`H zu%yKT^!KWPwdEux-`tBauGz+WI^sh2xcSqAM<=5CR(+Z|Dvrxz&(cHNL(q7-Dazf6 zz`K%uc8vZ9cXBT-@(sgF*FiL}EAjf{XnHg*%+$xRmOJC0 zflZ_d){bw3WrAn$ZuC9qt&(DW-^sG4`Y%*``-$OcaLK3p(LAc|jADW*o$A)VGxS-epy(+A*p3CWe zXbhkZH?_$Rj>{=BDWAN0x`F8&ljHZ^iz9QmJ!RQo1arAQns_!$gnivx$?@QiWS;9T zu#=P04_ZH3$tGcH_A#5w>*Vwh$JOrSX;_-+Y4MvpOI{r>^hhRo6z)@f=SV5lgT}GhFqX0kJhyIxiBI*xgUh~1yZM4;&=k1t zJRSD!I|A;@Ho;@vMA)2}N-DySf>L}RiCTA`+x?i+1#dDTp+*=UEqhH$J{dz$)kMUA z$siJagv`HT1Xt%L5XlB}ChS%oskf1&DuuyB(q$%pj-M^ap0|SJ&pRQa*&6(=EQOQy zcSzYOAy}-S4{w%;6IUlC5dXUzc8%O1$K3bAtO>&~_OcwrG(W>Mmnu+l$%T1WFTf57 z33i*oIS6q$3!Vev@V+vQoDY-*71O(5Sf31kTZEzf&}?T9m&l) z*YTJNgE1YO(W3n%1{n9#cOvQd(0VDJb6tsslWlQr$75=~Ybs8wx`Ydcq)qi_^XMDN zIXhpw+OLbsf6^q~U&iq>&3x$b z_9~VM|Qxs*~m*j#;6aTi)9r`_63_cj|AbCPH#OBgJj%gtd zy37BP%=XF9wRaYD-ztHw-9B(PjKbTGzu@^v7542`VOHn)7@V(~!g|gb$41)?gX}yh zc45_I*7C9<`~6iP_$G?8VfOl1*wjwqdfxbh7K`EE@j3 zjIVZdC5_F`<{y3Zj~OuL?jxnnLziV0+_LL|D{g$wLsXSaeCeaq)&sqg%Pt}vz8E0YrS%BQvp)jvL4upmm zL-I=r&^l2^+=?#|hxtM^>9L8#lFOxNlpi9m%IA_vkJdB&E1QYE+6`hQ$RN88IWY6& zHZgJ4=~Ur_EULmJ{Ny+usay=M4^T(-9986*WODqhNIbdU62B!lqszS6RQ8ks@jv*D zU-#iFZK`%7weBWeZ%W=XCxaW=83K6{VS_U@v&Vulv82I5)$}xZvA%o22 z_SMDkW>5n@+7A-3qG_aJTncro5vR-seQe#?MZf2WVySo-RoV8C_gF;*OQp3iG=B$v zaM+9s#P!iPt%^2se8u}8#A)W*eQ0t2G(CJ!2=!g}<1LO`ShzkOyKM5%H|HX^C-%cY z{$UJ17>qmSO5ydKR_dC+0iS-T;1B7UWQCC7%SZVO=& zYZp`7YcH89x}OfM9cDZ%j*`|M5jeq$KoGHmt9P>*!$NZi8an{D6rT|Xk>~vHjvb84 zrFzDgXGZAU|M-6*Ip)-@8=OzVn08&3qSN<;6VJGt^x?o4x-QZS1sV5f>I!M>sENjE z%WB-_oQcKiws`QQ42HfNM?RH*AZ^F%n0hd0Vk>mP`t>PLUb_*-(!4=5;S^kGs)DVf zGa)u~kmxOWL*}0_uhGykL zTJZAWEF$FllQi^fAum&u$^32uSSPv_CO17N{r9%OioTa$ZbROlx3kb~!gO5!TLvvJ zmN9-i^-;6=F7@pe$2D`8;h;++b?%zS|MtY13MULOrW3c5+4l;_!_%ttI5|f@Oz0v$ zn^y4sk7&R#jf?c9P6{(x^n_khR^r#LHK*#2esj!_da4;3MPIAup~$Q_`f+eP+Ny5E zO`p%9&rgo|csrObwo~F~AFCs}Kf;-C;}Nq=CNC7+Y+T5)hNIUhDw=fZyD2GAwmP}^_`{u8NzQKJF) z+)&5elTgr|r^KP*2SAnMLFKFb2B8j7HvO46J9lsf>uM>lqvQ+xCX-`gUWwzlW;c}=0T)Gh?n!?-+_r!knyxWetVF=X(_6(aI$ zHr2N^BEP1t!M|H$>HN74{H~-*Mn2yS^|l~xTPcH+OlIJHZVtITI$mIXKvmE_VVs~g zQA1#1I!kb~e-IyKc4I=!CoD4=#Aj>81(PNj2y9O233g5!#gImQLCQ~Qf!=gg0aery z91hnNM9-Uu&8xQ1&ex+f*?lEV88-oMmq}q#-EZL~zw@{IteJCDQrvPly6^1MOhUwd)7gVg#g6lE9V@8HcX-JO_ znhgD-j-MaWqnoDF=%w->trL&>2J_5QvuYe%)li(e- z7`&fKLaAmH+2X~gyYGtN0;Nn`aN&g6LzCBFhEOxw_}df zSv+h%L{p~h!DKjue>X~z6{UkD`=>5gd9{%rUjsloyqpvVvvkq@1nR+gTKXzlN#Ac% zsIF@unsesDOvmY{8$Jt{50>D$6Sdg<;4C)96QvT%-jJ!M_C1&uL%@VqGo;|6rd zLFKd5>d97&mp+Sj3uoc&Z3cK($_n?&8R5EVyv+-D^HqX7xU4`gm5J1 z!R>b9+P1^#U3O3ywGM2$y*P$*Eo|KD23K|a76$)Gq^st%lHq$2(7nQ(|NPhlxN(IA z#d}kE+qxbQ!RQ|7KRy-e{Z7E(QVf5hK7^)LG5!VTRF%zctQxI*Z&sNXCJ@FEZ#g2aRGT^MwJb@um@7P8MVAu2{5dOvgEaF?h4o8)p=KpyxkH;QoaMIH$o0kMD`% zJN|i8H9JZZ6})cJl?TMATEY#wyyPJ6?$yFYu_5@OW-|_1G}70TjIqw}5c#7lN^P&t zA<8;R@W(`k_-dRZ&-12%3zq|UJ{Sc{f-XTo@JH~zF3Ao}9s(O9AvS!E0^6wn0pgV3 z!ROfLP@DV;jFmq@)7utU(q9d`&i=(~3TY@@d>E@w#nH%sJe;}DAB8HVaIvs7?hQ_1 zKDP+t!1>cOasOxfnwzEfKKn#=jx-R9i@TvO@iXN1=EJ|t3G7twA<(*c54<55MY2&Whr@$`Y!!=o;QuxV|{EPIUy$tqMrl!fh+vUv72R_=Pn9FB2=JTV=Y|k z?1cS6Ti}QJHE3x}0~^l)IOjW^9nDl=Gn-ej7OS_gGd8SeN%~^8M9+k+^*3jS6O7pR z!9uLi@&S12_88c)7LYjI331auLX+lt5O;3@<6wD6jTeFg+mmVCrTNtT!gKmsA)Bf{ zlc{FIE-;?-HGiT;6?0F%pSlapM9V`}G*Zcj&X})Bhqe(K!B?SCT+aWlu%DU!t1bA< zE(i}BTH&fAH&I3A0$z?C!y9uu@Q$vKK&L4kz@#sJ_D*)9SPW7iW~`3@4YO9wtw&V${JMW*ceXrAshQ~s-r zZ2UJc4QDeP6Dj#R{@Xr)4O~C?n|KE5RV2_UN<~!5LW|$0JQ<@mn!(z?T#i}21QgS) zK=k`*=)1oF$`%`g!TETY`=1)T+8aV2MDh9iO17id-z=QTy~eLgJ;Vj)<_h*WJi^Cw z+VFr&0a`F3f^!vZsHVun^3b)IfAl2!%0(dlctHhn~yPcsA$4}Zq%QL6+)k--9qeMW*KuN?({ zKYqu78*+jK<5bj99mXd;UZ^lAhs9I>(6Jszpr3Py(a{F@rXK?Plj6ac%Mn^xM8dO^ zC&BpHQaIydLVo9k)1L4GWEtfdqFq!?1RfFyL@22|g+Xb?a^TWf}MRwX*4Cywfhaqo|Ks1%%NTfxGD} zX(xOfI32Y(f1qsVBRZ6}0?T3+(U+e}7_Ymh$)wR%#?kQ!ahoRwzk7uE4>OnYP1X@& zP(Ood67nG1Gr#bjUiKzMNBR8rcOhosu`)#MQx~;AGl{Hp`Npgl$)M+~&8cC}ZEEmO z8XeNg=<06DsGbj?M}{BLi>dojqGJiR&6B|RU(WQU;Xl4l$w}g4KbPxr6+%PpVyIBc zhc7|5K=OPy#MFth8NWr@!Nmjc=p}^(e~Vzl*e$TUcoQyhbD{P6-Ee$%3TPa5f|t`J zp+V1@8J+l(+_Wj?S9+wQ=}LcGu5%6zFI~XioAEel=UhDbO$Qk{Z~8AJmTG<nK!7ap$qx^TM?Rjb4mEk$^3gCok(8ZRAO>M z419XjV2A$;^5k^}k(cGZZn%@|%dRA|cRN5yf)D)lh=bsa3Yfd{JnTvT0;hr>fM04C zSk0@3icKfL*~A94gx8ZvHws9{nq>ay`*Z)J=)42D`rbGmva)w&W<<0|anEy;q-aNf`DAFqd{4w}$?WiyQI4TLLUsbCNn0RP&S1M@orPK0Dak4g!= zouUYu$0T8Y$VRv+?0I2J4e6#VewNyvl#w-46e*ABV&z>*0p$TsQ)+ zh}04r_|tNcv@^vtGVx$V;^W)&*&V1j_Wc%dqZ2_+If0yLeL|yjY>52v1H?Nxh7>>6 zr*FTT!b)ZXgzpZ4)s8W6$GnvU-V$=36ZQzZ<1R8kPZ=im{iektld)j28g4^(a@cV) zte@ry+pMEu_SG~v^&}p;PaTE2fe~;)LKALPH<8`Dt}qE*p_uvcAU0>!HzGu1# zKk4HFK7$MAk3T)eD|STkbn+&Cs3wG`Zaetx>T~#)S1kD8g_eBs!Bza^vwFPPkRe~R zw;xx_O7qV1CiA%y6nWpqqC3Fqyvms?SGMK!vmXJdX?7z)n~Pvpq*N4Jaf zN?X6+gxUWv?OP?jTwaRJX;*PnMZ0Pt3 zV`WFPo)6k!ZjlHZAo2~$t^dHpdE?ly4d0<^qY3MLRELcR2lmeQ`E1*34OS~x;GZyx zY>mGWyJw^roBHn{?2`~>LwAp6CrW<=!@L5>3+X2!N>ic5Jq#>6Cei@aKg4g$Fnv2M zixw<7%`{cpk+D9LW~?o#A|3ae$vlnaM6q20W==UoE}2~<@_J_UYh3}E=^_fUM#|th zGoD%7eTiI+7(v7i+@N1oMqG|_dsG*TR#5Y_ab^7|?n;8ya z`@P^_RtS9GH$y-gd?7az4#LOk_J+;~d_eb{RYjBG zWHNHK5%hj9h6(S2;H$DXXhe=+biSmh1)%*K2j4L`_*8T$aauU^cCECk05nn2Bf%; z18rg5i(#jNZs>8CF!vHPJva|r-bO>$f(?*SJ(}#Ff1cU6X+H;8K#z|JL1iKT$>LJ9 z`*941=EdW}!4s%E$qEev>@fAW1NMy`hf+=!=%P6ZOX57ytyveFvNG}2lW??6eMLoe zeXvJ11D&6r!wHHOXqCg_f6ZG^`$`eYoNGq+ElFtNa~oZ?+Hsr7MYLbP442ENqegQs zj*V|aTmB9D4PVEAW2$JG?}fQ^ zALRq6#E~B+?l5M=4rXz?I$1QUkCWcblkDtB&~g1nJ}DP73Oz&Q&el25sA2}gn>}em zXCm44Jzwax3in_A-OSPb;W)MSJ9R(VVKt{k61GXdtGI7mV!iEVCHE(NH<_q)mh7K< zkBptZ4`kg4^n85;uYV80UXdC&vpfv$SJZ&&z;p-_*#rjN9whg57IFSO5g~jlH~&}y z?YCCvsHKo;-}st{E%GI1*9z%Cr3N{@`y(BEl7_Tii(7Ei8%yR%FgxxZC#5z!z-~+z zakeQ1InO7cqVyJCj&ForjR~+zU|wu562470S;!75CC*y6sodlPxL7ZVhT;`wil-$L zxOW0B&Dcy6>*~3v;jv&beIpUMCiocBMv=A&4P@`E5;`+-4Vk6ZLtbXvV7JR|Tb#8PIH7NfbN8newkS%yq_qI4HQF-HRC1 zvuMJ)Z9IJ|z6C!81mkOEp!?!8%G?d!lCZgye4pTJ6QhCCi}&1bHV%cXjFn4nSb38&*$f$&g&VtG&_&(IJ1RZuewuF zQ2m-&Vd_H0@6x63-vx6Eueidhs~5?JJ0j#*9dZ-$t1BeOjX{G88rVK=3x@q^psowm z${S4#sO}Oc?&z3u>Xdw+1`Zd|O#Ks_`DYnCw__e!nyO)R)OhUpu7(H29@4+J0`c_b zC={#TDy*5c_^SFR9*=m24}9WK>|!y-{8xnA=j%DBliuHx`0pCbI+x5qi%OIUH&{xkcyycu^!wyc}HgA@)&Gd zs)eV^0ITSW}KJeCN7eltNmZ6s{0yhEl&*@AQESWx z0BOZ^xiyq?_e60E7X21R&{Jkz6}g3x#BKU@tDU=JW<)NS&E$r5lHk4@93OU{6Hl7T ziND{9(?rD4$*l=9Q-5OQS8={=%1FN4bUOnTH6`osKjVLomaK0LF1#o@Yd8qqR3rAL#!Y$?(bQIi%C+CNuU;_h{ zjW>zPC!rfwqKcEhg`va0P3YK>fc?A=~1_M^QxJLENzz57s-&E7B!HGLs~>%~By zJwZ(UO1YOG+UdD0QPc=Cz>{A&3?Fe2=j5uQvR5iDCxQ4P^eHv#7ru$Ru915$+k(?8bm7M||>j{{D;Y35S?sI(ieH*+z6_5LMNtGUO@jh^YMllfxuu-|YcIqYl+RoAw1 z?{}Hvs;?R-l^c#yIs51f-Q~ET;V?aaav6>ak;J7h#qj+YQ!c9J1&QxlPFKTSZm2>6 z&znD`S$6}mwJ#jMCq(1-lhzpaVjilkO%eL@Ht2Sv9uLg3K~BU6U7oq)jwQPp>5f;N ze@QjB^@TQUI(rvFKX-wG?qwJYC*h&60hnv~3eL9Iz~tycxZ>IYInk%!W}XPUH_4FI z^tEAawY=F`rS|NmOU~?0fp2y0Spe&1y@B0SZO69S>$9)?)LF}6Lw3!#v8;m$4_6l0 zu&ED}*a5xq?3bEWxFj#mz6c)0>c7)v=LUDeGa}AT+?EToBYhzE=MB;`e>u!`3j@ca z(co}s8iaaF!^VT=;PJ{5w2N;sy8|PMMdlAi#(1URLX9NS_9AduM+t15zY(Kx`NS_c zf|(j?L|^F{li;}uRO4eQV{ziw|7Emf25&~x&)p&)>z7z7X^6nrGb_Qy=N7RPkA|LT zQ|^m#6j9$dinG70PCCl1Y2BVG?po?LPW@nv)qw3gI?ika#yA+_Ve@ggc%3#D|FfkU z%x=LGA>>k?9--M5Q*m@{JtYSeaBYDE{+y{IWG!FN=bK#7BP^|=)@nB4)2?tcw$J5M z$Egy}l|IB?Czsnb)(XB+XR=(rfni1pvkDqT+}CTurgCHAIir~PKe)oYeLI4@wcSVx zYBjii8-WY-!3eJog<;|uPu$kA3LW-d!9j;}!kSfvlA+1CUeOP2{zjlemOb9qh`@{# zWAv~rq0{v_CU;w`;9~5uE^ZR%tX7?2Di`T9cUHNR@YX3LB_X~nXx9g($Zj6-TjIm; z-IwVr(Xkk)_mB>_+R}1)S^6MwBP|-Fv~#o^#;(yujeH|?Wp#IYJkb zhXaVVoD_VxF+@7U+`%p92&woo$jmLUg5lTV{L+Mebhdnl$IM6a_ZDA6UiJzqed)l7 zZ{JX*9m~+zA%_0)HsTsHCob?NVDTA7IO{ku#=+NG&aXd`)}SWL#Q`bMN&^q6_a?~(F2 z6}WfvCuv#qkf!SiUFm&YbjoOdRQsEa8`rNvQy)ikPu_;=^M2CX=8o8seGKD{3j4oT z$8l_L8W!O`{M?v;#v&1DvhfafuiuG+rj9cGx9AADcye825jkJ^m&j+lAX692fKPia z!Lg%GLTBzAgv7N#RHXzf(-{eoEBauk(|=GCHVAgJj)485|Hvr+1~O`!3$rj;lDxMg zOlJEzt|`5n3GB$Cua^P*6SIJmr;3R3w%;Vb_yU<|RzY{JjO2WaH!><`+-Q<@7Wb*x zo!0bj#>}c{RM)vr{Vp!RJ-4K(qVjlrSbGlhPB~-W2pwFKI|ZgJ)gdGH9w6NwBgh2T z%gnnv5%602kF@5+!o}3nkhx|xsMGg z^&yT1A3jI__czh+a0>cI7UH`S z4PMIVJ&yY($3N$KQATeBzkcIA?6$s#hqt`M0k6yW<@h;Vr6Pg56)!=<>Jl*W;9%?c zA$XZ6$IfJw*sq_|*qRCl*49mjb@DQ11ODp+_o%t7vgAxw-Ekxvz{{{PgIcV~K5Z5p zf$dwQ&r)|AHaKt&`$^7#U68BCPEoUEZx>Hu4>sR|3pzva!{Y&X$3B9Cr+>k_p{p=r zS%lEMvw`go`iQ}c&DPQz&JoLX3|XugCU_i-@k_x=u4d~f?6k0>mLV>5Y)d@MpB)d< ztJ@$=X(L3}1cS0zBJ?IX!TZp?a65e(Lw{^eOKA+8RIRA!nU+soR86sel@7+3 zG-H?gb6j!rC;HU?!9Q!VFhRTr3uXSp&Bx{Wkrl6StkN*9xh%;?Nq@jAuI0Gzjz6xu zuZhYhjtKpyENm7kL94c&pgLN6s8mo2oiAj66&JCwNpZ6HDEK(LSLJqY`Sdz~CUx zdSipnY=t`#fj83c=WI><3>b}s$Hd#?0C7>BPtI4H^luM2#H)$35Q6CD&6J~;l(|chFH3p-o+2q*uZf;;xIpY|3 zlqSbz(S{>ZRy*E(q>FxKa&sD*Y5Np0d@x-Ng?T!+s9*xx-4@5q6ZfID*D~s(KN`RL zHDTTpd46f0F@N~VG+uwQJOAYQexAMQ!kffQ;>*l6c)#cRd~}L3e{cI2JoG9D|7y;_ z+OhzY%{)PKAE=S*iqZ6wnviGMS!FFLWk6_5G|AermDf1NtO^|M@UDUV`MJz-LYSJBAD%_aU#uZMoIs?{eL($+Z4&Hr4n4L>>PrqK0@F zqK3fY^C?C1WP4tBmb$<&wPMzWI^lrqTpZkblZxD+h4UD$A>91XQC@JH-AbXnt#3V-eKr1^3>l?ZMx!wk}--Aw)PA``uk0)QxCv zw+IbwcH`nPMmT5ocuXG3rPr(X!uDcY7=L#bj9IP+B6SDguwn*mUfV<-UW$N__EE&{ zs~o+s-HFKUPb1U*C`0{JO=2%1M_0Zspub}lqs&fQEPk$o4{oq1rZE#U4mQ!sL0|gN z^C12R-iU83197S85pLEufq#A}i>$qKlniF&lSR`nkT>%*hVoo;nD>+>i(Gl zc!`;hDyHYKFV7QQrT|V7UxU=o0l$T=#A&K|IMvV-v%W;&8kKdZqFsi;9@=DfIKJRmj~^4|4Wzb52@Ar ziI{ikG@YF=g0`*frE|uerVCDllP@yL_^-Q(T0PI9rDJue&$Oq^&Y1Dwu(g`5_pzc= z!xYPx2e&c_(upMav^Cjf^OCq8oB~=|_0~W4KBm8GJ4kD}JbaGrCdQ(g;9D3=_RLp- zcA04MJIs&1_=@yUxyq^NOPeuwZ^Eh_}(5#;h3j#-hnuryM ziHATC;|kh(?y&u2Hki-$f#kyTWZ}cBp!+uvg49{~tD+AV>!aX{kO_q^g!rbh^h%Bo zon@(nu}|tS{m?nw{U#kZb)H1G?hgEMX+9p1E5~s+{Qo&Y7-$fHx@O8~e>a7y zO5J75l+P0<1b21FsfWfbqqUyul(xO8czkC!e^nS}7>oz0( zZ*r)6v;z24oupZcGB|nC9wM@>h?-yf%(Xf0#bWidT=QXqPxi-Rx1Bn(|FAPUj`u^u zN(MI%&%~|b!C0~ADZa1k#FpvV*deUXein*&Kx`-KHY~$M({(WMGmAc6dPGU!yqmpG z#}}zac*Crd)A*c9XFs__E6lp+Sdl3-JL@#}VQVk-%qXP&`=`;N`XXZTOBu9c7C@xA z7nrpO&vSw#*iU^)<|`zSEz2p5sM$HG^0ZyCvBA!8aG~a!x*Fmz<=TJNl?}n-BWP&A^RG+wkF! zjcBLni=By^u*xt5SNsqhVhii=?D%tt?Hv9WbsOcRim;RxVY=Tz)Jb4bv2Ql+t>S6X zhA~)mr;R#9^i{}B^rR(*p|nHbCaXg`wQOEVBl*j8VW|N&MO4rQ5w>_#|1BMueS!|y z=QH((Y?vf(!TEPtn|Z_hVH(pj$x8V^lK*rOcj`bUEm(8Wx?9Mu)_f7e?8`_`o_I|Yj*#$e%yHQ4&6g6=Ks`a9*+R%*3Nu4B> ziS0~=S0>4RG7co4%Rtf+TNrcj5QN=21?H)_puOul+?ZAk#zqKwqgMjdYrz&FtN3f5 z6M(k?d^qa`9JxuZo)5F`DL%&yy$Gf9TT>{1`a5UvvW@XMw3`l?_tNP$Rus&4q{P>=rr%G;9`QCK8y7da~eCUGiLJq;>s1-F;gS3KNa5W5q5yFhA$#(=%_|Zzuxpmf& zzeMp>QWy2uW{jzy%IG7pOe}u$o=QHrN;iCU!Idk;u=lnM1Hwr=j9Q_W~LXk)%>vAM-|4)0E0dxT;o_Oq_6tt6O7@57Cd;lNmr{H9raFpGciltlSu`GBZI&9jD=Mv^%M5_vlO?1Vjj#8L3c9`~+&*gGf zXAs@_N?h2f^B~#n zrezhE-^a*rN9$DgbELAjj;>8L;y#fi>o(UoT6_5?vnkL4*Ddbk9(jzWD?YuY56=`< z1olm&d5eOmOGy@2ab!9jCl*V|-W%j^yc@mPHj=xpBM;nvCG>?X$NZftjnWPEw6fQa zI&|9MzBYGZ=GsOhn=^5vu?<><-@#|cqtPo{6;(bg5%zPFadq}4Jl_(JD||;|!N`?( zTSgJzfAqj{$_W@4un+r1364JDh<6?RadSb8;D!H1gT;MtN%&q$}zNIkRUT%yHZRf~gafj5T1))G!)2{j>GEoSXFZCS{u2 z;zj?Pf0`>B8$vS{#naPXCQSeIyUf@uQ(BPfPIF3r2>iB7^lsT);#kyc9d{L|h-U+t zJzW>h2y0|~{cSQ}qXxSi4_V9IP_~X3Oe7;4CPT}{@emaBja-Q^fZCJRV461r?g#5a zP4P+iH}?gRSQH4WJwt%M7Yr5817VrbX>hjsA%Nz^+0yoYIDcaV`>y*Nh-#KV>WyxA z)N%qg9Nq~}D`MgFHc3dDSw<{AdcgdsM53^~n%uJ}w;u2nx@t>xsm#8aU|y*LZ#{M3 zvuF)joM{IMqwK+=+LcUukPKbP)8HX=FeW!gz&hz~%x}l-;G|+(kuv%o-Eu02?wQ;{ zU)l$tsdqG*TVBLO0je~mxrJIKr=a|8VU1}_!~?l0c;%iq{vGPzT3#hkIVXA0*FOxp zH+V9NJIshqG=bpHHemY54O}8P5R1PK|D8V!qBFbU$DV+Oc(4U9J;@S|rkXfH^CTTS_}x>jIH z+1!L{D;k0SCePmNRb?F_UxWWp2b_9d4UB=nwJk7Ux19P2TR#Y1iC%Nou~d$|c1w&s z8dC!odL7{Ye*m9Kvxt8~snwG)8|lHyChG0H1ve(|z?8QKQMq6pYJu=B-$=zROdRH{&NgJVie@Tu!Ya^oaruEs4S)th|4`i=+aj-LzXvmD`L zVh?lEvW4k?AO;y_ytQrfQkXWJ$2?!0M!U?i$oEIrNy%kfa(dPSYNJ1r#^xO*7Tgq~ zFEY&BT$x~1RePG2ZFxYXo0`dLOB-${Uuhj@zliy__dXSEUq_dQU#1QDzSjL37is6A zlQi-`HRbeO4!_boowf$gx847J&_SeKc;{SWcHfHhv6@ryX%$ zXvMOf^hRnZrx$Sm-Azw%Z)>k}uKRWnuac2ue%D^+_ueX6V`?Tantsylu`qr3j9QU5=?NQY`PEgzwRJKB}8QEoLh&&Z;GPO2d<D$xE@t_fht}V2s?)gh3*h#TOtR)i0i61}&$`s~ z95?0Xee%p%hKv0Dp2?K%XVx+8a7DZm@>HeS68AcoR(=XL3SEH5F2~@b~5a_uk)@YqgAqXkTfwy`Dy z;nx=tw63&01iRkV|g}HOVQdI<&&pHZYHrxfZG8K^exE)$(82tL%MHZG% zBICCU?0vOAG$Txf+?zg=20pk*&;IN(~Nr`R8a5ay{L0ym>%>jWkf>E>6=%zoZ6vi z__I(0TAL%lc26HsZc%`NQQn~AEDjxeazVfBCOB5sg7oDXkZBVO0f~7~7;p(}qE*1` zSR^S`T@Rmq&)~g(;`|c-5}bCe7*9-pjt`XM@zK6LxbTp#@NHJW8_n7{Lt{GH4z^L} zBaPI1a||!I*K96i=L}+P@t0{Hf0?oe{*u!5TCia#p88h2VuIw;>2+-vYCKmOm#(c~e4c3F z=e^(P4)GFNo1%u3<%C(^@xpERFjrLNeNSP}%evdQZ433EY#2YZYdrXI2dDC_Kt+SKLYO8fX)4 z#C?*bkw9iQJZDT^NKm667`shCyTfWE_9eYyran&H7sI|OLbADgwM>4Oig$o zKbbftpI}_}*OQWlPh{rxB=Y*gEVAuc5TmdshsfKAa&r_?E9zJ7qKDdqweZAi`s~U@ zQoHsp%@F)NEw1vkNF!(7*cV>mDVQAr~A?sxvx`(Nd0I%+H!gd z`6Btry7uwTiu%W0#Lgj&kx9N}eLUZTuF+zsMwtqBisn>LF^J&<_}~2@an^uvmHrUQO(S zWF=2%4r+(HpI$(i>IP8nfD?EB@1gobulm2pH*i|P^kN36EO}RhG?$Sms?Akx7 z;V+9;%n&``I0c^=Bya~*QmDt>OlrS`C%WNxh<9%iiR#@=)s%$IS>L{+)vs z^9ER~!GY{>E*Sp40M@qYVEH`*Y`m{S^^}W{tvpPsK4}0I`bZWI7fFa+H1&3HLZ?xY z=-+q(M=n2y0p?k#t~G)`uUm{AO9xPWLnS(B_G6NA6CQn@iF0)sbdpHK)J+EXygXDuF$+_K+>>3J>rM^z3ee=g(rm^M{ZX`(pvF;cBGi_ixgz z(@4^D0;t&CkF@ZqGW{gl6s@<$Wy^oyraza2uE}MbWHEsMCDq|)d;*H?LV&$>^xL~hrKW1#3jY(d(aR!xZ0xzUq% zh}~p;9Jx6ecZ(~dxbb>wb#fZMORJ%Ws!njXF8*dBtfhFvxkK0>xc$0&KA~~fBMcb* z3gwCa~?&u4rg*8)gyJi}W zxwAyLZ%@Y=TZIhQKpvV3uEOfLm6&tQ8E=gmhrf^H(x_~0vbyRovpV0N3~yH_Bfe|H zgd^TC-CvQ~Z}OqnuLsb_(r4+kwVJf%xhj$Skxk_9TqO$2d8=@)DEhmSX?J_TSv#AeMAIPE3%*A8?zm>X;>&b8Y28v{W;B->;jx%#O)DlR26Z{= zQc5Z!0;qqn8I_ou$%*~>Wqp3D37Ml5PZX|PrBa&(m+8JgG(G==b>gVkv`#^hiX8hz zx_=0+nJY`7w#WjaH~TVfze1Tht5U+wJ4hxRFC@}=L&X34FnPTq5{zC>fNxHFV0d&Q zVT&ISsn?Imf}OVH_@feH=QWe7KKs~O%0Z4~O}avQcDy2#btKLX$LX!LtMMzl883a< zfbaVs<6*T{d}-5yb$O{M&Yi7H_|fr9t(@Xf?eJA1zA2 zJrNeTF~JIhUT?v{nVaxz$x3V~;c@-cqsaQ7!@_E5yx%H^?KB$at<}IIT}QZyMf<3M zjtDL?vVok#b!5YNGnjBbo;!SzTHn20NoEC%Cx3F1$;6a()ce(+3Qg-PWTciP`5jWl zjNI!&WkXb;<@q>auD74FcWGt51^r?kzu#v)eY+%N>?th2+{VJ8^zk4Q`;t@$EImaB zBS=lQf)3-aOz-34u#r;%G3o;*(;`66DiypoABSJ6hhgN9AIx%F1)4rjAWW$bCUi!C zw_GxOI9LWso@p>&aLMdhJsmb}-U>&fh0NH~6fmA~2A1DC17%M_;FaSoV0eLX9DN*a zOC`hU6YlWM%ndYxonY*``7p{T3tC+xz&`6Zj0^UJ-#QxLI@b?YOnAi9N*Zz<4kyS9 zww=!NPN5@y?WW6QPm#tr!HZnBgicZVLGtSbzuvU9#CPv!60~s#Jbi8l&(Exa8~y4q z)9jtKpU|^S2n&O&_Yc8;p%L6}EeZSzHH_AQ8uFVyqm3%{%%oCra_UGcnLj(3c)p1x z4`e2Bw;tONc9bE8YY#J$*TiU4%LjTm-+&tEU1ZjLNu<33N88|?HqL#*q1MD!EIYLx zXz?}ho}2+$u_3T8=pH-^d>sYf z{mUNWuHSF*M8CjF-6hK({;vXe!4EWyzkqCHs=#-u!m`$bs1Pt87xe35$>WFga)J%LRa;A9f*4rPSWIYG9C1z? z4RZ$zAWD&kwz31DZ`J^-Z{)x{X>GwjHxA~W83Bsg;Se6Q4rUBd5-id|wBM$4ZXIdl zhp8D{vP}SOv9p4UcQq{Y@L)dg9RZRR@5uS=v5+;ZoH(Bv1CzT9=<{o#@aN?)Id%Uk z(~#58`IKv7z?>*tt9cq51H7^A;Td}H??MzWBv|n1BR%p{=z&UeQ79cC ziD{`fXkMN!j+OVr%!Y+HYI+??9^)`E;SOH0X-37{XOMB(h{Hy@xZ{H02A-9V8kY+& zaD6^XtaJk>GYjb3Y5>#6?}of9lOV)8kjdRNma*IHL-!t=LPh57r6IO)6|tGZY;#T( z^~e|Q`_4vl_ftxkp|nJSAka$Gg>$EOhaXJwJ`P{UI)TOPP4I7(g&>R)?t&cL1dcTW zE;F~nb4LPy)y!e`MpZZ&kVDpw`BiQq^rnAzPs2ER0FzV;h3++rpA8pa(bhWuleUw^CXNuI)uuyq)Yb4wvaSL*`bR5YF)rPx0 zcEBY}0+k6dAa862y0_QBqxCxAfAKRp_tb0!raX6lLvx^#6So&NJ3DPDDmu5rpIYl5zl^xX`jXHiC$ zeM@F)IZ3rx^F*q?Y0Ug8nBK-Yr$FI8BjIUf3DKM}Ea*+jOWnrY2@ zTVZuWuH3t4BpwsGSAcuK)O^WB?75wVI_r7-c;W)?De}YfCsgrPoIb_~?&YVY zO72p)mNRaeYReSof-HLlEqKJr^qF6cfn|w|C0q?-HzVR{6s9gyy(u2 z*QlGn0b{(-o|XkKgMGe7Ah#yjxZm50ljEAz&0R5@ZtB!JD`Z zAmgP2@qSw&v8|BoFp{QU4;Z0%tPZwDZ^F*|3-C+XF;t0*L$7ZSQKmp-NA`Ti zI+Z50@0aAyh1a8bPb+#C{lu!cTex@-1%A{#A$_}jXWc9y0 zvg01=u(hXv9Z0ldzuaOuo53Cv31^?$ z9%RKez0 zsr>fOhWzQcDSYX7aX#aj75`+_41PzLkc%Cb=5HFlMdLX#e6fTq-o&VR>c&V9h#JadCPy0e62?FO9VN5WZA08g1huN+kZ(zc&)KITG+>%=o?!&f;8>{an+BIs zr-IG$k)MdH$Wj_8`L#wsN8Z*GT8i zX72uxvWlY3`P6C7dOS3I9wUD(K}jJydv5bdd?a+N&N!Cfz(E6)@VbQG$}XcjUyp~F zD=2dHCEBbqz}4l`asBOUv@)@fI`srFib@Jh&Vh8YBU_bI%2mVu`Od^ZnE(IYuglA- z{zJ9CVVorq4(v^ z(o}m#`tVmHO`2LtKTi{-O3$BjPcm{jrvumMInQrg*2|Z4Qn6pf%@IFn&rUy@`n62( z3qK~+R@p={tcwIz7n3)WGC)hy8y;ES5coKgAXxe(Wa_YBmHd)COj`o}ug^exkQQu9 z34^lCe`J@PIJkFig4}I0A&Ab0rq`$7amHHkcg%yHoI;o>wgq~mgV3KuTc(f{?s&}951Esze_66k)s;Ni-asXM{s?OS8YKNG!pTJt?3Ov1bA4~+6CiCMg%xrxR z`(sD3?F}PY{lz_SxXOs#J9;||+xBqJrs{&8oE9lOxQ&xf7Dek{l2k@u>x??kM`n?0 zWPOcQ#o+Tan3y($G=(d{g*P)Gxos4zFCG-~lz+(%?eAn)lPX-7C?re9iNm?NGbE;S zE42x|Pu>0sywmgx^yHmFvVFV1#Eu~|aCJqVx8e2E*i zYtd=^E8INk7pnKE@^{;%`FWW_uf$TFf2>`ErsN}L3Av;1jV?HG=qgT5=kc1tPi!=Q zfLbpuqUBOm-X!QZZW~O)yGK%SU2!93WeuWSxjKJSL7RUqmyJinGf?Gc21Z_JK$!-4 z+zs(mB{GU6y-f$3|FL%F(Nx9nzbDFAkttJ3r6^=N=j_k3Q--1p%_vezsWhiSlp^z3 z5+YGTqanrlJUc4VC?X*wl}ZDpNrU?B@BQrP_T{*Mg%b&9+{Cky!)2Y-A+NCS3}cAhW)XM;94$ksyILk^mMHA7J^&-sxZ%@)s6 zX0;}DLr241c=tw$9a^Hrj$pe%?fMhYYM8*TR-MSsy{XTBh*4!D*8heZKi)wB>4DRq zw}5BiV;CkI2G-;1K=AG#>}Ku5dKO5Wa28E05tu!DtID*zYz0WLsy!zSWLU_4t+;6@|oP{2k(;(M*iP z>gbEX9CFBc7I~^306xd3fZxd%AocnzXxJ!Yb=8&G7ghtvZK-Ivkg76 z?=Q1y*9y3}EK$5*b{5qw{XhfT9I4M?er?{!d)znA#&f!cxcSgl(c=%#Xz>#Ro`0K& zjbEd&cbO)Y*4l`%^H<@PUs~8?egG|be{EcfAr?oB#5#5!&N-xmh6b1Uj7uaAADc?o zdhwsX>~4(zsfE6awFo*EktFl!v|-*I>i=ju#_{{88A+y?5^V^5M<0-HWjVB6)rFKE z+lGC|{jg~M3sedm!EJ4Qi7#6EacX%Bsyy9>A$(oV%j+=in4^sE_{_O$Z9NJ-A7IAq zQ~0Xq6zV?a`yKY_C^jvr?mJ@jnG}?tL%wgWreF`4NQ|#5S~1!3s?J&0!hOO^sf2lP2Wy^CHf3(6QJAzeuxq zE$$GSj*CZ2o);1{~IPYVMfESOrV)HLw|E>hDre_B{y0qKC3u1}or#r7}C@#yIxA=S$Er z9KkB>(PbSTO0cuGj$%Dav*5Kel^dj>45m1{C?z2If$O-0rO2AKD0a`@>hn!hcS{Qenk$ zAATRkVwi0_&Df|xlddfVIBEv=+fKssix=R;jvH{?vjg%^4`uUniXhPK0~}Ua4>=t} z!TuGWCpapDUTf2^qW2uOY<+?`hlg{b?K3%U;w~{ zr+pWAheivQ+wH}~pGA0ED;W>;ug2B~?s&j{JBDr9hQw$RR^0hX$4siDauEk8Ha!yi z>nx?=1*hrU2O?a!U?M&$72>XYx_CHhCaP;%peH{I@y0w6v(6vK;ZL{Xo^oFt`BIF# z8?!ONvIJYqThZ;}ZJhn12qWUdkugZY()jQAWy5p4`78%3j|Sk;=R1*`lY)UViKukL z7ME7^Q$AvboMsf)syYcK1g?j3L%Vn;PBmGiqAz~#cbQP$B_5)7lKA?+Cr!o?aMad> zTw@1GadIL^E8b@mif2IGb6aTFI{;%s{Kz$BbNG9B6DTTfg(DwVz)jn^P}jc;1lc=b zk%|%A5IisQE!YT$7TJOCy&FtG_B!!{UysQ)p4~NbsvI$H$t&NX{GG4G-4w~}J5K$B z^2MuyFOrb>DaCsI^RGN)7PsL9j~cWPIWGp*qYEgjha+& zV+p+{>K) znork0*h9SQD(RZz-x-%VBSb+%6(FQP26lglhKel-ph&#I%&Y-c`6+?U`HJ$}JN^)Z z!|pr}xK->k>?;{A*Tc-YA_vR2%kXtNp4&b}8!ko_lInw(N%@-F)a%t9X2JXddM+!O z&M@{N8@B)DnN{xM{slLw@2O*ywRfdoQ+QaBV~hAYmeaFmmzD2&{elD!?Iq2jS9^JLn47kc`p9gdE)$p01s~Qj4}1q01>dvOnfGaMKO%{<$v(z)ppUqz%p+%+0!GyM zo6N2+Cue!j-?|DFQu;QNjFcK6CJrLF;+sVJVYt{j(10EetD@z8@w`(gjT$`WeL^p$ zcd30x;T8wD45*4T^HbO@gt zW}FQ0Mui;lK7YGBW=V@UYp)Y2d*%siuHNK%I-c-4EF2n>LgD6*O(5hmGA%r7O{VJ~ z`MXz4&aR81@9lIjM?RhWIhjvyWfxPc(?{t2(6RWVNDZGV4^q!}uV|O^c&f}q&}6LznYe03mvRltpuXJ|pt#1sY-EgIn$| zqwd=lFp~un$?=zWNSFT(v9#JiMJ@z-z*(-vY^|~-pc^M<- za|jdN?LcLeJiL+@k!#6kXwMP}bo0w5zV?>PA+<_ce||XT&Qib+P68}>?}qPQ#3N}L zhw~y`uqUaR9@AKaM$Q5}=cR

~h6cm623`l7ME^{T4^0_s|r}8+6fvQnFh@n;51$ zlc4tV#N^9+W~IakDmC4f>A%@RHgLSIFJ&J5txO|rDgu~P8x8Y?>Cn+%0Vi&70NEqh zA)EW**phzG^Z5XamlcAdTR7OZo`zn#T~L|k0JV2_Lt*VUIHYC--V1`rZP9O19@|dd zSSrFyiM4QVwiWJF@gVYk$>eab2V<;WP8)CcGa`?Zw4?tdW1cCG^=J9dxE^!V&(_1f z13V|Fny*idEEewZTTcUbNt65aGPH8mGP0@76~+ZCkz(FMH@~ZuWb&U$E(VTJZ88e9 zI44F}G>utSw+x58-$v^zr=ZG_U07=E z$B&S5IOa$-YGmdhAww{Ci9QxReob!+Hqobvwc=T+Ii$fs9;F{1Vph~1A@x-=$Pd1T z@?qT+zV96ed)BRh%}aK}h9~2BR>&1*x@R)8Cw>}fl}UjYMU7;?dlO0Mc7Xu$gp5lH zfTYL%(DL*PiI-ju>&mS`aGpQcJl;;Q3PiRf_(|}?w7grt#2V2ZZAVj<+PY#?~7@=dk3{Swo&~1 zj|=YNxxilbX2eJFA@7wnqp{7CaM`_7a(xR!55@1H8?>)8%0ZFLmE;Jj^-&tlrYy!J zl$OvPh{oTH@- zck`AEH_hlaMykoc!)*-Af4u;*4ljoHKck^#@d0pMG8U8SC(<`L4J~JVktLhb0M`sRwKC zD#6c%(vbLTHTkhO0@n8C!nI*Z;GPo!4@~XAU|%BmXi9@eZ!a;3{aNm)x=6HgRtouT zcZ$U6DPSo-v-wQkA7wA@#+(y~Qlexu9mDbScV(DSn2b)|S8$?K8?D%8f#0%Mpn2N7 z^4Z!inK|b*zaY)t|xM~Bhw5a;ryP<{C0 zGY682LgCkDG3YLR#4~T3;q!SbR$!pUst&$}`0Yd4RTaIE7|X%peV3s&`mPl@Rt8>FB^It=-N1#I4Oe9;m?FKH+PWB zTSel^=ku8B*LM@yshZI8AA~glp`H4C! zW0eIJ(@sI`l;yB>jW#4GB$J)|{%^Xv3Ld;ZoI2EOz}+|YVZv%V)XRv(aX&7hc0vuN zZp_3>O^G;G0#Q+J7mXY~PtpN(NKm-se1+z1#vSh6I4apKR!?$syX| z=~N{qgFN*wASsnmB$PyypLozGn(@PxlA7=0n|k|+%c&=H`_X4~z(K_K*Ua$qTS*!{ zRUT)l28fw2Fh$D0>`m;(l>)6YCX0m+)#%%l+A69aL9((y4%aWEJ zIOdcBTCP(_-JM=ye>oMKvu@+{)5e_L)0e3KZvvO0H-^iMn97+S6>xi=n{i4JI^1^W zdl;{O6}zL2uqS&2%&JWjUeL9MMDJbj{Fwl5C}n}-f;&)^ZVxn9o}Je*jJ>l)gUwG> zW>=NSvNsZi?6BcVY@)0jTM*F&wpBc1yoGl@rpRlsYq4qQt5AYMQHC9|i~9UXkXVWif39FR)G`Dtq0xjj?3eGfFb zSD8-SF5Nwx#lgj#`LdPVvf^+q_|IN0G%S@%$lt(sSu;52k+GcbrC2V`ypTID&g0_3 zPH{VaXLCDt$8bKUOE`6s%t>Ju$29qIraS$(mE{a~Om!lsnQg=ke5t_26(#805QbxK zEGyHQwgx8IoP^=2{bcp1qcFCBgM`yPV7a6kUUq(h+tGjF=ATRm_B}*e8x})lZZg@b z^qKq^+ev~io6wIH({NozAq^;UK-u+i*gm}uR~GQ~S$-xz$F>od%s7J&EDmA##BFH% zN)zSYC(_3zqsfZ75~NXXCDCt{#>C zMT(xvG^NMqTa@Fp*rfn#FV$>>xdSCT-qzLon$2Nw{EV=+gZ`PJQ-*h+bVdvw8+> z)p2*UX~k9Z4PzOwVRHUw7YVs$FmiK z=5L~H$2`fZQGLWJ_$7JsAch3H=ur79e1_L6mRO6H!#B;VAUKl%_b1nDNM;5q2>o6paBjs=_L^I=)E3(qPaBu}SbW1juEFQ)GdspcY{ZB1uk?8j+1FFF=i z*(agTsEer3y$H9g2*n3O%(3)HHFZ*aX!R$)keOer3qQ_%B!jQKVf@!n@KCD9YCQ?= zWt#$5+CPF5%+%oAr+IO9zwEhrbynOnT`lg){(f}d_Y)_|EOsp+q$gRMSU-vR0aSbKC$e>9a?J85gm=$COLnX~K=nGv@XVnaSxKapLCQ zo6r55vx$4_ZpQun;Kc3w=FUAnu!8fSy@gv{zmt2Wv5u>^JHT~hF5$Y&)^lvgCeCiQ z2e)*NJ@==ZcNmFga=Te6&hmsbr{F(`#d8m%k5L?UTFIf)_<8)^M;p6`CScxQ3miT4 z9-ceS^T;FuX-Zf;4O~B&cF7&bfaN!+|KGdJM7zZ_>{%oYPZQwbH8J?(f--*bIEisK z6?nla1%slO;Fb{9`@6cp6 zb;}`k|AZ~9UU~xCQE`S1(Lc$mUe92E-#p6d2#&Cymq)US&F*abNjKJT-y!zIBTv@6 zbt^kN+K!FcXv+?LA<6!o{u-Xm9}S%gUywHw0&v|G4&Q7J!@brVI{UxCg0Yjii+1{) z-Nj{``a^f_{k=8Z*&D!G>S?oej3%4!&;WZ}G+4ohmk?_b2Fni&!1VcLa80ok3ZWiS zTB0B+YXc}bC6neTbErvCgL&FH;eG=B^f#5~h;Ig|e`c^} z@lp6_lLUK*cf!rOTFaq{-lv4F-Z41Qv<|0xuEzTkj#zg#3codM##@T&_)oBi%szXHD!7hk zA{DzCU-dlZ&E5=Zm^+vLu@h0b$qi(Y>@DIi_9ll+<>LmQ&DEL>Ua$Ehx^1774sb83z=f2cZbSP{3sD~ z5iz7Vtb(BnwWwEuA#v7jE4y9cM?)8mK$YCvLOO0Fz2nhI?P?}s>G~1m+u^fxOI{YO zy{ZKsKkLa!p)5{WAPFvE@{n=g5fXmKLc{G8uv`}gN<9RM*L#4aRT&I4za@H(dQfO{ zhfHOnMcmpKOjEbM$bQ0bx<8Xr?Je`L({l$7Y;>c_CVsewnT}0W4b)l`Oh3%&C}%UZ z>1ro+CU9FWIoNxctc`q4mR5csGCS<)r?Glale3WQ?K6hgx6;UK8wsJyYi%&#`o)e1 zUy;{dhB)$BH$CSu3CG@?j-4i#@wj6gDyoR7>Zo@5X0AK!@j5A3qBB) zo)w@MT?*vx18|W#1Zic_Fe{FqW#^gpyH`Y#j=2|^(fK@&Kj;{7OXIyDsDLHFn&pALawvQo6#!|3LStAu%vAfsq%Vpl zchYx?_brVSGY*m%Dy_=b|2nAr%;UK7mJc3~IfNhkZsGeKrTBP=Et+}GC2!YXr>os2 z{x4=Xp1Vv|whV)st4~P6B^h*n6IgB;q)Gj)eCdX&>%{#1GP6Flh-tR$h=glV%UPk(`7})ZMF!`n-sxR~52FXY1ym_`VUQoQuM^ zt?_(MdpgzV7Soqz3;8)WW4_O8OdLELNYh{ql*FjRkc?Qy^2|N(DfS0v6$N&E$7?9u zEYEh``3}O9qu7bdUV+OM3Xg|9hbf=y;8Y`pmMfYtZutyYoOBQz+apMo@+MFXxB-^^ zh0y5I!mkwuA<6YC=*r%OEczZ|tI~n|_(CkABcbWcUuL%RK^Q&46`JqdBBp~%#8x(( zUi0ju`k(8mnQ#&LDmy^Bel2G-xqfne)FOE8>&zDa19pkE!ppIN>_pDJYkVrJW-={qZBX zy;>KrtxuPGFk>`#_p&T!U#HH!c`=H!QhSEilP7ZbeD9&X`hLt;AIrtt-N%a+e=)^O zf@>cC9{1D?<(7WEgKa;~W4y}_^cc4iN4y`PgV`ws;Ky>0WYKs<7~AC@I~w=jC`fWnZ1_cjFwt*pGC!}v#lQ&NB_q3 zzx=HDi#W6$IF66?vr!{G4I_OXVaseeZpqo_=;or1XOd&+g4{ADPj3@UT5$x18Qy~< z)z=~OwIsXv?{~<*B*(tWvSW`IxUw@_4OyF6^6c`wk5IT$gROOI1{dkyAm;s*Zbtuz zl4m5Sz1c{*zK%g#MjF2={S=QIxeklo-okNTd4ACu2kh0+MFVv&y6Vbva`nj(y0BB2n_(p{-e0LWtjGJNL(lgMBIt%T+#n5znKXmc? z;D3jaIGx|dOt1B&-)}F*o|=3VSN`Mu@g?Hr>4jwI&dFjcS2yO1VKa8v8F9OMRk(k) zR-ER39Zu8IfcwR-9V(l0adzcaeAQZrQ5j(<>fecpwDw6){yGB|Df#f5UlVmj--aPD2vHX`Sdmu;+*zv4l7z>Q95s?H z8~q#3%nJtlRaar&Cm-@8v{bz8E`wF>esr*N8(PZ=@O7mmS99nu8aWzrEjs-5x?GB* z-5%k;6KApc+Zi;>iKg%RRmF+jnIfMn<`C~Y99FNN1@5uRpdJ2?Oe;%-$Q~hFzc`nS zmvV+T7b0Mx49}%V&mdU~Ka=17kBFhK1o<#z6pm#k;SR}UI&HcO%B-A$^SB}SwtX&s zke`CFPk++D`XFk#?kGLLGfMqT8blWE#&mqvBwDm}9*yMZ!v`~7(C3a?;-V`%h=GeC zroZ*U$GH+Xbuf_LTPQ7Dav+udzN`bCe2(t=9v|vA-~xS#ndH~PV)3PPM=(7-4|*Jo z!C?P8xMG_H_KAnVn9uTBabHQhsDp$*@`Mf7?D#qaLt->sM8!i(NtJgP`5j$GMikl5 z&?)zsYeQYcryiW7Bmb7sdZ$?Gq;{FUTYiGRy6~2MO!MZM+%YsXzLh$jam5m&Oj2Sx ziKy(6fEVMhku!f~KyG;}iL5e%xK<>;jY7$VM0K*IK8}1nIGXU63o)8IGei*@Td$IZ;rzQ>c0WAexf?!z^uc7hJ8?aD zn30*PPyZBd#Fy9o(SCakI&bMhvg#+#*sRCH6VKz8{ui`z=mbo3dn0j5Y^z}C-dQ^dK|_~+>*o};45x!K8Z zIV*>9KD_I;ebozG)$$4-E|`O-oAWRtp;dhN^L??lP6-KHG@VF`(@4Vcugt%M6Xaf{ z0TCS@2}f?Wk=k8jiD&-!^0PZPz;2lZ@UK@LzJ-2ZOskGFR{ihGZe0~%=Ep}2?)PAB zSXk3#*+Y3=%O~3E=z`mu4`WIEX`JAmhtncX;Ns2c*fiugs;xhbATr}?heOe=7VvG} z5>znhr`^ucw4rDUt@Mz9;E%jR=TA6%6)b>#rV()Z=X&V)8B6}nNg=U=U1Z`)ZHUP8 z0^2obNScuX%xl$y^LnF5ra=-Z*VZAGL91Xsdz@L~^@QqM55XUc6KIdZJL1!8!2Emu zi+s6ULIf`>#h#A~Xyv<7@yGBv=(TY?8rDXmXqtWF1yMMCiYh*D)kd?olQ7Yn zhoJ}~&{dRxUyM@m#=^a*{v`v;lU?!c>g#lgs*w46rHsbrQAV$DDSW{`QvZSXJv=^2 zf;$$%zhEPn_|1b+`^Qmfe(7;hoI)LK60!Q~C-JiB*7$CMEPh+E4i{CL^7r^#y6w#c z`tG0=H5>d*!-H7nc-vEQ((@p>Jv4=!sT&7UT}$B7{XlAX$c4;Ojuoc`3;~;5Dfnf2 zigYI)B$Jz`!3kMC*f_0->5y|FcZTPYKhvzhDQ6jd{d5{ppZbive&++v%S(yh31d3O zG8FDh$-xPkA#k{vcN2v~!=oc%@FaCRFu$TnX3|3<`OTjZEJbpq^AfSMKLYA!lVFI= zPOy=QglehH&|GQo1l1S{Q`9u7wuO#roUs6@E8t#@U zLh?lxv=Z&0HU1shJ>CRj&m9syT_Z_DY@4b1{A+Z~WgCnxorX#oomBDZP9%XbRHU~U zZO%mEdS3z7ZP3ElK%{Z6+J&1H3(6}y3P{4iQII!015Pd30e4QT!`?eN^LRX*K4(UU_yy5j6YtVM4IfnK%)?`?53yuO z89HR%#iuW_@VVwYG>E&*vl6!9fU7_LPSik2*KzdCs%NC~`7PS%bP(LS#=%XSc-~3f zLt32YkwhT0Mz@Wie$yCg*`wtvpM;W)vsFl(p9Hxrr7g;<;rY4i zzcLg5c@j0h8D!~b{_h&bgEyc1oCnq*^{gBuZ#)CzmJaZAxCd?5q9JuxGoSAX}4EhwWaWH}l#>@9Ko=0SXYB_Z5Q61se>!CR|>1pO_A%Z(S{`=)aEG&2}x)&}z4 zL``^AoCdCv`tY`QE`8(CCH~%@Lbo?(W6->vxO8s>POK!@IPMB&9^vo(;25-fR)*{I zhT+ik8)DPA%i^aN&luk|+r_%s$Hm8gE+nP0J>(Lfck=tv0Bd~@12yKEPMSMG>|F&* z-@b?OD>PW>R%N*_N^DufAlSxs!|%tx;Kr*&==@s?zgNeB{_z^vFm(bH&0Wj8)59R~ zS~sZ)^MUGJC&-0`%|tiJn{;;GX2iSlm_D6*K!#iKqv^X1`Zy7+ZHlk@v3X$y@Mqw{<5?4`y$uVf8`wp&9z&sz{qilD)v z*=5Ij9#Pu`tMSq~YgE{C1=G@UsI|*4`l!c7lP7P)dWe46QqKO>uPvQ&w9pG^=Mur()tXd|ubkSLd(k>PV8n$8o+`vYUevvuRqdlUcvvukPmnwhvHe&~gkY0fJqi8CEx0F{ZwhNa+H zyc@1vnFx~y?-TpbMEE9^1ocPxe!$*QP^GX1#+sJH*c~$5(;3oSjO7S!_sL59sQ(DF zJi4$ba{*?z8DihIC{ll52pqFHM6e;BEOJ{9cN3~1WHJZua{G9f%_T4{$$*!y<=AP% zx*?!nmi5_u6=p6=1MS`0;r8mo6;>r=OB+hY@3GCXW2h{)8`Ox8dP13D&nI5w7$VLc+tVKx|s! z&#oM3mbd`f=V!yO8THhB*&cMJD{zeKRumrFMXe>qV`Zx!wFuk@C(1Knc=Ny(u`HEnZvGCo6WW-vusc25_WTnK5OKx!*�hrR02kgsY>CH{R9 zS1-)xSu;J@Al--qKY0g-Mm4(b{(`f-ThSq;1am|7p@rTm-X&OzzO$pz>#HLAuRDx4 z?(U@vZfjs^!b!SfVlccfxDBiE1Wbu}1d}hMz}{_DVEUyJMjZXpahMl?!(zSYp~KR0AK9lzdO8UqX!MZ$Yg$YpZ!iVPUfEoN>e0{O_&^0@c__0w&zm4Xy8;Fd)PZDVB*^WY0`5z_$(HU0 zGRyfFImZqMm%%c!KC78Lj+Ox)BMuoP5msteL77|u7_UzT?}tfXE8PwG+bq}}6II!L zL6YoXyA*30n+0P=XTVtX1K?+~idc=*0G$T_wvTn_5n=@zv3w*}#}lcsHM)Ft!k(|q zl-Pcui$68ef~w~{19>b|+O?A#@}cly({`v*3xLuc889U^7Xlq0Lu5uN)Obh0C^>@JR-vsWYNZD7ycX{hi_(v;i{`@s3Mt$cawQ;M$#eFe=LWC^1sQXg3$7$ z1<_>ri|tfZ;7<p| z-zd?Mxsq6O@=W=xieAbs493crDhEkp`z;cFA(~V^4Fb)f=b>tK zDrg@)02@g-nEcHEpZYmq+29V@JHudz>lO$Z$G-<}{UoLDtbtYM=MNjQ(CF+OH2oZl z&pz1W{QmXmbD<2y8>&#{c`33G^bx=_GVU>Pf#|+{ZA^Xi+L?$A!~(< zLkxWwxd^Q%NaE)$$&|U~Nv>r_6X)UTFtczQynS{8bVaA&<oRKn~iT1m4my)#a3GU44aT% zHd4iJGpd;6?tGMB8*s>?TD*I%8q2TUplW3S%!x%)iK@;@INrkN$@qH5PTNteo_aj! z9V~#*?&&_j}VacM)r8i`y*Z7HXR4PY(yWy5j>S> zgX`0en{B?~cDA)WF?Yht_kE_E=Kq(-m5)5#%QP{#kZ=*Gf& zT6X&z9hEr?3u{>V;o1gztoJwlv{jarJH$oki$X}twwp}LctXcS-$TNM9?T5T=C&2yP1o*AE3wlEHAWvNeTJ|Z>8~F=x!em!; z&U%Hhc8@VEpbfG15FT>vz*(CMan%_)Zd$}=oR#K>`#f)8!HXNTV3#qvW!R#8{Te*0 zaR6t`Sbz#&Z1CgCkJRD4F3OkOqDq<{Y0zv((teX?ALpBqM;Us&>q?4Dl#v7zOarC7 z6_EL0D4eS}LM|WvLxu;-!ktCoFL8wduY~ zaDN_A>k|-W%q@}i%*k-=!qf8J*={&+;WO>1FQPY$Jh1};@!R!dXis-xU*a4lj`vDb zZA&3{Hn@TTKbLTGZacGt=RMQ==SYCH1isW&$Da|B=y^XAxAFI>+>cYJxjO>gPUVx) zE?oKZStjIQ)ku)o=LFxb7r@KXAX<3C0D67LQ*+aBdPw6meja(8zP;j#&7;qt+2iAM z&A7Q(&=7=5?~dU`{Z+gh=pKzbt0q|55>5V&SqZ(1caWLhE5ZELaVQvB2%ocdL7a&Q zJl6Wbt+N^MIQl9{HroLvHuBK6q=dSElf>eBDeV5Cf&()S(p|QbX>fTr)sGLt0^aXa zld6DYE(BnZ#75K{$x`iFAquKv5E^{Z)qvsi%>{H-?r3qsq%a!UITdf03>UY`+~Xa! zVtUg>nM$wPT0W(8Gv+4Er@37#NZ^oqB4r^`*`%VjYI_@P0fFlk2NouV-aa$-wM*b}&ABW`7kV2OJQM@CnGnvI`Kp2s=(!?Bo zMz!MZ1uQf!!SuvRT-4l-GB1yy^^7#cZ6opKi4v6YnTIC~cH!M6;ppe{D7)3WlJ=JLOE%_1==pPV9={;5h2)?B7{(a@w|1Zb4nr0=K_CGrQ|DV@yHH|Sd{~sK`HTj4@StrZFz5nW?hanGnfA&T5Pjj!~+pTvZfwPqn|$$CBKhp$Zrdk$cK(Xi}~6@K})i5c;+qPnb0`VHm$p&AuWZXYdP(rmc6}#GsnMMJ?for zRXQzOWIC!qWY=3Q8ZOKg-Q1+eqy_mgZ$C{G)w*93t+@M3^wU#?7=2eH1^1u}_aO?)x7wIuq2%A*wDsxk^iT)RVQ^D7q|EIw2wQFFh}+3pgW+v!|l| zgf`Lf*Rq0bsZXqi=RXv-?mc9sl)ggH6zM0F+rCxs`Io~+VgFURPs|s@9Zrm zBrih5I*(#bYHwtchFuaRX{{8M_#PBJ{~9cu<0LJro#P?u*~!A$}q;`xCyiAt3NaD%3Q%ArcQ7+^NHY|Las>Ue???zq0Nk*D<_|MnUqv@q)$AX9`!XJumDSQY?J=>ZG8fJ<-bOwZ2u` zu%$xF6G=js#Lq&7>4w5ZvttCeQ#6D@2V;af?c;?^)_uX}?S_KZCyG|pQg!RVzZePc*8vUc_D2XZ~$JE>z$A*3!EBnPu%p3E``sV+1+v!-cgw z%tT!o(xTvscS48LmjnmfLag3-Xj)q9rwRLS{S_3BS|Aj?_ZQHuQw87d^jmFDOcOK@ zuC}aZhFi%s$B43x-ijXW8_QH{ZM3>@s7=r$aau6JzQA&1-54v)<`)9l`DcV%vbPG) zO?fStb#am?ZGp2W_g0!n@C!x9>W492fv(I3D^0PCrHZ3rs}Y z*Uk{7T;SiM7IQ_j8+92SNoA&fiz&0MOPjeG*(eNYsI)Q)G!oT2e-Nt4d=dVrTrT?5 zHdR=w*DbmqQO9`5E))6nq=<%2O%a7fMKO(9yO>`U^^9%!EMeleqoPEcE>XX58p*TZ zTYiSyNzkunqUTZhqOj{nMDdpA1^xvIqCfXSnL{r$NcoMA%-EJn#uF!zH&?3|`!ynF zi{o^m#(gJYS4OnZeXXZ(Zi1ZXRRXM2(4o@!xW(JjWoyMZ~%rZTN(&WK)p8YN1(rNuA}RmU&!1{8O<1t)@L*w{)?jPj>qc%;$|{J+1VM{^Lg$$=T@Y$ zX%Px(CrL=`v{BvK=>wfO%ocB2IV@?k? zyO48H3eBIwlIQt8ta`E*zgu>qMXw0$6tX76EiB1-yvHyPX~6m;Ou(KeVB~q4o^USF z%)M!pg4e-rV=HXG>4=MDM&aM{CM?nCBk$(Jq?YMV)$>*9gI63Ry<7oKQIl~GMAM8l zK@?MyK#N?pu;`Qm7Phn3F=MzKqH15$ zvx+=Q?a3q?DM3d3d_U716~fH#eTQk=&S2LQbClktOm?k@a3pFm?!JEuJ&Ou3O+gm* zEEa*wj24T)zq%IN&U9Nkv^#M)+S|+nRc5eVe}~9oZ91|3Qf#4I6Q$qrrRs%3;8r`# zg!UAHQTk58!wX4xw*-~H2jZ^~CZRwHG`UrRbwYsYQyDlTDMqG~Ceb2=No3#(7~04{ zbC4>NVz?dG4<_SL$x=L0v=9rs4PZ#g3=DfDvHZk0cvW&81kV(a^pS2VD4oQ9IVc26 zLSAr<)4wsfcLmr?jRO>EH;IHM2ce7LU+~V{1NQ~|uzpnlO26BS-wOxm)csQGxF1QX z23;?5*=V3@{2JZYi6@MM~z#ZHN@USo$Zt7aYU9C!hk|0o;dIU@w zx6xH^4%H3_l4i^vIv-|6-?jhXw%Ol!VWqiXS(*$>Y8R95X??Q3@e)7f^?{SpJ4lmJ zf-NpLX&$E0U9LVUOMOAvkEgNfgbxaB+)t$|H;|oxK3&|Dj(&kra68NmM4i^7z^f^E zt5N`cB5dg3epOO^^$7=u!|C}!H|q6fC~DSnTJ&;0?NipGm3DXFlkExE)gp{f*IlGR z#Uv{AGon@2Gr`cSkt?+HHKP;9N8era$+Bw($?Ts4XJ6!UZ>{~w)wnqcKhD1ivi-5J zS(qU{-#=)oABm@mx56x$MNH)3SVr>6UR3rK#nCTZT5v_B# zr^KnjG=4IL!hbEJ4+EbuTVWiZ-_A#~;WV6Awt%`E*OSyFf9folgr$FbAh%`_1}qh# z_3>gfJJy39Jv;^>s)evBPaf@uMCfvgJVpM|q7ps>3?C3dK8I9P3{qx$<|weH#+ob& z=tA+PdRTQ~3mPrV#ya6+s2=wQHE-Ug8=`Ni_TWEqKX`yPm}io(YXPN=Z3V%zj$oVq z4DR{0(M9zhYN==sq<8Mgm)37atzh2QM1 zqH0McR^0Bl_%B}+DxaN)7hh`7bAJmS%pS*pWK;Hsz6Eqr^I0E zvRarIHx=b__MwIJIZ*%Sjg5L;sIaUYEiRwOFJW)c_DB;RNmM5B>J2pVHx;!&;sjqG{)nB87I$_d?af1t^{en- zO)xeeJBg3Q?l6Dmowgi(VsFuux0(}|mSd8i*I|0zK8*x}E$9{g!1$yLx>K%BFKWJ` z!e|Hca83$*>`jI*)vg#l@Cx)VRzq6KJ!p?v2`k#&z{1}R*8SPUSnmFhYsVDeGt9+| zs^e&+*2qZ9US?#xR6rut1npltV5m|iURqQM4j%#88@8k&mj*62{P1#?4gS8-MeBZw zvLAd6S^EMvUX)cLY+u}fOICNGlF%buw|FN_I&Fo^Z1R9lypq|!$p%}Lg0N|3JF3(c z;7V^TY%3PU!)~Rh_2eVKT}{}Vk-9q*Dmj(J?fI$0q6Z~BRq*!0WAOZ8EsQ)`&5X-bab~9Ya98)IQ*B>3 zEs%95g?+W;vv4+rJe8xIyBz#;s0Kn@Z!poPUV`FE!c#{Fu+2vm+Q-~sdzU3VXyw9K z$ra%HatDw2&ZgIHr*YSTB9QnHirYgvpnOgwcu8o%SI4W&ie_(y`8gE>Cg*|G@hc49 zST&@C4npCzn_y`@i9FNi;k6HbAnh|BvaE=St6s|NR7M(kREtR++wtK3nXp?&AKd$< z<6DO(urO;Us>ff!mBHp{p7934#>NX<186Ishi=x0nWan6E?5&PgQd~W zSb!EE%E#oSA5d)G$n2Uo!M#!M#I<%onC-7Z+AqY&U~UoasrbeS2TQ^8k?GLAWiD*} zyc}-7>0?GK+>ked54|}b8FP7JM(xfq_hi(eeta#i-E4v|WyufvJ8wx_)8#Z)ydmINO*kXiF%@_9BuA^*K%x*iUJ&XxHXX#SP%9KBDkq}xdT z-$v5x4WQ!*e$>#MK)2bwBp5l5&iB|-A6o{P>xKNZL%k5E^sm5FKNoUcAx~=ycA*< zy`;lQCfTugnh`q;PnCLLNPQT5H~yqy>mHiXD?*EqBlY2RFa`bV=sj0y{$iOWw+C) zL=wGxUQ03}1?2KKiPR{H>Prri5#J%w(u&7r&*ibHSsZscM3cX89<_xwQq8RCSi^6G zPs_DYKq`eJB5KVrR)?VLmIyav<9&|ji5v#3Y{AufFOzw-nprhPA9APZb3eQ5z{lwt zu7TK;cT{9b^o6-!YlXeN0T$Wy)46sp=BN^`4M(w@n@ zRdt1G%=mX1Ze!~lOZ}N%?4f5aZ0t%7`@Ykju{E4&HXv+lb~q!N{91R=Q5k<)v~vO{ ziACcLmt?#%_kk{m8}{!_^xi9Or94Eq-r5ZRxq!m-BSW6HA`r&9)e)vI$=^ z*v1tG?D-rmcItL{_6mOuoL>GCJd5gJ%K3|w>0M5b^7E)`ni6?j)}X2uK{}%tLDhSZ zLVtDO^7e1o({cyzEq{TFO*oWdq)h_Ho?}}5Oe|2}#w=HFf{nr^v`;Jz=Sv;Hkbyst zIB*sYEIa{cZI6-srS+s2w2SyJtE0-MX<%v9$t)YqVCT))%4VFH#YTPda(g7$-%@y!{7uYQIL=u>|IGMML!P3LUR9Hs63(r zwL-kQ{4a$Tjpb1h*8q3k`w6A>4RBoXG!46*q%-Bk)LGw(~%3eGU@kcy-YvhnoJ zC5@CHX3(3*mh|b!Fzz(2#zrjzlzAwN2VE6ObHz-0|Gqb^q%jncCS(^K0 zItl#u7;UoN!i?X?!PfHw^tvXJPuxM8j$yQBwh=}RiQ_#(akLAMBkRUf6qJ%fF4H@b9Vf7_nw9N#`h#G7qPk z%5Fuw`C6#4+XQ7LHqkgrMHkgw zKSnQBKVmY@8!|n%T&`kEF`j0sF??qt>hJlEh4sDYw&N|f6vpDJh640m@D6jj)nKrw zk7MlojT2YqW%fnuo5?2KWb=*RX0S=6I;@g|9((5DJ-Rq_m_l#I(!)QsT#;Q%xEI`# z7>duPk5_VtZQ*527!*Jvw+ZCNT+vA(nsh5qlfb=;bnx319Gy0eKCh7>|M4>HwG$=D z1WDQzvJVW`U5D<6b5XWp30;b}rqK;jH2JEHH+q)_QYVay{;{Hcgx*@ZAuFq$^ZPN9?T zZj>44%8n0iXHTi6vliN7R3s}z7Frr~dYdQXd$5cv@7TiaUwsq}N>os_V;+iaizCDC zbh>7Jl2%GDLmg>ze7af@pRJR^0nyozJ|PA%4^DCZes5|NHPw;@XN3@MqN7&cgk|1KPMcNV|md%LK_&3SsmEqU5J4Vm(Z(94frmX!Q0tN z7-|Xhe0&nAo##@Mej+XyDa1RIU!u%AThiuh!9BXWu^=Xy4oIJ&+iS|{c;!`m{GUGt zKK=#GqoYv!?H~vitb;Fh#=PN%Kq^|}MkoD_V5Fb`X1@={|K`MEs_P~+N}PqycoM8g zy)nghx{*dq1QYW$p7C>j2=5GqX>a%k+{ynK&DO2ODI;@Hy2lm|c|1hLH52I7U5iIo zD^R}GR9YoNwBB79BWn2ZfckViIR7pV3@hOGECw~!xZ#X!2h{no47;atX|J&pSx)z$ zScmhF{Pa9{HB5)D#5pK(>MfjY7XxS44)UA(m^@fv-dx94@Klh4x(E9}N$wiXuL(g- zon+i*u9i9gLW zxLcW)RzJsSVbAbq;yM(|aYT`4i}1?7Fv?xGj%HUaB{zxXcw6iXJouCi&-P~0tS%c$ znB0qHA^Y&=STP#ZUq=CZS6V&cMcmezbayNe4lS9+G>_kA0(g6!UT?d|+9Y0Nbps;E z<4-Wjjm{<&_t!XXZGhPN3pVVV#}3sRvFW`!Y`6Xwy4f^FO2JduOdjTlo~(|`P91=v z8(EYz6hg)v9~zw;M#nd7rfk<3`e*POA5REVWHCdXD|p$AWpcbdmvXZrjwsPGS#7e9 zL{hJ51kvQL43=gRk80$HU@La@fY|7sOf2_Ol4|@`gx;pX1!=qF;cz`6QXH!@%ht=rQW3Sy&VS_|w zgX|(D$cb;{2AsP`a$?_TyMZj5{_QVRoX>^%w?Bc^!+z@PdO#{=FX(BxIDLFCLx$Ui z@xF8>2^aKVRRjr;=Ju>-+&nz)aa1b zH)CqoQNW=MCOA>ri}R02@b(#xLS3K^-pdf75$h`4lJ^chbX2k9c_r+KvB#aR8Fci- zD%z)QN0WKyHkD6+eh8jMR?z_C5_Z7J{*R30yJ>8Ss5tBVeVAemlkw`WGMuXKN)M@u z{@Z+uY@4poKK~=MyeO4|{_Um5M)~CX`3l{$$ftyNC6r%rf^G_2;brydV%9@X$aSh= ze4Vn%*7qT)Sc|i@7t@)`i3hnC6kRQie$N0)y>X}z--Js0!;EW^LMOfcDDGM(Hn4|%l<8Pl^*VQ<=Q`dM^~ zGJj8E-Pa00*bfuv+!=$U?`{HkRWAcW?}2PO{g%o0oRxv#(XHlp0Yl8o{<9u z>&3C*_;Y;mCI{F4cMS|a{^GvWjAFD#HQ=h0F~&z2qk7{~tbP~Aw5wM$GRrg|?V12K zgzCZI>0?D{pm?WEVcAv9w=wA;k%b>pn8t%a1dCqvEVquryoQ zF3xKEj$-2(S=v%Gm9CghWvd$B)2*#%>66z8HvYVhu?gk)ca;WB(Y2-NDH(~( z<71wNBr<1jZX@BNrYM}M2O5fJ@x(?h8Y#ShR#QIuc_$Tz&dtU_`JLz`;>)OXWiU^A zudsga^H^@$0hZKnVb1&$C@`3dB0nFZhnX|jx4vYS_-C^XRRt_xP%dl!%$ALBU&vlc zTF6@RIL+?Tc^I9#4^#q5Xtr@9WoO@_J93U#Fj$9+65e9o^g8O&+(Ij={~>=s3v2nL zi=8^7o;B;xBH_p8G)QwP=|2lhDJ_JauSF0fTFc|DHOSZ5f;zA7VVT3p?8KgfY=Lw) z8rEG#z6)ij`qLDTNl0)$JBBc6@l6mhBM=u0oAGosKANqoj@zdQz?k(+nv<_Vwqb6x zHAkNOM^s5@YzZw@pU2~(dO^5a936$6@oUc%sxCL7j+RgqnDhfaYZWu$<`=14{2d*e zBFHX!5l^0Vrzr7IIxUf!L#Ir)kag<@Dw?Lnt{0MHRq|icqRt-tzWG0VblDDdz8g|! z%vzGR$|aBeAryZhjJAb2Qkv5mk`BzEqS{1y6dr>wJVfD@pA|E2WeOb%xk-Gv)s!jP z4|n62bBPQhBqGjwCq71~>V zmsI>_!|t&Ca6YgU`sVXfP32DP{e0NF$ zb(bQ(6$}OAn7`bT?HSzGuqCWV{Q~yYHhp%qP?_1`Yi)j3HrM>C_+GjgxQ}S_3R*U@ ziMpCL(!S)Slq4|<9be^PiheGXlx>ATMPakk4|0v0s&`u6d6r<|c9Y*?EYgSCqIq0# z)OCvI7U1sTpK#`72p+T-$E&1``dienWEY1#tW$}uT%%c$VVHAE6K%St;&L}{O0Ey1 zFBipWVaF$YJ9Hey3sYh5bTJB_K+2ETA-aD6y~YRObL=B#YyW4=zmkWZi@Sk)$pD-W zg)p81+nM)e{xrK+oHoXVBiBy=odO$R{c&HgJp7l$R$nEhJNdNp%vomF!xNAdG#_QR zZ^d($R8dc}9xlmmrl-5}NayP<;xr$?_KV$UwpW(q9IKh0>g&uNlR|K|T}7Kp;;Fc4 zC8^|@(VhoLGld1{a-ccA%@UzLmWyrr%It+gUzUvyWCOLDxjFfLOzx#K5boE)35{FC z=;#hG-z|r5(*te#L)x@?M3arXtj=;QXRwQEUNL*z4uRVHE|?TzfD3OOgWO(SIKkUQ zlsV-Rv)+0xM7s)*PE|Z!Ut|JN>V~{?pom8+if~O;ChD(G#w%~~v7+=R6FsntX*1Zt zl}ow=;r-v?sJ9|kb@f4ZgB674Y+$5JmZ53$er)W1j#Um?RGBo7KKU=EBn=zfdbSUK zzMYCs^+xev+zcAI_yOOox5LEKSK**?6dW_M$9DnWA^!J47T>xn;5YEm@X=dDHM zge551BacoNDaikEIhvayMk`v9vhog6`MR1m$+B?q%3-+j?j8v9zhzE)a2G(0c& zlR2_s2V?vC5#w_95&qbpf+i-sn(Hq~;Ua+)!Ru!R7u{mo^dguDaem+?`{UFoOZ+=& z5wcf0K;>^1L|M_WXi%scunJrn*8)OoWR5Lr$ zM;QU%3wVe}^i=Mw#Cd{|xZ0nCyST4#_r1q3Xp@Jnf@(Bt`!ifT`VF0(zhQhd4>z=# zpuE^6+`6U-o;@jt6rT2@Q!s#{Z4Pw$j0C;jz5-j+-O<=f6^*i_=~RsqH3SFI9DP;% zxAr7BZ#e?-X0PeVZ*_L&9tYM;VF?V5`N8@lr(l|#GjjFIFlOIz+|1a}&|d{wT4P7i z%1!v}`4BYikwt1tVLVdWnbv?-W{RIMW{GLyfwvVX(YKba2NIdd4B)$LLli#BPifKf zDQMvV3c*h5pRdN+C)i`P%~I^`o`Dm)^Pv(Qt#4vy z#u?BF)`I=cgE%$hDb{w^;gui4sOB;aQWZwH8gun%Q?eRe-zG?9E1H?F@sB~KaSZlV z%p-qB0aJcW!uoMBNHF}zHI`Du1EZ(tf{-9huPVgnCoh5emG4;g-H~nxXyEYsT%_J2 zoVfHA$32tr;&gd5n&(Q9=7!YbI)>e1A23B@5ngvv#!NdER5p{teFXwILsy>qy@fHf zY&j0s<#Wy!hQPlPY1|adhw5jCp=kAdJR_0=ADV^Gd1E(Jv?rogKq@AMdEf!XTc9Jf z9JzC%aDm4h*soj!Z)etmR%$*r+#5jQt%9^n)d!#cBaBKNg_f)&CQ6D69}GP)vui0X zX-UJE2d?6OPzv^Hg|Ia@7Vd9BQr)gX`MZ6|GV2EOLF_D3YM9EbdbJAAcgDby+fta( z90*}+)bNh87fL3~!s*MMU@EWvgzfKPZiW^Zh>5E#2)x$wsV%ZDk|bYiHP z%7Xjo^1kDgb-J7`rrf2fIlLVs1-s#wjvpM`o=azgj?wHng`}B#9mOjj;)bwp4DDS< zoLkFDE7_maz3TB!{R6yr?gENNox~8!D_EIx9bFwc80NVU=L(O*x;J6ea3+*|o_f(K zUtjF;i$;a13wg-giq41Fljt-7T6Ah3ULA=+neTSP8#k67Lz|u>7%?!N<`z#Ocx*~9R5q~225Z^q9a~xZ;87ga{*7^`Kl8ZJ z?~LBgSj;uBT)QYEJkF%!EGjAP=3h}HX7$c>0vpni!Q=r6;EJ=##NA-6aj}kMA%0K z{A|6XARA|t2CwEWfm8RcaGRI>!tn9e7__<%Z~NTF`x`6KW2_EkHQSl_*MBh~COctc z#9gkt{4GnbB6Y6R=i8>1N4hKz^DA<$Iv$`mztc#u;5aSWzMYI5fZP&8N$2))ru%L? z_n)*Q6Or>0TPpZy^B!H&uZX}cKQCYruNTPolBZO=b#%ab10D5wiux(yG_+QYLN6{t z|J}iOH2xr-yd;XY-p060B@6${lAzV+{$Y9T3v3gfiLINZ@cv3QTsCC^sao!(sm+&2 zaCaH+-N(Xz(jA}%9Cme`3;WVFg(1w00)1#pn^;h`aaA@G1XQa4GX3Vzticb z{WaRK?+o5KbO2vppF?NY9bm$R4l&vLazXQI5hjNS^SBCLkJYu9{+)Y>DR+52$1fL} z4x35gg9l-26NNa)uo{nNv602;82l~~o!{hQzf`y7;`r$h*wM%w|DJ#e_WLls_YJyi ztD$CxM-;M9kTthH#LRs=7c6wrpw*n$TR7|C`>oq?zQ;eNOm`N{OdN&w&GmR&;w(x< zgdpF^BzSi21UROO;}J~{Ys`EjkW|6>PD7~%qrzr0i z_@V6`o-XV__hEPBx|-s*v_oi=a~JK41n7@4A3YmeicRvKc<6cqDl8E~#m-x>{#q-@ zN56ys%gtE*r52r^rhrSM0G78X;iU~HQQyTM*Kc>hm&r@AJX6>R?_iN??F!IFLP z*mmv~n#|3{m-A2JNj!lz(yJi&b|~~jorT%%;dJffNwP}g<3L@EcSL!)^%R10fC+AmRV#QcZcSqBp>AB%smJoZP} zvi3z1d+2QjEAjgvJMrQ*`8U6$)~$l_pU9I&hu94ZR?y5-1>;eUkI%PZ&#jIchPG0d7m?T>x><1 zEU!oh)|!z*v<59&FM<~aPC~xpX|Uw&MLcIIL+TYDIGoX$t&z++s z?;6Q$s~eNIER4G#&x4trxsM7D?Wajm`{+ib1)Mpk3@yr^8I6SfbmLMcIUPJteGuFIa zxBswW;sf3iyNZKChZ(*w8D>aN0&Fw*AT|XV>o01YxuQO#qPT)Gd(0@MX%j3zIu)nR zmB&<8kR+@pP$gp~O-jzepR;l>yyhtOxpT^Ce6tcPil*J;^*^_26cj#h~R3J>e> z&a@1=Y}X`-1U)hgSit*TgmCwC5}B$>x7<_7%PNr?md8@J?p!*qV^0a`YpLHr1=w9P>@)marvHVA>$5gAZ(n}+k(DP!sL1@K;~6KyL6>9nE(EmnPr z6#f^_+~=cpQ#o)t$bq9$_`+P|MjI#LUNWbq^o?b~s0eHGax%MOj|6+O>MQMycucFB zUeNSe&$*eOxQti%EiR*5L?X#&N&RjGjVy8}*-9?;Trwe}J=*lyQ-N0f<5KD7A=LEm zLaXw6+_sC4vK*u-NSvQG1pk3JolT$}%!0aw9F;vz#)OO*bc?db4H=X1^1MmdJL3V) zO#Fs4WgHJ(eE=_`&T%6rhPk~`v22kSZy%^aIGYiD3aWdH!1qoey!X+^lx=yiP45$2 zuZbcp#W+fiN#S{I-N?1ZoaUdOO_A-i+nts+9lAEhNYc^}qJlG9-Q0%`CP>cR7O z22jr?9pi2|v6X*kvnx0pwkk3kesN0RYfB1LUlhi+K?YpjN!Tvzn$D~+mK0+5&29BfTcX1-!)B*;U9~oPOV+km>5AL z$DhOCq7wMdBTR^tB3OF#)_|0R>oLD9mYR$Yko(=^ zv}Wx`j4ybD;@a2HUtW(uL5?D7MTluoA+L*bDCL4Z9pAehBPTxqgXbAAX?!8@s4fV3 z;s6!`d+ErkIaG7okZuo8!Iz2KpmzQdu%#0)IqELF6KIFh86!-&xg=x8$6|gxP1wR9BFgtsX(nYZnnwHc+IZK1Uyv2=?YF4vs zmS2v2t9BwYsEk?1PoUO=N(^u7L)D2Q+<7?$n>H=Oh4bctwQDSQfgnFuIj4zJV;^Jr zU{{>wz0W_W{X`E%-)kmC-~Dv|r7t~gjH1C7T@?6P4T-)FnTXu81RXcYfbF0a*L`r= za3H!@$Kab6bvU>s2W$L~V~U3&)3WOYWA^eo6kW_kF}rhk@ZLH6=uWU+_$TNVMWgm( zNgDiSLA#Z5s7GrRDy~bwNb#q*XptbrG?@}3w~lyBGoG?F!kok`JpPoQRVbUzzSeMM ze<`g;Qz0XKb3qIr&l$rF7UEQJb2)|h|E8jVpETGb!g|+vk@7|}n*4h;S)MzB-xsaK z1ThU9trw@ZQh6#mHilCol4-e&HCY6RlYmGmmK{8WefD)Ikh_T8RkMUWH15p)%`v8# z4*Imn#*{R}yWn3|Hk@i~Vy1Lmp_$#w>Ej9|3jHU_29&&|(zTs*%lIVLeqM!-{~yJ) z+8>JJ0^neeKDss>!}B8{X!2D6vv!H%tYQJoUEPibO8k^NUV_7q_n{x!tS{;c*a?b zsya_$-PQ8HIB~jf>61r$KpJ% zL>X2)2BIlXTlu)^EUpi%$ECa)C6N9SeF~~jBX=84dVd@qh-jg=3eV%x)CC{EH-N0> zL-;C)RHIKE~oMcwiyaCJbLSC7F*n-4xoW)elYkn*i8?6b?oU-SRKb&-7Ld&_ln7j?(z zo@iV%bQzxtdt?8NADlCym28j0F?O4K1p6P$4@(&jm})47_NF8Fd+SGxiQa}C-u)!d z@R;<+|I+5taN@Wfq02pYNd9dTdOJ+T-KVF*@0IJxcKJohifRlbpURsih$>D5i+y2!#_iF zaMkYlto+xlthP=xdrQX=7L~1pui15QtxgI5Q#lOMp~IZa!J1WfaSUme_7p$)60NZG`Ux+5ycq$Zb?-Xzj~TP12#8{qL$moQ$m9gEXgY*0VN zOx$yzgr>bzkdsZ1`1Zm`ngKlep@CLUl2BPS84?Aqfw}HJCThbx=+)1L6*1kIaO)-7 zJ`zFKe>I@eF-hOpLo23TyjflJAsVcEkZnxX4Nm?%Xm_r)PnyhRjLp@;iOw~cQ4kC6TxE18o1o~2hqiPZQ)g=C#)QpPi7YT}sEf_^0u-Xuq*Uy@KSAPx;5=3%YM zUhfeH#fEHV+zeJpQiA165oS|F1X=#r%OEmC z34i%c!Uv}h(6HP-5^)Ts-ighu^(=2T?aO9X{jm|9EK;L|q(OBWAK59B44F7VoTA9frohQr+<(C{Q0EcZ*XnmSY2Exx==`?I|K;F&$_uEe)&hL0HeNk~vh zBOmb{K89J-46tRVHO6EtL+u&iXxi_N>!R{d)no~#?2N^bn(4$2^6*TBG(~3YBlEqn zw8^-H2NC+Xl8Z)|o%~76Sc51zZ8aqRU}G97&BVc1>Ubnu1@}}9(md5?^v&`f8O3d; z=wTlB79T~r3I+7+dp7Zh9;Yiu?;|YlLp4PiQt_AOaSiK1?%PN3mhA9!~ zQPocFJ~b41h_`WzTaE+9_2|u;O)M&@;%NvqxTCZMH8%gFrNa~C+9tr_jzBE=F%4^Z zdY0$H7P^^qn<_5$kw|n3iZvg@V~RXK?&AaWts;r08|0F6j69XUn?#-aRH-@N2*dB6 zgW=ISkOept~}%#7MT;QricM8*B(*nh17vhF{FZu#%%X88Fr-aGbN@CYCgLdX`U*+ z8@)wSy{l=U;5;Q<)@Q%IQf0pvX|da0x-xe2UvfU3XyCYI9ixWOgY@V1J_;#V08OkS ztoy5pq8rmOXddsba`uPcpYO0>)X09xEM@hMsL~uIQCctk2iuara0*;yIOpnLGm8zj zQJ};|>bn<7HHx|H@33^%+xZBqz-NQ$I&1Mpy(?b3Ubs5y{0wE*AAKc}d$tb;? zA`hP2F(>E^zI_4um@;65?(m9)^ zd{{|AN9|}#-UZ{AJ7VkM4los3PWR8S^rjqW|JQ>QnYolo)6M9qrYp+*GR0>w3A;-_ zazFJCS|;8L;c!z+>9<`f&0f8o=2qsx+LM#;!%sV0rndrD#B9Q7lPdiEFBKh%GVw-h z9LoRZed9|7@Ly>TxAa33Jzv#Km9sji_X}?}sM8DiF3v>f&jR%PZW&&yK7xyI2DN-s zq0|&fQvFg#AKsj$#od|oPk%p+_$AT9Di4~Iu@mI;CqYVNs=35y5Ajy6lhBKNlAhQO z)8=Y3zt_*UXxwE+x(u zgD2#1@%xpS%hSCLypN;99d+ujmf-PfCtyj!eOUQJm~AfkL3TR7NPOD>tg?6oktwHe zI7FKLE&P$bvVyE8JB*GklC&&Nj?&_$Q1j7zbXqM&uO&@b=?WLtDSbJs;xb0{Z^hVB zO9j?ynFE%0q(V?k7i{K}p-)+p>7e}|9PV&oU!+;FYN7L3xxI7flIC+fe4zk$jDJJtyIith9Os)C3WPkG{3BDDjf?`X$-jzk- zbt`cjZ_c7E!U`3PWsrR=g=eIE(aYZfy?l1VB3nK@zf}kwTGgQKmM=;jevEs)f8b&j zWvaTF3MJDC2KBAcQfz_|(lBN=nM%R(HLiHI>IDd&`3bkye1}tODtY%vE&P_+iO2iT zA>YGsd=Y+~aeP?Hwx37hxbueR+Bb-2qI3m90lwbpsLYCTz*&y17A4c%f>i} zZ#&D(Eq8&7Jk3}vSDqg8W~RjEZK7R4ku-ikfcmC|Bf9Z?Tw!sjJSa%-HptRiaS@uc zE(|O`w}a1LNi1^TM_jQ43YdJ91Z#PEWu^vxzRJ^(zlq`Xg`ptx!Wyrn+v2pvCvZ4! z0w?DA<31s46ewMW%4s6hUMWRi0mKL3z&w$m$-Yh9^Y@=kIB{x zvF|@^+?DQxR=0zoGC>OVRE;rX-D{bthZ4AT_vUb)IZiWOWoK+I>3-AV)Y2RH^^YKx zKUqj8{f}@h@00yBpU9{L9qQQ0m~YX!7E`O3>edZq6;@5>aZE=dB(xc zqPf`iWB`k`Z$R$d4bTx&2Z5`8z^74HY^%M3w=b{9NwW;_zwmHu?RbNM1%4#`m*<&Y znau1=9AQqYvoQMGoNNvWlgr|IbekH8Ht~0nyGn{4O^(BNXNdUx3+T(oJSYu)09Qt2 zaE97$9J4J)zM8pYBN%{M+f&i!c00bMe{^%h2sXp^qURtn?s`j=P}xEX^mRI?Ktwbof& zRDESx86wY^+LWNKT^LqunvV6YH}SToF3tK8OP=2c7$s{z_)>5Nc7B+Jmd46>*KIBy z9$mw%DedFdKB?jcamHYmaSYzv+=ub$70g^-U2Ru-1d3#b>*U*UZ=M$YIoiezP+>S; ze1e>Ze{L{8m*_yD(OZbS*Ux;C9c0R{abeoNC*1JI#FTW{!0ulsxQ@QAOv&^-c+7@D z?|3_m@bt#?4Y42-8;sf>G88Ut4FQcAa7=3o&+97AsaTc5Iexp4nSVEndA_rf$-Ffg z`7gbI%|FwjaH#@7z)28{T83qxPGIl%e&%f=KWyUhbh{i}8Jo~zrhKjs1aEl_Hj`e% z$)TqZG-`&fJH7DkiwIoxQ5@&=<)HW9Ve~9ri3>0v4SqM_jBk@c>*hRg9(@X>RJx+yKt;qb_^;dL`Nbadf&mVMprPuN(-|t zNs>rvExBiTQ@c+bZ}vBr=4`#gd`Mcrgg&zXi2^~o79c>Sr-yO0WIt{w_zoM#Zh++d z_bC7MHCpq%NA`~%t5iVj1AN|G^@B6Em*%w+C+e!qY3TCH_& zYn}Jm?`J>H>&f+-%S2s3?6`XsBOMG+K+S`R}7cZU-l3jq|E+=ttLkBJ& zorm{ZwXq!>P%dN+Ol#snK+jAVc1%X`-4W;*8GtY5p90INov?@V#OfaLBAb1Z#R1~?4N}FdMog9b_zBHyhi!x$M}Hr zF3IIeK!KV$xIaA!iJTX>uG9}F|82#eXEb2@H$70$*b4D^+^%0N2mj0Xj(^qqG5tpm z-dp<`|J@&D-mYwegw}NYZF&d;8d`AqeSNx+YDRl*CelsUb0jAoi6v$yP^I$>#_ukr zzZaj<+#3??7abFvAC}3TNwa9e-x#{SbBKK2>9dnX{rI%tAwMr7UocyYt5o>nMQ4O# zsiR^KR`Gf{FBPBc4%ac;Uzvc{?61%~%K$dKOb2!INM@E<9CP-%9I75{!Q$7x^lFQeV%_+yaC4#;teUOM z&8i2`PvQlBmYz*x&!#cvO0rCs`ekN9XFP1VdL9DVxDKq9i#~6O(ewZ;<1q@0Sa30-8!JBF& zX8aZpL7RItHq8t~h4LzNAtgbNr8-kuSj>Fu`^i|%yw4o8+s`~1CyKj{c7xGV8w`O( zFs^9~2!Bt7@@--Kk?VV5=QVRYBpuB^vpZSPCY}%c`a@uqDTHqwJ<)88HqBX=L}9C1 zDgVbmMlO6CsO}h#$|vGsgUg|9q;!+;d2fuOeO{SPS&)Qc@51oQyj#exxDI1qtY)eca|LgW{($ZxKFsed zVG8ZWLQ7p0be~AYPnP9;Pv6;aByl4~%s&nbw;xCRIf1Uf5=K#p1ndq?zy}@k7-?x) zbpG0ilU}@p!M=HzuAhiCePU#{FNL&EhcpaC! zUSI}tdU>Mes!YE72RL`I7fR#2;RW|Rj;x-F(NEuEwfGE*t*^q8pjF`U-+1ips>QQg zTTyH9DK^G!qfGyYv@v>AFy*cu^oCx8-60Nw?it$rxi23JTCU_{-$Pj}EpCRXfAVot zrwsLWr_=7AHw8Ig@&%tK&u2^qOCelN3Tu_pF@=?d*xV>sqmltux|^}{)^&W+`2p*H z$P3B}uJOwx6yUq67B0@n#yOQ&apR!@fkc%m=UsPU*6w*B$n{fzrWKbVZmTvFSjB*0 zw-C0q2^g7SF^+vO!fftr;~y%u6O38)3%1y#phK@N>MZSpeH{zI!Tva2P?<_@k;)LB z{1x;L*4HvJ&n*642fHofac0#q{ChtYwsQF{h2Z^6r&JIQRz{%e@gTeq z{6H{CUIkQdrlLW&KBy<|WF&a`u;^I{WL#)se!TI8(~lyV=#b^G(X9oJOU2<(Ru|^f zox>0VKiIOc78)~!aYW6OQIU}XCzTjjb&AVE9&W|eK4-D2KMfLv2EbbCHiRk|@aLZ> zhSfh}@OJuH=3$r++N^HH9Xm=vuKxsVbx_9pOR|{_IvgwDS{vsf9bvMrrGiqtC|<7| z%QV_u7WhpV2R$#YGQqC#pk%)chGgP+!K>bw^oe{F%xN@$>30)&&Ul!&{H}p%^@gP; zf6X5lJBJLD?wBt$Xgf^Ldpy_+{vPbSRDUK6&qlhLJwGuCh&^cTZ<~MW?ZUc>;FxPzL!(=h8y)bh>r)33*8MQ_IPhwCJWL ziH#E{i)G*Njj=c^+%yB4>gr+X-Wq(nClY-_INr;@7;?FlK;O1*qrc^5f{Eclg1ct( z1^FMY;?(N{xOFV&IXIfltjhG^cGRKp+i51LMTyeqVJTt*?hAZ=O7lMn8N=$t6fn+0^NBpMC!MtxwG&j z+^^pY#nR`%vb~&y4U6fa+CgFuCQa{Y zg{Jp^FbB1*m?mBp=SP{w%-0Dv?Rl$fn&^L%-~UU%m#kdM|L3`qLi=7JW0i<&N{*nV zK`#z`Z9$KxTHraYn3?d*Q_KdZ*GVJ|h{`=9-SanZGcG3EW}u74OG zRRm*qh86uLZXS|YMmwL*q)BcP^!~ger)TMaXz>U$OYD9x%+K zQnK~kP95)E=%Ir%D}RDvbstP&yX*C^H)S3gOg#+axV?`{xgJK8yP+~&gWx%}u;Y9O zD114F4~!q<&6Xh)=602>Ynr%oYa8yLSW4|rvZ>#73q7wp!k9R%V&-#96Qy#0Jlc8; z?=>`{^5hgKiAVy6Ii0YrYzvp2UP`jA=JZdb4L>hEi$mVEocE<4oz{H8!LCZwk{HE7 z@l7=CTr$0UDv$G~x}o;&l~^P80cSHJG|cUu?p(9OYq8-dTUpfRK=oM&*XkRYb@2O zKf_nQ!!ac<7}a+K<3M2^u6>q@l@@Nu?BH?^xvH?dLjcta_JCo+JsdtdjTG_%X#V#e zW{uVWYlapQmHFhi}1UOuk9pBZZ-KGL&6I(~1Yk(L|UH7@ojXA5#Q}<=eo=EQsot zFQ*(&HTuvzi?XM>l1Wt%=3vm*qUopB9T*v^BndhQ;lat9~J_TV_h23qC+l(tvhpb?k(^u28>Ijz*C z@{B1U@@N+1X1?OsKNIlOr3R*v(=(Q8R-m*~9*(yVrI#13!3pWB%)tUKn>o0Kjc=dM zW=Ld{{L;g?N}?StYrkSqq&n2!AK-7@n98^JU4ULEUt?1GFD%iyM7McuRNVT6Z11gs zyo_qlJDmh~^(JC$b~kj0Z^PLdPVhkcCj1krVO)IAL$aj=h-aL~IsK|A*}RIIxrfn( z&!Kf^Pari`hMvE!6@1cL0(Fjc@P6|de8280%C8wh&r^%=U??9QRi#n?s5cFs=_0;a z2j%@N!r%i(acrn1jylU=;Y1~3^ONb5oD2om9mY77a!lSh8%74>!1kg%y6tdAMK&LE zY93&r$^{hnZoz&XMH1;z!M^n+sONJ4-SnfOzjqMEt?b4tM5s0YBVK+om$KqUq4ZHT zJbmT?7kSZmXXAdX&1t}eyTs9^LIoVf5(Rg4b?}_lEofW55~5wqXsOo>(%aXMBSPcw z0RJ6`+Z_>@c&1?c(RNJD(4%wLa?pC25$CxWfbNP%_@i_tmUj#?4;+_aO{PA&dRN1D z#Uz+@P8)tV?*N;(J|tVegTA&qQK9KeMr$a7;jzPfVN)G!3+Ff#gYWwB5ALA5LS4Goj|dN0b@F;gnom77Y~`)?R_ z+?ILl17;5E(5S}@T{mC?b{7eDPaMmxjvu0atl>ObDP(NBi#QtvO>oVnK6_c3%gvf( zy%t|r6`}K~8hqhdPNJ}mWZMSuNlXp++%aPgeV!<=e*Ot>_YK0_8+T#X+0FQlzaP&h z``{dtEF8Nw4Rc0wQOi963?AY+raEqP9F@y&KBeFIyAXB|-P_gkBp;7-XT0WE~7^-C`z0 zOTre;&*Gw=LkA|7(ChYocGr~-c30R5_RTd7I_|GP?_7lG)=(|qbfXAQ#M+M;@#zHJ zw?Dw5whh+k$uWf^JV8l!Ea-eFq~8sBgr8!l?bizmjP9hU*d`kD%x6U~v@SgaOCW@ls?T)XTFkb3f5kgA+94N*?(y45D(&?KHECn_-+1 zNzzZ9PF&B!;g!pg(GUmc>rsqHc_k?Q@FaztRO07mk=OSQ-0}Ao{^Bx9)qfA;t=I#2 z!Pg&ORuu76%3FDA`!bm1p_i1<^ODM*7g9@-E(&cCMuoW*kbZAH&6J5E`QA7RTUw85 zr;5?9@Eraz5kdBTJySj|89bdOS*y(xSUf4kUfHC9W7H3U(5rUHyJ$vD4&Jo(XBfFU zPo#s=H?aQ9aa2jaN-h-#D8rTeAAj0QyIce5h`S1P#x)6C`X%^A7j*^4FXhnrbPe+S z_X-31gPES>>&AP_7MPA0O~C@oGtff>fug~5F~or;K9?ccEB2(%<>6<&@+Qr(uNg`8 z>wL*rH$nL80@L-y-lmnqou=Wt=ds!ASF;2BrR>t2MRe!V0=kftLvqE*`~y3p1)e{r zGoR$Ls4hR7^eS^GMdB-z7|Ef;&yGlo&TqE9%gOqG4PYdO4;#|=(ydaW|b1yaG z{;f&4Jfw;V@qfo8pHhcMni}-CW+J8Z3y6`t0J_g_!t}tcXtL*>t#f4egn;ocO&BXghlVufP6+&u1)UHwABDb+eOL z1Mk~pyXimbm^eyXtBKk0M1(muyNdZxr9=jc9O%BX4c$1OgJw@GG1V;)AGm1JhJ@P~ zxUC-jYxCKBkrMW3YbHC{WgIT^lE<>zcKmb6nLoqXk*Bfh82H@Krq}Dc@a!HDsyPPYm?(smZjM$?!(T+aPFJ>lF-T;=E+<{tN5LD$( zL^I(#sPV50dmP7*$M0)ke6t94noPo{GbNxh_cPyD=@j#Cd=^|4oyi<|wUL>$xRFt+ zR)SfP8I0|go5(oyq2916X)n?wq3t}f+-pJ!#d&BWaT=m-W?<^##iTpxOXp0YX^*xz zGgJIO-rG(K!LCF3IMZ$#cD&I03EYpI3iLET?1~^v4=-@hpgy=p&ZVi@e(r6-h!nvyUF-l zJRNS0p;^_D7;9f)Ji9Ce`&aD|%vdNx&Gwo!&3!GM-}aCR5R@>&-^v)# zffJxR-XC`6SHgY&XE5u(KB!yegWFbLLMg@vpP!hFI-+Bt^~p27rQ(04X0tEx0~^YC zQ8m>({ga=#T=;dmv1OQ63@WmUW5n37ZSSeN@-#%e-pr^->|=OuwCRM`ww^&Duxq&-T*C8&RxrUN*Z^w24(O z6{4{va#VMEJgwKf4X3+TL&$tv_$9NK;P7l}SgJ;HexYn@!X|dVdk~xaekE4T7e>v= z4d8Tt6Ha$sj!K(*V9BLC?C%v|NA5}Zyo66Jfo9|y{U0vbFrIr|D$=st$@IZ+8)@F& zOokt4lh)z_SaHbdJTE{6vnV zlH@@fs<+XH6;bqeEs^?30d2l%O?zZ-;f0f)=*cmBcVrO7M=T(P_Pum6gRsN!9&nLV zbedC-7us#{Ot(LJ*nD6%)z`zK;0$QqTg_?6H!z;dGq1|thwgIGm}OCo@@dy#SWpaU zVUIxKaslbgDWOxgkuBn<7H|?M3LdbrE(dX`mA40bZ*b zjSE{Gu;%SKxRa*C&u%j`)$!60SUN7@?`PNWHw~_1Pw9EFlS4e&6i-R^LDmHJ<=@Hd zga`T$4H5A?(Q4$|9-_p zQHIX!v!VH4=F^{@i%EIEC{EnU-8J(74(SEb>N7Uvv!#){2d=}vc{_3LE;oF*dlUZb zdtw+uuHb3r_xrFV#Qel5tD%>9qAY5PJl) z)tfM_T?wbO=whP51N!i;n-ZG3DJ#7Va%`sH_uhHX%rQ=Ph90J#TL(y0C=+~^ynrH$ zd6*eZq_jerT+2qW_+JgXv!a%j-Ey2gU2z52lcK27Hx`cu+T#Jq{b2We5KdbkAbI&@ z`XauUY$S)+R>561S+SiR^h-g_i%~e?UKuXFoC@!2b@24{BXB=KABSp|bEcf5iG#6IvlCG2o)3vmTG0JM zp1HpHt{@_;njAT<%#}EQI$JpdLuXpx6^`F4eeoR5oqQUPr6u9eF-}X~aR`@jdvU&5 z5}x}Zz;q#LP*s0KNnxi*y}E|pd@+I<>rBCZSp}F#-=r$@V>ENjdit+yIt92f^!Nos zuU>4Y!Mbo7$=OaRIvHf>kxhJNKh=oc273_3f1Hl~(pi#5eE;D3SugOb@_5K9y2dP7 zFoQ3}|BdPM-s0}56R9h}jr_G{QA3L{^<`^w^N%VuOUsikd!1Q5woD+l?3C%!XXi{4 ztu^_J^@tyC@rKG5zouLFzS4%*)95;v30ZY8mNvRZ3Bs)^_{Yss1-YM_=t<{2au)03 zdfZIN^5qKByAw%O)9oqDU@e6@ZzkPX4HD6lr5Wdb;*I_vs9~W&SxV!{oZDYGbZmx* zpK@@X>&1OuGL~}n3sEosK1>+*4~{<*#lD#=j%@SBw-ugvPf{Ii-JM`fPpzPB?+W&t zr7v5McPRbtz-ButI4!7(1<_*=UWI=@-Njt7-wn0tsOEWbh%=kKDA8nLt?CWjPe z&SegS#0#e1m<0FM326BQPjX<_P&6Kcuc2d@#Z~wD0jp=yfBh4=3{O3#ao(RuK`z%< zmdbu;Uk}@vPXgZDEzF|}cTE+3NkE=%64=}KVvc1G&YJlgM{KHCp;@QcyJe@@fM#7h zsJ#~VEMAM#?0um%L>uxSEfVAyRP$Oo)tF4`gYrXhm|=PnlC1AA>s?<%RD?e$7YJcG zH%D*INT8F|`)I$(WM-=06@l)KQw;rENwCa}GDI9HiZ2Ymp4q^9kp<8wAw!d<8d7OT z1pO3KrRrE~dNR6`F50+Y`sIso^;Q{9-LQp>OV-lX9V@9}sspOQGI_n0}i((!(?pYPU9|79(!P zDi@X@0+<4(-leG)wNvzMJS#h3_;<}Wa=B6 z%`x8MX)vb;e-HmbpW+Eb*H&S!8V`@HUyZSq4UCS{KEXUQDM8-zP&)bkJJx>xGjdn`u&+~59Kv=HB@n8U^Q#*B{iB{qvW!m95`W5c=U_|G&csvHrg z(;_^y)0+(@?*EvV%YE7KdseLR6n%EyrVkYH=sw-p*i7Luae`%m)qGFseY`gtT`*}R z56>F**r1JSle8~-yDLdDAoY{-}>c1A-q8=a63I*v7P-aH3nqKon36-{hQ z37T<$!E5xbB zdNN&z8PBn(jxxf(*8_9>79^~ij$^7-u_wzNqyOuoswWj>_AZ@-)SMaBrT)zA{#Yg~ zx}KzU8pyr$F!kOjfG%HU*jp?D`yJk3DaYRq$y26}%(}A;hpuw> zsu|O8oU$hj^wsj>CqFa|>1pR%$-Fb_TkK+-v)q7vu}PUtJ}$~OPLF1H7RRzquA5k6 zAB45%{J`+3KAid1MDffH8mB#*`u&BmUhEIJL>ve4(~fjoT8KOje8Bq@fSwP};pdfC z@cj`5bc(zW3xv}^>}oqND)9n^^ks0Je*_*~nS-upHStBH0UpU(kK4HK z+wk`izMf^s3}hxV-e0aUJC9_u!t3I=F6$`vLRdb!Z=1cLikE}LJ${>Q=@#$m^uu1`1b@yKB+MdX|m8dx`=;Y#~OY?83@gj#rFSH&^vcL z{uSln#|6r0xCO}JfC8;ie}y43oX@8r9wNP*q1WOrnE#YU%WWc9MGHx6y#YCus?*FX zdhmVWeWqyZMwrW;H?6BP$>4Q3Jsg{k$M!shKqeW=#EU7RyP1q09H;5iGf5(IBgtm? z(t2A7`hNWzuCE_Y$_iZHRNDjGuKgAZG#PT{a?Dw0-AMZ3Wiq^ZhpM74P}Nex+A6^N zvK-s3Z#GG8XK3S+X*9m*5vbkq0DLHp)g7a_Z|ZC8*?bEd8g?@>pHB+rL_FgkFOP5cO>CLy7Ptc)Y|HJb&oI9jzR=KF{9h=1dE|m(2)o z;Qkh#?-L%sVyKNg@;=bn)K_%r#zI=DzJXjG?xY*vQUu43`Z2lxAv0BJ7NMp*nf-SQ zTVl48;ZtwwJ#WKtOqyUUr&avWea}R%NybH~jktKQ18e0495-B&)D~XC%>~=x>-{X| z;k`qGRJl_aJU11bl2aJ1Er&r&Oc=*D9|zNGTN!5`83+oA#a;Gl@b+6d-|VI%Q~LBd z4c~93v%3z`wazi%-v1Cn@-#77{T_@p5&@q~DTs?b2u<<55E%ItihLAk6-=UUsX(tU zz2rZ?Bf{h=~?X%k&-+Jsp)yF2d^0ph6K3m3WedM~` zU3J-z#@QItlv+gD;72s+-^bh z9k0-v!V9$OR1+N!Rbq;-7BJ19#>43LVG8*4fz;0SQu-kSU_VHLS701G)^nj*XBH6O zz=t|bHi6-Wqww0R29Ag2VE)xYM8EU+q0W)&TPM&Ksbb6;uaBnNI^p0XVUFFf4=1%d zqIB#LTpB1%FD46-tkYYpI9|j^2PMOnrYYDf*e-~-%jbLa2MOMtSEbxYC7M;IK@%FK zA;J9+yqK^W&+8#G_KYle{h5X*KQqYZ_`WCO!XY)X8loJ1nIq}y&?%vY3b+#uBN8#= z;#M^IF$)_)HE~?iIWD`ZgMKasc+rfZB_F2Isw!bJxZ8#D0jIEiRvE7MNumyuee^nF zCz-|HW5Uc22x6z+sgEwu3AH>RfXo|sgdcJFRsb)TGOweu@k zr9faUrZB9qo(^mM`33Ab;|piyJ~7c11N0@chn&3b((d!JMC>fGxur&Tw+gY6&#%&| zpcFC@nMEh^6zN`AA67f~QViLVy0tP@87V_%OOn9KA_M%5jA?4;crq~GPvgFE8Fan} znk~vD>j8Z(|1+9*oXZv-VV%wExUQjMsw_T9uZ@n7*2^Kz zJHqWI-)G~V+mY;Fxe(TUU^&}%hh@c=&u06I>{yEfpJCl@F;p@)#y`UuwA^|xjoNOc zqf>X&{x5~JsN)Fz8YraVoG|P=k!8)xh(s7pA1{nJ%$}G+g{`wW~C-d#ODZbD4q9m(CTDq~6)?B(mmi5`R?UFBT6In(# z?D+KI#B_QR{~cBYxeJsg%cIr3RA% zRL?laG(9h-*~3xPv1J~KI&fNO?i3mpk))M#CejC+$5=6Y7+(jsqUdNH9+}&QZO(^r zc3n1p`LKfPh3jNawDANgjDTM~w+WZ8EQIBpFTuS$4x5d<&@V;=$IWhrb?sdsR#yZk z0^DKaV4>h=dn0dCN*Y`6GK$q{*};yR*M?f>Z{mzoC$U}cE6CkZ;}7*%^M%=mTxZTX znox0=Qg2GK0kLD)=P?rOwAfbDFsTyL>)tbY?*t7TnvU(PL_Kc2zu2f6X>JAaIGIuqaJeq1Uw^xO^va1yBI;HWO zMhdjruVX*R&S!t{?ASX~1TfLY3|?;^VZ=>!ASL>Y>D!pI{K}m*ro%otrcbsP^Z16v z=$cQs=5`-+Z{+$YvOl7M+D&Xw2&Ro1v+0n*6sj}JV}{50Go}abG79;5^poS~pOn}~ zD_7dV=Ezc}hdsbt98RN0oDU^pd;ooUQiIWb1?XxWhhJpe$Snhi!@rQKVFk^9n?e_F z1k=wI)5t!3GJW~=9z&wM=}X>1iW{~hSnWVZxP8d!EIyS#8&9b`1yXpXNeWkV;q-ZX z7}et5Q!h2?O72fI^u31@ir)#;Vk((E;oPn;C5Y>73y1Sz#n7u?4}toYuMLW!?ry=S!&e3Sre!nNpK;8GuwEQ} zAV;1D8)5g!jp(_%5+98z$Bq{wbjI%s?i!tg2dt&A`t4!Z>%JG`xQycK=nwejoh>d7 zutw*t+L*~-MfV@rQ27K)`c>ir->hRHvCI#O=S`qp8rpQg)P&9s2BKJm5*kfN#MAX* zRMeYGMe%L)Dkco1l%Ik~fDe9Eokp!kCs9-ISNu80AG_7Z(mpPeT&kYPpRc$a7M|UO zt54{_rj6Ga(IpFPK zfT>5LIKN{FKHB^Yb;iCEgvpH&1VyLvmr01h4+}r|`r;LwI9Uc8b?!qjFBZ&iF@pC$ zU7*1s4bArDGL9TW?4UsmYGw{F%iGJD<9Xo_=3j_k6W-#+)bDtD#~E;f^HB4o3%;3T z@gvy@jQ7U^usIcvCoN*}$0swK6Y!D$tHQ}NgssB6GkXs$wPaic{ueH*L~&ch2lYBpCQ!m&5xS>s(>Y}F6qr16WE@v(ZA1OegvNO!OFB%YfTM?S2#?eK0Oug^Kyj$FZ8XSCR%Mj2K}Qh}Yi`wdMmzDbSSI!Sa^D(?NRg@a>U@!+YY z9G|!uO%50k$3SM2E4A5(1yk6SFOx||IfiCz3L?`r@~ls;9&73}kIj;*Ko7}1sFAV* zM=A|5YH0;1DfWT$>NxuNbRN}v@Tl_cI@ULX$BH!a*h1lRn71bxLR7l-vXllUgW=3#WF5%gnBlr~NVy@VkBmX1HF%IcI;7 zk;!qYi!3EOjxqOeW*Y`Q7pCZXX=qH|*KV)8_?l6-^d<7Ux z+Rl|y`tN9gPr!P?pPLa3SMrRhD>-g{*Flsj>4g_(w@PTO_hE3W^ zKRH(Inut`Yx?2NNx0pcAjwpDQ4>cV6!Oz+YBDlV#kgdwn#OU)xGo*YD--fr;eewVx7KU&S;o({so5A?nT2 zri*7)NL}kAN?p>zJ0T^YOx!+sB$X6A{HcLssF_XMia%Fp;Hqn-=y*tjs<zJ9e4kp(w6v$h-3v>&S zLbZ9c;NN08Jtvo)uq%U|*PF=}c7MkW`=ux|oMS~BPB%Rke1rcazed30=02Mkj)5yX zhS;AmZ0C&(HsEjpd*P@9MpW?8FGCCK&3-da?6>jf9y=!xYV79CnlhVzq@$E~PSXGj z_o?FkCDGVk--koXuAzCz3w*A6nBudK&|us_`m--fu(EqB6KlMm+1&aI|4dgQ`~Iag zq#6n_WeY&8$pIDxu(a`6FadjrWSXv_ndMuQ+$v66Lqc)Py*N}5;W}w=Wz(g{O(b{l z6RkPA0W((Y#{9V{XysObQ%64I=g%`}v9KJSQR0{tBHEnxRR%kiHSpb^lQ=Up8M91FzA=o1sxmIWavYtUxgNU>@;Ivc85_*c;PvtiD4W`V zYez*;D>Va_Z?7W77X?&&J(IMz{Du#zNpNu76~_3NHW|bUX!>7kdViN=ITubP{tFG# z;#e#zC67_&_G(&qI-YL#c+g=}W4hxX!(^P3=YQ;2%)5VLAys6>(Bg!1q&0>I>u#zu zGP+8PKi91kP<9=wR*WSlj#;g6(30wAI?y(cI6OaEfQ1_q@R3d+l*R|}w#iQu-44I2}0z$OPcgjCeobRAZDmh2P!s4VGU0a_O&vM z-$D=g)Fn$_hE}6bq7lYj2}N&{weTxR3{rbep?vyYJbly|pH_(D>QWbGwR^Xq=~W6F za6OQ~E z9u~3u>?;~@G^3Tt;@NY2Vy-Lh?RqlmQzKhjUAi4^K*~lL6dx5&;cqRdvGOIIPWnL! z!oa%KsnVf&J!oFugv-8)!i0^TjFH-UcoQlj@c$JI{<}=Er9p~ayhw!|_t=0HT#ey5 z^Y`*gH9PqFYi8ny(PU;s#ex4sJdPPRhk+T7xt_QZHR{%ThgaJj&{I?bG7^#@g=1Z> zjJ3zN29fA_`Y3kIcZ8zc6!6}g36k@dQZcv7T%6=dG7WXm60HnpuUuo+ILXq(VsZLE z7D3IqOwclZ45xDeFE8Xe{?>U@bJGHPvoMDwcVI8rJSs#rpBi7<UH!NJ$E*Hrx_* z8<*hfY5Vbj%{aJnxf7&~G*SP&IW$f569jfQ3vw6w@JBm+_}&Ai{D6x6?7}mpY}bwx z?Bnms>@_JR)?H4M4QxIpID6s;W9aI~Tze};`}0*uO%ep^6i&Waf35mMzN@B*_8e68u<@ zzx@QQ_d831$QH6_$$$*5;J3YLQ1evoYQD!8pTm{K43TPCk>%uI`w!r z*Jo$NdDPv=^-?%FI-WwWab5V?Q?m%B^6zT8hakO1Moidiklf}dX{X-L@hY@o z$$YSTupF%2n;}SIIab|Sh?d73uz1@#lrrCfp3|Z*-|8if-1xz1E~4~h#5~g-Sh}Mc z1MYq!seiVtM)4lDIwb{Xa9qB}R$jQ*V=iXR+kr+22Qc8$H#%uIj&%wCN=*~(=$FTA z^6s}IH6LZBTI4R{E!ia4XMr$x)SS_>-VUz&@6(KXEp)qs>vi5zgpphpW91GVs@o*P ze&N_z)4qtYH+Gw&Rr?tX4E%!OR$9E83qq#6;gd`wr*D>=%tS5uLFRx?4Q;O3OV394 z)1>*<=qM8j$9+`=8cXKlYWM@f4!Y<&p`Y2AX~Fo0J>$ztk3wC0<~-zHvd6MOj=y+58oGC#0^k20 zFy!? zxTqK&L`Yx*#~85N=8hHn4&qnMmsr}u%`sZi26{uenO4v(fYyzy#~JHm&_O7@_pu8*30V^TRQ zObSQN;=$dH5cDGi&gIU7t*5mprow^zU}fg-jLkne{fxAAFxk> zI31Wzn;Shz^IHMN1s%n^C&XyS$RQ}VS_|v+dzj#0Wl%b+3#sCF0KePQc&}iJ$U9E4 z7vJMx&_}es+lEPjoR3cBJ9;>brGWh9=n;~J)^@_=x*&~>NGV~}co*5(%M&p6hC9CZ zkwZFg80{=8uy)obe0W}!?OUzD9x@-x2Dm>U=;)>$W<#{MK^*L^x-kXk>iPT+Lrmne z1%esLAK~krXtG6nDmOPHzmsmX_d@`!;Ce_|wD=b-$%yjtl>WN^|;YR)!_X*hgQ3DT; z!H)m2_SR8VeDU8e zAS$+gQ6!{0rO&(u6%;!_loAO+unW6TL_t)_BBX5ryT$Gf4D2EVu}~4)XaDY>cddKZ zdY*e(YyLQ#b7uCQIkV>z?>Hu1?Q&j<(4FSmHNTTMAF%H(_9u-MR=M*sa%X!%4*x64doVv_KEvyN|x*|fcx!`2B0xUu?^ zlGHNb;*kMzGbeD)T>-JD8-O_Gnez$8;PCnF2x6*RqQo5 zs*nzBmxfP^m-3WmNl`ndOGD$DNQc@Fmxh11s8~_eK?;79D9N{)$*XU+loxz%C7-<7 zO0m+>O%X6@iz40PsFbT!E@f~2A-zy;LaeF4-Ij$&FO~mEO&_h3^wJbkopruKdzY%q z`30MuRKo0>J3WuJ-+NcfamYA*c}Cmrvc(V|`OhIGx$cvaoa><`_v`aS@!RdZVnft3 z#nzN7T)%jQ`4=nEKAYcK*)?Cux#wPK;M6Ht6fYp?QX-Q&1j~j~#>-5k`=VS$6T1#lyyCXPN^Oi(abl9>&+Q zKPK5YAD46|rb+WHn@R=i<|)#8^^^iCL!^NfiPF7kvC{Sc7pd%ny7Xb~LFxU?4bq^* zTTiQLr*>_uFeWjB$(wYKE|z*JrAvx?|w^@ zm)@VD2ncJfIQQ+D>xxJ-4(S@RCfSvS+c%}PB^8pL)ekA*O1|s!axKLNrxl9fs*4qo zF)>a-GET(}B~CBO>z#s2jhrtF-}jxR%g{Q#7LCGoatHhW1)$^Db%Jp-kyJ?ai`x&4VFQyavQ6|;Q?5mXn2++hyq z`t=ez$G8`71={4OHqQWIYYlfS~pgeSKXTNIKK;h z`>aCuQ7Su*EM&lgyJT&9&f0l@dFhrQ-7DWD$se~#nyzhJG_GeVoQ@PK@>@r`&OH*M zh?u4=StMx6Wuk`}YSvobU}COlvgf*DtZkLTF~LImJoKhw-vJXTC0cy@tC8ZUNfW8Z zu_w~L6-T8I|4Y)Y@O+nL1uI=U2=70Wo+G8}N3^ApyoOjhzypcKxP5ZTa0w<=uKvMqixU_ZAap}$n9r7y8I8Zr& z&%ZV?#dSX?lY~ZRud$2{2_|{X5a!PBMa?h|Tsu25(cFNM<=wD3JdvtH);vfVj-q}x zSF@K$BZ|jKXSVN@jtx1fXqK8UmG4#K&d6NFu6YL(_WqLCTzbg@{aP+N?BbYpA7E_4cabp2lrO z+3YI(44TMEOIpf^tn_O+NjC4kAyaQj#gg17j75n>H@p-N#Iq}ZyozWGgBaVD}?Z6?S89LX; z4%NL5+(@$*P;GnC=DSd(>&l*DTMoUqVQ88o(~BG_IB3s~9u95M-s(n&kpX&?r3}3_O#_jkR$!HOi>NDrecIGMOg~;Ma^2k zu`#uAU|OORdJn95Hp`SYHeCtcV?j=Zji^bQV;^A7mu8l{_UpMBt7e;{TWP^{bz54@cP2*6AvM{7pJ`nweuwIbZ0aPm_MABu;VSB%a4NGKy(-;EiE!qew*zB0+fXgr zF;v->QCqAj-R?^L5N8eyb3v=S9p7Gv?>cm&dWHvo-gPJ8uZ-FqcSc0_!0n|Q$C^08 z7JJ@2btbk*=uN0O@jOS&E5V8hbFEl3(uL!9ZRjPi{w3~iNLFqLt2_p`a^~rHSH8HF z@w`Pjdmp^!j4W^@g%vz%Q^AvFuh?<*AsxaVa&r59^sCDFx#1Sw9^69f_8pX;+@qFd^ScUz39q<1Lg478y+FU+6PkW`!ir~Q?D4-&%ib4I zo_~YUH?9z$eu{I|=c$brI=|nK^0(nA{Vk4BUU3Z1XJ=^g>I@SGT)_X3NQvEMf|l6-mfMuKzlq_BTkOA3#tyIR z^zC^Uzx*=718$vT=H zLF|}2jNAB>g=?R)!|Ek2E1se_UO~{EO1za{P$2MvIYu?y8T1j=_HVft_>qF2RWvkx z%S54V{cFQ(j9OL@Jn;#>Nzc$Oze^v1F0L4HiyP_p_;1fG)?Y6}p?RN_=})oEddT{$ zN8DWToL{Y8{GMq)sxbBZleDpus|+-dH1UyGfoZUVTYdV)bD`}GJypiY^ZT1Ti=5p zxeBhw_2h7}2UilhWA(+8SR)TM>UHB+6PYHn9eFDtX?iz$qI+!!m0i4eV9}ok2M6Gw z--|lGp8VVAAv|9^nS02c(uAI*FYHD6M`xTm$!L6VXVlXk#Et36@>n;!h{vizMI0&y8Z^%oMR`9K+mco7?aj&e#k7|yD*OK_O4yzp>Io#_#1@GQ)<8CEE z+bhZQtzq24DvbNqqI<7`d%ExGdZ>;GE8pR^rH;CRZ)xjXhu-jaRF>7U@_H@v_gAp} zXf@ORRA6cHlDH2~nHpWeNVk_*w|Rn5$wOjBJQld7=je{C#rWMvwitY6K|wtkMPIQy z^o8xg-!Tb!i{Fk)uAF(!Z236}^Pf__|mx7b=RwVPXVii`~1Seo=GR^y@Qg>=S!;U2r{%Rcw zeUsUKF_p2R4luK51uuV!ZA>Cd|E3Z9dIjdEQ+c!`iH$pxc;&kmNo_T)_N4RuWd_=6 zS$KNo^X_^UjZHT&$aVt_7ez1R$p&^bWb@WAla|hTtjW&gmsTzj<2K-1l7rT%4Y+!& zW7(4o(qE_Jq@77y_Z;T+U5n0~97c><%iCiaTz#8P*|r>{igc8=CzI1Og@vAJ{Mej@ zgXKybJEZckcooUVH*oRJYIcb4R4mJ4UYY0(KF>gV@H*kQ z9hS;tjTA0LX7G4r8WWS&WB(_Gh}OxNWvnBzF@t~JImGu!=U8SA=EDp4wN!WkXXO#Y zCQRRFkScJbug>LSlA2A1|28t8X7cuOHoGQm<9nNYaw^xO@oobJ|E*(3VLBZqr;$1~ zi|d2ZnC6?#(HWU+oV%8`O;fpkU^NGXK11Kfne4o?mdhWLnJDx+etZ^Q-v_c-xk_E0 zpsXp^%+QvT$LPqZZFJ;=Df;r&rMhzd1YP;@4jtLzwuao%yrcZiqMaPnRzvVwJIL#| zca*p0bdp~@=_nuVqan8s=qxv>)slB98OWYT4CVJ*_2jr>GdZczSk@6(9HS~rdDmt` z`E0GO9P-FeUf9A&4s{mZ6s5ZI>DBsjl&`TovdUDpa4?cRjJ4#iBXs1558CqR51O)} z@SwiEKvRC>*hO|aXCVJk?jj$3s4eeVtR;7x-B~uW*O47RYsn)fYshD6JIFH||C7Dq zH01vl>&Wkx>B)!ljpXp9M)J`PR`SJ*rt)=f6M173Bl)Gap**jD7x~pYO?g&XNBO@F z9pw>!JIOCYwdCIc+Va(fdUD5dEji9jPwqUyP#&?zP&SJ-mdkeA%B95?a?S#C`Q=-4 z`Qm9a+2o&<{O**VTsl@ycAB9pKM63DPZ66PcOqtb2;7(E2zGn$oGpej7yGV z+lYlkUtNx|VWOxP#q-f4p61WvnOd8Ohiww~#69wMTLSM?;nO2MvwGKW<=}E2c3wr`oaN;8NTks+9*_O;6!|S=hWK zZsBC?F3Q5Tu;^_G!%vlP|HEz`&)m++#A0@*mN2-a1g9&zIbgbzEfw3D+hGse0*h$< zZZECsincuXjhWY*Pfe5ol-!{dD|EtkVmYd5c5w382-+sIxZhY zSU#N`3YfQfBU0#gY?AW{K9kE*`yyu67jvg>lfXc4W7Fdt-qhr=;chl2uA7P4nMZX= zE(OoCIPWL6riCPi=3*PafhPZLU}4@S6oU$AHKdT{=9z3?S%~iWO^jJpz~CJ_*p!|_ z(>Fx~`tP8l$`<B6l<_1`Sw^6gxme~dXtcko}wT?{PW%QcPt zjF(C=xPOo>3r{i8rj(aUPvPxziqb+SpX()d;d(mdRKli?e zlGDwf0sRJ|J!lAnoAn{*xHpNl-i%x$`qwF57&Dkq4}abt@F9A77^5$YAj>}-4ed~d zxkb}{$#imKqG(7P$Lr82I&F>O$h?s}`Z<*1uRS>M#fKqZ{iwJ$nD&hUY}ECoeY!6Z zvA#TN87#aRM)6{FB-d|5vizox@F5$_l4>u~mkdGMB9Qt&A$$o6Vn$?uz?ymSN#G$Y zR)t5 z0{hXAgTF^mA7gX9XmxoccN@aElo}?E6NLWck!X(^K@+Pn z^xQv!hXE5XofaZA-b1LHJDKAv#}XbJaP75y4@kcN!sHM^I=P#mDH$bkG@#`m53Wd=pN;uEByYHWD8XF>gfxW^Y6A zTolP=_0a-%7>-fISlX@%q{+r05_gYc*nnV)w@k*yPxLMf$MRd6z>v{nP(Lt^CBMg_ zUq1$;9#|_8pNC4MTMv)d4$mtOtYQ4UGYW49KR#Fvq+ZLHo{|?Tb4IYTbOfIt zj-atJRMZMa@n(7i23JE!7IRqgcN)9%LpdH8#sS|Egq{s#MraUu=R#OuG=gP{2tp=? zvfkf^nljQ`{3y1!AI9jz;XH0K z2Cw*F96E$Ebbbh@yN6@kJOckALENti#-@`WAsK#{Um3>Nmx0*K8qROWAa*VdAuDAJ zThEUa+>lVZ{2hbaq##=AhqC5=G^!6Li&$S%)>iE-$Ia;|uTJ|<&a7xBCr)iIp9^j; zU;N!l4msRP_I}?+cKO#z-t)ALd}O=29B=TS{OM&!S^G*ydB{o~d6I>$Tp(g2PhSJs z=e3!9u-Zsq^GxMWqfO> zcCy+Zb@_RfhFsKCUp}NBAl# z0z1gVb=BoLmFn`(;7;gPI_`#A3gbRKLg?Ur!U7DnaWP}I3}pR_lW%JW1wti=&^`Ch%c>{8ZQE?lK0&#LYuYwGC8f3IuH8kcnC&JXnD zN%sxq-2YJ~8E++PuXdCN4|J7VwsDXLH?b2Oq^@$eja}t|&kg0?&2?m-EnVc`Kpi=H zPG`Bxg^sdEM|F9ft-AbizqTwd(UJpu=*csJjpQvFrgBuiiTuRLOb+gCC2w}KkVn3^ zl-K&cr0J?x!f&dECg#uhQT`I&wQp#g@Se*rJ`nl!D}QEugWBI*u>65zTP0b2tFqjf z^p}%Xs&c30D)J*EW%K7soeZq2X1E*X6LHYD= zPx^TFbX9 zpE|NNj*owEN#zew@9P;?RnPJzb-eHJneVR~NS^tH@4LUUSiPQiF<MwVN z{!UP-lAP2}RetzFRlYh#MXq}Ji_+2`T+MrrLC8m^*;8){y)83vXwAV^-8J zqPr-`5%W~!v+7Fnfo4r)SE-4dui9LmlB+70?-gEDw$0?l-CM}F6PnAn-mA!26Pn0Z z^i^d`G3OhGzgT_wE4TK3WZkh3^wN2Q){ytS>{P>CnWKq?8Ez4eIF|ye@nhX~C^6xd6Z(hUv zERjpAr=Zt8ndu_WNdLK>_sYxBc1U9MgCq=!;<0~|NcH{|)YmQJ{hWBTJy-JdWjt?0 zfBfYALE2bT)m)CwxARx!_NCGjFJ*!d=v_I;8_`kg@Oy+q;w@&0a-`5hdAe}=hs|m9a zHJ5}n{P0U-koHPkTCG91-8#DdO=s4zwTwQw2JeS!Nj$xVwCQV!8k$PK@O8qIZVku2 zq~ZBa}mxn+HWa8BkJ;- zDww$Z6`rNfXtuJNNqcL#JzdoJqN*@Yc}!=uXPm8n!EEy?G}^uqbxE;q{C&zZl^2v( zzTo$hN0^9x&R_iiPT%9_s&b0kKV&(Nc11I4-&Fi=(2>epN@oYzG5UZX7A_EM4uT>3%c&mSDQSwnR5&vY949-X5# ztlnBB{0VCr@b5K&QI)j0{)$s^Z|J|cnqF6)asPW2eal`@9qoVz%VyhW(;T`Zs77yR*ZzW;lKxyn;U&8p$5$7}BN zf5YwLZ?L}aPSKD4tRK;v%XDMf*I}%l5+eF~-k7R+G9|nhrJ;kky2yi|%e^Qr6j-SX zu=$sQZ!z8J-rfVxFquIvLf@@M;>~zBj-2Suk4p-^EV1X(Bo}&=IuYX19qGSb+#KXa z-tivXnBa%4Zv^e5`{3i|&SQbm9M-Kjx5ftXQ^Z{ACtNwG=}t`bKu(V8PmPTiQ_l59 zS*-`(4|oyK$A{j_M2y|JC--j4!Y8yhYDMl0Tp-hWP%i|93xlyexj)01T{(8VI&4jk zFdM$_?8=ZL2OMW2;-Q}89QCH`sViqT2;EC%XQmm-c&(B-FKYN|8x_nlP|$s)3t7Fp zGPc5=@8g|#c*})0odh;uwga7v+)$YT4bQvty=_l=2z;($vrMy5u2iJBVkda)WnMPi z{A0?;8wv^nWP(B+*kd5^<8&X^y9w>X53Yo#$ozZf#xy@y%160yOItyJk2@EQy%1$P z9=%ae9W7I`&7HAxyK-Thf<@02{21-dKG6%Y=-z|GdKacWawN}LK}8^Zn&rXX(S2#v zy9bN%`eNqOhY2&fQF(h1^PLCKDXAAvp6jv5z>szi4A5<6#NHdWtTQxYM}YyuvyHf< zYQ*UChMaj|j%{-t$`-2gDgHm2KhQwCQ(Io#?j-OfozU~rXLX(-OXuk0WYC%Q>biJG z7!Vn3NYfemG#Ow_LaGsi{R9U}$%y;5y1dlZ<7br~rF%8`nb(=I*E*8?R*R0V>MT&# zU}M~W99QXt=?G1Z?H4!>BYlh>>T@MP2aoo;EO^zKJu144{M!Y?gNCd%Hl|+Lg!B>< zK7@(aUoEihY{9!BCiK{+&5BiqXf)}9MV_|cL~F8mu_h0M-|Vg~TKsy_fjM`ChS(&1 zs==0@C#w4hJ18DSpAoR8PQe})#ZyEJj!-hrf?_PlCr%Wc7jJN2y z>L01J?w_>;;>PAW7{6V@G|6Ve^)F64Du9L<$sgr)k z|CYXe{4MR%{wLLK{VL6!_e%52#ATQXcxCz%Aim0I8aExEn= zE1kU3gvT*IrA_U>NG*dJrB*fHrN1kb=r#VA^mA4-7Wk_1s!LO}S~X|XxMr+yQsVcQ z=9F*#CylrMCv`~rEuHM&l-_++SpKjnOKnwY*sX$!&|FlB`6XT6{zr;dZBErxH5PpT zE1A4iMQv|OR_ZJB>XRxeQ=4%&N(F~j&6w(}f``Ar**|Q?+FNSen9)>tMXHi>?VHry z{D)-i`CGcE@>A*~IAOY?#IJ~`y8jM=Rq7?_KCP} zA8C{JVj(!CrTdHU+Ej#+>SjE%wi6n(oych=LWg8Kl^=_!soqM4w3VXzY@zR*i|!?% z2fc0smC0NA9Jhn(3yZj7wHWP2$kXgM; ztb4oJQ@ESW=Zkn5uoLytjr16wk9*Q~#`fQc>55JCODd+#+)eB(+)jsYJDJmJA0s1n z@L`GAZ@+y!c0a(*fqNNZbC4+Cy+kD!(M^9Fab0(?#(y)Z7q+m=wU7Z*^EvIG$Jw+z zjt<CrN8|iZ)(HY3S<3n1JD2kOnjUP$1!g z^E0t%COswvW7=gfO=k|p;kqx)Tm~|Dg?;F-We}R8=I0bVfHws`OzY8`?e=c`tmsYQ z%w8n5?YcQv7^=0~QZ_GOm=bY03=FS*K zP3}-q`t&D4tuKx`14OSp1U22k^okt8yyf8n^B6$Ck^strgUGD+N7Kuj`0xIV`xn8# z#Uq$*Ife{1KP*Rh@yOU0t1@4zpN5EdH=HOb5)Yp!-1UMv)G<`lJOlajZzLA#gV0&f z8^^Z+q)rcD{Yb&XYUabdUS2{&*MkWGp4)L` z#QSVtbaneOKhldY5B%8P%A3i%y!cbz7qy{%h*TQHhaUsT{OwKu;}H~;N21hc40T^a z=_+bv3oi!nc+)ui2Zdw#IFQ@N!UWzXk}Ia837R+=gVJ$yu^CT}|0If%BdKZ=icdoX zXS{@7=(7kB&jz!*?|23+AIqEoaqT?E5>gpXb(;vzU5}#CYCI!6LWr{%M?&W@Y`-~% z<=Mi+_WmfARRxgTXE?ha1#u%Ikc;Phcqffy;efHc9XpbUv7`82K3VV>X9{ihXsjoQ z@A-_y`+5X!2Pg4U5yIOaqj5KkWKQ&WdcO{(#r1KdyqZi$i>U+*8$)E%D5m;^GO?TB zHAapVT0mo2w=aTbau|K)h#KF;C<6T^Qh0d`dfmpeWc_#?PDfF3c|65^#!_W7iX>43 z-&h-jL0A}GMdKJRbj|B@A{h57jKboP82W{AOl2(XwoO8J^+fch&qDL{Jks_oMLmB3 zS8U@LJ2MW$!;9E6A)4(qv-r6s8s!&J9BLQI*LG9cbbUH{!O;XhnIU4{Y3OG}3cT}p z9@K@=abO^=J4Q0(@@UffMU$64k&VOWb0SJ`*$O5Qesl`Y{>;K~)odJkPUG+RDLAbj z!g2y@FZ7+;qcu;R#kV>VwYDwgf}dQz(Gc5?=9tVf;p1jA z!kgMLK+_!mXmft~=y2za29}EdDHYR+7F1AH_ zTzIZUzO5#!b2?*Erp4jq`b-vH@5PBiW6({+!1wJjZ{kL^;HIwLDDaT6j;I-#5WC-m zYK0C@|LU=6vEXdKHAm~9IjuFXje(ZwD zE}=K9YKzkaN1kiS{M0bT$=aIdlME>n^L#kMjvRFh%;N2s+sB50>n7YRHKTZsF@ZBo z(aSLub$319dYZG`!j;N%uC(xx85}DK&v+-A$J)|unk7T+tOb@`k1K=qX*J4#az9P3 zJkp^-vkRkxJJbHLw#Yj*xzS%o;3u_tFXogqN1LZjwV6E01c!V2#KoJkE<~5*JB@hR z+?WT!HVi*+O0b>KnkvZBLgeGfYDRK_W;aw;yH>A~F z6S@wy;Jk>5Rl_aV_Pz^;7aOoXUyNb72I*0{jJl{zQl&ng2etWAtBcvfF8Jl?p|@V2 z>#4fL)d=0XEh3iQq|2Zt8e9tPLXEiASq2(BH8G@UvoVXe88UT@9y?zdkkZ$fSHH}d zJH`|(XJgE!8c{BeUs5mpZvBmE8mrHjvpS?4GGN^m102PD6LG+V`G>98{6%> z?~mEBrL7a*z3tH6V@oS7Tl@rO)gs>lukMyah(4;S?grj0%tV^8j&sIYoEo#1+h;S0 zbrHSGrnw}F9_rupV!AYLrtoed>L!JFxfY907XYm1?=zUEaW@dFy$-G@aIcQwWvP zxo|ZVpZz(Ud{%(g>pYRK7SqY1gf$CyGTd(q=WFsQ72JlX=zLsy?Zhu`D_=WoL#we6 z>q|LkI_2RoFpF}>^=Rs^=T~4hX9Xt6W<~*KwHuh-Y$Gq;7qGT>CY7Ou3{l;|mtG|d z%iPNG7R9)}E~LGG0plNT#Kvd~YQEb9k8~Ri%XXsMaudl9viVZDfxlDIagABepnEw~ zv{{G4mvu~>BXCS7GFjDoJ&x|_804jLf28QS&t1nI)Abm?T_^IjR7#3glh`YX?3h&y z*|>(JQR19irlTk7taGns(XZKB+IS`tct-T?gW`!4nj+Q%6RA1465l_mRERpCb9@Ss zkCXYcC?QnDlnO$)#>a# zl|cvd^&HO30BsdNJ`9P_YwkxR||Jc`=o(C1$!{xdhwMmdjm3-alxR6vfH z<4hr5HgHNI*_-nD;**bEVirk3Ih=|qWWMO#k4nhq#l<|D>TKoTk^){0DxK_E5N9%KFXOP43*O~aH<&u$<&-=6a z)QBFv?xhVJbIhe!{H|Jz%wc-+dUE=waqM6!j_PS5zF*6Pn05Rr-oWs7xy){xM?BlO zSFx4*CpVHQa0X?^H&Wr4#R%P8-aKE&WO)NOKW7qnaxE>Nq@!Y#M$6Z$>C&*8XK8C` zo}WwfzI+ZPi9Dt;o4M20ATLhmtZfE6My_Xr<{F}u)^cV+DrRd_QAti^yjcn-(=++9 zFpDWq@|ozfi5uo3_HVTV+t^L)eV0p#N*)?jxv1vk^4B+?LG25;|0R=o;&{!8I1dr%bEq^ z_bu+jE=gNCI7WB}Ms8*P`VuBrZlYU?4f;a=D6-Ch52Chr?_+lqZ=Jd3Zo=YC<}_2b z!tRM4IgTd$`ewuAbym!@uqMyioX4Z}$eL(ERGkGWTb(G&bfVK07d-k{p}NGJr-|nL ztuYbUa%)~~?n<1U3qA>==C-s4y48;8PjO{j3n(sgp~+-{^HKjPll*{^7{Bp9xF~xz;#g@!%;lz<*J4DrsMNJX0uOp6b zlBf@Nqthb=r{B1+cZG-_(z}qItB>6XBis}8@m_1pyfkC-3vI+RO*{{44asR^&e2gO z7_V|<|2!*phA5b1ZN~@MguiW#=yk^q6D{G>`JV$OB34+n(30JMYRrn~TWpJc(haMiLec5yB52E?mcL5omv*@F>ggJS!eC#rt5YMU9l+D9$!whzAn!<&Ug*;52#=+_d zyy-KObEBpcv{dldR!tygKqRLwk7DA)iR35E=K982?07qa=BFmG@bC#v-;`&n=p8VRmxWM@P@4P!r`ng4-dy7ipbtYFly zYLeEA-lf1*#|sR0%BKhXOR2yu@Ff*-FS#}O5fkq}5q0Ydu5GD6|MhF+dC%GYty1*n zL{H_|TS8C1W!1Jf=q;*XPULH57(FF3^9ibtpD@sA8`Km zo-?l-8Na51p`RPc8U3As{$I&2ts!w{EyHq!USxJ9=Zjv_LGv|fMU@EEIPU%VK<5_X zSegQfD2wj}*VgbV;5ADHmVSEcTB?s$aa&Vh&9y$V%}QXj zR|w6G-3`Rm{bJe6MxKX@{(HAC1be(A)2xo|Z{E^xekH|oDtPQrf%RE zXP)u4?_>V#e#($nFZe2S+t)pONA{4n40dhgn#vD)75(62cpWXKRB)!%AS(9sWU+Hk ztn~WOCc~W{W<4ng?7@mn-B{@?`YFNP@YA*Dri*synsH8;gtT9RSG zna=i9d>6fseNLqGR!}O|dfhM1+*>PC)lK-?YT8rq%%0yj966}houhLV3>YSHQp=9a z&%!%2+ElEOHpGb>a6q=H@QAeIu&Cb$UG2)82C*+^Cn^U!@iSY&AvN*r801RlxsD>% za1@vqN5&*M(`2zLKlB`Vy3US<-a@x`zAGOWDOmSS#@J5AA=H)bqK|X>Kk>XfVa+E! zXAC3lNPKCE&qFKb?=}>EA!f8*YRa`b5i18T8bKEIw7=fKQ@ zDdSxD=H^28Y_X1(btOgk)hVCo!}+?NOo?*m%4HYM^!7mGx#%A`D)?*S$jH_`*nHKU znT0*5S>Q}Vt~+zzcf&}GQLO+Li2r@AbmmEr3tyYqQGUmoY3qcR&wdAHjB>=R*oofj zo!IN@!tySz?E2gtt=jIa+b{fnhr1E}#+Bt8WFmUI;ULoG^fn3v1ciML3f>R*q=!*= zey)+Re<%7y(>*Bf@6D-o{qYFxM_1cE3?1dhx3cb{-r&wStpTEk;EB=h{Jr3Wz!6znr|q~|z^I%^@>=c}OjnK!HQi})Z2%GLdNBD#Z(2+o#K_%!*mtlubuat!>F!{rDh*^{>HuU# zFGhdpLCR7$itqL(baD@l`t`x{u)DxpA~1OFqSx(#`7;+BM=RJ+=uFu_1z}PToX@*+ zIjtwRCW+eHN)H;w^d`J}H?}TvVNR|iw;qXkP8M9G0o~aALxCjDOM{Ngk3DWIA0c#W zuXLwyXK!-jJvn0zd)xQo+8mkf;&Uradvaz=UtHR|)9HL~{1f|9(5^q*278dxVK8A6 z`lB{hjC0jsE>wEsTkR|KMgj;HI6{LEZ$1@zusgdiZSM3Vqsoiy1>U&)9?sUV0IY)o z1;-(P_xF6+wclU#@dYNV_c#m(Ph#ZGQTQJn%c!i8G`k+l65$cG>{%SaM^-Sp#Y!Hf zEyd~j3Rde2{PRqqZ7F=iqpz(I+{!f0jZULPIfZYkD|jI-W96{L0%tLgVU7uGN{D6l z;-wsJ~5uG?@@vi5=s9p6Itsp4ezQ+=6^-*#DL6feDTjo%uX+s6-^cRg&_a%{no7nIfxVm^N6o5b*zAZ z{#=UP*BHvOXHst-%R}cxYzq>&X|RH3X^Cj3CNb1k=;h}oFu8guQqx45c*j$*HlDMU z%lYk*LQ>b|WSm>ZsiXvE)-B;(RII3JFUG9*0<8Be7Ch2eu8&&AZE@Y4ny#Sdf;i3% z7Oyjw5Hd854-t#eJTafahi7uebQb!4O9)*wNAxM?P&IT89$7Q#o;Q>8{?mzfSSb2$ zG3afG;pE6z!W!qZKX)#H9<$l}zc@PYc&z^aji(`LOG?@#N+^x{u-BOvEiNn#Z+cS+#xw zNwa4V(qkG&G^XJ+c@Anz=HpUlAvP}-;2ge?QO6hZ`;vJ63l}iFT5=Ef7qax9WCiZZ zd)I6pn$~j}yJQ}NGG@?y`y3|A-(c_Cx#XUhPpeu>L{~_pU(|e}X3Zr;cRt6T&t{AF zO#0oMg+u!|B4*CyPI@8_&P_x=WG1~|#L_xFiW8s0vHv-a&wYe*(k_INlSc9J`7m6r z__Ic5B+EaCQu1#UN>mt%aSSz!Bjvb{At@)8+9RimCNJky-^utT2y6dC0yB*h$=);{ zSI;EQUtWq=z(QKpPvG0Cc?=JqOJv_f2DOdnv9f?!n-*YJw2(EzC7oZh9GgQ+d9YzV zbABd}m9MZf!3I@=O8Rmg@!lo(@ISpMl53gzF&puTy72B)e~y&7lU?FL>H{Yx1Xz(b z#U6)Fw%pbBpu+`QA~P+ecij%>P?)|>u6f5DIkG|alMVXPqC(aYRekWi)fZ#A|Ia+t z3*Xj#30Y#n`NaxrpY)^bg2KuBJsEjYVaguSQUk1cbyh`QfQmnME(|gq$iz{e^qA#J zPTxV?I^ay;p`mo%?#hoF4lM0CkS8}?=-Nhj8}Tlr)gHhYdA^B`_W0P?Gvk0g=WVQ6 zFbDp5+p_wh9rtwYm|$i_+fvD7Hn-w%UO$e1RQR)1MXOYqTj|(xEmQ7KMpj&0ZAZYD zf!r;1M9e zSd`ldH*uh>U)?#@-W7|F?xLAHir#I*zGj2a9yWm9rv}mMjvIs5c;ln*&9fmg1Db5Z zGwP`RSnVlm`+@vj-(T{F&a$2x!0Z`=aGxH= zpe~*at8pg4L$a*J;sMEW#=X!T*E>#}-*J}3u_r0skjt$6GmN%7&f^DXNNIP95iaLs zt#VE>lE=B$ESD6G(}E5>%bRtl`C@y7P?OW>m!6@O_8BVTE?^Loht9$)_@-WD&xbsM z_43hgR6y*BTqbrp&#Wm0Y#GWCzcaa zQiaXSo7^j`A|+V-WfLoTy{3dseX2O1UBXKaH~Qs0!Lx1fT~$?|;fUL$_T4dUY~7+bfTr=j~8 z@@_9@x*nvd))98jI*dop)8sck$@TiD&|54WhilHzBKZ_%MrUy-I>GqeC(-MFiYIsT z_&5F>8UguSD$Ao!$|cU#JE|;Xc!6fM{8J>B3y?KtdNrl|qbCC&$ib!c( zh5hqlj$SWk>cMiFq?HoyoQK(fJl?vTXXwdX8grWd2QF~aKabikud>GAGBuqG$qp%$ zKBQdkx#zKB-Z}0jp66uwSz1NraenZ5p1wYb+T1g29&rKP&KK#~CKprda~vu-$D1i< z={NZlK8;S0{`WAk=TD-uHb<@{2bp!_5aZVzVdl#tyy|?C!ym(Vio8`f}g^fk=-Rq@v0I+MHqkHK(@U7sYs->Mq_wKuqsaD#1AsV-rlhXnXP2A zIy8sWCdU}|Jd<@+yVx~zAJzL0(_Puc0loc})0<%Av8%G1PMQ^Q?*F>^2|g=z-mgHrtJ};~tu|-Oig$+u7Sj{$8QG z@V!$Dk1+9Eb^ELQyP=Nb6*aoLXz*@g9jv`Ihz`+UVJ9_K9jc8+XnkzT>v3a~Ch3uN zN%U+=@2w57S=o$+=~~?0)C`x`&GAcbD%qrloaxt`c0U`kOsgRoZ#8M`ropEMjZv%J zT>c&!!Z>M+c3K0%me(WQQ-ce~)R{NCK8Ll{vF=ujFR#=H-d&&P-!$mBMS~65n!Ie& zNOF)(u=w1NajP22d!!M%=NjjV#`K%NmENsCE0wE1D$S>UQ_kN1tBeb; zjb2S{=GCc--s}dFv#ihF0}XgPMV$@-4J2n!p9ah7(RNxL8mFsKez_L2H`b??X&p?` zr9UK7o%5OXaqrVWvSba(uCK-TSJIKH+km0-g=@D$lO3^YH1A)B9*x!M_gS5fKAOy3 z(vY{i>vA-&4xc+~Fko$c$|tBXqPYfIcj~j}VO@&SHE{B3%Eq>hdDOI#4`wnVQIJXHI#X~Vo97e+Vp{U!3F!`B3 z91S5qC=mTAL(tV3#zfyCyf`ulGrhq~4Dn*$B_A&R^<>^TKbjImQFRar8IgQ+j6hQf zCn;WXO`5)Ri5o%t0?DAQmVDYI;p|z9-{Wi`-+GJhV#rwPMUTP4W;6}g#qfQ|XnaP< zTKQ)++OKU-!N~{qr3^r8%it3Q8Zre#~qVk;`@xiT+5%&FGir%cL+Upjv{1%ALR#q zdGF+p`MgmKH4GsqBY^k0kr-NzMr~dw^)&)mq8rAezy8Ahi9~C;FHW_>nIV3ubMGTK zvwIx>DdX53I+ddAa1z(dk{;uU;wg<`l-mTdO+=G!5X-d@;g|(Z#GpdF8-t?B-#%V? z7RE>)@n~}IM^Sh$l!I4e$VrH0r|8^!BF3YiG!{iOnm&1wyZahJ#=TL}kseAP4dF9) z2qvZ=fEkISWZvtK`uV|_+!?{Fy+MTU63^0XVb68t!fUc{rvTk27buY;G-M&n(2A9~kA`dJ5 zi{#WQWN^(@8rM`YqF*_a-L4a`p@Ks3@~62Ki+)i`?Z6wdUc1AkJ(aY#zDDJrJ9OD| zlYJwv`{*VOi>f*Fs0#J>l`Ov_tRLH}l-pb(%dmj@Kg-FM$Gp6uoJZN!BtNZS z*xM?$46J19zCt#iyNLS6JTletsONT>ItPkKXkWr$^)ia`Z;Fn1hb4Ir=~L@AAqTG0 zU;7rvhgD$dRz?fEGR}k-itq0-V+*g+X-Wx~-`*g$$8FRT?$W-uWD&G)@aXtWlI3?d zHoeWE+Lip6QB6Pf2b6D;Y{vBm6c*nn?9vUU7nrk4yO((4j9LA|h{yU}_`Xs+2}?R- zHNu!bk;a(sF~DYfH|aI(LZ5Is<5qR2%~F|t$xLrjAL*KXp|E9(Iodfet-2qXf6auQ z-JP@BOz2u^OvJ~Yl5G`YPJLr0+jhg#yc@-rWsc?0jddIJF+b1&HEWr#WgF4?Upr3q z>x6z!J-)uvr+-m5?5&MR)9c1|>+X^%cc9})8?5AaUI>vrWs(`@D|_?G-b{K=$ zIX{y~FkeWC(L@3k#xqs)Ad{oh$qSF?YLi5nrz|G!`~tc^N#Ngrcv&ATW~tMBUW+ER zHgyIb`SBPeB*?xqj-`(hv2QIrVdq4yMoT8kK8}z6Q}As%o)x~cxndfROUpQ(7tQ1D zxHvkVn8nk-bNKK31mU-cX7(ik_mm0fwU|W0uXxeN7tr8$9Q{P2oU6W!=X(~?%zQah z+ANTn{ye%2UCh)jqI1=bqy4X0e76%8L!*Uq?ODM@19>fFapD16%W^vQ$beSFz19fxq^NWMnTTHgqB1gw4=5GLi7iL`u}>G4AI)+Pzpr_Uz?6 z)Ln*&+Z2jVOl0E5M4mLBNAFs*u-Kh|>e2%4iPkmc?sObiMpGqPXPf_~@}oSE>$Apj zYv^>`>dj~TPGOVAEnvdQg`A(VOuT!GNSm^Ri{|t3j7_BbZQ&h%UP4N9$-3=ZMn&ax zt`A?xN8vprna$0TZDP8=vSfx=d>El#?Vpx>o$ylcHZnN@}+VAk>Z^mhL7tg zE~fz7RaY9{`~A7Le$*|c8`wY@rBXSnGnU&tg&S3M$))^91})_N*7u< zO+@eAe0>ayIkI3fOXR*;H!h0h6D8mLI}+8fXx{A)W!Ic&cAX61 z`iUSmR|Sz77KpA-AZog!nc?k^=CzSL{^QM@+oRBQ4q%dQ5Z&*G^7TPDN2Y``yj~C) zeZz4&7L4bH@vIAvqQ+?~AzqQZm_3F7lNix7{8{&R7<+_2U}`s%7Y{^t-z(l_@#;>= zA5KZTQ3Pxb;PcjDY`;51*h_;k{xO7i)n44{JDi)jBP3@PDD!=JeJB0MdLBZ*&Ot1A z5W^~qar|+MWFczE`Mu$3~?9?1dvl)TBC>zS;yCX@N zGK${4hO@&`UaPDJUf=O&NXJpc%^%KC$-sBW31UK*2+~?jAnJM~p`tG?5iY=gU-D8!xhdtK>2+!armTL!xI3aXoQ9yM2oIhGQpm% zq9OihX~X#Ewp7ckFiGE)0itn+Y#qQ+S08q-kp2xH;ZmP0?vEBcdj z+LfrKuADY@Wb+qcZ>Y=h(|2G*k`pUR+-cFnm8zDa?TsD6=Cy<5Sh-^sI*=X}K5X7H zkch%T(zW5kUONxQdHUdMTusXTQm)l6gT7bM7*@vQ6_3t05Z70zESc|Y>_ zt89)(a2qE&pK^ue=9hTV`!ef1t}tbKDMRf`anriNR{I(%w5l;ky2*j4 zD#AqLx=>Y0Z;vu23TLu%bQN_r-yk$v);X>>nbY(Rb&g!;ZIA01s9%>1MJ2Ti%W3R; zm1bEL^!-teVf9Vggx|t_LKUMjOEHv>6#uK@*DNP@N1?D<%Gfly3ZI;6PN>&#p`wC* zCB+o|y+m%OB1X#1-J#nRZ1)$TC13kKJcrxBJaW2S;;ZpR!n|`?vhS?0OY-EB*GbE=%U9GeeMZvRZ_aps${H%4@h%rDBIz%R;I<()F2$0Oy&j`zx#<6o7K-d~l> zx$hMXua8QD(=U|qhA))#_wSYaeLpLaJ^w0AFaJ~WKm1X)ovKA^pD)VO18`eINu$@>G3vXYW}^-=VcdRNw%vnIiFjiB?vz}|LHaMt!hUQvdnSrz)?U1` zE;~5voyqvQJ6I!+rB=98ICxvpIIxqy8ryksUYJP*yP2?K6E7BQWB7ULXjqyd{Z`u< zAGn3GHQP8>F5ITQJ2(`+UHFhY`Ex9jjea|MbvaX*DOrR?Y~*Ug&1~Pak*cZ;I(<)P z)1gfmPu_)x=wJ0sca!Y9UAmgKkZib_;`>`rRc)kJ@2!+Z?W5Q8EHbQe&}edyE(XVV z-$gQD*;^UuvVlHt*K_*LW|qZml^MUVnR>}{j@n4uoXyN`n#Rdf>-eO(p8gZo;@EaA zn{3w6ux298TYW42vo;IwSaNx%x1jT53r*W?5>9IxbAPU4>%@(i*op4zv;nQLwN!^y zf607QOV!WgCu3N3}(I-v%~PQ=Ji(hNbRb&UF0E@XkM|<@tlS-v6XG z@(*EszA--dAFGG`;=TD-eC~>OY}O}U{`t{2;abiw z(pUXpRn&KOcKOcnKfh@B{U?Wd)ltp(P*1h7Kuy)HXML5|3k_AguDWXd!MduKIdxR3 z4RuttfBs`f$#1Ug`o`sC`Q71Pu`K(Z+@A8>t!^H&n$eZlE%c)=-TK(^U12t*4siS5KANuD0s-g*qzT zWA#)I-_=qj$nP}_Q&a7H`IqTmYN_fZsH-O4sI7`pt1S&R>Z-K{^;A8Q>!^NC|HH&b zf0?86laDih@%j5#?&YbeZn4J)KxKxy6V1DeO0m? z?*R9Ps_pL^t40cEroC4S)t-yms_2Q0RgGq7sp|f0qN=K=smiX_RB>8e6;xGERXw7P zD(j${s$x}bRh^%;RXsP#EMl&)C+l zuhN>*Q01?#p;{upcjLgis>5pj1fB?$yn@WA_lGbeZXCUBqe*b_NAG6_KAr(Pz4NIwp~C8OadgIIZXs z%yP#d^wTB^OKOzl^21m>G@AAurtqRo6r7Slx7S!MZ5boIjj>dAh-O`fNW4lVWA#X(bNP7a6-Cp-v2=MBM5v*#&7Oo}d2$3r`vS@E z4&veCD7ux4Uq|*x+h>i&_j5GvwWIMWp2UtF@^9x&>CYv8K% z)pIK2p0VKU1132?ByY)0Of_!Oc6kNqdiSY1cauBIZ}ZCfF30*lVDyy-Oy7PFza0;l zn0bd7-@EkKSk1q?B`gcQPM*nq61LwWiR*M6FEg}34>1$IM!#{QfjubX+wweybjatL zPcD^i7f@e$iEEe7^ZHo<%HJY3?WmzF>o#3BJr*z1P3h`;ChPp?!Wn#w&82GbUfp2S zyJE7-E4XHNoenJ@F!055j@=V~r~hMS-grgmuaC0-d4)3Q6_K}|(|W~2$@o7cFzY5O z-&A6eTfw_K*O(RffNdq>JJoxM#p*8%UnyLk?%z3i@~dcFpEz#u0$=sVm^QhG`o)J- z&v?SF8IPH8So^oH-is`}~$tHxrNe_nOjaav=6WKp?ul3A+am7nN`$c4l)j0_Y5G*gUYNf}zL^E#TL<-%f@l0(ZI~vX;rs-^Y2FxSp@+9W|i)GikSiGX6Xf|&gj&DY@ zJRpPzlS0usBOUO@(Y$^=Ubs&&q#2CI|HD|G8cOaocRFvs&6EtuY#!f=!^2aai*QTN z+Q$<&C7zN0O-KF6T=Bh6_@q*Ga#db`5emoXaqHIZyB_LqK|K&%YIBo8`@H}ZPxIR+B7ZwPN0g7PYq z?i~ZkZx$}s-7w+bPo++$$yD8&%Sqq)Oz@h=(#$DL-aU0VaPGN`BV+hj{^?BQ;e=Q$j!&n5gDG_VK8edGCvYoa0_MR{ zY}z;$%?Waxr$n=UQy4?+qfm=l#?LN`nGv7Fi=kao)fmC&`VeBzw1LEukCN z;~Kkx!eeQuve$8LQJUx@>$o|09rN9j`TISM;QJXA9p234`P*gRyqkBOc2H=sS27Um zxch1?hkm3BhjSyVvbV6yFN5@B>zQx54ky1f@rbPE^QTlEbj`qjV>-_&S4f{iGT(BR zFlq2|7P=?#%~W{kmMQdzT7gb*3fXZDA{{>t)n1My6sl3 zC4BBGyp1<8YSc!kora5N8uu;>bImb>Rwq{Bdn27mo@;4wZ7a8AfAdH(c+(u(QZu+c zpB?o``_hszM_TZ>Sqlo{Thnow9zG}Z*qPOuI#qgPXtl(!a}%cZ)sYOo4lOKo$ZFM* z&r4gezLPfdk2S^pK|@sKP4Icvhyw41%#*qJzgf-jo!5Z*3mS9sQA>8WYbknHYi|B- z&3^xugd6BFYENta`RZcOUyp`yZCQ}7%lx|?aq}~vPG&p)9&3wT&z1~3)EwUn%_z2% z&S-NjjJ9f0TDO@z$A;qh)FQ5}E=!l|5WA!k5f|EE>e)u}*OK$K(x&%QUCGdB5t`Bp zm9TcM2*W5TwH-g&btWoGpFN8WSrH?wov4mvooGX2_13t3(n7mkYYP5q(`lJD?|y2D z-?16iqcy2_rU~{Q4XK{ogkiqTS--li%oE$-(XJD=k2}$EhdyJS4Drr1Vb(}P>R*+- zynk0xt=jVZTqnsI>d<$43o6fQlNqVY5W`k%@X(<_b!*xU(qZukA-KfpvR7$MZn_?G zuj>)8pe~I%6cW4B;8AGve#BZp3~zmdOPa9bdoL#Yd(H(U`IDAxqezx zxCDA1?k(CfOnu*zzC(HlbJT*##l7h(8HV1vR=io#pAVt#{2t*(u4v95oF#Yg){YNX ztx=U);_%RhkUdt|#VXi;bK=?pH{6oF=)1^6x-?w)D6_5^HBMBtmTcitNA_#k$=n>Y z&h}$(V|QBK^`PTsCm7?-nG#o8boZb{G~fZZofw^M#q1m#&j0DpR8MOj%4f?J@ZH&p zRo1S&GjkJ0tUCss2GBXfjWeTM_>tjBS!WyUXWDYu))wDU3L~Fe6IkgatdRcbX(crj5j0d*a0`8dQ@cz!NaiFUjq#+to@9I5T(Mvo#}@!kB-E9%L(`~i}2bzpxR zNBnMD(SDW-mBIEjH*#j3y%V9sZK=1-ovVMuOB?9S!CPL!^Kc_!Xm4t|cEd5*2vhy; zIK`S_c+CP;TXU}Nv5;N|Q~vvDLf}4`r7rBl)SFPOCj5g&UD)|hSPer>WoF%jlU+?I ze`(0RfyTtfm{4sbdC<5%;uGjaZhLdCH#8>P$c#OeCT#p{!oncQ+k|&x`E?8OvV^7P z-Ay>qMrg=;#Pq*zwC>xT@a^3h_@FOm3(dLzuRD1&BoCz91;=%gvFzIu!-p~(mdA@9 zF8xfM&1kmKj9bgPG0e`CS`lWl7VpVrYjgaq&A9Zv8w)bc=;+f+-rI5x{x;@HPa_)q zYtP}^ZE0QF4%KrbeQG8c@cM68Omhq|?AwJ>O?~oO zN!IAB0mIz7ir-dpMk;e={W2%1gEb%4^u|J1jn# z9CqzGipQS)4ADGHi--ewecs9M%&p{%-o*KiyJ%-6jP>$WEM1jGQ_1u1kJ`lPh;$xJ z$Uyf=I)_9P>%4Xy`+_!8Ynf?o6 ztG5u6y9clJnS9%S5RG}eF>^aWfw0s2HatL5^SzQ?+(wS>E)JgCLFS)bIK13Ty%Req zUA#^Fl{>h+Wj9-N_pruJzV5pXt;u_E>$IIEXSTAhdJ9SGv-q$flU?dN==Du}k&=7= z-D?}m`)r}>`u(hq*hR-ZySOHPNW1%4JlVOAt8uGdW!iBr_rz@IjJy~YKz@cd73Myq4jG8nvcddF9Zq`1Fy_?^$7W*?whaU5CHdY5 zErxMZW@5>DUYs8@0Qa$j>3nVwMRx~b*>-?%y@eZ9<%L$fD~~=8KrO*bSX0j2-Q6_{!{2c>gWD!*>BDKFOq#9V@3B- z;^~iZq*=8c&F(mm@Wz4PcWv2u!GYjWk_}sJ&-O`e>~ zLFji8j(nu2%m|zrblZio`<>XVDScYrUO4Ua#_6FqMW5X9JTrhMem>m4>BEpx;d1#8 zp>e3pI{h8R@8*noMl9=Bgm7s_6z;dDvsy#;UorC8j;wWhEwx=6cts`=x!QM zMpIwb$h`i{nKXe{QpC)40V2(Lvi{C!%u=OczQdPflP6MVTkdK8xDf-#Eo5xz$ViDyQ0D|sME zCPCs09LeA8;S~RjK>fMo+*X8B>uw19<0C`|kEC~cEN#Y3r}_Q>yjMl>Lpy->YlC=o zAeiqd5%m8U!-OW|F*!7jYaJ!8T^u2urxBbO5yZZqvEnxf=kCC1Ot6ojq*(}Is{&ZC zIE1#sSAYA(U%F35N@h!VXgPu0JTQi{bt7oMIEdys!L+Y4mgT1+F+DSeO$p(6EFI01 zKhcu!mL6@%L4DSnEPh$>8u|?-drSlq9HQCwGlt@I!)ZDrn1Yg_d@Bs0b7nA?C(AL* zpDdgx@ePW%a89EM?7uZt_JU(MWh8pyI`L}G2<80UU>^MmX3{&+F5<&kpBKWp_+Uo- z@~6V!e;>YB)t7zZ;&C)En1)wM2zj9q{J9!M$jI?5zY#{mH1VxgXmNaRGmMN|GPq7F z!ZvCX(X^FRs>qu1c5BWxYQ*8LtuUFc!z|Cnq@*_C<+5gMiEPGq@s-wk(3q~?^@)qq zl6_udX8+Kj`dTAO7c}Kd|K_ZV7S5u+7Aby>2|3x4TJhRE{-{Y??Pg5=*^ILtwCFjj zDLx0A@vWC8cY8EtOM^z(E^j2fl*arV+8E`BCh_fLtsT{Xys=F%-`R+^hZ~WaR-gB} z>eBD3$-XfSxLTphjJi6su4#g9gm{rtnlmc3Ijc8mF)gVfFPmtveT_PIpR2PYz7CHX z2ph6ggO=+WqkBPvAMdodFjbz9ZdI{mcNq}UvOQtf zjkxx?6MDnseq6ULtA+P-{IWJa>sm3gs1;c*ErpS*gY)=i_&se%dTw*H+v?zbU5C-b zo0HnCIgQPk5-Yewb*;d)(b z#h5$o7`Rcm5b}9llMt?U3}(mSU_wtxhe5O-lZ=A-cq^Fol11a7cwYKPuu`-u(|(gl zJ`>FCgW>G=8qEUtQM5Z3z?$v8c-INQWWgxLYK908YLu|9LU?QBN3}}`&5eTySnH3j zV*n$&`QzU&h@|;LXdNoovPHvjj2*^^EEOy1W4Y-tfdGi^Tt;)2T z6wN&%@H`O3`(q(IlXEDi-DvV{!nq{p|5qKEFX%?oZ|-RFy;DWUN*{vWk3lrKHH3RHBP4%1f__?&Y`POBzL7A}$BK42 zNP0AaV}%qj}dSn5=q{IM0owi*WCPkA$;rUj%vv z(rvyuR5GYz_-GPA=G!nDv>nIymgB^`HxXNdSSF7fL&@=l%<4Fwbt%iZ`g~ znu1o)DkiQ>WreQr$`384hSjKrFGo9hF|OB_p&Vby+;59%a#R@WH>4Lza*^GV6Y(Fs zkU00bIEp8&Ui$^;om$LL7SeUxBKnC3Zr=&fA9pR{?CQmADpO9nY#xJu33mvHS( z0(1H;W}o^}_EjJtQ zT*OzuB!*2(;*78K%I!<0^QB}OZ(hx+)6#*Ll){>NX-v~hl}W}bv<4+JvV1xES;B}P zoJz|!snl33VYA;7l*{wE?-oz@*^AiyYawYXma=Tb65%B!v2E2dT3=0~U4s?Gu3bt( zqhvHDPBYwIV3x~8jsXL6=7HYyb ze>1L`_n&Nk-u_%1jZ^=8rHz8IMFVYQX`Wj|Q3WM?l%8mV}7(VP_7 z3%+dMpO3;&=#e4c>!4zIswwWl!r}g8%Ew*3XlZCJ{PM2id+N@mQ{Ay`ELqeTD0gCDlB_0G$_ay9{FAuH`;!*rC$ws=-`?V_Er1dWM}j8w7e<~d znr3#)sb|gUu9nnnv>{u~jqCCpTd(o}H(|3j^JINDcg7jGF>{G6^)02NHN%=_FI<>z z=tQ#UQ-{yGQ8r)noNLn2nj!s9^WDkp>CFBW9yB@NO0Pg43^=8 z_vChrD|Q7Q!hdq6;;#)`SK097kRu(|I52g+J6)@U_1<$h=Ie&x{LoW!@~%uvb|k05 z2E$Y7IBXGS*_llodN0f|=PmfIlD&A)TJiO-LES-g_C36^;Aq%( zRNr60slls+Ewh5xs+CxbP9;8n5hvCxV6*K~!mll&!H*?G9hQ5+uOwD9TqrtW5=-U& zQcq@$Q({v1X}J>D>Lj`?SjNsL$&`(eYulr>B&wz2^(Ku-%~kx8{*ThBE4gWu%ycE0 zA)@;)d@c;T1u2a0u?&~@YZ!Ghg9O7gGB&NH|E_fWPHtf7nKcYsl1@99blKCcVOpIu zM%7N|^{5T({+33I!ZnoCm)XU>wZiO8B{Ws^hPZWXGG56^{ney*OylXc^_1DCFt=(g zZ~SEb(Rw9ScF8Q!OO=dpDrMe(ETtedote$n(e8vW<<6|lCCr1 z%%ofMMtDa$iuj3h9N6{8fh@27WVV(b=MQo}34tpc9qGH=4z*x;KEv(V>+HdX37&Kq zAim}e(orqvYr{9rjQZrl@rVJeb+ToeJjPSe#m(a!<$f!D43~Pd%e5DJ?WNy+m<7up znxM0-FU`-Y_`RX8blF(4p{^y@W+_-bQL(3)17<(%*?7~H-#uKoo#n`lEiQQca>jP2 zD}9Y!`Fp~ihF-4xXKhEv)g!oRO9rE8a~O5TI>NJB43rCY&eB_KOj85Vj$=`^iCd2N@kB>7!dezvVq-qbErY|m6E z$3t%@X$xzVE0ij(;Wdi)3VCe(3rc9o1?6ROiBk7>l`_-2R56%Ptc1?2R@8nLDMsm6 zl=dFymH9eFO1tvT9NmqJ_N zE%$EA<(!RDg=)kYBo!RlSBYzUxaxl(_Hp#}AuIMT=CFux}$GbGY3;n+J zX2cOwg5=-FO)#NJlqp6s3-NU@<#l~i+P#(aY!_2z%rvK6*M9W5+L!xA=DcXGFeu2J zyP-y$k1*!z3PXOUx5BekXBK94AyLzq_vMDHon%a$p&{L!4Vh893*YAJ^Z0mIN=#+$ za@dR*naTZEu8)O}0n1*>S~$ay%*_TkE;XWQt{G3qn-Q&JAs!%MpPBTfQ@8~QjV*-H zZ^3HO@Nc^HkiCBg+#a`u;T?nlFRbI4vgQnt<5naZ#aGF@$+gBaLcwohAGYo3ho4Gj z3k!SDI#*lem!hru;-syr^Pi5Y>Zpz?O})A5!!YU4yVFpW*}kc2+r0*=HIPto2}w(+d64BLtBhpJ|xD{()gK5s{pfkC&mO)1XydgAiF}( z^Secu*?$vaAO46jGd&5W8Y{$ZP=>Q$oCMqBB*GH&MOpL{VYXx^KjjGuv7{$LjM)pb zH6#2i_l^)-zE_Afwh6QJlOoJnOOSHl1X)|R7!y7q#oV@tvLg$`S-=fZR{2GcSwu`@ z_b*Ok?b!k>@T>saJ5_)^uAItdq)cTY>B218d>X5dnZ|-{Phnxze5~7QJnKC{8qEkk zHZGZ;%|A1Z6_yFI^k`w0s3gQhszq7PWq$T@59QGP6<`%&0<2zDm__aoWVafGnZCa; z6LFPbEMhuST`$4P--|QXWJ#9OD#D)n3NfpEA-35=fH_5ou;va?c4CtVYr8MT9DAf# zh@S*|vRjNPn#i(wSL9f;gFJguqR5KdpZ(AhWe;~zZo)Ww)X$lZ zD)Iv9D$Yho`7G4%P>=l{tpy4#a5)-i{$YWP?-odqWSG9w4&s62fw^dd#w89&pFS6+ z)T@6hW&;PhM-`7V$NU_UNX{~W^b0H4_F7?6_Z+CC%|+16`B-shF)9nGf8RJCwMVT` zLu-NwB{rzlv4lZ9MKM3OK*1FY7_PUVUY0c~&1XVfjQl$O3-QE%9z4coBS?~Z>A^G5 zdc_P+7a8LWGlbdJr9C~ydR?aj$vHi324so#*J&il#LLM z0rF@r>x-biN)&RVPNBXn3}eldWxbs^Xq}|}8wr5r0?Gpy2*8v?e;hhWHAH(Dd`K@8 zV;F+#A0vnd6oH2~L%@3`5UcouP*V{BgQbC(zJ=IwI|EQV_c*TI^T+cw0Vtk9j4X`+ zB**x|fLK{h(?VgS9F93dAsEmIg(?fj_gx|Q`Xw0K6+_S_83@g;P|V69@9A@IEGE9& z0}oG(YxBgL_s7t1mX6aae|#7XAg@m_gq9z}QQ0uO*%^YPUJ=mC563C$g*|>23X#MR zjOGL(LM0r}19ANs-#)7lQcVVSAxyHB*xSx zi88IN!t9iY5L5wE5X++b z$~%8iW=XFXPxq!2eF;_(Hk~;=5o0X}DDSj$I^+33Uw-1uOj3*ucM7ni=Tn%i>16h6 z*97L-JC$`CO=Bv}{OtK0A(mW2*UJfFOh0xiyQ0a@BwD7iOS$|^q+5u+%MoL{uZXZU zDZ;EJPn7XK6=#>HNixbRVBfb)XIbh}ENS+1mik77ZS50f+uOuh^9^Bk&rF0_mQQ2j z#p(X?NPz8^5g-65KTDUG#s;D$vQV9gY@XwIW=6TCT-9WDS#lCvr#^+59ilIrDeScI zRL1{Qm<7v;u{m^%XQYcUnGe$JOoj{_ek{Xu56H4XPMXzJ%P{vT;%ul;oJ|UoWE+o5 zu-9GU?Co}OR=`QJjy!R;KUI>&>d`j$Nw80j63k^tgq?B}Vj(`ltZ<(o3oZ;J7RqTj z3!DYtw9Am9x^=ex1@uh1fa7yd;p4g(d@G7VAx{`eW=Ej3B?y|^qhTO=3ODsbsi#f7 z&x#Q z0*Gbd$crXc6@7mG zJB!xzb8rto3*H-H=(Y<&=Gg$O&I}?ywI8-gAI0u30od+JJ<-*t@Q`|=fdk=?bc;Z1 zQZTH=0*UkHkNn;MyxSZAN8eCv9tk8LY5?RagHb#&5H<$>a9$V!X)*fB526f5@_MiJ zg~m%?srs}h5EYNz3Ng8JL*PvJ>eG{z1^!2MJxmK+Geand{1_7byCCvXPgjgK?dk)9f zT!j18a|l0h2`(Ql;oQkfD0Qcd)>#I45M_ZSq&HFAXpZ#t#IlpsN62Pf;;ERRpwt3; z1dXW2sE+|DePYOIgLBlzhkL5%J*|prDFdvZq>BO4;WR`TQZGp#JNqeDLXg(g!FsS6 z)WPHnw9b^%#w|Z1+$lE1Cnr55-PMHZWFs8$q%1~50~|bIO!?qu;Lo&zsI4jDw;Dj7 z`kKK4W{^(M$M!CHJQ`6#3)KfNG!)_XOa8jLPaE=%WVK*r?F#SAk|X z%2?*5ibG4~!AqGVhObG#cu*b<5^`WvqZCKTW92m+Y%kJBuz>-hn+>q-x-P6#4X}HY z9%TYh-+8JLWJMVDjIFRE!5Ct;`dI5@0NEnyN2lvRNJ4`=r)pS5uX{9E3-&si_&|>x z_9eQe!;|WWS*uhbu*VQ4Iy!iB!3cSmjNtrQ8wX+y;kw0$@*a(` zWF>LmolRjaV+8kP(pA0G#nKH1Xb;qYQ-KcN+Ug*N`sh*JCJ;ERiP%^J*!`ycu&3ks znfURmsCVC{3U4C~93It$Pc8KozUY%q&BLHKS~UnLsYdVO2Ixf8;=iz3s9Y(-;pQ^*x);OH zvjlO(Uzzl_4CUTcFr8Y7u7WZ=O)mqtu>?E`Wf)4S!OQMy1ddcxDUZ^T2xCi3$)LhNH5c68RkAfgHHK2_t`#0pfl zG{DiT4r}&QL4839A|6#@mT?WnZK{FJ*G3%N*NC*E_0Zm2hs?vZq@628?1Kt4xmLnd zp$L;($z$PIj&ZAMu{*yK5ndgb^Qar={5p}9HHZ(UgK$tAg{Z(FuJ(ULMEe&6-0H=W z_ zny{3(a6=F3VLqh+{XI?itNa1Zu9MZ#7p?Z<@ZE_+wO*+>* z9#bjnViNM_O~5}TK6t&Ii1L?w$eKJ2S*lZUPle|2DdIR`p@KJKl#dy&fIT~v@ijsT zIalN`SRh6EpXm^`lEkEHIcQ|b!s47HG~5;NNKFHWG?nnrM+Kd~l+pS^5sM0y;G3+7 z*FCZ*&R3(kJ+WvOX~Jc?22Q#u!jCkxmlL$${gHSzmo?#}rUG+I%J$i=h*zgnad{7M zY9v%}?E>ZXyi&)l2nG1`D3I1%oVZJ3m`|E#o$Beh(<+0^`^wm{OcgIhlwi1A2L5WQ z$ktN9C0jL&EmcN(qAr}XsD>}r!6ecUzVM+e&=(rW{6zDAcSZa-FN=Z>d3?M?8ATpS zuzsbA+CPfezLK(#n(5f`DB;B)WytJOLf;(~taDKU?>condMXDWI~jaFp$3yZdPrTM zjeTP}kO(!xeOYzN|4;|77-dE|YvR}l=?ZOB@gYSOHv=>f8KjQK&6@aBtd8MEJz{_u zLI0o@vhNtc+DZ#7M-QsKn(!>r#_uu>SV`)lV^9qd+PcW9FvP6wW;CBBW_qasR#!{F zlb?>+S4CJIQG`|!1vVa+rNt9=onf;@DAPDDuf6*=9O+Y?XrYWl`eui6E?)w&AK677NSZd%O^_ zcIn*klt$N0S(r|ih0AhUA zkn*ZTlp!#riHob$v3;Q`{Aq3)p{9=NjanGDPZ@71@8gzYO%J!Bu#!orOjl)pt8QpuFRYo|q= zB7OX(dsJ7dDruk8q1r&R!nInESWCLx>_%?y)NU?bu$en2*TsEQ?Be8iwsLkE9h}9z zF0OxX1DE-{gKHJ<;y$VMax0XEIsK3!F2UnB=QcXT73u%rY>s{7HeMg(gk6U@&1F5D zdwmZ#NwATN+|$IJ-O$cuU;fH1bsOYv92nxhdGv7e625YQ`-nMkwuciL=;QWYZ{udo zYvImkwQ^@m+PIV3syOTRGVWh(Dd&Ehdb*1%IHkB2Ze?5}m$t2!+br77c?$G!`~Uso z!WF)A%;y_-%Ds#8UNg*{srtg%iVty(8a-Ug@&?YM;yX8_)y3%uHE{l7?Ocy#1!sSu zgZoa@c?BizZ4srGhBixY%{ao+fZ(Q=;-<(JLSMJQp2CmPcjWg|O<6e5Uao;KT zR>i)Xd)(5>xtaBF`;vyZGQnQX`Pg^vVOu|!M85o|Dx=)4i(k3k%t0=1_b}K0_8-@# z&Wi&gqg=$?e_Unr53W?YpL;{R0C}S?+~?9?+>7|%ob-}!+}mZHoR?}JH(q6sTg*4e z9gAw^)|!6i%-Z_6S-PE^U11kjW75aT{Wri_B=&LNg+{omAOCO?9sjrhsc|@2Xh@S@ z6X>m^9-=8_W&bcn`x7(tcN-!@&=6%A+Bmmf1=go^u{Bc#QPPzE~ig zu=+;pwl%s`ZyKQamI3^Znc@v;w0|xzL$;s=1kPAeJxx6QGy^P+)5o!7b3BeQ#@RXp z%%gQto1+ePQXj-ZnI4N6Bk-vnX#fpyhvra5BSvVJGC{^9BlK2q*msNJ;AC6Un9cyE zSrd229y8(`Fl+rRe0^?>I%hjLs#EWUwC9T|%yCCh2LqdQ@XtycPYjKaS)qt)!fKRb zYyjI@(o08^7W|VIW^OS=wFmutn?9=bG$E|0iA-7>tNztRP(9_ZjUzU6kQug*8RGg( zV`6m}W9}a#oQ*TaaRq(MG0?^OIVK3I)rZzW>d($L!TisLSoFaV_j)WzZ%(>dVz`_v zWstTvM9XJ$bZcAStb-m(iy7v`S>tQ1JtC=()=B%e(Ao}bmRiB>JN0~PX;-Ny#Ds02 z?`(oq<7dF*fEC(b7(-_b!!ly`#GdC!OKuB^-Bys~%rUXg26ucc@z%i-Hr3XcH-85F zq?VwHyaNZeQ?GB*TPmdS)%ljT;d!FIVhZi`K{~Fo9p5Sd*7M7(x zf<#yjx*c;V`zaSLu212@lY?VF^RW2jJ4i` zDdoXd!smTCxRRpk9K*tQ@lBua(NFK=YIe?4DGD zxuYeJxm}J4;gvAnn2);$3$S+E8@$={65pR>Q_gfAJZAmR$NLKF3twVj{|hYs_6n;% z6vAux1Da|J;raMI))8kVYOEBBQshtGSd61{i1~LZAGLC?@bdQ?OzC-p-qd$ENo(Lb zu_BBo@2q)EHDxYUA}XQ=={?o>N-CW9rj5u}sKfM;a)`VshpkN!&MhbZ^WO?$O;=+6 zt1>(cuEGJ0QuMDY!?d_kJd`eho=FLgET!Yy{t9Wn&v82HIh1Rj)4J#tq}$%$#_($_ zihYgly!V)@`UVHZ^5OIJJ-)qpiy!w2@nql~t}QNxw8kggr1$K&T0S1Dv$wcTQvbC^yf2kaIga$OYDSbF)tka<5Fka7XfqSF?YZTh}_m4IUZh zCdd5ZEQyg)AUwvEwDEwa{U>*I%0&39jB!uoe{o5NcrbPK5O*VXkXy>%&skOtabwQk zIOnTBxxd1sVO>4O#j8ysp5#Q>eE7$0d_Bquz5mLsEC0sj84hzXi~ewW9;4hz%0d74 zVVJ9@*O+4Zk21^0;{x9=Zj0o1?#cvSL}iY0gDoT6sL6Qnj`AX7)-P_?$}ij%n_(_L zWjqqb|Kh5Gc#x6!n`>+N&2>%v%SjGSK=lX@_WO*-fy;k5UKw5pXz*a(jDMWE@*i&B zs$tGB_$#+)?q4qa{doM?G9LG@PJ-q$UL5b3h*5zlXn8xv4aoE0%70^AzT_|NS{OY) z-xu!RGUCkSk8)D6JgB%xJnS0cGldl6Rcj%tgG$jkotWG??+~X_fRj9Vq_xh&9FaFr z&ZJs#^Lr=?y+*2JA^LO*aAoZ~@=g`P#pWG;))i6z{{v~OU!&w~0doF+L}gJCX00s7 zKzcDI?JL8ccg1+3Rfd@K3Ow;D!?(%BDAB0J7+*82)_lfls*$`x`Vco`5dRMT0RQnX zSRC1jabca1G3dl`vLwY~??KhJ;Zf_kHcGhEQL^YBZRbypZH8!_2 zLUv{Y^*|ah@Vgb;|Lex%sy1SRH)E^nXQ~tH@p@|$LKhR?eR4H;i2L-RznU@!DMyJu z`#ZkXA(WU=+Yh#&sInAUld5reLM_VIl%exaDOg$&u&)@!>BL*w`T@Mjg*d#a0C5T> zsPV0UepEGF0%~Etv>w^1&2Uqw!TY{i@XoA7_^(>zJtUUrQ{sI$m0|1R3hbAxz`mQ+ z)bFW;Ffr>7OelrQ?+TctRA9VaEw=oshR@a@<(V+!;->M*estm<$70Adg#>JNv zxNKdFZHXV@82)WS}>5}zC^ z!9&;OjNVG*hR}7*pcMO6Dlybp4XrO#cuKW@zDOg!jBCZGi^OQqr_8Jetq6TKK)jkh zB+l=o+`R$hKKX*f8sD)YrU#d|eu2$cH?n86;G<3#W@WXbS+5Nr>^gC3-WTk2_=YP+ zg9zF22N&K9W7@+`s2TM`_)#--=Xa3jq6&%~B)g->ET8Wld{%e`(~xjq0VzOQ(l^$qJPx-c`U1A>i|^VQ#o z$-Ap5XQ&uT1EollE2o@nV(9;=!N;OToOkNRlAm3e(M^A2){UC8--v}UfU23qX((xc zWJepm_%vbY>Sx>?ZNMDeFYu}Ufh*iEWX>Lj{>>reEcyv=W#T#9A%??94}?1@56%8F z+|hzJbgq7Ksl?@>3d(XV$CK(>#PhYIVsitGs5kP6ryWXf+c51h<(Or6p>At0q~p7= zxM~1na{X|0{f003zmV%UOd6rzuxc70zR_1aI!wQl)&}c$-H5!>OIpAV*oO~N2E}KZ z2{geXtsUe4Yr~Dp#Nl-ze#Glm*u)wTM@E~O4eK!rJ{>ke*N}y)=&-&1+RRQ$nP!!0 zOtn;rm@>+2qLLC*|DnOwZ`Nep3R*1gvmV=Esm(0b7&57!TI{2;ChPZAVnR)-EIvb- zRsYjq$blJBeO%{Jzml^KWW+I#> zvk*{c1;H9jW1c$uc-Vm1%IY$?OU7(@jwxF=NuQ-JH)pDsP1wm;edZ-?%BtS$G5?JQ zOxse2#r@D`2enn0`*#iY?2{@J^VMKF=1Q#Tp*(vWD9_6OD6^g3s!ZsY3VSqOi@h45 z-!U>^@fUPi_-$?Wxj>WMve#uUb_Q%)s4mmD(PLSE4B3Cr^qAV1F}u`mz`hIL~O@%&N_DyjvvyENA62928 zqIYvy(v>+(di`wHI~8p0UURlH)P$`}q4zCOhdpsoWi`dhEcCDn%Uqz$-d8HJ9!Gf= z7@^FbOi^WDS8B3z*NvE+AHByt^g0%%EJVr@IsKGT8A)?Ze#+~b%R%IoH9D;*gr?YKd`$v`(F6gX|&FZh2VaUNeX29uBXkSs*yp9Pd^c!}+)xszp?h z;6a|})mnHYsEr%bNsBPg3=!igN^+SM=F>b`pL&&B`}H6oU;y4#IxyTwb+RsHSpL+* zV!3~N_2 zbj5RcP)E6SJ1i0Mh2gX=hvo4Gu)k@Hf<_J}c9`HwfIecL8e+hJdfv;lF*nEvCy7O@ zxLp?xD`wzzu{F%%h$9(33wI4D3+}8rT~}$Y9BoNFq#1a_XMwOy9O<`do}R_9^bv>P zH|DU5r#UeBz^2kHye-NEJ`P5Ro27^3bUn-N$*8^QiBEa(NIG=+gS@xmQC~h%d;udUzDx^%GGHlVSMW$F4JVTnW zwV??qC!274qyfp3>amBn2^;Rw^)IL$hSQoUd#M@O$LlaAPdvxd^@!Wrh@a=mQT?3y zwGR2jsVl(us`t1xoQo5US=gQW7#9O_QMl|iX>#5{!S5|*IlUv+a}k1xt-U_A02<3b zKvC}l+Ew1+o$X5$?0pLX%a_=E;04~jd55&0ud#n>0p-jWV#S1fC@mu{l5H{ho(kd0 zPxtiaAMswd5D_Peai6ykWpTt<9Qj1^uqy1hQjJaY{XKk@h~tREy}J%>lM?FT;+tJqf4N|u$hx6Ysg#R1FU8O-7A09-^%-_(p|AQ$O zL&)nI!bKlqo(1*5&#nXBDc!Jp)l6P6;wk-YhVZ01WYc|oqER_i-&A7wP8AlmR$$`9 z8e~&nJZ(Y^R0WfSI^RN(fd3QW3j1Qp*s5ZB=i2i@aH;rtLALTqFYZ%hmf zz~tNjOqd%?T&-|4kBh{$f*^=7Ow&~LEmvJ=H?wD&zJ|C@43Tg zzwbc?zbooWiM3z94^|R~v25!xoVW6)JQSK&D*HpRp5}!b0py_zgojZO6oi8i z!V`=U@_P844M9@6KYry9uX^<{n2z^_Z%-h+nomIep&wFD1i+6L;#@Kv+d|&4aQCENT^#((h(M;3VpXMt8 zP+RST0tsbQ32ULpNeN%CsiLV*74q7YZM;K`m;sbeUrX2Hcs1+~)kW@aO*BXAK;^Iw zT7IZv+$C+4R%$_2NCQ2`q|oD`fD@(iNZToce-+ZGNRh(GQDw^5l_ecK=>uu*W;Rb1 z1=LsC_DKc1A{6m3Sr)dX#K9OxJ`x`#eDu}9OIuCE+i7Fe#t^+UKdf=o!}@pFzk41CnU{gy@FeML-pC5xbKYItZaiBpFp zQA4l)XPG|U9@NDiHGLedH^eVuF)WBN!O|FF2RL!ix^ogaEP&>zJ_y_q4ome=)Y7@3 zUmuRCC8SG#6^4VKA~13(3{P!N5yL(ljZ?$mZFLGPD;jPeLZR`5@&PaTVy{Ffrr1T` zbxIg6Eb>Rr(om#X2Em>*H&^2V@pf4#dD2cG^CoFh8Uvva9Ee`>1H2cFK(}%L^tSs` zescu)W`rW}m>)`c0wG5_9)oAc(f8Ps*pyzx0HAziI{$M60-#tIf@cCj@I4xg-V7h; z{tF=8(J^9$9>uC}N8x+(Fy8R{Vq>EhvDf`k_uvGQ0#2Yp?HKe84x-L52%XZ$;TGrv zjVnQLTXq-^hmN4f^#t@BeDF-`7?wMp0AI!t3>bT%Od*tJufB*j_eQ}ZUz}Ne5(_Ci zT0HeI&NzD_wf{IquX>_B>kwXSJb=Sup4heDo#w<|cnV+0(f)}UdO@!z82c0b5IQvg z!^%ET89sqjx~^>Lh`{qtbbOx&V8?#S=UYu0BIlF1bDcQ+JCEU}oIkjHAG~^Sgz9q2 z=3eH9L)S>FM905uoj<~!1%ls?IDe@=uzz#{1J`{qr8^j$tv}_+P_Ct_H?l{M;_1FX zsBR5`($*kk{HM#zEeu%NDnqvSggH|>Xu_oKo3Mww^;vnW5&O(TYw_`hET~_bwbrY% z--q?tb1yCSJye7BcIdK8ecH^hLz8V!(qilNwb_?OT{dyPI@?>Jz+P3!v#o*3>|U=d z>q}N-KH-Y&VY?bTeM5x>`O7n&ql)apbQRK2sxVhoRp!&H$pmj|vxoee%*juerG)D< zv2}VZwOpS~ve#y++q7ArfezE0uFbp_X|mohHFh*wgVk#(Gvi<_Rx_x{2Kh8uz!?Lk zJkyvt5jIUb-I8VKFebCYl5{g*Zmp(l#RC&oIBLvJyfa`c<@DJ80(F-3LY;MVX)$LT zZT9-D3Y+pug-ys+X9@S!*oc_|i}O%m3#1jJPvdY;_5S}ZC@ou#+yvY6#YY?+rXTc%C*&If&3 zpBpki9YeM_*o-wMnz7e)7VJU0CG$A~*6TTgUF)}D8Gg2Gc!M2_mA7VR8|JXNEwh*d zuQls;oxxO;tyt0&#y-_^EF{~4rH&Xeml&!&c#W84vL2H(G+@gvX)pm*EjHw#$#e&F z*{03L>_w^}`|zL!&)jNYspX}z6k=X0rO5)VV*-3=I*J1%b$9zuByi;t$HZz zuE)4P)wq4S7URn5A-dr+0*5FcyuSnGJzbFd(1lY%pE2;R9S=G?@N{JhHiow${zV-& z1=QpEo-)j)b>`0#<&X@>N7soW9NGUKn>N2e_qLZ9Z}I{+BJ>Wb8#)Gp*g>rS2KRdP$v1+9#~!Y zf&%9*@Jn=KeR>np3P0n;%0?X6-wBhNMq(|}zHV%#9!Dz=s1zSx! zaq_=<{IRKlQ(p-jtt%)`wiYupYVokB1ooOWn7pnWOXW*Yno$9nPc>NTQwCo3B4~Be z;~mY2bFG7%c`fQ1%8+-p3hUxaF|e@?n}1g0`hrqovQ;9CxZTZ$pGXH$i6Gx9ynfY= z^vq5;Wp*Qfc{k}!`*3@x4{xV+pkj6xRQEUHj!83o4%Fkvi%LvOqs$xIPniF(3={JT zaOqtEdY4jGU|~M>4~pPReYrP_%8(LWir}~gWY|!Kc|avTP%lcP{vX7O zG}e0~=D z?1Eh3oj5+8a-BxDVvpuVOy9i|87@2V^A%+|k^fjrY8&d+y)i1}kBOFEI6>Fqh1yg49{kWCj4%Kr9uwl77Y$v$GTy{SWPIbp_6<@rUJAiXMzIbDL085lSkTlN| zMe1HqZ99fHkLfy|=a0uuN1*N<2wsnan6ZcSWX}5_TT1%bsfV$9Jgp7)ha)pA5cv`0 z7t=pU9ExBZIqVMI37+_%b_9Leei(=k#2ovRIDEwy=e8e(-C6Q0SNp*IaR3Sq`eXhD z(&Y`)dT)~tCY=u?j%zRkb_U?m+d~-7?6L9IVlaaRXqmMJYd@`nSd9agjHk>1|CN-X z?1*gUjMxG4=@+ep?aDPczs?DkhU@T#%O*(gb3$M4T5OUcK4q)}*y43Ku?*(8qMCm@<>P+fERS+<>1^%kg|3)I zaMx;dtXmDCHI5i~y9$b)&ctckhIel%KP<`xU#TYe>c0`Y+ljfTyB2(=YZ2;7jG2pD zAeyiR^Zq(xea~u?OmIfg!u44F+!fFE@5JMIZdkQpBh+12EHc!4bFiIph1z<-`nh zhDVwUN>%qliaaoNr03i7Ss%jGyT}tYh4e)|tGe17QVJK&X{9fs|7+sO)$3C2)M(*lWzq_cU!2H zQ!f3e2`1z-TvRthAN5{VZJSLDTk9-OaIeojy!T^wIxM5ABJXSRbPUQ}VvcP9%02^{}*w?bTsGd@V~o zoLgpzquWjKteN~qU(H}e+P<)lns}~ei2E-sF=34%_E8Ut`!)-2Mzdey>1`XcMV;;(V#6=Q?(;Lz;%5T^t(nLo|3XaN zEF5n%L*#1W+xgB!G-KFeYXXsY3*?lUBl7`pGt?YClx6jYctqd7F2cck3$biD^@_F- zS1yw9)-No+L8PdFEW$%GzKlv%^@)Y)pu? zK`OEFbpCQMf3F46KofB1%%L>X5QP_w;8s9vwpoU#n_&vO)ueR_Ho%-7s*7Z4mMui@ z=K@10FCkCFLLKzB7!lvg1n!eevAx0)PdW`?^4kuxvaPZEoHgkHX@*Y6)iZ+5&kL4l zdOHLBhfM)p2HycwXwrG`t=0x_a&3qgW`zsaX!gF?4AXT@!JDIvN9rc%UP`)5n%OT` z)+a3k=_Dy<%W$_Th9x;{T0ym@f(5!)0+(yZ@2NEl4j=8Xyp@h`DTmrdh8lZwl>RV4 zQ?vydQ)~bcTWGJIiJn?J_)FO17M)|Wcabl1z#7IT3_Cwtl23*Bd8DWPmt+Rdv&Ojg zfYF-62KTMVCwgN(9+leTMmKp;J8TeZVS@{n)LV5lh2eM$WL-7JQ`-KzHgnAUZ47VH zif;6ogBLeuz%k7fJ7dfdRih6JvQ2uI(r0KK{hcQDEA@2|nndq|Exotuc8E!tjRmu3 zBI<1*)i8AzW?@YqwJA4lBmy-qZx*@Fk+8a2O%vm5ap=>(An>YPi53g znHY>pdhONc{jkC3Ag1r7tn%^4F~`~uO!OoUs!;YJZO5|s5F9QKfVnn#G^j2MAx-w% z2VM}P{>hycK~U)q#2s@VZ1N@c$f{r{)%s)ek`q`gbN~Y6y-)q#1$KtJLMkDHY3 zejx(lAt!LxiM;Dor_fs#f{ZC)_~j8s?6yEWe-MG>R{>Z@+dgtX7)ghNA-6FU?WU)| zKP?K2iMbI!`5NN1h)EZJ9!{z;2r8#za5R##cTOYN6t{um&rXZY84##v&($W{7L9N><$aaQd9`Q#GC~YxlgX*@^CC%Rj(uCK`g#l#$n|9HFR%GfaJd<${wZ!i!8W}i*_l9x|)hT(zo&8=xuzFN{0SeDgreVP-c;cywoH# zhTTC?$X&P}&A?~-RMM7Y;2mxI#N)~0&^j#z zwe{rj9g&1TlZN&tX&hfEhZ=De7{tm$L0K8!NmG0LuL5>Glf%tbvUI(Z!H)9^IJKL+ zt$ix!BR%igdkP3`7l$M$a*YyP zXUpPgg)ClPSHzO()1lWX0^?y(%+?k}y`}{8=1XIz2l05PDiFtxw6}8?>q ztbnuR#jTK1L|g@FF(Z{w5Fv*?br~$WB1`>I;`lbm!P!cl=0o)TN@?oNh{FCpX-XAo zA9|%BXsQMaOC{`;B#-raMYIq@=5T;KzE{dY`;ZE@sjH&os3s=QR>6b{Ra_(O+SUZp z3a>JN%pU`U<>`Zw7Im~)9fhP1y+VEhMG-yfG3vvk#26jka^&d~!K&Rdke8Q2v%U-_ zMM*&RkPIgOR;D^p0TbI*V0lghyzYkBS)_#mQ&kMQDBv-tOIiX0_{ytdYm+iQtyX~r z`7RPuwQ+`eqYB+R$mXH`aJMQrU3Ik1Q^d@Z=i0PLjZLuoC%~en*rGyL5-l+?Z4N9;$ri#boiMyJqLY@FktX-{uZ}hqsyi_n%b~@HA zBi;8|DP($!BPd1`qhEw!m>`Os+l6s(8uh2Y%cFlx4*X(r(7mOK!%phhd{qxmHfw<| zll0~CYS0~2$CcMQI47n7;bih|sOzCiLL0Vny6{}8i)sf$czJ8#pRO9(h`YQ>UmJG^ zbg=%eDulPHBgRn$d`F1QOrM7b$tvI>R`boHYB){&9|<}KF4KD`N9T&r3epIBYa%aC zn>ecSFjS$w^lnwyIjci1R1=AfmT2E&MjAH*c<iImxKSh^z?dMpk*?62d7T_O(sqidMo zExf#$h_J5-c-jz;CHiq#c86f0);G|RoB)~22~hG&!m}F6OG-SsjxhfhLvtf7zm38n|%{XbLej@uEY40t- zS^pMP&)*>5?R6YFn~d|%j8Rdu(~cQg&6YMEFs_y{+7AL2(4<;X`ofFRvVmkKwb=V2{EM$2&4 zr-m{JYlwASj)nKCP`0rif3xc_2 zw^jHxmFBpijVOxkgV?c7go^c}AaoE;MFR*6>_`8RZ+LmIA5JR7|k zxGwzO-Hd>kF1&c#3C9Z^5L?!Z$O+vrmLH)^n?ZCK_hDlGFj|X-F<}B_;dM45iabBt zeCi-}q5>a^i!l9W2^P24plKHQV=gvgc1jDL&22*Z(>7e1(}%U+nz1dj4hlPIe=j#e zF}N0&d}`6Lw-KVRt8n2@6J9QBMB(K&%3UQ-jb9%o`Hx^72C@6u7jUb-Pz}_B&b&^% z-t+~ZmNmh>tr@vj8*sj-4$7B0FlN?D9EJ|0iS-Z{tRHR-qu5RQp&$SC!hL2J_6xM4 z|3x>fhq~ZL+x6<}XI$6%41vr(%0K8wuIvC>uMMI0#|Vt_hp-`Plxo9K)NdQZV*lTG z&^U;9x4&UBZ!cvzcj3y)9`Mflh8gd=pvl{Yf6H4jF0GNi--Ov|9nd>4grP~pIF&kt zWp%%xr2iLH+J9l*@eLEU4q~u05Ho%Rpi9XgN1Os6v&|odpMr2ZIS8w-grc5ihMG5t zxwk3|$sJKRsTKfvqm%!4KGL1;2a%n`)r$lRc5M&=)4-PGI=AKXjG`;l3w%J}(A<8y|++xIlEN1VW=W z1bfW=!5a~d-*sWwUvdifdQV~by=W}|9Z7ofDBQKAyk7db$8++RZ;ODwO*EGNJ&Pp| zPvN9i1k(G%Fqe@RH7*j9y`qq@EDFK-(ZHrqNYWf^ zBz(7rB5yPj3-TfnNxs!PlOiysH3G*>gW=Y83VZT{(U}~8)&u{`I}X9w-Z12EkA(lm z2)JAhN8XeO%!;QhkF^oF_bC#ayCU&qWfVkcUT>^=7N^FM4>gfES{tG;+35_HEu1v@ zZqfHwJnwmU{`a+P;rIzWJd-EWqyOihERQA6=7U>z?DTXs+_u?cvmwmP%#C4cZehw9 znQ`-Gnewdv|NnP=kp`P9w$sx3WFpmA4%vedpSYCaZ&`yIezLC_rfiveDJxieKFdjN zfUD}Dxomp?r1BJ5f5rUl64`q0b?Oc7dxS#vlG;z~c-GV$N#9L5msWmcV%smUP7^*R zcRn6gfhnv_IGQbVxya^>kINP~G%ZKWMwW@EYh?H4G-h4M_?4~NZk_#PQg@bL@Ks_; zS7sM04rgyR4$MC0@dp#le`0^|!Q7$J$GN%RlXB&jyJj!WrCidNGITw@g&M`EY&*Uq z2oqYE-MAo{O}J&sKB^?K^*5xgzRtYQtZi*kEO9Zry3sJ(sAnRvWqPx0T8ke0+@G8y z{X>EK@7AmA`L55iRfjKTd39dr%(d_1PQ=@+%^Brf{u$StYW+{yxl1*2wx_Phd2@6* zYkt2x=T%QGxAfkNtl-8nI35$vzS%waxZOl2r!eL-q*_FnhyOx0ZOaDcw%$1BPpAVs zd-Xzgo%?Y1nbp?Wy5<*f|FjOvamvf`O3}sqRe!TKHgskk>Y+^eB4>`;ve^z#gR^af zhv5<7lwC73F8hdFK(@Hr`J9{IHM5(x+{VK816)+H8F!_rnvnn_Y2!3JMO( z-~-KynN|xwTA;lrF6WBcE~@~6U5qa|E^Eg%=bXcvq_a!P?&d^n zx`6e0a@igMDct*^-`UT{Tjj(yzRCU=_@41*e1&(54r_I}k#k8npH0-Wf`D>Y${f-|*7nfHJ0+%Nt2u^pijI6tpz}21R>>gyavmaMNSD-l zF1PwWp4^*j`Enn-U&wj=Rwk#a^$x~qZ_KhBibK}y@T_n9I@qqNYiwMiN={E!1mlaa zWzYL|WtXMQV~-wg%-R{PX!)|1xP2eVbN}u8HFh&LmQQpXizDuGFtb%Z#0xdMkJKV z&dA;pEe(o_RA?yeBBbtlA5w^KN-|4QWJaVE68hcWKk#{c?&FSg&g=bpJzvj-=ZBw> z`Zt|8GD(V!^-3kvTXHy6@$C8~W)XL0xoEA$i_{1n0~w)!EB|mKAOH31%Q?18R z?8m1;tpCbZd@!{fkNZX8?dz$iuyGL7mW7DA;w3V4`vt4*Lk^ST@DMW1Jq`vKPlR26 zEWu`GHGU}?1RL_}@z>XpEQz;4zp6Unv*bH`@mq`aY&eg*9*l;*6(wjK;)d4J_AFFq z3)}PTDV%k-W?|$4D%+Q0@#jExYGfI;{xF8$P_vk;$p_{+N(1d1(pX#jQ^ZG?=h(31w#xotVGWf71+_Vn5{3DWM>u`vO8yeu>SsW+}!a6 zAN_WNHkURSxP2T=o=c)cUtOtVvH05SOVGugSvWuSJ8BmVWP1PAV8PgxjI_wJ zp5-BQ&dUp>+s-KAa|sRu&beZ)>@y)VUYf10kYyFH4I5U8j@k*qg6W&V{Ic>pxNwsM zOK9)~Nz1`3wP_6fJVgdhT$sS8hg^dp7fkSRq8>E8Q(&tWnB%T1UR=~VWz36LVG9?9 zVY}QxtJ0z4(M`&V=~X<&XHFrkD`XaWbe*=!PmaSY$yab{`y2XX@L={<%nR(ZiDND2 z4B4C!YgkTjIr?b5$8R_1Fx}MU?4v|0^}CwMM#kl0#I=j`PuXD@wO^jG^==X+F=&}7O_b#YKj3GB0u5&E36F*-&aKPT9-3*km=+dM~HHliD6ZGVmjD>tL^ zHi3IF8`$Se8D_mH8;3ZIz{iGjSc3EhHatEMWp!lOMmK+ED3Q+lj+o3nzxWhG4P*Ex z{xZyH-ETC>IfHTYvhl+_I~*W)31WHnmj;(U0)rH=eEw`M_lLl=jG~b-6sz$AMzEQYzMPBg^O{((9t0Ml+xU~ z5{%dMz*(0ZaY^WST)bI}oqN=Y=ie3MJI0MmLiub7WY1mK6*fJcrpQ3}gMiZ(0V)Yq3k&}uqdLE$tl}J#1KauLXZeZs6EU6WzIJk@Z|Ln%nKI)T?a^T;p#T3mOl z6cbidvlr>PxNyr_Dn0Bv@ru5T3Ab-!nfgSQeE1z2J&$AiZBsBo(TbhCg_zV+jKb38 zc>1I&OWqaFZdlG^nfD{=Zh;P zeI{_b_Y`5T!oi?cjz21{&8ywB=_oP(HR8f_F>`p1I0nyvt2QSd+)!H^3p)Z-V}o3H+7SCqZ+1 z5bnB4=q_&sykESVm<0SGwav0{aJf2Y^jPsMCmvh(ltD%NHTe6g47hFAxz;(4zznuv zL)$2}rDZzvh%Cje^P`zxOAXuEFU2putpm~~gQ37jhx@+r1cil)bcE#%a#J=DLUvuV zN{bYbR@=g_UuL*jh=znL3BKIuHFOX5gzrlo_@w0jsN&l&VZEd=^+ZYf$>So28#KY@ zQw|(5(&O^ndFYvP3(xP1!fD3AVD@zke@4_sRumgSfL$P5d6x>!w>0_LBc|}@Mh)hA z%&v&%^?FkG_yDzx3J`k~-j>{@A#~mwL(X&6De7xG4*G_75ZKZo810gxgJpgT6KAV~ z+spT){Z%6IsktE-?D#|G?;6HdTzn~P(oKTFh5isBslq(5kOsGX5a+WqV2x}%wOAHM zF8Zno!QZ2x{NZJ|yzUif{ilW*wrN;fa0AaM*HW*6&qcoV3)(Vb75%K&NA-3W3YFHm zBKxYHz8T#Gk$r<;)?pcb4M`Ty1a;zn#beUxh@x zW>G>1YPJ=Bak|QdX+GiVqZ8@$G9?U)$%gvV@5qMbpCL&n9Gu2#fzsAJobHQM@>@EH zUDwEEFItwu&|T7C?V!Z_e~l&AL|&Nnjv+KPwh^AL-3o=AB~)J07hZ_0uHq?oV6D@A zlIn7dzEis(cH$_L<4!rkm$nH&JFjD5K_2 z$(K{`ucw>LYiJ<@UAsy42MONy?=_IWtH68y=qKv=O=S4r46-O>5j48~M?a{%hk*G{ zX=P;)9MjE(YB76zzGMq=t<58wRki@lj)RV+v$(YTtq^dom+F_VZzNEZ8U!j1MRNSiBB|e+3blRw{j?*;T%h5r!R%jgORY&|FBSMI)F45 z&4TGpw{V?DJ)8c_f;2k%1(oi zZZN3YOYyrZX7dM*j^(xXEAXYx!|1WPcCqjGIlQ2%yu|iu$iNU-UNj$UF1!cD2`0Qu z<`v6;+DULCNDcN};!)xq!o1mkpxTR#xTzB&tM8azm9y)_IuF$SkJ z2E(!2lfYnqCMXv@6xrnF&^x{aZe5uGUSB5DZ-gXi%7unQE+)_IqW?*2)=3@g9GF1Aw7|YfY=Mdnk6crG^CMePx69Fr;RY( zW*z9IO@jgHF7VOEgs&m4pmrYk){;N;;qx$fEap-i48**&bqe=jpO{6jR^l&pJ{Em< zk-TboJLjX34Q~5Fgf&-80S}!A`I({gfu#YszRnXfdPj+5u>z#dxk2LRyTGB}-mvFn zE$w<23Pa=xC>XbMk~`vIbvWiN(@g6KWZfSA!Q-3^`!tRH~b;t;!UfC%27lkpCjX^wTsNLFT~@|C8%xs z3{MjOlGy{K;FY=q9Qfo5pSBgjA~`wUIZ|Xw72Sc1;v>+p{s<&Egu%}tv0!0yoX8oS zB|F?s3H<}yX|>!oGWf$mxZQ97RD@s{PZ_3Ee}fY0-CxXz`>L+!8qbGC+`?Nuj7V2JWWbA|E;$h^J>djIXmHF7x}~ z>CuI7)-{ukE>^+a6=4wePwcXmc7@%dx424S0Bl(8fbJ&;^QAg-`NlJ;)S$nB ztp72Ct~nYGUsF3F;POlyu&Wo+ujt|{yJooTmw+Q~E`W~jW4Vrz(y-RxCOOnU0v|1W z4Gq66h4}{yVQG*Wb}X8QX699Nl*}f4v{?gUr^VYJdS#(`)k4(vTrapy zKZ;wgUO~l{jp*YWN`J|1Ap1mL`C*(do=48njbV+#VRs)g(sC*uSW!vez0>F3-Csv8 z9jM^eA5x?zHdiox<7AxepvvMSr{MbQIyf>|7ekLu#qXnbE{W&mu>Lp2lI^CzKkq4YKvGqR*vEa3Vs^>fV5RBAezg7-;LE%%gD_ zivQ6Ub-`qMWIsK&;}+etD+uc|Cgb<_ebhkP1cH>M!L}nBW7b=MXZt+r6S&(-dua;B zXw8J!-$u;gTrnML9!^!~#-pCs1>7d_g*3a@P{SGb1>YxT>{*~Wds1zIQw9&CP0M@n z&;u*F|B?dCd!$CYz1EPt=3|1*ow?}d8HXe7Zcu+OA5s)nP7?=cqf^&Hd>$T#&bN96 zDz3v3WIFDSsHacvPtb79;?OpvVJQmzH5$`gMXm^mKI@C_AU~%HIT}8hG4RZ6)H_A!|*a)dhg^8 z!TZM2D`rj~j|Z zxm!~EvGnqDsyxXWKINneU5REOBRc{6mwMuF&1P{1n@?9e9>)N!I8^$!00*+S#C3uc zO}*Do`?_0&Nq({9+{qQC)&Fn(qlj$SEby^{r_0L`|!Z)M>csSu7q8RKui6{dA<_ zWHM%HAsOX)7*ke@9A^6`*gP~1rD|tzdruF;li_Q~{d=0iRMoF^{$x3Ppe5%2jofg! zS`C$YqDKBJn1jn4kIsQl z1=WGhv|=mbpc^$LFX|oFvE>uh```h}w+b++A)6j>J1aQP5}%EdN#Livm}+RBp@ zjk~kUtzNGViPJsqF}D-!Yyx>(y9wPn|>O5A354b2ibq zL2I~9k>l#T;0HHiVlydv2QWxZ9!JjHNRO7vk-tWg=w70Oyrm*qoUVzfr@h-VxXJ;OsC?2KkT=aFFE&PTYZ_+~pV@06=2t94 zotKAa+YPaASt`vE|I?!5?*++UySOpepK~vb+bjcB|CC;s^ptK8nX0OC+o;B&9AR#r zbE#t1K3bt^g-3N?3!P!3gr(p1aog@M<{ZlvOS|+J2|IRhc=c3{@KVA7)#Z(;%m#gU zdTK9OviUBRnyL=G)DoCkH&e^l2gGA4;6#a8)MDFrdeQqUeVB5EJX)qhN0k?m zl35PmiKksElZK9Ht}3sYqE^|dVfMtO-qK33^^fT*Lm)EWuTRF zZw!4M7)@0~u8)erF=6~MQ!ZApf~tSUR-^UxX|Cl@vie0iS&))S*26bad2FVj;(d}1 z)xSqyoKvD7_f`vkMUJ^i=2;p(I)vJ;8wyEzdW>H(SX-ql0K4 zTST@GaVIy-+|X*tV?4ij1$9Z_*=oZ`Y|Jx3_9%o7>pn@V`%hux-u29BlP^xz*&|54 zY{5{8L>zga1I}$P`6n}l>FzsD(^`kHj%S9rSLp%SbLcEOYma8Jo7C{HhZ39Z{glc` z+`z<)ZjAbO04EhBVuj;B`cv-&xzSijU*7nDpSGW1-5(^`;jD#hMb-vv@ySNBO{-Be z@u%1m<$?q3oapDG1RQO99^HP0O$B*K@{3!ZSp3VJ){jM(T_XAN6?K5V>#znJ$ z{{OJC_atlbGK3S;Gto9>KdQG^VD2S>jy&-kLyixJckS7zMdMIY)*h#twi51+n6WGW zgk??@IH_hObX^?HGYrQ3a4(Mh@( zpKK^#6I$mIy-PLNxjO(4*QasT?+s8rO&X;_E~8TCKuimrQ))bKGaim~<~;7az)@p) zoY}35dP9Do);(u@bg+)ve@Uj!_39*OqzaCIxdyuiO~e~@V`;(u$2fGN0Fp7H}@2Azx`ay8kdDnpI^X+9TD`WRWMTPquklegDtETrmC8<5Kn1Xom{|v-kHRHW;~?97OJ@IHxut) zEzYwx!NYOl49czoFTV{!b)9=?d~7!UdYxwFGvb-xJUOWtC1QxCvi<9T#{ zdL})U9FK<mnqH}w!1a@kp;(+s!{yB|*jWKbx$mT}LUOsR@%M$3 z9}i-7Q$JlgXBjn0`;Kq(|KY&$L)bZ>0w4T{#o$?T%-8z{w{HGytZ8&0?^fpHFi96| z{A9Tde}ClK|g<@8Ws?3ytXAhE3mUQO;sA7}!d&_zsF60uWo*CQv0$Rc4}n z6cxyEoK>#M!mi)Ll^Gu~Xru;nZ0(}2H&xKsqf3MvCdrs;t;!tRjWP0d23=mLgYT5K zp#8bG^i$esj5&56mmX8ad-_L(h2tH`Y^BZYWIz_uX|vgeeLHc{oaglBUtes`PQsl# zY_aV2OB`E&2fxm%!W^y&4Xz-Kt2#pGe2b%V?1F^HjW_VOj|$Th_s;)r){=cj%4}zh z9sN0V5V=|Zn#(pC&TVO%ECf&OrXGj&(L`=P*LBV5b!;-6(=%Ky$ehC{6=aA9b78rcfS%`@`FJw3>;VzeRWKE?S zgon#elh4WU_prE6ZkWUMuHfn2MPIn+AjVDbi{o}RmeKOS&9uH|D^>8nB7ACGXjS4I z3&BqLLWZ3tXLTb1PQ-h0r}k-(k75^|J{XZVwjy)mk~936`-bxw6GP5U3Z!4H+k|mB z2Pxhx7dpYqrB`+}YFEXRZ);A_>yt0x`{GQp(lL*W(^7%pP$h0>-C&^Yiv{WK zLelkH7DY=rx%|@|+fRq%ZF?zd|8@a>o;jUtKXjGyg$qe={_ks|?hEP`H3W8lzb8x$ro!t&e+=?^MON%I z$Io;C+)t5*3hh@V$jI>`S;pgyM3lzCN0(P@csZj9gb5KGTx?Y{?YaRX&15RT^1* zcQ*YSR3vgzB1wTlDA{q~6=$<{8p&C;06U|R1c>^vpgaTLel)_$D_^MQd7E>Og+`^D}h(@L&r%m?CG9w`hP zS0xe?rnadH5mEFiEyAR8ic7yFt?*Y7~vDurh7lzqYZ%KvBe z>}Wo`zx}zha1x@iR^M z(d|PzgM436Ae|L$I_{5Q6WqDk53v@qrk%;0m#QecXNUB^3tMNAspKAwMvhrN<&?_`y zr|7xRSdPciRzbf~0IBo+CCu;ILC*w<9K_L)#OJXyIL-{GJB|Ox`yNg`8wBk~u9F#=vgG}*cjO)R!7#w24Q1 z6Ku}VfN@@d;2kg+Vv7{`AtJA#zuu19^=@OS+==5P*w6+PZZ=poY}bd;A{W37M-r7C z4y3G5g^$;^>8Q^x;9GiB>gayjn9D#R^>!l=@fmfY!9v$H^^nZ z9ANn_r8`zW;zpdjPq@e@P_i(Rtge-&F>gPUxaEGNJ%0w_pPND1V?SsdI|eGLq%ifw zUo!1=b&1BpL8NHmapEqr9)9_3BNFOiP!jM2RI1(sZB>I`C#I9t9D;GcR=YNogz7p#pH>MdX6eHa%_7{L?8}eeda|@K zAdS9@63EyEZm>(t?3+dUQisFCp;S2xnuqIww7HmRwb$ofeN==k)vi$W;W+s+dLA*$ z8V~B71g@X2g@k1u#O{j`spr&TQ;Nv5Q<*_(y~JG7II+KS+hVxrG?r6#N(SXRkxk;B z!VQs}#l1dY3ICo&(Az0vg^8+jg?qD|=%&|0;r+Tga{Y`Kyh^zM64B}KD?Fa}+!_sc z6_=9Z4<3M;@(m(Pvw-66EFvc7VYpeHFnZ7ykcfZBMQHOf8Z_AYE=4T=*dm zlRZKqJ|_;UR+mBCV+CIA<#obOiz0?AY@s~=Fa%6{M9LO+k_{=tc+zTy6T!JzfD3B>;LV7#}jUF ziSUy29ehQCQ-P#v=VET86Ae+{2KSyB!))J~czc>FnOf~Bw5)wiqBP3rquUFC=Z3(# z4+wY1oFIe6Oq12tm#O(szV2%MDp%WPdK0om|RJIz`fn9H>Jk-%5(}D;8OANn8G>#*9a&wn1@^9vLErVx zV6bj6zl#roj=2-j@iKwCm#OIA&}7A{$Iw0Xp0s;u5j=|YhTZ-4|+zHYDDqxV*HQMrhBv~-ug}t16gx%Vv z$_K3z-(y!*-fVe+$OSxzKUPn{m?xjW`dJjNf|^o4^IQn|PZlD)u7i#IB7V@cowU)g z4hBE4<7*6?Vb*^+(6h4`4sv7o#EKyHx@QY@ey4<4VVSV7=?qDTJ_%*6Z^3V?fJ+kB zqRQ|w42zG$A31eCTXG2harzZH`MnKm_soWZVV6O}`Y&|m$?+ztD_}tiM;C-FqGa1Z zINWS&m0(qceZ|w@<*>&P-y8%ltPKAGZl8qd+KDi)Yb@`*+zx9V?q_$mCBw%32_S#`4+$>%AcQ5m!*9zQ zaMj=NVwyIGbKa6E2s*7UQ){VUap3 ztUd#GKDFXK^IKTed;+$H4<)%H67WK5JI?fpgn_2Bc~8G>xBJUsZ&^K)?R{sN#{IaG4D9bB_1g#~+W!Hi*ZL2}$`vt<#4haG_-{b}O&aXJ|$FUxP-l?_)S9)Qj3HWK=8AND+3%6q97LG9>C{Htf% zU}aGUELu4Yi>Wr9Wp*2;{c_@MUz~yD$V{|q)8%d3w81bk1yrXiutQpxY1ETOc#~+y z7nZ*!J5O!lU#EVBF&pYcJ#88k+66(oq$yh_GW5Lnr?O62JF@Q4Gy2oQjTYKGC+@GO z^LHo3pvu>9rrDv%WFBj=?!7G#G+!HCFT3zr-Qu_IXd=iv=E2@f85pKxLaSap7JBuh zaA(sV`oj&`l3fqML3b`GTO=psu2Nw#?VIsuRWc6CSp|L5j!?sgX{eoZ7!oUYaew@7 zq2}&0XlSR&Ys|kRa(M)+`Qwh!?@j*vZkyLQT6GF;v1lZj^?4ANQi1cQXW+ci_vn>o zU1qFs34e4*;w|?vKv9|ZJ7dj8=>H+_)P{_p7Q8Wtd#sKJ08+hmKn|;k`5mkj)HbW&B_?4H^NCe;?413a)VT+al=7s6*9% z`Q+wcS2&UE&h8C4j@6%X!ALa)y<0rltc4d@lBP5($m+&dxzX@a<}6G)V^5}@_7`l; z{Mq$V4W=`HGFyMykm;Ho1yhAA+87_@r8`lreH#rV$ z{Jz0LbKM1$%!^#tdQ;5ke$JKJ zKW?MXOp;;QrxJAi;)rnr4q?jL5zO;m0ZvwFCTC;|aO&l7JQpkaT(gEU*>Ax94w=X% zb}KQhRr0*hD8oMME}^&6b6{eH1WdIzV{Rda?B4dfX!L0hYY6Bduc{Bynfa%%HC3L! z-jzk?w>`qAY8lifz?^Lyr^j;batN=Ml{K!c$>E)*bJDZ7UsKoJVPNuYE;Dba#Ga2eC5`ztXy}=V=iHs}vPcIYny3T@Z z#vm5CD9?)Th`~~iPMV|lhb$>cD7`!TEgIePV8?bDuxE+n-^@tt zZVm@ai-G9Bb`<(9M`(H|=BMr-!4(75*ems3^wrW}%i|64jbSK^^(n-K6@!?lNzi<2 zId)!A8!pK{#|ra!x~+LIJ2jyLzdF{l);uS+_`*~+J?tzS<&()I*9}41{SU9pu0W&6 z7o@oF2H81R3&STbMm1Y`ye-5C*(yJAc2E<9A56j30d;VrVX4R`9LU~FnBezQ z>9F)gFuXLqL5rIPWAjoiW|I6KzkaCTl%n!bFxF?(vX0VEf%LfO)t~V9A(o$8$)-Ha zWtl@gp=seGP<%-!0O?XLfD?1?_8IuMR82Kj+eP!>L(5QnSx1D@KIhbhK?X>{R4+*jIz zQQK``uDEXe4St6Swk|j*tRDOQBk;CFI|-<1x6+Yw!?)A4+4;2z_`YNomFOJ;fjcv} zGxbUY5BNd2va%m&(ePzVX|DBHK#z!(q|7WE3(O|x4Su0gpWF&Sv-^EJvPk6xk54U=q z0?rt#O^=Q~CH&nSk4M#~!pz!7R3mgUJ2I{kheQmd(|1c#3HpcflLN5N;udZB*TCmb zp3Qf~x|RAjsIsa>ot7s#S(yj0v$Yko z)i>h9lQF1QB>}C^wBX`=NtCX-jop*$@r_XswpVVZH^$wj`#vwkU8lr6=pBMIaV!q= zOMx_hv6Bg6fyp1k{8dM=G3GvcKCPx?ev9xX*BABJ9WwIZ0PrQ(us=+l>6KJ7H-l;H z=&?w=J^d9Z-tEKTI%yygwUQ3>lSZ{m0r)gx2iSFqKFuccVU)Bdu6$A<7gCC5jq{NznSY^oX4%`p0zPaH)gr;+)sy z=P_+;9AJji#N5G?$pRb|XE*XIma;p0#cQ&%gzW8Ioc3@v++Nd(x3ZVeuF4yn)#m`{ zTj0RP`f<#CiaG1qwF|57z2qXT!-UH8K_oS2ALhB&kcE$HN+Z3`gY?{Vw)Wd$)L4I+ zIvi0%fA2DEvwaKx4kwAXsVUR)l;ejS&|*8^|0TA2cF}!HtPsAO#3(KZ!#f1jw;2Ps z_JuR2yVF_u!fkAodIX)sr;_r1Gpto0Xzn7;-^F?Ss^N)vZ^t71vZkBfFm)n*%Q<4E zf0%AL)Qdk?Z6Vo51FW7EC(~6W-GY2V3dvtOopFaX@$!9TY9l{@?U+)7f1ms%TfHZv zbf7OjdEQ3c?|jG4F0uG0Q<3U9&p-p;E@IXEootq~!S3f?Bv{luy2ll8zrB*UuhGi% z)%hmk-_=i~`3F`N`7%_kBab_9=@|{o*TD5EHk_o|a}t?nL*)&!N#>M3y0pX)C#6SF zvpe3vr^(WFZ*IbyB7ZW9-wqK0W^n1QBFw)jM>cmKrQ6b{lkpmHr0=2z)D4>@>Z*0b zaOn)4Y(sty%P#TZaYgR4ouTsRkG&f<^ z=ymXALkk_}|CHN7n}m1u^RK$=7^3%(^=wCKEw&x$#;VQgPnB;sVs@YEkpRN^{ zzs!;L`qz=Auf#K=&rJGM^)BfyG$lV4XhXB8S*-FBtQ!7%iH%0~w5Rg~k&%)>0Zah5 zd*eLGeZs}8XQbWcI(gm{Ne0A^AsbWQ&<}1C$;Lz__zoqcBwq)A(KA+?2k8ov0xM{r z>r2{tU>2CV#^J5=LG)zuEZX2IvRzvYsmjhG+!q}}uX;)FJ9MQm$gh*TcS#<7F2AGp zn|jC%lS?pS$}6rl-Icf>JOjfce4%{xK%gx(q{n+bx#=J$)NK5T>$-lBc8I_xBU6^G ztbs1k(pVRCjnv#~FD=!5Lqhx2=n?A*Vk)vNBS$O*kGjnwt1FLID~kRH-z~Ipt{Jsz z*5ZHs2!!=phGN)Of$JJC33*EUV?&Q!+Km1fJ3Fi%^#opgTkRPT`>u%-=Wm9eG zS+!`oami}#;;;-NWEt?^mM02p#%!T$zIMW%E0jFZ*Fn|O33P=-I4QsViw>DP6T{SN zsM*B-$Y%v(oc_IyEZ%8}`=^~$F zo4=6Ra+7Fk(>NhvUl4Jqsw+)SJ4R>ASVPD5ozQr}j{2WY!i*XU+?N!3QsF=9IN(2O zK`VuQS5s-1S1g*P7U7KhD=_ry4#791kv`Q_Ws}#>VV24>(P^y|9=tyv3;x~0%DI2( zsg0^=8(2pNo+zT1%BP|v8-SnFHq&0?L0G9b0#nPXkanok`;B4vds-x(Js8Xe>~ur< zsI9cCFCOpgkH$H>98vzi09@xOhi&PZsCGmKwLU+m+Ot>CmrJi>e(7|!#?_oTncFb# zSPd;2xByKR#$#Gm6lS^P(~2|?v>3F8Xs=MiVZXNGz%#l`>gY_CDLEA^OSjW;+7IcJ zteZHnHWkx1ohA<=reWaJ5*#{cKc3as#AW)UP=<~X-pcpVpqk0(*`bb78@gzIupQ?0 zX)}{&H!(Xzf>j;P#67j?sC}UoZ?{Qe_L>B|YVCtz=O^IaIm1wLni@OhlR#^xyvCDy zlFVq28ha$Ol%)(mhU?!DytBm}|2sDm&nP+L{gecuuBX^r zw*?nlN8yy@Xml?GOrKyZ{ES+Rd%~w+>ixU4p4_JkXOG3Fis4jp#bbKWg`v*=e!)Mq z5oOnnFU^lq1P^W^mK-j?AR}L^&h<9v^!@}!re@Ib_si*KThW)`GmaA3Y<&K^7T*MI z#|e$~G%!1q-hI|Z@07VBx6uy2Rjr}hj_<;}^>S$L6-O$rPsOaoXQ|qUJ!I14Dyrx6 zm&R0B;grLJ@K=@x)_U9FNh4X@k~W4usI|v6lH;i6bxSNzt0$U%i}B!_YO94eWiaVe z0ZH~U!dXM~@U)LX>GA8=Ff}Ee>T4C?dIw$98S99SuLX5FA;yN%H+DNK0r zO94&K*3#FG((K^AwHP{fBP~Ajo4OvUp)<9%;M)lkv6|DNk=oiQ<8lkE+!UspPp9>fZI2_6^*GQgQdWdBc=(@XL$%^OX49%rM7{8!0ra-yXXs5_Vfh z^sa!X_{^nHzkB(3_^>@5J`saQ|ES{H)xD^{U>!~T8OmiJ--l-E8OVv5$)z6i@cGbK z^fL0nq#04D8t?Iv7w|oUser(?+02 zb1tq@kz~<*xiC8`1lCh0u-V;5Kb`Cs8PTi3>FZAJcKs9hS$r6G*4`u$)%CDMavqL< z=!d=SgTP^o1{;3#61<*k%t$TC_Np*XBpO( zzPT|+xFdTEe_iC6kS^(${=S9F^y%?eRkTc9xaKXqmfAAJP4_z+ZC3{&R3;6UH zew`nUy_x!O^uYkutQP|BS0vznU;JtFDlt>L+#3gK8uI}i`FPqW5C|U!ygdziV6# zY0vIM46n~;Dix!yQ!7r47%w~sc}fdcGCbsw0ZUfD1NTE?_=w7WuqnI)FWMye(`I{N zNS!5oG#<%!Wk}K8lQ-Py!+6juavdY>TbYCJslJ0LR z-4w?b{k_VH?Go9p&&o`7)mgIm=}dSv?;%Vu5&MhQyHJyxS82_|7e)QJY7?>C%`w;IoXUb3Dm+tGvnq|b@poD#V0 za)Fw4W`T0RLoly8PscyFgC|4|wetyY3>^A{UZ3$6-k-XLoBIYr;heLyHrNz`=J`O0 zl`>lkllZ&o@1Wp=fY4G2WsvR#mTJTJkDrzK=(iW~VYVmx^VjQt z9G!POR{#6QtwQ!5(Nttp+~<8ADP>hwMcS04si~o4i)<1isq6|7_jzA;QE4ZoD5O#; z8cIWHefpl?@8N$Q4{qnYulMzOJ)f**_8NF-w1}-p%_hfuZNW25h(Br+McNtl5bkpu z-R4M(e%U$V-+h&+|Dlg`{5PK6wEHL7=e>}4S?t8Ti%}S@oyxwDI|i9fz7YNC2dYFl z0Oxuaw;fOt8LTa!RV5yo;afZhViT!DMNLgUd$=xJ59$%vF%1l(6&nrTlX9j znZ&=sQkyvH>^z(HVhpX=yMfw2kSFiH%w$eBEd(b)5+v-XBA4B~K`~`7?Xv#HaMAl9 zkU0w*R2D;NrVnYq{(*7+y_K6eFbY;$e1x)V-=NW|5^5h^1LJ@ZY@g@|pV>bMkDd|? zDoKQv6?aK{(E!ZIvcrK4FYtP40?X|R$q)S+vV5~8R6-9Kr<4p@XHwzyk0mtC!JYQ* zh~Z+#r?Z9k7qh)PqS>IsVBriqniBKqal`J`Z3Z56`LAb?n__MBQ}`8eP6t3RYb_O839KL>_85vpU7*Y`OXxcAfer@UjX6i+`U%^N%~Yzjz5t zzg~vy;xzhmi5U1K?Y)G*M_zRbl7^g5H4dI9O(UNQjuT!4e$v>OkHNg%3d6?Pz zADy<`2`trYpl;=d*iRzkd4g1H%E0E~c-s2OmCxdwAv$(NlwBoDsP!ULB4>CP;R!O{ zPzKDlIuiT&n^@V0tIu`R8!O7pL?Zr$YeXo~WYGDa0<0pa4 zr29%i&vBjD)p4ft0-n?5`lK!RWQFiT%w07gEM1gVQ~`-DJ9O1lfb z0}imRA(S{vbF_cN4bt~HgUolG!KjWkyE^g`&qOOhSV?*c=4br`x40rxIPg7cjFyz9;f+7#Wv`@%sk!)F1Y zvj=tlvxHgo)tbBym4)@=EQJS6LoCzC4ieX@_9@+A178xJpBKWx^5;8tpv2^hXcz!(>tQND}=I#d2j2T6m zz8Jt+OA}bl_v)g5?!#2cYMT4am$iO7o(+QyYuzZjr~?eG)G6%KWaSRaoHh~ z;IL5~InUQ{lu?UIrd7Pa};1AsCvx26oO)MvIneGO{5HYIy7S$Pq8O)mJCs#}~@5 z->8xf8lC5RREcEx_9gCBeHvY|Q=4uZ-%MoaPAY|BRQ>lKvX0L_xgMBC>5F92bwY+( zcE6>Qr9Kmbq;~>E3sWNT(t`#c4y0!%tt2L46L8ieBiz0|g{}6jtBeA7g zMET$>_|_Lf?_X8m=Me>*X8MFCWqlf8G75&;==6bE*ncDq zCnh)wgWjjn>kG8_`8<;py17V-CsQ=(1@_l|g z_dYb6O5N;(ndu26sfeW|M^qtT$|72^R0}UB%qInlPtna^R|(&WSmHZy0JU1aa#<=< zK)Tin!_{P=XVN?nE8Px5KBvHXX)TEzO{i>L2(jaP^oRKTZc)H1h?YLes6~a+3FQ;0 z8_!l~8UK@0wdcDXD#rA4{tI%$dM}h-4Z%Or>U4{)Gv6`t2D57nIc%#8{Hq7XaV>P` zs!i~0YAh{NnF+7lwdu;P0yt>@Oi*;TjoZI2h@2Rj1iF{s!^u)Vxbko+y*&3a{r(gP z9dbraYM-sH!6#;;YK>rweLLmOufR>gmc&x?2#HlFBALs+FdK5bXj8Zv7i;qnHD#aB zZI+UFtIdKQ>FFbfO9mO)gZ7X!PM%B%u&1s^6L|*62&(W-k$j422OGUZ~KA zNz$=1Wa({zpzc=*%r;2me3FN`%~s+t^4ACw?p;r-2NQX&Tqc95KM6Okn;yup!o|K4 zcw_jB&62>S)OOWv`n`54gb1o=-Lht3=MQ@_<9rOP`}&PG%u~j!&E?$jn*r2lPB=N1 zJb_5d&c?TXUr5`v8N#M%N9Y{(x$t0p8Z*k*h+J`wgDIoFkZlHujPGkpGHJ_H9DKQ$ z#+aniZB64K;6@SgpLU5Ue(8?0I3tqdq{M(tXY^^P2wtxQ&f^ZP6Kfes0 zMds4p-rMBf>D!Rwl!8aLexqd}C(*7;grS}C`29m2SU$Z*Klevbsf`*aQ}u*M44QzX zMj>hXc^)4JM^g(M5ia_kiVBfSN&1Oyq9JpWbg528gEWpS$k)NTn_XnaYYjLtR+qgz z`6-IdNaOyza=bj_CVysJPXoR3&^Wmlwe-f|9os#q`z#PIIUB(HEoS7+EL9q96~^;X zBbX1KhLHZef&81&fP*T^WSO}G{z_{=!_+r0|EC7opqvReKOV#q(`M}VTSPLgL#eB2 zICtfe03Xj$VV6ffA)kb*sQmPz&`~W4R<7uyo=5wbS8DrEm*@7j@;kUQPnD_VncHx< zJ&%rFu?DTbi<8=_A;#VF2i;HpQN{P^)YVFko2?>?Y_KKL`SN^iISLfmCA9YMXq z1ae)}VckkOXg?jq-LtKPmIvSH%w zX6&RU&Ix%(+s`ww=V~Z4c_LRoF&150j5!Bbi=(g4LfwDqbRWM?(0))xX6Sq3#~4X8 zJz9&s12-Ww%$@!^VMI?IdrbXm?1)3vDBNFTg}q@LxMRxefgQ6NavD?7^3PN9=0bbT z%MShyZDRrF9(|I^;oCedj_6f7l@f&rXwhs#qN(J3E_7J{%9dZFvtuvu{D3M{dt-+#j;1)*Bo;OZ9&!d(YUo_c z2Ch-;5?;A8fb&P4qo=}R==2c;a@)jk-92;a(lSIGPhO_gC&%N9@eTFKuYJiP$vV<^ zDZn--{0Ox$qcr=gJnD`}1nb*-@T4EXm8rLwl@ldJ=zo{4tLS9T?K#gx>;GjgocF|tZze+uakQ^+HNBiAB%7}Lq0I?-a=~LQxcBp1tJ{yMg8m&m z?39W0gD&quR$^ClU%???ZBhIX!x-j1tM8V}0RQRE=xF|x8qInsw5*Sym(-4-{L$n1 z@a<13+^>ZN(xbt(?KxdMD+1RY%0YXX8}RtPG!8dQz<0an69+tDi~) zxpNny)jvga>pw*VSL~r_%Xvo8sfB)amLQF$W~AaW-}6%NBnRgf@+_SBB>VO#xE^{( zpe0%kKY#C`V_cjt`tk=R8*cOavu&jDr2z7EIC!447$5R?rX5*YbcgFN%09SGM;`*{ zy^~98b)w+bM>NBTF!GRkhAiu;B7=1?|2)e)oU`ykBg}^KQ;g__bmW{ z+6b~O)Dz76MZDW73-^rs#Wh!{ahnWs1U+J^B9)U1vH#{ok?E!=dOV^}_+_paVc0pi ztGb>z&K}LvO+D%0={ZDiFq&HBsIhPR+lX8DY1l6B3B5LTH1qsSBCy^^EZQD3t8FYe z)fe4VdXz4#_`Q>Mw7#V#JBH}Vg--CuDjo#}Ed+QLq=M)_ZuSTR`gK(mRBAn=sgjE@ z>RB}GSe+(3H+K(dljZZ-8n=c18*_NCn=X3$s-uQ_6h`DD_{Dss%x4J}pMD385n_0A zWDat3HE@3XWK1#@P*eWi`%TwDTp|0qPRF8!E(#9kd8R`=?}*Q4@4ksciL>aj{y@-{ zn2D2Se85>PPigiS6Q-%{7Na;?k|qR=g`0uWr0Mkun7Mg0aSKf(D-^Ryro}&;qZuG_ zPOrg@3-43E1IKZ-gcE4~w;X+j*J3o^A^4pg!-Za3!0(9ea&obgaiN$P7+QU#X{n=^|fA9m9LXAM9f*>=A!#^{#7;f1nK7)3j?0hknJ_=MI0x1b94!lz(HxI8# z92Ts1{6#ybTqM(rK9TBeo#e-EN$OJ4Abh*w8gUO1gU39d*lzB1X5A|nSST?Lerfz= zxQQo;VCog(IxUvS#5+=HCrN&Xi;k#>pMd5^dkw> zdQ(Vcrj-$|He=dy*o4;alz;}qT551|Ej?lE&v)yF$uNwCN+f+T{{(LeD1%!R!1>Qh#_ga;c+sYV*~3i5t-MESD8vg3C;etfrbp3^wY zqidZlFn?qKlW-)B%Ss-mr@dFgITt1Bm$nyY8*~sW{&RS{tA)yYq${EQt2gg=OC{}Vo)YhyCBKTuST;IrHIp#AC{jJxrbw&cjO6=$zNwDLoEXrIjao~VFT|6<@UR|G9j*Rx+b z5?NK{tKcPhjx;XUV>S5f>_zQj_EqU>)~jm>!j=@XfumB`;+O=mb-VyqqTE39?>Rd6 zlQ_H9Zf2>FBgJc(;-ay7tvqM zpUJLjigs<=4L5%B&!Ia?aO>!C5--uuJ7{*&6{!jEs{cJqTpr0P1&Of^M{dAqr=@Jj zRbxCYF9r$+kC5uL*BIUuNSzGE!wx?_uiMr_ep^n#8%2E1T!r_wtltYW=Wi#k_}Ye@7*S4dt`%;$sKakS|L=n9mEjs52Aq;E6X^Cz?5&AH12vWj5F3N!Z9 z(gQR=?leek(Sa#5FY-LSE#&LbHp1QrgUZ|($b2=8mfn}Z{Y9Nn>`_fWyhvoP&rKp4 zt)Xm?tp)_o=_Hw262u`rko{-o$o9RP!e&o2gSqY&5ZZAN%PWJ~C!4g{d{rfO9_%7^ zgO6y9(nxxG1m7F{tqbRe)gg{BrX zC=|lqYu?m0-yFqvn8SM?T{h5vH8IRwO=f40gSMAGq}ZX0=eph7Ixxp!zDGA~T4hRg@HCMf+XDUnDA{Cmh*{URlMO1+V|5}TK)ka8R<4}@ky(de zql^_x%(P+c!dj@d?jcT^>JYwuB6e<^%SMXFvkjjYu|oDDcs*UiS~g3t>UAY>GN2s# zYvREpsu2`*N5I&dMIiL411(D}n0us?v>GH}w0%ACsrQ4<($*}w(ya4o+TVhXR&rpiljZmR{I?o*44Z^y#z;2x4OFBo5G9HJ9%Mp3y6C$iJz zyfACzR;ElKC0aE$5abrLgWI#0VE-@yKdx)$(>c8)<4iD}-@gl1J(6SmhzwowxCu{0q!}{ktQ1f4pXw!qkf^(zS!|H%*RAZbRc=UZIpQL6} z`cIF}o@Y#ZCkB$^^New9-C`Tw(t8c;y7q|*Q{CA0TNkmrGj-Tvlegp=|K*o;fLt9>LhkRd zpi4bkX$Zf^l(>{hrm*91Ujl_SFAjs!lo1&BNtLXM3?ieSo+HP{$uiRiW#G5o2Xe>m z2d(EFoi&FViPO9a5_v0*=~r-r#Nrq-MSCmxBKwwm!1o^mcH2POqYcpP#k1!;GvVN; zS$OPxBh6Ro5qxVZqUOsb$g+JgL?L7&QBjJg)ly*OowC z$7}MS!wp#b9fF@Zt-{SbS7q+B$)sYe5sY`#S!WK19XPQp&fNP+*=dLq7!!1&l3OyGB+@aVxwWW=+6 z(oz1JxcC2Nc6h%gT_y9u?bIuBgYRh2Q`OwZ>6+Xmqdv08#*-e-;z*C2HXU=;8wt2D z*I!9t;P3=$V!xMkb(rH!i6rLD&q3~bVI1sgjfF&hr!bK#Bmt2fWU^xk8ChollQpV{ z#gAuX${SgjJk=QEO->Ml=t}C@+C$q9Fl56xbJD&ng7MI)0asB9*((+X8%>nBJ(_$D z$vhDHJcDc&7ige+lO5dAc`mr)b%|Wa9f6>=i|kr-jHrxB;d|+WBsf0@b_Hfb@1`QM zsJN8dz?|jo1pCpzx8Y>^P%^1ryp=d!O(DM@Ph@e z2$!A(DNHV@C9*I|qz7XeXQpIICT(oqL$<1I7N%b^BP+C}snZR2vV7W1av}N}z2V7{ zRrOKmWo3`qF|KGbxdUnydtm$398$GwCEhR;gX9!R8uBuebP-eXDXWvFZ&^VTay`kQ zOFwrrY$B*MS%NwLy9pT;N}y~4qu?zEYHQ?ah~+iXviC3Ddp#c5FMVW=loV_W?1MSW zRM|xbm&2L&l+j zKgjcpTB?$7MD|vE<7#d`Br7+G$ey1~(4sLOycAR5fY%vrNy|LW*{+u?akvFv_ZPxR zrFi(Xb~C8SIFYehCS=|!K8xb3gBp>woMio4;$i(n5WZ6xa`rhg2YW|gap@9)XzP0- zb$B*uoIRasTG`Ddw>OaTtAW(n$(9ZKT|!NECBd^^ahMw_NA?Qe5Y5e@@MZQ%Dt+N2 zwDGyVmr>JL3F{o@$jB@<%>OZWT52NX{Az)M3_DKw`fsu+y%p*X-AA5(DN@-adEsWx zKir&X2-mmDv2r_Q*(1}m*vIcr|t$_o3GemYXx%UC8(H+0?sDQg_~&Hk^$(?F zSIK#(&uE4*?#tQa(NoypN48So^ObAdExI2!e38u?$Q~tAmq9dMMqMD0=)&`9jk+*$eXyf94rBEvP4e}6rc$(`d#z{^zkBW4l>;*`u7$C_B`DSv zM(;ll!wqE%*ch*V=Gi4P>f;}VlapF;N6QY_`@4x9Jw=9X-SQP)w3M^1vy))qGsLM? zn;ElIU##~_$7B2sS2H>t%(gFw1*J9EI_46%KQ>}(o?c)#Juc?C_~xQPzUwEu$c8Ol z(aSwb`z=~JRz(&^8IpYIQ`fbbmn8IZfE_7$@2^Vwg@9{6mZO1?20k#`@_2{-kJRA>{H*pC<-- z5a;PepQ^>-r&HCiq|1dJvr3L_+o%oBzGn2WZx%Ct{3V*VW;W|pE{ExS-y?m68++65 zBKH5$fQLm7ap++j_e=3CR-c%KA9Hf(hYRCiMAapl_}M|!@3tDQ==MSU>9b(_DT8h| zb3xY%OOb|u9I6au(`h$&w_NWkI_3C5>QJ6fbkodms?>4tFt|q!hit?{>rGK8nkw3H zs1SQPWY|Xq;;fR@SKCpIGf=;v3~aY=Vy9a5qfz}E=uFuTf2Dgkwe4p~OPd1NeHehz z$Kt^Es|xGEN>b&`^Kfs8E$H9kyK9MqC=+m!p8I)-eHZl_mtBe!6{a`Q?*0#uQh670 zK4=Ly>zBZogVJbddS2vZb&oVZjTBXnRuUb0D#glGFUM_J2BffR6(;j+^MO1oR$=W` z-hHIRuFJ3$KD|PCN1G!)nYIcXA`~&n-iZBrbGGPi*fJ0{w&URMIFWtmX_3Xai6Uhc zADUm|1IkJva5(xCbYDy$4qDH_T2L>l%o-yqFDjri)2Co)!9JeNFH2Pw;^6axm&|ka zK3P|H2pAK-msx)b)+u-qr`H=H)%-Y&HaJ1l+zT+YBAC4?yuZi!E zLv;TaS&Xm9q%DsV5dK>UZM#!=?(8C*7EuYlVu~2Y&u!sVp`fv>gh;A=z@bT(sxRct0#+v`Hxm?(HW{xArKJE>~a3#vHLjdoU7Lg++gX1#?B z(Y_onmH7mi4lQ0h0U`wiR$(boP9R$D3wSA;|@ufB$S46 znUrc{J6HaFEG#H`PS;0WpeL`o()%m5@Ui=1l5cPajx+)*x$Bbf=Fu2N>wO~YQ6J1& zR_lOWdI!l{R6#8G-R9rO6jFHEgG--riiBz9fPc{z`rus(<23Y#PB7mH@p_|hOZp`` z>Af+iT-Xnb-C3spVIl6=cSSh<;-GMYc`n_&D}k)MnnxW=9mtr}M0)SH0#-(|SgUjz zS_=c<3C7`gasGU^UIZBzI*FDNpIzzBgZ>ylth?7ki|y-$+a71rgo=|;EHwrSvQ?ov z_y~8&#*Z65jx;}M4tF4E1Ktxw@^`lyWUW#FOrAIe3cXaJ?D==%_1zEbj9Y=}>7vGE zqw$$a1MhjG5>J7E5ju*AwRo5Q60F*tvIU;dvYdl zQm5~6En^yZPoWaG`pR#>_2+4Z=`@(C77Y1|(y8s-7HP9ChCL$CAsK>hHYaB!NK z&_TxonjAA>h0Ri~df*YLPPtEhxsRh^fulJT3jyioYv@?vgo^5psQGpe&Rg{!r?wx1 z_5Zo@++$NS{z)V)uzbCZ z6FhXFH}{UD9}X+x!=ZK5%GFG`tYQ+i%ivveX6?4`y*s%Da`Ig1frs3<53=<1fDAp{ z<;(cjD3cAHp2BxC>gbl0Z)y62D7xqxzwh<<%SF%e;UugjX&aw+zT}b5xyh_#G=h$MiiFFIlj)EBxx)G> zb^s*zjB6^*mUW$_L);$IN$7B$h~ z&>T7^^dNexMYC$)?@DCpS7P`P`#)EO5r19taqt{hKe_I$TxmgVTsvOyhs&d`J7 zn8|crqyiOHYtfAk>D;P#XJS?PjN4M^&+Ih2Ku5`6XEgLAxX#XE;ik5+T-c2aDhBU4 zo4-Tc3J*sNmpDnvE{StyR$FOYtd(v5E;FI`#b}zX)Iq2FzUD&Dl+w5m{rcB^;#|ek zv)sVY4(7YhLC$Pe34JYA&3(=j!}$p;CY)%e7wR+XWP{r2ntMyh2W>TGe3BFuh^uPm7Z-F|U?0tx~JK4|?LOV(nYqK9tR=@UZfOBxso>v$gX z?^$v-*s(9g8lSc3VjT+A)nvW|OOn#OFp|BQL{%bJ=Rb#uR} zT+pWcB!1smiZ7j})0y4dab5l<5`1S07S?Mpn-yfyKP8%&Piv)Qp$#^#iR8AQyMi)b zBe1t;8@IaYJ`wq8P(7`0I7KfARp(^mynK1ANE}I1J}HT$MFyf-KW}1lPY+(6xe_8% zD{;(yS&^*371!sVfcFu3IIk`pU-_wn=i?n9oM?+z_Nx=CKP5EQEFRt-E1~mzn#hZL zG1T*M7Fn(|O7yYeEYBQDrB;EGc;QJknNf3!HZ6Tc^)$v{{ND&PN!BIT9PLD+&?B_$ z+!mOxJPSL@tj`xDf2 zP6jMC(&xPU)XssxaWx5%LTnZts_ZA*tlg>J9v0hj4e*lgBjOWviGT0?;GI|;j#?rM zQQ4IDmFp6Dqa&~xPT{Hiy`uCIF;U4?Gc+Z3^tJg!I$@zD93MLw);?~grmyyrnQqF^ zclS1}!Vjb{ioch7Jrd@5%Mr_77t~Bdw4d*a8GA>9%KDjL`d6GQ;k|fWTnG6Ywg-xX zcF@bWUXmwA)2RP58M^frZKO}uNFNrt7ZSv0W30-aBtCw@we`0tk#-)Xszp2p+x?892}&?paDMFE`m z$sxhvrS{-!@{Q>*iv$m05x-|RgT=#I@P1a3=&Rt3V9l*5*!kUARD3K0vfhowNfpjG zJji=LjRg35bPGuh7^FIPE~C47AssI!f$@Q@%%5m6eE&C%&KtbQY+84Rj2^oOW3JZV zh{Ruve(!lGTiecj<8v&V^UXoz=`5H#*AZ*v#!#WMH8_Ny!_|MQgfHD>Y0e8D6pPWu z4fm{2UE4_%r9D8EmM7D%M?Tk7$)j7ASyVD2qB{HMAB zE>7P}jJK`Ay)l#V%Y$5;yFY?TUGOGzl~0gqRpuB_`<*2II6yxCnv7nC`Vh77FBLNz z=7t$3u$gs%XOv4~@Ul>}StBM2TKW(_glNN1h%uHs#gd(?mtkUd2|dyv2eoP|;oHq1 z9R5BXiM0A zi1E)a6T|%Rv~fcz#5KNyHfAiF6&-_TU57cr2@VRY;<2T7E$qrE#`PCk=zbZ0S}(N> z+0baXbux*cue!0^a27YMu9VN!6_K<#DkMHv8sV21dC_bNoxuWbyL~l*!vVl^g-Pb8 zrzH6NVKQlf6Eyqo#AJRBRmrU;?tLBfdT$)jTjGHFmo)I%;ws|ev>lInj6xmFiMWB` z-`{&oA=-{VpRQEI?zcWrck&bM72aSD97?86Wh~tnv5Co;cM2wTTa&bnDG;gOfLh&! z_;05i4J(N_)g)oGY9fHLHT@<(4Z532YjUAysN>w`VUd`xdaw`k6JnRF;rvL1juZ{u z?x2ex6psD5iwX-K()CXlrkF z<};~|qw#8SDw*k!KoV^2>5R8(Twh8uU9Dt*iO0%_|H}uQi_IuZa^3>vMG0iFSUPs_ z9h+J#!pe*X3m*1?k`BXR=^smK=SNMCeL>?fK7o5@Jz%B zZk*D39Oid0#yVoSt#&1|+hU2ZPt%Dmm@WYVPg!iG|PwtN$1Z=6Y`=5{kq z5vl_F5@WJc(hnzkZ-bUxO_4{Dgh=#hA3DCD&G=|vpv}M3M3#$&;O1d*YJR$gd$UIe zVt$SiDKF2&^*abGn9+&PU)JCW?ieb55b^H<{?0wXl|JYDLn7-j&}?2#^=CSR#g@y0 z={_mYFD*;ER>`twHhzHvdg||Q51UqVz zVO*mQsEm*SMXwsJ;JhoimcJAn^WF`!GkEr9b^`e=HO#d?2>{EXDj3%w%O2Rd4F=>Z z;8wvG66tsrf+Lc+;=xv05a18JVY-kY(8Z97cQj@1Ez+ctX7jBk8tg4@LR25+<(nyR zcXKK@-AN_k^H}J6_7zlf<#|`(Xh?W>0lY`wgyJdbVE??3O<#VR{n&7f_2RpE_VAfR zC_91RdJ||W@+{25=Frx-9g5e-kotijIHg<;6%!($+_;wB?jx}N?k)bEQxDZ*U!mlu z2z)o>f$=6aIQ>Bd#wwnJlWy{mW;qJJe_sK6BtxNRe*rO)D`d3SZ3F-AO0u=55d7{d zfa;$6e2*moMyXDL84ZeXTj&MaZyMpqpf%Jk$b#p43P_oc8*CO|0H>dCfxBx?;==Gr z^vOg^7_7}=hW2W+cb2F?!<36mRF4*Lb={<{%N$}4oF_5SgRuK$9uy0*;jz0CyW`eT z2%WhQy8ktjyeGwAI4YLJ@UFBaFCw79B@#q)B-l@1f{6(V;ITmt`EhO<82fxDUuG;| z{r0!O%x8IEpKZq$eUM}ItX!ZgZZh9NGyq_6Wx}>Oy@xL+vSF6NWN1!N zgGz-8^6*?7jDJ+f=koHQ^4)&u9Pa_$E?3Cv-clI5`7-G9Y>tNqufXlUPsqP@ePmWj zAy{c!!s}bvRFCf;CbHSk_FoW0$%nug6MtGJQA|4hW1&6zHF$Qvh6-^Jbn?ATX4Yhy z9A^sZx7uOmZ9^EX+5tzx`OK1wBS`!i3zI#|A!)~b5F5j1FQngt%ti%(rK4C!v3htn z=QG$;M}zyc4v0!rWiwJfz`+Ymw2c2gJI*d+e;$x$hnLJ@rPqyRU;Nz#ViIDo`Pmm( z6#ENwy2rvJvn(cWqXiSb!;6Rq_zNGZpNGx!cF~t@@em_d1v))<;BT@yn_=${$veMM z-|*uQ8}gWlIB7CsNiPYR&UXQ%7vZOb8W8dfx|{s$5PZ`List==?ME8m^QRtYdiWNy z_NlAhEEgWFjsfAIl#PF;aE`?r$d4hLGFdzkUA z$S23%twQhj>rumK8$DH0NOga@3KDX^V49>dmXySCqJ=IvTDhHPW=Z31vras-+n!h* z&B8T@mx?T2FQE#XV&VA@8EAaugXeiZj{f`4$}uYafuJlTm`^?%#CRn?f?@Nf*5>Hv`mmTT*CACoLa$()o=%_sDt~jveuv zGdUAMHhcfZty|PZ<77e9kq{@kqGm2yVDCy-dd?QDDm4PJ$Tej6`F>7gr;tR}Sfi3( zIyc{^kh$)422{6hgP#w7gU^(+aG*p+yzf`kCF@=jlqnZtj($kjh=SMyLFqg;U=PG4E}ipyG82{zt&6Rpt2f))suaJB-f#dkHI;b=bRo6K&tK zn)vh2L z_vpdPqtVpvOfBc-|AD*8&%5UbYcQH;8D=++hC8&5dFMHrTJkeaY5FgwM6rsFX|Bfj zrPE1!l7Lji{fEyIPorW-9;_P7;hvsYiAMuZGBxX`5x18XXu5GJ9}V(AXNOuLSv(#b z(v6AXhDaLQWra=CwFDD9WU&2oGr4xdoBWcgCkehkXxt17@=P`jdymV)srGof_**4? zdQ5|VW^~iuz1jaar;xVlM?|SP2Rl>KQ7?qJGl1jSi|cix@pGLrT6@o>&2vw3sV=ABhC=~e_ad01|4Gr{WdShN zG>)6=!p~@l^Wi}38G7XOGa~**k7qL2(I?lt$&-sWal~a?=7FY^Xz{@$R6X8+i8PDa zj!$Iv25zPA+{8q~cUzd9ldW*bAyVigz7G#-Nb!#92as%a0781F5h=HJsBa#L+bsHt z!uh*+ZpuOEm*Ts^y6P-^R}gt=O$8mVOi^%xjOgxueUXQ&KWdDhh>Id)uMgaS0PSfokS}#rivCNONh3|Ou+7nJaW1x zN9ZSR%StL1GP`G{GS_;fI+TmIdaM%Mr;q95`Rp;ie)=yLt+L`*ECVA`$7_5*NMqU&mI~`>`1c3s|$8 zlUZwx0P<(h7yRA-g~y;A)iU!Q2e2%hI%@@+G$JBx4Lny}ZNZ3HzRkREau8R2ZbB)+QDC%A!{&&2eE(Y; zxY7UcI9D#pc8|rWS8R~>EfDo6Mq%-V222^RBP#IOPtNLigPyhq8*{&gTlK?OG_<%E z&$#`EjQo6@xU~yao{GVqSP$5m7DW92tD#H(JBv}dyJ_F-C#3LCF~oN%!`i(2WQMZ| z9ykSim16QC0E5uHDrCAwvrBRce_hmpL+zuQ)S$7G>3_Dgx=pi&6ViI@?x4r^XyJ~l7r{vSB6Tv3A}6A6kfEejIwJWn`6Rs(QpZ<7OZF=4i8ury zoCORj%ER||V{{s5;e9?9aP)I6DGgeM=JWyF-hK}I)#gCjXGu|mXMxCtkrz#`{fP4I zbFh7%Hwr(jz_r}f5cZPtmyIA7v=qz-ra|)FrzmWkOR+Rt6mxPCJfE9~GK0G4 zG3f~!hAU#M`3jM`#C7GW9=V4TKD!eVPV%tkYVPk?9>o-Fjyq;YW z{R@AE9}*Umk%e=ZoxiR#6B_i$6aOfFhp)zwpdHvghd+Oa=l_qQ^A5-IZR5D?RYoEs zBa$K&@!Z$-6jD+t4Jz%@ASERlijb8pyU2==nQ&j{Wt6B$KW#;WQVEq+MCHBT|2q8B z;W-ZXbD!t={eC{!U0@OESsKoBACW=tdv@_-MPjj}u?T;V+Q3~8IcybcQ~Bie18j3i znctvy4%sAof$AN1?5KPTugm9rDY`T8t9vZZ1Yg z9wKjLdlEHEn3<4$fCdJpLa=cQs`qVT-g>M?R7C=pBr?c+YdLAu#pLup3*_@Al$?KG zK)grwaHMPu+N@lL3>2>t#mk|1$hQ>Fx*`OM5oC)7i*MV zX-i%&{azU5$Aj_82Zy13|5uKWrix~}$ukbzOmno;(R7I}K|d3NV0g6?y5qQn z-xhcckCKm!`072(1+}#(-F-Ej_$P}y0{pNtH{*3UB;nIaN+>5fh`e3;hIpoBG3&GY z(96hR9C25OU+nz=Z8*qf^Wr?!z7OsP=O1B5NI7#6!k+r7T;w;OT zk4HuaHNdv_6Je%(sJL!VxlH!}YCf*TUI1CXuIo#@Un!kP)LP;bw(59WizWG@GmBI& z%0x2+9g%a$09%f zP?zf&p2CVbQ}{ABG|^=pKa{LKja5ICMxOfxlb*wAtW8ZWQBY0bZ68%9XFCnZS~7;p zCe@=kioHzkvq#JzuYoO#20cZ_TBk1ZYM*pofl6b0*M*fY6fgT=Jl#xUMHMWmP!5*i2)}xBfuwo4qT%;XP?$vl^Ch4deNR#%RU7;WuX84w zse$ml_U(9kl|EK3jpz3C%h2$_M3fm2gq{byKrLftI#&htDL#y%oz0JhhBofOUJ&Igbz2G$#pG9YwDI_RyIrCNHIU5=|#B_dm3H4kc z!^`Lidj}oH4J%(@WzJpRMkSDC!W}k!f>t^u z9L+Gg%`{qUMSGDHST-sWkt|Vc&UHu@xH>^u?rn0ihL0DCO+q;*CfG5*^VrcV7~OZE z&>r(0pFFI=PZFsC?Nl@7-e(0vQQx|j#^$SQkiomRL}c1 zSY8@L3X4NHK6x8>-FrhrX2d{QOgLl?UnL!tcgWKS#1#$R_Yp%#ZHK|5cR z7JtZ~7doE6xt}j-->b8DQ`^o(p!61)zVhV|$`wH8mJ zF$M9b7 zhb7X@nJ!>_Vj;eTOJT;{jnMosnKB-4!BEqJI?KjW1IM%U+%+fKDp|v}X{tg~o&d6V zb#11swjE4mW-}U|lRo&u%cIJ!A7h#!-YS-(LqF%P>`>yJ$mp1N1C?L%Zbc zsmYgodhpygdf}L6m3f#AHT`uA?#CA)oqI(4I#P;7s^k2UJl!tBA5^smNb+Bo(T<`zk zov=&FoSuver8}&$kv+%K-;QU4lx8McvAY*Crk;c!TQecEY7^Bw(TlsE%!kNcMOtR; zg82Q3u-nLh#Hv0+_1tattWG3sTy!1!G-tt{r!_Fi-~v_MXoRG>pJz;{Gon&u?7-J0 z*tzNp@7QrU68h^GRF`sd?rAyR^oaj z6;9tS1;x-XAUp0-rGasJXXz?>rSb?35HW#M-)@n(79Z@ma3h-HV*q+t837jQ_sCOqC7gQKN4kke(pI5jO2KRY#*^WF*IHkB&upeBw@{!#KkD+8~Y8iv1y zY{ik=LvTTN8fsNuiKCtC(Y9@i&?m8Qrm*)CHVNDaRV5kNP*<8i=wZTfId7poSqu0_ z9nRx_M$y>5ESmMT{)qEZ9Jma6AX3)!MhQI&@$UmEI8XI44nLiZ{nU9_#5f4cUP?nz zE4ZDlh5}zpUx3d)AAk)q)?Omo`TIk1U=)noO3DDdB5Bo{(~KzyzxWEncdW?4 zGj}(0pFsky?=$7gPVB}>iBE7r;~?{DOD4|wxd->7ZcIh|aATSoZvXiiD|Wcyp`2vm zDA|ZSv}^Hh{ku3NV-ub{$qlR3-a+4A3-hPR7vZcM9w@cD39J4w#|oUEhj&i^uS~Y$ z$KRhsOw1kdf~`vU-&a-4<8Bq(qORetH}~-3u5lc;DhR8DB%virn*30SG<^Qn2>v&# z8f%Pb;vMpz@Xxasa1)b&Te-ZhqIn7KixMEB2?QsMaL>i-&UiI^#|D;ZIKAc?UUyI% z>wRfP5rTVJn@@^3vSS*5^15kQT~(jItAt}hsLPNyOZeFHQZqUg*#iM)1w>+f5={!i4ya@{#8jWpX3Aoa?n+07;1m zye!+2kneU_SzVN0GxUQ9)I7yernB+P1{rkxR5-Rw>BOmz75L%pazxDftHI&sz`3D26{Ep8L%Ty)bXL@1ZL_qDD(fEhr9)4s%6TWWDbu4fK8)U1- z-)*K$e!hHC$s4Z6B2AL~f8!JlKDmkyh_>PM&5n>8ACI1UOXF#~g1^iC01Ha~L z@_SDI#rl#wzSDF!BzBDBjd6F%oA=k?sBkXxKW2x|dWhqF`zujk#}h0OTZZ2J(&GD% zJtt8aHdwl69Irc{i_Z!cVu9WqJa)N*ly%R;(;pP$2PeMcm&TpgZkZtad`kVV>t9mxObaPx-N5mM>&OJ(jQnjDrbgUeb@s+5u;r>X+QKCcU;JJH?XL`=Oz8SJ2V(hiGfFKHq=;Ik=`O%En0hLE-RVrFUv2(&St} z*Hi02JG`F^4@bblxGngAR25zpa;g%Lb73C3(xkV%gb1b-KW zU8`5)5N9Ra$^D(R#a+Ua7j}?RWhrux?Skcb#ccj0SLi!e4qKKxks%|FBO;>7v2Ou1^Tme2G~plgK>y4UN@FQb`*16 zyI1<8LwyR?JM)N~tG>>Zp2b*bK^+{Ov>V)wk88S)z{-#0|A)uHA6o*pL6WK2fV$H6Nvn3KK}MJo@f+K#!F1YkXg=V znD43zNzg~G#;hlk+6mFD*$R`3#o<<%Gx61Qg(U`dkkxvV9eARHt>>*KMN>YI60sP# zTQNq~_FCi3X?^f>*;;cF2pcZr#uH2s4WLHJi7Jb0{zYu9xX zUQZ61{3{+sf6(E}J?~-~1_qfOE}LQUwA19!UG6*DVT6o|`oq&0e~_DXjf@2*vn{nj zF!Qh<1X*l?^j;TiZhnJg5Eaz?$rx`i4nyt{2hnw|U(ogLWTk7aHjn{-Sa!FM9sU!7 z1uVs(wD!jlxo34;3$<_F-wm908CxgABtD#FxkbGN^#70Q@tUR>@|pD%~m_Hk%W>E&E8KXg9hruL&i4_QH3!&&>4p71-;>Be<8Z!9Kp(4vwo7 zsHe9*o%$>a1oz0$SA`1nfc**3D-@yaCX?uHw2PJ&PorNE5Bh!?(FCWR^paiyMM3`b z{hwph#peoLcIh{~p+#`)(M_0raw)b=n@W?jn!wy8v|PHMb6wOQrP-w7zDyuZDk}>yL-{->J2c3^`vXoNrUqnce*I%bLHZh{&cV{A9Be` z`qg7Ctz4=_1^Ne3@dg%ihNsrw&txOkN8n!F6mf;ishrmyf|cM}xOjUuLw)2Z2y z71Zp51NGcyK%1gIf$l&PDW4om*Kj?&cT2BVKDQi(tzR?Q9o6;ldW0~48*8Xhe;*VX z9YSATo~5HpT# z?}hFv$%7%k4sMEOL3yz!Zhtiun6%=Gfd{2<+5QX3{?kTOo8$43@Lcj>`7#*ZP(!Ag z6~VXEGmz+@2jP|)^n_9dl-`b|K_hY$I+Z}c=p)U!VG#MK0X3>a{hamjs5V(C_ z2+LNDLXBlK*cTRno=qUI3pMF{+huf_q#6jXdkK7-*KD`91>A6}2dj57Ao(*I2G+A6 z(K??#l6QuwQen`1eGyZ)R~jNCB=v*bW)rr&8*g^^Y5GDr$b`&gTZ_%WwV{y zxdl_PGD}z=90faPUE-N~ay()WS-QLXCwE)$N9V;aLZ6O2U2#PRY_qp8+;$UYB@aWO z&lz-a!z~EO)P@D7Yf;S$LAv~s0xhzi1mbc@oWn>Pe72@Qnq?-`xtW37eShA!P5xl_ z#S~h`B%M9r0_@DlIK_+5W84gwe4>7s-qrVFCu7Xnf2M=lF# z+Q*o~{wL{O*ySyXlS=nat=V>vqopX~M=6L$b zSEfRbO%S+0l7$<6miCDLAMYN{n{_x_w^~We!nvFvt<<93DjXpA@0=Rv<(6sie*4 zKRiE4AGC#nL3idPe#N*cxn^?_S;q+BaWPHeC#+lC45p>)7J4t@2*>sn^PV4&L9>tE<^^~3GiKi;pfV!|IT}vKmnXU4ub*DAlfIbZ zh?X3r+}l7d%ru67b-^TCEdcGEd;=D~Zm#GF4?<&8HxR;?CW5Xygg#0_W3A1V5?n?- zU0nvZ*mF6>@aODPCp8{>V2tcsvltuBT!a6ZN22xe!Z^ZoJ^rohguJrEvByOhDu*cQIQST{? zrV9-7;)0&C&qE@Sw^=dzylo~<(5oeL-fv~`Gei83FNCgl&4bXK1WaZw;hqyk$mMfn zr5BfxzS!GFI0zxyl{w5th}c4-Yb)ydtAQ_ zGQP8B9@e}k&swFd#PU(C>|{Sryz?eEOMg|tQEB2tSYi{I>dSe(XTM@aTs|VnyPvSq z=_Sa$rI&3KZX+}H?7*fy0-$Aln~l3I2C<6Vp51&tjyZJ#3#`A0@+=mjjBi)C9&0Xc zP?SaC&Uet1x6AS6x7jGrelnKTzkq{gkDxhC;n;A}6%5b&@b%S~v1_auzGK62mIqW| z-F8H5YPkFG)jU+W`x&dEW`R47NMV~l9P?z`b+j-=5UW`xpwx}3=-rtqxa#yqZ2Z@q z$jvN3Pvc$iv3qK`k>e~UFE(N?^?9H{YDp(`l0B;l@Me&yN z@#uPM9I_z+37pUb^Dkl8*zh>ntGEZGO>etW!8Xr?`Eu#6be<3PvXmTE?BBtD#><1Q&+IsAnE^|lA}HNFwcWJ<1|7US4o`=LGICu?xfmYr)oh17Yjr@Dn5 zu%p8TTEFBGRrNsL+Pc%^PjDS9yLJwygl9sv?poO5`jRw=oQ814hs5P^9dS@qCLfk; z0-b<^=*Sl@xFpsEj~`ZoE}c%l4AhgD@O+{fIzcWiy2`#E&mhzKBFWOaWN@Cm6xO@* zAvQb#Bp>e}1>yvDsE9$xBvo+RyB@S3Z-=Fwd2s75*CmM92Svh@(355jt}`MCld3<1 zXHymHT=|dqBzz+;MYyhUX%X4C+XOH8@E$x8bU|5*p<_2D(QS#tL}vL{a{QqIxCmW? zMJ>KSE?y-c`0@}e`5XEhr0Ja4UUF1roGc`P>`g06`em^wojFQi^HDDL@(vTjIs=%x z*9@#lJ|q7yA0z}g?_&Ek67VaS=v4cX2fwT;RmJVtnS&-o#djfor2LC>+U64j^M!ce zSTPJsYC--Ub@HZiH`JYsX0si466uM0!ft-a+RFlY@2gJ~_g^43UT(-w#{!1xyO<+> zj%?hN4P|SOSXA_L1`< za-`C<06uavz?q+`$-|GnQ1jw7i5EcRfWQYsqQ?PZ!yA%v-0G#ksstt+XBymF2T`)jZpJ* zlx#YAf;O2%z~$3A^z&m|c;(8F{djKr5#*;ufhF0RO!C5yW~ILP$IH>6_NX^ z2b!F-<9C?~#C^9RXT(pEU8I#v7+R3+2??amcOl$Lc}7Hrl)=QJo=Hyx@9q^c7Bz;Qc*SlwO4!QD)nUICiED;tWD|18G&y$`I!ynNkKx;dR-MN9i5;KDo zOwuD;Cp6d-3(XmkEurL1pgEp&&kkffFX5(H4Q#OeMKWSz4)6MAQ~w{4OHRI{`m-(%^`KNw;*z%G+w_$mEU~M2fC{au~ca$jOz@sH+ELy=&pqf zRyXAvI>@76)3m6&d?RzHbC9ea7Uy5Ry%q1voddSxa7j*oHg zDTNi|WS5Kp^tc>BLY|LNM0W(*$R8v&%W45%sKS-elC)k&fUmn;kjZs2#@{p3(Qifu z-XZQCc7p<}5j}=>`eXrb_#N6~*+}xsZ{X$TMcC6Y8^>$DW+roezXtCfv^CKS)^!Wg zT(u}rJ~aWio=XokKz}^$EwZ#}N zL|sFUc_g3%93#0g-h&Slp|H%++tKJbFJ4Pd5I6x0#B@$ooJYzKCd?nbe5AK-vT9at@P1SOPw0pF6l z#P8K(JgKdh>@d)VD^q5I)1emFb9@cym2+oK&zuLFc{Ee`*CzS=)v( zjQN}ooO_q{c?q*toFlfy1!(AC3EC|ffRs81$hG6kiJoc|2>i7rq3-y%FtuP9DMmzL+ou)m^BT@$ z?IQ=PRi@&lea4k?;{ueFnP5||e9)cHr1a`eWGJtKQ#7=wMbb?AE7}?g8q;BSmOhRC z^B=s}pbM*3^XQ$d|IaZQ(A;N9NANAeFg2v)vkmE6_X)ned_L3<_x za&njh|NdTslyjlv@8^6r+;AIs%HF_bUP4GC;R?i@c>$7djY0J9OD4~5fW0p^j>q0- zVN-bvzFuMoJ|}I#FkE&h@%m}tg>tv1mcQhSdlT!JOia2w%KB z8-Bc6gg<;+Q`y(8&tE4Pgs05pljjC;&~BLmUg$DdovXm*h830lla|4@sS)_;E6yL| z)4*%|FN2Iko3Y&w{Fw&!BsMZXM2<pNL4iyT)C}e|FUhd zvgsYNcW5EwxLFxJtJ_QrmSQqJ{u!CxGRQQ&k6;zZ74$+~2%p@;9p0Jz?;0wjqPLO!$;qEeLhq0au$6ZNS3qIjQV zq%hYX)||6Pd9L43g<%TpTo9{w)f-kw*K#{`TIgmeY#l6GmK>DIJprYac zkGHgAmKVfJI5Zb3^A4hT?Jic4jUlxms@P^E56>b~_@3TbxN+J|e0@6tzd4if$BRoC zHPIMc?9h#hu6GbC$^pFj+YPi#2vC>m70$&f4)e3* zxx2YP9;s;~K06cOKe&p73MG#meTx&f$w5Jqx2Lp7Rj059Od;YI};`!PK zn=e=3)fRk)6|J$Ht5$|%?Cl5B#gj>`_co%ij{$+cm%#JCOy-@NMqmEdO7<2e^W;k| z!qIM7I(SJA@uDr++cR~sxam6P##;rP)@TWf%3c#0hag`5WMBNb-W;{GJR&dGaXa3$ zVRZ7D7TkL>4O<7QK$J`k+gUz{Chjb)+_KOYsU4LfrQQm}{fs)8oZJI4K~?08bS~I` zPC|npQ^*AyR}z?>fd+bFh;q_h)~$Itdatko)d+I?H`{2CW7EiyV}UR{B*Aqi7b5<= z68LgP81LdZ2O>3*u-@)0*lG&U?*FFIy`?*GM5PX?OG991wH*C4kwJQXcUaBZE^^$w zkx3m9;C@Z1dQHBBFSr|miMgEKDJ&q^#n7OgJfB z%W+)2d2H;3({Go2>+gO@0(i$0LmK3e|MBCH%i2^zyG@cMnStF?T{)tQ7Hj# zocAW)c^np)y(XteMVQiB1QJCCaIdzOSnW21&%zx4)i zZk1bI-=oNWdwhQDG?GW>lMvoX)aMjP#$O8J?9^0rb>up0FkVYwVgNDLjVSi{XHc!# zLbtm;fb`=hV9D}8l2G4Cwk5dZ#OF6rM7%z>|L%*fJ&YmexP5P7qA)U)&LrY4NsQ3& z10-~&5`Ft_$98B?@VV%LL(Zmjbl8Pb18;5>m@$jT(<0xO`GCF6=aEzsVxH&BW zF8^2rHGQ9n$(t3Jiv8l*=|AXqQmDvI^_iC zR9H%1&CUZw+ys7VxlsQ>5VmDyK*I2Osy*xrg8S|0p`+ojGxZwsYB$9Jv7Imwqz9wZ zl;MI~IW;>JLx(=^p)0!<&=YyNY~_hYkdW}E<@*lO-mW_MJ$ouW{9hjYytlj3$Gm`K z*v_CT{gEJ89s%pID{9SJ!rLJ$Nb4_J&}@U7uvzT^G@5bGj(ygUbEgOPG)2QFKp~^iRkrJUCJbFRKY${`(4|xGvB&D^a?237;-%VR?y? zj-V5kNHjY{kWTtDHs+WT736p)Q8iCs$}?rUSImN@39g~*5~`?du{o`N`&>|g=lkijeac3<&SNok(!LFA#h2kVDNDfSq!tmm zQA%HJUrv9xoSFJ`45M;j?UhU3+j+$I(cvOp=J>*e{n;e~Z9tm*;)v!}_65SCfMLn0R(^6Aq zY9+n0@<(VV(1%|*mhE;J_*%d{AI1QSETlDoEJUo%A}iOH!S#ixgqXZ{bQ&28xR;WlV59f5dD zjz=^(3f$v;pnaRjSY#PK;~+q-r@VkU(bl}q^aG?i{yKM85Twc=4IXjd;qX>b z@Yt3PIadC#i`;^jViHtwdKh#r`UFwKW-xAe8KyNYps{N^pL0uMbq{fvsO5 zVx1kh+H!k?`hK9xMCqNU*XcyqA4mu^Ll-wjLTIi5`JJ+u`#;0s)A-*0#5oJk)}MhkC|x5)BH}c4Lj|0vI0ZJ(>)DW_7vah* zQ~K=7BUst62~1qLJI-!y#MsR_o!9<>nH<03^F~=1b#|m4J>OyOMg@5C^ChUa>%grA zlTo<%JMzya0iIl$#JSSn!0-M_h#eaw*M9#ZbCr(3E@ccJ!oOj*LlUc**9*lF3y5Qh z3Eju-F-+dqLq>TVq)-d$uHH{&75hOUehlv4&mnW3ya8d0lkB4{Jdm85MC8)Vz-!-U z;yk(@S_V?+{vVML_9~7#s?DHpJH@Dl$0fMimjPaFo9TzJ3vh9k0Trou4K!DacF(N^ z(cwz?ka?KnBK-rqo#HfFcMEmPm!!t`9&_`uK2^Uzi?xtX<)zEo!Y;>R_~0%G-?Gd} zp_Vi@d!&!t>(`NG>-K_Vn+D`xeL>K>t4P>h1N;-ekP63cHjT?br6$}VYi~b9k4omC zkcC?83XYvI{#UVFSgMludSeelKN?78SPGfarvfugbKsc#by$10mYdyt37hPV?)W_= z1IFAQdRGvPzD^~o!RmNE*ROxjzZ5yXcZ035oRbENV&V2^`! zJb*Xy)?|OkQIsk^1C@CDvNJn}Q3+BZ`J+#Xh?f{R=t)9HV;O7@It3;v&hWWmC0_M9 z5UF2a@S;b;Xp>Jg`S0T_STLUt+t3}N(o_Vs9~42qrvOUTx!!d279h1Ooca40s>iN? ze}fYo5R~LPat|0@Xbn7^zK6UmEF^jFT481AYPcV=fcy|!McS5}g5;%saJ8xuS$&>? z+w9iCuVzm;_-Q9Z?~sSvW&T7wEe>omM4`J)pPju)8P4VMNk7LNeq;WD(S2-z{YotG zK$G~s$3C>H)an}O>TaPllthLNbsAPpBP$=A#CfVsH>J{zfn z@Yys*Xu37`d)$Vt^R%E)#vk%MgNetgMX0$hl(~3In9NaiAPdz3_`PJKm&E&WNWpCsJdyBd1q6wpT?gkS%YAbHzWX=B6s0L;52Z(zYR``Z-jlXJ>YSWF1e~K3NpzK z#M9rAw?L)~X~`Xh$a^xduQZh$yWh$#dFMmcrRQMPJR!bfO&1CoxlDwc-=euoR^sI1 zqflE~N>aq)VQBA0G~;v%FGU0SoSR!l%ztsOa`O^d#+qj&e+Ys z$?33td<~GEDk8M<2YDrZx57pJ3$YsWhxZLUkP`k!R&RSl(7-doqU(rqH_dIM*(7Gd zn3Q(=1bEA3;6}ri zv+p`{NQll&*uH%MeRlo=`!duDBwe!T$7EmH_pcnLu9^q88*@;X!3}cLBLGs<7sBlL zLVQKEnVD|h1l8A)iJ1!v!lL1(+wC6Uom>5=^*v|WyE7ijPsz~VajrBW@*|A!Qs_V0 zM}mC7W8t|PdHHPi$|_X z()`u)Y2RBrT2Q?o_S9U3tU@uWVekjE-3aWj(8ZQ3%gC1>zd`X)15CU{^vc*=swJvd zS+VK~%yXU#tKa8A$4`#eUSG>Go3@hebA;)}XZ6T!NQ|^s4}#j%>Ew5FB71uV3sFOU zR4KC!jO15gx${9(?hL@N{cN6!RSI?gGMO*2HXUv3N`SL(Zi2r+9XVDaO4XPj;L>xH z=Dyxf)rBr0q4mC0PF=og&m(>Mba^c}Jle}BZoU9(b05NaDHS?duDjyyOKEs!Jb=FY zJYf2b#qowtGj0YE!nfq)c#pRlk`Fs?KtP=!O}jb(X>CNGa?^;$m=@eR z{)zh>>>=OkJ?ww2$ySU9a6S4l7?2F13*92<<&N1LJMkujWZB}CcY?2EFm(=FX5eU9#u`&qgTwWpzYfb6iSO#1&&<+!!Mb{e1{O0eIQOn+&;oH z7ajUxeGc(GC5>;$j-ri~H{gvA3u7w5VCvEZ`ZL9GfNdB4RaZuHpASNalP+CW9f*aG z94D9ObMNMX6J*Vqf9Q8nG}JmAhEEFeblHR}jkn6BSEZg~qsZUhuL;Fj907ZjS`h-@T1g^uKhv=iNQnGayCx+2q0ct}^hu;7k+M zK7;S~BWys(M~FStKxVv=M8-lK!(ir9NOe=BI}`oEy5>071@|HkePr>@x!k|KtfBSS zD$qRS2+0e|79lDS7lr!+IZHuwh7I*w6dp~pq8lx5AQpjth z2JNeD#)eWUG&kcl95`FdpdE40@6V92@8>{DMUg%euVQ=%mp`51k2%vWEJpW%UY$=3 zE`ESdG8e&o?gcP8TFh>kHVd959|Dt4@%UHi!}5l?{$$&t0n`|pfFj2XsQu=rU>$OU z#K_%(&`IHV0hhabvfvg8Lmz<}Ev2{f>cQ*J5^lye1zn+~Fu&Y|Bx`Y8&ZoM--@KCE zN_qiB$)X_EpGkzH9AWc}6rxg4Pix1VptM<@ns7>q!6%nNbcP`DZMXyra+iSj+38f> z@i=_UN&pGp5xU)EBe}p1z@cSR>Cx$ju#NC6C`&EIjXP>lyRb35pU8q%EiWp5%8!1t z*aoLMZRw`c%kWUQ8ct=Z(a$Ewsl^+yY2& z-vN`S&I1uyLYGacp&~{JG|O;=9LWkGdOp5b)z%7YZWyVU7-7(=lj89HIs&Pnz0mjd zIu3m`M)Et9h~$oTwB>gauA{$TTfQsM)Wg{2$rQL_%z2Qqe8|AfIUpMLoGkoRMh+Z6 zxNf8!NlzqT$(~#I*V`Sub++5ETY4q-Z|1YbDgOXUxr{ff3RzL#cset8z+;KiupsaT zY~6Ml7Lgecy=xuWna{Dkf(^*zYvVk-*>BJ$O9eO+-VA^AqM4NEfa*U~5hSC}u3K(S zI@+^&408objh!Klm%@o!u|LAjD^YOo80-FLE|lcV#4>KlWT&NJ9XbrJu_+d%yFO$0N;+7H!S7D8E6fQCo~g6xY8Fd>kG3wsw6 z{hUAy@BSc5C0(}LBF1DSuK^vg>_Jo34YQ~379z=^T(lwe9-}XP9iN&gCKa0#VV=i( zWH3JfhH)(Pt)7ifo~Gor*fG5LZy#dqKCo|>`c`(NtK+>b*=W^sGa^znNRE!*XCHld zOcd%q5YcoU2>;{698B2>GBKG%iA}`9_ruV(JDX9RT`-$e#p10kMl}6{Cn#F~#&$}l z;X;rsT~K_B@E@~GBJm<~s4MZ>caPZqn2H66PaSjpfCvV9uE>Gg6)X11-7D2D^RH|3^6Y2IS z(qgRus{JqoyCg;s`Fbe8};yT^E|?OO5i;$gXGj| zN!%(`fL5dgFy1k1iRzhWjBY_OyJpgUWM6U}0>bl=(|&FCsK+Vd_+Sm%a`gv$jq^ty zKRFq%e4&KLerTWK8UBwzC!)K76UF4fq|8w#B7@$N%dKYFR191j|F!KHE@wQVv;D|TIvh-wS9|k9K4B}G9KgVuym}mDF$ozUL@VGg84iD zN70%8Q}wlB7|E1SWXPByB`L(&>)FPXNJVLoiX;h15}J{zWXc#qL}v0W&R*-J$dC#d z6Q!t78Z{~n?|%OQKltpk_gc?$-`Ayog|>@+18$)b^Z^H7=$UXpuZ=&UzbYf}%3&HP z(T#Mq_fM=~os1m?1|ZRCD_x$PKqm&wqE)ow;l$rXpfL9=hv~QpwkJh^;?F0g<1rA^ z&vrh??KaII zoM;3Z=cJJ7_iQS&ndO3YP9WD8@4<`R&3Wt2gtcXFsg^w%^wYas+TKu~ZvXTZe*YIl z4(*hq7wb=xxE>W4>ky{{J3>)Yz$I8T>k-Q;6X%UiUq|sgX`1Q3PH*!rq(6!LN7sB_ z3CdwFVIuh`3~mes$D(3*uVGE+j>y0zJwxCv^T#z|v7leULc!x75c~dxblrct-A+z&80;N+5Iz<^K*tSZ=%xSGktKfX z;bFoQz54e#I$T8x&b^g2p=;*@S@I4RD_vsUqn;%7tQ77^jRB*^b$ae!Hp?J>n`>-!2o0%DLz8n3JPK@pNv07PeR1N|@eItrEC8vBnRIZaD-3#i z!;`CW_+w%+9?Dfj^Z4!1C7*+EYeioJ&g-@ZzGQYGo-oQ*7(HVh8M7LYg6Z#XTZ zmvHXU254eCtlu9rz}4hzxN(Mu^h|xQQr=UGT6gK#bfTGtI zG{p&mgkx!pZmKvvJfjlb{A5D6@M_7rnNj$G{7+EeQm~y_L{A@CO2_d(fV|C%=$B7S zY1jL`|K}{yU6UO8M#>);i?pQE?_44uz8`{~#az%?pFo#Bo~G$V51}%x3%b5}lJd-H zEWU+Dq8E2#gRdG?u|z5;d`N*7+0|g@osKtk>40yn9k41j-upNM-qB%iw9l^9@_qNMA+IG3kJXSKx8Cdpc~Y;YYq?-v8U zT3jMDT8e$*M`3A!8nbxKH&A9Nn&tc$W@X-DPE`HDM`mh~ZoCitH*7`s_p3puM*vx} zI|^%`*2XF0kucBuK2YcQ>8_X|^vrn``H}Djeph;6xdUgJ7barUB;`;7QpYE6vR6%M-(O7ZVb=Mz|A^l5QJ4R@>>PZuqAB#ok#uI zQ;WvdgaBva24mS1f~%e*P%vtTjMr++reiUnZoYyl^h|+&56hueiVI@`*67;zBG`Y< z5H7p5k_Y59PMyw1*O#jjLpuXriE#k&-6zV+V8W5ozO$t8{9<@h)rdO3PNK~NI{4U~ zDIE8@6zVN*f?#nQgvqvnQ}-EU^Y=INwq=Z2{Gb-Swy%T50sQpa%w~wGxegOo*U*i_ zXISPNf_#^5EO>*DUXa#=lICwk4~LW49aI%rez*-wZaqxyn9e2L3s%9Z+65pv%3v$G zPU>LhH8@nWj$AmhoG@-FoJ;q=Pz#!!uxn~J)w#44@?}NfK`je_Eat}|Eve|S@D=!D zbq$_KyhmAq{P^)l2N0Sv1}7g3o9p+%l<{LY6kmpmHu-|=R&VI_-Gqp(I5`^T0;7H- zP>=fl9Ox;T;RntGXshdWbhwpu zojoxEF|z^cma+uY9`D1^;h&*q#vAOtP7G6D{^R!SUxDO*V_385Ebi^!MDFg+1@kA7 zFs2p;H!FkT1iOz&zv)AM)#aik4`hfyo**dI_(IAFcl7v#91@WihD#8Ie)Ppd0Q~^j zr}L9t4_CpzhE<5VUxNaKU*kVpe#3Gm9`+=>M=D9WFusF_0@CV0e1|uQeGBB@*A(>9 zbq~B$si1;f$wcg(c75^;rRnqwEln&<2Vooldmr#!-*l|n;XxO zvIWslSP}rGR66LIu)njU4CMaoA^urt0M(DBm>IXt$XYFSS2Xw?+Ma)a&mtX6UJQcu z$A-bdDh#4`%);@^XUf@N5mLH0%A8J~4e?G9IMu!g?(9>c_vq$xW`{b!%a9F(<6Hnm zW^K^)EeWo_n}ZUHUt{U!RN&V|gqur&PxlidEvE~+Y=z<3WtyJ3N}O0nU&BXKSHf{V zWzKt7Q4q0LL+1AddBTs*!ViT%)YZfPL3;XeIDoFBRqOx5Gh{1Zxt|3~s??^I-@b|h zkFh?=y9q$pZ?xg&D>&2TD|N~$hq+?n3M#ec(3TLw99tNRqptC=>d(F4RvOOSt&L!P z0)r^tT?lL%+4(eW00D{3h^d^*41e!LT9&~uHtQ_j<`M^CLr1~$$Sq{BvjtjLiI9NT z;^gJRY_M?lAt9`*fcjK|pG59}@~6Ydrhx`o-EZi@hN9Z`xP_p$-3Pw@Twd$to`z4j zedm&#$HeCTCy3sd2Z!cJ@nS|4=_vkUIMkF!wiGnuJmqhALc$z4`%_TBr%7hlfj`9I zV+-ow|AeNlr=s926I32?0A4h`Li5@;({Vk`BuXX|GfFJ8URDuteY3GuO%Hl+E(UTM ziO90ujD1$wDjf7Nn==={-{5}gp>-%aFtnWPy^;zIk1j%kf-CG>uMcrH=gIS<>+#;< zW#ABE4m+n$!=bEND7(cEqu>cWg1um0qRF$l9ZL$&5wJC{#7mVUVE)PuMsIR1+4OZi z`Wuo6=+i2av-}U$Y(L1^X+4aB+zQZ`svzjxh=qrd(I~|?6YI_0jT3Aaz=Qfw#Cul> zH_MCA&wq;OPbll~8`%O>i5~FXOu!^r1dP)@p&izFIOeTAf|3XP&W(X3 z?6aCL+XNT=dZ9lhgh(&2603DOz2M>rG_wy8o)-*;NO4IE&>`)C9w|A_esP&vK+D@fJNJg7j-2NuY8I$UG*F zDWCTpo>n?juMZTWy&=m`lW8b5p~G^>K#}+i7cti*A7P`AJKQps@U^-j1Zl7A!>8zC z)Zpa-F^41|!~G=b@vtX09(Ew*dLHZCGeChRcfk61FNt&$K-c1jaco~GE*)hV9cTgk z+^+xy&n7_AXCrcWrcduk6b3K78#v269J=!}P~S;o$hI!VhV~|~T&w_DCC(wLUo=Q; zMI7!9K8mJukAvsybFgEY?Qs8Vhqdty$Wr|t+0Iv1YdJrY;ZMMfK`vblsQ(p#6 zcGmTos+kVyD7Y$Mq#s=@oYzLGO@tdgv;aU+eYJ9)9Ai09Cj3!}~#aIS4AvW-q9^A|28 zGw*g0e#H(_dWY?jTDj553);CRQj6(t-2?FL8AZk#u8>ySxv<(T02lBCNK?=l$-0qD z`qQ%^Zd{QJ3;zO(A2RfP;}qzgw-L_MIcx_niZ(JTq5tw;&@c9+;*E`$xM!R#Va`+n zY>`N(c-@3TDZLK}Hi z(n%)=V5n?1xh8)BeenB=yPwyh1togSUS(tQkE=$@lxRTb%5KQ+Eu>Q^)(mk+8f1&CK;L^t1Q zgb4Fa>N0N`WqgDJt;0s7NoFp)+qgv15AjIj&}H&D{v4_KH;y*fIg`%q>cq4(hP;%? zfsFq|Sq%P0NM2bBig{OH`^Woe^m!#Qz^93v$x`$~iJwQaOsKCPJYlZ-BAhUuiB+s@ zQS9eFh`7!Js%j0`?77YEzb#oVz76=UbA)}W+Q=v}1m-W&hM5*2=;_BK=|B4 zBw_C+UfK3k@;zt`8K}$SHO@*0a}RY|rEnh<^>B!fGJDooy#njOTBu&;NGgBJ!}o|- zs8xU{H-lSl(eUE%3?x~c0L2Kzl& zx6cC(H?R)a7Y^{qqaDXp1Q99aW|H$QhN$gHU^as+eP-QLTIKl&_4Tj@-JL>AgKXz*6aGD&+SOBZ(<52XCk9SYefT;HI^LBix=dQH8 zhfjQ3g7&v;rB9!!AhjRU2)D?Gw{_?|gtz{|`)pRj-dHQh6lj8%O5LXRwF+}_1z?2qGA;Y*}e zyn>3na-E89AEK5{?ZL85oz%x`D%kb50hV1jk5upvp@wA#sigyPxIoXJIy||Mi8U5N zEt7w!l0W6t{&H2cT5uu05-)}ec_Qfd=^E;Hd?JM|Y{KbsW_Ws`H5Vn%gqTaYxY}tm zSY3OD$J3wUx>K+5&yG~i>J2G4_u>Ugf3-U`XY(fXHuVs>=FcIk!z-EZoJh*fa4U)# z=|jJx{ov=sDKKow!gGHas+#OHlz2~`AEkSJ1XR>BkNpL!Z&Ag zIfr~ZS?_B)u9SL5M!R0385QwN*Q1wc>u@nqlWU}6wbjrtQ$=;j+(0hcNzAC)J7gFu z#B`}Fq7oLGFymTtQD8w4hy2NB1|)3Aw(?ulaI7A=ZIsE}T*78SUI(Hdzk@kiQm^ow z*bHXNuo5;^4PZ?7oy7(B%Q^AUih75 ziqgbZ`JT+3H!qpqWBXBukSH_?@JZUAK5WK>foz=~R44C2N3ZYD&>Otzr^)m>aUrPPB+dw^X-$D(Y zDZ}BiT-3zw$&!yYAalbCX1@AdG-G2PHQ9WEQcDy<5g*qg?`$Dtthk*?@&jx;=M=uk zazPIr*oU|Av{16}FKqvE0lAcQn_@f+@UBm;c#fzc68ols?X(ohx(zJpv_OMosvi5Qqg4p$Uf>$+J98=J3g!;Z-_&CX==yg7u0^immC2)L=PG9FpD&cv?PWZ5Q$sSxWizV%E=W#Rx#GOOjYxI<`PyfO7mNj!#+fZV zL?+%}#>y#*lvA1k=T@#fCEpjxyyFu^grm%O=*u&dVjJhR?@vnD&4_9W$-pfIEzGm> zHfD>&2&GuN8a>%|0<&vL-1e!7`H+RsRF4T$VXsDgTwlrUJ*LJDG?bxi%LH+|;T#+- zkw=N0;!zD>e^LBvZ&SVUQ&b7R4{jRzz&)eb$H?n0#0y2O(GlSb$gZ@HSm-Acn>qt< zm3qvT{Gvh$9Q?yPjgLoN8>^Wc(;=LMkvTkfm2XIDU@uYBS%M^`dTW{AlGxd?hzd;L zXWsVtGRFo-YDPQXP-ph!v)P6ckgA`@dNKRBkDuz`qfzUrq^Fup`m=?MSav+x^6nQt zm#&FkuLgX3gErpvvYmM^-cQ+2#!-`LMbt>}M@$WiAa%F7VBt3(5B_V#&+aWieu9dW zYghwkrVD0S4ubHr)SqaFr(z+i){@4@_Rfd07aHJrV;2f32*#&lCa_8DG6?A12KOH|z{>hB z%%e^r@~iCwWpJl~TbHu{->npc)WL^n_&_g{m4rF2Rw?i;caq~B6GO%%!in9u0;xIJ z2us(<(Wf(dDW};J%)jNQuyf&Wyz}V}f@$3=Q?6{aj zyxxgVH($hUCslF0m?&v3`o&Bn*>gNEnd9aBa(KvJ8OO#p;jyRHsI$hJ>9LW<7BQ;#Azo( z$W)~VGVnTPBpA$pf0uFHSPM3tQ-Y`iFVU>2Ey$Wl#gg^A zP~vlSJilH5KRB9&uUV{Q^3OD&vAJdFX@(&0v+ZK+UiFN;FsvvMZrpChrn zScoKBqp0XNpRh!ZCwBB)i+gZ5xM~=J!dl#zQ7#u9-yQAj_G+kZKNYDr0kl&)5!C+&p8C=7f3ED&{m2v?cs;+C)T zn5wLk*lnVi5_uNONhnL?niTIxw>GUuVv}8X+vOo9mH!y(ST?~`X2fxrTWD>-G(Yb3 z%)}jelW5iTb);>#7@A=qOr)+9Qr&^sXvd5j=(tiZOe^p~m((w`;%zeG&on__0*<3( z${TySy5N0{Yq5)P8g4UTufgyN@USKlO+>ikpBMk)f{g;muy!W)AKr~8#}?uS zF$r$DyD)y2or($s6`;045opzLtQhzYcN*DKt@qlnxJD!;A!&r?-1a1U!~WpW#HHlk zQemh+uEwm~U<-wBTku~g3tW990*+YT<{WWl{YcIYIN-f8r7Wxr6UtUt$R(eOmWjcg z%LcLb`3h=v(i|>Vd7KF}6~u+2Pw~XJR>si(HJ*xeAaef&!HwY8j7X$G?caCvNW1C< zB(<7{Ed65`KFM{=`oI}vm82yKS(Sn9?=D4-^kb~O^McXf4f&cJ^`9Ig{2z5RWhv9* z@q!9lm&Y7#wL|f7`>@H{E8Jy%VU)n`eZaRu3JLw;BUPuQX#=H$Omv<%wJENM8UIm) z%N9Ok+kKlD!&Q$M_bv1BLH8Ky?Z0UpSeZrnjC{hy&uXyc@G5dtkY(}dT9F$O2{>kX zG2`Q?4ZUWUpySm-Fsay#XZ$#Vx6M!?I~Ar;ynZ!P`inu$AN=8|HoF5jc^0d;EJlS-G!=oZMn)`B;d5=iP%1+Oi;*W(%zr?$DJV$vm&jgjb^l zpq*TV&-#jj$Kg3(lDLpW3J<`jZYn}!uW^=kF>GlR28V)KkfJ`9xN6xWH?7T3>QV$H zx6{zT3|naS`HAebV(D496xYojf|{Brm^u1}vr*HFj+2jplG69ER8I>oIM~9!n(fT^ zrZ{kD%K@dpXsGzXa**`~nB{wBupp`uI2E>qcI!-}yNW~M_)Q70@LdGIH@I;c`Od)W z4ng?t7X;UL&q7BJrBPSB>p;NHh5jL?N!+rw(-MDPqm|Np^v(Wp=(@>IuKL~sNs$D| z=yQQpiEp4u&<6}3?S`%ZBXH5wqEG0vT~5w08j2RDzeQDph?+629F+v?yk9_y?q&J~ zL(^AXT#%jc1&G!!gBrhbXfrtjeMd00#-auV&07G^uZYuS%GzY7_yMB11Lzt4H&LhV z8u;^25-uVx$np7tgKIS;9#4YHPenn6-Suvy%hB7(^U%Gc4;6(kg~xRiq?tt0yM2ma z`HpC~+Vurmohs0fyCv?up@)L}4dAH8TXf=^CuF>t2`eweg8x>1NFpEL!LTPifdpxP zGcOPo{SS(d>%+6|IjDKVY}_970S30o!MzvZ@FL_Cv#$095>i}){NxwIYZL>HUt(d2 z))Q=%qE7U6z9I7;z38?_I{MI#(BXdyu=~GQn48;=zRGw~k}?9oy%&VPE8f61=d_^u zVlgT`p+Ij}dd#sJ`~_9@t6`C+7i^hV2KvHV;X`&6`rhA)=AOKO)iwHHJo_iRv;6|; zZnx3;eLY}svWL3lAqqvh+d!(O6wDJQ;3LS;2R%Y*>+!>Ec4j?PSK2^f?tM_oSPuUf zC&*lWAC99tFm#>|au)XS$9g&JxAmkcwT1K}%bPRc`Ij*GnH3E~ zi7dn9dl=Z5ErS!2rl|4P73d7~r*(zb(h=hIB+dQ;%b1OVIo?k}P0Royno8jKQ9XeD zZlH786tM3zXwv)(H%*qqP19vyBR-6hL=8dZ^JaMdR0_U!?u1?$RRGOI7-jP~BGD(Q z9iQ7E<=k27uzn85kM&@b>25}AMx~f7do57Y+SjPJc{!XI&V#WPJ*fBgOxkU-08uO} zXP##`%yuMb?EFbaVTT{iw=O^%LTAI63&8cPx$OR38ObM54dJ+OYa|h z0(Z}xhIhXy*z;*N3Ndm->_-?D9nzu;th!+L=P+>QSEYNO7Eoh*Uosuzl29e-Nd1`9 zfnD~KWZlhj=7nY&*tfF(^O$C&^jaBw!sF2FuuyQ&ZHM?bpO7reMy)Km$ZgEIiX>Q8 z+GE*DSj(^(h+YSDZPJ|0WGSLVU3(O-(SfW#BC=?|5v^=thWNQgL{xJtHhwgZCj#OC8t zFP1a>$O!F8G(`hjsaNMR!fqpr*H)ff=_iR8++?AY| zmyh!~JlsCF6H9;7CGVrlDenj_q-jopUEMhl2vwq6^~NFJJeJwP31n)-KjMcCBgoxK z3gnn<=Ft7UI8COQ+EcQVG}oO+{RPR!(~I9R0VYz>9C{wf1~hTL z@H?QQz1Nt`=1f%D=n8+NSAwlp6N>4ZPfQ-BP?NuZGtwh5C_cdkgf`v>d5ctRu(FJ? z8_tIGvLWo-DG4iC7vd++7o5h{D*WJKByKj#AaF+y{?pQjHNmym|Lnh>4CUBl+Ue_{V6u_$rfXW)ryBH=b!s5p{}F8w*ckfTfSe%?GV zN^Aw|b+yn|o`ITd6yYRGcc81k>T!P5HkP+_7qzW2SCv(-cmeXr|d?k^T#+OB!x zwAUPv6HP$JWO~?)Iz~Hi`7}<3_mb?&ZAnxyfqs&EfVPxM|oLyf} z)!z7r`YR>L*rF-yNC?{X?jg5Prk&bls)!D;jBw@Wg;>X^1IMqD0b8dSJY%ap5+9x- zdTV6Y(kecJD z0@R(S$Zk_FxnY+DJq2da@hFMfH0ejhhDMXZGxP9+V_ahC=|jjsHFJ-15%0hMg?w1y z2I}9~-QPzpu8seN$3ip6H$P!q^hJPp@~IGh+X#GPbO^1{M)-@97%W~Pic0IRpkg5r zD*D+is%gs!T>V&yyl;OIgj z7~>UTSKSe|dzV98MhG_aoClGw;_+&mO(azBGSL<5Cl@nRc?qV=>7sAB=@Ph1?@O0%y zcrnWrdHIPed4t_0L~eK;{2XUz`5OYrRH-d?Ie3q>4(%sHS2mK1iG;~suz*|*w?O%_ zz96*x5;mUwo}>g7@kZXX61iEwxr^A|BO1vdIRVvF%2tX5zFI_d8w6>klz&KG`48F4 zyTdary2jfiD16gtDuav$@bS)7Ct`a{n3EQKyANb&3x51^n) z2}0|q5Mv`T{C2|>i4#4^(-)8?Lbe~UUeylXaLYWN*K9=+v-1qvy?Fuh__l<1m=V*CCRd&JhaR3BviP{qht4^h-T?9%Jo*_)m_%&Ej%)l7xAcs zEOY)xigzKl*LDb*Y!ujTU1u%J5Bh*b_KK1mzmw#dzz4j`l91e-&saY07KxZTOC;x(kzJn-U_8Bx z6pty8@#}>wYhf5qTV5u=9~)BA10uYZXZCF8j-5?idQLUAp2u zR(!W6KLXz3sCRYboQ*P`k#UAZ92Vp~|9h9@3TKmVl8eafKYNMH^SL;?-~mazzCX3jBl2cz(&id1~!?rM~Mxt|ow zxkv7LFvN?^;a-xJ=WPg>Nm?e_N$!T@TaNNk#X!=h6{(M1Jw45V*IAw%ay#VF-i}Iq3w~!{WN9=jv3leE9Ap_@C$*%YO z)UGqJc*}xg%xLg+q$6ODH*9@N=~Q)~q!krtg~I_hYhKO3&I-0ee-jI~XA!TsL!?&7 zji}3*(sl|HSkUqaRTv~ibmGiO@CF6g^evNY?Op}9YESYuo{GJq{;d4V;L}!DMoCO(vBf_meTL z*@Azs*@DtGR-i16Yxq`PGiY_U!uv-7RJTtfN?Sip9dmz#Eh4W{CMRRi58p9#aq|fJ zn79mgO8n-yZTee#VZSt<_Dx4wbF4U3!vY}ZWsKbGHi4dj3UD?phx|iXc*FHdT=^sn zY7W}sYs)8?mHIjOqta4lt0o^1`(|o$4+T%XQ^Ad5&}rAljH;V6mA|+HwY@h%`9<@Y zpP#Sbz>Ybb_CqBoFH{`u-{VWo)E2~x_(Lh#XD6r!mL^E;$s@En=N+6Vm`9H_qyng^ zqV9JhwIX6&%;VY1P)E5o3RqoC&2X}0N-jM>XU3E9J?p=Wm1PGrNB9ioB+kyYz85n~ zbQNHhbp%s=L=Ik!h{4;mO62qGHpGU{sr3@d;G|A)n8jN=xa&?Hq9*RgQ~lL6C%?!K zUK*=HQR7xzeBmgR4rRdW!#7Z+MJ=Px&dp7FJdx>Nb0oMcABD&5AwT!cV13MXoXd^X zOtbHP)T=HIim|EG!`bS{BQ^}V8_b~Mb$21&-Lu>hA8(>nmW7^G-a&5)<}jx_uTc8q z320mH8K!Y^8x>KVfjmZ^P+9X;Q1Pq1$o}>%B)NPPty*f#d^ELVBzJ^SrnMPpyNLrg z&GkO@@5OUQzhsDVPX$E0ozb$=45pcRjdIRuQ#WsJWfndUMGY)#e`S*qr26TgN;g^5 z_AQ=SBKv`g7}`m-n?+&it_4^;{{_|evJ<(>Uqf21KhaJehq=Ok7uiTf!L5?hXz7|d zYL-qLHQ4nAt-r7ga*Zp{vMOQt@Jt0s*UFPyvTDL$o+YI}gqDO*D2lNkps5!_Yy|4yMk2EiM^JVVa6wF~tu`vCNGE{Q0&x%S{kPo@~bF zKJtR??*gqs~yqVACCHd zJ>`ZQJz^Y2ByfJ%ZusxW4BFw31d@AdfXCO!p{MCw#J+Q|NV5QmYwt$mt9FBs_6!(Z zwE)2-KgOTEZ`@A1Mg4G`qJ9_WB4KA`;BJ+J)u{=L-RunPF;vcWB_Fag+6&a6#AnKh z%>^_-C3S$>jqB?2@!rMXsMTWE(3`8)s8q@lon8@+9D|cllmYwqU!RoHq?0uyZ2jiesxQnA$sD4%ga$JHW0-n$K*aEqpvhpM1umzM$=7el7k z=Ob(Mk21P1M?z(Cq5b(WC`kPT7P31zw~AeGwkpdOY*(jlE-pd(7W^cf?=d5`zXEyH zvZ0uhqR>|!3{E`-)TI|H_`I(ZwKT96uRpmA_`PbVOYRquX4ewV=ioqQVKg7}>b4+r zCfpSN$X$yjXIda<(R|cGZ$N!R{A9teQ&>emj8gPuNVm>o98$CgBm-h#rvG+IWkLac zs;_}xE}`&sqJez3en{-c3yGfk7AjHT3Q-A}&s(@8innQi&CRSx#UsYe#3Hl}p^;<2 zH&91rHH*;M7r0nP^DX*l&StTX-X~Z4{Gg{qi=Nt~MP9LIy%Dn+geOvru7!3}nVON{ zx1|-dCUQ`}h%#|7nhh4i9i;HmPjvWLCIp2pA)BTu;eMbXUF0&3JJ&2ii|a0sHjW8i zt@r}vKM6xen%^>uQ=5rlMIgAOzrfz^LPgB6?Gm=|)-Q0%LRxc#RJNk5{)=37gM@0?8X z_xUX3nwwATMZ7o~mJVos*MB5bCc8V=PY(lDP}puWzb#{39YUy z@9pGAG*EK`zFhmk6xM}NUsB@8IWGWX|-d* zf=J2tH0gK#4{zA;0q;}fkr2TXnD5Ie)cVyC11lWJMpao}@t-xkKTR^IiE{<&qx8w)!$+}+BiqS6 zbDZUJE#Rp&o72R_%$RA6di{`N% zb#)T6w;6Be#-i!b91yNl#{yFAXn9QpvPi!QdkPBJK9L+R;uqUxU$g;7EZj-7*XN`E zN;SwovCX92{~VD$@&Tu`UM1i8rhg@x7~X3rBe44=D?rq@=o*_I^YA{~!DJ`|y{ zrmp}I78eo!c`4+m#uWO# z^&^C&$o089At+pob+$?JgzH3jXKajkzrq=Nlk6s}=S-ns9C$V7*j>)!=$06bso+A)zh3C^d+6SQ-ey)Qmw~oo)un%a-Ar zmh4>gTL}I{B!OQ%2DjGok+LOWq;JC>EHyQc*8TYwSH98UDa+0vX?%P{u}=m{vSfHq zRvsYbiYID=%YQSSNBQuqqBd&md=QFiJ&HQdUd9J_`-o9jKE60Fk(_ubLQ)=3NO#jk zDnv*hMPz>9E|!n{mr4RAnjtG3o>>WOFb0b#hoTh%O zbrP%K1oGCkg133$C1n8my&c{6xutJ4(BmVTl zdX2mAw|9zI`*9`Sa!8%r+BSsa_f_Dk(5Ix=r4Z*Pdl7#J9n$X8jfyXdlE`H~B>Uh? z>}{}-@O`R8t={!m?n65k`mzK!$K>Po)!*?`+oh=Yjs|gRj3su`x9~YmAQ>8OzF=GH3Se+E+N1;R$(>y@lRn z7VKqbZm19i`>iMl=l% zYip3mq{mcn#BtKj9*C66=fi`9E^7UU-Q@Md^ANXOi1*#_3-$1H8Z~5|gPWJWqk?62 z5_h&2{@&UZACgugj!G`1ROT)+Zuvrl=f{9wsyXreSd+bXQFV{1X{S_u-LkuD2yedsvRlNqbf;D-{8U5pb4OgDsb-#)|=#fkY3#M66Wj7 zqK^smqEjoADdj(+=+mL!Xiv$1wb|eW&TQ^JqWL(R>vu%=;yTzYy(4YNbEj8*>cF>$ zFFBcPQFzjY7%qlwJQeT$*l9hFDpqjn{T1*0(+@~=oO%WUSj<$CECh& z9un&2k~B?0HkXn?3%SmrtG@1_4WoSEemi{l7vGbX=F_&v7CeZSi!!U=^aOdq`KU5PkM6#Yh?E zphFM3(TRWI#Qf%{V_q!$nrK4Tw)J4z>l&4P@)^b0u7aenH}J!q z4+2)QT;}l+iZ6xbLi;6_036s+-+jvdKRC4_75yIw}7wfE|?RiNsPbHuuWeRZai86 z5^~p>9f!)GcgGztQ&6NYsfp5mdu}t%pHzYV9zYiJ{ekm0X7Yv)m{IGfNLpR-HXZ0u zKsSDS!14U?nWQOrd3` zT*wAh_P#Z<8oyRwMTu*l;(YZhMOCeNC>>K|`+zZe1odF3D1aj0h_N|%3g1r=WAdIv z;!`#JaDnCTeR*#H$NUW8NQgD`xz8dRr)==DO+)C@rk|)fvlKsNdk~kYrR0m2C;esu z1NtpZB5J!CTI(oGXR~Jm)>Z!a-$MwRh`=Jp17N4b38+|V4x_G%;gH)F*si}8y-t?` zzKR=+)#^9YjX^(buDt?=zEzN~hqA!BPn}mXpofOrvtR&*A^4USl$Ys3K=(Js+-eyb zIaLZ?75hWjMFa268n&Krmkj-cW2t`@Vlc zzj0EdD@E*9+^>0#s#m0;svnW~{BI3|p0(SG?Q5QwFfl>Y~5Si%( z);AuazUz#IMRqZ)|`fRn?|D@K}jeGy=DI8q@izD46&cJ2+M&< zq6SWU=VYmg;c;N;gygYLs#PDGyvyD$INj^hMsnp?Fz; z6D4#f6Gfd+Wu6p`_ zI}@eqDS>3zFLa}T)r(CYz|RVuv5(0JhgY% z8idBhRx*Anw~@PeI+Lkl%jh;zp!aGDYwQ@{>K3G-q=_LEe9;BFeNYD7lj;z9CmB6a z=|pGg_2^C67Mx%d1}$^$Qa_m~=HNjk&`c7;Zk}uKPK_Q^7R--w!EZvo>!RmOzAa^Y@c zJ*<>P67@#^C`$Z2joc2zq1J><#`UTdY;Rf)dG{J{{2eFOHFOAxtzmPmKTXJKbqzeU z+Zu&l%A!J2u2Ro?SHt47Nw`O10jJmZH)XOUi`tUI!9qv*;P1&V%$!y>_i)x3c|tX= zq=cYcF&bozB2nkLa85D1lX_{IjoHWz(u>5 z`unL4_tUt}|17?*--_|DD-u^diz;6fqqimrIL~q}Xk;kh^h=g#_Z@8p#dF}i*A&*7 z?FbfnZ>ZtaIq1M52QaO=M~=_9ZM^p6U9?)!1`Se4Xwa#R@>7q8Z}zsxMc^#emCQ$K zRBtj0M-5P^$`rGxqJ{gi)*b18{fZBM?V<9V)?mHyr6}}tJDNU2fxzE+*r4YM-g?^> zzB7Hepsf^Hr4KW6&ndxrk>%*v0woZQ=Oei@uHnwa*T}(WD~Kn|f%a`HxVw+5gJQ)E zj>e9~So%>jDCsys%c33Bw-5(1T$N_5QX)u{MHP@yi!w5p;mC4>Es6K)%b1EvqfS0Z zMyF>6V+bUx*Z(qc~2p z1VQy1dk&iu!1&9pq>d;?QW>8TakcDkiOXn6sS{&ySg9$N~L-iz43 zGl?vGq+u%cGW=S$61oBnQ2hLL=&FkYQjmMcfKDw^OpHSf8|I<~1B7~>*hGDO)`>*~ z&*e6|*-eMt#!jV+_rnrNbRiucf9Vi~BJvz;Q5 zK6Li7C<38d=z+#QB$J(t9G`8*mM>~J-@?>j#{&VZ6Q6{3%V^-)_UE}dH5qJIxCVFr zdcw%*T|>3vfp~}UDfGZA7&N?Hnf0kO_F>tkp?-nzM)xoyRb&t5bKPK9MI`;n&K3r9 zOz7N}x%6S3C$N3n0{9%Whn`p2P4DY2r1kFe(XTZx!@iJyI@-P**1LJqp3-w+KuL~w z_FD`vH?rxQeU&tAUq`>{XSoJB=ja`c!SwA+YdX$3oVJhnz-EjL>A#MB5Og7md`g+d zuZ0+B$(}{apRR(wDks{jA(%e7Ee{+^57K+%kI-JRbLjEwa`Z9o3Hn0OcbYT$mM&Nr zLC^6w!y7ZSQGG-)#5G<+8(I}fln4g9kcUiC@=D(BfnU^%i{d7sr_@aXJt&i6nPap> zfIE#Wvgk3XA8^b523(`OX@7o8`rTv+{Wx8hp3U}gtc%a^I(R-j`CAI~#gSRqcZo9j zA4O;4PSx9mVHt};2o(ttAra1g*EXaml16Ess0`6Wqkd(cG7q89AXEqyXTNJxh*DBY zgOZd~nlz|1_wDZwa9!te&OYZ|>wTX4-c=>bh8$34UzfSFZ@Rr;S=kBL_jVhsxu5X9S?mza_q}t>Fkd$CM>aJS(h+bY=7g%CSFScx2~h`Qsp>ITQiAm zh+WM_DEYCko(b%a2^9B7yMWEoCiHxpf!W6|!=T1?R@rhpBq+3EdCGo_?nq^4vTp3` ztHW9T$7$GP=)ii)*|C%RTu9ee3%o42t~;!)*(tMzvRAtm+3g*wY}}wKYbI3*3;K;% zzms1e%-WHCJ;9mFzbVhB@|CbTdI8&Inaa-iY|E0>#_WYLHlQmOu^(Ui(B5}x!alJV zv|jn*+RY2$z@q0ggqhAC=O1Fx4mq~aN{{W8mSzWjpMl@oUl7?X(&YVEN4D?gc{bKR zg*|j_G8^M}oz>eVV>{?1Z#&}N9r)$&1V&Abhn#f5S#O(k))_W^kS^BedFdcnGDhv8PmAy_vznICPC&TD4u<(2QP zWThrWfJx~$m}goE$sHPOwninqu@Gh*?J}CSC!Q+}60kzKQ)>&<59c;I=*GiAGuH$6+=h^GgkL$+st0%J( zHw)N<3)ZqF-qYEZre2tEE{^^6BAnHcOJsW<__OY6AK>q`p{()q6EJ_C3p@PjboR~m zelRnZWINw|f>9$kvxghjuyJdg*`yW&_J=|~yGZ6Ata#ziCWUNghs7>otpkTsi_`1b zH+8XW{aOq7v%;KJX^&z@evJ@~EHZ;1W#d@+<|!;o=i=4%ggtO#H7XmfVY{T3v+cVy z*r=yV*?-(Uc2tG6?fvBrpf&3ss1B23Ekudz3#HTW?dfuMO6ffI@takwVdZ;>znu-{ z#zUy;TN66*vH@Fne-ku*`~zY|SrCm{3pXFSu#0BRV7GoPWXIZ0WcP0mW)Dvq&Z=DZ z2MN7eGH2;H_TIi&aLUwU>o=;h-UlmSRKiV=s9b<`0~xT*Xaa7UAHlkQ?}H({6?DcY zu~YB0GM~(zQJ>xO=(Z=F;@oc^n4cvZYy!WJ0sc`s$@`ND=TSK>l0e7yM)dt3L&%ZO$K(04V*aSN?yltOzt;#8eDjd8}WUmb$NFZ zKK(ft|3xi<3q_}J+_W!rYHb8f`E*Yl^YuMfy|I=)*UTYr8$$3y*%sP1Nt1)lo9bYKSAW%*kpS zO6qmCz^3u)d}evS;A)n_*oHXTX;)5(+d0mvDUNFOxMIYJd-PF$IPD95N8|V=@?Y3v zV)f@7C%60zC@l%aWg9H8gYnJH|$QcR^{C*zwdjkIrN zCRMfDO#RY(se51w{yXvy8~sk=>2--X^`J5T%)MLS75rrimrf@|T88*-YXl7x72|<@ zd(h+XO4>Rhg{qnR({0Br(6n)Y`otB`Y_nugIgm>hACm@kZY>#>=0aD#Euf3F3&@*2 zlBj>^G;RARn7yVukggYo-2M^oXkZV=Y-;|>Ei8%$r;JRpcCZ{$9EPIIMtM@U&lggi zchU9nkLm8V1Dtk_E;>cFaa&G^>6D8<7}=^uYT737BvjNf%4R%O?>Heehy59gYqj(X zHw*7fDI!mcHMxUvdl~KDhT@%vPKlrV7tE!_8qwdWO9kmz5A~B+%8Y9ZrQ2_x6`F%0 z{84(w=Gfr~&#{U0&!+oACOVgnu)9E-n`KD3fdmG?5>j&GJ(J!uKt(<( z7;!z09+!5ZhJzwHPpzFQ6iBcw7fVQ0(|eMUWD3t_caZKrXL8Ew0rj}-LpCYx!YOMc zaq;ZW+~MdLTGnu!u9~u*Zta~c{wBYQp15_9?kyaPdYeUfK}MT6t`hP?mJhh6L5{fj zYc@A4!I%04OT#X+J!n4Kk(syJ1$&xvsY$;Feoj3}kNx8W{<%3F=W&S!{JTzz6DBaM zLJ9d#c_tnFzK3pZP(%G?jwG=yMQkwj5z}I{m26r+m)I&MFyGD`!IF_>aB5YJ&{NJM z&+jmJzGi^#zx0)R8Eb>_rK7QK?sf9!;}+C@S%=mO4wA#I3N&DzBDJzLWhzR3(1jm$ zxLFSrG0RMbst-3OliXgQX1OuSeUCvoNqHd`d66a!+@`G?uQ0nj;>kvnEPCg-2t5L= z$jfCX!7S!2b=I3lBBtgu_4mJtTIzQM00<8sBiPb6CKjIh{+qw}dovYw{Rw8&k<}hDs6E1Hu7V>WM znS1_7Hre>%mN(1jIfs`_&`mKcKL3+BG4>+ZTdT0%mga0xyfjT4mSKgN2%_|Z z8L9EsaCYqqFzB64-l)9;+XNGQko$_P%Ti#AhGpQ;geUmiq=o#Oew0l)xs)9qwUaH) zHN%Y&5vcd2Riw$gkhZQbT#@kwx<96YetB#MQ@?ofjhg?z8rAM^4t!8^cEOie9hC^F>Q()5s{)u8M%M zku8vVT?>ZPe+PZjB2e5K1iDjfz%RE6ejEP7-+GcD{bDj+b_m2*|3&Bp%Oj!NmH*M1Yi9M(Ffi_SV{q|40;4jWvZ}ecM0SiP^6U} z>R|s=hF%);53(ktVVQ3nNnGp+BD|f_o*&5u@}vu zlbN9H_JZp;AI2=xhfE8AGZxycvsD2xp(qdmdFlG;I zNr8JY&*_vmwdB7@e@LrNWPLu(z^w}w!yJpnY+6bfB?TYh(mx}R){BF5slOcCwE>DH z;!vjemhjsTKqmae$;0k4SH9kW|IALJQuh_6SUZz(+aL$2zDY3jk|T^=Fa?$jdSmO% zPYnNvDAv^=NWIveJ{yBgRXOt}Ms~j}hvuU@7?+su;`DSkHqmNKa_6&M!>axfcG=MJ_QTy#Er}xu3Y%17=`#)Saz9EyG^w z6yBrnns8?FG}tn|6jk#CAB)hBp7y+xB;Feh$!sc(363Snd&sHObhS)osX9?#-fDWBYIl59$%&xAlE*X z`mOfF&pY`|$21nqrOU?~L;D%Q zO?+dakUL)wd9P$?ChJZO^VI3Hz0I`m#4~cj--C9T-6nsZ1<|OKQ>12B20BSc5IF^Z zad2lWo#>d2YTM;#SGf!cp1VLSffeYKS4w5vHSnz>!#g#r@W!PRv7#=L`7C_@DeQK^ znCgS{YJ`tSuUrdXS;jNrw}$Y3%8x}|s_Ux%E&PmE5}f%XTk}bBau_zgmPD1Sr!lO3 zJkBdB#K)g4$U3Vi%=l(k9iVcIwhlAL8%J~q4O7F3t@6C~f4MkieFObm+adU*ojA7% zPwA7}QB>tuDM=2$Pybpw;M3}*H1|av{rW42yA$U~0^iN&&1P@Gy@DqtXZ3oN*`I>! zB3mYRmlt{IWQ3Lb&(WX+~NqrRoVdebRSoGOs{nbk!MV znqt(II!P0Uj9 z%k9O~`0ohv(O?bzE)kAN$5kO!{ST49G=gTvz8CGhH-)~Na2Pa-^f1^($Ti-UMWat; zBJX)e$+yK5(0pYe-G1E!Qj7a9RyZ2MY%OU*T$Lny8%Etnsl=F*zqvUJbH zb5!m`Hqt3KXq&`Rx^MChDj%su+-V^-nxD>CnA+0=?pAQ*gbh*7@E|ft$#jw>hg&|% zVTb>IR9>MDkrJ;-*3N5KR;$Hm9B*K@@`q{H4H>Fdpn;C>PcykmrC6LNN2;}w=*HFd z%=c_vlt1uEbW3|0m6AMC<+N-cme~`!Onw^{?&E2g*AU!(!kC)d2$@Ei8@Ts!0IKZO zfY)Jd-1T!WMC-pF!P_~rXzEOL>g7SXl%YDjWc7SvFDr%G^b5&!$s=E{ouuavIx%&{ z!|BAi(NufWMfxbmiHZCmFs;^Z!{me2Soit{4pDx|XeP~MxbdcBN4YUgy%de38!qFh z2e)xf=Tl5EufU4m*`y=?BA5Nef?8HN;X7sv*t@;M=8|_<6$^_HO$Ge#jldM=on2U{iosJ0q}tlL2jPeoZdS zJR!KzS>Ap4S(Faa<{z6r!NkKI*e0XI*KUi)M{h2od0P>#^Z6%u9mV)T<}BWQ~2`LBYLoWe)WjdXqTsAI{$~I)>R_mSW73 zG<5D;#)sIC;^h_R<3WMzxukj_R!*GG(SdNZA8$wdm4)Z-vlH6SFXJZYnG@xv4>&V9 zirn1BK}6FRTDf%rUfQe5-=A}ZtWbH0k}t|Iq52B8EZ7Q7J}o%a^e1(FrGlS-{GsX# zJW#_$pZ7iod_h1OW~5m1N5(usqnRy4(x{g%zgdSWD>Jbr6?y;Rf|qJ~H*Rb7z+9#hr}r9#oLra-ow&oCeH}iG->;^F-(rMZ&AQp>W>gCSrxi&?yEeGJ9)t7y)u^5P zD;!lh9n%y;@wNUcoap!iTYr70PR|l>u8I+_ZYaUeJinZ83LVN@j*Z61l9|kosb1*B zO7af+S$O+QmEfvWM!$FZ=V#aJL*7=wt;H$u3w}EB{T?oyLq{VHv%XGpY^8WMq5*%t z|41~3t->Q?ZFu)xzwo7JI4*WC#qQ)TvE&U`42nNSuUDGmU}6aOCvrF{`EJ97_Vu9g zd@VL693vk!h6>!6Fq&YN%r^$72~L=~SW$lfWLFvTbt}g3P4|(z&BU%JPxZ zlkgBb7q^!Ug?O6;e4ITRmmaRb=o>#!t53+dH)KOv{wuLIZ)A{*HM=8I%l2 z!kP!aP&4+vkn4Ml=YJkVu3B)isBFf`)3-4>!oB>;v}@uh%PKP0t5RTQ+LAvfesbN4 z^P#&dl=~xnjk&%_mfX}^jBMcuqW3DEGq^St97VUt&k4TV=iq3%a%&fUx|xeF<|*-u z?G$*e+gs4_p$oqGDuNeDD`~p=H2#>U8sDMrLN@X%(aS54Ak>mI)+(4#nZo^yzE9nM zkH_oA?l>-AZoRX>rYfPA0$J`;_HcM#xtjUu^N*gkSD*(L zpA*ZUA1l6fx|*g4>@_EYV>C(SFX!TCj2TPJt1f9;fL6W(E-p!;v86|#d!sKNzc5u` z{kPN3$V9HcZ5HR)YeD~O9i;c0ji~gnlMLs6S8Vu^K-S1;^76|KOxmwQ9Lo~GdA~e1 z=E(!AKL(y#nZm8ni=gc6L3s3NH&hHcN+undH|s*1MljuRSem-#ET9i`=ZOPX522@oEM>sthiJF* zAci!!qG!x6V)^w9opGR5dP7qHRgevVgBll8jH{1%PK5N3dz0c^n_Nm-JSuY*B zP?IzZ-$RSldg*I>b$Hv<0S4jS)N;EyjW$)}XS=%ac4``AgQ6RESAIA>hjFO)-WOR3 zdm*EiOl)nUsN#rqSaDICXgR52R89;&E@+~6+y7A8lzr$amqqVem(Yg~BI$NL1EM^Z z5VaSXN!15zkA2csTF zlj(Lgw9~m&9Ba*TpWBs4a!LzxW+f1}@KZ2uydi0nJW3zN9U&Sa!CYC)TDm3if~a17 zHx^8p1fe}inChWV&YshN5U*?G0`JJJI8=#^EsngWuQ7xsI*4813Cc*$!k*e;7`<;Z zF7I|0pw~Z++t;v_8V>QrQHsF~6HqF+!4C1>e;?wOn-7rh*28H; zmQRt7LFpS4`Af$8Fv#;DW`+_}bQVs27diXmp(4d+W!t$1g(Td20R6Fr}HQTGYf zsCy?AmEC9XPAx_J+uccYy#Vvwf5r^QpW01}KTL%MYfgfZ#~9W|b_FXDG=?3lKElt* ze2D&GmDpkan~D9Mhi>HyxL+X8A00f+*WEqHt0YJ9_brt9Hx|$7hC{u$Na+f1en+?F zyt`?QYla$0eO-q3n*`^G_DH(#iY499lZWTEALHp&hjHp$BRKK^A^c7g8?mX8^;w>R zGtS%aU*1^rTe*8^JuHDfu-=Y49*!hEcGC1`O%}ujmGE;^#`9@4kMNYI0eCx~gDorC zaBQXrzap-NKe;KH-#yBYH=j|!e>{5*BPSliCHD96*D5!dk)y^RE|TKya%7=jPshf` z$AIKkT9~0PFa%3X@!0Kl+*5vzT2~yxUuGHPtIc6xCA`>;f1HJT^0DmV!_Ii# zd;-2JZNOGFKV&XJjpnJ%eAlpi-aqUYfMgP+Kh|W69+Y89x$wPRo(I>)mopoFH{j|2 z=gssdVDF1uk}_5ncJDdH?_Bzi|K7EU4?Gjfn?W`|&2t#NBm7OFKo*Jyr6;t}m zYAyx{{-TKi?Znl=mR5MBVe(EB{>#+>-skc@^zN9AKPT8>_vCFjbAcL~s$}6x%O&VM zFoAD>(15Du`*6bKM_7C96kgjKhH|eviCJwg9?R-N%YcUFb24$Wuf{{`m!NiY3pW>^^J;3bywA4Jyv)CoeEN(^-sXBZPMH@EX8)~#LoJ8V zxg;O^2#Zc*%g}u4aUO~!d83<~d1odbi)$GiKe3Hw*m>dN+cT-R?iVVNaSCs~(8sWj zC17@SA1~K?n4kN15ASg83^|i^leWG~=UiV;!P0F#sFrXH#|6dF8_!NrUuFRpdUH{Y z{fZ?uH9AH$R~4`G6CIZFvj6`K=X9gYJH+@E$I)qXBX%s)VzM|(x_Q)jnsmDpUj{@o z!H16F4wrsBzRRD)Iu+72vif+w^fIP8pQSZ<2ZiUd32*(XLAgc#pz)SR@G#=5JU`=O z?kvuzJb?Y2JkR>=#9N7J=)Pz;uYYC|e{sx7lsV$UJ6{;Ex$?0GpKtO*^KL7?EMz@z zo$(Y7B&P7edy;s$M{D`3O2Ait)~*@TxU%MmEyvH;rpKr2Df7qItMmV5Jw~m0LxuC# z6`VKx8s5%M!_@X9zFaO7Uw29H%R?sd{s$YdR7RPv+AR3kU;Lmy!52^8JB~dK7x>Q3 zVKuEYQ!xAce!9YZ3C1hslA@8-*m`s;mh9S%`Pa_yZ#;(9R8?N(2Traa?m{O@>bbyY z5?z9qVP){ls~3JbOELDEJl5)$^EQ`<)4%tm&@}KL9e3?IXgfTl&mTJC%Uc(?1Cv&X zYNfQ9g_mcc?%{oO9v*WI4c#9w znpH|}q))ay<3cUEh}EYB@ZZ&=WY)ntnv)ZVZr=LnG$vTc#I=z*ug5|E#&&wD|0Btn zY7DkHDs-Vvx}Y!`MIXjzg873`sQa~vSN755{c#EA*{Q)YfK9i;oMr^0Gu zEjYMiIMc;wz>3$$=}fObjAEM$c>EI1D9V@U!?d%sn$)0o>|Ffmo`r;)D{7oy&-Udm6^#FA-)G|wk&JCY~d0*C~g?ei{Q^@ynD|X$F_4Yvp*MGJpPc#U(zV8X$I~4gxTuUFEXU>4>el2 z2QM20ldBGMDZetH>T9g8ec30%y=m#VrF1B-m~V$`J-zWiG8C=^G>N)aDxpcn z53c9(eL8U!a*L0+;EdJs__+2FeeU;0oU8DhK3l$1G*d^r%3%FY!Rt5|yekb*cfbHU zsaV=9bq3`%J8au~N)cdfI%TbcF z?mOAZl+Dc-zyZpzIQTtHZjGT_??^0fXlMGCm1CN@8D?x+fLCYfakfnh@#Oo5^x2vD zP;c>&gv*5E*ooorVL~3|U9>@=M3_^mvoMZu)l2n{Qu&Q5X!?|sOmpK4^8NEadRr+Q zkE}?7@-4RT&hROBYHkb~&mQ2O1qxaDi?d*7a2fubaf=*ySx7l)H}c$OGmbw?8CQYN z7(K=wV_vpl%woiMi=%Mce>NB@GZ*s*gRx_29I^5rPL%{VPv2K{@qw+AvE<PmEtAlf84~q5b^=)N$O! zqem(%?_S8Sn(PL*mXyG5Eh+Kqal2r#(@9#f?-(4rCq^G%hR=&{#_{4@-q3V8zj9EE zF0wBJzoK7c=YoEa8g~z5XSqWB;>Qs8buT9VcLOC}3wcy^p4>S%h*?$l;j8lEZ8G|5SP$y|LjnIxu6rT3Pp z@=~&2=(uwwaB=Mt_|O!>{}rq8)9hw2mPEw!gYLEojiK2sHl=p9YuQ1 zhtMurE3itsK;H=GGJPjk#@l%#-!wy-EmiwUU4&UFm+i!vXJ$gm>=?`Tz;rd<1CBB^SD75){9jMRsf`AbsUl{P3f-Bwf{m4@%O6h0e9mclHrxKR5{e-;HST zqi9^LD(pOe3Hid`3q|K$#_{%wgXld$2cNrX3QmP%cxv`)%pY=w{;Y2z_v6PfdV$%X zs=JEn8ZvNq)LqbV&Lulngo3(ee|7Bce9kUkj~ZW!g}rr;Kz-X;`iOCdiscrps8a_V zUnF3-N+uohA?bw6v8G|_erO?j~+W@jfiO|x^!C=WX;^kpO@+Rq^>#V~( zN>1iI*V?nv9l&Ux>m$CMr?_*|*OMis57FShGfg$kMw25O!TOUq2CghbBE$Y4$$ zIk89vn&-SAn}^iGPhBfv#+wDF@1=8}*S#SwJ$A&kxtYWc-6CH0=`aTYk&wC}k?B`+ zrtcR?(^1`#FvmBZ-LN8x%nr>YcT#}$c3V!I9_5nZ(-yLm4a-^OoF8P>#uV05uM)h+ z){yQmnWW}&HVv>l%s3wr!OR;OtZ&|AYFw7ZDzuJbb+^aSoFR|d?-q%`Pg8+K`+mc~ zi0@?6_wQuIzxB2kzuy9@=`E0L)(u-nD`V_DTi|}W!04%` zv762sRfRFj#NA}s< zPoOk;(5Yf;5NtxHS(Pz{Cy%m+mg%ymFNG1~(iq0r)`NAH(XusI>IF|8>4C5H7_hu` z1g`cAeVpMv@H=lEI18)-wU+NJjVy$eRlX2X{+y&lJS9aIl5FTPFWag)Gg*^ijZjsq zK=0FwF!zuX(VMs*aQJ-^aj2dBeM6so_H>176A7e^8A`5y3L!4~+0yp*-SKSN=37+q(`NS0xHYy$8gp$8wl1>WCH0N9)|$=keY}IM`0y3pu9kwO zx~Exk_`Jw5a2hCe$KUv8UP?{6zwJPkII9;(~8haIh`$iCwgz;kP& zO;nVVZH0rHZFTy8v}lL~HIUP?4XIORLwAlQS`G(j@x57aqDx>FuE}GwOoY5`pP%i% zm>^sKq#)bhp%V0V&=!)^o5tNP5YDy&tK;RTK&m+*i*<;xgrj?G+1B-Gv#RcR}vF1AKqmPaMuYgl?A+wptr!+uqr( zU~9R}%XandXy|`^m{r>{gOee*$+p;wBr|dj2JSd-6O-3QzoxIF?tc3jukmUm`}$B) zs9PjHp}7ejxc=g<|JO~(lMhT(q=s$5&v-aXQb6<}i&Kd2*esbBtdPg5)?S`L%|2Aim^cka=oW!ND)KDAtP48u__IukFIEp~+QxE;< zZ^xR(B(gWx+rf)rr$8!TA$gH0Wjo!vj$LpxAKEM>z^5z*Hu!I1FVCGwK6;J}@&Z^@alI8k)+kGO}e$uivL?2c=-=fmtLKep!`XQ)gGn3jVLbQ1*`cJ4iaYo2*v2 z&PZ9EU>n+k+4lQA?B}h2*@3Z=LQWwR6rC1wcbkKm@O&f` zVWQ7-dNwf%+E!~4msPgx_Vjr6S@9g!pjVeYAdUoGNm;g{s2VK$lDL?3Pu#G$oJ>1& z7TR$G^sct#qdz<&=WCtFY6%tGnd!m?UYW&rj}v%hBUZ4RIdwk9OOk(3F2OeH7SlN= zuZgaaba2!E&JFetfz=KE#PY>RzCGcJz#h8;a~6#TlUcLT?sqF$dgcO@zuk&EOZshW z%{=#A_Yck0C2dxq|U;#onP)c)ONK=sAB3o>>>pU&xH& zmDDZy_csoat&b0hk1dg7D|WjHoK0_BH)jaUlpBGvX~BX&;v#WgR)&)+KGS_3A8^E8 zeQ;OSk1hQkgfi^MoX86WfEmA!8us@pYJD zG!`~3DMcoA2qvbc<4nIY7`9WAjkKzV4z>uKPCn&E7b^1aWFL`f9p_2F!P7A1{#7ck zeVZyf}GO+tIf}cbO$*_BAFz7piY%8z@P4z_}EKqP)NhMBBX~!+$J6N9| zZuE~_H2f$mCjN_KVb0?-<%I-VXTk)f>a!c2UiLvasuun&|Zjj^4LP#QJBR zgv8dcCIWYynU+i(IB9ltDnueeDDiRVhXn`?K)2SRIBPKZ@(>4-plgM`V6X z0BFoeqb{y-@M~KH#C*6%p1w=N{E=V4Gh#L+L>bl=&LzVm#Gs{iMZ}tqgysorpt;L~ zTHc%o`7ia@_HFk`D5F8DXFXt)eiqOoxgpFogY~TU?!P3^Ac*@h>M6|?h0?X&U+EcP zcOfb4Z&FLo!YFMySX&nlK1ajJsU%nQaEK6__xga1eFNjVOG@Q??192wHN&8?6zGfZv05PyY|5X*Gphi7X*>rpJC+SYOYf< z9E0BHqTz=o#?b04{gm|t5BoNe&UQ!Ev9uGI%ReB;_Aua#eJHE67!I8vG-HxKyeWHz z;u{~ZcC!@ArYB(A=Qm8Ao(k{MY)f@QzS51KMR@P{3D%Mf0h@hmL?#azwlhMX8{3vi zlWm`%YyD+(ms!F~pGd{Kc|KIJI1@%Ty(6|8-V^EE*LZGt8+`XU00z%P>1Rs=STLfD zRDKu&A8m3lyRjAi?=<6Q2(F10M~VBN(d@pU5@`84iNEmL0X=`#vF*w$x#0PlbkjIP zrsiL^sA$zyZvQK5lDB*beh9tFw7Eyo$?avNR*DkyI%(Mc(3{L_E+>%|1ynSzj=A#h z4->2_C%$oI7=0|X2dCQVke&&jxU!#(j6u8_dMdcnJADhN=E6SeJZ&7EFJr;EJGi6X z>WiFtOuN`mU^*;}iKV0K_mhj&O(gfSCe}~vVxE@gF@Ifqn8r2cWXY{0CTN-o{VtBA zA78f$%=mU@PS&q#DXpj6ffcO+gXTF&-28$_*F};ur^C4Zji;HH6Sr{>9v`HYj`wNT z>=9)3*a_seo*FFdi)QXM)`}EvOciGS5GGf=o@-l98U2OTy0JxFWV^DOX>aS|79G|h=cb6cFMsWARL(|Xa*YaovP%)ti-f*{`x>sd@Uuv1 zRyB8Y)oW%@U|z;Vuc7Ca6Uf;4=a?W{FY-2K7*)DEC=#EGC3as8aMq`@H2h%!qu)`* z)i5&|DXCMO+m&X{Crc5y-8?tbi6MKVDylkd&FPB#$2P}TY@k(3_b^ZLb~Am2It1J_ zQEc**F;$p~4Nlu>|5H7P*O*5dr;Q`ICC|7Br!(9~&v{hl zRS)y5$ck{T-n2346sb(F;9~A@+^b7PTy8@mGqIqX`7-VTaZ#`!`itI*)vip0CdO5{!^uc9&@?=cTq&uubGTGu1iF&4AE#aq9qMt+BqwRlX-QT zhE*@5VMFc6qt_8+^Pz{LrQro+y-9$uEv+Q;nR?P{9zpMh4d*tB-gCj~LI>V^9Jz4Q znKOx+N}QT0lT2g>^NeT2rt~xbeDn%{~ksz0*+!KxeSjzChN5q4C zJMprJAV#xCFae9BnTshdOx~(^GI`DiW>V8k(sf>s6j&tSYEKOqwRMY-4bp(yPo%;5 zdm_=_bAi+Rb(H&Px)t90Take}9lY}C5OGbILIdAB(8r2-LiX&d&5y}dnVA@1fu<6s)xlH)oS{{n?DAIF;Uc{DQGo)u$JHK8OtV zwF>Wn7P0r&fo6q;Bs0BbfYa9)%JljDNi82ce3F0V^s)xp@ep;#e`4DtO5sT z__EFgT<0V~yunm(`4a&fL;`2~Y8(;WdIh@kl;G$BbJ#WL04D-_pssraIDMH!#ZSvY zZH@+e_|^wnIWohA*W{nR1I+iXV-9-FeWa`NET z&>FIL{sxE>oY20%j>E%AH;Bhi!9FP&0-gCn=hGtrQlD>t*tA@z%lr(BcA2mPgU`qa zoqmY-*$=8ew4qb`g1|H!!Dh7>Lw)5?(kL+fbPY!_C+;05_e*Y5^@_9H$ygS0h7V!= z&qzR+nHDVa&;&2o16@j)U}&l*Y8)p4AJgUlFPhBQOm+na!wa1J9wVYS`Z95xvlRTl z4U!k{8TR+sG3>9cN4fsL7r-E(3X&GYl6^^k$bpXA0w=$U9KN!M{ydjT)?}W6%|DmJ zxv6Vmyv-HTdb%4jevW2M^Y(yN>s0!PeM;|~(1F@BM`6Lve{kvJc&L5W3oS=KGFI)k zY3lWfQ2ECdB8OSSp-2_-M7M;}ic9dLstdHO|AKeSViIJ00~j|+NFQ$ir&oDH$iAc8 zxZ2~O@Ma>6pCwCbp%Nqok66zpZ8oUG8 zZM-~`l>Gw>rD_Nrp$n1{3+PUJRp3Uaz^{sSa!1Ypey=|W4yqr>Ufz~0Zgn9Nt~0^? zh#hP_9}Q~nQ;6?Up}%}h9wh!$GrtCgz{HBN;JWz_=}1qf7R?`tWd3uS`c;zz*42=& zHKUNA;qErq)=$_RFMdoi!8ro(luOXRI`G#T;J2y$kaf>lcrTwU=Wkr+5mCQe=gD+@jo z$rFjt7qJG878~M&!VGvW*fV@X=E9r@x!@-wEcRpjh(62PA zQ#_8zBWIWfv4}2TA4Qt;>lwv!2k90y1zdRM6PbRuf((9~4iby{$c~OjWXdoZ>Yg*e z?KrfQyjWikBi-%UtY~?*yr7zhpOr9Tw|3GJ)lCL|{AF%8uOh{I8)0dgB71J_X=weO z3K!<2kfadGs2sDR3DHa8{0$HCe5xy?U5S9O`O#3Dw~_7E`N8dQS_WlzPQZmPW8p~f z5c1<0g(Jz$@Z{Y|u(~^!j4LvPM?a%L(IK3yx0N8x*BYVcU=5KIErTC!c@RGCHN+kq z&#L_V3m&svaJ%tNJkA_P@mUA*_F@VN@4SoWyfg5N$WgddUj?sykKx%T*T{ht{m7ov zVVj<;CDMLV*}_%kY{#ej^ssauw@tNqUNiZd)v*c+2y1hg`y+hCIBEk!GE0 z!&#fu1MDaBiBvsfHkY*SGLBvtMQ;DiqRXou(X)HUqsi~Pi0mlb=W&;r|4y5{l5AkC zH;;v|lj}u0)phv2uL60kJ~8h0tbi8{33T=pDSksz6b)Z3g{O{%!>HZ55VK03l_Rml zEYrA2bqqXdMqenNNjQT8nbj~8PQb}v4bnN%71CA4gX6plm^v^Q zqieU|nM2QM!FHh!UwnhCoOlqI|Ck6)j#+f@g8_^!&!E#D{0HvFzGUw}A{f015cl7% zf*@55Hq9%aR=Wg{U>h-tl!Ebq${$)->5dWi4B)TeX`iXD$B#+KVs$dq~C@P zpGL9YcE{77v#l^eE0bQ{a}En1OhM<6WU{MKnyncyWmgI;7sK-g^dMhPh2}m)_Uz`W zG!J9Is&SYU5D53DZh%oGku;=5p3m0vBB6)&VEpK#wAY#?*JM?(!21Jt@NObguU3tB zMH#SbVH>%5zJvVSF64{nS@SQKM&q>AVz~7(2et@#%e~7^GE!-WspV8d_D`Q0Nto}1 zF-MA_$>0z<*jr8#E{wowHUxWGW%x91HY$pqQjM8YX!GtaICA+Q>=gWl`)5jG-nVY* zdVLG7RJ%fNIP2oxHV=|C?kfq?Z$^2?eQ1>-!aGk=(M_@emS-a>?UjJ)!8G~;3P7=b z1ICqBkh-a!g5%;Z9{ux*%>Lt!(Y!5;DHK@4TgLOxkCs9J{|O)NyMqZCoy^FgM`@eD zQr%bmj(*#jhK{*e0I%yXQn8xY?wZSJNdvC5{STn)G|fCU6PJ3#;?beGX!Xhmi@ip{ zqUc=gSdokl9`cZ;ZN!XgcO~49AauVFLaY6b(Z5@V!rbzOWZc|JoVmuH7iUI*Zo?b6 zTB-#9PMYu^Y$^7p-JtR(E!inn!Th@C31mK)qlIlbb#6OA*JXadw(XaY+^@n*lCQvZ z?m09{9*_5yU%<@Omx-pE7DgrS#45v&G_SZsRLBBua5zVl6@5uq$q7jRu!4HtS%T*i zrGPYV$K+Kv$#3mmy7N{BCTg?9d8ohDxg1r-v#|%u@-=w-1VoLav9$PGGD;gYQl3VT z)S4+M;k$|^$jnEHenq}5U?SZrHJPs!nCNOPO1y-LnD=_K8t)qHMD6J{IQn@ItPU2O z?t7i+n1lOqcZ^$IisM)FqUM1Sn6rBoXpeeDevf?# zE|F2>{TmaU-2Q;H>=t~lAJ-oLyBaxwBOW9>8 zAGSs>2Ny%(Z~{JYV+&C%u*7#Yw2}UkYm7xo3GZpc4g6d48q@YY7B>El=l1ynB<$`O z;N%$aR;uB8sCwu@M;WFwB3=Dt#fNt8SlGE}Wq+iWp583@8 z3umdr(j$_n9aSv}a|0~RDd zM76+V>ukJ@K4%taZUO(>jxc=33H%on z;4_0JSllZXB%jQ{S!T1*t1F+OE3grIR4u@<`6P2%pb$)t2tuZ&4;ESJhRi@n^m zs5~VVnScKUiswSX&0-m885U%_YEtp5@G)4uGXuHLRcEvB$$-OnH4aIc&GLT|ysb?c zZ8a&N9Y(D@k{}&w=~%FTqdhk!@>J+5!s5LL4dNTx!RBUYy>X`?(a+| z7e~2_r;iiPGRXxW&j#|*VFh?GtKi!wNiu(hAJl!S0HLs}V4%wR+ApP(7W-x3&>RK2 zhC5l4$gkvav;k;^nc=He`^iAcM+oT*h6t}%-lX4IkSsg~I~!Z5vPGd#3A#`;Sw)r{ z6oQ_NB+_4ZnI!yZ$L7ZQcuCto;&H|Rv^A0$OA#^l*9uPvE)rq4ggru<)7sHihe*&L z*}=Pg_9t^YW(_nA{banxrsH>`6!DWe&$v`Ag)ebUVEXL?SUL`%1j96p);5!sL(9OT z{0|zPNP*Uqp|JVDX_&Y32k1q=A#xSF@Q-u@l+8e0lUb|!srH~%kes0bcjFUD&yN(MDin^i%>t14>3y&hWkFM> zEqiYKCUw8M6!y#lCmKhmnB`dG=lTXKsNQvhX=*|9z1{4jD%a`4xzTOFvY!<=UUDlLCXERme$GNQ% z_dtA@DiXRS%k>i0!rtr0;mC6V99;8=x;gg&y1_N~t6leluf73pUVfdZDMW)Eq%%Eh zPJnb)7iD~QBgD0OGg7h6P&;A>N2Si9@VPF~erYSZ_cNB7Jn)*V`WwqIQHn^dWhO-Q zr&UyXPseHX(%`Ty5R#1lL*;Q?-g0RNeB7+S{>wK9r)QPO=8_ypHEV+d_j`D%eGWKc z4VHYlgLJMs1B=bIAj>tMIEn?CdwqFa_5C8(BelK=8J%+>`za4>uZxhII_7Z8J{WZ@ z<^kfBlFel`XuYfe&U!6@<^PPK-YXZ#Vbf_)?VJm_dI(#;P=>l|n_=rn30lTEfgfi@ zGi^p`kQOl$UeAa}=?#BS(~f55zY&&acTF4pa=1lQ{O*uT#}LWM+63jyUsAd)5)9^~ zkP+u|!fc8AP z2?c^Tz-Reg;`w$iGDf@5dIheR;joN<)z|<6oQkO6byM)_^X)`(vlL0q2?J5(aIoBA zg8r2p1w8PBTo=j!quq^&FcPeJwKHMXXYs_hZ$%gW7(z$T3>?tSWpnz5Nl$PH41WH` z@ht>l`CuY!^4U)ciz~t^C&T_I(jIJu{5!MJ)UlCv9ZnXA34 zK&HEk_+@=Wl|}L(vpo#W|6ssOEeygsnJmpYT-)Bh-IE^qqPBA9$|P({~{ z_7lFIIrteEkX+9g@^DZN{cQh8q}?>(`P6;<#+~(5%hdZg7F0em6_f)-(+|{$OD6Hs zFa;%X9rIhuzafSCsnpiyXK453Qld1_g8Ww0k<(3?C`U#Tie6>GKA+{B^Ej4Vx8f+w zd>$OFen;INOro47Hs+2*@C;^!1cf68uvftJ*p#F z(+{H+w1sHknL8uVaNA+Z?XU=U z4V)vHTu;+Lw+qES79d|sp1_f|1{7hLha}4rm`3jm(%cscSDo$2qmgOw?953#E!6>j zFI6X#ulz{L7(%1Zg`j(eKBc{TKIAB~WLGA_MtjYmN4bt@6s{r?>jF{G#!blOgb-Qt ztr?jH&c$itGsqNiAFjLHho%Xuz{3Rx(92m;OeOcsD5@!EXx{^X#qJ}^oeLR*usKM@ z`LX%le}6b|kQF2SR0AoNr+}vaZYUYx;|rp3#OOa=9AKBk5Q90W+{>8!`uCjJlzKwk z#1FKpeUkiCnF>?APl12xQWAQuo+$F>qi5q1%6|+WfnoXx+r< z{_h4Pd+Z;hew4y}feW>E@GnuTi863Muk zHd*4LhSsDNqqQ>67}sU$P?jJ^9%-FIJ7Z1ALN3?Zwf7y``sN7k{`wPmi??8HRUZ^{ zy^UO`9U=GIIPWp_nsQgX1X>M85YG&wC_y2Zc<_y{yi5moA9#iqLnj)$y_~xL>?L{H z+=_~$jG*JT8Q$>J047VMk>JeFX#FFu`?%hoB$-G-x&JNdoWM>h`P)=dyuX&TTD738 z#0hdC(3U@paV; z)7u(Q^7`{+wd_vFUGt7C9+hWjzOlu76YdfpiFvRoK^8)OZh|*4>csVQFO2RB!jGps zrP}8=F}FGQ^Ty99=F>(p$$@nb8R47ZROe!j{iAPEA^R4izH*SMpFAUxf}dE@cZE8sR^meMsqe0J*Vr z6Z}we7yXSf#XB2^Zmv!PL4Wf{ISdd61E7A4TVXQK) z$==^`9!@87+=?B_)XuzL^o()s#Y1PNEnczxCDt4ZL7SxXu;S7ia3^9uIPVRE2NrMe z@;+xQIbIG2-PY2{O~P16!k51CX*PB6m=MyZq{yGBIVAGdLt^PViUi_KF+ZgXzbUzi z6k`1Ed z)leoCUVI|01S?z%!e8_jTq&OBPNayLH1EMikERN;0e4QgloacVn5WZt~TW**`s zJWdlP9Ume0^37PWYzmH(6UUdw9xw(t2W(Hbp~AdadfkFm^rH(`nF6U)928xT9W5I0 zz~f95p!b)jM;B1ulAaX0T2FkfWnll#*F@<3(Ta@O*iE<;|RVJmWhvY8Vi)9lG@vePB^z>m_xMzJI@6<>kCj}_F@#`;Sb}tGy*Tkcy&wj8% zViA-++(g$^X5vptDZtZTj+XXu{ohLg*!{T+UQ*~uF8|m^=G>k_!|Z1`=R+lKUwjtD zl+U26S1hMb{V}9P7I&g8{8p~F`j$Ru8bpsCF`(aF_aZ{43s9Y_2l=q4pB$D^rNew5 z;~6Sew2o~do;i0L_AFfi(`Smn$~BGr37J8(w?ZDD7IDKxU3ajc&_!fB;(;}v6ROdA zH=Y&wk$625!2=PW$r`g^WbjrFzxpE1PpRWNY-hr;kYyE~6bZ+YeUZd;P!2k8I#5sc z<$=CP98ubui82juldqPq$&S-UiJeI;eYNX9tS&7^YstvcZ;l&Nv$BH7G@V;G{q20{ z=`6;Fx)*|#3=gw`14PwViRL+2Gu{biIK}ii+G3)BuKxXwx-Yb1o1I7T$}gNZxiXr$ z!R~@J(f84Z`)QaF{En`RHQ|}(9`kSSenVE6{iH6Gq>_&*C1`W{FA_1VO6PX?;;D zDCD#6Arbe8hmRkou_e0=@aMI<)Niee$UbTo+gjKGLeKWYpNq}-RAn42Vjh9)7g6Yr zc+WjQOW>v%$Coa-3%94Z!#|IFXu7`^#B2VLJl}P!kxnO(f8YQs)$52&z)^^#YCzia zJ?wJqfFni&5cT^8OnWWN4vuVubL(Rv`nV$)CNF_=8joRU=S7gsEP$rK6|B$hIV{Ni z0>_XD7+t^6+~tfH468cwY{Ke@$r^9ikhL0a1Z{(tW#MorR)XHD_nkbSv5-7|vkx{k zJ^;THXJP8U4)mgZIqVVCA%{56oJMgWx$-mthTpP8!ngvo)+d41iwK+-KOcrRR6)p> z?W8zR3Q}+H1M8xT&`(8BE%nx%1MLvspHJXHP%5cyO@tjkPC&PXHGIA!48Nb{Bj08V zXnDULcqV>W-P{-CKC83(C%=;7rYaO1!Z|sTmJtIrM_B)NH!KJ)C3(@0!2Ic3;Cp27 zhHNq+d?Xvbl$V0k?QLwp&P$MHJV63iUVxI)NaoVoQnK1tgG_!;Ba__$Am0)T^NR(c z{cQlG$XURzmX{!yZo+OnHU`D!x0vqJf#4}TkEA~1lMQE+Vc_&#h&;2F?I+Mnit zw__X$9;k(R57)sEdQA@8jp8f}#-KH2jL}{E3MMli!M*}%Nb%*d--|Xt%7xRQY?Xkj zG~38*2U{4s_y^P+9=7w8|m55-+Z z&=l4Ly4DuZDJ#xKb~xi}MK4fj;2F61{4#WSM?lt8H@M}UNp=UhgzzXFk9e%z{ac60jA&%4O=gF1FNa*gSO~xSrs{cRHAI7dZn9o`|srJxV||KARYF z-Ioz24Lq)$g#saQ=m}4SJJM6x?Q0cq?DQRwQYH(VA85e?|0bw8TM8R?Ol9TZDbf4I z^FJ?rssSj z^VVO7(wll@mqP{g?`|TWJ*?RC+WUIjHZiuF}}1QG?#5MqB5 zo@@6rwYCo6SdvINNOSvxsu-|-F%?R8z9QZJ=Rt7WLN=|-6N0P~$l|g27@;4Rgxc-x7uOgfqZ2+SXDR>?^P7K@&sp$$Ppdm2?^py!xa(X7VGRuK^%jbc0 z>^->ZJC8jpDv0wp#^Af@iLl5b38!q+#@}~YaE`1DEY|;>%u$G;hCZb;@%xOxLvR{) z+nR<2&psq2i(Ki2t5Tu!jV7MT^Tlo(3XzpVHwS5~gWl0Nj1M$nnG0ue%<4L1xG@qk zmD`E3-AeMpK8K9yZow|U6G^|j9O$O~fJa(kXkc~~juQyRMyAPFu1X&7-P}oLj#<;| z+!x_a-Z|WRx&^uYxQE-hobMBL8|?dMJr?=3f-0Obj@2^v;p~D*r10N3&*KG`72m9g z%|z8`dutY4JZ<3HoUi0_#}}NeK{zk82>IEpJW||jK;-=A zz`HRI?5JqPWo*arj8-w4_&Q;?8Hw1m!wL&{+u?~#-YDmQ8QlFThfNb3A<`fcZ@s>a zG(LKP9MvS*^7=DyO1=ZRU7m(zoE({7mucL5I|4G}7vdZL>4KHebb8+*KF(ZK#@l6U zLm#>pjXv9|5&UfeT`W$7^rKl=L%W5so9mDJdMv2aQ}gk9#Xr=#a9^ywKMyUKRe)YD zl7$qNAsF8lj`G+wkSy{DEw`S9mKiyq?*II8#a=b~x7jP~dB`1?glr~%+LW=ezyzAt zTnV?mr@%RmFS~BnF_e^g3*Y#90o%lipq09PjK}yqFlh57s8@`B5&95Kz2%CZ?X3g% zi%!_VMh6!?F2s7FOcfRDO?_0dB`NDHkWkrwbo)a=Slrx$&i1vC&gbq}SJfL;Ifmfd z7Cks>S3F~$9!ml_r`KJ<_4uP(EUvk4fQ}j%ldSzbxY2W%s_i*|``#$i>0E97Ps~xg z{NzUNyDNoj701xbFf~%M>fo1NhQ7?@ll{Sih-xmT=AZjO z`OkUC-{M1IdE-=E6ZH*S$3DeVj2%GW*E%e|Xe~Zgl)#^2(uW=8eW7(-5}uYRh&OZR z2a7KYaqsvVWNXUeD2|0%Bg*YeFaC!{uZZD$a0%P5mch1)i*c!zFYXeoA&J>mw8Gk2 z{OR<2DqYJBe~5@d3+L70R+o8rWhoC^?IPj2-$~#=7=Cs# z372ciaeb|PRD3gqYWn1kCvtM|-k00K|MzaFJVBU(lSUYiv?0ToE%<0f5wwqZp~bT< z;P~+m{H@E&@xX;uSl+xCzn9jd8~23J2@9if=Cnf8UJ!`m`BKmtyqN29N1+R9kuX(% zIdLP6k z44r-PTF53<-Q&DNPanaC4ncOoeKRt5Ner2P+!*&f-iHUeuY*ceDxP(^hpc9Af^Fb* zEU4|nOonk;aV}?70HY|RN)g4Fw7bJm$ofe?m=#FD{S*Uc?jEiu z-!%;-miz#YcaE8vw)kWCF|uoRF$qpBr^FLRiPnfDXlibTME!hf$&?}(o!rbWKNrOA z7fl3}(*m$o-i>wsUwG}{`Y$9Lp-SdNxnTdsT)Z*qB@|d!z{97>klQInIXzzhFY3o3 zs$qn4po_9zTz6UWxgg%sR)BlQmVkDSKa{AJF{Z;MFz2N(2)Cud#~T+Jdvyh14&G+2 zC)7jcm^Q05x`D`z+=I?DX5gE^?YORQA&+J*f*=3+q3Jt3iHYHMc<{BEY)CbOZyUNv zf`Al0Y&M@k+tra@MLcM)--ZWH&t*-QNU-W0_qo)k3RNm`@0s-vm}vQ@aMM~DCSN^e zn4(vB&g*;#4mgh(#ZuHE(N6lejxk=*fy81^oUO3h53${|6n zmGv>K6g~r&xkTXguQ>k8!tZ3?iB@7Q6iJ%7cj*bAd~__Vi+cOX29H0uL44d~!EidC zc;@T?MrRu2N>zjWZ$A)y{F0Hl-3o8l48d%d2~?$a29gU8f$%q5_$Tv-OoYB8(?sG) zTe2+cuek}I&#xtiTdK(?`B*0Dxdce zfi4{wb$>!d;Vfe3sR0)i(;;%M9G?0{mQV3ENU=zLv$9=jM`YQ_xoBdr@)J2T z`45rHO0X~ZDLi{8#eNLk2G7FkvC>9acC1$z8P9G34G{?#8sd1=AOJmokHXdZ*;Gh& z6ff(51Ql~i43q}-h|Zl1c<1^aJrsCKR+=lpY&muM`;Y1PuG1CrCZiH*4csJ+^+hD{ z4ae6n+=egOenAFd>%c+#Jf!KGvzM<5quQaw}#$qwCMqhv$vyg!3~#zBWY z(^pRt^yi@3i`8V~#X|m}78}ZKYYn<7&Uu4pv&dYwmtjk*`7v+0`Ar!NB~0z?HeRu-6Z+~aj($IJVZMqe zkcI+8l|-yVvCj7!b`E4HRjp$$NOZgfCJT!|N8j$#hM=re^JqWds=+ z60<*tbF(j{!qWaQrMh91!b&ryE7p{8&|E?e#s!ce=|r+dp34sl%q6SClF=mZ7fQJ7 zM=7e8^ZI04$)Q#0RCI!KM`0tQw>}0f$+n_wx&u+tpM7Yvh!$^J?ksAc=@Kb_O);mQ1|oH_M5YE| zevkNi(op5gn=38>MRE~T>c77f!=6Ipw`Zbpja;&IXdQECPzMFKFW^ln=p=q-pQ)); zVE+4aCAC4Xjj6h+%0IJrCx2SSf9TDDgXp2|7PLD$o|wwl6T4*+jKz{ZDqA6eimS9j zD@sJrP=OfI@en5F3X#;HNiQY;A(M~QWRR7TBV}ByfW&X*P_BM9$aBA#srz3MD7OjV zX5ic$yTEnlPaH-OHtJ}Qa+0m?@?bxugPWX_Ga&y)uG1;Gj#TH#?vUeO=Unw|u0~G(=QtEGS#wNXq_C6czDc zCc4z~nd+N!nJWAAn-W&*V(KI9sUvRJ_$m9BB9YR)t2$GkrPiI8TN7yD@|+b?u^@wP}XSv7zAYI@H2{zp2eKp_C;KB`?&#UP*#l70}~(O2l%W1JeDAc%mwU z=8PEU&lbK*#l>wx?U#YK`a(NWePV~}Zib=^F(+*DLKGHA+#y{mU-*Yxo+Ev$awM~C zGr6=m8Ce8LfJmo0IdvzGYUWr;yPa;6t@SBL#P1jNe1{eldE62r9>|h1zjQJ^SDN?Z z;ySDw(8zn$^og`7j-x$A8<`?GSWU)vX1TdUzxblD>h} zv_2t6`y9U9f+6Ow*(C4pjt+8nybmo@+en4J+=0XD)zI*v%g9nM8%!-(v?2ciIu~qD zzDzn0trfzcVtN->qd@d+rX8HBE+vY3$M7b49v1l3i8uW|j+f+hlLfucsJBK%=t$LE z{8B9!o`(D)dHWaRYp2J^c8fCXz0?$i`kg0g$FHOQLwlIev;$yxNB}22mcbPb6Z{O( zEb6{h6nd1u6`%jQnf%piK*HR9=}$->@`;zkk?+@%<^Ke6?Pv%2{bdHun0^;Umv@kT zJZ`4B={s4wRubMVZb#`CVTZRbkHZy#Ibf*Q&!o35$1Wc#sA7x5jIR~Kk39!T?dLt{Rbwp7RhoGfh3iU zQQVLJQ0lY08M;Zz?nJ<;rgIOeKuO#sp%~ok`d{ z_cfJyiQ`9bZfKVVAvD#ggQ}Qaiq_gn;I@}FOm43)=V^V27XNC(3sTL|_tSak*~4D* z$47a1&9@2^c>Xa}`#cq=B{q?adG}DzfwNT!3I3E|*#<^v*)7CqWut@QkFnsFU<}<0 z(aDs2s@x<3(Y7M^_A+~BID8nJ&dx^Tlp%T*GZk&B&ETKdSb+L8WQlW895R#9!1rH> z<1U*8MB=zO%xa4S9WJdG`KOQQS(xxdURN{nw>rsRf6m2No`C-@=%=o#J!fV<^atY- zOx`)~W^xwVldLIIiIKn^d}3E5nD1?bYaTMVlG}mq-#$s5afrb>&x~mKy>oG$?LsbJ zd>ZH8G{)B}R?_N8i}4$?e7t9;EOqurB987`P8J0{Mh!N`_+{8iJV$dMF(3TSNIII( zJ}T4bH_}q5-uf$k^vaRGI;9r2eY%bX&kqy1pZ7uM<68XBkf6Bv{c!Zp2pXJ{k1OR} zF#pCq)Fruv7Sp`WYYk07GY%%<&wbnQztpeDP(+c9AB+aqt&Sj7d4x_1vE*MobA`9w zz5;E#Bu+(WtS&p7t6_5AdP+;@!*2AYMZ?Vj~bg)oa3To%Gz|`9s2kP0-J=Q1iqP^Vx zMHkbLGRm2mD{`UMj$=~Q7{jJfZMuNDOG%XVlC_=##7O=FQdx5oj(9A=_lFgj!FzMa zz*}ux_)(YLx>T07Zre(0?up0c7p|bgPP)XQ;|RocXwxqj1k>88`N-WQ9B;k719o3r z498?QVVMjbrOrPK7uU_DeFrq?@ODjNvvw8n-0&4GY2FK8AM-#_UlTNwd+>DOmw1HZ z-|B68gCBPuMGNZVI1jQ9E+3goI}SV1k1Ue0-z9>6wJ6Z-ivx)=7RO$UK0VfOg#1lz z!okOWF!IYJvB%g$^!1%3T|56D-z!3fW`6yJJ%dN+EiX6YLzyA4c+L~jY@317Jks!y zKrwv7Z5hp`R#Qg3Rme_D9y8YZ(B!<6PRwEOg_?M}?w&Y1vRM?CJ_jyKp9G-+FR%+b zhE#ra0ljq%Ds-*D{U(BoH~RMDkIG%>gjy)Aw@8PM4qixKyZ#(IyWc0KOTHk(G6xvh zQj0?j6p8zP`s~|1!Su;FS8(U~0<^6E5UE-rO$%tM;O&Mtu>L`7?q0PPU(3w~jc#pP zn7gh8W^KW<9x7nPqUF%^Xe~Ty7NrgIbl6spxpWfON3>YR@w)O9=zb{&n%SaDk2_6d6eUNuo32 z(e@@k@;Q4x4zaI*!~8-Lv*;=>$3~9!n(u{|^@@Oui4wN&noY;6$k3jhW$lWX+O_Tu=lS)Zm$uKNyM)5&#v`o|`K}GFjaY$blr$YTr3-HykY*iC zg<;7pJKUi113Rdck+g|9C~}$#vq;__+wIXrrxw^D*AOeZB6~L+4Az9)8?)(w#2Iw` z;4P|iXBXDEH;C>DNYYJ*xjg#`Jvw}80Iqeh0XOOWWa!~FsCvS^Z9+u?SDD@WXOa5AQqpT(L*5DW5WB$NWWlaAacd=5hG|F*vNvc`kC(NPuh@)Nnb*6wD_vErKM$R+q5L zF_cD+knK8|L|(?38d$N9IdU$9WK~pfY`%}gX804&-CCJR-twI)l?x-Rq6E1TcNHFV zFF-Gnc8Pm7g8nhJ^A)u3`?EDbq@Au*|YxRUp(uJfO>X2($&(!&=!fuZd z>ZtuR#yLL#BQE4OTjzOaR$^UtX5JsX+$)z^vGWE9zV#S|=OIWiQs2eEhC zn9+i@WVl_2{D>6>^>G`9C%lO?zxd2py&ohRy|QRH@*J_*mBS?d@FwF8-Q(@Dg)u2?x`LNrDNJoAX`I;^a zEc1$-b7~>^*_xO)?uD3nT<%^? z9Nhm%Qsc4`Z118;2*kR-5&J{-MA3H-H6EM;_w#nb z#&Zo&v@x1Ump+A!JKm5(LUQo!$_8qcwlwLoeL<9`8^heZRUErHp84}^F@*2aAfr;B zxjKgeboVJhP=gEUh|MDZIm#1NZ%r^HGulVZ(1_T>ShJ zqxfbAI;A;@CKi1}5p#c|j0H)!?r#8oK+3@V^zo{M2p@+NfqZPMxJe8!xS9G7eiIiR=Izdanu1OpeEy{?<4wIt~g$ ze{j9R*_^9Z7TyQFCz=w$NM`jo$|{PjB@nJx))v&Yf&#iDrbdyZlM zWjed=sWXjr1n8xX!t_x3Rdl)}7az^PkMp{(qt6bd$aSY1?kh_|+NZqmHC7X+t0WW4 z2lI*621B~IZ4r5QYzUi7sN&<>RPj6sSKO6)9)Bt;#DBvLk)_U0>h6>zJm10wADhjR z)L;+X)prpaG`zs(1uWEd#FNF#6LH}0I(*8+2Z^gqC!e%DP=%xl&N`on1((~?b7nv1 zxS0!ZU;84+HIG6n5-hH#AEKfdmii~EOyi5K*aJJEc~iflz5!D#u+0`LW{T6T55#f$ ze6Fj=-$ty(&x7RQEzmOC7KR7kBLV3(&>XyrT71@lA8sd2leVj_wB*Q zPJbmeeKeN06sD{8UB+>aZrG(m9}Z=;iet$tf(H3JD-Zt9?@5E z<-vTcx_2Wkt+7PfDkfFCRp(>lDfT3xD%$jGHJTDcNzZg&RT5r<2df*{Kv;+e`3cgv$1FR73^s* zhpX-i;jw9Ucx_7qdbzq16|I_yE#|hMJKeeX?~F4r5FiEbN3zk>>$dpGFJUHlZV@U; z&%o=~pJ8lPu-NSPPPlA(2uoD32o`5Bhlgk3zQ!1&A>T~?n=MZ4g2dS#dIgeu)K6^1 z7Em7oHSw>m0K8mv00(~hfxS2OP*xki!Yc`&@s9qd_^-3<|!6kfh z=4&KXb|#kkrTq}5*9vlMVrMEze=&F6xM4ZoDg3Q}9$so{h?g0JG5!OBcy`zqX)yOZO?+bw~#qs~>|zyB$t+OhcDXYq7daC){px z$5nB!$oJa4uxjWx_y_l+^c{r!>OVog3v?2r#wc|Dh&x-o@h0(Y$j5&!N#P9@qvX)b zNv7gR5><4gn)h9$22Ixr<@UiTc>I?c{CCa@_I~SwS&OAX>Yoe@DtQhKi}+a@*kq0c#FIi>_PJ0hhdSO z2fKZ26&rQ%Ea!K44vY7s!12yl%yUiTt0 znt6MZ?!ZUwNU}_~3l2mzkwpvgE7CJ;V9Ewj@S1f2nryFPzZ_sNrCcTNs=jN1wCB9YBbLb9@l&!D>^C#Mz=9 zMUY`{!TxvLk&RMLhU~r+=nKq7$M>C~MrP>a6~2M^ho2ufF$dt?ULoc}e;DMft08d% zoClu1L5_EcbPDC#j!{CO+sa*Ec@`;JCKPJg}rZHBVL&SiJBvV9){V#Y&VXtQS=16JU3xG|LMW| z;C_JXlX^J9ZS5HLnxY;X?Y*CMH49)wT7~h!d!6LX zPCjL})0Vlua0YsvngBK&yDe^e5eTKlp~6{dq6Ms99#OKTh$E^8eO@I?x z?&F9yU&tr&8mEzX_IsuzOaz-74Wj8Ig)oaOU`uc7v5~J#*ef>E;l{*j7zeP0GK4kFuCRS&aMh*H{a`HS5APdB zV2UlD9AC5wC;k+K)kR6Xj_dk(S&AFBQ%Z!L?Zd?TcMBu>_s5QKIt zL_z)!D5HH>QIpgn;xw6rcL$$?O{esbbb|rAsSEIM}^6_v^wioC&!!=}UtY&C z9^}c*7)g9^y%T)Wu7>aqKX^6P3}!7)$d6@r;ed1iM60f(OE2`n;5UD;QN78`y0H{? z?+`{_+Uvmf*dDkR!6#3oqoC?%GK5>rV6q=hhri0t@cq$esI$uvUL4LxQM+ZK?eSMq zNex4HIma)3TFDQ07G(4miIa|rqgZ9tE136wF}pZMm|btMiS>D!!iHZCfK%TqS?>r@ z&TX}`s!VkaHh^RZnvf$;oGwu_B;v?;ya!DDm4z2~?hzTu6yko&8z260g{WVQ!|M_? z*e8q&G_N#8X@15~YaNJfzFR~1&NckE((>fw{GTx6+%mX+r-?Oiea`wdd9xCYd$5v6 z396UZ!i8Kf$>zjQ^skJ-2?J|fyW=K~*qz0R#-d>Xa~?)Z17VfoFW5XN%IX4a8{0$b=G{bXZcoS*B^g$=>jJy1 zyB~sIsj_3bGucfVyV)y`g+MCLoS5#GL6(96VD`n7{cziu*K;I}IUOsBA0r*YQFP(( zx*}{M5srtewb@%EXW13I=qdNMaq#FY> zV+P6Ajj~{g;-PSD4*P<4id}wuEo+N%p(cD9`!*~dUHerC?kVPAd36I5bLBL6HV?z` zn)hJ7!xS1f>O-;4D>!W52S>Y|V7O`>JI&XfG_JRWC2wWOqrN6`HOHNJb+p0VX*{^x z_zvxFnFC7&2*|c|P(uIKfZu^wCeeUJ_Q?nF^}RBzvfD)%Tv zvYWSckfB}+i25}HzYPo`-)~95(f4YNj(0H9K}NHTSecW)@chnoe0Enq6Lp`Docc%6=mjafe{VbWP`L|ce%J8TVn>qW~U8*nn2W zJoc)dB`ESEA*;g-ZrOQ2=HD=M@Kz}2UHd?GuelE2_pArUF;6z9P% z$C!+gUq%q=%+ghh$7Z@fVzg9z^eFWPhR*b9X zC9ze;nY@3k47g?WLHFvJkbQ9nib_mEN8lmq+_#lK7LiSED=95%Raatj(;7*rF$kUtWIq>;FQks>Qwj<=R>g59tRS!KoKOy z&Db8^AF$v3o3(71LMJ^jWqU+c;g`%eRtS_mAqrkUPf7LY`!J(KPb%p z?&*LNRc)gs$2bwcz?MSSLRENgT23VU-;J?k#- zz{a|{vR7R)Sp5RA>Uoh2E29?8rtWoSKQ}Z%qM;AVs)n<<4{gBp^)^uLS7B!kzk@)j z1UAV^h^1q`!Jkk?cGzewJLg#*yV~_AJ4I5LZTu?C){AJd+T{`~@6!UX`CJAS9ht19 zumO9(oME?khp~D4T3Bzkl09?87i?aogWoz0_Edc`sBwLqgFo&;?g2%1Ep}r|R&yT6 z=}l~cfD-$q8M7bqL)j}UOxcD{f8k#57Z98$$r{gD2X~Pw>$f@~UQcz4t%wym!vu$B(M|Ds}hFuCQUI$N2Gl?KDqinRBSIbukac zzgUbhy{4=3O1~BF1y@Qb+3g~q|KhrG@mrJBNe-KoVsqN^`gM;f6KWSw`h-O(rJ{^V z`>&ab$H6u7$biBr+iT!;4qG2k9>(`mdSnk)R#m;KOp%MF{(O8}N$dPT-jFm|N#2j2 z0ZRC!99eNiai4Te*>Ld=53N2%`P8|S;#qN>@~*v&;+KYx_3ds-*R2JV9S_#XuLEZ) z7lw>h8W&iuuQr*Mt=m-%U-F8+oou(y+69#qH+GjAA|Uv(riZ+T+bi zt<|TwpynzntI~Bh|-j0_A``TFHb7(g0?GnhHO#J zivyHtKdUO4Cu%4Y0y21RBNydMavI;9>!GwB+gIt39HB%QYbpAzpZLVVI_0C#QNFq2 zlYD9L9iA~Si)ZwhqI6s_P+8xwv+|&LE+1R}6!-RRsFcjl=jAI_QS?)@c((g`<;Iw6 z+_BJ7rDkZXvb?I3^45R5d$}#m_>Gw5Jf>3_6;X0!5W zzCqcWa#%_5U`oe1Njx@hFE3Mc5-&3&mVa7*l8<&*`qp2ado&zFy9OOV#JfAY-k z$xz_sHa>F44F012AL9{^czH*e$Q7r{9Z|L$w0o|zO>T! zQwznTZ6RLpc12}Yol7ajyU&I4zJlAJmG|XYN8ZaH_NddCKzd(R|~_vP#i0cer=|EIusa8s4`in)QCWPOiAqs2hz?4O&g#XJLBp0`(a`t1Np`KIn|{8t#IZI1G{mz|m0)d$9v_f{BxJJjQg zZBNVHx5w~y^HpP~Paa^b;HWF=kZ$ZhJBmLz_==tOiR2A*1$isCrhMPwYAmNSUPH7* zlyT;ik?^E^b*1ysg5XszTGui)Pxty{9|n!=c+S~u`CEP^`Q?H`DU~O+l|Lk1G`j8_ zt?M=4P3~8*B-^`oiR^I09uCgfXZ$m2F1Oq2&vQJ=@l75M@NLC#5SFcCD@{qvU{{pC zx@Qa9j~-*Lv4>gO*!mEyuPpyqyhnE{ttpQk^N4l8*NOTUIK_sgwl%WMMlinnB{u&~ zGE=^lhp9hu*`QXT%E4oP%7$}gm2LNb^0O@u@Q$WXUZh%IV}1W%_Nc0fZM56xUh6}S z{4T1o{FZ%ZxzD!A^B|YGUGRoZ`2YO4Po1EXV<-!{(26w=oC*FZ7ubw9C*`Y+Uc&HJ zyZG1uyw~XQbh&U&4`@9#l!@=F!Pex7Y)Z9@4d2GIhCQmY;Pa76`aE}K#r{Zc zS8a*==C&nywYj_5{<~Fn-PpG2m5yEPaKTu# zYlNLxh+urPKUlXubF|#MP=eAZq$$_!Kf*s7^t#z!TC(7gby%zLCVU!HdkuC~ zo(|jgErLBmyUUkOWkZSJ2Q1HiknHMJ7hlJG3^v?}Z9P_rNA=x+-v_(R|I`jknbdFt z6Lv?!u)BWjW4oqo!%GonR(k;UeO<)lE6d?lBtWIP%eh1FB6-luF=Dgn?)>ER_Tq*3 zUU2-{Trs$Jw$9z{5zOuI5#BG_FT3SeWLpjofTD^0;Hu*i*0NG%?mc5W%skxzoH9y? z#g})47Uz5+{ApEaoI4((I~3-xvvad(?F(H|;F}ulkyF^rCb!gBGH(omq{BDvwX#cauzFbK+xVWJ_3PpkhPeh=9W z4~A9|4VTwKKzIDkraxZ$D?i3qs&-Wf&t1W9+&T!q@imVHeqpd+#czP0V?{&G6INqo z0^|0Jz`5&gdEAV$kiG7ld(iEL@;{qSGFuNkA6pC+XFs1Wwzv}`E^T(e80S|HBEEMQ zGoA!Pb~U5U$>=5iUUme^=ZWrBjjwghI+o{Uw^S3WFn)I9 zeIqC%w`E6qYysVq;$q;@c)UkkEv6J+1by1B<|F6)g!&)qit?FKVm*}4@4h3nf&|xZa7!-33F|p z0xu6ffg+b(#UQ+r>1J7$kNws`fq?n&2|u6N%ITmS*=wb7Up8JZ`!j#_)!WUU|t#c`*bE-h-pThiA{a37azbHss(hTa?og$uUA0S2~6o4zA9JzgYduaP( zIm9IG;omB(hmslf#as2tqgs)!s9!(IBPq=UFf}fLkJ*zu@5 zO{$1r(+2WaANR96_xsdX>Tpj3~ z+f?!Xa~F2R?vo4dX{Z!+t;*dxf8qPSx$+$eS9Ot1+rYTGTbREgSI%m&juoxa4o>8s zfuOA~bmL6hASn>P)7*X-+%(>Wb@~0_V#9piYwARpowbF3DR|qsws&7%E$F#y_n@-8 z_I4c@=yXwjxa={z)#U*9POUF~+UzR!Z14zJ$UnN;W5f8dQQvfOR4lLlPnevXJ5=_M z|Hk5rI`E{eUVQ(be^_3RqtGaDh1l=ye)&Zzz89PRQr`aKDSmdM34hXaHE;O420QN> z2nXkf!j4E!z8UXd-*G_{kJsZpjn^%~*9NZ}TaL?xSAG+L|Ea+I@&2((a>C)_o_F%H zS|6eN_0v$bi#N=eW($JdD)w^rNj~`a9oBzHb#by?HF3niE^7%%U;l-azu(ZrI_;Q`DZD-5Jv955)`=Y$QMg*VX`&E88##t#+JVHKw;U+xs zN`iRcyHkaY-FFLO`yD$W za*Bi4ZOseg%y;+T6_Gu(|MH+k}JkG8R8 z^EaoIzz>rgXfzH^Z4s5jJx|Cr-UYF6{2uVlkm8DG&Q#^Y_(kxcrw^Du9{}4^&2>fM zi{N#5yYZp->#`14wz8N>yMR|6#5%s%bGyOJww9pT9XHpH;|x6ATD z%Zq{w6jpo-^pn3%#rsF+E`+>ngF76sg=fyy`I^z0^7bxSeDHKTb|tT#Jod_UcC}9( zXgMc@3y)3mR^v!uE&@N&I8L5-uZnJ1y9#h{&2%}W-(I#fu?jmma6P2@A7i(g?UB1x zuEQqUm4*6;@kJr80kF#bt^9D>XFeet@9mmpYaDT@AzV#f#s=Nq#5*jWs*9B>^HD{X zvYMx}%U&H@Y;5$fBQ%I=*QFI-)7lBN1~5rggoDy`lufr`o>emf$Q|9QEO z{n%9u@czx>(Bmul^-Jqv*7$F7U%AwR7O?X{XI+Gs4WC%`NlKKhjrc72Ejzg2q}blh+-UMH_+__V9%Pdw%e9NL3S&0uj<39>8{lli`{noNlfM4p z)2r>|BQDUa`+{=~#j?2Bg4>$imB z`x?yuQ61*FoFNDZ6U{9&oFLu^icMVoUFKi z>jc}pJIOOPellLXKOJ@~8w+I`%;)!wjp5zLM&iq9IWVf<6M4)>8LBwVgBuM*x$cO2 zy7pT(!i$L{I3(&!-LI?ThZ_%kjP6xm{QC(8Edm z%dQn{bp2JlS!fCH+b|eEmmCD`#)Rk+wyls4^@(9a+IHhLd1V-M*8>JMss?oydBBZ9 zO~B+(ouB>Ph*zKcQBL@Ll>eyU#~*%N$|eo4ktY{&U}Y1V!QE~yaP;YDKJs`XAGKhf zu42sw%xNX#hiW9UYPA>e;;yTBo15d6_G6-$=Z@vfuho6l%ux?Kq&4)6tHaxzTh3bv zco*BO-n?&@KKPw(KlU>gzk^x3E7mKn97u1#a~Au^-nPlSP_I;@Z_APV^P;Kp&!Ttu z_Zu#{CS5%E>fP_~8k%(=asM{Cnx79(TN*B#CLES!e0}<6z%KqaYclL!;4Uw79-)i7 z7Y#S&yTFElV|euKisHAVldOwcBl(cjpf#{Ne+i6otH`AXs(aj9h)q zes<#b2D#Glqwq3e7xYPQ!rqtM4%ds`XJy@+DuqkB;`NW~LcgNLRzra0ww

XBl<;I{Ep%{uTV{4uks_lC`%#i4j-H)D9Q zJYy5QHr4%~_*s74K-Om4Q<&zyiI+O`T25X2Ic50vR&uod1wZtz8XOGg(4yxeIJF}U zwx%4{+4hO$ag*-5=hrOHU1!eJ#e6LwJ{wzEc~#{kKXIuWpRlzvukYE1zdOH=XB%X> z`uP=#O=^~WDe5u*+}BxbmRnXXlQBqM_s=H$oX%z4Q@kHvR(uuZ)Tk41W5^EouW`j$ zQq^%>En0-7-29~5Y8VUeoOiGd752(c@!C3vaE)*N7PG#_~5IdC;yf^xI)y2|tGiQ|hVLaAeA zjot3;;PZ>8LB8DyvEi6Tc;B*4;p2m6#Bfi z;|Fc0^3j3y`Q=S%U>CZbw_o3rznwN-e3%|Aruuaj8w-x&o0rSPIrHX<8LRPcl|z-p z$Y%XOI`9?R)jliV=;13DNhlzG!`qi$4qd{QHD}_}*ixeViEg}huTn6;=b(JO|670O*r!~gq;}xuxjHk>Mzh$`NlRl?9%DHUdiIJSlyLwE_mJ2Kp0x$J%4%r4WBRKH8>rQ z>ZZ7NU`hT7P_|A__*mhMu}{+yiv6%aSn#0-M5PynqpZ9*s!??e%*tG$II(>&e4e}nrdIG4PHS4?EIF*wr^|7r=7V7o9|TT9#z-zz5i?mx4|)DzR7@}FN}nv z%l-N0(gMGB-CnFUxB=Kj^tjl6d9A2%g`y z00fqr4_E7L(XG{m%f6?2iv{Nm7vH{&PD$(U3Pa!H=OGK_vDa%p@cJKj!ktSw zw8~P%?+bdd^P`=Z!_-)~9Z;R$e(5E4ZMj!hqRKERanwdUbiEL4iu(lTXLb~u{(Qo# zMipRbQv$%dPz2b#t02Z1_Q6@VJ>Xtxytt^7zqrb?jF|f%0j~I%*!e;-J6Gotw8&cp zy>2&VUFMb+mo@ID^ei(TbRMg~Y4H{aI2q5%99_T$;`M$roU>R-*R4=9Z7q}&u7dxD zaOhvbL2OVUP4+)tT^xU?n>g!iBN4CqAU?orkX6mek*h{OhC1gyv%JlZU|vZt(eKbn zh>tKpp@y#FmPf8)Yi}pfF<}ivef5Q}@Rki5E5N%Z^~6#)Rp`CR32r#n0ozwMS&4%O zpT`9zgWKtf z&mRP#w!6WjNK-y<*f{VR=O@lIhKM<-1I2>v_k-hVJw&W64jcRng2#d-`24(7*Q;O? z$X=TbMP}9%4a?TC4YiB0xVk?;x;7EMy@``Ymyd_wCibvS&eQo0i-k|#2Vr~Po-qD% zI<#+%*WT}z?;aP>N<4Sn3CiWq<%OpdS4R9J!_A2vd|Fjk@s@2R@rARK__5q(am1S6 zEcSFbOqjcr?HJcv{CTB@m>uORc1UcF7xs=7rx{L*jw>Fqe%piC$*x1-^tzhj3ztj? zSrH{>My!N?zO;clP4i%3+sfd(v@pctH|5J-TFTCYEjYbjpsQZIFx<}@0)^Y1V*@i? zVAS40^4`=z@DBgh?N$Am-0HOr3^}wLYW~cGF+H5cr}=ttaJQ-!& zVbF(Vu+uXehW+Cpmf1a0JpUy?+*4w>D1M0*Bd;Zh;Z6TA_mskNd<|I^ax++~YHdMx zzzFlBW1w#S0O%N57b^I*7CY^#0Oj=de013oU|XlQIN^0gQ91Gtyuf?8OpSOCtqs{M zu=IFV;^|1;(~vOnQmbnadM6e3hrWX;Hf7|WHD2>fLl*eW=?CYN{9%1_Nfs7#f!jB} zvk!ya#7<9Ih<1CJSnOT{2prHz=a)MnrQA4YxGD|6^8l|c-8+lfI&ERL-^a3c`7{&xY7PW7BKSf#8=*!M$_>)NUw>F&kGi zheH8yzCT{$|8fTy^<^q->ud);QyUu>tY{{Bjo%ED9%sRmo$H|e{!o~;eKX|GE(0x6 ztHSlLVsKVhMy`C~Fta^V8qW816FY8u4)1UJ$v=&H*p<8w4mv!QBeINaa&$YRM}B~@ zX8%$w^kfuF4*0=b9~DpOS=GgjuSGkkt`ge~nRvHa2fp!lb1qIlg& z?0tJBC{v3;x3k6I#8y9O)P29vIp_@J$8Lh>XQr}I`2Sng&zo3ket_7$a6PffqSj&` zSrQMt=_EehGDEcWUm`v)RZdKbsVNp6`&xFzzX`S+Y{q^*^MEyLbdcD*21JkQ4$IH} zU`6JX6kjfS$Z8(TfD#SM=$iC)f$eRac~oXJOe|A?olVqPJVq-I&s-QaFi24ti+m*cx0hs&`q!6?lzb_xrD0;$iS z?XI_Mc)%wXgs<`Hdo2R^*a1#O9{|ConAqc0kZ39%ES6r^S=`y>J!=Z?&~joic(i#L zd>XPH);}*P4|$X>H+C!rwcBL!HqZ4iUw53nd@xqr-KVoSXx%i3w{I%Oy|)$D8+J3O zRz~b1*o(Wj6coqh{DIZ>D`DZLvGCgahi*vse70mx9?M^IU5<*H0AsgahxxTL|>omZ2I)Q?Dv3q{NcPtte(#m)^xxoVEszMo!AP;&Uh}< z<&Q6;1LrZ2{uDo_QZNLn+4p4q_upV!bW85G+vSn>hCv-e>EoG$bfN;NpdD!3k& zyY|VF4=ilJ=3d$)4+}ENvub$4#Q1h%rBS)8=W$0U|2cwPsa6x(H*d!h^MA6DwQs?J z3H4#*o%JAH=mghmc!T}2WOi^*We9V81V5koz}Bf-@$=F#5T2OKnr-d}i9WV)Wa3g@ z;h#3*li01SaIZSB;&22+L>$)bOP$9ay1ZbEI>*DR`^(@}W-a*mbPg+VZXxr$T0mS~ zvY;qjD=2nv{sS(Q!1pTl@8Grzioi7QK`gS5UJgFG5bW%V^I*vyTF-WYSug9Pd~iDp z<9>KBzb%z@|5RQCy4E&O;d+UbW5b8CS<};@!;fl^_azVN+4X?Ii_U_5iLX$=zckN3 zKEwE@4gUX^SX`Xh41d?&C5Xr7orOt-#;{Arm&uEx>x+NBosf%MSqGb1|D$VKZj{dV zdsX z0*5vQ#MXFS5udU)Vt%ZTINwEpWcQA+X?8jk@#+Y#%8h|SYa-pBgqMZr;a5_g;QMVg zPMTP%z5u*S%4DCHlmO4^_o45?5@H_SyY@sQFR1cq1b8bC;KACD(6*b(nrAj=-zK() zxKg9opk5(7aL#>Q@0ER-TdFryPg}1$H78lVl^6iC*Js0_Zqs3I)uG~x!bvc6a1K-F z_!xhs9fr{d@ZR*<8=-yoq0n}5CCGm^7)si@z^W$?lyB@oDzqkqV!}0$ox06ulrvR;+Oo8QVYrv+LwU{of9o(9h38uX3vQxM- zJU;pYB8D2+KQ(JZNYf7RHnF^zpV5tt`?3=jPv{SAW(D$e+jL$g)|EHS$RKrt@broE%NlMqxS9r#lcf646I(R!M zmQNn`n0pKyq-6FjA@=ou#lN?X=Y@ql7{0TE@?&Tt@x-VsNPFO*%&^-GrCt=0-7oFs zfj-ON#onTPz=!tiXM8oqv&%4;a_A|GymL{GO@1l2aqiB0_ln2w16+p{xr4y_n7jCR z)L?Kc9ide0)>7B>axZ+{VFSyq+LRq#5C%WH-{ghwj_3W3JW}4<+NOql$WT&#RWp`3 z94{ZK_>qsfoxzKDDhqkl4SH9|b9mi-86Vuqi$^?Y4#OI*=fzgM zZUh%6zu;9B9sFD#r0W*gQn}l+Hux=g1Y_++DQ;eEm7D3={PwB+ zO16iK0`9MjVFjxyht{lSxf37qgkklR?``lNM%R1s^Mj)y;rnu))-zdoJbW5Fua}A6 z_1dF6N$bwKmaeR{ThdlJi{HC^vVS`h52Wx@hn6XRjXNqMGK@xE^c5d+EtWZ#9?N5E z|B+16gs4h=9f=&<*Tc8DhkR)=M?d!gibT9vhOq(PuoM0S>E7Pe3aa9U?Lx# z&3WH-e2ZL74P*xrZ~;|+vxs50z~@#5$Tirvp(KJfhz_||(M7fx3a=lx0Mzgaxbu&>Hv?Os9K<kl!1nL~9yDe?q$GUR@s5n!;%8wP(%c|5k|O=d09Rc+Iljhr=sPdsr?=biXDWpK+l781Uecm0gNn=%Q z@jIP%PHgtIZSs)@74RC8b}(S(SlQt8RxUgt-&kbi4xT%rJ}6;0!y99b?}f1+lxA1~TI=4=A;yznH4` z73XHOhDxPsu@x1*!|dqmMuT%7czJd+3#yXB`wcyq;@hb@YwR{&ZZsg8w+5|*G zBPjyrwpt<=U3Y+w?>hi`El_!I?Tflsw<^QWRtkO(uQ-3=o|zJnGK+@{eU!5A$1u4` z{rUW2d>Hp>SW4IDXAD&GIUo;@yT__X?qc4~B(^Wfzy?GPUH;&EjPcUz<=iH6 zDc`5|(hc0OljnC2;yXIe;XW_E8h5Cpc)OSvI(=|;W1)^+d34+d_gUQ{#mv*8;?&?B zP-o|BX!)mz-)-TvJjc|u+stD91tFTa`y z2Sym!sNM@?xX=#zi^tfC{1dF-{HEgA-gEi$$RlE*mwQ>WdRKK>8xwh_lymZE&!>FT z-pbtda93!!bX`j72K>y_kC(>5_D;~XkBO;8ZTaTl5Z=7<`;_2aFmzV8bli!Ll zfWz{q;BtSb?3#E+9yV|lyK~|;w|iAx?4R5pZf2I}v$N|%m+z|~>Q`a0(;P>h*Sa0^ zs?mi7Zk(_4nE1`uA$Kg>+T2HMvg?@K|6v&GHP%U~qi*HfW?W|NyUgMZ8&y=w?$72E zy1DVY4?SV^yIJsO_BAfHSjVpf)KWIy%;bjy+bKi+izyTQn(%VcK^`w|0NeWal>IA1 zl@4t@m6f&b@{O-2$x`E8eAeqa@Z@9&j7eLBzq=T@QgtLWkABaS*M%4_1U{6{AN|0; zmtV%O9{RzHxt!x6r7J3*m2?~Ye0wEe|Xb;mFIk|rMxTc1ijy8iJ>upI@=ZTJm#ef-+Z?- z&+Kc%ye) z9lKLm8@X87(L=8s9KBo-N7PeX_)Ok+=2hNcusz=xRZdCiQCgY&t%5RnOd9-hd(Yeg z>WRf(cycy+fH<(iZulH0D245>^3w~-Dlye=!1F~u>{qg=+&`Vlo3HMqq_v6Wk4HPe z=*P|ZR`02DzHyV1Beqdy&s@bfu{C^2Zd)b4a!Vz@P%$O{b1`MB=Uo2CFV|hVIDmKj zb%Oi8sUYu&?#8En-YGlmt(5xCu7Bza=hCUt)!ItdjqA#>H+AtoAn9_`YQ^~XQ@(f~ z!hO8b_g>t!q{2%+(R07=BF}1>Y@Fw;Q@kBoDTgv_m2>BV`NV?ueC4hpyaGJnemAzV z8BJ{A`ZZhS_oyH8!^_|By}vGs%Y*F-7@H}pTR!FwlYEq{?+@@hUDFhs$1X~t_Ei<9 zgPoL&TXU7X70>wgvisbB+z;g@C6Z^I)F~a(U3f;ROy28ojIu3Er!@Q0M#+4V%S%^C z!fSsA%6X0E@ywEq#8c}cL}l)2E1X zPxGn>Vx@wz*v81yO@VxBpQCcj?rzGCErL?vOk2J$@2DI%7q2J1@|L`C|9E~cU^X9r zE)K6BK93tZ+bEqE?3b6>cSub?(H7s9she6Pe67;u#~$VSBp)THT?1aDh%5&j2;sIp zsz6wk^3W&#G8=qtJMUz>UT(6yIDX}*rc$H+Hh!nA7aw+SE_+mNf&4M?4F6z??%_|uM)3!M$%@_TR?1hK9?IpN;}pAKHzmV%s8VC|abyAn5f3ojJBmM^{gT`uwHk^AVBL3~Il4>|YcF8*|elkD-T0LP1k%fS=y z{(!Ul8>>cUyML(crMuU1H7waxkFUKQ%oE!=q?$UPQ-0k4rXMH%_8tr-lymf^?x36O9ihBwR! z=04qu6!m;rvW$>!BM^ML*%lOpqD6)OYg8571^=l5W5>-LL>+I6e>#O$<}yONp~Sm6@)O9*@(#qJ=%KonNS)KDndd~tfQx()iX%{@{Y)> z2p^IdN47rATtkzn3&^XOqep47Na`Vxj`Ori&5toui+FpKESfAFdl1=qd8jzt@i%%Qpllv27m zk2n=Uy=ahaKxq4=n(I()6LKzCHwt0hZ|X&P6iVXdjnS2g=wFrGm1w012A^HlU#N*4qjLfa$ctouSS zRD)t%Fi(#lTQ8Yy38jh{PCkn+;E+^YFU8`}tN*T@VtqW#e##XP+B#9%Bd_(qbx=R@ zn-Dbg66z`{C}|EVN&zLURZx+`V@tY^g)5nE=D(w@ohK5eqHe8EIg`<*JlYfLM?I*% zHHJ8QtV8pXP}d`M&K{ZLIbx%e}s1s;C!^yVrT5aTLkWO=^eWO9~R6|nD zeW@1>vP}rRWZp-T1`8yVf@+2yCGC^8Z=_3RT}5dk9pQq#B$NiSF_!k7=4em_^FV3U zsUFRnhPIdKW}SYTRC7M<7Y$Xy9|+n9iFA}75C+mIM#Xr9Cb6G&p2X=T+2`L#UG^s0f;O6y+eK^ncG)Nkto>olDdUV-0Al2+Gx4^eE4NA2TN8DaVA+ z_62R4wNN@llL*?*z$QM>{VBa}S!MiDIyZihdJ< z`qSJD$ZK|*Ige_o2;wze`k%FlqdC&}?;5B-)gw+XSBymv_KIp45GsNONv$TWfnrp1Pm0qcObGfNOOHG~{;7sm2W3&@sTU2Jk1CS!DAt4!RC8TD>a;&9O6o_)XE??? zA}C%(siKt7Z#`!y|2z3Hp7P0Wus}koM<1ON*1iJ95KntboPlfv^nr$WsqJf!08_ioiiVDUbRYETEcXYsVdREk{5Oby`39 zsJ0#@=>!+7p(m(lqof*?XFzF>wji0~1+>XWgM0*P@_g^6%cGZJ>Fg)FUMH z5v(z^29Q*2}^CKVixAH35+WwIbYiThg z+gfTMZIaL?s1}Jj#aeUG(&|xNEzTmTmvv7CLP_|mqfa1s znmIv5i9HM`31k~2GpDT$b&A*Kq2f9av_9+5qb*5hkdHvN2|+oelN3-INGIUD462|F z^3(BmL6`|ky!vmhHh0NV8+oclbyS1_;Ye#CTQ%<=)s_&}zM77Hk^%t-<&kYbSl3DM zCgKq)!g>s~^%Ad|Ym(nOCxLi`RtNXRgfXnX$}PBwR);WM?vw#OR6OnCF$CbkV8la z!+)f8KEw+a5T~kWBS`8cGfXI{p5{Z{fVv(*Lp!%sR1m6!JkCY3NXq}aR+=LY>Op=j z26e3#`V7cf`x|hb#MAz2^F&=hKjE*9K9y|Z)&KBl(lAK>&P`R(kI>?fH&|`dwR-5X z4zmCDkWeQOhf+e&G0=`3=>*F#sc0g!^^!zT9_3i~#9-0&R!Mc+n)6(+u6BG?v#nx` zHc(QYwI1ba1Etmvr8Zxb!?A{{qK!Jm)6nK_=|i?8A*Z5(u+~NEZ*R&MB#gm%X%cM- zLAoREIq7-=`nA5~LLY%*OsYACbQ;7PR8$B!R0*MyW`4e)HB*j3`nRX`oT2qNBIxlK zg_w+>^UViAoP?l0^t?mEB;gO$++UA6)g-?}5@R(RRq98aHhSf!+5oP-Vp=>`PNlWeM~qD^%SXwx80lFVzTSS^ku0f%0~AF5fBPS2s%Iui1S zur~b-qRofmEIp_v{Vq#;K`~JX>tlvuh@<>4PxG2@qy2Z^1gwK|)T7^outy*DmJr%d z@uy_whFj`rwTZ*p6ibqZLDhsg9&2j0WI10@TH}4N4*fkMsOG((`dVMiA&%;skfV7? zX#eejIvqcn2SJZE`6*YD&}Y$A)G3F21_befWUgsIN%Np{g>>y4Lz{-GnsX?g@-;v5 zCbUUc5duO_QZ?5mo?so4YSAQfob~u?YsdANEW8#+u>@KR`UrxA{}9BR5L8e@hQkWfQOb?gz==NY}F?%(-h7}YgcAV^3cB!o5BTIUevs;c?1u17!RP#x-T zLR&>pj)XkLkzbI^_4TNe--ME0-!Y*sp-#M`r=Xn&!>K;<2IS6C9n`6|@E=JW*#r~9 z2XiT&_D2WJ?HxYFQWR#8VIANYZhn7(MEA ze2Etj1_>qh(xcSof+oe8$j14S)Z$dLZH<%Ao{V{d^luM}qvsBi#8bXi*KFiW2s-B4 zV^%fWDoPjR^eAaAG_0J2wj?1`%Um&zY#Q1gsg^!yYxyLxuPR|}^bu5q0c{#4$;?q7 zjZ-a>`l3%e&Lol7@=$6V#Uu2V7z6SGK|&CZJ+-x4>X4uEwIP{nSmPwLDOZcP`p_0m zBFILadeV79IAq}|&tUb_8Zpi>9aGRRp-wjSpc<2jLrEapK$7YZP&Xm8c$74E?bxHP z#R(RPHXRoUrJy2*kO-1FhhoXrBPdRL3{rjKEH(9#S-1A3T*^}slyeHZiqs_xf>x&OvcPns9ST6t^#rRkPAV~l{gw8u1P-83J)YF;DN zAP5r5G=x>6U-KbHK1WZpkLqjjC?%}pW2r^;3>arZNrPgnu^Oi$iLsO;N&or<ME`RO@MqfIdetBsP@q@s_YM>t}P zfRZ2)JW-WUB5y!jwel2)gZ^$3By)^OHTR^v;h3k*4eJvx5F|6Gzg80^&5xenRMqTr zL7n1E2yO4sFJKRAya9R1;v-$NRpb!XW2H&#OFi|HITtN+e-f%WPmnAe*4K3GO~;J# zB*~mh`DB|A2GprG*?I(V1Ruh({1l`34m^PfLG{ z*Pgq{hdkv=7)$Y59EQ@+IIB-J#|hHE@h+He%~Od-P`=g^CHb{=kq_r!9g~SiTl>G7 zNC`o;)c<*WXl^u!C#V=hwF$Hq>%6smq^PFW(-Ln$X|lv?^P@RP<{087@=+b+Nf$8Q z5v3j_)iF>k`N@a+WYj4}LTEXpOXfM7FvbU=^+%~ko@%QYLph{txzfKm$Y?{aS`>`Y z)`T|IH(8`3@}yI~Hc(Pul_Y|A`;%r#JVB3~fE@8u)4EsMJkVubr(lUCzX5dFX=&xAU~lWnaxAl>{tBuM5OyD*Lh#Suqw z);Uu>8dOJ5z*T6mtsu_>-uT0 z1r>oqJ8sgy$3jJ&d=#rlmyl6M@{QZ9QmxUK8DCoxd!y<30PA=DXIU? z2O0C@0sBlwI3iT^Q=FEAG8;MTbF;Mu&4FU*9M)oS4x2p%Al4H$^-wJmp%UBCjVO zaVQBiNU8`ItYck+9&G_ZQX+|SqInunr$IHX^QIW8NrPeqgkD7oq4~(Be6+PVoSVVI znUEt`_krd>LsB*IcRnV`e2&n5s;E;u?IB5R5Qm`pR8JdNi}Hx0dRi>n1_Z@QBynEk z(_+w85farUj&zioA59bL#OV<X)YL;}K=+Qnz>%(=csFQ6%zktv< z(s7WkSCK}LPeM>E@dk@@!8+RW4%#M*U!Z}yAmKmse@95D)7&ScPCV5RP&cUm=2QJ_ zzYSuaCvAs+PyBg>ytngW?Pbt-fl0{?TmW(H0OU)Kvu86h}JENpGo5yor3s8xVQ| z&QV(@^8X#wm;4l~#o*jYx*)Hr7>T?Aq3LK^_ft~QARb|j5md8IwpJHw8BnJ|d0KrH z!wJ+E`&f^GNiy?ik*7MuXQNL0MYf8dx>_C7sV@1cmfm9Pk&_VCK{mzFS|~=##kw>| zr#WgvGS?9-^{A$R(DJQ1&Xu?*OAK1te5fDQw)W8Sur~d_K|M4Z{q|_9D77_XEo+`C zA&)RwV#qI`O^^^|TjxnO4ei`Qo}SxCmsImPK|KiC-{=wsA}HU0ei~Fqn-l7)gg-Es zd@6!;Yd`W?LAA9xVJ`Uvv^5ST)zjvNxXH*v+kv8^`WGpm&|LX8d`mnG$#|{1#!53YhS(k@7y&GeUc{@p&BTo z5Y|3El#BBbFo*mq!hoQ8(Hv-y)Jw=AOcszXASh0cl0ZHF)`p{N-u=Xj4Azk#!i5lMp^QACk0w>P2uw z-8#r;kj(W>D6M${>IAJP?i0mmHuh0bHzDYKMpZTU7Esc>T@aK*$B530_UY!{62?=W z9^-~1Pp^MSsGATJYeJujx&dK5uTV1|3khQpf@-#-FrI7;$kDpB^9@7v$WQh}1*HL@ zc{Vd`JR= z0au!wnKSks$x^R88d;XcAi*G`ys|CXEM-}iVZZ}7JTPDpOB*&js|qzN4Qd#$VZ(xc zo)r;$?|Wt9rStCf_pYCa@5`*}uI@ew8i-u`dtT!~Z~L3q_ng+bpS@~F;_^M?Bxmm+ zW{qOqd6wonzJ2WdAHKmuD>XA$z0Wp+An3#zoGf)l!O|*M>)`qPgiH&%k!!dVpUXsA}Jb~7JP_OjhN3J@owPT>UH+!n> z=^5o(KR|922iqPA6gl3Db(P<>l}mmC<4SSjqj%NjI+MAMT|2;Opz4g7`}FK?rovwzl4yp{B2TQ0qLo z?%+O^-+tixj-8nK;!_)7z`KwS^c%T{`OTT+uiEm!6T9k}tvqh4ha7vz4|D@QyB^pF zUsrw0wQHbDVt?+)JHs9cyff$dY!cjoXEs;8IkWx?@!x>wa1VKMocr0YyLIo|+AH7$ zoQd{K!tY*d?ImB%#8&r>2IqcT`Dq@M#M;HYy}rPaXYByS?XS*QJGgr{r^KqWpZ^?< za}(Qp7}(|m|K+xxQCdq(I8ToGveC}6m-k_=cE%3uKhU19Ve2c_ii+d=j+i^|O5b+f zqr9CT^lJF_GNyhyGx3>mdK8~na|d+*q-O43#uCWw=m5rQCcWPFEaSzoF4w#la@U>z zJbS>khjXpT{HSf%g&+oeHzLvYK|IIVr6+;dJWT7{YZ41kKK%!_v6D}1ce~pQ+kP`Y z@>n%9iH(h_)ehRIsSRxX%igJBe@4&M+x~$qSKHYgTXA1&0ysl^py##|-L;im@0p+H z1c7fKITIxI03pavwBqk?YKK^Hd9T)^y?G<1L#>wTT*4hA@){~LW0 z{f74QP&4uE7w8MIwte{>-@d=2|L)KGT=>e!)ytZU)o-9X@rJ&BU8tUfYJYMQG;rd0 zb-#YjtamRZ&~nbdbLrjKH-ItiU(p?X)&C^L#jp227v~3@0pzMP_V+N_eFpj$;_he8 z+SuCjcWs~;uU>z~HV1d)>M_W@3wHI%k6yLgSA3v&Pu_pVx?lUa&lmIP8SAhA{BDyc z-x=OhC*QHSJo3&s60x5TfsQTLyxf{aU2`t+BvvpFV*7_5 zJ1+O{zpe8YkGSFvGgq%3%Wvoj8fScfv$H+T4WQV^nD^0R^(j8H>U-Zf`8!&!J@u99 zv3uL!KIY24;Ro0fv2yjSua;J9%$jti7(10q-}htgI?^X%-A#WeUAE(3w_%S z32bWyS}8w>xn}~zoE+6P`~S4xjTmur_Nsbp&toirdrq{mCwzI< z`yKfL?0L@8uf@Y&ed@8^f4$SbyH_k&8`yFOb^~XQ9L0WTZ@Jnh zHpvZQ30~R%)n!P4|C0&aEHy2H$Vs8`wKXJ z5VN0RkE%zT7v|MhV%qlf`DCowjaXot7v}BPL6Y-5x`2T%PZ)PrM^6x-_Kj9jKCvcl zYnFBA-SoW~YH;H-z)nihz-v(usiVA`;1og za^+Z?=I|>Pi(fhTfj@zF-V$K@jO=sY`P%L4dM0%adE|^giCeE-AZM(8eUtO^n&!ypIPo^?VWf>$A)tfbos>e0}LSl z+>vpS>V%K(Xl-MG4)ldS#)iP}XnTA1#P%G4_C8*Lds;8gxclmh08QkEDpqfK#wUK& zVVmoqINHR1rN6uro5U6Oj97Q;E)Bcx8-2&0xj^RqT^i)95%r8V&-%=^gxmxT$U9m) z^_`XL{4d+Cvk-P}kgInbU)}^q68p@uX5h=|@iEKu_8oWjxhHM!+8MqJ0mnP=`T8E( z$Jl{AYge@P6`nc!_!;Qh#8M69O!eOujEexLznqTmOlME*2CnoQvFh{#JjlOo-qhBt`^P%;)z?aXsmW5+w%z74HZJ}`iLcvjC*Gt|tT z)amvsRKF>N`;M<+1X;We}zb^ZA}b_jf9rR8mIC-%JI$C_=>sc{NUME(-5A50l-D1_8nZq12tsUgd7jp5e74oVEo_gwID;dk4 zYI4emwO`9+SDaV}p82x(4`cR3TccFp+;#Df>yx;!Z;KBYvo6r~tUT&)uAEWZ+&M3i z_cJH)4#w8<>~-#WV0Vx}d&b6&HGJO_&lTpu+2^{?+7N3ykn@h%d}8)-cHIH)k@5pi z( z4E09fcQC-o>1$gXXkicQ)l*;K`wVIpw!8t1pP3EcS?*;lcg|-~obAi2T6x5(4kf>8 zjK|0BlejW+?Iqt|G5x@o zZ|#MC?odD}=0DH1XCszsxaQlo$MKCJ7ks}PeAndHxzNt5&z^kad%uR%`H2nVBKLl1 zXI0Gj0kH=36I3sK;Ot4PBtL25btdl0x&{KwTy|pHbEr4MGZ6Dxsn5=R+wNVzU*A0Y z8B^O!ZQTGbZ~{GvO|*BRK4)*lz1z|#YieKbDHb|0B@)23N8I4)$G$($2NM5oj(YFJ z8vFPj9=rMyKIg(Ld*te2K7ePm$BUZ!>JKMD13dtG0-yiDPh-q>KRMQ(SmOUG56IV7 zH{cwlc8b;8zWP;f4Br{C?aC7iAkSVW-{6SjpI+6U)PVa%ey+d%H7`GaQqME@vx?Q< zvnX@kd?1~ZcOr5xBxq2n~_(|M-oaOxc>0_LjHO}nl+Iz<PdB zh()e5c^8Q-e2!B)7AE?=A9iukBiAgsfiK_MlH2loqCwlJ-DA~x_741e?sCqagM8=< z&O6bHpRWdg?N_~6>dY{A+*sDZMPv;pQz?{0N!Igeb zdt$qHA=cek=N-#){|k1<4)P^vbNZrc;g#(^-D@mOX6=X{z+OHV?^?KvaeG-ifp!x?9{Ainr+!KD6PObT2EI~$?&lKg z9^ao6egMV$bq{kD3v?1=A7Q`w3Fd_HvICy?@oE1?Y#QIOwz=vKzB4PyAIV(YSL4|3 z;#nWS-R|mG`_*1>>hEn3SL_*b(B_qqi$7xBnSMPxw7rBB$xx5bt3j7F?D*!HU$Y;zZT5tD09uKB7(Cw>XKKA-3kv>gBJ zpg7w%XU~a0z>M7syI~J-#Iux7ERra?fI4^CbN#haqBBTvHpSk?lpgC05BfEMAI5&( zeYA7!^D(|2)Hf>`yZryb0{9T^h6Y!ncp{k?U-U zUC-o3Sev*t^$hVJ#4q3u6#E=0j!uAFHGSLqr+3-->H(|^bb@-m^d7PJ4XOCo%`u)v zw+@U0&EyB-T;ZL4+5O}NXu0RQJGRd=W7(C*KljtgyPtY&TTWn?p#Aru{XF*ApAB?^ zs-t#lI}5FS-CRH8-5i{Hr||8Y;5|;N>C1^_n-6SpV!m_MD#p)qsdMbw=WRW|+6#{S z>PyYk-?6Hj45#R&%1ReZcP&FpzWM7{RwL3z5^Jw`%JX8t>i+Yo%0)f@6j{Z=j>ND^WEcrq~EX(x!e{qR&URJ2Y-OSgXd@92^4Ft z*uq`U8DFvGi8nA}!P!Ag%_8232dMi{gCoXXm-y`aAntj7`nLKf@gw%XkQ>&* zEzrY0Ef$PT?AjxV$wdn}YV*pST+8(=eRl4o@8{ZgEo149mg`*Oo?VVoeY4Cu@5HvQ z&dpxV8o<7CLmLEkakPPU5;HDe?M!vzcXaWvgB=q8;SO?$buiSNUp=zbOgX>EGX?@^ zm*mzc>l*$=E?TbQ9Cy=CP&>n&#N~Bj0qWh?x6ci|?ST$ZJJ8!6@`b(J-#$}cP+zsg z%GL+=@%eZj^8>DUpGU8Her-v`lh!~x0u4u7J&8Npd;^EyckKAqC8#=TeI`GBTl+NQ z^eVl1k+{H~^?eold(jZFfiHhxUvIlNzPoo~0qp6VhAr1U&AsnLT_+x(&Z15JcX+vD zRSU0V$BMu00oU&ids=&X%OCg&TCAMUb6`8Gpbg|r^+v<@ZUX;E^~LYG&M5bO`e7^e z&%Y6iEO+e`Ylb~b$hTN~SwFEm@D4|=_agD_G0=(bpkbR2HSnJqs521j`0l5l+UQB# zXK?}A#;sQ##KdprZ>u(P^|e>St{&K(yx$Y@yD-KqIWOew`$T&d?FZU9;`()QJ9Ys1 z*4FnDIb%hCqNO-Yy z^~ZLe_0Cdelv?*hE_3Avw)H+Exq)5hM9VQZf&J@UKpXQJxrclI3L3EhEf)WkSoOv~ zI}B_klON=bXBb%LvZr%|J6vyzGst@$d5xI5YpXq9Fbf!a$iu#%e*yk;U~lLv2=bGh zH3Qo_wQpiuThEQWJncF!b`Yzcx1}pj%sw?gH;1%$;k^v~8wE^!`(XWd zxb}S)Z07~K3wK55_A5@T=i<4ud^wF?0TlBI)N3(6I2iM0LD-5z4jAF zuK7Sarv&ZU2RcA;m}9>$#C>P#ZphiQIK+;YcV@}6Ua38{--9qHNphg-f{4iFC$_x? zzWqh@ysaATE0FUr_A{Vn0{duNn_ysjw{nfAzBT9x>}7r3Gyd2yXI6aIHsaNTKHBzg zKu#QAF1BJ`TV1=73uExD)1F{#+xbe*6;E{a3N$kd)T)&cAN778$ZsIj!T*4>YCmF$ zF9tB@PI9|G`qZ^8gV>o>_vPFjOSGssGfMK~4rdU-cTRn7(Bb)r4Q+7lP5Z>_j5~Mk z$$NRStGSv9-5%Bmc|%(ccHK4H^Lah9x?Fj&p1WGHul2=4*ADbCZ;yd)xsYhN`ZJ)Q zi%*Y^Z4dc<>!wYwIt#XZ^J#qBwWm7}^N!4wpgS;btoo3@b53ij&N#IbEtseSRNq9icSBpFexe83yl1etzA_UZG@w0V^=)Wl9bLWA=IZ@GJ2!Im9B2{5jRo_x z<-5as47(A#pkIMC`bzQhQy=pFUd4HhxHTQ^*{mC}wD%@D_@Q<$a{k(Dow{nN^BzPH zcYc70o!|vqZC~>peePC{b-w|Bk8S@Z5OdC!YmIX5re(R+v-0q)mG}95tzF5bIr_At zG0RtzZ)Xl?kt3Y%oY;29wr1cLM<0HoCoy$LTa)%9cevWt>nBhmHy`_OXDr(}fh~9D zs;#@APa{_Qs=+VJvX{0qj(&J9qz55w^Z?a6(e}DA1Kau;D`uU~P>J}MUo(i$G_z{V zH_*XQ)6?he=eOecSA65vr<|Y3VRxX^J4USXS)unGXE`gLzXe{Hm4d&%O3D7AvQJ?2G-B+m1zf@AuL4bLED%e7K1{lQ{m<7*g*{k^{~#UxNM>j`aaL zzIl6e(Db>7+EZU3cWZLx^lNYY0LI;=t7GM`iqmqv(-U2sAWnaIW7{<^7k?6qol9cJ zZ^YjBV&5QV%zOfEQG1ggiLLJ}#Qz4wt9Jt;{_$*m>skVBP%&!N?$N=q{oFROb?4+JIp~PGF zSiFvIK!4_Xw*!67S$g|T9e~|Rm}{M4Z9Olw2tfV~R6kxG;at7biw&cJq#kn=u|xhap>!yLBUH*6(i_3db3?uFdI zws%8^Hd1F|SMwL(ENec`>Iu<0#5&Iw`LVa3AwMh5k@t{`ooM@>XPG-sw&u!td|PK3 zvHBV-8{b)e*3M@<$k|Vhxrx?(e;>G)V>j~3#rd)L)@Wz0a|XHp^lk0pXX*Fd+a5m= zZ*T*2_@%MmwOr%d_xnncEV3>>y%i0 z|9ZiCIZtu}J%M%gY|sNJ8T&qS-AQ~ixOo1aMTuDaj<$Z#vtsx@yNa31HA~*bos7YM z-;a>{58|Tk1pTV3951*jS1j;`4D>z=lKdUa~#DAvp< z^Znf@FIOzo0T1|H;B)W99(y-?>W>|lH-NDOfi3>OL{5Mf>$5-6-kH*adSI-9dKT>J z+zx!k?c3tiJ;OP@$dJGfc0pgPK~JDx:U;`F}4OZ*W_zwElteE;myJNP?hS3TSg z&a8&*KWF0S&%FCibQkO|SNuji!ZqCepL0y5XR{`75q zA6nPVgR;qcXMR_DhZDc<2|eM6011lEKI)+lijQ`__h&q?Ye(#k-O$Fo@2aC-zrnYM z`I21yjGf&sa4Ju1kgq!I60|)7?ap$?;pcwG(e|2fyt_o7+4k+ZV7z9kfxTkeXNL90 zUr&zN)|a5I6G49BH(-zI1NVjabJvosd6k3L0wB(^ZpOB^b~pRV8^-Xp>-S^1xM|#a zB$6NSJKDH=+gHCpeQMh`);xIw9KU1x9F5n!MAQG>rdCe%-5huHy=(Z^-Lah~rv&X> z_py&0Yt9})PE6y#J9FmP@j4%|w)56DbNF=^=m9z~Hgol^{@9aPKQVOMv)^OvQkbQIe}y-9Zpx z?nGNV&~;Drx2L{!6+;_)7j0lJ&_PW{Tl0PvW8{td4ydaiJF1N(vEsuU+DH!o$FtU* ztBJ{L*fW>5M-xh}d#OFBJ$u>1oV?nFIg0j!e$KqZvH!v8*~xj=)(70GhgW;3bDaZg z-F57D&G{hk?UlK9uAT4$|HZyXV(npG{>V?WKKZ_z1K;{1^^CXQW9QKc$HgR^r~kKJpWot9tr0eX%)}Gkzr3eBCFU#IO0Wd!fBw`}V{} zysNZdu6K3DM~}1jAahA8TJZ0Z#p?EgrM%C+wq`@3DHPHaiaV zs~qg|$+v$3>#a2&YWM|I%hl7qLXlrRwAn*D6AcD-pq*u0t~QPLGXvY2`fkT=U;y_^ zwEX)GJFr0mW8241ur`5qf{E=eappomqzK3UTe@|S!Zo3BGGdQz@fbTOp+8xY=z)ti7?X#ADxvd%6BESG` zYx^X4-(Q=Ydz_uk$$25x(28{v9O>BgZaNpf{+-!!{CqTY%|zR`d_&0(M4r-P*^h5K zhwof_7bhVhBl!X3*R$g8v=z@klN#XcI-9ZhxDR)@06F#&yqb{=a9*4Q#m=YVSZ9v1(fGx#+q};tdSy9=o=o z&-1a5(vu&=l?;A}&GX)$;1$^83Y@1{KY{0UAM-O-{7su&#}9OnKhNS$?gE@+-aghE zAK1nRa8~_U5ZaxCSj*MNy&kt!*Rbb~+t;4o(Ta8gb@k4V_rzY-4ss#z)pPsm1Y^Uz z{jE#TZXY}0;Viz-{?ryH$pO#bsW*QCC9!+y3-I6luJe#r>?`-q z%o*#m>+>4Kg!xx$>=U3Ry}-bBjx~p8uAVjZ+5vj3I<@<`gZg?y1MBPV@*|hTg4{rR zS4zY}L)+Wh2esP2;`rWyLB4Iy3;d}M$9QNXnWgHxzuuczG4~&!!#CcuRWJYB z0NQtAPvE=I@SW$`0$guPyPMz9ejffFD1mRkUp5|Fk66#!`?Oc&$Da0UGk>GTIre`A#>Cs##gXS6bpm%+>?_at zE!Uob?y(?OUp;}`s-@oD7EZ1M@3$UTc)&0Rs&k#f)18}|^{VB~r)r#>Rj8u#_Ar=NLa^0n=8YUC%e^3guO zQ0I=Js}~{9(&4+4pMfrC=kNIMwU~-M2QXJ$>}D+2KK-_4CEDHV`Jk;4K|JAFs}5rI zE}`wAtwdr|*Q~|1wxQ)8F+M>Hu=~EkmlH`$36KD>U*`H;``MOL_n>xefX(qP25~vw z(Hnll9k#MI7^cR_pZ3-HX=R1DjBgSa~k;-2GoYUx@1`xtfLFHt)N3uKt~vbL4ps?lG~gt=+MMys*Z5ICcaaFx0vCCI4UkdBrC& z^Pa(81K+)b_Q_xIYbUhl|IC%wK@+>Ke0VL^&V#r;?IX`S8E9wM83^Br9eWxF#v&KL z?#(!RCUN&ja`qM8hi7+|+L(Ez?%jB<_jHiAzjMR@7xG`^Aa%a@*!gg*@8Akgt;ls3 zzdNk2=aNgiv%G^Rev36L%&~Le`|GATf2|(yMQn8Pa_gT%BlgQz>X>P*-)D`SXZ234 zX=w9J8^|lczQcV38SBjEwVgk;foB`&05jIE=AOjwzz_0X52 zyuis=_vBf$m7e^-KQj~G8u>3@$!+n>KA#71wP#dEEW7Sp8yK^%cQDa+Y9_IUZBJ)h zjG>Lm*A8Hx?{LQU8`5L-Gv9#x0jFc%)zp|9z?dS(XET$XxAqL~-H5%R>x{QmUo+UZ z&OyH)owccF4N6eLD|B0T&1LP;I z?L6??Es7yr6}7&o1AcGFD&t+Rkw|eL2PwTMW>Id3rw3O2w%QV0{ES*RIyv zXW~~a>I$dhpB>zB=*Qv}7hV##=5WpbmDqRVNG3mSyU&ha%@J3-kF#bh&MUS#YX&jn z`u6oKXCM2zlSp#jlRULO-az0#zwQ`1;9ZE{vF+8s=|9N57(?vc18{%#RO~Yn19a?q z?tzcktNO&+Ine>!#d*%Lu5981{6_2z1W0n`mFJBgu|9M1ofAn6d!jqAW(G0DIr8kO z-9!RpvzYSacpO(13k7~yjb^OynQ2&uYDlrdRu))E^fW^aNMntcX#XL zRv*IJEf+d^#NyOBwTVsOeiQ$dUN6{5OsU?95sP;rr`>X$w{72A`&{t{THeE)lFRp8 z--#vuh;>IDUAwe3%i0c-xV`23*)$&DiCyn4@1J?^pv9^Q^aS30M`yOmz3lplpTx~y zjbT5qkGO|*Vvv{H$jduoTt9$0pUK2dV7+_RPQ>NZY;6-5ZndjsesI zG@&F2fQ(oiZO;*)3+Br)_6Fp6)7pJ)NcK08^8MV{i ztHs*kg?QLU$$d6Vxk^y$4(8PQ6T51HIFh;dCigq2+E|>9_FkXH;hj47J9a~?8RU&8 zaNj@|Cs)jzuwO?jJr_KuT;s)|U**u=X(N~9tks|R9X-$ueFtyAxes&z-=zrZ;JBah z7jOZMc+I&j+x)rzX)dWjJ5QePpZb+rwP>IVI4$?R;s>^WrtN%DLTm!(s|T+Vcx-djuGP53bnu8{`^(2chOH#*Nk7T-u(2{k}hxSL;E-z3dv#;d$*N zzk0H}=c!r5)E%T6-m%Lk);7$!X8?0E$YFPMM?XOQ`4?K5xj61UV$Dyq{rV^9kB!Q= zRcl{!HIE+l(f2&+KkyU%#CNX&6nQ<%EZ+M}_V=&by_1NwZ-aXW0sqXaJKgti?&I*! zoI$-cYTx=gD4Q zz#t~ad$PyH9{6&bIW;luvu9#I4lhW7!wGEBK%JAmBR_iD;|_cd&K&T%fb&Y++Bz@V zv%lxPV-WDf1XuFTs2Y5Cc!Dc9apxSdR|0F#{^pD|k*oLkq&vO5rysj#l5>WAt@8}( z00X$Q_OEbWzy-LEy*qHOyS3oF=#H-YMw^?tb}%-HIXlsVc*DL^Gw|j4eg<{{-ai^*nMsi%4STj8{K>Q2os3I)}X*aru|SRpVD3 zY4`ul8e(!i-%t~8xz2Cic!8Mtj-Ke;@Vy_z?&yium*d`n7Cl<^G-i*w58`Fxn-4Yb z*p>P#yoUc5Y<1?!Pjo*ZZk@YTj9TxgBv)*E@4c%2@ZdP})LVP89(WdeyU#=yFIOzl zeSQv5x)zyo{0eEdFyWlUxG#Ejwukr`{9d$`x+px(hi2 zJc!A$uX`!(TKmPa8@^nhcje#>V)c$<`)A~LYlU&|NQqS^_wZ{T@e*{!;n$yM-SEL> z_p!d?&&@miJ%@d|JUJpUw{|aMjhH)Mz`zgqlbF7_7HgJwXnq3iUxD}GIU@J32b(L; zcqd-xM9Xof-1b0U@P#uvFlYbVuvXvPP`h(N$8O9niMR9 zwRi9?xh#2qCGH+0*8cVi(C|gwOMd9Yo}h#3O|4o?aNfO!J#yjxwbfqUg>}w*YNLJM zjLAKH9`bMNv-$_N&mlmMb*}oH!B2d1Lk-WkT3!>mW;iFXgII!z?fgV5m)pu)^x3}jU=qubHW{o|{ zCa2v&3po0S{oEmQ^_OFhK${!r#5d+l<9GT_bPzkU62G3`jyJ^T6QC2TyGD0o=ozi% zTGO#>58^N68-5xCN}ZWqej)b&&QeFL^Edn>a?QVz*S_Bt&-uoOh*mQidcgY*%mq;H z_{Mt@j|FG-4SDt*_*c1Tp~!i$mf26VbM5zv9=T=&9C;1Yv;TR$e?Q>O+;@O&t-LGR znflgRr#9z|FQD!**FN66vG3xxzn95bC$C|rHfV61>AqzX3!vYrb?)JwXElhuQ0MMX zpvbfS0cPxTq2*aWzyq#lXt`zt_QduKiEjAL^|LteYX|nVHb5tKC8ynhoF-y%2RcCs zI)L-Mi&tR1z4fgN;F$v7I`itO%^vQqsO@FGeOv$DMPl1$qU~{!gDo$CXK~ktJ#y)e z-LXXzaAx2GdBy{sz?q5fvotrr#6LM_PGax7;~SH!c7OS=)&u)J!Joc;{4*lP@ySF`R4{{Tj_npx`x?$Iyszt8+z*g?` zF?RNdSTn4bKS5$Qv@(((z-MdSgIr(-cE+mp+;Y6J>Pp-hwrm6enl&lkG03xIrlSOd&05Ko|oIYwjP)J_JUR9?Zr&b>{;PdJ7rFM4^T59Y z>*WVfGS+#jC$`V{aLko_T|4Rjf&ClsJU@N==-){14`#H<|t3R*DcgNpBVq4S6O>Jt_mrpsl##*j?!?tFig*A8Tjo10n0iM)d zf$@R1r`(P|fqICqYmeX2-i>!uyP_NUne@f>EY7heiRnLX%a<2Gt~0FH9>CAF+A}tE zZfIL)&qNQXJ+dlVmETO{!8#$(&oS~t;wc27m$UR8^hSGid4 z$=KSsoYT);m8#vdw9P9s`4Q`U{x#iTrt=%9eKC`K2llr1>}2m>;d$>J`1h;-C9poM z;ati&xA$$|Ok%)#c^7m;+qdGlO8N8{v7S|;mGAWM9cl>&ZN7se&c6ct8js|hXy?`L zXzh}(pL${c^lhDY#HtHo&JhtP^n+Lz>}ITqT)B?zZpMR{aq|gkr%!dokxR_G>2L?A z`-1NrYpaJgI6aI)QaV-o8d+M&&U1^Ot~cqi((5y(}8 zRWq#X;0v4=Ft%?z0y?-*7x)bruX7Q51?Kf%sTBe0Y|yP2%~-DI<=NgqBW_RQ+SXie z%f6Dg&mi7%@t(%?fjz3{o+~GTdnz;5Ipm80st@|)My|du_QJMbppExf@2KZ8Z(O_1 zh2L*0W^ZdJsC~#aa&qt5v0(ksMjqtlSf`ll8yV|aCBA%7IdlW=(zTJ3Yq{n*{|agk zVl!6nK`ihS|8j%-w^khM+~XJW?rXmzaqL`uUFbJ#W8Po&j8%8h zC*JYpT@Dvtu2RpE_>8rKeG~MJf8=vN^6ReP50JoGRZHw;&l~wq z&H=XpXE$Je#5zM@3(u50Zoi5Bw&!|3=Zv2BiT?ltIMbZ_SRZI}_NW}THO<)H0Qqwk z?Wqr1qJhuVd>BKX)&e=&iPjcJoGZMiKwCf1wIkY?I}3MJ%r{Vd&~iqsyt?ZHXzMzV<6OmBed|*VnmQN1?hKB*33=`%>Kt;h+l?~v)eC#Vt&i*=vo zjW=|n-N%^knD_Yt8gXOFjAg5v$mLsSA9pt%U~HXepIzj-<5z6YHt?0qRXd2ub@#eQ zZN%Cqxw;qkOY)qf$KFp89OH_)z<&T|z2C7l)H_$Mx`F-`jp%nU1BW{2`u;S0`<=Kw zBG>yp(UnJE;Eq@`jO*LaK0=#nR9XzmyHfNM)JQASKnd^+i6#4dl&}+u(Vcx!qu>sUixSpd4B!VAF4k#J^;D10J zz}kVf-Z^qQen$tm)6@G%V)mWrlONdTS^(^cHs`MHVcxiBXn$V4daSvpMo!%yTCQg^ z*LVFS_I@7h7vjEG_OULAC9u}qh~3{1F?&rl7~pp}5%}~l*TDc!Yr%J5E{XZwcxG2U zu?fuE^W2y9+JC_BRN zKg=M0df|T9v3hw1d%8;kMZd*92im!3j=ip%FDGsvJ6*9Su{55oq+=Viz5#o^QU4q8 z&jalhm9?F z)A1Ge@wvFSYKwU2!7p!UFa2k*hTO3zZ&SB~`;^8)(;w0#HMQ>nemAV#0Rz_HF= z)f=vJ692_Iw6*phYJK;@ZrD$1joZuF6Wh8;&U3j#)!+vxA!cpHnt#Xl+}6qobQ6f2 zm(B0F@`IQ#K8X3>kC7KZZQMQ+j2&yplel{pmz5BKQ8}<(QAHs}@d6 zKqzk_o8FFd*$KDecADjo#d4+Csse7{lGr+JY(XkjURda zy3YIsUHhQ@%*r|QoHcUw_Acxr%y)Dn=DfM>hOYZkgF$`Wy?2zY?<#U+}ai0k~px&bLF zEl)g|V~@ZV4ZEqGi?-%h>^pb>&n;i=uI2|k-+`*fw!UU<&dbhuPu5KG_OM^a)-Mh* z|9cQM3+}-0;EpX{u6Uzvf(!87eg)Uts!Q^L-*?`b1K z`;dPDzi&+9wJXY9GuHbw(Q>W#-t1{?pfAMjS!cqw_thBnCudK&a;l!%i9h6HH*9}R zw6+8NPzV3B7xn-NI)0rC?GDzPZ$SNeqXs;|Ijg%wz(Lg3MXtM*bL5qt?mo<+-Twis zKfUBk@)bkNZLxf7{rf<2jpvRBT22X^`|j~_TXm6WFoK!HCOS8)bB6%d`D=PZJIguh zmh1C0;F@c2Qae}wfld*tzxU(&^#-&(i}|_>eCq-zbaBIb#Cjv2{l`px@-9U(t1TW8mbb_E(S}Gu_oQoxj$;lOOQhdl0|iE5;t? za>dSW`dtp_a1{HO#F`yw^NO~-dS=XaPx}mpLP8J-cJbg`_Gpcx1Zl%_A?*E0^O)ncXBV_75@q*d2<6T-<*9Kb_#4c0gS1iheKWM&rIJz z_wAcA%)!-{)36(Pyny>g+~myO@SD|u&6Z28`mdpv=Etucu;at(bUrtj}<^lO$n z_#g0{bWiOL*Iw3m?vcbnor`>8+oN*i-MQndOQ?G@9;l&Vl@j}?o2eXl3aBS z?VK5V7TSBWhjXk?AaBI#+v#EMGY3X-++V!GasEtt;NOAbUcVe49l)3~Ua;LmzwVNl zwbg^Wb)IlPdsyRnoO8kN!5v_~zpJE`o~vgBTx_-6iS0~nYqZNJ-ga!DwF|klG5Hf+ ze=do>Ft_EZZ}fTM*G}*h`Of}9?g2dCz(0K>Fo`EnoMBJn9bB=S*kbPn>~lriOYYg< zTI(uKPRT6w{v6ATHLIDkAF$T<$(S7hO9LvtPqM&(i(n zy%H}zeu#huoY~MF$UXH~HOoGVpLb_C%Rc9BouByjx?-D?Q~OX`JaWdrllS{y`*B;^ z`d>i;WyE^d{?Ny^r!@~S@$Wa{NiB8)^&2^L1M=-3-~zYiqX+Q@0?s6N&K<<=pt$6c zocG|oh8@`FuYY_F-p7cQyTGwm)yM}gz+J5Oy=&j-fj+wo_yL+a*1k_`=sWS}j=Q%t zO6e%G&f~rhavf;f&zYy*9qezcfqI7MD|~-17uW-+?XBGfGnGNTJv>9kuCEMsk_WHk zv};f7iEY1TjI#vV-c`S2=UKoivEI5Cz**SFF4&&y34R0itsMKgmo@gPXV}(0&Z;{k zXWkyd-f{;zh$}NzZ)cc4JnPIAm)ryRZt8b(H50$Hv?9>q;Gu+|ywW$oBF_hkPD=Rw|HH5-n-<@Q*0 z^(?Wg|GwWKe|CMLR^Phf{B6|{ss69zM=ZN$>nEwLCprCw?;hv8`o^uT{0$EFi+bm- z+WEn;SLF>8muEhMzPs-D_I3A8UeOoT6K?IA`+Jge+iQ|P?Z&0n?22l0k)+*#&2_(Hxo#LX+l&AWH;&E>Km z#N-Whq66*xKjhL^-})z7iP*gSh#Qk*Piv<=fbZnI?Q>w?f#;Iz9*Va2V_ea<$Jo&K zGvhv0OP#-`zTnGi=tS3j%fXIVcYJ}T{bC$lcZl8dqsAUXpS-n-`3c7D^ZoT=0N)KE*E)H* zwOr3H+@<1QcVBYub7m&~v4{2eN~hLc^u^4j>#p(bYrYvrzi#ez$K0#KYihWjSxytg zBiG!!KDK@r?Dj^?o}O<+s~T(VCvRZ;uKVw<%B%Y(t{u0Mb~UIC0-f9~1-otSdp zTy1|pW{tTkoQX~_h@C$62zZT{xl=!|y@Rf%Hh_D6-q+gi#NFi$Ot=BAx7FV{i7#)U z_1)E3=es(wt$k2q{DJK(`!q0twmkPy^yPPS?L+@YO~X&)?5X|uhBlt7kJ?<_cYO35 z#LXqp?$l)L+4$x}SCgwdi?z400cw9{yW>R;b0(ZF$SZzVy04rOi&tFv#2#qp+COuD z?{B{$#+8}afdNWmc51!%nw8Y!+wTIr-v+FgZ_LlD`U(d8J6ik(0)N1B*9X3K0OyJq zVs+lOe?5g~-grY_@ZH6J&PZa$`_?V-5$C*;x2Jiv_ZHNuQ{OZ1xcfw`^IWOx!~-bK zl51R!Iel$=H)`GM_}0C^73Mq8p2RymXEp4SSar1voRO=4$|oo93Ey~$I#w>Q{{qj? z^Tc+}9b55^>b&`}hi88P`^vS}9*Vl+`&{-BwO1_r%t`zKoGG60np|xA94#k+S|n%* z&~iHZO5F=c{MbDPw*3PAss?iYP~TRsJGQWgzfRhF0_!e7F(-_x?dy47$=h3AqT6kq zqs8(kvBNhnSB`o03w7RY$4+8b{5!Te<7)3e$b~wIgUgP|Z}n zR4;9KqPY0p$2spLZjJpWIrlK8E?}FpU%kIuH{u-(AV2lFC#Cj?HQ%`Tmg~$9db!6i zhW!Sv_Q&@-K-(JUc^|biegp2E#Q)_td?bPg6M2e95ftZ(=otzR+x2i!@_J`K1} zZ2vp<3wSjatM>GhBks3#o=(2{qvcu`$1z0mIGtORe=*z-y3 zJFw>LZq0?5V$VCaIXUiPZ#gAsdyQQEJ9fhs2?DtD1^o1lIGBMZUcJzYycYZqyx0Sb zTyyMczLPUQ!2e6xo5xFbmG#~$7@3SvAaWU{L`h^W0wI0s0C_6h%j|0@BFI<>qD=QK zQM@wgW)6aQ0O0~MEC@rZAK`ETb50#Jt6RZF6k1UUqx-5r5V^JxBoZL^`(11A?&kgP zeLnooGyR@nt-Y(d(`{mpU(m~eyR%5Or#0PxqK?Lq=d(BFzh`dzFS*{gcRwLsdE{2u@@qfr zz)ontBRlPAeEXZ5EL!h66aO-J_VY8>BRzy0keGH}lxZEg*6X>ohImv`Hd_t)z>i+Gr0KRMd= zSJyZ-)=lBt-&%WT@*PNENiJgOL8#{7D1Ef&VJtwWhOg)wpAKASI7?35Cz>s1yhC#d zxQli21HbM?tN{Ilt}%KBw$D?pXEWDZTN+0#Cj)b}UrFilNAZ;yP~cN1&t6IY(pGDa z0(#(gLjwAWdqz36n>9X%Itv{Ew%jJCecgk!%BEP=T$TC7LU7Hv+!wniju zd4cvGvYJ}aVUOZ-=?A}ReD^4&eNGBF`63bbJHegIXJTr_bJ*LNNetq?yC5AMa?$G7 z#Om`dZGKN&$;9nr4`(_{T?K7!W**x;16`j7b}|n7((b84mjTxGk=o}EZQVVNHlDfV zg5JRR3g4dIYsUB7_UW+oPmd-1J@|_GQ~(mudf3M^RG*l0wB3C@^v$KkfOP6>G=(K8&;(JLaR!Tx%a~?_o!*-brb@X0*Ixa8{t^2o++&oYIws zh}M6sGw*(i-zW8~-dEI6{b9b=m6op@^3Ka>dll>sJ7L#a>@hAch>!V1PR@jQPduyf zt?huKE}#d#&r>+=p|+3t00mpvL%;UHAMZ$>HRkN&%vz5UWzhD`#N`!ZV6{La?pq2*v|OhRcM0lj;`Ua$X!pN^Z;gG^qU}ERAX3jQXA0~i zZ^Q&Y94Q$)mQ4w()P3FI5=!AF?;$fyl?#o^k7?;;B~}G!5(xT>=`-Xp6Z@>2X?S)ox2-5 zEtl}8*4P`VI}Z6;Kg6RnPL8%;kIuwai&lSp=+O~(z7pi@zb2>H!*e;S*Ct-?CR(oG zpZvQCdl0u*QI}R@j||55r(i3_cd57MbkX?EsqYDTMckd;*ZWQI`_mBQgu5Cy@7x;4 zjsYFooKi~bS!2|xooU?K1l*+`v^hWD^?r!EhloN9j&XZH?6STGV z$k?^78DjEh)-toiFoRg-9~t%|uT1cL|GLt8+pot~1L0vmCLJJ_p~ zfrDN7#H*m?^}xNxb0vMxhIWSi?coe_ffjkNcj)VPt(=oRQhU|`=P*68Ic6ky6{ye- z;-KCQGsnJba`K%aUtK^p4zDY%CK0oyN2digy5O77YPjBkJpCTj`S=~ME=oJ6x%zz2 z#;ld=-!JXigIETh$2uW@O0Ir)86#gny$|Bf&hP_tz`N?m%Ucb$dlv(p8{0h+ zy557!5QNYYGd{hYcK3yxJkKz4Gcn^+I9)mQXI}2v9Q;fkD6 zqQz2aT=&Vu>?=3n9|!i3<4%fw3ozCJ&$T3Ht?_Z5{X~#I0p!=3Xk2;L#DFOrXB2(; z>H+y~E*eh$WR|{hF)<0$*uHrdLv^r}7e*$V9@r3QZNxL-6@Qumu zf*7#BaXGUiXyCjcR`(|!(dODaV^*TP+P$XWGLXnA=IUJRC9x>& zE;4!vPv4%_%QJ3GhqhmRF4$oV$gef*u0NrsC#LNg6>Z;{QUG=W85p-V1Mj$g?ujLQ zC5V|Tpze?Eh}HeDr;B0Mkgxrs&5&nZkKdu~lYq6JV{IS(VQ+Pb=9^1LTYh)skt^_f zkg=T|*nuso7#sNJIxxj|UN#=>e(Fr-_4qr)t*JAInsMgX-<=cOo_rp#vM07ZJM_%{ z;7T=oV+mWCqUH7j658h^$9nw?0@Qm(1^J+^C+407xVO9&xrBC3t%2u!xg8k)Odv(e z*Rx?~a>hDfUF|v8flkDH=X_V}V_i=?1GV#&x>w*cU%s{Vj^Mk8y_dubTFIr&N@#PQ zUGZKN^IZ)8Y(4lru>zKG!~_^~7scnVMDfS4n;PHPVfZJ2zZQC@{xgtbZcV(P6IwfA z_n^e7{@UFD^BJBy>;nHz&i{s~v6_F`B8gJEoO4DMv$?+jXhHP{7?brAy=tzSYq zVoD#aNBx{**Sj?a*ZK^suf5O-ct`pPU8S_Ujw7w_E;@+W)7tubEY2|_Ze1Q=uW@dV zEj({ez60_KsD15)odkV^@hP|r_^joM+LJ!Mck)ZLze9<^?gpKRE5;RP`)ukS%6xFu5M`Rsz6=Ti5;o$O<--Z!>! zv11I~1sahX(LQ(9L}|Gm+dEND7LB*2R{j#b3o=|`%@mGYdnA0JEkCL83))^wV2vE> z#%JLfIyhAd{qSd?U@PVe*a3dVtaq0T+((}A6X2x8so$GVjF*4O?VP$NadQdZ9#zmi@Ygu^QgUf`>@F+d4%)pooPw{| z+qlTsHHUVF`zotN<2lp&@OgD$ifKwFMy?BLi}e}yl{&vOz=Y4hF1 znD^m6Q*>|c(q_A}yDY&31i5?~*6??MF>@)JuFqH>$|?FF z+t;%s>?_gQa@@uGJ>c8~#=iEfbJ1(__M={u0f&j(Lmj|<`heqHHN2jf{R@78L`+_w z&DVQFoA-{a%i06nQ?506sI$koPa$^ZQSt$hzO;RJh->%Jdb?9#cVOgJZEU!wt_`*54r*t;L^skUQPgW0pv~Pa2};EaPf_=(Q>9o+L`68fIW@dOUY%JM~@EMUOpdp zb5;iKSa-#?uEx;bg?1{f7OP5u?=vp}9qkzc?)ZHA(teHf4thBDuFrnp+RwfjZH;}5 ziB1f=5~-i{AST~^>=z(`zO)`if0StV;CGvOdu8UjpLN}kH(t*-}}r~H-{{T_;=|`Td%GRd^ojVFmJH!=X+c4Sspmqci?pBDKKum zcd1UG=(9_&9{{ayJb-;YpS$O1_o^M(_Ng`4?lS|%ywi?Y#xL^dm(fe&%8ojQdd#l0 z`OXuH_gSBLscyZ~K86}}fLzk4KFsQf)tv}ipTHW7eXUW(^QAJ(qc#VL0rK>_(mvzX zXHbY4SDRD3S93~VhWlY=!k(!W^!U8S+CGLE+}Bty)&0A|Dns5rJJid!wsO%)9BKOp z+C7A4Q>sm`B{A=!K7&FJFzyrD(R_Z^v-u2-ySsG>dxf@7#;%{wXlJdnumg1D+^+-n zSL(AUt&hC|7Ruj>99a`JE34G5p#*C-Z z&XKgi%KQbVcs=<7oTWC`!&6)D-s4U|-X7l7F8n1(0$*{ypUo*)1N+#+cn_Q*GBNuo zs|6;QnJW!T?323?>jEM%&_Fr#Ny>anUgQ;z+sF700IHJ*P0XHES$+d^l! zLlBSBYOK%LRnR3`-kBY+-n;Pm*wgxPuVdtV9!p@KAaA{#dKbpoaTMq17d3qC8pBU< z>RH*tGxpTmUyQsS?hZL;xs$W1VE5Vr+ufYAM(==HPtIO-C+vC-{B&AAh@0;UabSO8 zFJta)fBh=x0M_*A1f1twKl{!T-eHMT?~A++WZ+D5&e6`N`3dp8kJ30RY-@VpoB(xp z3_XbZoXqDT=R0`j(bR;TbIkW>dlhu25A?l{B{&YIAc##Dr{px{ z{EThigqH7oYwYbEP0>MoIruQ@=VRPcKNorhJYbDtOuq+XeIn-kM9jQ%jE(1!C+fS7 zo{%rm{IQq&6msrk{&KJjRzNZCd#g5HX9YbLjXxpQms97G<2>g$dp6W`_=?)z(?y$W zZ4evp#$7t{$H>byx5iJ}rF|ChC~jvS%nBm%TEO$xwJE$tikuZDcY|izN7V-5R#jTC+g+c_c(fmZ(O~` z@38$IngYe%y$BulfTOIXKd}rx9?~t#97sc8`!ls?`Aze zzI&EPRRUXIj&thWVB6oE_3DJ~qMiHndG^f2>$?v}d$wrvV~E?ggV(FsqsDzM`sfax z^F}H78OY0}&6iuWfw>MOd!n6J^hIg0z_z!1;khzeSq}WZ41LsabF|Nn-`)0_qQ}ql z*r)a&=QHd|N7VGsrHn)Q7s$E)N=P6V3BwTf?{^SaF(4)h4 z7ya57-4mM{L)SWbI5(3EpiU#;S)YJ)1zv!je8v}@php?)()_@;pVCKr7RKEBDo_T7zON0J;Z?G3R&aC3-e+eYcEfLGJWA zd86!L?<|J-*!5oV*TzbVuZ@EVSPgLB9PQT|f6eiX!gxJHX?{Xm8|W@tzM$=I&bZ$d znb;J)1hqHu4kY4=Ik{EPa?H!0EQXpGIOuu~w0%debEk4kYJI6a4)pEcfgKP)iD>1!r?jhG$1~cmACQy6G8Dhq>abmmF73?K?0^F&9HK_fD z`YFD>eWyfEJe4-1!?w4vUG#&F4(|=V;%CubNnOyunhE|AI9I9tjp5tNp4OGp>M~l} zT@u>oX%GL7Eby(#z&YkKFka^d8hb^~m?-Km3rS@o4?6 zFW5>h?fJPwy$gCILEQOj&u`5b8~5_rd;dNYr7!##YPCJ5Ir#y~5X;y}AhkAdhdy#@ zZir3jVa)mooFot1-o}kt>-pU!5vz#yPFS}hp3zAl#wrR9(C$DLbdl69=i&Ex^Fm9cwI}rx@XY#h73Tz~LjQVib`2?e)lXvP}AevGm|TzO0<~sjRi=c z1RCi|JCC)y^srYZW`2V0ENktry+dqeTrjI6XFnxJ^CxhvwU3x$7j*4|9``SiYHd%f z&c(OCeS(yE@~JvuK^6YMdU3U$D7(tQN{e^+;~6}= z`5>3kir&`9j+s8LZ$0+WRd-%$fwu3-gM0Jrg~4oL4{qdMG6t z>K$TRCrW91)V^@LA=VqGm(rcq!=0)W>W{Q~{XW=%m~{nPsrjJ(g4XN1?wdZ|%W~*Z zcg9Yo^;l!uYedGDUwhH7;CDdr8F-)OXQ0jx@*sfu9=%4J8+q1G;0L-F>9*0?bdr8f1LuPZy1siP_8eSg&7q zBxlU}Ag0t@(HLfG>LcEXyvd-Ab*0tY-|yUd*Vr>`dnv{*1NL-JwdmCH(B`exwoYEg zHt+l6`{c8TA?Ek1zEbxojlV}+v4^|b&t2qq_;uc(J96d|eflA$-^1Rv7!N@J;xsH{aG7BjBK5=v`=X{-xKdNKV2ZEi@>M% zF8LliE!wZ+?ri_##5=Tgo-d<&U|j;%I@6drxk-H~Y2tz16D!8h9Xc-tF0~nd4X>UF zy#)4Cj9EL*P4e*VC9E~3)S5z`o=QaP>)mO4|8hD|=MA&vj`v}%$YHLMqwTNm2HReJ z88~RA&MfE!PB2HU?=JZ6pbmN;2VDX9{&yrZ+I~~w1v>#_b%%nFtjJBkY|(l~v^((} zQCgidooBBMIyg)6^=#NZkYjK6Q19ShCa0u!kJtsfg1mhadIhg?@a_OfGS2 zjIW7hbiF4y^`D2iGr4sRT%|8G{+@-HGu+cy){k~a`b*#p&#ONJ zp2Zr^sLs)BdDaSf)kbH0pF@w{#oqzu^*azCg9)*5j?X9&OY+L8Gsvkvu?*^a#d%ME z=iJL&6|}tz+IYQp&f*!2#|ZKr=<$1P^3JZ$86At3t2Ox6CwahUR(aUgkM{Rbk8#tpUZ`};r z|Mt@ooq#(ScZOo$TCa~C(dL>WI-P#-2X_I6%pj@~`IBoOQD zU3qAE_Egu;47Sg)%2CR+7jQsVTDYKnPn@|lj;=eR>s^pr0pA7t?tt~8ajex=`XHuF zK7e{Wt2=e%%-g@-lQGVk(DL>`-HS82i~Dy&&YCH{62Xr})ADMsL6iBed-Pnhp|0MY zywaYbwHe!ZL07Raw)-qWofR!r(AFC3zznY9J0aiw>?Ox@>f6g*d+ZF{wGazzcdO@* zR+EX>dV16veD5R)MDA*VQ5sLZCU-fg&wAj&sFDLkY@9JO58TyQlm>hB*!DA~UN71i z+{cIZT4m+^jtpqVaT)%Sgon7yTxH;>cE0kQSI{Dt()1B;BcSBF9>(JIFP|tzs zKF%~}Z*!rI?$t<%Q}a9cax<`(9N{xkcOV43T!;Z{LQOmk@J`%eNv_ABQex8(C z8?ATHKd|k+hHtO>ywQr>M7-`<+Ov~C_U(uzdN_MFz#X#j0L1L!dyw&KoUppbS86YK zJHTG@yJ$VTGVFl%-0rc26Xe}Jp-&KVc7dZzjpI*%z2v6SV(Jy~W9Gn89ps2x@7b;E zwWICaqm5;By$|#hUqop_&mm( zJqy?!cuB!Fu9!Od213{$H~Jo)A))cU<%zb`mX-8r>6GjbV@ z=*0M(u@uKK_Oe{DLhfW6M!$ zyqwCJx2JJA)+eA}&(4nSS#4q+$mFe2wB?UyDFqwfJ2h@Tmtj6TR*4qVwvT6aUX^HP zJ2?m69`^Oj2_5(=YxIX0R^6?TvuB5v*8`t}@zJ-=dY_B^l)#=6D_{*gPl^WX#`zs~ z?Jpk`IfKpx+gf6QJ`7gCXEOzItqpkAYCC^M&a+h*ho|ihJ$4UAUIuk0+MIQr@n|)* zAGYx-rTyMG0uS3=>|KB{wKdgWv{~a!`zG>%w%=vIGbp9BJ=`Tg#+UCi_n-X}Jo_xw z(YV?@I0lpmf3&>)K)v-WdhkTKAEgL^MAUTLFE=9y8DvZ0<^Jk8x*v$PGB5F75XjKIa)& z5x0NV1_{_R38WPKzyJk|e-Ef%k1e*EUy*|(nbd_ivZR7HD!A2|InfvuI^v6yDiD*1)>U)f~ ze?ogtp1J0nN2~+hw_Np@_uY54BB%NV4)IzC-@fb9eEId>;9H~E*WJ3(wEc6m&yXDX z_SE)_1#Mq@St~yg)0ejb-m$SBq|*9j>=5`JoEVxfIHfrP2afpWUc&+@7d}zKs(d;46b(FK^tCA zTtDzVmwiUQ`2uPma)Dj*?31z08&jeTHaY`)rD&*o5~~vYWX6@`OKI>ORqU04yQtmG zJ{i>AiFL#}bD)4KXlLrzXNmS+_JG_eI+a#a&yQU-=QcO~eW=`e7v$}CnA~LOVL$I- z4U#+oN8U;fy$ZIm1OoKHxHF6k?_!+ON6S~9Huq|8zmI%pUTH+Bzd9-Ra=9oj#; zsD5cZ!#L+C_&s?!`u3?a={bR8-sh3g`igfY-(Tma#;~0uXHCwUdPl^~uZU&r0*+%Z z!5*;R1U&=xFyDbS{&*+#+(TSWBIoYzG}Z<29yqrUPv}~Q8Q9ikv>f-cx7?beSJEaY zdh86`L4ShQ??C;14Sv0AeB((SjaRVko$$R6dG2U!y&vM*ay5a$Frde<|dE7w`nM`?U(^xbU-Jw@Bg zKEisr9q_)iSLibM8D04SA0&MD>(M7?4{`5nX)S&%20l9BkLR$j-x+~6ZjW(q&tY#N zFUZ$9#P!YhL(F>pgwF6+pxy(qz&2lNE+CGr=+`rn&)92pk5;0zef7O7_fqE|iF7gS zQgEpAd_IFt4zZ5d6zl@~9s?!dm;;{CUiJ~jdr&1#i^HpVwY!64{!&2Cz^?PlaAveI zXDee(j#e|HkMR9p$|()TwB5Bs*Z$56;=rDw6k^0Hj~&cu+w07u>_DG2$i!AXbK&4&4K5Gs(%=6%`KYFrS0w+ zDj(n2`0V74C^6g-F(s@yt!_NOIqM@@FMWHvf0femb?0cIbhN!!-koRi4tjFVa37^F z)PZ_8K^z3Iw_2Wg?^baib3IV1j%wDuy+7-m*Et8ffblMkm(ud)dFIEQz2tajikuE8#{53=dE3L! zk9|9kOWVyoOSGMOVl|I%ZQy$k&XVUG`>X4I*!Ho`Sm*<#&Mee(|0{A6a10b;%LRPU z4f#G&oe}!Oyz#w|Tc07d=H&*lqz0X~9O~*0@cebA1G6Cx;&X!8ym&hDcXJoEzi1a59S#QP=)^b6Lj4jy93_o5~wG@o-0silGm4K ztv&ma+Gh{nUU`U{yAuC6{5`b0CT#O7{E@pRW_+BL$R8#z&$x9W1AX(gmp}CI2 z7ftVwSKLPl{3tD#!$`a9aToL2l^@u^^VRtaYVke$l$d=I&~|nTT(mthxD1SW>+B`3 zpnD*9A_vhH86;wz@u2sj^$X(OqqXC?vo;7|yq>+Zm^+%wK)yX!AeF}VjPlLfTlCoG z6>FTWt~+A)rM+8g>;UHF%ZVdspxFD$(I!_)>z~mn06Cva-EHvgXaDKZ*1HfNcXWoi zlYe%N?K7!D-4g8LpFkIKYixbL53G%7ey0uUT(mPLaD;X8M!4e)>;Pl>Vy#wUxVr@= zF?)-8XPmzy@BMl%>-AG<&(L|sjrG8MkSo9$=6h^o`WfvVS3UaC_Ahc40Ffg<#i`G` zw!d#NC%1CY&M4%1V5}o{5wR22Bi{WKyDv?5Lp<=UTk9_Xeg?H~G~d{4s2%w}BX{XQ z9$-%CN}FL{#a>JM5AYsy!52rdiSGdG$Me`Ti22_Ko5An#HTYdpcay4I4q9em?`3>sVCU7R!Cw=rk*=mJ*cJ9GlpuJM=F0C$(;{^s0AZsiXg z`wRJE2LxRES*sS~oIKj}syUSWj(7p~a!z0;?6L0Tvl;HB-dVsw`qJKs6s^}}K@aSc z{@9bak_)qfJ3I>Sf;M|OaO(RY3GS3w-3P8ZL}@d|=cPY8E$(dViaapx64B1m6I+va z|262a73Z$7`@jadG+Um25~bPhtXN~-Tq*4y+;a&|pBr+J64X$e(I@*B67V}>?q+S} z;mdU=KXdl4&jow@-jG*i)H!7wEh2evA4U_{%|@_fP8D~cLX!t=VY9mw*A*|X7ozf)^uq99HU1U;J$gV z1FdaMS5m9NU*Y>4-QQj8eHnI7ZiebZ%#N8s04_$l}wT4)>ZK?fqh(8seFQ=+uH zP%F+~({l-|DZqI*wg*Z&lacpxU15awxhj_$@RXfkM^wgUxALibLG@IgFl`* zA;#rY{;=m*=iW6>-6(-=+&Kw5qWxL;*r&(O1DqrOFkJh~Pw0#ebT`B*FWNnKlifGYSsFc(1WEYDS+-JmPq9M9AF2~H}l&wAhlu|mvw_UZBExR?6`_7ZO06T5QJ z^_h}if)(gMfNV}Mvy0SbqQ<@Km*fUq^e(8g;hXEtE74HUrC@`Mt#r}$k?UUibsv2B z`aaK$w$@m$4X%Xm`!UAqo}AtHAmo*HCT%4OpvE~XQ0GLOk<4L>4&=d3`rfB!vZl0o zYG)q?wMIYMJK;Sn2d>Xi(LW5FWA7dmknyXK>#_ZPs(=1dXEHZM<9jCO$rq>Zq;Kt% z9-d3vSmg&TuqVXr)q?_ReM#;4p^FH7;91qy^#d}#@D7qQv_S&e)=wAIqU|lGO0->8 z=7zZay_b%h{r%j{fVTCa9{B1Krx5tbJU!$&e>}^$$N2osSDPBOyi6>tIjzoq=FQ2` z_Vbff&_n%XG>@ipKQ0n~#1~4aI8SlqkjQJjRaVl=CK6By;+Z}hb10YfOp-)_? z{Al&=V@<}-YT((tzcR#P%&CF=@y@35@Qr0`cXhtneN*s1VBWn~`3c5 zz88hK-_M~DS1U$mIJ zOtG!WV5}?H#yhk-1lR%cb4gWdAM!n@{G;vOwdJXuV|@m`|4aN-nm&%-fysceW*0wG zyIQpRz@8HGu6t?|evs?X>9lx9Y`jlvd-B%!oc%qtbFFuWlh1JA)c)i;^YG<5X98Db zY-52gXmjqq2DO&+2Jw2I#O%Ks^vOG;#(H<0k+X-rI&=ossMq9sv^jUNZ$a+>e|@#z z1UnPAruK*5;~RHw#iDUC{^h`4a_yJkkDpC_dsUxQ!Y;sgP&4+_cRzRQv7K$+yLRU# zuy02$&~o*)*YFf;61+Yd-=0%)9VpoDXiwu~z4P5oPCYL<^YW{2Pi@ZDjWO&J#6c%Y zn^||pcZWdxEIMuS!kXF>9mM7B9BKDc^Kdd)5%c@G$F{$H3id8B@1hFY8FDi?uEtMj z-^nFN_|{Lfk*@G#fe-3l(RQ+~3~;7B>YkjlCm#6qEad7O{DkeDq-b;PS!2d8J++&? z>@U{X;xMqUzdy{t-Ya0;&Jnad@H1K{UGM|kg5MK}Z$)cS@zrI`K+}+;`W^B)*?m%L0hUZLsPJvwKdxipMh20NY-?)3q zNx*#~uorFT9=ikXQP9@Pt8r?rPsCGceX#0$V*xU7$AWgR&R9v6A870A4sh0>?m_w( zb|3UO((?Avx85^d1pH3VCUfPh6NU)VX&m#DKbh z89dJsG4zsynDfqRPEIttBQ7!s;)Qq*lwEvbPxC1NwqkvhHYXOG2L!Q1OuJLVT>^EF zU!OnPoWA@HZWzPPAmL|yV6E{Hfi1U3XE2KIgfsl^RXn3T3jT7@_!G4C^?9NbIcFs7 zsh@+-FS9m{otn<;?#Ne zM11_-kyClbl``}}d%h~c&vFX&xxjZ1;&A$Su^?=>!QUv>|J7Nd-cY(;V0tb z{z0v~)jI5em*5p(tVjD!DmnO3T1@R-doSbt9LMhA`0ukNbO)T1%TPnuyOmR8NAQ8R zJFdx%bwR8E&()#jSm#+Mz?fX^gdITbPWEz6MB^3o_?_jh=8e@p_!-+deQBQ^d#&*8 z*TpaccWI2i?%=PPfp1h%41D z)of?RHdoL+IQiTb&6e-?yzfd6Pnit;tubDEqjR(#=FKG$fSh9Q8V?+S^wH{9L;Var z1s(Bp1RzoT(cZCwUX!oA3mRX$2kB_*pS7*E4>P-HJIky4qHCNx?y!}rOWS8kZM_Hl zB|YpnBi5r6@O*y8JG8lowxhihw)2Fw_1z?Fyp#riPoMGK&t8Guk#k1n;~NY7C2*#` z{Hn3n*tP#~KI{Gc=gD(%wjfr|0)MugL+P4d5G0n*lIcM@ROku19P4C1DJjyxfBCzUD#@>%fG3jgd>lony|J`HU6; zUMmMNpGxDnlW~9j@eG+<4+?oX8BEO$cA&k#@!5>$?uqXZmv6l~0r!(5-`+KjF2I;M zB}$vAzoKSFullYLx6Zh=?yDZ_-KF+S2z9kz=(L{h)Po?W^ns0^O54poNtB@vdJ4Co zeYQ1+u62P6oY#X6yQ5d_hiy#&&us15I(*}vCxgmC%c(+ciJj4kbF6iT`g;O=#aUI* z@)hr*-eI)6Ul|APs5m=C<6yO18EU+D?Pzd*5%?KY8{dBV@_OLj-m$xldrq;fPvjF? zzGrn$?H#mS^PRY0KpA}cb=dYO;BH4-KE^zJrH?k>b85@&gn451*<)s6B^ZY`AJ|jp zgN!ZDoN>AObtmlAkycxKVkcsX{p_Wl(5FYchcP*En)bPPKi*3hEq8JTe%%wk_3q^T z*dxKe9QbD;&PdUEm~)?9a?_(jj9fM^mv|4y9SX3ISb_SE4DpP1wsmzMv^!ih@U6|* zYkfJOfXZjS+`bS8_6=>~_L65_uKkSr`#1Y#`A6HX^4!@TatnGk@Xb~3;A@Y2205Yc zY`?SYQFll8pyrtyp+eciv>RRSXTyH-jbSvHt)>Pwl_MFvsPP*d$G zX8*B|eCy@eUtRm5YYn-AZU2mxlhpK3%%wsIJ?09&_Kcd2_;K_S_>9&d6B9}let-#a z>v}N5KJzF(+PJZV?$B}+KM(GdO5<)dLk;&ON(0R}+ z^U?065LevGcg8R`aY%>8%JmxCOQ+oPxW&b9v%TgbOZ2MRE!KN+~r_H3TnT4B!& z$DTd98(>`l=40Roc6~PVlRqV%hS)J;KYA2C+S*Lq9Rn>#bn4RP_vo6VM_}vwT(zCC zCZ1#9pv{X;jlYI#K3k9NEYFsRb=cmAcc))-X!+J>;9Ui}-UD&>FX&DT_qrmMO8XpT zY-hP!564~!)aPH?JZ*b-Kq;loQ5&;&twGnh#A1kzydW$qYgHdOs;%$dM`~zRu`^taZVgG-=;D2mBaobzYzvtQ4-d{ZO z#O4NDkKf$m37gGd-{3u)i!XlA_GdSLxBc&5dBgTCZ*|rF^PlGlV|n71GOwPV|t z-{y|<+rPNm|F`X1ZFa7F;?^hb^w!P251z4k#r>|gedka9%=Z8O=bzsG?Tas%pZCm% z&DWoQ*Vc3X_0{vYfBV#3YQS+~?KQ_PbuRLyZ)uYZixas9rZ(jY|^Y{JQ z=byg!#1DPxlK=82@0kDgKRs=J(Rc1P|Hiq;Hh=J>Cv0wizc0=|yL!dmiGO(b{(XMz zCR^#zpWC|RnIGDE_+wu@Km5yS?}cCdf$j5t;?O;wbK|!j`ozaSGJni>-n4yI{O$aO zA3S^ewU0e_`=R-1P`@$Ajy%6o1u_>tB8`Cs~>{eSrS*_(@=_LrOc|I+s6>%aV) zTR-*8-`@Ys6TiIwODF#3P`d69?|=E7pW6SgpZV8Io;ClA{k^BWZtLauyxGC;{L~eD zZ@u>G=KalnV*h_%@#@X)-(R%(?YMmZj9Y%_&=q(8@cEzG{@1-0k2C*T47H!yfwJy(eAc zr~l~X4?JuC#Gl`9?_sw+=lsJbuD5lo*Z;u$kKcC9dH>aXA2$Ex8{T~X+?%cT{_Tk z-M{I(|K-rJANit%v=~m0K_T?3r7)ee{+4uXxWN@4w`opW58% zO;>OJ^?es^-e3N6?``LQ`_Rw5>D;ZqdhZ)Izj5@Ao1grtv-Xc(_Lj}R{lR>5&vX81 z|IOz*e2jzxhYM{>J%N@A{I1 zdtG_f!TiQA-TL+C{=)q2Db0_Z_tN>NZt#T7?Y?oz=B2lM+~$|Ae8VO`_TM*OdEf8t z-{;T%bpDvbADTbq0k@lf<66(#yy01Ywg1IC&F9y<)|<8-bFGKWAA8P4^Q+GNq3x%= z|0VOE{n*FmANqsO&HwXJ57~a%r{1~!_y6V1n+yN#CYzf-`1V_Gxc+T6-}u(MFPU8R zp1se%=+>L7|MDCA&$;GVo9(y#^5)w8BRBWiy4n7l&U@hI%6qN$zxt?aY`^EfzI6X9 zAOHCLH{blo{fA!v_czlMXYT+0N4_w>`bIb0djB(ick8)Nf8P8bAAGaDX?nuum;dI+ z_b zkG|*T;z!?a`>W|Wn>T;>R`aWW_~F~1e%lQ;XZ^%9H1zxT}TZ; z^W)#R-Tte;_qzLk^wfvUzwv8}t(QIdmGjsCR&1W|ln?CP`l5I5U2)Y7_rJROk%POw z=eh^K^rEx(Z~T=9T=L{Ue%0nt|Ly*V-tvhTZT{PNdGo1Xc-iJ-UwhBybq9I#fVbaq z^R06}zx9ms&)9#!ZJ)Pw-CKWobM*-Q?hbH+lNzD__0g!KXg<+Ji@Y^qrfx zzUKX#&)o0x`=5T+yEk`z@7)hBy!6E8$Nu^r2k*G$Rhzr4U%&bI{Gx-e{p$9?GatG- z_>0^B>el9#H{5*d)-O5q`ghYp_b)oUKy=z%ZXzvZ8NcK=O(_NvWwzi)N$vDr%xUi^`> z55D~Ke|P8_w~GC%rjOiw{)u1Qd*n3^Uvl=--m(9X*X$ge^|2Qm-0VqT-Td1*4>-8< z>+gH;;Wu2pdEwW#4?WCyAO7ZD zHqW@x`?h}W`0e&5AA88&-~Rsj`@4VskNf}e6VuJJc7AmKN%#23`4_$O*Z0qQ%(FHx zdf^xL4?Xn5=e^+A^Y;Jc@2-p|{7?{#jtxx@eS+4=W><-YT$ zefo^eufP1&`_H<=ZykEq!*90r{I@*#&=<~p$=>5%bMwu+9{VHP)5kw^e*Wxtw{HFJ zFYRA-hbuNWeeNgsKmD!W-TcTo_uRbnvP<{=>i6cGa{cFTZh4!pF||dOq7gdgvcr((n4B_miA6a1F7fSCo6n!YZ;;L#6H5zU^UkESq*m0vI13&!*F-( zL-x#LIi9I`$s&Jmg>8v9z>tu51?Lhc7JCGc{r&yD|f4_LRbS}R6Tq2Gw z-pU51+-9w=H%ZdD6|nm9NqE(6Bs6t(3cp@25qxZYg?~3}pfNNU1m6mf@f}2~RB~Zh z$w+uL_bG8mC?QkAW@CNUCR|ou&El&cv6pH-f@abpJUYH!Xp240-o{TP5{F`8ctx~$ z>-ZnxWX%Cj7Y8uUotyP({)AKrCpP-*2^ zOzC6{+U?LL(t%;HJN^;dR}qJwEwkW@Z~@f+i-DiUmb~^_Bi=gkqvkM=2U-31u(8~N z=6oK5;H+bO>qwuZZ+1UV(j1B$DeT z^6=O+1?4uYuyLoeM63P>$f!mie7|-M2G^LgxkDpZ(++HbZWqVP8DLy zxm9G-xtkI_zjEP5k2NgrS_q?~|ABkYP4s*H4%3phKxx_u+;eI(3jW_n!*)MmOnM$! zG)ogaB&&qC`>&FDr!2@VG=`H8yFvQ#I|xv$VK+2yipt9u)!<(@x*~QKJ<+v}PClYf zW*9rdhxajXq;;aGRcpWwJn|s1>n~#Zau0+@D`D9MCuk~EqMo`L5SeU9*I24j32dQV zqoXlJ%?LWyWRlh&F>n|&q3|pxX2x0U!ia@poU{eI?H>pEDkdw@R* z8-n%ZTGEuS3hO5Jqj~GIV9@PXta9>Lvbu6OIkoa0IjP-&gQp%Pw?$?0D#?rNDbiy7 zjFzF-wUgNQ!ajIk@f;t<$l`C`7F@o!gzVAziU(IOWr}H4qR$IE)aa8U${zVxqq{C1 zQ%4n(_4`#>icBm%lT8qdL4YLVGdSb<99Dbi4Dt7Df>^^eK66erpJx`!Km9O566%Z& zro(XMt3>kmRymp2704p*r@)^{3e+uE1FRNB@(E>*TqV8(!?Rx!xyU4#c-Dn~YE`0% z70$F{&^_qqHCISak7F|@*TRWsFROPttz_*<2T04aVXQO8L6nmz5ErMoz+_7u*k*2l zlL8``)6Z){hr2p&oSh~<_r1dUJv-BtA==?q-JG$mKgB*8L`dt zC%Hq`uoE&*V3qDnIB$0wYCxM!$!-TSvuvCI8qX*L2MF_==?g*Vutffjz{&=&+dLa~e87 z{Uu8tEXTnrPH@qp2abXa8Kb%w=Osok!zIc>O=d0DPwQk#%^7UxrWjJD*Rlcm|}CSB|a+e%(~JroWWt77AqZ!kg9Chl!a zB=V_h!r^JY;_N9;Snd@)@$|b{ScG@Uhmyl^CgwIXkpF?p<{g2WOF5#+;H7Xlp$YbA zeS(zhL#T`BK`ow$u<83tNOO279G<@esyoiXv+W1LOEVT9?wG-xjc$W%-6tC6oj`v| zJHrn(Kj?6grm=}0bh7#(D7OkE0~hb52g_pVt}&E;=r*R*(28!foJ$|7ctU@FB|4|= zIppnqh$he6x$?}{c)qzB@;(iNbt}}l)PWoZU!oyr2cdQT9r&+Zp6{Azz$Jcz=@lyz z`VzlD11W~P-#y7S=l#Ob4O)=UeGqcJYl-1oAKGuUG4%+@1@)*;Y|}F>Smx3P^}@sP z(6Z$qqh1JoRQ5s9pJPz-=mgp7^#~q)p9-tTTqQ5o{t}Y(PT=6^P=1n`!y@TIapx*I zyi}_U^)L5=@ug8(rM3U@2E zlMBJxn4)8eKORqGu~`(#95h%h^yk(2mF&?uO{)511$dZWgh02^fUXhLXOk^mWwRAx z6E8xH&PmueHw@)=Y{psF2H;rf%do*)h5C%W4;Q@NiY0A-;ZM~qu(X!O?|ho@tWPEM zB-ujas4XDB6v-&I2PCcT8*4d)?D_b5glPa+M& zZ6SB>XxKI*7j#-@2m|ljWtH05AWg>6@-soWN3R&(zBvwCSRbywppxlq+<=exCg_?e zN48sO!{(!Nshj0YTK_bHb{TuYUUxTAGDn#tse2K3?^tHqR|^ij8o^@XHj^~nd*pGu zJFT_Pz|5vDlGQ5Bw|z^*{Aq2(?T73YB)Qu)Z`O;R~V?0XMgm3%4 zRy0-df<=pu5%xEWWIfY|{ibrlbUj0sP@u^VE|#Gmh76}a^&McOj}*0RmZLsOeOda7 zKO}H@GxUG8o;}yG07DyX5_Qc`*f+TuU);RIN==rNHOKQ|^pZ#LpL;Ic{C%F#qvP1M zQQo}YiM9N@rVZcHJb=oLtt2DOLdb@13J|HLfIGAVn!7-a?{IEma|1%Kiu@J|w(Wx< zQyXBq&j2`MIE3o1$p)1IJ^r^XAD&rh(`np-zP9QC)2|vdPE(WKwOqkU4Eyr$>kTeOAK6IS0VsN(E$B?gqW2 z-}u>Z1j-c6$6FKHpe=bFGnKce18Nuvw2FhEga~+W^&$9;XreQVBj}nCC0af=8z=Xb z&~6pbMDNheGj?kEoLxIlvIRB=@LOg8PL3@u6+0sDdpO#cvrNypB!1;sOYe7-CU z?PQRhbDaM5KY>yGGvSeE8>G}4(BNUURBqZfI(Xq^%*&gHY1ZS(tFLz4DSHGz9U8{3 zhPi^X%oXx>LIiwV*Fp4e>+?g|ulW7vhw+}g9M5Tr;Y%0Fa!0wT+;O8b|Dljd&z4!! zZHrv!r#-kYFC)rSceZQ)4&@zf+yfT@L*IIN-+ z9gF9o)Qp#?p|p-aG+lwS7HWVnZ4R{fofgyD-7MJG7dUA6op}D&@1H`O6j%1)^}6zdd4uGYQ2+MOx{nA z`z@yX6U^uUH4|<#UX#POOf>JW&Xp&u;$&e6HQIfi-c;O1{ZzwfnPMX(T=z!za0z{~ zy_w#XK1tnTwfSw|eC!JNCQMrDz{*qNg++3OkfE@XO2(Pe`O5de`t^0J+tf_FNhzpX z{bi$`hvRLz-zcdX1*BmCE%>TN@4v9;mltm3_KWuLC439{7+8ii)77}7K9p|Ud4f7t zrBjRg`|xG#ATW~X@va@kxMX`JIjr%Z=8|qUUU*6P;Z-r*WKk_%-O!hA?tS-g({$GN zMLoHAPNBx)vOR1Mc42G!4#ROjmI%R-2U(C@JF^ov(rIIDX}`l+&~J-Crq@=9ddb7d z1*@5K+301|UQw0?E{vtyOuMLsoFfgedF|908Z+-C!*s@)wjZ=M>h%^6EQWaH@eud`s! z@bh^6dOR)_E%}r$!}z=nui=+?isot+&^Vq9gNkISw3i~i)iIZ}t1XetO)HedtkWWf zgFM-+usN9A;#QM4tg5E(>#&*y`Ex}%tG`0*=dFUL$?lqs+>6}Li6+Y*-+)TxaEZYJ zTVe0rQptdbB3ncG8-kU~bwO>wHD=nP#^Z0Dgdf#G_)wI7pJZ?9DIaK;8UcjQu1H|01XmZsd$gz(3$GHg10 zAohKCK+o`W@R+8LkyG`#(ILbyjyH+3?HS?{a*h?3mb2s^^@7`+6nNJ%8m|_#itI5!Brt{{I6#M_#9Bc zGp9=M%sEZ`a=si2SGR)4ua!vFWMZ724!P*)hT_0$F!|CLP`g*mo|X28yY{v)WX=Lm zjP1tFl!H6%su^#O;{ico@-06dh92%sDA=1 z+4_ZiNi~ODl{8{tbc3zJV$y$a66zOy#QzNRvHO@eZ@ywF*=jJHLRSM6PB{&;=R6S( z$60Whmy5aSG)?|BFdXZ}tuT0!C5>3v#@x>Lr=dRa;JlpSSF_oC-!?_x9TTu_pZW8f&XA+@g4Ry_Y^6mUx>lh zjnE-=lg#!#2Fq6bf!t9xWLHu&$WlM{`bnX9Mehp^8(WHd>#o3>9WTh`iTZ4ao-C|z zwZU6+`lD}8JU%daOMac{k%ZI+U~I(Snq>PlOxNpXd*Y9gRiD=j4F{uGQRheYSEG@% zxje946MPKT6*mxT*@r^s)?J{u-J7+mOk%&w4Zv*sQTS+Rz(&mrJw1fT;dwF{83Vu?? zg0I}-Ox%uU^nNeDYqEP%G5J4Lmi$%&md(w<)=Vv~7$org)fT*M!#UQHFU|IKoTBeW zegwsaV0wMSL)uy}kiK6s45WhO(M*@)=IDG>+JBH;Xnf3eC@PX!t9Mb)9VgI2DTa*6 zF@}m+dx-qIAaF77!`(Y#u=3^sd^%i_r3L!3G7T5uosu(s&|(BP?VQQ7;co2o2|H1% zq*X}G$`lQP%$R?{0CsZv6<%Y$h~Lpp<2SZOkWZ!KaL5xS$hvQcQ%uiF+$M3dF*1?n zw7&2&uwPy9Gyt>6|C#X#(xZ4$%NyLc zxE#9Ht%fC`YnW%gDc@NB2w%UNi8GS=;;(L1NWMRsb&s75k?I%7(jN+RZ{{~XQL~Oa zuQK93J_G6K%DW(S-~oI%=*7oO8H0JR4zNwmllbE8x_tHh;XHFuE)U4?n-9l`#y<%e(N)6AfPE8p1P#SgzeI&HuXHLGJE9&YT;OIir-z<7_xYyOZG;s!h zm}|wAJxB6{vIM@NY8AiwV+{Wu+`+T&{pC@ieQRw6BknSHAzwHB89sS_6zgyNW*25E z;O#>teCqa0-n`~MCfMrMew6BcKJGNHeC9w0%qSocn{(i<`58R+`5bE8DdAPQ#=IeK zB>GR*Aj{IjaQ(?hKF#em4`Vv`YlA&;tCr=Kr6*BkQSU4#Qww(1$M`JiE4=yg3%sQ3 z!QKSNVauo8;{uc4qyFfneB70EKG4^Re^#Hx!~dQ@jomwX?-VMSC*_mSYc0f2_Y{zX z`K+JcX`yJWA(rKjXDiYL$ZuT>M@r{0ueV3p4x1!Qn)wUU>?+}U&I%~_PaQR+zhhVE z>zd)C0`S3+cI;8I!oHEF*!*l*jYaV-apb>n(AEa z1_M4tEf41!9TY#@ROL_W)p%NfE8muQm~SfdS?i>rP^ z#^!8f^Ah9P2CGJvZ8@98xw_)O_KVE)n-Q14e}lO7^n=CuqnPx`{lXfhB1wLr0^eZ! zTm0!+FG>YmK)anOwx{l&uer2u2pky}hDmGJVV^$P_;mCzY#~}C{DK}cYrZ8KF25so zKXig?_64NC7TI&jc~px(z^nqz+0j`?YpSGq zhtU>EFf% zVq>$Hz4-f{X(Y#CtL?a&aWZwJ_y>wFOif_o;hO+&U$d^`=3x1JJbbS#6Kv09iupBl zlGY1#(yjgioRvc9e!sn}zq2FxJvAHdm3YI74@PLZdOLoze#l;Um<#7KZb8l8 zH|Usq2H*7=N_-2g@zHS^zBphx?}bf}BD)_|zZ&zA@5b;DrIS2+Tne9EnSp0aDo|Wm z%O091!ks}QA%6HEcqDb%Huhhk#1zuV%3CwUr4dV6HD-!4?~H&i%TGZ<^Bn3lWjFXM zyRxY%-f%JS3aLNZMdCIJOwyZUH7kE2$=j2_<=6+-pmPBy_0vGDBWaL23!&+g7JYZ! z2ln6F!(z`=;<93onxz`Ev2VYn_33g0?Guv8)(knNisWFhLq#N)@pLcWp<^5pm zrp++4_$stWM%mVPzYzw{vJh^s*$>*9r7*m748YIJ;(tXmKsN3P?8$m98rPjg2-%36 zit>R>Sq^8esKOt^7BGq$O-J!L^zP#b>ixHf&X+Bs8QQ7f@h1r)6ieCc1&5(XK7$Re zQzQw0Ldd9*Lm?&iknncSXtCgw9y&PnCUent7PiU@LB(<#E1_2)5AbgK3ogXY2`}2ULrfcjTSeIl7%D->I<`XM% zvJ24ExmRF??E*UKohq@9pW2(n8&el!KdQXoG~E}cWS61+f}I{QogFFuO~`5udi0Jy zefqcvZdDD%`vsAt>YNlcoOBYc{`P=Uq6qoSkV@Z`r;k;1_?!C;(0r#BjJ)OOO^*~( zwl)Lbd+YGI3AYO%KT)WK8e$^reZa>MTAxkjSdM6E)Tj=sjoB{K$`F ztmw=ULb9zPx8VcKIPek3z7AO0N1Hw>j$^U56KTfZJ=Dv31pK)3jU2ZxB&T(A!Lw1F zG^T%qGL5m|b#V-8_Ep8H)+^b$Z}(Z^HAOVovJ#)^PQXpZE=)GXzvl0yanSxJ3YMP8 zrO#Fzrav|q(%I9)z{O~~a4&ct1ibP_C9f8oW>t>ypU30VD;uEJRD+&_#HGW+A%EsJ@O5jzvA@5d_2p|AzUB!FU798qC>pc3qxV)< zo{z3MJ6&3s`C=%RVin%$Wxpm{%O$(ZoW!Vowy^SO1+jWRi~T+nN2JH}r}v!Z&@l&o zgI>QxytQIGzDj?C)AOdV8ntpV-f9AdY4^?y20NMf<1OoqEElVe&p;)4O-!EL2vJoc zT)R{&6nN~z*EcnwC8-FU64Y`2;PuRGp3LcrrN`;y%>}gP*EYK6UMuV=&VC=>V~F!qifwl)FltXJTeW=?vGrBrZk-{pOhc>JiD)HnPFs+vIm5`)weD<}S}YCC zP^DK!xYF|5DKOl6C`f+C!6bs=`lg1l zSffj!jzye&y8}Ha$}Xt7@Id@`__Lrh7FFkX=vBPP^jt zWrf&u{xv)OGm4!%B+E|QxQ)}3UrcY7SJ16lR`hiAXNa;k0+1?UyN62g)NfTdw#AMI zg-+v%s=ysq4(9E{5C$&jhF^Q1LBO-~@Bp^5XQC6kw0@d+LvXL3+%F=qKkcymg<~?H!UXrq{rmtSH5BK)6m99UCz7WTD9L{Fxd(FiWsmR)Q zeY2fw;w08>8;DE4M3NhmGKfX@b~utg87e=Fhw7lK?1bkgC^SRTy+n_H_>ss)*9XE` zvmDaCZ#(vkTYztj*HD#jgK2?PG#%rs#nM;xGWFCmqRTxC(QZ!;-mEG^o$KdFn#~ai zof`?Z|5k%9e1q!U0vul;;SW7-p`_gwOg9~*Kh|s0BJ0bTHFqs-xcrA|s7BIcl_9u( zs}*j$DvO`O6`-H_GWK;sjPUU8R{Eu6ADuS&BYd!p#-(BQ?Ay9eaHL=e|1e92f3CB~ z*S2{C>YLc?fjMGg@Eu7{vI72yNI=z#S#WdE7YH~!lb+h<11_gkplQT3++*%QpS5qL zRtW-?H?}3+!fu!wzXJ}F|HM~EEhNdLoV<&?jbu%>})2tt5Wl0k-<4B3tt&lU};-2ZtNAxby9`IAfm!2Ixpb>RmVbM?r^* z%>{TiWHC?wzJ>Svwc>|997Awg!LMABm-(_cGQtL-tG^}t2@YZ`D_?o zl?2~y^w9P9cpf}{1*b+kc!^&x%Rh1!OBUV{E8<^bj!hdTwuRtPe>+lP_ZCf-9mQov z(ULW-a$-TT0;E|+qPyaMm{}LZrFwbCqMiO!PV*?8R^o{Rb9?XT$2YOe)d*w_w7Bf& z3#e$L1zir8arclKl=mCN-mEDl=?R76wA~K0dfpLQe(wn7GKT!+P6HmW$cfjT7zT%m zCZmEa;TK1LpdJ^#P;GTZJGI60YeiabBK5hzb% z2RP8RkWS5?sR~0)b(nLlvT*Zp5$a_GYw|jYpQ7RuEIg0lc4?*Ag1Lz=yJb3H76i4c{ zAh)sQJsa+`iQ4w;`rt$OF+LO~1O7{ zLt;S$tbb66!_9WnpX+TvW~`5J@!(=^;$+K*90B|(w}US>aNu2rt=QF_jvw;=q4dX! z-ZS6;Szo`^(wW!)zKZ&R#khJ}2)f=lg6BOyk;%q)Ak@um}XVsO+%epO)`Ee4yXtSWR zF4*vQ4@1at*XeNcrY)>rZ4@CPEDXAf`@oG>AHL>G@9eGgg!I|_om|)4$&~kwC%P}2;QiiNaP7aTRCiDZ zOmNG9qw*7A^7lk>;>7UBF!s|baQ;1-4v96Sr>A@(sorT2z2_=iozp}tJ}ic_DFY!Ra5t)H zUWA&pW#sLgBz~?D$armQHYuq}h-t8gs~10s-^Qi!#@ui`y?8uNt&Hajy&BntYfAKw z{#t$|e+U%ieTUQQ#&S=m7_Q(-cw2fFbN=TBr?Ly-hV_1KZ?~K)yz$}J^X@?Bj8rHL z^}xDhABbe9AY#^3T(VrI_P<3x_={oR_)O2==u%P$g_fgfQTY(Qy(N#IS)#*V58uOD z9w-vyyb^JNbOv2!dLK4v#&gew=XhA@ETTKKn`nNeL}!gE79UQFz1KI(iDf250cY%zk{)$Z#IHJPFPWJcXgXj@dc?Uhea?wnp(q7dt<4Fm0@!46Bd7L4$x; zSiiFl{=_TM>AT!{#nXPgV26lvG^Yz6x(V9)tH9{prtD2jP)-IhaJN z(upnCVbW@O(sa<7zYp>k7YnL_rT$=?V4}i1SNEpifA)Y`@LE3f_mJAyr6#qDj;-gW z|DFq~zfJJ4hYlY;cntp0i@}!%+<547tJ-hjclqz)H8}O{0f?P47MFc^!E&l!u;z}B zWOH6O#66OtFLqsETRM#Bcfpi?N!6vhUL~`OL4)z%ixM88Hw9Ch9GRZfKQiOVDYm99 zPt1C~Ogx^}%udd|#9pB#Sr>AXSgMq>jXp1##54?>KFU$;7GK(?GlZB#->6ZC_cbGz zJJN6dZ7`L^f^yJ$yc6WfqQ9;u=hPNbC9iXWb^avueYgT=cYbDlPlSQ_j+OYwHHGOu zos3%7?t{O3C5%=7#xnfZiyD=Ybhq77x_U?<|M5N$*W~UJlw4wHU`{G?R`#N!FHtDX zOC$f{l*s)74eXX#BXpcvPSuBG!=znb;ckO9WIMHs*XG<30u(O72h+CRvviV1OfaJ9 zO5ezqDjhm;TRr>^mgd%P%W%E3F&`gq!Osnz!)~G%WeD+Dg=$utwN8^AUKy?g2-|&t$lP5p^UJVXt8Y zPD%BpD*|VeXC^^dB`knOts(ryRUMQ!@PLW``odf?0-UEt!nI#@;HBG-etdL??Xsnk zz6*9qVggUW3Z-7BrPNG(R6?ID`wvkjgx|VU7ZUf zp?Jyu^E5e_ed!8O6jEt=!VjCv2wxDQ5uGNt^wCK zdDwit8~3&*3R+{rXzR!cbZbBW9rf}(o%D7XZJ7NJg44z^DZ2+^>qH&lZp>w_(H_fd zZOgD(;VkjpWQ0<)27^*z2{eCJqBoZG<%SKWRQKNj3!8D#2`1E++#~&>hf&3x%M(v z_FqEeqvDB+&2*uu-~~LN^&9f`WU)|J@7~NwhRy%v%kDmPB^gm~(O_>N_pRy@@~@}D z-*xNQhTHih@q8Rx^~@e!c4%|^{)T)?y8+6oUm-J-&%w45r5cyZ+u`D+5OIL}4Gg?K zRm@yoXZv<_1(TD0Bz{~nngrX#3ITpkaoWFlF37Cq)2ArX?Z1P_clUBJ@`DO*8m@@4 z19TyOtUDff;Ex*p?*BBWi0PrmX6hzhTst= zU(}43<}W9dkjBM2e8l?$^YNgoxA^sGH5(iL8NWP=;8HJ1jZM;9%-f;C@2<7sJ}dh19Y<~O zv6((w?9~?=9y_xecHOM*Q#r;a_T|!7S8<=x3Gib;yts0}bHVI|F7IlQN zEH~C|frn~pwDrL@QE}e`5&FVlRhB|}RFPo&>YJcuG77z_tZ3Vnj}ZTDFHvon2otOa!0`8%VfL7Qcuc(? zd}wWg*>(mXNhn6+s3mBaF&~ZBO2Op81M#l63wav%f)yI;;H!-DHKp?5Ec3p;txBCc zJX^3x^1Ig$F<)pUzFM6i{%HgV-+Y0LX}HN$FdY<>_rVCAiR9UW8WO$En2d;3A+kd! zvaut}@xw_)&~iOYwoVRUijH~2zGN79*j%cqte;o&<+cMUICqA0&#NL4@Aa`vse^s1 z|5IaoAx0GE*F(=CCTf~!6OHBF;zpK&t$u23-%P+s_p@0+;tX+*hbwVbFd*YbxwGzchFj=Fg_Md?#upszBuCu=x}9VfnFR&~ow|!LKK%S8^h4%x;sU-KZeD=ppg= zC?odD+=y1MJp*4J?uI2gK7z;ifzZeKw6Lp`ki5fLlCWkecKyI2qSd5BmIhnmuVz^y zU$CAmdgDqK3EOB+&@HlBOBzE)?PXK5wb{lYmjtm|n^dkH1!JYX$flk@FedI0Sv9$y z=}xeL!7^E7;m<<$E-4hU+|G!lZQ1x(|8j43unR_Rzr}vo=0VWl{&cW^G^oj3fw^lH z>6j7)8niM+a%tQVVf-~M+o4@wNV1Iuu^Bvo%wM~p#vo${bDfvLZe6`3dHCl#(epnl zo_Z9G3a1>f?cr7C_-mEq)caJ)?%D>SAz=Vy6lp@ii{s?rI5Yb5OBWQc+)gsK8-o2C z3y9u4QPh*(1ilMjKx3^d{G2dck~33e3wu=A-59W-m^z7H zYX1!3lVc!tUNjg#3W2ffq9Db47u?KufI9Q(*t+Tey&G$~o^{q-^l~6_54B0h+eJiu z?jfST;3fKGt8$kD1Fm^%3`~~ZLu^}ii6<{gbB{~6*yR~cy9e*!ehHe+O#S8vBiGCn{7dXf!Po<^?cyP#Am2hpw;JNikrI{x6~g`bZ$NwF9vC6j zB&1bF@%07|QOb$4(2|pEa%CnnS(48NOPqSMN+oKbr$!fCI8N@!?h~sluSlxSE`qUU zw@B9R;oN_Dq%d3jD>-$nfK6*V0!hsRI=fSxeRvqnn|vHn+Y*H5gGY(A4HNOihlRMx zwTGBqT1_=i^}53Wf5a)K_Sm|0tT0#e5V!pi!)0sxlV>}0`1RZSc#h>vZj!x&9{=LS z{r;%&{c{z-YtII;j|l|#vB{7|#Mx_RRXYKcLNXM$vcx%LQ5_6;vbve8cF?_pad^}xPvXt{Xob3FL}N2m+03mN9ZM89Nh0B&XZHaUmHDeMP7f_ z$L5Oo;P!6>xv%V~%w`rlK@LBCwgJi6-$eh`T28Oo@|fQ#ynBNKhdfWdcHwd^Z83mf zoKS2*4rc-0#(OOI#q zD;aTdvq69Jy?nu0dwxw@9wXusdDMFuOk9$P-8sFEy}>P55_bV!`h~$-)o7;L>`cS+ zj=(SJP~P|FX3^A38AANV!;W$-ocM7GiQZj6yw&#LL4^QZ5Il<=+!iP%OnM_U9$dx@ z#tdSiZPLu(M01V$cQ4j(a6U%Nn}ShUzs1>|Cs^X8_4xUMlhDwmg)>H`;lk)`4_qJem5o|)S}-i}gsR{^c>dWiO!T<{=0pk=yY(>l?-I3ZZEed+)?((9KT~kK|p4xP{4}@#Hb-e9NHOWmsf!9xj)^B(`q7 z4r}{U!F6H*+2=J;oHSrDUTM+9{N0z?E|b^f>3{_!(XX2gKT-^uxr1TU!6J~IssRRb z<3Oh|2~sZ3K=r^)m^7`OJ-(5QDsDGPa&CW^dZmypGWVhNUBjru?jhWHDOXkrve-CEr`? zAmH;N=J2Z$*3VUeVFy)fo;E%Q?=U4AVEtYww~ixTQWN1zP>qoKp%H)OHSY!$u2Z_%K@A={wX-_oLiS&dCcpf* z0@S*-sNXedDmTzs7(^UFwzpd~#(E{cuwWmbwmg^*w&*D`r6yV-b~jl+(_>jYS84VEzB|a9WLGQ8V@Dailg~w;c9o2(2TRmz;kv)uFM)g zEmFqxn{Tjp3&1`N4>c(k0A|sIw zeEe3YefAq1Nw`d*v@!5hI&A7qhePrm0wyEq@+hvdqN`s!lesom) zN#U)ecNZJClKQk*LG+M97_C2)8m`<<^}Me``Nn+^_2MV|4s(Dglcm)5fg&}Ye~FCh zKE#%EmI|GF{=uM!{$M6A4XZxg6DPinBgW-g&}j0JC{EUaFxyEiXTy0i&t*DW`(3Md z-nb6RxeH)IbML!cNf1?s_r)T)IMTf9C(&%SgxUgEF>6#k+pyCf{u>es%V+iWBaSX4 zRon7}J!jG}(ftLxc;Omx9@a!!=3W+znh)c|;eav<@5xZhbE0;)5*@l&m403sOlhpL4I33+KrRp z?f#LXYvUdqSSHW4(+)w|m6sqD>qR2Z{s#;4dSJNaGLY%*>%V_^mDGiV2uEgBv-Zu& z)k$UpG4Fl2kUsLG82huJTF&|ltIujCTc?Bxu4R+3fBQ51rk~F&uHPh)j+vl6pfBtf z|08!dYLn*GHw5pjbn-N4Ir;1!Og3n_KwQ#BqWk5tINw^6M)n(x`sOq7%##;H+wBd+ z*GDr?(S=mISKy6aMj?^7@ItbWTh|TeaU_j(4(R>;_B8Zf)gc*KFc}Z4mW$0pL`<4> z2EXjyjuMxtRDHS~{kx$NGKNcI?Phtw`B1r-aNU{9e_X=v*_rU48arU`jbaE$iieAv z2UGWdbtLbsy_h+25Ud{mj_GUsLWB05+;wLe?$nAD%}x$tJ^C}S=HN;4*6u!JAAQVv zjP{b$LjZ%DJDKS#ky)@Z{I<)e=FGOC5S01_oX=as)^umsd_MxdIt_y13x|rY{Cq*u zWeL%ThLOXIrlXo`Ld}=4O{8SZ7MNeOn+zVj4Jyv%(ILmbLvd>*@aRy8NN)k3z;$5f zV+&rXtuSTa2GVEGen?)H2nD^JSaaVGqhxi#NHv9NrNuJGP%~jtW+He#s361b zKN6#BN8nXpf4bnw0BSkj5*Kgu0JoaqWZkSQ5I6e(+h~&5xySXovMP*h( z(*E7BXe9O7ZlVP{q6X!KL1$v+rD&H;wE}GDiwD2_BAZDZwcqvGzgXM z5-s-c#Ba_4;I_02+GL-C!-H&gG$9}Bhimb@MWtvGGaD<1$D!HzKJ1oLFiC9G0o8$b zdV3K1T)F5Oj#E;mM(Us7L6is84G{R6=!smd#f|^d9fFUGE|AjAB`miE$e7j1m}8%W z^9PN2D2>=U5V%$DeFh=WixI4dkhy$*~er)KVnYTX2LehW6&^gH%xz4AsoFt1=l96!7-lKpMT!c zyUQFI0P+V%(2sYw5J&uv?bCjTo(hPk1FP_Rml9W6QG`SHoE1&|Z0L+pF7#iPI?Zru z6bd!tYRr4POh1hE`M5jRaMh*NXdPmY+eU3*nZd_d!p3-Vwtqh&J-3IQjQT(>FIk0Q z-pkqLo8^#Q_Ew^C;E3oh9^hBh{CRalIgZn|5#E@btx1(kgy4=5+}7I@JY>A@)q?}ESvmZ^!}kK{Iw4up+gSg z+LlZnvdNUY?YfIOIelOwz7xIP+b~~+7V;_IlIeE{C}y1o^#Q}--Lw6`m7k)g_a9c} zb&<(Ci1?{aijS{2fG?ZW$d6tZtHNj~O-@wgA%!#0sh>CM8K|Pg_{-HLVtMt`j=8Ap z5rl>bMwrnUfNyQ2F>T*47G1T6jaLd3*SThk+r~Mef>aAJj0uO|&n3NG{!088u?&xi zqs2z2V0Lu;VX^sJE;%g)cxz0DaIvrxH$U(FPn&35pykX}oF3uL{N7G~_ya*>R9}dE z_KPXS5958+rt#>ho4NGQkNn@Nezk>bUhto94e7`jSvqe*FDG-15lif&YG(A2h7@OQ z-ajjsFE=^OzkMsA>n^L)l_fo->~(H$)-Z~!(>Xw*jMnhg&-e2c{de-kcXq73F#)Y4lS;S2>Lni*SCW=A z+MSU8uT-MG{{{=}-El^QZe~d%ok&+4-g?gWJpOPk#icpa)+m~b$?INS0WTYX{KvBKVxerN1OUkN@tYk*^ zjNkM7|NZYh&pG$FuFtigxx3(a`4*(dB)|xhLY5htK&)nF2!Dp=lAvu0q`{_&B&%15 zlUL3I-X4qN|Fp6v1BO7i=1ADVx&Lvr*)BT@4kkDXKUa8*Dt4mi7m!h&UVkBJ7gDi1-o z!ckZ>tP#aIr@+v(gq0YLA_)S=Rs^< zV~s5H7p}jp!Uqq^BR14*YW8u-NC!~JmNDp-Wqh%2^9~nl*+y4apIl6rE z!)xesFA~&DQ%MRpq~mh+={lue)KxQt&KFP8Pb0p-h&m;@y7?CLzL763PuoKJhw3>_qGs~f7GGi2$-{uZ zchSa@GP?JCcihtLHgULTK;LRa&qp;yjRl9h8H?JGWB%nsYGH zr7KK7(S@c}CBaA~OXgoZkyUualYg-`aHmQcv+bwC7>&8K>|POUis?fqzuiFHJGAjc zO$sBHKKNj;ChHks4Nm9Zf}7<@EO6Q&ypvb}QL75sL?v&iu)6`CL!D&XuRmgK<2-ST z-#1(~uARMjB%|&P;kY8bk?CX&fzSa5p>@&+(SEEUseSc=#yW;kyT;9=COZ)J?pCD@ z)&0T5Ig^?8$bp<<4SN3ZRQlyurr@p>!>&F*&pM+v@JE}kvAOfD>B!B2a52PJ()EX6 z@2L;i+YfP&9_CD+TD#Hjv!liNPr_M{%~JgOUl5pOGjIz&iSBJDAhe@{csp#xVaU+`io4$|s_e;(W8;E#84 z>|NmVPA$X(``3W$w14dEkMS(%*)Uk4Yz?oIez1oD>7-!UUS?bn1FL)zVb}hd#8R&u z4k&74>%JRw@MAg7S$Ybq0^E6EW-zVq;tkoF<<$CkJSbS3!1W?yy!|ei4KR;~{;s)X z?=W@p&dz~M$Z_YJ4===|)1S<0P%gJM8j02jw0|pe3sZ zy$U0sTZb|v1XJjq69`ei>%pcb8g>30L*oH%M~ zhLE32u8_^|T&XXtCi1mkAa`R3%^IOZO%!)S-~MT2hN&Ul@%aO+R{RNV^$TI2@dc=# z7K-j^J^0)LE!?Y~C{Fa6hf3P-;q4+FxN+)m_f?uhESSQ?*T%vZQ$jcU zk0l*O>*&DVe$*jvAcXbli`GWJh^t2@33%NuygD%r7Jq#YS6*aOr+gWy@3{&rlQ+<# zu};F+wS%Z;ULKv-e=rW2-V3^RVsA2`derZ>&oK;uS&DN8m#qhK=0Z&IJA~wVa{~+WmP9=S?z=-BJGXn zQAqx##!?})Ge!*XVV{WO(e&!ZxUpN351Zvs9xk7D<7b?{lxSgi2b zkIP*G@Jo+wpxVcUj?31eP3G@FciKe!E!y*(A-9l3<+Cc=CU#CilJ-iPqP|9F(aWV0 zb863lgZB}(BCiIoH+Q8U=Jll?Mw!r##z{Q)bTHfMT@8CQ)cBlH`z62ZV;rS5gD(6X zPK!S-q0d%U!=s-^p+V;j48EVt_`MU%Fn>GN>hz<^VNa;VpSN^OWB^XCvB#rt9-*5h zW6v}dxYzm_ydl1x{&xLN>m~nW=Lj|aNVfyc7d%4G+1;s%H>J6D5}R-E5{&fxjk)%k zJootna9UC(zHKgp`z>ye{zQN!hRG=Na{)U%MLO$@nevJ8G)uZagCxH~#;FD1{Vqz- z@t7nG>sH79;4yM>ENbzaUOi^+wBo+qX7j$}0WNy8Mrd2}25lAG`M9_l`gVVCE`HQbrHug*TS)0e_2lZok*OZU1vV zcElW{b$fYh)>CfS*vYf}yFl^eCu~)4FW&igJdZrGkT3od#-{~e!#}rf^Wr&)d_|oa zPiyVUJGUor@58fDK4>d1$-KjF%u(T2<`>{!?JQ>TekmF^Br~i3P77gX)A-7x6}Zwr zoJ6&R;E;@IsBODNsQxyEll*jSO?pt$b}a(hW_v@w>yOEsR|EOs`X(H3uOGh|ae*XP z%wVM#4zUSFd$ExJ!*WlKLoWJov3v>lik!@2hGhzK_WL3<7qH?j({Vp7#+Cu>`J5d+Zx_CuBc^^;qFMUHtDy0KaaZ#kW;f z@|sG2x?F2E{ZPM%nhkEi^op^l?0Z`Fb4V&~IZ%W*Gp-TcN_Sy?+9)>bxe?xa--miy z#emo9JYw@L6$gc!MeQ>KdBHhdo)>G!5^)vreQAzs9Dd++|9^0D>oAaQ+J_swtMF0B zQ8=Z19q*5|hT;8U#D0 zN;>A*Zw(=|C7Yy=xGm_EsM4C+p7iV1XAp5Kg*P0D;ys}cPhZp`9@goQoe|%%<@%Z= zGPWnAr<#$8XUlNY9I3vtJH#xSbh+USW&Yqzy=XYRSw0|s2eVzg&!*w6E0&Exd_C3% z6Z$&yMb-Pz^Tr)geM}D*NOzC1kI1gC?p)FW}%7Gq)MQ6ATx#UUw+FR&=Xz%C2X21XMruUE`&!EZS=oI{v>1j0{^V!!jkF(mB@7;L>q*XCcK_OG(2)8lST z@77ZI>Hibdhn$Bu)~ng4?N?xDcg+%`KBI}!bHFe6Gtp2d5y$L5NW6ZXC0Bd}EWt0R z(IZ=QHVLIQWr5^8kECjkGHBe!9kfUMR@`*o87~gI2-9X?rn^rvC!4+{@R-j_TptMI+9As<$C2MhP_pvlD(s9ER{`1aY624-l3jpcsyzu6C5 z-(|w4mCIqCe=buiERe;gZv)p?&G7i#OLF|$Ahz0dA2Zc&Wrb$VvTvpTf$_*oaOyBA35las9fxPx$gN&xGYy$a{Q?n*CSY6b71rnG(0H6k>m zifOGkP-6#LCT7W8YNzRMmiPny7@kH^!6dlsb4ZEfD@x_oI1P zlbBgd7F)fIv*emdG_X$^Z5M*+bE6Vc)L2X=>O_!vbAMB#&Z{(RaSfGpTx4|1D7X;k z0IGJq(A0S*t{+#zLW=f*eOL+6{aej~e6^W=$YHE`q(>}Q$!X%}P}+T>Hxw-Y31_#h zgDL%AmT9 zXgrujo@NEGCBaH1PYySVUOviX;B*T*Hy|AZ^uq_Fjzb(inut;l=NidOY~O{6SiBWGcS`Qc>Z867)+IX_?}8+Tv0M} zPEW%@4oNoG7PJ$~maA0k*~L~T`vJY0;7S*kDp23Pm28%WBdU7Lga2aG>8jGJbXB>U z?b4#J)G*^e+VFY=C0`HHY~iZdW4H!8xks1XQkMR6t@B6+gk#THNT2lkMdyxIx4n4X zjFdS3V7m<6L9O~bOt|brRliJu_x;x3(*@%st$hiLtkMVFv)1&#-KOv~Q{tVO3gnjB z05Zyp@IXClzIaU^Ui(jx*8R@Ly%h$yT;Uh{Q0dDw`xnY8dt4_^j@m-T{wne8O+WZt z`cB#toFdy#?Boxfy||6cSSVA90P~Hf&}*55i}*5@kJgdyv2Igv!gFo9YUms~)vhl+ zGHD%u-~Thdy4VLQ<`yz}VFY`Bp}%0SScFjt-ZWS@gzlU>ACA&paIokMJJR6Co0C7Y z$v216NWTQk3uDZ+xF?m5;*iYu!~bj69_n}46305CnQlgxJsr-SbAAbi<^3T+>`N2u z0wL~;I}Ga{0=aJI#5~EP*sX3it#Qwwn_n@ybM_19S1^PQvX!(j|BQL(`F1vc#$u+q zYX{B!f439R4knEgsZDSQo%S2)-&~G+^9Iu^RV(Pn2hZTme+f+G=qABOra`|N6o5|k zWu|vtpH_J{!0daqBxKo6)>E4^d$U6@t+Wdj?Hl0NDkb`4)Kj>$U6U>hpF_XRcL&$q zyRB`bEa>P>QF!=ZI(*3=K!>N?7*wiY^{bC_t;IpuYW@# z2JM073Z&!QZeY1 ziUHO>JA(V1RzmQ5Irivr5epq|V{+0>EFRs4K|MCI;u8nZdFXC-(n5t;o$k*+pU~od zYVSz#@NBY9*I2mxFI+HQlf)x+M)9$aHHcn=ov?4{MOI~U1b5tyMhCliVQZQ_I;DMQ zi)y>!tKs2jUVN8zchhCJ7taLoupa$0O@YQ3oPaY{ibU+BY~8J=%xl9LX4Nnq#|C#{ zYL7OelVL8rns=T4OF9USo4S^agq?FQY)0Gsndu?Byl-oZ6YFGupPZ^iTl=O zVJx}D#&qhyY3>PXw{3;Vw}Ir>4J7pME%||^6=KKei7@>?E$Z>=68X?;jQHoN2T|6_ zfy!w*G*Z=&sErI2CZA44^YW8mHsFL!t(hb4d=bG;e5n*|oCJu^O(l`Gy3mw0gcZp3 zXomj~tkW2W=k{CT+rE}$!q_u#@|Hf0c0G)K52d}#PA3psq;pwmE~YEpV;5d+XIDS? zixcijykh_FqK@7l)>qaDdvCZx|H;Y_V80rgr7b)jhN_L{Usm{M^-2qY?cL0wX`?7eP^xKx_hd0n|IOc|pmyKdyAuzWbfM)-p>XWcRJ<-s zg4KJwf^BRh++Vx{o@`DbkEtUJ)nssZ;Us#s&vDS8E5RhY2wqwYAjJi^YI4Vm{&k;8o$G-fJS~0Kcj{n7kvbgOFdFO?1Z(-BZ$uT?sWB$ zN96EKEtYk3D{b2??M}bx(S0q+Fs8y34A+*zf$wigvPp=jRya&PF<6l*c{vdGJYD#u zp$VVY8yCNta6fkRq@t3TQXc6NBB($hD1hSvR-Q znDOfzo3KI+Jm$IK7Mph5>Usvo%N=2#mjiStRDtc?LDYfkg6_jzQeAIKN8IQI-cubx zcbJUZ_C1Eq3(fG?D?cpeAJE~f78MZ6>Qm*glNvbm>50kXO@~*)GO9&22F1xg+*H`4TtT?jU(Zk4fk?MQ(YcaX@x)|SCh536KF&|J<|<6-(?$^c;-0VZc7)<3PW)5 zmQnovu>}~BpUMxb4W&m?!eGa@d1S`79HCV$6yNpE#EZwKGP{45nSS^_;oeRgxb$lb z9IAVU?!6lD&w=5*)Gmp|K5572FID+@%fZaPMhl(fi8%OeEDTWSXLCExjK^K}6&=H$ zldz6DLWl`Q6iAxcLsuj&!)4-;qyvAVKa*7{{^I3MC*n9a0^SFl$4T-`Fc`Fg7J2j$ z>NJkC-i32v!H@T3ZOzdVnvJRE`gm_$3{hIXkRSR{ zi5+KGfZNt)VpQrc-SOd)*0vAUPKh8-mG@!wu~Ia!4`hV{ZNRl<7tFb=Mdeq|virwo zk&Fzf9=Ya3YeSB)d&l>C{5kXO6Jw)K4gxw+H54tidaeQ!$Xp zNn(#o81I8HYm*jLygQ2<-c{oB&nodn>D`!Zm?w_N8H$-p_v25G>$vUQZ!pW$#Z`kM zm~727_@?WJY192NQS~;XyXqiFBMXMBz9lQJKOkRw`k*{_IQw%VlvUnRCO?m!g|c`? zR{zk25uP`Nj;24@FZ?mu-?}f1e?Ee&K644SSPX>;%6^h$I-2bnuEoM8d?Zs{+Q}L9 zi)7{rYi1wnfkR*Q!tfKBcv!Kg@MC{8>JR7*&6$4%i*0+w#S^V~!1DyWv`+yqZs?6F zMyv5q;bqJ}*oGNur!ddpFpIii#}4mGWmC5AVxwA&`TTNCJ~6@tFM01+UH7&N|Ek;7~sg$gO`ZR(9zD z%i0!;Wjg*)e``Gnh$XPL(+oeKJC5TnEAXS5U2)5nqhcSqJG?Z|#2Cys^>fRP>G5FOwXu4ZCn}V>oDbU4v8Js&YGd4*s2EMFP%U zBinZvus$xnX!j%r2KKuEPRc8B(YB%ZQrb6oS}2pY^?(!Y4+uk^JMq{f-S~$cN__aA z?WDYdh-~IF)^oR!_~^H*tUmGvYd$;&-WA*?PGl%aDoPTEzFCM`?Gsp;$3pSu;8HQ* z?<<*tkt!%9-C}kXQ8>RK5!NYh!bQO+uvD6leOQg8W>E+0^tvOqR^>_%k#Ia?;wkJM zmMrwT`v;7ECPU}33nXNqC#KeykhFD=Vc8cGx-MW3KgGkiUQm$eZk3O8g~qZi0L&X$8jKrw5$}bgtf{KrJffW3}Qf8(ykcHI)D}a-%0h-Z@8&v3H~ka z3A>xm(9Qj&`|!0R-PQdU8~EmraCB|6u_g;@{)Alefz-flG(7?zY2P& zRujj2%fa?}9$j+JM1JE}4%Wo}V_PidihDwz5gpSAlKgllEYsKvS?9F{y<<~kc^8jM z^`<2a*wspQFL_Q1{#wa?E~pl}Ijku;X_rnMEpTh3) zNmS?GM!XQSoAiVyr02we5cKdh_IKFN`%l}&=bjt_>+14})=gFFxossT9oq*#{I%(W zk@NA@*yC*R&uGxk(Ss@a{e-9zP7c2kr1w(9>|A^FtC-HV9vu&wVi%|iI)S^^8_DA%!J;Pr=Sa~WLK_9d`tU7j3vJmene)G;%aw^i|9dS=NRE$$6jo{O`zD( z;f3V%OLRUPPd<$JM_$ayK&nDnsP|xufBpfU`qjf-v!l?XZ8ZpQyV8Z%{lPl+5Lukj zpD#PTivLO1qW^wgg$?J&(PuIJaPuh@Zdf`Re|2?3uSFH?ea~tPdwCMyOvu2e_bl;4 zdkOpR^&TkKZ30#QQM_VOBB~g=vX-*15;yD^%AaN8V7vWf`@g)KaJ2p)%p;h2mgc;S;Pc9VFx_qJ}OWuw$E zrk5v%jkt?zo>$5~RcNBG>PMz9H51)0&VdS*ZrHX_%J{G-flC1{P~3L{>Gk$MX1r*Y za71il6T$|vn_blSy|3Q%Xx169>o3sUxt(la;#eMcXC%f>yv$C`%)k`FttC!TbF1faJ&gFrx`~4wUf;4KJ|kvR#-fWhs2t zaO9^7z7ky{NvmJDjBWGi0`#vHe7rJ6*c6p0=uf$ceS3|7#)Kgh`X8nDQWLnws|UCs z!i4wp+AqvGu?m(P=t}y<^`$nDK+m_v&`6JVcIMC;_E^#&$n+hc#Ci-Lp&(+>ij~Z< z;X0Zo&EdsrgX!e$LrCxJAv|uEAs=;DVhu}7wg|^0NH`w^Ul)c!!7T+$Ik(gXu0n_w9K(N)Pf_^(P9|w)R5#zi!lEAf-F< z`p``pHtd)7Ht4<12+moIEfwga43tww}XdB)7TYEOJW+FHg9L$D^;ZT?w;uP_aWrB<%8y;KeA${E95}5 zKwb||C9@>&-S^)MAb0UqeC53udtK7T*e`nYMS?k|gp`Ag#Ls6Q+qZf)vus`o`N!QLdE^9=XqAjko8HL}N?g*~x&1INZ#Vlr ze-oMYRF_00cZDvuQ%Go$hmYaRBif?9F8;;lM|M+jP0)YcCR-6HUAF7 z29&`2i4kDdH5adq8bFF)&=Y*JOv%Cd(OCB?P_}j2busqI zXV&e9!1f#phx!^z($wn$S-!k0uK%FG4$tW$Z+;d=hCk^6sUtnv-H`{uJF!3MDRHmj zO4DKG$r5=?-2}Qu^$*Y)Kk%brF}93+fF&*yhzfm2pwDypZa)HSOm^%L%d%ZRmRLdv8ugTlF&6;Ae-cSy2pX<=Fqeob1j+}U?4P**Q zz1h7B)_6z03NQ5Siudk(5mtNerKys~qTA>&lC<{;x+TWvONX8aD=82AEa{$-DCTbV|fUndLK zj)A8qUXUf1=EJd?qh!j`CHSZuM3;qbaAnLv@{kRM>5muF+86fJY|tY5#6lZhPn^Ie z&OFQh_7TY9v%kp9_s&r2s7uuD8cVai8T3d0cCzDgEYX^x3Txv9*}#lFf=pA3i%U+> zs?1Uvy;_N`-4~7HJ{I8UO@_S1E}0xsZxR~4o$1-{31F~gK0R`uV2aHmnipXX+4EC) z+0S&C6_^6yvO*}FwTvbFw-2pmcHql{cgV)HRO&zF3~iC}zue9}71wmQi5`DK$=O5d zbZ%B5tP`T4akDC|Ew6-_3xjdZ-KjR(qryb3qnlVyh4bWv3MWr9^00KG17Dl=3-^x; zrddmzsrxS-I^cZ;lj%p2nZ}RgpFhusYO{TU&!w)IaxsTghMU93%NK<8vnx@%E*~Ft zZ^r`%E!nxg0_z^yC|0zE^TA4r{GClGR8JbiH!NMqts*A#$j3>1`z%{tx>B8Owbqsm z9-Kn_V$ZXkM?b+%_ck)2+YXjD*9yese7rf$9G6|1E|e&IVhdM15+)5f0~xmQRruCLS@zGZl8YZy^|GD!vmD}iA;!FCMrDDIBd8l=@2Vebe zACGv~m#->Wz|GD5_#}-y?%d6ZNBtUyn;jE)>*DkL!uue;;%1THVieETtr<;%nlo_R zwMo2tt;h?kFUtQEda#hRX!grn%A;E7hDqTcr8-hB&KllIE{&7&A!3f9%@;YFSRIHT z2Cb1f9W-H|JDY9(Sl$*Q{@xZA_qOH3(#-)PRwXr zi_iOx=8pCY(eG3rOey&%$7qqnC04OJzMpa7kk2gHL=*KVePwYQj>Bv{$y>MbG``S3 zj{^@~!;HcJUS+QVO?GdkZ!7|DBsxNkgE9V2dxgCRu40$Iox+y~#-jJaVX(?L50itk zaLv=*>_F>cU=|r@cyS`u)R#h8xih_KeH~YBcb53Yns9wh8JlxOsKsrCl7Wk`?VX(z_y!B5VwWM6B zKZ-k`@v#Y4iqs$*&0k=S!9Ss3#dJ{oas=wnz9#OSE^?dVi)ikVfzj*og_nkA;7*5t zOYT5=??E6SNn+ttD|8s&j*4%kdq^jhbsyhK-g|YhYS9TEnrjo!Tz5LELYdZT8qm!Vfpk=- z6Seg{2Bo>4;*$dzsQdi6_;|K7%N$S$Ch8F)8>2&4SazkVH_U0Q&kvXqF%DBCHpGGW z-)#4{5S+@dikFRtNPMb!ki*Pj(e-D-_JREodZc5Ni5WGRDuUD5+t3`O#hPDy5v)cz zkc4}xB=YM$a`M+r7&_Aw4F;s+jrrZ^jg=|T$NM+VxSoX}BP_w&aU;30ppg;V3VgAu zH}7*J5#2T%hEj!SIC=dU_`G;1jQ!b{D1KT^)^2gO>F$t0<{l`46)UFEw71RRUVMY( zH1r|&azku372JoIrgehljQ*GreMK&B^&(qqO`z*lcktPBh6n@w$oSK}@I|6O)9I6d ziry9Y`tDna@p>2%-3{QPv80jkQIq&4`Q%fYJtSQk0jJ`W(Qcs;7^oVN;)WH1|4v01 zcyk+AcDn;B8`jawGmUBQ0UJPJUz~WaT#;R9Jx=cFsDmTZK0}d$%#XH@* z5QoR%BxZd-o;A*hFI{GWgOi41$t;10wjW4;5<>H(9`@2PCpxNN0Oa49hnK7N;+D!Y zxaYk-2F1&yoDd(dm9n6Gru<@hUH8y^PExkfdIhRk6%DVw`$MVZW$!+Hs*TIx;eyMz zYPrSs?r zJ=aO>yw?w*VR9C4-FlXfGM&puZ3q+#)gNH_-^ILO;b9CIeu2fUzK=T-mxpYjxIG^2wU%RfhbEn3 zhT!<5Nemd2#@(hC@T5@>`J!jd5IuJ=c=cZZEp-m)&x+aLZBM{buP2}X?z?P8Rs+fG zwE<4ON+sVHDN%M=ozK~}pO>eO=KB}+=e913`0P$2{`BBw@x;>trn-AI*_tyGdkm&* zu3j6m5yPnev|r-fShB4QFrT8HC4sMRr;B7;O@@o>8 zBv8eW|61}Dy0>_P_lBurRj@Da`{FCu7OW5sC+O1%KPeyU?0b@X(vYae)_|yX5G}4; z6a>jjvGJHbY-m^_1YKFi6>oX-QFjh9!<9?f?WSXF(5z~hdrtx0O;dr!NpURvzHYm79;7z;$n5z(Kv^&`R=2wHXY?5=%4D z^K}Z@@#+JV^*Bzh_i_gPrwNkZ`#fnpaf=O)e1qS7BCsZZC@we}&y>@53%*~spiam| za$fO@Fv+T3UM#VXi`q|sLgQ5Nc>XMWu;rtWz0{7(-ZTVuuXAT_CiP&W)%S||zUk~+ z`*K#E7=^XFT;*3v&19uB9^&UVCrJ17q7o64ZVayyMTm>`g9P)e(E+N|;twE=0yhL$`k^U^9M~&@lN6&TPxUKl3&5`Irh8<8h00T{sk# zoQm-D??ZgrgOkj|<{E3faRcQ~4zmS&?#bRr%!@lWW5w5s7lkt$Vq_{3Lp6BN7?ADJ z0EbRz^bH$;@e{5I-)zFfzBiRIw|fbDshlX*mJGseGowhWhY#6SX9*1*MyRoIB9Z+H zCq>d5-2fBBwyMv?jeXL{s~OMOj*=D@;dYRO94Z&jDrCwRM>V3y`%0PD`*_sd5QAry zwP3?mV^qIiCT<9ModFaAhrI$cII7V^+Vjq=|C0XpXGDt6Kw}74ZE1>@+2Ob znZPxDPm4qA`Vy6Pg6<=K30lYZG3$^F)Y6;Eeh)cBc0>%r;aQzJCZiLDwyZH0)2hTWk!$Yq2>2&#KJm6R|b^#+x0DgdTwlh;a1();rxSeG3FZc+Nen`O59pSmHW_V z=R`1A6+zrTs}kQGK|*WLQp%n8Lc8{UJaD~KSU;MAJJ&CHpHMFRQJqabZ&DHtKa*kW zp&v|sYO<``%QYl)xCN>H5MyKO8A&!vo@9@_Ys?|`Cb8YJl6>6Tji`3-0Yx8^iTj*R zcFL_6EiwE7&s(>Wq!s-!Ao~>AQs!7v8r49p?zTd3d_Rc^l8Qc-892hTO2|@cW=ZV{ zETii>S^ovQ!0vS;5xkViv90Q4f2ko++}2NYSfqm`X+O|FaQcSq=R<28;pDGj=E+P`qm#J@ezN7Pm70; z`rC0f-~RKIIJ$!{;QMD;*`^yM0lQ_8dSD`B+f>2iT^}LrNk7>G+e-3pq6!JJNEDiU z7YSPbjbTQwBZ~K`lwo1yak6q$6L@4jW<2N~t~8To;sIO5=Khm0-0l*}w{w*5wxnlk zQ^4LfS-2W{QScZPQZjp*5!H7OU{NcVvtI>vd{WK;z9DrPAG0%w_}%gyeug{~IL zzwH-n2WnD>5%%~qHU}aEW%T#x&ad?e;ol@~;7+4^_;BYt;peG`tm26_o;&FxBuvPL zOI@VQ5L3kqdF%~$t7FQ)?BiEnF(-vOf z@1nx_=j*pocWVK(4eS&bjK0C!OYidW>r^hNKMmz|4l3AO_8g|xMdAXN#rP_u4@_AU z4?B`H;FFpY9Lb+5R@^PY!_5t3{iH)AcFh7=S3L{i&K+GCHhhn;CToT0zQ+wb;|4+e zqIYERut>;!my1>p%2i;-eNlZYqoC#=b}RkcV!3?pb55av&6( z?Tz5~$6ov!ThDjSF6UVqSNXj;zqn=pm)!98L2hQSMhh{aO!xKTnyz>^KBdZ!LmHMXB&!?hE%FhEn&H zbExr%K62-maQyn3mxVzj)MJ-H!>`elXHUlWNR~72Bhd<;Rp2(RtT+iko%G(RhKR*uQ0u zUnhX|amk-5d}l#72E%W4$z#_H+_-%|-g9QyXTwurp+&p(@XxP=ZYRE?GHb>y*Awtw zD_cS66DJ!$#wZ~RWoVcy$B@@Yu5 zi0r@$-a7m!uFjSVE3PB6+Hi_w)TNMJN7Px&$2c%ByMi;u{YLA%!T4H=nKHWhUM?QI zFKapx262DR%I?&=mUN#om~QayO8Yk{(#hqjsD!)7PmKL|O=;SNo|o{F7Xmr2drDsc9?#U~Y<<5NsG^GEusJbi}=o4i~d z)UPGL-wYK-P9?#^?w`O<@;|HzHiCh9(egvWUhG5Xc0}@; zbH;p7pcTFk%_iDjA>xBsr&)9BIQqns0r!rAS)Ec%y77o?v|=jh`{fDbHTR&WUPgk; zlm60w9t#$C4~ldCb%=42Uo+}I4?JM52D13MLVAQHD}4?uzy_vxBSXSE9k47TEnhR@-H&${p?i3g#tF6}U)wE0(? z82+hs1(%Ik!0H+Y|*7js8`;V58QGeZhe`~0%!|4!(U*wRtw5^@Gu%W4i3Xn`}=a2FHwxb1^*9v> zavn!oJ|6+cS(8~*U>BZXI-bMQJf;BI*e%`+cfNI@!D^}S=-n|0TIL64E2hF>|9Cce zV>)YZU5>YWm9ahfy!>IB4tDw1gg3_b;;)|W;t8KLc|_huY`c92*MITD_|l`OH1Zi{ z|5D%+-&K-7x7QNw8aJ$d(HGo<7YSQ@!o{t})G%sgFFfegM2O}R?2_}Coh(AK@S-6P z9k3CvT<#!Sr+#1oouc^TotZH8YC6Aj!JgkP9KkPThjD&pFP~>Uogdm2#oMNtaKnCy z@)I$K1yMGP-~XV<2focDi6N%!zaypOwTY|5yRE|7C5C+Y_+upBW&!)UW;>Ya+=Q+z zQ{n5^E+xwEJNZ(_BwkrzbYcF>bEs|I3Gbx+$#Ti782ofJKliW@&qTjPl@2Yu_g#bk z>QF=_k1y!&vw}DO=)%LhBng*}o)ouczaaBp*x`f4>S*7b$qOa)~9I!@_;xRARBYjy;DzUaaK@Oo<@lmm|!@lpN=eD+-r?hWk^ za-SQ*j$vM)=@Km#cN((&fVp8tL~IIbdFP5&g~|6m^>IaiC)uR*jv*xLuii!t#0O@*K&_E(eJ6 zz$Gv(scb;@s|*oKz?9y4|Lz?Y@P? zy|bWQkF>GCm{PLw<8bl6{%_gplS;gM>R5iy;2ZprvT|aU1yY??P0aJtY?7EhldW!V z#vx9ZSTxTC?YJWXY3YKI_7R{scNpzgeVd$qr3?eQYy|uDW#HRig(%;d1q<>D*vr46 zXeCahtyW>coeS{esb{1?%bwKPloM?wLr6~hfg?4;ab3n+;gDPr9fuT>zrR#PizA_U z>eOR4cyd?#cEp+FN!n)rq|vM#TfnydI?hrb1YLA`H{} zWv$s$AbqPT%uMLT-%r=W{x&L@-Kz_K;5(52mY!1WoLg+%H!nOhdnc@aPz=ky+hWAw zZp^sO2#Ot5$)pQ1-YqyrZv5>yb{t&6*KW4J3fwG*(mYG1PAtV zNG1;H?aRmJKVzf)5TBR5W4&8X;N+`c*!y**tof%cS029!Hd|Zr-8w6A>do2GnXJSk z24!SlPH+CT<|Lm~Z!79AlI9`Ll;C^L1@@$I18&Qz!}n9&_{@GE*sLa7Sy5EJpm}~D zXpFT)`TSrM<}2}WO)A{!S}vMQ3&Nn@oou?+Z~ReU#&-@I%yoCaEmoU)LHu+k8tq7$ z=s7c(cubg!tK6P518-}XsNIFTiXND)IRx*|n#>jCsi>bDA}YmSJ;WZXgba8q6xRL-uh*v}rXk0EYzEZMR?KiFl>S-ftPz>4Ml6Qdn=K>BHZwY{_4+ zi^~(%3Fb{H8s)`3XyvV6VaaQKt4#;2qj*~PqxfP1?c}M5_uL}Ko&0(EEBNjc=JNf_ zz4@ZfDg3=dA$;FoalAtUZx&xxhte;jObcovxdp3UaC^Ti@YQ}=@)zEA&mR?d zT7MIGJtjDQ;VXHz^D{1d;(x4s&36^(n7h4y@wQFN<1ciG=ij$o$gh3%g#WYjF~1|n zn13`_ka0Zk3D@b`Sb^u!K3>GYN?!C>V_uExM83H}KX1#Aqr8A$HvH5#BRpRPVcxF` zXSg#3m|w&qd$_8)+<*8%kpwxHuF@(C-8U4UuH()p?vqd>-k>w zf;6bc4_x7WW`gyY&2#C^;N9LBX(}fRGvwdsd=p6YANhTiCyJ^9-~ zvXf-^&fO+F3zaY2*%ZwC*)+&|A|%fr?>bEJW0r8c_@&%Q7pIvWbRB1AdR(n;Rn2-c zIBq#ow~;D_v*xgyB0i7{|fH%|Bm8`wGG@YR)TEXol5*rSzrF~*ng&B{36O;`Hs7C zbUI)4i9UbSe=}dI@)~}=I854iG`J_{Z>}%8x|NQ|YCz!I3tT(36yDXvQrbq3<>079nY1N%qrf=%sa%Z~jYL3C`+cSEaav`ae0vb@Pv2IamVl3W?Z-PB-Lj0v`)r`Fl{LH<(>@7uey;E; z-ZapqHhF_6U|2bsK_C|54uas~IqUPp|%8ioHISK|LE zR_9x9dckGK1iJDE+V!ut)ziiUn07+>V?&4B=?hKW|-K zXS-ZnB9Xh^T zf!mce*z|ypq&$K4c&m__sy+MRKf-Je4lo&IE(_Q?6V|vjLiCb2_*h5a=-&wvhsNUK zxBAElQNxKze)#y$ajZ^ELOD@0CNlF8s}!0GtFFA|%=R5%4iV?z?-2CgoSQ<0iiXq+KK6N)F$i{P2~rzsep-HF8s7B$dneI8D~ zsDT5gT;a;h`)tkSsd&_)7H@bd&`I0HNp?Ag!xM!n-+j@BL zvwSf=@h4h7DFw=U7vRbH* zm!Dxd>ThA%87EY%Uymfx0dsb_V4p@fdWz1+cR^aHw2+HG{x-ppj9sv&LlcdsEyMp- zMc|Ed8Tdmjiu3(wKWBkbF1xSZh;9zI(e0-po&A!)xstFKd0L62dwQ6Y7drn3f)2)b!r0)4sA~k`{Ocw$Vxoyl~f-MVK{(yOk?g1P5MUZ&VCnE9R@;~jjh;VY z%-0SuQb>o;-?QQIlWKPQQUw%9_QUAdWY|2{79>*kz}jEiA%BS^6H2oe{GTqe#Jas~ zySg?G$SuOH9+!an53()L%{d?%W73vY$cm+=aJJcR;57Ex!OmZQ*v^EVtTs>x=N`y{ zc^iCL?T{u+tm$My8gZObF>`cj{{SKRTA=mi6V&|Ggf#(?97o*+p!mI$ov15j%~p9Z zPoNnqd3+H1)e$V-KVdcwp=?%eBZU0R0tcQqvp$yy_b-b<^sSA|ZJ8s~UwOe^=`F{l z_x8ZDoKp72ZXbm2F~#kUJ@6>b46H445T~!lTA;b%hN}imQcP z^Gjg<(+ZfrekEsY1K7Lf77g=jy%6%Ol({ zA2qRb=59FbJD>S&NI}l&m8ki1HQp|q4w9GM*!(?`%=;wKGt)7&_sb)6$mv93-vxM6 zV-MVRNFkM|OuChKhAxDS1>FNhtmHr)3@iA8F3$iIL&Lxut6}w5S$r81f+96L!GBf^ zh*jHQ&TuDOowOIK8>--JB_C{dYyp9!8ok{opmOSC;B-&NCaD{^*54BMn+3zlq*drU zY>Yo18sga{G; zL#awK>V65p#SW2ZKN1K1jq0H584vrU4RJ-+M#ziw;GB-CLs|7z7K8TD$Mw zIQOO)-4M~B{{|=1h_M;o5{|`t3yU#`CyVngY2dvqN8BVKh1~KcaCY|wZ9yJJ{ZBJI zGHizLQ{O|-%SsUQbAUA!2XKl(H|*bPjOlBw=y9I~t-Y#EtF}DH_kD}7{zVXmMq%f_0@(g94ri}Cj1T(-;h1}7Q1%GXCiyvZVlx-Vi4|e9r`eAddFd`s}f#$)CEanOHj3G0V`a*2HrPl;&7ggU~S9b9w|o@d0xc^k7cto z<>zckPbw?%a)ne|E>v^fS%#$_XX&H|oO`}WC@!9b3%8YF)1q{=8zV$^y@Je)O)*gT zq8;=WcK{9sb6!`xWR*cpaKhp%=f;%@?B*R;_RZHEOLMp2F?oNy$Ik-gW7A=4R}I@* zIfg1{SWx669~Od^ud#tlaX`B z63=8>)3u)msKhOv-q)?g{{~LrbOCUCe6M0^$02L5+9mG9K%bpj#g`X?2Ju-8%Xc zSC@p++aphKRN(++@t)8t|9BjjW`)VZJvdcm6jLs}#tlVpvG(9)6waT3jXl~pzD^9Z zi=W{|&wRnYwVFn3ui~K4MSS9L1e5pXKu*a#Jhgv3E`246GUgH3r4`O;*x<;yAG?6b zK79{2!gC<&LK1eZ_h4H#E5pQ#;^@2}m<8?l!w%S*;Nw*h_f-pOzFj*Z#a5A z9^L*%VPM^Hw0I(oBA&-k#YNQg{Hl+1ZA<|re7B)#_r=J-9UOAno$y>~+gv&)MB~y0SAd1s9{VZ^zN=V<}3EAGbK-(0%F;Ms+e7Qn! z*TWoFb_lW-a%8C;qY%DEVqQ$7pw}^`cH=VCl$nS<`QGGqb2{0~PsD=)O!~2IE?Q2% zj8?mKiF-=GA3C(1V&<=g$G7e=y;;%Vo@}S@0NoxLX;&%v^}4hu34J{b!i64zcOJ4F=woz)NLw@$DRQj1h{YYo39$)gqn> z)LSrYP>i1IX;RoP!SDCJM#gTJX#Eid`tvP-%H_|HoopRueSU`MCrz{a*WtQ5oABM~ zljs`WiIp3E;H4)`IEz1(vaZ@uzrucsP3VXElR>C4(1;h-tfQ<)P9(Bh@LAV9jGyd@ z6RS33>sfttdhd&xw^ML=aVIf~y$g{`6CP4xZN892;Lms9UzebZpF*?;;hZYeX7&*TeBUWF4 z)G~cMKdT5Wj$eiK=e$w-^%G?4Gz1CCb4e#mi~25$qRsQ~aCZAm7+(Az`zELx>no5*!#p9dyMw~7UXnZFO_o^SmB~lTT@h_Z~x9z2R<3A8GMG1F} zZpO>Y06#YvqDhPl&KzvRQ8!7tBbfQRG)<;+$$tbr?;u8+>*A&RA3$uTIA%Gm#QY3L zj6GzG8qQU?;qxgJAGZ;0WisHvnnw6fV+!Uy_=?KXA2F9Zh|9-T;g?@UxaNEUUfYw3 zc8?r!;O}bz z6Gz5=3G^e|pJW%l02w(WoV9d4_MM5tSJTw+&4EIQ|L+ngtgeBH!|@O(EkT?y#Tb1p zi$dNlrVB0F)WKDvSDHJ>+Cr3$=7y8EV>PXb$fX|ZND{ZUq{%9CDM~_#=G)(fqE90D zb?sD4Q9O-WDGYm#^~1W%2c+UrPa<2gY0T8+G%jU6^)}1W?!S7dQkjEoo&vw+Lw})Y z$O_^$^zp1uB=$?J$4Grota*AKx*yzyKSwq2Y0wzVXmNzuPpwhUTb5E~CR2`<7F9e9 z$IFR^_JzmV1YomY2>vR z@3}8W@7ARhrD#T5>UF5+d@B7Da11{wCy{&NLX0i^43>s^csjtJYCNMTAU~F#1oLpl zw<0VapNX1+XR*C>8paE4!wYSBBwBcs3L-;@>ye8htp@nd=rab&hT+@O+b}{p4;2<) zgx4Vt!EB)_W*qv3H^!*6RLOIGyrQ^Ax6f{_O4s~{3N4dE!SltzjyPe}v za&89pKA48e`OEOUy%32X9>UuNWtjR?3Ebxx;n&YS_}<)-n8R$!nC?#89g?K?OM<>+ z{l&HgofLfg89f$jqh$}3@VbULZdSVwiyxMu+sFI3szieV&V2_p^KO{)Rv6n4FQx6} zQFP}YpVDWVl2$-4Ir!(3bJ!{R`MH9|hdd`EWg}CWfeEJf9zUn5L_?BYYE7+OE2%=& zf(q1Dkk_ag$@nW{PDdDSEF4RcdncHRtByBadU}L<`y6QU(k+y(yq~~IoSclmVMeDA zHCdg(EgMYnx1t_ug#JKB`F5Q6TcEy^QpT4_s))feuxk7$JWGfMgSU;~K0%bC4{OnW zM^|cpbO)VQ$xwfj8a4G!LZjzac&}wOil3ZF9__mHWrHsL?TMiD1^Xy}uM5pgHOKIm z?P#mDip~wi(=q#W`fG5M{MutFB4Zb|r7xu^)^4;xZ!PI#AStXcL!NIG#$3LE-1}1? zP|p^uE+s?gjr;Ug{~_(GZ>P6TveauZnZEFSnha3GY z9!F2@W>EJHC5jwg3uz~WQR{RL1_k<)w|FoWjCCbT7fG5reJ%xehtRUSnV51?k?Mb~ zBmel@@FUk1gSPeKKi6{fyFQgN4eaQCdIZ~b?-6IR*iI9>+6yd_-DKsCBJj`LjJfjH zutj1q?7qw*Ec0NvWt=Qt$sfZuPQStVFgA_jn0Otx{Jw*yQ=4#Ov?`5Sji-iHooF<2 z3O$;a;zoP{=i+XnRaYa93vEE<L78{og;EjZV$49)g^#$$(KQGbCxzAQcq zW%tX-aHND1+G0s*{S)lJU5ffY_Myn0^CVQ0LIsO`DNa-tRpQ2>)Eh0dn%07iRmlP$ zqot_uVH%Drk;1u;+acoYQ+#rF1iPJ@QK3kc7J2vJl-d~dDxOX^szu3oREBC2>uAZJ zWRh63l~is@ki&~XoOU@&&?CrDe#;|#zO@ij?EnXYXQ6cb4Tv~=o8DT~QvB33GMVo} zB`XZ6zgLvB6r5?^JO{dSc_~dd(n8(y(wOu609b7hv_C!H;Eni8ca@5lQ4OIp8L0+2~DHUtdnxt1aEQ1)DpS}<7iTeweA0?NH>L}870XdV z-59T3Z3Pi~0q5Y61-ps3jhOI-d;js%Wvuay8eucZH7i%CKvL;}24 zXjBY;jhT**wyWUQ>!(2IZUU5k$-oH;^DsPK9Nneg!=In0;DELjWO_8QYcj2zT!W3A z|8kY7ymA!fXZ^%0r+DbxR)EL!OYvItVl1iq0DF`w;rw$2_;l3<5@%^a{nquA<6}h* zeKROiwt`f$vq`!#hIYv@eD>-U#O^o-w|!%2z}Jo1^d^#vzBKFY8sgkwUpNQ8P9>g6 zEq;!efJy(pvX<~GoTKKRoDRW!AaJoKdvN0vT+8z(F~Od6|G!C;eBwMR`bxpAGiMolcJOknfnHz zbHOGuDV$HY5<&qh;@*_Zol_o{^?xpfo$QgWvbrxvX06yP$e%puKA4#$4=g{314pf4Qh zznZH!rV!Uh~#3o)0n@D$Z_jEjLGT7z2_ccdn=bd`x%nY{y22LDrwp#KGBrh zIL@?ui3U`zJZ=1>wwx2!l|qZGa|oUtq1$bW^!8OB&hw8(g-~<)^Vo+hC+#Em&%<;t zyqjMByhwi=ZsEz}<4|re@f#wK9uw^CRi>;)=)_OO1W0h#k9JaNYwoyF~Gk(vq3Xq`@_lZJ7c zjtKGf8!_SYY?^8%PyQQ!Ve93~sB4~#pOh@oY<3KG?Mg&H-ZrdA0ThdwgR%~@(RU%E zyARbUb)hsRm3Lw3a4FhN_C+tQ8Sap71_#AZc(}q8LT-nGRip=eP?SWj(+8;Wx&gn# zRzuP<4KSRunOV`JDc7;;=M52s~9Ub3{^CLe8U2pm$K4-8%OMZn`hTLlzU@hL{p$ z{!wI(KB;i&fdkr0o+vWiijbQIJ z3pq81kF$1{HWQ(IA;EdQ)26`i-N)oL-4SOgY{z;AkU@> zJ%b;Bdy6Knl8wTBPjgLX4peYz=NYh9p^7vvUy01*O0Xzp3OGwHf$}vAV0maHigruV zTzO$qnI?qKdp+@Mas*!eNz|q8kLMJ#SgUI#A=-}x)h`m-4KRquqqrBhMnj5l(Bdc(ficfq{P0;4rM z*|J-?cxk6KC9W-j498AT%dy2-YBP-$LZsL#&kArUmlODYW`lgyOicRzm-E@-G0V}d zhv1rr>`bFHo?h;Y1_zJ8?&=S)v(E;<-sQl9o)NfFv;s3LwdwF)S7Ki4s91p0iMmb3 zwS`WozjT0IGYesoI#K{1#?d_~+(AW`AlfJeh@fB>OD}?c~so@>Nv!^EuluS<2k6 zoM!R7F#?R>6q5N}bba%daXy@Z3NL#s$)87`_H3q4`?RTG_yx|h7p%WLA#(XL3A4Q; zaJKSwR1m(7jVCneXbsXDhds=P`;@~qQwNKLI_O)m0Y|t|xO~F}Q2I|5HJUx}YnvwC zG&lvW%M74VZ#FZ#x)R=atj7mm6i9Na7&+a0hlf*dW9e=MdU#Bk3{T7B@QE@ws&xp) z2Ry|+D{An{N^8t~Fq6&Newn#GxDTS$lj(%{So-F76^%mr*}v&-%vbFfXW_c@8~aCy;YnGWu;7#c@kzXqLf0|}7E z+;NVnhmJ|ifUt@BDh0e{S_KckYz576E7+n}S6E(nAIvq$<-GV5&e2q}WHV=1;)2TS z7`?L*w`d&4qE}w%F6fzEi^E}#08iW?bqKt~GO)1kgdo$@7HwX{;ORQRO?Q5Ptc^Zt zwb%)+|7sFHXF_K^waCmzoc>KRqRF|6baD411kWWjX}%BjPFX_+x|NvxpbzM{G1LpN!EJBef_c{o6F)(KkJL6R=5%NrCG{%Px_KY)wag$EeVt7=waw_m zzIi0)r$o&H->~D;>#;RX;N!At0N<}E1*dstu)NNZO@6bC(t6gA_v0z#w1o@V6GcJ& z$vn1GCJbtR)N|%%u zg7Dc%U%aKKL|t#Dklv*qND2as0_CiGkB9lg z_F?*Ky9DpSL(A$B(eApnRj? zX&NT=wQ0CEas`e?NMW^E5p@6eoyl2iLv3CWoBBKh{_@L#k2k<)i4j&jS%Pox{D<)> z#aQ}d9qzvV4X?GUQT~1;(RD$rys?gHJf04{`=l}V(gd{2QNeg6cRW7rBIfTML|fA; zoGUL)cTy(N1lO_H^mi=wxxWFmOGoLEL>U!#WRr2^Vlwayq7U1*(B(6S!8IVB9Wb|M zMqe7BQ(puR&UV8&%DeE^uR|C? z>{)eimV1J4WnSUws{80GH%|qo z& zUfqK#r>&9i9)({oC$jy*>QHeu1;iwiQE6W#zB6k@oscRMiy%4nvFjBZSJuLrCKky8 z4s}D0coh>7oe4@cNuYdoIThY4qanvZDzD1MV)q=>JSs%)lh;w#+$M_Ck}_4c_<=L` zX;7!yJd*6{z}4QOf;wa>iSG(yer32i_M<$I8>4wjt`MWmmDR#jr@5*42RWP28Z^XaSRvftN6A9otQeCY=(!%UEDk;l=D zwHPGxf-%WKxN-bADh({dzdOtEtlSLhJfcS$Yd4Ty!+z2pu)rUu({T&82Sxhpv4J;+ z><57E7bY^vlwpo;=31tlb_N#&p2QBTW0;`V#OeZc!6{4$%O07~!P}cD^7tNlJl+A` ztgZmP{j*W_zX8m$ksx0s6>2)^hvm*0c)z|CUlj^r@Hb^l^0CA4-V$1>P(v2Ye45bu zR%?xO;f%SoC|`qp-*JWWTmCrn`neD1jPJtRLkeVQI~MSDg~`Zt zRn9YhKCb>5gO4=QvCU}->Ugcf{fV3K>5p5u=Ji>8;o^tRf=BD^kTT}HZzfBarAB!! z#w7Y!mjbkX==?KZy6n1DaIa6`hO7`wze;E`f3Co`u6-cC+m?JYUO>UBwQOwP6(WImE%$XS9b48)uRZ_a({aNb zL7)Xi><*=-sWFsoHG_ihY0&N;(sWN@AKFS~;Y96=7`?Iuzt|+<6s@}u?`}b@OIJ{S zvL(UBc@(;Km!OvopgIA*RjHVQ%85L*7|zA3!&jl?eIv)ksX#DWiNi(iGf;CbAEK*% zuoYwSI4&T$p|aZME;ATXY3Z9P}l-_gks_&Pwv$KDnZ~$llEE$s zvTk2Q)8@>hE>fp06VkCnkVCVn>^E|)d@!s$1CNy-!_*R>FB2Ejnx{koXNzfn5TAC; zyGEWZ$8cHZWbBHUV{l_EZJvD(Me|EAa{Vbh%FV+!GU>Q|;30lPRXnfTh}r=Uj^C#8fTyp=eFOIGw^?}8dM;LgKG z6`oie7md|Zf%us7e&(4Da3X;sp=ThhATGIam$@qqGfzE zsU!-MZLt7%x>8D-0t{6mbPx+ii^-@okxnM;r6$3#obfIQ;LYZ2c9}>)!3j9#Susqz zah9g$M-q$RklmX?berTzJI&LnS$ZFX#x|2j>v)_Y6Aja~CF-V;1)r&F(FFRE>p-Cc z@ua6d2aR@CL-~>?uo;u7{Lo(NC^I1UNu9VQScNtYixW?5J6^q0f^EX$WZZEOcjRP3 zS(+gzPHV&~*2*-z-i#WrIMUu-M`)343_Vqk#!O*pN^?If_^vpOydK3>C%)pJ5(d|% zaZRQVv>TQ`{jizHD;>pm#m6|?CtM_9-bd>7y+W3SN~U8c z9j3Sfg2^#=@zAnNI(cJ|#*GWXB=@m&aOV`-w%P@KqQkJ+@D=JF+CkwKG1PUu2A3qC z!Tr-FQOknmw1WruCs>;(FoG`GUBcy+Z8$cz1RWLk;1$vFD0@Pbt=OG|-QzD{+pR^| zsicHPRf|AuNt$t)1Rr;uUydflQYJOO&O&b2Sh)JlhUM9Q$K8S%L64RnmOeWMTdcJ> zHFEb^U#LGAd|ZNq^Gfhl)L(p1rb)fkaunP;i~2a4B!8+BE5doOT=G2HY?(qfqhC=Y zTbm-1^yu}Cn^^r>a1H;S!|zqi>}jq6h^JJ-8(UkL(PN2gUx#3Z+$5UH*-UbkM`-wq zE+na%p?1P#>O7EwLAl3pwaz9eDAq^kqAtj(PJ_JTBDnQ*IvZ+~$Ab0<{4btijFbZS z3;58Jl%`NBXD8MBKEpRd_n~${Ka*I02BV4#AU%I3d_NY6HW#u{snj2znhj&u_9$F$ zTLy8jF2i*>KFo7G32JM<;PmVg{4Tx?75qxjSS%28^btQi%!5ZBSDAH79VdFe89w@D zjDgeT@!s8LwB(0jzJM>DwLO&{EIUl|ZZ9Oo#>E)0T^6&WUV>7wA&th4qvE=ocu1%P zE~RCI_*X9&D|-steAnTD$CL5?xna=Y?SQ6KO%N@+1onzCu;Y>wREt@V zT5iJkYzA4&?l7Oy*%Y$m4=z4;4%2l1p|S97l($dE^G&C*%{CP6U(Uv7UqZosodom` z9bqfGt?BS5eX{3Hr1`#|umR8D`etWz+qjVyUvi?A;%msh@iMcFZ{c`^U*b#@lEQ-X zwJ=4>0t^&|V3`M>c?7!%V%`QhUuzmo z*H1^`*1lATaH_zfsh&7p$^55YhJ6tpr2m&_fw&S4SBxcAd=oh}1sHCX zhtFgKF{xn^iY>RNu@^09;fdMw{AB|s3v!o~jwNB>hn=|DSR1z~cSCREPLdt8qS|l; z8vJ{LF1(8(=V|N7e3BgKteeF)jNCH`2wzWfk`A?Gi&t$Gaf< zw+&9WCV*7wB2avpjhiQ)#1W??Jk%^om6_ME#FdX#I*TyuIAQU80bZ&rO^}@WGwL{DZ!V$2cz0B+RK$d#dTf`q$7J8Xuy#!o{>tXzEZ10EYxWbK zD7*so9$(-*4aaxWxUfNc7V!NQsiL_NFLxioK&h8dETxaH$9bXMs4f&N4`YKZv7EF( z9aNKb0omZ~?BViQP|zI@bEDN*`e8$GC~M<{J6tl3wY~=u1_J*3lv?;=u7dvGo`Qi& z1RUyE2iKlx!Go{~pzZh_GfG^sEKn7{{TGR6^#xdnnJ98CYp}%c5K7$(#$MxdnCpH4 zIBJrR`mF#RZmz*?fj6LVN=#M8vll&8?A8Q!XF%& zC_~N0ztARKp57@6W?ZYEqG#tdcy6sG;5-&`BrR?+rBl`%bK?c9r^*rWy)m9`c?d@S zd+~F+C+@RZfRg9CP&;fn`u0lWrdb8-RKi1cbI&c7sV9U!aza=XA&h~z0k=CH#7kk` z==8@BMQgu;y5|uH(BFp3_Gx18!fe?3O&e9qHSyXB9=Jyc`Vyaq;4yb6y#5vi`@9qI z)$K7fvgZ#Lx}HGO!|JGh)({ND1#@}l1iZ364BDcy@tKDfBh3flWTB`+F(c%Uh6W{ziJ!rAKe> zCZc-#ELwBKgZ7)cky2ULjCp~Q2JXE=WSYn zBX>vOy4N%O8aswc5Bb5HS>v%t#~0__YiAP^rXb5M##nw8`qb8-_Ka$*bx=TG$8C5e z@HiG;I|my^IB0u;p-lQbTxk-IcHR(Bn}A|CZN7k@rp5?G%ok{BIc35P)WF87L~h7pKy85Wz^hYiBG z|I0o+Ie$DhP5KYcU-=JwpVqOiRs9^Xlf#^z`X2O>7d)q~FTyMVhfij}7b7zCF!!xG z9N4r6q*rplXJ`q0I&`0Pth~?4@9X1JIu6V9cudqahB8A|k?KNmYBF2{(XS#wT0IYB z_od-3o8xdS?<}l|>SHgSYeCMRJ|?2|7cPawLBNXfAnh6f@!nzJ_;)Of*%QKEr-ZZY zUy&@w#R)lncOYkHI7~EYgc&bZz@yQ#Y*MfxELvCz`8UR5#4kr2)M5e zOl}>MY|4p=It;>?-6_Mx|10P$1nvwbIx&{A&%dv$0$499VI${!iqP=c*q6tz||U< zP%sxP=lx`lwL4kioj9zUI0mchpTOfaSMm026U>Y`57q519D^mNIG;u$SjyMwv`g|M z8qKOe{Wlkx%E2a+sB5L1H`}k{k!5i>Z%9#q%T7YCx7|=7Xy_EBec<}2HVf{^;+$Wg zfM;So_6tbpAu_g)iJY2LvN03JtN|W3t(|q#VJsz$0w!`Kd55Oh$2R>HxBA=M^q&M#< zt99eT35}Wf!*;AG@2(eed?;R{tSvU$>54MwR2PJG1Co$PP+(K1h4b(rI}6 z2`X%{#%a-^u=P_4xN;oec9}h_JyZ#qf}ScdhEZdCFfBWmg+-4uu*zA$hj5sRsR6z? zu<2`lEPrQVnMa3vh^*a`KmoSN1*2ceRgBS?MwOk#aKmR9Ox(6%lMUj+D~+fx;O=C^08_M^ z%#K8A!`4GZEHJ*Cac9hgEe&Vk*^^|*OU(tYk{feM^?>!uMKJj7b$BV-2n)JmY z7+QFYCC4m;PeKh~FsTDi!fn*nufrAL-7F(`8)UEUfRqj^m_1(};>iHcShlhK>!nyw zVkYOQY6DxTRml##4hQI5h_(~v{trXv;g4k>M)9({*783XGI+v(KDos%;NlHsY zTZxj8ogKv+(UQ_s8lojFZ4FKB(%$pE{{Wxo^W2{MH?DKe_h3wA0IYg<6ZdXJOO#M1 zDSuTGTb3(}`N^L|pM=eFW^Nmlr*1+|n||n0ZH9uCKdEO0OXYrz&Z#Y0(u?^Obg}nb z(g@!}RMA5HY);d(sLM2VkCQO!TOsCYWYNhv&(JgYEzazHfnx;|MduMLVxB3HI3#O< z7snuY>^OXO@fFLp@Is2~yN52yd9RG4DrQMriXh?lBJOVDkdEg5~2iuUF>l{)%c!CE0oyA{IujE5&S9xhx8?N>q z!?m~5xVTj!hdLc(1rIG3t**V%Zt-!{Z0WETOtB{_r|liL;ZBq38Dl$h+f5>dl|&hjkJz z_0OPDnx;t9UW=I)`lQtTs~GU)5|x`Brmy`+!8@xKahbX7{xpFddiwFITwVN2QAe9o z+i}C^y0FTu7B|+Vky1#alyGN-&^(+B96sX7T(qPKyy+&CCxr33jIrEbXBfAxI)ZrBI!GR; zU_5s)PmG+(f98zgj~{l?^+CJDUsor|ci?hZ+3Le+ToZjaF+#$lqtyOE7+EaujTaRO zFkLfO&L_MPXLlw`Gk# zGKq}fUo*$?Pt#VM^t%%;Rj$FEdpqfgyv{cbdWK~;JMiz0|3Ud$9c;V)Mc*wA_!NB* zbyFj8f22RchOEWfdRfa>bfRggtLS>J3nm(Fz!K9&yr0{O^>k)q@3y~`J|~e3!eZ!k zxg~9j|0+%NJtnm(Un3M8TH&-klfv6HY3HaGXT6{CG_lT)G{=}o@w->D%ZxP~x_lX5 z7~Y!{V{*lptXPq;vm2RPZ4^;7TD<6$N>c)6(}B;QC}fQ(T`lV`4QZPpHa#<8U-c$@ zh}eLH1t-Y8Z7K|UtFWeJrTF1IP0q9{VVUy<-16FrP={g!CV6nAekUFpeGd_Narm2B zjWUPbs4hB(XT?8BuYH)9Ezcw0>F?xSb){TUYe&!i92FHU6QwhGU*VQ}Ti#bIP;KRk zH+O%MtN%XAS=oghTYPwSA5Xra=EN7%6ZvXc0$;K&L~%_Db?W6IUGJpDJ*H>iRZTMm z#n0uHvYwmsj@Tya4_13U!Y%g=c(C!eST%gIFrIZ;)O>M=^~E;0JN^mzU%F0RYLC%7 z-xBiY1Ufn)luUO{AoslX`5HM}gRl6J;e*tI(@S6gAxfDiIap&YIUvXOr9J9(aw4}umkOQrOl_zRM_d(HFVuH9aDTBP?>2eMcyi-&^OzuF7E(!P4+@yz9m-oZKl^Z z^>OR0HdaplMsN3MAfn|OozFi?tFH8<+Z`Rrb-c7ck>9OBL>o^PQcrk;bG8d8e{3Lp{ycUy;_ae_K7yr%}K-z4KZpV#~ zn=Ick4+qkqk%A`3XYk(bx4?E;3ViaaG3Veh6gHN@Wy>y6Sbs!X(D8@V`EDu}1tnql zcKPldxQ-ra_>gv6S8Cm-6|Vi$M#__R$gX)u%{%YWxe2GqzRw3zcpM^T{@WoM8wX=y zbT^#3sf87@yQAo(6N=-lkzp5v$yZ0=j=%n#8x=R?$Ur#r`0Zim|!B@FVl z=W~~GAs%0%3@JeNh38Xdf|hU#^~K~sJ2=lbka-m5vLEHHP?+o`?G96Lv384;eZ{9R zbm?7$K90fiPx{c^D5vbS(x)J|T%F^T&;xro6d zJTYoPCcQr@FmJsLo_x^2&&D)-8NH8wPS6+0U#D~R)zxhKIDy-#DzM6l7|6bUI_vls z{%)2WG24nGbBgGl`VV0eaZHTfc8@Gduh1afYI^mbBIVCnBur~H#m_bm#mBC8Qu5=y z(vX{-aORi|b=YxRm>5nVv)lv`_~f}*Kcz2K-<8C~OUuL!*_%)$ZNY|-?pW^Jn%P2u zRVOr{Sa~EX)o;R<`**2R#5e5oI)?flm9S{{h890k!vOhx7&cRrBRo$)WoH?5riLKU z_X&+`l}1gYbkJIUCJveTQdVR-|Gu5Tw>yvLUs4;6|8^Yh#%G~k&aW<-c~o?+KP~N8 z@5Bj*{~{;yJYMfmr8XVA)0CI8UOXZb8xJq0(PQm|!aVt`nf(soUe%a4sTV07$rU>O zKg7pfuf**qEy6Iuk%Id5huOpJ^ejr5`p15wp>xL4hN1x!TXv7~ey^i5TYHN}?+gmL zt40eaRMXI&M)XS8l{}43)4rv4Lhp4FUDy)?>6S57N8A((T>nW0iU-j4R3lRN?nj^d z!|2aCeSA%8K|%C+N(gD7Q(q?I$l6?}Yu%+yOBRW@*UiYV;;gu%xPlgsX+dqMI@{dQ zmv!+}$?sZ$`1HsW+k29=|d&;+hYl%H7jxd^f4^%Wq|9|8%TMg5$><6LO@mrZfMGeO8I7_hgV}#MIen? z8%z~<3~)AHm+gD~#K=HRZX4k#mL5=+RJLe~2EW0y^7R#3l<0#n$4Jt+u}k?#$tR@j5Kt{YK+HL#LZ1ZNJ9h)o_wS&QQp*fLtk2ykt?nz{*i8G1XGvB>2&I! z?1_kShDqIc?7Frc0j6JZYj+pU**BR_HfrPg;&ki?IE^+*vW~ZE3oNv@B6!&_?lU@? z$9jhF_~KlgSKJKcciXYaM3*8zkEWJ&3gk1k9ZU8)>~>3ox84SCJ3X9}M)lzw2SvKr z#ax_D+9!IP8H@P+|KZxS=U6a!8=h8M;Nr?5^zrw0nK9FjJsu9?sN9**-j##RWrEG6?eTv^)6}_=-*PXm@-XD7ORhtA`xSVcn@ASj zB&pquwcNiniN_@b@~eDp{y4M&n{C&EG7I2Pw-Nn)*TUP^8cjLw)GJ?2NUH^p^s?l- zFG^f}huQbFDR=yJ5#92(((&6b>CF*+_=+qt)wUPwhOL*LzwLzEwcBZqJj31oSC0(t z=Souzwn`nUL&$T*0x7okL$B)|6CszT%5%>#n6{@i_FP;^hIv|a$bc!;+k@r}RilnZ z_tCI@KfZlQz((0`XSi(;lp=L-+ItCVXIsb_FjF`V`;1*`1qeSGfC@#KH|ikgYld6H zPUk#^z4`(FG+85Ga+T)YbigXv6TDB;5%tf0(jWVLirBBieV17BpGESsz14<{h&QNkj(VP(~#leG$u9| zQ|4E~>zeGlANmpR&s0Ir^cdz($wgDjE>vZnLFc^`Qkncq*`zx_e?q`~?sSp?Qv!VDk1*3H` zvDs-j{VwrC|2d~IbWeZSJN7`zU^o1H_lifnlDS@gKk~<$y*NQ^EjQWymO+S zOLp%bF42eH3GY4D6cM`ywqYZX6O@LedlylnwFR~In{iqXvd@W+6zl=)EJ!+ua*CG*`EwRX`e_{gTR4`}qz$>`a12xTYL;N;>3v3n(o zf69D=&2d<;I~M=R#a?Oa_ri;7;2B`bKTr7a>UHyYig_W_HD!<8wpdhL{SUM4^5GJ+ z8gE9E)*N(mB7clqN?UK-rRnMMa$m+5v3JJ> zdH#@t53N^V%X&jLU9gblifEeo;V<3oJ{{Q)^o7}gTVnQwcskR&3JLXh;WyHhQ)GsW z>ijTR_fqBU_d=0Umxpfc>TpqYHCOh?W9?D9WdDJjZ}-{`o5p;MY`a0kc4au-tVbVz zd0m`Qi9sdS{4Fh+e-F*&N76J_f6$EV``y@2_U8m;IAO?t`S4hL3np{bVKey_p3Umb z{iE7Rwpp`9w83n$%1xtvd!Ari{U_XVJVRasropMv7k8g2@%PjY2pkhdZ|yWlW85Gz zKeK_l*~^*#UBj{AcOwP%UB_l3o+q{R;lu+btT02)pbwsk-;1|$F`jUr&WAbfTnxM` zZj$M04eFlp9?ug}(ErA1QnKicv-jgDc3FkkENjTlG84TrUe32nOyaaW1NMG>9O|hr zFyTr9b`MS9J(Z3w#b;Z)ES!6ii*!ObCV4pDFgN3=E{+_~!GwRW*^Ys3^PpRtA!aP^ zOlmG?d1O`tU;TE9C7mN2wmX)ul^U^SzX7~j&y2et-;LC`*HrbdkY29UKyYm)T5hOf z<)16`??f@#TPab(rf3@Bmq}L?oJ7F95;`!w2c2Bl1daON*#KJmknH3N(BndZQ1xa>9h^cv$`T5(RzO5SG?&xfOmLvcxV#*;!?0XvjU6v1R>*6K(NHA2p2PErzp-eh1`qq| z1%nB*5jZCX)9Q7&TuYOC%xukjBg{B=q$<}AYsF0~1Nr)miQK<=HV<8x&F$1yaiY<9 zUbJa0@7gztKl-`yImaQq_3uay={u2sJKm%66Go_N8IB*?yU6axP!VLLA=>61=kc~j z*mQ9Pe}C)G$CeG^AKE<_A_~?USK-aLEzmsGgWHBn?0?IXFDk!5Se-f-h5UhEN+CR6 zI>TK4MF-sJ$$h-;*;WI{W!@1SEx(GQ7dfc< z-b=*HzTtt{r}j!oFB-h%fjjc}jeh^GM`@bvXf_^#LG zlJ45vcsK=Hr9p_RdJem`quBRfF7BOpgQveVS>3iC@8NcQG%OS6LYIj5llzM8 zm%=bkX(S?Enjo?MI2~JNMf(Evr2H;5G&LwoxOR1vwA+2fflDj!c(gud{pyZwEfyFT zX$9+wUSw&VCpP9f(E?TZyVpD5-K_{Y&-oBV+6JsUPm@a@bme=q?b$%~rhJ?jOoq4H z;@_N;80~Zg^{Xv;t+6{N8!h3oHfwqKwPcy8a|AY8tC6A53vq^Oh|clERKHhA<4ayVEpP-sy6KAd<-HL&DG`x%e)zad7i-rQkjjw0 z{B^lA2YaY^~PToc#hO6oF)xq#^4?}+7A~d}7pp5!;Qmy$H=b)B8c$76l{Ogk< z7VDhl0*f=eO<^4eX|!Y86LyGxq)o017qibn2W~y*0yaI@=8Yo_*tk)h^IuAw9bw0p zE#G3aTQ{s4mO_Kx)QKa#eYneFBhI{f2tQ}}aCR>r-eF|VO=sIu@{tJnvwzbF=NgJ@ zZbd!o_Dc2n-|201U#c6_nsyoLLiy)98sc_aczRXS`H`)~46QQf+*M0)`@v3VD>kD0 ztDUH~UxC@JoKcwcLiCPd>NIsC)hY~z!)hO7O!vdox`U+E#+HIRqzg|uLo#NhJvOd6 zioEhEe9~qHXTEaflZskAHsvPD-cQE|!*Pio7pQh*)<6M?0DGthcSO-F^a! z^j9Hf-VKacuodg#rXy#%Kf4^M;0;Na`20c*jPGVZ{ik%0{=Drcw3Av(UeQNHW&JZe zz4aNZ9OWjJd|2`*;Dc$j4yDd%SWb%DqPQxAg z@by7$dB-kIr0tV)y$e$LYW*VK6_Ln+Jyx*KrMuv|`P}y3D!y0|!?qjyamkVq{Oz<0 z-}|W`Go^Cq@a2<~duKMD*QFxhz#LR$&EnFZNvt|8k@xsc(|cfJkrHKCpu4j!2rZ`~}T2_kg0|QfxRmnA~gU@bb2m(EM+OXm(mB zKE%p%2(>J<6t2RhADc=2WeWXGR>7}vscd#XgI!)F^Sa88n5C$PgkEbYBBY4QzpNp9 z&D9iGa2&_}Zb4#J5~{n&d&!Yq_%|~b^S1Beu14jocU$f;TNR0^XNO`>&_dc+ngYxT z!{>39*f^#WmFjE|S$kWFPx?jjIrJbNJUEF9Hp=wCbd-qiv)9SvN>^Doo-7s?D2Ryu z4pPRXCDPSFucdJ9TWD>g$5C!~FjgT$j2`FX{J83bvx>~it-5oW?lw=Mrh92v|GWn` zhg$H6p5Az_e*xhK6gfv~q#obG5wOjWZ*}-5oK%V_Pva^zYiQv1+9TNc@;cmWbmg$MvcROmOBi=ll|RdVk{%r@5p>3!7hm$@=XdAxF#AZX_c!23c{VV3 zogqxW-=Q^cy|5_%JiImxz>j&6G}Bks^C4&CEDJDi$x-UwWdLo8pC-(r&cbq3A2_N8 zipQ!3&`aJN(L()UBvBm`h!mPnP+EC}cDnh?~v2b(VKz$n}#5ZN)nY{N;ceUpG z`7XTKwG8qiQR4r4xjMBT)091-Wp<3846*mpyY)UaZc2idSXwGmauzUZ6MN?MRp zB=v3GmafF<;Dhb}1Z1nh>WByKTzDk&uEz4oZK%O-c!SqE@Ji&e{LKU!!t(uuwB??Tvu-gzdL_K z;R^!&;DM+-7r=G1HFxRvbuInskLM{F1|P-_uSVFbK0xRrby($g zgPHbsD)~2)U0e0%I+=fS(9)UT+Yjas(Vkq!Yq(K0g3~@P;357eQFFZwcL+4$qr0Xd zFM2MN{3l_jauM`R^3l=r5H7bbM_AD*j5#zKE60{2MDYvM?K<(5rA{)RV=^YFhvLeDLT54Qi6_C#|0|L_+W+MO`qsZi5=wyl#rXGfvnDNsG#l`1sq#T zw^j^bwO?+md~5(44C{{M{{&8bTMf^OZ)Dc zGMnTbT*6ziVOd+A_DR0?+-{E&-8oQub`6*A)#8_#%wq5}#sl@isO%LE)!)i6)UZY5 z;T~9#*AvSk&7r)eBO(vo5l4n?r`N+5V?m#NxH0<~O0!yV(Q7sFxLmI3@;4`I_m6bn zq${j#4&s+m4|+da5eL)ap&Z_ko>UB^3R%0^6w?(W{JlX9V{obAwv=l=Rot7mmwxyT zlKo~kNW!~BpBVy5;r+PW_g>wjLM15py}lh+`mcmf$d zn}R*JPoeBuAT7Jti*~e)r;@?-B5R8=>^~pJzoMP=d20<#+^&S)8x_Rp{6A87+IMm7 zq6#K_nu$s3JF)$y88zqB(9HR(Fr(u<@iy&@cw%swo=QdH@r3RaJN5@%@JW}(HJ+0$ z8c(3VH#CIF%Lb6_pQX zQuguA*wrT+Q}+A9Q8Nsl8{=@gu|l+r4y0Gbx~QxeAdOw9CKl0Qai_i&&2xH2Khig2 z(MMS$ERIKnMtfe~&{ygxEhnFoQ)NB3hJxH%C^5wvMZO25I0akb*Y%_5aM|5Ce(Z54 zt=+?%H<-$2Kc6mGblFGtaNZJ!KFPhh$5N=!Y9{*6OeFsXe_^^}v$Jkm7iq|k8s|}p zs+>Adn{O=o4xPi-p>d`ZPio51N#O;u@^;F`dpXlJq+iWjJ5Bxd1-? z#{`YOCbihT<%a`z@wF}EIM-hA$GdHL*vCug?<8jmvwCn|gbELSE$2dAc8iWp3Bu>$ z2zI(E^U*$Q@Wx+@c(j=se>RH8wU!w;krV?f-i*b0yM>G26-le%uk+mF7a+n55vb^Z zv3okx2m5VOmgQk*pL9!#nRBFWZm`#nB;^b*@O7t_trfnw0ina<~qy%NrP`WXK_8_j!EIJWE!eoQ)qUPt<3V_g(_ z{@948NpV~dV95GQV~|`}AhUid=wzo&WWFs4N2+RZZQxrJ%~9u9tiamoZ&2zu6YqoL z@jfIQmeKtY|DqeJthDV{w8BkDvM-!_r%@G_i!t^|djywGYm14aP_9=W=E_nhYhcw$ia;DC-rBQ z?;ZHpqM>}qpd%0Y9!=LD?IB-xnQ0|t9e=3}O6n}IqiPGi>#K=i*>4ir^F38Oo`5@E zaY%CCft5P{dx6DS31?omon7#XtVY~nLnb3 z7a#O+W@jvoQ#nqd*6V0ns}w3&dYz8v_oRG-qj2bX4z;Z=;n{`HG(W2o7HM6kTQ_7D zOy)z;vrm7D*sF~3Ywc*>ia|8g)dj0_WxeB20UjH6b#w{d56gYE_-b`n-0ZNEQZ6;n*RuvZYLGcEchlh9usZq`_=qOGZz9RM6o&iv z;@g&Tm^8FQo3!1OGtiMPxSqxKmfKj9xlR7ui>Te!2)DYExcX`=R(wo>%E&65&ac9q z@`IR=co>6LY0>Q}P~-0)dSWbh9m@;}6|>H`l-CiOCR*sX_a*i4*+=KyVrW@{KD7k4 z!MJf6@Ke!3gl>Cm%6Lrm&g-Q{@1M+jy#CjDmw7Xh`-cvUu_yV;7pl$6YQvaI@l$`2K1G`)mkIhz5n(tY}Pt3u4 znhfIsrnqO+k%IQriIYn#_@?h?IR~%@1AOx^{K{PH%6CARSOZ^cKMeLUMbFno)OKG7 zdOz1*#;mZxp_+KO8UpA?Vg+S%CAz%ZJw#p`A&@#@JkbEZ>wGFV?KY-*ONZ5u?0EwyDv`u3==Ch>C^GE<3;-Ex#FEndryu%K&!T`rbqiPkgsYHWsX}& zww(^r*3396cNV0cl#GE5p;$Ph1E$|t0V&H1Gm2$?y!{T@-@6EDYdfL;yau?XuE(Xb zli{!<1@BM0L(g3q?QRc+xsNVNR2nGv#2@Npc#+gUl#}J9Vp_Rj6 za8h(`uoJ^fZy~y_3{QH;<6NUT4Q$q>mNzzp5%WY*n4&oJ%v6*V>%q|Q5`BBSj!r%F zr#}-^X@0Av!fVcH>F^jmX-IWj$*@a`(5b7Y-bdDel-0@pYc(y1)I*KTNUG7)MxU>X zXyMw;(hm=awB(aQp(Y2pu37lg6osj_GBY)QBmBK~|IJdl^#oPD7uxu3Wv-TdUl zBbDXcpQQooC=7+Q+a&SFu~xD^?&`AQXa^U|`43npH-b+(PU2!82fk=I5o6;E<$T0J zT~#4VXma(|o|TZij%jF~#Ci}kpqvm4hgHiHlwv~1AFgmnh7JV?= zj6eJ2srvB=xW7q_-@MS{EDtp{vbE;nwhy41@)$BL2@|cV(4p@ac-{BI?z>A+qp=vi zUwyHz+78=g?%3X~?npO~&x|)*FhgCD)h7JKZA%T#4k9_rHW2yY;TUCAj*+r=b#a+7h<_R4+pCdY+=p{Y5eM|Z`8wn7e{F!z zqhrueJAvKGqoDC?0u&S1qMMS0EwAL9+`Cjbh+UYuuL2PVi}B#t1xguppFHg!Qybqx zy1Kc9;%DupOSS4eXsZVQZmYxr@efeA_Yo5OA7b*jr?l*#%s#ZfLi?sg!`&(tO>!Q! zaElsFnH!_UvpbC3ec+(g4}CTRis3upeqcYmuU)~Y9nG{gB#5TgY@kqu0SFy!j>viP zTugl=I(XZ`P4?fhzdU!?F&i!IN8!VfO5#5GWb$}=`7KgczeD=`EoeRLHQFS0hCD$) z#qv6^d_*e*OA<0+qgsjJ!)vk2 zqX5Rg|3iLIceGKQfVV?N;>+R_I5YkhrX5w{I}_eh@ZBCzD$T?4A?AEPOOtQts_~y~ zk7-6yIa!S{#i>gtX!ufn3>z{Ug^>!f27Q+H`QE4U&8-ofluDQ7`JDdGK5UaPl1t1+ zv)w*|U+7C3_1g^>+neA=<`e3asgI5_|G&HO4e_*gBYmjs2hZG%n78c$zRc*t9Xgq^ zX;Ob4U8j!RYA?*nlFyLEGHK1$HAt#yu>MiJMYY4$$KKsEFFOfUCI!X@}3@j zS4L8-IU2P7Atdt~?nboZ3vy=NQ^||Rjf~>sgN=}Ha+@sn-KP!mmCe7+Ah}m=7zck0 zM8=8)c;3pu6^oN7kv+Wk2N<*a&>M(y)8n zsN#xm2z_4kKv=;M3%23~>#`iJBd>&eMj*ejOv+U8A|bnJM3Q8ileiy-@#Z z7e%c$p|Z>z4)d1%HXmc4aCrhLEh#3Yft7ST_$qDc(3Rp^S%~WUdqtFy7p& zO}iAvlb(|{T`n}njysX$T69KaT$j(rZ#Pk+=6H(F-vMin?_@fwSU}F?MxHrB3BDx0 z_jv`|AJK?iYY3y|lci54LDDJh<5J}1-ux}S1eDuEM@Bcp#Mcf1SAFSkUUztDMB$3x z4ixTB5E~v?i7W>zQ5Ok&e3YWz>F(EwXz2M;k5M(>4AN z&ECc^DV-r}WxugURUJukSJ;O^Ly+A$nOeW_5Od{RnAXLo;=y%2asRst22|P7T$L_j zy}LTTcst4*Gi}kyzdu)?Mo(b z56dEp`yc6U$zVE=qb>Hlg;42!*;#+zMyD6wF5tl#JtYrHEr+ogK77>2>oC@u72?c;VtAj$gA9troh9iBDF@>$M7^jvl75?egL8y#iWS z!)dd9Z#vS(LrhtBOxiZLg{ssgSiP3nzo}#XHKGX`5 zF)yER^JP2CKGTz}x3qF8416#9T|H@QdwOH>yVVpdP&Y);RK*HdtnQ4wG|Tn$n=DJ4N8bB=TQc zNu>i8L2ckgSPzfa$eDb@1~F3qF*WPvVSsHGzJ{tx^|X-Ls`X1-3Ga(}yNZ9U`um==0R1SFqk$ zDzgU@d38WrUijZ5*=LYPM|>8GqYL*+BC;Ewy_*Bcc@K5aIZ0)yCxq&#ky4y?2y&kF zfLN*sk9L`KwOflg?)6u)YdwADi=ZcpKu2Sl{d-UA>vsjlJ zE#2l7RIy#gbjLiF)RJ`ge7Pf!_|=^){P$zSTXlRMbXaTT%@ zFj$HAj5g+9HY!|m>Inj62GF%hv*Bm>nF4p5q|ORQpzC6b{Xb69`RH_hI&L*T?=y=7 zXUEd|;52dgqLy?y#{=ijWnoIreY9;zMR8XR{63pT(L1lmzJuxW{aBHX0%Kfd^T!ilGYE{-~0FhdW?!`V*q>wvjoSO;SqT zKv8gdmzdUQK`VBckkNp_)cx{lY482xIpg$dLUF$=E_&AW0*oPmZ=0`b(k7QIa=_o5(k+9)GXG-7Kji3 zhKc3xQ$=RLSW@lW1E0$W(4E>iN_8JbGZo}smrhgpT~z|qX1#+E&tT8Z1zgzmC{H)3 zchj`yqOqVLlsi*2{QtUR3 z!&Ae!V8v(-!F$->mc3)I?YL!~CI0ORrb;PQ6fNF}6LE{-9+gdc_ordZlPYXlvL60D zgXpC0emWuh^CkCk_>W3NVN98v$7oB}Z(XNR{XCIqk%nI3ClD{K#mGS#G-Y`#0!XKYk_!+cYT=a?~BRhg;?@?Hx9S(6J57}`ADAbV|B}>RRjK%tC9if;t9P`@`gkR$rl%Fj`pXbkDaq=|^mikETe_fUK%K7Rewxztj zsFt7V9_KX=58#TN$#wY_4a@Qqq-z!n_jI|VWpcT6tdhHKl zN)n>?C&eD-MyMHW391MV#R(s4?@R}w?#p1I(+YI&8gIMNE zGdDF*==eqO7~7T&Vb0y#d_dA*FKSqwMm2->Qq&?l5iiUrVNElwd_G>xUbc!}$lSRL z^HYkyyHUA#I+C}jv)z}+vd==^)AT#RA<7Hi2DHM)Rhhu>4?4O`yQ=J>%hb@e)$HdAI|pAhV2 ztiS{AKERt=YyKYBiGRrZ_N#>#Xv!7hZ>z=|-NEbbcI4*87OdaYLIzhKlf#DgsJbwif_}`SwJNet z<@+iQ7?#8zYUA0#RF^~j`*OZnIB&e$0jCV7lgZ4kqL0iNx|OqE?m?5ePbUY;@7in( zQk64^MY)(qeBn8@ccrK^*vDVWfHdR zcEIUFQ4nPO&3-j+ltO=ZCYY*&LC7LQ5!gRl>Bl`)IkYpPUuHM<0JQifaLK#zwgs zHTGjEO?47Qbz|K5c9{PBXe5=c=TI}#n0gvd7k^XaZoDPFWO7&5MDwnTfzzVU`ehSG zydCY5v^rKUmX3#I|K(iL*$86OMP9rq&BZFGSgh;ZkG@v;AaPlYlo9z!RQxi7hv5sd zSMfvY^*gX~+ao5cf&AtDzD(?-W$SJF!arU#c_fjkA=E6ED=Ji~%jwz>UY2hdfyFp_f z#*jhu7@Sp{E#2w1Q?y-Vh=1DM#MIWU=)n_BI4Y-#a43uAVimdVtQVvDM~LQ_*J9SO zR?bUTyl@W6>`v+DG*Ozjlh$s&Ou?qF#qT<6F(>>E_5Eigcl6yM`{_!&b;UZ_FMI?m zPAhW>jNhw1prX*74CH^qmH6La-6(K=UVpiPURcNISn`FlIZG}MW9 z9n7%qi4FfuIY6-;74U52SuAZ16XwS)D62w3P=n0$b3H4b$Q9eYbS!bnD3x|yIOe=E zDw(WIS<*FVOhcb8%wQ5wjx z&jv!OtD}-RU8%RpWv5jc9eGjSORTWy1D~Dm<^M&A$UIskJ-Vfe8?n(8Ua(x8kL-oo zX$5juq>~tY-U(YrzoNvO`=o8rHt-l(N<)Q<6zi}Ga+?B{C~1qWajwvr^oR@&B{-il z_d(@=UzF8(Ih`me!J76dSnx(22M->hyyJtZpVw~TeC8BZJsJno=&$0WthqdFSWYfc zwNCYS+<0(>5-%>94h>UH`lonM1g<#N>tducNwaSY5qI9H?(|u)%A>zH{zy~0rnZb` z+!#U~)$K%`=?`jNl}rj>|2S`ynMO;i9cjd$3g=!H*2DEJj~f$4N1nDr!`u z2D#%ndgEZ~{Q0}H-3cxJa&a$oUb-NA`4A}0eL@Bu6vgywN2$=(mzFL0)!Hxw?E_O%tzv2@ndRtemFT7w{>zZ zl_=<5xhRZ0>by$MARgM14BH{e4B+K$TtGXqwG=QA4@&wla9lx54xa2^74@1D1O8#Tr*zQn{??k~FQ6JGbA+`yz+3 zZty3nI=-5|?8tEGlJq}iXBtk`_rHHbGEbolWgap_#M$e;4k}71C5l9*Mx`_+` zM8g7!a7g4h?Hl5QK)l}vcCVZbL-Q?RMtwRe+OB}CWtZXI>MQZG?Us0WMK!F-F9*f= zMDTo^1X4vaKtAR%I)5<;wJKCI+tyBF()W#n@!b`ubzLEHZB}Ny-}7c{XtlJh`%}&Y z`AxC2Ho9bI(|d?9-*%Gs?xZh|sP>}!S6fl-dlqdSO<)cj$z!_1S{Yv6I^gvmg7o;k z5SDofu6-x~^MY`A5$6Rfp9R3reW5@lYdHqXBbeEC4_5E~2B9s35WVL-lux}6r$@TL zbZIjziW_Ej4eeoyrtz7dIt%Sa2NUe(tWsv`zOBF$ZhPY}XCqu05y5dVYoPi}JD5F7 zL5tsbG1I!Vc?WjP$5>LK^gdG!MEjtl1{^&u|y*Omr6phSUlM$I~28G8L!6L)G zp!&0f88(z-MhrGGyY+^V!<~l+*ikgWA_eB+cO$!tvSRRjn<&KkD?+NuP2RHw zCU%lL>+M2!EVDDLbmon!jYB&%j@mJbe|aKZx+qq1fHu;1~yHCd+!`{jN$g}iwsno9%Y=@jxn8~OHoli*G+ZA4yx~ZK}Y5ZUh#x} zUf2E8yvo=L6tV0DQu$toeAMqV&+LqmYe6#;`?j6gz4WNnoF&n9ZClphJ1_nrJENmRY=)u#0`#k38QQ+p1DQ>WLJ}?#=t)ur z^JBS_-38}IcF!tXc(bNVMB&emqm{2k;m6vq$Z1Rw(h{ZMdGHV#yD=H^u4%#Rz>6R+ zeGzjHJiiMQNp%QB}k<&imX6ISOU)BX0)d{V&drDVfGo?5jlJz64oWDk9TqM#y*f zDx}yNjqWR3LF*e?P?}eWf)~l48lws5X6G7a@Yft9(>n{*NE2qtP98GY_=73x7~`F; zo`#BhU!sc7a&R%c72U5(N2|`pGev$1jQ3w#q|coR((`I~%eKj_8NT2 zDF%o8=FmT76eUc~hwZIbp<7-GOFxpwpX4f_{wmkWzUc=Nc^Lz=T@KBBTE=r zLNqU@0Igj93<+8r(d9qJP`dCPIyrq6y4&~xjsE8deP3F6tCi;?M&>7K8?Hbl^Ut80 z|Kd=wTPzyc){oYG>_iJW{cf{!5%Rl|hdeHnAv^cgXoj;4Q>^nCb+>jPyGI@C8TA)BXzAuYhCh;q==gRd zz1I`zwCAGciCL)VXEP5Jc*ybnA5^&IF!M)`V`!dJ2f7zf{-piLYSUe`eYYkfxv+y* zeW8L|9~od}rv$QBcs#o29#g{ksj7Tx(CU&t^u3;AcZ8{e z;)zl;mt*I>)4ReZbJW>u2Zt0 zlqq=|&uEOEWO#>W@M0YTnS*D0(M)kM==OPm+V?$XhDFVIPOHT1G#(Y9vIzj{d!H4QXYQuL+ZH}ah)%Jp(xN1E%U;drwaY!yAi+)&G9j15wl8`3=}w89=P zact*$Wp3{6u?;Dk-(kM5?1JliEP&lPn+p}C!3|kaJUCMskCT_gOXYi@-M0#D_X+~j z!3eMo*b4UK3ku(<0cw1G*t^pkh246M#1c&4^JWZFjxPj8`adKttqx^YSJ5)fIP|Zs z56wEL1#NXpVaJ^Y^xtv~RQKO%G@&;ZEtb|q^Q-fCq?1L@j0w`UP(sN)ADEfpolHc{ zSzhG9N)%=igceXw-r8IjwEg!2fS9(u)YcUw__3No9Ti|a;Brfc0E+MtDmt+xWgnZ;v@1r25l|--~G7*$)Ct& zR%FYe+a`IaL{0*1e%PU}Ps)%L5r=`c<4AYgIrOk>0A)T1K{qX+CN8bW@_lvXK(^HTZ z{>gYd&SqBWoafCsdIm`qd_dQ`q~Kk2Dtv1L)-JPE;3hn=vTKWJHw4 zc(YhrW~0#@Cg;prMscmGow&pl-WxAHX8a}r!*CqWp#UyF?R}J|a88VI9WiDcrZ^zw z4VGHyqZgcBJ5vmyr|FJNtHT6dhS58o_udfn#`Oq#Cc7T_ zn295^EeKuSlY{z?+992MJ+$w$9CB%FMA?~I5L7P@H_wgZ7}B?qH5p7Ovf zIRrk&9cJv+2AM9c6==LnB~!7N@@|d~Weno%AoTVd^r*TCok$oDpHi|=%*z1e{df;A zq3a!QSM4lhx?GLB0P%!JQ^XrD0Q%21KvY zhW3+VcGYfo?7UJhGwL;4n6B%$c3}IcQbXlpOI%msJtD?In5kf5DCH2 zSpZ^1@K}V8MrYSb|~m&C#!QGf{G}HCn~Z)56t?(YuW- z+7)&bt?NFHG)Xd=e28Ps^&Ucn7QdK{b%}^ExPZbQ9YX`%lOXWTBp9dr6SbtLfy1mq zC{a5E(KmHrz*89(SG`2098U4BnhDJBmW3&^MA3~1ZR9$shPfNH0^Uf^fyV}tkbU+G z6V)`#1b%$L`0S8@9)m;ZH@DU~2YrGk3!lQO1*f6v@&?$q{ToViKY)tlmVwqL4EB4t z?DGUpPp@i2?Hf*@N^J$0&4_@S_;+Og{2ofFTZhhxL^0{h6VTVsmgw1Dbu?>04Mw9B zAmIHS)I9w)+LiPfbr*<$)VIZqj-Eclf4Po1zh*K*N!+t-=ndYTdEUsP+6?tPxx!>r zKVe=5>|$1l#xhdj`;ffLHRQNf3|8;YMsl3zvia#5<|lU@2U#m183PG)V{RvE2^XM9 zUMy10{EZfI-IF1|W028`ZJY)d3X$y#Ao)9=$>#0kMZtF7W0@*6*t#F(mT+42m^C~M z#qf=_0Tu50pE$k~Rn>H$eNmGcyhfejyZ+@hbNyB&4J;T=&jOb#%fW@+4|6BS0aC>% zL9c;%uj|P8v(dh_DZ+i~Qp z-K*3*JEI8;7#;Wj7<2s&CUvz7?*Z3G^dWyKvwQ9)^kB^)#7kI-Z1z_2Dt7kpD)(IE znFK5H4%-^>8r?j3W1n---RLZ&{x6>E^;^K{JPXhsX&$00)!><$9yDJx01r)IM6M_^ zCmy;n858C*I+t~r3`;F$dXhV{(Ndd{YVPOVIk^~ZnzsP8?)uEEYrDuC_NZdKmCBj^ zA04Q*BTB2epB_;IyX~`mPj)<|+B1m66@d=mkY|G0q%8)4O5t_PxYDiy)m|#wfeRpxk`2~PGvIMVg-zD zPAk(MKY_8Av101?g)p0MEJSL_+tA==8f6JhcZEk3DSvX32+|Sz^dyhBgK|e3*kCI*KScC0zTWV(%^NtxZ8DKQi zB~eUR3|g3#i1=Xzh?a1Ekilul#n}mYM}B8Uj;f(+dri@JqZw#HrW*>|>5p~?d}EUC zXrOP$y^yHGIlGJyWxE?!rrHV3bs@`$2ODnMz|E2%UgM5@yIHDSnTPusTYodx8Oyz* zY*oIgGCOqDnaXT?#`5?rTc zHS_5y+uOwWr!MWUP6KnP{e&DbAn%#mNbZ8D@P&g1NfoHTk?pD|_oY77G^wRzOCsCu zxQwXTOS4z&EYPr|869-oK+Rug&^2`r==V!jLW6t_8Y`thMSjjB;coV%DRnyO=&^>? zb2Qk^zZTQ3JF&F=#VzvYj1ju%Sq?4NLJgiy$-T?gNq8h!@JPRJgXQRdmI)Z4;g~Z_CTaq@mo)Kvef@%C0 zO<%W)lb-rUlIb2zeUw~iOUNerLDfRIez&Lan~i}`H(#5~s6Ii*?yRGpihIz-S`#u# zylDTh1QmQLrgbv6=snF5!HnRi42g~<=YPE<=FYCP#6(}HNFI?=MFn^kR!M%#jFBl% z=97_yM)d-2ZK-3ZwF+AB|SFC7MI(G(~CUqV8tjOTB`}c2dOGcN#VG-bP<*J|aq-=cUtX6iv0s!=@vuq_jAQ z(>;cf|L`?BR9H$YG>m9U#4BLs#@K>o<7q+|CZAtyf>^~s7#p>wrfZ7mf0B(LzMJv%Xb3a|W90XwZ?4gMmN2>(`Ap_GoTWKS>`z1TU8 zSp9K=xNd3Yy5uXgO(v6ML``MRZ(l;W+%#h-A=ljbF$BLHl@v@vuI!ap8gOKF8>#k_ zqCP(3g=f>uV8-e%Ap6_8$;?rk26s(wu5>wszJ_Ykwon_&zI;z6-fm>cDH(S8zQ<^H zu`%*K_M4o)n?b6Xe&P|MOvMBDvH3QkZ0d?h?CG>1L<5z`IU{e_){qXWqsEYhr0{bK zd2HpaLVNR7$c_PVXkYhMuMgz*{O0ihYa zP`B6?58F-0^^y14t;SpEBo}A=(`hMKrOv|AJI$$yZv*>uu#Z&8FF_&K-y*CO0C^sb z%naRw?0~T;qj9W~8ZL^b>v?IkPwNHo{@F?Ns!i~J%n2w^or@LDoC7*=T3~V;p-NItyT*B zRxPAwj@%%Q5?5JMBRzINLX*9qZi_Fb{~<07JE3``j46u*ntCys{^qp)?Q^a*>rIY_ z6#Y>+_bLsI4iv(H$@Q>=%UQoU_YKO_EkN9FqM+*~$EVdhA~2Nkr?X`9>5Cb^sK>7i zIQafEDXq;TQ31h3Y_UHHDKMgDp&O`ck~ZC0DS<1O#IiZPq2yVSCLO)Gn5Ga3q5T~l zVbrOsME<)6)qr;-(VU|8j%<`F5W%OVMQEo;Iw)^4#5-@6FmPv~U`~V#*xY{%Sz86T zu(KVTOx??#ZZalAwY79yavlpOyovs~X!=FhlzsN0nBEADr~5{fsEdOM$@CvjkNN9@ zh&|Vnd?|+2JJUfVNCnx`(GS)K8rk`;xjff{+tAyW4}Asa*v3ntboSdu^0xgTo$Osh zY9jx!QMwaoZ_`4WK5~*?6wRZC;q!TG4*zWaBUsOteyu`re+X$fHDOtO^ zoa~yChd;;H;@{ok{KnCVRCrmIiY`CSW=Z6di;-e9-QAIOHzD|pvIM`w*^qytKpXp? zisMBDh=S{?)v#^3D#?`)Fi(^7v822Ptk+$QYhK3^>#%H68sJ8&jyBOvO-i*Na-A=t&Y-UD}96iu%#FX*=OS{3x&3X%Jn1SOD9OJpk=nj-+7i zHhgG(A$6|%#IDCHAt6lG7g{@PPr%yU};L{gGN#uX!!gd}sJ?RB?v2uY;h7GxP(+i)-h{X{hZIIz6igVUSu}&|~z=*a8 zKgTcz^D-=1eU8nf{x^!ux_tum=~jaXpZ|$Kg%sc$`020&5G4 zq2T>SR0Lh@1J@RI*T!hX4{~6u!}E#J{GViX|5>=|G6#g;_Y?2_0-`#xon-GXCpY>I zkv;4!G-2Wl49g7h>VkOqu%?@J81$pN96fO8+-Q&(x`rf`uJP{e$RkRH_sI8-O(Zn0 znA~ujPaiGb1kdgaurK|5(Sfj!WcsYr$k53h8=1Djv&?)piF`Gs7H z2x8qi?ya(}3Qb5khRa_~BYZP2I)8mT81g6KiU$wDAW4PB>(8Kh4=)kPXVM^b=p)BV zYk^@=O}zGT8uSSAIRjoV=?eEIdF$k0q;Mh*HxtF%1V_OBF}IlBjvzndC)3m>7Pp)# z#Jla^LEG~pS}#72K5*q`lKDYUm?1^uwKYlIydwP6LmgjMw1Ku?q8LwbgLKz&(zEI< zTxi^YzFgY@FXL57U0*EG*s4luv}4&kg^N%+&l&!G&jyc}qtNs*iH)nd4x3A!!&&PK zBr?1VlH0uL-O^eb5qMJ&D#mdrM5@r`PtN4xtgqy(`DrqLZ!9~uzK3Yatf4CkOK}V2 z;1xvy_(R7c)OuzkDIMaGT`Cp4Dfc>2gw7UrmQF5&tFi3#FN1JW@&x?8a|>L4F2_?X z>Y3{eN5CeqTF~Id-5=xSsKcVQRBLY?$z@E~>kgv$zya>;c8%N9au_JJ%OcEtRr+kD zCmDHj5B_}egfpIDa5^i4jV}3)COJ%mw>%#9PVYe5&hG(l5iPQE=~2QK6p@}|-_caN z$4%ANbpqS2o1m843-MfjzyJMyCebejbu?;{kjEL#rZ-&4S%LO^@q$0zUck4N#|h0du~B>s9%$Q%uUSXnbt3UpqpFlFZIB?{ z?;Wu)?;5-KMm@Xe{RSLj{}f7(=MZzZ%P?wj8i@Ti*!j=^e{(d$HoY2HH{=r|c~_3w z`yGbI(HZQ=6?TxmFbC|ub%EqL4J_4~%Rbf4;n;OwVf;Ui4f#-x%Ez{Go^N^Ld+7_l zW#x-K{h0+xzpwe#7?sRh@|1tt;VEi~d`O`Rh|8Xu2 z8b~GDJYQ0wgv?yg7j`caI}K&#tF0YJ9-wu^SVx&YyIS|3u%tP#Q^J|PC>w;HldoXc;UAD`BnKTC4XmkX9-5urKo%G0(NM9Ui{ ziElL5srZb=KP9f?|8~8{T7_!(&2=R#f3FLCRJ4#wnFp0N)S&n6ggmXKHAGU+f#xjL zC029B*nL%hka|!-^F?o6!DZckfrR2RdZ_0n6^$w-u}{09+V>Evcx^Vxvd)I7E{E6B_ zr6YbcZfLzA=+GWx<1Ng|uC2T1N$xzH^KA=_UoD`g@I_j2^$z*a&`!_z9HNS;jqJtA zQ|Y>wc_iTL5Ghi7O$P1^p%NV-i+A%(+Y^!a1X3_T86PP8{jP|kw@+m$jIGs*j`>{t%(i)BBX$Nipuz2S5# zTTQ>I=hKc!Lb9!37wc?WgRU9`@+AGdsfWJ?jd7pP?r*$Hx+Be4-mA^T(&jDeu}z2S z+%zGo_BY6|(F@Xf?IbBlh(f08XM#K;`ye8# zx$$fj%wFOTaclM9r%@8gJR3`k_Zoou$bUprWG?0H{m!Z-iqN`mu{1dD9l0(YMR#PD zQe(43vc>EhnPc0Hehrn8f#*Qqnpx0p>6=8(Dw4d~t;%j0N`_YsuUI8YP-3zu?vZrG zG7{FH*QSmRa=G7pJRUzjAu1qj}pSouTNkxcKxS|RcenB@4J&(c~fIV z+b*M*;VV$RvoF%VI-BbY!MN(-JdCGYAjJkc=uJQlJ56N*UC=fWPfUIf8QZHMao%{k z)!vXwzAj|N_IHq7DN?A+^E@NH!He^xY=@OyoAIq-EwadegecuQ%H?e45Z|q*;G^Fx zIQ{zwIe7jW3^gU-cX~;M`eva!<*hJx^JKifeO&V?z8W)b%VAjcCyIpG zsA3Uk9!?yKfz|n5*zxrN@KSjskmq##RfnTli@tcsureUWFD=I%o#{|k8wvTA2H1n^ zfSt_ElrN-&g3-no&6^+CvOe=FS?`EW*6=|z8g83HhW_&b%ldlO{pUP3(DMs2eydA0 z64S|GnkBn?C4n^q&X744fPYPq&A~mS1t}`}pJ%LlNoQ9=8RB%gMEl}BR>|lEt--ryykwO#^jDE0 zad}vXMBmrp2@O9X; zz{|WX`L9D|`2%@JaB}~6eCo;-GPGBN@hcN2en=Pc8*h*_BWJkQZjP}a6>Ci^z*9yH zz|r1?n#<(V{U5vO7X2(zp85n<2~LtlE$-yk){|`JYf-Fcpht_RHZ+%BQH11va&Q4czrPytKh@atKQnV&rD8uiZHZY%JEMd*`PWITkrFckM15Z}} zifDhr=-5VCd3 z;bfB@Lyqhofh7}liJ3|=S#|v$iu6rFlE$slao4Pm>#yc<~NF>k`U>4na&QJL2nz4u?NJvKvC)pSgzHF z1u|U1xXJ-04h@n8zGJlav7zw4=BdIX3r*-7>4v8E!P&HC-V^Hf-x`{{eh;nRJf5yz zmrU2B+$6r%H`&lb=h+D+D~WRT{lpflgMuR`xq5n1>p(WR1=>4p1^yc5sWbwHXo(AZ@HHd2(+hp!?`}hC*mxX9BD9!)PkVU0t)nxxoS(zp z8*898E}pI_I!o7V6cesre74zW?j6$qH=86z`9adZIA||Fh$_7wkQh#*dEvbh#)gz= z%;Nx_>e^zIc(szOD2SuOt5WE8&r)t5o=ZpLm8i@=Px`S;j}mQ9Dt&wgOe}SWKv!Mh z>dxpUTt{|ZDj+HQ-;?fdm>T`GrUuf{SmWPhypY#|=IJF zCW0tEi$G3xuSv8Pk6dCpsJHJqDss7pZsYh0a!QhXv&DmW-+~ra*-)2#c5pTIX;Ttj zkvT)dr_I3&oKon7*URYI;5IU6zbL(P`vATEQ;I%LGe$!$iOhj!7c%EX0}*aHMpj9* zqnJg9=@B8HhD`lOBKJ#@*WU`sd~SvowXIJO`g}4@T4^l^^Zrgw%TA^CX^M3F^Os2B zz;$x(%Nja%RTf(-AA_0JrdWJ#Cs7%zfhX!36g4Ig$M;59;cfxe(~rioZU^A$q!!+k zt=rLCSuJRs(*ersz7x4*Eu8e|7S~4~N7pL9!#Bku`5oMTwvI_7A~TDr_nvt6WJm*C zJF7*NZbpOECms6XX(tW#FJyER9+Nf6tt3vV0`3$Qk!;VvoU2T?+e7%f&~Uv$1xBIDVGIhs;w^uq?2OWEl&IhIlM?I1z&1 zd^tg$9q^%FC(fc-zHgyWIvv~{6(PV#jA&ZRVY~Gk@$@EjT(Zv{?;3O?8?W=>$nY=N zBO#7Ih)W3rzRspN&EzTomt14a76Q%)=!G2Z_R(Xxg|B(-khoVCKIO*R6)cZL!Hjz;_r`eimw#c8@2>IWB0HQkD}qktUXZsua+lgUdQ>gl99lz z7)L#7f*lTh5PM_;)>>=>-q!EYy1Q4|bo+M}MOi*>) zbdGg61&)>I2!q`Z$ zmx=|Yd!ji;#Cf#Yjo~Y{zree{=DDX!+exE$BQ0*z6npZp10s#6p7mUfR+5AK&J62bTJI$h)Wgt{tRF8zW)K6-rkJo`52OxRZocEg9SY!R3gc# z2L*APBJf+iP^g)IlW3olpnKMKfZoM;*m+6|GNx`K9^$Rg^=L1bIpa8srsv7IY=EC< z9*{dBA4xBZ=(<@GS@#{|>9huY8g*Wb>g|fB@1_OO69bRY-IPA^x^58-JW&c-e>FjU zaTyva<;13Wh96>Fgd(+R^F?9E`c>4383zgEoMnCW1W?g!nG|N2#YEz=9@tO^E zdPWUBW%z((2_#8NSr*+eeUgxi{0jFf{bUWlZY2NRRwA9wPl(*TjpXXJqiDTtG?tr_ zM(a|cH7hvvLY8oY_v}p@|)s%gr&Y*D>hH zMd63F%|gEo*MuMTItm5bR|w~+I|vVNlf@@u+E_nZaeN~)Mrb2{OZcR;UHDB^9B;bs z1}hDxvUivUDr@tOUO$*YZ*5(L=k}Oz>zpBeuk?!k?#d_EIGfOqffQgyzd^*ZOtN9K z3MI34(ddFI+Plh&K6bSbCb>=(&WQg6LF?jiT*)sK{qHH{e%?Z^6lM^y-)hbE#t!Vh zPw$!&2ImOtBbKma+L6n@$K;Z05?;T zwt~MRy+pZUHhX7yEn_g-kG&nbf z)Lt`%-ka=#b-lU%%pY#-x=fDVr@J3%90-BMd&-fPVkf&~x*wP05Foq7J*2SdEbC{c z28z|)s5kXCtMKY3H}})!cbGoGfzLWgzIYj1mA#z2_)m>izOEso_fC*KPZeRMkrY4S z{X%~9@fv({AOL&&uEZ&#j^uLTNo4ll4`AwLsp+94B>T-3?=UT)V{@{}_xFHR4kqv$ z3`+PhHK+L6Ifqf=soUu7r5}uOfEs?(asYDrez8G@OX=-if<+Sa`O1|R==;hf_UeoI zY#$SXU(JlgYu4N0zrXs(nn)dNqj8q3TH40j<&n)?T%kfcC3Bi>vc4cmcMay=H$B?; zEfjyhFUQ~f{xy1}!ryGr`wns1F*C=4Vz7%n;crDPE%cj=BZS+#k5t_eDpDZ2# zVzciVteROP&FPuZ>bizE6Ks8 z(hOEbZ3(HnDNDDUwxr<|E8(a<59BsXqYTAq(7(#qXu@4sWb*~A6k@Q` z>>POeFdyxi{ew*2u!THm{RkoZys@&ZF8{)M?%Xun6<+kGa{20EHZe0AHnZdSr%O|@ z=Pv^iedREh!yjcWOPA7+?Gh-&Bn0XH-3PXJj98Cj#>DTwK3TU*nQq1@K*#&wx(Er3 z_{nf2Z@R#%T9W^c$Kr?Mr15MY1ISob0qZ#a$efZqs55;}a!>GhqV8^N`AkdvZQ3q; z+$t0wKHm&Sb0%PRk}@qiZx4e-L$Gvv9o}`Q85hlo!aDp!`17@ln-dq|FCLrdmeFz| zn0}Vb3VqRhFl7>4Ud;U;Z&LQP{$GLn^$NjZsTh1+Bb?)Bs8Z3qzhEYkgk83s!;LYu zB+y$2AN;U~x9IF?{O-{y*t|rZ7_UNjtGoxEZxGMsZi;6rI@KX}*&)2kzaQJ{O7m~1 zWkAEiVI=gd5KPZeAiwuLM+eV$62U5Kc9NGh>~YTl(xib$rmdhVGE3-A;W!%Nbcr0w z|3J1Es=%csCgkm^Kzw5zAFFQuh02{1z)q$B9)6F9Lv#Uau*MPJ{^$npnnGB)P42ib z{RA1@)5-d7v>+E3i;=@z{#a?bCt1JD6%z7Bp!m54f5K_(;|cEA-=H0o zg?hLsWdvS~Q-hhSq*(>anfOfXAAEZ9d{94U4fU(7@r^^(a4haH&g#j;+F35j{THL)-(g%zn+!_#(X5`(EF zL|5t!dGBUTd~}|ZfwpTfCH*ACzLy7^Z^`7@{AVf;HDlFrl-h*{)=^&Hi)$$|syBPR#J)FeCHqAY}jb+ULzWjk0~ z-T43FBkYx^^|E)x9(`{mS-V z+4-A@V29&Z*!VgM8+;9i2RmNk%Z*BS(ba{xMM(nBmJ=n>;xa^X!x|9We}N_}$|M|w zn#{;6Cy{&fsD1qt+B-o+n3K~_$~R8MeL4m3TvQb)TM}+||B>#E>Y))qo5`NcZS2a8 z73}_IZtacw!pt7aCZ95z*}m~fD8BL21a0d&hn6}fjwTK5UHe>jud9Yyrlf z!*H8CLB3wk$QRulTGd`h^t;mFL8>x_#6vK1|5LW5<_SD>wIDGC_SCiaHR&ymBvV|X zn-l-GL7<&7G~qfpcXEKsT^oUS??XXyZ6{f)RZD`_NWuMNWonmqgM7*~rx&$Ok?9(4 z^zTbUd@#|Ww~TbGLAhExrPjhx06rtsdOp- z34N?ROyZ(!+49+q=*amOV9zs0O%qSU>Ymr|GDL{Pvt_B|%StMX)M><}_3-qC3@ta< z$%uS*88Q|!&r*nrd-lHO5+$jG zM8uC$DV0it1|hRFiHeYUDusL2x{V}BrI1X`DGe$`DXH&%f5LU0efL@Gd7rn0@{6-H z`ATMmcxB%soNE}2(&lscQD4>hC0B|uSNtfox|>XE+ZNH{_Axwh^9`O-vF3ZFMZUP@cQdDGDsaWeIn;jR0J$I)%D$kMaAD*s+}m9V_2XxWqB;=%v|VtSwc&F& zFX8{)b(n9IJDny5=@QqLLY(5@3$SD(*gRDRU6Q~iu# ziGV9k>F$Kk<-4{5KiIm-1pNgh| zxj6gAFiL2Tfpn1q4%`o5)D5Wp*eJvwG4O-7%p5j+RVZG=dno52nBl4pGyCqff~BfD zTG%|}Iae>@EFOyC{bDY{r$_hML)ST&7CIRVPjqAKfffuldWVt&qL_6{3GeouhCaQ^ zdbm*5FdbCqgya_{3OI4QldkHt-OcnvgB3yR3>siOBXAqvWszy-La$`Kt6($Y?P^EYXDy9|}G(xO!G> z+$B2(Xz=Ay=A1@08CEf;Ef=M@y>fsj`g%S)|Z80dpy;nP1I)Gfn7qH?DCB&@F00Ao|s_ATW+O{ zVj`Qz0ux7(_cQP*9ROJ)xSSM*RBj(gR<#J?m3uzOo*crwu~|> z%V9^5I5pa5!*w_t!<~NXEdBL04RkHbi0oknaymd3lKc_ zU7WIgI-T8_$2urh!?u6A^h$08(M!;UneAPykw_6$6EfkNcv;gsj`5s$vD>V(r5)$X zgH5cr@p?gcA+Bz5fh5iNbcXWTFi<=ZPJDb9kWCsg+&MKh)W@9AvrbdECVPyyGYloT z2Yl4%+Oa;WR$EV9Q(jwn_vW%|s;&_C#znB65#n}FFyX$59Am#lz9c98SCVbrF>rpp z0(^`-&E8sBkC&GoU{85U)A5`w?9K@i#Au=d{Tn z=T9RoW*y8SB20$|I>?`38(b2UMrlqyN=E!iP|IVg) zZP}oH-UWDeeZ=>}VIt+#OTJxPihm{L@Z8{08sj0$H9arG9dyc|9{;}3yomYSx1|%& zGQJD+89Cy|USaTY7Cco*BJ;-NFmK@`G(WHmntrNt54_Oi?l@%6Etk=Q-qYL3%0&Z2 zil0Rk78;QI#nsFkD#86aT0$GYYID7_Eb;#KlTcrNo(wf)a4aX~LGc$=Dmx{B-Vr)Q zHX7ZB)eF~Sa9twDB)}cqr^}H|tv%$hTL+0>CrT6KyqPZ-iZJMlFG-*D44Mwsfcl_0 zIptMLr*!#Fn|9%q5GvW676oGOV)3Txbf_1KYJbA$PJBUHfYW zwNbsuN=MtUC3es2f>rGWyy6b>+R2|h^p@tHHO-})%{V}-EJ*Ebb2i0Ig}JZ9vb$WL zaJueEkhaqWf_d{PX3l@vkp{7l0k+% z!^tKoWpZS#8U6Xzfm%!e@_6YZ#y_#G#f;jd?@xYT|gx$K)lzK^XYE?3jZDEpPk z$leC08inb$-!62z2+)GxbIIL9Ms%n4L|PGEiItgx-(W3Gs(-?d8kZVV>&SejBC?JS ztlU3hpk{!-3LPk`rN-c)48qUjn z{e;YSA34^GTj8LRBP~cX=T7Ky0oTjVc@rB~!G<7bvN*7Wyt??9T>elX_)>(Rq@|qX zNm!$T#!a~E_56mJxPcNM)8=QlCJ9Tg06yHjC>zR6xYU=F?YCRd9y>6Vm&60>I=jco*~u-aqSN-kmLg z@*Hhw6Xcr|_8DT2q6iVBRM}I8(QxN$DvG@c2hAKwy3tUF3_KAf3vz>q|K`;wyebCk z^WVT>@qg^B_z3c4$dqn-vWAWdh0+a=@3JdWhR6}6MELl$j+{E0$Zp)iw`7$GOnmW% z7z#3daheaHdgL$jwyl_bRFMwFzUp+;^CjU-jwi#jH_)?pZRxCSHlVtwi~Q{RNz&(e zaLS(YsKcT)R7Inb)=BT9E#IZNp3ZOSo2=8=d43G$oeBcMfeS@G=V4*`UiM&5JdSnu zf^c;`BpvLgFWAQBRdxT)X0a>H z4-v6v-H^>QSixPREnUQ_I_e9gSjS472=cbRz8 z-|%bk6f97S0qfiU!RP?6PZun~OV0V&R;CE|D|WN%R93Pt4AbFL$YGoj<$+$3UEsoq zQi+b&q$lJOd8GCpw2EGm<#Gm8w`?j#Wi5oBt$SgjXA{zU3TV;e2RT*o?ALZ(D7{k) zfuVO{dFw=aXwGw@_TU}QOGgI6lArJ%ta9QP-I3&r>6oKVMGMr7H9&=74wM)r(&U@| z#Py9Y+5T0W${3W8&lU)ZQFXSV*n z6!r;~V4lu#m}0z-n#D^2di%l`XDfC=N)(+a*s-&JU0IpWf;tDmv*j{>u7KI7fcBHB z0($E-dvdZmv2J_A^iGUpKgMR$mZ-zDm3M@Ex*0@uAJ3*cUL7VRCXK$lw4Bz+&8A0{ zMrrcGYC7HU5-n-gry7q0@8~8fC>t@IJ8QKL_x^P;+Rsy@n%^>Lv$PcV@k0UIly-`& zDY!xRHh6P)e2?LJT#Dm1N|lp{2velyg8$cYJ^H&fiu~O-nMNF_pa+`fbGsYdxW!M) zm}iPT%*V70tH!RQM4DQS}b0T3SM3gP3Fk!y_Wcp$Qn)kw-r>tQ}ueiHIW=%ZVbI25D zP7T2`Kf?ss(@w71{RVDmD{%Aq*QrC^Y1%$4%+-(BPr|F_v-(HxFb~7!xQds*(b`>3 zwB`8|(z9+Zjs9j$1HXqu*SrCevhzNvMQgfDeU%`Ot4l{*rRiwLM(Vd%ljkw*3N5*~ zi8jZlP!|Eart@?hGP2jBQ+x$I9iT~58=sK8z^mlzxvMltTSmYuUZQ%xZ_w41XQ<8K z1l*|k9Ry%I*`F>-^co+LTgw{=XFv)@rCE^IkinKpVe%|kaF4Kyrn~#hsP_VU>aO8K zfBf1_i~k1GM{|#W;)83{FF1p)%Pyd+barr4oELNVhN^If+JC`;)){nKYC73{zlE{0 z=wcco=R(EO5VCB`b24x@liahN%PhCrPs39uaPW^24ANgYJOBNLV?<6-T*!StI(-4Vs&^FnXOX#w1HDVb4dj)PO7K7M=E&t=nJO& z^9o|Ntr+xde~~?k0Tlgr(1@T)@>OXG-u;}U(+=!VNZ(oW# z7#=%x*$;0C?ydcUlLb6m1}@lG1umDrlMsU%He*^Ec|0Wymb3SmT9cEYEg=Qo`(x0+ zcQ=;u2I?%XhtVUw{}??>Id8 zmh8tJB38I>{S3_OUyoW(ib+L$HAKo~!N^oDOy1bZ3Y6pQ@mm4pxakB&VT%kiAXbKT zWn)mbRRitE*Fx)3Io>`4S5kC;Hf)al!|utdWuIu|fJl!sI@{bQFUv*nS!_INzPisl zR#(DIU(>^CyU&IjjrzEC>?>}19fMDo&ZTX`1I+R!!TEAjg4(tH14Yexw!_Q@+rH03 z?kru}UT;QC1@o4LDXyG_8hN~q^UZMAq6n{5se*9YQn<_*!nETis8n^4WCuEv$Ip3e z_p5Lkea?kG3)oLO14EgP4-#~t-VE>;uvUCiFZ|+U4Lb^YASSU8+T?}!A;&&o&$^%R zb7vw79Rqx5KSi+r-h@I5(Rf`{;OAKW1)j%Wqj`58Q|X5lu=0Z+N;&?4L*Ja3gD)ds zUS||xx&oM@+(9O(AsB?~GkNcwxZtM~2Tn~5P*!IS_aauaW-m?&cC_|*Z;CkX{sg2p zRtH)$wRqh1|9CZmd55u@2tD=K6Sb>80lu)oWuG0vmQEy{A;BP}pNijKW)T{@kBNSB z4Ch_=ZKajr!n#eUW>Wj4>HQaDb(X93*gbRW7^PpP%<#MuxY6Mn4lj6s$&#&5SniF( z9aGRbbPIV^7eb2vd}A`kBbhD!%ZY(LkQsOF;Vv;|x`OS&p&)^`;!FZ7dTJ*5`ecO6 zv)@QlcOIns*Bdj_cDIl=6T!RJn?U?@Z8>`W<6!cX+1v(=7J6CsJKZa^kQP6;29u@n zII*G;6JHwggD$>7rzfFwlm8jIVBjD%gtob*8;Y@{yph!Sec{HXnS!F8>!mzDkRdz3cK}*z_i31&x!Ne^Y4ulP%B@t&B;BSd7&YhTxeW znEdnAbkP$zSo1s>pC%xap_xs2eHL_+i(qaoimX|dz#aU2F5fo53}4O8f}F#m@UfYI z|IEwmZDlVKrjy9 z9kQUeE1JF=_o08)oWV+R3wg?3WT*2Vz}H@Tbe)}mYE$=+B)830*Uoowwi_KM(o>V! zv||(LRh6&<}BcKsbB|_1<~(<;qz%ftH-j|Jd0W5IM$bica_cv zvS(4)5E6#l)jwk0Ss9dUU&n~pRYSvmQNBoCA>O(E3JPYGfNB0E&@a&gA!SAKGxId* z@$_Il+}6S8W@8LpJQZ8TgITwlN;c6|9A=mqsp(l{h3^F(j@x~1s4jXDQyaBV zWyu~0`bMDQOfByrKDt2>T^uGn=2<#B8=7!szv#_~6G3{$l1I$hk}h^F_a)Z_0YI{_G@n`W8L* zj3B=vVQGgbHk(<#FNbMtzrhIoGh!c|`GV;KbNCiFqWC|W)cO1sQv8`!FHuZ2nGIZD z$t>r-CAH@!@NJ(R#FYoW5Ye5s^zVfB+0=sWE%$&X)vKJ@vSLC?y2M!tY!|)QcmRn$>P!2dvRA&wR zN=U%Ft&Afa=ZpQ0yY=XYYh-klci8Ooy(36R6n-Q??u>J+4JWCAXePPa7et z%7FZxvI;a^&N55ho+my3RZ{zb_iQV>41)F6z=vnTD16L|d}&uEFSV4&u#FZpYJ1S{ z7WWvxDhYJ4NEfhp_lVEKIdrSx8CE3p8T&(gCc~UNN}CGp$Q=EH;Bx9Z+w5jb;x3EQ zrGKRO9!CoCsUVZ3=4wZM{`%0P|4zfAG6OPLCPSu=eqld-)T9qsSLW=x0Osi{2YR|x z@PK(o(&dZe=(mbf^nL7iy7FENZcr`7FV>}4cHsdasw3Q(Xye%^c=ft|jQR{|0HjSxDRV z`s1X{)ljl`2CRN6MMkBclW(nvHqKM|tl?^bABVwu`I9ii_%ckL{S)kd1~XO#r5J3S zh;DN}!qY#B@Z!D;IzIXVvvy^Icw;IGD~iL@$rW_ePLEnP2zsEOsG#V11%WRl2VMq$ zB(>TFtZGp?TU4-=?^`^L|96Qo|Cyx^Nq^Nq-Z9n8yZwYrjVd8~w0Dpf1+`4dsK7|t zRf3`J1!T6J0n7Xu*&G*sm;hq`VrV;UC$F2hWVvV#&>OF$YqE^`QHk@&Kd4cx)DWtD=9Hytg z=Mq!tJW~Cx2}b4%&Uua+VFX>di|sjZ?PV(UENvji?w?}bcyr0w$rR~yDSGu(HW)-0 zVqAGM?RYqW`)mFwy1X=&I4E|otE?-?_7}J60vg|O@>~EKYrE*&hRNL3F49~%6CZTf zp1>Y!+$+fS%2RI3DNd!Q0VzWjdQYZ_COYoKFNK~&$JGEjL!K~4OlMtbmCp zpRwm%{8+Cymzbk@kI6(CE3zz9j>cSAfl1L`5cTj3QC(AoPT7K3t>ba>^=<>XTKJUg zJ0C=wYfk{VFbiXD$-;-9*N7|^NNep}lBWNSY~SEnS9K{AeLKUg_8hpu364x>-n=)( z%O7k|#mpb>@d}96CL}!?6L6n)EX@}?M7L{MlWE(nDLv~!{>3h$%c8!qLIzo6)`{CZ zyDTNTu;Cr^VqiXgifdry6_3?&&mLiSY|Y>?>&%#B`CD+<{2A^aS`M+_6{v~TESymy zgs+XF=sTmCjLd|0MBC5=)Q6?0(*zm1p)Z(TlaXT-Q>IX};}>{FrKdqba+KV#3!$8= zk+3dyCK@;k@{g_)&{n99Jbn~Ky~jMk+U5$GoEiuAx5aoDM)Tp1+BH_Xu%ENmya(dO zq)=s40wdnI;ioqvkQP+I^zMDjKH^oQ!lY(4(aZ)sWCaU2HEq7I=o-H8G-1A$-CyWS z{0R#m$N=3mAMCar!50fzqz=)583oW5{0?Lyo{|Gsf0NR&SjKQu4$+?C0E^xPk%-l8 z#4EqB?%JxwczIG7UYB;n6|vs_;)}6HrYL{%IF@htk6oERiH6=a zwi+9)LUnt0Oy%lQr`!@|n`=dFNNkf;Nwpp(x!)vvdXt!Aaa{-(|s*VCWuYuKDwM_)YB<(ArLa9NY7*zq<9gPj?yoN|$IVr4Vn2MCM1BSIfLfe5J~sXh(HpEFQ^1n0_MC;9 zCB;y5?Yz~VsBtK2sI1KpA#nZHTXOEn17ao*Oyk<`Y+YFwqqFKb^zN>}6AOBlz6~s_Y62%u!+sukq+-?nBV{>4z0eq;f(3~@Vh}6uDo{@%;Mx> z-snM4+wH+$aW9UOW3-)Fr6}n1+UbCaml7D=OLtkP>?2?^l7dcj7?UpQ z@*jV{2w#qd7Tg2v-9uubAUueNX!1;1OIs|zl(dv;5q;zCNz!UM>$zsEqJ^2I&kE@&SU zjfM(Ngbf+s9q#XBXJ=RPd=^Z?@hb-buAc-K)BUKr?=aS?%Y(@KheUZq1(xnA=9Rj4 zvhU=sbJ&bm*!KGgDmT{R_#P!1+RzU#_uF7z!vm{%t77K5?=ENBA)f@U4kwkjKeJs~ zrHqiL2oaxjnB=e72^}f=Rz>+NDX(#5trsSefOEz~$bKQhfiiY>ixl1SNt>h>U*cs; zn3A=1hndik3!J$S32g(_Fc|o}rv0RW;5|E*bzXCbwS7`ceDq?-v&okjiO1V4KOXZ1 zYwj&3xy~E%lL_N$n+`8FM_LKr&17r^Y*D4^a=cw=jr#<16L0+!ILuZuQb#>W&D>@t zUp`pSrTzwFZ2Ms4un&8=T@AbbxP#qr2>NDt@^^g(bayawuvD9lxLgB&XcPF~E!pUO9JTp$ z5`8$J^oZHMNFKLJH{-&e(b)7Q9XbVf*A>#H#Bsd_l~B^^)xxq z{_0J2PV9oA-Kwzutq60_LKPo!szGL-9jJ5UF>`w!8q2z1$(+?NF6i`m6gp7XlHrdp zb{zsSJOyb2r@350VFogQOsMDwLL~vIo|OK;GlspuVRVgb$=KDYoUfzh9bf-`|O`qv>dIEf!C< z30$R5xj=&jJ>6AO>|Uc9l>K6i`qnoX$!#BSNoXT>tN+1;DGPy_9Ld|C_5qAqWYNoh zDOyiT!hdp;arJ5~NFEjD2fWv-b+1tb+d?sXZsG^Suhek2QUm7xJHVFe#4u~-a5+g| z_?%xYN@S9kJL;_X05hMMae~Hff%#iM)0)TLJGebXkEPlcLIKP3PJF_1I%QN53I?gSmr8c0laAa%ZRO-!20!HB^|G% zNXE3K5VceZ9~pjRXE|8YXY&@q`?Kb->_-x)9Z!U`#LEz0eH$!06)~_t8Rsf_!~8-w z_-StJ1ovZY6P6mcwz)2P?5e9J{du*H(JslWV_uvmTu$i$>a*!@?uHRZ`P= zk5>L=VvXh5uL=j@w{{*(&+g$#PIF~f{*`5yWQ{}O>Eq0n)ginSej|0ihmVjQM*rdY zL&fk)kTGT!w1anCW!=OR$doVl!CsYF03AuNE1M50(VOS&_O@lxVC?EP^g4eKAIr?di&lTx3Hwc{eT@=o*)L_j ztcYe8Y&p&3BuB%QzlkJvzlDH;IRs{7i{Pu+0Whxr#T@nQ<_WXv&{zGQ=ins7i@FiR zz9^{1&tlpbeoz87@naC7-OHHRQUv#c#+U+d zVH<6wS=nh_EVP^ijg&!{Ii!pynf8ylI zFU-HYmsja%iowtS!Mw2JFjHwFBNRB#atOuX(R4TVrLH3px#LZ0lke4fn(JDH&(FYe z`mE0QrxE);Z5O*dHM8zy5FuGUG0@gZ%-5F8J{T_IK6vy8Y+IdIgHEJ!-R#~|`KFX-|h%$a9 zos859Tk1B|kLGtNewPrP9UDTO=RQwZQfY3Z~+sf)bdyB=RnD%=S-bU(5dBIEoeWxC}Y6vdF2+mGYybwANbsT;eM2L&@+1@b!R4&!jK0Z&dl0-c$LP$#&r2fs1} zmmRf`dsv@;@lpY1j5!cqUnI$BT1ansog>N#LX`FFCRtLqNn+JQwAic$a)RgN+^&Au z6j1;Xc}~p4EmI-&ieLwr+rc|ju0U1lMN#KcEIY+}I$77!Lt;Fq(9v`C%%gqr?1D?} z#PVwtosk(%3tgj0-7F)hA53G8CKi(XPhRw>{!tQT^_)y~3@013Vqhv)+Di1$EV}v+ zC5k;JgSj;2cT_TCwI>1%_pZ~02(Q^eou-h3d9W42>##vC__lUFN=nXdQQ8;$r`6ysZ&*G>0T8wPV zA~rTBm>L1k&j{@R5y?v=On);(2~?DqE-q!%obnk=Oegm*n}TbSI~djdWV%jSk)m%2 zOz9DIxawgDQ{o*^IaQXg)H8thuh_Do^;7837frHqfuWU(bEUwCQALhPJtZHk*3#82 zZZyY8pRBO^!Ir-^r%ro{Q7`H`v=|9I)iWjeIveuQS8f2M+K%Dv$;;t(&|2L4vXs#k zuq!JKO{i$r71Vj{!e6q6!+$@gleeq!5iAqr#-~15fL;{|SXZcrP&dL6d#4S~-6dqs z=n!=3&A>-%!%DXdxSMGDr&aXl$^GnerU=Z_+|m1% zKXubCrzN49bi+<9{HA-0y|+_~bJXPtS$F0k>VI>^AJSgrYHBXqT3AZ9y6>e6k3NLS z-zHExaE&Sb;)qi9;pkE$OGOUyi2uGng041Y(sOAEvAl2)?6O^;m@^34dp~pVpFFM( ze@+S<^{7g6IzMhr0e?wd0e>)X8P00vvHrh~gNyqPa6G>lXC+T2XCKvq@`XUWc`g?- z@>iqC3pI2zJOuFxnOMKc1`nR>ffM_lfsVf@zW5x+L?3Hpp9&n4-NmDL00c8A4;hsA z(#LCN5zLO;lbMQ+li>7j1Ip*s3I5K0*kr(ioHuL8jN|voV{u7xui+81sJRLzOMBwa zIwQ0TkS5{IxwLS`1F(%LV=tILgjWxq!P@M1@KI?BoWJm#S>|p=IB)+#>s=OuuBZz< zEBnDew}Rz<{0CJ>yFu((7qRCgk2e^!}XLD7XgVowgeEHuJF(f0TxkT|dL+ zY-OB1#{)cH*pjCT%H(mS8}s(+T=bM5A}Q~qiSGJxGRNl-3M%%neO&`|<;K!O0mig@ zZVcrG#$d%754^H079T&dgKo38JTHOJwKO3LMlVjls0&L#eNF)NG@eX5S#uKW^AX`h}6 zM&n514i4En{|^cE^5^(RdSkm=GVJPqMIWxdOW!_tFL0C!a-e0$7+vXBqGCFYdu%L{ zj$I0&ny+Hmo-01^%et3&wbT_<2SqSAXgO*&RKbw&6j-7)hnDQ#h5x1GAhozl9fFL> zY0gY+ot^;v7ES!xdJ%3-mY^?o=@AUQK|SqMX}M(qzNT5wXCvry_X>wW^|^R;0*^{= zoPY_sPe4f2k{Zn44p!TE0uJpNYj>Sx^?dGfcK*6SPR=?;HaYW{0nP#90Ws7-vzpZS zZ6iYa*W#`OTXKKN7xv{)F7rD~jWv)k1Ft40X3ffB6jkYfSfwuVvUWbW9SYmkKqxqvJh{Jtm}_J})_*M^rhA;^ zNXmlJOLP1x*+Ne8B-y>y`jm6x4pG@e@RnZ;dgk=7v-jyzmpQla+L>Hzm@mzD+ue@R z`<`I_F$#nTGSI zQ$P^CcaE|8W?@3-pKxZ*n72Tk<+9?#hC6CaPHqZUMN2pi=TJp z((Qdtq);OsC5=9j(lBk7f9(b9n9@SS)VSoTP!$39qbPm!5LW-5$M{=rVwzRN$vMS# zwxjVhd9hA|%JpQEzYA>XXa6xKDCMJ7=3*^!PB#i>9D7}-x~PV@xSb$J5)GN0TSmA_ zE}r9=9}M)z%NpwvVIi zH2;&Jqkk536m?KdNCGb>4D!Ocd1Uvj4!TxEl+6|Flm zAz$kh_(TZdH_f}yI{zOW6`U(ceWmnjxfA!+Zwt^-up#0rq*Xl%X?%?=(kKXNDGc{f^Ba)k&Ax*k$bJOu5LUJ$*yj;Sj3pwG9z zpyMK^t#lqM!N{>@Vtr2s=bh+hdz4&gwt)p&oD?A^eZ_F6fWfM*|3GzECxWS^xa{vX zJi5~!Z?w6w@>{u98nQ5>(+Uxkeh9>+fyDfZ4i%Q;&{kJ%uG?K}; z6JEsckl=kyDFnTLh2b*cn^3u=8(b@W$bak4aVGZP<<;)eW}_DjGe_Iv*g8GRsszs_ zI)zGBX9lf^^{Ft7_G@EZ!&}&ffn=DpbuldT_((W=ya~_h9C;h;hTd_Vuz05=iQWON-YjBYsf{x(?1#}mbR?SUC4`BsVgjny zQ43MRntzLd_U*@D#ytc0nPbXK(w#-R<|#r)ojsMFa)CNQIeC;|O~d}GGE1MWV(pF} zwi?xqW6D<-k|iPncJXTws&f07{*gMXKQZy7ut$RX`j0u+r96b&bWN5@1dNa))=A8f z!9>`fmetW7(t zTuJK|1?K&!CQ>}tf^O9`g+ETEOj7+*;?P>q?txljT8HXZJmFWG^L~RkKK+ z;9L%h$|lR+iP5fB3HssUG+N&)0jt}s7|GcooZG|}lov=r<8v*mgfAaB1^&y}Z>#K> zkRyF0^F}H$gd7s<(N2^tIkv3Dg zkNrNUbq$A8PiR3-rTE)Dxoy=?q--rBvO(0+@srZ>+ zBsgD}9v2!V7E9OThPW%ZHLC!X<9E>I1+nykj~De_(bj4i;ym=GZ-RgOq->;sNeM^QJCF8Y3W^{lE zstL&R<15LJ9VW!#kQ}Z(5Q^@i8tBy9E67^fQG1zRj9T_}X5twG^4*EW1gj&M+2_M< z`*VZI>k;hTb_>(!%f}!}Ns(FGI*nd+Xe5m$_c;QK7U8@XCg^(uzV9Dm{@Ztk*|Z5@ zlZ7Lc+!KS*z1q|}%8?GtwZQdLhOtEWB_=eFlP7&z*lu(cmPr=TN#0qwTCWkm>6`=a zX_WnUy^c!EKZHNktnkK)7JO>9n(y^kgAZbXe9xM4lG5Hxn)Xa)mxym?n3u-5E4K{7 zgb0Uqilcu!JDCty5z^*hLst#)VDsj943%<3*S!>H?(YWa*zI`t^UpLU+#X-df%VB?o z9ko@TfN6o!xJ*?XUin0lKL!!x{Z&^cB*zI9VqGw|_YQf`Y`VBN(sf51}H$ zbi&u)U^HLAB6?_0TaVZ18F&Qe+jIDtCf+o%KnQlcn2n3{EYUN$7N3b+!o$-f`L`XU z`F&<{`S+TZ;?_UTw-(;oc5DlPs8 zHU=A}r1D;T*v9+vS(16Z{50O$`U)By42h4lG#2hlgo3T9oK<6T{4uEs{QGk^U{|&p z@g0wcO$#QG1M%|g-{g8+J246ONC#u9aXIw75zI*;f_Hf3QpVxgFmpbqTyXi%!POXv z$%YyzL#+r_i8@Vt$MZwrx*Ed9?-TdO;pJ!v=fLsr-^ho#}J(d88Cn1L@X1sMHew|uzj3H*1T*Wp{vf5c-;w9y?-@Lw>eBT zHuj^YP789PbMW0lefoWKAd1|U!_s0QzJ>l<%zB=JVbkvrVbOQwhEg&WaxZ1o19i|Y zl;Q5P3EJm!e(`3O~pn{b<#4F0y&g;j@(Sh+0+ z*-bMdMpp$yw&+#6}o(%q7S^9|GeSch%!#UaE&7wVudI@FHA(>n?5 z-F9E#sOrJ~<~XQ$=fXY>GZDCK?IAuRkJ;1kmdw19%X~DsfI3z&xco~heBGjq;m>Bl z^4&jioq-UFi7vs7BOX}V!lG5_OfYqq<)1fj;6Kf_=5MhZ1dB^8yoj%YS$w}15lj9; zf~+klS@@pgs(8Z6?YK6_u6r+Nr_0lZqBkVgAdLvNF1RPCABx)wFxpTOSH9VZZqG_k zwx5RDR$zE&L%xcl)$J%ngAjVp0&! zxM@fh7mKle|Bi6%#|8yBzX<>A1TTEGvY%bM(g?jbUBhYnWby6gI%t2a1tOubaGT3V zpSfG`d(~n3bE-I%U!+d^#H7g8LkY}+f~~~(&42WOlqW^0UG##h7l~Cg1&8VztXjSV ziFXbmZFB7DgZdV7&Lfjuw;+@Jp5TD(A;%!>h6UByvXe&T{ieSCl3dJjqJGMc1uhIV z`s=7RJ^r(u26KMWk3pim!EZ%uxb+ftrP)MMcHNn9%|x)d%7%G5|06U1qFdB6 zGt6Aba1mtiEZF0o&zXs?4~h24sU&WjAJxxjq-#=d^UiIJXTn~(GR60UIN!PN1igrp z$ax1FR`S9xQY0?P^@(sN-^mWRQrQP*x9}Jv`$|&OdY2Ro&0d zDSFK&$Eouk#5j_6F%)JQ=?~3R8;2B5u0>&Y_2Nv;od>AWOWSLB{@zahM(8H zDpq99kKbixc~kmtNhvv`H;GNXug5v!%j4-UJr2^MD|p9ecJt&+=0kA363P4%LuBXt zAQR@DBaO?;N%x%pXj zXou3EPm!XftVAdzLWx3CLpb++pHe6qC=E1(lA_XdqFp_WP^x&$*4ZFHa=@~uc%GxBWSQ|pS~A}g z9*O*L_6vE~__;w~bf@wDEs-RB+-&ys)lb{4r44ZRzd4}1@d*jKp9;kS-{y3q8SDQv zjHRyM$>xa|JPqiBdwyl%VQI8*SM%Wuuav{im&5t?@&z1El=Hzod7zP8%DzZzS%y;! z#Gcv$F;P>o@9TjqQ$UD>ZLOqVs{XObV^lHI!IdvvxrKRM2x12V2e8C1H<^9>b0NQf zjg9T|6$^eY!#-}t_`qZ$oj&{-=&oM|!<){N9fQ}v@j0Ia$8r~r)Ne&s`)Qc8B!!Dj zs`y5s_OmRznuWY`U=Nbjxv!ocH{^$*VAxV@cpzkc~Py(+$a+rC!De}>Z zLKnU~0tcUqf+sB#d2Q8ZJ~?$T@Avc}G}bWeetC_zc8+GCFV9MX1-JaBzfhdh3nmi+ zV1v$Pafa<2am#I8@w=6jX>}RFSzO5f$po;dRY#cWuP#;cZZ}MqE-XZwpZf3t@H$D7BiJ8swaZp&w{91e0`9L@$k5jwQDj$z^J80>wrid6^5v7Gla zS=@oEkk$7Halbm2dd@$Rx4b~5YRFOdxih?IODf0Y@3pNQG&%OpFdKLh`! z{qXNq2NXxHz)?Sa@WAOgw0ZP-k*st-o9L*7u9mWBQ`ZH%M?VobOe<{HlpJJ|qXRH8 z{wDTVMdCNrzihJkBr@eslO)S#3SM$?#(2ARSg5CmeT#?V<})2o6{>(LBhF&_wIr-@ zSxD7l$FPgGGQ6_zKh#K<;IBVrsCG0#@+#Q?-`UCH=_AW9Ypy!_Z0;tYM}5ofz7Qx41JtqK(hyd8%4s)eQAwh!r#L3UQ>^J4j_GsiahZHy?5^m8`5W@!))T>{arimZ6f7d% z(hRn3>jhY-ZVFyKs@P|SuD~N*4(79hIjL=6Yqtb~^x9AijY-3n^T`;vz8+j;O|a&D zFuQ+gEFU+^4_^sfQf-4}_}`b`u(q%c-i%#>GC-^=51FTaWiH~pph1=H?0d+fBMqC~1wTgrfZSM)!*^B=B zi_uuJP&{9AjM!kY6gL-cBL?eDU|Pux3}=u zZp8`tF;IL&N#vq%74ma_gDlwzGLZ>H+oYK)ZcB!@7Y_@2smpjhB@eijGmN@>1y1B% z#A~zeqT|eRJgyT9M;1?l2SRS?{BkQOIp>PiJ4!HUMmc&-rJ&VfYok9z%GOo&<4s$% zxg~z5s=p_Yuon(^r}8MDnqt9bPG5=@|72l&9f7p#J!DjEKRnd19m>A|b^YN&lv02ycTsGe&09P(6h^PLjj`z!F@ z#_Ea{UYd&y9D|wF4;2{XtWD*nM1(X2i&W92!hbTnsnL-vzzzb)t$qe?8? z^8}ujsEYfFezKvW5JLW_;k^1a#87NQZUDKdev#i0R z*H9W?K3Q5S^d0&MoISM|1;_EhAvUV)v>@uuas3lezdIo}WF2K|;0i?NMIj(s)4v!BhCXpYW zuo#ahes_inPY^ij9&e^`mzZEWw{;#`AIU`j?C03WxE<|oy+HFjaj197fLA0LFu(ow z^z6wa%*1*k>{}d29M{#dYm+#8yXkYW^qn00(`dv@RMxYx?LpkJ#DxD%bB2Lu8_@gf zFTDA+9fz(TPW-;-(v(id4p$vUZx=&$zQluNcE4u-4e4O@$!p;LTF7q8X2XQOA;nOWfrAlnehk}VmUMsTO%|u@zG}J?79p4<2yn1{#eMGkCs(r+a&`UIBhgmE@}a1^>E%6BFA?hJ^*X~ z+{VVL(KrU{VB%pr*5Pr@X0^zcUhYn$eH&YdL&+hWJO4XT-{ULH+Wy#V*cV%tIxx#M zMxsf>yJ6%beH;;)0%q^^_>=8&yx*heboi&IWZjqBw%90-E2fr9Qm-V#T9Z$rtEWjD zMhWp8uUPi;!U1;1{4o8wVj?Z?vyuJt3Fbfi)!0NKgD~XccY(V-1YN!rgYKd@7La7k zJM7Cv2J@ex`k9kBeo`}y4tWYbwKu_|cs4b;{YO;TF^@j>yGuQS4ua0?cwD6~0q%1E zcXrJ~ONAQ{r#pe|_cdm}+x2KjHBj`8&^pCCnF9&`#ZZGmZA_sXh zJ7CuR&(ylRg=(c8g}|;(a>_l3Y)RJ#-+lMMKg|+XR(8Vmup~Utb|2PRDr1o53+N-; z44;hL!K!H$;jfd)@2d|1O4~u*q#9-hU7-3IzF?y9o6Ly40cS>7h;}F+fuF%ISyx{% zpQAR7J~%B$mVda+W^XcPD@uS*t8Qlw2U^*Kz4kPyT!$wWpXaIX%rL3L6zl|s^j~Vq z8fWT(C|?0zps(mt;8ZrDWde>VQsn3VcyqfQE7FmTKv-%MwFXkseP*=kN#!F za}{5}nTn0nKPa7y3-)Ci;lp_QzF$mJyMvrtu%1u8>{B~L+< zs}V5u%w0AoI)wT5j$*UV3+(p~_wm6Vj!#PhXv^d8^z7vbhFAYU|1FMqZcGx`)GlF8 z?i$QiwVoAs8snYt5S)DD5h=KoKt3+{&H_{4vgExbto)q^%x;N=xjXvu(PR8z$EFX$ z?yZrO?p;cUMD@b1KAwE=^v%37-j(l@&m?cP*0cSe%y{!nVbsw{#a&Gjj2&SC=7X=1 z%I5}>4z1xHzE)xy_3hS&PI?o ze-4@D{}#P9KN8t*>uK8ePw>voi${bzF_VPvFth49L@RuSCN(klZO>t0pVfG~j1^aS z=8Ql5ZsKh}1N=Vxm}ugSB6hjckOv9x2bDvv2;Q(O=r_|F%cA$<3;#T3^UPE-S~7r! zYDcoq^{LRQaRLsV8ieOoZ@{YGXHk9s6QTEOiN#g!s2^2Cw%!(gX82NsyelBxrN~6J zc|7!&9kzdW;;@rOnFzViGZesV4wyLSyYO9z*ZMZsirGDax&)d0&+GDnVVF0?m zy8*xcKfhPKhJ9a6Y4g7*$$=jc+^?sT?4ka+x6d|wQg~3r+?v_v??+(eWg~do;R>yf z4&dMsPvLNpEj4}A#J1{}lT~YnfyN5q{ZB4IwE03PUY;ZD4o*e0jPvF^+^~ZMsvKkG zCG~W^$#kxy`4OxZnBv=|(QuntfokM;db%Zw4;DIstJa)mnwJtp6Iupyy|n^E)#5j9 zTe1=rzprO!`d#K__V;)KS;UKb!{DLdg#KFK1APy~z+1^^Y%p@8R+k-ctx6`U8rx&; z+G|WF`4%0$n_%v3KNhiK5I(N^Ez!BF#8V$l;O**ReEt$6SX=j&=%4q16`jj)`|BPk zogsrOofKGkgpEihB!Vr_xDL)y??EF?6;}LtPVUb*2zWhG$N|cs-u6MbvA73YwdbxIQKHpT)k4F!#r;mgEMOt;SqCkxbTG6kcXtZ^aZKRAg9F(h}-)!ob z=~MvesZh9+mqpGvWsz*>LN;l>3cpj{%C`Fox3F{%HY3ddi_ZRp1vP#|_o|wW`4k=2 zZ*zdno*Ny2CNWsM)DkaA49LSy2OhRBo}WtJ#Ede$MHyehA*ERt%GX(8hN=;)Sr$zv zJs(6`>ZH_SUNSe2F=q)=7P0bjGwQKxBpA(FLrOOElJTxaZ2mDrp<7!c@_XFEisYqq z_Qr|$GqNv^{eBBV9=s;A@-I@K2st+5_)b3b&SA2r#FtGmo&o_1A7R4Y2N35dPoL~> z1;r#agx`VKym>8VR-Xgei*DGX_zJpR4^ygooM{?ApuOp*h}GZ4{6P<1*BiW9Yp5yWI zYXxWvOkgW4%2?1T5p(uDz?W-uvkPlaOYS=j=T8d~S+)Km`qI-44C;{uH~u8oCgxzR zeK_ijj75{JO$^4)k{G0xlKH~jV3+DA2uc~wMz-Z()P!6NpPh#`6X(;AzMkw^QXsY4 zJD1Bix3hqKLCnW{8vh-%nR(5>&YFduxl+y+*qo;gS)-LCuQxuTe^N%;q{y#?=ZoGF zozHimDju+zqP3zQB(Ow?T1)ip0^OK9rb9GZX4%NRQ^zZ(Q?M;jLwxPVTl7;1;&vBfI9*=G zv&`(7sIf0o7TL0uJ=Q$aU;|%bGz!=WbvzIu;&j_;4*X?iVtY8vn_|8gEq=WthZIor~uo zJp#vw_rrE8fi<{%g0S0)fYo6aiLJFa+SvRD(olf#Z3nsjl!xrpvM`qSsREbRnu=3% z9-+|UgZ%94bm@t2Y~kl-&|WnIG6IjV?bc109qf&&%45(-XDO|PltQ2qhlEVNN^2dr*#DM2x^)2uH(kP?JJZ@(~nt?4gtHNL(V6`lJ)w4Y^wM;v`a7O5L>m9y7b zp9weEbMHsoQAI_1dbf-O>^>oi#=TJM8&VMCv*XUZC-d-`9K4r!8#KAWEt z1$z{dntC~wUfIjMmE`%n^{Kq0|83qL%(*Buhy76y?y4znJaoZ7+u26pbi;3L*7miW z?|rgb`tRHVX_1A2$UwRl_9^*5*s2}8(&PY7KX^s>Za7Ofv=gav^h^HNYBNs|ym=!^ z2JxgvnSAV@cpl|8nNOduB#l-q<@xPde460m-tUmg%hwEL>(lOVpVAQCZ|H1(?~@Fq z?%X4pQT~|7c6ajqo2vNHW%aD8s)v4SR%Zpz#_-n>*@eOzmzCdzQGe*mm6!h<&r0Up{l0;dFKGKEs~BixIr%M=OZV-7voV z!(9G5Y$#v8B!-XNVa>I(hVmWt=Ioez22=mhBr?d=<`ZRo_=}on+*(at;;|vSc-w>z z0H&(gw#bdjG+wf~KKB##9(YeAdY}T)`KPJ9^ItmkpCZA1Ei_H3iv4zqXCs}?L43j| z$o2mR-$uH^w`+$HKs`c9s%i&q*-;O(3jU$<_&r&L%g4Wl zxy?10$o9}A;>ZU7`bM`cYG+~DQfeEN4I$;BaQT8EPBHT(yKWeY%$BRO#+SJ)dhI?q zrS(FzsCxmloByQyRp!Cd=g-0NV*-o%*aEs8!p*_<^+_h?~BK|ai=K}RySp%0T?7S9G2_T@6; zk8(X}y6{fB8E^5ccy8$uIA-*@czJS!t@F`H+ap@};I`);Om@{E{pQI*!)9A3eX|L! zBcW%NUXs_M1EQeje^j(xlP&!73&yMxct)mj{B3Fxn{s3q=;xi1Jo@^IC2!hF^M4e= zwOyGwfXZP!@q!$z;gc)vLa zCd|!(c~j0pYSBUP8)Jd0(Svc|4_o4UtuO0Y{!H{O>NPmL`U}dnbTD$a z7UYUjG*J^s`H4)5XJw+WW=t!|_IwV>H{#ImUI~0!WI$)_xj;LP>0r;kqqgT%#xwIu zeK8u26#Kj#DIP5FvpWRe+r`0#^jGpTtiASG=tEz_?LViIuFLz#u?@Ml!YLWPJLnV1 zU}58KzZb-cv1Er=yhuS|DlBMorp*ufiYAn+k;`$L=!j8aq7I+csP%dY#$HCVBJlSmEK*rly7uSaMLEAkJXsopvfBwAz@q^63{{9oi=`1bAnVw>v$V#BYwP_g_yz2cz@!Qq-XO}&#i zb;vSP+J*|Zg78mfhOPI$gJ4sjh-Wq9$=?T~M9(AN&>L5E*}sHhFxhPnYs{CDUm1p& ze@6jxmVN=PBy&ud>H-nMOp=)%i~sgbLHDq^_{l#UC6kn3!N7ZTmQ}n>r`IIXrlUbd z{O!l$GG>v1AM$9;hEt-yiH)|Kb#g@$G)nQ%U>_XvyBR-B{03wE{E6p(xojg3BQTBA z8FQm;<%-XP`eS)ek=~@HMp95Pb;RLWe~IyGSCRSftCHS3Vydm6jV(8)V#D58kVaI( zc5xrfd~=la=PKMve4FN!<${~cRGg!^8d@gm!}Nejc=ux@?Y&Y*bSv+X59C^LX4iR} z6v3l;_l_!@oLeD^DA`0y7Kmw-izU5M=K<+!#c=n$43=&a+&~|PQIqQf`D`ObYf_F$ zZruGYI@tD4G%Md!67%-qgi=c#+8Hg}IPx=K!8lLmIOCjXYDa}gZTAvpedC+t59>>> zXRSsR8Pudmdk!q8&wX9VK-W8> z4^0l_*MfWG<@J%E+5eNsTx|%hTDSr)TvEjO=buAN&@q^rVn$o*n?wil<`=V}>)~cu zAF`cokW8wPCsF>NVfV#-)ZncP)C$al#lO4>Yd^wz9*l+Do>=xR`#76fSqpg+DxhuP zF?4s3f%0dAz9tl)f}3q8AMKkT;t!=C$>NQN^F(9O#evahaiE;yGtEjkF>S3U(@ z$wMeOx10|2QeewGzmd^vEbz?E?bz@3ZXDSYK;kW!=xdFWWO1q!2AwQ|a|8A+m%24w)=C7}sHNL<-E2s^V-pKDx!7i&H6)zW@AC{IIl zW<@f{{i~PQ{nt*sCL{PxnF2$!E{Ps?UxpL!iik_$By{fYgY9`<(APl>hDGUMsMD^-)74 zkxK%V>DSRe^CX%CO~>{>`-rQlJ}r752TyZdh`rNto;7Fzm(8|dLu_va|1%r7q8ju#jach{C7};J$R5QS@v`@eKqYfj9FrVQ9+Z@ zs&)|eEvdz6UL2>WiLky(5AQBlB<%$%cz41#P%R5Y)zZbHO97Eszke4VIcXrc)ndW) zemiEPr`OP@@|`{aJBdzoXs?b4DS!H z*(V*k4x8b{YAbY)OeP^)=AcC^!8fCdv1G<^EPL23x%F+Xz;RAxn>`1h`>GL0h1}`+ zMoU~Cw+`<`I-x<(4*ce@3qQ5Z#D?^#_{dL$*;;*ZrNDr9{OB&Zq5i&T$Ca;8csCO* zKjsTgc@?{Gc||+XL`A!-X*bYvMn0Y(vv9>PP4Ljmh5H-Kao6ha(CsLL6OYAWjMhle z&&(9SzDx1k{7rPJdJA#<)k~ZwN8;j)Pw4UOEk1OrK)HL-Fg@@DzX~T|CYgtQ%kR^g z{B*d}^ql2Poz1teu!P=5IZQg0j0=to!aDn3Fu63BooTMdn-|mY_O|DE+&T*CHjZQM z)rq20-Kp@&(H8c!>hXz}&Ol;a7dvQ|ETM~@io7+aK>eqa)c?j=XrHtVGzEWySX)Qz zJwjXD89>CJcitDRu0H{~msBu2O-)=CG*av^*-hL>Q&}A3V~5j=bh%6XQ8HL54Gue6 zVEheb!8f)KR*$yhev6lY-gbmxZ-ODvrv}%Z&4!6jV_D*v7*W34J9u#a6t?_WOG2~s zF>3pG+(&Y0bd)xmed{dCD%vUfG-&|njff)mZl5HEyUsw}8Uh~8=it1BGia{3jAh}9 z`08>bTi`jCm}xvgmD^kKzEcdfemhna8f6G_9h#UhYXDSa|Axf0I9TkO2`}!b;ec{^ zT)kxpdb!Mpt+%!_)3<%dL1FjmxbXscY;clwzum^7yta`!UZ+_6j(H$=r?U9O20KWc zR3h?swdIo>28f)`Or+g1hoNak9HjX~NcKm(2iKhk(Qaoi#DtE(u1ANtjO%Zr=_Nc5 zP6`A*orApZ_aAJ=<4HWuXc1RwOXT5wPV=*(T()OcD#$2m@n0+YGwrBiDtm7d_u81i z`pq89%XteqEg3+XdW^}%{AKLamz!j?h7R^gy2$I){CHx)JTl$xBl{ec!Hosh&nzCn zw!JH2%{`h-x8(-wbsmCAs#YvT=m5Ogw1N3+{D1Bm&tC>lt(J5CH=93t>`cRcvLxkvY4w;c zzNkY>dZ2VO#EuboPzka)rt&a-cKeWM{O4-+q~HQqPy4{<1Sv?@20Z49li#x`ub;5w zZ)4%B`DAEC9cDJXp6_OZ`Ka*~E zf26kHwF{nVZs&>D_j2xRCSq3}q^EXcVm<{@el(e7RMn0e5C(WT+jv3mGe?AI_%nxC+So0(qY zk9WmV>wckZhF&|#J3U5PSY|GL8Z|%~ZFGRVR*Yo)dmg(tFPq`zB|`36o;~;LkHc1( zlS|RJ!KmIH?d->(z0)W9#-W&933uV*rdpQhEn$1TX0Vw;KBM#QZ<;oP=(LL!P2=0MR#!!`!mjtzOXsL3chn@f5o2`-!Wuw#*JcUmb=4Y89SzR(4K0& zXca|FZ(`5x8?&wvezy|3UR zI%(TCZ2_+A4#B0DEcj}}Qe5V_6?Mjp$8?W1SZxLvv?LzWeWh?G`66D0dQ7a%2iY7$ z%pEZbdh3($^p9-(yJ{>h`MDF*lXhUyih8h&Jr8RImXQY?Arc09QsmzV8gHke>jx`r z?$;%9e-K12Rjos1KWA(oUjwCUtt5Jf%*aAdQ~0ZqF3$YDL!2p-D7M^r1yi?he138} zMy=9^zb%pU)tM!%YljnVIVX!L0}WCCd^l5%7zr}U=@PBZKVl`*^WqtuN#Z?MLWJzb z9!yftLcN(SP`D%tKll2e!FD6@&|A3}btIX+FdBshQ}t1adKD+RYvP@=IdD)hla#Gf z7Dt3xVV{W)q4nGg4nI10zVZ$(dv*vOnKug@g_}HC`3Qenrr|r?4LEn|Of+4UfaCHu zWB-;+lw&^ViUMQP?E-24wM3Zn76@~~0&@KNIq>fG8!1-H(HAF`eM0*eOL5t(X-sLq3diOEtX)z7JqwPKDfNrR zmdW$PXfse;P`#Q@oUw%)O`6I*@9Dvpm@Oor;*qw&)}9zzp~{UON%`! za!Kf=v25wMH!yI8B03xtJdCHFQx&tJ@L$G1axQ^EwZ#eA6L%ZBgznG|SAG4q$_;vSxX-d(&738N+jl+7t&FF`g85kS7wsYuKm=`_%KvSN&A- z>e(YQ%Jw9!D@uVopve{tlYy+4p|I)FW7^PN3Zh#=4o%3!+OKM0JA)pOw8S(z>~nv9 zbkj37kB5TcrvW%|Rao&Q42SURhrwoEJDk=VhEgqkw6arYdB4^8GYi3EIY)+@W+&3| zF704zIEfw$tDz6{MpJR3l+Ai{n0*KfV=I40@Xj46OwM^Le9Iogyf5|Pqn;$PrTc#t z4{N)^u1(2iGk1F1cw42>O2O%|`_KvK53=lo##wr9r~x0-@RsGyUCECf6G5-*cnE*5 zTO8A7L4RlZvXf>PSY+}AD&<~$NplkSIvb7+$NGwkf48CMVc^FaXL6P2kJ*Tfsr2L) zHFnffmYoT`#ayB!Z0(7)eCP`;uHdzweadKK^V^Hb>Q|Hbaos*VSaUcx%TeRK3q&k4 zI=1-a-bV#domc`hIN6hQBF`qEZl^^@hj~j%`p=t94NFn2>IA)jFpM!E2ufYygC%RH@7PB{tWF~DD%s9>w?$;b) z*CQ_S3jN3Y)iQf-=NmyTkBz6r5+P4nH50BXO{QC3#)^Em=FpBGDdh6gRKdkwCW)0M zLgMHcIL2J5m)UPBGqsH>Z5_wX_I{=si)$rT&acU3OToLkW)ci&N+C;L{Vt0ADF<1j zSK#j2i3t zRl)2x8GzXh1EPG^nB6!L$Q>g&TR?}?nmPBW>k%iqyYCkEqW8DxyYVGCX)7P-7S7k7R$wCT%%?UG&|eruwHNQDg^+(ec+%FEJ9SrN-~8z^;csNuUp z1KHp;8N{|!mcJQm!5+VP&*!)t;jMlLc+SW(RCn9}Nq5&nk=)wbw8<-y?VmG`MiyvM zpWmakFcFy`=a5&3Xs2HemU z!GFAz#GAe*3v~98^OYGqNVSo5{FdPz!d`afk=;;a<^oxHmJmIlofMXOfZyLY(7M4D zMu$~^hI=Aeejt=;<(iQ<8}>+NBt%M^9Al-em&@2&_LVI$zXk5Ec7bG}5p*c*A*(eF z@XgK#_-$q;j?uZ!MrFKY7h;C-Pm4xE(SApmyJ;Pai9g001(xHme+K;8lcC%!ERC=G zFNeGNXWCvkW5F+kj^;;dXK<^3iu|jpH+S)P2}wd%b;yoylGD$PnS9k4_WOM__5UUG z!WKjECd~Z;ssL<-YWI)dvIxAxbyYX5UeyHYw z9^XOdowfqUm%5UEHq`XAFXdCHVNghAlAI{Ib8Kel*W8T?9dX7b^8q4156 zXNtKEuZ9S_H$4xMen6FFHVA(5H4>ooB78d%2Dp6;TC2XIciVi(<;z{5I(<1g$L7O> zpLgJ4NF2x%rjagq50f%VS^S7T>}7x&Jo+38hh$w~z%f(40v@r^R|fGMLs`Btv4w56 zoy%Z|2I}sS!2AJ2Np{{BK4Yx~eYLz9Hk8kRRE;3!o%EblAFd&ToXokjriP}39)yaW z7s*hz2Uee-%6=+zvw7EVb62Av?)mlz51Px^sUf=rA&3!QcU*8>*$Hm1?Z(*lb{|{v zrV~mBsYO(~EBExN>f zLZet(_i)_1?FrlxPsi_Tj*0d~pJk&hR-Op{tGYr+@B2{c81;_LAb>22K)0Ok*D_! z<7;mk;NVv)nS;zUCa0Lo^Pk6Xn{SQ0fIE`!gFmp)H#?xloq|KSJf3Wm#T&2lh0NC> zrrcajCcA%wcg|fb`<5r~?apVTP*E!VW-0A7yb14Y2f>mGfpfm`u4L?IU$E-DDPmni z>D#b;c1PBcn}%Ix5q*sKtH+IC+2aIHigI8@wv6D4K|JwQ77p}}6+N3iP`cx0iQp^U z&veG5kOzmnMT=a&!TN(O&}Zlb(DuqAPba5=OwCC!snLKWLj~#j-ViRcaIvKQks{j= zeTI#Era&v+94~qvU_~Z~+@Q{`0-n|N(AK&q&@O!-S*P?2a$6W2d-aFS)|$xnE-+`d z%PL8!z-VwE7sOY+`8;8Ck{l=p>cc_N95P=;n+$OZg?`Giko(x1S-0Hdm&Uh&ds7`9 zZ**1kua%Ja#;tU<#dC>q0VS{hd(RI1QAe6=fz+xObS@RbdOd;L^SG5A`)$RiwTx$H zt5;L6G1m0z+!&&D7(~wHDr88>P<)xHi-P_GJK{UwvC!95Q_6z6YBvn-8GyzA#^Te9 zrWjG;&IT>NNM&<`JKW7mc&JYB@5(VyA+JMXTGJrtiz4RV$c5DJgRptp8d!bV3e|pH zLF?iRlDW&&*kXY-_C+QYSKpqDTQ~NHz7yoQG)WiMo=Ik9p^;1`d<_GoDWch%j^Hef zH8}CD2&Z&!Wl3FilI{m*=&eR6b*b7xe}#upRi$WV9=Z~;{=~EL$fKe(;aT!7<`|4^ zvSaqsP5Cw7i}d6VFVG5e6`t)aZ2ypO7U(6%r=R#vC0|sDZr@;(i}nI5xlp#jP64uK zYjXXG`%o*2g4Fvm)hbb==-Z#3&de806mAXgGP`MTO$dR1pXdlzfgkwr23Z$Y55q6G zQ_RKMb?NA1ppA0rNvu3boBvqwg6-LV zh`lIuVPyMZ_^)RHUo6a1nU-&CbcGq&?MOU!^(0{k)hJ#udWX;*u)}F>yV-%Qd)U|R zso-MlkKONLuuNV8bG4$`o9098x{bgW_@>VHRSe=Eef0TU-($RPo|x~uE$nOklbfP8*}+&C#Uw8_cPH4htn7 zpuOn@e0IK09cuKTIzE!_(hg;nj!k6Gom}v`GMtd+aZHqRlRbGRbcy%c7n!UZi*>ci z{MfKowr2Bs$**-2ploU&MDTI=X_*JSY*j=5RrS!Ze=-_;7=yimo8eO*AqO_(t!=u- zWHxa^D~{OQ2!-ENNb34Ja26c(l8${ONari8j=Eqo{`^w1;{7{tQZ~Q^4Fb2wdI^>( z4naAs2e?Mvfb|wRLT=o3QL8AD?lAmQ0jY*SP@KAX9@jf%{XxNeFgTydu8K5P3Vd>1V=M5#$SuT zE4r@Yn%2D#eBn0n8vY-#p7TZ$a(^id@V$>_an2~?;RBUJyXowL1X$H)kq}BOCBEYg z&_u{rUvX^2YXJRcU9qH>DEI#cGjY>ePqA~^RB^iM9a#3r4!WaLXvMb-D0SE?bl}FqThSy8*dl}W=|dr? zGnefezlo-XD4_k#XiU3y5JT1;#ohXTf`_ezCKp!Nu1Vt9FJdTO_`RN{yxfL4-xHwW zh8oLznhr_@Sx_)F2Ogc?2bO3JH5H4|r@JrA+oXl@Nxs-REDN$-3mCp<~%*rc8Z-U%d{OSE*H(ZDMw~F_a#41u7^GK_n0!B z0`}z>MRPo4sdr2Soio*fbycLZ(wTC2Lc^V%Pby=_jpZPvS9rg47npy8`h&w)MfkZ; z1w&EjiPSXWzFiLD?3-1&!e-Z=*gt9IkQI0xMB zaU0(3qUbfPKaNt+=SJpNx%=XD{!g+UWg<7=uEplqlk*I7g5yX|ODOhL%%L(JrtC(@ zFIb~z&1m>HN#KxUw0YcE@r(2Si4Ao>p@ZiE2+9-K^90DPE@krX)OBW&H3}Tslm!l+ zBL6W{pAGHMBV(7jGK<2GTCU`Gj4%)|Gh}Ha6QS{i37pw_Rxt9`)A$cF_@3jc{6o$Jv}|bROUL{bOJ=mdtUQEw=4gW2p9lQUuXJ88#Gi*%d?&2h6f%-?Y%4V#ptjx<(r4SSV%at( z<8}h3ee8mn+oljS3WK{gxx^`I7N=Q%7;+*6#{sDJ5=fFD) z`@r@0R3hnYLOk>rB$;ZQ^c1oqgNTsHImGV=&*prfI$tXn&9)x}rre=|Ghqd|)aOMo`Rrfvacex&jPT&&!w*tF!K0@Vw3Agle#Q9Ji5UF%JDXrT zhF59oa&MJv?ASC5@6EVOkF<{9iY0n{$n4#0n%V@W?Rb)X*_g(gcCVuSoeqHely!W$ z(Endi=YSa_Et%}3Ja&K0FZQ5j7fW1x7`BXBg#AY4(sL%$*txb8IO&iLF2_fL{_a)y zBkBaczjhaW-~NYIFE7z|QMsUO`i0dd7t?J&!bs~SA;;km%D+|T5e2P0X0^PWJbB$i z=FH7xu4A?dhNvXGXVd^wUwXj}fw?lW>N4#Kw;_oSPLdx(LecB>8tLX)^Q5QiS4n@T zhO-(L!rm(im(E;6T5sN0Qg&io@w&UKFeqcBc*IwCv7@sbzuA4BcU)EDW=H0b#!sm* zSrg$yWQC--@oSNRB#$=T9?v%V$>VYDQqi!fXJM0fGW@XfgycaB+3m}H_>x0}$CvHJ z{chh;^fV6h7bwxDU794WZeMXi@CUY3dlJ~sw&mAKzmbW>S8&{g06ceRDfgN^mrI+Q z$UCjQXl*4C>RPtc=6@8OXF!eb8^^U%w4_ohB9yf2JokMXvJ2^Fm5~aORkqADwD-`^ z(h@C|I?r`IB?<{4BFfB2RuqYh|MP!)UYu9wJoj^7*Z2GRFrUex`Vewt$OFHgDl06raB4UG~bU*!o3>O1THp7dZ_)TQggbo2W>yI7QNU-fca_DTA3K*)D22 z*CaZ2(ipowj4o6_4d@&L{=!jqL=)bb(wy@$v@yzD)*ba%J2dQIOnQynK=95bxR(%$>b;n_;mMd32-Ii;T=R2A0 zk4c4cEZZ`E51g3TfDzm+06xlcAXAFl>9Y|N-0z^q)+!v??u*>dDi|$#0U4|XlP0zs zs<&$3G@k&x&NRWHJ3K#GZJe;LqJ&!}XU(1Wh{bObDa5?;CEc+`LFn$@21Q@effR<* zYx$P=EQg{-1Q4Qci*G-}{F zY=W=sf%>_)@JB3u?vlrkuN1l4euki|8%=Mz{9q+!uA!CT6NO?G8Cbk*8Z0U!V6rn2 zGjuND@5Cas@gVT_-#CzJK0x(HNeL~qGbz(9CQJ_th3iI!?4BA6X1aYUX5D_v4v9sv zRr-g8;jfno|2oysBAZo^a$_SZW(?xfyu-X<%m}}KQo;S}eYo!t>p0)GX0+FOMniMt zg#XFSqoZwR!rzSBgxj0P1pG?lb1XbhQ}P6JYVRVt?@KO8sfcH8sxCosjm6}I*gIA= zpaATTmcqeBmtc0qWw3M6$Lnen={5n{3)Sfa zU1xgiV+fgl^E?dY&A_pD<+$_H8kp=e?gHNb#svM@NY`ryG6DC-k=NgwNs@vzUE35& z&W@W&UuJ02VIzORq!Fpa*m@`9@KB066naxngG9P;WHf#G_Xd+tzl&&JI7wa~E~2#b zJo|xvK38^_)7#Zwd46a!`8f2CB%g6+jvVg?(;hwgZGIIo&mBvyEVQIPU*417+-B-| zAe8+YCx#B+{IIOtj(!J@OudmqiAfn;yWPM}PviuVciOC-N@N+krS?pTZ3%2R&<}rp zWaIf03{5yPif+Ce$8469!IKlBq4&^9^54O+WY?SRn0N6E@u)Y1vx(k}zR(JzQ`?B^ zM1BW8;6Q!F40#@83wUa^qiyGHHtNYgc>27HtyUgSCGB@I$%~c9jLJ(jPPV_<)1P`^ z@~&L?^=v%Q%f$UeiERmN$5gz*|f~s*a9T#Xs%h^J@;O{r8oRCX#vh8kURb$n0?X}K2Ku%UV}~2&w?#W_p*fdB@S2}gg-`M7&Jp4e@~nX&&xvLYi=S=`&W+i?NV&% z?-zxhQ)knr1+%qBA2YL)2Sq8aU&-pNQ8cmbD><2-P8PJxp~n4Y^xRBSj8(k~Z?1+B zr!EJY+o?hqpBW~8*S-^Ne%8c<-X~ReCeVcst57?l9<%q!qpHMLa`kcm^Y8Ekc6R3} zc4EU?_H^VNp;$fN?GAOO7Y-}Z#HnlOmUKOO>b4kkEs%pAMHTqeI1b+Fw7{mrb6NLu zuI$Ad_lZZo3Eg(Wibg+i!rQZE;HZ?#5SH)}?B)ou=2$51n7tGQXYat{7;_w#7C|H@ zEWqzEULefvk2ihz?harnKmR{ie~NyCoEh^Z5M6O-o#Qn(x-1Nn&RPekb9+ z(!#!B6M9!Xn6B7tNPhY#2v^2c(zks_smz|UWSet_V8x5$Ot9AhIO&)NJNOBik_MEGPe^e8ST{{$QZX7^v1&lu(|#pD9CMv z&sUD~XHW)y8haln@qWPFNxG;L^$Z=_Bj7*#RO@Ryy+u9!vUt!;mLA=Bl9rgH;LF}i zxY02fpY8u98jPLAE^}*Qc6Ihazek2$tb=gs2{bP4ONsTM_i)>6K}=Be2G#v_QM38 z<0!e5Z5>dfFMg~ca!2H;^*$x$+S$d_ zSoa?>(2gV8S6b=j59&03K`=~RYX_XnSo$rFXEbUR2_CocGg9e9jMd?JKR=z}!u3K} z68HcDFI2;$Gi~6r@eJ=!o`T7%gP=J24*O%C8e6bb9jj6jh??YOM$k5jyz5^;O&@He zu}e=Abx$ebcgJE9`>30otGmJ6D7nET+8bJ1W_GiG&(9)tgK1=)p8wOj5xGGa0Mz!pp&TaD2lORGBssZI>ESqrCpwaNR)iV6QxV{oEOXGFFlD zuOkI=PI-(@R~gyKPe`8&#cLezZ$i0NS#*!lrMntV(3Lr<^x7tl$n#yr;UzctoRS;4 zxk}lvMn`@|=J|2@=i_oo?`dul<>KU6^8t|wyE zgCu$VU%q$z4gPADVZ*%z@PWF(fn_g1^5SlMaNs7|z7c`$h-U8jjptmP%^+9XbpySB zTXVO2Lb&@!Ggz;@9kBhl2n`feG1g}t?i=gP34>3FR%xcPKfLD6=)C*| za<)_Po=G)k1V6!jpHFj>)|}^5U%uxyZJL2C-z@kHf(*N&$_9AVE2Kq=&{N_X{5Kqh zVjK72p)tkyJ1L0uF_h-qT25HkRT~MSx9rBNA2UVLUI0gBc!xy%1-SWSJ~Bn;amBj| zYhCqJc)gS0?+jMJ-k64^2gSL~7w3T4hktyJWE4I=_5|g-Zll1x1t(k%z>*~esIeyn zoqi^u;g?9pc%>9^yGEdW=VWlPxq|n{d1B^fbF5Ka1`9^b0LieWBJEHa*ks+t8XtSc zZrC>#{ygauovD{%n=ZDYTQz0$W=DyRjEN^v|NUei3=80TcOZ;*nFilh8HuWWb82g@ zJENJ~ZB(tW#ngIf&gRA>?zYNn9J~;TKNNG(i#d-QvSZjRu?&!LiGz@)P_}z*5R;j5 z01tjEV>VkIW$xT@#KH|za8Gj(D&|hbIB#)IaoR%8zV85M^jcozt1OER27$Q$q#GXH zeHltavmoo)GVE3j#!ys4+jSzQHm8L-<90|?&K2UMnw5CxunEpMcZBp8m5G!WeZ<21 zk682Nli1Pu^7zh8Mx-wLSv3FTTX-ri#fA6Jz-9M4`QG*!ve&GDk()LPVwGj#V@e&c z)gM7V-vPG|pTy-IWmv3R!yI_|52_xtU_f~)7eBj_n-J2*+3Jkh%zy6KE$;0Qifq=^YJYS8O01c!e~c*~!E>Fi0un1d_u{neeYXrNP6x^D;B zoZ*I7JvQSrqwmo9DH3EBD8sbBa-^_S5jE${M9;v|Y)6JMJlI+ex!2=iuvZsc76+q2u`X*Uqt0fSYm0{7YUAxO zHMsCpKfKtS#+LkW60LUV^wy+ZJtahh3o&Q4HM-$*(j zlt~=9S}?Kd53_Yx9!;ENAl)vANUAGSo_9eHamsXsknb)ltS8-`@0jF^QqW+jOTJD1 zMqb^urL}<@=t2G-IOkuV;CxwVjl~*yGAlqCL%Ve8KZO|j5T~%KI)4a){wxrsY%C@( zyU(#%KZnTM-8#bUrOLu}^JRnvouzDuO(ikx8)6P0SV@aZUC1lnK+(94Ny488X9%B) z#t9|-cT<;a3G8(cW2!}zP0!p43VPiTu{aZ^uWKe!ac7DDuml+^pU*y({Vh0->L`v9 zc*%1-iW-_}HeROFpG4CYjUCLy;0d(P@g+@e*aFg%?=t0nekAU43`5}<`L5PY+Vl9c z^`a_fs#B<+dyk4sf1&?m|4`vaOHkZ)lSZ<+bj9g%XsI)#G4Exe$3afmon|X+-*{N~ ze(ia|SrbX{;~H5P(OaUt!I6rxZqXxCl1PQ&JTN=Z!bnXzM||TZk<#aJOs|C|b>8)v z4ICZ@$#i2JA(O*9E&b?@eMQhQL`nYzA9{P84ZWhWfOOe05VToesQ9&*Do*LAeP3lT zZH{^d<9)t7(m@O8d8{h+p44C zj^gmm8q z(-Jir)UJz?n^rRO9~~t-wNq&5p7r$b__0tg*+onqX;abO)8wGL4rX+iK(M9(PS3bN zouUnBoY^1J9Tf(5v&!IM?RQYr@IW7#9K1BY1tey~p@IJ?P(NzGt~K0>k6l-zy`KxL z-M)>vdNU6WibmnvyVux@e+-%BSuP~kZxYeAG9tIeo028Fr18$Fjo@hh+}hu(i&30b z%)Gc~MUsoR!J>mFiRMV+O~AtA`d%4VNP!2Ycajj5fXRJBp(AL$cxS zZ@RN3iJDwK0Y#1cJz%~cJ6*zR%9E+49uh;x#)_bQ=`IlNWd{r%J-Qg)pS1^Hk zxmmDK^8#73!Hztym`_GL&S!rv@MWY6&B&VxmN+Ca2+f62jM?G$?5Lfk0-bkpFrM#x z&7Yh}6Ary6qpIeT%G+m2{_cyUs%qwipbw`6D{n}V$G^_OdZ`p<_E0qIY$QScGmycv zM$?de6;98u2&D^y_42>oRrx!G>se;Z1Ql^m&r&DWL z!Gtp~E=V8$`*R(SId9>!-UEzSrxrF1NzqEaUj(Dy^SgNwn|39eCEL50sUyQh`L!#^ zF{ck?WM&8^M_M@+Wl=VrRHwua6Y1v!BLn^ist8 z?xTbu&MSmlrahp#{G2mq@n>@N$tim4-#;=k*PI4Snn|914<^2Q_&La&C??EXS{SQG zg!$%cgja3Pl9b?R>OKAj&3#=(FR1UPw!yx{mG7~X8$~jEUXG_{jv9*`N3}9-9^1&R zgeUZcrVI7?wuI>~3MHIx1Z`LC2ixSLH!)y{A7$BHi(MmPHhR{AIo?%{P4zA{dY)1A9V)rkS z&h+pCgRptvVxhs5hlG%CXC?~Y#qVbCzF7yYlfIH&?{%n{j{zkj3GzkRh1|;?Nxwc7 zLhPASq*dcQT_2c0ZMBAp_1H+>5wQY>-I}a@wdOH5`7Efxc?sd+<2FK=DJ}dO%h0u& zTWCMe;x+v}m9hLAMt;e?VO%2C(t$1KY41)AVRN#l(CX7XVd<`7x^ZR_c{Hb(2z%cO z`d@0$oUZLSCt@U??v4@E<;^5^?i?Ez?ns|@J|ZSRPD6*bK6FIf1qHPbGFIyyIox`K z1kTc-lNa8Cmr@SsUng#Ds&-Ql?NQDw9@tOFm3`z;-veSY?jl`wsX??#Tb^mN%O+)S z@`-CQ|G6&*^LA#F5A#c@y5Svi)H4$7#w9}1p)j`Y)->>yc_XMP*C4mFKe6DP!Fa4n zWea{K6N~k)$n!hM=o{z5rJzUbg#tH_?K{i9-!l)c_H6^f@sYT=FAbKUGJVdINoVcL zq0nJMm-gq8*Qz^7it0V`^RhTC=ry4wopDTNOFOH%Ko0d>Zj+r2n@MW>eDb#?gZxqx zrzMx&iGG_gI9wVe5A#>kmTFxZSgS;5yzZs4w{Fr3!*FuPVkEUmu%mk!ap9poZ|LEv zv9zvVl3F+U5oL{N=F}ernrT)=CUl=A;nxT0n41si>cAix8lFWi4#-mcD^3-EOS0o+ zi$yL)_F#6bkVegy7n;ubMmv1k1l;7U(Afa2##%N0xv`aO6-1CdPc4K}E_y;vsf(VF z`a&CrJLo-=CAztRCB?5vMtbvHH`F!4#OX;mc@J;YAj1Ui@k_?Z5ylh#vAgN zH<6M>C#j*wBzkktXgUgl=&C()=-1gtnL4Z|vuhtSyGBnYbw=9Eu~bzeyjsUJWtY?X z)F>KtI++GMry>v4LHO`P4A*S^Lv-Ir6WiTim=Px*P@n7P>A5nXzue;JjFtJcsqHLO}6*WfIyp*EVZLBV)O(u|G#gn{|zk=Yx*Vd zc%ctBc8;V;88TE)UW-=qo~H3VuLZY5@&yZT?IS^a#`b&K3fh&hh&=dkn8eLqM-sAK zMN^eRS%F)eNd4eKEcsSWzRj3My}z8Ld7*)HU=#17dohi!yctOJcaI@6Rcu(JEBhGB zrx!(?nnUEOMFaCWbt1{~*i5z=S~7lV&Oob-Vd+d|Di%MF-jEp~Og&abub1wo{y!9~ zwWhZSemNTn67^g0%+HJb9U&L@@(fm$nypM%(iU>~i9G6!RK@i%e5X3Y6Lz>Kvignn zB)=wLe#K9>K`?2pH|^$Xdp$(VVlt z;iLajZ2hFq<*Dm%Gu77OzPK^;Vsk3p61at#GH()ghlfD#h~v0CO_M9MisROW*YM{x zVgF6cAcd8cqg9@6Geej|EpNCxqy)g&nE6Ihj`;Iy$9aoW6_n670B zYo`~IvQtW=%CeCe)nvlPt_9v7)X2nr)1~tlE+(tW8ldBxGRE@Wqpsf%x%c;9a;x)7 zF?vTLwrm`N+~H}Ur*Izbd>BoA&m84G{NB&8e5~~4y?Bz4R8Ah6j-b6ifAII$F6@X6 z!8=}rW*8gLTe~S)7JnJ)WaT+iD|7CXn*>h%kbxJ3acCX;3Da}DxzkM+oJqPiH>7t7 z*Nsx8pB=_x#H(^Jd})m{CpVK5H%iE#gB7euGK@UnXWIKa4?@mJ1#a=ac64>HW&)Ug z+-3L!XDXe<>(6#EyDz_BeCzF?!m|sb%1RiiXIF6C`PtZ=pNh>pG4M=kf%>} z>0(3ke-N2rT!6ov0-j#e^Tw#N#i2 zL${hIN)Dfb$|=|J+TC`H;k&`7mP+D`_DOg(ej|0*n!!lM`r)hrLp*%42=(Iv@r~?Q zd?irrD&edIp{LG2+7VvFtqJDgdNFZwk|ZF5SL0XxJbd65y@z@ z@hMJ6A6EBj770ubW7rjg#Am(~4UJn*e1Ctzvk^u3F>eYdXiH%?4eQ{78R2-r_?F=3 z-aR$RpR2*>*co`>??a{EJfsQ>GC|qsHTd4tL&K@dNX3L_jPm7o%o8P7GTgNe&E>+; z5EVJ4365NAhd$>u{hr8_e}7W@dkivvQ|PZ#hm9|G)%rxg=qlH1=mEk=<8aXv!-Iu4Q{K-d%DJYhN$K?e;RF z-Cu3U82<}om8Gsenas0!#^HO;2OnGcXV+Un^oZ%y_IMLmtT+QUb82w@vv%lM ztphr*bz$=!X;AUBC3$DQvqIfxM9RvZJ*HjB>YeyhJHjs-yhl7>j+9(yjd*YBnv{ib zaoaR}`qY>H*rvyG2E2)hNQWlpG+GC2QD8$Y*U?Qz_2k>#zpU7g&Dgs_12r?MXj<=h zX3ZpTa_jQ=TF0Bk)Q{)vZ8_pZ{%R`rCYRgr#XJijb!60 zZo+>vL+An3F#6v$YvJ#3En(!FHu^;24GoCD4M~Ul*h3Rzt=lholX+LCu)^C~qKvqy zbf$U<R4paf2-T6e~%I^Uz0}FX(B6iRUGDB$)xX|&ZgDt`WfxI8k(S*LKo;J zQH>XYu&hNJw^YrAu{raFnzm$B5MK0=@# zLucwY($(pkNL1Acf&D=_!u}od^ldNmH}62k4aB3iM3cRrcO8ZT#;|9VBR`kWs%v>1L6NP<%%oy*f0^{>OJnKdVdA z>O*p5&R`a;F7KyD&-IH^B`;B<_lh9({Te=y*-VFbN#WteH{hq6I2L@Qv^IP-z2Cx6 zhgsPqbLIw^uXq{6$KAyygJ>K+ae|Kc5J)2GlJHHj1wNT0$Esaezz%&aps^0|R4txA zd*&UdN=wCr?uXY4&4RRqGiCqLZ9{j-ql{Pz35#ee&xxPCd#mvKU3;PL-ZrW-x(Uz4 zmSN@Z01_ZsNHisN!JcP7)JWZgi!c2^m%Sz$up2?{Say@~yfbT)&3-nns1|&A2E!EZ z%W&E=4eZQhh?DRXonK;0ul0?k*$UFIJ>!`rBd!HE+#irft`lk0IS1-^aytEO$=A@r z5I%;*l5N6hrmsJa>{>!_+VpUgx?KRHmw3@y8F%vI@fP}Yl|E@>4cIxQ6(oLx2d(If zp^=A%=!BjYYA^qpT7NI02fZGW$rH691(i{#t$l|r<2)jp|JRwj4!Shm&6pIldDxXQ;`+(Oy^ za6qpgKQBmzn!S5O+_ssbkA50*iF07IDIFcouPTtw;(`)iU zsJby2gL#%Fb9*LA%>KbP-xWvMxXD;P>k29shp_=0lF>{c&ZRF)qyDOE=#@uL8P|+A zjCiCRh?iJH63CvrkJzB7RK{3oJeOnIkM*aN zxl#jlI@!^lagEf5bv3hL*ZOU6%XKeC>{mh2&^U0q>w+zF?J(4ABlXDfql5GKU3lvz zwsDjJwpx$DpKod)F-n$w(~(Gy#TrBW?mBcn=7%J88RM2Wo_s5l;G8j_73-`uVyApdGdmOx_PvmZrS8X)t}~!FVmZ8x-^T1|*$lZNo;kjE zfa&}FN|4qYf;(s1V_bwYWS=zv`=}y%pkov@BN6nV%}a=RJOCz!7-3>C{o0PZYN@D52&Vb z3r$t}PVw7)vc*0D(0i;v`|%CP>TQ5gs|C>J{DLUtOrw(QRiYLVM0(wP>E~oy^61$Y zQCY1y7T+F){q}qZV9Obrx-gMWeGo~qZt%0Xv0sVNz8d=^e1P+-G_DjzD@xH2cE#6`VC~_9xuA^{$AR;Rh6o&a}bI$ z9E2(x&4lclT*f7HJ1cNKM7@pl7^AL0GUbIPt#XW}P7|etfzu}lt8TP24eMv(2aO(B zI?b7`*x^E#9Nt5}mQ5nT%RYf-%~`lX*QhdWB< zboLGx@?w=3t~!}6Dz*rOEx*n(Q;&|JTDHd+zphr%lOsphdr?(KHUpC#k z_6#k!Hi?d~@4tLi@?&+!IrfIA{ki9fD!c)8piX|1ENK?O56;SWHgu2q5(x z$gnQaaG3v}jJmppj{Ec7wy$2u3;p?j@`OVPv9SLBQK8q$lPNXy9 z?CA6K>-?M*>7$=b)a}Y5s1Immmwqi|$Nru`N862|SH8EB!C0V5?UKTVo(`I#@eAtb zB(k%K*u^LBu;vmL~zSCcRH$;`N4(@B_7GO?2wPqTD((%lyi z(hoYp^zp+l^j=p1jrLeZWvVp+uZ7oYT33n6U8d7x-)2!?Q2~iGj>kRM?O^w~9CAd| z&b0mH{RakOWK~`xm3Pgc=lV+j&%1)8(}0yNSM;FpApl?W5~L^n}{F z-okwGC@TJI6>Td0$!w8qV*i~!Ml$_w6H}hGbz}H0bIYKM?=jiap{2Ss&-oOozH$__ zjz_VMyxZyMOAne;-bmdsf@)lyDir#4QSXzFEFZYX(cD!j%<8qHtvq<&&Z(~7Xi<&gPWSEY}`-!{;Iw3dIa!h*GwXLB#$Yr4y9eD zn`k2pl9XS=G|1~IJyN%ccK`JPh6`t-Pg{_E>y9&L=lQY8O-GsUKJ9ea&|dgyx}xw~ z*=90PeFHUi38$AoPlvH1k^ZT&7q+TVp~nmj;qQn@nqr&9s2p~oCjX=<5BQ*c7mky= zrwZtYX@}u@SrN1FST?J0CY_q!kE3=Q+K5ZmCX&j}yfsJfAwNr%s9}c+^?$ROGOe5F zVgCZ!={6hJMc0rDu_SPAdd9vhe^|4%vyK_H`Yz!fbP?HveL+t&S?QXwT~nDexu3Bk(%VvwGJ2*Pl8!T2Z*eWCfOz>Ex>}c46IFI zC#s+4GXgW9&_c+bG9Q3N0rBKb#8iCRAr8xiCE0JLcFce--*=yoL=xQp5V!06z2)o* zcr-2^1kby{?xG4f|H)#1)xKkAS?kdox;dnHNfcF6Y+_$DJcHM+bMXfMoyY!)k)b)7 zyz96ZLh^m!y6-EpeM&l3WIx2pgm9d)cp2OX{0h@QB@m0#SxC**xp`OWQ7fp6=HGCq zdqO7Bvx~CH%BnPy=sJqTNX@`m`eoQ(-^G57P{*1HI(W}E3+=wuQnQ!q8TSzucblni2phf3iAf>z^Usv zo9;%j;~P=zh#@A;Q(%KPf2uwC=AmH!hs*TDlXvvfzsvOW>|E}1a~uajQQYSDH>k_F zccf>V85fL^b`>~$e{Nj#vp6r<>mX(x!~*lcpXa~x}SGX?Gkt)-6Zt!Uze zDRhCej#c191l@q|xaPzdE~O?NxATn8pVscMQd}JMMtEYPC$Ex_+e2?YI!Qm;J*Rhk zzEK;ysh}U@g5I0FVDZK=B&Z6>8`Bif+UrLr?W?DiyROn5hQQRGxJ)$O%)wh3^RVY% zJTdtChIANig(;JDaGn2e(xOyOtGch!e0d|X`P3|8dTR!rnOuTZ3bJTEydBRslrf7G zn{h{UI40Rd;kvprqFu8E*gEwQtkkS#68`E#%lvhMEq*_6rQ8L~U49Ox8U~T!wbtzJ zjzSRK%tCd!DoE;4W8d9=Bxua*WxkY~<3YD+sH~^P?>i^cxt6}z|5e0RuzG?Iv-O#a z#r4b*!=JD*MF|t41L^#PRO_a6X1m6t#=s_YOSu6yyC1D?^aBzZyHA1maWERz~wj9jZ$Nq8mT&n9pn` zzh=Fv)l55HJJ!k%y$+?~T9dLsj!+`0x$l5<$KaxZv0SFPfD7b%1D`VUnRTT)YggrRpMm?rhYMklb*N1CE+bu9J-#g)-Iz<+8JCrK@3Yp!!V_g{~Nswpi%2;Y01zZ zCSh9`qhRrg`M;|{rPG|+Dy*Pwv%ZqE+o$6F#v#0;d7JJVT||Ejnb5pdn#|=zVQ_7N zG)`Q!gkHL@LWev}h=)GkMwXh5}+cy@Qmv z{zu+suVh+6iUs4J2h-)#gW=pHCHx|i>&z=T4yLV+_gUb2!`8pD^?G#|Q);ud*qB^T?Yk>Vm;jRXnG5Ht$hk zpw6QhLS1uU?ADcZ#a$VpI={Q!b$2;=)Od)FSa=8eHtFC0wt(I?C#slgY;wsPj%K>UAfQyyp9gWl$iRTB8Et(Zj5~ z%swjjz?XW+J|!OBax_@cR#+YFA)GW`Re1Vk7ga7eOSsczD$gw8#v@ zbzhQkp{}O&mRjChyKx&kOUV_k$XsCU8*AXku||+i84G*6>LF;4v2a4pW;(ult6)HV z9-gxiDcusXXPT6*XRxG_povxo|>Lxo{W(qp~ z^24U20x}+(iMq`@o`3Tivwqv6)B1e+JL3Rmbv=SBF=ALTsT5}K{s_j>%fWS4AzeCU zkWPF&9)h@B7!}2L@Kk4U&U&RHHE%6;c8)yEKbDUgX3w!kMnd>kT~Fxtxs7VtDA6Ff zN%%}~3-pVwfP0`a^^iVG8doF<{`)=zri*UlPPPxLKUiYiC?9P1kLTGz)}r3_5%{W4 z84qr==UI?DXu-BYN*NQXEf*&$%(o==H%GIp#VpC;{Z80&AO`O5n8;q>d38!r*39P5 z>Ugu(iGKakCrH+|qK(G0=<3`ZBCq%-NUuotE%|ipc>Oc z0S+l$g=J#7kbiuxV4aHsJE)z?*6j&`kTGH8sht2U&wV1|qE2`^J|FCceZVL@3J zsBIA}C>iyUeA3k*leekz4)5>G31v%CoOy!@5>8-_{#(Y}vrM3WX7!UJUJgWeMH1)^ zXW_DE1~At9w8(1JZN^GA5{jop!c_f->;Z3Iv%1Q&@MFsL?Vr7j-F z{B%RKeI`za*4!n@9uGR9qC{3Y47Asy{=+HqD-I37$N%@|7dGRv=&fmvb(W-7cQ8Ft2k)232rIwlgU ztUPgIRTMkrV+Z@Aq>`LI6hqGYRWgUZ7Yno(&xh#FMUb+}mCwkJz`jQ!FkB=c`*`2k zh&ch^yZkp@V^&&R>M$vj_sY5@7ND~Ua`){44ce@!O%g|n3|#`q&dj++s4 z16w0Mvv}bobMMh*fpPx{Mm{%<=O(7(=FrEWW$I4$HB6vhTsdqD55d&NOx*CU5R&Ds zA^2DrbMM*@n2;PPs-1jOoQqMB6CKLtnK2vlao%c>M!Sm!d883>f_)* z^jsn>vykXLh!H%R63x`Tbz@yJk1*9eqePAe!$_;9JkeV{jn105h(_+rB^s}tvUF(0Z#IUg~5-LxIFG2~v1xx5Kp_7-AM#ACKSrW(GLh1af&TZKlhNjQ015sry=AX$!%%sND@=NWI|ZO0VvSC3GE#cIJ!6nn~zvBkA6$xX8SupOU$6~ zLmt$PvxMTw8EA9LAH6ecm>P9W+~FUO-@2Ue?s-iTcxMusVLgvrOPhrA3~lgZNG|Ty z@}{N_O^NQid~!160fb*&gV9S?;lQs&XgRR~vU~s>*5xoVL;St{S`?jSXhg2{PohUH zHnA~eBsV8Sjr&nwfZCm7!AV971YTJ%`{N|q_u~!8F}cqA{ECJ*N(Wg%o(Y)g-vOUx zN057HjB~ukpz1dh_Ro=n#AQnfBu9LN@B`}wu6C=jtvL%k=gcH=mmbz++A6Y!Ip%_- zJqGN+*?eG%Lg4;G7r1r&25~NSBb;I?>?zfUQ`D4>^bTa*{2vRVzRzHtJ70l%r6W9N z)0j{0>DJl7Z)#j$M8eQxgwHAjbG9VHg^InnQzi&=^)m48Mm_2v=SIH?qG==c2&NS@ z*P0jFiu77yStEWoaK!&DtJ%ch!OUcO%3NI-FZmnPvpAL%tYkh#xzlM))2KyR9q~DA z2OBf~Fcu!QjH!7ftJb~_Mn-wVOr5Rd%#|pb=N?bJb3@_BD?>(FRL%QV2uWFyN6vrT z4!fQ$qw%pnsj9o8aFInPbJ%ek<>MW+k!Q*TE5*@USI^J|JEq{fg>zYrT4iwip@J_D z-RI9|KZvoHq&tki5OaRdCjLwdGIL%sCu?%aC*=-u^->YoW?qC-y&A%E5_5!4V|h+W zp%Hobi6vr!JIwFhd9==HEuG@_jyXMH;jq?=Rh1&{agNvX~XADOEQedsr zS3$1V9WZ4tGcz1BP$NB>q_z|Ru6hm|ZNHNljYBkKAeBt*PbZ1K>Y~a8byRJAEp^(t zm)=THqfhkCfR*4cyHtFT+^-0xr}?a2-!3O=xNbU)*;`4@(gf=VZCYgY^n7@?B#&9b z5H4DbdY^UW_Mm8yR?S|{`{Ow?KR(>va2J^>_?3JgJzLNx|Lwe zhwDT;?F31>HHTU{@50d|74h82d|Y-!ng(1-rY~%DgxQCe(ygcJNp~PeLW<8)zh%6& z_RJ638CyqdH;yD*J@yEs-M5jBfr&K4aR(jH3n3228^KejS}T-V+RTm^M*{gdQoBMO=qM7-OH+K=(G#OO#$bkKGr?N`>nwT#;>KW}z+sTuC|D)_o zqpAGjHjI#nk`f}M5-HPp_H<-UXwaZ}LMhD|Dox6eS(y?lm7!UNaGt%Tkc#Gn5+%`} zq@+ordY||6`{_Mv`N&$%S?Ae%|Mq=fSHG|~7F=Y*reSzcU9zvVoBvO1HfCv;@su?R zy2KKCKgK4(BG z&z-|Wz8fcS!$^L}F>Gx62#-#@isG}A=q0~@`1zA7B=pY-B0FD0jFz;*X;QPu_OQ!v z^@as3@`Ewn?s9^&-g^ zO6*vI!cFDyWFI~J-+RF?+rOLb|2UFtGrv_5s9K8IZWEIC>;!TDG{`#~+k%$tSwe%; zwQ015(CKpzCN-xvvdZfR`1jq-{1DAAc>7{m4F4+9WilcrlD{u-NZir1Vi(MRh(yay zXQDxwC9FBbj3j2vVLyJZMcSW@Xi}XG0SAL6X0wINZK4KWEUIFy0bfg;zYd^OEj#qX z*_`aOQzVXObg-}aWxVLm2qL%hJbSjekk>X=g8I>!{G8C`)3bQIK%L&Mk)rLr`TYBM<2dfr(B|mnEcH+JA_^J%{cJuJ{I|d78_d>|Ot1dk?Dm-~_1uZ+YR?VltLNpG;4ejI)-YmGOb6L1>Y`|v_1S;M@kp&Y&$Hvq$M=+tH$7C4_A|~{yb@UF_cU*u&0m5ZJ^xi$xZ{G5&9^*D zL)NRXt?5c+cws5l^^eEPnJ#`U{Qa6UnaB(H(70Sn-11hHJnpzi7W!Kevx?`qMtZwA z-}0h3@7`^3O7MAN;&_V8o+M93{M;c)Gp)m4yR^uHurLy3YD$Lw^CPW$=i%g^o(GX|sYhA8QhvQv<~w zaj9sgsWe$VREgjjE6Av`MZ{8M!m@(Z`Mb&&aeT=#a^3I;ma08OvUauNiL1-q_VimHq zYWZf8NCb+oLIThaXVHR&#vlw;TDbop`oaoiHj26~8Pou^+nl=6@4GO?j}99cEG znbeGnBPCKp$!q60{;bVdd^qD4_dZsQ-_<%A6`()Z;OQu$61JE01ih7<2;WcFHXh@j z``Oa%>hG!7gNbOpz;)2^&1Z|Wx3hd>j&$rTr->#)Kk&TJjr?yY-S}o43v2j;8r%RW zUp$9QnJ3L2>oxG(J_b@<7j5=t`(yf3`#z2(`-rF|nLM8K4tFNYFzxw=@bw6Fa=v8< z?edz;a_73S%@5LvsuG7oS8bBmp7X)qPfy2<+hvH2?ISd0*HMYMD29wZ=EW|pDPi_K z=a~7lemtblgxsr4BrnA>>`|LBmJ+3qFV<(tlZuCA;%axYvs;nh7aK;W+*763_bntT z_U(i-86jkZLizktnylU-m`{BA3p2$4l6$Ct?3Q|oJp5dkt=tgy(iWoXJL6b%tvi+R zoUe=4)D(&E|r6(!jpZ_z$AefHAWnlx!e;hMHLc#8_5-HMOtr+!n0j>fWe z2MmcJaUn);9nitiPjR{I6a1@UG%4t{!7c%iy*zh?HnkrnpYb*VN7UhzWmCzft()mb z;Uv^G*PJ}rVt|WwkEZG<8#y)pho_b(GyQp5EUd$Y1qUmWmoW-rq3TDZx}r(Ot~q4S zlu)9S!6DN_g%G`SqP~)JayRH2arSK@AC_GJ`u{>v-2q23Eyq>dP;M%|NQC}Nl>**c zJ&s?txvZo-QkDJ*E0*}1H`2G20c5M^ZPL=*LfTwgF#O&ndAeo`irJ?^8Uj-B<5qJr z!Dk_{3|J$$w)TEew-jyWeOo*#I31t*%{#wvT>lOf?#NclW# zY)w|;t+B?WXU8?{H6VB_E{OP#>(9_937vHRnpnhp7fDFfH0qiu7W-f;@vswt;@Kx0 z@WsV(LVH*PRcd`qszn+gX9CgWCk2qrIHHifG?B)}P&BJel z1L#~c2Ru%)n7sNofsp&p=$^TLq#`Ac7}<8>4SDCumBu{cxUhyy7%49`Bn?*bBq!dPJu!RAh8}KWgW+QFm)}xigQa`P;DZxnxqdb= zH3=ufOWef$Uw4V0TLg>mOx7oAvWCQ7PDMPX(~(R#kOe>0T}LL?tyt>E5&Uy{8A*Rw z#1VyDoPQIN$m{#a(vBb!kQYy`23?VeDsS`qPg4YkD-ctG6B1(l2A><;N9X@ZAVmvh z#jfC$UhB|8~3U z{>Yc|jqED^a?7(m5FkGIIPygvFklQCuf8#nIF|z&dFlo!o zkcfyLDViEbUYbgn($WI9Yp@YTT1C>@d}&grUx57-wzJ><75w4PmL#S>8^xr_B0sC| z$X0OxukjVmHa7#w(}8W6KP~X^b_A2o0%dBH*Tu8x!FcGK5c6xpiQIz@-Xh>#bbrLz8vL40WM?w>bWMhFs}_fdg!YdLz?7Rrt(+h-S|f+@K!W z@K05~4=kEBKLqSbK~cvFBk zR($tD^4NM8)@`_p-{mi0D$@>;oZ~snUt@ z%a{1?`ihdJFX3t1qS(j&Hs0wjqM)jf_IpmixdrWLV%}rieDWNg()WoRd}~Oq{2fk$ zQx@WJ*N(9;r)Fl2<=L!=C&V{&h&ZKsfE?np(X}suBs`~-tp2rzg=FZlRWEb!`UXY% zqUbJp^YA1&zC9b~|Gb55XKf<4FTY{aMFnixhUM(bpcR_9A^|VrQ_0WPX-omHW+~;% zsPE|K2u!)o>x7sRIln-9UEp#C>tzwww}Irw5r1<0^C}W{GY#3VEyL+)$B3iC9p0iN z5Zil*$++>r;Fm9Y!nb6yb5jGSRdN5+5h7lejQ=Jo{rp3Ya{RDwd_K4~Hz}o(zK^-6Jt#63y{W|#MH5v@sEM{&;rZ8oR0z7>CCcFeJ;8fwxJZSA7 zw12)L-q&)9zr7)Y{x*-IFB+fohfE$(6Qc^qexy;)ASd?svpJR1Qf5aJ)Y%~q5e`^7 zm+FR9&=aH2Q^U1F?o#n7-*f03y_vXxs;4^gr}IR($v=TF`S*iYX3oN_zlR^!AV!Y% z9(Z^DHtH5TiN7rLBC}LR5t6jY6oq+ev&!!IcH>kY-hi`ZT%uI=^?aHa_yji>}&`z^^gU zg7*wc=nJjkY*BDAUH;ad_HO!0H>x{HdUFMj* z(L))H^zhyx>~NJBhu7!wyN(*-7*#i>aL$9hcpl33o?VNg-yDWLm(&DDpgKHM8Nl0w zT$VhITSoudCrJ)x+T*^9)sg{;P*#?|#qTLROZQ(4$Jt|+;p}nWsE@GscsO>D>S2nUUPIn`BhkFmm_uSaK>@k(iIziu>-aB(GGO zar!>QcOUFwZD|u({MTB3`06XTU1}_GzMMc`_Vw_?o{G@^<{qdx)`wJQjKyE49Oct_ zb@KQ8NjxC^4|QkHB6m;5z&p+pC1n8@NV39jk}zy5*|{M^VDZ?qtzmxbbh9*ZN~=J> zkBaE_Uy0=KLV58`qjGZi*Ft(HU?h5KI!E%TbBp-NM_F;jaYK^lb%ngTl1-q>Q1b0~ zEgmV%F*Cdd;j)<>D@Gdl_M32;WuRxHs`41vu!=SiB;I#G;h4FZVO!J zG6G$mse^4Cw%}Q6E$Fwt552wRBA#U^f{w}g^kdV1^quu*x>4GK*iVfmhK~?AhRoS2 zlV8X@t%G;|K8hH;9zv{R1b(!vga+9iqhA+HpkvFYiD%?Dk*M*nh^wi<q@2WFX@Dpx5@}3S1%w?Z{ z@1<+AHnSj&s{?B@A&b&LC+w(dVUxQ-7u0xc28%? zZe!64rCqeQdLRD2sez2ISVg9e(P`bdR229wmC-3O~rYUSuRxbYQwTRf1*02jBq)X>*E@hvxG+9(sEIN2` z22nlN$J@Og&)n-w+3c^%{OOid`u6$=YG0zt)UNc>Sp!1vY)>T&`!b&1UONkF=g0F$ zGa?0c_8}TE{*_=F9nOaL&cT=O3rDMpBTVhqB{t#DV&+?-M`yR)z}Gr*sHuA~CHA+d zbogER-;ZFnd6GY^-L{KKEcUXhB?V39F`d~0y^(zG{ z=LkNtaZhoje>Q%fmq?ShFXMI2H}c-q0z-YD3Ogxu3Wt}a(oIhh{nD?1cQ*T`}T!|)yKeZMCQ)HB2B2YhjxUIF%5I+IR72TN9qU*Uf3MBZonXW-hY z&CI%<(~9J|^v6WNVy3Oq-G%A7$Iv&4>l8~uQ7Qm$v5{z*P4XCf2V8nEB{?V#V( z|BBs{Uck|NH_^5^u8ew*WU{guc*VPqs7LOWWNyzftToAvguble!|Q$7KV>C$Y+wc} zoo2`^gmo2G$FRmkTii6Uln;nM!RsQy58#KA`aaG9;T`Wcn`%_bA=Q+Y|X6D{f2r|S<)L!}+H{6oooY&uSn<-B&~gYQeR!9#CK zvQMSs$nt*tWKl93Oc}~%3_D40RvU}gn|&i|&t;GoN?&pL)z1Q_#EOmJ;z_NdDtlPB zo{b665jW)DA?}OLlB+J>nGBXDD6f z{gnI*F~<^^?E{V&e02FMD7pV!Cm6pu4S*zjpMdTR&Z5rIv-H=W{w%UcVm? zeK;uONbSYBAD@tIV#X%C4aJ?Qx8a*dMEsy!LHzvS95DrQL~HqCq9R$#XWkvkU(l?F z`2p)u+}uCtRQWjklvL0++os{8RkzSqy~A{5o2=ENY4TPneQ(9iFR#*j|H(0p0)fe= z8cK@PH%VTH-J)p0VXP@Vl{^TF!s+t2h|#cY67zWlne@4wAGTr>u8jYH@k}Rimd_x` zJ%52XtgvRy#l~#-;8`Z+rC1UeX~4pI9jMBov*PiOT*cYzbBI>B7nObZ7{#Ptm5g$^ zj8}$?CV^Y_P|T} z@^P)?)pJfPzW0y3I5nTV{kwtfnW2f_1-_$73FpNg&E4d-^%6WvKSc}&ti@<@7Ws{C ziHrRs#Es89#Lp_<;xB#$B{v?2v-67`_}VY2yhCv)-Io4`zqJ1+?>btH+iZ{H&$F_K z+WjfyUGqIYy4IV-sW}siOP(lwXf=PaOBY|;=8ydsKA>~P>+zy{Vm4jfm+f=2!?fv= z5BeL?P;M?vP z%VsRgf|}cJ&rtoS00r@Ag8GpF34QeUr--gnYs>C2jUF_h+`zW$~+Uxl%* zs?nu;7d8lMzcPWI`>@^T&$RMI2L&}jj7!&I-n}nLm9ChSe=r~&yc2PKbC%w*%wvyn zIBlTAN!zk$9Md#_OA8E=sfwjUU{S7O zb0;q$OnX^x_(=wWlGMLO_Qqi<_e1@dYr9&hC}~ltLfj6r!61nEuXs<6 z_pG90gWA!vrNj7!V_GcS&VepnaSNZFZiW97m>*9LApB>~K6+gwBGtfw*a|(z+$WZ7 zi}X46EOG?{-?o}^;R97Mxj`2;JxA-SM@ft~HxnZ}ck#JRTZmNSNNk}XMaK_lP|NpyRGN&%gTr?a zsa4-m@jxQ1J`;^~&rBurj?0&BOHqc7>t*=1bsWAeImTWd(O?z}KERwmSLr$zA)oo% zoSvBU71!UY<>g{{_SN1UkKOnkKaYQcJ#%f5-O_l;9Kp+>bb(_o%D336_808JUp+qe zi8?=PgE!qJvk*ByxWaiCwj;T&4#}vqRy1?!bN=uBkFY3P@Ic(k#OE{ZdB5vBP;J!+ z^7>B&HcpeGj~x=|vmZ6|;q6oO^o|0$r?H6Wug=9!PPWmR!rVB~%n7Uc_P~|h-DIM0 z?|QOjG2Jt@8U1s-&puzZV|Rk4vOYt9+7dK|Nw!vFKQfNA1*^02BR8piV>6y`_dZ1d zB8kyK6=HP4me^0&z+&HDXF3&2c#)BT_*3f;@#$3tVi^+!v09VB|C>0In42Sc$B!l3%m0#j+SU961ta45a4O#WS;#{Kyd=@tO=QuEEBMQ9Lbo&_ejYML zl|@WoGKbT+>}uk%EtQ1TMH9=gHDp^?A8dVCN!!;-Xw1R{`r3CX(hxF!qMt$@X38;~ zP&t_vj0mNBS6<>>-{kSdav!OvPL3^}t|V4giWSpQA>xa1QG{N+gyEqaR6BhI>58o? zxiQC!%?s<|{T>~|#}6gb$2U&WzBkX2^G+o;XUhn&>(HrUyHGzT$0*3!*iY9C2Y zVl`@OWKT77PB7P%A!O3PW}-hcl+h;P4o-V1-u)(#u24UN#?Ki?;wR0-539dYyHCC> z_Fy^*`|+1eKk$*YdVOP$Jj02L&lF;N`T|{^nom^9nvl+?P@)#{39BeipygJ1tSF+N zm^~^dRV{0=l=4H;IH8whiYAJSDzD?5rpl7uom(YY8x>h?Un={v=OjzfKFmgZ_()YH z+sI2>Eu!0l=-34h_y?coV$-iO_;OY(O^!YxE_xj=4iqu~3*Q)#dpc_|H7vk4<_CyL z!(wqWpG&T4JY~7t$1>Rit65Uq7JRe4oNk%qgD=hOhMPz1W6Q1IBdzTpsh`m}W**f; zUs#_fw@aGQ$PaaFK;WZx3Amb%;`4098o`yc_7?y3>}OsQIh0-0+|G902*6-@KJIJt zrg&o!Qw(Fo-Kc?d+mzs^kD5x-LdwXrxEH+D)~&Su`UN3_?Zncj?P43#yQq$S7V~+% zgH74Gi+^QP%x-p@;ZIEy$$SF^_Gp|7bJCJ8EnhN+&f2LbE>|-ZSBJ}rhn=55H>>ZV zCnleec#Jv2*Vby#*uyudL)s*oeMN`%7#v|<`YkwNG9br?P+a}^7E+bhqh&o7sMvJ{ zzIIZBz8f>2UY+KTNBS?oFX1y3V>l6-IzEy-Rmq{vE47LK`#;ob{4zFsiZZtQy%>Kk z48q?B&(Q?K*{uEYQC4id2$$Yy#LMSzp-C5I1-J7`x-N4n)7!31&Q8h0=gKF70N=k@ z``{UDF4%&`B@N_>+QzN0M#FW_(p``4xOzxD7;d-$HQtk9zkN2*MM91x`R`WzeEBi5s(us6d0J1dhAbp&+iUQvS$pYU z%YMup+R)F#nmGG%E&eXd^Yy!@;7{vPQ1FPwnxqlXiX z3QUJw%VB%HH?S@_js^G*r7Cl>Xi|_WGxONXX2dw~)*QrI-M$pYsSvZYKw@6+LFy8Y zk?V@`IDA7qn)#!cc4?I(?@RSqdv6A=olu4rFKS@#i49Yme3B_-xzXg!*GtOHVvx(_ z2%@6CiG)_pK_MRYbfu%fDTz>Kamf`pM0X>5aD4*X?_@~tKQ5)S&h+yZ20rAGXAI$D zBk^80Uv}EZmCb1mVL66vbk*=@G+auX#sh0Qc4$3s(p1Zz|2Y;HFImg_h{VYDO7zy3e8x0gZ_D?%@;CAkxrKA0Rm?wYm5i7+iRP>7^YVdF zbY%N)2^w9^)0=15a7gUwX$M6GrQ^IC|SJd=u!kHuHzT|K8!Cm@zgys7sdT3 z~9>lBKE=pdT^-^xaHx%u; z8+V;fV=LceG2=^??0!iIzA|YY3c5I(iaVFn)*frYIbqJk=E+Q@#g+DW?qSas%3^G^ zg2sOviuYPhqRZ6)JsP2dmnGh0Dq7*}gkcd&mIT4SfFsF-N^q~ zExgwiN<&-5;mOH=(D9$Ukyb}If3|y6soo#^(&((v(&>){hEmfkwtH|Bi=I1^9=B1# zK^sotR|=%`z&zzr%ig>;79-4hT_g@k_&g!h-BGm zsx=w15o@Nh4<9$6{UFBK4Iey}8ZS0YBiSBS3U#b{$iz%&s#-2?7OS?*SS%|wdd)d&1 zr|c@kKMzEr!*BL8XNyRdyk4o4HrQa>JA>G9;&jZcy(DXJBI>YNCXu;1nnfCGGhCv= zpNKbNw?Yb7z4|S>y2^~!MGf#lm-ph6sm921a0-i#j7Dpz0@OM1~XeO%wV_(iG^k?{oAiWJXSo$k?sPwam;kAA#^e0bz5kXraHSVuxqc3 zxQ%BTY{UD3DJ%Q3le+4gqvxK|^w^#^*!R2^>GQTh+E3(=Y|bwfo_+!xS$7J(T%wE@ z8eb*W)-puttqs%pZ$Fwg8sR(FkKhd*+eq=3 zwpslRbsn9_hgeEckB1(7`L+u5z1y5fkB_3(@!rfj@C*xe(Bi!(?w8!Q@Wb`SvcyaG zK5n{u13i;|E0OnEhE;^S@Rk|l@Da5qSiAK*dRr$;mDE>C9OX)}{p&Q`K70;7Cw+sp z`5&R*h621G@DTkL^c`uq>>{s^MUg*|hlzRQeiGyqNSgMx;)}40Z&9cNjzwRw#Z+l> z^v!P`W`&Tj%2;CeGLEfXuF77@jifS}h3JN{JhD9;f|7gR(U}*D_->8gyr-rE9=mWo znQu~$`nTQ0L0jbL+OaF>JSR#nP2LNaT)oVCWrne);xVYB;Sth1k}L@<@aKb{-IPct zcJm8j7fL4gkD(X;%3-C)(@@o4SCltu4%vCT4gZ)snGDUCLsHgEV6#V5vFdwf$oYy6 z);gkx@7ep9RQDlHRi0#d5iU}%I-?ses*1*e&)rkqaS?auDhdC>Uxw~?#$cw!emk58}7 zKx;w~>XaeacC8-FL|L4v&1i`8>u7!&qFet}K_zE7TqhTdMmju(p%qX0ku6@hJ!Ck0 z872*VBkhr+p*DFj5RV=8E@EA1K}`2NLkniSqVs1DRhZwL)-0tbE%m=Y^Jfo{m6B3W zkorHbaZ*sq({G*6rrqAA>pcTJO|eMCnPW>}$#KBU%*ozTYQ_Km6%$a*^-5eslgG7! zx~2?Elel)v$(v?#5yuiO>8~4HvvE8K&q(BUr|suPseb3?Hy?!0p6J6RQ|G~n?X$r8 zJ4+yav=mC_>H>|`m0;srUziITl3&-+JWoqX>na`Nf4j`3+{-K8JTdZsRU@ z2Z6h3p+HS$3ao!q0}l=Dg36QAL>==^a3#y9fQawY!GR6K;i08g(D&~@poh{xuEj>6 zX%`L*Dy9R+nA33M{S;{9c^qtdZ3qnJeFH(}yFrVip|JE{64~kLf&ZFo!K{n=u>ZUv zh^w5%>A*yIduBK^4mbwun}~(E&Qa0Rz$KtB-#pZzTR=dHJz9{tSg$Yt| zZHLv7VW4e}I{Z2%21cjmSbVT87ai=35E(2xCaREMV=32*K}JUr@X7iPWcNvd`~?nh zbCCfE?lR>LyprRl%~{0xS)B#-0X^LHwIWU_R1MrM)8fXJSBvi0Y=o~AKU!XqZ024n zID)KuVxUkm9W3=%2F~0Ku-&dscrQMHZ9#IJ?e!?jBiE;M{-(ZQ=G`q^3tz{zgkVm5 zMxG1F4CMxrn>b^wDxfTP2+HMOfRoc*AQhK^=r2Cp(ljv#yeAmIBlm2pJyBce(647 z-L8dTljn9gPf$Fs958`hF0VnBv<{r)zX)Eo&w-_0kD<^0*KossM?mY`OF%i{De!OA z;S9>G0e1fbR%tqmrq;}{oS!DmwG1;y@y;7it07(n%Fb7241Vy^YYQc>6NV$S5hF~I73C(s-24UPTog82u3fzFc( z(0%?8U@(3yFiRDK#7{F}YoR9mGg}VIA7seeO&W`<0iHPU0iusjVegY>*wv!}BFRXg z6y^-h9lHl|q79)vih}tUe7UoC9l)Nir$NQ1EHF_f7<7FO20E$Y@H{=?$?vIDl#>m09T_>S^kOs1KI19)A%!fv|&O!};PfLR{ z2e{4EdqC8NVyJ!NF+5xG9&RnW2}EBfz|X6n!DyRToMylWu+ZKEj;}3-8xyCaMK7Y! z;JzSa(s>#AH{3<4SK_(V9Po(X zU@!}TWjjuRpW4+xf9X(ouQ(Uno^%j`NzHKD&IRyngD*T$>jNX(rdocC>=9YLOaRu; zQ@G8opSa7P4b;e$KAprNt}h_%PVYyZiMqP_QW3*9b(39H9} zF<~ljeeg-B&`ve$^}7P$0>vn>;Gibda$N!6lso}ui|t^Z%WHTKPOE~C@{d8E(ske$vk^@Xm~VNj zbOi8pZs4LeujlqCx`Vg>HHzN38*u}^>3CIo2EPCK6u#y=5jIHD!QMSvK=!T-E=9hO zd)uW0&JEj*rkxK#E-oJEh1zSNIYI7WD9}1s49Ju& zuzP_Vkk?ZOk1a-nIdK5Wa$Z34dn~8$^CVinD;0UcF!Tq`1tI75gE-r@V9wJYVEMlZ zuylw9^mxVKP4jR#StS&%ek8CD#V+XR+Uclff;r47-3>!+eWB{N$LQnB&* zAEt+-gPNs=5L%d9rl`epd$TqJ7cFhhPG5<8*KznN+*gk7IQnQ(fJ{1x)=3ELG_b(dAt9yXPgh!lauOpadw+#fnuz-)( zxWoDDmckwVNnGjeN=`dQ1oV@xfWFl=pwYP*T$ldGEheg9-1G=A`LHQCCPhHfX%Db0 znZntXB`|AI9xPnD8J?dQ5BX?;OE7N)EIB8FeZwLkf4vT@+jj@3{QUtO(H~$^Y6Ne% zIl?PXLx5gfE|}E$3Eb)%jWqshpbdHI=;G-mz~@&XS5c@2y5$vNK2v~lS;jC*Mg~@j zbztZ*ZRni=L6E~vV7Dv+Y{^dps(SGtUyBDb=X=9XE^*MxAQKurya*K4YQY38Sy-H< z3u0Z4fuo*rz}~6`2A_BgtIb}(YxC1V#jnrcu9*yU^tAx#7Rx}Fy$hH##0ZY!M?i6@ z96a)0Hjsao3AVc)1Q$N^qvP`wakjMvUVSDGMkgmiBinfR>7xvYP}T!^yOw}myIUa+``zUmpMr{lSC{Z@FJ`kzA?%Zf@M(b-?ec4cIabgN7NQ+=Wr2xH~54od2UE zV7+fHn5)_XlKwmfnMFfkM-T_MxSIe!MN9B*$s%B#pbDCo30w z0lO9T!H#Xg;B!_X7+k*w7+l)}wmTdInr+b_MLHamq&0vYW28Z~>^M-N;0i*#p8y5l zZgAj=H1sZe3xwn}*lj%${ykeNa$gY3buSv=dh4EmaUnyYih~{m*<#pYss(=*oB{ze zd%$1Sci_F`0?nLf^Vs;Gd<9>qLegqdtX~Vo) zVJE8G1)al&q62?FL$~N6I4 zJPu~Glz`*&3&BzSR5&MZC!9U+08GA=1hYRaf{9r%FzHA%nCPMaXGTnd9>f~lp4r4b zNjCr`^A^IZo8zHOdkgfvXaS``B$N`;Xt@eGpysS_uo~tDs-HgxiD@I@v60=t{d}wt z?aTmSTCc#HlS6RwyI<&#?sxR%^AR|AaRpRxR77*W6avdX!C-aZ4)C)3Ivg(f1uwRZ zN8sjt@akL?=pQZv<~GHFneQF~?}HCOLF6!~6FnTd_WuQ1si)wpp>1$1bU|N3szJuC zF_3<7hmW(ppy82lXm;}~OwEl3dec%sZ2t$Kxr&1pGv~u|!hPb^S~b*B*a!RSp297* zwV?Dt4)|?!3Tzy55B5KehX>7aVR`UfV6-(GXf~({uC+9f|4aID8Eo^9LKOk2$oOFlDv(u2i%S2%ztaec#1fz) zy9x~E)o^Mm9pDE?xw&T}!7ud-V2@oqh{{|BzDYlaVxU$wSc>E8sKvna5lXGG}lVOgp_*l zpurH{7{3x4&X&N0$w#48;w%{SS_XEEdkXl|{$O;h5-6_S3NJ|C5zgx$Ip}A@>7VQ7wxuu-F+&B8RN4f4Dldr4 zVE{M1MiD-=m2iJ2M1fFMX*mADNNAgQ8>|?a1GL_`aKUAM+}ye$;6sxV+`erk>>eEq zoBS@JNnfRKaUaGp6T;v>pIZ3!^?Ud*)q~RrEa#3E&I8Tg?sJ=*72p!Zo$&L53EZ3h zso?6h5OBkymHSQifqLnW;BNVMuHXM1*m=zW-W$3aJzc4dtYwp6p|doUn8kyB)ewMM zRMAjPQ?#;gCW_NL3VwVhux|TtcqcH(VzsFa=hhJflsq~>+J6>syKfjAKei5R8ZHHs z2d&}yx%wz@jtMdzql#RrPQjsnGvQGAeel!6V9U{OhjUWdt=!b9vCtz-tdWxifAupg?CVNR6Hb zVrJOFzBy;1p+Yy9^tS`-FZKq5W|bE1cjk)H7ZrjNXFiD@$JT&OVT~Q`T_q}9BoEAV zZUfIUFZk%(EEwVq;I65%aQZ(BXf@df9xAedHt&YQ0~hkZo_!ElHTVEo#b?~w)ziVv z>(xN9Zzi1m@;ta~pa$oJxxm)itw3~b7AUEY02b25@S)mh_)2RA)ZTv-EWH;6HjlUm z)Hla~Ka)$qf|IAgsn^lKvGp7q; zg-H%uU@BP#wZ&%8w9^h+hc|K?RFZ(gm2R+kydKzM90iOfX~C!&PdT~W<2f{T2UnQb z1MeQNKuadtBgIFD!SkxCAYL&Os2o`j`$NXUWvQ<~X_F>*ag;Cj;!*=wct?kGiJ#9+ zIJg~{Kahv>Q~H3-nt$M@y%@O0uY}7lRl$V|!oiD}NKpE)3jDU}=PX+90NZ`?u+_2^ zD3tnxe@)UrZuVq=O+RxHRYyU4>RV9UQw{c+ID+hRTe!WS!ofU~e6W1iGhh)v7N-3g z0ecEgph3;&=mrq7NkKQ@f~}w7!NEqT0V_H8WIfPppaY)%cN&x>>BHs9m*DR48o+GY zZ7$N;8FVF;fEUA4!Rjnua3msw>kBR74CAsv>Oy_wvQYt@NbP}f{pR5E&~e~c_)8#f zwhvla7sCr9K11!&5a9NHG>lQ-0NdG3uGwWe=TV^qc7;9%TZgH_!;fX5=JHT5;Y~M4 zU7ZZB`kjUkhW-Z&CTSt{<@=#rI)?uq&44)xQ!MYLiA1j3lezWr{c!bL6_gzH81{@V zg0ia)!Fb0Am~~r^oBc!q?A}}q9y^MmWq2gaa#KXz-)ADNDT~nUpT=nAi=*H{B@->$v~R0i!A>N(EtDCcATf;+Cc5R6wn3KlA-0;iNO+#yZ}eDN~`BOiHk zZSSsv%-+pVd*6NF1onXMe~*HI@rt4gKdy6j!3uB}wd87So^ka>6~JXc7W^{X1^Q|V zKwU&ASj=Yw->GXr<-8QoumM5o8BQ>8bqu(ta0z_MJqNC-qnelG+xSBAi;b`Rj*t7$OLj=(it zhVXlC5{Qma0CVj*c<7`qbei-9)HbMrqBvtPu+0M0%NW2apPRb~9=pxXnIu*p#UjYF{sqn?@ zDX7>x0!7^DfRnWYVAyBj^IbL(sO@{rh3~(>MGBwQ?)Y?=$&%si3GYGb@u%R&$GhPC zqn#+)SQ?2J{0~Lv9ai)IhVjylhNN9Iw1nuqpZk3>vO+_o6p^g2jG{qCMLP}BP8y0r zB_pJiY_gMVWn?QtMn=ES?~n7(xvq1a&gWdG&-2{(>%QO8eB~EGPqB%`5{th4`?ZDi zLgF5+J#2)J6)*5&Y9fOBx?_`+4}TkL&sX?c@tGRqF=TTSXo(E3yRn@|j;P=-Kc3`v zJt`Rb!yAvfhw^8_{@7nPf~{^Yqyy{nMBPRG1#K%H;cPqhKEF+lq4_K@Fqd6c{lFYG z>uKnhlW02Nh>f>y(3hsIf{y%-_WT}k}8E;_40w`Mio&n%(NQ{!+_e+!D1jO9kVNAQe8#(ct| zi;y&`gWrSyAiGW8V^5Xes@jiQ&=VY@A3ia zP1eGefC$Jm-oe0~FJUqGHWvB+!|f50{NFEEY}TKFLeuqllRFYRDhuITHV2scLSITeFJ(o?Qv&`DzH7X^tgX!+3l9cD}Fw6Mn5*b;Z5!D&2VESIuu;`8mib3}H4-crdM-@w5tYLk94A)&^ z$BTlkxWkovyc_%&iuOae?9zTVD>WPhQE~x2t+(LS*A#e!w*vPyE#MV1^ZD1=x!m2b zSo9)v5A}Lm4;_>8f)k6Na&a)Pe zN%XS!HEFC?hiaKTHoGmrZ$plxs|9q`PLf7%TLaCp!t0CsVd<3+_9yr%%gP9$#78Mq ztGAY_re!h58*5n7`bRdGFS#<)CzDu#jTy6gb(&R{OQ2|=Fehj%X9FG2FqN4Dgn1x} zu531AW_}Y%$jebphaCT1{TW-+%aI!E4ViVPp~H1}+vfdv>=BCvnrm?(ekiusdLV0j z7$(JCB&7~r_`Z&TyU#9Wh`&^@=?pe43E>${d-!TWU%xtf742MF%r?E$q442-G5c^2 zxy`MmMN7v+Zs-a;_TP!~`au|a$q91n_t3&-#tELD#QjthcBR;c#eu_4C zfAwLfM;~E*rq$4&3x}ygZw5U&KbIa9jTN8tnb>5|>>hX{YPV|IAjAvqUWIcO8jM z_x7XW@L`PJw*r-Qx=`G(7Y4;eco~<9mIhOdU6Vth%T;LYodO(NYmHkspVElTGT2<2 zjnC-fvxNU@Ow=}8y>L8ZpxrM*m z7Q?NpWBL7xWyp4YO<9A@*rfnlOkEd+FDK{WzLY#4_Su3r8andqj-6a(LNLE(YQmpx zP~i(JJ`2}18>HXDReNvpWZ|G2rPu=QA)CpK!o7JEBQ>z`8We(uClx+BidtUtg_IEgUT8#P}b!wF~ zz^`j|LN+c4W?FO6GCm6$sdo^&{}I*Lf2Y|8RiHllt;opRjh&x)j(s$FOUGyR$E1Iw z@Z+00oh^){TWiv2%KvxPkt?{}qs@HD%fqPN`WjPkwLagKpc=M~x+b(wy8T_?x1Tza-Op9Xin zq5SdEYoeGw%B-A7)AKd0_Z^J`O?$*GepTX23OB{(1`!x2`wXk)3jm zE_PSSnsVCYG1_VZBATvYdYd;bshC4MT!v6VK^ECBP{4_EhS>7zIFs8&w5(w|jq6F| z?>`m^zUn@1e|9iFT(=PBrr}WNFYK&qpsQ7S2y=Mg7$d#J0-tL(8>UOz#TT zB@joV}bT!yoh|k?8kUeYHKw8__T;;gobhHxv9KzbTs$c zdWstzYv(O{M^WLnBFd1Eg=2sg;x73kSZfbPeK4lBFQxQ#=UK9B{Y=xx4uHqe*_d!5 zj{ZrO)3u&AG{0#q9rhegsSP(MH*h)Ri(^o6DHmSzylsMlzKTo^UL;+MLdu+Qlj>a* zu`+iSOLWV z!e;%(3j1gHK6e$&*J|PLv@-hgaUzODE?7BfJ^YmiK|duPZsH=Gx3a{pbDnV6n2dp= zD!FFuem+Dln>)FAvLORbh|&kSu$wnB>2T^oy4&kP{avm2-L~<3Z|_*1U?j{{+b2P1 ziX7O9YMN9WM48`4kW6<84$iobHwMq(v|}&X8Yog+^kLR;{~f#6YRnWH4VZ8K3O?I2 zf%k3AdFWwr@KLP@NppDYrdEFpI*y-h24Jl-F5Wn zb}p^oZ-qVHp}2Kj_`VfaVq>i%Qm5&npfVb|5?h5m?XD;h(=lDpwB1%$&~wv$$e1=9 zZO7W_=E9+fwlRfR4On_(4qp5>8B1<{qQVn-2;RR4x7LWE-~C1u*4#bHu%TWw(P9g4 zO?2YQFO|7el7z^7|s6$5O7 zHck>{2pm}D9$#G8DXgErJTP76B@Nc{qxuQo=~759-8OhieeB0!s=$aSNc+;*{c^15 zPPvVauC_>T=uqmbVo!DFF3+$&r{BHvWD{f!3-CuQN-mP4qH_>C)bp5XS=Y&f)EC8dv+RC!yA zC0^8~grqJO`e+h$ZZ{?692IgBJeH`gxiBm8M)V<9jA=VcM}D3mBd43>bub#|(~@v{ ze-gGy7LeJXB9eBvOvi-%@rt4+V{+9%6EnmD?ggY!8axWg?R)iQT+D@L@v#!pyJDG)FUER4!G_8q_>! z-uLe;d~7WJXwc)-SB96D{Dob#&=D$YO>a`?k@mt7xc^EK@;hqDclRj%Y@-2JnA(B3 zF^%l1*=YJ~Yedi2JY~0jXNc~|XNoGuz2N5`o#K|_e4c)79p3Ho2i;tZ0}rl}+t?b` z(&)q9d`U<6))bsP=!>l1`ZTV-l1a=jWZRaNiKl#s7tf#LF0SgVM8EYx*u}$=^wSh@dOIBokED~&%Ah}D2NGrp*U2@K zxLbe9R5piWfiqT59gWXcCa}94fh{^SfN5hfR3)EwcP*qbRdN($7LSd~HY31sEH({Q z;={}O@UW0K_#xfR{0&5GdhU5~lj%~eqmO^#)>82A=&#&d= z7T<~#$u`_uaS;<*6Zp$(;rxKgG9Iuw3%&tKa2pj0&6+2i{$1wxva|X3n}t~UOAd2B zbddUEPmEb2iSstU$SAU%?#hfKk5)(OG$mZ#Adj?r?et{2;4xm=j3Z@M*di5AT1U6i zeJz2jdAu6;(q2=2r8OOSe3yl<60uOq1d>alQP z0%nDLLBPNaRN7SIKtdCIo|F)mcuZqdWuQKGAN)EFqhCN1CLen+s&4gQ@)dZJM9tOw z?w4~s&H4+U{8SRTW;yJtf1PNVyx9np>!$8)2C&$d%^vN#!|Z&msrF$eL_b&I&VLIa{X7FhBw|rkJ{wxGi|Ar? z7F|7XjtVOY<~J-*uIB`!=DmEEK@s1hvz!kdr;ovX1$Lw60L-Tz=b5ugd1-1Gw=OS( zlYAz+?ks~!lMETH`9)7JrbBi|65f_xLfzZPXv@n&?dlVdFcr=yv3$r6)}7(HkK*}E zr_ub{#izLQaTZQT4kn{5S6TWeb6R1%6fVA2NZz+x;CkYC)1j^0^Y&IAX*hr~(q>V2 zl?lprY4b09GvC>Dorfu3<}TyU@PGj&d}fvl9M&cZevl`vNPGm7`vMo>pe8U@f+u++ zfefa;qRnSB>D(SYn3twv?N%EaweB0UYkk5zg}F<~>=D{`De;Y`rsBib{WQ+~DGgds zY@=`OCz^aj5=(DqAzogY_nk_7YEUTqvC4(qcYLHitG*z1@-MVDEyf**J#;B_BzhY5 zU{O#K4(wS8jTUuGI_p6#g%hc6is1Vs#G?7B1Kdn(;Nvhz&{Ge8eE+ zHEugj5>@JuYqEy3co{~zuE2Z$dXDARAH>)QEL<*=V@jqj0TKRI+B`-uB% z?oGYXqD z1Ur7WP}fOS8mz9)wX@#C?negim^xFMyusAn97HX%*tpibrzVD@87`1RxiY7I2`E4e{D!Kn{d_BlvL zzVBswy>E$HG`uPJp)+}InL$N%{>Z&D9m-Ks*g92>#z`tuYpXGJPIyJP-Op2nLLKed z_k{I4KEtH|J}P(5f(*DkUDM~ENi#pOuaroVDlCw z7xv)KgK?~7*HD2ys-;7<(y%kRLxbmrQg28$`|7uo>^BdAt$q;gA7@TQ-_>wub2TOV zS>f*l2fR+IhD*gcB$Qmo=XIT?Ptr z8sOyDfY%e{1nsJXe(l>!(;K9j*1;GmvuUSue?eE|7EsWNNP7AEvUqjSZSkRT*Tw3# z&yXF|j87`3k=uVSY1!7&dIJgkyD|k4sBn569rz~;kH!F~9KB1PB_nCWm?YNLDJBO| z1=%$3h1=@_Bo-FJTcZo+4Qjk@;V53)SDSPe_obKTlG)}>g)m$C04;H{yy1KfF7I;2 z{FD$XOLD^irQa0n*hY=2pJ{A_4CX}`mknSXvJ!{= z<@tMmIj*&>15$%i$YH|+>T}l|UzeppK`RoHDf1Ej!5%9v*`Ryl4mxze+~$>PmPmJx zu~5avF`>r=y9DLmU(mw?Kj=f#4GN!?46U*#{7P7W`#UDXY7j@P zsUpe@#*xC%cq%AS#JOKD@zc8tr z9K6m647tj3_?HHu`QK!ue%y_Rj)hPw*nwm(4VtMQ#BO|kDYDppm=+&aOT=u@Ytg+=*~L&M-&jNbJEb!xNF;_{Q`4aubH1U-B_DT2$j zffywbg_yWRxYT7ME@~o#UQ=W_EyjJnCGh7;s5sq3d-4`^IT zE4_&D;xa3a@#T|l@}DVHko$2AeJYAEVcB7Fy^%+mI;SZ8oj&jFugfn>%kWLBoalg= zDjqoZ!J%n8$V2e#hbG3*1BH!b9H@gDzp+>~Y&L(h(~uwD{}i)5@6eUJBed13fd)Mr z#j70Lx#mr9^9&mp94W_Ly=wF`vZM$86G`Pp6}|YUOsm(+(TPvGCYdMS@uQp5_GGql!<8T$6ylOY4h}=u4t2R zs{Bf?%Pc2i6!TI%#ik$AryEr}$k4ukc7~duvHKUb^co=EaxOlr^?_w(91Uri z1FcDuq!@t_p7!qy{0{WuL0l->B6s4aX%e2zDPU^q3Zh31ZKL)kTtaPD4Rllwj|`zDvX->*{X(9p#|_H(@w%8)ZGuU( zw3>#j%i^fX<`tf$+&JG0wX+-Ok7pej4^@SEV1JBl)rDr!1|;^Kgo}5iF(4h#oSsF) zuX_j{Ts|A~?IrUW;7#&T3S494Z4{N~qJHKyj4l-6=LK&xdlb{e2_I>u)J0M|*^l`| zr-_p7@0|F!d`wOM$X&bbIX+(&+n) zWV~ka?5|?3oNLHCJDm996{C59*nw)6Kh za$WF-T(-*M-On1BG@L}f;Ai`<5;ONL7nxn(820yEgUGV|yv?K5HMIG%5@t_Tg#NSs z*!6QDJv^*L3JR(uG0cIs+T^e&4=P1(oTFed#T*d}gnRaI!ky=zDYioJD(_9fY&%!n z6t6>fHgP%q$=ou+oAv&UU(58NuwTSJsg1>YI9BvolzsrwER-!Me)V5;Wb1USO zIY3@wHO9c6ZjbJz_3DOL{QeIe6}S=86YZpBRK}Jj$&=@-D2jcxjW3l*;8Od-`NL;# zkYPKL&lxhGhtClh_*QkUGN3=-Vph(oWtLODgCeq@9K-Mb%JFZLB>()nFAaV4k`{%> z;pms+{B1=ApZvO(ufN-$Wy?3Rmx1l<#H8(bE1b-drFKi;^e&Ur#C)V|9LPtH9ZVNW zWucZ7fPA@V9Gyp)e#HVs&Npx;O_9rJNpOwZqY!<0EyjOHz`Rs7cae~P2KHbgr z7mr($DVDucCN9i50=bW}eERgK=&zT+e6Gk-zj_-|>~zM};dfw_xqy4m-$8|XOUO9a zhbrev)8vy|NapJgYMdsG#jcr1_;(sjBeVHnryO3fs+`Nz-l8W*f^d+xqwGl)-|*@b ze`;CASDtu8@|EA{mzx2yxh`#3^@qm)vle{8EF4-g1c^o}u+$Z@>?3ZHJiEd){}uB! z%GdZ~wH6)|D#@jC6nIh9ByKrs5S>~)m5zToOg9au@OOvo`T6xeeB~TT-lE~dM@R4H zuXhONxdaZxN0AsZIYYURS2}MSbBgDzeksh-hj`EOi^6>(lx-ACQ%cNJ`saKA@f1fE(n9a}sx{mz|Bd}t^a8&4}vto-;q`3S!eHMJ-z=I6}f^h`Rvt=>= z;Y@gB?!e67?IM@E87xTU3$4*Hgv^+oFur;edzRI+siZ~6OH$~7VI3@b0O^jWq}Tlz zd`bp3T$zdC60Ue0zXa;}q44QYMMu;)#2RdZ&mDLCINO4I(Na9cz6#S+f-ooy@UHfx zg7Km3o4_VVOv%Ou33Yf1SyQF9Ywur{aC}kaV)}1U>sG% zv~0$47GRLTM{bYh8;Vo-kj^&TayW*8+7amOR7a)903`i;Nw?!7c<{@O{7*!@)f2-4n+DdLZK8Bx3wBs@7MlND!^c}Ar3BS`e{K!)h@ymQEGDZ-m?v2c(rr;!;tv}1k-b9nB9 z?``3zN_K#c&UIXtjDYT+R{H%ExuSD^-kcKeW=TNJD#gt01bSJ zj!YA7^Y)jZCm+G+&@`ItD0t{To$TTUD_n3_M|MamQD6-FGP_n}@cIT@A~lT0{T@t1 zLgWxLKa13?4^Y$oeKg=k4qC&9Via1g#L|I}##HMkNj+N!@zpx(_@l1F zJfrbGOOc7CH|l+HJNy7`ax0?}Q7$bSy_4i}o5?5k6X^u^!M*ZkIvH9;b{(bSMgAwn z^7hT*P+d)~(qzDGcJUmiJ9QtV~5mN)#n#Ge+t<54n6qMWUQH1}l;-Ig5y?>Pl%UGWNsZ>*p` zZ|+lCf5Eemdkcj{m+)KWJ0@74BgG~^`dEB}wF*0XvZONCFjM1c+gD>f(6FE_y{~Giym%c^cZ{tguS{6n-u>;W>AUVVRqW@6r45OzSl*QbBKi>*mJ2}a;Tj#Ruui#TlDe15xIRcq-)wq z^h`~Hk9a$V>n$429ddH8SD_d`1dY~B^8n=;&nFkTzL=FI$z#fEp{e1DL%F+A;&}rx zY4@SBY7Optzx9MU#7;u)qQ(v07t4vGx*ean$-6sywAnj?!+Jr89~>?)uCQtX)sN z<*%@|Se4)ZCeJ0_OTx;)3vx3(a6QPKL=TS8gd6ID|9Tm(!|!9j<^4$Bpp8uZQsll5 ziOYpX zY3grttdm^E&t^sOGo^lf>5w}Jt1E~9z(qKB3%418QkSa%p*KV+j)-KS|s$u(fAEnN1^2jwEiEhVRwA-kVI^9bsKE9sJQinq_ zM3Jxfqs3=eHDaWG37J$F;bn3y=D$s1NuFuMo>}0;(6i`Pl;S0MA2DieHBEakghGeO zP`=9wTv_Hv%?c&-xh)7$!vyX`Q@`T~=9a)k|VP@@jVM2ra=fC#BrYVwuf#U#zyAUS@q z_7A*NCAs>qJ8+jSM1uPv^lq<&1?~Vh1b=PjMD!@{#f4!P(4%k>3UXKZ&uh2&-=G`( zijBYwIA+;YScyeXkN*=nt+Qvdj~r%0294x)cNXzgChNJ6%xw715@954aN=4ng{cjt z%-yZb=Xevge>{kSBTdlyvkH2Kj^Gni!Hs)aTw5RJ)!8UYE6nA*Z9i{LImK6Re1;#x zeqvqq4{Q=yFtJ9j&EBA78w1C$bb8`#`g7_F$tD@lO{o^v;Jkx1-L4XcloyI`StN;j z1E#^l)fRhKa9B!zqPryz=x2c>R;4rE>br_hzM9HI`g&llW+@%tI*k77Hyc}4Z9-_) z1mxabi}LI_*emUd0^`M$vvnvdl1dQmuu_M`dTrSBN@Dkp>3pTtFJyMc;KfEy*qGcE zG9uS0k_nkhA!Df`p9j0@1Z>reLLiUA!p=Y;UtP_H1h|VDf2`&OJz}2pU4!>|uw2+X z99i(;A2ylmC&RVD0b`oQ7-M@J(Mm#g`&A8gYBX?pjRf8nWyc4<{Q~!<+o)fgif4nw z;@~Dr@tto<;^D{RG0*7}rJ8?a%5uuI*n0)5YEosBrqr`dua(%!!Iwn3Ck?T7*9<(j zwZhl>0+Gah{-xs>j4wXRO96=M5`bYY!)?wk>J8YWUD<&NI zL2I5_@XC^r{OH{2eDjO9&_4SS*8ctYzY`0gzH&MA0uylGU=11+v#@kRI(lLhakf2) z{0}W7$>k={xUP+{Q$)DjxQ_o?5XP7M7sh|S`^1iov1G1FD_H-RzO-!HcAC5FJ6%jj z!SBq2=pTCxBYq7*jP4fPdzpqX;~!K${~ygcYb^Nv9tcSrj|ck};pU&Kv^D4;ZHl`? zpR;_~$D?U%*_a|WMctj>NO$Ld9M*9E62_ME@BfO* zo!-!(nliTEF^W~6oyNK?%f+gS`^9$YW#VP4V)3BKA7!W05c^*t&Fd?RyT6n$dap4* zxqc`g6EuQj-D%oswTBvq{~|4A4b0d!9QkidakXa()aA?(wtFbHEqy}yR}^vY;Tgnl z8%ASJh0;(%NqCL<2IE^_QCRmEb?AeAwFIeeQ?TTW1vKw!;d5B^kC0+l ze(~Nd?$+7DUph4LZ!Wtqp#Cua9I3&xeh;}y)pOo>hNik)q# z5*b-TB=e)5#tXThf%8UCR*XKSp67JCZ#13Z%Y^Q&H5-DX^(*QA4BBNpN~~11>au& zAe%f-jeRlS!FJ9Z0KX+y$k=EXwR%L+b<1Vct?fkZjt6OD>>s+ursGv!8d{yxQ1WOe zs?`r7BA^oM6%N2EXa((?IG(0R?xiPD2Qg=FA%bhtux6DE`aRl+=KeddC?poiD8Pza zY535aBQQqpv?{qj?ebejQdeDQ&)R5Glv>J->|FVtVJ3X5XfdA~Je7Y7(dIeF_Ml8_ z5hjey#irf^^r1MJRv-IF1_m3r!IZ7Mabz@47#U9i*AmEP&~Dlq@RG;*T;wHV_ws}* zLm{(M@N{o1q3&N%D5#J{ir9qGQg5Q~<4efvd5J$|BFwMf0<*B4Xg@uV5;iSnC+oJd z-jxyjacl-3S{%h++&4oXHU<47N1}O60M-{q2|Bd~PBfMAN9&8Y*{4GOcy@u1*Zhue zhh+KJTvx2C-$mq> z^a%}DCHUuQ&5){G$m^f{gk}FL=n;G>iPP>-Dp(Jrjul+D|9^bq&nY~4bqx-Us(?>e z3|1x(y^r2Zo+}GTa;0z$N_DZ#^&G8b#@>WG5Ab2J^c(xnAcX3#d{WDHfag|}ijsckxqn2LRHP2o7Lv>k=wAgs*N z5i;C)*j-YIhYEGLvqO{r8#JCze?FDZY0n_}1y^at9KoX*AO(DiqpXM_^d_Mc3ux=cOKUYXW)j;biU^kiuIeQfOP;bJ20FDeOwYET;2!Fq`ve8kqS18jKxo)Gp!eHJ^soJdfDOa(Uz0f^2N1rU`oO9U=Fp~x zb!7IW8)11J@Ovf4$1`Ydt&*8cFmW{lHWUg$iqg&mxiQwj4E@gsji>MM(FVB$YNwud? zqaBH}UuI&%yBIjeN1$3em7gBAgL{SM3p!{NztOpkTMl#MrppcS<9-ko9=BvBm+NRi z*IZf^Y`{!@=aG2SNzz`dj9(s+O#%M^RinX^K#-@H0<6@9_oOINUZ;bGLL4IoMDIpjI>j^JfJqfn_9lAPN_ zbM+k%wl*2*mgyK1C&sp}G1#eS2PPL!1`i%mRlD#znJLX*3A~0>{6Jp#If3Rc(4|}6 zF)TxvfBsyNqoZsJsUFJ|$t@i(I+(|pv{II6+U#V}gW(D6Q)3`~idjQ1i>A<;4KZ>agT4w-@TvH7$=WW=*GZq)HuG~_ay7HHz zhN@O}tlo@9*)5?57MBs%-h;Pm6?pPIPt+y|-svbID-EF|xrPK|@^uIb2bABeDi{b~Qd zK}^BVo^7@x(Y4I^tZQu%UGI>j)hCVVR(n1RzPVhiwPn7z)7?#Mc}o_xkM7f|#nxy~ z+(=IT@ifSL3k~#E-Bj`T z1Dzc9hE2~rY?H8SEc2dUOhcD8(5IrGWTW>S213TrW8YQmT{j4QW53Y71JCK1?|8U` z*x~I}J*1ue0R1oR7<}s%p8fG9qoE$;vLl^z|4YPbw;4EVNzjy%C+o$#MPH+A=;9k$ z{LRniq;Pa9mB4{%@P%KLczOCC*fdt7H(ThnaQ}I^LCG9t@AF0@dwj;3y?2f3%B?@M=d|DD9n?^|iJVs9@C$QodyS@v_(VM|Av*LxFh zx#0af?ZKd97Kr{O3|f<&u}9}4o06fzDsAmWFXAO2pHV>uX?tn%5kGc1Q;D5%$r4?> zA$;imc_iO97*qeHV5QJ)Ch<5PSA90(ne!k-j@m)#XLhn(8?TVcn%Pu0Tv;T4ZwaY6 zUL(cV8SJ#FKmGS3gZf=mh1UBAv}M9x8X*zER;}O440hWwmseI8quEL&!#2^*2b+1G zmp?!9pCUJS&_(i3*TO|N8}B7w(s|=d$bM7g7ug_;?GC{30Yb-J+hy7$s)Bu{A}={y zPPQ6{>7S78P5vk3f!&Sx6vZvV{XPbX%X%ny`68M&z6n2$IY3gcigx~vBY9}VN1wy_Od~kubzFU)Psj;V#Zqf ztNxTlr4Qr}*6Q%DH75Mz6@4LNqfKSOOQ==VikkFJu{P@>8u#KTBp-A_N6?L{PIpoN z={w2w%w;+_ww))M%Zj@yO~o1)uutF|td%$?g1LnsRh7#U-lnnYN!Wd0-x5 z)^w7}Q(-4*w!@^;DKs!*HA{WDPqd@%EqP23n80ar_;Xkl+Z?{5VV#&4Z!x97)#w62Fyh%hT;!FyZfWfwh<~=n3hf6HYlaTX!THX4q19q%%}^Dd z8`79X?c>G#VwerrSZTl~+_FaZi}6se-U#if=UA@%Bhi+-vh;al6tC+)olmbY=5Kx_ zqv*5*Z?~P#Umf|$%A^!&@%wiq3VDQy-cCGPZX1tDR^$pFKVgVyDbH;78>L_|h;5Yn zk1VYeVg7R#6kjw8J%+Q$b=E!6^GOBlv}7yEpX-B>5^Le~^(?h4xk(@LHj+X_61@HU za&HU}4u-6xr+uSvCCVOMDQ`u6AM~TDa!o9-P(;7AmoUKf2Nvy!qp1oJl<{Ue`G1{- zMl~&r3|7X2a+0f3r&AGIS@D>EY+3(uDwydE zFJadA?a-hRvcp7uUTvWtrrC5vk*L*G;6p#$rVFDQDg915j@c&Q#xP-LuIs}m%tEoOZEH@Hdt_L^h*C2i3Y%OWN{=`=kV=_PbdS z=b4pC*0Z!xG9V6{KB|f)ZdW4adWmX|OTe=x2VJ_CvGKk>m)2j)?~iQZQ@2*)+iz7a z7CINVfA?hZW!CKWOiQ|{QA1}>EyUx~vfN|)EIRbXpT?dQ)8bo~Xh*_5YMQZ|TD96m zG8zA|hd)Jh_4FcM>sQQYILzaLLmeqk-GW{!jb`p^l904fnQ!;C0Umw;IMV&x~8 zvO5M={g{ATDr4Y&TMm60uk_c}RX8VWKR$8I$DMaK(0|Sy)bdRyZ5RFt+A6Q$fI}H9 zy-^IZfnj(qA_&*{Wpc!Q)X1t`bHPo11tkA8hQmi3q4PB#=KTBvs{d}mh+7=#vXOz& z=*MvAyDP>Q9mP9*rO>SV!KmO}q6xnX%gWr+18wylj(PNaVm0!Qc&OBa#A-#{tNe&_ zT5dXQsAVjLaR<0owh2>d3pza90AhF6;pI~ec*JHc7zOKqjAc5^pSBim-~9s@4kTgL z-g3^}q6y<{6fp^Nj3u5k)Zq-fyC2Nxp4Ft z42*4rp~+aHw5k-o`TZn(J|X0RXFfTdn*>=8eb9PNGBo>7z`pr&uuZ%lw&W1#YzZSX z`@iCaZ7Ou`9T9rySRTw&J;ac}0ID{0{Zf{1s*^6%mSHbc;UgTN0 z8;{f^!3uwQI1xTd+B(c}aMgL3(p?CZCrnV1w+3$P>myfp$8byPlDW$7j&t*?I`Hig zAB=v@EM^L=M>lfE$>yWOyRm8`79SE1y z5zQlFtgn>{T|eyMWi~?Twm2}G6^@qU6S(&r5535;OuPOjfY$s2ut#kJiD(Iiw@vye zvR4LQxqITltM0f^yC2>kRzZ4YC7$rQ4-X5Xpfti2*1esEfjd~2xzGTIug`+fOc5CR z7X@KvGqFxA5#$rffxqZD_@8E3GY>xMD%?X}$lnC-UORlRCIgpM=EB(!F_?Q(4esX(VW|K&+X(qyt#2Bd5a%Mf+I57H8 zk-GEsu&fK~dTmMvzc$v@**=c?o&MN)JPmgwC8Jg7ebf+Yz?4e`IBQ=T&hNZ~>uUt5 zQ;#Zm@UFq#=boVX(=w_C-{|T%A?~VULU`td2WWO=^%18H z2`1D>pZP(X+R!y-7Zwh?ppMc`EMv@+%vf2vp^!@hE?Cl=i!T$o>=(puhX`C6S;=;6 z^62GV0)}OhbXwSY8Z3H>8vU9L{5*l6pjZqSB(u4c66Lp(t2m z_VZbobF>PKHx_aqXVh|ojjnNP`vvF|sV|6GT^Q-h-p^(uETdQnr>(WAwSX0U7HdOo zAC&{gG8)XaX2W(ZD-4u!#-1e$P_@(-gcq)Y@2B!W#8R3{1fRq?M#d<6X%i;NuE3TA zJ5=^|rxE`|=*sIgI2k2Pmyf)}zS0ov7%Kp;L^i|Un}Tz5Qs7s}0I1oUqN)9LSaT*D zL@i{1EbXNC&K{?gHDf4Plm-dz8D#G8U2>uQDEQ`^0$dOQ`@4(iza_GCJ}X=dnuSuE zCt=jwJ%a9Se+DMwLU`l$GBma9pek*@==l>uJ-dc0<$YMKFT`^O9E>zm2%WG%}34JZz& z(ceaw$;ZE*o~4n?Oa74y{)ONcSxRdxOX-sANZM0*3k+k; zAth`d84OB?N6HIe#XuEV&p3z8X*2L_$RK*}`U&C}|3fA58+dGC8kktDhASKbg&tkF zvRs3fsY+7I{uEH&t%%wmOt5Lk64Y3H1fL!TmX$*EFO5M7Uv+8_5rc^Z?fCNhEi4UJ zN2jh4DETG|uiPzRK=UIRJ}^vbn>_LHgdT=|8z%#PuG|M5hq(W4hI0d02LBu5+@1V( z6VljBrr_>&I9OZ{=WP~KkyHbEJdQ){Kl+n%P4hw2FA1jAisP%-Y-ap%g1imLh3aFO zn6u>z3Xk5wEqR(OkFpRF|4W7)%9}wmW*vrRBdyT2reVTrl$SC0y^2p@5aTA8t6n3# zyJEn#o9&j%64BvNBDS3L#`H92bUAhui``|hyjF+?|5!&O-|eMaZ$5xk*Y6UO1GyX@ zy%Uhv%|4sDTW|%3%U_KL%=k!Xwlunk zu$xv1gkiRyj{Y^Ve^|K41ZsCTzfX`$-T}%VVAPo9OfCFs5weGDd;~wj4YG zGk;wMlLg)^Q<_4~E1$=vd|9d|8$ffarQnvkD(ru$0xAXKAkDsai-Q-C#<%gPJ}ig@ zj*P3oe((QHdkUhbQ@~Js3!@TOOwf1sP= zdTklmH1USy{F_d9ZSkWG<|%Zzyc)PH1NXRn5rnK+gid~cz+uG=C{xtpj6`dblvO1# z?cV^o5mF1PK93<%F9QD9oP*|Vi&19g7)&u|%vJYOcc_e=r+hFW zQxcCEj^hpc+c>qg0zYn-0j*i>FmwNHkU2XKcX|n<{`OrE62Bap?L1&>q%vGRZHWoP z)3KhNn*&jg>mgvuEljrHs;jhP_3ir@?3IU8sV<1Omcu(Sd7S&;DM*HIKy8gUc5h;g z1dlDCus9G3+}6Rf)!U)l&NE^c{>hu+S|Uv{I|SDcRjO*h7sYXmt%!C#CInE&A1d3TAimDb_ZnKSTB>PNb2 z+F3fA&7u85=5Y3P6q{%+9V6K@#jq*#8O{esdULBSomwwUEe!MVKa_&AT72aE)e2Je zbvJ~tb7WU%1+nSe0}&MkMEub)vTFA}P#$*%kBkLi^-zFFtT%!&$rcD}wucE+Df6`y_U^T_NT%N$~aV8|dH6kKY%y z!EmC z2`bpU1$D+?$D%3h|ARRe|K4&6r#OTEfEARt`anb31elD~K+n6Au=iswNG=IvGt*75 zDI^Tq3b(@KIDtVGWoU})C7V9qBrXltz$vl`M(_<7sUO0SDc(3k#R-#i1aMur7?v47 z1~bPtu+l1k=`X{e#_=*H#YSSJWhlDuW(>$HH$nI1Q|NmlLJ#K;qvg*gY)JJbhn^I1 zhCe5942}y!=5`~fKdu0q87G6!ZwIRSZ^pPtF|3U5h2qFUFrKOk{r9FpCru+-*OF1O z!vW9TutodVJE8JvENo%S@yVbyL}}Yq&XuAmWQKtRZMKr2|9(uRp+Wli`@1AcO?-j> z#srBuH$i+-|p5uvkY3R7ps?Rh&YjA+kxPitZzNu~y#A1i@;b|w`q9fx&3 zZP3=hO7e~8ILrP^CYxu@hKPa_Ff=0y-2a6^@`Iz;@^dpbY+QhT>Yu?!I}+xrEr0{v z`LNF52t>}efZIy1V7mM*XxWqv8R9QU*9!Lj_w^(yZ~x(j&sXvJ`8ss8?=m^QLxYHg z^1vH=TUevA3fcwNLbh5kZeNkk`Z9&sls1LZ&A)Kb#63Lr=>r&VmBGQ6Q`r7+3w6D1 zL8X*tQ1O*Pa3LlOjuSVy#=Id#+h>9K8)Yz+(dAscv4~S$p2}J8-3A-aJHc5VV;D_O zgA?EVLHng6tXHjomHE-2X6*@z?e^$&e>y6Se`Fq%ZsYypE}XJYvSj%$5jZ7NLEZ<< z29X*a6qxo07E4w^ariO#QyT!WySIX7V?Sphs&Nz)OgY~Sf+1Yd0p>h$1owh5A{X$G zv`O711^v^|&wv-*qK82JeJ}cgrTY=@X9*_hna_GSL+6j z#x-YxA6CM<9rk2eYz`@{ON7~uwJ_Y(2eSlUf#`u5sOChl`L8`5zo!M;mdu75(xQyf zvloY!xuHj=8#>waAb{B8=q5(DfUZe%+EJ{mb7PfLmuiFUI!8IwUGbtxCx z$7|qpdkXaM{KfMUTGTIm6%`$~!@_~h*z?_u?I!dTveX}D>3oE?1F5KRiGzW^m*9O*Vc6%r z1J=nuhTO6$TvmSxbq;jmh@cC8o?!2<*S`3zn|04unxbf*5w1RR2WNhKi25!sFfE`M zOoXyQRkHw&d)8skQFb<*Qj4odGaOI-0{gqga4K&ni5f2@YFTAOBv*^ps!34$*e0~g zyo&AtGRA%71gbLi}a~7-KIiNzF2d3>>jWY)8 zA!~Ue@TrV5F8^%26L}N<`}~+atD-@)QU;`sy(bB>Yw_pJBPhU4!^Vy*=r6no=N#Tb z*A{b}tmEO%!9T|9wFBq3(nc$-rr|2^O7}{aBiN07- zOA1_*O-?=gV7#i!pE$-laORIK;v_h6s47vTK^tdKmt!Yzf8KxC@n;6gTE&4F?-Rz3 zR3~SJrKy>MC{4Tf1V_IeN5h;X+`n@V%8KoQJ??4br%oM*TYZJfZM#OR9-X6)F9_5A zB~pyBFpRgW_27-)Xu|!u8DsIIpPku@bY1tchwCXJ)RLU z$uZJ6@EaEh_hEn)V}^AZf*G$R{2*6gcSQv5-5!8?=WVd^>s)%@ZUDpLcj1m!X&ks# z2Ua5K@Z|9^ymsmkTCu-J$*tSC{l;n3Y!Ae8x0A3TrijS4$8vZ!AicvFpo!nCXl88> z`L|&Or10)G7E>4S=|{jq0a5#u!$)4=;V)G=HplDd+dywq@nnfa69!>Ja0lc8j*nu`RoV%A+8mBC6F!A+x3PCXv z_{GWw+XC03F-8Fbd~}FT}_k0mjW(fz^*R@y5qCu%8=8HvcX$+12}k(>xf2@|%8x>axwy z@H-ZCOQPYW9b;dod5^ksRL^b}l?>j3l8r)mOX?VS(zn2y`5ipR zv@tZK3-=1i&fWs(!uQ%Cazx1*G;)IZ;ZP4#kOF*kYOv0m>#+d&^4p zZ#mWB7Q@_)F0gET24q+Cg8TJgykfyaH#zihzG=rp^gt*cXwKtITmOV3axjP-*7YQq z+e*6GeL7(2HPVxt3rG93k!mm=tGh5&_jRJD?fg+c9WxKCB_!duNj~5 zlcBGE*wHU17tyV^zrv}_-@s4K7$1#|n_P_3HVLR|hsw%v=yyuPcz-P_*uR%* zJH%7o&M~-nM*~Z?gk#reA@Idb$L~+);EQkW>1q;-wl28% zB9JvHQ<4E{}) zxkwS$z&k|md=#`_&LJQ6UIUw*uJ~(KE*Z4)0=LEq@R-+61ZH%@xR5PgyIluo68}Mr zKn*O;4Fj`#@o*^X9xN?$;V8d8!Ks;)Av34GFj>B+kKq1uP`0lCY1UC}%+A4+AL?;= zYYV2FF2nHEi8y&k9~~UKS-&sS=f1&Y*=3eElXta`x!-aSoR3#E*(O1r+fI~ zW)-eo*?n_D4#($rdm&<<9Aj8W;6RQ019`;eA#PN|4UFo7(VtxE8V>4$%3( zuhBbBRn)<+f-}429UK{sMS=R8Fp+W+jMmG*98-7tyFiOx$ehG5>9bTLFolNmAENh< z2;oXT0#=+PV(wT?V*-n4vDpb4GjI~OpB}-uK~HMdEQc|S8Phpli0{qA;E|dX-mqAN zU230UQBE7Amq?-c96r_+i6Gl*=Rvjh25e)zzn;2Dn7QFLzT26OQ#K}`^Q^bzKzIpQ zT^NJ6j9r|%sS-S&h+wj06D|?Cilf)hVli`L?)AJ+$i~@lH@<+~13Q7Q+5(4;3!~Gc z!!Sd^7x0$_W*_w_;Lo14Y=XD1^ehCB`@mXYDf2uZH3~mnh>}=2X+SK zfwMP1o>+(YGq;VSw{SbjyxvKQ?{5X+8(tt~aT7LbD-hvHZ!*Cz3=i|9P;jps`Ynyc zuI0SUJGlVHPxX?mvzxG^?kZ09wc><)4SLUci#&=F^z1Apn!vcm`}sfMpi~5`KimW! z+2-iXer7S=B9wc43m=|q1A}1=T=HTbel3oo`QycOV^ud599j)~-U+~A8DY4%eTH-7kkP9xh=F=xn$XTm>Z-C%IZ``P{hXGVY{OJdXLbpbB*0yMMYcGPNAg z?h|BQVqB;v)A3G34fy>krAlx+JMHoyU=9LN=(?2 zgd-l6IQz~p3U%jzPvrpkKb1$N^hg}`JBBZRUBP`Ucg2>x*ks&f+M46zb@? zkJgL-M2IL+(aWY;weYh9mh}ma~^<8J8ar+|Ks(jSMF#54c2xVtDu}PE-7)INFLYSVr9}N6L ziBSa0%TF4^SJ@rV-(gFw26bsv(HMSIU&{8|Q=oa%2eN3A<-w(*K=yP9*|<}NTmI<_ zm0wgrBhB~V8|x(~nl6QVS093XS6^^6-@i0b{;>wH0XZd_v5E7e42LADqIZl@_^5+TMb=4Kmh&#~* z=Ot*_@?yLhe+Ug9G8gd;UpSH)jBE7w2ELd!MsW{m-q3B;p)?q6ukcn^gRd|dayerl~y zNz-B?@J@IbmS2d#raX1-!iJgLw>u{2(f=~g=4w7Hm3T$=?7WB1C(2RyWH|0wd;<2K zd&T*&>8XkR_f!n%TgvOhOzWG5B9j52 z7SnDrgmv&`CEK^uZ3gi@pGf4m6Y&u9CVEENsNFjX!D^S`50{^cyCtH~GkqN1jI>l- zijEw-jp7qB5Od)zd31d%=miau(+{ga!_XNeSnoXtZ7_k6SZ_{mf~8TOQ1jm$hbzrp?A3qt)9(KRV@YRNLDibRIdYu0RHw-F8Me&HRv47s1_J3BZVMn9@3ekK|tA z-BW{TG%*i+BgDYeKOA;!h=%KM36gwIL&1v+6qS7Fg;*n66Fhppbn+xagyVTw)sO+7g+0JqzK!d2qlBwDc#3;Kse1zY|Jh(QF z1FwBFz(K?r4C}&ZXYPM=nb=$^vv4E!x?F~<%Cac8Pz+T^_;I;tHF#;d;~wIX`2Ylg31 zt^1$~udgWW-`K77!er7*j4DLn^-AEF-BmuS>8Ddga6k{k~ zAfIwt33|I>e~dCNP#J~Px%qTzNg|bdluGqE=GYRffbSW1qe7mS>$dDA-Sw-9x-Im; z_J(=5Ejku^n4ciCt_?!6X5haK%gHwjKBA?+5uP*d4qq29b}}xLy4iD3I5-bp_RfNv z!cEYS`~YqdVf_6j3Uj-*p+n*he93*q2?K=ThFWM4j5n!}{%bOtV*+vydN4KP9;Wv; z;)Zk6aE_!TDzIKLdB}Q;vlED9^IGPZiiajI0VsW|2d6Z5LCc4SWS2uZaqv<{Q7?Yx z^7;w0|K7pmzN5HnJuckqV3r*mLglq)2v9)jV+*D>g77xJ9R!Pl&x zbKj>3y_zrMd5shJ(7OVYx826$Um8G6@dKNMPDh`%Tr%svB#!hYp@nh-=<(lxCzkDC zTTf_W(KcH2B9Q7oNrcn7^Pt1f7eqSuVk}L@ckkIw`Q#Vi+dc%fj3E}H|AdrQszGy} zDP*yX9p&VsA@ec*(6qLsq4cK|+@R)K&S`;2{2fZ-o2Vf~71;J2>Cc84C+ z-S+{-f(nU!cst3mxJKSL`@_kKZLo2y8blkOg5FgJ{3o#qojfza=+PAjS|)*)&t+3S z(cARP#3X%?{~hP%^3stKDcaiZ0b;2ef&Zo{+{?OzX;WXKP{>p|XD$+<{DbgJ`6Hyp z=D~rhIz`qY+P#$8yFG=4H6dUV?g>LT7ceH-X4HQ_SquqctWjcbs5n0XcYZZY7!X;2t?}YbWg0nnS-XbfV{ckJ7f)ljO3^7{tAd!PCB@ zMBwo`IQSj$siPk2pq1gadS2QdyPD3J8A796g6N+?Rp_$X3Kjd#!speAP&@G)q%LMa z+X9x8?Ar_KJD4-pIStr+6dE+QGMDCGQ2i>3W*5id@!k)_HLDZ!SE{0h^avE`CqYku zG^ln}a&}zc2fO#8xL_=Wv_>rhp{13uv5yts0 zP~}lVL4Pfx6*-5@^m#|Ts%&wCgD*xrkHD1M8t`Y+YDn)Zh1csJ5l;(YIQB9CP6eBT z$;SO8X0sKkk2`|uKbGQmHI`3U&Rmk+XV7EALEM)$ADYyXVM7@&YMFl_QZ`1=XMGpC zDqg_h8Zq>f3D+#X40Q4W~2D()wsy+2aZ18fa8W9S!K^WyaS{Dq~h+pLOk_O zk9h}opk`VLDh0D=Puy}`e||0&Zr@8~!_25#0#Xs3Rx%~pmW1)7+{Y5lkNew;joJN1NMPu*Ahj_kpDjjzb=dNm-#&z)*3XaK`nCZ64 z`gf4FJqf<$cu`rIP zo6Xp$7KU?$!%;(%`7-oVN1S&o0^oPlzs z9q?rGKe~&9)KYi>J-GHOZXSAyKjxRBXsj2sYOH}z>!d;8oI9F)WUPLLZ}8-q7wUA{ zp!$#+D!gumYz<#H`rQ<^G)v;Wp1J5MV}N2c^WkC5RkG@24Vn4qIbM7ohSSzQg`2V? z7(l;admLkwL^5{{T4Ph+N;K|235~_KnKR%HFe3|1*r-fvpP*naSh01*Mjo#I+#(*cqvcs!espm@*$GaDhn-I z0P?iZbR{~eS>ZiDX%u}Y&OC^b@SuLU4u+-m$^%{bh*nSu~+~e@U<4x#(RfoCG zXOW%bJf!Ke77?9D#aG9V;TM{SrfOz*pJx@^%o!rt7IW}qhBDUI@#A(|4cdI>3m#q3 ziUD%wm=em1=bo;HPs~4bLVP8vIBvsjjg5G+z6sU$6yo!~V0<;*7IXPJLF}^$y{GsC z<1X#TBU%>BH5djLJlVcx?f_8~Hi2V~@o+4|7+npXF$cFD_}$B=q0wcO!?{4u9^=Po z7h~LhI2m8hKTQ|CucBS7%^R&RNDoaj#79vU#QvT#ev0PD=E|=yy}%N|JrTqY16f_r z1s;#8V4xro?1CaOsA?bbXE`#?nmb6b+=ugnHIUZLzPIu{xUJ+SF8eSA4y>u*yvmE@ z#H2q5m1jb%cXktIE{(&?w`F)`G8ZKlpM55B^ReynhtXtMxUU zGOUF)tQ+#XN&uW3_K~@tD`Bt1FbPqaK z9Rx~Tg`nYVcF@m)=;Wgo`r}>(DBt#h>CPJ~z;MmS-81-!f>m@UW@$AEx76F(y^aTrVCgYjueyDF=4^J9zLHe4Luu%;=(z*c2%mux>ULPfG!=S818M?e?!(skd%noow+c&e>j41%) z&q(6F&*nH<@`!UiUxnz+OohfD03DuQu>ILu5SU$oy7|>;ZQF&;Be&5$ITqzT!?8`q z7ZX=6LBV|zcr-d0O1Bt-(f--+U*;RqKh{8!R+W(M>)Wtwn+JP-hhnG^>l6N&Pahkr z(H!NUFd-<13vMmP!@5VI*vAx(q7%sQZ=kbk&8fmELM`q#K!D3;kl^=$66bKb`j!rz z82X2s%OyejfrrV)%vmNGq4p%}vJ*#aW30(G>0&hRK8i_r#{)m27xhVKr2xBs$A<2PtF?!-aIZPLxk9)z$d=pqFxsFpQyo0m2w}HE7FqgZz zB9eP6Q5*g(Dj?gwSd)8G=0Ti{2aL(Y!JCJQwCBAvT_8P(c;E_(er>}`XDg7*m_eSu z3?*APDw5?b^XZ@B*)(ifKbCHJ4z|~Fpy7E5+<>)Mzv2KEFD{RTjO1X;w+7 zR&W?^2~OqCVMzkz8L#R2)$8%XJ};6!t$^cEUx4xt({c8P^Vm~h1m>MnIa2SJnoMEu zW*@#s5PjehL?7fqrPx@Q-jNCm)MsK!nlgSBoQJg_LMM!+=(E69mR02?HL~l-ibt88 zfvQN(?(FHDl#!F<*V5xyOYh=}4c+MEsEOC=zrZ%NdRPcfXrA9kl*V#7=GSfT#D>jK zzyA!G7o~>oS`~2n!FIT(sRRL*!zAb9D7oYUw8VctRer5VUF09JGvi}S3A=}TgdRZT z@_HEE`y0|6A93Ejjo_fYE?F`-NCHIsiR3mD=*W|SLDq$=J(K_~y)qaUs)}ozopAl` zLtyWL)!OjQr4PayE$KqObn3FbkxF+9j zi6bdPBH(IjN{;GD!{@{8z2|Q}; z;OZ!fb1F5Fw48zmf8=pZ{|>C7-L%?9f~zIdWAst zu`X!0T!>p@8ROhk2ZH`Gcb9oNoElyW@aG^HO87v=sZV66*A9yIZG@S#b*NOE5q%!1 zLwlmQB_Y?&!lxk~%zdwf)<<^Y zck}ajhxxeDa+@H#RRkj#E9TLgGpKRy3Qj+=1~Y8*v3RRK>ck0}{5PY~L~WWr$vG@a z%i6SQ(H|q4kfK4ta^D00)@+;=)nwxKo0&_e}6 zWUhb*pBo0P`3(E_++@F>``~!-2#8x=g$PRz{Op=Z9*@odr^$S%XzhpMyfFxo7sd`{ zUBXdj=hk~QB(TSjG1?>WZcrv_91wyoQG{kIZ_u(hhk2^SxKFMSZEJ$DFq`GIR|w-- z(fgpGa17S-FMk=n)PqxLc zD;+VQ9wBq+JK4#0WeNIAnWt?g2yC7LO`gy3vR?tFWW;0ZHcw*Vu0~|%dXkr%WGq|j zfvZo-V)GUi;!zvS@i6H&NpQPP^o~5?NI5O$j5zKfN(@ePHSjIx*MddFCN6^mFH|_* z{=%f{g#_oeR;G!xHtOx(XbWOed>$G7jGd7dCTWk25w|1KTK)scr+r%z7Fgy73wOSf?O{ zvjzTKpddX^P5!$mO~(EdbMhMlIR=WQr0iP@N2zirNBh?aPG^Y}C*W|s$sJ!ksy8G- z#SeZ$seD^Zb+<=@4I5DA;cPNHT7e@`lWelt`WP5ghX z@D+*Wn9yaMM~i-d)Zz#zZ$1F(E^pzf^g$Syd`GI4dg0Ew5AauyACso>LBSiAJ5dwn|X(b@@ug3;_+7zeZ^t~;R z_of?mc|V2{>)BZB#n^;7!RT>K0rsSvgEk%ooUq$Yy*KZrt({xwLV*TywO!d*XGVwd zMp+~B!bqIVe8KXPigtLas}-klzG4ZRz5iD?2Y*-%LW+Jh3|i$vu5k|QHg3fATn=K!~kMc`!zPt_XNGyJxw8Om~b+cZMUj9REa|$KKOM_e(jK9(Ehs{#p(B zX2tOCP#&0^eoe>~Pm-m*1`faO!{K=kk+j_S-sZd`;`2`!{C~wlz=1WWmt}+fSrTXx;t5T$cR7dJMNJA0g&Kbeok|*>sKYwP)yNji zxZ(XibQTVS`aOj(Mddr_%u~gM(pMqFpqrnTpRzG zH2ylq>6+^VZ}ZDg>*akciaLiNHUXvm!uVe22bh%>f{uU=zD##U@9P?HV~-*>7v^$GDL3@-DG^eF7DIzo5#< zEb7T-VI4d#@paj3s(W@OE^BkfmCEe17F`IP_7|YGj&-XuUcppLb?nW^!k9y^Ns!iD`1B#85UJibZ&Ewa*az zzQ}{CZ2y6Yt_Uvf@W6L$cdDB(0X$1=(SyGlKfn4-^xv#w`PvJxYwk{v$}0mYc7tBG zJrNQgc0i*Y#dm+5@%-Q(Y*h9`lRSB}bN>jxl;tryOA7KDT1mUA4Vc(6mPg`CFb(|* zj}dh5WKOCxZ^iHfO-26>Gi;K+ zh6^n9=&oee*WLXR)`&1JxwA0JE|Et~LmreKKZ!q={l#v^;cZM}xn{u-sCu^q-_`EK z`Ad1I)D}Bhez^tDl=0HHdLPhTbljxksWp-HTMseIV!`pS4J>?>!m=5L^pBwhEffx- z+r9K4i1pNOu3L{%KRoFDh8Q|3ewMCVvxJ^A3!@HodGzT|HFA4}0SuQug2dV0c$>Kx z9ws%Se@sUmeoMrRo1kg%FglJENnee4V%vDV0yJ3T0#!2j6Mxh<8xs^EfkaH zoWY7`=TWb1I>}hmNP=5t;5^xQl9r?byJP=@D!Un2^TGgK+&poU_+9wuJR8@1JBSua z)3MCpEx6wQ|GljdTv;wtl6i9)Q#@hS0W(-K&lbG%_QD&R-!S{$EF4VPf{Ujd#lZ8a zSk>T7WWMbuE#k+?Vu1i~@MUg{&(l$5)Dz4POhtFT^(b&c1uaiTV5NH{N}T>cq}q-X zXCE{2da60ttUE!%io{_w@iKnuOvVGoi6~?2fVPbUL)&NK;Z4=JaepRSG?(J5s|)cb znTt1PX<%%XFSb9qjV<#&;Or;?ToE%D!(J@MY3F@t$1z#jY8Q@gu5QJ8J8MYV>`GLh zDx#?NIUxDxK|?we)2`patr>>2_t!Lx5lF%N-^XzJB02P4xgQ<`$`g0pMfB01ebj7n z2~Cd=MOiktswyeMJ5DjEyy7$pscWLH@j(pW9mRR(YiawsY%EcoPM>Pq(-WN(t-8&@ z_$U(X`_~}%zbq6HW<2JZJJ5jfpXEyKlK5}ixf{CzxlaeTa_ic=F|v?3RV;nbi}^pI zrhSKHdbaq9vMj8@AzFX$EG_gA#H8&MyXGmP$7@-1DH?;}@w=eDDF*Ka^U~k{#A(#s z3DkRi7{4v_L8m`5pgouk+Ms~f3>s+5@eXS8Igi#A1fg?91|HZGjj`{`p)*PWwYOU1 z$md*CVcD2qJr3|}Y{K1hkFxyNQwZRTql;79sobjPwAOnwUT1#5S0M$cbEgktJk)n+raKJVJWu007P~Qkce-`7W2azZxb{NNA z9mgYY7vY&_viO;GN!r<5;6#lk3jA&+&f{YA#)B`Ycq$l!B`=_PTOxk>IuFNtc3~9j z(w+W1h8w)6({po!=$7)_|50=v{#3tj7`HPbL?J~|q(YqMzMf-L5@}INODQE^X_vM# zN)b{7H7SoX%|n?se_5sCx0`w8#hqcbxjy#ORz5bHtl$}l=|p} z&?PD${rsCGm#c;XwF%tNVlLi5bs*Ja4uAH2KA+Mym-}BjhcZb8N?snsn5tk>&Rt619=g$= zt`drvc9II$C}ZaL#ngRvqS#knB^$ln(tYnTr3O9UqN+*Tii)|pkX33I|#H!D79 z>{TQ`T!dX-56RTgmR`h2ND)<(7Xk`9g^o?>$6^znxwOUG&7t_xj|1lbGE_SH64OF#e@-B{x}+$w!sR?Tzn3zRtEZ;;=0iiY{}z(@`YXMw8mmIy!08725EIs^G=ChbQ6dWHTZQ zE3!6AB}s`gp0DkT=v|@cTeBOF1v}yQ!#VhGYnn}#uZG0l?jY@GsAcY%8RR_H1A7-e zl1%OG#M}%=(p-TXkdYjQPQPquh@PRjvtSfd>LX{7;7VWXj_y;0cf0cj?K#pD?ea_D z_+EjZ*`&w2FW2C@`rApHo8huxdiu}FBgN+u8faL|*1i#Wh+aN$?xljUI?m|K6s*D! zbtrxNM)Us3qyH2!e`@xn%uHQ+|Fwg?+dPoG`t&95EI)F!Tgu~h`g31rf3D)9kM!%p zaW<@2Fz+K#A0m83iV2uKfzwOfV$q?wOk=z4f`Q)8o-&LIy%8Hq(?j(XdfN3E{ZL+nr-#Lks%i$-7dA_ZmzA-Esf(y>&r!;{A%oriCYWz^ zl@w}X=~JXNDeD;`uhJbi64#+nR~_La!HCMEUBipXs%ty6s}%W)@hbeu zkDkz5C>0#Qxfqwfm+kXtKldJJ$FyuJ-_%zVesavHEq1&V- zsyn3vw0A>sNDAa-&q4ceD4Uns$liYV#GHJ6F{AS!s#ZUMZn_-ILfUDUg&JC22Gij- zZ?avrg&@wYyLG#vjS(_m^x%BI7Vj`r<94Sz=$Tg|t-X3v^c1alS)l{JdTbnz@Ar?6 z&+IAY@k3y1<-n9}L)goxmn@>tp5=@=Bbm2$zvMQ2sOjG*Dv%jMJKrDR-iG`6b+vpx za+DJHe(?dr$9+JR=P*`%>^eL2(nqk$B;4_#yU6=Y<##4laL?#VeEiZA+_TR@_FjX? za=~4C6RXRiX2(A+o6D~RcjqB14f&$JefdIzEK(F4mn*k5#OyPH0^G*ZqvikD_(UsO zSry945W${a|HfwD^q|N+OQ_jFk7`})sDIjE8ork>TXq(3j6k0DHmHam-A7A#G{%3ThByWM4%8s)GDG4^ z^T>|6;*ahpa%;X%o(4Z@NPbVecw>iMZ5sIG@tum6meMI7Z-%eFY?{k_TDGVUwSP0F zJ&zYqzIhd1mOqB|p7-$kb5ghz#ZF&28f_Ia*sxwvu+hiidEx|GckMMxNN<)jt^WtF zZ?gQvG-ZCqO-h;nhEuCbE3;EI<};MPBEensO~%mkdBeuui~6;Rcm z4C*Ytz^0N@trhHs$}x7wwuJ2&Du-(a zbTKlv3+&Zbu*gsYrg-kO#JKNDsp`7PQn!1KQs*o3ND^88txY{J<%fjSR`;cezFo+` zbp-cM9>}HN<#^SHSF~`h9MT^CpgPSl^!ddDmVYUny`M7`6W$EL?+f24#nu*@vJ)^& z_5m(O=m@_3XL6JkeT4<}FzrRZk-3s9mXY!f`e&LWxBw z{et56*H|*+0h)G?-~%=|i(KwdK5*ntdJ$qvpCl96J7s_FvUC8SHKr43!^c8ri4|Ts z`JimuJnC!;C;Oa?P$lb7ge)X}W)#*38Dja3iX?v)-*BrX8 z>p*i(WH9T6KWt64U$epP_t`TQF=Jizot?e5j^)N)V?X*CQ=_XnxxF4h)$xwd%&@?s zpEhuIJqPur*?93|J^aq?rWMPbC^SN!>TVGVeecrwFVzyOdDR%wq>fYeiZuGSDt9gK z5A8z%Y`MsP{R)|mepdsq%T44P9BN2u;Yx}cHwvHsn*dX%r3m`^mkhLRp?ni496yfp zUpwJAyqd0f)WhvRd9Iw34b#1TG$*GYD@B{?nXQfA80DEX}QR{i@WcRoKUX0z7?)^ zYY=wL6bl#bA&npz9BVG5_fs-yn35B1vt34(%KoG?`~t08l0}<~rlBcnFWlyDk`})o zB{HCnQmv$De1E6SFDlyc14`j!eNjBGcbuj!`UAMdmVSJ_o|N|~^2M;#;`QF9FwLZA zlsr=xI)(-~p0x|58%=%FIwFjDkOFApiIxK7fYv5utv( z_Wco>{mq}ObFR>af(PWVO@-orD^l8RZC>9RAo`Ab_|*Nq_^|_h(d(1wIIjLq8>_U) zDQzT8e5}nkf4mA8-^9{50RiE3EbPw>OS9Ovj~tB2XJP(}wM=hBQpSv0*z6K#2Xm3$5iM8Qc}Y}6K6Bdrb4 z7_Y?BUPq#P1bm=%xI5-S;&y1mRRu3N6rXn)I0WWI}P!k-< zjxSMXy%tKC&&C~yNGrsoNqP9R=MkQ{eMZOQALwvVq0?(a*x^ROO%ZcGQ$HOv46uOQ z_H)o{sKJ)JOQ`?pK_e_C(n~IKlLZ&}hPVt~Vjjrn%=07ZStF(yJzWyqxQcZDD`qmG z%O%rO!_YA~7(UNLFYWM88rO9&RHl!{DlJVM^?gR}1D?~BCpLW2c?&K**NUl|A5!He zE#z(gM2}~Ro}Ifg@-#-#&f6_`vbYZWKXxF+D-w>K8Q3&vFU&@%W4w}-tmd~$o3wHmt1!4wLeRD@t5AV36WaVhDq!1h2rr{eYk$UNw((MbnkQuZHe}v z!wRE$&b3k8!^(_bE3$%1xhI-py>YK@It|f{rRHf#q~yF9kCBeg3->{v72>Z{^rasB zB$W}1$@zCCscZZu9d$zJCD>uV9XWhiX^g|d;BvXp7*PKw*)x&DLK zNV}YY?hkDQ*Lpab*EZAZ2~K?X_!Ye6!VF%Oy9Kce#=&XlTD)F#jFxq)rRaQ-Wq)}L z&hCNO`q2~(qm22kyh*&KE|M?%8H`PaVYu}<7L%VX<_GFixQT^Y2nVRG;+2$=Uqyt1LD)t*rF49>y*QSVv*B}3?=(hZjFa)X91*+VN@GHJ^4XqbH)551Oc z*!r^wta&lM%q|sP)K_$8y&Qs;lJ_>C5=|yLsHe z`zv2QZzp>?s)KzD)FSm|kEtWp3e)4KpsnyethDaqvTL{K#?2;Ld&Unc86y8XW)*bW$I`IBQB?e(UhIV(q-hszrNdSY zlkPK1#UAg^SX7|GclWKPmG7E`$Hf&})S|iS#auo?_agr_b~JgewqgAXPO=Nek10o2 z87+FUaCa!A!ezTDz_Xg_o`&%)=VtJiug;QATUThjEb{`|`)#^+uf!^@5g#*H}^6gX}t<;)chXBiz* zb9&6&;s)_t6InhlQJwGW=gV^@#`7}yx%^VzR(d$M2g^;|Zd)poLOA+|68_GjQ!y*> zq0J8;@)zTjVE^7+s0gnI!?0R-QZHSzhM)gB^a*)S&guv0`lWevIjKYNjw&drNSOs4 zSHj&NmP{k*5XqA?ZejF*J&5UEW@}WYE!pn3oV`4oO@~Fs zzAkSidH!C*oHNXr@5PI(-vD3w_+TqV^lqRKuj$a9qk)fs=gD5-B6c20M!We!@NT)x zRGLrsrmaDM#xtn84dVI<3;2gcWxOQtHV=@k=2uV2@wvU#c!r@CKX82&Xyt0`4hX`t zPy12PsLmCu{dh(aL*s==1k2UpXLAqf#B>>Hc4-@5XPiJ``MHv0Hx-)ouQ#5A?8G3) zOZd6{HNK~I<(XccXt%qDjqg%1sM7!kVz*&lw`GvYS%`FVB#%L+8;lS z&KQT&I4=WOKkq_E*2zdB>d*7^Hbv>Cj^5I;n=^4}#0hFw?Je$Ksbo1%WRTLR@|4y6kd=f9ub-LD-KxYc2mr2TdBs;dnh#GQXsRSzRj;yv*BdNAClYt4Z7GKe203c96p((cx_=9L&!si0&ELhJq?@i=yUApt(f=_lmypYWdoKEvh1=H<6$vs5=3mhU!F^;SYO}^c8*k{YK;YdMFKj ziX9zq5Yk7?RbFpG#JSJ3dA)EMwEl-tv-e5H=NIr-bZaT`A$c8rU5U_9)~CApR$h$)>P&=ldAguBHJ20{zl$_ z7p^wqCm$Wblf*|zEPaEhr&X-^wGO>v`Q)Nt49m@nptN~Akktia4SnHy+6pNf!}#Qu zh5Sw=aldC*C?iIN-cHkJzm7a&Cs&MQ-ii&(z%+x$-dxRXn#41r$rF+}qH|6*d}7Nh zWqSd> zghb}bI6R%0N+!cqN#T@)e&k75XvCmKQD`Xy&A%5>_lFEIzFEni-@0g_%n1LebBr{ z$9xZyreh<0(O`IZSRVJ!gwyGhjaaZuaJ?RfLe}yQ9e8-2*2j&)p&8n?7dp?e+rv)N z+6HeLMlU3zi(gB|*jH1ooGb1Zg~4c-Kik|(7VU8-aHB|tx|f%*G20v(3!aL&k@k75p(xpkAw4=2j`fm`K z-AQxAUhcN=oNVM{J!f&HE7{oo$rh=*KG2qKuPLfen8-3%A$s>)y7JBsojZ*9gTd=~ z>WnQI7q5kp>lL8+VjGRhoedtIfY>AZ`BUYcJi}3Rb#664vu7%L34Wy3+x~E56HvOPDLQ3Rog7Q(Yj{Eh@*huOHD;TaLfq!fD9K zPqb3Z_xH+4l6* zYX$CAC1Y*(1-RF(55)`xrp?R4gpxUY=Daa$h~YnRO>&(OD;nAo}W_stO`o6&gDT`GkK3DJHa!}=Zz;a zdDFV#yvK~oEIBKSHgx$;Mq}2ZW5y|3Jv^NK8rUDT_fDYy?^yg5OzD|z%V_JqWOm<3 z6*8J@ac*=NBtDDC@W(6~v^xexi8*w3OR?lsUaMq!d8i~PH;%r2nt_xNzo^UlYqYGO zn_zSAq9^$$X}#k&T6j_&)!QR5aNsj2^d1^+O!QJ`3YLrS(5l^jsJO5Z zY9U(?vBwiCuk-2RihLSdUQB0(4T48M1>9*oP8X_AvuDSf*rSA>EYQ!DtIb#C(<@s= zW_%;}4_m~$Dq8Z;@)b~ZO@-6#Y#fTRhv{HV^mh13U5saO7t5hM!|D%gzUd>Xst?kS z8zEu&DSpgvE3Z53%15P77iaJ=$%(?QlKyT*EZRF%GCV|y1$>XfhY{{#cQgTaqmn6R zw%A)Aji%XkO%!)|gka#w(d-L<;W#b>`LCyfyn7;i!!j(|y8?@9{-Yc1KUmNcck)ERGl%3ANz<#*k2vb7GHVKG?Hbs!?UTj0I8r!+3BVEJRG z^Q5nyyz;y?pY*pu%p`YH>eWKBy0eE`467(BNeSf|I}!eD11b&&qF_}f*M7X8GoHbR z$O}eg!x@-&*#nQvcl^$Y&s_CL6L)URg}dlMe7$mvHr(!{Bg%TvKB=-}liq@neZx~v?A1EUquXVy)+TV{pJm(}6${Wp~@QNZb*>!`e^E|a}F z5dU2^7i{U5^q-Rs{T)@!oZYT5{^TjO=S(5pr`u_6cVFba*@vT_AHXU38OoK;LuGX& zT($+Xo40qfDSNshaGa5Xs>-1V`8okRY)(U3r-JuE z_E_i=C-UGwQ4k@=2l*UFuD>}YF%S0n=`@j*HQ_@?PvuVicJc_vN-DOyMa}aH#r`iE zHDN!G(Z{c6c3U^m^}X@@vFhV{qXab%Ygbh3M4kF5_Tq+G`F6YXfUR_6Un zi)dDx=)LX>|ysQ{c_tHM#1;dMqgjCBuX5 zOi8{A=};^;{;`_ZRhsh>-+kn~Ko$PIg`boEhl|1kw>sGkOOJ@&-H8PDOlJfI*?h;D zhz2BwS@0P564F{RfZjZ5CH*g^{KR5UZa+DiI}2;Wy3enn==mA111~U}?Jn%$kJW6y zPBaBH2GG9FeH1FIh)x^9n5oo2;rk=pG$NThRd3h8gOY2P@Only`oc_+Q?J)V3m(@1lQJ#0kR=ffjQ+AydKz5aEW`45Vhe8}6uUVfW^ zOJ?11T&;@wpC#HbDFnV2iri<*cQ&9$9`$N6#pZL>l8AS^Bpv$ys7Hz#(lis%AnC!6 zcuQ!EGp|}?4XI8=nnTlNTl$WHyxv>U*bN}sw_IT<(LuB z9fO#7CV$AHuK|QB)<j_kRKaU9ct#sOBH>r&(p@(j*B>&9$$8aK`%OJ25P+FH*YSq8~cNbZ4wJ%Nwu9JU&@dR{B|z-kMCW#&2iRBP!C! zb)%#+d{#;Ie@w;Kg$r@Xb_tdbj%IHQZZpHXk@V|G3DrIqyo@`U86 z{N9@R-0sjb6b-L}d`&4z4lH5I>%=@VavxdV%!1BFNBs4=PqW{yqNRs|;CoP!4~W%~ zIG+ro{uADl`-);rS?$C<2PW~yk9YBez}b98dp~aeqe7xqb3zhmxrU7x=7s4^R=9n5 zHdaLWvA6ar^mJP$4L?5_r!7~YY{?qjTXBz>+_n*2w`am{lSR`6|HyDdE;c-uLAH!C z!e5&sRWIN6c>il`xI-JQY97fPM~dA}oibkCuVSyALP=-Bb!s&C$J^4$nA_|CkGsjZ zld~J~@5aMI(-$GKE{Ld5M^VEgwr+Jdidv9NI5rsSTaF;`(P!LSCJUF%QxUf$8&|JB zqGxNqQRCwz8nPr3T}C$3+nNY+K0F>bNZgHLYjUk;DG zy%C+80E4A>P%-Hgd%iZ8%0E=nh}6Mct=|-`qdkKwevv|9odMeByF)p9jO15xnWXyN z3$`cZ2Yez`_*MUIyt#24wpiaGtsw((bmchP$~obZb6V*%MsU#no3jNq30nN?<4pdq z=M^5+qk?DU9U!$eza$RB4@sidI>7L_Dkk3_h0})%Men1N)~sWKbrVKT6G~~VxJ#TC z^Hk?e3(3EW9L{^@OCmow&=Iviq_jd0d9|+SY_-DX`=2GwE*5mxv?pfPcS^3k-cFtc z#@M;9i5WfHC)oXZ2>p-YRlr>2YI|X&IGg+p8ip&|M`6im4&{`du+8{Q_qQ4gE^|Kp zQIg`A)fbxF@PcIhzfiN+0m&5CZ<5eQ9qfI!aD_FwvU6`@Y(Jju&ra!ou0I2(c(PF)(~5pa9l@%IPEv+dV_(GFh@*pv7P#W9jHIIx^jPf7JbE2x zvXhpvqa~_1ul0@U_ZY*#qk~LF(KfD|uIQG7%mTFc(j#JWz&SZh1x zDBMx+_9s!!?`HaU&!3i-_MvN||FLY}?sP5xGg-vu((g@&*xst^WFt7ziUVbt?P(Lr zJ-35Ywhe^qqHR?4>ozq_Scc@cZDjp@1!eWEWJelYp=Uh{8xr?p#rBi7ht7r5)Amhx z%xWoYm3ZH4gXmEB3#MN9fUH$^)AqOhsJrGb7QafCT_0_O=B;LkdQnEs^CqF|v34rf zYa!VWZ^+g*mlA(2q7wmsaO&n0xZZsNzm-$4!My^rza}6-%~Z1dWR%3Cj}psikg}*$ zcdEO2irx<$PYP>JusH|fn5pVhIuSOK)bCcZn(}rQ>TOEbMkP^n#!a@zHk7$3+@N{? zMX-UFyx91IR*YoB*yC(t$@s!}I^I8$&ZXMYqbX0>82up94XmVHLx*7V{N1)|G74Dy z@j_bKZwML7)v%c=D{0P=SP7e5E{O<@l<*;HIL5m`_FXRuv&xpdlsAzy{q&TavmC{| zf`TP8#Or!r%BNXx=26>|ET&qppV?`;vGLLs>?1vAF~JSYp;?ZcXE`Is*aU9VK9lK} zVJIzif~lpI=m?J!Ec{eflpD=<>=}d}_q7ordO$}{{Di4sFU!yA&7B$|Q1E6Pek~N= zc_uphB9AZQejMJteW+jZaN0XBm;Urjq@RfuH2=FE%=%{WmVP<>!?#1+$7&bVEbB>q zoiDMQtHks1%ocL4)};#(iRhi*j8~gg`1vwh=ufi2k5gUn*JJ>Umun#`K@QI%w&G0h zP`p(-jF7M`EcK4HWSXi6d;hu*Nw*cUblhiOYC|!(Z#6c(8^YBDgDJYN$g_N(0MAet z$~>0O##@*(!%vFTveTKodTgWBv5RPXs3#Rf8B)BwuJrGi-qNr$M$$}i_o)jm5j~!8 z(d90sY>|0iW`3R~i+{_hvnFz*lHvS*_)fYwJB!}89-y7G^T~X`C%W(OgEnq^MQe%^ zD0fQ<4O=R_O0@xO$e_2jimEzfUEx8_!=6z|oybB6HhiRbmU!G5gM0&LXcP+<*(QHH zG?|UH{l~!uFLBJb4OZcAA%Cnd4WFe(w_S$Nqfa#u@+$NnJ-QpesAp z0f&!5-s=p8)@{c4sx(adm=4DPGt&AtoBegUC^_5mn;H+VCHIz@G@`nc_8j80e|Zhl z{Ij@g=Z(Y;v@(NAs~%I&=6KpO#he=LR$#kp z43z0QR3ohT)py(RZl)aGwkh+0!xXsH2pK-)UJAu}9HY%;+EAY_Ja?j-$~+#DZ%j0_ zMb2`a_Y`VBy_9>eCVpq;A|9=Hh}|f1Vn*kdu;9Man33Zpw*JjvQW$ptzEzp<8~qlW zUhZP2mTjgr9)_3|^o-42^^7@L9jAkDrP%c7IgOyFl%=&o@asMB{EjQyKD4v)R3GYA zaEJ;*68VnQF1%dtD!w`$!-Ls3vH12)*xG%dfXX~t-s3dAtGh$~Mju#MfeXt?Gom!D zmE<0h7n~o9wZ8Ft;mTrjv-dX9mLdi#%`r+n3jVoXuy6`RSYyE);d& zh`N`r5zh56z94ib_qgQDVune`C-o=&G+2d_Pq7%Ya}A_+oAC8{3q786nU?!ppynnRANOE%<{9CeR^}#GvJlXkfjc9DQFd@QEPw34TK_!c`kfMN z6DJ-nQRGE_UHRM}#kg;@8Pn_U@Mk#}`G3mg{L)T49`_DR6CR=Nzu#-Qf^us?3ZtpX3Y}ZLV znDc^UP7uCW6e3+Ah!O_Ju+dsO>CA*Zl-I3@mW~;O<87;0FFPGJ+*glcuGpevyMUZ3sIfU9Q7E-qC z8Jf011`?%e^1g0Qs*Ajsf=&=UdbB|FiKLR*hq{o&T@jOH_rm_^I_x%egv{7x`p{z~ z_Nz9C9CQzqG?db7mk|2?JC1yZ3O3R~(XCF@rfsF~DEvb#LR2JdhyO2Tr+nh_;}R+8hd3 z{76AhlK4=aB%c1+iOUVxPva)WznQ) zq>iy`W@4xME_B(QhZ%=1i(T#jn6&kUOTti?2VOx%<{y0U9l_TswqobiV#Gx&!_Xq0 z86O=&UUPzJ{^T$iD?Nmz(tVhS41DhXR(zh?kAMDoj?6_q#cX5*WM#W?jdnl&NbfNB zzk7fi6y4_yy^HvSwrZ*oPREfZbBMH8V#N7XnD=?QaGf8N6z+J)^d5WD7?bUg?Y$Mr z_eI}&SOJZhxr2r;eaWK#TEXL+@RkMEL8*^BE_S!anv6B%Rk?@Xgq=VYpN45#6j>&?r zH{N&<4Q@4JXX_-a=JJcOtS7@PHIy-*xlZp5?$SV zKHy6%|J&fh?~Z*6%k*SyJ2y!%Rc15#9V>Y{*@4|E*n^Zg$vCleuW&Ci8guI+>*hF} zRZLtY9cXDI^|MiuK3%;T*ZDMbj~Ric(bvd2{tCtSJ3xJ3oZt;V@_A%_41aTJCK}%e zM)8L3INCWG5iYj)G^~^Ii(=3#N*@Q#meSay!?fFBKPA2~M8PFf9Ns5rQ}+$%PtjWR z{$+}@5xo%C7zatY8$K8*LdCw8=ATQVfIWpQ%fO$!8@iF|sT5X(3Q5m?$&%R_pKM1D zILS}0OyR-L*Ye%IL2P}`T_jU88nZZ1cNOw8TtgfR4#w-r7BA{4*Mxr>B$F z#^cODKAAn&iD4gKh&yZ9X(o4SIlXgF;kK*7`Lt1k#Qt_D(kh~mdG8^VEd@vG!)X3q zJX3oY6hOJ`5?;*AM~L4pY}+o&L;8>8O{o*-7J#3b%5l*Z>N(Ha_IMKGLQe1&W{IO=D(_kW2c7Lt1cLY z1>P$usKS-KyZ>4e;JpFU2mGcFI%Y^?B~&=0Cw2?2UQ)LhsY2KR>8r?7(osW<#UA4> ze#!rbB7p#|K0r6aryZE!qOSrL>3(h_T%EYe0C?pU^XF4N%e;+sxzf3N7 zw4gQ46~>K+QClU)kCgocpV$@i+;#!Kx8rKrKHf_$lYbkulXn)#^1RkNa8b^I@um|r zO8BKJ6Bc5#odIlCbw%}aWt2AB;2}Xoa=#`rO*d`&#+(5srhn>c%D9tuu~qT%N)&pBCTky1{Vu34z3~l3zm`--UmC z-nPX&EOj=Q2u}LhV{=&T{AgO)$zZlO3mXR?K!Mu?I6POSAn#jjrq^dWWi}tyfiZ{` zJNNPn)okH?PuB7$fjRX4hniP9{FQch{zW*(eq|Ss+m>P$svXUieJrDbqq}i?wv|RT?ab_VJsv1zy zl1eI{vnl-c6Ed?(rVvxY&l$quAiS~r1vlW&rdpSY(Q)?@7;!7NK+bajX5Tc(?%Qk;cl#SZJJ@f;lbV9AE%n6vWx zyV=w8w^2N&8-HhN%scG=kY(*_%6YSeYPHtkX6<-nmM+EB(dCjab0?Erz)lR_a9!e( zoJ{lnE`Xf75uI;-!b%-CvJ~0w$jID-S)&?ZU$jBIr$~_j-wtEX1S+$CMOvQ>QTIn% z!(I`IShWnhbbh-tTGulSE9~AK-PdL-wDq&$-JIOv- z$FDEDz^%WiNF7uvuNSNS-U-F>vXn~aY9S)n>3B2H7BA5Tqm6tEnnLPr$22Dl!4}IYhM3QIFfGL@#i0_=}oaCX0JaF z&DvO7%~KZaZk#naH6_xkMQ4G@(D_hNb&PW(y0&~DHk)SAD93I zuRj!=ypz2;l}3^?_Bj1_4EkpHVA>i&cvUo2{>`L=NuA^|z=YOrwS@bEwW9m)k0+A3 zIG*;Dl2&}P*=_QLbsn<7*nU|O4bOa5_U9R!5q+MOUj%h;o=xM{>qCF17w+ndbIkHv z2(0{yZb^^uRcAeN7o^~o(Ozh%Um}&Z?X)qgiH_}`fny)OV2Yk2ub?BLD^~J&?HB4-tF6vX2A$vrTTh*s~G4pw!t7 zE@?jGGrA8RFZoW3uj-;-T^$}u3$S&4F7oFc=l!~7@*c54e1lgpruDyu{zqDI_^#lI zRvIJIQx5$pj-Tqjfp0eU;a$qgP`%(B+W8e+*=i`g_5KEL{1`5{-Fs=$s(Gx($o-N# z@%a=y;uEF3Hh}$;9(-GCD^>}&;SY`8Vh)!@=3Zy1-2V%vEz84)xlXX%?Z9iZCH!N2 zS01oL?88jr(CgzZk%y8)yG9@-t~O?t?efy_jz)eVt(iBb=TN>~iP*_$!|PBQResK7 z25}j-kH$VHLrlZJmn)zxX2~iujp+IidC|A{A@!SGBh?#GBu&&FhXKW3=$N4~RGmKv z&ae~h_X(qWdav+)=PB&ftAWbJN))b2fJJ=-1|RxE6(u6aezK0Twwm$6dL#ZfRh}Q} zWsK7Ei4@#_AZ%pwk!`aDUA(*zUAvtRtJ%i?&R@uRhz!!YIg#hQ`|M@?U&^`K1Jn_{5wwvL+26EOGDSu81ifnKLK&}>PIgk{^1<+J%HbeV%! zvy#y|MxFm1RZGZFclpOCg%sbJ*wMCPS>7Uq=9-iuC4-llpHs>vq18COv0os86oDA;5- z(4J)R848)tspo5<^x-Ii2HZo{gD}>);HGVKY6mm3ZliqL-}LL26lq(`;J;iRyhm50 zoScKd{`MI6O9S_!OkvlWNVdO>*wP=ux8ahA#+5zWLZ#%DfNMv z@03L!P3>|0+a(%MaEl)MJY$w~sz@7i5#5qbQ7dJT|Dr1l%AO;6P78`A*Ft)}23KZg zV$;WH_@$k|1@;9?tL1o|=_8U6Gl|>9O{DKle64naV8Zw#e|vYF@o%K7Z?4d=9hyAm z{X8xieU-mhc$S%{t(2G@SE7U7ci7^aCgfhKh{gr)F~w~SHe9cv!Zazqh>n73Pyw^p z{)U~bZJ^_RiP#=jgeGyf4_VTJ%N2e|yeRg8p6#^ixE9@>7eV%?RPa=V^OZ>#@u&X? zgxnp6EBFGpA{BA`uPnZ|t76c+<>k~k&Uys_7YS6>1qdzV>_vbU`>Ahqe?DgqRDsi zLkj-$m*mrKk;m$uVrCve@ngZ~3h(jkcl){CCULH--vYljWn6S{;cL6w@DazI`MgY1 zl!RZRgDc+9)1Xq;Ygkvdz4u}oGL})W>=in>G7+bKo|Q;kx6y%e2Qm9C;PSPb`F)dV zdZ5guPKz6?uh*M2500 znmV-EXgd$I@#MXQi(~n>etdpz4p+O-$(PzTlk~dL{td`p!nQMD(5%gd5mM!pA;tgs0h2XeOOtnDtdiFQT{um#5~61^bItx zUn{#6eOyu}@x}h*faZZb(HT+ZSZB;f$hz|Vji>0v&{XQYt|GVqIjEg}0*^;$bY*O4}!rkldh%|veQ;4X-d5z_4{wQ zfvBc`k%LkJot@C9*Qhh|H|W zrXjPXB|;=Elo1sTT1LpMkV=w}cSd%Ry}sw?@84g)Joww=?z*q*I*;S^dOq1Xf9d1W z4bp4t6nS9D4IFfy2)~$LbglLRT^0GVsR_Z@qb+h!@6$0!r+~KH=%hgOQoI*Zj z(TOK3=+BUG!d)Crzb=n~{izDfc9Dfl05REg2l9TlpJXGo#Myc>QdfRQ)sj)XvTZWn z2-bn@|8n0N4fIw!lWNaCqsD)ud6#Z0_^)4+x%x$h2hr(pXWQU+h0`UwK{8F?s6ZgmwOv^+}ca8hbi%me;K!ZIiIiXF%K_>eumph z7vAGsJa;#_z{5Act zLLcfsV?8--H3T*;!M63$qUU7_&Dthf@GgU*Z>WGjo&{OaS!>TrpsD|KD6{>qLs+j| zd$wx^JGSgTHTN&2w+CJc&aW&64Jt#yv`bj%Sd8G94q7WdORXj8BwyGQ!~U$t&X!PY zNGW6&ryOQqhLtf3%Qz}LVnJ8;cOzSmy{vJj5o`E;i$1BW!6r!JG*#6Ulmw=_nXKKnDe6<0sM)7 zKO}EtFszp%+AtT@;!niZ9{fv#C8ds@Dj4n(da_{}j$RysPYXuj(NzbopF5B*@V4Po z{x;JL_L2$`G!VEq2Z<$#aID;k6KN9Oo-vHSI^e_IOh-^*tSxz698Ia$C(#K_b#gl( zPw)Qy;=TU1aIap~{M4Ls94sgjJ-zezw&*^;vR}A)jL&iDUBTJvVZd*#^5=WUB;#4z z9Qecz#`l={c%e57mD6@Wa@_%b>tx|!l|=VKcH?7d6e>RM#pSX*n(TI+l&7`Q5`%tt zwu|6CjKM>2jKW6?uh){lkitO@_eVFl6cr!j?@`2}Am640DYcs%?m!Y{|t@vCR zX6Q$fuhR!w)u065+PQG?`2~*|+Whv(ANVwXCSJ~7hbLVY;mh1q9;TJg-$i6{lSQ@k zz)y+Bf3;&aMw2Lh{UO%&;*{jxO*!s!sU5S8>+s0(D(xy)g-h;OWJ1sO|<#lZ0Zp6-(R6ocyKuaihX7HtHV!l(5MV9zO^WdX@Gh6 ztGMs200WPwv}U=e)VXx9ul6q~RIQS#$8SW>#SuvCVu&ZlUP>3MzLRDxc_OVfN`v@z zU}3-eq~B;kCq3I}sc@8ZJ9z{eu9{rtk0n=pV^33}Ybj5~2K%C`X;O3{U6?b6&Kqh% zZN^`6AEk+K<>~y*y*7AHvk@6gHQ4SrN4F!I=)lcz9v!)wA2smcT0UY2AGsWrfqO8k zmlKQ!nc~tFE1353!2B+;Q2MhEH#3W9K#&VMuP(x$%ne-LDUQ1e*@1(`18gweCVJcX z!n1Uo@}G^S`O-yHr<0GmoqORi(+kflZTR@EjNdYw%!9xBqT#Iue)Ta#rD8Ql)lD9y zThEugIKZopY~>*auW?kx1iRBOP@DT(ma#noiyb8hFs@?*QU>AO9^s|>+Yi6)9itWT z;dEm58ft&#MLuUPk)JrXJWEJI?zywrF{cH}v4o{MAC;ULb)QBke<0ZkBczJ{ z@eZ9CST-yP4{W~C$DJ-ve(R2@Lt9DHq7T|de!KX%18#>KVN}z6j4$v<-4;Kzns!Hy zM*`K9-lLcFkq$I<=T^oteC@sS(9s7Od#|CIlor~zeHc~0JIQtoj%EWSG33n*SVFK9 z`#wszCpQ>V{i1=C5VH*HqpQeZ+<3Y+Uk84oYvlBDEG(QDosbQs=q3faQa%^gWaeSr zvnkl!o`$hT*{J?<5&ucY;_(VEZ0I3&qmB`XNzF!Aoin(%DH3N*+~M}c1xrTWr4PC*yJEGQiB&FFO&>uZxqE2fGW4i`IsfnTV& zFdHh-M{ppgR`~bic}c-@ald+woNM*0Cc%ts#SV8!jqv;ps%N{*pGrK%j&K^+gLr~t zYF3FfGxoA{j_n<3Mc>VM^*I*R$Kp|zvV?*Q=F$7)V2YlzmR~Zl=QRs{VMkYUSjU^A zVU4a}&&ffdr9X~6?T4`|4&l6k8;;6pz^XI>0TOq#$0bnMiXo66lSd2x?jwh-PTW+< zk$21)$4v`LVdB37WkY??_iZ1%uGWWUBoHuUD!6bYJvRS@k13jTsy>0NPHRD{I*i+x z4d#t*-Ff2J{=&Daj&R{X_;k^SlH$e@&CMn6OD^y}8;YY%iFlfu$M-~?RCv{|)Ew2kP>opiEYzKbJYJKfwI9@27dUI%sOz zd)jEWihSH#*tTu;tj~fj=#?`Fkv{vNqEbgY_2a1P*9BCSsliK}h11Nr9}n1kLGYZ9 z)6$9pHpwWTigZ6o9u!@qoC6Qweyo;jkj$6{XfbePyrrnestlJDIja(tRY zSJ!Ie=7AW`!yv9IVu%Jr||Wq$)7u8Zq>Sta`3lIJtbq0K&mjFyb&Qbj=Gr+#<59-z(ALFqwap4&iQZ z$^~z)maTp=k6r)ghZ%n=;dgQeWEDg3-XmXhN`E4iouP%7hlqS*9A1d`Udu&?X}@MJ z8=xX5(Fy5C%BgOY8Bv8s%TzR-N~OZ|KxyEhcxkO>m~@S93yxb}#pv#SFgx4HR+Rcu zii!tGXJWF%@+dkaQSzz z&My4H&qo+v39NBUfRc~MuPTIdwS94XZL2l^=W_tr)0`+EF^YL7cu~Tncruz2LqQe| zkXyI_#rK=2*=i@I*BJ2Mn>KOnmPQBrCb91noROb+h{*;){8xGdFB;s=gLTI8AR7su zi0^3dg4481VBIct9>J>xGwEJnk;C%EGO&xB0>|R&FFzn+i?7FwS(ut4P@srgnVZ%rBBH#Y26=h zbXy@DZudR$?Zi4#9PycUyM3d|$n|t5YZ?Wv*vUHe33fpi2-VR<_J|=UZ;r)+k3kqH zvkO`kf%q=wb6=GhTfKJ-yD*M1-(rzPTpCIjE`*W85jnnXc?-OPP9c3#AH1tD!Kj7y za9JU|nqD>3(WHh>mH%kjh!a%%eji=YeS}#TO?g-CzI?{zeeABwcFrC~8LEEFuSpwkOUBUIh30JS%AL6KArjmEI$^SPCbW+v;buhyEY@6O z$4)j%{PT85B)-2fBTa>qV<{eN_)Zq16R5Xc0<{ai^S3m6oVoab=6(5%*`{OoEa#y- z`rc8zs6Pg~Ng=o@^N3RS=~2t?WvnVq1+kA^V7+`2zBKiKfmSZ%H1wnfjWp@ngjDHG z^C;=lk>SX^ZHk+b8!0fXg<3v*qa?^*WFMvbPLdcpitf5F(eMls`G@phLQE~gU9 zJ$esD-c-NLS+-&-iJDe5|8%|cz%Q!!LE>53zMQZI>-sM*;|IGIC%5$qJTemwM z)J~y4fkAxQxBdLpq~qMp%8L8WI*98VdSmqC7#fmvmB!S4q40`*#D=KT?H+{Fmd zJ?0=;uxSQ2I+J4SBX;=vR0>`*13T>(3J>!Ld>*wE-IX7s&mVby#?qG!QL2-?+8QBw zUFJq(3eD-_eB5R!S21gmn=#bklxLOlr6Hmp{M#{o0u~vr+34&_veJ` z=_)gjdTXx;Xqko6+ya~^-aNG&Z*P4#ZVZC8`wkw8jJxl2g zTD*6aA>VtrYVInSn+B|KVhFo&?-;xBwOV>9twOq$&PqQF3qtS>1H`6$qD6O8DPcxB`NmJ8 zk@2B;y`csblQg;B87Ug?+FBw$hM{Z8y8GrvW9nCGkv@iRhd?g>Z1sUOeunwy0 zPE#-0nXu6mUB`vMq8IU~6rB#U*neThh+s&T^Yz9WMUiyfxXrcsCq zS%YZXgUIr}$akCM@xy|fsAzHszKi3Kv~Ul!K6aHx?|Z=`uU_KUmS^KdykPrjHPPx< zn{Ywi9AE8KksY)KXUxR+%c>t$yS=9!&ARC8ZHxiyOHg>D0-A1daK6(8dyltMV@v^g z%lPBTcx&9wX{A%Q_YgB$#2&x#b9kaZm`?w;qTrAeTJm8seB!NP`11>;zX-w;dkGvY zG%=+6BiiKVPNP+OvTD(F85(7VJGW$Eo1INnSN+N0PzSqk(wELh6jEKd9@dqlVqd;+ z8NC4tRl9NZIkLQUO9MWh(PhuH)!50pkq+|Cc39MTispEoWV=1m5aSYl!@PQf#MDx`)>w+I<7ZQN;-%N=VRh{m4qM775weoU3_Nn zFuuCkmh@IGVS^LeCB93YF?qEN*rHI1sXHu~fD`a4Ngu(-LTK>cWR~XMjSaPu!Q=;d z6z?#V0!l1+QgRP|cup%wK8^y9F!DEe%d|&;s($9PFQE|5t87wci_IO9&5I@qNJdjnOU7-F>g+=^xAysWA!L$si(X2 z_KhO+`B#9a(#;^ZPh|PEgLaPDNuMGn@^06K9<1D$cZ=QvISpZ)EDKT*whD~hR7MKGF4 z+2}awoLoQ)^LDV{T`7_&f=QQDFB~c>&I@*nD;*dU#<2f|#I)R*tNUMunK&QDC2wU# z#@Fm;F5M()9U%kbjuKjBBTFw2X!4nbhP-cbf8LTco~tkagCN^1B(0u-^{1ZF`K5E| z&HeR!SZ@cuH2eedUmX^+-gt8R*o8IEoq!oXMeneGCI$aaq0g57j3N!#qEEXpz-R`x zRa8HktC5)-g>> zQ>r*NgpKU-!eMuEKDs7uM|SKe40v%2g_bs0U>`}3uO`uFRTsKZ8Nud+WzouZf76%AlXCf?^F>SRB=6qure7w(t^GdS}w5CwAOjkG8+q#r+32 zqECNaSS=Th)p>^nMRTzo+Xl z!DyDGdSjC0&BHy6pIpu@ozLOaYfFqeJPBdvyP)yL1DgDO7Oi|cfE4uaQrGpFqOV#` zJ=Z^E(v(t`d44~8y4Q`q+?piqOx2RUx0aE1)6}McHRq^f-hJ$Ec|u=@{blzDWw9u} zp5w%K)c#Gz5$zmIANzw=Pw7h6qZR0UgE-Fz{A0h)wUTG_UuNo=$~yG3V0BZlNdK$h z6E4KjCBHM|d*dU8trPo_G3B@`=AZ6v8FcT@brwF)nx2$6GP&Xj?5X@c7M|3?jE%Hd z_Q_CcacXp!IWIzT(|R{EFbm|{{O9qlm)CK63na|li>#@$dA9a%2iq$_lHY3_*oIa` z?j`pdzrX5npHN4>X1z8aA6yOP?km}rEu&ca@m5Ki{V15;Xs2HOmuT&!aQvB~$WQte z^7(xyU}Mh`8jtOCZ=)tXTIEc3<5u8EaSvRI9ZDM)Mbh7;OfotxLgEuHrGX;j+b-Qe z(M2^hAYX&kF6qKN>Q1n!SGSV)qh2CjA;&X1*3%SYJ39VpkI0P7Bm01x%pfd^)kmqZ z=P%PG$IZ-`bh8;153#2NS~Jr_H-^B^Ta$OH_K4NHSf;>kP!b zGP#^CMFrD8;W&tS)SXJ!%A#E(a+Z`9($OR*>IM7HI{xU^RINgfIZ!NsSCVD&bIZS8HvQU?tn+pcyYyoa0l^aI=ZpS(nV zwrDP{pPr1RJ`tLv}GxduiJs+nhAK!WBBn+X`C!F`P7`l`1dOx z$sVN`fM;C(VJF|J-bK2(K$Bk%u;XW+oAT#_tt43?FW3d$8X6(qrM<+t_icaSX1ROI zVQr0;Bl5q?$1d*J?ct1>edQ@{cpZqwAQk;vu%|{*G34bvSMnp0nqAxLnahJyX8XE(0ZeR?Z=hliO&l z{zLNRkD2cKcs6LnUM7tkNf*XWq&kIgk{>t-7PZUZBf6@II?o`p=Mj3e-avkRB^B2a z$?6^#9;ZTz(iObW+)#|vJWc`AL-FBG2QD8L`++X-#|HPZX1cm zdQ&kja5o)S{ELj#p4|2NNvx1Y^J;@+{%1xw?-{WNp32*?CAJ7dW$!cdX$@qyzb}G> zrzC!RGOv!y;udNi_~iCojEa59!!M67`(4Vn=mV>`Zi~H`2A+E%>l;Cq5Uo z(`7{){>DXyx34xAS-x`mT|1AwmoKNY!MhMw8jT*mbuj}i7_<@gOcx!GD8AzI=?Z0|It zep(-ys?X{D{QlUad;lg3tRd{k(&0@c4H*mRknUgMbiXT)e(nvkwQh7VJ&F!`9A^zv zt>L%12z_lYWA&u%EX>~uhit##;sP6})pkHf*$agp;H@ zMeJ?I-wo}kpEe1XYDVyjJHz<0yZiXI--~I&Pk&6B^#BuGR$`jeow%7FCBK#O02Nn$ zy!Z}eclW`BnQHvovu^x&>KgQmdyY*TtaKI9!ow_wLL>=n>*c@Dmt${hGFIt z1>D$|MB~oSg`!72CZ?ufP?ZUNlJ>!3i%QI~{>ffQ)f(dUWoO>dIFR?;KA1P{cj2y#^Ka`7 z_>WZvJZg<9zm(sCXKC5gP##8cx}bR-vV2FNEuXl-i9e6KM+rBK;6B?29pzhjWsnDN zpXb6o8tq8EroV&ysv#`V)sUH~2GRYu*08x%jJ(NZFcL-GjbF`iy1JEy8Hi`)11&^8 zjVGPy(WIXJktV2H!R_J%r2gI{@{S{*a%e36)kiY@l@}x<3*NCa$~k;!L=}%uy~A&e zilT`!(e!m%ELA%kpv~R((J;|1I2bsD$t3lp?y@f^O7EFu$IL`YukmH<>5XLRvgvWs zps8Wf4OJfOoaP=ndnFRClT%nrQwZBBTzML64CtEqX&Qe;4k=wEkc^U|=C=*B*BH~8 zptUqTE{$C8Y)8Aupq;(C2PK=gL;uHh$V$a$b}k9Zk9NRNdo|RMz|+siaL33*?lQK6 zZBlb#eGSG4c8M{V;QDUbs12*JIKKEnCyHF?gRB`Px{JL^K^01Ba}fJc@U34QCP&4An7*!D@P6{_ zW!*#Bd##g{=Woc_>27>*{44mcJ3w_dt~6?5BP$-M4AW0(bbLV#UAKv49dSq4&aFde z;9Zfg>hux$#v0N*b57E0{~S_w*TISV_utOKou=u3rK>ABB#-ncSz`zD?6-{u z9a+O$r_bPt!L#`AaVFT~dYWpa`GPOpL^0j;5F$7LX9{Am%S7a@<K5N3cdy~(m9&O@3iiX{&p1p=o(e_Z7;IT=iKi>l$kSQ)OuY9~{^R3x zPbU;xM|DH$?fn#=F^^V?URckFyOhgY*queWOfmNb4f^jB7TB`u&8O`J{S9&nCoMTd|pS> z{p(0Yd{4T+IYv8u@6or@j0XN5LMC&(X{&WFj4N)UEzNi7??@T)+wzWC-0woC(rRGv zx*jSCB{=%ZkLi17NcO3 z4%9TPFWEloO4S?t!*Nb;q|P!$^YA`Y`e{CG5dMn*w;7~AXfX9_>`$>99D(@^T|B3V zeozcq2WQgfQ$NXNlP87dE7Px%8O&_UJdqESP}sHZ^s@E|^Q6bjys4B`pRVU??VfVe zf6w?Hk(b^&V>fN7JV`23ZZoUUV{GrcME1|@1x*hrrh&h%QOZp#+N_~UHMhG_`nw<| zGc1&ahn{5Hk`hU2&T)!~5DsQTLljOpMIrtAQ%S5AIsbdZd_K!j@!bQ^e3cGA-(-~h zTtpGZZgf6t3^_HvVp&U-$oI(vN{=e1j`%CYI#S5)$UzE;%O(AcbW$29+~6X!&dm3U zYoUm49VLodrb+F?;+UmYcbfUagT~LfOI_Z!Qsk*0bZFK#yq>TdOIM^|;`jN~XM*qx z``)HtyGQsJC}tvKHF?a$cwC;g8pgVMc;L5#9S9AU{Cv1kvOPeX9&}`}!+U2k{i+vi zgZzD_yZ<6f`d-VPoo#0;FUwHC=zMmj%MLc}%{kUv^%=cbpoqkUD)?KI&fXVxWB>hp zCwV;Bnoixj!$!TZdP~t)GdNLVOkCFWAaTz|R<048wOVBk=^fgB-y4|8FF}_V^p_zyAT9^L;UzHBxM!tF&_FUhdMinlIzy`J~zotm;$(tBhoE@u{C=b{q9<{yL12wyy&Y>r-nFTQBOM@i+8Ba(?TFCx?^1`~^7 zka~UyjhXR}MJVJ5R#1gt$xNd1{6m5roXTUYqqxWD4g9I-3N?!?=-cKXw79%yb$fK! zv4Bd+*oL{n&nd;I--0O=H;HDN6);a{CFXW#sr2Glxr~YrN5yW<0<$*#p`Z#~GWCAKY*#n1q2IfRJ!&QDjpC6M z&FEcBcN*5BMy4x`QLm;A)#X}P?5eLboQ^Y<3PQEms4ezr@Vs>exsOzd>fS!+!ym8R6U;StTz3BxA;Bs`Rj z$CR0ZCH|*R5>I})BCAd3(Kc@VBL^rI)Gsj9l*12Df|DOE~3c<}# z`E(V%TsraELy4zFCt}m%T6|V}hlq81v2t%N=0xs8sX_+LQ@KbXyLCkNCX#KQS;gG{ zWKmAfI6he|kq388<0}eBLssO7SEU-WezTlVe(ngx8Xlzsi(SZKN-y;6^%i&3Y!H>S z6BpH=VssZ9=*soM$6IwOr z#aJZuZAWcI3@&bOLck;~-gA;Xt|?xo^hS|W54EEXrMdK>EQFZ^Z$ek!c6wq}$1?xZ zk{(inRIAEcS|imL?7#x#=dH!#AH%7-dLXIqUrRP)8=lz#kbxX{HUWQ&ovUfXYKK{Cq0OI1h{~07{OnJM)I&% z2e|+DRd9Hkg*W44pxdF0XNO`@dZ7j#tInWH{dL?jxi7L1L8wh#h#Sd&=zi-+A zk8THHaOe`U9`ENnzSr|j&(x&Zwg-6MSCM>qPAHeM7i@pp017!XmD0b*Vx^z(ySX@W@_&4Yo^BJo4px(=5X{PF7`dy^NcTE_=kM^Iy`(-2|?T3NGm)c; zks(=^aKXnrPxZdp7_s@RU{v42i@^qbM~o)V484cWglas|H01XF<~;3p9%&xjLGB0Q zsBM=S+Ln*Q`0`|E=IdbR(#24a-Hp``CvoqMirg>l5+bW~sok?mGS<9?h0feavz83R z>Q5K2X=n{%yLMn(jshRjk72WY7FCr@B{FtIA5(jHhne8nz(XQ$A!e)^3V0}bR51^Z zpwr+iE{A-k=ijcgcLUcs)Q4p7?N9IWhC6NC;Y4pt&lT3};!~);>q?9EyD<6vM9|)0)!FT$ zsJyGBxM3DH4L%JSM+L6AO&{Z?f1v?A>**zSk===MI%Tdb^1_et%BKSz#qFr^S%uiB z?=Ug7<$C{Opt$ZF_SGaJd+7|Wp+bs9t2&g&;rg#Ag5!b+@ zr@>*&CDgSK;q9sS`N^1eY?Ss*zG(-1y4J?NWW>=RTV}?@h<_$~X$Sy^(6? zYhuMbU9eSiFk{tDe7ZIii5s-=?v526&=Q)E9#21Cc7^)+Tktw|9>b~*VeqUbEPnM2 z@1j)r_{ZY&GaSbgJRG>}O)o6#I}-GC0fKJGk^YEq3Qie@wiXM_oiPis=l5aw`DlLI zb041`a+g{Z8D!Pwva*T<2q`+u%B)u zeT2iFcHCiQh}F$NFPqcYx9Tgr7mtL&_;6Ht$G|(QhE*RZl3284Go7an{I77|jw?y! z)`F!yUNIdD8xNrI_EmO%u?lUMq)_;7C)zu>lh(HQ;g%kw*`Ja~O(hxuTJch=8{4I6 za&c1Ykk9y&*qi?gUdR1MAEnn<66sa(7wYRVnY+K;&1c=O;v=k|v-?&P*xkL3OkuYYiB8)F6Ua=BpsS9-Y-)L+c%q#t=Jp0#g=4Qnfgz>;30ptRgsu#?pJisg@yk!kjs0`4+Kj9rbmmuNwc}oSi*?^plcRB2~DPaAc?qSJJ z4s=mT9X57<&_zy#mz^mDgeO3TXk@VU26^QlVPdGmw-7?%c%?(!sTel!5V1Nz`k>>w<& z8Y*}xr^sk-A9!C`A(=7yCrkPFgKBifO9KFD_RYRhyGQZtY?1g{*ZM)nawsi3DjXQb z7b(nQ8^)w&LGDx~#^?@);$V^g2p2nar3xnZK2Bs8jtV|k2I393BE(gSK<_rR=uPBK zE7JJ;M@2aJ;xgm|V{u^1DE{b8cYfrWj>uA+ppkD9sP0xZ%AuygFtApR-{o>i6nHAud?(z#O5n_X&-AGm3oM zn`yelnE8^=Je^#@ij zOyUQhoxq4^JMh_YJpP=%Oh%heQl{2K^myk=F?Dt_=y8+6y?~+r(UHWtL zDt*iKz5(>Ysh--5KDf6foEGuj=amWVT*ULx$XZFl#UKT|;8Nk(P z8kaw3#p7N-6MM!+Z2htlwK4MOUIo}tq2deQfTZJ%ObE$K)7 zp3*-~bvUf_5bpljxOQ5VMzBH(dp-!Ug6lNWU(mB0Hdqi zaKU2&!nHbSqN_O!W4FPzz>V9v-a~PJZIqZ6v+chOX!7lCR2g#%$9h#^(zWaOol-+j z224P=lq}>=mBM|J6BL4`!}UTVMdrEE{3;u!GrT88SBkuX;!~2j`9BxuA^v*B1-_|A zBwgF8B^iD>Sh8jg!`#0!1^;a!R^JnRX)Q-S_=JRSPPWFA(;oP0n1cknB)TSkx2s2$ z)A3o>yzH|(kF8t96GQg#{Y8_xvDIh1yH!NzC2r#%68H$M#Pmqas;+N0L%Fgk+u`6J;ay)=0uf37f!J$^kxvLa;wt3UZ%;uU;K z;`e^HvtIs3 zM=qK1Zsp(b{QCoZm}`K8l^^IczfPgW>-k9Q)qM2D**xfQA@w_TndOh0!49@c$#{`5 zId}G@hAU~9!bM(KDG8_3=g?<^aMHbdpPJGZvS;p}SiZHGS0>f~o;UIRN*OYgj8W36 z0lg+=1k4dRl4K>i-PxUzEcWsV^JBQnB7Z(@NDNK}?#JAkEGWqgpwoTySy9)UlD_v7 zkQ5Vw2csqkC-EK%x)n(yBTQ&f^&M%x;Z~;=7em`rR5OO5Pn~Lyw)<7S| zZzJ^|k15x60&UaSLcOep^E7wH7Y{b$K^r6p8*GLxS~Bo?yp;+T1kznCu?GJ=hy78o zWbv~vu$TAL>6d3cEB(~U9w&tJS3dqcpl%CS&`zUw^=0IqRz?>Ww$oU_@0^sooW7Wh z;<^e$`177dd`!>jJV!yWaeN$jKp_M7=#E)R4Rm!~K9{M<;qmf0yjJQ7gVpO%vvN3| zdnVIo#opu`B}?ryjoA^K4AxsMfO%#7CcC0iI+L@Viu^Mu&c>9gUZ=A&;Roqb#4@_v z{eMo`C``P!2HU@lhRZrD$^4nQk_kfd%PeUS#DPZ^nijbDl&u#^z>2-h(PE11R)0r4~a2RHv z{)#e<|ByeZ9ly?B7G1=Aa$gol@wr#2_K>{j_=KUdPm9QApFrn4G0U);g$M7Paj)D3 zyDg%TxZf4lpSt11vul##KTeV#BfHYQ$;;_q*d`ilX+#&UX>j|de*BxF318ePI$pvf zsyQ_kncZ_K$;uoD%+ulT9KkBK&1Y?kk1)GRIX+gqH=h-%%wM#&&@Bxc`ldaRZTc|- zhCUgn(9`3Vwk|ZUwKwZi@s8Q(grM`eVBN5V^kL30Lw4hK1 z-%h5g+emX1{V@4z7yfGKZIrVx$Sl^yRcSXk)@h-mQ$g}`%|pqBYyBuEdjlVK0sMhm zBlfReNnJ~8n5oqf7G={)ZK6|p-?6*!W14f<&_4Wj+6{b5zbKjTIDs7*dWRAfWB9qT zDLnu8Gk#P&OLt7#g?jfalg6t zIpajNs{IKI+?qlYuBp(|S3B8$qZjxp97J}KFz&wWC7W%cL#-NVwAx*K{^Q$FeRVir zxA-45U%O6A9ga}jzZl7^rd^WzXLm|0hT2J{Jl)9rZ`V`d%E9c(#pjYBeSLbOmd0E3 zVnuf#gb%zB3Aw$MFy3#%pQR+TOUE`#W}Mn4@#>YvbHbx|X_5z*RX+?Rx~GnOd>>K6`q4n| zEZj785_1pnf2l7aw=@Gb=1U$69Tq`hxhaqnjwZJ`9FM*ob69$=C;L2d8XMkC7HynE zWx_USoNjZ7zG}k$-nd5h6^=*(wa-cVJf6*%U5M!DzLrGTE|M(l=TF0zJ*4BE!{Jf8 zg&F+)Bxw-ua(mJ}>Da+IlD>IBt$lR)E9;L~d^{e7V@IR-Wiz>V6j6(c9y1Hg7QV4S znlx5XX|h*YqD=2DGXMKKX{?;QjEs_!%>VOaAOjhnwF{Ok4O}#6p^v}MAgM&c?4%AP zv6qT}M-Fq4@&5n+*&k_FJYewQvxZjgc`5s49eo@O9mjR_kZ#(fP&!@ZAE(%L(j|sU zr7~{nWimHw%YqJFDap3(R(kDm_p%AG%G@ecxm5E>7asUCrR;r>jAOm~J$}f4S((9s z4LDcRkN@jAyVU;eHtDNp*~MyC%8Fx+V@iyMzb;KYw}~&4@5HamUec3ER;2;4ucSJU zlZ&6+zgiM)IHvSZTYTAkorxvya!aK$+XhG{F6moZ^U1HYZ`|(ESL@|V{Z3yfd*!sP zENS!s3(^4#V`ge2+u%aYX<`^vI&ua^Gt&?@V<$c($H zFOkj|JFC=0!L;nXu};}dr(x3N7_Sy%zz1r{jfrgLRMzuibGjT*;5`AIobV5(g)TD|Bym6pc1}Zy$ z3yinQkhZUz8L9XN6JpbZU#2b+o)!K;v4!d|`nx^ev}F_?_&XDh-W{R68$@sO=iP_w z$;PZ}eh)t8q5>N_7gE)zWct#yp8of44!XMf7}qJYiw(zSfFOPXR*&5dUY(=Zy>2`H zgLSR2Ec_m_KDvajo72vBZ6A-F=WFA^!|%W$R2wcvcS5Inm(>NEAH2@Hgr*- z*&u{{s3Xk2-vqwTa)J?w-AI!6g6_UJA7)&625a3OLhQr)xNmzLEqyf)?{&HdUr#-v zhRa9jskM=?Ea@v8HQNnmSGtL2_`T@XWHlUbrVSzc)P!uW>0S|_p+xR{jnmv z$kV;_^!#{8p2P#e$SwBXU>E|f>Ef(`D=1#x1}9&?%O*YZs_%W7!o4?I0*;sUA-}^D z?#;NvpNYzWoREeE#cTM-CF{WF;avK5g)PcR(}#%4-EeK=R@mC1OhJ5(XmC6Xs{&-0 zL3SqW41Np-(qh7)`JX^~=|wJtuZZeur%_#JJ8YI7f?sBeBe2=P7B%#9C%0MAuv95> z;o@l6V536~MW#F`%dtVhZmL%>jQ=QoL|>|tuuAU$)?K~Odg!Vtxmv!T9r#s(9Vh%m ztJDGQc|U}1S}ev_&(0&c&qw(BuXf^RS7zX*$4wT7M^1s6;3tGV%px^vo}~M68aaFW z9;atH5zA;LkzeuoNTqusHVT-EV>e{u0>cv!R{9&=OZ$Kt-G^cNih1OYbpu)$XiB_v z)Y;97-MGN>DBYXU%1>M7!I}G=gXvbXOzpfi$y(q@2DT~^uZb$S>Yo8V@i+&okDf}6 zQno>5@df7e4EKnp5Z#;^_(2B*NPaKN=IyxBh$# zgxAm0G!8+6ogwZE^ylnbUh@MC$Ahs|FItfk16u<1ApPfOZX{$p)^8}m`ToEy7@N)K z^_N2U&lf!V`=+4ZXG!D*060_e@WKGd8bpDJIzS0;=oEq!s zBU3NpeUd}|UlqBM#Od^>RwXxQ@@6<^rAao&91&%gt;6Zn#`xce4%xYBDrnnwQN{9R zs;Xa4!_^pmw&oKA^nOAoEo*RqLm$>jS-`v^D5T8%NULUPgSw_KxGE0O38@ySWTX{e zG>sz>mkW^Ft30?rcntExYAG4ZT((}pcWb9wn+wFP9lJM@CkL+b zuSm|uVdJeyUV|2yx_XHGv{XV_n=7c()ME6myNBoVB$7C_g~6QQ1af`Q(t1*yy!DI? zTGlC#<*ZYG$XU0^Dp-@^Dq+~Bt&L9-??!QojX|-MVRczTgWc{ z>J8V>PfOBXPF#ZX|Ii&NA}@ZpNZI@GoA|`wVoahzr#;BAxoRD_fr+{!Bb50$oxxR(8!}x9B13cTiirkrU0=xB} zLQ=!#^woeVz7=5t>B+eeC((mPZspPjhid%mhcYJp>g0XSXp((=77kCijf?EP$|jv8BmkA9hh z7c~q`q(p!m7VaLS`~uze|`wKq`xY%cISMbS@dtTTh9-Ec(~4nnFrLzil&Z$ThrBlYaX~uarN!Wr z#)nBAO2!E%nyKSJ75yTRfGwZGsB_g-@bs;P)hYzHPd-wwkYo#;NAkeJs}XOocVNy6 zg?ztzbMfx}cxvA6&#zFAWOHM0LjH0E<}kbrk~V&ZsQ>a{efc;RptudU+l^)83opa& zo7?E##+7v2y>hzT`41h7E9n;$&!#Tj1gee+tR;J_@R0TiVmQT^x{S=F<^P-+oZpMv zxBZ3px3Y24lH;70u@yTEQ((;9JE+&(gIzg)ib>0D=E-<&W52vC*xIjka5R1d+Nlvc zH})wF8A$`X^TrUCxd}=#7qiyzc}&Mrm64Q-kP&he=66?s+nr#x@zo|!_WuawJ(;k) zED0w{E#^rd%IA+C~k=P}l9l z_?Pif>V01tT889E?Zyk}ef%%ZPQ4adlND&pv?F}sI?-HmZ5>Osjm10fzu|<2#c*h9 z1r&_Y1+mR$@Kh%o9+Y!%Ent|QSLp_>a4|@)ZNsa%&miHO20s^WgU~lWsQSzVb`v7- zAB_hfLn`3R-Pvey?Iw5@HxDc5?F7rAL+p0+M%Jxd#GgDR7$s0?Sa)#!DpiBH;o4B z<}?BPJ;uXb=arEAY8&0Gy9SR{*Tsc{1KhA+Hx|}tFy4?dlv(D}kVV#b?rBB1u6_}U zTbyWnZxghL=dkJDM19gG3(>RhOKEqN7JD{jDqFnZDJYL$OYxda$XC7xTVLPe`HnmR zC-Vs;ti%kyL}v1Bl^3I6?*?>evrzEgvmCz5mRPJ?;?2CDS+V|z1UjwJnoXZ3Njp?e zFp2dVY`jATHRf}4#W^MBcYh8(1>O8&;ThUA@D&2vk3(Zq8T6MK!95>e;BRn;*Jg5% z9+b}gc&G#4{PftL_@s!a2?9Q<=80q)_I%KTcjy1W(E%AXoHWbVXJ@ ztWC3E(bZSDoh_x1JT9CiS8RvtS61C_@)~FzN@e%1+Oi4Xwi3r#M#S`& zf^gewPv~A$0-Zkl!QH_L#m|m`#28t2bZ<6a{KP!!a!`y#PL2ZO^|h?0?E=1kQ-;T0 zJ%H_gi|Cszr}z%HFL8_1uOeRlcKmFq4ST*~4OpDK21kBeg9pJ9(C)H{XR=5g&x(AG zB|UQC?R;tSVuvKjm3ao^_{OX@PLi!z8HL^{yOBbzc;cugi;tE+#m*00ar>znnDy)( zJn!7i2Gc@V=x+}u*X_)7f({DKpSq8&_uZj#-vVH)UJ9Q1eJ7{7D3{1pO=4}@7g$H) zQtUV2OUG_XghlJxAgKKjezA8pmOShWW|EI_d;Lsi=5LSedoqdR%@9b;;_x5;D=^LD z5;SeS$sb5B<*!(KA4kcJAy*Xd;B`x9V&6W1WEDjg`0G4A66XbAT^$OGyI}@qm7j))kL^hjhv)(I5*mQY+9;2W zwr$};ZRByyd3#(gag?q55l4&z2l1GXr-)WJBDzIO`3<7^DWt6!?U}j~$A`Y6r+kLt zv#k)ya*WZ7QD@PVC(;n>JdRY&>c`H@a>$Y?DWnau*kJQ$qHwSs|5$2-bG}^WX2Joo zvKXTmhdc3|hBZX8|1XR;>7e(2s*t|t9?Wge4E}B-Z4!`cPfDu4fV9U8ehse;F71EJ z*}qUF4R=4_2~j5@qzE&UjeqFA@+8>oTE(Qc%Lw-kpM|X^KjF)&Y*Ck88Z=AhQ1!I8 z{I!D`=zDDe=Dt0l8p=59(p@R7bis){lSpyev(fxeR|@z7*Fu=#qUhS3%YM z3bkuc;64~$KmiYa(oMlexIWdCxfx~R+0SK|R8AJLRzHAd=qx6hJ-ZNc*^a_uK0xZJ zX`mOZ1g<-df~RLJl9NqFAKz2_H9496Uh4pJ4T{<0+57S7t$py-8z3`UUSKTmh2|$! z@n`xw!n++wY}X79>Q83TQltA;)k`(;sN=Hi^66ZbT=J2ft~-yjJ%j5%v(l6IKj~_STTS=**f5(DaHr+dKvRNHwxixdEpjPl69OumE_=K7oYrYK!RX zzAeGl@4N$pI}_>TJ}sQ0F@a3iPC z>%N zyCd)7pf3p^CZ@th-DAxB`V(wuL_b`^ZK*T~duN8b7hScAQJ6^zQu_o0?vsW6VyZ%BGfJqF zv{A5NK>&TeTnB$%odfS4+(OTc-U-;mD17bLJSrvXEqG>ch1aIv6UbTQ@g5!1U{xnS z)7bHkcxBqtq4xJbsKFDl#aSEpxWtRE`kyl7{fH+y<>$!d|1`*P(Yk(J=Y-!q5@3}m zbtvlm%if+znp|Bzm|m$}Q{X_8jscPd8m2M~lk~%{aW6?Co)+Zq`MDNvV^d zb+I{;O_Jd{tl#5xFRmlQRetb$#E7-Uo3cj39hCl16uuUev-J7bU?}MyyLzJt<}R=R zGo^Ys^){XLWbR@+=X?=-dGQBksSUHfUApY_?~9o4P=QCk-N{}|48-EAH^X4D;wR8n+0AZ$D}k}UVwhspK~^mYhLzG$P%~@< z=Tb_Ux|o#E`H>&D@^wusF zHlDxC^rS`$j~;K}mVxNroUg;OOWvcSB`?r??QW|P`CgcjbA;XM{02^HS@dR^Jk?)% z64$jhu)CrEvHUkO!XFa?VU^u7)?@9?6sH!jn;#~!+D{@)(=s03Wld+N&)ML`_%2j? z#^V2ePve9}`mBWe4u+q$CQq7B9dct|-O2e+NTQ){N8VNq_`k;}P!-s>B* ztJ05&?LW+({+$D#CcP4rYMrEewoBA|y+}dfHw@5(k!om?>=gXk)rqTD8pB43ab(Y9 zH~eeiOwjq&0AV_I)W1@6#}lTaxZCbbC@u*QdW4ma?ZmIWkHZCZai)53Hf}zESWJpF zym$ITL-)@_=MGr2Z?47cf>A9~I%EqzG2bC~>N0qjycz1AYhshdSu5_OKxPC}E8 z9pmmlGiOWN4B67qNOrEwh&3IY51o3NY`3c(>(ljt;O=bdy0r;j%h`j=+E%(?&u-@M zcL0i(8!@|)E*7M45PaT-GJjida{5y>4K`>53&q#C)1;67nBR;Hmi(f1jo!?rw4SF^ zvI*Ba)i9Nlwv6|cas?7sxtCR!;Q5O(@JmuBN*}*tXdmFOrhe3D)oieR`~%L$_Fe<4=iN1>JG-RA+ZyJl>txbPj4^Oap?xH@5)ZtyJZ`^7&c%=Jq}C?(&71$CI)N(|X z_0`U&@n0s;OU}JWbN4%X{aOvx{rQ92BO*LD>}wFEm^HAM%l=Vs(>h4`UB?dU`LZwJ ztJy-INamkBo~`u$334qzz~7>g{V6GAkFrbH{_Ys4Fx?F6cEvLL;#sh%DUh~Z#>wS zUvFS^L^&u#Mf3i-&4Go-rojp64XDiLDSx(E0!x}~E`0LFODK123EMrs4u-Y_Fxk23 z;4M1KoLpj<<&Atn<;5nnF7r2DYITGiOS}(adFE{PbW2z@b00V){(`{Z25#vEX*RVf zhVAJ+O+Qx8WD$<5*qjrR^!KhHxapIRFKmd!uL2~=leP_1a_%DJRnXvkQ>*=Sb*=$3sB;0z~?xuC6ky(sTPA%ENP1~%2zk?FXFv9CF2 zpi0~bru-_Ra%Vo%FGdMecKSiMZQ4Qan!JO<&6VuW)Md1+!3iF{{0ChY5p2k=1-86= zK&8vgVfx+eSSsu(KUe;PfQ_?(f3{EgA%1ae<&`^3;>!t0n_q-m9+xQm4HOmb4T;mzk zc7CT7D<^@|lMCo?Qw5FH$IS05h4oukLrqfxtGLn0vfGkbT+TC;-IYK;W@fT2N3ZkG zxcgJ7Rml+jL|^!1Tparl;KDfL67wJ!s?xqX>|7w zwA$k?{nCC6eSM#Y7nwV->wBN#yT%JiaQ8yICQ%KiY_um2WztF9nH$vAAQar~YWa^s z*5dGq2S`U_J(PN;K*`7j{5LBO>91XjrB+=B^e}^|blSq+zZ4gUZJ{gwMscN^Ie3*N z$)2?@W*P^i7&ywX1(vJnf70>T>xLeS-!%tkPLp6q5;fSa6NNaa#Ee{v59ez4G;%gu z&qG>Y6IEQ&4|M%8rsH74mgzKsbm|bSSr<$~1tGBe$tvDo{w}h!`60(iTm_f(LSnjF zn@*}3MNg7%NNv(I^tU7gy??3z(njTQMpum#j+4ZTW3S^wPm{=lJ+XMDUk*)%YuISx z6%c-qBk_jc;bgcoi$2;R81&X4U(&PTpyOs@HtsaZaXpPoPVD2Up8rBM^slq2Z?5AX zZL&DdT@5caG~xfyi^X4CkHMju4bU`xGA_S=9Aw;zk;AJ9Qruum(sg_I9TitZ^K=Vn z$t@#e1Es9#GXbngPbTXRBXZJ5NWMEv#tmxiB3?6&6MXjMWhmZJi<>qZ;u{9=Ko)uQa}h-`9*G z|G8YoA9P#D^(VR{$)q1$zEg!YFWZp_{}IX@Wl4jM4k?z%LS_LcNy7IENce96OQlXC z9$PsijMs%l{+dw8vT@<*SnR%P6Imd$2cLigSl2uq`t@9y)8c2O-L?@|o?AweX4vv? z+%0JT$ zb~_c=r``+DtN!n?#{cm^B(;J^#g+8}nb!YEV^_61O$MKsW@as>gd6W*x z)?UQNOcshny6wuwpVZRq0so80gMMx}k4^kkMs|L;B*V2Q(7CO-Wb8yMYrgs z_B+icmRnDNhn)!n4fvQLNtylEqiT;zQ+Zl3Nk8D|`t~3erqW?ILK;iUXG#1;{;o9lXRb;}7(+ ztaba@1F(ZgyxK=cs zE$FUeY`qxy&qx-GCTv7MBYKgd^l89}*SNoau2kvNQ2}gg2fG!wVT?xrY;uZ&NeK~X zco9>lBJ@1)78oG>bJ*^f9u@!au!C`8n`>81Xda;>HC%FF?Jq5xg!(gvST-yrWw zCW2})lq{qmrdtVv+KWJQtr+|LTnGlQH3ZsT6PTYLV`JY=L`rspNK5MmKgci&ZA}nI za)Z&#V2wZ6hKhJyC(%BsHQ~v=7Uy~+&0tnt9K;vjgbMxR;Ii5k1wc9OIe&y*UZM*1 z|J?&?xhQOOPNc<7l&8fj447e%h^5at57VNLz!r;RY_yCg8(OgdPnspi>IZtEccCWu zYCG_(|MjD`1#;w+NhSo(3WhzK#}mh%KO!D6i>?}@!HieG1cfyxK{A5FUq8OVxef?F ztvdw^&YeW%l6hD&|01*<58^joT*zKj>Vf^GP%76^MK^j{GLuab?4NTX_E%X9Lt4fx z$#Mq!xMwsdR>s3zA_*2>A495iI1T;s0nd_9;SQQ?ru}{&>7=ohkn3}j+p=Re{v5CX zNi1)H4UWh8DKlSz+wuawLZ%ZW$(um$s7as^u$=i0&cman^yu;0CaSPf^gU#y*+j>O z@OpSPHC*)rCK$^-D@tLg&Si9EhT zy6ym{)gcbozdi@8NmZ(i#VcgcJr|b-rXlMx zf{WF{AhZ1=Sb0Q2caH%I3^9O^t_0lkYb*}CJr6Gbcut=WT0`>V88|F!5#4&{EPQ#H z1y2jE;t;o9Dx!37NJk83JFG*M>yxNQWH|8%TTWu5Mlo-#4{%r46y_Ion>=|a%S74&DhzH zc22oq0^G>hjekEHk5W1;;hxiK+GV2v(Fd==Tb<*$`kW?=t~?1{$wIiuedMp7_7|FS zqCxJ0FSwMqpuD}aiI|oGU+cUU?jD@XvZ8N-{fH6&xr#C%GgJH~RE1dR)v;8&RyxW+ z3`#{^_AX}yxO?_#BDN-)gzKfCww`FZFGGut^SXzYWF?|qSHqdVPc~JLafSOU_MlrI zO; zR#65jwjF0aE_U?KxQCWSH4+abbpf5&U~w1l#=DG5YsZ5Q;gP zn_hx_x2T|ZJRNx1x0S7WugiAMPlA1#@~nRUF>JT!C4ObDM7E7CrnOankaF1&I_J0= z``NEy4M{VZP3%El^vNt#WW&Q&W2LxbG9Nj6%Z)5bc`|dFti!ep=&+(6sbId=7wUGU zuooBJ@{$9m(VNi$+?0DhsM|fCJ=N~U7oYWkvYICQl^PD~|Nh{I{MtgtD;}b*mj-yJ z#;s=y3%%fsUZ0@u-FCS6uaLdI_lEoGI2W3$Ia(Vs0(4kh< z4{TCtKVb%EVDy{9Y~aA?dzi1txJbk-CG{qFU{q2 zgYF~!J01L?TS|P-63j&ZUbe@1FMb;5j8*>SajKI=cb}IA1aF#8?*Ed&z3z;71TTXX z%Cnf?%&9CP{V_{Te!%z-^o7GUUztVH84+9AfQ= zKDNCGBZV`#sr@-^u`8!YZ8Ek!(}Ih&>yb>zIVyjA1iC^aI3==&dorAhZ#4&^#h)zj znu&+F(SZ)gd3pv`QS`@ZQ+{wezg>XCt~zY9mOjcE-HQ3v-5_aN$-nU|3q?rA!(WSe zSng#$GWXTOw`B_Oyzt$0kwg@#>D0oL-K1f!wm#W@uO5&1wqdW}S(qo9PfAUf!v<$f zc-+1OZ%e}rO4ho9-mheO;1J z6Ceb;fAUm$(;O@s0ibZ7DOmSLGsTIO?0mT!T3B%bY3x-%DW{Y9yQ_Su(Kk(6y5m9w)M`?g-VsdVd=ki8-RE%qejy}fUdQu0 z7Qub_VEX-a9h`nMg}HQ3K<=oR>^u63emdd6ZeIvswkj1^+R>OLt}jH19~NSZ8Dive za1xAI7IS^B_R#-&6dShhg#3XySZmFGvMXUSyW%h($2woe(Q||F8rO8B+8IFKojJz7 zn|YFVCSLTQlo(zy!JIrhet{%-T%*myefVim2`&v)@xbrr$TGGL+YgDs z^G$c?)`!-FJK~S!!ex<^KpfJ#^ii;H60E8=1+Ah2rcrwpg)Z5N_kXzzKVIw7eKGI3 zr+)LPWVANlyw()29p5Xs>1z%)e=K0Odjc5WkHqU`-{EEJ^4MXy5N_|*^<>u_bG+H+ z7q0(2f)B4b26ydtV~zMc4!cHNlqhDy5YEMz6JOP+{Q0% z97pc+Lh#`)8YJ1X5*_}Vg#NPqWciM0Vx@JK&lc||ZZEd-F}ir-l*JOR;?0 zP%pR<`yO{`tHPs-jWFPvBBB8H!@1}qWP__W$(ttPb(|;RNp*M7(!?XU%ua{pI$PrM z)iYV|-RCr_^a}Vy+A;gd`JD0S@%ZQ0xkTKvhE$6eDPh$EPvO49aad~dVkYsu2>s%1gY#Ka==@2N{A8JSwok#A z9<*A{`ADrJe)&6a^S%#w-J@x6`k4n)s9p!hwoG87_syeg6AS1-PzJZ%=p3l%N1!pM zn%RL@*P**hntjljD!ldEj6LnFWe-Ms5@(qUY^w7(CL#Hga$e2&-K86-U~?RE=-vwv zO4-!p%?h@kXTvt9w18%CKb^ZG4Il3d=UL`T3RkyIX0_cnSjO8e%!*~h&2=7dUSmDk zGAWCRb!0%?$7INt&W7}TGoab+J{W4~L#WJCQK#`7l(zi}c(l*N*6O2#e{}vb*=^#& zW!Vo&e7P+X`|8L7Hq2sg;OOELIp93$3-X?2*Y5Xg?u|Za>;7 zC@qt&-w>+C&YIl7)!X+0n^46@h3YY_vitnLji>2@?RT)^-ZSva?FXob#ZmvGmvA%c zptF6a2v?}f2uqz734i;W2rEp=nL>^PQ}(xJKmTrFlYc0|ug+DpFfoR%Ees+*s`j$! zav5N9(t>F^S@KTm?PpJ|LfC_@V&)M(kH#JwfalSRnY?H%m%f@P)V=?lntAlWJH`Y@XhRzF9|D%i%a! z#fxUEnhaRJhXvdgv1IoLc`Sd&JjM&tWbb zOv;7Qg7~FS+MvgZ?y8`(d(Oc>YcXD-L?eGi)kKtDno0fEg>VUtPpMQ#FEmC7`D^F= zWZ`?xvcc8s*|T+1Sci^+P-R05Ys(eoD3@%(J&HHr>amq9`}|)hQ#l2xVHaujf^qQr z-g>THxfyH!38wuLop?`>EYrMq1xuMZGM}^4gsPELsPZ6DX#9Q|TeV=Ar4~l9&$G2@ zX;>}%c{sqgr4)zgy@RmX!yK=A55A{;korW3GcT16!JZ%Ckl*)~PEKn^u4_cy2qc?+ z$aTegkLU_VoTmvVtPCMdvWj?cxvQY0qn8(J*bjG_Rbk4WB{7v}<{=ccM$fpQ&QDQGVXEVFzqp;%n8^MAxe70fl zYglV81&VI?ho!eUo(mKr)SSn zJf7wb0Yjp!#-0t3b~}g){jPFfF5P7N&D7b8ZWAVL`-<7l_y?E1<-+ipT0zbG%gmKm z!`#ocGSlqM5VYQt;TCrm(G|^f2TIxC-koq^h5)%;I*RKB8W75BMGjvy@g?uqcymA^ zwklL7yL}vi&ZL-6pn+lU&vyw`buSKI)uEYGmKIr`5CF1ac_`*BYbRoKm z56;}fET&YzPMdw8z>cyW)$uGVZv&F|mBlNz7-T*^Wn!(dveWL~qpB+E~mIpHX;V2QCtAR=`rLm`nE!eLqBj}3bQ;-hRWQj%} z*`))4(Bd@?r<%OQ4>V$#hlqJr?AOG$uWR`JiL21@OA6fF-cNk{03&|Q&AFr`NSi1e zP(XU3tcvds{@}mFAJ5#cfy*v8W3@-Va9=tdAC*g`UCXS|lRA0)%~^~5)-FL?iqcT* z_|tUOjh`?tUy9xL*v*t`rU>)$B8BM5eVFG{$%^Ad%!uSn#*1u&zq60xoeJ%6c=S@% z*saJ?T1GR^Sz^MVRrlG2iZcB0UKKuEUX6#2e?WWhjc4Is7qEhvHMm#)H%RAeKxSSv zeqmLQwL4bB#BdKhW=XKXbkhjG&te#>T0bGpZQ^8X>l^xPelaybTd;kyB2KRB#c##4 zNYL&?kSOr+r!z1&4|qr^yY*d^97ERQ!)CCJ^R1XFB&*C#3} z;E|RhQa81LKlyDXO*>Kn?}GGk)=e39OY|lyQ^eP%JdGpfKL+sB%S!wY78>Y>Xl+hR zZ$(T0jEC{Z2fTR^{Z#xZmHXRfPZi9sBIdpGZ7`AtB z=1aDpLr;d4(7lVLqI<#=y29NZ<0>@sQ<-S?tNAzq&r zK@a)d1<%(fprthmn@c=^vetA?xjPR19y~%Um4;CJav9W9dj^um3-MCe2v_@8VQ0H# z;NJU_6W^VPW#8{03csXSRKzzNp*ny!?l{8@u{ORz+cs$CRr8ClW|G-)#-ulMBic}H zj{>ijpbgJ6u#D_JB5!;fZ@f_l*3X-%$8vpqJf(zb{`6*}qbX>dTEtKBI*KmXJCOLL z%~azzAAE*J5oLRkUQ~OC6zHtLY2J%b(^NZrEbbyYdQ}#zM~$Jq4WpQh_8ROp>^+qmBC4OsKOCVqHRg4qW=r}jB=EJaqt<*1!!$pi@cnhEPZ}NfR z;xu+|r5GfotmKDHZ5HJ{6mU}Fe?eea29u7i#I=$4nSI$qI5FG?SAO2aoc1`v*U2IO z4p}4d85?nQ>q*f2S4iiYjbL4`V)CRw9H;&LfGbsXX zKF*q?{}$&~s%YUgx7UiiWJR{+HX$Xi|AVW#yZ9zkV{l{DeDc?MDQkTm%jz1ovXBM! zcut4}`Sv`Dq^zprAL)9=kM_C4zu32jE4`m9usb^$PM4jfYHTd)6J=$5{nSAZy)`0L zx}VtN)HZmvDxb<+j2C(4J8-ybCx2m79#WBv!PBDM$xm-Y#^-)S*T9-rJ#YZ+KC>5J z^+6;%ZYt*e?BnVGksxC|Z(yI3mh5s_GIQ%PWO!#c$Oc7`c?x<=J^d25#T=9IJbk>m z?i~E5XHKL8Em=Y94lt!Ukk|5p-`gh-)kzEC@W@J*_qZBl=BIGV#-VJjf)r_%AH)|e z!-(0bXCT%)4UF9;!sAPNf~M#+NK*WWs&}}MkG9Ehe7A@_;SXXHg%(n}nopMgazp=u zK9I4aK9G&O`{1>+9UarZi`WP>Vb;5HmVN02U9*>?KbBuYSq`n-mG3>+MdXk9%(Mg7 z!b-03`4}vFr5RdZ96{>u&G3MH56}NcH|+Z(&K{)dfYc~$vVUO`<;AYR`c>szvasY#4^oVr1!k_!3x{pr+ALV$#! zMxr(6g@+#46SMy+$*pM#MBB#(ca?S0qFD+^VK^Qn7QCV#)pBX~SwP>VEv1gm3w;sw6B)C_;Y;W#MX9JM!1hu`sWH1&2BHWI|( z=^M{sYyBa*_`NL4eESb=d-jjEzBS}$*j}eHA}{T?LIDacd`vgIIxNUbS_Wai+vwN9 zF-Sp_3H@*64)FRb=)PN1*ru!+lCYo@#~CS;jmyX4-YNe0VC_suPK?LzKhMQIC&T#S zhg!MYUtUwlErB7@NzEE3*%`p(hIG`A%WtBiGTSH>+{xV_y~_w~p#%%Fz$=g2*c$jGf*H^?Mwy!n{O3lXWj(V@+*M|>SW2={qJa7xjH7-3SM$JnG^??pjpzb_>`z8r$){~sQO);odfN?o7cvEYz~aO3isl z-?xr#y2GKhpL}RAKOd*OIsv858Tj&}^SJ$r6Vc8*1lwmmVatAogRZzgy+4tM4@R0& zON}6uR8vdo1|#J3%n*9~PS;mUEuqd|rs8$44$%+wjC!FgT4$|?M=E7WK*9$sJ|`Q0 zelUWTzFmkdqodf@bGDI6HSSK7M5uj86KDTt-F1 zzvxQ5ndW2fjzMhp#GEMwH?i#Cu|%f51KJkP!8$WqA@;H;m%=}UoN2eh(-togHFh&l!g>(kfcc=5%sy}eNwVVh$J#X%7{YQ zp{1dssUoscDhlay&-)}JyMZVq>zi3dh>V8c{rwA{$NAiQ?)&|EJ)iL4#AEoam4YvJ z%*OZ6t_kLi&A9JU4Z9sMm010jlbk;5*pa6|XW7mHMN0#kSz*Z62mVLCBtDh4SrYCn<_+2*#u+L zU0LpfzF=T^2LGFufqhQbV$r8!UU@N%?dr21ySryH{PsxLQ8^oWSk{2Y_zS|ia24|M z$OgGLKgZx0g@&tx;iKMYDch^a(;k1pDFbJVv+foNFAq1d^-mkbmruuH{htNo%*w@N zN$e_iL3G3T@=?NecXKT1kxfEws6oJw`8dmO5qET$f(sg2u;z~Rjrug7B-K?x-)j6L6O{%Cf?l?Al?!zVx)fSmaIGfmPMiwi$ur;4{5xrApqWPrLWcy!r zc=Iuy{dszX1+V!QnN4LOT{ef0^!(RiSriz zLiB&;kBhu2NaF+*L1Vc)R2BV&9|!8m6svgB+NsQs98+N2&_`T6I0Yl_7KlGPie(1J z({Y0O9@MssmhWl{!yCqq^4Y(hkXu#Xg-_nA3I3i!UWI7U*>|p!eYclG!Kh#G;+YPF z*vG-^baOiTc!AuN__CaeCs6NIPfQ9tSb1q1Y2VmS7*w^1q;(_H*Gq+{}^$i?RB_wM2k+|>4qlNt>9?hO8ne%?e>0&W*TemiC$h^Si`#I z>{9Du;%TG-d+xhJWbH&|uarf%ZrV>bm2RcG-Jc3IN#Dq=od!Z#&PaS$PVl^JBp-Qh zH^ZAd(PMc!Q=UeZZ=^;3;EOqHB^;pMd$zQKm zDpW2AkY-7te3;cdwE3JbYwH+EHk@(AtykOG$tO!N^BR+nSf43ubJDW<{=Cq& zpiR`a*(%=@a-Xl#PvFX;6<<7hHhG+=0f#EP(@}PJh3Nmvg^;FK=>2GiV0p8HY${y@ z8t06OcheH$T--?xcFq^aeLBK^jGHNwSE=yiC9`N04MXU*|pmjP}idj%$^=(hdf5nQ8(AnDFe+he4(_He0rj|+m2Ht?S?u|+4Yw7 zKfsHF_1_82cZNzFj&ZOp^a@6Kuj9%NgVAGiv&s9EtE@Ib znfjAEvN~U9Ztfe%1U44Ys}uu{s_==dKRACG53V1aAT=6Eh;bc>c`_NxH|${cQ3D_{ zP@9Elci63Y5X{<>^u@;5b$IIU7Hqp1j7m%y-6W>Zs@t`2>sAAc`0fSM&MRWx5_POl zw<7H`&M$* zajl@_dr|x{qgc##%13L3Yn-|v)gvgi+Fr@H_%vr;j)qLqH z8(y>1Ze3eCRodH?&hSgcrpk%j+cq297jA|U@4Y;BWI5{Yy~rAkG)ez}6D-?F0g?~8 zg6u;HDn-BJ8sQJ%1u^EUEoafEH=W6sO9d?E;SiiYx#e*IqZhY?Ho@a=uL^9(mW&3PB^jf(Vu;0r%)&7fNzBv563 z3ARWX$O|8)NEuhc)AxOZ>iSDCv6liRZL4rqWHF98o`E{8hIEWiFv$2xyQQx_kibu> zq+8wxqW#zrzNJ*+fRm|wL_a+US>J#;%1iM4`0nVY7R$5M{Fw8YDKI2;0qngP$`%X{ zfqCyP5c^7ROx>)F#^zrIO0PrPM7P0>#9f`)!PcS2OfZeRc$>K@p3%Dk3j_`h_ z4jJ-Lom0;3EkF!VYNp!p1OL8X$;=MUE?zN)<2KYvP=Oihu6{QKqL8rO#^7|{sVl*t6ng~A(j^h zEkU`$Q*^94k7aWo;!3Nb!nfWDSZdXSFONLV=1Tp5mCs`N98Y^%8Xbzg4M*@N2I^Rn zVu1^!aftO)JFx4%7d8&xz@$)wXf;KeB{Uck=l&zfq+S}(xbvgrMDvA#Jb>JswG@-@ zE77v0rLdzuiHv!Y24-#(VaPuvxuk(-?n>v#k%H;Wxu-F1*8NX3VEHgPe>hkfDdI1W zQ)HguA;@V90$&q}8PP{B&Og#7B&?^{cVUY-rO*cU>8@ZGh8$!khJ0j|bzQN+>a#H7 z>?*ctDNX9q%hQxdWC%^)*7 zwaK3^-xx&K;e^^HIDdf^hL7n@AMJO7vM$r1<>E4?edZ<^-upbu7#IV78?V7sg%n$P zLJUhg?nV>?6yV)lOYm&%iyd+0WZS&OP?{JC$|IjLjpp<8VM7$H^!r5{5|VLoi~%Ga zj%791oaq%;6XvmW3%zS2KulDSU_88;G4E0c4BH6$hefzDn-g>0EI1RM0q67!=>Fi5 zFiiEI+~ML7D*y8o?tJ+Til*7nA!~mWAOsqU7hk5hE!Psv&?03ks zz5b+tIiJZU>$N{HH|6T(RJJO(a^-7{hD443xCVBc-*k4B}3*j~j0!&MwAkht0 zyzGxhQbf{a+ZT3w%s6K1pMN3AT9GbSN+*W%HK^_#Ej)Hp1*>P9iZhM&h`P?_$ig>` zq;Xyg<`(MV#VJGZ%AGv;oWBPY2mzCdGS)@O!ZqRqEIaU-ebd@Txfu znJ^sUe~qI1&1*?w{cuR?KAAijyTtC_<*(#xXAsSk&!N39r-IU$R%r720{hN%$87UJ z`tTpZ@cS+7tlDjep)&evRT=CKb$~IiDq)1yaWd_-EpZ9ZMw{B*xT3wD9Pj6kC+5}= zQM$`u!D7I$?PN|%2hp?|0}eV(M5}Elb}4&;IcM_O`+AYj)QaYmyLJK39pzLGMG2^AWR>+>T3^ zh648Sfd)S(P~QF@J#)Mqx;FHnv$K_8eNh4X*q+GJ`#6z)Ex$;^^U3T!X5-mc8Zc_; z4^sTsj1882ldg}#B!|Nu-qyigJ$deAi*va=bV0G)|`;`D-CV z?+U5Po&e`P^2DFtjzhfddgw5lj^7oofMG>Dd@?l$e`bmrif-b?gS~m|l3rkY#u!?) z4B4E`UGV6x->_uuTskvm4b>})!+DB*xFw6`835FKlvi-HvF*Y^19tF z+_zk|bkPaCz5D|z#FW9jXvy36rV5(A8ZKY_^J8KHn`Y>oprz1k{i>Z`Kl{#fIQ&s>v2iP6wOS`NHiL@8lZu z@^FRY7vWpe4Ox9chw#$nJ*?f5D<9Rm0edd}fbUst zv7<)g&qqTs@5?KacE4Td=M+oBZf&7mf9ONzK4o4}X^8bX-{D0`SL);SP&gBC8CUNJ zflRl9WK8rFIE>Pa`dSy_HuyfwDf5Td+YRV4EWw6rIyB@OvPZK{^VVm3VWQClJoF%# z@0fNJEVk&7kNYRHQ%N~Q>C;kDKdn}7eSMH!a@#8S;b1^I2Bm?2yDBkQ7bjM!nZmv| z9pa+sULe~wfIjH-hHsy5LA*^r8b4H|NZk%YH2lBPU_$tAMZgO;l;xWg_#^uAsk!qKzhTSDx^)`>4RxG5OnziWR z+{;X9Yq3mm;9AJ9*h^kohR{drPYNSaABhStJ(>IM>9md|!bqh8dOywC{_n9nG<@|` zYVgm3)_hi!_-!wEvXL|^>2r$~-r9@NJIyfFygPqUh|r*gq=(G~`Q3ir@MF~#svhKM zzx}*Fv|W5lm6t82lWJx3+V=6(dWj6s_W{_uD%yWGnQb5SNy~nB>M(ldnH4>EOP${B z)5RV>EVNh4E|g9Edz!9PxdVH;`|+gBk5GA-D@H$gB&bpw8op!*tvh2#y~Zl>&CYA+ z4(s34^88xxI+~9;tcv6Zbdy+RC!u21ceF@8g+Bv_(Ei^Yn6=g{cDVH+VKMu_)}xFbu0`De?!~nB*FE|iu_PxfBNEwGk;`^kH{ zhk6A(66?>Y@`zrw#9ZqV+hsxc>4(3CmXD#hWY!6c>^q;*RwKGcGm!OVb4h4Ptr*@f zO}x~8ST;Ri2Rk(53^z1tr8liy@J{MgGHv)LFdiEznuHCYAKWzQ#xOH_e0>`0S^@CN zxf7BbQqkh+f5Hdfy>Q~xOl@9qN*xnZS&G9t(XSN!@%o@V2 z(SQ%~tq@n3e`S*Hh-pPz@$dinLDAVKY^j|hI4T=s_JbYhdcPe?E(J)LuZirrdw09@>5=?)1rm3A;3K zq1Joqby$=7XIjyL0c}KBnA9g*zo?cFFf`SotIywf}3EjnhkJH3O7{D`$=g z0gXP;iqd|6S1@t*_M~1bZ0#)!4DF|ES_pG>+Gv=;KzpkuBUb={FN&D}JKks1lRqECk7pbY$RKZ|GnuTKZzp_5B)^^J9Meb zcW)9{gLwSARA43_cFziuZ`a}gw~MfM`!$f~+ClQYEF8S= z4_Zx$N2;SQ+FqBqW;6akLS+!{xH5(ff3OL9I);#W-jCV2gbLBONS(%cjildhePHt= z9pL16&ba}Vsthv;OAPM?w`1zwo1wH586lSq=re$xJ>7MJ`|q0tKnpLF0UR*U6gX38S) zom?lnt=$5Dnkj#N^ENMaeZdTuT%gJlXZ>lPHEhJ^uQ`z0V?k#Y)_8*+o zy9Kp<_4wgYnz*n&jeX21#l^;3@Mo$m6rIk2*fnxeWL^)>KaP-B87^#_|pK|!`zVPTDKA%?f!h* zfy;Q{<4ETER}-^ddSh@$32UEykM&Z|XXZz&n90eRWY(AeF#Y9Tc5-M3yZgx&50~g+ zbmuQDzci1(EpufjH}}M#Id_==Q8;x^Z}b^ciZ9JyF~_UVFfUx<6V_a3u1*`!D$A7{ zE;!A1en>*?j)PdO(G%wi$#DJnVjgW=!fHO3VoOybTD0nO>s>m$`gt@yUv>}wEKb1C z9f~}4Yy^I;+yq;{XYkjXuYpdaG2|+v4jvWDUK z&|_@HtC6_hx)*mDT8xjz_9pG8s-eiLe4gJGquFV;U<4fypFWF9(?XCe&AvTmnw>x3&9e!zu% zaX8P6_zqrVX~$d+{vhh^{|VVTk6Bam4yL`;2K%U*V7yiV{FPhK_j(@rE##Yi@VN zsq%8NZ(SZU)Ej|1PZx`k>)N5|s6Fj6?FJ-9>hYDop34sE923mun9FNwwUJ!gz$R1x z-DIUhX13fXJJ!ZSH*p)>`5Xan%U4j75m9u@?~~NpQ_A3Od57Ci`|~!VdE7N~0$y}m z!Dnx>!WC{rblY2o#uF9zOex=Px6+Q8ZnuQnu5%#9;2Mh>Qo!IN%BOthMG2eAM zmaA`5rd-#P(Y#}1P4@`)w5>=y$XBsbduwEa%lE;!LSq`0vyIzEPZ0+#O_0ZzJiG9| z!koIN*wcFDDQI1lK-^y_;D@4G_NMeK58td?61k*+6;f}0?Zy%DY5$w#pHVXObk+p# zGd9pJseS6VMvyMj*$MnUi;j<&P9yJXfm7#tzNx*9{OlRU9!uVzi$7P;Q~ssUBI{xA zI3SS@@mA%Z9OJ0Hrxp*NufVQ(9TO~dT|_0jtIWyJ5Pfz#KupFF)^%4DZBTDVlh`)w ze|0nTJC_U1#u_}(wE_yIT|jS(7sYCedXciPvxMQFjHF!Tbw1qpF7|p9CQ}`F5H`|R zWW|e6R_lEf4|<&e&4K1T+RG7NTdrU&_JO2U<_&Jc{2|3P60^Jy&|LD8Ml9?`uf=V{ zZI|m{`mifd_~;Vsa8;!xL*uaZ{a?s;eMf@TE|4KV2QZVTYq*u3FY2!JgNE$=Vp4;Q zOdR2iE7#oz|ASp%aKsrd>+)IZfaLQv&xg{8!5`RY*-m`+|9?tw2TMIOjy2EwOZuL$ zmRGk6vgkQ`G1xg88xM!rUG;n`v}dixrP~yRhS&A5{?QOv9@`T<*PC#6^%nNB4Ivzu~{lh2s5j9-`bZQ9@fu9f3c7aLc7k3@Iq*OB2g^uu#qI&fY|`-+yfD zKQ^)cI|tG*p9cB2SL&kg9)bIg+Qrv~d5E9PTxmEEE`Mbc$}r9mg~{2BLqn6 zh@-bA%%dhVno-g75VX~Fg9%d#Fz`qP+{)3%(;=EXqNX0ZI9$h)iY>D3SFVX=$~V}S z+YzX6wT;M2FABvw-Qf1o`S|eoDA_8v1N`v%Y#xd#eDn2GmL_GbpGTd??k5X~S<6f^ z%;FZ=_0fs`UNH^e+iBbHZ&h&l*0Wf3BZXwG?nCR}Y^SdAkEz2;iizLDiSnQfq29er zC|Rybhriw@r#Bz7fw`$*uuwW1DZ@PIJ!0vT+Iui@8Eqf{Iok*sthAP%Z7sAhdA-f`+K1D;tRa(whQt+yVy5;cumzV zt5B5>a;Tm37#5G-MkCYwz-XF1-4yFptn|u->aN&J-Be8J+QJTar)NS#WQVZx;b6Mg zV+0L}8^}Yp&4TaCGpLVKclxk-1eG<7qQCQX=(B!t^d|cZ3LhojG01aPnhKEdC={DOF(tJ9T&N-Khr@Dng|ISW0>HdXS zs`aC4wP|oCuuQPpy$s|Y-|$~rA$H{ShD9v`SF8&W!Xv6kL}VJRSbK&#>ztxl_p>EI z@FuFYVppl#cLey;2I)Q)7;&3LyRsp)7}q#O)coJHRfE` z{vL!}IV1WgScy+|Phd+|rwV80Zl#s66L@yE2YWgriJJWmqvOU+B-@H_5{vVDRorD> z0!{GBqOTLx$g)G4TtBg%tzXp>^V26&ox$q#;sPD|!abH|(p<0~BhcLCiLl-*52`Mw zK?chO!ysKc$L~6QR&h^UVe?8Dp+ASEFS1RJtBhjq7(1^Xvz0MpN8iQ{k2*lK3nReEVmC#_(o2tvN-lPMo7{Z*_p(plEs77On-qgVr~c&exG< zi%}!^bzs=dS)PTE7e2P^XJelBE}(F$DkLRDs4KYOp`Yi(FqFY&^Y`q)VONX#zTJem(L3r8?}>LU^Fl7Ev-pCLj*2M zBX6C0;lQv*{M9ade(kpO%$}1LO_YzYIRlH%y7Q^)4=@sO zi&Qxbqdo@)(LxO^`>cMeA>`p|P#-fHmQ)Tv+gLM_5!w}Hrr+@MJ}D=+Hj}OlZbok1 z2cqQda#-Tc>i16MnIAoPtCa%OD-GZ$^P{-yw;WRS2*4-cKKz;goez3l$cF?)3Dv#a zu((VELYBosK&o_CE31HU3R7sIk0lBEID)-yXcLI}Pjbj4i=Q8Lg)6JcxlhUi-ao~h zPcR=Py!04G-c3A;e;tmpVZ@63Y*%7~UM`g!4deM}waMi8&2}_4*e@|IRKQYxiHsXm zkI83~7%v*kQv1Z?_HF#Y`q_}SVMdd}P}y_+tHP51O+b$2T;ydJ|Zza7EdcMs-^HudA@hYjTG z>^6(19g@kQ;1Znp*@nLx^iFUYbzFEmB~>U5_OuJ`Z3Yjv&V%+(%h5vP8&-dHBy7ST z^ltP)_b6Q~P)rm$Ob+7tg(IO`*G2LwkHciEW*oWG@|?+TOv3Q-2{?sFcj)h~7=L6V z@p?6dj2!n6{XC0!)7-PdvZfK_{Cyy@fwQU9yV2jL=Hi~;`E37NRd}#X z6CQXE$F9e`ux5K67Dx9&B@1QxZl@9+o~MIHPK+myq^?HUK3k6I|3T7?b3nY867Q7W ztid;f^&L4vShwl{nXq#^XJgEucHTNhmTD&Uv~86Kl-i1g2?S)dG8p0zts)T z8UBG$+e8w%uULpx*X3rq4%qtakqFH$yr@*-8GP8tK3BcR*z{w}zwtj&^t_4H_s0rO z|LNn7KRtL{*G6txRl`TWTfoN@J|Jq@i-`B;`7nKeC5hM^NN%q@BN|Aa>6LnkaP-G3 z=9~G0S0C+G5}CA*`?`1IG171M*SCa#pS{V|&F3+9jRWd^*G8O{Y`0a)ex3Q% z1^4KMvtILO^5apN_^h&(n9qu4#<`<$b%W7`PTQMo+0h|*!o(P|4t&7opkCZ)tshst zK9%q2as;=!D)H4_C$W7qj+3(ZP`0)s8E;8ljwA7I_-VpH@^q2JqrKOYMrfsC%d~;? z;don^Vpxkm^#i%}>pD!IQN-trbm5Kr_ltM7z2HU4Q)C(8!FWi&Q}{SJ3C5e2VGp$& zVx6qdcW)Ij`we6JVI7v72*WwswCKOBKX7^D5q8_C4SrkQ5NDje3C~pP$kX`UbSE4W z=W1(-%UW7Q!!JHCZp0G4EvyIn6~w@Wk10Y)f-c=rnu90Z{*Vl%1okcdn~*-eCpVLX z@moIrg;5HF!5{>nIX?v&lc(~m9va;0(FF(|* zrCfU|?^-Ek-JVD}!GUAMF2n5j$NkI6v&D&I;f@q;mpP2vWvr)e14q)MA9bkh_leZ} z&NYzqv@kQ6u^C@yOL@C#s98UQc>g$wO)aOv(7cX~sab$2-&5Jrqb+2>#2&Omw8z)# zVW{fG@wMP9vBxPkZ#~5lEBw*DwN9wN(-#llPQiCNncU^IWr<-^0>2|=an$Df)9g}Z znAfrbCf=w;bEn%9BlZaU_(y}rHB1v!=Eh*JWkdOodDE~>w>SPwSk66*Z?P>03NUT+ zVc3X^NVgY@S-|dL{9B6-ew`T4Gd|Cy$t~efS2c#Oe6WsRs&?QHW(V-V&7=9ZJA?U? z3lf)D_dFi0NXL?2e=ysm54&Q$8}~+);%egvynOvJSU{fGy!Z^51u1juj1Roe!k#61 z(p+G&ek1c!?a!>%KEb7b!nx&XQ*M{uO>C&tp>zFSV8_Onn7zwSocPOx%hnH+_~`jC zHn9Y{xLjh(Z!}0g1cH5J{rJj(S9wr>>0U9gJD8V`Brj*F)9FjUvpb_w*_w-6h@004 z4Bb~JTrfL>)-LzO`*t3XJ;ecfjs7h}iQiCycH{aT5|;-oh@N*hKDTNxANF|<*H|2c zMNP%xp1((kR=ztI3LTiT6)jcze|rr#0%~xG`6Smx=o@5Cib-i5uvjmBbw`Iq`|*sqFXFMyCIv zKi@uX1y```%DpNNz>mr^Xt&soe@;Kb{@8`ej!d^AKTgKtAH_ylwM!9dY*T^P?`M$A z_r0ite-+&5HH7~<+m83Qhx3ME&za^WBQpMz4meDm%)H+#Vo~jC3|#-YILA_xYnBg? zviIq{ewq;-{W+C?Ts4CSsps>~)A`u(WIU>U>&kIXUmjDqgs3|AVLyvHFb+4<#N)wu zNvk^-@3=#v`V`r@|E6QXSZy{sTjGu--NQXc?iZ&Oyk^a7tU%+`WR|914@qk4guY5i z;=-1my@=n2tYi_H;~|(2rN~rh;lgOB8~;ST1p#_eY$4 z?+#fuF;6(QSxR~Kn2MpY6pXnW20Ncy(x;g{;A{80qby}G4zcdjHvC z-9_BhU_U=wM#0r)K3INVjpxFKke+{Dh#IMPMN8E*mY7vfRt!wXyrPqGBj*)x$2FQf zzgk1$zjli6d-eiD?Lrbf`~n;0-6+r5YR>NAK(??x22Q;4hf4;o;4~|W$-`3EgdX`Q z`#l%Z*B*nXy?2A3S0IQ=J%obGs@!<(Iu@{Z1m63iM3uivof)fJ;@fTMcyIP`HujJv zE>nHQLKVE=X}%7t+oBBnd+U=+@rAP0eU=jI{R?21PbS$Dc1-vX8V9?k9uR_}=}LGC23=z>sM)Q9eG2D<%3u2`y= zgZEc|VcQkD(W{>tA)OY6TBbxmAl0uf?w1G<7XnH`DGEd5t=&W>>B{hvlk4YCu!Rj-lwUuMUX~0|02)Z zRB(3x1k(7aL>{xH5Bu*@S2{If4fMZlEXP_Eyj#A3yB`UaGNemz?fa7wOKJfe`O}0h zt!lLT{sGwOau^LfL&5654tQprPBM)>;X(T;xbpTqnb0%~me|A!=TQ;eJ@1IuCyHXH z#b$Q>tpXl7^jvN{$p-X(L_*mq1-4>$6RY}V2j|}9z+A;mXmmA5SW!WC4^TBJ(25*x|9_2z!OPWx*u0TBGy#R_zB8XOb5U#jj3;Nk7X>ZdenDDrt zZN1M^kSFv+#ol{Se|ZVS&IlDdZ|k6<`ZtulQsJ@oD`A>_Cf*&sl=pf&kiG8ol`Zra z1My%C zdKa*8%YrwQ2R(x2wBf9u)0Hw6vSPXGgTxmjz`>di{YUk3AJ7ckJ={-=?y$J?g zXVFvD=W$DuJJr;Vrr)P5qgjc^*)zuoRC^o@U#}k!O^m!n({Fol?vBf{4fEWX`+a44 zKu4Ot{P1Dx_ZgGZd#1s`hjDz|ie^;a-h_(`<9PS*9{g}^1H4(+jt9kV^hLO?;MrD5 z-z^)35mP+ro2j#@%irg=j=RT!;^`))ztD@$8`O;|tzL@>Po?jPndx+go&r4!uO(C4 zeEKZcjkcUx1OD635o4Fxk{>_^m4kAL2APS!Zym(kwk2ZA)P8tABuBK^;Q?RA4Z;ld zA{;ZYL2?kR;1;P`XfQ4kih|eS%e`-*cJCpwQn5eZ6+8)NkN%6FOq|f?`BGrHN5QN} zNf>&9qu{rjyiAue5MCSU+qL`X>k;eW>WSrW;L~@k|0-kC9UGt|pf7K}YJnmBOEAAy z@&S}d+`i3961F+%ktpI<^5!Vfz&!5dZ2j2kQnZ8 zzYqf!s?!Ol|B)|6n{Y(7FU+1D!gZbv!G!JmF)O)%>Mc$YFWuh@bJZ`xNb|cebkz;u z2P*N|&HcEfr7v&G>xC=cx{>;a2Z+}8m9#Cl0wV57?oYFia7uFI7_`2`=U?87c>{`J zUq3f+N&ALrbtn1aP;2ggDU5e~Jj$n5CP7n39vqF#gh{s)FwS!X=I&X*^al-K;YD7o zPma_@ocj!vWfojfy)WOKFS$rQuA_e6MI7dK0m96;Q{Vr(a}~VKrsZVw9anv6`K?XV zB6&I0_EzM}y5GT$69V^7wFKv>zvR38^ys9n#`NjhPK=&70#D6N!bNW^G4AtU%r;XL z1Lu#YV-7L$?pFbRJ6K8HK0S>6H||B-lIvhv`XB4*rA!;=e*(WZ&)K)GU16m(uZTF3 z43={OaA>KIY|4O1XtkmO;x(*jxAlb>yi^Cyq_bN=&UdMZAnRAoO zOQ>*q5>0bI01vXSh%+#XTlKq!Muuh3wWeM6a8Z=3BrBMmSY8F4Hwua0DSy;1%b>4P z_R;2PQPNCt47b*>;+7XoNcIV3`26Pv>Z^a2ZCi1G*U836`@3K&F5QAnFL#Q+>!sO4 z@H1G)m3a9#KcuzEVAmc7D-O-0FCwqu{b><&^ohAt$uS$XERA@~LIN`qOL2jdlK^FEPOQC*@6`#fyH&>^>M`z)E|c%Cm-=y6?z62U zck{TQe%x@(d{E9ZCdZ2MLA&b&*fTX&oVenxK;}xlg_lxyXx$#xcaA-5KXXV_STC`j zKlG-Pm-eIXx#ldS(J99z1We<+oR7q5p^rDrCf6?sa za6aze4Kd6(ojto~MJ=~y2&rqiIJSBbJl}X7(oHT(j-h7pBo7fTIi$<(Z`nypXC%s^ z7n;$QH{Y?ue>8PipiP(d`AiO(Nx5vRS8U7EGU4t8Us4vi3%V>C2I}gKvX|C6;A7=Y z%lo8aO>Z5pv)~8uUmpRF_dNoojU!>ieP{AwqXF!Wms3;GR(N`#3Ie|QQl&mSa1I~J zHRGGbBUfX|qn{b9%O4e3aDBO5@2P*m+2Ai&#M(o^WH0DCYZlS-$|6mh2H9;oW-2__ zbmY^XY(q1NJ7cBnCweKl195l48|RAYs}NnN+U3b6e@lbVJ$p&8x(!UQ{w~^R^@Oe# zPw8KeM^vnQNN>jXq9v!|IFO%CW@P3e>A2pcV znC>Uu&e{bAi;Zdf%Ry59HIcTBzYaSC`+#;t8SLq_z^t-twDvtN|EgKa{4#q9ZQe)3 zo~LY~aGAFKwu|rJocsjcxo#g@=^F`#O4hI?Bp+_&yUHGK*hNZ529Y^)zL2w}b6Ay| z0ftQUlHJk!4xbP7q<51{Xz^NAHk;lcna+8H`x`?>kfY@8R>J*xDsa)?2JZ`@_$DO+ z7Mxb+a}PZx^H;eM>-bBc6Kf88LwXb0pT+EFTs4ff`bpykOcOEmyl};|l4M+~ktIg- z72hur=&H97Q0ae`z06j#4_xK}n_u1#R38+8-61_{V<5RlHcEcFgnk(KP6x^)U#Q;a z#p0=mX)LWS2U>IVVExfb!8Ff^8t)Awbk_i!l^2X74^E={EnJk?NWnzmQ+ zYtNA3EfsqFj29WYb0l0U?+up*CK0#(>Fn5*F<51N3SA~Ni<(=)uw+C7@f*|^_MTcr zujrknS8Bh2)n7X(xD7NXMGmVkCX%g1)pWV&LO-lO$OOe&+8QP&UzhI_-cFcrr+mN& z)ORTgGP?zIs5k&aCWq6CAFbqmT&VQcwWHZ{7|nj^3UAbu;rrvGxKhI$+Ey)<=D-g{ zpL?eu<7gDkZEd3kJNr`A=p+RNh1en!0qX7FtbY)_`5g4 z*)P(Z>HBzmyWf!feEJ@IeH?lFa~Jw&!A!{Ue8rBXDdM`1XW7ryqsTvLuO;NE@Ne%D z@MKXLY)TyoR~OmBw01?z{I0}m7WFFre&(3$h~^~0!sa9?emxoX`GjM*>W&Z+ClkWj{#(-@d=r$@*JEl@u5pMbSJf=)miRR50Wj~!H|JXd`$CU z-gVY}HZyuD58z4s$?knTetx7Zuav`)kPK#4|C}G%vyD%;3*tKtpM>lKm%!343??eX zv(jm!Agt2}s!dY(^2*&jGTfOjmgWR08Ro<{<`Vm~(6^*$MgNkRd;Loi!`GwUzqvA} zS-FDm!h<;apFS8pJ0Up#UBhQy-30F4Gez|%8=jS?LFZ-0lQ#-|Ayk@WS1;PjYd?hG z*w3HH`+Jw=g#l;zcNT%pItK;4MaPAHcFQo$x0;i{{37lRqPpl$4xbuM@{OwvF4l4%Gt{($g_464d zcF=etTM5FR6QHp{*FbTUQC@mx1&OLiY@UaahCAT-Jvr*E%?Y?G=IPqsWU)46a@IdVq z!T8K`vbpJoct(2X=Ew#>{aH1bv_TCL4>_^plfBuMgRf%_5=6B8_OM zr-A#0PvpUI1JUYYhWxa;p!M=PVfXUqWI%1ZtR_+& zc(pg%*YhCeEAAy*x4Dz__%dOQc!aI!%(b1PeqH8n>nY}>7n6=Po%sCJ05a}M8_(@G z0uz?^fsT)};Jp1>mN#s#?AQLuwAX`+Wcx!KOl?fy3r=?B1Gd@mm|#uG|JNT$?_KOi zur55196#BsI&rYwNLX%q3x{0!haX3rleeyti@(Y?3JZM;Sl+V3g8Jts+1wEb`TzDq z%AO8%*zga#wG#e*C+B*>n@XMxxXPdZ0kX4Ue940;IeftJy--;q!f!zZ)?L;V*9~`o zvZw~(_o1om*Q_V-Yu$CS=Ja^<$9~kNxf{6Ter8U=skqTZv$#u@z%CavYISKi24`o8 ze|*l91wsaY)^7&SYv_(^F3*6M0sEM5_(%rx^At2 z%Q~V^t)xMM`9t>FN`!3lBUDw}lP+jjPtW?_2fJn`VW(9u@K%nMa>xB7bLufU!v`owL9fZyM{8Wz>}XHFco-`hSBr z(cej|t2&&?YL-{IH-Tvbml|*e@O!y188gsy%9qTO${`1wug~pm#`)3WvF5IUpt?k9f4V)VEt^*qu z!}ckpfJ65+RDR$+E=V$xd?&Bu)rn)^*r@Kz^k(%1^CAU$a8D)MA+cX@+S z*MoGuJVSU7XQlkQ2FFmEf}QY!H95c;dA2#x?3oYkz6x^Z&(yMaO7x)GxrMK7+Z$Elm^%6GzITLIATZp&F?u8PC z8Q7<$g7LK)k43hILzVYyv{bShT;zRVw^#=AGRH{}Q+}3QcGX51U7|2=@^={RO-BJ< z3CJl-$eh!u#dk04z|YppP^G|eIIGGP-$_Y_)w!GLDcFJnUFC^|R2(YXeicrebNCBm zfhEr^MJImhVo`-F*zfCZe7Lra?pJ%w?(5P-Gf^U(zNJs>gFmo;wd6=d+f}y4>LPRR z?^0&3@B=ygEtk$%cm)M_XkvG-?=ba@6V$v_=VZ)Ap}yxAwYjT+f9WVg{_m4u>Ush# zOArV9`}1Jt2M-#ZnU7?jNz+C*AH3`Qa{PVGLdbpFL{4$9Q4*IRG#*$$r#E#nPfzy4 z<)tRHq^^l=jJXWbB_r@{MJzM(oI4pGgsJM*2ng$cLU&kvA&Xsi&{sY7j5@z(8Vw#` zb|y_C7iw-{@0@&`b?5-@(~pO#a#7gQ+y(dFj%43_KZ0X-Kf-?PV@QWU4J*Y)fv;FH zyF6?KayFC2)i<K)Zg>l&uGuh8XEp9reudVT zYGH-<+SHEcj6W*<%EfH^%yGFAc(=kq5MPyIb5j2>wC;b7kNjRqQ(|gB`@nKWWOE>q z^|?u}SN*0sXPVfvLF2&1$Q$Ojej+!tWU-2hH#bwsm5UD#h1WZ$W1l`dENP~Is=RaI zZ`A@C6Bkb|M`w`{=aa(A^A6FP!#_!-XEIFSbn(mdsW@5Y9GUuf5nZxqI~6T1}3aZ!|r<{s^lnTZzB5uf^4_ z7M!N0Ioor08(Cew41~4XpnUJUk!o@j4eI{pzb2M$8N z;2w;6=-~wgl6Y=X8CogT$!I@6PM^AjL)L@$upnY0$CS%ZpTMxTp`vo9Xm)1pK2bot6O>Y}w7Y{HiVTDYILv;RYF2^lE+ z^js(szN8Nds@Py1KBE}tfy*o=;+cu6tdveT4Q=^LiU&S}^*1|g`K1(=$L=OGGA?p! zC-VHmH8Z)o4n@`_FNoQj@E#>k%%aA_HzEJ(J7nLGj{@Tf&2apR(uxxiYo<)q{w&~L zzezxMH}E@*VJ)&{GS70^{+eF=?+K-Q7C=i`0##Am&Y8cACVR$A!n^zSqOWcIGhl2R zyEny>dyzH{9=x7G5AuGPe~lV8?3^WhchL#Qz8NCvMtm1*%80;f+Dm3pxdN(vFNMqF zW63LPNl*$Nh6i&b@wG?$v1#@$+R>;0nU%J5Ok+0c(MTWzJKvCzqt=i%UW_{~=Etc_ zea^nmip0MK1fi7?yfamW-ttVQo9g3mQ52>dKJ`-3!WrE7SBL4AMiUf2KZE{B@x_0$ zBIv25Io#*SRm{sIBNUhy$b1&5W_MiNODk`mp`jSTYnN=7QjbC~(2 zIgZ)5$p$t~$c3~yIS}5EN?T4QLtt+ZR*9D&D^Fwe!gCD$*FA_5%wu3qS|VLv z-#|Vun}$5hSA*!G*X%gA17te?w+-sx1XglI=EM(G@Pqu^|h3!55G)YE+W}D3?4be?x zzREai<75SU*U6J-TW*mfF`F1K&3d|THO4`)xo|`U=zIPiKiE`27c0JnfZgN3U!s7B zmfjW$g4WW<+DqsmMNPPTp_4d(gP=0H5K;}tLa?1R48C^e#s>AF&l0ilh{>Rnt{j5D zMcTN|YZoaw#}f62{QdHLE!2PJv#G;>QPd$%db{r>jml-PQ=%%UVg9^|66&HThi616 z;Bpf$WT<4qnNGLj4yZfAzL2w|Y2yb{oBAKbSMmGTTE5?A7y;v}bLsc)I`qA520XvX zQti6$pp~74hf`zdxI;&A@6U<&)rt9Fl$#0TUN~{L(Jeaf%4Re%!vw!EYeWn9bMdd- z7u`INMyHm&hq)Iw5)ba^CMtKrSF$tuFCE6UBJzd!S=qi4g`Ij}?`G+`a zZK7}NuF|jjj?h|(DjKqZccP5*pelQm@xGPHYJwG)RWj=YLtZ?o{3h>4pQw=~JK7Q6gM-9~NYNphc(7(uMPN=nkcY zG;mEi^skM;|JJL4S^Y=Q%M+4`6-8v3B=2If7iZCta=t1Zqp&;s&3SHZ5!1Z+S zTpwD^M4Bg%>euD0!-^j04|c(p$Fwo`A{YFJV)=eW2bxpQ2Nl=#fljT5ZPlkmR6g5^ zYk0>qcj_W&z^f@-dFFb$VZQ@x{PPJW{tXhm>e&j_C%i~j(oAq!oDWqh2--5v(}xco z$f6bE^oOfEldZ3c(zeUct9N3E&IVbCv(KUNXB|LgIs#Gci=;V6o9T0Sf>a_~8AID} zp~1~EV!Uc5ZF{(lNL8N0SM*I0PTqy*9Czn)kFnI4ekKat7um=%9d^BxPno;@Ka#3( zRw%RikIjb3$As(WuVx+grK04r<*fTXNt(IJ&-!)WMX0*|m04mvh`YtaIm$tZE;4^zLai)k&3hpW1u z*xShqLDk|GIKN^+b^bgg+Z2b2I1M6|CkutXQ8>{g4^PUPj@!l6@si)3=&DOEdUrDy zXI|oPh3jRkX7dOQ%$KERZ>NLdf6{!{Cy1{4ZA<0V{doq~dw6@p2$o8uk;ivS!Re_0AGYIfK|QsZneO*2HFzwD$oE@$ar$SCoXY8DKiRS=jBEk+H+ z%OUXmFZgEm0~SB&0FUmqbj>a~PE|Pu?N!+a@2WJYtaX_1;r1H@fB(YjG$P@j&B}E7 z8A+UX-GH4^YJ;vXOkkTvkJF86-c-lAgm@X9B3u3@(*coxyr(Z8-!v=1rm+k0+=qtr zT5=e;Z+*$^k(?tu61JOeekP8L!d18|sWse&@#$0{FN3LitOOlPwdwbcog_Bs2Y~`z zV)=LrlhNuMv>sl39W&H6NL%ietPh)(m0kT+}sLm)_bo zjJ`%MA;N2)QD&$Sn08%&``c&XN`~O`+JIAP6<{)xNXL5zFrizga&_n5v!6XgVf#ik zTIvfh7W8)EnN$+&`@-j9a^soifj!vA_Nh6xYn- z?j9~6R^I>6)LG$7@SfQuCbNa?>6aCH#70{S_pHX-1l4#(Ss!liuVWIA529beF{JFP zJo81fOdt}RCwP8o39NOzNw2xDgL2#HP%Wp0L*Hn=RTLO`@^E{p?At6X@8AF)VY!6m5O91XOf0z;=Qh6;(P1 zMV>}*D}=(Iwh&0Q4#)CmR^n~D{{xX1h3JKw2rw;DZ1=5UVZ+Jw%%sym+}oxyHx7>4 zm>t??lm1~FxomZl=~Y{UJ>LS3xg3m=mIvTg?|7KKCk)5*Q#!sz29NL936*b>X~ThM zLOEw+M&&~ii+nI`z1hQ_H;JW?&#qO#p?M65}j zmOfYkpQinR^^F`F8C(VuFbEO`I<&TP3jF6Nfs5z5pf_XGZORM8gh5aJ(26q|Wb2hg zGCw=ILb3^0*q@nP5piV-oXJq*mJ3UWwNn_%y@>?9w@uMJ7gO}C@hw9%o->~-_b>-D zrwD#m+@iC_+#$9ny=c1jWG<`87ZyGIOR!cloHH(j^_|10Lb+71&F~IWtQJX3sy)af z-%u_sMV2f28iR}!;yF>}w@A~-k736~u%_X&*ekC>1%r|D+{;s+cpqT{<)jO-`t%Y| zTe1T^*q}$m|A`7nWQE}2g@c^u^%PPZ?TP*{Rp{66c(OQM3@JP_<;Hp+r`{bGVfLZf z_wst~g!chTbhB&FOo%9RAzHsll5( z+~dIvtV*N}PkZSLaTjP*^QAkonwc*a&h&-3fLqxs#-8|>PNwL_LI3g|l5U>LeA$zY z{u&F(@2nO^`duOvtlUo9`Dvkw+D4}D&ve|~7ex>79?3IkGP(cQSUCB{4dhl|$(1fV z%*vt&fns{3 z_6gZ3oWd>nwSdzfjo@ybv7;+*%!lvmcY*CuWvpm5M6y0)1|ZH;dtFAW|1SyY=s@D48(}m+6&Cl{8Q|K(0|PO#B70O^>aARV+WFX(#@+EJx$TF-D~(5v;pkrC(z4$*Wo}$DfsuMv22Jrth;jsuHsNylP*TZFFTWSft}o`x$nuq zOpK+dBdO}rCnCX?ovY7|SWcNOHQd%<{SFzy=cAC=f3na+K9geg# zG7$F8xddX>9iTpYm*96!IQg?)3hPw_LRNh&DL3^kKlw%)gd;|{XG1f*sI|f7*E>M` zP&4#;FA&Q7DQ6cdsnX#`@$}uoU1Y-+A7HE&dFI=U91J(nuUD9`rQUi_We?^lENw%EX;PfJnjmFsAR>v{BZp(`>Dya*$v<*@kV zE5>lw9=;cT402XR(rFhWkgBF1w05ZTH+dT>JUksrJ42X+o%fOD%r^Lt>Ve3!Lz^>^b6O(Q=uCexmLr6s6cw+VKFUvK&WH2r}dU+ zGvUL|HE?`yDmrJh3ZQu(>XTba-ZgsRJVW2s*HYQamZgb9*%Cg2nUjH!-P@3=jI{-j*Is|X?P)JUlh@2n}+Cw zXLr!x&{*1WYdLiHgus0XCFrlSKnv~q$a|p$&WVyBGY9{oGq(ak={LX2-!8%CiMh~y zGxWGWA?YZug@V^e43yq$rdO}m0<-u!T<b}Lv{E%9WFRa_T7C(lR~#~(LU4Z_X~B%qK?oXIvUh2^)vn5Hk}(@ zV*qP5&Vq=!Jey}l6zym5U@%Id2Me;?$KFno!b*sV0jZ9eOTf*64M^%d|w^9Y_* z{bl}C8G_tXBQpK>9keI)ISMyAiv*g7Au(44Tl|`c*AAS9*bnUxDxra0`Y>DG1!Q>gO zG^j0KkkKrJJ@*APBWW`G)>D)UR#n5VMg>0^V#1-3}}o+absA^$wj? zM#|EgnQ`0r{%<~o?tjx^HiXlq%j#*5=P!24Gb8-Xc01a8qK9YqRzc&fi)clB9B!17 z#=~WqQ1E+`z};DIr17$n3jTO1y|(}Z(o|G-H{1+Q2ricP}) zqkn8mn8hz8P<3tw6Vramrtz{0_ScifGjhryi09d_%8o%nE{TH5#$-63um>OI?~V@3 z)gZk~3F3PPh-!Q;oiH9hK#MzSVh>y92sIhBLu=ZwJp!;9hVm_YpYOEshmc^AsPBUD~z z4ql_E1lh@PNX#`1baV?LQK1OCm#>`yCEXRIJ#ulcpwUCr|KM*QNZ6UJ_4&vx9QFxF2J?uJ?0nyPX!2NO){QI^7x)wa78Q;6X zvuO_Z(SI6e6-(f|h7Gq4ytu~XY&zIwgc&4{H@tYshBh)x{iTUWeA8`KPxK;Q%R54* zT58}`BWa}fr@mmFl|3!s{fm2e|DI=C9~E)c#*?196X!FB==hauZTM;5{D;~Y7yY6VIYufx}us&sbzWwgmhi?iRn7(4D2!^L$&Fq&`$ zOf{avucu$(g_S;9`K27f%i~G5iXI9t=|dCNnG)k$N#u8fK1t2jz&`O$$e5+4>9c`{ z0_}_cQLUI9n4d)PmLdr_npp^;#aD%n6<;Ala~G{pPGI}eVuT?Xm)saSO_I?OA5Jl`FS$k^(orz?O5qZ$*B`Tp)CC9;N(m{HV!<``bK7 zU;UMWbAy?vQPF`W`b*OJM4!YQmZ0C7L~w1BIk?rjQXi@5P{uRg zLamf++P8$e_0gSkai2mPXFGAFyEkxe%e(1c2NN)N@}Z++{xPygbZFV5g3R zSrP8&yg_=lq>}ru-h}hG`iX|uEGDn^?SrlFSFx=t%F4{FD@j@EMNV$@HttI>zZce* z;Czld!&QE#ptL-SYn466oeosz0?ZUDDqcL~0>affw}!#d|3gX6iZ zq^HDvxCYW+zM6mX^mQMB0NBDLEkO{c~- zvZK4Kgql`nM9or#dIw*GqMJO){mnyE)BcOGm6SzMM|7#|15dOfBZln=d_Y#M$RY2U z3y8V3lAHhYJhvOaMK8tFxqH69NtOI;#zx^FnOAX~Y;F0*95a6iW0kDwA&t9aV#j=HlyRaLl@i~Kx2+>J`Y6XLh7uWXXIIi6P%0~2ThebcK<^3vr{-4vGf ze60q)UQdbQ*u!K^5}}V)FQNO4Mv35_NyT4jLB(Uc58QF>Pju`wWp3+u4VXML5_zvY z4%#g_MC3v?-)&Z++B~PoiQk!_$=m4`;Y?a{pa7^Ut9>OdySHk^B6>|F?=g?BObFA0yGo)K(D|vlmI%GLrL*{CUcxGh;@znK4iSO;$ z?~Et;6m5h|W`v?l_s7gNi_i2`%te|N$syxU($vmkCjIwd89i7s6CDT)=N9{KpcmFJ zr*nHA5s}1XZn51wEbV%XY2MFTWp#a2tcKhjJGCyH>j6IR0TGnO7* z8%Q3w-iLFi%Rtjs6(1CP2d9!w(Wm@A;=cQ1nisT{o(lX7-~OIMpK_w=f$oNRArRwK3;f z)`H6Id{lH;RM`5=f~2uA(rv-0s_P_Ot>?mqx$xbC$Z&8x z*@}d`zj3em8SZS!JoM|H8MkBJ4Ek};DzZG~8Yg4Fll!bxPB-md1HrfNQYEo|I?XbN z9GkiYMaDFt>dC3X?o}VCe%4KTyY8!ylkz5ujWjrwYE$_kcfuL+ zP5vCU?=K)#myV&bC%Ld$t)5(*YXE`{JuY$38$Q|7aPleUq>Nrve8pbD3yt4keJJs5=}chIho@tpWQhC38sMoSX6(`IQ8Zu<>)PPeK6 zc3fS8UTS*VT%wP8UX%qA(O65P@8po3tTB}vV~Hx3wW94V|H#aPDM&IgnQWGmA>Izko}mX`7+m6HX4G;0O3msTSc!t#}OCrmPf76c^ zIdEo55xC|d8sl_}R`auLgVe)lME5O>-g-pmMf-q9+b4W`hd#bhZU(QLE%65Tg~WR3 z8@akTglp{j$vhZxhrtL%P!6qvzKb95iQBEvyjzS4zuS^cUKNaCN-O!n75ve10=CrbB#VQcg@3<~5UBuP+^=?pZT;#Fg;Uh&(pqWS-0H)1E+R~vmjvez zA0Tv`F-o5HAG8kG)`3fgm|_JX-t3-(epCsl={*jJ7J>HSg*ax5H=N1{WD3g< z@QpYX{E`1=Sk?55JTR=G+N+l%HTQd{skR(FE`A6kY&%lwFelCuJJGewDDdQ6x6ccI zvVKO{&@XiXPWk*`rmeN(9_YQL-6{7_b^d1T&F70y_b@y3_yRGyc9I6oJW6l27~u_~ zlK7HG6g%&oJB%rEr2-`tO^3q}qqCGS@ZZ2-xD`aG<}Q^j?&kK+@I+2I6aIK(sQf-epTXH{YN=ic9f>nYCE2 z{49Rq^#H%F)}!1~` zPM?|da8eI(wt7r|nTt_1e<>mimcprVOR!}5Htd}AOSlZ%vdxk0#p>sd5DKn04;6!EWy9F%ow9P*bof@4Nu?Ca26 zjB6tBwmn<87jI5+9wBnjQ`}C*X{L~$UZ2>iix%K6dx78oMsc%3d+5L6E@CeC3Ceu) zaM}AdJpWWN@=Q!brr#q3Eta12>#`0K6WE59&FO-=na60?<#A+1`vdk`{Y+#Zfa$I8 z<@jLwXV`Qt5IgorQjgAkJSJZnXT?zbWllT%JhTNbkZk~`kG5Fomx`CfJOiDDQ>l{3 zIGUL{#&-Oa2dL@IShCk!2|ImWi|iWifI#Cu3^{Kh)}fN@aqClzeA6m2_TzifFrbER z{*vMn1$$sl|4NeZS_E!ATF9RZRa;Ye6RIR#2E!A|=z@t(@L$4P!QyanLEvtEEHyeE z>U~tXur6_`HM|e`_AQ2@Z%r^WUK(pJH{kx+tKf3}S)(4d9ok}4pf1ggE;!65$2)zu>9{X2sI-XDk4e+5uhmJYw} zOrY^^%2D9bB(z_7GXAb+0OI+@Y*FkDti3ZHHcb#Pc0aYS`jUEl)L|2DliUdbqXTf7 zxAm^Q1y&!h#-h%*phWHj{%=we#H~Jr{w8`-@2Ybw`YVoXv8L@R z$**|7rydUaqeEjZr;?;M?!pXAXbtjZ!akIRJ`2bG^rN?i9#MW0gelvixE^_hyaC@k_!O;lUJpmlH6nZQuV694=M($}VCc;llJUkA z-wXYP=J$EgzXA_z^Wq8on`KNR#_WYFF>hfhKlfj%_!+jw&qQ0TMDWi2MtI3Do?kWh z9Z2jsjwhYC%d{>GCq;wb%IAg#v)!us_|bx!=;k9|Eb;pZioQF5tgq|Ao*OB+-!TTK zVF&g}>lipJEp7XHeHSi29f1uawqcvbJos632D~=)fDX^4O!-sH6pVR=;wK8hQ|~XD zpP>iyJd>c>%n`Xi7iVXEwPVD}#^L6MJlOJb0~Yc6Lc=?|;QZe(tSNaCzdlMq+%ShN zcRCA#fLL5AFe8fw4f(xTJT9q8fZ7tCbFa(iuue9C{jU-Bj?w_hY~lI!=3x+~{tZ>? zDq)j*LD;p@0GqE@$GcL#GuAebU{1UwF3nCytj|fjbWJW6JM;#>@q34l{nNw2A=7P@ zUK!iId)BXQC>^j2auH&6X}tU3W}<&^ z58=dKuy5n$bCTj?uuFXwfgd~ADT&8uWvM*%4is|t*Svsv4)!=opRI(X1_nkJs0~(FTahUJyyJ<{Z}g2 z91zO2+Ph(6Cuv-gYK{DR_wk+wU#@9IFL`?WzA(kE1$0V{If-mP^!w8AdHM^@ zc)XilJ%50EzPgRRGH9?t=YwgqLl5BjTG-V)g!(NVM9RFY>FJ0Vsa(VFqHVu3rwFWK97EI4*;j4FYwdLf!=(VhXPJTvEAN9Tt#yR zSvWm~PAM*-AA|+;NhJSmar>e0*K#pP>!>H0HGIeY`bjP>B!&|`oxtr9&7v`_d=Tsd6OO7@STsphY*LucZ4p~s?y}B?6s@$x?*+<5RLqQ=D6-UdPposZ?})9SdcYvU z0Q+ec(TnfvsQp5WMMtwK`2qBHdh{&n=T>ks|Y6vthNn-0d`q@ew`BUnKg zNG1C+wO>S;sUPOyM<<@qnP11!IExkZ#bqT^U{`r0Nt(X_ZPb_lk;Xkl+`F=e6*Ww>a4 zlM1(pHcl%#4_!TVoHh4OM^0V!s7!Y{-IbzAH!EKujVU@H{lic_9Q-Ym?6%L--a%lurF zwtT;^-OwIh#rSiZ4aZive#qoZuZ!}It{EU{;05M9Kk4GCc*a$~l+2KmWm5S(`H)H& z_!lbC9=EgP(7ZJ;aL5g1tf{BbUyhRj_eyv@IuEw{Z^lklKhQSMNcbEY3sL-ib9>D( z(g`1^sLop4y!;)+9(zg0yY3@WGb3S_!EJh1>?T~k#yiLJgCROLfVuKCPEa~l9R^bW zl4m*QDwsB^Tsw2rw4{0Bk~tXY&rGdFS@44d$TnAQ6dF3+=vIOesj7< zGmIYjdV@^ba~d}_J%;3eid4COBh77C!5#bI1xM5yVMgX2GH-h<`^e-6+0zz5S6CVX zYkmpcS#qB0EL%(tuCb>bahhEEJtyj8X+!t({Jo;^2sr6c4(ZSKLV^{at>iPY7rGLe z9oKaP)$%fM=j~3K@~oWt3h%>u`RydoU=8;MSy1iNB&s(z3Y}H2hKagSppfgxD%47F zOIswl;UyEeq@7}%;fDj<7Q>TN%g#^W@@_259Sg|i=OT{rT+RhpTagJPDad2HF)FyR z4=Kbvryf*`qezAuct4AiyxK{<_Lkj~5iD1HA;bS&r-dii z(8-}ci2@HtnLg$+x*Fb!W=uLoAB4Rmkb8g*|BIlm%XDB}QH=1cza=~!(})7xbLhm+ zOXwuUv7E^9-|X_`lOZ9tA08YXLB;eIc=7Y5*V{da`a+7{y*vpy{fTIINDRu`J%!f# zY2e*UXRx=v8_=4F8F2or8$InPN!`{;GAU&`pndQR^4)Gn`#$QDT7^Xv7uJ$PWDYaO zx1NT~IssGPPUL=TKBnWdKEcjxZ~k1`aQ;sS-0>AcQrKUpo}^2^HfVC^f3?ubtqL#{ zWl6{V{>-3THKcHPED75HV1C(++drhpo$6aax7@o(m#lWA39gH%k75lfzy6e_bVqRW zqsvLM#ZR6;htap4y2RnJ7MbAG%FenSK*~n8z=t*S;qYJ)Wu{JpVF7=SZII{Z@7HMi z0~I>F;{*L6|BY_fI?QDTd?w=`DS_6@7ht^dDQGN`1(oVTIP>}qGI3MEo(`EftNAjW zHFBEq-;R*T*g?ibOP$;|%%_tpc;55#Kj`emF~s@!Cnh6h645aog1H55>@@WfoMusn zf8!EZ5HJZVrS5_Q3+B_tJaw9Ve=Fy(!UrT3#o&ow6#RP`Pi?Cl=^=iumULB>>-=#A zlw2awe@<~Y=Y%9SiXVr6$e+fV%S!RSt^4sw(LCmhYC4HF-!E|g@PJ%9m_-l0eL^p+ zevEG(x5rOyzk;uvE0x|5LJkSFz-rx8nou22Z(XvZ(d8NF67SrbC0|GeTX?QO*na$` z{s*WP>tm0)mpC~#4*LnU=(6A#Y@TI4$9(IAD;O&u#-8a zCJibP@$^5h3c7qy68tX2F>@ugZ6D6+#otF|u=nU*;;gU>i=Rx;pbI|L2v7OR;L_h!@C>Pw9>s8E+>wRS zZcZdKFKmTxwE>V(;ti#5G9Y#BgNW!S)B45w z+!dZhw{NO5mv^I+Tly`HNL>CzjWW$hLi~7K7?%ziCtM*S={sC+PQ)8d&A=zpucK?G z6FJkf57Cxqd@pccGM@NbimRAxg7@g8qEqR|>Gc{tFx)F5ly|yA7tRXDi*@B}JFn~8 zx?R?0w=FzDMnq(}f|_&K{qBD#aC3H)Cg8^>1Joj{DatH1goa)Wj!w2u+~S~(a^+p)Z13e4vzdowi5Dq z-KK7^A9lbaYSOrNnmV4c#t<(nlptNn8;GdZJ8I_8h#&ski4_9eX~&-3bk)WZ=ISYN z>OP}|Oy4w<%TeFYNf}q5h$IMi1rhUjrZM06l-Y zgS~m;F}P)x;~%x_1fLu&&`@XxP#VD)cf4XhmB|UqA~RTbJx)5%zKo=j{ph5ms zUqe?PN#Oc~Gu*rq6YkshL=@Hch*n4+2a)~KWCn0BPDz=`$~s5#M-kT%znybl^bh4e zn~H5)^5J@5C#qMB<;Lg##|(ba1lRdrk;9FZkhJ|BH2hG6)PQ+Hze#;;k((tqX_RHl z@NB%7F~YrQ4(+@j#ylH$lc_r)!bZK-qPl`+ay!Wityg-CGptWz#VlnkF=Z}t`J)2U zYHu_1(%z%)8@*)FxOZrHK+E<3(~S=eI%3DkXNc3>3xaDsa-{6qD0bleUw^zyu(cTy zP8HWDr|*sskN)il&-9`#dv+1V`YgQUJ=C2^Ca@G+!mHGGNGmG@y>LGX?MvI>{f`h~ zFKGai(=l*(s}vl`G{BuoqNr8L5(U1=LFhs*@8NrbdaQNfKz9sXQdmLn6pEmyjoOe| zUI(Y&w}5c`M(Bi-s9YzQ9FClfr+!Hz<8N(-qX;Pel}&EDrQo|-lksmG7SEQ+gF{v) z(ZkU>T-CQ0@Z>0NF?++TRcu8I7G9?&gK9)PBNd;M-;cdNDq!=iuTWy=HS%kxDtWDZ ziO=LYAco~T+W#5wIX+!mQ;SO2ynF-|y`RdxtS@lTPb?CU7HcBZ575J z7!u6uKM(E0cW6@eM>?l20=E18fH|XM@Pgm@VA*z-6KOg}7Voy;j-Iqdccw{We{Fu& zg&=nF1q^~H?)CaP{C*r9h@>quVH`l!}VH7qW zBS%}sz_D&D^gn$?T4oH=Q58GbJ!J;glA-8N*^-HujgRuKJ3*|1GyGS;YG!71m( z3JR4(VE!&crop5Fg*&9v`rH({cG(Eib#Fa+`p5z0br;}Y2dhEF`80GN+<}UkKC&Oy zNs?7h(n<2P>w-gH)k(G01l;~i!&dA?74FoIf%LU+Y0usU2u#Ez;;k_Zj?@c2E?9>K zDnl8e$`VxQIf~?FhtP;wawz73o?xhGEW2??ifQe(B$L|0N#2yAa-C^INNbuIX8q;p znzt;hkah%Tlg|RVofdSg-&hhSx`_<&`L(*)UU*`cfIHF~k9OI~L$bo{2nV z8)Er;VowtC-!DQJPv`qW3&f#*zW^fJ-ZGavM2v*E4V(+@HL%OF2$hfcG zc=Og-IF%iXr$1A+bzL0~X)Q)nsWA~}yxquEja`KAsp=!?pgOX_#S7gX38G8SZ=-uo z1li0>`wNTb-ljPLFPWG1Iw1Pw2pFmEB;U$b!jbfk?2_<4w8_2&e2aZy$yQ(dy=Ff; zDLIo}anTMGazuqYN<)}}=M{oP%9X`rU`n7dc&d4qy<%h7ID zfpKVm%8Z*{2MT%3Af{$YdpAAfPKA_k!%>Q0?{oz>EN;cWyIipoU#T7mOG2SMvrJa= zs~}$gFk{`;i1LqI`k$in@Tcnk<2c!~$Vg>{N{EbmKA$5Zq(o^b(GVK?B9)d(HX%_N z*<@s<;-1g@b6Xl3qDZBb2JI^2kt$O`?zPl->=v6iIK}nwrgA~lfvIOGTq|X zx#3x)F6uYATcSaSOzZf4z+aT?ABlb*yM-ptcth?E*3u6at<=Q%IQ=)|$oieRj}G*| zLXR%GQzb`DQGw|rLK6IOQf4aJqQ4z~99mB#3_Xyg=NcN?;6%fJmXJ&4LfkSW#hx+I zp>%IA&GeavALpmwebWe@kNF<>j0DbP+i!AJ4WPMo4iub`qur&kr|b_tXVP4`$)@xdZgplk2eQYBGdXz0bD#Mnsw##F1xSka+bPn5L-K)%B~zES{AB~ivFr%H^Esvy*7;;yb*$e z^mLK(Zd)*RFTyT^DXfPi@7vs^kBVc{VEqGgZu^>Juz!@1XkwWuTP!h`T@c@mpS=2o zE^lw8??b1d6>)$C|IL7ixti>#co+7|xH`1^^D3CTMxX6kGDtT2#Na9=f{ZF+@Y%>K zOuPGLfhll;U6(%Mr@MUE`0tZx`KvTmb5MvyiaOZ4-pOm>%AAOLx!3TOczzdHTA4Xy0W!ON!zqos{AmA24@aL7G5)E@y<#3-lvE>A_ z%-{;ne19D2)^x!u4i>6i*&Us?$*ZeVVB0er zG-*$Vpz^aF+a09APDr?oFR%yjf^sod(7c?mh9-31-)7p6BI!jlYgTZlgl_&`4=#GA ziP*(JDkRv?j&~eyk}^laW-sS zs=_UjF2~lVYS;n@ z5I2PL2O9AF&Ks~3#Aui^LszY~g?(dw(zY#oAmDu;ua}+-!o|-?qcCdJ zP!FzUjWEbFUhZD1;N}FSk!$X${Cko%wCE^f@BjQ^-$EU*y05@I_4*8MgDc@g8oz(s zItZ!O8dNH`4a9$C;Lq0ALFvyOSatgxP474g^A#_Wl%46csPzr07}-woHh!0UXa^ak zKAzPo9bqn*-vQZ0D-XT;Kr+W>4#n0 zL{599MHg-<(DNfN1rKx!!B6ovd3rU4hIozUUN5W0##6fSbjvE@H1h`TH&4NlcKT@5 z9s_p#-k+TC>jRMd5yQ{Ie8D}?0RBCh4Ev^3!5{gnWP6y9$ed0^eP>xOH5$H4JKGgoOJfIm07V1>n1 z%uP`x7B^^y+MCzliAg>h87iio>&vLDqZc^0|00#46X{BEJ=#7`Otjako<^>3#G^f< z$?n&K$S|Q179BlEdJgE(Vvh>4%&i0`=47Bdo=fnD1`a(8PM|F%LGba_Jn9v3id&KV zmYFws5}v`c2c557Wd^EsqqsD9~Y_P%$1;OYRo&uUb$MXF~U39NJm&WQ$qjvBZs(SqC zlB#cT+Ok!4`MB#cS9D+!w#Xg=exPQf6wu$%Xvg^3t?2+#^cB@ij2!MGj#M&H(hwA4EIUo zVa=c~HVe1RLSkC>c+JN%L% zu9nuqeR|C{-BuYi+-`)K3}RSL>ON;26variI^!{FluDLNWn=WjZ6mH`!eH}bn^{E$ zFtS;XWrUZ=mvNg&!#z24f7SuM*Zu;nko*Rc{OC+NC>5(t(P6W7hrvK^H=XYO8dfE_ z5PsQ4bDt(s{iGn~_gEK_+L(#y+d-I^><21ikJ3FziCOB!=ct%xM8{+&di5y+D^4_I zOH8lv?p3W_gEr9(u+<np~%ElG^&U8~KM4bH)enV(Lsncuz*6q!yLRSP{7HCjd<&kcr0`keRFu&^7sV#NAZDHktbuG6zLS%W za({U-vOGUkz`r}6Ip~1;T?}aQMia<4jAk2VA7P)&?BL|4uEygIPA7)HUvh0NSA{3R z6$*~|qKqRmAojKct5n-BkmxoSESnHO&c7`H-_~YSy5%kYyYDakdw(S@?ahW&fo_mA zX9hgZI16s=n{fKhG!*d82pjaaabd}kjEzkTOqbHR3ISZd$ z`Voirsd61I**G}e00&w;M!gH;$*p-=XvdX_IB?n@@Ob#C=}wD2y?mvRPe~U+)Uy>x zz@?&ng=?6F9*6MgJOkEE<}34WaV4i;pO0(Df5j0+CzwrT7w{FSXGBrhNV=v+g5#ox zq}*aI^rS7OhPhjysHulE6?zMK4>Gg$Ksx&E`oONl?1 z?K4f_h#TWE8K^AS^-h~5b3i~|wI-7i1f*IU)iI@y*!!M*fsL3xcdg3a-`--z< zcUs4>cJ}K~ZTU5%$oG2ot<@1-KWEK#oQz@iy;pm&<6%B0oq%N;rm~YyD6xxD zDcJI}a{Z1x-Zj$#`ZHxf^_>vg2Tg*Qo>}DFvUKhb-;KEbAePOxEk%#7DdF8E@0hRO z(lD;mhkqJ?veMo&RZ~Vm((_B$Mf?Pwwps`8U1UW<^uoaN#V`K-DTCY78jIPb51=D8 zo&B$eMaqc*VD+EAO<1rSdq8Y2Hho>sMs?Y+rlKEsGw+w!_B{-XsdS>T!>hTQCn#gT zSWLJy#~ZAY)F5-GHeI8+0Ak;a1$jwBI+QeqP0$>|q`i|H+nSFK73r{>iH~OBk?e&nLp*!x;Aci_1h+wHd2enBr7vYp9M1 z!QRI&u-CS&WLKKx(DG4tar&0YY~1+(TqZXG-@ptf7P1j3OUC2dcENU!Bl!P=4e%;xp_T<3vPNXI=16@QzKOry?` zRokm@li(bZFr?fqm18JLHURHZFb21z8|D1WT=rjOXbE*mg-Z+r7J& zjGV85(p#gEX+Z*VOqV7>8{&luHxDzTbiLr-9cNUvUV$2ADnrs&8c{uc*Mg1kU*HEn6xlcnv33B$H&V`7k#uk?&d!U=K4t+?K0>jWR9p zAJa~p()N{G^IsXtsp}-}*QMzAz0SzwwFcdHzzIALSV7x}HS8)(hYKSLux&UB43{lp zY!@2g0jU>=#|c0goWhIyV%Q?nA>^=JlYQp0l{`A`#w{IPfG2-3U>E$%WDF1TeJVc7 zpit{Yw4=?yBa8QLd+`55&mh{IaSB$otrDCYmSQ$}ju7X@qwxLLarXJ9d^U7wK77e` zAoclq$f$NDav8AXYIg4Y zbh3zNQ=s#FUNfcvd7hm_j*dEj%yv@H2$rNRVxKvAGho{k!tpDG27Ime2jhJ=n^D=Z zk-2f4;*Prvu9fv*58l1U{(U+K3a^&HByb?&m(^INhU51?LfrINA8~PsB&IkU4+RR? z>cftBTc|RMIrW?f-n~VWIbAkXW-(1Dm4#2sPGi=n5X~S0Y^X954&SyRV|E6g|Lz)B|-*9p1&eLfy8_ReIoBkvNkLEitlTvyQjQv&CtS(^uzd(#c})~kz#zu zib^K!Ve^Bd+5LA_uyu186#W*W)+O&abKM7=BJ0H7akD}?kNn9UVI-oDN|^;i^YPo; z%ILqOWIS+99jl*FMot>Zq-;VZy0ZNVG);Spb=9Lm)uV#6yIY!uYDlxM%S@==m9z-?it)L-@zT7V1ys+7)`Q;Em+m%y-*^? z6)xRSCeAT=SUzh!8hb7mht1!Nr>0yW1v~xN%rFzy2M6M`ONX$+QWIA0TmWmVXvGdB z%h5BsX*?2TDb7Awjuk6Ulhem;f=Blh8sFy623s_+`!BQ;&T=ES&RPvw4Q;?3I^IaG z-3u$FtRsK7r@}&qXCxTWc}< z>J?&DJ9pt-&ti~xQ!A`ldmp?_Lb(--eQ3xW4)%9b{aj{aXEAP0h#@1^_H6aL%UJWMB{+c*Xiso}f7S{R zaQ`8xZc1dupJj4-cg5(+<_XxN^Rm!o!AfGUm`TLiSHtan4ty`rlbfiS4F)Rj(RCdu zaxL9clu&l{i0?wQ>Wii5WB z)jI&jH80_PjhE5YHyKzHZm=<$3~aw@!#;Q;N!{)YASykVCJmHg-BsZz-P97#JzkFO z_TM9)VvYdLCwS$zBkwcXjc@Va7_hv`d?%6F>VeP+dCK8 zC}xpiWE!F-x=h6T8-Rzc2sXOTUyq|{<`78a3FRy@Gm{aCqG~J(yQ3&%S$6fdccB@nX&0yrbF{=X9ANgLS!B z^>Q`ocKj=VmHWYTk0JF`T0-Ql$D+l7H&D-xD@?iN6-M{lDx49);I^MVU>tQBX8kvv zMtWVMHFi2weo#@Q7jv4{mP8X*op7#Y<$A2XcoS!6Q_4&TuH^#cL+H!OWAx+UTv{OR zL>eO2U@ty>C%8Fg&hr?bJ_>lK>m0H1@hn2Mc(KycTO&^k6^pQ?dl@T@m zyhYDco@_LlBA|sTW9YoOqewr_#o15Ckg)+(g5a6KaA(0um>*F9d1tNZJtbFEy}}*x zFHU13H#L(g-+Gd`e+~{V+Qi;Sy~#PuXvd>3xv(=DCHP#zD7vjni{@-C6MB^%Bgg4Z z$egr_ZoeOaP5J(~f5~~yq~{DH6V!{rJduQFc!SfXP*i&MAy~dxMZG`7(~P27^rGfJ zP|UBV-U`d0x9AP*TV6|KT^9?4eD>|yVrvpxq{D;-pA{-QwsR-umXZNu5Aw|@iCA=9 z!*|C%!^geUKuTJPS?qO?D4E96QBPWV#(Ea6d|X5VSG+~W^MnuFg7n@aJvV}Co15L~>XC>UXe+)Fb&INV;8sOunj_n`WQg<~i>^0^TR0SPF zHnKO7MUJ-69|tkpP2BJ-*~4&Wdp(0{k}xr=0cGa}jCAyRW=+KjvNY$sjaYm;Bfe%f z^%&3l#?toUCL>dN>beyhqV^kAojwTK*JDU`U=b6ODmd z$or~G(`9#%b#;q{7B0-c{1yFiwE_p=N3H8ct{!fJUxQ6h=B--7KDHerJ)8Sqq~ z2Za`<&{wF$%|1Gew$B9g$oxO}bwV3p+k@!1?po9$wFMp81Kcl|t+ z3hK$Vrdw$IKR2K1)kSgcR|lNctq#SMZx8$dBk0?k*_$E2Ph2aieYnlTpam^=%dYyJT**)x`-=4#06 zQ9b)UtA8!7YL@^xNQf=ovo4SiUPn z@?z#}UZV;9Y4Lz2ZfIo0+c(1Fx6$-ebtrg$x&VEf7tpF^X9x|_#sQm~n4U34SkB}Q zI+JQmdy%4O^X#ekvgSfMCp3xP6l@S||HscXUqvA{po{EDl!nK`04A?Y2c+72(92mY zdtml^ls}iBGpT(;1AgP_N1JRk!API&*OeBn;JE{@=0!70%bjq+;2+wUl!SfvS&K@x zmcfb^8IkoZKU`j5LB`F_N13_^)UNV+3EmkvEd3IXju^|DSmp}r@H3%WMl4ESqyl#? zEryCEvb0L?0!=-=6c^v}gjq^@6x%w`xSC*`*QihZ)>^Z(vWB@JrwcH)cLg1M&~Kc--Xj&7s00W@r^}q`F_b~S?YJ`0i-@!Pv0DpCBYv9a5kS= zvU5&hTHeItbum-f73>4F=kl1{fu7Z5cIV50F`nZ%S|-_ z*jHDAr(d;&@{Oi=9(9D!8arCFL4b>sr{d&CX|UtWb13kuL%L})P*L**J9Wk3&HLw5 zsYEHZ#$+k^Z_6!EIIhcldz6JrGKOsKs`sI+!`cj)tk0gQ9?gEE-{EKHRJL-qIsP@K zm&`FaNwxHM!agZU=FdtUELU(5xY`FcufcY*(Z*I?$f9Coa47{eAlhb?Yd(5h>O#jL{UnzCrx9CMr;_)2(> zstt%n&f><)2O067fIe~EB*Jeo?FyMexLQN#zjz&Lf8NBatKX521rsqknFSL!o`j+? zfwc0kIi0aN7u5DVCYj%caq`q6>>J+6c-ZW~{ofrS?3gdsALO787h<_-{WjO~4q>P8 z4!p-vN@%?|3IA?+j3@ozyUzTqWuC4ktP~r=zMT-m4zvBZA!<3eRC{38PCeX?4TxpJ zMO1J&flaqKiY=}mL`zRDLG#aa!PEVLT;WVrND${4{>8F9({2=``8cA;&zjWU<}~P% zVSMqKfb~}NL#B)mL=N^4i<_f~4EfyjenTa`_4+B>I#S8`w(N!KIcab~Dna;T$p^%J zSsi!`2EuK!wXMb z|KKAQ`>>ML_VI;;A2k$t>(YqHqi9;t!a?6RPp zgjz2q3M!3wv(SOvxw9I**{+E*4_7muv(?EP$xyuRmM4Ak@E(fVv7B{9Ti8=Om2ghq zc}C%TGFmO_M)k9w;mM9wIH^+#((W7q0b0i7E*b*`8y*Yt11ur)WD1VDOt_%^Dy8Gr0D(H!E>(3%+VU4Nc|u zDV6`G!3%YSrqcQ3Vo5Of_0ulU{+dH|`0mX|p95&){(~^b#{d%vclO3(Y`VYU5~wYn3&T2x=*7^p^lm^8cmgr!EJCcYKohAeeH3s$IZM~)q8s8RVjF6Io#W;K0$I_@@EzH>Wy zXr_*$q6@i`O_R|E?Qh7|NtGKY4n)3FH{r8Wx7e&bBTl5%4uPHAYl1vexN;_X z$9yX^t4ENbQ!P*lzaVIvKc8$_V?xx`GC1D92OA6G$+(h@Xz`{Z=0CC5WJyTAO}y?I z>(~981YW1sHd*&}K%=iTD_I}GeXr63Rd*Botv`x9d~}wedigFJxx<& zZ*u!x?vP33H@a!5gIDsMyp?~7QSnww{Atb<$Qek1*19{~9-bTd%6|bj^RtlAHMz_U zYUIG)wki~^y^hY<&`CBJn4^{BW})?~-AMO#M`CTR0+!p4LD}wT9CqKHZm;!+xqYSR zeV!_pQ!N3PUU#DPd5@`7?Ik!Ldxd2EJC0Pg{zD(fE#Q=2!Y}$ za7DqLl-%w`2NRByk(QOn`KmLkwz`KjUJXJWdxILwdj^?UJc${yaz{@dNOEsK%aAQM zEd>^nXvDj99q2-yRosCnATUQ?O9(20R@oBByVM!izhyI3iky@wpmD z-GWV#g^@MZJ$cUN)Z}-9tdhxsefFb8ZBid-%#LBWz|F;BJJYb`3|W%>NdX+}mDpjP zZ#?>$22MCOhS`7XICJC35fZR>75V>lvvy)SJb5t-8gf!#%vqL7d~igKk@HyDECYTf z6onM*WjMKR#TlGaWdrp4@a!0Xmw3A(G!g{S;WaA&2U*#TlY%%J=zsLcb&x1 za>{Jd_7YTR77I1kPH%Gamm%-d{xLV~jx&W-hH%S%Kl!;+Rq)|+I1_mPuVB`$F-=EV z6@E880BYV_(cCACgbvO%C^mN-S#^Ww6J|a@dB0m3!>n-hbn{Bo+7rR$gkL}%mDxn0 zz#AQ5{H@b+%1H%FIx9%ot zv9Xe7uU`!3N{y(MbsL<|l%_8%vdE2H-S}2i4lt`4s9#$L8d;r3l|8~~Zq-@REf$FU z#i!tgTmdP!&w=z;iC~>qNMDM+(32D5X#PKbuBR`8Sf?1_DzQcMVRt^=c=j)KKmG^J z_~?YIH?QM6poO$hRbI4UouX(U>JXh=D1^5o>hSpeVra}-P1D}3r8h&b(~fyJsHE>L znlZ@HA9gAt^E6}XpKnHw@0X+pj4~iK%p8p?k_3x82K2g9D@}WuPBWcu(tSC4)ak%9 zs;u;Z5hW#)DENc&-wq0wO-Z1W)yp75LI%#U-@w%`hOVpfAlW~!(seFvaPn<580c!D z4`-jiw7g}UU9ZRV_K-rxwmA~jM4PxiR(zA zbqQnV&r@dPT;O)CA)RWqg1+z2qH|6kf;GDB)F*nD$Z*#m`uo6YYHKVndhD=;w*UJG zR<^l#%wQ)j?N+2|Pp4CZ2~%lBKn(Aa7SZ?JzVvKyIu)Dq5mJL|klKz;D7pIfneJl*0d`_lL_J-ZILgwg1Q@EN`gSQ_nlN>?H71rX)iUg9ibSoLX-vM;bNzgi%)pQe6O%1fwSdWQ5c62{u`EO9r;-=hROeuyCP6Q3!H_2%s6JHotzzi@p- zp89ypiS)>AQ1N~Z(knG#%F)d>AB|se5PuKv6r91!UE0yA2g78Jmlvt8i$-H??8x#{ zbEsU;DdK>)Oq?i8q%_yR{zchxr89liB1zXl&V9G6ho%) zKE%?}GsHOL56^pwr3XSb0vDS?mM`@MgWfW@d2*7VYT{lJy~~I=ua*#H8+*_lyIIJ4 ze;an+cuKx_c5qQeEJU8a3|AJYVld34w=Mp0v6lWW7B24n)jq;aZ*l?xgW1TzyX^nH1DmW zP{-^m`k{RVxt?4`OjTk~Y1d}*CiOAi|M)g}>1#v+OspW8e+G++j1unOw~1}_QlL#& zt*PNFewL7y%GG?INwztAgJo|iqbHnz+aIX0HnktQrwcQnJTehim>d*Juh|V^4W_WB zRvJz`(WL`$209Y7&<58WlJe~v`FUn4m!8|jjMrsoJhPf~uDS^CZ@)&Hk0&uV?c~_q z(;mPYbtSU+&=ojOtDx3m1}KgDA>7xl%Wm}!v2pKNg2jx+Acy&rQOMyI;(jrO^!d)iCpII= z$g^+*$?g3Qhh2zec8%~{<^FKg_DzauOKIVEtrW-PiX%FU}K}MS(Rj5WI3a zt9LAcnRM&9LCI|Hq9X^Mb`mzpRx5kYd`P{c)$0$$?*Z{d-LIv8JG|+`ZN~G}G zB=#%Yik|Vlm&z12+w&Cga`lIP_9)h` z`GqZi{t@_82E&JrMcy34_`$kc<1= z0Fe?#FIa)tj#lOSG3S`m8U5U4Ekn@yeizN~$$@c5mA$-tBFcL*8gJx3vw~=Is>}|2 zc)**u|9HuCxCG#uw+GmP6$Z>DQ*qwQ91WXV7D0S^EXaPfMRQJ93R71KnUKAIIZd(H zrcV(Q1+80#&?de^xWZNi_ca_q&b*gntk9TdUVn+QHlAR@X4|lRgJvX*&L$Qm(lFNv zW0I(eGMbkQ`W4y*H{1^}CDHqtF+CYv==NhoIqEA}^w*VG4(AJ{x<^6!pA#@Nr;k%@ zDCAy+{|4H37NbkEA<4g=9Ps=ooO3c0uXyba8Y*Q>!TWro-2afQtnfg^o7Hgsp*%eJ zG9H~!+XM-{Pl?3+TqMY>!OM1xhIK7l$=C1dY+-Z(N;2x;n!Y+B=8Ft_%R`p)uy5e( z=13D(xQ9F&6@-g)auD<1MEYd^LY&)_25T=G;BuE5oZhg9V6qBvi`DSE^h|X6+*#z^ zy%3x9UFMP(YLMi0C-D+Ao%_}Af*$m4$FEvyNpPb`=sghxap}!W*)esd`lmeFdZZA4 z9#q4kC>#9LYa`maIu;^3mDyUS!}zknEHt}zHY^!XA`M-~;Ps7voM}rTiooBAuFFUng_Rie!%z>ci?i13oUvi3GOD(h+OL*_~EaFKh74g z?-H!JxRhQJy^?1;2EQcg8?Es9M*{@Ler4Y_diE!m&f5RHs$0Ik>)g`7*b{U8?I%#&_huN=%r*4om<)ld(dy(|9c-7c=1s+cxDETd}bPr%_=7+&`64Y(Cwg`I*SsI}dR>W=)PJ5`R*hidLLLQI;D zl&q)I7nCxc?%xDUnMXXEt5mQo{~SD4P3D@HeMal%jsi~|No2i~Lt`ILp{`FSkj7dI zHaJNa_KB<0kR|dWYw-%)TA|LenrYB#h-r+D0=PtJA+MQjU=^4R0ck9-VJ~TY{6!!I zN}}u6UgFA73cj!7aJIr5fU*p@-{MV&qr$-}J_0&J)j&ty870yQWJTLH!Nm4<`VHJf z8mx&ZU4js?zyKUzUJNUQM$jzqfx0P;D9Cv#ZCDjg>8K!BLM*A^q8*jH@SViHJ%o`= z7YV#qLnIdUqGW|)V!T~nD4+5chPOW=;fwmos{4{yGi54CZJN#yw;A;1b{YDbQ)Xhd z=F=r+tLc~#bs9e$itLi@;lflYRPS|-?{k~bk$=y*A1y3BvcH3zo_7kFgwCQz^d-=N zu8DmA8^dR%>Gbop`^3;}CzJ5-hhTxT5~uw$5U3A=iyqr)T;>vP#o-nB)z-;aX>$)O zYAhys`QyNRlmaTdr-J*Cl~mA6fDTCqc^wi zqZM=RFjWEv)VuN`2~vK+ENj>Z=SMyhd-<1)=tCK@G7q9!7n*6{FyPE}kI5RN&ERi( z0Gbx35|73*a-?G^9d-XJjppxRPbIY2T`Q%~i5nKAzF7i#9J-Nq{7=|9_9fgrehGW* zy9nbWa(D-{8ZFZ4qu;JbissC>5k0>7jNZyNqjJmJ)W8+k z%b|vtoogq{J*PK_&7aly>E0wbxYxzz`u1B~hV*Wz{3K26C7tm5?MIn}>oGWVx(@rU zri?VS-2j2TBkg!FLfsdnK}q;`#AZH*SAWbn^vQ{-E!BaM+afei`4I?;qVc6=GVJaD zlJPP6k6RgSCX~CJC8!BdK+-nFc+HF!7_+&AN_#reEo)uT*8?M5z|AAfl-OvjKRk{$ zi&fDRzlP}P#!ci{zibszzrVe~y3ex(KTeyRe_m*0L6Q?t=3U{`i~sM(Fq;!KRwJ z(cKaGRJD<3GiF)BaNm49_oV8uyKqla-!<+0@fuw`tU_m-M$uK~ zbMX{iMUhxpH}l(8o>Tbp96Kb~qJL87z{Pk2`*+I`+!4GNANhEQbz;`DKdXf&Fyuy>!f7Z=TUV4sAY7G~v7aeB&zr42zynN7Rfz}SDr(rysqmYaBojF$Z>Q1^y zKNQ(ZjYUOSGZ-8_1mpduQH9eEWXf`N$SQk`&)si_AtxQQY36C>-t8mcvS|c;Di~!u zfA43U^CAxoxaJdua2Gm0Vic{qQpnDdHKH}1s_g5{mBRYkqu^qwgkx%Uvo5<7@&CPg zzNK-f+1?K&`R0+6DG%_Av7?y*BLy}yNsO9DYhXE*T<*+sBNT4U!YRvOM!xJYel|Lm z`7+KJ>J&Y=2eWPqaB~T>+VZ%-z&M9hX+s1s__!Nbm?uX>IR}lMO3bBk-W%G9Y z#S$jd@g0w?%(xA_gU!htUM@^QXLx>StDz-M^}k6xmi5qgZ_8=zW;dJ9i!Y%xKFg7L zCzI>HF$u-#D+zYKm;^OVZ{YgX$q*i+FGx5aPk!%~f`zBOKt{|7oc=k{Exl#vP;eNT zpqUB(@*W&4cA^0W51c4$}Sl@F?2|-RnPz1|4KLyMr&8`sE&U!1WVE1U)4S z{$1pPrH3H(>m*_num@T0*oHU%8ciRkH{Qka38@k-1VW(Ut9t>zl0XLMYHW!=dqX0{KTap zJF#))pp8)9j1J#>hIiPMVoz5o);0M(ne)Jp7$5vWZa8d(Z~T1IW?DCLvzWswT^_{4 z4Qtth3wCjiz0KST6F(T7vPJ=?;GfQ2~QTX3gcxD6t z+xR!)eP1?1v3nM>S$R#cLjM|3_-_%H7a7UAIvr%wvQv@!`j_yZYX(Fui^SUUfz--- z6nC;(5t@G8ftO$J^KOb4#CS$3+;BV$52q)R=RrZFM9dAz`xdf)zQ(Y}9U9n6z5U#) z31-O0F%r)99l;BKC1G99Be3Pr3iSE72ji4Jfn0jpgY71KA>(b9Fe+R3p|{fr8Z}K8 zHvM;&Q#S2E`TNh2L`N~vvg?V|&*ds3 zah^N0P7&l@3a|xH$9}8N&IVp6oA}vaP8JSuJhrXN1PCZzFHwO4nYu#i#eH(vTJd%U!majox zF9Fu}EF={RCFzRb*SKK)Qaa(n1MIm>3}3X3!{w8zQI@g@pI?;*X^B$87C&uNSdFpP zMsGZ7;}qO_rHlOZe}ONq--9Z7_B~$f3yWs(oma9H>{K`ySgB9!y(=Lm*qb|IsK@o?<&om$TX2tu7EQJP3O&h_u$EaAGLXN`X|-ojuiGkAp_8A5n(JUMn+agn z^%u~*R6(`$c=WkRi}_bsj#o`J!Yhp@AT8so0{g(X$mi%VDSxg;Ep;_08MguVj(bS_ zZYaRoTxF!;Yf87zx^9gj7jWC{KVT_7fsXY_LRu;TTyB67KKP*!Uzc9WlunxorSHmpsb-)$0cP8_~4BKS0>klz0vMduxm@UvyvzcT87h+O!)wIhtLbQ%1q7d1-($DPrAci(8<)8UEHt$p1$oN`^Th--w9`E zZiOUn%Tt7To$J|jy92Cw(QP}im(qZHyQx;`Fo<}li?6TtB3h1!A2%eYY~Ik#+}k}$Xi1F* zYI0lvGh+|a#FLGp-p?K){a_`g{e3dK(XtsXytZR^&`aE~!UhI!3}HG_bII`Ha*id> zCsi@V#8|6^b5&Gj(_F`+OFQDwc8lk9FkYTacsPdqeJUb<9vIM^A5QdW&;j^)>K_RZ z8wtF@co10#jLD-DX?aN##A__4zrFX16bk3liD@UuzL|SyJSWGr+lFBG4+T&bZUz|( zqDfP|4tkPtnaGA;gdwE|S-F`cPH?J*-<@+|`{{6Er`kdT+;os<%sMu~{Rn$~I)Uw4 zc^YE&=F$+XB<{BkWx8+Lng6O`bW3nInfXKs^y;hOZQWG%`;995dTtN>&ga6sdIi?6 z%lMu!VRjbn3KkDfp@&KNoXMS7^fhY@ULovAcq!Xbg-C09P_G#O=b=HBe7n%hEw_ld z;ZbhmENfJ?ShQE4%?T;SQ>{>mHg5#pOeH!S8&I7{@kF#X)P z8IP^CM(JhgT&=4aCue^NZNGMj-O{|o{O@Jpkxo1Ddu@BjKRaB=jj+gfAOa;^ilcLn z_EDAgA?WEuKe|BeA+kYd=+ifiq&3-*1^YBG?t#F#d{e+ystu{=)0YHE%^VoF@+G<& z>;P_;UQsi>P56*RD*VXNV%=9#X=P&_ZJJ&|+O6WDL0bmbP7=Y!{Ctv?I3MN(*nxJ3 zB^@$pGXM8m9s4mvo8>zO@JeGB_k zX;#c#DjGJP-V3tC?^K_Vp_A5fD_vI8F+>g*w_1}7$5U|L#m~}s*%0ckIGqMM$55qd zI_T_lcka8)Vx~WsNTUbSxmmeEtg8D7eLue&O%%AL%Zj>4-`}x>bt~f+-OY4%aSv_s zxqy0)^l|gPsW8Voy0p(H1P|7`!TW>z(S#}6SP5JN3w=Z{ut@6gSc&3YIW+aiCp4rq ziadI8oc`#rCow&Z!fsX}G>Zpm)v93D>+DP|rfgs}D-BpWk*JU#wvWb$4#B$jr&;H% z1ZwG=g{=zRK}Jc5ewkUv8RuVtq;xfQ(=rT2&TECr`eb^Jc|eE#P1QrNP)kX^*t(!>d_GjP;8t3uEPw(m7>6>V+ln&UM<&l#6o_w&qn7uxh z&SZ*9$g)SJC`+_~Mtg)XO@S+v{3skf_<4nw82wZnem8~8th+#xhHBwmRWU?a^pSSOzoys59Y=kOtYGm3P2By4fY011bm^8pZtU%&XyLXwWcN#VzER4QJABK9 zU-}f&_jOUEzF&(S)f@t{O7FSOKSxp4ZVB<)lkQ}hauk2zc^Ub7-~uW*xd#2{7JOUz zMM!+n4c*zkm}zKihPuE^QaX7w+vM&|%RU@}iW6ltpnW50kE`MPePvmAO#=GU)knse zgpv^Dbh3W$FM3dI2s<@y64hL1M?79=vjd-BLDRD#;1;HZFMcg1o$+cg;c6o3dSFId zzxD9*j4$#2@)2U&Gbd?SOby>L*hs1eR7uw5E>X&yE>c?L4OJ$8sp6&_^!QLKCtl&t z-O1SqM}JMhsZ$N<#y6Qb!Y!74lDPm*_GwTz+Zz&QrVC#74ysb^KwXA!7G3`kL02mK zq2FUO>D-wSwAIK1eHu}R){py6zFe+_stdu8o|yv0@{I@=b}tV2a4l@`ZdqjL_O zMq}!8pw_|*D(9Ugucl^$sD3j2eqxreAKwTG@qH-Ktr>=_iy_-yjNwBi>qyEUGgz;o zjBKR;gApGWSjISJiQFv}k%Yey^>3O6b&t=ZHEkaB@d6RF6d*2sp0P}uIn>``9!s{Zq$hemQTGBp3KM+M-W?t! z>Y^o4@--$09v$N^MlZn&U+*RzM|sq^^%6OLYzztPHm9L=`>^_ISEdzjPS4sMz)Nl7 z*$17qtYY~Q@=eK!4djMU4}F3UOishb*LKmVr8Ds-owGu3V+b6Zv_owFmZ4e9lBRpQ z<1od)FwyKf$hBR^W-Z6zt+NcaAF!dSWiL=@gG{-KX&(fnOvOu9Phsn~9-(KP>gn%E zJU@G*445AF;mo=fnb*go>}^sc9p@#*p3iW=IQ<=6I{YqMY^RI+Bs%D}S!FE9F$Hv1 zq|qN{{V3S=2v<&?qdzW6G|_J;JagJf_U?T^3+`?sxu0>lM($$*DrpeC(uqa;NTRe5 zOZs4HDcr7@M3-ln^2QqeQPua2{^?(K6FgV(Tz$)RBa{Y94@!&|TtZ2cgd=4B%C8EEoo_ zrsp=kLsyiX@xh&!AyvH;hVK0aBC~#ekrRif5FQWl7vV*@r}43pQhduV3XMtjL=Ahk zqiVrjVteKd&!zi|?j5fb_+`(L+}08rI=6{ejEE8EAN?W{37pWIo8?fHX_EL~R0&yC z5d~kH4EVA4<9V&=;V|pbK6KIK4{U8!#U<}#O88a3wntKBWc+9p^ z)Lw)$EY9Hj+lP_xO%7;obqw}=W`IZiH9_UsHQb3E>gCP-wWLo**3$3fZ0_B6YvykK zlJ5N?NyYQMU}4ifc)ari4gAwep8T_i-mqSX*!>Z0j`{_sO0+?hZz(tMEkVH!%V_i0 zO?Y$5Wn721!1VZ;SVx%UE4@~R`=xTwZQ+EEVqF~WSwR;DeWa!Jf&A0O-Y7xiEPucM zrFi<|RG3_|2A|CuYBj@QBWSia!p?K~_-dpawstVZADpD=NbP56HMs|cHM_x~C9zJ)24OUZdORWlstWjEvJ4msGqOAgxPP@XcG@n>U3sYnb5-R9?%g$btGgV=w%B3czpA)0!WgpyCvfVy%Ow5$d~vq7 zEGMmZ4R5<_4kxFrfE(Yf@y!{=ILcWTXI>bJc*hgy;_rAa-(M1)9HWVLHZpK1T#c8M z9L9Qc)bOFAMtptdPHeR*rUr^42M?NIDQ) zEJ9Izsw6&=n~crM-oTVO2z+Ipi-!Gn1fR40?+bAVGp-+&&{Wl&L zSqN^3yZcd!NCJ=V{tW9iqwwla!ru$KjOP^RAm(oh&)=!x$2T46etSuD&U%2XC=C)a z_F1Us)&U$H7>Q3lnu7nD+LOM7+4!L7K8zIRey_fo(W&lIFw(o1{#o#e+;@9P@BbG< zR~KmFr)_F@vTGWitMUM8rRLxa7Zo~tcr4h5Pr)DUE(M4npTMW^dIwb>vL(F6%Q*)9x&vO33d%Q#mi>Q5ZFN$_yykv z(d}U%S@(wX%REF%R~eG5r=w84|6vH){ukO+N5LTxhDD?7QEJ+1ew0BXK6H=A5{ikq zaH-%Yey@jh^$+2n3x>k#vER|gRcG1lC1+vFkvkN5r_xhGx46NGhwnjIe9c7z9O+sG z#jk~*(^!x6ldJfxZ}QQz&4w5{w!zZCB~a--3kwAX*m~E3NGT(%=554Vbetpl6SLuD zK^XV_;#JuCED?262@-wjwV*H+ILU#D@Hc?Mev}Vtby09RzY*I13!>brCZQK~4Xr7U zCf3RENbCDoH0RzS{^ltTex*;QzZMu_^z9m-l+ysg#X1n=nvFJ<*kMEK5Yk`fNiuuF z$eC$#xht1jiAU2EB)Rbt=^rqquP!K3|^A)+bA{`qOH% z`D;C~za@jm?#o1dDV;QSRVEoGp@#b&Yy+)#1yCiYP4}GCqvH)egWaM_z-EQuf(OcA zBo~ie>_hR>$0F>WeFBn)CZSM^9{9k8;N!!>@s@Nc+BvI^h=z|v&Po=@TIe^<4pQdl z9d1UiT}GfKGG<`DSQ&Y)xI(>J+aOTrp$~@SjR#X3hs?JSePG9EJo_bvWd^1*}Kaql|--*kn&BT$3cYf^uug z<~b&Kn^XyyO2lH@hugv1!3VB9Y=+P;l6d7VH59w%H(KyMAD$MZ^%P!tfzu69!Z0#h!mz;;R&|5_P_I_9pe-D04u0r}<9KE+k7i{D5p~CY5&$mv+ zW$%RS?)+PXF1wmLyL<8*vev`8##d;Datqj4#gPN^#<7=PJ=Ac@KDz(?Xn3HwhK-%< zgl6YU;%OFR1n290Vx902PSz3l>^z*!H+n^WcwC`>^H!rz$O}E*cZkaa19(3&4CJjm z*|Ndgu-;Y^Dc>D{vrU6^d2bR(^c;gJ%_sOds?tIi8FE!n<1>wchSV0L+SS!mDKnb3r*DEdw__x$_$gP`8b-5HtLUlTiR@dD zB&=$!K{HPOB%ixN`3)2GX+p~!x@Dmx%@QY4HSUn`Hnhb>tEKVQ;BC0;V>tH7+KuIR z#N&Z67}u-M#;2`~kpMBK@j2o2g4KMQzr>Z8T5Cb^0~OL5Gzgj(P=Suw>MVbpzeCy5f>BV4E8Og?N% zzqL)pg7pI*UT;{wNIi};?5q^+8nc0kdfw4+-4r@o$Ok3M&1i4jLA?EBDwUKNiH(m0 zVapM3T+LK2G`q^aC}7aC>!p0dxcVY-Sxn?JPx z`wq;s3K22_r@x=*tWgH+-1%#C*0Zwm4Er-sJv5gFD%a7^1C#0X?=dJau@xiVHLPW? zJ>|>yQ#;`vkaaYcDqqsT1yV0Sed;$bu*(ItGw0#n^(*wH0Hxg9!yB|4kWv5S>bIXr883tsE^eJdKB4Os0R( zNH$^_PaZVPM$uuaIB}8^?0@?aIUIX|jh_C%+GoeJUz&Q%+2%gZe4;;-_{$RPQqW6etC7ws`H_*Qo|G}d}&&U$@7CNF7PTOZ@!K!ZlO1yO=6yX_lab4e%xdWUJsaq_VS-C6 z8pzurCERbD52UDJC_b$-6vSp-bh`Nrt0M(FKyk7h%&PxMx)&&-#kwi@#*V*mWs5Df zKd}i{%$K7MK`W6#?MrA8`bf7LzSGdqV|13f9`~oOm`rpF!G>plP|b-PEjSm8SLr2^ zY4gLt;*%6pvriM|AxgsiDg=MlcVXV!86ETSJ~!Q?g+JrrjwPPfWB;iS$%BGN^!}>_ z%yH5cx@2=Q?cNhYl=Z~qVDk;Uc*_u+`|&Wf)_aHgkH@{3*eI5;8-9={#K3AXF zXPNomVz%t(0Q~)M9OM?ZW3x3PX0)Xj`TlzdD+|q;aZWNGysW{DTh%EqoPl%b6$ll2)u%nZ|(U0>KaM;aZwC+a)QQENwOhXr;veE$FDa#+c1`5%4FIm`l zR*l}d*-x#Ty6B>jmGs9A39RvMI92aTrv}es@h1KZ_I@YA<1F1+{d7h8EMs^I)-$4TYE@1!g&7^#Tz(HG0Zkh@(D2S`_-<*{){ z`piz6_Wn9bpE(>~YI}fEQxfpaTM1ZZw-%Ni{+jO0+DMGU-oX39D{!5{6R0rH~1URjBqYiIF4JZ6Ki_D8>3*TFF16v?23wjVS(8 zKOgu+9^MWPBHQ9=SYJUGh5y?Ge774;Nc&GECTD^4_s#5Y5H-L z1Rm}+NS8bo+!4V#cxFAt628%NbNxp$VeB3O)Eo|Ys?g~v&gGq=REWMi1Kn*?@%cOb zpfGVASPB3C7fn`UhrK@VbDEdn=jrb3x9<&Zbfl3GMp%A4@EiD?r^wj3hxTW#L3L-$KyqI` zIH(zLX+;?zKM+R41XlP@fmMxEPSB>n92DU^9(&vPfLX2*Zg&_9i4q>5tQ!qsfyuC^ zBMBv}FcD|{EQixcVfg&1?=Uy$1isQF68yWS_|y|cEE`n@PK&mnCePCl(Ibn~%C))6 z*1KWxlx?VYzK@WlM8Jv{Q6lSi1CVj}Dx8$-M`I7{Lc2DtC4uJSxoY<|>RoV;m<;j7 zW&b?s@@y-lGj23Jz1su(yFTHr%$|tS<{7~KPCFcbG6@Gg^@megEqv|L)%Zk)Ja+JM zA>%ja@o`lKxb~+co=zA&Q=!08f*q06fx~EGZ9YV{JMwqr9bjc7(v*}R=IkQ{@H^smCftNW1OLoq79y^+q)o($P>uOT+~AB<0k z1)1yTxn;#Z+}H{K;r6AblnFeNcR%NlsU{k1^%)NsHO85G1-lZLe>N~tx|h$+ETq47 z?WZvY$7$%BZoF>pSlrZZEzJLBU~*T}N@3b_+`oMc{q#!)Yww?rj}_U&jJzb6xU?7b z-+Kv1j-P_%Z6n~RLnZz_!xL{@5lI^3HNelm1ib&B9eOChlY@~YR+z<)+O3L6X&-pb zpNYLqHj!t(CQ$lO10H>`;jX(>iM<@Aq9>{O+)mTws8LM_g%PsrLtIL1T z-6`cX^l}u}eprI6C%543uP;R7$`nXdDuplhZV;(ciuArdf}y{z!s-8BQG*L+?C6~D zbbn1d+zojFW64^w*X}HhFyBNxKA55DM<>y$TMOx{!+X)44Ia=oItIxU7QjJ2C-y3N zHvK%LQe2rO!z%J)n4X*w+kT*w1_?~wxJ8M;R%X&)`MMx`B|?1X%2?LCCX4#TjOO0n zE}`yk!*Ir?I665v9m2Zx@e+R(EcToNtqGPab%h^)k*Jo#>FObc9 z=b+%x&RqVK-R$)(ferT}jxO3ggdJa`IdcH{oz1bDMOD1Z)muDJEgedHims zvrC^i-poN{-6HlgSd$(9DZF(Br}C*hUzmOKGri`n$aX!HVvdShRAH1XPH?oxh1z-4 z_fQuJ@3o;*yQecB=Sp^F;!AdWTO7?_UQJ!~61dAxQ|O*lEqdbTH^E`JiiBVB#AORD zaCksCouoU1yziSTv@ivwVvuLk7C}j zH|dFxo$zh8;5s*L5&!+ZftG(h3;Xx3BBn_pP%+;O^6N&Rc?K!y`ki$2OLT(Hj+ch} zGvA?%dH0Y-U@5aayPy0$uS$!P&X6Fh`E1C>qm(>SVi)ps1dqKpeIo#`HfUE<`*BK; zrRWdEnJZ9UTs)nyn$c?~GpB9~>4Y4S5^xSFcZ-}anO+a8a=`y}(= zroJXy8|^_=K2*Y}t|M&PtX0faGlK4aJq0D#_LJxzGvR8LJT$kBWY3gi=xK9x2+lG` zWykK(T{{-Bd0~d+(7<%~+jOb?(a|RUM@_Ogt>+TnE}aND5}#?AbSC{RID^$j<-qX3 z%}oE!dy*KKNv-5mnC^u2FnPWMt$wtemJS#0IYCWi*56}@yqt_p@7th%oge8R3w>(b zJdN%#7d*Cmqv7+d6XbUj1~p^r@(C;eZ0_YCEP3MhxSv& zeCd@_^!{jdw9Wf5=i(8``=|J!@ekr)-u(<(G5H>s_)`kqX(00 z-XZEJ6Nl!Xd&Rrp(d5AJ(`3kT6ZUi8Y!)wh7#{grvUSfy^liy9G&M#M?idN(o@{e2 zH?NdCf3^foY^|a#^|$C}`WU?ncnsS{V><7l9kouBr>kZJmVWv{ zN>|Du=jm{ZUP2p2vwpdk472+!m@FIhO9048bG3jA`YpB<||+ zZW8$BIag_G zatx+R#}|oAn#y2Nvl^t?I>O$P7)a_<#0fLy`BJTss6Dj<&c6=9?(>i1=MRtJjib(R zIXXOVm^y^KvGd|8HU9CnU1j|45HnP*-ANv3XwmYUpHTkjKK_O<7g@6SBLDd_5XYW& z5IMPnO;VZQ-BG9Rx$^)PP%fEE72S&6A$T+Pz{jJ>AQSpn6#h_{LvJ3%q$}RT;&xBG z=GhC7zFTYAko=&0!qpVI(C;c2zu_(VIBq{A`CMY&6IU@T`G;itJH-CMsJvs^MzDB^1>}slEvYp+$ZOk_PbD)*-k)q@4evtVa z@1v`47JF^#N>41yS;^TJ)Zn*g_rykUuz<51=1vw?{v+Fq zO=yy1D4jd)1?Q*Y4Y3#6X!(s^y3%|LIdl z#_gF-bW0-md?AmQyWxu+f4Y#RCtr{cTdjfiBqGQ6s!X|R3HjKbM!H)E(Ct|s^pp8U z<|Z=@kBHG?`;{Y@c;_9#)As>ir8FK! zYdYcALl<*HcPi5BTGwgNrY2ZxUJFAK4_Uc{jDSYVskB+R%lSMSDmqtihvfeZ$IsS1 z63s~|A-4Lj&G&1c=M1LDq5iGs;FIYi_D|d9xGa;V~K9cgz-mlppiCfPRWi z=O!H~=MLLSQEMbY@AVFY#A|;M&T2-IP0P{Vb;)GSP9DvCt;+k|4@F6$(P)p?jq z$z9e=OwXlrdo$KS;V2_qXmJ$HaSP*Pj{XOGdPjiFt{D3Ak2W4}9>SB~FIn*4u&D6a@PPhNz6hWe1585_9gUxJC> zuRwBU;FT!vSrhrAZjA0HIl$|L(U7z%30Fyvwz4)LxM$BNy6$c!$ae1$*acgO`u?YI zXZ1IjF`*E@XxYvGCzFTF`qschtuWMf;{$iT@f@c zG7R3rO)lE>ia1BKnY(oP9$H(p5K8R&!Qt0p7^p87Z(rtw6eC99z$J^oCsK;;4Rj^a z12e>?H%%efc^<5sEG2$_PK*E8>%tqFxp3Y2(PUgkIv=}An$rsir7x5PR`Wd_P+6Wu z6?B2U7Pw~ib9U1uJ<_z;VIrEm$AbQhj-@_kpHN#yD!TQd34~mVA64QgmOfjK79KYu zA$@URpOVD?sLvK32x%)T@sp%BN1~BYYCgOfG{AjV#&gRP2H}47UrHv&@}u&ObJMkB zh(kywzq?LweyfP-xWWlYqoS7Y{38Ji46ks=|67e=Rhxv_NE;;$)^Ae3pJjnsZXQu4g%1iHO=5$lr36J39h z%MIFCmCx$CNLP9>PO~ox*L?60zy0ltTvheSA=@A}!mfj+bA!Y{Aay05PoOV-17PCg z*RVzEIJjnn;7$Ef^j>}vUU(xDclH(Wl_VDOb>#>VXY*6HjK^2sN{dvC^x@Uea)|b{ zAr_{a(6?_lsX@kBT>YaFU+k2|nynY{_^>7Dh{R^3H4sftP<>YZxso$C3K8~{-&4K0 zhlI0gG|p}qK%e$T(>K)xka@fuy}c(RFj+Ta{ojdrLtqEQl`7&gMQgY{x@%Xe#S{8tqB=Mc7Us^y*g@{cHA^wmF2A z$A)b}%ba%M@9NXh!im;6bj)#V({~=WlVlwJOPZeRSU_0JPgo-C@f>N*6*qnyKznoy zdA48@Nonhai$hs^>hNEOoiP(B!gF!6t;Spj=oRbjg(Hh z;ggSV!{7EpFnQEW_Uv>SnYTB9U({j)=f2Nlf0NDOl-U$E_W3^e_I#JfWr_g_Nc=?3 zX?an-fpu_8SDWlhxIy~jo^$8AYDk$zD>6L0m%g{IM%$CyU_y!_3b^=~oL+cXSWtXL zu--DZ{>paN6S@EgnTpY3?MPZb^$qFipNrR+Tqn8eJ>-bRQF^B4I*l*AM@_1>v0G>c zyqE4md;G4Wi#74cOK|VCy^bt@Qnrv6-*!ZaBP2-TqDU%v!v_u@Y=$D&jj-~oKg`Y& zx|^I7XB&1Eb-uRa0v4#CG2=7f;P?RKQ62&5C;Q3xWh2O>h{e3~e>~D&SH??!NgxxA zqlx?nFPeI71+sX$mG#s{P+5r_l=vx~zw$~I$K?VTX{?|YVLB|@)tdI|g^*EhJ7MF* z3DkU=;HkWKkf>Oy3jTxb;IPC3dui0rxCiUtiG(}4zH>hNoRUtzbaqnHqTy5~_X=%F zAAtXaUZ}-A#_cIo<;zohQJmU0WZ9cf_P=w*vo}cKw2vpiDP|CrypzCkV^gW4+HxW> zV!$%pJDqb|Rs_e)jPMi3Y_Rh;!e&*g@Ve@q)bqFm+WmSHE#CW$S3Z|d=bG4p#ggN6 z!`D!NP5A`(27z_N>H4lm0rC9X%9`WSuXu zSRw&Q&)jI^uyCjiii1^+Y9uhDmC#i~A7B&8s0ky7O?g zS3dbsY|U!MKM>tDd&OxgsZ)vDlVQ!5zhvFTU~+6$GV)a~#(CF{z`XC2?|Fjp{xf$$ zJx-6TRGP@7pZp}z3c}9szTs@dMiV@A{tGm9yFB^)O&v$&3A^WB*PtZ49hUw(LVJ9M zQ6nMy;fFJtz1xUx{~!&oO2fHL@61WkayO7jehBtyiQJb6p)@+Ukv>k^0@X^xXj5tw zOHOvGcy;zK-8C(Z`%n5BpHse&+PUqfo&A#3&nSgm3cN`prZ~~OUjfwi@!^_?7A(!_Ky8?d@t~^|0C~vWoUB7W4bZ!Ga0?!j@}PdVnPv*9K%{{?w6T( zk?kU+He)IM(cFd_HuTeP1=g(g#5VTn@@j5O6=q$UaqO0zD%2h4pw6iY%&*_3q@)6j zUE;A`??OIfIYmeJ+tZKFH<3%(qiA7aGwm>6PGz<}g!V)GIQz-zmQUV4MVOCN=>#~c-o&C0rO@FOJBblX!>eT$^2=wO;$p0C65a4_QgSp+YqW}q7 z<>a@(fl1!7L=^b1+_H3cHoZ79fiH{+Wy_5ZS0IZ!mcLtPqv>%be3Qm_HYQt_(buP` zZ=^i@i#MPy+c%T@$8FiAWsRJY#58)($C(f8=n&DgolNty1gA58I7~`Yql4%6k=8%= zsaf-UruVglQ+TIK?|2EktAi$>^Gur_ysA^-_am(NGdZ{Kg?R~E|R8kq1?0K zcd15&NMPtap}UK(QQzt6?ETqH)N@gpZF@5Tdzr7mn?JpT7(HETyyra{H=~HI@>#$S zX>cZ9q4E{T`W!i@zL}Y>p3Y`>7}2!~GazJK8jHPPhKe&2Xw1vu73UP<1V(WuJUXrq z*Tna@xe*V@+PA`U!!Mg#_up)Y_h=yew#P*L?>C#~GOS{TkRgmnQ)Skw@$8&glx1VU zY;@aJQheZShghtC7`;Bclx|X9!u(f@#ecd>p!>NKD_CY$kudr%ow>>a#aM=sO!tK% zUHNe8+Io)WDPJH)Y2~zGeFEIb3Z&oiy7>O#JBXF=KDc%2A)WB#Fni&zf$z*5RbKlp zn)TcGMBrpwA(Ef4aViLkLQoDh@UTsN?#HkxNpfqjh>;w$y1r!=aCgt zf(*r$lO7O*TZ`H15xaQrsUnfFSpfAc&?)yCUBd-hIMFjMsVMH1IzCys6#uv8E{8+4 zMGeM789S&z)5pc2(|TI?L_{gP>x==PUxjqfi!*H4+kN1!)(;7JxCc#P%7^5v_hp-Sx&qT4SN ztjbgDvBt1U=KK1NZ$hvDZLhj?wHaBz~Luxnd2vVM?A)|M<{Q%zivlzujf@uXx4KhlRf3cbJmAp{k#U^?Z;b=k68iV zvWG!jiVZg^ESyZb8_7*-Qbn!?CD6ID78y*qh%+|q#;?@;$kshqAU*jN-1^W9j_Dtu z>_Duz-gYG{b(6vd!@aoQml|=i$vXU0Udd{!=s1~ElSRhdlqI3=x!6$m=H)+MF7t=e zGeLAoM3;SeM?5=jlhh6meoc}o_A9d^$wvIDAC)mQx>)9Du~?A8Io72=LWcaiCD zTaqLD7LE*ijgEcdV4ci;vOaw+8`iR4=-N%kEB}Skzb%u9?(1~8#`V$=NM)S`2-sX- z>S}wcEOkbxID6(@b&t^mXW$-qQ?Q~&Rx6Gjd|DEF7Wj#RYumtte zX+V3_HX_yIsxbM@KQu(S4fH;rhTHDP(Sk8+=;DH2A{(nje%&7<)@~ljcTe7k7fp`9 zdQ;c1jG?I{Mq6;N_T<2gH^K09pa5QE>9dSs7xA#qT2}f4ML1Ra7@m}+Lv_^rdDBu> zK5OAvNZRrnd@3Sv_0e2p^pkRx@|J+_80T ze%wnxD&?U?pPkbMtAg(^+tQHEuc${}d3t#AydGA$_#I8#;R<)ZIH5P+4B^H&5xFkx z^*hfwg`YpwBlg<|(5Z1}$Va>7U@5kR`1IK{a?${4GRxp)myITQUl$;iy-%rhcOYHu zc8iJL9${CH52YQy!tmhDv)C@T5o$gv(S4ug@p|lvFBR+KE}aQf>Xk0r{C6mpqVuqK z&I$5kPf@vpy##;r!gTy}?PIZgdMI(wI1l2m}M(^CPHjbTK()I*htrc}#*-r0CtdYgqL)9k#Fc1iN~; z2p<0ZjJ>w$vC0X3=*^E>kUtqK`etzmT#k;$4i6WhrzeB>pAEKLL#-#imM3)8ymcY* z_!7{r{s)h4RO6@ZGN7!a!6s?SqNOEmAT?(n-t(c3tocxY(mQX^y;J>Ii;n>oEUx&@ zLNhXArUWK>3M6ZOKiNLtgI<$}gl%5xc#4AzR=;A8y&K+Q=?UK0q#_yoBQtQfwgL-Y zri_p2E3uK!w(@8HZG;mg2_#1?o1Td}jV6UpLrY%lpm`&%!S3~gmY$Cm&_co0IV~AP z<^S$OTcswB>Dh>;G@ijDS5)!O#<;+tstCY?V;GJsp-DDg@N?)p-Xx}uOzL)m9oJrw z$_M2{d)7(1>z^yWc2(%+P4b4ym|>!f*8iYgZ#TDm)Nho~sEIBY+mrKOOnKXC1>_rO ziv7)Opy#4G9Z}T=eo3Qgp65oe&mW4v4OC#W*?-9Np$jdS)y1LoM{|+l`zzqqxeu3| zx=3YLtOwOuhhX+JH;C6Y!oKgnq8;KvUM0!~$!Byxv*3f+EidHKI-jX&=T%UE`9;7kl+gph0)JaW zg*t4Tf&H`2!=I8Dz&G09C*Pc{WOnFVZ9JMyf>*i2pm7X0%Hj*1-ct@bk2bLATiNJ{ zF@f@_f7XWY|k(cKY5^((IxH4k!bT zwthlV2aRCR=nJ0^cbIDA60mt#54rk*#8DkEYL0VDv;LJ~T~={@c>X zx)lt-GF}yLICy})6%~t;cC3P#Pdn*#gIIK4dnatb)hOyaX2(J=(&*c{Xxw{G_}XH~ zipzoRpL_~s3m&vT4PyF5HwMW|ze0z}2h=hC5dFMu7Yv+cRO~E|JA=K!hA-e&>{Y?L zyd&`BhA~!k6(`uEtAy?G(qj=RQrK{c9^Dan4alpPsG@B#s%>s!`|K~m(ecHo{LN7Qo6itDX`>`oaVq65ER@IR&F65Y^+#~zns;bI z>P1l4@EDh`x(WUnK3v!6V(PEF2A9SAU=z9!AH1T0pQd@^c+b_?#w1--BlQs?EFH0v zTLN52UW#L6X5iu%N>ooN0q%vgpbkw0e+P!5fsVO2QffYX^8N&~%aSE-o*~?r5M%7R zZY|dFFk}}TywI3&!tP9{1#39AoQ8E+vsJ0`cw*>Te9B+aO3vyGp8nerf3#4>HW7$r z{!_%uDi71$I%#lqlP}aPQisKtG$E-u8C7muOfplqQmqIf=c(35>1yR@%4bEqMXnk2 z%QWD&#c&^OQV^jcN=PYa+6!qq{ODS z+-H6o*{~utjs89}0y|$j!mO6WGArJIHOmFjo7M|xRnjorQyj-kWNt8-J%$yBo&aue zjKKoO1UCo7VC$d4E=gn~C{r2yF|QC}4#~ofzlNgsyep9l3dE}euZylIba2NWcp^jd za5{6#LKydO2<)s(!(DcjxH>Zqe{-FN+&3)aG?k4w-|_R9*3VeFWrrcDQ*p$dZ!UqN zPXX1dXIy=)A8k03z+T|naPa1Jkk{D30v~@NyH?F42dwwfJu4bu@rwrVoE5^Ht_X~w zDh0gsOAb4yu7#vrSMsw0&EZH>3@w`(%uAkGPLp1KMGL29fTfHwXveJJCHo$sl8|xi z;GQ{Dp;(sgeJJMlzs*5C?{&D<50{olUTUCcu4Xg0y2sS8?G{9s{DO|%>CirPG<*tp zM4Ar-kV*SErr0@`qocw^a|O?$!Ic|gr!iOg_b$gt?f8XM`ot0%d`1;Y1_Qyce-jRH zFk((W3|R6CO@@0XQ<<)7ocWJT?t$hn+WyF$q8ry~)0kg$BG(I()bprwYav_OsKkEO z{G&(uhp?3XKSbt<0XOpGMCNh(9C_O>4^L;*($;Nq?9UQAmL&N67L80KgDx%NZ5?(j zWJoC0{2c%r`!s2hgciNA71P1%F(1;4?kfF)M^B1i z_D&^sSU-fN{gPs`$#uv~UXS^HOlBJQE-Iq@{g|X3NOp+t0@1z=#;e3jRmYc|T(Ly>VQ!vLh*! zN=UTx+~+z^Dn&9Hic)D)Q7PKfPADRT%4(o!SogUO8icfmkEWKAq`jo|y}$p#^UHJY zbFS74s-zqEwd-GY?ZG3- z`DR7044MWpy{qY@yTc%|=Lqa4JCD%?8ZbD+hK5~9rvru282isnIB-^8W_8V-PP`f= zL<`$#;CCb1uXO}yZHYsfv_GFZs+ibFF6*KuU0iUi8w-_$CR_eWdEuyyX!k;Lxk;z# zP@{N~(pkqW++UM{ms;WT8gqJXLo$DTdl`}!BQQPIhZ*Io@w^bpJ!p{5r}&-0;MHyT zENLeV?pFbaXG(tP^N0E0{gMwisE&!BRUm9(sSq;E0Q)_T=cn!NWA+s{USE}pdqelK zqpy`w@$F>tbaE#d)sibv|Kk#GYP@JVZM?AUy*nA*EsbmvK8u$nUz`7kBuu*76V>Mg z;w`DyHSG91S^I?$$&nj`=l)zNqi(}Mltob8u&p%Zgg$PZp2U}2nZ%8ra1mo6w* zWfk{pIt6X>*TKhfD|{{GJOW~-lEKnk`2CRpmg{;`(A7y{Tw>*!XjGAt{nU8$;fP_*qg`>^Ra>9=Y37JBr? zRZk*?+(|u|_IDlh{v^2;GxNypZHrjT@oOX|%MCtkHGsZbmx?*H;bNG361-{P zR#Q45T!d9=hcHVg8qQmG5t2Se!-w{{r1PKGX!8H^>vty5YFo&mE*wPj&KGzLl@?I0@!>!7= zb9F0wq9eJw-)NH^ZS&CRyD~Jj7K!8ctzcERyKuXyg`{K9e5?pqEHrdK&K`~UgwSsW=GO1vuRX2!q`BuXb>wMmUlE8XmFLf%Jl|l-xTelkZcNvr58*9bCrboq$khemk=1V?Np28#6Nweb4w;vaQnJ{2y2RpCn z$paH&xrfIV$z%PMU2(mM7OPD0RpV~5f6*^-+qbU5Y-!i3-g1}tF>~Ai`}vdBNPe)- zVtyo}8gG4h$4YHii1R|T_(^X!o>&(_x^$+K^G{xu)oo}Oa~_|-93OL@ymtz3ko=>G zgA)0jhG~4nl_N~?QYdpQ@e@z#Sn-;$N*q4uGh2OhHW}Ki06tfHidXa~zFVJ58h@{6 z$4!UxzSa?Zc8`O6LSO)99UEaiFBf(O zTC%VfC(+(y2}`^=5WO$@vthHlk_}rz$!ptLV#JYT@peK4`>{n4LO~ZUH#}zih9;(M zJ^|7F7x5PgAB7iQmr07v3mE*_9&gRC$E=QQ`02ZWna2qPwkqJzRcl$Be!zW$Q$UYk*7Qf@0lOy=;fB-(a ze*!9dM&WH#M?v=%R(QQf#c$7uOywKK49mf@&Z21Hc$V8db-|c1hCII5fgd_@8yoB0 zdFn%TUiNtxF8I}jw;tWgn}%<}!HI+6@Q740L*c3*(-+X}yc;uGB{7oP&SQ`I9NzQ( zJiZ>BxUS;?I99zLlguXY{i8Bq=Ft`$r!~GVzRq;U ztixWDZ?U-LDJ-(E2-|O-U}_&D@oq*7nVsXy8vhiGIx_J%PRK?a@lb>R)**OI*M>Bz zzsCKE2|^$1Y#}|;QmC<0DqkL4A$3?X&|$TyFfG*w{iYgl|B)Ky(eWSn*Kt++=f6fi zd3h5zvv9y8P1W3M*cUNPuMQ`@>;{H!xAFW{avmS21?oG_2wkf)iRECm@(ScxO(bPo#xFr}Y|Hh(K&x5?$C0+6nrpxk|s#D?kA>oWn>dr5$6`WQN zrYpW*B8~+&u>5mxxN+n>AMCXWhyJZXJu5#vmYgf}k=PQ;UYd}A5q-&#iAu!UIG%f* zOkw7e|HC7?POM*w3i~pviun8t6@o6@kgYsX0Dl~FG5&c1pCiptmNaV;9Ab&HJP)u} zN6N%?!~3vHhvQ*y@j4zpP?6Nu{t-56d}AibgW1RRB`8lf6N~DOz@DXpp~v4avgP>! zqO!;r){mP-j<2?7)6KugM%4bp*ICoB!Q&EosU|aiVkt%~nvElh-|@^{x3R3_BJ6j~ z=0}#T;ls3!@V_@x`O>xDg;yni!8^~Kj4!ojTldtX@zn`@rjrr>HBO&vIA)dhT)Uej z_gjebj`idzwy%X3-dT)X*ofmUOW!4I#kn(b;gq7}|1>q=Ex%2r{>w#NKX@Dcwz`e` zwH{?UPA<(iows+{IO?myao)v6yXqp+e3c1|(fjWmJLHuXWICj=34wRp$lCGnA@MD`3QN!{e?zwZ0pE+|3 zR7(!>$+kv3!%&qkw9XX|lqi6$=Ry4XT8mEDo5^GSX7by%SMlFtRrEl8cKOYGVs8+L z)-_SQ?Z+$T_GPl|t@Nj2qD@y)dnb-Hoi(6iHm7rsZ+k?C$)$p0(pO>iA}L>LSHp5g% z|8k*V%SdLL=7Cz%(=h&mH@WN49hU4478G4Z311x_;qTu^QR}8QPd_=GM+A1`t*Y_7 z=36q^uh$oT9{Ve69c6-@ojY-9NfBNgG~3p6wILn0dj@r0bdsGpI1*IWrc%#?(yH`f zvydJ

gVT#AT$4&rm7l`Qz@B5zl&&v&&zwyC>C2qc9H2k<)qCh$Dd8IDdp!S33O;(Yu&Hgk$5**bOsKiM&a z``DdEVbKg+@1;Ood#T{R+rMO6i!)iGJQadVJ@7*M5MnVZ0zFgG%EBt7Y_gnUQEL$E zw>^x!p6@~C7CO*u7b#m39$q#sB)imdg$12oJpz0RK2xLURtQVUq=U2mFrO3+{L%ah zzjj)&AcrroKfsRci7%FV@$aF_*d!P_-HW_D`jXvpp261Lj4z#BX^Vc98Di9zOXTR~ zBIwag4b8qR$E4;@WPs8cHqStZb+>83oc8-NH<#CfVR}1gow@)XU7o?bG-Eox+i4o> zX#pAYce9i2x@17xeta4667E0tgu)l^K+GFQlcLwte(PdDuV)#vUS0`S`ugZ$*b3zl z!9wY97v87%FC!ZgNX(cxxbsv_*OguZ>$71vE?^|zY_t#34JTF- zU!=#|G4$WJ06g&Up6K@5g0B5uE^%khVz00%xNfo%1}SQYw}S+Y&t27RV9bP;vd zNrIil0hsbei@to_3AY^OY;6^kwKv_N5n)|;;wdTT6tH!qOo9X;ryRp-RV zb*jh}1Ih4-^<+wzu5d#66~<2Uq?*~sDP1C$H$798&kXrO@4v_+y{2T7w(e2lv0x(% zedmq_5q-&c`~E<-anwDk0gG>ill#xC#jnyl`S=MccZ$}K-=3l&H`!ND)zW*=x(XQ` zJkOK98$A&HrE^04?P{7bs2lZNABZ){D{z0!K725~40?`Dpal<&ge7Lv;mrFC48g8A ze9jH_VE=wl&Abl9KL^5;nOgYfkT!}A5}()l5^T*&mYkqSR7>Jcx)H|)xYalA8AQ$C_KlCoju^h;y_TeCQi z)Rz;X=y*6YIG&Eti|nW)Zl}enw^>rWC+v_oTPHOyvaSR2;eu5SG&}dm=OV`iCr2Jj2q=wa{<73sn3U!6#Yt5vP$o-7l_&JQPkgT2uxCZX)t*NN0uK^I<(cg7%}c1A^2l6tTb z=^+(_z!yxtw&r^zKQd3VW=-Cekx>_iLf_Tb;XA!N$vanNVL2I^**2Y%lM zfx(YktS$GIV5O@-j6%8zT{KQZev2wK*D9uSZm)vJvJ5&Q-(LQz@+f^A-<`JoFo1q} zdRUjbp3jixTr*xQz#Y@}!WOq1;)g4We6mXi{vape_p1%8W!XknHsvL8^LWqJ+_mA8 z#a~Rvauu4*jVnvqe2u;9lEBu5=};AZpH)hIpLrQ)S@GP*g6@FXcu7n0L)_LcZyr;? zyQUW4XPpjuEm$3?~wJCVMW7c+&F%e>w153aQG zD*vQoU9Rxdw7j3=f8?8*6{Z&cCKFnx34hW*f$yE+RJH2_p`qJg+Pf@)YlftflMlOr z$6Pna@%76-=OHM;fdOy+H0Pc7P|L zedsN3ud>JY8pX)v$?#<>z-7G(R{ym|Hmhso2tq z4FzyN#}=+$)Q3wQqj|y3?a(cvKf8HS6+*{8F69oNKMU67_2iG~DW*SAa*Xe+!ha7kpk=^o zayC4d&YdZT?Fu<)dg`d;oOq9`7Vo6njn-1tQoZt5Jr9EL#s`MXiW2vV*)-N>6;AFZ z&~(YivHg58)c82U@Q;>a%fC#fv#E{OY3||Gy&8G8(Jj3Ea+Q#u`AWF)?+?2b!MIzJ zba(13$9JBIVB~w8TsiShnBss$vdCuMV<9&>R?X-E!1a%u(8OyO z)scAAZw<0p^SWBR?=^}p9aBSwj~t7;{|;a=qlOB5c9_G5gOhN}CtaG8JDFC_Iwf%t zcX7LLM~E!{C3UXH(2aAqLS@o?)Ri&;69*(<%G;y-Usl)h@rm8p3-f=ZdvFY_H9Ri< zvG8HW#Up6@wiI4vyq7Q8>BgI9RI`SJ!#H>R2;Ma`AFe%gg!4-WKxV-^sMhr%Zw9V} z7Dq+-u9FtfqPLVpnf$PH*{8W|R=FP^U5Q*P!9(0%YsqboSK`iYZ*XhR-Fyg| zaQ(sojT=*#P&RW+m^%!2Oom2rMIYT@Mv)9?#1_f zDu88XUr3T#o$Q|ZP;6MQMr~gq#C@y9A%DWbTKycgA20-;%yq~5E60SyDph)+(1^ZK zdIE);7O~|`v6X5z2H>+e{gm9TF74|!=z1S{DW>2{*hmRXH9!?PMi}q z_zwilt`k75EJx}Q$g%rXb#l&jkLWT2WFZ+tu%h4zd>g@8(sh|Iy|kHRM4qI6Ce8H1 zpI1~Wu6|LD2kxp*;j?Cy?%yCsL}<7ARwxER|CL)fY0Hqv~3QfZIK z2ar8K7XF2V7eA2DntiOe{}jAs|A75^AhGaGcJu51 z-NVRPXN9VgPk48=oOS-pWvL^ruuZ|3RrIKZ$}4BYaiJ1(``2z14{B5Qq9>$mLKp1+ zSRJ;9oA4SvDND9(1Gr7k>s{d{8g zS|nx-2K2p8Z@R1J0m0~BF}wYECRk=9gOg_Hrf;^z@qb3Mj02&vyVH)7 z+x6Z8EMJ7)|K-ar4EuunCyqtOXM2d=K7F<&c(UwX_9f;sWEXo}We*+|FQIzPc4|}- zgo(Tl_wMtB_3m*HkE$mNL>A378mF_R#*W~=emDG8D#DKkN0-|CRS?okCczn>{y0|q zneB&_iKtoKN0cWXl^KmbWE)g7g#J~yKzpkr^>1sHDKI0bw?1Zjsje6fCZ30n;~x<- z&E+)e=2tTN{&|S#X-=0X5}K5uN`DWpBQw@arFLUX=@0$o!W^$@aLr={SVl$yOdkOg zTUQG4movor|0ZM6#4lxP-FlFh2HynV^bq0lqfdh4fgI4N$zclJ&(h9@3Yr>nnPz@| zFC3~9N0nTRBu&@n%cc$O0+YQeWj8ZkNo>|mI5+hTetsRv6pnichBw+tpGSExZ_qwS zYyN}xPd#Vb%Ocnv(H)DnyOGnouj4s+- zarrUaXt)K3-A*d&Cb_Q5ns$Xyx7o&HHs2t|ugZmW?lQ7+;Rein zQ-Ho*W4NyQV(=~bh#yuw!2T!uiJKMO_)OjR{A1-%dUpCOvHK8&&d0_i@AoTGP}z#( zy4Bzl?NDAMH9uZwOlNb-4w3k19lH7UI5?9L0j|>QJ$ykZv^sdw(2$itC6N1#=%>Ph zEzhy!fj(`RJBBEg`IW6zjKG>QCq73-i-0!TOd8DBL*C zv}7k(ab*iOm9J#>i;RU{BL~6Vc1v=t@1C-rHs_dvSDwUEh!hR)yTT`!j&JiXlEr>n zJk^4+H}+>)RpcIsGx24%i7srxhajBzd@*)wQlu^F1pb&LLVjcrl{r3yH788)t9Ddb z_?tFaec(Idr)Ng10;IfN-ck6^s2D~!h48o1Y)(IaAiPUH0^?O5qf3gfxVdvLhKxuQ zj&17AT%IrBo?i>4c~pM_-eWO07YVIOBzICBa34+O%7+H>&Y~HhIVqMnTn-d_vkE|n z$L6J7&_cP0=9(U0S0sLhjo&aBlC4E`4;4w-!|_&R@4G~{?%@mQHd7lG>i1CfZC+CMr&$0Pqx)j{3vG z?{i`KkvhB;+RgT}otn^|Un%VOd=B5Ld%&EmVdPG8IjpD^*&5?TnD)Fx(0a8S`%Zg` z+fpUZmGesa=utl0aDPq~&S(-h+m_>!3A*T;aT~+z+F8Mh7h*5-YSt~wSvcaAEl%w- zS@MBx;7*FVIH6x@ncEe~b2fCcusWv>&LpdG11Wn^R1gm7?|$Ltd0z+|n*~vy5`--5 z3d%_)tn8{WzdqXnqEu4I-&xPV?{OZk&`cF|LgS$*el%1$ot9;M`}0W00~iY$)}!q`hk;fd)X$eUCEuc}@^`-JN}WO+A;m^Op#xG);7 zsBecw5AM)0MG3Gt!h*bBeSj#u@dTG1Dddl{4qSKb4SjxwVy5qDI(}_5EmSy4Hs1XV zmRWMT<6ni4;n|Bue|rjBt{oTl?y#bdF372InB-e;yA3ZKde9O54Vc_!G5j4d95S!e zz>=wn5<6}PP5;%C>ZWUy&Hv{u2K_llJbzySKZja4nq5PKcC2PsdusC8exqRNg^xmd z|L;PMrzV-Z+8w7wl;P$A6{7jYlkQ2nOJ-;vhf}9pWWP7>VPBLR$nBgs(D7@8B|pPq zezO(uYo4%r&!sZE!l(44U<}LMZ0L)la@c=VUGB3)UG8_o2<|W334i9LOU0MoWR_AE zJu>|%jlDXQc2FPcx^NrqTvkVOu5SV=<iQ0a^qa3hc64s96mY78;m6K1<`;XW~U$Xu4aZ!TSBc#&pr zyAB%H!f5WfiS)uM3wRLl5k8k|%8$2(${X-EJsI?XJYLbCS@t~#*Q~yP5%~}F`b5%8 zn{418e8`@+T8M2={gt!M*`w=v9vo;2 zM0;@_fN4FPvdbY6gJMW#={-^?tSP;GzK4ALgLL|K>Unb9*hK!}V-;+gk_a6ihRL0_ z)WcFosr#uiSz_ZYBh%b>k?%9x>8q+S^7-{qbb%3(zrSxQpK#PczM&N9#LJdo60|~g z=b1ho9Djjo8idklKLd;lS7EuPZDK|@8C@9|PJ3H)p?xc+QvY7|n78FG26=pi_Y;p| zNd84==?U^3)k!oe{{`GGohVjbo`&aY-@va0VFntf^`beDFIc2QxZ*yjyI z8TX-Q@4BZ#i^Z>0TO}e+gc#J0zG%MheBY-L!2~B&-{&3d0UgqXt7vXk5uH z+a5oMLhqY}LeWJ((ZYKKHLC3mKh3TP*@klbODQyO`ym8=tfEtW_mUl7n{72_Kc(%D zZ?U2=lBZ(b05T{^4a_4K3u0ajo+uj`JrQ}&h#*_Ubc>@q*TN5z*=%( ze=f_lJ_b$~=fiB{d=e+DfgqoHqP{7HM)XmUHyCVy(INV@RNaXtxEu2HQag6%YH#{O z>c8)(dqMsNCE#jBcX<8LSLUQ|M_Zu__09{0Eo~{Vy;oPzKQ7TJ>}J8TiGH-+^cY`(!AtKTIy6^No)O(u-d5F4PY;a313r&o+|g*^V{4Z1 zxATs$WcnZIwnK?7xh*jwf4NeH_2D#o#aGx-QY5>u#sJ>0Zh-+G?}N|rBd97PQxkQiG1fQ5S->4^$;xwt`) z`^Wc|e{|K6@3mPDj@q`=FmfWq819AEmw%v*T0o+juH5@y8cnh)pci$V(V+IW@N1Hj zc*f>(nbHIoL0fY;JG-YZYKr&B-)YT6zrC4kh>wCf8mGw9ynax6c^6#Op9i5rG^x%@ z5i~o(L_hWO^7Rk?$-`X2<@i$aTkH46E8F`@Jkl(>>$|z!RVf0^7fca~{>95q#NA*W zEf#dqlb^7B#U$9XsFn8rHIi0O-6UM!d4U=%yh42s`?4Vqe$g)z{AthH44k)3O>!US z5vP;m;aBcbR%a`xN7kHW`@1B|t+~B?$Biy><+&rlGa_AbUM(Q5o)76GF9)znPos@H z-Qi~QZdqXGwX(%L44R`AkUZMH_mNeiKqo0rm822FIw(n5A?bfShgcKA|3O-fSO48={|>-LR#xgP@TV$8s8rddV9tS zd5ha&#r<7^Ov#U!ZZVRV4|_@tYnIXOs~*zPYBkhzf<5hPEA0vM`_pYx#*!tMG~`ZF zH^}(@Pk1)?I@>vTJpTEV!$S0X$)8MqOP3FJreAh_ruSCerM`8cw4dH%+OB6J_w(10 z|2*j=Z?0G&PxsK0KPys_58LiSlfnyOUcfAtdhHrDdN77&1#hC?{KKhIcoY3suOJUL zD5Oe41pV}ylPGs%dMh-6PI>i=4!T0*$^T5`Wj*c-FH@$IF;6cDRX&zN?7Qzk;m4I{@Mck+81D3ieKzIXDyxOd zOfpf~D~JD^^qv>A$azla0lclaTyh}KXOE7^K`rzv?(*v@*-h)o+ zc?wG#E+F^)j@|6jxM|NsP&?v7lW&YT)-w zyCha&9|{XLSwe!zXukNLI{zJc7H*uG$J~EKj3Ns@kmVEoa*$?i4(C z2ZNp~nO}Elw^FOec*ta4cP*CNeyD?YnuL~2_{dgu)RXc>)$H1lnLK)U0zQ1uh8=C0 zWuW?(Y)tNn&$o=j55A4`gU=WGBL6U5ImM2opZLp)4i3Xl`;U^*P4{G5TD4@Y?;fz( zdmP|sndFz)9FIqBM`Ga8`_L3w1G4qksPWWJ`t(@={N|0|esc}n@K=IyHC^%djF;@I z*AcNacPM(-okdNfG*MxWp5zG|4z1qTqzwNt+&kF_*KDc~Zbvxrkzb$U&aKBG$2^TS zN_$XM62^lpEZLy|YvMd1htLi?`piBJ$m9wf7+8j;lds{m@MzXcUyn~(lFqN>kH9Nk zqeTzH5Pa%B86#)<@_UPv$%(o=G~eul@oQ_CQ@ABdI2a9cueXRn_g2!D6KA0&@;Ebk z`3>*f5V@jx-jM}w<{_u=l<8*pN-8U3m? zkA8X{jvb!9{8`|7K7PRgR=4XQuX$h0XU5KCHXZszd73FZTzNsTRqPELuRa8=;73knfkJIS4Zzz@wvgMvmmDp9v z2FXJMVL+OkxkpducU}AA$M3(z8Nt=e|ilu=;#dpaDP+=KZ`y!+dk=4`tDI&$-C7RFvHM(v&D+-Hm7V7{BkZ!0O}; zEN$Wd{2W&+Ub;I78+WE)boB#T0CDt%+5$4i-3?#pbcfRC{ds|JG_7n?lD{6jjD9@d zho^4&!8{)5!uowriAzZXxH+6+r6~pE^86myY1)-Yjv1V87eVyiX~Ry_2tMrOH!|5{ z4GfF^55#HXiS5V=(I-RKXOso>sU z8Qx|Vk|K#edMn%;@4nH%*s*8D#pJW_+-EOl?&}S|7Wp$1UwgE}Zum<`rGIKK(%Vxy zVMDzs>FHdKzlwF>Rrn|Rxr;geQ0_p5(3@gR#wk<@K1UiogQ(-0TX5&s21vvRG47rN zmXs0tN!5q`91du`#FescO|)U!cGi?4JUWd zj*%~HY9KZSaJHi1W@v*Fe9Rj5U+=&tdr=+*4u)F4!q zUTxS-jV@7^p{R|&Hcx^;V>7O^-b&(J%E;-lo@|!73bfoyVMCgBke{>cI4N*}>clKK zr^Il*-#eP~%9X;dEH?h-4!pDdfaEFcMgL_u6U}+~Fe%`vQ2j8X?B=?aSXrwgTwfXs z%T_Dmyp0FN#ow$-7A$XM`kO59>B49Tbvg)coAco5PE#s+9j2ScEhmqp>#oD^O_(^{ zm#XZkFWEEcCeOAgmJJ@^!Y`YT<9GkA7U2qE!*=~(m)E4gguZB3agtihjiCSax1~1+ zoPxZM*>t|;G`eEjB&qA5M-_+p!{wSVi2hRqdpZI^Nn$j9H}N8s`ti89fT5X>D~b_a zab8my7QWeqs_)0Z=i!}Xr@0Af9G!#i9Y>ha?*j>Dlb;`X z7w??+m-sZBnf24L?0bCxyH{}zDno5~WLgGn>^l^$;S#vHPmPh`I;6{y&$x8YB{Jn) zE_AGX1*2@w!rjwVu;%A|w!W=5KK?S04+yRV`=zmL+n7=qywQ+O(Cr`#=j|qq-DB9} zr(?ihND(&f?|H6G(o=@;It_8fs$7}p z=`6@e+6T?!{lG9~Ir~;=Nj~q}Of)R&NkOX_E?V#h?hFh8Ma^*R5%qu_IgrMbPwPvZ z?a4U*zh@G1_W!-(RtUJblyxgR%r>i6ksfR6NZXb3F#Wj(ozzR>FzvYrL-$-IeSDJP z#kx%5EalS@;^V<=@kExZ^92%4&!X14-Dvs1#Sp)v6a0HCQ~hnH;D@W!BR2~Yw|+PR zhsUSEJ+-0G^~_V!zpwOvJ)aBR{ES%V>oYP(-CU*=eTa;jGKCIMn?&nhOrgV$o`%JA zC0XOwk9=z$Sa$hX5w0EV4q=Up;locObdv5WWwsZf{GuI=+Bt+CQtm~cdz^&FUq)d} z(F|6(us1#UR|`)~&BY+MCvYLVD=9I34T?|pf?~8A?f)&9o>)k!u)a?8N>`$KA&x9& zdlc;0YC_0dW}cKAC{3kuymsqXbWA*r&H$V&y9uBuHB zZdyvZ9!n%o9KA{D?iF->j5d8fxt65XekE%==F3W+RuIdPF=W9$Xn#=eSlFa?L#qhMFdKi7F92*kC(S7k8A=H<%h(`ykAc2MQFj45yM?Z(d)uIFP@wD_WvUAdy_66}1S#aA5Y#VzJYSux_y z*P6URWBWq96&;5`5`XJVL^}3Ie`x)17*?mGkn({PAH@S2otulNFBY;>m-eybZ6d1L zpWz!~R`9FCM{t7)L->pr|InNt0xQWibkF&+sQ<1Ci%;#wSq*_g&ID~7BHR@`&0AU6 zty<)M!u;TYRVd>3&cx4lVV!ig{#rE#dl>D(KfZfJW&KMq z-)tXINHOILADlx+TRr|^Prh*Eqb;dR>%s?04&o%sCHy3&^EG5CHDf?gw5@ZjU7{AnP=5WnGEmU0|Nl{m4v zvwXm$%X;R~BbjX7rp`kz9xRJ%HbsZ^GZzLa98?;T8An8_GB7X0V0-BCy`m7IBN5XS}WC zn>A9oq2mrLELFjdcVBVSNNe^h^ax&YX~b{C2Z>Q;UbtjtG2WRs2lv)ZkluDdXe;;T z@kP13NA)y5cS{8sqqq$pl^e1k_xW7Bk&DySLg3yBGwgQaGdu0D1C91=Ws{z^2^-An z$-_B6PzsH6h32VzpJTIu| z%8g7rSc{f99y}I-dQPc$ZR{sj>$*hdykr2M8XU_fxxBzgIr3jyFL6h9f=?Wp#AaG| zMdiFm2=3Uz3)kMkndaHh-1w09J5a=HLAhLI-e2A}OR+qqze#!Cse!y>Yd!y%n!vRr zewp?9e9X8t0}AJrlDU3GnA@)(+k&;MjsjGEh=Gb7^YG?5Irhl3lx1-RnB!(HK5gqm zcIPORj}gjw-HQu6b6X!CYS<2MZLN56eI(m|Igw2BDj+@2)PhUtQ0{AeTo&AT3s2m< zN`4qz1HTbL+*5rUZalxOd~5qr))MiWxx5;ITds_xHDhJ0O}$P0XVC-~BggZkZ3<*s zTP({=+sO3hsfagZbNGmZ4+IbBC1uM5va!;HA0B)H##V%&(h!~UuY-#Cjf3-e!z|tM z1NFH~O;r(lPO0FREqavyu-(YqKfEX7^xbJ-%o~i~cMI#{!|;Ji92~iMj%&Q^TW**f zi*8E>bDNF3_=>GY>{oLseCVNoDQeL&)y6!Sf1{o69{f^pHaU;(JEoy|;cF(lI|Ub> zDIhOjccDXl8+lLv#ZbROfteb9CAYs$fG%3yF=q21jE%kqPs^^eeY2ZjckOWaabgKt z=gx-ghdY?BK^6@Cl!&T&7W7z5wP0_wn;aTkNz*F#Femk1q))H!%thxDUcZ}#|0Z2! zGe_+qGXJ;Y>4(MOj+4rEzU|7ZuOTkEp&`EWwWjKq+lg4(4fp)lB=+K~#N!^Pm|ABa zQdu1cMfdFvjCl&-HfACn)|X$2NM(b^)Z>OOf5h7J`$A7BZROLksmhQdAN5q}S>WDyL=h7dg6U)un{=3`A(9o{L z^(srr`^TEz+w*PagW)=z&2$O|VTV(9 zZZb@bpDl}#S#5vJzSyl1wZ@eZ@1N%xoo{;4QIQ*9}L`?@j`q z2e4NuBHAwbBIUD*z%pR{mXv_AL3@mmX_^`iQpk>^UC@6b7c7kFW7y)8itCAFnU2F zJZ>)$GaJMCm-aJw^07VpW_Ez@^fE5jUXTZYn{>ssdfK4%s*U_p{8;8N)r>pt*at&~ z&xHV^dFXh1EDpw#Fkb0BIkTY;2K=lf=Jo>w)BcNXD~)xS=?fcnY|So_e%V4wKHNo1 zxm@_EG6p&oME;>@H1FQ_MjRimi~W}+vKaeDVY&3in>MUaaO=Eo+jW!!_8k9`n5rLR zHWy>jdG0XU@VWu!eeX|)FR8)!S68qxbRtpm^p_n?=?}whNqu`yM~FMHkS+KUh7%q; z;K$id*}I|vbkOImWMz3hY)JJGHXb~PnJH%E*`!d>*>@%jKBh^lKFp@0^v;UomTm#} z-O~TVzg0-Cb%jTf(@2Vh`&%lxM7J+{&87tOXTTbr_rzXoFd6gl*@qNQO!h6)eU|B$*1nIu#*w$wfF9IMElg>x#LQ7S8m)5RD}H+Cko3T^_sroiVfPvtX)%@zvxp24ZHA|Ga*10OC#(Oqxt>Das` z=$&wqB;Gd$?b~nQ`JC6N6!Mhl$K=0VZnZI=purt*MGu+nRjr*#hyZc>oNHD+8r@>#17u! zIiAleZ{VXYhtZu1FW5`(?d;j=UbwaJEb*r*w8rN->5yF@6=TiFVS@zxXyl0YC%Y5l zXLFcK@m+Z7vIX?g^l>S_go`@@xVnOXPkO7gibWak(s!Z-VM)O zFMvSzNfa{HK-=fJZ0+)WZ0M}M#6#Mf)jqX>8Ed-W=8317&AvBfPaf*BNoVbuqIg$y zP*{f36>V{T^m_IroY)$S)hf%gbioF-UbH5}mcHH60?&F616Q8_sO$JD#CIQrW?$#= z&EszH>76>X=Hob;z0;6oM(t(mB~JO=RjNYP^C4u4>K>uj?U}??;#-|Pwwuj!&4N#o z&pm2;8QZ;x!u@mu@Z?8GX3uo;uE9x~0exeeE(DP?a&7Y8>XXp#x+Z=)u87~mFJSDb zQrs>vrvReMu;;417tb}Z{(lg|bi9w*IB z(wjeTJ~OX6&p-A1iCJTH;H`HT`GDt#ss4~es(9xMGCNHgs;od)mxQnjkJhn5`z?a& zwCPYg{3|=v*)9%n{tF8$Ie6~eBqYz-Om&hE!c=7~v@KWzGyXGVYJF^l|50|P;Z%KL zA4X?rid#$CSL8D5Ogo;wpNF)sknM+7Qg`!lXK{;nVi&QE}Nh>t6| zF%DZES7YArgmvDlfOPI5?7y^xDLw0w%+*p8%SRN$fk_yyThFB1_ISg#3N@^}-VV$M zCYSCmyCH<@*g@M#Gq!M{3)`UfK*zqw0Ke-=>FG;rH&~pRG zmykX9-lc_58NQ7k$?;)VB#F%ILkgby`yb@k&w&NgEr)!c&SO z;ojSE=+GxbC|JekE zmBZ1= zw&|~hAF`!zrFaFFFBMCE{!%GR(Dg2(yA#T~l^jTX1|}1ul8DmW3v|YkImkZjt}tJw z67Q6D8}`)aNR$?slGi_#aqW=H5HvEGP52iBE$9Yxy)a}8-n~Vw(zC^+57$9Yb|)m% zU4Vlh)p2I+d=#>O417&e1$@7h20a(iKevxS%ji#N^P49qURDEtUNVO741Yu8Hs?S= zbUOHu5`eTpWaZjvU`Wl-zwL>*SmWszGnf2SGa#!Y|&-^PO%r-;&b#gU+= zAn`y{KJ^no)>II%n|Uo@*5@g>}OXMRLgRb zLhu;7Zgjl{ps3sf9Q_pV1F8EZzxXUp+jt7M%&!v)=0`*G8e8_N^@lk3yCW+0mqDjj zbzt?Qfw-&54UF_lX;;80EU!_9bko#H?bBZ5sc#6EZx_k zxQ*o;slee`UU;2M9-evnB%MPtdCpmt-QN@d&mSq#)1TYX^B)`0g{H^o*!*K~=B7Q{ z{Ne_DF&Rpqj+_2{JyON_u41O2?YG9`V8cZ1GLJ|0Ds%D5vlmP(SqzY_6*K6 zSe7!+DKqfAqC6}Oeg@cC{^DIVjbYt$Jv7?g5%Y%T zY`~s0C|a*fPt1~GzYButzzr3GU4RnnEggYX`;{b%%htoYhbHhT@iNYjc0wu^Cbawg zF-p5b;M+=5Y`bg@*1D=kYQ^pR%Xvv`O65u>b}GVuhrDB#{Rc72>S4^PzDnw@sx3V+ zL4uV?FWlm#ssY7Jm_AX&%94v{+MpSrp8FL-zWBqUmD$Kwdo1w2Yoz&YU1(A+hL39+ zXuqy4OGgXQV_ARtC$5ez@9q_1u1DZ~78FljvkE#}Hwzv{wzStV7y1IFf3KkkkI&G< zrCysQ5p{cEP4F~U)LsGS;y(f1_X)mB?`Iz_U5!i!8)L~pcN}6?hCTl)7Y+@Ktyhp6LW|on#R@JFzZt14cycLm(rvR}~tl!$Xuj?ws} zAI#CQfv%Bv!rit9q0?22^{yl$(Pml6V$N5{a;m|v&o5?nzw+3bS9U1>{Bq$;$}!y0 zlt3o`F=a|Mv*6bZU1Iuu5c`{v%c@mVSb_W@oL)DY`CIEEzfUq{YYXhk=7r7^*NI26 z?2lhz##}2FV_OGbZ;W6EjB`-h!CiFjZh7o_ybUKMzrrag4_K$&CARdJM%lDCznSTD z6&Q~WiVye=#Vaa+jPjjaI<8-p6)yWFag8jdpI7R#q2ecK+`#>?1Il5s+iqIe`V0Dt zYgl4oAuHMS`f|AIX>{hzFUjj)(R_G&AX>Rc>TOH<3A)m$vHZg}iA3BiiRbiLcN4In z741;W4P|Ni1$wt*?L$T1#H=l z``Ak-1~%>-g*LU8;J(L4B-0GiV796=TP-d_&e{%Cr7;36`s+atmb*)0$Gzs`*G`qR z4vb{I5qsfPBfy685oCtyX&mN|Nz-E3VU=-A0F476~C*OW@0T4;F3^xNz{rm zp`vdbb{l<+c3!^)wn+qBAN`hklMI##gtgyMV^DwYz%9?4dB@1NbG;YAO51i z!P=}XwUb$z%di@*8LWp}u;!k7)a+l8;BdtS-Se4*G-~%kbcO+YIQ9^%d%20N)L13i zdv-JODRN*kdwQ6Qc|Hv5d?<(pk3uraw$d)cHR>eob&iRQ!K*Y|;9x3d(Y1M!MGMMk z`SMRNYtDPP;^+puMNw>pz64vCA7+ZvUb3)rCd}t=6#M?)V-|2S8y;LA!aQ|OFw4Xd z?8sXiX7-;yE03B2Cq!qc!}DO$sJ|GFf8UMNo~o13vS!Kh)rQp3U@n=M7Xb>RTO=E< zI*^Cz?XcLw4blq3`6%-m!T#MR1X2cUyzVq2J}?&fB*{SD&kgii(h&H;hLPyIa#IpE;{JCUZhy(!FGY7%djPIKJMz@iA2ve@JPqsXr`2Nv)MCqI1-wJ`H@=rt0bI%p zysz^Sid%Y?_gDI)cQ{pq)FOK&&SAO;bk-)`|J>|HHg8(Z_gk$w{3Bl{0Xo#uz;w< zFSOA_EP1uO1hyu8fwyw}*} zKdp#(|24_J>R4ow@fk`JFQZub-Sos1N`L-4Dkz&@7b}jjAV2z4;opdbpzkypzdE>G z%E+<7+oc)EW$%tc-}7%Y^z~KY#Jd(WXuBntdtN6~bk+#p9JJVgkR&{_Wfoa4b(>tY zn7|sI?*=35L@;|@PVGL8z@7e^==jT8SY^;?GHEqWTn=o9A%jzp_p}LCvNVcq{FE1n+DFrO)fI!@9Aw|5X82`Bw@` zYljQZW?QrIdOz{pvN&j0&!v_(A3%ML4^~{Q2yt$%pnuqajYK<1`8`u~k?2-prQlo)C|gpZh`UnvQ~6mp zn7bwm9H}Mw`rlr6KPk68+~RIPTPGt zBro3Wqnqp<9Us_OkiVaokPq9HnfHkO;5fDlG<8QvIYb~>c%-54iFFe0pi~DM^iMqN zkfX%7ZWH_#&e9`W^Ux6~&$*YWz?@NO!i0{uSY$JhiVV$2b!RtdzK@`_-PcgC%m(W9 z?JB*r^9gPqolgB3Mz5>R;P%z^SdeSN6QX|683(FRPnVWtm1-Voaz2d?^@UTF$!XYr z$~bHqT`c(!c@E1-U#>0Q|6pN$0S)VHLZ3$M7Rswy=^x8OLjJ$4c$0rFbzd5cN;m99 zw%WE#^~Y0Kn{|qfdv%^99j(vmwBKo*9JHasI_t|ny=(&tM!Xr>PUi> zve$9&nboZ1-wJZ8ayv2!c@Hg5#=+e6ebPNbDijW1P96`EB?)K^cqunSN@atPeC-%K z*pNiRhcAJf^;P6gSS5aVe1mjOH!l1R_FrY#mnXUG+oIQC`^<_JZAoMc9nx9tTMITc z=^Ktwo+jm8J%=qi?W8*J9kKZ{kE~@1B>1FM8@eq3!`>L;zQu$%Ra54EC7CTVn8aST zS5iHX6miFAFWNnGD5-mPP%w&jhN1v(_;#V2RNGmyNeXL8!^nMPRyk%Lodb!i#DkTo z>a$rZRfXH9O{M!SYvOW#BH*1I`%;}$c41ajSz=!Zke?4QO!Samps z)y$T|>mCeo^m?Zb=kDzx4dr{7$+AN5qKDbU+YW5Q)EK6AvI-tanGL&cS(A5`ir}9g zPgjZ?;ET-{$vp4FlFRR+>98fwXtK~GUiicm1$=a-2mAHQdP29CWm(^4FP>_n-Ps3m z@<0`A)qj%J4Ld`Z?u>@j>%QZ6J@Me0Vog+xo4~g&0DmlTB5CPfOb^AfKMrxQN-m82 zr(Ak=Cwm(mA(> zw0~U)&sXQ;shP4YHQR{oUgk($nc(7yi&OiSBGlT}(EF(-xnPRYmHCaS^# z)qSkN)R>h7p26;62p<=#kW;=R*e2O|Eb?p^DAm_NWQrkad0YT#rq*~K`v7xq>anBg z*Cq3(EQW_N3y|hd7bevWL51Ib{9w@#_B<~e;sZm8i{e%|({q}%{D~olbgW2ir!w0p zoRqr9ji7O#mXw1voXXdg(04LYK3bN*axE&^_xn58vYHDxw51=WNO$s;h0IZI%s9u) z`WT|2zKNXf8pTZd&LNSVEc5pA!msOJ3EP4>W;nx{js0*RMx1gbi_E>?*xx;5j?N0< z_v^pxXW$v;HYWqlG*{DCp%1|Grv^#tk7T!O%A^djZaB4b31n}YN}8s-p?u|yGtOOA){VlZJj())A@%uvnb;GXaqTQek)w|%?5*a8EC58BC2HgRj{dk zCJ~0sgp<7*7`lVVi)}V+_lt=*BK;{gztt?At!GG1_$abylUGo&T!O+NQlfYK2JJN& z1NNq4@PEsU2pBX-8QzCU+~gU=>L>7(wjB2O_W~oYzwmwfdU_-%13_H0Wa~`=3yXE& zTVxze)v=(Vc}-}Zqq&rg94d(jI*Gp3FQ8K3O()P!~mizyN=mUFE%bSIG^VSl{_C2rAmx6_SOVB{5 z8uJ%v8V#p<3zO;E%Y)JG&OgHR*yR}Ar8xDK0^Z{DNU+?bgr?p-hOT{#l+@{r#c1_O zIO=Z6I$H0-v43;R2F873iLL@m-GPXOznd_+IuI{#&SG^F8esHDJ>0u#C62G0ix=v? zffYw4palbmvzgJ^sP(uJ9c>tlDvnpvQhgQHGcN+`%zG#7o2kW4v^zs1n+RuInz7+A zTkKo;oK72FjtAu-95{I&ep$7*Y*$V_6Cb_?FB%6C(}-}~Jy{lUU8k7uLhZ7ReG{1a zmdRMtxtl&dg6ZAOeq}~!`^qd@pRy6UyV=1~Pbi*#PvTxu3rDm5(it1RLeRH+!mel^ z6z83Yjm|jIC3FW){g{I!(rnTC(cvtuzlmMYoz8BI%4fq1BN>S(M7E0G=~m;%&{;K| zR4=SzRv9iNkeZ`Hjmv1Wnl;`l{a*fowK(Wc226eT1U=J9N1IwiP@Y&Q6z{aARBIIZyI^7h$d=|0G9Je zq89L1keo*}pn8npG*b>e_be7hwv80e8u$ggH>kq;XQTKR`Ga6W?if7dUNsxF%ugU4 zEqKHcWg$JNpZXOJrxS|P&|Z0MGBxZh{CQ@Vj~Niw-G*N{)bAGZJ7SORA&CR5T>u04X1QU>8}}M zkkQLZJYuE}j&_K~gGMWsZ5XOp#)n^m?K1HcS*1~*Cqd-a`ypghj|Sql)euFW=b%}6 z2V!R?fx7$~>UUfX+t~cV%QfcEFq>TH?R8`2-Jw`AErAT|IZE_54~B&o9NENWQYY-h zCqkf21-jsU1hULZK;*3e|DIpMu6?S)yjg{ax(48pY3EUE{s&>a_X8T%_ef%K)dVS} zASkja!CHM1TIv0Y7G+yXLeMASH*zKu2bR!16a1x}p)!mnU&khc=3#C715_-36MyNS zFYT;ugonP1aPf<+sO+N?dMI3^$>&Vq^$2UIGZ{lt0!&D(ZY@~#BXBGJg%hGq;>=HC zQ2H|;s;iRN-8Z@5CY>dSW}c>Tjb8ZDqicfGh670deFOxABB^-DtvTG zC|*hz!Nh$EVD=#szT|!760|46s*_VCCkzh2O~WS;v$hbAjIPH&E!FY$(FQo|rj!r( zSVDUw_y*os&FeCS?bTDx8_)517RkVS+z=>-=rb8yl_~h zcUg$x8z8$to-PSmfX}-8lkSQG@UucCdL+^wjv3z<;7T02X=MN#RHx%D=dx&L-AgF8 zI!QhE)lyl%1CW{Hjv_Wbr5&BgI96jeIgq5l_e96xBQK9b`ze8SjI@HUyT1yCFKqBB ztNpO6YaO{1;6)BrXdu>`A#r=3gkBlFq<$$bI7nj&c*mWm!E=16VVX64JF=OcTNR2I zwU*;+opK~HRhBiVnlPipX!w*r6*RWnN_yKC!jzdofOpy9S2Yyrc{EacE|0$RilPG# zQDkE}3G0?$fZCFaB-Ox?-E_BQ1K|S1k1~cAXr%qN5{E=cR^RFL?#s< zu1upvStH?+q=K6J11kBi56PV8(ZS7=@q3e>5~z8O?fw)Xukat@d2_k}75s#HDgX4g z@dq5dwg#0|d_+S-7Lb*(CD@^*oO*gi5%JX-#Bt4Z?7TdO-s$-!SuELyf(LEJle?y2 zb1f-H_i!&h`|3UBwq;`5tT^nKF&fYM`wjFCZjmTi$+0)qeW>$iKIC$pkotQd?H}U| zziicD$+P)1>w*mavtM7xpC1JEe{=Dycs(3t--_dX-Pu!7C$l%qlVM6ge z!RE0Mk;XTX{l7YDrYX(Sj5{Ls#x>$&NAj4d&ONVePzmsU>}3|aU>$9G`<9;Ud`k2Db2GAX#T2J` zEPyS?7E|Y}NC`N-xKdhEMAeJGN`^JhrYF-QX{PTraG91%Z7TQCTi@&O_$^|*f1DM5 zYuJirUHXU`w(?*uUyRQ_v}N-eM-#yyhV^O0z!T%KG}}{|Y`hjtDudQQ^{_omb$u6@ zdEB9GH^bokv1jnoY?)-dj~40OY$`dWn+?C-ECZWINnkC`CGOShVf4zu@GN~Cs?o7! z4#pmApi>$>(|itwotAndML9Up{1&=@#gGcpdCAoBc2UcTVsXb=O!aif5rbYoIA*U6 z8f%W?u<|Epu0t)>@311u&pdE!ur!mU_FNdU&xF!9X7J;gCVSF-h0P6gVdd2pG|6cI zyYAsAt}wfZBu+)-X`?)OHIiZ%-NTZk-ejoAUnIHyQ4E(fMR4TiUO25-0GrgCQTM#K{9kv1*3$pIcS=Wk+NaP+>A|(>#CO#E;vAHlWI_LTj{f&w z75!8bgfh=P$9c=;=<9=plFz+!5y}o_-(RnUy-o>KSV>EtY#2$KKLtwGAyua1xtKX6 zrT}L#5gsodASCMVgt_iH=yl^1dcbfDebuhRn93?9H~R+=Qi*-DJ_}Qt-{Ie@-Py(h zBUbUp4!&OZBDpU#=(SBNAyRtn%-yesJZO_-Oo|B2<&??IJFB6=<2kIx4X8(}1Anm0 zhs-qtq5k`u(xP;8iCXL{y6JZsnz;2no@y}yngi6x`m|@#KHGW-GhQss?G{1X2ssuu zS_xhr*O6Qf%%(Ph$*}Z4seeuGG4v|zhK9pWc@-B9>~)LKuu~Oy>k=PQ`(i(~^KJmsi5o*?QuRUOzrnQFH%4sH^BB>vcxt0Ck^SzRKz*iU z1R8YD44Q6xE<`Wm{l zR_X;e=x^h*H2i}MdXdmdjl#2`bzT=5qOO1|q+Vdz(Z;wu%pdON_|TlevZ&cOS`xl= zJ#sFxrUv~w^zoH}?DpkLlI&XrkUA`yjQX1ay*s21He+?t+jWFR23oU$+v-pgm9izJ znZd1v%1pLv1p8+8lTQvXB|!l*aLM}t=vVn*p)&s~d~awH+9s!h;V?r8$eYbf)nagl zcN9MLQ=Lug7*zJyMA{wjnk<3JM!|Kd7GYPasE=tP%Lq`^Bg?8*k(fi;IP$d?xp2agM2s6i9A%VIx6>AYW*tV9E`o>S9fT(Q=I$gFnRpHhi>6Itz;?`f>Bfs=pyvuJr^w z$~0sXt2Rpeq^f0;+qKI&&4&`#*aR}wc_9Qi+@)8jyyVQy&FoZ93QWt5W===6SW3-4 zHh9h@u$p%iEZy2@-k?V`FlPW{xfBQ{v2wT~`>H}(*4ZyD7x`? zJ5Darpu-$jfX{tQF`tNcFM9%O2h@Xe@o;9=Y0FX;ZwA=`=aAUn2Lfj3GA)JA^u4wr zd;WY6!VmT{r=C{uUvI}EUMsLpb2~Q7YnOEPcL6eUZ-Lw+W$^Y^g4*63Ch^$MQhwOs z&bq;@s|i80_<)f*yD|hb{Q|4=LYRO9%M6!{1Ym zOmTVyTTb7QJ)PC)=>f(%rrI#4OotJ6c^>k=#<2YJHTIKGk7S`x#CgE6=7J|BLNiZ%fb6`y|MBHCx*K8Xqqa zgq%gO=z_bMWW(E1nw*+UH7BJ)_snW2pX*QGAJxXXwwu_*McERKh4Of%M;9O88c#g8 zAB9t18F*spP1MvXiybD%gG;&(+B8zi;Cns^#yutMWA#Tc^_c_>`y5!qHejLNxhze* z7-#Gp#-=}-#(t$Ju**BusJ;FuI^|G@xV20yT&t)=K}T#^V&ESkIZzgNYx#qzkui2G zn#O3Wh`(@Wvfvl&gmk~EV7B!eZ8iEKJ%3l@9k0$yo@ZaB9_XG>RsU9yRhfW(l62BM zC6kyePA2t(rOv|3BG9OqB2gR`2*vz0`fgD+a(%QP73A)N0aHg1z4`-;^IOe~-;aWs zgA~c^(0H;7zrcR8;w66D;Eb+Qeb_-X~?$G0+kg%4b2Ucp5~Wx`N~L?TGK3$;5h4 zuf*KD0jay_lBw&zBbCL6$@C4EiN>nYkTC5HUjIRtbV}#VnVUwDyakHH@5(XqS(&iJ!i@qsf#EWO-8&FR;x^%97l3Wf_ zX1iudbMVJYF9q;x;GXBX_xNTg72bv5=PwhryRrMv3jJCZv zX1*J&YBCeb-^L0@CUyvm|0&~TNP#Vyw2igxR3T%=i19imCS6TApr<|wjY;RQT;|I=Dwt+8C`gE){D_9S{d61i^pB zGx3*wx3HS}IMTd#H_UhziObcbyx$ZX6d`{T-UnR8mXeX=-Yi4%-FOj9zY#^|KC^Y) z?PW`<6JOFLCQp&=Pj9^2SC@QSGmNO%hoa{$Vf1fL2>E{T6pa158ejeFK`tf+(=lJO z_|ReY7|w^I9EEO}`*;OQJyt9DJpTt>;U{t7on6Gm-$}}*iO0*v4q_*^JZ3j+Y*;AW z1TQ-#!`YR8VaTUibW!U+2vzvYGCb`Zy`=ucix#`dmdbVLro|#+;62FkifICBU%ej> zb$cSkop>;f_rus0hr!7FW0@qzaI56fr2{Z{#3FKBa{?LS;ZZtCdOaDP>I3reqgYPB zU%K^7y)eh8%)$9g72GZ!!Mx`l0@YLVz;f*_H1AwDSh!!|bDl215C2BM<1S}*bgT57 z7g>e1nnaSg{1jxmq6aSz(~v6Clp*xxDYDfm5*AH& zTMuf5UujoZ2z;=}W$K%vSnIBFkTe&8^@08L{ey<97%ns3)y@>f@u~6v-S)}a>B6%*DG8lhr^fQoX10%zoiSOj5lWIt2V*x-ItKoxl%fD*+H^PE?A;@ z?I7>v&_Uf&lJUJmMP$d8tx`^9Gzr{w6^A-IFwega;loxYKA_&p5Bilxzoe~0eO~dT zeAjW((>xH(BL>XseYBvqmLrXz&t`52hKuGq(X8k1@%}%}G)pavCQ0v=!}PKwUmb>_ zzD4Ju=}j-a&)>nG?hBx-S_QXkJtWbH8bNMsvV>e-+8-}Zgi6(sZMQ^)Np=-?~q zZszh(JZ$G$P@Wpce6*ABx$voMyJao1R9{bYU+u#q77iszsacZLiEH*0W6#$SA`Ef(a&F|4Pd9V3SUj@QCH9gY8ubCg+< zjjtK>;bf5`yLBxcY1MszY366C?<*I4OvV%XgS>#*EyI0Rf`p>EYYz6B(&wQmG+%5ZDEjQegD3yP+fXmgj&&xdW2cwSIz5lwNHW0D zcoMGOy#W5Zun;fmC=niOxzO9G7eFNK5PnH+6FyF909rDS+MgNA2Q8?p7&oE z_9+666M`3SVX-2(R;x>ElqBSjZf$)DP@lAMAe{Qc;Wc;^~pd}go--f{l{ zuKBnM6*_j{M%n>~WE=R*yuXyX*x<#zKH%Bcvt}SmIz|+zelq-M$siAv(mD-Kr}b< zIho zB%VCVYJ*Fdpx8uiTMcq7`D6yk=Q{a*u_D@xuM4){_R>H}1io@5PB4-ByEV_&;&D#a zf|}P+TJf_2x|{W=clA)x_$HG+Ic>{WshflA0TIlyE<*=r$dRXaG~mR(U+7enrc_k@ zPRLL3f+0q(`0x1=e8(#c?}*^hStTd)e!m3ZH%#boOvgkHJW_#x<&tiFeo^Xul2S@XPk`18Jis-vvrds zwebZ_7}|vm*IdVXO?i0uw9{nazcNr$XvBe?S5Zx00^O~xN|v0fh3yHgcvohvurMeP zebmdr@6NBop|2wV{HBx7roVCeD=+X&{7vUvel3b;|zPkJwGT!+{=@!n#h$t4o`;( zNA966@k?x_m=2luI4L{d7`x3@QqKJLr$?qzR_1?y+-47umyuCcmia#qLm5HFcgrgO zfY8-8e!d~THdrL$?63pRi8w(dnl{-%#`FLGC}^JRVDKtm6kC_>ATy_%PY<*d%Dp}L zYu|5kGhVIY^ZgR}7`txH=ue>_zhW3VvAm9}*{a5G4pQM42Z^~&RmZq}-t{7%Ogpr( zJy>YBiRV5mPvtjQy%X*H(!mY5SL_#>lB`BOK6f8FFDIHXM@1J!pU+9{q@C4z@z8&u@OEQ42S;{4>9Bi>IKo$4@x(<&5AE zZ!7fqJ>(RZ=?jMQb_gQ^j&XB)9&*c{L~$#RG>ZbZ-{5AR9>v{BFmfn-5yq`Ji}_NG zIBsg@Med(W0-svFly4uqLFipGLm(@k^Hwc!!u*m{!RdO8!<@)A+N4{wsOWYTPRN zdb7=8&4dBGPQ-XVb`x-mQhfN8tFw5Sv+mr3W@{mEZISRxZoh;4$P&?&SDB*qPt^IL zB4>WejdLQKth?M=?iXh@N5bozekk%Ce#c=_LkcIOcGY1TeadO5ckuiEo5&AcR>{#d zbzIR_W&Y`klS1z9dZF}Wp-7?cy6DbfORhz4C-*1YQE+Wj5ai}{a^w7Lxa% zpYbOz`U$FkGq}E^alHRac|K9+Fh8X(R6XS7?IKhHtK^{!N!`d4rrv1X#6J}3Ux;KTd_wFvI(T?gTc z%xs}^%W>h5?jhl3%y)j5(khYMO;gc=R~@3|Q=i!P1|JvQh0zXc4ea>%D(Uat3BP;r z4pb4Eh60Lf1$gv0h32mj!q>JvLe9tnVgCF~;jz0Q zy74xXE3H1nkNz}QG|{*~^l|hHE;YcIZ`g2=e^Xq{N9pb0AE^NUSIt(4v#R0r`wsJ` zeqZB*9H;UcKCgJaMdf_QVKeTA z7*6zBo^RM4$v>1Ab4Jmr9A~zQpUMR}93I&$GIH1FFXCvhVBdee~CHB;mtH9PV` z^;FK$embvPFr6>IRw5d4u!|d=Rlp~|3lr&Go5uh5Wf8Bc=geguw&Gv^!u+}k{v7}2 zo+xKk7+)Rd$tf&X655w;5nk3@6a|f2CQ4E7;=E?Z@q7P>`HeZx_?vVxCx0=8voY4; zXHAghJ6iX0ssrNrxBDGLostAD`Ex8EGQ*N{7`%nMrKQ6ETQ-az9AC*Nbqp2WEeYnL ztt+@+dx!J425IoCupdA2ji%r*s3X0e)JGXB1xu(n^2w|yq0bk+(*HwU1rGYf?R63zWq-pjr2e9Luo3fw%?4}2CK zEm&?F&m9^s&+Gcd@%nM0-2D2_oWo=uNKJ*12*%VYz-fT^m0_tm8@U<=Z2?(np^}r zv|3eUc|Vue5T6&!)F*My`7V6>pFG}YUl3P&`YorMwuP7PoXi)`3g!oxH1fOlT;%qZ zL~<*FG+2;^sQ85F z$-REA{BXK3O+yyF6HajnQyy^^rM2AGi0IpsQT{(xkD*X$5pS-VhlwNR65+@a0) zt3-%&C+_2Cwq^_8Pk!Uw7dHt#v2B9y@nXJZ#}-al-pS2QKgrjRY~;-R*YVBvw?!*l z9r(K*=lDDqBG}(N%MUXjAXxt@5Y^R7->(Z7h2U9=d;*E)EA#Sst#)(a#>8EG-^daC z*$pQ-=l4p20*X zoK?`lq%{T@z5NqXVJ3C-OlH2^{|JqoL_rsn2u+)J*TnX>#-ULY=YnV8AIh zs~*I*$!;_tPz$Rk6JLofBkf(pt;hGU-m%w|l)9U8ItO5hP7DrzX~TiLrkp)Am1{=F zv$4@sc7LrTjaB)=DR--QQkRYxY~CAPGe=?CP<;+spMb*biD)nwf#)yOQ1~_$Z8y3i zx6z+`%p`Qk>M~vFT0<>!QgL*`b$UI!kaFaXpnk#$oOM}>30~1CyrD{;;+ zW`n)MW%qqwHqP!W9ppSwDxJQ8`%F4bPtB%~qw@$j7By78f!&&!>j2}UwPK3P{#2=ZR8>iM}15rIZZOB_9|IcJNsfq{9l^usfzs{ zULvl_ntM)}%-auoVOUNBj@fTy^QE7K>!$VOGX6JNhGt=V-8OhvY)3$Y4F^UlN@GsB zO4nsONNcX$yi zj2ib5KCRbC^FSRL%cW$KA$pZBVqZ zo~}lH#8f&k_zj5~J?@rm!oC-hIArq*o^JXU1yxU}WflE$E~L}D|Z;NnfJdL&v&g( zU}LNzpGY}G&Y1##97f~j&I-uQ?Ztc9o3DR8hi+Tf!%4@KkH4LU4=3N#QN^9~pX)Dp z6l);P^bQ8xh(PZ>lkp}j6_@Ykh;{su)IEOT_&+nAwz(9ho9|(XzA^tD6T^{Ft9iUE zAIRv{eco`tk&`;Gd`TxrMB;T?Ef|E;W&v7g`UUW?u?ThY^R z7oJ@;6Me_^N1cBL>BS!*)e~dsz=oG3TVrmEk?^qls=UJb2-1rOBh;V|pKjhoDat3s z<|S>k-a3O9#Xn+&ech$S!`JY}%oJ|9_KcO~U(#Xs`;_8of@LjbT|KMmXNIr92FMeK)v{OWRNsbfhl@7-97_*5lc_iQq@uWg|I6O6dez$+Lz zLXM|SwB@$hAMonI6}H(P#CJS1xW{ZmRGr?zhs2+ z;|RF4i@RJ}kKn;y#n3WislygC>3~1IrC$=3W9fqw)Ll2@%u#7n-#(1w8Z)Wm1SRy3 zv4?Gb5q=fL@RGZPZ!m93EYfa^p}l zdiUB7e}?bEmmE3%Z|!jGxnIYvOMU2->qWYJxf{lqt4cPU?nFwzTg8d|Rvgdo%~AHh zG3AXbZkvXoA$=V3>rNoQxffg79pIKDb=3dD2-J4n$uZfxhzJBmtS4{hK;(2-qyB!sNX5hnK_zQ z^t=TfgJ3Rs*NNLwJ$b>89^6eyov#i_rdL5PaM{C;b$1;@R>?446|o-yFWmU(#@(!A zk}mp&A0=(8uUNjs2sWoclUohVw0{k>XCx9W%M3CY`lpVzuk|~PMMJWux0I$~VVwlmT^Qy^#ApO;Tz{yTLz z{dlB?Tje!WxY3m^d)yFb=gpG5xRyfd6S8bJbZ5z)@E-i4wV!Nnx{57b)!ANFW7Vb? zQdnppH@>`&K}!-TCPQG{Hh0OXU`x@wsf^6MG^wmGQLMcZL#ari5UmC69Cbm1)sx$VN{%L-FIB{gXQydn z-FxU6e}v)QY@}tIVvf>7GKDIV(!cN;6HjT{oEF}As2**mYS1|uE@^%`i0{wr#LbKB zVQ#3zG60zJ?9Yl@tCgfGQEHs-Z$rC{CvoKV8@zGle!g9L94<38(ayzUjCx1uyiI>F zYNfpJ86F|JrT>&HpZZB+qIXHGnQ+d!?8jc4!d3s#i@m*R^cX8DU6ls?JE>IhY?q|} z-ZW@htI_v^siLXdW@^#Zkl7k;mHE!ufiGC|sM1U3IG5$0`iN2hNlB+;B^pwje<2tDXxbt77yp zRlt@92l$|ul2ohjEsou|6z@}QIOUtzc(Vzo^V;Zec~81-pG^)r+r_+1b4c&!Pw3uGmf;5ztg2f7R1q37KV zc%glb)LdNP_=vE+UYDLv>LXR-71Fe2Zqm4?Q)wru(>9Mn@p*I#M|8bROS(wKqQG{# z6c;aiE_V}s+qB6x?18v{X!(#pG zU!qCDlkIvo&5v-iR|VVmsTf<*S( z+z8hTt0aX{<(xEVE_!X+Evd*c6K4ItieVve zMJR{9cjm@j&aikI!86b9fXDUU}GBBuBv6_9A_x>ox^o1r&;%&HMd25#i#AY z@P8@KpD(L(i?0qZQ~CglWWko7x^n+O9X5@f1luGD&CRVu%&GC*eb0Jc-(eiDEPaY` z%NNm*!I>~D*~liNFR)$SPdId12p>NKbe^>k2GHfv-I~yVYV2Pg!7(~}Io;j>vL;t8aouuTVut^ zsnJOK{)qS99W2dzYAEfZVkZ4G_!7_2T_sFP2_w@Kcw1JOZAbHn(W;!j;Rd#C zFkqhntulT5I%1y;XN_g;xZftnYZk@9FFXJNp;syQj0^JOwGkQT%9gi=Vf6Tw;%ly* zWZ%$n68~k3#kdn^dBMOzQpW-I(utc&NdILon)bw7Vj$T;|CQK4>%ZQ3yF;I!H7arL zyJ0Y0eU#s=(P8uGP?Egwglg3i8ju`|w1e*`s6B;#;|vXcmt9yoFmZ65fGZbmC1e(D_#_!H0F3w~qC>jJnKEr8mZeIyrbOo|&OVt~n zo>cJ{b}89-xycwIpX_<^kYn(@HHU1Mjv}XPHXPoN!LQy$@s8Jiy!`iX&Qmscn+&jM)FXI)_@6b=7nlc6tsvf{Q-H7#q zrtqCrzhM{t5}y<-`D#adRGj+@-=6m|vLu*{CdcrXtG#&1YePuWmlL_>!SL$hJ z(&3|kuTuq@N~P%3u~F#$rzUoX_v7xT9XV^zDKv)|;#A@kF?I9@NsvjMc=VoO=IxGZ zs^*Rsi5wR@gGa88l|FcTMSAGo8R^}oMlKOX(tbgAxG;P+oBF@Oz&VllGiDQ+TKEy^ zZl^V_)hH-b=CiKK{Ij%~GMjgy{pvVqy4=CxyAC?QuZggrhZlIJcpS)&yoq zt_xLaKYT#a>2x`LHGf52MmEr@(QoPP_urVX{sbN`R>4!tm!kOP4|2USlBe#^;8Mf! zkOtfFB&$BGaB&zb?9sr1P%Sc@bAX3_>B~Xw1w80mlc3z4WI4TsjxHz=Q%bs$R==fq zzF&zI{zbw~Pm_iy9-|*yyTbRiH!}Zp;X|S4q4cCPa=SS|)!LB%^)hCiU3u_WlF5Nt z>$%gvWZWB5LM9ozBm-MO>BetFV@54aUb%y9=Ime}_d>YJwD#tK z@{PIkrR^BoJBiGZY zdMab0wm$n$Ov2@|^>~pSg|f<-oRa3i(lNWx!}kCeI!5y0=V|c1(g&vco%zsL6;@A` zry=+MU~75R(8HY8=EW9%S!h4^3AocK z4?4We52~5U(9&82h5K(%sdESeC1VlY7>=8Zv{}C`j(e?E774{Ew7qU79`cd!68Z9H^3PtJ^l0fo%W06#}i3<}(@x}vb*!fEx*S~m?mEA+3xV@3I$1+r>t--4H z1a_TwoM#^UBi{W>7b7Q3!pa8`eA3f~ZO?nrrqolEkl{@$VzyH7k8R@FSVaoiGmf5b z+=;4t+HCveI+6lLz^ldrMuu(}Fmfy3uo=r{vU#X|MQ2Hn(<&%@w&V`(%wM8PFg)X# zIIHtq7>>)PVSN?ZyY!~`IYE_QZClK$^U~<{;sW71(2|O}nj$fJ93C%yLsw_a;1bir z+^jK!pE`BJkDe2td3^--9QX;zo)hppG8{R15?mc4kCLC7bUW!C#%Jt++@6D!ROZ2o z^-9ubLt^>4xgL3i4Px%Qh*uRZ5I_5u6UV-ReQ^ZeAK=5$OO7LV>?{(8j!;LH@tAgI z9Nzbf;M*_V*=x)vT(sN8-d6+I?@=B$$vh>;FEsE(YiB;&9Kp+$gz(t{8y;(?!AWcS z;P&xhu(@4=8UL=rR?VAcd<(;b&ayR^?e7yut@u>nUo5JPm*~e$rhkvjFnrh`oSoVR zQ~RB~t$99G-&#m&cPnx6^FfLH_+nw{{YNazn!r_Bv)Jm#Nqm3lAOg(FDDlr+e6()G zvFNQldT|sV&|8LHuJ5t1=SB2+Jd?%s3mh>!k{?G*k|xOwlUhjMazy`aG(%qtTV6O} zIV%DkT=3}9GJ4%BL|7;1(vj7WX}HOdlex?OIb>j#RXqO;nZ>qQ$(WvOBRX~Ji|GZQ zkQ{ZD?DoCGnPo%RLN+h{-W&qg!p?Ys0~GwPBUYR-#rcI*FsK>M2ZPPH)v!_&rp+Ti zS3yPf9VDucL+Ed3RXnWTjSsK(9l*Py_C=|ct8)Aq@-qn{!#IGbyyq z!H^G^F}(Q$JX$vJqse_(=a4n#EZd8`n@53aS6UgV%ntWs`AJR&N?Rq|G^rJZ5xJ74 z0C{NM%7W8l1Kz3X$4@q$!PJLS;9?=q<@>cbz~~?*&E3m^=bCwIm#e7ws>-9I>v6-= zm7cFy0;4u1UXpqTrI%`;Dht*h)O|f}Zdb-`s}uC(cLW_YRig_ZGD*X|C%x@bL<2Ko z$@y9?S*N_l)Zdz{Bl9>osYbzLX*#0K`q8D)!(jbi3cO~nL8J8`b{(z5=e>SYq4y@# z4v@pRc15Z=X2X&TvNunkYr|hjmIwCtF6sCVV!d9cIF)1B{A4&j4z5PGnO)Gkwpn;; zy3qLxYSfmg0Qq%w_!4W%oU|Pmt_PvI>{QvY%fVX$J>cyRcoN_C@NaNkjjonq?5WwJkjWf5o@->)%g_0 z%+uktEr&QaBZ+;UEymWI_qdQN$H!&5XwI82*n8v%)IRGV@z*I#a2Sifo9|H9XbH{YZr5WJ#M%7a~@WjC?y@j+^eIwF3rn=H=fCy6g^OU2)DrzO+t4Y<cYfHH1R~h&AJIBV#%kb8@1#Nk=v1U#sCcinvu}jwRsc!wHlxQb4SaX#12RG26 zqc3RA%C2~=W)IVTQ*bcgECpqTi|TA|ay@zyYWI1X9oSMUI{Fg9872=Dd1 zL`(4jvG(3+at|9QN|i#yeudkVc6vHlt9xKak7<}Su8HnE+=78Ge>)?=~^8UUZb0XC%veZ|xb0r-*UMjQLp(&NnvYz|LM+SHLHPwH_>cJDhb zVEF21)2M~P;^LzYVtzmaozKaorJhbOwXndw{wi3R(;sRdXVEW<5^*!v33dHsy5TNo znmDX4KBO1nQcg#DA9oV3Y6{^uAe3xZYv92sL)`lO-sbG@7@BP)&nt5;^*%|va5Rpw5Qtev$iAn zc3hi1iaM~4U!u)kuWzJVABUhYZBFmz&k?P8sP^j5X7`%uxFn5EVU$RDyO27(wB=*% zDd^X(!V@2bVakhep}IjA{|&m0jdeL#-rtJPT6E;vZ$X?JP{6vI_w(kW6ZjQ2318|i zqkiFf>hawiav5`J)^0;O-k{Af6YaQu>^5ro(+Q2+i^*0ZLHn%5XpwnhA8Y)kzw`cL zYcCTH$XZO5vu5Cnf);wjhX^MhS1j(K#VWD)=yG)^2Fx6fN%;y`l5fd>L-qJ}mMM-e zdWqYzQm((67NwpvpkWuyXsmNzI&rIz^oKekctim3YbRPKhhV7tUf8;=X3Ik+d}jP9 z5m|kdTyr#_)@43D_%sjdavw27b0g+0eTZ|5tzf^zLb#3DPcA)WKCQ4`fcJLH4fCY^ z=Wmc@`BqHqlZ^#&w-C0Z2|1q5SRA>J;3CH-udDEwG7A(fxQA_$R_r)%Sy&g1zz(-r zFdNyMI;w5MR{O13qE{kXzi*`<*-?1)=C;V))F`CO%4oUqYS=$~Ko=c##ndDdDpd0eP##`+h!qTC^xFE1P}9jY)^8nZr< z*5A5~%pDHg92-h{Q$o=EuMKM~O>nJ58>{71Fv#pQ-P#sTa=$EK(_ly24;GMFu^}`T zY@!`~#~|`VFKDeEFD~TWgSpofzNfnq`&N!anZ{1)$0@?Ks9chjG(dEI9*vnpuW%7R z;_eNdnOc;%^KENnDXH1Cc$?WS7EO7~DQ|2h-nw^RvpZ6gY7R)EowT2Xo|7rx0#tgEJl4xazPj)Qo` z$xkvp%zz(jZ{Y@6pKaWAEoL3>$US6r$H$IW@3xSglNWP#R~O-S)sqzD z7QjqsdNl z(5Z4TtDpHv*2i8zb6gnjHU3F~9rB30)iCaT4SiT@gSG*`XjMwNi0hb(9~(5dB48;- z8fM{e;c84WPNo4jcA~abh2{01(a%Q~NU12o!oEEzduTC5f0_tS#lCoCbb*tzjrvg zx(&K&!7w;nL1$}6VTy|qTbbR&jBGic-?$8WJdOE_$3xBA!q)I=Pr@zu<%?BhfE77 zUzW|R2FF8FM}s@N+OoKu3G=bCnU`Y?P24evM!iwzE9-hNXUp?(y=$`HFW`32 zZ+V?|cWIVf3TLKXqMVvQq+UIb#GxHwd}JNXxTeI)%hyqV2YEzH{{x+fUI>=ms|xRm zaOC|>vBlpLt^c;MRM&yt>N+4t`V4A@p->H3huCkqXwdD9eRp@E%<>05E#E=)2%~zL zmyE2E#B*gmjvM`!n+z8Tvyt_}e5fil&8e2S#CD|6>C159$qDG`81W~oPCWQ|I4xKp zWiP!!oECprnBAK#K6Y6_;rea##OFNbb>5Aorj>}lw28kb>|^15foyA&DRE(qc-*|5 zvRhtY>#70VY#k#0=6X}z-MP@sj)iWF8#c&4B3T9!KaOOUURlk_XO#F#y{y-d8p%Fu zF3{n58TjKgfgNr}k;|{Q^kz>cto?1Up)!Z$s{3J0Rd1A~b%K`T0X#bJ2S@wHa?2-2 z*19&D|4#pj>BbuDck+QG;82_x^?5KYSKNlMA5SU#Kqpuf?BhnqKv~XPpmfFSLDIjY zhe+3#T$Y}c-$s8jgD_X(Jm*yeZ~3%2EXEsjX*`v;@@nzHxx&)7f5 zf`?oPwY+NUgE%$W%q_~9BY*KeS-G)D|Pr-g)>5bB>EiU(VM2+J>n zQEB49Lt>|}jmtDP)w|59!TUL>)DbVl{7%a~*Ei zdMU<548S|z3%JpBJ$#nf@Z}{A%$2G%aqC{1UaJk|{{6&t3wxR|cq4v3xq%TSMab`_ z$_{T>thOG47VEosy>=#S*Oa3QFR>wXGYZsqV!z`QD9W~ z4bJ&-70)i6#O;C(JTGgmj8uuVYR(I|3n@FRkUJ*`7TWPd! zM=bAio}P=oa8&qA-ImIjo>f;S}8yFZLKi zb?7U^tn%iK4iP-!rXlUg>`6u&ohdH7SajBjj@>&$@BB4fT5G}i&L>D(Z-|B) zfhZd=j9)}gW{v4iocitv1syt%z*1#y9+3dQ1xxYTeGO8)JIJ(`R!k^xWtS25d`fF7 z*7VZnwCEpra50uuMjYXMk0>l!p~6?=G}%8s43i5U;cew>816eo`NfNI_+Tl5&t)TH zmM7-S=R0k(8A%>l!b_TrBJ zbeL;ILfg+AVyX#Me@-CHR}q+@GKHFc4wHGXqEH!W%Hs$6a8uw|j{nsn>@G^E?%N;? zA|r0ocH?J3X*@wSj^_Ahpq5XgUBylk@~Wd4=J`(KEWRii=h%@NyF1}zt2Zu(l|l9P z4FtMb@;kkqY$_kcmzqOu4#iwWT<#kjO-rOFx2h;SM}@A2*&xVdAL&tZf0!kRoIPrcr9Dc^JC_BilXXAP9`FFH=c%3YQMV{9h{~*<< zRrnBf9j#Yh(wr7`ynkB+|MfGefc3DtX#&pln1w?P?@%PmIPZO>2ehSU5OnM=H10;= z*N1V)TlkGOw3%|NlNrwrv13>Ni*#k?d|KY^3(fR7fHk}K;QfHf@ajh#Fu8`Mk^&CR zGr?zv6kcy+B`xbW4fjU`aM_3zTwi__r!Q-87sD>RVrw&P{2C^E<4|^|{ue>9lg?wn z#L0QUMDi`o}*I`)uwS;_h{Ha>@jlNh_ zBbJ>SFE&&sQNzb_$&B3}Z1!qxgTuKHIxih8p4=>zo~W5BJ>lC$I`BgVHZ2;BP~{?0 zUNVV(*_cTbYlaLi?e!YziI#Njj|%=iu!HthTWDWvqCOuY(L-Ss=vOBg9)Ez0n?~5Q z!ioksjzDtXZn8ORA7Lk(A3X^9Lf>92#E;uA+2O!gR`Pg=HE+jByB7A88Vye2vZKGL z%PkiwlKG*(zumzFp1pbPcL!WvI*`}zJ&zd!4bkbF40urWq^XT2bbkFw*1Y4yMK>ex zY<&+j@6p5kkzSPXZn*UIK?|vK_nrKqmrNJa38TW0WkMw+3PTc{sU~QIn0dSdCrru3 z5f@9iay;AmZ|2^i1Gq81qjbQqSNxyXQC3p(38DT|%>cJVC)o+(37 z`((JKyr3)T7sO$U5XlgY44ad>;aHMrjo6!Jq?u>Q2`5t!d?k_`FOMdj9!_LgWFu^^ z`_jrDd7@q=k8(lloPZ^WsmBxI*R+>9^o2bZ7^!mw9o7tRLK8d6)|8nHW90Nw>NIaGiO*;F`;ryB$1@p)GYdFxOC0wu>A;`%{NS;g zZak_n7pnDV;J4HcTT?w~f3bxBeeq$h4FmW`r#`$rItA(TyP&zNAp$~U%_IRYEfaODO@#n}e zhPO43T0e&~9!hz(TM@h_rK8|$KZGvqhxW^H{7&A4MYj^!KgLZP z)TKY3pWm@pV9|WMO(JTym8uH+_bXI=a)z73Jd2?M{^F zx{&8iEP%A}Jk1~LLwEAtid7z3q&Vma>1-QK)7P?%Nx@}l;+QgN9|MV0+qMgrsmSgu?~1m`H`CW6SHBm|8&C4ZlX?Z|n7RQK+?-HKU z(wEn|TmbS_ctYeU_y=Xs*lY{z-W5xRoxh75$sf8HY=r}^x^U5$ifEySxPOWW&bcD< zo`iEi+9+Ocd4%T{edf`6`#Glg9sLWKk1k)V5MftOvvZPR2@Oc+lG_8%4d z>NIgE`W&@hnu5u16ZxOPWd8o{GyWSma7gD1wbH&OW2IA5zjEW?e>_zLan+m!B0pP0I2?L|0vAv z6j{ZC@WQ&B3QozxZ&0OV$;VW>|EYv_U09C}4bAZS(F4_*Q5d#L<}aIP#!fxAk+Q*4 zn62o^$351N{f9nq4vxT1qZd^Ds4qTh^v1>3)#69e7jpf28vPe+raM|SH2;!}N%I!y zRUyH$^(8dIp+Dv4w$eDCc~}_wnWF0gkr`WzZD%@QX;%}Z%OAj1t6Jn$UBlYxGfDng zAa38uz%`GaxOhPUN2Q89G(H5q?j_^#kTe{H0=$pa;N12keCu)EqL-==)#|Ep;6r>rLj1mlpl# zNn%(0&^Q7M8KW|L#Zd&QE~URsGpT6Nd-|K?$kDQzaKVk;ys5hn?8d3UZtx^FPZ)@} zk=GG^A_A6Prns}`CKW%Ar>PqvgmTj!nq_T)nIkkYR%<_w+|P&QMMZjgOF_0zcR(Dm zq^-P$B$8`zJo1c|gsVc$sRz4N4aVhBI`B%{3MJQcw6*Euxyfm)S2o6_xIk1p-yka_j5_q zn%D;bvFPJyFqXCWEb8J1W3!bLR}Ejwj1n7>=->ARzp`1lx=ugyVwX#n|r zABM$l$DnR!&nEkIS^0V=Od8oNVprve?#XtjbWSAq09$+wDMF=cGtDTPjHmIY=#uFt z+5J6LxUMm$k=^7`*)ofa)H5l6OfMMU3B~Ot12JLm0&J3YVuOeJoIfYlW?GdJHYBKU zV`QG>)mJZix~voav)GD~KLX*~^kMY)36g#8!f{wvLUyD4j=iL(w_HCu~YoEw@v>KN!UxhR0^w4!)7No0FQTXTy@sul6zR3Yc z%2hE#PjG;NBm1_=>XS>y=)=w1V&qIep_H2`nQ=8vGBK+NCvrl$U`q)!!>&@m=S9@B za|8XT?2d!o2V+FRFRUBvP7U*ZQ1Qz-*ea`kCwy4PAH2qJxvo??Hh-y9dSC`C?TSa3 z(F1xI(k?zu4<@}FuPkfE|>QsKcp@CN~YQt;a z!(y5U6&W$_==_#W!XfP}nOJ4R-_00bo^MC(ziTL*_=&uNJK=-#P*j>0kPMRHQR|j6 z&riVkiRbX2>>N0Fb0vP)mdW}DC5$`y6Seau1&^|I(fKVQ^kdK zS`^-Tg<3rKQ^tK;O!;<0IOxovJC!jsx4n&O)Ge_3gR`Ny!+|p|{LCDb>UGVsOc~%UqV; zS&W?6f!5ws#l-A>$SCZOt-SP%0BddGMo;j+#nYjdsO+W3ob^7K)b_AKA6!!Hxlx2 za6}wR3fjbS&llpopFL_OdST+6Zrpm|E)K#^`Xo|cI)BvyY54J>;$+2j@Ty>r*&U7H zDx3MEo;+_qS}B&g>k23RH6p%4J|4?w!ai7n;iLOXgXE9$oY#pwVoNhy>h_c_Sgt9R zORg5j{#aAnLnS(JE|gYQSwYQNkxeq~X=AW6zKX4oo>ru5leUTw*GL)_t0fkitiYk^ zgZbdWOrf^O68<&5c)oBd#s%bH-;<4OX;Me`ANIoKH6|!rX$SY$%lL4`D*p5KIt?3m z15004V)_|3$t6`s;c#r6sB*8AOs|_uAD(2wixSaL9SKRTI+PYTV!Oq0uJhZ&$E&rW znB+qqo{r>pp$C)=DxubC1@2$JN)^^hJo4LbRDAv*@w+4yJtMv8=?5#!jGGP9FJDEA zlZ&u>(-)aDWAP>W8KNnWovYmV@X0%*ZCH)h3O)Ii?=cD(n}FCw8_-4bCG`uviCG61 z@QGYs$_$O9oP$SX{8k5g8ovvsseQPxcdgjEyem#MzlKpS4c_^w7yDfA!@DhqV$Sj! z6cy|7oSsIK{&m&jafBA#zjGco5eE1@UV)WArf_?H8}B&3kKfOEk1cA>JWu}_8wc*B zL*8;ExhTgAZo*9N&2jYohSKy!mEu$SP06vT9WZtMRD3+1PjZvf=;!KDWV2CL z3#sgs^p7MBv$jTOQQ6$V0y@o zH7jh`NF1ifsWH^OyBwSzK863U?)*N{k@vM5%GlOW^pa^@dnYWA>Cdy_9UYChZHKw3 ztcJHv*vr{Q=PBDai{{MvM1!_10)L#2u(b+k)Up=U8x}~Kc)!@|VI`~6j>CIa8^*X! zVE@*&Z2DysCq}r?CI2(xbbA49nRo=1FZ}6^-YhX_Mgoof`jYZ`zNQ;~iRke>R%SiP zhpVkVy4^QIr;-q?Pi&w!-*#cOY)1R~;1CSeC)jnpjkmQjKUz=*Ogzp(tMnuibGo9b z;t#og+(5Y(ev-r8t#~|UE&Qe3&6mvstQL?El%7;O(3K$#^zP(7vS=EH z3kwE7?`;Gc^f%CQSAEjZT}|ehGLLeTDl{L3W4?@wcRa9%B;#typng4PgbsnXUw8Np z_)2=N=P;=DHI};$!oDS;6m6PIW%uQ<^vNXjp3)nu`b0yN7(l1!0@Zk5B6_6`pYRy? zW-7v~Nt60^{!c8@KZLrwU+B{6O%$>^7>)BEp-K5UN>dI~j*A}tuKrH$Pn)qfXBFl- z?ZU4z6MTxih@m3`5nZK%adBtKU#pP@d>n&78N;4`c7(4$ZOdt4e|{OATQ?iOr+Oj#r6+Wg(qvj*HrLqRV!yz0RyHt` zUcF)^EmXY7Ted4;$ZkE^InhA%t2)8QK?`5MrQ_}5<>dDA0J$^~)k2?4qOOTUrf-A= z?u)+KCNf@d3oYxvj?CJ%P$|rjR&2~`-{jzKRXWQ1`O-u^Q+z#ni>z<_5^G!UP-2z` z;yq=WOv@IG)jp1$k2;HKIWNRn^<*($uB+(3beH&f%Y*)9*&t101B?w5@zql{yOweF zWmi+s(JX*p$nC?2ite~tG@4eOe=l-w%Tr$2a9NE~jqCIO;-`WQM!5UXWSUQ(Q|8l} z78P9ZNTL%d7Bpd39!#i>X?Cdd^L#g(?0gmMl27GrQ z%6tqyPmDl{OwXT}-bfMc%Wn)II+`cKs{Bw$x_G1;z8Dp-1`4sO{{uU>?7`7Q-jX zJ4jXbHHjDJJ%~6IE7T8*-zul5ZA}!qoL@rQi?XTC^)g9@%0s=~0oSs}BiO>5hWDFH ztgavEgqmd<)HeRH$cc#fo z4>>lB93y>ho++(LpF<-S0XmPbQ`?Lhdd{-B>t-F+J6|p-u5dunbGIE7*Oar(KfP+h<)E#5?MY$6wH+|9Uk)VxIJ75kGC}J z#$gy~Xvw%sXDnAghtGf4iTM;ysrwh>>Rfxt%^A@)hN?UY`kCyJGKQ1vuj{_C4d+k;=z<0Wrtd2S7L8~Pi&F3U1wtp2l!Uzkc-)fcF0 z))iVgBMkRCUV+N$Jjus#8gw^H^M1(H-ax14q5wE zb;#}JiPHPY)fBz$3PrRnqk9WF0~3p(NwM6YuZcfvT`8ntoKVqvOeHq5zTaye-u^R^ z&Rg?_Q|p$n#{gM=$2xQAHAOe+9Jh7}Cl02v;<=Re;0axgzDTVu47Fn~Bs&eNMENK? zN=|B|#-FJa>J=av+N0IRYwie|aBU-Q=_W{WW@L#KCl{m8FdX*ZcHrRbbPU{jQ!=XJ zH%+bD3gg`Oba+A;y(zB1j4NppiRTrvyQsvQMb6+IlT^g&tVyC~fstg)=AV+SCu7A* zXA7zb!CBMY>z(kIdX%>|--kOrEt22c~9`blpeseyIx19`1!JKm74#t}5Sn zm&YM@yhWtjU@`EORD@65F4E5^;o6^rSWs}!=GLHSNo$ritlmDr@Ut^`>FGEgaaGW# zuajw4{bg)jJVvtB{+)OjBSCGn+?}ri~>{eBBW*OWP#Frzz3C)raWLt3dAYZYy8% z2ZzX5x6tjod0p3k?A|RBPR8M!oUoqjH|I%OBtKwP{*ZpHHNffB$}n6eV`LY~BYeyb zvaIv@ceK8QKh8Fojzkr%*>ql4R-X!QfLQ-Cos&CiL`&j{I!O zeKHu0DKn72BNX-z`rz=x0vaJd7m?ZmS~3Pp!=XUf9Xc$EZ+Idxwg~Tz_2PcEYjJL@ zFYj1&SjM1aju8W$XymlQ_K?|BbMT=g9V%T$^LiQ8GLon4J`j`hn=#F0iOm<4xE>R zl^c7YsE;1!RamjXaTQ+qXAj@UHg@~34PtW#&Fi{ zC|)=q0o}iMhhqADap3ZI+BM^%MDEWZ$+H|CzP&gFeI~b4!e@EDdp`+=Ck=7-`()nr zH5b2(y3*kIVd!pYPk~8y#5$Ru)m+_0x?*^g^wRx4(#F0!xTSwFu5|c6N9DmAy5;b= zBl>8sA0}x`|0<5=^yPQ!-Pxo>gKIa%<6y@;glbp9Jar)tP>kYlwQjs{ggH;TxS02B zoW>ZyX7l3C6Y=WyKf3E4&T55G z>|AKi&tuMDOzR{{j5ijmwLeK_KDjUCR?ECNdS_W{)G(<*iJ>%3QI_`*w4a>%E}+q! zcOc|&CHAeJiJw0L@Kth>?k+cnq&$(&$(3PPOCh~J@fHm)mh-^m6WsQo6Rv0Jlk?R5 zG(PwUhu;A|*xw7b@*imFz|&+q%@-BvTktMPwst1*STsGBSJ!4>{g*Q&f87EH_sc;c zTN|xIXG8Cz6W4ZI!LP>e#s>Qn6gprL&6DxN-&PNYFzPJE`sHBt*`?I6dzqx}z)afL zW473Ja*LSTa1WA_A4s+?M#qttsPn4vv~!di#>#q+a@jgnseckr`l#Tibv>4?Si_fG zt$FLapE#Yd3l>2)(d(8x4S&%QsY-8YM_LSWlw&cIFRH7-Dx>3AQEfSww;a~q8 zQ75bI4%_wRshb?x%-@kMFDUY()O6?=xwGY)k1)~lg8kZyI3Dw!{97GK)qDegG_c}Z z<0rvG494sF7(TE*o4wXTShF2xq?t;1b|PjXq%LVoSC8=5{9R8|m8 z$_htF-B}ZV^a}B(zX^LR+k@ty`zUw+CdH!nTD&m$`4;5)`jEwfiP+n-0{>&^y#I3U!ziB2v`eHi(vXPi{$A(0 zi;5y8BOyt$GLs5f87)PVWHe+X2`Qq4jIu%^A}h)$BOaNV&*u;5hhF#hv(9zS`}8ht zj{$DF{OPj;+m`P~^J-;W6|9G~1^TQRqQjA+Gufg~8`96*P75p&k@dx2cI&pX)P3(? z*_v*Zq%SgRiM3j=nE#DFgg>F;?akEPCkEdNwqVi>f2f$ML20{$6}5GA=C0@$+}4-2)YM63SAI)1_tR*1 zW*+tNJU}-Ni@dniJQ$7XhaJPjH*zhF)J;<8c_;9;nQFYNc_kuNFD4~+C++&HQu6I9 zRIx;8U@k{#mwp=je)PxwRm*7Dkn{9L+*1Y*Zk7f-{U_c|9WkNgG9~PKLC5E1QQNfB z^r)lIl~2CF>dehBe6)|M{EpK8PTOb|Zc%;9A@bB5O^OYf)KZ*EjjQI<)-xYaFt-#% zBjfRMW;ta4g0N$nHG0|hAnPD=>T9u#3jYj*zqKMt)^)_fXc_gi{wXD1>qvtFvZOA@ zw@DX2F)iwDPtVok#M$pfZgW@CPt##!Za0Z|Kp^QGuBA590);24PI^~0jTUz(kQ}bw zB)HK5HC^CmxYFHA03?F%_uncmV4vCu7>bS#;~p5_&r>hK@a$h{WbEH2C=& zns~z<3KF2Fo<^B^nYeOyJi=>r@!{J+Bn$8M3v0qY`ANE5@Q0>DfGb_JC zHSG%wV-j(6ydr0|D)FBUwFsR$i_YrKpbw@_)bHMO#0f9hlgr()^uk(l=;A<)PYo$d z+W_B7OwirGl?vLW(@>M6bnK0|qmJuK2eLQN`304dqM|N4=$o)C=rN3o>!bj8HL>o$ zNX_4aFf-{kHKZESo`>f2r*szGpKwQ7diAdCn8<{+@v4>CmA)g#W7*UazLXA^*HX(c z1(*pon`&hm)a@Dly4BG*|@{=?2uW1|f}{P_rj{tiIjiXtq0Rf6EG z2(0h~suDZHdHyN#cFo08uT)&rnS}YHbEL&7wUWutrBZBNTkdo(3$__f*fK?&RYy*P zd+mTse=RRwwUd8{xvxcKF^cwS^TDqF5dO{%+TnX>mrRj17xpK$pa^nJ@h02WaWwGJ z80u7*L|1*f!G7u^a?ZU)|Fzta!s|v*^O9?lOVWIvUFX70QSCYS_gxAwJ4wB_j3>wT zs*;)~%d|}{%32Jr;N(m_=x;tw=n;TvKZ|gpb0B^#(301!h~sDLuc2M*9-QsAp58gt zOZSZL0n=-dXPS)|$Av849K?gmmWX-aIXbd?0afQrC)M}WxEGm#qK#s{+i^3k-mVLs zF>5iy-cr)`)RTtf39e6X;oF<-j!nUeyeMEI#s^-YQ%^N1^ngD$u@clr48r1@PUv5> z1>1XNBD_&(>(5Gw)92FcA=9L<#WKpxX@lY$Iy}5lo!^9RLf(x@JX_A3|6dWz-`d0V zuog@PZpOWt`mE9V7~Q6C#l%)0^fId=J+BT_^y?C)6dpz3HfJ=4=~I&ad4kXe-k-Kc zr=|O$vcDW7%a8F2rP2I!gF9D_h^Ec!-ALhjAi23sBdvr(bhq7cN===HH9{BfR(+V1 zqmwDQ??tJyF<$n*=^2u=^`ZB7JDH#A2D|4*P!^umGNWpEmnfrQgfAtS?4aeT_i1$0 z0?NEyz^dVUm|QYgKG%jTSG4CPZfD{1Hw&ZuKfqQ0DwYYKp~GKGy!hG{jaSF=jr~LT z_2m0_wdOm;&>ngxw9*~kUI_M>h>=Aqd@y1g0{XY$&dWCG_X#9*QFK-BoU6*-^gaL>swNP97CKeIn(ZE|qC7AH&rPyYkESv!R(TxO$t! zo@vov`sfgdKYzQRe4QpL*L7vhgdeDSCHzF{RtP@+fF94fL@!gf@ervqD|tW1&wC5e zG3XZ^S@A^b{P!)b`Ky6FUnj%0L?7k7FH%>1bFi1cj9U2EAiUxZIog0&5-xV}sHM>cVr6SkbJ(}tJN zc>(RfJD77NLwYhuOR|@il4kNquDxH6#cr!*$8J0pSz}AKc-xx|!;7&(PnDGqKjMZN zB9rPim)>;zDWy10Wm6jm?l)vQnHGKE4wcPVUaTVZ=y6a|i4T!(1cXcKy-lHhI*IO7 zL`mV*xn#ItD|uz>Vs*<*9zOXbcf9ftj~1!&tXH#9Mg64OQ{(Y+^=rI7e}sA8d2$K6 zC8>#7TbmJ$NZVtKm&J!=cIQr`qFje9eZzTCnRo}SUWz@%S}6Ia$bs8$A+nD+Z-hF%Q;IpUcUqdok_w0=kiZ9F`vQ*rcl>ZV8{7;;blKb!fmE6FWX_qRKxO z>yyepPpJ6hqszfW{5|&zt!KW9b?45J>(ba@p$ShMwTf2_j^Koi3jEft0P{BpHtfqJ zu9T_J$S60-xYUD8E^kM#{toC^FZh!0M~mET1bn;Lky)}19e5fz^33@d0(;&$QTX+k<;|*;W%e`;ti9J=9zm&M*?GZD?2JXlBv19n`*VDMM z(Gr2Kza`a-G_)+|9LFsxNSqySAG0?8;7y4wfX*mJ)njDob`S) znIs$I-0^O_uly+5p7Ox~)490+Djc4=zNkKC2zko zu|*0?{v?HUo=u#lM78r4OM`9gsSZ>55AJixBu#lAwNfhFlSsZ98iJ4Gg``e~9Q@i3 zs)8SpwA?9oXbC<6d zg{i9k&@GvdzwVpxQ!qguE!~Eu4F-0~xZ^Ps_-N8>mp=JY?Cd({caO2!GA^FoO| z*@~54ES6-h6;k>N7p$!sfo7i_l#~BR`XPLN^wWmN8_(d;zu(hawL7F6G8}U^U4!Rg zYZwgPO&d*|c=ofcysv*GFZ^=?Yj+3^RY?SyepbRw!Fmf#TFJlmsBm&mH=baUCh|8~ z2y~Y5%avt#vS&E*x1WSv7X@yTb>g_m8hpG~fgkNR-W{_Jyphh7 zKahG)bYzpx8a#22EB0rrp?$$L()&;=+xqVg>ubczKf2_~_vl^XHsw#SX$Hve;4Pfh z2;!&xcCkU@Y)mg=y4|Ki+GMH7y_{z7KiMu`9)AKmE?t1D)h-jDzi$;I^2RKq|4*xJ-|VcDgN{B4i7L)=e?#gIqqpW)GHck z<0P=E&}QuaMY5S-o+Tbk0G}?B3`(n|`Fx7TJGbXG8{hHsEhci)#Q)g)?E`!pIu7>7 zbg-bJ9^+z0^S7q&sJV0!jlD+WrH9~)G@a%bx)S>}WD9+`H>+)H;z2s5a{r!YnEYHF z55v_kH!hwJ1@Gjme=eeX>qEK{r2|E`t$3}@Ve_r4zY1(+Yzx?H%|>!6(OV%SPmWS>8PIkO7brLhAG0I-J#zRN-^I~ zL(8qmb>e&LPwm36ZQeKfBz!~L{M3;2_Y$@CzDr>T+pvE7l~~*Duyp439m*RS!yaSw z<&%CY^No%`?1-QAci%6}TC^DxOY>kbsVxd}|DtAFIp+Nti#1usP>L*|RF`a+xu>A_ zjZ~TGcV+lTs$$ClXV}dCg_U>PvDMYrf=$?je3!kryh(WF;|<__(F3l65ggS!nnw>< z#!eG2An0ZyN_`buH4g1;fpW&T3(Dx`> zC*)Gz+n((2Q^cdC-<+O4gZk`$NBF*-lAByrc4v?dThE$^vdoRLt690Y zo}PgV=26Ib=fS>8A5q{X-u$V(d3JRW8tU?-S{*g!nrx`4nNzFVVg8up%v;=_BBXs4 zw*KtME!*N?^B{_%?)4+B>8bQX|D$wb_9N?neipLZ4Y_bUnoeWiM$kx`B)mB3fbJ&$ z;YA-sz8V~b{)M|GRgZ1>(p8hKuNUEJSSPxe(Szn#gwmN;`&s4TY%UtI1U>({V#eR@ zlykH}O1sh@e!ZiqD$YsD&?F{Z-WWwWXZeqwxPuRO?@WA6wxx9n2Tr-Y1Hd>np z&T(PK>-z*_@;R>mNkg->U;+MECcWGJS#nnLqkWZxyqlZxHcLN5((uIuxc z*O{=)bs~pZtEBTIEoH6kMc${QB5T-PqaV|C=!R~*)Ur25TIS#^f3-SJ?)h`H{LsNZ z+@;T2OwYO}S-C&vtB-Pcv-5l|eW5MQzLPG~|KKkzioL*2|2^mRv%R@^pcj^ZsU+=_ zSIOy>Bj%0Sg;#GLP~;;6c~<5PcG=vYhc@d8-*^cYCk06bYH9d+eF_-P`f=-2-SuP6ln+po$NUf+4N)3?qws z@XXbE@*PEE<;~`va_jmZq;#!6jP%-Kp6L<3`7)o|)@t%^m-QIgkWG$1CP{f6XF@$} zBHXr2CyTK0SmqZ_cdIq&=>AvOuxMVU+++R=qj1=WRw$G9ue)xtB`;Vbt4TjVZ6GzXbER^M+kzOsRqu#eQ*m=}k z_ITQjJ%V>pcZ-oyf;^F|Du?padna)I@ptLmkfY)qGzSF(=U~8xT|7{I0w?|Yj8Tnh zOgqFpa%PDXc~=|J0Z-{hMS~RTy@n&J6nMM!N#Vol44<73B!|Q;bV6$kDH|5RE`1k_ z-BfVtXd|$zkV7MtIj_`{-`9VIeV2#y&C3{@y2{Awf(cZg3jWO92dEssiQUI8V?lA} zrNbV}rAZUzR%;{qUDiGvIx!pb_C_PUMj6wm&Vo_oQ~Gji3zg@p;b3_d{k1+!+f8do z>(N4bv|Dfu)Z#GX_y+X%T?pT}(bzfiyKIdAV7}e$FBk26$wSma(a-o77B<=Qxu!g+ zG1!olx*#DWPO|}%C3In7z4n<2OmAhfQjY`)xyy`p~7 zk2{8V!fNfYaw&IIy4;Znm*# zllzU+#vJ5sMuxn-&X<=&B(PeK4ZJ{lg*&BH(DO-y*6;^hS9F}$$cJ&>n-z3NdMvY; z(;^!(W}#Fx#a1foGf>Lh#uz4ek`7O=P|x#fh>NzSXm1y(FfAQheTU$D<#W1j?nCW+ zXG*=&vZz;esr=_FH5-T4tMch-RqP};W~(ke@nu*YmsvOQlHdE-*6K3p>ujg9bdw5t zw4=*rxk8Kk1m%(>wr^D6hou_$=9tUoZ^!YozUo}z_>lKm-)7}&$sDeA2`=jfVS2bS zC234VO3Of4hHD~cf`i=jwTXO`%YQuAfdnV#1){>2!gW+6za3e?R-;@vx1}6+T4zW# zeNwFhDv#6Kd;Yxfg|&R#pri7zGB%HjCJHsUv~4zBrSRKXKne6w6-`hViF#-bAnyVZf>hJmhXu* zx@VLUr})I9YJxYWYINrZ%fZI`^AY<~hu7#`N4rK3oDiH&yZ7cOG-p7NcgyZdjQPLSxf2dtb<6aiR_g1_O$9_K|n-3Ey zcDy349aqDEE;=Yt9mlR_JGo@r44z@1E0sx(lZ7v;|1$FOAdT~}6lWfgshus)@vh5))>UfksuTq5hVlVkt zr#|xGe~xp-v&G!vcZv1Itg55`4?ONPpVODS^N64Sai%<&dnzx(=ROM%QK<-n?w6=! zxt{z%v8Md~n|I6y?U>|7yvqJI7W(_k$BdpP*IKA6Kbmub_kVcH?n}&B`O;c8`QU?; zHN(jH#AoW6yB8B>yD&Yzg;FZ_!uG%+Tp!p;XaL7h%sM zufQFj-I!}DC*`9udS>a3sbIYyhZ+NV9RlAqq;9Mc^ninFAnaIYKo;*3RKcDV2l%uxw5*#;U*vFnhNrefg zO@Bibs=l~F6_cHHyp0FcOYW$b&Rq<#T%C8Q@2(B zq=NCS(#}clm^2^{szv!2QEo`_x_zah{7`C}w-Rm3JMiMKCS2P%1*(EKzq7U%to#fR z?U6{G!nQ!|Vg^}idmu8`o$Ji;`VH z*2d2Go;2#f9VzP2W@@tW#+U(pIUsW{_cCa~EZtmuSTP7EIzFMF4U=(T@nOvF=nRdt zjvTyQTlm1+ zmFyN9Nw3;RQbKuu`cHi%B^W!Ss{KsN+FFmkgA};TvH&UEAVZR0`~#Or9^OVt*h;vR8-2opk?4&1O-W0pWIFB#bcofo2eVAUdSeAYoeX7Y2+G}+Je?gNi*iIykp@1mSNx6r?R zB7cwUf=_*Z)25Ii&^-5BI=;dXZ`G3MkJ?Gv>#-N1p^u@yqeb*9Wx&aQJLb$?B8~dE zj}#7V!TrSXa+`j-@|NNcTzX!cy(%hj^!E*l(rT7X3)&;?UiwZp$7(na(D;TUb^b`X zDG9&9HE4R&KyPgqzFGT(pDEXIS1)Buyc_}R<2x|<{M31st})k_$tV z_&~rS-jTIUnw%oTgwU7hkR^xy$JK(fFFMkGbcAF5PgqXt$U1jQ=xO~pjB1k$4U0o` zO)xY@MvJ-Ul3P?frvM$(J|SH834hBsa>>gM{C8g=Ic}JX5jQhYqSpqe>eL^<~v zSBmZhitO;;Y3S-2BRb*=F3=5}uTO>6wfE@iqKj0pHl2D*zDz@v@6zN`cd#+eowY?bRIkPqnxk-ylpl{4TF50b{~Zm_?m$H6gGj8= z<%nx9plhjuM9)ky6V!mt$OM`Fw1pU8;KEgnWz=hLXP6Arz}2)ANVlsoRX>aNzS$0K z&l>n18Gw$H6Qq-N`=mR8;gWqqEA=?EA9k-kp~I^%Oo`fvyJt-BP+ynph91V;ZZb^m znk#Eu*@?gB1#v`<6ZU%LA@!nQGHp;`m;37clC=fLrcPS(yOlh@nc?j^EtF<`q6dZ_ z$ucCL8b8m3{%*0S4Ej$xXgPt59_yo4yzN&uC~%s|d8}`1i==Bpo1T>q?|C6un{ynO zS4Y67|7;BK?Zi&;>CiFOlIbk$L^i3Nu{|kRdNxXAt+u9M?{9Tje%%NuT!#m)8i~N8 zO5Ax)fMD2-r1s8dDQWX!nss=u^rS3{y-a1O@Gn)z7F}UmD0kwcgyy#kjopWqy!x~BGCXVo` z+e4>@FGlHak$VY?mT5VMV#0k>zI^`^O**#^irFXNqhxJ;Kr>wmjaH_vmz8+Ny{V|r zi@^LwZAsl#k@R1$CXaRV=~Tjd><$g%`{@gDTiuHncZ%n;?{pCV)tK7&1k>qzWy1GS z3!iyreA~9GEN;UKI=Fc!wk#KSflz(=qW@Sr8=5OKZ&jt>iHc;Ezm$in`SGvc-EdE7 z9fzAWV10oOtj_yOcVjGYNiYl8YAD{R^yaQt`|)4XApBaYjQ*|$lHtr{l=^ifc5OHY z{hR{(i!=j6affN`GnR+6-E~vAsCMdLriaJoK%+g}kXIm~=Ex_U7UfNv3)l zxBrai)VV%%`)MWBG;hW4qnc#5DV;(DNlySEulcnL8H-qoC;nPY}$!%gU()@OVmKDsl=0onX1qsV(oX{se zwyGj1$_IzdbfpJPD=EvXiF&D;NP7!s;KiSRc)T=-PS~Bsu=p-~YiXUNR6m(KMZRaJ z%RE%@7${go(#vB@q_i$+ST-e&y;gLQUiLC0(ad3=}$9^B+q&FZz2F4zW+5B|C{W2h(U|N-os~v{1|A5wx<#6;6w1A+lXCjmq^P z+n8+Xv$UDkZK;ndmjd3q&@FGP_ z7`cKwzU;wGy-Gy3vj93ab?ALh0ot9-tpntB(hG+sI(PAl)Vi@NAE@2VI-b6me`YZD zW=*Dw`fN@Jw&YdW(d3wU#d>4weNvj7gi)L8C5`Nj()6GdNk`olSI<48+R}Pyo{9mz zJWwq%Y(dy?Hxp5Pt%Y_pgu+(8K=bKdd{EDfCyyS;TB`!M`o$=o=$Sx0&d^;faijK(v1X37LajDW@EOJ7EJ4i8*Tf_cAhtsIjh56q+e%>yi=>?jy0|D`j>qK|(xUb4;8P+P0ZJz*w6%}y z!14F8)Q~c|k|6x2ueVE?Yf`0&9b7PF#sPTP@5S%ahe@^XB#cSg$hkM8xIp6|w^#p; zuT^(2=8HBLG+l*3Vj^3#*~RAXuJEerPVzDncTO0w2VuekW#ssVI{CjN6|XtC@9TxU z!e=l(Y=@CJPU|*rBhS#Gth}{`EeiW`zW-c)Qz%&SX~lvk^cZKtJ+QIQa0M$pumTE+K)hwZ^c)zc;M^If|cJY4Vl$0Nk}Mq@obPp2_dX6ZD4T(mH4KIFljP z%ySZ0Y5Z4y38Q|f@q%Z@oNKMaA;%`N`UK@ z(A~C^+!hkv{(dRnW9lrA`A{Hc1~I&$Xq(_aeYA`Z7+Hut$`AeS z^x>Pghj70j=Ks!w2-Dgl4&EBVp-YeQ6g@Y#Ro}%o`@cg%gAFHfgF37KupsVx=`{PQ~CwL z;^Yo2sk@G~KfL&CpJGlt@{m{8?Us+T(2zeXi(ob13oy!!;GS7|_#_yqduC1Gl!jBH zr>qXTg&G_Xavc{um3c*RG;G87pnaAJ+dMmuvY$6ZUxf=RSO!7Q+zf?Fw{T7Og*?c| zhY$H*MAcNmZtD7pf<<5Di`(Fy$rt#U_zep3^YM4HK2Mz=LfvneL(#A&6})WEs=|Y8 zS)#!sI{d@2kNVtycqP0mjW9`m8S;%@9RF^VeD>jb{xskbURE5!%GA}UcPPQBup@X- z8znqhjZlp?-%G)y9+GMH?#1e<3Xp^Vla%vvH+I5h^B$eBw7n z{5!0J>5I( z%w_4H(L9O?OF;X|gDB14M{`DR!El#*xFLOm-PudH==Ta^raX~3c5J20s~tGt%1Fu= znPuHRK0;^PiPKg4p?I+)9^JntmHYeCWkoln5A(odk2-A6uED&3YCJE>gzev}lE$Mx zRQh2sg(IEh0lSdAT;vejbVJzCSm|wNN9s2D6WN?PfG_ul^20BtJScc5{1C%R+OV7ue|3e_&uuD} zdh7DM&nk%O?}vgLf+2j%oX!uqEL;0=t1QMn2BG@`vA%~3UCufnh1G7PX-0#vcS!{{ zj}FFl&zp2+ujq!(+%Fj>ZkJl`?S`lw)4`)rUNKRY0$ELcbife&$WRy`-IHIVy_QRR1A!%=fWS!BnJ zMNYpb8!2h=iJvnNzb_Gwt!8zZN|W{ZkV`yIVNw4#PFAvn4JAZc3<;_bY(~( zUAa1lhq){8i^KtR)#5Pq^q3|+*!_@tEcrzPL%$<9u@cqkr!drb1ukrkp+0hbq`X(h zjEZ#XF>p8%;?A(j-V^-cRt5ctk?5(yL1|IWUaqqZ=YP?s(B}L=?E98Qr?;ikp`Afw z`X@p*!e^`WUUMjw&aS4%Kl);9)iP|oRtTH?t+?^yJUg^q%G0~=M)>4B>%p&;q(bkKHe%p;KD3ibL)?Sh7FYeYO9oQX@aru`?6+?GV47vfkDApiXL~7vP4Je7yk)Z z6#k!-5#lGhg*Wn|Ar@R;{~A*Ed3@2|kCPVSuC{VHGQP(nxqc=}winnITN=`46i{seg|$^i@emkjN!1*Zo0ij@zYCXM=h7*k>GfV!P=4 zUW?>$4PJ@VQLrc%<)xA zdDGrjcva5hrCYl4lIsCvx^Nc?MQ7Cr=Pb!r{MIq|^myr)Y4jj<5vIlp_MD?BRIP!b$AtrqrbjPT9NnE&eEr*;h!3|kAJe%}@meprd{yuAYxY0${zG=dp z`YCb91XY}dUO_4N%=OJXNjbt)VF8v7*EJ~3HyKSf~do}yJ;3WN|NovVb^gUg6 z=`fm0`uF0UU45}2R)w;Sm84#VC;3k84lH;7B6~BbJ&IrV#PeU{>0UsH%plCyT3x)0 zyd68DK3Qn!3nS3)`BiGi!ZX+sDb*YvN4qUHOQtt}OOwlP(;m^){9cCnS90#rUJ5>}M~ClRrmFJo-1GfOZt>p8y8Rv@ zqBKpqeLd3Jt-A{-m4B zhimt9u-$A3;wL(f4MN?@;pj4FKYigi)U0qufv%=_WABy??%hQSUC>49G3gfH>tD&< z7jJR+&P9Ctzqd58)s&P*Z>7i666i(K47@u14s~Daq+vO?WiR^=j&#J5MmzvTE1yd_8e2 zJm#0$Kcid}SO&7OymX|2~!~2}}`1<-Zwfoct zm2tO;XC9`tv457pck)3 zLM1bfGCIdl@4=xmw~8jK=sCqQ%Qv6or31_5ZXSK)(PQ1nru}P4$Id{q@75DbCqJiI zKU>k{q|1+mw`YE=g*0N=GP0ljLGa8IWoWlwI%yCfn`-c%^_vu3em-nB?u0nu-YI|X zDKf}ypL8dqzxveN946iJj+4ek)S|gc_`}?X;ck6jp08ShS-wm0)Ypmoeb^6E`EKg` zQC+^yYY$JF6N?`2NAL{Qm3(kd7+>jhTykhEx3>0~Dm$&QNA}UAUgq@JNcP11C_3J( zp&tnz`1d%PU2=wS{X0AEU)a-TWL%bf#>We6@32+AYS2B7QLn|ksrqQnX^^$L%cON9 z8*xRaGbKK0E6d5RU>}_v6uu0jyqI}hVWtb0ARSU(e;h;JuB2z>H>EB8=gHG7)#X

lFu4YM7f)M*>RKT#!6k!YwwKVD)|8Nn>$wSpeZ_lP3F^yUE8EvGfq=i#}nl7 z-*J;5L1@b{;7M zCb8$71e$%P62)E)9BD6c{Yi6nw5pLJn{v>>RAduo7W3OVcd;UHAusGUiK`ABh3mK# z2ormhd0|ItvqzO{3$Ad(_#K?4y;$(+L|)r#4EYL{rla;Q-hDrwzRn!V8lzV8fsj`; zL8}cscK3kS0%K`~{}%KLo6co7qNJ8vf*})F$^*Wf!2I%^i0%-N2_lDa;A19la~i{& zLgrAf#V;{pZZ`LfRwCtDMr6EUl~j>rhra(5v3TxJ8XT)ZB1?=}y^o@?U3X+YeTF*s z6q@wfo(@c3i0{{DV|~a)nm)fRvX>;t2G+MFS>J#783+j++lEK^VhVyDwLn*l8 z7VWX`!GHP*pXteHdRN$1>Y&^tjXwTSrrn`d3fi8H5hGPNKy(ZnW%^OXr(zsNJbn&n zN7~yTQtoBK55aK{8yUcfLss+8zuHv)DhoabcVScCB^Xx`%y&fhUDv-USUYGe4$e}+ zibEFUCHcWr%$1Ym8EDtPGl$AgP~;ntQ5ZXn-n}%JCgt=*hl8VdVehAW|3n9lnUIaZ zf5R|m#zX00sy090m&TXQM&sk)XK;_-gN56w(eIiw&SoV@U(<$g|BQHglyro?mS|(X z&jzf@>`xJo*I{E%1-^7Y2#enf{q46Ve||WD4@x!C*uwS*O%g1jcT1sIu><>W)>6n8 zA7rWQfcunO;VDr-*M^6<@3)&(XAGvh8cmdXV6Zgh+jzG7dYHc!Jis>3dRXdAU?njt zzSVmMb#hIX&M#XkEuY^;IvW-Zi=egajBb3{qZ>#5bHSKiC2XJc@_r?}oyi2JEls$!X4`kh)MlTKsM!wEp<;bcY zP6M58_1ZC0JF1qYOJ6qMRvP5Bh;?Wzj*22bn{i9%D^KC-^s4cF%-(0auNjO_FXeP?@!F1+(h z2`Bm8sR2AI_neg5TuJUWQFQOoPifx9GZd2+LR+p0-r|gr7`N;W>BNV^;^haNMInZ& zUPG>7JJHW9!_z@S@nDz+r)^MbSkF2U-k%{RZ6q_#mE+Y=qMsMY5cgz zhx@L(%Oj5M;`oWG)}1>Kq;1U?#JaVC$A>(OANLLsy4lj9M@D#G7lkoaz1eE#2=1uA zn8&$)A@3KfFu%7jFgxv_qqFwHT)i!;e^BJ4>j^w-`3A23)E2vy?V(un1r8CRf|+T= zmV;#M9P*cJRWA*q4%>;Q#szE8=)31yoGfhslCj8KBAru>Bo|$M(m2oueGE?{qt68dmkF<|JOx9p569G;zWi|3Og`AW z0neh%W!0M%Xx9=`Jd8Yq9~&x=V|N5=<{U+rI0fO)7ky@vYh)|>4WXO|Z*k4Y6zgO1 z$h0*ahUQmkX!pB__WcWu%NHSaQsrTOI_!Dh5p!~%!zc6y)`aF`(sB!aJ>H!SC;g&6 zaXldKERk%XnGk=rptxlt4)#{VFZV>`NgL7Wle=uKRfqXgc<9D^^1Xhp zeB{D8?iO-}+w46>)u}H?ZA3l&IQC1rH0BD~S4miZeivt?CGhh-3S2eu4WDR;kRKSG zCx2CNPZ)KNVC2KqTo!x*LngN4&!Vqy!-)G-CEGwZZM$-6}R2MjkzcB`IHUd&m*dHyaN1fy3E+=qdW-z8itLUl4pbU-mu8TDC`iLOO4* z&1W6-xcX05PEwB`M>{oq_~(qhJ;q|#moB_=$V~R>*9G@>$B3D6044-2MxjGB!tYgx z8ASnG|K7_z&egErh7EKl!JhWssG+LbesCBy0%mt7pnc6hSwW}eiu9MJJ2=;YJiJ6~WCoO*+@*5ZyYNLIccROP!NTDB;+1I($X&blN-cMz8U# z+@+AMRu=Pypxta{&>(HmTt?bAmr_=TRyuq=Pk4{IvWcghzQ5l}zJCH}^38rU@TwZp zv#ZHsdRJ%;Ob|J*9Bf{3gtqMMiaCj0@yPWWg*A)+dDS9JGA*I?x^L*fz(^?H5ZMeD z;VpTa0n0}gI98sEb`KVy?SK~R^NktSw#xCc9Y(9A<%8^L-h#gH9;J`7>qHO1rxiG3 zTLMipMgB3S0S|h2!!6;bE_Z!^pN5m^Wn422Trvz=2?w#PQi0EQ`GBs2zsc?g=7=|0 zrqtqho9?H+7T>YR3&nnuF5i?>e5|t+acd`>EQ;XyZ67q(^+vbVOEB8#2-5ntrz7{W z=<@IRVy$W6?$1*Q$RCU3xKnf@<}iX3bCF(@MZXo2X>Fzrh6K)*LMjG{j8O;5`=6ro z@aOXF;h<+1PbwK!(G zEH*9ckM_Cdps{Hbx=)b7$WleTY!?E98n1%qS_3pujfT;)IG?9>Fs{lS0^e3CLawwF z21l*L-bcy!vf((cc%sYDd?t%X?Z>RDPrwsl9W?#hDEMBK!F#XwK%!F`yBbs`J}tf? zR{e+&Hy*sp{b%`D^fnIqN^Ier?jgcRuT`e%4Go~t^;cM)bY}d}8c*=uX@R%3??SzK zwJ`q6TDb2P2QF*wfcDI2I66%N7VWq|6`KFi53gvPKUIn1c{ZqEf4uN?fSNcn{VSdR zn?tc`pV0Q}I7;s`jD$0``26N$mK!7?q>m@@m?qCKTb!Z=AB|CZ@@?qt84HswgRuMI zLfom4fc5?+V*Z;UV%mUvwEr3RKMCsC`%VuY8JALGz;~*AlSj#?4&t;lTT~l96_k$t zW(&ORp-=Nj;qSd(d{!k*;=L0Xe_xj7{~1h|XPrjbp8ZsNCXL9>ks{P*P@hjj$R&0c z&gdUXTa<&TvxBoFINNAbxF>EI{2#7p4~5?=SHqfD7la*^!_axeCm8DUj8zPlLj$FH z*jc0mDxR~jOCg`{(dA%umlo4jzHJuB^jJwo7oD?K61S^G(M{*i*l@`aw`EMieRBQj zrb;I|w;sm%G2P6$_!x^m)hT?sFVM0r7fJ1^AL(;{<3c-4v?#vF?!V=V+|_l^F}o9f zn7`xupZA!y`x;6a%89e*NQmco-?R0?P;AhbL9L8nm>QJI26_#|!l#3<#G?heH+w<% zs^O3|wG0C6&O_svmEiBBiQAV=M)Tjb?31b*PK|8FAXr6%Rrz~%=2{Hne5~=mTG_Le zmV)>1%Vwr&{P!NtLW`YhcrZ(z0=uhl+skIy&=Z8R<}DCXKLO1*9w&Fkd08ELv|L3E z*;2ro^$zglyBeK5vxKVLhm!S^1loRgFYQ)#pwAkBBc@G5r~jJ3%ljl&c-_Snf#s+@ zmvD+xKCJ(6TL{hP?>eu`vE*5JmMGfASIcmjSc>4O`*ieN4NT^qP(5@g>gAf?%fCAzxF8S%?rCC*K^Po3 zH5&%n9B1q5&4u1~@=z=jCVWvoz{<2^v1xD&+Ah+>m^O2qf4CUl?Db@GUw#BLyKuHF zL7zp*pTeaNR^jInTF7`-HQq)F6(Ttg>uo4Hw07a3&%>x^WjcD!m%<4_?ywq_z-G7% zYwTQS7QB)%(~Yl%-LD4;i-tr9x6B>cF#kKu^Tj>V-(o@6txfQ7rmlHd^g#2R_tZ=cgJDp< z5lao4jPWOS<32Kd52rk*3)}erq0dX4+t*BVEYcMB+HE4gQPJQOw-q)TxG^=;J~Wl* zeYV-h;%(Vc6x??K$_Bc^amkm=W1=U_y!0R77(F_e*@~n89>Vt;8=*4I36^ICvmX}r zlz7IEh7NsbwChUFX zO~YC$@xO1Ju|&4?C}{+h=zhY|lLJw@dM3E&DB$kabNIAFhPL*NL+bR$z#9kn{9+B{ z3UcE5v)Aa*?-07>s)lF1`=NXCA{cp8i*_vuLwDt|*#E>*xRD$P*D`Lh=EEv*=KB=R z0#p~%r_2|X7M-Kq{;4=EXcqR79)-|X6;d9_P*1{WbTL$CN2L1@4~C-tTN6Bg;Rf!# z_Y1`$H@cH)Pbd0&(2F(;+HUQKb982*MolRc_Kc%YiBRh1Hpx%X6X+VN0uI8VaW^mL{k5yP*o^Ma?VBK_EZ3XS-EIUrMjA6K`t z+)I8!PP7l!%rpn9sz+w7jx$8%N1A( zpRhN9;Q1+R?_0zE@@TP_+C&D=?}Gc${@_0~L~zOsqzMfps7^sZhq7jz`6UCd`bfa@ z@fY#aVJA#2-3=1jXQ8usCsT^b5PUQiW0B87bkl2qanr|%>kqY3+_Xg$)ov=5xQ!6^ zQ7#n^D}>QI`r_~(SK*|>3ly#$Lre2FP+{A{*raH%*=x)aBE87{j1`^c{Y#y)J+Rqy z1zS*F#;j^Div}Yfiq%nD#ML#%V(+(!qN++V6@1LWz(FA(>xyjZ_L21K^hK0i{T?Rx z+0&2dU+{~54TO$bK;u20;o*3GAFg);Bi%!=q`E+m*?L&aSREq{e6A%<8!!)AOD3YS zk^y>7+#xiyei9Zg%@igsKL-Y!rJT8WEvC2^!qAU$Xj&+bGEwdLiI83n?N8>rJRpEB zL9=-if0xoj_o4f6{`{#JxpOkb-HoD=)pm5XVh`wal);)o8L-m3n)Sam9vn00!pnhW zOzE2nY#TKO*0z~oF3(A|jtIqoIWN%ux&ck*p6v0a`(ex3H(<3>9}~XMV-LEJ&e;C~D9f*}7+p)_s z0;5*T;__|tarC;Kn04+h`@7-_Tq$%#v>6WrC7YpMP8P${&$80rTVR~xKDhkap1r>N zfw}3ILPi{)6{lI^`oAIAvg#uQW^oMu*OGuQLK>Lxr~=dnDWhBvXE8($LN%RX=x}5d`VF(; z%t-FYK9a~H4-bc-DKp@(u_C+t;s`9y2*<-gU*MG87m(8{hCfROvUPT$F!RF|sM7a> z<-1gH)8C0$zmfY1=N`dd1@G~13Hp$B7OSZiALNT^&s*X?=a}n;4hx!6Y-GwKlsTOwtZ#FUX9S5y9uv2 zIpevKMUZRC8MBW9ykfvA;#&xRb<p(MWBcKenL$`qnTG~Lf}s7+8dNFEN2BMLz^DBS{7E%1UsCSmH373LehQa;?y&CVc`UTt z7aZ(c*n#*P@XsU-ubH2_TM}IupNF0$g}7X;85X@6Kp!fsaV6-p78Sm?y(mxBYEPN}`k7eWZiUkNf7yh8 zS@5vr9=tFe$5iV~@!zA3a2N9OQ(#R> z3BG!nhqu)OP$Gpp_Rg2XvzQ}LvWELXkF3EDW4A+C3B%HXw{f(pFOHWB#^-K!nEb&C zb80kjv(rA<{O1Z({PV`!)8pY`t_psb=!Z)uaaI6SLO%;7T&i^qzt5ddP9EQJ?J#?2 zGOtGKu@PwRI2%ZLKQ?p>qU_2mlr*iGv@Rc};3X2I_|KJEC!NJ0JP3Z1IP+!aGca3j zkI{++STW!YZVkGR(bJ0Yecmz9{j`Vu{Ta{v)Jt$2K%qD_P6$YP1A4C#JB^P8qZ9A4P)&4Y6eMJTYD+omOr9f_Yn2*v=d&beeDw z!?#G&uL^T4Ib{gB@mr+P#&sGR$eK}6RPr7?xFH}v8R#YU)dF~n@{ z*1k69I6Nq8!}i8wq#0F6XZr5Lj}zzcd04TqGTVF1Z`O75=V32?{bWPm=IhgP^Mm;1^(Fji_66O%{lq_Pn3!jwES}+c_Sjh;F!kSk zT-1;ELVEjrqYY!}ruGaLcL`hQ0cP4_s&CrK%7- zyZkP!O9-VQOWVjpqMhb?neg-Xa%{5H0=3Hrm}hVcb9ZQEQyhM?$C_F&h~)|A8Kb5N z6*#1`4=$*3q#rGZ=%{rleU9dwGH+*?ovFh{=dZ@~`m34D%4cR{YQ|#qm7DDI&L*KU z;229?lgoViO0X3x6IkkpOTroBc41-jbJPzgMXjj8C?8*ikHXL4tv@Sa2lqS$bh^W0 zn`6vDPK*2&?WUZY&eZXi@X15&AxZNkm6%G-{N+883U~I=PK7iAhflwMe@s^g8NR~gU)fBe=(vT&xjgO#mRga z`qdZ*Ro6h?x_pcpdH@G2c0#hsanzYKm?TFQVM}5^QW@(+sr`+ph5JpD&a2=hWB%Mf zFqhQpXVOSM2e~oGf({*zpp(OONNGh2G;w#2)8|j@SM(7$zaxUpD|Rx|$f^UI%r77< zCk@vczGK(^Td4NX3KiRQ@CfJ2)|@>GN4Uq+swW!C4QhpWNTppHu9Kb70Fi1%y1suV zmB!_B4^$3FFZm2!cBgUXM*co{GfsFk?7X04(~lyiUBmt|D{<9lSM=qJZ(fE_lm+IjU4X!F~aUv7R$X^4pK7HL1*Z2JhY$+%Af0lk;^~S9WoJ1_>5_6**8?L z8iNCP=6R1V;*M2>2M#I2)_MFH$|%$0s+^;oY^1yT;Xz|}{Am4({l_vC(jd%6{h z$E%|M!eC4|(<~IO`60Ng%QMlvK!{-VC_Uj59+S4koI1o065k;-*n+LPJ{lL53UKuD zTqaQ$i&L7U$nofIGL!g)<1X`l{^e14@2VXRf7TxlY|~*y^;)3R;)wD4%F(Pc1AnCr zLg(-n!6IiI9N8qmvB{Iz5Bo=~|5sJ`I8XsMkIF}3PbeOFV1pAiU*LzWu~=R@1ZKY) zj1Th+;O%;ZuDLZR{X+p=x~4)z1GgI=(4QER?@2w5-h@LT@ubPYg z89ym3W-9fcJ(xCcdcoaNJJI-iE?9Lnuq%~cSpRd4@Mh;qZ1c+j>>P`!C03{}u@93u z*ZpbHYv_D86^7+>z}4YNtaH39-3=Q-1MJ69=h^9?q7x>#ZSEFMJ&74_)qai159$%@ z<~_%n1q0Cc%Wq}_Lm*SW2@(|sVXL@>?vxD>-zxhOzp^9X)9(-3@%f-@(GLi*9t7Fy^&sz*3Aep{>5}sX za>*~o_=){!#-|h<>!XN8J&D}YgY=}I2|1HFG$bbA>uW7oxZ@%|IJB3qgH53SuA!tG zpG-Et7&&pDpSfEC+B$~;h%Pjs+lSvb%28v;PTFc+NCs`ksotg!1;@}VdD^OT+ctd$ z!)WxnvxHil)8B6}`A1Er)LCJeGxi~jJaZ5|a-Q;>?@=~RT^d^sbwmI7a2%g3i?Ue^ z?!2(Yt3MxtNAh3zeJvft+MU?%#BcannX1#miB9ja>1z_T1*FqoMEN4@Xi zOwy-QW1hjVLheb6t3mZYRygC5E=t;@;DnpXG`i(9&L15E&!2H-^%!|-mHmcTv5%cf zUk_8?J%d+0mjoBTBapVb5nj0O1Ao^CyxP1Om+yLqy7vPqFmyaMO*{j>L$AUP=?{=u zH;Bs1T6q7<5o7szWY;SvupN6Ee%u*_?;CWm-huZt^ItKKlEdJx^$y~y7Eq7fL~**D zhv@e=n5oZ7VB<`7!M%GoL3T|s>k}3v4DnXP`nD*HxN!j2o!7?Z052^1VS)$bCga_+ z_2`{Ai}r5Q7O%g$Pf|hvm8>^~(!y-0_Oybk&-IYJcQ93nKXJ2pG%oG?6g4Ftqe<5o zq4VQ1SW|KY&f*GK(|r`(t{uh37sGK%_j|}W}tZw=U zqTw$PHDfTaHXdZXTzCDI*57i)>_i}y$Le29Z?C}+Ie7$%E-xEY*o_Puc zwDiY_8g)!_4&WZ4y?8o%6I{HO3=dM*q0YK?FyCf^bwfE1v&I+uy&nY5mv5oI1wRjF z$%_Yq&r-yZW=xpU!^R)J!$(eGP@lJ&>2x-+;bAjbzo;?feoLK>>?^^x)fJ@a97qQL z-NW*4cS-wEyvQ28#5tWY@cF|kx_B39H`zkg!F`ZYI1tv(h=PIU7r<9v76X^x#MLVc z(5s{fJ@0MCbbb$1FK=PG(=GVt4a1p{fl%MZ_k7$xU(sty{q8X`Dj7ms3ONH)<2v)O zWc~pY)>CLv%)(T!xg<1fCvT%r z@;VtQx+c#PN4$M3j_@8}K7WUuIRBK0d3;}dlyfx|@Aaq8e13FWyaVzE-rylQ0moNl zVd(rUXp?^n9cwj7%H$1wT&5xZOqZw0nejlkW0_y=YK*AA&hE4pzz*G1Y+ZE%Zz~+2 z6YKU9?3;qE<7e>wUSC*j?Tvd{g5dk&30SBrLj`MR)B3ub81+p8w|(1Ymh{+@c^qrS z54e}4Z+X%HbzV&wTMcGCIcWQ?jg5;RM!IWSN%!PVTzjh(FZ>coqrqJ;btqzOd%obx zyAn93_YtUq54n}zz!d)x^w$0dF66yD@3_6tGMeYuuMKDGM?8bTb@iB+cpMZiJc65B zPQ$~=t9Wm*78d3QvCH*O;csL+%6jP0Hyt(He=!29u3W&>>q~LFR4`d(%pp6~Q}}r0 z7tGX8#ln4(!rLc1P<6hncr3e=9JNfvw2xi%GuW72zFPrZoIMn-I*39hEl2BA7xr?N z9q!yQ5!YJHfcmAgD0;_JdQqN8e`PPTyvyfs!-~ChX2}p-6PAu?efMCEs6+MV_EK@s zPwW}sfR=oYOT2&6(6yk$6QRwJxI$R_YrSqO{L7|qhadjOW0KzL=M$n^t0EI z-~UJ9k^UAGt7*Z4VhV)c5~)me&3kNFY9X$d3ZQ9LJ>12yjBM`qWzR;X;hNa1BrjKq zp4@NZcdi3M_ZY8Zn7N}~n0xFWV(ew3CG$j*wyE_lZett&pIh!N@ z+$9`&v=kHfs^hxPDX^kfg3`KI(b)LGG@?6)g0Ib|TegZAxKx^K__@1pLp{ijeTDrR zUeTgczSzuX+C7$|U~}LNxy2GN16VClqMGmE_Ig5e=`yi*hjlYSk>vgShW zM3BYbQaf2l%|jNF{Rz#Ls_5+-f@*d#(E6z)RE6Z|hzRnfC|mGHO}cFE_lb ztwz@yWJr4;gVDDim}Qg=Wd8kU!uyveNw=;NuPO74@t-CP78lX^jvH9#@{yb+e$eQ; z5Za<>Pv*~OkYVH z@oDEZ7_#L)Op)bz*uSP2EXld!U4&}crA)lDmacHOve7&xHpy-@O4tijyD*SqV-qpl zAcY2RlM{`SuVd>3Up#x*lkufEI@|q5x$uveE_;@4*c_rMS%s+F%+CtztkFOw59D~4 zt6^Ru4R;JC8xuonRZ}JZd-f36wG8RTYWiyVn8sf`fSP-FKh3@dy!T|n-ebzt@<5In z*WDL7+%^R8pppCW|R={PTAKN(HF zOzCc}SZ==#v?s{JILQ+j8$AFe)AkDI{v5%br?z5F!{yIkyI|nKkv0cddckUnYx>_m2=e{wj!iXMnsW7F*K2#B!KEz!lZ|%pey_54!en94h{Grb?&V z6!ANlsB$FtHxxkK_9%3aIgHie#dusvf(EXRp{K^_WV1n%462Xf;yN#!?HGn9?^=vpsDWb$veQ3lvcj4}OO|y~HJmIl^C`P2;qsh}Zh))#%6EB>Vr3pX6aUtIe=bbvi zTqhldnlnLw&&}zNc>wNh@`f4fW67X+4UOBAgtKLj!H~@xsCaKDjc>dT(ZPEmzUV9H z$j_wmnO=07J7P{9i^OaLSL~jC92cI;gE78CX@1jCQhJ?3t2bGYrnd`TEe^o!S1-ZK zUJ-q!U4(&OWJuBHt?!Hph>Ioty4^N+&MZ+G$Xreqv9e;XbfXToG&9>y{DT9nx~gicyM!Xk+i zFfHGZJzljK`%T$Jdfu-f{)rW+TJ%J5LKPj|r_NG_rT{!H18LL%LMd8qI zT)#q_{;a%1B?*zFxj}*)dFOb5;&CjD8b{B14v^-rhcx&~6Lu&iirbDcQSx9d%FZ{& z;&2VzTE7RDZu7_feK%p0PB^^N(ZJ}NRVeK{n$&qOeB9!A4Et$=G5RTVw>OiPYdpc& zfMT<>7pZJtb}~9_h=C1(r&0CO7HWPT%w6;E@l(bXeAKFkZ#Vzr4v=*ASMdZo&he*j zib>R(dmZDLtWq3ups>?2CuP&WQQKe4jV#G^qxV$ej~c*T}n>>j8Ja8C{3a6AbTlTHp)?E&BbPdo!-hp{9`u=vA?R z{x-a!>^X8cXI?T+f1*iaoT4b`^k~X>n1wX~e0FEG2W{v(rfki^f8YKinF|@@>k&v# zRKH+;%RR7++60NpoMSKb1CI7RNN@PAye20dW3prLc1nDM&^PD^ob z{D&8?ENdaGY1UxlUjN4bM)S_MWf+#uXoacccB6z_5UQ7+gQpe}7_jX=_H~;>$A5E9 zhNw+9=01VB3nycCS~?_2D8R$ohGXVkOB};j(ojnr3b%{?l>4<9p0oG7lfmJ4m5*e`tNvLAt>d zpn8c8d~!E{o3+=mp<*@3dU}xt&o+E^u*Ci^!r@Iv0t?Wz1Loj{2ErsX=3T{w)0IGd z*g7=Je1cBp<7vU?x#Ti7lZH<^NRQs9p=tk0{E?cEaeR;1@M#iV>mE=0e~B2hG#AIu z^2VbDA2G;w1bvg=j~VAw;I?}pMr+to$emc>|BS@KCz&wt@-t?_?1a6OJI#E2b;x4C zb2=S&kxI?ou!3KoN8SyG)pJ!*a%M1kU6rDF_0wrik3fgr7(E{FkKBC*h@baKi%&)? zQmbbiov+EJ`OYns2t15(+soYuE5soS! zPva3jn>JqX24s6jVN3aW7-6^pS53T$($B7d%kqUN?$^T0)_-B-x&3g(Y6u!HwWO5+ zxwLu9Li%-UEu>y<2Nj+vxb$cOY5BkGV%{{Pj=r2&=^Dl)(o>AKG_GvzUstlEL=s;8o^`6~FGo(8Yt^P#2ZHt0Cp;a3XB zCC-yjBO;K`Mq_ck?qL{wTnbHnzrp@_UCch?5Ja0#M|V?IJY2aKgQqy5>KI#0Dlx$i zHxsaTj62jA^A5V?N}ShJ20I5e@T|%)=ri^=tEvx%?*oQl*Q80 z{WzBeppbS0{_8f!X_4a))}Mz5y&oVgX$i*o)iH&+K46oR&YzWiVby0d2pX`4K~wBc>Bf`%)ey9{ia&5ThasH$p>Z5 zPsPV=-{4B%bSzsJi3bh$V$H}r%#Mk~)bknO!Lw#*H$pH~ArjpDufSckDsVXd0Egc@ zhsILBc|Mz=@}*E*BRLmey*&g2Q35CK9fdUgo?6l!tu(kV0n zBq_@~khW!wrCr9!_AvQ-~ltv)V9=-`TG>?a-zaC&(E`#|W5@F7uZb;kr z9Fp`6;I(oS47C>U)x$0D$g&3h4cH6ku5)JViNBl!D^l3*-%Rg~2Q;2p0rpE(@c8GC z@Zhs7&a`%esD@VXygL|2MhCI5DRt~W-*~pfp`FEVI|;hqA4AKai5S(p1{2mu;&H_MOq;d>Dx}3nr-+#dB1bh6| zF%`ETiU$+!$=j3Qh(~$%ZG3BAR16-C!3t7z^ZE(gCA$%gi!Z>Bqhry_p$J`4f^aq8 zJ#8Gj5Oo_ja^Cz!9Nu;vm#Lb8incCXdegy6 zlg2b0zP&#>Uiu1A{Zg2wWCpkx9EGCbY;-d0OTP}zqUWh!;CH}M>^HOr9!+wD4=U%u z^}j+G!dVk;X9rTF^dmaHz7U&o3A@(p1G#J~=vk*rNzH>qNq0x`?4CjP-^P%(>J@fb z@w>1q)lw3q(wmC6O!q=;&O|n|p9a4DlLKyjCUFkpcAO|D35n+t zVP$tA7+?N~oxOc&Gxt6hdqvTx2@aGWehR}gPhywDF*H(t4^xa^QjbRx*`KJvKDrK~ z#ST5uboyc1%;yfxmd5xcm(NU2t;3$I0@ROLfM>?Mp@xKBnm9g`K5dMK%+Lcc(!n45 zn$BkHzg>c5eYaqCtt;#<=I*@h7GNCCIUY6su+Og?{zj|gpqMN0fNO4 zMG|1iJE95Wb9j4wGtKgh$A?9ysPtSb_A@_=_DA`%^-BuObbP|u+aAzj^B7~?AJF3o zFY&hdSoD4EjHl;w)^(2zxm5xN9iEN;_zmu)ZKH42_d#Cw4=nrT12gt`pcQ9S{m6Y! zL+!)Fl!Nl-H+0%W3&(?Eq&P~P=4wj|>XLBH=MC5)Rfjb-d#Ry@v*$ZoAS*E!zwG~u ze|}5S-EspuIAjksm>1LNyi}~OEW<;>YmC~t32K8j;|KezIDFSsp8M9tt^9rIV@o1B z-oJ`H|2@a9tG1*(*q{4*=TN`Obo@AAB%U_tN8WG0vKd8JpsnmIW~*q64L{|?x6f_q zVx@v0Exckg8VYd4`Mv08uo8^=)H1o7Hf;N=-EiyTQ}lCv$@)p=GVcvNP~ZV%yZb(F zk2b}b3BSNcVFLD5J_5fhk7CCF13WNJn8w3vLQfq1WrQafAOv$Um_TaymY+UWa#(I6efF{R^4ClL8ve`a>qO-_!E8 zjWi_cJ!bCRPrfsQQL>h3G5C-;$Sa2|HWg8bCuhGeRi&Eiv(P4Y42*u_ zO*%51edTDu&n}h}=;%qypN*#tp9ISDm`AO54&pu;-lgzK0MmJrw0PA6nyH~Co@`PQ zhsE$Xq{JikwRawlx^Y5q=#^);l_Z4G%gb?!f(*4B8iGUQp5XD+>v&kN12s>nh#L|X zijGW4Jm${5XMA=oe_9c}=rFszaW34PIuI7}8i(C$6WrpF3pZ*ULBDhhu0D7YCzfnS zOWy=En;i_3zK+Gg0~l_SO2MMh2PiG|9E#DE%;Nc3{9(*@7xVm}cVr3Pavx4puBg+0 zM=mpXyNd8oiXv@vTER-7n~(m1pOY!_zp}nzaJFq++;lb`C{-b)}w^ z>3nXJLX+kfQWK4%iJe>V1@Bx*|H+4WMn6!w$&dbwT0)KA)bRQG7ON&uWYa>~Ik|A03S{U|;JCLpVP>d;eBn}P2 z@?3q~dT|0e{>g)I6B|rk@)dj!Tt#U$p4EBvf%&9MV0-B%R0=(Y*FSH;(2=@yO!*9@ zuJ{gCPxy5=ITWW^8A6cmPS6hvf`#8S@U%+=Hdpyz#e|WxcGP7usC`Ax<0jyriQBRC zoj$C#>q}DdH*k2jCCi%8fC`UX>8R2c+T5NG!z@DaZU5u&)xQ#s%Wg$kd3*A3(WIC+ zq4d#SpQd^*rc)gi)cjbUhIo|V!cn2v@pdME4%e~hP$l>jI}Yn-)#JV~zBGgHBO{aB zKtgp2TD?ohlVhwg&+9Q7Cdl!<-Zre9qeA%GkJg5nmV&?IBnotPPghZ9J(Wnf3KS<^3Y8@bISmqd|8BN zo?nAU8!Tw>%>bHnZaGPOUWcz+PvGvB#W*A_0@{Ce!}x?$XkPUYEAB6#X?tSm!6@E) z*G@$nqfFi@r~&hf3&6^44^Ei15A}*WaQC7c*z#H(Blvrh$MTE9_iskz9uPz${utuR z8flgr*bn-I9A?v^j3_r_I(cQzr+)RZ+;OafbA&2ZvN{y}^mf3rj#@$c>PM8}eMgVb zR9b7scjCi2t7wuvtv|4vT)&yn6|=cu9r%cCS-DC$SG`OKlyAl=#qD@!&qipnC}s{q z3Otl+g;5IsSW&MpoAu}rQ+gzevzF+QIF7S~vM*r6&(YMK)Q^_v2ZB^v0K_;A1NCnc za9H~{wBC@2TINci|6&#lj;|Cn14pwv^LY3A6+gRK29m@1n<#tlKCqWc7*yAUavPg) zweCFTVvx#mbSwm`$+Dc$y%D~CJInOm3oQ<2>6)63BC~h(J#~IaLc(3ZbAe(=Q?btLAMyypMJrggSZ~0Q# z^`9fzhJHX5`<+#JNcG&?8 z4x40_XuShEI{vbVDn(5HV-8zR9ui_MIS9@(@4oE0g86)lp3TGzs(_j_n(}mwXi{WQpgTmVz!r;oue6LtbHOgP9 zzu7j_ImO8G=x@C38-rI@nqct2sd(`Z?{GbTMp39NN+~#zZfO~o*{#4>RtXgshmw<2 z6*+t6aOPz^71S<4%u2zV?VWgf|4}$_T#kY~B}B;qbtLiP6U|*dSS+i%P22p5YS*4d z*Wneo_=S;klM$C$2c!Gra(EwH%wmTR!A{?KaLDD8@ZNeM zsZ@2KUx7J{-`WU=uQtMzHf{XTG>lFyy^G6E{(|q{kAvdseXLJyrSK)^5w)EQ7Ms^@ z7G15jlmFuV7`;gcr(X)gmtkS#?ze{)DNF(Fu1_q=?JP6iV#x{*h;YKO2Gjb8bYxwI3WBc=p=8#ld%Wz%Xh*@r$^w6l?$;vu0RMNJ8kv{)$!$0m0oUj^w zD&(*=;3QV*hNF$63H{`=OM)R#@Sci9;YZ}Pd)8YEJTGvOVNgB911R7VijxS zaMH+ucxh1-Xv;mqw@XX$sm?>V@-@rKKjhL0?ivx9%QC}+^-hV#8bobxG$PKbI zZ<7|bJB$&j>1FX2SBqdAAU+HQc==2yz)I5dn_!`r^%7NzDT0u>#Uham>2tah;yI+ zzU3bJHh4QpA8t)k5RVrxB6ukU6dQHF9yvi73*yo`P=4^blI(fz6%_>Hg=pS^CQTQ%e8jix6A zm0w^NZeC|6QV9yzmV@DdW{^<)f&c8jLvY9>Xp3$atlRGjd++{acCUVdd08ba9_j^g zZ&h*GoOp1w)WeKFulXMI2vyaMq=ucxz~tf)aPL|Vr}C8PvFl9q?@VA*{4(&uhCtY) z=D<8A8PoIwrP#{e;iKFYRG_euWY5f_XQvlo{Kg$1J!}b!I=X>=7zg2hrgo4v<{7&> zd?;HJDi6_4@@8jyHjh_5EM@liasa4>jsllLW2l_?8h&!_-MeIGmXUY^f;IA)>Ny># zynl=>7W%On^Rv0HP~H5%wqntCr?PnIiWJq~I|ssuAdm?wqVYS|kkV5-T=!d2s2Y8U z36AQ}TCGV+kDT!E{FQi2;sX9r$OGr>J#eA0oI-Euk;>Cy=x=d>%>KmE!M=IuT9O1E z@iFkF{1#~3eT|hrY_LGKhD|!}B(}YOMBgJUsLehb-#&}PZvoNRcEgnx?$jdp>=LY9 zGmCdDYgz8xvHX0Zf%)y7G%r&S{nsxPCzB5)y8VI7->c9`#{`?^-Uo-RZ6IgI*4i!?}Di4xvE~N#7lIX;(SXBA28y{}~Zy`Z_df1F$QfOPk{)NV{4aT~?Liv6r3T;&>9=?3 zInm|!VS1d=C}_WbBcxyPV8=W6;Fej>*bKd`0$nlcJHae$-`#j^V%RxJ{=Ds83&*- zcqqt7-KRNwJt$fz#lp`<)JJVO-EmZ-(>-HE>xfVEP%oOghOH1?>p4j{;w6rYy#ose z9uO9<9LQ?bp1`EP@4?T-m!;ta&`A8u0(Cd?tXBf1`F`eI{-ZE*TCVVSY`fX*^I5D$ z*+JAUw-L8wNQ>+LZ5HHyWiUklBe+HSFv`mKL`sM3z>w%?I@WGoVl>VU^TiitLHM-)P zg!5)K@QwXcs)T}5lbGk`RQC4mQ4pW_!3&)laJf>6O6oRnV6{6QmeeJQwDEW(s2{0K z;^)PePuL>956Zl4$j_T!h5CsRkgR2oU0bbCclAFkULHv*vWBd`YPIlvR~op@e*%lX z7Nb+D*{!-PD3X2!`6nl{kZCtz|NQMJF{P1BPtL*c=3G=b{zs@^o&Z}8 zy}{atJ*-Ae8mGkdB}wU@!j?WwaK>;G+4mO0(lMH7@V*R>$@d6tB?d4~u7mAe`vC5Z zwxo2=Rie>>9Kk(%hfp(4m6_jHg09cas5^83J*ZV;6K4KkAxse)lrr$ep*>`_<^r9h zF$@d+1-2iHi&ik3-L|0eErEVcoe6PyXYjw{eW**R3FfWKMQ8sI z43+#X)J+?U0V|FWRxD>bTm>v$qD3Z;PC@S>eW)B1z+J^F*@fyV=9Q!O)l)4s#n<3kJht;M%h~q>bVD`=AbGsk_q}=V7Sw zu9(&pT8k&f>aeth2ZZYDQ81?aDI`Z*;;T_VP;0_Dh#LDEr|jEBBV%SU;an}m6>de> zseS%O(RsM__xocI}Ht$v}sUODp?7U zU1TPEN2u(~{O;c$&~xc?)#kg`}M*;sW(x)a14cCr`ceaV0=-vo9b)j$a{J` zy*T;*=SDt^<~c0qomXgdj5*p3?n|$~hm%tjK))ez+!dxq|7vD4wYg1f+nfuic*>hI z3DRND{^xAbvoyA7_jt(F-4CA&)o_|(Axt@!&#c?|-I}v=H5~3zV6Qq!`a(~z7#E1= zAEcn+znxeZXqwVb)EH2f7?J|#`wB!ajTRs=2<}2`>-49gcd*wv| zgYbOT4E)bV1u{58&CYf(?ps^VrZ}B|%`5Z3p=>q1*`z~?UZbGyEsX zhC@N_akQ z<4dDQuwhOTdu8TLq$Gk;d*1}@Z#N)TDUtgn{m^1^3dwK3N~ar>NVfbcU6$frz(x0= zo9CPTudYBz_$=IKpUk=Nx5-HN4WCQx#eiQnIOEw*(6OAyb7et?Vnjbf(9a>28!iP>*WWd)n#xt}@# z7PuSv*YYGR)n!RD^VmAeZ^8$eCduViS7yN_L!%=y$?4r$(g|><)YAE2H);~+rF=ox zjx6w#QN_$B#h7OM670%zF>lRM(jHb0HwK)7$*&tBFk=R-lJO*+Qg52}UXJQO_qY=n7TSnq)AMM{yG!8rT!vDnHB<1Qk>bxb6!%RNXraw2 zx?aVfO|^G0(daF<8+}IWmjh|V8W-xj(ufS^wPCsnzcV$YqwP*rv{ZQuDV-nidc{80 z#{Kc7i`pRSgCplN>01aTWvIO zNTOry<<#qwKhFU!rQ9zwDdFp9Y%Auw*+QV~jR~~j-x-PwE1;Qkc`l}PA7&(M!i#?E zaa61VMQI&HhB-u{Rt1Ljj9=YKX9CeR-HYOGuK1>Lf=Nr`id5iAQmi|W@ zpBQ2!?+(1z{{rj3_D3t5Jvdj!7PUUjXG^|m!o4+8SbKdg_EBlT4Eb#8H$b2Mb4j5& zgOw=Z1?L{We+%cb2Vv9?Eoxjo7JJ3rL9Z$w3V*4E$9Ud-(f1rYxp@l8y1mCcA?hTv zCkI{oOeT0bm^P)YLEn~k^m7ZrC5O+!T8}VHem;?8cO>H=Go(#tmB@8o3apgBkEuo8 z^y2vkFrE~Pn-=n}q@pZNsrU;ia0sQB?q^C($JwE!b8v-k4kWm5XI{L!I@@k5PIx6p zRz;ix;Ib5Vui^Llxj%5gTf$lAl5h;?=MMe-1?*M_gXyMotS-(LvzH2J^m-`VpGi=0 zC7I87gK_M>QF!%PU;1=anx015gG|8?3KY)apYJC?D$EIMvUWmK!*h_*mBEye-sE3n z$=we>=)B!+lB~j!E$ReJpaN9`-K#O$(-x> zc!eQdQXEe9C+#RxBL;0$YK4by%b~=r1-I&!W22@f7Is&$rxkzklD9lfS?Uj){J%qs zk0;vB%cj}ad3Nqne=4508D(tbSmG7VR8>mFK+bjURrdv6pKgMHrFA%@?KU2E=JO_1 zEqbQ1A6%_&;OD;%Si}F<@Qab)q(2N7nz_)Ypjc{}mWJU6|KaO^| zxp&7Gi=vhgz2^@3=O3YP%pttR_r5=kTtU~;4()o&(!uBJ@PHVP56*D$_6Rjx@gfK;7_M9B0^xIw$;L)x;2x>GJ}bMF-|-JqTCk9l)(Aax|d+Ag&Y^ zKy|+cHndd_-S@u1L!YDB>3JIg4pcGC-+*Ie}+tECiK1-*CdP(OBSJ1Bu(Nz_+fgFt9TRGoDKm;R$B+AeYG)xIo>RJZvydKyMrF z*%)Plz4}h3rel5-d&LJnG`_)8MJY5QZWi0SVjBF%=Ub|s`(S9FIyf7Ahdu7#8Qy`$ z`1Nf#cV9%IQgb#s=Jcf-SDRQ_uP30hWH6;`z7iTMU$L`y!r;48EAzf)%+|0sEJoc~ zut>YiT>C#|XR;LVpScP4Tlg6bpG^i=m5F$Nn+8_AjKjo3pYfuuNNZ=GXYx;$!qC;G zc0d?_3RrnADl{u<1Bu+ICW?}&-IMp-G-;kRVjxR z3|@lsR0lDK!mkkct`t`}C}QPjxuosG}DI7OlXs%3CnKZnw}Vy%(ID z_}~8+31$&Pp^wc=CUaJg`B;^3x5Ww6aAAhqIiW z4C#w%D2+8;z&pspXmL09_1yF0-3IQRtvn7h2dhGjy@=;}Ex^uewv^eLM@3tH(cSM` zgz~jPEI*|;?5O@K1lm>L=Z57p>w*)r_87phPXr9%bAFF;{pqmnF4DTMkDu&6Lhr$S z&|AX>biEYeeBv?;j-3If3B}lV(@?s>-NMr6>tGOPn#zV0fXO(0oI7C8`fI7uq!JU}dD9QyVA!{#3*}n-k~}8|Oz^x7 zL9;W!E$$E+w(o*L;ju!R)B})rYiBtMHkiCp35|D4Q?G}^Df|ghVD(lkkG_dR!X7}w z+^_7%kqgXmcQ@x9w9sJr)5QKTY6;v-D#OF*Hpr3_1A+J&i! z%ds&$jY21elSS!OxLP>^Jl^P_P~i&2p7*fYLz%k1ZieJdBWTsXTpI8?jb@bik-4@S z&5$?4kM^njdnip=8-i%<`+GPxJ%wF+B!kt{)M?FWC+gO^g0GhE#Kv>Gq4g2q?TP1E z)~Zrr+Lllp^WqdvUHl!^?opu4ZhL9ZSSzY6KMDW$$4&n6ij}VjMdje>&@!(T>aWRC zpRS82_#|PYt{;_jPNtBLA938O1E@63j+(deyYY!iSQVs$eg^N+A|;F}TN;IFn5cil@tk9!5=D^JB?S*WbUTUgDvgW37)#nkArYp&1yJ3%nz2CUqeGRAJQCFfNjQiI6Fw4J+^&=pU)4WG%E{qpR9_0FHged zi@tCz(iso#ZN}9a{-E}*h?;dLqJLu)EXdZv(Z)uoRpEo>FSvs!aXGU-x{ytpG7XKi z6;W;2J#bxQL6-yiQs(;J6g{?(voVWBWt%8*f!7PFTiZo)#x+>*eg&S6a^m@wfl#z@ z1NA{xw8mh%`*|Mui9J}@^coutec&i}t8CNV2z4jMQvQY* zVh1+j-|LRxu6LVVy4x(gTQPw(y3SzDK7|A(MzkSP7qndbpt*P>jvsmq?@vELG0P^4 zEz3PbS%y4>=ZE5qPZ^kLC~yzyXmGmwfr6}Gpi}fVCObu$J+XFzVwa(KYruBA znZxiffBw`3hrz_#s@P!m5^l%8Al>h+WXT;|`PzA8ua<{Tehq?yorIo-3#sycHkr>I z$$tA4;T|npy4qIA-Ijx~P14O=mg&-sf@7ptP(r@9&*Iqs#?hAmYm6H92xD6Mk&-$0 z6l5L4UP(Jh=($3AR!;mkIhFbc*x?+THkduui9TlDB3kc;K?k;BO@TlBb$<<+dC%bK z)`2)q(1KQ{2HYLJ7vILdV=X%u;H8~(0xr_x2 z`-CgS<+S_bBRDbL8WGwB%0$@OLdoRD0-kf^I9|)txjpu zUTHpmH><>Pp7P+i&w{NBIEszESCjLfG3a~V0AGGkfF);lvV-z~`b{VC@ZC^$_QDYi zv6)7e-y3i8&UBT+ZVwZlotsIOa*8;p#hm5d ztzijS6~e9Dcd&DjI<;m_q%W_zKkZU2R_}}9+-fJXzquPD(-%qXe!pWd$DOVl>4*=z zkAVKoD9q=1Vvo`&TDWj2&H6ePKPXRTO}T#TL-};5*fo+I)!or|q!ir!FAy&EG(hhz zg!-8ZIOd=!q*XM6HqX#0di+N1l6urr5;5l3YB(_H1f)CoqRoKoxPN~(9FgPujU%#T zF82tnuST&yzbE0~broo7c!*WChoL3sxNX1VL?)4?sP*0iH=a7g9@_=NvQ2W}aV-}T z7k+{Q+X3*Z=PInT*~U!p33EvrEh_fCNjiRQIM{}XR!?7wnTKDBZ>1)I%F2b*a_t69 zi<*w!G3PMf*BVX!S)o%+D|Bsjf%D6U(B=(}wAkYpu1Yk)jcA>{PTBRMH*^e*eA^jL-r@3#`aL*d*?=z8K3-2I+FUb&BG2#G8IfrYp>pf z+Oh`tuXQ5?NiN~*nm8OjUk$b`xyFX>w`48_s)F6kBw@O3oiMc{Q({%BhT+n8Aa-Ir zvm06uwF)y};>Sec_}FzgrLzGRJqTkKnlD)B2}7os8867J&4s{UTVd2zQ&#ZtGWP+7 zL$i$(IOz#!%{}OW7Q5KLo(QzGO=20+QG(+4N$AGCA%67}LG8ce7(ciHO;c(?%Po&> z2pkCY_9NJwY2(Q6`w6U%lS9LxNfbM_7aiaC8KU~wqlZ;Le8-y<6VF*tYE}u3F*^hX zyHwdc?IpsXv=62yKA(je-cjmu8Nyyyjf5k5eE$CW4ofzZz>1G?U>}+ez0QoGyRS3Q z`}S0b==lN@9jxIQJ{BCas=?`005e~o%SL?D0J96eOkU8C#82oEq$U+e{yAEj)QmU? z`6{(AzPIOF{`wdp;@A;yZR zGVvR0m~e_Y+H12KK8qRecMY{``Ofr6@u!&9(2-6z#Id0a9GqK@gBEX z5Yq%Q@+9mw=pPx8jUI82KZr3Dht^a2pUqyV9SycN%PYkP@2IReGjUHHe5%8 zvwom)E$1U1M9|BPV6$hp2*>^nBw2--sQDurRy6Jr3Qe{M8CSbZ8(*vv_3rKx2`${<1;Kw-;&91wUH>yQ z`wHaSB8#J{4%11)2wI}sfKB%kz=QGKc>ZhlHYuE~EYxN#aRzM8%SFukat)i=jO=<9 z-*N2KXFqCZN+LBz;)?W0czX4w@Ukq6J^uMn7}2#_@-r}kx+h8V++7O3Hz-B*g$n0Uk+DbFS+reegF5=)pmf=$GpWu zDf+bTq6ML7A*8am6n>$_$J;jYPj=+mcmv=r01i%yGt+$zbL`;OjBuBH&P zmy~M3-I}}VaKmRenAc&7W}&NL>X&(FK42p0#N~3QnG?0A%_ik@ez+)lljNC&Dir9t9HzVe^adBuI>=nxkd|SkMg9*1D4bpUk(0?k+QElkdjOqj?@M^*-?wF zS^dx;Nf|ADkHECK=h=qzyH3k^Hf7!I1C;jQH*NamL)$D?@b~?D^i-Nc z``8_d8KZsw_ei2M6?LW`#=_eaHwk6vsL{ohtVuv-kj8}W{AzZK;E z^cO{JKS5U-htqm@Ps$$RLv7>FP*24MDk#ZDvBn9D8g_s(EyJk2qDSr%^PFhQSrvAhujhszeVMGt!oTQ?21=Rk1A3lg7l5Hh zPT+9c$+Xnmm&D6GSUi0ly$V@FhF|n4x5bpI+q-x^d@1%isfKBw4w@lZ6fxO}R_6$~ z>CJI`?|X!<;AQ$+IRRFeO+c}0GEFL9A-USSH~WyfnjPAvEWWB}qAKb~&MCZ4f7S^q zlCHBW2IH~ItqolsM&b0WlQ>)D6;>TC$Gc5=B4j3sX)j~MAG+rWgT9i{>nBwIVLw@h zl#y-y2fT1>2A=IqqHp#u>CwiiD7uGG+7v1A%+m&zW)+GF)5g-tA~nk16MZ%j&xeR0dazim@b1%}n5J0m^SFr1yx-2#D zGP__`#Ts;O!^RdRs=w6DVw(2Dp@^NhqW=>}YVAi|uIY5g$BXK>+^1s)8#$eQ*JHm*G zHSEvLWbE|C}=y{5+pC`@sS5!0R4!KKO|#ys2jii)z{W-KNm|I9-T(9Rm06 zE=HeXUHI9&06*1SLdBM0%%T1-dlavW6T1w634L%@_bD`1S|l-MJ7D6tB`|B}Nid)E z5LZYp(@y1Wrs^-`;oh~OaP#kJn0_XjI{+_1#=N_1V{30bF_?cpol}`h=oi9FD6=yjl2@ht>hrUx&z`^-4Y^n;v?8AYyq|1`k$R2_2{yR{yCydQK%QIfx zk8$1cE3l?tB)$uEqYuyWgopgRIp%98{I0Kt$C`~ej3!ZY^+ufSnuy6)U!x>zz9g~K z64bsg!08!wXtjGbUhJ2L{uhP|`~UM0CN+983+}!hYE%T;D-_V@!An>gc^uA5DPr}O zT(FnpE;i8}o=ra@oLKpm!SZ+*`1um6J$4#AjK`v(M!Ar-d=uoF?S<33&M||e)ofPO zbLJD>1lvESqRSa0>UV7dwBND7)N8A_8$(LSJy^_Yixl9$q-fCP{1h>55ANs;Vs}h} z;jY{xu)VzgToyrkohA& z{B2|dSsoV<62fTV)q3H1dkU17=JLBnm@sDMJl1@95EvMhvD;p&z~4PYID07wvKPJnsZxlJ4#Vlc3MtB-zXvC-jY18lYq+j{2Wt6n_qWkFY+5*h zsjN=LEmAJnXYfs|aIVJn2l|ju^?(XIUh*uvF4q4(gMaR~Lyq%ADBdOm&#Z34qJPKG zxPbSu&38hK-XSLSdjj-)jmERG)}-+I7)kXv#z$M^anu+uSo_Ew^Si5Y={y5UYhH)v z=6T^8k3Vp!tpex!`15l>46KamMQeKPz}(?I_@(U-R(;-rL!LZlhF%NsN%&T{8PVm>bKzYmJ~JL2~WT^#wb5i~A!!>vL7xTISiFZy>twc}VgY%mZb z*N0-L_ZymCdzeNHOrke){@{WoN_b^tDxNDeL|>VM@NQr=teA2QHzw^tJ12eYTsj&S z-WZFM{^r1QtA1>to~h8RGfk*%>kWzF(Xg2_e#O^Wpz2YFEB8LcG9P8MIvomG8MzSC zB?YCs3L!VP5_Trur}|z`Xi~*sQVD&JR53G zB(e$nWboemRcNrV81}k9!}Z)XGG9B1?suedF3*3o(=3rTH*&s$H&FMCseFI;10uN7 zMfp=F6AU6DiMyM$ZzN#U&}2HZ(T4^PxJ;A$YB0;V4EH%YVBbQKywA2`*lb5E(UrzU zMY&Mf;eqL|zJl*VeHfH79*#Wjq$@!@-+J;gsk<-6JAw1Dx&MA_?yZTZw3b2QT4T22 z?j=0^Mwvp39^-+($H~>LnuaWQp!yNpX+!roY8@s^GW?u2`gtY18*c<9BW&qSa0z~1 zegyjO0g#feV4-8|na1+oWUcfUA4HABm<^o!d9)V4FI5{>0N ziB!|E^kZQPxS#HTUO!*6XR4)GV`>fR+*>)?Y#sRQJPv=?`a$Qq6gK!&G7GVZWx|MO z*y6Dfy+5CZL<&VCIU@}8bH$U+7to{cFL-k;3ZKN9LX*`?mM@>pqNkmJk<(sa;rrh7 zVE;`h{HV>I7}pE>wSl-VxGy#hu*Kbf4uQ7Y3$X1S3t^{LKW_c}%xwAO(;QiGX zuY9SE^Y4EAt;A!EH8|#$F&7a;$Ylkfl3b6M((2K(^E-@lA50yG zgK29;4eiS~fTfuSSx`qX_aRl{Y6*AatvbW4E`I9XRfB8*M3!Tw6wJ7osv(9YlhRp0x61Hb|2mNJYUCqD5@rE z;@Zg?+`+q^%{!LMwkeN=;qJXzenSh3iR+8KuW8|n<$Mm_O>nO2tT5}wK$iUG8F>cv z5fzW!pp}DKVW{#xn7cEd?cwZ_mzmdL>g8CLfA=KxTO!9gw;eZ4OG-tfuyDLnWshfi zUBbrGZsfXS2VLFUiE~qOaGu(BnD=NA`!_n?REUlRrMM>{Zqz_fi3xyD z@z2s5Aw!Uf+U1PIEmnc*0ZTADl zKX8N_8AX&`exJMt&LWQivBK=Jb&_K`Qj+cI$6=sKA9_5%pO!yfLA`^e=#abv=fX;% z#;VikKkW+NHLFqO@6%MYbp;i5n!!m6J`-1e0|WeO@tkEQ7Ps8w`ObkDCL>SD3fkl` z?J3;1QAUGrPMA@UjVW{W>5}>g>J53q%>3P^71IMH0lBN#+t5Sozn5yX?PHO6tYDU? zZ8MpkxjtaAq4{t>BLz+`e@}@WA8|v=UIv?0=&nB_>6#^YY|VX4=s1eg;uqo96Cbdf z^8ma0D~a;s^C9)0C#>4k9}hM?L?^tB6IUwHgkBlAV?FmqT?xng26dQ_T!#I;TJYtF zweTfiHmob@jcbeh;)n~UFypWkuDqfs8upE#nN!;_Xy##Iu8S+{Ix-*bH}^+LpC%}I zKahQv8qGS2i`WdK0qkU&Hgg{Qt;ONU-)OJ2CO%&VDmK_m~Y=n?7iA*_L_HJ{AZp){VXR8>brto<-5|pN%Az@ zd$rg)s6l+D6(#!4$f3)pGf8d11d{v5cmJ*_*u$SId6Q4kq+XU(KBp0kid*QAOcn+F z*pHewgGg>uJhcqV!&!yHar{vYjOq24nH*a$_JpU1FFUM6&%_BrjdBTWIKLJh)@iXh zMmFqOZmena=UVpQu`Kt>Y{Ha2oa1yb7^duLfZa;}p>2vZ$vV8i3w-xCPw>Vg(@fDc z!%NsOC>9p*{X<`a0XTJi35pxq@Q7t!JUm+lTichgl85EkZ;Lj)Khd9rE^Qd~Wf1Fq zpbzNCG_WU^CZlZi0o>KacSWb`FyFEh3-4K@S>$*Y+`E;nUb94e8C_2&2A#oxqqfX= zeGiQDF~l!5k=RFj7>?t7M<2uQcxuiNJf0s9v9pe|<$-6g-L5}H+VcFwKkyI#=VEtPw^pr}ct#Y7<7qy;9#06@D&?UVTTI^M@b2JC<2u+| zKKtl(=oLJhZ;6U}tMOdaaICq#PkePgN}RAzT5Jm4fzJNf=xs6rBGydxTfGO2FttS>9JL;bNCJTIV}`q zR#Zc;XAhbB76mN3yBfcNn|bE&`Q4m9`KU=b&K>ey(7B(XXpSwmt*yW+P+_Kj6pzSB{t_;Fc;!x(e=7x}y zS7O?bvI&pvj|5xI-f-&kLs&oDoL;%z$NWiu@TAX3Ocyn_0OZB{nt(CGea2-_xgc# zbS)^(i5AGaOfhE&$cl5@gu9OZ9K7w$&Vn;*x>>xL%|jWNW^ha-u+ zcHs_3C5YLv3)*%YLdWf=>|SwyDBZb=wK%iv*w7#NOl2g6RK3709S%bG(nU~^ z&e;;n_{?x#BAK;iQCI#1I#Fv$PjqT9ls^+!k_FS>8_ecf?V}mr!fEtYJM28$2tOCz z$5F}BboKdOzURI|aSlW1#fwCG{$V#srd>qEx9W7xLW_n)C&H#n>Y~=s1{ywiJI32b z;Ps8%52C7%vqsLsEhQG%??55%7>xw4hgbP7dMtnPjc9@fp+Ykq&azNiyFEpR4p zgA5wkKb``|HGs;8W2lp)K!-c;vbvVVkW+L4mJa)f4R7`6x%MPdJ}!;YhZV_bt}`W{ z^2Z0ymtu&$EM7Rdl{zgp(GKpy?g_sQA*$1O=Hv+seQOCb-)x4`tqIKK=zT16(4mE# zHMg+eK9D=iDfRQr#;H_IL{m-BzkypejJNR>B14-Pf4?3l_(c1Et+(F zB|QsEQW#l-KL(l57~^epCQKLiefkHw!76Oc&_i%!hXk+u8c1R0yqB|6hSrFMcr=gD zMK2WZZp^@_wmdKiF0BBw!GKI5hpJ4EcY5( zaOEzox!y?U5-;KdjftqRUzL&#{ith<11$;FApfU1EJwdv@?d+EFtPoG;4sdC_Kvtu z*EE#G)o#33utT2)zkd$a3+3tXsBI)=y^c1gS!0%|241(ygs%2{=-a##O>cByZ(V)B zH&d`~-ZE;s<%%z@d~unD2(jvusQlLf+SQy+kw-4$xUI&Ra7Tr`SUnBz`pc18yA~Pq z{*<|u8l@!2a__c1o}ZhKcIzI4o9ts$*C6V@Cz__lyJC-S2{P3T5cZycWtw5|?S3Jw zvP?#kV{v$Tkp`=J%{^%b&p>1CCn0S8Vj9ge;3u=k_tXlgXxP_Eyg7;5>p>X zilV!L__9Et+hI}k>+%7L;T~*vqdWXP{SMWh0s5?3g0qYBF=b8-jMe!GI~F1a4RM06 zoG-VW?=p6Z7clPRI9lVm8mDeLfC-9mRKw2`E&D>)rig5bty{k2)_YShRd#{4Nf+4p z`b@ZYb%M}SBg?`}obYw(KA6)W1Nq^JwD-yoal*xR`dIx6GxiL_k8cr@O8r>t`&>!< z<_D6hleG}LW+scbTmws63?aM63;Nr=fGN#6OjS=E8&8fxtCi(w>+v44Y$fnGU^ewJ zjm9Yr&zSSdYp`V8NEYzut&reT&hIvLG*j1rvxn>Hb>=h5ON^kapYq`BcO|A?caEt# zPNAh$DP%EShHl*EXF>y0?695yXFNxN$T_`Bm5nhnXeyPCPo|P=7jpG?!L(l~uygZs zcHzqk`f%bNJ)IUoNlO3F)4~>YjNY=Aw%*il%w$ykcL@4Cv!-6Vu3^yQ-LUNCcDh$H zihi%{hlgYbV~PUzOeGuBrUYebJ`#Zzb3!29?MCTn0}snh8LHu5xyQw5*|lqv9iGRC~h^R z!1~D)zxFW(M#$pZWPb(~E9i#tcT|;4LamEPJ~kt%KI%6bmldK(oD22O8b*16S*Xf= zA02mGATjA5q3>pL<(-7ekTyt@)y0&e0;s+^0^QVn(Z4+u=T*&RCcnnBnZeyc{-V3M zG;o1v7*Q#PPn;oYZOtTQmVkl33~(Et(|o%l4-(rKnEJ#Tt-77rs{u!F)w7j&%QqM5 zRLA4uUr+Js=5@5^gDtJJmJx66)eu*xDB#6sv(WbC4VXHl8vL>jVNcvIblg&aUyt{v z@1_~3eSm0nZVf(m8iY@`ba7V03iRyZ&bbfk@wT%mS|@FXbC;gbe%~6(;(nxMr&@*0 zySBrIs$Fo#c`~Z(nGKr1mkV79A7L+N11p(+X78I`338LyvV)pa1m%Z!adXohY~C4w zyXI`eE-!t|T>gWt_Ls$-xigr_`-8&g)5plL-zHHfZMr!2(qXg<-wvayli`Mx7McnJ zVMcO)+}b;fX8cyBqJIVGFfIwsEzgI}_`A%+=M`8t@H@DT3TxcckJyE$+<}*YhVr`f z$Lt0!fAOg%{elL3U6!9#K4mH-{?lxxbpIXkSdjKEJ?6 z&fBOoxr+F0iuC7i0CnpG-jleFr(8J?anT%}NuACrWxc_D%3SbyTFB1!y-dBjLn(XZ zG`c94M(tyG-$wrc?`BRFpa0$`p3z<`+8z8#&ko6ovAc7qpePt`>SRIQ8GrVBU>1(x z=al@uIpA10nq+$Jpj_W(`1!qpl>T(%=%=<={rL~nm+PYB?{+pZqn{XSAc&{*+NjAt z9_%h|K{w?)xG#S^vv;s!2ec8Ccly!S=IbbX{WVipAKaV-?)c{hDeioq79`<@5jPaGhy%4zT$rO`NR#;5I1K&w$_;AzVAbD$?q$u?7Rh! znnyv`ENQR|(`G}CCqbBOD5eYu!&SWN*rFcn+x-lsx}Rk zFJYZ39gN&=L3T(HltKnLzCQtXORI?O=tkKd1zM_l9R24-)3RTgq(3tr{@8EFJN{$& z_dB1(CYWG5e?Fb(`KevXdue=20quQz5kJiC!HdVM@yXLv@M~TOnd*~qYGMg4$Xrch z3QB3NMG{CojDVW)6PXfcnu#vY*vS`9*=yUYI6LAz?s(B3A1v}jSvzU`1`=kqxt+V7 z%xT6L8|;(63xjoZ(O_w$kXLIc`ko#q+O=hfcT_=)j%=r~+Sz#eeKp=Y`3hf8UQR`~ zM`Mp^9Qb>lGadUx8V`N7VZ+uaLDIkidhS1$0=SB+DejyaW?}&g$1TSXiqfc~mdI8=yGNgn=!-DJNsKv`NPj|YN!3aUcN%ZR zSBsqK*1k17uYLgCKD>i%CpGyVFc!R4$+3OmWrF^RT+XXZfQW`kI5B)ScE}CJ`rJ5Z z|6NHpHKORrNd-#(IEs=lZpWFOhao^ym+n+E)H}bTh(&;%x-1bLU-K^^GVo zdJA87DT~#k+bQhRP_acN09PyRh7Eu32!=kcG$3^#*xxK1pA^Fu_N@aa z?IO-ad&mB2EF_hpO?0JGhL#AvIBVfS_|usRla2G}Qi2QVMG3gsaxE43UW9~%V)RJP zgK=Sr@VrHUr&$9*S8X1QlMNJ-H6G#A$qQ(x=Q+B$cR22_=mpk^vhde_F|Bh5ro5xP z7yf9fX!H9cIS=zDyNkNypVW-@Kccv|Bby4O|@nR>9}|yzhIt2J4l~iQW6op6#;1ZxSP_oK}sY15CjD*JG^CF+|6shw*gN zUKlp@K6Q;&rI2TNR2C8|tUISCobCTfQa9RN(0#f|D2vu+s}h!A!iYyOZ^0IpIOh#J zX!sKbCo}v}S}EK~G-odA>x3J|S8#oWCh3*_!pC*pROU3FB+-Uod)7w0ExnYCL>2aa zwCA4FA18kXTX8*L<8-zJQ*8KTnC%=b&su9`=bX zgz#DoNcyG@mcwP>!XVxW;7r!7ONUXhjx^*?`O989=L_?XykL$g>B7AbiKJiiU5E7lzHSp_)J&Qmu2Q ztXomgxwa9_f*bI-swHLzRN}HepK2 zkDTZqD~Mi}W2xs6zb}Q&#_VO8AZ?{fBi==lvq}zXXqus(-Zfk?@DlvYbVA%B;-#J8 zI9_uIzWb|y>z;oVrkM9;;a~EZccv^Ea92(Zikl zd~aD{^4({!+s6@R@h;ZK{{2C>WC5I?xfvz?p)|hj9mQ1Cung;y!nVIMlIK$o&@+R> zlo)aXi&RJckD~K%=kj~wIN4EAiA;X=0UY*Z*dw_Q zzWe)whNd7imMFvDmI25*I)Kg{YFwY1<88!jz$b6D*@Z6>wBdmQy~>-8t+#9G@SN-9 zS6EE>&42JrXgL}^HbRY+y%>P5LV;tZlH?i_tF%kFGl~*tkXNIpS=<>~^eE1Hl8OdD)krj8GR;ccNsm6L zky64@W|WeSqTJ8F8kS?yceso@PlLJ-#WF$7c;=9I2+OJ+*q#1lkl$yH$#cfDdz{4m zAG0Ebg}bmDQlRLHkLd&leHMRpI_pYZ!Miyuj|F{0s1V~vVVPwhN?T!Jcn(`JEsbgq z*3!hz5b80QjC17Q3pKT!?9+Q+wd)SMmN+yNcn?Y)^x!dY&vos z)MJmc;gVtq%o{=3gj38a?-2Ow1ft;~&czuggVIvg+}!mJ+#G9#nG5ykaAGu6tL=qK z@iBOB;~|u>GeG~)Kk!hX608jKF)nx*A96i9<3lUK|8F3^T4e?e61tcW(utGW)1Ywu zJX~~gBK|jdHJdasm7QtJ0oUhqaHvZPPa1mxf37#Gzp24b#+Na?F$vZx>f!3ypHZ3w zFw0>W9EDxcVYt@28e)>|Ff33Chx5!)bZHvrrkRO4 zs=&Iq=kAeK#P)U!Q%BMfl)6=XLnJ6=(Qqqv!Kws42J--v3I#RVOD<{&Ej0?KlgC|Kjj(`X_ukUl0Q# z_T!VnexL>ph{nJbdu~s3AOuV1p2XWF7oag~ zBKxI&k>elTW*H9z=-#X}TrzkHr%0|N+rn-5$WxN7d7XfL-c2Zy=|VY|PD66MGVZMQ z!Pq8KR$A@OwmXl-1tCGWZ@LBUy*!MsWA0!WlfwnyUGU3{1*F)!n}kK9QO}#jx$hC)gQr4_{r_ z!4wXR0lUBQyrLHyIG+1ql(Jll332i;N9jA{7i`6sa+A=HX>%so)DDGYDQ3u00IVoc#23V|=;!E#|AuUbzKPpz8; z+fQ2p;bZph$Te8rI3Fc1U&3ZPjwu*e!)$k6XSLCxcz4=3`n&uCrtb^IG?(*umSg6w z706`Aj_rfQ|1#nD4#cim1!ygP8DI1Y(+c?&bY|iODjL^7g4q=`=SwSzT^#@whZ#t# zIac)_Vf@X_3DcekQHya2)17Sr0$T3qHR=jjHU)oAZo{(odF=3T8m#k{!^G+uCb}XJ zHl-OM)YQT=&J}vL>lqXYg=4;@1Z_ON7n^R6fwb!v;rZ=HFeddYB$?;a+Vp#r*5C}K z4~@ar@(v`c2-1-!hv`%QSbn;H0sbHl=yUhpZNy9 zh^P%puQ&~P?>Od~_C@XFI~B1cl@Ik`H63Rz;j| zc|4j%gp=^?nsyX_?@UMdr*U5Fea!F)<9MSgBvL+=l+s^d-E95vb91Ss@jmuG zK0{lwWL~tHg7@gqt_@;z zm2;uv$t@K4b2+KIS!2^DE+05~0|~0c!}Fshc!T4JFIvv^K!;NBmqs4`{`(rnG+H7m zJ;OnHI}pm-kLTwxvgMeljyoPeuYfEiRT#m8r#-M;;S`ta4WrFlyHQbqkxsf8UOuBp z0Sl+{W9HUkQKknnt#9ZMPiVwBGIadAQU29>I;+e1qeHojT174A7oJU{SC&w0i4%;x zb3t=%b}TyWF$R8-zz*)NFW%1;-&=>%x6^&JXj&V(_zRNns2UwA7N;k-@=!5qEL9jB zz@D*+^f|wsj_NK)mplvL>1o5mPhMy|CYzF@7g6}W46-R?DDugU?d2(uY(ON(rfMhq z7H5pNIg91O8)(JzkAQN?=<{MWb#>U|+S@~Tet#u83Wn2`httWTbvJiMJBAph&<$)?*;b@+ohKAO?XyPD6B2z=jOC^gk!fL6%dI9df zvyo!n714y7B}}Sw3@#b+!IcGf;PS@|)ZCRya^c;0U6-4sM_k859V?15{z=XX7JRK$ z9;7mI2*qB!z)Fdo*tuAbteyeQe-w*O6F#7v+bH&mmvHRpSj-idq+$%AvD}$zr}aNH z*-ZFUsvaMybCqt9o$|TcsNkd#~wg<~M zDACg)b8MPf4#Qj~z#?TFW}V2#0zEa#y0j8hA6{cLAsiZZcazxdb#$$J25xr{r7ycx z$gBGv_WivJUZJXZbjM~KXjh~!hFkDs>K?4${fGA7vgZHJyh#zErF6D;Du3RIV*Y69 zB>1M251nr>;>f4Z_?bV0#szBA8^I&^Gk7(c2l=AEkuoMt(x7Ut2axLa8c!*l#Hg4^ zJiIJ{#Z-R--?#)^AeVtwWyit3rw%fQh?c+QW(a?$;g_Y0VcXs;_}4FVX9$Tw3j7NF%mWXo*QS27OG!zsG8Er}qGy6Fy6l z-9}WLavCak7sB(PlOUQUiSL}op_`i;8de&^ii7V#CuuTPa_|52>kgbZFbnnSb71MY zKJ52>2FLCzFw5>qq`b?QTmpBI=w|^`)|a9!B0Qq%P#D{}1P;j7!^tEcydNV<2K!Rc z-uMp=Pd|p^5>>(FP&}xwo{nv<=isA8A++sXikho^P=8-F=>I#5b*T-|XHx{}D-WWg zw>0THhtRa)@l?!SqIY5fimuJY8{*}dkh6+3N99P)uM+=S?WKl1Q|gebMAyGYXi?aK zxAd;ziAy~w)|`kkC8|h2_EO90O~lXHOs8kQ<@JLi4nF(@9~C}=93(TH&W${Mp-9-0 zr-4$1=Ge0=omuaG3Bl_^#T$iGI)w5KOuOIIcfwJ;T4x~ivFrYQ85j$ z=}`<6x`bo?Y5mZgr zrCgUmoRDWmO;X-*eDVgYZzQohx1z|f z|31g?i(@0D!WeNboFb#z=lya_#8 z5rX660zlbQ2WIxw;X%b$_&s_x9(|!kb~}WqZMQGV7`;GOACBGpTLxF%K7s2bB2m`S z1T`JA;a&L@bd%EqN$((V-r<2l+8KE7^AN@sA*~S`OP@IZP8GKu{;*9BL&^d7(SaEZ^ z`PKTcTcn&Fl=8xy9WUVV(R(OSTnMFm#^L<_EXX|ilya_f{jZ{3G}13l7e2)ya2>pt z9LsUgrWNF?UNNT?N2nv93#WJ$z{3ZNNsDvF?$PMLzgM`PkX{AG@?S!DLL$!FCXO~* z+L#jkhSdF2DdmwZF_!>XdFwD_ciO-_w-0#f=@mS_Y5-CtDq#{j!|Aro>qt zlBUl5CCi|`DHqL_{Dbb9ajZp96v?5Ug?&;2y|zD)qbd#0K3TwO5itxuKL``QUx3#Y zd-oUf*RQWeID3v)}@@mY#bPU z2ZyfAp!;=^xMKGS9N1b!i{{7B)+`aamykf$s-r15Z!ATxJk8`ho-*XdhYhkmSi)s{ z4f@u@+PWZqsE8?Fb73Qm8CQl+mCW#&4;KgJiDzo1~H2$)4gu;zd;=;?}JH!~CA zhc|aVxx0o0MEuAu(1ZWoe}&(1DnwNId&dk?PkABLnQtNf zN{6|<;AuJBoHBBDj3?JA z1$4K)1W#=IjW3Mv(2i+=bha}RSM3w!hXmA+>VsI)i64en-CtpC+kLq4MT-BSUWxxT zr-+n|qRHkABf~lEsM}&d7YqEcK;$mORi|J;=|p_Xb(z|^JWgxf7YI?-WKDC9v;Aq4 zSm%CccIr|HZ2oD>HmEqjzq|EV<~D?a{}NH!?=BnXIg9<6?Zu1rYv$NJhv>R+B01$u zCS9Z3n0PV_yFC)w-I-b}VMa8t0@b7Us9)D)J}v&ZNn;+*;^uhmpZ+z^DNfbLQce z{y3POHcyS8(@u?hOyR+7C4yF#FMThYrmQ%yc17kQ;AQpcSb+1)0f1~n`iMfE5nhCPgv^LF_^pF1WPt4 z@ZOye!*5~VaZkoNjPtL+iyT+^NRbpT`Tigr(mH{+YVSZ1)Zs2}?zXRPAFBxezcH`$75qW8QJ8B1oP!h8zl3<9xP+J{*f8uWf&31UN2ceaYH* zgeQ)hrFWr?NG9+)>Y1y02t2y4frmBC(1>GB%siHlugn5@fqg0L^VJ+^ni|C}eoTes zbsKQg-BgZIx{DlSxqVSh6znhp)UjHE3BN3P8ejbw7j4A7yEkFCh#Fa^ac2pW5mWog zN0@xi0@Pb_2}#I^W^{gm>&qSJL+47Ib=w3hLJdJW;R`r^8^CQnd+|s9cO3c|k1cj# z?B@3{bbO>nSL-6UXM4%H3&qHNyb4W6Kf2B_+=}{VBk!9ZPWDly;2-DVO_m(4G|7V^ zsaQBya}5<;IZjZ=VR(IG5N?QvVUo%$_*l3eY7cjVw4V;h-VeasPY-cR)FIHKZ8&Mw zDztmz$X0Co3RY8E(a$>x5_UUcZk!>v2i3;E2X5lH4YG7;gA%rMKE$8S`|!x42kcC_ zIK-#4ah&oLR^HnQ-!2s320{kc@i+m-3{`u0mZ;3XUyz zhTKXy{KmTm^AapkDC;Wjn#HlUEMD>|PSt{HjTb%~{0S?ckAO^$7kIVrf#s$L@kpWv zx=f#mVq0=SX-OXZ{&*BDmlWaBdwnR8_W)vtQ?Xw(3Ey?Q!ZUAubmzJw^LFR5IB{7N zpK$>_BB!vCQ+e#^TQm3*qris#o5!9QbN;fE5%B9<3!CS3o%xS@4kioF!l*lP0|QZ*mVKNB(j`;CB)trt)v3YLp37J` z{}je;xd*O01hAu2ojz`<1+NYz+^)C>F4{@5s`5ol=5rle6ZMasX%)iEBzMxwtuggg zQv@ydO^{H&lP9pLpIsij3;)*NHGO2<$HZ%OadN*2@98CuC8;=>+z#lYy4YEIXA{rg z^L-DCTelekWe?%Igo(^wYCniHNZ^_JKP=&~Cah6$f&IVhO@HbYv0Z)Vpaj>0klaZ; z$x9}2eHD1((84swOaN&6#X3$m!S!(>sMdCY=Vssns_%K&ILFU)nafW$*DD)VPPTw; zoD=hrs65V3Fkl&Z?Myhp9qzlA@=o6H=b1L9GV=JuvnY_p%gxO=rOlqTJXr^8qirDi zq!|1C>NXp_CWbK)iM(0pBY73vbMKnw;0(4+ zxS!n%$mT72QNs2OECQ9)@$3fYhRbv3!Rmb?O#kr!=X`R9@&!e_Oz%w)ZIcK>>LsRg z&j_<)4T&ru^)+)|kO1%7hrl>EXhy5`a@O(f3~PNJ#qzJixOFbx-bYoHQ=SYdN7p+9*Pbm^SX1svEf`9*mFbz zCWa|P()SgR{PQL6L*r{!`aTBJ&16hnr<`CbO03y;Pc6`u2m#IAr}3Yx7{5)Zn$D}5 zV2qA4T+6u*8yd>tNU$c%Qk{wC@9Wa!8fTti@qXr|-^?01q}WBDNH+b44(h7k$B2a6 zI5MvYg%^E#8KX0DMFJ+jX zb)C0m=6Ckgr3Y3DFXl}Q*bZp^2E4`NO)J&wSk{Gbki4^;on0db`**rRZ)B}$%JCbl zeKWun^_jeYHNxPz=nSq|EX6zcNd!87Yv7rEX{NSUuFkk^Ho_Yj+s=E!t740_PqICw zLaaHV7gTwtaJVg+^{Hzj^VK4OmLMK8PGT!{&aw5!xpk)>2FHxJES{D<6JC-Cz5YvZ zdZRk;TxULSV~!SVZ?xd;R7qeGZ!ZCA6qw#NEHYjFy@I7~A7VNumSVroJgPQJ;w{~k z#eJi!aY9u%$2gnLBp2<4`lAs%gR!BoD@GP8Q~FFN{xaUe0%$!L z!kgNB5MB;F;kw6;yk33}oAYN3&X_xu?R0z3lxr+;*6Y$4VMVH@|04G>hf^}Bb#E#b zuiJzhIsw9uo<^$&sU+^(!W(DxkMoD-W6+2`uSv)o+$yR-cSM?}d4LbAxOb_>Bb~SG zFu=xBT!(qq5?1_G1A5w}aN)8@UR{#{`^d2ak0%?N^vR3!awiosxxXja#be8v)S7W{ zqI(is`mq&$55!>eSu>XCDa6|rB*QWjXOf8DF+9*30g(3yGJve+&abPmfv)m^r@OzAAdukpg4G9fsW zCxm{wUKp|R4cv1~g+Yy{V4gV_UM?(${L&?uGB1XXgqqWP>4VhP=s>&s(|PxLfAZ=o zC*l6bcDQbjBD|USo%dqQ1<2VIz}`KSg$4@6efFo(CvFTu8lf6@51!4<@2x-0gBIj` zk-u!PAXtZ#{odoV2F^{ltOF$1kHDfiVc0Wa5(dkNlCtM`Y8kYk+F?g5PqhHa!aOiH z48?iNm*R(CBGfXW1g&l2Ar0@bvCZ8)gPqMhom6kg=4RYe_f$}``z6eMaRC>%ieplC z7PM`s!d<(i$msb36t9iP!vD0eoXa-stp0|LT#tE{_zdj0b|0<_-G$NX?yzm;Irhrw zGrRh|1ak}~(e_(97|t=Y8{JPq(JCRdJQ54O0U2;3MFPFqNA`jv4F7dfCxg#i?%6>OV?QKuo>*77VUrF* zvr1U++#@`T4dYp$MgWmLvx-O4_X z!r~bv@H%)rYvX);%XIQ!sBA9UmXD#3{@K(hR)u=s))W6}C{4fBLGv^@$Mop~Si-$O zylvz0nb>t4ZA`}LLSm4i@|RulaY4&MFZdlT0L!OEU|aS?2y}nLuH0IJBAaxu0;}Q5 zqU$W1gGuFeNz;ge2PGaD&{lH4D`>k!40=rXS_K*^S?#5X;UbxG8yLG zEdkZuS@^(s2O5o>q0o-Elqu?mOTr{6H0}q-g_gzG_+rwyRzb(A!}ykW>-d`*b@)fN z?ZVsR>`)7H=1|SUQHo1h*3wEtPa>e)`9aZ*wfs3M@fX+yUiN^jNKiG!}SyE zpy{9$hHp(lr>z3qIc*O(YmA|kf&yNcycIjY+z&*33!tUihDp0{jD^*cF)Mx(%CscW zx=WAXcz+f?x-yHlDB7@+T@uXSa~1r)I15~Zxsmcn8hT%i17%-rbP8Z-vFJI(?@^$l zpNzT&FEDHK?HmVp5OuB}<4F$Kq58FnbjU;!`h032E^`vPHC|>HljiV(YNYUtof;&y zEG5yEn*63?`t16sG3N~9)|?!}%9`LP8ck0wpWV2Yf66*}r~LP=T^FRWMs z@|846a*PP=O3J7CywjMgUoNS|Dca$Xz#6!2eTKtB)LT=<999^^xkGjME~JQUeR~*nzxJXJ$A@S>r9$sM zr_+&NM%X=UP2OReBy8Vi8t1tIX0LIEO`a-D`>HeuwsChWP0iR;{R{QtI0n-GJS_Rq zLc157q{&+f+4KQ1rvJ;H-5GL$N`>)w-!dJqOWwlKuNTNCwTj9tS2GmIU<;=e!#!mK zplrNrTWDH&4C20~Vsh+_M*aMSO-km;}jONLHh@q{H{ z{D8}QsH$@_eQSiiNlWd$f-=`-R)Qk>bxE_417#ET}0kLBzc$4<@1gT za~w0tM#|i96vq}%WuL`NF;F6za$XitLO}>^kDWm8zLul%p*r|%8HQtChNFIj7+=^> zfgcig9|b*k@$Jg8Da=g{EgWN+!{ZRdT1kvltU{l2Q8@Ic0A}mvLVk)deW=VLsUK1J zaNlAa%JhemHZr8KbSlTvEW$47Iq*SO9ixxw(rVQWygAm9Ogpd)htj6BxU5XYO`O}9J z26#iW4(<%D!1s#dvC;4_Ym}QmOsCvmArWmDVF10xfm_PjiGL+IV6TI`39t~nGSdQGFXUBIkuXqljWmK z+$q?JS(2l;G)Zz!K_))<(eT>x_hIdsf2a ze}8zEBQIdHLJ1!HehLn2EP)-bGhxOX3tT+#9%aYP#nByHw|3QWJjeAW&2pnD+$@&1 z>e!=4f+5~<P z1asbtW9G*MIIY!-2`|O?OQz}a#b4j1N+&gXc{LqFM`mDK&Rx7HcpGmH@kwM6k8hc= zke~ha9s z5>JNfEPd7m7P*Ez^>T4qJJ(Izkjrsx`smKmG-^zaC#PkF7$lqxYh4PcYp*utUs?;> zcdSGK*Nr$@w*y2^?ZP2d9|#}R#TkiXP@cP|oVfl5+1zo26Z7t|S$dM7p_qd6*4yDr z!vy&Hhgk5*G0exRg56EB=LhK-@Qp5d(?8EkP%C^49t2y0x8rd#y>_0$Y$EC0Xg5^k zxxmFQb$EF_=hB%k#aB{kqo{2Hw8sAgT6oQ)GhF`CD~8L+I2eKS!KI{C@P>L0`jbuH zW=i+jLEpIfaE6IBIv<~gUXTyPN)aq!=L+b#{ekO;Ov29J4>5Rm4rc9rgL|jH!ok|7e}2K% z=@X%*OoQHq1>!bm9Zalzi_0AEqT+xmX+9F>N6Fsi$I400^wd_sz)Y@(%VjE0tC)~Z zS1i}z4y3{5mUP)!hx}C=aaiXQ?0XUi3-}={v%eb#xqWU{QVA@6_zG1+B`EEe@fIaX@ZSoEK7OdC(T^J7xV`0vPve>C$x1jU>N zy8Hp=U*=P{aw1L4)Z=eVyM(Q)lh8BjBy6_4h{LPq;&g9iiu-g6dW^%_w~kikd7=s4 zCi&8qljHf8-HwzxDoK;LXO{FW0+V}_!K=g%1QnBT>pu-#Hm?;eu7AUYl`3>SON<0+ zb;y0|K|DM?4YUG&g4Spqe6`$$AI%0)s#%aPG$WHARq_Wen=9d-bpq5MrOrCNj>EOY zIXF{*2(3A{krJ|V=(|%m?K1aa{t40OazB~=PCbWe zn-9`|zFdCMaU#l`HX!ry9q6Z|&Hl!G!=%KIlwvm)9`U--ueF`D9&p@fnYol}9)vR_ zU*WHrZnQAOhoWbMG90>v>9SX8i^MSyc9KBqZG+Wb%P`(@8Acz8M5m}f_&LIdvL))c z{l^O&Be0#cbT-qKV-sl2Uk~aQx5iqx7<{u}E!NB}g@hMY?AvQ|8Zj~8Yj?h({1Rcl zr9}<#r_YC9U5pCw)W+<@Ls<1ycGIV|k{1`3jH_)@71q?Z=rAMU%ncPO5gc~^ni zYYWO)8BY6_ah!SEIMlHB#r;<2=z85%$|^4*z4h0ywAqYI{TwNN%M>&)na%ZnOzFkC zG_LodK*?NJ(D>$CuD*v&X%Da;Z2--;W?dZ+CT;(lTk-W6A}!UQBJ=C z|I1E}ITZLF4uAN`KyN>eZBL*cr}OmMwhWIukENn*3u)t@IBM{-ppN_k-22m++ILK$ z$VJo8WUV05nX90vYlbKPbYtY#*_3#x5$0#^g3DF&I1hm!kNqpn-6!NB8QJ+~6z5FRipR9&4QF}W!cC1Btb`|wK#^QaY zam2qPM0?YnNbR2-hzq^ub@^6QN6iDMWGFq+S?G zeGOxIN%Ek2@mpyKPAC~a3P93DM+3R#OQ?OUKBI& zre6<|NbJihYT4!r>zFbWRxjsuzU8C#Nj3a;Xd@iI|Choff6`fDUB2u&6@Hz5G=IHa zE&oHtYp%->NzI8@DWgXNAH9^PjqUmreuDFXrU{Xh=6;Id+$(nv+R@AhgVdGGVEI)8 zluRGRK8_*gKUI*a!CKnVkMQ_HOhE&mu+{ zRDAk*4_-~-EtxtG1%f$G=i-md&F(za2M1ByOf8a+(Wd`4UBR=@CGgo$0TU@p;Pn>N znkWRSz^6B_OgAk~GEMSS$Hfc7;O5GtxSKLnzKQi#$A_RNyT{cY- z6C~~53S9q29-@xO(zwtH{K55;a{c02L5~A(ov{XQU+Xvgan=i;DT<l|7Iab*>Z);NPJcT+_r$TUb09&DYp1ofM zyxH?dP0uLg&FIv70xg{HdeNE%Ol-_)oaktYL8naN<-8$K5zm8&e+ld{mx;M4sR#G$ z8hCOAIsAu_9(?&J68v@f9RFHb!;v!hVkf8Y=v3}2EemD z3*)QCV%|1UFg5%~e%je|{E{hYP02^M`am2!-41(&bZPA7C#W$glgnmJhw~?t*nx91 z*&ay&{^Nic^7~~=N9TOzdgfW=x=5U+8~%r>CIV61^9=(0GE`tz)LlvN*V55s!zb%a%yzP zqY3Yy=J-2O<*56!5SOf4i{`s7;zKU`ov=m;B^5+)Q=k-NUHr;6Ig9e6uVm6Lbv{M4 zOT)E$OV|m$Vbij=_K+s}m02|<^U66Nx~u&l`+QB1*O>H#Z9Epo6gH0XK5v(&wa;5| z#R~yE>N^a1b4sCl(^z)w;#0VML<*M6dz#ke06**DGpf4dP2(P}LWd)o`2MdEwe_#iB{OlAwH(e&JON35Ir3ADPNvDhUK=xj#uzUerCy`0#X_v_o0I=@7e(- z0>*Gj?il+ZVUJbsPD0Nd&bevz8>fvZqh*^6h#m=pH4Dx#quWQA^b=9O+W0Hfyw#SH zKH7m(zb+f&sl)sxd$O=Q+#c5M2veUw72oP5v76IovvuvEc)aWf=w>}(f3$zo&}b}) zsz}h5w#T@*WjW5hQUV26O!12NUA7O$gY>m}obk{Jb{f{=%)i|r=&%45FD`~tjz{52 z{c*7J*#Od_<8kK9W3WHMnq_=kfVunZ;hc&V8{Z_03hR07%laq49vP4?_rE5~h;Y{- zcM>a-C%f0>IIpr1Ed0XQ*J+|GqiYOl+U8@?cxCh{R-xyDr5K!43ki8wp~|YCg|_-I zEe%!5mKnxv-{zyw8evR^Lof>+0LAywjqqhubF>4m_N1aumniO5;@*q<+fie+9a<$; z!SJ&z)H0fbeb&<8{M?XGW(YNOB=Nv5DQemzNIADlQTyj=jz81`uZ{I#*@9pAETVK|t%o{~HTHw$KLq)x(8qSRCT8_O<9(Gi_FT;@Fy{>@RKhQ}qShQW4bl~gARc$mv14{7Oja&u1zqxxf0N6kEdL9M?^<|YB)>q=oNQjVy*yhep2MCE z8?%>Hnr!NUBWS%#4K3zeV|eceWSDFK1Np6>yr~DxBD!Hs^;hP{li~Nv%kmW`RMXIo zIQ;xyH7GYGuuUo#VcX$pOgl0Y)+sdMw$b z12*-a6E#0mAh$X7m^DX}4B#%xf0IL#xrcF8vH(WTNd zyW|U@;f5C-Q`Vv-0&Up%HIg)-ljso(Yrrd`Rv-W#X*^R`>@m)F+N_j42ZLB$Wqb~B?> z(=svrK9_sDevD)2Gau)x^+f4rKco+V&H|Z297Eec)bNyzH`>{H&3GK28U$jZ_tB(GpwfEPO z%YT9>-VhEoF@r4o***O9H5A9&n4oZyPI4;P}F>JL1%R)Dk;1tIs$ zODt9Op=-GpFtk>j%!<=c@W@lx;Bg*zzgs~42g|@iIs&$>GY7}1!_4-T3&x*zBJm?1 zQ3p4X*w6&pApV0Xr4(b((h_|9@(J77BSn+8KOkSJ&yXWMkF0{b$V1~0rkZ`jwL``D zDY73tHh+Yt>+hg#t}sNWXW>YD1yxNbM3uY+ls8p?WAZBC_W=(S85Y9b>iy^%5kyUQ z-r!n}5&2(Y7iFKnOd$%YWRN6GRxj3bT@TI!Eg*_Z+{c3*=Yv1IMwu+Gj?&-vT&MT5 zH^1}l0)G3+DDG?-jXwKUQHfR`8tp^kSAE2z1JmJewJs)deqPOc7Fa2C5p}D^(A+Nu zxa{0jn!NQn1#Ym#R~+M0c$=-f>Q+URSHGRE6*!h$n+`$h)F8OoBUel#Wx z=F;nWNxtObT#j>{h5VOqV50jtnBc|7;JAr2M*Jj=Xi3u;`8kvjH;P$y?Yu2JH5kQx z1beG{IPC$KlL|LQAKzDyx5pM*+wFefIi|n!nYUDg?S+iyJA?uMh`f`_hdQtN7=gF zQV6t*)u z0jEaZp=tKB_~y1P^zhd;TJh1EZup*nyM=A|Nm-7Ds=~1`OckE_*z(jS&Bp4Z6{rLg zDW>WO`nqdj-_Pb#B|iU>rUT(t)!U!-n6jd3KocO!FjK2u>W*B{u#Z<-6v_#(B$=4lYN|3JNdI* z&8Hx})d_7Rhw$acT3Yx&iq12T%kPcDWJ}6SR-`2&eLeofTijt_5QfZ3%pZ^OldMD31=RTj$b*b~;Hz6LIBe*}6cTA3U9Ye)U zl2jjb65IE^!oS{)=+4h)Pad?UU7o{9@Af{t{o??7j~qgY6OG6wXBEyEQwMiH%Hz+k zv*>*Fe0uh<2=fg`q4}F7*tj=^XLOH4(z`Rb>qa5In?9WuyAG$hy63S4R$zEXUyGr z38(58qb~0V-jnmkm65;L>iX}{>05*@Rq3cRdp#OeHp5WP?rn3e;CuD?DE}{--R5q2 zoRx_O)i~o_{w?kvsZaLHo6z~|1#D?Lj9x}tus%5e-}-RoSL6(6ir zG4s3}Df=f<>Oe6p>$nZKOk3c}qDnTl<`Od>_K_vtHRU<@1oE#rNd-2Wv1D^LnkNZV zv22ELDcV_h;W||aDxN^9QA%_(U@FZH{R%I}j-#!KKvQ{-q(FSOU`Yprr9r0PxG)ip zHOr!my*vIIx(Uz76mkB{P)so?!p5+%G~q}%3NPN`*238+o?%YLbN#41dk4wqzs30r zIiH~B9~w_Q%4Tw&Tx#-5wD2l|sVZ6cXm=@^E;phP+QVo{votCF2xR}=RK=YxU-6IB z05Zv(PG4VLL9albt@&4uH{8ymzt1*!=_Q8YcBAl+(O4)t(FFhWtMJC2{rFh*Jbp5< zM~w@yAbmRoK5K8r|5RPMXXG5*e4K|Z?V%uB&iF-WJ zV#!1b{gX;_=V|lj{V^^cl7+Xm;=n3@J|-;9#wi*6{q@oTZhk7lgxYA%;~6bnDVZg_ zytx2=Z9B`?zg2Mdpa~u4Gr8x3#*?e}Mbr<=LpL*Dv<ioe3T#5NTzOUiC3lcKSA*UeQPQPYEcD9s#N)cUiG`JCj!v$IcO($yaj_ zpC7hik(&qUEnC2Lm3qN^uj$ZhWkVZ{oJdAvCH*`cP64HX6u4(R#ilsXXVoyQD2oT> zXGgJVZVWc=S&4hRbMSJnFAB#qTT&8`>tZKD!p{LV+%aO=Yxa9C~? zX!uOPMV0FG)pi_~f1bi!FTGgO-p5kLuf&*Bf3RQ}_gz;@aKGU~>L01hT9Xsu-~t0w zF7?CL83q(kAx*nyUlLvYl)yp~Z7~1s8&PV3IXivImS1C2T9?iJgQE=K*HtzA!1)e~ z_m87L3P!l?_Xr#xA&I*_+=d&zvT%Gnz~`u?wAU|z>KCmdvvZ+B!kH_=p!{6nOVAlS zY-CTK<>7STnlak7w_s)XMNBOtbd?IjZ&Ul>ocM3BJU$!)*Gc2(E*ab$_8wGp!l3;- zXGLk0;hA-Z@nQT7{MLU8J!T%jnNt684vG}6-~FB4vD9JKqrKUN7HLu6hk5wxX%^<4 zU5>SHVkzW-J?#tMLf^qKLt;?ZiLk5@yu=IV+PIKdm~cBqy1g5V7UVtD3me(76&Gz$_U-D zV}$un%!Qsu$(RukhY45Hptj~LIPRGTg}SM@>5>^dv%LiR)(lloDv?LpU;Ge%5L~|A zfuE}4c>VZSw3+0B!#&qx^08)!zA~IfxW}Nbd?b7<@&@9r95Hh_j4@HiJ%h}!u;?T{ z$Wuo@i`^Jyeht^uePC7{iJ&R|8E$K;z*DJ7P|~?hKS*Q zr9)_1IwR}K{ytIhZ2tAx+kgZWnMqC_SgdjhK3j!It+7fy~i#; zPcqFtL+N2rH0|(p);QA`tOgE2&rLb-`{_k+uUBFQZ^Cfe{Hu85_+~u0cp?V8A3}y# z_!*;2G_K>>nv-rv>D0R^1XqKp@|7{A@O<<~aUELrAD=N?Is(dr8^I}gKH}r8tn};^ z*tBj8y%;%~R*#RPt{gRb6GN+%G1OWcD~4t< zv6t7FvRNb=WX95m1Z#?U7Az=7_Y3WJm9ftBGa4pe#gxpNwH7<;Ad73&$OB$(?*&i$2>ug359sjSB+|We8^Ermv$K?VC?-w z2+E5?sV@gF>d<7jbKU_a77DcuQN(JR?@%A~Kldv4rzwCxzQ?#L~%?D2E>_M5W zSy(po6|>iAhXXAsczfPACgx-b^Eaqqlj~>p=hsAt6ZzqW|AJWgt3J`4qf2mkvm<<+ z*uhSzHnP}TV$i+#6Wez6DpSim3!+6au+~zF{86qU=a9H?QLw^Fe`VmN6o z%!jn%Obqy*h;a=K&?|a^Eqj+!Nl`j;opVieDEG3+w>BCSPOD>G)^iA7FHKMG2sHV+ zIVptSWiR73p!fM?RGS;gq-+{RPthJqV(+o_<^ys2AkN~EJ}zx*7DHis0P5kMLIK z9H#dUqz*$yp6YLCfp;ZszR0#&d}c2HqWIa81aA2Bg3XpoM*m6wV6^xnNGhs_|L%{0 z9qmV%tx6R%2uqkkr8|sWRmGZ&{2@O4EHbNsbV$MiZw`sT!7U2p-Mn1%S2lzhn00{v zg5FeNsbTo%NmvezPVdODZd{;S1d z3Aga#?wwd2=YTutZ{-m8fvU54=7< zhO=^>+jppp?J8XvQ z5{5pGte_+tCBDx>hpw5BH)I%^o*oGX=9A#+5f$@m!;1t%drQHK|Gi0LH`Jz?uxDaz zlsj<-?XR!JsVQdgGd)S<;-=5$tI!|k~ zI64=SCneL7mkM+$K|tLf+Nj*~o9#9l&4$feCRl|Um^)~#Gym$OCB(ier>q2jYX7HB zdmS3!zfsl9Z^Z*#Ig|6_4(4P2nJ#$Iri*C?`e@Y7xo$_UP{4ULVNBu|@=@DE%lUQp z)Z2kSJf@&g-+QQ1tOkk?R%-GpmE@n>)-kIHakeARif#FJfX|XH!644e_u3^TlsE3;`!LQ&Qk*7)q$1u>H;!a%9#2&OD32mYu^n_qq4GJH2Bj^9DO7hN!XZ7`b>J&`3y4`(Xh zIfHey7`ygxAL6rUxEW{XkQVZiBW`Svf89Oy=92*mA&mD*n zLc*tf!TIqXp;Rgl<1duL`OtQ@<6Q*p^E*LtH@Ii#l>zHj*k>j>dP6i|Z8w5C(bZBi&u~MpwR&>^y`?%)@fOb7GK{j zT6Smad^H;7@E9qqg^j;;m0#%K!0?73Cw)7Fj~ z3s#`(0(Trfr~p$}>QMa0*VxAU@Dtv-vlGw_qZUZwVe9)SpH&QpHphTM`c23$osE-D zuECT|N|gBME81}$_LhC}=vJ^4e%mkM-8pF-xNRXQ>Rp4svp>+Nq7G_HbfZmjD|v5w zy=bcAJkje?o^5`whY!X~1m`nq?7H12)777>&BEQ)*rmT4MBU>nMOW$@P+jW)`W$KuoV`rf4^E~L%QumN@=J7l9mx0hb$B~lkzOdg zMEIA5ONz$Qp#-4&wlnDSG!yE$Q-fi5dCp{rINB)w0r~D~n$kW@NRH_uqoey_S;9n- ze|!WR9CDNm7*j4vnX1JsZ!e=h>!;X!z7Di+DC2;*>v(z)|8`EWqQu$OG$Y3g&t!UG zQuiX9K13gl&#fbM*^8tl>dpR13*3gs-E*;iS_Vc*=U}GwR&=aOL08VY`W7CHUu-04 z?5g8%BSryVExcJWMdJWK#y)IT=wHQMqZf9?}gdeIO#10 zFOSFQou&LND-j~&)!9}ii4vsRBp|QVp$@NhKraB*i!UN~g>y$3_e9eaS$JSxoPiYF^ zP56p~kFcY@3Th@wVDE5EG~Qu}^ICl1?!^vV`(O!GwvV9jAfN!b|IlOPDOeK~gn1QB z7?8USO=OdCTGnbPdwc;?X4tZGKS!}?UFnccbIGZqo^(XbsJ1nVU&kUiFF6FGy<)Lb zx&TiX6>}bM492${0TZDeuO5~mr|Je=;hc}18jmn?TROh^DSR*2aFAZnIwSj72Gqa!7KF)#Gogd)l-1GS7XEQo;PscwAb-d?qjL!{kf#}+4 zI9J5y<=yQVu*MJ<-`|JTYm0&H9zk0x0%=vVI2DJd;JrLU+~Z*bHMU~lW3LSIucBG3 z<9tlfOGJgy`>?d63^V1U(EVRFcdd=4sjA7e@}3%vvseM&FId902SL!kXgVgEzrypI zFJN0=8_!cM$EOWp&~%`a#jg7V@@K|k+>G%UmuZCl*96S#nTeYwdEt~n3Ro7?i7OWl zryD_1JhycbH^wW_Z+OMeTte`!`$HV3QjR;@=3>Im2+VUHj|=X|V1C?vjJxpyt&==O zTQ;?VYse?uw&|-`k~fM{JI6Au!o!#p;e&2xD&mo&00VFq;jj(|n`y5RGrY4D_CJ@axO1p}lV;PKi>9NgFg!t*J3IXM%`FM2}G zkyI$(zL8&N3+$h@4i7w$VlOvOV3$)rvTXJ0D1rWf``#aahV5gBI3{ZRz6Umvztx?CJ?|xI^j%cAMRSmBG636}WTl1X{ms1fA}Q#2H0VQ2Q$s-}0XH z8m(n?&RdOcsh*&D9=bH6vIci+MPc9e5t!N1j2e|I$?*OQQeDe4E!?HLHLQ>IOizSm zIY7V0O43HV2XMtq0pCOp2K~Q&V7ZW=#aK$hp~bo29KVHn6y0dVfsbf&Fcy2d^DslL z6`eUkUu(8FUdemO-X=_hc~b(I)0rOj!F4T!6p7)&VRLbX&usj9y$S;6+=8*5a&Ujm zAh4W108bdCK;OL_urW44W9ewtCoK=-GgMGtGEfkmNELdf$O(tGJOZoseXOV|2M7Ot zNIjQ>X#V3i__#|Bt)mWO!$~=;n#9j7>KBpit6Y>=X@F-drojq_JGjNeo9e=bQESmC zbRJbeJJlnpX8A=-akBz3i)SJylT|qO#zh2=TR7*LJPzO0Lo1(&3yGoQY0$Sb(8^t2 zIzJxbh75tOa(wIT=$jZg{|j4MzJg^)9^|MA_};3yclEc2IKaZ$I&n0J{vcIzw_EcF|TGZnsqwhl+y0L0sqvXoBIP=*r2zw_AcFTlpRC>wPvoNU+UA+SdhVeqny^j7 zmSFNQeR5J%6v_0WsQZULJN;uMKUWK80V@u}?~(?1F>@&4A9Lz7tiUTl&X{#7m=vF{ zp|g|9=+qH6()baCPo7jmOwJgsqlKY(dRZ4Nk+;L7 z#WFO?whiwqhT!gLj_7n(LI{dHN8PpZXw9WD9)zHZ=3yBjv@rP^> z``9)JUVa*c&sPV4>v4ak$oFpJX1t`e4JR;S!%iAl@sI4Bt%a;8?i>$zMz8$UgcP2+ ziyhH{U-kds2zl=CU7H2=PyT~nUm^ge%i@VrP5j0)4|`$@VUheslz(YTI!~O*`oawS zXJ-q2M_oXdy9j>!n34YXG+3xin_{YPD!-nekIBRPGt%tYacgS1zLq{H zf5aX7*{m--j<16dzZf{tfDuE4Wm~ut(liwpTBvgecQL+>(j{vTp1+LRM{(083OhcK z@TX(Ea7(@eJXgx&*YT_Hd5JiFxAMiVrSVXBGYPt9-iGh87va{QEqw2<0AC`Pu^h=K z)c-M+5(*q>zym&izU_?WFAu>Ioh6Wb&jVjczr}eTlVOdV5158_vY<#myl^uW+YWI1 zwuuvJOZVg2d-60?S^^s0{esC-4?re1gO%?y0_~8$p#8cF3`WnwjWSm-Hz|+ZIAseD zSDe7CF=;r9XL{~7yhNA97rp7AyHl2_L;g_%*xfh=&U zx{9~AE77glBk|0mTBuE419Gd5p)Wt~0j!zRpl1tAY)_bd4#Vlx1=^ z9Uv_H8>+tgfm=+)vDI|~eh6rX=Lgrr%ssWZWat3WE0)5M*Yt6~ekU;dP>a-)jp^go zll^8Dn)0iG#_j(>5u@ldTbX+!o)Sptk8X=H;(I)b1K_%@M6qvx(qYwaxghmo;GywyzJ-8&>fnMbF$R&m3tsg3z zN7JmgE;LSWC*a0<6w~g1kw;^iZKBT*IFmr|zDkKia zCv_39Z-FeXm@bE_PJF?FfeTSCQ=Fzm5IwA}q}p2I+}78qyX6KBKahrTKi=WRj(W`f zGYTxdjp38pJvd|5&8Bv^Vx8C-G@ECE&YZ1cdRLb|o_zw>ZMD$*rWREetFq{|7e$3; z7ucPx^;ldHOfEtH(RGzcJUcQ43vPXYFMW|{8K1?u2ZJe@^PqawB2e??WE_3-5!KF| zCD=^S6Gmtlg37QZlpZfDOqEx|beH?MTr`f9%bod}y&u+RxbPfqJnlU01$US0F++<2 zo_!dH%T^q~0X>6pxS=d{#xdL#u$WE<`P1Ed5!9h(0S9vq!@}%rcpoFfWRHJiuXu9J z=ZF}-zI+K6|J*_;U2-Hkqe8}8<;lb=lSk+bP_BI;9UFHAH`zDf#&eHw-2EH41zuv& z+XG~%zkxtmS?HZVNwCq;!@@U_7@d6ut{WT#{fS+0ZAT3FeraGb4%Hx}j6erVe>@*q zfuk?Yp?7mP(%z#rG^g+$rHC$q?}bx%?M@5kFCK$(U*ExxFD|%wz*lgU9)fGc7o$|` z5d8c88oI3>LW&#oN%odM%2nu+d5kf>Dd*kkIZ5beKOaAvCZoiXHz-t{#4-LQc~OyYZxWsRvIbL$Z-w{Bo__XNS^&;_WoMG-3%HBp@AVB3o^VAf?0W?y4rrPFBg z{*p}ltE=eh=zJ{tJqn-C<=yVdE3stzakMzM2*aG_f_MKvcqtZ)uap`gD|j$w%$CBY z{2_G5n4jhP>X8|mp>3cco>sbsn`Q6d^Vd9sw6z@HILkoXm|g72**kcALmk|z8-ia- zb+MuO7Yb4v8etx7z-X0;kU=CQ9I4>#`$-`XbZg~CL60e3oN11IED7mKsiw;D>r*i|a!Z3-u zuq^TP@FqMZpNd-qV>nVa5@vffL8QSRXqk8v4*Ish7jG3Bb4!;je*eRs%TL*Yb7IhI za}<24)ajpW6F#$=ORlZUpe!c@mInp#xof|uE<}fQ^)7_KOM$p+=1VM3kHNARN3b^Y zA?`Eb9pZ;lG-X3BRxfm-q~Wt6TY4zYS|5u0T|dLAlkT|Y$4wL!1mJ|j1=u!UpA`HD z^WNZ5p2y#b=cinT_;wGh?My(WDos37)(Lq$53*D|h(0X~rqTyJ`1sjYdN6bewVho} zO_u#=>N9}G=XRs`^pQAb@(mc>sEzK6=c4gdOZ>^1CVn-gcw@37wnSaTlM0RONoX6} z!M~*|-rmM^zF*nB{VA>p?rD@9)lejwkyCW3c)I9s-{0Y;ww zf@$o{w&8p-y7ucHViMEZA8{*gC(2^d`n^v7*B}Ub7LU2~?z^TbeN)gBa=HKqq=B&fX5l=M>u)58fz=|H=RkpEX)xYV!C9H#o=%Tx8F z=Gg(4yb9oy!#v1{e246Im4K}PscG%FE@8qN$%8C${TkG#*nJpoiSEEM*gcEKVI zN>CkI7_bup~P=1K@EDbNnv%jR1!}Swjv>`jIBwo` z9J|3Am7fkLrEni=pO!+=juq6Ebc|xFmy_8iUBMt^7U^?Wa1Qq(G4Af!*E0tfe7=Vl zjR(@#H3_uMc`B~oZ$-NdI6o@o6HDOp>usO3X@uqpMjx#4=APYjq=do6gy~pUnu0HW zi=#hhkCfHyL64|#aN9EitnNs{TQcIF`g>$^>l8_D8bF#A5<-FAWeTlQH0>OdZ}vsU zf|cG7L;1H_G~Yx_cjpDo(-q8RL3LMZa`U_5Z^ zG%oKqr=ndCY4SNaq5Gj9J<~T}rX5Ah&36N2+*nSEc!VB?Y(R%3HP$J+9v6&C!CP14 zX;xMq=8LUBv!w~Jr@kHhcZK2Hj}`FJ#SPsC;+)sFp=wJS{O$QBGB`Ki^c#4XiA`B*HegY*nUu#B-f6gu zX~jqIK(3pxs#j0YKXaEps>@@X9DvM|JIui)OSC+}QqW!VqjkNi1&HS>^P$S&S-XX3YqV6$2_ z_N(=gJkP>LZw|-5M~{(Qi!J@x8iSjHAAwwd4Qu$i0i1__qzmuY)4kLzT)Ua@;SxuP zs1O6~h;g)M>sB<7*@eDaRE7SWIEs-I(XWgO7;@?itgX?*zgqDjt-t0 zy^H7Ax@NEbcyZfT}nsi+JNf$RSF#+YcNJ<_xnv%Jf zynxv2M<40Zb2W9k(RC9Xt~bI4nKf{;BMNSdC5iIv z7cp8dfmlXA3Qg5`ycZd?@r1$Ce&VbZJ-!R+XjDH766TjTT^H<7XJfbMzY} z;^6qLGPeG`D?8`**6dpAUD5G0O(vtN4li=X;`^)n@PrGa``mqNXZjZQUP%STv5TR| zn72iUP$B%9nDB1BnJ{-&3I%={M5bagFny&K>zi0Dx^Pb%JLdfnEplhv z5OH{<>Px*nov^>;A#3YigNI^W*seF5MXRT761)N}gtfleG^ItH{)siASD6+?wSE%~ zT-zo3HNuR=H}1gdtFOVJOv_B^QxK?BE&y51h1j7n6nAQ*;tthMcwhA;oIfFnP5zhI z%IobIJZTM%{g+|3bvN%BFD>P{$m7g1eJR`;ahBP7j$#8%-Qb(f4)BsZ!u}qzfrnpJ zV8o{WOO{SXRs`7_R)euow+;Cani_~!Ic zbg50j{w{s{hP^d!7HG|9_R3 zvU?C-?x@By_qs7TBM7U~#^BMdJYT)41aI%&4DW?dwR4U6qdI@q5V9T zdc+F%j*WnMRR=*K@daAw&Zi9+MyZnzq02K>OtJBS4Ub+z`urLAx3(D)_KIPE>@(Dje%RPb@pj;}wUQ<2y?}MU zTw&Qde`u9ZhacO|Ly&zML_Zt}&om=JciAr9r9TdhL;tcPH)O$TaU}ej_XhrIgg{6~ zKhI)4!*Q!xP+!prY^S_s-pjm1uA)}_-2Mh*6V;&fy(t9T8H5#X(QJIXE8Dhr2-`Mc zxmiQ4B;-vVil&?EVC{4x%+wr+QwQm?--DVtqg;wU7=(lLUJGP#A*gRMndXIR;{}CT znCE*OBCoZ=ck|uQHSB~)*}4}@?{z`m=g*-0(GF(1OoCkZsVwVV00c$2!{V7A%tXtx zv9~20$L9@1%f(AEtJDxPM)$!stNU({cQpCDuLhPBUuK~Rh_l6X@uTKlmZDN2a+mGJ%+6+1S&#^JZ}-D|!Gk+; zqnLx?8ajNZgvMxIrOMc;ROZNixF)`sF`^mgzS&OYj%k#BMh9edR^d#}sC{z#6-sB_ z!0v@V*%IDwmR5+y%Prxcv+gVgN;hFqk})b*eFuwmd@nnh|11~7W!FSFM@$bU-hGLI zZ%0sug%K_E*#!$vX@HHDEqwL4$|Si9Dr=7<%X7SdbNn<(X16bNT+V{$P4ST0lZaN= z^r)V9y2TeSroV-kv1n8bdQCbDqf^tsRU?e7CM1)!#yL7I=^}iee3#mOA498y>ZG~$ z8AJyn&KR5q<@Q>l<_9VesJ@Xo-PFTU>ozEQ5stl*Ih=bDjJJmkp=FcuF>dG^&KCFx zrFIvvs=JMJr!-;LpB9Ky-OU!7*h6x|3wW^n9ezHv9fwT2h6`5aK%mzJm_0HVN0_OD z*&)vP-WiEM@5N%Kf(o9WS%KYaro)Y~{Oq#Xi27Y6g%Ax(dY`~EB;HOi?e9CB-CBtC zK?&HfUmE+;_TZY>YB*iJ?{-;8Y$zUz;!rH*+_`%(>i~d^`?GFBONayo&R{ z3%kVQaPx?%Saantt}6b)Cc1IHYIHBWORs=gU;e?fg?Y@O>nc+ZTY^W2ufh!P0?tNQ z4XkK9@t1?(R z;W-Zcr$7(qHL;6UeX#vq0LsrejAI5&rFqBQ=ycc=3~)Y%mwt;#W!@25Ja+?G9Ehf+ z_EY&hJ`?xs+5p=BcH^G`skm+~-+w;O!AoOIFi$B0{96X&uaY?s^bvUm{uE5f?1Qfk zoaLeN7Gm`t!o(C8=+15=6B$)<;5oDY*4>b8G9L}|1(>sEBrI+e(en-ONw!@Ce?LEf z_83>#v27-^fgaYHankI^;_oo^L?axQ_yH+@deJaSpn~%?sJQ+W%xzYr^wz6%z{6Ph z@B0RxBV30aa_O+3j*rSzj45XJM11`CD|)w7;*$T)p}vTBuAYeDuQf)v;Pgn$=A5bm zwZH6e;$f)SdWj_uc}DBcB=f!gNE*5C7nb>5Kx~i4-xua!#)VkiKl2uBQ(h0XCp+2u zzsK>{v)OpTu>lfgbNMVhfQoy5Ld4h@G*2%fr@a+eZDEM7x(acVUNcr5-+)(K^Fgjp zm-N=hQRpaFcyYEH6f)jIWq&-yJ2YX3P7dzUoj?*FR*?4KNa}eoPk6moQ*dk>E<9ZM zftEDBMX~Fbv2XAJ&^#eY_adTbr~NvZ)j9wZd<@VaBmmkTYU8|x0cfW!PPhljT(;W9hK0GHWpfz<_&IG%*Xm!ojXe(q$OT8@jKzrZIW z^65@|KTb$J$1Z%2H2bVt0xB0(spC}vyD;h{?pt)66uh%=lIlP7b2tqle9b&$_Ff2a zbB9fv{-VeXsq)N7njxErThzDWm=hjsWMH}IN2UsBjN^xQ2L zsqNC?F0A(${N4<|TEwG8sTqB-;SA{T6__5aM>S*SVNabk+L-#{<(_JGPBG1N-33E3 zs+=i2Z@fxhey9pdx5ZLjnX;+pqL+{zF_vELIfC}%XTi+ZRiGBjZ^xn%BV18r2Zh^SpsD2z>_6^K0UodLLXbX8yjMnLe#c0B z<|({aJsH<2){$e!G(q-{74Mh+!v$}L<1Cd~Fks|O?D{%}Yy+Cmoo8R_@}^__;W@PY zC7*+R=m(cE$yUEc-V+{v@qn|S|r?qT$e zO-7?70|ljTD6B}96)Jn8Nz#9ypjBQ*gB}#%q9-n7<~fSKOkB&_L#?sq-atxhYX9}c@GM&?T$GQUnRqfhv>pN|1>k+Vc*A=)iSqiEiEyf>TcT#g3!}!LPZ2q-5CjB46 zS-YAu6Y1;{RpyPvr~an^V;%sftYPL{XWW`Q3Qf+P5fnGC5tK{ig=Y_vvArPxCG1sj z+bBh5RyUemd8iKFk%wTI($#)VIyA~ zCwIcu4=YKJd$KbvIRkKu9Nrld3;RO8fNwz^oo$S!uLF=axJFTS+6J`5M20g}>nI<@A{!egs=p(a5S4sTQEb5#e zfZ9i{a8|qq3>$19q$+lhXZC8EIy;l3T{NjF;|+#NO`*Gwv#@V1V3&a<)$`~5%BVLue$^TLG5aTM z8qYIymc{V9RvNlnj74*Qoxv%t64?3o3zTY0k;K=x7#YsL=TjzPh5TT&N_q*7I~rMn zg$k_p*#QbPK{y+9k>Y0>aHsZUitBWy{5~~OIU!43rbkhJcPRSYcmR%tK`e2;zo@{j zi&lLMB2ngG`nAXv@5p?B)`QuA+CymcPFYeK@(nY7nejfDBxyFw(Elhq)3BPpun&h6 zp%M*-l2kO%aQ1rE+NL5ZnHr=BMbh9Og=E$|NK$Dcm9ZoZQhTjRWk?AL85+n~WDH4& zcfX(CPw%;|^SSFfr+wD&{C@XcTYU!0eR9TGrmFbq>;kxPQWN;p0@Qu+0Ce=-2lKVZ zAuO^T3ZxhW_??F*=3)l{{vzeD3iRKq7W8&%GCDk9DQcW{3(MbhAp0c&q;$?hlop`~ z=Jm<&y(3T3q&br_(~9PLv~s|!bA~vBbU{s-RnY6|1kZ6L+IYGK_QtFfnW@w8``c}J zo68WqwC)+4&UXN-)#Ks4tp&PJR*vVYbJ%Fo0Q?83;aQ_r;t8XgMJJ689G)ufzwgJP z{7D3FU1frMYkQ!{KL(~{I-<;}m$_6Jf}3A(_|?j805`E)w7Ct*B8L7AeUD@_L(z*^ zJ@iEF4Gi3(2Q{0|a?f5c&Q)@kdz2E)-3xWZI{PhfUDrhHbAJi^DH3x%H9xpIr3OwP zc1u404&gdJydzEH50QUvV@RMvBTQKl59Z=|OyNf&XuQ9I_IC`I#H=1$N0x;O}2Da?G2_HibfJVw82ses= z(Cgb^%W`!ntlEKmHxzJzh82L`4+o!r8FtYY{rmxvZKkF>VBM*FESW?_8|+&C2o)FDCpTlky_S4Zox+$UtjtN zu8S^CuaAe}(D!9XD9S-0R;@6qS{n@B+98(#;|Y~g6#32j@LGovc$>2`WL{AK|1X1J zU~x4pH!NUY);32UzWfH^;g;Hp+x7++wD%N8EPGrrHl{~E1v zS;{SOW}1bcZOef9ECWkxUPISfkD^TuC2-A6iKK}4R9y#C@lmNiI7H8i>>i^>LPhS? z5Ot9yjcoD0*Mmvi{=wvt%xApl<28JLTPmJ5(g1hl@_7EJ9eABbEOy`e8lUyMi2t0M zfx$QkOTOL&2a}1oqw@&nwwGb^<0W|D!3z9Qp2xobZNlqr*kIq4XYu8#cyXSej<44y z;28HQ`0C{~Wca|lICiku7pTv{r(ZW=gAL#Duirax`Hv`E)qWpOnCC(6PuoDIj+Q1z zp7((EsygVp91SZIM9+X*J6d)92sGwC1MMdp!Q5RLyR00Ab3|{=kprvoqIGvcx!)U% z>m2aGf#E3Q#CjB**@t8o6_E7pm8AUhb~3Ky9*%i+5`Q@+Lu#xCk=zDVGS2ZSE-QI} zPW>4SUsN+-RzNPE@K=Mh92Dnz;vIQ{ossB8m&TH*`|#>t4S32s(XpEvgw;OjlA=;o zqA~L_P91ds=h%q7K9|Y3dU6^*y*&><{BsU3dF_SyKO#5#kPW`J^9PMWNLOY$SdEc_jT61kSEZrb+oIr1 z`aJx()B=hR(ZSfMmf|6E zzToy~SKPT$g0I4AQk;nu$QNx`9l~5hF6W5wQ#*dSV@aXCi>~^vYKDI|-o!H0NB47<( zKD!s3pSj?=%?4FB-m>>3;^`8`N&NQd*K6*yU2^w50E#VUPG;GgLQvrTGYbNCtTpUM&CnX^cB$V0p~<^gyu z$%O8z0LW5_hm>9Ma0dT}s}y^1-Qo;f*(gnBJyR#6HtUe5R-Z_2CqXDS|H?#&3k0R5BB6ni2PR0_d$~6_~S`jd37eTd3_45Z?qRV zaMqBU*AJfeO0j8o43s~Zjm8H)M7f<~h<4BixV^DPvTooX$%bt!QP4*z*z0l<4YL>z z?th2iuglKCg8i@1n)vr+@hneLc(w8Z6$&uR)!s`_Nu&hKqNr zkkp9PxKMOh7B(56CA&@`o^r7K;boYhBKp-D$HTfOhG?EJ6|GwtfU?U|p{G*^ER!SQ zwVe|-w9v$6PMy&7*9V=QEB4J?Hbc>m2{6*sgxlQmkb5#Xof~%kCih|5L6r7uCUUxX z0KZ%=)caDpcXw8?Q2frD)qZWzn%wo%;BRXMJJpabABak+h7J?;s$_6 z=@|Z-i67Z%H5B`Z3>#&Y(cIDxIuO}9oP;Y6!z;FCpy7AYk={oOP;B@NYEfeDRVoxp zrp!ZnPD@a=aWuMBwE~iU55QCQ9YoXjpXL;H#o%Qf_aLB!L#5rv5ZdJqs;+)`l|nsu z_PWEB7FC$NF_;*Qe~d$%7~JE;Sz`1M;BpV4ER=`2ar5y*Z5cpD4cr{jmHq346jr!$ zjLiG^nn<-DCtteW;>iy}A;fIhMfwJnJJO_&1lAA`huI5UXniy%lV0YXgQLYa*#vN*E? zz5DP6rJc3K*QSNTskfy_<@iT#r(zB__3wG)(YgbrSt_8TZHLg)@fKuDRTWm8W`rl4 zxyD86?nDLul%PNQ4)QiU%e}7aMrcNxY2>{;)Gb#9?AAdpB&JN#R*WTVn*mC0NS4?d zoJA9-)_}dYm|bl@29KEn%FY}w(eO(~qhBeJt7hT2@JbB$%qc)4Cja3MY%W5&uNqK| zBn3`I9>i9Kf%wHj21=Sk(V?NUk!xE&;S>e6V6BPx z$j3RL)}uxYGI!&=yd0h~uK+t;oj{Un1MuoW>3F$^8j&~X6lc&oF}w+Z$*zD}?ebB>Y(F?S z{xrzvU4iCJpCI_wPjr?G2ItN(IB&!nJYqoT3dm%@OV zdbrPhD$cgD!f{amH7Aasmg`Yy(Bv?b{$Lmi)O*Cu7+Z;2*KR_dKNrCan>T2(M;e-B zdkO78xoG?TX~?e92ptr5BY651Mdu`-gK=`8@be%T?n{AL$K3I4nJ}ExcOBoI69Va@ zY(VSkd}s?lC9)XjgKAI|k{pOc_m^Kmdi4hAVeD$yt!X1Vr8h#&dM9v=h=nKXLg2E- zRwU`pMt3KyL{D;j;oI5Oz>DYcLZ4}{$%+TMk9e-^>1>445$+kuoDa$r_zwwP^Q z2x$9oNP69bLYn3v$H>XB`%$HM?~n&I_pjqCwojSD0nbc5oAV7 zAmURP1d17)V;#pJG9?1m$1+6=+6(IC`9vfvgjq zQSgXnbUU&I)j3w9XX5AGwj~l>jTgIILz0kB=05ZydkA^@_k337a;u(g*RojpdY^rK3qEi zhtM+kHc`wgSZP7*OL1=>9F6vly~pj$+=Fa>o=4Z?%8=FQo9O6ePb~NR0(|Kk4Yuh+ z!E48JwYFo#>EabD~b=-}bi9k6jpJPfg4iDyYK#tCO- z@SMY!kWY^Z8kbEe+KjnnI-b}KcYI_ z7f4S}22N^sqE)B0#9;IjZo%Fhw92>@OP_Ybo+3}7?wu+sJp6{s&n)IlSB!<5x^k%6 z8&GqPrv&>Ma2sdNKv!orNEW86aASYo;Cf7R5t-49q9)~|4Mz=O_1+<1Jz+FBiRb)l zQz+N*F@XErZjR$*C1A3D8JL|J1j_$b!$w!pO&ndzWtcAJEIxK{Q8X96cm;?I#li5f zVm*@cc#q~RdyYhcDLNe~!!`Wr;2K~3gzp-2L9g{RDrl9*2Fg!hv96dOv~&~Casd!+ z8VPpcxkx?n1?RU}hRe@NMRNKXNG<*{mn~U}hrUzA55K;JlDSLpp^7ND+y4~x3Q5rK zxejQLCM0Xe;U;j$rh#VI{>%_G!ctuOwzNyWqzu9nUY5YyZxTp&?*uXxN8z5uKzwDT zELpj9AW=Fqob1c;$9s$m!SKs|(768}D2LjEzRNw-`vHkwg58i>E$)97UqzPtwt;PX z6a>z2g)O>LU@$rY{6yZ4;MmXgbZkdgbQY4e-o}KtQy>-I10n7BcdmbC0(W|?F;ZI} zD+zY*HXS(s1fFJhN90S5MMqFHoFWFIKkhKDKah??HRfWu2W6meqZE7=MZmV%W8sRS z4L*fTagJ*|*j&{i`peRBQECyyya`3!18tyQ>j%28;|mj%M?uirwJ;M2SeXI$!mQTjI6A|)1CwroPN4Tf+=Vu%_HnkCEU2Xgs2V>rFY z6WkQv3hw95iO_8P26b#x23M_m?yM#!uJ!LwyEsRV5gE<0>phWw=5^Hc=Q&*ZdIoD= zJ_|Oe0~*{GLZ51brs#U8_N zZ@Af3$$c|d0JVafT4UpXXnfcptW@$G$6ZXok+b*VvVdQ3%X^k&=io$+rW(ShEh1lc$pBar zHG?x7{U7>VorTJ0xWSgd!(yjmCy2Q~h_ntvFV+l!P1)Kow=)HuS2upx8+j^hcgT#~wPtcYSYA_D;lYa*CMcy+-q1SfFB@4PW<+!<*;3 zB2EU9$ zd~70;o%;dlyBvjDlUrb_@f)=aNJL?;qoTD8c!)W=J;IuLk z&C)u8E(TmgcVkz9=W#O}kgx~)3R`gZ&tK3hvQd?PAA`P(4DQH-zbI)mLMdyY>iKrq>tE-0(O^49O6V4-Tr!sOk%nu$n~7$aDFJaZiE?rd#)9y^0%>5 z&n#eUgvds>Wr%B35e?=6I#%v|t>H43<`#0}rCy9F1Af8>-#RKw9e z9~^c)8QYddqNEvBXulIrY$$) zO#2RdL73Gpm^LB+X&mW*Yi~wkpNvAlk3U13^J2JkB_8#joeisI|3-scIW+o+Dfo^{ z$IG~}WbUYExc|l#ye!2Z8#w21Iv-+TF~<0OY#~zEa@aJoW~Rhb!w#I%8X>?T7ItL^ zqlZB%kmV{mBr6}mAEzQ*Fz76{5}7dPS5L%79Xs)q%_7(Qq$xfk0NznPNpzKq{Om6Y zpj7ONbDFQ<=9MuxwBTfZ8rj$DSTr&O>kZ->408{t(@I|_XMhO4Zr=kjllfHf1o zqPl223{BEa#_exWj^|8T@HSgKJc|}H(VMv z8QU#Uz>3OI5cX3PloaWQ8c4G8;^Rk0<=zU zM7@emoJ{#qoYa347KGSv*IO*H%Az??ZXCqbydBNOZ4cyRZjVI`YhJ-?H+L+z%NOSi zPX_0+{b-y~E;>I?z;C-R!-57av_$?Rx|%!-yGO3Y3!TI+VUoxa-`)d{f4ziZUl*bB zgT%CXBB0^DsVH3049Z?CfrhJI$l&g4Zh*QgXghd;bFc+i2+}z0g$_R3w-M{Dh`?L_ z=wT!tCsMXx(}}Y`u0g4)DE$PI>M|ec@7A?$iaW zbEOW+Tx*B_KDNQJs3XXKsxBv;K80JEMsV-vb2w^bApX7o1M+(y?wjU#p>6xtajT@g z;PEmkoOb&NzUNv6a*r+n<~+%$qp2irt_P7-Qo-W_8}a++G58kpgU)-qAZnZxw2ED? zf)Uo--M@>`yDB$$?)y;8OZ`H}C)RRV_iNC>TmOOL5G!zu4@Y@E$=pJ214zHWfUh5^ z#=hcO*fG{vVxUzgSvyrvVq$oQ+tMkEY&-qnTcA94{%MAHMc1KB%~X^UbQRtC)r!WP z8;7*k7bBDYH{9XRsWAMf$ig2p3n$E-jy)p;+~%o7rs{1c(fTFif@LWRG#&+-cxcTehe$l5f(fldC-Y-iGtBy>-6K{kjUD8DRyiReqt2vaJe~`0NK814IrJ*%S1zsOYK=(7x zp?>R~uyE}?aeXR4F*5$JO3odXhSsA3{0Z4O*(2p$!C)+Qt2KuO!Hu_~KSD{&04U$! zw&`6*zJqVTyrZ3HlcS2rKiq)M>HL$Nznae}e6~ef?!AXxhXBkS$isJMoaeNN2}-Do zfT&yFIhS=}&tp*+aIvSkPLbb;t(L>MKesuDMc=rV1=%QT$6iEJTZ$&P8-hZ@BJ3R5h)dQz zfhhlpqQ4~?mnbel!%j_t)Yt73D$ zJfwPK6*yJB<2J>{p+;|A2u+fdAz z^rFD7WSFt13$CiJmbCVUak9qRXq^2)iT$26;H%ybr(e5*LBS_Tsu+MH;;k@qI**UA z``Ft58h34-3X-d}<8GLyK;^6mSUofd&UtPCwabCvAFK?%U2)t@`4%K!d>F8?mu>BWgJ)UAPGb?j~TFa|p&qOu+d@ z`Y_x-8-|}QhxFw-=vZMf7d6KTg`4rnE@A_?U22BO%WiR{&nLpOa#`3fx>1}K%464+ z`B3_NGM>GB4|W=Q2A?nTLZ=NEA;qFfZd!Kb4Vx|iL6#vdm^YKxC#zyvv@ zt$2}>4><;oDrM-*B7aEpy9th&{jl8QpvaT{50=%K!}}`)|5{!LpTu6q(TY%*sJ|6w zP27jF&5xj6x7_gp|5VWbp@^Nbr=#Q3W}`vAuBh?8A+*{>a>)wb5I*uHQazjtgHP`O zlk#2|@oOngT`&l{xIaYdx&{!vp&y2vH-Rs$JZ_E7z=`HY_=V+3kQUv)v5x6*!ec2? zDN^SSe@I0ti=`pg{{bX4euqcq^RdEj3*2@%3Hjc)LER6e(aT^%$Q#`VGS1p~)AMsM z{$e=#??W}}cb$kA_dfvFv`P3xZ!kWnz8OD$_#MWE6+-EipWr%F4^lF%u=I-}JkdT2 zJCyyys-M(|cIP8pR`386Vidr#=^A>RorZpi+_S7|k&hKb;pov&{7dwcZut;^)?A&1 zVw76Y7)w_!IxJPZ%Pc@`tKP!ncvT2}DbEdAmJQRRPl8=aE}S_x7;jz6!=4?}QMsWe z_^%m+E(FGNiUZ=vf)!rjRhDDSp2Typ;pFY1fJcXF+8*5tjo=d1+3jFHEuG_xUo>{O5?1982dIM;nL z4)V?p!yo#KM5pOV2(~MO9$9rleL}G8s9 zPqCOOx-uMU-aSF~duJe>BM2}2D~I0-*3k7|EOHsS2)ShpM$hM)!-f;uD2mB(QPOwt z+yff;vU(uAZ`8-H;y-{_>mdBiuU>TNoq_J#6CpsiA5Ixbqs?kv-0c00u+KLQ*&Mxs z-exFa$NBbHIwKVG-*WK53o6hMc@S;We27B4^SIlu1GzM_*%bVMTRyu%qqQEb-`x#uKfO6RIB8Vsu_AR<&C6ojTSfQ*=N(X*x!#YAEg)Sq>d1UnA-HIe5U-QP{F(Fvy9Sg+q%) zKjMY&aQ9BK*q^=%-5niBGBX8#PT7MCL=VsKfpYLyb^|IH;>G1kB$8{dYfS&loFy5w zwhBFNxr#nbR0370Vn`Uj4u%$|qR`KPr0gZ=saz~)Q5VVS#Es!jSHIz6`i;os_IKo* z>}xV_`%L`3@FQptdEk!O<6qq}MV;lI6i{-=usro z-W6xo)uYgnck%2>DKd6B;PdEF;kudMNh_aGc5+= z8z(W`OZki}i#B1tq!F(OsKL3zo|smeo<-~3c7em+$LM>~PE1M`b>LM0!D#bk z4enjVUx~Zd1Y$CI4Ze1M2|Ac&&zWvXl5DZckVN^0NUn%oB!#sWk~)|RGvaA3mR=kCcQx}qxeFyNOza2;jH-&e_A=o%C0|)vfVUJal@L=vg7@?lS&D8ef zGb# zQKakD1@w$d#2;((aANs#v1eBT%e+*v{l00$=kqZvCC*A4RM&xYr!B0DQG+^OlSCOT zAr@4M6s)bp=?`yWNz+pNSSkoRw&&rAb<@RXF#=mD4kU70opJ7i2Uta%yN~`nh?Cl( zfZRlvZ|b&9l1jH>K#ydi`C0vvr|+!s()w&Hb5$a`t8Q^)lg@MBf(Mz(4-CcbuN$z; z!h5)1VLWU&Q-&s}I)Lp$*&#Le_FR#=IzZ}w-?j?}NlQr$86x$6UgJce%-SvPUf$b1 zjMlH+w$_N}IAp?00KfzxQ_Hy$sg?i#*C9zH{U?=!-7^0pf) zUkVK@ZN=(-&6l=LbRZd(56QW?udqY0FC9K(6f3!pnN)`*?fN*CO^6s+c3|rz)|{6F z|Fkn`XmB*{Xm%y-1~&AW;`mYzi)tZcq%=Kk+=Vxor_s5!ci8TX^USP1oBaKxOlr4g zkoa0Reu`f{864O}IW;@xHPV}^_{xzLTdQgNX-3qwl1jsNHQ*y%EyC#TrM&D>bDFcp zo}KfL;&sl%WpB_pU!e&f3_wt{*#3L*=a+Mgk;j~59Qf}bH=oy zHBiWXJDBdEWBB&=I_lOsi5(nxn%Jde(@BY0q|;oJPn~~)+|pmlyIq^fYZ=brJuVm$ zlaaO5-SHA_AFIvCxU1}DV+Orc+bXzQIAAwNW#ZcG#8Tg_U`f`qSei!&bN-V~=LYKn zGI3{VlWs8o9=WoP+hQqf+Y&m>t(z?TGm>|0J|W~J_X!#K#x$jBJQn<((qv6twykhC z{oS2I)gv{9>gOK9qf-hziY+G239W+8=?Y=-KLZwa?ySfXM`XETEjE3uN6;HjGVfLi zb!qyEixY!|0rppg*cVZ3osAuvPLk+F`J4FMU|D9Q)Jt0b?jplRBnr!C9s$>;qx5Qc zBKs7iUbctkv8w^{H0+)e{$;Zhum5n7oZEL=^fsvTg;H@OFv5epTK-np{@cBD=7|yX zc(4++mmNijtP!1dsbA5Yl95OD3+6B9UiK5xrO?ewbM^ z34(u@zE>{6N*ZmYOSVRd&XXZ@>W4kF%4anFU~>Vlj&rAe3&yedpquo|g9pOySbLD# zi1>lu2QZm=wxAxi#D73p#x~4Wub#*j>aHYOBb~{jUUlKu7j3*jF9-Wf zmcmY+5gZ>ZO*Y8I5Qm&}h^}8kW`9j8eWGkZ>okv=dRBz6C&v%5zmN3k^;=z~y1v=; zP*N8C{WXmBJ-J0wQmbjUVIsX2B2S{nOPI?14z{ZJEBhYyjUBg+pq6`X3%{c3N&>{s zePY1%QX@GRQhHRLZtg27^~p&n^*m`v8{Z7)*P0ko|JPMWM&+=u;w6x)j?QG#V?|u^ z_l)ph)Hpg$wvZT}a1^f8OY!+8743_AVG4Isr z5PzPmn;lMTA3hcqEMG-yezedXE+%yO{RK2U%tYW-ma{>D4t&Uf5p2$*IV5s<3ZYX} z`90G&&~8)lIkQ%yZt^b$D;jrEPdHuT9hf34TScgy-dy_VZxr!+V!&+?b3LJI~@djfdzq^KpXh(En)a=+lBzWRTGF zSOFuIMk+B`5s%x+accZzB^%uOo!ZaZM&JE5rHTb7Q1f#$ zOVhR}vMuw){MT#^-aWI4*pk-LI#E`6&m)F<2JfU5hpL1lr-qXO|7z)vb_F(}cr)iw z^$U;u%`@ZnSbAi@Z{bn4pD;4-lQ4LUC3zO$&icGxvW=G0*{>1v>6bVUGFda0v~7~% z-)yZRR*7Lm=W{sG+qjGFIif45?AgaAuRhCa2Y;o-l{46{=0W^(?-k53b2Br(e1blT z(qfTM3h=)V!NLL8u2c9>lp31y$Iu4i&V>R{O1jx%r-jd2%*`Pwn~{FOzej}H%E2U7>K%ms_tw}9D9 z2gKjjBQorgORtcP5s{DoO4%|A{%xShdVYULl}m5a5t{KhwfY4avOI<~pJ*mcLo=}2 z^Dv<|c`(~E=p^0G`7_VT9O2~YyLfj{6P`Wu0~xw{s&Hd#ApP+#fVE|PW{W;23oVtK zNNZdqc@~#P@0;;#(XV!Deg1|}?(mf!zImMu)A3<@3mmDALNCodfvMh$aizOn4B&5^ zPNvewed%ZKE}Gb$1~%t4splhgW)jv)CrKS9j@EJ{JkyWgC)5yVgO@2$S9$ru|oVW>ke4JT1`4+Gl)$BZM$aTN zWzs#8{$vE-61kRNdUrkFwX~Bw%~0lRixP#(e{u9eYZa}jSR&AWk@T6$A#U4WZTeK$ z%03-FM}u$up_2GQA!dOk4iPzGlUIJG$129Mg-TKE-74iW2c;pxCg&f5TikG#EuKSc zl|IqVkeRuI8wP0hYulE@`B(RCBn;1_*2e?u9um&m@QcOqk z7O-pnZ{cdSFG*AWh`j?}3S<9D(*~JEbf=#My);7t-`)gK`yaYAVw1q2!FBQ9apN{aWlWGGKxQ=KZ@wA+(5&UpOe{%edL*oDW5*&I+5#L%c=*8g-}~F z;%PC94tcPOt(s=YT6JCE=8;ewJM{_O>=RDCP$a3n)Ii4nDaSp=w}p;7s=S8XddiXy zktc25bmqGo#JVwp=lwP5k%~NgBqNb#7x?hSQ`YmJ{+RQxH7$7e;R1|k3Sem``{}|R zYpA1rdTEZ71>bu~n=~41rPWr}^!)BfX4zDM^H*!~zqumJ_=zmdngs74nS~gUC1_s@xqm)xvY-%r?Chx{p ztFTLONqZhHe((=zQl`DFPTo}TMAX1zgn{MggM zq~vj^Fb0*=Tf$2r=2M4Y)$SqGjke<%$)`2)1KFVAI!sf$miC>|BJs8*g5uAu{2h-b zo;L7&Lb?V_-LFag)Ex2s&6BBl-#&Vw&4+4*htS)r)L89nZ=v^hF6p{hhu5avq7H{+ z=mWhFWH6&rSg_(Kf8V`{fAc7YHayLsuFKU~!xDKm_I4sG+n2}iSYtLNeg>JnAcx=4 zs?BfIyek~tltFt{n(i!wsIn*9i}}i$1(DKG<8@qmh@Z1(N!f@ z)V?;GxI9o|zb_oc-b%&vdg&{QEtfI5Z>!i&hDcQ8enT96cptF1&N*fyTDeT@zGyjEB_li5z z{2^UmJxTagqeF8fC&+^(Eo9ohL}AYgMSiRL zcicXthPJuplcM+nVz*)>IXl9gA(Lv-HN=*0x9q}jgBQkpRU#9Mjfly);WT(^2u1H0 z^`6#<6Z5yz=Z{sH>fqnhmRIEiZxI^dG?Vx@ z4_d=)u%So2(0Iv~{;LTWuEwpP5zZrN?#gmO@eq*e*?Z|u2afkDI#)8&t(^oUycKe- zTJYPrxrCeUPxJF!@w}u`s`Jr^b@Ec|*3f(O$EQFXB6}JC2@W8CqGajJ@ehQ5Ibr0r zc_R6A1(TX%ROr2Zik5Ahz(yER>UA-XN)6aA)Lq%YZ~YR^TQ?Mv{-d+V@Xv$!)X({3 z)Q&-9wu&D)QI|p@YVYE6pKrjUg`x{b_cLE5no)D!W#Fx{%50JUajLTDhL9IuOl&HT z^87#ve{y*;rUsqD5d%Bk(^hG^PyV6^-b?ruB`oer6 z?{*shbom$lt*S_#_neB{tLBquwF64Gx|<5lc6ua!=0#ZZ>?f((w}P*JIFPq;8pymo zE)&yz^)zn#J?dKI#n&CT60B2?v%|}ZsoxJ3I#Jn>$W=@w%U-z&e|sHB(5pZ?Jfn&3 z=yhV{4Lqrzp~bStjb+1HT-hDF6KwSJ2UNPEkodJ$fZxA-tmW_&N9+=W<}`nLIn_tV z-k?cuIX{`c(1WW?7Zwtx2&zBc{{|7!gOB5x|gYYqA%JX(#J z%gkdGFB?fGDwxxon|g)HiCbCJv>|Neri08QAc^`+-@?AXx_ggmW17{o5FdJ2 zC**WF@GLH$_ie2a|3}Su`%~fM#=zs$C^m#xjy*w!%}f)lzl7376O7pHI|}SX`9wBo z(r+3YJ%x_g?`T>KPGMz)?| zPeOB;``?YUN%JJ_URuq3D?F&(H;J%z=@nt<^ueraP7@is?IKO@qWD_j9zk{4dfI={ zj20M)-Byn>;uG?YO1{0uuX0B610t@|{K9#BZKN9ODzBk~&(ylX`wk@HulU%qJY=Vq$$KeQrA@$3`)Sm6Rc=1aGbd?`p+m~jM8>J72`pxTUfEQi(WEaYi0)olL){nO zpwm_JSoUTsDs6UGNUB#SuNPM0*`M;+(!FPeyi)?c`@0H%`+gk{yZwMPj2}YQig#s2 zhjVGcH(C5g!I8|W1|ejQ0=qTSk*OEHWdmNXrOW3!5ELu@5p6AYy zvCpPLPYiVuwZBzKE3s#zeuXgMl_hi7RZQK~mDojv5`5rEFpasd%l0yJy7+Gl z+cP_i**Y)c$NN}|d$wTypPf1R-uQ^stjYdc9pSj30GL%T2r>?u?ZIb zo}&&U@<>&B60^*%W2?0Ene5MSmKdPJdtV<<`ufH1JxrDEs_CaQzn9TZXtmxU0vUSy>*J}DJGWGC8@*+)_ zeLC7q|7A7d`%h(vhr=)ObS>r+Ob79&25Ip}rdttB^^;-`;TSJAKn4_{-&2!bhVMG`sHzTM(Vja+YMU11>|^ zvngt9!G%B2Y+HmOGbff6ce}HOJ#X;CSwk4?d45F3rL_}%X=2%Dd}Hc&>iEu(Kc6$OEcE3T#;?9qa%W;RuYgaJ_5Ei2 zx-Dynak(`A^u}1yc05!_lnNB)ITc{-cM=MNYPeS(WALJgXw>;e+^1_TW9eK7lZjL+ zd%HJ`f1fmtHJ|CGcZ+V&H>W?)VI{$A3RIF7vqii@mL=PFD4i8v>SE`uZ!u4&ttBsp zj39F#9U$^vcWAC*Sm`TYE5;4>BN7i;Vw^pa*K2Ph^L8y`aN|Cex052e7n_LH*WfbS zzQJXu&PB414_(TVmK?$U!!PhQT2ds%2=V1@UBa?x8TxOahLAkXhu*)a&F{$`#^iQw zrVsZHWCp`C$#g3p5?z^1GLx4wm+0;IOa6PNX;)0%`~Ss1wrSFu)Ly*ZNS*o5d_xVK zQ}F1lX6TXXrW)V=VPboW_G~m^V*=OHp+{E0*lE+)>SO8DiAk95Wn=p2^=dMuD~)DI zeA)G#%~Z>OaM^>iW@UKH80s}QgDNHd6F%~O>?rLPhBf`7GGm4b^=G?5xp#n|@maj* zSkq4nsx8?i!>{iHSqN<_ZbnazE`f!UGyL;>u zX^XiFC4Fa@N3~*Eebl(JM=$G1{PYI={G1G%I&wbSy*rnAVHtMyS09$n^S~=)Hj{md zX>^*`bh`TYe0C#a0{yzzg}wNq&0=1Z2p-VKs7YV9k4`6vm2|GK%p84(ALvA`5LyO^DVz)Vyc8@;BcsYG0|0kYo zEX*ULT>s&U5v$qbXgQ{)aF`C*tVU*CJBAy}=JTO>XQ+C>7(Qm8Hvd}YHgWbXBKxvU zN$u%_v@xiO?rha&&;8m-L$NadYK=7WYc?XE%5wz8x&gHLPNbk7kW0!|d-1{IdExwq zaHetDfI~g{!84v z`Ux%97|Gv=j$)oa9^x;PquJVF(rnPWXklaaA#CrU%hdGqm@Oa8JT}Ob>EFu7QV;&& zyhm?{U(XP>esl;+IP6A4!p)hHR2EIiFr;W=G4uPZ$nM3~!lxpCz8n--OZZ~uJ^7)q z>1P$SuvcZ>XT>{PSG6()-!ZIH!?|hV^&ivH+QO*SJ(johE_^foMGbHEkl4LS z?60f~`vxnR{ujjf`!i@y`W%vK;7b>_XD~7Q$P`wNBC#EVNoP(EmA~jt_nvq`OKR1a zb(pq7&_dXl>MXqHB^M z#55OUd;78UvQ9fq2ykF;!i~xB*atLk(^Xn9!JB*(zT%<7b_%LFEAi|UW8&ZBA{6E7 zU~*^>JL{av%^AK?_}iUDhC3Ef{>KlZGC7dsekow(rGwbtkLEBP zPvB2ps3cMoHc~l zGe~WD9ZpUXS8vpZue_lMF&$-8Sf9E02+&2j$ zLKM-*^e=2ZSS+dd+rTas`;zmMNcMC4co<*h4VAsu(KXNCB)?$@e8ABUQ7qwBrF zPd^4$olB#GE`4Zx$#m|p`a3z@RL4{9k4W#{Z8|3#z|@@c*)i`ia8^qI?PEJxWO*KJ zbqW)vzS&5lk6xgCWg0x>-z@&}^(a2aGnk*tdqAGrvv_k!5>MX}&*qJ+A%pyAatr7Y z7Bu$7LyDj1%rQOIe(oxz9zZ@>zArpC$%osCg%F{V1;aEX@bK$9vWXu{ac@5eUS~aE z>jZ$!2XvvTE>rj_pM>Xw%Q0-nX`CNui|d9up=thJe3y`b#WEeZy2TmS#@~QsQL7a&8a2P~tfK_$Qpb%?>6I2JyZ8m?4*jYm>*3pRAE!dGaJY$u`5<;xy5Vw{i?y z!Ug{$7ohT`Dt5o@#{9DB;^lY@yW0nd(`yfi8@hbOuSX6Q7mQ0rzwjt*%J+a@MQOM{ zAx+dRD#^3XUe4ySzm2c4I;;yf&XQQiz)>Q-Uq33JK%!g5gG+X0rx zJ+aVZDBd(X4VmZegL`KoM4woJUgz&(=-}PB7LMSF##L0cM~{2zAEEW14?^9MYT=E4 z7>uz!3ibUI@y3=(QclSlrnC%!jpj9QG`LtWKJPA!9B~|U=0)Sa#a(3bR2w2j2EhIO zW5D!QBy`rEg^PFeVC^3pp=QxkkX_RT`IClA<&Fsu{G$Zbm&d`+-W$NjET5Sjbb{vD z^Wm&tCM;Dp0pF#SuunBf7;*BCv>S+E^rkDIDD73BZG4j1t`n2I(kqDGsEbWQV=!g8 z7oMtbg)={{vmK+AVRzvtz*lcT(cv(Jd(CH`mWU+@TVvtS!8cIQ-XAT~Qfb<}*TU;x zV)IN7lx(~`2iE=D&u;WmWsAS9q#eG|)KEEq&1)Nh^DRt~Up_9> zcxd3N|MaBu?G9l~$WHhplEaiWvf|P1N@AsTS!g+IG!83E#*Z5lQE8qS^;IrF+R1!q zo0UWd28CmZ4K_2 zaP203BvWkcHC-6DO%a#RUPFhktVY%ROXzlO1qLRa1lPYE!sChB_)@kIex)78*hf>b zqT>P!RG&dM3$iIznh$TCrH*iA6)04LQAu4Q2qTl?OhqJd_7()c;|=*wr>&NF~5TkTz695 zz!vuR;#Vq+3FD2n*XW-`8ih^YjaAoe#L-L4#S?tY#UBmtVn(P8rcbO9vOGkPp*~Ky zU$d4ReUOp{ykwq#bwHwTDac>`!_rsk(=UrknCIdnJQ1m5bXqK0#|}W9I|kV3JQI4C zm%(vmU5!&sOTfP7lCa%J7VjJnhLvA4Y0%&| zaPDa!z8QW4#;P92_l*YPUBA7>+NY<8uM|B+i;4&Esc))e$%>H}-r`Js4YQ$N&S^p8 zdxP+(L?ZP z&GNT~V!C?^?oSa|VBlrpu}df%JG2a%)S86A!5VPOpk6SVKAC*xi+Py;Um<62Has*n zk!lIo!J>7s5b1CovNt`1Po|#WVk^h*1=fU>?YT0vies`PX z?5;dkt$2n%2om@MdtEXKF~a8ia^S7^Mf6Gg7A}GcNATs!1Txzq5e^wP3EopOnBSjZK7F(coGE6caAz&e(jJL- zou>$kyate`!yU===L30WOfz3}HioK}74XF0#xy0h7NYn1bN#t%_>;lA;o8)ZSfVTe zk+LZt66nrd2g%}_7eg>_^dXqlrb%n3HS%u}C45QAWxn}o7k}~1A4adAiMRW0;WNM0 zbA>KDKJ4@&zJEt8{P8g5Hn-Ebp1CIl1)t$p!XFA*=B2b##gZ1>mxcAqhjHg!8+bnp z1t|Zj$=40YlSH}b(%Ha!!sOasoRJ3HKORH#Iu6056+!HTT9P1Jxrpj|8Nii-skB1V zmibKA;hVMpP|HV79ZzET^oz3P3OOPC&rAQZKAW;g<-}U>73T7)tqoyLFS2Mr?h3lr5e}UX zOnL1?J3dz4mRALIQFHAG+WucP_1P9F?35dZEk{xy^G`nOdu1<^9~(?5k51C>p$EvO zcY$zC-;g4>88yf{!GfVy`1bq^>aAtO2bLGpqg5%=EF+RQ+Q)hX+{TzrT=}?#$%wAK{vbiu~r|#n75e zeCV+fUZrioPFVk@<71Yyo~C}neEtnuhiHJ;ByBJmtwWl@d9>qLwP@DlP+S(gjtMJN z1)bcrFx_o9tr|3*A2uDx!{=p?_JRq5`dwfCAmKH**!`4Z_7>n1mH`iy3{Y+UR(LQl zp3bHylAf0aZ~9xrj?69KjlJdg_l~_hY54#|x0j@REF7-C+QY`!rwZ+XMlhr7Fhy20 zG2i=cbSqtrF0Biq%#=n+=c<{c3Rc1*GeMB;N+Rx+40Q#4`IV2l{K2Cpc-qHYD10`T zsrjnGs9hI?h^4#BI=aumnqM`Lt>wfYE?vswf5`CAGh4a3R1?>ueN-}MY6Vk#77TCw zUywv?Kh*^xD>FRGybdH&eg8(XU8Km{8;aPF>vq&K>mC!0IVilWal-VQt%Apbc>16< z4j1*#r6n8t3u*gG1)Y5=II1TU>ZnFAFL7cie_spD+Y_1Fl4Edp#8EimE)V=^CCt%T zAe0%ppcLE@v}d?;Z5GaTjn7N5o!c-=TM^IbG=jQy6mu`OWceOpxI{zx?znIrnwIy2 z{L)TQ+xFo&BQu`dbH+e^R=Y3(lbG!nOH!Y#$io!7=ylp7VWZ+cDptx6c84s3*3+i! zzdyCY$|pca~f^)OCr^x3oy+tSfaDm7w>JJf`i`};F0K~tW{|Y zQ+m9Ce}Cc6cifM~iDR~*_RtI*ZE}Tz{Ps!8Dii5~iVDx}$R_7ChWvv2LEaLui2knW zS3atJFYfr3$+k}i{NW!%H4h9a$07r!bO5hzj03vvAqw8TMjGgHUozY$nP-&Wqsjj+ zz-t^QytVtp>};wTO8b_yaVoXVZWKm*T`Q=cQ0DD(eChU_8^URydHA(-92OqqpSt|0+UgP)}4W;tQ{`+}G<{3Wds3lI0u@RNF zd%!~5gOs~mx)-DwIMJ5-__kf)+*CuM-0qfChrG~Ebh9J|V|~JD!`Ob1>&iDY^O1IrTEVbGs%K$d6W*jMG)vU_ zW7UJ9!(S5vo8Cj<0O>d1{Z6C2cJa6g*MyfleBk>ICBCyFgW}ji+BIf4_cK_<-^cj# z(TisDmYP58kLz}_^_q?k+Kfdr4;xeKUwc?Qye~z>G|_<=eLDZk0bymaeE-YN5gp5095Tea2*4f{GM3_8TTw1YuQQvpL%uS468 zqu|e$klu(6c5ls7nmo1`I<3d^WOaG&alVsl?>G;~_#qzl=&XZuL=+94TFt&*sG>P} zQ-!?WC!llZN9Nj;Pu^`&)OX)%b}b~1E?pT*U+>BCgD;wxp6)Olc08ThOWWzm^K8-x z*X2do+CtCA2^93mmw(b8!^8UoLa^Z-(w{zu?Vt2Y&NbZOYh$xxAi#tcFj^^X=Bc&MDL~fr@hGJ zz(l6lq$c(oa{~F%m00|u585lt=l`NdbL+>sKlT;&CbtTOBRA8X#|f18J%FCPdrbl7 z!+F{|AF!CS77yS42=W`oV(bxhT7A)xe=N2^_j``))64=k@a1PJ_U*@qdJf}X(>`F# z-y>z}9+q5Qxi?qvNu*wJTSRfJn(mENz`f4Vxcv{|V&g*O*9T(S7)Ok%Y$n%RzFalt z9_@8Y;d-MJSziw^^*RLnrQTjzbk`h2b#|n%wu6FS&*DR74CJk&P5AX2F??tL>#(=) zY52bM0vw*Wg%7Ovf+1H`D5E2lzcjcm6Rtv#cz9` z4@VXRbZFL{MH?9W2534KFD{_B!VxOGXY&5_0IelLXY6dl zAJLZOpIDRh4Ev=Mi_7f}V&Na@olF|IwaQl5;{6}joIHaTwY;NlrwE#%SIds^V)T?# z6Ng=}z&k4vFvnVzw|Nfa4WCv*<>{f=v-_^7bBznl3U{SBLD>>~I8XFZ!H8@}y`~|e zeQdBx1Zz36j4oXs!Y?{sqwQP9^JhngNYC1VXqEY&;QZB(d0ie(MlRdg@eAWgN!$Rr z(u_zNdx6Fd6*T2?Bo-kdLgsj&E!BN_UO^fhd^n#zN5!!3PZL?cS$4Qb zPvC={PSWdj)ii%ne=d7Ri?=Syf`I1TVB3%Y@82YWTk}+QxF-*;PSls|35+F~vFqUC zeHl8<4#D*p2k`v<3VDf|c#r-~sMR;cu(~9E-uD(iRHeaV%Wsow!&&OSv_+baG8h9) zRbg>uCd?2L@!k5t;wh&;;cQnA96RN#;Qw99&kCXJfseniW7}}DXlJy)(wtNuo0Io2 zDZdtCO^+``3kEayiJtb3qy?J>!^$;IxF+Np@rX&ZN8N!9==Pwv?GwP>^fJ2ZO$DoX zf4u2k39I_+!NQ1_OmjmrmKTl(m&Qm@>hB$>h{vhy%ud{v)JFZa)9CcLLFD%S7(E-k zS9oZlg3cHB3l%t##5$8&1+-O+$GHMkeRBNIU9v4w>=_t$7*5U z_LDRzT8?CDX4BT?niQ5fi{Of;L?d+x?bYi?Cu9xzjwC&JFy|s2%Nj(R{(7@jGN&m! z(37^$dIkGNKNd}K@T72!e{?_X5dLi%D8AyWEdJ588+{)~zEK6Y&T7&7oS8IuVJB;BEuqpb3uqR7gTduHg?|m> zAneXPc1WZItw%SLmQ>eQbmap5S@KEL?9@pU`*hHOONRVP&r6!PP=m(113fCsrP(>> zM7_K0sYOKxC&{E@$*jTHY<`{;T88qhG-b-W-a?LUkuXa5LS2y$=~Qr&%o|@LY!&HX>ZC;^Inj^LT2@3p zmP-)pwuOEkC7f0g*g{9o) zm0jtDxpSeF4-<;IK2xIMFhSY27Tk+fgvIGv)VXUgFV5Lb1Nt10Y_8gha|Y~(ZU4vAyoU;g1!WF(mITWx&G6)q$50@#hpTb(`TZehXqyzHdgOf_ z)gRR2l4>QfXX$rLvlxmCZRYUw(;iSOuE60N-NnUnPjN!rICQxDkGcy@U`!YX_v6o@ z^lDLA^)4-_hO0vO@NRe%pTPTk4#p*8l0j)kjc}@$1>8LsjOMmk68%AO_%6~wtmGAGD)^hSx-G7q|G=mqB>1-3g5!@#aTLfv@-u$|(Gr9VWZ7aR$S zQeG-JV*n*OIFgF#IIf?SONKfzBpR&1m!4{1Ic2W&<%u5dTlE>PX6VqeF;%dsT!}ZT zY(~XvF?eH93x+8@q{HCpsG2zqyT^Z%7>&?|wl^19j7JPHUlS^ms@s2rO6aBBS+aU$ zDpCzm!Y0oKP|fjx<8t}rb+?`V*eryB?Wbr(ryg8*E1jcOY~u#O{V;HJ25Vj3hpK-M z!QQ8DQ)cTxG#iqF!<93kuIdgtH8;Z@nO0I=_6>?Ef5QgNe#mR%aLJ@_T&+^WwDRkj z%}PHg9e9p)MVtcHAtAKtu?`)T-h&E*$8v*~4u(x}{ND^kzQwzN`ZRLMi;KzhEz=sR zB(m7{Qo84}lF3q%$P80=Ld}%VaOk5F9jMV4q7KSSx@?+Z>R}D>7`Imt67v;&0>g0L zt?{s4&77Yz)#ZKGd=myN8;*4yp7=tnD5x0shRGQr^qNki;cDmrwh%=Y5yh`^zB`nq|04ZUQzw3g={%$R2;Sz`Y4a;aO8T z*(@^SL4PHDy2n>)*y}H`8B_uLHiXlwU*~w^rpf$tfFai{*X66(Xue~t3FRES#2;wf zqP@p!h1^~@;OU&pY|4_e6nRIM+G-C8tJjag_|>N=CDC4@aDEP2y*Hz2>38XQ(F5|+ zT19?Vky1=00Wy(XuU^`vvFl?=LP zf=%~v{_4R%9_n?5zOOsU@0|+e@?KN$^fMzWFZ;*xCM3hB+x|GB#1Z?xy~>yFvEiwM zbouN}b#!Dfa5mnSXPpSfx+mwM-Y=ay74;{Jryqs3-km~HUK3V!UqlOyPN_ewo8nZH z;Ded9c*Of;=$$fv`bNbG)3akqDx0M=oGIMWRmGuJjNMHVah+3+-14dodM6xV!^{h5 zKtEe5x-yIM+D35qL#x0eGajXGaMAvW6R2kMGUl;YhcBfZns$hhgLMTBj}=paM1|iF zx_HKoEdJUm+_5wm@xaJutSeoe_mgU_91hq}Ql2chcsOEKwlf(lbmqqs|Inl4DB*5q5r6=C@itNIon+%c`P(v&lot*+a)ZeB13p2v7I~fsOm&Nz^i_ z-e&@7+fU^h{H8q#9e=lwYGg1w-+n|g zsi6)k??|5!U03NwTOrIH`<&zt6_b1Kd|{olGb_1#Mo@iIF5G(amX8dH=XaEb;hStt z?tJZV&z*JBzF}#9SKvNTcUG z$9eBO=*#aZ{HMY|uBDZVm*=j8`L1~^{WYVIb8WOHsgc@}SFlfC)yq|ck@Nnc(6@Gy z&{$*#C-zqfvV&*he-tBF?hFwEM-&K&K6f08iWbnmg?IQvf8;?rsZ!sd9aMNYv8|&w zfZNGy{MgT9e9f3`^!NE>Y<*}TIF@{%Rd*Hm{jg;2d@r0&Z8hVi;tc4rDq)MKpQ2bE zOu>2MnEKkof~|WNu6rmJZ2h=FXXP(S9B{8ztbHWme=|vH;c`KEyDfo-7{D zBt*LS(PEz70bB5sMLTg-n^&5LIbOxqff50YX{bsA2 z|I&RqM`5|aal>9 zVu?xo5y!C3HWsueoo+3gN`EGcNW*C&<;EJ*7qe4B!4Y@Tj?bnu#fwNS$QnW`wOLd^ zi=ZZ&B2a_#iIy0V!SI75m+wYqI{7qvNhVo+a3P01mn3MC05i%H=~+z% zFPin6gn%^s7M)GOzuuGmwpda&E+m6>74Xi$gFfrNrAvpMctxl?r(*}nepx(4z0MM5 zTruSt6CY8+#Cp29Eg7c!c2h?0`{Xw~oF*!$!YaiFEV`r#_MP5~KYhd^{q`I*&nW@l ze^be6&pN(xNgwW3ITz|mmQKk~*=wA?nJDG zN-E%+TSL=!|DxX``g4iz6kPuygsy04(dX_AnxU)BKi$jVaeMRliHIpM+_ng-zw5C5 zIa2^v7xJ^-*5veQ**hrj%Ioo+?OGqYWWl89;-e17jdtr?}pJIlktW%zGO4Vlh| zr&Y0ynR0yN=?0n?dj>WydoFZNiUP~P17IJ$i7(Y2M}ytcx!lUnf~$UiSa(#NCx7XO z{wYss)sakgsn2878y$r%2GTq0ga|6`9u$3#8V@;(F9_4uKNhyFpDBHR7Q*sDeL!!e z9T>FF<|_7rOAg!VcF2*$8Ic%ki_C3^)ZH z5cTT&gw`&bLvJEvsZUl0lef}=3k%d(;HLRFWyVX<#dQjlkeEx6)ACuIU!x?t=8bT= zdq1XHXXA;HH`oA*X0w(`-$EA$kmAjU>_)huP;oL`V$?Mj|HS-)iUk+>mVHrVeQSWw zV82)L1Xl3Yx3PSV-fQ~3_F`EtX3bTkZ|3qx$KXx#0Q~XwIwi}UpfuaQFf{iO45%2% zu7$4`v~I++dkgmvRlJ}XEkA@cE)K9TF}F;7ag=cOw3=x2;3vY_%Du1!AJMF7Tk<$@t10` zeO>~pYL&vVM;j?<%1!oj`dMLt*qNSO42M>W1|e(l0GRu|5*pY%+^br{l(&zd;j{aL z$`BECyo-SP#nW)?Qw_dGO%^7r&c@02opGU#2yN%}5=->TS@ggVj2vNywK@*i;gf{T zciv#)*?P3Oco3I1zX0~h2jABmhixwz>YrMHHoG=s!rD5to_h~%HoV7Oj}o!>$7}fI z=}9SXZi_tA2ER2_gMN$&YCBtq-98T$SI1Oi|DH);QdLW1>R0mDmy*G&KubJSM@jt8 zQdj)ZW*_xl9Zt65$#`YACdS)1p=0AM@R^iC4Z8yni#`a3+pQeMW~U(S&`)7DH3`Rl zM~I&7PN9FBr5UeBGwI=0H(LMHA9gO(qBrlINiRW<4~q21;1?Rgq1SFagx{z6L7VBD zL_xgwp0W5yh$m`lC_`hM2Cq<)2r7}&+1wSFHh;$A3Nk^=f|cp(&RJI7>{*Fme-FqrCThad8daN^0SRQ0l$7VNWx z5uVasq0$SGl6#ja_fMsjn~iB(TqLO-(}CNQ7m4yt<_gy=j)6|Bf%x0uOZe8em-x|w zi8%K02HcpCESMa8%-(!Ug00_nh>YK<;pS3HFgO0h6s$$O&G?&S?2b7&TjM&Oxu_vt zUC>I`-#kF`y)9_gR)X@bzrZK_0T{76!nuL(=-R6Tl%&x@ec$$m1#5;e)zQQ0u8cPY zv@7A9iU=ed8Sx#h{SwE)R_vT`k^Z$eOS%sqhuc$RF~(9I6)z}oV>dbaYO$IMMs~5D zW+k+FIm{tspEnE?J%&?H|5ABoALw7JOg#^_u)aDDw4-S~2H9P~J1pBC_V z6+yYB!|C~N1HtBcUwEBv#n$v=aDIdumd2=yK37Zc%;-aqRTLv=dA);}gVC%aM}=>_ zZH`;V=HcAZ1^Df@)UL2_9=6(Y*uU&Q8j!M-{a*A-65aO)D@mPA{e3LR(CoXQF=R3v zxVDho=X}8Gy$3Mbssx)X8lmP$4fxsjqyN$$Q=obSl*!M-;J#usl=OhE;sY8oXM+^) zlu&HIVsQ3+Cs{M%wdlma!?4BiES=d_$b9mRU{CdFsrE)6bf;Cr!O<__)Bp=??so#( zdiQ~0t5RW8+h95!wo#a!a1d8N^Ag{8(GnMzZ^Ls=F~SJbH!ymcn(!~xM!37nT*%Yj zBN={a35~Hr8l0jC|2nRp+ukl=_9au{yT}gJ_jtn!e@pDQe2}n9Es?2@t7Q-E>ja&+ zb?{_F2~_M}AT-1cfV;~7gk$L|q0TaoT8_!mk+*vM$RZg&rO#ltN2+Uj8@d2I=9*)U z(_nGo4n?tXkyN|6r-eL%H2FBM7>doWbL*+*G3Oc-I*wa*s@hd_xaB6>eB-I`KNgZsTQJ(~td{d=f#%$5&n;XF2 z$O2dV7*4)fi*Tw=3C3KF#nl(SKvMoj_@la?y?pW$ZkWg5b=&8V)O=fbn0r)cADhT* z0(MZMmCC3pMIRxw@V# zo+6L<`ilK{vYIv(oMSU=s+c%;CZ0;rN1@aXdyZa(Sov^RxXu6{UmitMdLl_|_5@;+ zT+!*cuJ}voc<~E7pvGeEJWr8)#E< zzAX1UBh4uC`G=J^d$4}}MC=~hEqZ~veC&a}RI#~INV*>d@iM{ex$;5SQLr8&ULAs# zySFJ-{wa3r8snoucIfY>jXG`j;n3{wkm>A%`2~IO&E^PN^d=IQPIhOHIwN8Af1dET z-({FJEQGw$PGPg=5OL(^W=wb$feLPgaA>Hjpc-FD@{jd-+~ZNayJRHlemes)|LMS0 z?IzaVdX}x6a2B-A?t=Z=(mbt+X~0*7gZDTC{O~TBDIfeSTycwqlDXsg?!QO*E5&Zs z-tREYyE=`lsV9KE%T*!&PaxHshQs?6{|W8Gkw#xRA~GrKPo--k;OJCqaYL<3XrVrw0D=^^${)-DRVb6j|Y+WQQX|Gr%j+0UBqWfd~4N5JX2QD)uCt zoTCqyUm~)hO z{NG^8(~aMNNfvZHv%Ih%R|NsQ+H>wH!@+Of|)Dn z&2JeJSG<9M?vWBxehiZC+0oP01L?4W1@9+a8;gAp;lHF5I1*4tzm7Yg%#a%}>%0Xo zKpi-y`cbIUNTzvFPLxta{=lu7ePnQTIQu5 zDMZ%pA{qJn%;`rfYfV=qdN*AXz4jFC$}fhZ^v$ePP6@q}G{HLVD?Cuq#5S=hnOBIJ zadI){DW+o3TPHqkt14tBt77)&@hIHcfU6q~v12dbiam&S@fRf8$qU#~O)cNh!D>?32j%;06} zyd-iSx#(l*KXD*BY&2h)!nyWoMJinJlSw940g_}ZoN$;|YHfNz@ z$yQit)+*U^c|U~|&qnV-t}G(oniPEdLeMY=a4~fi)eoD6Cn_IP)h!2JoGk~cE9F?i z+8}r$+6RmFB)~dXc?j2eNO7CKg8w6_Uv+DLaCFi`@%TNsF4F>X6+_sU$gTM6l`@Q< z5$>>U=`8a6)K{9tV(IuP{2lXqIe=ZsF=QvFH451yd_;q+bl66}2+@M!S};g{4RgES z3uj;U5vp3(i+%I;ds?WxL7TwOs7DG&RjId z>Rvvg)lqrOBzU4@b?zErhutJ7HI8C!OXSHQR6+V%J%?O7iEzLExS%ep%~EcKIT~~* z!Q7f5EZ5YP?Yj3D6qehI%1bIGgD6#Gq$ToojK{W0md5hC#r>l=7iUTBa+A|{-Chyb{yEPtCx(rvY5Ol4x>8l z8p%k#X)JePx$E@ z){BtpM_Qms+Q0ZPY$spP-=YiqY!b~XUIwKmxsV_4&5|UW*z1dvG34TWwx!dP7Noyr>x;Bu zQ6UeJX`Qbot&na$Un0LYpK8FUg}Qvb7(I(JMu1Dm7^*=yjsi7 zSxNIc+{*R}+ zCzX`-_X~w@jdIxRD+fI}FW7-wz1jCUQXl{GA?SFZuc&XAF68~4MLT3a!98}8t~ULJ zQ7a_$?yLdHt<7e`l~(h_*)61cHiJg~kmBminUaT(PL}zp-hkq9vQ)Lih&|n?gj*9E zM9$+91f_m`CC}O)NIO^U1($4RSn~6iU{N!JzBw#l3&vEi4QfMaYLN_||KAFcrf3&w zm2%H9j`~?-@(&SA)HL1(qr3d@g<78Q zb?i|V^YsRp#Z9HG;9%hecfo&-pJ{7ND(RQh($Q!GHg@wSuv>5(e1Be%{J!@{GNdvH zavD$D1uWYw_)D{xWThI6@3K+6ZCyRxT`-zVZpwjKjtsVJGREv1TcI&J4JQ8NOk`-! zqnb;p#I}s_%3)0A_dPgUsDevBXOgGlSbCiAD81AEFtvnsrgNYO)*ICcH?O{hvie1I zSN;}FH8iE98-r-r3v~*f7Y%(+FiKqU2bRc&)31xGnQXp3a?d?wch?LBmS~63tKXBg z=U{&O$y|)~-9jNnAL#Pt<52b6RH}Jb=AEnNfkokAA*5-X@IJyCHqWtu2hUX`4!2*6 zK7B9*@9SxjUryPq+`WY@Z?og)XURx?DKFve-EJ72HVAo7s8IHx5(a0az+a{6~E z1X;hME?ywSoZe1Jo3GKS@ni78wsX|KdN@Yh4Tq|?2UwwlL)n9GlcD#8HPE~!3LdZ*fCfawF`A2*dC0b1o3hW1qrnlY@!Y?UebL2@}GkPZUdgp*E z(@gl;>;NwFXgx1^w~SwM?8}c&=*1%obm`vwLW&bTpevdi*r0{m9LwB;*wvI-Y=8eb zbhJ*3`TVCKgvs0yuCz>I0ZTM#c%}kqIcK3)e_gaH8_M1N&Qtn84=5_m7v7fE328|? z*c^Ws>ffvZeF}>OhdFCu+p`gHw<8~pycYAOz%2T1G9I$lu7cMAcfm0Fqal_G+^OHb33M@%p`&7(EC>ip|b0~#X@ zWGHu0CgIdMVX_pv>>k8n=V}#Bt6$UYu0&?qP%aD}lnUcEC&0Dj5HR>wF6`dr3WJ=Z zaP@|RIBc~Vrprv=_kaAPtlYOW<@!iz=f+|*SnyBi^K>NH342JD!S#CQ`$lIAPuUTlc}qUVF;#zy>OWr)X`E8%yoA?UjJv#Uz)g%hh3;756du;ZbJ zFfAfN6gTV}RjwPvFV_C0`tak}X?+H5ZJ)xM9lg-m)Eo`iS^Jx{NrM#>ABn`cvaPB)A)SSYBEJPS9cXT$RS3RL)_j;=ZUgDZ0R zP&M$l(0i@|>#07+)Yu#7XIBQxH$H<7*~>6eem|-GcL^Ih3-ROW30R(W7cQUJgVUON zi960l;pL>+_+)xlSyYxP-A7CzI zqeOQHxJu8pf@U|&4pqj>ttt3Un#g{*+8s@gE+!R-l4u7jNG5L20PW%(q$tfenxj*| zrl`ciu%Z#LOk*e(e9VJ_wKoLod@acQrcc|QrP}MRtx%b=9K-dWL0ZRC@Ra5w^)iv> z5o%A6tpBuzUfIN;nK%*;j+l)m>#ktE>sP!z{Rt-cXkx~V1;Q*LneW~x)mSNsxozYg z>h8aR{&io2(Md{pbz~tNJ-Pxbb)s-ap@uN<hq2cZhD30GIPfNtDM z_<8BQU>aiM*t#QyXOEb~ZC(0&4o4L$6QO?ArXdG>x(-4jQ3LMjZ-r+k}R0MxR%j@p19M?IhiF z?6RZuKEBENzP3e8!)TDwgz)(5YFOOo2(%jN(W;gxn6^I`hJ1Mrzv|R5;bN-9QJR1F zWu~?;xR3N5{brx!Zm-$G-=`)tr?!JhUPQu*?m8Hx;T-ItT3S>0_lMWj^Z%jew>=P47G3HuE) zu;|7Q`24(C%BSb!kW*nedu}N%oVXi1in5?OKamdhied-NI|bh@R$ydW4)yN_z&zcD zLZ5l|xFyINZ(H`EFC!;Q7G>;!_76_rc)t=bFb#C|S&WH{Cvx#8cF3POWg9Gn;*1)3(G>BDM8K4wuQq`d=SlhqL*p_g5&l!BY6RCY_9mkJ5(72kcqrGkCev4<|m(k!CXW#j>p%vDdHo`aZ4Zos z?&KNJd2u@)G%v-_up(T&ybcSdJELa&NOBt%dG7EYseZ4c3p#s=akQ=q7BBoCMd#s< z_4kHxqau6DmZVZi;W_saEh?I%rL+`DC6yKpn~bc85|uPi%6QIwo+2fMB7IZZOBy1g zJ@oth{s3NhoORvT^}e#{ia|Zl@3sl+NbO+`{ak5}qdv` zMjM@6ad(mn4X!ex+a}J!8@|gSHR~k!JG;}^QL#|ynhm$?kAqZZ4_z7|MekisfG#5k zx+HHW4!saA{yxJLoPu4bnf!Q~{3~8OJZL_y)fSVOEqlnDOTJ*SY%K984}}i;&Tiew zWYp;#5B+cb2R?su%5{tu;{eYYv?%5s`1|xhi{)B6tT~IS&ALrq7#<{=f?lx2Oaj59 z3)sQv7SZaIH8A44C2hUrNb7c*Q0o|VD(7i}|8CS`o}Me8ur?dFdNhc|vul`5@leLi zwD?iaFx+yd75si3q%%J)rDtkVSXjRYOykCHM)Veek+=;m2d|+&8V5kq+QDMA%mT5S zLpocMew{rjxW&F(8IUo(Dx^@U0_DtOKus|UiPpF>MK(dCeBOStjReO?mpH zEU0ZYp`O;6Om)09SoWR)vSkzEnW3O1Fg#|&UnhAor$lqb`82U41tdA{_5`B`IJ@W zq+`cQ@v>9Nkg1$3vM!7uAA`K9=`b&l_gACWwD#bbOM_8z{+}o!aWJsrYb^Q9Q5d^P z7538|;<|2G$;ak;-kf3zX~*F*k+@B^BgoVh~{sm z1z~cv3|jcDEMM2V9r~Vc$DxPDu|*sE5Vm#OKHYo`m*HJ#Ix`3-<0?E!qeRMgUozRT zF+460(795C;)}^l|MLOt+&K+PB877gCJXmnZ$=ug6ZxyJarnhXqS)07uSQ)UqeriS zKF=9wnVgGp_tr4mnUNx=248}$|KW8~jXAxX{cN3&-c#R^%m6*?D47TT{R~)5oh}~o zDa8L&zMy5s2?j-nu+#G{1pDvc_SMt*mYu4&;-3Z`I9(YW3KSrC$O``4?mEx8)y{1W zP2)eGSb^++u|lW#US1Tuk6)WS2tQ{p1%vrAN>E@G7Cyx`eNt$UO~gz2T_lHF%3Bvd@b#W1BR_7vP_PIK5-JuMI1&n zKg2KDJz)70pJ7h-X1*p@(5*j9Nj$PrX`RDts{iIDbys+f6V_@9nx()A!yVjj#!OB$ zt>`C*X*k;Y2>YY1%DeBL6-VsU=gZb=^QH&cEHtneY66~##e>brzhA3CrEUl0s<@L# z_g+{%!k8ObT)=E8K@X0QrJtX~!2Z40q1ief_nJqL(26uFJNYQFzyHx@tVRsjnIJ42 z!|A1!ZSe1A21rkS0h@A_DGN{KBmLIUH)+@DUa}u<>xR)UISn-VcPtNb+QcvIj^Rdn zSvddVEzET|gA;nDF#Rjx_^@^Y$e-(A&hB-vLTeCt;(wAKIF!M^P5Fn5dXpgh@NAGO zSB5uZvOy+PjefCer?*b#(nmr@Wn@?%+zNRD3B`>t(|!S%oVtMre|n*~P=V9n!+eU% zQ4*Ck4L^rlqH#qgoLgcC7aQK9Lumvq9_=TJcv%3APN#|YzuS=YpdNNv^{3|X*~CkI zBx@pGP}hleS2T~{Gyi9JYW`9@yL$m{+m{VfGS0K_E3S&e(FafeD2GAg8cB9wGFnzz z!unlD#V0;nqRNkt_;~(Yo@}qm`>#ER84mR%m(0e@t@3F4d?N z+nE{Qg6Jrian+3kzZ$}G{*&QDlTx6tas&@F(Ir~_q}k{3tLY)t4ODK$3;MI;3z*!R zjfVu!yqan=GjhGe2Hr8J2c@Tqaa=*5 zB0sEa1cRQ==D(5};A}`WL<;$j;w3lP=>&avSMgrtQDX(Y?kQN_KNQ!j7yACzRI!7Z zi*V4G^{`}X6Pqh=cFBJWc$jGw58r)@@0q%QZ%Q8w(=iNMCq=M7Q#_!dfs4b++LhfHD&;(9hEfF751IX>41l@{#P-y7Kmu{BjS1P{2 zN;^BcfAt{lm6X82a5LX>q=$XEF_I^x4CYcN&hl!9hx}V}HquoMnB|oOT79<&_IzYs zBXw!vX9Fm>{a|L(JQ ziwE<`FKc<=h0*+i&nQfKYeg%HJvlinEaY46^xCTzbWGJK3=+SC*ZQ+q-Skwk_K-BL znf{*rNR`H<+3~nMdkx0qSE6f!9*jlMl97k_(2k?L=+I#9 z5Nu5^$f?q{VpaON%pJ_adtkt$OqwpNr(XX*;J$1>+^Cf)e=FR(t>dDwdr~HLjDJpG z;urR6|2BSS&^@9#{V9*uJ;R?QlyTLCJ9x>qrCg_@3ulzp;x}m#_w$|0k2zal_o!p2 z=xTyDbE07Cy+b%c$61(Fzd@~;DMJ3n5)LaZ=h6qPxlh4y(b%8UNptudzUizw@3B{e zl-?0SUS5cVjor)3oC^s*HHrnTb^^JuJ8W%i7^E)l#r9Wg=w78Z*zNy`$S*F11zIy` z`N$BQHYJ-Uw6*ZvNk_SCK?FZI{S4PPS;pt>&xLEI4Q#>87OWPsm^@Gkwm%#V>pJGq z3jf(6rwjkEY4K35CGg30KA*$#H$8}%F6jL5IVMb6M4xNTr7^ZS2{7FG$e6gF;u%OZKzJLfC_Qg+7a$T=K+?-@EEUYL8Fg z*Nso%LwOUv@%Aco&8mme$G3@nej1->P>Zt-#&R`H37J& zw{ur>RX+Z*A!nb9QSMV7FX~#wf0Xz@;foTeZdaiWzT08X;6z|v`c%Iz68(zDqmjJ@ zoV2vZ5dv>$amRV^87ATr+*Np_z89A1?Lx&GGv0l49hrNynskK-*S6j9>|V@k*!FTE zJrSCSAwgkaCU1|~j*jK?SE@py^)T?4O~VVOhuLgPMebU75Pm+^zZc7?07)!72XQT_V0m~LuKD;6#( z9dVM?LP7!WUe*D5B2oL!IQN9y}y~ta%<{Ptf=JwRUJK8;=n<9O5iDHW z6qLh%i3<`h;rog-JeqArnZq6UrMwu$Ne`d^jbY!1O8oD-Hr6GmaFvOBQRuB>o_}JQ zte{zS?c9n_R*r)k%|T!)xQ>2qmm-sW*FnVn&*H7}h^u4<@~Jx?khSxT_&}pvRu~eF zD_#3@a(_4yR8{@4Il z|2-u^PD|LoCpXEnCxq$*kB5)>Iq-oKJZchv-YYmdZXn`euTx3&Q#W{}Uxt5*u8PCX zHH+&CWw`U`k1Sflh5yy7#=B|5aaQnQXlpoz3U=M-Ho*k{E<7MMzFCW_-23oamY~01 ze26b+nZmefa<~xe?D{_rWzq`MdH8b?3W-0k>A#y*xoL8N+KcNt?~{s8jso*~7nEJJ z;&?htPZqAnE~LY+3UkURS-#%$8}2Op#3mWu z#?rGVdH)Tu^g-Jvs{C{-y%H17efeYF_EkmlEWi(5rKQrg|LXC@?J8E}cM3e#_CR=J zetB2)IPTtU&VE$Q;SL*4@SyfZV9-FUcDOaw%7scH>|k^j(v0$-NH>U8X!`$z~ozwkfoT5rXZb0V<)g$rH{ zImt#OX!82F!Tgq*7jGXQ&lR7TVzOKWiwL)d`03lRK zLf0*l5e}WVy!o0tuRrn>JJvkm&5b$`^aDgM8nf{6*&8G#+yf7tRN?E5Ug561FDRvR zl(!Ao%d>5+;~VsV$EARUH=g1Yxq*Dc)DXU?qn`gUui!6NYx3~WAzW{O2c~SYLMgLX z?8A?#q6X11GDJy<`O59&9q;F|Uyic;Q=b;j7I+OS+@@pBlN7YPG=v){x4^@bB4j#l zd~TsRD)pQa8ATirojffM(Oayce0eE(`@@zDqEjGvOBPQZsvxO$ImaKWZe@*gMx%OM zm3VK&LvgetH_^5vi}(H%X^xSbqXDwZh;Va7-90%3e z*)Ua^*xwi zvKP)o27^oX0_fIGX8*KT+r6!e0X^*mh&^*e$U9!cAJ+$qxBi$*FYK#?ex0iPP?WSJ z&_f?%);tm3AaN+yxeK|!Ew7BKz-;~bXg~cZTyu%VJ!W>a_mjZVU-tj(c!c=BwKm|| zGY6I2(?r?-kaQ@7;{B&*Ic<-Gnm_NE{c3&GbSq?IG^W!#53I$nyV_V+@c6QaT{q#- zkr?PM-p5Y%7xH*3ZRvxsAX@Tj7s|)Z#yV9$qO^4w{b!j6d+%Sv4{?*A`uqxh+Ruoe zM_E!dRfRh3v4lZYnRs%EJ5f540b_oBhbBR@wLgAb(E3#%LnjgCE>@D$1+wBtpR~mr z6^-E9iV`~Y#~8d)c@rMRp2DveK9K$1!=d17spzCa1YPhli#CnSL4$dU)b`#_QRcb* zwD0G6d=Q!;US}}_l2$5$@`6Vs&E^4{@+g5Pu=^M`xX7+;#33>=zyKZvD8hwFm25{u z9V9Owy{!Xh-_vkp`F-e_u1sxT3H+<_zvZ26 zGw70~QdD<#APJjiBGH}Nf-+(DG&&}QZufjk#_edP1^Nqd>^d_#=5sy{5%$T6V+N3R zNh#?a`WU*Ja&c{%JEokxMhrX$0~MOM=N?(cWP{&fjZ!@`+PenIHXJ4IRW#{kiG>(*|rmCd8xp%ER7VqEJ%cxm2SY( z9AJs~1$|Q@ft(V$Qrs4kGDMbrK;H5YYxVvR^k|e71$_f9%)UtNn4{0>rPurJFm|dTam3eRoxh8uU`sVa^gv{sy^L3 z-xxsOuUJ2 zX2vqFqBjs}CupoLxolOo6PHqbAR1t)4<83Tg87?6nY_(GT6Hgvo|Dza8Ur!-9qEEN zXIVQb$9Jr3(QH0;L^emE&sMeT7x^#sCmedaTb#rzSneYe=62%`t2mti`=VqZZ)B_Z zVsI1aE0@u-TX*Q*1?{ZGb_h+{WC$${w_&4{2lH^)MluWZV2Mqsz?&=v$u4uII9ZD8 z^fY6WS|C?Bc!xMYjG#7UPhsRK8T-IhEyy>w@u=@-AnM~!7UP%%8@8MCpwa@I<B}b4Z2QzN(1K zUN8c^^vSl`x5)HCyiy^9)$m`i^kcqJE2KWk4_1;VmfbkLh@;W`R!0o*WZ{2 z`?`Fo!5Cv`sUA-Q(==&Lq=bg+9i~5vlA&gnE1jVG0-mcBV_AnPc>0F%qpkgUK+0$m zcxpDT4^pOqx&!&dD__{3r;fD0sXE%?d+~v=t8gRw0}^9b{ zM-OYn*Z*eVM9D2$I?S8@l)j2F*&aN%OaV4nUjyAH4V+SKhq_hi1irUo(9J|pVgwu7ts3M?h+;9h_5*1L-hR?C3FvKkYF@Has0j zt{XdI8b@yqEfjoX*RWAa2Xrp=hmaa4;p-S|GIymo<1S5Wd<$RqE`%(9XWW-k3;$wg zVNpdSgbkSwdYJ*(G%1rf?KdX%o-f(W(j2nNT%Mfw5R;sTyUF9Y5?IKqU{OUYsOCnq zitG^VFDFNv7ET9{&4-!cUMza93hsIli3c7HhNu~Kl?q~V zR>LuiUlykyw_}<*@7PS^3q-VO7@CZrcQPozGw!sz4XMUj;+`gJ6AE8e9*qg2_+gNdIU-K{{~+R6dx}OLH9P_ci)p z@!^iBUt9#MX$-{+$1_2NY@%-fl&^jbkv+f zG<|+5d^jfsc43D=PD`Dp{&$>y?5?1d(Msg1P_O5mu@hqtKgRn@JjI0v-?5x*ca|&> z;a<-za5N_hyAJ$iQo~M&_KU8TYdE-ot9~G~#YE!*GJ(lm_$XfCp+>TF))I~H3qW$M z25!HLB~^Z(Mdv?^q~p5#+dno+q_WlPXhn%3ot8bH#)Y)m85i`YHoqKcs?dhhU||nh z;~Zi4YF8H6YKD%9r`d){OWBAvUDlyeFCKMDN$hU_fcW=cz@7(;VA0>RnB7M+VQ1sQLjOu7^!$gd=4 zBKPue*>mGoX839ZwY#@L^sucRLL;ICe_tf{#BYN=ZVE7Ye=pcCaHI2^vzfiXR0|*Z zliH1cL&K*3B@QJ4?CJ?8cA#AapJ|@M;zvu$Z*eoYm~j{{%~iB}zup11`R#^)C)H%T z@^Ct4?O(biE7X@gHBsWVXr*v z$yr0!=})H>A4-^hq&f@nJp_M*|F`Gf*NKZVL)q1>V3=V^9~*5D?cDQHbUq} zDUNOmqbojjQip&7YP9k?_Uq@s!}gtm3ElOiBh?1mj~qaiGt1c8m|-A#nN74E-PzVL z>zH1bJo#<(g-seV0zb;lCJ7rmiQnDm@&(cRZ4*CTCl9^tAnlE|aQ^Rr8{zkfaqe=v zX`#!}o;a}8^9G9LYNtSdw?SarAtEt{Z?Y3F0<}Iz2U& z4VxuTvJM{iqJw=faD%{mFAqkiDTiE4cA>3dA)V=8;r5Dog9 z3xTZs$~Lx*q1kqfay7f!M!^d;7&rV`23~KVxyNE*fm!t_R1<5EcV($-OD!7qh%{;r3cV2 zTZ&jWQJ@ih58&CSQ+V-W1qQ8&$E{k~Xj5hi@49kfOG-G(Nd#8zvi`Vz-Fz0beh?_$ zuZL+G3SjQ9hN3<1NKaY5UG%(}c=6OV79!*?4n^6K*PX_sYT-O~`}c%WD?XWIIY+R- zPbzHe_b~B#;b%;e9ZtuAw5YYOUg(i27VlRPL$Scv{CDI$^#9rnDGz$Zr`lrC-&J7b z`wpU&^9BjJlqJ}HQ=?-F=X2_(%B$XG(FfXxq2_Tum|NzPu|iM4*LlLsSoDn92v6eu zx}k8sN5DRbq;P+XhIp811luShaMzuLZ2F6jSeow618#X?qh4ouJT771PmY9L1Jaqs zm<{aV@=0XrO5v?@VLw@Maub}~_lw1MtFTSCEy?SYfoN7b9D-e3&>^gt9cU`VN9MwP zXMrBwxo8W{y{H7=o>hvRlJDU^qsQokzR=^bjO48!K_q8oP|jxpDjb)??&A`sTrq=r zc~lWcr9{^9Egmn4ELdKE(C_7?Nk@liq5rBg=&+(wJTcgf(p@59?m1ELGAKcrsFJkK zZYS+F(xCR>C(QcW0iK=r;M&LxqBG|Jrn;Vj)7D8$yW@^HI9CRqeEZDI-nwGR(2_F$ zQ`>184Wg+P@^rMx5y)76TEqtTW22H~$notnp>F(mszZI-G-5ZQ^l}7 z&>VutMNtLk%~a~=YuY{c9yLu%p}TKQp%BPfy>%abV|68*_v^{ZkKjCS@d@dPB!^TLIw*RbB87qVfP9TOLb{(0!YiJy#OVk7!TM;*-ey)n zTd9dK^NYhc=SN)a{u>_U8p&F;zeAsCC_kDRihH(~W66=%a9({1Jz+bG2b_sO;@iij z{RzeD0Ww^A{WJ766tPhnBgI;y)alvUP;w+imM-f5-7eB>7=}zz$6ZI>vu1B^>e+Qn zJnwlnmtQ{>|3>@JJBHF6Y7npF{>F=6fG1fzgYJacbj!3UG&koS-z%>m$r-1GN^c2$ zuihwH>3Rnm+hrtV#W!Kb5`p&@Ev2V)O_4{qflRN38>vIavSN++>a!4Jg{b#LjgzawBQrHuk&n|fxOvwGnE9aS=>2o3T*crMwA~39bpGg zv4QvE!Lnl=e%{5obK?P?mN5!l^zHb~73cX)qo2g{m@8F1pFrNc%O?9>WQqHUQqi`$ zOSWlfzOl9h~wi|`19=Y{g5%|7)UHi z#OoFg<`1W-;i}!=ZOxj(F+Fv<@Y%lP`985=xX+(_pA-%68Y7`TRF@r>&gKpat+{vp z7ueI`NX7dG(c`{$(6D_I9u4k;`n0QJK?v3YN!S}(X=q+(l=z^Vn19U@`HZ9&nmR~@kP*$q*TaA(@6(n@wIEAr7( zPVo2JV)?T0P;3qviKZht4E1q`73D&wc2Af{e$yCgIKmBb953Q7{a6xs!H`%erjo^; z581<`W4P_j3Ecd8HJQHS7;W|z{7-+vxK@j_#OOgZwK4bNA8Vs{?7j^8{(LE~4Ryqi zHk&C0|8;2J|{cNFoSM3{~2vn!q zHTUq*+f)3mz6u$?`!`5cP2f$pTlmi(8NBY)BV2t`L-L@%&}%1jR8KAN!7rA^e5BQS zI^c#3Q#ME-k$Pin`?g8b$cwY6P0b$oSgb4g(qSP};g}D3SM2%i$-#mi`$AL-#eAk; zlIXAS*$3ot9wu!J9(#1?H@QXNwz~&sJg7#mO2R+Q%Ym_ttN3WKtVFamnX^oRI~RAA zXp{H&aJUR4wSFZ5cQ}a_&VlX63Lu?LhiOa4iTWwK+BCmki`(i4GP4W*Sf`;T8t5xQ zSNkS@t~v@o#tx>{?c;budq4U+>KA;CnS|PHmXIrwg$*M^nEK}zxXUk#-|6SZ<410S z&F6H4_tIJLsU8c3bBn}I@#nZ{*L}{$&g0sb7w~-XSpNFjJAQbQCV#ac4-~=%@MhzD zF!>Y@_0K245t$IWYjz3jTl5x+WOBiGhA-Z%`Ns~8(Zq+sEv2U@Lu4i7b`K1XAIC9H9;Qu_lw`d-pB-Qc*z(#+YWK7B7qfe^b0@e8FB^1Q>3MB zq_946`H$QI5EPXzv5l?iR~9>VoE zgiC!&WwH+8pnm5bNsz>rW>@_tw?;IBVeU1YO|S6JIor9=YtFTQ1J~bkiC-1#2($L> zy!q2z{`loY?vuC}Z@>PI2}*g~qgh_kQQyI<-fHv9llRcK$EtDFyUCKnM}j2D?=2dRIfkfT|hwsdv5HzxeCq`8{T__6E}>x=6`X+tUO zo$bWC+q&7~6`RS(HO=^xY(pOK5_Vo2zz5Xc;}>JkSG%r& zjnDQWkGd>gG*F+Dk$uE}?=|#O_{`p_OvZ;IGf>YbwC4L>Aajp%wbTwgHlqerWybL9 zPLt6h(3i*O8pGqrEmUosJP02M{}oipzr4MQnqQZ}w&c^`a55XO%$P=x?6^e| z>LtRxdbIe?Yk`w{sTW0RzFfmz7jNe&!k@=)c)(IQ$%czp?AnZX(jfoe_~EQP+?(vp zE;)$d@NhZm!Rj${gfNRfbb(J?W=G~9&ct~OgiCLwEE{ci1J)T=pyoSM*k!UhC~4xV4fUEh$HTgA{fwVU>8<>|Rh=b_Q>W>+tE(2V`1c96w(f$@NZNLuaYO?W!Z6Va^NjFf7X%Sm_13@-z<5QmNieXl|k796)2>kY2TH(bVptw z?2|gj-v5$t{kCQ>$lE}B;5P0Ncm_kgwv|gyDdKW@KlqSW8u0vVBzy`OjK{nkiSCiN z;-YiU*-A|baT_-QtjE3qwY$HdvoDe+-~CC$1|-lulNZpp$|1ZUbtr!{UFcW4_8rWn z2h-1LhWJ=*DDFF_f~CpP`1q0`L>%d5ImguSiQFhW@L(*?{Nalh1y#5-Rt9T#Jq3ds zg`!(WRr93f9X^}ys4XYiAk2?n_wMsdVjqO2WEoiv4x=!bVv`B4xL zpIFOJ0`} z<)Q+Wan^wKW4FuT{(lp zUk!OW%ee(ZUfVMl=?VPCIwOd83!=T|L+$5y|D>D4M&Z_`bGXvbioLyUjD7Ll;C$sJ z^lm2jFU^R2lK#Vt7dD~7r(7fgXMWCODek=R6duWVjDFABF@L~W^b_>oeR2QTSc^4S zxKtazZ~Vkw``#?~_6o!s!4h2YVg;t%R_1vTDfoCwD8^T13K`)f(ogY&kQ+J=_Z3Hy z`-0x|X-hwP4^y#r=Q6aM=!Ku=juA=6_7M2{9>!g}%+f#H!S*ZF*yk_>3TZw}*x|}D zcQv!ABQ&t@=V0vgJ%~Yf?l2pt4Jg0bji3Krixvr^Fs)@4ak|n1y~8`1kL(oT4SNpL zO%RJMBGBx$Hdk_gSH5IJ2(GPK!=|}KlBZpQuYKJ?_M3*VXHRlil0^tSdN-f_@hm2y zcV;x{g)AH>UCxD;1+FY>jA37{;`cQ_z->Y`7%2CNLUwqfa_4?=)r6U-QF9j`=Cy+H zudSeHHIb_w|ITI#e?z+uGH^F|9xb}0OY6t$LT`^b^>Egs)pL7j#N6R@O6m;^_pHMd z13BIpEK}^qs>6q&-bo`-UTqY~ytI)Wa7+=Et=qs* z;GccG1kjPAO8Q3~MZq77-o9~Mw=A3&+?dPrWs7n6;}|?+bQ{JSHNf(F$H}E4DRRi_ z0C#De&m~j03*8|O^kAzNEqk_JoO7-pA95y~X$Q{YKVR114o^k?TS9SOT0EZm(*UdA zY+?n%8D$fE4J=d@=~Lg$*tDSzZRWIz`nzS2DYgx4EqD`Ufzx^Gg&J6N6ayT-iGh9j zY{1lX4APehG_~F$! z4}4xW2y_om;*#xge1518A9Bl*TaCSmKAG2O_@F>2*Y86)fpg`Rd<>^Y^rN|DXW_=K zQtX9m%!%dTVu7!Js<#si&jjG!xEJufW}4`J_%Aj}vKu0lh0JAZEtzgH5S6-zaI>E) zu;1%$kj*HEr|;9n!vagORC5t-U(}!9v5Lpxi-u!CRV?|^cMg9I-HGdl3fyLX13xty zGRx+vA_w2=I9K5-j>><-lIA?dr>$>b>dAOCNeD)DRNzuhveZwp9>xp29_fs8;y-VS z@u|ZwzFSfX#%G7o`pinUyyGE`b&tR#bw%zWc!v%11y;Ok6g1v?PhPC-$F_7?p5U-+$?e^CItd?<}jId7josUAt_&OhT|5tGv^~;8Sa&$zWx`< zDaYSzhNdb>|L_8fpVi|>g~K>B;xwK)s6&U%PC(iHp@I+U1|Lw^&PTjRP;0?~sD?Cnc_LZ`MeKk~W`9Lm%1?y(6z;9>F`NYC@>@}@K%d0{**qvjc+**DjN*YsAjrhT_VN6Ro3GJB< zFAE*dWe)TbGB3JZYt43;TYr@pkN<}==4$gXn`5!Y>J%tUAI-RzD4Al$XN$o6W3hVnU+czeUjU64zX3XeD^%h;y zEME&JdS9}J*gEu;F2;)LHE5XjnGvV{zW%TXN9U}k1!hFQN4o#oP(+yv}(8WZ;)-)f(0=*&N zt&#+bf6XH6Hx!G#FHFQJ8-dU8vXA&hpCj+K5%jQ$!Rjp+#bXZ^qOIIT!NYfh=N?6_ zaaNbFQMmw*_T<0?>O%KS&lcy*cEYfsX&5?W2h-Q@5pNv(g{1B^;zd3SSl+fFB<@NN zv|Jufw=O9M^RC%^U&wa&7FmTxufB_}EbyVWLr#*@0!6g0#uYY=jDer0^Wc5`E3x^G zsqi7~B7gjN27C_*qx;iUSmSS@Z}Y4-jGGYvXO-2cU4s%oYWkY}TN;f!uO`CJMjtr+ z<`|lkr?X4nYsto9!Dm{b3rPvy{8y+OS{~J;uPP1_!|8VP>Mwa*mln^@*je%?StqDf ztQHI3{v7W1JBs3`3$W_)SL!)YlUl6=Ivo?B;#56c`TLm#zpli0u0d$2_KBAVVN0w)b$s}d5wc(dyQyFhZa|z zxs(4_kp(ixSHZ=)L~46C9?mqKW5L_buygY!@TY~fa5T?`W*sdM_=yMExm&taD)b(H z&vFDbj05*o!SvDHBv`uXGOC(|^Yj!WzO(oX_zzO24;N)}$3spyZtg2l-l8%X@mP-U zaz2L>)E;w#(ak)4vIl>+U6-j1Jx2Z2-@;jie2kfMj;LNthoLVHP$S(TIC}d$`r9Rz zxsIox^wSZyTM2sInp1XxACd(>z#QmUZvdCu-{HfY)o6Akl+1j(6MjEkMQLXN?e0hv zIc~Tq`ZewdSYFnoCXdybgjkXNCZ63gr@WDzSa!P&8K z(05rMj#j7ROOsCG+X!^=+)_G8wud%#560X5c8F6Z8N%8tYB2vmD;fQCBb_4TlvNKD zh!Xwgi?X5}_)wpHJYG0cMi!*vk*R8u@;6iX+TP``$nYG!|GbJWIITjv?Zu?$*K#WA z^5=j3v~YO95p0pYL}pxE&Mo)PES;}Pt-{=E@m&gJ0TBb-b&M9 z|2fm53#w3N5r%fMXIPopHk@g*m@Vo|g}e+~@~v?-ztmFBqjvf6KFe+VY3EP4V)B9* zNE^Z#J0&`D)d1@C))}rjo`BJtYlWPbvt()E4Lb`Y+`D`B38hs#iSD{NbbWyunHA|wrd*MT4v#MYXG0PF=lBu^t+C+CW?zKiZyCg!H-Mr+Gj@f|l+yu?qi-JW}RfV~HA(ek9`S1hP!R6aepZGrgaip8+j zXc}C94scw-2sVuv2w#j5GR*RsgWE;6z1{}LM{Hov4MNHJfW4xM9e(WD!>#zi1v9LJr=^V0$ZpUI*A&eP=x3I z9fyLVJa`a$2CkKRfX(^>V&_pyt|S@}Hxqp#t$B|X4H^ecUq(ayx8v|BQ3vdf9iacU zkAf$u7GPqt9*Xa5B|19#)V-mKJiplsOYZc8rWvb*zBU&!#W|H})F{%TfMk-f)|2g3 z-pJ=VsceoA!EG?%6vP^HK0UXh6AN`%jYV14K^Xj3t!KMzF01D`m!)b2!lV| zej@bRoLov?j&_35Y%iWH+<`r&>QLRiV`)cvHk~0A#%h$Ec$z^AOk8;i*5p+|KwUej z>@vk~2P-j0Y6Z$h8&H*Ep&xqGEIQm;U?mCnIg7I`tTN>_(T2S+ZvI8mIysc_!Upm7 z$`@jrz-`2P=Rtxm`$K0(1JV6lNS!=)Q-hG*ICPYzD4@idUT_!yZ|*IEH}3stT(B~& z@G_(~KMtX<4h^IR8+P(5t>y?cf$NS}C<_tIqsLPu^w(Yw`XXc#|MTcL4~@ISl{1p) z=!+L4;BC8n(3jlbQpA6h_LFFJXYsne4zPXRL{E#aP-P-C z?7q?m^~)#WooFEZ9y%0OqzxuDee=+5vmLL@n*>2Sec)a|4@?@e2mTyRM$ujoz4}#= zT=VfEV3(0AFd@UM|R-@f$eW*l!6Jrb0I)?EL}698;*~R z!~*qT@zg|DoGqinvbqIt&71v9BQ~V$Zp%=xxwPdBY;{JnHJaLpPx7ysp(9Da9Y8$}R z`nY(*j!byroq^qNU!u=)ZT|TFSu&`hT|8&&Cvx|H9G&+=&fojTn>5gp6e*RI5*qip zPDx1!MJWwsm+Z_GrJ>R;Nu>y-DWY+o>x4wMC^8yCMkr+SHovdW_qYClyYo8dx}ML+ zW6^NDV=Rl%wwF~LwP3}fw?codg7lR?sHst2Ojo#T!tKyHnCd(Vm-TJI-GWPtING7V zei~j15;7dcDQwEd7V#b9?c`j)BJo)`-6C|fB2g<9~|u!{?`dGcUZcRnH9GIypW3tvI?l3O_R#hCqGrvOd zAYH}J*k8uzLl>CY`a^8uJb_Ie@rA5!S&0`z9*H*jN|J!gaQrX)2DXforB=e+ZRM+O zY?}6(6e;f(EjDuEI%iz?CHH7j)Gs0qu}AR!!wVR8FKxO8B*4Ze}T*k`Nh<}rK`Lq<8qO6Ci@@H`QL1UCL zw8LNrFEDi<&mF@D@ueyMVV$xyAO5ZkKPL*l*@rvmj?UM_G4U)cxmzK6Q9*z&*o!Se zacEsCyqij!VDu+NJn9i;^*V1YFBf_)0oTUTOB;W~`o*KsL2(S;+82)RxhJPLHKTgwWVJ3p&|k9Cv#%lxy@h;%a9n64QE&ZGSXJc;i2Up%Dda<(eo` zw8{$({pSca12nlr%ULuzr^r5jSB8H9!^LhbGhx?-S#0T{esWRoj#&TiU~oFO5M-vj z!CdVP*iNxu6r%`a9w`E({W!*U@TUk{P*O3^N|8N{Kxh402aEaJ0p z3bpb~Bo+p;?=RiUB#33D^(VnZ&D#I{SO_@yaY?A&YF*e!Gn<|Y{Ly(e)+GYDCAcQ^sjV8iLNMIwpDN_$qD|BVTY^|H}8StFJ-C! zhl5N%(+PVj9pJ}N7l4BEWW=J`f}1uNY5NKBnRSU|tlCH@jQB%Fzcc0&?Bcjwdon&Y zo65H}sd1b6QuN+oRix#DZ``bsPwDCsIf45w ze#9SbATT;Ah)y5uhK;jo*ujrA#PaJX(7&{XZFp1zFy=fd{gWfUmZV7j+Mi}8)|)~4 zRX-SHb(P#0ae-ug%E7yrub|SdRj7W{9D5%o5RD~52kA)>Dhad3pGO!@U%d)rN0_j& z8dEXyXDLp)u!J1#yiPn89^pwRJ$cs5K3p(BmujtDg0+fPTwXVv82ero|IFzX%evhJ z)0s`oEM*;xOVfebqXt1_pafJmy(ba|GHBD1iJKo>AuCK?!ipKeEWL-q&`UD(uFH5X zeo+rU|84=Vv%5({Of=oO>L_gd;R8`)lktc3Wi%U=YPD$acTy}cf4xrZ#>(tjg72!I zkjEM^cKTjA7yeT8hdZv)yk!PSlbOOGHHIZL6lkr>K4Y;$XLfkyF zo7l*dK+lFE;%4zh%*qC#_oXVb*mnxv5BmeAbN8~gt}vLTtpI`RuEWUnpGb|87HFI; z#am|#x&KQ!%HONPJgxmKU3(xb2vx@M?;@c>KAxG4+D*hwzr-2V=fwPanRxuS29e11 zkJ!iF9D8&n@u{gZ2~Qq|r}qpI_3c}X=Uj_PqpcZC&`w}a0TzjDTlEig`Qri zF$T+Q6My)tkEeCNW0A8qzUWgX+Y9PNuJvkA9wKx=oz37-eI0h*FBR^`b#Qg3SmbXp zhYUG;26dN8qk_g7kX$_uq;w9lwB8ru*O#;KkoR;v+c}Ax95RH@zI+{~hNq)jlr)Jy zyA92cp95DJWjZ?kEVRH>;Vj;T(z_2~$VFYGXa0$v53r=RL3gYMOt?nobzf)3f78gq zs39;)ql?VXS0fd_CbRQXf{DVMN-STx5-XO~fl*^A-hHzIWXndeH{N}akp8?n;A#lz zwm0M7mkp!)H3i?)ib6;`v4h?hkB63~FqY$c8n-F9kc~?fz^5sgS1hhzzTrF2czlAW zwdHAT*3)2qBBO?-$2ieMVNch^WSO+y0xPw6Eq*jAlHZ-=%oYjn?Xip3ibn1n0y6)K zi2b3ztV=CHXBG3ol#rjr1qc)0J?BH(?QZP2&F+jv^L&UyH{`kB6&0 zLJt1RXQFYyfX$lMf-Oe5qE8d`L2j8mcWo1PD&a?oQ;^Udt2ZFp_6_i=SB;uaQ-vCp zX8e0o8`FkW!4d1dU}2F*j|T*zzre;)d||@onki%D)1myAlpNOjt>2TAM zQS8K%SiHJ2lXa{*Pd<7zvqC)&HhzR2-!Y|Ce6gk<-10{7u?a(X>%swSllgS!I~y?9 zE(EtGkK#@%zZ7Mi=nJD{&gx~;FCu+BPMBguGf|W@D`7}h6tXH()<=w8_$3&lA)Eh>( zd0pq7E|Fa8$Ya=ZS(C4y7Rfi-$MF1e$vUlGLms{_8dY9Lay6G+vT^bq%Qe@fMJ737 zRR1>*%4VjpbHAmiX^kJZY5TzKPsF0@ks4+$HIkETS+u#SNN21tMdh5UeW)hIi z^HsNr`bM1KD*2||Sl%1G%VlWelqIy__+2$R9Oi~joaNn=~pmRS&@)Fl` z=+%);U!0DmR<=zry*Psgy-J`4qA>a@t%v>izKKWur$JuL6ZomWw(*HGr!!{diw+if zpcQ99RgDkO<%60e#)9c3p_2y5?~>-E1O%60wK37o_vHHEv|FG1Hk@j*mY&gFh*G%V)=F z)cMi1yx>tIZkpyJmh#G{Ie%5CZJGk@I#ppQp?w$1eS@HTZ6l2FmtZf1dHdpk2N?L` zA>Ohyrw1>a661GuFsD?W-=C_-cgEJS&(F5Pw227<>xIM8rcKmg?Qba8`Hl`tGJ!cd z(P{oZ;Mg&NzaF7Ozu2kM%D0F%wY#xOHdSE2oZz0eJDA?||40s>Pj^)v#VoygSnXIO zuAMcQw_Wn!r!Ou?r!GCQ#^43?#`9%DuOSYvrRNatIvaL}tI>bmvx&;cNXQVEVzPNM zroJ74P69vDZ@_-umgHZzd)f)^YF0)or))wm??b%ur8VkdD}QsLk>Az4WK}&2K}z*Z7C{ zs(Zr(b4j>ppCkV3peq{w$BeI!&4(LK_gKvMzoh2JM84MYGB*gNeEx$6(4&7IAFU|n zvTNk|`5%q2?}m=x@f302Q$wh6YXBF0HpZlt$H~$IgJ@?_I(s!`FG^P*L%-=aF-~AQ z7c4ZzAJwUJwbL+ab0rNr9wWB8e}vDc=ZFVwcnsfy8HsrPjC_0*!(PQN=lbz6Wa&f$ zQe<3B-VYDLeR{&KZPsrhCw9T{^Pj=-yU$?qpE`&YcuKpo3yFMe3f$RP2(ho;VfASj zA{X!tEq>-Bz}^>+wt+;*|sigXR;FmOJ0lvIE2SFC-c9hV)ySJ1r~- zr$^_mrpp`>dB(2aJbJ$Y8GcQH9y#7fo;L5mM~QRTqV@m8E3ye)?Bfp?<8(kfAQdNy zO3{qiLfkG5bkdZex6awcF)|>Y@2KqpSHFI~xaA%998$#h{0rwN2Oj77kGpy2 z)&SmnY9iB1DxllaYkH~JC_#n#oLUJ(ZV~^_;07<=vdf@o#xRncE1)z9K48S zmlpH!trc*$#Dgzwzl$$stHPvl0Vum)o-e)MfK$Dzarrkr%!N}hYW7>OFAJqVoH@mA zd))HkBJmZUB}2ak@?hP;@c5TGgey0pS%VSC+m+zfwiBp+%a+GHpnR2c9^ZShm=_hz zrQzYDXinKj=CG5J7g2R0DVYSSG$xi5UvuM*5B>S|Z<1UpJ`MYS4d)Mzea7VnZsPSj zlGN!{4`mJq$b!D{G{fNp-}sgC3sNcIA~%=rOBe|)IwwHw>K@ErKaLN*odRx8H^cAu zpJ7^R2fc^?z(-5WH>DiKs`b&_=bn-%FCm(YTP`q?$O1l2q(wbszcbb8v*6@k4_>{~ zS#Vewp|YqBzPu9i*|nG0-y*3x6H94cW;%gBOfH4$xD_;T^?jH%Z6?*J?toJV)^X#D z^I=EWEvSEZn|(Q9jDJ?Sk*>ij@bab%TB}fp;bSD~rUV4@fEhRNlrTR~s*9tO4?PE? zuLj_6GLi=L?Vx%qHK~k#AuOAg24NFE!nafF#rqD{uw#k2EZ|ZCeUKkYm#(c8X@;-k z)>Y4}R!$oS2}#17`iC>Mub)8VMt+C!C$>Vo=^-pnm`ZK5UxR^^E*~*U0aXMpnC9QDTtj{fs%&S)+1ib zu-5Q4u=Z-~qF)6s!~Dg{z-2y?mfgeYpE9JaR`t|HM+l;STuNt38)4^Z2eRgUDcc@n zPgBI_MH*l3pw`ZDJi$Yrx0aj-#ab`gOFFFHbi848Chqj&`0sRbj4xf6vXEwK@1sUjBFUs_vc&DEG~IMDk=VE0 zLfzaiXnTUvNpth)itL;8)2wv*W5YjMpn9BQ*<1R0{WW24B=lHIuhK`(o%Gc05-L`; zqLyEBY0Tq(XzPEAS0fXcgKQKHc9bChc2oL%xEihV?Wgwz2JYldt*HJ-QRFvUf?CFn zx1KY1wRO^#Xlv^VL+d9u0$|#vGwfvOD>~)jb9%JqCFp^@M( z`_@KKJ;wx1HjRb*3$&@|w?5T1x=BaGTT)9+e|TK7o+^CyrUxtMg7enZa4y_Nw0cnr zH5@B%eWYz4mF>I<$!pc=2m2SyBmM&X9pFG~$06*WcmWpAl7phweD=9&C{8o}Eefk8 zZ0ZbSx^Q;@q`q|qiz~70T9qMn%g>~iofdHP&O+Q!c~&%Gfg89oX_(M_2S)1?n&J3A zf1D~i**R7G{nK$sDPKe852=Kwb4Ovx`3tbNBORoNyo0aZhapdxV_y4D6Hgzoz*TDZ zX~LKqYI^cA)hj$kU#0wp`I0Beya&^Xp@TR0M9I=)U?i{^L!q>-fP@EzV(Ow-Si3g~ zM;|s6a)!sP{?^9Rt}!dErx@(Du6MAu{u<^_&*LUq)Y#3MF$O|yjtcXVMkrLDz;tZ% zVfXd3D09n{T4sqc{6Y-2E^eU3`NLtDkma0)S$JTm3RImv2-#zNK+g6)dgLp>?7TEG zZ}A&AFXW>e`c!#S;4R|$>kht`%7gIpm0$x*%#b3|PpgAoY=w^pvYB8#o zrj)r-#anAwucJAgD){N!-EFMZ4rWu=jK6g0MnfhDSa7|qDYXr`!hAEM@rkJ`9X0$O zo8!I)9nZgENp_dWs#9@9wLx*Yc$jIh5+%JWdDX+2Qy{-{{C~)l?xIaY4)$wAi=@BGY||t7Zw-OzRX)Ikg4u zhHJu%8{QbTa4Te3e<7MHOliltv9$iyR}${+j9rHnY0?>i3lq=?+l>ZLx$LiGUcCWc zlz1(^m;k6Up@FQiFoAJz93UhhgG@g#j{df72QpKYD*UG`fa7FnPyBJQtnXZiIdTwk zE21IY+>lVzw}!0;+rU-cZcBx8wJ@-vv} zGm+Zg+)2#5K8u2OMYCTYLm?ca!PnZ=;GnA;b_Dze|J7|!JLm^Y?ixj(_zo0b zXsr{E-o*tLat6ywOM^k0=4`O8F%AE=muc^ahs^N{Ffup+&kC8shXz^Vk&;{3f3H)Z z+xM82OTP(~sQf|GFKwcV%?WJLYd6$;@Kcm9@Er>}l}EBAJaC2GarQiF3mHF95wsV+ zg)y=UTv8?!r-r^HBb%*3BiJ0Sk8C2w-n!_WY6mJoW~9aHIvnd9OWW=qhR3q!#jgF* zj6UB5GA9qA+Pl}V_w;U&YTi^hX_&_zeLuwRTz9Jd?^7b}h<*T1+8n4};(F#&sgH55 z49T+fx=?WCI0PvV23284baC=2_|tugp5KrTGi()kL9LL*S2_xF*0zGhj!`&I^(!?$ ze1!JQ@!*R^Te*>)JeXgrqCfAPqPLS?!D*D@=T4-u+8N12$Mge+w#dMIi<2az_9+h3 zQ^fMNgXGTUMCg@Bg2pBLV55p7H2m#_+Y6du{sLLtxy>54Bu#~5@lAvU-^6R9{qRbC z3I4Qs4yHo>HS6+c>a)}f+B5~HtNR)7F|nedA3xJGm3G!Mn~&336=o>LB5<{#v_~Tr z_W6lxXYBbcw%$IP?g?(e@5L)=2$!+$U%j0U`|${}OcLPYl}LI_xeuOasp9b13Znl| z1H!pFogTWEj^uY}$FD?cv!xJR|9dTpZ5#^A&%c8QN5;@a&IRm`mm);HPY0z>-%usl zm@EBUfrHmNgX1kpag_d5(DakwC1vdx-TYTPe@OtmvRX(FY?0y5QwwmN%M+4X97IiZ z6lvA9C(xQ?{!~Cy9K+jvtlxp zUyAT-6 zrkLuRm9hh$SBWk$9p?7=C?StG!^3iCcKbypthSkovFd`iZt!DxvO-`({9R7h$Vbz# z97)=C>kAYrxX`!;CF^&q@8Z>cze&J*JNDDCk~CHlphq%9&puV-OSJ_0`iCP#U!Or8 zXC%^VN8)L2X$uRn+X+!&f%KPa4V9a_kWNX>rhZ+`VBmio?TSW#ZB;VqN?1(m9W&@% zks@k<9H*mIaMb0MFz4ww+A|&LzLbralUiie7%TJzo#Sga|1rmvvx1oS2UW7>a3(#r zawSb3k&69I*U7b}TO@6N8FT-01xHH9^O=7ZVeB_em>6x$)jog68}4I--)sbU=)5L3 zS6K3S>c1h@LVsUl|MOZfF?VD+V{>5h{SJ7lw4Cb-{RzLzX=pxpKe2VMM+c!J z{b%7(!ropZ<@`GQ&A1D^MTvH-lLymYZ}@S~2~N%;LRWe(o{U&cCFkl>{kv^=PD!|z z57@+|cpA*Q62m+Wb&I$DnoFNpWm5a_heU>YQM2EF$>h7y5WKvayzx*K8HED2y7i03 zcJ70KLPyd$D1z@P8xL*cV~FQsvB+}S2DoPP79KiA@cMgmxqOGvhwTt@BoX_;$TkOU z6yh-W;biJ~aST+OtI-tmBJl93gZ0@m0^4aO9VhCb^ILjhx5QxUjHG4O+vY5>{xh_Q z4r$ay?4Bt&hOdgNGJA1fQ3>SCyboKR{1y#zI!*p{oQDfx<7)5L*m5`hejMw)4>K>R zapw!WiG}UDnpKg9U`qRY3|=VB`}=P24af4x6wjg1`BR3T^^b*Ampm3pop_1iY>Y`2 zSUe3wskDk4ZyKh@E!-3FwTmWXNvaSv134ZZZimfXl-%xXBpI%H^w^^gsxQ2e8`qzq z*P@TnoFjsR=GGy2e?pP??dn3+);av?B2BI?3c#-w<*;RA3Xlz!{Hjbc&a+bG%^%`O z__mXvJChRSoBiVE}D`>qh4Yr!vaI$m&|5UmS!*u*0CU**? z)FWq~0=agGG!L!6h3+mk@T296;CCNPCv*>?Vcnqu(sr%!b^hRe3{nuCrNe|^{@~dC?dG|to<=GytXf+h=JU)rPZq#Di2ARXsx!ELT zfGSzxVga!OZjy{qgnvxwfT?>2Q&Dpe_{|;zDNT{AaY-@}FRa9u5rO2tf)6n6X*Rf4 zET@TEuHxC0Np;cfPJDdwO)i%{ge&yyH4t=t+TxLxQPu?VtT|JIGYf>FL_~&D}Ocwfke8>IE z?t!vF8(t+7=t98(IaEg2Mc4192mPEu>vTTp_ONC9ch-@OYpH06ge`OI|A`T5y137M zFtwAjqWLGCd2se_{+}=}t^Z;~uf;#XD>{U`U$(%+bw{xJ_XM^eKMvn|Cc^p-NpQHe z7M?UT5}Qd1{Fwf6QndG&INe4AbF*eqqdkxDr^Q!ha>^KbJ;hkFMTqgl7_wEjq>0Q< z4YG8NCcd$jq<wGo<*KS%#C#eV19Z53$ zsKZ`vG5!ZL`Y;Jzy!`+fxx=VWl@9rEY6<4dz6k3+tr2}4m`oZJvS8TRboTVMByO){ zwcF~RLg7DixL;OD4xdyEJ0g}4qAB*m$jba9eWb^Nv%M1$&VHPb0}s0@lkBx+v6C1@HDTPbc`R(3FjYM z<}j~~`=Dv$EFSbBoJZ9~VUfgR7P={pov2NR+QJc*ACumZrSe-~>-03}Ry<73n}!gP zkjK8ayaIJ4<9WrA^IR0MPvBGkf&(u`(frCbo>wT$PQ$K3=Iki6uGgYLJNMA!pJf6^ zWgxx0Lx(;C*)|;{4+QH22I>8o$_?EvcAK?G|i-wUzzk zl}$44T=RkKwLb}e?`!cFT{}?j@_K%4(;BX1CS;RCb3tETj(V#YL9UY<8NDP5A|K47 z7r%_5eF@jeCKY8&e;W(I_SNt!>M*Q#_Qul-hI8u$ljuj;PWT?X1uu_%hLMB5!knf$ z9#Wi$PRD@0`}GULpFY8^Axb!@*9M&ns?fi!oy4DVz+L7|(EQRFyk@2Fm;r{+UD8O~ zD~{2saups_6^u!q&$w;kFgjewiVX>MGv?GXAD8@>gTtGYxhvJ;l1pT0mgYYAWgt!SziopX z>GFJ9)*4c^tQ3-c)0uK`GM|!mo{k^Aj69iI1&gc_XoyrfJ&{@t#dn%ny#7}FDMevX zojzSy;6o+CWT9Q5pUfMo3$w>fwya9WuY!*?Vk#HAUSAv^&c`z4J?;4XTc7CK;h379*TE3{p`W@`ov*V8f0r~|JRrz{zqeK8A!*4#Z4{`tLlLMK4i}H}aB!!x5-ep6g zyP13O0-Wc*2QKu)qJNUWnO=~^KLxi~K3XvWKfcT%C8|+ia?BS#-%+LsIb-a?Fe1PJj-Iy#HSqHzPrP~)pC4{wy!97aWY6LZsMN( zCA>Vl1;vi1z{?;Drw{o7bq8v>%@n)3$3?#|AWrBg2<-Q8e?2UBzsH&tpW;WKB3x)* zi`k=cNn?l&EWOkNmhv1nJvdBv?unt*)$;geLk@d?!HwOGvczYzufWNYG@jiuoa$w# zP<@l7khJg>C~a>gj_N3I*6ry00RXqt3UKfY3DNRl`qbuY8Chdlk2lKF$lyO+Q25*$ zHkS>if9sDzoyI+;`9_XAxvzm0rmw($aVHin-oOW~D8x9jsezKm)fN<+F}y z)C~;(i;DUUIMQ+(?U36>ZR#e`5&R~23jGDiOP|S$o0+sp^_SqK`V1`rAJEV$kHs~1 zVqL2mFPJHKX*G|)^FeW7r!|xOx#5YgDtuwruq^x=H3}u<)&VJ*KpcH1L2!7WBI}*4pg1g020Pm^|bhuF8)m2Rcu{iOEYbTBcn1 z4|8QkUyXPoDfV43fW=w&F*|S>&+tfx<6Ff1 zfZkcYYu;SmH!_PKmzKfLktp`CdXDoVe-QP}^f_m`^?& ziar20Dn@a$5+khKBmB<(wxV^il&I~tm~VJLq3+C&A$9%p#-Pi+MzqX%haImfAtK=# zObV8QRdzbqxce&BZN3BZu3lps&)&k65RMT!AK=C5p)evuj*gik2?OfyLz{<|@Xh;% z29GX_o+#bJW99AS*6ua9;fNBRUiKW854C4S?rn`nHnC?sbS;-6*$cx&o(J`CTd6ePu;(0k5aq!B!AH@B36j(;sTF@44&^51 z-h9RTc$7bB22WqVAziBo{ErQt>ewy#>Z$~%_y^{Gb`o~KpTd2X+i_RHvuBiCj3w>i znEKd>4T~5>%cuDAtJBY8Om7(Z5-VoO?s+&c-i-Gw&gE`S>AbV{KNeY3fI}4+EZW`+ z3$n-H<1b(Fe1kkmZ`&`P8aj}KN*#fd4tA)#qX3T$I)sIP+*z>lMH1R`N^rS;0M$x= zbn(;S7V}qOhm$vYZ1>|Kryud&Zx^|nZW)R2X(kyl+p*X>7vv2X9@JQh4|X)5zOV;~ zAD)Jfw3nc=Tn2eQ`wdQnS@qw>|Gtmp!UgZL`{qz;* zw+p$Os*kYgZycz{4TV_SAU^k#R-Kl61^+i&iT{f+74r6%NZ0h;;MbqdA1~g8n-Xog zO4VPuk~*9(`Zb!DHB0icSqu53iR*C9bTQtQv%}P*##HT#FR1MH5IBH+I8R)KQr|wK z&94!>bN?1@zyz6g@J8M%bV%LyRPupQ``K>K1nf1E<|&Ya0~QS78;nka#<`cUyyv?3 z=ZNVz(Z`anN_Q3C8ND6MGS-6s_*S%kIFF9hY!qc{-miULWdwy`MkG1G1kSCvN*;R> zs6BlYeu~t^fw%jh?D}VN9^=?kuVNmys|?Fe=Rw{ldFwA*46SFUjj-PGErBk!`wQfG z2UYTTLW;b;k*#ZkNZgh*Fm`%M)Ze&@zN=f(LAo-ieL9>5$xyoUqdN&V%ZEKm-8{;# zk$*QW4P#;U5Wqkb!gH!3vEna z;xC5{;!MFSRV3VRj=r4<0eJ=Fw$2FD3%fwDWf|sphlqy%Fy$d(1Ni>2iI}Ig6Ys9w zj|Z1+Li3zeIAcUP*T3Y&UBve=BJVczemH_VD|Yhm_cM6xCU@QytSij5Vqj-mv)}_9 zz=!6Tf@H%J93HTWZ+>}5?AEl1o!IS0zLXy4>c`Iu=UogxvGXW%HvWNYRq9xA!#TWG zG?D*SDaGKu+xXDs8NB7yJ-qhurj@ZY7te#9Sc&8kQ=RaiJ z!7kH*Zy)&tFBur{mi%nK@9te*6IRC-UD^oAORg}p%twMlPH>y0Xwm+-3{)?)#+$XD z$(0jF;GG)-o9F7FkZXb}dk%@$%Lc%?3-;u~5Cy8~{8ps1O%}rhM!Lt_W|T@A!CfzH z=AV}9llVsu(fj>>=siXczV!^@?$sQ&{V^)pNDY(mZ~j8eil! zi3b)=#3c^jaru)HHsX>aM$P_B+Ri2knexH7HEjXr$b4b_dX8lF_zBo6F`L&MUB=(b zdU09m0x3&xk%W?KU~TxWHtIz``l+4~xT4*-&}IfNTU*a;4=8ZUB57WcBg+q^9AXcW z&%oc3CG5w27Y--gaO&`#XtT|LACCVIXK7Gua~S5Ar~Y1gX7Cr$NOX!&A3SJ!XS`gD#i1USe$IcxU=pyH|PG=jTi;2;E z?{hXgRN#lVzh(Bu67-pI23cDlgte1xdHIn9ak;D#&%9EE?b)WHj=%)jJ+wQ|& z-`a;WbOn~_vxzw4)Em6D_9uJ0JC{A|*&s0YcfjNJ?WlcN5BlAviX%O2*cQ~pls0NcF@6mVb1vXlzryxa!bgR5`2xCAai= z+KBxq_8rJw6ONJ#Hx9AM)1K5;TMrX5c%EFP`!Y&iLYy$g6H($QxSl%>GVO}=#&~nw zF8PBD`==%_ZtjzNQs0R2?VF^P4W{vD7vs}9UmUlp5q-|RMu*Y<{GyQ}PkUs9*Y*hh zqeC_bQUPLzz%Sxg3s2&;Kl*&1?I7?GJTLn^W^?~719>4+!Lf~pgzmR2$!rcJPqLig zlT#^aQhr8`#jGXWCaZCIlQnM5I>6@nWTTSz|Nb6?Zkx@qxmg)WsyA+1p$jX5ld;Og zovR!X&drUBAR(@mO>UAUkFu^qRNp_+`)Vc8S~~&mZ4%y*W~Q|A-fk-Y?=@*JoQ?yj zC*(+Z<16hOtiPlH?o2L0gRC#?@r3ETf3`IeqfjnqX~+ki+7FY$E+JWG#PdH7;l@6r zxLWL7K5F=GR-yMD1MK9v|F;&jbX4SLwR|CNSSB-^c@4*Hyn#vr7cN8HgJj7Ei9&nV zKu%&i+bBGqv=i6D&WYdP#*{&zy}24ZzHNY@ubaj9XZmrMak9K#G6Gi)P2fSvGhzJ5 z@3;;txn9Kq?rUMjONX!Fr$h;msM5i7-_PL_R_$UkZm)58_ZPuQyNus<8(KG0eHTBq zqliyhe*^Yu_JcwC3HY(0T6E{mM83~>9GCI@f}O34G4Myf_;r7NZHn!2yu79dCT~e) zxx)^kPN)%o?p7wgiqZI{Cji>MO&3qqo{o(NB>083elX0+j!%z?Vj<_$`4aTuBQ}Xq zyEh1Li)ZnoM`rw9U>at68Doj63aF-*aRvPf?qlJ`rG}29G3qbjuKQh7c&Uit&8B>{ zb{klT7J}^{1w3DvCfuj@aPJ;tJlNfcds{YQlwJllXXSA92m4Xu*%M}|dl7cpP6E@U zMsP8iM4!i*;)?Wm(tcnEvr3A?XR+QeV^}Av!fV|H|>V;z#)Ms|NQam$G(^HX<{P;EN$M=;_26eABW7F7c?C z=k`s*tLd-ss^nf?(x=9=UkUdIw^XP)y;Im1m80RL=lF2Ofm%Z|X_R_87xH$UCfnzX zNAb3WwY#JiqSC8Q{Q5i@*Y!@u7a=m(6|O|5s*j+7KYT<*175?yec8CHO$)`c<+Kxw z$e&x2aJ8fSmk6LZj}DciEtO&Id%hW4sD>Xu3n-Q_rkF` z_#hU)ibkXP>0}2hgf%l8=!Tjdm{xU`a9?NGXS|D5rTs#ccuUAVqfXrfM~iNe3%Kl9 z%$HOq^Cc52@z<0G_&zR5-25OD-IX6>r_d3-*{w?>Z5^mwl{Zuy2Ma7434VL&Fe+tg zWEF7Tg?5zer*CyO;Ni7n+1c^&q9I0%YDyN-Q>A$Bl+WIS@A{EwdsUad__&GcCybzZ#vF7H zR?z&bH{eaz7*r^F#^P_qQ~wQXY4i3GJhbmDT13^-UC&Zr^To9_GSXT!_S$}`w%DC+ zcKw9%->vwGR8L;J{XWxm_o8_l)2aLJ<$RGzE%($N55J0g;bo&U@4c_gC)Nr%GRGPG zhl8xT3hlq4dLGrou6FaswRsg*pwKq_Hr@jY1XhpHIl9zok2t94#XY9){xl` z;&F#<5`I*%z|nF+Bya3cc1cm2_)8xoFBG+4ocTDW^Q!>b9R|VOHK}Z+pF5iQn36Z; z#biZQH7G~fqP~+h^t3I7^_h7P_Ui={FG~TZYkOFGQZ;P^M`Cx@(5IX&spnb!JXXPjEjOLd0w6(9^ZEor|+PAw8>O{%5W=|E=mQ3 z;kTfnvM}8TGU!Kx*T8*21(f0hU_;&jh$=5Yo9$Kf>HV2>s92g_-c>{+T+8YB zVXll8YZCLLO~m=0BZirtwtCxs5r$9v4y`w%Yi4hCqVLYdGQY;*5aO-NsxFmTy}zhR zUYGC3CGV#aqeKf7={{u^{(?t%NgYWK>m-BE4ko^4L)adffg;b9iMX<=3^rI437Oy+ zu5uz@=v?N&X=g&qYHq{rxtXX|7Y_4%b5U`?0iyCK@jtfCVQDOXV^mxFZK=YvUP-sf%X==bm9t)3~O zW7CERd}4LHYT&~QZz^-We1>gtVfan>hUqp9A?`aRNZ^2zqUgH5nibp(4Vpi(DfV?t z#(gah9(0M%i`xjZ)26~0NrdK6V${uf41tXeV&iq|q5ZojQ4Bf9zIhu#=ioXk1axLD#7AM@;e}C=kXl+;c{C1%lxh$zflb!@i zXl+>lpB6d6;PO=1hks%F{Gs4x{8kd{$+QF%^oXe zTUCW!zfR)MujfQ{%P78?Y);otz6brc(x~ha3iF2Q3B8A{y!2WzF&{IU9cs`+R6mm>M{&t$XrXkvWD7-uhz;$2rtKpeOj zR&Kli305(1efub`C@==3;<|8PL=hN;)xq}9`>c|}`>@P=JpZ_>5BFuay=*C6#;Ol?jE3HknwIRwFQ@*3mHm-2>5q3~aIKaOc0TJyYS z2(x@wiw5E&S7a*H*o=HuGe?s$DsJejksOq}@VFfQCb6g8Vt+13T7D5F-)e%z6zng{lCtIRH# z_~8{yj$Da%BO4*DrCq#nS2FCe(c=fK&Y`O?&kGK&hIBDglb-+`3=J?o?$DTDVig>(%O9Zm!rrVFTb>6 zf8MaRr_x}h*-vh`4uEAJyhyj;YFqpC5!&zgPjuo?9j^K}5xj>_f^lY(|Bs?G@u%wR z!nh%Grc8yTG8Br$J!{{hh!P?VXhzaRgET6s%ppXELXnUNarfGyL6Ro@(mYB^QW+{y zsowMc1^3?bx#yg<*YiBzhx};J0iz5v@?e}iNb}vdN^#wA>AybOyLdKk$x&eImhFYB z>kfdy?YB@Z?~P)c&8qF~epsjS_w;h(br@G45A~itbi~Rlf(Z+2sqT6uYH@Wg*I~4d z3x99}pL(_MJKShE;^aa+6l)+)eI?wOJ_XH=g%StGkxgSnus8XB^$z!!)$U%+^k-F_ z!2eYab^n;gM&P12|Vt)!q5F@A!+o8-=D|$I7m78$V0>;_La4kVLd~R5gWmw)sv1^k!53#Yh^;QR#1z6+x zB5QW_f(CmxGziQ?mW=qZh1;@aIQLHv2{2_Kc&Oou1L}`W9Hwn}=qBg&e)c_l!MT1K;dV67PORq8cZU>wdq7cN&iL zWLqp@mjLAcsFS<#<^q;w%I7k2@zvKmuvf;8-+lAW|x+ z{cyML8>(<N(=WP_p`OlujU3S-mi(ejK|9dMGzd^9{kT^Vz*Q6KHx5A^hiy>!70*bc=(s0?A z+@i9t+<&=W>G^wpxOw$)5%=Pb=q%5zsfbOb*JUhm-}f{4VZ|1DJ+KPKMr?uqmN(E@ zr%z(rn^nMV=r26fZ)>vbn^n z!Iq5sYyqWPtVHHNTIl`lPw3`T`{>GxSGfs)){9)OFU0$Q1K`MR3)Dw`18#P z78=@O?b$(5eNr2@^O`YL-#?n}&DNsc&idHTdnikNcan&b5SY|=lkOHfMn@jLjtwyk;){Y% zO|s(4X0m6Y6gl5m&U@EpKx&v5$_|gmKYIl5(#)Pq9rjHSAijcbws=YI$H~%qrY+a0wg_ zt>ScI;<>i?AKbz_N4e?4{?a#6mGn@z87?YV$(04E;#X-YJURQ0m9uURxTe;E$-I&H zJ}ZiAJUAV9i)|OwXK4%cR-tw3KQmmpaSsjSfEEqR;)dT8=kfi;MJD?x1V> zh3HdkPGuyfV&8so7=BY598Zd1Nap;CB(?i*4j#M8=T~oskv-c^qTefh>(pu8sI}(+ik0_rC6|V{ zByo8-YkwRZ?$05{k!#50pgCasaRT-jyvCG0aiE<(k#tC&fM+57+y&d~qJgWI@l1Oi z?-H)0t+O7A44f)O8Y^n(DxKln^0|BPX5}ch_P;IcpY$l&_pyVk5slzRoE}A9&(S7% zdnXckBR9xBAB8!-2CV;G5n3w`iuSYxiCTH@>W*qHZoY>TH?C=2ks&w+^S7t}_q6i-N{({6{&Ffd^(DD1dE{qN|YTjhGu z)|mY`Ejy1hln}>L0pWsQj@hEG+tj$NO)4RRpt+ zn~G|eTEXWJqXaV@pK~qcV=*&xu1K)Ll=bLrs1AA`4qtTY=nf4l8q=@>-_`vVt+6d2 zKEs{KhOY@|bKzCBkJw%qq3{4+-LxjE_R;WfY$3j{&BM$`e6OTW77ZxAMXffa!k*?6 z^wC2Ts0h=Co6j7n?fgZ=Dt&;zo^ixgR|I|2AJOka0G;<8;3lt2jP|;Ki{Dyd_J&A- zL#KIl>Z6UMF#ZSmp(RCrxMV_s?^(gah)nir(O=wEY{HJtxdy%$pTo%e6|iu_NL&*z zmrK5W5fdU!VfmCiPQ%q4WLGzHE(LG7`F>rT_-r*ceex5Wt0<(+^`_*Vqb0F!6_GMG z2jVvKAefE41rL;_as}_*VBT_jy6Egd6lCwlo=6)SeES9N`&Ev^h94%{d*jGCPX)3u zSIFIy7{*<3ZxKkTU8nm>+NhT6TRMIyh0moNnOeIyif&t9O|_Q}($HN)BJQRPbr@@g zQ)^tgmNE&!qrEP~BJ~+8zF&ZDJ((;l)`tB9JC=F01ztu+W1f5gS0Ad3L6f^gIrW!m z#;OLx#1TNELFd;(nfRVz||+=ZdeNnE?~8BWO`Y5c|r z4!m-?lA{`U`9v{|)I5SQdU6=(J_{?2eW|yS6}U9Hfqd6K(q1tLHtMZ3g5O;(e4HaV zzPc3`UKgR*$33{YYbH#Mlcmzh3t8EsM7rh@Mcq|;jC=oyS{9k3nXwwVedrw7Qsg;t;2W=4Z`rO`{pi6mX!LyBE;NcGbPuyku8Y0EGo-tXTs z_oh&Czi~Jzomz+y$Y(+}1d*qW{QK^sBR)$=rWO&qSyzfQ*_#uFRb}z$&)o3o-86dS zS^`dZYs`(U>f!dZma+9~?TC`BE|GuG2qsy^l*EZnGl?u@6ugKo7D8^sE`>?g>l}IsCQuMro(yGQg zoR^s^?Ac!dyV_^MzF)$c4R*_GW?q|C(-Ot8;M+5qo@Ejd`#G=QQ|=wuFD9n^a!XHbZ04%kj4{!t*(Gy;g=tw|KA_`jt5GrX;NTAOM+e zzR%I8pHn|nf`8NQ(0lt+(OvtrpkWw4PlmJ!O3m)DQ`}c>aIpe7&5>iXwU1NDiQ#y( z#0cMZCbO=N6YREbEj^ew8ue_av;3M<3@(;&bN2PYfZiIMoZ3aRl~TCGdrf$wQj2`M zm_^!}&*3=vHb~hmP6}=>VD`52dB#*YyH|LSZ4a|!<8E(aVcj`csVm9k6+dI&!aUv& z{DjW?YQ(0Y7%r>PX1=B=%pIP9iQF)fH|;E>{2f*Ez_Npwo{$&b9azr}6%WBv(uMWI zDVr(r9lZWCC6{?tudT)+QhED0X%@Rni)2q?K)NJ2jqryZa!OFKV?ZGBK!nE)Y6Y16&c>vOBY5eo1yX~2s%~h4{=Ix|yr+WuU>Hja^TfH6i!-oi;tx9E zv>1G{EJR85UV2r}8g*A_VA!TZsOz(aHSxSB`IkhH?2w2xapOhRGiKxO*b2P)vy79{ z(t*E*@uL0@1~{qg04!?S2B{;*h)%2ts#g6uRnXh^ljhjogzfI~^w+iv=uXwx7F9(& zZFq{tEURE@Jx17eN|DgHK2YMkp447{NyF1Ga%y)@VX&DraUS}B>j(DY-I_>xzH=YzmLpMl)e0P=I)AoYII$uy*%;cSg&@Kupxd9(G|J0Ect zZ##G;ZbxLDm$9L=67nYfB{?%!77b%0Nz^Ah^8B(d%`+GY!ye1xE1Meby^=aUynYz} z8<&HJ#QyLdGuhm@k*T0vl}(SxnF*>hBiLJE4c#%x8;rsYNaubZU@I0A8@oJm#7s z2wv0eSCBePTZqBDze_52G1mQ*KrNSZcx6sH8#m$(t8VC|mKLYcSiKKJdGX-Jk`r7< z%XHik9*oKEwRA~Z1SUN*VR9E6v1Q;3s)UB&(FaE`KwbxA1B}Tsl{CC9z7;a29wH0I zTeFl z=y`WZ6tf$}?%w-DYsAyoMiHR@xD0{Gsn0k`FM{>{C?{q|`asM@g}8iPMlP)UL(l&? zgsZC!*ah_m=#jh;UB~cjf%_xayFN3NOG`#bAH!bQ>fk41J*?U0kFGAA5Hl?sE{>c@ zUOOFtIs*@$^SXmgTU>{iTH{gYn5EV6l!-)r>>(IsT7`dmM_`5orMKI{Y1O_=T6Ork zXx`cxbjE=wb;9k&qcsSadQ;aGmf3FW3gr-5Fc*F`gtDDfBg5J{x22AlUj0@I|wlvviPC@ z4rvQ-fnjk@#HDZ|dE-%xdJ0C^^lBvDZrUb_Dd)Ll4>v&T8CQ1y^*vEau)A=Hyp{f^=+fCqDrx8 zeh;lwn1U8(&XefPi^=j8X5@))z*CkvpS&8LwBgpIAXJqi*OtN4ol=uv6CUYW=kQwQT ztS0U=JrWc{Jid)#u_w)#@35&v+b0cnJuZV&Pd4D42@_b6WDc%a+zfmE?IvU0#mR(u zrUK71(?nP7$Dy6)4Z%GAf3eB?HEG%UgRHzuS?-fF?963(POcYODow z{OZF-gvD|OG2MbgCnKTZmlVcddy3b}vqWL@z0vA~0<4@V%g+v4%Sp39o5yO< z7b!<<*G7<(-|NU>a|?J^??n3VRAPs?AKMbq1icJZ7odH1jq)A+uaR{0-j zp)-j&{!?c8lTM@k!Qsr{eYfDf*>4!=XuztB70kO)7EizBJ@A$-I6PL34V#ok2gk_a zn%P?TBrFEo6&k6OT|SMwJd11Y?GSB9?B}|^#FMs$0HUIl2LY+-IOJ)^bj3|!_@6p1 zWxoX+?dLPxVpqUpXCGMnSYp)@Kb(yFn9M!Cvx3_k@&nDZ+?f3uEj;*i4CZ8&i2k~) zCyN$7!XybpkgOj|()Rr$62=u|)Y4LnKmV0&ThznGbmpP5I7jcg?2dgRz#uQ-?#n*lwO zJ#l-;LfjR-l)Oo(Bq7Q!T=kuEM5qvmS79OeRVj-6g3VYnmX8#rlJ-sx@yNQbA?NQxohZ{N|`u(|jVe5Ka zpLq>RyWWZ-mOmn@39`ZyV!ugbb_z`2pG^#w*Alg6Rif&33RU*X5cX~(m((ADb%9gJ zI^hX{;pcSG&iXUxy<#GBQVn3mRYIorLmW~rd&9xK7xA(4CxQMVJ36G&T5a<0Hg^v7 zNT~lMxMaE%H`Km*m z@s+qVauK;~$M-op57F8pYjTFyR`eylLQbC~Atv%MNTVkBDcR8lcT4yNTt_2c8Wi99_!8_p-~>a8HWfphxO3lv$F6;Y$l(V)I2je@Fp8=HKTIz1&I6%mc~N4Jn+q zXa@`v^mC_Lz|yZ_;b28Qir>DZXF(Z+l}%v^6cQLGss=LNz-{o*H4K; zB0oNxutl!qMdTM4x?c^a_17?e|Is*N^DXj4Zh&mKSU^VYA5BiZ?*O9_2gv-+GBTIX z1fM@&OI0)-F*K`~tlzVVbbdd@RfndtH0wNOsrnOYKWc*e@*%Ee#VwvmtjY}EnK7%7 zObpxR4u+d=5K})Dvfx6AwR_TVVXlTMky&KVHkvM}YFZw_&hGPIux$?esK#p(PlG^x zSqnY2NE1Is+kvZg92I%>!m%zzo=5tfJCAzMw{{#^vn>X^qwLWzVGmpuj)g~!9Lj%6 zh6QhLfYm4iQs1%X<3OV$r`ywtZtFE4qJzS?B?_A9A<;Z;>8M z9pQ{eQa;mDI}-#iQ;ve!wJ)F}b^#A5KBniauVBn@p3&Q4D|ApAA#^&DK-Q>CAjOG) zNXE+m;=koQIaKzL++H)4JonoP*}0S0JCo1!Wxy#>{v#2T90`XH_c?eVnShVK|Hc~M z1ESg=fmnWH4DJyBQ2o=i96jAf(8iC}H5`AoygB|AdmNvRMHhHYtK=lvAEg1_Evwn0 z3Qu-5+!pKA8$eU+9pt(k#%~wLbLyr-*xXbGGv;gIsT`ia^6VEE=W(C=FdyM){54o1 zHJO`Ta)!(Lnn<-_5^6NZfq1_yxJ8ewT`OJoAf;jnh^C&tMH{;&R z?Qm$1X0_Dzc-SAl3c9YQ&@pzCNMw}^OWBS1)iDO=^1d(qvLEPj`yi{R@necZKCJ(c zC7V5$axedSvMrNj*ddV`KHtylhF1>a_5Asuqdb5i{$;!EZseGelPm8vMQ2OC< z4N7{{V^Q-POdA^ydFK$M6>?Fs$RG1%`8iQ61l^j-F%o9uf_EBBbAC0>Kk`$cR8oVNLj z9G<&afW1~@@#*5boKfd!c71IGM7Wyb*8VQRXO=`q7dnyLSN*`AJcoy7MKI>dO@W4s zFLbocgq*xf^v-{C;c`e1ccG*n=MFF9v$Vs}{_ZvwdQ%JZ4a{JH$t8ibvIq3PG@yG` zT)EcLy=X1B9Q_(z(urAv_%G)L?$sC#f%_Do`a>SB?|6(7j^!A%cc#c8x`4eH zSF`Wxyq2vg0DD6VQX{R*ZBg7rc3fVG2MSYQ=j#&sqi8G%QnkUD)@xkgvd8EkxR07` zhD>pT2lM!T3;qq`HMNnsobjy)Q1QCNIEiiSX4^w1llF#f3e>Jq7%zsyL#L6zFFbo3 z!dPGCbv&<;!rs>f!{F>Yb&9G=WF?nvK~%3 za193S9#5t=Td>NF;_O&(31++hp$V<@CTSg3@ zKgL@l9N6dSs%*=Fhp7Ig2qvl41L+T;9*-Ad`KwiECshQg(ko$?Y$H_;*F?QOX&f6f zAL~nOaDDbHoD;l~YO!>zU9%DYpPAP^Ta2nTE7+=K5j1MgdsN#}$?d(zGY?!g;0OO@ zu+wrgKM^u)byi>(@fV9)9>*Sj{AYdYu>tHhu3{P=Z$Lo83AShk?|;g@j^F0afjzh7 z*vSe-E@9Vc>Ue7ei<@0VzwMT!MUe{Byh9zYfEKr5k2l;4GsMot+MGr63}!Sj1m8bA z#Bkhx*z$e~>;Co=kF@rIigc!+(me(jS7l<}bdK#=dz*%s&4|XGUJ;<5lMT{xs^ql&N5Lh@^;ECw z5_(QfW%HZE@x7r9yR$&R%omkVpMAYJes>ABynkptYScE|^+BJX{>lV;Hal2v;&A36 z`i0>%i)Wt%;{}sfoW}8)0wv9O<|1arluqQJd8mjTbX&q6x5qGt_#Ec=ESUX%S*1o8^L;vhw6cY!5tEIE?WbvMe2z(t~bg+^a>0!C?&F zANj@-Tig#JKgm^JOq5`uIm5|t6D{0d7S5dwEy8R1TS$=gZSJ>D9qxS~jrt0D?7M9o zepyt`JsLfhUFaIa^qo(#v=41yyQmw6)ozFW$~fK!Kw(DRC|Yy(EI023uRUIQ53M&J z(!}HC;9jvA56s~i2?zB-WobBB`xN51stBzA6^=*ZUtkZPPpu7jfbs@= z@hi{gZ^|5jzDwRiNP0a?zjP6_nkYFY?@pSJCqjhWTkA7h^jN_y7bf2;P2LpG1&_nu zu;^?g4s=`=nXlGkhc`Jx`&Tn|wSE~!DE$=FKPqQ$rUpO_&u~1ppoQ<8bzo19&1F5h z>DYSQn3?`uhyIrpapbiJ;HI&WbC~alKkjI=&>Q(&pyC_w8SM*`3QnW1mIm%QsLra2 z!bP8J$6>x%I_|sv!uqOn7M|6L!FM^)nC`xdE%~m=4Eq#uRgNSly)Xq%kLc&R_x!;qykGO=VVc?f`1?FzDA#_S3bYD6kQkI*CC-(iL+V4g&m4!Z-9G62! zo2_RGBMn%>i*7cSuNx|HTt@cID}``bS(5aq1%K4%k~BUqY_7KsPgX~8?oWJinWPSG zNs^*}6|K1I`#NzobwiV|I?V7Lz+DoB_`vujW`%^|hQZNn=AL7yDJjGDcx+%%pC_{` z!wlKJpj{l+X|gYFKd?+|7Mt1hp;~2)f@rVI0j_%BI?h{_$*oNlG5eJ_MaO^Fazwh8 zJID6|%f(M-FRo94>{omTtjTrIjrC*366fefg%Rx9>sBi9(E#7{exrAjJw!R)K5Xg% zBWB=Dxy3gP^8{Qi_ua_G;HXX;9y=DpPpRoPQ4cW~ZAQ~y+ z_ryFeQgAOA6Lwn=t3*pI;6b3;jWQ^+avw&fE@d(jn(SiDXj~RtFNl6B#+@lx4|2*C zqS^%<+DYVNd(Z)E`RJQe=fq@TA=yTR-+s_?vuVV}>=AK_a0a`df$*-&5VcH8iEjK7 z5}i7l=&7$|`wpwYn=eZ6OKL0l4Dg**4Ss?!Yi;6uXbk%E=k#ce7&@cZ0blT)U*VNq zu=2mZpfrW=JKQsph;30|I(+_hV{9_KefSE$TJC32{g2ToPz^oxEI||`gwO3#LhE{E z;oqcX#8dAY8M$=|;f^??jFJ};FBcMWFbst?O2Xv<x$`@tgO;jXi^;A<&VS7Jh^1E78d1exO$>GSO^P9BQ@h zAoJHJlI%!rvMqj^aNP7I!f#VTgu8rQ$W%ue;ZUoR(9!lL2J@__{gW;d>EVpbd$*qG zwTVG{cnYL6gb;&LC#X64k%(O6iJ#5@eD0YJv#JI`a!osT(cv^~K3gx^t`rO5r*o*^ z(l&ru{a~GS77qHo2?V~~8FT#Gm3bH(62AMq|Cj4gBOJvM95wG3b$=|cFB#qy7PaZ3WgVOCJ zWJx!aG{u2H^94?F)M6H5{M?>i2onPeXkpMi(sw^s^d?7yy-Jh7f3XbtFO$)~a(7{A z%mGqwGo1t|H^C|K5Qu2f$N4L4h}{zlVhuegx8*Vpo)?g##R{BDl?hq7VkFUMip3xQ zMv-58rxR_;!H&9U@^Gw_FnePoIoob6ES1&~9?5+QtB219CFP^=>CX)acRWdtB{a}+ zD-P0$>0-oYyD6FXHjPNuC6MF12g}XvFow*q5$2{v2nBHl!f}3v!aMFx!uL@D?an`sw*He-mm7M9+;LN_f{r4Wr;B(N z&?ijf`M0uuL$EV-2%`J;3jQLX~s{o60)o`kyta_2Nr*PP(e(0MvhyD30Av~|GEj*^EOQ$<| zkqIV6P-^eat-B*le!1*~32kwLwKdB0X2N`kT>e)eig7{fvb)@3y?j_~%`@>E5<$*t z5R%*5aIAVKT;H-5uPzV(`^vZyQVo`^=g8Jz3t{r@M6`YUgBW>c5yh%{kQkQ- zquj)y$WRsgWU2)x#Wg@F>Snyq(IO2;_DduN5i#l%m*|xlo{+$sYG=}+&xpkYSEA87jEp&} zNn+>Ak!_z%$jwGM?9Xn&$D|MYwr&Pu<1B#atV*(fT?Sbi>`oLu?IOJxNv3SEj1N`VHE z_m>jr96!Z#ME?IN zhw>Y6lyN^idU^^1`hwZu)Rly+vLiv4-&h-@2g4C>X%dljn=aMVBYOm!K!Rz4_<#W8 z-nVjj8SQlN{Si2HZ!LtV2|-!>4@4|mO;nSN$bzp`(0#Io#trb9rBFHYWQQSnxz-mG zW^W?Pb^S@i=u0pzQ~*!2*OC69DB|>XDJegcLs&o^c0G+I-=6o96suQcSilvs^>Yur z8F7~UiYx@qV>}u1ok^lvXVCGzMX=D;out>^C9>B4NbzI|GBxocd0yW^nl<_7?BOKq zS~a?GbocIIw>;DmcP&E;%OjhACksP*OOCwH53jQ@5Tb1#6$d{nk-rQY`Oj z-C)b+{M^GHr(43OtG(Qftd)G{C?{0-(I`@R{D~Yb^5b>iX~OZm7N(saNYWN-llT02 zbKC9nxOe|2PR~_He_wqAo8nW*Rd`K0x-XDjQ&mXKe_ye+7|1{Oqj>FWK72Zmk5_Kh zL#z5U^3F*HQX3VBZ{}4fPvQ6A-=x6II~VQO?}R0_2m?07%x8KPd2f1#kt^xo$`ZFn}MtCtS*41&R;x?C7f>svjSF3ZwR{$HQ|A z(eU|AI^seMY36lygR^5rKRhhi{h}tk$AvMU70p=mAqy67bE0{ITJBNCRH{5e7POns zLbqxIB{SENhd({Z$mrQbeRV!;uKPeftd&PpI6{_g*5|uW>hTq4kM&;+P}Wq)+*1ya zXQ89XfLlAP@{1BptNlT~6zCAoj{}%qIfOA*qY%n&fZEg5*xGW0i2j{~HInlnPVNO~ zq!mkISKlE?REZdUJ&B{9XAuS6Be2eJn6P7;uCS%|IFZ`uiw^wnUd5K7yQUh?Ue_nz zM&#n3^BnFAHm6T^Z^V>`kyzod7Qc^{hO8l1YFl<3{dc$Eccba-U2vz!*t`R8M(Q!S z*<)bdMjwI7hlAXQf=!SvnFQJ^_-6~$)A9Sec>TGaUbt^d{r>D06pXxpqiqc^Ma>=u z2k!9M-z0j`>OJKf5_m4w5w^pmhDDm`vNx_;EZA%_>wScHN2VGtOXXo%{CD_sOrFfC z6k!F&wYiH?hX#vVl*s}Nuabb< zZo)t}J>g-$6U6tNklCiDvdp1{f}${Sp$u Z8*>#fv-~`THscx=d$MrPNy=C=zIyG$xmNhGzN7iV0#dBt}TOeWI9yDMAjOIA>>$^y0ubL4<-OnP=T_Z55 zPmUSfDqx4#Ph)UcpB1#f1`i`TgYhv+&0~lY>4H+48;NiA*Zdv9Bx}ZBBXP!u> z!sYLqPIbQeeX=HlSb zEI4mHhD2r*!IHz*@yLYd*2yLkEI!?rKIZHJ+@;4u%{6%SjvXg_ZXyvrFKP6$XBZ>jg&F~Q7+AxtTA6dStZ zjhQybp+dn7b*U!WOn!{V0v|x^S4Yqnn}AZkopIijdV0g>0=8CaLe%>Z&V0%ScrU99 zCMOT!q`R4Pt*IwnFpl>G6*pLmk5a`@Pbrw^?n+O|-@<3wUZnrE6(O2wv`ugcCtW*> z^5!=%(Cq-`dv?=*XJ(*j{2YwCw~!uhT?&sUHqhcfm+8LYGFT+nfJN1kY<1W#G!tLU z1`?&=@Si29&+%;K(=%~RwHXt*%))b-5!@m5n}YJ!cevrhR>RXuH!LhVgNHYJqYs~h zO}@CCtJLWcg?1ePv)OSFT6u(?88rirq_tCPJ%_2xcrSF#o(THxa-gW6i<|G&g1CM= z)_J_7&cEV`M1mB#cKa0M7Vxw4A_bT8Kg z;ASfZ9j4oulZ)4BqDkx&Vq3!{kAxanlqw-lT&sZDcbt1%R< zS(2$Ce%JqkYH8;Wa4lS%_j%>|$^J{F%{d%Sn z|6VY3Y6EqRF6Dd4d#Ky{E!Z%BDyl{Z*hHNyc5l)JX6-efrMoPJ!`DthVxc)Zy`r3d zUi`+Jmf>KwZVqWLwj!-3U!uf;nQVn&0=jL#$@dOQ5l2=>rkTp%qhvjH>~|y`G#mjg z`*lg((r%vb^qJd`9*-%X|6rJd8G6Ykvd1rDW6^&-yE_rzTpEBkSZ-?i514->pF}B02xLZsXNlJId_o|d+zK>fnDp7kT!3YacKY_ zc3#4-|0b|tI*NU0sOtGnN$ZdB-idvI;hy`be)^tj3tfH)!X+S=`|5YT9)$ zff^l~!e;T@pRyZubjuw*oLh65(~R^X2X9EyqrWn_ynsr0;jK;1@_vi+ejIgLocjftO=eISD|li4b@C^!0%1h(C3^b9@zC78!Z@)jIPJ7X{$uT^<(hs(RLhd zG8$FSUV@27BcZ;%fyy>@V&o?i)?7V;v1;Bgs9MTA+~mmRq->gcIRWiGHv{E2^_9V5 ztfB2Hx`97t4>aSlY(BT)ugANB_hZt!0MPeb4P~E92!|>}qq|(>W|%BWbj&~}BOgo* z;rWsiHuCy-H&=DW0$)6@-tgdI{-lLM-)|FDk#M&2I{eR=KKSm_9 zUkHI~S91n^qlv7RILqM1)E7pmSpqB|G4Oi=`8xkbXKx25ha#KVxpZfBr42cIW4ky=j(jTb+`y( zPI5%I;u=&v&F8bHCTweStzh!TGJL9;z%+i$WZ!uIlxLm=_f&6+z6lZ;cOqCtzzP2; zTw|d!`=nyQjcy)JR=G<-{iI!N?DuhOr=Ao!V82v!Xix?Y9+QOX8%kLE`v_Zg_XwWc zRE9f?C7BK3yQMGoaDC;{u;HFMw|1x;6Z((wXTas4c=#7>)pLS=0W~j=BCmhhle`I9M0Mv!OypT6;c-*2GeL)Sns`I@y%%V-O94-3 zI^w4T?!?JtFZdkJ!yR=g(B8L1a5}sYCk^YR#Zm+CURi@pYd%}8{%|D;u_(0;(>^MC z{rWq1rK^Lx+uX!O_quVrwk+k`I}Wn!6bX!8=)op!71H6YKk;7HY1Gk36HPNej8kN) z=~}lY+TQMnKIUCqaiImhVD}N)L+4N;?+H#`eC{`_m%GzjK_bDCs5>Tb7JEF%d@lp6 z6@_puzmm~EnBQa9t!3v^+(>|~H<7jRLMu@+j@8$o;(=DQ^zVKwdY?sW!!oNkB@g28 ztp{+a`FfO~U(jb}I9uQm$mFZCxdAelJ=svo3^tsnuC}}J`iV+bUK&Qv6+N@oxLp7_ zX}7tXT5qUudM!%mr_hkTKJ2m=qc6gCMx?^G<=4|ptH!EJlvDl4q!r|Nk z(-#6AlX7~A*J7vmDf3#^Req*Sr1ztvtsQDV(4JG7^ry;4$oQlKL3)e8a^Eq?@B9UY zvEHPidm*vuBnWxppz1WhCH|0N8{-MP-f{}#h7E~krJAwkJ0g~NW(=+vEEbhas0NGs z`|zc4BhD<*W9uZwa_ba%CSbHLS7|v&ohMe%!)0=?VQRnV+2+-_b5RBy;rrEA-#iGd zDmo(XVJkSxb0@KNqcX{QFol#>26FmZpK)$O1w=0R3QenkG+h>u?^~y$*|10!m9-qB z?D-y*Ny*STavrC>&K8tj3E{}PNbcml`*i7(&q!8fkmuM2!)(&&((xZ*{f+~4=I3_w zte8blOe{l*{>gaC!IAdg%Azrw^6217-V5D4morq^CaPGiinB`q-Sd<=pW#s`u{nrD z?K8yvUa=VR*dId#yK&qXY4+FMMQ}wz3tx;phWmW}q1n2luyt+}XlYBaee&LPVE+7m@{CCBR|!`6oWg`Ez_pI%I7VQ{za#WOW5*w^FWLjDcwSW9mqj>J zO&wxeo$&XLKC1OO9$H=1AmJW=y?i6#fR7r)%Rd#&F0UjeJCBhJX9ZXqy@)t!KLptm zlAyM1Ir~P^*sTmj=oO<>EpH#oC~4<(2jWTd$9#C5Nof0h3(@uq3-Fh~0-aofVfAwl zvQhUmH1E=*d%b_tTRcy#LFmD)mft9Fvde_V*JG-uS1NFei9OBz`;|)-Jf<4mU4ojX z1bR^ZhG1Fy9XfjcFY0~aE#HmS2SZs-7@0hgxySMUi5aC(U=jg4&n1$@*MGsE&zZ1d z*$vp8phb$t^uY}SchT(0AIbI6V&weRCs^8`1aS-8ncdwu^sV0|3aod<>a*k6s3pM| zbK8ylG%~_0vnJ@!JC657xp+P`o(m11f$pofOQdaJVF`0Ijr7ZvvI z>nsS4UoM)P6-8@W$75HnEmM70h7ODJ1a>nga3%7upjRpn0&U#Ec)@YlbUBRpk}+g$ zP&_z2yM&Wpofbt*iG_WcC!j8G3TjMHVn1d-#2u*>=(H)5=s&a};x##}|5_bLd7UHs zwr3J`?F}SS-U=+kUE#?@3v6>f%;a*PqGLfN(`bGQN^aAL-_qxTk4?|Pol}JemM7@@ z>C@3TU4V-x6;j_v2k6^YH)t~V=DypcV8X0%T<&5^+P15hQ@QYpJ9OePKFq&`3&!Z; z^FawFp>qiiEcD_Q+s)>E$G5OBqYh`7%CNaZ+RWl$InI9-1+)C`;KU&*_Ul$0wryO9 zA0@LuX5L&D6!eC3|FR8c{0)JRi#_n}{0@|^`T(c9E{L>bZ&oKLe*>6&l%#y(xyE5U zk1HaXYc;OJ)nhYpiT5bh5gLo)15?S#!9m=CMbqA? zVd;cC)V?>K-947f)*X0{mOF3Lw_C&6CEj;ll_yOmC&jbI5MAaw(~^n5T>uA@3IV)+ ziUw_Ev6^RPjr?m5Vrwhm$Lr}RQSg#eULy^mFGnzcZw(@CyN`*u16;~th9?VWun*Dp zcv#sH#~41Mir1bXbC^QYGpeZarGNBA)Kt2>GDEPS(4Vf`@eg*ch{X#sH@OLe{-i3= z*UD6)m^wb{qSHUOpt19H>d~&v%FQQ$s8oTxKRX1!j8f>bX^~dqIx}%ai4-0f-a%*E zEM^m5MWAnyFWWyq5cg&%5o&y!hNqw?FQP#dI4PNh+e)+2^+iy3rV^G~j$r$`Cb4P! z>~tc=o?Xto2_ef9c|E01kob5x%#zh7)jH=m^UUK^wQ$`3C_3-B9N#aFm-bW|B2AQt zke>UTM`p-YDUr&^h{(t$+CxbyC9|oFO6$4KRU||xE1N=nBYP)Ae)sRc{;SvP>ACOg zI_L9wzp<~Ic=jr7WeLxpvVJR9vUAoy*ypGH>A&D*!m-b9C{?LH_B&pH|4p+7>*(oR zcS8p`Un?n>^ItCVs-m#qRtg+YQ3Gk|I(j)-Q}B&Up{~tpC|(3XQ=tqp1NEfDt z$@@MHOlhhQ{0y&#@ZZU(vZ)4sSH$p~z0Y{oGe1e#!rN#?kMQF|FF16wjs|-UBK3-K z!l->i+4IwRq_{4Yxf(Nynv_a8_vBD`nvY($4BHxqM$iA zooY-i`0CPrG-va0=2@$X0h4#(_UD($2!F!h+tIN7s;}_r;4d^uP?y?mYG8|fm(q8i zAuw}V4_3EMo_lWAqpzOz!h!x7Y=byQSzgVs?8{~B!~VePjvGS3nJ>&b=daLq`5>)n z5r6L;V^n$E9V@#J!IM|)pdmpQ*ST(lg!!sM8`@cCy|(KYk)H~*W#Fs z;@qKGob@n<2L}(sr_G9RP&;5mYc*pVjvEW>Z>sWDH}&y=v%FNJ#S`Sw1LI#s;Jur1s5_zu3^Wp%{a?Ik@@oYcvm#trx61~C)<2-xC3<{<)pH8@R43H? zKZXUWyJ(QgMYcru0qacHVmPFbCZ%Vw-4)$%cGx*qRB6w`Qfq1PJUi6eD`@?-mZn0UTBy=;#WR&6~BUuO;#G;{Zs z6_1OhM-Oc1$AVQqipWRA!ebx#qb*%AAaW~RB_pq$jq8GwQ zn}N*bN+^4GbTb_K?Sz-o0#NN#qvYL&D}r30AB5Z(j3GJ?VNcCGxLuda?@oTs1I9hU zb>$PK6LTM+YU*>^Y@|RcS2j{xuW;GRXR2^DQ3d^1DYA_jg0MAR3xpj~nQK{#FeWfW zm{_@0n0QBy78lr4xmlPHIOZPhh>XAtLkC>nQOCUh#6sAR7|GtkYizS`FIv#Pi%f1S zP~TJWLiW^5Oj-*A!KGIC>j!F|Z6>^v;) z=FP&xnjH*QO@Nozk*qeS3OYr}VDbGqx)1ZiL0&s$4SKtTI<0=N_{{=}IDCPoiLQcR zm*ui6r6v^hXQ2?Pr2+@$>SJ!E9(1589=Za!Wne4_88JeM{(VVGhY6GS{*T_@SuZj& zKFM~zdP8#pl!Yx}+o3Z)2ILI8L+TA>R=sxf|33Ho_vLDhaI`mAY43cXc%&+}F|bALP-KNtHEdag7pErg~{ zpA6ZBSK;PyD|)fWm(_KiWr0sig#*b-sQtYV+OvJ&NKc6{`o$fvykm-YPe@@;&&!~( zW;*T&n?`c7E39g7Dn!poMe7~Mz<7H!<%{zMPxk<5_Sz3G)lWl{k&0A*ki7IJp8@xu z#L?4~*^<1ejyUJmF&OODjiv^QGcoBk);?8qWIKwxT9qp}r*95+d4(}s=U=4Iq0Ks6 z5{0x}Wh7acILEk7_cdm-hKEJ0{Y-&O@lj8#R2f5-o{Xe7r;>|-70q!xD_rZhOsJDu zkj*b^e4*w}x!dlM)e}o_35r9Hz%G2aAr}u9yTE6+BQQ$rZ*440pb^XTxY}INPdYlC z9UN*1w|7Z|d1Xr-{1w#1bDx-VpV=Y&T-8Iko8OH^SnpxN&-XGlbt?)jJO+=>c9FJR z61RM*ApP-nr8L3!8m_T?D|}j`iNp9LjJUr61IN99BjWw2 zS`5;DFJwkBJ80U+GWynFAO!X*mMoIZrLmV{Xoi9>42;skP>tR=;=LI}Eq@QiuV>Jx zlZt#&(HQP{I+GrTsPhjCQgFQKKUh^_%;(%dt}Sx#pS(GQ4VA~>V#Fs|$@p*d&3C2n zXyq~6fX}6}_9fah%XU2Fm0S=u zEZPWHdREh6Fy(`vP2fjOUr_IsBDxsuE8LiB!2SyehPvZ!5WQYraG7Q#?knWDGCYM!ceRqI2%7*+)dD$T-ht%#a=~G=0Rv#9# z4nOJZ9}CFbG6{BVN)m2)J(krjI7(*%a%5`%6|&*;3Rq2`sbKrzwXA+|vdrhSf^d9; zv*f>Z8cb)ny3fO1sK3sGo2;KKtqFHji(Erv}A$R*;p~17f zeDShqS;JX#+01Rnndg9QEXv*tpJm76#&!$v_1Z{371qHDi(7ChWDCWJo}`fn1=3ZX zKplOqlJm+@H1%6w{INt8N-r7_jjVz*)2_jqrH9~rmmTitl*4}`7?muVOMSJnz+tf* zJ+kS6BW(hpVftWcP?8|s+3_6hPUWM~yj1Y;+kwBwguv9fLPl+}sINcT1i$zIf51Ygu@wX@GQFLOnbm_Z;?b{wS2TO1b>Z-$IswHQV^9KhGI7 zf)_8-;1+xWB@SO9QSOblCC?7$74YV&sUZfIsv432*N;a%wonU1WA^>Or-#qVii zv9`vv%zZfRfe`BER^agVavWPcKppx7L<%K|)5UxAM`3MPfpE;>zUs${sIgtRN>Eki? z%{7dUH$-ifCv4A^z1vnSP&%f!fp*!D{n-S<1-a@c5-V9M2!b^s8EBpIha? zBFT3p=&jg=Y0{tXjK~^lP>FjwNoiMy*4Z zbYYMzVEmWzqTq+@^yuEScib%a6=VTdo@QXtvU50OixHUIDL48iD?4ZD2O2)pc^S@o?zhHsY$kKKFH z{(xxa-Fgx>HqM5+O-_Kb3!vB18N$ZeeX^fkZL-3WVKV3aH)X=o!?K88Fgehi}j?d4@!)1ON2MLq$I zzKLw7))aPPzbfVw{(=9V{9ocBE7oeETA+*qu%>Eyf60 z(}c8PnqXeppG^MUAjLJdB>QYm_a}vtV~r}~-pA={-xZ|Ce#q8IJZXpDUi$CGCpP`@ zerA55RnYS(qnA0A4sP}@g;_5<9WPE*d0zPzxVWo zEOB3>Y;pQAATftHl8k9@^ciThUqr#5-6abe+Jx_-aroO@2l)7V4K2y)V%vI`Q>9K6 zc#C<~u79%|>1TQMYcvQn*dyT2fxs5h&36O<&6p7t>XWVsU zIF3xOXAkczm053kGorQqH!h64i@GotTrsH<|s2v62mk6NQKp6Q=rnn9N{F zDjd)}1ZDk?(zXS0GApe#D38mALm}HJ?$}YuP_^4E-MLJ*%E*q)*&gh5Cju{=&Zp_Q zr|I%4SGL{b6(souqssKk@_y$vg_W6msX%0LZ(FN~T@g{J)TagXLqwiM;Ce6-tc4A8 zf5N=J-Jtxz4q;NFJq}ObPlHNFP>M?q)VlV>8k=`QjY0(6eYFUduV%6aok&a_e-vKy zOc&10y$>fUN=Zp*XY20GVJ000bZPlJ);{bkbE|cuo@Ns$>d-SWYrY9uR$W4;KUrv7 z7ALd}l&A0UXJm`#hC#b(2D~cR4%y%0WU*ro%G#EH7BouxV%k_$Y|OMHpJ%=4T|^su z6%axxm8yKsvQ~;xDH8UZDr34+G3s44V*6Ifu_hN?c3(YI=KpPgq+0oiEcD4hhn4M@ zaOZh->C#^_@t)2#FqoDAa*H=g%;jU*)yXc@%|i5h*581$KBmmINFNW~jRD)6J%xtb zs^GaYkUXoF(3|%Ppjq`&Q1jW&*p4$S@x9EULp*O6xIBQ3Q%=f~7LSDSzNz9p>c z9|g-4Ukag)*623455E?%5-eSwkw(F0xHZ=d8}d#G)7HiE{e4nNv0$G}*L*)4vPKEZ zlq1MuU3W24)5#RSRN$NGAK}ewDXLVM;xv0h{^r?{5vy8f(WafBgtw92!sS$ZN-8kr z!_7N{3h&uWp?5w}(L`)p_Lrp_oAKy0AKB1se>kFeoi%>Zr>374)a~{RcF|LGQJpOz zGwEo)YSr!E?(i9xI*a$y~Zo4V{L`R<8Nib}Wz=+?wnd<8f;mhIacwzW@ zx)nN==N>jhg?lP^qvI&Of_-pqu@MEVX_I`Ae1YHjzsf>WJz3oH!yt9aq#xp}%xvcw zF6>L;iW^0SRR3^(Zt8j7baJF`Yy{RYj|*NqVwmox>*ZC;0ZVFo;p8oQ&?=w?=0%^NFC|~$g~-}-vYJ5E@8aO1 zei5xZ;!cJ4v$@H|Q+#WtLIuB_!go%4NIQ*Ih<(H%{FLHJi1yjc94Fjh)8=E)36mV;Se42wVbJH- zbdI*s-76u&su%Z2<8&yz(FuW;;Tmjk{cwDV>tMOd7P!8lKL*%VfM;iSTAZyR`1^kZ zO_6cqpRdb%YudBm!=mTKu&jJ`bDO04gCq-Y@CdXs_@xNz_ z*{BoSFmbapj(K8_V;e*cZ(O_hT|Pit#4O_m(JgDWED?g0BEWQTw6MU{i)1^>5 zUh{o8&o$A;)0Ocs)4QE+$WnQqGwytNuZb8j|22euo`pSQzr&e9im3T95#qhi(8sQ7 zvVZLbL93@jkC_H^NPQ)At4>9hpyMRZGc_j>Scn1S)HL&`Et-zOT5xyL0gu<`; z$kcYHu+Vp{&|PnU@b$x0w&Gof>`c@&Sp8|0xaW7EF>T4Hp8N-HCp?5(89PNRj33vm z4CI$S?14|JR#AAK-VX8Z9989ejG&6Esp^Go1`T0W>85MSqZ zSLtGs4$n*(!v6bvUI?>`62{(jm2HeoCX1=oP?{Se47*hWNsHxh|I3Y7Y10OC?@a~E z^smf!fE|{du@Js*T_!t6JwUTC8aJwg6A4}ym9{`Nw!(=7rjzxmR=%ER=ogu zpDdx_^(8jp=QLSaUY4+T$^e!q#7g9i_rSz^`yJ+{;(l%JGC%T( z*vE&bB`@Hsy+%o1tMn9<7xrS6t*z|8f*F$iHGwo%e!I-~`P;G+YsP_c*gLwClg*yq zlouA|q|&REsyMPzm;4_`azAM;Z+mzImUSP*S6x!TZ$G@?w&pCp@1rfhc6t-M_^N^- zbx-kHM!e+dMKkW%8o~}#3Gvq5OZ=K>Ys>)4~@pKyPtOjJ0C!BlNP4* zxdOXA#jJZrCinNb!fp5M{pRE0RSw_^>9>+hqx#&5Z=rFUr`icmWDO z%z$qn4f$n5AMD{Fu(J_U=qAsHF#l%)`ke)@O*J%so+kb_FN9wuMl@G%GI#ml$V(*$ z@y(WDB12;{ey=SDo3E2$?zuQzAiC^#sxe++q0S>V^n&9t<4L%kCNvsEfOPFu!RnY1 z*qgp&OYWMo<$FzeS9L1c4lxo=_CF=qIR6#4e-?e9%l^Q`dHG=V;xNTJR>QgMAlh@K z2lsfhnop`fDtt^ZM{^4oK6cc3I&`*<&b6H2Hg2C_;Jn^&b<;&KzTt`o&zvD$*K2fi zVGsp3=DZ$m)O%l-EJTN~V<*yu z`iwx>qxphT`t2eirax&7ccx)|`%w9#r<64+AD+bz!>>sPARtl>Hhs4fnnjPdp3@B6 zEo_3^gQf5{O%IjqdgALP-^97SEsdRUL>LpPiu3vn$E`CjfnUl92oE_V@Jsv9_IHXP z8{t{L`J~84TwjWhy4y(?M(Rpi&Y!@Xj)So8{2Ux9UkW!O&xmu2v#{~+Tj;)E26vp( zPPV5;z#`GBqVQO*{MHvawCgRh6nET$;iL39AK}gUR5flF-J9ZH=wrm6xA10p9z3%P z#tZt_QOool_$(QR=PYhOrruCq>VKcMKkv;=g5N=$_^d4ml;gjg&a#P@d!X|N3A+B1 zLv`iuyl+|(*Y59#3sf%RNCSJEysQUyicUo9W2urq*K+CUy?9WFn1=;DOUP%-Ng**l z3>3GzVqEHfc;Vj?yzg^X`0~hze{onVay-V9je0L$f9(p?3~0us3ae1lOc&Ej`%-hA zEuP)!0kfM*B=6)uO74swD3~ei!APfO94ux7RhL%dh6_{hz^qu=ANf2m&%B6Jb(dgg z#dP|svQuWMtsywhYXo_RY%DAcM6c`$)U(k>-FOXjIp+3qQ>}kp>NYS4O z(3B}W01NC2A4Y#n3aP5-A@p52UuH8h2<^7az^o^KL2Fxg%-pmYMvb3}D?GwsUWh-{ z1O&ij#fiB1ad+wZdv{U))+sn5I*6lAj)t`jZ-sWNB6#*w3ny)B2jgG5qVru@_9o4b zI`;tmv8@&Kc9zrf#$goPjKaZ?uhh0*A1{0lCZFq}g1n_Ki@9w^HHKQW$g`JFohr_) zW%drouBqV!X_DwBc_YW%d~2>$%k2wL-xGq?MDn0A9DoH?Bb-sg>Q$38hYx&147 zZQV_A?Lipql#BY`=U`HFCRiM3!}U#Z(z&?{q!agQO4rt(6rJx5bpNxP>{_CgFgE)x zRn+$)&nf-URP#6t4zv?@h2kEzJQw?_E8&Pi=14Dn~CYlE*N@DkFIxyLzzi8 z-f~ryANin!D$m?mn6@RX+cbtNW!qzp`%r1K>j-J@ftu2NQF{2MLCjd59L5gzT!uGX zLqXf;pdd|A#()p=XxHp!Hre!{%=$nxn=#=$Sob{+Z#3eBQxXm4CNY8z_nqK@D+KRV zGa$dxihX$dg$!)MdELRm+%CV3#oW)t^Gj5*MW+bU-9O`nE86INR$uIA<bF|v@bdb5`J8w@qQcm%!IS>)^;~W7dB$g`Yrf%x9Ellaz{OBp6!q$+2|x5ezsTVc(L0yg>OFUp$ZAhH(o$Zt@Qu(Ga%E$UaxwASu}*>xGP_pUnF z1-rtLfd}Y8jw)Bo&7d^{ZE==vJQ(I#L7aS+YI%Slki20Jq}wp5&I5X1xsoLCiBh5p+l?in0%M$;IhED>iyLJ{TI?U zeLw?a3u%=`ciiy8n+kSwR=r^w2g|M4Tbbc__m%ihu?dXM*dcm~JsiK-GEZVD8Lzd z@L_Z!R4*x~=?!wi$ICCVI5iQ6ylTW3{;ha>+XlWcOYA^(dyP_yFL-=^K7JeW0Ee4R zfS|!*Zt7$<6up}ynI`{3W;-W_npZnX&XzV{@|kP6wEt3kx;hSz3aaq-eF>z$8HWd- zc;apKOQ0keV{5-epu|A>yY@20-U*>I?pHuj+Fksvw+JoAtZ~YQQ@B<9ANQ`Y#TBP| z((1wLJSa@yigSB#q12PsKMqItL<*SXMy*%>k!$`qlIC_uJ}*sz z)@A8f{d@=dC)Yr9VH7?ZHHXgTsGz#aZkegt9r!726zrZ2rt2q@!0o_Gs(zkGgB7Nr z^P8Jc;kJlpo!ERIpzK^iQ5iXnP zc@6`mQgzC^S3=SMg5mnML+IpJifj7phw+~WavOhJ^7+?;&fmO2ng{AIdSbqo5uaALCyf?&<+5S-AUj1Fm2Xh6TAI3#lt^SN@BU8ytX(cY~P zbM_nd{iiMU@ca&0S_w3zx|B7(F_Ux)odhj9CbR+!A!zY#iq^N`W99Er$>8gRzxs(x z&y&y+HH~Wy4CRx$mU6?$@%)ZQ9a)JPe~+9h$@_?6;Y8{?(4MS8$DYh*7t}*Vw%%fL z=BD`D@Dw)n9EpMNH8JV$SNQS0T~hSH5u2`6nEFQ_6;It>_BF|7iXVcgS6KZ(w?_N%WL<$@DX#>Lo5R3f zItTmjOJ(0Rm7zHBJp38APU5$8kmP$v6^uIaOE7q6k6KIhXrWOh)ZYmsJ93uK< z3`=0icW=a9hh_3xb?BYed!aDK7+Y3v6;_LXT=MFl!m-OKDBp1#EBdHO-yLUp;PibTxgMn8aB&?ZXZXIo1Y4{ zjTiHQ&VQM7!!Y>r?R_>rHTzW%B=Z-wG7owLv6?kzaK4`h40@0v zf#KWPBWW}x*m}X(#(~)PWHyd?tN_8o4x;4EI$XcxBW@c~i$_@!STBAz3#gZZsiI_fBj*>=M*L< zI`TfA_`xdwBtcm05DKi%!`FXD!TrM(g8P)+bj+&u zb3*Z6o}A!ca~ZVf`N$UY+k$0Q2(=Yd%1Tv}Xvyk9@MrNt4Criz@wV^4eVZfwd+&|m z#iiI#wi-;@hwvfRzv%6vCORRUkhzz;LX?67YgrH~$9%~0p@Zi?g$JVS20= zC|+N~&a75vX?B&u^F12)#QQ#MyjKUZu6crA{4sX+@Iu+{gX6(FQS{FG$Fr%8`{)hs zhnajU#7U!YmbaXAZoVs?JFyBIlZ;*&QIa`bqNeP5u70^E>;})n97R9Urp&-OD7i%|qGlf&KZYU`ssL?g=(N6<`@&2hx!} z;D`A!A-)x5M^En&B3`|4X#4#TPQ4fBad{75slPnBI(-#1l(Sjw$~zQwyFb33r!PA` z1N|w1Pbg6SVSu0etk6-IZn5f3DdMc8>??AS%O29@xMVKXKRi{j! zC$oIgMry{7Aima}RXjAnckk8E@Z16lSStbZSC`=shTvoWpQ4||7f!yu!W@&+uw-Ea z+)>&t=FX!?|KAo;tUd)3HG(m|U@2(*NrRhXT4kYX9uzh!0WKRV;beOcoUl??c$Mxd z`|#*^$N!dud*%womHI-jy z!D2C%d$lv4E`Bs;N$*DEqQZE*{=OKNPh3eVeP*-$JEzI|r@mlKV^W~H(30I+7B0-( zu!D?mMcH5KA#ySpl6Eo{2W*F&_nk0W`96-b-;GK$W(r!`AK2Xxd8wULS!%Fh4}RTNfYyfHx$nhW z%*a!o7YdKqk-{b+qvi&yS^fkZJI|nEZ8+-8dV#7In^>2(q3qq-ouK!*0)FmrgZL6B zcInD=)~E9T%X8Wdm(%U>d%UBVQH?_1x;1!Wtp?0E^_iul8sL)&<`nBvOS89BLVCwv z8CM-nUfGqnE#Wis_85u7ujFC#j(gyF^%7e?c_dXiDBx%P5MgF?5**lkPAEG6Kr(8i z$Vt%nBb>T27AH?krhfP6Fu+pGYV{g}dwPsQx5%T|sJRp=GJVva8$)ti62Wh-HkDI< z%$OF3$p_TgSU+N~TW+z9KXc*7jb^rkJYc?ajAR5q#X2g6V()}nXiv0-?Rq)PXZL85 zcKM=X>rmHtRXIQ-B5hnhA4pcgqq@B4iJT?{FXMJ_pRaYxZIX#Yjy)zr~hj+t2 z^<{Ye+eD0=um%@twXuL9;#u?E8?>CV9F;P<@x8Gd;jfJ^n{AQF@|TMaahnJ@*BAq5 zW(Q%Vj8oe8s`)>`3>5LWBH};hcZyJYg zHA67%Kn6C9-H5*TXWK3SfQ!G45w}U5(M$&T!a4>UR562DqQ?srS<_o46mTAY=yY7Tw_ro|%?WUL+ zn1v>iDmb;)f^GGWfjj=4uxF?^KUHaCCx;fWV1q;)Fn2P}7_=Go3=2Y&hMzEA^!o(f zcV_$UmVv4G`EJ9;qos*dsHGz8C4H`4)ns(1|7I? zWh(?9x54+mQ}CUuwwMMh=P74JZh4*xDj$o%VWVZ(;&vKivc&hD`8~KRU5YzGV&Ht{ zFwCEzAbP$QLU?2vpHv4yxMRY2V6`+ z&2K;OgwajBXt!7RP$wgIsGx`fnEe zhv1yH67sm=Y9FS#Sa>mO2z8WxhxkY0E#B?DO!t)$_O97L?^_PZx;HN8ogYJGJJOX= zDd{u{PEYXtehy=uK9k;G6FzeJ4?5vv!ppbhl4fuwnBHB-olKOtzV0bf@G>IXFh7!D z4oqqmbD%XBaOaRhto7?7ja?LuKmuuOwi6!vw-#~24P-l;H z(m?mxWnqA+o4||vv3EubVaT97$`UgY8f*^EwYm-`WlOQP3~_eJb4b~%k6iqDLCY** z%hYsi*c1iJi%MzU5GmFvuSQcDr)6&iOpr^!)ai%O`}AV?w6`8|GD7I;?KeWJYc8lP z4whZ>m<2;0#tOQd7m?Kj3AP3G6z@%M#XeLOT}oKNH8Td|Y^!*BKkOD+T=fRs+Dq_Y z4PvL&0kplRg&umB(7d=6AI#R*Vr5yIY^(!9?+vf>x|K(z!CT8hY8h^pz;?@ zJ?T=1HB1Qq$$qwl3cdI(iv80{SH^K+=A;uiy!%2Xtb`Bw-{X&xw?$jOEMo8sZ* z%>r66?H0^_HijP%_p_7z5|C}agDLSyt>T{-|3|b7RJwjC9j5e4{yOAwSBm4!X)h55+D?W~7m=M(6?B-bMfsSM z_(6FeDn1fVHflc4u#u)EO6iT3aH-sBRg*qySRt7Hh{aA7OMDfSio-*Uq=||S(hVkgR34iEJm)Apx>yTyK8bn!eT;1x z0~r2h2bjqZ1exOSgWpq*M@n*WQ)IG+!TwmMvIK^Wk11ZUo13*KBmzoTHyJ) zow6rH(SqaotaF)#Y*0%SNw)l;nTLM~1A6rqdryDJs!PnRTbKwVS13waSuH>haXwpc z0s8E?ikG9Ju=~s-c=wwnw)GUtCkqZRv5Cog9E)V#+B+$~`U6clQUd?yoWGhJM{9;2 z6%u-?LQexN^fZ46gU-3b+-++yyek@py)1<2MKL3`w(o>K^Yd8xkX)*}?1imLyKush zquBS9Ico0mp>gK@u&B(OA8oSbGN&<;kccdbQ#Iv|iQ$mo*8(*wol)_<3HG_7h;tjv z`JsXuh%Z)`_4{uDRJ4RpMhH;ZOZXmNBbYQCBY%_VgZd*+$&FESp|0yja` z;WBuzLJc=z2;5I*?1}zXGOe_r&3(Pt%u-!4vT`K*vFX_4auY6S*yHhD1NoTt9xU#a z$RBS$!X5N;dAD)L_{5Bn{H^vhRKDDs^k#pc>PN^|s~6KfSt0l>%n%;`_#+88wt|Ny z?d3{kPx)mhwTeMV?Cpzysod^skQ;A*WMdoPjl4Tue9#E3)BeFbb0<1B=9|N({UO5AOLO>oi()=6^AKO;`GkL4 z*Ngk$N(?)CUGmL;zvM^5GTQe091D0e0sGI?f!mY%W2NB{h|04NpR14(FNur-3PW3!}iRTKFhHlqb%5AUt#CD9+Y7iP^r4o`C&`5t*= z?!Rp@|1qUEueoCnXCrHbWh%GXH0O`hm1a#V`XP7eT*c)xVrjFMHS;RmOzZN-lZf!6 zz(EBLyV^zyEz$S!f|b~L{Wt`A?D67dD++~p)lWj%ezEtldJk{fVprZeak{Wy`T|1i zJVmZkIz5qRlrmPHZ}*j_?v=gh>)tY^*egU9D6a%ncc-!9;#1VVbuG^C!=U@o(YV)0 z{O*pLfhM!HczECIm>yV$N7t57x^$5+OkSm9(1}r01NIOg&x5R;jilin$)?CB!11j>AQKw*#f5u<&E1a{

< z{8Ee)-xnXjeY*me?kS{_(*tUp3M;o zWID;U!a*kP4Nry2dcHaZ5u&R!@2co6RanTDSEy6KhGKR$UgVu*^x{s*?X)DkI~(xo z8fEm!r;lrg;`XA)F!^{V9P!RYy@KiNd#E;BC^g|~2lvvarR&6wxGSw%bqtoWH4vP6 zg`IL$d2#kKhYLxWf`39Wc(-`+C2zXm*2S@K``ZP$c|;;>|1z0=OdHML{};*E zE&E2lOpN&I3(x6Vsv5k}4}wJ~5%vacq}9j1Q|#v!Vfx7iSkkME*9V7ktqrPN`g1y8 zb#x%Vo;?QCp4sC>9aWrqD~A7@)R$kIKZMcc)39^S7=GFN2)*5Kfgku5!YBUL#Bh%y zkQ+BdSg-j5J`Eho*LE7xwnITeVU;{Jww9A{>K^4ghY@r)Cx6F962qPr6mNP{Sg!Y5 z_WPIx{*g?_CX>tX+4MU6lsm)+Us0-P8OZsRZ*#ec#a>>0Z8ij38FTyiTz)np0@hk; z@`P9I?16tZ%U+pqKBicQG3 zJF%ysa>CbKRrq6lOwzV|5WSnEOgSRst={DU`&@6v%lieA@{40sYY_zPlM*Do4&Nue zHZ}fGDUFUT&7^~)e$w^b+7ywLN-y$67XP0o3C-r zBj)^(=<9HNWsOZ1t1I^9dRH_!PpViye|1G+-%%BZWpnY5qq6koYFlag+fNubWDWf1 ze3OixOd_Y-di?RVCJI?-M29OT^LOo=>8gE`a5jDgHoseq3D@Uf;I&c6>mSH+j&@4w zk7;Ay9~1DW&LH|AR+(=mdE(Xm2jNxwXg^K8gxX-qbAclp$~|_UCbJ9M`L=y0xz^tfS`_WUkDeT7a2RcjJ!1K{gIiq5YAyGU3*mQbF7h9Gmw5V-hcw6_gHOFYlON|{66bO9 zeA0_Zs>&MXV84xl`;?{Bd&_Hi}p_;zN3H!86`>F_E^#5twH>5=T5$5v3iA(!D${PQKQ1qVfdw& zIMedKP8(X1@os`9>{joAF;>3tMX8*I2j>bq!=q@5*DSVkZ5Go%y$r{T?3IOa4bY+4 z4Lf!#Q++>U5WcBa9Gd)x2kpDUhi0RYtG^MazVL?~)SH~AYT(&L4*1442vTz=@$!I5 zW>@f*mu4U50|)QruEqCwc8m#+G)sYH`BQ|Q59?%wXUc$qCp|iJfIdW<@M->I_)?n^ z=x==*gvt8+){;x?!tPBxZS^7cBQ%DaxSZhj{l;?Z`$zb)fu8&)-J@>Py79583FPPT zg|1dqGUEi%QyYAN3W9X_F56HZZ=c3P*dm_%{<82tiq6EH>aGjJLW$CVOcjMBN+M_P ztwD*Tgec7enyEBTDl*eR2nivI(8SqmokXTml)ObH6)F`{i3UaA@B0I|E|{xX!~_cr2jQT#n{Xq2 zjlU#q*!%V}Dp+pCYm!#z_xLFmOz;NN^2c;@s*v{(U#I?C%AoqqF0tPWVTb-+iQg%( zA-|=+5YMh&^MaNQDEH_-&Mj?4hYx<(@opgwuKx%htb@hRLpKRCs0$FZE(nyWx8eq+ zdFZH`o4@mMs?fGJ|c*!PwzgY+GnKi?)`lS3<(WCKiwHL^(uY}&+ z+wjbe1F-q=aeQ^n3r8G0j(!@Ou;I^We5YfMi_&dLRK+P6KEe`qJFUYl#^IPVb&HUh zc?sWC--$LZO{Z6sccZUb40?$i@#r=g$#gftFELI=a(wVS`g@xhBy~mbGl41eavZ^` zfGV-$;$Zly6#)*a8DRD<3tr|FiI**%1Bv=yc;b71wmcj@pxCA?#|>_CIYGU)yA47xXbhbCdBw{e>?Ht%Q^SQZ4gzp67&`IQ4Z3srEz z(K{qCEQ2nQnhUEZoW{8q5l2m)4*KVV;LH3ytY(7=PFTEE{Pg^EDmEI;_G(U|fv47j z^ZO#v-RHhAR%$0$IF5y1@`A^6V;!{Zp8~UWZh=|8DGpkgi|r={;VfYu=s2Vlj9Of& z$|oHnwPXN28nhcX{O5>Ytpdarr;_<9uVC`Jq!1iU?}GKENFI|;;D=d1fva+X`|S&Z zn2<0Scq5JX=i~9xuWpFm6^Z(%N1}KC%MdwY2pXmCgdI7mTxsh7s+`$P6}v}6pIDh5 zd;S2b{0D-y|4%ra&tb5>FpE;O!h7bo(RQX6{y6%Wf4O-J9|mh;xN-<4zfHr%0~KNC zp$D|&)KOGgunilV8*o*K0xBuL0+>Awi(_T5ZqqY3rnFvMtC~k&{(H(bRaesC4|-^G zSQot0ek9&g8AXh~F;s6Y2evr~ge-leIyGGjEf z=h@I;OLZ*S9*@TD57EWg2j|9lLHNiY@G_|v@86k?U7G5W7E?>f)J0;XlglwQq8ie7 zSYpdNP27?cgHrRf@NG>htljjN&*{qIW%i!=qBXK0PzphI{{?d!%d-5>l?!Nqs026U zt;Raby8LTb&Vox72O{vI9nYxZ9NkQuQyqjS+pl9`gbbeB^pa-Q9}qfTmuPNsE7;v@ z;eGw`;oj?ffju}$vOxDYRt39ZhLsxzdFg-(tDv`=V(9V0ZJ5T|V; zPx-!x{K)^yBZry6s!ygQrlCLizR*Xcw&5MI-!u@0&rgB%*%4gVE}d#j6sDTs&Zb3u zgYu?Sl5=Sfj5*{3?;fbY`VHHlup%6GURJ=vM~$#z(K!gsJ&0C0aad{M1+5i!pk;7` z8uUx$HKr?A=faybQ^gL_E~)Z~LI*kcUI(PS+Cv(K3_$bo^T2C(5HI|E0eS}vhG|a( zcUR4AVl(G5h6fJCu)FX-&+P)R-_3}{l+YSuNUyJ810WZFO6dpBe;8^pokZ@u@6#a50o^=th$}0yF zmMtfWzb^Cc>-$NXP`J1wI9t7v+Te$!JuH3Y#eY`V;Om(MFfFzPBI{Sd?OEwC&^QMp zMacs9I2CRCw}M(&6+PEo$4@D}=B-ZusHN0$n5AC9*XXYYA}OYOzBPik!9A!sY6zp= zdP9<13qGlOhu>$};@sGH_}(}QPt;Auq6y)2RG&VRZEL3|@~mJ}g&Q*zG+?zU+;3Y%_PNN4q#KvfTNe-qIY;3hRb_hZS1lN61i)sWaBy1I4ObNp zliop@;!k}lXy7#&7lp~#X2~$!F~M;CzetlA7MKZd}E_I%hV?Az;}xI)~l;n;Ik8vU)-qJL!?W(%{4 z>|^8U=oUG&7a|csVLE92_XP}H`5fjRHQ@G^FNi|JTdG~{C_2=&gFaa}ga>@wf@vBv z@j>4gP`ckE+)*c@Rh}^^7nqn+*Z1L1--~$56Y+=XH7tF)5*1c|g00H}u;1z$;qSQ| zmjuOuZ}SD%HR2E4`|F0=y9QuEkUu&pl=1t{cj#uBZ1YcKE$s-S^IFhe8=g(K_bmhH3+XP={l z>~;SXSTR%`cg{PE^Cs27J)@gs`lDttH)k^SYqVj>t753!_^mAVh!InYwPJbgqp3yJ zQ6_Kjl)iW0$lL_intHYq%gx!(?z8WFQ&c6*cMBEfWi!}`am!fr_L*?9r~qmdTDZ5~ zE!tiaP5(_YW9&c-%MVsr8R8Z7!F%3cEhWJ*RXeE2S`Sbrq+T7?en)N z`sHe%kOLDl{jpjWv;U}A7|hkTs5KLIGH2R&&7#@3{!bq<2tUFWY`a6ZIxh#ussSi3 zbq4G;Kx~%zhPyedfcf(Nbo4SQaI4$_T~Wuu!SMpP7R8W9i4(|xqRS+u#8|v>Q8*;p z{|8CQ7W`n}URqGIi%k5wjeU}JWG&&snYCOv58j*?Nd+%uF+;bp$r}!^wjw?9^PG?g zO~?SDNCTFaoVcHcDZI!@rPISZMD;z>_|xJV;-n?dn#LrtIhm$R{B{}FTro*twmH+C z(=St7^G5n;&J9}Dw~{D7`$?bgDdI94Pl=+gZGfr^m&NPc0!5<>^kMVgP@Y;iLmd3U ziXOIcgpsp%g6~~b`g=zrwI1kA%dWK23y*bKkD~{BG4Td$T`mK9>vieK$XVliOf-z218C$+X9izm3*FAK%Co7Y6m{tPJmGweET+nK6WD^(ek@dfF1^v+ z2Cvn_dAy!0^M52Tk#4Jqzm>$1vCI8gZj3Lhy>gxH)E{JVVs``^aqI}Cr}C-Kwc)g& z_m=3z_y0r>E{>u_O9OC+;1vC|;Q~oW+zqwIXTf3rU^-XVi}v}xqLpbNPC59K&fi;2 ztgk5%zK?ZND2MtKA;IHw$y4&B;>z%B`^~Di!8N{!?Me-=$X~0$&%LH z><;l_LteclLL(B~UggqjRsqx~M1_@V<$XJKXu&mO z_PoiLjoEOL=GPXB^bHOQd^|OBO>hZcIuTEu4!co_ToHe9Iwl|ON{Gp85%snoM5btM zCmSx+lIANn>7=|XH2qN#93Cu!Ca<#sM?;O@J{ZrxACsp4pad?;&ZNs8jpz1m-F&}k zG2hi1D3ZFWM+Vjhh)zrq^MlF#;j48a?W*UZQC?Q&#|_-ryHqb`vSc-@{&*gac#ekp z32IDfeIk2ZEzPE$3L^%JX?))Aocxkt1MbpUNk7kDnm1;`MDdHNY!VUE#uqObL^MD4 z=C58<#Yd$vNKe*=@!l`Vrm;o&EB_3^M?((gXFKUIB{?|=@V`Lz=t(naX$`8dasepI zUZedtPv$c|>53EAl=EkiAilFkMEs*>lb8Mf@R3ES#8dDUHWYlO(J%hdpYE&a`^i$G z%Adc;0?o-llU|BmXg?88U9*J0+`ktr!sdhVEk~--dyd}ot)z?VUlEhlk$lE{!FTp% zEHs{(#3f&*QlDK(Byq@`JQt@7K1+KHGZ8u*eg&p*+WtID*Lp@)Up_@2o*PA@hqQBl zPZ9O?{+YkNN{SAimm|22Q^8ic6FzR51yj;EQ9ZS5t%8O7y&T3Z3_4Av@hAxaxmJ5iegk zvTS!R?Ymn}Z@w7{k5wA;UfV>|k++Iy42uq0P$MCN20o3y9 zKx#9mlFX@Sfg>Z+gqd_U?={k;^0KD%qQM)YUu8+l*BmB2*1;rmZf&uuNqb3%oapO>c2vQ5)V>Ds`-D0>=|4r<#??4x9Y!#`$tmF-sX7NuewCQ-or8HRPJw5i96Peqa zsD6V7C~DoOzuW)uwbI_SH=>yD*lfY}7oF!Xc3uaaoYAbJ+J;s<3;^lCG#=E?TeRHg zqq(we8Sk0w29J~+A^6~LKC4kgGC%!*jZq^=W!Y?2(*GQrSW*P%Uyi1C?nKa(oT>E1 zid^m_%%zP2H`2J;dql-@AZt5OPs=}zp+QZWaG)yGacbmj$O}4OMU%Zd>&#v9`$S^`C$oL-k#ua$Sy(my8r8KM1!?kS#D1|UYxp*V z&C4`qUv7+{e@+|G0~(5=jaiqeSl^oM`t3|(v!dxW^9=rI;xumbr;u#7ai>*y*#bY(qqv4n5x7J%?o-ja zP11Def5)i7x=S=BsyMG%;3JuiIstpa3&rs(j7f>vBl>LjC>F_PvZhZT>15`}4q|Ek zf+k~g=TGB_fTO0l@(H4pAI(H>b{mY{-a`jhEM{AK0;$YJMP_8_OZ&1)%yX;_+3}6q ztmsiA4U5@D*T&2bln_5ar#5LU}_{xdlZECaXRyMn}P;|GKJ3aVCK;_99^=qF>*>iXUDv4 z)jbN=b>HUQ*}?faw}0`wmrudia1%+fS}@8+_kfc8UOGp{l0G{1ntB!&lAOr5G_OfG zd%w0*Cpw(@Wa+XI^-Hir6bTKJSCQ2F$zZ+Pko}U`%YqAxCD!((82j-ikz_mL)T{Hc zOW`#pwfD!6DX)ZHTnH9qbfNvmRxG^|f$8-dFlV?f!f_9*w%(61lR~NTz4LrYamfj~*BDCn{B9xr71SK;bey*04={CD#Bu9yj2{=MIufk^v}8!l?Jm$IB=O?}o z5Zu#e>8Ztqq+d(6*xB9>RtNn6uPS>8X|>0W1b6Cs=p|pYJOpB%_~8(lKAz-S3KgnD zz~gT?jru!*mOT8&BNoI!&E;@d+tWg}tXIe6zaiWJmGEVpI#?#|f|2J6i4~bmM=b3k zPk$|8cE(0jZ^RY$ODB>{-adwg`=#K@5*bNZX$sCsoQC~mh6_>iCp5226W%x2k{17i ze8H^=w6orf>W{z7pFC*-sn^nI*;k1Vly~CzdJ%7q(`VDF9VHI_-FW`^DN$hmCcL}; z7MOk)`cM&?m|hx2pI3Z`kY!@py6Q2T_bZ1DE#62r+3Mo=RCTsaV-e1e9wvEn(;j!Y z&LRVTY@$&IqDbMeA;fX99U1BoO0M2WL)(R6Afx}3I=iZ2ZBaDcnc@Nu-NmBS$zvrQ z+P5KKtrcsyRYkMcPk^+=Z&1tX1$vDb&vb(BWBb=9;IOq9s$AnGCGi=OnHfnqa#I`p zS-c46T^|SL&RgKXn;BRz@44vNt^c@flNIinQbY$o{LY(WnrTFDCtPBmXuaxs-fvVi zsQI??w@z|Qsv-a`_+<->-zMr`QN`PS#X>}WJ6aXHNj&vtNThU-e-zunvQKw#?+pW% z_0$nZT`8myo(JJbRWo&7Whn8R--}z!=CjXJg6JCGW>~xNGsyRwg=b{4s9xSCu6jj; z%i^a{!+RE}@+6XG%Wi>_4-P}(%9}9up9v}Vy93N@7B5=nMjVy<@#hmRQYWpKyiaC# zUhloasR{oD(8ZNr?BUl%e2x1RDz#-K8P<4MG}U4$wIAWZmtUQW^Gi)|az+TXXqyFU zc0D4pU6vgzV|clFJKkJ4SLlwJkUze8;Gy`OCe9oKNiLDJ>|Gpf8JPmPLUujmToVNT zlLGJU6U7Y`BIr3aQPf^OSa1uyR&Lz?bi!nV$fPJ-f%g;0w&Yu zBOmG3dK)mjH=iHXQU%qW6;yiWVwPc4$`>X4g1oa~_}oI61y|=oubndH{EcFs&p7Fr zPz2+rHo<45C>n0>%WC?n$rR~89MNeashK<(mkenm@5eudqeq9*IZ3G`?d^AHnL0$s zQ%%Gx_qMUdeDcAnnrC*gyBgtK2S{L!-HHdpxX4G3m4(eT+uSQzpZ%<7b> z$ImUKd`JuAEo+7uyE|y-*T3{cwTooiKUvAnr+e7Ut{>#}q`7FN>qE-bU*W=RHPlVE zpq(3!159{Mf2jw+9`QBaHgpB1KN*A}qOV}AyarYG^n-0@-Kk!*CN(rDgDnx6T&A$z zTw(YSe(};sRx`h) zS&?ek*Ekz9&gawJCVG%0Hv@J5Enz0QM_BjH?YPFH9wL%CtQC5_{Gpf*|9+PC8cVU( zWw~^Yrv)GIz#Vh5pTOAR(*8ge}_r9hqL&@*<{iZjjdFEhXuCB!5WDS33bd+tqKbbY0 z@d4?q;b8DVT0C{TFHPSulG?p4;|VkWQ908T+~d0kO1{fs+t?U9vnHF;{-FLVuE`jI|d9=E&71-rB_}ah&iX3ar6Ae@8=S(veJjeppW!%G>+MA&7FSrD% z1Q$l&dC*(*noM3Q+@tNQ_}Iqf5MnGx&qN*9S6N2?zTu=qeUj0wI6Zstv3`DiUp4rv41l{PrP?i!U)aW2Vzrv z7oV%y2u`x&a7XeQ+&XzM?k9)fz4|H8RZaj=whlM?{f8UJb5g8wY%Fv>=D!538x)E~Kwf$lGu*Z-veJ)nE3ZKJI-MDq5lH_#90!fZ=j-CE7PqJ@|EH1cO4i$0% zROD#}_vbzZ%Q4+h`0+D*cRvB1N$+5s{Wx?jF6773K0leT<83gn67*Kl;z(-wx&z|;RyzMxC+c6v8F7ALwk%p-2zKCb@dt{%F8H2QW z^i0rMu4Zv8U(O{LCS)ujs^g^4Z~sm3GC71tE~VhihZ=bHt0q1)P@pgN#&U%>5)i`Z zg6HuBU)iYvBj!cGwp&3^oh^%ZtK*vah0#C0BQp@Tf`%Dq+8^8pHya_6A^uaVc2UO^&z+od9 z=1lMad@~UO6^@IJ^xB)>OOepo8)C!}KP}NnsRN=@B4L8|IBXw39@7^IE}?rvxw>8r zf9=)5_XgjF7PnXYdF&ig->{iWf7}BL+*c5L=L~XQ`6U#X1?IP2?1rwIToO4r40LAy z2VuA6NtmM>#QJ^{f1eaA3eSzDdW8+7$>0Ty?RX+qxMzZ{FJ9xG{13uAbrAN3WnzqI zGunISg65GzFkki`by0SwD$C7a!kbW@H>f|TkMxK55%wfFtBjvE>k}`2WJ?3TR*G*= zx=-r2*wBd0qsYf!8%6zH&hbev-@utQ*Py{x1tcAN#6$L<5&EDWux|;cm$w>W+jbEq zpWkgRX_<&=S*O9eeIS*Zvk*SYUm`|*9paatAM-^$*IU z=hBEN`YiL_JD3*m3AzqALH56|kTU5dzvB`CCPr7FE>RPHZXQiCLf+CH_w;bkzAt>^ zZX)VyJ$yC@lRBkMrOe()!Mz!)WXlpqUz z4%%EbZY^DMY6_1$u#!3i&xghvirhJ6A*q;>DQZ^mgVRnND*Ig{Lt5r?QEDl*8-9US z-HwECrxsFcaTJ^sgJEq$=wWhs zyK-`&R(=%rm>%K$^%1DqKjDitq3g5Z@gfG#Uu10G5a;AX7EPp6H++b{Za)h~BJb;)AP`d|= zKJhdA1mb7ba3%Kw$)5X|_gni?;1)^YQHM0>=+MCYs}t})k1#&?a1z+56JD9QhkERg z#`W$xpfIqRlon@!Q+yK0ygdc$j4L3c-U?#otb&90eFQxB0KQS!87bZihm;o|xKHIM zI2<<#)X#qqjhY+<^_yh)jqxL3kc%nK9HEEZL)KwLq7x>lrlD2H8Qeo2qSlae7<`^0 zoOvi@*EmkOoQcsRUP5fvHMryz4iT9vphvjJB~0;$u{9rIRp93QacO5zA>}wK%vg)7 zd*d*2%rVgDo(N|xwBW+qTcT@eq2$G`5aN2`BrtqVE=@25df5w-p8Y2tJ?A3X>{G@Q zebeF4z%5|GJOeI=tX)jS6=Kc9Fst7;&)^%zJ);bC4X)+d3QGM&$D= zYOlaH^exO=Z-nnGS7KZ2Q+T4i8jr+^;LhH9nDl2P&X^bjXV$0V0lgGZ7#1fC#Xmz? z=us~GkR;z=1)Xv|k2-Bvf$0r>FsQnbe;P6hJ7x#tvTd92_UaetZX=wZKj!1q&*ma2 z&3Vu^p%|Xjbn>!_3Z69A1FHX}fzumJ7 z)1RN>eH8-``>G(nHiXBC%3=KMBy400uw?5|$S{)Q@d@|yqTIYdqv|g&eR7AdN!A0; zeQ)8wj4L4f;2LCZ41iaQ0?C{~1H~7wDzH8;W0ZcH2Q}ZngXfDJSnGM27w?qi=Y#zC z?$hbe_v<{&wswR1;xu@f{Rn0-V|4i$D|&uc0?O)|s2Kj97B;-%^Q^D%x4*0)_2WAD zUONLO=O3gVl34Pz|5bi$Tad7?w1p&{NNAiE33oTlgJ7Kve39yA_%%KTtW2h(^VbP< zp6(awlY9=;95=#O+v|M#a&PLBX^&Q?CP?}g9G6t?8X}qIZA|UInlhmTPn%05=H{#Y z@TYwyV8<$aJ2VL1NoSF3S%c{ug{Q>&RSkEZvW2pQ8=~~SHzK7&o-}QL0AIbXo<X9wtq*bbo+i8zavBo?)2YKLiTF$2PhxiK6=8w#^m=#zz0~zg zysk%+cI{F}liXNT+SCgt<5ICicBIHiX%P85N{i`eZl}x4C$dj`smRzpi#GleI;L9z z?^X^(1=Sv&HaCI3nlg$dZ7BuCVddDaSqOE}dbnfS6TVv}8jDv%V9u3kIA*DmMYpHw5>fnRw0*>}ckT^1w$>k}1zv^Hv=gXgVgW9f_K_yV0l+&JqlN!fENQxe z!DmZx;4NR=A~{ZcWsga6Z|{yvHPObC*Hx)h z$5(NOW)18d^qFs-CuH#_-liv(d-8o623RzBF77F@#y7%!Wa6SgT=sJTW_NXRjirKD zxwHT)nIWz>SH$c64iW8da!l#wXLx*C9p|oCgr!CES&J|OUcI4)&K>@ljt@{k^|(?b z%Y5;0;Aw2Utp?$KW#$8>1*6Ri3tXI^EcjsB$m_rQSdX)?TwW1h)eoavEpE`*ulvp8 zhlE1Bt}7^hx1tF=m>Mc&L&Eq%9=7y8*?cA(emJegr6;oR*-~o^?l9u<@?CKCT^5G* zZo*rp$?&cL&EJ`tH4fZ|-Syet`lzICn6crV^V^ z;y$DFxN>}%FU&VKg^{QS_XOX&DSWvphdLo^@MvH(W}g@6=}4*>=g<4dzp63D*?spGLkvJhDy{Ajg_={43lW~`v4!e)Y7H9 zD(Tg015wL=5$;htFMJLYG2nnbJPIj;)~i7v*H=LfX8h&mLWFDGqX%TQwHDozXbz*s zB-2&8nPgDJ)%+$8D@?o50bk{=!?$Uh@Z_@P=38@nprOv427N6=#}jgrpGrI^M=z^N)h@kWDc8&3zDnRlNC(97#0$NzS>ggHOX|@TOy0+`nl# zoQPbE8JQO_QpOcMmfa_huaCo|H5PQ=Tm}!jN1&yTg=pNX-|%79S<<5x4W_mlxPkiM z*nZs@*Jpt~#Fuuy4j|s77-kf;h^PKLM=euN@g?(%#FIbwhkkkqQ&>*?MW9YxkA>Lz?d(a6^AOeWjwg9K8mEH&;T@*kAByYct=XG!3H8UxlK5Qy{fm z9&2}uhjigBm@p)sc;(8_x(CAE>-8#}`P7mcr|DtIdn3GIm5U$OF2+JL2@aSNjZMuu z7`AdSvZ0ya8(K-5-D?5&KZ3sZU1H?{h15o=R*2X6()h>zsHeLT-A{Hx-=lsQC9;5N z@ zWKW%VyW9`b*^|cu)yL6`5evE8bcwk1LJTDSs35un&%xOCGdTU{EBGEFyj{!p;^yT6 z^!etI7&_S(-6p6@bo_gPIX6M0nm;8S@pSNtvG}o}4UcFmNb*BsFe2{(eyTbN#FMjr;>8TnV{z)@b~u5MvbBl zNDUfDLw_GYwVg3I!6^i$l(bX*>#G6WKfvdMlh~ZF*O2OG0!*sFiOb{`2(?_is$2 z?nY@-Dq5NM0;;O*X@X~nXyKP( zXny@A+AO_=3c~qw`pzC09=Juk?(H?Y#lbi)_)Zr$8EM*hG^i{?8HK*dM#@U>z#o^4c-WIE2m(&bBWUqL9`IhlwmYww}i#WInN zTReE$rBI*rbL4$V6%f1QPp6kObnyL|K&qEI7OrPn;Y6JWkP;>AkcG4AsNGroQ27>XBq}s_#sg~p zIv3*af1nn+CTMvkl?Mm(Cqpizh%!z@()EVPB;e#~9&s(2Z&SHIydan~96Ce)4*rka zxIGbW)YRwC$uOlZp;=sQ#TNR6$YQwg9#;7x2HzvmbV+Oh+5f1PXimBWb&_(rT3MX0 z=czzFNBhFC-cnj&yqS+G4KJQ6{Jg#{dN_uB%ur;cOsnTM7ffc+Bs?_Rf zNz)NX-ldASzsTX#^f5SY@&mdq`4G*k8pB?fgu(8Or64y|6Lb6jp$CotXx?7~w_hKp zc|BUtnsOA|ZGQ3F|Dt$z?K~p3cIEx%)N+GE(s=B>365AG-P@? zjT70tLiZ-^JF;IqXx=s6PwE6JW@t#p{Az(g#|~lXuM8m%{0ZwLRq21iKCqx*1D5Jn z@bf+?aA;??_(<+2(DnVxlZR>Wg{MD@UV1vg_7YQAEm{N)>$k$_+e+l>;RE!Hygti* zHJmyfUo4i+T1Q%>l0k8O5uKhhjIE9KqlWs$V(+*;e6-en!kj`xGaDE-1!O|s?s!O^ zuEnMZ91>ab75eTi#{HXz;0(t&dVctLRL(bJ2Sz?8Q*~qcKQ(3Cooh?Wik|TPokrsM zhd+SF<~WEqA4EN*EO=M0o@nIfVQ@0n4Pxp{^Nlnwa-E@1Nwj%?(S%tG@(RL6)8lz_ zXh(ljR{uSk>>j#|8?K!I-VrMFwqY4}Z#@FnYoGGY+UNYk;tw=OV;Oz7MIWoov@vhW zM{x2ShWP_m(AZB=wA-)YWB(GTPvtR*at%#;&nkCd)TZ{VrEEezTQ-l(i^RzoEop$;3@x6YhMMNie zYN4u-4}VZVM@`HGf#V9(Vg7*|R43&kExoyrX2;wmZ~8Zr-uJz*EG`?CKWL+E)wyhUo*zy8 zDT^ihQ!st)D=-$i6`gw*B6>PAm(K36qp!<$QknbnNz0?%q6ghC=p5NHuqiJVEw0-N=f=h4 zlMk*mXW~!M!Kp9Ex0%au?&cY^>}ewD9=@B-oFq%VBg4ozg+BN^R(R7dRm8I&rm<%e z2Vtco9s3R(#6|i}5TwWXy7!(K)G{6uy`Iv^i?!*`zgA?vZvk(g8qUXU@DZDwUI5?6 zM)H#2dYIw80Ye7v=W;cg*t${)bHBR4Zfz%iaajOVDhhnYBOk<)QD)d19>HXPO}j@*W6^8E!CFbr@tWI8B`ujahueP_{f{GUf`5pV1=% z>AL8#BttI;nm&1;`TY;#EwOz(U~@0gym=kczrPd@vRe$zm79t75^uP3=@}K@J`I!p zlf}obnsCt3k6H$5K>Z{udSK&Y(oz{qUOL88xl{@7x?%vu9(&kSaX5XaDud+eNAPU9 z%5Tq+!C4o*L3xA}gj-I=z0F#PC$wqtw5Nn$HfIMr%t-rfDHgcJhQBpl2yKBkp~O%M zmc5mst#Y%u!`c7{os`3GZNI>$6bf$fAZ1#qR8UfziI6DE z#;Xfi*U0ZY)u>dYG?)lnwXMjC>fk^^3-qKZ3w=8ss7~<{B^>DHN39hhWu6pT#Z=L* z6Ej(hQfndo=%GoDAhA;ZRtzT9~m`!}F)xY27ptZ+xF5);TAk<&G<8hgcQvNZyd@ zzoTg0!};XXSp)D}FrRLVSWj0^ze&z{o#5Oo3}Pb$_GGFiezlp2b{$c$c*}koHb6p{ zZX~pYRzXweL^MkJOD~LE0Uy>@6Q=`{I6oO>KGg0U6w0XM`{j1@VQvhh4o~LW>KY)q zWEE~Ja)3F5{)30+U&)?W!Qqp>0+yBcK<+9fRy(bjW;rWk(}-atFwzNhL-u3Rr7K|X zIt7nTst0L}P+ah3D)h=~V@^mue6}7)vu-rNkExt)ah4XCj>Rx;*T4KJPP?J|s2ij& zXrSIHduUjVG;KYn!kV&$8Qhpdu(#iOdd0$%7-!j{+$IOI=v^A=zuSi!w-AwKzbxW3 z;RgR<`jn4w>?C^XE5R)Do5<7Fk{8}$=O+t=yP%*a3pH=hTwQ{!k&-5Y*vi3a3OcmRjD zx}iYi0w1rX#C*O6kDio9@`ERUMR$Atp}JaXu}zi@yE$5X&d{5xz+BpDdI{#VdeW?) zf&=ug3vHH^lA$NQaPNtZ^v;P=`ni2P7@C&}^NZm$<-CaY*H&S}CyXQKEoIoZt1Z+} znDtNWXUXm;USnNPbl8r3)%>{Sb4BIX?8w}iebi`Cqj<}Jaxke0=F3~0*{gry>|$Oq z*<3soJNHYoxprd&cj-4O>3@=@2X|1(%ODPhz4X_s0@1s~EUNV}g7(wPQI8OP<9pX#0|7X9TZ@O91{J>cghj)>HS-59u5yBfkD> zJYOZdntHT_(kVlF>3HcFsxr0{?!FNCJa_+4+adYf?}`NrYJ5+NR*Siu-Z|QR^aLxf zJ}C+~_YU-%UFfUZ!vvg=E&E*Nz|sr$vSDq_bl&uOqJ8U#kV6;pctNj0^u?S`9+J(* zC-~F1-s`B($2$6Er3y3OqRPs@DOeb6m9~&xw2&9Ax1nRb&6#R;ABz$fvA}h!$fb@% zxIbejn|bLMy(+&Aeu##$us$s|D7=`4-+av^uY>76sgW#V+gD!RQb&31Ws)ihqMdyP z_*wTn1eTp(1)n;}xb2ICJf;@xE%;88c7)THLDoz~dn4hyOGSp(+eDsc#l$Hv0V*qf z!KgZ(`OdEw`;%BzJ7*ZSkz7*{_tT^QkzN7;BetRsDP%cCLy_S11v^I8a%Fu1;Z54S?V~h|?p{b^^jFf+^73rTvkzpFK{_$qaG7r2)hEtc z8%P$mZij%(4YcE`4$a>9g=mIPV4^w6tjnf=4E-Q<5_|n%#pKy+XX;Sc>3ED7`o&Sj zZQE#$L>HgTsD#AN8N~PGb!{vI){m%JC6hWhafE{Z;Vhr7mc6Iu6oC8O}$yXymbm`|JEZmeQCm`j_RVD z=GbKaBy){HzPM)RX-?Wq z*|Y5j@vK>?IM{3=yK5FrjVAfhU8}Cb<*p!M7huNb?&}p;l$+>I7bWr{a~ysA>N*`J z?4a!Q8EqT9pREfW!3HTw=jVAia`EdJUN&hSb?%I(tA;%x)|Xu2WP7bh?Dkol(Y24< zZXQAtbF{e2JzKu*ry38cpGWKRwlVKwX?E~$6)9RtgP+mO(rNA)#DFx}7{W)zyz z>t)vFYybZ{vRMnG^P`B(HBWGV6vW;N9iRzHm)Y#0(?L7S3L*!TP7OHsAAhpt9|6rp zWApu!)8>HFEY$g6J1a`blai`!_L;Ml4)Ud>Spl zDj^rljtD$7NhypEljdupedw&%3Do_)u(Mfwnw&f~f;AXW^T<~oaJH|Te%kNPmfbtR za%=7A@|8hwNqsz(_YS0CS3=ojhhelqRtc_NiRMdtj9FaAa5hQk{i|JnNZr+blA6#g z@j!WjBi=L#7e*HHn#)Vb?kqGX zT!detAL-s-Mr`ai8MeUKj~#n-p7q~p!n9xIfU9>rM8#Oc??D0Z%uisvR1YAx^haQr z=VF{>oQyMj`{U1?LQopvOUB3ECrSO^(7A*1*bR+~%;{|(pDEm9Z)@&ijyoLKhUZ3b zxWN?%3?4!|V%q8RS;n+_gWyzgJj<*^SE26e9qLP`(Qw~dk)FkC7&|x#aI*>9rTmQA zjGjl?y{T-3bRuaSy@8Dq93-`4kFyzjmr=!!XJOETHF)~acx=0n4#`rj00P6}?+96@ z@ym%NZkDm&KMq4cuo~O$Qv?6zC(+br=jivPZQN*BqUi9&0@0MTb>!Bauk`Su%M_;U zVG30_EY|oUGw$zC+2L50+b(1R4|dVf`Y9roe~6V&3!}~b@~Cs(U8)lqCNN9f(SSl7#=yu0M$HT`})23*-jbL)qbvBgl@;CE_*5m6*%%*P;}w zqvU$dXKs+f^3Qh8rx*KmnkN~#()eZ8>_Wr`8vM?KWgklwPl@wmBTj5!)$epLwJ#fI z$>gDesSA$j{0E(bE#cVjUbyDa%$v8SimLBR(RCBl$prsrqMZh&;)PkIkk|Fpy!635 z3gg51tmI&^()|?L^FI$Xz2`8Us9{g*9EQQ1>sN4Za26>P_$gM&cErQ-CLeCJ57xZ2#Pk7I z#7A%T1Ct6J7@lfRRJNaof6g_~bm%W=g&5+zlL^E*&5+LSG^Y>Mqp4}5;9$Ms%Y$n^ z@bS~P5}n5L-2a}??cA@3M#3BICQUFGf7Hf9Z;#-aidnc%VLD1r+lcZ<&2W|CWYjSy z$g3Za2*+1smgyS0`{e&9I`4;?zdw#w8kEu!rJ+4gLifB*Dx)YwQj{WOhaxjX8Yrcd zmWC*bN@?8lIw{JCj0lyi6rp^C$R6MK_b=#&dpq}>*X#LwJf`r31v~J0-zZ+9Ys}Lo zB9m=1!ClWY*~?Fp@nG>vp1Ljqb$ylj?QOB#u5t)p+2jU`Ef-+Mt=HIS@(33v*pXYk zaX9|HFt42+2L2wh5O*Y%H99tev)MRE&}d~U;}f`#>jqwI6VJ0OtodH`GKSST(9+$D zlk9TYDUDTpo#uCT<>~|&`Q{pwm8asHH4pHsycE@6{|oZfgV^-mHcavPgMP9xTz|L~ z?{_s|E9NGO;N`%oi5E<(M!T)3QfU)aH9s6%b<1^d)xGJuIB`x2zrQJzmpg4m57x)d zcBkMg%P!n)8vsLJ3!d|L%kf@zFj?=n8kY+!Pu+&MXcKRV)~PyN{#F1zHXs~FMfQ@9 zu4|ZW@?2Qw;EsOY9>l;m3ZmcX@p&eeJS%!ASF4ibDA>@5r6y|D0rwYBX(U+23 zH$E4;K0hSuR1ct~_i}i*_baw`b3S6_TFkk3Oz;d@VE>~`=Hsvq`^aB%d0rq^b_bwF zP&~YNFah8F*us;9{k&nX1LV}K;S+zXVELYBn+ajfy#Z>A9giu|<;Cw>B~z zwO3+k%SN_ygfW-W8A#`u=d(Lu$Af)0?$n#ovNA#flPQ<;wG715nadG@j^n~fU>;wJ51q6zC-@T2fP*BpBv zq-rB^(4{`sFs+1rxYy7AzA@t;PdB46eTY8Ve|U=aSe`Xmgu8^j$4mJ!7+l@~Yj-5G zv2{0am!%_-D)T@Ihs`{^Zvbz-ri2$>{}Uhm+Jl?6t|W@?(RlcwkRcxJ#k-b9;@7@y zxb~yK9UGa3MxTN}YpV=QPM=ABpHaa2jPLBlImdlDPlUMcX#F0kP3j_7&eYQ?Ae zb)-OH3iNmvk=h+ASm*Res0z@gty>1d{ugHWXwDF<`cFj^wP_gc@6u$CkH$04g=HA6 zH<|5wwwhUv6Y^#X4)|c_aMBQ^1UpLe*~0PHD!Ew*2Fq?nohZaxmmRrm*Fmhj?~Ub# zW4NTsF|77&ha`bh5D=lk|ICYnVH@m`fBFv_t6lMB&~AKB74ejfl*smPBJOzj44*&I zLB*PZFr?rNX_GTYAF&mm{csZY2N>h0Fj@X++imvqxeP6TBgJj!wxI6oE4Xd74xiDt ziG0z%DPC#zhLsj3;)q-Nd~tLpmI=8<$179tPm42|`Lm5RbS2`Zr`<%$I!FBd&PCBd z;|+M!Ru)5&8gRhz$2enk5oQ*CW9>RQxJ^SEM|Nl9rMI8it0AND&%QXOU-*xBPfUWx zzufWImEY)7S%f}uDcv9+XIn0@{oF3+1*8t_M!@A%1=A>5>(4fU=W6Hvdj|2t z?TNS_+HkSEEMGLSl)LLGiywB+!di!NTxTY9-)^XLiM|_n{C2arL4%1a;0&yu^MwVD z9Ko}2CTjl>dPk3RSzO9KvZ_JYr;gph553sVOSc&y?^wc-V{?rT$~HQt!7Gqj~eGWqPo`A;y(K9Rac#o~Jt8JoV&*E9=72c&kIGF_a*t6w2P?H^B7MaDB{6S8+dO23|`hJILM~EVb_&TyfnIo zd>hn=zMCgw+~;8!l=PC^>@Fdr)eD)zUT5~nKL;)M3}@f`U$GxA{E3IQ4n1rwji0J} z$drVYVgZzy7o2ssB;u5N(3I4^jqsmJ>^hqGm}c|-Y0*qnSxd4 zFSc4e2#eoOfis%|;o4st@zq_juwVT8CP zL!A6cDph=+z~6P;`=&9xi@yne61m`Jl_QqBOXz>ogc;tDnOwhg3tyY;%-0G$wu{my zal-xth?80f?(^kvYuIo;@P;g{+$2S_1iqWu0Db!L#2cJxqk;1W&7kMYn{f7IJ-TW@ z0yX@o!F>vbvXoJ$$4qCb++Kr*?So>a>D5C1=H}(!)UrJSd|Wc z+l`|w{Hay;ZJRZVZm}kTfgbSpEWI4J8^*2d7j|fbZ%Ud+N6h`eynC;bz20w_QkNc% zu(2k}Dr@k)XfSxa$%FnCe?VdK0nA%m#-p9|xYzM*sJbcKy9o@E8nc-i8xZ!C=IWfXfdGyEXr@+>uLQ`A$vv z)~HOij$A{lkelGGdK31Z9Yvj6V$eq`4vn@*!_+l83@&R5p^_#t%5zK2x}X>{|`tL)qzNv__z0Up~8r2glO zsQW~HDy{qtmA;#By*Ll-vcCiQp1qvSKF-6+6!B;~+5oBs2S>Lp)ODp)5OT(3dbU zGZ7vxze<`1{09YVb*ajU=OEgUg(+pX;I4reZomE=o_sQ)A@WK*t51e^O)eGPylGBe znD|4P?JL-F%7jk;;Kh@^H{t-_t@upmFmapy5gmNBx%Ev$9;7GBRgSB}?{Y&LkTn4E zSNMYLs3KUJu@Zhpn}WxFCoV@v&|7!M3Nut)x~0|$s+SA~^Y!0}#c&1T>@6TwXH-!2 zPz@0u9E|g0N5EB&Z@6W)8ogYk$)UM|-}3b02@mFh+k8LziEpFjncL{vBj$95js_$( zQ`#5a$oKfq!b5+8>GFl?f_rTqd|P}2c9j|P=<>rbuGbSx#0g9<=n2$)Orx`=I&#{& zm9GfB$KI*6z+^Lj%;+9M$JG_l^_znzTxo_c*GnM&yAhN&HiP7U3wLPIfTa_;y|OxAHqsUo1>Wqc5Dh*Pig=Bxz;TW!0<|1Z(&ztF z{8VcpT~FMwTJYTMS^SsW>+qo#LRYo#_!;Uw$&aqe7yh2UE)8}(Km%+j#%#;sG3|Lg z@pc&3v}r)K6Hcjc#{zBLbfeCgvkT$>5gTp;3xc{F4_fBsla)r-fR>(+D z3D5@ZtZ}e5PaEAfRza8kJ4knLhapoA!^)ngq^IX7KGtuC>&K6PUiS_B>y-mbPUzCd zVJqpQmr4*J_X>Y3_u*e;y@~D{W!hwIKx^M=P=ksN@NaU4y;VW*X}&j=t3FOUiw#Lx zrX!WM%Y%|t%Zb(DvnZ`=NrTc1>G+Ya;rS;ScFtxa?KFD0_Axc1J(MuD)`C^C1N`zg@_v722Q~ z?0{wc?O^;%kM2>J#TyPvQC%-XHpN7RWBeObknbgR_dLL@Pmv$5{{yd;3vj*cJGgE# zip~`Eg0#O2O>fDfKVCR6$(RD#=!OtGc{*+9nqVaT6V&}a!HG8~FhjZ*#IN-DTh(9U z55EM@d#M*#*W{7MNe1-j)F600KT>?M_sRqfhpQr+2LrJG-%+@?btlU4y>zNp0hHC$ z(F0F3scn%G4fQz8#^|*OjsZpL*Eo{QyLJGJZ@$OHFBL=^r|uJO_%$%}Y8ci}4(8+M z6@%I~ReHm08pISB(BxnV?pBtE`8|^$pt6+Q_FD?FR?#Spxma~T$OH`xhvI_@P!*|2 zEkZZK^5N%Tyf%aU&hrrb!H}+-vEQb!kWQEG#K9Z(5UZjmus=8peq<@o?xQ|% z=TjZ-X)FTu)9#RBTLt$<3SGg0r|7WwJMe8yFA1~o68z4aAa++X(j$>q7}H!w5O8-l3>b4dD}^$a<*T>2ZN@}S&`kUVw)+S=20(YgyOT~ zXohDiz4>Z4y{r_%4l8VcqP3;8wy=U4XBdlquhzp)wgbqFXu-$zX9(PVE%Zdxki=y* z5u=i1k~c#h4xFwcb>1f+zik}7Hb#=C`zs51z{BwC!e#s~D3(q8Q3PZ5RI_150%K~* zb|Hpi^x>#cTj?+_ohMs2oeacdzNcEuc{?q-2FDUBSlF~UU=yP?sy0TkUogKg+u z7(B|J)kPncPB%JKLm`q&jWj1~Im! z2-%|v&CLOZsqI5)q{h zMEmxmaK@s7IX5GV+a4@-A9)Je7T*-I%U8r7=H%Gqhx*~majjrAxrSU_9Zyc?Or`Bl z4$+s&=V?KB4()rbPh+KvVfE*&^k!x&DcmO9u^o;;?xCS{wIWtJTGULKB^Hr_z-%<% zT!5SWFEF{~TKLjKfr;|XH-%f zWw#6!e{9+SGw(FPht2} zyP&$KN|-V0Gl@T2iRbM!7+X~jqyF6|)K8hHwqm%}}y6>vGtyE8uo8FbUebNxosB@nFm*>yAR%Naq-cc|F9Mvk<7(cMI?-J)p6gjW0+ z7=bRk1n#QxAJNfFqQmXw#09G)(A+==$o?7jcpt@&{5?*X6YQ{YLt? zvXe~JJq9mA3gN{Med>HL70fj>Xr3Yx#aUmVUe}Cf-Wn;cthfaeZd_m|ejZ>Eo1d}r z`BoS?!x^`i1&}>T2J~sJEHt^bg0a$II!MC5a>)!uTqR@zKW2P{<#X55b$xfhY;Yr+ zczz00u8yTmpN)lg^=o+eFOI#OV}%3s9cawyW8~SrqoDhD4=hyCpjHc_;PBG@@H6-a z80;88qp3dDynSMKQ7&dch+&-&EG`DP$INN14 zo8aP({)(CS<7)w0=5~|**zcmwZH6Rqlsxfl+5+-30ix$UVv4Epuva8ab(&JC>aOMV zzKH=GV->jQx(VzT@_y_4uc6G0WXxb3*7-kYUb>%|W#d2yAEkneW(~vUwm9*dOAX={ zm!%l-<~&`u*^J(f&taby8j_Md7F2a$4z)&Cde*QLF1%5tO{ET)H}QbD=;a5Zx9}HO zwa7qNbw0H|Y0p#yXO=D8BeTmg;JTx`@LNBU>b{a7my^#B8G#dTxmp@^gx!6>aR_R+-Eu>1P|so`hOfYSxSwen>#?nbRy~uOw(BInr&j zlIfanvKSjJ$0fujP%`cantXMFDIXGuN2edCCpr*|kQ#DCJebb7CPQC^7DM5i5AgEM zLHHgZ%$l~0LYJ3Is6qQL5}o-Q%!Z0F_Uc@mU0X~R>l|d84y6*qo4aXjp(5RIWG-l3 zG2=>G27tr|Yr1}j79Ho5N4-{ zQA!6sz6;?7DPXl?D6KMUgS%_&Y3ukjvOy~WlnP!zxwjMFboL(p%Pb@Z&e@QQTL;n9 zfyqqjt}C6EU=9W==djOtYV>5?bMan!o#Al7d2qgot*zB!)9n9<^X-gb^jcTy{WB5P zh~h;n$M?Xez_m0WV+5VNX8>LDow9%1B;cO@Vieh3!(5Lgyu9Qu+m#cCnp^L&>i4-| zo1@Q$1-IbJE)KyJ+0-j04{YlMlemzpp8VmAz;!ZaDwXNjVCh2ZtxbuBbvg6+_MI8C zIpPGNpZ599H1?o42C{d!QSE>i@LMJnbay|-kd;~YA0XC%EdrEdN@Z3qrt7bc5spKKioCBV)z_bRwChQ}J9}7KChj%O~w3m#&K8$@A zGQM|iJ%C=5EmXo^$Z)MXjILkZNp-b?sB7>M5}WfFWeaOqM!Ez1ZSsYEyZ6A9TrXnx z#|h|?z1S#Uf*WpLrURNC=-KobQtiBh2eyV$ue?e)n&U;<=k2G>etl5qy_A}=1XOV- zfVXPrV8z%$(5vZ5ZEHDe+A@tQjB=wgr_@km8xg$98R7>&9P#ejv!bkDThV=-Gt&y$ zOe0iI0ZZ&5=X=MJzm~e}tux}IH1b2CeC!8BKL-4tJ zk6W7k<)3Vt>8x8^dFv9Itk|lXUvjg zQ)y;j4HUZ$rl+j-64k?#n6c*kTVI&+)VOXG@vY49;dmDsSI}?ydNcxaWl%yIR~NI`>+N z9+1ouIL~2JdEZ*P-n|C$WP;$?i4pXVp&?x+HxvesF~q4J+PEq}m+c9;FLr+T0+Jlx zLaKH#EXt6=9`u2f-?6ag93;nqhp;iur!Hu|HNQp;&O z;Cam^YIC6!;#|65+x-~ch(B(qn?K9yNA$&lH=*$0p?(98$deT0nb+Lp?A-h(9p~4=o!t!kR&&knS3-8 zSlqU-bfq5_6uiSzpOo-s;dqR3%Ykb{C)4f6+Q=HMClK>&FCLj2f|nx9V1kjhNNs*O z#^1P!Y44BYGJ)^nRDBSJ?Y&5+%+8|H26@!gs1u$v&!YdePk{XK8=0x<08AX4!c0!f zh*hEwu<2@-n0HYHNQTd&4w~EPKWi7-GeVYX<~;$uhypO(;7Sj;xzeG!3+QCi97yqc zMQn_Iz)~Ss>h89hhMrmv|I$y0;>|T_-PI^MEa5YRUJJsnS|{mSRZa1i$VbJ_`WDPV`UmW_TL3>!KY?6% zHLy~?DVC^k=ABka-1ws;?_;VwI>eB_e?Ewl(rGXzeo^H}Rwf=(*TPgz`HA=4EFtoC z6RBR3E6hD=4DHehI3@i$#Kff18u1GHcj^Uro)>@*cCW(ad@DtPUEFD=zmRP>Fmw&u<6ttvGcGTrg~^3%YC!}qL$cT$O$vvuIG&6 z`%$F8xm=h%T^Cqs$P-VW;umgh=GMNAm=T-;vlB+~xr+C=?ruxI^+E~^o4Jn_#f6~E zut?Uqa2kebJKzH0J9g;5CzS(&UqY0P8V-Bb%cK+qenyNj4-FDJizD*T_*X5gjh;r+ z>)zu$XVtX0B z;n-FDu|wc(#qK9-9e!1=jSpvEZp5NvTo_S{+yKfJPw~v7Y{B_;13#QAMJLB6uw>yC z;d}TH|7>7{p##k1xaR@&f-GyuVli5Q{M^y8%pnL`kvyt5`a0{FN;$n+vmOU*T z*8jJQrml*iuZkCvA1#fX&+@Li|LO_eIJyXwUt9$Vze>?R+Z;AlX$qgyuSuG$J?Knf zuX<(O7?yg+4tqmavW-3SVCm_6*4+M#?6-bNn&i9$7Rg+2D@$ZL&*GqSTr~NpSOD&; zMuSqvM)G1xB`7prfTMc3^wUZg`t$Np9-zF7KV3eUC$2Ij7C|-4Hd7tP96g2+qYr_` z7+dmTO+L(bNdaT`1GurVAGeo=VP8rB)g4v`5>{*A)YoVTQH`+aHmPNkG`EZ9q*-FY zccEj*_K<h?@5&N-cCiVA@29J=z@aV!FOwf?w0ef%ZyvSsc+GaI6 zJGT=KH%C&Hj~b}cBjVo|l%V_KEi~+p;5Oct&7-}-xV`m!m@=XV5+z1L^Y$ZfHc16d z-B$3X@Mm~LaxU&0>Bz2rv8JNU$#~8A2!5G41Y3R;g8!Reo{5k)zyH+RZo|PtuXBgkPT5u~RJw+2tmWh(nEYQ7+qZHUFBx?c zdrv>doXq{O^^YNJku-*N*4>bBFaws#PR8($GuUn8ZRo<&Ri+ zf`MqCvpy+*c}Zk?;uyp3A~NlKGTkxAg+>+ag3!JP&}#NxjK*Kci24T64+TH=X|o*P zQYz1tQyZ}Mz&Y}%B#f@GcuY3>nS<}=xj5z4FSt}$11rDmA*$`&?4{02wrJ99kQOpo za;3^VuPB4~-p%7Wbz!{ytUFJarzHQe6tLtRNU(FkMS4a!C`|(`lqbTD`(m;oE}IYnPGD_Pz$SV%OUFN6lk$z1WP_Mwn1N{)?$m{*(kno&Rg7nWG6}l2pziv z0T?f2jRMBBvCC^lL7v<^8s+(xpw1|6J+%i%JsQln*iWTqAtP;#CZ|#9vUcnU7V(e+ za=hXOp*8!SvclEb>`urS^p>c=o|^x6KuYMy)b-SQ!3FHOq|0NzP2*zkOYruwHA!vp zN9ExsF>d@YY}_A7R!koRFQ$HD$EBa(#Tl*G`c>E+t=%qgMYf{E>5rK8a5K-9sl}mJ z6v`iah|xvOjL)}hgh1EXbiuFlbnEF0RH-i#t(4cIiT<04j^lG#ef33rn%sr9HokmZ zj<7RM+eHt!1PXjmC$85SjyG;5u=V985O^X7BVI)k&9UL&_A~${eLsd4jUBB0%zgHA z>Lu=aub#g<(S=XGCE^_ThIwn_u}{sKMon!OYspLVz7s9zdr*PyRvb#Kg1)nlzb@lJ zS6!^l)kQC(Or{xm1HAn|!4M3m8~zQXjw`BAPkkB7EtTRA*68u0F6n&8aU+_ySD2;z z{(%Y21!AYgQmjEQ6U*<7hrrkFl_9Cw_+(58^^*~JbnSZl*X;p(m}(V13Jb%VuTvmA zTOZE~rJ=NGyddd;5$f96P&V?$J-W!woOwW6N6 zjvPQoPl?6U)8lF7-e*X6kH8{>Kvt|s=&29(Xk)dNSDuXGE^_Mp(gG2cx~rb#wxkKiknow)AbUe;oL3p@;h;EeM^YUCOPmmGxs>K+-` znwkIq>?Kvcds{s6dL8q~9f5N$hT*&!8R&AgZGzsyJ?z&gYtdxq)1vyo60AR9g>%}5 zvup`t)Nk0&=KRQFv*voBafiU^#dIRtbp<4cK7ov&Az*%J2v)4sM-QS-S&0*!DBKxO z3iA`--EF`tq%3KJ&VO1^oL8Ir{TUCdxg1f&*&8DT9=iv@)<+u z9rgmA-neNq^pqCXPxnJ*`>#y;`&3X{KY<=zT0#%D`qB~StDvIy4Jb%O(c%B3X;u7d zFdr%n1B8yC{$mT!TDTM>Q+4s#xg6BnFqHnZ{s=|SXTZ-vKjBb!CF}Veh11tl{3^`a zUwU%(Nm`nQuQjG|osZe7tU2hu57w z6SnF>i-8fhs4_wSW%k@=a( z=$GL~;J-sIw5=xsLEH;2sAI%8;ZOe4muAh!^z?mE^9wRmA75qu_{&uW+8LQ=h_Bzi)+|ZA0no!Z}zsk3)Z>B);;@psIuB(mdTLvi`d{@44`p`COEt&-HuRDJ@@O zAvg^sjTG6pG1<&yur?#}Ran)LBWz%K3cEV-O~tfL)*}1+4;d{|A#y_R=j-W% zo1izxQ0C_{qF7f8N$V~_W?UQ{?XZOgH54$VQ5meOuNBciLNxG`CtokMng8iZ17)TN z)|v5i^v7LvP^%X`{c9clyXqmlDr{lLjCJU6M}1O%>@G2zYL6M_^4xjjWV$wA=s=lv zll)I_%IB^*%mM@Nl6v=I2-?~Swo3ZoAz#cUhBuNWi~fmK#>cRtvDIY!6j^%gPum10 zp98g`(OB7+fCn3wvsZ_%veyzj;Mhie*c#zPLI>8u=b)|le)f7iAvYD@eb>UTuZHl1 z7lZKfH)9fAdjO7h*iJ|)V|{Ryq9(JA#&130F;-s_$(kmwO$qc@+xGn_z(f7)Qo@ z=@boVO#s=K+v(q}@ua3C6K0u4pkBgZyqji4iaJN)MDm+u4qeY8Du-j4Ojw2M$)zwp z)ei@JxJx`8`dRJQwJ7yS25$5!;eRdfv3avGe5uf4%Qf?`JS&b|lDSXrMQVeSgf(uf z)a4IMYp^m`m`B)zv7TS*e35(+_Esk0w#^>=h+Qz(czl~Yb5De6gAcIL&`<16zJeQH z(*-Z{4A}77l@1o>%sT{+O6Yz~Zf9=J*KbqdlV;Umnuap#d4B`P+6D4qt@FsP#|HqW z=+pg+Ux*U(O~h631*J7s_uBFAHH{aqsQgwXiiSmMi#LcZb5<@OlZ+zHKdGvK59 z4T#j9O9#JB2GzRNXsB`k$}KgaX@S7kkSnGQi*sPiu?FJE;=ty=Pq6ORZ2XY11*iTC zrK5Fc^pA^%YK%QMW9^}?s8#zVuzc=&8% z#2PPm6R+p~6v9!baR14AU~N3SQ1x zLT5)TGHx9uY8kE10v?PK{qQxV|7a$rwiTehsS#IO^cBx#Cg6Z@L*YIm5}*2X0gHdm z!N_a1O!@Z?o{!F*J9t3$ar)}MJuRKJ23%ss!?}(LM#@HRr)sq94B&N@CTW zDg1chS^QX!B(q1F=dDrYGw=QqVUapt@YacsU*^Hjx%cwIGBNjW8HC5qycdaN-{Ygh znp~}bpk~t{}PHyfu;dxJHHR1RL|l9hA!^ zTk!{XGH}_VS|mf&`CE%TAfq1>-}HPiw6q}^e|KW`rIo@*WyXnDmB@Rcuds*L!+ZVj z!Dy#4+^qi%BMy4=N;MI_d}cwf#uJy=$EbWOlUs%t;n6Yu zOjVfw>^U7rs)j_;$16SP{U7^y&%;&t;M_DGliA6(`<;Z??P+|`hWEIktD2+=P8W;% z&nSBKn;a}ngB_n;;7-e3=2C7A-xicGjrmbjz2E@cnezlU?J6VLAN2s`$BW{o*AfGx z6js#olUA}+&9UHOWB2se$L#<{ifc+sn^;>eS&}~@-2834})!@Sv;n*9_I&1QeBBmjF&FppLed~ABFSyJ~+bVG1X+9 zR4k}X*hgCzY@@kd^YFN07<20zg*G?8Fjuz?uvO@~_{KJ3%y|Po>X;<&oV}7yW(-q< zWr^a)OtR?l1|S#rkP8=@nfl@#sP!g-OpjFHmb1>lMb{(PT4BxoWTtb+n>83QQ=QJ3 zIh@K$OrVYB@2JyZb9m*X%JWYZg6^~oy42N}tv@z~-dX($m48QbTSK9XpViDlmUvN_ zb!8a))dDBJKFV&sn{M-bqE7=9qgSo72+H=K+jJbyz9A=RjF>R(E4Ho z@4lGBq1NJxuh&&z+^FR=n^uVGMk`SHxEzSQYCxqnO%RV1_67YbItrVX*_RpzbfP~1V?T8MRfPdf^F??RHJ$o zZ$8w5tHN@ydS4!1{5TqqF7B(;t2qlR9O}T)<%3AKcPY91b}*mN=EfT&bodF+SmEr7 z;u`iVacA;0JQq-gSI?WF#G@=c)ZoLMpQNC#qvr!HEnse zkG>V&D>wZW=*Mm;T9WaK-1@N!K406wmK~+|<8~?z8cxt;btR^FJ;(T)qlwJ@$v8Fj zAnwgN2M?s?gO0Wn<~?j8=gz(cW5ao@+02{1H-5p5^dk7M5*hrrB^qaU_rkx;Juv-x zEGv7^PNMEzh8f)*#lU)~zqm}Hr_1A7p9}-Fib!-47gIq}6(1!npTHvjzw^(P$ zaOy&ox$MOjk=;RpceZWgQIB$Y`~pwjW}-;Gp7#++7l%P|Hx_Dl>kR? zkmPX_`OtUkdF8l5_~s{AJh+8^OQKlkvIm&wyqPDv+Hr0B zD#+7Jp~Ftb5%2f==)l|YB*QU@kMh-pcH!RBP;NuFXI+CWjYIh3+8Nxh?-69L41`3} zkFX{7H0Dn<SY98-%R`_vmEXFzCb|eHnMnm zIA%Ny=LY{R$JcL6`Ox?KV7q%N`ls*ZBih8GHStIIro3}0&u18?;R81L~MwWsEDiF>!n^>1Z3@XZHY8)(ecPOpUu1#SLrlnNc2`2p9> z7y54ot7*fba_sDwg_Vmu_?3O;+@SXcX5N;pT2`FMb@#RLi>r6=N#-VkUveLtn{oxW z-#deg?VjS(!o~RJyfb(HNBQ%T2jEeh%OXWEE-s zb!(n9dKMq-HvrFjAA}c2r}IyqPoY>sk!IA*p(k>KXzb1os?~h}D<2Kwue#LuK-Ca_ zdE?-!lS_^H*&az=`uHB{9=xH~FSb4m;<%=j`mIRDiszcY_suJHR+z8XsUz5;xi+Pxn96X$# z2JL5J>HMjOFv7u5tn#sn^l^^EU%~s!cx7clGiUd@NEwU zT2F?LQzp@+;Zd~8(wq7`tYGak?%^8m!RUGa1^&F`$5uP6qw#-yXt?0{*>x>Uq;)c! ze*LUy+pKFsQ)P})(Trp&9~KWi(*vRL+DfWZv6`-XH=Wuq4}`%LS4H7E4`BM_PSW#W zAKg$pmadvEoGlM@XspvvTh-*F^j^kUx>9{AZTd4DHmc2`_k_I7@2k~#B2BW=*%{yZF?_K zLr-I>y7M-xk^L-knW#rQ&qRxrlfJ`t(E~`T6vMe?cC_o;kN+`r9)2}{VH|IlmQqnF zEvu!Z?s-mxj7V82LL!?a#1A29N|Mr$q>@5PL*4V7WG5k8r8JNhvNJ0DzW+h5`|93v z&hvae?>COf$rT@e5{F|$n&G_Lck#t5PvG6~97yRa#g&@E`#(iSB%EfnS3i`%h((r&K_9-z3>~B9iVs9CK%Kk&a z4Er5tM$o7g3+SPn(e&siFPiz}Dw(mp144#9g@Dq7WYs4d`gRh~n6)W(+dXHK{=EYs zM^%?z9iLAB<=qhV-H@QqE3BdNpAXD=cMME68c~Bg7SwH`Gw6xau((b*1LMBa@E_mk zio92}{q!BU6O<1#vg6@{)Z5bO!{ec@Mus{(5ZGCjC&Aoxg}vDfRr>%_V!!>6$bQ_G z#k80wfde<@-{b|Kv2GDpcDaOFh2P1NAqPdaQx)inioNXb1shyq_7|2Mm*)vL5mp8! zLHMshbVkA$sCytor9(5(_fZBeNVvh)Jx~J^A=EgvKZu0CJ4SZh)1X7R5qak*e6AHw z@zYNy(H*lm>5PM>70_8D4!|7eJUTBWiqtDy z;w4u@xQoY4Vyfp!qRZyO!CS{+b3!;K#|@zWDhlxCfduN%ccRb9aO2g65N+Z z+N2vcDYZzteF1{z0f{SU|3eyU4azmejHK0$sQ2 zHC5;hrf)h!p!vW>`goBHbw00XUs;t%Lz1R}$L&t=*q}tNRyhI*-9TgLGAd!*4v_qX zRJvN)zt+*G;j4prrrR3&uks0#)QyE@{e|$KLznnu!aSUMRp`w>O6B^)EcpA9^Ps+X zA9Sp&hk5U{$O|nWqIq-*PMe!bx(YNQY4cLPe{($C?VSWY_XQqC@nrnI?lNwae2Z^? zMZn6B-^icxowk?5;$YFfkCff$rDv@CsMGEo8VAFuwbw)2`b}LB&^`o4_UAHJ)#t3~ zmI1sTphJZjEDftFhwC%dY5x0PkRm?4vEVt#IZVfi zI-R8NR4NOdF&#H+hGEArW1hQ6ojw~ifch0l(e<(Cq1afLdYI^f;lDEQTRw;SsRh%% z2gm4yC~5mkEdg}yxI0kfD~%VGYoTwRDXp0-Ll-A?0()-nva_lB zQa2jbcNCua{uU<>m_p|qE~D}$onSH{j`P<#@{`mYROvy#h!Y;mGr4-SPG8zHBAZ$0&`PXNsw z!)SZPI_gw7fi9IxV!C%d;C#qmvgwfp-Ci9>YmfhcqdqTT+{B~o=+;1M4{wD}BP6LR-eAKOI$6BlRh&2WCfS~3z#|2o${AfmVY~oA z^9yL#+C=(hSqjMB_o9jR8vH8%f-iH;xNYYTCYd&nwj{McSKd-CF?A!J4j4=~KiW^d zI?j;Xuyvp}YqWh}<6By@<075D`wh(uRHpW0o8XIZUc6i{EAS-tQtKTlbVh`n{f^>F zI(f`=oV!+s1e~-1^_*|yeu^}WSJS{IY5?-?7uC9e(Hd6Ot_`?*4&R+^`kMm)Q?R^}Z_fl{Vj>12^(PC@6OtDN- ztGMJaFoXPusB%peTknOT#dUx2*Lg0OJ~#xkE!5F7@Dg~qo58*N7IcP|0ex^)gG#Ck zy%4o+7#F!6Ys#!y-I|GHipO

@+v-oV=J9w%Bt!ZGAS&Z4=xYc^qdpt6+=QXtE}M zKbY-aPTJ;D*j1s!sgF4ddP>=KURp`ClNT|XAaoEQEc?O=GHt;AiA{{#!#&Zd{i+hRd7z&VVxx28xL{sFdh0O!Jq02^I#P2!=4Vd)L&MT$IV}ld&yx`Sa1up zV?RLnqB+dDZVPwmT!Th0)x^tO6T}&J<+05r2LCg>LKNG5ahR7n_fIRt=n;ja;f@5p zx3{A{_vYg3y~YS-2XMZ-(64UFK;@({e86i5tUp^#h}e%*RJMz!D(GR2+EdnI-pZZ^ zc|woYE<3A_vJmQeOLSqo3w*G8WScf|8JK=L$Se;f;+genY;QmiR=!i_O0EZZqnihJ zPuYxH`c9*oVIgSgtFW`7S#a@rG_>-IOl^;~FjM%+I@%)9yEqRwjR`NcIyn@=ADtyz zY~PBUP8YF>pGLz=qQb+D4Z_i(<>cGTy-u)sEQ)0;$_WY*(hgM*@-4i8AR)HgQ8a6GQQ?uPPt0|fSX44BuX;_sdY zY&biZv%|XZd3G{n{Z;@C{0$1;a@17Kh~B9m&ihO=xOtQ$|9MspoAT0el!H1Abn!>? zz(|l0I(vf$v+_392hOH!EE1fv}-pR#o-vh(v2bC@FgUMs0)i8Epx#|Q zemGMa^s7fuU6VYTRqsk2>yL8LjX`{Y?N1!vJ&a2`v|%}UhdpbqvX!05rIuQM$KMcfgnl)++v`Ekf{mh*Escodmf}doNGyn#ArGZqk#z^Rf%+mD zNO}E~yty0={cGjH@oo+bwG|Ul#%*|-*9q+dok(o@1eWD5i}fD%yz9?#W^qxIi9{Fh zn&KIh%vpu@kCoXUxAWj0F&?z<%F-&(r~P|fY0&MZ)X&BeZnc~u+X_y@3ky>cJ}MZd zEIUQoM|k00^))bkST7r_H-^QYegQYj-jmuMbvBR_+~21xt~OXLn(Ux07HV+m9o0uKNRx{|{R~ycxw_ zS1`Ti9sIqpK~Nu0;mI*u__szWzVdP=bEvpNda8UeenvDlri|voSPoy7D1pjmF6w++ zDjsVP4cDZ*;fP@|(LeVdO6qfn(dqfPYCs&}y*k9eZ5o^IJ0HzG#O#cNBHtoEjlbN} zhp+Vnuk4u<`1JcF>?tY5<(roiZRaR>HFXqZu@9_%|5GySrXBm_mW;ad!q92nWp*T{ z3cemVjkCwM;<&IT*yOK6N7^?FU2=7(NOwco6K60ccO)O0a}*|6BoL_=YC>N#6`uUD zz%79>SfkWR0@tpG#cSS@FY8iRL-|E^(lHl*T)K%)6$er2VH~-~9mN-e97XKV4+!@@ zNA|p$2}d&I=)~|kk=@t3Y+yzmkixTKmtR}wd1NbhWj&3=oo2E>V9&%7(nS!W_xW`*U+QR413#tknka=FuZjZS(0pw?**1LtfHa^0>9#iq?l3Bui-kzR(phdUrNCAr-kDzmpCLNKbNph|g!SRMxW_3o$ z=M_p(FXOY2v71B6vForRBnmz{OY+z_7tp#^3&FPfd{1Bw2?;W1A4PGHpEI5ey!o75 z{qF@@%a>x)k}B5v{;4=dRp56nnt}mG!XTn3SAfe2yXqg~$%mI^WY({FP$v<>Zgz}f z8N1?H(n)We*)kZX%JhlHd?-dkbpU<&AB^iTY8u!z0y8qc<^M{#ri8K&n^CsuP0!PqWoK0w(FC0^bHqe55ct+0c^ zzV*0gN|I1|-Yo3+C8*M$e&W8Q5kFg>$D18<;Q5A7Ob@-m_<>0AaocNTsh3yk@IaLXAh zvt?jyVH6DN3d2(~q#;(O7A^?f^1;C>bm)e|SUq7LF5jy~vorKCMrR$~-Jl{?Ts#~{ zjnU!@t?E(f>=V5I^#-m#^BVW)Ey1xR-$m}5^{94Z0>7TJn>VU!(bI`f#ebe_7 zxr@N&n`Y&QLtRGl9o}+0eMTj6Q81WF*0B@CI(Fp=@od(kM0)?!Z8TUnnNi_v?OXfK zZpYMfSh`Y?r=0mLk}Zkn8MT5V*<>TB5@uX`kaqZH)^yXm{_F<4P?7gkP|fF&jCi8ma>^?`l(E!~>G5xNBbz0<)5 z9q(bPHy0Jr}z z2*0SA;~hsCE)jT-t$PyHfMmq%QJ!GMjaSo5dxrHFETI^sH!t<8a^w?83l!gBoi)QfwyZ{R^m zGq_7#FV4AbivgL(L`Fl^@Z;){P{%NySuCM@AH@NibP$Z zi_*ob%ntLwv8!Ja)$L7j|AYNx)ZKi{6|yKZ zDujpQ@6BkQa|JJl9l*o&Z_u~!8ZP`iM(FdZ5n1aPI&JS9nAwoVhYgga-$I0b>bA=` zMTLX0og5o^e>OMzNO8@4M!)5qVN-tU;h%y-bk^eAR7Ex&g55M|g0C-Voa$#k=dHl~ zcbl=ndzrZL>jBJ~{)zoPUoUoh^@ce7xKVniVJuD`IZk|Ny3oC@dxk&nhVUbe>F_CK zF14~Y;m!>!LxPy2oR~kAL+mD-5|0xejPHu~(l^g2ew(||~fxQ6B5d=VYERo^NQR`-cL?u zA=f(Cn0p=M)K*{dL-7<*roSn(o0*79)C!<@N(@FDuI1}ya(ri5DeifE92H+&!uXM4 zDAFB?!6Qn<_Vcr_z@!UR&t&1regDF>rvnNGBZrRiWzeAO4An`(;JOy zRB28bb9?!TExJ4fE0R8;+Z$K9aBUQR)5#Yynm%~r%MrUHUvj{5&IM2$5&{L~>d-AH zOrn+;aF0WGaWWi-Z$rZQqOilPrRWGcSUzPN^gn~%lPB zMxD-{BuBp`qW)*XyC0cj=d?)Fj6Fgldh;+yHJT4wxtVMIZN>UmmGDq5f*xBJfW9(G z@Nrr+6lr&&nsXrknW(|N&W-?u`Szs&HrGo<=??r;!4BT2aR}eo525chcA><&GKt%v^xb+>GU0hACQC&_;P7~?NxV!nM-QiiH1D#R)dg(i;wF6meJ!!DF2u{V z_M&I!%5h@o$7A1}~=e&(x8XW~Z< zW8Qbsk58T;!OQ2)<#$$Q@kYxlJcjP(eS@Ci$ZI})ph71%9j%Uj<2GZ@w&Qr}SU6!82ttkiiT^5IHnl}HA4w!h5KJ*myncf9hI;|f1!p%`mEeZ#p9m`*<{N{UZ4dvGs9D}hN zBDrpt18TVCq0E8zIQX?Z%xx{g*PkY%~`_0PP!0I}bS^HOvXCZ7k5#n&yz&{x=pDIU#cnI=Z~L4OMIsM`v8^Avb%q%<+O_fPby zFO=?xzYH&RjHeQd9Bod(Rp%a)J>0!`nX*{s%B}DFaM#-9?Y>~+ycBZb7NCwxk*&A$e zzt?%(c*K-SY$#wP%nTQM>*D?en{Z*)RAlB2tZ6L zV+P=nPjN7!sYGltK^5w}HE6y~Fn;%9&>WtE{VSq*;{5SU@7*|fdc+?!0+VTgnF5vl zrNr$|dGL3!O=xFs&)fHF@%NjqVT^x1+HO^%5B%h@|NbPp*Cm>c=rW;ESCd)Ih5Hyd zZ994FVS-))1OMB{9dz$jIXd@OAT{0QK(E}==3Bp*;(-{T8<&5Bv1ZLQWSxQi0{Rvo zP9|urDde;_N5ay!;nXVO9L!m5O+D2HVbUpC+%EAQ`W&-TTv>dh^p{X+xX&i9~sz86i9u>sqYI(+uEPF!l8 z2>m%LP|2VZBG1pEW`hS{meE6|=jleQ|Fz=2k!zWY)E!J*T_YahTStsPDPW~TB(YS< z1CIgLnA&TA(|W?iIh85saY>k2-FpX}kJYKzV>_OD@DU&FnTn~0VgznqCuqJ`rVU@k zSbC^bET?CIuNx$J#jX7;XwXRP$jW50dn(waxAJs6ci{)>75INU2J&YsLWCTCJb(2n znyWVK$DFPOFg4;nT}cnJliNRw@|JvLlP1dxuC8di&PI34?4Qb%*ji?CYXC%-jfRQe zhf}v`O+4S5%7Z>1f_ItU=tA-ne(D-hi79VEv2q$jS^pJ#$yn2;Ccogg;1zHVEWz1} zm(zTO43TVUAKRf2O^(X$#Iel@xMs{ttf_9q3*RPTf!YxMzN?Vg4^o1^PpirCmvN;( zrpRJ+$Z^OLaxL$6J|S+C+VD*|V~&0*q&>%*l5cl__?n_yx@^DY;HOO1JlNW%X1~_t+W-g zCk*BIGiIRi^s zjBCD3;0xp2$@WV(uv(bCTz1L8F99x?<&wphhU-)B{koL=zJZGjN8m>i!Zazt@uF?F4i5iZsOCx2M1o<8kzaz5(?#yDED8 zM*@RCDY1p3?eHn%9J#-{h45HeW~X|IsTH<}^st*5iE>1KtekwsVrIUML6+<-WG$Ta z_qF3rJvHo4IgOeFlOgr5Hho$x0jIc~xIjwSr=8cu4Q8s`HXsAX6~|-d{6y>qyjmY?YFGZ=t6-nvL#Te=x3Z6Qlkg;_dyD)eaI>78%h} zX&2ruxsz*3S%I{s8Wb6KgU?)D`fia89rdmS-kz&~TVcB~_RMM2d1lMk_?^HDXJ2Da znLi%xmf_pqOyU>YE8{y*tB6?P_CjOyffI zE?kS%fxWE#%r>b0u8R-+;z4HYQL_2&1$e$wm?e!mjc9jzLa*Tp)~RkLRlB^= zHt;4sCZ*u1w4Uz`Z^kjq4_&+cdFZxbs8*ST3%euvI!9^z@Ijex+FpcLR*WYHzx{xx zJCmX7s4}zTC_Sujht__KOZhT+=2ew%K zBTm-K5*v^B1ji(|@mV)4u=?#ZZWJ_8FEja8Re7eNkSCC7(`^xaRR>(ko|jcjgze()%W+jabAzmw&_~ zADnn`%pTYwxI;?j52gFH0&wE4?$Y}KLt)2YAB-z9fSsAb4755Lho#?znl~HxTXhwh ze*Gpq+82Tm3+KTVe|1{=&yCJhG~@pTmH!BC*33;82<;h(IV%>9xxK9&o_fjAJl z%Qe|WQ)i}=_COS;EsM$X)cGBWH>7IZLs)WBmaWj)ginqpu!DX!^uSM98lF5Lz9-4i@__ z!`HQU@XW*`Ah$3X9S@&`dmpvXSy7kfT)G5@!(Dk$U@L~Y4W};*+u&8n5%O$}JdfJ( zQs|!CN2T=#@S$cL9Gj8^-`1+qpqmO*QNx=q9->Msot?3+#)mI@pTX26i{V$vN#^wT z8oGy=VVit2ZaBUWefQ5MtH;mct27>S+h2zGN$_a?E;1L*RecD{8V`WvnBPF8Ir!Wg zOD|*&ri~xBK%$2(f4#jK*aQdqJ4Txy&`{)G%UsB}%CllW-Dxz}J+kzh)Kz9U*N7Co z?_sYxmNWg;qo{L&B7L!QC{4Z4kK5y;SbB|^HG02*;q!i#b{{E#9p;}vxzh^Q-5CYX zW%uCC&Rx7|gD#crHlh_uFL3o>hDA;qyx*e>4{NSvwkNdUT~azP`LcwUt~@~#qMPZl z(+M;w^e&qwFm_~iEnu%_+#}oNgxBPVD9kZS!n_m{qCIT|IKJP82HRrsW#|GvZcZ6I zTx)GRV6Y;yD4hYvMh%4zQ{74?oA%?Gx!ppiA&VV3Dncn;!O0Dolr~Y`jRG9@gR5Hn>F&6ySV1&y2yHNW%8ZrwDSZb~;E*W_q z63qoKW7`pKKk*^8i)*K|0*z_Wpku@*o#Ls7!{CGeR=V$m(0x6m0hhccVu|l~k)q-$ z7SJ}7&$fu*neTRC#m=ixG+ubuEBt}4XDFET58z*{X7j?>CHxAvAWhOcM9*Vom<{nI z=bN;dU1lzE$^tuyEk%NNTuUS^{SU17d%^O%sc`TyL7z1y{9dj)Pk|}a>T)s|nvTN7 zKaKcP;|d(^U`iePrRldCEgpRJ3)+P_@JI7}aBkFPXmnBMH8yL|*FF&D^|#|i$?0g( zmfCnRrRDN{P=$3~tA z#aTo4;^mTfG%?_?Qk((d($^?JDTqH?-7kc9h_QF@vTT zO+@{?D|QM0``2at$&D3K%=W=kIM*Le4>;sf$qCoN@zr;`*WC-@@EU21wKpKq*@hHm z31u!1Aqljn$=Vm`>R^oMvMJ%2b+a->L$|5*a>YI^{Xa2QXalcl=Z9 zg~9Vw=(^y6c&%9n6O7-&x+_X_Ar_G#P1$f}&rFscX%0Q7rFq&&N!~N3mCoF?ik@!} zQL9ycX#chmpJwZ`SC$9at59=Z=UU5Z?yBSOW0!GH&{Tf&Su&=#?PT2>h52~oLLTZd z9J?NGV%q7mSsuA6+V`!NX*_wt3`gH2XDY36&b`anSz-=TWa`POz(u6t)JvvW*3Ui< zOa!ml7g?<46MQ}}1U+ruld!XgMPr;7(_@~^@IK}?D^N7&n__g}>fY&m)9oaF_(vJP z#5lh@*+ijoanN_vb22Qa@ys1QXE-(8V~!caLY9d*zdG(kQkK+J>?=mTUD-c`JqTz zZy}63D}`Tvx$+JPCma`0M3R}aN{xaEQfR^+yU$~r4ZlrkoM$CRPo z^p3FeK12idC1cph7h;u~WGp(F%7&mOKV9}1-+U{;M|ZyB+t-G;BqI`U-52(`IW@3L zb23P8I}U%n+Oc-tO7{D9DA_*xCM-!-PRk9jcTgY&pw;}N{vVb-pIRNbkYzkXK{$3i*zlQDSL!RqU?L+54JRlX8 zcvs-%`i10cd0I)z9d@wbqjG+)GUE zzjpMy+$okVwWa2_G;!ohe=wb_LZ?q&1Z+|We7Lv+cD(#cX69_5wmVNy&yTURD*ptv zTe^?-9-2ZGHQiy)qj#mF6`IHbrAXL$SW5W)`Alp)Iz&!NiTFNf9WI@F9E&CI6Ag)K zQehkp``B^K@gO%0t%v}H@(`@s+yRawY_O)`0&d%}j=r(~ zkB*h-fniy*VBi}&Sl1x%KbGd8?NTY4FgH}-C8*Gt&q26U!i(FdpTojM&ZO$)IT-4> z9pf&?vQJBUM5os*B|8rX)7J(EaZ`^!ei*66>?;4-6)J>5als=}t=LGWiI;($+Eb!D zZw75lRH1d>r_*=qX3^g}gg)JWDOh(n52Kn5IKP??*7FY1Zv)ftQ(YLTzbCv|8(zae z3*oKsy9cD^H<7zlYGCo|Gz`z!Lf>WBQG+Fo(A^PBr}|XUiWW!vkn>Myfw>nw?UF`s zTq&V(vr6HkToKfqlBI@wx6-IZCU&DnZp6hnicE6Sg6nef&{6$|>G-&_g=cMr?}al| zf0Ki8n?UruDUC$$pO3Q*WCYgQaJ=fbm^e)J5$+U8%prXzI<_}4GygVXH~kj8>^=j= zKL_Aqt3($%A1?+G$lXEW(LTq%w|^1jsa ziw>!^yUvo|U4x`+AuPM;9Ny`WUN_H+N4P0_+j(82cvy)>MV*25wesYC!FiGp)kiAF{w2}6 zspQXeIn>`?04}S)kPfE={4BxnM*9n9e2Oz|X=A(j&#H)t*-*Okc?CRw6;67HFi&tf zF1XrrU|I5bK7Mu+EL^5RXC+2p%Em)1z`C4$yT5~MyRjU!B|k&o@)q)XyE*%i@tOo@ z*RdIjhuKQ?3(Q<#**s{Ig}{@cu=unlk@}C5XKIDeYu3&tP8bN1QDVT`H{sw%SNwPC zJoH+RW)9gY#NBiZ{YkWF*2`fu*6$cmyrYC>ODB?jxn*pB;0+}4eW+)n>DVAE8wQS&tNZB{h=sTfRK?=vXhK7fZ`lAvyx!Jzpe7!(Arrq!4Xxbm}ARBWEb zUb$5;*(Z|pqLv)pbNLwkx=6ylQ|S0bT4}&l7^X~CJF4PZdo8ljFcor6m9W&E)iAa#T6AJSD(Qa21&00w zQk;5{NopH_j|zc(WCbgAh_P)pS;c&)!2PonSZ1XwNQ$j7U-R8A{>9eddqqRp6rYb3#rMs7}A{NONzcJm;6ci{?fAaj>=JHQlo=cge;Z# zxe@=N6ow-gs6p1`J-A70B&g_*BcX;jiI;p0u^un*EGFj2K}?F!bZ&tYSqH`#37&d#h^1-oB85k7;B^rW&Gy)bSS9TZSN zn~W}yYi&oU(|#9_h^m4&<9~yrun$}|^)WO?j-vCHTfwW~0C+d#IeA)o3&nN+!M%Uq zS(B*&oSXX=45mo(et9MOosmbEoL%y)rbiwg*mnmBRHUH_6l* zOCg^>3L1AQkP$``$u+xia3?FB1TKgn@1=L+)3O{8Im_d>WkpQm)lBTS`OVxku9EHr z5due|xil?pDm}hG4*QoT5_hu*@o?#Z^iN44#5?Q4pD$rVX{I~;c^8T2kKTjTdqq@l z=sX(m*NPfXu!Sit(d31vKkQy2CEV4*nSA39ah8zjxqiVJKNbzgkfjQIB5~lC=w!Of z=K~xXlSGR3R)dz5A^o-4gXUDZ!dzQtD4sP7^VGB1cw-r=J46;{83~L>t4GXo*i{($ zJ`KwDXmgj^+agDmB#0fBjaP?7Yu<@ZP%wGH?)um zzuN5TP6dHyYrdV!W-nalS;9tTorEJAr=a@F2{<;}4|3)=u$9BV6Yt%GyEj^Jsq167 zbkt_PdRz!*N-Cn6(BV!l)nwO4UcoVv_wl-wH;b$q4!vuRK#_L{q-^Ym-txB)J;4>8 zom)nqStf|IvuvR%A{1In?Wm~ZD(I)m;ogp|q*Hr2m}&HZquFn~ZB;>DFJ2-xC>ui? zHjJfLA|825<`QYG2~J#`a#p6gfKIm?&t(joj&lQsOc zpN}(#S$*?Q$_MWA z!x z64%m^HktI=e;4V>6X)T-#+fL6a}l@kR3r>hxq$rA7v69{Ck4s)P<&_bWQ|v=N;jb&BeyU7>jjVd6n@qnX#?aWu7i6gjvd zk$C($MHh}up=9JC7#6F}Z5}S+{rhA2)77E8$aN^!-PHu~MxOZJgQq;$LS862hw|g? zo&28DbHO+A7$;UtfGhX@irR$V5%)i1X}geH33WBW+KocbW7S6%pz?(bo!m}3pL$?> zi!AZm7Y4&_gz)*Q->~SC0+xI|0+}JY;`SHbT&CTD8wftXQY}93`30`%SW0s5wZb-=p)?@bfl5!QgP)N-@Oaf*nB^3XzHija?9WJ*wVxWx zXa6##9{$ndl;u&VZy$jcx6@e4V>L`35{u&(ZN`m$ciBY03@q|LC!S#=$pxw-7`&2W zHz%Bj#nR!B+d3BDkv%P`(4--$0i^LvA^fyXWsXZW;C#hktiE8xD~qj|rBwmDBlMib z8+Jj9O)MH+*a&@*(=g^=6(snk^Su^5#k|MFhCiINiVxJw2a9n68_r+{ zKYH^uF8ZC$U%0N~Ate@=_0fivslEODYD2=dsXoBV!7N&ayoXz-7Kvrb=7d7TR1?5~8g)C}1#hcL8xyOfRA7IvAbGf*Qb zT;w1zT434h&=aYLi1h4*+*Lf0CoP!9x6HQY#*xcV%Df6!B}(y}$t_IuXd@Z8Y!1t` z$S1oGO(l7$=W*rtJFv&Tm`J#9gVK+uaq{E6hzEwlhKYaJ)~FUdY^02l@A{eYmk*+p zUm;|Lt~=$Lf^#z^37kDFp`=Ix63^=L&@T)56c06i{75Laq&8!~lM0Z&YcK4htJ*zkjJR(}zZ+ZN|Tl-^>xZf_5aN!kv&Emr(+kwn>{%_n%% zln(+!(FtC4bg>aKH(7789&~<}=i_FU!1{3)A&5$0qTNvLI6W0L7Y3vC?;n_(k^?(o zFf90$3`Q#tL!oIH^bgv}imdzL@BC`!-X-|xGYfEq|8RVqJAn9g2)*$>eMXYaLE7OO zY7`#ElNOPIz4qLZ?$D60| zwmS-C;n(ajVApUV_ZCIRw}p~13X)|lgDuPUm+Xa6UJ9(1xAQQty^lXxLt zviuZZzSAA=eGbE)8;)b6iyJQcEWtl`>hQZ~^qBYR8C+4@fEzD0!#71OIR1h(t>3D| zFTP&Pzjayj(V~&~YQuhXK06w?pA@aBR-@r3V!4Uk0q*dh&^P-2oLt);E51K*Jx>kX z$Q2Jr^J#q=T;{hN>}&qX{2qK4xfEoIhlk!Ro$Nf2RfbM0J?34=VzwYi@4gH+$$>ED z#3hXUX9j2D{c)w?NA&#off+Qb^N>a@@@(5DxDir`AI7W@pOv@7Yll`+k9t*VaipAN z%GQus32`{T!-ZenzKoY-B=8YRiZn3q1gN~q2Uo=}II>)sw`2>>shFeOWoW%!MsX1H ze$s@cejV(5TQuI>-gpmUOI2}hd@&|{?!Y)%;IF@4L&-P3I4L!c^u|O8 zJkaZSZcR7YBJ6|6;&xH$6=fD)cZKZrybhU*69r!U7_bv*v$jAv{#andeje}#TmQ^P zjjRN^+~x~xpCPbJrfsK&j&Xdz+_l_hYYoa4Oo4I7n#HBE$t-W-dR!Kh%1nC&_Un#& zV1MfXwP_zuEi;qp$nf2C@gRa~gK{y?!Bw(Z$|f`XYDhX)isMe zikd;QlYX-Nvl_f6Q(y-r*g<)x6ulW}O06=J$>PaQTxF*=Pdo9IY1_#YOJ!O5Z|Yh6 zRd@%MSV{B3FLSx~V-r4q;9gSOqyzJg_oMq(U4j2pPk!{-^2)LrIMOFc{mTDgL%9mS zpZ)_K!z=`D!2gt;X*^Zm`@fALbLNtiNJO#kefCrIhB0(r*e|0XWQ$n`(F3^x}x|YJ#|uFKZ(j*+e@ks z%G0M6V%E>!LbSib5`{D-(9+o7WYLNmT2vK@6JnIPpvCFu?@n#5I zmZ4KMTKu@ABr-Q?5^>+UjUV5AggDizk?=Q0R3$4LwHl76O`O@$I9VFuGqv?sOdE~ze5BhcC8XQp-NMb*@;)N{>>0atZb|z@k=<{>v6WKyi zVmO&ZIa-nRHkVLJl@fBZvY?tbGikc+QvR^j57ZWV5AC{U$IsqxNDf=4@ZW7q3Gy@` z?|V%|+YVdeb@>btU0O!GxAoO{dZrVt-5bcwFCuELDo_7y4MMKLIz%_Nom^Ntn_v1X z6@3!=vbGV6(7tsw+_e_UWwl{c1i(#_k_SmzcigSG?Oomu#NO=lg6Ee}sNM?Yxet zNWYPr7*o%uMOU-`%^ik5Y3A@PLdJdmh-S1_qLd$49YYT^6{DGvWi;!_N}3TOgSPq< z;je;*e7ai@4nEq<>u!xi+E3!iNY^!_)Z{2f2D9tlev9CrkXTOTZa#S-TP^PEzr{bz-%)m46_2n@m%4LT@gYkv@EnD6|<8|C2%v{c$?3@e4*;7dGQ^ zs}2H^&UlXE*xGwBNAWdY-R8^N9K6IRnmm%RLajokYu$_ztZ$Hnd&ARc&G-sh=Q@V^ z*qz4hqE@2wqLq`i6SZEr@Vjpe6mVw3Gbq1D!#MU3@x~>BpR$s z;H%3Qaq3TFxh)qWxnkoTkn`*jes6v_6wWzgYCBNV9Kat^G zNs>Lenndl(C%>Oewr-JD^hbXG?FpoASjU?-DKf5berA|-f7+~@}Tyt$f$G-e)FS;uW%ZJ2ZXgqO~~u?{W`!&$^8@= zxM<+ju967fScbw<{0Q@I0Z9(tMnb1|i*wec5=HJ6vECL%e(j7V@ZckVMB_V`Uj2&S zyO@%d>-KQWY&Cku=RKYjT!h0K-{a@{2#wvGO;-6=QoqEd^xx?Q8u9lc4T$fon@Srv z*5V)9#24ab#*+rI?o|5dW zmO2L&&3rC)dn`}78oWvW%~%q>S41Tm)Tr4v!80hi7S)WeLfsy9Xvl2~QU1CUtr8q) zg(Kyt(F|FeQ7-eb#I5P5&%}=OWIm+-RisnNdBe%mMFjCXQ&3XHM=~_}4|Wq(j4PgbW67zbwBrxxGHAzzgJmuoBghFqb(({eBfjB!B+uY{O_%}=yW-nnrDWu zHh&|9UT$RRr|BrmI)?ky;)OSb563-c1wi5xMQSxp3&j?diSHj6PJ9BS$S;0Scn--# z;XWl-%kAjGbG~?w|5s$Xy9yt^@)b>P{Yi#DbfOL^K6vKI$@ulR5#-Mef7%+;PAo26 zK)L1&ZZ%kr3g*qj(aMQri^ErP@MI>bEgZzHiI?z~^zF$bO&_g(~gDk_}UE_q2Po-qGXw_h>3!UthZ;E2{Sfll{Fa{?yMVNICX-?1lGKr{;?&G1lU+w2 zqN#dv^j`gCTPkC5;?|K7+Q;-*V+B1Wvz_j(D%%rb1 zACj+cr1-9RF=(WU9-4hmi<}*?yKeXABGEBEjQSi^qsx0Y(wP?LspKzn8rX4>Tzxx_ z{&o=R*<%T$UaOKwO!FaUJYtbkbdz{pQlR+5drUXVd(hM$w@GQP0WrUOiQevBO~bsl z6Nh2JbiR{5Rq=hs|9KriGp}_}i%eabH@%2lHSi$8t!~sgF^QIi*;D;pd#TjFB684e zFK^bk5p~u&QYGX;GnDqyCCi3r^1&9GXdt4#g?04EJn4GP%9Zt1UpCZdnd#KmozHq6wA8^t8@a;ayTBT}S%)^}9}!#X~Lh4EvKTQCUJNW}l`b zv7o6skN_O#m-V8}C^x{sXWfS+~5uBMggx~HHN98^8=|-tPgqd<1 ztvWIwj$XmkznT?LKOa=QXeZ_DQRenr01HCS}Z#SkxjkSqGgc~Qga6?C6f z5cx7qnY{D-M24UK&F2X&M2R&|NuJ;dpXqgrJCmn?&+cmBPT0!|zMv8^DL0ZlFMCXE zuAk+V!sO^NMxNT5$YhwM66LU!r8QqM=5(Hy~3spUC^n0qc4#h;XugF{F zY;r1I76&R^M5;Nlv~|=py7}4Ox-u^%+UnGbCT-h=Yz~bi3)iY+-MQ~kucIgF`lnC6 z3+GgJR^zyis^6^DOZUd=c*?~8;gWs ztPHKQ^r9h`R?ye)ZVQ=?honXvMU69sI=U+rtr(jq+G_fY&YL1pf21RoZr<5V?~j$E zc4?KUdYU!0_jfm?Q%aL;571HW3y5c+J{^nA z>!ZiSQZ>EpXs|5_4TU}Bvj_i?d`Kp8cM8R?s|8Nw`w-&Qp@(Cv_X#W_f%n?jN**P4 zp%RlhJUdH*+MIWyUpF|?bN`l7$uAaE`>PSH%+sf*^FE>CrD=5|g+|fm;|^$T(HfG} zvw|M}dzZhTnU1>F4HxDa!>HWt8eY{s3dfIEpaun<^j_09vTV^tuJMA9IXJIO99>e# zkwDCE*=9|bUeKrar8l7vH&60LLyJg%$|C)W$Hg1@ysDjlQI8P__pq0@p&3O(MA(qSSmAy4^r#5X!}krsVg zEK75?Dbpp5wj}l4DPbOX9Um}^A$5zk3%#id-lWZx0m%|)VO%?)I8S3A*PvlxlpJ|NF7AEKf#nb?M&;(l%%%U$%jNY)9m`UxYBA?BDi zo&BJiIISLsj_>nC0{olHcyX3#qgH0{%$x9+ab$PPn}pC}`4Vf@huM zr#~G*RquF^KU+T&BZnf6k$=R))OlR+L_^$bkt(>a?qMInTX6Gk9-6WEoalkuaQbZO zUh&egwrJK+I%l16kS;A!spsxKria{e(Zu{w#PW);erb=uSA__>`vwoTWB&oH_p2SL zne_0F+G}~wXist=cLp}lK-jHg4C*}jk6&J$Kox&Ip)Dy_X!|43iWop8ty$#v;TI%Ze<}9JwIV)C?vavL#z^fV&&>;8Kx;A; z_(`)qag7d$_RsbgcrPc>L5D^>c9;V$jJeHqM@%P=+z!%zaTzE+$B!Cp^uZgN8qgFY z1y1=^AYarNCUDS2eB2S?{Tlg<)7|$&I1~8~*LMyl1;ZSv`2*Sd?E6afJ)1_-lUG$p zxb0jNTRoQqEI30WKP8eAp)2Y3@%Chq>tu4QBcIQp+lg9KuHXuvLWE+NXQo{u>+jFQ zflFf1n5%Ee_T_K*8+Y|c`i4njH{&sQI+&8SRRN@`Wf~=kZ}~gkRV4ITE;&#{}v+O^HRg@RDaxCrioTPd8@2MMwohSNIrRm7%YG(Ega z1@~Irp}`$Gr~rPGk+#P<1E)+hZO3>jUlc>FHXNY>28z~y{3GO%;%V>MDzQaWBaL11 ziCZ)5Eq?(OaNHbey77D?YrrVd{^y>!&N&P%TinUj3VrQh^+_mBN1wLnPQ_>DcM@gq zon-Xp{iw^_jts2yB^L3KrFMtZu%N>f<(X)O|bJediu(8Pg|Tw>*|?6JO$V z1cz_vsV_LJA|1yjEkWJ84v1R2d%5!T1mu-wf#z>6LC*J`gy*0^H#u0+RkjM^*P%be zeHWzA^kKKf8ZwGR=av+=`f@3H)3T66yqF^1TBXJ{=D5(^Dw1@v&J-*mWsjSFsgXxX zVf3X1;in2L>(%Fvu&)xLY4wCK8r3N#6<=S{P{G~2A|{w7R>l!yxQc4}bP!BGB08}+ zl^AW7~Iuyk&9$4)isozIz(UzQ-$Q*2inKGflN#fA&afQM;cUS$mxy zWjmTk*SPSq z!mo(SCNj4}Mfo;L^xw)#{_dJUvSv#$Y4W^CI#TX%_j7l1u0>6}Q9uXZ6f+U67~LtH zlf`qb7ourG>Pz0dC{1kosG3-wF(-}nr%{GCj}wjw-tNBTs81zZ&T`5&){qK8o4!l2~k+oPxhDw)5BqV$lJZ= z_^9rMc{ta}4X9{)R5=KTS?$AzI8CEcrS7Eg)uj5;^E&nGCG~OVWkL!A)=*^e;vmljRkCCB?IzVT7-6UhbO~EfT zw&VCWMkJvYuu5?Ywn$b(FRllRYMac^VA`ON$%sa}=4HH`l?>ImBPMp=bZJ9h8Gm9? zD>)x;N}E<+B)WB*sdVNivf%XrdQ|9d%J{9JJKB<|xxr7mC7_Dvjg&<8TV(OkD>C?l z|7#M~*3KWkEk$-)*@@3?D<-D3*7Q|QA*%7vqv{t%(UkW`h~2OoBu+0JskA!KTMY?f zzw>Xnu+fi*^!9YBakp6{V0;(E@Vc!G;fWSRncnC&fj$jTYk+mj8Nm zJ-;PNn6;hmDz*vVPi=>;P?@gZ)NtP$O8i~9fd&b@uC^WNNu8#z*3Bj>FV3e62F~&y z{oV21NqJO>8%}PvyrMp9J!p}5BW=E9g%dj4=%I(FaD>Nv{P^g2F8f&~T3MOH9kQ3C z-xOZ*Nx$adw<&+b%y1QQ|Ah<5og+90M1nV`%9Fo&R+41=wj|5@9jRW-I;7rumt-^z z66L2xbbWO!xhD8^b!QypB3>Z6ac3KGa1j&n&lYrHvn<}V%86@9xhqOj4dH6;Me#pv zpAn<9YI5OLEIrdEN4YlCCd9(!adq+D0gl6G@lF-rN1S!v{xj z5w63;V?5gHo)_1X58fm3nqS0&vycYiL9z6?oU* z&`;Sae7Z&06TLRUeaoFOU8unGA*zTwDx9Z{SHV&1r=zb^N)VT4htGd}iOm&cZ2nqq z!rSg|#JFuGH_Jg&q?DpTF4%?R0@ZH3)oVMhYW_hq-!JFv9wZ^@ixPNk!baiTbOfoJ zEiGjCEO2H*6Lwmki*Kur#a?^7utAy%1RU*tas`AV-%4fxKwV!Y(jPprJo5c`>*LhQ<0$WVBXueZIy z*`H3~BO82hsr+?r&9x_}v_Ti0h}?~Xw};|OhR?}ga+qIVeoSx*7!g|F&%U`j4)wT4 zU_;YoIQ`oRY-H1cJWv-QKykP&eLc@TW`4FE(-%Z)TB(? zLp~sB6E{3g<|8_!QjFFG*yFC^5cG8RK9Oy>Ce1&V%cpOX#`AEu$n51>GA)nt`{Xjj zXBMd8ij@`UAaewdtM(<4`%a(_7h_P5ZzuMRmb1B&M6uN6<#=%ERnBFcB@XmTMz38{ zQ1SH?qB4GvYZ)~c%O>@s-BNq8Ci#Y$?9=$4^k?+kzZVU9Nnqb2Q8-Afk4$$dqhCub z@apBs=th^8knQ17kZS>V>wXC;ODL<|ceEYBsH3D#xOY=Op@x|+i}y|Nn5Bqia}038g45i`@ww!5@k}zqR*yefaTNVa z7kJdeRLOa?0@>R5;x}y?Hf;5Ee5ET7Z)*s_^Svagw|y@+@`Pud(fYmOe*;P=yeJ(> zJuGhFR;scSzP+Z zl;0y)MVj&j-up0ZY@s2ApS5{&X<7oScDXM1EFa4G+cQEer_Lv$}itW zO6_BDb zlHRRja0*(7bQ-2|c{4@$jin5Fam^R`&&lAV7s_#G<|-4V-)4MBPZ-j@Z-C-|i4kMt zM!f#c#Rc8=SW6s$o-c4h$NZLHbr-=qXqJT0^hdaOPA-0P(h@hs%3|~N!FZm24*qj~ z6+Ut34r&kDfD}^Ku+FCzbP72lcGy}}yjcP- zYF~qFKeq|<^9RwL%)My)@Gw+Sb(Al4y(ZphzJe=XST7z3tKu>ZM)QWQ=E%^j5LrIy z5?#nD=Swvlu}s@vv~O4*x_U+h+b0-f_0mXWJW9B(hlxn9-U$t4#EB##%u#K=BFSu5 zAzrx=BxqPX(pF92$L>!>8rkE8tVkAKczJ*utC>NDdBo#8(NXxzoH+cYWie@Yw@2D) zm_Nd9z}Y|A&^0a_9r6_6i(fb4W0xbbR;Cx0wPUdRA2nhTI1w54kHbrwk~!I1!%*Ot zA%4c2B#~8Czo=yVAC&y@3=$o>%5C4ybJ~vGVrSVxZg&QAg2 zz->R#;@@-dw7L7RKPX^%V0r|?arthJWQTKW=4Ebzm_edTR* z!=thDS0$Uz=MUmQiKWPS^C|Q?_X)BPx#9^%s+`}6D4wmbM~?>O@VTERxcU1|RIBum z@3EPLu1aYL47e+3{X_{O`JW~c?bgRDZTd*!mUiU)V1P4!u^2C$6N!g{-(&VPhtsOP zad}}gT7KLBEA}MdZMj*vxk4V7zFUOtOkbgobJyYUrJM1YJ#n~Us?d{q{uFH=Pq6Zu z1^C7N8|Yl|Hzba<6h+wXK+O&*C_glc+p~5*I&j4ZIR`8vMZUlA^~HYp%c*%bqeiOO z%(iU7E1IqGqg-9C>!je73o}3-d)?U$7EvP3Z5=n+P&kjSAA!q?mt&*h-q=k^+eWw0 z3Lma`idOeJqdz%joQ!3?FLOyUI{3-> zVVG3;V!ft#>{C;XziiZ_wiUll3HiG+ zZdzD__>|T*F4*QEKfxru&hm3FKQny4xZ_GZ2A>M_mTkd*-*edb2#3q^KB8~Nsd!Yv z8+<2j0`3r5;jc5daWcwMwD5Q{_D@p5``HM5@_-mO{hNXpE$l&0WOC5UR0BNjPMR3) zs6%l^pV0X|Td|>kG}kGeNj~=Ii1%(fE|TF7k>J8SKJCwX6wqjfPDCWpWe26`HK9g} z-5E*#eiytIPvTHvO(gevnIkpzNI)uSndI3c6LjfZC0BbbgLwIm;gu_;&=1~?d^Kz2 zU9K-Biq`jtzDzjMIvGx+-#_5%`lgdN<{Q!d19PzZsC%f~OA*TijNq>hJI8BFnF?oT zQ~1t7H*)8oE^!ka#k^iHR~ORHYfU~)`m={2kDpbfE~|mW_`gJ2&F63iZ;P(FZ|AM2 zYg2W7XCYfrCDwJHDz;bjM<-pzqsl&AVmmb#?b;BFuE@zC^`f(6hwc!H{~AY*|9OK( zT0SH@Pvr^xFIlphGsP1d3J81t2)92-9lx*r!Ufg(BSVh@d#M~lag#;;OKkWb@L z%7iBoPurEqV8sXiVPhM55R@;>IZMz5*^jy2+|PAOyrhU5vc?nb-g9dWD!3N(nLGVJ zLNqzr1Jxm_GS)}MG22TFwb!kl|$Bq%V4gq8*4+kaML|3?YvOYA{n zIBpq7{Th6@?@Qj~*mu5KcM*P+6T@w1Ux{hUb$)Q~3-Xx0AeklwT=Sneq-Wz! z4qcYU7J^^03OTjG$#_n^BhkHBf$k(- z6IlOQoNs*@sh!_~)S`-nlFA%+^sPibT27>`rkS*qP2;74Qn2Wu7doEcfgvgyYtEU7 z%`DrAk(E8Kx=fWDt#?xBiyPzVKeX|zW#`bL3gNtUTMX&;V#qzE*8-E)f#^QIM%Fa8 z(uK=w=$%n*#Qx@M@@z1Xu68dYFGr4}Vw(hDJXYsFiq3_hF#|#EK-C8~shbRxjR|mEs z$^GS`4z=Axc5RHnTPfo_Mb21L=@-AXt4dTrZBgRIz5EjEYbc^xnKU@6qQ!fH_-g+P zpL!CG!(*C2-jvw8)}0@6c<`4$qW`#R;iV+~#X%`Lj85iQi}!WU3H}>^?N} zr=-^5GycC(bkI-E^|cy4@$N3#5%L?I(38V!oIS~{ae64nR0!@ABMm?D9<#>qWnm{tWsM3y{q{bj)BOl7TD4DXaZQQ$JLgDT zmaoH;h#?vAc?%977mMSwZnKNG4T$#`rV}D$V_zL}LWV5|NlAh&3gLxwi{eavXa8R0 z^xzS<$7U3As7xm-AFf4C-LJUiLY&9>;9|jfxt^458AeJXlZb1D)qed{$Y4WLp`}S zUGO0ODd8q8HN@_7W};w+sd(_vH@>*Ief!) zjKEyq!9qHt8k`=ai9xkSqBCAnSjgJ+|INvrEHL#y}MMQJsFybfTisXOOmE9QiL-hPvGN!cTBe=laS7 z&YG({R=j$V3!SEh`!`ha{B;wWuMkBl8kLZ9##*6%nSvdTzUGEzc;mrk4QP5+6xv%c zNR}G|8t)6?s>+Ih3zQ0CmbbJ`ry%48yh=tXiLXTE*UOb~{KI~<_P<(^eEFW(#igAzCAd)Q zgUV#LnIqq`v>6>8^Md%V)}v-|1?XjPB>SRj9vQ9?jQ5|E$GX?AlQV^hv`pwzCz@!} zEpu$SQ@0;;NFdpTk1Iz7H%8N}u7%{ln<#F6X|{MS{mPFDKg7omg$a9k8>W?HipB*#Qzd{DuGS~FwOdi2;3{<5HjZ4$ zzr`mV64m`!VvQ<#L&Y01Rghz=0=L*!6Px|Z<8&2+Yl=2mP`z=jbZ6pB>atOS9@!Ah z8z}JDH*PU5o%I^8(v!eOe%jbyPM6>Ip__}@P)}^_&53>CEu?Rfj@DlriQVVP$ZYK;=QDKlAz+$6TPF@xO~8pQhk0yg=vDZBsm zDJVNVlGW53f{xqQ;oZqbX#cqjVjrJo?A~a@7k^cFMvb9QNg5L5@|kH#3h@053qPt> z!IZyc&|1|DGh$<4S=TPmET?cJx(BqPp1}Ry<8Z3a5&m8Agu}^U@OfQ0?6Qf6mm5+c z#D6M$|MUWqzj5sCMXOmA&Dkug`4cXj;lbqMMesQ=iS^&)!9Iz{?9b2r@Jo{d`6mtf zcN~Eq6#_q(t%O@m(;?Es5GpKn+5QR3*`IG#uxrxY*oix}*me5Ppwl}76lUFr&X-E; zJVQ@bZKe-vUNME8Gl7GfUi;wf*etMWYX#1KHMrV|!0Nd?do|9UofkfpmF>`Gx6S+j zljMpaN_`7!AzQ|-H;H2>PngTv)Xiq)+6>uY%6;&yY$}_1M~Pi>s0(_cZ^D6Jv0&vB z0dGfF!dB~haHUk9-Lw4y}NlvG~X>k-$BUGcy@`i5&LK3 zMM#~#6YQ6D!Ri1Bw(sdH*ifAZI&m?86I-DvUzJt4GmdRAp2B9O8L~RM51{9#8?#~{ zn^`(JlNt4bXSUz;V*>AvXZGA5&)gJ+GY<3%2N^NEJOITOKT!9hmvaU!#1xR{B&|A0B3J_pRU&V+ckDrR2ONZ1-?4_mKK1G>JQ zIVio0X`EvT+Vxu?G^`Z1X(vJ4XAJbVBg{Jg8~)z*V&ArHWaXw!XZITDupX_?;lY_P zOs4f!W^Ly$W{S@^82Q2ov}fF6+Dt}*#QivEHf)2D6Qx-BSE}sJO_w3CL(C4#)D>lK zvJlNrG!jh;lM!ta%4YRELwI+0KcpXPho-DIV33#u`4aOWaGn*o4%-DAHtdJ)QN^G! zDHjy7r-CPck2#bj4!U+u~%+ei*9BF zh&Ha|M6%&2x-6@iikDI_>CXZ zZ1`0vcHWPpust*s3<4s-uJj7zWW~WDEedkG%%NU7)p}%quk|!+z#NEJ!K^O& zZ9VbW5$i9pI-uI<4s*v&gY`0Az*<^>==f14vHBVF@bU<_drK3#BJ4pe&Jq^PcZP~2 ze>k7I4=TK(c6?5TusS5=EnhljM z&ftIA6t*v4&(s;8Vp1Fin2niou%WM-@#eCb<3|%?hgljEKu`z zKfi~8qeo!V{TWbMJrTBT90y;g-UeSQLpHtJot1kdV(;j@fXWeR5L<5u&);T4u<23vvtfB#eC-Q^g~SE&&kZfH+9xmk3D1j@kki@-OCJ*Yhn^} zhr^YP=b2>w57U2f4rCmOfm*x`R2DCWbD_F0-#G`~2LFTEFDJ4i(K43FKFZ!0nazGb zvz-0#HG|#xtd(6|r!2}wCL+=99`?W2D;Z0_I40&w26KQZVYc)JFcbcaW7f3y!n0{p zS-1GT?A-I&?A+u{>|P|rT6&2Yx5Tk9;5{Bz$r->{Q71EI<83BGO#vQ@JmFA|7nJXI zfJ?b%@aTC9^KQHuY+jfQQ+}?7yn_fH`Hq1Ce1$puI|?Qg90!fci*OlqS+>lM-MiF? zecqh}-VHCI?1dETv&V#$xuL_h=3Iuar}E(6T`Bg!I3IS?b9*-5Ns|3JG6g2p><1## z0?#aq;dy!;#C$7*V5LOJ3TlP!gVJpEURhSfuoF(t&4IXa0q`z)EK3}=v*pc^?4PD> z>^Wo_TY_F{YA5hF znql8i6!ePgK~Lf$ymz<(1*Ml@uSN>^C>(&yH_2f7BoS=)6~Wk+(=cMkNtoH61Pfn% zhN1B@*#>S7TkQ55@-HUA`3vPt&O;V9r*4O@pALh_xCN?C9fO&C z;}~0Y1I91wq4gYC5AK8OVA0fBU^~wq-tLHobKCa7gRB_VR(z2S$!K7i1w0!t={URb zq#Nt#5MV7Xy>C75z+{HYo5I*j7g?Y24YGa^Iu(Mo3gEo!FBl`I!tPGJ0Y_cEpgvZe z{iU&qoqs=q%^h#W>i7SF4Klfquvim%4zGY0G!58?wIDrWKU~e71JB~uz{968uytJ; z*a_>ZTFrU*GL!~U?}M4E-MNfp88Bt~6PaI$^Oz@tUX1mS2m5(V3OkG4%YK@f!fuyO zWcNF7XFK0zLFc?1FrfYwo)%w%_L5pa3s1w@*>Q}n%RT0UX(!|9aFzLCQqIho5zj1m zHjNn{e}gG@k%R{}BcS)sR4`ho2(6>ynO)o@(3@lgTZ@L6j!A@>d?20q?DCxPsh9z) zj?IO}E>CE2UkL#j6QJ`}1>}q{VMBhKv*|xIS>wL%@YduMOu4PfhIp@Ml|-vp&rpUv zAf#aLyzYRXzd9L@JL4g(a4{(VUI@LrUBUg7HCX-WfT`;&*}?q=Z2#}!tp6A(ws2xQ zB%S-oj9=meGh>S&>c(04Fu4b&{Vjoy3Ij}@PAH@l7r=zO1+Z+}CV24(AnL>l7$c(a z11g~US{~TmKLQiBBm?!;V1KF4Wv96)3g#!3Szg_p-Fjj^JBVhno9}tDFE(nk zIhn&48#a?UyCjr}^NM43dhKBB%9!SH)~p;Vn36S;l_O z*A|uT*AjKS)E3>Ep(|RI_Lz0vW&o|a6)-Dt82j^}3j59Q3S5_32qVX2GK&>PfL^^b z`0kwu%bG@jg;y=}O;HI7O>Mzpn>nQZnFC{kUEo!lE_}Rwi}~?W1Uv6fhoA)uA>4ns zP>=L76J=7F8W{z6Hct_{9*u=@vos*JT+BROq6&Ks$3wqMDs1jc17F){7@Ry0%69H& z&m0(HYs{rYbC11d&F?*AU$`G;J$IJ~#@m?F4p&H;~BQ-Gu6;abQDNEUv>CRm192k)lVjz)@1 zZ0u8OwUeFJO)FZgqf|6tL5MzCy#352142Wt7(d9GYzpmdgRYPM!*8uKG~FCj=bB9zb6CICg3A0=9d} zW_JHMRn|FQ5k3q1^3?rjp-M*#*)oTqEqOH@U-E+a&^a5fPY(iI>keaPkAmEkr_Au~ zA!c~z7&te<13azlLAXptNFP19E1(hY2$0K_N2+mITj+ z;k6OaDB%It7Cta8$Qeq78p8FME~B87#pp`rFz?oRF`}x?%(282#^vHWX2li*IA9|W zZb2q6w01s7IE{optOiwO&fp~M5u>hchbON+!0)3Gtn2#4FbP30<4`d~HxZa2Eytd< zRbi_vUqRSp2X=;TENgMAjx}sQl**zr@?H}A3q z!&FDO(c=e_V^=^((j*AEoB-1{55V3AY1X*wH|&y>WA#7Xg^3oUSo2qVSoxMLc7Jsq zyLw+VoBj{6<8sWILm~T^(*6u4tXteNb8*whcP9gdQ6x| zgV|m(j*&BNvJU$Bp5Y#;!#S64OiYa##OIiUet{{>{~}_mmZq?VjXCV)mSon>FqZx1 z;KGhRI+yu#pqPm_D`ztP+s<@{?PabmS<0A)on!V18kDtfmcf@ONAP<3kGVYV0@JXt z6h_^VXPfT*fK78>fNg6#xK}+_UXrLPDM}k!t6d$tMz--6kRIxq7)9bq7xWUd5_vWU%WGC$U@3&0@z`cfqWiJuuulh_xAZgssuY zW3i+kTjQ$7_Dr4uNfxP)>~I|J|4fJ45I;z5pAIWT|3K$uEc;!*gk3i#k!@`9XD64d zvU@Md!H31m;e*a0$kX}^zl)97Xc>LB_n|4Mtj_`)i3d=XCWa#oDKKHv0_Z&50s3!D z*z69DJ$+c^|0p{1c&xrIif2wFL?kj4QHD5cKlxEANukh8q%@&Hltz?!ip&uiA_qRORNxG-Dl4Z^vpzC zWdRH)yvE3IBWlGGvCGT=$^PfSo+#4DO-m|E&?Nx_SvuJi%8p33vBlT4pr`MH19it? zV`l};n(f$CRD-2nvnj-9J1HgxkoxKWNaaE(oD=`R!c2pb9}7`KXd7-0hhd|>Fr^5% zQOnKq)K+knJZHwzKGjuZ{h^9&+UAB2d;z%ZAAG-Q zEOgF!%-#D7|Bj5I?Gvt``oVU1x`pD}-0wI#q)izgG-xVs41Fn&MdFu9J%pDki!n%%0$EfSF5knqf1){&|Ij1TCbUqpK=R_^M)btNQG1n zDAJ(!V_NNK%3Gc1&m(b9-drC;-bo)mUI@PqRw$oBVPGk?N)4lMu>iRi<>6yZ1RI?4 zkNwg&LW*e|qQ+L>?7kpqR=?@K&So=K&mQ)%<;4jjl8!wrcDL`^>d9ZwfT zK5>F}mn@!m2&3ZYBK+v*_BoAstUBa`iOrgD_fbN)hC9NKR^UTj2@bo(!Leg81bdoD z(O?0OBD{FszwLQv-86Z6>K$bMTmWCCj^VL>A&Pv`ki_Ya+jL{GadQ~!d=$@8UcP6B zVJ2|yvBn%DO;p=$V!OE7Z^8UqSTlP7bq&q9xh(-BaaK^D;fb$FQ5ZY!4vxJWL;o!k zq9spCkoKOD?Bew#7%+#Hgv*fgv_^dG)?&4LG*S2^0rS>m;NU7h$keUGnLVk{w9q7Z zGducKzmY0*cr^H~2VM?#*!%Q3oIVxds81j+^38%3r!&r6FU@A!7c+~ZU#!Aa0d`_y zxMf?!QcfRZH#S%zLaP=jPaa{}B0r4$+5!W|X!cz1BRd-YgjE}=8`_o7?b|shs-Aye)&-s)%lO!AA~lo^vtV zFCCk2r=YOj8m@Mk%zK*(l%>K@JuwcSIGxV1`w(jPjWFFB1E|iE0e5^B-gw$$m@R>< z$tXKI=mOh}MikEMgije)pG=Cza1vl$o`5u4V_IV3N7NEe%RHQ^@wFC87OlH1)|l1aWoc6SwhXVsQhGbqNUBZI9lo zQ!uwP7{B7Bh=wgmNp=pkI4jUXz3WiEXn;QgB{1*#05id2+bOw>L0eVRBOQW-Z~6L?Z)zGcQ}7rfWxbnBf=~d+dn5_X~ zDVXT9g#M*;knVVKUhP8>USGf$T5gg|M>~$Ne<3#!`i30>uP(eXC}W2Cp>MnXD#mF~dq#Rw$C zB{==#C!9YAVA(cFGJR}Jm7{BD(+W*GzU?W-O&ez7T~i==kgMVHOlbJF7xkX9rOOIi zm~vnz6Y%EN8^c7@ef2}g9$kz)T!j193CPJyfYFpRybd~x#*2Zl%w1PS*~gIH@ts>sA`pM61QE9VsL6PM=5y(YQ|n-Y@5W=t zy=l1A!)bQ9#n74@2Ic$r*sC9c(75uAEtkK<%p0PZ?^r$7wl9mCehyN-!XGm1silye zB~+IkOErqYXrIuCg$rvjY4T+(dv_eUnOl$|aE(PdPsE87;xswT{ z|9gtPb8+Cjm{p0NUU^u#-46%9>cLZ^72p0DQ*Q1GvgR$|MvckR?gvH4F$#jvmPVAn8cPNW zYv^*uQ8HYjPr?}!@t`{mYDKSb)V2~%HRq66z8lAmf5G&-)98b&D(zdtN3Mx&h&B(0 zk+ds4PtBqJ;AEN*b(9*`SX01<$#hyqf=V7bk;qeRQo5r+TkD!oLCKJ~m4mg<-k|Ul zqhbo67x9iX{q{fHus7j7l}+O?hQEc*;_p!N{XEzX18(<|6&d&03Sk|;EHaL@8zjj__KCQkmz$;i=3z6-9c}hE`Mm8h$aIrXzm_3?w;on?(s5X_B zoK>S|hiJi{=``RYlsJv>@j+{W6|M9|>L;B8`z*l+e2FDrrpuW}fdkD;%g zdSn$6L{1SQ?ny#xTn560Z*x7t zSv)zQ4fz#%5Y~=EKs_IMtWcy}Pc^!@umD%YjBu~aAFMSI$_qmf^41S~pBrPt!G0Db zs))#6bMeWEp-HHo&1#5YkIrW@$xGspJY;|hsa4=B19GnSvAvRu;s5L)g6<}u=1m!H zHE=#&b1^&QP(nkB61>nkx;(YtlX&xk-%^%M6j@m8#pq@cdZ#;qQddo&KVKv$*Qy+U z+Y*^+RT1mi_m5d?DWmz!1k6Q=sQZzGG5vGrpM%VgK&=@M9Y>P!|NJnfACY zRl#BsBC+CzAcghMCq7d{swn&eJIx&U^&Ur3OBp;qrsGe}Dg0-47+ZhI!U4?&O>=CR z{MkPC-LZ&$|C7a*&RNS8k161v<8fG)zQ)AZYp~<|@TnPA5Lq@Bu6-P{@aHR2I8w~g zhh8(Im??2e&(8Kx1eeOA=<`RWHTE%gebKBaA`{zIi&3J564i;xQ`a5| z5=*PbnS@@Zr(}gS)jC)|Gy|Fjt5EUcKXm`jf=uwklZ4g{dxjfzPyQh@57vxi$?59fS^eTT2wlF3#$Vr&;noKYp-LFt)WzFfv52p~0Z*-RtbdS> z1;>K%U1TmgCkC<`+d5eNIA^q*JE6N#1RD?CW2=`-VodsHW)}ID@o`McV0Qyk-F}PZ z3+iBk@ImhSlF)gd)8Te{ApX=C7;H?$&YDK}*-ob$e#vyDwVs?$am zBr^Lw#rzTAtr4oA`%g@uy7d^gaUM{_6@YKQK&naGOmhwr(0w67(NLNqRuXn!K9 z2@;*=x!|wXJ$xVAi;Fcikh^~z{D1YJ^kgj_T{!`hszSJZJqx{a(HOvHG#s7}S?;Mb zsEfp(Fn{ceG(}mYBBt(qgcCt>R5~!5B-M6N)SCc`(lDXTb*mtSob*TK%kB%f! zVggwh+!ld_A8upAc4hi?W;zw+y~l>u3{ET5flX=#ZbWT{a8(4V9t6YTAP;r-ZE!d# z8YbSik=f9MZ-*+eFJ&L3RZZx>f9EM}w16CvFVc_BRN9uli0)0zVrl*wT+crP>JcMM z^Y%Bk`E&vcYCVP6i=ntXEdUGC?qaUdU$|v7V`6?dCQePmk7r(Z^1ua4(!J0mrU>g} zTVOdF0ZES(47L_yY|}F=uSmmfxfWCfNs~({k0P5bNiI#8B1}`UAu=7)E?q@`wK$2{ zETFmPwo+%kE}44_8$^uNWhk>?BCf~To3|&}tJxN8eWM=k%?-ef&|r8E`{PlwKl0A5 zhVhXox|?1>(gWr6SFV)i{3;=ijVHHHmTX|~DBG)0!%T##*(a}577?w%0@wNDu2ePp zf+x{jUkh4uSDa?lmBOJejPs@5L*lLw`CC@NaE&$GLizFVBp*K7F2YTvXgsQm#-j8L zR{B+1n+_cg<*7y2|mX9wLqo=FWy za;c=HkWw|1>5jiPy+5)K5)nzrIeii;1sR-1T?oseaC|K`q3gy+sWfIKRf~Gi_Tdn6 zc(Rys4D<2no)|T$6HWNNiCph{)3-W~_ZroL3+LqpCq^PrARA$iq9CR+7h?_%a6ZX& z>bs@R^^`pFRX3ovs#mBS4WdP#FOs-R1$o}BCB4e4w5jD7C2ahSBdxmhvQ~$*JQT_E z=x3N#Rp99TL3}vIsK~*I{5~9^{F}#U(I**N7&M=^lO^!{t#WuLHeKf(yZ(x|q3$B@ z&!@*oTg*?tyJpeetXVXTnbInCIr2~!!?@>qPzc|E`B84THRy`aPhybwD!`^aZ(uL; z3gx!l2%I29!I776=J*5V_2wsQ%O;2~vBC#?2Tc0G=`6bYAXo4Pry3OLappqWI$%v= zk#f|vy9!Q9KVV6s>nTXIjyjedr4KPjnNr$) zHg&L&{RSj?QmJit-X`- zFruc5a^aiot<(tnyk;gk1{Pq^SwEbbkc6J!SJ)xpvCz&mf$mmIj7!{xBS*insv{RU z4@#0s!bdSnZ!Aeql%*R>@1Rs*8a-XKoAmaE)6}b}6z3d5|D98#*7`msF>HnRr<3tM z`vy)Abs-`r0qzurEWx*MKRk(Q|AJaKZzAsxdZe&dj%T)f8&C842Hxp{UA(xPhk0eQ zb$B;l>mw}72L<2qp<{6lWiQ`ihi43o4y=Rwo%fKtCq?UD$&&wUO)87wIKN+pkUEu! zMOg*-XnFzN<8I^4ZO-G9O=9E3>)1~VaZEqU@hcmL*!6M2Oq;)r+3zvKTKBD(BI5_M zlBu{q)(|}XENrNn!N0JWt&fnDZ2kK>zons3+ego7n` zmu^_$Y1a+RDep%9axuE~bPV5!}#}{U=68}%*XT*dpka=9GBB2sdv9RHi^=r$&lb9)w}t*&zJrNw%FL=u3eg}zn&lu%)-f`)Rdg~5&U^~X z)N+RJt<3dIK3l9Z#6GIzvq?r;%!x_DUCkGV=R~3ZO9?VY&LHY=AjcdD!@Oc8W+hm| z$m0O)XZT>uA;Ql1KGtDskDX?+*nB|+EICmIx>StnM0;hN7 zU53`dJ2=xf1cj|jik?qSNocX;8-;-^SRLsM0 z_y{w%3&qm6LbO4VpZe^j>7OyD_n2J7SS?ZXozTb6MnA|(ha&C_r>BN&fX+QxstC0r zks}tgMBRg4tzS&Lq<+HK=>`7X7A6~IVVXStE3S<@g_PwdQL^m^DxCC5*<>kwaz99W z=ebaxtPq)XUcfWqA=qExqukFTG>c=%Y}ZBO-S>%z7Z8QN&ocBK^u*`}U1UyjVW&=6 zVdJB1C~FPCz?xVbNb|+@-M09+!yU8!6`+Z$8x}skk0FC$q{}8FMI#9(PmRLHdKukJ z-9zge5~%QY5UtwdK@+~+qb)JN>Bs04Udcig-ple~n&o%`J#x3u8vYRtoS&#Pr4X-U zBw(3hO?=4-^kUh0y5fJ6nnRy(x=jG>PFezaqjOjgT!!e-Q^;KyfG`bt%v@>=!xUtQ93~ZvE~*_#8ia6lh$TJ_RpwCeu7y3Q+rnGa1~o zfCpGmQ;mu89ETO~0vh|H(0Pxa`m5$pA_FP>T1#cEt7(#{8eKYnnB&#q6WeHTqA z=bAw>@wc*WEPAU3xE(iPdNVaf@@d=*Joy7=a&!^mQ}H* zmp-u#v*X#Eq=T%R>oC9fD6_IxCrH}!B1Q3HXsT!+E%iyEnqe;*zs#KWA9ta>{#x|w zs3ds{{Dr;mMcD5iWO>Vt@v?p;LfeFK&{7x?{U4d?(rK7*FBWZc?&7(|UyRzkhHz0N z`ev+0%DNha4nIP^Zy`cnC&KT-N|;{GM9G9P)L}H4bWXNlwtpS+wf1x2Q)4Rs5JCHH ztf$`(X3;}a5fYwTg+0pu(YW=Iv{Gv;85Eh5M(rZnxNrj9_qKrEc3)gxx*aEze7QO% z3{kTzapW5)E5wKVWbNsbiwz}w)}?uW1S!>W1-#1Ipz&Lo-tRM|Q?_QbJzkJb2?cZ7 z?IYX|>%gehZPcv;?OcpEse(KFqs|WZ^rgxcC~s|8yYWNgoP#h~x1h6HHTF zjTb8yV>Ra?UM-QsOV?faQrd`_ZvPM-(huoVosj=^9H)MHu@;eP=3@GuotF^C7)=S( z_FiE-wA~@k-6yNXJX-a06(!fJP}!C&d`fADg{w5#1PfEbI*w5t7({pX6By?kl4@c! z{keaLZoF_NftM~6s6LlUZ7#B_#luW%LLUo}zRxx+yTk&Pi?a8tvzbA16Z6aFm zA$)i^*#?;@6EI0~g#CEf&4$LNG2ZoZT;McvzGvd}@7zDg|M`oOw*^q$?nqVd;>b4G zmwYDgA)lUoTV#bqfaT!>!wH{pmh-ZO z_o`v?J4N&~G;`s9Zq0UR#3{ROv~yZy`|+JpH;&QGtV#^_3Lc(&^)wf24P-q6keFS8}nRJq$ zaga9r;HT8JLbN$an~Zmlr-i+IbhA^I>eqF_@YNLhH}eQRQd&-0d9&#mCjzu+L_w$I z9n$|zq3%*y%Jb{TzIKkm-CGUcokU;}yg%i1TIs{ioOS*GySSBv+{Ror&Z=tJu6`eiXv0XGC z4+rZJ9id1kJeBCScRTFIp2y=75whIGqu<@rNXAo*~EPp23K>905+v=M*boPw9B zKUOR=M6&iw*g=VsWxVO|tR-~)gAR545v5ldl?d2)0AH#vK#q4D5^7tq@-UY(@Lr7S zbDLR}>>YOBoYMu%1DV*+ZuVgN2BuIy7U$3ZWTmna*f}JHSDI3=uFhk!lZ|n)APooq zi$}JU57!8-K&<6tsQin^VACY}xz2|66oio1*kGz_a^}`AcRKQ{lSUrT;O(_B<>}O@ z^Y&Warb9tOkQ(ogXyG-;-L?>Wr!9lH&nhf>JAoAWy=c|nFp^6@M0#)6lJO^XDk%1X zx?&!pN1|}nArd=XVz?ZEE5^SHM~zG)PAgU5#JVC}dt8bWu5M76ehOzlUP1BnXim%B zgje2r;7#meR}5!EU`;uErb|+ym@DaWYi{eSABdiD0%{k2;CI(_YF@s8j!1Lo+`3ld zUUxrCZqB8Vr?ctpS!wF=7bm$%<#6rKg8r)KV7hN$?Dq=WR6C(MIU13ZR#W)21RC2I zKtbUvNX1B-+%=M)769wv(pPUCRRWS@-2AuIhcvwIoEK8Qtf z8Hfm$bYnf1@*+_sQ2?JAy~v1=q*srhqv`5-OiK`=84lv)F!?PK21Zb|AQi?ZE}<@8 zfMm5s@!Xc51dj@n#I+U(guTE8oB33uH=m|SPoR&~k12PqVv&qJh;X7_m`>pxJPg!eJBo<_1#pi&Sb&KSHUGj7c+6iQcL|LUPC~EZ&ui@ed#3Olk)<>)*v& zgXQ?F8-cZ=9a!8pghpJpkfD;Z0>>&&S#{WWe&Y18mY4LTSoLR$8hV9<2}w`L1q>TU-0$=tqfwF`YQ z3GW`4%kE&oUn>JF5vqMGL`FIAl4339@ zhaX-}iG=pAP(0ihIvH!FK){ZD~d9yz#QJYV$(Yn<9OP(foiPI%XF2D3V z4Gv4%ut`aP(u8$LWTGvpFCR}SOK0N5oE>l;4S-p67+ycNM_=1d_DzqE9?V=xb2(n5 zuS%Zk*7DP@*Cp64okO1|43Vn;Q*w;2pvx;RkXFrFYK#@6j7hU-t>^?g(mM#rqhYp}aNk@K_faclB0G`MFpZ2nB&0Y!{a_Yle_C7 zJHmPW4euepUYMTTo<^6_p2LPa4*cs-1P(@WOn@twJ>kB$=aHIl{dx>hZuvO&C==6l z(s1G88LV08fvwSTtW@|GdwTOP>-Uw$u;UcWc74JAzB&iTQ-fGoD@=Bc(?}{>lVsj? zBK&P8T5t1F%p6gA&Kpk|zu$9t#3DSp+6xrSAV=MWG?(LU2K)@@+R+E7{N~5_GR8wX zSrysiuCa=xYuTTZ{>)}E=L05<#gV(xc&BcOMhQKRBO7M+PV;fklw(S-MPo)A;PKCOP#?O8W=;$7`8kI2r%fW;l)IQU{WFdIW5ye>v*js}1rIXXJj3!9 z8dx>R_VepwlpV*#bj~ZdV$XfnF{q!eib>Un@W?F$iILaQo?V6mV|T%TyrE;#f;qCE zkr?#?7Diw3zBV5rsmD1YBa@lk7t__(fz6bk^OHmC~zI3@fMiPX%oBkE+ca0S?r1p#j?Xdf9rls zT-*s)u8)nj5vD`SZXuyi8*09C%h8dc}A*frtoxkN2nyXYBP*OWNgmoN6s<{G}f8o=lmdiDBTN1+i%Rj_XHDJ*vM)( zNU+J}9}TY5PcoQjU!lLjLKg7{mZH6W89wa{!uVuI*cT|{cU?0}{G^IET@&%4Vm;PP zwSZYd5A*nKfXT+2pxkMTb#-nKbw2>L<`ozWsDCT!9#g`e*s84akn0y(O8PCRC8D9t_XCvGp z7UFLr5YFWeetRv!t)r1}a7hGjd?0#Fcfevu3+txy;pJUBtbVGBFa6TE^}U-_7+qmI z4R>PYn@n_ZnQOa9fzPFq-s_eSW+E;<5W7+Pw~~6pYyeNAW1(Ga`m~q_)X| zoYzjJ`y-kp_9K8PcRgbQVPo(|S{#-m-`Kbf;cSbkGPIw%;hj%7 z)Q@uKi07=tOD{Rhu{eRYdk+zua~opg6JTM!2a!_%!HP_XJ^u)gXJbjnLx!5ROd}yp zacY*;Lfg|&lx`}-$m<(m<1V1hJrKir=V2M#h}kM5IHWrO%lA$A#lIeb*OCx=p$3kv z7cf=iCY}`Ez=6DIu21vD+l)tu{Md;@2fMM*IT|7=Pnqr>KN>8#L|+{0$gj4R>^CQp z*!sWJ_lRUNSW=SnE5%j4qt%WT*3Al$bcMe`Ry60?}d@z1j;w0IolNvy%U z4dFN+9EsGK**IO40r&qOFFJ1zcnVcmAM_dq+j`-2_&X}MT!wd=6L|Z*pwOL7kIVY&67!a$s~t@! zsZK_4VJlN^nTDi#Cq#en!+}@p5qou*&E?i@nFrnMTkcC1ctsVi0v0geBZE1!rXj4> z3>@jmuI;_doO4^)uA^6(QO|wKbCcm+c<`IzbUNtfkB20?B%c~J{^9<5P5Rs;PA12C zFl+b+u1(9s!aWN3W$%u1&Ufs#$%b0{0bCtXMf<*0Fg}zEo4zzIo6YH|GP#(N&Gl7Q z^|1DtNUu4cYn^})Da${`3x@z4dld)wtN9$anF&5FSNBuy9M1jZ`mesZB=*jkd#^W> zq=+@?_X?8Ps#f^-Z=+WyD(J=4JaYLPNRo38(Kfb}-aPh!U)^gQl{t^+r-SgV{S-R? zIzn8~oT|2k&~VZw61QGLj#biT-$BN8Z%h#u;p!StOQ0y76aT__ghmbJPBf@dw5eb zZt_0AW<=)9moNYc=vz z9FRQT6qjZT!bQIg>*8hTv!EuYoheX6f&h(kxQ=gv6|6@^34yc3!5wnX^*(?e8;`Hy zZ8%Y&LNL|NvwBx8Dq z%8M`1EVUAvq8CJWH%j5}DXvE9+6^%&afsQ*vBh@cY)a_{dESxP3Ggv$!5{r_Wf7m~-~Kng+6q*FiEY4nf*3@bTdIrHrkJ zNR3Bonj7lIJHS9v0O~E=3<_yU62GlLtH;SwpL-+hDF*|m9LVBrKBMl)_XBUC6GZEbFk&Ks zV!4y}x%VL-p zSo55#)p!H{#CZQwUy|FY4$N@Ursr$KXpm<_f{t^@Cc78!k6mHYdsXm4;1m1Bp0hA5 zK0G*6%i@%TV9C`f_Ich|dMpwJ$KA2HYaBv{UPF-6;$v!*>F*XX`k5t6&n(-p_sA?- zzdMxFxfv>t)92B*R0EpSD^7P)!Vvr?4zoRzFwAiawp&(VtG+BeZ}XEmmzCKis!tb1 z^oZ?HqeJu+@8*UhW#}Zw_iw}gj}EAl*oBQn8*%Xa9CkWAilwAK<$TlAjGM2(_Hdxu zDw9%V>>ESg3Nkc$xfg<$o?@u$ECwb&VQ;p}qFPx6-KpY;DX~S`nw9Vqk3fOQJ&1aY zBYBQRliSN>w$7eKsL&H;P$Z3698*23h~vY47qD|pj7c4^#e&gzlt8R28cp=4Qm{8pMhh5% z7zFv9_n^KEWRl`c3m%1$&iNEddANZ@U8~6Ria4({O@_z6YX&d$xf1WgmL?jkuz>7n zDq>BpVoO5`#LbhCS>(d`tszvkxPol6S}02VAtm@1Q~LD7w6xF%BJ-tH75q=QPoi;}J_&)u+X&>i^KlmUv0Uv2lw%WM zTUAKz|Mk(M>EG$8Y!6raR!~;NVQNd&MP2+RyxqDI`brBp22Bbxs{gR1TOJ~Bq6#gZ zF@;jhjLGiRGWxe&ly0e(VcCNgWcT)BReK96E1O|Z5RF7Czy=i|^11j3Yn?k0YxoM{ zGAD6`iv`*Jb0Uuce+q82BW+DBT6F#y4(~n#9Xny-bJC#OCORZE!;v)k)oHCk9&>)x z&NfcF%3#KaxuyS@#?!MbWlj(h;>vJeC<8_%XOZ&d0=(}0->1^1DYFmJ#nyck>&a>A zCw(c`*OcNPhp@xp#Vo$}9y339nYk}ZU?;b0F|iHp951I#Z%dcZUu&Y~>k7ncE`XRr z3r_44p}nu(;Kz@%u>b#*Url{9zu|o3oJ$zO8_@IzSlFHp(dpWd>)3?Kwq%Sn9YUu> zF2v@hVb4CUFLeuMyl08*dR`P;$K2TUF;2`Na6OwXu1vp&jJf=+BBiHJq!oKPMxnV1 z-4Cx~f|3*ke;UQgGgY{d6_2NJ-824bp!dO)8GM_7z5K`V z?OQdX&(=ZYaw>E*Rv~(cEqgval;y80W!}4!+4PJk_E}YuJ?HeIR*6Fxdovp&>x(h{ z2*C&3s{)9b@f6P229NEMC@iZK4LW}a{DWh^Ct+5oQiK5&rhRz}8vCtUFt2sV$ z@+po#;y7|4Ii#FOU~w0^nM0!}?nwP*D>n$$ex) zeqy*ZhL6*)>R60^0vouSM+*=0^S<omx>>>!@QWPMl!oLJB}4x&SHFN$5})3 zHumbkcZ0#p8Z2$hK{l>3nBDI@&QdC-u+WWhY+}npCKZ~?9GeQ5E`KE(c5`LdbGESw zN*OHe+hx`9S?7yPrbB^jtqWsvhRd1f3n^wD7Q*Spm!PSf$Mt^^ zNcG5p*$Wfw;qsoH4eF$stWFCTYSP)KIy8i_L=A zBsUqK4orshLM22+HL$QR5eDC%js^@Q75+$^;g$& z*#QsuJePxtyE7|28qV$(Z)I7#=de*{JyxwC#~Q1}5&b;?`{zf&@w+=jY}GI;GKNJj zS_`EOSMj317muIZLTuw&*c3QGYeoTWkrCv@+#SaY@c2Lxdbg;N>*pSfImxPGJ6MfX zJ+u1L$S%xlVmt1vV%eJlW3?m6-7>^xYBFze3Ooyr~iYNvnQS`_duerJsxv)<7RU%2fJAXD%IAQK4%F+ zT-M|Aep{GlKVu(??G65g|24Sd_Sm4l@2tUc`!Vdm3@?M~v-oBg``y+ z8&xZXRm+muM*i2V%0L5BrB;~5@z2&#t6C0!>o`59QRI-@$m5H-Ok^h)Zt~w%9)*&$GFl;itF=oj&ZeHCu9If60X(JnWUQy@L zY#O+>u8zHQYGEf{Yhvb9PmU+@fc^I!T-HMdwJ%)R>&_S4yEMp_&icwyJY>;#=NL-~ zG~w3nd@L?##3zYMF#VE-BX`U&ZsRwWvr7-*<&xOX&AzD$nuRo%ckE5cE~fW9oL&C8 zpRHyA?BOmycA&+8Ju?);BahAa*|HkE79iA28x~;=%-LatsXubSpVBxiu04lKyLR$u9$X)p=c5XS#%HrEuo?R&`Nl0Z4UmMw` z)oH9;t&zE#sA4Lwm1PUoFfVa_9QAz7@*n?Y|M@RL(7J`VE+&JpJ5o4vUI=bF8{p6R zzqa%8ki+RWZ*%x*@v2DtIQNB3;${o|TCoc$T<&83icox&*M_0{9+;=xM#JbW2rh_& z?4?W`yM7Qui*;ao)D;OPC6sGBGTan1!=vgxh zpylxyrSqFGWpX?I>0HM+s~wp7I+lIZd%|vP?O-J*^Vx!`(@Z606Ps7L3X2Pq@psW# zs9k#uuc2!EGn$8VDIRVfbH%sKe(1U2jk@OvcwBD+K`$4So18-TlN0cN9?0zzSMiX0 z#~*phK?V8bUNFNWP7gO|;Oc`8eR}woQNs;w zn#$LKFUM~ioNBRPI|632lqLa|bClDkEw&k$OKoS{EzU68b-66y*8`?_u#A;VGGn8q z6KGI-B^k)uQ^#Ey5^Z>cZAuCxzsrfGt!f1+2sD&7oMtpA6%>43%d>>EEuG^f~<($t!*%Q{^wT z_)ZqpN=~Pa)f|Uqv7ZXt4^XK2QL;;2KtGD!GQ-r_cu-`9jFUW^(lEdTnQrDODvqnJ zM_|@=5c31V(R=C)Qvdkk){p91jVP#$=s=e=kFZ~KG z^z_0(@;ct?Nz#g$cJxVZC#Cssr5Q4tXn2_}^@<9U_OBH*+8IhUa_eYpkv2U!(Fqey z8|R;S52pvNU{&!$jPS~l>T(zlJmu&}=teTVwu|lucv6I)J+Pu9 zLRWVe%+LLSkYWq&R>wp4{v6(lgGs#Q^9p(ILmPRGdEa;;m8Cpy;WWg_{DkM-LDWtY zqO{esXn!D=5AdCg9t(5)%<+aq#~yg-u0xLXd)DvfhMNA1csr{TtK3CtVBs5_7UuS8 zJK$1T?u&91Hja##W zHqG5gYidmCe-xc}T+Qzr##`ExB25(yndiCgQ<9yLq*Np%S)npBGFuu-rAZ5wlpmM)9InU?w-1l|8-%8rD{4-!Q^VLpL{_hwW`v=gm8v)dD z)slXD?_%@LiXzqj3p4#;!0-D?m^b}8v+7D?qrF}+>UzV@o)X8v14V4E?P1?Tr{MYZ zgUC+I!_1f0U=eo~FCR?j%AS2_l*)ka?HY*aJ;IPh$`q9&LATf+rjl=n16CVQ*vT0? z3zwohR}N|wDNu`Q!&-3W>tHRK{9K(Ht3TuYi|J$|8%s+2BSX9<=^`ch&)?04FGCZj5clvhd^^FEJVy%ES>?$*cD zB>~9v_ec3Yo-N_K>Ypy7Fk3Yp!IJN=w*C)tAN6AQ4Hf!W&fNkE$fGSk2LScwjr{r6gRGnqQiCq znmSjV=7cDbN?)6y8ijyS4 z!CR;fJ%zlZ5An!_a}2|;WBaLWtoa-XkCO`Q!b&q{_H-iKT)L3G8WF?{_@S}pl>`OV zno;_iIpmfYNagXKba$sJwcpT!z0Vq$N4r2EIv@SPH?c2156d+!GwTH^*k1=Y+ONWx z(RO(AUJKjw8rZ!OF?8~BZfD~*=4fxnmb#5)DbJ*_Mt?axIa|oKbsxUU?8BO($v9ay z7GIm<5U?NvE>g`fo;irNXr-cjg+8qIIwP3RXZ35RB2#Y;*6NK%l3EjsA16Rl@NCZg zT7hUAZOnC+gHyE_X?R(arLR3{x*sM(K|VD<-9lOyyV<3Nh3JqxhTxl5P`5V$X##iX zx>*4MyvshH&pD3Y!N_x8(c7E?`-MBO#4jCs`Fzi_= z^F#+-7d7#9-bdR8qj_SH5^><&R6=yRb}--{Jl_ z@qMZidDUCc&d4=1kg}1qE>EMcpT)`Za6PglRmj%q7e3f^L(B0pdhPi0HClphSWloY zyPfEp%W9f7e<8K13?b#=o3X+4CKfnfMX3C3_{G11+@oFaT0Ea~JI}Bq($|^%y2tFv zt?SHOb!%IZYxec|(P@J?5 zhVQ#)80)D(3g3BO`@&-S(|3q+qImc3(Mejosfnbzf74G{4dLzD$F%>p5;9*oV58Sg zj68n?Eo1g$p_4f_hfJqyIw|yf@)g?s>@i8(*OQOWLE0nJ%f1DBz;^XY%&wgS&(Yfu zsJ;jeBgGJ~EEuB`_G6{(QXCWu#n*tzSZJP!DUNNJT=fIWqtvKq2=8j%xQ8K1e$YPC zjjfz1_UF2gO3&HQ#C4XmXxnb2_Ell$WI0k=IhH0)H=$KOzd)h)7?c_~cR_yyb%%^3 ztM9V3H})dt51j(7qC!M!oI+(pHI!HLXP;33X8n>RzcqH$uWLhp-_4@#IqOLOg9&Z< z6H4jIFX=>@jL<7kMyOxaMOoG1jlKXIrq;6%?V`(K4ySr|xQ&kCNiunfYPqi#!gV*4P%q67U$%pZSN_2M| zLfblZvbwpIf?IdcfZb*iwC$sm#YA&XM`A|Q16=ixCXH`n=x(4Y&AD(M%r}>LUDQQ! z=K@4p?!&TODF`p{K)7}jn=1O7J#3F>4!2!c>31Q!y#K4f)afW~Nxwyfk>@C>DVAE7 zB$1nt~dQ4x~5`7>qzX-MBNf$yECEVDZl zTlkD{vRW%nKwA2_7>K)6fmsB${!aMxINioJxOs#CL5Ep3I}md(vpX>v6j4=1SHwlj+p0 zm>Q&+`P~hR=4G_j_Y3&q2gHGJcm@MP7GKliQ|j`g%H=zNh(<;hmARFfh77U#p2l8q$OL!$@!M1kN>3qD5sNQEgBRdAB5l&*3xV zH!W~K!)E|qN@R24KKiq55VqT1IBMl7VcYO>VMy71p{U3UVRl=bu(!Ds^Va@_e7`tZ zi)c}DyCSX4`hds6w@gZF5afm$;EA3b-rIj*O|Pw)yD^^)9De|>z@Kn&>%-#QDp;sT zA2W1{yCON;}v1(5A@?Nkn=D^{Gpdb@6ew&7zDo>)v3+8aLUv zSC`nPo{j8DP%<-qKEQUlDB$YP!EoU{weeH3S=IIvaQ7ZWUHm)|`^|sX79TKWc`&xG z{lL7A$sxf~6Q9?tg7uPgbZ*Z=mF-k`&xwHYjUb%sS%RB0H2GaUgN^JB$0wede6get zeY0O8nKdGHS_oF~y~ai5{p2>_PCb3|NF>LYp6&aFi}?@O5ZND0*>o&=SJ>cJwlnlI zB%v^H2TQk!(;zz&GBH|8t#+4FUItm&@=grY~8*60By9vzJ3`-_0fSxDqN zhfCk*Vc@vTkeY2#k=RgqjMJt|>-Mivym^nWIH(Ek^PV@|R32 zT&Wzx^W*0DmLtvV&IGg5`TqPq8_yDhyqIy8Cez+D6iowb5Ix)%D`^qW6s*SbtbeSm zPYl!c%|&~R4Qdt=?%#^X(O>Sk_rjko3>nRP082pE&2ZIkFxZe1c3(N0(R2w6)k$W{ znNAmp~xE4QzRoOEbeyzjk`(fDFeuW(hlEmbV6YzPa9->O#vAAg$S?6j|$`-ev zGr9q^JUEN=e)29)>T(*fjo(=nPeSSPO;}4_#oI}#kh~j>%fqDTPu5_nxNAu>WzA`x zi8A?$AI7qpVniP3#f`-S-Sqo-DjS3kjWO8S6h$>X7f3PbCQbXVf)bzjlOyjc zitw)girO-qG0uhKB~LWmvqf9L5Znt@hw1?VLe9;F*C!LsVsB%%W`V5d_-rInIF8x- zLW_4PCl@(GZ1DnwZeNb=UHtFwD23pD9=0@|z~002kn*Pn4&IV9ewh)Km8_+yDn}`| zJeYbfhzVtbEre7#@@^f`O?Unq(%z?~)N>TE4S$3vEcX zHG_0tX3&Qv>Ese(4Z$>-DP@^TxJ z*n1iGYBVXRcOCtwv66!CttJ4(_TX-9d=z55m`I3M!S6;(2p*QO!q`sC{(MoPSMkkq`LvK&Jx zZG{u{^PQu|U%qcIKM1+>V`!~@f`hZ=X?m4DxfC~GQ2uFFdGj{Q-`T>B@4CvSUTtKr zRAbmHK4XUC7mhW66?L*?Y`JDLC4BhEK5CDCpv zYB27EvebL18Jt7ci9F_;IS5X>HL&EG1_m8zXHtrOY>L4yXf$VIrN|ZZ{QiOaOZs86 zBpP*N+JVhuXq}4{B`BHGacL2H{wx-=7bqibsTHKo>cYKr18Syk!c{A|Jj%{J5)_UZF}OId|yS`di3) zeIECGG_Y#Tc=Af#L8deJP@3a%I`v`(rIx(Iyap92dl*8`64L2O&INMe3^_fo#k9Jc zVPRkdN)GY4-dI20#SMUg+7zVt{6JEX9-VI=PEA+ksrunRyooGEXyaA-`AAaeeftl+ zEUu)!$SNurafCFhMv_s86Rn#$io!~yNHA240!KW>20o|L9;`?bPSa_{aW_)r9D$?! zef-V2IH9_?sM+DiVrHEwqGs8@q|F{SToM`&;k}LWC8WTjN#cAq{j`jxd-G?Kpji%A zMOVSKa}6q&E<;D$8l)y^!FG>09ao)9)oZqqYJUJ_3|c`E3;CV0r-%vE#4+1Sj=OXW zuy55U$muJh_|HZ3ob1Ne+y7x^l{j5{D@Ja2|KATCPvt%SH2nJk603D0cO8D8`A~`3 zzmw?sh+b0KD=M6s_?p%~D4>7wW%DyDF+OmK(p8w6} z88@?yQu)mHZ!_~ce}+AJKLdxv12E>=8e9>E;i#-H{66Wya;6i^Gd-aHd>a%Ek0Dk1 zAw&cDdFnX?uQ{jhp_e~CO`8p;FB7n9eFu9JcNaMi1hjCt-DHQ_{)#g)j>z7uoX(^=W! zX7*X3gH?e|=^R(P1e55hK*|iAMeL#y^?X>)=QgYPzRM5e_54v>$a9tyi`yy- z(6CO7G6Zw!^k<^M+g0fFiQm}r@IGdhhvC_dT71&V!jL7Im~6|Lq#t=lBd1v~wNRTg z%3WFd>JYYZ)k?Pg_HH&(Yz7?}#+e(w{xolu3z^=qqcaKrpv5@@|L5fFdy#-Y=aaGj zT@ARZ4d1HQBGBwScJbcP8rL>Bnbg3q$^wJiDp<^VWr%Lx$`bo**s6EqnQvFA;PWWX zrWj_ACs+A9jAJzC+zRpEO3vWl9fk&39${><}{wq5CibMa3NL| z&4>J?!>ObjCi&SiA z{|6b|nz{y6w|vp|i94f8!cg(Z6xZIDF^|9XtZC&*#t9}&Jgt&>c{sCQoe^;P+Ke}W z7vQoW0iMlgFtSJwcBAER<(4bP>(0cTRs46Z_K9T-DPVuDU5BWb2$@Y6BkGi*-dqXt zURQ*~0YyAIx(Uid95Hg=3e-pjB6_(aI&v+jx;ma7>mMU+-obU;CZsWa^;p)@iDb^X zs94bqqwBeNxh??0>zkpNorSgtQIbegrHK#!;-=_hxMy#Hs?0?g@fpC11M;-~@pH~; zy@gVZL*PUlwrHt7cB#+A$@Mdl>#2i9kE_|_*jS8uehyb=W#RhrG(5@S*=GIi`12wM zpUR|YQi?3OJh_XZ&dKlt8@WAOJnrDncyv|{=O{pTSl!sJ%`v%3zUZu%; ze&n4p6-LvNaJG(fI!;Do=BrR_o-q`AYmdXIyBXSYpYWeSH9YN7p?Sat*@ZXwzEql? z{i}w|zDqcGHV;=VPH^t17{#=$ph3);zJ#o$PiuWidi5-N#&^EvyW6n*S_c+oN|9p9 zFlxKefUx^zUh7D~_UW7X zEb$J{MAssH%VX^Pn2Y6t&5+)EnI0woM-d9YNh9thJs5SHtUPzn7U5ACj(dVwnOc~; z2jOYhKKKSsgR5W^y>>iJZ?wQXjdGTtyPPMm!fO!8^@q{o z4e50GS006}j3?bnXIiuDyC8vAd#37}u!~Kr*qiMREN`kT(|J^ari*`YR=f^O@EjTH zRao#d5?SNVlZ%$LaLovLp^mMDFmP=f6)VP4mwE_u8~TC0a1ckfL^oR^o5PB}S+KLi zE*R z7n@mq)8gyb>GAbU3J8y+vd4jxHPwoCW@p3b&>JkhT8^sld^lfALr(MtynJoO3PKh! zGoCM*B)^6&88Bw$hpq^^9a`AuSl+EUpo?jVj@a001&5kX%<18EwqDmD#Hik1-7WSBavlHI*WG}AqZ$`@>- z5sU2TMdM(a)Kvl}7jM{?WkctDCt^N}Z5n4SmkmC5A^w%S0(rzrp(~s}q zEPMd7J0}sD7J&d=o|m)hXSI9}wn1wvo-%I)@;mks>vp!qhY9z5ixGMxpAv?Yo)PN4 zsSuio_zTDJzUJf!bujqU$M@X57`{W2WJY(Q&h;rvdC1_TFb`S$KHmCtC}gGEn9{FH z*chQsf3;@OvTu%58l*wbct_~C+6`7EDTa?@)$#n5DnyDq*x83UtU|;Y<%i3#xU>)| zBB?M7ib4Gipz>KOq?*2B-^ov4H~+(qhIXj0-UU+-ed!IL?6N?55$DRCQjXMHEKk=m zFEWv)N_KWmK2wTJWQR@PFolMrEO?Ox`^)^<)yP9^?#onW@?Q+w6zt1XD(0fhi8BGi zZbL)vIJ9sO9p5Ff@<`FjAxb=mmjmdbhKQTU+rhuto;MbA@1?f_6njeI}5 z;&_||CRgEafiC&!tI+XRoNK(~Hom-zKvHcT#EjJ_Heon%TLL*99ZZMb-Gu23GkiWC zg^wk9a7G<=+P#6gS}686l|y=j2pyU!piTGY(TzjnXywR%FsqUlR$refoL@9o7*b*) zj2bsY_%5T49GWvA`erzl8xxuR8$oyOw_ zKbZ7fIm2|%am<`cn6x<-dtXK)VnZ18X1k&P?-TZLxjzm#T|iW09zKX>V{b?}hWT=D z=qU+g@O(ys@@$NjGlb;^8QkWMOp$Yif{SM@_`b-1UCekd7!jbyl%`6s9cxcRQ>6x{ zotrU2tR1tKWaHgc3-r9UV;Ye+Sc!=k^js9t#hKy;-R9`C*~@~z-(XJK?d(9D90rzL zVye}LS?#D4w#B!RwI=N2|F45t!$6_HLEBQWc4`INUYNzaM!aF0LZ$Fl`!-WQ8_SkT zSwXk&8j52@C@M;YN@wu?N*4FoRQ52P;lnx0Ss6|4!?C#63Z;?~STsQk#zlLeeuaA^ z?N?)+rVvj~d}C`PKjT=>Oj5gZgw}avla1a9iZ!vKIbS);A%dShlfI+VfGb6lApO231`lF8aq-d+s^m+9vwvj}Bw>!N^`mj$+5inT#t3Us#I&JzVHaXd zA_t~W>eo3mwcUnl!=-7>h>hr)Q;(&a_g`EWOO`|x}GA?VrpA^un(#ueFP6L(`UiyrhnY((zLeC$e2gTyH}sO4>; zNYx^ecdVy=g~w!Qo=W!W7Syp_3@u-EAZulcq2IKTy|02LG`X>kj6#&ka3)jmEV|&g zg^s=tpxMfdx~CmOzVIb}f9XVzzXVC|`VSAEMj&p87fPJ3!2b6&jBc)h=vCS@arqN+vGxd={c-^#cwvMEt(y?I*vW$ zyp6@)*RW(wC_YXJLf#QGY)#>e^xiN;^JOsEUygtK&Y;yH4B}62 zW4}frb~InX%kDb7YmDHmhc!%cWj-6{KMYfh9HG742WouQV({n)Oq6cJMu+bXCne%r zVI3Adje-`Rd9LriNX+yBl?>dY&U+upJ}iym!g3JnElMJ%|AXi;O=>(om%hIEf+3%V zurCu8SzpdZmMcAzS&Cb-q_?tc)aGv_J8z6o>#w;`y3b8G;z^)zmASbvlBVH+W)KYK zb5<{FMa%K=bT(L(mTYifx6WN**LRws<_-6&ntAe!$9Fc4cShG|jD|9Ia*gP5f@tU# z{59`m4(i%$jgBAt5SGs_Zad3vjk(Lt)g5K}BD>gxz0s^M$d?r*H!y8>k9j$8SJOdf zT%W<2@jpP%8acCGsKeRE?_hq}0dG#HVXIp*Dh`C;qeuZ#FCT`UUZ5bQ)tE_K31K=J zC)g#4Qa07vk7@cm5=f>DVY?nY63op}WedD#F+t)B!8!#a)^48Gq@u4&hq zwqrDB?fS7`qb#;(Z#SE?Y&hD_Swpy46<>r+0)-!;?8xp6mig`k3)jhJbv0|*+dU=h zeUF*eMS|$8w|x2V+;+ zc4+uz;=*$|tg_|*u2EZ1Vmb|S8*I?g!Dl>1s*ntdqpwezsdU_XdX)8tBKCfu-0<$|`*NrIwD&jhCXUkd(|*|RYlgW0At#ezL8e_4&0 z3-{`tfbHausGIT!-ztLP`B7AG$>gdaZO9mb#fO=Kfs{#t&ue@IYO6h1ny*Sn(tYx!i%*Q;HMW{d8#?&-ZnDB@=?hNPmuH+1~9IHUwDW1J8 z%E8B|$w&(R(y z?Nh1+rK$@AMsA}8Vt1H8Vd1ZdZfQ2SeSRfs^!?zu*cS_UCT{!Tp{U$6QJ}Nxf#CAa zp=@ZQ3mZFN#};uX`QnpNf{1-eY{67R!LI}z!L?my1sz^Nf>VcYvB970VOO{d##!N5 zC!L5tOQKL*?k{*HZpYT;J25xaN$j_gHM{L$%{#=rck$GdwWUvGmP3=cEA%<@J~@JQ z=$_|ibScwR)yBM=0(9gZM}lr3)(36Jwwo8Rep45WzCXeF&zo^*mH^j;@({o31lN;W`kAgnD6RF_VT?b zR`pb}3`KcCkxzi&{G58hewSuJ=%;IfNA((P?$cV9v2!R|CYeKf;znrwU4#IU3#{bI zTUKqh8doRpMtA)N$aHh=_^H=SxOxgINAqWu>T3Q@R))`x>hyC!lrH#ZAt<2`JzHvF zHBFP8EkPpfy?EcCz>c-&3ZgXf1nbYv5_n8#5R}B65}*}U5KOnpuZ%NI>!;t&3@&yN;bWH6{9p_znTXThiX7ZcTggo1 zs@T|?Nsz5G!_By6mN9G+OTtZvyLRF}XFtTuc?hRw71GS)ea$yzIFa)N>JB|{XzRv$ zpL;xuv;%OgK4QS=e(Ws=rOqE-#4rV)bl6czvd|jG#G~1>*knR zI}Y(vLs&H(XSbG!qqFrptDJwF#kD#!#f0aAuX?jtM!y?7QOF(4@?ET;I+9JbuVIyW zqA(fM$^JxkvIR;{S@ZBf_Tp(St>5s2zSzGeCVHI;{w9+|#YT#AFr#YeKvIp}Ks%-o zjTxs;S1*-9{qcKNHgX{zI(wj{YZEpFt6=o9+pKAN8*@-_K)gjHmUxz-_HZj+E9T%N z@6YHhx&!-C397qQkIus<(DlF;<5!2_R-Xzn!@)Fp=P;T!R+1bJRY7Tn6}iq$qOr$L zQ_|mfKAT)Zg4!NTDexriO{pY%_%PjpJBjq`(;Ky)*sN#G`7}`w??^(Js~`TZnuQ`W zT@3XUl8-583jEqYt=8*EMaGh(C%wUFi7;5VDbV|20?xW!My0l^sYOSHx`N#BV?!ab z)bf$tP=am4d=ZskggJ@|q z^d|~)M-~cS*xeFdK7LhL%K4%Ji_c)YswBnV>cM5{F36j4&bk86Jtm|wWjj&$q>hGt zhYRSM8cw90WU-mMF#O*)zS|KY-P)f>|Ivbdd>$+E{ut}tEQ{{QAt*PVfm4;{7&k>7 z8I8%_(JhL8)GBHpRXp^$h$-1 z2ES(qxo=EqW-PpOFX6LI8Xj$*gtz>;th2R&HH0eRv)N>nJoZB2mU6@=Mnh$VG3qu3 zVC2YXTv`wfxAbWklzX4Obhu+HwQ3%qXA?FO z#hUZXm@;)_b+XGjTM<=LhAv?P&vWK4^+z+=NY#=JV+@Hb(F{Q^vmg zfz53m1&d*6sDF1B59Y;lU)Fs5<1?MgIbXPw+@7K?E}%UHV`wJN0v3eaMP9=&;f7J( zLaoaOgp;#Y2uG9~3lDcZrqJ_Cc}~6nnnCRt;{F0bI$N>+kPwgOpJ%6*y<{U|uQBuJ z_sphB4DkzZvL6cF>`CuJSO*`$!kv+jUKfJq$Lf$s$iTU?J-Bjo1PM%5kaoHog?!W? zhk4`3B-4=;(@kjJfIJQP@(yAn&mhz74;)UpQFglr&m+3hYkn90C3O*1jojIBV+g&s z)203soT0Aw2dma6VWWWp1fw=%_U+^7y6J~v`GY8G-+>suADFUc5li=4%62@~GW zJSh%>@wjQoS=q{rI($)LmWPU&webFM2GS#!AeVQJ9l^O+5w4h!xD%;C+-Vcz$vrCsm^sj&HJTI;zw5E-sT>>oG^$Y zvj5=z_3s$P+0n0^mtxd2?pwEcjMl3k@kO@@!A9Y@*Sj9y=kguD-38pTK91wh_}+Jm zH|MI&$CIUjDEt`;Scbx1WG)KNd|;Rti}k6ToiCz7J?2v6q2COTF5W>KF$U*Kj{=6A zT{-O(??heTXQUtJc`MTW-OEYhoed4HGp6?0-|*#b4kA~5!;l1BvJCwS`Q34-p%uux z^^EPQlBWglS5lAmE}D3G3q^-aq7U4MvKr6g$;OU%=6@1 zqH(Qk2hts~u-MBNjT-*&e!-bmzrxUQpXaS_B!X7C!fIO`zO?ZBbxak0PN+v`)RG}8hT^O zNV}^X@BX;c1rc?MEOG!c@5BE8O%5nC}8F#T+h3QsctXeo0N%aw*}aE<}I@y zy8~%yrD%Wl028&YVTWiC*1QyhwTT0ys)V@Is1F@CQA{tXWQ%RA=x05 z${DmV6_7Y+2DtKWm*NRp-jzf*2U1A9(u#6is`*|MSW@?($j_qS7+ciIq1cA~;MS!dy>OHslt7S+OT*XP2Wm!pMCRxU&F&y#3uzk|ZO z7l@BiqDkVdu)2JneQD6au6}>)bxlXW(?h7aZV0dS(=q);2+sWtL#lHSevOSl&S)$A zcSVAo;Ev~ECt{f)@0d19m#~O?vFxLFG!!PA&~D)}+I}*WbS*>4`13T<9G;F;Nd?*< zVoJZVhtP$i&yg8iiBR5;a<$;Di$KoX2~EVWj$EiYOor{2ODs%69b+Z?SW3!l1VM&TKVk$XIcDQqyzt_(;;|01*iZ2oqKr_h9{ojpSU!cwqC}C+9(7+9RznX3H+xo zgT4y6}KRi2Ei>&t^q?1<;Q(To7m0oe95?NFFc=I=N9KyRdQhaFVpr=Fn9UAC*U=nwC!WPq$D`c6 zrjQDH7d zLthAZCqthcXFS57bxK%}FkG0avQ>EhhM&;t>26`>DqZ399oNZnu{r+ZU0TWO;`E?j zi`=uMXpdJF#(x)Mb6$9`-*#a6wIi8fqdluMHes6k>evJub69TV%rPrlu;I?=cH#Gn zNI(hC29L9(XmP6QP4J<^The#UC~gL%)dg?B_|sL@1|KN!QglTWv$LoB!zQ4ZV%SH@i@Z?EB5 zQVjARiLoQ6S2OIr&1TlVW%-8h*qNU**~6IG2;ZNDTiVz0%(4Vu^P?f5rj2A-DX{u! zXuL55cJ|s>V&IDE%dW5<%e&{}+nI=O8(X;67!!2|qsR3M>-P6!KO-X8-HYA=<$uhHKTF*)-)OL=XIyCJ?9wQ)Y|GG}6b%phbXi{RNSD>N@yjB%CWxH{ng4otIw zx3>xE?gt^6cQAVX?7%JeBN(o+5<0Ql*z@X}tc>p_=S{xCDkLwl_OM{4b><5i_!(gl zw~BfT18Gb}0Il3MpAt@3!-DU{uU>nAq8rr+9k_y;ZQPGuABioul<45o@uYN9gBH7s zP?OU-47GL#=4Qd8jn4pi&g~bEin%supkQGQ3;idLeVZquoQ=g#VIS+-na`G|gu?zs z5B#eYX?rQ?QoJtJ-9CxvLmtTF?*hM9UPi#+t2nac9!^|53%}#>&^*(RBPSamS@jDp z0fjg;%^b~By3oh{-rWtnpYt>WhmYGpX4n>3y&6J4qPCLu*eE*7_i!VY>(XQHtGm@P z3_@i;NORxE$-tG|CA$%QF2fO;Jd%Yb!T)fQh9H!{D&3Fh6(% zwvK-Ao3aY(^{S{z3C0hR+sNU5ip0&X@Z*9c@mn$xG>QIrosLC2&vb zYMOYqigUqu?@o-<`-X2KbrYV?ddR)pX58b~qeEAD_c-rALptf%g}qIIc<}Zx1{uD= zSZc#k@n4w2_g3emFXNi|FjB9#quLMd26T~sKiaWm^)gy#6mOnHMqVNGFT6g6S%FW(F zv_PL0pH?EdWJy}m`5&ePhhuq597^o!aQDVTXw`_5C_mSC7hJ``LphMBB? z9h5BjcOgd?E}App-FOyeXFee*R-dZY2}$j#I2pTdVp7`0Y*@?{mbdF8Q~z+EX*;Ac z36c4H?|K5I0TEc$=#Tf_QK($L7SiMOsgniM{z5<6$2pgh4yp8h;TGBw-6~kP$b~7T z&1J=HGAw5AFM-UD6@v4e6FDU@0`{r7@Huu5-N9FpJjM*2R}a$$UoA87c*+t&$FY(x5-eOrlnR2lcYI(Uc^kV@mWd|WagLaUZU82ZeTLgff1#yX z1ltNfEMMV)|0JWC$Dzw?!p~Q%$@v8vYhA;pxJ9w&nWC6BFc0EyEqUhB5l(zQRo6ca zei01FLFWJZmMsR({Ir8bl@t~dW zMN>LiOO7}#Oqxs?4;^UCUrQQDHHrG_n6^y z-)7{`@qsbVk!&3qipL(@Ev&@Pf6E{6@>iv<1>a#^l7alJFoZ5{gY2s@)S%=aDr7PSS^c~5s-(s-%b?&m^8H^V*Fw1rsehb5~hBMb|4iutg#s&V~v=J(U zhY1HqItVZRbPy_R+Af?fA10J&+bfixH47qxB5^7!8Yfquf>&}5DqS)#z3M)5nfsA7 z6hCIhJZm!k3g<$Oi)UxkR^Z8}vq%iNftVrR(7&+(FXHyV>fmmcF_6zDPH$#@W(};P z^$BZOpUQ4~MPmNG`#5s51c9O5xJ>Q%)^ZeanVt~iyffL!b$C4^A9C91oC)m%JHr}` zX?}%IcF!?rngJ!1%FyRU!8l;|lEggfQz@HzgBZvDc3?Bky3@6)-H zY$HCL|HdB6D&fYYl~^`n9y)D1naOltwnKV2a`h&nP$U%HIcb;`Y=yhGnwb6l68w`M zLNW7As4m)wM#iae{#hQb`SIM!Luot`bTGHC@^E>khv=5C>|R?wzUwt(GSx$V^jj#m z{zkZR18xb6$+4r0u1l*3Z=Kc_9%%hSqkf&F@4vWPLzVAho~2;6Tn8?$%7XeZ39R6I zWuHAl>}}S>pLA2sKwW}*YZa_ro5hTkI~n;sXQMYt-~#XIh3Xz>g}OkAXdBtatLfUiHVlY#_#o zthHnR)CZNeqcDmmAwx$CailE(TCJ`y{Vj{)Eq=_Eb4Ope zUxz_UGt8?iaqvU}JX{HCTLM|WNG03K{gn#NQLLB5S%YJ*z_we4Kg-0}`sh(?Vy_Wn zU5^AaC*=z!luP}OqVtZ&@@>O-W*J!#k`Pj~JlApFR;g5^sjaj#N@d_NTvN=@1OnA`gorEzOM5;j_((v@@;ThYc?`n3})sKb|8w}dkl zjX`YY_BN57Vt?koGuscZ(?KXT5=WzxLzMjT0j||Lhd5ZC= z!4SU@$JRGCup5r`?5}$VQ)q2r(^eKTk$M&@3dmx|bh_9?c^TZ(>0}?mB3L#*%ilK| zO2Zo!Xp4tFEoxjwGr0@RXmlI1y*C&ZJDhQWzeD?kIapEkl6^n*j`ehXX8PRoeRcdq zOq(bL_Z8V}n{FR_Za5pta{ONP(Go_tYgqc5bIhXgFe?acXR6QNG9~@vZ0ACI7M`KW zCOq-Q{96fdDoWtpmeY7*o{!7#9kIY(3Zc!M>wkMT#P#CP6}Sm&KhDH9em2^@?K;L} z^`os)gK0e9J-l*Vge!a|-kpCPvZIpF@31}QzLJH`{B|}#XAwSg&Z}YR6GRQaf$v+n zOL4LW=i|L*dGBVx;n5Zheq+bIoqT`wvYyE|9AL*)xNE$vh+Uml&gO9rUG<5%Z1u2{ zOg={(?gdJS2$w?Fq`~mf{KIb8ZotwT;aK08&bx>F&%YOh*1a-tzy1T_Q8F~qS&U-k z#c0M^BP!{!pdWi2=%IBcb!(*1A%`3)$*CserRT|b4e$J01hTm4yP5m1JeGD~C!0F4 zQnX^$Fp|H)NF!r5S+s1X?6~Fh#j!tmNAxiIJrHEn!(#Z%?PXRb8*1*>Zv&O8q{x zcr1f`^JEOu5wPpHEan#)AVy~*<`~#wisPk{a{-o zc44SDzgt#2V~>k7W_r1zZ@z%0SxM-MyMO`zg5lP(9v4XoYbvHt5x+a{I~71q3u1Wx zcL!a}SEB5KPSGCQQ1*CmG%FoGoaNqV7Ug#z7x|t$j%&&>VX%F2rOr(wV$0(`ImX_V_M^diMXy8B;namcY){fdmZFo%8Nj<0^@_aNKUO{b>?X-owLk>c~408s3w%H`(b=xFDoA&!fx)LK(F7epxu?` zw11fby-(=E4N*Swa(P~=vlujAZ**WRWN&_CcG6mCs~`{_j&g-M@)wvwd>FUl}x?^ILPj= z$zMlspuunq z+qlq#N!_gzC9~n!Z@Uqt)ghQ~8w!oWaF{HZ2?0T{tSZH<4ehviKN}Lx!B}p$6549Z zX^c@U_hg6Dm2w)mF| z*Ne(QZ$x56MxOBwu#|C?mt*~?jDoWu3*)Rt(jwmDf?bIlbz&!{B_IEvyGF7 z!$3(4Cx4z0y^4Az3N3#n8dxO8Qf6wibyD`C^qOGYeP0Ok^RIED{1M{5ZNrM6!_YZL zmCQ#jqf2cQX@s#hpJhEm+5F>pBXgFCq`opa&bLkpo{0SC3s88XnW^=ULCmrSRNjAp z7grh(bNmVHl1|{5;UjXpBqNx6$O&sRWCgWf<@BxEgWh+@W3}3BJmhoP$H&cKA~J># zYheHNUx*KpIhc3(I&9sq!{*j$40sB>jxq$F9H69WADkBkVd>%(m^@G#F=iE_h-I%t z8q5EQlDex!1J9il^(~ebrB3_Gs$cNuNv9Pw2HRkp#wWIR_9%8ZN)>^dH*vP)Y#dA< zf$J~(b06*m*3y3~RWC}W;DSSx5$#To`7HLdZ!?<5Sfc#GC8RCSL(iZfSf8K6nIsHi zo$XA0b_E-9qm|jO><2fG+pKHU78d&TBnxg9;3^{vHVc=F=bZJLa?I3cpA4Z=%8@c<+3%~p=V0&vH`!hI(#n&Y;`_y%;-e?6g zO05ySx*UZ=23PR(T|QKMVi8g^10$RNG6k=(Q0AE_HB}9ql34H9_8xwFH}CbF8( zp=^5pQ*4z_0ZUTWWUh;(S+-grg7+Qee!e7>dtbneev=SVEQZK*DfnOj{QBhhY>+do zK25=fK_ijx6AIZ4w{SS^7OK8w!KFSC^Lu#~`NLc`zvLiu{G!g@PaMR&Ui=U_4ErX! zqppQJ19*Sma6T4?PKMKE?wrxCVtrlhY}Gw|bR3%iqZZC!)G)`Bknfz2tBMnGTaaaH zj>MVDI67Yf^5&13*J>>s;Je=ZdE-ztTLXJ+W%0=WDD&Cs2PwG%JTYts8~+bGmcPVp zjmxMHjb#B_OIRXZV;e``WVe3(W^w(4ncKuJ(Fc!_EWG5SNbB=*(ZABAA~$C>QC5vA zq|U9w8~&M^TDTa;+mtY{e<{1v7Q`xfRyp0XkY%ZCWp5WMvmRS{R>VK&QB4_`_@xK| z$|;;vw;mH-{$yqrr&*xEH@4bI5;NPo*zhUbXtDDVYck<{bfFe1PpWaOuKjt15Zf^aoAx)b@3qk^K4dD3`(edg##zEqH3HLL2IH&6PH38&A^d3xJHvnOfv3-5 z%z-2*xy^z6S`|F~R?hZabY@W$%vr}r*y6%|`? z-!t7>nwg{|!-mIRsP^;a{W>?i`KymGrxtc3Muui>TteCTOUY#hqh#LW%(t#Z>W?tC zCZ~ygcWmKoe?L}J0XEU5L*%Eu3wcwMaBfo?{`D(^+R;5H}B)@@MEpJo8PGc30AD@NAZ!FL}?;ZE;)UY`(f3SbH3NTxsgXe-Kn)Kx{Nd67auWiRz(%@{C z?`*n{Ec(9WvCjUVDI?BID4aT0c+{>hBrTH_K4u@HqaXj^SJM*Ow`Uo>QQ>>i*8WsH zArHn&NAd5?TI8Nwk5Ee&cuaMIzlAb(D01H!KbtBC=EKFN5xRW#)iTBp&q^xUnyBqK z`}QGf_S9pKSt$1I8-&Jtn=!K{4*X*R>PQJ8!QUuA2-$1iWKC!`*=i|m&FP`5G zLdvgQs2L>>t9=qAXE2_7s<%>Z{}B3Gv5rP(%TxS?7Uu1sgrYX?@zXvGv&~*;Q5C+nm*3f_hcl3rV2OZy2iSaf^&^qj4w>m;=$EW!TuE1;_RE zY>AdR6wXGXq~tlyY8Bv6>=rz^EsMx6hcTh_2W}YXQj_c;N^@_5(BlfDb`8Pb&r7h) z@D=t1Zx54flB2NkPFQZj#U&F5^pzQ;b) zu)dCL&(#r&EvDf&Ln+RuKV8s_!DpLRoE@)C{R0QmB)>a&rMwgN24^uPK$b@R&?VDd zJjZcKnr@X}hcxFQxX#-HnWiAr#M}H!iXP&NvtK=TMe_Mvf>#pG1V^b7#+{c!Q zoG;@yoDTU6r1pj@uxxK+-KPh^YnLP@)m5>^Mcz!?qd?RxXNBo`A-Kx#;8CTCuyad- z9pC*NIk+2{MRlBmEJL1RjGi8xMY9_vXtO%cdcBy1Z>b}YxY_~o;d^mKQU^v`d4{d+ zDm-`eqm;>Ac%xW_RRb5I;cF+}4V*wy();O7#$lS|QAy%bhv?qU0T>i94;?epQ8bnF z!K$x8+h-d*oy_18JqkXmdtn_Bhz*6CVEv*Wk|u1y@tQ2y?7REBxcWnPv8)&mOjJYi#Ky-wju4+5vWXj68(H?LZhAAac$%#Xk=c-FwZPZ z;2l#-{>+-qI|!kVjmR=Lf;K#hA;ThX+M6+zOkXv@!S@p)eh;C6*{yi==mwS~tcI-D zEhZfKhEd$*mE7e^qsFeG>ZY-z-&l>sdtTy`43S;)IC5xFqov%fvq<6`Y-R$)lR@q7tIXI$2ksu+3$ZH=$8MJ4_VjE-&0dX|>;~Yuro}hP)_0+v!u+VsHqL6)Wi7;`blAwD^ zg(i6}r3X7#QtZ}M^w-v$mZtsYoE#_mW0XdF&OM-3jass(E2ilto;3M=BBX{~-c_a<@_zq(BTrra7dH8Ti2|A$n4l6!igx*YdJT>Kx)2)}#G%pG#2Tn(k zLmRsp&;s`b3Y5rH$=_i%9eS~YM(vtOIVB;SkN69zvxn0Ak8>!}WC_n?3?ru=McU@> zKs8$r(&o_BG&$aye`a3c;7INp(QQXP_vcBU7N@@-B&g+3Arg0~(B7GADP>gvtt)b* z4F2Bq4SS2X>@Y3(*G}*9`w3YP6XX+G$>gIaU1|Kr4v4nEy&)LKr!Pd6m_9Ohwy|Lw zr3tZyUqHx2{m}k|FW!HyOf$MPk z9%xC$$#!&nl`Q2OsFPgcCTjc~&c7o`p-45>nSd6 z3a!c1V5T2pSY>@Fn>p$V`}jGNomg>@8QzHCU9BRFJ#iJjW}Ks@aUH?w(eRt5in%e&w!DULR!R8Jt&zTJ z##5_PG+kQcPh+j`BIdCw&9dhn^w>%iXr;lcAp!$NkHt`aRyUrMf};*zD6UzE(G^4R z&$f?+_MZ<|@xxHr*~*LtuL{E_Ej`$hr>gQyR$ZD(N5cJsx>-DeHw=J zUSFJ6SRRHHe;)@lC1InlH(p)bg#V`a;4+`Tr>tE=dsZxTwrBQ&|IZ6s%K_TZCQK0T2TK;c61w7WI>q$M(D9Ye> ziwyMgjKW+sC)zX1lMI}7X~KWSu;?H`)OJSbIbJ7}N4^&B_`DX(RwN1^yE744piJ$m z6UeJ&A>FuQL)(Uop;WPjNb~0Iy{I+}+wd0eRnNh2`ds+*jOLz*K+e!ViiMZ>4#V*T z&tk0NcS<#;9_!B%yi-~2kY}u|Ig|Y!cZ%sbuY%zc?lE4fK=)#7XtRSECAL4t1Ag|L ze3eLDWJ>it(o}Oujz42K+bv}=HThj9=dGnQ!{soobzMiw%Al~zfvozu5{6$>#N0GF zM7-``QcA0sYLF~$V-#dwIN)K41+))OLBpC~%+$^TH|`w6Macr#|EfXKqab|QXoi(1 zRxuCG_IYIWgm-Z4F{SYc_K%v5yxu1EPIfH5HBN(Lq85Ie41ke+5S#Bi2YNnn*f=x~ zhR5O&|2YbN>z(lU=rQ`b>?IwmY9XEYD|FK{icD&)s30|$$u`<3T~&?T|3MeyR|zs4a9?OAEDyCz!_2 zGG^DVi~Z?C@biW))NbBl{VRLeEa^iCl50eo=q1k^7ol zLCCLa#L`Rr`A}MlH?Bvpq~||eJ0vB{+qzMpC9cApVdI2&K}JYwy-E_9zTB;I0l}V` z7_t2ddOc1;y4xDIQx>q%oh9t3YckWE63))f@?%vAtJ&T@6E@*~0Mp3KXOBv%nNHF% zHaYnK`83R)H%k`;p46zp$%$ zjlnmoaA1-ThMYOhMnrvJr_cAps`xr)($&ja143EBvZILJBu_6-_oG4IZ*e{}n% z!SM5DxMwjD+pemi*h~d8#G~#(YgG8o zl=B?&tJshagV@#FB++x8VfC^~z~pxR{fj&Yk$WskK2Jl9ZVbEQtbmWrYUs|}BS6Wjijsq83W52=oguJdF1@j1v8 zm!wKN8LICZPhryLG@13_ag_xV9}~umr{*!e^d>gz(;!$5?P2e)cd`*Ze$eV4hkwmc zxZ7n59Y=8-Y%E2+VG~*_?&8tWH2CqJ`L3Zm@$0t&BBy#|>KQ9&4^)O~T^m!%KFbc= z&cUSqVl*Y?I|hnmX;^MI&iaPqUDrOG`o{Tuw*BeoGzr3dWfE(z;OzR}aB`SO8=dSa zcFaIhTv3nm=JPv+Z4NNx0{9qBvu9Vw|4U3xy^D{_wK zLu*3@vRjR4@+WnAuEM)%8b_e|aRV;J<+6FVMq^aW1YBCO5mm<j?0?T=-5i~Oq3Ax_XxuC$l-!#L>U!ru*EIS8hnfBMXX~AVosmJA=%leGBTy| z<}y;`JjkC)9pvYFh8q7fqfLG`Fbw7yw`+wMCl!iVS7|Xf}mFZmuD@-=}UJn z=OMgD)`m0u`y7RWfwxhYngJh=buh^`LsFC?f1VtGQcf3~X&8OJI)xH?*O1nwL8N+e zEp8cvgMCgxv;RxzjepH|Cr2=TU_1JyFQok`fpk>Gk!qAB=tf{7u6MlR-5NPk$x@_{ zRVuXYrxNWs^nhpnw_$1YS_uDC8N!o<{6G&wUz-M3M}5O+P%8Uid$OTWtf>N{vZ;<9ezowxZox zyz8((6yrJ9dBcn6C|+R7yHZm*i`0hx#Ym8TL^&&@OKgdk6h>{AK!2?|=6Goi^Xcb- z;qk|Kb|nhRFU~+xIR_IjIHK?GRLb)|L@DDBQ>Iudy$-uZ){47GVd)~)J+*?hKKjB= z{>o)HdH4Kp_B@^oJ&L&|YLxezC~%4%-Mn6fWIH>|qgONeR;a`vj=MyR-kpm zNnGL%v|?FbSY%}42afYOVh{>+hG2Ql0Nh@h2-#DmnEU-U&ZZP{Z}dUvgf*~J1_N+; z;7ewjoW)KYk7kFTTQNz!)ik&KA(8S^lFmIziHaNP`&LuBBl`d;bIoW)?;Ps9pid*} z`I#i>6oMc>oBp;1!l&_aXa`Z3Zz5O-Ej(r z+@7dfv{rqjHqkSW@XcXM-by0RL*C!+A|(IM%vu+L z@6AC-svilhwGUX`qlXy%L5G?is!>4YPn=wP2RnWyz}IjUC2QtT#V&@u=`lMKI;Dh=fs%2hNVGk}!2Q#mu8^WBv#g|3s) z!q}Y`gkj(B3FW;_!r^Dp!eh?ks3?v{S5-0OI|ooKY)@HzfW~O#-96}=bW~K-I9!EYj!lUW8M95f0iUH?#8f2 zy>fJE_oLVSzu>g3Ar0DNL?=UPF<$cs=Jo%?*=fI^U;YabyDB*cDHfK=dQ?#8N&YhD z>8ko!N)g{iUvI0>-PmweQ+0!lGLB=%3gZ|;uCPbLRb#~tJNK5{uM*JNYV*3)n;?qpK+ zI{3O>2OU*YQ4z^ARGHDN=xUCT8RwX@3Fcg{2u)51}-moi@|p!=)6@9y#HLsxm!G!=_E&yA5Myj>HDbFV_ib}A-+UJJG2a@>oR zq)nEZl-^=O(ZY4)Zem!knF8b#V8H2C;Cyd{H!Yea{f%I&m%Z+hxeE|kkJU_=YkH+PRP9ZfX&NOP&Zi%t=H?A)~ipf%5yLxXOBj3LI562b;jw9c|2A3X&VV2tttX?r3 zzed+GRU;Msy(f3qH>E=jSwgtV_6oQrA03<#&u?ojJJZWsX-Rav@0(- z$H!uC@7+q8cwFs^Aym;YEzt&I*88K z(a_u$2iu>ku{2Tzik&)Gv1>J^SuMoVZgVUOvctER17Y66JrfV$!t(7dOu3VSrijn@xo0r_kdvcU z-KV(!I~0#jEv2Z`JM`839UaMeOJ|ykDav^T-RsO@FS|7H+k6Lhtlt8=`@=A3;yL#H zM=-*hPC&G}1K0J2k-7K`5_cI(;SXKlubKseb5%&vsmALE5Ae?+0aoF>5B2&gG}qq7 zM1^PYTl5i+=htBA)N0PHm8VS-v&r@FeDYW_kT$*Hd&Xrxa1iI`)w4a&du7OH6nyU3 zJ&MTNob6rJ$_gH4vZ==Hoaga@06od)$= zKr44GqWU~H8kXrzXPN~1wLrwaX3l4>MyFZVp@-}VckN1Rw~6*^MB~`I&&X97OjkH! zSC;e0D`Lap+!cohYLAg^`Wd@?+EC`1ftjBt;xK0iC?Cjxsqqv3dGAN2_2Sgav;2#? z)v=y7;4S|wrz@Ps+bhMWn&E?^a_?CFZbex0J9|4mvcpa_Y)sTfX7S9CqVyBV#g)I4 zPC3!gR1rOLm87}{d9+vJFYU|!Li);=s4kN`svBfT)rx!RrOrbv@Emuf9mWZZWvJcK z%G95(N3-u^chn>Li7nVl=okne>$Zl7Yf+dOPn04X((h&iCu-eULbPnrB6OcI!|e#Oa9S zDO^7r0w&nNSR)={19sxm{F#XFnTaKw<6ZJSoO&h-)p-4XO8WzX1TLYD;T$VHC zmh|GnB~`LWA4L_#lgO-LEjMzMELNs*U@(7^+uWQzYe5G;FA}4kIsK{KO^dS-htg2quh&VsiQihm_?eNx{u}d`s z7M)*ABe<67=|nQcx1P*u*9Uf|R~vV{lpw>;_66ekZ1?X2u#3Kpu;3QBXgr2-ZZzCC z>f^7LI-VSe#sKaZROfDxiMIFgLg5;YPm(~;+cl`u*ntnF^I(wG#l8p3C2b3CEBe7k zArwQdO~HSBKOOSch%(+zrN1+_)AsFqX@|{ls;e**p1l4`O6Om`*cyZliy>9Heu)d z=WsUog657#DC54jmBZDL66FPz$eC!-Gs3d>)l9nm95bBT1aajNRQ%73syfHfLeKkH zs6s z_Q0kcy#N2Zoh3x;)5S+&wB=bXU330H?C5Vgygifd7ezznE9biVR&g#`D4cQ`6ldw+ zxFFV5vI^&Jk*q{Wa!&3Orx;9f z^F~U(9p~KsVXiu1tlLtSHO}#1yW(Tm)$~+$SH+JR+w5YaBpNWLM3xfLmC0b73Qha_ z3>PMI2c6Yo^#8aI!NJK0IhK!qU!Nl3_kBoQ>0m+iO8DxdhoV`ANVL$zv-h#=TI*r9 z$13Gqjz&KIScHVDR%ER$FsTld@F8uQ^!^$TZ16_te^LmR-iR+Z<@sE`xsh}v}4x2P+ zIcEcR;dzKT>0aRs%P}z&^XmvDr_7+nND=Bg&S02z9iCk(!!zF`DC)SNp+=Wv=l@5u z(j&8t{**%#Uqb5yqD%#}LVL(bse7Cv15O$7VQQZ(yejI6;S<|*bfAEKQ zh?C=|(d>PL$gipm^VikDC2ayNkaVF3IeK*dFTeZx9>d#d)+pDDB#*KVdQv4O#LoCm zdk>zX?RU-S0f3nb5eSppLUd?{F9Avd+=b82* zJLZ?%XeyyFm(+AJ>CMmU1BI4ow7ShLB|$Rd(kS~XD)cRG7vkTd0_opTP#}C!w&Ax!j;w;q`K4(}@Qw&=&-h{o9&ldgB;vTEb7tkQv#d|B9>qX~rcb5g$y_XYW zI%fziefC29&!NJV)h+a|crN*@oriv({9#|4i7N(|uyR;D?o>_T`Kl_!s*I;`1*>S@ z**%nPJC_!&@5b2pCbp?o0X55Bu-k2=%xrKR+aDv(3YO0z-%a@>%q}3A?sQ@$ezd7m zj2xSjkfqy!hIg;9e04pJn_R=;kR!-iew4dbrSa*-Fa+CQ6+fpJc$jz{$fJ6 zHYN1UppTY1q&(vjdum~etz*N$zMMhm8&BBpe#xX-FJe}s63xtzqm8{kF)`=`clE^L z`q-5;$10B;8?REPW;&gku#J+Ri&L#^C*p=NGTAwj4y1K+A5{+K{@sQmQ9o!X^SS*& z{xi1YuFJG0c+NeH?`rBo-61#O&HX4rJD^awFuq#&mBIUi{dm^mZyvlRc4KCUHhoW6 zKv#^1(Rc2-Nx5%<&h|Y>D@=r5^j?I`Jj*m5Xv6%@B)rk)vv>Y~a5#S$zjFvHCpfYx z4}+N5#AEF8mz(T^`fc{2{4(=(mVseNF!qjph>JeMC^kxujFRp`()%{v_8&?EwXA91 z1Y6pFS&jzJ-;R?1dp-gL+SLC)s=O0M6P9hD7D1K7xEr^7SSvddtBimwW7u!*hySzO zw$zMg%gQ&i)ZHoUt#L7PDNbPP@66?H`FHHXQvnmt?Zf0y7k=Ij#31f3-h8oy{S3Lq zHi`dY6{;2(@+leD_HY)`I}cWVv78Bk{SdtHBirtLgN@ajNP25ech(vLB zzOOrogL&yFd0dGjd3{*(`xc@iu9HT-f$&e$N>B-$B`nGwAw)>#@jmM@CUx^In{dw< z)tiSxVy!i--TtzJYn|A%fj?MTxDmvD`eMOwZ+zUMkE9!3I65Vf^WYN^H6WMs%g$rb zjFEVK*pyBC6~%taNtru%SpVB0pt&2lo#xE?Y{|s-VY}ia# zhk-G#Sg3oQ=(XxAQT8A=RuEgxG&$qo?ZyF2e$)=$*RI0=pB^;Fe1X}X2WZ(Np!umH zqQWO))k$qW4;T*SY=w$PHyACq#<)LH@K~L|hGxpctV{<3pDg8_Q^2V5 zGc$7=h4;E4(2L5&u7)TKFbqcA#?gGXw16dMZ(#Y$OWC4(Z`h3bJ1kUeIU8l`i2$1f zc$nlubiEC3JU^3gv>vCOWSO))*fZZa_R_CE1}_It$ss9?08s0UlN1P*$Cb^UeD3cm2gS*vI3=gMsuyel#6YSLa+O8S0JT zb3@0GDB%A-`xhCA&ba{W3WbNwPMkiY!yMKGu!$Fv*^atX?0k|v8@)t}b*)svzh4Ux zmu>-Jl_lOr+Q8=2FZSZ%W7Z*UiTn_E1Zu2B#a072=^kQZ{g+_d({QZx`G^fUlgM;w z5*=*ZO*ibma?j`_@;{P5@-nToVc>7t^dp`A+b@P)j{LJYq!JcG+ORJEDoV}P zaNIP3*0uOj{JS`sdfSiUl$2dPSZ#ve`4M(GQ0z_f_?(}{DhTBYm*8F5_ACimL z&UbK1{vvKZ-HY)CO0Zuz4V&)ILDQJ`iPkoMx??y z2b079GJY6@dxr-`svU#M{9EWTjY8fmGaTX^(9W$=RJ7(56sqoG%sp=eOEitfp zaSZdzQec`Khp`&=2vIqPZSUT|%}17^9297OpCq}P-bdJ@aWwG0KTSQBKuSIFq`hwz zS@2nM$YSnI`aO}mtHxnk|6L-D6G0K$H0%Hmod*jV@ZsH1xHZ=8z+tbyQF87Ypgs)N{1uL@0eI6X;K$J^lI`M+U3+ zQeA!^CGHzaZ8kL+KWZdKZe5DIrt#?5k$~tczIbaU0#2kc&+X=raD9%0kM(#iTb=Ss zrD;HfD}MY}fhz7D>bCxhjI`6Jy0;V0rk5e|+bH^a-u3 z|Gy^*+~LK2bmH!`J!}UhBnFbp8V9ns7(y9t>!D@aj2>STGAQ0oWeq_z?dDWcNo`;` z-d5PqI0FMhcHz#SS$HtFkF{Ln?~a*=NqKWL4c{J3#wqLQW62||ikwR4LL+F|j8NLc z84V-WkEf5jIqzZk8_XRbP|+wE+I`_JZYHHc&(RP4=VxKt2TfYxu!w9=%%z5XMl?#9 z?@dn`3ae^Dg;9^w1i97MgjZ+l1l0#eg;Ji)P4!hJ4@(hQ`y0}W1}&;G>4A)&2dnuN z%<8=IndiVV7GAuP-Fz^I?JoJutoIl|xHTJYf0y9N=+Rg-zKV^$cU0v1Z36qx_!Rqv zBP{w-GTR*K&ffclLwv0g-8dG(vl4gc+kf|H-lWsCerYt+C$}MThB8^dR3^>6zp!~- zJ>TnDknzWOVl$4@%~1jLY||$C>!d*bDVNzPT`}xy91L^b->Yrxk6miM?Bf!1cJ|Uq zws2@G8}Z>Q8?){yyR$usjq!TOQYTMDOwt6%XZu1mehT_&e`Zh9W-%?Bi!A7PEoU<8 z;6Y^o)Lr?mwxxmvF4sc$Dn;zj9*ogSDkzUQ!hYoV;p3;5s7(9;^Y|XzN|qt}P3|~s z@e$@BR-`j>CimDf(zD{vC-p*1cQj{!38PE9E|g_OnP;=m)ix!!SNT32H;e@EM*w{H7GMK(Vn* z`}2LafAe!zQ(MeD%hT9%Nf$O})HB}WpT~Cqm5{Y7f?Qk_9t@CzSCkSWaYWs3tDJ6K=Ha?gpFUuqUhS<1Z#)7P;`lENmj zqnxFqL=!e1r3-5DWV^_XH2TD;E^7)&8yS)L5N#^exDV&+;mEzGNB53gqH(XQ$Z_oh zsw})o11gP4zhxVgTXJ}x_da}vx8nRY&YwBxi_!>Xq!c=VkBRu4QWlOUI@#Vke3n*n zNmN4i>`=^4QOAizEKF=CXW~2;Qpd&Ag}b2ef1<&Lw8t$NI3tiYB%(q^eq zH$={vb5Zm2IF4T_ht zIBFVZZu5nnLJ;;Q^E}vygG}aCCNp?&o+*4zXKVjdvkuOSo@>YZpYyVykta^q!ba0! zSxBo=<)VpDy={>1F`|)|xX-)AhgPb@lF{F;G;`ts}V(>NGm~OW4N#b1P+id+XSo&SLiJ{4%ySe=y5>k-#ipJYu@_k*suS5}Q;O z%l1aEV5Pmfh^}%#-`E`xd*X_eh}oDXrpz6=VMzEOOZuaIDKM{&{%vig9nsgQDaeep zKAfhLL0W?Nvz5Yu>@C9PFT;eegwv4d6{EKO+}FCI4a@j;dxnz%-0n2s=hTt(<*W-u zx&>0~&_$%oS*%;N?7+!gc+y&mIOoH7e6z~CU}NpP28Jz@FJg=`a+lS^J;Do{Kr(HYs_GBHu0k2$BvPm(h^edTZ%&pkC3>r z3n3vR>DVkolK=9CpKVXVX}B7_@e|Q|5&ur{8T6;A2`JLn!Rl&1y!(`ZRL;X#FQh`} ze-xc}JeF@8#?25i5~aw>7P1HUD3hbt@z5u&K!Wy|D+I5$oCFcqSyyR6@F%4i_arAq2gF9 zoYEVy*@EwKF0O~tfWvs8a}jZ)BQVqym8?O1F$-Bf7MBOd zpnu&y%t?HNWd84u<~!p3Wg6(&#F>EiIU;z{evCO6iSdhM;1N~@u>+6MBBn+IN39^O z03Ff@=eySPmNaoj5REFiNu@DQ$vJx^souYW&GQ+Q%RP|#N{{Eh|6^xA-ehVACb5Ha zo-((=)1b4)AM(pKqV@I&_}rDlduwS*)^j6+&`^pw?N4F+yjQxh38t~Ju;>5hE#_TN z%X^K1%XO%mGtf$ZOC#*F8xW#_GY-6K{B<&-Kb~d|Z$9G@=@8jOk$&P7>RkH}TSIH{ zBVZPto0(5O(<y2s(-Rg8HXcAuTIc_;*zW5?#}9?c`Dz&&JATrn;x&QR=**y=fw^?0F^~Q(Sxa|L??u%EIr83Y zK;?rpNHzICC{4MEFXMEe$-l>1sdKQ@Ef|Jl79&fIziToNv*f~SEdRF-RxS-k#)Iil zw$5g^{yMU@{`*^}VfG~x3V#T*K3-&lxSyeP z@C0ngS_94EczkVbgV)_NIQUf#apSa*WwRPBiL?0LQw=U}53tOA1EJ@pg0h?G(5gHN zr!)QtQ{c0QvI*Fdyc;F|^5MbvG}5^GzgMic!LG{%p;C z1$QMKk{PT@b9fFjVuJ^r^R^_L!8O!wHb5|Ia}h3$BthY(xX`PZOP@zg$HYT9C@joJ zDxcekmn+eZbA@Q%ag0r!W&o9*Jhr=`lP$U3&9XL@u-}0@QSM%dUmlsTmA{XLM`|H8 zZ6XvzXW7JIocU{G0EuV`L`+@F40>Idl*uk^e7qZ@^gr;7!eBbEFcViM*R#5bzcHjy zm8!?r(3+DaI1lj8v<>tD(qnGS;4O*PCL=LNILoMkw| z0QQIbktRQ{zMW}FCjOhL>U=ecmBiD)`MvCg<0`y%Sc{)KEFgVc91?d@nR?C$a=YnA zlNDZ2Z=<5{b#)5wWL?BETWyTtJF7!eDseBW48_;BA#FBLdT|Yt2+m^iNxPZvnrJrL z*p?k{dL*(tWhYuPGf31IGD+0Cj58pI9~3q0j1f6}-h*=sFCimq$y5? z+yl-YVj@^=GlialH7bwzK+}^u?LWL=3g**s)2@*jI6q@Kx1X|{urg-Z;?5RlR$%uu$Hjj*`Hb}SU9 zKqC?DF43o%-V4bnBbg@ocu~qjPZ$<*wzy0QIxarJlNGIKf5O?&*RR3t|2aB zQ7N#Kfr5XdQIFN}@b&9?U*!(EU67`B&Ra9r0&r^@TC? z&S()16b7+b@-6Ho&sKhV_?p#wUu65X?_s*?icHj6FIsDF$in~VuyqJ_SdqR<_K zaemrPtS{SxkqhnX*LtbvBV(_Fl@9n?sYwKZ3TJ7jysC%FM-US>l;w)+oP=owrkBcVu`L zRQEP!-|oR1kK??bvl?Rz2P1jcSlmw&r+cYJWNP#hH)I2#b9e^|xEEc0#bhd+>q1fQ zLa2Xi5=C<+^--QB@AAEi0G{y~HOrk&UEfS%M%uKsCI^o~>YytH3Rz)HTY8&dyiox^ zxbyBTcAWCVT=DgWT|jJi%cf= zWs1>?z3=dKaWd{+yvr;vm_gM!pYQN%F{$G<{ELjqdMamHdTeF2p6l2hrO|9!T^yU9 zm%x;6j%Vv40`dGsIfUdatmQ7{t&6YY_{$>no-87zKYwUZ=U-}x_(lmIdnv}en1Y7i z6b%ch7v=4EF8W)SFZyz5j3_TwM-;sG9UE4sfN{mP2%O6EQg;LKHborze7|?l? zq9;_n`ar>B7f|BDcc{9|Gjq8iFo^Yll_t;f7;(1ta(~R~+J)VV;FL36|rT~SvFx!0fjq!q!+R0>0xpaO?j6<^Tn4?Zc{a_ zKmD2f?zGaiBWZNujw?O)l%=O*$H2dB9iD$!&fVbxK0Z`{guzwj#`9-Uedn=p(g%$1 zQ=pp-dZf>8p+@@**7OY_8-?MN7Wx63`OinmHXK9ln31SvHqEZuNa6{B)YR-oK~>kV zZop;oD3BGZ!k*2|)|Tf_3Oy?`QE}uGtQDl_e3~NNv>HmE zj`rf3PcIud#Sp?!T{K6^^1Qtsj;?vfejhpl2QSV^Et8^#5o&a6vOGCeC&Tx~Uv}-W z7D|+yF>l~9oU7xXzl|J5q`A=yQ7&bTxIp92JfqF}&2&JR+gkdP z=}J4}{RbDEe#dhYKW(^1m;0U8*JD^w5AI**75ql-MX>&ZuBjO~^}rrXvINV$kE3PY zS&Yf|n;ug=` zLW(PQQK->+O5z#F5v}&ZC;KF!=5(~+wP=}eJaU3?wfZ$Bt^Ip=N;EPv$#jHntJ`uJcNUCM69j#DuR z)a!>u!Rs+mGz3ly<=|njiG9~i5n>;O!^inMcjpjlv(qAt_loqry%ZiYQ)yW9ezLSW zMeZN(Q^Mj_l5^Tdr=?Ed$aEDl+$u-v%9qjToekqZP8hTFDQ!;F71qx+5}q%S6r>ey z&@XFy`jgy`-W87~J>>n@vD~Nif0l`cvf)hFz!vewTcx? zjbg&8^XyV=HhVHv}8ijurE0MiK8gD9USf#TBhHhKMVmI5dqttDlYyYPPt(odb(5EusCDWt767b87>-NcY<< zI;WaIzMkC0e&r`5WCqjo@v6iZs+jM;5B8a(scd~D6|Tvk{i%thF+7Y~)=ARys8?XK zc-QyzALRCSW5|(X2;arC0md`2&@URwCVZ~6cRpSy3_%e0N-o>f&Y6zl~Lr#lNKUWH^4_l_d^evo+86Drf=Gj3T0dgS8-m7#^m)!z^4 z$Z(wa9e|_LtkF+z1FP+5V#^2KVwS_}nCrDB>u6{mhQwC^GA9jL<}t82KB zcLFs$2eYD9nF_Wk()6fKwAkLlkh*b56_=*zhqjZ?ivrrqE>pwc*VMTxjaq9Yg?rD( z3#**Mh26UQ1*Pg_K`U&CaO!yu+gsAY?isvf?t1swTaQz$*mFM191uZIk5!U!WfIWOZ*g~b3XW~e<1*)$<#I3`t$TT;R4qeNjGq(~b*3FM@{bxXZ!hG7feL5Y~ z8Bd)hc9gf|2V%9l(ATL;B|76s-+VZIyF88_7xHIaz+jTCa;B)NaEg!hCd(_s>GI_Y z=&qiHFHiXMP$h_LOCu=1AeNL1y=d6sY8I^igIRo2MU@KY%xxctfir$FkKB5wT5|@; zs@te+?L+TOW%65i4rQ7GH9ilZNz0>2a(OKz9%1p$guoHHf-gvF>Lz#1+2VN zg$-IAB~smzi0W11w0IrQE;>t)u1^)#zgUf(eX2qkIX||IwFed32y`3Z3Kb zJ@(pxGS@Ak%M9#Tw%521(Dgd&NW0Afua~kL z%TBP*;+0Hu)-3L+A4_#l!s#Ui)8-coY0JME^yi2e#VhbLbWsFdTOlH^@Rjt`c_B#$ za!*Q(4QTUjtp1(?n?CM)I}?c5gXTEdG#idz`0RCKGo~Grpn~@MoG-s0Q`#=D1p~cs zFmF6}4_V>Xd1Z*kSFqx@2az^Tnvw!+sQsxn{fL@JqtoJE z_i+B&ITSKOg|scbC~TuAeVt@Ru_MfB$G3j;d*VoV-CB>pyuBFWn}Y4Umz{QDDu$U_ z(TMHgPpK?4MsWYcf`Rl7frW*m(n-=H;Q@aY~-#x9`fU+Gj}T0=`7*3q8WG8#I6 z9K|Z}d~56s8sO+em5WBvp3C|aokf@S52cQNhgfdfQ+8(kcXndS zT{djM1?Hof%5p3>LSEq*Jg;Ac<=Bh(Wl@6R9xL&8UlFTt*T>{ZlknGJC#Em4M$yb} z_M}LfmY-QpId9_WLE>?`U6Mxq$A(k+>{3Y2{Dr6%o>?5icbqHwQ?ulIj((rqD>8pKoUNFX#hyAvvV}3TSwP7+w)Xl?OyDkyW&T%j)8sJ%9(ALl zliy7gKhoLzlZ7sF7Yx`~;li#l!rS`CG<}_bmpM65nEDu}y)J;(9E6`7cb^XKU|XLI z$J?B>c*$9!&0l?B<}HP;uO+N~^mjIdXUE(U#=x*Q1X7nxaps8_Vmy}MwR#ffy$Qmo z;sy9}T@v?K+u(H10aWn3o93Asytx{RNsk4*s=b9T>CV&`zJN?7+R?y(dE}q-2|XY6 zgqlYY!oJoxA%Ep2!OeQ7Fmi*oaQoOGIF8>2{WqM4GLq*c`bVR6tPA>OmW%9HzY@(| z@JzJyvN>}Ma$qM94`Qz*e~WfNnsu*rX5yNWOx7Zt*}WDq(@_ohGI}uOjIpN`n!e;P zbu}5yRi!@)2l=e}6RyYK$J@9XoVs`mRo)RW3DlzFwy~tMi$CwKrO<^M{(K)3LFR22 zkZ?hZ45sPRlR1OHzJ0Zc9U7Z-U8(20Z@x3h8{uCjPYu z2IAi6FZYUB8!JJdpYvBo8zAwP1jcU_gPZFornKGy_lzfEYK8~AuPnv~p9*IF>mJ(| zHXEWbZpa-4#2+4kfih3orJ{o{+b&JxV(qAEY834)I!r@dH<4pT936}KO7)idLiA5N z;f#l+u&VSi9qPEj&V5scU8!t{XcV$0q+m`H=YJaTtn$AIjQp*Mx9)Mc5Xkx9 z|4N{2cNB@d!+XYG4@xQTF?6{jB@AG(&|OgY>ismy&gIf?(lDdsCuG-Rl7fGRcQ$0O{R5y=Xv zl;{~pomVo*)v%auHS5vDJyv9!9nP76adhjdFWDPA(AU@jG=AGzwj@mlQ;xjmU8mP9 z%dm@0e(%9{xCF82Kic?Z91OJ$@u)6}!i!%+&}zlck^67qqReCNtbBoc+)pSowGyk> zu7TO>8yHehkGb~UxU)`%X8){3UHA#^$>t1}N5jcqRgJW=<>>HJK7*@S!p?`ZvPe5^ zT;sTnPn@$+vg$88H9v)Y%1mHST*Wad+Zbm8zOwj~95!o{BU--f$NA-%km`RFpU3kI z>hcI|o##foUHEfFBAc!*D5dKYi^;zKI_kEQVE)F7nbXzfEPXDq34#s#D*jxw_*XKD zuD*xCOBphm@fYhBT!FI=zxy=D(I>rU^v&%SeOY*)UT5Frvy5?cOZEwKO;m^67Cj8G z9}Hz@Ri2A_$Q1Plp*wE}5?+?$@7fz!H0LIYp7XPInl@JbXNTfZgi>h(yjuN_E%ZOY zx?@Ud=2#A4Dt<+u3cC1xAB!ncsSV#QG+-;|_jjg+Bkj2wRlS}=(+*Ci&`%C@ zT*HZWPt%~!Z!&S;Pl42X#?!W+;gnsNOj4)k(+BNTmN^9b-YBTK>f`=b z9lFKOb^CTsqh;Sjv}MC^IyA2vdhe3yR`(TZQoPK4Yey+xERDo}c+*Zd9nzdMpUlM# z=%f2!+MXgo`PrOV;`kYDniJ>|?`3Q>nN8xGXVKYH%G8%QO|UU27s7V83GM!0g(7cR z>ob2Y2&ZC?!~NPlsE7Rr!wUncpRXw`58&LBjqR)}$qb}26Yk+taAJlnmX^pM>9-h! zQ8Bo4`w((<_F=SSGG0v%!takiL=(PPGCbbSym?3LA@}4SeiqDR4_w0PUz(JDXd10p z7eim)Mv`Zb2|eu*qX(;u>Bs_mniM>d^1s{geCrSr>kFg_KQd`^Um8iDPodavd&z6+ z0=lNs#71=W$FU<4P)Qxad7bjG`0$G9MpUpNif@_0F-^43G{qEqeSUvvXYwDnAj6>m zG3Soq!l$E{dN&5DQsZD$xsiRbxW%;p$v|w;JmlsaLxplYPVXJbXAav?zC9SRzZS!j z=ONBF4@BE+?wVbB7cYKFQ>Ilvn&A2$isw}0$%9!GTAD?NPw%I@BlnW7QW!OCwD00{Joow#v1a|q=BPZ)b9sa$gIs9PRTiiykC_Wq@T7YZrdRl*^R*X( zb*fn2A9wuW{l^~OEAM|l4M+Ig<%uW~zfaU+&4oThEK#Su4I`E7LaLQKU zzK{L`$U?e5RgLBE?s`5GY#m0gH!q{`qrUX?uZRk^{KlY_-_U41k~%FMsPvd6-Sz`j z`@Q66?Y*c8XvC++R^FZC4(ug|;TC!bn^u~!hOxP9^2!FLeEl{fsW)tv*l~8MAf7uB zUtok$j!;){ho1{m1BtY4iYfm{*D#f*h9-x5y#!4Icu~XPxw9C=f^Xw zecuBU)gq{fJz%ueLzDw;CYDi$$1zhxBKoC**s`w%0~Ji&mhM=*b# zi-Bua!AjyX4DDs;;DPU0p7IH;pED7!m;mWJZAhB@5LV@PVYTKAXMPuN;_vu z8e6xKlg1Ldu^q&kr6}-nKKwgs(6h7zaemoQ8Xt-JkCu3L_&?TNUK$A{Blp+U%TDbY;V_P)ut}bZBtW#&Oh4)#rb&61W=x3_85MqiNO!)d`KGkrCQ+Egh;?O3+d}aFvrFfRn0%y%mFqSp_7C|x9(s~9p{Wn zJCb#m5x{7ysZ|!8hnHE2{g8UV4bVqvH{;KSl``XGP68Ru|^fNWA_nqT)K-~^8D$c zwgqicN}-iEcGJHEFLFPrLvyDolFlnRh^{)rOI8Qvv1W*H8OB}YJuHGVJ+lWN!StjC z?06tUx*yf)QQ-wNo%X~pp3Bk8-hyp!_Ta?kCD`+4AZjy`QDylKA6)oO_x)JvSgT4Y zJcIU7S&B{@EFrnWQ8XsbpXNNXrv%Po38;O9?YlJT*QhbnmSsyPhxFn})^?;7&4ij> z4GK+;;0=Et7a4Nj$(v|&y35g1^J&ys=ggTWiz!UQjQUNyk53m4K-Y}VFGmcdy-&3$ zl9WmJC4WBYso>(~b(rbvgLLlcdi}!>W2X&+j!iRmes`cXe?&C*uO@}B?@yIk*SXti z9W^Rf(AJ;FY1g(Rbj&l0`bjRLUc1Y<=q5{7$10J#^?Pi3P>R1L=`eK;$6bC;+#l48 zzU+6nwDl=m#vj7|`^Lh9vpa>1f)qjRVUqA{bdFG}=P$Gs?1jq94xAr3fN;Qo7R^v2 zRkJoMoIe>uz9(Qr(+-GrY(>nLUFfs3K)dA;w0gcmL76;Fouxw~EBaHuS~e5n+ zJM7c_L1?^hh+o-85Pf^fwhd^)NQ(jdTseXQW-g(BrUBG5WiZVVdkz)eyR2O?oN|{~ zQEiY0_1egh%;w?LRXCfLK3Gl5KW?GiE>H4(tVVbCykKU5|FJiz{ zGYuMmgGU*r#98utwg-NTFG6pbCT5JYf^F0eZ0ZcfaOnb=W^hME=1BaOsbMqk425&h z2*k;GBXVd0RIgZIY{q}Aag-g>4Ls31O9v;n*kUF3lGFZZEN!~Wds^SIvhOEecYNZE zu^gnykD-#2$uuM|iQ;Rsi9ZoI%Xk`nuoA;!i9n1N<>GVe9(;O~fRK;+=;J)2J&H%T zyW}eVwe!5g&0NH5MPZxwe^l|%R(RBBFO+!L3L0;Wh32sT=+f`e_+l3d1+#4Yb0|gN z+(xX;os7PP15g^wy(%Xi(67T9Ju>sL&UFUHuZ@9OMib)yaA)f17f5hvgOW=WMtFZ> zMt>dARyP;>oL%wL-T?9i>1)5!{zRcXgHG13I}~* z7f*Mw8{@vP|Frud?npUXJ+G0o{Q8l&hYBr!@&V_39-=SBm-mZx(U&+2b@zZ9f0rP| znlthC=R&veFq?e#D$|#~%_e;9V_*HHu&?+y%e$D%npSdl?X`H8KJ5ULIBCk}Pq=Kg zC3z7f+Y+#b@8>Pnokagx#hBS|EzDZ#Scp2{nKl78`J0bQX0xz#GN6jdGjN@GWdz={^89bPb#RgIPFh$Y} z7NdyoM)32Q53AWxSR+$`t1_J9uHwso&%pg&hhd!Y3wh1_Jb&*gCWLw5bn1S1K6{CO zK0OF8@51P$yLhO500C2sv1#`lOqTM-I>QzC{nQYz+w0jYhozh$T8-74A0fS`0b;Fn zJi`-&J)vjtW2rhd?J=V_rzer=rJ3a2r%0xEy78?@m!i}SXmNxYop+U>ywoaKjys2$ z4sxVwatnvI$KW0J?EE^j66M?}vt`**GJ3v?#5z_}t(O@Y?0$-kVm!llHxbK^C1I&+ z4Ay=hiv+ztZ1RpoX7#j=nNE;`M#EdC>MIG2`-$v+y&g(6vv9?B5C1+y;CDR!9ck_S=Y8K~s3<^)Q;} ze1}KmQ)ne0#(?{)kRKk3^O99)F;b@x8Eq2IJ5o=o1)Un#gAJ0Su{xs&Y7S>GYpxhg zv*bIA<@FdTyg21FZ6wO(48~Ox+2WG%Y<=5N_Otd4>l`S-K1a>R z!sDqhOfSU`>l$bqS7QA&f0S-@qr^=-KlJVdS!V2{iRb3fk{c#;FH{9T@>E!*K6NkYqcq*8PXaBgE*3dNan4Jj>7jo-k@$i`pPR1X%yp=Ni$-@XwO$M`hH>*Z6OKTJL(>Gw!VbM+*8=YxaYFJBcdZ5 zG0bT+l26Fsi|TFGvDE?L+ykxZ8^6yGzi! zZ*Q^6p$@@HyHUSGpaa8VX_{XFrA^DAEV3k>UT@ybOF;Fx1UQGU#kFso+tr|qyVYtW z*?$%#y_rM~nd;OXHHc0(-^cN3-zeyosbF3(K+yktm9$bT>A;NLG)PCA)b36tjTlY( zY5om2UUehlR5^lprpQ;{m0~VWAnLTCwIXFQns^n{^Sy*Yd(H_z9zGSKbM6X-?`wto zKO=C(BVS=1IXhbmw7rvaC^tJb#+=T7Y9eKbGZTQnS#-@jm@<(#*4 z@H}qTJ>~ASH}Dc=A$h?Zj9DMgknn&x$^2x6zRI{%ql?r{H(A=ZaBSJq358)AWHoXW zk-ZxAYV*%bZYb%9&nELi5v|~y={XxTNwxhpK5tw@tN0mG=~XVh%?l!4u_wi4X4Deb z%`Tcxu<-Tjk!trCY^n<0)Z-v;v!3vs@mGi1i9Vv*ew7I%%a4|guX`Op6R z-ee7za5FskJQQ1o4aafkRe0&gnQBg_AfI*)o0KAP+1dp+L(A~j^Z*VX6Ngl@9CGJh zWe3NlpsTzECzgp(iqk$w-its(#(Xq3#&NdhIqH{CLh*{6sTP$=NAJw0Q%}ONSwTdx zmy;-_c`IdDs?sv`AgD@Cg@-{4?>r8s_P-;@V4*bU?ZzWA%~*KXzE!yPELgaHdz7Fh zts;1+eIa+{5+-l*h-v3?KE2BqHu26omaUY;#^)Pgg8Kt3>1n~6$DE%Wx(}V9A$T=h zoc_dZpf2sToFS!2>ht~}^XKD-6<9ca+>0Mf% zkxg?BSfk`-1l0JPWs8v+x-!&YX!wH_E}ssav|SJ$r^9U$--W$(!H%H%>B{l)LKlCqsEO@3J>PWYz1GspCK(75n&7%XU4|i66-CV2|OR@(vHe z-lE{}B~-^;Mc$=c7<#S1%n`{<>XtlqYMP)KGTbY9g$2kKu!~DoDQM(!I@UIi=5Nv? zuN%$S-d2E)S=~${Fbs*~!{I7d%J~5wpgiFSPO~IwFnyG21*D!}G%NWxfS+?5fzdV?UX3t1PZ3t6-Mxe=K-* z0^6|QHF7Qup)1Qhsq&*Mow+!GiW+J#JaHbouhPg?XFg`~)&pQK=pkm`6Xte1lPwdg zWp$xnSl5JeEXeaH)9SNeTRxt}qj7v@zqAtK0SBS2pUAu7<8Z0p1njE62wX4Z4lEZW z-j+qYxf~>x1jFNaJ3jWyhojpOlq^_>%1Zt!>X{3p?jtzLd-ob|PeA6?6$EmYZT}7{ zdi`b{b)5C1I||A){p$-{GicqTer^-(U=mL;Hf#{vwpjA42gs##FeO*Zsb2Gjo>%;Y!N zu%Tt;+`ah@YC}2WtgRc@Qp&MoUNMeQk^Zw-zJg%ThQjoC_7t zfoM9`&DQ%aq0J*(snqK}nfVow-nkI^?KO(x;`hQ*j^{zSYvA#!Z2Y*l3Exa+!|sqh zP7E(Wj8h%D*3_fOI367qlVJVdF&L@7M?dZW`YS4gbN^DjIm?}Xc~Mxky$d>il5{(2 zJlW)Wkk(uU8r5;1eH!tRDMb&1#z($Cj&#F{9$oBPTZS(@XP0w9nO3tO@bSM3^ZX=? zj%$V*cdsS!`B;WtCx%XcjxCyb*gxYHlgXQoi(kC4{|E1lOjJht_vdV}?iJEruPSKV zcuOBAo}_)IyQr8ZP^@P=XI~7Wk0<+4%b`PPn_dW;Cl2_KdjbCz>e2iIlZmZQB-d$4 zWOT@g`g)$?9CuskPgbIR$!g?bFo>q8J99u{A&!g*=CX1zRfgzwkEw*{L2Ea_`-0`G)T5D zz<#-UrY>`lZJphY_jBLCQT!-AY(I-l)<^O2WHzibJjmi{Bu$^XfgEOf(D>c6=~jm} zg&NLagZ^5u24>FIo(pEpvt-$Z<7Y+Ynk|qVVL@|`nUd(bG6^Le_|(St@AG$pMwFxU z`94JHZGo%hDopM)#(EQ9tj|h<#J|H>F_wE1=C9#>VsUi0Xkcwn5Gt02;O)Y-SXaQi z7&onO^7TQcY9x(|w;I^p`%_uR-8IbhstTL5bOjCMIq*iIiL@$8XsK8ZO?|qJZnXY} zoPh(a+iFZCmujc(ma-qxA)xj2$E_v+K9rA=&Mp9|^} z4ROlD6rLCb(OxMmss0X^;p5419zUxa^URc)JH2;TrSCJ>VpZ;0{FD5Nju3qcIj2rG z?lru7B#FEK<{GL>hUKx3?E{3Nr%tZ zj^9{JlH7H0a>^odx-pFYyOGZ4a&0iZoe$Mx{5_cwiX3$}#B7%1Jm_(-ei4YIQtlr} z$iw%Pi3my_A^i7gyI^`XS*To9E>w9m319jr3n7;KQO&!GR(wxjH{>7UI|P!tsYcsY z4`bES6Szb43e%AkN6ghzjH&psKX%5b-gOAaMMZcTB~B^jo!C*e9dDM&u;`2BEJWfO zdvIC~GM;kie<_1)$eM`>yd!JDIm{2-l_};^e?BQK$Bz^*IBq|Mn|p8LK=F0X<*!0@ z&IXKFD?=VbM6~tj6lz)%Lg~xG$S}v06!iD78@357?e-<+vaysMjLBrzrB*VXYeU$= znZ7KnGM)J(?_v|e(%7J&oow1`1x$4ggQRLC0uD4m%9=AQI4{a0D3>{k|734C`#SKR z0|q_j@A<6>|V0s`g9g0y#~W=E3v3dnj%lBQf!C}{q;SG zot_)fsal28&v+(ZqY0ZFE+g%JD0W%tVX53UTn)}c!=fVG`Mev)%0)PzaThD~`cc`k ze)RUM1xma7nmNi)nB46s#5p?)Qt{@(;h3Kk?j)c#AO(YCYtgfq&uRbl zBF!)c(lz7RiLe6JtJ=**@4n5Bed9UW(EDsp#TZ;)vj>eA;_X4#t59255_6qwUOPbEJQebL>Tssu8pg{$M&tfM z_@5cU>iFjs88w)l9y5==Ray(87XwzijjdS&Vkt zV5Rk4OH_DuxhQb0r^xW%9#Qo4jiU1n(%6-oj_sF=adp5B?5qF5G_^z7u7+mz+VCy= zdgm!yeUh_$>?WZ=+!c>(JDH4$E;@>x__KKw8hO{)$3+XtEs@Nx^gMfHQ^tIZGuZs) zGVGqLCJQk;%3Oa`vxD1t2h?s5K0V=`sv8QZ?@3^xp~KNaVd(glfR>wk&@@>JioSuP+P`ysi~B|Dq3{qRZ}MQ6<`miuxM#X1{?!SSOAoWux+y0vrnYbDm1kk4|C7wPdW!7U6S& zIKI`nqbg|*3cUtm(6#SOeq1Z7zqyCa`*ML9ayCU=dmWq9zk+jFqgmXAeK-@-iRyY; z`q`(#yJr1qVmxPdeUd>7--#N%ZidDCMyPldAf;&`UQJnn)tcw=?3OewjaDFsLr+oi zHv}8Giy&;*Q+UfH~o)i6LWd??f5-35ogZOY&za$LM3CB$?g6N&I+?eWZ6!X zMel~HZZhQf{X2NRJ4SC%XQNF0*@>TrSYUVtQ&e+jgE9xRd z&lOE`OknWqHnX#^!LMbzIqxP3ZyRRA?}j|?tIp;#vJ|}TE5V~9uTW{ob03XH!(>YZMiO-8fEG1K&7hs>ODJl^|8vK)v1Y|_RA}8p z?+iIgxiN$ko^-&wzM=ko*7E5og?xDwz_t3g0eEyv`iuYFvSb>Tf znl}%D_su)3C^?$V?C-?P{7d1e+kqU;@30xGN;3`_(DaNi&^^R+Wz*XcKDZyfzjqmJ zNiWd3Bn>zBPsFcJclvlE+@*|<;RDIzY5mX64#m_5;SeKpTguX=`|p$9lq;2|Dte}IQuvv^0=184a@ zaK`;_Y|LwIWQHr^*?e(G|7l^RhN*1BCPnl$0&44Lp>6q0O#L!l=}z5K4vo>IW5PZ+CljDRFcxijG?=$XOes}--Fzf zp|K+mLfk`(#N8*6*b*;#E3zjC>3;O>coBx14Z!2P87N#a7c=*97IgD61a>Q9h_Ni+ zGkZ`#m^;m?bD;tMKdWEdhfir2q4%fTGCVS5aO``V;{Zx{%#ZY?ty#q@xoyMw{o;df6-w{^j zqP8}R^VKqtlCT3JXZ6Wr#8xsbaiFC?m(Ze6MseAV_+-0~`E(XA9rsd}^lLwxV&KF6 zZ5L;8eKXjql6V&E$NBCDLzwV(EaxvQWzkDrA^RkYGdT0mAxOrQ=gV zIb7iR(T5HuaLYY`)-Y?hG+t-#X7O2T2f;q@Crc>JWUHh$vogIj+Pm3xx+Sj}l6ym(e+I8+55wBVoxa z7;LS?Kdk_ayV}ogmxN*d@#8oW9tu0TAh=3nFYK1Mw{to$g>^ess#JGf>>Me0)C!wg1ypOgo=r9$HfWu zuyqThEQ_JF?J5R6ti-2rH((TY5&}1#d5r$Q)05BZ=L{xSz8BxN%>wEZva!Yg3+B1J zMbYML*x2mD>~?L)c9>(LwLQ)=Z|)qhfxFUT6kL7J)Ml%)amD;+IwEzb65{Sef-Ul1snNucMe21eRXz|IFW)d?%3w2hP+~^*seHSiphl>FoCTT-K1Xkm;>7 zWVJfgf=;hhf&p(GlsT`!d%smM@R*1G4iz{XG_tfC6>Q(@04#o<0VO(#xJvHfK6Hbv zcpbvzByX_$vZ>6dyMoPczr@08x3Q|cNVd{J6$k3(W9y&QP#dR$(nATX@^u;{l+r(jUNCoIkcp*Gwb$}86L zyj3S_D7?iYhe!a`8mONufpKVOgFl|){~r%_KjH&>G4Pe;G-k5fqx{*5A|adM#T{wS z2GOK%!|CHj1zHl?jip2VaVzOIJ9c^)c0V%UeG`4ylutue)hj0PwG4JIdm(jv5XGPP zh@IyOaCpcftbgPO2diV)I+VK#d~~tHriUdC8;*Oo?qG-8UpVTEk(xvgG9z*^!ke?o zrt|$**HM_LdtpGu7`b}VXs>Q!3gU5W;KOro9GwEZfxiapI<7wE96xhbsXzu_qB z_`$5le`4yUO{|FL@V?G5<-U-^d`5f}f%oGetKo*54*Hn0wHW(;4W%4OC2}~ml)UTX zNhxAGY4Bf*-%Fq$1tlbtR!1{_U#GY2M`*$ZEo>^724m3$s5otd8V5y`WW?~F%OHBJ zq{+SIHZ(%PgXC1q$tSxROVadUSH2E?YxunGR1lhNCUHN*Yc?bDBrG*QW5xYu0Kl6_Le)5mKx;F+_4AfC^>IK_dlgV64>sgTV3GTF)qWvTFD6M=kW#vq#bI&-t zmGiO6*A6Blr=e7S@E0t~2h+ypcx-%{hNcNOF~9#iDw^By+CK@GQaxdKaw7DWER0IAbs0W8LQ(BIl?-~16BHw9kgX@B>CGb3 z!q*Tt#qjaV%*ljHLyg zE!Yv~f)sm8*t_k)m&8!45wF1n^H%so=fVGiEYtZpgPGruWN~}dn2e?}JLumaXipT! zo)uHj(>(;|d}L5yJQ5F-UbBxLS4ga~ixwRJMql2%qQIl~X_bT(8j*uHG*aawnFROarFJ? zjP9`$FiDt!*EPkcPdmahE=zGdYcR@BN#M#d9dyjnz$xd2$a}m5-q#dRWNr!f?>TrZ zI16prX!NBsoEd+IX*+PXg;EC9M^?~+gd6?Aw2aiw6(yBYZASv?( zGc3~3nS4%=HuSF`=4q=SL+_g)(@dHL?z}A6cRZP?_S|OQrX6Klg6^@^Egfvzmmh5N zLLj}jQ_X`7k!-WaR6&?w{Bm5>-A>4i^0?%Te;d?mW#Sggz52b#Dv^>S<<3{k5bAbEU z;}A2CM_hOc8h+a2+zvUcDqjn)>-%Bwz#IB2qT#pH5)YJZnC$2>wk128xw!VR_=OYj z=X?trGh+-yXU~UNnJifSGW5^!=b7=nIJH!noYQ5=RL2E zGOT%*j_Qe}n9Y5N^Rv1zGcuHEo_)m1x|H$q;8*6@1(;bqV2jS1vK1E5Y*+0~Hf7=^ zmazX9TQ;VIJvw}hEm)Yyc8NqW-_}^RIkS{mc5h>S^Vb5~wWT2fvmTn8FCFeftmQO=-NGqgO+n_l99q$abK-4oBUSh#${StwNvKKLi_XsZ< z-$7w_72TSzCwy3IBDDW$Ep(AsAQY*U6_&ExisI)}QlT|Dsp``4)|W6lD#yOKMG3<6^99>y zI0yopQv@TNCke`SZeYctny50>$LH`-m^7da?cjH;&BKi?J;2#bnqQgw$_f_s?;)$1 zeV(}n$I;}fFTC45KpQoG(_PaJ+AR@5B9~*>aHDSa<98Rk@TiVya}K8SxE8kLjRd|( z=;5`?A7<||8U-c(*!t}tR^HBIS1UpnPb*`!i#{r zsPY+0e)mnusNISR4UI^*`5%%}`w+Ev0Y&czb!^h0J!T{6$Ieuox2_lL?OV%kt~|hK zOdER?SIV4xF0f&5Kceo;JW9#&pnCVMq$stE6oTbRev<U?BKal6OU?nfoO zHy#J$-DNn&Mbd7=IBG9(q-%~b0)gf~!PQw$1k6E=9Z@*JhHp5=#wdzW!#Fz{Zxl%V zX9H<&bp(}tGo>vCLiW4vJX33xMyka~yi7L1-c*8%?ks#*dlBE)eZ|4vNBCX!8}5at zU|zKf|3>i{WN0xS{yoDzaECE@su^y--h=SF37BFZfZiFI@OaV)zxO4mjrL}X`FG%i zIiJ}^pBi>O@*^9!^d8Ha(rz-%EbZbSa|#{9B}M&_wp_X*8?Fh^{Wz z!y(_RD7Yj+ZdZoT74hfTv~MLgifL2g@R=mSGsnv%Hq!&G!!(ZjrL}nP$#JJ8jo3Aj z{%h8vQ4Rt+*zy4HA=i3-=*BJvcon)cV-bR5^kdQYZV+P9zyQ7)ezWS#QQB-*y9w2j$P-mklzE3 z?QUU7Jg@oUiVpXxn?v@jC0unonR&ZC+hOI;WUWrImzQeT#FIg6QS1$t(eaF3mlKC` zzBtmw8(2kM1e5+UnT@orp@)BeQj6brYMs+YH|r~DUBDJP-RMQDo@S8LW1eB*GmE3^ z6e-ue8ZOhFkzkR65q&3cnKKj9+qie}%xHXfeul-I8`Pf0eOM7QC~t{9Ic@m|>;3w8 zwB{gw-YDkxp7Su4h(hmmO?dx(#_s}#q_=`|O_x~HM8Qas?%E~cRt#jEICi+?6uum)UoQt#^U%=V!Dqg_D%sk{g+CoZwJ#l znnp)fj-$h$t;k@>EZVq1j3$~_!$LnCMN=w}&m@RCP3S)#M>=wNE?J+e!=||H!aA#J zA!WW0exK1Vl-WCSTJh`#;TT6_T9WQTx{G#DQ^ao4{TabKl$+?7yEPtO*vI+(`=IjQ zN=P_NzDdEeevP?DoA#Rq63Dw zJ)l8srUSM6hf$(W79|8Gl7;LvT3{eY8yi-W4cXJ4txIWFI-@PW2XI3?gUWIq)AYLA z!i8FRkZ@?ox| za!j87CUGW(@m$n3yI^^+7tb3x;iUuL{jBC57v446zd+FQweXf@ zqHtz)9CuTl5Xzez6zbUw5k|yWKx9t@s`x%Y!=V&g-S6?8fi>RW(}eHbO#JY>1i{U_ z_;d9Jg8O%1_rL-a2z}t@wH0fR2ckb}DaLFYjlC^|mm3aZ(};A)Y~uctpDS^9@+Iaq zMv7jZG9#5r8?sYaOkP1>QB~yv^Fs@1<+CF+PCbQ&MjxWzUc7VgISZm1TZD~{8N#s5 z*}_5UiNa|nn}x}RFUfN(_jz2s1##xMCz*4`4d%gjU6WfMFkSDrtZ!f!+fzzp@$v|5x^|xLF6t?Mpq1`%Zp+P@ zWn_CTjHWurP)+$!ni8>@N@Bhv?eqqU9+poBovO)V`B6GFY9tkp%frg+a-6>~llF#L zlIpzyJa1$Lq^7L7563v^fNDZ^za=!<^UwedN(3(_OOgN1-f6LI8_7amyqEI2FhtW-2 zFy`ecSe>sx?ud&RR5=$LME|i#OL)HN?kU!38qe-}EMr*`p9N!N{4u%YEQbBOj>ekD zc)YI~XU49@%1c^!H+~+JQvQQth$mtdt?+nsCmR=$j@MmpkU69tjf<|qTY|Hz5AK7_ zh*wB%7)%qcOHpsU1O;~upk=Kac33=z?dH+6a+D6eHknOp>*i2EmJ;2Z>JIxesaW#5 z80Az3!MA2C+!Kmr|FvL(*JzsHZc3e|oHJ=Uh`vV$pf9ox8J>Kn{Z)+q@NR$`pC7!y zEX=$)9=?uCv0H0B_U*91hfR9WXv=4lp8bT*T~+F-9YZfC>67y!H4^@ag&Ch|y*#^w z0_N|fx&?vc^KL)ccMYZz3mbY66HfPc9-;IPE@W9dn`Q>oqVAsrUEiZYZ{Bwx>gpMc ze!Lh@_?dl_?KYB5NTvSqS82?uQ{-4OgHjf~;q zZ`l@kUCgst4E4To=-n=Xig86u`qnoVA18q4UM0Ly{K>lX@3Mxk+($3F7LnaYxGOq} zyNxm-9K?5yXVjq_FG5522xthlQ?qmi<+I0hbD)CuNDmWgc`O#%4fPQYUluNm)d?1^ zTc{<>y#5*8^Zp>)c`(HU$dYl%e335YWf4xQ=3aN$Go5y zb^$x*6k-`aH>CGQBPVSMp1;n4Q`v8bzmz3ru1F862GiK?C=|t%^4;@S42}GP-Or+M zI0Agvt;k)Tvq{&>pSY7_Z)6z>%}o6wJb;@yb%(A|?y%lCeyONB3J z?}$v=Hhveqn9a{l&km7c;!U!>&bfK4j3&p#(&rRoy6!X_YQb|b$aX#6Uzz}w7o4dd z_K9r^`-6+~Btv>Mm5rHc!sc$f3l0CF#7st!W1%cn+KrO%j{c8W``Wc3N3*uiq{ER%cBE~wP7nnq6+cYPryZpc8o z%_Hm-9e`=V6>M541*NbGIy_ihIP&ceI%;r=evG!{y}S;%o}CYOe&-g6Nk?v20+L5Z z<5R;loLgE6S2b-9K4x%b%o@1yatcr0Yw z)loiaGy>hTna_ch2uLdA3?lws8Pth`sZu2S_A<1U-H^Yf88>94=|GA!`EhoX&aQZ5 z+`5XWjo&cQa3}>&mZq>rW9Z}6+ZfxVN}t}?P_Mfk?UCC|{+2?DoPQS+mye@#12`kR+tH@!{D6&>9!i-BBu<2_ohM!1AfoK?#-mZrFel>n4A4NMpTawc>dwL@fP_>p6 zweUNYjbtHa^M1^AKNBkY=||C1IR|g_EOM5fhqfi0Pk8qv5x3tYHp*ebls@od zE=8QQAen=s>5|ni^z(jfZkR188PqA**^G$ zB};4Jc_0oWXNIA?OM)bwHArcmCN1(GOSjHE1@_7b502O)w0oQ=^gVJx_(1KNP|na# zxUk_cA_6j?EK!LKJD(zWvn=KBsl>PooorhKVZH2pXlyq}?y$K~_U~cQ5w4gnnTo`g zt2k%y8f~L4U~9)*Om$abwlN!6h;AVh;j^UhwWaK`nm5~@AA_=^P_}Xs)xV3REqYNju_S~-QdZHg17c)5IG(*s z3uV=}-m=VKF?7B2XA_QTve)1Dvo@92EGT|B@+~D$DlroERi*68nvn=uC4^z!A{^r^ zvWcRA2tQ48zI3ww{O!94Nx%)BFp9d{jI(U*QP)dQ(ar&xnc za^ZQVggfwUYJ>9pcAPp;j;gn|kbbL4_ddCjo2LgYJhPUb9+*rrKSomG+Gw^+K@CeQ z++gG!53}xA_%sP|C)bSkVghmLbtRU)>Vsml1SPJxkE6f(X#0&NLZ=W1q00_y;m;gp z;i<2;X?TSRHeNi0_XTNKn@|kvwnP+Nox-2}Kddh67^Ld0s^H0Q8L9VCSO2`AWg;{FP?*R<@I!oS4QG>O9#t?vS_jD!{4Evv@nT7l-6N zA#_g;V&Ba~C1*aovRFhn{r8cWwJ&+xTuqZtyvEcM^1=n9orIej_6t`uZ5O_XH4~Oy z_()$Q6(B$TC^D8O;jTXaTuj^!i^ZIc&4Sn=uLkbNDd6X)OHBIRQzl|`n&t9yuu6*) z-Wcyh^2{Arci$PJFMhJTxB=vUT|pu7Yw6DO0J`2BMWHWT=-aS%{AWLh2DGM-?adLS zcH$Dkypr*(U7BJFj3}UeFD-uILvr0c23-5e>zl2ufRj3%4 zj{i=DV#$7ncfO3GU0-9Ji(H?#l6OJjTMw(gqE-&e-xf6a}#X7@N(NJNvrXgJ~=9 z*la%}6k|{+6^h$o+Nhl{3D2T8L3NQm=BvhGUT_KS4Cc9q=@Ct`cPf{m|d4whK8aUZXR5M-6iX}gHsD1_wPsS zL;j4NXu{{GAGvF$1y%n3Sev*NgNn}L@R-}E_;3M}4@6;S(+PWmliL(YJ zcuuO2)d%>p4mB4D!^$|9gXg2=uW^5D8lK1N!1^(t2hDheC$Zvm;*}<;R14@>>_aqi zPK43=;Uri*j+BQEAbD~T=W7WN#h&H+P@3B4K90v={rHD3Oiab*-`f$va=ZOx} z%r`-FY^w$7h1`MF<_#$4Iq<&^lK3+pgI4KvkdzyaRn_CU%QuWkyvby*j-Fs6PlYnK zf$>aD;U-hH*MgRt8j9ykK;U^z^rik}#>GZJUm#3!cwRyw53RiSam7oVcNnV?v_XYl zUU#N*y@}NGq?%rzK2G+}qv*1JJM}M77q&KA2p>pT2;bKX6ZT>lW-G72QnE)#5$_L< zpN7NpGudTXDY~LzO~;*%kpJU2+V#nU3=c`s&98%Sw`mW2BRAs0f@OHLh%?U(M{p;4 z2HwVskiH+!4^<7K8AmEHC)W;BeM2DnPlP6ye8X_%^JskUjlkL|_;~0ts@ez9t;w8Y zQZ=0%53Hvae#X9ZW)@P~Gcn5aKDrJx!A|f4Q+xTGA@2w7ywReY3m1{w6icf2;aMot zQn)zVsPz0MBfypRTq_L^y&EUCsFoomVz91wi;o(G*d#kj2c9IrS}?)qn*;Th(Kh<7V#AT5kO zjLV>z8gZoYU?)F5n?aEuAn6+0EzWU=q^pb@1mz@o!QF0U&lEoKo!IIo+{$QZ0wAl3CSVcgT?vj_SYuE zeBvXlnl^+aUMTRLsu<}t--Th;LF9FcP{im(WVPChCQe&MYq-^u?RetfoZH4O7Xb zTc2cSCgW+hvGDbQC}ELxs<8Q3nQ%kzJz?VcU|~^qIr3h7#!mMkW_JBH0965A;X`;FmB7@q5l=N@_8dhOfru&vV&sxvea~ z@(}xC7R!A1C$W8Vm$U4MsnFha3lA3k#HB4G=}FEgN?!LA6&bmxEdPyZ2XyHES|#e_ z*{{c(owKu#^S9DG3D09`{P=9THY$<)%Q%ZEqMT*5Tw#my|FOSO!=X3k4NIB2n&*vn zGJnH0OsmhI$=P0FWgFv|UfdeCm%I0zzv-YwZaikJvxG;XF7I!AV#5_RSbof5X7w+M zecsc~Lgej1tId&}eThxwd+cN4zgSAlaL%AkWAB6hV=2>vV61r-A0F`D(lEYzdXs@^ zYqw$RL1hZswweqdc+>SGPPF4GpOQSP@&Wyq4wUo?@TF{n^+=U6vrZ9EX0mVa}#fB+bsn+UUbrsU(VoVY^VB z$a9u+S`ZmphKvn<7<6SKy0&Cd^V|-4p*>vq`QLYH7*$Al?Vh9^yaD68<51(xd#Edl z;pm@;Lp<|%UFgEze<@~4q6zGcbU9miOB$0kW7*J!quAqc1J>H?!19hwX19CGfJLFyHMZ}P)`_X}JadTbcBXhQk?A(= zW%FipkKC)n>`sya>wGaCdE)}%Gcp!R4%={a0iS1It7o3CbNO!Ch`iKnNKz0*v9YBz zY$oSE57?0N`?K__=mwn*>Lmdrg&$THlTy(}RK6*N!Kem&*!mD(s1(-sHsg!yKioB% zM)T#jP@YpfmBgKO0e@56xn9+b$8!Sd0|v5qt1pjnKa`w~$t zbi%*+!_b#I1@b5T@UO`ob^IOjd5jkS`Tt|kqc0-mAXF`msMtk~il>L!)DG^Z-~s=h>RL2spd?A-~lWA4!%(ciU40-+5NMZlr^MOlk9QDGIu@ zm+AQ}W^>nNGw+<^Jfk&cY1`Ydi~9wCDrX`|p&dEDa zc4z0hQ(3ujBwKwkh#l{ZWB0!svKrCP4AYDdU7&}VcE+gEFv9VBqwwzb3Y7A5cmIZ0 z_@yXCS8IyEepw^$Q9aY>|G^ezRDj(*%UsQLK8^lLbzek{Pon`&4udWo5D`;78VY*WhFwvW!V8j|vsk$4@r7u%+;L9x^p zyneF&Gto<;O9ZB z2J(<5sz4+D>*LRQ7uFr`;F*irhs zrd9@Y@1$V91b3g$NW`rh9$0%mg|odR$SF{t!Xgytlm0K%`^IANn4$DcYBqhlvyg=G zGpLkjSZfFIzKFU#^V3UaCr6dD349jNv!IwcxvpXJca^fKxsu3~l7`+WE%Y3a#K(IV z+4ZZ_u~&5oSiUJn%n~9!a5%O%arVuqJ|>ZAjdvO*P>FMZxbI1*TMe^s%`aybeM zmf%9C51wWiqf0ml|EkBq>&y=9vkF9F;1)E0?_q|VCzqyZKrW`c`8_6tj!ivE(W@hA zU9}5!@t#3(UmYUf4WYy3#`NHmkVY>XL~BZY@l^IL%x>Pnqo6_D3B`RqpKjx6Vg;$o z$_V=l#|w+26@;giT4=$k2y&Bg!92dlG2PC&+%a*GyR!gIm--kB>|~{Xmf&B3ALm(B z;QP_jc*VOzA+a~mAUcvdhsslp-b`wV9#3f(22iP*%I?)vv8C#DEKvFi`|%>2(U%D< zrv5B5GMB+Ibp`lX>LF`WBU@0#^Fkvy>t_XnXx)D}vwa+9I9+6+ex~e|{2B6e|4FLe zBEq~~_b9#P44t2QnC^K1*41v`)eGja@ zRPfAc70!hPBQkR;Cak}~E)P{iUEn4RGjoA;CSj6jADga_!;CKE;mg{8Fu$clt8S@~ ztg9sLo6(H;GyCw(wHCP+f6;F%MvIlWm;J;=ycYW^=-ar7$y`ZcPhO<4Q7tJfIVqCm zYp1~L&|B=PQl`D#D&+V3DirF2u~v}I?)r4Ii;E3#z`-8nwc(f->W6Ak?u0BnjKl9d zq3?AFPp|NI{_$%plQX*uzQ;5F);@ujbCw`Q&P%X;`dEQdVlx|fsfSBn3?S8@jEL}Y zXx9A8wlo-GOJO<;$0uU&_2cM0#eHCx7<$_qSf=Y3SpBnx#pBH=>_3ExqdTE+eG`lN z(#lq^=wj{n23X_Ycy{*fRQBKdr&u_ZNdK(`rM+29BI_2A8ivy7y&u>x2VH3WP{YLX ziCF(t1Fu_dW0@=q!rw?}1~-9GLox zg0z9vn-URheiKcW{C%;3XPj9mVy2ov_(c_AE%vZda>cA5OO(DC1C80OOl#DC?EE=5 zR%^9^J=?gE**&&rieF{f4nJFXZMcZ)s#_>-evBbg-r>>14A?qv$L$|w7=7h7B72%( zSbP*QA;5a+Xw=2@L3+U`l6|$5L_?2K=0WaTJ84L&V^7fQ_!G2OCYx%*uh50saQfG` znx*I^vL%9Ec82#Ojnke=@YXYPgA>#_>wR;Fm- z-F6x4eCXuJ(ChGaejj)b5$VA+X51INxE%tweMZiJg# zQq)o*sRz!bg9_VecBB!N_=MuL!b>=Rc@LkxFX5GN9bv8B*uTLH$&KlptI3^!W%+m} zb_rX%OK@*UIx>gdz+)%Q)sU^g!?yj{(5Z)L*9$m%K##VxSkl?M2WfwP6g7%$r(jFQ zj`~-z31_df?ZfV~rXRU%PUuFa(_o1e{^_t1O@X_qGrZ>U>~`xc2v$v?<^o@mnZ2Lp z`#Mt1%=r{t! zi9iibPo_dOPzPGMf0)~YTTDOt7xTN8$MPH7SaWkB%NXv@?}(lBs$5DK{qP-q+5Lpp zhu)>Ts|(3&O&Wz4oTv7mM@Y`ZpN^U+@mOkFh}R^B^t^Q1SnY}tnMsj5(0_l^ZO zx!|L672edzQP~n>nz~b+wpP7^>~!8id)NWfIi1KbZ^D3d6~gE8zCrL$>~maAPr8qi z;kAWi=xRg%sg+}1M-?^v94VA7eMJ|9g~ZY#=-I`=q%*CAYBfJl^|3};C>qB(Qvu|j zHHV55eK0d82BMR<;@C$ggnU`US;}(|A|FXze7CB-ESF-wou;q9eQDFev9#%EI;!^T zlUS7}9g2#juP6Lye;3i@iW3Mv{~2jJMJUGYIX0J`#RC06EC^jm54&T@>9!M@$S)?- z`f0Rs?`3TA;Ge-$3hFl3Uy-35H&`P&V5y+ zYvVX~-ddAdq{9f6|dV}jRcEtp$<8G&O*#^XKnWUMM z0sjf5yqCr2KH}rip{R+M@iO?Z?>u{N(t!>!f3i7SLus;)smwH=yk^XyzPv7UO&v@_ zt`4T{w^eBP;9+!4jq?CBO33!zOY$kXPtJ!6D2MwOgW*s>Qf)WW7X^w?PiXMViK45n{c4CD{2bL{-iU^(~7wqPJ4Qmsu+$N6o zQ~3;+-@$&_9K{-I&S!~?M0<@4=LE0ATK{R7zSRVS9?GGvD}d?GRU_A<8^|+bITb__ z9ca;|t$S`FI7FMAoR?ARi?#HrYzs{~w1{R-5uvyRE$r-#K#U2$hA#%0_$8Bv6-~So z`2&)YU3Qnag>Mzd_@w1)!NL$=_cn=lX+?f_FMNT)H;a%Ye zn=(r!TkI#Cq9>n6F&m(Y#GZIn=Sl(xT2rE`gU==#QWbocl&%5@~NSsqTy1Ada< z@pmNhql#Y2M$xi{Y8>`|gNL!BX?TnXSv?iyM%~)CsBL13Z~^tct+0v z(>*NE8m@^Uf9G)L^ifPN+X{926qv8;!%8-aS4%po|v+816oybF?91T{0nk~!_{KAjyr?X?~b75k_*I=r=W046Yu5nY)yRu zb^BkW#d`VlcH&a{c!HmgtOl@jK$0xAULkk?8yp#Z3=X}oATxa{b!i@=$+MlPNS9G| zmJb-ub0%zigs{|yDE@N+vxmf?VbMyQxyA3#!{_0M7Q>*$$~Zk(7E+(CvNMf0prfTu zdUXpaxk!zsrjDTb&Ka2J8-tTGct+k`lZJjBPwO4{-I_CDbQJ{@dC7ymE3Be#Joiu) z%w3@+t*EsXl2WlQ*|z=TT!-h-P_V)IzkU!dKR|}3Z`1E96%;=$liF_wQtj?cXk6+* zTIwjux0E5@>URiAtVV8k1}x@PGTWSfHbv_n%P#0)*;DxM^Ag#xd)qL*Ef(t+=Rop7 z4tBjN;~6F!BsyKeU_N^=oH>||4}Jy*o8QTZb zqsdw1D>GjBbn+gdCzl`;RQL)<%u*L-WQvfl5#KK;3@6ze&oORT7w&0a#*2#SaZp@Z*-{_N%V!SVkn?Pa7tDqJ;P=s_)-N zZ{#%K&A*2poHK))iaGY&o`aH)z04pl2#s@PX>Wr$c@N)9SM&YJake(SwQ0t1jj^&CQl`XCxA^k5a^GnwOv8aDIm3HD?^pa1gi>%%#* zIJvC^%hvESb?{fD-h2v=x^zTjcVmm)3<|THL-RYg(kJmfwB!6#Y8v!`mCYCmiCxmj z%X`OqwQE?(9tWltatv#4ci>vUH@IhZV%5$@RM@QL2)t;l;$~&EUcbB550{$ zv9iYp>)Lg2Ja;&JwiL3BJU_QXN=)dV*iTu4i?rl*5*@enpqp~@Y4@lMTA;w+rEhl9 z_Opgm_@o$j9W5~aa}o;IW~0<77*8jz!Lyws;MqG3bLxt*(&ZzJd0)>aLr7ZVZlm}9 z8mv%l#5<`z{L36oEdv8kf5sUWjpxv?VJMw^W=mgOc9UTGN?M`v0jt-qC5h#ys9mC* zqQeR(KfsADbcoYqj&@iFloG&_*vo2SQmGHUe zFrnnM5Mg_9q0sf~P2o%1Rw0u=A?&kFg^afhd2uH}vho_a{C@c|tkmKla)wf=!h!V+zOmS*6|! zmR(fBRt?lar9z)3&0R~e?E&QL9!l;RI+SziA|#cR>6!NuYSx_0`?bSp?mynMk5;0d z1Rv6zA4Zywg6PYqnUp8pfva^f?11Gfw&TQ9d~8y|(%?d7<-MHqZ1R~-5$7bEkHKuK z(YTgA1fSD*?@u!c2FHJ(W8oXbMN81A2_=XsvEcleYIf$=H?|~l9JXQsPUl_4Kh)h93&ZkW-l{5`R?=_4mCbe2cAwS>nW$IngqRR)UMwn>H)7rT3m93l2BW=$kmn#mIyI_vw~BvHS2|MR z`AL+SBqF@~bhYsNqa8xW)D6On0~$h+%dg1hlOulWT*VIJY}R$05BR(fcSZ)FX$jwR zH}ky7IdSsn6sN>1qV)cTALq{<#-_Hjcx_z?JIS})-&+Q1SOf~$j-zcs2-%T}TcPMxlq9F>AjAXBn z5!%ye($FH5B1Ib7B?=7{*%=L~j3_JK^Zo^r$Mc+X-`Dl|>~;P^LSv@D)3Oq)`~Kpy zM+;(-pW@*~&f+y)%iR;{Ol(IoEAY)@2@@k(kiIUPzgV1SzvE!@?j&+ATtV)hYY1%s z#%b4a-o74noj*iM@mFZg@3&O;F_3n(b|LoJJgV@wC&7LRS~J*+Lxvg1QlCN{YtK^T z^taSfTtTVty~(CToSyBug=4F}LAXSnTDPguvmW9GplZ5>auw_c`=*SpZ~%}i~2JE zLSvTnS&C`xQf4pKIl#Rwns>Y+u$|q+frtudao3^Cvm$m*;yivC<)Lv>9lFAwqmMH| zjB<2PV(5+HRY!0Em!VL07flAvaNA>wQ;Xc{;m~pn0pd8!p@@CH3a)dO@pm_4n(gE@J_1@8E-PAf=WfY zTAYoRjwtLgZNoFyv1FCgh@+lC(92i=`K+HTD%Kv)mmT1ay2&te=wQu>QS5I(7$!aa z4%f=RsGTWFRWsfoN!=eG{f}d&#BV5t%%UlqJ?V3hKWPmYB_$h6c)Te@E9dSuY!#w+ zvyNf@a$$7!@NU6WXYx^%B&}mSFP!NEzsa11J~*Aea`%V9?h0yD%pr5TB^2GkGx+|6 z7~v&EgQK}CmS;k99tR=i{94?1N@0$z`OMR_mx)$LVW#^n=JqFm1*lAheZVFRR3Cxw zOn+Q@aS~OS12gM&*y(T$V^VW@moNosDj^UvQpeDQn{d(}Pb0#u>77vojrOW15r;@p zEO#Rp*EZTVS6A>^$Wzd?KS1!lZj7Mj@E+u#65>M3DT+G4y+;DZcU^ppZu~W{tXzXOTYmx^*H5wb=2K0r;MhKTvnCxg@EX79?G5C(Y3 zb08*;6R9NFjH*o){@Ikr z*WeeTIA_DX3OAnJLC4&+cvN>m)eoBPb5g}5`97$iFP9%A&F*L!j1~ux^G`q-|zUA+x-m%-rN1D&> z{!7Ai7Kn?pTX-)*j`scDPEJ!jsr&B?`XpG-bmOX-aApmY@PEaUFEz0Fy53B|)f*R` zYw&wVEfNmj$65_xS`eOv>>1ribTpx-59~-#x}D-)oTCr!^J(mTeb)Q(0Q1TSXG3`B zI9J-AJzYP6wTw?f_l6-9cV3=&&7Uke_o+SMH0&bI(I@L(5|@z?n8sI-=wJZV#eBzs zD*pR$YQ?uoIU32Ce*qK3>8t2TduUoXw`B&nYCbK7^7#d-FNQb~<9W5z3$Aa9;f!Zh&XKeavwD z@;}yk*oL|5Ut;Z3Vp!k2FlN?$hRv?qz@7%Jg3sTZxZ!>Q!>8wH zN>GPR559hSiHw>wwAF5dNQ@N|ANHBq#MiN$!k?^k_9&#-ck$h+8@{Mo5b`5{J;EObK#XE=r%(htQIO251)Sgz+BkZ7tr5M?2Rb)nO%n4_RX8kcG%E zDa86++>?~}0nHO%peiN}>mMw^>ccX0ZQnua-L{KT|7;+V+e*v6Xpw3xXTC@3BPmS> zQF|FKMgzyWLumEw3^=?UO2aJv;?jRQ}Eq6b~L2>R$y!X|->XqCy>R_UFboc!uI&nC7wvA_@?JrKr7ef1#DSb%WMi)HY z$whcQP0+QbiM`|Ku~Hlp-+YI4%o)dhfX0yE{XUuWP)w{`j}^<0^C1w{v_{U z?CVcpf8)Z~#6y=@fLR!u`B|M6nz~`=ZTyB!4X&39vt)Y82d%zv8d%5CVyPbrq9b@j`2fqCshoRxtUDdaSxMx zT!Cei`1x}GKX__LH2vC7KO4<20B%oW+5~;jkaJ1Y_%Rn8c+Zc53Z; zc2zBhb*%_x+y880eSfdw;-f0q8-2v7g{>&tl#d{x!*J=m$`l&S_%3WYwwLja{0UFo z9<+u3wG$}Je~QOH((wE0MQDabz}|U2YCeRqEuX{L5BYWMy;-fE(Ux0!tDgqx<&2&Q zoeBq(o|}j0iT3c%U4p&+3h=+ambngn##XmJWnF@1wtZeci*$})gGJ6z80U+8o=4gH zAPn_+ygOez3FS78Y?Fcn9(N6}UoA@bwp0}BY?P1#_=Cg*Bc)iF@SIjJl zfo#oLn5zX~*QO(gn5GMjurOvGQ_H^Y`o)Gmdd!CWiep>m@T&9Sb@1N5g!>RRVUueD zyC4zFR=&%sJ~uMu_R08k$`*g7m}6nddUW+pLhZKO7%&s1>?d+0)c*y=gEiRv#2W_| ztmpsUei$HAyjE92+>%yy=-o%=9L#WImM=cZb69a+#++RS^tTHf+^}1WLJJYlEWYY`SFP?J=PJPSN=GBDWbPu83ImiW@DU!j!G=i1LiY{@M+ zhXQ3E((oI1=$7(BlKy#$vbuN%=n(HF{Pu_E#-#`jnt`7Ua;RG_&K-D$R5a9pnj#j_ zlMYSVyQvOsv4P#Wn9C*h1-mgI8-9RX_aQMwjfRNsCSP#*L^4rCT_YFqA_CfUgT8D@qo=|s#8I3wX5r=H3_Jbigy&8nn#8M`;Y8Zw# z8zS9H2K)c}#=3pOSmN0P>{@*n&l_(cDzp?2-0QKVE&)eZxM0EiN{EzS!G`dg*j{@P zC&pVKMc(@+ho7hNJK6 z8VncO3Ug^o3=yA7mi*t`e1#X4tT{&0c5kO+A9N_&rd%&?jwrj)s>+&9%dnucFZ8yK zI;%HBk@r1^55j#O?+>(@+*K9P`+F~z{`2G<#!#pTso>j6f3{8M92>EC zG%krvf_H`@swQcoojW4FE*^@hmJYZ-WhH)ZosWnGo4H#|4$E3zGt>VRac}k@v(3HB z+M@2U&vrrVYTpIg9PosG?(d`pQyWS6S0?8wPNVHRuab59QyRZHoAkb=@;j3bF+DHt zUUk5J&L&dK=YEw^Z)}rsz+9y^rr&h}0}(GccaHad$IPZFCy8=>hSITRNr*9#q64bq zDSVnSDMzYM?AI_TpZJP(+^?(=8A9Xty3hloP)Z!Q3C~Ay^s+}xu;bBp%F6#hozu$s z|M5GBtldRlP9@XH$cq$E5l7=STxq++E9}|07h<9*xO*!W2C^xbGCdsp@xbp9p7f)w zfD+}hX_(4&8lK=uDf+VXFXj=}>=;KnuWTvPWi#c922!Z*3cAgEG%w?RV#Me_=rtdJ zsc{pQsyIXa=N9^vmP%r4PV(MU2q~_2Cc{6MkoId6eRocxlHMfp`{zfeKdqo2PV$s< za~9QpbRknO9je+ejx+{4P!Svh=k-1K^;?_xo|lYwxzSP!&Xn3(hukPj!Ox5ff_o|D z0=3g`1k&S%=qoSC6&&#MMrcwl>~vd^x9ua$T_i|NgLjKd%GtQIZf4ab1CKGoaQo3e z*4(w8t-Wv>3P~NXUp>3Cub^3}cG>WPnvRhE9d1B zsjkY$8nJOOUtx>y9??*&JPRqlHxeE&;J(3&Sieo21akVcZK^9Vc@OfNWXjsCWAy<`s-^>*{`hVf#+A07Hvfqzh9Vdd;r?a z?{hs0bTs5F2Cb){t8f(#9o9tgk7>}!nt{r&NY=668#APepfGqD8zj@PaQ-$l&TnK6 z>APXktx7*_r_r}#?-1eogy&RcVIXQxMLMP1v2eVFQY8NmMTPC-yWp}-j> z0%et4!Do9ff$sS-%38M_yN>c+(<)`UEGJEW4~`_O^QG8i8_dd_OPK!fAokzIqipx( z6NHBCkD>;t6NGCVP`}v%l$nzkhvwmlJOI8(E2BP&HHvoS1PgKnQkoCX+EPi9X2#?4jboH zqBlQxEd=s$2>Tm?dpB0#*+e&(Zo9{9))umE4NdIIT7sY98?f-iX@tyQhE1ub@YnSV zw1+n#jQe<_{z#ENci7$8U(K4<%VT1`D3J7#MVwA!1w|`ZM)Z7aZ%>7a?nA_LkL`t` zOGq4NfgQ)TK#$MC-dq^Ba*qM3mmn z7p1B7LKMPT6JuWWVo0kS8LW<=%ZUy&sC$s6ja-R(_cSPfdxkw%KjODE=j>JmA>)EN z&b}CnVTTu>XVeC$EZ&CRFgaXwJHwIjCjHHHSO@;`4G&F3BNnS<8kb7wlO$SK43^)u|}wNGh?A0#c($v|U*sb$O2t4z8AcX z7ZdaFo_i+yUhs~&kt$q+^N~HG3)kxFL9KOAjtRi$SIV%N{)hAGWNG+(VKUGBgIS&7 zkPP|3yyG&^E5(@=uA8ajS_XZ~Nu?zQ-gJ9$Jhq5WCL@J|ypJACL)-U|Zm$l_`ZJCW ze2b%*=Xi!C)rXFJb|I}#DJV$`!}-M>II={FqTY#+&X%vp8_B)xb4n@xp{yYKmWW_! zLIX9w_n}@d?t9ZaL%Dt*=zh#kDmp(zkiW8;Mm=9nb1y5x<);Im%dW)o90w?Cs=}lp zkYP&-q8Ii;?zjv^jN+X4-PU9q$Q>s&&#^Lk7Tr!XCh{6iW;_0)y2oW0S5u6=X~XD~ z6z7jAO(pSbY&6z-akjEV)RmCvL} zjCX}zj)DKmJCMIIkLF4kP<$KckA5MXE55Lj{c>)HG>g_;T7idO9ink|N*AH~7fk z%jZkVuv=sr>HAEf!+Ukf@3juK>vzMqeJE5nieOs+&tYr4XF0-^{4Np6wED(Sr0h(R z%aJ3;gM`WIXa(8b06#z_I-LgT==da@XKgg z=ORtk5p765b{ad%a@m5~(eQevgXkA)aO|5crj{$AkY_xy{;3n9EhuKD9YrnskIt9& z;$mG7GHXUtS%xBwJ6(=^gIG9i;rva*1Tu1Lq-#;!f-&wK1?O#}qATxl$k!6nZEaA~ zZjILGy3pS6m36+%W7hotih-ddx|{i*tu!8^m?%zk-C)-)WwCwg&)N0cPk67Rl9hOV zXTE1HGS9n#Z1<6N7M9=67KuvWq{=kxYZrx3iz$j$&xLfV3Km^>$x5Gfu(bB`Ottj^ z5`SLwe>*7n8eCrGDO=!Sv<$Fl` z5{InKA6ehxg;?%z9M30);6&PH)>}B#< zR(~fOGQFM^UvFX)PM>A0{tvr6VJi0BTF$?ddthsS5b-H8_%mr7R=nAUihujCOlCFr zDa^yD_TOwd1vD%g|F z`}2+Q@K?SD-{rg`E$@$kSMmt>6@{WJSMf=%0@tfTkrQ$Vx}T)sgt%b++W98Gg$P~wvc)zjR+ z`gJnx6W>VEZ*AzRi6)Jh(E$H12T*gs6DLQUMr7y;3`v`Y;4)#1&CO-y+hg$QS1yjU zR6}RxWt5uifq_py8;D(mdHFmak#rjS4{(GxLhYCZPJCa0hLQbD`1TTfeqxOdX$!=98e`9I?*D1bViMvX z`J7mvGH3cw%=Tb1cRNesqn41gUkF7vv~Y&k0DYM+Ch&IZr>`>6B({QQn6}@*_vMv% zypDGVFL737TMfI*T}6(yA=KR+NXPXKlYZkididibMirjKusvn)kMG5LE&lhLHY}% z(nRrHj6$a%aVF*F6+558l8RfVJbHcFZnYhzice$i$p-FxSYS|-(r^MBfNdf z`Ptq5>`m-e#C~h$zU(X<97u=ZlXNKSZ$_lrP?)z)MS#&9JQ;5VRx%N(N6)dAM18uS zf0Bm%3!qQQ%jm%ib!rg*gM}8q*p#`vvzu*&4vjyoZ)7$*{N)Ox4T1d5Ay3mB?L$`d~^3CUFaH*Yju{V|j3H)~Rf_XL`>_b%q1Kg(QBzGRle zA2DNo=Q-*7hSh%ZX0bvAP|kjgIi~NRs#l6~hfK7+@j?6bF;wHJNuoEV(r2Nmv{6Hw zwm%Ri{mj|y%@F}BbUnfvw(GEinXP(>$w%}q=%%4%NDuB=@P2D%6M8k{QJ|uQnN1>O zmp+wd=5Wt?Y$tB0T!3or7}UAfvY}T+VVE)$DN($O`|~Rcd6mh&z86J6qXV3!3@}%9 zCsH+R5h%qStCl(tPd3DwM+}>X2%|!{ooSr%W`irPQTc%mx@`N4ZXc^BnXC-*k2yyD zrSjApxR?^17E+qA9%T<3L0P&DP<$(kD`Dn{S2DupN47XLTOSoYf7z5|DF~;W$MK2} zSQq;p*2|=+?%rdZ?-Hin4HGzL)qw6R=}?!SEcGn9&U5#3@a*v;yuMov`)idb&HRfQ zJNP*+MU1ZaZYB{wPa1f1irzIIqwT;tvNSueveBf2D;a>g6wVhd9+-dWDmQM&+|~4ZyZW~ zw&~QD;YOqDj^n>u8CW#<3h%DJg2mP|n5(q~+vU~C#d0~FJK#drj~8=>0`hlhubqAcAS=Dc?hXp)4kKb&V_{tC_JoI`gi21zAL znU_!``x#Nq;?sLrLx>o@Rq%OMsvnj#j3!wVN7~@%MQ`pz(Y~rBls&Bt_6|m5{M3$q zDrr&WfjP8vqbg|`Eg;7UVH6k@N3u!D6c=)W(&`2DVE0)zD|9F}N!ueqVFfb!)Zt=s zh4pNVWC!1VW3Ly@#Fk&WNKTdkRdukwi0R1VoVP!#&!W004gYLTKF7U)*ZO#xJc))tHrr|H2FE6m%iE{^)Q62q+c?{*dB(q-R%$FjMx8l?! z`UA?rccA`l6CK}~Nym-S$?;+Wb!$XZfVmG%3-4rMs>87HwmBAbo#gwv6gb5>V8(kN zp2sUkSH?(sG-)PndThtDmg6bQPfAd-WrN_Yucu)97*9d@EGNNx$%z8}p%N%to`6F~ zo}fOo3oRq65w|%A2ikIB#dpXWO%im^V+<|gd*J-IVmuz}0-bjWn06r_cGH_NX-Xg?50+s8{I6h+v z&99$N*Eo0dXxlV8AymunKL69`q!x>j zXl@zmUTQO8A5XUCXd%0Ey_8umE@sF6TCpLa2Y7$72%|3dVqo(_EN|ydfe3H>Db|Ho z%{;siQpAM;A^h_bgTAL4Eah@gSHtIcMnfr4jpyagiSZu)MW{7)Gg6d9q)j(l#F?sX z177TttR5Sw>x|}80l4rep1aaY`D`Kz6~`B&wW6HgffZ2L?uwQFMdHoI>$p_21fIUT z5b-n-?{9?S`0zy3a;9~j=mWG*5=Xtj7*#5}U>M`Z*}?~ThKNB@!5_;cd5?Hdj*{!l zXlvAZT3s@hx^^ULWs*+`oR*bo@b+V=}^X z&!9l{J#>{n^KaiHz8@??e0mJ}wS@WZZvo`o{BT9m1?S(4M2d7fJG5&j)B?&;9@m9_ zt1s|8-+;|8{2(o@h8%Mbj0{M}s_IO1dWGX(A7SFW8g#|YAl)_A6z(&bmb=Q(*uUkx z|L_DYm7{3W`9^G<&a?4ezMTK1huJ%LhpB1~9l5lYM(o}~g^2=smm@@l{5kG(s)x_3 zHk|g#guJvj!UCt`P_QGLD1Mds$rQ7H+sas==p|-q;m7)uRA5p6AAh$^M|PqisycSz z^?6m8C2Yj(fMWE^yujE!Em%A^1D~4~BDPbBK0ZE5Up&%DYT6^RP5Mc`&#L&(y^6+F z-K9@)BLpq)rVD(dj0NX%MFmT{jBw6sEAPJr;I`%o?uT{8iCzW#*`Y#XMjWJQ^|N&5 zekw^^I!E`0E~AvsA@J3!fx#O7PVIOD(UE!Zsqx42#Yy1{(cT6$CFJ6&u_xmBp3JROAC+I{LifZ|{BE8=!^&+aJ@hD@=AEtu{)9&HBqYA+ zLfU{R)j7$M^7)r2vCn~3ff6nJG@X0kKu1N?XwkcR9OylXkzUXFUh^*0Hb20~_WNk; zE5M7MF7#xZP-&DEW#m~?_R855Kjt3>FKngr%K2o`Tu*P4iz%-pk>1*Epw(*z*+kCj zzfi~XWc6Bjczql`w6w62xj|TGcpA^oN1|W21~(3jCgCO_GUxn&4Ytj&Jlulc(T{l6 zxtP1Hk7CH-a10ME<3AwJRAmcO_o9Dji1~(1(_UiH%b9ev!je4ItSIo7Hs#;u`6>TQ zY!Pih%qMXQZJ$QB``1%ct2*Tc?O-jZeVEGT3U-KPu~j-Z*opaWY^=p>Y#Io|iL4mJ zW@WmBL8Uj!E z)tKWJjs;Q=;S-;MZJpckc>WB?rXPo~STwe3WZGWYj!P1ma`S~Q%xZs ztidy!JxsM?4UN{kNPC4ZQqPzKKCeDW?wjY6*Ha0S_MXQnk0vyAqYMRid_-LQ?O#(Q3i~3eEfnfQ&fxg{46J$Oi^~1hSUT|pzRw7Q zcR~XH`8?2ce+0}gPUS8EFDUw@!9(>StmH4?)}0xU|N0Jz1&VaKW;N9WuBMoN4T=~R z2+yh^RP8d8dhRcxitAJ8f$$J2isG)lZ!38BAp}AP5U9^{2&w;MM3f-XO>9NTV7rc~3si%|&{QdhiW3$X7RGzQNethu;<$h32%CN|kBKBDvd=qn*nYPLR;3ieB0Ps7 zbE-bn6&#Q{JQ9y3!Z6fUfD5ClSY%{7^HY$4_J9$_nEUYiS`fN@D_BTrH+$irfcVa*V<6LBprn>ZJ_^r5yfVQ(1JLgOWK!6%@xU{sO>>l zOr~S+%3w@!xrURR&t{uai2s&xH>*)P#`C^hLr)L%orjTRz7oAi6Q{gAt@M0ALoj&P zSWx;>U(nYuT~M9(faVM_#a`YI%kD~qoOm5pu1Urm?m~SrBG_2)fVG{im9%rXg!?fQJS#ksKc@;vuXGZr-6TPFw& zTO@e!LQU}cN&)ZWn4r&aH}oED!_RpeFmY5UWS%;qZ)-5Cy^+KwpIgkvE3RQdmoKpU zPp&bod5XN7v@EQneA2F%It@kvlI>GQnZnis2x!>Md!m8pbU2UOtL^Z#(Fe6>;!wJX_nbr;AZhy=J#8UaUnma6Pcz_g z#s<#^g0Mw}&p^f9*KZ+ip(B|H@L8A7{MU46sn!xd@g%id7RF zkWw`Xp%=%%LDm${M}=d8#}#yBh2li*a5zlej;7{2$aq(Sb4yFG&L|(h{+`CTszz4i zt%9sWhN%CkjC+q-)ScaZyGdSetiP=Y@fxR=puFFwy*gW{5#*fruUMhQld z_!B9ro>vJ&`)bH<5TeBoW{``)aL)671KD|@P~x11UABL)IxGqCivkg=-K4&a+)Mc5kw;n6)+RRQ= zFJTj+Em_M$Eq46<2t4yL;aTfN_;YA6RyXM(Uuz^D`KqDJI2aq`L-FeIC7c}Ti4$u2 zIP87+P73!E~K9sEb zI75-|b~f?<7wyZ@6;cXdIBJ3g-a-@?G>JdEDjgH0;aXz2el z!$Q?~JOV+Qmx%qB7af>q3WQ824FTflOv+V%Fm)xi$c5LlXCK<-D*4Gj!}w4*fI{pD$} z{UC{p2h+@Y7rN)AK-;)u^Vo#B2%5T^bB_a{GBOv7gp=TYWGXsLg|Rr)1LglEqVRPI zgg0M>ROA}GRJ_bQDtN{<`ZlYb%AYTvwQQl(XlB!KniNIf(C|scbnuuTJ#n_94G&d# z@118>k58ddbIs}Gw2#<2B_4mD*$jt#N6!{kQ&X7A$sS%`la8x_5X{atJWHv9rY z9^Qtc*;A~xl%XnquHIGoirw2NfQHO|G&t;m<9t;dEofq&V@DwB#X1a@9YniN6bw~u zG5*>hvwpaQcAU9JLZUCo;Zhd$8LuW4k9+v#%%8^<=P0f`mXvxuX?(sKeR3{Fi05=9 z51fXUd^~am+W4TLkHbIZaZ`6Ab=a4Z&$&EG@kyY{_iJd?XCb;+z&)KS`1k4O2-=#a zO=6KCw`<=}u*(lqS2g1_=Y84VuYl>*DtK&-tM9C#Bxwy&EqMfgsc?uW z9wB$j>$F3VK-1T*CC`y+)IOyOUrHs&_M$wM^PQvA@00lZ!-+eZjj(pJ418AYgu_N} zKFI+r5v=>Eox|Hp)h&sDYkn+-anirx^<#Z1#L*%K!bt-q5 zjHC~T+F@|N6w(@#NmJL8Ttj^6Z>=XC44Xms?>6JL(rsq&avc83Ekuf756ll4;8yz? zR`J)C9W~BjCLs-MVsIO4pVZ3M_!h9`U%jyECeQredV*fwgZZ{S5{Ig1LRxT|#hOmS zlIQ#35SxI|XXRMV-Car}$Dw-SDTLVt!9)89ep{)drm2BBDBOaL*?3xhNrP^k8$%~~ zAM(YYJ&3>g0nfKgB7fffu^z5M5kGq|Vf#sV`+Q;ro=Z{v-zkV~yp9=rGf;5D0(HwC zz+#y!jn&toDfet?VT3Kox@ytjj?n@IwpVarpT8i_WuqX(Z?wQDHIww;He=`)ezx3g zPFKX1k#o&7ido0+Qv;jX2H$+{I)BMT=KW+Bc!p?CNFq}!*F=++7e+kW02!6Ta6dj9 ztERQHq7U-8zM6CMRz|=xnzPKbKC(A8N$kofD~xqX!120tIH@L~XR zstFCB6zM_oAFOgagk;MRaF{+q@I}*2pk?kYNVJ$Q*e)k5IDavY4)vsA!|Jzuha^Em zRn#bBqZW-=^c21ajACXLEh%aPGyerJY!K zJ_ZlJ>_Rbr?lv|nQK-mlGI_F*CVW<=ug%;?yWg6w)~%w+!&Pb5+dDY@EeSt@OYwa+ z(KnUDGE9IHm$+~zTa{>@&;p0m2z&84<6srL-mgbtk1uht@!nd-S`m8 z42m|f%)q~TX0ea;Twmts1+bfX=>j44D}59T&bX)-teb>a8(Xn>^l#MhnQ%vuKQ_4- zLVBAFj7A;eK0^n7=CMY2DL?n{-_ggpg-o|m6h>D@;7rF9&T%os?^z9O%cC%sy0n#T zcaLPBswCOVXN-laO0b;wlOSBM4K6&xrFPK-XZN~7^Er1&f9IK2%c!TSmpu$(&ycJI8=QoI#Ax0s>$3HKxXXkc3I51H;-Ih;GH2DJWSO)6Eac+_@Q zKadIkn{u?=M1>l&CX-H(0*!6tXZV3B5W2(pA;x!6uXqz>%kt6roo9}0hmiHkHPjS( zm>h2{rW)&dJW=D0=||78zg(2gq5`oNw%C7E7ZJx4;eYBI9vT?Yr{)b*GU!Z+ozJ+# zF%U5`xjX7=BIh0@qEz1->2uBSr$87(l6Xz}OfoB>K;SY=O(Qawp|$px}Z z+fGTgwb-zG99hhnLn~ITCpjBun$<5(;t%rRHc^>cS7=jp0Kc!8l%mo32-0IcV6U5p z+qtO-bWXt2tZTR;w-ZiMRWRB&k?6cBHLVy+tM`|{JJba=7x*qJx`{5n>!+@lT_k?> zE}0KsP7fbfu@%QnU{rYm3KtjS-qI&*?~Gh#IIAn(_ta93z^~7 z9=(3fT{Sl095C0j_;?}^UI~{WWOfSIwIXQ!`c6u^(nitmdGBb*3##`IqvRi3S-jXq zR&s&&`6_QQp(E=V6qQ-w#!A+r{~wOM^@3Gr2>f{_Hry(oJxw;D+{YKGEcph>Ds87L z>%L&%u^+I{5B{#7pg4UhIrPpV4<}VRdM^(>YiB}*KRZ7@orQzjWds%^Le*mf-X=A% zh2MJF6!D>G+)>X`C*EKyh9|R@ua_uvb~m}jwbSXGM^r0TMXK@#NLNytcKO$E*wpo@l=|WIf5D2@nIK-?E;ZDMtFu2E^+wvw< zy>lg<>9nJ_-?gdOtb+G-ZAtFLAqq*1pu~1R`cUmm8EqnDa%nBxd;IV+Ac|+eyovib*Fae%xsiaa>N}4@K!+)&=-4mLyRzCHj{>i_*Hb)0RO!^3>aj z>$7v={xu&S3eU04B_2adSE0eG2ph%@BZ+o#>J;ySyv1{fgvH_QnUiF^uYlIt)X<%j zOv;-POltS7sN{?golDoIt2z?2(4Y?i58HX)is#rn02$q2th*b9i|fKTW5ji>ie=-F>{?rKe1hoiet{V`%dRZ!X)>MnQC~2B(%d?zMm8m&gi$+FQA9I$IB5gI&q!fP}61afM<~@SP)pQnO@Qj7Gj(}Xu5cnqZ zuFR=Y7SwYYYEQo6L9i6%e3(F5E;Gna=o<>Vg=zFi4bJAAPAA%oXzcLuboumcq^9kl zYc8==BppY4bHmBeHJt8v64^B*uwd~XHtGs@CUwZ;l&2U%b3)mkXXPx}V;HjfyIJqU zKWybP1-zMC%T{a%=Vz-7>{Y6U@S1zrR(v0W$)~WCbJ?`tnDhIO1NXNb!sjFDut;h_ z&QhKa>e55!)Wt|P+kn4D1k2nn%(bBbfzpyhDf0A?Gut1W=S1mSXYphkKVL38O_TkO z(F&&x^fO~E?d=&)<9K#SU-}iEp5k5BC{6mePn-I82+@q{=LqCkJ*Qrf-El{naAzMa z{5PL69TWuz1kQqG#d`%ZQQHMKZrKZBCW;GQiezDqOh4@ROVP8=nY5r%l)mpgfs-c` zSn8B0Ho82U3ID8QX1j{m=6wOIV=V8kjuof6SwFBPQId}R{fV2}SD>>_4vyXiFg9F> zt`@$xSJptyuP^NB+!Y9&p8=ngTDXsVft6EBQJrIrw?6I&mAM19?kjZHKf_TMgOvuXmfWSDi++qjO}&w?3x0Y7q? zqn8pi`}J_}rU7KK4Y0Yhn+deEFsbeegooxJwB-$&|9<3NgA=%RVmthVIOl-xaethM zfn&@yIKJ3{^G@oB{9uKG5_cRfIF9P3O;Fe}4k2?!(8q$cWHir{e(yL(4)M7(Gwc+t zRgXe>8{Z+X%7xd{YFvMphRZkiVUPI~_^vw&mE#vUt2-0l9b%zSzJhZ}d7?gU}dG?b-q zwHDadtDt{sBpNho`2MXEH9I@u`!tDnb0^bQ+m+;cX$w`guOQ1C26T4p14NpYKrQkY zg5EY^1LwL)m#xD>2YbX!?nZjHCP}-kB89qTWVn+1=eAbBWXMb;^1R@*l5iw&9)8%g z{cw8iiitUy>{81OCda)ZQ!DSV=_#cwBsP+5EB_xu=N*@08;0Sw(x5@1qNGLWz0d0< zDus|0MN3kNtc)^C6ct5FQ$xcDEfmr~OGqh{t&;g=B!xo0`+q;e`##ToU)OmahaG%P za?n$niT3VOsA#R?9hYc4xv_&gd3grp$#aZ+_y%L8^Re*RHV8a7!nW}do)1x^#V%I# z=IACGt9Y1j+nUyN+@Mn;o%G61UAWa!OW5(WjRsrV!SLY@}8OVPWC<}itU&g!L;~BT`qVP1Xe%l4t+o?!vjlv7-ov_L&XU`7lv;OUYxWoCQ*YCeWpwV#pVBZJh z-!53ad^pPNFK|Xd84O*0K~~}zV<^j9#vQ?nRgZABh3~(NN>M){2yH#&$!OJM{O-Gh zO&?UrHCT@RslPzC$RSuR%whYy_^g{df`=cRO7C@_U{Fp5E^asm(+O?#eePJH`J3Uw zQ%B2aBQL6TvYGU4@n&+K5Jf|8YLeZ$T$n_zfQ$8B*m_rCw}cMu9qmG%cZ|t$*+qDY z8RE>y9YQ;adf{%(FGA0o--Om@uL_Oyw+NSPoJ8$@?v$GuMBUN{NdNU7Dx4`tg_H9z zYLOI)N~lqD&tTepV=^MW;sxUQi|OUDSSq&9BWd4@G%iL+V^(g*&oncL@f?==t_!f5 zagz6(T)}=evbwfD&Ra5}#@{<=)Qv+VR%cFyk3*m>Ig;cmH_;Yv6S7b#gyk}ZH=euk z zbHe}9ilA!2uMywM*}bWE5nDBg+UyO;F!3q(5yd0h*PT_xaoBXaUYA2lt-*9RVJ5c?WhMd1eLExY6oV@OA~xDP=-nSB8tC z+8*D8zo{S^0pl^N;e=9>WGjFlKlQ- zh!b9u@asI!(T=jNfh1vx{RK?or4%{WC5;$n8GudIy*&arEMdscIRVyfiEOK5jHMz7*1=YlCPS&+2K3YOuq0x0rq?&m(rkU|2{k)P@Y= z_jSJe)~aUrL-i1vxdWz>)_BqDz|_At3WRB+aB#jQzuUcMjZRBg(uj7!@1P?B&rBR5|9Q+JHiz*?bdInEShoI5Kl^O8w!=RQ!G~`DSop*jp3SpJh{(TZP z@VQY*k1E~!G?Xrk>xA#pb)>4-a@how2C?3h3()sxCBZpqq zcDvW&Kv)2lWz^$a*(Yq?tVH`YMv-|A_XxXgC)$3NluRNiPeX;SFVAI{OQt~gqa2Ov znm|^~36O5fWkszQm`#v2=Sxhb==X`_CVHH-Ce5Hxk1H^UcPQwIByCSNqhE))6Q(c> zqu&hzO*zSI7Te;UxWj`f@1xvQnK=HW&g&K^dyKZ{a=^{aWJP$EE z;Vw2D7|ppOQ^@V(58S_`jWaH`EH@<%n({xn+hYj@I!Dq}F=sl_+>5m4G!kqZDr|bW zK-gO9A#5~e!p!Cq6!Y{PI=il*=Hdx-^d!N?u$>)SA!LWAUm>rzLxdaKCJE)Y4i#pt z*h$~KEpXJj9NE0@*0ba|rdbb1Vx2h)zud%*)_GyQM>wWD^G2`ea=0GSL_yOF_OrPG zy%ys1(kmaY!@N1$eIiE9(}3lI4xITkhXO8#P@ruBN%<#}piYC<+H$Tzpd6(d^Ybfr z+MYbDPBnVDU}mnAVQ_*L?m0_~@0jy<*-c#Cs)EGLij;d~H|Nv`kPmm~eLby6ZrjeH zwo;FFZFHmo&LMMt^%N0%_>9MWH(K>y&>ItdVasqmp)lqlbu{|Z#-3W}tQ!kSJ|Ay# zO+`}feC*xPD6kkZPO!D)G*UU+X}s4e8eyJA#ar@d_9ZX!k4s0QbvLY&#YlFF6z75S z{T{w!m_$Q@rcj}q1l_on24`t2l8QdecLEV~+{cnC3adD?`80Fgkpg*F z{@i@ClQPfp-@iVA_NGmyF`?qfo9~6VjS)ycXoL#ByKOO(VfvL-7-6i&yVzpXyQdX* z@D!p!2QYJkdi;7nbri(WTvr>is?>*3%|?9WzK)$)!?0(;7~v{kj3X(HQsMlkEEsLF!=`^ru**IHx}yit;~_SG&SzuF-1YyLJT7dTx=4>ztsV3 znCofiHFYFQ8>YGy1tp zi`?w5;`+8P_*!I6r7{O-hs+VWR1`^9FB5%OZiLqx?4fbk9uKTnW3`tl9CE}^>3<3m zwx98#rV0HAu41QnK88tcK(lc)1ENAQ~osCSC#(6e8#A8vZUtl1aaNb zFdMuQ@oQ(HcnY6;IZUGr4SOn(b);jn%qe2dXB=DZA$+)`M0l_Kk?`lB55lwEzlDis zPYHLPlchN2sigVVotljgl2=qTZMPx1p>>VDemw!fqI%eK4AA}}iP>MP*#OTLdpwy) zDLO_pf47i+{ux0txXZ`=O99JYJq8wsrb6bfIWC)Mp?_K{8zMRr>Q_?G%AE|M`F(IJ zc!|5;lHfb&C2Siek&feBvRprnPMwvegulh;E6gFSif>e!H&}S9LQ44i?>$m$^`Y@= zirL5RpKS3NF|2VFgK5Z3jyMQo#zUN#zj8X;_uozCb>u1AZBv36ZeJH1J7Is60$rXntdm zFr}-CtP|o%XQDmi{j#va{t~WcCEz^Y&$nza!KlPL?5MpoUQc?#hIw=_qhwWh?Hr4H z|IKE*;}e;|y)3@-KhMs$7qiRTw=ySN-VM@uhct&j$i1RK1qIUd#eh3jTlk%^`UFWW zJVk09Vf49eI(2Bu)6BI_Y{BPxHg}E`wspN>U5Db>*bOGkp)>|bY9iETHi)xN{~%63nNnTMz*g7rjbNtIrPCLuyoLI=*=M7t`lHa`vG(=yEB4_vG;Ix(4e*ZP|vyn&dk6GBVd=-=qM&Z~PzK^$!N6WK+@Z2j& zTazTn^`B4m zfyLbCmdZXSi{oDg?^m2-P&ujrGu&kNnx3;Moobjf>pOEAAkG@;%Z#cyn?f;uOS#BK@uve1pgZRkvPHFR0+&zQ$e|+D;Ad7 zVoHYw-u~?u^u#$Z-MkoPJ(_3wN+Vd<`{^vIX$s1|2jWu8L2SMh#QnD`QE=4|F-M-W zTD3tCEKx^wvmD9~|HlrUJkRtSFGAwU16+M3Nq>3&{;DybmCczCRy_;RJl7=mG#o}N zW04S-j1EILX|>FY7zJ7944RU=hvVkh-6L>x8whB@tSZ6 zj>Dr1&oQX39giJ4AlsM)PiJ%jyPEs3<%qL`Y?vsepXGGC1xc75k)thEn7;hf!2z8Z%8P67CN zO$}uJ4lYwC(S7}`l=J5>^_bkI-_=Q^v-LP#8`??B(+3NOjvFET;I1fKy0CTsIYdB@VtF5Vem#a+kt53x(?2Qu6F z%;>;93^DRS^uQ@7T6MvFwJ7cB7U4XIN4#r)7&+H1ku!ZSswX+)TKjPPyFQ%Hy&kg4 zIaeURx)nS12T}fKO_I=NH1h1lAF~5V5F~t`dy*J<^sOI_l3K$QsLIY>Z7Vz$8O?L}67jSo!;a|);`5eKi z9>~(RqZs?M?~pL|EMzk-ou!@x*hI`Jve46LPwAMhvj^x z6Fp}T`XgQ8H`oQG^7g38GsUaS!4O;ZfyvrV!;G_mI5BrWZuiZ?GTmX&*`kBO-Z)4+ zKLqK3c>@ z@ci1@e9XDaGomrU(25SlYEKu;PcVm>m;}x}Uch+ck)^yk%ToSbWIZ*T*=r{iCNc69 zi@7KQ$z|sFJj)*6wXdurF=+i(p;#YM1$%1_Kb{U-BG z`N34~i6Fl}o-JFKfaE&vU^V1#P8^&O`!&8t1(fW~dX>o`;{TtVU86siW z*7lVZ?ivqQ6CI519Sa@qidk&e$i|g6fs@-PU;YnL|N9K1^3$MhQ-~%`#qQ(Lhkqdh%a7!Nk+h!op`a)0>AdEUadu^hCfVt9?&dLV_|fInQ-|WSD}aJT4DEDePMtQXpY*2>Yd@}S#tv) zPFFyyCKEf|oY?+@L^HBN3R#r^J*!Ydmj68tjv$r6|d?|j0YT&!wCUj*5 zAwH95p3W|UU}6_*y~X<=@26Ai`#H3m&r#1Qa&EN6J*Wse1&gwD+3qA|_AgzTje;cG zn%OMav+agJ@r((Jo-l!>KAXp$$oR6e8fwhJX9%`$GsBy$rpUjw2NK`4VQ{;Qb*Zjo za+;;A(xaMZ0-v*lJBCo+sDfmkZBFJHUF<4iC-x_?<(}TG`=<=6uu{NT-dCI}8;Nb4 zuXxL2HdbU6GMB~Nn^w&Cz^3hJOt^uD_34Nm?tmGuW!dt9Fy_A~hsnPE%VwSDe2|p; zEXY9%`E8r9a@Jn_lZk+-OD5cZ9l_aECn#xKH@!Xem)s1#lg+mq^kUv#%Dy`u8;WAl zlM#u6f9DZy?u2yD8kTrEf}Lp}1;5taNQra8B5Nmv=_umxoKRN!_aN_q*s>e%GT7;$ zXx1@pE$h9K!W@6+v7djWVd8R+HKb2xX`wxWkf=Q7%sC3A{2$Z1_nPe;a*M?xiLO)nYXE~~@rRxS$?DAr>stoY2-V5K}EQa{>35X3H3W+0M*>y_=HZw$nMXw#h`bLjn z>kIk?KMHpU!ri?E>f;s*3WR}zgyJy4sJ=vj&FT39k2QaoxQ`yjzSqW>IpUo2#JT4S zjx%wOOtyPr1*`qRI|IiCaVE@Is1<7>W1hzt1AyaPMD>EO&kRqW*jqb~|Nb zzFh{wt5+cK(*`!P=M2k^JeHSQB0L&abIgNy%p_O-K)O^;H-$dlSYdN)gQD`%0DR3VdEnr8M zEnxSmn+5yAwGlYR1p(^5h&6VBW8E^&02PDFvRJmZw3$6A5y5s5?kY-Az?J=#Ol?OL z#MK`lxb7jEA}hd(j$lnXFl@|3+<&|g5A;W1oA5eokBwjz7nZSg6Bi@MWFw}#AHx>K zD0DeZRy`=o^}ni0V2%B5(APmE29>V?|(=k8#AzNHB7Ong& zUD5gpGB4H0FnJc`2I!JvZ#Vj`^0`mPKBCYh%DJ3MD*`>~KIe|SkJ`jIFPC>*Mq!U? zHk%YWo(Ve7!06Xo6t9z{(!27s?t%!#Se$|3^$K>|Ob>p81emCAh1I7x2fP0ayRZL| znT2o`(trW#T$FK=GbD!CXR*HV$t>hdGy5v{gk|l&%4DaC^(MxKu_6f1dDsaSJM*NGoP_ZyFWrT-ifa2 zIZ%Gw5_;o4l`MHCBVcm6z|!D}peE~HAG3G=twbGnq;)Krhkdh*Fn{(N)(;6B z_ZrKy?25Sh;34xE$YHO)tYqK+>0vg{S^vG7hoE!S7#Ep~R}*L9QnLp;k=DlM$4|uW z5jaI(Z&y%?DU=!cX?^o`mqyF9Ts=tE8~fyqcWUUy*5E zIYlHZQ=9b?>XKbXS>cKl?$Zb5^4Az;{Dt)$T!D=f86q;oVLkL0yPp1tt$HZ}iOdC% z7?aHRb`LQB=L4Mg-NN(dIT*6^I}#_1q%ZqAP;~YPlGcv^E-b>)#43oWRm0~~7u0`p zH}R@)XowA=D|(yBbnSc^%ef|wHQc@NJq-#qt0??w8tL_>lGA>7$`sm?FVCLI-9L;6 zySRfP_#`gP=AEI8J+SUxh?zxvCjH)zw(@;^Sp}oD783LW{n&3%g=>6&?0M`U=|^Qz z^S2BVZ&sl#%^on>eH53u|Dw{0^Cxe3Vi)H@W=-tF3B%zePk%6E68BE)=HjYaDvFc~ z>1Nd%I$n5>^0dNe`}<8aVW|}r@0OvkKNd7+j|3@eU&aORvnV)q3R4wgu)X*LB2wPt z$ifyl%AZ8|;R6_F<0w>|d`)=J^^Nf3)wjZ7OWTA^k_qcK@$&IDEh9UgVu%-e*2b;#j-OHgb~2>gDiO0| z8l{c(AjgMcWU?xZBrKI_+4~8?w==_qv*#ZcZnd!&`cw#oYC*$=;qhKjG|I)?d*u3N#n>c$h3a|3{!~%aj>UPFv-s945xPi5ExdJzQE~)CNQ{CMsnE8Dd z-_s6;Kkq(_J-!*0B12JnLXOV?PBHo5!}zhh3@!ZE)p+#tJX|~~sz;)(iMy0$D$piX z0najMQH1wXJo5>L$74C6O}VY`>SZ@!UW}1&=tOm)ep5G<=^wy^w#R5m{)st{BuV65 zAI|5b!AHcNX|60|F;Vf%*e!?6O1sVW>8)X?7r~op#<+jT2jAQm!#VdATYjsCDVA+Q ze(NcWyW5H}&lTy_MMmM*2UER((cFDOH1m%;74f`P&*ur`)TB%Qu2kch%W|rlypNJr zZK3w{!>B{G43`!8=dnteTn&fP>@~cnZdHn{=|`~C=^3Q&^unzE89b}w@lJ6EI?Ej} zDPTD3{jrUmE;_=->;hEDUIPuxj z#C9DVXj=gpc2>3D6}Z;fL%t8$ob%}OSFUXvXP zyDX@ppMu->Mq`HR1_X09w)*aaD44tgV}0e2aYqcFrme>c=_ITTy@}2vw)o=wg@x$M zhM6Fc=hqKIZ&U&_(hq>H3o$iV4NtG`#k`CV9N)hJab`=f*SCSm3+nO0+=z4+tf$xB zF_d`Tp9Y1=QpgG+9xKJ+#iL3Dzb}L8u#0$Ey#TRoR`@t07xn9Jz}T@Gr)Q@^vx9KI zD+00$YhhDZhXoxG5aoR5@EvZ5zTLuFc%E+fk?Ano`;Wc9I~ZSd%9%-SDGYe-c}lP< zt?nF3#U;N`vf(m9ws8+*?r~@>s>aujHaL7#CfWEbY+EQsV+LE0vV;ZwdLT(=JbTrq z#@(&4Jy_N{gi@c~gS-DZe)dYoviqm`&-5EJdEdkR>tdqsibO(QVrlNSaEM zvpb``Vsq)T>ohVF|BJ>(zIUop=NTuW1~WxUu9T&iiP!M9F$ROn>!A1T3O4?3gIGc{ znhr(b_ySG(nRb}|gz;=mMi5QeJA?a<6zJvI6skyUq3zWVC{yw>-7?Om#x5^95x)gV zcX-CP;SnN#9LK6>Ji8Y@3(i$LAU8vfN@SMMe6Pb~cq@{+UcSV=@@QDke2-cUd7_8^ zFr%dbZ&M?&dRGo+&-{*{hCfJ8dWYHn#7XnwS?pQcjc+HWaR$*M`rR^u%!0)@N4^pp zv_jG1p+jn857DlPaWsDS2|BcABrWo96WnL3Sigch3%InD9eEeVB;`l3Ka_!QYNarn zn+1p33lLrS9O}38!Q5o%<+B-dX`K%pe|4B)p^NCP@^p6IEtwhLf5rBv*0I}Z zG3@Qt9qhsyOC*Shy*r7vyVwn$!BFU z+u4@t{cPZ=B|AQ84*8aCB9BZ{nxbk=PaAz`R;dZ8oaVj&8KUW&C7muQO9x7%NUJ58 zd%Wb4r{aYfa6>IWyS;nA5W(dV7+(IIeOsA`-K;f@V5lbNvlyAIRW0gf+11QzqkLk zpsafl-h7(}rw7vX{?%LxEE-97UkzYRn>wj}E`^cm9+Fedq+1HnibX zBU791N-EKhrgz9|QX~&!Rq74v#DYKzl(%!PiQqM^E~v!s$dlN}v-o2ZACa@2xiE3X zYN6tw-9nL7;ldBTOqhKq1;vtW+zb7Y?*Ln1apf~!yYO63q!{Wx@ZH0KH5mNF3Ds%V zsQ#dZkC${IYIzF%AFttS;|b`UKY~jmXCSNU5L+Admc%<7M8mxr2F@W>+%SkB-83xeP3A$i(EXI3%mAp@!@TDs%IvC%;@t z7`&aH-d;~nJL&`J<7c=I?E-yDTAXA2$77!ReC(@gF{JdSwE&;<1%wB}hS6cip}8{g$LR!Rxm zPFV=+vnC5;Z|Vwd3j{*T^&crb@;p;(zrm8ft6}c%2{_z17ANl1u{S!7Ow{2d%a8G7 zU+?%bQNLTPZBio3nRrIfqBetV=rUz0ekyF*Gz(U!JB^)tv=g6W@1xaXGI{adgo(o( zTCUIrX^~QRchtdPP$B9M-$sM>54_r#26>4LY<1D6YYa3)S(ywzt5M|22k7GLypa+L zG&!OT5r*yXSuM`9kB^|3oQ#@hNpN47g~O)ju`010`+L5i#HURVpX9(^2NcbV`rd4dVod5zGk*7 zX0u(r2U%>zAvUntk9DQRvwpKU&h9?Va#!AG-cj~!)sxu_nT*+)8#0CDKP*qU1-^Po zNIDaaWWGCi7dR7prkbXwH_-2hYOex8a~F`!>OU;H!zw)dvu*Ka3wKBbQ- z#fzgZzKd=BGa7dqM3FMY2{j|?;k$^lmUN17NW2u|?FyjKYKQoCPuwus2aAM5c>OF6 zKP``8Ohy(n3(jW^9yKg`P!Ln|KEWOsZ)NYVj6sI+Dm3-)L3_j@2v6*R)KXityfcB^ z$wW9w^L-L`fV+wwg|Wq0Tv51-De)ueQceSkM&E<3!w)={OMy&o723>m5NUf9rmCm# z;-NpbPn`naaa~y6vxo+HET*CqT^dxZLp27!VPSic=kDu}sk#@Mk8LSl-nfsJf=KHsrO=v#Hwu-7_;tyjC=HG*X?nn2h~iJS7Jf@8z;^S0tnZ>BR`f~Y%^m~VWf4!Z4o9hJ=yFXy{BQnaS1!6>hC~5+M4M0*{Rjt!?Z=hH zinJ-xf>up4rsIaH^nR2M<=K72KYq#aZwv7ijiB#%HG{ zJXgrXw~uFWGie`wbxo&AejY9I*-TB}7m(Z2Db&o_^>KppZ1Hwc4B6Vlc6_X0?n-Z% z*}4bp#g`7&s<9LAZsg;!{binoj)$v`4MzBXV&Z!IntF9OUYq>KYD)51PiiPj>CdI< zH-1yp;&)Vg`!alBp=C@RepZs;Cwz$P9Sr|8C18@l(t%|q=0Uab-xg ztuoy%A~JymJ^E@v^LEKl-dJN47CPX+t%opUN;DoB_#iaS1nZQ{p|R-@CXYYOzYjTZ z-xH6%BuiY#j$y&tkC8s&3#&dMceT2P8#ZsQ}Kq`;H%XnSiU%kHyu?tckwqSa5qiR?HIgVvWs`yvQXrB z9nasUVBD2>L`M6;s4ItE`Xh;#ode8&?|nA={6qGF^AZF~{LFc|7&k9I#_iey1iT2u zJL27de$L2TF`sg4){xsF16saPj$ZAn;eCD;p}XB$;k=ovg-3>23meh}!t?3hD4RPL zXFffId82qQe$Gh*zQ2Oo8-fs7(avK|sC|Ej{hNQB(X=8$$yAQ{ZrOWRxdvwMRyZETuCXH^&pBDlA9x-?A) z_Q?qL!`q+k;&g@h}onBw3fyfAo$%qDd{I}|0q z-8Il#6#>U8B`DsRht8Ka=y_s_+KAnVn6eP1S~j>jsu9|HT}X`-=l!5}nCTjaz6~Fl zI_Fyc^BT)c(^Ofmmn7Sl-XXYoR|>6v`Mu?JGGhwvY4 zlhi7-{o*~Ls(zg37uen2GGuz=1AKhOV8oy(C@c-c4!2b3>2r7a2TwfWpNA6D*PN{` zLs4<>@w@RkOkajWX7_0fbjnhH)E6xN^Azs=*Z8mB3i;Rl?1Ka$E!hhEQ3}b)5}1>H zi$!ur&GL<(vBFQ3RvU>?+=n0V7D+QmQOSD6Yu zTc}3=z59aM7A>&TO2k$7JJ{&;3q9eyHz%6_zsIXFSH=-fXK2u<4S{4a{{R&qnnm+! zCeq~-lJv_W2UdxF5Uy>3p(kh1cb`Mb87H`@#jz{9YT5heA9!y59^2zs&RT{#u_cek zQ#tm*nk&Nv=bY zO>E6#-`ew-#myru?&Lz&nsr}b8xsnh3Q<}=PMHp+kEW4RI&ir)0=Kuh(S%_sB-j~8 zx$F0mm9-glNh(vURV8!Xz(0$Rm*8}QJxu%UAbCy+n*$B-cKaE4b-uu%%66QXavLGH zqadODi8cAiVdTC__WV^9bIi$NO*i@ZYM_8tl@!wt%}Cmo8Be#)r_rge2{iNlGyELh zg4ruyz^b4dQSJ2*TTqA(9(vHbVT0^D`uKEIfRdrywenF08x|YkO>YcTjW1%w_)si0 z(!f#9`+w5+3Q4{)R6mWg$p7*l(tXZbUg-w!=^pqu`V>mbm*D7fCsZAX#2$5XxNiK7 z&7ET8l^T!Ot$8RN*#XU>XHed|iflgxlBbC=o$*knNL3?>o}@({d(4q;?}4eG4dJgj z15ax@nZmsHOfp7^Mu%uoOh!2dZk6ECrpM5*=)~5v7pT7_pbLMT>G8yk^kV1|dS0qT zzn+`I`7h7!yUc<>NfqmBG_W*h2#N~3p_DI1UAlF6Hu*M=r#E5gfl|oGhEZ)+7F`?? zM8O_`^uTl%iKLAq?JHu`&UbBk!>Uout)2 zlnlhpQb%0xRzN|yBp%MJW&fFMM?v3L48Hvi29skErz_-+TN9{ML@={G-`FjK-|UWT z88Z?+%=U+Ru|c($FwRz!PVha~RjF@Sd6Iu0Mx95tLm6%tzk*%wP26;7g1Nyp>~7`r zh4Gw?F)E3?AMK?pW6j9%`Fz@5JcX{`f6kKoIr|~3g;~_6Gr8Kk%xUFOmZqT$w=rh; zGRq8Gqla-%Nk5x>=?61>KLy7g?m%|@BtD8X_8Ax5722jhry9CVl7#?j0)*h%{0jI}0Ge{MzSs0XOX9?Y39 zqIAdf0ak^V<4nL{Va(JqLIrs#;iDiiVN#Nu@T0$&Fu=_iu8BS<8#xdB3I?&=&gjUQ zfMt>4_`0|j1|n5RY%Pb)!5hff8HJFKKCn~cz0>#_d@MKxtJ(~7Z!m{lLL&<+E@OcK zi<$P)AU1k-81pc`&iqDC#=Uh1_%$IKm`%*0P6V=VG_7FAP!?!6hqf_S$>w`rBlvy?l-^cQFbW zDNW07T*4E{1Dq8s$vNyJsZORCZA$l0SgCMhgC8}n9Wz(jPgLdvD4Bjln(l$3( zrU_wsl|RP{#3{0}1vxp?7{*5zEx(m%{**e|x;`4o~hzCp>a40Y;z@nKg4!XhL1 z8B_$$l1`jOYYR57zqVjk-dFs$U!HQzzaq)-JthT;)9PJU5h+^@vw4s4E+-b_IOjmlG6j1r zVz8^ApIMZOpu&Q4&uS}JL(o-LyXZJ`{;G~K<@S7DwH0D^PTV&tgC9qKGO1yEadyEK zKEFGNcJ7fcxv~%v9|yzd&mg*FJ)PeEw~ds<1E@46lm;od(>b@zbSb5tKI{_}>Ys0? zpiSBI@I@>|dhUmU-f0B01DIHDhQjR!ar|8brrTLj`pN{7yb(tMoV92%!;hVXh2g_z6Ddawb-(CF+M8BL)L-sZyhc| zQjm}86}i0k$T_heLaFcN6_T`1r$&k-8`T3;uWC+uMv6RpHxD&$41wh%pxUO6Q={Ip zFxh^VUVao7{_haCUyf{3#mM{JNu1J5g{98`5-;At^LZs$x~d3kh6Q7bcrr32?qj#f zJ#>G+i{{K1s0<5+ehmMA**lyw8h)YuXan}uJwRg5Wr&nH@|^W$y!#_hlUEv2aoIQu zmg|J^FfXRke3|tGh~r#MI!iLx!_-<wn!mFV*L>-e> z>tc>%&Z`;!Q*<7VSiWr-mQ^ANks>S7#(iEdN}{Aelqk`n ztd>fXlI)C7sO*^$l86wglvFB3(U6vs=C7hszWe(FDtRBC`@XL8JPt(K#6!z05%yvh zu-do?=eG_pp{?_&G4cW>&#t4uX_3_B?oJ)rGpJKPTkYHjGWw{iSd0^?{|Fl0=1;6h4Me!tHEBoc0ujS=LNk+7yK! zHJy0N?>%E`^N@GW5vxZPv0e3TY}m28Y^9tiZqZ$KVyh{;n#uiDS;MH)cpe$?Gf0ff zKIkk}f>O-_WIuC4WR5mWBf44fhejsqsl-(0*kaJDnES9!VB&y3B1bvkOqm*@`)^>_ zD{Wfy-Ix?gdZ9Fgdso(glUSAOF8zP8n zIVCt$eL?W>=>mOY#+&h*{idn=YwJKQtp$>Mwrh^p}d}W#1_qf+7 z5w@1Sh#lI4KX0|DJYJf1-;Bp&Wg%LXy^cQD_>fwdDXm#KohBGPhKwVR73v0HLHjX0 zEDOM>4eq$3IKae`+SsHQl4yDDhrb0*@E!cd#5$wcr`ixENr~+4wQ3gT)5vC?tzui9 z^4P_*JuEIp4!-jwu%kl(pQoH*f9Ciy?`!g?89op5$4o~B@4nrebOP0!RdRYv4@*{E zg5b?Z@gc(tp*zRobl!coQ}sR@*{zN|=Yvqbv;-BrpTDu7gq3LihSLvMvVOIaB#w@z z@g~A#;LwP#dj34K5C?nHcGR4dq|C{`5Hyi9Cf0n!MeX4<|9~jXwB^j6mr~SutqG3$ z)wIf1Q&4f$Sn!}{oItnmCtd11OVS#Dm{Fh!s>uRVW&E&st{L*Ce4FLO$s9;rk=+5jduKgdJ|U!E|$vu%)-t znd_`t_HncsE3gS?Ti^d>P9bAZ94iGyhfWrz9LLNy9fgfVHICmdLO$=kN#yt7i)Sp_ zmv*zuE9cSMJ-E=OG0;!G@aoN zqMmHd#<|{rntj3e&pH;@xkxB(>Hz^LY%eqZ%$gN*ZaF#FVFA>yP%_ponBfn z?U3ibciypx?_f`Mbh1FtL3X&kot>%r!PNQuPs1yQZQlBweHyex$B|LEUONnnMC2hp zGK)>w701lSh(TyMzXOd>fqVB47V_jY>$}W5WM_GI$$Svux3#GBiaafAy#}97CB(1a zj++N_A>7WrRv}xl_z`EnX5mT9gXH&DGHq{Fb z?^mHE>^Ix0mBV&AD#B;20qoO6q4(}GpIvWcj^-QSq|1qyBd$PvFZVh`B*R)#8Ompj zuy4s)*iAIY8V^sX?Qp`8n+j;Riblt&htMBCjH>%9P+GYL1NY1@x&JVRr}gk`tusEI zcRlTQ-AT`5UPG*N;R_sC?ft8X~o&l<~;7c zS#yhfzvX#X(Gn*zg%IX?iM4vBvn;-EDw~=LCC@@+4KIZJ4(`F_%oEYaN_ZD_9N#kb zAvV;8&rzMx(7y=9_uDb?sRpex(4eJ`#+;GAkmhe5#Hhc8YJbN#@TmAG5jNNV#0ftmbd^j=N-p3ix8+x-G+6l2hn(W4t(c-XUn2p z@p<2QBzEk>;Nlt3z4?JH*=0=={Y&Y~xeD4-ok!(80c3Phm4v+IF(GvqR{A}|zwc)- z{jCRFRmWrUyJlEPjG{}y1KiO%j#^e((&?Mr-*jpPM54msvHdKTH&x-zgEWl1yaWj< zwy?B`L{CQuR?0_0a%d6$8{k}^*K?7x{{#ldaLxPZ>#iu?=57F`W6y!hbzr%k;+}W|kb@uDs3aG9~e6&tixwY2t42IOr>kK&wbT zbDNS$CpNUvciy*u9(aMi^RvPxo=?5~A(DRPwve1tIh`3W9OHhgkH*6^vrIqm*abI_2mC&*%i55hBw{5yF53iJded;%wjTMl9_Rx-vOGp-hzr= z9i=Un!F(sTm&AF7JabMvhEIy-S&KL5{&fuphL$2qJq~YnxFDv@ldPHPYYgLQR81>+EPg~=Sv{?Kmrt)g*pdq~BF(*fY5!R_ zQk-T;VXuB6Ot%1I73Ps*^H!pV-ZYEfv2J_}r>KilDbI1OphUPr;P3NP(7fWWV8?n% zgXDA91*S7|5dEST^VKD2&QU}9>#>AnL-eSB`&+h?&k&bbEXDhV^*E8Wkh21WkRfY> zV#8!~uPMe@%lFXxa0x41{NU6P$5I}buu`ibcv&_cG9J>nV)~P*)lQ?^oQH2-P)iyb z*J$D23Ti)-OylaWV3@KNZJljKF2+ve-8zZnYCiBgvOTHH3L?=|KU(%Kh(;{hMfaP> z)9;Jf%x_l%8!hygJvVvA!Y{~SYV%oE_p+FsSt@~;R83SJP(ajPV72gTCU@!tzCPxR z{D7hKO;eI0UHeh~H45wU6}d}Q1DC5fpI~$VHg9Y}-i=m_y!n~k5nG4F{EU!4d?#K^ zHpN`;*K9n`N(JA(foa^Els!t4-dYWzB*Svt^V~t())vr@sAAgXUQ9>7pQ7_S%qUZ= zizQAoM&HR4M47k3*tP}5f#J}6{}eAZ$5Aln)sDI1OY4I|NcQ?dnmWf&@KFA!Afh%> zATD}PP&w_8K;p}A!KZ&g_&(5yj(i206RAd#{liIQbuE6azXNk;85*3bNG7utsjzK4 znXhU`V3`fhX=g%VV-==K+{f;qN(g+m!0fUR-Y-}OZ?7FV=sX9*nkBJ{Gbo1N3WK>% z9qM=ALcq0Lgoww$=Z+PkH9g3uA&1;r&(gy;IW+oe5y|g0q(UP{L5Wg{VC0TcL9Abq zAa8e$AX3ptQ1~zql9Gd1GDL?$u4t2{qZBpOmqRwKjfwX+vcM6^EOYl&7E|_IXz+K}5IQKboH_-ev_~e9d=?)dkLz|+c%c}R~BWu{+glwA*va~Uf5f?QH?c#hMrWoe($IC3qTLF>nI zpXQ=8v|23Ux#&ZfG};v#O$lYQhhTkME_-@9Pk%WHv!m}M+2+N=Seb+o`#i#e**70S zbK_Us86!{PUnNLlNheCCgd;sl1H*Y{FF?D#w3^ zCnI7UXRNO&hyJM!%sbeO1lfIf!G>bWj5RP^aTrDX-_&}r6T6;qKCZbNbDmkqrjNbH zs&|N>BjXJVi!5esL9gL3RgIdHz36Mw33{HLLzTU*B-|v0p+mhefeNG+;4 zdqKfhn07sPBEv&rlxV(;pG||QRAL6*n#Fg4+@JR6Wi<+xRbzj&4+8#};Kh#-xEN^w z_k*@*P_{m*@-2)aEm%_vom}ojs4l~&sD}t?Z9=B+ z3mg-wfq}d-ZTRg*H6z_g{NySsSZF|*D!&jrN{1#i@m;a}SbFzGioS%`p?K;RzPH## zA+DFm`{`{uy}6o7$28D1pCz=5=OTO#cVcjF6Xwf5gM`vqbme=(vsWEu#g-_az7_Xc zBk|}-2(r$9Vz2ywyM3+@I0T}mF98=mBtUAH9Zn4L&se}S{=2L2SL-_BhZiGn_$s7o za5nVQV&-q@HAHTIGnXUna6%?T2C%xi)d0nE7ksZ%y~=qsYS$#=2h{2Gp*yDerK3=I^mlg z;kUFXX0F@=6aI5L%^?m?M4~aK$OL7=LTDEqLVx89s5nrc94wV+=>qO;7SyFDWA@BQ+A@th_>OeZ>xw(nWb}aU`fMVd z!5Es^-%Gna#RU1+zfpJj4Z6D4mqzqUV2}82#O%&MrAauvHtolsHZ24l`wjm?Uy=Gj zoMx!aA<>l{RQSk%CP#4}wwWl!9~w=o)x~LC^KcSfSBjkLwQzkcK@Tj%=?w2ZY?`1z z36;X+njk|;S|)U+WfSFkI*`S$Ib_c{j1T=%(B@HzsDVDbxi*FjXWCN!R$ZFHv$?WI z5?FoZGp2Fx8QU}KCv)NV2ovvEWJJ|rqE;;))^IMx=|ZTfAAu|Hayu+9qJ6fNv{$>9 z+?Q5!SMFgNG)vd7R2jj%EOnXPL1*?QcO2^+cS`?~;V|-!(PVnc5!p<99(In_BmpO)z}UA|BsmVspz`a84;x<+G|v z({&h?umNP{k6~x`Kw{QPyy%oe8BJMZ$eA7}^=KddR`mz)*&zC`2{Uq`Nb~s)gGJ)L1?`-?Sc$T5aFfu$02R4-Aj-U|NPH({?vwB<^t;x^5 zOX$u93yQIvPQry!wA*<+sY>(Na>Zt{5^SZqiX7VhIf^dbFlVB&nXGBZ9d_$r4-1kN z$2*f4rV}X6IycxeXO(1jAiav&<>av+SC6tK&tEW4g-H-~n}x1JV_|DP0XhETk>6g< zq<&muqNlr9=&nh~8MztB+N%-za6Eh9xRgm0JF(gTGuGbY&RX6@vIM=eP&d)0l?QYv zbB`(!_foo>E0W>P(X?yK9*Sx_Or^Ir(t%Iz)c9l}$*t052}>+kSHN`^EB23VeIx~i zV`3QgQU`{-;CJ;KcgYL;Kx5}wzS|DOyu|-#WwwPNRdbDCL4%th5T1hTZ8HU}D~Ds& zk3D#7cmhc!;c)WZjLvDMSoqYH>F!Bq8w=}Ln`1YtPwrwdzsi}2yByX%vB&J20Z=>} zgXy0m@nzp8^vum-fdvw{-9HkuU;bm>^4-j`Esf1EZzTbS}4V1!~CCNFcvvBNr8 z+BOx9yX!FW>I*n&y}^Q}hgjQt2}fo&lcj@>;IE&lpn8n1AYdz>hic0Ssto^QvLUh% zl2U+G@pug1;03um1Xq!BOvL&*yQ&z^M#kl^;&C0Ue^D$`+^ojL+QQiv`@?KcyEA)h zUdv{E$Y$4qLQtlD8Lx6bW9leXTCZwGI$}~JJLM~E##qzg^x1TGuOm(W6;1CA50iAg zB^KB79rcNkG}%y|Y{p8``U5R!v|fTT?G&uo@flI~Z{iByb*_1J9ENuyq1}BMp32z> zJD7%X<4dvU@)5*c|4(0Lw?1prpT>^PpTg{R&Sf2I4cKr=Rg6iCdT4jNo^=Wu!bQ1YI*Q|KP10#R?!}r*7j6HZ2wSu3xa;Y14+n!)@!6`hc z8H>WiU3kmy2QlsU`952SJf5e+-}M#qTcOF_=YbG2b;g>vewmIa$d15zq(k> z{7cMb_;_s1$i&q0T*UYB9`w&asE@A2{1`tB|L+3!eC@>w-ea8IIGluEy~L+6<_$DWnM^a3znotdmC4^Ehcb zZ7M^~uR7sUZ^U<*6*zYMIA&L7LNr&D&rcpB>7)?bVls;P{gz_6V`ehlC9_ze{uuVC z?k%%=J`KD1Ea*Veer)->3VIE^Bb_=7GxuyqSWz-O4@cp^NxrX&jYjl!!cg~>_-3An z1v6fv#D6A59nd40&mgzn5Xzr>iKMR#5g2{>M{iE%&^94OC~iFfW2;zb+zG~c`4`X% zYsEF&yBPmiK(o3cY5(gq3b2i(qY)02B=U!ylA8(9-xIKYgC@Ke1)yhGBHny#Mo`NP z-t9J|$;o<@%)eKfm%o8O(I}g+4TIOiuy$1>f_Q$i?6f`4sBj+kr#!q_9F40{wRo{^ zD7{?t4uZ~JrnG+<%J&9fz1JB$wDgj4sE0OUsBYJy6oxcB?NEg$;W5Xg( zG&VFr^)qLy<-f*{=x2B@T#p5=pAj12Mb@jfQ6}xAfGykT>mC!j-_O6_9yXlAypDeC zTuEmqS&^gPBwBNSCtIxD%q}>*W^t*VEPrkndu?-#iNA?PasLBc`f&<%tN4E7rwrxI zt--N`P6TjXRsR}23b2{MImecCHCvvZ*aag;?ju4RyV0?WGcBygk_7K!iT+xQ`odO3 z%zS~ZUwB`9ZWU%!r6X0_0_!vmqeaLUCT&@`;{Fe>Z7<W zF_D?fQ((`<=Rk4JF&yyQi@yEGcpt71eg(;hb-YWXKIsZX9cKs}@8}40hN%j6DR+_} z*??^*t6~i^15h`hkX6=(3*-Md6?BJ1MIJC5nxmgKe?|fk_p0pB00|lru@8L|7 z4j7z1hWzjhCXl(x6wd@Ol~My%Fn137_gtRMFA5=z>rcpRZ5yqOE1{Cye0n@IfJ|0! zFC1rtg}l1Wd3pCxr`>>Y^$pmtXEh6z>|#&${97HPpO@iWW$VukT};rveJg=6n!vB4{s$y91!VfGPBpK%P49TCV#%!1^W zAT;p(wXi`q3g!&pcZwptbsa)W_3mRt$5i?*bCe`P<7w&ZXqvt^nEn|Uk$UYL7UVb? z#g()1)F+af$pc#j43 z|G{edUl=asu9!AqLHe2v0*}66!N)IIg0GoZ1p>2BLH4d}+*r^LiyN}U5f9uI=09ES1z?rh6 z^0B%i3dARrgC>!5=3_ERJW3&>8OP{>$|4FCv1g?+ne2&d4O`Onp0y;zv-yodd_SVa zjw;5o-+~gRd;0||jOk_ZahKVsyTQyjwwSe#e#1nw)lug@9?zVE*|%M@*(ld=ruMv@ zHEKvgQO=it7AtVjSdX18%VwT`2HA<$0Ty}uK8yJtz}{E-!r!|Whq+f`ljaEe*7OnH z2A2?asukT?swA(eOG}qn(lLeQq@tlrr&bDBCGX?Esa3|;UQaZqhhoZFZS>9`jm!qQE}xNlK(k>~EG42lSpgNF%>;y=>fi%qmI;2&Gx=M62-^+>9W#ZpaQ z6q=bKblF69>gjIIab3l9w(nsc_qH)Z2{4ZZF)VfM1$J!vD2%*og!?Wkm{3;1nnG1z zYp@X+UwrYKB|xI98CRSuVa#XKf=la2<Hb4SY`4z9D>t*kvJq;0};&{ zT#{|W6~7udtML1B!C~}FNr1FtA!hF@#1hU+2pn&SGu6AHcJ~B!ALZ|B<7im!wMVXd zuYOtlbe7P_yL}f|vU|zPS^PmscI{{@^9{Gc#r515Fq!bBaVoZbYh>GR8zA}GT+XH% zj;z(fFiiKt{Qf17x0?jj{|@49Vk(v!l;N>a0mkgN!|4i__;x0>%X<-I@ z54V@+gG-k$L6B`2+In1Ir{M8q-obh{4xfj~z^)_4~O_aN&Q{>y5wOhC)&1sKEUP2wiTJj3*z zRU9xvN{( zGOg|)&%FlyzoQ{@?ymm7&o=CM$`NMhV$Y`9&Sj?S#n{B*IXu@Ogb{rU&|l$%hwdhv zl{O6D%U>`zM^l7f<_xa~8*%*pcC722jUDTiaY-r+t~2W~d%r3r4*1a0Rc7RL>@W0_ zcGB>KSX$gtPyc!Jl935#%U|F-3O=9FJE%qlyZ|3>@Dsx)WJ0Mt8S@$xC^O80I*pvE z>6j}WnED@9*g3%bNg66{)#8wKBGT?gV@s_IoR_V@&Z+TuR+|si#wxU^)k62h5p0)J zgzlUTc%Wntx(P(_yJN@6H>|;u!H&<%58jExXuc0t6Z?zNF^{l4RRcjWq1ev7v0lRFJfx{Mf{b!#OK|e_`cyP9>j%W!22C83uI_Zv;>(-OVMQG1UP=`XMQ@1 zaPW;E66-JH`L)mZnRy;Hb$tIn%aKWM*~*^F$1zK@jjTDtm`xZX1WOHl^sQ2Zd)_1@ zr_932XafxM*P_(aU=nVKAd}P(`W8T9nJa-5#y?MjT=Vvqhq?v5y zkE^V@R2jnIF5DCDfR#6$@Z^gSK5gSZzpfrEI{gce?S<&C{V=NAC`!%0s#*6-F{r(= z#vAb{#9dp8J9$r7Tjf>u+foM`U#|eG^?-lS8Z3FL0;MDCF=AdE45GZTe{vX#hge|r z#M|tM;$d3wzLga4ivA^Drg19AY3dzAn&Hdc=T+U<6#f7^*XHAQ;9=ybJL2)pHB3~$ zhDq7=u<3h$v4KuuL`@1}A1^3Ed$~6{5BlS%k2?ai!Xc}%0l)q{#jyzPNIo`$gu-T$ z^d~t|TTzISb4%F=`{~#q|D8Q{(Sp`&b+~L8;28x+9Lmc;du21se%|Bv`*Q3Hm7sNcd zDe&ohE>KF$64-Rd;y?QHig*HPeQHL{F4lbgPIzAM(Ae2<^ zETB=`KW*+MVCz3UX9m0@NIUA8mSP_}Ww)8n15B8rdJNmrk;8V6Nn=h2kF&<4K&GSe zlF1+b$@F^1|)pW9-(kYb;j3 zlhq{EG2h?+*ov7xOk$T4oIYpZ;@!`%w)p|Oth1QiY0J5^+BEWBIGr*Mr=go&s48hL zd6)fwo7-wMb)15A_9axRv?DC|9ryMYqCMse3U7XbmFNeaC6b`wdQ)k3!UW2D+(RwT z4F&akmkTCLcM^o_kYM}yPD+q5fNuM2G}i^;&HMuRv=$;jS_KC$M)2HlITm(bhv&Zk z(CGgUwWTNFKD!L=#m^Ba5T~#6B`Bj-oDz9=xp-#1{yf*AtjH&rC5i4}VS$qDX3-*j zwJj|yyhIwm+h(EL(FAus4dorLd{)1B8WPzd9P-M6>a=tC_;WeZWgfHPJJRSz!BByx zhNQsncMpA$NuzVeZ0RQNyWFvj;r_-%{8Nd5=F|#&JspKR(!5jiFrTTk++|P72iY^8 ziMf@~%fue3!rfpNYT|&>RRI{En8Dpf#>n010@?2k5EZ?G>ENH7Qh$>7IvAoL#?<0d*WcUx3y=FYR{_@CFcr^3OevCqu4r;|1sY+Frn-msKd1x&XbS39vjrg&FhN)pU(ICLWu|zHtVKp7&a| zC@2P-2Jhnbg%LEL`)^ZU48k=0FsvF>;V-9$n?KaCyHFGN`c}asWD(>%(ouZx5w7pN zfSvxL^eFKQE;sDqU65kd(OS*o+O{$GQ}#@vRFk>4{nRhHrVZB#ws6tm&)4=y1l#Sz zj;hJn&*#3s;em1dj2hVHj`}K|?;fy&v$HzdHm9QT&v7*8@IUu-0^~X8`MBN(c7XHb zZJadWTVV~se!wfIj5QTsL3rO^OuxzRSwn>BkAo`Nn8?#6`*>E*vv+F~#v?R!GL$|U zqF~Y}{CDe~{>Bwj>{Z4VHZ<9v)sB0jpVJYe@8jqR;}bV=Bzq|Q+OS)UXX~B3l zr`>}ka&OSH(e|v_^^tzu>{{-R8-kl1T}8aR1b z3;&)aL8-L?XSvI^^`kH)8miHKH)YDx+=r1*wj#KbJEz~(;z-PQtWr;geXJ$F7lgx3 z=?>2|UB^;>SGHK9grT8};Zhq9dzT&fxx^kTn&;sr=dZlK%9-6)xR*An4Am-$SeNGv zH;HL5pR|_N*_YGZrXq?EE~3gwku+7*kZzkOqO$ZjhE9k^5zn+mz1auj$%CxNuol*) zy_j~dAIsWBDC~+3#g%X$_#BDR&am#074&$6RzGwhteW!_De$1`xa$S)#QP$Gh0st|#@5mjl3XhxIMF34 zu$&_*__W|P*=;^Xx@n_Wg#ALcAT*TOzo=$gUxl(s+eO&ABgf(Rrj2uvMMx`l40PY~g=5@3Px$%3zx;3E8*ZaVRYSx7CVJGQJ3s57(jBNfLuc zJRsgdsG4eveK}jP&S*Lgrr&1=-$`-CT|E=Id!DUZmdzg2da$}XCn>k)C0UBxrZ-Y08u2jC9!Uv5Tr{Gw>_L6riX+YtF!g){XJqw8uaDtJfY2hY`iTqKJv zlPn=y$N2z!MkAy$guaUnr;ACKap$%@DmOjHA`1kAp^%K)x(7*xY?1+ zVHNWAafHISd}vCCgHD8FqHQQPPhO07+t!n(c?0eEuaFisAEnm7FzSD|gn~;G;BZHr zZjI8SzNH?tSaBQA*=f*wy(FYe{fq6drD@+>KF{fEMQCF%_6-vKX^$hfd_Usd7K)z@ z8f_{~3Ev}WkoQ78W|z{7{3P1)=>)O4o2Yx_3Yu?!n64}#a=A2#K3|t4&8YhjxmpGH zX9$4Jo8 zz&zZQ;;f^R*UVOK0j|&6iz#siJohVu&VM&5PLM+1r$Ah(%z-G&N4U=v zY_7Nl|IMRm)>bVV-n5;pM#a(Ul!-KBaGb!TFF>#&^|0XEX=_3HBP+q3HdTSfoEX?C z2+@uIb5i7H(wr*pe7brbJN9qE_M?{|I5&)DHHp#W<&R*&nOZ{j`WSq>58vX+^kJAb1w%sL9^JV{Xq_?zb?piWeSQiSU-D2X9E%gHxho@zQ0Q?J zLD$XbL9H>V^z+=&9!dHXn~6U)S^{T-Xo2|q2!V)|onX-vOTm_B;{={NPT^|#D@;8p zOo!HMknC46+GckYy~B2}w^?;es4AVM8C+&@0uj_@Ik8zWdQiH_`E@bnnD@2{u9LEm z!FkZWUz?EV@e8}hs*|SN7W#Q;6RBG3lW#Hq8s^y2CT$%m*OI4uBPA)tB3%ighK0=R)_G>i^Y`IvN?7Mx-b;;I&0f7Wz&V~5xlodao*gGKbj)Gs2#rI$ zu{6e;Pl0wMXIZ>lfSr-Y(O43VK{s!#-(P|=-5)W>a!8r5P1XFnq{Na3m?m~Nn--`!PzN8K} zYW73H-2$6)Be49)ZtUnkiLOtnm^a4_+ML5)d1fp3wO)hc^(M}QJcS|At5M*18`F5U zcX7{VQVYnVu!GrjT7>W3j6CqF?*e}FKAgOvG;KEegBqD*++m!FBU?t(hGYR{M@=T@ zM>3Q)>=~Tw_F*ZzjT_qM@a}vLTzBUq{hkH7%U`jWbGEowJ`b_~iNo>HXXgFAg}r;9 zgux+0sm)4@)(_L8=5Mmxdt8V@e#fchyQ%L~gwvh3uxHg%tlxecvTBOtF+GT;oa3GM zWj54s&Yb#}amR`0NIG?89NqDKg6_df@cAzp-^wTA%5G`0owc9xCm*CoUMtA*`YckA zHNJL=M*jKySimpdIV=hNJFZ(_nF znC>o2XLg&>CP@dH<7Q4iucherOfmBCUqKyj0!VN5BC;H7L`~cS>DUs76l%lm`}ufw zsRe!q?!jSk6ldc6Mb}wd`pWqyUbl5=b?$Wf@?r>Gh_j=lRhP-R_zgLEcTwl>V%n#< zh5k5(vooDz@o<9_RD7D)&@A2~HaX6YhFM~4XanyD-@y;wh5vM73Vk0d#`D-Kxce*} zr+rJ&_5D1CZ@vaA=P;Bdr{l)`=XkjH1Ns|#vmmKI1GXGG8jzL%8i z)}VaFPUs0HBkDp4CjNek5-Cx-w0RO8HAv!Se7mmK=-Po@GJl z2$m+3Xb=3H)rf75%Cy>Co>uY9u9JKMqKA#b90#6j3eAPg&N@hiS0SyZ3|&uh*gu#1 z%qJ|J3Fk$#<{fTq)p}=nT_A5jF3=wtgvHL$E*r)nnSULhFR z?#}G~+-KImdDcaujMebm^IN_T^5o}um)Zd=>ro~-1!4MGAxlx$i;&!ANLC;B(Ld!l zx^&%_8k$T<>2(=AMk`X|HBT~4<7}q#x%BzVLaKH9h=b+a2_0yJDg3Uzx!n?3P1ZPI z_>>7xT|i|UTjXLPTg`l=vLWis@gUe=_{l0T)ZAKF2kvUzaKa5yN{&MCD64g z#mWP@s5qSrz35z=NeV^Ak(&^jJA}frbDb~UC54;}wOMP@X$ac|Y5mN)qIyZ|(K;#oNn3Y5#PL-`LN*_HJcYx9(=DiJZ@T{w2G! zyM^{utAI~*rDmwSsLulz5 zs0Q64GP3m>zCvp%!O8sadzox{e;OvYdCs`_ewdClA8EUyAZl3SrR`ffHV~aOl&gJ%;YIt8^!IPB)_q6gc+ldxhi zef_?XvL&_YmlSuiHBP6MymL{$cP6RVsnKHfr#L!xH7z<2Oxr&u63&Ox@9i7un)*L% zSg-*}H&b{fyab00@(^R43oVUl2s_%zM$`yHEaV9b&icSUUlm2DQ5SQ)uuuPkZj1h# zo|pR3xnK3)_en87uUh?i9TvF9IpA`ACorMV1^M<~SoUBnmQOjy{Y+X&oW`G}_8su= zIgU|j+aT=e#C{xI!!}YUTensSp7*}8AcsiyLi9B|{-uOHTNuIA?`&t;_3PM`Ds$#K zFv!CFRPfz+I249)Cswi-?i+C~gV!f^%y==hCmhH0C5I7w{RDh|dt=m3TU?j*$MVsB z+&3AAO&>f_)VdtLk4<3V9S9%KIG9SN;;T&(BKs6jk;*@l7Gaun(T41ntff6yoTx%f zm0Z_sLZ4bM{9cRDlOr-DA1p!>a|>X2uK}00pN7TA=eYQm`&x#U;BLPG=A86LfXg*B z`toyhVj3i+4q>*90NuwQus>QrZ^mTqe;kX-JCE33zX)dVaUafWf5fb7N)*DmHWQl^ z=pMgItzMx_9xk&fE>N2cI5R1rh4(T!G<=736Q-S#r}Np0w8f75&O4>)#@Az5qxBNy z-h=2?tH8dNvzT>cGXif}KzaFGl3lcu?maj}9=XdXM$Vkd^4qcdj25OeZb7X2e4gO} zd=1PX|BCyIR;jXcUxL~1r`c>*LM98{v5b8?@I>EQocm~=4nwfSXXawm#BTm~m;D>a zV9Gi}@OkQdeE4jDMx8lmFfoCQz5+G|rQrRrhfwqTj>S_a@#oEvbX6D8oIlSf~$cG*w5bMF28?dEx9mp1UmY#>^IFENK?w?p7Omqs4i7q0ZzK zXG`|nw_ww0jD7rWbc1&{i}@_GbKL^SR1L6kXEV|Iu^X~t^%$ak3#XP}fVfT=d^d7_ zimNcigx$i5sIy?4JM=?12tmyyXz-DtKj-yH_sm9mv9h*0aav#{zc8&R@e^is^FV&>gYwK|E58rS8ddZ#Kck$T$GPdu_!+>%m zet4arl<)_Xx2T@#<8!Hgd^jCjrb;@ov*~<6I4!;!OW)jL=&S1?QrJ9^#x(gb_q92! z>Ov-ak$j$=9h%OR%k`Ly(*lG_Bw^LhR17F|!z!m8ulVk<@Z$iV(@voHRtah>eFn=f z3S@Dk3IE)#!rkf%k~&)Pw7417OMYR3K`mBq6n=oES~*BAK+cC!Sn2wve%ODEENYvQMKQunSkyn8)rI_Et)T z=^eezyX#6Aamf}78%*$IiZJ4z)vzDmj+1U-4WEUaC%@iID$4evs(9|zvyjH*EqgfM z>m)X49Yeot7~WkrN9UaZT!kv>s>{;qNu%i7d`bFf`GB*@#9?<$2Cf?8;oH*7QVOft zY2HuH&q*P-vZr)z;b&Ssp@y7v%Shoy2C?+t*f-yt<}S0KBP*q;hyTuBm-7sJvp=u} z-C9^Fqk&>qo}GypWP2jp*cO>(*tQ`7i;rYMt@9dUPd&kc^l-dAvT z`jn{}7K$<(-VD;4zZBWxH$k_!&Q0en&}cM)@M@Xl>*8X9ZO zu=bHGCHLEsK|&A}_spVOpZNQF(qowBUnJ}9@AT;DR|;0*y_Ik0No}Yf-D;Ggr1dk2 zEtjEN1$Usz8OMd6;vtg>{1=~y2gC0}*8B;+4>rS}Gy1=L3lvNXx*_Pvc_HxD`6W2z zCv5OH`GTPC5qgc`SPE4!?A~u^Ew${xL55}J>j6i5YCdi z2Z_otWbr_UmOs-Xsl{ScZf!gN7AJ6Oro1*CEaFOsj|5G*b+{Db70&!0l&C&#jI%|-hEHR zlz*`p!)NOI)ed1Q->>zLP(sl*Qq-AIRN70@9(eQS*!nAt)Z=fTF!K(UX~TQQn{^ zx^>|`9N;wdQ-&?n=_TXTL$6t|I71jRo?oZ3+!j$H(;Q$`c`j`CDu%e1sn8i~0dH$2 zDHFtXNiXNvQ#X1+rt>cJ9`=NY%`zbEIf*8vH`4XTyy=6zHuT@N*>u>7_bBuq3wetf z_+at?=JX7}#@}PGXfCI7%ZozD1}iX1L13S;hSOPQi|CEzLp$oOg#5&%`=8vP=pQjJAt;u@Q3q54q2K*_YMo#AEpiOmYso~e`%m_@)Arr`@uw% z0{DMYM=gI>qC%PN=zvEMTH+XwZVW6%mQ%J+)5bAVh8ak;7~?dVHc;~pf+|5zP&<&$ zbprf=Mb_Mong0ez%GyBX6sG}oH^6Met1vq$2b^#)j4E(GYL)v`yYUQ|-mo83R_K6< z$WuzMy^LD9>Ls9IestxY0J>f@3pMVZfr6G?hG=egb>i+mxI)Lm*rr?vi7$kcp*vw$ zgfWVf2}OK6(IG+5` zCDLd{H$sD-l99uryQtJ67`+J(LaWhjG_C3jYTGGIcTOtMbH?wZP!nxf(Gv`LH3wi> z+Xe`*^#@1i)v!Qz2iFIZgR#GMzKWjMRbB4>5RYA%wuHX9harpY@G*~#=fF0-izM8!W zZ9cjJedPtBR}S%rZ<{^(pce(>k|of){XP`U{togYTwb-M1ZEGEaXC;46x*YL>etOc zYZC_`^+_V640HWmD4`5maNtpvGS!BE;-3}18k(f(bs$f{rr)(Ss>OMaKY zY)1y_u^&dRdVkR$>n?P7Z9V#|cpNR!as%bq%Rtng!uV7QWEpLTx7-=oXgk-1X23(Q z-)uo+VF74qd@veZz7o}C^n*>i7Pn(WNULEnnz2L^9TIAX1zcZ{V)jnU8L z=crq$y;NJ^FcrPvCG{t)l`>TcpsYg^;d1#CXyF(V$u`&FV?i4%i8~MNf&>jb6{B0a z#b_^EVLH}(3^BJe(A&gp>hFg>DnsBU^=)?@r8+Z?!s@Eji5P%<@d6M##j&pxx%s#5 z0yxg)I&P0~-jOBSQIUHf(&VmF-Hh_T1HK@)>g_kQZ*s@*J z!7L-V%yWWTTYIqI;07E%7sSKfQf3FILDTtEDs9sV>UNtBHB&&7deEAV;-B}RbD_=1 zS0DkgZed7$=6@(x^#^o`I3SznhNvOuH!K!<4ASXW;eLrG^q)xpCFeqTAd(1k&D~+Q zCm+oB4grO|%`nOL9nO7}LJwsn&|vCym<%g`C*QlF$aN;NTd9g>1dM|$w>Pcyjl#T!wlb`*5JECV-m4X#pEFy0goSZEa(*Q%k8eP_|7 z1$&U}9Szj_T^?E5iXdj`HjXV&hgy$ZKt{jAP&}7Y6WpzXYUQ-hRIN7(@K}NF&KKY^ zwokzSRx@nz=!3trD0HB1E8=$wMXGk&(KiWUv|e~Kt=N@KAKy_%3+}l^Pit(YZQZl! zOY_Ph-c1lqiAthGMHSQ>CW5fy8#v%rOeL@Gr7rhQgAn8CkefI_HF5c%7vd41GSUu$ zoQAM#d>lGznxKA75WMNNrw+xvpls6lLFZB@b!XKqn7glqvRx~MaL-=UG#ZKM*le^d zBZkY{>7f3){gC&%0oJ|jghqTHI{B}HL`Oc?1F;T`%cUcEmkTK2RRTJ4b~7^1T8hRL z1i?Cl+YjuphD3?=@G3?RR({In__dcPHPOXTEIc1>)c>LK9l71KV-@wLCk5>8y@Zjo z;%N3+8DwX658}K1z)o~47>(P*?7ItLyzd~W^sx|fHxk4p_j4ZE2Dr=lxnoAR!bI^T zWwiAPRif$#gBjels=f`2@@n9+Pz>bR8$iTI1N4c+qMx=g$YkOG_eiluhmOrc{P|1Z zrTIaKGYW?(u8*a8EDvhGbN$=D-oo5f(r8)HO!T7203F|BhnB6M!M#U$T7?Rw4=;

w%L);$vRP;Ugs`1%0F1ANGFQUpD;x(81mTEUD>uJDW7kImoN3HQ#*A|+Z8 znRr&gx(P?_cO&5QagO)>))h*dO<`8WJ1Vbt30%-mg!SfGaPHbFSivU*9h|OK^Jga% zTBbn0YB{JZPKTHk0njTZ17B`8!>RrJXc5ap9l2uYjoiEc%b3}FO4B1o0knv2FrByG zlK!YBN9WXkLbo^{;h6>ZVX&eT-qFLLJopS$|HeUF=uOH=zLnCm@uk!>f+_uNN2o^y zK~%q`8gy{y^X=&|F!iz${=|lWvxqi`O7n4DWOhjNoEt(RA!w~$4(i?gA4>S$3(dNg z$Ysb33HD7xUMfA%B{&2=@53SU^ds;)BaMo=d`u;^7H--=)}O2zalUwuMW#XEB4r$vOC?T;wMy4C?7K8axIHy6JG!;vS%* zweY^d3a$wR!fDQT!dKo1Umi;!+k~&6b&>Pfj7^8;83^>-7JFPm6lMHzN%%IFU|7n7k8LCdQM}J3Sk*vUZRod4!y zH@vz11IFHdf)ssB#E#rSU!veul_D#B6!7?@- z*#7U-{RQ?g5)}r4=H4*t+!Uq0H=a72YsT@XBf;G1IG1&dh5R+~pjZKr*6jnEAJo8l z^-jnecnsC=W5BU-Ik*byq8^tMX!q$LWcHUqUs@j`y_y`P9L%ElOc|QJqed%-tfkd+ zZD^`Oh`#q+6ZSZ`fi9IcE9iuJqc0*A<91Z`;32Z)GE^%jFsjht zn1W;H!Sr7UNLaeS6Ur96mBj#;>q2CuFI<;B436JdLiY@5P`y0|{1%;qzdnzkO6V3G zbKD1gueO5j8!m@ns)ihvg&>17`%&sx5E_z7LP|ydA%m)DnC{mPUz=sntOrJ@@25I4 z+V>hB$o+#YlDm;qPz-W#_e9Z7%E;RE4dA24V6El@$S`aGp-&Itir;P68E_in?!`br zE|+iTcqP*-3PB>0%PySV1cOI{(YNdOkgHQa+I#jXl8;M3^5J%1Ai>jDV6K)p$`n*X`mWmO5D`T@ zZtkhjF9k#A2?$ht2g&<7;m}MWWU=ij)acDcCl&nAm4)Zf>qbxHwL=A2y}AZO?k0R4 z{tU0gpMhWH2v|faBl&L53l+J6V%slLvcIA!3&SAFGbMsjU!X^|MEk;cAlJ{PUI{{u z_3*N`8E%~;F!i_{?r&Ksd7Iax|G-G@rvQ$ZO;W>W?7n3}{r zMv-|>xn0yOG_lnQ9oN~0j(KRJkiP_a7|wHd{5(?l5{7o>>m$2keCV0Rd+5$7fzc1& zVYmDOw1qN5KZpd9DXItgVhd0+eF<$I_qlu3cgV8;18+ALfmrZys^@z!rPOaoEt+jh zExltx6#4Ic8X#e z3{J13jvSB#tE8DwSULw*G`ylDHS#F7x0_PuG9Pxl1Ar?zzOzCKsN^1mk<~WvZ5ziU zJedp%Tn;%+Ed+jD7lErh3<_7d3@E+~W86-ymo0>0?!5Lla4u>J^+4|0>(G1QbtwLS z`(f&bz{BtcLT(|KLpv$4RYg29SsckzWDOsGSZ+!oGlr@=uVD`vOtPk3e#AFIevyMX%e| z(f&m{Xdj6%`iw&q?YiHDjx14zb8mb=eB1>dH2Z^qS0#7fIRTzZv6SnqYm~XyeX8`& z9qQJ|X@_9j;cc+yS~zs9It?)qmSDo!@2T_e;N}t~ zbj(Z?IX;#_ujD)7p^qz6ypEtYEv}{hyi1_!Hdat6Qsq?hU@R4z;76@o-Aa{}eW8xJ z^1;R|SxAa5r6gSKp~brf64MDt_m=?)ECPdeTUeLBmnv4gN#*L?p`MBkQmfcS5O_}- zC~|{3=DZWGJ8uTDb3Aw$BnbYl9n|{?Ex2_i96UJoUdXyg5DHiWRZYLC=e})V=b(Y? z%e0YHz7DG3_91tRUcqsLZIo(BHl@7d1GTxD%h+#}fbYlJs0;SXA^u+~Y>ar#^}!87 z#jGc=bWtYwN!>ul>O|?+Mq>1ZdJ+1J!5hRVhM{&&=REjf2(Et@My>YKk?deIwD(0r z7I&tl=Z3X>e-2ZxZyCbN4K`qGeiWuAOd;drW6G8<4}Qyup_UUfk&^0k z6xaF(9%-d;UcxU(>I6-9+8EPy-ukqtiY$Gc>t}a+Rz^Lo+yjQG3iEaUO&^&VaM~tEs*GT-Ns-g=#*TqPyw~ z5q(b!_1*sg>2e2Qq4O2^#Vdx9q-wYqlL=D4DbTu~Og*R&2gfE!=-90b?@hEI^w>O@ zY5mB^&`y-nxh6@O-_@pE{B)_?u3bhilIkhjDt@@dW%eh-`5`K8h-wSVpi<^vr0&E` zP&vtODE|W!R9G<&eyS~jjDW3BgNmSODuc_hhk+W`=dzaL_V;z3q&y>XsqF&Gs3L0( zs%_(WBefty!=a2PTqm6>q`7Irw)@-}7B2=9rC-#u8LA-c77076?|`D#dpPmt5rmjp z0~>OaIugwfb6oyWNl#P&Pnm(;RtolW9e`fL9HUUKm*ToBsBvWpn0etcHIj4&HkkYc z!}Z1}|K>hqQSOe+C#6x{J$?vs^?*4V;ZW6|1s1+1A#kAq{IUrJ_knKM6DEKx8$Q8* z4TC^C7Q^YxK5C)9F>LDr_|v-@*z2od*OUZ^=A}?wvaM9Yr_a>!qxY$o{yEgW+HKVA zLyO=xHS%OY|oQ)1$7=ZOx>cKR721?acMSfXQNR8_t2y+aC z5%jaR@^u6NNc z`2iG~s-V65)sesf3FL8e6jT(CfTGxX&ZBt_f;MNF<8yoYAUmY#;}oqvJRv0AV>9s*iElhlghwa{w5 zA8N8{z++VrtZe4Gd>;ydzluJ{FZfMSkBcbj+sTyu0y(PwkS!WX$V8dv($J67;i$~j zgzJ&waTLgJSxC>-WP z-f4B9q<9n#d~KsbN{1*iJRL?DGk7u_1?o>ZKk1%r@YE^|RPt`Z_}xM{rO)+wh4)c1 z(~iK2EnLsSjt?A5_yvr+6S(8J9yUMegh*{w)EcAD`GTZSdCM#`{I?v=+W(~5?dCw7 zI}Mt79Cu^seo!>mh6MFQ&Rg*c9_Vs2iDbn%NLeMku+kMNplh1*Kcez+=c9lK$|)iw_E5 z+;0PWsksoKyA%dR6v5%;UFs&y2k%2p!Q77XuyBO~{I1l3w)gj_6Z{-o>A^mLb$elC zhYh@uUks(<_b9RECJ3?+NAe5iqW≦QIY5kfvh_f{W(THmlrdlI2Njn69C}7S5)5 zbIZ`~(1pPFtrU_KD}k2i0?F(yINKcz=0g8bwi88^#H0mfw=$GkRDPa{$u^+|OKG@d zL*V|%JC0K$ii~6V(2`@Pp=2e;ITaX!&5bkAA8H{wXswNsTt9=Hu{s)c*g6OZo zARJ#-2scvXLD~8*ym`I}wF`Tqw6Yz@0nbNXM)_QSL@mUak8xd-f@t-kSMd7FHHhFi zGHT=EAjx^fe)nm?JrIG20cr3y?xY6PKN@YG95rg*B0+tm%{fjdAJ@_P(CFw9el91x z8Jv|d^yX}a+>qJuuJa+~v?_thHJJygcT~8J=+!V>90K3(n7|o}LTc*sQV5cDg~yzC z{{FH#a9Hs=^<8E^HM8j|HF!Rrny2YWx!jSX7M_+c65gc=&W6UI=(-AcH}-+Wk>A zDaMfFYX+w`FM)s;MlihZH>bnB1E>AK@#PL6RhJl~8JmytYj>iy&B`D<%JsC|&xVIt z1+Zy%7UUl=0w(7XNQKWtbg&_c71lx>oX#=Yo($RBb>Z*FgRsEW1=2fr!j9ezfM|6n zRhkLILp#8zgzJ`x!0`RQ$JE+VL+bU|892&yHx>jp!^riA;9B~O>o4i!G>cSRBt^9UL5cHoXs<;#wM29 zu2n)kiTOk|e;=V93q7Z>-6hH|P62-Z-UY9)14Jd*!2@pJcC^_ZzRlYKI(GTo`MVK9 z7goSeKUW~$|ERIIja<*&B=j#_j`T{-qiEh)v=y16_-%=(QsE(b@=k;07#PIC!HD?d7+1ILxnjnEeA zzN9wviKu}?vp(omYeV(+Mk> zt+PXN*JaSz2S1@Yxe5|iyP@abuOfx4C+PFM$H*zU6s4)yBQMUQtkZypc%yEcjC*ig8Y?P_0k4C)I&?)6SsO$Oz3cklVDcWJrLj41YSLvuqgN|lZ`GCs0*i0eY0hHWEf}UumPx z_p8wZu76{j6axqS1>x1t!u0v z@lBu3_y4>U`7qz^LoWL~j=Jja-R--ZJ0t-zG^S`{+Sr)ptyxL&IsgB^su@1sm0yy) zq)R5e$0x*ipVxiE%>m!>p13)@QkR*$HibOw{NEt$er^S}h{(l$kHmRTl*D*JI|uMw z!zrxfH;b1&Wx%`AqRadFZW`}LjyNwpUzKMRK7*%{@&Uj8X27%hZOc27vxJxPbUE+C zA9ebHPY-=3;VixC>PtE$zyW`swVQ67D8ho8QMm8dOy1*4LtfK6W8TC6 zLj3EPFc$u`9e=Gd<>h2r@N(Db@P@-Bd0Qez@SX{2UfFvoo>ZYAuY3&ier%WIos^!- zySvJer~SDTi_K`l$EKBF)5Ul28uMegU`i3g*X=mAO&_QA4dCKm8oXs)>b#Jjm$BiB zxx9O~mh#eOEashTUcg)5yd7V%xPT9b-oa&wM=%rp5&t#s!Ht{y@ll^2c;MkKy!iSM zPKdI_6~)n5zdjAm=+4C(g!{40A5Gpe`RP2rTe7_WYNzuQ_6zf*mgL|%H-be)MREA8 zO1i1Z0`HsefvtQb@$E$?@DH&r`uF5aWbGS^@7Zm@O$P(%%7@+b?nRTxFklz`^v4as?2+KT8*doLXB7OQizwde=$!oc_}ZH=l(zHd3`A_&to?4`=<%~F-4NczekqWm?pqe{L_G2 zq<>eTICguxh?nlOnkS~Pj3*kU$*WuU1MAIsgj?rk z;?Easa8>;WY*yTpAKHag6@_`kcLs0vh!St5 z?g^~;-w~|G=ZK5?1L@9z+h}#2GyYtD0k5=<#n#I|;Fp^Xc>8uP;Mp1o@n)YC!aFvs z!)-~mG!s%wTb~Ta&WB_0zR?#oZiTCHoR9)RTWe|xI>P7|OD*x=RSwL}V21wRt_W+f z#e^ry!$19;a33tSmaJMkSCuPp8$@^?{Wu$|79KM2i_htjucil)by6owY z$_{i=a2gXC+ec^H{iY?izNa0vhv>QG8))rEf+R7`z9z$4ji?eDNfq$Z5)Z>@7uFc7 ze4oV_zbn8R|H@EJvJ#Pf0Q8*ye$ndg6f=3Yot`)-Oy}HYX6ig1qJ|Hc&RzVPmOqqDSFTxrkN?P}%ik#BY7+zE{Y15<>V^ugpESiz z`&CiIdlk}de~R8V?HIG*P%HiQ^-#_6&K-E}Qv>W*W{Ee)OXD~1BDn4WExdnc4NB;& zqeX4|p<7BnbT`1V z*LKqS=MK_OBR10MZ;h~Xn01*$^hW6%EbRRX zk8BIT8J=?3Ha8A`|Gk4=JuMm~71z;^t@ZKzx|^sd!WDPBPtlV}+wp`-1b(pU4vy0s zp!W#Z+3Mf&?}{xFAYf4N{F_J&!g>boT6_VU@Un<6bpBSVzB&G zJFnspBY(i1K4!ERA6lJ7_m!vMbL(@kWL^NBAk{#tsO+y@XR{pLv=SiYXf8P!w}4Cp zI+C&YRA%~BDdL$fNAHp`uC4QZSQB%^s&;I5BE7XCi%wozNq-Abqc`fc*Q}9~!A0*5 z;+{N1oVHj7yU5F6Gy5PW>3T@bI^!ASu-G$tSCkmm33S5Mqe6J)R0__z0eEw82d!j1 z7q5NnjZgf$!e}oxCXuYG2TD=yQwZts7t^SWEk5!Kf}>CH{m1g zo0)BEh42+MOj5U{Fz<4G=zXGNG~&53aR+BI`%err$!oRA#MWJ8$BFMXu7A?W!QLCp z=+;6~nYRH33R=kF2k)6#$5TiYWHDd0S@QQ~7vYm1Bkt1o$o%p9L@bpM^zlA%pL3ba z$;~EB0)xb4!*d2~%1Q10QWAIbJy}rvl|+wMlfuPf#PP>6vYPB6SNFUlmKK=wd2GT7 z#x1m3unBE{=owk^Y8Jb7b_YRQ&Ji8)VzPWyBuSgUn>;ZMBtCoiNtIs;kfjCBF|60 zB|GoGAWxp=6W6)A%=-=DwORJ^L^LW7eeQQ8spmq7IJr*_e+njH?xDooDTl;uKF0_K znvkZ|Mg&eiB;TW)NiXNWP73fQHEH5R=rl#>jWvvLmL8GWt4G`(pCy-9myyt|!^~Ts zGG^rdQU(OLBqbkO&VLy4JD#3no`Hw7JaDYfv zSCCh(zX&TVO%5l&WU_Ou$-0DjqnFb-j1j4XXQU=JfvjkMPg0`35KHrK#42y1mTyi3 z!{=v5;u1EJuiWp~FN!A9x>Cu<%IhTHjyXeZP9{-mE6DS`QG^<~ORflXk^2`LNUL20 z30!uF6m#D(^5{3&_q2*!|CUUbzBQxMqI}7;*lIE^fay(TrL=5dIB6I7z@$TGZQ%k{ z^6%d&hJVzOuv95?P}mfI{A5N;Uk{_jsR#6fC#!Lh(N}utS1`Toc`(|qo?oLq&xsU_ zz9Pa(tMK8@dAPjyF+F3uEBDOpsuiCz1ADip*LLv_Ga-Xd7>UH$%;oxpBsTLOQEe3` zJ}dWO#T5Kf~rTIj!CYLzQEhbBD>tT7>AO@9idr0m%88PQ3p`yfC@l&eo zrok~HYeA7q+LN`3N9;&ajSmS(Ok$jF%_iF>6Nt-)65^28%{VTk$?d!yqzA1d6&kU$ zzSubuk#UCf^jVR>>QvHuG@FF?caz;I4TNSNGs|x7C6Q|q$)N97hMB#Obhz-dVhW!~ zeajYdPjpD&bPFS~ zdK|4O(IT6@7SzTcwqdpVJ{ghTSH5F{PyEOb$8q(W^GVcq{ob^9qxAEs5zp`-3j=`-xP28fmSc zyBGyOFJ^g+1D&l>PCjnBL`qkLkPF7%Bz@*d^5bU#iRsKDE33T8jBCl{*774HEOibE znO4a-nIsW0_b24_ooysCDV)66rba@O0GIrgBZuZ#;>n;fW=ESb`D}HOL_B|A`)1Bd zlCtb0c_GEmPFU%YRRL1$y|E#3z`>ea0eKQX5K2!Ngb|;+LS)L$fxN}j8MWFqWW(=S@UxyDOiy~_!Z>XXkHu14*EmP1oMc`qY*Nza|Sz0DUzH^ zIYW}!Sn}#k3R(R@le~yeC3Q9-M7V2tZNlMqw6o0K+FwbswR#aeqMdS$wp45)uDbV# zP^=WGv<@SNpPn!}D|eE=`>qr5oBpIE;Su>Il0roL&yuF~(d1~vRq|+aG&!W(K@6@H zGPPUPS+82CV zZzqvB`JHUKF`o(YHoy*dBJg0(dc1W&gGh(A5#c@gWNoGad1lr^e6M+s_l@H0Xx|ND zxFL$PC^Ry9FGOhh-!JHUIt=r2>u_lNPC$|AR-ZODz^ETi7q#7OlA z68rM&WN+F^@+6Y`UN`EIhrXO{iIWGW<6S~kwaL>?lEsOcM8w^OSoCcnrWP?|&D$_0LOq-`XKf>iW+8;P zBAMhr;jSmE!<4(;A*s$~M71KGiAj-Ryp2U6uyyOm{psP%x9w$Q;(Q@#)?Puje6=LsLf(^0HDjb}NjKSXtd@va?;;&L z-Z2NwHWHKCeC9t}3no}4kz~d7F+vY^l1Fy`n2V`HBujZIDZcAPUYgV}GnXopmYzb` zZE=^h>OUshmkY>c{omx?>DNT=%psy%Q$t+zM2V+-EVFY~Fw^I@hRA02GB*-05i2yv z=!Ld2@2?*qo?T0bhKmQi>rf7g;wfPDU^!-0SQq&+YlO&|{UraLJ4Y^g%ahN?y@!#h>M*N+n+l~-g5JoI~*IP`-O+{L?brBg!5GOs?j*@%Uc4Xp~XRWo$O73*C6`kC*fOZnUMB@Ae zQCx%=`%7jTtFl*s&69aX(mqF%&CS|mKxHpF3gVH6k|8X zin33GrCI07dn8+Sf*JBZ$JDmX#mx;OK90Ev@oWcb=EX(u}+My^z#c9Za8Vkuuu?DES<&V)aimLSTC)alZ{ zaT-%>eTO`85hFXPJ`$4oguL4Hh-|HYL?n(CGdWqFL=+z-q8W0e@5xcx`(Qdshjqua}gbnMF?BEg-k=S(5hy7Njm(o%yW4f{Z`rXa1d5WA?lg zCZ*y$BCz*4>G=^#Z1m0&&Hk;VF6KgwubnK;(u*W*I~d{zFX+>OskDEaJ3j8JPJ0Ek z5lzqY+IHk-*@KIuk^_RuwtGHRIVg-%3b z&JwcVegP4FZ-bY$i!g(~|J90v4gRu;A(g#@*x&9k{)Nr)1HT$(D1Q@i%{;>#Qsbl5 zF3m^Y|MW$J&GsU~*MExf$0uhYg4Zl%z_UpAl{PZO||Vm|Sh zZI4y?Uol%78imAIdfRHfR>Sxq4&GX(L;kOjFUqP6R&23 zqtvsoZ$K*U?CHQ%x)R~P&CiPNuOp{YUy<~^)%3o@BBbf^R_v#Ta5j;~DwkDpu}U$% z<$ao&6_rTBiGOd^xE3VQePW?bWX z2U$pn(jP6OndbXVwd)m&@zS;xI8ULF)?4(A_V<(`FB)f)L_D9|FsiD(o>WQn)m6z` zV>1#|cZ~^@zCv=xZeLiRVM$iG{$w7lkQydbfOu97}NE6#Jq@hbeh z$rlwQJ!TULWrp^vfr@%ukrN#s@;2Bm-c$T8&hqgwOjr*(0}DlGSc7bX=kBOlxTJqPx~o~i^3P; zZb@OxIPb>M#qKyrIu&0DOT!5tJ+N-hVZ44<2!2~)h7B&nV5?8rSPTVX=P*fpz%vQw zEziK~<&y9h&oFGI^^$H1&&2M<;y7csF#fo?6k|@?&x+@EdolC?^gcjo0 z;j);jcg0~P8aRLBLLBo~4|f&y(7Cf%`kGJ-_6$qMkG#&Ka%h>(?~24 zaTdp!?Zh4t6n=iP6dUWM;*NiLxHc*npX)z`OLj-&N~;#E5qdi2jW<*RNC{u z20Cc5IgVN#fd75rAg6<#*tEVDPjUN`gCA?~pM;w@z$qP%P8H+bn+tIeU4xetJ-|L6 ztFg$OTlka6Eu4G62!9hYz^)40uzSXG95|JYqdV$w?*44-FOZH!zS-mTuT1cd#kx4l zLlK`j#g8?AAHmekNUY+v3Xf-;!{he~u=I|L*gmWPcR7aOnVxx=w>}Rm1ytirwhs8w z)>$|pJQUmgpmCv3E^g_(j4w_w*w8B(CoUMLN4ayTh4PHI47eA`i#(Ntj>FZlmu=yV&tg9f66O-p+zdkd3r|lkc`xAycI}hUt6o&V$ z^vBw+ee_sNHm&;RFa7kCFxFkW3xD$Iqz^7Wf)}jtz@A#Ou(MnVU7Zt#!@Y9wx=n@n z$G<9kY5pyI=V~(^4LpQfpGM&C$Zy+ARJ{7heZ=)v50Xf4h*-&WxhSM$Q)6; zH@pDvth|nIpLf7_bQ5u`RVu!-J`NkX9mlF&WAwtnx%k=lTDnn?p+AhS$DxZeG4t~b z_UZG)OPu$UJ0b=oe&bw*6YNOOtV%|IH8&4=-6RtBvg{V+39@|f59xg+PR9OzA({0d z#H&P_k!8-1j(vKhNs$nj?x*DCC6+AvR7ZyAOk-!rsIxt)3(44r7CIwp2m3=shea7i z?Do$sMD0Zex#m8XB<*{{h#9SC-dc#UdJJP4Ubn74&K2JdQv;facD-L>Dhv%Zxet6TaOA zWa4)@vimH@NXHAaT{9I~;UEom`0DV8P~mQ)1gx=d*Ud*0ZOXm24j+&%&NTk~MC`p1W(#MS-TV!wF4feWMJU;AY4& z(aLOeWI74S^(9-5-L8qd*j{`7=Ld!tyPSE_zmnAU1ry)%xx{;I6iHep&)RxPvhR|( z_q4sf*6~vbA^apL+t3UdF(RQ zmOY_t&h}5wBohBclQT0!*r~0vSy82aQnuZc#4N8S$C8EE7q*IQ%8aih$m$bWwx)m- z#{pZmX#souy)FA&awofN$vN`;OA)z$zJoa%aGG#9bN10bik->Gv3*6mSYZ>2eORx; z_8m}TpVST!KLQ5Qb;lr<-=_azfYi?hT(=sMw#O(#r5 zJ-Hp4M%MigXV?9Y_4oeCCNry{h*C=K^Wr|&D-B6jiBMD`iO?R}lV~WTK|@r?R@Qy4 zJMYp`B$Tp}N=vjuMN!}T^DlgVxPLy6`<(N-&NFD9Vf?p_KI!gSq(OCTV=`L}z z>ooBr16#4@W-IX;T{E%!elM}@Mlnv2UXAN*7NXSm1I(xNngmmQi97ONVRw=ix>p_( zg`4|Ar>YECZ%q}=(=!2;Mq?n4_Q8#s`CzYBPH*HZgG~5%_;1Tmh%7UL^=1mNQ{fzh zJ#J*HX3VFisKWZrw~2bQ9x%UW(rN#YNBE3A$T2m7shkj!QrX2aLm7Gn*1Xa_8d%x7v|aU z%P9nEBiGa8D{j-C3K1RtND0zT2Y^}faIl_lBphBm4s_Of3XN5CF!$V8_>`9eJ52N- z;!`Yj_pcZ2S=T6PZ5j@yCTG}gO)IuVwI1$dB*1F*l`yt`EGQX=(vF!Xl5_MF>?ggH zJF*{GW&u=PS_OUk4Z24WPdZXam$Qj63+1yeAdf`tyPW1CgN~O1L;9va@>zqsj5LmLpv?+o=)4lPf6Z% zeU{Z;r{Qo~KA3W+V98v{-b{{#sdLA{oMl>Yc>h(ZpISim*C)~gn^UYZ+x;znmLgCHAn_UK&8J+467K7|?q2fxdIF05zC;*rk z{ZfeE&(Y#Dr-hzME->%m*J7JhiOe@5i`f;gg%iRs7#Kbd@)t(HgUoR7d}xjvNg$gM zr$~Jwqp9ip0MVuLloGGg8MNDCC7hBNnPw;X!6kz%+GOmFkL|o+ho>7Ivy{>2DREwseEtx+gRM#C1Btss9}U2ODnkT~&fFU~G*N5>)c81lW5UC>Ekzn2chZR0y| z*rB)_A&5%}DH}DY50X^x)!HDe(@! zdUVVgBG&&tL_G1`NO8o$muSgVVad;aH2(Ak+pi^{?rbHjdOr=Vw*}$+NJ*dTTjkX9 z^$9vRCtrBuy0v)5AU$#A7#Fcc(=oh$rXJ<)4H7p_>O?PJS+S#o7wQ)3;X0j4^p)+# z?*|8q+u8);+-eX{EgU5d-V~2w{v*bZzm6^g8R&ZRvyk4aV*%lE_&G~Sy!C!N9`{H= z^M^I8hHOUN8Hw!MP6PaFy&daAmkB+-6X%F^k#k1dE!{KeOHzfM2;2t?I+TEOp!z6rm%gL3vFzyz*AGQ|%2rv}? zHdYX?`eY=YtNZ{TosbpJysaw!Q2!O}Z%T`UV1!ublCn6jLs6U>TZN(eLHOg(2=VT@ zn&POl237`ZCH#2OP>i2;1Y0UgVAO*s)OQwPFVVo#t%LZBB@5{E^R;x>F^TE*)=(dFSMOV$(x;O?48%=>j7XTGV%mSz2z{z679n;wEu+vnlp zauM#y{D_v96~v#%DT{@DiegKTBGfXU#=5RP!C<{i%ym-*%gf`~qxV%3U4{-0IyFZ8 z;@c~{viuW@+cd>pk~etvpMxy9Xg^gtbDHjX-Xt7T)L|v?^FV{|N~kpJC{EK<7l)pz z!5+2~y|pAf*Y3gieCi?Wnl}+e??iY_*uv5W$KeAKgzujlh#H-wFe1~2_5HX+`2!T~ zbC1y!39r@q`d3NVL?s$-SIzwX&B3ZY+Yqx(upSo|H)#L;7rT56ze+z9ln zeamhYc+r8Gp;kjS$I(JlZMfqwi5ZR8L-&Ih=*X(I?6Bh^(b9(kTr&0`dlD?oTuwCz z5C43@{v6oN6dDWYjqxXGo>U zl6&66HcRbc{o{v$|E>s{TDp%pgi$s@x|P1Yu1A+NjD-4GvuUs52X+JHQMT6^Hy=}^ z%Lam(;W;Jt{^ljN{CR`q8|h^8D%UXup);D@e$B3)4aJ`Ri){Wrd)(SHLAdZ?JiEU1 zsnt%KC04Ovp7_M2R5Z3fmQI)1&Z3{p$I1IfAQyj;{gSCmL z+AMZbOBTKM_S5uYo_6~sFr`6`cxc&K5jTD*OO7d}%DPfuP@+wL_&Gu5UOo7h#H^O4 z$YJA^9yU6LFpY1?IHa?_SVI_%U8_&9d0%d^iJp(?`=cgw=gCp3(x=57CK84+4>CHBnRw>Um8QS24 zhlzABn+``NEJUM>0XAiAC%Sna#@fSrZ1@Xb%pGxo?N7VFI!@QHQ5I1wq`XRaFu}3p z$$|ul7B7w}PfKO$kDS2PNCUP98$z(UCzSPdu%d)~) z7tO`ob(1kPiqL-HeD&h%m1QU{+>Vav+4#J&3P zXo(X|KW`MZ+wUXNvzUuD-jlQ8=6DRgg=$A#U?yFoKL>xl?_(Lc zE%-~KrwPoK=xJUL5jTYJIDMuc`kW5HM^hiO?N{A!5d)UGeghqO$C;U{OL|1rC8I)u zJFc8E0#_7Bi6%K5!-d8VaazwuT$*=*&8@tQ%MSsn;S&tAjYho+Wz2jvf*r6=WC|~4 zpxd4tcKYjV3=K-d?EwL}$*&mwx=XONIt0JYaYi@erF8V_xzu%J37uH=b zuyBcvm|Ol6O)8x6P2oJelbwWLw&bA7{R|8ZIEMw#ys+E-C`!M#V^fzZK>FTv;d?nb zcCWtzH(q#!>(*DG=KUjhR0J!_&FIY8Q6g9C9-(qd zIaTv7K;51TY|g7cqGBy`Jg_9287z9r?n%$F+W4oH<(E3M>Q8=D-`5`Hk1$L;GgVZz zx{4(fq~o7CN0?J+s%VJXN&NjX4Qpk!aIcL#hG|6NhKc9!fx>PapQQ?=rMpnQ>>>6W zUB$Mu)mYjtMw`6b_$#LwE$7+cxcCB$XfHswb+PE;dPfx66@~dbjqta$HAbJ-0NJ=% zD0j<^8aL_UFy&nIeX$DbhwsD7^Nexi+HENQlYnhYN^s@WXV?*%jWL6BuyTb&WVfjT z`O*tCWTO^lRDWg@dge0gosZd`<>#>DTN|F-FTyQ*f@xLwX{6I*(KF`=?in4!?!E71 z5!2<-&3y;zu~K}Nc@x!-+{Y(#O7O(N`}pSLKXj#6@uP1J&bIx-R37OHqgI#;!@?@) ze&w5@yVY;mz{50rb}a&vS~Ho!S{|Q0{((#3EZ&#sQ$B|})><|)TBdTtsTe%yk& zF>l!Eb!&v_&i85ZGgo@|NgsPQgvYj)GN!lQ8m;U8S?zCQRMz4fJ?;FDZuyxb$#NIK zL02g_;ucO<%Y9^aZ_P!jK_3XHrjL+aKepsmv`+|;&&I`~f!OO))?C84f3xwSf(Nw)EnZ3!|#B^1zv%#t5BIglLM9o7o zSe}GOJ?~~s=gQi`;8{tckaucUDc@4zuI*#cut)`$RY{;8=L%|_A#h~IR#5n6O`F%` z(uB*0!D_N(J{^THca@}0ZwX`pGZR?E;lL%0SbWS6mU+ko<7P^5 zxDrR8t#;sGYztB!bFGq`BWd>TbkV;)HJmc_92;qImPOr_VQUJNu%WsH+*jWK>NgZj zeYV2m-X^*(DHB?Pli~aAIJl*03G&LDp?H`re7uqkd*w?+7E#;Tmnr|TsKjJeZXkd- zk74k9rz^NVcZU}%6Ig=UIW|q*fz|KM5{=)X133%a;rv%UP#xg_W80-67=H^F_^%Re zyu64u?LNd-e;!Pu21d|lJ0ocQ7;^}ke1$%Dy-5c>mA0C`bUYlDih-!}SHVoL9)c8d z=x8}Hjgn+wFFQ3~^v?1>`u?>V%6}fjdIKAU7C$~v*~l1B<MNv-DRwNk#YeI`Fm3;5T!c)tIMF5kcW zF)Y+G;&%p5;XlI&{%wRAKkJP&f2P}k54AMp^A(i&VCOOLb<%h`RBJIL-E)JP_V?lX z`U;5YYlCZz{V;F)1KOxQ79JSCrd~a{aAb8Dukd0Af9~ow{=V;Ge((G3aOHV8bd{IE z)%ZdfCg=va0a^ZF`b9W{@i23<8gG=R&bxR0f}#okz(FgBU*Ir<*A2AiH+~4=)x+F* z6XPhp`RM_E)Q*k3=YyLto;?HIvyCv+APc;;B;Ua!W&ZKV4w&<741e;3E$`>*$(Nj2 z$X{MJhIf_U${&^W=R1}L@`vaeUOGpUH_?^mzo@^0_Z!Ul%jPEhTB6RccqGNkUbN(A zA93a1?U>Eac`}Rtc4snAY?9&80Rikd%U~8X0Dn&fqL-eBFJ3A9RR=DsDlDP^! zq|E=lK9(2uIPx8B3jC)wL;k(0fcJWy1yW77p+EZ}Bp$yDNuOIGI>?ZJ^SK8gZZzMr z@(mo)8^rfZyf9l78e!kZFA#i0kvE%O1;pAw^t9!CuD3D& zvOu2iksp9zS=0Ex6Ww`p<@vnLv*En*(b@cRcWeGa0Px%bb3W_UEl8;DhPV-VP%)zw zr1t6ak#B?f*YyrOXJpB5-93%}UNweyP+~Ce_#@~Iu;yJ#rtlh`&isK(llZNrb})ZO zAC0qhg*w9;I#2A*pHiK}n`uyftOf8I)3@<;IkS2H77c!L%zVD-yaAjW8v@V6x5JscZU*UH?tC0L@N=g* z@NvQ6e52}0exztVzdqE5-}z@T|MJ)le!Ze2f8ETM&)FHsw~kuIXB0c|A9!81`U=N8E;wZfOvm(oBciU8i#1&G2;VVOZ@R2=`hfob#3AFhny0q~?3@FXDxKqP8kO zMN@`vvX@i^s5(v?6LoZOMdQz`JTy;eKCYRPd0}^xx1iB{E%wv>;`>W4qp$v1-DHK&}H`@ zXk=Z27i#sOH8L7X{m;VKT?rsncLDsr59eo1PY2tl=`cvcw)pP#RyJ8!1}jA;LGjTX z_&xqOjJ&b~e3sQf#PU0^(>MT16s@51a67aHzJbVdRiN`qqJ_w-0u9FtVA^&c#A~Hs z=hZ~ed4Ca3xXuBMdk5gL!whJjbsvsZD)1%qw1tzhhD+Jx@v#?4? z!O`h7R69R_VS^LlLSi(0fA$)N{~OOQ+w~N7m&@=Gxup=azzVclY~f#46BK`c52Gi@ z@RkXFz*)7@jwueX_)j>K9T^No=dvMo{5U?jBOhk{@PIZ^3G_<*gm;WC!va-_7UFON ztehalf8KnOM*S#+*wGJQ`O{jk-gp!E=(C^=FR0q-7CLVE8ff@%6Syb#tl{V!~r1l3OE&0#Qvj(w;K8bAD+HP9m?(PQMLKwh z3jcYyF>kx?4-9*DA5N*r@jFs)z?cvBVCCxT;Pm`4Xj&w|pe513=NyHik{Vc~*$9jJ za$$zSYejJgMz=XbJs6uGTjJj@!`zyjVznH><83T?t+U$ z2I#M4{Z#M3Dp+@+2^yr!L9e(BDn8$bwcqZ*z4raW3g>Ee<+%qvUv30N3pM#?u_pWy zi*ATixdlJ7M#0zCP8w3(EcBS$Mz2+r)7=Uwpe^y~#mrv?wYH_uk@*FdeV+olMg!pT zHi9)j{!33UTT8!nzN6uqSE%0A=N z6SV1XuQMffo@V@LV-6XIY3mXmzCWa3nU+H(_Y1l? zRSV+&$U);SWtcq768=`Ng02T+AY_>jXq8%m>iRVLef2ZCPyP?pj*gnEUdj43+YuM&{rTL*B5!P;}cXxUr#It(`+G?x$uilo}~si%|mI*5(V5PZwo_S z?xHvBhQf^6BAOVy5Z>AUqJnEWAaryF^@U0F?aUjZ;)+Lf-RyBd44p;kjf$cN*^NS< zq&C_xDx3wb_{vzr8@4Axd{ZIR#$B>BQ1@uzIiVHZrk(n)qTT#R-L744S>^0|{{q&=g$*L6^@p@o&F_?O(AdyGYM zb@b{JB`6P(@RJfgd4Y+DzS-*t&M_RAJjkR?B0}G#mC+5hKA^U75`O9U&GuI?b|`;5 zO<7>X7OyT9om$Z)+*|6;I)YP#D%mM?^$aWFw$^{t@XQA~f15Pu$VY)rWI7#_y-!&E zwckp!Py;qQWsBtea%hL!N$TYpzz&(q(rMLt!fhTYqG#`S(#!F$>GEweU~`Wq+|<;*8xDT?w-$!@H4)!R05k6X|E?ygZ3JvS-8g#{7*%i;W82Howd@NXHyca6I3sM7_|_8PuPSNyR_)_IA2Wf-;CoQoWSrWBhhH~ zc+@c%gDJyxur$cr7F0oklRZ@w?bS5{ewmG~wDaYgW_$hW(Og5QZB+!*#ohaJotbT3$-W zloA7Uyt^7dERR4ZOK)5fw;xU4SD;pNF0Ma48xtmGVDsh&Z1O}Aj+(j`w-gP*MaHwx z_|z!Y;9Mr$otnrNJ?IjePO@fIPq(4F*I8`%BPHQUbMa6$Vp;tQ+Fa2~%LYxu5H$pD~Kb?#?k}ds;w%Uq|gHsSc!`l@-J&~ zcJpxY#Wh;ux(#RPPPHzo_G6mFb8C#lY@D%tstmq(?S$SNpP_j{Dk`iyfUGu>Nqn=c z+nSTO#s^_YSQL(}j7FKsiS%AA(Bk*Ap&@@9y(Yl|C(5qHr+=DIJhlfvjBLlbd@Q<6 zzla<1uSsYqOFUuWhzbv!*crt;tZ{1w>c_9aT^%_X+Fp;--@n3~lZfY)O~wBue!|{c z0T@t|fjSREM1k|AFs?m@Ve?H)=>3kZBW_}jN(zmUo=&e87hz)ACA|FV7*3yOOqF)$ z(G9awaP_%fmXSY7ly`M8N`#TH8 zL=0en;|1}!OQC~}iZc?%PPV|j^TqfrxfxB)QeH=~Q{y}; znENx|_>EK{K9H6z)5gwuU&lgqXKy+D20QxGDKaXL-hLCm!boiIqYA%DQ1m%&L-}(W;ffuvS$+B(5|NxEB3YE+comy z37<0&-<)M``_x$F)15fiUJ7I6U$Eyv-84X=CkU2!RFW$uiE2L@;+ff1IH;%<6{l2S zlIbN}9*~3!?S|o$y%o&B{t)_0d~_nEH8}riI3`GGq1@0r_?thCM=IYlhex+XW7;HM zC;5|ja90WX{*A?DmYodZTv2M26sConqLPd(j=N@zt!GT>;*-CuQa2yMy22LpI9-S@ z#(u`h?xV3aHvkv(j7J=kg!)=v@omLmF}JZ1*K9qFKYknI^NezK+kPLa?6N^`*W-9R zGYS>{+lO|8dn9~WAa*3Lz+tC`qHF95OuxPeJr8fhjGNz>&HejADJ}<_=Z52TS2YX_ zd&i#W*J1Iy4E(+}3hUOZh!!+RJbJ^w;n%}4INkXRhM5Kk>zg05nWTX=0mp^}ycIcp zdB--iWe8stnp5T8ST;y$BIZ>@3w^I=QFWIn{HbMtzp5;7)P`uZA9NWL%}bGvtHbuI z2e2z46*Y4HpmR|J%CAeri95X5&_pw$wIx@y`|m+o^YR?N`4fYkmk#5*XGcr)6WejQ z-W%+U(8YVZ3q)G~>5BueKfuLD-SB+ZFiiIQ#D1x7#Ncf^k+OKl7=2Bm(U@N=?qbk<6JC$u8cO~d9Y~p5g2?z z;+cs%$etEWWT~yksP=}z5e6II1H}##E zCBFm@n61V^lQUSH@pCrK&J!aS$>DsL=_MONH;T->YlV**M~muo?}=!q=`8<=ZEMf6{aI(C;O?8|BTUE!WVM`m1T~ zw=u$rwt7_M+zu$IJ4d$!T^4>)h!c(0I|Vpd7d$OyV!Qt!Ts18mJ#Y5X+JgPmRmxTP zOj90%m9^1MEuIw(ekF{Rc_*qjxkH1Bd2lHi1P56bof_gKIzMCw4G+9eXICC#juuy0 zyuuMG?($)q+O13GYiwZ0vPNN_#V!f|{-z|X>mYZQqL8+O-M1%)G%;?hqV!+C?o}3WYhD@95quTEc)O zN*H4}6g{U8Lz%3fG(4-Fg>5~+j(Sy>G;)Kf{m@9ZNBMfgCxgQ;Odh|Yxri-^*8E}F+bAk z%L+NzAO}>hb_Mw4n}Uq50{m&pr_PlNVBLS4!K5w{2EP=-iI^%T^SD^#E0Uux2c3k8 zPiBI1TnHPGL@cj$8UY)R7z1DF1Y&PxXmYfI<=g(FT7PrsVB<)3Qqm7h-|9Q9K5D=M z-`$~KEG53u*%D6oPC6a!n?P6ZQbL)(6e`AP!&cD;I(o`g`gBc$@Ys#z>`OCaBS!X% zyjz^umbX_$vOgx2bU)rtamONtZp_oJ*e! zxd@MiZeXM~2pja=L3&OY=*J~f!z(fHJ4YMZoo~_}BFPKB^*VfY| zs*F}nS_qFlw}bqRlgx8-0lV1VZ*^d{v1sH~RoLupLvL?Vr)?gpkP$o)g714ng{3!` z*vtp@Dep!1_FNUkHN2+jNBUWO)PC5c(<*Yzk%giWXG%);T%ivXc33^wbC`vEdCMYC zMlwyAO^2kRCL{trt2DEO zSS7Z~s6cq{$snt(pEk1&siA0*qQ;_aM#GghE3n#d4wh%_hk);iY*}{&bD6r2F6{e{ zakuus>Muc%v3L>4ns(9pQ_alJx0Z5wE_6uFcz8;s;Y$2;%9hB$d6iVUcv&k=mgIMg zW#8#~|G!izA`vcCDTBwF0vdGu5IyqIiQbvL8+^`hpvOy&QPU$Y>HdzD@V_LFJu5;= zZWJtrdk2TZ;!7`SWpD-Ep*0lpLw%rqb}BX8tuM(G6TsA3T`;K`4PrS7cHxLLYW`b# zsm_l+PmQA+uCxe4HUoT_Q3Z3C2Se!gTw#mEbNPDbP`KpVMW=}@ggXyUq)+cxQ)l~Y zwBWKAj2>T4*PTnEC2?)zA1%7=DcXsU_JOQu^Bx-cbnQfa`4Y-H?+NxClvesen*)13rJ`D!N$Gf5BsVvSMb^-NFYVh&4 zHWWrKgil!oRO@g)R1WT<%VqCkf>{bqclE@M)^XUBU9r)>&z~s_+I@?AKUbTLww&}kueIG1A6MqhEFKRHo8#D0uQAsXMXA^GkmSpG* zZNe9ceWLBpHDUAHNLV4^#JkdB!CLzaovac~Q(})nx0wLk-nLTv%_(55aSZHEZnY{t zw}R${RIynbZE4V#%}^avM)x>K@SWA65>Lu=+UR|bPMTx@Q&WaQ!1V|k^}2oLF%d)m>j$X7L^DR~)Kw0LcXf$~#AiZXb*lM$UXqZmex?Cydtn&nL*VTkHnFio zShraSdEu96zSBDJZ`1%=j{%yz?=X`D*Ru6g#;aeT+xOau)Xql2&oj=j z>PkF~8T3#%C}{-qDmhK_{C>p#=M3C4<)Lt75GGsAUu#8d9X7%4_V@9D(jsC4Iv2YK88cvt1onfdxfT* zWfi{glng(#?hB+yF$ha=hIe<)g2wX8Fn04M2zerp#Yx6!(O@p&cT3=Z`J(~nbU#SV zmjsSVe*d5R(aD^E7yLi-N8ffha#KFoaffjZCl_wXIUG^t0^g43ivDrjgb6%n@^~f5 z$ec{-%qNh?cgKqWRk_jLTAbPn zEiTab8*!;{;d;~jxcvr;xZ)mnE-H4uVE=}_g7c131@q8HFs^DXxqfLisZ=~bW@Y=5 zpKE{{cE_G8NwwqL`Z-RCG?Uh8FG%E#IxeS;hO(j&N&QQ%=y@Oa+$-H zaqkZKaswVNTw;$8=VY>mJ2!tLcPb{3Tj()B@_OYH)f9FD7G5{mHMei_w*^7qW+}hu+oQx~wF4|ghLvGk|E6iqb zlM-z?yGLIM9WTw5xXW;>HNTS=9v1|oe?|!Iy>%BP=)uIG^tH~Dj+V9wC*T; ze^72r_-xLt(VqKn`V213@;EV8+(s0h?Iso5H<3__Gh`9ulb7YC^V|io8CMcInC$9P zA$x9q7kKww69l||E651?DY&cDEjSn>M@+X@69cDCqLEQYrcJp?^eZ*E4;yv4{(YL9 z&W1r;eTpGBrD`H~@7Dy*@%UJ7(ZT`2+k9DaPg<4iPcbDO&hlJDm?o!gt;or~{zl5a zjU|VU8jw+vb<94kMM8FdBezz`aWm4pNp9OG@~usld?o#YLQ$9CmfbJG>Fct@_f@Mv z#o?{s)U6_cxY>}L@iZsjXN@Lv3Wk&P)Lvp{D$V^a{!0{H8p+NpA4&U^LENK2SuXUs z4A(ocja-lTOx!bCNc!-HWPQ5{XR+CYtMM@4zOB^fJ`Da%#{V85!S=o6ab`Q2A0*A) zc`%r}nmw4CbX}fHesq?M(@r8MSEP|(7Dc=*V+1R19S{Uut`*qL`yz064kwB#hsleZ zk) $@6VIcIkbPFGQrldk_q)>?ceUslL-7u*%ORU>-I(jpyhSjjgCd8*zi(Gj0L-5Qok8Hj6 zfXLr%A_L!R$mFA|$P3>^L{V-TQTOyGBl9J-&8CsGy1gRF=gY|2=+|V0?k7@K&`eUQ zo)YP_T7spINzup$WYmmo;ve>l9F0=s9xqnlf)4#9rl0)D9?5GF*^>DmU`qFJg224T;ozLf*7ICShlq$d(l?{oxtew?~<473p)6;&eERNkh0z*VBks0V9A_kN^w~-F2u572hp0gh3uHPj=ViFkGSjlki}*j$;%pV z5;dlr1beoU{d@z#1-HoPA;iEqz6k{|#w+s2&Oo*hqu9Qp*ey9aah zodM^UH<~MG(&LI|RFTcv)g*b~Bl$1=3Ar&`n#&x}=PuvS;p+Dd;_@uo1P=Fp3ACTT z7ffECBZzwRifA_cBfI|fkc%fikq*2@7CpE@ZnfMY#||TT%FA%S)0H^)24yZ>u8nLn zoI^^TLrH~F5Gi}OkVs|!BnM4ZIgPx*T=&o~95wp<%+YI#aG5#&T6XBe}GRwM0GY8~J(kD;bDrB(q(wkTlD3qIFhGtW$Hy zrkzq;?pY=7R-YPocV#chsnaA*t}4XgjUI7v8A@uq9*_xtddW<$F4Fw1mXwSh#A#iS z=R{u>x%N}i+=9w9((ZYISQ{K8A&!SgUY#=cWyUZrDRCJ0%vYU@2(%&6dj3Q`)}7p3 zFo*Oke@(vp>LWf6K9YzFA4%!YKV<)TS+255hWk3JlW5r=7c4$~Lr`h>LLlYYC)kox zMn2d+A~ z_sDpOvpPSRn3M+*ec?vpwTY6Fi8f?{!Yp#EXF3`4UpP5ypH0$5NyM@%ji^;f_Ox$2 zvAuGX^bd&WqkM&$X_q2yEL81n6;CDEI|h{U(~lGyTt zr5r;uf$(yO6BwR6y6iDj);phY8r~E7u(!7I&xQ7$R-HS<7<~=ei zv5k!Q`ka^*JS8WWN|QdxUH47>AX2hsK#+7fn-n@`lTzz^QX|ME1HIWK)I5V6)J`RT zCS4@SHGP7N^kL-iB5m?7VFIa)Eh66fWn|N;d*tH%G%{)@l85sjkdT=bq~ic11DZ;t zc$zY~V5~;=H}ng>$2E~^#|{#h`kajZd54^Nqs0B|*W@Z3)w!W&3f$&SX|nsC0x7iV z5JU?<2)GDUZuPApT&Rj6SKlYk1s!`!;^zM#3PGRAf0Yg7)u$U|o>ME?yZ$913bkax zf&HYw@dW9;eT>{`K0SGf z`%%eisA!P((9qISsWg7=jEw9pvy4c#5|I^AD3Yj@5J{yZBT2vWhwJi(>+=1c^Pcy8 zp8NjX&&iOb#ZMI|HC~Fg?RW(5$_m(P)nR2uJw~)8sGxE@CC*f({9Fm*FA~NxeI5kO z;elmK4g1e+7&_l&==WM7nwU5Y`IXz?|I!q<4_jd6y4m;@Aw>V_$k7rn1?p}UB-PHB z_`13Y^(&u3#3>&od;Y=Yp8!oR6Cv04y-3X7%(m`2!6uEmutu{S_G7LX$R3w&e6cpzba1IFF%q z{4elI4xp4*jON^tBjg+<&m?52-! zeicN7h(Ra$9NMPFLs%{eil-ltbVwI`ZFA9L=!Wek&M47xM%C|=C|3oR-d%)A z$_u$#dM2LiUJ5ZUD;O`g1T8Sb<6So}?@|)hAHM}X!>h1qO2A~(BBU9WpjaspJ9{R> zNoO%EpBUlWd;{cUCFA?Q68wpHjHGF)sQl=Q_iqz%??p0nl6+CMeLc=!J_;A}z4)}l z4!K!bcyPNA+RoX?SepcHqyi@FEkw!Z41_F zB>=&?F}Bd+5`4RY5u_IYo{d+azvd2%XJz6)g+$b4dSL4ECR9xyM*W{YXw|mB+F&1k zR5?KC^jrB9d3bJ@?$s%i=kkb6r5*| z1K%-Ts2vJtZpn$P?NT-~lx=3Ee0h-Oe}caQ&k(UM9SXyCEaS@&_HM>)rg|WWZ5#|k zS9mNY`bI;4mpg_+5)j^y3jx0Z1ig&H%q=C5?0bqsL)9=zDL}@JGJJ4tg!%nvpn@`Z zO^#wW6I0m8j~sSw@_iO=WCX!BGprt64vis8G@38NwNf)w*8n|wV#wMOjqA7b@#||j zzPV;W*RPO;_&;W9anD(H^ec95_jl&G^)DOwECTO`W9&@WZRozp!GV8iC|D4UX=Yb3 zHQN;@>`r5K)OG})n+JUr8zf(_##I+1C|yZ}nPwS+^Gc!3Q-saOFJWLxIOxv}xV-g* z|KF>am=uObZ+vlbMgU?5uHjr(GzJuIV3MZ`o*XuT-LHMHkT{3~XD%U9Y89N+FJeiD z7xb@rB0%gDT2~ChvXQGvZge1O`B}t^U&XUaN069ggL10`W)U01gnuWo{_q-BZ>)<2 zCJCS-s_c z<`EkTk3WT&=$(VGU$mZ}L7f(HRgXdobZcJ>2&nLoc3bZeiulFc%f!N0Q&h3!+7I4^tbQFt9&cu1y-`n ztxuWh+gEIlyom92`$w#r6qlH zsJKd=)^};rc0nC_`%s>e`((%{Y9d|Fk|7-%QF;<8MGIDpr_=~F+QRkVx(*M(yPy>B zgi9gb5{_>@awKzp0_{qbC+h|w;^v4!;n`6b&M=10x+RE>+CQNz%Zr$OBnWPD^H?fWN5m1T;wSa2K-fB21aGc{;Omo7y#Dv^%q1UhS^OVTdW z$?xM7>S^LW{|a?T=IX+ICP5=*K91iz4*y>VaCX=Vg$|1#7c!fIKh2}rjhqfCuS=d? z6RFE)CfP03qUbgW@{5>8*W(SSKVSxB?9w8~ani_06~m!9a`B;JG^fQ2; z4it4@XW=HS58sI|LL2ciOc#zFGE{j(m8#{`Y126=s@^w%g=Rt|ziAA`Lv5I`wFl$5 zy0^GNkQR%#p;|EovszLyd2SZ2CFS8wm>0yW{b0xKg+G2QL12>>nQJhz6`e-zj!IPi zCkkcru49ptFD72U4k2|_I(S~6{;r!vC6kqD!rVE8@Yy8f0g8#7LQULt84;XAvq$x) z^R7HSRe!<+R*fR1y}K4ty1oS!Two-kJBI|; zn$rBpxn$g-O=BavWY?rmYkFtU{rR&gqn4Le(oeXpeTV3Z_n05eM;m1oDc?z$WzVV!Ju@IO&r1T|H77 z$B4U9BoL!Y%X+4gg@g|IrH-ScYdv5e8jX?$ckt`&hj)!64d2(Gl<(?fFPHq_e$zG~*jToh}@NLCHJ3sL6%Hl_F?YJpd1f{rTm~aC+Ssv%3DB`Rq`@ z)VWqDeP{`8&^B`Amtf7>*KE5bALMUJz*JQUtp$2$IXMq{jV6Hc3$869mRXU{kYt`1)q-@q3OVR{7Aiq0C^AC8eGCm z=~R}yqMp4hPGW(bd2FZZerA7RBU5g4U~5ec*on|^M-q~qBh_duq1<|>N`GhtCma(KK=8QLmeSpR-O-0io; z{j&$~QU5SBR`0{MweHBiatQqfyy#hzjZmouwHG~jN37kvlW#iR$v*Q1P znbW-hrU4)J`rK7`C2?m_{}n7M-H2utKb*CQg4+vk%=JG7tI=Wh)p(E@r3|of%ipl3 z5-%|4>zG&Mi%L%?I6C++tHQ*Rn;h)QNzvS)FN9mkBBXEBw|BTQ=-z9gJMTIyw7 zySE=AavL$9cRy5|j^X2>UGOm9&-|YTux&|x?C=9u=GCKu@BUioADe)dP<0pVT6)W(x$pSY9-B?X}1+zJk&vctQ znTAOprC*v%!??AZ9HY*$_)w-#^Xa!)cQ3dUf8brrka(8=<2>e$#w2|H{@cA6 zFc=L|?~*%X>}Y!P;dpMpW%T4debj3(QqBtK;% z84qox-YJV{OudfXT=k3{9Dcxf4i&OxT)kNuw-qv-2hqRaI6BL(;p&xGWC>k|lX3)S zYfPuWr}HTDstILG7*G2umJ$VQq})RrD6?ZVZQsxY)#=~Rq{wMZqx^LF6KJN51&!=8 zCiP`nwCQgRlUK@P-+1q^t8O8zPiPYouiA6<*?L@<;*9X|#uyr7Fy}Q#)UA~mxZ(%H znlS9S8;<+0*Wt2*4lX#1$JR?ykkS!C&RuX~+9yhc^+ClNFHFZ>MCR640)OKYV9XURp&cD#4eUG$hm6HM$EYPLVd`8+w zX3*w+(@1=qEFBpePkr+0Br{2#7OxLxDL)ID%;*YxZ6r6B%!G$L~VcfqK2UTqm$@3rHpLmOH_M@0w z{SRBZ+fg<>7$1{vqi1a*{0-b;AS6rGDig@LMTy2)E0cn*2wf?VqNcx+L{fve@>7*O zp3bC*+q3A8s5ZSkJ%;u2GBjR6n$E6JBmtkv#FH|Y!hRXia0H|ArdlNEuTOTGL}NpG zq&TKQTQfnUqqFIYpDG%Rvi{3I`$heHXF~(xpXiz2@Bl(5x&?R4tuWQnMpk4 zCPboXcQAYmA~7g#N&?ZV$a>2v>hUoppR0mo@<5qHdMA+eCq?qyp-JQX=h0MQQ`!FDg$zGZ}?>9}PL5Fqdp5TDVgFCRE_Cr2NnmWqWY5I-{bmp`u#pI6S$F%Wu zqE?MAT$?~XgTj<|Wda@1P$ID-N|dkY3vSv!UM{T4R#n&KQBmr;u*F z6D@pNm?f=DS3~DgMTYX03nU&D!$q-LcH1tIt!Tc*j&fc> z>6C8R>g7YXHWj%`qnSf_BAeKl#O`b7GWsS$lRwDO<7{O*%W1r;uernidK7flJ9B;5 z8t|HHI+wTJ>F-m_lSCU#S!iRo4*|;d5bpH zPE{fEd{0Q{2V(lAAgtof`sogCxEU3PLdlzO_*{usho0m9z*lsg{fyz>c649ZgwMsM z_+xGWyV~s#IopT9**x?r;U~@{y?{g5edJV^BiF77<=0BFYw00a-7v?&gbh%8upFcl zj~zFwuv_FEX8mi#ngJsk`DsN#x~nMK-;_>OxZ}n2;~-mabQ(M2_G7L_AJU@pW7=dr zQ-<2BjcF=2)BO+D^rmtl6$J^>K^}Q(HrApu1q!qz+7C8HoVMa01&d=Lcrbkqb(Sxp zPZ9H|;F~U;PuC=ePHxR#T0qC9&!V`6ajaJ`hP@5&W~Z<6!YjK7@AazT`RN^YWT&C( zzh$&x?FM@4xQc%KSxg4=elkz4w^BNzj1l7BUyYHD_ohPo|vd#oeLbE~O+(l*lmw1Zl1SyRzLAyUy+q+m;V((UA< zq!;{T@JNDwSxM63dS$xTBSw?T#OTjLVLEVGjQSCYg5(<*cRv_8ll>5z#fS9%*G%H- zBer8x2Ycm^i;nYI5T5uBE`PgVwo8{(ZyS+F{!%jL*6MwwJ2)x#00z%0AeL5*{`Y2d zv2rujFSaIo?(ZHIaUO7(AdTa^y@xc2$d#O4T_=HoHv-T#A7aK80<@QR5=pI`L;Sfe%!U5?DR40fLR(i7r+I~9S9=f~SGkg-qdWaN z=|*P%9it`Rk5ihsFF8#0BgY@Uq<20A(*;8j7Z{F($DARo=|)rEgwmoU9}4$$qMZqb zw76#zEn2*Z<~%f~MYp>!wX_wTsZFpny^j^CM@ZoFd1@=(LncC7>Co%iS)EIjSV1!n zi}zp5Hnv>HZBDmqugHeww@`>DnvtC2Msj&=MHkadNNq|0DoniaqVpUoT=rp2x*zRL zjieQl;k0^95Gn2SC#5ylDJn6X7FYOE%WHq?`58^)af2Qtgi&)Q4{Z)npc%6@X``e9 z*^ZkGA!$Q={G*2KaCL+@`cc%Ka9ZLUMBal}D7GSu3`!Cy1GHiZ+t=wB___`CN*7@EDCQgZOa$FZM|HbrTG?(VukX)C4iMT)x@$7f^t`JH&M=`3_sZb#<8Zm7h1ag5AFI>*%wL7ie0 z(b|uO%#ZBJ(%0;7NEdq+$mtDZrZi=d34J}ei1@b8q|saAq*yYM&UR0wiG7pF?{)xX zX+_he-U!mP52Tj;DwHxho>rI0(guBDT9@ZWTki&u@$VqYJL*aI9t6_lhc{{2>KJ;s z@;Y5Jm_V5-Q>k^xkRGhkq6*%p(B9XJs_U)DzR`xUeOKs@q6eLG_n;Z`FA$&HHF66J zqtejp^eH2lPDs>3>q|eh=l8&cpO?Hd-@r6-2vaqLsbFdw6a&3TsVj>*!eOk=3h}bo2;<4F6hP#i@x-l2}8{tN$b~w||F<;vLG@5i2qv*}KP_jy# zN^M(~lfR`cO}5xX!Y8U*WSkI{4~(OidqpYq&mVkhT|$K$H_3vlv<8s;l!!uU-s6gY2kifRhB zHK=kr!vqxDD&Ws_aTKq4fv-BT{xGP!z(5q774L%Q=iB7TOMaCLYK3qTW&14`YE>a zwqf=0m#FqDK@rDIeE8l5n;9Qq|M>-0pUcL!w0k&MkcEdY!*T7zBghU^W4C8DwzOAZ ze6%V0?wCNLZzW_!mg0nHD66^V%dRQm`3$tz{LFR7&Qua*4`eYMWikU*IbPCqhK4D@d6|C{oZFaKt zIa|q}gKv)cFnFJXErr3Do7e@hz8=ij(S~*Qub`*M195jD6o&HQtUxO}H#Hx7I6r-Y zLp6lC&pi9jdls0*i=k*fNWas<4EaEootDjF9Mf55MIk$UBNBKUl@97F7- zwUFRhfz*HsjvK0n|Enr&4804RsE2UNtj7E;jTk<~?OS`B(EP3kUUeVwo96>YI{&aa zykqR7nKM;CVR5fPJ)M8#-f{^+GQuyfvO3m|o19G@r4C z1$k@($8D_mZyc>jlBdrWO62uWhT;`uNxe;nl9IGY?W8(Q=5)@}T>omO^$$&bEjU*G z1ivSUk*Pc%{R$XF?~AuMw5A(}+WL^w`3_?ovnqe_17^l{;M~u9Fb}9=29>Ss^t^X0 z>gQ`#``Zl@K4n8YBN0l=V=*aOo}!pK{d=oU&xSp>#0v~I`s9h72E)g>4=wfg58SKt(4zMV$II z?=t10Ah!6TD?6YV$K*I(Y&cDhT+OAaezz!X^m+yR;|)+&sfE$4Bs_eSg3$F`yr44& z@*6Ucch#F2Z475#XA+qC9CyaEN12o+Ym#=k7DXN%N9H>GMC%nO$61^T*eHZ$TG?@d z53D!uDLbi<&)k14M_lt7%#_)Li^hhSpU!DN+L;Jn9t!>e7f6pt6ZcYHI%TUwHnp67 zzQGSqCf?w#V<-+*yJ16#Ih?vx@w{IgjUT$%{j(0x`M4c*hqmCN;9TTX<*|Z#QR-?$k*kkBHcbbn4AT< zrk8jp_y(yjYw+veXWZpjh^}pe5KSIJ>LZRx8~O-YX^tnz$idY?WyBpYMZlkB&^x>X z%#=a!%x)}ycp43<+cEpe1TySUqeD4rbUay(-dZQ2PAnZfo71uL!d?7S8^antj)^}a zLekSj>9O@|EYj}5^2RO*#Mi@Px*(nP6(_6rLS(ph2q_ad@6hxtK8JDj>LypHEHs12 z>$ynXwg}tJtU-%KsGVaMGuBAaxFw?Wc2J4BTqe`0gOkatP=)Kcw?hA{6LKsLLGr|L z7&L?;^{fYiE<3|taV4tvb|I%moLcN8h`DoI)!SnE!|<6R)JZiXPyG!JmOaGNnrE1&!9zP%3X+l72$mi5gx4}Zh7tea!jN_rc-IC;= zFGtBwexp(<0Y4u{!uC=i_8s-WPO%1DEFHwW7ehE+{~b??-?QJsYB=#xh1)}b=GsPV z;XGWb=s{3zJ(Sv?GP9{a*xM6dnA0gylr9QGIj8rV7Ushzy#j4JD)6zP6KN-VF3R{G`M^kKF~Lly;Gq z?p){i+Y>?*hnu-VtjDifadK)ErMX)rD3jx!;U-E)YEv+gv&NEEi^aayxKcS}k9Y#i4ne<=@)6J`9OLoVxYt9$2F~c2ovyS6| z>tQaV(mBg+-ePtmZWY^h=pZwC#6u>PQk1~eA(rRAp?}Izyxr^qy^0f<#kUW_BhEPT zmD8(7d{D6089g#Tu%V1&>2LL+FZ~;Kh)L5q4>?L$AWNG6@zBy7NviHsCC?cX>CQA+ z`e+k?(|++-X_kW6lQ~$G7>XB1GO=4V0T);hZr>H8sthp_c`iVyE8Ea@QIn*3h@@ZW z&^QM*dhK0{L$YnSAovqcQ(i*To2&nIUSXDH6PEpZ40BFHQ{6e9nqJG3On|F?8f~RHGHcyh0s|e47u(DyV(HO@n4Zv(2CK|uV8EP4GsJQD2i^yB=!+U z6xBHwO%AGplMwAC1nb5Hcs+RuoU<0-vcDmBeSV?m*$CFUeFtu~VN@>{?FUM*?#N>_ zb9?*ZN5jbdC`iq(|Do%57tYKrfz-DOObK}c{*r9yKKO@S^TkL^@E<p5n3@S zPpBbd4i-n{1tyW zM(s6UKfWdOz<}e@oiG>uR#V~hNQ=7;+hJk*6V9@jQ4*7uapRh#c^!h9YMwCA^lUR$O=!frwnohEn~6=ioR8h7g5U^VSpO#nlTrilS}eyu?PHkl z@)@^n#He_O6xGZYrg@SR>HbPhiW(SC)1@Rx!i<;Ptwc%4NQm-R{DEWLFuD)$lcW$I znV%7$1tT)3d@hJX6Zr6OzaoZt=W`7FTHLYOi|yfy(KAB`jXMRf&9R>y#b3trSP7eS z6!7<=7VP&-LdhDAnY5Zn`ODPEd_;wQJt#oz(WeONFT;wH6|jksA@c$qy7yC!HeQyc z8NGZM6qUuF8RMWf%7gF2BCy<}jXry2D7wqQYE~Dkovww}dS!I|TmVzg#~7bqj)<-j zm=vbsc1sEp&C5{go{b52@1uGuFJ9ZvKvCE@VD>~rr@O=H);644;E3et-PknV6hj9# zAtL%H+^=)z*Us}O?mLSs3a1e)WQe7E>ai~62Tol31Rd*o82<8M_WA#@`horIi2Gr- zaY;TF-G2Cg+u#7Rl)|}~h95oFo?e(xT?S`D>7v$V* z#*|If*wEaF!mMiS{G5&-X%eI&Hi3G6NKtG854G0cgJDNGV&_yqzn82tM^f zq+S#XePdCO9fiGS!u0lvBxkXT(xOcqlgsI0x6Y5@&8!jpOKM-MyJOJUqO5oFu*QAV>k z4XzU3<7_ zy9FS%PaSPK!nh#U#|+ka;bB<2{|R>y z8B&hty`|9S)_!nV7(70Q;@59)Wb^wWGU_N+F0g~{;1+zFdIi(C9T zBjS}J)Prmg>OSqDjQdx zhC#?h9HUZe@c!Nk2skcAYtLhL_**?&HKUb%?X6_j4Q``b;3lFbhT&G=X&kjV%1Sz& zS$DuyHd%WQyHv)H+z5SCC97iGgV{J$bq1@`f+6}TfXlF~LY#dt)PI+OCEkf|Jd!c5A5IkMphNu#nz^D zu;locT>Zw2z?q7;IHZf)i8Am7L3Z+TBpWS3$Ptc#URMAYsaHtf{tVWYMfmhG55-PZ zSbm`pwwjq}eBg$M*8?Hnz%h~>FL@@`7YlTwklUGvGBqFAg#U*r`EH24JC13uJH|>3?jt|!1vXZ-!tg;c5;ba2=Jy4A&h#KuyagKt zx52OK3Q9|T;NtAT)e#{$^oHYC`13I}FdANa;^5bkkF}dJvDGjL=a&|v;&?4=x=Tq6$IHh!b-qFf2_laKJSFc0|?5}}}5 zh{2opVK*rQI%AJmo%0||GJVJRpM|rVTW;acw)=?h$-~dO1k||OKqoW-)}onEu8YR& zfdpJ#R*bea*)aW)%xNA0@Ka2M2Twc_rYGW{SvVfGrQyY!YNRj9!WEmFa0^esBCb}L z<#`ix?YWHYe0A^(spGM?0h<5DLMbJQWBP6(&o&IT-c_tu`8WIPTF2(b)w16~{0Kk! zo0b0g#tL*kGsAHv=<%C^-0jME8QsePl|qmn6vj9;eq7DsL47?Ra&$By^jR3&dKJ*{ zvk*J>*5gfg0~$BwU^^v3Q0xvm1(K1Y9*RR;X5i+Zd~`2<1U(II4c$z{w1GlQ`cZ_B zO?mJTjfUshJ7}B}jdP`LkZ}pblpnXyC6We1hZsDbUB)T`(wJ0N274)7#rhH?;T)(A z(o}*3w>BQ0$iuvtGBoXcgym@|IFtPh!($&X-Qy|FE-ga9%SiBXHPpNBsd$qV!R1-C zkacf7uI|)8=22}V%ca2b)IG@Dzl)~_xV^O`jeT7i%4YJoFmdihDJq(MtgccByASeqtK(+rMC+!#&5>cWoeK& zbsy)O?<37SjnkvAa6H?7w2A(Q)g6SfOaUC7!3VoD5-865%Bq)ivB*omnU}?H#v9nq zblXayy!t7Z({4uBogz%T{}LrXJ8`+B4PQp;;I}#l{9mi_=<5q8e<;P`;V1}qq~MPJ zZAeoH0{BHrR8M37ILG$Q(JWn@=VDVyXi7v*!LvOGqryjyt4=}nomnk27!iJ`1u#}Q`#&a$C*d}ATjL}g-jmo?tr-h$Ah>#$(w zA~?Ex;L+6UxM>^?nI2bMzZn66{1`-WI@zZdcX&6%!NHZw0eF^TsdOG@<@;dslW62^ zPR1@CZ|oYrfJ4D?xV|h7uD>IZW|hR$3_IBwtv+Uv^`7Z*`}sVZeB>A1#MdEb#Fal~ zG9G2@0Dm$|ZVhJghd#0!{nA)5dzkSo5=YRjqiB=ka;PCuaLwM2g4xOt*O5R+QYCv> zU&YF-ld(Ie299~f_*r!a&Ku4^-jTca1}^w4?g}euHCS+Y8&8*6$P|%8g?%p9S3GC) zu0LY4zZEn0^-fUM^FosCbv#K8L9vH3o>fGk`Ls7=G%liic^T#pyaY2V!I|4}DE^R! zuk|(P>Bz0~O(eN1LX z51XzS429VN*m}SdlXwqe$@>c^D^7rvSTa@=w*<-OrI1%Jn#Z3J|D%7j_n9IC8821>Sd2 zezBY-cQmrRmAUNNSPpxV91LZV+3kpKu&cg0|w}OlMp;e+Jj?ZpHiA2Jkb9#?tr{XeQr;(OZs1O8Lpw z?dxMJX1`&$slSC3=i5RpWaYd|t5jIY%L;AP|oZ0^x zQ-3{VNA5SUhj*W__K@#v)`j;h?cW_HSn!N>YD;2=>I0_!^DA>@^{gYKj7^AKfw8KS zs9kUvLdz~9cUmM?IXyyxP!+ZZW#Y)%Ml@Nz!ITs4aQ;FG^km(c)oW+Aq3Z+_9P(ik zuex9}Uo>2egW+l5f!&&t>=as|_&OU)jttJ{4{qkC1xvA>QcS!6HdlxNVNZw~h!r7zlyn zTP-{eG(h%49egpGi3>K-C|AqD;)*!4FgeJMVo%`)_+C4Mwfj#&M(Q+X zd8^aRtz%qLuTh*NJY%YVXf1rM;6qd)v!ILacxV=QTFZy zgyb6`GFpLtULkasNuocam34VbptkiZ&L0(^zdF41UhzG~&3?-U&egEPOIsLE!gn^> zst8p%2^`q0h>5dgA#uzaAH=u8pkNK|SeT&pIp?*MkE0Tmab#gHLtj3=hC|bLOmyjm zX$8j!x{8oUybK)|mZ#9#ab!2A9{Ue{N5rWSs09DTm;3@4%zBCB&u@_u^BC6iKfq>h zKcr;a}4UgZfeE%@LqkhkrvQ zz!2sIGjOKJ9J%6iu`xsr16zbKYvVJv|MCyku$_-|7fX_0sTf^vQlK0DV>obo3}sw4 zOSrus?@~A&*zyH;X4QfBWI31VP@=^dlgN%&om`ET={Uzdv7J&>FDy&h>!s-3e*)wi zD^4eTWXSymH*X={2()=2y#K65vBpx|FSg{em|V72d?Bi83=psJ0t=Md;KTV`8R`{S z^0fpK98>=FS}%B~|G{0G2k2?4!}c95@T{#vZOs{^PjiCIwBrcTaf6oBRo3@1h)r=` z&$Om_v-pvh?C_N~w&rCWGnzEa&T_LBL^zfuA)^Z$*1m*<1n1>)^>)?CVyOC;!XhsO z;)eI2-<*xvoF_Z*s-DSYkB4-)F!X#CP;%RYTZfUbHj0ARCVw<^-oV@N>+rUTgi6O{ z7;f;u1n&2Fz0s%;49BTvD>U440sG?uQ6~@1xA=mZY90z8elq^_9=Fyn#iYwCAs@L2 zIlHG}Om!W`?6zaIB9Vi)w6W-&BxEvA3cI^Ex1A}M7aGU!Ko3D8Rik>j=z5wUb zDzM;Y2@H4&;R*L#PlFfStT}Bd)(5;9zBm&cf^sfz`msI?CnGLmK$hd5aRm1pkK&cX z0a(t8#X^&G)cSI{;ut?Xeb>W=WTs=}jt-ha%puyj30h)n;d0ji8`sQ)ltMpD4Sv8( zVFZ`7IR9}&fO?O7&*$^D%`lzBTPPe#b1W#)QpVH}geN`1+D+{pW%`>>~ zdknQOuHQMblL;Q%$u3X0$mmH3i{>)lKK)m*B8bZ<(H^8y6jXP{z;a7Ge)k5$F13?Q zPk6^X9dEHWl7%cKn(OZtq(fs~78HYGVfi{s#WMLUywkXY(K!!sn!g+>+iEcA^&CAvy%6_17FU8}@IfO8eHzY;C0eq# z+s&Aq*)_IqZ2=Q`_mv4f`NLvjf3dhf`wzb>3RYRyD#Fgf(7ny&#}e)hj`*win&|{ zwtZP4M#W{}IVu1ZS7A8tNMMYc_j5t&KBfr%&&j(7U7B6>c^jAtcj^NH<|4sC8ib65 zbk31L0s~q{0ih%Y4g_Q3a5)$-j*C!V2w;Na%0|I3E^Pea01*V64H$chn;Q|saGlj1 zj0a*v8;mmsTYlpU8AOblOBrD>+}zg}zo)^+dqJ>S>e)7?LO_>mv{ zUk{Ie=+7TM|G#PM%@(tg2_=VTz;TM1B?;L*SkAKhMFMZW(4QT(D9;O6Hmv_{Bxgw`0=;<^M|khy1#h% zk^l6u!-stDhabP>8((w$mf!h^#f@K;~`Q-}ZaCH#N%@BFsI<)^;+@Yd(Q zb@=p;y>*CR|KAUP`~33bXMM&;9{>Cc;`q&;|t#O zF2{HKgl8RZ^Y2LfGJod#{U3el@yYLtLtem|0d6?zx-W~&wG#eI{yB*zuWPJU-27T{>8`7_@?I^pMLlEJO2D@o_qWU-}^qtzxC12JKp-T=O2Id zU%cP(O~3Z+PL{55E5JvETZ=hZlYM_aDCN zJC?%_f7ri1eD>%53;wOBrw=cAtG7MA|JS_3@uxoR?T_F0^t&Ga=@&iw_+|go^N#Ta z&pZC3k9@D=f6bpWc;Bym?%|#P!UrBc^^5-0;iWHl?(xI;I}bm@pKt#AzyI>%2mbTt zAMZTt#m6^&_x|`3@d3wA{g?b1%y!Y`9pZ@6a zzkL1Ehj0FXpFRBC_x!}+^`Dl9H~h%^AHVdgKj`?TpFJKw_&XjwKIh}!n!l6qtm8lX zp|?MN^e6t_;qU$=|3=>5`QYQ1|ErfC-|>(9;p4ab@;4lw_35ua{O(u&=;4p^{}6wR z{}=u1-~9f^A9&&Ojz9K)JootN{M$?4cKe-s^ko_{MiW`tOha%C~>-TmAq4_eWp8|B#m-`G-Sa_0sky{==oU{z`ZFGk@k=zUvYG z2~_d_^>3iQw0-6meAb`-ufF(mUh>(W`GuePk`H~^%Rcz9|l*Qs;% za|i3>7i@R;>@OqVb0;xpD{7yW`}df+0e58$`2idp8Gp%fz!T=eLhXwdUt5#+0WXBev~=p2k=~HV9wSH zulc#?G4JZ5u-(~S_O{Mm?qg2JeuSKcJ%IfJF0kEUPtAgDz5X@YvtH27Y}h+^cSy9g zS7_g{y?#t>t+5Y>f16j$oo?WKF4ijcbk`2wM6BmSJIlU15TGY3A%(>PEynO-;bHI)|_q3+N@7B2cCENncSDf3houxK*3wxlgt9@eT z-{8vyzB9Z}^Zo24dduDB4DysE_#r3ES9;D`&tRSPm4n)1>iyc=x#rxQh;xnw+js*J z*u=GK7N$8Z020JT0Bg-#b4L<`Yx3R${sPX#{T$g>&YV4PS8Kbxf?Yjw&bYx3bAtas z$F`rlbaa4RwY>I@95G_%dtQ53;~uqFLlbugdzfSV8h^{NUlWM-009*31kRDG`t`i} z3$Ty&!~3+=_bfYRUC0@54ZmT#lePAf)3=9vb+oqo4|r>vK3LT3Z+>a)0T<-#K;<@s zc)*ufuO$D))K8q&I5Bq!_#}R%jW*AI^?dOgb_YWn@jjj94marg^J4b#YpnHKPqnq4 z$yp6s=-WeXvlsXP1K2~@uan~(cU;ih{bbI3>#XtnK?e%nYF*|eia*s1` zj*>BZElEBzydQHrFegMz7`Lz70`#>9uunrf)0i{NEy;OD<_Y8Kl699o`sOq#pP7B! zr3YNjSzQfZV0O(U9Ps{}Fv<5>E0q)TGhtpoO>XNYH^2fiTZ#20F_9_@289`tIP&PvYh`Y<Dd=KV2!qUXK+JIxvaY7%yE``?Bru|#+06=?*CxVkc$WT_OG4bJNy7XH|L1~Z_T16 zw&i$cxe~N_=as8n@EdSui)ot^U{792%z2yL;Eiu*+2(D}&R9~Tu02kzdiT|Lmi#tz zz*nw2U>9O?3)(z$I#_^lZRZ90%6;6GIrf!Hw0qWBjUm=3OY#GVnE813pJnmH%n8t9 zYWqL4_L0AwhzSLLJ4?^Gt1Xb6-oRHcK@Iy??RpMp*(cFE^0(Po^ZCLp2_592 zZ_&rxz)ny*psOw>xAhqK`~O$ZMsA0 z#N5I6$M3b*z#STpQwON$gzFy8F{h(TV)ii4n(etQ_;vQ2H3MI%TF$@aRqvR0WZeDE z^s!g%AQ{v#+aAW99iV|5^0e)HIl&CEe>am zGIHLpHKOuj<p>6_b<{n51>w!Bz?JsR2>W7Uc6 zT;=XM>wKCuK97ir&$}~!&uF{X$aNQVlGDh&0nV@fXvMrGH0J)}_T1Aus{5d3A;+2K z?BF{(K+n}iU%@9_Ltg{G2j7w-CiiAwZ*!bEfHSU6#IfDap4!%iF_8GyT#REc*w)Bt zcQ|vM5pwu-k5I;hu@071yI;tk=AAw*zxwh4--Ea|sD1Ds-|7f8{EI$%kG6N_e6IJf zn;gC|dG%OrqSf1e9VpxG&Z}q0yA#CZ-E)V12KKoNtQ&&-L5?uL;hQ7;p0J(GpBI$` zDd74vbx&Ag3tR^Qmc$0ya@IO;Pt6u@@Ij7w3$}7?-nzHGeZ0#Zu-;ss%|NHNfFIFF zTzAGd_5?btp;$o5 zN1nct$q(dTKnv>89bIS1xzmDgp1PX{Qosp!v^Ic#5PLK5P6A)<`ZV{{SIqIgt=XeH zaczI^%Cnlc0HvN;&a>4$zzzMtKsU6t+g+L8@rzGv0jHnXb=S_sR^$fQ!8LRV+8zbZ zhu?@VXt~7p49U~b)Hutzp2L{=18ra5r5m}J-@o=J@`!Kea)14dxto36 z#drz6JDL~RVLoEd9&290uW0g~88qVeWwqtq&pwG>w6_?#>)f$qUH6~?s^5^ z@zu(iULzjNjhuOPzB|*d9`MesKAhEh>*~JXJzL;EhTYKiQ8FhwdJww?J9F?$W12gF zcToFan}2A8Fw#dH~h42<@G&N zqP4BPtaa|N*YM94=XdOq z;Cugx9^~5BeEZnLGg~8PKflHd^cfgmRs{Z=<>;MoTfN$x03|t9kJ+=K?Qgv9f*wN- z?Ytm%zt?Ggrg!kx)w@UBGY+(M_V9D%yDsdjq&YG3jwUS7w)EH&e{w;{FwJMh%eYm#N=xq>}`*x&3*1pGq+~J zDbDhY_1ws5_|^q_#Ox=R*d3Vf*Iw_(z8i0T#o?4C$t7poWyIXCo(c8>l&WX0wZo8+x*JIUF3rQ(48Lj zJm{y+HMXpA{UPw>>}72S=J%vVpYxh$tfA$uiQlaG8}IvG&ye^4=4DRU%RWKP0v{`H zd@%;M<(JRB1)T7n?K`Nw0*d$RzV=)K8;qR2%rUOt;Y-X%?7i_BoD<%?1MF|l9R_?S z&)&k`=9W#)&OGAgi3J3{J8tLea1GR~b;pGqXQ-b*UxSDDcNgoExU~ta4^V>EH?|{p zL3>tZ%XNny5X9|mFa040JgYG=Zlmw4e}*O37?kG|8MBA80`%2y@Ys8$ppoP8UYG<_XME#s2 zW)0DLZufV_x0ib<^45-+GqsyQuF#QF%ejBzJHMZ%@6Q+P)!|d^S{C5R>0osDnCtl=+{cvOU(Bo(C&JF9{!HrJ>egM-_QK>H`nME z6V5P4&UlBnPJY2&umfm!;^m*_S>%#fq801+prsy?Gq&}9n%E7n?sEEr@BLU~&!9VX z0qo`74(#(oyt^fwd5^)5m_6OYTsd>(y-RBwynbQ__Oil1S9TZ9`AVigfS9wlXKO#} zhB2V+8Lp_S9CZ*|k;FSN@18OAvf65M19)dmAS3tzyLvKjx5k{^a@PA?-PL!v0ejd> z8HxSM32QqrU$Mto&Uj93Z3D)NSA#H)UWhs8E_Kc?z*RolewB-@4*O%5ScBxMjpqvM z?OC<2oab#@?TNftpP4zp9hI7!7=ev0f7_Qn+w4Ky_h{RD!MEm`-o|bbv(|lv{gK); z=6$`#of(twZ~=Tz{5@Wumv!~b;f#5Xi+!-`9?_K}hnVBDN;Oy#wdOm2fxAK5`wZNz zYTyTGIiIa__v7g!=EyzJvK78s^dyVbS_53>O4&Ghet#M|V zvtZj}fJVG}#j45K!P%#F%USanerOXP9&1Jr%7OYOv7_Ruyqw29d#m{+sf%(-HRPW;h3>(^px zlr5)v#QfY0^gZ%DU(5L%7C8MxJIiyvSs&Zp4c=Zp2gM$q+Zt>2ouSq*JLmgjy*3snh4Ohee0QvUW6f2A0UB8H0U!ALwWqV(*S^+YPOYTokykz8%yEXb_LJAPUVaCj z0@rJf+#uF~=e3`?)+R8fe`UV59N6;YZ-LhV&uEP>t}Hostv%uF*TI?md*Eqc5BrMq zX~xa5w(2;8lF1Ku);pnQ#GL6F_Ye6&-Y^%gp*^$v>8qVxabio%xeZ;P6*z0EHm3Ho zTYXM5cbn%-du7geM~eu2a?5A8#5dPH7UG`i2D=MvXNv`xThAszN?b1~I-;c6!Q)#<@Au(y6g*Y8KOM<=J*1I~FPW{%q4 zV!_t8r}bBGk(1lviEoc(jTw)a*mhBG7{azqE#Cs5mWt6Kt_bIA>Caf7{(*Rb8+ zI`iae2DW`dV0U0_0DH-o?M|!R(B{ft=ozb@aP~C5V3*`n=E#d}uKfqTb?S}x?)*&o z`<5ME8G%iHN83X@!))jHZil(+>>wt8-(O<`zJo@rfGVA|6M+k^?2E50*RU1&5tCO} z3{L52{SJ(c#0L$#qXVrNi-hy+*mY)X#hi_|zUtJR%h^d>TuyU8RuXql5;{p7;GYMCs&N<_}&Pn_hvjTZBL$n}*Ew6Tl_^~M+A z&hFq-*bPL?T4&n3VsN!9RP8}*z?pA+XCD22;m)jgw{Bd$ z_TTJ)8z6Gcu@>7ixSRRrmLwx$=52h)HaleE6}(Gv_n1&trF5?Ha^N(A#-ypPclz+qQ=@11vFh$2O*K{n;JRg*E0` zt6ra&qTgcr7vgt8;w#4LoNys>a@KCPdFDGe*SwA$pk}CZ&9@G}$L=&DJx*;B>p(q- z3GXz}EoZJfIBS8kc7bo$2?lTnbqU&D+kI)<=R(dEaQ{utC#)nl0Jc5kI@}`9tS|-! zzCCLG>dOV_!2RsM#M}#}&fe&^|sRSDlxzqTS&y#Qpo9Tdx-L z+}1d|&H`uOyZH7|y#M<7K6=F5O)juExJSbuXy+MEeEaNyGxu=fX<%;Epu7*y7&&Kl zwD-_szDM@au6#AJpz^mnoUwvzY^#d4lpTv%sF?r*Pyy)PXTyqj^A}_Z9 zzc%<>-Bk=?SNgYuEnF_gR~}<*6HfUtwH4 zeS6-kF`uI~=47nrlE>CJ?tS|E5NdOM&vw8X`(5A_Yc?^j1wGJZqj%(WIbdJ!s&ep0 zPJ}T~Iok04J-FH(|Li;vKzr*=#(XEsn9tJKi23;}8rwPdd*gd|4}WF?PPnVmbIs8Q z-uv|$YcYSFkmg`p=iS(|$pQ6+`~^LLImM&xC*Oc~VQ=Re_q|T^V`$ImEN9qT-aZR@ zpnW&ww9QfMWll?gWbgy%IeXOIOOdB-T^Q3wFJQoFtBsY`wl`R?6ZmZQ_Q9_Fd~gr? z30hE(Hb=X1*I8wAt_41F{@s56A4%Xhe4n>D3%2)jwZglRGv;}$kmYo_QdHdE}SmROZQh zHqSD!FXTAWSZK3{vr5C+bKocW+V$??Z#DY98`jirK_0L^;TGV$fRhjWYy2Vbt=oDn z_z5acjTm#rHg~(PM!mV7Ls;WHwLR54_>OkIG3|@F$Tm~^%_lx%Ac1{H%)Wex);L~0PiMRD4(x$;?aF=HD?!h>V_*+# z@46(`-&3IGDPXQSi#pd_wDtB)^dcXv&Lr=7@6TIf`rEnfc{T?O;`LnMss>-(@a-Mw zs}nx5y$_!02DF?zRlYf3iCGiG7GR$nczZb0`?~`ETt#$pf*f@dh;<@>xyA=@&RzVb z?S181FNbZ-HbY&xXmh){F=4Ks-2rySSG~OxmAl@7&)^26dG-JU?C~!fNB@M|#Ch&~ zP5c%ZV1H+Rtsd|Vcoyv|Vt0W%|O6JwsWoP0=vE+V9VRbIrSdl zH~N+2ROUA@)QDdL+{fA-dW$=ASPO4V9bggB>bc8Re|&p5dw^^0od1?kH+GseXS8{P zSO<5pov$`;d%pIux98ZeG5eX5=mFey>u0QiE~vR!<6Wbjx4 zcGs3Grv?%9dlH*o`!!GAKGs~|%o(6Gyyv;!+1I@j-`z{F7hv2x zdF@-|c`riVogc@RSGPOlp`F#yiyEk{QH-^i99HdCXG`o8gqqy5wu08L&sEHvvFaQ6 z1OsS0Ynv0~BnUY36>ZPbV+FQ33%vEWK*PR53-hk!R*6{mhOYlj|-@C*b6&|`OLKC z6+f>%=brblW98+VSmW+&e`oICl9+w88=SVX@jYf=pRMOJehZA8vr6MD_FNKUi(@;_ z{Lp6qp&he6*F0+ym@|lZmK~_i%pLCQ)aKU|H+l zT-M%l{@UXieDB>|t1jjqu(gl*TffZtb+(h^GqA^=xH$o5y!LAI*%(Xs0jw#>`JP&@ zj9kwc^nvU6exIwp_zV0K=4umjXYGIw{C0ajn_F<+wO>1Spo2zyoMx_jIdjCEQTG9t zg1k+V9On&qbFQ)7VI&zqyZT}qzah_EI{w~Vv}geboS$fOA}8GAzF>Rz<|>}M0dq&p zxpK}La2?psJ2=Zzd&W-q2LBkGV!s8L7wDQn4`%Q9L+F$X#GbZfaa_$!B1#Wkm-`lR&8h@=?wZzl|ZLRiJ zQySlzfo>pUJwMuWm=oHb3vG_FL^nNvoFATBZ%8aP2I=`dy`RTXmFK-ZZV-B*j0zE^qf87H2c(N6wVrB3p%x# zBLdFeYGJS9+7$1%uR8{@26|3C)T=E&Ko{<^YI6&4f$ckKo}zD_2(ax_yRG^1JLapE zCCH=49yIKhQ=8+iYNd9_+0%aJR$uJ8H)199sh;uyBN9cQq5Y85elg!+cfUV_T`LB6 zMO~0%tmE6O=EC`&+uykL#!6$#8QXe}oZNt0fVmZCPwfQGvUkRO)-{K@n{2(zkC-`u zZJp0Su3~Vd;XQZtBi4aAbr$-{>#KXtuMgh0Jr(DVoH<2kgBFu_u67}(u9}t8cCU(~ z^^;g%<3uW^zkmRBMrg%ZduF)P85nb)ZLjKMqLMSek!Q|8Ux4%7GgjMYw#`fOuHidc z&iTUrzK@&V;1u&4Sb`kt7SQg@d^ta(_r-Y5+{(uuF?%FB@P%iJ%|~y~rW0%UigiNY z-K=XsTPW2lrgry^Ek;gQ>-QJoU4(IF%a@=t=UsbW_BY44cb0J8)fqg!YQJBXkYoM@ zf49c$C1)@DB1KGhx94*0O~KqPx88W za?V<`$rnRA=PrS*cuvI{<88%~*!0}^x^p<|FYwNnTVnRRwfgcEgHLi3-OvM+pslgD zu|S6!4B%PiMFU-pJaT8wP4cT(%w6hi?4c~g?t#xESH1*)iP@(o@4{)GdEbdvTEaE0OM6hTrC>d z`uN$d8R#X}_-#1sP23}hd&l-w?DIIB^%*nYo*mwPWy9BQn-K;lWy|HiLcfe@~9t*KyoV#6C->=_I4w$2CwmAWsSo<&ZjR0tOt#=wf&G`c{>Z-?+>vO&{e=YhVbSFQ^Zy>Q7=-57gV~O@%dmUKxiP`Id z?^$WBS%@xx*Mpu+{d}RyQ#8y@L~e0Q^9EAL>2*Zq}aW`rOTn z&BiSu$GAJc8CcuUepWlS@5T1HB69x@?Bz`L0NY&_f0}*dlzN}l>%RJT>3<=w-aGSx z-uO!b>;TFYc|OAq+)Ei~^UN6n-`)NCRf6rTyC-5ffBvFxzPf7Q-J!TAISMgt&vb!T z@6paLoMtciEsw0~7sTW@I75k?_nPRMPu{X}Th5OD)h}9!@dhXQY6n_dJ>-H>K;PO0 zc#jLdvS*I`06k}(b>_$=_AUFf=T;}bqt3i*bVJ`GM=jTaT=+T%e$}eio@$?`d5XTZ zK@UGu+SP-ZBZ!0Y69trrwGO-Inzw3meGc-XfsC26pzFSvc}-3bbM8Q2tby<3T%G1! zHMF)ptc^8hP6G2EqwZ`DzG7_2xx<;*0{o0PLqE}j9D9l-&_M2f3|n6Re(%$Ry#n56 z;5~NiPOhB$1a{p$+Wix~#p--H^9KBziQC`L?SQ{QdmoCsR==EQ+39095$hcEEqHe^ zw%|Kk@!p&vw_ojz+j5*8#Lc_9jh*v8J9dJBZ#+TGKwE3BzGAI8!t=VbVy(}{uLbZM!UR2`V4$+zj+9;n2=8p44d%Y|mOb#MQ=YU-V|1Ehl$bV_VOFH|AbT&>tjFjCl`dkk|usumJNLx&-an zZs6Q6V$%M&a4I)ST5e zHh?m6o~iCK=B{$p6W!q&_2#?#7E|}roKd}*8RXP#VvBL|6?+(KAk~QpYCm1=Sb<%+ z=y$j?pYg-F&N|cA-u1lKe0lq}n6o!M0~uZ);5_8Mya{r(0QIeR|bzk03wp8T#2cBeTfvF+#c6<6f=?UqSv?uQh@1xoZb(b2IP-=QS=jKnD>M4QowoB+m@>KSWJLVyJ37|WQx zoHZM-t{!mi-0ZpCI>m&E+Fn0KZa3t*kG0M0j8(XRM(=a~{GX8+P_ zj=6P@?qGbY#csLw#q_!O{Oa@a(5jC{lJo#eT>^*?LB&C?Q41sd7yS*XE*(rdO@3O z+;gVdec-==@>k&gY;5nj;)w=h#WlQk2h}I0?=$Pzq6B>=cL#hY2flF5Cg%EVGT9e9&<%Lj zfNwEt>aO6Qpw=A&Ubut)4SeD&`q~{h%e(~)Ir_?pl+X2hCU(tkGY575e$7*i3;q6( z-LUQJTxYb?%&XrIA?7_>C%?qJfA97hRB!Y`ylT-2yvr8!1n#<^ui?(vv2s19?OfyA z9T~TN4>qn-=X|-4Cr960rS?OQ!2j2%7k0zrJO$6KIW_ax@?Yn>(|{pIrT!0 zu?DYb_jqs*_OPz*4DC5;27CvKIU}aO-Fx+fKa=nIu7EjJ%k02T(7}?e_}%TvtL2Q@ zdy|1JUom=kUk(2r`SmpxTj}{=pCD$Axz1aFyIsSlygt1=WBGSxylMz{;!94htI4(A z+$3gwfU2j5`De5>13N+81$vVk;@7~qd3)>Vd4;Zg`1^5e=LWioe1L&n=cCR$HWt7> z_FQ6Od){PJFa63TK0w_&`R-s(-$~=v+b`6>cs)-zZShfe=GV#UN!^XLh5_ql!-E#$b)`lGRAbxdf zB?Hd-ieWcw@33Rn-9l|*&dxeuJb|3z*Zh%_UjiO`6YtBu&g|$qt9_K1b(zz)U)ADk zyUW=+Y;(6BTTgS{bGtkF1#Vb_-Hor%&sgsV&b&^4bL>CBrmboCDag@JQ1^!Rj<{mYu;K5qt{d~56X>WL3*|IBqmdj|`g=XS>@@asJw<_`7? zY|qieFfv!&XMJSlt+#(C-`Ev30hD{_k!4oA;U~WTW`_B<@OL1v6P&e)H|)AY zO!MLnn%VC2I2dppZJxH;S?ANdzmK{DmGfRM@cR`F&U>-XQ((JGpY=k_UF7X)Y{9ng z6-YT?5BDjJe?!dLw|Ui^Z{008`5?aa)waK%6KfZI?@eBA7=!Pievchq&YA%?@D=-S z&(r(`_>2|LQO_!H?##TzcK`Y{D!Q$?=2>fR`|OErv+UD=IW6YT>(ql<^LkGHyyn=y z!P!e~?TBgjRWG@o55E%+d~<4F`(P_|_QW`~_NVX6xP7g!ebm7E2JTtQog}ufPYP`J zxPZVOBGx#zJ8v=j8t*Hg)|e}gz1Ls0vmTydj=dK0730<>SaS9>U)#^>z)tJ{=I95o zuAw(>M^AmR0C{Jreb4U#?`Q|y&HUOe=lgA4!yc=~t3G@3<*&*2Yj6+tmqXjXpuL{M z8QM$EUK@X*-ZKnzM_=rLmUAEDH4FM|3?wnnzx3?1naPs}KB+ zU-iUq)_iBWk3EESi^w&HdHVM8Yk)e{aONt-$Nc?$?`2>s+CINJC)(N8?QRo9o7aIo z{aWw-TfKIJ-^-)TcUN^J8e2csc)|~O>lSSDoM&!=TjrvTTi1y*#HdkfT z*EyM0bEM#e&wQYjo^#ILyiIa)y6)v%d(@6-Yq!1RwtKy%N97V%n-{ZCt+kVF&gSXnHoZJ(P!Q&1oQRqmhbXdw0g{S+dyy4+hxL_Y8978X|9P?Yh=h zJZD`aZr=q|4q7hNO28)?oZW*S-e+S@k9nT5;S;}V)rpC=R(&DgydekFevh~Szuze4 z*1L>vjXBEG_+o%MOA-hDG;_@D#FVOmx93yfo_vyz?lEn(^$Te5RU-#><^y%j;_l>c z;ag*_vC1RWT;o@0-<5{#9r^X%oCJ4)oN;Fc`V438z^^>)1P$MN+7th%JJ>O2_LX<1 zz&7sr?K{X(C;R}$t+TKBjXoV;EID;xUx+vS-P)(=BTLLa#;o_(pEt(foudw5|0__l z@qM16cEp!2U}w%bzMt-JP2K{W*@>5EQ&W9UbBB@4g1-gsf;tO3Db6vsB<76TCt*&+ zi8;$%2XMEBz9AMlYdhNh<`?jv#&^F*@63Iyu}_1mca>ac4> z&tXhadw-Sd4#*XGnHVwa>~Gxwi^w(J9@xek7&&9+={wtA_Sk_Z@CW=JTlfw>c86Sa zz#CgY1Kx$Rw!87UY1_+ZV{hMgfBk(i##t9)H{cd17$~FGo+w(a;!VdOoz@1yp z_wNe3&JA_F@(ZznZU4(@K5NgRZ4Y-C=mj|6Jl{`s6EU?hcc}Y>Q+tjD-k9f6+iwRS zF?FI9_d3hx`h6ZfV%DqcYXI8w+yL{sJayU)yP=<aYJ2hQ`p2Jl>W(bl<>vj*@?*4eWW_dW1=yTc-f?6HOM%B9{jKZAb-svmV1<4Nmy z(8C>vJlt6Mx>IcbEY6vjd_T?exSzXTfq`xBBz~7XdpXb99c`b-!41B-=K1~pDL8X) zp&!M+LD$y-@;W)*Z705iPjo}eC#YTZf$;`+*=$`oBv141mY8*W_yOuJi2Io|)`7OY zoaJnJ&*<+t>bs|($woYcx#R>n-nY38-_OuM_uCV&^*q()RsX;yuGq`3fmg(O0Q?19 z!PTDF@{bd5*a_@^+49*f@D+QEl{ZhcnCH5d2j(tv*!FV|XASHNI=~t1fwc)1Vsh@Y zLtn4)djGMVdxN%L0Pi4xaeMj>*vpyv%3bWIfSmX0*Bj?7G52(?xmTc(vz@=gzlPj8 z5Buov$qBIC*T{Ju_i(2{PJ(70zPZlo;1PPcpZhw;_!Icb9{k4mW^1UZ~sd(0S&2Ml zt~)gJ0I5!Ta33P@tvB}uB=!OVyHmFiZ=27~_P7|sR;=G!hi>3PO-IWs_B+Gr?~HAA z@VBV*PR+6A9`>i)nYm-luQL+&4inCuudoN4_F~@J-7&?dh3snTtfseZ0>A&RgIvYg}IcaeZ)uALN)Vt)OVH!t85{W^Um=gb}cqvVxOUBo=wAa30d*oiil*c;!m<#zP8 zkGnQF->YkVVheu9mfx#^`IU>dm$g2#`_F(}1;5^G+uwUc;nyB(m>x^}$r-*y5dtlJ7fy$$gy{a$V zz+OP*pna!q0ryhe>jIpuT%oP8pSw2g6E=R?D0Q6!z2w}vatT@IUiPtP#ypQRli2<= zchNtCr@=ywGmP!=1K;Q2yAsq??>m$0=8{v-i=2v+Be&Jz)eh!vJu05;K;D(P_P{Su z$9itG^Xxz1&8eN?6F5WNa-OgDz)t-W+P>6l*BS5?dkO31t#8=Z#>u@wcYJY;zuo78 zA0lVW-3Pw>o3TB|9wfe~zKMpjr@3bvz=>QVlQyD zAGR^S<~UDo;5QNYoU>-Zn`ce|)$t=RR;?{J(Awso$=5Gg^~kw{v26!?Y`zrO^Xqxg!=0=(miWGZx4;f8a976Qw;k-=;S%lc^4Dwb9_`O>6W!tKyyyW+)VcON z&FAHQ?(IJ2i2z-&o8Q}=dNQ`v?eftaDez8toBl-D&P%Py49d=>n`*#_Bu6`X@lk zIkN>dz;pV(Cj7(qV!t51&952E4|sEhwzUg-M4N)IWUM{eZFhU-+CzIGrd>k5cj~V0 zS$&B+`vyLY<;<6}eh+uKGjkHoGuW>_FSxZ&K6wAE+3t?;d?#Y=WUM7nvVNTLbw7Kx zAO!A`o4~tVR&Jn;_f?mlP~_Y)AAaDUvHNKv@vrb}Z{p6ie$htEH(qxiGV^{sqqXv( z#uw%{;H=kC=VxcZ-#!z2XMsW7+`el42K;*G=hYtDJyi^9A)cUSF?ZX!%Ym_J>OJEH zgj`Yw%&~85AU?2-Eirk0&pWI?!J)6gJ+KG9WB2!`#+lAYV)afFHm;bjIIr%BT6b42 zYplhbZ%qfzS10@eI=;42=MIdVce|kL4%T~5d#;>)Z|Ngf=rgd_0QQp`#!u@R;P0+E zp0ShT9?rN%+plWS0cWhC?d92wnJYKo0xY1(>nAxUe9-Z4!OL}UgKfQi^at?-){Yf^ zo)%)hQ`WTr$Thy7L2FyiJ5RK8mY95~<;b<~&_=8qK@2nyYQ)@FVQc}N+=LSY-&~(Z zqWAPw>I`t^*E!)wuDR$2YJYn5fLD{R?8vS6Cugkk;GO~jXRW=|;uie%X?;gy)^=m? ziuYl@yDaE4yj)@{7kusdJ)$)h`1Op?#yx{OwwU`FS9*~FZudU&LA}pMf2;EuIZwaz7K8I1?HO?Dn%m~S4emmY=dg!6_L$K3%qb6g zf>qng_!D5ruX`u9Ga6cG8?&!+MSj^q9cb!YZQSaE-{H3&+I8P*5UVfOv&}p>pY_;c z%(U-MZ{=y%ZYozXy@lYNyQvxhln<^{The9xx8N1In?!?)K!Z@Vnm=as7& z=F{KWoqWvuN_2yBzFc4J+6T^>sv}l;#0OZw_Pm{a1D061(}UO^`$DV(>m#s1&4Sx@ z+~#z0{CW2Pj{|eJyG~+a2>1p9RGePh&MoIT&F#S5>I`UWe127Xn)TXa#Wt(raQ1Tl zD`1Z;chkngKIB=eZEU9&NCdb&zkYV?0XDH_+q3d=p3i<;owe>E*8&sNS@GSgfv!f} zaU@2}=WP5IP~U*M2V(ZRIn8JGH2eZrHE>Fg`HXvFyCc4HguTQLnCI7zTfooAk~81D zYhd05B)s<#*n=APuznBDpw5GC#z24s=Cqtx@U3rap0n)fJ3p|m(f0E0&2{DtzVU{? z-Hko3KoEDvW%IG+>|sAezXAK%W8?LWCFtOMnsx4T1B|sb=DzL}XnQ2+;NHrcSGn9t zz?-jR@&k6=4bR{iY~$89pgrJ50&Tg|+$Z5ZqqTBD&O*F;F{fz*Ys_1+-M`nc4&S~V zd;9(F#D5Ck+RKSJc4F_*BCzdu!FG=Icg;atbB%56&2sqi9T>ZT>uCEd=#8)ajpck7 ze8>7}{VZ{gVGX`DgB-PUN^(AX&%a<>t8DLH-3y%0&ipQRRbv^TdGu9r;8~DZN+(|o#=jtc%9sVXAznvc3p)(iY9C`QZi4Eiy^c8%>ykjEw zKWl?J`8~R!&G*a;wlQlxLmH#6XS?7(a%X+s#>nZ!oo9Z425--f{TMnxiy1dIu#I0C z!?w3|{WN|1n3rmhF>|zirsmtj_=UV}@9G8jIGi)xy_2WtTVs#2JUx_=GjEH@XAlsZ zRk-Ya_7nJxe_UIM2KX3$fxzv95BQE5B;{Xq#6W-@X-B!<%Ehvl6JQhjy+u zdG~GmG2a}YeJ95=8@IM*IS(xOCFJ@3=#L~ukIKd0QEPn(e(i=(AL9j&FBuu6#2e zZJ({z#s!?RBspuJt$+2D1DEx_wmwxvO#>0Lf9;NK?~T*GS?|PJ=XF8Pfli>*bA_wg zoOhlva~u8w>RfR51^fO!IoXb`ImFy&tbWy+qVF8^2ER^Q<9i`S;1l2GZMDUb=Uye` z+G7A?idx@uY<6I`oUoUE;5*BlW}j6mp76{m32bT>b;4|Vvxyn67+iuae~_EN{=O^q zeSvsM%pJ@zUvZ{={JPlC5!1&j@jjPap9Q&rEqct_wo3J(XA)~M`?S?oJJaW_4&eK7 z2DiXvwbLBz#hhHfE~2XjeP6U=_HpjOR|j0zMq4B7=UqH}U-At&LtXXQ_DH$4X6G*Q zu>yM!XTE*S)$U-+jc6r9EfPhpU-K&;&e#HinDLI?(e^Y}9J|!d2k-;9+mh=(Pm%*? zZWqLy?-?WKobx)bA@y#+Z!`QFIqmgIh@F)y$a)PDF+(X)2vEY@s2%y+k)arlAO zulpePO8zus{5bD4FyPE-x#qxyx%kepSM7)IF0uLzN$dg4J(K$k7})z0TtH)A;OhA} z1KF*4#;i4F?t;HPpTHOT=H2(H^-KfKnugYPzXk0M_j8>W6G=>7-`u)OIP&C<@ z+dMbC`3-$VUh%jCF5m{(_UqW*zkEcidh-@gb;R!LbuWGIqVg$G?5SV-5U+l4_Fqo< z+7-Xa;g`lXCqM`FoatfT%sEqSj&XZ#cX(lb3EUpryR5$IAOf$hXM+8d`M_R&=9*d{ zk<4B>Ybpo6=6M&ytrgC%bH}s;?X1$6YWtbD^^|kJi#?FKTjKKf?+|+dXL1rfjFFSs zrr?)-n!M!5dG^`~-hQ6~_?VjdZpV&K4_R5(vK-D*5#&6H> zv)0L*(?RB}caC-^&-lQO6~2Qd_$_AKoCRAXu)cs&FJJ)e1fH+#nA({4;cI49^?l80?bkF+syDDdH~OWuKS|SXV-FmEe!aMUH7``o&Bi^eCz6t=ryp1oU-PW zpR@}=j-Lnh)-TK>*2uA+d5yTe8dz5NzI1HstVwP6Rsz7?pN((Kcjy9kAmA?K`0uv( z=T7`{dm#^&ocY$*UgTMCep0ugk$VC5ReG-Z_&w%~gi|7BkJ9Yy{{IhU?+J;9v$b;0aF9!50X0WNADS3}~I=V-Q%u2@Doyf+PGHWKeK0#!TQ~V-$2S!NnNU zff+V~8On}s*jcZ8qYf{QD%@d3rVI`$^z*F!?y8c(hFN^ST5J8*-@V`WoO`>vHDiDM z0j4(4Zs$Epu>#nkk+kNp(&0F{IONUvw`1t=p{%zwW4+5@iT z>SezF_b={HJ*^|=*{jyPSo``J3T(CalGtDG$XFhqGiYad2R^43_}b_bdgO|Cv^o3S z_Pyb|)6ZZM*LQ|uUd_}_5UW>W4>0ka7wCGfxW40O>I2<@VqL(QQ$6sVo#?wcXzLVt zQS+EzyJm$Jx9wQuCQ0tU^}~ZKK8D? zu#MY4&@I=`%LhFI+k5Zo_;&B|_O(}HSAX+?GX|VF#?Nx-8RSvZ%sXOub+!z0;H)=x zw+3$W&hkC?ncBmCgS-jscL9mr^^rOkdpOhmFTnq%)tm-r-y1dVdPavlIRW3WC$Pr+ z2l@u&daQl-85=U-^efI@_7L)s#6V(qL0&lL=9 zWY)wVAVC-Qj|K0=dy;F!0zc4$`0iOX`sBHXvTMB4vh&!Dm^<`X{Yw+G&VB>i{)ykv zQybL%MyvZB_%-{;x5of-p+5F7ul9#iA_2310`_cR0%!dQ_p5syF}YzJy>mRf(sRvl zRs(fLW^3Q6SNrZXYRJ5_AbTJCS@)p6 zq2=Ap{@!1ronua5H<2q3t9YNygcEg-)ZKvd63+8ofamD=_ORDRV#|vO6!&Vm^zLtb z$N_ijvErL+@b0ePi0PYSe|K@WiEVs>4hFu@yK=pYmv1|qvG@Jhqk%{?e0yEr3)KUz z{y6~Jya4Vu(3z{pl*d*NbVFAkddWZ86CH;i!84(Uu?`eD>pX{dc_*Iw!uvXGtP=kl zq-XNJfDwx`w|df}IQ%-}p`BH|=xyErihRU3-@8ftgm)if4com+V%2yqKO_6Uhq`0? zS#mFXT!5l|12fj#hTdn=p1JmQ5B)$J|N0)lbzp4ddJpVvU&Z&r{CYp}<#+zBsl-^b zGM6rn+6nK>pss!v9Dak_;tYFv4rdSS6WZ_phIY2Q*{6K8?`6c=Ez#w}f1hJtdpi3L zoM~Q*b%u`iPVG6c<3PP9%J^XmNQ26k@6@#DK(O z%=L`+J8^kgJsbRfuWoX{+9t3)pLgw?$944|*ghlsc5G+ZXQ+WX@`&>eXjdLu+rEkR z=SkB?dMs!g>%em=H*^FBP#ZJHGbH{1PuLUAyHf`+Ch8r)f1;1M#?`gInwsrfXMVuD zXGeQK&)A>97tr7mw4|3Bv8Q!!#6NpFAdhafi4<~)=82ILa`8Q4R=#makJiD7#OYI((+4iYHNsW|qCHNSG$ z(Yzb+3E#oM-*s{gKhUR_71s`I@xv?JpC5JbioQME&3b36`#>`2)x+8LPK-7r685+Iu*In0+RCfD-F+neMK=?@ZfV`?S~A z{v~Mm?jLaO-t`lp2e=b+cJ)9foH2Q|y=Hzl|FGR(-kc}w2k6-5i-DcM_x2fIdmh}# zV-CE#INxVo_AjfixdXU&C#R{2mE@XXjeV;QebOi9GpqApduIDoe|SHSf$vTaZF2N? zzIM%Y7BP2ID>GNkK<`){cA#(c??DJmU|!uhJ-i2H|DC`2$$Xzpi&dw-!R=?^Z2QQY zXTPCMj=F4k&mlK+<=}noIpa7xanEIqdCm`P_cF)cO_0-K#6$zxuuJF(sx&wR25 z-SMBX&)64W{A)jK{SWZQ%}wB0^=pSX-XrfEdlGM;_CVLOh*fJJ?`q(i`-E*j_c&v_ zlXW47ZLV{a9;?nCci;{+Cs)j8wNORmG3+LC^~S0_ZpXcN4r|TxT>E`<{|V>L9ox82eJN0(a=7F#Fm@D+JTl6b?^8}gnz8st^4K@o?7>_r}4`1{Mv9o zgPyBr-66a><-X&!Pv(LkXZM;r2d?%qhSWJ@?QXBQ635pTo+;3Bfj!U--#zuK*TI|b zyD#K#@DonSfcx{~e&EeD=jq5X_A7QH&)6q(u-hfv=<7cjYg1EmYDAw7-xeeRi)@m0} zb${g7dva*=ALMnA*!%g)M>hUg^PN?F@H_ZGzA?E0?3rqw(G#(WeMi?bM0>w-?x>y7 zO607ISh0rf-qjDTVF#QwyN5M<_KcNlzP0|p!=ydn6W<-oH_vmLV@xTX>z)VNeB&qb z5(KvQ5!lAh=nL?S_LDPL&j0sx8am+Hb;Ub&Pkcn}&qg`L9dG*dvQNh1oDujF|1-8Z z#!hI(dTWGp9`L(n5Q_shV$)nTs4+KzXR@#Mvo?J?aed$UkHCKNHOm-i;B0=byWM9U_;M5Eflgjv zTO-%Nz_w;{i67w3tdYd9&AaGhdxnW02kIPc_3Lo^4Av!Rz?{&|)ziBWQ;xn>N38PD z=F0gl4f4F3&!EExT=hZ^^FV?damD%d&oF__T;cAS1cB#jv3kfkqd08Ob^;T-?uwYb zJARgBMN>K*Hup2%s;3s5^S_Y5cR z>>zKV_4ob_+q*EP9_9f3RKK*kd+j^6InMOCbU6Ee0OswU5B!necE1ptr_H@6dk6j$|I{v`7)_ymTaBBB@0_HX_!3qBXw6}C@bBzt`x;y$gN8bE~?e|~< zH*h{|&sjC(8?(mU-N(FNfw|s`aeE1Qf0z5y*W>R{vF)Q|EDW@@6TE>hK>2>2x$e84 zmwgpy*S_?3uC_ZV_H_s4g#HLJkG}LWZr=%<-*Vky$9GnsMH5P{&(|k%{eM^J3GI&Z z8Eck(I`-`f2VefNPY?@Yavy+F&wyDC-}`h=d)apa&r;`sn?T$DUhA1W(*ylvN=3{3N%60#BpZ0Jzy#Jk;GY4_=z59!CVwD4**v_?w z@5(eL;5?i5jyt#I%B?=k8N`*D^e6rR_EwAu>z{zN6@yFc0PZ}r;r5yIwauCEP0-i< zv}dmSvL>kS`0i_N7u0(0DF;0F&j%`g_G2x?49+E znVY~}-{`f^XWtHQ&x}Sg|HHOs5OcnJ*w0w?)u-<>`mTM{T&}Bj;42YpmS+s0I7``g zW}gSKkt=`iVot_-FWkla>WjXRZ=H}UO^&s~xf3+tT%q6Pav{;XUQT~RTQ%6be;nL^ zGp_|ej7Oq>0U?9r`Wz;D=T;y5y@z*OT%h5_o?Dvuz7O+C$T@R{ni&TLcY@oqtgUAR zH;DOs1MbZ_v~lmvn0=h#>?V`|dE66ca-KB>c@RL!Tsv^AcyIDDn<)4R&``LjolgC|$s^E5bnh!Z==o2#Fo zWBWUd&&XJZ8)|x2{OD2p!?#!BaMcS}{)AsO&W-iG?m5h<=aXyBAg2FJEDvt)ROf-S zzJrOa-p{uo<{01gfi2A0_wjWe+{HP>Yj^y>uUu@!n#k3o_Q7xX&JgY}fO$Lqb(THI zF*eZJDe}n2KE1ZOuj0tB9lxuy*O{EjUbic8YMO9Qb)p3`^wrAAxLht2$sZB;^sGE$ z*5AmzgB^oi-xKH{uK2l`iE#yL;JifdSpz=~z9~_Ee%;CZr`-qbdFQpxp;FY=%580&{hX}=!1Fanya7i?93i-Bf>&zUJ+Pg1h8Gja znbU#2^(Uyg#GRpxJnn)%BUYZ=!}xW@%=LS$YT(){Ey+>dVfS3UoxRUearhEy0^9y) z@+bBaxr%%P8N6rYb#%uMbOUO4R=!Z<^Zp3zQM0JI;WuFJ&bP+D2Qjbi>6g_Tuj;I= zyNFejXwP+_pRxM9p$Ghr^mgA4+;t-reL_v|4sRdf z9C0V6U%fa(Vn?o-mtBYNSweoeZx%wvh zK|lX}=NWCE23I+p%h`93v37J{=enmowQCMF>Hu%>p2=BfY|lNwBz8hu(~0dq4PK-g z8L{S>AHdmu-a2*@*R`uVKICIPw|Xwbomsu%n^0oyV~so19&q;7c2+|tI^b$YIk4}- z_#Je)Acj4#FIQ^N6HdM~bF54F1_r!)U#Kznj$ON;jlIF!`z-$pASdBZ_)qY@TQ%FB z@W!1x58K#)yP-oK`!lw`M?TQznXlNh>X>6)#y9Qk6I|p%Q{%s*PjT3LKVvOczkx2u zUDvzfGc)d))YiG%MLx0N?0LqY*w#H^@BYTC25sKv>v_;ic~GP5bJXs!X55c{&dXfq zA8<*0;y3h0?MmL6_|Is?y~KzGdslwqW82falebvEEnvm^(r7gv`IwTi@@-ClJQrJ6egP4(wh{ z`_C8WU61R^9q?64ti#=ay9#HPvh+&JfG4wo;Xnmiby|g3OneEr`Hy7k65Bhxs_RU!NLk;S#vsIn> z?rC0wbEg&y=1$;uVo9z!a=XSoCirAMFy@}m*xp;iPhgIA0QD0w_s~x;LHUX1jN{<_ zndbL|dBQ$-e4lsS1-YJez<1&&Y;zvh#U-}Bxs$lMsef6|v=eKBy05wTAs0N0SL*B! z%zcOd0s9Zo?&Uc=uVP%JI@;Gcfj>cuZ+n`zckvkvV(#vp`8z`3@8=|E4}EL=yJmkM zd4~5cCb?&{@%we%hqljX5UX11%=iDlrvTO{&g`+~4fIGzIBR|&?}6h2BBeCzBpfIE7Bb%*#N_sj9mfAC4}ujsF5aSmMdgb(bI zwJ4nknnXz>;}x8aMpDoSNZVf z3S-asJ684VFo>Pt-Ou+|eFJyUL4c;FK0(z}XT3W+$D9YSzuH{$)IZ_hf%|C(+PhPm zBj>+Y{@36~Z2NU$);gp5$zz+J^7s`u7hiq=@Ba*%z;{NVYe#)hk}KwOHGkiQG5ZWT zayoVcRgXWl-2VB zzCDePK%qUq^X#v$p6Ct&)bpah+FEn(z}Yj_-2IN)vx5YU{CBVGGg|+*a`^TzHgeUU z=@YU##IG5#a`b;B9{BbP{JK9l>n6T2dx^SJ zIQJ3eyn*U@@Y?bn_&n_4&xj_+Swtg8bak%1tTD$O5~w>+szxK%yY9|2B#@)8{jB+7 zpF=X%9o*pR-N5(WKOQ+9-)9kMXZYPFr$6)f?^IropPOy*^{k2ctcQI-;y z^qFIA7x4BfK}Rmr_u_FN;M)eKwSHFQcWhu^;1c~abxDr)6Sh6ItrO*=@5J;yoBOs{ z-%<5m=q7)F9^?`KVh*VDa0X)u22chu-=+Ot$=$E?0rsvwxpwgm0?uh<{)O>%$1|=Kv9>^J$n}&_&c=uX$;-a&-C^84DZQ%_Bs39YX-KtceK5fk^B+- z4{d%)tiC?)dREutcPQ@#+g{cfKLdMtFSP@qhCgDVqdUHL>m9nE`~M884>5Dy(;X5x z)0jQYt-f$~>_)Erj1ROi>#J8o4!``YU7Pe+5X7wTm+|*V*?S`5F6xe0O{U_IZFtUQ#!KpRa~? zr#mt8%{8ZHpl|s0Ebgd@ji-mxb5!>gcnBxxaGIGr`r+B!IJp+)VPJHtk zv4(a(&+U6V0w0)n#~0=~_Y2zE?I$@M-E&uf8wHw})p*^nkCt!zNPCY~asWpJhMKRbvcWG48yH z?&wZG?Lhl1g!RK1yto~;&JHH9jnG2 zl$~QAW9Dg_n+s_urJ{9XxQAM@QwDV=a}>_AQjcQE0dQGKbgZc9)< z!B+_^L0!$ju3fB4IN%KH0{f%&=mFee5Zm`y_A$@zffC~WE@8d3C3)-_c%TD+&JN&t z-^erHz4gs;UV@7}{%3gS2e!Hd+d9u;jc>~)%6Wu^2%6ibr z@!6iS&69J7ng`#sLH+l|(c1q^zB%5-LycJH@0Sx@Aa)_<*?$Y~J$CRr>?gFpH~7zU zgZ#k$JM2&Jk?S2*H)6m4e&r1Ozal5#5|}%|gIf1!aL)LAaPIE!cqPQ{*uEpqp1F4H z@P7k;x{?LwnL66I+J1`9L@_^d?Vw+GU=8>mk3Qzqedp@oJ_#4#4f}jucRTUj#n?p+ zybtq?sRy|ec*bYqH?+Lx>S)C~8M*iqIr8rQ1HAWQOh19~8@QOqbNLH$Y_7fj4f#Q?s2zyg|M%dXp)LPRO<+G!v-7>HGk(JR9O^9C|2z3-eC0>f9$|y=S#Mxz5>pp04B`_4Y8=yuwjC;3|*Q zo@!;#Htzcz*=X`gKvtqptpI(22eM_9JR3vwmo+L z#P9g-cL$Y^vS!4ZrR&{UyZi5Z+UGWQxgPm>+zUG3fA_jDv7H?UGFHBGt$jzH{fp-; z9ec|Nd=TpVE>DhH_pom26F)J_+;-U23oh|Ji|76s)V-t6;z~W(=dh*+Ifq1I0B2Qi z+&Vx0o^St*b$8}iqc5B>@CP$?oqKk;8~P5+v!;RCgL_Eqov&Rx+k;raIQERSS4W>; zkAEM69>fDAym<}Ve%`|c7!!U!&V%dpA8=3jeqL%0H52X`-=7!Ox{LSt1$1C6z>K4} z`br*J8FBP-5A54fb3%9g5V>ld@18+@9x-RMgReM!j0tx<+XoE&Bd>URoI%Ku=REiP zpoX(oi=&S-@h8~xJ+J3XID0$qbX`BoI<{xrymunka`i~+oa1g=>^}sY`%JWV5$42- zP2$c7{0qFh*wg;)bpXjm1F-AN@ZQnWb?HgG0qf52 zJ=adDCdYdB7{H#^$_KnK-@k9!yO^(--$HA}dX!mrKNF4~chImE=hgej8N7!XYmc*i zh$ViYwfCKvGw|=^)Y+-6{j5QZ7muI7yKZpydSK67XG!$t%@5*2zRsb(w#qR+i5XY+ zTzA^frsm{&?#3H-fgYYQ@O!LUd2_te)J8v%Gq5Li)_dIp>|XnA`lT3+e-?RzFd28jjl$T|BghK$*JpnYc|k6zeH zj}<%P&s_0@GtXIh!29$ZO6f#%4$srD70>G|YX{hO=4^k*u4iH&pgqY6VoHy-mo>_d z<|W4U_}PZ9csJJG;H_`K{_p4?*w!R|fZEx9CwzxRLr2mj&!*3SQclz)K#@4e%9o!>L|_vJro|Fr$}fAe4d&42yh{a1hZ zyTAT#|N0OA>`(sWpZw9E{qaBjqd)qy|Kv~p^q;@~twa1;N3 zJN)tg=MI1NPyXY7{{9dDU3cK`p5Cv&v++K9tatW_yYd-yxczKctYkl%J2?S90&~27 z^S$GVZgBN|B@dTq@7@{KsVBI=MdAbN{RW8wfCxB;?ECyK~06FKg=jhu_KZ zJMj!3N1nEKW}WA2V303%o^!0d8%O)kU(GnSyJoI_4{Y}z-*J_X|8#JjToKgY;GA~@ z?x|hRhPY=m9zeUnImg;@X#ECetiS(wBkvBpi^r81Vt#;*?_Phxe*Qh*1N#H`8T@*{ zb9om};0zS^F?Yn$p3j=MD|x`06TZ8gv6H-kEv$Ve_5(FHv^_nG@8SKrVC^K=zc+e- zFYtpN4Q`^%tr~i1Cw!38h-sVe{S~K%GcSnO8T5g++=(0^XMKAG!T;b|qJi_YA4l#C z@|kb`1eFhGTzllI3v|V;hyR1uwTtugm54Q8|0nYN=eQ4O>wZPs>w;EBl&W{2{(Cw6 z;)t8m@e?@r-@JbNyI6B3XjhOR4vp_Q3f$4K2l$uN{!c*P=T!Gcox1Gns<&>?tK)kn z?=0{YcbKuX+?zEZU(knSu4iQrbE>y-cw^t6yAzi;uK{zXI)OFV?$G6czWe!m*q?&( zsY&=@OgrI$zeC=wC4Y;3&-7JWqeQN{fu5jz^nOmYKYlYttvIP;aUI@g`waZd?E7oj z&Pd|+9H0U7PVd)e(7I1o;^2Wjf*f!^>`DFw+g@@%YQxP~^Sv8u%u^y)PW8c8*FMzw zeeCD>0Xp!02KoKvxD#`^KYOS<{)L>H4>$1}y3wb4aIYt7C!94c*1Ul&pLreoO4d#AH83@T07u(-1C?>56;*R@8OQRNWw4F;y>H>^%HI0`Lz}ATytt)_6h7JkP$1E z=wIoVz`D=q2^zdR%*57L;&|^;hx&jrA1#dAf1ra}bvt^*w$Z5#?$mZ*|BAzn{8lfw z@9}I7F{RGKb8xqJudB9W`&k{a_G)N%cpM00VQ8Bc(P|T@!f}YOy{j}#6&tcyuxH=zt#M-O+yp{?<2qdnEAZ;4`=Y^W~j2&?h*1J>$#$2+sI*estx) z8Gi!iIM+QU2rxhkfOt0dvB!yAvG3hp)>kgJwZ;-{EI{2e+!Oc=CR~@lt~t+e?&#hf z?Y#~`&KuhJ1K7*jz&?SmlJJ{Do9{ov`F>Yktlc}>yYkMQ-SEA~zap>h7wyk{cQ>b_ zy+_a9UijVZ{rX*?Q)l0Re+X^R$R8kpVtvHgd7`yTl0Sj*3G%R8V1pAqoj2eE)UK$6 z>j!toK9gtP4odT!+~WdjKlmn)5#+zFnDgY#@pl{TE#8l4i6;Klrnh#)dXDzc9zeZg z{qW@?*V(+ohM%C~@WxAWopXPFDu+3rm_2~!@jJmD_N#s1hBjyitW%!JGbi(y$KK|C zoo5m`;&J3wk67>B`DNXU{U#WoXQKf?BJt@Np|dd29pnMn%2Dr}nwjwK7M$sI{j9LJ z=d9pa5v}9{aqsaND3NQA&-fpK zeKXcstucOwvrogHpyF3tZ2RpUjn~=G9bAr_%ws>aoVyk{fA!sQYUDkGGo0N)3Hl5_ zW4*h9cGf@(@5FOmfIW8YjU0P;UTYQeX0F*iR$phz+sj(xq8&P7efEuS@%e54Ol=ag z&*QN5joYsTZOok#djR`;4(nPhykQ&L{q1312ipD|oM`QYJK={ma1ZmkHoTm6fqnG{ zaEA|I9zeCl*jXJeK;^+D z_QTvfc30y&v!hF5^{P5+5|&sTy7KkSgF9g-kXP*G?$+(u=56?MLNQi5x+jQS??csJ z$p?EcKLggcqZW&Tb)M?M%=%oB=e-YN_V=!;4&M9@N;^-Rc)c6w3z%^3ZY=OC2irY7 zpLgfHOtd)cJABE(1^P_A{Sq|Lf&4)0$D!?|yxD`A`n@4!kOZ;jJ`u0`U}lH2mwJK$ zv^|48tec?4(%L8L)qxK9#BSIfeM3J3zbEVYV5d1?0Po@s&cGht^#IP2v%bL_OW-VH z0mk9m$NdAp0edUv?{l?Sak(Gqp|xqqLd?ViOogdA<}XrP5P z{!Y?C!%yHdec+qt8QksivgUN+aln2`#M;%I_PTc2=N`H0&h$z0l!0yEJ{=qp*Oc3H_`D6RBdfw?}e|yyPA?}<2 z=C)Wf{2l%8fPZ(?=w@7ied2l_;QppBog<(%jO@{kvAb2h6d5 zSCbRy%!NQd%Mo`Ed%C0X7VBwAQ<}-xpv_7*hj#+N16O`vbOp&LDxmo7!_yTM;>sMd6dM`|h!@jOsDd%3mzdicALtEz#ndAjIXmkF6*l$7Y1b@B~ zL;JkE{|ksC*PHqXXqKs3$!wWKgbzJ-Y4}h>+kYDk#ncUewk}_{rxDuz2C_FS^JO&w)5O|BpRHs zt+nTr!@q+#xPdNSKgho}g4BbOLucY?qm6lHdj`He-O+P)eETcf391gx{j|;dDsb*N za_yl#Q;$+dIkLE9(bg!SITgzMme z{r0lBKzqkK-`*EEpP{+d*;_HkXE@Q?PuRws(eMYD*yU4aZ2!>p9sKzP zQ@{=2PAB|9%zoC%S3bJKS)&-MIDU^+=Q*rx!1EY?T-S5nbxBNq#&KrSoe}U87`Im4 zj@~r~-yHb~+^M1M6TqB~ext_EMvm5r=OFX{1pxa zwVnJp>Kodg?ql9Qvonm@-(J?7v4#2e5d&V?^G?JXn7N*73)%pG_p<+u*d1*@`9PoH z2DZNQl@mEFR($Vd&yVuJx=x*UlGyehXnR;IS9=q8pAT>oPMGJO=7u)0L~FZCz&pDo zIATM4ORjs6GuPSH>f5sfZA{)B{qFb(I@}%YPK{W+Ed9glv0JS7(09;#8sv3>eF10e z0Pghw&d5B@ik3Tr2H)}312{|gK3J>x-4NQja?SJIc?Q;cXC2O-H!$IReu=Ie-is1^ z^X2MZ^?8mFtz_Vf6Y++=qwVYLvdJ~(^J;2x%Em2;HPhZ=;?G>1cjMiD0oHgwdxs=% z0ObMym+z1FIoCOA26lg0z09AW0b_fI+Yx{Fy7mq8a^1_oP9XP0p6{o&&nUC;2S6eh z@A>?W>G(HtW?*yvVI7?L2{+Ml9s7&bCGbcGnk5 zEWh|X?qzGj%An4(T)>QlxUPL$vJ0H`f!(%^uKR=QU;=Xnu%>|RuC+hhiT>8NS9t6M z&vJRp!rVR0O?dr|_E~D5;RAohg7Yd4*TDWPm^kJ5A5H6fBc?7jlCNPFtrhL2CzrwI-7I1IOAdt+L=88_f$5Vz3%W2VD7JI z_xC-jpRMFpEiuKMmfzLAEZ>~UBUb(4FZ4+GKo5CfPNIDWI`$j-1a{7eT<_DJ0^aX} zy|3TZp3^>+N4>cX-+TK&d_T7V7sS0&*OYA*zRb3i&a;8+wjJMTJ_WpY{6dc)Znbyyj;j3<~=d9 zfy5rUYJT1vzR&h%T z?dI$b^iRZsyz>Fi;B$QfJ;*tr9pu+NvX^_?V?V2=;EPl%xoY$~82Z%Jnb8repXV~i zUGBgjW zY_+^{A>X;8VLNMr4xH0tY4<7{*Lu%#eqH~3b|dD_2_HbQPY>|RvV$CZ*t5Z_%}H>k zR$H#Ck-hJX)u(nNZ{~{G*Ze^|z~*6>kUNqX+GjA1T4z~TfB%fxvF+`>^jJu^%7eDn z{@#!KU(|5UO>oBFpG$db?dkq;Xmjl!z&_va{z{XEO6={KGFQCY3&@?-@W$+KPdRsVzO#%8XZZJO=E*lW zd!Mm8wtmUW`Z+T3{X4YZfafD;z{y9f8ns*l@+bQu`%F)}w_NZCz0EbA*eQqT2Y%%G z*;E|1{sq4BW9@ps9(Cvm*T6_PFhSre_U?k-Ef+f4oCNaT(TMeVeK6-UPVu^apyT#on|Eg$51$Nt>u z*pJs^Uv~5TH!qI7I`@(9&+z)0iSK^m0qkYZPL27_*zd=M*tR`alGlj0tzWP~0RQgY zcRYz#4|;{R9QpQ|XlnyoakhP{N1}l>p8X8+;CjFV_v>(pt!z$v#EN-e);-bp-FyD7 z8#H`#)gA4bPUIC2r@ik=?a%Quh$DYs+fU9~`4V*P8EdBf_U;W%G57JZp5p~=ea(U& zzex987gf_?A8}>w( zT=_HBeu*9+1UZVkc}D#Qus(r)#pwM--gBbGAg=g5-GUg1{PtZ?{qUV*Oy3&)6ZQ62 zc3j*05q|!y@lJktB?o`t_x$!OZ*Ujl6V90O2G@XlI?I{PQMUl(*B;o$C))ZzCm2B6 zd1~h?&hpI4-o;$S{&B!wqJwAh&FR!OId-Z&@a6Bs1KyakjEThWU;rhfROM8^B%j(l zecZ*|ZVtA&HRJFn+PhZjzp2H31Li3+*U#@nTkjbkz#b#uQcfK-my|aO~M);1M zf#>iYksClAz@783_a!nw~K z4D5#PXnXFud2lzhJ&ddEqg0+}(RTi!o&EhCS`$EVrst@6vG%O|T>a#|t2lBzueRSy zgZ#QX%1eSjuf)-{3p=QtcOqXIdBkGP>W6k8cdpulyHGRXtx-2%zB?I{PkikgQ0#HW zK7sRf#q?WX<4?3bB3Euh?Bn0}$Q%Cw+gZ-(=sF|Xya$}UJBT>?aYpS9oSE8i=Gh~1 zp?X4@yK~HYB4+}3*S~062Wl3)wbuB1-2>k5g}v9Fp+Cs+eNfkXMP5>C&w#T>5Vya- zkNw5#;+<`5z)es+(PwZ6RY$xF_d%TZTk8K7`17dd#hTrdz5?I8+6~`b?pHKDKiU`6 z`vw0Qy8+`P*4-vr8M*p9+rGsI0pK0k*BbeG=s+uVud(=<5MZc}dB%$mIK;XEwwNmt zc<;B@_5JPpWL^hP*rH*3*VQlA{aNP>dj#?Qvs=3WZ*1RDP=5pNCvW|XZ@G1!^izy` zU%z?934re`@5J8jn`rF`v>W+8EBgkpCh)Zf+MffT^&)am1Zr zp8g%!)829e+~7Qyaec9C?NzhjI(8gzSLcn*VLtJM;?x8=gZR|Pt~~5d*c09ydy0KmUwv}wUH|3K$2sle?&4++ zym{8!ujiWiKfJEbn*AEwgvm8%5`ja8fa_ej16!BAAtK9Zz54>?`MFYfxbK3u>E_gFW2>R;8%U%9`;P~0$pdw zrBC<;a{TY6e1ATG-(q)kQ1ev|+kM>ge=$c6j9hTP2Qg=M>^~vLyZP;VIJ9Dadyhb~ zcjS8h`tGT3|JohD^5M#d{|pj8$$!S~XrE02`<`Cc-8`^^SQFU7=i|M|eL_D0&!gXg zKR;Un>;Q`U$i?gW9(DGcbzsK2KhJi;zsbYN8&5(0zO&nPz1L^>27JEedPmOaAn?5x zpNsr}8*uWzOU8t~yszpBAK3N_@^_!KmYCww(I>QR?p%86*nj7t=OyH6Kv1py?7q?XvWYL$2Qly z^IiKJ(2uygHthXASMOYBc(NC8=8anKC&{fl%EZw5xjd{&$Xd_ zcAXsiR9|v;&oeRa&a>Muz#wj)rcd1oJ>k^O3DCfkaiD!sBj$T2>U~#Rc1v}sx zu=kJGwco_{d%?UL+PGQ-Tzfg{FgvjZZ?3VeVe7o@;76_+=LWVllURbcqdw7R@Bkly zIYHd>{)#>H)wy_ceed+`b;7P^jr|k0|DBcdJd3+ufPUrhetg0{1M3Gh@7|M#FW*4I zRW9)acev1o|6u&!_vgU*RYP2DFXugg{nXYxqd$VbdR_j`F|M89blB<(P@MC2_-AxI zBhKMH_}+UL4fq^e@IO%R3fwc?7qsV*uQ%%8vF`{@LsL8Q{G;CyAM7zW%^B zb^;UIy%O8|cGo9hu6_M}2z>E7?5crRd{!e??CA;tyd!gO2aMaZfiwAmFXwmKKL(i# z@}67YxTv|;wTt~GG4H}VIfJ@8^T-Ke6Z?TZ!JYVz-~lf9YT+Ie{J|?W(qq++#5v;r z`-X(Km-C!E@e{3Ix!5(A9Q&URgn6KW-@LB-czRiLiYMgf!n>-^F0nJH-qbW|-hqZ~ zPsRT3Q12}MUlO_&~pK%?x@ZH zce}24q8~ZVyu;bszI8vbX8#>K4eUMtOk7(`ct5jeVD0|AcXHe%h${p8AJRuZK}ml5 zuHB!n7Ay3QTt8Q}AKU=extH7>xW|MG;10%|{|U5QH34tzCy;Q~4sgF>Q|CRX&G#My z{RDRbEmn^K_rN~^?ccqwx`ywW6!S+M_m}H4aekrqFY9yTeOUh^T$3l)dG2vRC-AeR z{KNO_KKFNM#_@9;J%MxWI}Y2OI`y94nTa3hM!bVilizafr*BN$$aDS&U`@~UJFofn zsyT4>F1gYs)*cgWkC{scw*KC`9X&l~+q!7W#Px;|cEA?RA}X$Or8}g4z+?%|Y8&_}TxzfzK^s#V^{#I+zEX zlh{g+)w|vo?DKWSJm<)@XTmq&-gOqvy*c?tZb_`Ua#Lihz4U{=&PZ(gwgZo2j`K?3 z16$tbJmfd7$9!_Ex8^TizujBy#*XGY-*?YGC!pT@bZSMY$#JI!0t|d(){5!_f5r~* zlllc7JOO9=d7aeUt~`VH>+?6Lxxo9GtLJg#ba>|mT3hj6M=rhvHZV5To(HYS2grnH zt}~tCPQFi_nBVW>4ko_!=9R#C24{89!0BbZH|=%p-(vMqC-DI04bXvS-tSkV*M6oQ z{%7Dmlf0T8Yu~VzeHDA?-^uH->d)78HfsXZnaR<2hW@YQKfr)9N9nP8wOn@PW1q>> z_8#1?VHa|}Gl_Nw=T5Xab$`rJdho{v=Ew#6TBr0}_tW5-cA)8Fd~6W&=Zd`s=)}8r zwi3@}{?w1P+l~C1gD>Zq-BV6Gz>it_BEEkJqG#$FIV@8Hvb7ueM5f+caZoc zK^?xcL?pF5x9WAJj<{lNk8hvP8?k@<{^*VE&jjz(+D@+hlYI9|d}E#|;5sq;8MBWw z6?gVMYX^F)xIOLB@F%dYW}zRT^3frPf1UGk+><*cc_7un|A-A>{Sz4Q0Xi}JS028- zFGqZT=FhM2;7-i<-~A@J@8o0k@GPB}eLJw9`rnfCe}esalH2v>$h*Jx2+pYw>?{1| z0(<)$&AFf-U}77u=Z@`n(A#zGZe9nAdz`IZ) zR-K&p<}U8Ad$nBr6TG<(a?Bg}f69(^Pvodi@E7ne;GfX`o>cvQp+<>SZ!BNdt^-bN zsk_9le)O5SZ1s&gwX-`pO(2>0a1)%sSGjUtV9z=)u}*G#zcK><2kaC0KwkAWj%2JE zfo|m4*Zb=S-xB^1`@64sH=wROIG;`RB1av$o`Lt|ybHMjKS5W+nPWfiX<+B!H>768 z%C)z(=3B3Aoqd$rDOS(3bAa^?&c1*9y0p*J8TTWHjIX;@KWh-@d?x1|b@tJ>R{%xd zXLJHH*7vCI@xb=(J_7H?&q!)tSMIJiujX+E^Sm2&iGIWPydC>|B@Ubu<^cCo>@mR&KXdhU#>7@e@COJraw5+e zBbRPl!+))Tb06!hPvHH%0q;`YT=6)#fgboBc+QAcdHU`@@!kImcm}`semyV!$|cV` zQS4juQGTBL3kK2yu)C-@yfNz&3~;C4wSB7XRvdf<^FpgYt8HE9_t)r&~)`ZlW(seC$U?s ze0w-s-yYT`c);C>S?~AcUxNPw1Tp)ZjUm-DFvx#`v%fw4J*em6FZdI@0q@Xf`zzS{ zs;$ph{qgFV)Qr?y=lO{|ck!P&+{;;A9M99;_Ve%B#6E!m8vaNW@GfdMWBA6L-^^2v z_)Q;i9W+8TnF-jwpY!9Qfikta@2pm&Th`#`dqR5b@g&amjexWUZ1=7Wsc9u zn7?}uaC!NbW3Dw1boCB=By!Dof_JWQ>&%(4VruuV`Ru4I+}GR&My?(0XP@)++r0#K zH=meVv9`mvBi3SZ)i+n))62KHllTC}w{F-8>~SXd>43e>$s`8Mk#jdu5=XDZ6Uc== z=FTMl7kZ#Yhf}lzbkG7YZ(_@P_lcIPxOOZq)WEz6JlC0gePP}J=4jX5ul}7 zeD^3GyUua=?v`-&uy2<~GD=m=%vG-z9W?NK^d5naSX<}$u4g+qYu>#-{tk;d?zNv~ z?Zf_q9Q*F?g`dgs^Wk?+17QqYuGj|>|3u8QyMIO_2oQ7aT^#lZNG0meid%2*x(BrL zcAb5j@mPK2%SRP=g%3d96Yx7La@FrVSkKvAg(6?MiOsIsau?zqdjOxiwH@8m z=4mUQ?MW_hfI5La zjnx^@Ef-GMlel7^1ns)=p3t>N9R2V;*X4EfvuD*1w?_x|JrN(|1-{Sh2Fx?x&*|ok zS))IIw!g@=34k#*!fWTn~08pjfi481tl5N%$w z%{1F=Pi^Y=i_E}y+VJlhjAe0$&U1K(wh@T?8~ z+kQRKbN|lkvPP~}ta+}~*iWb(@amKJ2k_WAb^Xp9K#lRqsqb(z*R1a0{T2rw;IBWe z@9UbA9_9wn*2-mS27NRcYsM$Nnw<6-_cN~hPGHXg&a7$%_>G?R9^gB_d9`2e zNMCTCAa?`q_eI0~*sFTL2XLQ**#5cir?u$Sx!#Op&EnlvUG6;AzV=UG-g@Vlv9t|a zz|2HX{sQb1po5cl-(Si72A-eRo&nc@+JRN$9T;D*E;-S_joJbJg8E79Oxgin+waEl zHP_&)7rg6@*g9>GRo}36=ectFiEm!N`rVle&b_qe^_?eTp&8?g2^uvGd-~nb&Rn(D zt5%4Vd{IY@NJre2%NZ=m+bD6-ep46E$R$ClK&Ub0X>g}2J{Z7B&3-=k= z2Y3PVeoo%;Tdx0R3h&Fat^VX9)?UZTyPs$ES@?G}pU(@pMHAGJ@c`+tMG0Q)97&!hc}ZCKyivyXT58~lX#J<^IdeD@7O z?CQZE;0^tQ{0q3RI%gV*1sC$Z`?ZKCvCq8TANI2*V)Yx?30%eH%AY%OZ~OIF`#q`k z8MxFqhdtD3C-&@<_{PFLJzw>r&U>3d9oF}&qp7}V?D^Z~)ck@!fVD~7&}!}5uzv4e zKfw_@Q-@P?Lm&8$UHgW0jpPsD-W@w))dv352l}b>^0HHfw*~d)fO5 z8c6Un=s=C<^o;L7e^9U1zW?YGHE@A0;LmyV1K!wk*{6eA*6!H`ocx)wCkpbh`9?gz zcc0eG#sO*teRH8>tEUO}b>0C^U+Y}P*3_QNF!0~dIctlBa|UDYzmKUd*IDW_g!L|JCN!m`ueIs}B=`-@|NUuw9?zb`e6;GPPMkUXTzBFTtKMg0 z-nozFg?|Iy%@s_vUF6bg+xADy%UN@|v)`w+#}8;K z58nlSy~CbsCVyVM5x-+=KX~=}4LgA~*51$rb^`ZEtaqdStPyDF*AoVq|8l=O4sEc9 zI(wt>bNq)-jpFrgP3$||%=NuFh_A%2tx?xEXT3ci_w~=#{X48HbJfddP>=OqI(>cz zjrDCmBZ-A|;`D9buYa#K(fkVJ-#*s<4>|mIFoHVP$hGg6yCcrc zxo`Qg^~CPw{3mDs-@%(X&_RIekJkA0L5}6~m{D%s$2+;1@B+ThjsJxA9k_yocV1Uy zYya1$Rny4%bJ~4#=j~rB7WW4K8CYlk#0J)x>fIo2z|YbJSo4N; zY2U$j==C!ovHyzLIbI;-bMfwTA#TKKy|>yM-u`ND;05n_8=C)vUTV@9f5Khx{kx5^ zpSysorJu`nPvvP!j(NAJ)p42?VZGgHDtnZERzMU<>G!Jg@*17B@I%q)6 z6|MJkxSEAI)QKlKQ5tyb2B>oo);#c;>)mNs>*U>g;v1hD`GH2LJ$;Dk{|@ONzF+_T zq_L;Cf*woX6OLsZz_{wM>YhM84&T;io`+dytUm55?|07UjfVEy z)dMZL)NcIEr}clo!Ovg+Kn)_-nVoT>o5Wh*U(e+Xa_%*--kb3SYY*QQ`>1oBXW@6| z<9qHN>m2HQt|LlT_wLyvS3LB|CFpSS+PXvPufSUKTKj1OG~uGh1D(tJseP>j&U4$a zr~NX}e60V@+xNn=x#xtpPsiTm*vs<RSNvW#MP8|96!>)4v1%4*YMx`}8~g3KEUAo+Dz}ZY+r3@2gK^ z4dMpOxvw?G3ADmq4S4qIfo8xvSLJYB<;OqlX^pdGayRHd4gYr# z_tmf7Gkljb_-yN)z}HUf6TP{JZPDnT9*HlE8@46?qvr2Fwf6791z!wvpklQw_sAe; z46b10!tH~3zF&1d17}gM#++wS8^D~9djOxKIcGO+_yNqjN*na!9t-pxkEi z`~@3e$b$|-OYW%#-`}5q!>>CG`oIbA+deL}axTwe-`Ab}V)eT2xv%if?JN)SF7=V@ zA4_X(C!PY%zWM{WUj+M6`*)xAec1=@oVmd}+h3YD$Lhq9tDkda?ET=*yA^QG*5LX^ z9viE_JzOK#`LtbdU+>oLeJ<=VW9>7yhHRa+!>e!Dq0jj}PXONmYtLDNy!h#(t+~*e zGnQ|O{?qzCc)`{8f;{U+{J5|0i_iGZVQ=d$#AgqCbhwnmcaFq9h^-mm)IH&X*mKsL zfzNuQ#=U*c_8J>4m+qbXAm8BaZ(QHGaKFj3JiB@8?OppMF=uwRgrCG?!};X`H1z7+ z&)MY6eaYkZT)etT+~nvnwk>Y=k#~7dSK!$)*B(7c;Mc;IRtwbLjy86#eVzMxCy)nP z`#jNyKI`loxirRa%G332ku#C!a7gg2^#$lH-L9^18=|{b!IdZT!@8xsk5JZ z1V~`cnkTSV0yW}4`JNoscgvpYyI^nENc>pOdTw9NegpoBwceW7ea&@V#+qdqB*nQNz<9p7}Ac%j% zCU!=H04}&5i&Nt=9*G2NUw{Akd*IAnz$p=lq^4$$O55O>}SI0#cnh zeCuD}3aoKOu6_KD?fCYw$I)5uEDiX5Hpul4VqlNh@zjd_Y~9=EQ2Xb4C+pdvsr|`m z?JIAe2eI=se7PHO^|dE_?Tb6%>*vL`fgRLXS2}9@9S|3v%RbR{eD~0*HE+M4caIBo z3FrPD&K0rl@=4CS_!E4+WAOHB*n#yvJom)+cYV>#!F4$M)mia9>v_J{``Xi9Jj)Bzo@mc= z3iSfcUgqs(Y|XFWzxqVX9onbvl3dlHm9uUF&ycY-=&!`y!OwR%dEz%>_b9+cu4nKg zIp_1d_As_y3}ROVIbvse2aVYJ1g_ZJNcw|zUvt@`3F6w9*|aU!ep-8rzM=2=g}tgD zwscrm_0Xa3jD2tRWsO|(_>2mV=C`UvxFqkOm<@C@!H)Y;!2&yV%*Nhj84>iN!0 ziAH}0dUMzTtnrStXVqT0X7?G6``Xuf?@i8rEdbU&39Ji|Yo38MFK5mf>#Um_aGu4z z?b+egsdpVKCtN<85#=ISbKIzn?VB$xea%nck_+J3-@p|(OL5Th_HhPhu370-I^rQ` z@B7M6Vr!f`W9>7s4H%mXElb||01Z6M>9hX&`;{AT)>vEbf9@D7=iUwL&(?Qz>L)o7 z_??{j05vE2f$uXD`W@ZKmG`X9p#FvLcZpCZm*gha=abmq$^Ac{*3Uorv2&;2b=TK- z31u_c=g)gzebCT6=%N3_zugz?mpgARiJjT+hCwXqSy9u-1@JlioA1}}j^BM+&(t1+ zToWw)K$pPtK7cE*=Z@SjV$Jaj{7-7`0NxsB{#omaRaZXTufTix4C?uiNHo@#mh)f8 zO=DMp`#PKYK`fjxu$c?iCvo-E(#IagPqgp!v~Tg`ol)PK7rtju|G2Nu_LcbP>nxx4 zeW6$5U0dt^Pqgam9@^LCvj{W;bl@CUFo5%R!3@6>`@5-_=<1y0?0x%aHEYO&nU8<{ zAOGS1{TKiB|K8rnfXBc7kG|vgp+6A)2kSp=|Dyfv|MK7ekN?yE`ak^Rzx&&N`?vr2 zU;fSC{PTbEFaO!U_$UAL_D}!$fA(K||IgZg`~QD^J=2JFXU_NoaCYI{{Pm}Go*Opd zBGGuuOL)lpk@l>CF3diU<%==8tU>QD0<#`V51L*(M+f<1jl)J=Kz z40%E9`{x<2YKY|%tM45eyWa3;l8dkN58_Vj`2yeXK>JwdUD&(ryyv%KkDu-?CW402)ub$^0$PxET)K4R@zJ@Km#w&mjMb3mMOuGlkCZ?5hI zuI??#gVST+&u?ODtvM_1w1dlyfpw#AUcWwz%CsS zM~8K9S0uHdY3X5{`q92DYdzyHKx^-xK^M$gz0i0^Z{VEKKK3&bKeQ-9Qu3>NFJAPUEjqR(M zpo7xp(uMIIf6rn($puI-Km*=`x@o-kz*{RYNe}jG_@_p{X2#Y&q+QOY-qmw`W(VAl zAc)ObHv$VLXxom&2SM!2T9?+^3DiF5bs1UwjU)@dJTok6{yL1%b^37q+bRd@MVdk<^}a;`I% zadBAp2%!Cy`oE%&pUF4soW<|E`fMT@XsQRkTqbtohE6=yqnq%7Rii(^>Ej(57suIN z#KQX8BbN5uS7V)gQWu}r?@|72sQModxI~HI-eD}nw|KnhI7QpX+i1H z+C#qqXRQ^hW?);A0t4LxxPQ|Eb7>CLIk0NWC*So0jv6)fStCy1yDis%eQPgxXT4d= zY-$2-ZeDJpPtfGRw{{Sl51<{Og9Pe?8f~3JH`hD~*YJ(Ct{Ke63=MSrnQJcV>rRQi z3;6-}R0oDJ@LUQ1MUOv0P5dW{6#Dj6cD#zp~eF;IBTx<=VgQk{)H5b;4WY-0J@l?K{>Lu`qJi z61zN8B=6yUJ^vm{*Wv8r8MN-`SZDYyRY9(60WpuY3&v2u=doy zq18XD<{|c;S}u*c1m*^I#Qr>B>m7ts^RR{$0iQbuXa9PS@x3eSCwceQ{~0bo$2TrZ zUO$ZW$$N*7JJw|!%er^s+jD>`IcJ@6!2Am?fX}264|#CzSdH-cb$xW5y!vym_BCHU z@vYaFAKP!NkqZr5xouml=TLe4POm3u8}6yzdzsXzYs9V|`~fEMQx5-vbv|b@&cpz< zRg0a(6YCz8$G88Bx&Y>^5d&X%UYDHrb2<80_ivvG+*hsfxo>Oi;m=)Xch&)?rp4mj z+j*uOHsL!s{+0X_K7hRju{}C4S9{?*ml|tp4(x-xX!zFrfS=gFT3bp#)wgu?yU^P`2blQoXaCYuqt5$Km&7i6t1;(NtFL{? zC!n`=?x{Wk7i*u&t0!*oS8LEuW0txA4RkPp@srw`uPtkxcVlf2NHq2^Uog|+!NT#MnPiRQlY&UV|rzXLdXqxQLNLo0WB z)vU(sCGTF&rFD%UmpE9@cLkHUW6wV7?UPz`)&xlXT=i-^pYy&p8r+LM&a7U|1Q%j! ztapK^-pTJit(rl;X2uRQ)0mzS%Xcq(CGj{GjlI3g2e@t+ zpS$usam`rsKJM#1u&*Cw5=8?sIxL$2<8!&NpYR9P`^;Q@-{9<{rq1A8 zAM0HlSntVsegdaysL-ca(50;~V^lrFUj$a=E|tY7<|be#e&|Sbg~b_IgmO#@KtfVm||O7h=!hJy>%A zepc1$KlSZ*-+w)$7>cRk^qz3wBnkF~dZRxa19asg-Wi9IvRSG_uTXH0$I8nN`& zyisdUZ6{V=^Cr9;nE16nR-I??J+M!}l_hq0H?==AJoW#$&gyyX-+=nk<@(G7vCDnL z1?}Yn9lraU4>;?}hj-=zW(I$tb1qk4?QczwbtbK|`dmJPj&JNcYrSz<=5^Km;b$z~ zp7yxyT4${LCw4D!wAu%J!@5TUW5XJk=W`bAnSBJ(T>HpRe7Rro-*)UiH$nV|PkSG1 zp!NKI_MEj*=NTIO+kL&G=6$wjzm9()r^dUy0{2t*8K|?b^#iLvW6h;+|0GWQE832A zF8zjY-kxe7SohLSV*7l#3%GON?r*JqF5n#m+}PII0s4*~SZ5C{=&|$-J3-GC%e%+G zx-wT>eyn>`BY(9Q%l*&H#u@aw@BYM*YY%JGdbdyV?sLPR=;iBd_zhjtvg+(;N8-nx zOCMPC8MwE#p3mBdRTuc4)%9hbSf5Gve3A>`IZl7)vffI>h2IXQDLH0A=|d$7j48^rcfug;~W5bHb>TW3aZ zt$nW4Omr^mGgkc*-<(u%*R5;1jFPdwRp_Ux2(^ zCKfbserCPY-B<62K5*ZkvFhwUuoHXo>gA2CSEt53Zp0#Xtj5}lI#y5q2rRLCC9!p{ z`|3A|)BK;;9Oms2a-g9bZ~@Hedv<&2s~PxjC#Ls=lNYDX{_oVu&4gEP9(yzie?v3j z6X*+T`_>IL`1Wb|ffX-wlh||n4AqaFD{WhS1C8fz} zS&)0-+q>$C-NT=Q?xC;V-z}_he*0E0V*mZ9`+MewRcoK5u3_t)A*bFs>}$_3hu&Is z^{!ym7`yke?m6LV4`_2q?3rpF`qw__dVIY*InU-@JhApu?;Zo+*d8O+JU4uE_P17z z=jhlV=lLgb>9F>l=J4$s_zkG(;DO#ceRHDr-&_(8@EP62W~_Xj$ryb1sZscK#?Xha zg;gULTKf9$st0k+k9Ac)&eLP@TF=?A18WccX^d|l|63dD23&``p!2_tJ)gOOCU(wV z6OC4F=m%hBuD(}t-rrMgkXB27xw7~5JB>9hS0CMpUq2IL`Sy0-s^?7hbB%z@RhQVd zeYv?~*e`qf&!MYNtnVyyYFnTYo}Jixl$R@u@6s1T1lxf)fYbNR8ouvXBev(8KAdse zvwPqAp2@nzp5NuQo%o8K3CIiUBeATy!^DX-XV+(C@<+gNhFXa)cHCpdYMT3C$Klv> zCOojulemEg7{u17mGkpp?Uy^_oRj#qdGil__U*aq?P*QJstfD|D?%Qqzkck`Bk_>W zbrFJJ3G%xZc<711`{YxCyLzsym?X3*UF@0o1ym`vkV;#JBeqG`N{)^nv@SZ*UWv zz(z29_Hk<&e5Wu8-OP%rx0(rRVr%Sm&@#e}JaV#RtB9pVSERDTm+n6ARR< zZCEiuY-^1@L@i_X^qmpv?Iq{3E`iqj(>GWBgBZB1X}M7S(P-68;HtX(I!|%LrN^G$ z>IZS`oP7CMb1%;8y&bN>i=mCL`w)GZq360UxzZ6^6M+TJP(0z6`8k6**OmSJJB@K9 zmUZq(+Y=%HYvzveN36KP4dBW^3e>yTx2yPEd&IWp^=oZ=>XNtz8D{G{M^ya>aeOd4 zxR4j;Y-j%rIOexcfI-{@Yi606+g@L4>8HlJiEX*gR5Rh%eRwW(5m>kcCx3i(t`YnJ z)cUy>J>l-le(48guZp#?`|1n!tUctxiN}t^e&1+}h5NYNza1o6Am@Iu{iBf^aJ94_ zOJi)ED|7kA`VCY)u71smt2IC5HlMNXwqwl)ym@V%LCXvU@-FkP%!P?>CS%J4(>OAm2b)d++_fXakPEl}GA(mz$f^b+LQj$kp!v=;U2>KI$XZ z_uLyAb)EQ}!#;H`t#Bp(1Ab3c9)AF9f_mXBeQVTxJEwg0cQ$-$W5ZZDlV@?M&0Ly_ zw=ee(c>0w#P?7U|_NjC3TsRAU+q^ng&67JPvHhyY4`VQaIb-d?z;|Btb6a~Fmj+XN zzzsMN`E|dV(U`n&m5p-i4{M6ce`ve$AHm6EC!G6MPM>qjg}z{wp64_`Mp5;L8e^^d zY0Jk>dGvuz;s(?VteRTee!1uPM!ytQ+traP%Y1bP&$0Ksvd?qRcJ{cUb1!QGG%y0Y zp&!7!%evR@)3CyKQ`=*m^NOF_m#cc`08esa!auA5>dlpo{?1S{qiN&@w(IBO>mKx} zk^B6C2E2fM)d%owrNMXE(>N0w5adgD#}~xrYM=Nzcj8ZA&2MOa0Y3rzKfw+3Y_;qc z`1To~le4e+ik$)2=QFUrRk2(*&>u5b+`3VX1-M7U)1Nq{y z?osn%&0AaFGsX$9)eGzDx#GG*<~V;Zt2owkw8ow_hx%{)#yUOpUH<*i`}2+>cU=$z zXR+6}P**+G;CmnTbjhpDT)2>{=X&4o7ey9Hkd^c<5GU@|(uA8|H zd(>IsweH`*Ilnb>=Q)%mK4)v>Pmi{j_xc$;f#(xb-RX}mW1Z=GC)jYVu8x`1dp|c& zwZvEale6AAU1$G6z2^@wk_4E61?mr+{p?xiWIy*ZABoS>?}A)tncaSs$JUBvUeuoV z;2mGcHS(vAe$AioLXb0TDpRszpZQpweIr1Ir{((R68*u*yDxb@r@N$V2eQQFW z8LW3c_th75c69bhV{+D=n&t`s|1gI?6N_)X{saAhclJhXk2DV_ujWQB@vm6>s1xSv zZs7X%2mF=T+01FJ*Ltt|i6+q7Z{oYBR^FZsUL^iV`hc3Bet}>Aev6#Gdss6;5;tt! z*RC%O_h&zK&f&iD&eGu?*s7zKJ>p~cAlHbiAHH+CpSAYtSm#Q7dwFj6nef&e`+|D| z>Mq!yKi22TpTBAb@N?=66WAlLkt?tLk$lH`hfiRC&vEqU@5gN~a~T95tBQpE+wrtoK&`1v;^Fy}-y7dmm?i`(26Oz&o%$fit_0d)jN_ySzv9B1Ep5 zj&DuV0&`W9%((l?yO%WuVrLszWBtTh>phIkIqw4)C%K089=n!#>|4*88khPO>rVWB zaQ-m2d3(%UHSQDma_(>45Q)!~Q_nhg;EK` z)ayHo^=}}lafXC{!FQ}Mwy!nrckUtJLmqqT^gZ*0SDV-&XC3wWzQ0;$vzM`ROt=o- ziOt1^y#~I^x=5V+`fPhlZZYGk0%%HFeHdJsxn* z{~7ChU++G-)5E^gIQbt|y_|R8S*(5HAN>_yT?YfGch-j*R>b;Do!ZEK_xlcSXwR~* zvv=$a_QTr$%&*QlT;7R$y@3{c|HK7PIc&zd$Mc+>YruJidWL9zg{v9h+kMp!xJj;q zGoP`mV)7T@KCke{oT{t zd3PrM(GPre-l0ps&h^LD!})7Ab9)}GJsODodY03ZOszTx`&+Lr;%kr6!lfMfzI}T- z=g^AW7Blf@tX_#9_%+k^1M7QhojF$q8sY5jqfHVlt?>vfIqOAf;F~;1xF%xV8{X=X zU$s?7zQZ{~;7?;PwDyzFbq;-V@)It$fBH1y0iRg?E|^isyYyRbYfLZeT{DSe*Awg8 z$9z3&WBlTA%eOy0#^x)BZ%!^j?HR4k!k@KzV!*Lu`-vUVYRS1GR-cBgJAs?{=RNcD z04BPtqfI4elSow1=%|)ER zy!T;W_XuL|QLcF7Tt7RW@GkeU-aAlJKJj(?buVZJde3x$cb|^mweb2s!awk@pn)J) zy)YBseG>ad&fgVYz<&PjZJjlBX6A12O; z>)hK~&t{EtxGJWuv{-v5kk7!H0|Vch+Xoyox@R|!FCX9d_E2waVkdFc;0M0?hTnl& zefgTjkoc^LZ>@Q?rQ5!B9ZY8O{k{U{a`jwy{33oj2O8o3dk*^~uIQxnecq}bF`O%MY1*gtyR~)BQ9OPR{JqERzO`^EpRt2l_p2P7{3Ld6dzkl`HF#$kT9z8QIuq7a z&l$}*XW-AEXTkyZut(rKr}`UKjogJ?)xy`Fa3^0mYq$q{I9pJ6!!~pY?5S;7^&@r+ z<$2`g^t-@!pNT!FzPZoMUTP9FaOO3>P%nN1?m6J5{C(A4;UB=yll$JZEOqYf=k;Cg zzPOS2Ebq*vM(zi!y&@p74SV*eJG$?)VeQq)JI{CY4{Y5Rb?08~@dEZ6prM&qds?q` z?km>3Gc>Gq=Ij&3ENAVxH1>2JYfewC`3X0$RY#4h{8(qZQJcBAvS=ppW%q3O{@gM) z@2Yu;?JW`v5Sx44mqz;&sPi-XO`kc>-PvaVdFw~6eSZH~HK`>RaCYnG4o*@I>46**~z}X~xn!xA^vcuf$I=DME&=c+R$7QJ&k(0gyr_5#+c z?LeErobv?!nWf=7&rcugbJ2~>a}U}#YmAxQ6}jGJZN|PMYW+R*hIYcKcZPv|$mQY! zoqe4n$r)dP&s}YR2F`5G)rsA!feUqI(YvfI4YS!RbM4jeJHFZs>^Dn|{a?Fw5MS_r zu_o4D6V4U6&g2=KS$+lytluHl4SeTs&rQF*J0JWw|&$G==j#UduM795b0uV{-PXx>)~A$DRWOaE@B|@^V%G z$E7i6FQ38C#?nlD??h|PeT>apUs^fx)f4_`6TkM|`FHG3z*;r-`7#5s+B4@9-#fF% zz*^(^E^>*_GAGQr|Lp@GsPUg^+53fkfYPC_IN@8?djl3hNtj-3c1!wC=>!f9D6*dV5ZAMJLyR zcjW4^zF#i(z^_Hl8t+41-~QHQuGs@?FY_;WYt;pqn=hZ(nwlNo`RZ(&YjXHc`gj+v zZ_i+z%O0N78D^}#YbGe~+`T&RjLv-n5Bvrym%GOO$qn_K$^I`iKAW?Wocm3zeVlIy zVtM^0V&&Y^d$52bq9 zvj@I8_iyMX`Wybk82$=;PtAFc*1bRwCm6sx@eKBNy^*(%J)S^qz3X!1T&I5K@;ygA zW2~HeJHNeZR(xy31}?$m;1bc=hAOz6Sy*Wk68YV{J{6Dr6I3g1dt!e ze8yVf9Zh0Uz8ty^@9eHKM@xcOb%_n~Pjvd`swa0hh~3LrE95FhUuV(oobeNXqVFJq zYX&jTL^SecakXmg>3!7-c=nFOf*1N1xd#Yx4QTCctyWl5&l|si2X)mah(!|b{ii-q zqaUD&ShM=v264)PGq2|?AI^O38H-bE-t*bl71)g3-x$B=s*^wPeVX#t!51(t^)Aov zv$RI-06(H1#AiOW&Ts`Cz7d;0pSzlu9&f<=wudXzKlUE6Yqi|z;U}^8lAFKiCmOMz zSKG^+od4{i?lM;I*t|V%A6uMj#s(Mo=3MqQ_AZMHNa6t!)E!4F&O4Vgr|;hSTK9Jz z`!_8(&rogT*S*ND9-+rVJ7U?6wU>PY)ETkH-no0(ckG$&)*CFg$Ttv@;YsFgo+ znHxY```CBKP0L;_HZU{Seb`4oW9iinP?O8hRwffPg7SIyf~c2hI$Q zW!=cH??>YMPS$?pd>{R6%h|(T_7K)tFF)}EsJDN|cR52CW35ThfHl=K@u@L4Htb{l zjXG;5zIyqIZF1UNKX34=zV?A@vF0%EIlMo8Ykt-$DvjpZk>zdZ+>XmEY| z1K(%h^1dE-v_X&nTl=uD^?sK-gL`)Ja%vJ^y|MZ?tmkv7Yxwq^feaJ}t#!5zr+uMu z`8&WBL@s_{?de`}9pBntu-;AL4|4)vtG0ug>wGVC?jc;}N|yv1Z}+A5j2%wx6Ug=gMZ;yuRp^P%Ng4%zyrN& z#G1#MDwk{KFY^!|eMfutaL?*z9=~RBFEoW*{RY;V(|G$cI0Kx^o@%-=edH%`lcV1a z&iRiR=(s(Dn3lDR#wALoRtM-m^>plnm z8`_ay`xx8D{-S*1j$eCGuI<#i?0+FId}lq2=eFllAMFL+8N?Mo^vO?PAGHDO(TL5} z9-etG^9`tXd6ufhjwGhm86N^)t@C$$dp-5Br_XissSZD4zek&QX3tTe&e>kT`4g>m zwPMxwTz#C$Ifgv?*n0c9cjf8Zi1nKsvHi=!H+*aDlW6SqC;W+ZAN%MJ>=VADy<$Tj z{`9m?-g`)V&)@K^Z2~{AnM*V9g?jfluP>+0{T}cyI3eG}_Iu)ks@1~RbM$=p8RS{^ znL!TT8v7VO&}dT*H2ikoXT(=;F7>g-=3VkV^cEjcZuL-azX?i@4RE7Y{>i@b*67#F zf5egtc%QrXZ+-yJGO?~YLo98}J=?Ce-uxB4b+vMtz*ZiueT|Q1;9uF(InKSAPtc>) ztS@>xe-hi@9&KY_XD&GZ1q2#xhi~}S*;Bvj@CSHojs5E$;F^Fp_kg$0jP(xKN8Lmd z_-Z|~zI+mUhGvX!Jc#SN5KdhJ^V-N&cY6AJQ-^Qx?yWWVGyZL3$i=$9iS^6_>&%H2 zJy&erbtTtgJ=abjzc-(t-YHV<)yW%M^SZCuNEK(lKqEX~C;tQi>|OWBoI&jX&fPv1 z%-c7En01wBshdd+X!@Xok$X@2b$oNVQCc`>7;xrp^bf2^;4IoqH0Tn%feGhb4y>p; zw4On~<+_t6TGvw!L~P`$H@?8z%Q?!2YxwpMJ-+S*Lj!1mW&l559X!DeJiv8Boy$9O z?zZzIm$qX=#KwKi;H(KhK_}OQ^(^<%x2JQvhX`P-);U_PGvQSqZE2`e@9(bzJb>?{ z?~!%(^$t%z<12BeM(Z+f{}${1>)zP=@<9x*&e`peTNA{tnlW*}XY?=d2Glp|+|N4E zf%Vop=L^U;;2tM;{eWYQSpD9K{hWT=ue9Xl+$-=q*8B7smL6N@$EtH#H$l}Wb@JFJ z_E&Vj0P9`m&Rw)vy#`kOc~)w?d$}9dK0Z$|;c~;e3!L+Gd{<Qx-Ed3CM_*8PRP zJ?weIzrr=3pYY;>@A3{jzvpm1R|LKDtLBZ|51@LZ`3--7-$4RrGiQJI^t^R1_~x(R z0eY-|Pd2bYJht}jzV>oo>pzog+gD?465Drb;Trgfy2RSc{2+D)_Y3%^7GBQz>-}qM zj|*}81jyS>D|2~)VfcqOa zbk5|m&apt%3a-Rgh6Hlx&3Ts{VaaMCTc>7#{ z`~Hk=@aBDgI(~w8{D@XL{;d0WLw^CzVvRa`{{ja7mleyCb9VRY*dKxW+r#&Cl25FA zCp8az?ZNBb7=r(OqtW+{2KesNdNwcf%+%o;zWbVY-z#w^4r6@hO=5lbe-gWH#BzN}ZhN^gfL{`~&|DsA=Q}kgr+b>%NWk%`soC#0O2dV)f62!k?hDT3}wj z^hy4{^zu`{x$4w1#A z?!4bmJyuU^)CBgm^HoDWH`>Z}tZR&UMpIG~moz`vl1id#bxy>$-n^R*}WjIqKX&A7IaSdxLY1 zvhYdXUTV}f>{A`kzJRr#!40?q>l`C0Tox{KadmF|2b@2fg!|dMi=fwi*Aovp`UO^) zpTPdY{jOl<%B7q>J&)#{e)U;8ebsE8+9p`m82^q=bTEluA8Vf$Uwc-KJe+>r8TEho zo*4XOJ#%++8x8#vB(ZZFkJ#@O++R}X&mw;(7;4BB@avA+ng{cBUvdHFhCO>MeTQrK zr^moIcJ)}i)|sq-!#1oqwHa$h=e+>$LhS=9)FlX@e}fxXbv4V@S?@kqpkD3!=ByFs zAHchN0pC;icSilo$JhD>Cj!h|Hn3{i*2(MVw(k9AK330q7x7&^%TzD*f4Z+3d>8E# z)Y-e<5iQ(#ubelCU0Q3-Typ1~zrZJX_it*z&}W@lZXfIJv}Z5(ow?%2=1-rW@9VxE z*y>T;_=DJGZM~bZ_DXEcfz|Jz<;r!e_ocq}Gp3I{0^hyW8jo$?@jEalZ|xh_KF-u) zX>a-oSe|<(4w@IR&Y%5f?@rFz5v%sp-OxC%=S(!}0?5~#)VPnon>dp_?V+ZDD>?IT zpabKJJ~P+);2ktZYfZ-5-+tAHSWehaT>|%c>{>s&4V=C&b8@V}cio7^jNJ?V9oWOY zyoZ@s^6K2joPB*SYhQ8${|bWq<%2w2Y)-BHizjDY?D&P>LCbZ%iS-T=D|bbI?)Vpc z^YUt*#B~nr;mqHN-#|UbSej1k3;X9bbS8RiG2G8>#Hh>X#+j`_^e;|)zi8FTA!6OR6Sakb@p@*{hAZoh}~m?2XSJb=7Dq6J-`LN zT6IkyyaQ_<_cfFE&nJEdp2^wKkh3wZ(o&!cgoW9(|ANcS1zJrbU@<9(U@M~7`6LexZ zZSAxB+spUcoO9JP$6i70L%w>Vw>E+L-$CHNVMT({VIROfU2^uCNxeM+{e8`##{JA! zEgHE28h!`bC;8ld;Gg?%v3zyzsV44gt|Sif58{UJibNkc#6G7wE1`X^n`^oHR19a2 zz)pDgbAv2dF3%YfYahQ4&okA!)C9OuW52fJfvxkyPyF1mXK_#asqI+jd7*Re zTINrTy&31?>JDN(Q@MffEcTz3mAkJQYxY?C zG_3mXKJEJ>R$KQ1Uv)d~*g6B&`3HGtNn(Fin-91vzW3-p=8V;MazQ+Sd;^tB{BM?L z^8Iz$PpxMRY$NX5PW-xSd&4=ix&d$BAxGS>h1_~<2XbodQ$De_gSl_PIeVP`i5sg~&4SyJe$_KtHR&DV~4oq;P?gi9# zFyP$p%x8So2)z4>toQ68;ME6Q+rD+7me@MsvQFKguHLKodgmSAoTz&vc14s~&jMea zXu0;SS&7vRIqsq0OmK7O+FtfgxG(1($}SDEAJ=4Anz;#%?)c`*G%e|ufYlTbw7Qzt}DJeYkI&F z2WZCoEbUIravy7N#A+k4;9*|QIF`=7FJKSPQG1iWZoivjwu#={B&SvXG=_J{H{eXZ zdtr{)TJQ50V2vwd=?1p;#yVGEe@6EJPvAcGv`$~WRy=_HFCVK8W!3qAAH$hjth%mc z2G@-Bdzn3}FL40-yiw!c3G6xGwYU4KJG!>@gPMn0@*Q8j@ZOU6Mox{s^Vr9n^}Z9X zs%5r_m6z-KJ72LroVDgeB$hh;;yoX}sF|r5_}*J$132p&`0mqs7LDb6SGX3d&gXXS z;9sFL|Acc+{f6z>)CUucSbG?^kF{st{Hd$+pbPMz#~V=JiJjLT&M~p}a0X}n0hrHR zXygaJXSb(#2kz&3!;e^ZX;1g4p4i{W{Rk4+!`h71-#u^lRavQK_^gv4NEc(_o;=uRp+FH5#8+#Yt!3#`yV{1}B zR!zsRS~U8u(#4v;^!WZgy!J%bh(+YG1M8j09gTLvclZmnX^yz=ggKt@-c^IEdh-H4 z;1XQ%$KiKB_3~;bIO_&y&wz99!1u0Q$0t?>pQ$N=; zlsDFP5C_@TXKveptyyAO>s>w*`@3HzKIq`g{~&+(hVNO18uZRq`x2L?v0|N3_j>ow zm9s|84bC&39_njh6S%79o~NF_T&_Khow)}jOTOmE-r%cW?0MhS1DJCb z_tke_eRW#=XW(l(aPIr z;MYB1?H_Pr65Hq0Uhx~eXMErrpZ(>nwU3;&M{j@oHLNRg^|X(@g|#UUZkw+fa;I*P zo4|L+cO%J7;BwE%)kl2?Ef(a}t4aKZFZ{kV7O}a+c4B9DKIf`4QZnI(TA)Upyyuo5 zfdyCUtd}3~3DiFCt9M7^3?k&UXsn;W9__xqv#s&D`g@E&M`nKQ?;dJ9Tw+}{Z?5yb z$a@y+?O!_Vh`o;-Gl~W$ubrUoRBI35Ds7fq8vCmKvhYU|o&D%rz1-u^fP4Ud-X~VH zTpIlw`#56&?@Qmj%X&2JwlGyzS=sYv(n>r zUq275ZKpPofm36T^Rw6fNiJjU_apUv`&xTiTO4-61+d2htG33|R{H>ItS?|1ob}%8^JC>l?ESIzJ4?djySL9w{RLQi-Em-pynW=I z!Ta#;6R20?s&{}n)T?t{(X{w_7X8X&Yi80j@xeG8zDuqL4pxnIr{B`SsgIzpq}#J=gEcKT)e+9Q?qywrRnL)x)#iz#B-oItRAN-PgNFv_9kT z&epM!tB>amb?_6IcSWrJzY)K{XHYfFpuG^EKF&}pRxjrmxpWiTh+Q+v)LnXa*2^V0 zXEPqg^z{DR=gb&tb8*g4&knBP%iqxI2li*+XG`n35QIQLAf%h}!Q3D>{L_aI5^eT?l7bk@6vf4_SN z_L1|94I9TNw0;o3u)_Mw$AUe62fqN{sq1}ds}H&+T=4d{cTX&E4`cZOjGfc7)V(u{ z{>&BIBRxws!2eESxhsSHjC#?~3FPAaAYn zx$o)kZ0cPj`6JND;ZI{$qW5mBEzQT8!JaMgvCb4&VVwHeYMt|&Yry_WTS23^ef#Zt96^AL`6M(D@xx&n{c7|BtM<)snD;?Z-VL zvr64(=6ux)a+*J){#L|522LC|oXQB40 zp9t~ot3Gqha*yJ$_D^yhsMS^v_=*3-)@Q6PfBbn>8sZ6VtB3vn z^?m*SsqNeW5?smKPn%fpreCu@*WV)=I&18qw!Tw@+D_~)fwfa6#TzuuKT0p~ll09a#ploQq_oS(aV2z+_*4Zo4^_}VM6u-BBsukQn~d28%@ zcGR!;fv-Nk8=SRj8aB6YpUm~f?&IvA-@}}-d;0tG#C|4&oV(cb2J9~v*xG^K)eG+I z6!?kOKJHX{d~2QY0`~OtGB>p3^quhlf!{y}*2q^L{|U?mu&2G!7<&cws<(lw?-P=~ zdH1;icSv>n-RwQp%V8ThI{$MB=h#=9z~g=GXI;eF6Q*WspP#Fr|M`9Z@9+V%_OZWn z%-Q1&Z@-FT^|6O{5`m`fMXmv3wa&JF2>l(usCj@>)5#|}HQHavB~bV0)-|lNZrcVr z@wor!b-{PGwv%(N{RVJ`Jduyx^;~_RQ3G zW&riMaq88Fy<^p>xe;IG?(5$_oZ*i4=sKp`9$M>_VC|>wO4qx?xanufA7?qOE^ET4-n`ZaR=t)I~Y56-pXF)Ma{TV z7wb;eo;uj|`v{w{_hou50bV@g6f)&V1bw&YpwVcTf}ARxP~}{{~+@xBnnM zbL{7?M>Ej4XQ;tiSM$+CtnZ|Ep5&|veD`?bzwxc}XHQL{>H5@l(BRa!Tz9zPo3rM% z`AICV<^_Mj@Awz|2JBt1-}_gHfpR?Q!1ZlK@S z9p&vkbMHYM`1TKdwZG5`-(wGvx%h@(8tjceRg1n8KdHS~Z!fIJx9Yj(RbTuD)YqN1 ze?=c~9`EHm_ms2VSl%Aab(TksGb7imi9K)C)DS<>-0+vP*&h7Z#%QylK|Bc`9U-;IDkyyB| z9{Ap;cE`8p1f|=)oV#Aoi6B-lSG(@||M9)m|J!TtBiyBC#hRyIy>i(gUmD_?M?c?X zz+cS;bvrlU{aNvy?W5jZsvmVv{2MuIJb|y)V-NGy@0XRk5huLXziWDz!2TO({qwT^ z^ORWc;hc6~GbWsMYW?{La#w5z&U6nk;RoEGz#8{;roK4uS3bNut9gOgn#Qrl*%|9| z9ry#kyMz>BUA>{V}qH9vvz4ZS_ApV$X>n8!C? z@3(p1z+T?x8yN5n1n{k#?diGR$nVep3tsoVKeoQ7vHB<0`S#Pd*8|@gDbH2 z0P-WSAhfyS%jd2q-{kqu*2(>LvGn$@I^sqy!2oi`*1gemV&{qg<^tZ{M_WGp1n&M9 zxO`u`KB(zvy^DSWFK`0`zJIPixBPi&#(WQ)QTvf=@SWIQ^xeZ=auMrIookJJ!kJ6p z9n?R-xv$?bPrZx!#AoIQyuRAnA+T`2fvHa~c{L3re24cw>KYivK%Kds#2n*@^*$5p zvE~UbAY;9U`XFw^b#nE%$?4N)=-(F}_|`v(gZu=3ZxdUGoW3(V*4j?qJ}>-cZX@8Y zK&_l{M>B}s$M0?E!`sK%`ks+%|0~?r-Q$P8!QX(r?dRtd)Y!*&HLknC*~8u3!TzTo zQT0&kK8eP2CC)@=8~|&rb3X7pzNZKOKwGnN%^Bo>-TK-C{{jN7JJ`2@x;s{leQQ_j zfP3B1gCq`OYi~Pto-+d6z!h}Rfc>rQ=%4u3KCnVRA0EEf8RT9O^37Z176?W z?ovGn&-cODS>{>-toj$c{=nC|yS(uvz7YFoXPtzPg}@h2a`sMq>*XHA-p$%UY<(wJ zzu$lIQ~l>cn)qkpyBktG+Ccb$&7u&(aAYrLO1xyV*K^3SzD);AlmX8Z^33w*#i&tBfC zZ^)^wKKO}$<`rlBxq2iu=ko}8=G({nR2_bUZ(0xhh<$%-{rgzL!EeA`_HkARm-~-9CGpJl zcUfzMeeC-HzM=KviEn+=Czs#?oc{s=uJ@%KXi_b)^USMzg9+a^)SWx4GuME;)_bLv z{pIwF=N)e7<gWFUuwSZU-zSh8#P;dNvojK5`mp5+t1J0|Myn7`1V+T6X zV(*84_NmVUtuv}Oxry(2!k^z)#%k-&R{Fcc8!M;ZfOpVNQ2P*ms9`rb-`$-$epAb- zz2J=_cF#lyS6ciA?{~1ihxlrH0%rRTzO((UG3VYP;4Z{!JZC4ho|*j4ATO`hI(x`> zZ1qR?fgf-ebLir}HE{M$;64v*ljFF%J>BO8jO{((yvG&TGl9IlU!ak%_`duHoHZZV z0Dl?h^6l}wulS0!mv^!Ln=#gY9o~Eg4Zp$nSUbGcaJ;Mi>dvw5lGw;K%bkPRe#cgx z*trA9pZ~k6-Cn*c{L#q`G*7JOeP6q_L7Fdm;2hoeFk}FY2C*j){9@k2e5y@n;-brd(M6A*-~cb zsrdtR;N7gT)*huLZXh=7bNKRey*;Y>47Bd!eP^zHGFD%EdgOhd8T3Whu?^0Cx6NPR zJ(Y{S7q(&(x!@dkf2t4u2fhPs5UZfZr)nIFPER{)r9$6v)sqeMm~Z4y|=%|oU3-i3BL!{ z)xn>!^0gPfJtFDBp6=g35)b&=7vDbhK5(U1r=Ngx-t)3~_5!w%aQXv z2fkW4wIb2}4%A%L1RD64I`H0{dGF(n6AUyB)Lv+`H4obQni)U(we4%GH}*lFPOWo$ zt~&eLC&{bz$Ti^a?cxc&W5m*`AM|(t^O5Au0qeS8{}(lvjc?c|@OQ-JzIULHzIFfj zS)o4QoH_9m*k{10zX8ANS2)q&JGR~wJAw0Vz&))~^TK}v-@rYqjvcJ)T|Gwt9kNXPyc(l$FzkxygxUbz# zj~)ogsqMg7>H@1q|Lj^hxCFkTwE;XM7S!r@Y-|`$a5a~$=GAu(t}J}w3-=GKeXJk) zxn_C){{Z*x^7@>_et-vXuR2(JXD(|`b5GzN)gLa*^Z8xi?BguY%(aXA4E!thvQcyK zGZ5Zat^JKVu{{$s;)4ggz3f}x7iOyW%s@tdU_I6g=MAiN{w(=9b`Vbyh3`*v^e0@l~uw)QYTz~QvU69j1B>~H*} zxBHLyk$1nZz3}Z(HEr3ROKZJQKY^MjJAHr-JhewIn^;fn6g!8`S=QTUfWGk{=iB?{ z0g7jrnhD)SF!e zKdb}t1^B9ku71R)XXKi7boRN@^W=>?8e_FzCw|O5pUc^yn*-KlF55T(S0~|q`O$C2 ziW|JgT=mEwGb$I0|BJi%XWp-%$)THI;0Mt9z4dpXnwkZtwsQI$f1dpNX#GB5*Q4uE z>Gl1`x3_ok)Nbr|qwWp-9PFDQfN=-*^v@sh-qu|L*k9P=1+><9j|BC(llRn4iFH5V z|GypH{Wsu@Cm7)D?hYP%1-POc0RiOfF|p>_#uJVCw)>QCj+3!)B^SUqyMV`zJNW>f z8B1Gl#GFBF-kI*9uI};6))9MmkMH8iDD(A}#!od^>)pG9hTeC*f#T@d$)Eo2pvK&> z_Hyq5v}y;OchlO}9t}Hzb!}_a)EmorV)eG}9BZl$Yac(K7y9}R!r3E$bA89^iSOtA z0M#E|f(G7jBEY}C(;GX4c~7o4zuea@wNorE$Qk>-`klJ2WnO(g>$?!+ymS|Y|h#M&NA=GTywpD!+&AjP2RYA#j2azdM`D#Gn(3)p4J>V zSKjyR)Y&VsLm$X}Vb^wQu3+TS1R6h!#OkXzZrd8WgE=ApAU}*&L~L)9jhKG)CnruF`4#zv5rQ`@k{!ebA6C#V^@ za`vkei&t;Ix(~iqczMrlahK3BfM2Pl26J)0KJ8Rpz;Vx1u$^z`nPgBx(pvBx*? z1Ruuast4Bl`8&{KkIaAOCY=4{eSwV$`yL%W z%**3j`@p{d{eEA5l50T>2EN}}{Q%~K{1tt}n(J84%oW@7_#;+dxfcjDH>_vmLI$nC z+ z`pip*UvtUZ`}8EDc-_&@#CcEp2G)KxhrE65Gq4Z%D+v4%@K|-mssGFVZ1H{5nuQYa1v*+H#+dH{WFg0X$$`o7S|-$ykQ&OTqY z*6+RjJM>pA{In0+L5+UZ;McvZ0}b9e0ghI^8gngHPxJnK)^`nl0N=~}4gQHg;e5Bi zda6IXeP^t==gwpGOVgIkAh-2z`qU1@E!R9}4Pd{*zC#?|T716~SL-vKuZbOUz*9W4z3&Vxb07Gb zJH~JLrQd!JI8UtpW$7XBjLOqzBt9_jk)Oagv>tnMyl>>vTURIc9qj3j#MLiXe51BL z3;C_7Jba?56XbyNN|UYbb8#+yXRi0PZp3;Y=S&c4fVFB$k8d9}B9eITW?Vj8ZVc>z z_g$~RT5Fo%^xz-4-b!5p^Itt1XUAt}b+t3r?@VB;FLvbW+3}A)oc-Alu3b7Bd)i0N zS)pY&>-{XuS1vI6KuzjD84c2?Tz#l+f z`v&ghee8R~`e(?XhwxqPGl3edKSv3kAkcKUEB<4{yXx0meCyQ6*N)U&?&Lu;$6WR9 z`RH{>*j9`mWu{p*{0`Bm31Z_BAH& zc~JY6&ZU%#d#+yTh$f-5tJK0PA zs@?u1zHQ%K+~W=G;k+Ri#K63DgM8vYut99DYOu~Zb>Gy%{e`^(b^h!--(J4G{i=o? z)i!vKdSNfQ8N08RSl)k+s@;h_>TkfFJ!DqTfV;po>=RfUvG(|#IPqUtYwbM+Tm!zH z`A%+N{k-MW1hF~&0q>0JO+4Wn*6+SMKFI%n-&Y@V9R&HBi9a@{wN8EFPhd~uz)x~o z>n47~_l#V;yFKpfyDlH_J=TB5d!fcYYU=zH+t<&!Dv2S^J-!#bi z^K}JJd}mtYd)cq$+Sh)~IDWjBTE7cD;W^beSDXPsjWevV{)N^()CX{`XCz##@1nd~ z?;2QhqIN+uz$C8U3E0{RTfJl5<6=F2f+iB&_nYkA_IHl=zL7KT=pJ&!TI-zq!kI9hBar;i9bNY zx0m;~@6jYV^J>lOJ4an34r1S4o%fY9C-0m1*;vyBdHpAxClg&P&BV9wujJ(1;Q|8Q zT^m;4T4CQtUfy}7ASZ3?gnOWC#Ma53-?7er0Oyv5oHHAC0Q;+VR*N;Kd_w!ydt=pitem?fsD8w1p7e26 z5;qXQ{ug*{m&1Rdk?X*{^e4IMiSD{%=QZ+!TyzzaXiSPc#kNe6S%U`kf?)dh0Z+-L5a8LgX z4>;$jyhzQT>E!*`1bX!URq-L`g!8M4S#IZ zd!lRfa?b~Pd%wZg+o(-qIQzL{ec#{`+|*((*ubi>U&gx2gIJB$dERZ3GylT>9az^u zf(e{4uotW~=33yhf8d8cR*g_6?CbjsteUfj+Pc3M-3?e5Xw|sKnKj|f2as#P@4y38 zPbf9+<(vsRxZKz8kVZ~)FhBrfVUIUReD7#oom}%m<-hK0FJkjTziD%6)SWu}JF|m8BY*Hb zwXxn~ViOGDXXIxXx%QKDhOtMj_mwY>TAOB-5Y2}?tWNv2|^HyD}9dsh2D7n zq_MsO>jv;xcR?#J=Ud*WyB-c+O%w2CHL+@N5#XESs5KX00N=Cni3R5EGqoJ!C$o&j z0N%m*-rL_N{*3(!ZoqzV$FirInn6v&pMfRkiSQpkdYM1JFZG>cKYJv;oabRK;O`5k zk9yBc{sHRy6!@Iy*4Foiy{xylCvxu{t0sv%zI&Q`W9yCNeJkt$b+8>IuxA4Ia!%m; zE>GYrIW?ko%l6|=k(Kui3z&ycF#BadCjU!w2 zoA}PTp||D=Y9B&(&BOs64Dc`?Yi7gB2hg{dnxO@r9xJb%pm?lr6Il0g|HOB$ToW93 zvUlx_eW87W5hTrF6VyB9`n>Gb;k?HUt#HRW_%l}CI!`8j;2XB9-Tox@JNdk?IquTp zN3S4nO`^GgCWz~M68Pxl6He<+Wr^*j{{W}X?@OSm9=wD5^(5xp$W?8uyGO2m_O#d0 zs;|!uzJaq#=;x{(o1a)UX`HyA0c#5QrIB~fSo|P%Ms9oV`9f11^^@Fn@2u_AdaSK^ z)HnRIpL05~`I>``Sap7mCuhDsYw8F2G{z1z=kswNee=TO+a1pMP3&G;dqk`ma@ITZ z2?AVz8hdoCxv~9*9dO^U{tle@4{VDC{l|U5UF_9>^Nj76K;K@T9xGqET=V?7w#I(W zQ)^w@zWZeCe^15#O&zh2?;vvRG4ZW`-{G;sz5#BakuQxpj`Mu;h}FmY*PEyX-rIPh zo7np9pjX%BsC%OcVtc(dZ@(Mg0jI59?t3OR#`ba#^WsU~<2>ivOJ4tteSj7# zUwh=@>n>;-aJMTu=X$I^d%No!jW&ZMerVD7dtrb^tW_hdH$UP1d3KM#Kz(P)2hjKY zdS9Q%le{yn&0PO)>^iuFd8hvMp4kqyfZEo!Y>ps?Y>{Lgsb$`Km0y|>$ zt6F`uqK>-baXzj?74cn+aRtZ&m42|H3Q!r6MJ@dzZW=hQ=|UA?(Cd^YjWBE z*tua}efPM6h!tPdn+0Ux{~{*WbHhE=5MSZlsrCuV;5=$Q*2-lNfV!jg@F%X` z#Fay%)_3h#&)LHnnZz4S8||yl>1}<@)#maW{ukh^O%NMrpBby?fbT%f&&T-W8vdDS z?*R5Wy&9*(nd`aV-Mix#&mFBvFoAwQ7M;h~c>xmWw^*MYUit1ISHQH~{R0X=u%*HG zGrxd3dwz%Sv2p`zpVTHi_X+&ECw^(=;pFY7uHmb#?-bn0*Id%p1xVmA9egeggX%i$N@6qj^XCj0Nk} z4^RhNelDGywI1yNHH*0R3M`QK=kWYICUJvr*e>=yzLoII`0?4DJxb3`)>}8U)XJBS z)pw?-gRNb1^;92lX$-VA|5*FQKYQDAfB;>~T@vek>cpBSH^CFWa}Vt#=XcYYHCJn% zdsu%}Lyf;@m|{bFSpAQB!}uk{_T6j(Pbf$XI_i&m{4={_`gHl7Em7d2GYV1yGm7 z-dlU}E!IryuH=jd*4TP$C)Rhfhn)9T+hhHA)3B>A+!G{xy&rdR)`7ZCY_Gt2r;5>8 z=WZj`8=W2%!#}{sMn~h({sCu?vh;E8Z+DX9oHNM{pfzWo7x{}8nEMTDZ@C2C-QNEB zYp?%oZo-*Mto!~2-XMrY1DRNkpKZUdx%O2rrxhdCY>JMG9s#VqeJ-o66I*LJq%*v=?*W zZ|yIw`ZfH^ojBLM?c>b8v?ENHm*8%Kt1?oE(XkPcF ze_-utT|;xhZ@|x1-d_C^E*JlVvtFJ3{LC7$^-p|rBcZUKnd?mkIr}^|+}E7i8LR#V z7ofv6?Df7tRBx&F1#iy&L7Z6Yiz6;e{$pdp^;r74i}x|!J)9N9-Z`-Ys4G8z)HQhL z1yt}1dvmd}7g}{=x?(7lxa?W>l z;-CC`_wZ-iTxsp4e_xpBy=TXE_1G3mGxX7Yz^hZ^KJxBk-O*&M_xbrQYMyAWAgS?; zzz5pvec4F*{TKb<8rn{*rW>QtmmgSbF8D%SgPWj(8=SES^YnBFYimZX&l+a^*-xAs z6Kh@RkFVXQ-^nFGtnYQv_yON?%?jedtp5PNPg-|M;LpGdTY7Z*4`ci;a^C~anmPf` z{{CI%4R-|z7rE+whyQQiS6ts6e0LG<(c&i%J3GlWtlwRCu<;#}?g@hY+!$E*P$N2ebz1rIvG#ex+jro5$H3}q zo5*G55?Eh7?mu?u@E+%?srlp^{?(pf;5Th9-*|#2x(Tng;dlHlM|{=K#V0nvFb8kF z$A0dkzP_K>KevW--Ptz?_y%J8*8GZp0SWJWo7a|4Xj~_8K@P)l4n`V8U; za$9G%1RQnMKk(tze(iv7uDD$JAg=s<_4E6bKWUs9;2$T}{Hjgrftqu_sUI8cpspR{ z=DCNOVBWyGYidEA+$L5&u)_VCelC0NIN<6Xpu7H#4`+@0Rh-NI-?_#)8goD2DX9@n zaK3c(w6~`?;)n{DmVM70cdA@+Cm?T)^#dfRxmt00ZC|c|$fYA%^*RY+^ZLx4JFfY1 zoUdJ&UA^L`noqb~{qfdG45&S~?Xluc9Kd{Ouzv65L>Jx>%pEsRtov#+mu_GKG$3cr zIRWH6=7s0HwSIBAw9Xp%4b<$w=cwsGUf*+et6A=uMI+ds&o6L-*qWoSw}4ZBd~51x z1?MZ51hMMzzP=yT+kMS3XKw3fR~)g&*#p!(tf$B7^N*e6-Ityodn9N;zw3KqYZ43W z9s1N+U!Ng6`1hoYw*J1t=i1QtyYmGe;0BESeP^xw8(3q`zBN+|9{2To**}XsGuGUT zHDIPJC)b}D?h+xviS@HTw)KrFpKITmh5kt&cX2QKb*%j!*w@a<+fUupf=O=deCEo1 z5MS?WZ)*Zwf+xK5pTHgb^UXO`ht64QCkWyO>h~6Utv%(eZLxH|iCTO80o1%fCto1v zI}hTJBewp5Cg82l%^hCN1-`}VJ%5VN8)?HHqI`58&?$C*?xqtv3#&u`_R2YdTP;ZS>I!byx5L9%sD34MeX0w;>mN zdG~R*0DeygdEX!d3-_<@tFC(UuD-1rX9#<`lX^cp^^?51U&%i}^^~K(e225fxz!i` z)HL`PygS^0I}EtK0aJC}@A%%O`sCV2-Z}Et+cUrf9`~_cuHlbu{dYe1uRW>z8>k6v zy$yc#q}Kkoo$uD`Cl=rS9&7(&?Y^`5gd4#7`8M{us^JZ~Jia?wqp!wyn6ZA3+2aZC z40}!N(aW`1u&3ujt;f&8K7RKSM1s#fGgNSL&HWSJ|9d9p6D|S^hI;nWe{SEs%?~w6 z{@kOjdS`mqNzS{f8~E}aU;PXIyyZil868ak{VVz6(A%eBtu?RSo{6uOPvGa<;hG41 z=2RZuW9^IuXIrmU-dPdb$67yf^;V;AZO|)?@zwZQPvVO+u-~xWv%#D1pnea*1)Mv% zXTaOj8vWS*z_)f{^?UBOz`A4QsaF%`fqmq~#BW&P+lRgf-(JT4-Q)$#`wsR!pI2El z_Bo%cbrY@w^@F$xPt-rxTjJX5u{G*5SKeMp-0zf-5BOn>-aU?Q&SU(?QczS%zB}%U9HJA5AW-DulKjs&%$1TrhZ2g+ruMot#5JxWBU=%U;=+` z23B3ismczx7jt|bFYL_y-Nk=`nh`6n#yWf0Z-5VQ{cS2>wC<5dh!nu>x~2ZkCQ*%##-;;nX!3%?>rI*bl`u^l~{T2qknQe@Hy7|URPkQ zsRKbgh@DgKi(fU^jK!%9^M~UtN)P8-$P16}WUhSPqb0Qg4>$3j_w{{|_e8E9#%V8V z)M+2YBbJ}$@jd$TW%1jFHTvfswIiJ0b$1Ze1M8mp1I%1+aPnH;=?12GkU-wvzOglO z=DL7$jywA%&NgoMRWlMFoyQ*5RgOJsXYJZ=E) zJ}<1t_v+|+kj{07C!B9#AL~8l>@$J0&RdQkC*Z3uzMOS|PHmIR^?Sv;$*E8HzTqA( z;ECWLxisG2U2fp;avpnn|FRPJ3FrP=-{S$C8N>}ZN51Z0ANg23Ufs`mwHNR1%<~;k zJBiiXuX5M~Y6iS8?@?oo=z@L~+n@O8dMvHG>6aeuz^}W)>w89$4uEyf=Y7X;Y7ei* zc?IY2~dSAmz55a`{tDTPNY<)lIB3AD|x#AMob=>>D+xO1*PRj)^liPf@zKS7z10YK{3lQw>hKfa8O6y1=gEJd>A=79 z$pvUY?SKEia)UU*1p4099%}Rl);qbE|F<^H)$eTV6Hbk}N$fuAuhNEJyxx8 zy(hi=PS`8SH86qi64>*`a>kx3Id{pdaD7+qsiL z|C`yrs=KfMuFhFOu7d%rwckKfK6h(y4V`Dk`paj`eLV$bB@4oa7U-FCSsv8W8fc+_n*cf_CETBq#yV|Q#;~EEdK(h z|Da}IMHBV_zJc5me_}(xsaNMtzQ=p#VQWT`lULjE{rn=AKC$l71ZM>QxUawaTXO>~ zm)6?ywcGEB5B&GsKR{Q{QS1G##O~w2tAD*`QnK^b2k`R@xay0Mx6Xf06u*5!eqWzw zk~80cZ>@jc&iW3_1u!>3>NCUGyOxa~dpwxoyE@Ar0|apA8+z*-u*cBn%;Xkf@vR@zT0d2 z_VJz<(9u|%pmrcWT;Qwyow{Ga2T)ru2L`ype|@IxU|%&I)So5f{jQzQc2bk%>S#gL z`)0s5s@YhNGlX0hYB+Mv5OuI8*Akyy)C~EP!?umcr8%*mLz1)3V=w&~1RIryx3Btf zcw+TIocO+hxi082xQm~G?{RdMBR|nw(}8yvRZBll#Hy*i^yyc?k6b@n^C!M>tPu@V z9j>3B=)ixc{{wu0(?k7*+ym4expG85W-_}@z_DMT^}C_9ZsfuwZ$F{m5|Hou;=bN` z;5#$0=EMiFJv+$6`VR0O{iC-=yzcAw(+zvhfH&vKSbgL=8h7kq5|{5i@a_8mb#igm zh}f}t_j+r8fychUB(Hq}dtL8s9e88sB^aQ^s`Cw=-~l?En)bQ%#A==4@1j9$y?wp! z`3~LC%Lm-VHefIJ@;$8WaK`2y*bA_aP%Cf$fwiChg`D*bKY{)2@c|;2&i|WpiEo|z zxK{%`06Ocf9pLme_UGU)kRXV!SkD7HW6w$m{0^o*aHnr*{-5uQyWm@+b%zALm%g*@ zWqxRhD~IoK-?~5iztMQNnX6Xa>0^%o_No5V`S-EEz#9ZLH|+DiYtiZ_)_b{g)uHQf z_HFv~8N|N9B(68W`guL!>}kHksV!{3_SUDD_i}%00?t0?ZuT&5pFz%Jt-6S{*Tna@ z=L>!2s*#_uddW{=4jbDa#O=On8=Q0P^VHwFUGS~D!FTM@)E;mdYrnvkyMEFWPQ5vO zIqSlh^A)4%QB=)U50c(x@)H+AM z>fwDqeS5ZCnowgcu}9w?_7e6F>=oSwM6TFe2igOBj~#clplq(b>eXED>+^A+ggbBC zIOb#gK`ZCnpJ#g?>(mC&7wX66&S%(8-X8dK=c`}iQ1kv= zpOf$Ii9nlr>-UbmiVKLXdF~p2)~vsPk<_tw-2;E1x6YcYn)~i6$8m<*I1 zkdOV@QB$86?_k{ko+?~r|yyKU5xGVJ9%vrvF_rVB>v5MxeepX=6qY{dp|Xz z_Jo@!&Q?9^K5+K2&ONW_8b~;6&2=C@a(!mzoL9Tu*Y_>R`LpZKVI!}9(}%Z*J(F`zu{HLnICdVb?{eAs0q=}Fd942(T*Hc&^$E~BW5B7e zL(Un-UC{diXMDptJLR!gFz_2F4sGDuZypY74`Hp^`W|7`R_+%pnB=t`egL&`U;X53 z&sa6~u-5xbbiPyD2yDXJqb$DqiS@o+9mkr8_3uP)c(vC40lcem!#XpRwcwAK_b zejX!sy)?f2700`tJ~Q`soc!B9pKJFZ7H!+QgP}1Rd69|b^xz-x-tBQ;b0gOt4eQBR z`OD7tqy;dZSo^uRyjo#TH3MH8*nz#lcYJHrSo6mJ*!~ND!nIg+YCLCF=K6i7M&JGe z8^F`izrl?@a?bKQ;OV(~`zEI@;R3yLuUOyOyVU$xwk5UH)aS5$dro}sV=p!C>Yn;x z?18hFwe|@7k;L$|7k+{nKVp4#@> z{apWUYwisf#C7o9YZwE0Ya6hqGmP~!SFedx=T36wAN0}h_})Fx>rZ3Ox%+vOh;>JK z-_U*y&b%|-GvGV6o7)&zYy5L~0PkFRB=)HuvE~L^=ct+Z>Z}dw%q6)CzDJ$4EtkeU z^gSb&F0tnxho7Oxp^r5c>oce4;`B!>O}!1iyHE1|nO=SI&pUMT)|f9n8uir&zsVaj z&v&T56T#m$=eyZcT?D%A!_*$`SocP+rgp`bv(8;>2l|@}HTcsU8t1r|rziga1L&X6 zrK36bv9B}V)I34pPqg+*tg+e-65l@Ny<_#m{}W%RvzPf0_#Gtv0DqzJ|IhRn*8b*R z)c7V(W9~EXulNnUaYLg{t0uALe+B+`j18oG;*+z+ySR(ogd3ppje}o0-pm=oermp9 zC$XRVB-RgL-Z$3X$l2SS&B;5jqi>%p=fAUj0Q)6;VAbm%v|Rpxv)}m)A80S2!T*l^ zTOatlEBc^FZ$&Z;(k|9SESd^zZ{;Bv7mFU*A`MclYnm4J*!f;Asqh#a@7K>zn!p zlRh`ByZJ8m{foS3U?WzIwrx1axcX6Rz5VT}=88?M{W|tVJg{oaJ4@TOaPQBhweD!_ zEhf4r*1ZFNV2|d3uciUN)4p4xnON`cUe4Bb_y+$%=iC5wV)e~jT6YLwPc_yztb2Hz zQ}f~F)C>^#*0@jNJ74=i=R3HAeV*jKOCx8W4rh-}oLFnjO?>mtw_eVCVD;_q{Rh7F zDF=#!caE`q%e7aBtN6aYqm7&z^Ib4UEB_ao8~WOjSy$qKzhKpPhpHi`-ZS&ZE+fc; z%3&kdu5}lDk9zw@umZk0|1)aeP~Z8!n==!Qy#_kb$QjpuXw4tK>ejF;5C$@Uwk2^YIYdz-94r=u$+?yVQ_)4rs>kig>yi<>5?KMTN zzRvoMyijBP1Fn9a;nd05vjskQ;0M${RB^W zVqwp+_`XA81HF5EfQa?yls&9Fwf4F0*f}R>t@{OioOhtU!wu{Wy~kN6-$G{1Q|Igs z?2|zPt5z#lM-8*A9l7!o`-XR?w(G6y@D23)dINR!zSz{a4>@O@p4v(Lgg@Hqvo%Lo zcZPGOZz4hsXy8Wt0=1Kr{C(-njU?ujUJGXrYn&l3{7&4!3%I|164>Jcv<-g133s+n z1Mc1BKmvOOR=s~0c>{A}b3sjzvu}f|?+MoV3Dghr3EWZt3TMy$xppxZ0&W6#>Dcy@ z-*eTfGjEU55I17adHeEn^&5djYt57c_L}ewYhJ(Jk9Y8A?%(z!7w`l7BKFPnC$Lvw zt@*iDy_`1CsH-~@`;Mijt`R%OygfbE$k+QNK4(Ums_Xgx@9j`$%)@Rm$OVB}a z-0yro?jk4D*y{n@slf%DZ``nITdtp7=e^Vfz_(xZG=%fs176$B%W1bxjr_n)P@m17 z6U0M*Lv10k^z+Y#H>mz__A%dp^Cod@IInhrZ@~MR^W5%huMt@8=I7nF#-0saiJxE^ z$J%vbJXuxc9U`s_FHKd5~H{Xenx9{8RfYX^6eQ{&r2 zt~j@?&l1ffZn1h_c8#%jxBlES=;2N;P`z`#iQk2e&i$H*W!3v;&NFW>HL2zO?63Zf zRp%YtQ|^KFuBG{0-?0y@-+%cldHV#~39NNSY|i||SM#7|o-n{U9`?ocsQ$N}Hn{%8L{<31e>^1|BMft>j(`lFMdXr5^PjbA$7AAg{2JDv@5BnZ0o1(Fgd8~cvrm$DH{U^ALBP4QcA$5@_fPzR z|1hSe5!-vldT;yLU(UJ?@6U>T)eKDst)7k(%3G(_W6uY?^UT+q;n&Xt`3HSEvAP#|&&b7}nnrAYdp^;-i+gz2l+&V7Zx8JR zH+=8r4E6GXO)Y4Gb8`CX&0C|Lz&-(f1J0^kG|1Iwpz&_n2JGRkmpgfEBj>JiJ%JqX z9%{wE$vbC~Z}4gdh;7|5bHyo-b`aaU6N@kgPqZzT#$NT-a%j~UyPGwBFRhVthJDkR zoV@kc`Ey~dr}Bx0z12jlT@qV4c>4?Qq`d;`2O8hP<1Bld?b*$V(X2C=Lu)7C4JG_$2#w-f8rT%>>O~VZ^Up(taWz1A-+A!C)dD}T%CZ& zw^u_G*pYC8-V-b2|3v2-yub%|08hO`{Io-^oybH#7x3I<>joc9f|Fg*voyt1#gpZ zf$pl7oc$+IXDn|Y_g8z~Ks(6SzF1G>>itCf06k>aQE$I451eVOHE(@(k#m-{H=sVP zq4omb&>G7(;6CzIk3R&w`P0MSHIvv{ZO3+O?AV&Cz0uXlHRF75)IP}HHmtj%t(~;& zZBAVR`vmy%;na$TKXdKau-et`{RPxi4*zoBdtl$-8*ta>=hBQ^vpO{cSa*JQ|3vp22snG1x1Xmc z|G;-%;(uW6DX*^Ximj8&_|`Q8>kb*~{pFh)V$b+ozoX92oim@{y?JNOL>_B5=Q}qJ z2Y)^zt^Dcf&y~LWKJ2r#_IljW0(UbWAVD3hoN?`IO#S~oSB>A(H?ZFz9$5Dnxq7Iv zt^wc4Tz$?^JyuPSQ`e}S*uOsUzb~7K52vQ;=}}rZdwNFv=<}t?^*h-A9r!=!*S7|F z_0!ZIp@*Ne1Jx62e5LNfPW7%>-@@7UDsUH%eMau@V=-{V_WIJs(y5*BPkPnc@4ohS zR*>_IApS%Hi5y_l$?7Bze@wiy+7B_nD?(9^wCawdECqSPY?k? z+<|<9%OGYS`yQ=taU~w){JH-N)Sr*=gF5*adHbsK?*JW4a;XLWoIdXR9@tKev&4BP zXEx#hLtZ~yalcbv-$$$(=Xr;g=-5N@`3h!&~1>Vmk13+&^QnD?@#a_soRZ}uYJ@CW{k-5;C8PkH<+8s}Qu@awxlE{Ls@ z@7Rj*jphCPYbL(E{tc=RaRYCA=yS|9t9W;4#*1ON? z`_%Wu&S3|5z-O+SiM`xgy}$o`Y~OM9iKW+9f5rX={O-O%{j5+L_w~8k+q>7E=$$*T z=k4s#fN$Y$>I29N{WJH9X2RL)%xQ2PPQIg^p!5w(F5u5yyrcW7E3lV6?WLxJL~pKY z&<^58Z2yTJ;AqsDckdf;Zdv#S|Guw(?@*I)*4KBBxO7@Hes=0#SmXES+Ns5QL$%iW zj(+#tInj3kXYL6%z~gg$9$I_aZ;*c(3iSi5;I_2<6FpSkjWE(P!M3ER2T>Yvp zRKVxjeWLvpT!Fi2z0<(&*oKY3f&@KrfxWb8d|&tP(MmVq+|3%fv2D*4H*5g&+T1wz zGC#>_JGjt4wC)v)v(|iQ1)7oTUP&zP4%P`Z@*d|@ZDND}0bOj~KF(4vZ)`8~!69m1Ha;{ufFl)XVaXJcZRk0 z_s^Qr<2x&J@p4ZzHOqI!_s^QM+gBeX_Vx7mKg7H}h^*O`rnd(bOq*ehA&`uaXk(0l zj-I~Lcdk_{F3ljAF~SfCgp6o31JWQP5Q8ik8zJKwV}x0jwB?l{BSB7iwKA<^9+fAZdw*X>W}W)s3mh?X^*K`~llsj~Up)8C4hSJjUeM z&v`F9f9OHJWAA6>tVXPGUH$B7t@|0j5D#r+248a3cQriECMS=$wzcI)EAcGW8aJPK5btok-v=-r#O>jk?B`Bh|A?9QJhd;+!kwHa?sAZ?=n?DAd#<0ko_7+nevnV% zinjCiPCZtyB=!U&*S$K=8Jk!A;o5(K6S>sq44%c>{qyAoJAnBMxdva3HO{;596HD; zXKZ`f+gfdN+RpCS1H4?9?LFMU9aIe*x!MT=-yZdSwr%42zJvA(pzNAX+@1YzEk0wn zTzu~#v4wpVdyiN>jLC^y^`z_hb^I+C`psDNU$956-bu{9?q$7q)QHvY=muBc`?a5E z4*bjJnI~?ahaC2D7wc}oJNXefqhiV znL+FUC-6^l(at;pc^%)K%q#kGW`5+&$x(W)xG~SM@4>kP-&*%|*I$8iybJp!2;iSv z7*|H{1#~b$oiEoN)V|nC#NrIJwIku%1JrKB?XPW4+3(gTIo}_7v|>Md?K3p&=M`Qq z?hRW`!{6E-lbG5wInN!Pr*AL!3H-VjczxT%C${+lwlzK09(4x%B<{Q$yxPC5Ue+e@ zfmY;=&^r3&>Sb-y{&_w3-PdPh&hM=da1?bv;JwNpX!$)LAkH}xyMU_2&%{Ox@1UV) zE}px~v!AsO^7hm2;7;6r+KRUQ8gRaK?x~JgnrwZxdMA;K@Ar*-Idx9r)+u$)WG3)? zrK9bivCyt-4`=us266j7@Z}kALNTB8V79!DEoXv&Bb186ZLxYivF)vxOA*W^7HHqg z8&Ilu(lZvPkR7$z&Tw%J|`02xRODhoU@%1`0ngo>Nnue z4EqPRbpzeday^HVNzS`LH@Iix+5z-I;LDk4<8q(Iwc%e54%tZ!Bp+{qey*fX?a z)$ezp(Z_zpQNxiZoGISVrk@K{Lw|EVFaO>u1bK5+$6Wmc@BPUcD<8hP>w^K59!HPF zK4!Ro@${_$LEf6;;BVV_^+2}+sSe_YKe0W#n8X6ubHsY~cXQZJ7@Ggsdki?h#E9CG&1E=`X!^B%xGoM)X#?0r}6P@;~- zGcSK5h?gXk0}Wk3xrep2Ph$Ty?0q`%0eZA`ChQ3Ul%VS#0uPjZ#5{}VxBzqG$X7q+ z7nhWxow4Avvw!VKPCLjc70WUg2YZ4#2R-e*rMR(TnICuAiM0dvonZeN!s(S>`O(jO z;QKwPe4Dd}Ge)ksymr`o&z;}zkLRo$RD0W_G^Vx6&YLr4--_ug`SCfTZBKg)V9k3y zZu#Wp`R@BH&YWf7zYDfC@6RJvuLoQa=7oL8b%EbR@E52!`8|90D?e9!pyl0||Nd@Z z`@ByAdy=nyhflWtKIB2JlPgXxj|A@ zHm*k>bJ+Ix-fn2+hV9*Wj>MjM{G8PX_H5{j_Uqz1*BK}LB(H7kj_=Q9`zZGE_j>nl z*dtb-2fp_Gx_WkUiSHa|ynq0cSP*l@#h5<20rxopZEJ7nk?Zfn0__ZQYWw%=`kj1Y zTj!oF*3UwFKEW^GNALqUgD*hovCiuJ+jZsb`NVdX(0ASpY@Xj|>_O}d$2-@SU;FTk z(2v$f4{ET&UylSyQejNkJ{&fuXqps+0Kbr56%s={qvhA`UXa>b2}&W;W)Qy zs5d@|1@LZ@oVZ|n-xFI5V2!ro?8r4&yJLF?EdXLKpNBIm?=14WhyCv49&)hTMl4=O zC)nCM2z_=E`p&DF==M!cp7O49+0M+seqHCjp*t~W590P!%Mr!{-9a37qPu{*=bQH) zOOmg#_!B+A7VWyXLz_=vTwC4TiQf9}`ulZn!x=~1-IR<{RqKu&Ka3q`EBi`sbM&=N zp7J1m2cE;d>Mmp1ax*{r?^^5LCDuHB_mER_Ue~j4#0t6YXQDfBw=?kBH+*YGtX`pJ z7yTeE-&mqA@bAEP>+50rjv4E4YX9r%H{rXF_42KK0eI;duZEdpyP;nW>Gyct-Ik5^tb16hI`sWRDU@3orgcrp?zq3m=``D z^Hry&pLZ=^v?FFb6JN>tS~)ors5@|Hwb(IhwcX!mt#0`Ci7Q<8FfWWb$C`;AAmFQw z$<1g>sI|`o-hp@cqZ)}^b?4W0K6^D_56@?h2V862;RZT+XFY-ZPw0NwtVhzp-407h}oy0q6`0k+g?v;$i8`S*IZ`7eXIrprea|sEKvY(qhekAX% z9e8fJcc7e!jU*RqzIzR9XBsQX^(;JR0Q=P(;?_odQ^UMG{3IuD@7L&Q-#QO*c|F1R zaENw}-v$XMNxsdr)Uz<99W&op0py<*MD$9{uo*dv|-b zJp#DbM9VEc9Bb=Y5wpK2$rZO}f!gyiV-UMzdq$tf%V9UUug7k&_|Bcg1E}ScPmQsD z_&-r6SGcD=131&2=7EkLx$2B3b{^V0Q>rF01R_Vl5bJu2CxOQ__=U?E-wby6#1TTkg z&pH=10hAlL;wKL1;C&Z*l+7M#KGxaneY-B-xyBW_WfSYzbsxF*^DOorz?^!Z<*1d& zHAk*}y|WoB-q0Ay8!!GUpDhbEEo&yI5?Nr)xRO0%eY*7RuALUze{3)3^dY%FA2~-t0(Y#Lwm=A zoP7HYH82+Vrz^A2a*Y*7UkUh$KA`P>6D_JXsA0d3uXu(GjDuHaN@8Fd2XgP^y_<$z zeDnbN5r-ykZW5bl^ZVH~WcPlrfz#p0FNyV>8d|QXnI>P?{`Rz9ajt#>WAfGd6WB-S zr#8})F9DE|t48f}=&?ZR{W%rKxc8JKIHPAWK8f46?vuQ+28=86>-^{o{T2Jj^ISd; z`D&k2$94x}0p#piLoOoce8jyI?G`{e;+f0{e3AU6Myyr#S!R&S6_;U`-w9s#dy>%E5 z^7QT46I(IvT?J^d;`>>6Zv{9$3+v@~v^^RqpZx}Q#RzLR5TMR{)Y{)!_A#&D(F5F^ zg+#2pwU5{3?|s~HlDF0#`*~TT*sJVVcE!9q;ty)g?f1sm0BH<;XI95<*fkGZUC)bH z^+o#(6zddwJdL64v!8EZC$R583u}J?)()VSr+B~OLjDE@xZu0PC$zmfXaUGs`y<-> zG#B_U_!E2v^1Ls3_P7&!>L)q0y-vh7>@$y8C+-aEN5BQO3$Zis_h{?v;h&KYv~g!O z?BZevu?v{k?&|l@{fN0!!gcNoF?-p!!8Ja>317%DAI4rk{+!&t_VMRLgJZ2d{If~9 z&UU7|J&F13^&7bxx`PCd>*`~@`#r!HV@D5nFjqeI9o)<_+uZv*1t-NTbYLHMuX8z`zp?I7w6AB?tq;AKLO_@ z@whIYv#s?^jXLA*W`BF%(eAtB_BMWjcLQf~);X&I&$u~3Ztru!w(oxa{yDXMIx+i% zIehv1*y*IDxow4tPBRaeZ0yxW^ zPvnza!+v1fOJB%wmmuEJa)fbv4LAYJd552Ybw6I|i+u))eS02firDG6fn%-i!~9k%Cj_6fLaaf#bon6sv4#JZcr4`9zTHHkJq(E*Gb zE1SIM?zy-=ho>>~#Dulmp2WS|lH_aPgO&@G!M7d*$D== z{hs8ht@kYRKJ(qjnipV?;#d=ufp!H7+IR8}9>D$U3_(tqd;Pep>O;JCai4^vzzg6z z90}s%?ApS6u$MjbXRf|MEcJnXc2E19zTrD__ddb1#=K{4!1Efvg9~Wn?OD4ryXE1l z!HCuGAs1iNvwFow>{Wg6-LdA!+OMM@#LmE8*2*`oZ@lgVzvDaWLQKd>ps1hJeFDxt zql1{e_06gMEVWPF&-d&3%h~mwzfo5^kXI_7>x@3*5_#hAN3MCbFX6rGo%{#_hj{iq zd)MCRV_lEqJoW_oRhz5d#IJg|4f_En(e5#GeLn}Wz;=F*!{*$9?fFk|>;B0*(|Tj} zP}`?!(dO)-9^fG_SAOCL_J!C$+shvIe$kHg%r95$JYZgX%fYjcoXW+r?dQBCZhino z?d*mYp}qpd9rZn@IeG3QPV@@<5i^%SUw%DnYAS|*re8-7;F%iQoO1##zjB0n93WmC zWAu_|u6Dp5*!$kS6Md!jOAMgx4EtEuz`#~3J=UG-dq3{)9SwYE?p{HUvgx;?3Br1bv1t8g?U}=z3nn z}5b&#ON%FFc}lbE>x=9Ce{xF7d-AL|CNX9DXUz}?QT z@D9kCi*-P3$Cj;{T&^)K*}*Z|gQH?TSC zeID)NL78x#*L6Sg1KWJZE>0}F`k2ReUIIDV9SorC-kCT8(T)SvJ67+BuKMqfJH>U! zkr;^2(!M>9dTrqO?AL@6Yle0E`3Et1lbDbrUwZ)W^$dJ3d?zZ8IuTQhc^8U3XDn@9 zYUjFhdsYoxYX-SIVim`p;6j~ZZh&@Z?GEg3%y-28`g^y!2XgfcpvXCaAhvnN6l*(B ztaGkA`+J3R>>N(H>Mz#`l+4`WKXAlD&-`KWmH{A@8Ee;xmSfyHbslz8zXD|D%60UJ!$w;lN6bDxr%ueB$|kp;opGO;z03!GagO;H zY6rO{?1MH|JnA06+6(#)QXVjGOzoM4zmJzBUqQywd|kX(bTWh49W6(U1KJa)6ZF8Q zr+ZtYjQD`LI!i9T@9)5_bD~Ssan!@Ne?PGA!CCI=c|7l(K6_rsbv}9f8+V5tv&KHo zkUPORkZ5Dpc&FCPSi3gj1(f-YuRhT)f*{gk?PPDydNY@+#}m6GmcO5!Gq%h;EVMoK zonfzd#fDq6tsysJ^-r{W6d&99WhZ>DnF628o?@iI3z`6&pMv=eo zCu8l<(6!sot42-`Gk>;LL62B7jmwW*9J1?qQRiOr0#r}@iCy#2^&Y6{7ogr_eD}BS z#=usb(ZTlJqs7_-tL7Dl*dX@=6T7{&=Kr(q*VluKR`z|>{(%n$Xt{c|!>`?sdpOfG zh*#~=-#ItnKIYVke!PCXyYfe_GsyMX2+w1U{Mwzn)K*UL0?6^M8hXUynVa|x-+9Ve z9qXCOb*8nQnEd-IEr7jZ=gsMZ_xm8{ZpOW{@7~D6tA1+oQtv7l*mddH?Tl z$+=_Ia6V^SH;9W1Ib{a*z<=-HGkMYQ6?4uhP23*l5~%ALsW;fc{WDwDxtDh!IX&^L9gzMd_K885`*8ynaI?VkGnd4hjG^#HyH zH+<(^fajAR@XQBz??K#LN%9rceH_|7e*j-Wk5zZVZrJ+v>|g@z8Ov{Skj&#g$IONwpd|PT^us>j4D4%c0?*LUVi*Tcp!8V1D^JWipV=Ke!41FXstGlG=iSH+ z@Dq@$t+r2pU3Z69vuCV58u4MCJ9q8?_B5}3A@6P{;0}J43+wDFx5Wn@<&-Z?@Y`+Jga^KXLYW9g_9T=av z<|g(73_&c(TXO~hto0p~FVC3bo!QTtxgC_`bBN_%VHBvg50arMq-VtJcABn2Pw@&$m*pw$Y zgK>LzP#pXL5-1}}mH(BR&%kHt8DD^BY;dXv{bsDQ`+gY<{CXeS2d8?Ho8bI>k!ZwS+#^5-<|+o?T0g%BkYjF(oB@o< zyUD>mq21vHZLNFm^B4~}9jHsxpzekI$aQwlyZ3>y?#{eiwEJi)`aO;~bu+PlCZQ%? zJalLi3!qr%9Qo=N0NYycu7T7C0nAM>4%E)r*6n?}zO{(?8S8u2_t*N$V`?iId?9xS zU9gk=DxM$vCHAhL*aLL1-`VlHJ`+Yz4(u7=eBgnu@1fWAJ>~bRBBuui`U&JkuHNoa zz3|n>OCR<`pWvrHT6^zrZf2{U)z%K6t@eI}eCrkG`+cldtPxEFe}NZjegxjJXTO6E z2Dy6Xq%yIAwO_BC7x;ZseNgg}ng_7o<`4Kk=mUE?uOvTqx2JvN?z76%{!ER#P4L~@ z`aJxEE#H~)E?4~DuYIbHnA(2r;7;sG%(^ErQD?^Q#K(b#R4u;G&s@)AH|)?h2AY~3 z)gza7C;354?b#n-fQpe%a@zJPiPf`Whi$)sZI7mX^l@+Zx&hxkYvfI^`(>`Wz020d z?bU)jdkbqjh(ntz-qG9K`Sswj`^YuxL@coNKf(PKOwjPHQ9Sz%$ny>Y7#l(Ih?^_^ z_1FjB*a<&l*$wH$+}D2gn)vP|SJ-2pNgep!leq@;tB#rr=tta~vo^mW)rYehtGlD$ zgxB=3OR&u+y}p32zFxEk+nc;kQ8SuXXwIZhh^+;@>p7D_*T7B&O z2XF#GUMNW~-w}5)KY{zUxAh!4wp?dkzzz5;pZc$BUZCC2UC-EoUHz%E&m{I8`o}9a zP&abZJmP(gz({}&pzo}H*v1RF<|MxPsW0fUpULT{3pkJ0#qs=~!60UBgQG3n_eRdM zzJO`W9P!<=o&z<;o#*+*86+`#wAa5{H;6Z2PEPg1?$q3YeIDpvz&QMd$p7h;8sOgc zaSwY4#ajExtymD-fZMQzdAa7Anp|L)#t-_-F?T*@jl;Lb+yqsRZ`~c9u-3Z$j^b6m z#p27I*!E~}&(`2uqZk(*)H%tYfOYm5=Fs*}Z2hgZi%YJ0URT|K@AI#EY1?K51Mzt84Qz7O_u#_r`gtanBNWyZ?OQ$~K%o0<0SacFW)ATwA0%EN2%(P#Hk zJA0_JRqc+cMH5LL9cp{AXWbpKIv*kP8}fG@R*5;^#R+`B^{X%Y+t;((JIMKcoS+GE z89)uMn6{G;9NL;pwZx9uy)NaTj;^*Z=x| z``;Pd#Q)z8fA;^m!*754=l|^YfAa6T!@sZjzxl_1_UHff?{xm3`wadL&;Pr>;r+Mp z`TyJh<%j>`Z}a?r^XHe}{LSwF-`D&<{mpOwF7N+$zq$O)&Ohq@|J8r`zy9BUtNZ`m zZ~ydf_Wt?({`ddrKmI@e;y?M@%>O5sKm8ldfAD#@evO%sY5HIQ zKYK&u`fU9byFRm___OJJJ$&@V96UdNtxcf#Ic&fES=_PpTOJzQp9>dyIL}@7XM6Do zDb*id1P<8!{Os?CpXi|t&t4U;&;RASn`ap45ZLMoifb&et^I!4_CHdXXgjX}s&e5K5JllkKLEF=s zPCUWDHfQWi-a2jX^)vPta0g9X*RGAcz3pp{AU=pW%bYs25w-if-*-Ts_oLmgJ2~wR z?}pupS#PeXk-A%Q-iteA7+tfqtsUrj*!FR^C$LtIzFPPkx|*CbK9l=JF7X>$8OfI` zKwG>2KB~PP`;i;*(JQc(nq!Yx`$S?Rp4tA2+S)Tb(XpMQ)~|a$I1T?NIo|^{xLnak zFZ%?@r0%et`GBXk=R_;j~lgXgR`oBwEOW zuP@xw-xp`#6CY@K`r7->d*WIoZ+;rK02{M%>K&6x5pX1=U5Z5=D14-)_wwVw42C< znuBR9Ue}-XJkT`*&Xf3lZWCYpLf-niQ}Q*3n7oV?uU)l?nJ*+hcBt8VVt=OIdG2OT zUgo;{n?3P|G3{@A5wnl`cXSE%46F@I>-YYbD*-{QK7%lwoA@mO_3mX~b)x;u{t=W^hf5x`I@iQ@ZRWo)R4JNR6LqE|o;DCDmgyoi>$9qGsASPdNXXo$#XL~*W zT)pjYy)(4q(As|QTPwf%MJxGC+<6mZuDtx(2YogV%$FozW1X*J^i2G5#O`oC`&aOO zrfju;hvV7p?>peU-Qz^u9qrXXfKT`%SDy#Au;&FmiJz|Uf&J=PsSkcb;_g`IW3JC> z;4AL&z?ZY<*mr_BFmmzsGm~2}@(GmPyY9iW(X-{cA7k?M<#*6FFf-Rq#+>1Ek)v(S zUK3xtyDN6xF+9&wv(Ps%iCJ4>EMwGzmB`ggzILKJ(7(e=VoJo~?R(!bg<^y0-M<(X{|#mZSTfqkvj?jV4EL)*g{)>a=NT+bCCno!CR3u}N-V zi_iuQbl@4}Ia9sk5$iL{=Ww$he)T|SVjFWG(6A>lQ8oa>v4*tV8d~EL`)dwQhc|XDV4o=5U z@Ph9S&KhtXLR=&cqm;WAj zj}`xy*p2uF{BXc`@eVv^%0t&oZ0jn9Zu<5|iwTq-3l06FwfgY%_Zc$R&#Ur@4}4_= z+@IGo_8HW(!S@W4oIE)J8tC8#Z^!i`Kc2VeJ-Gw(%_(>JvCi?t_N>Ml+CINxPxIIl zbm0!hKvU;J*T=5@$KLL5aH)!)65|E5;{jeY}doxWV3pT0fJ3GIfi9mxBhbYj*_aJs_7 zuUX7**g?DlZN+ymWAQp#InyU%#e5bmmv7w!Lx0EUE5GR=oerQ zYusHJ>-h530X|!E#P|34xGw%6Z(n=XuJ}sbMSR?|eHy+yeTCn0&9?R*y{-4nIRj3T z`w`n3&oqhES=6~WSnuzxIXNAiv6YNcs}9rs?N@v02ev-)S_j2q=*;$WBH zXRMm9*i}oy+D_aV0p#4wqveRu218BG{vG%%Cc1q3v{?BtrW|uduAUETk<8Uo+cUn; z*t6^fXXoy4p49r>{{Y-2fnqOhwPLRmoRK(zuU6hxZp5lro0nJjk6k^eJ%e^#J6c~g z$8HV3?uQ+Bt1Oh;_6W*v4AmgO&ij`!u=iSTS-5tdm>!j%|I#;U#fvMr&dN~7RuD%Uh7+1?5+KBI|XRLF@;?E=j0^G^%GZL-%zNb1b zdHD(C2GEXJGwdOLu@A9*2G3l58bUn5<+^Nn7yF=Fu6m!DcVXQ830qzF!@caCjA{Ej z#-P^vK`eov0cViJ8ouW?E?-n{a^@#cBFV*ym5pELHvaQ*Z#zampYi55;v>Ik_iwbq{eCGBN_OQRTzUK|w8EWxAzM;u&ZYLJtM`DT2 zqz5?RtNlGzea9c@b_K{+{F%o*bPIs4Gr$@6VGOwAuSdPRT9?G__d?u#+%@p!I{OT? zJ3M;`^9@XF;{lWme9qm}KweNU2KMu?YZkuq>^}lNGpaAP+$pdFxJRP9puarhH{uiB z@JnAm-XGry`PS~!ds=TVdERS3 zVnp8mo~-it_0JFF#X3hv5B#Zr!~)+voh6Jba^0(;o$L32&r|-6fA@C0ufztqsWuPf z`F=HY3EDpHZ*O@Cyr0B3=6Oo6J@*gf!hEi<&tsZ#qPY-ebk9N3MG7yw}PTb0>EX^Z@QT)j-K4PV5ZK-_TFcu^X_Tv(@&L z*T4(dHN!l)K;Mb`dy26*^a*_dO6?X$PxSq2bmgnRzFbf}^W#0JJ(y#!8}QG~ii?ul z10%r|C~#)1UCd{~`FZWH9ZJl}j+jZTGe^2e6`ku+0JoWu`%^~yt=OvX7d^kdhSkT^k z8tcaMwz~=Ticu*(&n>n<1DlfHxT2ViUD>L~bt{=}N^keO2Z@E4r@7R5n+CSDVJy(Bc zc_;Q7zu}r(-HGN+KY_lsGJ-F!%dd0f^6Nb$ zp5~1q*67#sz@LfDF7+&pW4o(T@$XmvZD&@hJ`cM#J1Op9P7KgueLjBc?>n}aoC%Db znO8IM-LHev=Hc4w+jx6hy*+c~p=gT-cxuP0!K%CRy$ImUGdjsTyXMMABG>n*y80Q5 zkMHGQ;rV%=BDLG&h+!mT*ub0{qXPC1@~}A`Py=|6Ii1jXg?Ri80Y^5`v%fscD{CE%kj+4wWoIZ)F(M>Cvjn3o;88} zz*e4Uce-8IGqAom_z!Ra-lscxCgodt!UfSi4N>CVhOi|Ke@g zyGJMIPAB?0U)!JQ?$fFFTm#OpXy;d7bc=OXdndL#*Z%nKXsqI-lsM|oX#0&I2+#V4 zt?zyvtvF}PjOFh$*`tKGbq$1eF1w-S1iIb<&Sd_@{=}@Q9@kd2`U&J*;Ossf+g?70 z2JTnt;hfQ)(f!@odSUL4{{R#FMqZAy{Q2760HXgLb_Pzax|)gGVm%A{%IWwWyY2&f z^Q|2sS08sW=Ij8U)c944Rw7qFcWQyny>ztkl0$n3a_V`;+SfZU?jH3V;5w(HooT-U z>WuH_)fbS)(Vi!;J2+jB`(nO1;|)w;PFO#T=W$MKc>}-ZQtO;fOx&?=)J<%6{Yf9V zgFQU!AXobzbG@@hER5&MC$@a+-C^ekapxvEZFzz1oyd8=7j60r?=bM)L#`aT?x|MH zRXkUG1VP}OlHhBu^L4lnF!4L_+6VnBu|ez;+I>29fZ~w1pD{UQvqun-@cA_xr^h@chLiZ|*;PTiU+*KBK^H zAmRQ9JmVc?uFq6^$kmr`-xpwwVqA%YXB=&>2?7j!x!UgXxGtZsbk6>Z|?JZdci3Wm+y>3d$;m;-xI$34RX$GX!)Y{ie;bS zd``e^&-V|Q|cTmpI4Ol)J$)2`m=hCe`q7qQ~EqsE-I zjkvuNUA$cJAK+Dg{6S90ormp=Iw#tnPl=u))_pZ3h_~ySYpwl8u05@Ht}eiv22%eIeHI2YRaWBPVCajCzYZt19v|IXNt0Uo_k-< z;(WQvCwzPAi=;Nu?(1{%3ZP0pTx|`+0Q$OIoF)?%(vh#VmA=jgIq^9w0n3K zP$0M{tNjI>}yTUMcYeG-3!{Y+~IWM0UEK|$6BCmpFvz0^KKGb1ehX^ z9@xG=`|gssXR^24dX~&oM)Dta>aui_@W6!?f~v(`~+v9Z|G_K=qt~=RNO~i&vn)w z_~yU>N!&ACfb~h<-tx=^@n7H#d~43Ye#Rf5!7&z~B&cOS=P8lPm#00bse6qzt09B9 z^Y(moy`RLbIUjS3YbRW5CNOqqw)^e5*6cg#*o_)vwYNEHZp4Q6_v8`x`V8E|^VqZH z@%@f=A0@L?eTi80)=X{s8XLghMdaRJ*X+PHKk>ac=T{Exe@pG3?a8mbuV{L7eQ<%} zPUEok8}K<6FAgsK8?lz_ZYJ7(?k-n9qE%jPu0WmN+roIj9p=#;?PrYp+Q&2egw9;t zftKT0+);7A3=I5<_Rp%k#*A?7{S4&>z6yFM565&OCOBsX&Poy5A} z?vzf>UXkmbPS}d`!~k;d_P}Wrd^PTcta$m*WoI7IWA_;KGfezJ{DgfshF{N$n6>ild%rGioxYqwZi|19(h|T+bpUkPm`d?u)t>3& zeK{j?Jtuolpm+xV9A^K!Pd(S~PJ{UEx@zoi{XpBx+0K~w37&A(o!AR@L;nHo=h+S2 zfpgB+bq_(_e)jrI+&Sh?AdlP`+j#?jf}ij!cK8kL413Gzz*v9@ytm7B^|Y_OjLA3e zJZ*afIcq1{IX(lW2Vl=cH+euU*IdQ5;VAk8D4BeT)$>qZ>AT-UR1XyZSC8#sXm$LE!~?A<$vi@IZN zxFg#V{llK%MbSH!2NtL_8#{44EjFLhF0Xw zD6{nD8~N6Lz;nJa`QFz$-kJ#9>83pl@W_Gj=UWCO6`ysv0@$Dx(Tc^t{9)Oa?VlQweI%t zt7pIkgWT@@iM-sF3+Ce_;D_C+IRZ|wi{qV0%AJPEh~z>NT;G{0+#h ze};*DhuiV%{;{h@F0h|DdpUoit@G}kal&o^5M%aw0>$6Ed#rwP&cwyYwPT$}&T+;@ z&K?2xg8mGg7oa%APsDs~_A{mqe4oR>&Lj@hi66ikbMgmq`D%IIg|+UgMEYfMI@-GL ze!tF~aO~xsDE9EY6D<-na3SVhcC@`}PvRAGXY2s-g&cSAE(60Q=s_ z+ov7cxV4JAc#l2SoJ2p4*zRfU4!`HY!ygAvu6^Y=Q{SH@?{`PcS(9o&k6J+9&iI&uCbo{d_6eM8pI|sXF0=OXKZzyTOYo*y^VKt^?${WKgze7-^$m1t(;Sea$iIULTCVsz2Y=5dTW21`TIeiT?;W`NBxY>~+HwYV zgX0c$2b|>s-&|OS?&Rzv0$;8(%ulqo@3z0o>(8VIzB)Sl=8jJr>r8Er0N>4wxj z%OR%JbJR!>2TzS}FJsoxE}4+Zfn$1N%nq4&=D=4`};a@2uw$ zPi)0G7h<(%-~)33Zq(Ud1d!{j+Oe&;{C$4ymO!C)208)S{_48-T)ET-CFCzhP0wWy zbOO&b(f4h-EbhCX#Kha$J+Lbm>vPDQzGpeZsd>cYHR8@zGFIPu&nt4| z5>3wh&7JVA3v7FpjXyyL=^QY!;jG!#bP&->wwC9)^{rJ~R~kEw!O{15cqi?E{p#Mx z`MYSn!`Q}#n!XLKR6TX(w9WfGd_PXaomsmOcb@(o?M$_P@v#;6kaxc>PUM=W_KpVV zv3fW3`%dJ)l9QLj?ePL8zT(d3E1FmXmqR;y2iWpEQ1(9ZB8ca4f5#omb;kL~4{U2U%)3(~9@st$ zd&q0}#za>q+B??<#~EhA2Kk2ey?-A0 zdPj(RUhDPc+OHiL_!HZ+$O-&?{^B%h>?e?V*8=U?^5Bvi#5RoE$33gIE#J?wZ~vJ_ z!m&?3wDYtB-tD?<=MH1!D$o6F{*9d6ZZ5IGz%IT%_Kqih7w(P7soDC0O<&(9?_rWN zcCt75z?bj)^90(HyfqCw!LM=!v5s$jg1~l1{Q>#`=e$2hY9?{d`UK|Wm0bCXugJ0Y z4PU(ThzBt5vmS@v&`J-!0CyQ61^zf-zRpjr&uI5GuGBrnnk~=ziJxHiERA16kDkkn zKx0=gZD0@i#`Wb4Z1n_nhj2W%yfls$9W=FhWF%kUoxv0A_r_k^lHT7h|B3B*yGsY6SMU1_$Am4m`}8DU;DUw z^)iOs@n^2@pMfn$za)z-(8;YzJiDRq!uj_ z&Ix@7>IUprlIv%|-bro}t9w^-Kick5XS5zMe!;fqh}ExpUMpRn2U^CQY2PlWvrp}T z>D{`IzTcnb-Cb@2#lydlOQ4u@$IYu94xcnh>aOSjNYC}zYTdp=zn6zT^NcGIe96V} zdqH2m^4;5dZkbTxgB$pw`l#XGc5nO%S^)Mb9(nZxPNMhS`uqaOv%mc&dcg4x>Tb{- z=m)u~hu06j_u%KSwS$;ji1G@|R#)sb!^Mxa4op>2d6YZ$IyL z5K}t50jz(Z{r&Nu0&DlK5zC&&sBiEphIXED&r-e7HCGknK#4lnGfDIv-atEd#Oj~K z++h;0{%FN~^+&sVogG_|S6uW24Rpbt@*=DUoyplCO+d+gm#@D%s>h5oxfk|5aMlZicz1wiB*Q(vgFZ^s`8 zpwwB5zGrFZ1orLV0q!7j`L!!~ef#KZH!$KjH@bSG z-K!t|)Yqma(TcSv`-cut62v#ZW5B<_IqyC2wI@91bnJn>-%V%AZ(#2$SC0Ojxby3o z$wPV^JE5yC(MTj;;6$uiefPDGJ=HY>+E2!fu~P}Q=P>WHP}f}ZJGPwe{MLYucw6NtQsbzcwqXe)BGJF)78?K8HQX9#cs6JOgMnm)WB)`2|+@JzSs z>SqtnBVW#*A!p!k{)>5Ha-64bas15qdCk#pxp>7PuHLgw*R|u5n5g@P?|eTC{8?Rn ziDk6vpV$LiZT~u#Hv9_+VtyZ5x9`K4xry%RK`yZaZM=G*8?i=g0^{Vg^|FW^`|@U6Y=TI{+*;`Wi7;Euid-i7tn+OKWEv)3J6d2I8-nzJ>a z6F09OzUv(h^_fgtxQF}1u#%(^qNAAz>J zs#^f`1dUinJ9nV%Wj?U&VQ%0n_MLg`jP}0dszt}vzi4w#p+sPF|K@JwGS{5qW1Dw3 zrTXNWX?|c&wD%fL)7LmRiYf&C5zG0)SD9d~JsyH8?Iv}etOS3R(U+?jYc ze)xl!wsC7aFy^fL>*C5cmS{y@1IE?n0PLJMi?>$iZrBfE{&`CP{Q%yHax#Xv(;4h%XfL@N`p$Nbj;{ayOO6t22lI&?!2Rv* zjt^?o#rqkJ^jLS}KOg->z9oMF=l8AS@R{TMiSN19r!KSh4ayX(enO4%sJM%n>^#*pM3ux6+UNs?b?Pd&T#xa!1KtR#))096MKM;f5QI>1a|c% zF2^3i{E3|X0($^w-tmirGS?FW*rU#aYn}cXzVD=aC>d}(k02KK548SB35)OF{~8zW z(PGU`@^Zft8^pvUX5RWcad)Ua@IxQ?UR&qBb+_a?HN_*RcxV3)7jNh!dD5im1!y3c zf53qR2M#Pqa8PES=h}e<0tpTX2qX|ZP(VOHAWSCkAz8tJ0UH*QH5;Eo4I2&&II!VB zJ]RquvSD>HS>b%a9YMAG| zCuG+1P;+1NTdT(U32%LXzlk%pzOnE$c5<=*P$rv33<#_pv(&#vl- zBUW$o*8kW&jKezeF0HjbW6_A!IEU7ARsTO0XMX1D5%@wb(An2}_v{JRiR<@NZS`;t zH680c{pWYih*e+Dl>G*E7vctLA7c9(x5UNoz_{!k-=zi4A^-Zvet+O>Q+#1*l=cT}D@L5H)(*~P$@ zYuNIQ>0M`Y&sh5>yne^3Pf!}PvCry^0dKzMg};dVinTNGshMi9_G|bq>*Ooe&y^3X zOFLq{XU6VPv%)7ZHm`NxCT9$aCvN!mk#kvpYORS_zXzGQa*y3-7?W$q;_9D%zv@Ds zrL~{)lw~jP%=Pbh@*|3_c|PjUTRX}1T>0xAi?7CI-q?Emf$yp~%B|YK@8Chby1F-J zku&G&i4S+~uEpMG>(t2Aeu>d{1^)x~oLD~#)dROa7l{q>&dO33i6dh7!PZ;?kCr<#tCY1o~!#L{|nOKL(iG4 zQ){hro^v{z_t33HV?2P`=i%@>IQKR1T|L%5+Li!ooxLYEhy$D0c|d&b(3%chBg=d| zBYSml-LuqO#4dSfn@Rq@`qwjrAHdJBb^7-G6TjiRm*;;amQy3=QfsfkMy{Ucp1-`W zJ?$}pymv9wVvEyS2dwYp8l1Wb2B?0}jhyjFT>J#i?|X0jrZ$%~cb?T1z5(+Atgjq? z<F3{;;|5!hZFW63XZ5`Pytmdh0)6g3YjbbzXRUqR(|iN=@my+) zgHHG^)R9xufOhKB!&)`2mt8NHxp}ybt`ny|^9eOBdsTdV;s^q)eXXnKKNP6F2fBmbI7biFLNzcIN8m_k#awE!JA=I=CD?eoJ!qeJ*U>C%v3~k`G{= zbJcA4rTNF|GyhoknAs5j`}#(%Q)m5m{PX?Fo}T^Gm@6$kN3M6lvx|<_8U6c@GZ{a3 z&U4&={Yr;@A-31d&0`bZymJR}&5yN@y%BfLnuYh;Px`%(@1ACtyKk<$0`BCFOqMo^R=5zIU#kW_IpK8F#8QU}D z1j~FTmObSM=)_;?a|Htg_#3!J5(D?JCS%o+D$aVBy(X~EGuD~ly$?0@Omf#?8=QSz z{@&f%I$!R4ArA0L?g<{C&fQks?T=;8+>JVWzQA|v0JX50pZlBtv1{FLm;;{uc3-uP zRp;f*_N=+E?pb}4d4Y8S)J#zI#v)ekLQS0=s(5?knlsQ0aA99_9W?Sn-Z>|>-c9d6 z_r7ZIt^Z4nJic=#zB=pM0j@b?t!EX^6$$S9`v~^*vEIHNoo8tHRUa5)>s-fv_=!g! z^%=`{thzznZ(!h89oF9JoaKh^>iPK`)mc-|0L_`9Vsg!zT{mLY)_vp4XRdQR@vp$% z4cJ#-ZDQ+=W8WL0IrkdaH|zdEh%La+o$J^DrN8g9pmohGQ?>RH*42FUs$~r79&-JR zX~YHM0Pb!6mHsZDm(N{pgapV9e7R2Cu?fsqAI|E}pdg;P?n7%o&#$&0Y<^(nss?-3 zz_-@i1X}03F;h!?5OCt3a4+&KKd+ffZ%*Dm0~nW&{Q>Vib?eB99%!_{<$jI)MUBgv zT)OgO=})co(>&0gCGQ+J>)Epr2e3}AwD{_TeCoTuD?#jMao-wZds%C5^Xlvu*otGn zllGy`b9nHsN#1(!CrDy@=JqqzJys8V@4)XZ9si0i-*eSJ;9Y0G>P7qm-4ncN z(Oj^e!@2F}EK^S4%h#Qe4*Z$K?sH%HK=V`sAD{tyv|Xpx*k0yqUiM4;D;ljeSNgSB z=TqnZKMiLy7w|55|BPzgIcwRV4o=-b;~t5B!&)OBnFe9{77@uJ_FE?=9B5 z^osA_53L#G1K(cOcJ1$rSFay%)f=nc`Vnw@S>I#5m*x`G!p>ah$XGS%8n~#F!-@bc z0V~rlAXbe%Iv60}gnoh->gt}D)9-uMcn-gJKY-`F05yLTUxAvIYi9jdatZ$pYwr$R z58~_Qp5!ATMWA871~>49vGWM?btiE3Jm}&5iJ70;KLO+f}pU z%R5WGYuFdi;Owj3882}5bYJ(fW@2B^T!DT7eRI|%@x-^*yuFOI!aa@6sdtUwO9GaA z8COm#a@oK(;>!Py&%U)czWpY-#9FIXovSB4S~V_b^PT9i^dpzvnnd>s)HmW57@)Qm zw(clb-;Pxm_S9$jOd^SyL#~q>vHZa5Cr~3AsCO27o`X7hdd)D^5qV&YDD?OufOme-nqT+A;<@) z`y%JuYOL+R+}ytJgRylk@2Iq~_VY~6mPr!Zv4fn;bE!S|((pfWr;nB%CuhBJ&4zW~ z|Ng%E%Qf@By=p%+li0m#KjQ25y&HSYK)Z2;Qy*v&Zs0%ZUwMprpR2sOdWVQ72x9Zv zCW83B-(%u0^tqw?8#vp56Z+~#eEvRTz+KhA1xRgTf#=YQ7R1!r_Z8Ud{BGC$gE-~i zmwv*XHT2^|%m27rYjVvH`mS8_Sd-u>RmI|9Mzk#YV7i&M`gm*tb7se9=zUy{p5B#gZuk&E5K2)sb9%|e_WA&-t_#+or zcdG`j?wCG5!3$iVneg9<8?nne;n@@WLmnGo1QwV#cK^iIbAor4#L9U#d#V=`G?C;$ z;#)WI&%YC>w@$wD#DlymV)6BiIB(s#+^%_Io!R?*1@3e9!>d`Ueo~xFySHv=DxT<=F$N33>Yzfu1a{08h1 z@XprB8Cz%0KSTTvFyL?S-iiH`RfqNW)a4^lm%vrr)>~7xflYFt_98#v z8o1I+ow+ybuyyZnYLoaWkG5e0l!fo)-@ps-&jPPl^Ed1uAF=vRYy;k@ccQOV^Dw>> zgGuZ>fpsRozoZT~t6_I!@iwAgz1E3x?|Vx4u^$NPh;JiPn% zKUUsc^&#&W)L84B)_VSSfM(!N?2Vl0XcNc@mo+oWt!L@p9SqR4K^~3V1oa-l$#?Li zuXmdGfemdgU;Zm-z&g(|$T`a^uy%qAXz(-EJMVk%oxcHV{=$0qwa~oW*K@F^dswIc z9ZlxSsd*vx4(xr?NB0AJ?pUo{0}puTcmR7n)xlq}=4YsKft>_5bGb zhV8)blP>de_UgIj418y6@B>^y3OMJOfxqFN#!ck%U(t#H1J2xIbM~;-n!vxn>pQo5 zbU4px&VBWrzvDlF^=b!p5ZgNhv3-nP#yxP*bo?v!4NQ4@e$^*GK!Dl@>)h(q)qN7H z(ZAvQEVXKmwRRA@mvt@n^TJ-#GOv5cPf&X$Gu_u2>=VSkOX^?XT_63EAl3|Ti%0%r`l)*M+|J z?qJ61(Q>~}^hr4D2iE0YJyySltv%Jj*9tg#y@0@XX=g5t?@7bAM*^RN_owEDU$f&s z;LcJvfM-cG%{Z|jcI)eCf1Z5Mm{nB{`V(%)+dDq^lniu>5XWsoMR^GE3dlvD-ILY1jjO1K$88}#L zCieQiYW(-c4rk8$v(Fp&3j8yu+!r+>hcT;aw7L4&Z{o|_|GXb^K9dWu)_lh%I7@%z zXa4x=Cvau@Wz`1OIAYDLPPk{s&R=4m|BZQOwpu|>ov}KXzVkP%^Qj5^#8yvdrpCNn zod^9Ot`&R*RmXky{|p(VV!?w^WF0P zwd!3nS0B&V@FP|q&ubrhdoKN2#P&$6`(5ce$|lo-({Tj>=|&4yxIX=a_TF-@4JiD|Am~jYC4*Uebe80->q82 z_Viqt>oeBWv3H#nu8|ks!{P0naPB_^e(i%_I{L{q;-Pg#eXV^0-(J?{_MN%z9{U7A z?$qGbU5;3HSwAC~LA~?-4SvyHL=;|!+W#Mk*gZNLZ_mMm2Y^8ty{2plRJ`?zv zH}-pQorAi{Sx;^CMdu9eZGZiS@A4khzgiFe0&m~~V#7LXjO$tG!`VOZYd$R)z}|^9 z=ezS?-q+d5RIbBY>+Co3XYShPaA;ii7{F!TTK9Io+LyBrV%O<2lR(6RI^)x)QR|8* zv(5o$AMfWI`1t?-fu#Cv?I8ED4_v3F`r}*oFpuwi<#WzN6To?D27OjWnVQWS`xclt zh;Q_8xkm#>?|$~5@~jbM>N7a;g=h5r)Ax=OD~!iMwwlTQSM+}a&u{*#7T$Txd2YYI zbznY$bNIbQO%S)unU7qMD^Oo&BTjPX&R*b~F+F@A{5heY`0A#JHH-J*9Ig?ce|D^Q zh}eBXn}F35#HxK-M;u^mn^@P?zOnXEW1rL0e$F$o)+W|`r)CnHYxmVtKj2FPU3Y`l zxC86Yx%G!U*8R=vTkm>m-#_0YIcvY)*KDuk)s9@e_vrm)EY8nd%|=}Ft7E=87kwJM zJ?!Z$+5_VzVsZ8iP%+lnGj}bhy4-&zrDx4#4(@+{Cyxy<$ayD;KkzlTqHT|p<`;LNum zW)AV#a}2oZ2Um-@w7F{a18!i|Hf%jt_MiB|GkFi)9KKw}+9&br9V6wt=bgJE`2w8d zXxyiX{6r@f18xHQyXqXdxmdC1I6Vg3h4|Fj!~LH?zI4loo0z8wp{AjhzpJ!&D_@!qM`@OK%C2^h}bnJCg-_|hjV`CGj8CvbH>(tPwshi#gTJv*V$vn z(tdegf0n48xz1n@bqQSd^Z9gvpHcMZ`wjmMJj|0Dpla@GzLtxxXF#dF8~Z-dN8l5` zSj)a;*~1={lLyw+*|3v%VAa%JQs;S-*qX1_U>nwa!=K=(CfECBKRN4M^}A-zma9+0 zy58uKvECnhMXqOgfp^(w05u)kwyoay)n86O0oHS9L!Uk6C#XeV@A3);Tw-VJdITL# z-p~B0>E_`MeAhg?cWlo304)JazxG9Q10x~t>pd(Tq4l0o-O*N0w5M0%Ph;Sj9_TBF zb!F`5k1yXr^-PHS%3CY<0_;)m2;2xXaP|-ZMlPG%PPi_wg>$JB&g|#%N{_HNv3s5V zZR;AG{j@!>)L1X)ToY@2^ff=d?B{!C`~d!0^xwb@Kd4D+&U3!r;owi)&Fy*A)(XDn z+Dly+v#Jl;%=LS#Kk4I)li1pt2is!Z&v)`q@Cw}1I`^_>psC;c$g#%*uEXj3&Zu`? zu^qUF{atEiaOOMj9?*YcoxweV*gEx3ki_oUa_yt0#j5Mr0m3-j;;~osftA-bYy$a? z?YUxSsWZv_W1ZVR`sS>6Z}kH>k1LZevF7pJ2x2uag>?DOZYGjF}m z!953l=HfcmI`<#=^5*T+w8WnMO5G$5V*7|r{AwLA=Q{eYW`F67^U{XC0xqUC-D zaNdpjfS(|>N9{p*T5oV{AY0&S7M=d=E`ZUxxu?4R&E9B9QH|gbI#y1h+KZh2JyhU~tIj*~Z09|)UR^;QP~&qQ_|`hJXZrHK z&e-uge1Zn74WK`PORna0_Jkl#IpC7380!q~p{Cw}ZEyFL3#>DXmp`aAPmOn6cMUgk z^$T)A+(6ggXjqYn0TW;A-Mid3150dAS7~Cu7uXBD8u?fJ7x38H0Vir-dO6?B)vsae za{{lv6Ax?eOJ~mk7udQpE%kB@|6+~(@YT9Uqjum&uDi2m6S26dzjbwY_*Y;bYt>qB z?4E(20^WOdP7%cBDv$4HR^FcOQL_YJC^&)zsWSXL6tF zzvnx1xQEMJi*+xN>l}&w0q>dYSLcmoeFo~(cn4DpHv`M6yWl?5*w-@-^3H3|0cY(S zzV#36MISzZ+{o2OT?e&d+uY}24{Lq*>>2Pi556}u-3SzxijmzqekxKXz`s{ z%@g|o0q0(2WA91qp7!uw>>*~rVJoLkzUtuYXU-YaM(q0|Z20B}*7ttkTc@UU#Aj|b zvEz>au|IwS_o#bmEPb>SxbKXmInT|!T(cjup4sf-erNu;?|)N<|Dl##P#47ZC@ppv z$G$JD{DiZwz0{7)6Rj1{xq5Z%8`{%bt@VxlPY*vUA?J!U{{_x7cus3wBiG${pACP; z>Lp(bo50VGocn5fEUVsL0XMJ>xX&d1j;(($gEKG8Pf&9tHE5jG*k@#q0BVdgSHBnX z)<57}BiFq-mm1-+?%Z=(W8h5o_yKyZdO7O`xXOF?tjOIj);#9l$Ui{C_s{Fj=qx^S z^X`>!_A#gbg8vGvsabY^@7kIK>L*s*@O!NNs}8>bYh5GO494EgLp?pkPkitD0-Vo# z`3rmlPw)cje0LhrXM&b%PWRR?#CnInqCYWSJ)$^Cb7%@wc@kZenH%TT7CN%H>~?4{^h>+%*7A5lk57_T6ePs z-};`bZej<#XAPiU?&N=3V;jiSO?pe17tCqp^(b(-Yl>y^5+DpwO--w-c;QJ1@&HHx(InUa)`#cfreg-=8xit@Ca`tn% zzxlthK2v+xvtgZi#`3MXp|QWcj00%v*_p$)#}(h-`$VqTGssQzEI)_V8FzeZnm%U; z?8wje!#PB5>;A@F#M(n!Gsuyz9`sG~8>PW-@TJMm&#GJ_H{sO35Px~!=Y>=MytmQ} z;##rz+85$&_w{?XTU+oE(xt^z-cX9*fuE3nX8+W;%vCCPmceq%6i`-voFSO<& zSHD32PtbvT$PeHiF5?+{F5<-ZeM|T*nA7~Jl{>X+d{)+|dEh&%8uzS)4PxtEf#xT^ ze?M)x_hFCfrGUIM*S_S9?fC}h`P`=oR*yAH-8sJYhNgPuI_JPTZ^s6jz?S~rs?RGq z``ODn>&@${ckjBt?N=>1SGR{YR?UUlk<=1Db|2E~GeMs^mljR+A!qHAKK}PX)*65O z-5tNlQKJoX5ld%}#6O&TXU|&r4VbT5Lhs2rqlY82)Z=F5mUP0y1)iX?Vz`e9r>{rn7YsP>F_VhilW)ctiTsrf=-XBpdKd0{Z zH}G85lRozFnSUX#<~&#Teu7%Twi|D7et#dt9gJjatUVgGIQ+mD2|Q2f5>DLL-yfXE z{XCm5r(2QYU&J3Y6qt#>uic@}H^cMa!s--gb<^7aXR_8M6A z=3STVn^R-0`#*vEy3EO$xBk2{4XzWb)4Gq^^L!@p0Om!Y{{bH0c3=8-mU#i6=8|5& zujlXGb#lf7D=LpBf#;F4XDxi^xxkzE9aFDHUVGlVhVNawI`&Wf`_h;TV63%=P*>+9 zI;vG#U( zZZ$4_`?$9?_J0LcoAAtH-gpoXY{%A~_!G#R7v6)pnS{i0_U>3?``vc_)SdHM_XGS5 zGF$br_DbA{|HO9eMN1zw+R&mqOT9V!Igfg4|K)xCdCQr^(6V01sh?p)iueg0K;Pc_ zGuJHE)qce8A#bcb%X@9Pb+OKD?7nI{w)$WrTk!+yc_y5?NgVonhKjfEpNp)M`zZ&E z$FcaKNMl8;`daT;Sw%tda ze9twf{seW$*tT(DZ@)LZ0PjNmIC*;!TkC%3tg++LbGK(@uIJ@kYRo&&(Z8{W=N-Fm zqA@?QeqVb6xtZ{+b8qp-dZ(PhHFD`%EX}~4?*SeEv3s65tgl%;A9?q5FYlsyCH+9< zNZacL)cx)nd$`X;*Rb}e^AXDhsCsn8BPkvH6|VBsxSTIye-FZ1C)Aib^_}>YTK8&L zYicGfFcuADe)`tQbucy(t7ZaueRKLxdVA)E-;)1fPj2FeF&clKSa19jKk(Ie;?m$x zd}DdF2|Ulh*3TgJ$$VIDvds=&Dt$Rh|>ajEhIXTk3ZJ$zvO2uP3=dWHQw{Uc5}w8;>+3ZVH)gtg6(qejkQ<8b^Hr^ z*~gjeC1>p3-o3Eang-wSeb2p{H!z5;^Sy4se#NQ7pICbbHs$VXR`0PET*NwazzOvo z{)X0BH#XP60AJqW$ouR>^~O!~>LxfVa`mb{-o@5@++(ac)!Dm&|L=W& zZt$y~8t3{3Ts_wQ_DoOW?nO8gL=l7st9$wwR{`zdpxo^idtaJDo ze1QM;eg9br-{)X`7xU%;<@sqb)}F+qoqjpFcc^0|8wtFf1RJC`|s*ZV~0bDqSm zNcM`=&$Cty`GET`=x=a|?;I`n`LOnrujhe2iLI~m!PmZU>a2CXdcTM#`FgH!gIEN9 zos}M1`(5E$tbJbK)zutufiKKg9lkZb7xGiYvVkovoXazLZ#5&s-;Z)|QKre^oK^;kV$@$GHhc}ChdxPiu6_j1`k_hY{HW`0*`lUnZ-JkWLGZ)mmZ?eXiH#=F-A>w3jnlS$8jxC4Ci z*4Mj)9dKuVdFQG;zItn~aL&^3|E6BOngnu@`|l3$-kZNyDh<5P^62cR)}`&>mHg># zuhN>wHy-(!_dJcbG~`_FBd7M*NgTj<#Cm5FzxMil@txTGZ+pJmeQ6`$@Vy7;x{!BY z{fYj}>iyMS!9B=(zn;B(IQu;1LC3dF>z*%i*dgMRcZNx9Uu}{X&e(`M)|xcNsxx*E zHRTgt@dMsH>$9W==-b!)nIYi^ocgL$gYSOk^t<4^a&7w^U+#u}z$LL-m$AALtB1a4 zQ`5i%?)BAL;IfzbzB%u)CvrIP>xx``YbO1m-yJ=+>gZ*k5O7+5CcGNo7uQ(x=r`wn z_K@xt}xE zv)NNl8(3kiw(GOpqs6B-@P+vV&g_{&A6p>rsvfy=_L=z2*#5C#Z=R$G?H{;e2N&*7w0Rx_=x?x5^6g5VxDD34%CX$5#hRmI6OHp(lg41;Ppo&~ zIXqX@!K-yS{|)%NRPWRL6`g%tk!9BNmIn{;3eN8zN*CmQU$BRD!o36h1kTXaah4<} z_Xf4sJjtmKa7APMLoMf#>%^|o!H-zyG`6Q_sNUEIu-D6d>71{@Sz|BzI%@*8 zH6!-R`|9tRGZ#9(u(s|NKfr%S`~L!0%?6jTG-qbdcY3Pn#A@nU5RZ+()}n{+p}aFY z*ND~AnhWtHwuk*rtbb~*=t_%ypm&BwK8-=?xKr&5agPPh?4I_nr3I1ed9?TSE_la3 zK-|g0CHWiiMEi>M-(y|gnR$7?x4759|8<%BQyn%$u3mq`)g9eeZo=0yB-F2%GkAXM z8r=1LX$HP`*s*W)F+YIy<{G}3;6gr=7(DQ6#WL1g7g$%s>fxU94ZUZyN8K^J8vUt# z-?fR3S)W^b!`f5GJJ-ki5_?ZH?qRR)zGR5kaxyGG8w+Vk_Z6Wc!oyz}_kO?l!W@Lf-M-w~~O`<%Lt z@BUrjH>`PA?SIEIclF9ozvE93K~FG%oV9_y(Z{|G|9^X5{tK}^)g`FAATD1m%QHBq zb@CDTEbHWI1z&-4^;~n*djh9^Y<%>hpM2z=Y4^AGdZ!2QdHd&zg!}ali0aQ0LEe7f zzyq}WoX`ENvA_G+x8}w79uhm?%vn3J{(alnrN(~RiS4oGdZ4+${RPxpo8;805pomj zXZQvGD=@DgaNhBE{DH=G#dlWc`|^$@ujUCCfzPtG=Q^*xy*<0@RU1ox(+8QWUx(Ls z-T<$6@~q0o($w#q6OMTP0%y2bM@~G6uRxu94>-?O_u5viGph^Gfw4W*He%-+LCzT4 zuz|Jb8#!mMcZjziMN>lYim|oO}lfCJ0b7FyG_8dYbF{|5!7-r+pgyKk4IrSa(C?922N@$%`HY z`e>^b?GOA1a9y;y>g`eIf}7xaU%iihZr#Aj8F%ar_)Lvm17F@{e^Jk4&sKXR$=EYv z&DXu+J4fWYzr+gf#C?1Y-+{FaJGbA7|HiMqxTltDo) z-yrvjHBJz~C0}}SEmr>ycd?c|oh^w2So=VyZ;iP#&m`yFnA0CvS64@$h)?vS4sC-= z(6@!DI&1Aa;QxvJ3BCgL0a~sz)qMC9&NKelnofLXS6{xCd98iDV_EXfJYwBV=6WZ- z(Z`;ZgY(QaFIw$|ILTK(tn+rz@I9|_eJ>!kXFUh^!VfUPlRo;M)4c`=U@Y|gJbrm! zzYBW7cV2Z3{|Vf~wOIJ6Lj=F*8YL-9m^-2utyTBF}BC`eeGrJ zN;KX>fxLaxzhEN~;FlI_?-%v>1ANrL*R#O7pZi)HSnEC4&}T&~&BS)_6W$v2wXje0 z*7@!{fPFe>F&?}YyzV!tL>8D;7$GwBCfthoaF zxUc6_Jh{KAJLeqa?b+ZvzB85Xmzrsg87Fzq@Za9oc?Wy~m;Idkw(DQ8t_Z*EdUHOT z+`mV_uQ`yo)Y?z3@Q?MJa`Wcpy;t{|Xtf`|i{JepSm(5te%(2>*2`Z(os+l;)(F19 zk6eAbG4R~#Z`Qr<_s03)tLFYWOZ-w(wZyd#wUJxjR=-y`_iOOZnSlnlf5#d-yL&#d z@-J95ZA0!2?#$BVS?V)z?ndi8$JTwv>Z5N@QMK^a-H2PF0j|vT4(U())7$s+3R7rwk_sQuur zJ$~PDaqwyzoOk;I65pKf(ZHX^tO&oXyYf!;djhP!%d>m0axdHe3p*1DFz542e0!*O z-w1NRx*+!N(%!pT&#P`E8e;1@acQkVXD;9#aICx<^{^(3mEp1tSl=Uy&*$&c84uHNq7@c%Zq zIp1Ho7N2_Om9t;qPyCLr#<^>DLYErX2>c`m&iB@;^UmPZX)pBa_zmBmN!kJWecwIS zUiNT~v!6BIf2f6T*1`>-wa-;O8t=*Vli1z~>h83xsz(z+0ABFjU#+O~Vf|h5O&*Op zm(Q=>6L{yV`@_$uboKQ9JMm07H1-VQ!1`P!zBxI2`8}!QKd6&4_YK{@!8c<2$$594 z@tn`utS=7U{Hag$`U&a|(cQ?`zT}2HXh5z5eP{44YH1V1y2Hd5bw6;P{|yZD;DKMW zw57wWcWrB&Row(X;Nre!c5flCMmvayb--LRj)jZ0_-Br;Po0ogo51)5Z7pIs_sk#% z)Ropf{y~mdSU0o+UcGzjdp>ib0k!7-1ny;xnluNd@q4THK!1Q1TZ?_7t={wr;wcjD z-b$y|JsO<8-;cZ#(FDu6tTAtW)xbT-pZJD<1@;p5vCcJuFF-4LuJ^C_`{v+Z@SUsn zYLuwmJ=!n6yl1L&VQb%9wSiTursbNYdJxxaaOM*Xed=AOpY`giK9=oRYX&$yUa9*V z1n{$Doi%FpZ}(-dSo_Jl*Cgj&39dlCEWGvlS}}ktgD)U+ea~rVT=sU^|GBw3Ya-tm zSnHkJdiy*3#Gmriz7XHQsR?*%tZhKeW7oL1{#Wn{%-hFU-n{yLU-J!e_H)_qg}kvf zr@uYCLuWK+y*V|B?c|=;0r%+e4ftmO-!uKw!~GxkHJd%;3ve0u*opm)&V7vKL*(M* z^{Wnk0Q2?i$<6JLY^6!;m42ni)S|C_W@14LtoWU~e&;;kD;nSV`rR-Z&!|>dcOxFg zd!CvXF5tw#_x$#(dw}!&0bE!8So0^gZ49iuQ2!h7cQvl!5=XFl;DVoM+~2+G&T_@p z%T4S+*MM=&O4xIqwE#+tN7S`u|MH1JEy+Exrh69czZQ{IBVUf z7CDz(CV8xVw58V?v;PD83a-HAJqVxo%%W=F2z(HUy|2%sz4YbnGqL*C8fUIOIyT@% z7tCv|y1=^5Gab2g%%(n1jD5YY_}t~`V~=_!ZJl$*`gvw;osqhUZ@;hX+2!%A*Kgnj z{sI#If?lpP#DVYqsMR0%?)Lzb*go?9nYkG=d*6KBGrsv5>lqDv*UZ%`f&iR(a>t&t zCI17~bl`H<((d`4Tez<^BNo)^*UD9IuLu~y>alvAxaG=MAN)za(bL|?T2p#>^&c_5 znuvwhKi0cd^HLw=|Acc^?`5EK9#`FOl)K)|q%Zi1&N}@c_@3K3zo#WOupR3&aUZoo zT=%1OZtTeQ8GBVjtZstphjkuf=Xl#1`v(4(_w|`@c44mz)^)?a=*PON%)KYJff-ox z_PyTm>1Ayvr&f&!Evv`kJDO9Qa09rP%U(TKKhJNj&I?!X9%p`4BaeUX!1vnEoLbkp zqlk4r`_EkaPdMLkf1i+W-oeva_y(#keFnaD@&VL-%tWkCjeX4<2fnLfdi2;leHwdu zPb1ekJ*&T;I5S%_K+`7sVeKa;^t+(%|9EFMVEjU#5cnC(x>wyzTlz+>z9*4C=j-?b zSZ5#OH|ywM+|FI--M~cSs&nV6wdVldr_Ue)0Vr#Xu9I(C?#G;Xh3i=BuOJg1$km-Q z!^A)5a1S*DPW1b-C{``D&h^V4k#L~;!q+|1$GzJ3+vyz{cnXv!jbd+{iX)UHBfiJ zK9BqQy@a|HK_1RtU7w{tf_m%^G%rBTb6xSrjvH2u^QQTr2e$4LJK>$<3EbcJ;Nz@n znP;d6?S21y3aeitZKsr}tL-xQ8_jsP9_f9UGf7j_uboV4nu3?>g^F?H{X7e&T!f|NKsX zJ$-)Q+pl_Xe+_>C&nN#IR?P*xfZ7wjAMAbY_aA$>pZwgl_C7NQyglmKU~6G};(_a( z99A@ZYbXAdUf;n_;CngXe|_$A>*QnS+;e2B9v$vRZRMPuSl{mtxiN{|(;A<-u{r&L z{};Y{8e3a>v`^xOHMZZEcl20ib;)@K_xC;4ciF>b-OGLPk?Vf!mx7%6vcwna&AB4h zJ7UeipV)@=+}@YIxrYB0n0NLLufB4mjRSbL8~MzggR|CUt$iQFPw+t3;Tp(T_4?+G z>+bQr&p14>bE#45zHi%if3;WO9a*pDf^U5x;J^!<8u#%$)|z*I{SKU4+Y`{sT;jV% z@C76oi4OvNd0+b*+jnAn?4HznH$#rziFMwA?>zPj;tsFI{TtSLxhL_&4&nxG*6g^> zO6-~JVXZa#V`~EYfd2{PTw34Bh`m3y_Qkrd&&YkNA8~@2>ls|hP2xse_k*>M=T`5T z)S6epj_6Yvk< zJ=x3U*`C0C0yJ=P#&tjO$@QK8ck_|Azk8kffgeD7@X;?@=}&KU$DjC3B%F4ynkn`f ze>sNAX8`HN0p zY`xD(f5QE$8^kF`zJPIlXHzdXcI@-(`UwxL)%saFHQLjs@`P$;ULV9_ zteF$uS=F02=S+zo;M}eGT3C7ash(JuoW0bW)A!uom;FZIv!29$|E{~ZuNivovjrBv zW1Xpb;oBqN-Q)Cc#P)T$Pv(k$&DWmt%;di6?eQxg@b2mTckEZ}$L9&H`y_VU$z%0P zi|zp$aez7x_S~JkORe*rd#!#+BG%dE2fjU=<3T+2fh%%o zz}v^#Go##vxb`Mzt#QMD`PJ?k_3mqZt=RprwGTS){oLn-x2Nb@Y6B}~u6evO_YC5J z^-NcEuIia9mjMBVD~YYSqpl39S8r#^v`j z>uOf?&S*dXt;t#S6V9I2I-kAd`;HU;y6evU`Tm=CjVQC~_gSh3yw=)|wSU9jc6`yI zm-9U{4*c2!U*6So_4uit`8>m$`8%rex-xA-u-GW@_z$!#y%7Kya0br$+^cXm`R;F zaaIs+xR=_N>+bC7JVAVsC$4i7+c&Z1oUOo2b#8J#H{Ss@4coQA`*1&X*7}an05^R3nefCNc*c~6)2gqzaciz^pYStQZD3vY68`+O_n*Kxl4LA6 zOWfDlI}PXRACHmr*cxdadTmIl85%@J#SW>#ZtufW>>;jHchT=u-eseLgA zCjQyadi?<=@bhy4Bca~`)?5>b!Cv9Lf8&$u@YYx>I)2;S+0*;Gk$VDbC;7mhdoe$W z>(5WvH+3v~2l$(KBm~R*P;2adH{uQ&e#*!C-D}4`{q5oUfhNGm3N-AkwgL5XbJiUF zKdJFO3N-dN*MT*{xB>rN!guZA9>nVH`ARHYBbVO0>9OL3Q+Jiaw>CkDt@E6dxaHE8 zCGNm_q2Is&xvjBhbX|5Y=Nnnn`iyVidwhbrQ#1iD8eGKM*XQ5y16#fEwF&G$u{Uh> zSIgd4wLnb=nft$e!m5)qZ=L<^Vcl2a&<8V@wqow8$)k}^bl#_&P}8ivuiE0b|5V3% z61(3k@cGLB06o{|rM*+XW9!{VO=1UdkGW%O8ZbYxBZ%+oeW;u<%XuTQ%r9>b(Ko#J z3q4wXo;~r+JLAL-;(@hSqOYDic3yM7gI96^Jl|jP)cG0LzQ8;C4Y*#g_VMnWTa9~7 zY!j@=)w5&mnX%t-S-E1xga-yGuV3o=Xii8dGDx^|3-Yy za>2iuL;H%Yz0h6Z6K&OT4)4nT?rUGqdtd zoF=~KcFua|wbXb|)=tp$S=J5mVT`X%7&}J*&l+&%zu{YJ-vc#OgJz)lPV6lDJ`?ZQ zT6O;XdxDAI!3}tpk;Ge*#A*}HHRJOviT~vtGFQ!;S~z+4m7nq)*FD&$kr(zm zeS_FDOsq9G`vcD||F&aw&i4vjGnV#AzT?|>U^~`YZ6mgqODc)Z%z zIfUm^tHyp$^aJ}sZ0!W@Su5B5*}n-i^)AQy9R7yRnoex5o~w7=6Mlj#HDmMc;Zo;v zUuR0}g+88f=9IGskP79=RAu*<7e+Dcu`AS{n4ITtLFE$SErA>`?*&$2B8lI zQ1558?t|DFFZP8OPja#AC)Sm*dRQy0^?n-2TwH@!=YL=Egm;d-ud}&-7vu++_zmcr zbN|4%W)QpNTP$ANLESgK)!(pRfxo}=`?L3HO$W7N*XGiBcg~%}t{}dF83fv-mb-8U zx#~rn_`kxvJ$vcNb?S^gbK6>N%QfRS_)dJ_o$1Hc8+(?@Q)AsApTP5}dxFC`W7Wbn zG~S8xJC`+b_MF%buK2>f0ep`p_9@32v46H;6PRmQR|H?~@JTQ3#lF`41uBNJW)O!Q z7{t!#ncj?Z&8C&V!r5QnnJ;RH&DX+4eEvMKPXf8VbBR5F-wgb72l{FzaF2#{AMK!r zGe7U^|BsEj)3fr#XBKnby-UqZfE}RDWG}cMYTy#ItykBLK_pfl>pL>;>u22+U!A;r zCjObzoS&gDKrYn5)qSJy#xCn}-y7D|bH7WR)7}B>{{m*Lb9m>SSgpO?BeDO4tDbVy zKi*gV24A|Gu-qyL)JF~I5x4r+3{yyh` ze_(x%uAhG_eb3zkt8eY;(PC-Zt}z!|(^#H;ps{xX^V%kIy|V*Nr@q6PcOI>Ksr}Hy z54h@qo#cdPl>72dY%jPZFBjOlM>P8KfhJ;Wh@H`zA&9NhS1auQMDtKb+~m+$6To{^ zFaI5XVk?F+U(X!=23+=%^UiuKz5OomlUU#HZO&=04yjGcM`CipN_ziS0^x^-c$4}rr*(cy9*53LJ zB=O0~R}Rj)B(Fw9uD!IzL(7U-cTx9mO|BUdegO5_|DU+G2bDBS6ZFoa1eH-n7-bZw z5FtpTjDm^Q)Xw%oL%)#}OT(5CMjMSNBL>w%X=x)6g(#zqLsJxqC`2F{ku1YjU2OB z>xRA0v-f6ioddYE8EY+TsB7yh=KG6Jd#{0P(|(WL+N+lKI07<_1MAi)3sNe!28nKL zAYIPxsyWYfWP|SV99wjC%xNrWx9@ZH9RAAw%)a;bEbLvLYs*pFx&h``!x;V6TGXi< zw?SXQuV8F{&gjMmXzw+V70e0dxPX3b#&}=q+IrpOnCKPhJ8tcM9_L~ibHZ-!iQElY za@=1c-HYe7*E)wWJr|qzQQ2F=^E!jGRq*qnt>OzX#?P7c%-PuYJJQ9zu`kGk8)6!= z#)LZh%{8`yz3<@y&ZbRY(;q{+Hz$#0f!gD(iS6^dkz3ItM(+MNJ|BTi`vxcKG-Ltu z`-v@#OTv<^mMPkpNpJyKW%6Be*P-4``dZn#G2FdxC^DV68bzdKG<7x!{2AI7P`xN zp4oWw>~CCyHEVyD@R@re&S!4EHJ2UW2(t?Rh-t!2Lb)-&mIviA-Ove!D!KJ2|0>$V(u#wT`Tt*h>P zYF$|tM^I139t2gZs6YbhAGpF~;@mcN9Z~E0Mc4N&cP^Wxl zF3XrJx-l;MA7)+W++y@?=@5Bxj?3GAyDVt!ozPf9Rq1&Usf7l|iVNWQMGakLVKMrXPV=KAF zRM6)=zJ~sHD)d3byfN0esec+t2PMm8GEiRNBtg)ULcjhKg9HS_LY6N>zexw zn{k!Ba~Z3xxknnm!I}9A(tX63zdtMVzH-z&P{UP|8szwQMFZBHZ?W~~h~E|2MX$ND zM-2bLIDPEZUHx51WQ;kl|G?g`7yRDoULSh59ns+in>k`_Kc7B>K+lQs70Q(M)VJO@ z&vPNiKMVdlI335`yu?VdvZv`3CvPxyMgd+kuMHKccGzXRNZ zJw4kEtiLN`{ug(|1$oejS$DF>m+rAAy3)1g++v*5`L!v{x2C!J-3M$?p+Lrn_l^QR zvF(UocWf=~1+LoJH}r*hF72z{bK)-G3>l-RzWxsXdF&7P-3R_NaRnMN_V!-PcNU+Q z=k2xG+ovLpRrdMTK1Myy8~WHg!&>(7T{A~1##w8N(XZV;{)|zu={L^zE0E5)VY9Y> zuce?%h4r(JXHYlBGd6s3F`lL0_;tn&-I@FhobY{3u5&xDcNswvE0bTawRo&muxa_So;?G;wbY&wfC64)et4vgvchxEt#jXP$WPtF}4T zd4XMNzN_VU&WSv0LvHXIUHNX>GN^r=wUPFAAL_)}&)kF^dcS|}*%i~*v}=oz({l7T z*7^bF8`IGJJZRg%80Rk7yyuE7k~npwc@12~yD!ktR}7Z^JgngzTO*;%9CD9+Q)h<2 zrykfYpfp~)vO$427zwA%Xoc7SP}e^DsUXqc;jla zhn|lu!FXr92hY9Elh`X*!@11$xodArVtapA+S&KoNc&iK2mKw15o^3Px}Q4t(Ab}m zf$Tor=vzL{XMCBl)>}2TH!y#T$GdV4>m88BsGH+FE@NHVwJ9?nXEgRguFqB971?vq z3v9}@V$|+)p}U-^#lTsd-OsG=q;VU5XX|y)wY4L0K_6p&J^DJkpR4}8fIry#OtcxB z>>F&*pC5UB#gxRl>{D~p4y5x0HkZCV`zthb`@1jb3C>n>^ysrq`@Qkl&q#Z(&zrjT zN}Tc5w2m^Y%kp0J8@pD`5tqo<#F=|Cp1m`i$gxJRKmEo1^lNwSUQ;``wj5_&zJ7Pa z1@*mNQ?2TFXW6w8W?4tYU-rwUN^pn$j{&~otp1u7%B)*=5 zuM+G1THp7`+B+!3*h{^~ve#xE-^UGIdBuN%3%GC7udP63Zx2@>?_oFX)~itNsE5t| z?n83S)%Cf*;xFjVYOPgg-HpAxf8TxU74|OuEg$cp-jdxAfkF_?$V`G56=_UCPtbNavd3o3k`=wh=xrbE&QZ|=^jSmQ``n|u^k1j$3_T8e%kgJL zfceL@6ju^id^&cdn9(%I~z+(gOdD% z>~~Czdc$@@Oz49RUB63PjbV3hyf)d?p(u0IX-Mz4Qp5QR`%MBJo2v%D=wolbyg>I% z>TTLQhx#6S%{c?o`!jyMFXrffiOpWUHlhZ-p-~_W0bpuIy`!83X;pUdA-Z z5pOT|3HmB^oXPLPqAiEd^SJbRhMMDki}C2rXuhijzd*mce7v86Zmi##J(idcm$mo! ztiMyTYQV3%`u-Q=T8`}&qt7dYIekr=@0K~+#JPGdv0I{JPf#B4_4A_lb0Dp44gD>V5a{{?UD<0+ zTSabQTtgb0pzVg9p1Gi}=b*e|Q(A8WbByt9Cr~DK??9ii&g8roY~~nK@omt+xRNmK z72W4kV#MjUPQ&hse9T{G&loY*+0d2tGw%V}AhEgcIp`ZSZPRX_o_j1kzsF-Uz9OB^ zWnQ2w_ek>^`WfB(4fG0$UF@ms|2`Xk$?@G-v6*Llpi7?*=^ZV*aRnOm|E4&7)^RrD zd{+}`Tt)ZH=B?+wW8ZZ`3VO>yMcVU(O%717bpk0lCAzWNLY>}QvS-yCcVVnHd+WDe zqHFiAj0@OJdqclR_wKg9_d$%lhE2KdrLf#@pwIf|yW%*XIeklz!#Zz(%9SEoPgpU)GLHKv?&Wq-+>xEo_!|2xDo{ zpy5wr@K@s8dyi*)g}u4mk6kI+SB$+k_66C`7USKzx<@}C{X4!rhA^My5AN~R9QEww zvt3JjU&x4|79~WEJ_Xq!H^6!Hov3reK7+F$UxQ1Z%NdPvuA1G{=RDty`?Tgge)IcG z>e$Td|Gf(7cSpq^;OyS*1-Yk2ugfgf*{Ui!)*tW#p#lV`Bj8ENhY`ptQ7XP@Be z_nloG-+SCe#=x^3Uouls{c0X@ z#&{2XHul=Oot#MQEc5%yAv|x72_!aW=?lH8Lk3cgW2TZrKHxK_;;TyfyYF$+7UbwR z-`dWXIq#c&pA}&r^CLz-&+l62HQ&|ycn0cNuOtMp4($E!$~&ZWta$^@8Isrp-JI30 zvfsl6`u!P$-}naBGv}-fxEs&Fc;~E%4O$=}z-Qa@+~+t)VO93V*n7QanN#On=6Pq* z<2i8pvz5A%6@5{UK@7upa9;{1D!R|cv++)}0sxj`r zVBd0_Cz0ynnyZFQ?vXGe%F@?cScv=zXvTxePh?MhIejF>+EK| z;-1xF-V@ST>xVjXcF=drHlN*w&HmQX=TchBKKAc5*f;cw?B^!F+vhya^lH*Cfzmwq8;4~>}KA8Ag%gEI9%F$O7{d-X)J+*_mP<;XRE#T9JkD_bDIciq+F z5(WAhkP&15VNJgq{2VCzY{d1YgoHkSjQXBy)6Tvk1Nt4acLwJPDEXMD-P*<$^Z?%- zV>jd3+y4f%_gHM^WtI_#-altZYdF)MSo^GHtUY>vjOKeEtDboYp22?3ZTtyYu^ZRe zUyj(>`)=shCPf)Y?7gl!x&7V9@$RJ(|NV;b+{74f%`>`v`kq+ho#VzBbbZ=NjCvJW z%~9sSerGUdovrC3cfD)0YEI`--w8S3P3 zOU9_+nnea|Dd=V3GU3f5XapVc<^4Y|dbU%&Bh zpx46}*beULpsTXn`!UQADi*gV`6?!kz;LZuGwCY=h;&m z^1{C7&HE|X-76a?dvA4Ck4B+WN-Y3 zKKq)p@SS=M_R#0b7S$WUoB1 zPxM6Z^0t&;-r>gPavpPhmRpWE?FHR@@6Dg(%>6JY@b&ZH_q@s8XK@ANc1TwZK$nJY zpFLRLGgPE83C5V$p!>1y(L*0Nr?f-{xbV9--*sm1&nFe>ikLp6^IEUZjc!f@du?Ey z!_3oO!G7MWb&Ypv-@rbB+!N!u8u}Tz5gR%Cpj+3T#`gDxz2~gRD^k6n*BE|fviJRV z-j&z=E6(}R&DFo=RQ^3Xz729>p9U4w6Z)Mf$<^m)Mt^^A=~H6Fq4gYl7`H*dm3@s_ z6KQG0}`h1lbKMOfaL+|-H&bDlY{TYAYw{INBcZ{*laM@l_N8QhIF38&IXUjZT+g_%KNFM4 zRaa?E>zE@aVq(TvQ=7T29v@xpzr2sb)^%pf#09+``-acCoFUN*=ySCk{nR&f?|O?7 zrwsaAPrad=E80BE9{g@}Z~dZ<-x%u!cwcSl^fmTM-16!7%+@mRavZYHh)sVk#hR79{-@Mx>|NS@Z#WYupqhhpt>>v+bz5{-DbGf{ z`v7Hv_hZe>QD@!9j=II##1wt=EY4y-?dG^Dwx&Hs?LhXr$UQaIS)FUc_I~B470A6c z;?Zwy`#6K&*UsfyJ)FZ@#xyX_nFE{jT?hKiw zFP~e9u{W;Y8=~fj?{lzk05@VqbM4 zA8T2|{=JWOa{8LKjs2G6vsTapa>|0ep)2j_pF@1NbJ_#x>hmN@j61F9*6H6r=zDVf zJM_-c*jJ=yH{NIIjDGhY(9Nr$uc>oCJA7LX-=+-`Vp5QWm|mOv^}mDItK^8&zQx~< z-|+i)a*V6^5_~qk2hO41>rMNXGGNSJxXtuuNbePJ`s)pE>gjQNFquZiBk{7oDFMAsj4&#uqb*zf6AK;3&PG2-m0 z&AaS2?8?j`0~zdncAnMubi-D$X}@6Wxn)ZH-d|rxV;Ynk{+HOSZB9QoK4S`b6?=7m zWRKKe(a*#h(~?|dzdPjn81x2P63>0R>=6@fKED%~zn%wSUX{GAEWl_ZGY=D?B;Ap^DCHB+3UA{p9#BrmpPYdw~l`A+c`4F zy}m^^_cc^(#_gtmV_(qwGedg6dvFF8_48!?f?mP3#q62$>`0gXfP@kw#+p0(8~Yv7Ijg<|WMgkl{S6N2O8xo*-MQR9uzwBC8_=-%`+UKcfn=@N zE;S^v(ASh_a*Y?~?JH-E7=16)beX$i)jdN+dJg+;=36G3XmCY3>#J&&6fb z_UzjI&i0)@K_y3NU447#wSDjKIdjVq(|zdPMGrIvG5;qC(2V$M|_XExqm*6i`f z1a01tJ>8ca{hB(-I+{Pt% zM(yUfKeM-&ImYPMewtF-b?VMh!2atqW9@@B;;dElf%)t6ZM`0Y-DhOYL^r2Q{bt%5 z_KHpPuh~~W^AdX!zXRhoYz1kI{)WD!_JoV^#}QM^BhL6w?;f8|#qRkk@iG0*?Y#yz zzssy=?t$1nd*|EOTfgZeR$r=9zu=P!`Zox0^?tov6{zjc%U(UFr$eNo>tH z_Pw4G_T(gfb>EK-y`WpKgRNqdhP@%n)V-eqXJW3#nd2GoyVSQBRAeL&_w(J1LyzO= znfUs`*3i8Zb5BU`w4rO?vG=p8zkqeSobx)1eJ;rT5kL0AzOgSc=Csx}!P@;>KE~QF zW7O7uCAYEg|3vgN6Wj}TA$#<;hWAF8`c`x?yZfnkbq)S`DUvk9gR}X9vqo?vhoVE?$ULCnZ zHe^*2;|vvPys^%81IG3q==KxqB)miKHOMv2+zqT_OhGpA46fBXgM5f7OOo^JG8Q)D zt-IdkpwGG+`&2Ne#L%s^CnvFOP>ApE0d|+Z8jmye`zFTz*3oC~NgJ`+FY4H=mF)FP z0sRe>_NjAUv-EiH&O5V@dOtrgp1%>Z=2?FGCjPaIso1QY=+^4~yLrF%1Z%&7UXOhV za3#bzueBO_f4=DJUd#<}b=z^oU9r`~2imRGd*OFk%V**D_YUdaI+-|G5&S~zmZOGe zHqLt5*6Q=b_*vIkiZSd9RQT#{%z>rvYYy4{e;#LV=3$TE4v4ppduzGuVb2Ck z`xvznX-=Po*!Aihn48G5Gvi_{)+4WzM zNv^tcy41Ho{(79ZBs%-G`g3P*ZR0lN&8!pb6N=0p=DUKuYx#Sh9PfXN@jeAIu~p~` zX?@v~@4QOqw?|+%M_Zyd=!?C2zr(?_V(=Wwm_D!YBk7Tw){>3KP8U^gz= z-_x(67tnTM@ACJLNR$AUy)M}L{7C0M9Y=3{=2mPQ(wv$jr|azX_eD&v!M=g^gdKMK zsawZhg}wB1BK51^;PW%3&wQ7@Ek=H#Hz@3l>9vSob+lD|psljk=KIl*`WkGozKagz zyVvL0%hb=>lRYZ}*k{i3u-7N{xgJaJW3D9l%v!RC>Yh0No9{h!8vB>X)uYdf&3Fl@ z&_7G$SdvwQ^R)Rc*>9n3g9_H@=SR;Rx!MC<)+#ybny;?Ez=!p^PCsQM=0u+F zlCsC@Bd6zE2fMM6=)phWe#N4Vd5+j~oNt}S`p#_b7NwuBA+7N~AKiCl%h6Y#@#gm4 z&I3IU+ZLmLqF>27P`|GjXE)D${jQ8Lmo<7WdzXHFF7-XNN&xkYYvAl#j{bNae-FbK z*j*9Cl6wYg`1d5dPv46;&RRLn(~t-3#+qke@6x;1R^*BBT~Ce6 z$IrDs2@Njfx%VEW$FaArd7?e;AV)9H>N$NT zuJyj{*jv|Gw3%DnyQ2R(?rrAzT(=xPb$h$4-(tLv$`kuU)*LbFf$p6;#|a9y!d~hA zHT2rZO!BzfbqDI!aCIAf`|LqF@#fpBFXXzD8uM_xBb12G+1KYR!pCZGo7(lW1+z|Q8)knxzSzAX5F{22XJ765J*Pb*2L<_(xQnsG=(DEvTwRa3M{ckq z_HXZ~hYa@OT$!cUPxdZ*_}ys_pKnDP+rS#m=sX+y0$Ywc=DANWSM25L_ki?lEl`AI zZVk0DYuE4CWHM;S9YVHt($0MD||f*;o4>3Nba25<_mtNc6ig z|CP98?=z3%=w-}K9s35xuXn^c6|@^;KYQ(`enDSzov`&hY!!-<+><&hqKiJG_L^gF zEo*Bl;D2igb=Cp<79Z53R+@GzjA`QjeWy@MYp!G z@9%;BF5jVb53h(b&sc4#gr;O!YYUW#x%Nx;)^^T7iuo%~U%4Ioh;b)=_RdiBVe5J9 z^;gigUpdKL9!GuW-`FR(YR-6d&&1xJaVoh7a;y{R&ZaI+i7l{MXNUAVuiqW}vp(2s zBX&X7=*OA0osUTLEvZke&+`OgBGQ)K-cFz1k#!f{RA%ibz2~*lSuP@cK$nrD>iFX?0spUwt~$&wMIj? z&IO;JmtcQFx;J0Hb^Y&E`n4Bq6})@p`x&X%pM|qwH{ZMI@$}fKvsPZqaR+`z?OEBI ze>#quEl7o&aTo07CboWl`sg?Q4fswJ_AdKwIdZokp8n<-U%}pz=mnaR6^S4A z#(D?puRH7U>;s>+4QY;dZ=Ck$?E9QK-$&xsGb{LeFKo_H7O3_&&zQavm0H#`e?$75 z-8+|e=iORQo3VXP_Km$c1(|bym^lx~J|lK*J&sUoHEia%tl3wNdOm|KhTT~`^GkA8 z{oWhDG3u$rAH*8p7d2X-K}e3hwkquHqkn@?=8)DpL1FLAd-R6icy)Vf*Vg+J=lK%p z(x*MLZ^yfUznB}tZ>;_X8x*Kw?YY+$vtNlZr!ue`*BAPdtFQZ(kFz>wLO_K#)bz7< z2A8qoGq*;f2iU`0^YpJX`TUH#Af3;Bk0a)NF8(@mjJr4PIL`3l>3n<1vnS$vlocIV6>2As7bt!13G zt)ahw@1V=`co+9zo<4h<@2WBF$5}5Q|8C5In9AO~{vMPe&FOjMIFrkLLW6+b1E2eD z%kus^Z_Szb>{0L;Z(L>XY)u>c0R5+F+orACkHm0(&n}fXG2R-X&iZ?P`q;xbYxtS* zJu2vHIg9IJ+|>K`b&i_7M?BW>{jz7D3F-G~L$)Y=sV6aEoNLxK-Z=}>KGv<;j^lkw zbeFc6lqF|OpD)L({h6b0aHeiswf|h+FYPzvI+Jld2CsXcrM83$-nsh*dvxb2n_d1@Fz00^B8^g9IMyh+HebleFkMDG2zvZwOmS>X# zRO}H*dKgpWNZ-5R+k!n9*L?_Azl$7s>WS{G&hI^XKJ{K3xrZ0fV~DkuJx|kapN6g9 z`Q6Ch*q@JME^FzpknwnL=3Vf49{V=-Jr4PP<V9o!4ZXI)7Eygpk=d$5M`4&~|i=(CR`cH{js zzjJguHshVg?>qm@Zg0Q40{a2`4f<=)-g5X(?y0#EvqRTjg&6m}K6;>A-?()qYdc59 z*0fV+%~x{R3oGyy(5G%q^UTq1ANT4BTaFm}c>X%=?}$-v;2D#+21Ps4`UM#=_WCaR zKzoUSb$bl^z-Mpe7k8tk&!ytqW;|K%J&JYA?{@S)$ILUQ zf-!Hf?Tw4k$65BrQA1lJCSixIv3K?m^wqPUm$|HItUhCmb=D2r`xT>Bxf}V$6>P2= zk3FnqeQg0Dzp)a^#4hSHrl}|YAkDF6kHzfr z4qJ|QYmdI(TXk!gFB>tH{rWy|4wrM*oZoTSdY>3}b^G-d{0i1^{*sU1GcVX1+pyU) z1ieB8HslR9?^fx1mipMQd3$YaTa3^Bx-aXR-T4aw7&ftgb-Jz zz1!mP9q2W)s&!^Vkl#vA3q^KY%rSAH?_~uH(ovHgnX~eHzD>wTwoV_ zp7E`I_VR8DzFvb`)^r&o*7{JtvhVjxt?onaXO6jz9AmZ|HSO2H-p=h(w{IjmSSJGg z&*T1lclJ=Q$p-75FWCFt>2n4953q+Z+PjZlfxhUu*y?e_*}waga3I(G4c&S4^`*?m zxf^=LW_%)pdj5H>pquNtL%Xnl4_dW64Cvvp4X}=l=*7XijUyNDQovA8418n+< zD=~b*{&Q-Y?{g~b8yJ@{bmNVy?DcCe*v#qndG5fzK|7AmS3d*$N}O{veXwW$bU3kE6G_+Wqfs#%$2oH)Iv`em3eS``wH$=&5h^E$D&mCH|g+-Pj)U>(YA| z(_q~h*6|KvE5}ja7;C>J*Yn>%xf`{d!*eNnT#V=H8U1`MoBqHT*z2!P|NKAx=wJM& zKmQ-)?jKROcSTpFR7_@?ZRiFPk{SXaD9r^pkmg7Ex9BY^><`{p5O1$^A2jkSOtg{SF#2FTwZ&Jlu}?J%c&x zJYTc-9oOzX9^g}WI(dRL-a5`|%>ZkPdV%vTJ(E4`(2ais=6IeqrDxLj9d&jlh49=KZA;WNBo8? zU_YPF!|gb0#pk^!FJS&}CeHN-_*d|J>elt_XYekZ>m59R`s>L{WP|J6>+>8ZIKu&& zGa&=C*Qx7wZ;YS!--7*4@ZxqnC!dk;uFvTTC%DLQoX7kf+)NVbOqHH#Ur@i^jqfJU z=X173g##2NHt*nj@a+2|cCdCO-*vsSGg^CpJI?q9>G{l&^W+Bk2HwGI*xgbC&fv4j8n72MB^*ClT?*#UJ1MlE1 z6wbE6UOoBe!#%e3Je_H~9W#85{#)>~b|prf$sW#RAD@f8mHJ+Sy`9Z-uFqkmUxR|r z8O-0o84hz_k=E4zM{+OlJ$wT(W9)qcZ(zN{uh9#v+(w^%zqichd!1u0^PJBE=j8hAd=5T?VhkK+jCFlhem@%gS(TE~8;5AY1viFf9^WvzNVn?i;oz&xW)| z!RAbRFxGSa7Jda^K_%uDSn~=Gx8u9m?-zA{zI5&JIisI(?N7*6{~7rl{9JqkjhL6< zY@VzCOq)pFmHq8_hn&UUzX8wdS>M72<~WB0_<8Vr-mmYCGn~LU3`8~Zc+%jAUAw&I+(LBe%n-Jij8d<5o; zxyJhpox$%o<)5H|e}?hS3Nb;C3a$ft^XtTO$^B%Hv~Iu^3N`HeG5CG*YhaH=dcO&;;RfvWCHw()@Eh=Kp7WQWP5-ap30T*Bb1&-D zjH#cHXJWiNa}MwnK7+5|1Uq7kH+GwIHDqD0Pt={=S9DHU!!wvi``~aWA0rd6H z7&WYKZKbuG$>(6q;g&Olf99{8*Etg2Ow74L183SjBL_YM;~F`x^}S^Ln*9d7#Q1y! zGT2LjJ?!8B+AdS-OJcW~_q#jdq23ORYj6eiHE*JyVLv$_PZRCN>U&PD_2>U@(aqOh z=;s-{KkvoQg5Rev@g>~DE3l^Vm3WsjpiOD6@1J!ZZpY{K@6hkSzLnf7*w=cV&*$ZH z@fmBg*K25EUF3!o?^c{kS+M!%H1WI1&y(l+e#%$3>wj_2JE&m4jatiYOn~Qbe(&2etj}Hl{-fV3 zViK%t4OdOxgIg3hVQb*M>wkVr&g6>p48Mo3;B#o?ZeWdoopE4k) z{sw*B&uPXe&1=L2y?w{T`OVk2=`gawJGIqsOi7!{TaT7-%b5Tr0hV{MYGx$z^3~z3U0e^on-nl&M97w0m3*I{v-lo;4ezjj^lW^ljrF-& zQ{M@mf^&NZpTkRH_LG8a)U1$GI@cAvj|+T5j{U4>PkqiM>K|kK0_>~q-Jif7#+dg5 z{1H4~$>FoM%lFIQ6MPmM`UUCdyAbO$vYvIEM?JvY1K8s=JWSpq%{_zfXNhs%M4s@O zyW*cb8~0Y}<7eqRa8~Qe4y^wXHfPyx$DcR!T_z`_zpr}FK9^6R5_b;?C%D3X((>{C zULJ`RW?MA$8*B}n^9>|6ah4r;&%c5I`<}sj^*MR}?l0hZzJT9=@rS8@nExrZjoLw7 zvB&uXJl~!edx^R+&LQd-FfQi4X6Y-ayTJ|ih39#FOAh$ly^n^?XM31ft4^EJxjjb# zpGBi)!Vcetyty6kj&*%sz9PnZ`3!t!58&MT-hy#Y*o)6*JM!F~4b;6WOf(6zRLF?hMv&4)b2a1KhwK4p6}u_iy3Jvq!$A&es#~dx!r7 z=4{;08F>X|g7u`p1w5Z~dDqV4^Y)yc?}ojz`Yg0x*;jHJta)FdTlWT>(K>H#$Mb%U ze2L8%^X&IL|x7QiI2X((+{(JZvsNi|T^Is>{vBm-P-GDaxE_s=I^SlpZZfU%@;59e5_W+;V2#3%@6J=zC&LZ~=3D ze%_a_;=l2qPUh0?Q@Sl)k4`&m5`))a(_w0Q;pU-&@&*5-< zyxSeRXZRL=fZrI8O)6N^d;A2P|0D1oJ%_fp=zl;yhws66+w*FF1NLjj$6D4sKzx8_ zdEcHNS+IvSE|!^zX5s;g{2Yb2~nx4@mDtX}^Ns`>f2qKh87oK3FTk^YuHp z<(-58e~$_5&Y>(|-v%rGay#nV>mJVdPGG#!&&Jo_^YmGH9%og40m(coN9mkjfoJml zu+9PY`leLZwvpQ)L0uU+o=v%>KiK;OO7GE`WYyeeUNaXeKJVdX%EYFx=$k(m?eH1v zJXc8Y9Wi&kC+_w6z2na>e(t{o-`zifpO*mNTko!*dp|$HzkwIpX8&5xgr z-?Lt!pWqA!*unGL$M4=Nx-(rSfy|%?at|B2^Cl?8_u8HfUj^fhTe^M3T5=EX;09hn zBUZaHdpN@hmaWZwBF(Y(UcY&8Mt7g+`r7UI{lI%t|NY(g?)VD%!2SY$&wADq`0T_w zm&vNxX3azveDWH$DLwxa>V5*gulKiOuiW4PKAZkek?+`7+}N8V)>odvdwUK6dusSy zDc*s1q0Mt>GtS%!>g(>DtHDpih~GiJD~+BB8&ojH`QO4z*uftB9ti!nXkf7}q zyoRst#!U8j3fJS9BWHbU1vsPng_z&Mk6=#P12*IN-uWDxHGc!vGxqoJ&D=XcCgKv%sEUhNb@DYdogD{FZ0fD zg8G59w>E2eCZC_r^5J%zKWDGP{)U(Y=TY~${Tuk#;AhO|?j4=L=YKzAugC*@hHbkY z^`D@B1it5fcEq_FIMaDb=ND@f&@MaJgRv_<(dB;1f;@w{#`vtx@GbZZZ?L@qV7~qW8$RC`pGAW2qwi>g7qEu{-m^VoQjd>ydab0! zp9h7SJK}sk=6KJ(;~yivM{8}wIFEH2x_9b5+uy#<7;pvk9h^a1fh5No0TrI#jq^L7 zy^X7I2K9xtHgl2Qr_ZZ`@532>n$kFZa)vYqoJ0OCasK@Acku7Q`S(cAP$1ZQmYXT9 zXCG_rrwrr?te3ZAZsi56^(*)SyuTmd8>rx&dPm-Ao6=s+;J$))`w9};)ZZfC!D~2! zG0tFr=ctp5lC$>jt|TA)oOnmhGqWd{(#8(I~=Pmws6Yukk&H4wpLe98liyY^2Zfk9~ zkDn=P-NOwuxXN*Sb{~)l_OwTUXK;?yqp-JT1*P`i+>OtmKacqg0-JFsP+CXbJ`Mam z=YZ7L?^X#npx?Nc;5+&nwpnkF9^g#pDNjh_FH@S=`;)W&+_=*Q<{0mL@Ex4tDQvU; z0iSw)(9iwumU+Q<#rM-*;vB24xmU2q8*p~_#+;ynvEp9)1&sgVZhUX~?49iid<4FO zKCkbQe!u-5ehWKttaAW$-yc6GK4+h4oAVn}V9mMut$h7GhWIoGyt8}kH{cnRF5juR zWoEcydkGCL&~uU2+Tzh?u6J>m=r`6JXL>OycjNtZ{uA6!9+0o(INk&6oxogm^Uts+ z{{VjWeHWG9*N^Zga8BnvfVmYepzZC{t*cF6VMbStF>fHf7gvjM#to^Su;%x9`7B>k zW3$e)QQw-@xAz%Z;s>AA4!qX{Kil53`JU|ttYhAFVlD5)yYb!i`(p>{p8Hqe?}4Ag z83MiqzaKB4&HH=L#p5WiZYjo$k0cYPqfq?b7>YrWa)h{=2oxDWa$Gx95 zpT|#7;oa@1dq(cT+4TD}h~FO-n`bm`1N&Ov-s+W}3Fdl_XKWYP!w&q6{0{nO1^Fp> zCO=@((37k=Ox1;_6=~}*sIWsXEc$c2NQ1f%}yTiV|XTBRhfiqas`JC+nKTNwm zpYcXrIpdRxUO<`P`{rHxE{Js+STo>^&3pSAehJN7?#$2k9{bQBQhSD916|4x4j&PZf69t!3SfxX-Yi zK!3nX?Be@%Ro;&8IltpB@G~)I@H6AHeoM?p&&Fr!4f<#3d-xOz_&Gd7PM+Rj2j@C~ za|ASI@g6->7@H%TX;pAG(J_+Q}Pf-~9A=kgKy8EJ3lXq>@*-lsV4M$LqR?yAW3j@Y-w zh%;WF`R481{@q<=ew^6!NVR=nd&RW*?<@eS`fS zeDaLn*OBIH6MMbA9p4W=E90$mfd=}2i|vn~-7_ef`Z1J1O;TX+Yrz&R>3I71<)f@?|Z6n!i|?{DCvJ2v1mbRN&(yXN`D z+Ma(uIU$|RKF*PF0nhLw_=e_?1oWB(dx&vS|Ax_D|5J&og-*XR&q#pV19Adp-r_i5&CSc`n#~1mn%MZUf~G{!e1w7S@s^ zP1p-ipg~Jc&|txU1PvMtNH9BEu+qNgs=nRT-nthThPPPhhNgwgrBMpPhC}kan?yO{J+NTyW%Xb*nOXGu);_1 zS)KQ8UL(&M73qv0;63~S9@xd4FJRmZ&Zg{(JoN?a_4$zYGER8|XAxuM16<*(W&WL> z{;u@D!T$;qwiV9$_H%y&eL?RA34J;2ckRdiigfk{_BKvi24jLA=Y4u_){ytG!Wp0J zQO6wmZb!_geY`)uM}Cg~3HdLuf$!Nr!e79<{2Q>R^(RQwv_}D@zJg6(Lk9fmWjyPD zM*8eqB?baAT#4Pl{2F7f+hNo0JKzj{Cj9I_jDz!;m@~1TVbRCF9jMzYA?@Y*u#Y-F zyo~c0V{V3qUaSFk@SZ%gp9kf~ebleSojA{DFy|SZB@Wcz@4PGEXVlN5GEvL3e1|k& z%(aHkz}Y64!M>jT9r&4EiMfI@f%jOMr$9PnMo#!Yt=kXc+ zq$H+7ff=lsnWuqwaALm+=AHVW%=O#PXWwt1mHh(jr~Z)~Yb&j_!8u=`>pN@P+xzml zX7HVxu}SZZyo0qKP(fc~)&%OlzxB{F`#{%!<}cWcO?7-;zk3*KhvJU`ZHvF-unYQ-@jn{H~16S-}`w5@4CS4pd!ut zGco?XV*&LmHtQIx{vX)>XYlWEdxeu4>Ju{F3Ie$Q87*@p~z^uNbNR`o+_PvqE3R>;uc(;=-rgL$H^uY&ja9s&|f zU=4kDaAsxR2OCmmYzu6Vp}_;}RpEyh{P1h=JIMQ-KR2v|I7dM;eYsV`R~vE_P5-?f9m|fd)0Wa&iBdp=KH7dnaJ3EpY)6GO{?Sm zoc7dzZt&0k8S&js6@vdm2KyDxxWKMIBkQ3r_^sz^F{s!IdOG%9XPemv^nJyEZoYeG zcAt5o_`PACpA}dCeyqn?@%8htSN5JrQke?x;$3DKd z?XkfdtS{pX%Vo@WL#kil4h5!toH3(+0Q&^>oH0Q`59~_sOnW|Xe*2xV$r#~o?6blR zCJ1sKNT0p$?E=OZWdFHf4tk%5SbHs(arU!jF&?bx+4b8?oBPnWf4k>YM_+vvyM1PK z=hx=!fi$n%kY`S*<6buW6I6W4KKK4D-Wt6pK6A{M^Zjb*-HzY0iF@mc&(inhS5M?{ zmKFI&c*Ayu27k}Kl9Rxg{v3Ap_}tq+V&FN?`|UZ{SL4Vn#vo7q^gr;gdtg*5;uY&7_JbOQ*>o-1M z2!|Hq*(nQT=(XxNOW$KNFAj+4_xWI7iPQcDes|xISJv)b7O=OqTr;{ojrDBC`(9~Vkr^2hjPpy0-g{h9pY07P)=x2>v-$>C z$k2UA=Y9h}Z>cHX9)8STdM4LK=yeOx%i#&VH`AA!1r3;Gfe8_lo)5T zMt?Tg8YJxE?8c?1@%ge&f*X83^!_|}&zzxxbNDRGTaebC!Fu*h*aLk=cW-{dZ+wqC zWX0xt?#%Y82j=+msMjX8FZSjJz7?Fq9u@58zQK(cb!~o5&9RPg+ItStzOT?@@qIA{ zyLpBE6`!$L9iRJj#Ats-cc1I1@4OFK4|_q{-^^U&K97)VAFo~H2>6Z==QLFJ4oc&yB%`bQ(rUE znm432Yo9Y(Gsl?4ckwq*BgR-`%**KR?WHXd=UmU=Gtsv0BX)NG*YSJgrVSFz8xu`$X&yFC(Y1PvDtu*0`{T49=>aV$5J|(QaJPSI7J1i~&l|<~-h= zcW=IT62#i)8SLxcyD9JyoHNIGX4=j1`2|=b4rXn2Jj;yKug`n1zcu=E!)MLHzMxm+ zdA`K{JX=GTE3p+BV7^?BdYcklE%-+rvG%-c-@nb&&td*wht1fT81uteb$r{-KG4CK z0s*~N9DR=1#J?nXT^(=h^M)xhwGnyYU5FGj`9Ao_PL9eI4`->OQ^$ z-;iIx=agRPRYzPyR&bTesO1?7DEkaZ>s!y2>$vZP^nIxAW8AgUnf^@t6<@H=P@#aI zqh1FY;PYO=Gh7ed9QWpJ9@P;l4>}Q<&`(^a-`)7kIIHSMYy(0aLd-oH#mYsh<)X~Q?+0Qx^ zS{$D}WC3Rnq-V`YYrFUTDv;SHe6Ci<@8ln_8`B`+U%)!*6`M8sJjgd~6(4rv3VKF2 z#ySPv*k`D4#r6Qt&==C$=33wQ^S;&1G5-cDlsZ{W~J(_MV)>+DhXCtlMqq+VWvb z)gZe0&-X@~@po$c_%c3=_A!5fhVBfX(Y=GpzJT{L!7F6&uC0-S zxMQD)ThPx~b2FIl3=Nz;*!OcHjcJFh`|++I?Hxf+m^9pV3#O{e54ZNq-?Oq5FPp2Z23lS7NiK?^VTCwVex_{#?g- z3i=nYjx{D^Lnd%G@zlVGSq{saWU>GUB+3BKkM{6CB`$V z`}al8zCe&C)(rOAonZ!Ns`!*2-S6XPv;KxXBkkipvrk}+RX_Fu&L7y7zuYH%+L0A$ z%%^=kkA&W^+poTi|80P_h5Zfk5qCv;Chx@BA%|^6>MtLSiflV?w5-! z_1JyXwU+11(CZ$0L0+l<2e6O4UdG>%Kad$Va7Bz~=$vUEGv+$()c4%yZj82yytALQ zu}`3HW^aCF@AsK=YBMe!zJkxYm5iQZ@OQ-bT=rQT@ryacwK{yt>#?uI7|N?Vz8p2lZ|oQ9KHwv< zH*mg+eMNdNzE9d*-mP_i_kya}defQ6I$Xcg<_Nds5v#+!Gx$s?5*Y+O$6U^G`n5E*Aa$sHc3<JaKv|Hc3`n)4yBgYxMo4}`RNY7D`mDnWg3(_9l9^@aGpI*k#yNx(& ze}SK|S2IpGAIUDgl1Lw7lJu-5-pn^VgyZx_YmeoDWp7x)hzmrI9_VGOC6sYiqy@8)6 zbDUHC-OGr#cP1`CdA*=xGp>Vp5N~hmdv6JxvB$AhdZvCJ{pil&Oc|ebim>GP8e)}QxL$6EIB4(;9N;~v_PYmblMSpw|iU1W57bQ?9Sv$B7Kb{U`BL=Ee?fA=(I znelI+bXH?B1o}jrdedf2Vw}x>+7kOk8+xnQ!MrQBitQDtZk#zvSFgw3yRydv)+cgf z+|kP3KKi_eg5Cb^C|7xSLeRT}AHVNp@@xol_HQ00JUJIMc zc;hy3xi@yk@168##Jx`F&SjrMOhMP@K7;wz6=Ut?48~UWE9i4?-Gp?X;@h*!=$_f$ zy*AR^XD~k#AH)Q@7;mqFbamg3tm96MpV`axko`WGyWq3Oav2yCpv`>!@|TzKzZXvU z{4DtXW$BuzeC2XH+r+}>U8Xu}FX--jZR8ECX&qxSbf5Av?nR$HyaRuJ&B$P{ z?TVg}{Vw)VN1HL)J)3hWjqi5!HwgAOpggmA6}>jzkeFD zWn%B(jBm&rn19v?YIw%`P9C_TQXkx#n*$^3eD?YZKIzXTj{9T35icwFOe?-Cx;g6W zjzkY+LO$>rYm9jx4*knMV$SED+51cyIp#aRdHS!|1KWhmNMjOui&0}km;Sw^4gYju zCiH^tTp1F&%bI=mdep*ZT#AwB8JxwvIf1U<8Wm>t>hopX_j!gOrlX%83;v4DH9cWt z-(uj|%ujX9;MqbSdvhv!k_s+jOC34-sIj0vwM0Mr2OUu-;N#3_gtRu73^b< zdt(atU7Xp0q|2B9d#&0)pXh4==$^T;*Y4a2{dyU9WsLLp8SqW|v30Q9 z-@X}c5a9hvsiQ|GKENKz`7)$_^8@K~QnnapN#whn$N6UD+1q;V6EfH1oa_sG(bn_P z!#zIVHS700+;Qylh5JS~$9r-fmvwwTcTk$UAuINTo|UdTeAe@x9^hH@2Xcm=UdA~* zi`*c=73M=v#8+6^J4aK$5C_)rbC&mUr{0HlT@RiVvin&%rWjnd+39k@t{xJjPAX?f%7!%U+<3n3P~GxsLwTF%dqIHqu&F) z$L`~N{jB)0%n-yC{Kom&@H1eYz;@1Nj`KC_*UNbS&G)>nb53JC zyJxVTbzIYBNaLl?5kq%&|DNQdF$WF#_t-um3;4`l!Jk{tNP9T5*!!f{)*s_eD$;*9 z;{3+8UyxXpy=UGa&=Y9OG42*iukHLrJ5pa#vUfJwpu!BE!@pyCXft-l4rg5~{ka)y zUV;@0+;+|gy%*i>-G{b9$_>v`ACVvmIwXLOn0`?9yLbl)Lcj69_?IiIx_2=?04 zwVnNQjGE5d@E7(0{e4a>bBkpayfKug&LmI z-l-1k(e_byL00yklwe)=V(nGi@thkrXOI-bZhQgn#rOov%Ta?EwH^})i z2&8p3WcTyARi6H=-_H>Cf;~Wg!tPpkeDMq05pPUX0$A4f?rz4S8{fd6ciI9y)o~|Z z*t<#{acA%8a`e^?HPZj%Zj-h;?B4$yRM^ag`Dy&jyQ3E{M&E+{M=&>H$ljOz^fKNP z=X`|;yld}MJ&18RllcwaWB0pJ-*g$$c$YKvcMtgt&QK3qjpL5lH*|fz*Ykn41%27a zy=_SQ^f{2}0(IP-vHEXdoy=a_bi}C7_)1MZuRmXO_Ydq<&@-_09(jWJIkt?N7q zzxDJdCRs~ecD|Qa>V!I z8NPcN?|!{73m9vTXSJ>~TWdXJ?~Cq!0_%FtMovEXrS7~1yK6x=W->lT-GU5w#is1_ zk@w4(AtQZ9KY~AJ{dsjed>hht(x0{Ji9BmNkMxDjnt^oYt|N`hFrz!8xt_hUkAtQS zyo*|~6Q{3Y^X#6-Sr)j1v-mz)!x)!6HfY%5Y5X@-`dP7g7y7JgjbNVw)R6_OIYYl^ zr2YII#98|p>T&n%Gx2F3@j+Z;Z%ykst3EMrhO2vg6}Ag1dt>{%f^HA@eIczi5hp!P zKe}humx(b~3i~s+->Z8_#3tmsr|~>`Jas(3dW|u6CdL`GArp8?UXRd!8%$3Nu*y(>^|@o?(UtKCeClrOorK#uxM(zQR7CPw38nNBXWc^cCq` z3-UcAY%NBddHU_8t_&q)aIUU1X9PNH!*=Q=M&3l;S)tt}w2lU0>T>181;j2J0s1^QdEUZ*S3F_EBR+UcuVmzl=WC?X^53t5&#- z@5}T=4G8!rcn5cUSFm=C<9iJ09EDtatmwZcw%5WI`1{Q2;CvZd0{7mpc|I3wYinQ+ z_c!o)q$Ae60)D5@*jG^Ap<(Y`rG6^d%?<+FhPoUj!S-V9)MH7Vw>@6&vU`&KS=imAz=cvj5A|w}1cR znHR9nbm)aR_q`tSN8 zgnL%RxCd+6&-xA9icCk}MID=D^jGLJYpdiV=Oezd4{#23QTKOu|J;_*)}&QV6645p?(MR3j2VpWUpQ-d-3kP2W(=UcId{a_d7%`J0aK-G+5bto&at7 zuIL3mUXD80`uSq)-@S}?&%S}pTx)qZ1!)~Ir^LWG_kmq~rMBb*vzq(b3C9zfj%F(N!$l2y3(53HmK~qWAEq+?mOtt5X2U+S3;Vb z(Vz7}W?$lnVSj#S8*kpKgh03MLQF#1Zv}nkBw<;*fw>8sErIzi{np7)locQHj5Eix zpT33o8SHgK#>@EM5jndtV*Q3RwxMTza6P0m>$^e$=T-NaSMrRXz&r6==DN40^g5L| zbZf2PjLz=fn9!z#gl-*uwT`$R?=#0{{vFh<;rXoLYIV$N9djprNLR0|tmIM49)4%c zkYo5a^c(#6G`<7uqpg6tHD28tkF-WYfA>N!xb5TT?kn=|+1pFMv$!HgJ!_d;*? z^84-s(&c;$x-l#GeUy)Vj^VRbso4KI;?245)Py95-`>tKk*_bKyR>Kai?R4hj9TW+ z>}Txu7IS5O8qe8xKs@IaJohJP)C}@Z8v2C&3m9*lJ%9B=3^-4>k=MU7kSjJR=!JZ1 zRB&$fKss+mUy#~T9sSL*PXoUf%rX9~?HPKGHttKmbrRB=D+CyqV)XSH^d6X|eblnX zGq_*C`?80<6WqbO*VtN{J3ELa|9s9tiUp!|*mKfi?t9#;ty__+p zw0=}#SoTSA#8h-vt=Nf?7K4oLbMoxk%!$)?%&e`{)aTLX44y~6t~#_5NU)~!XJW4q z@XI{w*fS%Imz8}%YID}UknZiTt^3$pD=A?{Z-?HW*M7t@U&XF33p44r*3EpRHvR6e z;C*Jf6fA`&kbDpzU zw_q#S#F-NGXMxXmD)ghjM+S03&QQ^{Cpcq1?%3cNZU^!5?OfLPb8e5%;O8&MpDy2K z^bW0W-ByRRrZXGoiaLDmJzt-hywkoBAK>qY&U=R(*pN4vclOjc&uOgKSFGz?=QFmh zcj-(CpS7I7RQ9Y^$9JITGtbrY>*!Tu{Ep@wiqGu|_NhYLwvX6EtoDL@AhY1NuXUWM zpr6kp;mg=JY<=xRSkHM}=JhkyG4qW5d=A}@-5%z&Na=rh8s8zyW!#-V>;4g(!9JOo z0)g%vF6W9E((@#Am%S3Kkk#wxW4<}wt8*IXonE03V~+M`ZI^K;SbJY{XKY0um@A%1 zn||-mGuflEzq9Xir6B{~Mr`2w>4|vgcWNB?j6C-nGJ$a$^4a}j24IXgLx1l?S^?Mw;eO;{%Ukm)` z>jjCf{{y{f7O+)0iXhTXSY#kb%!xHfY2@j{OyBp1y)TQ75DOcaHY*Y#DA4 z_%D@#>{GBCf8Mdud4t>p89fnqg$)AkP@#Y`zCnf=oTu0P z>*#N-345wFpAvu^O?6LxII&O4H&^u4`7t>b*=Y4@E9beH?-*uN=b%n=uQ z)bU+qtQeQD-(WrnZ1%W9BF1`mST19J9(?xpIA=Ly%#i}tRAywaNzMl69VHigm$jmHwa=6fyRkR$y%p`o zS+{~cJkN|>TZ^GTv-kI&1?ju8fPVWm9DWp4d3-;eCMQ!k7do?H`stN);{}Y_EEppZsNFb?tM@EbKjT4)@Md<_|CY* za_@?rT5tzHkFAbAo-v`m)@1U2lE5W+mO}}&s&k=S)92c&pz6X6KgEy661Sh+=M*u z#Fzph#8ss0PK$|L9XoMT}tTZJf72-J9z#L%&qW%E1dBeUH@mWSFXdipeOX+ zgE=<$pbYGRydS!DDRIoreYp?!;PQQbgBonW8P49`@66tK&)|ODzr82DANEAd0?ylr zwayLQd-~lAXV}pzW+>n@HfI9!X7);B%&kaswK=bP!5+}xef;Xys1V@2RBQ#^&!lr& z$J*w)uGp<*ja~~q1z>jtvBt{;8JuAP*Q|^Y7ZpFcOWl<#dLWOSK$=sr-S<(i5M%xA z+~c>7Gc{ymZ_mKrh^>O%6?M$zKBMudVzc-Gc0#vL|$_^hp!@OgGg?x7N=eF1AE_SzbJ!dA2&whg^O zk5%sYIGZ@9^E;n)Z>P?7LpQeKGuCC?7i<%_%opQ~U!m8Ik*CesoYgfWoyR+H&J3P? z1wY4E$Q1(q0q7g(J%2|+K<|(Ad!WJt3S7bYeta2u*06q`AL;y!diEn<=u2sZmC zq&8>reD>`%kn_QYuY!5jEQkDpeD{P6c?Um3zjJTReatgsGtb@?dqAoO=+3E4pD|zW zVIS|mXU&ixz-RcljCroebl?oVe#OR$%0G5BbbK4+824V#6Ika4J|E})3;Yp2?Bgza z4MJ-j_dgRifh*T8zV5p%#(i7cciK8WC-WQJ&|hJ_jAv4b_s#f&2)cK!1+${u$qA#_s*409K7*cfQ4V=CHSU+5%~hZl~{r z-gzR|I<7i$w0{J1JVQmgq8?|%)>n)>+}PX0&&CaAaBrS}pOBC}7h8fX%y1>H#n8`q zb9~nNWCFh%&gaqd&D}?y^9A+n>CDd%*v!+`8ys^=&JX2*qA?>ejuY`1UdyH878gFm^obK=a`o;JSS&89X}TW+1EbudQJ2e-+CKw4>@PCUOUk5-aEK~(&tyOWqcWFOuFEM zh900V>cE%=&Tovd;(J|U#Co4TFY~qeYe?n*inOW&9qpA)Vz5vcGHS&Tk!;wd`q31?webCguj7&sogz`QO1> zGo<}^p2VBmz&Lerl^C-a?|q2%jJ00bp^iDs&FFq!)E~$RuF$Y6-MgIAy?x>`{4;ww zKX2Z7B`!h#TqE^c!#t(?gl?QM?mxf^4VyLIK;I4N+Ai$5w~f7=vBqE7|L}sJ7|-mA z;2-tdT^9Dzz;k-X1%E)sW^CS>?*-kxuu6=(^!`7=XSg0%?JMnrMbr$3s_;bYB8~rLIbbT2;j=gqcZrUOCZ_gkZQ^&jS^S1s31$#duGfl)B z(bpTM3?HHQMryml3jNMv)G^-P zXN}B0v9Czu)a|uhhHjmN?in+>^BHG+Lf7x_Mvu#gS&=@k_mGI`y^Uq{Gu4q>kOAh) z1B`9h&iWPI9uXtA`!2+THhs=!oxWFx-Uok$zOa?bUVjXF%in0*vDtlYq1ab^~?C&_c@63v-ll2{{pFwGZmzDYouN)BP)CT(!hOD zj}e>6@A23YwgO7$c*W=Uk8?iY$CvTH=c&fj@ecQV^nRDfvq!|x8+q@Z#^VeWL9K6WqZb&OPtgz*()Gp{mE2 zM;Z8i1|P|r!Fm0I8;1LDxFXZ`Cw-tp~_fz6tUeM5R5)_#ERmUe9^_&@v! zoX_Q6ea2tWrwe<|G$YL`ut8t-$cqtg{d^hEc0q=AWP^+zNPD(A`YijXea_MElo-$H z`)My@oui>^yB{{|>Yw2T8+&_dSC?mShJH3=1?%-aKKn|B2J4Y`KEtc|P!D_GAMpvB zd-ICaQP+9YwJrGjI^58dyr>;A1qpx?pR^W42& z8hO8YI_7mxoXffVOrAL<#vLmwxYQfC`uPIM`>NeMefG(4g#^}}pT_@&Mf;EF4S9z* z_=$ZS`POpYz;2#3X0X@uKJKfbdsoJJ<^}vSkU#V5Wz?xal*8`hBrqVZ$eW z&n2**^Q3*8$vWrx68as#HJz>J5F4<9e@-&LfPZc>w(g_Pgmi9ay@NH??Gw~Zkim0l zyTYUoetH?Q$KhVF^`9NspD)LE5_#Pp-}-LEW%e20j9$RCkdx}@rN4n^D%dKx?!+!| z!*0AW1vYd)7v8nAi7N;HsG}EJU!HL$-h7``f386e*e63nH!mT34eFWa`K}Q7%xg%` zZLa?5GBBs&^Uu@Tj7`}5{NAA8GhSOl4`@3&m^U4I1h!+fI{G+!hAVjPj6DZ)pcn9W zy=|uswg%ppb!Tu^WgJ-3GsR`FAbo$V?=wvJBZgEj>Y$8@&a$3$67r24-$Cb^4*fGS z$Ec|cNKk*FV;SGUl{jz?_Vt;_S#;O1CG-@|&1)Iy6?EeM1K?d)k1a*~l z$i%*)i$6!aJHLbcoOlm2vVeE7LMC2$g$0t3n<1bnclJD6dul(Qvu8MS&5Ib%$NRG0 z7tmhFDMPCoBW#Wte{^4-gESNIcqZN?U4gL-h@oi*)iAJ24y2mJAA^wFMSRx)?9 z)_&XkI6QcG$p} z2?Cr?n|sfei0yZ}k2Cjka!1+7ipW)X} zBVOMP8u7~CW4G4V`qcIu@39%)7sxO~17&Wb~%4 z1o!qV?uj=>j7dAVkj8yo?2R$623Gd-FJfLV-`){pf_)IDoDW|<@_i>(Y~GbpyS0jb z@cxSS81G&|S|=0Jds06USE2X9p1}MK>Cb+3X>gv^dWpTC8>O|3@qN?wJyLs*S7H;t zgRGlk)C=^cpK~Y^`vvZYjKgLvV>4VKUB+Gbd2hiUw!>!r3^(@PiRZtuw}y840LB$; z4JtT;(t6hmd+tW8eP?f-4F0~MTn?Mx&l9?{C@b*=5_ana`e{%2dmn7Q7W;l4;xe{7 z`vro%_00DfZ0N=pZN&B2kiOIA2Xq_pO5?9!y!Dj9JiZUk~A0IN{=(v+SIr~fqdvcyo~>jkiD(ptP{Gq%3tI6Sp_ue+>z$j%ed1uLNf_j!5N zVE=~v-3#+VtJI7!!;Kn#wx98PCnvwdX3hr9Sf3BGRrH_mS$77{qW=qeK58bUGuy}6 zuJ{Tz^Mk!}1$$S|p|)r13+cIyb=hNM@A;jr20mo2heWq_w)v-Zk#Fn<-fITq%~MaX!UL>V zh;Q%(eUhEHN)7m)dN=mY@*bRbzThWbDpdbAK3MIzp{b%Vm5sGWjyzc?DtR!#;0AmJRP4ga#pa9wP$o|F5uo3F+QjD!lsVi z8QUlJJ`d~rIrcrhp?kmP*z26t`Fw|r^*J=~ov`L-n8~%Cy$iV&j9sC?1I+Vh>OaB< z^yGdY=e1`eC+e7e)$Sb`BkVWud;y7ApHo$T8S^feagTpyZ@zJ*4&Be9@!Mt8(jL^X zc0hlw>=$A(zTO+X;j3Uj?dCK{#xbA!sH2az;vV!_dxkHNp0FdWwZRqK8=s(pc5`}f z_A`1z>Q5E+QfgyN)?Zj9l-{`wsR# z7kf=e>zr6O=_5veLY{ZH?eqbC=AAyz<^5J{4WH|Z{sz|Dz#Pw!p2mCSbIdS~vww~@CRTL;$DX0EnC zH!h(|g-}0YoY9^Ow4nC!jwokhX6)9|Uy;UJ!)1JcGS~4=dyc;1@UuVfWquj+w2GvT zbNHR)yJ3yOelh->!Tk}J@!LBKG2VHvgEVJ@+hv@kl4sAAeXmQdvciTfkovFa=g*Cg z=455W9~)cck;#^9@#TIb&b!eIKn`V%&{07En6>9RhA(Oaf&nu~p=T^bGdV zpDW_nTgQB3=0pF4{O;wLhxloy7I>eHeSr$jsUF8Z9r|hW9?a>z>$i70v2LfGoIV5o zhCX9|wvT&wp#RsGaS!P-?%i{*@Cw?T+gXBq?H{!5%tpR5C>QqX)>HT1SNIG*6Z8BW z8apA~7r5X+F*%!8FxP!KniG829jyW50+Q&S1r03UPmE=`O z=;r(PtG-Xp=DC&j-u9z@B_6Kq3-J$p#`T5GI|%mH+O#8`#~S8&K7FpTkNdKR_mwIj z11VpyFOWf>cOq3A`*lD1p;z#^ID3n6zpHVF&3QBSdP4k!&HBbCB{t`GMIE!Mdsm6Q z^$M7y+~5=3-Q%y&`|Q}@#~18K?OCY3vJdojLB^QJ{`tU~6KJzvh6MH$;|e5b`>4gP z+m)q`yWH?K^a?(2?e=#beKkgZeg4_cGdi0&fzP;r^=aIzck65q@O$SKnT;dfy6Vn3 zLmbk5x8rkm?Fl^~ReZ4QV@CT{VgH+#5u?qyCT!;2$={%%cN>1^akdpIcH^AK*o=PN zv9Tij2me>cJ9opb&7Rs^_EK+P4fX!)txfy}B}V^-O!&2Bqmfb7gFP-s@qvMsklmkdu3KdprNxb7h|dU#kdoZTfxwRUi6< z+`vDV8YBA6$w3Uu{S2j!cl;Cj75tf|%{rbZvp2Q`Ht6{mbZpku?)>$Tfo#~#KhKe{ z>Gv$Yhko|G4`V&=ud&(NJZtvne(bN8Z$GoN7jnK}t93l1jP&lDwV=0P&r(m2sdEGE zV&8h`)-zxE%Xx}^?B=WA$gxL4H(nXY0?KsYJs1};a(oZ;-FE~$z*drf@ahyAWd{l;mN0@LN&S^JFc;jCSVZ}4@IrP!Fey{egJD#FuzjAo}N*A zfeGw!S7(+uW01}`Lw_dZ6l@8cL0!8w5@hYL9Q#_w9opA^>1DhJ_A=hS${M5IiXP~` z)9Dv--PzYMhv&RQ9D98kJ;&%_{`E4>Z69k@7R=?bZt4ly=y*~Mqa`e`?zPNJ-vq;C=33A^c}I+ z`E0Ef;9lx4_UZ|&KkZ0#Z5jMN^t&}9#hUij?)he@*o+t7t-`)P6Lf1nkUjT_T=s=H z^D??QE@$@K?mg=S#wPamOX${gzJ$*>&!OMDYe;9Zzwzq!td$tnd>J*Z5$Nu{f8`at zJEfm%d)vqPoWXs>w_4h)Q?M!RU9c(jHTXz=BhP1QP3zmwz4Li@;>vZrYvzl~x;e0u zlOdrSXZ>Jr%_}%t#TTIT?uvRn>_}_fcI012zO!_lIP>PGQ8Qte0`~m^=GxDizEjq> zcaN>3p1m@*c^|RXbS`b?ChgdYaY)ZlklWMv`@eF;`FBqbbbFX1#x-NnXYAIn_6ka8 z_gOoGF$t+|Y^YAKDTyey@?{Vx)9eIJ& zpOGul-!;xWZPH_qpYhqRXhZK3-7^@M;EsNS5{DhB9{997Ux4xUsYqAd$9w0y7YFJY zUiWYDzs$AXEBG8%u+M@_JMriZ5_@gd?B_z3ihUn3#yw**{|f#+@9*K4HG`P`879wM z@4L^B-M#nbeLTp25B9A3kYC^q8`wkN1Vw*^edJd3jLo{|cS84Lb5-J69eGM;@(lJ) z=)Imc)_EsipQFz`y6fvXoY!8?WW9~Ob2{HuTj2lead-G#wPK@IAPaa8?_q|_UitkC zI?{K=@0Nt!=U2hp#y%jiSNgtbtMSdRZ;{KW>ns5YW^Abrz1P7Ohs_%Lw5w~E^ZTUx z35k8!8?w(~EP91IH1?$q%yn-2dfpi_K6|JaVqNE5+hZd3BY19eJcscay&?Vna4!2K z_8WRWbmJ1bzI?=JGp|B{gxwy_`waTEZ|L5?xji47OW%sGT-ZpA_j^NkHd(36R47;o$qJg>cbP50cT@n=k2um%1@en$HFnRZeHpFO0J z7m(n5Cdw9Lc4PZz4{4nDaF5@4`U+ar@%I?M55EEDY1kX`JN60xc}DHMJ~n6Zte&y= zac=Z~ec`_`0U5uwe}%rn8@@X|5*2NZ}#`u`+HiSg`VD_{x5L9jGoU(XPTkJ_x%7QO8%20oKS!W4*g_=+0Qt zOC9%dWv|WI8urr{_>%-RkPVDmh|^ZsC-(X(;%UDa@MnsIrg~|^TW&NtIb}{8sKb`F^8;yAKeu@|ETPl+Zn{Tg49>_p?hBIyn_3H z3JLnTki|Xz8E(*h$T;*B+2kQg*e2a4?{Q+Od@w4Wi>6}sg z@2ypGoFg65^NF?zdVltb8U^g*46dj{S9+&!_*QW5_jv|CukWEih8x^rqPKQGCyDrq z{(w*HH)M_RJn{D4E4Ew*?T{O>&Q#np+w)$T6=#g$%B!$1~i=-A}~!y3{w{dd}0po;RfRME?xlrFQiR5;fD)amV&HY4;4Jx(* z*6i;R()m|##)dDRVE+d_Q^*;=bM?_62)E_8N8g>|4-1 zuhO}Uao&W?O33H|>fK%u=-QTj+uw*FHXv4SY5g-o1=DYo*_1#GbkKZ`j>iC)k^xo|ubR;|t8}{oUW1 z`i-~VC-m=LMvcrq6RR)KJ(sHn|3D9<`cF?|e$Qi^c2}!oZtI-365}4ez8~CW!tcH! zJ-0E=>D^R#K*Odli0OUUU)j$)x!9~Rp%-Mof26TFM&2vF`DJ_t^gqCO@5j&bc1gXb zGSpcSM{RU_7_aPgnOA8{B3@r*?>Q2(+XEfDD_3mj3vAk;1)*S$XG-YC_&N1#{e0TN zdfvyn4_%w_=34v4euG{Q+3@Lqp!fslU%&Z++GmXU3p8T1g%ZDZ@AnFceP77H{xiLewO8+lJl|dOtYOai9(=fr zv(3m1pWzvLzcX$pr;gsr!0vMr=UA2W^*&dy&jQX?G?Q9D{N2>{f^YV7 z3(`K`)kI7KrExyXSJ0L#c2Gzgq`6#`|a4 zKf(F;UUzJ7@CB}5+>Bfy9nyPU!F%y8ChVEG6X&Y#@z2CL>nHFTIh(b&NH)H|p*~e#cMXzA3jeW$?7yWNuu;DXbIYWX<%(5fN5qnxs+ZB=oHc;2r>d>9% z|EA(?f#qnfbln631_lib3=9fKz9R7|`H=0RsvgaA5<30T*OUTp)1c zM8gjz6l^H=t=eeEK^qDt6ciM^&y!gteK*6e*5A8+a#eNjE!kmX`s`=_25!KeJP-V+ z1Kk?tix02V#rE@Y)+qS?0!9w>2G4g~{2TsXk=7{qeOCVo>UG{D;-{?W=krUr*0Yv* z5BMwanB%NIuVu$h^1K(agA9DctmiS$?@{MgZ+X;3-@pxA_URk%5c)eKa>5v z@o_z7glk=Eb@Xp=o?ru29~gH$`ztw-b8m7-ZG7?)zp?vu-q(9wfVCH-y~ppZ*z){b zI;Z>gz8jG5e166{oXbva?$yveckMWQNIBM$XWxP@^i|+{^~C=bIOB?K59=p!ZTaGc zuI<0spFXtF%8BJH0Jwe{e~XEyU8;*10VVZI`)Y`w~ldVxPUwG z?tXsEX>bbA?`K8(^aZ*(-qAhUv*OcV<7a-?rl#@@{#USo!9^DQ&Sno!k9oJ282YaB zGx-VKUP1gjy3#rUZpP>#65YHcZmmdcu*yNoiR+9n$SUxAmv%^N++Oh`%^f+1w5D9o zs2?aGfpr3mUd)>4_HZ6&_sF?nH*37i=k|o-93Qa%27EuYO8`!w`|l;zzML`RpauPq z9cg^@+~-{qpRlKA%nesw931@-0Kc_wz*=&LAALjj4hu2sm7KAPe!_b8G3QMlaIOvb z9pn4?1p4h`UIP7{yt8II{O+sa_nm)$K(|LK!HRBN{f6!{>QUxf2k{2koEH{ebKc;tH``9mFPjx$-~tmk|lYo6Z)*2{qZf*+~gLB;RR?c2Zw z+`tCh#jo)H3H*8cckpNM4^W8bQD@A7tqgqT7o@s%)U6wd4u0$#Gw^qOm;HCohv$_s z=Z7+$VZs@Ajyb=z<#@lAWz@Bg{lo)Qpsy%FcfmhupjR+_@T_T0;%gv) zxz@MN1FT^5E%?#993WQ|5kKEyx!%3%hyMlKfw|VxcL5EaCzBX=yy4r7?Q`B!)|hd3 zXWb^B`EKZbM_SL?=IamR#D001K6|ae&u;@BIR(^1Mlj=!54ev-owF&a^SsX4A4q#? zn)-<4p62>9=lon0IPS|tIbroRhpdk4Nx@EWj>IhFW|^fNP_xijDkd_Jpp z@B|gUadU1Uwc$F4{+3A4@jKTIcq04*kTayZImidnIfb@y^AdFQb0iKKK64) z_0bExAO3OIC?jY1(5;^!4*5opC(!m;hz~DwcKtq^6{vS0&;Ir+=+3tM&Nk+5%N1~1 z0OQCGL+bJtz-{rC!?=PIsn$A}Q$C}Q0*G~9AUI&}L*U9k?#_Zv|&h-^}{yoO~ z*>eT&z&l#Q+VVHxPCPAsekQGHJn>a<*16+rKs)fq!Lw!o_V={>`Mmz{IzNAPr#JAk zcmeBmJ{NaU(F^e1ee83-3%ap_t#nTNl{oyw?PFhMKj2*2-qE}b`4K3stuKgMH<2x< zeZ^)@;6T1p{ zXASGv>vC}IGy3CiV8b8ygnoDNz~_GJ;j@qa2CP@~|2FfD{rbvWK>r4(1fMPZ#;iN; zhtK)6&si4iYMfX@c5GuSkb4Jy2bk|(0^Z=mmG5^?f+w)Ay5C6&&m3_%r1OunQzPbn zBJp6`=N%-r&|iT)&+j*L?9ssuZYSRGRqO=>y61cc&K`38dslgJrgvyO8VWCHWmOHSK&QcfU4KX^m8$1gAQd!pt1E^+1!e+LiX^A_eNdFwuFS@RC$ zegbu$u?y0<7GO<#nI{@3*O{kuKI8h&KKkW$ecaPJ8wmNJ;+H?}`!e_7OwJuhx%M@l z@J0`8b2faJ*ZC|vwm#p%H&B2z7Nq$@qCbscmmnAYS1|6VTmNy)(H^q|`Bzeq-Yb#r za=eS^7xWwO=Sq{4lT(n^xf64)@m{~q8UBd0w|h6oqwelIcn445$(TJZ*tdP!&eQQb zgXdRcAP(D_*0S#tcn@cBwsDZ=uW*!NycbC4_CyRZ{;+n%Hs3nN)Sbh*#tHt+pLcke z1IA3qB>r)q_m@96eV(DBJp)~jZC{2TB&{_$o0|AKlK`6(Y5*S22> ze8!YMr^M&`>pg_~`p4_fcg$Y(di*SfEB}0M*74Msee4zB0m=bwbH<4oOMDAGzULyS0SR8T*TQ7w_$fJm&ok1O0;E``*C0 z>-WPpK6+wXOWk~*o4o?NVxQlA1)scC2_9epEmmn(j9l-4mGz`7mVTIOEFb#nFP z0b_wKJflv`eD#X%9ZCXh_tb&>`#wJpe`1!w!A^NGw^RDvz zY{?7!+o3Bvc28yk-c4Po3wyUHKaO>49-#IHr+@?v_-wTUP9V?cc;*%T0p~j7+6|bK zNa5Zx@WEL@1y9iMd5-|*dAC4zu%f=N_!tzy|!cpMbZZ4?nsyRirhHsYjM^hJQcaU8H_tU-L$tm^p)M7XGuw*gf$J z{7n4B3i^4E75ytd-%E2INbCCh;eP<rPw3ytFop4?xC?ulYiz;x_$*q^`_v!?

$J{gk(Z_e9$Jyw8e$0s21zbA_`7 z(xYEpe;%?Rm)GNW3H9r&W8Qb*GrI%(DE|hx#e5GsQg|otwt*E`J3t}k8MBjb?Up!z za&VB=j+ipB%RY0=Q`hHyPEN%(*LXvo{pH!SgK92k%+8)OX1?46ci_wwzqXfu71Z3Yt=W(TZh`=h!*1xFKfKI;f0S?jJJKHJd0*!rzJ0!<&R)S6Z28u) zk9V_v1vhfO5&H~Qa0kwH1O5!rHnsrcjr;}w2cWIr`sZh&P}4K^hd<-!nNv6485%LU z72O>u?Pa}pU~T=~xH>aGu1K(e0yf}IeAimcJK3)SV*$)tkoq%^-b(BS${}z2+=cs) zKVpQt_6s{ecDNhT9s3*$c0t$Y{uBE<^0Lo;C`-;=+S^(Dt}(A-8@KP}I%}C#H~NA7OvvfxhTNH$RcyBajtaj;e9df0)?ob%;T(|ZVW zwkvV{)Yj+yJo>Eb+znm3VL!3W5e1mHBJcaWgEb1=uH-mt4O~E91LMSuHR6{8H@U#~ z^^5w!@iXK-9-!o$yCTag9OQ!ES)IZ99prV^wZHW@pnluuJ>+>;?G?BSZF2)?TX(t6 z`<*eRx!$LMVIwDZ{3r5`c`eEKJQk5x%xV=zwegsxN|t8vn2k@l{k=F zfVQ%M1U07|`tN4d?!es5oPFNSy6PR+_e4W?K65JY^qezPWQp0kUWp61iSNwU4(v+Z z@4#3G{tWiB;ph7f-oW3m^(T5s08cpQw$JAi*j0bdyDr4soi!T%6*;~ic6e|SD90Sg zPW%b1Ief^!zsw!@4CQwDanvF9kc-cALw8oc)4v^ZLBGfeW{-?2v>bn6u42CQRGk8_Tb)I%S2 zW4?2Kk6ys0pZe`Od)g=AX#dwMfA;`#t@QwJ;0XeZnc(~G`g?VdS3%%YiUK-lIp=j| zXLZien_TlN@ei-_Gw~$m9q!;W2(W;E5dXGQ2l%=1=Y9(6Tk|83SAlt*n$8l$J7`L5 z`En|GBSj- zI?j6m)-%2!A7H$IcKw1L6_sa@-wIhZs;9ZfIfNd+t1fG;4aK5z&amZ$Oj$X9oxA5 z-jU+-KJRUBkGv;7>+ADCUWlh!z)~3BJeFoolP6uKmqlDbD8VC#{Kax_-f2|aYO$KF2J2>S3x`v z_}+MrN}cnldspvfuZuF~o&NetPwWn7L;ebW!Dk&||FgDwC(nJW2l}tbC$PSCtZ%J5 z($BE-IZuFc==K>WW)JV$a?a%E!go5f@wg2F_lapo%f+}nn9#ufA==Q?|x^Co%L)_(`q5@!#0 zW!$>r6LE9~#=1^Av+;Bhw%jM|94HZhps(9%y`FNb8@T|*i9cO;yBhA2x9iP z4n8^FC6MOifhX|=^qhQkclYI5e&on)pz6!Hr@-!D(+8Y!+?Du>?sM|lc$bQA%)&fh zv3>S8&_Ebt5BB_Y`te!2fKik9#P@;N8=Qn&fp>m@|G@w6prbeZey0C#U~b2L z$3J$#oqEpiAn)xiv_(aCcU2uYpM3*ZGl&K9f^5H!<6H1q-}{*J4xXS83%L4~e|w?V zb>8z=r29H&yTN&Y4(|=QzRr6lHN4{;INv|0VZUXc`%L&Ny16&}0lxt`0aj2Bjw!xN;!*?m@r>Cdf@kaLbx+=&JJ^){de=ne7DBDiJv`Krvv+UeV{0@eRn^D z#aPU~8@e-`w)2e~+#9<6y;D#_j?b}z1`@jgc@H30Ti$|RfO>+*{*&Kx&S0MbV$R_F z-X-;;o9k)89|@2FwB2P_M?S#~=xaGSN$jk5-)Bv0bdcB!kZY}qylJERcYyx{@7H

taK5O~*Deyezjk#m)JFq`eryv`?1m``}$L@&9b3cJkzeiss*ZW)J2_E3QR}DyT z;yUlC{{lKbXIHMs7PGFqUxCLyax1W}_yRUyO;1by*x|%PpkK7{FUUI>arPj`8uk)# z*w%AT_VrzO0{aHIfevpx6ZFh7dY@9hJbM**BM)iMryyp1ds#A#*^e^bnQ{gmFl!=6UzW-Qb?iHTTvH^cD`UJ<` z=BnE}(XC;>j=b#1gu`rd#~qNyoF|F7laYryjyc4vX`MURz=xMPul?^}1LwPSAs%_; zqubY>1)qHal$ibB(EZ-dIc#0-wim2u?p11xYyxdk>1I80^fJ*oql(s zufr+8T<=rB1`;gFm~{ea?hP1Mx3=;D>?HztGUpD=54qUZGhdi%Px%d9=W)eF4K zjt<%(oqOcS`EAx5?*O*`jM-cLtf3sc!FSelhYkCVR8M0ZeFf;-+-m@9P`^?5!Mg#+==L8$Rdw<(2q;fEwQS4ie6Tct^^UzW`^G?>ldA zXLRQqF!zkL!+*KX8V#NjQss>u9e)uBiA5H3I$Hfxzd!+TkzQ_9?{8Y1k)> zopHbOtR=TAvGv`6wtIBWcWtnM?SMabtohi>e3#(|b|AmPF=sun!1X;$;)QquWdm~R zKH;p+yn%K|?Q{3$cVND=SVP-WbN-yPPeGUC3GJiC$u%!?Vlj3+D{FY4Mvi-U0P{Zq zfA{$nY2Qx1$Jia5bGO6^w5?;`uJ1VB!(4mHOP~zUfV;4!^IVYDeg}T1TwaMI&HVya z?0TJ_y^Q%>3bsCbG-N^AbJ4F4c$D_;V9bI~Ud7gb?%VtzH;_d?uvW)DeGR?fb5`%{ z_uvhD1MbaQ4e9Y7qX+h?PdjF<$jLQsE&F=kj_pzE^ZA-v!FOPf|M4>ayP_}X1zwNI zE806;_Sx`zoX0r=c=yQR9y^g|eg;4qYjD(m?<<^+?{=Mgcdiz5KHrf)yv%VEq!ampvC?J*9r*2`aiM=yz~(yz^zB_c-5m z=WoDgW&bfl%${=1N#dg((qqk(8%SWE1w4Rt7h|5l#%Dax-;qzS>N`)u0nSnPS+_<( z!KSxolMh-TfpK-~+`+e<+Q41<-o@+u+0w4_y^ym2Z9mV}dVukqsiW_ye`d*8oU!8d9IKAJTfaD;$tu15LF%`s z7I$9ZqSJ8q!EA9>fwRjLyA@&Pc(EH)jZ!LRzXQe0N=XYx0vsMEE&d;cP{o`|k^gE{`GmakE-@pS@U|(l+ z-T-o)d(_;~fk(ghR<@Y+ypwf*Jl`Yko|=GLi5H}E=?iU+dG_-RK6(P@akg)uyiA^b zdQ5#q&z!T5yWy`u&S!Io880#C9Pb&n_bcY&H+I&1?(<&Wp(i2aSMcdNKQk4d=ZS6Z z4Qa0Wifo6^I(OhP=Z5aS>}fv{NsQ;JY{0u(%bKE!gQI^1=KFl)n|t~qW=z`|w8uMn z>Us3o%ltk<$mzhE$_6TM|3xJKIdH+&{{RiRAMf-8CFU&F@z^&41>3pRm5XuoJF@O` zzZKga9oW;F+Sah2_09i!{kdOr*2C{}`RmJkKYacl&=-(!8g+_*tG*!Bhm0vh-w{h> zh=b=W&g#3;V)jq;4t}ABGu;j}q+Iv10sC3YUdjr5ZtkGRL?F#6#Pkmx-}^rA_X%0R zSwlIVDZ44~8Zghgzr4=(>nm$U<(c0rocd-;iCgQ;(dRDAH7B)^-bdYfqYtx( zF}THi?7zIu?-ltgK7WT7=L};9#;ox|KDxPIfO^B#`ciGcf~E-`g`C9yZ4?i{Cmmrs7+FGy!d@Ewl-TMK*MfqMz;(GRYD#>wM3YUm3% za0?ju+8}{G^PS1M&bxw#{zERWv%hyY=Z3VdzyCXjd~pHh$g}3C57k|0n^R+QmGV92 zXnTglet>_z@L51c95v7ja5v7I_~qaCdEW=p*n%AQhP&ybx4uTq&vzo#t=oVn5*{(@ z`?)jU9PeV?_d{Qi0XlqZD3|N3HFk*YOwR9qz92Vn2fmXxAkR7%bn_DgIo1nMfSfTm zoU=xY=}Uap>vB2P$(XUR_ndXc&e45OpuO?VxIeIN_UU6m(u!)Y}&*My5~mz`SZeFCFgg<4gC{n z-~tx#02^?J_R&6{kMW4v!}@nP&S-AKKl+jrhtC?FUgi`z=-#Dco3|dYmOYgTNB+nu zlk3lx&<~vD-03@h@8$gpSb+6CHD*7#p3yt!JR81I3)`IY`?F*F-PwRYE8ebiM(^W1 z&if9=!FFEjT2Cpb0q^S3&V+~k1WJ3{kiJiyxTnU<8@br_TfqI0+U_es0|kD7^|W_< zIo~b&*s~y4{0)6PKWyhM_%db>rMBln{2TZR-a!XnzzuxbXRiDO+nri_0g-cd=V{pP zZ@tbL&OUc+V_(4on0o_qD=6fSe585@+Z7ea^FGcl$Ga$tKJ)^|+5z-=dh&B|Xc9M1rG0S*5Rso(kJxfAOw*iUdl7uHnX zfjc<+kvZmxnBNCICr7)YD?3Dow>lD@VS4#_fr0E^mv&lo}4p0h!+s?^E2%% z&S;&07ufpb-hs7pN0O7qfqD_dJF$-5(2bSXIbT6H*CRi{*{hP{z7|k`@v$TA>&&y3 zac8(A6MIAVY}g$=wS)hFro?VJ=TLUwj*PkM!9ll<`_gBv^H`@Q7CvW9Y-cEk4{hEt ze$H;6CtRg`=XEaY+>k3M_zTkd&ggUUPSy?JjJ{_NePHbEI_LB~YB6>5$Fn8o_lCby zbogI@{jF7%*xt=qN=#{P$vNXiKf0$Uf5iMgGFSiUb7pOG6LcFGzLs9@qIEojdcnK7l(4;Jk%8O22#Mg|_=dF8Y#MFU*5o z_L;ZgH)fw8F2|kyioF42BOXZXu}^&7-(1h>m-ls__m*cpYcymDz!^Qv!yjNA?3zD$ z5BwEh(+@n}(|7&8|EzD_8@7Et6+LsJBke7g12y0P^S0}}gFI)e$W($Ru)V_#o_*y9 zD8P7Cf8n$D%JbXjXV3RCfwsJXoHebr;cLXK?dicE*cCbYC(^xPkN$GeFXXxJ^W7}Q z55IZ^VI24waz_`ig3LMF1KoM-GY(SD*>BXwXFYT5-_eD=mDaz!LLp|ZpKG7D|9y=| zp65poVl^i6kRuMy9^R|r@5a!b;YrRLa4#K>JpJnSY{=Vn?x>(I=-M6J!TFv9`U41g^H-#~_SdJpBR$R^pj_vS*0*l~alq$d zpV2SoecrLns`Of?g5cnE-1%E-xReBF& zHGzI)KVVHc!o7)(KKdfno!dSC3VzIq-}x5Mfc3?wgVgVLT*;Y}G2gvk@CClxKJR0H z_v1TmpF4N}`>)6i*mu+`^L{tuw}P&`&RPxq z8~XXa%PHW3&)gsHA!f@tzkAz21J<#|7v!=-;?z9y@qGf`&Ex#m$e5f&zmOyU_cK1t zf#dv2?K?2{$1z6T9!O$hJ7gf8@f|c!fV|I0=LvIwXY4&o7RCIeP#eyyp$T8-fJ|6nm8$A65*>m2tVmIuvQwN{+25unnTiX*k z?<9Blu-%>WK9%szP2l%wL5hT*a0}8J-s6tHsUt-vX1;SPwLNkpW`0AKoH6fm1ICmq zFm~?s>vg^-XTE*>oK-LmzCgbm+~tZFeF2S_^LeL}ue@m+Lnd;&V#Dj`eisCKp#K1IIi8mM z0cY&<&_@osv4xnkbYq-Tn(uDBr}2y#+t5qSSP<*@8yNc}Cb!_Pa0-6&>~CH9#sc5) zBjxC`rnM8-^t|H+;_M&n@@w4fpa_Z29J`SMrdL0~_)k-xG9dXUuyz`#Ujv zh(c_^zUU)X;fYNhT$DRJ57JsbSD%eLI1z_#zZOTG-zL9culVDTg=5de?!dbC@{E4y zpTKv0FOyee=HKy|qg{Y{OVoXS2CO^oqL0|9jr6YajLEqvU*^w0YdBv6Z%TL`?;79^ z*PjXh^L76B06t6e-Gl$VK!GEy?H(4;V)}glZ`g?~DrkugHux3U@p+8NH{U%3_GT{d zy{qc@d^emukn*0u(__9L4XJIr{393NLfpFgtyz#F(e0_;(4FNYkXw;|e3^F(Z1*$1JDBB%{;yMaR%62t z*Lv2qp1ZNH^>d$}hk{;n=6H`Z2E32<4OHzb^P+FS+6~s8h5r%eDHB?`2P4hXLFA8b9#Ye9cMY~$o=p#zst?9zE|iYg*BYTzrz&d51?Ap8h#GFOUWtyjOsJyy&EugM|N<&Pu5-fwY z@cFq%>+`xap{RKG79V8s%D|!I;ch;}i?%$Z7>j1+>TB`tmS4d<~ zNBCW}g3A?uOlZp;?<_uhe*%H-3>~yQYV2@1@5Vt*3^3jUb>hPxv$j3Y87sb5K6b8xN>8u|+4i3hlV0_21oP{E4N+yHW7$Dd_z`E%^UdnUdPobwJ!%o_H# zt~qBM&wS^3?)vRjkOAzep6bMJ)Uv1bD!#6tXX72Ly9nc)ME5!Ry&}(dz&d}x=j=Tu z2kMA_CvKf@Ac?h{&*hhwnH$*7;IW>%zJT8XC%V0z$NB}_!P_C_RA5dj_cC%$UE7%X z9clisV@f=uzVDVXdszPgZouc_`#E}$(~|PwvsR1k+A#4qR#*zXCP*LS0UM*OA^`1ddQ6%3h3;LKl%zkvmKtkDDR zZ(6g4`f{D0H)pTl3FKMBjW|LB_C*<;AWW{>tM=;oWZ%0t(VoH--*GIuP;{+;-yeu7#f2m21b z9{v>>;M0MkO|CW0@5_$Qdd}f&)){l5dNQ~}&K&gxzP!|j-M|B%y)K~bbHCoNVf#!f zFjnAt%*~xz!2Q`z{T+M*KCk6Yhga~qlNC6l@3QZIC$^!R zD`x?pz@69%z7Xrd9LPUoH}n7%IFEd#d$zvM_73DPK%VvWT|j^vDC)$__a4q2z?i;@ zPjs+h`?-364E(zdyZGCX5)n)*I#j@sA-+q{i<5qU^-eZqPx zh?q6M?z4`&df>l+1lF(Qiv||r=7e#iyL7km@1TP3puF(Dg8v2nH;`w10gXJdqPtV2 z^$W0;wau5W-9Qa|`bExXrHq)h>y95>hWjLuNd z3vvb8xzFc3>Z5P3^Z%P!UX#m-IPQ|@aI9I;8!~~s&*p*)Xy6VCsKf)D_2hboPsTvU zUXVAifd@!@<{GO9ddzvP75b3wSbxJV2NK&I+C!gp)%{+xws$|dLEL<6+0S<-&_xq) zeFp00)dPuick){jBc901K6UE{y6;N`chJBBIx)Xv^ryb-{G5y)qz@S=#H`bRGC_~y zUc~g-y*JB2BXGpl_dsd|H z>!^jUzXd+_&=Wt9U+`Pcd%A-F6Ecl#F*^tJ@XB6H-T>H#DKfB)3I`($f zk&EsOResE8vSBxL=UC96Kxxej@&P<&U3n#EJcwn?*d5*64m{>JkjlW`C$|{Kw#Ja? z#w*Y#KhQV$|9n9LpM~#^_-|m}1=!Qr8~6@_JTdNDX04Ve_*UXgxnpyDRt=wi>&fxo zJieh<^n%<#qT46*0qw|H>xSPQUx0jV;}tkhh3DP|d|&V{`>bbTRz6nlUIrQt~S#QHedQXr2oyjx4 z&k_X~w@xDMZ@&Go&-okDb0M$gP{$u~#T}S0H;|{EIcURozFXF)@SR0?uLn5is^r}E zIlq0)(;hiWee}_P!Il%`bmbL4`iLV7abw!*Cjve3n_GZ8^XQW^B>IMa-f7gdFYq}! z^BvgJI_GD5C1y|S2B_!_&NuK0tRV4wUv)WyzvC0wcVy4QM(7)NKsI1(k&7HB_!HRM z8rG9%4?j147CwUtoZ%h($VH8VYYl5xVBDTPW-arr)3D{q@frO670y0?2lBJIVE=D0 z^5LJg8adrOYI<+&(`Vgxa$EkcKD9sN8~MbOSO9l_IsDpZy%O{N^nUgUV4nxL0C!{F z9o&HX`@`$}-}(BxxVu@v9ef94kJPZ%hTV`wCq%>+v0DIbjdoVxRTc;O!V(GRX1=MVJEdAEx6toXkIIp!`PfV13yeCvB+=Z~Cw zv42Cij->rA_tkkS>@<&=;K`H&tacg1=eDR4t8M66V+M=7NoY)zT^49HP4y}Zp7}unS35eY(>UC^Q`Smne*K= zW}gv1{3~)hba~eA$bww(D{Ia=J}2jU>gSBlLmaz#r8asazJX7`nM=+(_9+M2;qxozXAb$i&gWADKf3oDxp{n!j=BMN0}JsAI(+Y#$OqEc9r#&G;J--};IlmYKG53{ zKYLa3JFu=k=M8M<@V)V!wf>iV?)HMTW`IQ>-0eEwE9Vh6{MNK@1%a<16O@>~4(AiF zhI1LWrbqtmz!#+TFW?TwcOf&`*LwQhy$F2T3kYJT|AuY-3OXp}#N=D2fXo@Y@AD4J z%lv%E{RW&Pfi>2Bas%Djy~F5*?Kx{#{Lky$pYaua1Lwb&I+y+R1?XVp=Io^W`<9dE=mu?*NR7Z_y9T(eF$iVxqfVOqUchwP7U-W}j z$vhY29Rzp>k3;WA?(MJxyC)+7XI7564!Pnp-r--sd92ew5;xDirFKsL6RtVlL;UhG z-&_0MfpaN0uxgWgK`y{Lr+#@w-{(xuVK3iTIR(2Q?H%aiS7M(I$anW0lHxu8yg+j8Ey!jUIR&c2Pf^(z?Xl{mU{`}1tf67!jP@0t_tG4Ok8%vo>v7X3Nj zAAfHw*!K6%`aOepWdv~U^MUT~0+~k*_$}s)-_a9T!@2CYq6<$3KJS`y&)(l$`5{MO zn^$xCwXM~#%PV2@j_>w6xuTaB-eqJEi<~=oVn6mdYr$`x+_5X9b%njGpTK9M?JV;2 zbHe+bX~>BA^KRTH=RMj!cT|X1eCC^{E#Ew`=)=~3SLetbeUn`D z4S(ztM8=#qY(L9E%=+>Jcn9qpemN`1DC2&Cyx?!dp{{`Lm$GXP7-|dQxyn_|J2Q~6S;H<#D z@~q#HayzhRBCV^>`xrNNL%)N(&e;Qh1^TToZp8Ez&=Mg140dF}7U!IS-&}e2^=yaz zGzRSDy+0wx%$Vywb@Sw`z&;g(z+Qm!^u*7*VLZ@_HhLp|0rD4k$_^^}_$;ZDr@yJ6 z&k4DaSAcxs%<}XtAh7+6h_gnOci7I*4v;l=VEjU!xs}+MQybm-+8*P+b5G#zgn8O` z&~w&S%GF=u=*{fTBVV{T-*f9X_-9PN|DNExKAs)1KcfE|xD$JV3-CVb%L`mi=CQ+| z7XAltjtb5r$DZ!92VP*`(Hp+p0M;J|ZVR*o=;mjj1NDwv@oPVj6<=CA0{r;A95Y7ZkDogpvtv6)iX;x4$U=3qW zbocisderOe6Zl8HINk&N9ld}%y7;ui&&l^bNi2Xo?UFODH0E4S5McaXj&G62S#~%{ z43wD9-5k+!&ehP(_1TQs5FT?r<5L>H0pr>qkv)zYJV$Gl9Vmgj7(Vm>KD!df{N&p6 zGt$rV9c&=+dG!DCGVRNbBWB!~GgfT5H4pGUc64X-MBsNOY->Lbj2?&X&K4l=wvV)- zulNd3nit?Za0c(-olfa28%W{}_)ax=H+1*&-02UJ9FH?v$C`I=0rD20uMi*a?I8mx zMjo>12V>8OF-PD7@|E2Aplo^gCUj z&(wH`n0d-b>LR;3u;=j6cS!69xB=($Ue5BDS9DPF;2|sdw&9cSXQP4z_@4M4`dyO5JaXi_M{C`Qe*qhCAD@BW zZ*NHR^;ye0{;p%6g6^!&Ajf~#lc)de6Z%1m$&H-+WuH592j^X^`4M~t?#KHEkoy2W zTk8a9)NvmB7W^HbKNFwm*1H33XWFjw83dfK==vJ^*d5Y3>c%VRSuoJ?7j$FJkmx5G zx;}FPT{y=Ltib--&ewt(#K)fSot%p99lfhF6gVC3sty9b`>goPk*{w7_OQ2pduyAU z==Lh;O`z+05|>-?`OMt&?R7r)@eJ`LF?&1ngSfpqxdrL$-CSbUa4z{zhi3mw1>Sbwlh2D1{yJY zRnU!r1^)&7nAP}k&x<-3TxR!-8i8M%n6O66pU=d)^7IvA_D^7r{tg1D7m)jWHlN_= z^L_CBKj*sRSGUg!jBQ}l$m9DEyMqs}v(|TX-~9so96!LA8EL(4`4XB&fkk*hVD!CgnkOe8^j=Q11few}f-p^RV_p{~ay9#`A ztat9Ipx3|#@>k-cCpGMI2Va0aa-Z*%_2gb^V9V==?fd63u1{GHy(B(R;GBEH8oG5x zEaom7_St_({0UZa?GeNu$TyJUiVtky;~^W;T<=}rJO2$-pkDEHupNHC3zqAASDr}q z0y-G`fNPy`uR~8_53mBc%1}2Rb5?bI&h0z}XlL*Eyrl$NdKI zv*O?L+KvtSktcu5$6W)Tvj}UA-CXB=ddykevCqHBMF#$gUO?Ta%$R-UIeWwQSq6B* z5&oY1pWueR0Bc#>zSeH&PY~oT$gjYEXB$3rePYalOn429opYBJUpY9=_X*wm$afdk z4Wyh3I<`IJ8oQt`z*#n=u)b1%0BgyuAVJBa&!OwLXUCsl9Aq3AK6KxQbteksyANd} z<~@4MJwDLw-GDU;{(|nT9lL_qnG2}rVK;2K?#8(j-MT^CT=g+0{zM=D_CSA+9BYl5 zsY_xE7%p-zbQcgpSAq*I`6Pt=QH1s@4$a+`G$0!P91Bu0|h%k&ob&4 z&PW zj^w$I75M;u*SrJm0utyuYvCQUmzXu>$PF<1t;S_M4uIgcj-E_)p|G3T;} z{hX;HE7Dx|7r@w=C;xJV1B!@Qe|)dQ(>C@7+^M|=jyxD6{!VNI=J`I0$k|7ZwLgIe z_`}PTzoPe;c^&-;77&MDzP_IH0M2sm_1ivo@4Ysl-@8@tc7;oi3vvY;aIQ+e{q)~~ z^#XVwXD#^T>rd>46mm-PAEdd?=-!-3y&*ezqT9!rD)Iro0&7^?V=Z&;+6ix;^A}`j z1Nj^APeAFh<{NqdbB(Q_f-n1g*PPFqey5h#Id8|drtiWHXj{vDspk>%J1WpGSNQDX zjO#0L0HNRbnEz!;XIj7${2gC|QxaZGZbi4}12ps-as#85d%VuQb@%~3VsBs^Z2J}X zW9QgoM&fzI%=?OchhuG}^`GE>z0P~RUx@)vjUj&IJKH!x486e%VEq8*Ia5LRTz38d z=N@|e+-c_A#e=vx-d&$(^v4&k(`T<0yn`o@E39|9PO0rq{kNoZFaG^A#KB$Qs#o$l zSg}3Y<~OACcu${m;#2ny#>^iF8Q9Kbu60IjbZ7q##+`U(&R6i8fA;YEEgp>2IO_iFC`eDu>CE2q3n_`ZGyU%&@^7i5nK>)HPX%6|0C`Thm61K$_llLh!(ocX>VJxCiT zsFQOiiM$Y(SMWE1&pPA8te?mNVxK=l#ym$%e<3Gh_A|BtdpVQ2BOl*6+k)Twx^w$) zpn-4Tmse&1zDwTk2`Z59f3H%J=9_2E|Hsw40!g-9>6#7-93T)tKtN!F0|XA(pg{wf zX+>It1_A*Z2sGG0;DQDL1`KH6)^_b$*|h;-1FjWxm@v>-*$}NrlrUjIhXoTJIN#^K zcjcMurvI$hzaIXGsI1gccc*@cAO_S=4XaLFkFAGiml`!b&B1?~mNsXGGcy*BvP zJE4mebYk!LZeRmfuma~U35Zt^yVOq5z?kK{_s!|EzjI%xu>flf9_k5OjQ5MIF=w!* z*v@nIM8BeIt9xGVujb74d)M0f?B!BMpuiP!?@BnIkpZmP)ZsaY=O)*Iy_eo?4$;0@P6c<(T!2ZyahXv*4AE; z7f{g0e2EG=wsOq1+X=4&zyHU2ki#6-an+o?Z}tnr)N-*t|NBb#l=|gpPW3;v}xmahJwE!z2|>=L4TU(;GQcn^W@ma zn%1(WG4u4j6WhR;4cS43XS~BZA)VixhP;79zalr_9Ixlu!+nhNjxzG?Ywa6o`f|?K z(c|%<+rzlE16hslu*UHnP+Zu4M!y4VXs^hYv!35CfnH)ibL2@M(mPgG;4I7TFW8A) zs8f+`hv#bT!1k&hGrs|Q+1GRLJ6~Jg=SP`)B6iw!|9rHb{k+$XelZsNXXZR-w^jp6&)Ycy z$k$&$0)58D`Q+U1@|d*?_E&Ugca}F$fV~5Z{_5}&9RBMbgFWg-%>7x<{C0*57IgDY z=*~0V9zN?_fPBBRob8xXf5mT(_bbP^bAFD~eJa03my1VF!ncM#ZzzZp9NkPEO*1@aP@(~z!TUgn)d5{t>Vm$*MBUXkwC6-aBi%(0G$nD58_1^Wxq{4{o+ zwF{hre$)tKpetkUZ4;O zu%O%1Gx^)o{QTUp?e`4+1z$m0!+w<<^GeR#j%?WGJIiNaufTuY&DaAR_hjt~)U9Rz zqhH6?r)=tvxld>5Aa>ZNCN}Z{T%&)?Jv-CjI1}qip5GA_tl%5)@6hI&w;)Hqzf683 zzJU^xubv_SzM=$UU-+#R$WBb(kj8gA&;z}I1)lTB(U&_4t0n&cV=t)k-G@AV4cT|y zy}K7@QyyoW#I0kzppS+AhZjmCuRGlPY1b=q0hy!#?ROAC_5oCN;|)2uXL`#!xJYL? zfdcP4(zEe9+w&2|I{CiO@|ZmHt!YjwS)Ylko1@-E?r~%21?jn{2N-97U4nYVtnVHx zG5a>8e&0fZXJBpftfd~iE%$~lzdrUJ%9wn=6EQ2&WoCAW}Qkj`Sv8rPlE@H=NGW{&+Ea79dC#oy7d@Df>o`_MPeTM1&v z@1JIml{{-&d+^ZT$a@9>HsBr;G~gU7D3AHh7U~4K?$z1_-B?4{Hg@zk_Z{f3$b#%( z)I%=hB~riGu$7+cA|L#yhw1(NCrJFh`8&|IUnS4F@=wT)eR|BZ?fCR3`1CTl72UI_ z!1+sXen_Eh{DSoFFR2gvwEO+p#_!<{nv$6L8@f6E*%bPKvo++3uf~49yMNjF$NWVO zdr#sWn5V7Yql_Nfao*1dm**tUciuojw_e8Vm%D9m&!s-fI9KxuxPl;N{Xn;dvl=tE z;J1E54h~YETzhqJM}Gy*ru6gqj=z;l<= zfjzF+#~DY$W7U|m{3BnkxJ%Ky5B3YVm#)|+JBV(Ab<~2RZJhv~zq7o8Pfv3%6;6TUOzQ5#J&ZH9HfY4n z9UOEy*6rYqUnwt z7O@kYJ8|=kYZv|a{Q2A()@#TMSb#Ovh2Mu8_&2%adxk;0pm$Kg1>f=;zkAj1EatA@ z3VeUR0~P%ljJ`2v2<^RIL6*m?+jf1~vxfQKfNyRC)`^(+cknzPwcT);bUrZETnlJmQ6!56!IdQ9C~$KF?b zg}5BK6_`8zoJ1zL!>P!@E%O}oyN_R9nC)qv(Rd#*YgFWgc!TFYPS!wL$2|AhjlnDE zo{MtZWBQ%p4Fq}vN@q7_|Km*_dchaQk&V2jjr30JX@1}vdFb-=IoIvH@836$&!yjG zorV0|<=6*2>lE^w>0h3v)HdFN7#RDOi|<0rxY9X|Cs1A=$egxZd+86PxELpA^u#Ym zzx9>cRS8;5e@7m+oWM4A>|foQD{vlrSjU)qmdim(xCB2Y&kuE`DcQ^aFeNw(jW9;P>dB^99{mT2LG**SY*o|MW6{Hux3& zpV1d!@1*vDu+Q!*>Oh}$oN1gfwz0Nj!Qa5qYujySw3hiL=1dJ~?hQ^?M_TtdKWlm> z6=`ficXoBpzn~{EvEZ|Apu4P>_^f5VetG6}=irT`qthAoQA)EqfSR3{iy9;D)tGkZzaLlFFseqZo|ve z^O0Nd-GQ8gXOAm2taAeku=ZHQE@$$=74(?j|_m1C~3Zs)r9$Jhm3ygp`MYg)&1aE`%4 zc=jtkx%%v}0lDS`Xh5Db3?DrABwyZlZ~^Wmb`1W`Ye?%fR;1IeW|<06X@t- zK^LBZ-+j(-^j3FHaRWVO{cYFf9dYFqd;>vUd%4Up){yn$>kYnBnJ(0Iv zcJv1HZTQSlz60wVKK%tKFCgFx&!7U&JCJWm{GLMt#`JskvB&f^PDGV?n*L5K# zry#9;1BtyLh4BuI%k%DkPo0KbfjO=Xc?Z9q$rmx__s@u*LCrazZ?h1qz&_Sh8Xq>U zujJo>=aMnc(zEqkJQHQ$_x;|`{j)zF^PZeZzqUO@&oWj=KYU+_+uOdzV@E*_?D?nj z+^4x4x-*S9LffBr5_x>SozvN`4+t##wS)r}e4d|Yrk%vhb7uEd1%BUKAXjYp8*tVD z?#;gc=L`AR9p3Pxn`dn7AHVfG2z<}PoM{8{Mf(lMoMR7#91)F|4(qn9~wPt-_s z>y1UcV+U|Q)++jf-e5!jrk`AW#zyV@kVXGvp0TyGJ=Z%3U>$Qqz%xET%zn-&{|b_r z=-}Ab1%FBY0~#pxq`TLtnKQ#1K%rg>&OZG z@~z?g72g17lEloHXTID93cLYy@6U5DG2@r>tR=4sW^Kft@t=S-T;smZ%yM> z8<L3m=dSohZ*1SFDCo|#qTfLS$Gf|t`#S*7xyy@>+V(Kl z&*S@f&UWmxVjBx!P64^=p6hr|)C~L$4ExvvwtQn1&NJva_i6oxp2QOWSK!>v`vz{n z*yu~WE)Q(@oyoURfbp?{cuv)G*^J{8v7;hSyWjrtnDaPC6`niNIss0^M=j)ky}&gG z$g#ff?DB;DfCuy+dCs|jpU7+Y?X7Lij($Dk0|mVSb$!k!Pk+sMR*9}X7IFFZF~`q@ zvlP(r`{$b6OzJ(*f%~%W;a7M6*1w?#`luIEU$KRAG<+gJ6*=#yBMWvx8cXD`two-3 z`Nqea=1#_Y)#r+Np6~D*@U!60P;W?URnUMur8N_%Tel;f?+y6pg69zUugKdD_t}8U zeP+(zFO?Vkv7ZUPRX*`Mux2HeAczTl9a~tV0_*Bu&J)%s_&d1bA3XN#vUbb)JG}K} zeg=K#{(Siz`3l}Z24r{_b#OY*+3f4`oFb-wBW@oNIcKqFA-=)0P6Ng|aC!d5kF#=q z>mA(Vvus=>u@2-FZ0jUwz&h^P{9m5f8}Rd>{}cKNjD9)u+;0=amzTL;_wHUjlYa*{ zbaMjfOcm}4J&?ti5)2OVa-OlVUvy>1?;Q4DknY3pwQor6Fh9$z|Bva@wzvCy16T0r zg}IP%hU-Gsn7!@a)N|qjuV@2%7}sY#Ip6VF$9TvizG1(C1jju)>))P;BNyN@r=V{S zty$B#yen&!{LOpA*YJ&d#BbffLF()HoX6P$Y27<|1AhO?cQ$LfjvjtCFF?NY$U9*D zW1pF`zvsX12;?!-`(vK-Gg59vUWr?CL#p3)IchwZ6P$oE-jr}FxxOd=%rp1LeVk|h z2^sj64ekc-0t!C;=G?Ve#;Rc7uSolNUl-v1TR;NNC8r{_-NWx+X8skQ^9O$MhVJ}w zKEYAi!=8V{PB`T}?;~=?oMW5`F>AO-dDiXvkOS!Y-?6`w(&rr35kG9|SicZ= zwu&^@KPy9Em%tCtSvuUW;10%u+j5>u1|Ix@Jl<VTD!K=WfU1Ih;_KRn%mh=Aq@bu9)W7ZvijxhI?K7r3Z3%c*{ z4c*?>v+oK0XKFZ`oUtcjjXalovzJ_TD>-^bHco)Y0jD7bLILTF2K)Yd6{#|H(#IU;Ij7E zUkr2Xw-Zl+uY==l74tyP+0UM0oEvpYfCn}IGT+A$3;db$u8&>|JoyzUU#Yu+hQ1+j&myX`(_8t;Ud{Hr#40+%zG=UmPZ=+ zkl6ONpStM4dd{;TOMdjXj^};@{$A2I;5;Yn5avGS`#HVLcioAP{bGMJwsVygSkv>l zJ<8bM1?g<&HSn3(1*nf&=n1^@g3o?|H2)5+#Ci}zdR~>-q7NCntxtQrxhV5okqL}l z;Wc&g+*?IgZot~keFH{jW*Uw8YcDL;WTjNBaA zKYAs$eFJ$%dWP}~b|C#O_Ppg;{|@ZYv6U5RU32W|Ij>;ZGP}`y8Cn{zn=o0_h)}?mw6Rj z&%}XT>sfn4Ph^R?Pj&s{u8VSh4c3vp7D3kso@=1)6bSYtl=E7%YuxZtGs~G6L|*(IFGs3 zd&B;Ip8c)kzE@yO?vOF(x8Az@PRM|hJFeKd^UM+UH6G_VXF+$q4*cGz*!J9r+f!a6 z=AGKR8D~y8i9f(1Z})rW&)~@S9IYeQ+RnOwqn0yQ;H>&wvD@#+qkqZSqr}{YwFa0& zncxQh`8<2eb58p^pSpEN4zUGYPJm(@DV%lOgMQZ7E4f3`%l-A7XJT%kC$fPHxk_tv zeaJw{G2i$G+~>hf;+}!HfVcCU*BWwt=VAecI@UKwTaL4M7s5WG#I#f071Q4y`@a)L zUIWf>2jz*HNNb)zhi}X`Z?3ah%RN}%v(x838u5lN&@1?Qo^`BMiQSO?OlQ3U0^hpp zatgL}m8lLwJm#CQUjf61zmsFWxd9r!iu?vVCu7e22JXN)$JxNMmc8`7g9TW-0(I|p z^uf07(9iR8uYE;V^K&b|$4{eOAEx5teC7xv%5|9m1JnEwg_ z{gXC%&Km*nTgS6`107u?a0d4~_Nom0Kz{|!J#_Xz-r{3^_xM?D@Z5*1Vq3ppM^4zo zS&dzc0e#13UEimdIg`1@{JyZZdIcx+ME009^E~G%Ec<(AmwDE1F?(zGoV!dj)cRYvb)93)pt}K1+c7 zn9cs}F+Wq*F6eznMT)T>q&>cZzcJJ8{Bh6L8~bOLz~8Zz8S}q;H(t>ju>K0Zfu^6F zf?xgyJ3=Oq6)hausabKFTMeuwM#M8_Y<3O3-}_Oj2YgMP(t?*i8v zNvvYaH+T4rF^60^-c8_hMevWD-}^V>es@{B17oN2I*Wlt=*90jo_;wK#wVJ%;JpB zbwgf&z06e~Gs3lI!M5Ip9B0kUa#zsddoK1m_GAzHZPa#7a~i$`uJ71m<%hJMeOF>5 zpPD7+y)5{T+SMEoyR@#n06pg{l^XK(x#T!U2mT#EUcqPF?dR$TYZ%{M)6sp$#?`l7zan3Oxqo_@Z`OF(eFc4-?|$YbXI=N;UN7Xxx1OIr@8c^-d?(~5 z;^th?<(X_ z>yEUZbt7=(w0}>XhCR5rja9hLroZOQA@t+#QT9B65)8QnA z=88&u^n_y%eTj5#&*2Lwz?rT6AJl2Y3&@z~GHRaZpJ#7KYs&F#yE+I^uoL;|W$LHL zJiCtXulUCI5qbd6?vPjV)LlW2^Z6D^?0;Smm+wpg?%4Kr2K_5~5A@wX>hAgajhpxU znEOt2@yD0>dynG|60W&&kI$}qXxQ#;1Gn8aF5lIHnRdUrGgYML>+jksvgE95zV(xM zfEBy%_Hv&0@Qk#dywe`vknUqFq`G~+?(*`O+$+*$-4-aID_Jdv7;nHm*O;7F^#6RC zaXA%O&-gpiIjrHHo$+rk^E>Msy0hwYmVk3Y-+;@!0*ocH@48Z6i;sO+KY_IZ+#mBS zD%=30KJtX$*+hxye@B0Q!6q($yd6x>NHl!%S70rdy?p!k-9O^g@N=3udCCIhy4MRB z=Z(DWy77Wu-~{@}!OjOZ7<)IS4_T2Np0OfW&SBrfu7M8ZoroQ8t>d$varNaqWuTXw z=c#WYC-IG3{My#ul-SlS0a(Ubpa63^h@7?Euk{k!85+9Mc+)4MjG7yTE~Tic!q{60Kjp9L7V{y66{--Y${ zxmtY8s(qZZv$dQxI`Yt!mD~*ky7P<$-@1OjuE4#Pob}zW-wAKv242CZm-%`5eufJ+ zkl+s1$J~#-64Q)Xg<#N)3WJI8ZTmus(!d2mk1&%ixhfN^Wv*L#ay8e4$ziZs4~lKcm#-@xf<&Q*dK zYdupV=bvdAv!8x_)|3~Zg9hxOy;uv3l{029=gydR!hHrmJ;6dY;tAwg$9HNim$3@Q z%#V+KoZy*z@U5+H@R2vLf{x!>z5~~W&-e89nEPIl*7vUd32aaEf8Tb6(||ww|99|) ze$f~6yT#i268nU#prhNrfB?o)30%hGF>U)6a0Lx`XA3a50rl~AISYMp{-~YAKD`h} z4qm`v)tEihdrlZH;ygdM3-TTKdH6esnCDQC&T`CkgJVx|0ZH7??e>^;ovGt<);qoc zC(!VXm9w^f`+KKyFFVhA0R|sg)ssHNaOh!$2*Lfr}W&NZ=4e`dvyFOK4%%+ z+~dSnd_i28V|{14!uPXOf%yf$T=Pdy;?5>dRP+G$SilBu;0=`X+>?1H^t(2&r@1Rg z^eT+yK1Z3$dgFT+zj1jB(ta0Y(MRo$ojXd*c-ifchtIiu*M9~-gNm)M0qdz>wb5PU z3^*tA?V-N|`;WfJh`A5Hm;E`pPm^d(5idD=z!-jG*T?)k9Q%96 z=kH4re?3U_7X$?aJKqJRkfW`;L4A8#n>a zz?v(1A#T1L>zeO(h z-eLhO7&QY9DF2)E{G6dBl6_jo{&G9Ib+k7)cX`;>v#$01bJVxc!53hDkZVk7?i;$x z_%rwpmIsai=;8$AmYjQbFZNU#yMqccb{_hskDP{2f9!E{tUdbRGj3mJTZo-NqI>?r z^Z)cT=M1>QoXfe4&(f!teedvmfBrto-UUw2c@NI!SvW(-?+h!tbH4%a%^ZE7!B?Oy z|E44-s8xVc+r3qEm%bWve(&k1mH4ea&PmB~j{y=GyMS>{=UK5G5>?Pfjxtm^=iyLTfghu*Fr3F`syJ$$3QRIG4*5a#J*S$&J`)ods-f| zN7?y-JmEj#cNWjP!A;=V7#Hqe=^O`&{)CUT*XV<8eg)Q2R(wa_z)oy)f}F&5jk#mm za=yOI-}lS4xB1R!A3uls)ve)Y!#dV#%7-ati9V4V(ppYBgy0mhq>d((db);^tQ z9cu)({ti!=6Y|u_^IV*vVHXjT<2>poq&e2Lj-RK7Ue2?Qb|)U#9e)@D729)i?^ier zkQ2H0f@5uauE-7CiI4kWt|YGQY&T;1oweb&b|AHnzJ8wc8SmhRU6EgrEAYOIB^c*b zX0A1aHJxQqr?Z`?wsbzv8F!vqrL(Ca>pL5fu6+6WAdyotaAds#VcsQ8nNpq zql$}-@&J+$(5JXab`c`7f`VE zSx3;0js4DvRdny&`e_WGw)H#%?JLOSoRD|W(B*d} zw#(cOw)1@VC1)M))VcL9;2C^+nf1*Vp2_diVeT zjETJg^YweCKR@k%_MkU>3sUqw?!7y!^K6euV1#cENUDd;1+?j@+?Fc$={Y^>^Hm|N1h2W~#)-XOJ59x3;wc>9R)T-d7noT`}QZ zIa8A3%opJN#;s$`01Xs)4Zra1n&(`~3({KhjrSeaNObFuI&iGx{hDX2fCeJxyxL!g zCm6GokG);b^Ng&kzJW3KW1e3n*ICL7H4{nPzVbb%6G%9tmvNRcYZ%)e#B3mwdGr1|N6l8t>8QE(zyA? zH!!%^)?cxMxaio4^b9L?0(9&K3b6JGegSgLU(k*J^$BT=Wz2o9_^jjI$sIuVt~TJg zTfY#iNc-74u&sOO9oxRnYH#BejP)P}GO@r7T{LuYoP}}Up!t4gGN$g#pPoMUIrzqM z&Sx)if%ju3V*gIgH+&U1TY}Hv+hgu;K~{LyQFo>aj+Mb(YUiB)qK;4Ld8xNVfDY_& z1^TtC;GS1}V`f5)SOw16z^~wMF9aR{UAY41x$d$ge%1-w`EDS9wj96v|40q>@h<3j zL2o(I=J!UB1B@5-oOAVEH>O^-&ukev=0WOrZ=S^+ID<0nyowYZ z-QU%@cjFB#-P%g1p2Iu%7X=M$MSB3_iX6 zPP0}3&$|{#c zoMI>T4O_qKn5o4cBX0#RXZ4Nv_YU*zCqF@ny~Y{;KF@+~eAH5c0*>B8kLe5iV+CbF(}^4fXYfn31`-hh0$`s7}bLS7M%Ipe5_?wmIeE+i9hgZD_ObVK;3%sIvBIS zx3B%2(VosCDxCF!8~EKKPT&nBxPb*6vsr&VQ%9hmkg@BDJ!Z&R*PiA!a0gegyzuWF z0P`yLSn$VQu&p)fA?1h!P5o)VbM((Z-u6I(Mh*9RK{vsfPFcyc}dJWTEM|idS0-F{<6c_Iyt^Q=TWyt1x4&jTh0dT-DC1k z=+;w~$IMl~qIZzMI`Uuf8*A7Fdn{~cb&Zc)~ zJ>y4O$GcQtkj_-W+j)N1RP%_H9rkiB_SV;x@L$0=V|?Gyt<}(P`Z+VX*1BVNV2*tK zqZVqQ8?Sr(wCnDt1XSXq2YL3juDw3J%$!938QuCFZoqM^d%iiI;|<(F2cE+T*i)YM z7i7XMz&g$$w8I!aYj_UM(6B4mhOvT5m{u$Q1$HM}P`9V?tZr707!9Z{QBBX-)(3FUVhk_nTk=#=e5TKkYe>nP+Z- z2EG#)1@xFT3w-UsPS8NjX^;Ek%#4{|kjiB2l!#30SKk zZ~7jSZ#|cJE#`Uj+~>r1foG2Z1&sOB@jKsuHr&w)-uO)$USC!=uh*k z1G^a8;XQlz=2%~O1&w$nJ|NeavmN%(@dxO@wGkJ_TF%%r`bj&cygVQRbMz&sAe8v8 zKYj|G3sgmKha@VU1d z(>Ho2y~uaY6SxA;?B^FaXSjm%v4`)|oB|p(g!b{y%yo_ioOJ_t;5#|)((^aJlUwB? z69g!rClWOB)U9cq)64W#^w|Bs$5*st`d0Ld-1Em=jhLJ<<3FbVN_+*@vWD;F8@6>_ zne%&Z(+5gSzB&5Lx2~U4`!)C@7wJBYSKu<{(qCgjZamPDBOm`lJdxJDqdQLlC*=b! zci=hNzr(SX+yDh=-{E;jHxTd|()w4>bDpC!%P-jSs_ns8*f=Z7(LJUK@DQNT!>j;$a7`P{hPDM<6f*?Kzm5oem-+EtoN)!#QcvOw()!0t_RzoKv$u6`O8l|=wf*~1!A@Ykcd(r& zyi@;Pazl3FC#1ffv%fv9t+d~d71;1v&OH>QIWF(#6(l*%R&&nd8QOn+nfGW;1PLzq z*1Lcf(^enx$BbF0pf}_dsl6(xt3K{L3gq2~%)?IH;J>}h^Y|J0ci?>1bY{`uUy;tE z9oQ9^D~xw=d(1cQTdBf0{7+947xd4-TK=xY)pFJ=vELWAxuZXF1Fyh6b-AD@u?G*` zm}~5r_p_ZbfU{QM-Pw28Isd(pH)H7bvA4Mmx!}JT+jV^|~1oCBat?PT$CpUJx zqU&!t=T_by^R0Gy=zjkFnM-@rFSB@9{sPzj_O$K^u5-Tv?HjQIcg&ByX|wd7fZQ=3`OcyqvEKvs3M%#$Ir<{a6VJdi ztmKHY`#8GVz!gYUp|df)Z`^#q5! z1ajT8bvH2d$F%w9h#&I~ZZL~+&~lzZ#~0}KzJpN*U&Uv=kdK|%?xP@GEl~qS2~NbU z5%?FUU%cEoKOK`$>lLO?8`uIo9{CM+t#w@;v7V`{p&s+2X&t=R*yr8e3P+#=j zC+~Qt+hhKnb$q|%oL`@JY=3$7axdzE{!YGkU_E0;-!J4`LBQ?6Gws@Nw2c*H18?e2 z^Lzr^S>*+?5p#wjSQ-175c6}pL^1>HTkXU|UE-~IdvnA3BwL7f%*0?gI#46cX?@6%aRJLgOd+nkZ- zEX4ed@jLVe^xVgOmH6~Gbni;r9*K0B|0ibJh~4CX5_5k&;i7loT{PfY;k5X;7i(2;C4MIs z#H_2&IYu3NHf;AL64*;$MJ96O#EdP(11O!LD&fd0V0+BDZumz}Z224Ty;-xMhkTYZ z*-v@Q;Q1xCHPl@_KA_LO@~?7$e9z+3lk8b<)XDi4+Riz+*3)+Z)(Pwi7W7We7v!pr z&sB5A6X~0?uX||X-~}9JGq0g12;lvypB^*khArgEv$l1H+4}fmosD-ZG_5B4OD|7abz=wXkQ*C)yYAib{^7wAC z*O;};8P+rT_^e^f<(+F={i*T#zFdZUC3| zEB zlp|&iXxIVFai$ZfG4EA-)AyLX!ZP3cN^mD;?Eo!^AiGIKlby>6M1sM6{xRZ z?C&vo1%9C0=Z)A6oIsCh`|qp=F=tAod+{^!3M!lg0m>Ph^RSm(bL5W|Q}_GO`ZWP( zZO#Q8+(0jIJMoTmE^AzOIrH$73J(tak@DU=4W{SaZd9QSP`RI~?nGdAn^+1IFYU zi}OUq=ZcuM~{w@Am94#IjCE)`|c~O;m4M%-r$zlXTq;u@LhJ?kZ-_qxI9gM z>^96F@5=tpW4{F6&lUK6qQBrH^}EHrxTpIruMcwIiO5-Z#pj!B*rI|CoI7HkoAEc`UEF|c%;tRd zjF~siDhZzV`fyLcy58Zq3v!OT@D3VC;zHZEwt)%)Uj+rdAmhBBJ-WSo0~zQ@doI8n z{lc6C9jKcx&)65B?H(Hl;+MyqP41|R)Nfr^jUmQND_r*`oYx%l{rSy#)m_#;@}AG= zIlnWNS8VIM?{PLr&)B^;{Bn%D*Mih%zj~hY8S_0kU&Ne4Uq@fyxwOr5{dSfcapCL% zGN-LizWJkXeAx2jS1{(_?tNpPCn zns;Zv;~g#NtH7^7_Dm-<*QofHOJsif;qv%k@14khdT^ zJazNO+^Ej3&pkPNfbC_zGkrn)6ZX%*w`&b?Bla11Cqk|;c11RDdz#-f&UQP`@Au!~ zdnR@P@*DgbQwF+l4(}+9iF5gJ77>%b!EwKy=?3KKtH8Ym=y25=F=zNhZ_nna-_S4M zx=Za11n~ROeEG^kjTODBKjz&hzOvgLU9P?K*+brf&)(K<_>G&V?jDB^-JA+S{b|l> zjx~Kl-_3tugD)UOetcIsTO)V$LMFK469qls8`pkAzXSa)XL`ksm@^DMdII$eF?(5C z=quPiB&v4AJeP*lKAmU1k%#Z!p7`AesQN(w_xgtHU2jKbTQr(c8wYLTn}vM<`#0CJ>gXJP=bQb{0s5~>~{y&-++EO+6{gOi+avF)=2Oj zpMLwub7{W;XE=eLv$otT(D!4X=)RZlzKo4ng20!mX~Q;*Ludv4(Azrz`vp2w1K$|&IRckA7_DF zce@jl|B2ehTs_ETCTmx);!p4m*yE1;6%=wJCfA(PW1fS$e)lr=fZl*T?I-UF0yv-h zb^jGO!wI~B2G>080LHFB-VIpq=wq$%Ifri#mp#lcIp=Us#scVX$O30O&w59X|9qk+ zD8M;lwTO_o-cg^+#>%=tH_^&)NE}Z|Dg; z?+SWO&Y~W3zQE^Ot4Qm}@$ARhxj$Bo89x!Xzw)AdP$TD?Ea>*uCoVvru;vvh61{;j zJ7N2{^i?osh$-b*tK_teN5BOxb9?OjKw2Y!ae4An$$k=DzOx(a*y2?i3?6=E2YzPc z``+I{18=*pxE`~9B>CvpU4VBbd{Y}J0Z%Z4Ys`V(a-O020ckz0{(ld zC(zKnE6>r-PlBKDTRURLt)VUoaE>cz=o|c=lP}NxRiyn77+_6kt&mtC*Ox5xaR9_N5wUgrN!?MBSnCw$f_=;9s!29C3|?gjYs@V^6R zXvs|I{}*^AW*=ANte3IhnSS&RoXMI>_tNkwmz|^BfPQnVW&G!-xwo#LxUz87PI6I0s5xf)1dedqZK#-C@q zl+%HGu~rCeFr;J$Yb@G zcj(+dQ{S_4S^ET9um@%81pBHnbLF|lIUvn(&l~75KYO0<70C5$jR|X%r@c=S$JS@9 zp?d}{@-85A`rpwzSin_(;3p<;NL2G|`7q@q_C;_AvjK7j*J2_&PH1yR2co z9<%R|aNdBQKj)CYA+2`~PM}1Ka)en6~$@0s95Hm&cr6uJu;Xh!tQx z@7$Uv{DEA6GKqU;3Ch#_nZVk{3+Tk1p(i%@?L5}VMgI<7A2WY7hVFfPF80=^+;*sc1J<#oygT{=JWt=DeQ(Hyl(V7R zQ|_t`T-N+iC(uD9W}Y=Vu;v$_zY;gceOTAMmfimqpStxM`V}m|dDUG#=KoD(Yh8i; z?BjmL={)OKd}WX6t6&9VA#cDrtZ@gf@s9AB>)bA9bcP~w?nPe)ws}ZD>ydEL?K8Lt z`8|-pdKD;>JZCCT`Xa)fVtKxIeq5q8!=R4)j0kNMqKke_p=b& zjDwzY4t>s3cfWHq;QcD~dnR(6<5ysP^es%3B)5PDJP-56 zxgf1$UuO$B=>9#TKW1z|({cg#*b^W-Glg&eug8`y#Fd2ZN0dgDuAPJ=V%Q?i`*4ob{3y`k&Vwx0YdsHso= zN?g06_nfe=wtbyx*!ZsKejhGR^G!E60SdOgoZERuebiwi#$I#YnQqU{WnGs&#EIPT zSwYX7=PTFmw=baYdM74M*foBmbDqv)?&yIm*cEy3ooPK|<3DCKN1t=cEg{cS5>n{l{KI4jB3BIrEhpIR#%MM`@0_Z{*09s59ke$1IFvSJtDjEVFPt#!;B=+=Eks&C-7#mnX!@B&pC~W0 zRv~8rX`DReifwPDHQ&H5PsETl$U|xWD@3qIeS2xmj4UVYW%7eY`kZTQVZAj+| zVE+!@@CDFzA6M`z@VoUDDLl7#@CST9`afp9j(i29CNcw1%l#JYhVE=GIiAsif7~B_ zdFE8$3>RX?Pav>MFw<^hjeMoGMqOg|m(z3NS9I$d@9>0m#zJ24S7ZZ49T<03=hN?e ziR?Q?vhjn}( z&a3YV8u$X7rGtc1l)yc##Fjm-uYmv;e4e9rkq2CIgn1oVK?C+03!d|7>p!ul%~+5J ztQ+9;z_%~?uF)@MUH5JME7H6_y%0y&Hs(BEK?A-w?GuorPdmsjU_qDTz3Ho9Lsu50 zy65H{bmR?qW}f5wt~*-~a*%S}hs$~ey@5ZH->@U5wEmIjn>v9NJcH4jxVHNj#w)p= z&zR4iaGcQ$@3i^ci_Gg$SZIX+0>OujYgq`T-F6eh)jmw$(oP~Gh=jI)} zfdx2sf}&2Ib18e`drW=Is10YsmodMGuh<>kx)r%$x17C;3*B1!0?q{l_Sk!wF+Xbo zoKOCD^idbCpPh;?f%*y9%X<1;J;?`Ecs@wi=tE7vv;CfM4;LkNB9A`y??EooKdhJU_L!f|cjQX^Gx&mTzB3hM zQv&U=Kcux6P;);!eD>8gR*)B@HI>#lJ?8wz4btm*qbA)Q~I{Q}*4`8)6)e3Qn0!d^i`_bkl6fUh7Q^S{Ub^isw= z)+yAn_8X9EUIOP_)ZvVN$ohZ}zJU&mt-D>QsqLQSyK2nZ3!JNb?4CQ1z5aidoo$R{ zSy{&q?CgTDLP=UcSfIcqF$CFL-PP6I0pqC-5|iC%DG=1C)FK<@C0JR>elqcf#AIVI zDMO2+F>ocpjEP|T@?p>zPPH+#nrOqP`GE0$Ffu@5hG<}nqW*sO+?toJdt3X!f1c+& zFaPH`FZWhW4G9t3F|Y^MK1VObG=KiA_b4!Tjry&ZKmI3dqvrohsyJT(;@)~Q5`RXOGFZalpy7g(-Z%+y7-s<|T$_4K@FFX9U|;JYzcDp1PVPCl6N_GggkQMkrZzG# zU+R|+T~y4yS9QYfTCs@;62A3XOXg+btX4=c#Ai$eB6>5>YoN`V>$-FFvlVf`Q6=`l zInp&-SDVB6y!Dmw?ir|uDY3wMLuJY+&>ua_^^B{G1+|jV%u}w>iNpL1$W@Y4;di}q_&izL=f=-!&v*c= zrM+nN^*NO>^RDY#Zw?$YWJ)NJ{{mwhJ>9)abNJ+!SCsfFU@bp?>S?w`WMF=#eFMv5se#-p%ubkv_DzRD1QK$S} zH4G^AP9NsFAJ)@XGq;}# zsogtt46J1@OQbcG+KdyY=gaxrP!LR~y&*hS-YMACMJxZSt$1AqT+tGPQfg zJTaHc7-d-Hym*rTKWGwj@hI~T1Mi9ZEO#WQ0eS^|$Cm)k}Z1!{cO zvzEREwjr3W*eg(DTl=<7a*i=2c{8MIodbt;V#<8QN<;okwnov$Mq~-zahTtg?w!6f z;P+eey`N1_=oPuv@$LrX5NS{DrMyFI10aXP*pzaOCA#akMgqRuTtl@Y&=}Ty3*C3F zyj8@`P>KaP&GVCT{fNyyxpG1}ascdisRa7$(Y%sa^*IQK_djFC*C4{jB6X`70n^e{Di)i17onsnUGQ45wU|R}j`7ykZ;!6y`wnwGbBCY+#;M0Pp6uq?x`wR; zHFo2h(Z-o4?_ibJV%-q)ov?S;QUdSC3>YV@HJ@f%!f#GwzQkvoYpz?n91*`Uju~+! zej)D-a2*`C(t73y^#yiivj^_)H0xE&_oYNO>p)t=zMT6mayU1}p4^L5aOE(|A@YoQ z!Y^Ed()rP+*f_lnl9f+5!V(vaSBWB$g@5uTtJFme^4ssgA zmXoRdM172*{*-gwL+h^jOYGJ)CZbm$fr1$Kpl)r(G&$Jpxy0_faRH3cKX2y|J)#Go zT!Lm@*d6A|Uq&Xt823M++nZ?41L-hcPVKJAyb9P)#1}-&aDT3u2WoK3_-XDzX|CsL z@5f*O_fQL#q{nO zxR2f|ZDk^cVf>JJqa*F9;fwgt9TEGLlgwYo>~~%7#de0zGjmVe1HXg#E?=VSy9xq1 z%3s4?A!i_AbEx~h$h9|h6Gl+Q{4RA3_7p$`7WbgxVD1`@k-j%7a>KMe52P^zWQ~1} zU)k8u<+Eqw6ZWVc+kHnbfqeR`@80|OT-VzS<_|MJ0x@OSi))F%yc@(WK#K6O7T5bM z&|l*(u?M95?s183oIC}#A<*yHEkTU=yyO+e1#C)vHE_M9HsIQn71BP+RgR0;TyLRG z4u*ZtTEBhNF|Wb;O-$==?jSyM=Rh9M-G1T;SlzuxH_<#0yT5*WKahc zx|joNR@i3x+qu0&Yzf({87Z&(XWfL)oB+Hxx2D-vGLPuO823SVXI!%!a+)&-a#hTg z-u1;jHsiwy@yInV77UrYj+xsWZSn;43V3&1&j5H2j4hB?!4lYCh0T~DNa)@NIp(X- z40#+e@dr>2v57K*cHh+Nc5ZG-TxjE(+}7rLtP@1a92I&K!(3mZBb$3L4b6Gj-vX4x zc!u-VUNax8u+w)*&ibAv?D86`J%W&cBj)_(87B7zaR0n(#;j|_X?_+?k;lOLYn}>@ z0rLV>tDGx`DB*DJO)RP-#oSMgJi}*y)^d#nb9tIP{N9TipY{rS1MKE{j-H>rRLnz* zm~xHTigO1;Aio?7V2{m+x#nz|^Q(jvy=?7r)?kipKu(!jA9ML?=j^p3CjZDWk=Qv{ z0(D~(@IFUkv{g( z>jbQ+-JU9A$W^Y_pmbR07akJL=SkX3wqPJO{0m(=|qq4=@?jJjua78%w3~)VgK44)=V*Tg@gh9c$%?9TF1Mr z9#%Qmy6e~qeTn@HBv1nHJ+Y3Hsx64RVLe)-KQR@l&+XN|da^ckB#xdycs#14RY4R_2v z-hjIX=F0C`M&SMpK*FxBO-_d#A_D8l=N>w@7wfoJsh!WzwH2Vo=6m97n&Yiu&Jrot zTMzU5uR^!(RbWrmD#sa9B|=;2e63s49%^ifoH?*wi42GNyK`VZBu=}ruDOY!2S=6o zgLv&J=9mTZ8Cdso8+{DlqhN?W0P+t(M0Zb>iToNkPl=dS=Cy`3=gg}}ENJ>b8owmI zOxYHgJMx-CEj=xz*a_NZ6Fx-HQ@gsAJdX zSTH{Yju><7F?0RRI*$Bos2K4L5 zo4_?hU|hm3zg({0{u0vfPGJi0SHSoIkk4L>v6mTGwlKFD%tH&;G0po>#=P!R<~OtvSKuF@8?ThxzSScL#3+lkjd>L@ zgJ!>xA3V&zd)*?&J=$mfJD}->xp51m9D#Vhr-#+uJy@XM0uFm07zYyi8Ftq%>WBn2 zb~#li!7E?)#nV74&<~SZ8Kur)2NSm$vh&HI`%r|d2`120x-v#6)`oqtqvR!Sp(lQjspD{nC~4I z!)fLW$TH9E^p3qGwnQ3}Km?wrF>^2k+U!MM_q+fVdWH0JYetN@`b*>- zdjbV^>*@1+oHu+(YZxm$ix6jzKkxY;`Ox<+!cROmh}q*0Twx8*$9HbLGVFyXpZ>XD zc;7QmJ@(!wKm6ol2jlVX=wN4WG&v}?$8S5_2|soDVG(GrarIsQcl{TKAG}llnWvuqrT6{fhu(Jh=F8T+cb9fv;Of8liJyNb z`ya&Y@#iS+oWUEO_=k6Y>H9+v10DB`dr#y|iw}J8eD*6o`^n*YU9Y?^^L1|uch++6 zd5$N$`;*PPvKT{^LKofau^dr#AS{zK$l_`;(2O^>6l3KM_yK#)a%B z+WN2Fj&^tUEC%i zr#ATK2f8peXkjdzRP z$;Pwx`VU;c-TjChT|jg&+BmiT;~)H6Pe+qt((xU#pXjKUXtcxH&VSP0`}c=)PyfI7 zbu`-9-Tb8e?B9R$uX^rLFVW84c=KjYzxwcZb-!u{fBC%bdoi>; z^bn0llg+Pw@A%S-KGS{tTV+4d!R9yrf8753Bi%3m<=_Q<9c_;`zeE1|=YIMv-5m`^ zFYF`Q?Rbs6*iE!tbrVgt$2)tQACR*z|Mr)=mp1tEhdVpsJMF0B1M= z6D^?b+u!WDM<07pH_>?epySm3`isB3>LyygbI?t+JKo*fd_(^5 ztFHXn`uB?apS8iuU(!u9nT$G~wa>lz$G*@_v|Ma;6YY<9L2m5Pdw%z&UkmTxvmc!6 z=x<)yO?1FtASWBo+Sfn*#}9N9EnoGr9-@P;Z~px+w!zcgM0>?#Z+r8h{l-gP`Ul5Ma`3U2cP?#jG&<<`67j9QAN`H)XYJ?$qJzy>?Vo<+Tg&bXIe7GY zdOGrJ?Z(T$f3uhW@`@g!or8{dc0bY9SNe(eF7uIn)Jrtp-r2m67j$&9ucPrcH|hMR zNdNI4T|jiOxwL+wt*`cWH0k(U>tEW@1wu9$aZ4&i&wEWB?JvV!|;Ir-g@%L}`)@OT(#>M8}`}>KG z`ib^CZua#T|H$8UziPMssIQ~RsN=`F3-;)NSM}`CWOsM-SHFIutthd!EYgKa=PD@WJo<4?d*r K?fl319Q(f!{~?Y5 diff --git a/trained/ai87-kws20_v3-qat8-q.pth.tar b/trained/ai87-kws20_v3-qat8-q.pth.tar index cf9b2a6e2d0a9941047af5e75720c07b6bdc5266..af232575028f21ea161272632c4ea8cabc5465f5 100644 GIT binary patch literal 1815458 zcmb@v37lNlS*LxHh{QNfQh*Zz7%0Gi1I8)sGMMC)2uYw5D4GCbbt$%poj7*YRVfg_ z!(~gvUc+kk%P#h27i&(5P3+As_RB7o8N@80nPFzW=Q-zIb+=q{@_oPW{P^E@|G)1! z_ulI2wt&gTQ}%^$(@o)VYY^TxoUHD9;KA3Odj9<<7bowUz2>18AAe@}xIL?W@}u)` z$BDP1{M77``yQA-X(OGu8SC`oVr|x>)BM;tp>fZYvdH8kv?|S6SVs>V6@975~ z{N%;_t?pYq!Z>%`duH+AL#J=Pw>tI8GY>v=|GkS-51)Rdy7zSlZk|8&jyp~~E%$z; zI{B(o^QV=Pn_v9UBPSOhNjc7&=Qo{jw?_^i+kfQlyARCY`ovQ(@0*>O-@Ngp#r)|X zu@N?&^m+4JygcLcA8^8FF6PhLoDWXC%la`t<3~~|Ix+#*yjB18{LFQW_OkQK7aq~srMf~u>Z*X10J~f<>A0$evjPuZ`2!4 zdgJ^96)G<%Jzlt&$Ibb8{quid@AH5B@on}W7|lPZ?BGS+4nEi&ytuHxWHEo~=KMp} zKeh+D@ymnsp&fbuARRk+_ua=1%{L19%R2IxOa5Vn{KFUX$>#hc)*sJ-yLO(@&FiTT zjgF2E%|EhGzoMi5D5*cXP=Cx~eqwX}vFjh719z=^Mn7)v;T+-dIn3?-_|oJPx+b4! zlTRv5?p@5MoAZ-5YO?MbedXRJhmRdTa`ect`F*9y{auq&Hu>bz)*FU2N z?z-+7ef9PHhYlX*4!@@GKh*KRR{ngyJmSmCdmhry?#R1`^vLKSA4Lb|pHs*`wmmJu$59_Xboj{p3k&rZb<|%h^_LXtFI~)+oAdv;{`on$_96YUy@%ru zpu-%_mzO4A(KY!>n|xJi^3{v^W1I7@xlxmK59!zLZF2PJfrC7!^RFvSzP@Yn4L144 z(&U>K^Ja7Y&Fdf1gV#T#-?EpVn{d>J^ji!6w{`sAF8_BF{_kAO&uz}X>qa+W=OO*> zz0{n~;iLNx%)iGE(w8pg-z#_IG36oszLyWXhxGePlOI^jw>IZLxc(tMwD%$Xp~v+( z>JPLZE?fAKZVNx^7JjU-|M+5leslg4>mSlXdmhrC?8y6vbpQTC2RWagD&#-ik^hY3 zKU>J(w3w&O`D*?79NP7eUU)q9-OT-Sh5F|^>R*uh7Yp?-E#?g`~>Fta8U&}q>G36osn~hXHNdLC<_`AjY^5*>SPrUQa zHS`Za=~%rLDov}W zB6HvBX=sb;Ca#;+TXS9Cdd+iN-Awl0)_zFw`By#NNN#-#TG#p+sJ4D4vb25{+M*h8 z-K^e*>+a91o%^--dqMTKB<}5cWd9MqWR9x0Gm^XBiq>^~HmY5pgDhR&9&J%Qm+NNr z4qVsY%4^Sf^^WB1ZRp3qkwc^Eos8rsy}Aw6j*-XxDQ~Ra8FOQqpZDrr&`alcMO#!u zuA9}nab5T9@BN-zy*p`poBMPy&bLF?R%oS+xJ42p6`vesP5ppSv`;I zx_4sFx8Lf0$m&}5AKM2;ql3o|Rqtygx4aXrYx#btw!8~jT0S3bQH{86R{OcGJK$a4 ziK_$TJg(t>KBB6FMsmYLXkEj@s5U%;EDevMEvjQ&H>sJJ} zTYZ3$-1;80uJs3^+WG~^()xvHiz;&6tj1i|w_f-594C8k>tjdwWIlAL`XD2@^^4HD z)*p;&>lY(S>zAM{s+V%ztUiS6y0_>0Z`IX@lDN0)kptY@W5=ouBf0C#(7LWKN44vR zAxqZ}M_W`AuA9|Ia9w|Mcix^&a`rYndi3y7ex0g5(n$U?Q@sM!j??_3qWUPz?x#id z(deb~$Dl2$6I?f|kL9}V?b-VUy!trO_BQ9$J#zTyk?P}($(Hp^<}+!H93!Oc=wTG)oYC8h7X~24PT3D!-tWj;Uj2^YQ}Z5nsZ(E2(SGL zUsYsZcf35p{7IxbZ6vo|pmnX!pxXL%$kO^#&=%FFa^0+6&vkw4b#Kq7k$qije(W7) z*Pm`AxBd*YuJxm+w*C*u()u&e7FEr4v-&Kq>)xL0zwTF`P2%3JM-Ck0+w!RT93#2w z=c098KM&QeZ$Or=Z$w*EXSr@xpU-vu&E0u>zJQ#)4S9eMjgF0~FEo;WU#-3f)sFY) z-v+8L#_WF~sJ;ZFwEj}GMYZI*S^Y<@>t3I|zZz6uM%v@LA3bv9Q1#_T@^-%ht=sNb zqPpF$LYAIijkc&B++v9o2^4fh-Nb6KzqQ z$?67s$G8;S-QRnZBZqzo7IZz`kTA+`dlDqZ$nO)kN;8i zb4K{#v$3dt9@UO9|HY^J16@>xt*=d_(`_`}X`wUHuC2 z>)ReZ!XLV;Up12V`DXp8C+*Ujn=xt@5tJ-x2|N>lw2!N+%ETh$*M z$({ZLt?TrssCN1@Wa;$hXp5@lx>@}N*WI5$9O1g(ZvK+A>re8?vAg*J{#QnF&%Z|N zdj1WnJ^vP2dj1{SqPom=v-*3k*Y#ZYZu|pjkL!7e-#Dv(G?IJ%6I$2vKcU+5KO;-e ze@0tWSGaCg|HAdebN8I+b-&O27c%y?;|~D&_nH4{B)9u-XkEMij%vGqMV5B|hPJ4# za^0-{2iNOv;Lh9dKMC&b#M^MUALswe$jvWL&stRfj%urCZ=^R={~LoZVo!QQ^?%Sx zyZ?*Euc}YB-IKVUIl+H2%l~bxdNSsTXKFoo>?ps$9T+`)YVo>*)l>A}ivN+>;=$MQ zA5^M)cm9J|wU6kd)mw2b|K%))s;6e})AVjU_0j649JbTx#i?2K)|emqyWjehkpBzU zsZX8Vd-8R!s&2-+cY{X`u*IVXs;3*_bw9AEZb5Z}okq4Aq3 zSAH|)n*SqS*XG{S^XhH!UDxUG=2xdVu1?R!>^nUNy>xneG=3Z9%5S4w z&ul#7(dr#h?s)aJO}=`~#%SlAdnYn!$DggM+fZ#+{&HQtGv>9=v3J2J{oWOg-$A+Z zJ1Eyzui47}{m1-K@a}{=?)O05#eF;GHSYJsDBSObwy55l>t=Na*E1WpK3Y8w<<0D6 z|NguE3I9H1dUP*cRPT#wzw!sp>Q2mSUuoVCqx8EAZBad+>t;3Lx^6FrMh6eveWcn? zxZ^&6x{LcD<~8m^7=`;V+M+tbb+bCk^~}aQJz5<@d2@R?Fgn0va5tIuyM0l;KdSx8 z-@&R6!0dmyxd*+p`#`is^#ZP&)eE_<+X`Pb{q-;s?x@G8yQq(2_S7GQUZ`J$wx~Xs z>t^+0u4gvh{n6?rC~tNr`;YOV_fj&ub{~RTPA>nE<3lm~KXPoKmv%2hTU0OSx>d$js!lqcTFQ9ex% z^E`hHnYQEkI)Q4t?ibFF#q1C6}Pkj=-P`?swQQgONv$~(_nT^q-)hU!WvlIUO>>npSnM~UG*1rnXcHNi%2QaTa z%@1Ofey>JbRIlN>Sv|yc-Bu3eKMPl{CERg8jJk{a5zK4cGmOGLM_W`C*UjoQ*E1VO zAFUQBZ)Put`6K22Bh?u)JvzRLy$;oW-8ZpM!Mye<^Qjo6-|Nv9)u(aYtUjIVy1g9P zKk_U5XAthVA4T28{U0!|aepR8;jYmZ)n{?ttUjCTnT;1bT73@6o7u~eV@Ho1JWzcu zne_AL=;xu@FaI2U0|pNPKS$q)R$84!TU4LVb+h^cuIskJe;Ve`>D3n!?r6UV^&0IL zV-(siK`XRhipH;bT=_MR>zOkrZsDiq>|}NFwfDZNdgRRPboFKUKe769u8&qZ`Ko0lXKmXfr{NZ!n_`Ywb zz8Zi22P3Coeaxg6eDLeP;Gef2elX_Hs}v3>Hjr(ac_yubQd zBA=gs!an%$gNp}GK6Edme4XhpoOGi+m8!gj)794#_{8cPxZc6>K5660{FQaz##?PX zb>nFpH*LK2#?2c~-?(Mt85_^sc-F>X<83zHcH`|fZsj&SXXEWRp1bi58}GRBP8+vv zyz|DpY`p8naO2%J-hJaeHg4Z|&yDxmc<+rnHlDZfJ{#}5ap%VSZQQl-d_J=GZyeY- zxN&IX@WzpiqZ`LI?%sI+jStwkXX67mUU2Hn$us%e&d#@*#p`FMs&8bp?qBER$XzGz zKl$35$6l&$B6i|!*vx&8ymnTdI(?ddQgZL<`yM!Tf7K9rwEAWf{1-#7J$d>yr}MuU zs=g&hzLo1+^lyWzZzJ=8x8&dFRNqeKhwc8QeD}XM+WF@H9XahS|3@;_cRo?Uy6@cQ za>2W~@5lecr|Z5jf7cTYz4LANyK~ywe@k`U-`Ky0%m?0*e;87I?~PmU{)6`S<+Qcu zwf@i0-~U7f>(1*3a>3g3dfcCUx1MO|o%8y^oVND7uKTO%50Ux6Tk>xdsvo{_>)n6C z{E?it_Po~r;qgbGs9@cB{a7wodtQ(GBj1ld(a<~Rbv~!9J+JHjjP?^`KJb?O=kn?& zZ`^wK->iNrr>#A&^?zyl=_e{!cV0h}3)Y_3^9(eQr z>9$&Bz9;_Au+@c}^cMfe)StU?`nps4`JBG%5xC)hwETq|54&?hznGKOoX~ZjSHDEw ztDpBIeo*ZCCy{;Yz5gxm;*I+5{GV4{9(L!Dem5trIi%}8nSPJFSHER{nE(A7_1*o!^anX<%^|J-Sh{rM^mT{yhdF)C zAwBLR>5pza?9L(maZY-IKjZ!h39o)Le}4VbOmd??nf_Uh?z#^*_%r99ugP0?0&UKF z!apbe#hQV3_WYMd&zwH-RK97Q`qV|mcl*D>^=S3iT<^$V+V4A4Ri_?a+`D*Ww!iut z+%JhA^~HblBRcd*N!LE-x1vi`5$@b5`CA8&#$-H zm;a4?_4j(yiDyah(1WwtgAc#z-q)U5Jn+c<)jtqA@eBelAD_7Q_=mjW-j96ghrIX| z)j#4sbLOte#?x;4_dB2Q8xKD}1Tneczutd#c-EeOz5o0phs(b^fAR6Bhh6`IzyF8- zGXKfX|I&|Xizi|}yoQ;7cK+hy`*$<%YVri?|1bUdH#q!};Qv#9wi@007k%SnAF~fb zJkj3ozJBlY=3jfi`1s-f_j~tP=(Y->eiLmIG9t$zEWrpQ?6bs|Ab}ZZ$s3Fr8zgYP zlh)P5wpJlr0k>WVVH+8MxmUndaOZJ2kP&YAhUDD=?g5WsyKXuj!m$pwA?>#$b``vo zyq5vFV+!FCG4m&22)3{r{1=Hom%LX1XPIJ8h&jg{2-9WVLzHP5*Ea*PD-r^S|AG#~=^I9nAir7Nax z1|?`ge+D!1jk}Z+?9j=~hh9hC0sJHQ61MMya){4*tWce(Z+5zi}@S=_rl+>;#En3LxvmgmLo z4B1CqZhPbyxiz|bZjf^1dxGuW%#$-AjSE+b)H8LP080hcx{+x208t3kj z&zt$QC&c=d&kaM~fL~5?oX@;mk1y*;^UO)gNSvG*Xwd^cIjm{Wb5C?<9O5^>0mjYK zHzj7SIX)-q=8cFAfSk^lf!=g^+7aEkLM-_;@r16lw|oJZr$6)Il7EQL9`YL(Q(#TL z&-i<;*%%$WjpU*GnJZW9a^%V!#NC(t6MWVLq>#%T=W_qX%u_DSL55iNudlHNSp&JO zYxRR!7aJgrD^sWUIaBODFOp-={Idf82uy(e)e~{0^R@UqgPu1K6UG{1tv0Jk1#-we zvj*Lqx{Jx_IrZl}z?LhKGsWNEQ#m|0{Q+379O7%Wx!3aKxOMt~H93yHI~MVq?~FAv z64w`y_Mf2VnXt{1BiAF{OWpZn>io#5^{g54%8DFt4#ZgahWM92E_3}%-v;*8)}Mg= zU3Fr<6+a`vMX>Coz5qPCsXD8PWnQ04d&u2_i@+Z8_r2veUw%2fms?;P?7L9D4@C4C zu+I>jS7%ul#vxn={v1xgKIeeFhal&W7sYCaM_nHGQUmRLmI+RDm-P{(STI1ZJH199 z0evx**^LD-Sdr&5pyd9Hkm?PwAu^>BS9dk3oVRi7tqG~zC+!IGEXV}(kDNvP-z@Oibr`9pMl0ez|u z-S5}#dx^ady}>8P1cVi{kSoUsXxG?nq;_2DhR7NOd<}7Pwt@3{&)io`rC&4V-gk&d zdyGI$?POK%YmIb{$B@>0=7TXDz_*V@+^Y1m&S%=*?$`5jHlO1bY=Hr& zsj**Ju`h55*xNN76X!mJ1n}iE$Ja(|5P|U#K6B-3@Fo4A=O3=heQ8peMceb&wzo5% z1ra^hjzfO-b8l3+Lg(8;&On22h-{J0WvxAOUAGKceSZscF$0{53N3;xo_u>_hg;xy^OevG8iQAOe?m0xqS`T#IxmMStdf zyup|654q;o*uB*6`Bu{%CI%cl7hq)mYE@d47GAP++M5?D$_Od(H-=oGD zIRV+9bwp3V9C4kqxN3SdOc6rXh7_N&YrId)voZrXZpx za_py+e~NF3lyiyh{szbd`aNV0OxIcBbr+MfN#!h@af0pgGf+@9WcF6CiEWXq9?;2;yVwP zV5nS`pKDi;A(s2}GG>$4n$Nuem%x4ne!n+bcbV92`2AU9%rliU!Qm_e{4tet@VW6U zeRc-aH^_WG)ZYp+A3mRcLP?_)RN51@%~ z60^q$WE)?eZ`I9#tckTi{wcO|Iivo|Ac8Gy;hFj@ErDn1-fB3rU!DC2V&+Uhjn8wM z0p~jhvBz8oW8e(M96pu2y$r`We0FK6VQ z?!fQ8a}MwIKA``l_}sUA`Fl3fxjf6u#JvxGXSs@hWDK9*QS9X$_Q-p~mTMb1!5{Ho zM*19>Z(WLfw%|8^fNky))cCC#f|h>n(Z2c{B{>1zvvm&r`!1BvtVG<;9Q~fbrB#77 z5&ITm+eqypHJ;5Dwly=PXPxJR=OT6syr*0EXCTCK|7zq2yCyH9+e{%yz^k z)+)(Qsc;$_Vn_Wz+1J$Q2{^C&&nuN)+289(YeFp2`RrBWv(EES%9o{ambR-+==88Y zf>t>$XQ!|3&>#o+jm1>Bw9QZEz+q4Ga&PR2F8_2@=5QA0a}GI%+Q?knfjlk1vPfw3%axA`c&S%ym#UawzV$jF(=2ApzZ943?P9fmS?Z$sPP$> zr$OePN^EEGSsEeTpR+DOjqXe>ITLi}4aga=W`s<{S@ZB40SS`CGvG?IBMPtMZO+BO747TyhJ)2e!xwa3*v1T`2EaKx*p` zNo8L7oGAf0l+GMG*&>5T z2#^=jc#D+3MjCey?n~L*O)T@vF&yIu_EDdKQD-OODHW~;8CT?LBlVB8L5rSOMa=ThWAHCI%A@?4R%N_>E2A_R} zpK~MNxoSI0B5r@za8>S?XJs!j?PT9e-wfn^Qse!o$0$H4zkRlWeG<^;Y|di8fxcBi zoeJLs+j}Ordr<~*^$$7({58JWxNxXP;QT{;EqVfter)H_c1COERgcJEoVu_ggx%jIcLDWwdi6ab!Kyh+_$q6ap$bZ zrLS{1!vyHlwzto$=VE=2Mm=GI$*P>U_c5OVatN9(W}kkBI+mI&(Q9JYxX}~7w)4xS zp3gw;E-tT-V>k|Zj)3km$Nk4x&ZR-Bn;$?_Qlm_<_>F6OHtPEO+!}vAJLJ?}{Yogn zm}p|zv$fjtU0bpAX%ZeekO{Q0%1_RKV)5nq(B-d@5y+qS@WQfB{jMxyR&##nbo~vm zCjV^c`n0L5LH4=!%em&`k4Wch(DVDG&5Ef*)G9NzQDw(pYg3!|hc53Fe*JAKXSqdu z1|o3(zK>Tx0z*)f*R0CV<-vvWGh80?{C+I_ecn7j2XDJj{!PI?)P2PL-hDgJ zp5pgs=AA%JXK>%!U`USN)z6{3mqg4QpIzr@@NE$j=J)%V5UarwjMT}SjFDgftkXXL zt~z4lH{PHJV6ER#oG+ykZ-~3sC6LG1)eh&7^3BHOyM$-4jUIt#;oN?|xU3H3@~qt7 zRrDbk;cvmM~RiL6*!Mtwq{TZl5{lHaFrA$bjD* z`L}c!Pe?o4b6IgQOcZ8Un_7BJui(f96{`~%?;+HcB>V^3Y$b85a4Dl=V+rxc{ z24p^b?#r3%qpr+7(F1h>Y3*c;#GWZZ1MHRXX{%2`odEgHDsK%kKhkHvo<0{ zi{2o^xIDKB>5|L2%nhmVn%`)%tm$<*TI^a0taYhJkoU}7&MAi&p_}ha=9_Cj=WxkC zL|2+`%p7y&bEQ;bxgK3Ub0c-p9AF>w5;4yv_7Y#ezZp5^w)o66&F6Wwc zz7$K2`Sx>uWv(G7=*Q)_{-jL?_ z8RM)8=?rZw^-A~NkR#01j!0(;vE)Z;8uSDL`b8j5?zJjsV$7b-(Ldw*w#lkAy`4oaf1X5aeIx2U3#GRGVk~FjndmcLzx5*F zi@;gT4d|YQdad2%>krtOv(vTRtG)(b&JU$F5p(5P;FGhB^fX`o)agj;8eqJ|o|0oc zVdwjzq(|uNh&0|H%~96aJ#Rp7fzo{sS6#g4!0+D8$@@XqZm=g^ZbY}P_nE7&x7BNq zI$vLJO^eTXk3MFuzFKr=;Own7v+2v{gRMWWG1pj-ecG%POKywo^~{OwJ<0bQJ)b|) zSV$#TU%;m{*YnnnE8?lVqwYh<9gwpzbD}5Y7O>8m8u&~tf#2P(BB!7R*3`i7G`Act z-wCZ7fw(H)-Tb?bHGa>rE^`sL<_g$H-6doM*3DMs9_=GPp>F|a9wE(R`QgWDm13IrOq`3$6g`BHZrnCi-1O`rYlE5_q1EjmzJg8l-z1B5T2UH^k)^ z_Owo}7|T6xk@ihs3eJN8n1DnqKNBl*!8T|**xR@@SAhN8tG0fZxg%iT8L+qU5KB%( zw~sRpiP?LKY_UBDefD&oE#Mv=1NMvL_#NH6HkP^>{`15yBNK7^c^=Mc|C&7G%P~4} z{hm>Nudx&HiLo~Ur1J;-;u5y!FCIExW^RecNPl7ja0NRcTYU0d#J-BG%|V}`$JCt* zdZUfs_*wKNdP1Le{($aYCiq+GM(EB|qu1#AE@S)Mz9BcE%OBB=H|qG@;}z_Hp8vZ8 zy0Z+iMMSr!{c|n#xelMUJR@SR+mDyCG*=Gqhxbi6M21*&pM{jlOi4d_K-R!BwDvZ* zhUgRHvFx*6m3;-|05td;;^uFIDQNLI-!^D~=d2upn829Y1_FFeTUDA8%4(kBILh?+`HG`*RKq zi7kQ8k-uYEKLIs+aardqc|PO}3_yrwKjzMWduhM`c&2f+b8ZRS`_Li-GLW+b=0B#4 z^z(TqZa>dt3poV#vObZM*SPc=A?3-nX{Xjb1#D|wZDbZ;FLh(q$Z4OxM~iKqE5)62 zF;}#m46KNr_u1)n+Tq5P&o1|82Ir5&L#Nw!itav}PS;n1%!}XJ6a$Vjqki}5o@;$z zi0&CqJHO{Y>-3z59um2nYea9U37`S?YlJp(vMTp=*~u9)0_PY2&$I?@S2sns&X{K? zZ@`X7=Wz~ml@l=Ne9l;Rc1>;qt|7iQ?#w8Ue0`4wJ7G^&^gtTRy|p8pz<3MH$t#vU zxrc=A46gi~aVB}i!e`ulT#KbLyK~81W9NQI-;aG4%0B(f4#eG8|2gBHoyB}Ve`ZMe zox>g@eAb^snm58f>GX(<_^i8#U;Pt~lS96_zROqfhgdixdW{}|yxQAH*Jbpp==SvQ zOKmLAi*wr3nVr%5VPE+ZQvLxl0OxczIgHo9m|Xfii}NdZfxQB8pOpsb=azfW-iPk~ z?JIAJMW3NZa?RJ@5;r$tn>)hS!~*REw(+^o2?*pq_d=ORzdZJFAD*wDb#geHXAt3< zp@&tuf1ZW80e=G48q;?c+2He;uy^j)`Hagm1p|G+y(;rrjo~@o-3#=HKU^sAJVGqr z8=T8CoB(G}#4Z7IBDQ{e)xi6e&wN#8nmISXnGyLwYMR$H{BWQrV-HY)p zU~U+fd7RmKT<&Mzg)&b{WxqauEphnsCaWMch{8OZ7=sEa2Z{d@_#cz$X25f65_^gTOGjz{aTivrUXM}Bh8{aD~lru=! z&f$5;WnCn0o}Bh?&>Q@hv3;iHY{0FZo``#&13o#dn;_4FJCCE2=f3Uh{n|nfKuuj- z75=)5d0yVfC4T33-!)R$a|x_*#aQM`$ianjF7CCTc}8yNa-C1Vy`6jC@xtFy8}ZxA z{ar!Y$2kV*&NW2ZQx5CYJty}x0-j$&PLMTx)+F=>JpugzpM9)zh8)ALJ3Er+XUXU| zy!h;;PuNE|Qv$9!me>@(%YDlofYSQWF6mh~haC2-kpY~;-^OyTa^yUG4Zb0A(!~>Y z(6=h*D0fWdjO?eKIy;WbJm$M^xy)^Vwip1p&8ZVo#4<}IF%O9}q<2*xP+#WT0 zs|3c>mwk?Y_HQgEBycZs$vpr*pDngAq22d2o?@9luQAyGkBcA8uMIoPVw8z-pV$W zdULIH<^j;3$Zy8w`&^6kzD9h;%yn;5kiZfcpNvb7!SV9^MeG0)HCyDg!2HW#0tVw! zvyBw{sGB15bGIrrmylP$Hi*DJqcL2--}$WZbEsjqnz()ZSvv&h(d{!Fvo~-V+)n;Q zq&|5DD`q5jAMmr?`kHxNTg2^i9$e{kXB)1{cg?u{U~g+{eCo;o zjHOQYIhmg`k|XyaXD}}Px8wkUn29{POKv?bTn+w+-6PjilWlUWn-FjD$vH&tpDpJ$u0N7H>tYexURkn#bF9k?-8IE- zQbCH{>;}-~%G}msw?JE&5)&=`3U6DBXxfXr*vooQ0gRf16 z!(8>83wOc}*zQ$d=0(KFg_Tmt4;@b=&zc!_K7TkG>=tMb@f+{wZc=CCTT(j$0e#p> z`5om_Vx4B-uARs z-9DF)Ghp2a*#Nl`wzZR0(QBmnavPJQi6!4!d0J$^w}tMKZ$@4l%RVln*XRw>In3^%VHn&awalgfGc`)G?^$I`z+*5tZZ&sp7EIfm#7 z*vloi`MD;Qb2m5O^Gp)bSPNWr>h@}VEVi5xJrT2~=iu4QkTc{Ic^0|sWJJnme!!pW z&?A^&%j-GhcL`qr319Z3H^6%RL-UXgwsP9l*)L*S=N`>#u$Rb$)Yk9WJAZ>Te|OwG z`^eoA56DrMrwj=fvPIUQKL;Xt4yEyHGL;@%D?G8B(`+}7K5UWD?duGlm;Cl=@p&!- zazp_69^g0lQ~4&!s zH+K8+@?AV)@4Ha`n~oe4VpoCxUCd>T$TK5d@(1FLGM3NCtHk87ejhPElLzSb)_;+> zJc)dNZ}59_B;Mfj`}j685Od$Y`w6{&KATkfn_Ghsarw8Au1om**)>J#y8+LUkF6-RK zzT<_@`yj7pX^ro#e?N5gdyq9U;`jZVkmKJV&CT}?+1G86=lls+qwf8Pkz7_iF7<6J z^AE5abucA2q;gL7Y0>lhu7tzBITzc!T9DTuBS_lzQi7l)K7m&|MSSY`v?BVKZzL>J)-A4q`Yk>^SQV`?Onb+ulnt!7ZRo0DmOs7`W8s_*o{G zIEQ{7|DD8#U5<15j`@D&nXsMJ@5h&sHJGi+=a8TG?yZR>7RZ?r5BflUUy=D)!0q0R zx8#ewQhCR`H|`$-mxwu!Glv9N+r*sdF%ZCXRnEs8_wQcL;%`6%_I5d& zy+)u(W1+JL7_X(-reLKrpr8XF%JM#p{JL;ryYZ7qT&-xUp z1+F&YAD6m-)DFP;)l;W?uZ#~tgKvs7N0}G0#-Gm;89@v3O5F@AEb>}x=gnt;KGhHO zN3cY9X7^*Bv0C6$4@mR!eL$M4-YO$tpLxE{pYSyxfEMI48h7sj(jMkq`^>->yWA#) zTo=jl3>y5R#%~|%T-lC=yXkUkZ0n8LLyS6U&IAmBdIZij8+V-8{c2;GM?Uv80{tEW zKF>5E8=yZWB(p{A08gt8q7uI~=1t_mZnZ(`V9eYA)U9pQW5~UFvRvADFe27a<)kKFk5w;*hBQ($6S^*b%ZOGGqP?8z`&uotUM%I#( zx$vncq66zT`2d7Fp(wYHqo*HRP zpEE_IIZY=g=zeGQJ#;3ys=bmQ*XUh2-` zJqhI4>k_&>C_hgfE4)LG`FsK%nkS>-aWG;WRhpyRyeCl-NqeNKucx5Y0<>JV7o8|OC` zf%jkt@)>kKYa)5WE;q)aYq#j$ul)Xo!l5olO3)d?qjBceXb1{2={3%K9xZ z8<#n2O z`eW!yxd-UxwdnFTK)pqp6Oa+)d;CHqKKE_i6?F5wA2W29v5=4;f+4o`xd-|-d3}v@ zP4G>SBcy#BkjNi&u@TZ<0hxfca=MyWa{Ko-I}<)&j=U~8BWQv9h+Vy0+5st-IRWUi zf8H~;x-pSrdFCRr2F~y7H3-0-Z7MUk=Nc&8>kJHmb>_B6_jVB^;JsRcMi8@qlM09R zS39)GZO{NY1G&?2`JB9h^cn5{ZXkCJl6-K=D$gOdCR-Hw@ZP0=Q zW+0#0D&*zfw!sAWOeWFsn9#o)ay7AV571rteA(j&dqkb~5E&AP-QHhV_=0}&qA^yA zB{w3Sp;dwjIrf}&`7>?&bqbl&n9qOip$~XI^4eS3V}R{kc_vcjG2dC_$^I_h5Zgj` zZxJ-)<-X`y5>LQ9(RR{aHM(_~8#`b-ljm`ERqjh)<4pFI-`a?-v_8*+F1K6>sh-el z5YatDXRp!aw|1fr89)n0U5+vLZ#*Cy>|7TMSHdy*`PP{?#jk#jSl(Y?VT+bngLEJAHRLGetMP~9J+PPPL!|Sz1NhpXCZ~&N2Wsx@VSq zaL?3JEa%}oa>!}#2C41dUIu601A4;m{cfmhK>$;H_K-t=BF}TkocPZHXP)BQ1_Kbu zYst5t^%LM+=Gg1Jl6Z^%B8URt6;OkKoxl*dtWB}x%!s$>#>}zax%M6B9Eq9heZKWV z$r<9i2$n$I{ne}TouNerYJE0_AQ5*~&twR^ug-dcm>e#924e2pJwArtzUHM!4F3`s z_xZ{5j7yIx(lgmdz4O)7Pw<7~B`()3_~p-iP`C6Ex3=fCue@~x?gvDCo=rd|ps$Uk zP8nDDfcuN&W#E@GIMru?Fd5eN7_I8kbVE&4*1-o zy+qT=*h%-QKiiyn#8z)X0}`m!@%xU;KSD-)dFhMApZAOI{PtJsOI^H1Z_%4BRwJkQ z^xI24;!7Z)2c&)MXO4WWGYc}}D`0gTXv zOWCHK8sjeOnuHQ}X34$q*Z8e59(x%>UhYG`J)jFhQxZ=nYuR{$}V^Qw`qdR!o9M9;lanWI69%twzQJkF=I zrx>7H7eI=I%N|30+HWd(m&r?@=R9W|G{9VcPC3IAnNsPK`>x9QxDWeIKmhXOXJ=e` z$|Hwz8~99kRyCaRwB$#m&wz7y*5}maK?bC#firmj5^zrSo^Ocmz9V@paIOaUjM?Wr zkar038L|K6xYRq}3~Yf*oi8F&Ec>xuEWs9#LwyFEEeUePXbeA4pCa>I$N(a~enx9X zEBwHDC&&TNSDV+#7QF%1izyg_gwH(td)`VpV=QMNuXD<)w1>5FIfHYmn{Q8R<=58& z{e7Qk91MsrfqBkquDZE-rH%(9^C7LV-aO}=De0eLsq_9hUramukWV|WSafFz=Ce+{@k0GJp-~yoCAJ!YkQ9D!=3OOvyb)K=J~w2M)-QJUheLZ66NI9 z5s9v?ZlC;KP~YJ5nalO)TlluH15%%S8Pv(qDnTG;3iA7ey(G_hLZVK%XX0{9@Y~mX zd%NsALf2=mbv-8djQ4$sL$ma(BJ1&%;WdZv#IX97Gj|O{2+%&qCi7>=|+jYIOYf0c}DIy55z`iy1 zzT@Rtn2_hY+u&1AD}bCJM@V~a17}QdTtRonEpX=rI6Hlb-})AwK%UY$gf(*2`0X8$ z+BpYX-;5kN5_$uSIhT3zUIe~#H5e-02XXTzt5R#d_sf}$tEU*alzE+{1r3-1x#e=z zsc`u$T;o7*fcGO3ckbo5+)JMcA;0%lS%Y&cctJl?Ko?Um1kT%#C(i_21=c&8vY#W8 zGh3D3ay1}!Qhww1a}WJY$|(q4tm)){e9y!gl=7xT9#XhBdHVC7p>F}X?2#9`{cB)< z>!(P4>M;`Yo{Y-$&X0==F2++Q;_|u@-A;72Oz}q^T8S1&k%TSS3tN>K6|W_!!z{f(K&PQ z$zk3la2B+18O!ekQro;Gu^Eup`J78WV-4_`Ig1@)shN9p9A|eAS3&>IxrZfa zs2S)7^4Hk%PK^n4dHZ-mPx!K*c%a6(=P%zB+dguQu&)5;ux^RZeEl`H{t%0wLEC=^9wu*AuUS`*V3NzOUBA3-kdqVlTpX0>p7S=M(s?n`z^B#)cZt(HYh4 zaTY8=i+u^d(BA@|W%KsIbrFBSo&q^j#3trmC%}C~d;?P3eF;Pm$dlJO zu7C&{{O)ZZm}2`ZSZgo&5;9REmpv04o};zu!hMWD_TjIAGuHT8;C^cC27e2TH`LvB zynEiT!-W#F&Y6ecDwu)>U;ep4jdoaJqeuJ$bmge?)#w4}yCzYcS=%`WAjHBkL$_Dc z`ShDtcZf*yojL2szUK_RjfFcuU+8j0WadFPZjZY2<=)t?lnQ5qbbk5Wo3?yY^xZuI zb$K6=JTS%ftWsy^bLM%j6F|yi{{|#<^Yq0I>h^Ovw>1&le3x^ZQ^yi__K2SF$^L-2c$ga1YlfW64(=JYIOH!ojlH9+zzSgNQK|?-v-vUAb=5Y&Iz#YB3O3% z0NL+%I%Wo7y-?S8c2(wXkTc-tca5}n5Uk<2ydRg57r_=d2T}wHtjas?yWFBXL*^y! z8QE`mym0J8*4WNfQ)@r#L;%+N+%?qpv)+EZe6Gkf0+G6_0P0GG>l&$uDOElyV28^{$5a5X6K5vZOZ8#|+lF z>VyuIqTMCcr*P!=o!V%Ovl+-6Lte@j)WDg7u~cklHzl4A*EuQpZA3AS?I4tEHl}^#SYj3z20TpBjGnUb1>hC`0#l7 z-Y)l5aN7lH$?^F%*UuYgOXyqZ4bpmRYy9#ysq}YWOa0`WMZf(5RL9Hbweyb9{rt1W z+RON^Ah$roKSkb3&NhDcQNz>VOW5`>*6THPOZ+nNSr-pT`woG9CP+B~vRR=scdJfq zo}W47yn|eG-17*1hCC1I6+FnO4v%$1d_!#O{W)|AJt70K20bCwvoBJI&zdQ+LALnh zbml8a&u9zTg6E=}ryW*19QL~Dc=_zwLbq-NE}=)H=V{McZhXcjNI5(YKVMu!beB0P zG7pG}I?g3(t0P@4EwoQPf{6VR8> z8DC!50p$HqI|9zq#*&wx3G@W=-i>pgY) zkaD!z{UWE07{D3LZ_sOfX-Bs<;>)$@o>$a|Ovt>D&S7l=c@8+`b(Wsf9G7+S<+F!J zF5ku8Z+_;ZRxWv(PM>u$k`v&VtqPaVSPc?r1hG6TdG(kcUcqXvy&1u_b;2${(uL|fvR7de||o5#G~i*u|t53<@*nr*h(rhdQ3FS~me zA!C#G?}+%vKO!UZ`>M{leeZk5emb~84|VsKRp&etIL}9L2Gs|hT4(tPoW*xooPc|r zUe=i&*f(JP<$%xec3pSUu>MT5&(q6sk8=1OyPx@FzdYd27nk>7{`arzyG_l=)$c1B z=LtB!_XhDNpx@z~QO;-d72mU&x8I%k0v^OopnsOb-*qk5-)H>I>$<Bb$9oUW)Z}=auBUasm*!7M+XL{JXw9MOr{{d$gH?;0G zsP`-l>)O12z&pn$^SORzyXPIwb9#>7z48oW&GqN_7xKS{cZR7B{{-jnCZ0d(bAzAw z&Tj1eTK7iW=w;452Du~-todKz?DYoJSsy^1@qoXXgLCfzo{0a+`_)5zJxi><^Zx}n z;d_pc=ILYqz^c9D?_N!g*t(wUpKrfV)8PCar;~poPIcr1dndNf1C8%hBerIePh#io z#Mar{c|E5X)I0lcaDRARzhg}}-w*HDx&9X3xWiwLnt?y4sq-Z7CHzO?uORSG-~)Jo zZ`U=iHJ-!g{RGb7Li`7MKV#!{@#gI1_d;Op;qToS?49_};BVL)aor8p`gA5RLBe&c z@Y&hJekXL#=8VA?V$Uw8E`U~C;6H$frA@dAT>E_*vGR2fyz>wAH{jXi?Bg?e27&(u z_X#9*K2Pta!TT5x#** zoi+Au_Dle!c@lfUe$c-zZhv1ic{ua# z-+<55xxc~LLkr|iSm(-E-)kNF8SXP&&*P4WJ;xJPyy1H?qrOga3SEF z;|biznJ+E5U3Vtus@hyLcvtG|F~AMq8RUI0Y99O~*NLrFTRQCCr|yT?GhEOO;4{7h zwUeCv?d$yZ6AgqGzI?RRllYF`a`7|Pz1z#>K8bJa-OOC`RUXb+M__qw&SLK_a`E=9 zyc`-^iIa_IV^adrhqOv)4@aLMI-eqxl2|=&{}#XHIg7HsSyL{rYU#d%%0{FW{_y zUEfpoXh+^!HO9{3y*z{8fM@*tvd;a64N!N^9-jM*#(lgG&;5zK+*jb4@7N?S|LXyJ zp5Slb1L(oNtP6RyyT&tj(13f$JNHL$0S)fs3v+^Ja04#q{{l)A>+U{~djj?uaOT`! zzr(qgXS1jC$a@a$kONhZ<_?_WSI~g-n16d&^Ei_=)~fqv9wf0noYy=_| zeP^7(ebxE%=Nt9`PVnaK<2mYk3f}V$&_U8GuB(SL4&w8X^PKttXYK|rpjN_zM*aeN z;FH^*0c%7%pkFz(t}PONLk_HMb*WASRdJNX0wd@tR{zOI@-*39O5uKok= zM6ES7t95(j>RY|wN&{z4`wjdAuHCct$<=?~tuuQ^8S6dih(dsY0WmGjTK@Axs3 zx=w80>X-P(zIUwm(qi>vCz?(d#{=i1Bu#(QRasN2t7-(KgJbZ=BQgLBzWF6d{^p&!&Au`_yhXPfZ$ORW2JtojDm zu=d%2sXYSDTo>-kvcJALdxZ5`V*QEl{aRBCo4{GU&rjfHo*t*`!mrpfuvT>Zp--K) z+WxZ6s4raRN*AlolXbvzHxTgd+tJmYFY9i+A7_!@=SsBJ)X!&pXV9Pe^zd`)JwJny z>mAwWJfmxJ+F1AO4EKW%Y|X-%?P1+SA7B{k$I^G~0F9cOoxSaQL8m^!jaX|Rt$eM( zCs%d)x$5jWK?9z*=a1Z(9`5DZcM*XP?(m7NmB(JO>^$P{d!Ze0?rCnu>S0fLXB}`2 zzYEUgGw^pA{lE|SXJcxNZ@~E?R?c~>@iYFf-v7{e*IVDvdOy!-CjJB8^L-|^cY@!) zuCuwH{m*bdU*AJ#`CIb;$9w9?-{D&XRkLf!3nrmykNoeVbwSG>f}7jXQ2K8oZ<4a zzMuS^e9vw#_iyiy^W%G_-SZPQ*0oqO2lta%(-=eU7@Q`>>_KLg|Tf&~ej z!#z&M^!oL>&grxF4Bm;_3B0>6Sog7ybNhGCY6rgS13BM|KY-r><3{dxX#N6Rz@7LE zyyw3_a|V8XdaOOwek1S9a_(<_&uhOtwDu9MU(vX{kB|EF@!p;J3C#>*{J`4#lRhx# zoE<*#y~`)!`{8>B`f9yX`#j?(w!=4kYwy^8Q|nzsuKn-$KL6r^J@5zC{009;-kvQ0zWWLH>sa%`-WNIU;eoAruxia&D_3>A zLkTZe`4@baIp?#-%ynn(J%Dp{Y{Q;`=dpHcflo-Y#v%x-T`gFLCLy>djXT*4e6udmgnT z*KAtX?h%Rq3y4)SiLI@A{DJSXPEEr$Y=VwIv6aJ@8%ON3Q7fnB!A$Py{oKGL_XcOJ z-$R~%e;50HKY>o&Co~EC9(sdouWU~o)cLd3dOrsft+2Q240i_?V*B-6--T+z7^_Cx z1j}9-aPrLhjBdh(98llP!MV>9_)U(S`}lcvCB9ts$#wQSoc;Wsj@R|~JJ$Q1Exx_9 z_V3|O*iPJl@6#XQ+{3fJ1KyeYpRQPDbY}PUOy9r^a@5%G1M%;v@v|!TXJC*&!^9=6qEbiUN{nPiW&ROL?URUk@ER|J?cE1PKbgLGI7r zub;n3-ghbC8~g?T3BO~VNzT3xICVGd`E~t&cAViIzIpk9_JmEWdq3m1!}r|o)j@kt zFX9s#Ys}Siz)s?Z^{g&?-0{`=PHQ{X@1lq`%bT2ckgLD%f4%$gpYZMB_uB(&@7s0V zU(H9&Z^WL%9`4n_d_e>Lk-8hX3)~0bvzXMF`${~pPr&=^aeOymgIvJ7m-}edTM}?o!G}Ss{5yKlbkbshWm_v0uTH~{{$EOz`EB>p5=^h_~y+yr#)Q8&R+M# zIXuhBJhmOYd40L6BR`1cN3@m0n%{is$lE^-+x4%<9B`9Zotj2$t^s*54x71ZI2RI)A{D1bFV?%LDO>Py@J^K`^(Y{ zeBt|}HRtMa%!Zx7e(vM4?*!V^(tqE}Aa1}r6>99c=TCf5F?N8xcgLRxXFl9d$mP+C zuxi<}_B!@(|5QU=$FCe(?L!TE?utbH4Cy${C>@@hO^f!KW$s2c}w-2@5L?>p0Pz!|(__cHFF7B=7_=mqp$`#j#m z1K4MR1UF*$RrCC^ey6dw-<1=XPvURDermsh7VBr1HH|(GxB&LG{^#eYe}mhz`F%90 zZ`clR?g2WSeZBzmfBw2=a?UrnckdVP_o3f^Pv8cxK7sqZ!JVkHr}{H+zjtu{bJ!ie z!THaF?)4k7z0Sa1)rV&>5AF%>84PA{#tUe%ba!~?_veZA_VZ`f894JD&N;jj>;0KB z1@&+6e#cw;1{ciZOiBC)c+b$#eIxfb@D2NooOfbv5c@O1nugAwoxdmNzMqbqH6P)> zf`GpfPqga0NL;P&9=>|HA;(=d>hxQ#TF+>mT5DY9Mdrcd2XM{&(SPe_tlo*Wzk5a= zeX!+7AT&Ug6(=vd-YSo!k0OKEbbOJ*#_nU{7=Qj~9AkA6Vybo*>?9 z1ToNFuJ~Z$@Asg3DP5WXT1Y^ z>z{yo58?(QfAs$CW%W5@pTG?aa0m9U&jNnC9?u+WFXwlU08`M{{R+%GsP$d2uRR0E zCH8S}@-F9bF8g-idGrJBfgRYJ@xhr_Z||0?pFKqbPun+5{zQDfuKnw~0^fb?XRi~m zPvw5WXMNXm7WcV;J5XnTpUHpR-O)#<*XT5#h-&(D62sIC^?~L&b>fxMcqp|F7 zUq5&DR@ZawZNCYeO*^pq_N-VPXA#y0vGw=sn)im?dv$y@T^-282ljKmTEzC1zrfvq zd2PT2*8bMp*Z#sYh&;0U; z{+!tNR*!L6o41D(yVgEF!#l8M602)ibN092%$1W1{KPuT$vQdUdr9KK8La)poeoCj9T-bd?u7g*;q?+WW<#T{Gt1q&1=QO{NZS;;JX!IoRU8EB8^CShWq` zJDb?Lcl@%s_LLuLu{UhO*B;n_+p`wu%)r%i)qC#qQLq0AziQ%mKFsOeRWEDdJ(o51 zid=VhhqLEjzLFQQo-tmnndH&9BG&i4nhTuw=I=WLKf!=^mOV$qZwcV-{q}+j)D?H+ z{|>wdYo1@Qz@Nc`TF-uyBWG+~0bgyys+nr=y~n`1*PFaPwQ@5S zZ~q5KV)?{Z|6P0NZr62Ya}B>{#7}BmBi8Ia*EvpTd^T#`Bgi|C_t4OEVt@Z_*oqU5 z{JtA`@1yRCnT;nfc5m+?RluG1Dw>vs`Cu$T+ZNgG5>aGoYTLH8L{U4 z054~KQ^Q_~U+-Zq-@Waj&OL>{GX}N+_eiXD?LF_FdSLzwbsc*n_AGM#`vCu0*S!jG zTIXu9{08?2;vlxhHFEjI(NC?u`&oOV|7RfI;LJIjy}!ZFJZ8j}hP=!3WiFjHl_PE< z*1q=9|CK&_Zud&KI5d%H@$dBWzG@!)0dFsL-m&X}&0L)Q%xSGF4e0s?D~n&+rJsewy!K5O-X z|Lzt4PmKUeoLHC7zlmJ?Yt<)x)DF;If7FhIhI1d^mpo$M*ADM~<+GPMW9Rccli0JJ z^?^C-T_gA-*Z09!a^+vKvFZaZ$a{u?|JScrWAbu|pXi-Se}W+AUM~0A+7S?7b42xb zoyIQj&HTUywj~yHdfma41E27H@4O#5>)l_Q#6mv6|NRR8rL4y&du4W{x*PX2FR*WC8n+}^`Gke@rp9Qw6_KY_8DfuGpGdRCXd&%}3qN-wzISx3^`6|z{f%cX&b#jT_Od<$3+kM? zG~UA*oiq6!-qF=N4&@zMche?%e7V54<~tuhzN06(+7E8OXD77!_L}hKjRXH7M?8;y z<_5fR$NFvsoV5*`0gu&}+s~{t((vv#vDGIe7q6z~fU|ceHfJBt<#Ip$0)FiqYu3QJ ztQ#UI2Z7(P7xZ$qa_M?3jd_>z-sm4u;rIS^AMoa#q2@wUdg42|nXB#%-@oTg@_TP} z-PjuLwrN2EIbpqP#NvCd{^nfOD{+AR)HUEde zB(}%IdKVo(@T-n}?Q2bgH>bV>`<=+CwO?8Gd4e-{27{b@YGd_$!251DdAI?r*B1fK zP~&|j_4?+5 zoU=Hmw!xLJHHTiVU_HnkW8a9KRlaKJ@pPq*oWDnTX6H)q4cIftIh%fh37(1XU}UR) z=I&Q~IdBHn*|!D%0||`Hwd>l)nlqTd^QiM@w0+e0-nfT2;ZJNHv3ACq%R3nOuHxvG zar8_`5es*E2Yv^h)m3K>_#=11I;T2ksXNE7S#!OYRZDKdKeX`m{xqyU{LNbY#P9mY zobF?P=c=0P>iq;4>VZ9Le{$BloMXm1tM)fE>gruk0B28oPwY7S&&0J}k(i6MZ(01B z1%YuS4(pE2JNNV0^T#~y<#Ta)KKIz?4tZkx`OlH^>ecvJvd;x;ozM4#eFhW0W7XKB zVrny%F0kHx!%m>JHnAO?-?In#H@KU+Tzd_yz0`RY_Xr>_C)93@zIo#T%m=u`xxc=D zw`?!{eeQ;TfwR70{jQsE^0jZ`!=K^pFJBAmIU-jt^X?V!1$cYvcWh$y8=4F7ZacBL z(wH*_U&v>!XZ5(Q=l9q(-0dLud}yqh#Mar%-&NhiGbfxg4XicV37lK&xjHrvP18@t z0QPXthA*eir9a|$w>X162DpRz&cu4Qcp>M~0^@h@m#*WVuWJUiH{zLSV$J)N_!*o) zf*Ah_(Ycq;&A#oZOX9DCX0-smM&R8?q+gu~|ecj7CbDqh4 zXRfo|@t=Wvo#ge&=?^%6ZuY}B?pJt_;5U7^t^fG4W^s1+a9Qj3$RFur?~&BT;+j63 zbIH5z_}-;w^zM94XSnJe2=KjM^}l<+_Ot#CJV2na-g6H8X^d9=3H|eR)vHT$Ajl;+ zQ6uzy25PhyxCU4I!a2jhTBkp;H#ID4JN_qQ(1|^RXQ_L^TC46%Uf6%aUC6md;x`Zp z;Ec^%?=$TOXU~s7jd{;~e_6e)Rc~zvPlu*q>pbYxIJbB84Aj|Yz<2Nj=X$=b?^b7Z zFKg`GMXqN#vGz9aTo+)SJ@xNE{`7(lt-3v{@5co=|5x~L-~rUxHv?b;xV+E)vg++o zJ>~J^z?(LXcMI!m&QbT9OJ}Wo-6?xJho4LLRp%ZJIKx>R#K3tz5)bS%cF*>NU$yY+ z)mEH`Pi&1n2e^O}uwO0w`;Os8uD!iCHJ#Yn2XIz-`*~OP@%-`=&ZSj*2G*)=Ac*TO z;QrIzp4&UIpY_IfVxNt*_Aqw?#gk7s&uK697jl80aIVj4K?g4TTB9A{hBmN`e5zxw zvs&O@o@?vf!+qszA8ey`=IZb79s~b(SnC_s{pIZ0LEsyA;3{h_);?

cB&uGda^F z|3vJ*1OLS{0Y=F}xtPUOpq%e$FC$*(-t6tst1>YKSpYS_=1J-z6@2u{PymxM2^$pvIC)Pdu{DmB-XGLuNT|O3X ze!z9$^4@OX8BFjM1aKB-7)ftqtnct%p0idSHn3m-bN2fNdct3S>|e2c(EWzp|IBV( z0RNfi3H|}pPN09euFv-N_gOnN?zgRa=dXR*58vKl?ywg$Nj$I(RL+>Z@8#wuTrI49 zt+DnFZos;ZJ%b6nBYE$?dd8ZqsoQ?TCw3NR*H#{X!pk|w$v)VpD|Z0iz(+Xu3GfxT z-!o9xK!9;rn6)RfxU9|m!KrmcQUlCg(0CT-mlMu$2j)bBYuM0&9?S%+J6{(jymcKL zSasIA=fGEQPxtZtxZyuQ!k>s6wu8{B(f(<@%bBZ(eckg74D#;L^|AW({stVrob~eh z0a6=n^=6Nqul=B1a?WOrHbZvJKLZC8hYjHQ?OSKSm+Qn^V;}oZ_*%JoKhygJoWH^O zIrnTe3-J?vnq#>~!@n2@IM#?`pChh4(D>Y)!M?sP&Y;$_?LFdp>;><=<&nRkpGged zSHF|fKFH}iTf??ob30cM4{Y^kAA7pw@9-lDK*u-MFHNl3?&vQ-T^EN&o5apldNlRi zh{Ig0@03sY6TboX6yBxv!upx5@^ARxj6uM;pSB6O{XX~hpLafQR0lfN9A#5b-Sn(e#KgdYLN>d7D0#?cf1jDEm7M;yHS-pD1wFJg+PeFL z^PYr#v_4BU3BK>E4_dCh%sca<($R6>TZYwxa|9V zQFjBo{`9hD{5~^12AubF0`{t(|Md8Ze?Q=JFmJESbzcWA=H)XM zccMmk_trN-u7GK|>hAF4@B`by*4WE)lop-+o`A9Zz{*{~XYd95{&mf6ZIE~Wz~6T& zw|h4D%E8&|nR@g6;LpTKtfqMGtMtU4=M(7U)tR60KKp0js(Z?nD*}GRk6!z}8r(*j zH_p|2;8!20?Yo!$oqiMBfSQq5W-Oh)Jl0;`n;PFc;k~GHUTeI=L>F-Cw5}8UOq^Jd zS2wUV7yiT#tgGe-G@$y>{{h^y`k=Aid@XEGfUO?dpSVV@5&O*S=bWX3x5n9B`}z7@ zTMan&s6j3!<81!rT#BofBYTSm!q~vrR$F~+(W$_ z2#~pYx~Dnsv;kvl-9LoB=fZk_+n3*~{9JnT#*?`23BQ8|+{2vw#M&>9xaF!hFE@bu z*Smq=fOnUu8&GRcYinuYMy}7hw#Cu}*7>{}*9lwa4}x6x zUNL70a`u(?-Lr?tTs3}wb$rhv?>$tEpCEGGN!=a3tMb^9$9Fz8KI;+d?^Y+gXYwvR zR}$OH{!9dSHs_R_CD_#eri zfO>1=uq}w`V=jTq8a3`O?|GVeIQh$!`Z#K^TjyO@Uu4#FwAQF~AAR@oj2H4QeRTos zQwS2QnZ%*TI{NK98^q#vJv7)man0S z#GcdNfvOf=<&y;1``xIq$Gg{c-WK$*4;t+UAm`bgS-#G%jpZ9ZuFL;~&O30a&!7fW zAGn>vE#F%ExbD^lPGUg|EcuFePq;HWXG{Fb;qUo;k9?+soIRfLt0%E*VccC#>_iM-X4|{TZOvd#*FHcijaVpNsc4 z(Yi(+d&Li}eJ63qc^B9Mb)Nm{%D(TZ(}GFfXY9WNviF{?!<+N(3eWJLf&Btb`(}*q z4053b&Tf9dIls>{;k`HKZ}9i)I;&^jS~AtA-w&VrL|UV zfD1hvxBiJ?yR4?+G;nKMu5mYb|{#^I|u6Z_we z3~KA||B0Q;TK}6~^ED54-@%^KdQttzci_9Otr~jV@f(=AzQ=zxf6Sz&fr@cm?yL5G zUGJgwJ|o|ikHG!B7gy~c$NutYM-pSt-~oIl1C7gl+_T53F=re{+^}lg(-|5_IeK^e zZ}9pP&Uq5kjEJXMZ&xu)BZl zK|JJueVotPJ2->gNACQxbSP?`o-1E^=Jl?6kOc1K*#>szLSSnzIP(=_2YhIOGyLv- zs~NnfB$j&sbsbHDj^E%eVB$M>G1j&d~4k+fIS+#wQuB{Pn~(^P5gP}CVnHYRyeow?z`Ieb3@Z}z2BeJ z&DA}q%_C-iSIrSW_}V8I zzwi>n$qYi3|UBm_&Y@``Jd}+7W&xx30+rvJ%G6o|3*B6T&{it-(_F@ z>T&e_zArsTE_5`t^4J@@=eB39;LikDdzd#a-`?=Gw7K}-+#kPk@~jO1SaWI3?K21Q z?pJlj$3CeROwa;gcbxigRf}IW!~>`^PJE$G&hKC2`VQj$Zp0r!C)eS{#8>}B-ev5{ zTzi|-ul`!@L4UxX>2;>=6WHIGY8$zFF8K1s4Xw}d6Zi&dAM$lCs+Vn53QWf(Pc0D+e5x|aQ2fkPOSX~X!=6m z6~~N;kJg!-t?mKp0yohF)>VDth;!*X_C}3oOZ?})zAiU&4C?p z%!{3%lXo6#+(X~hbDiyRU48w$Ilp~eZ|JO5ch|DtiM)Hc$6g&jL4Y&5>Vavwe&4Rx zx`2Wk*m>A`7w{WgEo=utA2qcfvG4UDR_8t1r@=Xw^SkWTbKRji{nC+(SeS73vaSP{ zzCQ=GUx4q0|E+1gM^^;IYQ47stoLp_S0omgbB_f2H*f}xxaQ*S%zK9eJ96=IqV5Q5 zPG5V*cI=5h)~R#9#9ANlPiTbqL&bWg+brUY|U2?`9I8(#Ar)UAsjES%92N&1X z7sdCz?DtH5_q9*GUvTbeuYeoa1p7Hoxc$##o^xwEd`G8#0)4f{u1xOby1v`$|IYz8 z;2Ka9^m30sydHN9ew@dzJq71oc_Sptl55i&dE3UT89m+x|%nSzk@}v=dZJ? zrKhudzOE1FEY9wX_Ly+aJ>gE+H=y;IJ+Sf@V_?1SXZ7%0#O|SWZ+o^}c{Q#G{z$+I z^8s?rP&vX$ynU^or&NO)!sI>>2-ZT|GK}QtQv@M*Sz^x_j>ceiGloAZHJ?Cu|TWsAmVK)?6IE zJ*GZdZ3nl*|B97=TsfB<*n9R^e(g!k&wHr(hIYVvF86Ne#re8o_i(QP=kvA3e9eQE zx3@DTuvZ*4p&#oV{JYns`Akl}`rx0{q7nMuW8#Y@M|^^}r?dEct#=>wY9_wiZ^o<> z8hLFdKFgeXIe+GS!@uDVEwD!yp3fDlXG;V4<8FEocW zz_<1Re*xY2b#Zmy{)L*$!B6ayIk*X&^-er;ae44sYdv?vKOZ&f-M3b(yx({Bc)YI9 ziF*ZHkk4G_c_Ow)y@U9!Uy0pA&Cho!x93%7ANLLP_N`}yMt_jI!OQ8(oruroa`{w$ z*IwY=Z^k-@Jw4+C>pU*c@)blbUF~mtB^m2`-aRh|+((UhVSPVz{@u6v4y-XZ@two^ z6BuZnKe4VEi!*Odz4aZxQKwcehzIrtXYa(fHn5`MyFA|$@OOeIbk^9zdhfb=;y=ir zf%-vgeNQaQxm&K8J)<@5VIP-zwZ+l1lehPNmTFwiIkMc;2@E*3_Dihw z&R}iL7;C=<_Xa+JhOWc+>v{(o{u#gYwKsANSYtln+*g=yz?r`HlY)yp?tq+rL$B6eTJJ`_ z`Wv%fVy)HRQ0uVxua_WPeaBY9T!y_Pb!M)Ew+d#sr>#^XyxYT?Dx1Tk~ zxmv7e;VhAef%V$J>QDTS)QW-c9^Ridt`UoiT()|_*+Z??WnPTK?(^^2)j8`_gEbD| z`ASPKg{V?E%*^@4(n&)ZxA>0#-5G(eMV~rxmx&1?h~*_dw;yM#I=Gy@!eB+ z7VYh@?$L-lsQus{pzbQrf=TWUr_Sfy;k4!&7zwbR)mXhbvHMuBf40s%@voqKbn@2i z{Dd>^@b-0Hd*AUpR;bZl;6|=qXLS1=P@ClK>AcV60_*wQH;)?qdUifj@+a~Q>ryj} zWAWAweCMC!8gM_)W*;FhXCI%7zWa+t>`eN1(DAJu)Y$hdhhE-ZYRz}}?cZNkuCeql z#^?f#HO_e69P#g8S5LX>1*hKmgnK&k7xUNwYGFH^ae^6Zh7+lF zuy4n2SaZ+h>R#l@+tZxA?CJYpKUXAwfG*&I+`#{zU)RhL#BgbzUVEQ}zZ_a?9#=GQ z>Pw40Kz;vW8yH%2-i3GIS$sY%R&Tu0S=Y!-to;V)`rrot5hR?s%@1O`x{1d(H^HI$PuG*dR z?gp9(Cm;Cr?BMNDd*GjrdiQ<;o=>fHu9@pSsUETFJgd3^^amR6RjYo)+UpMIp014L zyN|t`yY%wlPTa7OKeP?srOxH7=a;1^ZBWZy?d4uW4y$(JyLv9&?qh8SAy2<~SU>Ae zz@FAt9XVIVk9n_`r6Y^{eyV# z=^4E@>#U#TteN=Mb!?E=4&tE&J_G;#TEMAyu6j=T%sjDAFXVx-wJ!U3*2JE%mn%6T zkKN0;BGA*<&#CXJd)cRYnA0b(*81DaI`;>xTI(Cunwl3r@vS%S9h|SLhch~lGh6?V zN0(rN0eT+y(^hZwP2So=--+Er-aP`%9r*V-&Ntw!GuJ=|2_^{rSo>)^Xh-ZBJ!?yV z-{@iff*;trfAaRJUhq?%*$VKUSN|F7Ec%INfLhq<1+R6l>W}TjUHi+b)At-|>?bbv zBtPI1SX2G5pRVf}JY&^N@}9wE-P_?uu01o38PL0G-#Gdu7QOq}Cylje`_{+uJA9oJ zJAiwt)fXMq+*sELIL|2XGmkyxK@%+NZuHVNpdH#j{rIlSBX-Y;Z*9jP;0YLq95oYr zQnzicy+6a)+?~SouaAU=W|t8F#@x zx1-LSdeKDkKkU$FMeqkbjVE}b&OPn(zrP;;tvR;AU4S}gea625c_G)x1^yWX{@%-Z z>hB@1YY%nZJn_&cU(YP51J(_Y`T?+k_3y#_EdDKUZ}SsAKrLFe{r6P~ z=Y5_)#~1p}a|b?;fUC2;uHOMK@YYZGA@J|$yc3stW6#nOurk+Peg|4xOMNcB;yCUA zZXB>j!fUnEIdYHvC7_Oth% zNBz&w`Gj`Bl|ENZCw4Du8osfg1OI*Q6V{&YHGuUG;Je|^qci@lN$Nxgey6uw`xxts z@?-tZH;sYIcf_-M7t>l`ZUjDDVBPa3&z$o9?4NLIjh(H7i}shLi(GeRu6iLK_zism zIqd|#Lk(+f=Apk5!&OhXngx49dnYHH^WE$E|C|i`JGuXjc;ef~`KoWMz3gX=cck9u zdjkXBe-At1<@R&4-sdp{`p}{c{0W-49_RONh|SyIyl3y=e#Akp31h5U_f;dWKCt%e z*a5urnjwG8R{Icl;OE7=aQVJ?9`7u{uRu-df;P@-lM(D8Sgvy`L$gARcm^cOPrL8o4vRXZGyY*eBp8Xd>a6G2!=r*W#Ry zLmyW(u2{NV^Y34&NjP-Rz&-swa1ZykuXovFy%%q2)St2Yj@A8wLZ^5FK6<-?}32z{m$X_A86da-V4tY3-UKm zXF@AK6C?1g@xRA1uRq|=*jmA#peL54#@P4a=l#ve2Plm(YisN|gFIHQVa4Wq5CU^@ zVlVe9ZLI!xV-Q+bpnq9EyCZ+>vCm-d0Rl91J&&J#gwFZ|XU`<>-DwkzI`huty-#9! z^RAjTj{WlBiLGl|Y8$?~zz+RbeVomiz3)zM^XAki+ye~a8`k;kalWqk)C9TG;t$~Z ze(yn$tG?;8d{51LK3AP9R)4?0daj;<)vtHR`N?_qapawA*ETr&J4@ofk*n`ueCORc z^@+w_9V@E#h%ea4Rb$TmthvE|029uCr|UcFtU+80-{;l>BsCzxjCHSa0qkXs@D3;3 z_gS2iW!;^A_EKlwI}qxfCBPu(_h5&!#@Y2~zBS&R*7+Mqd{^e$OReWLo`DZMm;K#$>*QTKUIB5eyV!l? zJN4=w^pJO^0OiZ+#@e^;87_#`M3TfGTH=9k&iByx4D>H*!LF~pj@W%pFRMW#5euKXGI=&!+lsEvF=B{Q|pee>Coupy!VnJ`wvJ zqs~1-4f&3J0`_k~3=-UltEZM&{mi2u^#jhe`x%d1`}+Hqy&DMNKmR$qbA1CBP@f4k zar9s=pLNxkH^(ZScF&#o0fBX?+nOTem@nn2nXxp^{~LG)PoN1jVII9b)%spq>#{x) zCzf??`>3~Wkn7+McLG{#T$PKZHO^c;+`}55M__kv{e*Kq|9egQ%XMsOoe?-g0|QPp zpjO>~bVtXG9qrkk;Q6}FXN~8a)I6{*dwBo*E~;Oy=QUr}cY*IuqI3UAPL1)6x*&Fq zoynZKJRle3y^q=(>lw|DLmSu$v~tdt zX!bq+#VavfBbQ);dS`O=@;Uqw*tdfq?@V8gxS?xTz^dn-t9QmR!!K)3_j8RP4_tMo zEO*^GzUL4w_fp>k>+YVp_BWn!d6Jx_5Gd0pUEF+3z&W0CJ(H$Psh4u zl&Ls#&C;nK@It*acoyrL+<^gKcR{Weu{}JKI`^1B%_O$YWzQy9#pTjC>j->MJuv&; z&8e+^SbaI)+D_q+sbcUJ95_Ri@M7^W|m{sK8X|f%x$j$v~>q?licQgo_;Uy z`Lvz9OI^iDpigbXx3&YHsdwlZoW)*SQ$D?{Q|}sie16c`JBj7g=r?$6Cmuk4f;fEl z==l4LYA^1o27iJ&GqG#r(vDbntVXU8+spf~K0xt1kEDKILZ^1$iS?dq*IP5m|NN}Y zsS9k)X{-;Y-nFe-d3CkoCtreCZD|f$_ljS&)*Z9kW5OpGK^oh@**vknTKNgQr_=&D z(X;`s_^g@7{(%md`@ZlKSThfs_&4_;-gow(*9~-X8-Mw_`l=Tly=$LgKM(r~&-@L% z0q;8G^Vk>4<-RQ!9{3%9fYefR2JQ8z#Si!kI#=C)EWP(7tlMX>&mjNvyHz=O`*w2b zz2A;a;=ro6*7GDd0sAzd#%10!P3%P*q=B{7Pn!!LsQ116J?OExAa}MeNPK&>SUs%k z@b+=<8+bC7|I?bm{3MR++TZisLCBc{epc-1a$jrB>kq6y_qOj|`Yv@X_#d!u-5LJh zy}}D=FCjXC>QO=5v( zX#ucmyes1b?rAT%-A~_~z5HGK;}KggA7E?tUe2X|0@j@mzxHHK>+J6=&Y&Hjp>u|b zZ%_X&A+SA`rt0vG+o7B0hP_rdhb&no3Z#HmYc@# zf!>}I&RXZ~_-c$Jj#{kznY=ST0edt!`*eKmB(~ojR35$oXG;BCI{gzpYF4!NwBEWr zGy_fGyCVGKm58^mw13mJ1jamm*=SWB-YG2CzVFfC8?imy??JxwaCLstBUgPrC-_>q`q;P5;C|o@*mHCG&gh-G zhras{Io1jOOikVW5!+Mk(1Iy~8GtkBYcmg@(D)nh4E8H6`RaQmAFEG4thL_7K6jm! zoU!{^XI|Z&#UA!GPA$1{#O~?$)}5Lm4@!&eXd2epMBq<2weDe$1ZwQv@V&p99p6|@ zfS#>lxkhcrexEPb?^EjRb*D%5p=Z?+4|r>|0bD(QoYi}@-%Wd6ymM~-1%C#$hjm8# z)yj1T6KkKQ1p%zFR(=rs9cRwk0)6&!X7wA^_?|e|1MA;GpW)6}wXO^H31kq!+f(1Y zCTNEr*qROF?EZYI&&sjCdY84%>%R8wNiR6}bq@OkTm$y;K5Acb_P3{~`2sFi&wlvU zcJkH)Hjmgo1K;!5*F9@a;sNaW0N!6$cU|}3JQKL=ZJkTb-gorQ;-2Okyxh~le>BH_ z9X^0-1b-a)nfS&m>udioSobUE%+BB8taYyrKY-^Hl_OT`e$J3b9D#;dcus2*&Szkc ziQllU%H`s%J0E)eU-56oJPY^mXZ!=$*L*!c^W^N?;O^wzr{jC4H7^=-Gk@G);;Xr- zrGIF_fD82YvBvqFA@JS99`?M$b>J-4sGXqB%RI&%Xw{$8C!v6|&yJnfJ_$Dt8|v+Y z)z{v)PiT#Eoq>I{*8jYZ{E$OCV!@fj1OXqpVs&yZ>k>TZIk7EPAMZ(ib8+}u&)Bg0 zSp~J7c!25~t1hv-A6}hdCJy!L)L*E*6YKl$1M2yZ3!uhx2e||;@9B)b<*`3=wpe%3 zs2}(VKB7@;y?@spV8Uz7Kf$|a!|!0?tAD_|SHLy=viNG;Q@p7OeAl7>9$(iiJ|lb1 zU2c10&ZzUpKVl;1VuCKj8V2I%khVQ}0b%pBHrrf3XMt1^qKx&*Z!6 zo~6I8^CX%ECh)!Sd$q;dU(S6_V8YqcS}};NbJ<(1v&EtPirvq#;opJ%{l5AHB357R z&nu0SS~T{&V7-qssC$FE!R2Aq?0(h-ctWe~z*_g8`q&Yxx3f0<0GW&TyV3n7oO}8G zGsp#2l%BkOotaT-v?+P=xzF7k@39hRq;>YhD>b$>BUVm!IC#c?7^^c>6I(w-b@cQOmfkxiA z=ErZq`|WV*8;D%+eonxg{XK*A>a<AmKo(6h{I+O1GVBAowD^cibUoiL|;T$k_90e=sz9)YFCd$HyTe*)*sBnP^drL~9sT{HNDcwjxV|9oui z&PT5OCN_cJtLC*n_lRSUwseEI9WYmC#>#siBimYseeAuAuS+-MmOjE?Bv9aBfp4I|pk?{C)D3NFtfCPD3A_SJDTX8>*vOuyzhoH6we~hIU^V3DmB;ru4}Ij$LH9CBr$8xK>emB?moiz zD0K($X0E)?@(HZ zHllspk39P+$6oRhxWfcW)xvv#p6flaHe$`X!&mC8+_f{!$!&uCpS$~drz@O38iS5@SG>}SofzL}&t17d6Aw^l zXLq^UBe1u~Z$X@VM;{W{6Q~CqpK;IS+j9VGlDvL{WAD2>v{Lcdb-8L%9`M=Qqmggu z+R-|Fk?Z+f;JKsuNo>w+8*vWjDd+?`C~SI%}4t#hW$PRM1FB|?PF{ZGoQKMgN8Pi z8$s@D9$#B-$NseA@(ZYrJ!AQYt^aae^@F_iPf&e0pS%_T@$ZO|oZokQto_X=w*PK~ z-)F4#`Khbl6)t_`cwTihjDW{Hdst_Uu?soliZN>| zPyCBH@CF8G+9lZLy!)Dm_H*Xm_LDon**s^qeu6u`=kWKH zzI}5I$Wx3TeUCe>1Ya)KHRpvM&OhTTYU}EL(E;40IS1`rduD77`hg$dM=6F-+O+XmuT(lSv|LWdCvRI>$)31OGkg>?sMK;`SqNwBVK3A z^>gfu-{7eyF?&3W0qgbM*PM4z?=`dRdE{XiKZyMvjysyOCV=yjoUpEexwZT+-~#@^ z>pF{eOSouzDb{sh{6x$;?G_(9dI#SD=S}0^FZ;wNK*P^N4kAc>0R9*0{^eW!L@exctzzH}D-WRyEk((e2tw65bu# zK8ih+$Ys0x8NU8;cOzD>BsO;3_raeVm80fKAA2ij>>ePOzQeV)5o>7Qfj43k-*W_7 zB(VVYa)(CD?_6#FAJEJp+4y9_t-Aa{A3U=d3>HdZzS{?|bIG2evS#pRs0G zW9&y_6Fu5K=}pzC$H=8@9bB6bGZ<)ciepaB$sG^ZUpw3obSyc zgV=}#b1%Sst*vuIw9c9*c#soyrr)i3_Lpxjd%OW_^!=IAfPLjP;Jpm(T=qO_(4BY^ zx88Y*b@mFh{o}Q@e*>(W#5*uP$Xn-oQqPC_%5Q&?S4J+rKTpi-cM$aGv0^ocm?9@) z^-Xl(8*9K>sg2y}=bbqx(3Qv5FFVnQy)Mi;K6f2oe(g=af#1m6*Y|v)pZIFUyt#?4 zb7JdX=+V#}v;Ycqhdg%AJ32t|NZO;v+UvNt#Mh2obI$hJb%U6;`@X!k_HRM`J8!Y# z=Dmj-2x1fHd&UC(3EMk3fuol^rDC~eH?^Etu-_;AH}GChtmn!bOKf=)ZQMQYS2Qv2 zNRD@JeiCz+z;;K)+Lj-EZg9N|^G|GRlf0Z3>v@fpzknJMv=(=ijd6h?$og=tk^=$Cc->^r3cIEpM7;vkeSlxj+V+jy&28z-h$mj?stRa`B10e_!!)X3uF3$a$fk_vEfK))~mG&oj9op75NPqwzyCy zSC035!gtOMG-5{&y!8&i`SxD?OuLVt2PqD={oKX4d?oXPKZ5}M>barM z@`=@5xifZg2-g`mcjU_1Tig5aE+UBUvpT2Fi>{vJjG0etV}b7-XczDmXPJ{{%>&Lv z8w+B}Pt5DNe0y4docFjN`JH&8oinj-=wpZXyQG)gfu87qV~>tq9DMr>pmx7hI~O@> zGI3(@)b2gd^6h6Vfnu!A8EDWFAQ6jGJKz^5SIj&19q=7g4@~D={hYSPO$&|cxIg2^jyHVwI}@{}6ZeijVY|aXzu+t8ev*$4 zFhK)dysjS3R?NLo>)hix{bM-J5+`yOxZ373TJ`#KIoK70)3Ej3Q`-gHoxe|ZMMP{`2e_{``{r#QY`uh7cb_?S6zkr6F`d|QO#Os>v z+5a8(Aa=s`ImuJLBi?fDo3YOC4#tG%^*xx_-pvzuKONuaWnTX`^odx$&)XT!bGEjS z>%R82PmfjeN^R37KG8}9Hk^_0fM<;B+To5?jE9`q_eabg^6tbZcFiQ-@vH6?oBh%p zS{YeteJdD*B^D1Yw@31?*p?!7}``{ci^^>@~uh{iD!2UeJX9Ot%N=a8G);6%K~s`Jl&?lbZKeA_3u&JAJ- z{7lKIU6MGkul<~1Pi5xEXVki*-hBVYH}2jC|3`WRXkd^NksmOwpTw0LzP8WTy?ma2 z7L+quuJ_=bMPTF0OSD`$E!LUUqN|Nla|h1wuIn>qKB?a^xy2_YXKZaBd;51!xjD`q z*c14E=tnMn#<#YFnjY)>xA7OC*vDN0ZOp$<&^GRKYWdM0uVxH*GZxSO?o@M#O?~nM z`-!&qS0Kk+U>{O<7HcoL`Yrh%&^{Rh_A+j-F!%k6sV`SzpvwWBe0^rn&P-$Un)pp0 zZNCZL1^eI`GjH#XR^E3ZX3w|JaB}sM=RAFLfsR-`v=exb_R^PIy|Ftuq|Om*5ABg_ z?%i7KJ9Yk?o7flZqh~|3jXT#nlJCymU0|CN?$^*ASYsdgVJygFC$?JN0P+I3^NIL~ zm9y6yzOe?J-H2It)Z^8k!7t_oIbmP>*<%JayIN--e}2h-0pE)Kt&u-~eG+Y*u}&}T zq|Tos=8Y={&wArRF<$q`{JO6sXKeRy2hU=iyAASwFL`Wxn4`A(X-}8<%hI6Hx2N=e0|nHK--#ZgQ-ZV?TQt z(>LGotqt@5XY8XU;F+_h{Hh`5uGU^a68CH~$Rplc&y{;2Z@(LO1N{m=HdmiI`44RS zC>Q)2kgK+@^YSw~ArN!V6MjRB$OUuO+f%MH>z#gGJHW1)_G{$r`2|dD`?yc_#II*{ zG5_%pG@rVAoFMI(C9%uG(DjiQBVsaxE$DNa@xhMKwoCWMxcL=X)bNj^YbB0rM za9gfe#L~u(&yIKV20UxcQv>JdS1tUi%{9ZAb!QOREmmH?n*-X;u}9)JV6F!?kTcN6 z5_lG$uYHYa+e2={*S>>^U7C1;lCAAg_k-WRi9M*(5992keY&=~r!+>LGko_FZGK>T zFW!@Psf=9n>t68XmE`KJJ;SfNXJ7#LRf->Ic8IO(8RotVV(vf8?YuMPy5EI9<{C)g z9fiHIgSwupx4AkebDA9Dp3O7a?}>JNrTi-zS#CM*3#p8-TBp3HM zpFC^r*AI<-rmlJslj98gm!QqnjM)8%)qHpZJM{P5jAhph@{?Q`V@BoBN9`mo&%8af z>wNU;*f)Ek?cp;wCQtnr&a`>)dG%^aE>79OS8}4~ihp~*&ah`m?sq1Lv45yJzxKe^ zPoO>Wo1O9Ny=2BTzH{;#u>OcQa*Dq3n|m?CJA7dK9^65YPjvm>6pQ=qPS*S93+u;~ zdL`FvcS-Dsy(8N5--*AW8+AR`yeGc%o$32$T&}gw50KdAKW+aj`UHLgot)x4|31>1 zhTYK@a0Y(92Qd-Y@&@sURx;6^(>)S95}UhGCurMOk63jbt#ALbas9jo@kzX)opA@w zKb~Wdv$v8_s(gE7u9+RJ*sJ*LRWbaQ+otcg@oM(NVg$zsiJ(IhG#P*EpvEvPSA?M!@_E>v)AM&g* z_v0%*eCrdF6gx^KR&fWsh8Eu)jS!zG6IM@#=YSC*Fg3<6;71 zL2QB>^fdiav%;%`@9_q?t z-gl2l%o%EH2Ka@X60!RGJ?#R14|=Y6eFm`Gfh%^byd2~1A87ZeUa{Y0xQRW;O>_tL zuV)PRf$eA2-Q_hfK=oo*?dqRP^R;>$@w)r?nA>yZ?c*H>`b2->+{Bl2!}eYhyBlMd zGrl?HafJ^a#H?*1mzFbty}X-Y4jHjx2hSOD>$!54(r}E0nwWj9bN>4r!v2}`yVt#6 zuDpG%&G_&Ke!~}CjmWXz2psGQ6l3z9$6lLLapt{iIQE!mB?2Vl-V z_rUIO<;@Mx=Q;JYt?}>Zeb17Zy`9_f8-CppdJ;G1d(qJ+@!jN-+z<(I$-VxsEtomD##tG@K~jLtVMcYx|oZW`ZuQ2VVP`0|b2u&tZGd-!(O zBxcUpnXSZ_zS3jy-}fSZzpgufpzrw3a*jLbS1rE%Y8JX}USOMlUe|u|<+MQWTwv?l zUwh>0OSE?F#Lm=N+j7-eJBX_lITJJxx%}e}gTBTS)NEY)99(_kA59ro_xh~*stooU+2*may&-@`&esCdx&f$6WiF6oP9F!ca6CQj=l6}5(5F; zBhdHXcA*v8U2wp&aa?PcON3S7`TdXy+w*qR++<`&eWCIH&KZ`OMWv zzxIQpJ>`-!UmSI`ipldHJ(E1+?s37l#`t5~a^-XoTkC5#fsDk2{{s9zSUF>{_kd#` zpJOLBc3j)<=hh!P9?#lZkzfFSZ>VRG>(5?e&vM^RUWxF_()Etu9({dA-d&O}L2FkY zEq^8&C?ofKAGxA^ErEY;$=_OSQy zy|fb-=B*vMdk`wyII%E}9bf==ylq%_!FG>H&iE&A)Fd(Ml=m62-j(y_KG3xvaeHJg zZQm2<#MA>Vv=gnkUjuS(!~^Z!{K@O$zFo=xu=u{8H!$H?Bjk&_JoEr+->t{z^}O6$ zzZ0M4^wG6bF8gdf@ZIxV=8XmU1K!IU2;iT~?&ziu*Iw4#(1RLlo#T!|j!4j6*Y5?# zbM#ztoonx^QRTJ<5}w~H?-$8k-K@30yVf1Vj-lPdd|=D5 zzF*hxqyPAI-KY1|iBI5uFTi{E`Fn3k++CfQxpuKvodceC7;r=f-d)+;P3lu;FXxPe z0|s%q#>^L=Uh zYIDb&VLaFG9NNcza?e+|XytvsEVsTd)cN2{V&<)_-cYPF--#QuSNptqXm>t?+LL^7 z@I8k$_t&+1?SNnU%C6+^a=u?R?pd>Tyunp&_!r<_hdehYh#6P6Sg@b_+S|Qau6rEh zjr+NfU(beg5SM3cVt4e(I&Asvl{mf}a|1L`z2#xcKkln~;mf;R$;X;;@YJ2y0Os^3 zurBc*)O-O!tON6&zwV8*4g9jX>xk8>!OzVLv|{c-T|JlFcYv1L(Km7dzR01kv-Rc2 z6^(zKsbSl{&WWy^F<_0Kzh*7|1m4?Be4u_KCyZAQ=C$B|K;3dZuZtWo@SS~wpX38y z-kqFW(PPc>E(W&qg|<0iF3=^p``{$@1RePA2>Ex#Pii}OfFORMHlx*C_Zq|ozGn!u zcP3|IKh)%#addHr9l8_OK7U@dK7+t^)(s>z&hX6MrSaoEO=^Ypet&wRCw|MdL!do( z#mLpOB&Od5wll1q*cD?=X<|PC>)lPbgEb@8`Hp_Z<=Ug}RSqz&E{QeEvz$QS3(s5J zTxY8qu~Gd^}!#!u3bi~z1`jZL9P)qo{OK zj>We}y=Q1=%X40jt?3~*h$WC)v#~24E9N_DFMCuicIrC|v8Q>@X8a7la>T9i4DRd< z?Rp1e)lIay>P=qXUI&a%aJQa#0|Wn%t_|$z-1E`jI5LAT#I1=9ZN(WwB=v`;Pu&@J zF^LP$lH{NCNqum~4sArv5V2wp_`WyB-P`kZ>>zJ%d2)UMC-4abd1oAV*LUo3UAs+k z4UTi`9b(T5w*8%TM!$do?A75Lo7m17=p*hk_H*f8*2i`2;V$;*)_V`=M(#mOac?>5 z20Tm4<;yoFzv7tI%XLq8O#GwX-RgS-+V~S4aP4PZosAkf#wR$=`Q7W{4`Kn=GuZC| zzXNwrGW@bWSLzb9=hvS=uDu(GL<94p;`j+77T;Ol#yxkElhe>Ylm7xbw)@)ig75DG z+H!8h&!FLZzK))tVTXwIoUONyIqye*;1B%T1Ko)~z=d2r18n;_bHY8N6I3sD^?RHB zoiDF#PJ?5O`~kd|qi4MfKHo!f83E6}^6aNB$-N8y%V#37eEYbQ+B0~^wKKdo?Zn1jQdDy{Nk`G{wXLVM?uHCkMXI=m~&Q;_kI8)=kC-N6i`JJ{GR*6`uzGm?N<=Q2C%1j z_f&fa{!VOv?TVr7edL|t9XtQLIdy(wBcN@#M@xM01v}AmJ&J$by~x|+zH=Sl z9g~=~LEM?K<8rlc#NOu-Q{=g;BJX)!J9q~Jct3LPz}|I+t&eykS2L)0kB*i%(X|8l zL)_s&J?kVRl=!Ig`-gqP7;rClkZYas1p37xHnElJ8)$au`LUbv0x|2{TUckWf$ekm zx!F^Wc73+c_L{`iO6IB^C>xbxiGzt{EeG#eau zxnMs&uN~je^*N=^U7RV;e#gBVKf&uq4&S^8Ft&Ejj_t1Q9r*UQwh_0VdICQm{?2~e zxj{_6vxb^nJsb4Iw})}@!1nC+vQFPSJNOrTc|lI8-hq+p_qqB!%7=G6*G^8pca!MA zmp_4a-3!`T;syfrSbS&He6;+>b>$nJ>WB7oZruPUV6Am+>+{YLS$DuYme+M(@;syU zwG)2rr3T8y`dt0%*LuZF%10s~5WCPc^kiZ|l6jftD*~EUx#fMBsZy{A8B>4&VC76&vW{5Oc?g-?3{RwsrOz zJJyL+t|9OnzCAj;;xJQfuI6FtPh$ztsYbp~^V`>5n@it!pW;d)K0rzE6%1hS>W>}S z>TbGm>g#Ta9d$`OKm+bxHm>mwZm%n*Jwbqyt%n+@-;v9HW8QIBb!Pgzv*);giG4;} zCr?c5B(C&WxLt9QF=+F3HvBHAKg8$JV)ZlT^N!ToKiht{=HB30=X>JM9_P7h#OnEi zt^dgw@Xnki-&)~$2K+|+xa(toKdaVRBj+L?-HfqA-3k8454&>sv18`m?<}xE0PFJ` zJ+M!ER-G5$yz|GNSNpbg-|p1OTPK`X(1w2p&02h+|6&Z=y%XJ%=k~hp;k@}t%yas^ z)93U6#@yFF>WQ}Ac%aX4Jfr*TmnPSNxq%kny{>0avy4S6F(+4TeM39P+KI1i{%QX` zqxQr0nZ|Y1$+5RRJY#W*=?8kCOLFx+;x%jcNqY5I{HlSgf5gW28#0KOsDYe!ee99z zF5T+|)H}#-@&Zh7oX>f4!tO}H1(_G}~2^iwkYvie!qh@fo0DCEQ zj#%1x$IOTW*_s$C#boxxPfm!_wc_(Vosht&DFcJwY}sHa-Ep@ zz!%=D+~<|^0MCB|5AXs!ujl*~H0rGXvSUA^-P>5{SbI0bI&0j=T+M{%F7>?udw`DL z;G99j|GgT?C{_2Hx&Pc^j<~(#iU~duzkuTo?caCqN&Ep4-!n!mUPH7Md(>USQ+`7K z2&_}={{rex0}bpaZaeN?!rTN2JlD~~`1@U9yQ8(fKLfvXZjR5LT*vp_Jnr^FEaZ@n zfP*c*%Jn(K>SaG|>*dRlFO2C=Y_(_gIeK3E+j|f5UEL;~We4g%-` zs_vN`jL#&D*6#4D=J3(x-8Z&A%fa?Mbq>rPd;t~TJ+vQijLRG71TELT+QxkbotV4; zCAo5U>W#g@8NitXUpw`&11+b6?_P-+zpmN#msfXA{!CnPw|X|Aw`%R>9;p`8y6eFW zVh?&bPnog$jbtCt@QodMXy?cM)jUG#N!&d%(Aedpt2c4`bnx|xHjdpk&ejq0ov~IK z@xiOT@r`-j$X2x-|6Bf`kG{`npslm!-~_g@!nZd2H+=J_Z)%CwnS2+k7M^F-9-skz zXL)x0z&`pr*L}M^$<=)kJ8I?n9;okdn%v**-z|uh#MZgqG0!-#<<8(}z}fa_z!~LS_95ow-KnqLv6YBYYsU)g7;$-(*WPE9#~#aOpK4+=c7JpNcUCuGPEK*D=}8>uE6(6QVt-Fr5ERLsysn=W1KXJWw0$mT(6NiJ2F0OA+cV`F7xk{h{`p7V zn0dvyp3|Q8o4H~W+n(>hYzg32Og{by_@Fes>}URX9^R+6oHPz5aQ+K9|GQZMe#QQ8 z&<%8K@5`T+9bStCzh6)M1kSN<)81NM#mOlV+fOuh^+6S1KUdtEz#jTL7S5*cd+)Yn z20DpTliG+e?>NxL?bEcWSBzOBN6Gv+hw%sT_g?rtR?VFq6m!PpcHlDzHFdRP?AgR7 zcGJf;H}vg;@3R=_uj(DE9fLG+_jE^b1Ad=>KLdIDK0c#?@xV9M#o_0g?L2!<;IsVh zb^Uu??x5d^dv4F5FZZsGM6SI1%=oy^s$-7vBwoA0mEZ6Ox@#kHB0hTgew#0ebr%oj z9%r#vC$41jCDuLztwb*9C#W5Wsg%Y6<-#wWhI9OD&3U&vXbpWq8{j`8Du^iTNyTnyv*8EpwMXSm}HThX@G=UaVp{rOH`*u$PJ z_8nc<4uS8p>gei`D=&8dW2pvH9Uo|joB+oCyYPX2VU}Fqqr1Ja<(Lc5u!DHRHs?Lu zLG6j}IXt_%YT+et7wgVD*RZSCuBqOH6rW3nXI`<7{m#U_ANv*GIDLZ-^3xYdUte!Myl z&zTzbBzD?-XX!T(z<%=2uWP44OzE+54PBrvu_v(3dNJaohdbPz<6PwECExpT2knU6 z2VG|;HbFQ8?HskU`k{BsbJvVLtC8>I)MXP_JEQi1U%Bg=?T+S3l2k07dIEC;G;L}+ zS|}0gOfzvnfD8Q-JpvxKxr2AygYQ%2nO$5s4R{XaymbqlTr&bKGFrtty58sS_h*%y z-zSavK3jL}QgiY9hWUVdNX6O3o`c*k;T&iM-Nd^|eO=<3eBYT#-v3XH0x!EmnO!-z&CpYR%ireD%Sf z+wSPY9(zu4u=OvX=E$ecT+PC6O5b;Rx6iJK2}%|OfTiAHAdh5U`4ihkyrEAK*l;3szNimmv&wz-z;ckf@Z-R%q> z;5+gsv^~Tp?8w!_chtJn2j*|awg4VCQY)kB|Z@u&D&T`q- zHDI4%Z*&vbcQCO<^^4Uz(AJsvTuJ_R@QGXR?sMDk{fXOm0DHKveXI>~@||VB3-xk4 zF?$(v2k+E6xm82_Ld<;bn6^1#jrn6%)e?UY^Ld=n&i9V)z`K!SFJnJ`#>X~hpGmDU zV{yDwa~&Pz<*N^V?E&{rtX;{0k>rUvyY7`#NiN@5#M)8ro3;2S^wS#VHR1`}(YtaF z?fTgZxB8;pQ@eJ=wvRg|`UQOhht%A}10$#**RhT3E80GT2bj6;#GcMR5wq5F=?Cz4 zZD$2>dj)o)zreA-wF8(BeE&H>?ds0N-iVnSpaQj(Tg4 zGavk*&bXYCpe`5hL9OSs=124)?qYq5BeHsQVn^*Hm!Mm(1_K08Y7cwnx?f{0R^HwN zcy{;i*zeB+yJGBAakT%e`}n&j@1i8uJo~sqLkIAF>JBmg0@obkehK6rypCV{GOKzV zxy@^6?|h)&-;-RujP+ih>7B;#PjEehzVn=G|KlC7 zR#;=+9w9d8jPBrF$jNiIH5V{)&3VYfKBSZTJu&-V&{G?{0C%wNm^C-&yH~2oSyy@D z=6s(9u*Xqn&mbq{h%+&FtG@U>R?Ufc;x{me1-b)yiZw01$zjV)aHm$8xqNr&!1=Y$ z&Rbvo&`l0K#ZTse{4?+~64%Ymj~cm2tuyVT?R%umSo3?X9Ut^49)90BYX*Kp8*|n~ z%kw$;%&nJaUatNzxA^!C$QdB8>m2AK?l}g2#_DUlVR!V2oPC7fWB=AGIpPC<$^i)) z{?O;lRgeD!p4UBvbNozufA9B@@Tm(maka3|&sXLG^T(Og^`63=a>+g6SaVT>Ni4`0 zusb>H-monu_>dFBp+; z#E&E?GmQV!S86}6pXd6;$Diaqr#!V{@5_#5!VROy=k-k7&HYPfWvV`6 zaqQKox6W7rKXSpIKI4J*_a)~k7h={-aGWd18Iz;kunTYxe$Ta+J12QRkDf(8@a^@> zXJW6*x9$Oseca=WuXrx!j9AY>-EyB>Tgf%sIo91k+cn;OkA<(smEiZPAJ%hlu6xJ{ z;JFg5{J8n<>)pDalEmMDzUQcX@JxSLcO~bH?cTqDCrF_G4_?<@3~D{w30VK15^vb{ z)OMcnMolNC_}S?J2e^Z~%Sn9WYGdXf*H-_W z_GskCwmoknFT8KL;_|xs8~1k{dwRC7^0Ae__QSgKk^64Z?s9>1qSuV|?<7v*)_fsW z_rMN*M!gI1pVQyoJ=Wj%2VDC&SNwO?*{`crJ zT6F9gt5@~JxBfS;>-)O)U;lv^lCfrcR_!O;WA49(uSD<#J9F9Q%SO-GXF~s9aBkEO zY-f1Jk<<|D#H{;AaD0Ap-0ch$dxqZ_hcjcp`)K(;61!a2yayb){rlTL{HuTP zpZ)v)`u}d{ze4-J(r*9oKlzToJNpBA{6FMxwEwF8+5hm*{=5JFfBYAJ`=9^Wzy7no z{dfQHkN%6l``h3C(ck~!kK2FymlcMDJN&Qz?x%mk;3odRcKGA}mmU7_x4-@SKm6fe zZinLz{^h^?-~Nxk?EF9e$%r)oux zIp6i;`(ei~oA+4WFt_s!F(2T}JFkpbsPARDKaa+-liUP;U-7f_Gq?l$DfTJ*h+Pe? zT0gJH=Sq(KnmShBL~B<)apzC8Joh(ua65K^oS)%9r#9$d1~#Y|=Aki{#Of)3fCoK< zwsirN`s@b=s5@j2?FkZSt3B7y&t0>5)+kT%es7mIlK^nmjJ?NaV%Wky7ku;HS)jE$ zcoKKdsZZ~EMs3w(V2E8S-+Co-_X%piz_-@;9gcfhf1FW{x%V0PW5)t*K65?$;$a(W zpze&`;~P7^vE@_^x+OsSE=_I3`jNX&ti7xk2|Dl&JZI144`SlRInF?y`I8*(=|s%B z0GA!V?Og49YGqBerZa;S#*u}l#$I4xZ z%T*%zq8;l@sSg6*UF|ot!34SOo~>%+nhUhBrs2!8?+s+GxoYo6(N^S5;MuAk+j|*k zxy8$+2f7otrfnR)T9JRI-e(~1=&u|Pkl zbxy+3p2U0(4c}STnsY~E_Vn!4kDhX!e*+zyf&E$nHE?E+)#rXChA%oOE-~i}eCPQ6 z(V3pZI%oNgOl%?N0iNInDc-M$niaEK;A|Bkoc&pp5$&| z=Blx70B0Z1?sQ$Y_iTT;O8G%Q^ILl-_U*aOm1=wu;0}HU?rmT1E~&M?1RbD>B+lL2 z%e+0#f*N--W>2}!wU@C316cng=DT2APN45-d-&P4w><-WN4vZE2CVa4aK=fEZyzz? zH)2(53_rmDa`l^t#jjq}duQgHFe}xh4n5A82D$OTObP=G{}NeXw85(^EgR!51LU-ad!l zgQ|&--K_HrCy=0gLYb={W}VGlPdHwtgW7v~;Y6J6X&)3nR z4{A21TzjhJcC`-stoS!Lsn;U|h z`!%$2WA4`las4DGZ*JV+)l7VIJwMLc6l!ZO@doT+t-IRWUW)U2to_v1$(QRn?Q8sj z?fu+A5^Kcj9zL($_E+RqPvU0~x$dgHe%RJO$aS=~edQ_EUdWAH=QAI%^7gi82gbac z+x8oN0_(L~0wl;cLA>Xx3GG;SFx12+@O!3rV;^Cz1NX~Z9JM{2Yv1ZcKH!-%*3p+8 ztGnP`sy&I@vw^B3ez6C>wbs@C)`ADQLF^a(%?UZg75Vl`aDy}P#lW}kM0-a^jqhh8 z-!|7BmCQBEUe=z82fBjQ{Zq4^aKlx$TCWsPp+{E{;8%CyrV2g}p>(-oLz|rQI=-9$`ojB|x%xExW1i3Qq>s#4d3}31%ij7I;4aRQXTF0caMsVj zdGc-`<$^b21Af&rtKplsmuEWTdk3CZegksVio6zpe4?E#+)atxcSYWQjkq-l%n9p_ zSyOTRhChI!_8vUX2}FGCBfliJCU5Pr?;!5}@|-`_zRpV!$svD@az)wj;Ov47nIegi2I8!;|F zV#SI_Y~Ww09l7EG$2$EpcE^^h->?tfeb>kdIUuv~W$^=T{1Ep$=6K(8=V~wP@eO9K zT6-8z;2h(fKE{pNuc7r%*d4pUQ;gk+)yyFG{d(qi^75>=-`KW$xTCpFyz0>QNnkJa z1V_%=ll`c*u8}hqXt^EC4ehoO=!k`eHdj69qwn1+@*da|-GIHlJL4VgS+uJMwP_qg zYz}$?V@J*&PkMSEX$~!ncM$X(i48oboCmQdXd6{WzJYIR&*VRC$oEVUi(_tJ2Q>+* z7s@%tM2mHw>Z(J_8@YCon<7@-(PI*8BKAz}&-+Z;vHap=+eZmFCs2DP0bnoV+V0@* zK-Ld@&uUEm>AGtF+Ux%JUeO8Hdhge}8*sxqv{H4(bH7hP5Z|BA<|h6b`vET3HRF!1 zpWnpH-|+2g><;Yjp92Q)&aJm+?Tzo5%~>PY-ih`sN{|!gjahGB?T&7_a@Gud>jSQw zLCif9zv25cMZIMcY-IlfG&t9 z`I9!fshNAk8?pL%OZ*H1+^?&~I=RY?m@;zBdV}`@yzhtwXC!%h&#m3TJJ%n{MCzsC%G z#JZ96B7V6pu6*}wUH9$tRqW&Qw65MWZ1=O? zT61y&E$5EDf#>Jlm-{iFKsh+}nz44bpPVL=|AAkLRUF%Xa-HG4Gf>=F&P02EKLXDZ zu{Z;5uj;G9ANoiSzz#R~_EfCRzz~Rb#d6KEw%#}R*15-kbHT12_})o_=Pn`08*`6| zegga3%a}XJ@oWRgwO5d)c!z58`-k?_$8yg#*^`aL#R-yk-z2hjFTwB5_O zomheR1>8WAZ*umztdnn@G6FwWo%7@ddVq@K-}TYvjmepT0zOY;+Fj6FDIF{J0>@pd zN75fC?sTT7wUr|l_`kya1nlh@?Jw6ewODsj{fS#w{fOJA`0P+Mfeanv=#J>vLG$KyZ79Ewta9y4!H@GyL}SC%UJK#9s9^% z$PZwwllObR_h27q+pA$u;5~bW$B`#Bh}&DSrtUC5-cNgY_Uc3YxCi@apVZVTa>SXM z+m1We`>p3e-29M7+?}icj@$2v-FAM!>DVQ)c1yIM9s7%!(1Ck2a^?s57f^S?Y$bB< z5Ub7_=hd@^JBb@>z~3+IrN|kv_EpQ7;MgIE$@%&7x?AsMkeA!R8DGg*enUG$PJr6m zdAqJ-&%j4Iu_y2xK|Eshv4=C&4|0i~2~Q5_*n^l-Gl*49ti8|ju-#SIzY$aHb;PFe z+`zWi0M3-}4EgS9%`c#yGdh5GbOD_j?b;c;k$VAe!2TEX3DkSSx!mo3UGI@G{RT=C zS9dVg*(#@g(Bl(uW^u{iu;q%}&~{!Xar)o3eMS#rO~lsN0qR^^cY4}OPGBE8=737r{Vo%_H_L#t!dFPz4559XCxA$S2({IKStr$yu zYPGE?*|xbeId`+aXYqTm@dO<-;!O?YIwxac5UZZ%h$|yj&Yfqj_#_tC4J|U)9U2S# zIv@4wbO!KyQpz$Ww0z_C zXu!F02as#eo3U-nH-7Z$*xr+T;eF}b$GcT(Z~E9{1_9y|Tb^f2Vu}3(0hAtm5lEdM zEzh_!%sKl3UmSfJ_8{lZv9&d#B{AS^ImQCq;m&Pmw(8B-+B0L#b!YdLUy{T@-MN9l z-hOQD9W-ofF4*$T+0)o#_cSiI%SU=H$n#vz7{<_z*aZ680jhTA^yNja+P3Gaou?fA zjoCwf=BjC1JJ)C7d*->1e4T@ufY*u3KV8>-cj9+!xz=>FQaV1KN#v@ZXnDR%_VUlD zW5?9>Ot*jRd4V%y)g)SxXS^kd)x&qp|tKH-P?SNN3@SSSN#0?mvJzCu*~}i;qKlyRy5MKT8qezd1jkwIZR|$Q8solW_I8$g8Xq9=1Fo3* zf_$Q#bq3CA0-3ou_B`g>qk};%K?gM(PQR{q$v)=mF46XOhQ4zZXFurk6=Pe5SnIi*-{3gs(Dsp=;BFpx_8{Jg zi2)A3VIMHp6CeHwxL@Mmuj`Hz?KzxZKv`S&MSNhl+#Jk_U;WVTXdmnR^Zxnss*7AR z>?7v_0^2(IH}DH^uKmr~;|$~;I}%;bm7V22Kd&=nn%Fal)vow*ycf@T0rs|Uh@JNy zcm= zv2@M-9kYlfIKvY`-kdX)kt^o@HILW{+uri4Uy$3qj?b-J;oVeSF72%9kGA)SeHLf# zn4A_Hs-%Yg#%E$8eE_XhT|)^qtjG;C3MZKQTpr{12K ztJawL;ur%;=8Bc?4uQ|Sw&NW=MSR4a?{1Bl_fU7e^GE-VZ%y^tXRwcc#n5sJd#}I` z;tgGr%OB_zvG;%<%sqj+o6VDR^gG@e_px4yBoEwq#I|vjnS6;=f7tTsevNa6ibJ!9 z-&52HoK+fo0QYh~#W{*xZS#Ils=n;j@F(#E0q)<_2Dt<;@Jyzddp1Jx_+jl(4 z1yHx#_kiyB_VV23?(n^DIki7_<Q11R;JV$HHoCPDm?+_TXmR!=$hK43gRBd)goqK&q9ogKcu za?~C9mTOO+weOL>a8C7!RqJ!<*uDc*i|-vxxW}DWoSEet2TGmMo#W#i#@yL`&h*LL zcYs|v(#8^w=kj-ez_#ZTdlE~aZ{Ma(uNxe5Bk`FL*nzFDc2-Acw5qZ0$jg~Pv99If z8h<|nw&E=JZ@b^Xc1Qc!BLu#7)vDoGQ#-KRiF_xnjMzJ2S8s#F>{E9{ojWD)v*|l% z{UOdc6Ibl{#J-as*uDe$1OMpfJa@6TeT*wJ__EQ}2@q}P+Q-kb{Hh^d?*_3hh}+Zs z?w?!r@?Lx{?l{ONHP*^~A?{vt#~L!Qo!4^JT*x2yf00jIoDMry%^6+^dL|b70p~bl z#Uzlkw)mS{JZ#Ts-W{|njvc@qN@HiXlJ8#Ey$AloUa`*Rzf&=ZIosI4mUG;Vy9IIm zflg)#WBy%pV4uMk;EW`GzqXPWV(vePS>Ld`z_-p=y&KqZU45KiF??gS+xKhdMqQHs z3g3Bh>s`jKdxv+-xEP~<#i?@-Yt3tW2kvQq`(Chxocrfh*YKSo$DJEm%m8w*-SxS7 z`sRw;*54!ePU}D5_?(2hHDU>zmHG+Lfj(c?XW@)*|KoYifJR;|IvCqFeu1m@zq#V_{apbP4q zp>H4i${*JiOY+wGUNvm<|4N%!JxgMDY*GEFl{Y@GfBqBbfp3oo7_t0@e#7r*XWZe5 z68wf8=xHoA@cnNO>f7rRxu1b`gP8FSe#L)Y*K>6jx#|hzdEN(5PUOzmBiD{{Qkz)p zxHZn0=tTP*-g^?WpZTD!gFk&;d)T{pXz$1OtfA#Q$9}?EXUebl3O$H>XAN!7Gn_Xd z_X3{4`w#N|_Ywviw7c3T zjG4!_ryS8^?RMDipTuLs{Ts6rp3kv?$CVz$>{az}y#sf%x93%E*cV_0&OL*PuTEm_UpvbIe)f!+3;g4Am&B}z>-t$}&Kcjqm)o%)e-iow4b zhig9Z`{xgy@d56eqqhGUPVE5y_&i!Wi96T+6D{oTSvvWK?Rnd_{K&ahYgZX>Rq*mrb4fblTz9-Jp}lLI|h zt#j?6O>OQyjPDEH_X>Uk-@s6}5g#jdC+|4n%_qD&{0-dW3jEo3SE2T^4}Pb{+8cgi z8*uOc`3L^SkoTa8*u9cE5WpQWR?a@od;{kl8fTLC^UVB(hik;{KilVh z?~AXt#hPbtdpT=>2RQHjx}zgj+tCc5e`ct)-upa2fWCeA_q{5QP3nZbli1ul`0T&) z1M8{pM69{)7oY+8)9XV06HNGFKKLK-1Fk+RbHp8()25ak+}9nwix}`~oRgq|>&JS- ziFMWu$T`bB)L3J^zIl0jJ+{ue|L2eOJ@`LZbJhnn?jWppCqF0qzd#4-+8xBI>%_uX zJGIQ~*xayB$Ybs0J4CjsY4FbV-fzI(#@3p%m)aqZyq>*Q zpT2W{eV>-d@tc+!-^pCZ7oJY+@1%%TTYTcP=YXp@#`JaH33own?p(2lZ!iY~4S9*$16{oYBZXfHT~sVFw8O0kjh|a6z9~IgfKZ)|LiquHF*= z^t8r(y3wxd4g=((kTEMfrySRh;Aou)Oy&JJQXKOq5zx|6@_~wFmfC&=R$@RI^ zJ|FA%nms-0-7RyyO(S*}`3Jo4enQQFf8JOB(!o#oml|xxUO}qE4mju9*SZ1xnX0#j z{|N$ktc_f~e3v?Mv3l0$;Rwh ze3Mvn@anjmd3)NYVI!~{=bqTy%+*g@IqEu^PJB+r;+i&hp4z|{rNw&mBlOl8@t^wI zhtz~mP{~t6;#UoR;+t157secCR4!MpNWVX5)N3bpyE~;{>*i~FCW?GyZAfVpTSJ*>4Vjua0kYxw|!o~zh};XC+CdHW4&8a>s^NNuQ~iH zyvO~%;rl!PH{j>mbM>@F-5|DyyYr`y{c)xFz8DhOb>~-g#=}-OqZpKeeFl5xYNjl2iW% z+{60BKG53FQ#RJSj+9w5TdrM3PzS7YZ#BM=^Q@Koybp05Y~`3&`Pe(N$LTTA#OBpH z$IqmA;`-d-syEhr2WsTBq3?{%Tid`DeqiOjzwaW>E)D)y-(J?KpCCau_HL&?oIR^A zT;Qv-f71uG7h#K)3#|1M>-;CM-gD+#r{2%#=m&9!H`WjQj<-^KXYjt{<*uK zk8#U2*WVFOX4JvI0(VpIZgudhCRRV^CcYYHm>WR9V{f}w&UbixwEA8+N8cIeWPIHv z@Ly=ods#nn#rp2#UxuR?WGKb)_)}nlVr_fSkSLf>?gyH!YBB zxp%-H_{JTWi#@mAhS<9lPyWVkYJC&As!@;cF;<`2SaaP~47dQ^>qYzoSCHr$7~q{) zt@UcPVt{sEbM_Tzg#b zPd?BMP;<$5`~>p8i|2-&v2yC#_T`;%a{hN4{{C3A1{#qY4|D-O>;L&R)4kNXf5-nu z4}T7{#WAy!58$zf_kWNNtTQGU!2itZJG}o{qnd*In%I$@=Tk8^+or(EbIit9{@-Bi0@l z`GmmA+1FkPgAjQyNG4{;%~toP9;+ z`mS|yN2|W(k`M4CmeU%a^Ek2jkNvK=J-%k>d*=Y|THhhK^K<==KbGFU&cC9wCV@M* z+;g`-;q5Wt)t@@w;svw~jd_ono@@SZ_sydjvEH|1CwVm!U*5Y|uU|d!ulB{(p-+83 zV(lrfHh{llgV;M;JMnY#x&0gb!0+&K0Vck&XbFc#>l-_J1Tnt)(h&O&sReV#_8LIm z&&%5ST-)}^*>jMO4fXbN=Z^23H}Lq_^LRu1xRZU~vED&leMWFE{40>pWH#1gPiHk^ z|9K;F?dQDeNzR!Qoc?DQKQH}8POGkF!lyo`^4r!ubI(KL4i_+E>&exfvG%l=XXN7T z89=@J$|t^i&h2X}2S31>+3>Bc?;HAhpZ4{;+II*v4UO6kj34sod#+jbuRFo(dpf>( z`Hc0wa;G7O^^WEHrf7Y)2b}kFjyY%5NqjJ4y^p-U8qbK8KYZ=>PZIDm*IRf;efxCy z7qgCUF5#@VM*qz{*v&WaF77UG>>07@ot3$2&3_*Fk*j9l+sk`Cw|>U%&Fqd}Z%@v5 zai7`^YmW}p2dH;$t50#er@9GR?0T%nncn5!^i^XI-?_B#_B*h)p?Sb3zNmR{=B%rO z?~Zbj_*nP1zvuLGw*AD*TCDGApAox1eXKXuPVl0x=hk6;W8(n!&to4br1XezG|x{@j%l+2i~W}de0zk#OKXAytzya&>nr& zkn{8`Q})8#jMXEw;6~4?%k`b{^O^AKgnJDwdp7OnoL9ZE?qx1Tu6^rsz*jGCpCI;( zSiG_MhAp2NrPaa-k6h)q4Xj)pEvWaSu5`?raK682NuHe6y!L#K)yMrgBNq}Ye|9bn z`Fe|!BX1wyFR&9Bf6fVbV2$VbFq)7W^CJI&3BGk z<@S^v&nmDe%v|YY$R`oIOLt>Qj3iZCfYF zyQ9_~Gl)4gBXAO6edmGS@wJJaz;oVMt$AU7fB^FKd1DiF;HjN=Up4A0&s&;v=7hIL zY45x(n8bd5o%{p5gRaIpV6P+|SnJFOzPzXUq4VA3{A@?yZ)=^rI!}-NI}Y16oHO;Q ziI1gyVgCfyTbtOib4SzRJm#M}=bhAgZulL(fgW227YSz$IM;hU@jcG9*A=wf_r}`8 zx(T$-JnxwJ1I}K)rPlh3J$^YomhMd++!9VbFLL$wIXHX!{$Zaz*In%O`5oA?+@!{SFZ1Zdq=$F$&I#1G zd&3T_vohD`VQrTut`lp|)2IH7jov#>@CF{B=Yn&b=aI9Yv*fPocAQvqLe72l>%I0p z?Rxg;A75{V1wIA(@BYra-Pa832067a zpdV^Mk5a!7)LCmBxoV-S)}5?*0_$fijlNvR2G;+a z(|>MD@(pf+vavMke1Ge`zsEY^F>gQjNc@iPz1&IPUI9idU;YNv$op2^ygs~p)c2$D zyC2OH)Lbo^3H)w2uNh<2%U3Nssz!l$=R3hYKyiDXIeP}Uptn~Ga4h;sJb=1Hb3WHMeD_as0Xmpyv=wi^ zalGF-)hE}iC;jcwiAxj92664P<3a4-GqkxmclWLQoUdTQyIaEwd&{f$^KnLy@5Bw? zUIBmG8aaDLtR2kP4C037=>7X#kn=6&6W=-hjJb<76VCYj+*#YflY9VcovYq?xBC~| zPn|v0T|ftNlbpT0r#1F>Uc&uJevtP>@)zK)&M~j{9r$Je*96CUX9h6$4nK0g9oEmq zdiw=%2d!`E&zt={+Bb1u_qXrJ75lxnPbc=wT>GEzMJIm+)>iKi#>K(|&YAXm!Jm76 z$8Vd@#3#1@4ZS^jtQvjqZXe;@AHZ7S9qe7Z%kTJxow>A6{L6javEIwO?EAQTLzh5b z>$y^64|CQBxi?@RKNods-Zv1|!`Zuzxmfq; z`0_98fb*#LSZl8fc4F&p?9tHn{OS?J&J$1I4!QO1zV2(!0o?ro?Ag)W@SphKz#w*x z=LPS)mRS5S4nW-S)mirdzD>%*SvLi-`giiqdI5F%*N>Gi4Kw83v%?STh~2BXTkb_>*#79_NX^!NB4N{oV(fE z8hg60-<=mY^^<%v2khmXz;}lJ745)p2f2FL&ogt?+DHF_zT*dI#K-sduNvzTII9K7 z)nC1{5?k+&@7s8KE`P+T4|p~9lC!3cJidB0fuHztTHmF5lB=Iv_yHQZsTCgb4eh!6 zOnMO8$5Y=oHM#m$55hrgpN{o&sD8%mT08Uk1UWU2tk*y@50TJMN~_0DVZT8}=*9qf}q0My^^yN29^tGB^g zn?U`DR(k!8ug1CM%K`7v$hrT->N_{FkqgIHAMo~@0$)!3*uFVGM|IB9h8%v=dhjo3 zGFDD&&RX^E8E6K+r{lY4-4S2C{>-&QVAVg-*u(cK4ezXPPqk@I4Y7Ltj&F_oXg%^R zzxu10##sH1U;EJG0TRxW$zNji$E)v>^BYia{lysHquzebuAeJ>xe3$^VtF;**V>1* zz_|fGk{p=$fo&l1)mIK*otlnKtiKb?%U|(5?x{awY1Nr`_v(YbdcgZW(;V<#+Li#n zpb=uVm?Gh%(0-NSyyB7mPy#kpd+&$lM+XX<_) zcO1DsL+?6cy}AG0k%`8+;%|KWG<^FXUn`#G&2eV%7c>K_-$er6RgM18GCKnft@n5V z>(mFxB<74LQ$M>`@{`ye4fGH4pv9^w4!;ABJtpw8%YYEMdOA1oooU|@#8~%mAGy=V z`U~*ws>V9O@gCmoO0MUcW&LBv^?iarng&0JXY9LBTi+G9fJ;!Hix#bK{{#5vp(jw| zcj7ykU=Rz>H)G(w?Y?$(rygsTT=j-?4`JS(<{G}*vya>){tfG$yhHUsXN|cAo*&CT z(75XxxXN*!a5s>N6l+cO$2wo$*gX>Ng+`rk?LGej6L{YQ&Z}Lh32M4A{Pn)-Ye!^n z#A@yPBJWQ6zT<=+_{O5ajaYq*AH?>RzXA7=pTIfp?VJdFIPsM89zb1-)yE#)m}AcY z&JLMtXF2;FG)}-d`?GfEN3I?}n0Ip~I%oLL7}luQ208Dn-nd~WFutI5mi~xV^@)9< z?{LCccp}SG4{K`A*q<>qzE2~rqt8)Obx8nHi+04CukZ15p7}M?&)mJg0sB5d50xv9 zz#&(6|6P_mw(b+FkNI~Hz+U=KZ~?U^I(Hia|9W5FyTosB)~Qc?^}d(64%dLQeqi0v z{i{zdePHWzL({DR)~nHP`1bN%58$kpi<|H-Ib!wpcHT6Gx9&mgJ)Ga+-?3MHa`oNA z){f-VSC03I?K@wcJ59Jc`Q_j2N3MEoot$^BTD10hLleM0@(oU2op^n$UEi>0pH9Af ztoQR?auZnV839L)--7|v)O>7$=bb!z+4rFaPVEZ>s2$xE-dqP)_|F;mbr*P#`3qcP zz}cs65bM`> z9@}Ew%|6zSV{_fhyS43G8?gYp-sHY!C%J>BhS=jA`6kfX-~7?26Y>)r?SN~{p4jrS z?$WUlZS}*i7@J@M@9Jmi8`@KCz?&byeHvEO=NQ{J1|g^w?$ZQEegb!TfZNu1C(lU! z0)Hj$fiH5+v-b4C``p2r5y*g6z1pKUU!RjZb9#cn*)z?R7OKHl4|3jJ-30E`av@^v zSv7FpKe5)C6B8I~dw!w5lWpy4{f)lXsu{n)Sl;@EJ>Sc~AE4fdT0hGc;J$3&k6)nP zS~<_Tt8sm9aG86yoNq)d=Q(rr?OnZ+z`2~t2M+TS>;3AE?vp!D&yHX9Xtd`&C;Tv0 z4<1^My>9p&*w30sVte5~H-GjWtLOA?p%F%7j;$hvEINLo^#KL)o*Ii zyaPY8s$&=DRt+{m3v%#fiOsuLP-A^3R;MPgLC$^Ec&cyUfB^2HzTS)ahCc(ladHE` z3;H_e39jnVRUdd`{Xplb&yN1r`@)Eo)7rzC_HE$T_cBgkUHP$g8gjd)vF43>@1u2&wE@gEFhKpgaq`a5_p^EI*uDuL;P$chbSGo& z6Yc^$_G+2_I8J&SIxqnKbIw%B-j46cK(6? z4y>!U(SkZy_ZUe8ipN$=Pr2NPAhyr%Yo>QG*FkK1-P|Dd9Nsr+@ztw#CRW#hR?KKM z)4Vka=S~TZ4Y&sC{on`iM66!cx`+S0cr|nLYCK2pyn`O=&u-?d*Lu$feh0VB`Q}NT z@8IV(wyl}??qyGFTkIXMuaCv~8J%8%F3Ab!+0)*SU;5a~cL?)Za{hCg^Nbt(1fien z&;IssPw)OGym`^#1M7UXr^a2caQ3XeTW5dwwU7O3FG|n5_rMp_JJX&Qb=0(2zWtt{ z`oc{#0~=Ut>YcV#;~w2OTJf>W; z&~`ZEZ(xvfu6oII#A<-q#)j|J+B-1hosca^%&!o9DBhUOiV&`4MYh z>pN(`dToHu_vbApaiiXS?dx8t2ACVbSa^rpmCx&mulDqE&#oTMpBs;{c&Woa&;-^y zT>n`AeDh9jz}Y9^FKD#teT(^vFW+{owbyg&Lmqt{eL+t-t>5_>>mGi_-#^yp;%?4- zV0{Ps=v#C2`pz+x;8W96Ca`P@l9zG~dT9+B(qJNCr(OL7BDEwOMv z&&>7ro^B5ifn`VQ^vwro=Gmj;n{R;y9`o*+v25}9+5q0G8Sna$_&_^Bg7{ee*{g8^ z-r93NbpzPPzX$#e5_mp$B7Oq%&Q?3L+qY-p+smB4Q|x)gnyVe%6R0t#Z(dyPD|S!k zh+q3R^grSL`@fd0z1r>_IQzU){}a4{7Hg+QUEhAsHOKi0_YeC05AdfNIOn*-3@pbT z{XV&4fDwBy?}9HP*Z1*8{srt2@a|Z965GdrSKxQLG&?_b?r61fZ)*>79{W4DW@+hF zC$|kae02@jBV$>ZI@LFDneUz>*N&y9pYQJs`3^kRoP3MD1DyE_J$#RcmRtvF2XKaY z@B7=fFKh=_kZ7#ax9)n!vV&IGFVKhv&V7wDR%~4-K6!J-+6iv-7S@UpYqyiXkW=Rl zkNcW?!>VgQ-5}@p*j=^n;2W@?dEZChUN?C--_hOVtk>GlSq-Rfa8D4i_4WakR|7n9 zBg(Cw_Kn=UcKh~e8$*s6wHx*7Cvax*`ukMxAjq}L!(G7}NI3QGIkw-;-vRLre`3{X z1GvL$^Va^v?{G(FZAWMBO3qm#(RhNq?`Q1(7jwDx@v~9u_d^UcgPh;5C-`eyzjN*u zlZF?PbR|^7edyC%!#umK?FY>*VUA?geJ7&)FJ# zh%2z>H{cxap40`tx%!;o25}>n(~tYos59;&S1)rfbl%y1&S|lBDJ^e%atR;gZ`7Ex z);aEEzxqtD4PLF9BsS;ov%uQldET*^qrZK$7cDvOXHWY$OWR_-{Yc)7*#6c`?D0Fk zJ>NgbVcpgJ#1*LZz0F;)>YU|gs{aHXI3vIW_SToHUFCqgj#z!_-qbxngOl$Y#?Goc z}{+r z^wHT%U;H}v@b*cdHn7&(PhMo~`x)bV+~JDeoO;j5uXk(lpT4nL&xP7LvHQ_y;M+${ zV6SMj_SEIA;X^_?`A_ygOq&eb3IpFH0?-=W@``Y=;?c@iTYRKu=T;AH6fU{@)d|@L2{J^)yT!OR56}{hM`&J+C zg5P7=#8)$j-JxpH40t(pr`EW3z*a8T%p|stwzTx|J9&72rh~k-`tH$)?IGt}PsI9M zCf2;Y0&OQ&V^4Df-s2supNSRt9Zd_60N>gxynevh#~tc-332_5qZ#;#y^@>q`0CUT zeEWU+#GTmr<^sQIff}uGX|%j&=IY&vJu{Hx`ffa#Gl=aM_=#_yhdvs0a%#Lw-5c(O zFK>Q=z&|@#=iCOTrW4BrICr*J>AVNL@D1z4>NT+g`1bbpeX8HdXR8^_7`##Me@|;- z)diT@>eMWJxdCdw-Ot$X^b?%pEN42~&!WZZIk5q%K9>z_!VPfr#?BnTbE9|9Wz|if zJ?~Vr(0BrS`uLfgS)cEdC9A(Urvv+4sdE?g#*JF@=iOe!`tr`Jllwlgw2zOq<2S5* zM3A#4z`(cGyVYLm@SW?-sxij*-MmND6E~57ld*eYyMC@d)pN%QSDy*ix~UDoaZmsL z>buT>NK(Mvs)o7tm3NOi!CwU@6F1gB^Z42Ob}ul%BxnDD^{7kG?*^;ZJ)M664NQ3N zTKC6yj|Z`J_LDDdTlaQeomh3XlA-ChEU@(YF#g9MU+?w?_aJYreX0&k!}s$nt)R&C<@JHlAM(ZhSVqdf}5gPff6|!FBmRrLKU3}z#Llvx_2;Os zJ#*RAqV;Y;?n2Gen4_=0kn|yL#P;skEAbP!ld(Otp$`V}8~Mcdo{@mq`R*w<;2OTi zdw52y_q3lmdH=ra9AoR=?n|e}9%`I**}R|Af%9uNy1D&^Z>@K~f)0QEK!^RcPJP=s z^}fju;NNS{E}8qg4riaoed*lOUhY&rIp?`c&DU~9thvrtXFv02ANL!;KJwnR{2=DM zh~-aj>j#KvwR>4>IS;hr37i%1H+*ZHcm8~1jWf>7PHau>f_4Jy?D0g`h=0R+Uwe#L z{m6XwPdMkD-RxoPd03kZ?xC+<8(94T+(*BGOy0tJ{O{Dr4>*sy5$p5zzRehPIA_jW z^=;>Z*kes;a^>yQ@!!!Uxr=(@`kaW*{^sO=+xcluPQZ;^eSM!e{+%4>Mc(=wHh}u; zeeIID-bT)xwT+zo4co9CxYvk<(vZ?V;H~XwJ9#}10DZOULTz^ z>%e&jzk@SZ?Ke3zSFG>l?#@y-umfEJeP@|##O~!Do}TMIgW8K+th#`=CWyWNpWp{@ zmUCVL-<}VA_44&ygikQcVck*v4c~hCv!i=o;F9_YKe0Esz_-@#mo-n|XI#Co_HR3` z`RLz_xx28|eLU{waqeI6KfnvL1aKW(h|iu~{!R+`#@;>AsJG_;?xZf@%ZKydUk3cm z9Q?nr=Ir6Iwh4}T|NK$AV%1eGe*J9a>g`Pb|3)U({nRA2_OW*2pEnt32C-UEcS1Mu z8+HGnkDt4<9zeZs-LYySR?g3)Z@$4bbHLs&__5=_e|``Pa?X3g`FZGj?D>YDpaW|> z=1YTB|H6tZzV~{^irOnFh_#z@u%VZiw$y3Q1=8q@zHv`pE)&=t47~GDe(WlKd{sV)_WFDF679owf8~w zB(&E1`9=?L_zo{#_htP&g=cQ<3)^Dlwen}Dz>i!telGsIFW<}lzRw^xsBu@(h{cQ* z`C%&J9`{>*6+%xd4A6Q}RPUaKe zorU{M;tMvi)n{hEjxQ%XawGS513Q8G*X9FVAy!{|dbHj}ZM`Y_j$fY>*84WBI`{ME z&%NCJ0Xn>O9(R9{8?pF?-^o=>u3~b|vS$kn`RYTydcpZQ+DE`_pblH zT{G8A``EK~#`n$*jLlu~-Pc{5I|4nB8*=#Wu72X@j`agP@2hv_`k5PWr6JZAS2XVJ zT>>=V+jVkk&-cw9<|o#=ZVa4nO~4ty;rrcb`1Wq#Nqp{jc=d_KxN^AmvELPIo%2WH zbM!l~PQC9Nxq8b9d(|%38B5c!_Nv~A6*bRqGCAxyL*3?V9;mGg1rMFPORV18&N5?CJ5lAnI)sLd5#H@J-DftvTZts5e7b zb1?Nb@R0;}V&Kp7ATEyFnQcuF+oRqAjn+D~)-<5+JpH;iR^R##%I121qSYf-ow3Kd z2+Dxp75$lQ_4GaQi-XrTa6{)Vey=)lf~w8ck8J7Oc?51Q&Dr0$4%YcY8^x4XF5%RL zJgCo9OU^if-&OsClY7F8nOl$K`mx?hYj69R_ZSz4_1Gt(-1lQ^WAU2&*{vS zD;;rlx8^tETeYtJtB(f5?i zefL=X&D$rnx&9u~Ki1zz-mt#`f0qvP_chPEdB8H}`LR_j&`dHO`v&*2}r`cQ7{G`yF(0xpmqG$dNuCX@$F&lIfql@&IRgQtT%P%2?lX)kp0xk*U2v~V(+Jgw@&V|cQ`fj za<%`SS-ShWQt9#4o3Eb4bw@3GH~5Z!qn~&6&(ZbI0N5dRUadVRaRT|4t5@Kk z`yczD*6+Nr`o1;pW^7IEBQMZh(GUIKTlF?Sf!bd`mox_QvFFH5`Q4}OeDmOolMno{ zIeF1RfCLi^;BmfhBWG`S3j7rK>R#@EFQon06#$agdIOY4+3DVXXNVdY`<&r>MryVKY_n9tg)B+*L}Sm?COUN z_y(-kx1Tj4fIHX0KX2FZ6F5iz+{-)r^PtUKzqkE+xjpMG(0ZIx-;cmDcVO)uSo_*j zzwWwi?T0^e1AhRI*1UCdBe2eOmdAPSD{qek?mCD&egpaq80WG0#kW=0srdo^vQ{1N zzR&x5-+GHc?18U)<=5vMvHI5xcx`~1#~oS#Vr%P8?Cq&K@3@hLVPUyAnxD-JZI*W{0-D@^suH2`+9Qs(U0|6*t>RuIy;Oc0M$2e zK|^=$803uIVUQDba`l*j1r3~DPqms9px1!EoSw{dp1Py)?Y;r6wP6ljS$fn=xCuP+o{_u9vG%5ieG~5FCw|+p_xSa6 z_|LbGRqH+M5hq6D4!+$B_)e9Bv#-Si9exUI?n%LV(Jmu6dr)J$- z&1+j{t#!r$tTTQ`7x;1s#OIwydT09Q1Lu^+WGGm@k8fTy`e^Ob zfqi@feRrG0a{3K)a)Y?)eqV2L?kZpJXZdch#3?M%|rQ{lFS0zWZH4<*{?;tg-)tU8uWSL~9?o3s66Ud*BDYKi3|)0QCmUv!|MKicccA;$yu<$NHH+fxd7D_0DQw0Q=1# z2Bm{DuP(6eu1;TD_s|bC*m_gwAWkslfG2XzzcSZ1J$L)O3p2dC{ORKk_R)Iu?PK5E zw!xXV)|_#}p1wrubAvmYh_!=#JXN>nj(|u>z)^d+o@@3ST5FBf`nLW)RhPuY<9~zm zvwM7OJ)C-bPoQpqj{ihsocP|qfNyQWdwiSnWABU|;BXD!JJ#odnYm)Eni*~7w45FQ zy?q-{*ML3k;m=KI$;sQ_cP*>Wo+Am#2i88u6R1g`ZMn4S)qQ?8^w#%|&ZyqVo!#T> z_q;U;COAEUyuDNF$$dZisFn9;S?fLYJARl01HWpv@4Iz;^)Iyk8QwkQ)mjrk|I~JT zVV?m)pX29lt@=0WeV2(}daSW;He%h|yIX&(yx*sH;QM!S)>nVDiEpol->~ZIXyK}N zuDXWxjN~uE6F?m{5+CSa@5@j834Al(;s$1}w=j3~eqPp_^K&_O9)S-hir3$#Cr<1D zKY`plK0TWp{my4ydUmkyWA7CB>V*CT+5t|TpPhbY`_!rF#7}d~8u;do18eSz->`*W zddO#>yOZPSH&Au-a@Lca`3~>Wi z{^#J<*11!HKY%l9*Iat}pssT0{5hjn&hou$rm zp~g41rrsJ`^OG1P=&{hS_Rw}MbrWA(CzoD-;8*X>UAM-1`&%daz117L&l9A1-tvmC zZlJk+pxJ!K4>)=A*7WTgm$t1Qn9uK8Zp{n5H8g^M9>~=vf8nN}_VIDg`&(-s2?JqBCAMCkr^<+OapEJl?YadT0 zz3Fegbad-I{`1aDmRUY47#({OVnj5{7xW0pJ{al)t>!7LAf-73B z=(+xW=1#^zeC}L%IWWKr&3BM;^mqf#{Tt5y-p!o0Y7(BfdfvG|y!xRA`>DroxiGQz z&~~8CS^5W$kJZzA2fRJ6a3`;y>ah*$F24PFOZ%CBVF%Vd&K!9;YuxjVoc(XufNyYG z^$qLiVEw~5R?QW^H1PJ2Fa6e}9RA2vTl>JN?Qqs7R(@jj+sA@=d-Yh_%IlN&et!M| z64Gvn+ofy2H)MR zOHiN1$I=)(TQoFp@cLcsnt`=G;AUb$;>&q|@9Vq=@SXkhrCRSV)L!8_*7^&aJ2c`4 zzBA3$=LxU>f;)e|X!(^JC*V@L2on zC$PVM&B4}eYMs@I&)$Kr&QtT4+hXXV)4zbpr&bzR{|eIpdR? zn;(bYvu4r)sO#in`$7IYFm|Rt*T3GW!He@VlKAG;M=W@!Nj%H}zn}hhv(=b8c*B1h z2gtQgVkfaZ^@Du%-MoI9=Z>9R-~J$PjlIhnn**_Z>qHZtT=9zAom6)IDKySRpYD#=Z$=?L0@;%5BzuP>{+v+&2`{WTW^E)%v^JvJ!)t+`!sf==;0Po!#A|u7Sw)#`g7HYftu@n&`0oI&WeV!OU@Gslyt9OR}0KUUx*SKGhSEDU_tnaq(@&lO5So6;Jq{B6G z-lhJ#`B?jRptfmq#RD6l-GL|8K59dseLD5(&faS4^Dw_PS7(6#0-ma))*ci1M!to< z_p{f)m$RR~wqvbLFL?J*FRz{1>O9E30pp1Oymx8u=;5tT ztoQ3ct>^H8&U{Na`WNWyQD30PIVnf}9oSlQR(2{ z%;oB5ue0}c*90}+LFKT{Z?QfTwH|Z!@VK-0_s>PA*G->&Jmp8ZwcqLI9?y^U=97Hw zoa>F9XJ6+8_JPiMpPwb%AU3CUZsPm-*x$En_r1SQ$ z)_WIt`-{>+oxa}V0i0{kxUYKq-0%~wYBo>l=g*aWYbR#-yVLKN{hZ_P0)6X4B)G2` z^}fUt-Wh(^LM@zs26+NCel87-x=HSeMq6)&Z6cO^+qpq(PjLnIvBsU$>j#h@Sh)+j z7kT@?1ATeV#2?@ZX42Q(oix6&uwE`e2mYBNW6d|`EOYL5#kPd|ebse+@`*i=i&*<< zzfohK%g53*4}J@s)vFWF#FyvvC{tW{cm4`Da?O+TP6IBq@ByYVaKGQ+e*p7p%{S~j z@bi{yaK4GT9>n;GZ(rlvef9E=^$z6Jya8*kTHs85-^aQKekX6vc;LH}c7g=<_FZR? zqkqFs{R8y!_LOhA-w4|RU?;Zp=*ovPzI?R${r3&q;Rc*%CS{;bY)gOtIV{lr1ntOq%D_Nf~78))6jyg7OMmJZ$;a{=n@YrS>$YFK~wdXETVPLHp@|JrZls*z7%jr&(UwiB!I z=qIS}DLuWbb?)2Xs~@rR)O^-XHmI!{DYX1zcBZTCI9O!Dg8t7@>`x$eDtm>2HzGKV!ETA=m2 z`2^lguIGYtN(XH&<;fR^Rp-5*Z32%M_yg!e_wB`uk1)d`5B!uJ_fqK8rtA-;hUduHjdYTz$Q#am^&>QS0%Zm5XIN zw&j9#&aQs3@%3p$d*mME)OZ)`8~)kRI!`2Ate?fyXUBmR zL0#=l-aC0_EdNRFVP9gk=PiAk0n|0nlbwLpy&HC5<-GrhHRB0q>>0UcIJZC^mW{p+*OR(r_bHio$m$~Js! z^%tzidUL+B{WHJb#C**SoLqbTjqV*ZVC@wIdHdCV=+rvj*$?={PS8Nt&lL~+1on{^ z_KjS<9`O3=)z-&S&J`>&(S*AYg zGcRJzbieY^bP)8ISa%g~z#Xh}&gok{(W$j=m_v63NvyUG@r&A#M6q;d2f6wVz&CvP zI`|>11OA!Desbz3_VKa5gZSn;7~lzS-nS?{?_iC+jZ1^y&=%nQS@f>I#xsa>;YQwj z`3~~F!4qzfzsX_sgIsRgICAO}obQFTL5(wmd;?n1WA!#?-4!c}=MJ6`Xm_prOyUm` z_)mDPdF>F1ixuAC3AFzO-oe|pYTeB}t?8iN48HzMiS7k_S8Fc79uJ`2H?nul#&@2* z8nzC(@4&v!u_k~soSi_e{tcZ!qnWD@nbKM7+&9p7zQ^k49*z6}o*89oXLB=`?i;@K z?G9qSMeLk4RZq?xjsHgLz4W_07{r4(fqeq&4B?#vdzA;7U%R+-eg5#y>D2jd_OZ6X zy|lzHcsbwU3Vb`~1RD8^xUc>l?nyk!cOdT`&U)aVdNns7m*|}5o}xY*a?a3ia7q4& z?ef^O%OJMiw+wJTPiM;8!+QHq;65*OE!O;hTZ7g+T*-BKzhlPs8`QiX3vVxH+~DjT zx zYq#A0t_pqN+gEGf4%B`J#y4z|yBc$>ANsj+J@%e=ICy6SzPknJ=&Cn%fCTnbYiO-e=WO@-KYy&ZkxMwgqYq$@%*Dxfe0v>#;yXi~y{#E=>b~9g zd*fSU@2lGF`?=S50nQyFTD|p)xZ!tTzf58+v6^}x;t6lh6Pp`w{vNu4^SzR{*4d@U zcZZ3!uiqc_0jH+j_b$ZeO&h*Fv>x{!v1+V+f-aX^M^3$;{|z{&lk0F-;vd96tM|}Z{pimO$y?}cYHbX-h1HNuWkFlcNcrLT($NKP)GX*F&cNg0Co1V zM+Xml`%HYf1b&X0D?T;8_piOpJ6Bw&w{L0D*~fn${l5J{{6u4K(Gv-_k=OTqt*y@k ze&GKNJe3bD7#q%N_#W@?{2Oo=HEP@D1AinjN88nHZN}1C7v$bPh_Uv#;d_SKSifg7 zS6*9rasis)_{J?)ufTV#^t&Lo^d~y>E@J9Y8Fz@^S&p%qtZ{!}ExA$fH zS1st|<*oC3r5zx_+}4`e-8%2w@Q+si z-CV3apX6&tK6}rMUAJw|rFCA@-w0#+{Q7&^+Ce^nbJaYZMSLY5`0|a|`zClAgMf2i zdGpqk7IS`nd#rnZze9qVt4^!l{^irh*y9eqxxGfN&+t3G-<8wXnf6ZdY8uGEiu-UlZ6^&M2dn2CNYA5?7z8drHR2Dzr?Q?#f)VWXl7aHsp zsP&9oJ4dWLs;!>b(!dXA0OzaoI8Xiw9zbp7u#MPLah6;AXRPnZ<-YDy-&OJx)SGVW z-Q+y>ZNM3^;f@0Y@c%!<8@zkeiPif`e0t`cJA(Qdg17$-e7>1)$7dhZ%Kb^s&w2o7 zyKl#~1aQ{+`3ZLr3F|zJMhtYZ^X}T=>S)#F>SImfd*Z%+wtb(YuYS4baF>j|6M8l0z3d^np#C@DT^=9n z&J*5!11oOie2**e-r`Rn|8FpWy{vP__*maZd(>|5o=pA{YbR&MsQ1G* z^v3mbknnO^b_uM09qb79ML)5FxP!!>+qb9p^-dRHFW=vze`>8aZ(V?QQCx4-~@^B;9E}yfB->fBP|2p^{YwS}zJv@V-T{Dp77`mWjoBsn>ME-F4e zaUEjs79y8*mix$QYX&9|4t+BIft5fpl7s- z@AvQ%@iQFbl|##YfD6zbN6x<59oXweu4dxbJrFaeT(0~eU;F3kZLf)~l*S*q=KCz} z_|=bC#5&u9xLnb(?dR_HN@D8zUOjC0^In{PCTE`$_5e+r*hI?vw>2w5_*yVjENZ-8^%h*ZYy<`R_n^pa(dC%Mml~Joni9+jqv= zHL;y10^6O{HIrDQPQH6u@2tub3-bQns#*9A`%K>2Gq$`AF4&2df5)E2*(vd#@a1&0 zoRdEB0gO$wJ%#b&QoGM$OwOIWpBMXWZW3>~xb9%>AZG0i-o#e|d*_UKuN}y-pE`&u z*4jS;8<@KQ&tt7KJ)<0?!GEysO&uID`PjZJVP9>krY z)XqQdnu}MQSau_34}0dJwLQCiIKOZqRq97d5c= z-pgL@Fvweb!?th4>RImrwmaF+ovqzHa6_0dC(yq_VD=Mx0M=PUY` z#k1~&Z>>F@X}$O34#xbheE~{w0u6euzKNco0sAz#wL5kY>uTV+8hQe2YCq=vO8hr4 zh+V8DKfv?4?r_K4(;3#;SKH5#`#E#s51?eiqqpLF^cC#)!#?snUq{QmV>h(EJD%al z(|4chg`L{W*RL}Yv(H2i&;%lSb;Tss=y?<+B2J4JtioAQa<%e?sg>)LA^@#;l>0J-WE*b^vf{WG<` zD@srBGbMpT%)C6Gi}9B0J@ov$XYq)8#sRm1IJEr&?nEcBmmK>w{EoeK^~d#|&XSmO@6?!gw=G+P^3)BU{mq@w+SZ$wyK6m%=m)QIaI9&>JjcMl$>9tQ z+x~ujPoOyX#m6;gOrA5Gm(~(HzY+(&-+rDv_aJWnK)Z9*lB+sm&b61E%IU`H@fF?x z_Im+OU=RB`=K?0&6WU$1BiCN`OXAiAI0Mh8ZbxjwiNHUk=X%c`*R`L1-ANn-lrx>> zj@CQVoxFSZ@=lVtaIR?B=0;$n19W1Z-TuWTr?#K*-S_#r-Wkuhceh`MXKg3$KK4}X zVSK2G*(ZQ=wa4N2B=4^H^36Tq+<`HDPQdroJ?-CsJmZRS>+NSRIRih@-kG197rDgD z$=_T#V%If!_yb(9t#_A+U-itD^91T3F6s>E2HeZo<%*v`4Z1cm10)<^O#TDhfOov} z_D{GXum@Z@HLI-|)mxjKd1t1GgG=mAO-b-mC;^CDXU~Q%Q43-t&`L5hj-AogPPGEB z>M?yLa=kC{0_0VkR9ruLKCyit0)7I|G1N%Ln)mJX;|%ZbFR^{s@&oRf+J@G4mh~5S z58!*#k{sH*mE$|-E-%162C;xMfxVs8!Ch@mP2Y)~`oOrp7$8B+t5!Hw7u*zR%zbuRoye4;b4fpyjxYruOLdBo|P#O&{GeZv}H zO_zJ{@AP&*>qX_^*|%@dOFKlaTD3a|ID^D*=HT96=>>d0#u4jiIs3k>^DaiLnu$(r zXD53A#e9RerN`pSJJH`dV*wf%z#YmzW;A%h9UjE&XT9gvPivfuST8rgU7h&Ayyx|- zRUd0M-uwKK*vwY;4r=#wpPGYj&jjZqZk>7insWzp4SRwXAG5S;h8kEa$GtAVT6Ya% zU0|ydT3scE|TUeS(P9qmsF1I@^1*W&*`J z=bng{pxx~S>~}uMD;cZra#nxqTdqC>EjqYS-{RmHlZVXY2S_l~vG{>48gRr@QxDJuHQLVeE(7ejJ_~b$Io?anMrTk*u8|8+&lA7u z@fG9lQe5l{eYJ~^?!fvE2AIGdlg)BfYqWwXZ!d_-M$fKD($)6Sy@b;fR2;z$HPa^-!O%>_Eu zpn;xi?nLXq1GCPXi(?=6R=ZyYKE0fE2bCjV&zW4nJK>jLdzUj-FK5Z`#?_qBy!}do zpW5UyS1VEy~V0Grr*(J6Sqg=yI(_3 zpg6<$iTD%LyJ-xVD-iciyc_Sieh=dZbq(#A2DUJk!2A9T++)|;+kWn#?aaG2*nO&#uodzcT0dy^o(^++Rkxa60`4qz779{xby9ML7#wq;)*8Uf&LSJ z*X}#082pDdz&klz*R0JUQ*$FQHgbIyo!7CeFZw}kHwJ1pI|cFDfsir%i5}=cmn0mX zi?Z)$``+7#wVOH570!TvQxkJ%IqI1!W}TcDIpo~YobyYFKfxI^awoL>Of(q4^Qc3R zmuqgrdDht9J>)7G{E&+tVCIV5;mZl2?#AKCZQ8N4ar+jB8h0Du2I_vuxpTl1&Y2*w zJ-7BlF1qsUw#6NKuAVJcG|}Eq0A~eoCw=9NZG7LyNP@Ax*Dk~r`_5doHG`S{{pb_% z>KDhefMflQxH^c}nGWq-`_w+zcW{Oq#5%a)E1CR=RX?zu;W@lFYbLh6hB0%wX1bf) z2XH^7_9bpV&uPBp$2()s6M1``kDRdy$KB-zc9OTJak=$73R})3?tjPen_P0%H?+30 zwYIt9#Ik#?eiMHjNOQ#N?%YB9=;^M;pGYYVRb5y?1cO?eDX=|IN5SB90mCVO`1A zIM$880MFO;&Z!f>W?Fa1$iL4|e4k~-W7Sw^uaaEPcA}MtV;A+0eeqB1>)sRF-Q_Cs z+~W&#yx#=XhnRDfj1}88D@ZIC$NZMfaVO(G!-1}QhV-u8FV}oIEf(LJ!0+S~{i;FR zXN$9~F{fQVT7RJPh?O5lp0n-vuh_l^6VzQDwa@F?xt?)b_hk?L2XMD0*Zt!FzOhgv z&M+6?$GCO}0fu(0&$Qgr5B&%J{qPgJo{<_T`oDi&-z)CsXV`P>_sagxk!P=6*NBN! zBXR<4@iQU3n}}n7bkoiwc3pRJK{sNGTxTWPxI4)o#zB*d_U^nxKl3Bk{2jj$JAse~ z`k}R(cCPQT`|CP);u|x6A@%^BdS`Xw4S%4EL#+NixA^FhY;nhT*GbGi ziS|t86FQsERUhQ!UqA!5S2Vd6tJlD<-H1K0FTitFPdJ^p7~0I}YVPt5ZxVO@K!+kJr_o$a;?C?A3uf1e}@cdtcwVeoxuA;WzX^ci;^D zy_YdL37lu28=MPJSDctN3EVG%eFA@wtIrv}XBlz4yZD~NU2gE?);{=W>`uXEHXi1E{|c zt9kSeZSu~we*^aW1F`dsSexJld!XH|!!!0`?t$lk%fip3Cse`zA`zh*-6^mSb^tBuC zP7~Dk?%|JEcU8T~*~1-#e0Nlw@7xBAozUWfeL+X8vvl;GSkJYu{fkd5v0u>cK7jQD zUcDpO)-*WYzq9p!_eM;KHEW=Kw#MBxa^=(m1hGWhbH>Vd{GZ?lzO&tVpW(NoM<;HL zIiLB~)_fw?(Bi3$%vkTKyNX8Ko-NlLwR=o(B0i$kpWFCZ&&-^=*wdWzwB5sA+JQFi zJ=w1lOE5v;%Tt$6>;>O5gnWE!^#>ehJ%IPx(DvNlE%xYO0{eR>)d$;}Bp%r6>Pzfb zID^;?+d1a;EIA2+n6niB_og>^#(FGn$F{a&*ol4CC$AQAoZ-F$`-PZ2F9O?M6ZrlA zV_$dCx5jg7+avMC3w99GzSBp&_YZ9mD>l&+e}I!*6@>0LBsofV2Hcoju%d zqGvz>&KUTK?K|%-#>~qb=o>Y|9QW&a8KT+hndGdcHu z0_Q}m&-OR`-y1`EEZ&2>b_1D9yHnyT_8qbIbXEuY9Rzkm`#G(5f!!mIJlZ~!yk~0I z)}O2aJ|p#)>pHu2=Ddf*_PKe6h*f88C+^whXxsNpykUFhl34Qx+IYP`y zv8pHLJoEj!o?nueW4*R@10CQF$Jqhj{eJhh=6@wF&${Ab%fG;rpZH40SKN6l^S4g?RQ81=FY^% zcaOXLw>7iQY|fY()H?xh$1>JF)_3COi$~1_cVf;q<~@D}eRGKiw(;-Axf}Ttu6*Tk z_#@Z%pTB!V^?Y5mot%AYHZ^}Bt_0fHhy{82)=q5Sjk?1leuuyRSu;=U3-|&@3AFVY z$8*Ae0(;jScDKeml;^wU``^(oz;o=gnU_0(^R0KT@c|l$SUc(`IK%P#vBP)fuh^eJ z19#AgO(5SpzFpTY=Ifm!-pDHtVvW4$x%RSW#mVjaYvktiLbVr-J}VYyx-q48L|bau;H9CvnC46SlFs19UxCdBm$HoB@na z@)zRn71#-!*LUuI)*ZVE;(@krmvddv_nrs7@5ZnChpn$&yBu-n-oT`X_bBHH zCvxpKh>QAXEur`9hW$8d)z(?_h1!!ow)f(h_HNGc{qg(6J(8ULjJrn=i&%Z0_||%s z??+sIlHcOKU&%ZB0_O}m`O6WL>+FGkF_y>s))4#w>}gGyKl-?LBR;?+=WZt;UyeQ1 z_7~PZKm*QE+@;0pVeSdsu@jRM=}3s_}`B{<4O*@b|BY5?cUaz?J3{9vpkFaJ&!m6Ip)pDF&1cNmDUC~IMxK( z?;QKcxno=FvzGH2ZBOsPSf_S^0gPMIl7GP++DPQOSNBldA#`ge$Taoapzbo*EFr zdB&6ne&(7{0e;NjhkW&6W0fu6KK5{K#yamEJFdsyN3hS>U9gj9kv9)mSA2ZqyXOsF z}{VQZq9xaUvVGL zGGcLzS=)&Wt9(5jS)b3*Jg6+&3{#Wnys-JJI$?wD&LE!}~H`8vhHpgPKE1egOBipE@>&xOvYp zfbWpAJD9}X_kQ^5>VuZo$VYs<7ahA3^M0M*1iUBkjy=P^>kF{}&alq__D}s-eFEF~ zS?&?jK3!L@LCo_zFTj)oss{f7lnmmp>x`aVG-9W1CqD9k^9k%f-x#x}XR)RyW#9%! zJ%N3l-Gn~pRC_-AS?{b1`A*!J+~i|3#eUw(7rGB>u?8L?{=b%J&j}6J4C(-5@PhAq#t1X|l?XncUG z!}old>zx_sPVRohs}|q64Lv~#zIB_M#Df~YD|#+U^(Ai4L4060ZM64OcZ47M^m4bV zKk^O9Ts-?vpm+{zPVk=~^5{)m(KqL;yD?y&2fF6EH}h}i5PScMJ#xjYcOUD`J<0h# z*=yj}xyYICAP>ahw^*D&YtKVlCu%oS*N^8T$6f;+hrNAkJaf&L1KfYlt2-dCZ>{jN zm}tfNPs|#zI1O#RedHQ9)^fc&XSj)7Gq5|kjxX2Pncn(-|J%Fj@$dNV8Q23PsJ`;3 zt9Tss)P033ulk}-_>;WchOgLLj&rQ9dFX%iw(e0skGOj)Rm8R}TOYA2|(!Le3O#SZOVd9FGeG4ITKs$J2SBWGR14q&~~lRJVl zfcpjT-O;u_(G6|PIU=yd9atX%o_2x>w!MFk`*a?n*_j5PT*JAA_&sn>^;k)A#)UIgfIoBS} zYllCvFIV=)mwUr@zOaw{x)Zea9klBQXZNZcaeM9F_OQRXKx;S9w@utL*fWW3?HTXD za|iJUd22kQu|ds+m^BmVYoCej=O$0t@9xOOl^?(zwXM`9Bg#c(M8h_6e>#?9ago?)L3{)`w@EXKJzVg6;Es0rqR=a_wGcBer`5 zIZjDg*ILrOzWfDhMF3cji_-f-5bl@4Rt-bNh>kqVZ-8;bzxSO-B4bb5v zcoK82{Tgr&<2SVX4q}0}MtQ-Oa{|^xqF+~y^BzDkZ$IlBu`?LCp6wm`bif+p_V9Bk z&-seB&$g*!)!NUQioA%$k$1lm1HK3LPtXNC#aZX;ai&~auFrFl^Yi2m0q)y|=lxpi zne6`p8oo2lZNByT8AlB~`CmYzPa|$$dwGVAuDB_DcRm{f9lu@I-59HTzz^&jxPuG6 zJ-vgkN8J3xe~^0u{Y*3%;EZ2;lXExEK8Sg4^C$f8#=xDJb?)Vi1P^Sv4Y-GV&)o9R z_}cF7d}s96)#r@9fFy723wZnxGv0x^9okyCC*Tepn3LB66I7gM;Y?=?VxH0WT6_bM z~;*zw1vzkFN&+0AcqIbYpB+|qEIX;1m;Gd%a(yItTKAH?E^9=o1w<;#at&o0n#eJ9-K z8~UfW^?4fP66}8WgHtnzTVH=3gymw*WNvsRFowzd+?GDBiee1u#caL2o zch_zGxULzSGvHYJpw7E3p0ntGHx4H8fM-7;$GV86Z{(XG)?)3WK3x}xzUO{NKmWTd ze+}dZP$G!KuV;XM0lsgIc+d5Ir?zJ{PVD(g8W^{3q62in{@#aocvn-m^G!MUnpJb0 zY1i8S3`eb4TQ)H{17sdCZ1X2_#_wozq644h)OL3yV%Z&ivmbG7`2iZxH=gK$Z+sjv z^*mzx+{V5#$Ip)1_rT|J2Hw*H*w%GWNxD#~XuGZ|hzB#pd?aVb_JBSaovgeN1bw5>J2OjL&OBmj_cLc)xYGqRZ1c(pTw*ulAn@(KpUs_^ z-|b1f0kxm0_C_swCbmA>XO-yMF^FA{z1$h!S|x%XizFt)$DrKEd}n`A*1NJbiZzeZ+eZBXNSfXF*?%SSRitabV}=|MW&nc<$!DyGN4SyIJ4Nqr

Tm53~ zws$Z;iJ4FA0jzD>vG(fd0-T0zyeBr$Rw^HhZ!D6aeUx0iBi5Xnb=Y!}e7k-;#~*qA zd#wJx{1u$RfNTE-?BhLEJ$}`ppTvA8r<`1I`&%PNyC1qGKJf06m_6mGA;9@HqtCa7Bce)#r%ysqzB@B8JjeRl3$0!LoNs^4cc)`$&YPMNXnr2aebUW~cN z<>`q?U@u&5>XMXe^*bQW?T6>jE>?9w5 z9x?6TzrrKuckXA*s_EraZ^*pq`J-k;D*eAHk?TDGT&qSNkSITxzq+*E?YghXlvuB+j zJBd~N=QSsY1+lHYH_vO|5ag}>f^BTzdp3Js&=0ir)@`o!yT?WQl{(<_$Xq*BAL`^y zV)iqCgC|GXJM>t+2ikdex!CgEEppjsYTc>MN<85-ksoK?`*vdP<=%I+@6QuB!~X8& z9`?3JfDX(X-(tNvwe!8F7oa_Jok!i^H1zF?iMEgN4$OI$L5=U1eggRy`PvG(sv&0o z9&2}bT^|ItzWXT7>)NsIv*)@C^OeI_W~}+Ja4ZL z1lran+#cYM0J~vZt32@SVb1{PS2&4{{t907>0`W}6?z=-*~?Wju<5T>daQox0ctk7 z3G3+jL?^J%ByKFwqJwr_&(+!;bKaTwSFrz{Y^?a`)V2@WIU`o>T^j^w#&X52^^S^9 zYLXL4ya9V@+hrLosWEj<9WTq??Ar;>wW*efroK?bK-_?PkZ^C zP6zgU&!@fldDM0EPj4U3?>!Dzdx8#5)`K%}?jUBafjHpY3rPGYych5ZbbMzIwD00X zi;Dl;s)zX?HjHCe?gw3Gg=#Jn11a9m=)?(txZfSPuVT;0kAC)d&qVvqSX=joU(bY? zy{2(c=VpdItB*V3$L>MgJ*??qP;(*nkPmLx)yG*U5C^9lr;oAwVXMdMs`EWhddV5q z0OzZlK%A?eK#?z$8}+_F>e@F}{pGqo3(l(D&<)5rYrkSA8tL)zy!7RZ0M2xu2F}({ zUwzog_22f_MocV3f_~19YQF;ijMie++RMAR zfCm`Foo9~|{>0uK&y&Qx3*T4I<$LUX$_a9HS88JR?8F;>Nzxx!e+LuWKQG(IzM}dN zPw;*o&hUWKcgz|E?L24MOTWkJx#!!@Uh4fmKH=+&W}Lc+k9THb$H8f6@4>Secc=Sx z&C&P!bN}w^=JVrh*4FbT*2twAcoutB-hSBoe3N{Tciw(i?3n={oXA=C)7$zUsh-5; zG|7_rY@i`_4Ay1uXY^7{=)wkx|Pi(lq0cSkavHoo2UD>D3 zO3b)1YbOYOv~lYO+MOCOm%!MK+&5$9u+4c7&Xp_2KE~~LUEz|9`U z-Iud9VznpQ{(B$!#;v&!cV^XT6VrCio;C3^)=qxrv;(-ieH7~=(bWB`x5GAvpTZFZbxlpYgQ@0xK}=2o|2wsV*qPco>(QgQ)FiNny_7rodZy^i z3axtzZ2SvxXW2{JKKA&EZT^NHXzh|%d1FsXKOfXD)Z*SKpt}*g?F2o!YN!_6cndpIsx~@XcS)B?9CM4eZ*L9u>oPuOwe_`PbF^jxE1~C${;-woW^+oh?Va_rrEKzf;ekC3$#q z1MGWk@XROL`4eri#;x}{`+?}GWR1STCuk-}$SZ9i5*L{wm z9{9yUpPj>bj42uX5cD1BIxC!+gn`d}e}~?%J?j%xjeG%5zUSC?6~sN~6ZlNzdlvKV zZ~`@h_|9!zXC_>|(Esx%Fo1FM_P3{6&eqy;2FM_eHdg$GCg$^3j8~35l)anWx`SMK@5lYr9az&q@dFzv zew6us4|<++<`4a-e}IY+Pjk-O=U|sV`(?cce~R97eK)-ywRhA4AkMYl9jHfs^ec`& z9OpFh_E2i~#1KgBpND3bjve~Ix;(UVzJLpu#7^h{4O?yi=MG@L{lZM}^}2eSD=ss( zt@d+s0wY(Cs>R;oot(h0aNNgziyQd4Y7*^!!n}QlAl?rIx_stq3%O5gKxuMv8c49u zD%V)umHiKqv)>u*0er`_-JuJ5*3Q_*h4Dbw`v_k|+O*d302S^|CY9eZLo zY-9HF41xW?wvXcMdUt|%inw+1J2=Bvv=f*c+UN-`@B@y#AZ|>~jktF+u;o`DVm?RX zLHvw;$9@;=8SD2$%T=dukA}9V@xV^t9heK^i8d$Co^=<*%^P=?b8o;Mrg7{$wt4SB z>A7H^>W_AA#kpVS2Al*FyY@%>JI$DLji2bPEx-1m|G>6S34A%*b}z-*ho=Wa~?-XRX({hblavQce`%LTyc3@jCFVNZ@yY3A; zu;s|VAN)q#K614^%jR6L&#%M~YclBt*CCpcq=`ntHzH*P;|QGD+YJBWL?GuA%4cKi^R@2&$*C!U}kc%g@T7#nD% z2S0N4jYAK$yjZ;&cJ*WSLz}#@v$!ri(e4|>N@9J!eE;N6AUBA+<9E=Hcta-f2ISqy zeFD}cu|QiV|8&%~!^9YXRv!)=RW)MYF=&-Q+xh9cGbY`z+Jz=v&PwZ@H)2rtQ*G2Py7dbl~B(#k%X_403_(b1d$`lW*S{3xV!n0BeoEysk6Koj~#atn0u(ODXO_j^oc(D!klsU zvfi9_eP83-=R&NZPv8s!zO@OoC%Sr|Pj5$_ngMr^mn%n|#?bb>A9?2`G5xxWSa;{_ zCJ>*?(|m$hdGEA#Bfr1P^)JK~zssG!d*1OMmutSb*ewBK zUwg?*Y&i{kf@UA|*Vpymsd2BsuU&G@(3d+v7d#)&DtF(rxdyDW=fn@l37jLGe+MPRA6G22eVu8q z2YGkZHXcCRTIaZb?MQ6jwR*@SXTK1E3F;j9T`_V?1o*tsPXq5HS*RP3vBfl>g}O+hUf7t@)9WSC_e%lUGFh=!n4M^u~+3e%f2&X z&OC!Va4#`|HQ&Gho-c`6Z{Gfa9{Shi{{hag!2Qf6IraUB-PG4Li>mb*gVVqS_PUYZ zeFMKA&~AtA9oIb)GH*?yGx0s!v0Hnkgt%Is9M3t>o^|`y81o#WV^<$Fy^Kxm1Bu=| zcPS2`GqySJO>{8ug*z(lU3dSszC(Pb-rH~BJ7~lw+I!Uh>EO*+XR+4%QMT6ZvG&-z zjASm5^LzTZ%M+yWSbNqP>1U0x+N&}17xurdziZc@S@He-ZLbE@Q%%1!sF`uZ@%_H^ zzvu4EW}e*M>x4h>l?ZTW;s&xnJ>KSM&L4Rw&XE5;HS+Spe zUx2;#-c^&wnb7t#Z(RUIZEON%#Mp7!Tze&E-Pk8>_oZv^LsUfW$WE-B0XhEwbZs+{2y$eh=hFE>s_L@SfL? ze-CZq`#BisI!EGTyZhe%j@^h4u=`ner|tzjz`g@{?k2Y%_KQC7e*E{bd><#Uw*Fir zPr-MVbJg}V=05fQaGvUqGFNw{O+4WQP|W#x3~&bTxjf!8dt$rK0PZpI1N3M$Z_hF3 zzGqN(`U(dqGx#BtT(u|j?6dnN@y!{rau+zB&wj%?Ifor^L~*YlciV^w`vsVXZd>p( zR)7C|^Bt&_k;J(ZXL>JBP>2ooeCKrFz7ILqkN4M{&&FQX+M^M3hZ3~4 z?rg33szDo5xMr2_kOmWhvN5$yIX7j>g`#PlRV-|4|)MP&bSk|cjaU0Coy+% z|3NIl34fXgp3#_f_ED_46YsJ5{#ZM5-D7d#gdF(ZXObJ}@4!8+6Mk<^c|a`&a7GDs z19g9z@ZHh;`QUW&#uR%h86UqFyfb?ktGxQ>b&vMxx%{R+>g(L!X+Ozm9ffMZPVLp$mH^V(rz*9^2Z#xHPg2d{dO3UbEf%ve3}-k*uP2k`{=*CU_H zH|Ok5Zh)FY&Y1qh9s;|8e!#ul$GG3^nQaOBL~X>{OHS4OamBQ~Gw1A`jZNdkI!J8u z1#J7BsTs!bH{aM3c;Dvi;b$+f?R^I)^PoO2mckjJ!|CV-5?G_X&%Mt!iFFWQW?MVa z*O>iF5}zHaC%WOcSe$ln)fFda??69Zk8{hh-Wqc9=3x)DIgr?KprNZD>P8>yCb0qJ zD9-Ti#GdCp$V;$!+BZ1%(C%Qs3*h~!8|aC?qS@twKlG6q)Zu?UVu3Gb=10GVeJ3wh z4D5!ke!1qHjoE{^TxXki=Om^)@OuDub%wkoR=p1UiSK?F^n`1l%!NP;fNp2<_V$}zI1McEX<$-UHkss%kb25(H;n~|>?yqhBOkB<&Ub$S)#CbtX z>B$e^JZtX2yxO`z*PQF}ANcZ4`0~6Dd-&aKjq}X~G2_avH`n9@I?>Ej%un!*I?vK# z^)_$csz=-J)6TI-gw-v-g0LS8UGH zhz0&YCpg2immFg!Z2x^7#r^~M&fS5zdQY>`WqVK7noD#8drnVm5ag6de#loN`<+*N zCk>u+UO>&t6_c+Yv1ZlXU{B%=IN#@Oe&_6EowEa7dtqm+cV(ixabgp_&*8gq2llBt zbiFUJYR~xgh#!1+)|bEcY}hxr^6l5j+eeP|=ht=qFov(T&jSoN)|`Mn?3EzMz2M8Y zUkQ4EdFYzOea>9xGv297w9h%v&hfo4t~TymcklAA>psu;@+W9OJMys6J~#JrPy00M z)%e9jR~_+;br;q+$Nj_r#%d?@gcsyH`eMB{ar^$*vr!jtcHDa!V9$4+J`-W}I`T?HUpFrIWeiymk7h@CKnA`^Puzl7OTdo{?>z|2t^w731I|R1; zy3<@|k+1I^nJfY?W}z-_O?fz$NstI)o$2H&R!F(RQ)p7KC$+8&yGDnanL;habw<#JU;DzF85BK#2#=e#$4mh z>BNntHnE8umx}cw;z4?Ki`p8Ir9GHO#6zeX` zkk^58_MM!G2hfirZhjotGmV+Ie*mTKJ<445dGw*5eZ7k`54vE+(3)EPhZLO2SAL~`QW>Q z7-CT>$zg@eAVcSSbf#8C$I!nCwz38!J# zzQiy1?x?nxJ-YaLA&4cgR@|;1cVQpJ_!j58XQ+{S9`uuwv3B*m*2?iY)Y%9*Tb^2u zw&)=6C;EYYKi?5EcLVm6(}1&_tq%MFoY(ci)IRds7wYZP!To^mf&H9UfN!r9)L1_b z-G~j)s41Z8?8d<6N32nEB$+;!JC;`O}H@iTajZg#Tg^l?>x^F*n3y^(NDBzE3Pql)ZGv4Ire=a=HFEkxaWl5fx1MU z_#mrz)_xw+wPI;c0d;w)2 z=iPvP#sPWeCfK&TlicgEJDlBb5IdU#otl=bpFO*f`^S&E5i4##<5dUGnbxR1#{--} z!@nPTkni{t)H%?bA4hDUXRhyb2D%0IahAlskbmC| z-*d=yo^!mnj<#P1_G)1F=-BcST@s5Q=*V@K_H#G2GII49uQ*j&Wlp=ojh}c<08Qy*c|CRs4@tv3?{9nBP4Ie#DwtXT~mJC&j+z6>Xna z#S#s?e|uJ(zQ%g2cBs2b>%54&|FS?yS0J z>gU0?KFRqEl%D(mH7C}re&{p32Jri_`oVLC`~+(I-@ya4Sh0OS*2od|R9>i0V9&Z+ zb~7h;T2H=^DxvTx{Iq|D^`;bAb zACPa4-LsyneykaC_ioyoZ%x-n7qGiAuwRKJ>-YOR4%>Igx(QEy z<*{GraRCqD`35m_H~a>#;X7B~8Rk0LS@zyL$*Vj)9^{8+@4NM`{Cl)_cRB39wr|5PiM5Au`^x(&2itxFEdoqn?Fo+G zX}`+j{AacBUapH{%-=tanwv3f;ePfh34Y2^Bkmgk4sPb+yT2T3hdE~IM=s6J@-xu% zNIBT@doI5Bpm>k=?dS_|w}~%GVx8ABp2-jV2X+_-)h8FHA@&)d#nBt>pA( zcH-ZNC$S55!}iYHA;1iLa@t*++{Awu>;UfOoybwN>)f$E3+`>5yJ^c0as#N94D#IL zgPgVA#paAW?uM9geS67wr>=kf;Vy|O&Y7`(*Ee+LdY2m7GX*`2U$7PV_Njbg$H)HG z`i|W|5?2PcHTG~9`}mHHJkE%AU(fUY?%~S~YOHeycbP!3_XOq!sC^E# z`G@VFwZHp6{_X$vzyF7S|L6bi@BYuf`}4p0Cx87HfANq1>aYLTU;dZ&umAtb`u{TX ze*@a`*b9BR;)CzCBfE0HgSa>o^BH+2&!Hd0<(#ja5p;d-sKaTXG_m?Dldl+lkn^*f zXnV9*fZT}PfqMjS7diEunNz*diaV8{-+R2SXH(BLR=N)NGMDHD~G4GCpTmq%%;+i{=JA+27>R#6!%=maW0^8k0JG6E7vzN7%r{4&Ca_R}3 z5xM+3@!!3z=V81Z`HmiN5-oRYpW*-gwx01MXN@`g*UUeD+?Bh}ghNcu0P^)cpL)Wv zzxWON3+0qE(6KY90p6jx1ee$Kz64q^F6=YoqfYxt+*$UicL=+KvzmF0xVbyN`M^G1 ziRGF(h&61V$%VK%#s6>n$dB*nM&6z!#66ey-wwWWwpeFRXO37i5;% zyg{V;Kd6md_uT1e-dzLKJ>s9=7RO#4M67wvoZtr69R_yGbr;`X*W6CbI?s3k6FcP&Q$_05i`W-FLIr?W{FK0RTAN{iCov`)Q)rb7ZHQQ(FceJ%h%rgeszInu* zQ+ETm-UahP9w>6}Ku$wjKY{iccpjx}bmh>+&(+f%nl`YvHD~Lz5kE)9t+`>lhk17p zckG=Txom5#vG3TnJbloKJ?Z8D#?rh!{4Td|kXM`|^zEN$KkM2eV*T?ttWE4G;0$2i z`_@-K;0r%50na|Q!_V*!aqsN=k+^D+jw!&Cu6}3dtnc3Im+HQjyikG z-;(%^I&~L`k3_5)H7{4r_#}RQTfcL@SZ9u%2XXa-m~vx=J^t*Mm222_2K1BM7toD~ zY$e(O)D3vw&i72ySORna^@aGpv+AAe8Oz`C0k4Be&Kf0S_3*iyt6Fy7vx<{&z@9w= z9-O)>;^ywmDGs*0s()L&POSPe!{_32?r{9?*)~wm;Hdlm@c$p+J5+V(Bx%q^(y@TJ-gPgp&EB4oRx1Y%0ju_yS;U)HX7>;`VH>$_*XtsSiI zx$fQbxpUz6SUSA^|dU&34`d1K%@=Pt+T` zpp{6tD4J+qgy+^=@SwoeOUK>I(vt@mhRR~+uk>-y*4`CPtY2RUa}O)Q@GT{F=2 z86aNwK-zmv{5WFPchC+@eRo2CrapiYxz6ET1hb9H%PXH#;CEMx6`Sbl|MNQAggc0B zIK$p0*tJ_6d*M6VdGhzW;#?uOliT;*5I@)Q6R0<5#<7PnFxBi-f_-v_STOEg)x9yJ z?iBt0j*++D$fGB@x_@enwTePb^0pKBQS9r&F5J74B; z5B3GVSCxlSeaWj^5Pt>O6Vy40CAMfF$dAJh^uCL-$$7s6$nC)R0E%`=kmKwVU9p6& z4ctL7u5bT(7U;?mYq`!-{KGB|JqP}-ahJb&Ti=xpdzyo@pONSZ98 zodFH|6FGO%?!b9ls$VScab2IUuV`zlUmm*>^B&yG9_Fl7X7B@aH4?#(T-^7W#H<^j zYN>aR;?h^1=dgc!Te&l~GyHR>`4(%IeH!@jy!!7k_&b-xN8p11es|;n|5%XcE)*7u(~4bVjJ z|Nlz7qw|#>*ol3@_N>Ll@45ODt$n7k;|>RQy-aU4|mL1cV^EAeg}6Dx$a$Vi}m?^A?{u06)p(k!nx`R6wl=6U9sopjab}{ zR@COyPk8?M*7@J8J@5tX-eJSR^$e--yKvR{5;}aUbMQOfo<&_Br*B_m6-Q{ zma0hG{kK^ZwGAjDN~<`Cu|;H89UZMkaKzZ=tZl}8%z0*u2@H75V47{ld@y1dE!u!- z6f@>+#t1qF#x(c&Rqckk|GU;*>p%BfYifNDZ+hNPReSg8bDVpbsV2LfM~uG25`kR6 z9QXlyBbLZtsmr%f?9Z8a`Iwiw%!Te;#A}XL{EXzi z$L~r%Glv#e5Uin^5kE(#|6e>a{6uz9nMcv6}7B9ffv0Xe*WV0S; zt0EO^gMNkLp}+FH^qM&I3CrtWa-5_ z$$MIHeolA76OeodY1dVS*>#+9KrWE$TTw_Rz>lO4_mz3+THZ(9mHckSoWNPp zuOfunEy%G3KJ?s0P9dF@DTiG8bc$0e;bez%lH@(-XCU>-rw%b@_tKfKQpPej=T7Ls zS-_|UXOJX`LHwP7SiE_{;E4&{!X}xi!xa8ATCpt-sguL|tn%U1}jPVv}NO&}MY{M^TWPLh0l0h42$ zPRYb6=F9JZl6)tm5BD$*UR@k5IP8~Ng}@w%q*FW-oQv}^OD=z{jEfa)a=a5)Adfl7 zfqzUZ-eaO1W~XLT`qM~k_~IhHP?N8aWKY!a0xNStg!_U=y-WyklphesX z^bYU7WFb2j8l}!0v8WFpUna>caNs)^0jE=Tq8Rma0p7pI(OF31vzPP< z%H{kyt3q}-A4$GT(yJ+C_|=D4mu%*fefI&*xGuo1Al` zV{b2DlNb7WNhe!?i?51o$0eD)As>D?PGFW$LoxheNfxDDIOWvkm1Ljzsm=Lh((xg^ zO9ID9r(W2jbYeW~xr;pu=`22PI>p1^{yn8jKI(SD5egq?WG;BfcCWB!*-e4(Dedn- z;oih`@-s(RF3!h{PN>SyGeZ9apO1BN$=c-T52SBiA=r% zU`t{EspI(!FGe z$;WDn52vOYiJub-T>0V8+ltRVpOe%lKNm7#CjBHszEey<56^@A@N)UWl0+wOy7H!p zH(CgM?VUP-E6MUM)^R?{s&c67{y0?(9}-wByesY@5S1-M>F7xYTafOt7^A&qqIJ*Hk|H&YgVa{2pDLLZCjioPZBWUwqsNx%jZ1bQV3S z=Y*t}D>`;UyCx)b!G| z+Vc?W64&|2C70Didi9g63e9>1{c{&VoLQV~0I!gby>rIYCQkjDK&`f}NSp^net6XN zUb&Km5bDV9N-&cXW}mJ4ureP+$fs}OBB+Dzyf|fUB0ZOneqJ*0WrpSFJnIemxOm>j zwBq08r^*NTKGZ46Suk%+I=>&|V=vg;Wup8-aVMEcGPf&b*c~}6tR8c`k2W$9GNDtk z%>CP}@==EscrMz^U3hV|KYF0&*npmriHq+Qx|2>FI9{l!4~x5UUVFrL@!iBO1Z;Ag z5PBeeJWnr~Xodh+)|K?79zG+OL-SR{qeitA^Qlj~kgrJ%*-{(%V1($-_^5c!~6B zlJJ~-^mFoc%EkxUJuB#nkp~Ablb0+6{BWsV2pLelPWgSQH^uJ~GGCs#BsHACGkcyu z9cq%#XRVWTviY5Ux?(4kvXGzWI9H55h2&_(>5(Ybeh;d$X9+wrnK+%&+n)(y#{Tj< z7xT$w9_9e(^yPlMVt`tOkYoGUi#&Rw-JJ#lf4d6ieiFFPzB>*kW7hQu`SIdg3O#>qQUD2`+vxICj93@v%0;^XH!bA|mW zH%J#_Nn47WNzZ(_XGwAk0Y1{XxW_^g9WH%&mblAAc9>PVm+qb7(m8YXn#)drk8vjn z@Dpzcg>)w*(y3ke@|rR~doD(!~p`PwNug4w^pODCACh`C^bNM@^Pr#P$<)b#h$I4s;bqe&2 z;@`ZH?BU<%Cr%Ay;9~ny*G%DOrtGF_(~p(;m_r?M*^84z&jobu$qCHLy(PY!SvKby z?o2w8oJReBzdLG^!~Q)|pceb|;^W1RjC zp(=C=UU1^mA4v?-*aO(D#MUtkXMb9BWf2;D;F72!)M*NE*>@pHFE6y~hO>pko(1~Aag$?B;)!e)Gou$J+J5Ke zJg?|1@A#*zry7}Te4wE^oK2`ByHZaA7u(4mhb_r+KIVjTmEE2P|9&my3YjZ^W@~N^U;RmOPd~r>|7u(kqz)=La#AQO8?G0+ zsmXGR`^a(VC67G7dzQ)1o^siRkO^KsfF5#?0t*hG5zJI8^PGHD>9hNaM_%t%{0#NP z>vC$5cb5FS3$vsrl9^-$-9w18bjl{L-KQbDP>i}0>gp|^yV6%vyj>@gkD0t=0(_iX zT6|7jjECIEXLytPRIp5FYy6pawTNNXn>asY?dF5*fUGgy>i}?z{$EU~%IOv%qHFKnT zqI-clRpkI`%oMh0MSqBkQ;5?kKQ{ZwB|9aZWa6Ww2R3_R4_Ku7+=+w zgk130E>R%H@iTi@$tP9_a6u-(1-X!wkZ1P0T*%k%3%^L3$I<|M}}M%`{~aXIIT^bC&dM(N|e$;({ip!3c(>+w9O zkth7GVI00vl1=*PLjwb2G5r$B=OahPn=n=XeG&q0<}XOolP#C89D`c%-9gxbvP4d z%Y2+6Joo7MC7CyK(pl0AFXyvr<11;MF7x`QbFBgZA$%UcUC(!=(nkL^6@gg&=*L zr#U&+$>fJ$2$_6ID@yqIa{0Lba8{Dxxm3&x)Xs&l7fDu_*OffZL4fOou6hJ>JFl3K z$WN@Q+Tl**BbLZ+=Y@UCR|xQ^nMfxOZe9XEKV!PZA(oWL$~Zn|P z*`XeH%=e+p#s4PIDVsAGjZ~cZ!W?z!;;;%A>$hcyi&uCBg^(u>HCfce53fc&7xTSp zhI3}0@--A|3LdU(=BNlxdM+P2G3>Csbb*@egM4&qW%AQ+n(RV$wH|Fn!`aY>`fwAZ z7n~rSKD-n7xD#r5A;TvW(sLp7aj`DF8sf4z+pt&JQ_)?VA=DC&dQQG-$%9K>7XHMm zz9cg@aq3-it|0dlSfSca8HS3 z$#p*JK}t{L4|gL8zo~l63#dVixdP;)H#3p%6tBq+vnE>d74l`JkGK~S+1|%mn!dssq%gKq?-Vi@lhQI*I(~;b&$mK4;Fy+`N?hZGxGq z;)c%`*_=Q9`CTB9q({h8KPNkrO+JhInJZ^k4Q9$^C-_y@%jcvge$IC@S%Js-kk71M zaVMZN3wJO{$mAmz_!880iZz6G4e!Ji3UNRt)D(A;iI5|;M$3nv8ii9F`10qRbZS;w zu~uh2ayXABIocb!xp>49>7ll4>R?v{z8lwsOi1K!mNwi%pf@!@MKPq8Kg7sWoL<}q z{m6qyy$roo&LKTl+)095@TE+|bxyXxT+Bls^vp|YHKnuP!pHM&`xDvtkfDyG6ED=0 z9-ccF>oFI#6WRDfeaSE{`v8iCcw4F#EAU;cOMT9c{H9Q?#~GN9xrm2Z#rFbn=Hos= zm{m44oMQAOkN$xVM<8CvziCTS9MDCK<>R~JB;n-pPY^Qc_yCLfh_mmoPZ#H*5A~c7 z_U0wUO=RQe@8a$LUa?$&+fW^9XVRxhXK(1lk#G}X-qxJYCiwWB>Z9lArNo~AE%KR@ z7`&h*&n4&D-a9gUR)|-KPoJ=V`MH}+KHg4#I^h?@OJYw`APQB36MV^zSK6Putba4_N=Z4=9$f-%f zV`VPRP7bw0Klwv_#h5LVKNqM!`+hZN@XMC~@sQM`2K+*B>ILZC<~#YZS@1GAlI>NI z4LENvf8LhTyQDKKeW*(=XA97C`H3StB@-WYX`xmjPNzUVoI;?-L#_C~!^@;|7v16{ zpj_-baGf0WxHt0P(}z7f@p*5m0=ev=P@I)J#Rd56ff;<;r(E(g*_C=kA|AOl@tX3{ z6CaB@#F&X5xpZogp9`+!qi2dcq_`gGllVL#1$@lXB`|j;J1j}>Rj*#oq!@dHp9oGs zZ_mIS)NAiQQ;eE?W-|}_Vh`b*#DkAcPU1>`DY0{*CJsFG%nRg8XE~>Te6+jdr*U93mlpeqgsG*zQh z;9QATtu9FTIju#36A1j=NmvpfX{t%-EkU#Nd4|IV!c{V$Vd)-4l z-cgrq?rNH_;85L@{7XralM8Ue-AXpZ@$z|ml8LxYf%D>A96l|0peZ;nhl{oF6|&hE zHL8-i(7xM3JTK5Mm#mbrMBKJco|8Wjs0WuEW}ptwPa+#$A&}R;mzwlMacZ-$or`gD znu@{a`D_ZE0)3p1^<>4ns!Pg8PMB3XGC|7c1>(V%sGsy)J}*iAFu!zGy9f7CQ;eK$ zN$!Jpk#~syR)x$e=f}{*$|Hv;x0aKg=D6>_^3fnA{~DPzL(jW6T?rO zl{tmIOo9F6lAtM_S-p$slucc}o3rQHHN;JXT)ZxETp8n>cW%Y+T-4&fi}Se>xZ=ZQ z|JaF>kgS%zLV6}&m|b$J{FxB^ngOnpzYx5@JV<=(pF3yHnabh61tmW2g?^lKE}Oly zbHjPb=Y+z=I&ia+$6gBQ&3Y0Rb4?JsWKR+bp(Z;KhZ!2OGfDEh?_PKJ&_;VNiSUNT@ep%flo+; zns^~!e)?qcj>(F3#bxfy$N$|B94DRH_*hlx z)Gu6|iyHLF1@B9qB(q|3HX&DW>Ul|kAIYKy`y*Zx3MZe?-Us=ihFrOH@|Xi3I{D~M zI-rJ^U9HDBwF~((fgXvZlZ4yO&81TtKb(S?;QtNG++I9K4e{E23+Y_~eG379O?+Sa zs*?CYF7Rx7N&NK4g#>8vtn_V3smDFQZ=Xe*=bU_6%uo|DFGKvyNov9KLZTR)u)O{i z2Q&rt9+sQ)yMpuaZpcrBPUVgh+C3`LLp||W@R*PO?avUoLvE%XoEg2S#iCB)WeVuz z7t+Z`XXReDlOoK1f5`({#5h|dd!SZRv2MW&^i4`kkfcA{36taBy!qb)k%P|*o$~XH zPZPZS^g*Wwvvf%>@N1!y7v@5?b{(AUnVYaLTBkGub~*{ z=7e0a+$CY*qaXZ2phjoO!w=633A&H(l$Qkbt_tCfN;}c|cYaN^szR5*9Dy%MAEzAZ zhlTWUmZannYfA4HGR25fI|DkKN%9xcsYQ=OI&mk^t6eWEb?@UI3P~0=z@Dl6r-fW- z-#?N#^=hh-3*0v|Rj}oUgDl|GlhBiT+>euF5#x*!$;9jbBtb4d4!s_q18Q^$oLxgQQw$F8C6avZjAxy*h3~}) zb4t$z>K5X02KZ-{oJ71r$l)uNw7If5mr6aNDM?-G1gSVZ9B?t8^PVd3_oi9(I2-xQ z+$r!3gmagVet>?&$we0O(JOPY7yI(?Bx{Nzi3Oh|dV7Vvu?MZN7jd1?UgXm=Gb*%xP-xH8v6ihIo@b8J^a zdM=%s0qmz9?*I;cYps~;WOu7B^He3hWC2I~sz6>MU#=MY%Vig`o%BRtCg5E}Udlqg zaF&{Z`yw9dN-`6DGNGYZr;v$HJ!%JAKE9he`EvOZA(zffBU|!2*+M2TBX&jcn$Rh* zs*-@!lytuADZL>)EifB16fSZz`J9l6OD=i5BXEhedHgOw&-NMnQg5FHbHXJj%1y-n11zzu8+;NCa1M=nX#;Iw#BTVvK#_&t)e9GoTkjR-*9nteV86^LfN#E_nE; znMfxGUZ*&jBz>!rAqQLWntaSQPH;YccWsJaQ@kl3`Jf?m3-srny(Ej7EAl4}E5^`? z!}ZdUyeA3J>XOX7QY3SMy17%J9_w|=4)2wGPQ2|~v0jyU$sx|)m%Jor9G-V^Sv&*O zqCSgfiNER43%x2h7i+WcOm-r4OJ`AwICBH+hT`lsm(2<`J|F8417W$C!;J8FpV$+6 zu9)+&KGF+`P*V&YZy9XqcT_lrGMeY{TsH<6u4C*Mf|W@7Jc z+sg+g$;Z97?~6MwoOlB9ToSvp%=OZpI7sF}&!m%=`*L2%zC+bdm;vB=4MtW{Brn!pQL zE7o@M2aZeP^R4_$I=s-UeJ=`AKlD&67ZQQ93O*Olj-1&`K57?|04_f#ISau2%$`q< z=hmJvlMj9(ziXBCB@=vJ&lzz^KtJ!|-MP}|WE0EVe(|4Ht&lIlb4Z+3xVYEc&#A*1 zG6O#kXQGRb4}0|F_|7HprPl-}FgrbX#xli-GY5GUp&{T0=v~-~Q`^Z$ovKiipE>X}B?x!hVE}oaq%hz3k8N9&fKvlj( z@kFvqb|xJ@HJU<0pay$r2#NGu@tXWjk~+lD@w4AhQz~atlg-SYxO8e!Gm%X$2z6yA z*ph{i2~)-AJg5OL2hy{867u0yN}XQh^Gs5&Q*g53I05MeIEkcFo|g`f??@*Ix$FSN zW(sDg^6-fRmpm3Z^x*tJCUhgU&%#CSG}+V&`H~gcEV=MM)ttmx#GF}NPbQcbZ8R(|M8ek9j(j>CBG^kJQion`$%nHB9Y`w z=*bHE@$%`ub4hqks3@M5I<|AMPUxk4uRh(vszc|*xR+c!&Xc{8hn*vZLa54L$d?Ei z@G(E|ur&wZ^CVHMA%2%YzE|7{^+!wXNvo1*< zXHXLoVO8Z%5Z9NShHUC|3godG(g8i-bxVdENq9A(aB;>^x6H}qqgUeNGef;Daq(r+ zsf91>%SFjdg?#j)&+Pr+bIK8ju^;E-?*vup%+V0CQg6S{nK(81$YE9LF^?Wi`Mji) zZ&<6mL-{zPT>i|>c{jqo6r%@VPB%1u&QOQCF=vBqSj0%UdP~EU!N#ZrdS(%TY=ecC&6jOd89a-Sh94u-% zCtLZ2B(|4qO6H|}nj8Th%lY`fO@zZfspaKEZ@`ltmf~UGvN>Pk@R&Ps>LuBz$3FDU z1s3@LAGI@V7xzYe`cjAdLjGKOI0JmdEUqMcKKr|c znF6yCr(PnN3yJi?$GQ0IrKXd=Q#y+pVQ$S%4riWu=`0t|A^7BPDwiHLIEn%4!{^+o zPw(~~_-qJgAPKjk7_*X}ldF|`#WaI+aevse1V59u$Z}1dhTK^elBzgQ{^YO@bMhsGz8|wM~oT5jIyayQH(mdY$w@} z4;~Vmh3&j*xY#FG&Nz4*DJVaC%M@c?>IA-Q^2wtIb7aJ@#m|WeiEMHcp#VP4$eid7 zD7PvbJ(Hw2J-v(d&@X8zE}4sy2)Sa^LuVd1O`(t-mJ`1O`)KEdJ=J3^`gmXFBgMm} z7WG42nfR!cD9=m6!{$5+N#-wnd`~ASm;EQQK{zYL;fFf%A**nNth93{U8uGF+7q0j zjMJM%oSrNsXP-!t3n%PVdhTTlUb>TH#xA&8A%^1kC+@>dmg$tQzMaxDALoRNzG*87 z^CtLR?2W9cW-jT3FpH19oMct8_B`mI5U)|{iysM>{e)cEEazgJT1j2{MkVAY)?WAv+3YLK?iG>E?D%r&GbhK-Jol8M=p8P^YRx$MKwC*hhHe>K4OVvAygG}lJp^loe1oU`2tt5w%$$L$J~nie6Q^}R6fbf#k-I- z=Zt-vi<*VvoCW>RnaNACa;JZ&L0o#Uy?k1;ZRv&ox7w=7%6*)ZoJ^p8BIyNYY}-zD z=HoNcrCjout0Gie@ma)Q$nS;rIbu5j8SYxLQ@onsgy7TM32~h*Tzu573SKsnS(*aB z<7NW;C!f0L+yQ{cOc`~AM1VuhY50W+0{*0~n2YiDJj~XV-wR>+I0Nqy`OLzcxl?~} z=tm87q!*IXuF8jvA0B>Yqj!6+*tu+Ex4=6ZW|Ig!%^l82J)IN0{F2nn1oVb-n5!ma zUTFfkZ~-gxbLwC_7xN4J0@BNumzb(r>^a2MJKU!@%#urX5fit&uHLOA%#p}f9MPAJ@*cYU1v+(9OPQ!*?UXCub^Nh>}h+;6TJ+(O`N;7k*!$$5DE znjuFD_?(M<=Gj+$1JnCM44775d6{KsGaw7oH{A zoJps^=OZz%7(QyPF*N?J-Ogpt#0uq6mzk@wyCgG7bn=L2B?p-R`5OX@nZ1wq1)sk3 zARm4%9X>qXQTCk(xnj)CDtxrj$tN#!(dSAXnaJ19l@@A}TT?yyg}94%U6n7Ki|o)x z5-xrxomu$VTWK{mUtqy+PWq|K(BMsm+GN-Nbg7ABhV!l&x04zt;Quy*{Xb?{H?`{x+GOZ@|Je+j z5B(25_}bh5!~WJ9oxy+DzkSfm@Xixc?#BlIw=-Dp|2TuewX~)g*7^T`2LD6-*IjQ= z?LX@O|9<}K4M~TX|M>iM)H>@07VP`}uoeHK`?u?)gUx^5|9|BC?K);qO*8zL?te&a z-N9yEy@UU~{_721Z^(bMfBN5g{)eoy&bov74E`^E{xE~(9sF?LSYc4$9F0jQ; z_N4a**t`4ov~Mmx$PSpYTgSTpI;dl%t*RZTZNFQ`FHi2%@%HA2b}ZCf!zQ+7-(Pbn z+asM}f7<;$`{tx~?ekxpV{h2E+n)aL)pq;2W?5Vd8xBD)>qkUkv z9XqalZKsZ--rB6=-VJWB5B)i}TaSFmUbE8QY<}1+_V%N$u`hjdyFFv-+xDcfAKRtp z|GWLm{7=}o*P3phKYmf$d&YZasRzF>+ubq9zFM_*{X;jjum19)eSh3XcH|Wo+m-XV zI;sm!vfJMBPkX~WtJ@2P46)0eRI~Gp_}O0l(08`;f*0%_Pd;QfKIvtcStZFT&3yUoR?+kKW?-FE+>huv((0Q=PoSJ|hIIo#g5%9(cZW3`U@ zLQ8h6zS_W!Z})lJ{_yx`);{I!n46!q<2SpMaI`qmz`fqkYm(Eg?OMt0DB``8OF zT+80I=^*>u2D{o1=G)ZXu*Q!zS^o_??xB{wNxAGMLOnvZLy9O z{mLD4FT6;{Gw00RG3tX~?L~(yXs3_=&HVoHO!M)6f3oA3+s&T%&wcDcKdfs%e|$ds z>OR)qddhQl+^P53?Y@{|d%XXW?cDk`d*^4~?)>*n?X9<+Wcw_>t9|yTv3BU_Q|;r! zN89wNt8H!2U+jQm&$3^yJlRfdK4O2m^=EtUrtxn|ci8{6A{I>?^gd8WPLhG*>eJ3nK4uXewEX~Hnu=lo&zke~PM@pGzPREZ_R_ru*|(SIXD{A*uzhFDGv?!6huR;)A!?OW41j($7x5cuCuQ<{|s4f8O?+U+>qECtCmGVRnlr z$Jh@y`I9|luip0I=la|8fB4G&@cMgp;-JYke|?#bU1u!W@$00OIzBvjgN})}4DHx| zhaEdMJ8wzz?p_O;tKXT&{5oJtdidWiO+Wd~r|+G%tvO=inr6NWE;WNcny2IbhZpGR zx!DKy?gx6B4Ogg{X*a%=zP8h)>0xiZpMLf3%BFtA=jqsQzDUm>b5nZd-b2&P|8+vT z=psv}y{4XF2fRGcZh6O%_K~5lrdyuyOxkl&4>RV}FH_xDdc=rN()|v~($h9s#H=&( zxAd1;qtl(Ij!nP&)7|M~n=Wa#xcTsbmAK?qyzfBnXb3+ ztn{ki=SmNlXG*%#G5ylxznGeS^xlK%u3Zb6+h2S+9XjIv^tCguO@}SGPdfdu^U`-; zzaahkCZ>4oWw_dcO3n@mr=p&V9GJ@rw0q_x;bAUQ0iozP8@G=@SFLP7l5Fi*(o5`k1r# zU)3!7(f8>Shu)I@^Oz~=J5L)k_M_9yvb%j?#{cz+^pes4Os~GUr+MV)1&nYnoEqv<34uTNjN;GMKS^Q-i{r(R5NzF}&5^N>O2%H@Ze#;-S+ z*MEOAz4`R7(hukVHC_C-=Z4%f=>GK0p3~CHuKCEkGXFAmY_C7qYyN7?%|}_Y{_BgI z-9`;e?VtLj*Id7Oy70U+((_MW&78D!-R$t$rRh}+{k_t2d>7 z#8>GLz51GUuYE85@Z8JOk1n2>-u2EM>4+_ENMC+%A=B^fUx(cAUP(&tB?ln#6Q*L2fOKS@vMJlHI>`08fx z$zP_|Ep)Vb`26eAwO6<$y=nN2bXvVg^W7gw-@fOk)Z9Fm`SqTaOl8HZ%-Q!%NY~x* zsdUJP3z)w=`(rxx&^+y0>7n$SHAbX&?cX>3`zIHq&0UtXUp+kCe7k-R`^6^nrGqco zHSKQ9l@9AWPwINTIb{2V7EX5_G%+1_@ZZxF{`_{j^^aerPab<#deE5e^bezZntfLK zCEfkTUZ!=+L1z6woo`-SWp4YAdzUt!@B4fDnOWN`arRf{@O8d7`&_?=J^q0u%)f^8 zFkj9ZZZ^1Y8S~1Nv&$Y@}#g{i*-nf80^R|mk z$IQD;al!fNanGEWZu{Fs>5AKLYu3GPSF__Qqs)PazG$DAKG{Au?pJ%mjos#l?;B?1 z8T*($z8h^`*nDPs_oVS=+GPvb!3*@VTf8{fZnnYh>4abZnvU5uPj`H0v>o~0U+e}; zTy1+z+s_>M-m+$a_s5zauf5o;|HEK&Z;xZm@)r#^2lnhR{}|uLymI|@>2v#Bk$ySm zvGk{z>zYr`-`6ZT@*MO0>yy&opPkRlxcPju*!~}wZJ&9}%ys3;_U4Jpn+qPeD{Xx6 zcslId{mnyz{$g(IbE4U2=qje~x5G@f*FZBWIWIlws9Vz4w|G6hZ@sO|^@prw_B?l# znft+U=Jj8unGpl4_NRq?I%?_n)9c-g^t`S5r`IlZWIA@wThfnyILF*EabdIH$iZgY ziM!kWAM9@b@Zw&!->3a;e#H`Y<;_!j=|=0DYwjClcDrj!v(%Z>%~{X?Zl)ajjM;I$ zTTG+U*IeJxX}-B)eRIhtyOu47i}F4E(+8*0W59AqY+)Jm^iZV^*GX-zZZ(#6bk zGtV?Ht<~Gkcf^P0jW<`ci!Jy;$Z-|A3vM)Q30%@W_5EspAKI$r)go%-l%X2nyMGwU3-xml(Er|Bj0A8U5$y}kM2 z55K3+oVg+FcJ=!Cqfm&pvt6Q1L>~^E3+IL@h z!v6if`R%fIENlBbxRTv_l@081!^W6CJ6vX#9{sO0+4!9FvWKqF-5p})GXu@J*=V!Z zha=1>bFXb4J!E}z;AaQhujXFH9&qZ_X29`3m@D0u=A&P>vUhY2G#5Q}wz0o&Z0;Pn zq_LaqW)|MyX!GWYGt*>~^V5qy9GiAt)M2N8|GfG6#}({ZC$`eB=NX>1*4;Z@{ltUQ zFYmh~-8Ng?++G=BPBlNIm%M+ldF%SWr6;fbM*7je<}zCiOU>ldmoR@_du?;uoda!; zt9sl0?|jW1@Z$1jhvkkmt)o{mJ1xAY>3PR!^XQ~?%<{vQG#_m8d3wSsYne;amCcDS zZf5qpZ9g+(vlY$Ue_YY5^>b=I*?1u{VbbpA=O@ORBmQu+`Sk3C>>(dqZC0G?_;jxo z_Dx@T=lb-~M-MXJ_u10?zF20C+HyaWKmAeq?JduybFF`{S?2FQr$ zevH}oo@2~)FTY{F`*OPZb^jI2hD+~ko~`a?UO(y9^s`@1Pp2;SYI^3<7o=;id`f!Z zvtOtEuIe!EnzhWcyF6hg{5qdGqcOKRaEsyQ&nM4oS3K!hyJPcXbIcb5&5r%HFxMY( znlU|YGe=f_HGAy)irMni1?&qwCzu@;xxkD+bbYhPsPoOL?_}oFyWdD(n7FMOcI*}A zp}DWL^PF;pz3t%Z?5u@XvP-@)kNs(>Bkc;wU~~7s-c4WIJTYHxf2OsSBkb`bx3&jw zkehd3IMclR+aqSNpD#DZ>_3nFN56CI8F%ez4nKVxbMzt`n(2$2W_~kUncuJ4#=P@f zZ}ZLTYnokN9&QFK_>sB)mnH1i=O1VHKj_cqnQiwnhrhLjIsN*Z?Jb*KXR}3Kvl~3} zjd|&y-^~@%ZZj8rwV(Ov&u5tv54zthy5U9Umbl%&Y0{_Pc&-*zpGrxBVyHVjel-0rQG?_OmS)GMjF& zk?H;8ie}C0{%EJJwTL}@o(( z&ADBZ%?qa=Z~M%;-LA6fS>}p?+nR~joN3m}uQNyQ^1ivF?*{h7i@!F9?`ZAf?`&*$ zIqVM8Z}->D%(F+CFIN01-R}0C%>hSDHj90@lbx~M>h`4tCfS3Z+u7W|;)~|x;oq52 z_kNvzk?v-e-@Vg3z0z8y`@<{Kk9WE?z5TNz&9RrQWUhJUMDyUS#*DaQ1@q}gct@CEO)-flW$39%uv?|A$S;JQ|Yv#Sow}-E6cH8e0^TifVnR9p8(C+ZfL+1Y5 zN1GKk%1zI$FE-}0oz1jkZ!jnR`6_e$1}Eyf)plm_Q@1r6U%sOG;LDxN11k))J@q;H zLp^7hU3TkZ54~Y^TYEmS-<_4%doSG4PF!Q8`FQ-IW_XWfO|4&Iu0G}x^U@_d z+NUNgX)o%3seSbFtL#PHL+lxq-uA~gwzEfWIM56?%b5K(T+$>XHZ?oEKhi9F)(Yma z{^yw6Z(Y*-HeSB5e>5Ej_cMuMJ>w2@%QuVL z5hwr2m`6@FzuwW$9yRwW_SP>4*`2yCH=8eanaOtRYv12xoY}6|Zsvu9?los`e41JE z9~YW+*FW8Se8q?9fp>nCrppd6yN|ll+}L@B`K}!T!K9^bcjz!HCe^}nSgT~u+Zn)CkQ`y$My4Rv+;zd2oFMf&{@XRr0 z$=`a|M^^pPoORYc=Apqy+kI9$%dEV1KeO;QCz`iTywUu8z(wZsbAB>c{pEan)AhaV zmE#B5zDMQe#jQRxgTLL)wyx}9|2TYgd*9C!?N|Ce?SiJ>zXvb2A3Qt3Ty@#kW|_(g zcIKN4+7+iPV0+!TkX>W%YwWWJ-(b7;yvQEB>CN`-m!7m=ZF;xeXp580H{*{qx4v|V z>35&AE03?+3$ES5p72&bd)CB7Y-^i=cE2$<*bk@NWVb)<5qrR>Y3BXwE;r|le!`q^ zPcOUZtVQjVuV%y}!mU`D)mp;_tkb?uZBGCTaeYwW{UtZ3&O z`iWU{i}me0hu&`%K6kn~^_{Wi!Cg1911D^6PwRJrJ+}3Ly>`J5>@i!;+p$aYA~WNS ziRP!fZZ$myEMR-?v7){4$|Y^@3FA$l8TXqDzC6zKd2bne+SYw+)Ah3x`)*|~Ty1~b z`r;BhWXc})=D9Pw?ysxc9qu~UE@aQPbKP~E?e*nL=FT^qJ#(F9>`|&cZSOAg$i}aj z+}G_~yDw?K?K#+f-}6uQ=;Q9SpMEvN-t_n*}9M`l5U3HLIw*O(~g~uNPb_3cit z4YB7u`iSYd`h(`~>nE6nR$0WJw)EU~*q*D{@Ao;>eEGnh=EbhDrdRbov-_64?VY3h z+U5V4+U~11uzx?V!#0ln**x&(0(SDJ%i9CCnWmrjKbRd4O6>d(^{{=1oMARTZh5=K z@TKkZKRjVB8GL{}<%H$zZJ(WN-(BxYd&zmb+ezOYYQLR(oIPsM8%?iOE-;UcdDQsV z7q$b|-rn{;^-}w{WyaW#HhIdY<*@0epQ{KCwzi`p&ox$H*s?rg6;^B%MD4lVO}a+i7Q(C5q>yKP}758m7Ev_s9V zcAc>UPpR7_E34Z>r_}AW_l>k0thb+?u=#Gb$Dw=MThrsLU2R8u-wAuz?Dv<=Ju|*F z^DH;F-Raxs%xb&eW^O#RWqOS{!A!5+VV3*iY_sTQi`y#(Z)RUeH?;5Wc9rS<`BUZ) zf2+B0%1-u(UdPyP7C6vuztF>G@V+hc_~56_zT0>^eecQkqw1sffmbJ(51zT&d^_%P zv-ef=+Kn$*(r)omFFW7Wr`X?%qwOC~y39^aelmYqb{_lUx{sUB7wTzWp0_aPfB6sd zW9OoF!@tgBA0NApJ!7v`?XU@Z*)#S$-)`BtzU}e&%68Q7Ywh;mJd_?e&xz@E`kr*r zCwHg6UOJ!owt0>@ZSYH}-Rqik^WAnzNAA0A`p9K-rLPU!Iem7*jb{2bFPo3=I>W5A z@Wn&k>RKj!>+1n&`fxAvhcUy?{eMw# z-tk<1Zx~0&CM#rTg-CeLeVs^A%18;7(Nvn6q|zW2MOIR&v=vI*Oemv0l(a=f+L3R& ze$VgU&mW)H^PJE5oO9pTb-iyk`-TnkksrsZyecuTT$aMhvotB_-Ypt(w1kx6`_h2?K#G_ePm3l!U=IW>w(`zycIxpHrln^|vn!U+B9|Jv z{9g}S7k-C5ljWqYX3a(!z7}{}r!YKZ13Tn&k!fzrV-ri%#U7&o2Q=jPhbfnZ;+`TQ zN=uhTDsN*agYUAIE3vFZPk~vxsW9^j33g6)IbENlNOD*1GQ*q4D9#|8sxMn2PPDWg zE{z{AzV)1+$YjcpXt4eetvyNp;l`vJEaa49spDxRXu^*FPvhzpesU;|bMY`@`M3hS(}-MST%U)n>2-Mn_m*+hy$Sxol=U zHAT!vRde>XfOFugb{d|OZ8y#j(rMIyfrwaCP#XnZD;2MkZGGj&3a;$F9 zUt#LpJ#0;{*KEoKNz%xBClus$34>LR3&A>0)VREfqWaw;$3cVG{GrE~VUHIZ6LOXI zdUaFZylVQk{e{pj0rqQ+8q?Q!%Vc$>XnKPl>6S2DXzQY960$UYy%ZV0%V*!mY+^0Z zy{I{*njQX?%Phv5(6aC!Y{#8sIzIn6GaIH!O23{^n(R)J&|X7Qojb_%$}9{n@kaiN zjnFd?YqE2`n3rdar3?Dtjg39tHugb!nKsYeCdVg-%W-Zj#r?Gu_%mly?p78@I|qi* z{F6zvHvc0@JXA&2;26{{biv;1zHl&JfaCXX<5H?NUoid?YUjsN!QW_F(3DJT`$^({ znj&70RYByNMp`}V3EAFK!h#c9u&D4D6jk!kD?F3}4xJ&lrj690p~2^TH{w>Q7TnBx zBxbw%Vp#GlRBm#@%T8~2^KfkFxkgj(SI}GWe$A=U=1<97t*R3!eAs|q z1q%GzCGq*w^||Eju}Ju9kIGZg(6rfvjQm_wUOkW1?ncN@HA75-1Lk(P(FXOUGU?FJJi#@pZ)P-a{&I$_ra!Tw_q*2 z!oU~uTyx$N>iDdQs<}%caa9Seni4oMUKV~g2V-rsE$VyC#hFhbWHbIKecO{uQ+hq5 z@i`JOmX^U3g%8+$Oo#tiqRfx9{-K75=X80F8cgg#HLOwcmxPBc-q& z-c6=2Owk$;2LDki=&E#reTg+3TjjY^s0zRJUW?o86>uwR6;57@z@oc*X_Nm&+9h|B z)~frHfms}lkV>N9br5XhoPnLB0YAiq(%P~PWSTIdfFu(i)4C1UNOsB>TB_Y95e0`{s&Pc!3S=U=NFtqp|;=< zeyTL!8tK9P>n5Zh+78X*8%a%fDLvYsOra{TNk{V>HFn%5LF{j7);d!DQX|%Vopf$x zFZ{R744dvd!|=v5JdvA?mgiS#lA8wlb0bv$t)^oKCGh#4B0l$(g~G!@IJJHx4&CaZ zhx__KKS3X5A^TC)lYxKTmoO}U6KwMK!BB1==EcO&uhF75It&rWl1`==Y(5W*`bnNF~WcFQ5*JH+#+n8L+xH*ZYrX^5M-cEX{Y=WtK zqVYB(6wmJZlf<(v)MsBdJNFT(zk2b+(kwjFoQGgXUpyZ+kw&!JP+Ofm zMQko0l^+YKJ!3ZAa5vz82O08=kak>KB8OQm_sJ#W5$*9;!|?4&czfCimXh_h#(m4f6d`P(w`nsdNozGGq))|veK`I-aigIn zW|TLds4m(BGb=Q4rbrJutc8Xdl#*)63L?qTv}V6P#dmhI{6c#u-OxdLzsF=KW(usH z2I5J#(Hl3AZrcdDSLRI_st~nx1+qS%jal+4Xx=Z6<=5v^V3ZYo zm|#Lf#JoiL-%c7c>ZUB@w z^g(+~AFQ=ZN1LWIrk^p#pSUmhIr%vLh2-PjuRf?A^@}!5Y@;XRR8Xc}Kr0^KqDkd3 z^sR3s1rALi`9!sHPYiGi#aW>jr+z=8tfIi$dp~TqF zOi{^~M#mY@+u;gS8Q%vRHY?(npQwGexY4-2ZnP@ZoYXfwBwBTVP9|DYW&h{2y8bT7 zdL>dzoHpk955!eRMQBZP#`|jyus1V7Q@atYzS^U9m=slo zsp>wZ@4QF59KGN+9mxAn4}(mVpl{SpD|N(PFlZbU7WNj`s6X6g*35e0=_9fJ{xME$;UV7jClx)}9;a2QtIyy)Xvi2nv~~}{raL2XOJXi* zFWpQAVy>d)YAU7`9)i)?olxMl6c;Oj#ZO$&BIiVzY9}c4wDIjvdj6gxSM+3E z-(MiD*B9xiSlbWv7)33AC(xY(qbSI$FRCN;@ulEBeQHsF>#XUx(JJaG%AeV&K6NZK zRgy0J(?j$5vA8pLAbj3!MN7{m2t5aIF6b9Cc=DQQ=PA>{wqX>jxRAU$v&c)UoP`c+ zXU_xQvweqCsQdap8nx^$MW0+D-qp!aJ2@VQrsq=GglanX#1T0QCR0Vi1G3cfqh-0r zN$u=Qx-qkhsx?(<+ua#->)3HhOl^9ZR0R}S5xT0(&{8oI z@3-EjNxKW^tm6)P{z!&$9(1vf>-*7*+=DDIuafoomB_+GuTSb>Jx$+nkgkf}*y?#< zwAN6QHf+64`y%&JRfQY++Fqb7at~-n*+)|CFc<%WuUYMb_iS2-JnipiNc-LG$ozr} zJf#!y^4tvQ(>Hn``WVa3ex)#ixK z`#$44dvsNTW}3Za|E+0dZqqDja9RwltYZ|J^_KlT{+s1kM$n}099sM92#sFQNqJ%> zX348qS|oaOGV&g@CSn7vDX1fD;wi^ z-vRJV>m&A?6-+NfAh%CW2$s-9UTrYKhN_T#^d43`^)H*+>Pb4M_fR&EqvlJOY3V2v z+u*Kh7t7>5W9N^ElQNdqIWS^@+F_{ zZ$3n6t}$30{+fEt_(lz9uad2{9vb)mqvvjBuv>hAjH}O));(!_nkd7+7?h!3hZJv& zROh`r^||kOHQt)=obi8(l#<$;T*CVyKcp{gcqjryt=21I7|yx)LwfxQoD3&^r?CEVCs15Np=@5!t>Hq zENpy<(jWa{ZfXIkc`>LiQigm}9EN|30uKqm$`fj^y{U_ghFIubTP*$#`yu$M(S#i% z>C>8hG&$!2?Y&t>J>~alN9YX9FU-Wyp=aQANe&H}_Sk9Wg1UbFsI9MH ziN9!GxdwK0>0y0k2j$j0rJUS1bloZ&0aA(RZaIZR+L>fg&_j8#1?0$wBCvQgI!h+w z($;1v2CNbNA57Uuaea2@r`Z1q6y_k&M3Ju2HHC&P_lTQeSNlsW>hu7YelJe zF~7plj0D0LPbSN9wFMWi5s1bRJKK)&uLaQ5}kxJHiig>Msvllh# zQnxC}1RGPgx;36;#=|%-8>dfs!e+b?;>ERZu6#jv#_BWxgcXx^Y+9lFcxk`;Ud&7k|j1e1f^VAZSx?;c5Pe|nIXy2v2t@gREo zQjLO!7?T34CNs}&dhO6jZr+M?xw@N8{v$)vPJ3aev<}uRHpON6#q>|vm$F+D$vdns zesl42t+lZ3v7*nNinMZuA?-Q6lpr*8D(n}bEci#r!WYJOhCdJUG#-+4N)Ftch zJ*Kt8Z&E;8At{O4SfkxcTwF5)=8N89w_zJr$w~43y8q#3i4srx_!PE1{keOAHJ>s{ z)Q4&x;Cege$HKK#S-j zTptmEw?pH3TUjV~inQiKZmRMn-*mZqzi;?j&-jw=!Fb0O5`S~1;D@UxdP{%8lTSaf+44J- z_NVfi((OF|Od?luC;nib5uZNLjL$nQ!3&+Fx%>4OSncu~wyVD3`nnfzSS-nx^!kLx zoJZJOcorY0XT!^4FLGLA`B~X{ylqwpA01qbKJWIz=}R*Dlx^e<(|q_VtD$_=3=N*v zAjOZ!eTRp|XB6vpBV^e&BzU~UrW-dgX<7{ixyA8Q_NhGY%4{xC{{dxhTXEy=eI&0x zgP_79SQ%%bNfjvh8wiJ1pxIlV4_1-mW=mT!({eFR$He2b!hAe#N$1OSH*<-bb9wCG zT1alZ3X6X^m~EX5wsjSH{>_4O#V(|OJC9_y3^-34$fw>2;A)@Uc%()o-yi15JJ!tM ztDKX$rdJN%)HjJAKfaXfJ=@A3ubj@EHw;1L_HjrYIT)eCv(Z=jI5G>iLL)c{@sTsp zlgA->CXiS9jpaMP>+z@A5_}k>`C#kk(DuE80{v{XZ`q4p4W(%MbsX!{ijb@!!H-Ew z^SaAF@nuH<|9ZloAJKN>sy}V`mOhqz^e1J0ZM!@_GD3yF3{&SVn!)&0F%4%%&ce}R z7yk6c0KP4@FNd&%w~koAzk1H(U#vX&uxJY|eOHMee(Mio+woZ4H4-h@g7fk_vAQk= zFK!iL#o#>TP$s@PXCcpXE1Y>6t~K8VE33g8QKKq6dKoq!DzV$G3bJqh;A>Yee#KOh z4>MBaOQFd3%zcjtG4CDdtIEgyc#7%KYW#HVBNUXD;%Db&_?cWpx&Ju?&KrxhyC;f! z+!0!5^tr@wIi9)bIfmREhJ8;x5m^#|!4mrXJ6ZBe-7>sedIyT17o*-R5hfl>IR80? z+g=>XU&+VgoV`CXu8xM+xoV7>dH~f1r!f{T{C5}Qp-BVy>9E}hTAhX6-!pI~E(xJ~ z((!%f9+V}8@_5e}{(j16ZoD~#&+$#?se(VR9$m?oo!G{0mSyrk+U00qm(b*>#bpjx zQLfByF?%)yecV*=^tlz54P6TVaS1qo{UAE$>vO*!HoWK7Fm5?_J+D>QMoPXXzU5m& zV%2~6Y4rDu$6H!`$`!AwVz9?bhO6wnj*|}~`0vIa^xsE0Z0qYL_HsX{p-2e> z*(%6ayCCiR5)2)mghi6iY5A_%DC6m9n)n~x)$NB-lH(B8UlV&D=Al^eK8%;&L&Af< z_y9>h`fw_G6u09`wj`H*q=!Bweqx^+kM@75@D+8?DF(xN^2R*6CuYFZl%o;e;)Jo< zJCQr}AC%9X!SuW)Xv{i`;ai6zFn=p9o_L2qjd*(a{Rio_D&SiCAG%uKP4ic6z;UrJ zNuHp@RX_cJ!-EbOA1}wcr8?aAiVP;_|Dxe$PM9Fuf^I1(zP!Pd*DEB#N2v+p8U-FS zr;;23A5;6pK`6-Xk3P%^Q!KV&<&d{jKk*HXwC;`PlJ}|Wf-2q*4hH#WAicW*|Jlg# z#rc`IyR$b>T-74hR$u7v#D2J(G!@Iz<)9lf849_7iA9k^V|+g8b%^G@GU+^KE%f>1NgcZcj;n2gM|4$bWhdA&Xidg zzjp{i=H^jjolI)-dYmNEwD~8-Mbs^6l971PYu2V^{69%VA z!0~AUE_ygXuT9h+C+pzPPFZx%TnMR6#WRY)r*$tDFHHY81Ib>!@T<9$=-wxKI9eG~=cJN5q9LPpRs|C~W^84~62b;y!f8wQ5l-X*R{6b?5Q-TR9euK1oB*J0NC@E1Wd{ zqxSolivkSDyS`yned0+!eB(}SosGU0M4%GrTsfps|JZo)Hv zR?wEyl2{vQkBo7DsmD7OX62Xh?Smx79Z*AbgfCwF*n*DWw}=oG?8ZIUyp>f9^*6IEzkGe*o>i^Mjz^QUipydeFxAY#~bwMM>5;JZ5~tY459ZvZ-m9= zH`#45%Y60A2sEb;p#^_;!O!O;Q7Od8!{eMN)q$7=KlvFc|%x}!?d_1dd z{>Bd7un`8Cc(a~ZCE9K=lKy^ipwti}y8B9+eVlNUeM{`7d7i2?{>TwpG~EFHz2e!` zhB)HibWmG#o5{_Iqavw%im*y#GpB2k)SJn8l9(n8GM&ImB;)A7KY5nqV$ZI4kD~4Q zp5)o!Mf*H-==U5~_Uz<18hKC|vCeTq{KG!%Qc@ek;XkbE#Zz`~h!>qKeaFH&`w=_t zLGg(xwA^7RN{1~*KzAE3IY!=rHSTxPj?BNoN-|qUrG24)!6#k+zMF!P^um zeAt;xsWTkW@HXE6lv7N!`2Go&t{z9x z_gz`|tT#+P`Zb-aJW;*RkOVb>-@W`1Qpo%@^5o?lwY zj_$j_T6%dg^ZZ)2Ve3ZPUFF5*+}^=%Dy$~?%{I)-#Fc%mAd(uS!kjacm=NntRRhCV zQ+EJ!JfFwPN@ugS7bEDr`6Ei46T#Nig|HOo`)taZ12p&i0BjDSG1_zB!wPEnm#ihFxNMfx0X=#GGEZCed8|{$%-R8DjEAk&0z5T{+S~ z(wi6H=HgLqMa8rjci8yiiWZwqYRn*3OgF`QI4AZwzAg| zU)X}&LqbpJ2zKb&4k}H!$5vX2+*;rDq;MpI^;Py3_ewGyiqWCnu@02{Du^0KFt&59 zB`Z66kwr!MQ^4BqZ2HR8WZ0d_KE;}_Ph(=3b^m^>@OJ`Rwb78AdoN;jg(H~vTrDbo zBE#yBUKP2Y7h-=uh&`BglwH?Y&W8L;7xq8gDP%l*&juNL3(wEoWzO#o(2)_k^sTa% zMJ;xt_#HvQXTb?NV-cuz!ic6*L!y~D>uOro321&vjjZ1E&F`j|C^$#)Q02Z-63op+f++HAHqT#1%v+-3H6 z{3ue|n%Yt-g$a8{u>%vuJf4yQ>)&&hIkeW&O0&D{{Dvx)Z0AH3F}sBGMFLyx@`lY= z%IZbSh&OO$)e7_KGS1sJmi6D!$@>ISwfYP4Ell7sMY==)H zTU9!Uylj$$nCoL$eBLnnFa5Q^O5X@|ePqzNorA)kK?wt<1>Tl;U+~v9^y} zg!rU(VQJTL(eFLM4vB0bH0x<#Rw8R&HJgIRRnTv}Gt9uJH?4WOnS5z2gH*)$B_nHi|r8@?pW%?xXPXb0S4czGS_!yI93}UFs`0Mo{dm z6I}c3VVUD!J5TY~VY%|*bbt9a@j1rRntUB{h>H|P_%K!!n@WLx=Y_tpe+5@@9cHx| zvX0bWOsYviH^Bo~FxEM$KV?f9S=Q-%+Nz;=ydlA}#6}GMlc%$&l@<5Snz&iDsUZ zrj|ttWMDLiyjO(_skMbd)WBG>lx|>2ca~FP$Ux#>TAdd}9Tw_khEcHAM)o%_fs}vk zV3tFzgsz3OstV`PnOImF7qGDQqw`x|uMCcx@rwM3#~D4{|oIWBsRt z!j9(C*t#%gHT5L3xvmEr>2>VWh4FOTF_6`K31gY7z^d!SoZ{SOrfhd!Tth9&S?WcH z6PA+mLS5{e{fg=>s>#(;f`$(Bp*uAt^e)7VncOj8^5GWju!k!9`&W;tCN&G?Q_X1m z%{gSN?n-GN9pJOT0}G#uxrMFDEIi4b?aG|UfEGh(kmeH8gI}F_ihoEbc980)R zB810$6ml*U(wgThY5Ie^Gc&7vo7sB!Zuw)R`s?IJGi4p%!*vabL9hQ+dq)0%f_(T z`W|d+oIk5x=fu9OOJN6J*9o<0UhKA^Cz~!;#O@S5XFIpgr=qx6N=_P0x9&yJ&JUf! z#*ZI_{qMSkS(9qm-O~rz!$1pC>kMGhihJ3dP4`&P=eI(i>Al!|H)2m`onWa2dsyJ& zH0Iu8fP$49a7}$G&W*UsevU0-l9r_`Smv&vo!2ec6)Lh|F=ym4(FrmdTIh4pkd2On znD<=9g7oXzmozncxX_+J0R~a{&JR*3kux40`rq9o_PJBE-J_BTRlH$vmVO z^Hs29)Afk$4j;vg7HTsqn{iC%O$r;iES3$vb&o0c?!(S}4`eGIW;1=C6cQa=ia(S< zSL`>kjqcHG*{CJ#R_jytD)t%6aL8vTs;0w6Y7877nPHpOMe@%3K+f6bNbjA%zHS-8 zbh3R}XRlE#AtjQ1w41|bZIfkN-vzMQFMZg|5yP3M%MP~nuOB;NCC{3JwV8|tV|LBu z!tOF%W|pSNUS1XNttq{kZAmX?boUS&nV8Pbd-S5$2A^5u#Xszo^b1z>NS`%*mu2Jn zFJ#6Njl#>hPX+S>#%%b7hjhNUixds>DgBlzLVmQ8law_wiz8U&oTbdJJd zcT$5}5k1ZyjB!el$Z86}^tQ8t-;cwBmZb_icEp$k=nZF1!=spo$VN-(x3Rl_@37lZ zXPKViA$IOV8C#g8&Hnq*i;ajK!-BrXkYZN{srcs6`hF>_Kw$&xyFZW}J)q7`hd45+ z+v@Dci)!+1O<6$Y<`bju21(~TR#i`;>1 zPJk(Ut+_=@<7|<#z>FrRucIhT9ERNO7HEVe|-)Tjqjl~B7^QY-hgaQo}-v!29$ooQoMWYL>}+~*?az> zMN5Cutv4di<9Unnr1j93T^IXVdCHd?M3bC`k=5@6DmpreG(7i+Jmf_>JManJTeO1| z#B9Wdbz{i=r75YqZKTz4dr3C)5vg2yNXA=K#eHN-7os%j=8j0(?~%{Q@fNdh{D19p zl`UA?k48K+qWUultoEiErT$z*rW-gJ-|LHu#qxL-qE4;#CY1bZBt2f$LHRTCX-%`p zYTT$Gtz>yjy3s?@`QMpq$5cAwv4nc4idyYzN$y`08J~Bf3odHp7imCa2c2gBCTfsZ ziynQ-c*Gur9H7(I9EV31(I$}voiNfIaeKnq+uJvp{+uorG^U(=wUVdMtC3`L)P{bK zwV?O$M&fhLV8XaGR;-`E&R;VZGl3WBgibIrYcfgoTRoY4sUp3>i)lsuBYJG`j3Q19 zp(vdxRFOlZ@w=L$3O>@S2RG=dcMelHmdzs8E7B~B?=0=xdp7_5I!ZSWW2?nXc<`Qe zRAju5O%8a-#&}E8@oS3I)c+nE8}yx>KXaIk$nIf*o$D#K=@HxQqDOxYceCwltZCDh zC^B@Ggvp9xR<&4~%xCnXPCpqs-p7c9DKp3_bt=6&dw_~t4$>(D89Kb?EGv82n>OWN zX2WWpu$-k)^u{@yMwsj)xuLJ=n7a|B@6n+AttE89HHQVaZ(xREe|c|Q3i%ixrx={4 zmq|C7(LDn)b622AdxlZzRxNs#C$b8{IcgbYiy2bCXt!x4z4EQ1$MzLeG3z`jpFBXv zL#|Pm>>`$5T+JRy>XEJDCelyOr!nVy!(!qC*1`{vz3&Hl{2+z(+tk4_%!}wkj5fVF z_Ldp<^e5K|E$qPqMi;J~rEjJ0*-g*Ql&fuvz&5eBEtpIzmOP-azqi>cG3VmiZw}cl z*h0z26UZ$lnJo7=Qu(M%YH`h`0rF;)by}TngpQ{xI-l6PVu5x}m`logFHrZk6e{hl zk5?iC7qM0vJ=N-%F)E07>O8s^H<~V(JYi{ur^(HOw>$#ZK=PX5o+#E^jCoj{% zhdXJanFN28)QeX>kmk$o9{{^}9QIO|kvw$~!cs#}cUH`~F1O<+_-OuPkRNyT+`^;% zGr9M!MBYBb8Vi#wVJY7q+j`q@qvyn%FZJUAX?I|Az7ZKM@5Nb2+I;j+1-_xdkS~}L zje&VVnEQAR?mDeR{cw?!FHM5o*92%CoC(b-@p#ty5V7~#F}|c4|3v@5DJvIRk8;KN zPt$n$tg&3*HiTy^e=GXzUHCBh8!|Wc=G#TTY){KS?4Pt3<<+|}ygLIeSJHUzV`<#A za2;Rg{~L9)CHd-$()_1mf8P4qiqDKt;lI+~LF(B_1Wc^LkZLpju*rgJc$o0I&^UDT z6*bp`YcS(cKE57IfZz80n7K}sn@`u{BYS%B+0IV<_tv5OakUQLXyXE<_hzV=I1yux z>G3JO?6|vyBY&7Qo`>-z{K;x@U7oohVH-hF5H%CYSZoL=z)ibs)K9QQr{+Ms*=7uD z)k}Dt(Ta?%m1uXE#OrgGaktBXT;@j?)Rt-R@(rI+Tm2T^<=>HC(FW_t>u`Mk1T!nI zfCpTJ){Yj;d{~N<8EM%4ej;vk$HIP3A#Aw^I>T!PE0+Y#Ej4fRqPIQ_X0>I?7VU*SpgJTF4n(m~v9&S3sm!;u%cmf&Sm zH421!1aI7rZ#MfxwrdZH+BGpjxgSb84DqT@1DqmnBYfNwygXqoI#moQ>%@6YJdl4G zBI@tVG7JjKb0 zXRop6+UM=~ul-|rzsWH?vuZV$@fyf&l}7V9Kf!xmt71VPeK`G7M!J@m=Z;$jrR@3W z73YjsdKNHSZH)!dnp`hXiO=ZYk9Tc4iG5C|P#Ro;(u^#O**po&ZkafCJsdR)HY4M6 zA|}>6g5_Q0C{ZeSp8y3;6D;5P8f-Z9{$NZ$e&I(e)LgA#vfl_jjTWfx0=B4+ zM@QR2M4bu6G0Q}Zh+7UfrP)weoP;kifw&%?h%}{bh>R46`m`*AeE)+uJ~sxhe46pw z=N!HloxxY{8d&@&Lv!YFRAMjcV{>6pZ~&&`9^z|F6$+fT;f(TroD8qVHuIe*PF2H^ zY8@=NG7#AXmtgw!9bPwI!6+{W4C*n%<%IqSTilGqmG9Ac{~cb=^v7H0)#B^`PbftW z;&%t>a;Xw~-hIj)y|*}tT<0k0E{{VbMdQr2$tNu&yP<+4i zrE{V9$s7E#H%9(68!#AIw7QCaC`>d%*lf1`%sjbMWL)ipI_CGp$xm> zlx%B54?P~xa3yUFzpaWj-CxK#u!$aiG(e?_B&;PAksPUn`_g0Seu@o!+)gyVYdy*B z+C;N_mXPZ3Ov>s!MRTl@XoRB!?hKuQg9#&m#^v;W-Ud2%Hj93grh>OTMgNU;c%-N`N1YKt-DT@3LfaN(58m^UgQvVWdKr@4^r`>GnAxUN--r;q!d4au8VUV zj%+Tbg>UQWwOI=p)pb+h(B3E*SAhE)+$i@+Z!%?$G--k|B%Vq@J}i&s4c(9T`#0jl zssFH094Z?pYMWEc$I`fO3+T+`d}^3;jAiXsnyIn)jK7Rv2YYge{Q^Ty~_IxMarR0#(=_OP*;Rd;twNkmt5O~PXK4jCoCMuDT!2HXXC@63)PEK5mL0w4*GTcplGpebrLyx97M$o&Bi^%zg z9}Rh9Mh9kE(4^ZVX>;vfnn^{JvsL7VDu+<5MZGEq;DjgmKf>BmAx;?UlL0PM)$NbGDA+3^QqS=W+_Ib|=?Izu35;X7u^D z90o^d;n>h*n3V0LPq%K)yA<^>+p4m5z=;tk=vUfD)?bam?@8pFjE9X zg`=f79w(9x;CrVp>~GJ6&-H;YUTB1MrB<}$Fet0HJ(btyQgBx!r8upjSqF}g@vD6# zQ zk5^HWqAh+@exgN^owQg>3BFy2NdC%Zn(M6$e}j0Md*djzMem?9fsV9qs}7~OI8yem zvo!MVeDeJ+GAyR<6e2Q^UWJyTPNoC9b>b|BI~IsI?M;_UcT;APHeCra!POioB+H2N z{%#r5%=K>M(`G<|LGK0DG3_cRRQn{WcU z&E4>M(~aiF63n@~3taaUobKGj&MAc$ZS(}GU$i0RI0h;;5NDA(B4COKroHrnZgdHC z8H~amLts-=9>R2wqckZGa-x@ed*)os`62F46*X>sw-3*E8py}VZpSRS8tAgy@EWxR z{;K=nd;BC$)(*gvrBktDdm@zX59QZSPvq>4EB~>%4JPNyaOH0c4kl-!XL255FBZU` zXCZ1pF8U)23#kB4*WAEK1Mwa0#0>A}fmrv{9%L&ZR@xH&(`RGi(MqWKT!P%=0?ZA* zjf8+!6vs=5KC1z!QwOvp(vjdpW>S;~@!uNeTjA$_5^d;q*= z#=>f5GL}x?0NY0OCKUkl(Zk%pZ8vB)Y*K(_NLe*phAn^~qzes2M0jrJ4c|LKI)gaLt%golz$Dy z4ueUUBVPpVbFmoyCkD@D2J@WtVm&+M4JLk8;p_Kk@&$Ln4=rrNg?F-CIQ$lqg?4gq z>Yyp*|LA72HIB4;W7~*Ggxh?i-wRdHd!8a(luAVXs9dbwpTTJIOz1_WV~j)*UP;(t z<2?riAIQPFBgSa&azGh{;LW)({JWJX=Fp>{m-AArDQYoth6=y_paE+~w_)mD&58;ddH)v&%r0#95duzhC(nzH5jgYB;{-dc%ERZ4RI=6?KYoI1}A zu;S$pZ1~$lu5`}Oh(6lu)A14J_~_UNR^mJxwVH=?{c$55F1ta_S8e&5EMtCQzadu} zsLP*T(BM^e+Wc5Eqf(>4Y@lo(`Y01beV2?Oa|Kb8a-NTHk@?*&@>nu9?dUrEN6cTm zL6>SD@jH9d%4Pr9p$9<}cy2WnrH0eDiY@=6=)A+ZeA_Ti_DEJn$SyLH=AO7I*_2qf)=NjkvImwOJ zX|Yoy>BG#!kehxQs{G#k!S{?ZCS5FDU5}hz@gBg#K$sJ3DJ;ACiyay=kmBk!`7yBw zYv%$xbT^~BQI6*4sF3)7a%9v$TG*U7QrLTJlCZ)e2T2BnNVYl!xwRcwu&fdXoZoUU zuObc58%=vBSdikDOUN1L1nP;X^FUSc<>Jd6h6g}#}6>G@+V@r z_PE+I9*#mKf~PC-I2_5IyM8djr_&h<+&4g}j=BYK!Vl$1|$ zudrP%=5np8%)C4tQ!Rx~Od2+fI)zJL&FIi1Lvk@zqN@+npq^HM_^=#A+sFv}qW)mt z5g8$C-)K~~kB7*X@wn@)MrQ_U(vMa?tJ6rLxO`{&Z?P}k;Ja7t)S2YCe<@Ype#LrT zH8Lf!kL-<~G51-hLC?+r{gokv=S~!x??>~0J5lX#BPuI&qHVJy;GMGz&z5Y#iAU3s z?YWx!ekbGnW@(t_48g|OfvDSUN|ITMv_8a%R#!E%xMBSD?}?D+&q;JRT%2|fHz287 z3N-J47J2)qP=M)WcwbJ(zwAOJ-rY*O-DXi$`&vFPZN(=2a$G2C!1D9s=)1Hcc{H2S z12qlE?$<$NgdUFmR|qw+bNJR?kLArPX%(N(J<&_0@VZ{=D^Y@!hbo>hMz3bfr$_p; zDe=`4{8`Y8)20vbWs#&HGwL%w?otpGItSy)wMnS?nux8NA}Hj@CDJnfLT~?mBKz&# zq*Emih5wp(pWBBu_J^@Jhih?5y-~ZZ9246@$BT3R?QAhdu^GU`)||PO15Tz&m45_s9;i0f}(PEXDx6&p3Vm4X*sq#7uW9sCw+g zP0xq)`KvDSZm)p;%{tP0Gzb>s$71V}d~)Hu*NM(m+&Nhbsr0k>YWosTPR&N)*dGp^kN`wq zZlxy|RpGxUgO2n0x#wL8T)WXn!553@!DSR9*c=J=dp0t0hE3@ zL+Z~D)BalS+o{-w7S|MEG7>&pEkvJ=*iI2wi(0Q>9SBK1|tyntgC z^SaS0hI%K3)5!(rsBGnYoRdg^jKl#nTK5yxJfIbCb#S?CJDuPktE(zWVZtejt~yEG zdw!8B*V2zr8X&}+eSkqS)tc`}|@xzlj% zCG@d>C+*-Yw;@vh=<^9>NOii=FvVRoC{@V=REmo*nyB@6*x&NT*K4L$ZL247>X51>I z_-&$IfdG)WfbW9YDo_vDJ-|-&M`7iA?vPORT zFzoe~M#^kMH2F_QZ^>}PtIkHqBwsYHR>1A;oa6OvH451b{=N3-+B6-PP9%`uu_I*E z`Ib`5O6g@ef6WUoD4Vl1MO-&iLf$co-}jkz{56Ng%4yKe8cwlWqc|g?m}dMw3a3QQ z^Y7e;ZOSL;)$>&Pv@M4|b)Utcmv?Zo?=N#q}=uS@w@I z?u~-p^*|V|E}`QgEwrM)fh;?0A@af$E1uh8pXMf1zfOUy?s4pC_(fADd*N$WFzVCH zFnIBFlvkfXkr($_`vr4v$a0+9v;?-#M&bQjJIuBo02gO5IIEwdiD|hIo%9fq>(&FK z1`55=(}W>2eS|*7nG)B%D7bzf^<_xFY1=$>iihI!#c%{xM2Crub|TVkD?DYKY4oxa^j;yB z21LEV!xMjCxVI7RwUZG#eHzq?!yr6ajgYxX2oE}jJ3*c#-+GXqdE}6^j*+0{s3^QV zBPH~%i-+E@lTbc+9&0U2u)6ys6sO+BnfHPCbtoKHRdyo6E)zeIjmmTT5LPV(EeUaG zn!Kgkc12`vlR$$veWARBhClwI(3rIp8m@=%!!REwTB7hfX)0D)?886DY_uy` z@>%~#EHQFGRCY5h)VNJKPfKV~c@<{$)j+cGI(|>+Cwp5J6m6&{*TIu*?7Zss43lQT-&q?6B;MMKuHwknby4LzJqgnN)T#u67_56hxchr)Ysp)jl|O@f zOKRXWZwSA4)nM1o`3Vm)Y0aNqB$1v&ElY~%etJ9Yd!I!PyApA3NgHaMaxnGYQ9eJN zfsvLcu)g*JC}f~u#OJ6I$0Y>uh7b4?eg`w6xdtg3DO{OL zFM_GlkI#F=_?|Ez*?S{BAD1;RRb3OpxG!@o_HwDV0P_k$~9byp|Jg$~80$g?zf zxh*CKJK&4iU^tfxNT1+$avO)zQw!j8{4})>eoyA3tg%fjgAP~* zQS(=QS|+&$JMRYJ?`1daRNO*%=S;_rJ5uH{50nIMLdca6=yUes{nHO<*r_(EbKQqI z7x!~N$xc-1xll3RnQp9GMbS&tar?p`n2lA#aq}ZEIJ*z86f$wIN`d>xUT~(=AdDUP zhYK?eu>COi7DTwCyC4i(m8W54+h<-K3>5mx6a{6&R+!h+!|~}|WbDc%djnZo^=1J* zejdSh^N$c`z=6cwa4c%SDKIeU)`7KhIcZtuc-e<5utp&_Ae*z1`K1w|? zh7_j@)Bp==wRfWnvWnDLzLD~G@p}4XE#0o=dYGL#?2vl{v;IG)d6N-7xvS#-3~#zp z_Kqb^n@Srl>$9k;^R_E0B3W;4HT#@a!g9jAS(|1y9Zn3VNiUP>;*%e2rzB?wPMJrT~tPIZ7-pS(nN75%9cS;M}Of&awXRQ-gvDB~$vCid|(8})fN`*8CZtBX6t{;M>m(n?i&qBEGXmi?xc|DIF(+*(rP-rg0b zbfKZ73U}ppHc-xt#J>!ofMy+*{wR{oxsu3495vZv1sj$&`KIl2qUNa1}DPkjc zePV&R$60vD1|~l-g53%0I(7x|vo+@t;Sq`&avH(=IGw7vJVEJMm6tlX#7V zfBeBb);whU-!!s~8{}w%q6FPq26{L}hBPX_vC1?5*Ov4%L#+}Pe|R=oUCJX_2N9Zf zJ&DxYeL3569nn}#5?MT+HcKq0hO_V3D*q~S-m#aI=GD>BNA66kaWd=r3nn_zfsJf4 zWTrWytS=#f87e(ua|Y$JS!=W@E~KB;bv|ZO^7pcuwhSg|?8g>yPxrm!GpYQ*efFSD zgOYw$v7`&x>}TT}S~BPssVCafjXjktA>|qyrj*L2TRGEBmkacA&u#MBI*hG;T+HIq z7nAK7V?JjxBB?t&X}-4@`5(0;wXun$HFFKS`mu~%n)RHW-F24*bq%D153)2cRg$jm z<=WJh=NMlhp_p?V`eeUTWbkL|4Y^H$K{IJa=2JSMbCq1%b172yBPMqJf!*99Bsg}W zwWJGcCZ5KMW);#}FoBdtsZqfl86c~hJ_j78vWDmhR=QrcN`%{+?MtL~$doMY@M>OhB<_~P3c&ZymJfv(m7YVtCodeUv;IVQ6 zT6q1SLt`#K;^@sNNX6yGqqaj{d=B!^x`MYc_WT1`>xV0g=BKH zyF$BcH0sm$1UH?+klClPP(co*-s?S*m}Z1$JN)9ONd$c1I)6iq9##*cs;XF6)h~pF zg$2&ZHj(q!23qvSo;uS_s8ZgER*vMlvc@C|2@B!ZV-z)%wqe%eX3TWC07bXQ$a5@5 zRb??sPTs~m@kGvw9S@^lk(@ulIV64tuzo*`)+VS@f$A`dcu`8#lDTwo?s4kK8%>El z?^)C5O4cS$D8H+XzWhc~4jKs$8wC_3%W^HoMY^F`OD*#=N$`&(yTAmx9le8Auh)P< z)e#c=dV#ZN7}Bi}Dg&UX3XiZ`Y;fw*LS`DTJnFFcscq+BM-$a*RN70bMgb9m`us;3=DXx{kil+a_`p{XL8ks_mkA_jxQAx7o-h!=WAJ~+z zE6kGnu3N<)k@Qxc<#l5a`SAIy-byigKluRKxj%W?MlU#T9!A5-sa-qTquN9-DYolbg<_Ipb)^Ym)`3QMBMwaKKby`7(X zQ&>jf6!zQBlRX-i$3!2VX3Lsu*~&MUX>qX}HcZ)vt_$zkBX23HELW$KgNmu?(^)z( zGl_;;htTheXiAA6PZxiyFvnVA7033liR!nRzFire-&jG~{<|r7+#-sp%tbUw*gTMGpJ!11md(^J6-ns<{G2{~oOz`mW>qnDBx#&V zb#+M;wCgM#QXb4AHqB*IL~B_5$Nh9IrGbtn?ID$o51Ap)8dHhj860X}%xd%zwsBhn z8~$SmX%xn@VQV*2{Q@6~mzhrIqIOgAmRW4vyBs#-%Pjg8AHZ@Cda`+(jq&^O0Vc_H z)7z8oFlV-rM*o{lVeL`~;WN8zsby4iIhQnsWOJ?OdCD6eOW%Cw(esQ1D*Q5=s_)0s zm-Y;*ORuAi>;y?~nM*su%P6mS4i{qSl6-wEJ)6k;C)a2){+7x4qc@p|hZc*z>B4mP zm{LKnBikIUOrh)6QQIv|O4*o9=j;8*L-#A~e3r&}H{oolkuLZ2Tw{_tc_dNfNv;oP zQ^XK6s&dsN#eD}UxQo%#@m%+r?n3t+2U3ffCG8AbOd*eW?J&reQm2K|n}vIsbYMJt z`E&}OlM1B!K9ru!NG0FrN16M$7*;dKmKwGyP+yrBt$LygPn%6_)-qq#Iyi}4vUg*7 zyM394+bXtXKq@81uEgMA+|lc^5LW>It2Q*l!)f9+5zYE>ttIZC9nrIrSa zh+?@FQ7k?CElc++BbU%3nm#<6bjBM}*uhC8d0UT~!%k9DnJo?QH-Xma#c;{KNwMXh zXmpw;y_P2G8?Qk{pTfz%@E3Dc`peXU&#)JtquKDjC(Q7G4XseKr%`$S^w#|tB~Eam zon>j{(>sRjhDgzYJ3NzWKs5RNj;1vG(^Tvp#yR|xsi?(^eoyn|*&!Rq$8r+&WGU0t z)3Izwkp_7No74R;2URVbB?aM#*l=#6KS?* zlBD`5it~QQu6O%UW0*j46*>a#54zX@cpl)#kZQi7d+5fP@K}wEp-`7PM+I3$-j|B|nQ;hr(PYelM79)mq9k zvckw>NIE}%;z-mvo3>kW?ydE1stAs!w8hD^%4$2^F6(EG*N^awuW9sVL_NjC+LHFX zJFMVNEeqO~Oz;1ek!k?%Gbfy;y|+iO-TWNUYg6QVxmm0t|2XRyOY=D2l}T_;7b{bUz5v<57sf~d5V-Fl}|0B&FS#PR1$eEDZGs` z651EF;TNC5?uhY*gYP&j`M3ryPP3q@Z-WQArb3;)Ak4DX5XOvOFZ^gS6$0gd;Ihqn7`g@xne#EPcDb-2OH3$g zyat;lWnp_%g!h^4hg0{t1{LCE>{QxC=9LD#!QYwN!DGHYwB?YHr zD}~s74uaLX3Bt}tbI@+>0{zh=;3QmuR?c0Fo3I}@`MB;pXEv<#6o(CGxXg%}$u*vX zAZjy6Xd7RRU*;M3=c6xNdHx4cY5(DY<`ALdkg8B&KTw!6C>%E)D8cG9L!MUwuKMNi zS!NnEj`^a(*cyL3bYSQ393lt4qU+}=u1!fmm)%;_2KeGDR0O>Ps)Aj*h|u-)JkoZC z;Z^QVeA;v#o2)Lwx+xah^v2?ys42cI6^G}InZi;#2jNB-=bqg>#x;R4XjfQ)#O=YD zEpG?s=!r0WsEEF^zi5Aq1Sa1<1qs;*WG$NvFXJ)LRI^5mh$Ph32|`1tp3omDC-j@H zMb`>v_@CvPyMzUJd2Af&V$GrO>J+};aK(GASWJGaDD0T|7NL83usmp}@N>c!Ea%$h z%q$O~f3LdmAzeZkGg)5<{W?gv;L?c;50o*oLKbT674%ps9QG-z04r}SKdg<{DLP0< z5XF*>T0)_lm~hAB7Fzvfg+q@E(I|TyU7SyGVtyKCzaEPv4^psFem4^M{<8Fa1UCJc zjjS2V@ytb4&~lOyJ}Xw?A>RuwDqWA4%4;w_p&5o%=b#nmjF>TFG3bycj#f!w`OAHX z={ksM$Kvp#z(V-9Oj|H5xrn=Sl!c&_W(?a_3-iZz7^|Rw@1cOqi9@)0WDR;nCLpOd z4SPKjF}BSMM|?TAJZ1@e!#Hot!5j0~a45N2Vp3!&ZXG-Z*Sa{+_BxC$=*GB;hcF#g zh!w*eu+qUDzkP1OJMk=nQun~V_&AE5heIZGDc@s8;n>F{r2L+VcSggYrt^)`6ndyL zK~!*k^%nCQ{zGcd1l)dY%lSlu@V4R|Zd|{IdFKwIdy_B5t{;zlbw?!C4ClI-R*Xq_ z41I5Nj89gu?$P7;W0ZZMo{ruRAzC_%9*vn# zbw_=0tBvP|wQt6-vXKa|^FYCXxro9y*0x@RRAuF8V$UZ^QxU-k^Dh)?s7PKCf7lhd z_w4GcTvUHQjYE*qCZq_8h^FLGuuZnZ#Gpco+rd0ZFg|xJ z>La#eAjG#D!fdZ1hL4lNTHPEBG|UF}XJXV(DXKp7hpCCTu+osx5K|wAC$WU&!=2H;wQk>qR?hkL?GlN)v@~rwpR&o^rPLJK8nv34QPgMy^RXdN1#U`(p+_Wk+Z; z9{ZX)9Iuw>!8OQ;vyjF^CO{Wiv&^x?+W1FU4F@Mrr`#JMsj$Fx(>f3NsG zl26e(*0lbBG3_wXrLt=;=;y*$Br~Il*7SSBWPvJL#3$lSO9!o&d`uasMKrE(AM)?* zfJRR&!UOuqf-_b>EGQ@6+v<44J)hdP1_<4x26Im(z)PI>L!5JSgy(Wp55P93X*9}x z23h>&*(4DP7}u^2r>{ft=#3YumOEqISRceFn3CnjIdtcyD+$b(2JriD(J*&$07wYEIyUX?#bO^-m-bjnbPoUTL$iDUBG13SLi;Uimw(G z2-WI9&h0kX?G+WY4z|+SehbiQ4UC;XlvY@TP;Sn08upIpkHQdg_w}NW+Nv-y+l^;` zKJxv324}7$(?Pcm`sY)`CVw12K^;HX*xAP^LrEE?{QPM!i=xdekuJ_~CE58JxT&ED zlULF>)cld998si=7y8+Pmx-)e>;fwY-NO=8^H^&`IuqUUfbCaRz_2eHVQ0GuUp~mw z>M~s_XdFNq!yiHSUj+1Ru3%>FReIf}4dW9Us291z9=+z-H8Jwk=k$raU64R4wFRCwuWU-xx$`&F-Env9S*39;q9;gNMjh+WpR_Z-?dHH(!z zyvBaSzhf4Y`q=P`ADO4eV(QpDmR{$-qKdO7coX%WQVtG;ezr8ech|6oFJtM^h!^bs z^BU%#B1S7sXH%0Q&m-R4#gaehlG*Y!I`ecE&(7OG<4^n}KhF4(Usa6GEtM=T;UAmu zP>Dtdw9@2F6%@Vp6=en1kxr#7Y`LHN?fXTvro^8l4LO5Z{2xovmnHq=JrpK>0FLU3 zNZffI-?s(QCi`^iDn3MST$5Pt=|nc}%Te}yMF_k5vW7if%{5NzidkE|99d|I(>=ep zEdD9aey+2kDSHj+YPll~o)AReO?mJ1ql;c^*K*yq5?y$u&DqOxG;-;6regbsDVm$p zP0nIXJvI!V-^D=VgfAX(CffBo*);E@1C`&~MU=UoR*aOSNgY09$vp_0_g-Knt%Yn$ zKn~M=SI^94wW#K%3VpewO@nG4vtM(|*p0nv6c?$=cfPLlsXUv$mS3YJ&N9^zsbr?J zdYRoFYkH#+!^S1uU=Dw8v7RMY*|^Y4YaFA)EcCj&&d2&F*}A!p_;AV4vT2v#tySDwO!nLY#GJ#Byy~?|Pn; zd!A5Y^gY&9;R&0xTAz-rYhe#ePBD|gf0(953Cq(dqXD+!F!4};wn-x!S|dg2FYhy{ z1|5Wqw?x?v4-DWl^^SWZh4 z&Qi~eRyvrMM=NqoVe-6+Oem4%=K0)V;vsmVG6V2)oSUENd z2lS`H=igjGh}09-?inZCy-^IgmWvqk@g{~BJ;D{U4h&3f!(@%;Sbygk+_y>zqRs7` z;ddEc4=&(^g%v#PhoEH#L&)r{7%k-ky&1c)U+oiib#y&Z*@DN^$ZpX{~V7Q)&L&vD$LS3(MOYRem#Q6 zx?$$QofvNu2&K16aj^L^e!R@X)d8v4&v$sTEvmTvM;S#);)4AeWg($~&#nKh;J(mc zsL*VnR!S&sl@UHkstbu;QOFw`3tE+lOn$CKoaOrOqJ6l2hhhD&N%&>vju~&ogzRta zC@7Z@JX@mCU7Co;=i-muQn*Sez~zfPV*5Ee+3*^YukAzpj70p9*T$qV zIxxx{gny9}1XW`X;Y)^-F#G*dJhjf(b~q^v1@RDYzHA z9YgKk-?YWz)^ON`-^c48&G;4d4h=S=@pyt77C+&&e*nXP2~S7Z38k4c z1-W=XVbbYi==M5;N5VOr)EWU#dt1EzCX1IMSv0Axk?$QcX_Uc4thiss^OKG7cH(=u zIMhHE4^Tdb2aSn*rq!$3sfTBo9?dW3ddq5LXzSsEj|!Ggxttcq%NWrAg1<3Bp#)xaX-VXqG>Pvg>aOn87{mLqAZr=q=p2myNkE z%J4+%29i8u;hX!Dvp?lwrtyYyZu2}Y`K54lb%1EodQ9KdM}PgB>G}CDbg+0Z*NpH? zmA9|y<{qvi{(K&9x|ZU{_)&%`Wg0y$0e9sOU{e%3z9WCj z8q&3Er&`WA_Wo0ewRbP#^`m&W=*Ob?vk%OFO@y+Y9a6` z1`sbf2s@4|3!CllVB>HlVY0dgTKoRd?Itt4a+XHDv^l))4nXtZ7BU(5p49%{=04F4 zSg>R~V#daBZTVWr&CWuvx;JXFUL(qnYjjUvN1eL}G;KxU#%tW65lYw~nBmkeZB&fW z<}*8UI3BV_Sd|Rc%~HWz3n|#S9-w%O5_)m2ihN6^q2aeLYSxcNZ%;Bx1L7f?>yOjY z^61;pPs)8`5ca--_HKSkSowx7`-I@QFce!dd2KLtDt^i=NA(+9nCf&>$Sp_*SM{G$>QM<)XR7l2VJ*zxq=e2&?oo(whF!x5 z*fqD1rIgl0m`J;SHP8=pC8(4Pg$935Z=Yzxa)~RIHDSNw_d{Eu0ssVOjSes9ZF{HP3dYv12gL**eOmC4Zpm z-)@w%!VXcB9@3#oqUnaQwD;ak68HE<%WrDI=H+`<|7a{7S$Uq0$e!ZPpPxlj5ZaGB?IO8_`B^J)13R)HcB*`2Io7_`t9*Nvt}Kge|MAq zzVD)4P9G^gcMwwU3_xPw1^O~<0C^gHVoJNEDC*KT(hc8FQhj4+(R+WqkFAE|tfLsJ zBTj1_cwX2R1IiCNMi(ic+yc^Qpr0tMbZKG%S0>X(g=@?wE{)#R<YVST>HHnVFK~LsRk= z^B}hfZCnx^fgd#oXzdp-^51Mm3#t~AL8}>()4$QtIaQ>3Fo(39j4;hn334G-q~_B` zp+|B^>7F?m*KVS}{O|gdHPX}SDk^W*!=oqGlpANk^ZgD{Ki37#vM*rryE@pr%Up}# z)XGA6KIhDMJ=$PZ%~EP75u84!9Vnb=sN(0PP zNTcYqCp4AwLaUt`SxlfFg>w(r^^RoP^V5Q?vNqECSH0}*6ERY}_?ivff12tC$l}Aw z$Fw%pnCuIFQ>|12rEgkJu_fV@slSJFSzRc=Wf|o?jir#C0j%RiZWfl*z3vp6?sa(e9_s?D8*jn*5@Osk(rw{J0-x_)xlc(~hD;tZB3uqYL-d z==4rg((U`q4*M0+ZNBrr5i9WQ?Mbv=!iTcG6KQJxJ{ppCg1#7jprt6y}zDV&wK#L}C3ONu_fo<8diXS44|(&)%DY=LJXdwFOOJrd#C^S>qB z&vB88mx|Np)v9Eea*Qc2`$f<8pCRv?3u#2NA4%Hsj9rUiq>yC_-?!q(au-8-j~YGw zAVoj8S1!L&mv%f{PE*QvQRM0na&D_2W1drWKWqXWiL$157j#LZRFRTC>hZSThvvM< zAfuvRe8S$~e>?I6ykY8ZC*&(C2|J8!1ldg|g`+yjLa@#;LB+6~u6drPQZac*o4><{v}XJf zKH$@28KJV}J)%o~!f=4Qpzh4s{GU|>Y~vXoJ+g=i(ZswHCJ?=+j?jrpxbl4fZhz6m zq2Er}5L1P{?yq3acQaQ@CPOdH1qm~B;8|oSD3+TFrr)Os(kcabvao??t(T*m=YVco z^#(OnB7*cndmM2v!yK#GXz4E${HGrj?!=uHA{NRCo(nor(JU{#HnbMX#>flxoJnJ{ zWEj@u?ZMubb(pHDE`-$%6(%R?3k^rC(YI*~Do#(w>7SlR&U8lGiE;3mLqZ?tvMc3% z##YZkf_c*qsM<&iKbb1(wyL2i^$B&2k`n&;{|8a2S}Yk6B~0(I7fybgD=g-@cYWb) zH05d?^(KvmyFmvH4Od5V=1W?(v6r?_l}6>S`%wD+5~rtL!YMOFTQgt_r>vZOR@Bq zw_r1TjbIfgFHEhH#=dOk2`U{=y75P z{oz@lsv^Z(7{`0cAJLKUHWWku zX&H<#nux(%AJp}%o!oQ2&|z0Iq355Hpy6a86#R}7K5U5+x;D=gUW<7M@#h?cHOHq5 z=4U4gz8gmfgKOM`qX)Hl{rikF+c<^VVXjkmZ31mM^-piJxXv* zC!y%m50u6=U_;9br2p55*CpK)BzJ?%UVNfnUZ+1l@P|fbchToLa@c-L3(f=o(grUX z_#72O&{#ug?XSiC3GrAV5sNleZH%~Vjl@h{w3|=HX1=rfv(*SkG+tvZ+lEioO$d@d z&g=LMu$#C8ueI~B{!JlTW}M_2%H;@HJQS*fR-!qhh90-pkzoIfj)mo8QqLCj?3@mj zush`MIS{IR&y^ng3DVE5BHpwMrt^N#Xa`joN2nmfalG*Sq^xj!+i<~$@5Xu;KcU5A z&9LFND{fi&A>+;nMCytQpXD@#C~XBn^P?St9SkwVVkVNluEZ6^Ex3xwXnOn(N_Ped zNtEP^pF(#sL7w-*DJ{^?g2e1UOH?rgLqRjBLBdt65j)?~k zVKpBL>88pQHY(}})4b9{ZRZctv?&QR@XlE_J;s(5TMcDD2Uaupz!J87OEp`(SBlyX z6f+ZtL(ErQfez>E(~USC`jx@wuKs$gG(Cm|9iN6Dn=DWm_l(|ku47j|3C!z;?`%f$+n=~DIx!?n@5v_35 za2nQa$zzYTW7yMML)rDAn`ztNbPD0m;I5TJ==Z-w66HGau#8x?e(`H2J-V7XZt$XE zBHNj^s1%u0mnVjcAc?B3{$Z24wAQjPMbVC6$>)`Z2ZTQP-Q+Fe5xQ)_7bpm0a11XIrpE$+F>KCv_nG31w*Ae=mq(u)# zdSl<$5HgsQL`L%F6!%sIpO$H1_^CJsiDz|j9#B_Bs%&uh>OWoN4HA~VH zDPsTXTUlX}3|;REW~w2xnbqt z3iW*7Pd8lssOm&J_oEHL#{fQ4ORXWPF~8X0fG{>erIP8pE3tzMr!hZQWs<+Of(hn_ zSy}E2c33-)9X51hy23P8zi=ii^l@g-$G>1I+-I^~^cFiR-^r5Jdb3w?)+~G7L*^EH zft}gm$Bw7g(boCX=#z>jGJxiOj*T-su5^QPDK#g1I|yr_kZh~3Y&w4SGH+~?=#6+o>9ig+J*j6`{D z-fHs|hB8&i%FV|7FPW%A1Qu&Jp|&>=x2$J^0(mZOa**(`DOd;!*&@8TdPHbBxl^bx z(iftJE#o;oqhPR(a{^|n2@f{(z(Dsr_60ZN+VclEGddB*Pn?9fn6bi;7HOe#?+&z1 z^+3A#0I1meqE>4?;!EwZBx?`;1QY_?qsN!{OdN__jWRczy)lwt7Nj!x5~B4@AdV zKQtH|Lb&06TFIgy`@YWeIKE%u&3#v4N7pPNsB?=Db=gW#3K}KU>{=|$?d^s8nLoI+MhQ`w&uOP^ zI~{$_Gl2cp7`N2~xytr9*k^@#f4iydv$h~BR)f`@x3PNGBdA@viSnDV82#x4Y}9#= z`J@xN6E9vuxU&`>`MV+UG7dyRykasCFxc?RdhUq{Y*{Egp(gzZ3EVTi(LZ zh6JebzGGCG0=_jU<6ojI>{dx(e`Pm<7o_7quLQ8Ys>qfzfq0`j>i@pO9=9*hwmJhN zeNSxJybPIXz-t=?L2dkCA;-H74c^?<*kOo*Dp$B4_s1Q++i5eff&WHXVco7aNM2|} z+Ws>5@cq5P0Z*7bON8aA4V*P7K)rj2fK(+x?p6ubg_sF8p0dJK+e=uyH$sq^YcCjD zsR}182rCCK$AU-2II(yzK75tLrLX3gY4?eWEpOBPiFFj{S%_S_d-&}90^UPTqh6A8 zY3_^?Vgj?7O-CKu;+MzP7yO5}2Ug>haVQeS*F$^%Y+SI3h4QgVh%Ksy!qyAWYL}!X zo7JhtY%q6oMo<>desAs0C#_!YPX1#-hVytHfIIVk-N}OJ!9=)Lt)ZJrKJ;|`dU9#( zMSG^YFnf-UuzbfB^xoc$(rX8>;x_kLzSqO-K;D&1Po;Z`iFEE_BPm_g6S5TTgfYIJ zLdU6m$m}b{wuki)`+Eb&mK5QGNCQszjls7756G_Dg!(rzSaQee z6ThyVMstjTP;4h7EY~*H7Ydw_lP60|2$8k^Md&auXvf@?0nDl%R z+)c2AnWzWWY~p>VR!Jf2wkQ83LCHqPJ+(RxZoERb|^lK$jc~>|IXZMn2sT6|O@SNt5 z0E{Y2hNFlp@^j~6_`!a@-<}BF3;HNZQAKf<2Gon~5xAoq+gA3$O+rjqn8H6ZCrwd3 zc^*cI&7mK%TPdMe58h=;*|+!8Vx6n;k#+`^)SlxF@U2y zXMo*IK>k5PyvZ}dz*T0*Dpcp3p|8A~XG6CC(rNVg&D1pH3e88is8 znsVS4$?_RgUZXX(%o&XJ#j7!?aS#4T9LI-YS-c09iS|DMsOmd{F~M<2=z7UQj;hgh zcRkvrbrK3q2O##T3e4y{)SKSGXZKt5d}$>|+i!F#c{u#`SF)l0b?o8m=Pbrgnq-y5 z$tzKbGePW0@$Ula);mOF@*SwZU@zS}Urqb#o$y6h30n^TB(c^fG-X&D4IROAAFq|s zIA9vs->qoy2u7e+IL=z_<@aV|e7PaOr(hZ^1L`Sm!%fQN&qNCzNz;rv7s?Ozpyd@U z>^ENsd;5a?UN~ayoIosl0Mv{+#y_5 zvYEvFES!6}Ou3cLUDAR1{mqD1U5d$;g*d$BI1ZNn#G>tC5D(5m(2XDHf3cB_^-j~g z%!8D_MVD$8^)RdBi6psw9M#!gWX*LW$#>s2Oqsa{#R`FFKPrZiqGM39{Tz9&d&)k} zHl@vr%hM?Y^~+%*lJDt@$ODRbr(ULQ_0Hs}nny;(-`Ieu*O~ESIkK!64|Y!;T9+1~ zN^>fO1v%0XNoSHephuy4)vSJ{H!XH~#a_FNr%9Z9djF$1ezZ@8L83m6EGwX)$E(PM zcTT?#YoWsa+qAfK1%lh>QJ7##T`fE4(}^%@Uh|ff#(gH+QAV`F+JS1M_K^`<4;K=yts& zCc5uM<}1+9jvLIVe=y~{iBj2KPa0-3heFegDXG7oeJps#qIgfpHOGKz7q+oMCq3xL zD^XZ%y-y}#tFiW(JI$KClJ0%3Bh$qF+0I?kQ_RV$eKVyOSW*qY zW*Z-qBeyT=RH741E}X|uH0c@{m~t25WEt`vFVLH-!DQ(6kzSrrfYw(VWGY6}AC-GF zEhLf_+6eTYF~2sC_M~T1^vG0CnJG*Nr1RfIU>hicPwQ{d{lLkjvsj&m_Nc?B;SVeS zc868+ensc$?@WiELB#{^GABc0Y9C-vJGO75i<+Vo6flTprKr=(Db>6$IT>O)N_^&` z#yx6*H2u>=c13#-Ep~dr7LWSGj>>j251*6tN#2T{Zj!=G#XeFhkmc;1g(!L>VDXvp z+~+C*$A7oUb>1|b@#74B$qZVxy_0JG9cKT0>(~RsJuKv$21%Fj7!|66V zZSOO0m2~$*GaZi4p}StaBxiSm-f;$n;=mR*(@_EEBWl=>I0;&={*;}$ zdX#k7WfK0oN;kHO!A@i^4YWRm`=bgetnUXc^sc7&LG3MTsQ$)f z)~{5;v{$`m<`;~q@SzELOpT!Nfp?kly_d`>Qj^|x=2BunI!P#Ord^|r=(?E(W$4eQ zC-Jw*jXUNHIN$TBPcQpx@`P;@6QfeSDYU`Sm?pKG(tGm)ve>OoAy)fn5_hO6B#xz< zgMw+iz7Bbef5mL8=F*RLK1V96rQMUBPy=W0mQ^s=g}$Y)A0JUd(k*)Ewis>ImGs5@ z84Zv5MX&lL=)%4TRx(7IW@~ZQ!6JLwcal56+#Y~aWNFJL&OFyMq_*K2q}%z3>Q~Jr zwTfo8`PE>m=q{q}tV1Mmcn7uViBgEiE+)nEqVh8e$ZgF#x~$IU4f12irbV4@{jFmS zNTjzT7SN1kzGS*Jf$mSqr9qPyk&JvQRZaOv)%-crxI}|?r+sFV)`^m+Zz}1>jzH@3 zB`}}*hE%Q}rL@$mwDsY4rtjy?RIU-7*lA1sPZjBO^C+_J=iWwlW0IWcfz12bG__{~ zZHT-~wK0ci+{}rv`uWJ)Mn0%H_ z8g0Rqk0GdBR14pHTbi99N6ywF)FxU=b7s5JnR)5-R9cmKw(&0RW@VC8_h;!#`dLO~ zD?2*(3QJHkqkq~(bm_)X@}6K!^KzZ&2a6+Pp4YEY<$jPAb6Km!Gqy9%jr>+jCEbl5 znAScA8r-LeYoF@)-nNGHU&oNDn)bG2|e_!uW_(UmK+NI-^+Ar34$dewB0ZA1` zGjk~svfhx(a#W|0wzM)eNPl5nhohNq|7!N_>=6+>aygQmKuSv!)Cz8S%5d@q!zc{ESiijFK? zOfiH0u$K1GbdF~+d+cY^-&=K*DUnG}yN2O@(P8p3&!h$Wv&m272pO5Xk(2l~GO4g3 z-Or{}>BwD_q2;tq+MU)YUm=f?TWS4I6UwdZrMCGX8r(w<%gvyE%9&DrFQ7LX!ze%N zDVs7^iH@BYOWn`V$jn8QmEuc{Idio>1;=EJmBt}usjp>W49VPRO ziOGuBWPNidLN4)~v-&6+C;5xD{E;NaJChgx9HWIZE12XE7n<8~hQ2&cpi_|^w5M<> zEz$qZUVh7>!!J_Ebays+n?}>EsfX#q6oK!*hmiY$eWZGi(Ou((^pj`%hRxqkdrHP) zOBe4N9SLU<@;BMKf&kjPB9Ccg<**rEdgS!wKRTxRfaYucqCLM>&;f3#$2KzxgWG! zVj=l>6j0I=B`iEyN`CXtku1-KkJo-^{+7A1?i4v{^jSskIVVBd@GU*8&texQw6ZH= z(Uk9gheqs|fUnwG++j=D?Qg%B=(=k9r{hkUS4wIAhpPm6K0SY-LE2?<{26&16Sp>I ziL<`3uyful=W-fLC>%}sZ~Mr1wLB_++Tp^q5ccukPiCTjo)%29U~6yhW6_TW7KUG^v0l_AtI?INXYE2hS@<8&kZ6HVW{gR>0IkVNM(%2gZ7#NTJK zfOB%Bq;rBP#1EmjZIxvC@DCl#S4YXtwWuB!j>_0=u;+QI?vCwD!&{2tG)?IAKLgrm zS3pJ~->J+@gxOgpvtwXM{ zBy-nuXZs^8DMh!M9=K*ub#yuDv(0S6p*?IKalVTGMCN<-2(wu^j%>zoUXXe!qb3dV zskz4VDi5+-re>76Xc478I6;L<5p?fqG<(!zO1s3a(#T=__cCh>=c(+Y_;U$#&PWWi zZK5#3csc8Hb7I5)sMFrkRNC#+Ok0ieDQ0a9jT!ZphG|w&g-STvUhT);w`fp>LnQ4x z)yA?680{F7M81;~Dd!aDY?Vx<%J*gT{IwM#R;^)MONtB@Tbv}(cxgCnnSjpGhbZIs4Ptd2bgd+UZtJ(x^kXXM zIJbd)F}lf)96v7|E|49*G?7+$aE9B|a#n1U#!hq{pyEp+^usxyRxL4w#TbA3v+E++C1p_QlQ1&a zbDC0C)YH(JKD0%%g_6F!pe{RY8q>vd0rBVPt;tfRdgu&0W9~p3R%{`wPkl6+XYhhI zM9|}Lw@9)6C%LNsHWv|D$y|T@X1nJuV7m@IXLq(fV;M7(*)Ptuoq9Kn7Dt_Cx?zvm z3XQ`w)Q`Jz{|h6FZF?yEdoyW?HBvw_M@Ot$L;owden#f)^$rWAaj~XQ z95EXm7gNcyyP7&4XOiYfBic8Y^BPq8$#mBQsvlrRLwN3vrLSRq5BM2Z^Z@r8pT&$D zJJ6G`3f}U0?4ac!3LBos7B0!6`hqaJtb2uoJ2z=(iUk53IM2dm8s)7v<~*JjN*F9d ztv0HZX)&6TM=qhzC09vhL<;u>XHkFUQF^7BO-;5V$Upor9my%D{%t>4zWo=LINF$O z|0+RLp0h_6MdIUbTe9*=runCi(6yU}G~@kErl7isJnp|@ah#L0d3^=neVoGbUvDre z)IgBZ9*L^+yi>}VYMr&SP!%grUzF2u@jW#9bQ@Kz`pPQeZ?So~`n+#- zi7Z(>ZCZANCRQH?{S3gL#w&POdx22sLcgwkrpaNsEd0vir{yHaHyk!I4jH8mQ*X->ncf`cT!Z0Ks>qH{yrEV>Go-d)gze_2l{3Q(= zeTfXip3umW0T{9w$Woa}<$BlY{?Cgv$aWK1OO}#ks{$el128)y3BG-?nB%aCY|P^5 z_o4ylc~Z_w{|=!k#%83mN1!uT9y7CY8TzH3MeaYX=xt&$wSK$6&RrctU4j(VuNj0t zZd!;+9)tzkENS?qXnN)Ii?o$RNo4v&8npKs8xqYk9yfN=l#o1fOk{|OkB617HC|oR zqQ8PG6H9%@UR}tcq;h3!Q5^^6v@29~po9J;$RqngHT$TX&#tfC$6gKG&zUk?fvsAGHG z^yuXRCECJg!jWM+F!sF#f_oisVwXM!c1%T4#xGKse}YODw^B#62Gwr8%C;LQ(w3jL zG&gA?)$Z?O>p2VA`)(0^t&5@r?>DUa#UJ*lPKNT0E|TPn=OnD>oSF^sR1rEQzWW|lJA6Q-Xr=+;j80lpx#c3j|-;nZb}rR$>*w< zA2Jc+k1Rp=K0EcJjztRQB&2Mo>HulpfmI_DnM7unp+auADrtB5XFBOtKrQOBWVMIL z;J{!i{gaJTF(;5PCLB6*Ua%)eq{u;@&%36k(h=#aWLzbKr49-ZSs90}lW7>D8$vz{ z_Ru>|&h36>%X^ctDf}*&n z@Gk8KpY!yiBmN{dUEP7w_eY?)Qh-~RE*`rYAv9J}kSgiIj@8OS(YbTNfHOgYzw{wt zS8P99K7`! za&PW}DgT6jbOoeXD{MU&S1+zWEk?A5@T+WCD$cYH--k z`K7}N-xh4;v;9&SJp2UjM=iMag$cEfEQI^j+Jd*8GMtRxP-D+qiYhY{^yW$k_j9BK zk9!jYsZMQSZJM2+;lX*WdPATp{H34sQt`vI0A~|VLF20eF1{QC`QuWEQ#^qUywfhd zych@d<_gV5E<)o+Pa$!Os4(%ns9+iX1a-c1q0e)_Hm>uKvX;*-jjE9F^EqrcpFq#@ zqqx59Bus~RqUG@rbo>=?O>VUCT4bm&KT1hxRCUJbn%%h36AnS*Tp?<={-;)ic7CXIZ7w_c(`*%^e>^OxbE-5k6&=p$qA zTV!=b4S#iq3H#svL_?swP^nx6PpwMKa4*KrJ>i&@uo23mv*FJD9XX~(m=)`UaUX8M z!mA8Bo}9teJ{{p@f{xG@p)G`2wqQeVE@wa9#erYZFgJ^ZMc)DLrg%-yuMNgbkD=JT zURubXFp&E-W5Fp_|=Z6QM87?krA5L*&+Y?B9y(bgO&Ge)C9%ALTf+rWh0@o?grZB%JK5# zIcyu@3;Wm|SXCKN8o*90;V5Y3tz)9d`$?)+AaOG;@nG; zR_vwg0i~GY^%Q62E<)Nz5tDT#@m1jw&#h(h&f7_Bw<>@*=c~r;l7_*~VK}z7nKSVk z@NZNrR(u->$IC@@)ItQwF0yb`QbkRgEuxP7h0~c{B*yon)?+c&{W6nk;touS2 z8~{hv;*(^wznq^J>gn$FLYR2RVdwE2EDDi9mDgKZAJtFCQ-0H+tijkmRTTTW z^0D)ECF(6sU{6aX_TLHR9Gx62U1mjhmVko48qjcRhE(uli0MDafG8aS|7i$jHJZYs zS5wH}+K^UoX~B}`KUr_zFuEQ;k-|ScW7QH3?C7Xgwr|vAOpTL<(mlY z6Bj-&%7&t^2ew~4#Tg9~G1LMm>axN7j_FX_x(qWC24Ui!3D6lUhr=18u;)w;&s#pG zwbQ#vL-!_L@;~$2t1(&i2Gd%Z%f|N)BE|Ac7_=Y<_X~4zsyYu^|Ki&%&XX2Z3GBMEZC z_Vbd$h0&+czUmeVa)YR7fg`y&EvFS;|JWe)T(&|skIA`d^0To4+3Yo;V@EH;YE&|M zQaESf@km>dg(?1g&k9cP=X$>PzGcRrV|DA<1kR8=>DWa}1GXdT>Tzh? zyo4H^R5VYI!E2s%efKm5^JYXNKA1aWnwBBB@&j{n{K@7?kEUANXuRnT;oZduG)Fe# z;`0YkT~vdr87DDv?s+`lm546S8aD7l1skMW%-WuHuq|`mvdJ00*dCvcEHdpbYxsSZ ziOi0t!a?)t0cTANKBvKb1ZJ4TXWU-)!A&_VDdspfsopYcb>JCLvP!E+1`O2yke=sjTk5|c>j;{-4 z@a%~^GIw+LZD9o~p4GuF{)#~C^Fvs=I}+z{Hki% z1K&4n?8!d1I#CsGI7>j`HunWAIf9`Ip^&*0hSF2jyes{j{V9IN>`tj-LeF!$J6jbo z_8(!#+3k^^I&sulK{)1j6HAiLW7RovVS{KJRKh0-78X@#_sPd8krO-%y9adv+`;ER z0_95|(O$hVnDn}Z^eVIj%NYl`tL-EP-kvUeVuG;tnwe18cZ)0r){)rpNGjkyis}as z5EOS0#cPsabS{Sf{(=y0!1F;?A2ERUGcQaMMThDS^1t?++#Y_QM}3X-aCQmW9(0$X`LWa{bTP&3$x zP@i-R@Lh{zzs!--JOrB#8(`7rWrE&M2x(=qLWfrqz6T~@(d%d|D-;*ZzH~uF^EO=f zmmqjZ7OIuE!+XMMSlSmO+|&`dPWEs(t$`um2f(G;5)Pds(E4yP>Z~f^Ff<6!aX(Az}QsU?;I$+gGH_l(h`aexbw{69ZAB%;`57I)p%S2&g<4f)yyicKv zZIsf1f6@2*HPmX4!EwkX z#P^27Qq~(T+1F|2FCaVVB)_*V#gPr}IMN@5nD)aszGf=EnD@~92illb`INg|!`3FCy@fA7r;8P*(oHbe)wxk!%oVjk6;|_08J=j@Kg~J|EJj{4b0h6j} zz>06=-1CUDP|wn_f?>iDg~!O97md}I5+FPN5p2u8;mD5(g3~lb!KdgIyzdMaxnVPCA*qduhBJ$F#LD>3S~=(tl%G{R^mdJj03~ ziCDF?2-g;dLwR>TTH8!u`t1Wnx0R4~dkzH;zloS>SCMU4iaR`4?-`*ZEGVtP2ZKlO zy;*^b3ny^x)GPA0%Aup<(`gq!i|2YIQQ(pmN*pSV59TEl_&SXSz8QpPlhtAQRuma` ztWno#k9)bms-P(_xT^^jGkflONx-lRbFf#_51nPXNW1QZs_T2;uDlGnJ`OmTI2H5D z)?)tInOG}s1?0^@Z-XV|4aOm74flQz&Y<-@Iiw%X@7o&taXcUe?FwJWxMUyA{!v6r zlPcl%WGBp=j>30i19><$)0eCRv}599{K;-Xj8+ZY-`t|bxnY9o}%fauQK;-Fh2GFQkXnQG&BkgJ^b5g?>DLtWV z{Rm-`^$%?PHCAXVauW9DSP7%D_3^%P0`9CdgIh`iJ{eubtD)S%^K~Ce<#%G9aRl^> zns9$*8TxizhsGyE9C~I3a|QmsJI+CHLnsudJLA-%0Qg*s#sWPbxV@0Xma8%ta8VHv z&Su!FV}O(^_DC{32kBFJXq$W%!$Th8b14(KYSXOL> zK~rt;^BZTDQ8tcLy?|^$6O8U(z}DBbP}kSP$53Cq|1cWbEF5OR49b-ou$qsiT2`t; z{Hzu%|G1!a)lw9gTOgzVD+22NU{vxSj63xZYf|!&_UsnYo=XVhtLvcs?-!W)0ytT3 zhS;i5Y}DR}C12Ap;^i#3{s1QThQZjBv#=-^v!<6}>ZcNTkrOVx8v(tTnb@rM4L^=m z;7Z{?NIx2k+R~pCZ7Pd(;j0k-E*u*o_wjp%HDnZwc!t0RS4@v$a$^(s&(n2tU3^AYsxBtF{8 z3*|2c3DaFC2$PhD3#*?=3Vy-5LdD0CLXS%a3>L}?OGHW6`3y0f&_oaeGG%hRdIaA7^OIoxc|y)rl}^+J>Xhf}ZLoEbtFU^TG)3DBO%r z%N;m*d^d8nLZQBCF$R6?#%kMpXg~iMuP-mcl}aB>a}UIt5YBqbl!PJR1&jvL)Z z_|)V8XV)uuFP)5S+w!6Ndmrx|B|t(l7E?11W2ssm!jkX8w|*>oO-7&r1d*R((74bQ zo{Ebx$HEVTPtL*0<@2y&&SGTm90ce1u_%uHLW_fC@kn|gPW(Q^-EVvc?H+n9A+vKtM>oXcU&hE$!2!t5#uKj200=Fh5oZ^hHHSwdc zQinU0E*T27VbPCgF@zO>`ZUs<8qydRJMyy09?38CV59iNY8 zlVoruc|8lpgy=x-xO#%2Tg=eVHW}$Rd1hHPIfV)6a2+Zv6CEK8iX(h*nFLSHc6e)V zhn<|O*do0et-&TRiJT8XZ4)FbtkIR5fkTh@IkYQ}XozHl0Lt{cXAi&OCSXCmpRy{3B)Z_?(~y2y&uLO9Q7ZVAbSePj`) zQ9f@ zfM(@&7}FyINu`y%JzgCF5B#R)8AWDCC zl)fKIq3CQ&-b0>*hfy0L^WY3^;oZqo5}eug_$GFI??A`q8eH;Jgv*IH^dLtUIx`Pa z_ibsMkNHdk=J(K>YH8?f7>b!Ar()*leaQ7Yja0)Z44UJB7m7}Zbl8cKurdUv-@~fY z(KvcH0@^3)kvq4En)IVdlC$n}zrLol&obzkBZAy(N60kj5s8{slLVhPBut9pxr13a zpl^Z9>25e7zY0Zv!M&MAxcN~VQaA2Uoc&OY3ed$&6@5e<4Z+bp^DuAs8nlG}hw`=r z>|U`RS5oYd3*P0s8csiFn?dRNBDm*Wr?m8V8nd{ERv57p+_tU3O3Kk<=Ywh+gVMT3Mw%7ct?MES0c6cDZMw4L#$E{b!>aWed4d_ zc4acp*=XXK{UAJOm&Lmw6L7?PD8v${qxkkBP^JqWeTaZduM-Zss$+wdC9VgH;#!&n zCK&5rvC{^0J+Q*UaxGkNJ4;(7GARG06xNSC;yj_L2r&H+V z^g^0=x}EYb>S4=#Po(rk!F2E-99bZNm#Slt;li(_zU$$caU8kPA=LQfI{9_BklXhL zI$dyxb~f=|d!PX#KhMCkA$#DgH4@m$d7Tkcv2;iR4c$~q@#3#(W^@{I+aj=ZLl`#x zErXNudAtdd5f-*S!;KUTA$;O1#My0w%^&VcsyhUK{Z?A;@|&6?xLZ2@Fx@`!lTvTg z(VM>KnCab&b9wi0@Afaa=t>JS3Z#WQuP>7V&kr7X*Gi|u4&wQtlX$oNJdQPuMtQXg z;+BuagR-%>FgXjZZ_+rk?1o+6CV(0T|SX5vEJ^hLBiChFtx1YqQ zoBq?ePij4lJZA046eNgK?fhPW}Fx+Viw!WW-NlXIn zBZl*Az81{3AHv13>v$RR3?jpIu-aD}LAU4Q!Nzsy`)QAVzYgHttKGPewh7UtIXIHc zp#EYTa@;tFYoi}6PS2rT5w*lSspM_=isZe2QqzpR7-TyIs_SPWJ~s>)Pyp|;<9KON zN`8OQ7rFf%0( ztyTPcWgrbUOABN5hog@7PVU?>#mTaF^eFWOwW-JD>3VUz{@O&o=R4{4wK2#pute@7eZJSopuR&z zl>Fm8?b0cROH?D4PrZThbJ9typn$&Yucw_k-$`ct9g61p;-4}K_#-wJGq#W7F0(>h zz8j54?HBRkayEWNt-~Fjf8IOt9iAS&gU-Zm%sFrcT{)%jTXq)7RXZud_AvFBpC5Q5&d0};Nwnr8Ce z|Lu*bInuWqVD^D&SouQ& z=1KCn5GslLqjm5wb{%f6^1)_35v;W5=jo%(6kdCo*4iGSA00RNTyi8L6il&Q)e57t zw4quu3Ectq*!y2OZQOQ+#8jlQ^Hvg8=fp$$@iDC7&%g0YGw?RyEZjoRVgkR%HM#u7 zgWpMbdSO3|LN;LkyglS2afC(+VN}Jty#Z|xsnOsZ-TM~`J$@aqY6?PoIL`@4?ZC#` zNho_&PAA14Q0~iY+Hp4yTi=x+yT=o`lWQTIj8p`m!&-=d@=qIW1?g!tMf91=kXFAY+I2b8=dC%A9CXEk1 zOV4;$VB>dR6t}8^ZTEnv(khH?^~Dy`O~^_+M0Mvsk#HlPX7fd@Va0ly{O~MwWawiC zf4;k+D~-9w0^uE^g!0x|h_V$HN7fP2B6Z8@LN0ge4Fbi?y3t$TN!rr4dK;SmvJaZ`^} z7k{MP3tw=)@)?qBNkGYz5X64mjt#p;xpDZI2&f?1n)I!=s zakO+_q>hAA(){T#DTV|){3B_+_;SDE-jVURn{&t3Lg z$wI@L^J$&v_P+t>Kl_OOEbO8mTh!3-av*eX=;832H8{?5042*g7jSb3=G-_4xwW~t z(0mLH(OK|`aK_nh|DpG4G-AS)aPGzys(WBZ<1-ghQ-J{2^Xe#`ER8Szx2a%!F?r0n zNy85Ag+`he_Br`sXiEtGa1Q?Q*m+2jorWoCHV85_<&Ir57;;|Ne$hebVX>6oGLvc} z-Kn)Z2M)iB@RjqbQ*~o#a-ShNamI>ge-&9be4{xQGLRYaid1^aX_fLN+W+MhY4Pmm zyTSw1o2HE2PEy<%6p2;*nsS^w@-{uT!sdu(%GKe1S(Zm;;(XULQW77&_tTJLT6mDE zglJD0{8;yzc5}9b__6_5^6(%vJn^FOjtwNDI|(V{Y;pYw&sU4Y;KgeXXuh>ZwOSmR zkJKVVzW}N-Eu`)f{iOJKHEBdBW6$WD6uz^FmM>G_3@SrJb=GkYOaoox&qk8=X=HIZ z47q_;m_ODU>4|l8#W9b>u9T7WAsImDI-R&VhP-WfnAsyGKrwx0Zy#*xu`RU{YaLN||d+f@&dpz)O=-X_z& z(uXMi6#(1R(@5rdsMPIp_>cd*G7eX8?X55G8BfRa7cn#})0d*AZKUj#JjcqtJtwac z^YW@?p)tzzLdB8azTWoVz12MsARr-}@Dtdp3>3<3(+^yVw<%H73u=A|v= z^%PN|ML4@v>_SnbOqv~LP>3oa^Y)iy(i6v=j|VfO(Rpn2raNrIlu^`aS<8Zc>X7=1 ziPSvSlDrl#p*i03N$1sanl>YhjkPahRj;lvgHyMu*VhLn@fc%8~Oc?Gvw#pp?M)W zbfVXoE?wA1>ZOx-Pcfavo-e0?zrQdO_g1#L!j*d3Ub5eLF3d7gU|uH9EaPDt&zc#c zHq``TOV(0N-YFV5VKvog#xsw%)lB>NYZlK!$X_&xvNV0Er4p!~@Q%9qvwQWd7^dG_ z%BEG7vg%EsrglxL`(Q-rX4&kfNH4SJ%!!VD`D`^=u(NT+Y^qlsZMbJht3Eo>-Kz?8 z-a~^%oZ;U0*1_BtzMYPq?qMI2;@HCPQf%2}C34&5!%kPnu#D5km}#Xbnbjt+(!=R& z#EiFe*me?Gh$qv*^K$gcNtKR&Y+=cbjX<%_R@*B~R7#{T1bPUF;8 z_6PUa;1{2ny>C2I8#RUnd{t!K`U}}g-f>zK`GENfBpILpq#zr>ZQ zTljk&rcEC-YN*cFfF`OYv;K!;c!y#(+xOunTf1ihix*b0m!}u7pP5B$f$AAn>a&t1 zHmmY%Lm(9o{ma~`hsiD5#|%H-Vl(wmlJ$8$Be6-Q-=j1r_R=jDFwcu(-u05z_KEl} zfOC@K>e&9Xscgg3NH!^L2~EAMK;D~WNcH%6Cd;`ZZVCzXYE&_spBhMAF;nRNyWcE= zcQmFi<$c5R{?ya%PId!FQIhLx7VfM;FQb-|hLj8)9^1nXxq8yhBqw3Ep|LRKxRu~C zD;G0THsM6n6P&o|i~2&Ii8>Gl^=azDxmsD_^UKM?4Ap$TZ;BJ@V*j)if~mftFmb$^U^vH5 zm~TH#nEjN5WgiU%%_MU{+eKcO@;4c}7t>)sG#NRvWrFNVFCk%ZtnjAkH_mQ9g}{%U zSmroZa4C5KL&+xeTp1-u`-lsbkL83f!`2BE39iCm#=HB|BQX>aIFQcI>ctfpb>IRd z!jHnBbG&eNmzc1>ua!DMmWV<8Z=cVLp6?Wo*2l7#kp* zyuMOc@X=D3si6k#aXeR(JPe+zo}$J2GwfDeMZ;U3pN#uKD%LfWXL=eUQ}|A1)fISr z+9>=wzg_sOyjs{HwOXk28ZA6nqAPTEI^t{e5Jdi!!QJ3L7_m@6u=oCg^Uazz`>t z{0MmIDj_WFu#k`#ECk-N5%%bHBRy=0&{FpfZbtE(?VknHlyGzfg}_hX`K^q(e2?&m zJ06NKZ(==^I<^XzKP(bLBc}^@Yfw!VJ?t^kQ{C zwKN+E7o6;btEIz*3Dd?2o7)ux{8JSQJ-viY9&3foYO{sQ`OWm%;yPKHyyt%SSy1IU zN$0h5F}QOB()T*R!`T^kPDMjTd@Iiati)%dSahAwKy6bfwk5~Gy7CZC%|8#ryDEbJ zGD#uL=q|#YOCY=b9;Rnz@)?&WX7O3?gT=8J5j9Y7e)533hw8YSrUj2xi@mP4o$GdN=KZXj zQfh+fOeTD|HA0Bp(MrV|Z;)e)2u3s=#+S!ykul1K`yR&#TbBHT_+C-rk9ZPx{fUEY z-yw*ZaaWGi4@F{xbMbY;Wc@QjR6wloyeJ2r zO=oaF`ve^B{iYMrPf5w_2D$ynqowCEXjf7h=e@>5rnCra{ZsJcQZ0-+FGA7oI}qMO zA!CN1(@hEg?Rbn-?K1q1YQ~`zAL-eID>NXwj(%{~&$h20X=}C&4h}PdLx~#b-e_oO zdEmcjeh${1i>=M}FfDTbA4TWk*W=rU;f5$pMWsSW(XgWHyq;86*&0?t`6;sZsI2ys zXj2lAG$<4)DH@89>`f@L_lVy6{RjGds_*xC?)$pV^Eg;DV;ZY}>IQey-(+BA0soPP z>~R-7={0OL(?T0pA!fmT#>jR=n@a6Sod$=PwSdXfc%? z9}MMw>-uvyr}=!YpARk$@x`#aiK2@xxSS~gs1(k{izj!m|A}MV@xL=XI6$6niu?W* z7m-a}VufR(C%G)l0u!If^TbkZks~zZ=66weA9xxUA2wrg_$h8LI>8@~9A&jsHEvDO zW5-B&9$lBuqfcaU@t$;EdnXZ^VI`;wd4;WVf+0MjBY(F131gcvJYChB3yf!R+tW|@ zXQ;s*Wr`ebRD&a{WLZ8(hOfIiaq56EY;$xDulux*-`(9Ue3#k$!EXAi#ZpGakQ zamUh_{~X$xtvKItjo{TRrU)_Xc%|i>C6t3A0o#2 zB~~Wfhi=yfe*gUmcg_97e!lkj6-)4IpCGaW%KYT48LvLC#2J&s99Cx(8?5WYDqB|J znDu7Vd)-DK$AjD&zmhM{tK%mwUwF(Xv41_-#x?)dbH4dC9&xsjH=Qd%reNQ79CaFg z9%I>l(olAf9>!ap^|`u>GFNRgU>~1E?$o}YH~M9NCh!<2( z=g!U6tg1YXy9w9U(wy#G(yGh-69)63wxjI6u9(-DX7QocJqU71z^-W_!pUQZPf9=O zmgN$>-_V)Ert0#$pQ_w^?u1B^_9_4wD>PYmY$v{#uE(y1U9eenMuY9Nq4;_xTH_~) zbAJ)qy3P=13K!OFcj7~-^WimJ@L8V4;=hP0j_xe-sEU`Ewnnq(>?FP<=D2$+?y=*Z z8(boo*Z=k>eX1zpvrVDa&NG-5?5t^Jfhp0;n`xn70OIB9eJ+%kHrmrRw$ zlW4GBD0`NO9qhhpPD(EXkJ^lGt#MF$7C?Tx=hNl)#njnzGrg;u_|?XnyZNtelSVmbO`>_?`0 z0)|;!MQ-h4%DOa#tn?`xvU(=Y}8k~MP1>7MQ$A?~~v;~XFGb^38 z4VBQgGlTAg4n=`*(7s$BiSxJZI4*lX=9^unkWN==SGpnx?=s<(3nCA?PZshgyW>wc zKwo(m-AoUo?XT=u5@8=lzE_!+L(2XgT17@jd!3$hzb@cj2=tW9sDxi7BM zk|=fDe=WnuKKwzZxg*zqI)kV38Ccs;gcl)YB=y)sBl4}WzS9cU?X#Mlix#lF)>7oA zghDRr8iuG(q1K~|=)Zn`w5F>cy>1Spy9Nj7(}UwQrD-F{%}~I`qVLqZyoF@UO|i1i zM>1)AOqwR0(Bqy9!oD5=B~CzNLA&Bz$mhU2@*cdKT&^4=tJ!BM^JXZ9JzR;K z1@_pbnM2c-##7Iqh2+tHJ$3mKLgW6G(cgS^&V7Cf@7Jcoctk9G{!KwjgDXaw&q7ZZ zd3ZUWI2NWOl{jpeA$#4DVT-p9@KksfLjD|~ zlGv&A^G^KMfN6zkP6yK{beKCk4RKx*V@QKZB9@9=`F2!@IZs z@DTm{6fpx0og9S$ou4A^>roBI3U$Drr?+1|lhX1HaJqI8@*{UbWnB@EaxdV4Rh4|{@*|vhT7>=Q za?w<*3QynTl#!{1!3wvqC47%yr0eo(+d$a2T&E*8YH(d@&Ce{?a?rm~{PVjmlzz(K zRE9E2?X3i#dIxTgDH4o^K+)TFM_JcUnD+XBK3fv-=uZ=zTa{3`V2bcwjDym-zt|pb z%We0&^1Q|aFz8i-JCRp#Y^fbS1qoh|=oz}F`a)B!Bbp=Qv2mF^$F>;4D<=~n#r|UU z`3)cYRU>|+CfDhTySDZ(*e2Ft?Z(51ce{^6q8oW&p)1a9Z=kk`{?L0oAK7D93T}lG zCOKA;49?%T4}bBP}bk9Oy(1lVm;VBOc&_;R%iI?Q!If#^!r)X8#rzZBGT z591B}Q#iSG2dDKnz|E%KSaOhYJfR6DVT*9SJQz3Tza`o4N-(c}LM@5ssiM*yhi@9; z-7vs*wl(%E>f>hlJABIhfUh$J$AW#ZMWF{5uNLg=#d0u>+D8*7>0@!e9lj1$MY^T} zOjlZS>yvl5a$Vvtoo-=ZMlKT1?!jWQPsw|-01ZFAktI0PkCLOYqQhIriZeiL_kpZG z@d@sIOM#y71b9EpzzJ6;L^`@-_>j#g&Y8~XCOx?MyDJ|YI{^c{V{tW3%v^qbr>Wbo zQ@6PfC}PT28a|*C7Uq^yS(j6E_**4?9h^$}j*qCY)>(L`uae%(y;!h*4es*(V_=szOOLs;8agpoFAA?yZtMTsZdNk_S_&Q6Y(y~0mD4|D#` zy@&|##=h;pNqwR%{Em7Eu1OLeKJh?ctKgg)c%d<2K2)MiamdRO#fgvT(uo6P`p5=% zQvmH5!yvPC3MN;6LAaP@KB|gFtLWjnik|Y^sb^t6E&&_G{H1Pv0B)_HhnlKcI1?Cw zJQrDRF3ZIEW!K>O<{DPU48^{rNE|t}8wd2(;NiVWP@okKdi2K9*Mp&&G#LYz7>W1I z4BUR+n^(I0!>Uq!-rc%z)zDM;=@*T0#m5Uv z;o>t7%?3dD;ekjC>W556v3u+F0N4AMV@gU2-Va}jDVUF;74tFjkQ1J{#=+6D4r6MQ zabv4zAUu_V5b$;@lmu%u*31h!XF{NrSB$p- z5%{=QiQT_BLw1TAMt{nt`fiMYPt(y)We1u&eMalUI>d@O^Ri_(uvRh?4Am{LU#<_O zrw({^Di&FK|DiEJ?AH(4@xRxtn>(Jv=H~ZEZJ5H1$}`y6p);5J0n^VI;ra?KY%kgX)oCliM~0yOx?mGS9V>sB z;7iF#R24gOP`4ZS(}Cb~x<6V>M#EE_bMN>?L#~4=cdON8--ze<`S1v=4tha5Fa?7m z=b>JA2fV~*cIO>ijEI=TL!~J^!Zwwi{$9hk*RwFSO_4v3Q|ILENh1HuB(Vb)(scln|S3f=(UYlcjLbca5yT*6<44^!E%fD96eX4J z!oP+DY(KC97H4G9FGdkQ&cCTy(FET-o5*dsEXo?gF@5+>B-jnZfUQsQ@l+L7r)hE5 zNJ}j1`=0DJ5bfX9MxQ?|Ao7Z{SRY?#~J)NEyMe}#p0oAADnWQ;p4RtNK{Nk{#z}qaUPGMkIFFe z)e@9y+41lZb2!g!H7Av>!M@R*(ck5j)qjl&SSlk22b&`LHA)T7FAGLfi9FJ#+Tq)| z{-|^f!P`W^^YakD4{YsXrBcONr3EZ6w;C^!yw(L|yH^Ilb zQo9s|7uFy=!kGKm4d*PeoBStyS5FF@ctQDaKHMt{nfi(1ofQS$^l0q5e-^pj&R|km zG8sk*W=f4NezX8xG`$hK#|raQO%av%jlS+0h#^@~kW*R++?<8m6|Hbk$V0k`a1z}J zLD=%H$P+A>ry(gQJ-icE|II6*;t>g z9xq4J3N0=OUc%3_mD$C1FOo)OLI;ruw_BdK&#UVJR1BMhDh;w6axF}~~ea!$&jf=%DpD6b5 z5FWvX&Fqo$1LHUTM&b%$56@J7D{`SD^hD?9ax+d8p2e>3ptCG|AAC&!?=6f2`-OF60iO!+zh-3&lM`1rE>rdabZZ4orf=` zD%{Da79;yUMxeq~&XwZ0aihE>H+UpW`>q!6!}WMOqAMS(DZ{p*MNo7}K~wWV>=+i0 zGyjI*Wxug-7MwOW&Fiokn1&N8ZsWJ$1La-052yGiIAYcWhn|A@Iz@{M-FiT6huAY- z7>{FL!y$WU0k%{u#~h^+xO}^TK)>tg{KlJ~&+y@m!BKoeFybaVWWsA|Ha6zfV*if- ztgyR)?vlug?&!|nS1)9Xzi08{RS9ekrDN4#BV;^q#_h8auuIG0ZKZK+vh^S<-noPe z*FPXi{Te1LxJKLDn(5=kRvLOT1`eD2@%cZ07=_DYrLzvi;|w?YrywMC2^`<8%_rr_Z`Iz=82_F5! zv8?MAX9dy~TH++zm3TzoS&-cI=M4huhhXtT#DHcpl#4US2jV&pg3q z?>zK*JPK}V8t@ozg`0B(=dG&}pZs_j$vsp>cl9|m=STtwID!xU6xoF5b2z@|M)=%J zgZ*qz!HSE-C9#9>3oS>*^L;$HB$1uA?BNI}Q>^YV2(J^=(c{Q2>_6is7+Ogf-TyZH zA7|pO;KK%r?33Qgcw7!BK;Q{4pyx64RPB#D)|YYYP6HAeMFx8HUyNUU9S0&`V0MR} zm^<$v=CnQ*+#7kWANK@v&x~T{%iZ~*$buW*Fys%yr;RqCy-% z4&!K~$p1f&!qo{F_#^`RC;ve3!iNY~xrs$0Gu*N7bF3L5iR}Idj8E8s z_NgCmVoNJbHr69~oC42Nm1pj$7zTh%S4`i|n9GXKvqKMFvZK=&x56z1cU790#h>?HFTv7vBTFq8sVp(l?a3 zXeWOBo+0>%lc3p47xUk*Mch~!94)M}%C-GzHTFgj?HH#fnQHl-;pq82?^ zVS&9*`UyYnH#*{_LCtXss9R|?jV%_Qnd4XBFlrF&zBbXi8Fxsgp^#o%$neai+Wg5Y z2m9OWNMHXkl|5fbg)Xn?V3A-Rd9~8un~U-E%swn$lm^uadufE_Mfwz;PJMfiqh_Py zr0~|9qSI5z+P*Wglg85UAss>7ys+W1Bf zn%~D`(K#!;oV-LJmUHQ8KX1x(?kdh;-ZZ@IJw1$^j8`|quyf&Xj5{vREy4%)Ezbhc z_hsot()bMsdwM1$zn2n z9uPtaZ^|fa*dKEB>WI9$R}{JLK25w-Nx%Q9B5Le8%6t5sMu*1J$5&Qluz3ubJb8lb zZV?z`Z^mEjj7e^~HR*qEwQ5>_luUX)p@SNZG(k}wPU1|m)xZLoqhiU#!<(MGct%>e z>&R|2QL$q#9ca6Quzi=|>pc-V{TmTE#19UYF>uTdprq}S>DfeQ>e*ujb~mKs=(!cp zQYxZFw+3Tf>SwZ=6asS(7o;gSpmN9&wEcG<85?$?{#Y2C2hG5kPJzg@nTX;~UuoKs zbm5R5LDybZ(G0(FRA~MH4`1a1FJ(o3N{2jbmXTzXMsM`8tPVf^ZdDqmPWiL6&{Fh= zIvQ1xd9FY0)D9uj^)@JetUxQD@1*FG?Q~S!`_j$g(cAP5s+w<6?T)upd*L}9)u|-a zi^e!RL=hX#dSZs2zIew+0?qXtbIU4m0nZO}L3kKo%L zg6Z56q~=>fqni>1kEyd-t!PM(B1EG!vKJLBx?|<7@SghCKc(VO4cy-73bo8p7|@tS zPnX3~z3QaRP1cAfif^={T1 z+A5yAv$N8uA+6QQLbuwgQ`Z8kzit^|yLz~u9V_OfuBaZfot}Hk@n@Qe{?WH#9lH=- zK5}9YmqqPUVkv%o5{+23l{OwaK|a$*;iX_gw)LJ%V@GSiPPo@?#QCY{(g5u4I1a0X z`)uf!L>h5lqE4Nc(d!u@v@K^BnU%LvMbSr^d}j=;D|kp1+Oj-OelEhNZ@~Qv8!2`B zV9I`}NN#&_XqMhyYS>vzP4^XG|KDU>e3?b6@j%&>NaPhpqu#lkqU@?^Onx{;t;NP8p-i!;Robjgg#cdy*P)b2t)qc#bdM&ak zKN(!u8S4}6^oJO?y#96(`?jmx9O>{xNo_cOwMt0XCX~evV^hxJGGPu$Q$7brF z)6iWs?xH3u7K-QEv2Hv-e+xyfP$fUXL^LcYrK!8($?)42Qr@?i@{j!@`|q{nwe=)z zHq0g0M?}6ip?nN8rA!g7hL5;cLKf^e81t z6Z~dtkN)J_XhPO{mb7HsaoRo3g_2K3QMmnBdiGuXZgocFW*SLvD^qErp?LO=Dx!nS zO6gVZPrCg`4?CQ%kW--=<|)_Gp_`dB=j=98aoZxCibD5)v6_5p3Pl^ABm0O0)EIG^ z8XYH-jn#f?zUNMd59HJQL+O-!bvzkb3hq|PdGcFOO72(R)2-?aWPD&f z2+iVPlJV#%_|gT`o_Can)uvI^sCzW;Mm%|D=#o~?MWE&hP;9owQ2QkEo+dI`rKR-c zKndmDsT7~fQfk{}k0-Nw;QgR^=+x+ne@eAfDkm~-`IG2dQ;St=qxcTpSD<<3vSbz3 zh5o$UN!D5Jq!F)4$9F%Vny^^PxnD$AK4;Kk(@a{TGnZhR~mNhsochJ8XVcT4|d&VCu}jH2KXZN>j<9_Ls+K zwqhQoF5N>x{nJUMCK~^g`{U}AJEZzj3n6L)sZ&X|;HqWPO{YRqmn8fgLN}Ua?cuhp0GI=!yKQ%<3$-kq>yZTWu!HUq)B^PJw!2?bEaOsAAd~-%PVlw=&$;1XVAuxF;wCh^kOf6zm18C#D& z!LNU=<+Z{+H23IagazBc@VOIW7Gz;v{WF}6-2)jFJ9ODS0I79d5q<3??CW13_=^T# zQdrA7WfOR8v0xNsbP)abj_e!s3Zu5P)3fof$@5zUy=lC=+UEk|@^CY`|Np}TM; zAOmmOZLzqmf~rCuQ?DZyymxjVt`5G7s%blLU}h>FmPg|A*+?Gt#FE=SHKAeTE3^iD zM2{~QksMcu;Pf{bx~TgCRrvpf&mv=8ZPuu-`@^FK&hxdW$jLZ;-g}F6HO1J98h~nOrvY z476(m2l4X|%;@C{nMY>WFW57mgN3X0ZxH-`R-(T6DXv=7!+e=Ps@{7da8?(TZ#qwl z#P0s#fJ{DfzAGOVJO34l8+hCj@Qq%N=%Z;CwdvJS!PZptP?!(pYw6G~4}j;0&gf@p zh*fpET;uu_Ba?o>dUpk~i-cP*`U>LOZlkd}7OOfJz_02S77vSn_R|*dMLq5q)rR*u zo!S5FG&N-EgU#3+K;6u{GNUhvmcYM{zFZET1OMz-Msfc_kLN#3QKv5-y7S&%Ek4=pXk5 z9ainbuh0@W1V2M%$YA{Vk&loy3T(1S8RJtPVVzwhYzvp*q53Lh?7vG$&_Uf>eRMij z41MiJXc)eRhOHfz&g+45-?dONsx#-VjzYqvVEp(Vg}N)A;93vSAsfRX>wco_;cqN( z*TnODTZ9U}zf*M}#vK}rk{=T0nw^FF0&hstMf~1%P|Sdu@Dl!lqdgk)5(u3Z>A+w@ zZ^398!%Idv@ZguR{O3a<_wcFW88zzCl5&x|TJe~#IEg;#Dc^ko9mpd=NSlb75vI?&cQ)t!jaP@^#+SESPz=(AXbGG!EbRPEIcmp^|;BB?UfNyRZVvu`MNK++c@yf z31Z)G?ZtlUwy>Xh0#;2siU}@hh_7EIDQp}n4YqTYBCVEjZOv_V9(9|4X4bHhfu?kL zg{l-<=fvIr&E|hCo*X$piLb@wa#~~#yZ4jlP2;-r^h!$}CO9N7H2s7JBnsDa?{SUR zJ$6b`m*#D?;2-ghTpB67@*NA2>%SY@e`g{fxCwTj`ttAUc}y#oO5K;ul@1@!f_Ue(|~a=qB5PLkRc#LszVCpnmI;KRS1_}x2i*6Di_Ceu$KzE*}GOs_`s zs1RJW@xh@OE55PSjML>NuBh-HuO zLhrJ(@K}BWrnbT}x{jU%W3^ z;xf$VZF%Ohdft3;H}?>qsWBGGJZy0>8|<#)j!j|6X$Z&9tqEA&HJ;;*PH@59RPHye zi`0MKM-CmQB}H%6=BZ!0@#3rQ+(o=+wH?;ulX)_N9A3ce=55@iPOLGwgK!l#W4fLm z&z2&^-g`HC-zhrl1x*QcI=0N|&;D+x}Jj)9}%E0v){3#NTJJ|@9 zo)*hI4dKb_FT!x`MQnHxg3Lwvd{;0SYd@uM2b~)HSAHIaV?_O^ZuBr|`St z|9I?GQ(iN`h`kTb6uE0T>BmXogwt#iovTMYHAr6S+g(knQS1k$Wed>j&ND1*?Sl(J zvZ7<;fFU=`5U4s52UY=p*2i<==>~owSV_a*iiE!GQ@r$^$L&upQRUVKawvU5zu%~1 zOR6#+Kl4SheNqRqACrrZor}$1D*VtO}ng zzF4Na5uy0{ZzTVo6Y+K1T5(q#LO;%Yq5N1m?tiln z_ZB!o=j#utWX(r<_*ovV-(J&R<;zrbT><{Ll4#qhw`4TP3ig9c_)_<`D15q?Uo;=5 z2VL$^M;jTe&svV%{wLsFeHYh)wo<%nI_-OYi~>bp!dAx_C0nC0Vb>09FDOQsl|SF? z6H5<^@6xP>j`%xCglHd6YuGL zqW5^yn&)?4%v(hMFIK&ce9YfcRrhH4cpN3=&JXFQ)pi^^ApAfTacEqcfUQekpiMki zj62D&`d^U^Nf`r2op@B!K2KzN5|^suXZ!(tsk=?eQx#w~ zXgHd`?}zm=XM~CUtM+9T6uwy{c69G?@Qos#{ZPZ|d6!8_ip7$5El3P8W{1bGamh)Q z-8%Qqkjv}(~JkS`EOsS4GAHa z0mn#TV+0+X7m5!-Zcy-;iQtTRc$vEcDq2tRw`v9QH(kSS8C^~eGKcx*UYPaoHjZ@L zjaduDym+TPFWetQxsCqhu&EPTXQ|VNF@waNH-@BtH%Zs=G;MFKr*DmOFkZ~p!x}R1 zW!GgoDgKYki@V{RPY@Eb%PlDu{7TW2*Bf`>Kzko{Dhfo!tR&%2xdiKHG3eDb8Qs;cqf~T3$H*VYH^G#s zFc`!p@@_nt)^hd?Z#cT8!tO!=ZkPr0qO*b_X7_|gXlI~xq&$y`FU7h79bP=KBNqvt zp2Ir9Nq8|Ix0~{@ap@9%V{x2stZ(AaA-%aBL-_Z}ZG72%Ef4#(Tf8^&d0okVJev~& zkFO{3^rJa1G%@3Rmu*~UT!#*VRc-w8AD%6W#T$|1w*7e$GS(kZeZY=SJ)6PHisku@ zwiRE$yNcIUE)&_uJy19&JS&1@rt%^UbWDkTtPgQ>-3<;r_m{(SW`l=E;TCp?&-V(* z>%+kaRoMxLrc5*@m1ExA2Ba1k za3}Xoyf}70*R|JizkmiLovy_Q4PD-Bk&O27iMa9kG)f}2VXXcVM32tFAFtmiEo{e} zaxGqPz!T1|eewN67<9S};yo`9^Jo3%T(BVu)vr46q<0G3wlE8$-W8+ww!?6oxC5Ks z`9VK738S*S`0w0g4lZ)%P2L;%NRKfrH+Hq)Xygkn%`uT5KEsQ}zOBpZJbwPHT<}un zz&CUr9K{~6N#raGuX?~+b2|pQML;$z9VXYJq4#eQj}zy<=|^_)f>EBRvkb+_(>ZXS zBz9Z*5h(d`59y_IGl=I|i)YiK!R&i%JfFFz$npvU*?GSU z-##&izo!gmwN-0)WynEDog%TIJ`YdwoTB}al_hVo_s8r|82?Pu=QVIzRs8Z zGlIBOdlTwCy~R#4U%0#uV7YN6W&}UPw@JIPQ1&EF%&bS4;HB<+wp?T?@{u|`AA43e ziru^h8y~XAu(65gTvv@(~aDIjDNn`oy+8x-e z@dxKVY4e#>N0G;M!`nxA3n9EgB(@a zW{M|&YPDvS=^~>MbQQi?-?077GTyOl7Ato@%4;`7h@O1`3Pi_n-J}BKJ*t8GDn0(_ zEqEImL%C1xCQiO3I=7+6Ff#K2+9r13zLQ!}=6MS}JNDr5^%lJReguCIoX56|B<{KX z42MtIz`1v_xz4Af)cnv`x?(g?dirXJv~%qfZmI4n9Xxf1ZQqSy%g18&V(=fIXrCnY zdK(~VBt}bPeOja@Q#tFn=`SR`*G;HO|A6{68XT8DPqLjmR~kDcN*d_fTWTISN*eO< zm*}4llE`DYbg^$&$yd#QO~*{+ceAF7-i0++Y}?F>#wW1N=peq>MQ~1QkMY!Pr`di< z1(&opvyG+||Cv0M2U&*m;k#;5Y`>9GS8W&RoZ!7}dp=(Zw-_#6HybY9zF;MFGqIJ% z4@>4d6vzD^9p}3hvC=fNGg7hFG3nq2C8=hWtrX~~D|t`uFP%GOB&FH)l3Mh%r0?2; zq_#{QNiE|SN7WfgKStO|UOKJ#aj+M^QeVsK-Gewm`Jm{}9pbo|zxY@2JKm|GCTTpf zm-IDUCAWt&q~(z^TvpeH++Us8eN<2B;-v-B*ah39kv|-zt`FR$Z&I}M`HeW2-PGZ# zu)cg=XForDBT(v_0V(xG+Ee0~3P_W3oH zKd&{C;?`P8ql?x`VIThEO=EocTx}#54IC_G1^P*O9bBbMxoTeDrIKfhSwZ3c1nwAj zo1+*1;fa@H_)PI`o*8%0 z;I7m(^Nf_eE=pQAszkDPN|jCvKAp}PV_uavnO{|@vTgA=?$Bc#XHV_Qi3Ro8*Z2sz z4;M-?Gjj|N3H&~ir;?MIYM)C-a3~@ej<7=T2jJN06SN$RT`{+v+ z;TqDe3y#vL73-x_<{P9n{Ze_mavcXgRFU#W2rf*A9B%3Og7beJ$Gg$8+z|E!Z~Dqf z9skHl-)l!mQI|}ld5=6LbH|C&c;m^E$(~VCQ^O4D=53B9?P-d9e%Du?c&e-9t{@{dj3d@6o6X}= zrgQG#az46JPU@oCS4yv!V}*ym;NSKR!^~H3zq(ocsoz{)YAPdT?9`D?FEf!WZ|~v} z(Q&N!FPX!aPh!b#03Vz;jkEox3&!RqsjMMhs<;*-rB6O6={?DlW+i4xCKlC_Swf{` zrO`&^4rP?n;TgFeY9^!m*|g-|3L3sA4ihge!*KuYuur(iziW@P@B3hGzAgNZ1GZq# zvYEK}^%W)76;i$*(6Sf>WO@Zrf3c$pxSve_wdYXk>u`!U(`My&>5#Pu#H9saXnDKHL`)Zb?{wxT zYHu;Q{0$DcR8y4tJG!s1gWCT5pnukdq^7G$>RrB4X-qZ!O82E(N#C%zYaupI>xi*K z;sp2F74Ef;sJI`^&L3y6NwFE{sedHxvu|m+Mh1O1C)^u&o`&T+ik+Ah>c<0hy|u9M zL=CwN9!|ka2GP<#f zQ|Q2`C1g_=hq=jG@Kdc8yXJvtZ~I0I+%}WxrftG?+lLcmdvow;(P^p7rfav0C`0h@ zvmTzK{w)!tzcUBp*W5?oDskT(YQ#@`b8t>$C+<}Tlim6qq%u65HeHN`#efg=DA5ka zKfggLI**fqHl3s~jEFiv+YK&(>?rjHPw zw_!VIL1$|^Hd>clwN6s`hde57%caMg%1B0I4Q2TCCXXI{F;ecXa2{vSl~eXe5#0yR z^iaA#WiY}Y^+D&0jdXdhI`lTlVxM2RU`)hQ%g89Y-*Y#`8Smr$i|kqP?{xk>x|SpF zWN_#9^Bmu3jMVt^^kc*=vU3{%AJ0tsqZ&@hE|sKMagVIdhSCz#3gI6=LeXPq)8AGn zx>--FU+ki;Kjbm@jqr|lRe*cP`IuEb2`)NDc)7d&*C8DhxzSkO%L&WNI^)sIW*YmgmONga zpl-drDX{e;D!#@eZ$khE7xcob56kf3i8A`kl3;56jy_o2p@rhk)VARl?c68>3r|B_ zu@n1bk)2Ig+!rP9^)XOU_*egAz(DK)%4I`f)Kyk8-(S!Dd<`Twxk@Ztb{k2}-Z=SO z%pz>UA){fBhkG7SO5}8mas5Y7v`2xa2X4)aKQj8jiHv%&4--0$-R zZf+QW^i?A;eO4z7Qfi{7<1di1VIk#-&&@_X8%%TTj)=@!bX%ps#Xhz?@@Y1@=9}~1 zZaq0RPq<|qg1O_QQo)85{>DT75HW8k?8N>?a5pa}kWjJV3?aX53%Aix%0% zQs4*i9J%f-x`08vW!nJ`if<$H)k+w5QRE<|XW)UQ1$=Kfz@*g~P1y5F-rv@8C7$nG{{a`0V zIb{lu#u4mQJBz>dKCF91kF_p$6y2^w$R0S0VO~4XdbB@2C96X2bt{GUG8eqx34D5` zVE)bxN3^;>_Kw_+klPzbP z@%cIArX@P0hU1~%J{6jk!ppTp3toN0=>BF)6e-(4ZudZBKK)D0OEoZ~_Bj3bA_M!h z5|B4iJp0Ea!F$O}44)r}b1#n~EG`AfZIn80#Ti7csjh`n7Hm8Q*V9A`L zf;0n6hz=M1&mwZ3))~qfqH9-pi+*)!BTdsb8WA2RdJ#*Jn7R!*%K_^~a~Q7jz>~B_ znsF(Sl57$vyEYWL^|Qph)dEjMXY`-P6@32QhK<{5VB?yHD|5^7)At5>B;BANcZ=!W z&qGuj^quZL>V;_$&#>{`9(+Glju!pxoUAy8_geMm^!mP1eNX1nXYyh-bbaI~Bz^`u1ld1)xRb()Cg)ZGZX8;|z?%5m1U5YMjU;L|4? zjJ&uLB@JdsxA(!sMt!{d_y%!{D{w6T1*&Fz#>rv-U|3g&s>zEHbi^Jpw>M(%)m`W@ zUyDP$EqJ|#HhX87a#<=k!}vZKJ}e`Xf(r6B@TDJGDO979O*#j(;NU+RZq@eKy3G_N zC%U0X(+KCk4a1B4(U`429=Xo*dB>KiTr6+IyVhhNY}|1a3>=BlH$i-RdlH8#M{_sF z1^9FIKddQSfJ4oXaIaZ~eQeeF;o1#&zT_x!?;nOo{(e*p_r*A0!SQQ&39kFB>6(ySU|N&l6Gl;qY?^6G9YtyQy-WGtpizui0YqFQC)Z|cfe-Gq$nAQ6p*my5Ul$;&&df z_Z9bFQO>{0uCnRUjokKTBCkC%jn9U^=5EgQ!c|trX-lR`BcmOpd9apL1Ye=qrjmU> zyyQ{m4)eLx1iqRR#a=63!leBp_Bz#|ZHlZEao{N@xV_>!Reix@HRZX<>O9hDlceS_ zPm)O`$=bV&w|IoHmQOH`t1{&k#_HUdE%*ar3pj9!J@1vBE4&FTU9lS}jqW>An%_KA zdZ_F!d7UL`^MMFy=(z>b|0p^Se=OfNjLVj6Sq&o@NufOFao!5iQZ#5O8cG8xX=q0z zB}8_EG-yd_2@Pp!Xs4;ARMM}sXuQ|^2h_*MbGx7GI?v1koHu_ORb+8*yEx+?_Jo4pjB_6a(fVWi1*{@)V};&cM5kZ7{(XoSg_C2DXf_w z`g?&MT<~)nzZ6c!+Th*XJvorCaE1URypB?;FaCeL&hOlyT0RJd#h0I@_FLFE9oB5lcMIYvE{$Dv* zNtu6~{D$9AD!l!Sk~DGoSB_euDpe@iaLHsl_Kb1kW&MUoaXlO)*V(A8(J{l z-iX7tujQA~d-%zLS;B2Blm2oqDYITzvX~UaBPR>z@0KWTIkk*;c8=w#(-PQdhMBa= z-AtP9tRk62CrEN`@zT%EX_C|IeCf>h0;wuzjg;3j21>)?G3Kr-PRw~oCjb6WAJHXq z8$Ao@AEVLMa2V~J4|4~>bnJZmCOZk{;?#(5P`UmKb;k^lxmX`R%O)YdS+IN^Mq}s4 zG%P7;rN9eLs0`4@(h*bmU3?B(G~ef`UF?wf))bpIGX4y8!sdK)?5i3N8;7+Rtgr=j zmAOdqQR5)Nc5(<-;_Xr5Y<#eR67GyczKbo#I?m(^wd>fgDG?(SbxlJL$~G4P^6Q9kPl|~`PBkZXDv{-Y8YBWf^epx28BP$ku&EX zW=QF18{GvirX%q+H3VIjhN9x7FJ?@*jUA))xY}EVea3Eq{;BntZ@nJVww$N_rRnrD zwvkev6m!wIO3shC#=kA&@X&J;j*0uls9y}A1i$Bci~m(2IniDm=T$Wz|*m4 zTQwDy??gu;GYM+7V`0+1GnSi9gndgKM(&k|&+~m$+kP)fpU=VZwmBgCJVad-8K%3= z=<&1_v;I4e8<{;hPN6?))Cc2P|4c$T3|S)njC_e;SZYn)&`&Je$$kPsZce} z#?jy$@fj}1fRBsOyGNYp$&ZEeR^cn_Y9%s*8JM=a9Kk)Wz<-tKy1uvPe!Xq^!;q;s zC%vQWIXlU0X*)c7D>#qOZ;^dUdziJ}r-D8XxcA_afD`t{a+9GjI^7Q$OB^77xi3nm zP3KeZ9eF@`4>mOOhV1Q180D@N9N}2b6P&tmhy7e>qL1lv9xzDr!t&~ym=_w8gd$*)GwR0{qI&vQ zxDQ^!m!BnKK&)Wm7Y)Y0BTk4nJAtwt3Dm1Ko-Qk?!SPf&9Vug&tEVI4cOmaSE?ga> zQdmF11eY78ps02UaL)=V z-kR7_AaVvlGhuvD@Ega>;->X7d>tHzv@M1B9zUD*z09CD*++2C!4PBZ4pUW|4yxA< z!}DbukT+lo8?D?$vvoS7Lu(T>HW_g{hn~DUpbNXW4Mc05aO#@IAhrJt%6rifig6+n z_oR?QGT+gsXM@nPUL7XRIdmhX56UKWXCox@(bXH-$6yzt)b_%?OqJa?jY84Tv-Gav z5uxjN{Hkq-eP_iE#;OG8i{-e(=6Igr)JVRO2I#fBGv*d)AfjmFn43tQ zFCU|?k?ZNk{z7W~H4X9E9+2c;(EF|Cd|}LJu=_gfGLYjzKDpSQ@eCH5gs)t!klLLq zq;JFH;k_b_bo&0Jr5^|2o>nPJ{AVDoE(op_M{uT9p2zof;&Fd$F-UzYJ-Ewg32#P) zqUh@1{({+lk0~rh^yJ^{hy3&5__i_uD_(RJyD~95)3(O^((wozmIBjD7VMZ1$J36= zW1GWz;Ev!v-LazFSv_c={C(ZI!QNj@y2xz*?V52 z%{Go`YZ9(Tw-4xJ8%jp+lgK7F5Pv3ppqu}+F>i7MrA{srS=YN1vdsg-tz+&KOFPF85Rz`bPV=pN=EvW7np)ma;w7Dwstn4{2cJ3&7buTtFT|L8%FZ=|$O z3koZ>VA<6SkCl(ng9%Y$ZdOcJ?XBTaT83;aM5=)W^hauwu70 zI@nyNh-;OkbL}Ff_=-Jx)D{s##Wkf+bJoCj!DG#L7MX&y z0f=&Ihw1ZrQk>g!2oT^|C;sc$ujy4&o_C+dA0AjB~?Kh+0R&7Wz$=7wXx?5 zHC?!yodXASvf?RAWt^U>!(V3(O{cfpxU6#~Q zYr7QNQ%gGZ%s_hiN>#d><|thrZ6ejmjHKQ={kWz=gBPZ1^VY{5CEvl_q*b$xrL0Zk zIQ`yqUi5q(?+A+L*Yzv-(+;u2{oYa9+0aqyTc#=1CRdXl*WBg)l3@>-e;B&#VIDW-r96xddp?50~`1H7#@gC=QE6R98 zf)+m=+l_19tMKbCtE7&)ev-moe`(R~jqJL5A$Q!C!5Y0kKlg0|0N^i(mBs-jgTn{aKZWK&0et7F1vYK?^( zxu@{N=!(L>0`CYL!_nPb*)!0W7hS05UC(dv!#QQ#Fj+hwmpe+{9gL(UE9IrJ&;If5 z-%3)Gk?`nN)U)yUTGkj@hogFru(Y8R$E$3lQ^BTE@DM|(c9pH9Z`(sko9rSruWTE99cX_ZG+3Wa=NhufT}w`Ce~r)lJzEiND|MlC=VdlZS{?B}_o3N69dX^&V)h*Oy*0b1{6x_qQ&yN@BxTRnl0M`)OA~H4pyRk!M4PJev7IZ0 z(?dA3j)(Ba?dDR7*v|*WYD$U6MYsNq8~;ci$79yJb76^aT7Qb*1M}>qfN|=Q!hiPC zk(f17g5q@H3XPGzTu+hY+J$2KFdyFEtq@DXSCtLSxTx@gK>W&SnH?L zp(jUaqE)V7v~Fap4rXjKcpnUwuEaO{Whir73cU|GsILlvLxQ$AlZzfmj3Q2(1t9OH z2SqGM$0ge>xGxul7WX`;X3WLetf2^*Jq6=s znE~cqI*EYok!Z@Aitr@CTZog2J&>vraW>UFd8pT@VP_tmV zj~VQNqiTckeCTl$S6qU*vMPJMoCvcn`536?jSEsMz3X}a3Cin`e|9z6kBC5p$w)ZA zET$Qc1#2&C8@i=!CHbE*a7c1NO?@`yZQ4Su#%rj{`L8f>+a@@9^DyYvAav;BgtnT2 z=%Kd>@5Eg4*KM4oYM!tH+My? z?*NfONd#L-f>GTcZo}?V^h_)K^WH?SzE45ta80(TaG{;MwFRHb6vj{OF{()t-m@Q= zJ}Dn(S~M}^jv|a#9-{unVemF=Cp-{Z82!!)-rZLtMpf_t?Ox(g$9!~LCEQRE8?Zxb z4rX7Oim0`YC|ekSo5O2JtHvE&BKl%?n-?uvH4>vm4|m_D0XQ^51@~?@P=WFq^wSi+ z8JD?eqT@LBMd^zDiJzWT^a6MPNVoiyf#K3vA9BfU&;E|%PeOTlDxNb7-$M`pqs zwjP~=(&5m&6nm6X*v83(O-mZFW{?WX1BXCkbq}Q8t|R@i9Wni?*hl=nPfLD?nev!Z z6l6RN0oIWaizv*UnJ><^F7WITfvG8-P^9^o(%yWeIpU6-+Te%|Z@G7$Amv^FuIvfK|5mT&bA|@ym1-a zzivWg{z4q}NWsEYukiZPBEi!i2XC=&v>3GsJ_lC7*4PIN2f7FcPc8~(RnX9X*N9sd zpz6H~f>YLD;#+?Nc>3YP;W5x3un=vFGZ8O((aq~K;OiFv?PrguwMm@ors|We?R|RO z#fOe%&!huG7SM=Y198tloGo_^#VOGdODJNDvWMtu1R_u`94R&HVL0QfLeX;z>Yg*f71(bIrqsA}>PJ;`nV&?@KYkZil3=V<2<`6VZ zdJB!hVn}5nNLnaz;@!(>`ph>JZXl23B1`zn|3`0r{1JJ8wW1HQh5Q~D;mUJ&%!#(b z+_xjKb^cLmURO__&aTH>%@~>W&=GX~eL7Aj+LG^?W;#00OYolR;qBnZo#b+PSi?$E z^*@L4Uf1~U+7w#dA%*@W`O>|>G|C#;8wu@~a!X8>Y|E=mdMA@-bHQ(zep8oQ{bq1z z(h7=@>%^>Mbxb0C>f?;{>sw`0dRhv;q#}3MF{V%74^Xwy zNBll$K~4SAu-!41$Gx_LUXcery7l3s3sfkkzBiHzM+jGTCap7$C(|4IU}iIxG<%2+ zRO~0VjodGrAUKqthEL={Gv#nArX5$L^=4n|^VF32lMJ_Pf#*&$3XTkb-QH+kzAsNU zsp~$Pl|KlrzLG4IAe22%7gTiXM7x$8K zresjQ$8QSTK9jQ&BWU8RGgxr+CSP}7OK10-Ca-_z2zp1Tz2HP8KR?23y{M9Bh5ALy9$u^>)r((Xkko^^PZ>md zFAowQh@camOL2c?Bx{vM%KCj7LKX*hq5G(CdN9wI&NcPK>C)5Wkb9rb#mnL7_ZO&| zslbu*gLrDnKJ>d|&xbC!@ql5Qq1R5_H6}%INZWGiK2a5AeJNCudS*;kIc$SV|)4`ISC!imEkjh+NQ^rT} z{7ea<1xHR$rsHpljOYZ%*G8xkzN-^U&x(A zg)jwNHA}%(k=^NgOC5h5Ps8uwIoYB}TO?=@zx9%*IU{Z0HK7vLy%x!mdt9Z5x|)1= z$RC-rl{I;a;$IY~6=3m+D)83dn`7FGZs$?;f z1XFUCC;RKv(IS_#6tQ{`t~|Ix2o6T*jZ~c6a*7UC>_Dx)E@%CxPQTJcKA^`cp7Fbm zI^2DL)xmptV9yX<5ZQ&JJ>_|}$#ne8siX6g28fJ|9M-w_$KJQOkO}VnDB*`(IP5o% zuXbhy;hmeESHdU9rQ%`x38FhFasbIGyfD_3ySfbIxKSOkYgi}jJbeJU{zstEs=@z_ zh-NQ^Ga`e%5E{S7LtXI7`|i(%x`x&3%w_B@#+*jdgnQ0~s^Hvh1Ec^s~+7VOSES3cF{ z4^$(I*!;+6a?xIc#cM-ZPw)-f?N#R?Dq*~I`W%kiCdY;|%y`IvW#T=y5v~g+@Rw-{ zm^|R}1y*36jWYe&8P6A#| z4r0SVH$M3?g1?Ha@5@w6R#|X~+lM5=qu(s}c>4fh+8lJc7{x*VkeZYXKe6+6`z78@ z=gm;r=QSl%ug9HDA~$?iJdeRkIQyOs-xQhTAvto9g4P+0K zi7eJns2o>E^NV9>JGT?=7JuN=vkn`FY)AS<(GSRuV^85pd}&ey*|B4=84=I4y${dL zn#CFY{-fym$8oPsk+U+5S+BvAR}NL;iPwHmdW0LsZ#amaqUMyI{tSkh*o~jeT$y@KDG-?fhOIgb&PiJ7`Z$o~&@hhd?(Ll54kCg1v_M(TF@3)SGp7T)b&9mliW~!(>5suA9y}4*_G)S=& z!{&!#O;8Wi>ny;S2NtX};u11qXW>!I5dN^ulC?!o)%sQ!zH+k&VWNLEBXSMayEUTA zog=WdD8cZz`{4C45yojJ@XJY@u{LhPjAeqw>!Xc{ZKJV?PvOgs2J}4i66ZVY!{Iv7 zf2@0j1IZR_v_=WNGGj0`sy!EmbVr{T4rsG$LyU0%I!|Z(?z2j8yLQ6)uNl0|C!;}o zBTifU;Ng?0SX<=6?}HThqVUv@eRLixDF&;&7r09IBT`Ir}(VsJO z&BW+_kI`wvF#dh^5I#&uMzy&wE;q-}f)PWIG3qGN3ZByLkXXUGAHn(la{TmE96DQ^ z#H)u->D1EcI9+s2^zy>lBJ2pp4l?Ai!*0N_!$ph^8^l$|`*G~qYN-8*43&k-Sd#8ZEj?@~aL_c!Pu+%-rah^>LlZrDR3JON zu$wHAE9m;7Ci?X55pAEpp2oz@#^TKy$S|6Ln;V_T>ccl#7YA+RrCyPZs|l8+yojZ2 z>y=b6dkhtao5N~s3LL7}QlCk^P*M?2H~ZeDnS0-0f$>k-piAKtFiZu??MmoUxC|S^ z;_=hL0Tb0jNKtm2*y5SYt;~_Of3_3%k_8ktzBBE8WebxxuGC|XFO~ZWrs%UA8AjI1 zo)k}_ne9iB<&`Sgfjt^%v)oT!wc40jz8{N=9JtG>G5B;~4!gGI$-@8Mm+4=6O`C3d zk?YGkTD)Ql5F!|G6Oz!p(;4SHqXnbmgRE=9D>5nDMBd_iFL9wS?rb-~fI0eD)aweY zCMi&aXQgbR{*~CnG`*!0 z^E^0pMHA*;m+`q{i^*Pd0!8P1LG~k1>(Lds(Y6)}1G>?XZ|kVh@FcBhJ4;?udXkp# z!|fO`lKg(JAyv5s>a5vFCuO_Pc;EmEd!DAY5Or}66fF9q6KPmKPs*Rs36@17cYDTz zF7)UM>B?G)E}BJ7p(&KJL=jcb)R5P;J;$|1)0eKhWW!7o=|OEB$!{1TONgF877qDz zp=C7{cy@*Dxz!XAQ6!t9^NdE#`bQTMy3m+~+4REH6r)#dCS9?Qn%QDWuE}fZ$aqbR zm=Yy&X)UrZuCDkh7fpro;dDuGgX{j&q9~2Aq?^G1|TP0nJR@jbU*IW&aW^WQR{?(7#+m%v_jGm-f7%L7mspSXrg4n|3Uv`03HW z7E^jz7>Gk})rc}Ysm=H`$!8rP#~cq#eR~2QH>uH9Xi(^=zjPtB9F{$XQK-W-de@^j zsYlGGC8ZrPSM4u7z1E-7FLj1_Ne&HO>`&%n&2V|wGMb)UCF8W-)L)Y*rgf1lzkDAl zYPZTh)+NyHAS1keQch7Ohbh8p4rLD40=q;bY0o>s49%cimmYNB!a^i3?uhSkzi7`CVLW4aS8lvKfV)0if`1$O z@T^hS_>q+(%DXPW1Kahe%AA6E)#0!^Ppt8#mU_o0!&#Sjh=&c|{2?)s7i$$iARY55 zc(uVD6;qV(;%_{zMCGIJ-9kKmzYV^FPvgyPeV%z=^tzY0^T`v=2>fLQwUl#MCs>ZB z24wS>Uq86hpy8t1T#a^P9JoO30@Wo=g4q&n?yNTn-=7n#@1;SeCwx4<^Z4{51?gXS z7+UtXBD7)ypD|;OY>VV&mxCFTdkPmqHG*eZ@S)XXQPo$0rC&Mx$Z`|@=?Uj@Z*49v z+CpKdq^e%x-BRJia~thBs!Hs1)g!RA(ii1#VxctH4zck*_+Y;jXQrv)%b^VPGFIZ* z{>dyGc%3INdBGaL9Pr4Y4-O8Gz|Lq12WlUn*6%w!R_bB--O)%`l!R}|U+8eOJM6yQ z!>-?A7FOSlcQudX3j_5qz*e|tOa2qtk7&G9a>w3V3z6%sh|;c^7`{y0-99S9+}j#X z&lcg}$|cAikqzUNGK_E2VaKRdeEG~)-f!=V$6_zsC#L}8S`6?mbQGK)mSUn?14=px zCc`^()_fK$W)tyvo&65IUdiLKRw6Qr)mXzWi?7yaa>R$@T>Ui?EgliL@w*TTEjgrL zewLmbv4dIjS6b3C6X%9GLXqXLCVoB~2mJqxZH*_#hhw4m4k;IT?+G2FF}7U~HV|cc zN9P*M4^igmPHq@_W-uO&--c21fvJsR&o3M|Zd|dx`VtswTl`cVL})EBW?&k#W&tq<2UWe3m}AyRTYg zc^x>traxb8p3j}bi+M+EGaVAV?G7?=Cw-`i300#IAvgnDtu1k1%7M!7YUn+2L_qI! ztea2*&+#re?>-b2XFM>?pd2qxEAXUSCHVPa7Tk39Lz;UR?qjB) z;L+<2)(^DjEy>zg>J$oT?+i@bKasOEJh^3V0B7x=gdxUzP@{DlDUAi3=$6Y_H+Jyd zOQUf-IvAhhMq>Z6eQ<6rfOcgSwu^q^eJ?#;?YEwt_KRM&!yy`F)L;Y)}%JC6R-ub`ku71m{^ zk4>+^V4mWGtX3``7;Ky~X1#QgC`j`MJQ5%y1fj*tt> z@O)4+4Ytw4t#M^I8LEw#J0HojVVCGO*`W4sJywT2AX)n?s%|+*kIIkYjLKcM&-SLo%x&2T;F3esi_oSTbNE4!l0 zqaMiqxtn|)Uy$9pIkfy%G*;VYV9CY_Xy2BG8*-T#k(2?q=KiRdp#XEeuK3pQ3JXKe zVs3T~B2AZJ?xa{`E#C*to_oc7zW@mqVc6N_I}O+UkGjmfP4#w}IOv~>WiIn^M7Y{k zE^I?v(PH#%GUV%XC!%4BJa6__0FJdDg3is(!StXrOiW1YqC&CwO_ zJ}t!#XVEXP^?=HG@eH;kqyESM6u*ta+QWsIKE)58%QA4P)SfqN7=gD1(~uH;8q0L{ z)6Z@%Y3B%StnZUc6`97u!?B&DpfK7uLd+tsuEa#q)6#jBh!TYr*!y}T5`)sPr}Jjv z%07h#^|iRvYpL)*{h-Vf;{9y+13}6DZ2L=O_=oR6x3dzG>Mr4^RD$OrCGgqzhpLme z(Pz!2H2Iq;N2cUp+0#;Nj$e$`>J#8%n+CHjx-c~gB#-Om^h5M6!U`O4^uS_qU~WKj zQY>zox+29X6N^lXpwM$9{B9+~VssK-h*@rjEgi7m{~s9`4k7p1?eTl0A+98MhitwJ z{}yhPF8N99{j4eJpS%|Pwtm{V!na`{%UTx9V_Z`Rn zO9P>uy$1FFV$p7dx0t=Az%|qq5ygt|`}3Ddrq0AC0|Or7bBG7O8^`VE8sM0h8%{3i zfkiDf=+iL?he~fC!KWu2A8deW)L+Q$O~!h|nQ(Zp5GlRi(zlJTX-7vN^d2hu*p)93 z(vXkftA65nOoZpFDjfPPm^M3q(a&gYocMi^-VRA8pD;Dd8P-5FsXcNYbjPOfe-vl8 zkt72#cj>kbFAAArEQk=AG2^yM}oCXDZb)M#XfTY% z;aFc}yXN4=00$m@NQos!!5=8dg71`6$SzbN^4k_=!WB9md)@h+SM`vYD$T%C(if~UV(PjKnKCLp!uIZgR$4gX_ehqPrk zHLra^U*2|tvT%?I*A|^$ahCq8c1PP4@vdqW&I$Wn^o&kYFXN-6k{W}-w<6*8$ei*T ztYLgY4x@wvvPj10bmGtQ53O5-?1{d`JQ2Fkd#ursV|#j~_847EQ`VuODj4H#C9 zLA^R*(wdHVRXPx}(*Ds{1x<8Lw#E3M?fASb6Hedvqv_jCgwQr|&K!#XtpoUX$B4I% zZ$+O*;b=@yIAYhOQ9RA&BlA}SfRy$zwYnL7O{q`tJI0P_gd%_exvBbC*d@y)>gpC%Pkn^@(PD<-o+~AZhZ2v@Ge+q!sOU? zObr@=2X1;uef1L$mR<)%ZG=mZBO5PPU^A6YtS7ibWvZg5m|}&V<0G(ft+8<9DB(d- z4~){*#Wr69^hkP%YpSUTUV98KPn~&6j5<&GcN4)UpJI~n6+~E`78ws$z82Sp+0Qh2 zMxUX)C*up!xeyDSYO(6(eh4}Unm=^pE{ESkMKHdW%yTj2-|Wm zQN^4y__B98GOVVcLD`Pw4z}lcPgMA+iZ#!O=`Wa8Hex^32al$vlft@}bZ*FFOm)sh z-|!6B+8)5K&GFcEbRI_CIE>?KW3jtv2y%UGxJs~hQr{@>x*IAyFtr8ce`{f<{r_Em zIx1z(m|AVdhJT${C+{sbX7zwlq#<%1KcwDY8u9FY9A3=M!avM)k{Mg2{Orbv=SH z_hm=ixBo`FG=89sE@EcfL!7Bs<&HO-(Jw;mrE_lKTv{6$%Bebthgf#WJeo2ZYa~`TVhy{ap4JQt{kgx7{q>N3hY0)3$N}xOE7|LP|#qD z%+Pi`-sC>Iy}gRl4Y|;Zl2VXM{a(Q95Dg5TG`^zkf&)Ph7N9aCFkF|ap@?!5MIEhY+ZJ+Ph zyWu8Q^{>LE%q!SD>kyK5M8kNK*jaTaP(XXW|D;8DbGmX>q6r_|smU9^o3e4QL_}Aa zqtn|qaSzT#WNbDTlm_6w_k1`%P=S}TJT83%YR`&!m7OUPKAyv%pwoDidIp+>Zh}Xm zz(zV&kdzoerw7lTbKaH9V+T72G+KU8b-mN`o7f3XX6f7H=omB+MO z{Qhn2gjZl!F^(CRW73Zd{FxaCm2P4msd5xSI(e|4@BlxD-N(q0Z8)H=j(sys@bsp5Qy-uL+@L55!khY#w3-Rq*7e2eldq}d+&dcY^@98qvUovLBwM!I#CKLF zV)W-i%o}+Eo}$;eIqM?siZim}hJ0MSQi?J8&xOOJ6Hh&yi!FZ}@uun|suUYB)Vv1M z?=FG&+IaZu6<}HODfE##a^s*nY#o{gtQFFDmybNucJ;tEg$@|sun+Hw(y=wH1dd5dF)cfQ?|wRghMHxtTIvet zz#z;oh(@taCIk=@a(3wmx-HyvDF-n#Is~DgS76_~WoVz5DBkM_;dto;q(1McAtV6( zDiww6M?AZxro1|A8Y}yxfx=ec&YMz<&1$0g9-qnN)_qDm*(zL>pD;471DhZFkKFXH z)A`+U!bL3nKW5jkbU-n_uG|XSciFJ7+Kq=^x_GiuZ~>wolg#58%6F=A`7kqnnv;Zk zvo;FnL@pHXPiEcVM6TD03BVF4nJ8 zME{II_!e#im)VJUt-KY_(~I$8-xIN$lxNk09e8!D8P3()AaBwJyb-KJ{i$VGD&K%P zj{C5w-z9v>E{EObGve@|V%DW2aKRYV+up!puQoADGh_Gb1g!`wJo=|2I8i0ANH~wQfEy@sOh>@be$er; zK>4^gxXebZdE9{i*=68qUlmR{bqlL>#-b%uWCDvb@xJ;Ty80X!e`f~~_#^{%Nh5Hf zMR?~vlwzj&cD$Us4_kzjQG4$tEZVgn3xY1t5eGFCJreKk-nTJuR~9nEpJP~k8~u6u zjJ_+rrk(xn(B2zA$;a><9j`CP@vi4^{OTPXJeY>)GpAssejZw%CWo4Tfy2Bk3|deC{eokN z5%;!1p4QlPx`75gE=A$6?eJT87E}K1#NXWykuoM9PamftVPGkIL^jfS{89Lx7a7(i zq8rxk2)6xsjzROc@{2R0zC9gj=jH>j>Ujf!#`TD{6rJeSQMf4>pw7NBzHZu^ zkJS$3Z#HGz+xQW8Tl|!rVi%EhvIX__Fr=B2_T#>MDrS5US&I>he7dMB-wf%>6S{+= zMmVwIKyx0Gew3yj*iDCKoTa&4x|5q`2YNGzD6{sJtaGU@$tf6 zaW0+CPNNNN`|0H4Xr55KhZU7Kuv7SCO4{T`TMmw=c5`$%$kv&=T~_50J{H2!+J(+q z3=r)6GEGnNVhHj?4{L8LTS9+m6YbI10jd})0mXGVI`<)cKwyTZtMn~LxhR?8aXmC0{$ zJKA!mGp%c}rigiptai5@o9nmZo>sg0N6b#%_#l~^x44q(n}uXxKAJS2wvu6srC@wp z;mtP#p1fXJs)`sGVm zX`R?L%@PE|IX(FPtN*a%(jyp;ZbOgB7a^DORF;zYO*Z+x5mj3r!@{^1xKdaF)qNpc zGI%5Z_i-jWDL+QQ{RSlOxrv`{vr#NElMYh`VBE(>+15l|>Uie6?3P*6m()1;! z)g2`Bm?QM!^f3ziJcb)K1hJL;2JU$~4u;pSpm=Wy+{{x+uG3UHv}6rg?$M$JFBM7Y zj1ryRG=ZlkO00c(4qNCD$KCK`F7RffrLp|tY692P`SOcbBY4y5A>0xe#&x$2kV?%# z`ub%jHEgbwoh#}A`=!$@rEd4bVFdr#w0KD*sPpzYk*CqA1L7tVa8L62E-l!^QW^ zIR1kL-x#IIYTLT-xNo4mm6NGwfGxQ?3HG(V701?w@~y+k+#tAiBNS8kyrjV)Zsz>W z(~Z}^|1RS^BYG>AXyGgM=ts5*Eh)63#8)-45v)d=4J=66{JqS1x&^6JE0C%BIqW&I z4G-32LA_R)sJ%XI6yB)Ww(Tf=zYPs!clvMkV}w{3@TLKcFtQc%v|;72ULo?>Be$VB z^9)QqK4bNqcof78gHB?MaO`SOfKxw8$(%p|x$S7(_x5z}iY-;No|Ana-;IpMbta1r z9l1n9os%QF^Vd8D8nDit-WC`V_a4s|O(YJyg`= zUJk!ap3G@|ck}bg3Vtqfyhf8IW8tOl?&H?$twa3+2| z{se7>r0XU99xzs_Kkp)){E^RA1{ZnH>r&qJy_N^hJH`cDo^ac5Z46KBidn_JC{Zuu z0NG(KwYkb;mw!O7)AH;-G4(@!CtGf;P*eS^5uDztT^o+2YgNx9j7cFmUx+O zSx!g()3MOa@P~V?IrrV;#t%bhuxT44n=(_W@lGcxZ(cBzjr_1SEgE-Mwnt#&cY0>s z6_(Xn@>o{7|@qziwu8`ekAvrV<7ES{K5-Qc9Tk>nL_yW6u|b55oIiCf*4qYL0LUtvxL#MczEik1i@onq%}} zloE>#XWd{Ib(;TsaEO%~KCrXhCQi&xV#UaC(PdY_ku_=vQ!>N%*MeV}(}}AcnVXX; zxz~*^96sU^j~nL99y5CKH{~Ga7j1YytR4@oO!#5-5QK;`wZRc>@!jjkH!XYezI!pe zGIIfborpri?ns0Tz0E2GpV{$xdui*=PW;7RbZ1Z4ah9S#($jol@_8P#uSwY4*cCLS zAKv)>0MU`mvPRQ2#=5J?T(cvW9Zv1#8o_Ru^QeSfrz=Zs6E^Yh(tLLM z_K;)ho9J0LSNMf{i8G&sejP>c^~^SOg6J!n55TXa@z^-h69rXW(e3MC%)Qu~XPz_W zheHLs)j`ZJJQPsWr$4&1V>Ha~jK|Y~ubs&;#5}uV8b*1sj!3K zztwyl&67%hVUd^vd8E1V<7Ea&HnGQ-QBKGlW-T0$g#S+W$BU*Ou*ow*>%d+}ZpmeH zPUO$Gckr?>;lVvT6~p!CgZ7s4v8@OAxXxbg&}9@xR~g}`gZNn%zQtyBMb;Z*!Y`Av z`BA`C-ny}x3#KIEvY07U)1!PD4lX7o3b9f3rnKI2}9QpTUK2T^t|%jHVeSehF@K83=kQU8KeC36i?_k^ zfD0@Cn+dyHX()Mr1cSq$)5$hlY*?iViw)zEnLh@{u3v&vp+Bw}jzi4KSNI)!4+lny z&-%aszM!&;bbqVi=NKKFoIeB0TV7M(l;ya0=MG(Y9RSD1?r0=^F6gJsNw>E2FZ$KG;F^#G-ydB8cG^c*-})BLI@clTTv7x zg~)e*{{YqdzR!E#*L9x9VcEg`Zy(s$p^`YaV;73gs?(@JeQ+KwaCc}j%8wMozr_Wc zYX|as$x7JsS>)Wkqp(@6jk&SUh`T4V|!Y++3NBVwe+-$Ig+ADv`%#WeAjw4jjP{yPe6foOE zQh1{_77OIo!X`Hz&Hm5Xs_J#DZvR$hs(YRtFX4_c$D`~&+s#-Sm4;8z@p$_EAgWJJ zN5TjlT63Sz^uEsK``Q-}Hwj~Q`Cr+}LuOEzoyNSBMm*UXT5SF~CiD`|KhU2EG5aC_t3)~NIv;P`<9mlgTyhoe2W*BOhXR_D9$xzRDEy~W8g!G0UR`#`qssG%H^wB0k8Wi~Ehn{4m7dTQeY$Xa|drikN>a zjHxLK=xfb_N-XE17^b6fS1eRghoQ-HF2)~Oio@rZpsMi%X72jJKE#CJl=eQn_RfO% z$Xuu?MnkWFXK8=5v5b^e7&54wxdq>38&|wwHr?SYTQi=esy<Z?npez{x8G=YlPMFoV9Z?eoa&+%M^J!K2nXEl76Ad@F(h) z4QU5yB1=9YG~#fN%HoQSoKH!$zxAS%fH#WTr+C}hrA z8ZohwdYdw+;Nl|0o!^4R&iw4RwTqaBkD(Wb;?+Gg>XK1sAws^nX$xiOTyVh!jk zcWnRo9ZOQ<){u(wLAuEQrt3KaU2mfXDa@S=uj8taI5Po-zxmnfgorS(;UbBTJ%gn9 zYxujW5YbZ6SRERP549PoPWL5Qvi!uFSc z@hayF4g`PauT=;fzDC2?&K5ft>XXt|EfT#wfedBd)5P8PXqrklx$12~kx?%C9Sh*& zE8}2avW~TvotB#7`}$j)c0f7DhryuOP%!V%&GUAs9>uxK-lu2 zg&x1`f@WtW8ty+wtyTr}7bQZXw-gH=hLd??Fu6CC(3UI1NT45>VcLsN$_6yYQiU#c zYHwAPmzAsKRFG5Q2d-R4} z1%)*r=8w2gTYHao4E;gxK8Xk$-Q_aes?F0^=_6V*Ibrc;j{$#wHu zC`Q`A$1xFy)t(``FB`Y!+=kYYBS=4!2A%C|uqEpORE>_IYT8W*x6Mg4uL*x9T2S}m z_b?6k1_cG~c)Rui26bh)I`=)#oGa0$^Mh&923Pt%i~oGbY9W90I5N2lBxhh0U5GeE z#7Gq1&*g~<=T>H5#fc|a zbm%k`j^?6paSF!S2jOCDA!>75u=na6IL`RQofr3UpzaPvJ@mnlU$JniT!V_BQMBWj zCHXyarF$3T$ysdxeR%Q!f2>SsQ{Yt6@m8TfGh8Uocn+EQM$?*c6X@&29pvf~PlsBo zFl^}+oa`tbM~RF z?-c!u$)e5r#U$iCg3{vy*sol~9g^=!I`0b29Xdcb*PM--oyRbB(r#RvJA#~SmvX;^ zI^}evlF?&+rj$vb(55c@Cm}{t|8XbQtZB5dK%cBv7}67^XE2JrivH4jkZ&)h_b1aS zVof4_R*I&Fb5GJL&ip()-UKbmd$IG{6TA%n$o!1k|&*$&I{{XT% zci7JFbNH^#2A2CwXz33LO1^v>z5M&iH82cG_THS!tc9YnZeU@qSTk%NYfBylf4R}n z8#t1B%Q{K!+a4NnI}U4GqfzLUj$NkD(Y7@T2^&u!x^OwpigEAjMDE4aJ&Hm;H}dJZ z4)dXanx_wR#cJ5PfBd}TIs${Q%Hz*&?(o?a4(YyNJWyPZ>bNzybtea{CBNAs^WS_M zz8HTp=Abzx9@8Sv!tKi~tUt#&)nlwl)AcTsS}_i$H&fV=JH|LIZiZ|PQ#8IGh9GfK zDE=`+On?<;f4FQN{pbx7i4>uLx%HSV%X0=*oPAX{904g4Fn?AUDzAqi?_V+`7jZt! z>3(*F@9Ao+-!ZlArYLFLj8CD%VBf2P>mkqCTRtx?$ymq|$Nq!->*?^8zs3TWIbnQ9 zE}DjB!^^x78|oc+X5~MqS*^jIn910DWiFnMbVc8rTy`LR0((@m5$;;iD8Bv@%Z5i` z%$UV^y!#frJCM6}Pn=?Q17EQu30>6Y48*j&UuFh3+`Z15KXw6}(y&bT5 zlroY}`v!ivYRG)REQ4p|Z_D#hvGJiyTud#sMYux9{g zub5)2j|j$yR5Ha0bFF{Ryn|@IH=e~_jv)9uf&?Md^X+q?o{f$N~CF#{7T zrohiN9G%X&$jaxgy)IqqHuHwU9PXu+jKHH^+Wekgz~uMnqBJcLvF;z(yc;)}#kkYV zZsr%(Ffx|KRY^fQaT3~k*WuZgAee?4z`0QoJJ-m-E!P0kJ-GjS_-XF7kjAUKHi)_D ziuH2Fc)j8v&lztSBp}r8Mbpx6Z2E+k&>9kqk8#@g zGm|j3Q5oOPXhYT11oO1FL1$Avo^Y4*UUw51SEN8bL!54xZ^3qtB#32?AEX1~;y!$N=IhuWZ`{9|K+5$^6z1kcgW_$3Z<|bn#uP1K`3pO| zs0u>Zso}U>H-*k;%qGugCKMLH_krAzr|Y_b6gAgVP?;qqp6?V)lCub=XklS) zH(KUaBBtUCXEk(TT+l>1RLl1ykuzwk$X@!sbsQ}mbQ}8|`FHi~c+`n$2(zDDrM~l} z6gWJGM%+uM+WJVcNe`mB(jYR}Gm9+Wy?~`=IV247cz)SIXm_40%xKpZZe6=U%gQrp zm~I?7^~G~m%?65cwj*crm6SGAi`Mkn(x|s(G*tKV>6YP#%E<)7^Ym>WcyUj?ts2c=|MV8~w`t zLTSmOLZ!a6aB8Uw90IwEv%jCMHzFFIB2KFxUqwm4ZdAs3;jX0?YNzg_;M?nIRoNzb z`P*DL-#tjs?h_XVZ{JIP|F+YOIU}gri)i{mJ-W9+h9*7CCUv`TN`fzKe)kLCM#@qA z;1Yf=QKI9@s@P@;+Qb4}S$yr*9)o ze%VU)8y`~Qt~kn3dqY!$?$IolMKovSDLl1If%lVLc*|;VlyjvOS`uKtZZAGOn~wuS zCcu`z0OP!8Q)BdCvL88?&Z;_5fbtkR*YuZvS4vUixLOR~+y*hH;}}1p6~2cLkeuQM z>Jr;SM`AY8tdel5X?=y+=Spe*)3YRV^*D`?+d~e?Q%QWOJgIstgwGNem_Iy(+41$X z<8TFi-WNqj*cZy(_J)dIh0)|Q-n3<{IOzl!QV8cR%&qakzCtTF)!yY?yw_N+n2#4O zZ?Wb64~!Xh1y65(LyGzZ_(mMWCks_7reC3g94ZU-zgoFV!t9Ze$&A?|q;5n;^~R&<4TRQA)Kp22V4 zPyC2Vr+uRBwD6{waQ^253N(=v8W;7^M6WQxTO~ob_9IBxJw%LDG$g4rmS?!cP3hZ8 zH8S!TOY^*RDen0q`l8SGh$jq%g`pzCn}4GO7CMyl`0i9N8%%o20Yb^bFyZ;)gF?)j zI_ipkOkI)pN%G-!-Iq~O=g1kdt>3r z#qq+9$nUG;x!=t7GS%Pgf(mCNX6s7Ocl$Racl$J1S-hgn z*Ut*C3ettju@{BEE?-gdRE{=TS3~A)Af<%{()GA06u6@q4}bG{UBfNR`cOd=+KMRC z=sZo{<0w?dSP6ES!-aWTne<-d4dp7P)0HOyG*NUX>3qndzi&i^o+qg^C*e2g1kNMv zDZ6QCq7!-cZlYLz$I}y8L-%cMgkhbN1?gAI1YdtpiINh9@GeYmQ61?IjH9)2_o@2l zN51l1*aUz3Gs)f-3xfCI=AXK<(<|0&%$fd$Hx9D+0 zEqze5r45BM>6F}Reg?0DmHl(9&Hf2p&UiG8&!V2sJrs9bnO?7v;qKERq!T<*IOs4{ zc>H&nuxD8|1=&`U&(eI_Fs~mb!9xhPouCOSwCB)hy5A>8b{}>K#b&&2V!u(K(Y8VZ z_ZeMVyh$jLS|hAzohckU7cBgEvRHW9VImkgt>E`85t3ByN{Z5@^>e1s z#QC+zYQBu_oxPao&`(`n4=McdH|jM^rae7}NlE@9#YpN<;1fIYcGIMq73SpFYeEx5 zCs6&wv(#YKK=)?Hl3q?8&9cg;juFRcP>z_O!e4{`UUbpU!|FoE=%19BXe2CjkR~11 zPChGqjj{6w2zD~x_+CO&nD4t-u&ZAuq)5ySJ+zJURvVyeR*&); zMaur8Ov6^l(Vjj_YBA#Tq9T2gRFWirS)p6Amwrkl67DwA$*7Z@A7@UM+b2-x z07jwgDYUiTV@&!dG^7?&GiPRvIen3S9y^WJ(Q+uHQWtDulho(mz;OQoyMPiUHaCB5oDMSJU33+;LQSav;8kkMZtJUkgMoPS^= z&3cp&vg6>e1JyW_%trf?CdwqUKydXcV6$R9~DY%$2xJSDOY1 zhjtkYGal6Aau|2#2KI5T0QdRLE~2u{pUBHngAD!UNhR8bGBidAnZevqwaq}-s@aLk z1z!++SdLm-KjAp%ij`I{3Rze zEiLLK1r;-)<97%-MDQI&SOpFEYfdjqH`7=(Uz)yS996_CQv96hG%-d(ICA<6^*WCb zWcJOXZL0ip_~bOL+WU^bW}?E^NlHRw^fV$@H=6w~jAYJiqFZVq)Ux|H$vqoR!)MCV zuuN;(Q@IbC_r9a<@;_{^k>uR!;ew6nP$7P479FU(Kr8f`Xwd6s&Tweq9jqZF|IUw2 zPrXD>&7P8EaW9!4low`io+9j#d4OJpZ&<-|~0ZRJdpUzO67&x3>~{G8{tXc$ei9YwPOr&A}JPq_~Y zNcZ%6x)?N%K0Aj{`Gz!#95IukEqthF&wkSS*aa^oBNDULqgmE6!gxm=A@q)(FhRwc z0^YdOlLj9OWQE9#5~cp(H!;1Y0o5V(NH)}>0YW?MrW#Xi=2FV)xQ^Ui$neyAqKl$*uV0x85~tFT_2Oh`+l_OwL&z{mjaG>n)ADjgUDL$`Z%19hUD-@< zo~SEWw2lLGb z#k1(vl$!_`{{pK9_F-t)H1hQeApe|rnp!lS?4&%X%qfuW4f@Hs3-)wRf_Kh-$rGD4 zgls#VD23-tet%M>UvDQ-*SRqA)#V)YE$8UAvo8%)Iz~RemuThV12kyGIlAOhLO%bM zW6PuOSf4SJynCn6s_#)WBPWSo%D$uV!IFaA+yR1~bTDO#x{-wbak7r+zz+VLzwFH& z*$HoPbKoK>c4YKoN(*L~4W+klSMls|HEd3OgYTdrq^E5~BEDWU{gMkk&oHCt`;5YW ztf#5Wo_u|6=ww|m&2?Tt2l(!IKuQW#4%HGrBM0c8XVSmf^mPf=wtZ+ni%{5 z5gmg`T_O*br|-b__D>|8Tt*A@JV@a{9J%am$F0-%;ceH5FTb-WO1g~1+$!j0$R{Wb z9!4*0H7Q6{OmJxLpdyWaN*(JK!{ zMFE%IA4clHAZF74kDZKiVREPMG4)S_5m1!RJihH@FH1h+_y8H2|LZswRt6*XuQIjC zNFr-$owb^*I7?pLVVw|P%`%O&aOX)4yB;o&0arayqw|#|=NZw`FDG#OiZ#^jjhQ&l zv(&gcuqjSP*J2Q0;#%4Uv z#ouUOByp$l7o{qeJ-(B5kKsL#h1|6iJsNY>V_3GGE^Au9ijAK;oc-hZ#)iEwpxn`n zc;{4vY~-ynQM7APGl?mCf|YRoj1(9ES??IP-9BN%V7Mg6bcd-P&8p5`#vL+ zi5^d4aV`7Vy~APb+_NwibF7|y6w9(M{W_7oHoJ){&T9~DcMMzpyl2+UdCVktG+Q0Y zU^dqbB79Dx7_LYTN+U6sb0}7B>1K;Y$>Ob_8hg**1C`E+tYz?2HYDRWyPU5^dINem zW1P?b%1*IC8Fk#fC52;~+F86w4jaC0Av50P!Q{_$SPwk9ht+=)!;EFzb-ZUj%Zb)v z@;9GZ7qs$T#&VvixH$>7`XAZCM$T&*)xoxuPGQsg60Ehpe77zuJIdZS6fpC*PRxn- zgC3|B!^k-t`iCxJ*|jJX{YZjG&UvKI;Lj1uFL1t@iWBej@Wbc`@AU?+s@b_3e`?l>A8=D`+&Ysdo(&BhFV)JR{bov8xIho5Q{dmuI75L%)DP=hAaE8I^ zz1HF_`Rv1=8Hm|#&dPkZuq7K`Teodp1dqGc_!)SMowDADtvgDrrPZ^L^wJVVdT!Xd zq=Ok>TFZXL>$8Bh22AyrDGRBKV8=s~*ugeqw*Pz{o0~P0O=~*9N>0DF&OPwhTF-wE zYLAC9>7eb*ZH5z@G9-v4ymMtgRm$16xoS8%$AwAx%!Smv7PdujVH&fdQG2Zxa>lQj z(U@wMl%LAV!bdVshfeFp2qo4fzl+%hg|O|mUDl}wZ(1vub+R1I^-NrMFMIf8DzkIE zZmoKz*jmTy9NYYNGjYd?b?YY+1(ls*k-OWo(eZ1-|jFQ>;Uj54er zo9dylVFq5`%V3V{JQ2F>929C^G2f5JnCZ9}b|y5M&D@Z};vVu`RC_vmV8-WS!NzE< z%x97p<8Y}d5ci+xqqh7KX0FqN%|K`D9e5LU`^s?nZ5F;?xx|hsaZi$K7hBirhY2<_ zL96`H`Sc3yf7V3XUnmM|QnetT_?OA-7s1V*FpSmNh-t1DpuBn>7B8?yx_=~$lMl0A zgtHeleBP^kl3JAxQonZum2G#$cU^nv%XpzVy%KSnclom-n$ULy7zHr z66XbMGNJrXCyd*`yPePXLf2+Gal#E9N$kY-kK17BoeRGkXAuyzk8O@lW%3KIu|%vB-b+QmMvFmm;tX65J`B6`XgoY*g&N@p^Xn%XobF1^^)u+wi`$UcU5&SYuR&^O z9s6_W3A34<$7(jLLUGJ8?#y0~MTrIM;hn3jHNAora(AqIa}cSTCvtzv790`0apKlU zs4MfC@W1_7@^n7d9p8aZK0Unq#^H-3oXex!TJW6UGv#DJ65^d~Er5UOcI5V_?&5n`8rajzK`Cb--uXVH3nN}#h zYzB){lQC)gMrey&fcwl?BvvIuKXw69rUc?2m+T@Q5^Oc~${dBqq8*zBnGpM$I!IxJv*r6$$ z_w+o0&Col--WaE_p`z#5#~XZh#k2IPYnMZL*=_jq{&VBvhq(Q1B`z0iM@-5N6y8vz zEguXhm3fl5Mj_s;J%)7dGfRnELbq(qsN}6bEtgntB{Ry%~w#g?1RJRK0Y7$ zK<>*wQ|;IJn8+C!53b7KoSO}&}F|IS}x0wGu0S%Umftf#DF55bZOGLZ+PEQg^%s0 zFh}7R^5RXguXO~zi!Ft=Y$gJYE0JRQ5qB4Spe}_{8oFOo_?vSYk$+3E;pR)s&oe@- z&vt|kpO1Up6*SRQTo^y&KFzs)AN~t#kawsQb0zKJdh{^vnwDabu7lwDLryr$?`BJR ze{cFTM>_pbjqct5%T_dsBgX9)JLx(Ldy;_ZcbB1~VG`0GhG5tRM|`gK!jLU7SUK7Q zx(olqBEHl6F+CQKoz&@mkPc-XA4|*5ZARqQacJaRfQ0EOM^HuAOanwn1@g~cG%V*F zLwF%)C5&B&j$!WHZ7>tx&W*%)Q{J_eOu{|avzR_C7$c7dp><(ChS)!Z@0l}DkN$)Y z!%{J{^a>7z`CyM}I9{F0#PI!Vv8u-bNA(wC*V4`8qIHPgq!!XOgHvSFdzebMCD543 z8*oX7XAu|7gG`zf9ZP-${{{1T7I=`5I8aG&A2D3W(3%C298FvdGQ;di9+;cG8}iN! zS2-JD_3lc(EAvJ~VKo-GoPgQMYz%EQ!lXncI9J%iZ^t~cYuibpc`=02cC_(Xo#d6v zxU0MaeVmbUbwMgT2W+BG!BNybGoFsQWzwhar6k&Vgo5Pc5P6+@Rl4k zeG_xHvxaux1k7#JhrzvI7~T$oxmpN9Lkw`;-U|CdhGU@u-%0s4(26`6!Js%8;24f1}A^(hUr;iI3lXUAO@@pSKi|@D+trO_TrYt1B ze}K!$*?6|inZ~vlQ~iqJpi#qbCMZJ_3@(jSd!w2!T zn0tCu<{`Uk93E`qd7<)!$Sbsm=awC4iPNB+C)~+6X&r^^NJr&}E6CTnioJd_>3-;5 zQW$ic=BC(?jlx38zYTQC8Wj!_BXvtnGHY}uwRgo39j-x1-v!Ect-{r`3(!t`fcR~# zaIOA?z(QGiP0J{C{V}qhcY_{!?x#eJOQddb){)t8Qsq0I<-ggvV-sQfn-Hv;43C9vkUiRt zA#+|r_dL(SS^85*;5y3TMK5P{oS5B>nI@?slXhLulmu6-zR`V?o{( zhSc>_k;Z56?u6%L(p_dogN&TXyJ|Nr>9m57u>yg8JX2${omSv>bc(ZRzrT7wjUJ8UA$AX*^By7gpElV?#$a=6 zDi+8LqA%WmnH*Z_?__O6(|D8(Vi=Mat`&ByeK$542AJ7OaPkf5%E2L2rTmNpy@ zloF|NP!1)HE~GP%AU};CxUt)i_<8~JwH6}d(*ultu8fAcff&G9*)KQzgY2(wD5x-} zNAvd6y!td67IB^itA^nA@l5U$HKi`jUD$nR0d2H%B**tdu`Aq^_h64=D4#3mg&L5u znFale52N1`vgn3?0r|`eq+5mm(d`#!NJ~u#kF5Gx!L3;^^<0b2kV#nNwHUfQH{iTE z0y0T)IPl2wFA zEKT(UwS8DXd9$Chn_$5~RIFgkN%ULg1b8R3*Njnll}!!LAG!H@!sunqpM`isw6! zLp0LAlJc~2;UE4S6{R1cdG|iGmOh}!1sBOd{T9c?Hj zef4!@@n9>>c^pc&EB~MtqV(&G1UY?tin2+PWV-4vjyj#-Ik|hB?JY?UK6_Hxr8o*v zxlC^|_mN#}7G1r4o7_5^I1lzM7H_QKj^`MPt;wXR-_FyNCt=pwkCoRsLt)fto(DNX z2}&9CDL08WA3jM3Hk_c1{C(b|?1bA=C*Yo+hudQZP`La^I-06P-NOxNG~bhG?3qZX zIdf_Gx-k^?@&wQCjwEwSbINGfqdgfrslgxK)Njkiost7D~JuE3>~9B5FmL^OASd zeJ&%-@DyGx`OMt~l62_UH=J8|4!tF>@!Zjk+9vpt;kaZv9P}DidzDCS<8M63cn^)3 z7m(W4fv3^x^f;6Inp}+NTl68iW4xapU(KY$t?kqw&_~+Gev-VD7j3VMp;a4m$ge?} zjQrf_qE;O5NygzrV-YS%=ks3hFVt;Qq9ZSrsPYZp4-S8fNl`y>vQv}#a>vo*=X>b! z>1A|tR0O>~lSNrchNP#YP6xDY$o@Y~dKpagP1ce=NY$XA>@e1^yoC_%Fjjw;Oy}E9 z(juD|m{6%g5B{2wk1=PB2$N~a`~S%Q+%42D`ipkH-}E(ljlsvSpyI|)AT*6aEQ0B` zWF1}K;zTX)Hjv9%PuhL17g}@&Zh6n}HqVrb4~?MB|7r7%>rgtvccvC126XmkH>T@# zLv5lGJxP+G0Nc-y${$N+p_VjR$CjF$CKEr_P}75TbgO+6ojT6B{HNzqhLa?fE?YwJ z*ACF>E4l=;<2a-H4C=eD!HhE%Hpcyh$sZN!8=i~(QeSx1s0v-=KyO}0(Uf~ZBw7Ce zo%gD-tfwA`UrEuc-6`L47yaK8dhxs{?^Vgtv|oudM?HxGb<=564tFQ}Ho!jZJq}fW z;ByQUx~r%_9sV-(Y^o@gs(nP*9~0UwSa1%5De0K>p>~53We%34%YP2>pD}~hj*8*A zrPo;2r%4%cD=A~z9CGM!r{TB!DOT`+=C#!jB3E!Hcm+FmrJ5Bs?*V=dVWWpDGllI% ztitd-Qsqk^yE6~cv5BzQaTitM9cX+M%KB?$v88(yS~{+=KND`E`CBh6A6;aZTT+?d z-Kl5@=h^fL7m*zv4|}C3cH8|o)AWpG@j; zz6?sHV{rHF2mbrCVPu;R{WX+FUI1Zq^&o7u%z|;HAFY!aPZK_@hC2~X&9Xq@fGYMO z+#H(r0%oip0kf4O(PzQ?0rh)WfBS2e;U3Nua(!Uv{SXG1$~hBvG71V3FmBa5m_99K zhqbx;kNal5?s3oII34`GaTftA#mFF7fhHC^0b{)eoOzk8~&7X{2dRllStI65m zZ`kV1f7z^`S!{Rc0Ve2kHpEtcRuGm6FO5`4Z+?gumtyhXoe*S=k;c!UEcUi;8p_Oy z*|`u6kdi8n{gc9lQU6$xxhX2&c)&LMG;_P|%#`z9Gv|*cunYZ)_+jrL()WhNT`++D z5QnvKE7s>v{*4(s5Zq3o7;1}kd@A!#TL zx%3i?e)O=$gOPB%WsP+;W7+y08SKTv<80C*F-V)2v#O47EG{C`BiVxb5*%P7*|`U2KyGaaihrlahg0`IIz z^Lt<)dnxjR-5KTyr+)=3!SX)qyJ3th|1rp#@tY~cE8z2&d(7Qy7+W@?gjp<1##;=3lIrTqY``hWv`ws`OUXeiMM{197+qEIFLa23amTUkszUkBF;4bTxk zgQ;Xkv*q_Ym~ZWEmK8M&_u}eU?~gLpdS@E+_n(JD-)uJT)K2_9JPdB4mU!cLly#rK zjT2Kf`7CrQW^ZX@qAjjC(CouJ^?PArb&l1YFu-X3_rs)BNVL3(^XG%{$kK2i-Ja&F`y@p=jt?$Tkh3 z_58cA#^EM2T|6AlZ!9oh_a?{s-DSU4A7HZ*?Aet=-`J*=57|fYB~;RHK`XKpY3y`$ zlyk>ZUe!~`oqfWp77xYqnL|;TIRN_)*z+FOE7suN&3a_cvvBUETw}wtxjud9y1W6U z@%(SJjXT4JbTA|9MOa?)m7P}##=pFD1oF>d{2e}bejdwemrOwNo8$Om-@*p|c*auq z$|2lgDU~1ArRD?tyg06h33s@M_Ob--y1%o&ywjVds|T~)cg^wcrhxoUuh>s*J{$V< z2xCq5mwM$^DeCL~v_+JxiCCL!0$s{LSJHik1mbh870Lnj@(y zgY9Yj#2)c-DZ zWvMi5^9SPDP63l`4Y2KJEdJ=+WKYBF&u(+l~J%-^oUhtTVaPf?^re|;VAc`UoT&ae=C;26x{ETx)OXS zgcTBd@j5CPQDU<&JJpi&8TDbfE(TV75C84f5@ywI&t#`=WS(`?SZ1aT+dcRs8@S~M zYsyuFSk5&zW!YxfzQ4=e;ZC?22-u#KN413wCV9ML!A8Z*!Q29}r-GrVGzYSx|G|At zHdc$@L01HWYW)KE?bC#I!dT4Vyu{j;My9c47}A%HfKRa^&nkLFc73}ErXImOkAwf-!vb4U=h)$I~pxoLkR4RNXk?XR;_I?ZD#0?|iRnBcZpZWv( z&lN~WsijV#om8tYk(IIx88pqK`|=@F^lU#J^Ik-gJQt9oZzF^lH3}@yq3tb$=!T*- zUHUPeyFLfg8jA&#DEL#)U3D7Nx{40|UP;>y&!vfCvuTc#Hr>AzL6<)l(2q-1=tpPH8=ezL_d=p z4kpn==KzZBOQR!8rjz8LD5{)ZP6IzU&{x%1DwQ}w9WfOMlKlp&!}+K>u0bC?lxVW% zcuHF~jfTz$CE58$so-QKZD073!bIBXnGSb24{JwcpcI+dU#4}^yx->bj=ppj(z(fv zRAZ(oNWZH`vGy;hC90G7g$b0d7C_BaUR2^KPJMD> z6Te3)(Y_I?5Nw`vALk7XvT6;W zizOLp4yU{x%g z_;HI4KP#ez!fJZpyPw7sMo>3rg+_8U$iL$Clx`tIdZ$2Ei^tIYyk_iln@HPFIFj0n z^>pCC0~+HfB78BBr&rc|ro3}IHB4gUK5`|UA8k#0jjuw+ne){}q^Nf13ffnBj?5)9 zsZ8FB?p(e=isGlqW#4jA$V{gx{TZ~$_ABguz2=!@X^Ktgfd@Yu44%yI+&g9I%|}xT zxJ;BXM3YwZSWv@mo*g4Y`mD@PZF+{}e$AHdDy!296)!3n;6q=s!s$$XG~HPG7Go|b zQ;(MoInA)A10xqxsFfFOs~trj4!Tpt@98x9ffPMV2hWc2d*t0>r2G)4Iw>XEu|I^8 z+q0U%_gbX}IUQL_p)YRYWRp^}x&WTGW0xX*4P_Y8Hx?fL-0ne$tgY&b{V|8hCk$cu*BJ5r=; z2puy5ZOv7ps$d&xD^8}lR+lMqc@R~vuoaw}X9_bD+=Z_lEl~UKIy}x@g5ryZlsfkw zy;%2}te5b7cuzh}am^)X|3q4|D1-V2KO~n|S~O#UHk}QdK)?BCtvX7EhN?wTkm5pm zmB!~TS}tU^mw!&>`3|Tvk}g%K(OG5wx(}a7m@FrxNcMA9-AJL}zj8Y6evs~#o#K6% zd}=zM&N~t(sV~Zy!cTHfRG=P}nokzIhffm}vUpB>VnV&;% z%9ChmVlIiQ{-INbm&kowE7nhYgnD&VT2OVGo^-|2j`hE2rmT)|lXFvYXKM>b+N9~* zV?(;)@C-h$r_g7e4dhkgM{Xrg;a8uJUz+zYYswC48y86R9sB52iK?J3E+v>3$O&JL z#Z#HjdCJsCqmO;tNM#PszuHC7#TGZ3ZXQD(Is52E=?*F=52TXm`|0vH8EU^k^wCa% zT&8`eIe)bUy|+?=&21g}xX+w?E=(gs=a)3syq#7gH*4h3w`QoZxe+itj^(fjf1DbGx*K<(Nm-9?R)roCbF2?kyWjyH0A-HM<%b;r5)*rfX=VS|rIG3?l`DE7a^WM3^3HA`Gb; zA*lXS7b5JQQk;>tkQ)66x);mwwutYi#mq^=NrXJB^!R-HDxU9sj{%AOh^=@+%e*p4 zYDyQasyamf)!GGTGM+%y^`{_y<~uAr`4=GU3Y@lj2JT69 zaP-?Xv`?fARr!n|qc$<_em0E!6U;f?)Dk*mbb(rP{&McK0U2$0y>>FpyWt7+z-;il zyBu~}t%v707|um30e)CMl;$3XJ2O_nY#an?67gXABNJv!c7{DUeh}E}25pvypYTxWyY8iz#<(-=2ioE^6{eA4_-l0u&b5DoA6c?^W12etSF41@ z(q?1bSXUe%KsbcEHhayDP?qZrOZo>Mo-_V+l_k!wLBUtXZs%`A#oFpQpZf2Xp0INkThS?FIB$NBxvf*04f zxw)rsStXUoa*H-*mP_GHIiSsc_{(STM*9$d2?<2-}KX!`qXbaJgKNcjIkWYtqN zajF8ocUu;#3pE05zXm>`oR`8e7N%%^fyDZ^Fns$BWCR<4n=0qcefKnh`9%V!#XW%_=`M&{*bLdu#jq#-GI-uQ2E61PWN_vz z+V_atix>z6&pp;~=J7J9kFvzImgczUj|JY`^#Fy5q#@o0j?*zsfRZ|{pv4bzP*l=+ zbSgazwcZazo``^YUpsoy)`_gTTHvF97R-2R1gEC1fmqIuxAUz9{9NCF-sGI&GXA(M z3Yriq1u4ywB)+Z!eOUJcqjXzD4UfZurtu96P^81{dyY0mrmKxDwunrVZXk z_IJ7WT$T=bt&72WVg%TpOaQ4Tagdu40evN!+#d2{^nU*@v@x#(E%|#AeKRaVK0l@K zTzyHL`DG$5wXlSL(~RI!y((OmNJR?MojB(EOw^|vh~)MMBl`()C~$r$H>Zt8F&rb` z_{16HqI&{`t%^t6OLjoj6Fu1S$sDxGjW|EpR9L=V9==E1Mk870k(fmrYTS1LnPilp z0gk~D)2E5&d>?~$&Ra71CbxqkDuXxf62iNaBOr(vK$rJ8s=dpHBQCN~WB(m(`nDV? z-s2-7DFd`+=Oc7yZ3zmv!Q}|rSEEsmry?WBM{CUqAm2%#eNGH?1rwoIdJIij!_9kl z_aGU~Ysh?0J5uSJ2)r+^kOaqw5QyiYMyX_U*>o*h)6)p@JGuH=C=FJ>&w{4vIIyx? z38%UWz(O$^`mZLy4@*vU#0;8zwH~bw-2mG1Qs9@n6z;W3;p&_xaAJBttlGE`6z)z2 z^>1RZed|diI-Y~VINopeNF5TYDMdYJMo_xZH#E~C7ll7Ki_9DI(G70Lz;9m^nmffA z?#1W>?}#~U-=F}-h4rZAAIE|lkcQ+3+7QZ{4DuKEf`?ZC7`ZQiAMIRCi=#L$U^tYf zJw|8KvQcPjI+D>AfexDl)Zd+q^cxe<+zlSc?$tIl+a3ctY7PmHCBcwm?QBKiNUtCq zd2qfpAzA=KgdP=}DdF}%L639hyrbF!zD+p{~y^>*7pYgGgY zKh%V#ZgJQlvl5OEu7K5pR**gS5(t0nfyH*up~XcD`o3Dh)D%@vm6Hea5(!xR-xzwJ zZ-lMCO~6~URPo0wRb18i0c_{IhYJ$7kp1Wfw1OuOQ}WuuDdi>j7Y;!4Mn}#AstI0b zG8~b60)O6do{96f;a|K4Uemgju=?a~t9REVz1M2O3VKuK2ZSSc> zTVtw_$&DGH%;_`FUr*#}NF`7-kb{SwztJyu&I7r#9sKiZ;D%N_e458)J)K+uL!92A zRF0wLv>n8J8-R4P+Yk%a;0q%P#>v-FsrhLnA(jGap@$%A%>f`Bd-I*~NjSK@5ZZru zLF@i7=!^6QgK{rq%w=Ea0C|TukgIkb%rr*8 zYU>Yl&WB^Gn{xeE;UHK{rm(E{2^4TUR9&BKLA`h`gsOC-{PYpzE%F1U=X(IQ*aJG( zec}6)H^~1F_noJ=qvhdpa4bLsdY!muS_bSb+Xp>M^Fdj?4f1bP!qZ#L(7S{4R&C(! z=QgDfsjUdit~bbu%fGp?)&puUFM)J#N01FRgVA6fe0%`lFnu;GHk$y+Rk{$z)#Z8G zO0coO1tdDZ!?O$I-hyBi>@`D#qa@AJIf3Y5(>`Z{q{aoItmJV1~h{LM` zRUq^|0M;H& zuoB%z{m(hZ|E*e-{9PKq+x`}0Z_42_9jV;1Rvg+md<=C4-9>}zxg6DlvfyT<535?I z!;a5ZkgF~L6S-Ru#C=yI((3pZRlv5jldzhX0@mrCfK{S(@LRL|cUd{PW-GI9gSRk?vgwhF|%y0ufap&}t@9^)J0P@SfaQ@L-$U!k19pvT* z1@4XL%hy{-VAF_Hzx+i-+uBfxz9>xN_yMg;OyF+ZJh+jj1+q@Ma3@h2N+wspEZgJo zVf$nF88(0__B}xfoY%TX%nTCB^`W-MgkxTcfeUy2XVNhI3Y3Kt(Q>fC0)vOfGmeSQ z@vlQKqFozz!m^o8Ao-8WgR)EDYMwM`&_56LH)7FF|D{OGk;_AxV+AoeI57fOTgZA-uW+9dvzz4qyDr zNX+M#}Xjx`UFkXC`B`TFf4nu9ByuO22I?;>3AdP^YbnwH~Ih`biIPZjm@xVa1fpa zHN&)zx4>_Y2JUMS$7_jiF`{on+6=wu9fRoXDEObGT{n}C*v0(f1}fuE{(q4{?c zG)-ebc(o0Dk`adKT$mdfam7pTPseAQCt!&S(V!IQ4v+s?fgi_tob~82thKlRoBh?` z!}vHFnJxpO(b4#^T@+rCoze@Yyz5yi4XM9Nz2z`X4vI^I5=g z7^QJ&jS{Y$`y8A&jeGS0NBH@N2TcygAxkX=ZZ?O(ne|ifj9!S zbG@K9IPYyo!}--5o9_l1c-sJ}k0L?YIRb`+_}J#-O#Dh~8kS7(=XT~cz@Z=Ru$-IY z*~%tBo?i+SxR$|D`&i)r4ugyP$^rL=!zz(euu0t&dQE1-a&uD%c&Ud=-Ia0Z=ns%P zWd^YvGb`n?I0&apV=F0DJY7W%PxQYFV<}}^uXPfd*XF>kq3w`!Z71|`Jww*9P_TL# z3~l>Oae%)%p7CNT-uAu$7U$$aAx(jg#@oPT0_VKFy$Xs)xHDBF13er&jgCBE;Ezo< zObrQ#H}g~>Kk5OxA-Iov?k&a#mMz1HelEC*(@@k44#V2UHBg0m(CDctaOTJ}^quP) z^@JzEOz!-27($ph>EhO%O1S8?G46@g!v;qr@xJHI_(k6a%t%)vuQ4NF;T~j zieeDDeTdV=syNTHImnhyhCs7LVA63N87ERS?NdHF|3w!+`YeoFPfKHofFy{2dI$_` z!lCk)Cd{5Z8#cWMD2!GCcD)RQ>8OE@L=EJ8NP@PMG_c&PgxTN9_^h(~^ohLTFT}XA^8FV1G7DexP2d)$QAPUt2U)}`s+f?w1vmYRIks6-W zFO8eS#$c^^4>C@=jyhdq(3KaWaOK=YaCsmN(H)aHj@dL=W2FKP+se`Hu57fR@e0Zh zJpexv_P}eO(bt(8HzkDKp2E%Vq7fa5eD* zq~RTnDmT4E>IO=%!rm4gsxv`~8~{lAN&#AK{Rt)QECN|rTDedCbpReSF)EJ|?o1!5-dIt>8 z4FyX(OBCx~fR6qZ0sBmCG<3!jo#H$a{=2o&?W{0l>aGD6_vS%))p3XpYl2C8=A!+a zby?-_I%GL#D&kGeK_>rLE~nE1^>ExPzlBlg?55+qxn8Gv+3{;obm}J_`IUgqRoDXW zgg-KVz_F>E9wAjHE!bve%4M`zK$X^R-cOz=+9#BQ)Cc9!_G)D$9oU8{u4N#N-c+RW z<_Y?->kzWO+J^$z9njq8j8ZtZ$Mb+b^dx5zT0FQLZ8v&=^u9bm(ZbKsk(+b5ouGN> z(85GikkXG1YE|$SUwF*hKa!8SJEI`;>M4lSJOwu;Qjm-NR%Gvc1EsBB$s1T$#B*}G z#}nGT3Yq6>qdzr$(56R6(1)cF=tpHf>P$(8%CQqL*(efT94>=~nmm{{_ZZBc z$g!9nnj@XSn@H@eDEfIw6D^OF11nCuhN=$U^}Pf!ldMsRiX&S3_7t){XN!Vo?niI8 z>_PG8TX@D2a!7aUS#-Ap(BbH_=!WxTxb}N7DqmlQYL@)uwn)z+W3JX;t;BKrtZwjL zO1y40+&nzTzFbG9jA4gNwS>!1B8EwLg&`fu4 zp@$|oh@zP*?a(tWtIgt%KKir$I)YygTs?1wJPrI&^sFQ3pqe&pK41*1x+lWt=3sQX zWA;?wC-m#625u~?Y~!x&KulCNv`L3rEkZ0^2G{( z|7D_ocP^r?1R?maBa7P!=k9IF_fXJNS(LFv39TOP;yBO(-jy~N)EX!T^Aq*Zqn2PK z-2t%FRZ?n^ z;CE+t60be}SjC+stpUHu;s9M@tzbqft>=@OsSEM-f!nyrrjkVDX^~ZG4aDg34=SO* z4ZCJdC(D!S>35s8_}GOXIAZHcw%5m(9taa=UsR}5MTukBF({1;b9*@X_aD(l#~IkQ z`Wu!vGhx4MRmC1_-H5uBp@H^wTc)w;OKC(w#04JW}#`gW#&9jyi2PzT? zF+R!K?nxA)4anl{>zI_8QMh=XG`ap?BHB=v$6WS)!Yh(p2}d6qlDfDdTszMFlrkWH z3$~E4!W+Olc?R#aZN@^O!u;JUB>9_`^x>CoN@Ph#AnARt&i^Ifj9)oEBH6D*`M<2> z@Y5M(R58e{PP4HUpKCb4C?~yOY1dcS9Be@rb9u=(-?)=G_kzf=H~FMC+5vKoij(rf zgXG&zPTx?9C7zxFGQQ0mZ(DPWUip?uhmPMyO`$W4>9>FDCD%r)BC@fya(DRoLvuZ~tVv(=b%;n&BWPQa$T)v`-On&~D zwEI6{?AC0i|5ml)`VErgU6T@Vexix3iw&`cq9rF7!%Olop`5?Q>k78^^7Gb^rq#vvQy=p0WswkGWoU1%i9aX0QW-);utis$M? z+c<|9TlCO+{~l_}ZwI+Kn(%8uBwn&58AL|7TmpkQn!2HpjuUf2IcD0N2e+~1%coSu zGK$XrF_9c-2qrIrjY*3GgA?`7AUpjOtT(JrwAC7!_4CU(peCemIuERL2Shn}3n3CA?}8-nX_ zT8Aq+XZV2BUa}(2Wmodgs!M zz6oPZms$1tcWmn)+DHp*<~QNChRcEg&LgE>l0#dc?51uDn=l$)gLVGs30_u*p>xsq znIr9&QUCHt-X2>G^3rx4p0_6s^?#az&r2y{xkouzR!7O>R8Slv&#|8SXzU~%8jjVO-_-`>_QfDM%%;#;qW~REOK=UdDVe=sov92=iNyRt_WoAmt_cq z3d2Z1Mn1@#He~|FT5(2}I?)s6lN}da>HHtF*+f$jMr!y3+fhEqe4TP0JO(~8Bcto^ zy~8;ewRGSC!xXSN(a22RUB~WxSx5u&BAAa0u7G`1D$36Cfy^i!qP#^Kr;a_vVbhf$ z-f0KcHJE^Zd7oh8q+^(sPfsF!szdi2+s!OcO|tS>x*jIK_7Ln_;w_L><7%rvlW16O zAFoJQflNt!fZL~C#Tl6~bdrb-)71Z;;H`TNUelLP*W5XSB%C|w=FvN_J@Ov-3N0e! zbTX8c=E0@g3B*aJnt7H#!2G@xg>oNq`~T8mNVKq-@sQsmxUtI)iND;-&N{!HE_cxb zS$%bUBK51lJFFU(I2z$8g^En$6D{N;8BRs!3E|q(jks7!nVQNhqdvz@fw))-V^CKI zsr~Zkz$*c^ThYw8=vfG29$&+E%EM?_r#Kb5IZ92X7Ew}H&D7#bL8_WE{d96N*nj_s z^&Wn}gC_CJV5l~m_DPFo-8jIE`AEV231-YZel6U_YQ%Se6@XR}I~+!F;}nh+5v0gC4yp-uR7|JaOT=-~ zRYKzCxRb$`E=2y07eo&xGrxRJ2^N-`2o_v+p~KgA;DW{pM7q+AUI~vD)JnypEk4~; z-*q3egSP+@Kjq|kX7l+mHbgRFDt=z6k!5%j>a3_PA)#y%IB!Iu9^Cb2BG@H)qE+p<= zL)74}59*s}K?cIoX`10*_TpYz>BaqWZWx z;yno+Zy}DGL)fOEc;>cKB|TH_f|suTg;i($76j-o#=E+=lDl^X;bKCiKqW0-aK|a0 zo3$Qi)>tdj>2U^Zw@w&MUuwh}UY$z5-OnOQ3$K$8Q4WwdjbkMZye8u87B+X`Fl*pD znJs-b2o3H=RQ;s{SnJ2IM}Fdlh*iVvv$7@3)PonP)o2(Qp1+5gv#Sb9BwX+e{cNl{ zNg1wao@BC12tD%e8pr$GPI{-Urvc(}tmDDEc(?g4a#e9D{n2}wEppa`n^O#!{u?)F z*wm-E*!c~WPcSD5=ohJ8zmOeGzsWZ2?t_^f6Nts^6fEYLjd!M7v-97-U{5xlWfI1+ z=wYQQ_@VJ1j^~qOVk8bDoDLSTV1jKh{jRI*=vBB4CXL#`>!$S?=whn zdJNlesER2)kjQS2Il$P=T~?pP&E^sx%mnMQK;j=zNP2f)75FZRr*|u@*jrkD|D_C)dcW9@Z7KuN79hd!hg?+zFpu-mWI3TMID=XyEd2clF z5#@uFY`+hu4?5DTZ!bc}$yH|>^wwexAil4Mm{ribvNZ8$jrj=oqTmUm-;2MI%&jZ zFS5q1i@9PQgMgc9_AXvQT77*Pe;0(eY)XU)i^iER*&Ar)&{y6&T?WLF6+U(L1v-)I zMMjTz5gXsfG+!z3Rjza2h`} zb259T_ag0)lV=MIq-gwrFBa!_(-CE1vPJ$pU2^mguJ>MoM%UPre?!u=?5Qh%$*nC{VO1MBZstCWCF%PUXU2w^)>)1HanVBs95TE&SkvWx{fMfp? z=UdI6jU78q6S?D8nBblg>UiNZ{oQ&3-YmEe8eittX;lhXb%PxC=vV!Q1+7309qzER z*7Ol&g-D)4i34ws&sw@Vr5RVNmf@g_k8s^Ku7^?Z!@ovPlOqMwSaQdN9q*e&rhWa5 zSI-fp3ojfbVu~`bZw_Zui;85zwGx=w$8_l)T7Wn3BiR1<>#Td7X+w^ruHYT>h-toj zioD<2jvnm5jF8V+`Z>CX-gursH^{AG_kU2r>TiBAyGIo8jJD6zT6#0PzCfzMg-bg* ztv{#6TWu+?qgVyiKKd+P{&IF;{+(?OZ+{`7Pl0R8YBHEmV8kVtT|w0$19uz?Pz~;KF}J zMDbxc`LV;2wKAODP!ys;=lnW`jThSxwD3JXv}Zkoj#h!nkQp5`J71qtJ-gxBkPetM z7u5HRir`Hd!gy}841KZe11^ZJ!l%~hu-{WfiEgYXEefw-T$}bYqAe-J;o}STZ|5d< z>#K``gOx70`}!IDP4qIh75|La8crt`6Z6Q%EG_oG>Z|Pcwa)N)z@PbaRhMc&FtfYZ zu)$-yR0ElpL!>@@fkl~Bf<3ZY=!RDfRv9sC*a(u;a99-|+vS0;PI*9!6Xc<>?l81& z`N~w|uk3#lUXv`oB7dq|H9q5+%ldZ5k}ydvZoAeW4_rDymxSYn{k^sgetFsKAxUAn zH8qr77cm!`UK4BZ>7Cl3nBPcu=yv0*XQy#hYc5&b6ot(674cY65>KBf;=lL%>8}g7 zn6B-ybP|6C)ZE?%Zt)wLz44=voX=vfnq;={TQp0k2GBO}q_;9}=yLyUG zgRjC%D*Gd;B_r9e$|1xpZmN&?@GK3wvW=BkP zvx&pw$IMW`O(st77*pvkiRBLI)f0zZ5cf-{!TPNjK_1y;!98p2<2sLter=7{_}|6` z-P6cKonYF#`3$Jdu*Z$lb;!N<3fOw9DXQA@rfysk6RW#d@aGy`e0X^kP7dj&FG5(R z{o4!VdAS@|KQ(x;zz1q)cVpd%GsJgGG&58fNab&JF{*9eH0zoc_NGzz#GBb{*HK+I zv|$o^+$@t>)!AjGyl4V!Wv;`G&fj?a#(P{8XpTbO!I#h#ATTkdwolorgMP3aV5uWUat%mHwg)M(xv^grB?I*(?&YpF8|M!?Jc4b^gD*tw?=k4ndr?y)1-$LbX1?|zRX(;{&9 zlqrH94@qKR;D%RO|AjMtvgB=r2)>|yQ1D=W4kM%J%37@Xh(~817JNLa($F8%&$ji2 z(+ImEX5^d@xz%Dxo|P&QUi1z;JGz9L1c)%MXd+!=#Nf>(3ba7+Grn|Ih8%K>WHsj( zvV{vPu-4TUrZqDXX74Iu`pkARv!DCXv+J9w%ScF^{sirhqrlyE3t)^kUB~s+6^L4UFxC!gkcQS<@`FQY%Hc`)+4s%-S=#^O> ztnxrfM#M{`*dQ^~7uvq5mN~TN zAd|30g~c+WWXbnRyy5d2{ON-`b1a5pi9!>0>CGkxEmLI8__x`i3>~KBWC-vBuhE?R z0LEE6fGl}DhXlk6@ar9R*kHzGS~_noZ2E8o_w5?NIiFtO%+sP|!8K{NTTD+NDE4FA zEGzJ~^MB|i$4fXp@G2}F`vd=?>hUo(dmQF@3?7L1;FOFC^2XSUKG>8-r^tCyf$?^% zcy|!Lemo0bM+%H1NMNxw`AFVi1GcY=AtLxWu?n5a#NF{CzOPd7mS!`;4BfD@ajC!! zxra%!o(w-C(}3^5C@}}x9EsDwBBI&zhmmt5?5n=d>};3&U^9qetA$y9$USEB_GH?;Wn5rjy`2_ymC)eC78pGo2PgJQGgJNI z@%fRnB+_3R3%#iaotNjCzPA-z?aweWQ*zM5fjEI-<_9?R`I;bQ$`xAZTL77F=g>#v zbI5~oH7elVEB%=c+*z2;6B@9EtULu;-?oSG`XfZ{HC@LK>!z{xb?=z}B=71;k}+VJyj-7zKQh8vIG+OeFNC69fFomii~Bp z70xplrWFUPq2hEQ*_1epSC&hW-Z2F-@KcwR-kk!Ma^msApr@$$@(H4_Sd1K>x`&)7 zK7}PEO>t>+3YM*G6zu=1goEEuY}Fr1P5&-ty38+Aso)4^dtWiO3Hw;9 zP{MHyH=d+PoAbf`bT5tkdY(2V9uQ2OnTAV@&*DU(B)A)5g74WrWcy96NMn``86Td` zX6)Dk<_F5~#II?XF$6~DUf zf;FLA-7BcluUabUa+8~1{b05voTQ1nw$TIswFovEFJqE>N07{|YiQOdOb<-2!M52q znJ7<1`a#exm|-VL^FNPJ`=R;l$=^c)i2!w`;?f+!K$jNt{K#`bNYA2rzt`4iU_v-A zPw@_%o)^#98#*xk(9Zn))kXh$ex%IHG!9twbYt)A%*A zR8Pc>;!oO4QI9OGp07gB_PCMR13tW1=WO)4(3;*^RD~2JXEDNz53L$c;N80@N$2?v z(gt4-swn%FJ_=mPG^cei+opeDW?KE^IDs*Ap@lvLem%25w1+Wu6s2b4s)95oj@fbt z=-Qo9>}ls1M&&q|rzbQ)baoD{%?qrZDWadna07@)R)ZD>i101mTdx?i(!nt$PHAx@D_8q&7C>)>N?}8 z9M05vd}bc&WYW%E?Tm5D5}LsM5A44dBpBDN5jcK64ekA9Oz4MYv`B0fy=CpiNNwKD ztQfvS=k32M_;o`Jj(7wzq1BC4YppB|8azQ??~i3prA%P-^iR_(meKWg*B{m|ZTfUE%yXd7R;l}&9wU4ANL6+zN>_c|5gho z@g$fvn=aPfaC2l3wWSt@|L~6Gos56pEjs2b$NC0;ruujD!EZPgHcq|IJlG{f7oGl! zwuS{Wky)Q2tY z!}OT(bDF8G$~1d+qO_+Gc>Gp7J#o5)$&h~pgL;1iDkGEWU%7Ks^PmMYWlsuyx5Jt7 zJaLm57k&aCtSItqeN6KU(@??tN%WQy%QR(SO1|G`ViylunK;VwK425R(u7Xh)=I@rH-b^YH-PRvV%UCf$y zWrB^HSUU1tgl(9tT)(DW26C6A^UQvHs+UW06R7F8Ehw4}4 z9qy;F<6AF0XgXe-y~mx)Zi$3U`9VR-=A8onUk>%xY_oBy|HPWn*NhiPuTlVo{$`@JYQ3CuDBGz#CC4cqhv5ZmGvJb^P$oNtB$g zn2DE8nu2ZCH}Ibw_($B&9V1?k_M)Kuly@Xm8|_^7rhf3DBRX?b9+p<~_`_+ZIIhbL zF1OE$Z+WhdnD@Dpd$-<_=JO)_2iZ9U9Br|GwFkD3{{cbK%KX74BK!?+SRyR)KoBK2 zi2ii@!MKD5|JUwDGE7gA8Q%R`UagFB8#}2I98F3%@&Eh7UE3;HJb^eXt9v5Qo>!=`#dt>6i#vi@8kZw5xgmO8g5x~5U;X`$9WaMu}h~LHi>V+ z2ac@<-kvnjzgdpE#fp*Sy6O0$O&h0!9Y+P;`EVjR7*9ygB=n>#E*C$AS*qXxFvL{OFxe$zp{j;uR~)kBX7v&;9j;l(~;X-rzZO@5n^K0V!?f ztGPq{hWQ)oO=c?acMCp|M;^Ik<;QtM#48gck0HU(brFHXp6R@iw4%8cO}1F<&19Hu z)sCe43E6O{2xV-PW`-+0Nc})EDZDj;U#R)wWzV8;Osg;S=xdXhS(=15oQE&;bYqxW zL#o`vNO-Ym{o{R(bwgs)1%AWLaMNYH_9_)l{lPJTHDo4d^=ySK;UAIS;)KkMsBX0bRw#c8QvSqqmR*>XhS zyV(f$tbBv3m1A(6bGTsjVjtcKHlcoHi4-v!FafD|u6W7T+WNq}B)qR;2v-PRlCy@~ zTvt01xmdp?bFT}?=g@R=htt-6d{o7!#Z~a?W!n6QQKh6bu8Q31UJ5)#QMhq65uZJl zK_ogako`}!_^~z8f|Ef91eykpOiR2mj%{m13;v#fh4=gM*3Xx)w7xXEd~q(MZLJh+ zG^k@X7#YGuZyw$#6A%8cD(fdd6rxKcbjS%6mfmSqkyTGzM|)4PaD3Ki>N5i~02 z88aULiKwWE^Mjq#`2`i%@G@Zw>U!CU1`O`N;@*mA>NXR0>>C5~#?xXnsK&6d1-l z&WG{vs)i5ST#Kydd4!gjMe$U^UpL_6pJse9i?3e^qJ4%i45OGakUYPODN7u{9wUMFO+-QE5xz9th+Y-W!d?kB zB;S7@V^nkstuaaf<(DU!?k(l@3c?ZfD=tqbXOkrOhq#KUYC@7A_u?*|nfY}(-zAUv z;-!w4ISo0kb2$l{swb#Dsbrrdu;L zs7_KeZ2zyJez|`ftT=d@m$IRS$?w>}h##*ii?8oF6TIVZA`)&{TH( zI72N07PIP~P6`fu+bqzx$)SDxGQ!i!B=5cWBq*zmga`d5b3b^Kj?UwZZ+|9Qu-OSQ zQuVa1)PP>%GMIi08_+`?C$YEd7yN7fE7Y91O%N&LNi0$&=@KP%Y$?-Dj_i#k8P^op z4COtn)U6n1#?_rjhx5=MKmMS8&Tk?RWQ1DjM^3LlcRifTFXh{#V5oeF485I=`WvS={KJVRHF*9#6Pk7QF-}>Kg!g}1NISEa;C-K0 z60x2q5Vrpfv+-LNy3J`ddXf6%S6m`Fb#@F#Ni^ZNqSefsig+AsQ;sJe)5Xfi;%H>f z1}rpwlDtxzNDrJ2U=BT9LHdV6*hg(j>@PFUkA1O_&U~dyV>%{)UgdH|{j(g+{m+cN z9>2-?0cL{vr4;PGwG@|rNX4ZCPl={rJN30zL+Y;KjHF8pvrEd2qM0&SE0MSm8=65a7mdb2{69^TAl1gn+6@mS_l%*9Oi}<_<>@792?_u5!>FG zjCZVljzh~!sHRt};NcD_{BV&w(|%HkdEIdw53L`?I=P34sh%_0wX7CQS9ihkh3TLe zjA-2tB`VNgK)U@*=%c9(RyPhlh7d6qs&iL`iSl^BsNR{zBt(`&mX|&D-)Td?9V75% zy)u@0Xh|bY_(X3`5&2N>Mk6`i`m|A9IMMMHuY5JksCF8Y7l+%S_mmNS-jjuAE4nf! zrKO;oriypGCitd4kIAbJ;Cu)pupws??Gtt-L^GQfCT0Tf)4qEC+5@QYLq1n;M3BGj zrqn+|6oh_xlNkj9X7t<##&jQ$^ehc>_30U!bzhwA82Cq}dGqkx!V5$-w@Sb&a(Uiv z`{=x{F<4%t9DnRP&Fr7zj7L^=;5i4&NX?NPuh0zJY9GF3l{3SNR%GgvG`sgoxE9!Jn!Dk z=bhTem#A6E-(a?%ejm-lN5_S+JC`-MFnBgw?Uz8OYmJimI}v}l*MUDr&7(oBR+D|Y zt_BA`I!ubM=o6VvQLI+74)1$>p0RlSmudEhV}D;Vglwl}Wd5lh{Ozm_(WwyQd+q*1 zHhy|ZW@gRhM@E^Du92<8K)Dy^w5ZcmGf^@n{{ZQopG=k|%_ZKK`Q)=l5_vfIiY#>y z<6m}n;xDqE$#?N~hp|;x*set)EZZ`JX!(7@N*et*Ly||X*k7edmJIoHPlew&5P`KG zF2?RVB>y+Vl5KaIL1xi+L>`(K{ z_T({nlW^&{2*rV0>@N+|Mo~6gi(^cf;j1cPo!V;q|Gs&>CIKJmF2cLdj!Tf#@ zj~mqe>C!$%6kv3Td7PJwct6nXOS0j|DNjt!5z#Gz|4$=+k4ME18GyE3^2w|>1w z!?@pB=@X7qeH@8IST=LrxgQ=)EkowBCzG3xF5|s=UwEe+wz39dK%QH*5cl@)WSh`x zey4~dKW~Q)`Qqh3Or^NJwii)EW!6fv<;i8b^vG1=^u36rjr)=Uc|#JkG@A5@@yN*X zL&V%)k$*TrgwM+_#%9Tq{N-bxN#ytmGObIE)CA2Y>c?V8L-<`7;C2&g^rXnh*5^2X zXB^p@b(ZSeT*kTG+NAtS3;DC`4BnJ##I)8Sl3}Sv(jD}0+|ph+IqRyRqbN=wGJ6t# z*@MOWq%>>(M{W*&D0UHze=S2#aWlXd8oIQ5!WTi}0m9y2B8+p@fz&8-*-~dtpfNQm zFxqEI1hH*c(e@w-Vsi1hm(uJLp+oH2!DJ#4UP#LCnG@OIQWB-2#Fz51;h(zfz-rkY z$7f87uy*NtRzvnQE*QH@CMfPFA$p7XLR;{LJ?)eRC!TpZxtBy*U1eydPBHcU{xtcN=Y@qCW2AU{oAwnDmR$^HIW0AJ!7v zwR1@2F&naRx-R}(mQRcO_K|cOZZBmhio`c(5RKBUc%+T8v&NdSSjQxO_;xjr**Xv1 zoAS^q|57;JsaQ^rXy?$i@0RlKt4t?l&a%9(|J9Oayp1e++kowTWJu_LPhqy?aab{{ zi+F0t^HYs9*}+@~G~X!=*sS+h>-Z<~_FpGSU4D;j{+q|HU-c2M-etvx)sAs&2o3V% z>pODzl`MZ+bQAeoZV2=4*D&R-6R9arpGg_o!EB3t!&~QPk8kefG|*G`@U61f*y;Bx zI#u)<^FI`whhNX%7ss^+Eh-~L0~v`*eeSt$QHoF+LPp6-WTniEiiSu-r8E?kLNuuN zJ@*qKWu!zhvqvFcvSs`}zyF}eee`+XbI*C6=gVE;9NGnTP=!5=_P+6%$~*TKj+80U zyAKc0s}tW6rZ-e@oj8=IKNv@w(oRC>yiV*|aFc3gKBjTQ+gQM?U-ap6BSCNKIr`wh z1@u~^!pdHrlbD`RA!8)hA@sf$y`66dIr`R;H)cI0{#~l@Vvsr3yZS+5uLf4sqZbdU zJHex0YSYHpJ>dMIktI!eO`VP|qg(E)2{v0h;oBW^e#^oHeVQ`pRkdmqA3G3f#CK>9 zze9VaKBYtB-a|>T0{ykNmN|_aP<2VI2;MyFqzWs{gs-HYu6k=D%8eaOz0*vDi(UZ9 zRh9I~*xT?py+g93&jPm9{vna=E7%n^DAO5_%xU|nJ|HGp3-y_%LO+cJ`lt3jo%NsW zyG!$ES=>%Kv+Fpl3Y|sQubV9lwv87Y7p@Tw9N0kljTTG;P_#BNc_^TJxz4O6wi(X)5BBu`?mcaqb zT5-*YWf-$CNW8w{T~+15CJ5_07sKxIH3|&2(^6D6g(GM#;{$2r_r+y-; z;|JjJnZG6Jd2^v?3KfCGgDy46Sr7jCm)qfAVMd3O>&{!keR-d%>yIrsK0GA_2^?yvRpo} z+%;va;HaTcxA+Py2;5cG?_XabX6FE^-8d9x9JqxirMkGN&58UAen>maN6~h0vYRct zU+UD~!o?ra?8a9oT+!Q!==8H@vVwp%8aBh#LKz1%dIkQ5t)%b6>yqR3t@O|IEA*e{ zKil83KFET{^<+bbE$QJG!{lr-#f2s{c5g=v6khk1(%bD%sr-^#v|wTy{q$R&p3Uk_ zQ++JJZFGfr?Bsc>VRRa{;VrWKK@`YuuEtRX1E_<33yzPrVs$ef(3%Vo9(mP=2Aoo?$wL8Wh~&-*bpj_ zE`Vo!74Sy?DHK*v!d2#CKh5h~aq4asl~! zR0AJ;j6;)5tXlfzFUyDcD*2fI=n1XaG-IwR_&wMKzI!tz4rAQdR2lQIblqsix5wi5 zUT(A?ZYo{>Ya^{1egW2WAEQ>UJ2A=jDqWZx!!jfPNHoWc1)puj?C-w>aP=`{R&IG< zJ@hGSj2tTVTJ%#qH|U_4(Y_RC6_~J|H{THDvFgIJ%`0hWcq6U18_rHP6tJ~_598-V zJ?a^Dm)obO@vQ1YaA^H(TKw}Ay?gB#O0LCX$dM)V_u^SF?E%1aVXR;GZ zmWz`W(#2bYoan$!-@#&xD|#5Z{5^t=00s&Z9dh#IUSG?$Kt z>fQ?&#CgN5H+zLl>1E-Lm@GW|)j?Vl_S2RPNWXArP&*mL29G(2>)V}~!WL6$+Y%?K zTFGPvr;|Rg-=|ih-2@XoSpaAi&-p|)d-KZ zQDL1ai9Bx&hsMvLJ`X_*?URi@2~{}EuMwvgO_i}fefZ(tX3HG`(#lC6J+3TFCSg>QW_1gk+PV~ld?@4{00&SI$OGk=@Ja!)*QyKs`| z7_K8rbMJ^2hoXpK2*tMJ=ZW<5dE)bH1~K@>#RspWVVI4%LAyk^YaQ6DBw@dS zYiMY{mqhWdGY&UhK|O~UQO5!EA?n0%wp%f&GRfQny(;fhdDl$zczYaid<6B{5|7`! zs>OxF2Z-IFGQKgwL1Hp(7u31LVC;pS{LJ1qF{Est_;=1{HtS0=(KKBvUK*t(ejYjs zY!_sZnW@Kd*o*C~(;)`0?|+Qj@9n0=K}YFDrL8pUIH5&rsvybxwrF`tiMI4?X2vq_ z;CjD}P;O*P-*s-IO^O72l!T#b<0#V8Zy7rG+luQ_PqM^6-88|{M7VQ(yl{GMI@h)i z=HqM9;qIlgV&qSqaqWJuWc|`plHQu`V&tJSlE$GISnz~VfF8OS7dK0A3$qf=DZZzZ zd=Jp%mG_ykW*2(A*g!oFSHeSCou(ZsO77}cR!PeoB%OMZyzh21X|<~@H)>feoOnG? zm^V0tbXNXA#ie`DTzfY@zOY4D{rRv^ur^(&9Vjm}A08=ac$f*t<@?i6<>7Gt+7?`8 zF`t+#+hCzVyx6>^gW1Il7RzhZBqx6#$Ed{5tZo4nmpfiVb|D`l23yJISF+mhemU2D zH=Qr*7RbWq8aV3mF#2`4Hop|{m;e2eES{~P@NIrCF~HTv&OLNC{ux(|u(kqqoO|*@ z<9+AZUcL&HsQUT8i@4PM|`$uV+pXCWSbY$D4( zw`6zj>5&bOc};a4JV-k!tiw(83myi{_%p&sTL z8`GQjUVuiR9)IFh#VUvYhrN!C9VtkH^W}Di_*`*AW@JoTbizk}gx_H{{bIRO%l zbhz+jH~&1(jprxTFn6sZU?pR_4Lr8;jH?sS;ngzgf9ENTYO%we1;f}%$Db@j>p7Wg zP%l2}N5%PR%SHKZX?*JWPCVe&iRF7m{fO2pU(9#;PFJ1lwES}fZflsp&P6_B zb@DEJU`{l5o0`L8`Z|HaW>>aDaU%GK?nZB^CdX`schh|6-c_w26PlvEUKmtNEtOnr z+s=F?%Q3Bb0bVsuf!vxIJahhS?5?_lD%}Pcxuyv1=Ol>i!+0PlI-0@s=Bhqlc>*el+S_FJ1^WM~9lJ3kfUGVS<9mm{!_ zE07x|&KQ&b4!dtJ;u4*5esbJlbo^e;vMRRV)Z=scoyc;YeCQVSnL$a#A3KaV)enAN z`^UbunezElOL6mu9I7<0fO-gX=%j1Ag`ag6!o(GS=}!H0I_pt5Js@nOqQ*qw?jwYsw=8f1_!_des}WN&7pk?rlU%k1ym7I}*9q^L;}3zFh*&-X!?Hi-KBh zXA&1;!oOdT3Tc{qgvoXPQU7xxbk(yO_1CM_tA>jr&M3AUKl&@uW--nq%iJV2HjZV zN0sN5vBn{1Xxr9TG~d31(xtCBUd-VMp%w7)Vk`}uyoH8Ot%0$xCR59K>Ox)eO&ZlX zQ`k zT3L4VNuh63hU`9b6C5(?=rbJ~AtG?UU}qU6rOD6u<|Wg(ZmaqDrA=Q2(|fb1^xLOf00x;H9(&eYvH@fKKgr08no57u=wi( z>4&9R(6TEX9xq*qjt&uYh1n0fHm#jDykXRR&sJ&{oJ=+TJqE?6$LV61VYItxr*PH) zg(YO4kYDG*XYI|yC8g7N-PU-zAvm9EyfCLvoHh!s9rtA6)=^Nfd9QH4MNW7(ppWn( zqZ88bD@?F(!p>7S*_TN<*efI-C)Zx!yId+!!+SQI>ChKKmvXAt+CUwO?F9pOb3w&2 zfu`&jF6=gLrQaP+QFZltdKbUYxW%t&RXt^IWdGwmr`l7uglM|d_!ZsZv~%M>sj60<5a0)7dxFeo*MqQiW-DOaK8AOgqz;< zWrI9jTv8%YzblJgIQ59`yfJ~U%r=7^-$&6KersW2zcAtM`FueQ&j~M9Hu7Q9-S|8C zJgAki9}6QQgpg@ELfG`@5IOP+j96}r8uLo2WpNo546SI(if&QTq|7gM{lMHaF8rwB zLGGMv!A-Af^Ii|N_-RthH`&ewuVaVcUH_&0O+RI~T_c-ON+U z=K&5|3s2W1z)bC{xTfD-@JO{p-ujqXd8yIfCaV0hU_zt3(&&VQyR>YUIqvj~l^h(F zflub?u#hRC~CaSFQ#kN;C9wR=Xu+@+)W%v)9DsMldE3fPZhpP$3P6@i$#rU!pC zgkrRY8u^klgJu4N5kTCXOY3^Aaa>qY1FRb~f$PVfqw7yR$NA$*=$tvGd%DGz>m9_Kt& zpp%Y-Kym43(x%;$eR%!|*8bT}M@*CL`PGafd*56Dv$yV)&Fw+IZb>A|kIS(jy(9R3 z<9fd0o;h!|e1J>Vw&Llz)A$POC45S*1pjMWi<`)AoMCLi@4KDH)!HSTA9op!d|Dkuxg3`1!|KzFfT#H|K})tUt#5N}C!t$9!D8M&Lzb zkD=q*u{>*=jBz>Lg13$CvYQ{2`R@pUf2!@l<7N5CmdRFlVgD1{0v_~2Kpau(zgOHe zeLERev6_3U25@6zMK1KTWMv5_W%IksEcm@0_!vIM(Bd@oFSW;(GZXRG+*sV*xt9!F z{)NTY9z>M~|FNKYEml?EBr0}H#U4RDFi*x<950-Ve>Xd_1;(kQsW_D#*zk~?yCa3Y ze^!eX^_wtZ!V+qreur49fSqDbRdntdizilY!!0t0fg1UX z7m^)m?&55oV3&u@t5@;IBlhF!wOj(040XgWI`?}X_k8!E?sTE~;NJ96v{=jpTL zP@_8b{|bJ-_lr4Ks<1_k#_@c9-87yuAeQ*|K8#DEhVjv}{=@DFih`Br8S$Wlk;{afAxzj$KfPVzA_!mbUR44%+3mMfGbMiBzZFjf%hRoGuW3&M6PkHBjm}v78Y@~PvP9`$A;n3z%d1hluwxLFMl+yq>a#X=D$9^Dp&;ArT;4PP;4Phcf7mjJO_Nm*+`Rgmln*1pg{i5+h%Vhq2aX3BqKp#SO zzN2k8j=Bu_1A}I3LiHD2q5RTzD!F(9ZcsI9pPwo7`r7m2{G6)6yQA$iKQ>4nK6PQE zJ4eaH%JFpahDa#Q8Aa-M&%o4EeNorMgz9Y`M(sK*U~NepzU@w7MF*d=m?B^^nm5z? zJ@jDtgURf8;(l;`X->1}X0Xq?TWCGzJUm8CsAl-y6i^k@U*o zb5&+@yy&4BQy^nb4{ExqkD%{qBMg+~+fPy6E9rLC#iYN#W=x#gDg997%8@lMdgt_BBx&`nxn*^GDnZ%W&Vk_$Id4(=HlVpihNo-*3RPdkM ziyKu9q;oE+2%EKsz$GIjigCrPs>X$dxh`N6N{re44jOT%{Q>3-tz%1hr^Mq=2z|15 z6*X>lq)Do(wDMje-F+*N&al`*pWD{J(5^k0VPrxVSM-5h9XgEv{2>m?h^X}Mr2)IP zoMKh~>EgGFAeJw$4K=4vu{(jLpmWX^V;c9-U&Bt?<&QR|dtH~(R{Po1&DEW@mYUGZ zMi1d~dpO2Cts(tAn#DRxjl)p88n5*)<>YPUL>hk z^5?Tp)#5CP>^-cw77kV%!txLEhq4tRNTRJ45YO5O_%Ga^ z>nv|$wtC(;@^dflT~QZam}DB;}S*| zGWC(8p-RSlYFuyRwx(Ttn(7W>-@6y@KWPJ#?zqazWM0n;Lw8Dc$aug+b47fbIv$4_ zevzDt>jypxfsAk6hkqQMF*@4@AN*S-d7L2!7F|tbs*p>X_c}9!KF=gk*Ve&My=uwy zOA}z*icEOv<6!qXIUmYr>C=a+2a%LpvfRj!b@WEVF39?m3-_JxgLa`YEeM@MXI|BW z3+n$EIQM1)o2Fst!!c;N;GCHDE1CEGIFFrwzKQ>gQi8Qy2m8(3A^FhXmzESNLAb|s zyf|Ss#1?ANlA-=IMI{-tSDJFKo5T3;&$d*1PA0^zOlHF+{prCg?M(MuAKrPclY9>y zfvM^SVuidtJYVmI4~$Qcfi-5l=gU3(^_c=b>t8cWKV(WZ+lPSqyKS7!87qxH-A~$a zU=WYZs^s;TXYtJgCz3s_gQ(v*Q`#{04Xhfx8(!`^4{x4Nm2p%X;lflPYOof}p8tb< zomL3)7vQ%}D?6=o8&vz&i3>al99rH-%o*|$N4Lt+~;=%>;65$_HYNjPfTKl0a+|>tUCBq1%Rpb1HxV^q5E7pRNp+D@2DBa<9I(F zzp)iELn6d))*7&VLIVDmnTPRFC&18(krdZzU@i98t2&Nucs*2D#qYtLke5}jlv7a2 zGX<`0en$c=E`cEPEyeAxrTwE0P}j7VpmM*ExJo_ne@9%^898N`Eq3`)Yt3KXHK0gVRbRg-#VEVluco-zD(4% zP=YTBi|~Q{VKQ5vz{ST0#In$rWVe0@8*{j>s?cl1ndc)%SY5WTv0Ll*p(^b`VU@#{ z?IdWpEVFl{F?}+u44*c>A_Xp4kR_jG=ew{3hJ=+u<}eQD{Cd!J(Z|8pV>;n~_e(<6 z3t&~z8r*64lq@#cOc&o$XO3Y<*e#iFcE*|#aiB|IVqmMz9DW9nT9+wI!6z42ysnW9 z?d<{jS&5Z~D{o1x_dS*v;=R&|-c?iudo-BD}dA@4sPaE;h;^R!C zxW&$^%!ItH?OpkE#{lXxaSXk8HkJH)Z^=e3eF}ja*I~oaB6dGWo;4jh8H@;X#j|(Hrzwxm>oG(qler zib+G|b@F`4t4!1#5J?o{r-DV@Qpt(ydnF#Ddy|K`(X9RM3LFcM87#Ui28Xm^>1bv2 zj;>)|-jEo1{hpNtG>MNp zvzW8xX0YwOovg6mR<%UNCQcaHA6-lpar4KCkT|bJvN1ieT;QgM${XL zhi%x|Rc=Z6G%qqGqFnDP@Nhcbpy}m`@}JW-&WNc z9FZKKGC@2W;%Rqf;eSC{RY|lS%GOrw!^8LiD)yhKUkagrss<* z#5W%s>~^oO5XHGYm|A;ZyJdfGRlXUun;6Jgv__qgbpMwl`AkOZY);+tC&p$%!MRPLZye9Qlu9dvWdRldM$s#Nc z+|Fz*Z!oFX0r9$Ht=*l7B51#U7`}>aWbHL2{2JsY8I`&RGu3ag%H18}$hJ^vYO{_?e;11NOHZZaOo#}85xB>vij<4egxD0H@XUk-o&hSRj7(t%`jYz30+yF zsESI42Pxqx-3z$TzX3)2N=!_>%#vlfDE)6XBkuHK4U85l-Hd|oO6!7`_W8ZOzO>c)qC)B7GLqd+#bA9Ly^|3sDXd&k-VaFnAEO*2tS(q5A%DS zhN&3~>|z4KZFL6x#JSfDS^fH1pcCfFyI=L=w{yezF27Nr*Swx=KD&?%aSq@omZi{h zH%#f1*f&+aUU{}wR;$^T?g`RG>j&^Ioqt#`pqY>S5G#i0p5VVHe6{<1#vGkz72zY_ z^Em0<3cOLyJ2VDfw8&~ldnY<^iUO?bXrde~^T)OY?Ue#9`G z{Gpw!-;^QJv*%6ugWeO+wO)|EvoM!Vm7P-)Ch_u7KK!tZv2;##hPl7@z=+i@Yz=Q9 zN1r^c^35KFsZBQcenu?0W7Z`Wt@*?~lpA>ZMQiDVZbzwW$0`2m{1Bxs|V9{BEf0)N==6K>z$hNv=~KOf@E>60iv(Dw@PBz1{y?M)JqHcZl_uFe;Z)kUYc z0@yIv3ykbl@ZqCdqN2qWC`p!iUF|a2LwO|sWcAOJ=l_wB3uJekd0!mXKLrg8-okpX zO{i}<0KP1ENLJen!-1z}u*ippMYJ5w?B9RIJH=ionXr|wc=ZhKC8zLQ9m;<#zl=9# zw~~ot_OSlf<)CqYn0P4JL1I~Xo3s3ZQjNcMvNO4YI}h2Qpqe2rD^J7LV}sdjyCU0H zOWmYhl1XMElC&U?v8R*lIDodO0PfngYK%!L| z#A2nhc0=$60jE;&h2lH5Fwc|eRj!0W-M!!Z8H?5`W2se;pz|3|HBaMG5R1gmoe*W9@X**`Te=i@>Z_&GluUw_63v2?dMvz zb0O2}I_&c}!(V+2;&0dfhrC6Nzr;ekSRKF?M_S|G*M(%^!+Cb&9K2x`xWRUsot;9e#=tv?V)VcI26vumkc5PaU*bLmx$6IJ(r3_Uz=5Sw}7Dn3&(k+#D!BH6D64+KR}{Hey&mzmia7Bs`(pgwS3cYx%u&l2&p zWf>drsTU2IVn$cYYA0*u?xVie9lkld2QIJeV*VA36m7Hy^%D%oWCc{B%_0bIT0qkALSOzEXo$Ums4A>W1@1p9j#Rdnk?{ zyA5rUqM^3q3_E&!4lFpE3TOJs)6U{)@cyF>RZ5FO8VlTcPj9~a)paC2rs0_eIrJ{h zg=X3AUx{rqwXb?dZg0CK8DY2xTlRV4*0S;ZtYJTTWP>crH=f%$t;~_E?vo{Bq(We% zQZh`fc#6XNcC20R1FtVFm8{h7X}46i*Y@A=TYP?h+?tQ+I zW}7Us$!H9C%%99>e|Uky&jc*s<*YFHA9J!yLs?b^`#jj5-wBOFho>ss$1%3Y0 zhFG&zOzl?;R0ljGmsJkqRM;gsd#IV+@G;}ATgHOjyhgI+osKko(k0losY+7yFMuRO z6^a7_2jgbHy=1XY4Y^=0VNPyR{$u4ciTCA$eCyuxe7|!N$r-T%?oL$|+u2K|ck4IF zb}yz6!v^q{(^2@^_KKwM!VPwQPVu;S=2v#~kO`mWKb)^x*OwaRda!3L8d!8(HbXme z-tLK7kX_`Rg38>bjO;(Xk(|$;4lAvjh+5-ER<8M!*)>d}8(#X;$0Kv8{A_Kaq1=-s z_WX~|!hyR=bkh=D8h)cE%}M(SpY)yZcF}3-@mr6-xVMYX+@=P7cI}k){T!(BAw%l* z=RcU}qk*40-$K9|D|n%{7i>4TfX(S<)V?1Ezec|W`7sM5#pR{Mn0DJuSWv;9o%RLy z{h6@uwJw_9mVFm*^?-V{L@bo`5I(w}goD@eplRArnun9A_6A0bVy(ERb0H~GxJ<13 zO{Gs-3t|4%+N#^@53+&r5p1knJ51UF+(cOyg^->&*WDWWObLU1N=iKaWgwb~p78R= zeBP})gYKRXPuDfYV=uiZoVeYYxSt-x?dtd ze1Lvy%78t(3@?&f;O5mf7@!ad&#uVP9Lr$ZanOzb8QhcZu=xt}`wXD}ZtP{IR}ave zb{#NJ{twLBEzfP&ZN}Q!yJ_IJ_heMdNZfj-KO4T;OxP8pBXl0M5N;`@!^H>sRKc&F zzMA`^vh8F$_*hjzxN;qAk30zb9=g-P1GYm_k`?GGjKGC;OF8tv!5?^;VseBTJG#yk zzP}R5jNCN1xBd>sO^krTA92vumIAhG=2QFE7qOx_h;CIF1Gg+?Z(nN;nVbKEN1#s2 zJ$i!m*J}0$!e~s+KPW#~L0^y15!%!q!Q1>udh}a9t@KqAa$PnE3Oz&lwyq?IS2zTN zzMiG|f3<~;hUL^ix`a~G3$XJ`8~xzbMY>PCW-5dCp!Nr#)hQEcHm`!r+u?{)7arrXbt2+~8rYFO{nWgBf zXNC?zw&*)2ysFt3#4?%ppsTzU+~iK7$@x1l>g;yD|3D(HHAGnKk;fM7uZPvyYd}by ziF>B+gb%s#m_JPp^*7E(H|Ztt-8dE!4w^7c?OI5Z?XFIaFJm4m5~vSJqM`Li=p46q zG-dlga(rtB&Fezyy3!Yht}&y=n}0%Vzr;FFRN-O$QR(yMyes)#s}wYSO`x0KUqhbWKbTS~2ScoD5a~ z+vYe)Sn@>(4LwGiz9nJ1)RIQ!IKWQV2C&(nPv<|@rBTft@NrxTsD*AK_8Jh z_(>TeSM7q-<@uO(xDwAeYIBRx8|js?g^(8c5KHWjU|~Qok{P@Btsj=)`bFjqF$TUo zSLGLezl8B#H^Fj2 zAlfP%0kw<^5Pu{Q^j?+2k#TeB-bN?-bbJa8`RYM!eo0~Txu@7ubv(7c{t^2;(gfKY z1x|bJrDM#Z=(C#bRJzCqKRKtdlp$tp__O~c1E%Z|r@DDa?3?5JZKM{kE zQuaY;`(jA$H?nHaz_0kmQjV7C)q$4WV7}kbh|h~1#X9e4vsZoku#HY@#Qtlp*|D|h z?3%&4ah9Jp^2f*iV&OQ%m{wE%N-cVe8or{J|~umtA@bxx*>Qt#|54|RffNhxXk5aBvHM!p7m<0 z6yx8haxj`gMt)9{IU^0&$&OmPFZx}w*~UJ&U{uG}*qoPK_^Co)J}U?P&d$G zF!p5nd~t)`Lo{=B#oP(naNxvAVmG;s{3@}g5mO!TMrI8dJ)h3?iWkBY=PFoJI)mLC z+!IfMD!j><2$P;i;f60#wD>(1O!`>UVV%t|DLNYBS{u<$jAK@QSHZ5L8UJ1wg@)Y* zY}f@I$)3YnB<@g;$|UVgs4MlyfX@|h;=VU*Uv?A|B9~$Oinq}0T0$RgE1|m+?n7zn zc*>sXk(^JFd_nnHu3VGMe|k5&K|@r6bt- zIkN>@4~ZiRWZ4AUELrT2{kSD}Jzf+if^=92{C;tn__=fTXV@dAu)SOCJ4=~-aDIc| zCUGpZx(ln0jAx&fkBGPLMR5O}-uzQm0PFYXw>Uf6hFq{5501Nx(P`*e^8Mm7Qkb0% zs%kgEd9y#R>w1StK^dsJaW?c;Dwo~Q+wf}g7qM}EE?Oqa7`Vf&*c{yhj0=`YrZlOs z#QTeIE!|f2Zn`O5+7iaM|E>ozE|AwfbJRMAD%*rFBzbfP9Um|{mfis8AMNYRb(s$V8_#8 zXd`=DV)y~Fi?%aE*H(Bs$^q`@j-p}Jd+_8kDP)DmKuCua*WT8L9q0dHM9~Fqle&<* z2~qrq(n$OiaS6gJQsL&;Z0KG;0<`BwlQyesWFPS+-#q8wkC#`_a`1Dk{*#3Bbnao% zqtSHOd_(R!{w+(ey3K-~`jeQotvnjzc#z=^o^RX*uA3;7EL3LdzFYZ*DU+b=(;=Lx zya4|eJ!2WwJ=x5DCM#%$}n`EEnaTc!DVnZv@ZTtHgznC*Ql2XgBU z!9bNN*m-3(Z>m(J#ci5|C{06q?*lkq#ZqFw>XCTQu8p*uh$nRqZJ5u4m8_>q0r~4H z0pDwJ>})|43US9+Ps4FIXEEZ*7w=F(wljpL``PJV{uX@yV8*98&*U|Y3&8a( z69cupXusAUWUhV;uEfknS;RvxmEYGGApOWIUv=3hjRL3$hMFsv?Pf2U9uf2Dm2A+h1emn@8ce;f z0O$2?k&_M-V)9PI#nzLgHra`eQ%MH<_XFU;l?X7A<@*e%$b^Z8@wlkwt7K%}NQl=i z0l%0!R<>XVDsEE2dYQM==hJpHAMXnZ$5k*e-Vn=FgV5DjK}wq{nTmf zCN-)Pun_Kk)8lm7X10Am270%2!Pi-XX^f#JIL>QgPLIskoS&KC`{^e%OwGqRhWhl; zZfE-S%L@p}XcQ-Wk$Iq7g7|}%b7)~j4n4NgioWXQNrwr`=v_XQSe>{^`f5#Mk9Rj? zUWX6*b|uPKvejI(a~2t|OWE@0ayZ{<09!oyJvqC+7u8dj+FdSr$Vv}>Ayvz#Fy~Hh zw6pmk*)?qwdOlkNMR%Nd$fS9&wsj6qG#JjgLq3l79L&!h>;Yj1PD1~eMl`o07R-h% zCRZLhKv2^^2pmxe3c+j9Aeitn#S*Y~ybck?GkAo|Arsp)2Ckeqig!FCiGo2X_}Vs+ zyn1_B;w{BNStIypkDpLvL}*R2EBeQJG4JJB*!Sfyc(bsV#L2)OUrcx5zs}^s2q#~9 zWOf8So-v5(y?X|ee-4qkxR0Qc5H0;N`yNldc30e+<%fsf7|YO(Y&7Y$9IYT0ofJaR z$fzgN$a2QJ#ueQB#cTYvcokQ^IR(9i1QM?kFGgKh&&kmJ(wdSKsiM(uf+qV>Y))rk z2MXZ(*VRNfZHsJA+z~BP&+x1TxzZ>1zHx=h{?d}Wcey0WhU=*B!d?b_`JFeq=+Ugr zG(1em#q=!NtR*{-7ay7RRXcjV$c6^l$x*rUtEqQl59nL-1pdCCFP&$RFAe;rFTFn{ zjt36i#l|m};*Z`fuxEE8w}|k@;Ba&9zA}Y}Oex{^uHW#rhBEe1R3q;n#K0CcP001> z&Hpo5gli60v6KZR?D5qI#Q7@H{aX^bsjScL*X_&F5BLzhY?&KDPED$%X2@4gno6gf z8cB_p9mBycYxv6cWq7pL8=|>BO8RYJp)|s^Ub;~u5wwXC#wTb>Uwt&-BSz0atxM6k z@1244;vsj|K1RX@H%_9XR`q6|Ezj6hI~^h3X1Cac2ltuQsRLa8>;v3er%NSO@~GKm zPUlSQ$tS(+&2wJwWIuN6lcL9j#2knex8x<0J=2=mf)}#<#EpkhX^*EgE9WNvc1=?n zIHw!O+|r;nX|?Rl-&UTSCPvU+SQK!&C<8Ir==~frZl@`K3CYDLNH9B0HH2JRdB|?T=~+S228q zHkYdngp<-t{zWZ`@7)$9)m>;>%K?JIqzzC$|CZvcOi z5>6idPUNqiD)7f6)A-Rh1Ni(4tML4+O-O38A$^T0{gAy6<24%DzY%YU^(!N$o^}OK zUm67&n__TrT&(2#!dh~7-zs>LzPGYIRv*Xo*dcRg{(|S-^C4h|oW#Rp1GWsb#10%T zdrO_bcUeoh))%G9g(h-*#s@Eww^|>^%VrYa{a3SpS4I4=uoSZUPNZ|!<)Pw1WBT2= z7;exKn&inbw=LLL>S2$O=vxt+P zaZ-IvL7MVI;9L7?;i>2Sm}Tu_yt4i*w#&c9qj@T*D)ZVY#@=R|8x^Tuau~^}ehMSX zE&1DtflRL@AG6=;@w(Pb45+z=AwP6z+_6t&eD`7!8ZwJmyw*prt38Rv`YmYtb|Jo3 z)Rd-s){qdretb^nAnBt&>v_ytBVMR=h>bN&W&c#2xO;B2w8KAyhi>}DueJA*#*gcR zj$XrB4QB#Es6vz_m6_+r(AV!lMsF|;ETq>{WE^KxtwcRzB=lxG zpylLNiJW3K85fws&Lr()K?nQNuI5PkaY-XRwqBOwWbQ&AE_FtmcSq^8MmKiKT9MDJ z+rtk}RK|l2Q~BVmZ+QE(39mBm&l4-O`Nw$<+lCqw^OBEOB-O7t6xc-Lg=v>y z(47c0ANCg9Q;ab@aV#b3D!eeFzVe%eBVX!eO}CHI7OH;*QPrV$P${n;w;P|$*FF!H z7^b~KY3O83vC@R!hrg3U#Fu)Us^JZ#`?>$=ZQSy{KUjLlK&hb!nLGL)eX}QiT>x}s5q>0#A zqspe1=3#AU1pOaFXW~|46UO0E3Q4OHZBj&(qRu-rr4osfC8;lEO%jncAxitIg_QP< zHiYWD^UR@;3N5y5MYe3&h3wz^7j#`+=ggUTp5O1j@40@)yX8dW-@~iqnJ@=DDDxWq z_&S&zeIAU;G~>|q7B`|%um*R#cH-)_p2RfhkJ#XA?C(r|t3M{!`& zD5Bc8A4jx|5S4gG;(L6d;DVFJ6?<<(%8@iOA$&2OKS>>#^*lgBxFuM_cs{;w_8<3j zQX8I;`b2#H;XP*K(SqGCCNS6KlbNy;VnHu1lF{89zp>ANe>}w%>&IE3qRzq4__2zy zmnvjfJA<{oD~M8nD&wqkV40r^IkxH|N&9k}7~8;-XuXyq>d9wr*8x|>D!xUwz5`QY0=7i$E>lCq)>^lfQ0 zG{|m-&pKwPvOXO7yEU`$??>6JI4PpE?GSf5$BaeH9M3l0lfd~G>*&fwX}IBc1dbGD zH1@kL{ShmIeONXA5_4V-0`av2D?JSU|#EShQD)Z0T!fOPhYPXC*CQYbcL2 zm%aewkUF~Jj1-FV?5Z=BR^SQ>syKh=omA!E6mHN|8z|aTiw)e@FzczxESQ&*44(Cb zNsV0u7T2}dWb%%s3{7NN?dI&LLK3q{Z-7JpjPRt#uW0qMVAdgD&64WH49!^wM;<%C zur6upy=;^q5=( z(+sZ!`(}M+x^6Gq*wn|T%-;(JD()z+xEW_|2xljr-(=YlPHg0gXtuv4g2^c*K-2mq zOe%M{km(0z9WO2MpOVe4XBlvLKHsSB>}S03e+yW|WHH=)EMb*>(QKF6Rra?njh&9j zrgv|Z3GZMbLuZVcp=kol72G8EqBOy3(OLSq`4xJ*@(3lqKWT24HjK$kXX8`r*|AN( z*lX)VRwvcLP9#;Z+9qu}S-)8zZf|AEt5!%1JeEnuN-varI58c@Uwg<5B;o8&Y9W)B z9>FYM&tp&T^wZ(S5l}Z@mw7G~LsPdZ7z&?7ix45};Chqk%&TWFbQdwJ!4b@Jv>NVf zQ$-v6j9K4nT{c``%=-=!_7C-4G-p-@-yb-%_uD9a_sM{e@WEBTwcvqM6knXk|X&%zEOR`TSFH{FislEYvoWdHY#0xnm|MG%^aZW;;-aIc4mf!EUfx z>xv>WiddfcIPlrxLZt@}f)}>cpd&pUo;`g`z3EYAxPLj5PEIVS(+A6pTthmWm1g#Fx>vZdYg*_ZXMSa-KQI}`0eWIRSf z?#N4Q%ZbSntuYmBYOpUWY0W?#;ySLTcNp|q8sW6Yxj69EFbJ4s%hKPTWJ?s=*bu2k zxSp2|+d3S%C$}_N$MJlcY-Ww6CdJXv(qZs@slWzDBN@Nq3GBV4$C@>Eu{*AhS@~c$ zSo7S4IAv%vBeh~Cf7ym<&39*Y0XFRJ;}6g<5W?1+-7lPZh&t@gWjQ6!Sp+=}FnJ%_ zo+3+KtZzx~&C8br6=q1rdG5!*_FtmX_G^&Ksam8r;7Jd;`ogMh>NsLvI#$z($F4Hx zutn(yeBX5&WCu3Ticz}!7jqTv@7kZ-i_cp4vUCLg_w)h2ooo$1wgaqIt%fr*4ucu} z508|Sk=R&lX2D^4Y<1@iW~L-1al02MQTi;+Mvr)f{Ok-NTO>_OJ2jEznI}}me}J}y zn6cPZlko5o9nvXq?bvTeoS`OSjn%I~m;%8X;}4`-tpiFXE@c;s=dcNRM9*2!lIS4w{+Fzx?+f!ht}O8> zp8_Tx-cVA+K~Y09d#*SdV$ThhJa-6Z3;2^LYWhE%yMQOIr=;NJUInZ-&5->hFqR=Hmy`E0^XXy>I(K{mng!|SOZRylL{?okBi8SJ@;8sJ zAj!QtWT8|Pn;^PIbJmB!_m4kG-rg$W^r?^(U2;Z>X>-x`+wI^oU``VLd6K9#ngT;6 zMl`beD!ZPu1|3X&%5LUMNwj=@*f;M8IPfD9j?d^rZ~n$X%j*L6ZI3Ve{$Zq~uXrrM zCEi5oyb1Atvkr@`ZAjt;jB|xKa-3>7PI$I~1)bZv=+ zRXf_3T^kyVWu7)}kwYvu)j%5R}JxL1+<*x|xTw>N`VbTZrXBbqG! z>4T3?+D=^GMPs2VM&B5{LqA=EVW6uU>)Jep3s(hC)2TpqLn3C!FHDzM^h8O951k~j z{r-ulJ8M{<^m{{YydKT0$~UumU0e3{NdwC|UqFAqUcxTTSxlzg^58Av{@^i=dXkkT zpIFwe$rAfyh1zKER=ThC0_{68n=@&;gd?^sL9_J@A*<>PD24@~ov+8xtem&J%-gy2 z-HkNv$#o|B*m_pnF(Mo$Cw!nwCSL&SV0*l9T^d%4R>eQ3-^W!duefP32DEvO7u*W2 z!V8%uv;O2oEXI#!S=*YJ^Oy6qXHpo|*82y$gsx|_z>o^N@RH7Hor`7Wr3xHBWmaSJ zna^B#m_P2cOyu>k0nJppLvQJ*qI)ujh{Ans7IdZ(3>3>y|2jsaPIw~O;>#?vwT%sm z?`P*;R?`Q68sPT4Sok@mo_#E`X9Gig*&H1;vUtA)|Js}e_uPljd7;mRS+EQ3>>fjU z760Jw6m>RR;~pP0B@k`0UyU9dtYSuk0+@G`gq_^*nOP6+W8H@K5>J^MXo?Ppf6IGA zOmYYs>zt0Yt;?{KjSW7u_!!lsp|Hfn0AG2XfVf&CH2K01$s})6$+yKltoQysoKAG2g>t#E<`Q|B9P@>LGj9P?GE|dX-^lJLPWfmGW(+_QvOW|Xt*YRcJXVdOSXYm?` zD0=(48ElGq$fKgcU^pe24%bPi69(jo`p6{o_x3w9>5wyfl_vc5YaL)mHy>n2_Gqvs zM;(d(R|S%-as$p)*caObrd%>*C{b;wInCd6&q-?~P?4iYu{d&;$!XECF{m zEC72YLT{-rrgskC<=xFK1Y+!bq#@f1p}Rd9{XUZg760WIcD+VNlKxVIUrRZwyrJl; z?m>F%XdF9lepcv|oMT1yoRIfu!FERD(We#_dhNq0{`*{xvwX1`tFDYiN)O+`sPZ;` z&JKy-wl#pn#)U{@(*QlH)J1EXDL1OKipw_eqycunvuVS9|_o)huOd zb-NfHu$qXz&L${#k@RarP4O z&ZWa4E!Git?cc?9U!R1=d$>bq<4A!sYQj1<{A6B{v)QT(JK4!l7iQ-t;y3TvCr-_4 zq+X>8)T~dPw#>UCs{2cji;FYOeV&S+4$?vvH5a(GQLkBh_9^CZFDT_TYTJqhdJj>|9E@0hq@lUg+hd4{EUM~S;@sIOg&&5`=L2j!pRSkocB*=IpdUx z|D~1W>9KJ%zCcy7YC{>bopX(CD)tg@d619Gb0@J%B|Unl_cU@@It(9I-XkfSyH~P& zY`kQWn*mgIokWQ`<59xCYWnoVM0nM0B;*YGk$d?p{Mn-(`}W4++Mnm}0`qYMPn$y1 zrwxTu6UM^LK*9pGJ#gm*1HNW&KAKGgPvH62G}2)|gx-_p>!uwrciAfN5A=mA zcjEDuojP!RAdaoc_-gsmLIGCJZKpm3q3~!*2Uqa;0Zfot$GM!IDO&gAK6}*ginCjm zkA8h$D{8%-fvx7W@WX75&;ZVf+hVbSefd+z?8ihyFru9Ftde*t{x?+ zo;pmDGsczWWgC-2ZQ*3%*vn+GPNJyg>rLFaxfs9oj*?`j`AP1cSS7(WioE?!ZIt** zoh6@K3R)SIm4BQld75;d{nly&iT)6_+|HG`<;$=jhh2D~lnZ^?J{qq7bioZz1*ZGO z9k^7>5SC<$xgE30sU-Z8I8kASBz&mQ0l(|U`iHql(qr@`n?B5zh{7JT!Ddq=0kIP$ zKR#DWu1@!q{1IlE6Bb^>`(^z|!0%)l@2<&a+^_+sjylj;K3JG_e5H3g+Hieh3yk`i z#8kz0%xImR)AunQ*y@WPq*TNBJ#{dy^eo+zOO zQ zh#oG04SE0ASUpS0!^8vhrO{+qey@-FVLpi5T=sz4e%4R(OOd{2) z{3?2WzzD9(-@s?A?s4md9m2dbL+X|~HHr;4+VTx?*U=*Pqd2Mi5Z&;IpeYK`D9+;| zWJPER9IFC4U&uDwAxb^29R-+B2<|uC=sdm(CTnkJRN$;6jfe%Ung?8f<$ENPK8uao zyMe_V8b&L(+mMLlD6Hm(nSS|awlaLIB<0NriC^hmc8Kd_pT;<_@;~|TdagCwTvZK? zOAth#@8%}7`tyzlq|trPLn!IhVVpgEId0N3!_TM0vnSb~nEIk>7GJYWI14%UF5Ftk zEw9FhO^k5wzX4vU;1e3WqYBkl-{*!tDPqfB<>1Xvj^RHs+7dU>E7r23iM{%r1rme7 z?8NB%V41&+Y#dlfPE?O){(A&o#i57r*C0V)G34Ph=Tnf^&}bU8d@a5_-V-7{y3w%9 zc93(R3q9I@$;v#?k!rp@Pu(6>@J`J?@x8wRq?@KL3h4gJn4n=ZL`Rh@X*S=W8ogZNb_5UtF$DC6Bt9vjT6*!VzvNfSe zy?X47`x5qT&M{bo?eY3p8)BC5708_Zq2G%sc+Ftz z^~bUu(Z~46m4orY`Du9XYIzp%tq|WA?(-gjKk=;3qshCE{&1>MMk1>LtZq%O@P@3W znd6Sb+%5ZIRed7VHHP68nbkNYY7l7-On~rjVf46bHI7lei3e>RN=ELLlIU3yczORn zVxl}m;?{VCab?=<*|QJ$tmP|weAsB zH_#mMQ%{?~zg-O&yC{>mWb){Lb_n?v21LUL%< zC?XCSga@ZyrG-;h@YlxZ(C(o7;)8A{aj5f5Hu=IRW}C7eZzR9KxwH!8mTbX=$~I&- zHbPpjg2CgKDYlWX!Dq(K!+QH%*wD%GT#JehvdfEw;5Y(`O)h9}U6elp&T>CjCJd7WCOWD zOv={?;xks!>h}M@F;k9>vCx8Nw_C*ANnI>ACms!PX`;19+Q2PF15oG;oUz@Ey_~4R zJOl06@rG)UxT`bQs!#M@?GR+MA{y%DHq$PvjrfgqJT}RFO4nc5#in-6V;a&0aAbzS zuJ})v$$bw4ts~X7K^tb_Or;JKy)XkyPw7DI)BeCV%k|J}R$bThz=1RT_8gqIa9~p$ zNFxF+!|mQbFt#fgG^JLf-33pm^Bo=RsF=p;mzA@F(~X#uuE6mpjnJcPAaKNt;r(bu z-u%XDRA(c@Oqd*m9xH=qvU{Ou%t71_Q3%t!tZdXV7QOryGkmN@jcy&HkEBh-a(mR^ zMuQFWoUt7IQ(v$GyDu>0>M%CAdjxoy4Q4TC%9e#~V0!ht*m2Dnbba(>?5CfPr<7cxNtzC%W26bWd@2^+KmEo(AF9vRtLCDb z!*ycQ>X&rphoR*0^f`EV_Gaw+e5&vr)y3y-KNK%2kB9C}7{9xpj-O3%r^GpeuiGd? z_8mUNv;&UQm-hLX@wPDT-APaiu!n*qbM~vPl9_+EWoIJqfnC%fW_vD@DLThO`{OFO z^H`g820e$}72_apl{(pBHkcI}rNDunbYgfima~nY#}D(-!NWY~F!e?rIi7leKi}BP zxZ@FQ68D%z3{-HMN4uzQhZDV`TTbgPjArInBH>+*1Dp8cEnH>k%<#|w<`E*xhRRlx z!jwV$=?QTt`^_pL!`{drc;`Y#EmwoaSuT9=tqXKbTq|9Q1DG1f}Wp4Q@8a%{GC^DGXFCw6&TW&M%{wi>8W5h zcq;Sy`GeUFmyw(=y1}N%wxT<`!l;>94zBz?oqfHV$vi7pGW(WjMqZ3$_;MTzdlx~M z46?uml}qW!>n#wv;2Xru?!ZHjji&Ygy{A*dN3cDHlLlxmBbt_yqc7WN_B!mCU&FH&bj@ zmrRylMaAC7u?u!xyaC#0+gR*_0@#t4j;~$WhWch{!IH{@ zc>8h{GIsxA@ie>9{GrpoNV%w&E3p>|Od1PZeJ2L>T`k4ib{pa)>G5#1ycez7fl2=K zmwegWQ1P>3BYZ9EAA5G>5Oa$E#Txo!*x%=WX|MHc)>1%eQ>hb_RJDLA3H| z0c?%b=X(!r!k>5kW)lmau@eeqti`hiUVW0s#r#dYedTQE=^Mmq&p$*(>%G~5XGN?* zxVN0%Y=t-a4PqfEfMvyGLR{Z7c-m=>dg_vJno<_q9!FVsSuu#Z16Y*m85piTq4rIC`D*7|cwMBjFsFXYrxjJ9htq;UdUGosqfRL_Q56E@0k>4t^!(G3k$#+jf zc#xz872;C*AZQM%ox#yNKhmi2*ldxgqlhZ$SqtY!EV#Y+fN*jw>he-Wi>4RSljoj5 z&R$ixe)KH*Z{=1}XZea>@zIwq9CVD<`ufqf#t}Gpsfu;*-c00_m4Hxp9I7>6iS<2a z;fvdfxV()`bkw}rkY+Q4nwLLwU`3#AI7uESC#Oo#*0Wha~4e2*i1CW z>w!^7BkE5diNkEOLF4LC>X1;)hxGb^)(r*r;<2&AN&nCJ%$m9-+(mr2-qj z{YDf@v;ewR(VP{2+-H*pI{n2PI4|&&emdXdcK6$hvyWSWch@v}^wLqPZ?FMVp1&S%^N-!-mOuQG-XaVvV3a~^BiZenV-L+*>ARUn%SDf zdr=!WzRSA; z+NV_D)~iyQ`9p&~D7iwlQY^X36*;s=db^01d58Q9n<00Y3DLc0iV{cGiC_Jj&2Nfb zM}IF8(PLif*d+$&`R}X6!=K%^vZ8uGH>#sQp{<%UBxR7)n-;nNKVH zgLr>7hWwP9L%Tg^phmxM+^Huw(OJb3{_y-|sP7Vw`ZJYLh>H$CCL|nYdQzHux`+nY zyx_f4rqJQZCFJqUFf^k`2fzO_9BXvH#)_97z_SAiRQ84xa+A5u6}ufK1!cMPzvD~6 zD58r>uj}D&%!@)^%Oh}7#b`LX%o--JVJ-zzL z6{qjqSDlBn{5lobc}DnR(tfL}BNXvxWk-;DGZ)ci{nRopMzm0_5G}2-Clh@gIPY!_@lx6U_22<%&J6Z=B82iTi4K&B3F_z@PX>olz^mT{26%w2Ya8#wV4vQWK{?mL;9Kh=|b*x^+dWraXXE<@f1`)%tiYXiut0W zbLi6K&GqyI{lXt{sOVMFZcx2 zO}$2wb2_0U$$<@>;)v%PCDWp$Yl+@WyDswIJ(hF&QHiB(##5h8ZGO`I`ONUqcsSyf z2X%pAd`C%mohWb@RT&wCJj*%cu&13KvfV?EM9H)HNkL5H+$C@zm(djV7pYzdV|t3i zn0~(w)OB3oG>5rU4_RKkXn`L1IcDR(si)A_4U?GvLZGnDa0+%OcsUhqco`r|cIh5rxnZ$za{VwirD7vAJC%$+jX#BDmJ+L4GwZg;ZH+>cw^!> zINYO6hAsV$&%NJAeufj$5jBGJ;ZExD`xUnFnv9OU75rvDS3~KSCLC{l4ZiM?g~Y@d zyenG`S`F62(?wtDv%qBN-=xX&E48@CgFbO87t`VEE)lKYc#7I-9$|{N^;qpybC`4_ zoF7`80+$K|?#Z$yl$#xnLpyF#$FAp~UMjewGJJ8gsWIL&I3FJ_mY|gbJ+O1fWQX#P)EreFYJj_^Iz;Z~ z5u$Hx+r^c^<@9)Ut+;V+4_*E`2PQKc_Bmc3>s81aR674Wsb3SkrKVA!|@evDW^Huwo%`KdFYFWwHe z>&3AjzbjbdA1l^s5R5O`o&neBJi%9R5DL0-XzA)47(bxJuFIG3OPZ|3GHavgZ{NwV zwWgLjI{9-^H*LYx;2rlcdK4>o(a0=b6f<|<8B9O)94^eL$0OSFq2AXQE^j`AwZs;D z%yPS0+oBr*zjT{mNI8NZGITS zC9Y`}dtT{>NG8jEA0Ni*gT5lo;+tq%pCVB?(L-I0>u6s1B!SV{!W=S=v-6{FvVR_x zxOMGODE-eFcTLHJhqr~kYm3pZrVDX|>^4T~ml(`WzZ#!yx;d9oxfqv!^Dl zOxxC+)W#IzMZtpC!esC*(cNSn2MEbN99fW)j5Ja6S^+{nN5)B&6qFQSACHM zEu4-{RjyzME_6WQoS97O`WXCiYM=PVv>g!Svwd8|R$Cqdqn8_6Mrb<|WZsD%KDjUhCjT z2ddzff)3m0*#nE64bUdrW>HndA*|+*C0e0a&A-p~<~I&Tc*WyoP~M})X74zU27Fgz zAIECka4?9CF__B64Vn&XEUvQJSYc1M^&ZSHGlV}61Rn4Xd3N}uA9{J?9~6CTggyHj zD9|AO*3rS-<2C7croA$A?jHmda;HG)dl4*cItohlfy@;gSmQr0xMHxDKHlWW_FNxI zTnsAsxCyi2{-#cTa8v{|sy&A6R(J4GJIF?5$+60wO3@38KUTFbE~4!x*6?-PgP<)b z0K#KuLT`!<_kQSM`o33=hJF`zoE_>cYMVJTjc0fu*8sQ$!L%edU-U(9Gx{W(#3h%T zu_ZIz*r{3G%**BuoIfqkqOV!wTXS6L@UJoamuJSRlnHQ#CdGrM@Mns`rqi2L7~_}8zb2P zCEnS%iBk+VL?Op^)3t3Y`1n0bSpF|_w%Q!bsD{C` z?1|izJ9pU3snQaep&`ua(JX4wGZb8q5xf<0O*^JmkQ%`sp=>k>^^UXZ_&@jL;AvSd$Z}o~)83{EFArIfTissP zPp@4i>jk%N#q=R0^KdhI@xTe)=#;fqb@3u9-ou#O0STQrYBxTZyoz?sy;svqQ*-z*Huw zbBZNaZD12_egdU+$BBGl2?-svgM3p+CCBDu6XjkBF$j$Sg*_QK@QFJ%X^{u5wb6u1 zKcyY*_CkI=izQ9;WDUG4k(ytKLhj_l*9YMOAM+QCYFr9YaaYKjH`(Ol=R%gJUyd`m zY|)5N!q$ekNh0b(Sj^K&?A0fA>wd2|*7T~FF7*n84DSPMEZxo8Q;v{;_Sx2pJ`Axo z%NlOIyn6s=-gyqGb>rDmzJpv@l|$A>NXW2d%eb#Ha;$XqEJpUIv2&M|iB3Ww3Gvc} zx5u5}w{Z+C+tYV(SWF?z2rjoV|%Sb}hQn*^; zD9mvtutK*(_{Xz#_(R7+wh{w5{izNQ^}0@fkTM)w77J%DZG`W8wb+vd6?o}18*q$q zhuFh4a8t7b1+0!Bhxa;I51;90-EK0@df3y|WYf#B)YnvtEk2Pa^ou{ksRdKn@^zOW z)ijeiOjykP6&`XXrt#pm--0!Hw1e8%S&Yv-L?XYepz6Bv z0L1}sa^-FpaUWVioRXzTqy1#-wHD*7uk;64pKeuy10zO(ag71{SL(q+GK=fxotcUM zi;}UH3DC3tw00LUC{-d~{Hu^=!gH{&m1W&+-|(O-TBz3jFtJG;Z+%5CfHPcZMa7EY zENQ;B#B9t=XdhHVydK%%Pm)?#ZmSLlvLEr%zasE=8zKMF(=3)6J)8S=VgmiDAP*&1 zWq8e^R4CIpjVmsT@jOcvID9Y`RBsj_5x#+AuGDf3buqlc4;LuA#KXj?JJH;2t>{AI zCAyIeWg2;g%skSO`30XtErK)5VrIK2_UJU&8hjK~&(uSccrj#JWbu}V--_w`Snl!p zYEku5!B=$Y0+O2D3~1vmba31cYBt*ls?=iW7o#26a)LDuj~j!cJdHs4m=vq>VxYSz z27Gi4S?B#l%HZ(5;3hPBKTM&ksR&2W0psc0vZWrw(6)vDjaI1WomF zMK0w^^uW;|xRW6RzgkzJ57(}f-+C8_)3G)jHKYI!3Omach;?yQv%vMdehmCO51@wU z_B7q;2)g*Bvla!I(rp^%?8V(}P+2^WP3==9ZTT_u$%LuQuKY0sxh!Ea;<>DP;W)dLf*JSczVK_N&eeJPhU<1 zO?_81Nktzt)`mbrM4m@53HcwvKtC~WB8grgdh!T7PXz%rCUH`@znoc>nW;FUu69uPqAa@e%@}KKs%$?oM?>ot@`UWn(V3byNazI zKlw=;|Be@D*N#BhKcCT1M}K@eESjDhp-hHXW>W>laO^o&ja{GLPdiIvk@d6!knicI z-ol!?Y^?}`X9T`_D2JAHpTKvP3;Wfq2)cIuI_woTgd{Ev#jRVO;dvg1vHTNzEDNK_ zmb=5yayQ_9#cqU-i8JWg$ao~Kj1ixj)JLsbELhj}MmoO432el*bn}6aT$ZJZSS8}X z^gI3!!MLYJtP%89F6AokxZ?Hq zpQ2T}Pf*{GcAS*SiS&9R>1N+o@bA=9Y8$?sK60D|3p*!M@<$A=-_dCBx60!3)j|C5pL#-Q(BtJ4~z~ETkLz>i8n#?dR~G zPfNi}>LffY=!UMXauO`KahgnfSe=pwK5kHhUwl`?8&i_;IhRPbD`1|Rh3=f40+(AOpfJ`L`vp6&rtX!@r?nEM+nj{2aaH(IcQ^gmDQ_(^ zS4=+q?IhoJPekqs!e{tl11OD+C8bG+h=bQ9Jp5}T^xaVtvhgA8eqJw?n_U4Bf!cJ3 zWg64&KLx5fvzW?}WZwREB7m;I3$f!=7jky%&P3Eu1>5-ju-;{#0U1{W!LG$t)HeR1Uke57H6o(in}Khoq$N+Z&m#kSwcANFUz>!us zRj5UV4pNnv`HI-JrBgxq?Iy!%iazO!!u{+^?OH>#+R_eTD>QKkuH2@Xx=Q!CMiNnO0m@Ln#cRTp|Rep4&! zSguX|0=7txqzfdKR;GSUq9IdK>8U{nu*JVvs_}ie;64fBoLhGy<*pc-D4UFeY*&j^ zXEuV2;aL3OPPb_H)o@Tfs0kh!H891w0Kch=#Z5Upd22tO`0h@{k-{C13(#W~{&P_7 zn}f9KN&;k#*M#_fA=_8q$DxtlG$gf!NoF)N+gM;FYXzTTTApb8!YJ;Bl`Ps~(?l=) zZh|FfHJ*8LFj-?4hMQxT;Djl2Y4+#u2+gTL8~TnS)n6TSo&I>RRF6SU`HuKvw*&gI z!xqhs-HJ$m~P-ENGTv$tf3^A>VC%aEJp*L1t zG<|<9x}Y@|Pc=^C$Nh}rE`3-f$~4fVjk&cD)cA@9?)c8NSwx^&Dyhh9j6AU_yo?W; z?&G7AmhpcTGHFUotjJw>7+;KdO+U=h#3w_d$jyyic!;nU9_Pg1l(#FcDbOX&7Y|Z{ zwC!xC`Uqy~`3-+?cO_oeV@Sh}6n4rlm09b(htR`A$hv+#6x4r~(@S2Dobt-ix{X7q z0au9wS~QWI`z_?O;<%`%U%JlSWjI*K%*Tq~IO~uzw0RT0lJSz8KiU@mT@iUOdSu}Y1gQ2#6bwd7NXSsGmu7L1eEJd z!5+pLctx*XlR_n(+g z-VYD1Cd0OM6Tq#}1srDQ!o>T!5Y_bxKOar#S%H5zO8z_YaD0e}xgoWeo6Cw9o`m+0 zoiNGZFn;{|BwF-6nSPWPa~hquxS+F2@YE(5-CtzE+vd26KT18LvGNhZ{>@MPi5Z|{ zBj3}gfD|k`v6$<+6VE+cpI2vkIR!Zg-Sl1SMCigVX~9`QxCFnIRC%E#)mH96E{Bhz zkNOv(cjPXJ9C3=yV#?xI#8G^CLzKAC@)O$l`v~56^ed z;uI=FRt2=-94iN*BRGWC7=6Zy3*OLQXQOGB&!p#- zcfy9mQ1OgDBdQUm45}AxaF@J2Un)`O(>P_EIx~K9=*g z@7vMjn$1w%{hG#ZZJ~FhCc&}d5iHKEp6XnP$9!W4_3Dns|7`6@Mx#5K`cD;A9Jhea z6{G3=S5wev(@$uFv=N?>)FD1KVGSLvSc&!x8HqCI%)_xaw!oh6S8HCmlA<2+A6+$?lNa^ zJd(!!nFLm^?;$(2XyNbWb56n0uwz!b$Uk%xpU}Pnsh-nExoRVs!SrPCFu4e)Uw=Y+ z0+(7=-v?{1y$HqTKkJ%(eb9)WPv}X@3>-Y_8J%EIii0Fc@I6|FJXAP<)aoLkQNX{P9n%sYHz_jYK%J_yaa&<^6;mS~N843`P>h2$KtVg`Y~tn%smsIzj+2tO?N^vSMx>YNyoT_ zM~vdV?ANc_!)3tNt(56HKQMi3Y#)vlEBg^XUD>PVCAVS{LIG_ z*H2EyX*+c>>E4MdwqK$f9-c;-YNmMoe+e+?iJ|bO8&5jxQ)s+a8Ket-mEFrZ_S@te z|LXE4qFgqYR$5A-tfeX<^~M4I{=wt4?fGmd4=+d0O^s1m7=hdoTjAc@SM(r`gy&Hu z^u#uyqjY@}N-(R&k7VMA)2R`p#N`QhcDoWz>=CiIy`v#NQ<-IaR0BaCjt3w2$11BD z@vzIMnR8+i>z13t^1c6KbLwBh>_r98J>~|4y^O${VI1)e8&3YHf5Z=TO^8gp3Gx1< zPp-c^jMmSL0XFR`Hkc_#hP(`c`>Fl-M%R3@TG#_;ME@4;ezpMLpHs`dX*h(MziF@) zNh8QXbQOzj)3LWtJql>jgbzPAl3VWr$=V^)iDlMhJSx2tEAowapVV-CM7IjXOIG8| zE6Yih-9>CtScKc;Hjt*?Xeg>UL5J3;fOirW+7E+Rd)j++Ebu96m>P?hCr-j|`>t}z zwXO7-O$VJCZZ3)3Jz4VUatCv=^JLbm9%0+B>)99|2`_TXMBX1|iCpPf-t3RAIJ|=j zY=Hu1vL!MfeV00u3 zSJZ6d)p}lIDRu=s+~t_bKg|A%YZ7>VsVsFA$1dIVV8d=((Cw9rz%k@4Yz)`{+unqs zDx0OiRtY@Sdv~}uuMOe6K_R4ejI{RHSVRVo>LfX(0{6UjCKg8|p|Z{sa5c7F;EkPt zJ3rpjPxt@Q<$A-|=|*knzN7yCn{g9!vb+D&j{54Y*13 zE+1sxOPDB!NPBK4-nZLH?~>PK|IaJrLl}pMvn-igs6g!B)Wcz62;J8ahCLnSth)=1 zt!GcWO|D&u6UYzc;J5!8kbC9uq+t?UmG8r{Zs-3`(Rl}A`MqJ>ULl(jvQnv#c%O5h zHziVOE2W~HwubUe8fF?+NQoqo(NxBB?xTqIPFtz8lQd{2zvuUtfAK!&dCxibb$vd< zl(W-W>aTI)TqEgjF*TBmypjV+y5FI-b0uj~IR^Z{J48uE4N9)2kdYY~{L$1~Sl;bC zdWQGsmk*xlK2IBKJA zNlQl>Q2TGH^3CpFsOl0Ux!;gqnC8#`vHxV$Hr4=Knxn;C+3iG$568|yS(xVGBRs!c z4i6)c&A+4s+iwbTvo)^rg^||s%m-cNF4K(#xDg?ntn^0h6M?+d2U!|gX9`^ivhneB-oyJ}%! zMkxOt@69Imd&nH>RCunVIW=oe#g_UOw##iY8d$3HmCaFn%kCnSVEn8i_&L65&mv1C zzPU@$QD{+UU<>WN*sB`}yw`FKc)m;W4s6qd=*}&8)aO3S+nfOqaPb-igiotQ?CW`V`E~xVH8Z)|CFJ3SU6fSza zU`cBw4yk_uDc;x&J}tGdpv!c+|H=k38yGBx*o(}ZL#jT8&t_H2pQFLQ( z7iw9kCfBF0sO-i&x_)K?|5vn`I|tREPINDto*#^(%{Jq?9=mZ)A4^iuV=3`i^qI_^ zc?c(Y4~LU#+BnoR9T#lNWA7cmvQe88D_zTGg3%crUNV0y-qZhH5%_c^4sqN70gK<0 zBg6c#x7!}{QZa+K!-sMqz6g(vZ^nZvO7d9-PiXav5%O*m)<_Ay$@I#TQIxvI^I;lW z`OcGCT$bVqYq9nmo2*F8%GGuk;gDbcB zo8tHtVZw%wk)k-{Eb}>_$92cdXE8;SFk2^BEZM0cyvjwH$1Ho=C>ahdEiFH z(+#3f)T@Mh_x9pGhF?YQvQqqEFj)5TE0>LY@tO>LZXmRzFRU#8qD)V$pU(Te)Mnx5 z0|i#Gh&^4V&YnlAVwKXa%1n=0;=K$AO?H=4Zy3@Ho&w%CYqUysjN7*8eO)(5(hnsWd-k*`J88laLGrvvZLlU z+w(cLB2&f3eM|ICdcj-|>JQ&=o8_;7K2Rg%^xP-2mnq6~vW&Tz(`E7U=zi$&E0@)- z;G&MBwY+IW8IC(SOPF$RFg8|s@;mT^$*ir|)OXv(RmV+*ap60;T2wJM9O@>`D2B3h zEq7jEwT{W{%Gl3UXT+2G2gIo#ZK;+{8BY3YE83oUK^-Uh2){aKRGwLSiF)nY$agu$ zRc<e=8+ZjH2%*3~U$(YuBi>YBODK4&R^aYb_HIrbAnB!Av&g)%VA zSW1>plrpZLoyCtoFN#Tet*MNTx{3FCk0jlCjb)8LD6DLa5$}%I z6Z;+LLt8dvNOSFS!R_`U5>he=Ur%}^uHLa2U+V0ViMxC%9{cp=OP{3Tj$8Ld8^`P9 zXrFEvZSE-AOkRi^*0#xZg=`QmluKQ+ZwD&9*4f~!h8rlfNFib}Rxm1D#ux0VB;6xC z1nt#*Ss!D~%KJV{HX!F11Xd{XhYPlmg*SKOqONahM+RkAA1sY=1 z`V=;Dq6sk-XYsT{`=yK1nPU06PSM+PkU05zvUq%-2QV!Sesgma3p4s4R#o0*y1t15 zv6g24y-%@K9YsRZ)hH-dv}2_*U@znQRC?8hvdlqVqQTZzsHYQ!n`_F^f8YvK8TLus z;$A3*ojAedrMG3hR7?j2tyh4Vm+XZ1;V!ZZA*IaEdaW$4_lL^Wii2_Q%<(u#(SQA7oWY7P8RAF8tr? zY|;GX&C9+VtmFWIed`t6mu5wFH-c z9*5KFq#ku+JFZ@%ikE*+;eRgAW+#-A@ZXcBWZqDz3jF5}mg*kI{NXY9-{M^C({~IO zO75%nCdo%-u7equ&+hc=VO^HbAy@>QjyQ2V|m58L&N z)elV(mgvtAbA62Yl30nw@+*rND5M~WYa}yKd`{u0ADkuQpi;Wb~Xd7ITqem1mHJaV^0mVe8GWt(R* z#nbm?Rde^SG0{(0*jZJ0-!n<{{au829Svb!*m}r)P=;i>Gv7SF4u3jb0*%M7N!6re zPOE&dNMFkG1#ZOL>`WYNRt$Gm+kmVv3XNC3ChKi(uwm2kapRg^u zjw}yOwVJ|!8Z}sUKA%muSL7orf8hqRMywvYkKc@&!>4|hdiPJ_x%Qn0Y@1U*SZA+M zwRnv!4_lYY`(Ce;-HaPAe(;VFj5m4nqRwD+J@g27y!OY+hwk`p_95ogG>aA9-oR9L z<`X>w6Y+v!gv8tls_dO}AK(6c!>Y!(vu{mHC>^Ft(?=b|km^3GbVV+;wJZhyOP(f|A9RUQhAQs87zn!%2oDQK+B0TWXBxXPp3hAf@%^sJ*&@q zef1`OH#cF;rwC&GX$y>MG+@Ppo5g7746vW1gJ&a8;)*hBR?}WCzP*?w8+zgjv(G6O zja~a=+k6kHujJ}(-(1Wx_0pkjKm<2g6^WYfj+5C5_HeiR9hQIiEbHUYi#`7~66+F- z_`EK_oo(|($DwifuyhaaqHDuDHhaPd`yP0tyb#8`P2tY+Ui^4NA~xJK;0wHu2w{2) zF{bYn{#8SV57X|-yHsb03k|gBTj^|`aO^tyZSBnc25rZ%jP7*c)e<4>UmRQsJk6FT zeZ@VBJvgM7VP*0>EcoO@(q~k_sAv9MlZ};nevv${aU9=a9Lk^h>*1W3Xko*-cc>w+ z5^}OmvY#>k@JM|(>=wKkM_0PwXqHBA99Trt(=Dj8cPeL|&3wwCg?#F2DP-fn6|C$B zvb7y&d4*p-Upu4)tEQ^*dMW=mcSbzUTxCTe^%CijoTlTAAG6IR$~f)PJW))F6eq6z zC$#l9A-k1a*rsuMqHn?gadzrLHriQ%25AZK#qqXGN9jOi%q1tW?Y2I2lxol?yEy#& zwTOOOe3HhFFN6jy9dg~`6xpY=i_KZy1ti}#D23J${+Pf7rx?0@dM#)ueic6~Qj*@) zT|n#cG`1}Z*xvOnWVbZ;501GYHeoOJsUx2pDO@DxgzhHxdLy9gR}C0(TA(A7{b}*u zFlrum2m<4v{j6y67cp&6f^5l~6J$+-Dk*4LB`lxZ%u;8L#l8owqVqO$rkAr7 zbPe-a+V}a?>TeFbO1(r+9(Q8C9=5{Y6??=l#ckx^+-k7v?0_x_DzHCrH6-_mqT9M= zl7uU_1&h&%m6ikd3TwM~zz5fnP&=iJByBz_1e_Zu)Wl8)`7S4#7theLFtL1DLQnT; z8$5+i@4E}yM^03Xt@R;xil5=g*i2bn+D&r(q8hy^J>OH#X2UhJl77iD%v+cKc!h z>($wng|AM7#hEY2sP6~KjlWmPa=#cdKlqXOcTWZBT=9n($fPdWY=Oo&4}cdone1iN z1Y&waMxJ&UL$tEg(Mmrf?6dqOo|!hd;!&@&g01>7@!x+|!s6-b=wJ~=BF1lY@4mwr z-k6j@;{9aE+;Wp?4185NcIyIG7%zPxR&5nGjZ3H;Z(K^UOf6~O8O0pzIGn{>6R_c#WMAgAkO^-di$6p9vKaqDmhwfBsxF=j z_YJng$L-a`^phIIC@ul3ctxfaaguiVR12P^3K-P!irLqFCJ)@r;20C=}8yVNo{_-}stajmK~Hrhunk)T`NrmN~lbHE6B{ggk=w?)ymb_3hv=paA1b6!7Wi8qA2bVEQjriFL#aw04VT{YJ;I#%-5a zr<)yiW_)D2<|V02fuGf=IboZv-QSj zFg_5;Wal)zvb+>`Klu;Ac`%%~ya3wzrC@`}G&GyC6%QOgis{}lmG{Rrl0k1JrhU>RvYYy7q`!p+vEyJLoz3+8TZCloGNJisj$pI6m8@Csjg=Mi!6CU1T6x8j z89^$LQF|QLCZB~jhXz1I(r)BA1^A+QiWqr%3}%qG>}A?!vgBM1+P+;5T`V54K7ML+ z$a`1%rv4yU{<;iPEVbwjt5VRL^&k4bvxVukYw^(HR0z~P2ip$xrP{3-qBNHP@9`U; z_KF6Kk6I+#q>?J84&TRvXFBj>4o2KhNJcStF^qar06tp}@u3zm+|a`Tn`9#}9nTV? z7tfv@XqE4OdrF>^_F2C8gg-2@-6aIa{SYm&#vKe~wXx#4sINU&i zSE&Dn>mwfH^EgC7tDezLMc(4IQvT}dG&zk7kgxqV1Hb&507Y$wadGuG++FwuFHebq znDv=JHX6#qhC5PH@dsZWT}>BtddN>DeTUTDW9ZJ$5=Y6i8-8t!U|s9SgTanZxhJGXV6%I}0k!M=qE zHWwYh{QMgJQ(->SP`$_op%(knWjd(=bTUU&@z6nc$t#U+Ic&-$M)o%-( z`L8dg8q~4=tz+nP3l%!IcOUuvDt{_GO`}Ff#^Ro+QYV?kJx{~I1O z`Qe)N{ps+cZLFj~M%9l!lh@uh9o&$4ORf~MorkN<p-4vh$eL`(NyDJ;KP&@=BK2UH->) zntb9;E4lB>Aog3BfxCN+#u&3D;Blro}}H_%#oIx!Do($Lmn_ zmLJ*j+=ndsahTdwI8mYM1T-}(^4hfXEZNzd?tE${7q4%XuL`S{_k5%~xYFEW@bdM2 z>0iS~bZ3Y!U7K1@n%Z6qUvHda2Rc$vUh|%b7axjSUVF0iI3?mJ?_sY}C0}Z| zI;PG3g^Mpf#zhyl!G+ikqTV(i6CW1B!RMdY)0r7$6@N(9*VNHH?Z;_+oE;xEJVBoH zHA%j3!eV*WLM7g~P0q{JGcY#I1e3QPrQ7EguxIZI(O%;+))xix+Si7H|DLmWV2vdd zXC1F{o-@13-*;M-?}z{B(W7egj}|ifp4lW{X%1R`*@>)=8qNK95Z|eUV%G-?Xy?Fg z^vK=@XqtLawy3l}9PAq=PGA2V&7|4A^2bIN<#vmvy&Xb5ere*bedpm`K{ON&J_qh9 zLHu=~#1Z|jz_o7c)#e&@So*FR=duH%F`@!aA_qqD~Vh8X*k&2yJTKx1{H=l}PmS(g)w^*-mA|S=)v(Wd;QNL@_^a_TpXRlR@f4ASPgLW*))t|2 zYYL*mDm*g24#=-kOfy>x?&CA)gEg9TXkrhHcbXtMUMgTlMmf~kn$aoKPl1eog=?^i z9U-Tv)(91D|Fn|K=oW?h6N4a3(V1Fo-w*jl4aA|rnp?0Pq-|X+418S4Pvz@X^*WZy z4Hb7%&4_(;)sJ5+aeNUeu)0a?8>87uDet=Ocnwz{Rm#6KwbD?!*Fl%Ba4wap$EC|^~Tl;jUx=8|6wKc|^2W`yPXe;a}u;Tp& z0gQPrBd!NdfpgMwa#bgneD{!e=4$EiaGf77d74q>F?CUui*Z2JknpAu1r@CiQ@@K`qD+A!PuE($<|qXBwGWb>93wk`GnC3DmH_e z%{v{im42U_d*{*-)y7;tq93&_SW1t59t($}qVeji0L&bChYXp$1BT>H<^Q1*4w<9@ zx5f-2pT`dpMwP2Uj+H5$dQS!`zm{WIgeqU|dQ}*_a~O4r^@7)5KS-RgbCtCZx>9?q zl{9RbK4e&WLSKy>HlgQ32y8W?&9z2oadr+JKA+QnWBSQYUbP~=TYC>$%VKa{%_Xe; z^b2!b_TZXo7y5HT3Y0kPWvki_V5V@NdNuY2&oPmp9vnpU)G0ArR>+=@dMs9b|G~Ov zDhM`a_AtUSpM+gBm47xcq%yBWDt>N-Wg|bs=)x%axlrOOw{kjc+&0=*>o7e$HJ7%V zq*0~Ahsms%Y@8_dlXtB9hC5%+K!XL7XpVRpT;E=*P%m2#UQ&O{=H71vOEaFDyN{39 zY{iwYB+)s?l`vr4G4}M<8#MZ80q(g;APaa@`S#CzIH>Q)dflui(pM#|8Iq5;&P-rA zl5_UOi37|iv_*1=3}gGpR*9}N3Sr-z-ON1g3~4>LQYi3Lr)$PM1JhMq_%ex8aete+ zIQ-&VkiZb|bg&HsuNy1N7_tje*DBJKr!Pq5I#)P;r44eLdeTGw&u~HU4CJFi;L4&5 zu`M%;WT2(!3cKihhog|J^NkLU)4;D2O_+c4XjIb^!J(@Q7*@@|62%R4$*^MC+|j1| zs+2Kv`ubETS%Jcm)W6U;w2-)qi`b#GCNPW`iGv*D@Z{m&?BvM@g3KTv{hH=7{dw2% zNymP1Ns$qZGNhzx+$7jPau-bYUyKiS?Lk4IoCIIUVojUdSc+;8i7D*H0_rZX+?~56 zV0bj-{k%r>!H6H6SdDMQDm;JZ2TRJ2z~tP~Skc`QU2LN{>EDkZvwM%HZ);Ft#eM=E zG0-+J3*(QQV1%a<9M3GLG~yCHZ)-!@8FjY$?QY&LE)u_Nk0se#7h(AR59G6Y0vkH( zk7)aMEpxuNnN0E4Vo!P&k+xrnPC-rPSak>~cz5%*V1-k|T-h*v=eTu5mL zn?X8o^gsdItd_^fJoTy`wFWrgP)~aE%{d5HFDCBEX0W_>CNQdpa`kDbxO4$|xK5f? z-8JJA8*Ayvge>Yn{z>nc9rW(8GMcc7^Uv)XwC~#U>{MMD`#am38citU-1DSH*^}N4c&tJw=JoYsW)A}=O?jETMq{dzr&gGhhWs# zp8Dn~0W){uqiH{SbE6w~HV)^71_M}UVlw+0oWKV+&ZSo+&g=Yu3$(4|61dLDr>QO7 znO=|#?nk}_VJy&$4o$xPz$F+ns~DAQC&SlA4Y+oe1|PrF58pp)z!L+sAZg3aODv+V`N|G*oqbYvV^n%~R5HtQ7u=tD!dI2-uG9WTV=U-6u=f&wf|n z@%M?eOZEy_vDluDl$+6Pi%>Gk|0OdmaKTxNdPAT|8OHjIhNI`U!0jq2vs8BpN}c`L zm-*7nL-8nhhNsYFmy78RtE=SGOl7(xXBB%h`?8>$8V{pa^nj5KG8{PW5@?tHgPn_u z0Phcj6^)MItab^m{&j(f5nIWdH8)|S+G?6OLzi}0=mmX_&*SFTRB1sF(0#)#*+#z- z8q)CxQvQ>kqvH+ydhc=M4vXHDt($=8Ybnh<6?C{wuM^Oy{Ao6r6|m;lyU zl(cW4HE}y-1HQKJ^rxs#y`Y^_n)$-Pteb4g-SCVUQvqJ_Rb+(lMiHlUQx>gt#ZfU}WYHym!H#=gDu92P%We z!%M?>+27OHcK0c^q#Y-j?FxMHJUw*KwBb&9KuM{$+}lW(?r46?7S+3QEchYW-lNeX zAq&2ZyN>yF-SA@XN*cV!jGw!G1}C4l=C!_K`3s?hrI#vG@9Y%W{6nYNA>U8Yh7s9&rH|1UTP4)5&Adn_nHu`>e% zqwc(B&JI{HXArv;l?WO)w!x+3b~5nxC=heLk<fy_|HIzc^lUM1c3@a?bOp-$ zhw_Yf{rHlGRTRH1X2IRLDz1#VBj4Sv_51l4m@Yg+XL34>CONe=($GRS|g*o8@5u#0mb<2XFpz& z_zOvAf1D@J7E3zHMVE6RPK_{tYm?1r+uJa{ME(X3bs0cYx|{R*@ZDVR#Rl&Ay&E{W zog_Zb+{Ncc4>+-QmComr`Kjkxyc>z8;n$SpH$^R6?=ue_#=Bz5M=#mH>{6UM*b~z$ zTu|NSo1pY39wz>|L`v#)_{kgT7?jb8lQphj%KmfczqpDOh0ef+Zu9xyH^Br{`(Ul) zUHlhj2Dmkzc%~eH!QJ2E%g-md-vST(b7mw?np4j9UUXxI`D(%^r3`WF{&Uc7Exl_u zm9W&7?d*CtRgi6b0e(D@)@(AB_9^;2-)BC(yJQtE$>>54%?O}%9+7zS)FJS+NG8!Q zl|ak-9NM2B${Q}8Ea91(^3KdkW7>Yn(}B^7@>&&JC<8`Q>m!K2;g+)rmG9;5C&V%IX>q*nxUQq4$o z(Leg-RSJ5@$}vaSO7jB3v3FyfFwRe6x9=_Idp8#IC1%e2O==TQ4UDIzXeyujrwWuz z)Zx<^1w1=fO~9tVC|M2AI`1y6->BWjuQAt@AL6%{`}6KgvUzdU zXx={Z3$$7L;Pb_WVw*-D?C}bN4*&5`y`hQ8AC9DV?`rbh>$Q349BKWW@6Syqt>PX3 z_HoRO76w0BfXjVuqep)~s_DV-#TqqIVgHWZ?rdbXd$J(#VI=0Pv*WiNW8qlHX)=G| zVK6uz056o!Q_sWIH2L6rdfCx~WlCL^&$G7jdu_|fh5Kd9=n~;eRy(tG0R!=3-$G(~ z`Yz14@*0*sx!G55_&v9XU5UaI>e zXlqo7`#tW%{uyb^#I85pHY1(3_R7cpKLoz_MR#_qHZaG?kr!&@&5cUzv4R^uC>zTY${ufugo4|cf z-6y??6LHt2^mdspb*!5Pn@*==O}rC7I%p*t8r0#yhmKgRKL}?pPQb9jNN{XDAWV{+ zq1PG#N4=}W4@Ww1dv`4ysF02CSNtbjdq(8TKwtY%tq`|5u;YCVEn;S zFmVpn%N^yYYm}dU(-04QnQCA@^SQ#1ZaAJSD_| zN4=K(+CR2b&i~~p1h06?)HW8dtYN2E+uLL0jg*O*b3mUtacADVDV~fCD~A!G0n~A7 zJNc3jLb}^8rq!W7{7B4Z;cwS_WVP*QFt&1`Ho;bO&8O|4W7Um0l-twY_caLV^aQum z7GfNe3&FV(mrLg=H2REVxtu`D`{9z?lf$~^Sb(g_bZO2xe%c_0zp+c_=Z43@g*X2} zu7esY^7$+@{VjxR2NufeO8W|tC!^Rd!%+5jc`>n!F(e@wFUUrgBg{4!CpqdC(EmDT zLbnl(D2;i^*rd_8$KD+CT#UF{^D(Gn2XS00>)8-acB&SQzG7w>{6d*)-uBwf}xqJr!?9$mR$ zeOFla#1RgJW@Cl@E@*s6V8!97Tss>0c>U8j%IOj6I;rtZyZk`-EoA}9jKSx{L-u>a zN!&Mk06!ry$Tw_0LiAcpY1qycys?{d)$05W{OIvD;{CEJi1Cs7Br}qkgVH)Ubn2k!(KU@>u^rGskGZLTDgFL+D?MWv zOSfzm_)008n{KlchsWm$8nRuvYwJ zafV&X{~&G|yjIlK8!bi{_>y^tEiv4_3t!W;5m#@U3PG{|f$Il%dR40z57=>=7q%SW z#YK``%lJ4~EuF}_rBC7q*Qs)=-_O{ULKmnDwuaDEn_i2e_2dCRWrFjaOdiRc?@uq}kNcUxlx5^+*?=!qUNvU6v5><$Zhnq@gacNwC zTHZeaj4r71MOPEZ1Vd9|;_#O}ozu=5G9TmanI3$$&M%lhVFcY0`knl`{(%`A%?82C z4%c?w%ngS9;%jbw=I_0e_>gc-x?W<1zB4R?PLu5*cb^Vc(HoGSDaDbNJ?TMfEjDfP zM1Cn|8rMjji#rPfvEsx_Z2DP%>he+i=$8I8w&psYw6~6HO*fL9L#NQ=pdH<`algdb zx8+q)JLu523#eIBJdHoL9$a6flbznTu-QU~ueIwgan$?r()Q1=(zXXb@I#wk)zGCE zW=y~-=8QZ#^PP1Izf3d}@1kw9J;^J&2j7~)NXvjjG^1}6&7R~-FAN<3gPbmqlF$fn z+ZP3k+6siJSBfxqNgXRq+S+CX|t=uL-t^Z?TfA!I@`hoKS+rTfwjLXK+jpp>;dV((J= z)k>8nXKB;*3N!dmuLBr=U<7mo3*7a4kE-DvS9!--Q%E;f#@SQ0!@B2ddE!VFv{Nf0 zvKWbBGb)QD_OoXWF?MkGIur9tyJD2WF0eL|@&>NoAu!@UvhquR{OMYc(o7qjw{GDc z+oM@g^HOjQItZD8i@;>gI?_BT6JI_Zi=F2c_*Lx&yggKjKTr%nr~8roN<=h|jvNlS zumf5>xgWL2<;%pH~ubhkot7KIgK9}RCe`?fsd?I+&m5|;2Gq8Wq zNaj|178)X_z!;_VH0bhDT(a>PGg^5KKHDVVDYZy=YjT$LGT4i%m2!#6RlxeC#`A-t zukyeT(Oi3DCf&Ozp0@uUMVn9MNqz%Li<+X*`{#3#r9VCB`_T`8_R78T_awzG}&ka#1C%iFD+FilUae17W{Io#(hvp7DHSoaJkUA(->>9|`GCixHM9#z4m zmv&cb%xd{mdA1WmqTRJvm-D(F-Dx@{EjSksDjyH z6f_lj;#{4bY-gSp>wMu$E(}eDGOr{a+AWp8vtGdN`R<}SG`7)EMSZBgjw;dfy-im2 zsjD2_yFU(ldVodGI0*g~Tj;XwaZpiKOLpkjGsh3L!pTiL#D`ml^UuA%MQVNAl4n83?pYwA?N3^pffLl zB`Y8(1*=lmd%85X)djabNEU8;Y$IBUTUh@+iu6dQF*|rMlIwKq%Qb%w!G+>AuJ=ia zN1f?QK3>q^$-cdyKi_6U;T7e*rP%0n12YAmHz>D9Dw!4``oMBZi>s6q_L!{KrXI&&kBbx z!g$kdvI4j6l5gjoxOP-GENcEhh`KE{N*&hb=}Y0S{b`u)IG-iwosgaVGeO)`dKfjg zUq^ojLzolR4Hk24faKj|thp(B{>i3dnp2995xxv|6|RNsgC$S^k%0U=W0(6dozy72 z+tL+lyJ#@)d#$paWeMWCLC46V0_na$23Vy%7l(`u$F`RXaos&lcIAFGyAgVngi3og z-%UCET;^;h-pU5|=--uU>Mn3)h#f?&Zxj zwGmj?)1#tfOZUpuEi)jd!Ci7*Zh!{k%|z$m6!PfP8&Z~aLFnZb4nMPhF|B{=ktmGA z0n?GK8sf>n-s;3Yp;>s!&;d`IEw`Tn2VkoQXiyA-6LHSk|PVl-6#levfM76ow)GYjc z;vV+$J_6U~j36gpsPXhbDFYl8!5ZT?q6AqNjV4sW=(w)<<_#r-E}ID7e)Xt~8(~a0 zU)G{jGn~@LlIwkUAs^LWze@eyPTJ?}DfqGTAKSps@_yG`c!QJ?f41Z!v2BiHU4I?J zE{mi1tXdI9`>mv7TVFw2)olEvxQ$(!S4o~tTuZKvFGj^#@0pXM2_L(92z2dGr2{M} z%s4y%avpobn?G9cW1hc|9HPYE8zC0G34y_#37CCy5&twI2EW_rR(;=fo@;EY=9{NZ z<6~JDzVwzU*_NMFGMtP_d&V*uDv&ZEK}rB5f1$LtouCWZWUkAbd!5TI8|lE*j6QMjpV=h zVBSz3hh7_Y(IE3-RR%ws_>1|Y`Gcz>Z2XW&9`4hnISm`&c5*Qpwk;Fmw}0Vf*WTjE zy<_OY<%)vcm>cNorNldD-Xz7<-$cuu0a9n(2`%Fvq3U@xddPPq^uJ^e+ghH$tF5D9 zL&OT`vUQ*|<4J_9_tNZZOeTqczOORQBmnA{Plad83(4#rYbbsXqC1T5k}VaP^rhWs zhl2yX8Dr`< z!WQS1jfeC%d(cVzkNF=p;0`XvY|E5Tv3qU?u}^oSUl-gb-TV}o^Qdq*)BYM{;Z@8$ z&ziiw9)_lHlayy`z+AI%OmW}CR+=t`Z=K)qdwWY|rCJkl=<9`KF6|-a@bDa4DxBKGg z*_qUEX%2LmWR6ceB_FL2R^^c!TlKBCKGi(+4(gXKrq0{z#QWzo!8zu*`?Hmle$~7P z?}Pi(lgpRWQ91pohnf|&yEu|RmiC^V3m6t&(PdK>9s|$A7vbedH;8HEuz!*->-xGk zsfPeeD@(+&DSz;q<0S57l);y!bnwENJ9)!r+bV@{6>3x!18=0e$0g5mF;7>Ec@MDQ z=UngMSCt7ox>p8PE4hPPngPT*M^WdUi}{dTX%?aMkd2YL^+QhsDrg^IuX~;*jsJS! zRyP+h_2XU`q_Yst^)C^7e;fy$gU1lpUYxCK)Wa?}XA%4QX6ExS0mr<*jgJoB!2Qls zpww*vE-h<@YNsT861W8~J=cWM2G*i7A2EpTS$~U_%zp+}y>1cp)z?|KSsP*U`4Uj9{SMy>H{khVf1KUWg?~^! z!E8hSu~oNIndPD&@m=U2((DrjHm(b~PrNQIX-Qx`f6n5?iN^eZx-p-ye>qQlvkA;s zj>kz-hspnU8|L4L;~A|=xT#lvRQabvU$$qXUF;zYe))?mIFiRZ_Jwh$_u2eepa_Af z8^N~@q&sClI(LUAjT>_hTdYf%tBN&I8Xg1XYQ3Z^&KTP0$RXu%E_QlO#lhYxY@(E9 zwOyhC&#t&YQ^Qkq%1oe#*Qb(`R|a9_R1t?vE5IHfl;LyC1G4H=1+lkSgSU2#$H?PS zho9JzYcqGk*`NVr_H$ic9#{#vakXsQXiXSlnguq;<8k5ZM@(Fl%a(d|+4YLd!eK1w>`Z+PLH~r|TCQTgjCz*XmQ+j>zATn>@18O+3r;w6z18-UM z5r1n7#NvzvGa*T^Wmq_PDe1xZkp*mva;5lY=pnXhWguub_(B^r3#$$r!-C;+MdNc{ zMT5L4S7r_%s$2Xxl^ z00&0yfStd_!gjYjR8?<-xrq_b7_$j|mm~{=Dyqa+-Oq!Xx;^-wl3X@d)o7Ed4W3X_ zAqqz#WsjDw6@-tMQNK;f@i2)w4a(@p6q&-TM*VIeP^E>BOK({#d@u^(ZcH4dvGP+d--O zX!5&R28(V4R>f`VQx#ypwTgDmqGv6ldH)F==)JOlJW0$Tb#aE|OaBz?oLoTIqXh8m zp$;ih)?EBs;Qsl^HrQuZ3Jzt%VeE2CzO}o(&~2SPOPSY!2d+$@Q}=Wy9s|qyf0BD; zQrLC8DQ>5$MaBPd_U7SKesBD*v1G`sLWW2Q88SR;t({O(Xb??G5lxzCRM8+Ak|c>x zhB6eQ4EtI4+9{e0O*D`OsnD!Z(Y;JI+t}X*SLW88$O2YA1J8{)FmlgZn3JXo>B8Lgu{BS@FJEBJ z^sZ!`;u?0~_ZZGs>o)qP7DczXj36cN8hUC8R?X{`CxiUqaoVlrQ4*ub(QT&!M@ z)8(~EsCl0s-h5{Vnj9L71autEk9Q`+1xKD;LK9vUcnC*Pf4pzGEZh7!fZ-a!&wDGL z9l7br&N<}4q7SKX{Ba8R;{IMxYq|u+%FW=iMU@P@l!2v7exk-fp4h^58TnhcjHIsy zOr)OExFci0?D7%lb}bj1LAYcBA}WUu!&m=YNNn2A);8ARZy)8!=@$m%t6m+R5Pls;T4}POpDJ+D zgaCmJ{0^`D{tJD$afQp>n#UR&F7R6~nd5Vg>sk8Y&+unoJyieQL$=*4BY#&O#8;y? z^ZjjVB-u9*+9X0Y%TR$NbANCT6g%BicOj#P-N(5X)FpnqCrkQT-?Bd48x;2ZL1(iw z(4<2Tnf;!LEY|-wjnnzS92?zO{Pk}*@NYU>esKg|H*GlK-soYO*={7J{3rkP^L??U z+*ZzQw>}suZv&ajzqkqZhP<}_Bc}2r9OfqMVe`3i_GjlMh)YOj>xS-QyZ-e8sLWw9 z(9aA_RKfPxGeZP-@`N`qz#iS6D;(MG#i++QC@qIk#S`5}N&t}>8 zr^7$JOqR7d3yms1PMm~UJWrkT$UMQ6&Xbp?M{cje|4ORqsjfrZ2H7Nbz{Y_f4*{Fv zatXV2*^@@O`Q)smF zbJ#K}KrlEfg+X}tvT_*O{g|r_HiQ#>NBKpr^7KsZDOft`G8CXC`29C6aEh3NqNLqm zyXshYk$HpejyR0()@0+mQ~tr>MMvR#!7K3aDnzy_ZCLlp1Kc2=z}mk@;$vy9B<=lv z#th@&mA+7etNqTFv|fUDz9QyjEzAXHJjX9nE(ZIn{2baE` z!E1G5aBcSqI_Omq7dkW@beHGjzQrq<-C_D-m8Sq5%HJ~ z)h5Shry-@1AE@uIgm3HEhwqP+X0^UcAVY-^8qDFp1ux)%mnYfQdy=IG%Y)evz;3>$ z;j6PH`#La-Wt_dtE}hDv)AsKL2Z!_E?zV_K8*vd*=3ZvrGk&r)vkKW@eOG?Jn=z;A zyWOcW#0L2dO{Fj2O5>vK+eo+IT1oQN#FwWZ!^vmdA>sNT`XxIBb*};^&$%ag#YuZ` z)SEK?*qi^v0lvXp-arUwoxV>q2NxmjNt5~hBY$D*{3G!CvFo8@T4*VbtU$gVZC2;Uf>@#V35C`F58&F2_3n@>fQ}@FHL2{VYd_Bb1UxCb=D!To`McAqBMh1MwvM2Z7Qdfn4AW5$kKfl$6S_{U)O?d~lDB>Dq zrR>A%ufo7PUI+HhQeoOVPQi@D>zJ}z1n)JYn_uAioL{^=MyPEm;=U^%A>y~ zZyh&?OL1t!Gfb{Rb7BSCYxf*qxT!=|R8J6Fq!#gA*PZx52W6d-TQx}4^Nl1#Lqe`p z?}Dw}c@WT7j^m!%lZme0@cPOetdaf*Ul{|;3-&PYw=2M6Yz^0QaT$H~^grw?8OA)H z=dr)v8~KzbIr!rB0kY2^cCR}Yr8@(?Hg6xb_iN&}p1CXzni7B<&-jRYC42adON!7W z-T*hO+m3gd{9=ctVp(&xF7L9~l$>5Z4SU?k$1-;qosrL&?9LX37Hov&rVEJ4{4u0F zGlo~#I!Kt0S7h661c(l%4iXL7CM`-mWJPue{Pdp-Dp2lQA6|CEDm?bNEx*bt1YfF+ z!?3;$CT%6j^0DcBM7i->+JuS((Vazv@O0Kh&iE9uEeG`#!XN(;e>h(*mk} zWfFF7$iZ{nm*ZpwA)B$@7`_HO)BL+lsNQu1ZS*=w|D2rwP=1zfdUTVQDg4R3njvIT zf65Y{>3?cMpFZF(-XG=Udo6`L7T9C@EzbOtDqR?HSCN_2s*t0*Em8QjwP;3MfjFu~ z%W3j&TN?QF6ms(YB5r#h$nMe@Rx;%*?pt>WFS4J6-L0x|!S#duDzgkUW-p$K=~X%_dt>T*)q9ffCG>Y?R@HgrHgpPTug85exYi*Lv> zL}QJ0INcqyIP-EqbJ;vDI`*Y_xFm)CHNw>HKtC$lF2|Xt=)kO}XX%%nar}qgFTDSF zN6;$%0q<(Q(4e5fcwcZLo38PfK8~GAe7!C3&7n1Xeav@0v_b(J!5&OkghTKHD@YjC zLG>$@+57oYtfjpP)T6F)?5r?YUbLlc%MnhD~|S2!0= z0sUL>UHs^40_vX0L4~;p*OS3`Rg*UPIj#l=PgI5ZFL?Y%L60tU^hfWOd*G38Z&SH9 z_o=2!2v^ZtA?~_)g-`pol|~&8f@_m1Y3jsP?7VFWoAOqU`+DUkUeeaiOWimMeTP-%zDO2)%>6z13< zW5G_@&SHw|>_~i2E3VZyBpGc5B;~&ig0J3y*av;Uohedm%WfMs$+w8fEXiV;mZ#aO zgdt@4WO+8dYCjIp97Do-{qbn~LR2=y11(5bf;BaJLAK%(JuI*!Uu0?vT#8Z{Yt#V$ zxUo#x%9~uZ@DzPo^__@XY;n$?(}Z`}L)MKN4eK?Jv6_}aO!eJy{P@^@LVI#ZAuq7g z^&YY8=MVA2v{q&lD94=64!}chJe0l}i^HNa!E{Rt*3xi?U1h=CUq(HFnN#2vm8GhN1rb2phj9K~Ykoa|-1o@KHa@D^Zsi zyq_h(a<5RR-zR9ZF%Wz)tMCte0ykf(gQ^g9NF1sNo4r-Z2v0xqF(wSwc@8IcHEpr# zERK~gE@GZK74UTF6?T}Ql zqLTsaNOLEt@K9$%E&kvoCw%GEyGOzQK?C+H7n}ob-*NN%ay)&-648its-lr?Pl@Gk z4d%P7ijJNchTZp9Q&sB@_g*dK=vuP`Dfj#S~NGQ;`O6$&Jx%b7i^wr0j&C0N(ofLLZ< z!ATvX$dtTuIDau=FWS{0dQKW%KHCnTh}g+)nE*0);tf_~?$a}W?bvYJel*JX8n_8r zfQf0ABr-#h%-C!NO{#`0OVSB5mMXI;MQJEeOAowXX`%1gnYc}uaZqlThIymnnab2r z%+mG+(^8ISQ-^r636%+a_8JxF+v1LI5AQ(}3bUN94Sa)3q6bL6+>BU-zJXEhXV7f@ zdvJEveONHhP%`1y5=lix1Dl(aM=nP%X7|cRV#_PmI9=f$7yji1_I|gDllL43X*p@= zhFmptY95Ed(}gv>Y&%JNF^;zsYHo3HWtgy|bduy2jyx*|TY!K?=y^D0KAvtEp@e3} z@4;PPa`CW)BW&y&WBhpXG_te*FvG_C zEj*2%JBIL^ElZ;=c$ zm?>Gh$(^m6p-Hxdj=+W6LfNr)V^%QzDI4P#MxOpjAd8jN$ccJa=66RHq;7qO(%Ru{ zbKoylIAxCH+37!mgh7iftIh%4-fSc>`UE5Qxx?;ECAz`$2p_2xk3v0i@K=vYa`8tK zSRQ!d)Ny_f4q9(dbo{Kqx?BaG$xg**-p!F@oC}n+rW;72C){BFy~>4h8AY=BaUeVC zF#xwGjb_K|uVBM3DY)jj8PRl|fD<#T={eUF=ri6!gfb5)dx1!ysuWqU?KkpJv|uYg zn3C9}FYvXef-TL^#XeWh(eJ0jvGaC4xD|MS+V@?=9a-+|qReSj?X(v6PUl(kJ}eop z(ax+=7ejq@9;>{D=Y zi38KzmB^0Y3}Ba*spIt1|I4okb)IfHcFg_^^BLU_IXa8Tka97n;rEA=n!cY~xxt+O z8Q6lAt^UE=;e$wgyboJyU5$%W4q)+oMRxLXDX98&TS`z7q9yga#`R)NFZoBsY#*NFx7w zOF}YtNJ6%!3j5se3^@zVNv}+trnwE?Dw$EgqJPw1_`BYWu7DV+7i7)*ShBqP5(&B# z&F8OK1OAywxNvzKt;~AHmN?&rE#Wutm-Oen-F;1XJGYn$kTE!tIS#A~zTvA^HWLe@ ziRjUJHA(L%XG#8sBDS(j=+n(|BTcGl`1P0hOus*oy-lx%{pal1uFXNL;72AqI$y;4 zo^i-3Q}C%|7LwPV-Xw3wTylHFB%XaNDvCu?D6 zSy%jeP7bYe3CBZ>{BXr5Ylt1H&c^-xhT}_P@u##oWVop*$eh_HHe38se7pPuep+~& zpYk9ZH9LF4$&35oZ_Y8~5>`m6@*j|+|0NJDtpq6gvX0Iw(1X)e`^5hi&f!&lv^pu= zu;g;1qp@^tEUx)D4%bWw#i>n{tDL@q?8ss`ub={bfAkT5N#e1*?gOkeeh@S4vu5LV zZ(&NAt1d1Hl%mp2a6BFifIS&%#>!NHtQ}@mOBS`(le3A zE;Sl9^Du5T{s|wBIp8}^Z?J~%Td`7(8`TWaK%0Jb;O5#=QkbxZM0tASsG&~yd9f)8 z6y^ihhVR6Ab^2uO>?C%1<`10V(|{-GmJ|DJhsl47I*6vf-9W(G!#-+O0b8I3`nKRv((ed zcxBi$x=f{4?C`OQGg~;B{;nNQ-_BEk`n5gis91ug-`+&lhY5^~;_vYKZwOl(KaVMw z8L-`NTkwZ1Gszw2aNJoM$1bn%XS06>v8z`3L^Db7%8vhpf9W45N*VfOn6Vp)M(sFt z)@P*n`ZroMW+h*Fd?(UultLzVSL2GTM%-6igd?#9(se4^C=vrmL#v`HD!hRcygQ=j^DdTLtI~@O6Qe0WYqxQ@-h=XbtPlN zm!J7#f$h}wU@WOV^c*Lh7Cf*9DmXMuhg?|JPb;slqvn(K(fnH_xa{mom~5bm&2?vj zn#BZAyL}yfKV{1|$9~11^8E4ki`{hh{BbZucM0qjy`{yjAMmO9i^v#9EA+|H9&1F^ zZw4yrEWA@VBm;_x)`@Sri*+2a;H-;myI05H{%_TpVRhz z+R*7Z1?zksihBQ>j&GJdqAzPNQ|4+8_9GQ>Y^f=AZ!JfY6T7%MzH!K^Cy`zi`@w_o z06yWMBF@oYj^i%*(b?Zt^EZSuj^&Pv(DB3^b~bs!uSyRH*xAW%UU!r}-6F$$O@ctS zZ8Mx%DD=lZ{Q$|bE$qfEUp!$`8{O62>-aQg9TNR165qPF7t0Gw_Ui%*Kt1_B+3Xp_ z?Or_%ay9Saq!9!3vriaoD6s&}Y?0t525xM13OyV<0?RmdK0X@{K_MUemo@DK@Ujj44j!0jNLDN^) z;uMv6Bz|8tC}=0+LvsY4E?P+f%XVSOS1(u+_7+G*8#b6x$%XsYAChI@ue@5k*r^})>E`V$YHBK5V46ZK?6uH%$K&M9j#y%&I8F3OfXUBUw^H=Ca6d|2 zl!eEu^>G6iK9ORXC&cjMM)Yc`6I}RDk(Owk#O6nD;O+-g$o^xh#B`AoQOqmEYN7JX zcApL0nD7(--Kj`k4NIX%+#YauC(GbLQ+n|7=7Y5N(Np?H*9zZU&`h3gkQ0p`GE#KF zE?G4HK^;+X&Ji7~NfJG-i4|Q)pT+)WPb2va(}`boglO%RJt8O998tE3h{;6`K!A}9 z(Ge5!8m5z&KtB?-tX8yEYQLz^b(?7a#gpQnrt?U4k3G4V5KJ{cd<2iYxu8wIBIV8H zc*wgh`~%Izry6{4MEoDzC$QZDEGu#B^TFi!qep0>#SePr(L4UF^#RK1-9$w>^924y z6ZJT?jFe0^U|uuTSVo;8dt7KJ`ZVfSjnC3>+;u8}M($qDDn0I!KVysV;)KuO@?!(Z zpDhwi?C8YJKvc`-i^4V^6S=##Au%UMc4_=UqP^!h z-!Fg3vN&ZjvoQ=eKaV3Te$Eh~aCd0!Ddo%Lro#H$Z$Lu@$l|%eGjVz#7mht^6eRtF=by2N0T2j|sfh!eK1#_y+YCqZpO z4P<>WYWp%Chd-%DkyXy32K z;_vfaiG;pIt-q|{hO0Xc?b%664m6XrBWh%A^CVo5dz9>{8Y^0w-$2eCO~<(3QA97< zh`M5V?!ng@96OMJR}A})SDiT=Z`!{BIUh7+3)Vjeix~>+mFaL%`RN~IufrX3C*>zu za#L9pm;H?_xbVbLv(N*K_Kn2L_BK#wnU%0}kKpO-9z}M){YhTk+b+sz(qSc-CKX-ec4@DDsL~YkN4+x zP5nf3-x=UOwQSyYR}N)*S#WhxAE(;I_G`b(>Qm_^J2cC~k(!1Qb8y^ieXj6T4Ih#kP^5oX9{JZmX@&fYP9w)4a^W|Z*&{GN5O5e~Q6=Ut3mlR1p71DBBn z9AZ!WIDE7&%Q!ra zEr>{C{t$pp%E!{R=g-hBv$R-UfhKCMSU;X!~lrWQM3e2N-d)9#DP-wO`Ip$t8I|`_1JxGl{$U{u=~ZEk zM~*Px`spn0YBWnLyUlW+u3|Q>u4r*q~u_xJ; zg6C{VjWwJj?|E z{$weWdo`OGN1kA2_4^@hqzUsl`3q@XxJ=C=9QgRYRCI3KSV`|*b4h=`gG6z&F{^zN z!<$ZWK?)l!S#z)gTPot2mi;FD{QFaUHF5xto9@E&Z+OD^m};`A>IA48rPJ#zPecQeW6l3C>QL|c(3NYC8sw zzvySDo)}7=Uz;O|IXHq{2~%UQ%pTy>efjw8E-{IDM2HSQlNIk8#6Grqv-Jmc**YCD z`|w%FxgVPffwCH4xqkz`jAx<##uA!rZww>vrLb3l``D=&+nMyNRODTK8mVpzL*c{Cx7Opuhd2mi5xwNfag3jz_1Dktj zq{?^t*s7e&iu{k%u1prKFzBUIj1fy-zK$v8Jx3F)4?y;&a8?+;g{9W4!XG51%xiod zNw}Rw7M;(+Y28|onh=EibBiF)rH$9w_8;|~evZjWor8kCTkt@FH@p{^sS9-7S-AFL zCaTTEr&o61>B(#Hni)gT+^-@!5e;%Uv1}clnL8Px{O;l70*i03kiT}H<4VliN5IgV zQ^EN7E-;((9n}vglIN3)@%J}zc#J{;`*T~0xw1a^TOZ7>KNlR`x76r2-%>uNRSEQC zMo7vy8_D>O+a&*)Br+T0N~YaYz=ARcvC|6;;d|#<^iCs(br?No(g%((ap*ya*Uh3? zyHg>|a4LvhW2x-hMjCn0nwsuah2M6`Q1@;o*jyvf{n!=`_xZrv2qe~yUPZ&R{PoQ+Y#+Dv-H5iWaLmp~4sG^xGC!{9%j|BR@N+e5D&)H4Fxq^iwGPogGzfEXIK? zAy9UD7;Buw(Xhv@V1A*P_BytKcT+a=3NmKX428P4!V~<jMGvu5-htj$) zYp7(KA|_tZ;NR3;ZRt}2&wQ%D@@S$k_x6)jxjKuFt*S?Q$|`hnRUGo&e57WN!zs)j z&0-L$$9BZb;zH-{A*(H}k}1>8MM^)ECAW8}O6>4BNyp$eB71S9NILiwnS>|9zP*7Y zCc%enQR<;{dq#rsGifliLQ|I9e z2R)&?<2v99N&}iXx@5ynySN=q) z*Vl8a-HfqOiw2ge&g0HSy{Arr2{3W549pmzPqzP=3(wYO33J|QB&>5dkxxHKE3bOv zKX(U_6aBIbjt(PV$_)98iI&*qXNf_nu&unhT$*L~)&-0Fwu;C3ytZRJ@iwKvD9<*Ox_*p(xrjb(TjZ-|Fp|4JYI zh=dzMvdEvERpg(3JW#uOY>!;AbkbfN=zJDi`kex~y7Pdht{`u3hmw7I)9_eNZBA9< z!)3j(M7N?H$y>Gu+y2R8)mgwuA0&4ob7;V$q zP1(G0xXL&lPu#3aM)YaoUlmv3>fui?X|6wuy-@_Wj(vs6d0)7jH`4H{Z7Xnf_dKpU zDxZ_F7vOhy-tgOkoMFtadDQ86HkEC7h`R+|*^H25)N#pkzUQ_#e*gCg%r}U`h3+YI z-@0h@cB40b>oI`5TKD1$^A*_q)7f-+Xdrsg9wyvJGgcCG(&=#B7<|qxi}raH(F3=) z!X*P8h&exzv!Oy>>wOkZ*I7W;+;_vG2P0wmt(DY_`G85T&&Wgx7nh(i|EVb4bQ=V%b7ZyJ!&%$C3w%#Vp0Jn8qMq}#$z9bObW5@+ zKdECs^iH9AL$&7TWe*9d8~Z$1eSnhucx=xXRWV z{7sVh$G*|{MCCg;HgY%%w8})OO*gTh$4TzAc9!tJpN;pHXX7Q4j^TOF7L)rs50IEU zHINk$#_C7O;+O~WheU7q_WmFJ?_v^fo?*v|Lz3A2b=TPkg8&#?XU!MS z$fFanW!WM$FJ-`Q|pB1lEz#-?@~LpalenB+STDRx0ZnpcOT152qR&? zDb8S@@Qn={Vd;f6$Yg}TQyO#_=N=8fZ;B4%^!dBVko#8Hy7UjM7|6tWX&d2Jo8XSl zPv(1L1U~uc(eUf;U9>Uu5bxz}P4@Wg<4c+-y*cM7Q5cy>4X4bYe+&+jEwg8%Q~6K1 zxvpBcAx25Oaea4n^^`cSVqEHk4;OZ^=bLL;$@X;Q@A8^UP?-kxF`o3{yC3vPk>GPi zAJC7(z?mgno@SJ@)+9r;_&;b*GOqje+C;q%48bm&EW zaMfGNo5>oZQCH32o=dOTV3$AK3$$j7Ck?^sk}h;Bqn*y0n*kN+OJLqBSJV@sKBE}TQ<30&*< z4POylGezcKkkuds-1{n(@BR4^jgHKv5toed*zFIXsA3S?Hnt3D_>|ZD%^C{fD46ct zA&puzwh$?-Pn|NKaPytbp)sX_4;>zd?*#;t{DD#UvYsZ5da{PjoN!DWlIqTDq9@2n zE?;cAJ%zq=I*LENJ8_ij{3KWrI$M=*Ik! zPT%gY;diE2Bg>*DE_mo2@u$sh{Gf$VymZe9bUphFzx;U_`e^YEJHIL9OMhNN$%a|P z@y`VDe~GF5S%Y?RXOju+xAUNBs^_T65+^jOdkTFqV8#@;%JDVUZ)v$vKQVc7fR4Ul zfufx^lDdLFV(pcga8lY8o9;BD2`hw|kRdIsr~NY8_H#6TrrQYHX$ALA`vH}^_#Wn( z44~l1Se!n4vA~Dz;5SB(foHiPV3TRUCpE{RQ8i2H+vX*-d$9<N=s_W%N=439@Y%papMDI=86A1BZs$+Fs3>?>m7qilc|4rE(_V^A&!(@g#)%k zlB!$}9MkTLFCRKZe$~eCicx2fv~n9itkeL7c6L)cM;+?6jycIzXyLmKsZ7;WpPxFE zr!A2uVS_;@w_{5d_hjh<{A#{F_K3HD(SLSf-{vMR>_r?6F)XApJvn@)OA?GxUnQ^s z2P=G=_j0I|k+jtR-m|@A<)ow(6r}!dFKwxbQb8MoSFPT<+&Uy^bC9*8qph8dt*zZu zHzei%|M&HOebE1|u~sgxDGpxo9br!|NT02p1h(7C5c>`|9y07 z{!C`OL5W4b-UHS{f~fY8eqMI$L1h0#p1!_ai383xL#wPb3mTjVLBc%w-@C6J%QXtP zoz?|tYhx}v>)#G5r;f%tn^s{d%OmvitBpLjWgfe^>H^y#dyV-84W_$)>$AIZm^pl3 z19pXhuu*2Wcuf3PbRbO*J7k?lyFWRR?3aZ!&i_51>8j1t))Ye<4kfeBHG%!&QoO6} zD;_T}=aSmz!R&@(^yKX8sI$xxi-)&BPF@$ickL>-`~E&U{f@B5|MG-}UA%!#T{H&y zBvm@$m@V2dEf_DjItA$si6Gx*KEOd!b%|c%9z1jI0deWLCRlUD8eIz9%&das7>oae zos}ru{?j9F`CTgZ**BkL99Tmlwxr{4!hF^2$p>NQpowI>)Xk2j!lv34wRLc%wwq(!4rQk4BgtwK{;2%wm_~~9N%moJm z>e6R9pF5qVp~+5MImf2_ktHtvw{fH5IJR`Kuou{^ie1C3ftNmjYPMP6q%YcBa7swc z$+5sKj_k&5rhibQ*I+VMD<21+A4|$A^svokDcb(khepL%6PP>=n)k=}FpDDH<{Es~HPvZ#U@EeGdjivd<7nh*%)jo_hLqNRe14f4 zJ}F7Vp)b^M#S$63-wN>Ijf7-IU8FC@PeT?8nW!`*mx^E7;ZK1x@w;={WSmwcel%t{ zW*zsS^TG=!{SGa%-*Y+62(-bzauuA-%1Hk5fE7w|8jnY$Tt<}^6KLk95ESX=Ch2f- zk<8A2$aEUD@%%^&T6OvuKKRBL+#URhd~*iSC=FUa3&6*3XS$%_~PHK z_-RTjAAKVdZ{8>{6Q+BzU&6U;ghde4Zb+bJSqXe?{Q#0HtHIN>dvHu%H(XY{1DnrY zfPi=hwmy6{%bs+WMSAdb$?Kc^fWW)gHMJ7UwV!|xC03LCZw%_ZVuyQd=p;jSf|=G@q9$aEP@KFd!dQNH!0&8w7uJ#{~~aM%&9Zp;~6kg^OnEjWSY zryLY|=3<)fSA-THQ9z#+{-cj>Jw#>qTWOi%e0=AFz{>bI9;yl^)6b)?@Y2$8bfJbT z{1|$HTirCk56FH*V;9|`4MJ^k+Z1Q^+93rRq)ws}&x-J~pd9R8dJB687ZCqB2T1p^ z!$_pP8`#7JcKsmm zte(z^7V|~1t#Ivv7hL!?1v_}$!J9Gxo6I;(dY-1ko(U>oVQ>RS39QF&TPNd-v!s~D zmzgY5k_5B!9iXgD8>-j;MX%buU~2JkiiP^ZhYu(CSu1IEPv=Ry=gk9XYj^_<`(2Ja z#xUAhtj#iZCot6(#^jFOqKV#Zbn;wBZsM~{t}S?|eV^1yM8f~8UFyZu0Cfk!YWAIj(w!fRf0aPA4UD1+u+?02_JT8^KI=P;GrTw zk75p9zF&n3`da?}*7E>@7w%!vIh3h$fJE1EB;NiuQBHS-jk3B}upz<3*ITKI%LlYT zfFAj8OGjbCS_-gghAHLG#Z6HmPVFJegztGM?3LxPinR``d>sv+^3K4$<;$4o&^Da* z_pZ=0v4Vxt!rm|}k*z40!Cux3fzF3(@TVm!*@6qbpy_fFR=SXL^c(QLSM zAgn!kou*2H!1}`vw6FO9xxpjQoxGMdks5HcO@V#6hERBJ2Y3cQ2l=16Ogu@8z4Rzx z&GrT3sy5E2d0vxn>jJ@vy4iuaNp8_0%C7u_+(@c)<^mc&Sd(f*=2PFPd!T&bNMd`$ z0v@?UaW=b+$XZ1q!y9c%9K5SYSyeIF8g>eM?c9M+uP;ZxHcFFC=pZwgoX<>N^|KQ< zouDOX2>Yt#%t~{O*h_(_Hn7&7ZLQSf5|j3E|DvUExbrfmbzYUfpMM_MooK4QBAx$~ zKLH;du1^js#}Y1L8+P9Jn|nPm4t|_gXMdyAC14H$mz9cYT_vu5y-vCrNTM2Eg`w7dE*6=+%Mn|r_bUap7K&~Vs zwle<#ybAY4M&qZ552Wz4Zi0yVJoDr)Px{E83AhO^32h*+W6$)5SwhQ1!uCncXJZ?0 z;@|;Bma}S*Iq;YjWY1#}MzSpIXfRr~CCdG54x6ALbAjcI@;{O!{Pd#wq(D9kvqhUXnxnh1|%^`O}DPw*s^54`fSK&#+riOizwAVgo6&P<5LlbxanV z^Kvo#Yu8FxYJM7&=FNnRlm=MxQOHb+r!dLz2ax{YH~u!w7^Fr*>-gYHuSp`8SOmc$R=`qHclps_i7a z{SWELkr8#dXW_y)74pn_5s}f|DVeLhRAT$iO(LzMM>dqY<9$1>;pnsD*{s6<;L_IN zEG^BCJ)b3EksGBX@oK-s$q`rR^n03Y(er3(GJ7=M)}2gbUme9#I@S2!7r|q;d_2)G zQ72QyV+EDbXZCcqvZVaHh9vHbA=`4LiD`F^LMQf)X4EzTf@a4u>uGH)!tM<78*Rn5 zg$yC#zkXrYQU|ivd>MG3%%MNeo0|%oJZWv?HS@)y{&xbbSdtncQH5IcAvoScO+-Mlvw|oMqE)- zN2k{w;z~Ooiz{o&YgE0LQmfgM>A$^aab=4(zBjQP-xneLL~0@ zz`mco318Moe%qIj=n-zB_sg_Iw!y2w5a(lM*G(jl#^Tw=v>RiaN^yDcP|~AH zST^bqO&oL{8`jv61o>Y${;4}Pi=G2h0tI2-t^{&0A)M$aW@3XWGq|2)eR6URCQpn$ zIhoyA4P%VQ^jScj zZ(G7k=NS?AoQ34+w{YS;u@L`gxJgHw*+WnCHO_jIv1s2s6H&8+s_3M`8%N&#Gymeb zv>4+g{6nBAs&9JD{pye5aX8}snjGd+=W4>Sjal?#t-w%E=|V=@45fwE;YQ;_^vE0& za_HeT^gHkn@-{n!TmQR*e!5R3J;C7wZRiu~=6C4(Fh$1QKL_(7Fjxq)q}r}~$$D?0 zZ+`U>jLcg?w&ux0@h?l7AUxlyd~IrsmN@-F?nL)LwpQo zLijg=MKh-3EZ;?B`&MD@N9cj|R>`q$6igOOno3}{nvikHB#z_a$hrwT$l4mg&3R}N zi!V=y5B^(;B>Fb7RT2@U6AJ)+G$B^W<@or&KiFA*F1fJiI)1#iuIlOM<<)WdbX;i&mwJFy*yNRGjT?Y3-- z%@3UXw2Vf^SvtKrUX0RSw4*@<(O7Ly7q;n$r4}l$Q2Vr3NWr0&ZE#y4ak82zd6Bb+ z44VE1%dIV>*7wGV%SKk=F30l@3--X?L~Bi@qMJdIfYvh^?2< zO8n%WkRbPDygF+#`!9GZHP!6H9}dPaL)|pC=Fw5uv6T}gpPf#Ik5phwGQC7^KiZ3k zk1IK+dI!gNs=*;`5$h72rfz$R$Y%vFVi0ixT$my)yd6u9l@>un?jh>-EP(SU7O~?> zgPEFkIX<SHN&IP!~rJW+%6Gjc>z;xa|WJ3f>0 zw;u4;e>W|)x(B`nQ*rdG0$gtCTJ?|s|_Rm=|zd5w~HYeA!?nUns<_oyaq6ueAU$Es(oL?_St5|=f%n4J1I z+;?*?l*RtxTYK$cysIX$_$^C5#=Eew`60~HaT)twmCp2?vRO}I0n2_ZptC~evd^L+ z?CqmN$ZTXUnW+>==Gydnyy|c=nn((bqQp*Z#oW8C`yqFiEq6FehP{Xu z?nAx`Gj)rCu~Um!T_OQXat(VvJBi;X4#zQF6gH+tQ>!6{Y~!G(=y>0H)PGWrm1Nps zgX9muCTkN#>GNdTBMLX4on{kPCBXH?@of5K;diMGXU}4NUf@Jj8yxFS4PFBUo%gHQf4l%|ZFiNjlZNpnBlC zO!dY2-r__bDe~@4FSdDn8s9$U2y@dM0oR*C!>*x{DN{#E*3WjA`1vnomxq|aW0K3R z=U2e*23^?I8j8c!l-P5nPhH!%qmHt& zKMCxJ27*j|E!cD+9FH6N3qhlhJui7gd*g0#XA6V*4RgAxZ-3B+z#~tnQwxtx!msh> ziWU5y^X0TQI~m@gHH@|7g6yO$Zqw%sNSm$7wvHLhCgvxA+4X3=nxwJm`bli7Um^%l zci!|RIFT8lc*KnbY}$z1EcjRsi@YmhZw^XOEV?3O!nUC{*SB!w>Ki^d*oegI?ZcN_ ztDr0IBt9nfi<^UM;OEYt+@;kE>B9gGsLGw=5Z!gXCU?IYdMET!7L^#{oX`f4m0W<- zO}!3kh1Vg+Uda8OnMe)Hq;PujZ1PgM9LH*&rEeyhumWW>rtjt7jL2vpN-7j(-t!y{r82VGQJNYWDy1}p5HcD@TS|kp<$KR_j0lw?np#SVel*e4 z_}%}&>-+V+_nhbXeBSSlsXR;dA4?svgSomerxw(t8RIACkmt6lJ{I z{_A}Fk5;~HcEQD-~O8x{#JXQttA#R2I3 zWf6&cw}B09-^@MbOSqBa9zHRBGq)aF%AIU|`QIC3VeqWOH2uRM{x7+dj~wxWQ~AT( z?ol$&G1d+R#F zeonV0QFbH;w$7eRzgwI0W$PX~L<~tFi%m6osz*MxxpJE(O8oTgr~ZPkMzmniVM_DA zy`=w08C~|)4u|#ircE{@Xh3ERG}KExYp(#Ir>BPKGpr6mV@trNs}e37_F@*DS#(EX zFZz64cmDifC%bz-2UhIPb?Cb`6~fYI(tlsGiKn+Ry=Og~Ox9$uea1*yEWK-bnlFcR zw=%F@Cyy(Cw4qMiQ`lK(Ms%*GF}K29B)`TR%Dg+JoYF`-IozD~U$PV;CXS&W`aJ-J z(c>V*NfG2;#K_hbAE3Q1MPs023x2pS@n0RUupZV*^mb)4OEPo-{~~YU`T}KIQl(1= z-bvw+6a1j^ni}N%>cC;^zR8dmL;atrQx}{c~x0`IhW8KssEG8S?RA~wu zj|_5HJoPrtUeg~_9iPUw@ zVe0GE!|Bz69NdAQ*;d~Ptieg1I>t1ibQ$H9(azK}kK$IxWSFGsq1$E~FO z>lSW=OTKNDJ&*iLHs|D$F&}iHzx+wm>DxyvZM{jIai04g`54-6= z4et4q!AeIED)&N}woE=mYjR%D;TI2312sdww7i;4dODA<_AdcFOEcQDZx4F7cMn)= zL#Qxg2hE&mj{b=&NV3-nv{>6pJ}$6^2>Eb!YWH@!`pa7SxxJhoI=BZOx%9$&zx*-l zzP6Mxh{cam7M{uXbNah@pp*X~9jBrPzPw}NWgJ>;#RIcA+qXxF&$)jW-v5??+~~tN zD%P1#K59#9hs+mS4(6fxK0SCCH%wS*>5P4h)?lCeKUwAYqcG#c2e3Yxf#2>j?x%Pj z?wopt+HEJ9`}=WZh(VTYxXCh@-@^$uH8?L3-~&f+D9pK$+sd*H6VKM#C0 zl-t$)CzcP5Vr_y87S?&Q(euh#=)eN;W6#^z+S7%9bamu^ywYfPA6Apt+j|>`3{}>Qp5kZpL*w z46B9NWHES$=hH>yi8u{2gypO4&^2i`4LepwlnzE?wZcD$>%WhL=wC$^*3Kqu^T$5V z#-pWKD!h06!sPACAZvmOJ+eKSY%9}&D=)3-0e^S${bng9Xe1Gj!`hM}@-Y?^KZo<@ z?~$lm0d(x8Ba%;KBtO*Bk01Z%%@!+L@{Lo@@%=G!R88HLw)a0oPDj2He?3g0ehVt8 z*07y)@TeHN(ZGeCFI~W94%$l67HHD?tTar|HiCH{4}#fUEpRVLCL8a`NYbHxY3(V53Q}x4Z&3XLRA{lidnWu50JMnGufVBU8n|+Mbvn6An%zlPO73bn4VS zB7L7KY=3+KdQ3V9&F+1*}Zo_s)W9t;u;Jx>d|WF{S0?uWxK8j@r4 z?vl5OQEXT1QC7Y)45wXqBfFFzSs-)lAuEvlG#|VJs8K^G+dDpvHvhDs9~XyEr<2w+ zX`?pvGb<zVM(aJ=?bwCvUZb z$Q?oAnd=k8#@1z!=1>jpjyECoQ3d3V{fR1`Q!yp-1F(3B8{Y4bSUvO(X-@kKh3Bup z$}cnMDjhpGQliNZZLQ-=Oh!SV#67G}xr0H=?6Ftjd6L!V0er9C4>zP9*0i!yfRi-X z?(K*0LBUzPy>SC{1!U3lBjP}eQ|5=QKN0H^FWi2!l>{3$Fzb|Toc-T(cFSiGl%!6> ziH-*xQwzJ%F`Cn`Sm`pnI57{?Zba}SYsNt8?D5!3`zp?jG319IkKnGO?y_6EbuczK zksmye&i8N85U)S-C#$E&vW~QyBs@vq@xhd8ygl_Xo|Q6qb6>Xb*LHfeU)Uq~=lx*dGOT1{&4u0o`A zcH1pMILW`-@-axrrBhreWTi2eMsu<#b>}2wgEN6y6=nU}J)(p~|l& zcBM5Q){HUXPu#WnuZ{Bj;JGTyf0M!0?-f1JK{eq@Kqw0 zyzQmDUlH%t(@tVxJMw|qnY?my6bw)Ag&-TkS2=3Kp&r#JA8?kh41K^dq-V+p16L_4 zp2TacBs>WVa(wot3D0>SN0lB1)a&vSaJ?{s+!2lNP)Y;p z)Xn2gRu%YanHi93b2>cKjfP+^F|lqNlnnoXw~Xv?jpcb(b0r?kFZG~S!~JQR4bUA) zmcqhpZ+6;36?3K~3t=ll$+omRLZHfiIA8f3+)o{3Thz6&o8oO!XEBx3T1};^AFrc2 z#s8pc*HQYUaS}c0r%6jqrTg|D1v>NE1e(znOP7uep)uBubg!ig6!z1W7>{FFgLwxe z2aKnl9d}?~aHd#bnJ?bGK9ZapUr$bFrNivWRb;Djp%|5C#C%QCaBGDH9iqR2Htwkc z*Ar?`pMQ&tU1&**D&y(yI|r%eZBwcj63ACc=b7)8OJPZIH#ljKO#9|mz=u9JAm@V) zL^rE}CzTuz1t{F?y;>X)m_;|pU8SxME9fVkwan_+9pY0|!ECl$p}|?HGkIh>l`p*` z&X{h50e+9rFu;yJOKjV27X}7ar0G$!e!@|7jL^f5s zQ1evp{xc4qcMgE;$;N2qrA2$S zeS+d8|9A!e%c1)T-Zm$lE>Cg;?^Z>6PIV{0)IXR%9*`kSxFq##mQw!d^d4p}DNR-0 zD6sf_kI9>PW#s)UMwfPr<37K4@b~MZVQgW)!sIDxT+T%fQ)k~o^Bjq zBc=Z2XC{jXbuWlHJwn0fyZpD<$qeo&9V zwd+mAesiOF-(gYg{^s>q9BWC>bzSAI2_^h|$ziraa|PZDD3*Evzfm5O_;0T;-e-Uj zaq*HX+<0IUcQ;#27oQD=l)9^&p$G4Iy9d8I{2Y0bZ3UTL$LYIHO429B@wL6C@=Kw) z{KUUp!Ta>1ySjMCBLHkg*(#55_pmb@$& z2I%yMv1@Jl&e57+-J=fcNBv+^VFLbf7l9GP8?S*Q$}@@l%QWK|$QUau*$WYcT}2W|HJt2Y6j{2Y*~K z3uh_@aMOmj@ZWvT)1{gGW99Diy)NSlx4+l~@^fP$f0eZV z&6eS+zAjg@Dh}b)6$53XG#X{fQJke5*&#j}}5-Dr(X@=9_4}x0&RaDJ9?i zyuf<$7*f}KpIyy1z~MG~#F~WOE?aO!R=e~HTiWXu?yJ!&)O{4lca43COU^0s zL2i|Nh}g)74%EYN)t~(NYn#HTYURR24aNuF-pmVb78Bohvmhz_DEzb94;TJBB31%zab#e#uZiUg7oxJ;x z4#!Oy=VXl=SMcSht~oXs%!e!A6QQl=1|Y|4h)3r(@zFz5_>{U7qdv`GN9RWH&XgFw z`siWo`Vm%mHF$NQdfCLnkc<8J)s>9v8;#_aNv^z)OE$aSwLlW|_UG1FIfB`)E^*K=_h;rXa*!skRdHUHXXR-;ORZhVL^X&_@PR%QnZ=YZIxmzmUzt)dm za6bvRO((-8iH|;9xQ}D^xWTo%emMEKD~|XT$t#7UIC9HVOz)`x>oT_RHO?6rQa+rn zYPm{UQzOXlbv68a!(U!h@gEN`r~u74J9^>#ZPIt;q{5n0=7jCC$3vfv3?Bu{Tv8|tiUgl4uGIa> zE+?k}<+M&Unmq}#bc)`%fxdB*k>KA4g}_-^;R8ZQ<&2uibYus zLX*PY=#i+&mwfc#IKK(4QqMwv1wE&;{migm1BXi?tvEMX9db5Wp-jm_cx`Y8Z!XBh z$&n6Dwx9f*RHb>WLfklh@77-&=D}%#x!@F2avx4Y}BH2meko1C8D__H*n}x8(VF8)w5ajepLD}ht&OkC~ zT3?($__E{p@W(JwwLd+zY#XP|NIkdf70S?ynn-Es}4sXwOg{ zIp&ryNtuB1FIDV!#vRI?$6{c#4P47z3Y&|msQO0x{kTl z(MU|{$^G@N{ALJa(Iw8%|KI|u^-qA&;tXvfClYPiCh; zwIz(Eyso8v{S9!4l`Yh^j%RWY?D@iN3t{HL-n8FQHR|@g7~JwdLg=H-baIy&4ZVJq zh7V1rL)I^b-YW*M<`5GIZPCU-hF1iQ#N`-#?Udm0IvMh}1PF^X2I1hHU9c-#mugsa zbDFmCI9by+7tv`qUF{r?1zJ`Tw?XomzJ3UAox9kxKi)j~?L*k+aUGs7Jr0hYBG(FH zeEhZ+d?nY)lbsUz;c9_yZoZ4_E>FQ#uk?Arh(7S)mDIWS9EL-_b{BmgdBCxz-t_Z` z6!4O~x^IJOVQ-Wv86G+VC)K=1zsc8#*0&fjHfK2)X~Z!(yJ>JZVm7leNu-Xw_R+OY zR_v@pF1HIl!W_Si=Q-Q@(K7?gY1ttn-J@nu_e+~7o3Vl_@7={KdOzfGiCg*U0~!2< zcN=%HPT-cyl6gg!3UeItoqZnM!E#g<@})j`Ak&No%NI`YX@uH2x3^%^I;5 zBXQqI_mC8R4Pr^t!v3)4m(?Xl(i@;(Ph4EK*`nzSE1~NoJ4v zW^X-eax;q=i!Gu}uLZC>&Xuxr#gMwJs)pF znf+rDmKTZRrbHm|d`0s2PC^ZxyV#W39ctwp#Zv*&Mn2e@u6^xD4Dt?={?(GAr5pHkdlMMrtiN-t%o^&QtSMDvI zdDoj6kGX*#QwKqqi!ZFMvl9lMdWTPy_mGJ_B<@aXEKIxHogY^l$_tfzfcMfzBeh{{ z$JIdA^K>c8h6yZhLMgsDy&ti@8Gn5^D)=Yu!a}0NO_nLaK|N#Ir8oldL`?-kCSJi4%bmP+C7&CmF5S)>y1jbaLkJeEZ| z%MPQT<~dBazRTzKnTf?0o-=ay1E#--=Bl5hy~`0Le!{4RPm77?CXc-Ng?_oHU#E>P zBGTbjNgDfhDT~!NNGy<_C%KaPW&U0Ajpcdm6f|^8Nw()r)cGxtTP1&(O5a-46>gAG z&nuXjcNMBAeMaT|9el2xTjB7@Z(-r&GIDgCEjjx_lMj{p_jTGX{EN>NI;-&q;dRKa zHl1aw%%`A6*{`xb(%zx)w7fc`Wi89hacq6w`I`r%>@`OPyDPhrwcJo&*1&lgQ2nZd0TL~NUl(@3$SgC{RCq^7n z#>atng|yF>!ZwwpLe;Maq0ur4BKM2D^vg)p)!V|?d*owe?0Xh=Uvk4->0lx6JorxG z6khPT$Mtn%0RDT6OL9&4(v2@5%~*w~IK3mwk|X)8Yo=UpK^X7(ONU;56e{_2f6457 zZROQ%9^$y+dtl2snbe{E3bP{RiEPY5<{q2paMgulb>IMedTyVf6}AfpE}Mu`+WoK~ z_!-0=trUGO&K0zy-U}mR%lH+Icet@2lBE4G$BKYzA>r6fV%lv&!HAcq?bU2;9W|_m zLPumMTwM|c+M7xQ?Q~sm2}^;G52NY7GlqP;S0h*57eU_{Z(vu`Yfy9VZQ#=G=C6J- zd!x1-+FqR$Gc?-S{??N$weL_&K|`hz8pZc(Y$q#QtxC^nD$OoYd9eWz3LCs67pfxrFa~Zb`nkX2GF}vZ^(Zh*^WE6HIc)a z>iia1G4wg&@lz%yqY=jO>e(lpLw3xLuKZeBBE^hm)aeZ8z+9 z{R^IT?ga{ab~5$mKaRQzV@2{STE=iW32Jkp)*ct>kSkuKMe3x|aSz49e%0cMRlcMs zfx>|6d(cPi3Ul_a!t7e9&-}3h3X>VcPqh~`EgQ%ZQ#q&`s3y!1WR8)BE{^hF9H3+F zZ}RVTEor-c047`QW8-ljko`X}`c4ixZXSSXa+l$l>>e(4ACD*(kB7unf?j5<@ZIta znkcR2b&Ym>#-R1=dmm+T*vTGZ2OW_7!VcWwg@%;Lbj1u03-R>gw~&0)oxWXokj6O2 z(D`*2!MdiJ*dFf><8)Rbe)x@DKK`skX(hxs2fAinu#C%6mm*b}3*^!2J_ zG|@Ybw~m~F%Ri3NsFCqhyU$wON+dRI*iu$i@l!_sD6*%j@(|~J2I>c&r<0@(s@}CX zP*rskyGpHK*_IIw&ACnvb^F&7rwB#bb@?pFr^!{7r}H65Kl1<@`#UYUZgyIr#+J%njAya>w}2T3TyIi zc^p2R_W-2M98{RbG4>$ z+KoGIe}awAJJ9t-HpC|0XGJGZ%jW!E1vdtlL8TzEU5mExov{Jv{!+^CuyBZ39t-Ox z2MI|w+K%(zr<0bi1;mcsVPx-B%)5V7%6{77sUuPbbgwaAq#G&qANJ9ezrI6kZGZI2 zWUOt}4fZYRkK`z=MAxqg!q6pS>69MTg3a?+MB~jKvV3_ZG0AVjO-9A2#ops`=X>n^ zlC@%$YKlW*Oor|Le^1fws|qT9P!Svkt`WSp%ClXAMu}Uxav*=oG}&J^%u)B$1@>}8 zG7|q_@pH&aGHdlh7$i3UomVF?`_>-xP?Qe+H0uPv-MWU)+`kW<8|N~d`V}0f8A6+{ z1pT!i;odnCDK_dbZgD42re}tWZI<#cTSmZ}-v2Nn!;}t| z&XV#!{PEXq1-fYTERyT=1b*z9OpOc^m~7ihw(gKIx14nrJMNwqUHh!0-af`yxAH&s zg{H%4t2V(Wt=j&=GizSpG=Qf|8N3fZjm+Psl%x$-gb?ZdQ+Y{|o2bN*pRE!f@SY$0Kr5nlOq zL$GZ(a&-IN8~%MQ5UNgA!siZuA^nIeT=P9H#u?<}Y{lVhV)-CyJ*QUazq%UKgQOnm zvj99i(gyc^+F7vbek#6rHy@^yHIZ_;f9$2*e=_6N%d%xh>hVbTA-w#1Ha|4yBoq%F zMSr%|QRkq(YtJy_qaK)at)q>qQoOWllq%u9Ve>2Ki<&Y6X5 z)Ru#S?eiI|v2_JsrH5#EVh7&s3dginR{XKbaT25V1?P^m{g0| z^DCFoS&t@ze}+0u`8l7;OoL(c7C%^ee=PW(?k?(|y~B0{8dA-n?XYHz3jY(S0=|W( z8POobr+OYgXsm{L3)aA${mQasOPs~K?@t!ARR@BDfg`BjH>NdzdPCvHP4Fk@4Vx7y z2X80a^377_eRxxMK6#oQuaZmRLwA_)#&gT~lLB4tS>;YURt3V)NL#XPtQv_j7(|v& z8$~X84TR_%W7_^Q36x_;;r2)4$nHzhd)Mt7lYMO#?1uK^hx<6AiRChqd2=7QEEs_~ znug?`#tfn0&^NFTF9qH2hSc`C6J7UtCp2#>lUWvIVobJ(4|d#SbMnSwczrFj85z+O zWg8f;XAa*?USsJr9XOp3$s#qlv;%b%G~!G#eqRhOc8jA%1s3$sg;DtVz*{CO$iZIK zYQhlBDspy#J?xy(jb2@w!rtD03YT*A$(g3dWYJ$8DvXFhP5lYrtkF!X43ug0{8(7# zxsOgu7)>1pWRNFlC+03xA?vT+U`n4=$i`L34qYrK7Z#3VAtn>pUx$3i+Ga|l+?(jQ zolglKJ^|AlyHi(%aZ(mUo3(x1MXl~F0_!+)$z^?x-VBw4<}>4H?W`Fvrg9(tG?b^; z>KZY3?l0NU!N&$~uKWzAV&{SzS~WjHPR(m~rRTd7~% zD~UPQLe`s%!Bg8-Qu$sd>CU1(C?;p3kZB6fVmrx|vkfr1+#R3yt|C(n0;%@z8gjh( zHnxp)L?LwxRu$G4Jo=|Y^`mb>yp1EwQ#%Q7B_{j^)e`#8Vi$=zScW>&COC9=jbaO= zyjhD$A#Tu{OICPZ6zgi@gbyj(>8w9pwAihhZo8}t-II;s`MejPEST|S2VXOFBPD)O za`zm&p9P1N~P9~tqh7fEBDKdG-=j3rm>VZM$%I6kaEx4TmXyW;;)WwR%K z*ph?oXLis(oG|x;m*{-C^)%8&hb~migKLvM!#?RVQNFm94!-0=$C*jABm=qo7mGrEtuPp2_c4Q@ICJ^EZVM*3ijDj4rd&WFPaCsuIGt; ztr4iX&Y@(&5&C}17HnuUp|@Kvz>%4QY3UTnN%+Kv8f6^^={A^p#x@N7Xv8IFQjOypGX5|9hr)|)(Ttp zMp4VL-|+KRCEVe92irf~WgWYd*c%MR!Dj!7D&I%R*52RfsCr{RG}x=sq1&~ou~sI_ z?YjdyK1lpv<5B#$Mk2p!lfVmho*PZQ5m!7_7PMYmfYxpc;AUSXy5H6s z#$4_s?=Q}Vsjc4Z8L5VSvLg^R%La^INiH(abo%GzO8Q0Iz>OYx@tifuxa!LWdef)} z*j=wDgPS){P5*MbwB|W=TJe@d_&KBZe+*m0j77XQl$9C(5VfCL;*g7a#Imao-ROCk zRBzBGzba1(+;?q~mr-@N^uxRZQ@!pl)IIJl{ zxMJ87_8H8e@zZkQ&paz0-lBzX-%Z4`d(?R4vm@Z)V@uO+&4o;j8|c6NH69g2{C&oj zj-59X9*0SM!ytF+b7wL2AL&SYVKnCB3edm39xjBxh4R(&(X03@j*^(EWBZS%9exUQ z$A|(rur5=~{&)t@w9lpT+Xm2^zhfyIx-h@lpt3+5n?$b4_2J0_E}?F2EY2}LD{L8Y z-m#8Y2xV=`g7^FlLcZYvGDkNCSA?Wv_|R&ukS9G~r#p&E-zs5k|4xU#*1@9Nf2L$u zX#-SbT!SBm51{qiQP7_jCi^!m8=l|4Li;KHmc^|wrFrlc(*Ay9U2fWZRBSgqSQ~?F ziw5G>n|)BvM`B=4xQmucK9DP>6L7rbG`!ey1hQVMuus1Y*r3^VbXQRXZ5VQk(6v#x zte-jh6py0cUWc>VZ_0v8?FZ4fwMxvU;f}{gs0!6LZ{o9k(HNh+m51DYS>O_QhIP5` zCVgB?xWVtk{J(cm(hG7d?KnA;s`A^QbbkVbhv$>idhbZ&JQ1$ynL*d$3MiXVtrcf zu|uC8;eeBy>Gy@3pl_rOScce&b6mFP&QqD?w^-p@ zX*$%5xD2K{tl6QPiJ(*`N8^vhqlxMYRF?LLo1Bzc-lCUyGRU6GrR~F#ReSgX`Q?1& z-y9x7>xF@u?T#KF_Y=R_yHWl|46%{iMDtI*0#O)71^HmgrfcDh^zza?HMEPO?S>aALNaEZDd5oibGllb+U z3vkiKp7>8^G+EW<2I)OM3A2{fi(%fn?? zqOyQ*kXUb(&E(&N8KT|S4&0`r!B4N8g=gRh|Mgjyt2J%sZ?HS59$ZF@+k<6wOS;36 z3E%L|WJ|pB^Ezoxyaa=mw38pX&Fsde4{U~|K76j$=XOsWc+ewxoE!WX>n8Q#jYb*l zwDA!BUy83ZrzpUN8!mjd#2dS5{T{D2Idi+a{XlWcBY3|l8;W*rCbnzyVgI&vkl(U| z)+%VwB;gF+2{y-zcQ4?w=x?H__apK_sxB-aBSNHmG?azBfp<5hOvUvJe9x&TSf4P9 zM~oeVt)c6gK~WA{eg2q3ku*M-m@`cO=Gv(dJ3y4eZ-g zOkBt7^P(<+8)XFW=Q%dGF!cp4S)=H9^hGL!rpoih9}bZonLFvz^fs7X;6f;L$)Ps;AQCr#0yM$AG&Zi-($B zC)fxtIUew3JojwgNq;(zqQyn^G$+cCt9V56ga@OzYjZNI9`_rH>k?^Hh<0Iu&k1hp zF_F*I8p6|}U&8OnPf4C>3VWn{nM^tv&9odfg+l|X1nmdCg&)^US$X{cR&(DM))+KH zSzZ_T>h6GnKHXW~lQtHu_7c1Gx&g2|7QAnNhuV-hVco0sq~%UGC~?w&o2^eE{AYLK zmnlc}`rIbZ51c@8{5=#dO~CGnnc`BDad>Hl4qr1T0Sn{rWB6zjoDx*Q#$OxA&-CmP z$L`n1WNj-j_4p4mm3M$>?@#LsPtuALG4x|zA#S{{13k40#2nD)dkgNcjoGh=zOFJD zemP2_!|NSqW?lq~ua*!|S^y8z>tN~Yd3a73ijA=b)Xd`thUnDepyw9}%{>7>&COt| zTMXRZs!V%@YN6ZNRq$k$1(x*d#uLk?pkkyxY+dZLkHmKU!8dE?`2s2DHp5_kcXtv2n+W)``yBs7$^Jp49l+04N+8hgyMWFn(qZo2UH>2aGjgHtG}MR#gO6 z4?N2vcF94}!UVS4Ab~B7--O2s`}4;8`OH`UJyrz`#zIxe1y-QSPTfT^>hx(QXE2Dy zg{_A?(;c+dyi~a3+)wBp+=qLpJb@wht{9p39-fs=AU{hK>A&R*nZj*V8h>^!DhJjH zDeDH~J8Ay;^Kud%{c28b3+Cd|$PW%B9+}kfrwgrbRHqv2XVAd>y>zO13|&3PM#}EF z(vxqTSg-DD1zbOx4x9HBwr!hBO6_jr==v)BMUtrdYb`n`;}pGcARBAmPs3}kGjPE$ zTNsn_ktH;ApbjY}f5z^@!6y4bL;B3NG@XTyvj&%YWlao}n=IXE1y1umyAn+&K@S`A%TN8xW` z3KX2wp^MX^sJ5O9#j^G2v{{Xw9g_<+qZi_;dmc3L*(vd|?mqH)Xe^(!$B8G*3*nuc z^l0M#T`~_@l8DW4T`etno|WPCYm@pBoE$8DdmF?--uyP zqr{5gjSjEZOvUT!wP1T-KaM*2gRS;i0@u1y9Cu7=D#(AfhLqmgFPoz?pPcm{h70vO z$?|*Ff?vlf;YwNnb2l<2fuj{jhF(u17uEz><8q<KTi#2aP&qjbf4cY?S1Po z#?FQp*nDRTJ&*9xs>^)ETyJXg<2E&HvZB{7e@EAIxtNtMkO0RbsJT_irsiBjeUAv? z>3cU&b?*gJ_pIXvVKzJ`Cl203c92@NA!u|phSwKfCI=f|;qYg%;Jcwa^kD07ahn>e zGz}AbtW+n1h$4B~)E%Duyx~x>V4kc=Qb?q9(*#ITLpM+rlmsy~gwU|KWjRWByoU zAyx+J2o;8D!pfjejy{|3gMC*F?eX{oJtp~I`=q9!OH2dCMqGos&@Zgl-ahcRNS#if zqQqA`8o>MRxQBo1^00G!B`eiDFKB!37ffM0^EE@6lb?zG-M66*e&a;UDr#iM-HTa4 zR*3Na^%=OgUt*fi_zka4n2^6WjuXW|WsK1-b9^gW67fwg#QNN3{RbN1wBi;#FyK6E zp^?JDvDbx-%PtD`Gt-#@eT#Drbm#x>59Q_+XRn4aU-B#Il0(MBL^7+kiTu}M$mtGw zGBWp=(9h&0Kxq+s%);e3L}NN!YQ&4<<}vQ&o@G4o+DnmHKo`9+z$mr6Ijv&&Od zk$8nk30WljPCqWKANcW>PF!nnovrxVgT2&6)=lCy8XBJf%L;Y&JWLy1{;L++!!5Yx z$Dw@r@e;hF7*2s(=ZtX7l3loWk2a3pv{a7|vlZgDJ>7d(Nz{m$XLvS1RiXbe_84}f(6tA5^J;*;(c zw@79BZDnQn^kpf=%m2a0AInJ_qtN5?J@LcY<#@CGDz39BBHHh*$XfF!qWMe(n3k9e z7B4DERj369XJyk2yaH}!z3Bzzo8lkS-Z0p^H|^PC!OAY)fn04uetAaH@aRRffptRI zhgndO^Mfqv?L(#?P@t*Zrn6)HACNm43S_NYjxhRaqJ!HqU3k3n6(rU@f}&hUp3!_1 zzgQ{oU6Ji*c4H=uB@?OP*(`W0ToVhv=ds@({=mb(u2k-FCUuEAM6L}O48s=4^H~rL zKMy^Gs2k=YaZsnD=ETyk;a*ao;iR5x=nmL?}Cq^X(lLJOogB zLM-whLSpqQiKT3Xw$URg%mUVzjN_-iBvStqPpNXNDRW+S8}#HjQ1ck^wt5WNe!)ld zv!2F`y=*}AYh&;_5_J#sDbzCiz{|cEl9<8*?A$Vd|CQC?;nLgW+WSY)^WP6_SGs}| zUrwh%pWVS}%}H4Jc_5rjT~0jV3%n^?MSt&niae$b!uzbE>vJDrobU+wa$DXkeK(gk zU&59fru@*mdV!hC!G`ig$1}sf!>iJvT8Gl?1o1F$bU9$MMTZ3U~k8V!; ztNV*SzC~o-mhoc38WmbU>?pJI+KiWny^*Pn3S~=%oDiU-KbP;LA{I2BAcI#t6iYTf zgD{;e@QBbVm?N>cX76*NBVD2(YMm~9w{Io5f1L*B?2H@1jApt(EE!6dUb1Jm;dGCfeWwMkpyI~9tYDzpF%p%LBlEaiW@!qB@H}R z1=0q6O?qkk6ZG9!i8st7-q^F=s577!x33MBxc%dJ++Y=YLEf-oN|Ik& zZvx-(YzS{#kqVW@-KoOnUogAHQrx=eYr$4w1#5pkPz)MTLuSYqG4-cK#Mbf{>bqW{ z4>q62rZ3m9-=WH@q2hIB^}QN@P$Cs4$B_w zKS!>;)fCs1>re=qNe}p$(;iRmkv+z1>5orY^jPeCq8>4X>MPxbzkw>yed0FQ-TNq( zwZ}8NvUKs2W*@fNtvAe;pTh2qA0g#63W(eSPaZLO0AE)!8Uj|zar5qdX+?B0CQf|_ z@o7I`$7ws7J#7QE6W+=SB_Fdcn2Z#T94%a!Y{6|$yyFq0t>_$MJ-XRr9qsaH z1HS>4;9DO}UD1&TeeGcXRD0s0rg-eL^co2gHbZ;Q9q|34JXKy3NKcVB*fO9IFUBKx z|8h#u+I)hEUo=5B?F+~s2fDE70W_!kvi~`I)2JH1`0u}Y&^#C#WN4Cv#&hkxi?56g zR3em28Oo43C263js8kx1G?+C=!@2fnA4Fs-p-iDNq#~Jx!hQW7{Ga^qweGd<`>b^y zoYjM_);a6y?9b=@e!XqT@+oZvwo!8;~13Oqe>>gOZnN99oZzolWn@MV}rsHRg zA4LB8W@236gLWncV$Eeo>ByQ+NJ{;MMxD39i=G_;)!j}or*8tQ{dEugB1*s|Zy0P+ zv_daU4X|;BI;e`q6K8oDHrsYLi|F+w5jS(O^sGh1Y3F`Yc`}|fsf)=S@h1G;QU#x{ zFM_;*0sP;j5v0W1oQ%Hzo|-EZW5@leBtAf$TqBC4?TDW?Pv!{-slKqJzUT1&o3vWfD19mg{h^~t9u zXSiat04Gm8z*HN|n8T9<>I{kac;6X#ayf(BFY}jwj(59H3t z8&R`jw2;fcgRQ;>fvVP4CLc42se3W73y7x*%dL=G-E4gD#629)X-V%*Jls&u*_z4p*!q7Em%Atl4O~&o~iF>=(4vcOfnzM zI{iW15u`1O-PMX`IsLs! zEO13ymF~P8@%J`Eo}8Y6d`8W|6{icq{H;GZ(9uZh8lRzmj*J8iTSN9lcaj%#FOZ$r z-;g&sLF8hL3d`LqreQDhZiq~ zQ$w<0*0k$jdU^x*{p~_FzB`Ip)kHG0JquuoYCE*-%tPnLEFp^Dmk`g?y~KZiI9Y!n z1Iw7`!16Rpkeyx$HEWu%Ci;Po#cd_l#SAakie=w?)0o|*Itr8Y^gY4xv% z;{DUBXl-B=y;iB2p?xTu;QU1a1fiaHL>R)#^E?v ztdjzrmjLB?nW*iDhLF#TrE=;O{6#Jo_v8Q6>>f?pvmi!o`7*m2b2z_^JjlI}A zdKwxXJ&yPLlf|vQ7)zcn7)z}DbEzcmF`9j^2HgoSq$Ww(B&H~p{O4gve)#8*0vQ>i zxg`jGzTbrp9vF#knnq!(LnIb8{7Kof}Pi7 z=)=<>FWLl*-nv3%a{}SrfAcSMZ{SsHCHS)7j-NbhI26z#>UW&c)87P!+0)n9zkeoq z_$3zWRo9`2i-_ISenRR7g%fM(7!nvyN#7SnIu})wopY9x!#8@!%GydIlCmW$%MN4B zFNF~PNtvZ@y-l)Z!kI&eGTZ&AfEj%KkBx+UI>}gywtSR_>4|&M*8bz@*@Yo^Wswv5 z^5FzGyd;x*IVXoIA9BD^(eJ36qCIb-J_jv(`ls4s+;I>o5)z`@566OuRLba}3~k5fC#R8(n`dBml{D~Z-iwqP%phIv zG-$Hl8c*!we@Y`A739$UVF+L+|gxf|l?oXJPYmD8!h4z~;E`b{L0 z<>#iRhf&*`YIyiHhr98z1BZ{ei`oYaq1t=_o+=xHZ_2QBtQ<%&lcYfM^ zT{>w}A~$94W88V=9vFz{ydUUfsuFn~dsjeQK zZ0Jwg1KY8c=L8`;(kYr6EKkzYQ}JKzXjnAqA$sHdkQOA~N8T&eu&6PF#j)wQJ>nYF zA37!yVwXsNt`Q6Lj6)sikvO?33+r)9Mb*n&xwp!f(A+2U_>@Rf+Kbz?85rHq8I)HT=`~W@d4!v+>6O?Xs5e2)sr5eMaSGA!fp<{ z{FH#p3U^W?{nK#rl{ULEA`t&u;6)~X+6)JUz5(}eKP$^Og2=7=X~08k?z3+ySutfO zIel?DvFIK_RGjbPM;0Pr-fEyJw*sw8xP@SkDpg5+f{U39PFV6D>DjoWooQ3WAFmlR z(bB#6O+}P&4~fSQOFE!_!w=9n>zXp;|PfR6V|3*oyf^FH$YlHc3dZ(asXB0TgM5C1ULIiGCs6bN6UL@wZ6TI{)hT58oEcB&F(k}BC-tQrjN7L6yWFkUfhrA0r))vmr z-`@tShu^{D`7!X-+L%2R3$E{V?XeGGna!(dNqH}<5JK`dF z5sf6at;UiUJ<&|xGlZqAv1KMN@~QV=!E^9+fN3}@NRlTMvs-_9*|W;+>_g#KeD%>a zHsy^qsLhI?|AcIZ(cEdA+WI2iPm<61#)R_&_taQ>_*%BMF_WIxy2XtzyG=jjzhb_Q zd)Zh1JCvq{gIk6U>o$);i`?ub8-J9r1%t5UrrM%nlPs~xekqutH=qRdnE8;^pEEO^%rmh#+8($jNTa?#IRGCFJ+o7f;L37=;oskp5!>H0ZcqB|^{ zg%%a%Q9HE!mwmL(=ql zpJdqINN$wl05$a7!vfd8WmoKOvcd-%5}BE1aM{le2CqE>?k%s_dh@l+e~pN3o;g_Z zZEqh-`c%LkNUM>0ZYlZc>q;d4ijv-6izL*~LZVZj%Ps_rVG%|%xt$ZWC5mfPB*&+C zOSXAjCFS%u?5lKR<{`rQx?o#%qw6|;@_&isNs_?D-CBkREJ9GlSVi*dRW5(*(tO$ zK_k!~u_>gQeWNd*biu2au4D>3O!$IX%!z3`21=VcX3Ul&UxIvupZ&p4EIIOK!arEh zk_E|`t6+-I=dW&9Oq;c?^S7l>3th1Vq}nn8kKd9k{#beriq8t%ydHN)6KONY8H3dv zLj%j9Z>usxUB9X2yl41UjVt(cUIV>!De~r+88(xIu?I;j!E<#wTV(PKF1;Jf5{xh5 zOWbUH^Y>dMU6+YTe>zF{a}y5}W)O{QhqHrMYI&C^6Vm_Pmn=WCfpmYdC5G91aZg_b z)^s*tLuYMZ-mg^A5!YK>e8EuqF6kdm^SO*qIVch(WoJC+-*?dj%?$KNV5W?+9>+R= zS>hG3!ud4gCGuYX1`jvBg=X#B2c?}8;9~PzI&s@}3Lj?SkSD)srfePrnwbl9y5CR{ zuZHT!oXz61)&29bba z1(5zEl5^grpslzHo!V}u185`*X}n<|>D zkVV~Q&tL^h+}MJ6ZKkwkJDC(>LgeC}V;K5PoO-sEfA4>e%1NJsu&J@2ul<_eA*%@u ztFz$7(<@0quSEU1Mf8Sz2Mr0*!=^U2R5$n``n>cBX#N+1`a8x-d<;S)iHofy8(=)sd#A`= zZT|(uPHSoPKp)lqvW&BO3fx(_eDpgxo6|F^XPa_2u^Hd2K`LPgn!Kh0S}w$MsjesB zgjXVoyT6IVix*~8rR;;l+ms;p65G9!~ zd^_$V+THyKy?r_l|Em;>lne7Yf9Dve+@{C*{OGRU8mUT~9JAOC?>J_?dK)Y?oB&fC zFVNqE41qo}!52=eu*+rN=~sc9GqL(J={=bzJ)QDKOr3G}B7C0dLD?S)-wx!-b$w zenf~ga#8Jsqkju&PGAll`1P1RYV@Ha=P{&xshe!z-jJ13b%=9t9apqqJbNUPV~Ya` z^Z5G`@>=iW%X05=b+j27w`M8nbv=dn&C=v%u(5C!h-Y_G!$v0EIWyscb){ZNI7!*+eYE@qQdsb|3cZe-Tcr&3h-jDFpmtgk?2x8iNct4mY$*j zhi(4yhx-KH`{!N=?OTRh6#HQA!)NrP!E`Xa3jF)lM({b{##vWK@xhPliSblj^5gvx zmU8M4OZs*KyiLofoW8(dT>S*Ct9!(+k5gh^&y`r2^*p$8elataE(Rs@4EknPKdTrt zAfg&a@sGqLX8G_5toYKyGm=S%+&)H?PMxOvos@yoGet2A52MRPXJF>`HnF7R*Co(< zB*F`ebFs-l4Yj)2g^L6xRqT9W|K(zU>_W3}fAe?v6A^_+?YR#Bs$Q_na*pk=UXC~X z*aJg$>fqDH2lbPzplffPNAFzkL9C)JK485D6r=6gvSd-M7k z{U$3B81?^PqfG(07w^Md(_JQ79>CIf*|8Z@1DNYP5x)9*KFF^LWOt&cqt8OVQ@!a8 zxLAJ!h1i2^PGck*J@A9x{7;4h-LdRf+cjp{k<4r=H{qPSTH?BAoL>| zq&|QadF8SY)%DCPtdPx=I}0207qOt?m5g)jVm+NXEb81}HY=wD6jpq~TMzl-2t_m8 zSRX^4;Z*Ro$fqZV2<(~4XjuG5;Lp2dVufvqFhAacynnfu-S=Ms?{CE7;*fd7)OQT| zc8fxiM?Wl>E6*0`oa9d|9#>r#9s?y3KP>Bb8!3G+pyt0!S&P8;aDUT`4{dIO4&Qhv zd96zhr0ft~bP9z@(^kNNQ`JcB?FIV!lPT;}NP{A2IpVCKg>{>AlZIc=_mpE_+-_AHJGyd%0 zpLiASb!X98yIf%5q?ZoHucl!6+m`G=eiU8QXUg={$3WFU19#CxnS4)Gg~o<96krw) z$&V+2dq^1$4veIELdV%^vjZ`TO(x>yYslL_iuTXLKApK_9-|7%7ubU!BFHUnTlPn zs1x@a$Dq!j5Dg1J$kk0t6pGJ_Yi%20gLE}7Y5R&EMU~?tv-i*wOQf;p*lId$DW!h9 z(y8*jJ_sznidVf(hhT9Edl<#s#TF-=BY%`W zT(u6WV~TKfz!lUIS0l1N%i-AUTHJO~3R|4W=dO?c0B-LM7`KcDiDc`09!85WG5!9(yJ*Zd7~m<4uydGDXiei&#^GiIMZCq+1JXx) zM8$~#%)|ZyZJnehuzj@2xX=<%ed)kfIq{U*WzvE-KCmG;4e5J4Mmr`yr8iIisM`MM z4}YUAA5F1)OTT`-jih(YMIW+)*!7F^*|iCBY`f=rigVvl`wco!Gg-td?l5HLIyOv) z{|Cy|r@-m^9`F$Q^gC57h<(Nj9C&y)+AXygYSt!Hzqy!!bXT6nQ%OA?;WZxz?Y6}; z-FKl|k9V=8(jDxeupj*@Hxvq@lkmY{LvS8n4k~ku$*@~0Wc5OCRQrl@iB1U^XAdJi z**VxeYay93ph2EYcIOv1rorj&)dHW#nY8$H<9LZDaqJBx-fJ3&hpoQ2{GlTnXT1s{ z#?5CQI+bk5;X|xcKuEqenh%%mY{mX=im-;h!^Jbtq0WKh)gFFQSZ4eS(M+8K@Jjs@ zq$~b}9_P{I_qj=A_Leau@bYQyKiR{$`^#RUbZ?sA-?YIGB5lP>ADm$yKAN$w9pl;b z)lQJv{~Rs#xr6HUdsvR(e0%@m3HxGh4x4&6(wF1b^CQ2OV) zIc^ZH3OL0lrKIBi`;O3@p2TglxXsjiG$eUWW7yCuA7G3A85SO&$96b3;Z1(6s1x1B z84Lf=w1X{Zz2hsiuJ{rD@NF)Bm+69xCY95(=)Op@^AG%NZe&+fs#$dFaLL{AzhQV= z44Sy*F-L5^<8i+aVS`L*G|KQNANQq=KYo28HFBy#f)RzR`A`hn*QM#>Uxnf{?UUlf zL*8uj3LdU~NoOAGOZkX%xwz|(CEhi8FJ3d^09I3SM$Ikb$)DGdRLg$M^Qy}8Ky9N2b+F%97Zxt$nTSc*VR4@)89E^tHNf{&!iYCRt$ycOa*M~ zaRLwW9l`y6qyQ^wT;RuTN0^o(RLxhXkn*il5fGS^e^;R$pc~UHj*oY=!q}C>9`lt0t+#`2Key0sO`$+?d1986N>-V+V01lP}V)QRDfLFDU?T zJn?&d5uduU5A43iFn=}7KFV%onc?{m@N+LZB=qXS)x5+-lcPl`vu25h^bc^$YTx2a z>lgTZrztGcGsf~4qu8=dJQLq6V8^Cx;x(mi!k-hfK>o{jeCb&_)CZ4-BJXkZ>yR$c zs@W=VtyI~M0TuREOO?%OZ{SaT(PAMDQFzdKVB6DVB@YAR8PQIq4SQT6?T#jd1XbX< z(pu#Fz+D`ZIf8BN{7&tbw9wAzz2cecn`!76hQbO{AZ^_xw5ha)wgz8dFN@>YHosK1 zZLu|3uJ#AZJ-v<955Hw*v3uDojXs!3PveFhS)#^K=+pTF*TZz7``r_MPnaP&HbD&} zvMQ`E{~UDcjU)NNg41X2eYDTLgZ@yI!DXMq;m_M(NbS1;>rDCq-4Vl<+#Pt6;(6eu zzQdU=Z#3@Ga9lF0na^;2$rX4HK=Fbu;Egej@D}z)OXBH;$3;lz_G2(xX$g~iyXk!1 z0t$>$V3bTVJW{eC!G#8dyJkZy({Dmm>S(yh4#I{~4VDh+P`q~#ySGAvv~0P6XKDx> zoa6<(N_zwSI3pM`ss};CErD&KBTr)+OGJ^XGf-{KR^(bxL_H=KLSga=*uUnFum>AT zBr8|58+Afe;mQ{Fp)!&?cDWsYgFxo;#i394#^ z%+<$;xp`GU%Bij7b)Xk{Zrh8Sk|$85pXR(?GHfG_`<{E^qz+gTx<3rFN|%VHhLr@Go9Exw<;E+JRZ!1 z&hhN{#bn}_{VeX2CL2Ar60K-^Ao?qE$J#1qIsFDZHnRR0oSkUNZfE|)H>8F9e$`YK zpEeWcX%=Ad5I^3;%b(8vykGF}+F_@+t2uXvXB>*GLltgYVDN)|)YC2+&!4doE!FJ7 ze>oYl?)zexlsS^!oY#*phqTev<-6(qJ*Aw*#b6xh+m3cu%;L-W0e*_1EAQH;2aQo# zcuGCTIz_)>UHT9*@9bt0lcj>IcLKlk$|C-0b~l}|Yc1dQy+%AEXAo7M7Rx)^S+P`C zVPE<%kJ#&OAh!;rk*>cJNsJP0ZLnfn}|HkG*FJdDh&$aIH%jM(h5fXAXMfp<{ILGiMW_J7@)4&fG*QjbiLI zC6QaBc#bCKq~Nu`mZQp!K{T;59PjQshVK{o3J#B%@d$hk9Ezh)639@E#I+d(xbGey^+6wa#cIK? zSh+w{c)=GZKYfE-KFHvq<;~*QXN4kPq6aj<7gyziBol?LPjnJ zKN<2HS$=+kF1bEI9+yA!iUWV>Np_w@S8gCb*K8pHeaArd-*K2w?#iaT9f!NR7F3%( z%tiLXj_r}$1?m~P4h9m&B1uaOp7U)e30AvIYR@rJc&L&n`5glPkw&;7KmqsncHlP# z+9WT^llfODQvkAI0YIX}cFtlSe4E@GMluTOzH@3^=?o zN~E2d0@-g3u!ij#NO}L2E(z8}c@ekqf`4~W2l{}^77H8$uM(U$N|p`%c!{owI^huD zJ|2~o9R-)HmH5tTb+Yza5?A)<23Mj{jQ+?ifFGlb*+2Id*ru|a?f0I++SBGDVaqPC z)$TLT9w&yN0Fu941oN;Y^bRfI&y156$vi$HUMuelV-43L`R%sgHN6-&CDhWApa+oE zb&m^`w}RJ6#cZYdW;&#EE&pBbF8vr|M2hESz>1t#(9AB0PugDPS7n*vdmbOfx!!qv zS;|x7ZdJ+U&p1pi&$qzP>d91TQw7Zt?03?w`><8oM4XxE%S<~h(f(`hptE!j#68ib zNV9_L-JFeBg(`S`dI2+fkAYc+G^>^R3-cm9z(=%;dy-nfkF(lJ-BVV8fk+Eh>1wf$ z6 zr_MyXa}*%zQwk{7M#1umLE<5&a`?aRYS6Rw@t6kcVcV=~eBO2se>+Z?8IM1PF4*RA zj>}equ9OqBpU$AW<8M>{>SwfJlP$pWQIgObx=j0A3ctjuv8wo?B_HRU%%1g?ug z3buS%4vGjkjq*-(qcdHLSUl5XPRFLQPi7We^V?;xX|^I$l`~;IO{Pru;8f6jKzWa2 zrSz+ZH^=29vf*n6NwiiElZ3eZr4#Qp(C(*sbh_JL6qc*Yy_Zzr0kxI%_RFzgl46Ee zJ?BIomntFX+7;GbY{yn=R-n}NiLh&_8JhdT4A0)C#Lhf#hK_)zFkMc8{kPyAO*>0z zZeS$(?0pL|{m#R-mTVY(zn6x|&gD);j)SRNVxe|~ChJc(Wxi83v77Em{HW3!XmzPH zn)bO@q@f`P%LfHOA`PN5HeKd#Rg40wj|ZUSr#q{^F8KGd+u@p~EI39_CT2=wiF17q z7IzP0r5h8#*4mjyh+SzcjD}7}F|^Gs!v0fFARp>QqjqJpBDd8rRQv~43#lb;vdsL{IdrtB<7nGKJaz}Tk~!P% z>}*$~IJbE?biF*CeN~Bg1gxXzRXa*AC_AD%i?kB0aMi6y9$La&t#s$XK$YTRnWa0UOg{lG?8sl zAroE!uX2Ar+F*HvpEe?!rOF<{^L@0iU$4ML752lye~oZ>tpQ&9Cj}0T&*c`9M*3vK zfS5U+h;Iw5n%tmH{BdF{{<3EVK6QE&*(RNV1Li)1Vh_rl z*7k(^-fArQKqwUL(_lY7M6=ilPeH1(h-O-L)2-?{Z1Eozc4tpDq;T>~-Ypt>cDBHV z1;U&E#E7<9r;(QxJBiL6BIKgNSZ>t^xT&Yi$gk(LKjIK-A3Kh&%(8+XH`dXT6$zYv z*>zZW--Lxl3}cC}zoV$@i)d)bTPO|sN3Tx1C2G(=TCF92o)$EF;zJEbz~GrNYx%H` zHj6!|@nAFjWpfHd8+*b$pTlH{*&(v?Xc>7S;Pw`$2a-wc@qEcpckp?70yg!pgqHm# zkZC>&b+@U~pnMV8EQ%%;;zMMr;K2$@S8%+3c_)clsweo;Vwvi*=TMobN)~61!UN@m zG@jT9x~sp6P2asjUDE$hM!e8#JnDl3XXQ}cPhRND+a}S>j1{b)EJt)A%?v6od4uiq zOg#Rf4xT8A#TLyPw9hLHL~+Wv;Fce^ICOw_>v3l_kG%yZMfEt4zV!-;Zjl%mHbBm9Xzl!Se4+ z>5pomlQ}Pr=(-VNDApvg5?^vd>NTF&DUVHEIkJ(claA3IbVIu}_UPS*dh`U>GuC4) zUyVghUJi$TpN8#?-_S7!U6k~~j z7T-dN^5QyrAEq(OvN^D4?poORpDB?!2t>*zhg2l2W!FPSFx8$e$i3f+IHfbZyyFx& zX@J@8(xb#~(qE#pKLX!5wUeG_%`kcDRFXJ(ELj_8N`g(+!hiaf_)ual-=WlsU+Zhb z<%_x?AHSV==}jV|*G?ciHb2B;cs=s(eHvChGm+T+dS1OTLItnaJw}$6w3DM-v>ivZ zYQUMd90`8A4p-!lU=eX!*eAQA;NdPs*QiPn&1Dh9AbTkjcgm0*^*^Y`krIfUHk(Ye zo=+n#8?!M9I~e!(DTpUc!Esi@N!-E_r2W?^;#c*HY+j`7xO{szY2Iz^Sar9E9Lm3m zo0dr7Z9Y~Iu|$!Gl(rF>lkbS^N-^0#rxAZlJ&UhhHpAY^UHH4H7KzrEB~k-9{T`F#G5kY5X(+JQV^Vk8$d*mMH3Hy8SDAtfwT@DY`FjD(;k!`P&aUG%`A zePHPtXBa@?sZ8+R zqRt#DbD6=JNv!esVHPKFz8BXWgYE<#w#}Cj{M^?e^+YB+aM%W!G+Ck*b1QIss4_dh zK1OupcQUvj4)*5@tgP|pc+0*k(6vqA7KXcWBbNohWuMtdI{rG)35OX7D&03=%BB@9JQNA9RAb@+B-@Q_doHA7mH= zmwl-gn93{%Wt@qR*13X*-yZn;@IL(#c@%#LJBgK^c7pyr!CP|`;N8VJ|jUF z{7LG-H8AsPg~7q^u`QcYumc}e z*I*TT`%#Kh4J`ilh(7%nKr?*|*}C03_(y4p;=O%m98~V_pvfwPUKD1M+uhR1exa+` z;y8y8u{B*`9R%0Ld6O#_j-&p2y`lo)z zE(ctQ$xAgNbx)QxEOJBd2C^ zcX+z#F#69;m~F3=#^SR_X^nw9cB|Ed??V3CoEMRd?fJOH<0k&lY>byncBAoq3V7|{ zC=gi`ioIi(aBmfdihR$7;9QrxFeqaHjTs$-)TS>cJ}?-ubhj0fXIiF+W@ z&5GL>@dU1z&E$RbQpFilrJ-zzGRy66r&8_n*z2}4ptoQx+`JotEe6fQA?J?Z7Zu}a zfpRn#@$>|zym31SGbJ`FvkXsawc&3PFaE}>mwbuteQLSf8NP%Q+~hS0oL}8W$6A`G zWe?-yyv^xv&q7eyJ%pB|J_N@fQXo3|9{K|=Q^TBR_`y(Bh?B%3hZXI3`kEbL=|*d= z*{FyMP3@tcMXO=D z8-r)Lr;~cm;q3C`@g!E^F`5?Igv;djbIWAk<3|k#1Wx=U{2;9b>-0@<*teiT$XI1U zYtCal&{hOS#imGV{B8)i@Q;Su6~mh;txVsroEg49z-IYgUhu6t3*}o6Lft(n zIQgwFZV79~9ZXFeX?jxR^2&!-4_Sl$M7ywa2~G6n!+m(n9T)Pt*_f_a81~dO0~Ffokj2?kd}31#ui~|qHpB>X*78`>R-^nGP2PDncNaX3=F6@vyZeM#Q&E{?+%EOWGFm#N(LsmNskhrYL+CKF{w zI{qU=$&w+-sOUC>`FrHpl8krFToAN{cNoVW$JMU z>D$SB$mY%k#4xrqx2^Ma}yI{b=}4rJMqH|V@<6YsxvKfL>Qn>h~? z^1U;?*{qUP%%D$?8OAMT3oqxuriIf%kqssW`)1({LXLUE$T?VP`gb@NT@QzSN3!<0 zCGj6aG7Z^x-0qBid5X?4SjSuQygy1D9?1ps&lbW-fCH}I2!8fw0jq@3CrHcvk*y6)T z+%YyHqKz(V7z6eNB3S&njngk3M$ZqKhb?;Kapt$5NN>kp0S&?M|v-S#QC${9Wt6KlKyRJ*g5{Vl*ey{`OFA;V)v(7?-fj66H zUyE!0KEYWd3-~2xhQg7LMQG@@%UDCc3g58xChZy>a7V8n+6FqoP9)?IPn$8NB^cH% zet?VY42j$bf*W(Y;ec@mI`?1-Wv~MJboEj z-ny5h#=fJIzTD!kOYOz_MUP3O)&$}ukHI2f6zOgq1YcI=(zVI*pb~PGDnHSoo^Fat zROZhpsT3)x|9N`4$;wGdDJcoB{(q0bQchCKw*&+Rg{+#ea{0F96C50@Z6-R{JIruK zQojGMkN@u%{m*qgzjmYge_sF3^Z#@m_W%2JSaUXZGn}pe@2?}-*9;pLYvVz|*=)+^ zc=BuUc3fbh0iCWK`}E)>%Mv4etnd|DlhDJhU7$m@PjE$d>>p8!gW1gQY9&!iD<>|V zQjU5aO~~q64p9A%*#F!PvUJ!Zao+3)bkN_o%viCPU5qnh^NjjMXFcrLJ;-KP)`!q- zK{?{xNx}|i_5?7#t55CGBcG{#+iJvBiXSkFV!D z(v-2WjvI?w@D;zWm1Y{52?Rpk5zD?N6#dQxKbbWK?%j7K_Cv+Qszf+nAH0fLO$_n6 z(t#I-OCaFEC~|&C626er$A7+(gbU_QCwfP`pi-`sHWVw9SI3VaZv1gPvbhpZ_emnh zkLi;;&TiyOLoShY?8i5>rI=s+elldaEa{CF_PGaov4epwDu3Dz2JwYhYrzVVHqy}1 z%m&c7i=VNzd?KXinUX0wdF1fHP%@#;4g1aBgnb^=@%;}XVerZ*c74}=d^PMjT39DX zP8nUsuN{Pp^VEZ^&x=hry3_cyLu*+|b8*pLznWm3Fl7uj!G47a+|vGVS8>{=v8 zc6S^hT@hFC!d-h{_pd9EMd}iz4~-!Aq8wJdPGx!xQpD$571!MQhpvpaksSJAD*57bj@?UZ#^2^I!@GA} zg|Qn9$)^jlMDt<_^sKwXCJJo1uX(Y|{mC6#Fnceq3A4rXzFlB5XQ)f;+oUDEVdeC1 z%NV?&Mjc0$Fuc9)q4-Ck2K*i(VFz9elhouXNhW@p&d2-Pz~@^=n7MSZAD)*Sy4O9V zXN!dV`N(lBv$X|`><;77@-}Rvrh`QXGuZ7bS!~Z(d$zW;o_)BL!jkVHwy^&l&RXsR z_a@rYT{7TjGt;fdU-;j=NnIBhsios%VGO%kzU@^f@J#ST|B4re!- z>+$}kUh>BEKHM7e4Z5Z7z&E>i#{3%A)<4O-rqnXu!`W=q z=f8Nu@glry#ZBUWs*06=tOWx#;d}4rO8B!Wfz7IN1&!^?$k*X{tYT>z8nOQf_E4CC zPwHAC!$;lxzL^uNb5zQqH1-2%drLtNI$b?(5=UZ|NkCF?fH{6Dqu~}If1K2-Gjv0*|6909nHBL%@$lqfFB2sb2|i<(7TK_!dT(w7_py>Fyz}%y|WoSrG*p zZ;#SLNg8Bej{%c2Z33H*U+IS{|B*Ry^Z+ zJDY4IEvh`nL1Ux<3|kZAdb5g93kukIr4QhI{oATME&=ePF#JI?7TYz zI!scqO-vM<`ENAKS)B?+v705;%`uXr%C{vm&Uc~{f47KyLvx|^$Oqa|`@bkU4}U7( zH;$7%i--^j@2}Ha=jP6 zrEMbDKdO>R#&a=A<~r;8Z4rDJHYN#D8)4w(d*E3=g=L@j!TPsnVB3Qws9!su2sVU~ z+XoO1r%a?dMFw==F)5yZYYK1fY-#4N$2ssnd=ADaCX($TekgUAn+0k`g&}YBN!P(b zh8Q=Ih3YJM7&;q93>}D@V$YJ72k=|H>K&IYglFjRuMagyBw(MO?OeJRX!95)6c#$Isqw_(|#x z+oP3gje#0d1>x=4(H_MgN*l2gQ9+UB6It_Kx5tFi; zQ17k>Zr^p0o1JcB$2=iUtUW@N3UzSHi(Y1f`aQNXQyTWhl`?MyTvy=LYdk)mMLq6J z-6ax5@SQWJOdcd90qXo{MI-(SgGt!!t$`v^P9$X9B>vX3C#aj&Ju2pYp80w}12Qkh zqrT{UyuJ20s+&(0?p@ypvr@`1r^X&1cdtRqC|g{-S_b8Vtc8v9mI$AG+#;M~u@?2_ zU&lLZM1&*9x&H_0$2hnFF#NNLu&J;aKkIa0kW&$}(ps1QTVV|3o*Y2P$VZDJani)K z{WjZa;@!~*JXWnj_o*{v{>004?bWgTi-%nKT^HBzF`<$^ow||O?MQV+=KMXZc_bwqXQM2v z1`qr^6o-u?ZZJiu2OcgwiFdu$qeu+bUpwE)K74YM>{jRkjqLZV&o*WFWcME`4qwKd zKQq|u(ch4o5Dpr5mAG?l92=X)aU4vS(boOd=#YLI$IOq#b1_cjaegpaZKF%V26nC+8olfIYz9(M21n^#N~M!c#rSpf-LVd_>SF&lPl8Da`r{EEuKQxty*j?E0)2# zIw2DM6+-b)R1(hV0wyY12U=YsfJ0z`Vr?3faetu~Gw#D4a|8D2WGNI~@RRvG&|@t> zK^cB3|6_HoM#F{vZTL0f56Y)BW0%yxC7)&TUpOG-?T*GuTt=y%(8w8n&2VpMD9u)!L7qu!X?Sw zXt1INgLyKX7kV#rr*iC_`BPCb+8b5Ar{dgOr_c>G@#@=s*jGCRUrs22q#^}evsIm4 z%1adV8mWL?>RpsG;@ojYt*E}T0k5tI#}f}CQ2EMth?{JNojn_2^NTp)#q8d^1zThv}LMcC9EB+Q<*NNB!cj4=0w1MkuwZ!Bm$hW#hph3sw4 zH+%CY&YhD*Ak>wy>u6_cEKS#ZtPcS<18ct4q!R@Ikus*BPgjiSp*pnQ7a$fF{U*t1YCwR*kjrsV0!Qk68-2P zGkDI6-L*r9*RiF6U0MB@^{g|1iLVWKw#EP9ky;CwJ0}Iog2Lcz;~Tscy#kMNe$Cz^ z0@#0HAs$~LfQYLW80VmlDW;0(yls*&bf35|?5V79k%j?oju?xUw|8*wMHk$-@Fi?f z?5;{47tHGnc*gNz;z5}3jji#UB}m(+#~bdkq-%<$SlP~N>|~E@HnmIw+jlv@FByAU z?J^ds-&YdHfh$O#PpTaHO-3}T{MJ~CoKie%-i4>Y0b zJy{XhOez=hshRY7a^%J-GDgjhmK;!{Cn7zl3l{L=GP8Nw=C7-La~I$y!7{o|c`B)@ z{z;Gci}E!mh0xIQO~md-IQ+ixjT!&JpJ%mc65MS#$n>ckAbXpviDCQ+a?d=LSY8+) zCWRHuSKp`P=E+22$2p^v8xIRw9_o?H(n4ZMzqHObRcln{(oJ|Dxs=fpq4_=k#)`2Ynrg zWNX4A%KnWZS0xtUMXx#J^?C=IYgR>+4oad*@f&u^L?6<#bOyysgv_GmOu)&t?ZV2^_Ujo5aTj?KbF|ww&Q;>ckYWAo` zH8H+&hVHB2-U+6$bm++p8tc+TT773xt*2F_;C(8~vx$Isp(OGqPheDJxZb6m7PFvX z76}_fc1?5;$=Gw2279UT^#yd?1#_CAQ^+_h z3nn84EY?VD2Fdy1)U@A-@(-2M8&wX~Me;L+`ij~#>Z1t_5GjXOCKAFj>tA@_tfNp$ zgh3t8i6jmQhJ3chLW6fiMtudmQ;|Kw5W&9OAr}*E_jps*tWs?&Z=Fv-& zV&QrDR$5zrm$)j2!j)SgeCEvp{(4hO{?2$&Vc6eKu;ocSFJef6A6NC2hVD-wht@=b zTlXZMr%D)8YQUpkBhEtO(*LY>+uP$InLfPgU?^N9@rv{2ayx)IQDj+k2uU9Zp@*6O zn1>!IXlwQxZ{;`&!$u|xcb2OO(+gf`tO*{F@$_@7{GN5%PbU^YpxxJnXjlj+{= z3y2GDfl9`Mir+j1#|7iSXxCzLdBP2N9ebXzQ$y+T-Yr!BeK9+B574y*vh>ZeH*|r# zG#1P+hd8g>kTAHJUsvqIZ(m}^*VH~mkNg~ushvNN&2txepRp2F*wo-Ka$Swj;=+L> zb>WUfr}1R*HvI4}5wq`XMtVq0IP$ zjH6Lx$Xoc;xStBvG*K(Yo>|U*q|baNQ#Y>bwR>|3gc%bw6>~%p122+&@fJDzK5X`r zf;ifH^&u4o6mpreNbC>G!`W8qAUmoAGI3h0S&1xG+(WqOp+UTIK9g_8K8S;!IPUC2 zUg~BgP^@ZWbGNGr^lSzQEVQMb$9B=+1G{*SJVtqbNiih9b}uunc7*e>_d}(tGai~% zfpc7=@n-QSc8#h&^JdIWG9~XKu3ZgSw!{Rr$0>s2qXd{LVTUhDuR@taE*KWoKzHR8 zsL3^_yzegboV_7d9=5>Z`A6Xdmo*z75CflXjlmz67NRC|3!Ps*!U;u>;eP24wy#S8 z{=W{7+rWK49iPyuk!~p9K2vp$BggksVfIFZGwZZxV4Xn_V1^3Z>=gs;0~6>GHGSG@ zqe_2nkB9NCi}Bv{^WbRLOpd83pq->QmYM~k@5g7bOz&m&qzmIQZTm_L2nj%MtDmqU z_9!z?wUt=3=`xW)dW?os1PLBp#a#6XhG5)^nuG69Yl%H>n>HQt`z6V!ZnL0uWtFwY z?gsd_te9yZ5-0NJo}6PZlr7yS&csH2hXel&!HT^nnBFW=Y&AB%vKcdrA$?;xJ9Gb#V1a}V7L2-L+`dJatTr8&#Z`gmrc825 z#g-AfAVG1g~YCBq+=Rca7M9PlE{wns0@A+?WA!mySEqZQf2h;cipYoA8on0z!l z_n(lodo3c5D-(H#FpD{oE`st4R^w^+5$3yuA+KOZgkbE>yTmgsfV>n2!;^P0jB<4- z`TN8Uo^foT(FMIQMOm4~)_;H@a)o{6zmxiWxk%ND!^rINB3QTc2P-vi4h^b)N+G5Sy(li1IYir8ZwbG=6jQmbDw5VEsLY{XG}lQf4qu@j-5^~ao&d;b((Z}!ZzN2t8P;3{v7h~!FtG! zJxAk0l8EdgMRa=Thl)KBJj=`XdG|in@&*^2BXrdbYI5Ta33#!WoJ)|T!Q0Q$bG74W z(CHf#Al3f!>cq*OYefn3sG#I#5s1&J5RcKI zhEzO{WRLCk!m@D_gn@4#V(yP1_!w{o7B61_XL~oW>1!{tIiGTP=`T|l%dJ7I*@cs6 z+yLlWvJLw^N?`4|Vy0DYCtGMU14Fh}v&$-8!@kZE+!gAK|3)DZOXh z#&#e$DK$oFvOkP6iBKH=kuH{&L$14gt7O<;P8kjE_ru}v|Gy2*@_>@ z)>Yrx{%HrO$kb}qH=z~kD^u9kFe98LlEdzs`v}aU5>cU{3tvj)aXs!jXfMA77aNk< zh^gbqft(QZs3+)k+Yov=hxF{B&&;)yB09_U65yi6g2vjLOijymvMx;%3&nhJ`rcM@ zdto8@*(k|4XK4^m=PT^}jr+;Rf=j?kEF$NJeK3#R#ca4%Dd_2{BI!vxNK~B%&4TxI z?Y0$ox;7JHw@j`+vH1?>KC&Z&>%?h^Y%I=oi^nOkajb%+jPS*T5BTOyCVDvBWL#Bb z=tAunGN#v&*jKpYnvc~Ocb!i~2WrTfbRUxSVGzztVD_5Mv$h3s$_7hJ%uKtgo`}ICx?P6|uFYUSorZ*pg9jD@?^vy|*YSAtXcA*GN~` zRn#4h6I`1l&z4MU;!SBOA+=Nd*)Jlq;o`m>s4G#=EN2!#pX@u96sR-b=O^&~U8DT4 zKQHKH#VqPEehKYV-b12{ir{B*CT$8Vr1pnrP|{k-?A?@wOTKsGzlYAG$a#fzLYWCM zReOxPn?IuwamC5Da@?J(Bb47%;;7bU(AGbmK7V?KEZ0rN0F&?d--{PGz&8dLZGQqD z)nxDKN&I^q6ZlJ09#Pq}Luf2ojUJZUao&GDL>L`NZrxHNGv6JhyIjW73mlK|yyrvu zOu>fs|JguymvJ3*B`?QG0Q z;lcN1*lU}K_ij4kOKvxKNHhbr_v#AWBBs+fi{C@pm|Bwgy$r8fJVx(85yJjCPq$UK z5R2Yqrr6^Wu6X5zuexJFSuB?Q9dW&SL8&q+KX%o6Uc?5nt8WRD`#FPN-13T4hggys ziAJ=2%5(Owhb9@hok!v_G)ecu6U1@vCuaBiepY>c6vQ6)fY(o{z;f~r_U5&t^dX$xLW>kAS_Uo{a5DU1nu(0)e#%?tk8b?&68;j!Z4= zi~In4{3k(A?ht9_c;$=6<}ty;;?&svG#R)36|28Jnsc*0WNu%cM6Qh=L!ILK$iJRI z#_5p`z7BJS2ZhzRU*r-X zzuW;{*RJ9X|15m7*AAC%-N13WZt#58=&**ik@Va17~<^ti+!T63iFpe2WW|6-`&@y zAukyEuKOX$v@YQ8ilXUW$B9H>;zk@(kFkM`a$vIUJgnwDf+mZDg4y5lSd+JhNbk3u z)Lvcz?}Q~29rNpCep3m%hq(-4nmuIyupuMY&OKMiQ|2u1CG@4$!VE($L(=F8cP1rr ze(!MjRZ~UWQkD|qm73UHISKyU|IKFk-G$BjvoOfG23O=A=4RYHo@b*9eaxRlFQncf zsTqqIiTQ10jmkbeQ(}!rx66=QPk59DNbt85|E2eyak<*PE@WM*GR?U+f$rR+PuFRb z(85biMDI)yF%ZbpIho>A+_;An&s3xFJ!jxx#aQNEbruXx4a4L68d=|T8*){7Bk^C% z@$AoQ(2K^-xV%Cgbv=fl>FNP$xaciS`Y}jXy6=L;kvGYix4z7u)pj6uVjfxk$%uTf zmLeJRR`7N^U0~O0t;P_S%OKJ{8T$ABflaeNkumieY;(zO*eIXOH2!ys^=p0v!Xu94 z{{2EmUF$5Evu)(NoHl=wT@pPf`j4!ZR-mFUn%LZTJ*0Ekgf3mLK%ZCI(>YvLRobS3 zw|=Uwz+%curnJ$U-f1)^J)d<@zit~<(p90q?-@Z|_-C}wyNx2X`*5$xcX}=CEA@>k zCu8hGxjWBQ_*A=&^w;&!>sLO}Ijeq>*r)k$^65HwrIttyTMYRQ89g-eOgi1weu!;u z983Q>$nj(16iB||05SJ?O60Teld)>4>=io&JbgM553H1g+GSzjbN4;tyI+Y81-g** zeF```Y#-XnNMpueG9jEh^-t|}My6{9k^6WZ_B=a_4vVGHPV^&${V+t48wZJz*9z*; z2UPOpGW_8(1t)F~hVp$CO!w-W1jlpih0IuHbTo>}zJ5a=ScKD?WCCr=2!|Ix{*fwm zThx8*fHqaPp>5t;u)Xt>JslLy>}v~V_b!+Q?Q<*2$qsjxFllTHv9C3lVX0y`iL3+{^W%Xb*j3YT{BbFmAn(xOBn zU4of2T>iLeS|kol&}2umLh#DeE*Q_v55GsYQ0E^`WF0HU1U3Ysnbu0wXo!V3YBlh> zYz(TN3Sc!9jxo*Saygf*7py!Z&CcC#!fWjaXHUxxgKYE|c5eCyR_{wXUW=_}ByW#a z_qk36xrHN)KE|@nwb%os~$r)Ez>1_ z7XoO|H%ZFBSH!rDq?5S|W1w^XI^Mg`c6LBLlrA6GLd^?;=!EN>!;JW{H}fT!GpBo@ z(%}uLhKSJm^Ep(^sfhYc3xeS5>v&g|?jTK4+Qcw>ISIDCO5CI4;qZ_qbGYd`S74Y) zZx?kTC!}Hhf4{0{%lr`77|3Bh$M;;S;?0=u7zb5R zpNZFM9jItcLoKm0#O>Yzx?4n=T=M&f8!mODM;Z0gR))eWWo`@`X+Z51y z?r(`tOE!CM{{vRu=>^gG^_F*}^e?;rhpIsHXBL}Pr9&-`THw?CdT^EK5~`q{&|o%y6BEexl` zPb1qT|6rs3ZM?bvB^cI*QSJDX$TIQxZnYncbUcOATI{HC_CES^+5qHS4rl%a)WFDx zYwY3Yc3{!Ty+7|&5k_V|T5exQ-pYAUZT>2HU28L2ac&x_F1iZz(lglf=`qP$JR6@| zk3ox`T((-gK(MXuCcAmWkt9rWV?p?g7)IY=rg=RknJrTALw6>1>5j(fZnK1HFQw@0 zFd4EWpcVoj1(Ea@1I+0M|KZ{h%A9(U#(K2rV#GsHsw4N1?D&|=6Bq@t%p5Jof9(PG z!@F4G`hcN-PQSu5)wTRyoy&ApTNviEI*#t`c_M^pGvb_{<}$17UhDbCLtoPwx3p6Eue$M4TM zAK-}$I%TW|-(>nKHs$p@vgf%O)A>uC-8uS~{1B0WhkKG~uC+UVyyqFZxos1vEw=)x zZ-=3NT@1FKk`f-x`2pt~X2I~*5#ve{sNH}{F{`H0NOl7xW z#?eSHN?S{B_S~maYGontYYS10{YYBHO33M06)|@!U zp3k|%JEWgYX2d07`?PTU5uZW+oa!gBqGlWqVgl&2e&(&2bb=RfW{@#1rhxD z(UA$SS&2{M4-pF&F`~2m0hV*5YU%3<_}6t1zU6ArXP0I($y+n9D^nItHl#!TFvoQ) zoP_&IU!zP_0>@w0p?BUDlFMyjz)NAM*K&L2Y)1&V@pz=8TZv4*mduV#A~fYmDH%>y z;=2%4!mE)$&$YSCqXt*ja_@3xFj|=!HS8g~t3NZ1@fi^1R!eNZO#tr?&%ug257xaM zgMTA8(ajw(409w98n4t6;WJ5&J(NuL6i%k24;GSGhjS!B@D1B4)P*m$KZdpCLnMA# zBKv7Y04aYEk7gQoVNLum_``KJ-o8x1z;)(W^HNRNTwH*MgD270v)>>83mk{zWE$^TgdxrSD+WvZ*0bT~tTB2~ z44f$nqLHO(IJzUA{r2uPF@AQK`07;BgAp+_vyWr8s+rm4Y2aVNeUyort9^YM1O z82WzSO^&-4fqJ_YGfpCiCaiGd$C-Ymd^dUFr1Hzcwc_ne^|%JQv9g1te^T}< zJ`=z7_%X8T%Ji$zRa!pq5F7?Z>C8QuG&9AFiL0of|MJB74EoaE3CUREY%7df{EN#< zt>u5;o=q+Ou0W?UCwin`i*KQ@6<$PFu=76dBssysWUg@`IlDfA*bW$=(j+%@7EfYk zZ_%eWN()g}E*N>G6k&Wp1b6X+88RnE}eFMna6Lp`vAYkuY`)O4%X)BWm-3CO#T^9C&xdnA$qqJ zG49wY+L9PRx5wAw?3MSCciNhE<_bu{?IZMaW-8hKBo~aoCbAE=IDo~vv2uC@Y391l=*B$d8^;TdWH9>e(n@175;?DjnC-b!~yza zswh9>xjf&m(-K>jcEQLuCD2%3%lL)+TFq#&fcuuAG_foj{G=y?^*GKwa&#e;N`Fj@ zd8V{)kq>2a@6)*YAzE^`gy=5vB7>^+#J`fuYF59e8{R7N*+u8+gFDvP_PG3cMJxmHU z-7N6A%^fl`aXgA$UB|>c-UvkoA!I3+M-as9K_5{$%+*XF%#8Iow#ygYOSs*%%|hXT znyK*L%dc1x)Q?Tp^YG;-?z7@jaOW8*;o`(*3|Mzacx+&qFvnb0=#zOHQuF?ReD*~~ z+;jpee~$&ae=*qAaJ-6Xf8cFEH|$>_h3&3mA@IyWc-XK*__jGis1-9+_+(};mTtd@ z&wLeeiP&*ekUEHC4vfcB)s1NUn+4@u0aU9_rCZd_(lXz%XmMdL#x#ebQ{pPnZ<~kF zN{`s&sv}_2k_=G=>R8uz1GnzJiyrcp=uo!Q4* zloIO7&w<dGh=>K3@S}BPB8f5K~*Px2#7qk^c?2Vs=H&(yL$r1gKc>fjjSK3sE z*LIQT$_%vD$O+xfT*mpIo)Rma8ltXiKx;onlJHNX==tv_isoMdmowvpXJ07@=O=RR z@jR~cJV%^*DNe`zld^@sUA79Fw^HFcg>>9y{~jZ}m4)5X3ot~y2ty7{6b||PCGT49 zlT}a6nNaIvL~vyVY46V`r?yVVO>YiDJw3sE&KpPmlhA@GmutyL-(9#{nhBTgsxfwk z;nlfUoQa@vBixwmKuZr_5qx`dh3#&8MGgBE>4h>e^pEBEett3V?0Ob_5noS&H_6c_ zWkt+M5UYkU=RhUp3Qr4Fx`;fGhlY}FYl`g8v?YH9BU7VSDXvt2~E zYV24hcXttOQ#QiC_P>}lxSO0kv;ZEbexU5=B)*ziEpKsMJ5w>d&8lbHS(@nnjPB`A zp*zOJGXaY?(=gqa^wp;yJb4>8!n}(ii&qj-7q$ShGyQSPruWc0K7u!MW zy82lc-|pr<=gyYQ5~*KvxhaZbh$I-4+R=|oUS)DRd?k)xmg-KFZZl78N~gXhO{ zWuC;XLD#sX()@(`&^mt|b-L|Ev~S-ghwi?lTg4a9=$IS4K*KoZePJf~ayFbs z%O8ew?{Wm;=j`dyd}Ss%?2_P<+EK=b9Hf#F&eYG1n@QGeV3}YQ2huH?&UMD7 z6I*{hYO}=+OppB`4#$IG&dt?CVV)23T>TiylL1Ea!VH>er$lS|TLgVqo|C%le#T;4 z7HNxcCkYp|n3yUF+BG;#Q`brI2gS$W&zHwxk4iE~v7$6|&nD_)Bp`032YCiro0#f) zH@e719zQ6ipi0>@HlMc^4|6@DAG>BzMcG+2xn6`X^Q@ZwOj%X^x=xlHS=z!|!}T?X z{(DP?d28s->@RGmj68j~EQ_di)kT^kKyM z<2bf`fL*##9SUjPcwO#+|Gw$r<8!g#KXw_hXkBGp)^17EKK`!mKATUjj5gwb zTPiVEWiz%MyNjDU^6~n%g~AY@oA9!W%VK_MBT)`MQ1DlT3@jK!jy%1LS9ZDKLD4IC z@TE41DCEAU)fuFXnF#)qw!zS_6L!7WNOO;i(cf>klKMhxa%Z|6zx(xPnk1WvcJ%?E z;owX{>&4iZ6Md}Y?R~KI!W>Xs)&*YR(1%li(kiIT6@?9W}2kJV4m>gxL{%V z>EpuhTOM(z@^Sbn-G%Zf{peiY+w6vR<&lFuz$}EQhxO| z&0_SqjIlkeo!UwkdtZbL*6sq0Z+qBG%Q-l666Yn5?Z8d4IXI>G1;>^V!TFm*SQ+08 z?3Y|e$KLhE8#8#&zzjjiH-F~BrwE)$MHlL<+qI$ zZfRJ}bnW^?CXFk?b|;{W%3N9|mX0aob1_pP3#3IQVS%YC zHOWsTzw{J^5;AH+v5pd)kUD_o8Qgko1Vg*rggNaoAXn%SD6LRue0t8QeZfw3#K>>P<0tuDtcLr zgq>OjHJhAJ;FAjO4p#KWg4Oi>n{4tmbtW}ek0zEcO_|`raQfX)n=D+weXl-ef%nfa zcpR3?+D~;vXNPPwpX`PY6ppib9V*xoRe_s(50S2;9xz$74Azy+<_}s_QUj@}^zq+J zTC-{z&DWld+TCa1s-P2Q^o)XLRW4Nv*o(d{Nx06x8We)U$f=N-fa!O?az-m|Cj%&9SsbhgZU`28h@bRPW+QRSZ?u(}*Gn&+@@(_4wOxdA2z@aU1Q zlk|S79u-l%Wqp#0;~en_!tCL=H#2xRPq_sqe%eQ@W{1-||0R%*He2DzvIG>b7-oW; zk7DNpcNDuj0mUt<@ZXd|{9arIBl~92IL}C0eyxC>6-ePraqfAQu$bwy;dq~}oW<{v zI354N8b{`OgWR1;G(MW;J+y%NYIcMu&TWUN>f@02<_3%}`pP`@ z&xBe7mS|XcW9-HOHnAa+w?fs}dhN|H_IK-Y-kQ=g^oZ#k(zN6Ud1j@Ac@M17Mo}6^ zl#(#WYaNYv69NZ*72vPScQACCD$F`7N6E)!o+t}`FMa8@_>2oCNpC*HkFNntMZaQI2CA!}_ z413pHWg1VNqB3UYwE2_^Tx<-Z2g;iTjd7KD^~pMV;d};d^&X{#J7oDY9_jN(+P3f^ zLyDiKI+?GLIhG$V-XEN%y@q9nGLifVq_(dGjC}Gx(r24T54H!8En(wA3tS zT~R3M@O?`T#syJ{uP14h+G}^b5U3ABbJf69UY}&h*p7V&HCg-Nmp4?h$&D~Gr1T~V*-WW)3 zILbcvoWbl`Hjy;V|4r65Rg$EdI#zy70_Lr_f$7d1L+d3Ejhl?gw;$Tz@oXG*dh5r{ z?{`V``n9yN;0oFJ^cNG{P2elXwfkc_mmK>xOjI_f(U)&($=D=&TBvPJuZR>wU*BvT zy03XjK0?s;Ftq=jX)2 zpQHwMg~VE-tso@v?j)aHB(aRs)=%I8| zn7vY*f9YT{%?lbwTRj)ie<|@avQV8yoerSPDj{_ln@TmiBN;vQY_ej=lJfjcb6Es? zy7=ljnsy?VeUu(5n7MO0*1Aa28Bc=f;cFqdA!rJFA+U&OzPx57cWxZ!*ww?yO_xz4 zSekdwBF?%^^Eq=`%Le7|UuUI}^T%z~5d8D3=AGGENW8nb**N_=oNs7{3hVpyqJcf% zxiOYrH6pD4pUbrGUJ9j3k%A2-BILN47I8OdVfF|wF^YQA$f?Wn%)n{|xH0tuv!vY$ zAJA^BvQrWEiukje!md!G$QgVcH6PmVzKnLxT#gl00@m*59`?k4SzLcOid7QsAWO&o zVY*bz;e=Wy%U;uCGb1aAUnHS}FYeOCZkr&=t6R|5$2sGD4FnRZ=Rs$a77D(`K+;qj zY<=5+POK;%|K3eXp8sL5nQ-?U!}9#Tlur7(d>cE#Hj?GHtYMCI?qaR%XX1Z--$}u< zI!fOi<7j5g52}*r#((FxaXHKuB>0ruJ)Xaa;^HXOUV{mf2CsR(}VESd>c*MXAM3! z6bx>2GoaT9c&u;*`@=CPSE2{&dd1kEUp5FF9x|+Bf;jGxibb#gf+2RdIP3E>mRx&# zlzt4d~XfilLENe4@wa!_LyB+alUhBz-{CW#BxXY!UsTW9T&q&l|( zh};VSc`S;=;jBJAWn2myb)@LzUNtD<{12l|8l-3RBnfPbCJsS!tOv9rVa7C7l>Q}y z&)ltW&PplRnZJtdI28?!yETy6>B>Ab*$VYvV{yXaBJ5n01M^yZeypX*yJAIqnM3GbK=uD`al{sOOR)p_>HxD!;+#Q96swUKFWJ7GcdTHKsk zg8F+bVOg>s)tz#LQ3y_}Rxnh?sOC#xAi15nA!r6$Uv*sPZB8Up$I!#6y7VcnW}7S4 z5b0N;PbXI@J#HEvni+==Jdi^lF=$9?q@_i7-Y@+y@W6K2HDTn7r zbMWOkng{9cmZP8frfm)k66y5`AKJB4_nYd;-%v!&>L6&;w7 zpMoZNhw<7jEpV?sPFvn=q2=L@bbO=;J*D`Vh-$nh1rgEA1QiiH^Y;N$y)7E$9&nzq zP!Y&iwnnKPT%P)4n6+fT9R2WO7HD%_gQp~))C}yV&U-@Xgrh5A5tkhgTdRiM@~Ld$ zjRiPQwU2mPEAuN3O7kB@i|{?g=aVbT4XFp`D?XDlmtL9KL2peF=SxL9kkv=V(1Q;} z@Uxa9gz8QZXr<OCNj-Cju^=M(Xx0? z_RUZ>7Cc-4dPgPL$xffy5AB!fhimb4`h)E>Bvl?l-wIe&o$@MrTSs8f;f4m&lgNp5 zReqzC2;KIsmlS+Apj-3j(%y?Xq(^)z^My5`D=$7G^CyVFLlTXCL51Z0%? zQRDJpvv9xK8hS`#3|Tn)Al>yrm+CJYW_KUohyVU;!SoVsoVe>eQNM7M9qtpw?KwXf zpEhZ{w9gI}9^6h2h*$9DF)yuWf7;2Kn$)t;=7^0^TtBGj0i-&q!gk(fD7JNj-qtT{ zR#O3{YrEpKcS)r9#!qI~t0-DHx}Qz8v1fvnKQIa3Ub4H1KAD;x$uSe(F;(xn;rApP z+%o$$==YZ~WAeR$L-5ZqR_*ZZ0C8CoG8dR6P=31iZlm$MI_98>-qT z&KK;wMs?j!VZ@;>Qg>r3T{K;XUs(Q^1nP#9&!>w?)4U^8YvvOw)-Hl&ziZe>iSy~N z#9CSzJ(vE9x=q?EPvXVKg(UI1FHKchO@17)q7#R%vgL*oV8)dt6dpQJgYhn4xYZN| zz5iiC)^1W|G8v5LFDLgJ6kzspE$m*d#aBHYM%&#Z>8hM?`XFDQn<%oW&cr9A$9NmY z(VEp)=1GlVNSlD~hPsBOzeYWJZM+NO;{-79~z zdU=o4+H#u8y?IH~dc3eX>jIfNsgv}YWzspzTxh!X1F~~f7MOHqTKNZ)^%lC04pRBI1T`?YX`HgGzaLgoH)-(m`_dJH1dj51*ppNibW;GEc zR8rA?B)!EO>GFerNzt-7xYcq$>bYp3LSQTne`QHS_FSNapYD>Ocz3FwD8Sig6B)&v zU)IB8CldLWK$Ev7(lezMbj&L`8h?9JwUKHB?wh`Zy!42L&zDB5;^7=^8h@T%aa7@- z<@}M$X)>9adxwaqCK3(L^RQ13nfM1zctT+{Xa;KI>n&;I&xdH-6+9jOnGP^cdv#H6 zT@@zQYSWA1>P&o40 z<$WOkco+HLw1K^A(@sL2mO%6Q01T}+09%VEbct#{j0`>`T_xFcZ?P>&mscVAR20P` z9m(|%BIs#YN20ss;raF<=Dl%Iwdi+-m-wX^CY?M&f6P0IM@A~)(BnFo*4@sQt;OaU&+#`t*^r3SE9b7&27inYrn)V5q7Hq;nFSx>G<$ z-t31EE?Pf%(=#%Y!`TX){Q&9q%kBtCm0+5Mbv_x-0~Nbz1qX7&Vu&YKpFZTOH$ zym(&F8Pp3wv16Ixfvd34FNleoF}2h$A}3}Zrsa)1xZ7F4w5RomV z9U_VMJkOh=XrM%-GAf09OHR|rF4lfcrqeS@sX$?f%qE9P!_^<;tII4p!ncfk{!+W-8p=*6M zH0%z5Lr)gt;|(kx9TY4}l8{j>9v)!&C@nnp-RKYzp z^5k>0Av9`gJvHvyN;O=gncjCNpefFqM63$pXR!-t<=*8W>RpQcUDo(5(Zb&7jtdDf zOk_;bO3?QspPKrxAD>+R01M<7;_Z4p)Hf`I4Yv*Pzu$>m?bYujI!h>g_|lb3mi!C9 zYN}wDOA6EHb&Grd&mQ*r6+^gX82s??CSvw`$*Btsi)bpHiQK%0RG$4t{+?5#rEST~oiP?r zHpUB_+-~FI8&7$LFyHahkRWG{d}E}Hq+z?2Ay-s&oUF9H1FPh(gIoF<{Jds5ntxHk z_hOwOHZ2CU^!HKSMMb3PQY$G8QKm7)PDC#J6LZGn0TYn!%2e6fld42hvSG0n72o=o z%sg8x3{=#iEh`ea{2SqP`k!w^OnWkSaWBsVx_+3_XmQN=$}X+$TE$fSQpCXp%6oT%aDTg1YD z8F^u0Ev%U9ELgrtO88givct7_h8zu*C3!_QyQ_4QyjYvbBu}_TZtdL0yk7X3KkLX*o9DU} zrqi~<`?qBn=6DFJhCEO#-55*rHSnnRVjP)%nFwat;Aq-_?!KAuA!$F9`X<7sBUiXd zj#q`RS7@NAgdNqbK11E{5`8kR2tPb}jEyc9xG~bif%R#HgV*1K*8K|XlX?&5$EIL% z^$&>foJmwN=s z{vScB0oOY?z7h?bbX8>=MYTOYrltx5A40 zU3l0)gmW&ofb9GjxNuDw#~6g;M9DIwEj&Wu^dCO;ln6?R)l@p<5S7bG7aA^<5e0uUb-0fKd zH@I{r-@=$&&NCu64=a!^E`bR>Bf$2rZMb|;1rPh5hvY86#HFV&-s2KI{l<`cDgTFV z^Vvkb6h@KO4;7g2b%NY|_kb2U9-#+XkCF8Anta!3793E`B^RC>5JyjS7;<2_hiU~} z_=}OmCHoshZTt!)&0}ER_gT~;-jOC;olYx;n@QF0`+To3WNpYjIj;)!^f=_ZN#k?FpTfG$FPQvam4aB-f$DXoGT!;A zTzsuANjAJj*XxA3fx|7vN= zef;+;5ld|(*1yNo59jU@ZSQnA^kax2Ds z=f_#m;?^NzS7!+xkCqYdtIG6?`5SKG{T<}}s|2c4-ass3qp8o;5!6cJ3G;gQX<|FM zl*S64fm_Wz3_sI?-DdpvLBv?JyS<1=ccs%0d^T@�z?E#9NZP^aN?R6HZoc+D;wx zb4Y)y1$|I`mu|7r5dByDi>%XME?fa7RPNJ05Oq09k9+OWB zE4aB0;%rp$VqAT#oAeHCpxt*aPz4=Ny7B)k4T-sQ>)U%Y!dwL+UX;)&XJqLSGYQnN zR>1XM{pfE?aZ>ngocwSSis#*-mtXnNe{q*-a*HnYOop_;%fo4R);)2<}cHcf_vn{9jWNV4_8T+w*=8(Qoo6$qr2p zorjy@Z$bRcV`4652o8TEVd+7no84^Co8P(koK&Sp45d(gcQm~4%HrwM6uzYG7nH3By! zq@%3&GyFBR8^0XfMIOf3po|oS_tHPn;lfi~_5L9yw1tx8bTN@!)5fV^7SKo2UXxCL zSMtdF1**qOvUKNatUB}NdU;tcS9xzUan|3APi!v0OvPL{5o^tDWbZJ#+)8p-1w{`X zB}JJH9(06hBRMggOvE~coJ*B6e=nkBc>P7LTjwPFjh=>0t^Z(1Ll6B^r(u@dF4U|w z$31CRNpLLT?8ox&J=b@TYn%)T8tJGsE&xBJTA`kO6P}$fi~lJ|^BLj2$i>^^ThVvs zZeTMc+GfySD&|;ip)7j)v7NTvFr-OXP1ZXs#bSO}bED)8w`xrlQT^yc{>l3aJrk9w z%hfO%v(5%xx1T1r1`>(dmqF-AbO2351t=~gFzx#^EL4i2IX~k_uR;s6X~|8{4QfK` zw0kgBZ2;0&=^$$FM9Is$(ZA{%NSv8~+YTIq=F1Ow_go(Q*cS^nrp|bC=Qo(UzZ=&{ zDTC#Bar)X~20BINfzR1xpmli^FIXmLYc{|{0&?(TH zoK|t5A)MssouYc#ilROfN&0fg2!4OPF3by_LVd%Kp5}A%Qga)H&s_YOPp0dH@6LwO zb;dh{nrR8pb7%^kF~f>#TXMu?x<1!jaR^G@`+6GZaOW}-!UH|ee`EoA8N8CsMg zP7B+6NpDgz(M$G$J^TIX&)Tar^=t%fafS#tL-|K>h9 zWC=E&-odyuuEw!b{~w!g}# z+s|n_fZIpLII+>8p)-^FlrK&<4FxeTj0-`rV+>x2n?!fVtDwdtRuEd*$%)S`0VU`Z z#5|Ru_4ZlZo41k7)A?^0J-cbRa%3aS`8uEcJ?~0YJM`!yiTxCU6lrEl81!7JBbt4b z*-vb_WoP|3m6t*?>1j88SrANj+_^%Y*N2gU2z$Y#x$Bvq&8O&f#Vyp$cLUbuXoB2n zo?%cr#H@%-hJiH?U%%=|;vb=7d{HL5uIUlws*a+*2%G=-RI zc+o{K>uFKQK5qHVL*&Y~e$JyTh7KIgpfh)VBJF%0sP6Ye?#KKh_^8>!Y_Awk%CDC2 zt`bXnLi#?*{b)&iUaR3E^@ni${svIreiDP;DWR8g6z{A|!QC%bVW2iba!o!-_@!n`5env@w!M_d zjM2o}9Xa^DBOW6QMewKbKTdV9Q1~yBlEHKQ_rI?RIW$$1DC^nMWF2Gtc4ItB2Akpn zWe@Q7y~+2Z_tQVKqR3j2Dygf=;RmcQi1$M?cr@i7BXNHTe34iTuCLCcO~`isOl(gJ zM)IgYr=w&>>n0*G>I$mS@9u9*EznEmR8!z-P3GC1f# zZ!In3?tZ<`1#Jdwx*#o*Dn3IWo|2~(=RXtw-gc(R8uPSC+T>__xA7m^D(SGdKWCABZd>Soceg=@mLhkRAD??HS`mwq#n&53my1sWZoVs+Bq_JOxp}x_~`y4A6XnMyyt~v#i zwLO{fPwo=6=ozrKY$R4}suAq(Zy|w0;ateyYxJ#F1hwgGCiXjn;Npo8SRQl?4$rRP z4y2VcofdTt8@LEe+t7_yA0=UL{6<{ZI0O~^*~WU89Os|*k1_o)5{z!{!)A`pwLg|& zZ+-X=!A1Bi=a3K=QV=I$&!%Z01hFg=Uc6WO9L zL0te3t^^hMgW^IyI10DSpAqY)cJ>$4#kuMz!UQRGe0}26>T{jj2loPA$J#Y__E)HStqVqrXWHH@I8 zOgwQ={7&{}og#(%-V?L$hPY|kdt(3n1csU*wh%*HVKtR;Yn)AE>-W<=uNvvRG%?ZU z-QptYng5{Wq6D%jn&i9%i|k!(WKU$0@?>dR#l+Ahg^y{jZ7FRU8bv+57gBEBMshn) z0@D5)NlkaX=2wjiLkAVd#I5h>$3a-PLH&Z#?S>|MoY7dz`N}Y~* zsU_Sb6V46Gd2rs_7NT2l2}3jvaE_JR$mNcFcysR%{0IEQfwWKD%6x8s8)hKJ{3NzPteduSf}(tiP4;+)N$ zuil4$_tfE8GEno`29vc0VD_*K(R^-Cu+f11F~gNTcWXKO?1L@V_#OuB zLEd#WFC7mMRW|5Y2pjv;AKL8}l7(fP1Pf|&VMT)>-iTYpR5d8D$sM{ZNUO7@J?hN) z;p2Gvw+cEeQGruS2O(?CB4+%}cu-xk2YuGvMyt>v*zxWxDKI=N^t+?YjZND{S1;k& zY3_A+OjnE@-93O;Rs~?$hSOjRDX`vS7OUEM6Nfw6gswcB?75i$3r-)v-S--BX>B(c zH1V_4CC=Qm8aW*8IUefryWogL6MCLeVZFNFV=4KI>A$}7&-)BKtMCw33yzcQtXxuT zKOX)%Si;&&YclbKE^DMThg~Fn4XxJspv74qflQ$WlH#2>aZNFdeISm%ORe!pwiOO< zd4mT}$Kmz*t5`OuJ_}8_}&~1w@UhO# z)jtV(mNF=}rw95zD4~309T|DxqfqzJWoESe5j-_tkNFVoNw(QJkT-W<2(y``THncGl!rE5x@Z^iyU39U-+VfCw<@h{`3&RoW}&n{!rbOb*d*x#di$iP zg3M+(FzFy;s{V-xGbi9GnPhZ)6^ZLxEZO;SS@g-?Zd&p_o3V0|!2zqec>U~dbiR;@ z(laNr1wS&m;RtcK9yXch)QGW-!+)5p85PXi6(Oj3whrdLQ%8jXPxkENactF%wYbgl zC+ysCgZ9~{Qx)`lBTyo0DUKRz z0G#pPtl~8zyrZ~~e{S{(9-S;8-?V%1@SKyvfW^Nta@sFsD<0ry7i0GR zK^^v=;u!W+xDu-(9n4zaj6~0BzKgtXJXRkbg#%AtqV}uX$nodNLt92L*_;3cA4*}> zyJ^I7_hkGrd=|U^DX}YG_n-=6%odi>}uA5%{spw8G4a)8B4SBv*P)#O*O2( z9EPs`vg}aXS)8mR%Yu6sPF;`W#m|g^t)*OaXd<2 zcm?k}{9&qfAx$5e%xAbK;^~AlC@c)ZZW9%vWelKVAfAODUm0@LGmWOy)+b_MUDcAlQ(TC))>O0k0D z&yWt!)TF301*S~vuv*;?-7R{g^>^~ z62s9|QcRnDC$ql27DgtFB4>Hl+<#v@h|%+P%w8!Q`e>g&ty*=CIFE0H!%kaZs%kK) zIp#|+VI5VjH)U>??Bx9FJh=G7XCPXAGCig*Mcw<%sO1f1Zdg}|>R#z3nd<*YdT=6h zWyv1MKD2~P-*%gH8Hj*ABt|knucCtM9nkc66B*$B2G{hA&_p&C7AiffNShN5cBdC` z4hFk~ujbmI%bKmsQ-{%H18Tqw17FfJypOhp>;Tg;Nl5EKzDH9c!=ivJZ5jcsBCa-(VQQBrS`!a;$+?*v?|3eHbd%Mt= zze9Pvn~QLw5Us2uQL9axr4?(~4S8eP`F1HN1;} z@~J$5>RxrkFHcfnVp|se^O%l{+LQR7)W)rkM-bqguQ;I?9I%3+$KvqY?*r>rhfMq{MNdRdz0l@ zg?nviXz>>sj=Gc1`FBy_$|aOyw5YzNJnMZf6T5HRBGF?;u(oHSaYd9G(MZxICMLg# zdC@r9?T|=5jjzQmvu{E1T3h-^KZ*Wau#3j?7#*`LL;NXTh7Uh|z*FZ}z{nMSpf}|W zYVq9Cjji+Xn3t@;a#t399iNT!1|`{{RDE{V*cI$j#RZT_)QJ4i0!D4C61o4PAGYv$ zB;T0J@G&X`#iw6Dn}c)FU%>(=o?8nxQOdM;oFXl5YvIrK{Sd(a{|D6lAcOUdIO_UY zzQ0FNA~_93Dl9s#et;8RNU_h1Z{xJ-eAsC8Xm*lNm9=V|!Vbnh!MylZO#Iq_BQ)(% zcIgr9z)%0}PInwY|nLKR1 zOGfR?sL=Ow6Z8|H*QH0({Hm4I)M$k0LLGw_6pmoK(KN~=-hnw+24Lxq4o;tUz8~>F zNUVQsq|g7=z=m3Pfv!yxeS4^$rmb2{e$`5n_p2mD(?vJv6r)RYWP<^*n(YB5JG|%z z-4j$dD~f8VMbKD%KF~SGkos(m7d8$GVNc<4Cb+|j%yWz(_Zkn-6|3Hp(6)} zr+a@7QxicuvtRBA)NH@Wd})>f)2c|})+Hu{84BZE^m4iMJZso5CyPze10;~hihetd zgu^{YN%lM=?p@qJNY%SVHZ0jlAA0#g@grF>y~GQH(^KfR_FlStBG2gBY>wjZ*P@(9 zG;`_tS*Y}lAv<2oro|6b>8vS+khN3|l%rb7wk~a=#ViA-x@c-wA}>;!EF+Q_JitA7 z72^^n_HkfZKsT-DT@LkO)a%(4IHj*oPhHWcXUFeliyqv;y0t55%Ij7xcf=xQshJ^6 zdFjk|k9RZPyem^8^)DVW$-*ypZ(!V{*=%eY-$RsDrWp>`L08!r?F}bj*BdpMU3!7o zs+MpE$DH80=Vrm4^lyxPn+#gjXF;HD4>SK#CU{qDAaBl3!9)#7*0^;tJ8ZZIM}AG? zv}ex5!`mbAVu}~-nmfQvInY27=3m0nKj9eUwjY=M%SKf$9y=T^>D~$bvyTvN_8j)3K{UQ;$%iS2fXM&1lswz>o6nG12>Lzc;r6U@=D=JR z{E~DYI*wW5I=KvDe98tiI#prKj5u;zqmwyCr{d-L{7&P=Ei8O2g~nSf`R?>#O!{h! z2FayxxqJpF@6ChATgL2hM=5qCKQD4Tr_Mgq6Y+j>E1bPv3+G)@#FoSRQ91Ayylv+F zqg~RNKX(wP3yz^@mn-gB>rKWsqzbInE<=9TO+3;biSNdy;j@U@>}g+JcKgtAY>E>{ z)4L1t-CRG|Ga?ZBmR%!d{GQ2sq)tV^enY`1KWSX3{}pHY%Ckk2PvP=&;@pUZyGg)b|kmi-YVZJ_z;c}K5p{%(DEE_6i?qA$Z`fR))Iz3;|_uwmY#ajxG zht8&|%r(Kbbs;p=uZ_%bA4|i^$8o>zyn;{0&h&h;0!jO$KvgYgk)VtkZgcY^@_vLa z#_ZDIz3>f0salb$=;=~Yd54<5ZlMqVjuS1Bp>W6L4SC_w3Y`lE$a;SM!k&;L7xa|L zj$Ynbm-2$j%RXa1(Ypi{>WFE(H=VJ?QkeO87y0+x((ae`6mm(llk;hvNENnf)9B#| z%;DD;>Dr!H+W8}bD%p1lx^h%#uw@kyW}RYGUYdj0)p9{!VIchJe+3CU!WogIJt^-w zMSKd!!L#2$Vj{xn45v0~+xV7xCFD`NehCUb--*$GOZd@lJ|jlz9CwDx)i=eQs);#JbQXEF+W?%t?$Sp+e6Nf($aHq=7c?pL~}rM z)q7C77y;#R*T`~78Om+Yr;Mi~v9~EEy7NyFP34*7t;-}BGtrv)`{oy1m?8zWAID*a zvkLL9xj}Z66o9@kl#?`5<`Ncn3YOrJ#R#26PZ9Uwrsf4J;diPk?w|0KxChL{*}bWx`pQ+< zo~+E&b73TM)iuD>om_4AOZaoZlo6I`AoMuE71wpND>jAh+GI+z_e+WdG5-;dNfBg% zRWos!El&rW4p67WV!B+Ro2pN{PCxJS5~);IlGx^2-kUIqwCs`hpXW_dbG#q}NfoykfX@^%$e}d>!34MTRCdkEWxZZlg=bDv2r$KGD})K7DcgJI5Bp z&~5gbqP?be{8{@l&D&l`JwEdD#gsyH^lk@V@jG8~(grUy9Ad`8mTLdZA$ zEhKuK92e97lB}-&$!(}PLr(5rK*Opf>CQ3k==~kAp-+xpe1DQm_cfSi z@-C+DxfK>a$Q2%{F=p!Z?lHzT7ihq*Ub<64USuwnPhW>g(N{Kx#G748t+q(eaib?v z-|E@)<(vw7RY6j;%z8D|^HZb7j}z!3zmarfOoBsexIWF{^H2H(H5j+v6H3hA33f-d zkY(QygUnlj^4$cnf~k0C&jVsE7e`J8K7h-Wx4=E;JO-RkN9jp-1iB^!KTEX&)ANtc zQxs!LLW=PC)N-=WI-05abd*yun}s>sOVDoZM!ZsRogOOZy~!#$r01v(^SeG6!;Zb? zx-IOfW}pBa)OqL42tJ#>A`Z7_NaL{FehjhHVKdeFti9VadMZL%6jxnF>jTHpjm3Hn z&X0q*^bIK_p?@OXakYtr=v^ebxOQ_cAaI`psSZAzBoX$jHXw6aFz0?$| z#LwdN9qRZ?BL@9!7JzGQZ@FOdSqOTofH!?pK$LtBu9~;t@5z_&*pvqREGG*Gf6POf zdSzUcbPq!V&*EA$WBeKRg*uJ7Mbq<6(3q!YqTpgV(Z+Ab=%dM1oa&1|oYDg^{7>Q@ z4*Yn46X$n>>xw+gZrcEU|DAw7s`j-KHu|eo_LrwIz;=5~jkWb$ax?^$7GP+GOe{ zU$m+|i&4wl$i~z_VPDfFj0~%xse8YY>N~Ntpr{E9TE?Rp`l8P5Z2Y}i1esAa@JB6@ zIe9>hsh5=}5(h7GT9sPtZr;E2WBqxa%g(_GBG!wDq*sd7j43A5~g6f?r(g@f`` zJJ>i!i{YKKFz*+Vyu<|Nz=4m9l=e+nd(RWT>@~-@wGnvyLKHHZZ*gVRGraT6jq$2{ z1y0}F3H<31Xs=5Y1_fT^o>^=GGO`@bOHaia;NDK8j#bOG7)B?RxiI)az$VsYHJ zN5t#IEHYK?2pN2Enpw&3+@9a}CYQzYD{ONP!l7UnO!+Ctt&lJwvepJLXL>yR+nL17 ziM1mBTPm2(17Y~&))-d2r%#aIV9&_ph?6}5DR`mB6ZHcDC+{9fRsXxhRnGaw^qP}l0}AmI%5wqCVN#{x^zlcvjMbe|zFO}28Ijqe~I+m2_)s(B{m;B^d2JiRm{Clf=Y(5D*Fi=-0=GDw#?)MK>|Sh19uJD)6wikY z7#f4@ijnl;&@I}q`69QqX%+YO_7m{Glt#9+n{kFm{J2eLY;egdM|9g)3(Or!$eYy) z3X^(-r;crd#o>$L)KPtipSTF8>t{gpvT;0bbORJQEg&JDQ=x9;bny9<0zTX&zFQkj zY-Lu0)#^&_`*lM+ze*nue7Vka57c2VpMS1@KN}^Fn__y_RJ>|)%e&j^>dhn`D}`(cR%+_EV+^a>>lgF(9*O34~3Lgj2tyVdsoH z;MFk?jDD?wVT&j#xK~G`rPh!DJr@kscmgv%+~Ewn_?=qj0291comL566Fa>#bi8H+ z6r=`X%`RP>W+cwm@V*n{J9GHlW*XeTR0XFs{sFVni?MyRh7?@1g>}WRaRs{%kmn&5 zy_iGXHp`Ox2|bLg`(McK+>J3;0VzIk>uIFU9j2f3s=2QoqLt051adsky@kY&{%nl>S*p| zcQwdx>*79=|FZjGX7pNn(WFU?j%FSSpEyORAfqp6T&2lnJoMuxtTN-Qq7(4`N^60O z!9fSxmie&Xych3ut;GJrV_Eqpc6e|?1g#2siesFD@DI6)ySPyNV|z+C$2|{XZCHG} zutlI9+$+#{b{j$<3pIkr;Thd}Ot`9x&10LGQ@3VPzmlD>f&K^jtAa6T%TD|!*3Ma9 zoCwX0S?%h!-|amCogNs2`<^UVF~S5zyY-or zA!}jemPF3KKt*T)j*PqFPMG?{5Eu0KLy*s8vZld^4QsJseRl@1X|h%hPL&&}D0?I& zHde?F zvG%Oz6MKC6WDr8EG?7+nFk)jH;Ubm6E5Fo0C_6^Dutf@vpEKpW>Na6!tpqme+!GXi z9|f&m)sUh)hx=Ech=oVu@b=p-{Nt>{3Wo3D0r|rq$52|wxUrYaHB5rlrfT@&$|i^niN+h5 zGVJR&y12437uBXpGm5EwCdb07_({)8d%+h>cywPWWy~C+7*s9*5k0`ss`KX z>A+gdO=lFIQD)oT*Y<16dK?Pu-#ENZH-yMBrr>qx8b)rBU@zwOL+^7Fp~vA_!liD- z;QM?cg!_dO>+ec5e9soxIQJ)P%2F#RrD}9p!?Y?qY17G5-E`9uyP* zRQzo_PXya}{#tq)47teBP9-OZb*#cGdV7$kfgUR4{|X&)99!>6L}ucAxb9y zV8#xO;l@5oBp#~%)Iz(1s-*v*;+82ic9poOeVMlCug6UqYM>|zbkh~h+|tNtHP>_V zJ9;_0W#V+d+$1`?UqDAT?V=9ceCjq!o0fO!3#4^Agx}T*sMXx7R6W9&jykxFuCR{= zTlaUwyR4IZ&`>5}7H0HqX*^gOl#+PE7);c^idzF$lcO6RI^2*~hpnfc!6U6i{E=P{ zMsDu}VltsHHewQ=J1Q3TDc^^i9Zj$xVlw?#bCc%J?;&k{JY!R4Kb;x#iw+i9P|_bq z&rZ8QbM2*Q>(u+yz%P@&=Rv(QM-Osut9Oxp_XouKYy}xDQAy^1eZ<|FX@@yc*U%;8 zcEy2fogk55j{cdALS{`J4E<*V(`^H3=$H)tPMbxl>#swgaVQKf8wRf zI1W8Z!{_?m7*YasMeZ1?1og1wQ8j(=b2jZXR=~xzEL|s=4bgQW zQ2XREeP`QB>+D}M!~br>$ZtK&)7#rYZO2J0bvDN3o*DS9G*n=9>?L@-_yzB8kEEqA zm&A%G;ns{nf*(V1+}AfarXalHms2zZsej}%Y2Uy$;3bLqy&KbJya$st?x?)xFQV;X zdeK*r;skZ{t5<@7)x7`2LW1p4Ea5Kx1n8}lX679^#5+vuATC>mxfdn}l9Fn0Ry+#c z^|;f+9!Kc<7iBPzZNR&4OCU8&6SHt}5bP-}d# z;~ZKZ`UA}pr8ptD2*)gZ!flb%<6i7Cg8Yp4D3?738#q_N7%rWfNEMTo7)4xg{5Q`< zXhCDWEx03NKF^4Z149oZbbk1keEVcd7q};KP6yA_F@dzjiNuG}?=zUNn7b z=0X-ekYaQi!nn3t5oZ?q8=CA&p!sJOa*+ ztfZ=ry9M`-mP7lAXjox1hh9nH_wkP$>DQnTND7G&nooZw2nxvHl}?jDucrfQE0>VK zkI!l7&KOAl&l*oxF}O{y9hIYENk&)`nY*)y1PdkDqlwyV^t%e&BN2mn(lHn=38`=*q_yO+&UR55unKBo~jkB z)iq!akD7%(IqPuYm(`?{|9g8V5c}%A#BWWEpyY!B*ZfB)^#A>q(aPH++|2so z(#1M7Ve(jZxpNXewQpmx&$ux(X98-gyn~a)hq(J<_1y3O^104?3|)2O0{yk6kZzqe zM2;QGhewi9qFoy$`0l+NnWd9Ym;RmAGy0^JGo~&7ZjG+lh*o$^zO>>H2j?+eK~#-5;(A_mUSuQ^6_p zyd#G@U0>6p+`~98_5e!C%w&Dt-*8@E(n)Yf6u-+c#=+$=7<gt} z(M2Ixas3FMiM~dbnv^rj&4-whnj8|%pC|76R>QSHRng->Q&RJ32NevIlcx{u`wKiz8Z}eoYN7&7~*rx4~sj3H>s=v3poZ z{G9ec)UBJ0zar9ws!C{GcLKKAHDk-9+xW0n2iFTEMXI};sKS<&WM7&zCsCuto!J~k z?j-23yBoh?%isX)O*RvljMR0wF=jnEoF9O}Gn*Vn6kbPJ86E75x59l66EWK?3UopO z_}`r}S@Cs{yjCivH^Lg|I_2NOTZ>cSc=2f5C!T;?)(Y{i2|u@}*g!9wpH6LZ=h7b3 zW&+;brLQjWey|(U5x7jMYoH-=RUSC7zrxalTlCn8Zg0UL}q@Ox5crrBvM^&%XtkZz)`0X#oO_Zd;&)JjLYA0hik_*3utNUC$N z3wN%zVb?3&#^y8~`oYAGTJK%UcMj%YocL5c_H!jYA7~AqcQLr3JCxkN{{TXsr3n=~ zJ`&A4Ysk!w9{72{6wOQ0>5p=4k=ENzDjsLc;*=V!8gBxhYNNp|teLdk$f1tyNigVG z&wZ@ZqC@Wd_j;4@#2$TIn3RBzqITobQD5Nv zi_vU~%5ltf{>G$y^W^&8jU_I|Ui9;YTHKDRSSqoG?ogc!`cGHUNTpOdT(tl^)lV?4 z8%E)$%4?W5|2E$U972IoU*UG5;cr4>s*LKtOoX4CM1S#)u}AGPp4 zN7p==NRL(K(tGijxU9Y*aPg=hpM3hcv{Zj4J##CwY}sg9GT8$f-)|-1`F7}fFNbM2 zc}6VU^NDA+kh>~gOO$lYL@2z$KT9|1iGpyvr*Rzn12r+QD~{IBkQQ0|E26y@J?Mqj z<8+dW2`xXJO5IQJy{)aMNaZUj$e4Gx!gfC=Jn^*%WxszX*Q^xik@0r)*1=)=Jn$Ud zwro0gHfNC3dTZ0z`x)fV%u&dA&qjmMiY!fgfw?NR_%q+1zDVNz@6C#I#NcwApnH!- zrQ1=%&b9PN-Fdnvx`gf@dzOyc5(1y!Nb}AiamXlq%;B zg6}?AWc;`UdH{@YvcfE?kh+lXihW>AL5j(LdWDpeH~(0l$~KY>bn)R2Iwryx8a zhDIlgF^~3!lX~8z?bI#<^G)pFGd7nXQ+>U#{VF>V0w1h$U_G)_fq9 zt#&X=stM*g>5A-w6h(?-KT@B=%|fB#LGm{skJN77Ovi{G(B&Cfw5!Y)(rx}@Y?+75 zDUaRs*M)a5tIP;bP4E?&n=6Yn&dZ3ZPQ}o$v(Ld%W4L1crD(cA@ina|c}tCJRYdZ& zW%NS%WooQ?kR1K}o9jBRLRB7J0;W!j+44}G8P3*6*J4ea)%OUd_(qWXC2xt2XaXH4 zF^g6_I7xJV+$Xk?(iATGQQ1&wdRFjF@IThhJetbz?e|DBrOac5CNf2a=e~BOQb}_t ziY65rG-=XgCi6UGicFa*75llbogvboQqm*|rBG5Dr8v*;kMqxa)_T`E?>T#|{l{LG zweP+6dY=2fuIu~#e9NZUv!#bBh{)DX_HFZ0dU|p)>vXCRJDWxE?#FYe$E?Gi(pMP! zaW=}iWMNltFYaGgh!Yk>vKdaBtX7?@q;Dmr3)tpgn4MaM68q0nZ}u$f7nDpBnqQ%l z|5(Ajd7CjK{X2%J7@=uZJVc#fNo>U$5|t%O-;Qo3vNqEAe%OV2w@pQwah1UJ1r@#)$es=l*~>u<>lTs^+xk&VwV6Bgsl*bKaO z<`tfEUx`P5kH_;L;^_igT_{N8fUdk8qH)WFW;JKg>lfazo&hI_(AV8;(icb4x6~1C zuzql~vXz3KHa)iZ9KE0yfpL%YaQ5vA+OT#D%w2SYBp>1Y+F5?=rN~!Qr7VR#QxwKe zyEB0|_S-}rf2kBN@aPe$5fM$@_L#!{HJjj^#BXNZA#oIPa75qg4;*8mk$8_if+F*d z(5@p5tijFsFhxn6=#E>7Azu(F8%?(yi^0>1+Hj8ZInGY|LGru)u(8Wh=@zbzoAoV( z+?`y{lxt+O7xaucZf*-{SvnpHr))x(YHwOkdZ~#M(BYyO#&cH@EqEgZp-ein9kYq- z*-oa!QXM|*SxZc2?68`r7f02FmeEu#Uyv{EgdM-KAiyRD_TPQMxbx32?#grck&okG z?DlzVwDV!|&*>V7lzkw(f7wx+rHIMje&MU77Nm2@0lGYPD($#vMeDwlkdt%j$!gwp z++uHyyQWUS@0oXST~Hu?-D`n%`ny0T_9nUgxu4pcY-4ZD_)C`!Hn8&&e$%Z#8LZKt zg%^J&qN}ceQOH@w4s)y*&74rG0AVzqo`h+fcF7dhK%4GLk~Uou)E190zUxfjXxAq8 zN$WPE80HV9Y%$2og}?-9IgHpO!(^HoP+LDeY&`A`1LyhF^|lCHHvB*m#yQ}^pe!af zax#wO3E|6J0kdISJMH_t26N+@SQp<6a>R5hsa%joN7N@VY=ss2hL^zvPSYMRKaNQ4 zI*U=NgRJ_?cQjT@p2`G_u(PUtiOp#vCiu)Ga&{VzjM>fU#vUj_`HMuDlYNL(zxSj6 z>K9V~^V4`^>OYWg-G#K&?IzF3@(vxOwBJ*IGUmjk_)S; z%A-hFrLV}_vWpO{*A*mPYZ-}XPJ+fjHHb3Z4VoO|@xy#;-ra)%ynAA|nAQi6$#G^2 z?cxiVYX<5x%e9PLaXCr1{_tfi11j0hc!Vimy{VmYvLH%|?qGf_36} zp4lED-X+~*yi?kX!N(^T?#JCDMi(4;npP)y()V3?3ynEVM$!#fG~I|d|4c8eYtDyB z-nEcjxEwZpm*vIlEaX{)jVJ$Ro~NpQm*@hOe?&Sc7?^{myr)~N=~*R?aRi*^CAf!X z4mYEuhc!+;ZH~VBo6H;3H8DbAI^>9&z?ZdzYDta3T>U`$DD@E?Rro=EhYpe}Kf*!r zqc!L*yhuc`mS~@irGsfgbn};excTTWb|A}!gp5B<8tazBqidQ(veN)UB>kvTMy_SR z;}kmcB}?{+&OrYsrubU@4ZCQ|EiyA|KlwB22659bg2C7w@Sst^aX1@5n43AQUN%mg zF;`zN6eA?`pXcgjV?>37#KpN+|K*`5G)HLbK{pT2BRh<@Z}r}4Y+*6WVwSn7$>If! z(1!o%LrBPX-hbAxYgcv&_r^m1dH#26VE*eGOifJ8tY$A>VE#W_!~g!E|5-y@{q2hX zy#CMgzgvU(|G9=)CMH&k7tH#c*x^h&ijx~d z@WiADv~av*S*un{#hN6c-ts=XV4pI~KHWuJCk26Jnt(lxo1ygTRLac!NN*Q%dM~B| z4=5zz(ttfUv*|G&%6^C+U-B`QGt1zqV_b=U9dZ ztK#unng@n2{7pUGMbLBKLo}FGPw%9;!S1?^&^=xNv*W*T+N@z%tl16k@4tlWV-$Ff z*Dk=D9R*OA^%u$~WWg(b0O-srVJRzeDaS8h%(O0);Ofp-+ZGDG zX~+q7%k|?MPkTY&czMCiN?Yy&eM8fgnu6Bu*EqZ53bs4t;N0Q&^iJzsj`z-G4}SY@ z)wF*q$oN##X_M5+3gq_Y%VuDfa1|;xJ_I|{i=Z_B4A7HVpfxQAd_1$ERjH>taSsuFemi3X%NBt|uY4ONcvQ1c- zIIV0Tu>y5y+HDWEu2CSOR7_;b{29H;k)W2qWw$*qr<*3-#-`JT0%EyGpw@OwFlDAI zEwXVVzl{eN;Rki(QQHwPUYrO^28|#p^cv4Q4;haBU{iPs)@-~D zDt2X%YU>F}Ka^lcuo>L%P=QVBZjdK-rOfS9wN^^KW2nF5P1;T7Ve3A56t4J3TV9Lf zyCsJ(D^U!WDn6%6jn47oQuOF^PdS|MN(}XSdg;zcYa09D5kFWoqvB3r4T15J93R(` zRat-BO4U4z++N^Egr-Triaqm4!7)jAR#3=9HXUX+J}zJ+GcJ)8NoT?G*f5C@ae~uh z$8uV5RcM*Hj6h9+rEu&wW@Nt)?OSF;4y%l#>nq1$oApedyT{{QXEj|!x4dD1- zAvitbn6Ufb5{#{}7=6j*riaZV_QV0?R}PWmUQ@}E6d|}gV;dW>FOK>wiY9H!A?#b< z2x{?T6S~c@MZXu4c-Tn+wWck@CGIP*GU^|7a2%$YMtk{FKGxA=KSfY?%}Vw{p#yE} z=kN}Bmh@^ScWsosWmk+*$Hq^0D&j8tvN^_Lbaa*m%uJgFI{SGr`pKUZ9y>yQ_X|N& z_5tFx(StOuP9vK?io#kY891X7Lz>Sa5k^wJwiC03_Xz*eE3%7R9qFQg3 zyhca0L2R7TjL#1U2^M?&#y@&_sF3gx=Pr1ROCJeP zAt)T(Hl4$l{?Qm$`4$z=M&cW%ow&SD6=%V?Ha|ia)@p&zDyuUOBpC(!~$(zmF&E-trVA%z0-?J}+b%@{0 ziI~#lg{>aC_(M$w>vsy#3*W5St#isq%HaXVA;^{)QToD4*ZDH{ifrk}W0_X6{i#+- z>i#e>;~<1YWmxSnmwJ!=Qt)`%I;&JwZ zPX^X#AI6@~=W$NGAAVl87Q4PX;A{EwD#7ASMIOsWOP!xx>(W92M)ElVfy(9nuE;jJ9++vjn-I~s+sUG6S?P0EAC zZzCb*StvxVp8^WPTR1klI1GJP1||M{@^;BpD~TCWG%WEm)!Ai?ekTuMb(ICia4gx> z%ajUhyrAbUCDKH9Il9VvAKgi9Y2vX|I^|~qb=VY3LgW{-wfZr%x7nWrS_7V$yd3A< zS&FV|o3QCaD6T)_fF&j-*t#Gbr+p~LDU)V}qJRpk~sVTa8R?@YBF7$zW3RN(n z)ZyYws(JG!E#R-AwbNCs){i)oeQSQuA;ryT+r;G`e>#G_`_|*)Z(o^>x5luxsY=W_ zy-*@KJ;}0VX9_8k3gB2ZMPMGh4@QFTz)rqBJR7(GyL=nLtvn18Hl;w{pK-jP`~Tpv z^E>GLBE!>u`vv|CJ%qz~#Sj>?58mxbCxu4QoF3*D?UkCx3Jv?U__l zzt;DVGc99a!{7+%TzMOwRf+JX{d)?JJ-@<`tOze`T08ii{>SZK#UNXM7wnv$LQX#g z2PZq|znDp%r)f4OaW#&H=9<35Em6@60VqQn3dYCDP##t$&jY!ThPaEQih zZ^KN9eDrmkj%q8SaMvAAlhw?Wes-9fsol$Kq^ov& zFwVDgXuJGu9KBJ*Xk4(NcZ0n!KJqk;;$kT-z5B(^^n6Nn>&)m_ZwAYknBu><1Y{pQ zXJ52)%(|Z=bgq{KI$!=pMOLI@>MxFWGQ|}ixQ#M5jmALX;4@M`R|J0SRl&JB!BneJ zhX!gkQ^oTSsN~gIwCc@nqMj~?Z_{^T#Yt}eSma78aG2f}J%NS1`FL`gGDVA_xd2rtgXi>P9j)he+WlsBiau~;C*Wav=2RwQPVuJ#4(5T1+-hSy;>-CY6|&j z8Od78D1+3}^B`4INJdE>Y>eb`-2QF`oe%PGdA}Lwf31KKpJZa(Ge{cF{~!@hc@U@? z4s8)nNJtu2PdhY$swNBGvRPcsZV54_(va_>3X9D21;4wO2sGB33H%DKa?b zuHrl3^x;cnnQksQq{i`%LNakW%)%X<)>^$jjlSiFQfUPNnkM<8lz0VJK1s%Z)hp3; zRsmX7FT&R>_dnvd=!wCF^qvUESMF}$^kNB+a#4Y2qbJ7uEz$({^jhKfcL`p+tpT|+ z{1A#mvcU0h3Z&&|^K>-oV9!tn_^g#9N?QYQuKXyS?Ie#U)=k4FWnJ`mH}^Yw?4|c* z67hsgGtR9jL(c?f489YJ-)cu_pu7>e=pIBgZUg{*Bn3**pQ&{4BQ|GF41KL;1Bthj zpevielV}OXxa<@E_O(?+R4E*c40Pz!uM#8`WHDsefQ+5fM-y01Z_Qtb4z23=ebQ$- zZlWSan+DRKfn%r>#F5XdzA;6sYgwrrSzLdU2S3s!;phW%7^pi8TmQ{~^EUb9-BeL7 zA3}k6=*q$S_5kQgaDnbjS$Mqp7WwXTfm9lv1`Y0mSrk{Fhb zJu8A+Hx|+KVKeG{{xUnPSW0*44pVu%AM|UxFa%9XWcQ~wVjP=9}9QPSCjlN+VIpbkW^ltN$lPma`j09IB!`5 z@6RTHg=#Wb)YgF2jY>E(h637LhMO-wz{Rk?F#XOWcz3o8yyks`sHPP7qx%!Y7OV2U zchrO3p+9h%c>wEn*29|(d|2IE1^&M|j{mbu&>tNRRo^v0rdA4mO4O2!jO8A#`+6LY3MF)O?O9Js2|?9q$}M@#*&1YL!jR4gi~2`-|$y zOu~yvQ!zXMP^R-RYwNv@1Qi94F{~R=ue(B~3NxH;cq5l_TLg~3o473KFxbH{&sPns zhd<3(@H8h2Qp%&@U&ASAGYo>om!-hxg$o=qu!PMe&xp+&e==IRj{TBp#ZPLnCi6)c z1TXS~!Z#_*y88Jf_v$aJs=XVx34WlN^Z7caC?v zXC1B44x?(-!dT-NhfxX^D8ex}9ossXiRz9}pb`po&yt}njDejyE5Yu=WpL`B1M^6M!okSuT@n1b_QvIjTFuUxdfJ~~{fvGMg z(C%doIrnT}e_J}+8$K1Pe~UwR;dr=zKc3wGG@kAmr-q?_een3!nRrDIi`TzPVVYqS zMw$3v$T%OmJ@FLH>0g4n#S2lfG!fSdwxDu{HV(>GVcvj0F7bGcoAh3xhL!>)4?-ooV?&4bcpF4ujjC&+rZgEA8eUNSE5PB;kCE^dJFt`{IH zX)3V(C%{#$kIYW|M~b&SCA}n){AnR%k906|{m&czU@TJ zn87~RXC&r5m+iRIpG^4tkLWU1pryVP&e>al_x1Z^-<&TM$MpNCiUh~rDxF9#3s9hHF%TQ|Fdi&qvYECm; zJ@zHF?m32venM>HdTSyPKbbWt)h1rwxy-L4Q(&%z4ooyMBZdw<_?xx_q|0qU?a^+y z_iH;OCOmFU%$3N?WY4wb`X!PU+mWtNnj;#CWYgB+) z1Gmte>j}Y?c07_=jmvx8F+j!(p?EP};Ps2Nis-`AYq{i!tu&|O429Qz_u*r69;9(+ zbF;KG*n2Dk+KMuPf58n*89tGJ_{*v!qmEWh=p?g8>gk$QJ@je47hP>@NPF(;pwe75 z(rbK)mM(Oq53ZKe%-xjrx-}kjvRBdt`V!z=W(}V>`$fo3Rj^wi56)T6G_lN;x%Pak zrC<45Vz?%qbdO6XA{ zK6gE);o#Lo%ux2jbq85avuTT(nd4Ere;3ZY(1QQcxS1U@5r6M^O$WtWX}+f|t-kF~ zPyc&J70Zm+>;1jt)$Qku^ox%)D9;a#?N^~&i5N~Yuf>PwBC)07D{b>>r^lA);?LM; zwA>*C*Gwy+xBe`{P(vqF3>LvV_xbdOMi;$hEKYPDI5W22mQik7C*dn*aPy~GbR%C6 z%%y{&a$p7o)`XC6p_5^ARV!N~nr5jinqAGB0mlW*x5fVtMJw@Y52hh}k^GqIEj3v8*P~KrGHU;md7IHSM``I{_ zwHHHWgFs^PS`su*oQ&+OG@7+}G3aMC&IEjem)iN2=&H zpDa3iVgRnwQ9x%2SL|>Yq*Fy-(N7ht$jV(=bi#T=NWd)e*779W^N)vb`_IGrj{$J; zwFK{Hw;a#c)|{uh$&%MR;>;VrVhyjo;s@y2><8E9*WuZC0?$t*fnHoE(VKCD*o0nU z|20h@HPdzA=u2f3<1$_zFov{9V=;k?x2cgq1M{R}Dk`h~qY7$ItYWOU!)Z}2Q)}lH zxZ~ByF;ecp#@`j7r7!>w+sE>L*baeRygF~{%E`PZpCFuhE=;HhV=iL$$G`tPMh3;|ayDrb4lHa)Bhn@6@FjgNyvhlHO#V!`zx@G(8wA3&59VCXa57Og zh@)cnZPDypB#{}mMkmfAV6^5G-F(RbnK#k2cBKSf<<-*7KEgP(U=_*)XkzXPVJsh5 zjAK_-(7kT*7#47d?!U4HJrB(yJ5~N!UF&Ql!Dju8{V9JU^QDS)u^S6nk!>{4*P8m2 zhOlN+Uh&_3yF>Hl+MzI2#nW5L*-w8rv2S8>sOyt@a&E^YyfFPT`Q3N}`1>e~oS(pZ zF*t#@kkc+2YBWOJU4M@2IRid(_t5t)o#Y^~#&xFRm{O>PNevURQCbUkt{CMsFASEh zF~@*o)|fI+K&=Y4bGhd%TB)zWGFA^i_{Zb+w0D>j8;;_y3@{|B&jh($a@x@-Cpb*)^=%bqFW+ z>!V)zboAWlj}NCjq9+`8)5%ZdVfK@I{EJH$;fACBBneon_{aP4dqM`yUu8_bvQvq1 zRUdT@odM>5-Z7$Kp-{%&fXr!}#?EOJ?B|Q}dY~7eSc#|G(+~3|w84kF(!9CqjUYN( zl9x1Ihd0-%6BIgA;r8~gI6aw<{cpqs8^(rV|C}`_lPy90;&@~xUzgjnf0C;M+`0S{DlGo^irg|#0MnRmay%r0V{xRx z0)s@@rI}A+ywyOu-H{qKJ|Wq!_OniX8EosbAu{9gFxj=`1e-mtfVCBSPpfx^lIa2| z*x4fwrwf*WV|*~2_!kNbkn=4Qw=6id7=sfDqBSemuj6^#y>vc;y>wDw5~)gNm{9kUDRV;>+lHCvr!DCmbxpEk+I=pPbQS9_T1jR< z(SR|}qhbBbiC{E&BA3y0g7i&a##WrL#i5o(*mzn4|JH_Mu`xG${H%dfcKBiR9+|^oTp(9H55NcrGk}-O;#!~)I3UMp3DOulbtXl z%^kj-lZ4-+-Xy2*4gDrF5A)s}#y5t$@U7W>I`7p!lp1G_AG>Da5zD#gKWjdgnr@`~ zoJ{Dp2jTRdY%}ZrVGGv-E}?TKbrDI;I0Yp#Ig}EM2$-{9gIp4z)IKJ{IJiF$= zF>IqiY-c`M$7#F#{4bHXV_jAuX^yaWc^uqd|AX8erVwje3R!F7;MJOEaFFxRik~-u zzASU%^mRFvNuSH*+^T`zpdGlOK5WpifOduJL|^1P`7yMHgqeS5e&|V%nbWdp>X|(- z;QWB}?Nfx!?U#t;BSX@z)Jwc?_L6W}PJ?;kF7VxMg3mI4xU?q}mb)(H{4|3kc6&4_ zX!C`K7c{~4sseofF+e;Ey;$+laIUtC;-_e2QDeua`1xN09_bVk*y8|piD~1>1E%Pw zP>%hMfYNPTp8Ld0nAf-q^G|SD43e@a;j<98*}LOm+aen0ITlxJy2^HCX^`ZF{fZAbXRs|E&oPa^J05KeQ9&Wz}3gybhQxr=cYw11j|KN+GGJ?kKyYPf!2`at6fO>A1F*h;>zjoii zMRw`9du9WcV>YG{3(U*#!9b&A`phegOlg`82}-sgclj&aUiAokUVViLsvN_v`8+(2 z5$1&}i}Uur5a*4){|$R9hGAe^7TB|ip!rxFT1SlG!RzN#Xz4cY-ZVm|@`JG2Gy;#F zx`eCWKgSM>Av~NVDVUy7g*DxIc&z;{eh|Bie+-^ebN&`o{Z@%xPXah5zZBO?FJhhE zyFw3Z1YsQWgnx25tT=#lbHQUWtCq{_9CsJIXT`%i-ScqJO%|kdtI17%Ih+=n4x@i_ zU|-mC5MO&8oc=~Y&!0okwY~?IhYRr@R26}tWF4#x4uyYbl|amNh&&N3AgS>ow8SNY z-sbAltIs}R_4Yx0AJvRcT*e4C#%Ty#lr;onwKWAxFAsAW++u>N;V}Yldkew68Ipp+ zpEam6`8B@1_W{N1pQDh67h0~$r1LLkva?m^!S+iRA?9N(#P9tF67RHmj|L=pmKEAO zo~kMDhq@(i#kDcK_-(JC%P5Rvuw=lU+|{5Wbek-#y~*T1+{GNfs)BdMCt^?I5BwY} zElBV)6&P)`6x=)qf(Eapg09Lb0?SAZLCLOG%ybqOv<_{yZQor&m07h|F=hbY zM3ms5unYR9-@_gGwWus|1!uaw#^B`-QHYm}2j|G4>z++idm%$ak6S~@{yr@E7KjSR za$@BjXKH}joL^)he-%~R8b%T$Rj9MaJ^GOlIx95@jf(El@O=xA-z$gv zlOnLiLP%iq_#yV>aW|-fbGWh81!r9p!iHKsoR$!V^--#zdm;;8yq!){QqPd6O<~k# zPa!q5oQU&nHqnh;%OKm=oTPgwz@oJqAg`|gJd^dooEfE}zNcx8;WmEjt3DDQHAIUg zk5PB6G8FV>;A$-!{P*BFE;+Cq1M8OKoaP`dqgDl(ZbuYT>!x%kerjJ%4_ z1DotfIMp9Qau)_#hj6TDlOg3mmzpzOR} zj2^g*r#sJLvZ@QJ2v5T=wcdEcERMZ0*vb5i%VFpB7L&s-V~P2al@PVc6}lbmKz~XI zBp7Z7w@ah%G7A)2kARJTEX+^5z~yO1!Gf*V zz~i_8=0*doQHq55E3OdB#9PF-zllw=F0^gr6N4lfXx>{- z!cH9`*Cly)y|#|VpG>2(E05D}o|*J++APLiWF!0j%UPz+=O+W5TVRcJIBa?z4{{n+ z#72S3xLn;q-tBjS#a9$zs)ik;H0Z%+KY8jm{D8}Cyi2c@71CORZ}jXLPU|sSgle@v zrnjIKxv(b;C5jb zO4WCv%ea5IHnJLR~=TKP8L*YcoRvcmLs= zH~I4wEN9Sut*!LS9ZS-u-&eV+P6qXpGD%IP0WiUJ;3an-)_>zTDT6Ig^P?EX6??!P z&0wwvUIeB~oWVjDVEMRYxb#{Tu7{e!U)gQ&K}LxPQl>%S#K(-f=2kE{^peI5^wMyb z^>`+1DQ@Gxz_)9xalz}I7#dwf-JdVV#h+p@q&SZaXpf~g##++D6Q`jo=P3xjDUTbz zK4$Ab)?2As{UC!G88lB(2IZ6Q(}UIL(RPJ5?w?hG*JV3UvHd#E3^K-9R(sh6-Q`py z-WH#(aKa;R7GSvfJX8u@i#rFo+@Sy??2VN};oDC5`mPfCY^tkts*C1#4Nj(?B!wZW zdMPYjb{V3>MR+g5x}YP!7IF$-z_4E=RDTME21R9}?(>aKepgIeVs+7Pa5?H+)**u$ z)3LJG4)w<;LI19HvU%Pm(qj0K-|%Mzt(RhGN{TGA-=>SMo<9NY-#?}?o2PRc4mC_& zD8Mi6kMVm(I+k#KiM4|rDA=8Zb#bk@qO=^7>UQIr+4;EfZxW6)JK(#)19-N8!Gg|B z*cfh)W3@(Upxv@FBAIm+0`_I{;_p|kQYxgHQln{b*znsDTxM196kc|p=A982f zVbqA}$D$QqFu5xN7tEcD-?sj3abIdE9o>*jGcKNIBt$i-jD7~`3I9j0MKCP5jc3tt9&>c!IO?0; zOH0g$sPo_NG}JkU5!SFoZNta3EQb2_I>}KOJ=1s)e;&e$)lyBIdd9YoclN zkh#E7JjaDk8uHH)*ziBGchR!>}|&2tmNT8*;(HNoUT)MOAlrG%a# zNOQtAlETaq`q22MRrAzvNV4;S8ehs?C)-HvyYxz29x3b!!(g)- zaxlD+j+S$MPO~&xuaA_+X#;#<4Du_lvV~=#q~MYz6Ip%5s$=b4GGSaiasPM$TKsO4 zZqM~>B<~qn;;&5=9$i4=)vnm|<2$OJV)3zPHS!W}qO|x8+z>1*5amk>N<)6&uHuha zKGK6*9{xdP@h~(Ou0>|e37mD;88?-fu_a5ny0>Q%bET`Ee(g5I+Ocz)cd6sxinASP z$4!DXGb@NaSr5UlE#QmM4e-%xg6>5PP+TFzyRt=#x3FprFIq{C7jb<)&*;ooxE1yp z4n3FV1vpiK(Tj@^b=wq#Tc1Or!Zlcz^$Gg6$%A{2I3!DG((!#h&@IAg&WjV_(D*>| z`%@J0BjUI**a|-u=i)r`4oqwp7Q_l91?K!%d}dsL6Hn!Fy<(R0opsL5_VOkeC6>?|qgBUKSTjoI!XwN+Wi*jF*?=k_Lj-?tD+VmiptbwcDlFaNpA7;<2 zDB8Vf7C!grrBxza-*k>6X(;I-`nDZpOm`R@Idl_F`@aW;<-dq#^+Qm2D9qcoU>Hm` zK86+5w!FUeV|i=ly@b-?0a)Vyovk(2CmUU*XeT#&9phw758JJRj=9(2t9K$4?Tvv= z?VWJ?Koxvgqrg*KHJw+-?}jOtN5Fei0kpdLfyjG*s1`i}^Xmj~e`FUVDeJ>|c08oZ znSew>1M{|R9645-NNUYQ$lTH$G;*RLcMe*`I!4{K8qAEOi|P&W#j!G!xU&d-{v@J{ zog+$*iN^!lC7AQe61Wm)Noj0tS{7`EB!|M%@L{r>=(l zs!`%=?EyTS!{A(U65<@6ky^nOSgv;m&c|{2&+nC>jLS9IGs@$<-gQj*uqw@#eMG0- z9*<)+S28Oc-mnTv3GDMdUTndZU!-i(1=1I{mlYZ@U}EgP(6xMyEt2`g%56Ev^ji?Z z)<|~H75U|Kwa*+}taOUrTb>Fd11sQS;#i8ovan#t7nawl!-Ha5_?O!NKk^;n$DK9s zepVDD4w{0<#C)=QkvPbOjtSb_baBmBfze^E?ydCjj zxh(Q@Oeuei6E3ZgK>P7-^t84MZm>U%Ke@B!rsc7;eDtcN!j`Gj=S&qf(2Yl*wZ52d z=!#70N4&H#4_7u`Ld)Zc_;$6FK-o`DP%~9nu={2mUbvEtv8DOwwKpCo+$hE-zc9S> zCjqzFCgYmqE-KvMfbJ7Q@#wK#X!z{~jWCF$m9?wr+QX}fKs|yiiHPR%x~-uzW&xyq zD}xN~ta+&|3FaJG2Ax&M`BH6z^xFP$_-RNUV;iIJ>5Cv7H4qZ)7ypQ-)ZXJb7(tAcQlE1pNgB%aHX z6+D$|cD!BBR`E{hZs&#AS@0z8necipz6RyF70{Jq2e;&J(07I@5bg5*lfFA zY9%g(`3q}k_;y=-+j17Yb*gdY$9P<<(12d``8X=!fMOlzaRHl-A4|hfWnU{+O)bT% z?j~66BaiZ~FKG7bT=v=H82Us$l%Dz0UZc7$1Y-Il#%rgPUD)88YU zAG|-Byq9e2RC$LE|bS)1?|MFQ^qLF>0Zo|4zS@{ z#L4fkn}~10ZrGE{ClYzt)Nw{X&0Y{i$6B1Gm1|q*(?L0$ccB2CqwZ4Wq%ih%xFOwr zZzst*pH5xn#o%xMLo(KP7P&cI8^7Q~qBb8%)NU=X zrkPDNo@~H_1q{u4+YL7V-hsnIW!}SogV3G+0Ax03^IRL2@M7=3g{%K$c{9X3;p|Kc z@^I@QYwMsytd2~>@2YV$w5^()eXxhDj~1&`TCGoh|KWj4C&#}F7$&>dMUs`v4nUsA zGeEB{m|-c(lQ}8DQ=8cef9FL&(#U7XQ7wi^+??*$s#-WZSPd~&Au#;#C^(9yLv(E} z*jk1_O2iM+`S%bgT=wIbU)@BT^Xi_gyh|*u^Z9c3atM2OECe0nW^!&tbbz>0z4ZO~ z^lB+S`Jj)ZS-)xihg;xwA7T*WQh-(pPOKU^UB6@R{-B=}=sD>%?=ElB*U zBdBjw!qn|~WVMDAIaCqB?3^hEGTH(3lcOAYFtvoN)=!}~9dnqS>C;G#mO9y}K7l!> z`;T?G=?o2b%wb3PJh(aXh4l1twbRz0B;xN>Vrab_4n5=YU42-R>}tXFhMto1v-enq zo*Pd)Z$~p5ubyQu=0y^d10}?X=L+plf*|{DG^npX4^0#AfPVHv_&hTMmUmo)+_4q# z)Mq>RuT2KaOmDC;JPPuqTj9XDNVt=q2M2mNc1`GW=y=TKfu4K^4<6lx3E!_k(6)4V zcsdOzO)Hkt}=8fs$i+L+^VS!I*%H#?X zt|mgF$6Y6LB{(L@Id2G><^`{#LdjaE<#5(C5>6kT0_V;xg>Mt2;H+Z=lzp`UW=1?n zjyl53^#J-;cEXYVeVi9IhOA!~%qGrwNw54=X4V8rW0t}ux}Wpg*jwqMgROBmOK|xOYpX zkJcRSoDo8H{HU|i;qr2)ocTeD`OAQ`s)EKz0FQe*Al5HLroUB(!>7Lx>rG~08?pgB zwf~TyTy@seB9#f6uvg&7w-a>mE3xUmBbjgW3i=8pcqQxJgSUb>Z~M<|&O4_^X0J|w zS8E@TYqbv0+b#uFi&Q|TR0Xy9+wtJ~d>pHP3gAJB$PWhR^g=iF{onV zOt;QX#O6qjv*WBNa8;ZjkgPydI(i3(hB)3D5tK8C#1xrW>ft8@ zBcpSG=4tWD4_NS8dS~*|hqz3ek8ZqU70Y;2r9*i0Qr&rB&RV>pxN3-A^$e!{N{6pO zzaZ9RIW>B%kGnr=2t-d_#lCGk!Ni?v0)rSyf&1t%O7gO>)p(*{^5>_x&RI<0yWs++ z>jj{+qAFdSTu7suez4gCE}-qD!RVLPz=n|HF!zKP$PK+B`$Mu}+M8q;k#m40Vt#No zCYeU=yNl~5XrlI`Tc~1}i#X;UwSSUFL&1~=d=4eo-h3q3?n_!;9|M)`Vep;hvfPe- z1+R<@@QnHfq-rux%;+t&9FgR$E6ajj`AuZv_jQ~HiDQ2Y)uD{;8JrU>EVy>J8++3< z1Wk@If)6_L1tKG>1$KS&1>sprFty-1^E-PjIinx~#^XImoo@<>Z*3)(dflW&f@A03 z4gh9CGvg<9pS1aHBhSv~vHb&-HNE139zV2j#z`UEvca64c>4*xGWR7pwz-O|tY66Y zOP>Z`x!B^EjHx(qqLAu78KRl{p3;?-oABGEQTnImCBJLuEApPZU%WI62bD=fq(Zud zT#DjBowPXg9$E+!uP!5=34>(EiE*H}d^01SH=iU+Zl^_md#U0yTXYDVN_%eDaI-;S z+^%Sa`_(Ajb>kGeUoOPGHb-&4*(_YNOF-AJslcM7BD!rVWkX#R@HG8K`|~2u^NbJP z8uyr1DYnz0t;-OKT4?%ZBX;N2gXB=@K}rT~$=hx@=I03mHe*UQJ$vOGJum)}uFmE{ zOU0MsziEkR^EMNYuXe_;ryXoTLm>U&`J1|d5Gyw2C6(+r44!TmiO{|i@M5+sc-+V) zrLc&ewQOe1I6lbRRpaQZm3{0q7fbr3JD0yl)S5lD)14d{(Pji=-H|7`nwol#q1_8+ zqm(2e5?j{M6%J1rcl~eF;QlH)daskFi+4~mbCmF9mqD1#LfBj63x8&Ed)k!y&{AFk zKdSD+&QG_%TD=NvQnKJpFblITHh`_^O5kX9_%(S3|A)aW3{xd&GcW_AMUJ9eqa>=D zT%^96I_T4rapa}`Oz=FD0Ic_VHY9Zm)O&7#_U$L2>t`Z-+Mo%G4dNl%%o%cj=0E|u zz@l5mu#xMlr*3{m_T^3{O$B;%n%8UQe^GQE{#d0A|p}a+}DHB5JE#_GzcXPX-egHe*eMqoO53HeT~oO{rPJWN@j51 znfXCOs8Cu*9B(`DmZlVtFJjfC?}8l2mc1qw5w0M|Wyw3fQ-}K!df;|>9gMo32d!kT z?+r~Msl5u2u}}n-n4565NH>u@7X*Zk5?8KGUbS#0B=~_4|z4Eh&3{LM>_nr@p%805RLv5 z&>LtBYh3){*hwwe{e2dv?{$RP1y-;@J_4poEQQ022TZ#e2}y!D=((H@t{$mS@6Y9b zt?(sJttLPoo5Rf@KC%jCfizjL8*3&I`Xx3Uw@A-Gm)@Od;B*Evn@rGHUIpJ4E+?nI z$bf!O7JQ7(2mjeCAY zB`B_x0+n=MxP2}j6e4UuKKC@#aXjJo&7$z+MtdEo(cWsZErxyebX8 zHMf(8os%K|f)IT1w1ITatL-~+Eu3AHM65jRL2C_nFXZfl;hIQNZaRx9#Yy9=Hckf; z8;%8~^D%E5mlb}v25mN#;pm<50)5$G^t>)4I5jUDGmjVI+AE83-lagAKG&97I|sm? z2aie4hR=|~b?l;pH9}+6QTwaMjqjS)7;}ZOqz~uu>yoF6e@i@Eb5AM{n6p#Z(g52|}0;Peu zg2EOX!IEO0px@R_P-CPka9t`aa3-39CqZ(8vj-*$4sp8_-9%{td1oxpVcYSFj0t+F zEFmi9XFWYr&@_0W~j(tvrU-q)X@)iThGGS9$Ege_uY{5MUn63t;9E<|1v)0*ZT_|SKZ-uPZeV|~=vGP|AZxut8M zdBQiMPfCc2SQFK4iJ`AM?C@~43Mz2t?A?yN^nryn#y^;W+ny_N`p&B;_v#I@`^O7> zFSVok=O_%fPQzHuX($*CWD4XYcyX)qs7!SVUAM1+gss#dfkzUE%(D>iKg7)wKj%Z* z$6_$}kqm+-o#5xq)peExj%NQP9}U!C%fJYUvp7$7Ki*j`JJ3ea^Iyf#STdD<63pA_ zQcWkP-6@||tcK~&2iT9HU+C8x2I%)v2J=gmaK^WN?B^N3==K>pWXJtb;_=^UIF!a| zt^&U@lF7TE@|i4LiZlk>x&t5&YGMDx&HDw1}eI@s!v;?wGQF3Fbs*xy>dV|6I3 z?=B<5`C`x!ah;@0TBw%ayIKEw+I=KAj4I{hWlF-LupyfFYj>p|4-d_)p1s_ktsW4R%9_b89xf(EwwTFVm zLU3!ufVZbLm>%m(rDD99XqY)hb!!fw^s0KCyn>?6dM=+s@iEqv8{+-TbtqxI5d$Ft zZ^lKS+_IzCG>PCBhcJYW1bpgt5%sD)@DqHa%|`-w(-z9HlUioM9X%dNE!z(s5?V0R zXAf9@OMoIv&i7)p4FV!=l2qLvL{n6RbRLglc9m`9`o3KivU!6fplB-0ci&6~H|7(s z@&^?=MoeJn;0yL+v?WzZnuP5MTKFr{nMzyfkGb;YD727VmuJ|9H~wz?x;G1>JPkN+UR&Vmb8)io$!vlpF3jx?IHut9kJlz?FT)< z2cV+*GI?DR42;P<_!nLeZ!UcxYW5-U#eE|5zW5KWuQY??Uwg?!{V`&8mD2{qq>*V} z?@9l38(?NkCMQb8@OWAi+wt%;P7>ub>>YP#$v1V}G5rg*c9g{ahp%Z^-C1gJZW$IV zLOM$%00Y+K;s*XPOy$qPi(}@fxkM4?2@lep)1|2T?j}-bP)S}&xkJybEbz1YL8gA1 z1A4t9WVMk#jA@z?u_wM@nePd@f7(dml^}SVsSbNq)iLn*I;(!e0oNS4hN4Hl&=rq1 zpl?kw9{!L?UrA3w`&Ic=V3vnvAD7|ICrVgu$TD7gqwss{HB{&PHZdk~Si5u|zOi({ zc&+8w!+EBgo+o4G)l}5@+dxmY-=#*gmeAI~GHT`5L06x$0_oZtr11kER367dv)XoI z^ELwBJl+NK+||M7vwTnW*s{a-l z+U3qp&71@m96X_;Q4Wmd7g6751+21k86Ca(o&6>!UST)U5jTeIrBnSSP-tTw>P+m% zxpInvkIOt!JJAJIzSyGc!7aGV_%b%zC*h0Zk%C!sP6_l9Yy?-z?qk#B<+x$B5^?)g zPSdYnAievxF#g-_($T$DxGFRW*Z<%+)00LpUC|wD7KjVB9D0q*teP>vp%4oO{jgtm zH*GsECiu@$RM5CF3nd@3sK4_G8}Lya6pFV{6-7>jajODagt>fy4@=;gtp+X_smH4e z#8M@%~33}wXEc*+6kF?0B)7OcGxO=2Wgfw5C0JgH4*P3n)($NW+n`BwsR zuSx@Q+~Cw>eEOrfgU;tYBT8$7p)h(UM0njKpTbk|>f&L{*ljA9#ATEh|Hwe~w*45l zQHqYahT{gEOT@4>6IO2j#)^N>hxmW*VRp@ZsPz@)$Gw;53(}1EDDo1F4?TlTyfdIZ zG7jvy%p2A4OJIKG8yVVX57VEP!|0c-pg2XFOj+@YWN@tIP^A^{W=cQwn0119)=jt$ zyJ2zL4p{nO7fjebmsCw(4v&J(-~^}NPjo(wb$8{^a!wW&9#hBJX$^Qua1e!$2D1Wd z&NG@-h{h4gC{VtEeU;jR%Eb->Uaf)P%=#EXi&CuM%E&o^+;R`{yka6#-dRQ-Y6p;M zcdW_!!6?Y}+zW4t7?`NI3$p(uz>J?&@Gz+a>L%C0qLFeC6FUN(O-I37XEroSG?SIv zs>nReXr9@nN93H&RWf1SbK>>>pLJK%V!Bh|18rlK@VCnyIx4u!MtwG9Culw6DSEXL zktv-#Kb3vfx8~j>1D^?;Nizi7Bhe5Q{0+wOJHe#04_;*cguIk*Flu=jQpzpBr&kz^ zZk{8ev!}4GDgxf+s3@A%{+B((Wv{D_EW-V>w7ATB2bAzs!woYc&`|OyCSflg8>{E^ zDiq&T<>KEkF86#zf4REmJJNM~GT5koBZf)7uEu#+*;^KLG$(_~ND!BmoCcxKqTt}|@%(KC4WRH)n7>Q71FT9^s)_t*$qIMPNocxTz@B>< zL+2F9qjbtj)Uck1PnF{6$IGwiY1eq%qi=@g-GFYLgVbqYI&O(xiS?fr;y~q0To>+@r;h!nN)N`{s9+>WoO|{2~jps9F<+Bf@+F~(z z|IY|KwS!@u!DiSXd>m@O>%l2QSrA*=MQ*;|1Z~oW^fSju&yfxxm|zIgET54Riz4B- z!6SHjun7)pHbY)xHx$We^EWA~@ZX5b^RKCl<175v1OHy9!jY&zCehA~roHCjp_0G! z@TJq(75EsJ92kcm{dOU5MFM>-IY4KpuHbakyJ?W58b0TCv349Q<|Cbr4e>SXw<2e_ z#M;6F`v;`)!S9rXG*VXWJbL+AVq#C5%$)VeiFu7}PjIBta{XJas;=Qys2zK)r;2e2e84&{DSVwV0jbbgeE zn>_bn{1#M;@#9Wp2m?Cr= z56`=X@;xu{|a+zaPJCx(vzk6|YpfcX85yh8n?v!pGVqQMJNP3*jNNzzSc^WRn z+gTVw7W%c2(=#>6*zjZ+*t7w*#vg(QG?nugCBboJJrKDq54tu;7F!djJvd72Lmv=H zMg`Q$WI(K`j+8xq$nKx;m))*0NQ;OVmUqlWW_BoT8L_~xTs_tl^^MjqZKZv?YH6|i z8ame^mL8Xu!kjF7`Z6S(M8X?-i9Rt_$f$1^|MBghw>SmL;^nitxYK=1iB&@VAL?Bq;N?Br<3-aYS&8Cq$OB!!`&*bcn%1EHWT0G?PTfPQ>Gv^QUcZ4wtjZ{KOKSbG}o zeX)h4APbn+i`KucvXdiqVei>7a)d7qHoN!G57++C!7r2qD96(JK92BL z#qCv0=8?tIHxtRuSXhiEQ2!^1xVauBw~rnp=k{<>2XAa}`L!u{?wBbGF73ke#`@^! zA4cu>o6==Vx-0hI`bAcloCWm&6BtNMAKj93ny|dz!O}Jx=6e zt2SxKDPXoQ-$-;jrxSy*h0LaZadd8gE3uy`N)H(avMw4Kbm5D|$i9+iTl<2lUMDe&wbDm}!S_dcsPu*qa7J5^EqmV}MnQr@pji%^lZ$-_UONOHKrOB#`kl&e2Qt6t1(~O3&Y#B@zCUp^oh}F>a%=1 zonTa6@%m*gshA-I8{G|Hj>S!gPL78g+jC)_MGIt~je{ePbwR!&3Jjk3v%d(`Ek6Y{4+G%s=SeWWBvB?G#y7FJJ;V4^)$SZVzev^AWjkBLw8OBE+_5616T9Vs<=` zc3UnZlf5pH?GG3*-y#Va(j(;K#`&NgV@&ojm+AeAq6)e5wdCRJPvk^`8963%h1S>S zRScMkK=U(Q&}cnLlQ~XWukb%seQzJtiFLq2@{Jl)iQ`AUGNx$~4EOWK1TJ^M^>PUA z*xN&kQ#PY>qA{mOb4LHAfJz@Gp|58(4Sy+zK^1jW&2={Zy1I&cMp2j)#o**yEfr6s z#js@Y89Ieu#LC_NLANW7(Z15xRMEbHCUCXdrc`6Pp&)>YwVGqQ=y@FK;%1j^n~|!T zBky@M&CC+VXO**Y)8ri1&aj*6|MAD~=lA0?t{!WsHlkJ$+`OdI9sk&c)2s1&c}s&; zQM%5C?!I=6&Xv|gJ-<*^O>!DsuNYuigRQivHlOx;f2M8RPBkXf4kr}VQQ2c=XjsJ1 z*=}cNm+zcN$H&JI7_Ou$x7Ok0x30Lz<~RoRbkNsTessxn&LkFjpUvtYrdemE@E_Hw~Aj+2Z*j6Rc15 z#TV^?7*ZF2CtG7Mt?d`Sw(UcWBM(vGPz5gaEym_mmiYF`QIy#lfImM2TS_Ap-z@Z67;Re`T~8@ zR?WeB&v1)%{*kp< z;_5}J*Qe6QyS2!%DQBtb?;q@6b~XN9eT;39_(;DF7(l^%PY7_nM=W;cfY3Aw?#9>P zPlE@XIJg!zd@KaPJqp_+rTIH&JOS~)*FbY0;rAzc@$Wg?^XIhv2EQ5A&^^5h4*1GJ z7O##en=upbN>Mx z9D>Wg2k-yvO(W>5vhNP?31ny^yM8<^0@mm$=ex>5re^aZ|g&v{WA@pl+D6%1tpl9nu&Lo ze#7H~zc6U|BdlF<8_RKmAg1;yF7lWtXvmo+$Y>WA2&(R2Vg4Mn=Q#4xc}s}TZ(Z1s zXi3&^9vM3>L-j+o2n@fs28)6zq-W7;&@DU(RT@t4TXr_w=X4!l)=cU=Gl^ZkCgb^U z9!Vc~YW>|`nd?(#<2!aE+Vqs*3@+c?`RQGBJQ0B&i&kK@_(=?u{y}}up1|@S9vC29 zOuiMoWsH7aB4;8b>Fc+$&?7gQbS2*g;isAK`#~O*N44(N73-eJp4B0H12j>hPv_9?6PUcO3gZ)$=_Bdp7w$iIB_5pj#k)$(4ZA~&YuK- zd~D!(jSLhu%moY1I~&tdNn|#tLCRPqc~TinTvmQ2k2hJ8Z<3oC*BMVK-*i4&uHig5 zia!6>WY7qS0D5rmb(&FWgZpp&W%Ya};NX}F7G{T0p{UV{KSgD{`3J4Zm6JY@Q>zK} z+Y+GGDF!5`FM`B;0w3~npb|`>dY=jG{Z`3H9&;ydl9ZfynMjv&wf&>xx)62bEjvZr zg|3;YNB?^EQJRWm_6&Eu0F@<;qJxk&PZP>(y|8My>h`U+X(c0b_PW)GjQac zA}%-7#$=}V^U06B0A)VtsR~8a+MAW-d#YWDi8dn zQjfI}NqEk6A0CU?fWO~Ip{i*as{DLT7jV7UM&T#;yCMvqJ=u&Op1R}A4@)`yZ4+bS z!sYY1t1(VS!m#D%H)1Op%iK}Wr?yAK$=8?iaB#aBTuyHw_hud_zrH)1YR>t|i`gSV z^A~bi7F`wW;Dl(pi@RTSP8;AIm2ld${4({GNT?3OYXsIMO@XQJl*q+5LJ-?;3^7qL#NzyQa<$i;*7-@$F$rPJ@_j@H1w3pK z^TtNY*=XgIjT3Wrqt!2SJRh_V?dDy_J2jqoMn3@!uIk|xwuOeB4xo9>pJ{#O9jfy2 zJ5T$tK1^}q!4B6@usCuOX4qLk&Z6(c??E*waOhyGpCmHM+n%z+d+$)ek6HB4!$Zu# zqf)xgR~Bs#Ek&;}530m*C4%**qd8aaJgJ_F5m(Q1{AT-#|1NK(f1ENI^+l27x|a@7 zEIUjFTolQTOesj_Sc&~U4sd_|3V6LBja}%{#;V1XRyZqU(a@vwD?ZxW;l;+?7(O!* zW#2Bu@MT#X+cyLSyMr*w-wqGb8Mt)1GU}u_V18u{RjQtY`pp&eSmzr?^l2iq*ErKE zLMMW(Q#K-p&72t_(WT({+!vVMLiiPO5#s-(z{k3?uzsmCwDj14#{*@uGLy>_Ch63G zwMEe#)3M$z41Yb|hkJF_qVugV`YC=h8g;1PT)R^0U2}|HK6j7V7gJ0My^Z1X6i&NT zEKeRb%MjzDv*du48te@hp&qXiY1VXg#56VRsx79|7AeplmP_aw8#A0!v=QeQ*r2bF z8D1K>MDPBzUUBbyfxi_BW~g>N2NFw=^yY#$=N=WbT?2Q)EH_Fkps zpE{|~synpXx|q(rU`hkn06KoJHkH}`ie`{^)U#n4sk)QL9yYr}SBSo+E461(Ex&8@ zN8B=Ud|nsXY|}_Sv&nQieqlaLEN4$W3$-4OXrQ~zKhZNw#jta{2QK=pMGwz^Pvt&Y zQgI?pZgp~;)JqqMg0eiBXOTyCR|PUxQlU zZ>a8UQ*=#t$7yfx(5KhUQMxIRHkNeL%tgh-pm8@TG!v$giKof>U+qkK$1JFuIvLzT z894KMAI!`+0j}3$f!C7^1%ny*8qu7YWOyOi1q2a>PN99wT)39X2?CkC#ki0nTT zI_2I!I{xr1&Wn11NmBZ9f$jt{=RP4n*iolrb`Bh}7(%-9#S@PvFCs2@*? z+4ua0HShHZ4GEf#hwhf5Oo12r?0JDRjB=6T72|~(jvameHjcU#U}j7tKDKT_DTy9@ z@_K@x{MuD~#NI~T>I*nymk*XM4aIinbnL6Yh&+8${5WtCZ?s-PADai(=7T zkC%DwwUcpdFrO+ff61FQ(#DS8Uq!M`$m1Q8JUaYJ8p_?0iCIJ-Y&{rT(Y*Z~xfi7l z9H)Y4Y`#JSW-6pS={8irVewNfXJY_c2X;-LvDv%$8fiz-9HM1t6hQ=+bB}L~? zGD`&+R3g!c%jg;pd!1%+oZ$e-Z!UyT69LSrZ6_i_PH=3tEu^b1giUL1!R=pi;CR#@ z>|c$L%N+@Di&cTC;@zZgl{r9<8(cqE02}{m!N`^%*!XJ!(JFn%OyW3u^Z(o>+hg^i zz5FPdTN6j!8*f?LUAV{QJ$Q;$`GuI?Q-iPWJK*!iF={vL048&~$qsJTsYXu2(&GWp zNUuTTTrRUvCm;H};^DG?4t$@vgZR6)!#M%KHU!m0uds+<_yU*)P;}A zL2z3kn9vX9Z1SHtX4h~K4TzP&wQn78y8TVAj+T?srJv=ZC>O;Fplg}Nka z)0;xqXu;qD%seQLk_SR@KgmNLr}tO*R)PJOGSF|oFZOdiao3!C=sGtC6^J1kSm~i# zv4*pzU2v@N%O@uf$#Qmf{CWvC@ zWfe4@EsnFAuP}Y%63MP3K6H<$7rlF40@rGvqPAWWuu<_FTOoXfR%{WY59;(_bP}J0 zFZN?(FBpOHHC0eZItw+l6-KQSKqo8;D%YgKIp24X?3NE|ZIs)d1c7_*A`pM*1CkmG zA$QCI21*o&{f!cK{lqhPd2bOqa6R|M*yq@_>I4opUBKqQsaW0{i{Iil;STr7sN+(E zdHb*6yq=AC^~M1-bFoC>{q0oNRujea6tHxr0S-6^p!^ZQXFoYlS6>r7B2`MW)|ZkW zWp&K!B>mC&#_97J|{5GMnMpyR%Z&Ng3i@N_x#s-KOr?-l9#BnjMqv5`Jm zr^2$nebi+vg*F-(WBK1%%#3L6Ka5*|-rZVgxkw3%J56x&JvIEki%0d%W2u&#In~pZ z!_prvoS)_meW1UBnwP6npSF&QXO0rMy6G>wq5lz0?^D93choVd+W=R{M4*>i7>=v+ zK|#w_99DMb&Y>u}Am=%~_Na%Zt?s2AYCoyMBrRl~xZ&jKqIibeJ=mR&z|po>*ta(j zUnn2O-S1XlgGCf}6suz3k#y=DZ;sa$PqD4bq-f+iS?qne1(iRH@N!ry;i|Hv zu{v|9$bdFM7a5$Q=g0;ti(P>|E z9yeKutWm~pVh~vyluacLOQ5f%2oAO=k*o5_Y`}UU99}Mu9x)EM@TL`tEuMw5d+pKZ zdl7Y;7C;TG#n5?dI=j$xDtK(YL~3tbA~v?+byZ=Ucmz6a*h^8(0sEB> zlH$-+l)_Z9F!w5**ls~aGUG{@44032UW{0831TyEdE<&*EKOK)T6K8s`}Ds{=e)eWS7^f6hqS)ERK zs6Y>AN0Benn#t~4T-I;V81db`7@WC0Wr;)YNOF*VgC7mz&2|sZ2TP#QSTWjy{ZR+ zh3nu2$7P?`^@!{?C?su~^~`eJPvk`_R|l%6F_jBXvr-1GurB%=6E*V=Da!0-mN=ax zo5ZW&-dBz#>Zc8t*Jy#AffI<7w-L!6AdU|4WNm&aQ{C@FfA__b7YqGx#&Zj5F1m#7 zYohGkq3?8szX&=lSV7359H|UGjX3|+Th0|)@0PT0l;QTcZ$~jGa z_4Z2WFMJF5B?1=aH9$&XBUt2m!X_aOV>x3n=Mg>)y(#y}sXf|sjZ;3^99KwmZQ4lI zq9U5nYe6eB%gEt--q3sHI+W;sAW@mip>*70I23grE;c`h^YyI|-5m-=hO1%0zC!R# z{tQBS&tO>g4cN_>;y?CE0k05kXgd}M_wL>!^=f`#;yDpq9R7n7!rM5_#(UCL69b7> zXJB@Q6?_c}0k;jopcEDYV|t13C3}=CFOVk@^F84F&<2Q%5`n4l`q03u<`ul3M2|0c zPp&k0F`LfM1AkQyy51uO&W;q2#KRmrLZqC!?`N5codNXBCvgb)JjP{MDZts2D)6wi zm<+r>MTDd7leYKYnLO{8%srkyoOmV1sx&)?=u7mUE zYS^<^o!|JX6jI72^4C>J@NeE1;rllIh4jPEz+vTAIIeyc4(Nozjk-?KJNlj6Id!@s z!$}02-yCCYnO?fkEn#-bUR$z6!IYGn>#=jvx~U^?GH#vNL)YIn$F?n>saDE*j)Qlb ze%fRNa7doa_NgGh{KtX%#bl_Sxdn{mvS7=kE3oUT132!Cht$@3VC`PP$>a~PO(4PN zA7Ekfyde0YmJ3roPQj{5M+ly?4E)tKAh@6mbj9qT+e!h}lQ?b`D$nt7@@Qn$B_i^j zVO-h{(;)?AqW|X~X>JQ3o~v`1JO6r!#v3VEx>gZCzhbFeeJM`g%kf$>#t9t86a}l- z4Pb4goS^vQU#t;U7o4e+6r^rd6&#D2BGC9KDhM6CgTuUYxL_b1wOu&x@z?)Iuu3nr zoW?TiGWO8x3l*8*%64)g?gA}7*h+>ZWst^Ovl52Q#N6jB^u8a*uPqyd_)~GP+hq!& z=ApFX)mt|5Z8pZ~uEG09H{rU~+tHuXx{Nj6z(aS|VxL1Ht!A%b@|wx`e6}$94Chg+ z^`dw?VIke)HHSLSj%Ot@3Zc-o3zh{|L8$Il@U{s9`^D8@P}d6I3TE(wzBpUkxUWW~ z+g8|hat7tvR`&jvnfPn)2ek@KMSs_E_^30844Hc1d1fhI`R$57ZtXw~x7{?bL*goknGIphi@z$Kh<=h@4rmSqm+hyc1mYcLSEoI4=%%p}L)^yeUdwSeu~$c5bFLbl&?&TqC0^j&v*`HF8O4n7bI~ z1SjFOrW-g(f}a-_ zVXyENT+cE4cBsCVfy+|M zFvL_8$D}^cc$>5IS+i<+inKSqFQq^Yt|`%KrDb&Ys5Gt(JIM98im3KsA_guGz^=L! za#xdMxbHYfJ4NEjd(}vyd@hZKXT;H&D<@&dt`+EcSQ}RaRnP|ym*J-FEqFY?jLWZj zfW5vSut)VDx;n}UTED%z^PPl~RQngG<5dU>C@h#B=Ze z0sQOD0nfRmuwN$-ERRRR$`||L$}|lyGbw{T(E(ud{2X}dEQ4nY|b<_$G6dQ+2m zg)P$LLZCN2@+*&S^Yo+95RQ5|!&D~M67!9Bp|(XRCN2y?>g$5p{i0Z(VMDLDY@pqv z#;C0+Mr*#>Gnw-ykUha}#8tM0^~v^zOO>bLj2{nF9E(8xSqi+3vjxlPm)Q9SLwNm; zD@n*(Yp6cw4odm9uq%)2X-(IYUGugvvYWQj+mBAt(kfAusSKk3I7WEiLoc>=O(ty( zAyjVHUNZTrKl!h31-pHNGM)Fcf!sLWN>+1hA=7+OXdi!>l^c`hnaNj@2PRhVCL#&c z_r(I=sFP!07s5~+$3d430KYspC^>ctnhmu;M{5zW`1OUElBEbb1wP#UJ(FY7s*%~v zrm+2DEO?ov!I7Vw*FEVDc^{)m9E*RD`>MNOxxp0RSG^#!bBthT&v>|8bA#Ma+XQ?6 z<23%tcgeI8RY)EE!`_#!VjbG{&`+@;?BnQ5Br-}7h7(Hxe1}Max(Jjk@`2^C_OK!~ z7wVrfaQn3i{N%KvN1IGxgU=?I?1!*TJqR8T7;#xe$B19YS8^nK0sKAe0)IHK)ZOpH zq*4A2GlU1oC*$W#eT^+Sx{*&U`HB_M6Gc!#$P_OvIYW={T*&!iOPCEIi@@NQJVfu0 zff--oA=Gs%Zs<}Lv?a?3UJXCOIck5otno{Dz$YJ@b{)Y79-J@Pu?Jttm*DejVW`8+ zVpi5Yq_JZaRA}ZtbUeO`oT;1wecXTcNfZaw(kEo$*SYXY&I7C}d0^T$A5=5LV5hY@ zS!K5t+B2LXwXA_`9~vX-#zVa0cgtu3yOnNu?2d0nW6}D>a>OQe1ph3Aq5b$lquCXDzQk)wEos)&gjp6Q zaQjCDr?q=a4CX46Sj8P=PGKFBFl9TYgn3|>YBfFGe-;&ga1obfTb_wzMHXI2T90`%!!bM{7+*HK;>fvT3T0B@viKB9a@PQr1b^7)!1aRf z#rO*z%kjOQ=75;1Bwy*a0L<&9__oVfkUQHAvU@lkvuzXXy7v(7PrD4ebjE@CDlfWB zmYW}BC*$`$mH039F8=L(kLvaz*rH#6GsZtfFB~U$=QmE^H~a~IT}Vc$u0)!Xb_f+t z+0p4c7twh|p`;^5h8&8!PgqjJT*(&)&&nZk&byI_UfWGt(hb2&bsY3>eML5{sb=qu z{6{}4XybqRCuyK)3mu=i1D%g5;rD)1R4J54mBVXM{j?*ps-skBofi5HaxBNWVK{vw z4~-_hrgrNy=||~2%27Kj?yr>OG(ly=+%XpFdefm&HX7c>#DcluNpRH+0v`!aC>b~b z_o9Sg*MXVjT$(3yr%;4tg%`2@en4B(ooHM@A9H2k3!C`bfJTZdgZ7bc%-x7#=F{0M zGPW&&;7~DZR_4q!)%<32(+^YO%s6&`A48uiMHBafd89G=E@@VBAUlTg$)iikkZkP$ zXX*(QaQrxn&8Oj2us6u>oC=vH40Or9CzBHokq-~Q65pq0aAU$K+3NIxDSMs8%o@^T zwUV=cI9(e?JM`g@g9q%9p8>-&48g@m5)Rk2F{`HMF_wp} zRdgGd(tH&z_tdAFE?Y1IvtG}~W7=9cUSXJi^H9e*O-sqU+=(P7NrcqTN@nJ!@tD{X z-8_+#q^G*S!Us2G~ zHitYtw1j7Dez786LlJjr)=`Vr{kSjd6s8((z({`;JaFd$9k$^i?f*sXl+C%l=zY5K z>Nrl9riGilE7&Ja-&ubrM>us~2`sONL&o45$WgjU#_cu6bt}7QM8k2aZ~q^iz5gHc zLn4`amtG)@rY`3rzd`@wc7rROWKl))HU=KOi?=>M!HoWUsGKv5T^laonY4v?eryry zzGQel)4wqTC6C!bjxl>R+#UE8o-ldH54xptA$nB>1X=mR!IS>*OgjhMt1RG(?0J~x z?M)h1E{5yEp-`eXNP25O@`Uy@kx9{7@MyjZ>mSfUrU#s4*p2#Rw0aX*Ztw<=2y-}B zu#o+1Va%&Jszc=cT;R_4F0wRDh5OE$H!*u_3=YO22HZ%-`PU}miP@i-;gRRW+wMoj!%{x6yP^c^-x$ER5N~LW z_JZTO{!q0h8T@C>hv!1-U_5y(unN7*0kt~vaCs$j!d!)BpH1hzDl%rbynjlDYpGbDz0&eGI2|IZ!KqV(0Dm$FPWr`+zUKR#_u$ye1bA@yk)e_cx zCV>mGByq(8_OaRmq9YYe5;Kx%<^5-5@D>C6aU#TtNRWfQU&-^Wx0vlhBIGW|@VoLh zi`b7?!@{HY$SnI$TqZNenwHBUVValVX1*nqE}Z}hf)p^F6$n#;sv&F&w_{`-;L*K6 z_`5>^+@lu2+KDjQX zL^iG0gL=c~9BWaXJPJ-Cv*pa_{YEG9YR_RZO?f6YJEct?uHsYekrwi*pSyF7=EA|k zELdU^3yJ&Aa+$a+=x3zEzMKbectS25@Q>$skv7n%y$5_d(4*Dc4s+KIfm_lqo7%O8H@Vq=9Luw+?-qHob0{dw3 zcO~4Cs*IA0r_${sqr^1(7`c}&#Y_3n%{wG|v?8OmixlxQh%#v+Ei{yjPL3urGG!FT zZO2PBx=25pBG{jym(Fd(2}1v)=)B{peBU^3mMEmiNGKttNO7L~I!c716rsIPAq_%W zGP75aEfh*p$$IYV7)6w}hLA*kDWoZye$VeOe|nwsdd_o?>+^ZPy>a#{UtArSjuVMI z?%KxXh3729)e{SFONBJ1u3CbB*>QNJ(Vafkl4V~PYmq7|B?xL?2V47(z>R;qU{rPz zl#QN%vFS#jq3Hr=GA_e7(QwGE@_;jWf5=F765aYnKo2ezLs_NWm}=^U0p3edaqo82 zecwxC((Tb9D~!J4vIW0hmDA>%Nz6H|blQJ)1>0s^#D;e-fNYuq-V-^7>a<%BGq(Yv z>u*C~Wd%HyxduuZ+{{dK1UmM%!lj+B;ft>(KkC3F{#M~9Fs#@JR{bJ;DK58VF%}7i z;Z{(ya|Mj4=z@1tA*c&vKvd)eX*++Oq%Rhu*U!x(OAkon;x1Wy&-vi8S`x8gTPogf zQ%A)L3oM=Sn-+&lkX4KH$>Ysdw6pp=BPv=!5`QLxo3;u3j1j<-$ERV=ISTvr?*aW7 z2fn{5$e>LI%(IRGhfg`c^L`3rd50h~Qyd;KTqZ=+9M3l}v>Uw8vRDHrCyKKE!ikt9 zkfPHfhFGm?0~m3sg*2;haOk}aBju5BeBd?Q&Yr@TERf)Hv2NIqcpLN*g5Y!XQaIao z5xV)&r1Lr@@1D!RNb4$^x+R0j*xgP?GS_3?z#!-0Zo^r+s)BEhYJ%%hO9U+rnu5V+ z>Vou8Z9(mLCBeGM!h)_{1NcpOoM3&8tiVzw5iefJ$3=$@qV@h?^yFeW)ZP8pxb~433Y>fn3t8|lx_0;&}b8Dl%(&}mr+ zIw}Q6dSqa((sGzCe3=-ADw7|U&sm?pe~7qg8u4~gWa2GiP`)A!n*-kA8QH0Va8n_{ zHOC5cd@z98TxKOLM<4IiY^0f9Lp)L09bB(581ChifaZY%P-PPdd+(=$b>I&eS|GzO z6rai;yrRl|GxuTYhHxkodjXdxM}tz;IIfEw#%2j_VSrQ-+7H#EaJz~?^i(Te>3fLN zJUg+O>(KA6e}|FQ^;F}fB6?not}<-(rNd$d>~PC&jwL1muWBZd&4b0@@m>$Mu2=xQ zjyK?ji!7^WVMY4)oFuJzBjoQH0eQOmAWRDCBCmQcgZNB&66CU)R2!(Ex6li!x!8xk zUhPY|YBEV*Sqrgp{Z3q8D3O?+bEIfXH`&};Puwp#RT+Mc$DV0yxR4q6pG z!IY>5%x^!7rR6)&V5%OS^4b%p+3m*D89_MfR5rS}AH>zy19&mwB7)rTI(%zYh&F95 z=ySZDW)}d*8Ba!y&zd;wRf6x=?ZMwNH}R}PCcfx;ij8YyQRU8L9Qgd3(py$Ua*;iJ zVwb^+;zC#+K8e5AM2UZF%0zyKgc?743&#O9vgLOM#PYp*_VG<7+wh~#+VRce`TW?6 z^Z45ARK9>G%MY2E0C6XdkeK^>=#ons_~sSoZA-F2g@rx1Y20`0jcvtdds7^fy@PXR zRiX9U&v>ps2{os6q57vG>Hv@@NIxS*>LDMeL>THkXp8m5Oj$`#Q-8 zEFe>_oC5ida`-rw47}E3Ajzu)#XAq-wN47WIbR2V-drMK`<{@AgSCvi>pMoKJf1wf zR!pXgbQy~&-5_gYqM@Tm2V(P9(gy`?#3@x3zFp$8yR?6hKW%5ox}B~Nx6mJ2*LlNs zy*E(HWpS#`Xu-D5W6W@7FKOryW54{;0LddM@M4-h9E*yBS6q*{H8q69YLf&$4q&~F|Z~Fk*@noK`J+eZR4IDDgPECeEt!6z2puf*0s4x zyM7Yde`u$BGLz`?b`w0c`3?1{R7TfGVbpo1#_>{Q-E63%Y0^GSzfUkF@q4CgKbo_IK z+Gx$fEAOOyx@JC)t+6+ej*aR3dI!?rfJwRpHhJ;-D%50I?47z@t=khY==PY)C zzP=u!={toj2~_}j@{^cp{2*IbiQ+CO0A^|HPPw&ts|FeK~L_G$#8Sxm?3OrNtMafp+ZQPs<So^xKr zURc)2s43qdKHT-|d@pk{zvd2cs{KYHB4@+88^6e?@Bk?^wP);_qM3`K36%A?%F3U= zPxI?v(}Z9{thi%<1zijAE9YpmyNU?S|DlPdus~};FDhk~Aa}aMwG-^oTb;`no9JQ4 z?fGb{KE!OSyUGNdiGcp2A#liNE$sMw1$3JJLdM3Aa3hZ6O4?84Ke3t3U+{Skngr?J}nxGnPhv-zrgD~;E%?8yi~=s9S(wGh(y|46Q>JgDChf&zOP zY>=*_%KO|gYHJ4`*su+U6Wgha{zjZI&z1g8%)!lT52CMUD~4Ho!KM0jxIN-7p3vkv z3U4Rinem~N{O}-7_g|8UCvr*I!#S*i<$B5slK|`9nK0V66o&U}a5;K0uv`<&$o#Eg z?A4TE6Cc=a7bK2X=RkjsE&1}QiWsyQlC_UH{)p)!V)aZ3TW`Bycak@k>xsf)&a2gN zEgwIgc}iQC@UiowF0L-rMqP0^diqrf71|~(P!CfREVMHeY&!lIKV%nT(Jf0XO;N&2 zc9HOB_ZOnS;3$!Kww!Zqa~UyyI@$ZUhDkK-rs*^f1&?l`^lFZ~*)>Tpa6~}<2Jj1SL3~FBBo5z%q$>=_B1e#^8-n4J z7vS{TH0Zi~5MCCnhM-RZF#lUedL+3%%Jxd~oa;OKcFv&<5whHD!K^Bp6p`y9lHhub z^BOkwlR@1rAd~iyRPwB$fc+1eSM7!Qv3W4uSOaz&uEUD2qj13aEVO+&2>-gr!ITds z@TT+%Z>NkDbK`av^?%_hdu$GT$v*Xy-zz>yg?Hft$`n)Q+2dkjG=LNZt zIl>HHS^{r!Zc%hhX5xc78MhN0*FX9?RnEVTNwdZa-nKo#EPHW5!&nAp9jU=NUPUS8fa8v`-?q;4)_b`!5d5 zS4Uv(xinM~--0C_pJ|fHBJAHTj*)XBPp`hVDuPHsE#($5I|w^#)mbUdg+!91LI zi4gOVAZClq9X5IAYi3R{Ai6OW+N}?!|Gg`*W5vOpS9*>&*JuuYBVk^MADstPltjqcbXMGtK#zPkN&@U6&I zj9OQPE?h4A!Q>2DtFRa&)l!-A)s5tVW;{2Gji*w*@2OzDI)44$%eoqjP>C0zxK2(D zk4szQzIrV>X-XuWbn+c_GvzV`^ADq+l`f7vaAtgu%pxY|jc~|o9Sz91!(@JqB3HQV zo!@O)@UE+5w(NaOx0Jr4o+7hRS@9Qh+u;-$SyoDR18 zEnB9~g601EVS)O5a-n>Xl%DuXK3^FJH80D^>y^f!{F(E=bzEiFJV_*C9m(A6T?uQ0 z@6hA*=~a`X!(g&&ju;Yy2~)S~8h`>5VO}y%3K7<>k}s zgG03PZ#-3db%8nQBcR5TqinCnXKEhZLP|=4h}w)nQuMf;riLYvvejxNq$h)1sDDrH zYJMRO_oGN8-2-0cGoXKfW1xD!Bz506K%t*LNTxKfaT&qPrSv7_=d2V;YmHEQAg{{% zxD~ZmxkA76oMhqM7O*j&0B^d4K(DO^zK%YH8C(z6_COHqSfU7@+`bT8C=IBo2mke# z&~2BaDc?;2YZgUdSzjH?1p1>-g+2Nz%)~Xm0Vpt7jcc`bV}ZyW_NA#L20rJZvY!gd z<=Eg@!vQpC?!skXLU2(-3)Ut zZ!&%OaveE7@j^%EE=Y)5xcgT$_u-4_PZho=kwwo(*6# zJee#q8zF5qbD<`73O!~vmo5!EM}xSmcFM-vMD*GKDQhrhmuj4*KQ21p&2|eEPAtvP*L??VFA5DzNZahJs24~O@KNjN~IWgREDIAOT-NbvN z`dHncfC>8Co&C!F*dgYDKP{ptV|$P-^>rgJ?^Q7^r^k_Fp$g>iX98FHUEmZe(9JIuAnoT%wsHW;oDcP+z49F~*(HoPY|YIoYRR>4@z5ld30h`r z;COO0+zGr*UirsB+SvynPf_3-Q3!(BoYKW~26#>L0PCz;m||lH(w7fF zgSj{)eOu2Pz1Bru*v$u}3C|gJ<$L18<@E~FV?pwZ23)c8f^WNZVAt$W*tlpi`IDIl zK5Okj=OPad$<&fNX;NV3e4Pl#i9zCJcNiGA0mL>&!O731L^|gkX}`1;rg3h<#RK#e zqe-a|$4I^+0$<0TavAt}^k(59a{l5Jm{_-ux1|3bNvZWA#$FxNr0G7dEV_V39Fl~r zfLuc48;O{jJ2_8)*l}FS*2n5ZC0vhD;5>Iff0&Td9WSanIKHRbid-_W`5b#j^*)WB ze3;&Sa-6m*ZlmS(CG1yRO3Ve>%qLZK%-J&mZG#NasVxoNTe-ecLM;CNXNMOCZ82RU zf*RSJXHrG1Y3SbxMED{lR(-DUMdAhN>(PgttZwo%!v!1-{XyK~IM}GKhf@`o;iaTI z^s6m|Z1-^>?zm@R@YC#f&Mwv zBI+r9el4A9l}$s9uo|jz+maYX43fiXk<5AhiOi}2d20OFg|x{|Adh4vs9GKnnc0#s zS)35#_Z7^fnNGybScX2_y_)@{%41XB7*kWO$6zr!$adM+(6k;2+}S-w$(97F5g3Ew zZqIRYfFYT+`0{)S=Daq?4G)gs;2|5FpIuHnEjTxaO&9fR zF~^&Ak$9k}2x~kGaL92huA`ASr^o^2J9%hZ&UGwWvM}oA3$*>O8r_{wap&m}9J_iM zv$9QbJO2pU4s#6A*FkvI%LZ@Hi^FW^dlW@PkX*Y+4!z_$I)#SRyyq?z>^w*f%J#Fd zBU5O;{ww;;$k!~O=v=1c4XMuvoE?6yP3oio?Q{l^&uyw9D@P#;rqFgFDd#Ivn>9YG2wJ0mY zs*}glcjDK&8Pr2xn3(2d^E`HPyqHy+V9lav@HxRiUAG~;-4H<-(<$)4B$wWT8 zVp1CG{%Swxtdl4A7f+-@ha>3IBMvaXWhdRdDH3WYg~0J8uB^tt`Ly#DR`EQt=>CX6 zRF~oMe(TFI@Yzke%}|s)Yc-|ohbwtUteY8oV}>~Cb9{L5O>pux$J<{M3+Jv`!Fq2| zu77_T3OKjuj?J@R?*>^SB;v*fbGbH~*mTIguoZqjy8$s8Y2aDNWjGt0py0Ar<+SfY zT&CHZ7=8FjZT%{V{G1lLt8*^<_D?gr=^MvFUYkX%`Tb0J|4th1^&e4iY$kuxdaL4( zy(RhAb}`=b8X>qh7rG9dgVd_c;9pQlj=ld(7JSYHR~r#fpL`4KzN$j$#%XZab_pck z=k_~3%VCA{2IvkLB|6IQD-ZXtqG97*ahu*bD$-$%5C3z;ez-vYYG>eW*BUz0bq+q) z)ggr(KY=O*Q)Rl4b_KA^4j%Gee;=sYEp~=idc3VF@BU?GEbt2eFijEtX3xhVD*+NVtp}$kKS|T|@eupt4(WMR zMrIdYAT?qQ%z=_b`nTVMN!b3Dh$-A5Ov7t-=)DNo&2xh=y=@%Z&=X*)HFQWV#He6y z?H2N4MT_%!JCDC$l}#w+|J_Mn&bOiA4|}TAda7u-_A1UvE{8)ZTGZ8E5jRbhhpl@9 zczH(laDQMo#d{J)1)_($yif?!WSa8+>lW%Cq1 zn>NChq9#Z&iU)CbU-%3=z%EB2U(H_R$@MjpRq#!WsypHp`Bm}cuTd;XY2fF5F;$W8t?j5~B z|5GcXs!zX?RRe}h*a=s%qrR9-oNf&^@6*8X>~!e+_km-Le}lzGFTt0_@%+Dmw?N?p z*UkLL^@^@ZLRF^*t-J7mNJ(Xo&yUCGu?7WbH@HN@Iu~G}R3uY0WJUM(E93IzQ<)vY zGibN|I{NF&?W$tle{|M?SLBmd91YtrM!O%a!+zNw`g=z^p3A5}O%LvKZ1KhSS>~88 zsfkl$7$#6LI$g zQMU4|fL=3MLM1melL^5gblcC*tk}#0kn-U$Fp3LcS9}d=^q&P;S5CmNM+Om2PzU#@ zb0od>9%yTL}zW$mi{O|iD`1>b`@$J@iL48;u&9b+^+&W{rIecYR(+eL` zmM(_bpNhyj_d4=r!x0j3=_FCOb`pv85h$K<8g#k-A8#NI9t&B6 zpXDAnCu}Jk>@oCB3E0^eJ>qqKv32 z#E|5T2K1;&FdclYOiOl_(B?-Lpe%KjabBfP>Kq*DnX&~$H%0_o%amvbm-|to#$02Z zaOV>pUSvftd;jk$8dxaJF=A)K4_{;Uh5I$*I)MWXw~M7ByOz=5+?O=ty)ZGKf1jNB z#qq)vI5z$yX((x&1iOstjT47sA**09j5Rs1ah1=A&geo|tM`Z~w)zwQFA^|*nijEb zRfm%4(GW3z3HdW?lo@zgO6I!m1?|(^zA$G2JQTBFNa)1gD-!<>OvfL7zBFia0j=Co!+x7=O-zG?u==h) zj)*j&?!$|y_pcA%`%;uDsYZ#To6zpw8+x$+3@uO>q5JoHttOfio<h3;fQCf{>>Wd{VztH5Qag6)G;U4`Urk-_LGpwp5IY7W>fcZo#x{jy%~? zS4Qkx6lu4`8oa(#!1kdYiu<%+q5DN__7M||d)tlU-wM!L`w^DM$71`T3;1APAliwB z;k;){(Q^AK_Ut*8MW;4%XZdU?xU2My{C*HYcxmyZdiXIpb$$t``zDh6Xjr9xF`O?} zbBzCej{$$pBT@dnYGcm-#pQGU%cV7Wlkx4T5n7gii8j=<&_Xu}O62dcAws>(DqYU= z*IWZjx+-Cx<^Wvx>xHrR%`jV5h~ITzjIaDsh_96U8jdWS39hMM$e)jMz@}h5UEoUS{P+lpwElxF<7V3Z?BFQ$gRAGZi$f?oz%)k5+~eXn#B5z^ia{kB)aPrz$Z0v zev8*v!0AH#Jbe*<8k?aRTkGd+2xcJJPl$+*wdV%XGg} zArJ-&@2y~3Itp`ksq){Y%;8T+TE}<0>BR4BKfte%_2Ww%bmL39>G5l3P2v0ajl$(U z>F`Y>g+%jb^2ZsB=lg7Y45ikJpeLe%@h_I(W4h4zq@f^vM?V_ZP8Aqy zKEXY+PtiqrwQOX67F3jYa_qJ%zGMQ8E(CYI~?s>-2!>GvA^U%g|X z@TC(3+$>))vX_*;$>lYD?WZf+SSon5hKl+9WD;GMA+MeDIwe2FBmHl%@xHv^zxV}$ zc@~<2AxB9;I+yxt5l%)?pAj0sbHU5Q!l;$oO9Q#=>o*TqT%VweyQ&vble7OZA*(HD zTBQl*obyAi`s>&^=#O!GiZI118kL{yW!qe;(R{fn%2O#p!uS!q=_M!dT&yREvX&RT z8kr>sbK?uPnRRh9-NR_oSN7*@c|ERF_3|z2?M;GpAv3r^t zW@qERJXLLgP2UWiEsXhpND$S1{Q{?AeSLEmD z{e|{wHU6;HQ;0}*1@9Q{p8nS^x}jkgT@xI}4rw=%_@g@snr4%$QVjWVte@nC?j`Q$ zjOh0$9<#IU6g9O^rJsTnaOhGmjSD`C8z#J^rv8ywCLTh2zWWoCB@4=mW+sr@@AZ|d zqNHJgSO)1?noCj=MB(VCO!B5z2@cr2VlBnbFjF)yGKJ4ykZ1SZ;Kjm+q^tS_ICAfv zO}cAf4wreaS@n=4*rpM=4QHswIj-BX%>dU*dZXvzQ4CI#7VL}Ro{h)yg5f)#uzZfB zK;ZzF$6o#b%fnw|(c%hJ+ZKny0sPe)>!G6n38EdY&7&gXo5W|hIc(`4!aHBc+? zAiYDr@a4NLj5nxCoO%>gYw8ImCyy%=|&` zZ8E}}P)IEd+iBk|2fTV^5&oTX9z#CN#q}IRV)pB!*t;Pey?=8|kbWV$WZo*?mHZ^~ z<8C-R(L0a$&a5Y+UYD4$Fkx!(TAh1Vg6W?+R}9onMfFP|C}*?`w})>jM-!ihbMM#?3{?xl7e7@o<=1xn9PNSeTvq6?K{ZYfeTJ(WzTxnlOdJ&r z!uy-A(3!Q_#5?2)+jq;H9)FQX-KWf}GAx{lPxZ^GsHY2au&IjXN%fM|vYZ3{Qv&n7 z+ltmNlOzkhbjZEgi|E#OIwZgNK3i(E&-mwifM2(riK*mOBKM|_y=VN2jPw9)w|q)1 z&TPfSX?Ylyk9ec6it8mO;FZ(CSkLUnt@--6tj7yAPHx5H3y$Ej-+nmfkUh?|3c&YK z8>#k}1m?sl1uF9|os=ZxFrS+DleEGnnqBKbdR!COBH%yBE1~B?vi&1c2MIvf=k;qL zODqe<|M1{ujhW<-rWZ3!ToSggSPqiakIA2gE$~}D8!8`9=Q8Nq!1GE3XfI*lM&uco znKm7|M{L1CkiwinoLWmv5=}rOdpfj*Cm+OC39EJdH?)h1poBw^VhGXg%z&%+Smi0Q4m%BEwCK0D; z^t4ax?3C#!uOQ0ZV|(Hj=W4q2h&m?Evc&P5&(H(=wa|aPbd~eTF0y>;L|CSL8X{Co z;HFU+uwDRmo9)3|cs|VRGl93-$zUYw4(m60fM16V{F%L+1o|fOq(3(^og>loy4hfr z-}nT!y`hM8V|A6T#jmw{I`ypJYb>7>J>bJ5>q1&)i31S?=%tL*kt&yqZpDB zYC$9P2^?_!4nkS4VexP{d^IkC|03IA+q=CGv{VinHa;hL?ibjZKP2F8ZYy~eP(X5= z@6va3Q!t3X4sG|G#eFBvq0P06C{s5^Z%v9p{quV0)YL_nnIC3VrM@tV8o`yPIbZvX zZu6=`1spq}*^f+0^?|8r@0dqF^~uATHSD#JpsLf)98lam422{o;4KO6>=1gJ)_B$M zY7cKA#b!P1L&s)DHGF`%7`y{=oMwaBXD0xUb8Tr5 zme@olM{OacCF_W#Q5aqPCX=j;l3-NU=z;OfDUkSS6{zM}L7ytub@Yh=2}@5fI3ge) z@4X<~L(HL`bIgfOItF`8J>kFy4T$^pm7LShBkSdYXo!tHO?J~kKTSD2ACiIAF5X!D z!UmlSuXAqqW4w=Ria7`JT(aAyf^6d*gxWjHL1MBLbkNnne|`>XGVgL3<6g*l`-R4E z+%)xL8Pvv@J7=uffz1|$=)H=M8TVcAj`K-WD%8h)s=O-AY0XS_ZUp00JVKKXxsYma zXY!3(+g7cA%pMk-36IhgNy+6w^5jGqM8HF4MmopvRqAA)9~1)L{j#9=;TPHbM-SWs z%z>(!fx7IHO7&@cl8n=7g!}+=#`7I7=JyQdsN7}NZ9*Bdb^9Hvj5&zdU)v~=3R9V_3*c#)4K1{1-CBK z_5BywSxM5MaCj1^RXb9tz4L&7>O6Sm@?lwcEojCq2el`~z%U~w1157(X*Bcy{Z7SST{YlQOIQP*3%&a+uy)rSFc|8k1bVuM;yLSAg-h_^l z)fjvt8Eayyv0UpaE_!_u-W&NUhGXZl`}RVEFaCY^!v$CY63X=fNGv>bG8yx~9PnVjQU z9wO|_VW3kIz%hk{yN{=BJ2#LQ|6FL8R64EU)|5A;cj%!-Pe|+C8FW`@HN|lW^ntQw zRk3~~d*@^>U29%R+xMAb#0)-mhsRK2C`YFho~QP5Gw`GTTlSkx4YNrt1hho6;HP~n z1g{v+KNmZm|AsHZ_YoK2Tig)hC*N%Z-c2ss)BXiEFMI)0)iR;_^mQm$90T>c@56ax zVYoVKO!Y5TP{G^)OpGnT5IHy8<(Y`H%8aVc?Ae7)6_X&_`YG$R<_2*%s0l&3>0rr( zLIdZaliK%#T;9m#vu!gV^2%kBUFQPNMef29E+drVyB~J-Pl4vUU+LJ6dK#f4OHc3l z#x!i3hRV-4uN=hV&f{ls;u0|gJw05KI~UKWhM|9sJ8FJUC1F@jT*OyHwV4@gSLS{f zUwz>nl zT2g01PVHJonlDu`&gWa1l}YdDlm8S@=bav1wBseUOpw3>5=!(zULtPpzD}>KyiD!i z|D!+Vs$s&%iKu$+C@L=AkMA2ss-|WiB-?Th(Acpg_Q}?l?8z^$&Lqw*N zLNXbx7U(n2>+;AN2SwDW*~kXXAEGTgSR&+Q#%|jYLah8O*`(_=q`^iSe=a-8ILWIL z|7WW7^}WkgbBbN7X1q$IL5ABww=@BaxQ;+rgcU>|Tn>9=9#yT1H0Js;bxiJA9eAu< zPYyQtkmfln8BGmYqVds)wvOB&3Na^0=ch*2&G;v6xp9`=c=jA?eo}F9Nz@_I?NCNN z*3N;~N*ihODL%P$oFS9SgsIBiEIM3&8ND32HAbQuhq_wPA*GRS?FhrMbv;p>Q2KZ*zMXgmO%_Md%dtnXwLA@_H-_NSi#?c><&WaKe_%uP1JoV2 z4o^gQ;NZ?D7$ncRfr8EG?A7~FVYLUNNHT^;Po9%354og&-)WeGnu z1mfcc2yr?~T~p`N9r9Z+|5qwz?2g6Z-M8rpsNgfEwh3vQg16dZX!gurkd zqV-95#f|H_x{tBj%-(bN<<>yX9wZL#XNk#!mGEPOCFpT`w|8$k*!oumJa*YJjPmW` zxh;z(TYruRr`c;@DEbVwcoRgtZrtNt4D}*=*9hUvN+let^{4YL++({BrBcSz8r5bf z(lGZ8yk+l}QH?JuxOray?wy~HvVjhmnA}9i>&H;-k#%H?Nl?|Z$a{?N*G_Wstu4KN zWIbh0=d#!LN=|}ClkM{A}!DDaLwLGTH7(6INY+Jhu6#} zuO?1qO|~?$q3?i9N}PvLmTEZdP7!W8dmb+ookw{OYs{PRnZC>990EZ`Xkej&+tP<= z_wYyJdLIdW;7sP}Y=(tP_QLN?Yrx&w1uppP0i)|jp^>eCrVnwjzqJHBHRFL1=6rYJ z9Pdiop8UO)Knere$l{0GeR`>9Rmjt$B++FzT>AQ#JoLOpnmQK3xeH0qiD58vABE{p zXG5Ef0<1h&2&Vr|!@X1SAULK3pI?PDFPx3Qs*!=^<3x-Tzc8dbzntDol_Y61elrp~ zZ7I!^!WG45X}``= zop6OA4qnMb@$g2ttf~u(Hhd*}4sg9cy>DbB#vdMjod$MxkIDMVn`C-xF59`unC#Wx zz&f~Ck&;GF;?y|_g8G3>`~uq*cMU-9$0bNu9}1?(F2eu0TCrC6I-g?f8J@pn}YR!>Sq2Zd93+5R-XIUP?sZ`nX`@fDJPR~N#S zB7yZZg$IvU!Tmiy;cq|+mlZn+i+oezt3CIw$qs`ysdz?}uMM7I=_J=z4xFU_@TTi4 zV}Xt*zLJwe_Q)}ma!te`Jxi482*tB~A?VG`2qboU(ZIM1v}?o-h5NRlc|bm{2$+J$ zR%g-k`(IIXQ>@y~O?S4R&!ui^MkwT&!R{!Y0H5Yck_B%k5lNHV#F%3uyf~MXejqjpcVc(o1N{9W1^LOpY50Vo%G);HEd43Wgh+p9+`j!LeLwy%J6yJs zBl^p+y4afXH2&`;D8(sU7O2AL6b5AG(SB zKVvZQX)UV!t;Vu37S$%CbF=4YRAVx5Q^H}Kp(~7v=aSj$Po?3*Jr9lvsRQ-t+VCkL z5gsqCBoijRhF^V&;J4>2Ozub|nuZ3%FJ!bTzdwcTk?Erg#b;tjhA|2n1bBOE241kP zz>v##@TKct9LxBQ3W&Ek>9F;cIrRJr`xTG9~NORc`l#;qF^>E*|7hzjR0EaiogwcLzv(8mFG zLlNXWya-*HXCZczE@Vwv$92d};rW@xaPtw@^V%KEu~gJJkMTIjGuuiWa(uwnA{2h* z*Fh)~4HE9``5W2SBi4z4xFN1m_g zSds+Yd{q<08{QJ-Ey@^b&+)K2G~lSxH`;yVF0(DsnemC1gY*0eu!nUe9}2dS?G}3Q z{lCY=@A?N~>9-t8ZzsdLR8Q#0_J;~@OCXxjAUZo9#!F8Dvrio7K{ksN1o)Au#{$R( zr&emcY9+N^JD>b}yB%^g+`!su8}sat2;_RbCBoWsplPBt+!*Edme#(U4|pBWMQUE=^!0f*t!no6>v#gj@} zZKO+0CeX~=huMWrYhmHLEXa_Z4*h=R%-1C^Nr1;bx*}wND7R#i$V2a$E81GHZm9vF zM=;!$;eOL!4M5u<5k@P0xjd>d@JdoZCiey-QT2xGqPxg=IURC2>N-_eE`qj|w7Wf7CodVhbs%TQR2Kk*|BvN*}XTInebl>^H*v*^JSMm-6x-eKaTV0(z{aFVn57^ z%*-OqKX38OJZtFH#`|(o8bonudJJb{Sx6h_qVz}y9$*s}dT8n|CV_Lw}X*YBol zvz_S*|JSt0{V-{pIG>zL#l~A=OyYUIa6)$JPKJ3dttGK1hgtnhWLJW_UGFe%8xX_O#jWysB0oMDtJW? zSKCl_tt2s?{E}z(SqKyz*TV-^itKlv2Wk_-VMgmt`1akO`+p8}$A~Cl^AbqcZ8c^j z#gXV2ePs52wBtJG6{N@g4imq!fa)Ea3|9@Z7_WQI?5~(rfuKRAT21`iiGV z^slXD63=DP#JG5lE#-q&6L%rhgrPB8i!;8nsOr{=XPQe;wNy;-QTsElNfi_5j=seM zp`Y-xtES+SuZnae%!OyD_0<*RQRyKv>MDgk zg%j~e$^i9Y%GmAB#YA?)aY)u*4Ab(@fjfD?oi|fp^_x>5$n59NEU(~1XDFD|<`ZMV zK`>o059S#!g`Q7-puI1O2$L}BfB&E zJG-~(Jqdn^bgY-ewT$ei5_UHYoe9@&4xg;10#=C)!!wTs8nFu$R zSOL4)5sqx%4?RP#Ax+vC_AZ|SP1o!p#D0j_sZEDvPp^T<#h0MVJORzbH*n?)=McH* z1njc_7%lV#gL4bvW`_;St6PXQ1q>T|?l@)dDU&w6mvnnp2<2J&V3L>_gnpAH0TPec z`S#Cp#!jBMm@&cL5?{MNwimE*$IO6jLFSv*aw7-18^_Yf=a2_*v zMqr9_1ed*iN;Tt?$Yg_f(rY~tEc|Z~hv{x`@Jl-EGirtXevJ^TGMjI(L!7TnJK+__ zV7&6<5NI!23x6W)pg3Xyoz9oQUyJACq(?Wfcw;h_xyIpqjv@LrZYsR`yqZ+h$WnU^ zZ?fo@Khe9Q0}4Zd?AkMFWSLz)8}QYQ;^`p@%~#n4GU}|xj!@j%#b9;W3dTu#8disH zVk$pJP^;Y=4fFabEnYqkb3Pd&o$8I>=G4#^g-yh~tb*RU!S&Rge?&J zW&SlH({zTe$X(2N1R9xDxf4-qb|tQR+=NZji>SFXH;2snLEpR>qm|;hbj7Q+Xkyt& z@0hPg!DMb`a5DqLLrXDARtS6d_E1)#k*fN~)9s?qNrHnNq&KrL@yR=g7)*hglG^Yv zJ{g>U zmc!m>!SoU5;k)&ygiX_1%uESi0~d>@v*%)GaSZ?QWO2YaC{pYvo+s?sn8plh*&j$N z4VTcVrMdK9xFc$F4s)EGMO9n%F)GI!g%w=z+Ot4RlU{(zB`YvW{1^%n;^^HuDtMNa zB$a+2*ngI>BuhaRmM?N80iRlV*?J*tXwL-bI;aoXrh7n`TlW>a?i>u&a;XEI?a=YkinZ|Y3^c}^!XILJX%7AR6IEk&;L04@@T5Q@NZ!Ofi#gOp;>)3Pzk;F z_wT#j_5SfYYn?ytx@+BY?!Nn;{p{y6t>UnIOgRx=FX(;IKh*kB8@<&{I4bv%BObqz z;})`?)?93-HmvC!nsg5jwbxJy-PxSF20Km|FP;+Hd-TpWZ_bGX6^@DFetzccSB|}9 z8ApllInn29#u5Cnl8$qxb0&^-x z8^Bt^Gdvc<<$ULR=rP{>e6JARl-Y-Q?7JyE&E@iZa%~CEJ|cs+d9E{0$~l53{Yr`V zrNM<;(sqz{?_(`@kiO*l_*WOto#lzj zUgeDzr188TXY!8Bo5}0WN#$LQtmF2Z?+u^3A2iv{_4@X!ruX>^swmD-H~srGJN`3ulgQ?{MV#iGQ!L@cMJIAb z<0;2<37@+j|3q&dZ#qB*7K6gZp{W3-UY zOH;{9*W?!JCy)#y`47it8yvd``U)LC}JfqY=buM=v!xA!XIz2~_X$R8F(iQjS|=6em-ti4)y^mh-gNpCh@Wo8#+~#Zj`)=g>YbC&9CWv-NyD z=lh~$PJzQqPJf;;N4?2~bLrJ?PK2ukr)boe({57F_sze{F)FI#tUlktxo|0sGiy!? zM~2JSXyy8HuFmeG3x`Un_R(aN7`@EBZm@@Y<69xu;r9pb(|gHW_rF_c+ssSUvQvz6 zsbvqh+ESE$i`mQ#aI)e0J=wv%@nEDsYvDfD@JBPWx2l<+p-xj>F;1cfH zLv}plK^0#1!6;6aUj!Yg*vY9dlIHB-a=cDT+_*%ZcEzPXfB`L~|awl$4&_eT|0YpHW~`Bc%sE2pRsKbI>R?P+E+ zi?;q1;Xdl#glm(x@`SVQaog2=dB@gl;Yp<p}zS zV>U?N81LZxK68{~RNc*aep<+U+-8h(yt0GypWP$Q_=`83+j*Bb`+L|N{mq*>dw1v1 zSycjD+1fh#c^SVZ(nsE)@KGLnpC7OO$ywg*4~;z6`3Jf%`f5}N?l1WwusYz|3!1; z_GNO)B0q9`qu+AES`#>YLu^h4Un>#cAVQP*ob>1lBhC-j6pnp3Ki`9$M+fw7(4|W) zIpcS>@xNmZx|vI1asG39$vlMyCfCudhf4IcvH`cp%#-^{B^^_YR`Irriu0yd2=h`d zeyo|VQ^CzC+|ApWANNaW3BAB|I*`SNxzslK$1 z2;MGdX-|eQ4J|p?k{^QkJC0**)*6W1_mxqM=_OO2USfsVtY)R&W-%gLguutYl;mAZ zB)iUW(EOhj&KF*eV;<}1t|OK-`n)2Q&zee|OC;zN=|&87&Vb9!Jh1GmhT}gZKswuz zoSG{_?u3p)d@`68VDP!n&4S<=D@Q+iv+gl4>Hg;e!w&=j_62ymo|A7z-@wZqK_|}WUlaJCky=D$_esJ)$M<25VM`dt%!E>l@XaNbgJJ9Q!fzib(*w$7H;lqnT z;_yy*$Fqa=N~`FJJLWVX(2z!*w*prmYuI190W^oD;N`c+EY4L4_I}ecjC+=gx7AMJ zxUm{6PmLpU@5PdK&A;&S@CdxxodX|4OxZ;>ex^E8HnCn^v4_V`wn5KsKNy~o3TG>} zKx&~U+#t_^x9thsf6@ze5xKy)=7B+a1Nc1+qTT%L33ky&D(EMR9rlTkr}v)BeH;K1 zhXmllpXFryR1H=(aWP?jDGnXUf-SPs!G7QZ3CnIl6YX{!yZ!^eEHZ`NJas7AEeiq} zYB(0AfQ~vpVde}kY;88cEC2n2Phw#hpCp76A`P&;Z#({zh{tcYLa=D|0(_@sjJ{$T zXxNzrOTSf-OsP)t-}y3#$-e}0bIagys34{unu)t(*WumKDBK(Bie*g;@X=~rQZiu9 z-gac1b!e|6&XwV7%6?d*;KfLidOet5-82v%^R5GzX;J<4p{j*9dEDB#B00> z;QiFY*!fO)@WcoSzVd>Emg zLaoN^u(+BLTRciqs-20yq%M>MIzrmJ^Q2|m5%#eHA^eRjSz}W| z-ZpW`^qr$1Ap95-%)i5tZ<%DUdnZ|<6-z#f?1G5M7`Qoc4c>12%xEg+lI&s$s2Lw1 zg0s4b*e)(nwRVK$vLT}XNCeJxhf=o&Px={r=+bU)vZCIUc$z*SDeDSBT4E-?HeZ8I zw+hK?RReNePQlp zk;=%Q>Lhn=J|a%6jqug!Hn_ArhvNI97PrvV=GID9UARI-y+TNQ zw=%RfAA`43#mE}j64JeQH7I^RfnM1!P=32O{it?=iGTE!o!@nWi8{{a&-Ekl>azd} z)hs14b|Xx+@d~nTo*g`nTMhS??|>Nx+sNU}Hu6SW5xnxOnQ1u6#JZj!=VMlrJF2au zDTRTp$Mz7bDi+cF$t4LtBFUpqi%Ht3H5se0C6n_D$lUR%;Cx*M&#ZU`)5=m|*Umg< z*d~$;+b)K)QSs;>Z~^~l^3%VPQ1f71IW-o+Gg9zjPR~y95bN9=PmaG(Nr-iarAU{~OsF0(4PcF1|au9*?IV1t;Zy?Af+p_Puo* z%UtIw)4uI7bDBBBc-*qZI@1*xHM$->JJjgYUqBPg7W4Of0a)n~iZ?ID;o{hnkOU#1 z=sgcSUo1hV^=oh>UK@iSwehjRY#5!Z2yX}2kewC=RfR9$W>_yi)KujAF0H0E-aDbI z*&o(iONDHC1cBN`;HRPr3%&|L_MjHX8{Z+1ZD!!JqCX(l-Vf8KZiCY|#L4I9pBZ=2 zJdz*POw`}E!u(yEz%qOXblV;WH?MF!WE+aC*w64(_9BRWl*8OVX4tZS3d%dGK#*Jk z>u~W6TqaY8`{&=pYmX#Qde1!Ex7iJ+TlB)BtsRgdGzb<4gwWIdEqw571vSto1s!&< zIZ_y1E|)>}!#47Eyq@VQ*i3Q+b}>%A&1Uw_dN`_PggrBj&^v4pyqCR&h{8*tZYx8X zP2FgsIuH9Kl}T|1kk$W9fpa%3X}|emYGt#OmduJIbDUJ*r$8E%gjs{e!C%bvF;U{- zl7=7u=A-o5F8ryyOP-QpsMzoOG9@WBfcihT^ zzWhW6gGKR+Za>I4-h`H%3$Wzp0_d|LAb(91^%wkv)ajKF(VIp;iJqi^`_pKV;1Eo8 znvX}MIXF)}oy=AGO+2*H_-k$*CY>C>O_5)4?94G{oBt>FsaKs$<6$!@n5s-C_DRt- zAtOw~J3F#Us+FuuEr!h83Qbx-Ue(RenLuD0%dp5J1oV zIVc)&95%(2L5H9e%C}_Td2T53K0D)w%t7c*=KpR_tnpgxC)D9Kp@&{Oe(65|!_ndx zs<{d|=Qlu>Tm#t_b&7Smz6OTkFG6>50>plBgMu}dkh@?byiG7>41}kfEslyeJHFNt zKl9dO9M1@sER13w7{1QFW0}D?-&A0NO5d<&I?iM}ZVY7-Kg$v2@x?@cqav;skjJ%D z0^^&Fsb;YW4YoCPcac~F*eSB&C~&0(-hs2VKX(jcsCGyJ?d8+;z9fQnrXOpYA^ zOA~XDC7)naObv5BYGHxFbgE*h!1q^}LCL3KqWEbctS`2Lvhw|)y1*SaeOL$aJ|<)d zW5n!UeUYuDSqz2R1u%B75PnI&V7|0fFu|k880j6Zbomm7s$5o}4JXdQh(aX{EU5$( zUX8jNMDU|=3oNhL4i|Pmg(P8NJY2Moj17npmwA%p+NbGwGI1u3nwq0*Q#3|s=U~c@ zOE_=N4{h7B;?p$cv&<$(GMd$<_S#{Yiu*LsH-{E(*)3HfWFUDylm zUQ@?_e=Bjx?0I-1G>$#EF^9Foh-G#+;5vLeT?v;JQ^0idHBfGhgPBbiAn%7AS}xg% zLUsF5;8Q1bueb?=yR#q$R?wT9t!OCU|H0Dq5Z-yY25;GDpuuu)T=_;HpYhdJ1sQGF z%^Sj#s(zXHcSb|{bG<9_7SHqWUHPOlaGZ?h} z2CJP#U|v@P8$DmcTdm{xStu6;xYaoCViJB>n2FPG#v(EP!^C!5nXbt?%R1oqj2I;B z0E?ZcLH#=5qB9xkMGt1z;~3H#bdNmnzso2#I*S?011glRC81J&lQY0OjT8(iu+bboj3fJ-y{MK6B^c{kM78S7M3(X1syD zZN4Dm%Ecy^BD6J3!nP9@%>Bwpa?orIJpGr&fi<}rub30OZI~-MHeDR2wKlUWHur3G(kidc? zBqUb|-~AOpw)zKfAGgI_e{}F=><0+u{lWUi0knPa3R|X|!VR#1#o{aB@B7De*5)dD zhMmLrExr${4AQ~MdmD5X=_BMVz)!#StymI)v2vcbLA5GT9Lu|bw%oaHpS*7!Z9vM08@0qi&$m!lUHT7u;^F? z*;L)YoC|r6nng|Mvh5`vG9!4;KpQ>%WwGuYzY}^dhWT4Hk4ewFOO8KyNQQ=Qkdih9 zMsHA<7%eO&N+-6`aoZ!*_h%G!|CWu5ujS%Gi!|&ND1y4sA}~6b28-iVp=!xgkdt?W zU$alb<=rt*-(>{jzH{-dkUnm%RmWZ#LcR7_&_BYIf49Kpahra&4Xig|B+QQ7*Lzo4JHw`RJGlfp6oHF(`s@c zbSvSAgvS$(`wIvfzl9+4sFAsQZF>H3`Org543tU9qpex z;pT>HvSpx)DgAqyT|Sjd@*XTENsniM+_QM%E5L%djb~xc(L30s){ZykJVmp=zsP8X zA_NAlfNOmsG}=>?rgO(|vG#r_H;e?q`ZMrb{W#3>|?!n6= zby_{-pQ#kwdS^jgG6dl6)L8(1+sV`XcILs|V@!CmE*#qC0O?7=ps-RIrS@dN4Czzg z`Zf&l#6lFPvcPcd^YFCGKpwiwYL+&^9HWRLh1C%%3DEw~{6$5L(6e>F!_h1sr#*f%-KwK=WNSYz$9> z0iT<2tGE&3+W6Y%j7E^8_p$%Ubv&QmfGv)TG4^O6?)h{Ax$jjm^rIw(na;qGo}1`C zk%NjYl^A@W1P)hLfTyY;MtrNlcU@Vya`FOx*m4AellGzoI|_SM*1(*~$0X;;E3ztJ z0c7oV0C_JjSpCkQvDT4*GW&(F&`O;g)Y-nk&bG*`HFh_i9(O@jHo+B+GaCP^Xb4Wqt?3%jEf`MRe+v_vmD-rq8II>-12EWI zmD>I@r|&GR=|spNTrKPb$zN5FETalHEdq(}R3WmU{~6oDyMjGCFwN|u)n(#g0?_4{ z3~?F{%o>)l+5VPBjJD=e{Gc<5%aSB0_ox(hmb`}3y^YYcN*jM!RKnUdJHY;77BO@|Cw$f>Y4A@(p@{q&c=9r{=glNqA5aq7*eZ~xSyAZI!$4?bVeHfX^KRbVM9<{ zHwP?zg`g>G0mg1vjvqr@FuMWS0WVyMw(oqnI^t^TBmBrLD(ojK$eRVzYJ_0HIb|@d zzX=;v&x3~HC2)_u3+FYe$VV#;vTIi?bY!K&j#=Sgd{vHy?-rq}^apXfb`sp2n+VFL zkzk*7h@@JxNu?uyO$oe>1)Rqi`n?VVI)})u4@cq8md~WA$r67Un`7iJNetXHmASJ1 z0n69>JbTNxY$8&;2@Yr-gv66k#O+QCF>W|a^k#pAgdlU=d1Vtei~S_~KKqdcpPQN5 z%r&$$+k#fd7|^dC;@Efo7I?fl2Sx%K$T>9=Z5$=hb4Z-tK_R-&{u@euc*UH!vl(1t z&w{emV5^%FIu#c%r~DO7MkeFgat4QSSMo{hCx`G)IYSbX<`U-vilp!Q z7)&WOL;a@pXz3DzhNVSVzq}buog9!JX~k=eLewRF4;++R2z7tfg2S&0W;2boe109*RMOZlo)w$is-q0I2vCVEdn?EWNAeNaTPtjz(mVOhqra z?Oh2$@e{o%|&6tJZ#5#L|50>^Y^GP?INc1=$JrPYEU@>m1c)UE@6^-}U=;UuXz zR|g5*Q}OwGU6QOj4R-xL0R_wNK=Nx5V8$a!SL+R6T5G}i*cW&o;6n8uYg4xunw0t! zGb07Uq@qh3s>A2vd8s0dIr0}L_vnJKyE>Uj34j+xr%?Iy1r(fe9)DixBPK~L#5v*< zm@8yquHy)dmIQ%E+ig5MvjYrY-Xr@*my?QpmhijkB}n%VfMPQjxL@BuxAA^-k~)n; zHXV3l`)_hv=_Oee*GiNdMc{bS75G~&itjWsNWnQ9aM=A7l;3^Ft!kb$zaoY{d-jNUS4lpE|fvD1hgrvvB4#LJoB33xTLU!|y6+BipJ)@q z^?zCRub(oG2a3pQopVrWUP+v5r$Lg=3!dD}4H5V${|=;a^waTI&*gb|fDCW*kPR zPFJk?o`~_B2{^h^9{-)QK}FXHxbx&Uk(_^nE|>xmBV97bTORZp{i1DnfGjxD1pWR>Tb_WH3xs8=V5rkYiVT z;Y7z9n7Tj|a~(c|@|R!mW4jW0`{NC(M^~Txn^F%6BbQ)g>=(>yPDeeLa@_jg8Li4dC-TZrr5grK*9;lyEg6jKPD-@+d-VH_GccXC6tyeGD7Dk017B@u zocB`@iu?s%O7(I0&jHY>x&l*fUxc-H3Q+xBHZq+VnBuY;{4{jI$a?{d9+QHrfw6GG zR}jDDX2A)+6=;9=7D_tD!r2Qg@NTmJ>T#0c>B);AR8S9>H0MCfA3la*rvsDoACf$G z0vVT*A>^tK%uD(~Ou{VSuvh@<|EYn%ZW~zX-2plue?it#L8PFJ+ZT)B_YVwK{R{`I ze^YQUZ2{((JqIa^7a(Oyu*Ja#<7!W$z|K5eyebc_wn^edz!q#iC5H}HGx49^Y7D47 z1-9XhAn5c9#zVzHCsKygdC!DTiipRKYhml6F?gNvnrxN3K-OpF6Wu~lkO>n5f#NWz zoHhzmm;8d*sCO{iQkcHa(4<{D#x!0}kTETq&P=#XGLeJI)YngkzVy(c7k?Wwmqo5K zqVtZD-mM*Eyx|53>okVhnlmxKI|K(fSJ5U}g?)A343cuHhD7zq!IO^Nz}(M*71w96 zSN%IcyqxUe&Y7p|r@Aie&gGR%%PM{iH+Ta^-sOYJ@&-6JdK+?n8)CN97YOsH2hO2> z*tdK;_5Kk_g{n`}*>40ucXK|hw|We5E>q#MRU7d83P@8QJ7tF8b6-T;n&^0 z_)I{6%Kn}~Jy&*-@hp8Xiu8iF#TsZ=Zh$Kmtw%k9LA06o9T#5zi;QVC)3`+nSjTc< zz5Hi9Dj-bjho$MxSxum^lLtGZWpQIRfQ;Ts*za@&!gj{O-e=h$`lbt-yL6alOF1Gm zX-5wJ$bgSCxKMFl7%!C>^%y)x)M-y#ew;6 z>pY?+wt?jLj4;!+Zm^eaOJl5UUZ7KmFcn&+L0{}`gOjz=cqVr(y4%!|<@P^`cKuq2 z4)lPrl@YM8Fbv`hWWl0mGX!4_hn_$DdUI_T94HpVlu}Xbay7?2TsJJS5Q5&0R#M;Z|HPr-6aBP141g8lE!*l4s86)H`zIOhz}`X>XHRaP)7_yiF@x*Lpl{0C(v zYWP@R9mRDwph#sYsoPvbp89i0XyG@OZ`l;aE0PimWm|I1XbW?`={J+R_bI;L{}azz zOViEv`pk@OEz*>FisVQhK|k+6{OO#DKc9VP#`UAfdbfw14Zyqw^y-DPd zV+J@|Zi5Y{{b7spC=@pJg4EjA;Idhsw3Nn^QX@I|-lK%aI&^T(nh;ENxB`AFe?sj_ zF|2L{Sf`K$D>praH*>aNe%V&MYvPKADl>`9heFbPNdzt)^`qOY{ptQWL9}hQ3@UA2 zji!fPF|utNS?b$B{`&kN^L2#q$DPkmnKi_(3DWV`^-^5?%54tRF{ zFb*z_!2`ciuyCXew|MZixz~sA!EiXtT74eMmh?dQd4F8O+JKrnSD?u|LA<~?;0DQb z9BFn03-yDbmmUJ2625{@`vMR#pI{6wc{3k&47282C^AbvEhKGMabdp17 z8X3d&cEEC1g{3bwVV_eFvAUN>!p91Upa;cw6Us>Jr{nxlU#Nb^-+#3%0(}8-e7$fk zT;;|v*^@z-y>1;kPt3;ssw0GU50j~1BO!WAGO9hxLgTj?=v0@EF$=S??}-)OTq=Yb z>m{%^RSwgGn;>^ZEr{MOgQLlrpnP`)zdjl#V(7+ZPdsO)$r-{knFJ=eEtZjNU&nMg z2!QD`OClzu%GB5-!0g;~khDvNUdIfq9kfI@yFNI^RfJo<^`y#6nyhe}L7ROQ=`4E@ z`g_)EvM?rwSnk|Fj{jXtQhL;xOY;{pk!Bev$_d8T@`)IC`7Hb0jZ2JZoB`<;*$y@u z{5AD<*gwTVm^|Xt8BhM}YeLhC(^UyCV$`3iR$^^(JJUKY%eSVp+DL1;A0MxqIQXyeB(pXFHXZ)ohcaj z`zMtBS_&s}BVj0r1LUEe1o$C-4ODj1A3FJsLx_Gx?f0y_JkI|294|R`I#}q z(-1f(+Dx>*3y`w$=Xh#kC06ugphc4e@fjLpGDiO~M&GZa-Oe+3>t`sAtVp1KqPu8w z*an(Hy|7c#5g!$9MBHr)edVR3FD{8_2jnt1xQqR9o*i3MGL!Ew6N5p^15o0)Hd{~j z6yvYIp1HfBfOhh|X!Sni)9!0wq)Dxc3^c@&<+WO%F|ZOox9))Koy)N&Mh$=8{so@T zwm|tdCvaBlBSK$mF{UpYt22YqLq(tN9TKLtwnMmW*GCl2>9|SBCZmP4y+xPSY?(^Sl>3NP@k3%}a*@2a zV-Jx=(Xgb`9%e21h7nCKaN>G3npU3$U1KM3ie5`Lx(nc(16&Z_-2(Q;vso{mrLvzd z?=&mFlm>bWEa0)tZ=&QXgs*20!10$g@O7&yz3MMW7mq%~C?5qpeB=hq9tj7L1r^kD zQx46QN}~9(h+XDzn+Wcj#lOxc_qh8_Pgf?4o3hFn#r?6wPBR&?7Ko8Eg& z5p4!-L20HRC4~93F*qPYrU^ zPLm{AA7d^~jbffwi7*nkr6Z1Bt9lQ?7*k2OspczF9EEP6N}gIXoAuE`Cu%nU$6NC$L>h2dcL zR4DzD2S-N#V_%%~WV7cRGsA3mD41IekH)*;-!fC?TI6TOr|1s3yfPjF4LJO{mI8Zw zuaNs)=_Dt}ob2h)r{|_g)59^tD7?Rjxkc<5Epu^ZZbdb-a8)tm^CF#`J9rLXZQ^S` zD_l{~GuX`C`Vw=p^%{9JErZFH7lr%jzOYqFlO%ukB{E5T9!awWkvKVvD41;`vw3P@ zB)JSG)BccK^HOm1%4Qs?GQl8~1UMy~13&6)!RReR4N5GjOc_IObzLU{yBZiNI-MDx zZv`I@CxV{MTd3HMkX0T?M!xPLLO1Q;JCJwABU&jt5?+KFR+k+fwrntuP0 zNnhT5hd&PWV{U6dZt~>!MO6AAeR3S;U++Ood4Rir+{dsh7op<37i{>l2QurPL!IMN zJUb&4Cz3;`{zMdIhlkVPTo3$J9)}4tV{ka;Fp#zDU`L!Yh%fCWM!#N>OX_m`;OS-9 zC2#}--$r0*-xi49n+`vk*FteE(oKhr>Cz|z+8x5zI^@Y?cgrMLNF<_Kc@av+Jj6S1 z=iyq_h1e|bfK{&_!kE4WzB#Ogjm_6_&F6e zrjnjhljO|djWBir;MB(zz?z+a>tCheriCed+4vu`MSNVJNGt@4Krj5QmWU00aoBI< zMy9+6HYz}+CV{>)^RV&99}t|@3=Wb-s3g^Z_byz=rMj)C zW&H;`;1@Q}ZO0jnJt*r_g|}xs!cplvIK!hGeYc9yjhl6-fCt0pA+LvG*)6cSq!msl z9H*y0M^O3W2k1TZn*h760v!thr*jKnS6CKI=6wLm_p|WMl4$r+-wEw|<}s-XO@y7E zO$z<;Aa*&A^q%ws9hKSigvTsOe;U!D$~62IgUEil8(UtKkeP9F$ytd9L?tMYeQ|RL zBl9(sUC>$td&JMdGQBpi@3*0AJyz51+b;A`S16upzlO&= znCaJz8T+2$j2{t)<`O&l@N#T-XbxOn<7Irt=y*d`eB{mm3<-V{M< zI^M#(zIlYHR5?JD*M7ovo_|np{V>jaaE3h*#V{g<_sCwyc9{9S0S+7}hbPb6&~N@e zyxts+p8rY05y9I;CU}xWekmiey-UG#TOypWd&|`AT|?X-zhZtVz9l~EUXuLZwAx_V^VzuXkPHSmdC(Wrc2Y^MKW!Y;$LQ7rn7S$n ze*~_8nWvq>rY{YSdPTwl@mnDN9P#4-pWkq66KK8Cf?G}%^po2)s>|nhPZzC$^P^9o zy|EECMk`@|v>4joR>z?(gq}lQK>mCJ!}nitp7eX%6*Gd{AI!k_y3_E<@-2AvMKd0s z-iIaUzoYdYW%~KP3_TjIPGi5j!use0ShiFY^VCi0`-|qZu6G%|q^E?L4IC`^;EE#d zVKCDmmb~&egZ$JGNKNho8)+vTE%akg+bh6cHRQ;!8ywiW#Eo6+ybk79E70qkuHcTq z7_>QkkNv~oCb_W53NH)3W$9&pW*caHB8AuHk)Z$fF#m1vX8&NW!>%j)!Dnd#G}T|n zsmIcA(yIawUt!}Bc`GcE+=)TXZ<*jW1){9#4YF~YFz~n_?!NsQWL|qQg3|X%_?mP` z2yeiur~uq^`##L_Ttx9{H$M2Z92dNe#S|4gw2>4*%ebTD%*1Z8KqD43-siISeiUPL zQ!kV0M>$xNrGzi?B(Za`KeJRYnJIeBf=YoUR9tlyf2^9(e2zS{$@h>Y3%kkBw1ZIc zhzo_y*4WL*9*Ki47|)pwxeeEd)0NfWJ*bO8A|arhV**_f2+wEzK;Eirm=@)Vtet{% zS>XfxaWn!|S0e<}Pl43ctH5U2OgwQf0EYj&PX=p+k+oV4&$$BT52vxx)#8|FB_SBu z5dmzTFZAAD4+jcolGS&1k|IAN*y|7qC47E|t(76=l`aNli9yo&wU$ZcZoy%Gzi{Gm zD1NE9M?$YHB)p^nM&a}W!c_`{647#q?AnPh_b))_MStMC;90g#r~y-TPKn62{DqbJ z$6!{y0WrOJgXO;2jrGV`hiLxt2NmWwVO@O5a5RHpW$Fg-G){si=8y2{+Nr zsKTtr8qlU#1h-d&(qvr+de2}56)r7@g`1Z_n80BeiNcE$2>n@ucOvLlr?y*fqh)}Nvhxtj1iW|Y_*>a zy=QI!y$DsDy!;pz<;LMY_B`Cbd@lAEFTnVcD74O)h6;+J7~oaHf6oKCU%YXXaW#?J zyo{92YbD_wqSV6Y6n3~Q!bhX8!Bg7{q*g1ym~1k0z-=WvvdbEDbQH*<7f0FFs+-7` z#!t*4CoWsNy`A0I`;xtJmLR)IMv(BGdT_1oE!?`O5u>*S63^sP=0L{-BG__oLDMeo zg4+YqX5LGKNZofA^03~5*c=*#AA0X0bIV=e>`|l%L!$Jc=x;RHZHs2!2FRJGf~kcT zpuRnuC_QG#g$o=!D=Um`0?qLDOd&xDTQX)CM|@s(K>mM@(3Kks@+>~r>}4G0o1aCl zTrX@iECLAM3Hu!O64`zI%-!4-Ow79t*W<`|>8IH@Xn> z@9yMP;9hVVsU_PBmBH`E8PL%W1JMiXz}}4#k+c3p<;Hx7$?ODs+XAo?I|UE=d>O@u zOBlbav&fd~8}Z#Sf~#$`@nG#L+!H(<(+d@F%6>i%X{{|^x2J)d2L|9#z+Jdva0PCz zSp;f2bKxWGgfn78Osa`9dGXDkoG<7!GwLW}l{}APP02bz{5(&RGZvqSGCII!D=k>Q zN)LopnVpwPoH3rvzqEm5E>C9VO)g}c+T39O z-gS|Q(YntTogdA}ocRPz)?Ivzu@gjsSD+DyVM*WvNY=@QhA$gH#9s}zmO4{L{>v)3sZ0((@~rSU+Y(*!_TsXf zdGxm{K98xnyBsS_VE(48r1^uGyiJqo3 zBmFF#@on-a zA*g&Vh0GK!6ws|j+y4Zpnc*y2|2GV`UcQJCMX&JwtV<+2d^fE26hPD66-3~R04TmY z1lteYBFXE@;bWp4rm;RjY>*@=FV znidG#RtLb~#xTt72*h5cmFRK&JZw5228;T*pyIp-#4S3RQi(v~SC;`o_fLTIENT4D ztr*HBieT=>F*tc637X?hkP~tO)ao#y~f-v<-xUYqENac zjtSb-%5uJYgIV%IfmHBunGgIt=!a&zK{)F=_?u2b`h_1LH(7%oa)NZHxD<`nxJ!(V zEGLmu&oKK82#S}PAR9x3cVy~ube{2 z_+mxqTcw2AqkInZ({}K&*gyiL55UGla%fob1*Jd!#f9^fXwcekhE$IW_6t;H>M(4VzY;_ zSDK4r#S}A~1ToGj34&A3z_?2x%o(xABXR2Zy+IWFQ;g7SngImYahofaNlZiI_fH(=2h zK99gAm5jgfX5McVTTp+`kacR;LY9-cCR=*@Oh(3OI{CL$7zbN~aQWKDz)!lMq1RgQ zvUnnDo;ry?6_(>_r52blP{qei@1XP563~_|0`H9b@X|;P84QIinPjp)O`k1%k_T4D zs=?awKFIh!#Ny+*cz?!Gd~LiR_Fbumq@H)6A{PxBhh<>lxFwnIHAeDghBLEU^30M< zF2F&Ni{SI`E}ZV*u&iPgmhpXxyc1XRYf?*S+O7-1qbUFx)y&dq zHtg0;11y!5LKBO>kR4%27!6^jyf~U^e&K`m@7=K{ZxL?MKZNE-GSNRR8TmAHi2CsX ztigrPm+pjZt_85hx)4G|6>#0)bC8%?4|!7MEX!T{nTX5~(!+o5-ph|Ovte$qx`uQ> zT>B0gb5I}`f1iUzS9gHMU>{jzVuVp=%wfN~J#k?n?CB=3lp+kydw|soO0jGF1b%Z% zgtfjtuxAqkwqIn))PPoIM_(dy{aiVu_1=LkwGUv%^#oE8w1$LqEh70hmO^uj3Rr#k zN51VdW~KEPntf(nG5giJ4Ne5d!jbw+IKS8qJp7NqrucAv25kr24VLDQmp&#^8{D-{ z7{@O}fWnw6?OG^JCu)b$eSRdA;0T$&D4pz;;%7Zx(1cr8UJ_HHidsEt=yO^F=O>;Z zm$zRezpYP`Y>Ox8RWDE3=S*lGO(UVBQbdv4!myl*!Qrw#yj|8xZ2KfYB6dH#d;T0s z?T)~=6kRAEGXaNLi?Aqk5w0|!iqA~vV(yle*pjsct#k)S@THHiqQ)6_8qWbA>w9F+ z_82}^l8s;g{D-d^rRdu1F}QN-35+aH2BUR;#CO=7xwlx`tX0aN&skgnqjS6fodiMV z*iT0Bvl$tDQ9vFxW|IEtQ^?&{*D=Q86n;Mwhv)6mz?|Ppxu13zR9M&9S`X&3dcT&j zjZDOtxze}T@;!!Z!RPX@rK*--*?IC0ThQfS6Kdb8!<+j#s3@m}iC?CnyUSf-rE-Hz zW|Wa}wQg{BzYJM-uE0{~jpV-gN>WpwO^$E1M~?@cuwI}XV8Mq)T4`QnaNSJXEqvUdg$qAZ8uB-OJhut<1J9jPA9N9(YWOxzMir1!DJVVmaS0OjG7BGI92kJ(Z5Opw?ik=@K_qGk> zQal}3J<>phoXPmN_W^_`1;X8iATWAnMI{$ysc)?)X*Hz6r)E!B_&Wi@jy=Lp-99)c zTo=WMF7d*o_Ve02q?zEihjA>0P$NzlbC+47>OxyA`=Ei&??Xr=(TSR8SyK1eMVOdr zjAyo)wn4{Wq3HCtOP0R_j+@vz)E(#W1mTHK6&prMVq@Nt;Ax=Rwn9#1sgpf<*Q*v#qw zGABSTHV357yn%T8G`#X?3eL}T0uw=bHdd^Kp6)B4)w-grqT2_uHhoXOB;JCE*>MoI zJ$;rew$8h4}@5fw7zM!ofjgk8m3C zsu*mGS&mT=+W4mE4S9U2q1SE)XwuPiaw+$x8I#57&a4BNbIcjdCJ-)9EunzXRO+h> zqUagB(7%5eTfJA)Hno4e2Tt>OfAlXfo4nq^j^@YE29xn;OFt+%hQT%!2PhhJ;k=he za1WRFY`Bn#p^wiXv-TwF+>ardwyhK@Z9;yhJaKB3Ckm!IV*I8`@Keu&*DiV}=J1Ei zN*gIhNd@oe`H-R{OKy&LaP1~X6s^5#-u7)doP1LOJO4D|m*YkF`FbH% z2;IjEFSD?|UjlOuog<&77;-k=O9>_C!MJ5E@Ez@CejZX~2M)cVw8!|@vIVVOp z{|Zxfml^$@YC{3}MijZenqL0CLF-~mNqWj|G_ID$V*>4P`Rqzo^N&57rews<92mfc zUCU_liD-IjqDiav4PZs!I~bY^&{N6=vb8V~N%l_TBX@7z9RcDdCJ@*dNyi(asAxq1 zUH_dxq5sLzRzn3UcaNkks}|AVVIXUVdHB}-1eDfHhG-a}8sGou8K=o$mm@rxGr$b> zWifr~aTM`>4ds4VL2=zu?9)GA>Cx2@nivyBEwWK`N%s(ah?)%bDQ!&d<@HQ!@eogP z;zORl+(q8ZtOUw^3i_ z-2TiL6y_*HGwTkYpU=hK6^n6OfCsY9)Pd!|O%E+hcg@=s}!98*ucJ2!`@6zGixZhcI#!x=K~aq!WnV`Vm14Znaul2R_n>~y3(nI`Fg_y!6W=evw*nkbqHjH_|&m~nb z_*I>%1$1dgngi8kZ!lkLa>tx^YC2CQIvq#TJ@H$#3Kq*pA@h*Ck4y-{Kt3_{ieL-1 zNEMUbLSO9JCX3xd4`E7nBYWNZ8XGQ~%c}iN=GwZy}nN?F#ah~#geD5BMXNQAm zdSD<;joU<@Dp_<%HbrgeF#K#&$3~qmVhz8a#iogn>dvLsiRgvw3#F<|x! z^Mgf#%`ktO>39=*Y z!mPZ}D+umGl=g{4%`6GJ8z4`t=S2EEZ4LJA3dM;ZL?;D`E3*o;OG5q3%`1+JC#@T)V7yj?CB4!fyEPV$`N!wBN zKpK8))yIs}PsnN4UHTx^L2o-&qD0pTYumLI>=i=jS)i{vYgL^E* zXu;tPG-1P9YVKS|XZ^TcqDvh9C^&{zVu4I?h?@Cm?`Cst{fF4Fh@Y;$;iF%dbg`r6 zIcR9A;ryE}LeByGWtoYlPb_$Pm))5$LuF>z zJrwsHo`W$H*WxR$omiTD8Z|%Hgf8Vl9PS~(KEXY4}}jZ_SJCd2mjYOzv_ z_1J#pK-z1&ma;OJl6I#$S=}_EVP9+d8SF#{*E-UTds}J2(I<@U;|gfqUB#3Y4^Ujx z3tAe|PqI>j;B~hHW=@@q-TE5jv0(;{WmuB3Z7LX6K830fT3xs5h4fIVV{34)0X`ax4gW=>uvtavvyTrG^MyqP{AD- znIJ_*2a~C2@HQ3NeW8Mbjd<(CH`r6bo$rSmVGkPv&yJmii2vl7E!)Bwk%n02+?XI9 z75W47%zlDL$PJpSew8kS+EH0=35spuG8J#HVt{)FJpHbSuTD>gvFb9M-<%63R?={9 za5lDB^@E$xAfx*H9kkS|lZDS78WA)h)tfam+qaX{wy5I3CK~ zDz=tcZN*2LX$E+7R}Xq}`(pj^7CfXGiY~v)QN$w=4IdYf^xjm|KW2t0BUVt8^8%eK z_35129_rd+iKVM6aHFR%xvvdF%YRmAtM?!L6BEX>_IL67t)D1lHj(Dt(xqGPFVTx} zUGSfOo;g`=3*r*zK%@5nu=U)3(a|2#$J!bDRfRwjk>vC63wHI^(yR62a8x@8x_xp% z^GpQQp0y_a;#LID&&;(gv%qu3Qg~>59E=el+F}J-c8Jo(Pe3gjHE4tvV4e{1)_lPj zKX0dO6U=ctjqwVx=ECWa&pM2sE5W_)CA$Tjp6x`rOQ+(D%SotTScvVv zYq6g#Lhl#@-1|O~mIclr7o|*cF66@@3o9Iw=i&0))A(#McSio&h!%D`DEjyky8i7V zogcUYlXhQ(gpE9mtyxB{1w?yWIEF>KFV;!C$I{{&bhkSJ?m1!*?75%w_*PLKbC#xU zNv1pMDNwTU08hq|%dq#i;;!+}_`X?;st@&1f59*6m6^gm6iFs;E|+O~<|oaTaDfGm zWia{TC0Oj%#_ax7#}xdMf=q2^ZjJ%^Qe;L)KWeZFtuxt}{ma;yZ5wIEd0(oWyMd0Y z3zG9zb8_D@fi7hB()QIa=|){AZ4>5l1Gck>DK(>$X&Y$e?3i~GD%1atRk9Mas;CQ}>ZswCALbaA&*h=QrV`hoh&T4tU}N!5(AsW` z#jlQ`_3nq9|9U-LtW={L7LS;nOE*G)+G@~B3!n|Rs7G*ZBS-?DLW!zoPZ zxPi7c9JjJ^Hmm#Gf<0<+f!XhtYc}h#k$L}jQ&#Yd9lKRvGh58_W1eIwFt;&}ar@hh zw+j2P-S0lC&SA-Wq891&6P3jknm71NhYgxTjO0;WEGlz@g+-}M^!aJb)+j5m7rf2{ zwU;wi-|U#O{AJ9n$CVh(Ws;Or7t)LG&D0`xhx}$frU_$Fbhjmk(i2Zoze_rs`96*f zUzfyA&^UvBp{MYs@d(BSbYh>)Qe1zm8dkJ>(z^+Mv?^m6byP*+kKL2deU~qu`Bum- z%_?D+-?+iX|GWqm_0?eXj}KQr79{P40{r#)1|A&cSU@SP`R3A#dfd* zrO~Ya(+-HWk;R=?%W<@ODc*c}2Dfp~+`T_b!Q^5)Gb2xfQE}YGTVk)tbl$$qTOuw2 zK6^~f+}1BQ6Occ>T)3EI7Eh~{#bi<+>Hz-P^ zf?V(AQu(~^j8=^h^IX4z_tF0(y!$#GXJ(gSv-o^Ec4aS(JIB(ew>orf5zt9{c8Fr=8NAIuzM{GiEngyo!U{TW1Eiqm#xM>4MQ+`_BKQp=Yfy16!FiLCn;$I zstA}uN484S-X(wWYsNNkZOLU8Wib#UAV+nX-1ERjjJo~m7)hJI=DI!NJmqUAscB{+ z74L|p>WZz@(;Z3zKSQYIKs0?TO(K@_ovqa*iax`P zy!>UJyCIzSKw>ga^U^W1q}3v3LNRN(ET;oTJ(!A<^9nF}O#=29Wuol&38ZQ=4Rh92 zz!INi#<^!6uYS9Z`PZ$IY_ygnTd_i%)ws!zxc>ztnVbbVS07Lqz&aA&pV-gDWu)_RSEn+KQ@5CV1sa(7JD~Zp zCTaR7s6dSz_sG=nB>lHCo9r7e5-OD7#3Ny7<>HO@4JTOLs5mz0R2XZN;D@u<`{EFH zXL=WJ1~=>bI8UA|h&oo&$f+xor+%5Pk2f&U6E86Ok4>QRcLL{`OJxhB(%9v#F--W4 z$uPXPlxcTff-$ldm|gY+!mQ8Hnwpzb_M(yUci7XnBTFda>`FQ{MV0<=x{{aKKm4_7 z5*yT^#FpzxvA@E8Q2EtARBa{Bp89CPWfs(Nfhj)*#Li}y%DAwTc)Qq|OYgAWx)eq2 zvhazJ6u95iH1l!|GZXAsNo@)SWWk$*OVV-~&soRJ*W@#>{);euzI_ly>O+_=G^UX4 zBD9F}Tqkv0qm3MI$da4OPwhMbou5;Axt9cZ=|}G~mmV2I=FJe03rNMMp5qw)PMZvu zIk3Gu16jR#57tTYBTY~EK`GCkli}MtShDUTPM!V%)z{v~y_!lSxq3S7sk)8NewSd; z(=2p1+shnyR?aNDxE{UNrIVdy9eoWOpyVAQaQ1^B+MCv)|Asu28xZDt3F>iIpcbBU z6~pLlk6>%jp6yfA>I}r3EC1l-4mI>&ZiBmi z7t)8uEwoTokfsUlK%)(h&>(jjz0=}!<0v`&Zt;Q1Sf@gX8@ADU(E{4|;Tg{JokfO` z+v&ui>pZW>OOR`K4R_r%r}`0&kMdzQ9ea2Wy_appx4L1-x2gi8x>ex$&oHLO6w&VX z8)n9^sQH8|lEem2B)+IFlW+JeimpnoACgK)cKz znEsz0W=9UdqNq_&anJ{&j~ZNF`xC>j13ZB&U(mi@1I`Y+&?4a=-dUNAs*bVvW9J?G zldeY%Uy(%JZ0LQ}OxmA>tbfpYR?IPwy)DX5`!+}5W5KhqLF+xH{N}^svlXGRD3dm* z6p_)L6uSB4tl9XwbG*O%mqLGTA`N7}pnY#wk79iMeH?z3j8(Qxuu3D|%pi$lAuSQ6 zrK_Iftc*P7v{*G`w>%QmT#w?Vwg@Z_K7^~~vuK{ob=v&oB^AUulIzAvRP|4ge*f1C ziY7lG^3oKP_}I!kxgi7NN}6C3aU1>DX5wP2^Vr`tosC*=%HHRA90!EB{nnIO)VRx( z^hLdKw8jICj6yJjr$^>Xoyf_18BKTQLgxy{@JgisYUnpG9{b-z@}(>+s`JOqRcaJB zevB4Z7vU+{tJoOYguKaVoIaV2N3_zgdtEj(tro(d+Ba}{rVK9nVvC=9%J5!$C#)N} z3?-+(Lf4x+6z4fimRl#VgByiV%J@61sZ~G$RRfN<_Z9m_w_s!?=YL**0Rz|E!55le z(BW_i*79A&TOy8Rc)^ZTKH3sD$A4dS{zO^Peta1Ml7*+U2W8{HwRmdhIa5G4UQVTw?C$MC&e1g#Q!Xg#Z)tr~vC z#xCNscpB1->U#yLoa?!78s>J_i%((AzX*)`B@Xk_I$@gE4!nEOlNKpkk!sLX8Vbn7 z?H}{deNQ3&Uc8xlxl$=fr9?{Ysv*To9aJ&eOQB8P&~JT(VOMfzIsYsAI52^om8;G! zq;jl!^#v#7ak;P#Z=4b$h{t`KVA{Y3bj{4+axu4X%WPX_?{jrt(zOfBUiV2%SX(DE z=V}_fh?HZKH~yuf)gS1a*evF5WDuM@@&UdMcrquSB{Q>Fdw3khF{d;h#LG&Iy4Pnt!Fji|S=8Q(ObVcy8jGCFTwcP(O=?v^x@)Z8WG-5Jw%z~^+ zJoDr}K~Xii{JTJkQr94KP6))f9su#T``MIk8&+qE8v87P50~$lk3yq?cri|Zx_8Nt z!E;?wj*p?*KL-h)y3(!WiFhdaAE-Su#E6nc_JGPwc180Qc3kueDpX5Q0Ox#i-E;_} zwi~0338|4RYHoK)_2nVg1a<#`fU<# z9i|)Nz2wb*pQev52J*JnZYYVdTBAQ{HNBua@CB{PKjX4*Q%F;~ou~1smv{fS z9*jg=6LaM<&gb&Ns#5KYjCvxlhlU_eO&HbAZ|1#-$>rszEMT{F&1bK@W7)TMQ|SF> z6%=#d#W;U$G}pB9Fng?5X;v(}gr_V2$^59j*iEMv(f% zt>mb@l1JY7< z0WMR2A!sMIm5HF;f8o%&x&Z7)=V7d8BQIxv98(x73Q6~ecn>b8np?bmX#Vr#QZncA z9doP2Xy;#J+G<{kQj1K`c3m;;UZhG(^xZLGM1Xq+kAXwnbGSZkz=mt9ve%UV(6N7U zJl?x!yxFx03~jmzf;Pfv@>3fnMfxyG{xR0>HXx0by>Kzz46h#7M@u6vN3zbE!gICg z0iToEwH7_zv6d~ou-qcXpT7-Wc4**=OEc+eP8v-+f0g2tN-1#0eX47$r|+Epr8&Hx z;vLjz_P#%C?}A45#g-JdpWlb4*twHu_)8lGrcNZ2yRDpO@Db|OY$4xs+SD%fntKLo zVL_k~eoytqrHkjWYM(;bScL+%IWr%QztqNuezP%pz9fZmJ2b^#!sNkn%sl%Ybkdh4 zgNEIdatp}pvIc$F5kw7Tn<;cRHz#Sj*x-2(LM&CP0`d2sk#=junNd+)`JqiRR z*8*SmKGt_f7`xdjiT%jSXLu_Tm=M=Mm^6oDLGHcJ{D??1E3CGJfriUWLw5mlbag(u z)Vky0tr{5hr~|q^zJqYYe=sGUk5&KBLt?ut$X?A6^QG>?kM?0WaJUjSKJS8^&Q1x5P8aWq~p1==Jxn;gZh>B_RN*!lc1 zy6RS9w&px;|KA7UoBx8g^9?3m<}L5g+8*<9{x=vSU5IYyLhvo06B_^3M4Pgw@Y*nj zy|X%){rVt)b@ZBoE;k)fz&HS>cSbT=dQQweKT$?WMHEwfY9W(jEWTA(g)wsb5!w^6 zrfLq^-8Ul%Wrkv8R+6HCGdVm{qrvz#+Iss2MV6nYlcjR>DN~Z%w_U_n34M6PwQ z;H8L1E&Z9>rze0`qcmtU;&icbA<5rcLvJsM(kkWy+Bt>dE@NGb{9lR|t5n9>iy4lT z*W^Y&rXC}?{l2jNv@A?HQ_I`vzm(3OSVRKLw$SnDO{8?#pE~x$liT^zP@B>UCmRle z(8v#JzkiobzCTA{2N+gl{tWiTk11@+{;ABh>aby(!lzs*I-(3)1 zs!eZp&7|sddlH+|0v&NF@aNJc@KLRVpOZU4qqQ8Y?016wBWv(73xpoC?O?Fg53Uvu zF-8-G*cB#&l%v~26He|U-hQIz=O$CljucEYv&V(*-{9#=hH2Ff<5_ecgLZ?7I27&$ zMZZcJ&RB_^*Tb-TTO=+mtff0^)2P8Tnof1TgtZR;;3zzTHOkTG2R!U?9fx@y1I&!0 zzRY%AEwI|mNh2NwuxFVy;NC9y8u9}YHmPGem%Wh~eN4LFKhXGyE4MQPyfkwf{{B=0 z4za5stS=Qpp7N7oz)NfszK{07$B@z8h|SBx(R!~kv14+yz~Tc2wJo8YXT+#L`Xfef zZR9u-5%?nbEEaJrY904@OcIFa7_xtGR)jD`w#m?x@{PD)x+89mOTZHj(!eSeFs*{? znQtM$Gcws`-niC*nQf~Aeo7|H?~;v--G4uDQbaHMu6>99@}eN1pcF2LorebtnkmLE zhq{}&c`7R$bA19aljE5=KqQlO&y-PmH-!=I3MJ{NRn+|5g|bCj&F5bVXEHXm^3>@y zqpzz58mTT&tn~)c`2K+hN9lEb`T}~Bf5G`Rk{HBB#o~`)#!XfV&540Li1=^uQA3yo{d+eSK(zR0Vw(y z3Vc@g!KakthB_2rnqWC9FKfYkpL+Z%lZ0kxbf_-JoP=vU$u8&9kRHgb;GPAL_K>BWa2nvoesg-Q{m))>TWxoFJ%IkAf;Qxi@J>Z){GKML#T z_~HZ?Ssa=#fKf|kqpZn6_#r95{3p{&NBM@yH*|m&O!lG845Y3@@qn&r<&jsD!s@0$c~ z>iO}yO*%z=Izc+iPSK*Nd8FZRBzF13UO{FvoFnLIn$BS50=xHszNeL*iO@y z%qH~_4eES&34Z)ZfgQC*_^I_weTRWzVH-2eg`0)%B`#GJNzTz0P*ga$_ zH;S>-(=uolmsy$FJ_`a@9tP(@n zHqVpz#LyOJgug)kRyi8mq({q6OlQl~%-G5SYc^3Um^{DjqSt4D9(kgks5p%8^#jTt@ z$_0FVA0=Y2k}GET=%BRqYMS3LlPbEENnoBJjcjN{#knc?+kF?MCCO6UCSm$0a*Wjf zETnc*UAm_CiPrhPp`RpL(T6FByh^b%q0Ye9CCgCEQPAet@b z7=4ukvq|*CXZ&IJ1lq%EZiNG1-4C8X z^aI{ar;n7zu?`~56j_^)60DC|iuTi0@shYH7+lR`My+HS9Y;e@aLi%&?>aHlO2p>WV726DX=+}7UdtId0U>qzjfvu^G6kp4z15H z3_^ndm!*)n4!Q2)w1k^=OxrjN>`cX4FVIf05`4lLxI1F8UJW#{n6gJ7E;&-1TC~Vq_6BckxI^?(;rTjGB(Z&6? zo9UuzD-9e;#@B)Kk>|S*W#9ionbYr3-s2rsuUthpRb6RcKhn^?Lb_jjh)hmyp+#e- zpvcaR5ebav#b1-64FMulTBAa0!goQvy9fqcG+}nzecCWwLjMY~=$xww7AZR80?8#P zCaXhU57(3Y^#k-#uMRTa3*tx3U$FE`Bz`=ThT2nKp^y6j$L`^yY1@S;Ep8eE^eMkBkip!Phz`EwE%gf_)dWtY;TL^&A(V1u-hrt{y@II^p!IBA#}g zpGhjc%G6e6%`rPG@Z)L+R4V)nN+C;FpLRPoAjX1yIrkdIzDdTO?jVeh`iV07XYjg6 zAsWkvaGG>B@;m>;v;79N%7^pUZi%DBKS|V|%`qsC$CBW6A)0sP9mdU^M4fw5=)Ct0 zy72WeQ9=NeKbQ)I?=*1o&J5D;YoPyzFVN=_7yOqUg|F(XaMQ#?sB+2}dpyKYW>Exg zyAq7IFNxFKz2dMbViP!QECylw?Kn0m0zb{wq6If{7+<9&D6;4})o8_FR%^Qy8lxA4XzJ0yCbJfvN!xxa3J6Y`L!iSAOxqEU)Wep=AX{>+&Ee;4~E2 zGw>})4_hV$(fGgB?CA52?9_f&Hh7Tpq*yn=P4~N?Z@Unc@c@2wn@vnqD(qS(4*{2l zm;;5M(J7Bd`Vm5GE<`}nU>jr}wFTm`Ehc(5@li%Jb_PCyXHvnCb9E8CS+CEq!$FLu zekyM`NtdFH`KjZt1lc|37^=JTh{+zKbiIFYd&WbUAsPZprUdi!1;6uLY!1V_zqY(C ztv;Sk>0jRYdBvqkswW9oVmg~j#tt1r&Gr0SYj}7B#mu;^7ya69iF`$H06*iK~ z)s=ip#8?wAR4HMS>c%z|&SoD+RpM~M1dOjKVRn=!LBsup@K*T~Ggu;r>8~Vk<$_uW;@Ceo zws7+>TYwWptXW?(WJ@N^W(Ay2(9!AyBGUxwZN9_IthHtyFF3&TDYa9bRz1xRC?p#{ zH`qLTIy`MugU65R$!n0~JWRHuE}?vO?flbho^dkE)SV)=Lw5AykQBA4a6TJlZ`|Y{ z4xbbAh15Jr7NxiiCAker+lFn|YA* z+y;5G!HhdNu5AU!`W0Kf{;@dd9KNw%eYT~0OF?_0%|e zC}~T-JGJrEr*E)5cr$o+e=+Uvcxm?S;8R}XCLZsz;Z_JUkw7Nu34S_XMUNEa*nnkH z_)nknr5L5q#T}vMO7Ct%{+{#bv;HcRFfk0Y)HU&Wj5@ZNHgo!4KNh(fF!y52nH6hn zxolAebL{ph2u;3+7x@gBXHmD!%v4MmA!7sl`y&H!)&NRO@&-58nV9fZ3bP+b@*eG< z%Z#=DVN8=^u!x&~^tcX;-|JNAEZT7l^5Zlk=sZtcUKi-HJYL$bPr;BtbNrp@Lx&(! zm$Vad6)jMqVH4d8SE9RRpP|2g19RWx3e3>ChM^m!f!VkRq)x8HBI||bQya6*I*;Y^ z5^rw?=k?n$R`@Q?QC7mqO64eUNQl})BChf&)7PsfGItwJfY=ZNE1F*8lcoD9$0&)J zV5$%4A`MXg-zb(pUxOeN43@xg8)x(RxIWK{UsTyJQI>I37U=;I2Mgs!k)KKg8UnbRb z9!@ss!Po2F;F%eI6udQ$zWCf&?=UPthXg%atm9^x`d_0au<>(Q9`3*}wP z%)jR3@gl847@s}nxL#ZyrLR23SIei8-pClvHLAuU(RH|CXDv46+(x5a(|8#JG0ani z4B$&Br^h7+=?0Wx_bqo`aaAM;IDW)+{Da^Tz7DrI#bWPGV_GFYk(yWNa?D9bj`Q{g z6teFyC-Y5tGaPi8u%ZOmr*?qjZ2iRPXYI*iP814$J%uOlhhS6fVJdPwOg)pW$)xu# z^LX_VQ2&?&2d=L`zPn5DUBXwebk(3d>1b3vqK93%8%VrmF@+z}CcgI_Ff!Z=bv&XL=lRe?k%FVAa9Aog{@X3pn3uTGD)>Xk7sdW%W*TLzO4%8JsWAwK!<2gFi^X3%zGI%TR%YC^?VMqOI1GQ1tC;w02d(4F|e$c3cH+ZXZQQkCUXd zJA!79a;&|@i6qD$KvK?T^enB1+}{4C`EmTL>*yuM>*Y%L1kd2!#(H{@F3a}%+OYCB zuVH?D9xhi1!7Q$;%EWyb-~U&L={7pNi)~>{WgN%W;kZ;*sX??TVjq3Cj3=7|DU>`d zp7^dkK;IAFQFBy}q`7Rc*GL3jEX>6Ctu=V>%3V~|5aTj6W$?(C50msdxn9C5?v7PK zxnXxH;BE*m*`SYcx9$QhP@)?~^KiOQ8ql$?P~Kb&zRRw_*_rOhoSp-F{Nv26A1=hP z>MB$_qe<2xi>SUX979SQnT-jP&F7r?YZkN3&-7e%8obwwfLEGfuu5_@5_ApV~Hnj-z$$*cS-&0)r{b)_BcFAF8%(I2R_tP(4aOQOG-0lbb{ z$Rt#);-%!o!{0k{Xe>Gx-#_$Y4(-h46@DqMt0 zd}n@}JG&);+Tw-q$2c7R$lSsDs5Jcbk-^aB0LplO1$%zggCMgFn-*E(g7ZemHa?{G zSC>hBu$MGhl~mpC-~qySb#D zxsOa%3zFx%9PB$Iial+>o_K7?{x;BOEf#Kru$p1U{@gXj{^~K5f9{3_o`>-#$41qi z<3qZUbIHG1j_bqw3qfQ2c!=`?e60aCDR3>j%EA!O&ie=BgXdwh&M4%_od((TrI2+& zow}Y1a=K0@cI{k?C-;ZrM!C%>6edZkuC=&b_z?P^nntCbc2t`aOfx%<(#xz8nz*Es zVh^szcUATn7N&;Rt#Zh#`3ZfM{6jh?ZAqtZ2F=^~41f0+p`5w~guIG05BVTSFNM$G z79}2D$nNGPyME;zwXK2vtMYi({R7iIqs?5`$d790Eu^?JTEuJ6qfDC!jQ<@4|BAjb z=6x#6n$S`6arYiGL;o#I;=C=ebKwOPtMEsyZb1w?F_Vp(HIMxjFqieN=%P2X25AH5 ze|om_0>(c7f?+~kIPdQa)cmp%H_OaMOZ%(H(@(_ItYug-i(ytOK7@H^uHvV@Fp^C# zqgm-~B&YNZT2!usjO$YH4okw#@9(1S-VAi(59ex4oX~2=CD1+j%)BmZh#9_B2fNIs zVn*N$c+uL;6ks{&98acJmtgw&Ga26v)L^&gD;(2UL3x9M^lD!)_c?6EV<-F3z}JL? zukJ*lv~gyi<|9U+=MRns+0dpQF7w!R3STV{qQVX{IxF*vXXa)PKAi^WKN5)tx%|_q zg@q{2b+i4tD@`XOr_-pS6Nc_Izf``SE-X@-xLZ_jYn~-Tnsz(1+QhP(v=OI-)WjGO<6riuiPI{ z!AKb?_qfxjy)p6qdx^Hzo5}cq0`br3!IG^W6lW4jHtMN#b@DuOpRi55?!0DLwjqf` zo|n^|7lN#>V5b%Efa89!j-(Z-bT+7e|*oK zpY!~~=<&of`upr4IUB^#i>qOz{Wlx)bw7ZUp9kKM&BX%83KWd)K;?c_dK-L%2ESb) zJ*_!pTE3ZnaLfWL_wSJYeT=bJ(ldKh%rGxvm2h)>5boI(3*A4Qp+O`L_*L@3e~u0G ze_z5qw<3w(#gL}PM|0hbb8!)uv-;F6iAypj!9&eVX27BpgeHZ;OoI+aN2G~2nrsZ= z6E8C#Cpy4R6EkYd*iX&R%BjP82j291hQZf8$o$w|rm^ESW3);K`M?YA2^unX`KNgc zXKBLp#eh#{$KmQbtB7BdW2&kQ!n5KJ^h&ja&Q*nz+MMm!cG3?vmCk1hnx8Sf50=60 zmG)S_-i$fGbqH4#X~AHl7xra;gsFbJKnyHUtZ6$mGE#6P`U^!}{X(uyw@B7Y5u|SY z$LMh!{$h(T{K#phk0)r;qw74JvpfR)OI|=@!yY`w?T%B0YhYU8FuZOG2U|ZmlsIt& zC(F;miMCQ0qIMIvi<_YTIvH+PNRWK78U(fmqu$dw)Tp6HX{uu!4>}Bk;!`njqzeUB z_<~s0L0Io0ho8Jxpha#dHeHNBo9By|(`yQuOV_hO&vFcPTm5MCeIBX0r9t>x33w4= z4pr}Wqhw<=?w5UwcdPDz+L8w_qfnUZ{D?q12*DUHTTaLHC8NFals|p~XO916hVzfX zfgU4fmbE_m$G?M#K9S(R;S4P& zkjo6Ef5sVY^5oXYF}QvCank(XkiICN#^tnnW z?d1p#Z;_(mLyqR%y}$_7-C|0Qmf%=I6^1xSlIS!^u7`9QDgUgdrYi1UKIcA#rt!1d z0_WgUi7bA|K7nmoC(ORq^_jZpweiG0sGzIX3XJnk#*1x+)VU^+lm|Mf^Q9bSn|^~~ zt8F+PmV@EfQV5d%27HEZ@xGxMY3A*syT9FF%9Hz`c565OFfL%sE?6)rd&=PGQVsYd zW(?a>&*8!GR_1ihY}h|z8uln|M6aoJnD)AhJ9BRU@AY*^9{K}mj^}vG3iz1hPo?nm zKqhE@!g==oe6}K@L}{Q>iD=L~C>p;q z6Ur22CRC;(O@>MultdCjq(K@LO)Ba>zd!Cj_pWuNgRd|yQwL}7b*F8M%jwt?aqu#%fRlb8P7mvkhSeL|XhsQz1^aTr#8Zq6-UuRv z)pu#?p3kr|pc@3|N8ky$%&%!~(!Ep|@_c$oO6eWCv1L01wrqyYY$3cpw*`7yLS2D0&f>iK3j&A<|UBBDZA;^GB@}uE{p+} z_mM_#O}_Is6KdBkHD6F4E=X=aN*eElaN+5cTHZ5({c_*QM4`_V?)rhnoguE-aVjVD zUkWD?6HI3ZSJJV`pQuvn9I#S#1!;Buy?M42{O6Q_;zqt7-yMSC!B6mXmNaublS;x; z?$I4O51=I2hrDq(PBq#$!uW-CFg;ruSFfFoQ+`O36&KD6F#J5(^;C@Msd+wm@9UQ4A1PMe+POX#Wxys?o0DC;N3W*KAf2S zs-{o9FT&;vfgo|{6!gE+#Wxl&VT;ypI_a|wEzxlULrFf*T#!tgMUFt{vPbl_P&h5E zyao9-t&r->vjlsT@QM2DO8%L2)jC(|Z@GiMR*0c?lP1z_jVDP+emrD4eFsMk zaqJkF#Imc#vgNfUcwhVqZg`f9BH2qY!>kB?`Lw~tr$O+`Qyl6S-lp%ju7ar(5}@)w z4TukxB}>;({I0kg#$7uAIfsjpzi51xvmKRd)nMZzchD{GfOYG5mrKe9IF|XIgx_hS z{hDfYOTt)U!(~IEn%FVM-3fqTLuYW< zt&SOz{_y>#4?WKQ!mAZRsGM>c;vb}fA|z8Q2X}hvXz zQ78(tKl9mvh)kH$kPREkB=Pd(E+}3d08V-t*mUkSjO)^Zb55?XNXZ6Td1m&nj!euo z&A{%_0q8vVj!Ma#p|>|Jp?52bsND)fD)VT`_HQXYGkKZK2o zCiCY5Zz1%?ODJ$tfZgqHsjqiEPK)WnnMq=->--j6EiHpOH9olYUbTvorMO>*Zu1mCseGio7 ztW;V~G--k2bmF6+Le{2~5mDnJNLlt6%H#e*M)r5CA1%YW`88wZ)I<1OF&*7~&T^bD;lWJNo8uKFM%gd#(*EsBHt$_!`3?f#=M4~ zc%i_C@2{#syn`}yUf&I=e3vcZ({&oYtb{%~)I!GvPoeYYnc~K%ZrCv=9_H-iU00q; z^x=Iy>hWbFE;_shYtw^q#A-Fn%eV-8FII!0hb69kJp#45>%sCx5~vPj!Kj@{P@`1= zM})@1ak*^TD_R6LI#u91ycP7p9?$Iu3=^A*n{+FHTIs=wFcS#Bu?oJ&FN7cSDcrpN z5N`hb2{(j9a1GA`mOBnyb3PwY=>kT60;A41V%33%)Dn(x0a;77&j!Ay2`hF%@1!u8)3_gOQ+YRNR0h0WxEr>;7Ke=1-_-QaD*CK_Ig7br%i^a@ zX6dfY;JY~iIz3mwckg7FbN)66kFJ69!6$K`GzUdap2L&8!)o(cS>Eq>l32eyfJG(q z@ZxShckKTXVu=E#jID)J{QInu(*j2Emq67^1b_ZD!bzq+Xm%(Fr&b%{H1ZfWO!@_n z8%0npn%`6P>te6kOl&G#hI6;RfP^b!!3M{{XQS0n{MQABWCP)Bc_qlk-vN#B6tk8N=k%&?k^6>VdLaZ!W4~v|tVcOCv2$qb+;tRYlAs`Yfm9in{ zSt=Y&$$;kL!Vr0Pgt}%1g2LSyD5`QCZ!NuwA!3z~UL=R*cV%&SLJX|!%K*uQoltWd^7lfp%+&>bs{`?9 zQWCxkl*6j&8aQd3DaO51#YsH>WLe!TEIEAulIz-G&T&<=`cw{zOPb-QPY?8UThe!X zE$L-A#w9n#!Ox{Npw~9YchsujT*?K$^K=b9mHoqe1z&O2esT76_8`q1G=-DAzu~*q zRJQ7>DKk%)&Bp4PLf+a?*j{rAda9D)giay+I$HuiJXT^?!ajWbc^`gB_n;mfnN+r_ zpJuHyhQ%9Jz@xzc@Xr5AO*NYM`Nakry-tPu)*wbFPv;r_c1qC7&n>?Em4Ji!Z7}uM z0E{^*j1Qbvfm7r`*qa#v>%Q$_slJgcS2l@t^Z5t;+- z6}GOqgI~lN@Y1RjVmmvVDxdg4ZKpV~9-+DHWAhT$<24IAf4X6BzXQfg<kZOdt=s+#>qH)+!RfOoYTN+Twc~NrcrD56MSe$t=6FSL9G=PC zD}|Vn?|by0wF!PIPlcu#{Z#vG5Eb=ErFX*fz(Zc0)PMNL*^Y^)vm)K$*US-m%Q}{> z3EvO$Gbg|-VFBbg>|(X50qo!IaHc=Ki*j?v!oJX3^ucsFEa?hBjfe(5^S=s|7k;1y zb2Q+&l^m+`XCJ37@_zV4VNl@W>Hd&Qw0_72>*g-Tf!BL*+A9M%<{(FQKA%YFH8Gg< zRSx_v2;ig39;}UI=&{WhXTQ$G?SD@vJ`#neJbS|=Tm_f%x&D27>}XE)VmfN*I&?I5 zU~6I}p354=eBH+3ftNOzeRMId$lHQ5)`nu_^;awo*~J%{UAA}Hrl4gU?wVAEc6>}iyS9=;QAxm6ex<3n+w(Gm1} zkceyl2;uihMp(?#Y=@c(Ah)XonDZG>Nnkjomc#t#;i#%TAC|w;11pPp@U*;u_b8M? z$cGf3>B0MD0`|eiRfpj9mNgKxuoW^^?*OrfYccKy!xPh_F=1IM*{>=^T^35gO1Bgk z^E?wu>?7f*wkf-}(}E?QGCG zeOJ)Kg7-A6T@2oud`|eq8`!W)9L?^l;TD5i_zfe@3#2lr2$f{V_;Pcg9c?q{H!qvkJy>uxdmCUX~F}L`=W+f6<_h;-`{w|Nti|O zvxfQk=2TfVfvP|7!`puXQF~?}ULUuPdS2a1-E(rNNbfIDe?|+)twPr?eMrZ4<=Kz#QCFgq3%N{jCZer84UziP45S@ z9ikZVv6}AZd$`swgrMr}T-5Lm!O0Ru`0M@}*gc~hVsu1t+ZI!HaH<27ui3 zQy?oYNx*1NJ%~`s0G-?CA?|e*w3yE29@{PE{35N1pX4=~@}Py@*wjInPE*6hlEAYY z7NJ~vIsSN*kGbDZ7~!MqDCjeVYUf_2-(vWd4>K+**Yq!Rm4`Px3h+ph zJR9~kV(IdX?VDFi-DeoXvCdVH?Kz(B$B$<#(TKIDTqG<1#ZdEL{yYAz5xy;Jgkx?u zKv<*z=64QJsWFxGeAx(bZmpotV(xQ`wrKJU<{BgjIY+;_@xCJxQH)X0S9J`o# z7z-N1@XgG4thBwaEEcz;ji)5fCCZgOnJIJl5p=0j+BO}X-=zsr$sMSx7-oG3HPu^~X znswWOrj4drpO2BNyGBUbi{JcgavaP4&zN;PETvCJU7(s}sx+_QBx-%#fDi6iV&}Ya z*l_wbgywvML(;Zz|92K`DpZ2}yV3ZQ#-oAzG2Y9u4wJOD@rLI^cxb4HXOiB=&f}3d zM)e5X33G*_`JQlO^k*25Q9u009BVdsE#rLm*!PuU{fxkk4w1f`$|~Ytcra> zKjF!rHz1JBh2=wO5ca(g7aU!U(XO+wVjziTemXo6Myj5yf|)1e@dA-U#p#-K zZ&?|!3Y$;<%gZD$yYy-Mj4(bQ;eurgT=C6M51f4a7zSQBiaEAP2r^#o!$Wab;m7ex7z7pKg4B^0S-p#iRn0WXcUW$uHuaCj_M(P3D zG*;s$em&*;cO3p6r+^LqGcd>B1r;v49VV8nf!l8P;ktVn6@KnQ)c?~UiyS2+W9BOE6^a!S`rmVrqv$9h zAxTO8um8(KNob~!r_b7roBY;KTj#mQa~dIn>89Lt!5kYDTKWI=@qfP1|E}ZN?b^%# z`}=>-|EKHV{_odeW@g4;hw1-hia_07$Y{C|3V(GB-u+Ia&Mo~XiNo#vq1BpWQe zse~tW3b5qA7?cVM#oUqhJrrV96NKo{i*!1vx|j}bx{fQ? z1~S2d64u)Pl<(c%!h6*vc*r;t_Dm|r)gQa@n2sD0Lt|W{coM%%YlR(Gdm)==yVl4B zp+oOSl!^a_y_)tks?-an{*8h|ytn=3Ng!=!J^5ABT=4mH78VT$K}hX(dax`4`a%p* zsws}mX^Uaas>j&d<9^V#COfBa`Z z|8Oi@5IGZ_%Zu>oa#{AtJ_M`Z@@`Gh9r(9-E7nO{pvDPz9G+te+aJq-Q_LfpocSGo zXWaqEnLI19*9VN!9U(?V4VQ>G!`J8uIL3brUYWCxNja`ys~6fcr`smb@%AbB$v#AL z{>*OA8=j@*wI7f2j0qWsK=k0WS-~h}Dk@n@$?rgVgm>w!-~1B(c1=LJ{JnsiR^eV} zb*8X526o)vj^iYg(6rqZZhWvB(=he&Z@4zH4mW3g#oyxfXps?&_$z=|te8s_ zH%C!N*FpGt@;)s2r46HPH_<85QgH0eSlnlu2a#_!56ItGx7TOXgs*q0T-4kl8BMgx|o009fKlEv2N@Y%o&WtZ42U&8~XtNY#$Gf6SSaGW(q9-(*TAQ3sA@2 z7YjEN>{=3oRgeC|>8p>@!|2xlAocsZxWwB*@&~E7vh3UA0#<>Xz_Cl zQbRB7lH7{JiVc|5EW@6d>oASL5^DWen0FIe;(MJ;x@F@iI32nXnVkq`&iILL;)zV8 z&xrYYJ!e}NePE}xPt!Sx`f&MkH?`dxgduV56YF9a!amHN&R7$l>-sEj z$-6_fnBzVO{KWAU(=oCyA($FCWs+0&SBaUpJKyK7qjs&ciPjIkSD$&2$gZ>| zYCb;n=Hd))!pk)C6*`G9HeQnERYlV$_e!v1zAom~zN2~J(`j?!IPSIa7!FFm)0F6J zvi=Z9v_s5US9m2j8(kwd>L2KAlb<9@?Hql#tC}eM`@>oGj3q-|eQ5CCSd_}jCl`H_ zXlR@>nP23>?dttaWnmiq<>E*BXK7>4+r4lgR3AQ{dqS7(q9p8z7SX8=2V0X@#6I^n zjfgt~h1n@|*^JTjT8fOy?H9ZVilw*tN%5u;d2%{^Gt-JT zMFmq0lHPR<%+?1(-2|T9H{&j=_DRJhHsVz4>ozPuR|28hkLkLm61ucpnEHQ~qMn7t z=yPK#RL##KPX2-D-R1@cPY+SA8{Tx_b2cgDVz_UoGcf3-Fie?5sOyFKaHnlQwQDHi z8Xn$eZvFA-ks?Lsn>5l2pLyJzy#0a;A7|6oLtfkxJq_;QAz2VM>YxL?4)oiSCI~on zmqzj&_ksyCY2W6xoXvzJ?tSbah-WkqugK*@IYt+grWn(VNj~J|i6u(WwihJ;IRpp!?(U=tXe(bO`0g|FAwnhfGRk z#JPvx3k(F&6}h{);HN(z&DIVU@Y&t>;~jZ^t_dA6-NM~p`H~vPiP7Fs?!yGEWDf$PJFjoflU7*?nc!$`gck> zG-SrnRXft?B4_^m>TVajEO|@QqBisMfkf_s%vf$p`!jsGX-=+C)M`9M24>Lc04e;(4c?TFqA= z`6qlh?^8;wWZHSi)_lc9eD%W*2VYU?jMpU8=_)+XSD-BqHDP_f2iW`lAamzW61WW8 z!oH->bbri7TK=jA=JNe=FZ*G#a^_pAHO`IuaUqdxEl-CFxmp0;6XDyje0pW@Oz0yz}q&+9`}Hn*uc@{~*SD5N|(m;XYdUk&nZpAhSCPu9=T!ReEpO z_JUIUc02%m=c!;+eGTytN`;q_892=+j6REP05hEj_)0q#>SrZ_u&oqI3QIwVb{DSE zy2DD17&#DM1D#{`;;7N;WcAZIWJlL2Vi+Du7M*xQF2=v1tMq;N-(Q{qBXb5jweBHZ zbO$@GnX)k=eD^*#9oBvt#sBX2Lq&@ayXy0biA~_WN&N`L-}z3OTOe!;Ud$wCt6FrY z1X4Xd|FNBCoB4lCfL!kmxR^c`?~eXTHkCCI;WHMz2k!^;KmP`+GpFNIn`C(VFBBZ| z?Lfld7gl!uK~J9ny!t7Eif$eYD&+(hejWw6?~QrC?g^wXGto@wGVe(u%wedQyyLw| zp4z)`-RN)_O8gBs4#@t`%SFVNULpRUKGDw}%i%?m6Pz@21?{N)WJ_H#%y)W>>PLq; z?!rRwk~)s#t7M`7k3L2Zw4uSsAmBfWDLqFKa2^?~Eu%+1wt|tEGqvPhM|tMC%#y#q zbWS-+eXW;Z&n7*{-7W=dh&jqlk^~+5Oq}qwi>>Fo>LpD=tU9ipCEX5&DFL|{WLd>d z3zxyik~D}}91KfO5c;T&XB)~~q-sN|1SUIB;Ta;lOJW?m60Oc$cBr!3N0o?Kejn*B ztOt*UNpODY489Vv1?Ej#MDboDRg1I0S#y4pmH}%z>b*A7(be>ZWgg$6kDw0{t`QHr z8zg6%JjgaC({-{nbX!#h*si++YHvdE-$pC8OLH0Z-eixDcK*XlPNUI9-;h0c*Uo0A zZzo%}lz^~gA%_1I(6|*tM72Rc^BXK+-%(evQl1YD6U(UUmpW>7LIhPuZ>O7gE7Kx= zK6$V15ZQBX59Z3-GxxqmIN+m*+s3z{=YR}0ju*od3kKnm-89a>V?TEF%w~$cmUQ4v zGOl*?U@0f^sY8qqtoyMADp&4+Gja!T-yuz==WT)?^o4L_9`FCVGm(h>;m=2XcA|Lt zb|%(7hgpo0W_q4M=-%mqiqWnZUy#JP_A0;uG8aXs*uq?!Z;-a*G3I|fM}K-*Fg z=gF4weOh&F+4Tam+WwLuz9U|DPJmvP9n@83gv<-~0!3|e%srHi=S;$w#Nn&_fD6VvhX*FPu)yps{GMma zy2seE9@RCE>&cAlq#^}p%&qH*xg zG!h1ud4g|P7N5&T=G$(@vNWc!ar;AY^SlGNUJ!!TKMYve(gBnyeS*RZ$K#1PGch)N z6|PLXhU45k(J>+lPkZNKc|Z}~Dow}2MF~_i{yg>7&;Y~IaK8HLIp!8W0tf!Qh6|m;5IN--n-iYEOlBpq@+Fq6 zu6YJ?c{hXgMVGzJrb8O!&uqodohY3;T$TDG+Zw6!5&16u!HyhkvK9gQ|wHd_UO$G-Vs8!@?3W;^t5Apb?0FQHMER zb6~>PM7AOzg1M&!v+`Rr*zg(wGqyKmf7$}rZJt|rexe=goIBbgXREx0z+KKlv^^ZZ zH5#I8A@7l>9b`?4_u1RvD)wy@&m1=mNBzztSemN~hMT6tjgPZHTIUXO+C?aQE*~?8 z8*tULe2m<35?77(hQmGq{OmIy#J(BA%eB*C?z~ly9MKM5hd)7FbOD^woQatczIfjw z2xrvTgTfgN$hWhE-`a&V%={WndN>wl&hLj319`k_+7E?x33y~xDo%TQ17`>7fdAA2 zI>y2roV#>zr|ePCI^G9^z7O%8<0%Z!O~-p_$y7?93;ymHlU&(NN8ImU~J1Ko?meWkxT5 z2!{h-AJQFsZtBR7Qdp+D0z^f(!kDis$jht&K^oH{W0sDGslnIj+zV=u{B|=Pv_l$s z>i{j2I0F?gtKd-l6Uh599yi{dirdv@;MS?L$TP7+WVxs(-9BCe6M2T?su2aO+7pM% z#x3MocC)b|BnbkSU4j&)BanEV_q~g2qLiXLHYBCMlX)rdRxAayHLJ1Z&qe&WCmR>l z-DNzyfIUun%Fe}hW5nfL{CTPsgK9p~f=O>^;fq{a@oFor{@p^$wH#osg9aEqkb~^q z>)_Pn9&EF}hsMP`Q>Hxu#l=qI8nr^4Sg#95FDZdm;?ZP4y>T#hUA4LkY&~epQ zI@lge{Ss!tyed0f=2nRU`6+BzPX#1~I;ie97h1v3FIz4CQj%rl!c{9A0hlj54=~7#X7lEcz^6Pu31%#$vzC;SWE`T5+CRc`pd0c zDnb+sro#EHVdT{X3wqC91I|x20j1V*c+~HK{T`_}bSe)68lPidv=k(7_ycn^lF;V; zBHWZajGjxZ*_8qVpgIR&_i=H2BBq5qid5O@E@#$bltyP26vBb>1*klz0(-y)^>Xmh*bn-{T3|410$R&d?+qVVnTi2PQgmNrA-_-F2NuQEP;#XZUh_=tpw1j( z)3YB!jhtbR`Bl2eP!iTy`~uZatEkSJnbh&{Kk6-}Oh*sBCDr~?V7xz)dO0-H%7!zr zxIq!uUoJ$Eor~C4T?^1&*aTz$dj>|XCTL3^VjFK{s`v%)^28X7j9!aV)WzXwo))zn zbfbrS-cXT~N!0JM4e+O3a5OL){FJvNH-Qg`KK0|>$J*fhOA+cXs^H9_@9-w>KKP~c z-n*V0{3&k$=Xy@i{=q=F6*3ww3_gIo(qhPd?FtnoVz}X^GX7VUKwVe#(A`F%FkQ+M zer=nFscSFdK-OH)5W7Z?b)N$FPdAC1u_Noe$BSCb+e?mu#G74ZfmhVASk);JVlXTlZf@*R$@_VB;=m zPE$j-6$YTfubuB^sbS7o~j31Xiyx=I1Aq(V-?8de@rZQ3)5^da{iZr1Wx9$;ss9-e=^7Z6Ep6v5y`pF(4h! zFOqM{QFLRpG3l~RCEMmK0{=hF+>vVmWYn=_`eUh``IOmS=BX)$oUfY!bqRh*H%^kp ztgT&C*Gm`*Y($~=;4{I-q$!+N@5hA;FHF4hcn(q%-(h;>`qH)I( z)|bn2H?`MrO;=tKW2HXwW$s1ts$PpWZxJSCOJ|U&vNFVQ@g`bxZwd9ju1P=M*}$C< ztQU;)SxAokJ-{8EX3cGwB1X49pTR9?av;+mn}K858>)6f4tF0$0vkD3Gok`Aw^b+(t3pGV=NUWO@ zjdO{ItrD6jFSQF(nRulPBf`G97iQ;%emY2e>e;MiBx&#DDK;)BZBKSuH?|! z+tmNoZ@TKU0l05lN0j$wa~0i>iNlu{0;?m+T&&Uv@j8BntF4pdrfqmi>P`v~ZIxl} zL2@9KD4S2h{1ZsIXA$udb>&_QHE@%!UM9+q9Lb;SNE|z@$hPMZBd3 zq4P0ZW5yZo`A=oKBFlj+iwGqXWsXv-d6itJ-9=(H$_?1KgLM0nOnTGk7qzg8fTVRt zpz&xZh|Bv65_iXvjY^)dK9rE*k)`x?UKQMs9YfpN+~|wPpXt4Qp>(15JG$)mL#RD` z3L+1cfOAR`C@575-i}t}?BdsutFF>`cMU&ZKc|I9-TuHe+mA4Q`d28e{UI<9{vr6( z7)FLpy3zMaA=I(DhMK3ll5S8S`&QUe@be+rpI;LD3FgqgGJ>2o`^inYW=dvfPA76p zY`79lHR3*N6U{kfK$EKGfwM;>-K#eZr0zuGs1s6xlMlxT7CgHtP=239J6$eN<#c(_ zx7Q`&-!G7s^)G3E#8K|hpF#f)T#Uo zoqDi_sxA3LT8dWCWwQH0`auBKDqYGsIa$)s!A7or%{Y>}Mwx1S(IFozmXORHvDDqC zic9ZE=HBP@aT*S1Xj%1l`s3VX5W6)2O@9VKO@|A78i?Wyt&Pdr#qmTqs)ft2o-3Fw zeNgac%{}VBpp{5Bn3IP2`tTz60zEHlL>-kH>4GyFG&tfZxmYm+DmE2TWvW7L-)(?L zHoT=DDM~N>3M7J&caZD8o&Wt-1ZjAJ%GNHpO=3KzAG*eA_I~G-mR;hk z{f)@cmBk#)YZa_fki<1#eCfXBzU0#CRM0!Q5H_X?z$aZE4Hf@FfyPZ(7I1-glC7eO z+h@~B{H*D9;9{sTe#mDD?hpgIorpzUAfZ?9!_y~x)=X_G&j89KVJGi#Dks;G)_rf_ z{pxLC(kKJBF6zMcJb9R>tqtAg4{+MqLgqYW9MejsZOS5{Zu0O(CxGT9ucMcof8YQojzNMWpf)%S6!RO&iPtx{i28mi z|I>rN=Y1kkuiudWM8e52{V$wDqbEJz*GBFq9wLidHd0~RHuAx=nrN?>jb52$FfWJC zNq5AMznLevb&D5qHS-ugEB!?rh4jg(&yiS~A&+;=F2SJ->tWk&KHK$bEPU>}4sYf7 zP89FM$x;tW$rL{! z5}B09cMxrf-_4(#rvE1hYI1~vxw+JHq9}WMyb!hDuE+9=mb7}c3@yGBLoOarhr3Q1 z@bS@g+Mo89&mEtlp2FEQU~dl9IXXn1&H>W--zbdbI~vwYPe7)&FW%igoEGv=qO=9`;(^qS|ak>gg1rkmg0c2!`R9Rey_qaf(2 z1@tCQqEbG`iN9bCIme2zQ)m}{F4Dv0d76+@Vo4(upYlA7Gw3P%6gEVyf>YZN3z9PE z{%<$PksGIIK+GOWawpL3U#`%sv(LkBQzk2{Og}Er$7|@oSeD{CUVz zuJz#d`*<+&X+d($0i8f8cS6h@F(@dsRCaRhfh$edcvxk-XWvdHlt#q_!m(rtYX zRQeF_5+a+S`iwe^v#5vMIn$vzD~<+-EQAA=|LBdbv-DmQ&!ZlG0{bV~L;Z7an(?}j z>W&*nr&=Vl<_BqPQC>c4JiU~={v(I;YgkL-*No8IUIUo>)e_qHb;|4F{r$2dHO%`+zDZV*{`$Qz#*#nt@#9d{IF+_%Rnv1tqk+rdIr7KfgYtSw z6zq#c@7vC}-FzqRmat&|-T>3?B5Z>4V>n`*3>yoCQ0}w~JhD>&lRc+lqd^lL+w4Q5 z-@T(FhZH#JquJ&ICK-ahMd0!e6-9%tHf6wr?;dAP{?%|)Nk0{%1z>LlaW7Y`?ym9accw9P6zOK|I&f1wo zBT50jO%VgNd>eSSYk;$FFXDRQg9-Ux_O@pl>eWT##l#Byy5c$3d&#nslN_m?c_F>S zyR^F|U&Aw3&tqy_E&5b12K7&i;g9!O*vJf_ZCoYYT%iK*mTX{8)+Vu$8y6X|m!zXB zCsLmpOL`^eI65?M#>Y0h@mc`CKXsJgz2-Wov*#gPus5r&{DS9+MN~e@4<{lQFgII zm#ru~#dbVD!9MH^W{!^{&}_g7PgIBEs;F1&+ekl4xhQNg+xjS#xK&5hW!tIc&Jw!Q zY7sTqlSI9~|HK@ZXLzIa9m@5&;)HwKP*`vp*Irr88O$FR=+qqK#5Pq^38v4S=a1r?WSnq)ECJ#4%zKN!A@j*Coi4=HNG6=LuP|!ls`#?+AnU z+QDFZ;TioWTSX%OjfJz}J+$HH09o$;iO!u@fph9hQ7YsSJG$Dy;coc0~LN`<_2K(ToO znX7w`t624xG@M?B%Brq-YRftdIzFA9~N#)vu2y z%ZzeJyvi5B-!)GKu9k1O1FqXRyN0P;lAjglaXc3H^Gt2)O&`!Qpn`Y4od)^eqcJh6 zimFWTp=pyX=)M+PV*j;?lgg1K_vY`$Gh)@)T&BrZyuZtWI%*j`{D}3`-hjD%e5YpQ z7DUCWv$QAEm|yln=2xuAG$t!Bxjb#=w>N`!Et^7j)&WdR$zX*d1#I=A3ihkQ46aow zL$A3$lzdtPG1>0WDG?95m!Agb+6z$oDh3?8l`-tCCPrVBKnLH4^yH^P8nLmArZb*F z&7R;WuWq#HI|6(keSFzc1X?<8sBVBExX-Sk>bXKhRc{6T#Gjd248NlV&_c_8JfUY#6oTjT2AD7L z1y(&YWPwX3FiRs{_Wk7=o(HrNZ5=kFLKbC4u?t!4?oDj!y9KyH!~x}o7NTt13v~J{ z!|pqZGE=Q|l(~2mL%u|!=H$CPgXJ1*>nI18hsW`5JfGDG2}I+xUNXcJ8nw3@ki&Pw zaQI9y7K(S^BMUwQ#c8p&HFMc1{mF3k=q|W7HvxY41+hK#!R*=VP_}!LJ5!BX&2pPO zS%+*YsLV_R-HHU> zoy$z)ma+HIlFWLyJb&+*#Cm?-!clVn@JEaRlV}sOm@g@0A&*1M_nIiya(B? z^E!*lf6F=>#VvkR?Zs_VSK|IwAB+=Lhn6k^zg|ypLiXvyOyKhy9>E?Jzuk=85_D*C6?9i!m?dAkuOS zTx-;Z=0H7|dtDe%WrX%CzJT!Ge<5nkFL1n|&G!72Vf(8T*$3Xg8<%JZQ_76sW_tua zR*S>Rt=)T?CJxiS*+A~j zYG&EkegSbBm^m0@a>6tS^TJNqSr+Zvsq1%5@ye1un{AW=8hhNC_^dT~!QvkL7 ziLg>A3siJP@yYrEsA(yKSJ&3kXX1;fsObzEC2x!33aim!wio8i`wz!&i$=Z68EF4u zCmcAn16p`5yTqq>bUhP_|E&(jbzh=*|3(CGs#)+M-xHNO58*@kBbc%3B-{G=G;>$2 zWOfCwF=)mZwy&=jRi#c-?fP!I{NY^+YC2%>NCd3*iNa(#NfeX+0gZ`&Kx*Gb{4v;$ z!k%yNuYVf9CrL$X{dg=1*M(CrrNJhJXMXz>(#vXh>8k@CH2eY2{!f=-RhBJ4D zpVh?R*E3e+wFjj>2jK$cEm-i@h0XLj$c7J}U`hAhW1Q?KKKDL=kB&IPKVBf>TrwAK z?|zQqh5>9-{BPof3rhTn=TD=)B z+m&F{f)gM-E)|{zAA`fM{g_dF1e2Y=fgO{-1+_sA(Enl@tW-Y3J4kZy#o|EBxVH=o z9n*1QTRZ;#dYBFU-p#bXA7^o0X*g<>KHf?_MqPZ@!M@I6I`fSI4On;^lpm|`CBrs4 ze{Tge+SWt1avNOHv_zM#Bdnr$3X7HIT_LY`vGc=gShBkdtHx^Z*_j2Rg{7b-JB{Vv z`GutJ3z{3Qqn*ujhvvS#?cUlnJ6oWQPpTFN$r?PYs--pWL}22R){ z!k@2dtSsaPCX3c#Fid4fMgiNpj=#TdyN!!HFJhPp&v|uy0vncJgl*Pq;gfF>Rr{_0 zod*xoz`DOM;f)CX7JUV4V-+n9UDdM~92jTuW#@X9YqF88+2_e_Burv&#CiUXf(iS% zKL$e|rNgW=Yq&Quj5iH>aQ+-We6%$j-`2+B6->j`{GM`7Ts1go=7CG! z1L(X}395X4P{zlRoquS=3cQh7%|x7{CW_B$?!$?ND)?M?D_nO~qm6s4*+w%@mR_@+ zSuXiTZ@XTl!7keUa$`%94aiP9J*@|C_nIfob*d6Jcm$z)&DFn6bwcW(MR z;+-R}VAFp*KmAuFeJ*l`-jLaiJMM1BmnrLU?#v%hdh-eI2l)d}49)3z$5SLJqMGcG z>cy0`BdGrU3s&FPWvk@{%+^Am&GzJ-nA|Y@>U;&CWG&cWu?;(=IG2Sq+2XTKPt0&O z#SD2l=Dc5oZSgf=yyKEC1RTIkY<}6@=NtqZE8UzdXI)KEPnV`AiBHS=c zf=6!>;p*%XvalkJaqPB0n1wlx#gXP&n0`4K zjqG+a2hU|p(6)j3Ns8dIqF3w?YDrKLfyx4Kq3>4dL z&J4#GvhV(1aPLfh&9C&Hq<8(OUBM1>sGT|S6aiNJh5cmE$0oo67G?;FR>keMi% zMPx)7=ee)zK1hQUrLszcgcSWGX`v((N@QotXqnj=AyjCnq=9xxQWO={|M|c0)|+$A z^W67!eZQX%9_g2-bnQW^*{XuJoaNvgXGwzB7g9SBUhYtf0JpbM9n9AVLC2*#MDE2c zSaytqjo@QIET;F*5QNqBT%!p7YeQYNLIcoY0YB2I;VqR!}&Hw zariYHUw8qutN@;Wk>SRe|Dr2u^XO|$pqYN5G~;73eOOe3m(^A1-GYPkj^0e}8>zo^ z?V}nR-IPK595zy(yUD0*kxIL|dAUoLEav93j6OMqPA1i@9#o>l!TS9dvc9K*0iFgt zktj}8;|1wgy%dnpSHZZEcW7~HKUC_ZkX;u)F}J^06W@W;%($c`Q~zKZmL5$6-J(45 z^Z6*OU8@WC+oGB2qKg^jH;K#@{@29j>?El~JoF!j zE*!O^=_hP3>}dnc*yKc}&pt>MM&+nTsVinX2-92VftK8(RO{Llv)f#ROlHqzoVO`L zFQySo>2M;isX!}@4%44=qo{8Eb>dYl!R`jE2;=I9Y={z;K9uJ8WgVi1X=!x)c>z6L zu@Axxx=3{HN9Nw)WSn86fz#i|afW9t#i%h`*riHI{k7~`TBL!`uNi0>33Gv>B37jp8+mSj!l)bKmbCV>{z zxPAqP)}6*Ydwp6r%+UWl6ltk-AH+^q1*dIp%um}d7#81vzFlla-#8BcRT!W}y$Nht z@)ot`&ZSpYAYI^?OxtK5Ej}5p^ zM&yPtGgrHW&ao?@B0+~~0`F5g{D5+mIh(lNlOu4kz!l;O3d!#mH4twwfs1x0Qla~| zvD8h0o?o((HW)~woRus5vOK|YF%jiHEw~y8E!tvDdq!fLVNa;J?CNvM+ zkJ?wxW6DuOR6VCm-iGcc5f=?%YvV}{BlnHt^=&0{d=EqP*|q$qi8rCUNd`B}^B$L{ zYK*%xDFTIyz43j062?7EWG-BYBx18miE{08oa?7d=l-;!3qJ0mp)dUCy_ygjwo{gA z3Ovr4{r95L20vl=_$H6kP17eoE8_^C{VpQF-^~;+dcl0XCq{&SdXv3BqT#~LSWwwh z2{&tXN&UC;Op3%};&C}14hDRHs1OA_bjSpTmF|Gqo7LEoei0Ot=V5bv9BgymPE~KE zQ^iY8RK#^U)%&DM4GhBQn-&%3YL+dd|1Fvn?)PK#rzSa;D*~9(btbfYx(&7d97QAk zEQB-a8)4s5f0nI3(x<=4Op#M(0f;V1JrH=bNw~bp?J>7Q(G>I+?TDSw=l0f+XjuDl=d6 z47vB{6XWf63s$)NW^BGq1CQtX$@;`x=D%GzB;xdJ{9MI$gxEWOun{P?dDF}#`hk{yBo-?ZALh85Ux0y!aRj{#5Q&=32V`3r1@$%O-H-vfzxW-LgfYA zzp+cWr$3zJ_9^FbBU#4%r_KL?w$4*n?9KiL1LtwqU0%oak1a<@@1t&=fKW}AnR*U1!eTUmyAAaKj*ZLkhDJsXC|`H8S2 z@ETe?DaHZQLs)sRjwm?JhVug&u(icSx4UEL%e_L@ua}#KfpMe0iv!rn6 ze~=w7!E&h^N%{MH@~CeXd_NcrQq8*b>GVk2Fm;)hvyP(f>E_s^>5Kbq3Q1w#U*afm z6~;9(Y4EB#`t(X0eateE@9jsfPE#SupatlRR{sN#2Ool0s)~ z(Bm@TCRzi0B3hWba|ddw2*BOsa>Ozgq=fhX~J>IR&4EfE5qe+A}-8s49;23{*3kZbjd zq&P(jRt;LffPyY;JiZzSRvgELerY)8V*yx3ih}>!GG;PO7-Ljl!kUEtpq2F>7AWiD z7qyf4@RL0B{I!H(vG-)TTN~&uxd}01F0d{2CCu3x z?YW?^e*hkp3_?8!W3QG0+Bd90rPE?)GVcczvU_UDtc6zl19o6a`;rq`*Qu3f4J62?1)4}sN-0=pbxIuV&nI-nVozFe^L7KZm zMwT1)F`gzHY^IIEiu7e)B;7P=KyxP-(t7?R`d-bCa65lNYsw4d*L9+Qjxe{{ ztB!W7pQH!t{n6+xn~NUbjRNA=G^RF&@NAt)RF-SN0dFhrjRYm`y_t8Z zRy7}2fA0;to+phe?`EHAQUQ3*iuIJWUE;pCI>&Y7ImoR!62%=YFyhV{{7F9)Ut_bS zDY(h$9B#M~%auK}joWOl!d)?+-CG>cLHW`rAhS_}E8F~!>Q3ZR-KZ|Q6?}+1e=kMg^ z&YREu{FQKJ%B8qt@vrF3)1h2%Ia6*hj?sHpW*FbO{f`?U|CZY!RmRSlwNdk89Inc$ z;V#=3#+|ER$G!ffj6T{NLQMvoXzxK4EIvgra?lHRp1Vzji$dvhSxsuRBZ2lU_ogxI z`N!{(B>ftF56vDHpw;&TjB!fBuCoCwgLwj9UTQ{)v_ex-wVHvO*fOtpzW_dF0;k;$Co>&2_x>gl;VlApqDeI<}=k*DfcxS+2)ix3ML<^I0#|r7G5M*$(%7*WpqbUNmTkgp(6*;b_=B zJh`nHOR`;2U%D7Qs>V?yUy1rW)POqJ2mz+Xc*kZLwVULngDOL;bN32;|NAsG=?bDf zEz#)Qc^bWDB;%fdPc&Drkq!lv(B--9c&NArqfhK+_ht&*U5W2$j#xAmkGRXFN>SV& zkG62De5+w!dm>`q3~E-k0gLZTVA)&+HZ#5lOKu&<@!~-AeSDT4gypm|N}NvdoTN|I zZ=@w6t7)J*Q2Sfvl$8|F%ok%&^erEe$5m?ou%*0= zTb1C&y=5uO{k>F_d*tF6)m84JhYs3sAKX*ps=6p}S3B(B9=@#3RW*K38}6%vs^LS3 zWzUgSc3<#CULwBvxfLbGZ6Rj97I=th!1~Wh_#fMaTJmxX>XSxbC+j(WCmjYt67DE+ za19!#P~7@e7{ZncfYS65Qf=Le7k8E6m(@u)X|x$NM7Cg%XaF|liJ-;UCol`T0V{3` z;-0Qj`0_9bwq1UOgKWNXnNKeEjYUI=eId#E&!3z;s?6*hZR23%9A=R%4`}ZwATr$t zNMiFYI8&knbw;{Su{s=7l2(9&{A{QWH>QWxb?F|(`P4A|5N;{*LRWiJ)ZHM1KQ@iS zo_~FC)_E^p7g~dpi&kKas~|dXb08_*2Uah*h}mDW@#1JYPUAfV;-|cUe?Pm&D!R-1 zj#60HPB=FHy3Qoh2&N@ImAQJX3@_|=M~4o5RL`D^hSIm-EN>@-eU1j{`Ts$_=_=3^ z^Q8tA>!`&>6Z&KCd)#}W1mzWD@mm)UCYwG6rGv>3wo;ld%xJ+i%`rHb-%d6OHINnPcHkyy{H6MyI3gCQt{*4`-Cc~O9B?|Fdh z7uMp(Tk5QDtCQp@$$1G+x|$(r-~jym z=LUTr%;3Z6w@#Artcaio2mwH`9eF3DN27tF|sfJcmkc`@gNSPzH)sw%VZsuS#~5QS7# z)+sY`iu}EBhD`1|L;gIw0imJn9OO$X%eaz6|15KCW7#4gSx!Fn6p^w6*T^)}+2|f8 zj-~qSZ*9{dY~wS*gD>>(kZ&D0KS+Qc(KBG8P>-68d8n$Ji$e8RA$U_bxc8ogQHK-g zJi`NbhAhPc13*<~%TsqB0qSOzkH^r(QPj40pM;%QJdQ=1H-rr8{D|6yD9fdwKgeuoh$%67w(v&Uy2q z$C{bQ{q!9oJfDKG`6Ik!vK{sPa^YeGFITesBQ;97MIW+r{6(XyAT)X<$V7C5$XFkV zVDlCA*EJ!GI|FK-*^=vuot(1qEY9b(1Dr>Vz@#wcjJCraqRKST#E*?^hpm!|w9Lnm zCH4C8p~6NXiyA@o?Jujh1HSxr#7&hQG@k=bYX)3DqU0>NRLXgUB_BY?hAVv z?&RE`bk5ur+*{u_bDv$@%5}gFu%hWQT2pZPMXJThlE$*P`S|u3Hwvy}MUm!s49m{64!v-18`B^Nhav4q)m4R_`2)&nPPoq;*>4>fo z(^YVUxlp){EFQG_Bv>v#>)zS53Il#mg0XZqjpl2hQhr@@ zGV%kCTfT(TCqIJG(E~&xLW~TpQY2y*7d8aztjpQl2c*_4y@i znka*Uo9YlWGLw9l2_Tn6#mJOcIR+ct#Tw=~7O~xwW$ixj=2tJ={pke~w-ccFzY)@_ zeH&h`ltPPQJ*=^gg8Vg$@kdY!sy&fIk!6p;?p6Ycd@)LPdUJ{6A486t;7W$}@iS4I z2bl|Ap5#(h8#CR>hN0sQV7j>hws5N9;kiUK3oT%s@saqXyB*?ue!^;=B=|4No1G6- z;|`-k7;bzFhNn(IrjiHnDAwTH_?u|_Oa{d&ih;9P1UD)Q;0Ea~qAa?RoN2$pEZb!U z^SaJL^P%auOh6n~i_L>GBY%jNjt{vYWDA@%3$W^zINMdT!37;@=)YJHwiUD7G=U_t zN+ARH&CJI?eAjUm%j$9cuM6MU&7`Tz9ALm>Kih*-0E?@;Ve!-~T->w)x#Th<9QuUW zaDS3c?zBw`nP?DDVVc%aufR<^tGKD39yG zn1*dNrdK|RQaQm57z<$O%B{HV*-15VY7r0 zxX3Az1+a_k2KAl$2kB)i&^9PQm?aK9;O>sZHmxb7E} zyG(&fm>}@XM94jL2J|*X!og{a+4+qZd@&M&W#e1Wp(+N8`LChba3*kL+hA^eFf^u3 zgI8IdRTdTcjA*2dXjaCI(RDtSd>7fdo4VM1u(VaU497UQv%LZEHB7bI^@ zgZ0@LnSc^ykbiO;hE@a4r%kx!p%fiGGM(lLk73<#0rYn&lC@$wjNIH64DTa;AdcTj zabppQe&qw6U+zL(z+1RCRKha#^TAOl6P}+7#O z;rM1jyfn2H3q2=bb#OI!J+&l$u~87VMiqnf&*Rj;OPrM3?M4a)I}NQqQL;iMlu@?S zW!}$wL1a>z$he{tS?zR>tU4$G)%FvlVL*u4koJX{)?i5Z-V5Px9Uu0Y8-)wL)H28P zPIHQ6XL2?z=fKVW1o-{x6sXs!k*cEsL?nIzM5l%m+oP|^EuREZ{EP=Ufy2$&rimV1m!Gp1!kMM5Z)`d};TF+l>c7 zr6&aBSw2eL{A18B)B(z_ZBXezv7z6LJsaNTT+?{Qxyu{O%t}{7p?!W>`$Zei-G53F zp6mb*xf)P7w2^sKIm%I<`eWoJ*ue4F{*2*sm<3IlS*q;d(%wrSQR61U4q%|2xqIRBf-o#6%bSJAN2tQHR3j3*m!c*DJF88i^TO02Z5 z5T+`P?ERPz#^tOVCeR(<+GcXTbpGY6*xJD4y;;ggH@sn#s{F}py_HbaB#JGIR-;&S z8!=@w!H)YBz}|ibY13z)`|p}L@_ZjT&;K50G|xyHdE|K!7tHbm51*Dj2kD!z}v%SSmjN>Al-<>y;$*mMy|J;}vkTPXVvYX4#}4Z$VRp z1}5{Xp`O?{<7MQ@xc&rYQS(t|x-&3=Z01)rGJ(8YvJwQ=7J-~JKRI7N#^I!uaNh2l z&is23!$fXZB`GmhOx3S;PHtBtMz zyBTq0COGa=6|62)mb2!R3{$$KoatQ1N1COgNmFAK9MO`)Gn!Aym#=(qvg|ck9C{qC zhdzd!ceY?3bO!!eq=W5tc2{#6Im3l5h5}U1s9vGh$X$7k;f*(Ejh?Bg!5TqrFey3) zD?Z9FNrP87J@0rKv!CnXwz4QlpWRAMAIc^IZpvhHU?8c;RbvEeUvpBdvzfe&0Ee43 z;T6k>zm{PD%j@ou%**9W)Atodfr~zHURsM7EE@j9_PJ#sTC|DCuGq@(Zf@W-YEE!E zrdN{oHLOddSQ-3P9Ep&lG-EY-#;9fI8Yc9OIA>pAs-cN)5o2}#7}Ll5ozWqQ|gVZ$sZLaCbGSZ_`jWG-JL!%lIskn4{8YiS6N89>z>iM@yzY9E%E33x8xKSBUwIC*cAExsO1q1b{41ApXY@5} zxw{cJDQ?22UTU!DcLqFN#Krb#N)#o`$Zq-dMB_vU(NBCwZg3{afw2xI_0=4r>dy-| ze|;do6)H*3a0%luBF2>#YNyVk=jgHX^WnwhOep)H0_)wBaMETJsAvbAU-gAl-WCMi zoH;PFrv%n*P6F%H6L5M>8R_3zNq*)$BBdW6K=nog*f5)cq?pzCrbhy!-tgllgMn4qWF)eO zxU>5%UFwbh?v8=|Rn~31<~QTsD9i+%5@*7nvv-biX=>)kGABHusnkMyI`7RKDt@sB zx{!6GB@4r}yx**wCm)07oW@PoI|Dm8QwwIU zR07>kj^t?8FHVxTIJ0G?9r!P^hu{_lw#Z$k<~^}Ad-pLqtMnF)jEtf;h3#m((`&j* zC71dg@}|CL@6p>Y3+c&~nY4Ei!Ef*W!2%67Yj$-t*TX@DE293H4k!y?XHpSZzA1z4 zCT~b|!)@ZSdmK-b9mx`?H9L=69wp$_?{B zDB@z3DM$_LCQ^4=h(Wm_*hv0nHbtFeG&dD90_=Ns=<5+s%_xQse@Ad(bq4Dti@*=d zdGPqgqhRc}3f@IOV9uxTkd^AI+5T)ceHG_UH8v|y;|1ZM(P99+_a(s1;|1XusFDdV zC3m09;Fg_vP3wJYshpk;m>%#a+k5he@rMmWyHA8;c~FGoU{y=@?qq$D3)%m>B8#~q zyp|Dr{fg6{e34w#+{DcL%y#&r(#bc`Sn}*qH1YB}2LGf3V9J$sk{mn%!y>VuP|^T4 zLC?v{A8ar4=uWa^%@J6Btr$c@1ko-{oht8Fp(jLV(*6UtNa{c_>7Ps{=}rBxE~p8f z$=v`Q-d-q)$O7%#sj$mdg|;^LW5O(U?mE$qivCjA>K6uA6CE*3wHv}&4_HiW5BXDW zNsc{GB88KKB(5!p>}@?l#w4$iJ`-`Q@$g-HDMRu)5A#8B(oPw>c$ zA}aEdWbXL|gt@t%^!l|i=at_xR_62RoQgbr^hFJ~YWCCj)!9^Q%PAU=S%)im6Ons5 z2=67_ra=;CDN|)Z$vs)R8h)Ydx>u+d@rBs@vm|XJ;^fM91*rPR4+TwZ2DM@WQxG!m z#}VD{Yr$Dpf~uTQBgJbIm_2p{Fc6u-JepU)lvb}Kmy*~P%a#Y6IQo+rs^!C{5Bnjd zf**f=WIK{P@1c4U&@!@wT(PtvBig)JUdDa~b}n%7M-Xm3I!U6L7{=z>HuBqG3ci<1 z<1A%M3tnXfm=%4mNu(%cMp%q?MM@;=DAVT5fjxOlta1sJ?)B@e%`y`AThklWb^ zN3;jwUSR@Qu>7cn*?SSWk~rF|1G9Y-pf`zSwfyU4U5r;?ner@bsT84Ft3|1c)&n$H z+e|Jd=8&&%^C2xW2Eyrl?C;r%v%A)k3P<)`DqO$_3$%i?bt{~kS_13c;~_3B4mzHF zg$!o}!e`5}8=Edd|JG^1{SXZak|yZEdXhS47(ifU6X|;WlQDKmXLNY}lA`JqFj$cZ zo|#Q}^WQS2X{H35KhuThRtd2D{d^74UJN3Bh~T!@0eJ zV0>sl!wo3nyovE70?Je5@-s8?L2QKCvF#T=aDIj}t6Ff^7You|kV@h`?eL(d2sG;* zAxHMy!)jTEymdWFJhPIZ*kBb|z<$ z37yx(TwSpUJbevFiP#JB)q(9Aq!ZAYlTNycGUVr~!f^X(vN-l8W0?Jb3EIvf_cJrl z+dP=`i0ong3fJ+1%^|Y6c_z_0bCP+^;{?moHJ~DD3=&$5AYuC`QI%XxUfL0g&zRqR3rLrvEzF2N1}`U^U?Pj<7~U78>Pd>U;~76~ zd65h6>}9axY$E-@o5#e*7jeq=_i!AeQ^CJ{3E0cnK!^qprr*B|!AUi6?Z0&L%?%sZh6{kw%=#pzE( z#`Goex+DUNqT=96vpXbs?}Bz7FADeFY3)i++VwdRDtz2PyVnbTM+l)q{207ue#5jO zS?o#XL$eTm))}sY&ZhmaYmOA2^OL~Tx(N_}FbU!*Y4Ck*KKQfQkf4}qR22zF#aWAR zRd^ezRG$YHZD-)&Uk$2P(Tc5yLr^Sd33ZCsqvG27^iZP@yBpGg(E(ZTyB0uC$1bON zH;kxxKM(bc`-Bhee&R@PI+R!?L3-O&IE_c?ggD_)60p)sQ;GbtWJhuphA?-OB zwU+H&%$|nxZI6Mos1NWsurrc?^C-sdOPxL0jACgnu{)YYO0D-1p~RUO5?T%8lS?3H z84umbvd9*`%*OBC$Fa|M4<5L18I$|F$WEbl5^$rIWvd6n0-w`h?N|o7J**2e^DVrX zGYu;gSkHST;3anL9;~cDUmcxI=Wo-a`7tVx{o^fh5|#u17aE{3pbg17L14DTANNZc z;3-92-W`)}+C_Og=~E z1v-$-!+G#G?h>53oDCTlpMrDa82owimz_uc!q2-VvDxo43J$EtA}L=qd0m#w~@z@)04Cu7L$!m7_@M;I{>S;#z##Zz#@PdtH6t?wfLiW39 zm@f7QKz}+m6!Q{=v)at3CAFNzF3Pxfd<>R^%Hh1vrWnq;?7==3JFm|rWy*uhp;wiR z&+6^e>9`9`{o_iV`#+$K{2k;DO`*>1Qi!nh1uNE5I3k^!g!zZ*GgDDCyU}Z_{7QvpB#0>!*>no1z#g-xF6}^Z?TB`2hk%Q5L6~VEIWAld~o<+0GoZzfQ1)( zKMb^hYg54(I=BaancL$U$vs4N=0~#1U^=8$gg~s{dblLKfn`ave9b0`Hshvv)|Q9H zxAV|M5eaIlItM1rcwmlQE0LX^iNEzzFnat1Y8b31+eVMG^9F6QPCu8N?|e<3d&

nV<8DvyC z?hoP8w}Y_r&KeTQF(viQv84ZgCPek_0>$I8kQ>4BK#SsGkq!F}HY*`dG@41wcb2u4 z=7wTjPIzmt3wmM(mM%?ad#Ytl*ucrC}yP3xflRy#?}D%$oE@>EJX}zi|~lJ*vhPx8_g>S51sx6^I9I z3b3GG7CO~evaY>AXpT;VX{XOa-pmnLCNT$4Sq$?1k|3~Fhu9q%WNhBB&yMz|WOTBX z5q-N9eD{1Nh2KNiY-|Mx^YYN%eX_KBjwPLAJQwnQtb&ZhA($L}3ExhNV&@JuJg_vK zD9+7bb-J-wn-WSg%1)4hz;7hEfz3e8|4Uq0pW^x^*41?A57Y{HA-J}{b!|BeE8T(x zmB-;t-D!v%cnwK)W`yTKH@U&T4IbXLBu;$y$YnEY_$XS2O<$#H=t5^27_}7d+>gZR z`Yrfyt}2<+|CcfDO@MKc4Ejg@DV@JpmK&ubfE#0jF{^;>e)ks=ZTmSaXYnR38@~@a z=c-`E-%2=mUlG2CA~;LBLd-XRFjwk;V7!We-)m_KZ1_1 zvmmu34P5J3=8IoE)acyCS6X$wt`X2+ms1c7D`}$ zoHceF2!l;L2Vs1u2%gu?fH+YOG+M;K*Wy*EJn4?Rw|1(5?Cw1nVhdrEDa&iK&9-F#>{9neb318a@g4K=0i3*st>#4E3MGrI9tLeMyuF zNRuK`!GgO0^JRWz;#J@zn~iIw1S~(=RB;x(uAF-?qF}~avE+HM{772Ns49) zQR?j=uTqNOR(L7gOdN*CKiK;lHw)U93Sqj?CHTQ|Xd?IR#Hi0_Q8k`rmn+>z>zokq zzOj#;^XI^n_GQ9~0Evy90h|&Qfwot=cuBP$)yvO;ivZ%MRUh!3PdMJxNkZQ6dTcsX zM82vNlP0l`Fyt2jui_s=Ynlb#cy=6q9QgqqQ93xL{DHPqf2Y=*mo(5=lVd)bz?kjb zL;8#7lS3@uDa|Ygyh;^Vcr0`r@W_CsD;@iD&FYDB|1f@9^Z+FbMnvmX;=X?`z>j`X61CgHSB;T&yjcB22feW>+yUm8%Q zieGoJ&#mJ-!9jF02Cpc?6YKb>%j~5PHBA|;YMzt9vRU-2Gnw++OSrp#t8n+}|D>+0&-yugwx8Yah6g;) zZLGKk$CEVH1?988{EhTij~aQm7eIDK9V3NKk+bf`Y<*+J>8eO!8X z!x{LdUj_3XJ%T7XUY4i)kWk}t_S`>*)<;Rv@@6rred8@j^U5c0C(jW6gJNiJSxmf7 z9wHjN58==H|A@xzo1}X9u~7xroO9}eAgK?Xg{h7XFd7j>rUr9~rg1u1>3E7{N$y79 z{tEcu;SJ*p8M1Q10!FtN19iw{>_Q?5k8V4WdhrSyS9Iad$UK(O`Uty?@8GmejVR%K zALfk`@OUK$x6M}3WmEHL)Ng+3y7&M-&3p;7t_MN*r_I!~+lm%`oIx{OAJEW&NA%N^ zM!HrN^vmv} z)B8Wu3MQ0?9 zUHEOHnBy&flVg$EOCox5aM>npx?`fA=Ka_~AKbO16LZWd&$>h8qDKylJUEQDx`Sz(MA-vcII z$OIgJt%Q^-iBM@^g#xp(k-z;7e(MjWVF`h>y*rK49&s?sDuar}xfpi98%vyv&_9XY z0kJ)qc_&oxJ39|C*tQCnum6JIJg?G%`xzjTxC7SMC*VPiry#rP0c`riV3cez#%0T3 z^*=S75D>y6q00DdxfU`-nb^(q8CDDquzjKJRCrSeZD)H3Q>Jf7->PlEpWg*yN&1wW z`ibh5np8SclICwPqsG#a)aJ(8CF2fLln@I@btq(q{}U0pJwd6Jm4Js?T! zF3HY70-uR0VOEFgUtr%c7j)g!1K-UOAtUrGt=$?% zUsS)R?;i2vON$9y5_p=He;C6knEWWtAcniUi&Orx92yR9ifa> z=R?`^fdvwgbX=0G#P%M8;M7f5s690e{|ZMzmPaA1-kbqPH1*-mn;_`;`W3fkQhKf~ zlja*)z%-8~`0!&6Hof&l`G=Dzz3B{13ycP}cGfu-?uUYqh$C!A^HG>OI9}t&_iL8o z^(Tk0l>JUv8?kp0uQc=?%EaO!EzGW7gaiM6lV|12(IvM4qj-}s=zcn8r9H;llC^ju zWG&kH?nkR%OBvz)GI&@i1QT-Bkk5n8D8n)x-gERJ^%g(u`#B3fKJ>?3Mfq4$J)6Ew zrWhFQgg?Aa<0Rh>BDjFVGP%UD*|C&9h<-;i;sv=*l}T9Mn#r=5ZsNP^W?)#5L8|uM zfj^0=c6%PV!aPp*1VzJK6pH}eC=Wa6K5H$e-M#DshEWUICUFBo(OIRTWCY-}# zwU5}|juLI)M8QXM);YdE8HbiwP+$Ep`qKUa-Kc$mc-k0(&)>K3XHP7ICqIU~*^(Ia zbS>t07UMgY-89D58!p+ULu7b2wj^9eMLtn(VPQ_0LDp5dB6f+7I zph9>SZkY7~*JsMn?P`Z$=Vp)r2Qwj{P9|o#wk6kG%0!9RHa)-XIafX6o^jWObvonD?m} ztkX!H7WnXpBY1xMH6Sfd#PZ~BR8&fZ7G*|#16^}N5dUwXOI~43O#tQ zUA?~%g?foIo7w+p;vErIAf!&aMk~=j2Tk6UD$ERoB^W!v|LwjB*%i#DqKQP`E>=t3qWT%#FmK@?#8`a;m%_DV9ui7EkDK_cVgWaCsgI!3WQ<_( zCO+d@#LpN%bmG*T^XPZa8anuTKP`DRiO*bX;*nK0D4E;H=lFVP&8AT-DyNBthIQHT z5DPY~$`b2(H1OnU6WpdF#@bv*u+F0t5$${qiNx{{UXN&G^3{U(H-IutQUyF%7me4hD=lv3o(Lb`a< zo4!9(XI~%au>65(Y~eF)tQuN{>Ye$RqILkMZ}-RBYQ`wydxn%EOUc%a_ZBRFf=(BQ zaqCuPI;X!JKYibh6Z6t>t#B|5{CmX(mq@|4YCTl%7QqQJi&2CYV2#se9OAh_tF&}r z$-V^My;B5hrbIy5w+V3Vh$bxY3We$|${4>!ALDBk*~sO~nW4)rmY-XSQm&oo$Ft-m z)en&+@A^vEr9{iG6jOWY1-_$NN_jqCVUbS~_|%m{YQj4R2$I0b3MQEI_yKu{yrIg+ zzo@yZ9cG_g2YsVYf@xeVPW#pm8{1FAy*)B)qoD%pd@IM6Kide4mj%M%`5V9yL(w35 z5K3PKaZ{$5Vq8oZoHX6b9d#wxv}urYKAFQET{(iZGNdS4m3R20^a}iTa)RF)-%O5l zd*N}h?U?yE2i4llLFU&u$WtH)+dKo3{i-?pbw%8akF)u1y*W*?5|GN-B2qY5K$%?y zwEyx5yq7o%je>=7#?>BLyXPZ~8~8;Xb1$H8Z2`7yDaIM+{K#JW7@6|EiR-0{*}sRu z%<**w`}k3YMJ2x_$B^UHxl0FPWMyHLf;O0Y)Wb~Pn{w?%E$^0*roDTCX8l@8lFR0x zjk`2XnXHRpk-_*Ic$eE+2Mluwf*a;@;O{L3Fw^vhgmuTk=-3`8jVhoKN*8F}`520J zUd8ISY-VrQ?PcN19^zH!_h@E6ikzqJCi(U86!r8Xy|1k#(S2`dpJOlWomY-M?m<}e z$OxtMMNJj2bhA9g+f4INHh#7$;`u5(H@nh?#7^0fL&r|4iIt&X;RN((; zihT{j{W>GDfV%-+#q;pjMgx2tu8q!rN0{!~`HG$L2p-&PL}@So2)yGcn=woqRh4fer*Cw7l=Iek`_c?#0d)7xJ)n~Ti@u@-#?;sVDcy2$NN0g zZgiqeMmo;%X`wXRM7kofj3ymC&m|UHaHUT=pg=Ad!$xP|uS(w4v7hJNX)Yy?a21Mq zAjfV`dQ09v3aF(<2uB?#hW%&WgHdt-Ei;@;)wj1$(_dM(XwfTryQPUFM<1mxUu>z` z){wdm$MAEQW%%{(I6ObHnR!pBW}gDm*{k>IblKn#wO*Y=Pw!m8GzGL=zB>Iyfnb1m475{Drv zKA-Nf8hhu?!Wx(i%Twer@9-tekr_!=c3tQjnTMNi9HQMv@6Z9W5zJJu6D!ItV!Nai zb*kN?gWt1>gyZO_&Kx$nP?t^GB*P5iMCop0CKmpQ#%t|DY#iREfyoyr;Iby&vH!|* z>u+GTCdbCjRbUm{2k731N>=bHg{gT3vv=9ztg_6X^j3t>mf8F9yVP2A6AVN5j5aK; zyN{0lJwW@gQ8Zt<61^xN{cjf2; zh+{3Pyl;N08?(8(o*g=H4;$Z9NVZOyAEFrhDE1zqPgShxAB4 z5@=eT2KkgVz~_mP;HkO|9$gyXif=ySn8bW=nzl>8EKZp$8EIk?Dm<4Ild93RB>^jM z%g~vYMHH0aNH=-Doy~iWi|`I~7P0orA=vP4DrWFM zqB6>u`tJ@5#@>f1XA>}w-0`36dc1S8(PXyVD8UuOdO`KWt@!fba$H|xg9_q0?9J|x zjH?}@KePH^+;w&MV)uhfJRHw62R+!e3k!IDg&2*}K8K-(Cvn4~`_w!B6y1oApfRTz zb*-(&y~h7x{puQ+=PJhEGgbImxfauieNLfEL+PEkC>Q+cpI~5zA>5kP$R2--VlHv6 zOn#1#=?422rv9piX(g<~k$rwRcbp+Ukh{i%jN)0wm)%Ug?=F+Ski{e~MzTt8VMre` zF6W%_j9G}D~G5vHQ4#$d8i2-Z8ca~9qQSc8`^bDUg4-fp~edE_bR>NJJf z`T@}N?KDgO9Hh0rg!aqGxyQ({+@JSJYu*v+Yd*=0>SI`1R0OLR zn-5m}K4Rc+90=6kaBaTLaQFFbP)Q!stwq%$BfYlVHQ8`0g3JAv*YRf>;Xr^id8eH?obED z@|}r_+Y;=#;4-NfMTNR8E0n`g)xH;5VLeNyG84y|DgQB}r>VkO8bFmvB8U zuP&8)d&h};_sW58yNss#^-VaWlK_d!e{s@NTe&yiCbG-HSLntAU&^*q^ zR%x0JM*d?KaRLQ@AXBh!v8%BC;e7qM~wMA@9Odld9#D{S@9fVPw- zIR8S`^j@aCskN`P>G^!#{ldFjLqh*@6Ds|n_UQsxGkF@6#?<4X$_y+USixsEt?1hE z6?ApEHkBXs2aoJya7RBB;`sf2nQuEu-b$j=2l%<2T{sl^o`6h`Nwjl~F)cKkL0Nou z+DcRlQ>_i~VX`TE>$#4-oZ`>4ms(R^N)?9h8-V>q^Fh&m1il1e)Ej$MptC}o`x^2W zR6Y8*;|8l>`N89mb~FRcPP*XlI8mG+x19YJtYNhZ{QHZVPWH7HG_l`?DjsfSGpBp7 zkXBo^9U9>4iFR-i`wPKlf2lR)Cat&fqt9PXk+xnEZITEiKM^Zjw_pxx>ve*DTmmHx zo6yjCOWOOim|OWgk-NNEQP7q*fr-wW$?hH9#0F2su@Zx1W;^K&>zXx}S`Cb;f5I4^ zuNcpSw)itQ?Uihu6z@ZNJeqC#ubTFx+{Q|ya^43TjbvnPx+{2>X{NisshsFdZehe7 zZcXZBaC=aS?q(No%M*W^xWA3|FZf2%j<4u(eIFjw`OJ5VUSWtJ5$`9N@UF}nh^`RD zd3QI1Xyz6u&K-opAF6mXXdJG1zK^Y3bb?hKOlNvoR_tQTAvUC*z*4)6al9)4}q8tnp(Ig)%$xyedK`^bJkxCNa~2Pe!IU^PMU4sR@l)Bu)Q(%<#SR zIK=Kx@E}Q?W$cn>y_4ivn#d~ldaM~sb=F}{7oK9iOg=*YE>zr?fxBh)V5-6hq1{COy|N`R5`0+!Wq>eK++ftoUq#jg~>6%Cmi+95W#~xpF+iXDWXlRLAcRXOY6xHhld%3k#Jz zQ1Q1j_9Z0a9hXLmxp9NurHZg1YYWp8e(I)Q_6@Q>juK40ql~&_=23FqMMw$AfD%6U zxrWas{RrS`SH3ZzJZ3pAr(CdZ4}e00HO!!41v}qw$a?c~$@+5~4cR)Au9gPon9Jj@ zwmV?PJ8*0M6M`fA3pvZoYUZ}Nf?b__f_=BBWk-M3GQaHwEH>Yj`Xu(z2%8fmyZ!?&T<$+ft6-B{Xw));4fc>>>0-sbAXRB2uVhZQY+rfH8C{&iT0@68hx?=@$;u^aHfBNOFxu~Q$+aOZ%zuvT-TsA_ugQSNdX$l z8#2LbeYUt=hdGRnq{>4M)TXRWErS|(jQ72&CrE-xR#8-;mzn9*AGoT%O2QXE}n4GKmxx&;* zG;6#XzpHC71TF!H_hR+4QJB(3&#a!GbISQXC*|M|sN^HWV zuQc~*FjY!~k%R6LQn%T{pZ$&a=6D&d927O}_$^_2I84NJMC%}qw2Z^QXCv+BVWJ3`H8_=WeJWle8z)xk;w6F6sc}dpg@QL0}_(5;;*OOh= z9y-yziF`jrAL7Dj^;Q_yOC4*VV0!`-b- z#@t$cbjnMG`nXeqIKdG?j_^gn?x^GVvSuer@1BOjLPq5;a{mlh@o4OrMH=CkqiwF&Rd_>=$o#?tlK+itt)6_reG-mZ@tgHBf zr{3JhekC!coYX)KX{V`4)Ek8sj>3mBu~7f6nj#d{>E8PSj4+$U`gK3h!0QrPbR(Ds zCz#VxYh~)1J{m{vQb9-V4>G9Ks3mC-GgbU)f$F5m-? z7Mi^JDV1HjLH$uYmQ9bH-CbK=!2g-)p7IZ`{2A(lERx^(Bk?i z6bTe&FJ(ruM*;~J=_^6!t>h@&WjakRzd`m~5>*8G)4z_>*xVV5Lqpp!#z>JO&yA!d zCVzO(V*vhpoP?kMhG65p`S>s_kmuE%MWK|5XizbLQc?fWlAVs&G^qgtC#;}xVv;5j z6)}RXcLF(wfBE=+vnFLc$R+jrJ>(MnosLL}GU2XT4#gLM?W7Hul9PnXj^9F>KY)!H z4rH@q2c1&cO874Zg(TL}s+CWPi?sv_NQU(7+i=$vW$t{KIm9iAgp*BHf<|w5E?Ff6 z>VNFVHLVLUuGJcw!)CI}(zDpK&~@yY$q!iTu83RtTx{@=FeSf{rQDPV()yGE!&4No z?MNM;q}m8I8M72v;RCr^Ylje?HY%z#4xhT2R3~$q>Mrwp(y9SBos-x03U`bRGpb zk|i4`FlR#|ihQ|iv$SKhj=IeZd zsf)IwrLq;?IueU3ct=E;q8;4Tsf5c{x8cVN%IFXviIeAz#q=>&aZD=MBbx0d#otw-)!nXd5qSH`H?Jt&eXARdbG$9Rt|iD|6He`*#=9fSI7kN=4T*E zZfr@F7yI$rm$eGdfbO6w$epH#A-Y#_&&0QQ{;xVeZ}f)?S2JLj!*XmDya(f_UN|9K zoGv?PV|8>NjN)?^TJx@Q0lA5wKUWW*dB~&G;$L7regz(Vx{12%cn^nm7_C+;hh^6s z@rU~l^oy*=+da$a@UJ)IG`s}{hZ^~eOE@kswziUe#?Y~~B&U*My}bBuVtlRh$aT0dw=ep@!uGJgT?n8o{~EsF58 zBfoo(xPe{Q=F#Gtd(li&iKO-(p|IiW+=l>HI4`&fo<8HSGif&J>Wb4Tg;=!SP>4UH zUZBkNQLrnm3PvnAh!=Q1MzMD}yRh>b>woTt+m1J5Gt;J314>{ulmh2U6L9@*D|mhW z1)PvMi&~x`pf~mbyzq_3XNs0|!{s2gPt2n5l8{GyCJBf+38Lpvm2e$I{`WSg`D9K@I3=Xzfy(5I)GpxY>k+AW@xS82>U ze>HJZfC+S$TEf!0E8Mno0Hea>x$$oOoLOBf{%hGrH!jQ5%IvY^c-W8o-LaZGzJX9@ z`bfUx&pVQujaajnE4KM8#lRB+3{SP;6z+83F6n6YgTG6PEV#@SCHldSHh&1T2!hes z#*oZ&BP_OEhMh2;t2v=dZ|q*6_^J*xJ9h_hO*)E{t^~9AIW%GQ4jMQ43_ZJ~4_0dP zxYzD?kmY@c!K*niMdvXLMa$!UpQ(87mkCAJb_%A+YI4Iob9Y6`eE##d#7obPQ-{MP z_#5mEv&*75`GS9_TYDQj=Y)fg@@3;2W4laL@0Y{*Ga8&y@@v7gCUO2Qs)hD~Jeaz zP$H)SiiD@4u<&VUsyv7l=Z3i|uj8=x1Md!S7z0fopP=Mob9T++Hk95f1dW+?XcMK$ z`5dhki0OrbSw|cfx}%V*ot+7L^{mKrjX5>NWYD$c>YUZ%BEjhwLg0R25B#VP1k0XF zn7``-_hGLxcWbpg1e+Ma<7D2Ow{iu3x;KeytoX|Hyv&Dn-mcuS`QNxDw|8>}1v}vG zx^hs-+6*5=e{z>zCIJd9A=Be$xa5Rx?o+EQ>HhkM!>!^pc8U$N`dTBHG0B_rbV^0q zPzmq8UxY!?=iIX%FWg@H8$NWmK~QE4)IK}_Q^iNat|Q9uW{*5bY#9p|8&<-T_6lyO zX*@m5tiee}yhrcu80r)|K=YQ4p}t?Wc-#1;$>5wUZu34TYM8MHr0g_dtCaAQg1TTtJolD z^&1d0hIpH#Z4AcBRwLZorH$=+uOUh*2ZT=4ppJ?a7q&SPr1-n~P8$XC$>lkHdZ);9 zZ8#U7I)&5MtOZ@4TC7)y!`{}ZXfj_OMi*${i)aMF^z( zh``X7Z0O@Tv-XPaWHUb$iYmh4$6YgM(;JPS8!S=mcLgQhF2v<4U*X1nK6es066CDy zU_h}BeB~a3>Sj%>J&Vo`{pLI#d2DQ+l;#8>t-O z?_XPu$n{_qt?kr?wB^b$^@Aj?Uh@><)aALAO;!*Qst&f^A)skE1$%PcxjUks+_zUl zobiE&RA2akQcNXTK%6tR76#FbRG!`0wuRnM5Qz>q)1R$j_+?)fpM(E{r@zdmwa2!S zu+u(JUo@? zK`oAp7pAoT-ouMi$*@R9nClIX$H+e?c_!2Ye6>@Nxx7MGNj3(s>=S+x{Gk`2Kt zzX~^6Kg1wiC0g^Plg_qfQ`eM4a=v8^HmB9V>8l=GYcZtS?=`sapehA+>(Re8id3>- zE)`{PRJ&G%&i~y=x*Lp8a;YbF=#QY#rhLvzCC#Mi&MNL~lP>|*qY6= zSuD6QK}84My4Xz@SC3?44n~p2gl)7$!j?oI_b|bVem3Q&xT$evEh-75sNzN$Ib@!t z!*rSaiiO$STc>dRrz<$x{|@%u@#FggF1RdvK4xi5;9vVQUV`~l7kdmT<}tB4(eoAVpV7{{#E~peyyh|UL=9G#h;@{Z&9wK?UTv& zj!}XVHWroNeuM~Xe*f?_5Z65w!u<)0F||<@j5XvyD%1z=4Vytr0Pny2Qis+x5@hzq zijLL1p~B&rWF$9*BAZP~Z%!r}wZ@{^zsqQ$`i`r2(1u$=NwEA{GK~rkCz1&z<;G{& zZ}I_`Cyytc#z+h}ej5wUL{OH|OFGy8jb7d2-7#I=0^~iy$+HqT|5G!NZbYGfrY-&H zID$Jy-$J+VZgkGzoZ!YE5NrzhEQpF518mGQ&eJ*(0@h}8FI?0(VH)Pvo{z!S?`5dn z%Xif-kHVV5XxOyJ9G2{y$IV;G@ADVD<1Ai{rBN|R@jm>nG3EgU_dVlx&Li3Q=S$FQ zc@nUzs>jo@_bH5mrap8}so}nmuRKaf(y32m^I{YuY$r2TguZ zKvS&`(UKKGv~gTKE!82sQpEE{OJi}(fiCV(!#MCL+zi(B?WlL{I%at*Qr>h`3ff>n zS6mWExvYsYq%YHE(FT5}*Mp)bKVqNQ1X}Psi`3@wZkU(PDCuV&ys_e0ENYAJ>v#(` z*J~3qF$iViUE*9-wwcL^@4|v@D}%wVr5wJbh@#_^G1S#@7TtMYz>t?5ozSgA8A}&j zGua#G-&=>AybKl(&!u^KVif!(9won<;{&JZ(9Y*QJu1pERALNWyrW95B(9OJ(o<4t zuOUUh-!!gi3=5mB&orJWkXV%>c?L<5WA8ZpspW%HEK_jzw}lY3b1zu`X$9e67jBP~ z19vyfi2K!O&-w0>1*Mt#+<`I?R%9g4%6v4KWmz-b)+;CPY57!f(~!-x+s$$e<5;k# z2s0GA&i9l2X#UwlB(3;`bpH*I#-7)tUHpV9)8px@!z8}nI!-XE^O)e^@Qwoa`_RsuI^Bu;Cgio{h#WrG_YJS_d!A%J5kZY4%C50PAik;^&_#D5tTG zX4*&4mtW?jRC5$R&J_0e+a>{{_$qjH$Rsi&=5S;gH2Jf%Y^wsew zDFmlc*0pqO{}_)x-L0Ozya*HdO>%&DS%;Ndv){ywNhnu`_jKHu?fb7R@OpoR7>Weqw3J(R`9rp_S z;Yco3i;iPInca->yj4c|Pm0W;0T3 znN5v*7E`_Dal9;Hf`vowFzMA}ntr60WNtsB`!DO@UAq}>+n)e;54_|axj74FZT)33 zrZta?iJr#2deOrDr>I2-n?p$|STyVAuM;0E667a=ULq(@0X6tB=i>w@Po zdpS>L|J0W0v0HH0?FMkh*Fb9w&!Nbf!?N_Em>BOn8?b$ha#3sXh(st=y13Bj(Q|3p z!aYR&0aV$L$nRY5Qn-aM`_v-EHkhlU*q|l9L$yPZx~WWO%SM(L8OWa9wPG`)gjir| zJ9)Tp^!mFe{oWA3cZyfi2h~iP7u!x*Paf0#X#>>mzK|sq@p(-7L^9u8O_f=iJR@i! zD&LxbITm8vyp$h;(QBrF(xC`Ej~Tdgum-!IjHi+OZpvqVJRNhn&)x*RXJ?-Fvv{2t z%8;v})kbe=;eBnE9UH&`yuw)OB)&WBbe^_|JSJBIBlc(Oe@x=;I(8|)g2F{aSl3l8 z=IlVE{tIY|#6(IE^Q7M2v+1~M2`#i;fM>r>#X--rSiNZ@S&qGmyLV@k(DW3RI6sGF z&M9Sut7Pc&8w;L$zLg64E-Wow(Us0rI{K=NXA{n)U9uT;apf_J z8SPsnQ;TlmwHMAzKRba*x?f{Q7s@eZV-t3# zaXyzP03$nq)7ZgUIT{Z*i`!mTI~Gm`TCo#|*!CZ)GTGTx)Zj))&+oof0d-ls*cw{NBa zd0(pQKS(ws4pZ)YX?jvUlWv+=lWnaCv-v%mB_!#vPX)`dRDJ{IM`oi$(`(c|eIGp% zUt=M?#Qn>J$o`!rEpAa@C$@2H#F99k%YGaGM4!cXZ#wXb-#bbOkzl>zW7(gbiL`e8 zRa%``Y3K1kQaa8XD!^4iBS?Fw9ZR;D zlYwO^)F+E$*B42Q-ep8{n$}YKBsVG%4?$UvL>Bugngt7{Dn5DB> zetE3&`YSwZAxyQaHj!6w44DR&(WYm2=vv)1%6IrhJ!PXCwY#V3dM_S5w&!pTR|M|E#~{H&D+`Q_vx&n@gvid^JSY}C$sSz6PZ~K z&tlxJkHKyn-W?TAXZ1Id>DhgxwAzuP!IPd?R+Ig%5A^W96uT0y&&EimFrn3DEN^Qa z%N~6P_uc!1kGA!rM@|STt0tp-U;(IwHIuYqeao+IdeAud?pS|o_so{2Bbrtuw96%_t7l`L1O&@A6Bo>wshSCyZ`LhXkz z{=!?_55?H_C=yq#%z#IRd!Rko9Om~V;NrAklm&OJzjsa0zNXA1cS@$9Ew>w|T}wiN ziaT1~R7AIZCk2%YMYx^^lLgfoO-A~QjEy1%X<*!+4w5TYb8{STVaoJmyzA_M?n$vQ zOKb|f@tzBo`%K8pd?HyNk)v%n-!Q1^0Uo|kiT{3TaQi2va_+kgLDqCS?a`2?o%^1n z-Hd$x*)9kC#be-~`y5)dLW|x#{)?J5CaA=Bm6VH4Lu`!*eU<9KyL?_~pUF?IJwHz{ zyJe0bBmNSGBp%0zH)AnG!XLJ_nsVE`ss-i}`7qk71SYu^K%GuCMOx(3`iww&;7|tL z>l(TJv%9%hg9ngR7-6)t28y0}%&GO2a7l_{kSNjy6Z((BUvooveW(wU`?Iip;d=Dc zO~!EJSZr!Mh>shS@zL=Ryt$M2JT0t*zbCJAKO**WW|lQ3hs8pSqgQ8`*zVhkX^s=| zXh;jFrf;OrTg~WcE6{0o84_vlM_1juIA+vmxapemi8dcgk5!YXmtZy@C#jHsH_Is0c6J-Sw*~hV7GYsvfxuLL>Kd$KCg!2QYp_9W#{1!D6XAY0T;Grb?nHEEa z3%Amlvt`D=+H8$S7;(nYryX$R$O#yHWh`p;*r5N&MSQ+~9Ev=u#|$;f=T3FMLex5jLs*TpltqI z$Q#mw5$hyjQVXE(=Q$|!sREV|#$#gg%u z^e{4sZtUS%BDxPku;uPFJTPGpYO}5AsFXArH#gzU@=Gwuaw)tWo(WGYkK<&}!_e{H zK~&t4+gO^$!R2zES6PiOY#s3(&mx+iG#=|clc8aMEL2Vo#2?uvyh~aYr9wwjU12fq zW&ZfBMGr6gw80XgDsY-1OAduCxO#C88V(uYON&XkLiZ>*vzbtGtAi8W(!*`Jxr$p@ z8z}JGCdQpMeh%;bLqUhT1c72PBy&8RKS9=0$~R^1(2Z(td;U{S&yDApD~M50G0%K3 zybDX)3t`KQMEIv)#oU*sGjrY-T>JOE;P--uf{BhBxrkY++`8TrK~z?wU@vzD<-fV% zPoELE*G`UB-t0rYcQ>#sXC!_Xc@2BjTOq3NEIC|1PI7L!q;*;pD_;k4f4G}mZ+1I6 z_ufLo#&-0bltSXMUVQ&?3&lFkVKNnCSn{wqtFFDre`Utw9v2OKy7mpe%kdwEZ^*jwS~1NufEp}wl*oI95e$j+lqz7Mv3cQV~v7(rDs;be8T8dSE1 zfpwi9oY_!H3rYh?D%F|7&+~bCr6lgK*Gg{jPaWn{f0hRGQpj9Qmu=lA%1SnO@jT)E zc+qV=ZVa+P%YH3((A*FUE@IW{aSQk97b4U@lVI<_rL!h!8(xS%MB-7Mb5-Y;Il zku*%b>5i5#IX;+3x;EmpZJ{Kdm9l~Q zNBP|8$ExtR;T!kKvytob9EUB#eUMZ!1m{(J*B|g95h5BZd7?$Y7hKm(kx{%W27HTT1V(VA}rjY#D*;{jIPdyNWLJ`$LCVf4cG1i*AW9x^>!w=Ij`Q56gwHuT31^=a|DK{||!X z*~F=RO~mXaXK?3BoeP5ir$&atrr61_^6gxxgX!2CEQ&4`7s26OFy_mDhZ|PP@Nc8LK)j$s z&>J5q7`KHm;gdekWg3lz=^gNQ#1`<-IS6*T4XF5s?@OG&gJXH0-21O?xa}J68#-hG zqGyJk^_ufDw{$5vPlaDoTv`D#lH}9qyg85;0Fe;m8KzK~RnKPRp zeoY3f=voR>tF z9n@#(!IcPSh?mTQ?s18@>g*PrG(#FK>y+q5t~^yn@-Cg<2Jp$v84MTR<6b4U;Tw-O z6f~D(*wM4FX~}i)^aae)a=;`pGhE>^3|l6tfp4-2T#mDXtv^Q4h}<6ZQxK;;v!V6vpTJpblF&pX8>go=L7T}j5PU5J zBTA$Vi<0Q~l340|)`)FCi_vRV2-+AZfPu6E6kcxTz*mQw@@1(fLX&DmZ9sRQ8}Idg z&KbOxq8X1wXq3Yboc=)^n-7X&oSqWezIlL`y)NP*={7ug@(?#)Hiwf>u;o@{3_`@N zI&JwoV7lslT~BscXR`WgAY7cEU8L1SDqz42k*+z41+; z=9!5@#qcTmaS-33EHGiUnnEzq#T6yx7JQlt@N7D&oIAnizUGfC)Lz#DgXU*R%wRJ&gaS<`;N=vot8A^ryoZGK_9cPo}Q1%T+qiSTEEB2e@+ zj5nQvQS)9x*xDh;w-?6+86&XTB?<31oki^(xA2oyGTMI)!U45Io#OYZkjJGHP`5b+v@(uE^)yWgQElMP z)n7NU{~7^yk3Vw>t`c0Vxt>7SWT{|Lu!P|625FQzYJvX#n%F<}3zxgHlPlm^T15%+ zI5zYpT)5c-F)DXKq(cJp3n!r37=18)wi1@zcY?RpJ2B?tHtbEAi4Ut@;$Hh_cy-<_ zp1qfe=L0iPvT!S&Jm^Wba$9Kg3unr?8U`!2CqbpV7+l!c3B9}czV^O07`pLQAf&%T zaJ6^^C#m)n(+aau)yo-g)tu!$t@&_dToBm3xr`p&SMba3A~fb5l55Qs=-`7f^ueMB z;)h$I?m-D$-?D(pDsAb2tq~m=-3?9+x1sFxN!T@a4NSW|4OYFF3eYEjUd`!Xe<6&U zHF+2K&#{8}BmAH+C5M)7il+eSa7u+8@aI?@CnAQT*YUa zR*M9X_~l3wr(0SAal2s9UZ+F%A8#R_5kG01Y$q-<>O;qGcX7$Pe8^vt2ga%Wcz?P* z%(hqx9t%dIqqjQjOLKy~{M@zwL@7>6eT=fPv+1#o40mL0H-Ron~Bciaz+qiBtJYeDE~@!&IXXeh0&z+e$ED<}Xg-$5Onf9fm%e-r_^e2Gq8Eif)sCW4S~Sq}e9ooiv)|E|&~SlglC9v-@rzvJ^;xJdnSclG72!t(CvKPc$38(sVxMIJ@oG;3-WK32MleWT^b)cpcJKkyXo zHUHtw=^RRwOvHbEqcE$h7mj`9-M5v6*e^U0d(#5ZU8xDrEC1qNje9OAXiMdut8_x# zX#=d7AV;oQ6L5o&1g@%yfU(U&u#tNQ`&RO~c9C4H*_DMsX?Bo z3ZAMr$MJ(F%g)*c<czM?Ng)AuO%tO zVk~OdYhZ3x8vfi=gU7O7VJbI`%Xj zhN-vF^2j|tKT(WbLg`rTSwgd?C(}zMPb%ziWYy=bnbCYLmNe%M%Na5e2odp4!wZVQJ&zOYk>~DU?eo(8P6L}A!n)4bkKh!&G?|jsy?eQ-G5_P zwo)$td>lc8x7JW}avrUUSx?LVn?g1#dEZWZ4mlq^LAJY!aMwIJEM6swoWmHF5jIGT zd9CzeD$fT^T#4y!0hm{$&g3@=F`3hkD8FWeU{>)1fu>vlS32BA$1hx@fX*=b)Rsoq zx}#`+Rxs(El0x}3DO}3+K*NVfmKx>2p1-wV5{tI8>o;bx(9gfD!J2?=JxY%$eyX(+Es@?l2-Orh{ zO%~G?NiA~SV~y)7_u`^&Q}LkY1^Sj3K>Ny_NjdchOfA%rWYarp6% z9G*TTMXu_Sw0pEP6|XwPR+q10Z>*-X`ZP7XF{&Tz<*OmlDGXl(Al}uPgRXb-Xj(`d zEmPSX$?C>G!|Lj0TvynN57Sa% zWm`Nb&D#xK_pi~cvlq!eF_S8E^C0PH864v~Z&iFYX8N+}f~6i41YgJ6z?@Ipp;fe- zbG=!?8C~rXL|Z==1drNQ42%={8A+DQRi-|c`Kv9{!GJ%#74@GQy2|54!KC8TC-Ks(1DM{kc9yi~dyCm)hx zmOL-&LDDx$Q|m*E?5qEy=sf(fe7`s@M7D;JvWikj;knN_*P}!zNhKwtsXeqy5*gX6 zL1kp6kc^}yqbMb!Bn^oo)mM}jsrcQ$Kf&v9-}iN$^ZC5r77TF}N+c3w};`2=Rxu;jik0sN}gDgGxf_ldUf` zOy5pXI-&S4lh4ek)I&m7GyXYUkHIRs^wgjPqBlK<9fPkRNi`Ezm1n{-ole+2eG0X_ zsl|z%nHaYFGTZtol2sHrG7+~7a$a+q@0rKY=dx}hy-P$U`)7bVq((F}AJ-wc- ziX%rqfhwO`@L2JUcIrlAx_N`(%%CC8{T(7WnR&~!<)AEP`vh=Dtl|angYz(8q%7_n z+9~+H=p^pzRN~|`KMQ7BD8ozMSEFye!&oi!41V0cr%V33~= zZJ*lCjY{b<`ETE3>drBM@Z+tfm#(=p?a*^HzdVsEyE_|ulcx#1Ctl@JWfHm0vT(u0 zZgbNtsS#|dem3=lA2RLSd>fW^1_|Q2ooAd{e--<`7(nIahuq&0{)8e1==wYe+{T3B zu((z5LEcaB@2fvbUgkeyp%L69e_2XAoD6Ts%|xsB4$2Ii6wIGHoZIAWLtz_EgZ|bY z)8-cs@yWnd@YFQt%x|CJxpCs`P1!}$-M$I1-o=UgYh}Q_HX2}~H|%BYpXPF5LPpje zFY#x`dOzO1Ee_>N=b5(ij@!{rNob}p+q5Y3Ij-HKiVt}fw+7FU zHkla@LB-=ty_Xi@wUPj++G{HaFcU?cuqtk%@_NCS5gSN3;T9_I-yw(?n1e|Mhao1e zNRShyjW-hrN?zYKna1yWj$HSm_$dW~r5`TSVA4el2%ZC3%iO{Erktrnw~U~0N)K$- zFBhy+iR30Neu%1H-g1|YSDI?-7LvQ-XT0<4rSV&_@0cO}4`dV*1(6veS@YqUG+19G z0P_Ysx#A5J4BH~m{~(4xq^@zATN(wGN8;$K*93g@VjJg|cM<8{Pq^*jC)hcqgR|N_ z&vYE8A^66h9i2{ySOo) z;yBZHGAD?&6jUaJ@cvCJ+HyJquaA8JI$K9_^`^mGdqp~g54xNBa*Jjh{*r_HHs9ku zy1g@LdoO~&o;YzL#`{b!j|m|kScPvUj^%`^OYvm*QtpKgCpfahfOYdMnO~CD0t3%~ zbU>mPosN5QJ(iD=R=g2Run>pCYn4grb_t+ew(;G<08G?)!{vVbDEMyaL&i6K(f35J zAhc%)2S0BDoqjig{*FOh_|g{&TR+3VU=D29{#2mMSqOx>Ww6n6C&y&wa^ieeGZBNi z_s`aIRsRiRd!l1NS1!dgc*{rHvPzk*pV(>IbYm0~`yx)AftB2woTF^%^iwo{&TZ2P z)s0}ddJ327Fo*L~%qAJXLv(zqA?=)74k6KnAYwHD=evgESH44BTW*N^8jPqQwhNnM zF5~c(8|Vz*m!5jkh~|2Y!S6OdAmT$dcv-8_zJz?-;1`DfPRrodHXAf$;W&SMCt5!` zgUQ?0qDWX58u|p`GQ}K>)T^bNlh2XPQGQ>!P?;{x*oQZV?n6Oa7)(Mr?wqTkV1ld} zrZn+0u%6#=zFG>yrCQ*4T``RRR>8?>L~%EhuW<+N8Zj3KHP({H?{wP&$nBOLonAbR zrt*AX_5Y?)QELYtns$m(7o4P2uLL^I&w>t*48YcbK6v%&0h;8@U+Os}ylNky66u(DV7xF!n+&7*9)rOrZgq zI5fZ%(c@4Zb&46g#IiBHKCC3;C{zR`gGu;BxKt-i=Yq1(cuO$Ot~yF_-nY;sbp+P7 zT!#KBSK)lCQ|^&6B^3PPVZJddR0 z6A0RS1>Ij%1i8j<1q-ySAjwV-qMc14ew__VERbR+o$Kh=j3q4m0^!t{LPpq(yocuY1~-k9pp}~%Q7*9- zyFyBFu2U>pEG@vRoCMQ-)Iy!}D@i1^U10Y*R#5+YonX_?Fy4o-33ZAraogK!H2q$T zMvZszowEdHs87Ll{Wkcsw+57+`E%BZtGKciduibxN$M28fvP_i(2^}u)O4g3_w>3^ zlj=08@>D0ww`KfZ_6npQxC^ZfO|+VKrvGzKqwnP@wEXD_3LL(R_H`8F25Sk#`&VFr zo-Pw#*Fd(H&eF_gAI|XfRqoNmdz`d=0Cu_h!OBJMJnt$KYd&YgglT^KU9eQ(?;jwT z{`tOO_w7{3`LG2_#yLPNARCs(jd_kh?>KCtL~G>I%*Oy!bd6l5Mpn-8C)3(f&FgP)z%S#8Jo z=cDl7+bH4<^!g(Ud2Ng3*87lGZ^9Jo1Q z85eOWn6vpWfy)73Zlu*RPV15m_&$|_pyj7wx>W;2*~~EYAG~JzPNQ5qhR+fAI{t#GQpTIgR|-v5 zyVjW`R}69X9nZLxvIU$@@Jeo{(pbpjbLy!WhFczGLCLj6+)A@3?i8O} z=t};<9gN)t>y51i#oLv*`TeK3&MmUsrfmi=`F$At@Y;Y?Dp&D$vNy_V24aDgABy%$ z;iF4g=$m&K7tQg29%XGb^o+u{s{HI}=~6D^7@t$KKEXZx^hU5Q&xULLWX2Wg+{fm# z?HID=8OlWFaKi&zOc!pf6h!v;aW_kDa4Y$ai?YRaf!@Tk+^0E~T;a~+u=9@qqh!b7 zjGzHQM)zkfJ6aAF?w`kjQyF*4x`Gq8YUj#oJE7a<3=9-U;BD`#c;?M1G}yNpQaabd z*%#Acb=nVn*FKKMZXD0=gLL4Pr3q*%TEb{kQEt_L?>Jv40Edt3dEWXBd}ud{N*{!w z!GRli;`0A@-S@bqX7X_Uz;7<9FPiHJk>GU1Bw!QYpFL5!mEMK>)48}{yqx$AuR6^n zXMaiV+m?30UD-+8{NcN}qLN5%@5Y&sF3$Hxe2&73`B%8$o$e^Al7aL2JJC-E{+zP* z3)*d4ME%R#1l!Vj1m+1FIe%XZ%9@=`=S0L=sAV>GJh!LU+u}(_q{eiX;~0U8x|P6V z?lN5Cm4n07UZUC+V>tg7K_u!7XIpm@&rIjDj0T$YShJBEScs6d_yV_ZMLPFy&A0$TICgB}f2%r^OIdjCnBV1bmmpxH+Y#ne{f@7RlI_Dz-c z-{QRmw_Qo;R|UsXFL1jWn*{kLRrrpLB$G{JDD}o$L6Z4JZlE-Y>l}Ow4wm|uoVpr4 z&rU%7DH{0S%nP8IE(T@4Q@Q2Nq1*(p0K4$@@bP>E=wJE?#xebny4?ug)b?`^hxc+% zBZ49I=XM<9oQ6Z8Co$w!H;lg*4-Q^g5V|7^FCQvG`&FlK3_n{O*OH0>PMMgswHVvO z@-U^U13oHb(w4OOJj3H1Za7m6H#W6%p;4k-?WI#R7%EB*)A-!Mrf}-biY1kQr)Wv~ z4Kxrp!O@P{&^8!=AHtMyLQyI#7<-DADd^IrjuITaRnIy9$%K5#6Hu>lKyYQwL~hHc z(Olw&Jk%~5hi;m2;1a_>@5(nYIP(PF_whx$$6^>4mJ2P`^B7KP%URo;7ay!{V@ zw$@?F`+GR>Xdd$@{!KFP1L%3`Ubq|*0Hg1|1F2Dxd_MIWX*ACuR||JYyZ#4iWXt(X z+In2`vJwWoXTlxh3P#HUnAbXeW}vu{ZXV_tPv_Dww^x=lcb>-u-g;Q!{T?5>x!_z) z8SL)jzhPe*p-RsU)aypE6N{2adXxg?j$F!n)Lij*h%XHCnUA@fm!R*F5O50gtv=*LA5)rQdm&R63Iu5+v5q4IdfTI^@ftX7Qtnju(7MMVK`?eANv7z5D z=TfSDGcMX5h`R&zsCP;pX5bujO_6|iyc@8D-)FW9o#>>TCh6Qy#6^13tmxFp~k&rUdNdx3W{{fD_#yK!Sp4c^blz}1_BFqO|FkE(S>k>!i< zi_CGFd&`sFWX-4AhRgKniW_;VD^o#k6uy&Rf%o=yLi!#VwrqMo9l%QRm|~Bmd77wk zsug&n96k-@y~lfB!nkfF2-#x^K5GVGsfHAstNkX}Q{EvsxiNt$I?rM0ToDg4G<3cn&wA9~s`CF~ofaF>KfwX%d;Kb#T%(+&m6Rwbxf8v|aQ zud!&@HQpIsfFE{

frG-RiT~|vUGdoBp5+*V04i+RfdRX3aD1{!#P&Cy)Bjunj(!tS> z6%UgQq$a@xu}Ru0X$S6w7I+saaqe+njb=`2;tS1EAZ9qM2ELD6GO z7+H-D#IS-HG1L$Jxk66{meBo``^k91F8cXCiHs7QC6E6i=x!OK_3L)fJd>ex=+6qe>d{r)=Ct6b zjbk77U8AjMN~Ctz9?7pVKhk=gmDCszOJTANrRRG|%13`_Y1B=n^Hv)vxni7Tf22rm z@vodlrXM8li1n;@7bi*>@l6@sUqgE3*_QSmkHM#c`(Qzbq)%OT(Xfs+itxZhy879h zX;n2+=*k8WI)QQx(&j2dATg_TOzRRz@ z{zrQkx2K$!jl|5ZQAtyva?BoE$ud2Owl|%X@0z!i#CY_>DNn<6`c3Y(CZTCs5O%|mN?#zQuu97KnsbpfZR`x1gNM>B`)hLJpf$wGl2~FxD@B>v4>qSP zo>_I1c>#Bm=UcX7qaVIwpVTk2^ypdAwX$VwP~dJUBd---lc>#yWM3wozXRCK$QIJ1 zh;^*MCRe^~?8P=MD50yF?J4frE1EuOuyojd0^JWAOAEh5(55bnyZMZahkciWL!`*z?vsvDJk?h&!rzN}kn4s-J|WgSJ3(v^FYq!|ZpuqA^6 z@s#o<=xT8HbrC!%Jx=nEHbebqv6cT+MVI^3v%VIM?5AUjq+c;oS)r1wc;3&3w@#{N zmldO@VB|0SmNtKX~4*AW}P*i?E2N?O6o}(ml3L{?fi}PpPf#*YTB%oS|%&`H-$z$ zF`@AxDeP`jypr7+2%9$gJizHH`=+yq6&*Uo!m3BGaf2QrWMMo~CXHq-g}*33QO8W% zc9!m~OC+_2w@_BsRVw_LtZemW0E^Q5LmO_@O7Bli=9cq2@)O~^=*Xf-<(_@2(l_B% z`%yWUHyk(O!*1PS@7`XP-gbzT|J&ol)3@B?z)sN2-Vv6xCc z+A;0kCe$Y+lzMegXMt%G1Sj23_?$Y)Rcn2i#kFi&ou^=x?c&(cDG9V#qmsTVH?x~N z!xRtZhVrz6p{V(+&(9^tu{X;uu$&&Tw7qed;=I!oIi)&MR^POYO*nXz8Va@g%(nE?mp_8_k^r6nuu`R=9u3##SINR z%8gc$HoNR&%g+7AiJ;T$;J+UVxulQ{@R*_4(_+3HG24fI>k`SjIakXgdhcf0n!WkB z4F#;b{s6XX$tzY`d5M`?ma}yUPwDyULdqn5;UcHOEr< zIAgkWX`%_uOwT5@qB1$9usi$QPC>(;YfIXTttHh;ceec46lvkHQ0e2AdhWa?kpHgV z%fg%5vWL~9H@F{j#B=)|cgV_oK^f9`9j;VwZBKev8Sx=}&o^rdD*H z6;uDpa+W>v1#|A$m+juJ%iC%Ouv;U92h1T;E-F@+_ieb(EbT9`Kc#C#O;^t}uISU@ zR*}li)0^emeJ*TAS`N|uWO`fpktG?8$Njo5tg$wTF6>E@GTVFc%OeX}ZHvk5Zst1{ zGc<}d8eU?elBIJ)Z_}q;x5z#FlyrU9IYqc#I~J5zU%qCouTt68ofdXim1uhwjZ98Q z@IWiccXFsA(Cs=!EYM;f-CU%G1AE!MO@Z7}Xhr>whM{$+k)*S?hMss`r$=o@u)~fT z++=?woy@NypTBO>q$a`NoUo22`dBip!v$1)VyBe)sXc8-`9Y1=dFZ+E9=%h`B(`rh z4_;El?7BZ!8e9Hj>x_=EqenNfa#tIM?pNf*5f03Lel6`v)syl}QdsY)MttRr7&7WM zkVi9PnlW=ZU)@$>&qlo^HQOS(Df%yM1N<@f-E?r(H>i27$5lT&(Z2)flEK7DOt;%X zR`R%z`dgf&wxdLb%K4e3+bNN4t=Aw^&7R~sLY+<@+(+806WPS=Bc*k{f+)aig*5l; zRd%OsPpNhD1g3uODlO~&jdiy2rSLaF>`a^`hV@uL&E}^`&7c#tKQV-keUhky+hpb8Yw;5O>@8^&PoZ7)Hx;vj&1gl8%q%W?v-yD$bpNFt?`W36 z-g=Ca6kT6q<%ZGH_3xo{hWknm!XKw??=6k5TPatC#gSUzLAdBJHWdxD^pH8lC$^>V zP8rnV!x`r9k;j^ht}>0!ePqomqxeV5WAsM#9qY2;GHvUVrS#ixN$;P}=55=$ki)v3 zvUL3x`!eV&Gb)-d-SB^<%vm*FTJ>TbH?8U~KUiSFmli0bLkBBSIzOEGWe*qp@6Jp+ zagcmy+-_yFl0<`=EAL7`p7e>&otYyhZLog+yh%DFcV@<c(-wE`iOP3y~i`Tw5qRU9j>H_*G;5(NtfC8_?9fw#lPIp{~)z@Q)Rc7ex#`U zF;crri&#Q)8twk~NS5;-vuiJA$@^!XBRk^^?s&p-aJkHqwcjti z)74AZ!E5*BOA*ENWqUgNI?9%P*NA5?rl){^n;{v+AEu2_KiG%^Cu#0{Pu6E#CsKd+ zMj9W_So^zb(!0V$Hsp?`gPOd-Z7iGJ z#gR?y5l4YHy%g8p7Rn~xS4eY`Cv*NVkUy-C;W6ryk=?C^5});Ey=~QKVNY{@)^Qot zI!9rxf01%=p$WNkn8jan@opBbmP{V$hkw*NtoZ@!{=yOpwqJVOe9?IZQM|C6%2ca|Dw zU#Fg1JJVyW28uecSFxtg5X$&*3hR52v~{03ZCIpE>zBPyCZ8=+r2Y2+V>)Evtm!l+ z841s|Vu$Rw{R~Sj(-(DyK2?qw2aQTIw#@1{W>@VYn|W=dmp6*YJ5QUHJWyv}->cA; z!LgL-ugNQFy3)V-kvOXr5A8d1{|e-5M`cMfz#K^!G8UP_o5B$DO*I# zNB>6jk}#Yb5ymI(-p8j!K433{dhu~L^W~be+i2|Qjx01foLqNyS9Z&4fye(1B-08f zsjp0W$*SfCUGonprvW$>GU(caUgm zP&jR_uTlIP=|Beaf=Ts$Y}v)mol$79Q=aen9z%EMDkkQQls2X`lkvSMO7-!g1FxeM z6C;L88{PHDuy?R>`kWjDIhE6YTQceWy+GRR^}=@dCNFxJZ%XqcQ<+IjzP!5M2YFo0 zE$McBu@n^8S*mw3BF2P-&(P(MM^x(J#k{5YCM!4w>3Pp-wa z@{!r^D2&dcAbK$k@xpf4lsYzM(@U*QI-up`Uw5-Fi==WfwQI z3q4QJr^P!J4eDK~&Lm1!890cn{2Q>hU0ZqY^r!N?vYph@zZ4H<)gtImBw6j!P$V5} zPcbJ1FTXSqMm|F+KFooaDO@8&r|qoRfKl$p`2uBDqRcCqCfvtD_%6tNBb5j zm|3`uR_*OWjZx|tx#=jn9&Ifd*6hI)^AS{O9gVZo)TPdChmye!$gQ7xO2yaRaHv>< zIS-);G>1lb)nj7vX8CD&JsnHAgw?%9(Tj8^>2`F2eA7ONynMbXe7MQhg>}zr-0FFYG=_J;q@7Wyxw9Wv%=S{ou%Gl?dkHNcGDy`~P!+SKO_>~;#sR?2@;pIMWrM7u)naJ4n9H!#Db;vMp^XK9MS zPjjgK?q_lz&zV#lkU;u_OX1M|F~SyW%FaJ0$Qv@$<*iGP(deIJXw8y*rpzh9^}+4s z?$xS_UcaZZRQbECP`}1JcL^?6o9EO8x4sWHg z$2(W3dL{A^$FkT@`oxOVvsnK};(Z>R-y46J>km`s7uZssS)U*;{aMO=!d>~@#NS*a zR$ZE|Wy;$e(dWzF&g3nFjHHDFPx0{?YjHeFpO1EN;QcOFi5Zs;FVQ^C4R&a8pC{jx zYv*^DFW6(?_Y7CX)jmIO5xwD&hYuiCs*XY7{XhB^yVW6Z9&WA>oiS2-Z`Egd|*oT;oSMT9$)ok zEkC!RHBSv3$1_TMRAhYZ%ro^q@|5gHtmTy3l-Xto_Y^RT{cU}y$AQgk;q+WNbl`1x znRVd%!w<VUruT{*4Mvx0V${EnI8 zPu*w#o}2LL+99?#olUvMrD$Ga`;cW%YR6qKe;m@asz$lR@gP54bC&1t`!5rCgMNpK)Wj+%jbq63wK_ks_q%+z*J`%)SOMGW^qocKUu0$- zEZOiyv)G{W{@m{IAojrPEPHlQmCgHcj!pR3gU9L*5g)s!DhPrpZhGfVxtqkZN7(FEJ)*OZ8Esmht_-% z&0t5){9t;mZ}O0Aor?6R&3uLG5caLglehgE%!U+ca=pMFJiXf#-hP293=PyOUJZZB z7k7vdEHomg1$PwYx$PNc59iyyx93*>%y`1A?(|yKo!wk;n1@&<@opL+JlZFlSv;tf zsxux_=NGy>_m>U7S<*=7H5=Kxns)LA)sD;~Jf4p94MyInGL|#VNLpQbf!6#@<*kQl zOH)#xGZi(Ee0{5UPa46OO^v{mzFqhZ>&Nn~vbECa_rX;CJ)DpLL)ooleoh zEm2g}ss}k0+Dd~;9?@hsOM3NeDDUTIz@O$_rmh<=uv4=`+3Kfme%kh| z$z=dP`*SY0ZG2*z_)$lXx6O3TxE^3bTc)a7w7umD?h`Q z&C`_b+Fi5V_1#H&G-4uc&pl2jdcI@HUyP;ekILzJ@?V99#x$P!UWd)GKcdu&X(4;Q z$YQ7N9;4$Sh1_r6Jjwaub~5bgRDS$dq9Q}<8EbQ99`h~hM1x|sun*hE$whus`Mm)u zd_=c;<~S*U2A(!%DW|$hU%swn-`h=N#s|jnT`p?>hoLhIsPT)UcymcgN@+kNsmNIO zJBKEv6e&d{WD1c;6#mhuQA(pwA%!BTbkE+GBtxR20Z}A#BFa?4|9iO)J=}-xz31$2 zueE*){gjQ-kk{;1m`S0&*eBs(Uo))OXG8^_w@K@eGM(co1pjn-+v=?=P zLI>{=@z{&%zMFAoC$C2-#tJ@n*r3YlD=ppAiw;INWO5*II6!z@haPniks=&=PncWIVgma~>Y|vSW-aqP( zRo0{N?YSm6WmN|I%_pQd&t_pcU+nieffjOJMxLm2mul9{J-?3}Sw|%qQgr`x%-;HjgcUn<33Y3DFqx zGX1b2)`HAXOD=b8f8>!gFR^ZoJG^?JH-u|3+0t`gt-xlq|W7A1bj;_)|=G1>SiGnzk*d;Qr0 z6IBbq;7126{rZe#O&yI6Yj=_n2VasT#f^AM{wmP?lSC<4j$UwYAeLSn8U9)wAG~^p zW0(bzNLD~4Wq`fukBP63EES6saXTk0!d0%m>>ckBoS0e7mZb~Hd_fxYMV`Ss{W}Fm zj@^a%;s&H;E1xra>W!ON5IL4;hTeVEG(=O3=K}ly`P)z7V*Oq&X8jx}EwBakgkJEr z8;)C*JfXIx6>cx;fq41ZIP$SCSXq98Ut2X$xh@FT*UX{4;z8KB+8$@>SwfVjF`7B5 zkjvKXWaraL^6OFgk1~>$)w~>~`gXLNnl6urz7kdL8n1D1q?3H7a{F^F3?( zP-lDy`urJA7vu%eqr(c&SF!=_xGoeryVZqAq<=b(X zHTNPps@M+WkHypDI-w|$zJ$13PlV73>+z7iHjJ{KkA>F5XorU#z7F#xJ$u)q>RKbP zesG`6Uo@HYs#6I5WC#nTMq`cH3bNl^0)lgf3*x^|<*w+r!IF^%Sp6swZ#=yRukRb- z!US)a7OFzsEcn?)Y#~Zlmb0cqF>G{SEbQr$;*=bW$a1cNIkszy+$y!HX^S|0#qXXK zMy|(FqbUMor{jnL$*}Xh3RPH;$ByTRu)i|j*rSIY5F_u{=`(0jNEn{v2+x-AyG^uUVo8m?>r1kH#BjrBi$h&*&NA24SMuJ54T9k zlyf)GdSUAe=yxTpOoE|;X5NUm}CA& zn7HXbOno9wJI`w4zNieCv~4|3KcY%qa}!xbO9AKT5<=t_{USfEYeUOj{*3UAAqL8S zS;OW?;xnJa;7R%@b?ZC%8Q?>lMo3`wp)mN{GeCZ9eF&?ohnuFF9#RiyBSiK1En2@@h3mG*-aTv*Aq$d z#ykwKjc6p3qVizBHLv6DZ(?mVugT&qg{1Yc6TN<;mE(FdiAd=Q7w5bO*U&pKHbIy8 z4{1ZU#7&SCOs7+aCgK&jTvoj8HQ2aVfRVgEw}oC8&c6``(sBv#%PN5$T0H}mQj9RS zzlGb|`w5=?c@A-nD#Xy@F6Op2qM_vi;ogh;V8og~P-1e8^%Qo&(-`~8E9iqUGREZD zj1$nbb^$)}D}rYl@sLzN(05vr(AVxKs0GPmV$u`vsG5NW-h5Za)mS)M@CTgRFS2io zEa;!=l*%<1KQdV-Js3VPfy(*aBs;y9!;i@tq;`xCCEexh_p!e$dUOOeo1=+~ZvBDx z={kJB=Oxx<{fdekx1; zu@ZDP=9AFR2Z_zUd?1_0gs!u{MWe^y&`s?^Nw*CS3I50HFUz9_zKh!D+XQP zjrct*9-qJ5#WUvRvB>uYxnNSmxod0CNY^lu@FRuTMBaw;yso0WBcJ$LUL?~j4anuX z9r(F#0~S=*!K1&w;6U;|{IV?^j4rzn8wX|7CIxiADbGwjLF5$`afxp$ z=`R|MA2;#3p3^#P|8E|+wY$@imukqz=wld}mdM_SU*?`S4U#Yyf+xqsR+>GMqrD*= zob;2i!i*`OK=geg9DaR@csl(DHj8>$_Aw6cw5l5hO%}nY$t%Swc zi3H686?(Hw9CfeMVu8~SqAz&?q{c1-yLTeEG}RTgV|&(N`-LzL#j>yvyg|C*2|J$@xO=HqE5#D(?Xw z_9Y@Caf>~JGC28xK8!nk7xH~oaH?<~6fS-TAI=@6b8G*=M>&Pr`{nOJ%3M`+{%i!& z&Q+-KvJczP8V?ya;FMXeSm8Pc*QltWx4;>u1^9{Pdq2Q~Bi%(Aw)a88_9MDV9>lb_ zXCOlqDtx@~G~BIE7wLU+6xkg;jsY_T_-Vxe{!UUs9ZOAk*{wuU$0*T&ub*(d*l8T) zejd$&v+#xcczm!*l{~+-5a&!tWCJblao)agJW^&`Xj$@fqGW2CNWB8VdsCH8urR%yt?^heDs-MEYx4+}VaqZx2Acgijp0h`H zo)G8X%P_s=4z4YUM+d1o;em!QtT_4uXKDG6%#m&+;c*29ZA`=aAsnu)i69Ek8gY~U zWn8?1pSh<_!SqyX5!;c7VMpuGKT}S$Kk*|5n4dw>b^*wijS%%OFroL$baBqxeW)Q3 zh`--&LSy@G9QENa=4-{{$}K@SKJh#n?f8iwjHE=`{idS$^Skk{Vhi@22mr5sWvqC( z7lU_(qh6a5ZQ%U^f0jw0n!_KI*1C<8m8Xh|XQf~fZ*Qsb`2wbmIk7=irii(JyV$%Prqi|T`&h4WX0qSmMwoTJzS z$@TpIFDwNAT#W+9T}N?Fn*yFP{J<@Uu7j(xIhcOJ4Xcg~;^op;xY@%MR^d-D(O8J< zK3>Pi=KYniqrQS|kG$ws+gIFpbganbnS{vhVJ&KvcVNr)L*PBuQPi|3iwCL>7X@h6 z5|SH&-miV|bNDrI9KHqv>T0n?Mn;riq$YBo>xllf0@}#?DXVIuMckaJq928MxWoJ` zRy3@|rJXrMx5%m;>YOvyPs`(ry{ziD}zSE*OS*5iZE%Y5#vkpK({mm^CsnE z+2WC+rBeKNW94mB9r6(UKBp!c+ntMfojKGupQ3wxG$h|B#$~Zh_~)y*NX9r1#iP@4 z$@9ZzYAVXQ8g&Wo+GglAe0!jQ_st(ax-2Z0pb^G>={gXBv3zUeOgJWGCZj zU!I_2P>$1|tKgG&gLp+T5zkxXBhpITqFsO!OU$wK#vOcKBO#h^Acv)mmr*-QU-V|= zV_cN{66?ewVcP49=-@Vo-pb1noSjuh>?ZyvY!6gnl2@Y1TA>^jjTWboado8HBniKM zI8MvfI)Qv#Cs7I02ZIqWsjmD7l4T=K(D%6^-pWL0kn_AzW%2YB~i$b~N8n zF=L2(!S2wpb9dAC(9Fhdl;H1?O41=4FHpTSj@B*TM3+5vBllZg(P1BNlgR2!ntg68 z<*v!lqGP?>q-SfDtmhnpY4*^p>q;v>n=bp`OM!i1EcIWKV}Ae0UN=?>W@c&a|;up&(DDXW4KPV-}O@8XrO| zgo1}Xru5eXPyFS&neH3s#F?HdAU1QX>9+%ARJEFKzL3~YM~4re?VjxjE6(Z@&5u&E zH!Ccohi{mGhSWSNm}XB_Wt<^p55`lmmRD@*y&TY3$H*A>#W?CwC25J=NDTQs{+$K` zl=&yYHkh5FQDN0&#z!&YH@$+M(%ekd#}v@36+38+{V@7zy#^8cu#Z>|n?v`muLqxl z)2V+(K3#O>F&V9fBz%zqmK{D%dvbw{y19W4IGAy^QFlmre>bUYTMQka)7V9stu)ih zhlHBN)0tM6>6YoYsEoA^z3H#YG*@L28xvWar??3RZq>j{@lKeL7D&A#EQ#$f2~??5 zCmM+jWcZtE(zbOe6%H;Xo2!3vYP_efC|iE^fPD|!G(4I_t{oyTxzT7~Rz!O;=Fl8{ zeuf`!PnHQsP>bCiY~cGiy1yqKtt^TJ?L*7y^+yjOfvlkQN|WixFJ44;p(M5U4~KLg zFM8NehuVtbscqsJdh*#xF!z@u$LGltW%mn2t00ILiCv&OcYcGUl9BjqOC(+L(v~L5 z2GG;unzPG`^SLt~eelt=791iPEmD2sdEP@gdQ}&@xPL$F9 z3CZNjpI_wNF9%Y%K}0H_qzax_jHRzD?!v@~=Ol6Tak@SJ5ZzXEj1C_zgLMx&+4&@O z?x%eL$V~EsEaw<_o^T!BuksYOXGFtGUUT#Ee27>51E6|m9Q0+Ea{Z!h!VQ}o@Pl1B zsK!NHdAxKC#%#2}ixZ;B=A1B!M>?4JWG85Ov<9ROYQcqj;^5gZ53TljvEpV!xL|4r zk{?bnqrWP6#qtDHn2kX3d&{6Yz7#6G)d5GDgBG7V8fj>VmuV3Qq{?O;xFsDD{`T1fQDdG% zrjt0V*`ozV$9!PT?|h)`Rt((Lbj8{CelU+~9YUS=4WwY1I956QU?aT?E7wkY0ZGEq z@Nj%47un?t}|ct5c!aEefV-ZpZAl0xE)j!B$-gm#zUP zzhs(lLW}|2H;v}^9#P!+_og^m{2Lh^vkHRa)?&rSFeK+TVgHL-tTeg6{#jH&|7>L} zEs}wWEQl?C>_r~>M?jwMRVF>h9V~;-z?^#quw;!P(r;=gC3_sMrRfW9kxYmj-vdu0 zl<@KILtuDuF8nNRhQ8IqLAw1H(-1YX1B3Q3ptOiveB%gcsn>E?Sp+Y%gVDFw6>ewA zL0|I|m_Is-jkDPS4)aRcxHq%0w8|4&ie5wayz^|*Wr1L7%`-5adWA*z^SaUf0r+E) zE9_pVgs}(3@sj8p%T07vDm$y zh1RHXX2XL}{O23+u}$RkcUdxK;)|_DB4Nzh0MIx|;INNS5T6vzbdKy{k2(}su!9`> z6w6}R=_HVDdJgty%GuBxeT(R>EVA+LShA^499LaB2IU)LpsMX2j8SM|f13tj$=4C& z_Zc&$P_Ymi#kAP=kBMxp^>M)HZn!?QkqncSpb;<3nfDhn2z9G~*|%Q^7y6jPr|V-e zx=b*`|MGSCdOn@yeTac~va+~j8H2%2Yc3*N47MJU$HsZdEa!VHq2^0KXIVVA?b14U zaV`Xk>|bD)jjGTlN-?emluB z#9Uw~q>%Ia5rx&_6e61*LGMl{VMdq={L#JvYffd5OCCaQWyCb5$6Vm}?y2~RU4=Jj z1C@^62hFo5@0lU5IJ$pat6;~B<39JWGV1vF2e6$YX`WkNtOnk4g zs;C{@;--3Hn_Vu*@tVoqT#y7(+fmT3b6B7@$rEn%o)OGAy_?H2Zs&4B9U#}z0#d+y9}dw+!Sy%S@EcdDBORBg$gb46&gc@6uOc9Eq{9ZMA5 zkZT?hN`g`ZobBu)VZ5HZ@YE|+NXom-GXao za*Nc?@kk>-uq~^BA=)W8eRV1;-Il`bPIzL`zcP(T$oFx2AEN~Q>t%%ZrY(n0HY-{4 zevX^;BbfxPY9bN_XE;;sFi=n|WB*;#gv>YZS!&4_p64==**~gbFZ~V*RP5upYukp= zry6?Xvz#(?l$|6uTDFkPm9avaQOaumP2sG@OyndvBw=T*EEH1JLFbPoyZWfmBCc`+ zH#P4RnH8GMeE4pt)V#wGs`^Y|-ev~>0tcD<^95Xs;SXl(I3(PyzX#mR*9d00O(xN= z`viY-x3jc%ck)TykNec54Ik|z;H|JzP-1kQduOnb>_}V31=lc(Njv-5$aA?wd!{br z-;k!89|mzEzT4<+=SNZ_GmBZ={=<2kQzV1?^`ARPO38Bs_Z4NsJ9AxdQ~ThXD- zN(P^ixiid2SjbVXdEz)?qgM-mZ7iYbLn=&4@gV71s_bXEfQWAzjUk*D8z%b6)ur^3 z!a+B9FmC9|&zmF3&ZE0Hj}g^egO&uBXmf#W$i2nQEo|YY_;m^Q+HAI1xw=W%yvK|< z7wLf2n`pry$h=J|UUQ^P5i-P4hXsOL1I4lReGw zs}_vPe#!0Ko5@brWC&$nw=?mZUGO05AW?X?gj)0)u@7^2W{<*7niph_mh;?T`OG4? zoSy|Z!#~3jR~ZyM?Y20)eg@zE;pcN#N$2;wtqA?B0Yx)sz{L6mWKpd=H}aMX5$yTM znSd{rWth|WGx0FAU7m(a4??@)%g}3o8L`^=5?hy^LaP=9&g{$`E`Pl}OTV5#;wn;v zrQ9?eHj2NmwW3MY6o3=sh|q!0$Y_IQtWOKmRLd-KYdN zc)pFW_$fEmbQLR`|A%wY=eW0g#{a-md1}losKv)vZc>p3B>$TV9j?>3=;XD z;+C?cYvLWIek2FBz#e(cMI?X(-H6bIQjq7t$28nTd ziMY>nQm^@u3m*=!Z2W!rxTTyGh!&C0=1Yjl)~VPT*FXmH9O+WCpWN;>w}g65sVryZ zUS{E=MDi^nxZodK+4w|F;wV~8r%4tNJNp;JKx;l}t(hvUO_Ae$`-|aC-53(8R?kHh zY$DZ0nM_|omt{Um=LXzN*y|bp$cjDp$=SLVi=|`cSY*p9f--uuAf9ur`on=(q$LWc zJrr?v`5thntcu%u!jpI&NMjpE9pL^`8(_h`()8?n6E6A5dzPUCY}?~rf%3Wt&R6cI zP)XJU9dZr|-O_WZVi$+&EyJmWn<;{R4tHW?gFxm>G?=Wa0hi$wFymJ|-*r}jnY)^x zX~}l*Ubl#i_&8cH*qF;?tX>1=mx2Uoxv#h{H{S@P+>em5x@fjptb@BZm$5Em zdA6VRa!XuoxSwmZIVT<=R&h3(-mWtzokc3NvRjdSJhF_O;-tyf#pXn~G`ceIa|u)K zFn|rhKcFZvQ&2awgpAO=!a0U*F`$G!%}^pPsp8PSBbAN% z`Izl_KPY4^A!L8A8f1#g3xmyL@%zDx+-~tBRAE;rTnHQkYAf2=!XRb%mwtuyTU?_f z?2j;=b-8SIawWSt{hDBBaV5FI&vm74U%5r*8Ek!E4BpEyWK~N(b4njC3;a(}82fsa zFsybmp9Ozpad7k~GT!8?uq|;ei)yri$ES*kuzonO^c}EXV+_=7NTkDjvuMvvd)jsD zFyAfeN}LpJsNpoS#6_$7<4(<;!SHpaD^@yhc()mr3E;I4TgoPjt?Wrv8bY z7XO)O&;#MaN!;*mIws&d7vWWC@u@G0WaUIsY|o~PmTsW+qdn-t_7Guk${AW`l}4T} zHzZSEJYeOnal(vc4~T`Y4c%k&5enpY(t}A()PncLFO5+p5y$6{Yopt_6aXT9av43q z>rCRG-ji?Uj?_MW57Fo3=sNH5H2=MbtC4IbZ}ryFq30jS%|jk^%JY-tftMedo^zi} z@zx+ZGqTCU12f16xjd4gU`lt7A5UNCtRS4+c-oBu+SPKD?tU_go~}$JNfQj|r(u7| zy?4@dl==wn$VU_EzbBLi9n7W)I!8#pmjMmDRK+upCc@-u3EKayoQ&8!mu}tSN5c~b z$e`a;veM0)+_Tc4)1N7kg#shN_q566UCBJ!HnW%>U8O^xXzr#Xo*toocS~tdP1P0U zq9BMQOR3xJ#pD|t=CS{tFx~wIQ;yk0A4coa4RiFU(#traJIRw=@>QVRxzl8v*droa zRL4C&PiXDj2rBi{5MDn!Nzy;3kOzNx-r^@M`ka{3_G_z2y52g$-5=7_w{0;E9E_zy z^E9Z}?QZh^tQGYx%Z2p#X(Tak6HynsQSw}a?6{N6-m?9)`HVfi@!%NMsdAx1hhH6&TQH0scHa_H?y8t(s> zxQG5DFAAlovriE1cz2jAG3w?yzRr++zJ(jOyo%&%q_4(Iq9ybEX}{$*x+I?W z1Fd?(Et$TDXtpd7u8nxb>E3=qGNUAEgxzxbFD;pydTTgcwWWaC{&$EzR2)O+*&Zj? z%>B6j5^;;}k{a@}!is+BxkKvA>&fX6Z-}A&L26Vtf}W9`N)>KZkfCL_NT6;yaT+5> z+w><=zljyZdec@a+WU@NC2B-X@*Z7vtC+f!23PLhTTC`KuOn{!%vu+zNPQ*lkrM}c zxd~_7=?xzj@<=z-{Db8H$+|5=rTEU@&^bM%#pVUs;+s@yp81$Gdf3wU(jnAAc#r%I z%cg4UH&Df-No0p=2(dX5NACStPVR3X&DkGIA=}o;(q^?55;kgpvZ7@9FR>XB%M2xhj;BmVDlS z*tC{*I)5R*Ps);^GH25F#fG}%YSIDyQM9aC=S<+8eVcEe%IQwk@w08Mo33bJWwE|Ag_YWsG z!0VWKLJTO}N$+UhBS&&yk@lC}bp5syn38@DUR6b7R_jK-2VyS{Op`-fVn8=M(tych zgmB|cJRbNRh>7D0xn$mNedv^=$ohZ+UXYK67R4-lGdl*Y1FnJiqI_&DyaIVI3&`0G zx6yxDF2;I)!fl^D=%1}ppr`L4a+oz-q`4=7R&0{N*>Sh=hmH^x_juD=1E=Wz7-RB% z-c_PNYw$#_HEtgtfPPmaMKJz6Gw4b}+wO2uE1^lmQln8VpcFT#TH>0=0@Rt3#GY2% zV(Ky)q9BvaR5I`>#)=eB@^%Z-OJXQH>o6)tj}kS0)4@yA>zUV}54!JE5>4njfa*@p z_}^VaW zCeMsScfXdPV&G%kxhDku&gsL&ee3bT({*TiG7Xz*v(SGilo~CI6I`zT8Es_Y(Kj?uNF(271gV6_3wZg++VCMcpKpMF)y7qS6OPu86@0 zo(9PLq_L~L7ZVD$|jvB0$n8DMt374;S#2UXz;a=N38OrBNAiS=vKE8g{dUidQ~G;4*&wW0W1 z`xLe)>(F;sMxe!+?4$Z^oNU!_P#xe<2saSf%--iA5n(A7gR;L zeUY5ecVC*sVz9PgHynB)B{G;%joU&a;H>WjtdsJiODZ~X>hbSf?xANmEKHnzyi!Bz z{R;5U-YKHkhFH`ZT7!2+xnno)<;kHk=zVyW$h!AB9{*hmcdlo`oiTgqbxD1Aq%I@U z{vau`*WjNSt9RquT4P~U!f%)xv6SLo6>6z_i|5Nf$J5@it zJ|ka2>VJV$Zny?Iud54fhV`(0l!cDSAgd%Z*bQ0T8Q^zNBZ==0=~Fn1exJ>q*1DoKb0#e?|4GR6(dA{ELE=$=|KdT}%?du4%d3g>|6=?u8{+KviV zyVDG(6k+TtUm~~P15XsaAol}Y!6?=e^=!wp6zw0Z|4K2NA9Rt5woRtr+WLt4w7;-G zGM#EopH3z?ZfC>HJ0LGjnZ7bGCnt>K$QO+mBK~VS4U4@E);={Xw9cLw9Lg8OI}H&_ zo^z4#xS#tub}^k{<_mKgor&E~N6z;51Ni;nFqc=u@85?jLHt%*qIboBlPx&{;@OJ$ zZt`0eRc1)%a+k^Mtwv}*iGl8;+brI;N~p)302zgMP?)+J!y3oIm%S})zwK^F_cq43 z0)2Fm&m%p!23tlQ!Aq-;Q_DRCaK_1Aa7CJ<37V;B{4)gSf+Na29Et5=CGcF*jZW6| zVhv1bIy~ueFK7DW@G6sszn|N8QhY|rJ)PYQKP{f@Z#VG zth+i3Cj8m}-!jG#*Y%NXp4l$)HT@Sm`BEBgOT1&dmaM~DW3$2i@@niYKftV|9^|ga!^Yd%Kj<2w6-y*X9?@lg^=kI3~#^BA> zxiHU80)3OKpn01=?)y3nUs*Y`o~w6=Y1H1l6K~=Cx{RCTg zx(KEfcR*?4N_ex#goGZ-Wlht5vs{4?rl_W%{fYy^QKl==srd=X+L%ozek}xVr*ycf zp^WC8AK+~0ZE~^Z3wa&>9^4O3flEo6yjP)sL?rX1xUE7`k&wy4O9D7Ju^PmkKeC$k zxm0fAFxInClUkM8LFp_j%tICGU7kj5-<~8d0*g?5<6Qc&Tuij)(tFrd9*=pw&&V9z zOHdK^gT!{NAxg4@yzUDHB4J8LYIu?GH~Yyr2Rrh2Y%u|M4I1y+BS<}v3yHVR^S;rI z`1jj1dUKH;Jp!3feetvKsH81q7Bf!>RQv!sg2Rq-T{fiCNrF#;mo>fd7er#R@57DM^B@~+Og-1s5s96Ppu_JNZZ+*9+phj2 zVr`F!R?v#$Sxd5rWaqc5 zqeg21dafRU^L-BC?z%}Bl*`{M>#l)PQ3G6wWN0(`9}!(x2z(lps;SJyua9_+%#y=6 z7#aheceCO3=mv~U{(?7?v{3A30d`zD3P*Pt;JDK#(Dj7}=nj*@GaD|0L`yP#_9dJ7 zl^n+FXVtJv-3fith3t%VB>bGeh-CCz(fd}%!1%v~sJvSm{)<0^#<6RursgEtXnuuV zyzfq?D-M9|F(G6?F&z9n8pW?ZLD`ktX@#&1&6e1J(K98~?C)WHsX~;>uE00BV?@90 z4T!cDBjbMXx^oSOb*{aT{re#+yIct$6yCzg*Y(sx-4xv#T8Yn;-H^bhp#Ggt*kQaE zr;R*>A&m>sMX!!+*tHjTI<~N)_(W8i`HTDVr;7OY19iP~k*RLa!Xsl`P$X-J9lQQP zXL}5%rB+U-qK?K4HIBy^mdm`nj z%!Z@Lf~`@uOdYB(MB=;hwRrDSBuy%v4_Bw?eAPdmb)~dk9L!%W&<7U@~R@DAW#KLH2&G;@q!`aWgBPviZi@^xI>;-+GH7{_vj& zo{cl;AKt2e?s++e?5)6#Ra5EZP6sZ-{1*3AjA7)FRDsf^+q~vBqtazzJN85$A>1Nk zwEHg+Z(n@}m%nvzAG=$G=_gkTKioUX_Fl^*&wgm)>jRgeL_wNN)9+zeUj$)e4EX1& zN#s@iG5lF1LG7g$aIR0fg zNN!eFk+f&)Y3)ZXp4*f_hEE)TqTMSYCLtA@dA^0;{M8^MzLDNnuokwK#(=BuHp9ERuYEv$@fMsB_yklWWwADV82QDs#qI49;l+#s z*b*8D^>Ybry{$xJJK}I#yCGfPzaPGe{Gednf1sc+$oe*1Wl266Fu9}}1dFWLHtCC% zj|QpDjAKiY*GdP8K^k)9hbXbgz9l8FS_y8+7b|Q~a{o<3>EWNczZ5zq{P8 zB+< zovd02k8W%h{8?;IRHxQ+n^RA6`eh$@9oi9^7wjcf_owmv3011T>NuyG)XjI(UK5y3 zkH(Gb)5w7*6t?W>wb;33B$)d(VPbX){*ZbP2l<-i*3cwa+Z&CQvlTFWz5#g6ktJrz zyCBU!6s#^8qRSAUCsgPlU)=t38)s?L6fc7H#~zaAL)YNvS3hz??iHas{OQ_J^Il~Kl8czD%CXzK-hq>zylgWbP@^EK+1Izv~ z4Ri(?$i=PW=z%xRFou*1Uw2;>mSjm|eAhRaVjPTWU+&Y^u-&jtV=b&w%_S~_9%Sjh zt+=%IKfD#W4`lh7-|&+d@yVJ+tH-1fq4F^HD1MiQjl6vU+i(pg0l6bZ;g5!`x6VF&n-q8?i9cS@e97 z6Zum(oNy<~EJmpwh2>-$9Cq17j*flB+#O~D4;F!OU4O~EjFD8}A%(>k_}$_MMY1E} zjk(x}RI(}1lIkbbk!?u<^v}}-+O<#;Up?BwRaHgd8zTuwa1DY6({;jp=euyitN>zz zJz$E#aX6!QgLpc=2I+up*? zXatNX0ODP`3#>2uQbEaO@G~tpcc0P#8Nbs=XL$?B=uRe6UWGy4kOIES+D9hrx^8hA z6~QLjjpx(auw6W8HEq5MJhnX}xToe^?Pp3Oq1 zy`{o^lcZqin-Unmn?y|O zq8GZ$4q~(OHnMcf0d{xHH8v3z(SO=UxNQCJaIfmKz@+yRC-JD1E>Jr{%y#EcW#66T zdEgq0%$?g{rbG~#cUcbGf_{=@l~ksbRLXXk_rt7^WUTSbU{`;2kyZ9b=murpW3*cj z8(;N-ZLNq!-8%pWc|XSK=0ij=+MRlb=W-#oVz~T^0l`rX%#zPb-D|9}?aU$rZZG-CUV!>$NgMy8ED_Y32E2_Bj{7yH*LlG}bxJ2IFSqzd3>@9qzALG`kPaqPjy>JW{k2hB5 z!@1~lAZ{x|N`h2j_HtXyG+K;}ru=8vU(TfFUj+K64$R)B(7ZpYG_c+gopnaB?fEl= z*DD9eOOpq1yWk9z4QxdT-Io}ac^yqBsgdlSU~=lHmSBtOA+-Iymf0S(L|Kb+5@$P{ zli=@;BU?7m^oNG%T(=!>1b!yZpAi{oOQ(>f8Ztma_QrVm-GqbTSHJd)YnBVHlvBO?R5@ z!<(;GqwU`d=*v>z+VMSLnl}=gy*+5t&x!bReIs-x24KiMIV@Lm#`LALan_q*^xfl^ zT=X&%GUkLm4UyUeN22&Hp@#<4Pem4kyEIvm-g(|{_X*G4%VP~4OGv0>BpE&=M?cH$ zpsI^^qhU!D>Bu_`PD&{_Z*&~F+{NdU@48@{p#`?hv&Ug?p2ER~h|2%&-GN-Aa6EYI zE!-)eP3p6kk}tJk@T=Sy1@;MKwDelMGBW~4rz?~9N?)1Sj=S*sK@`1o&X|RGNYREH z)3G(JfjCY{BmgSe3sk8 zGp0QowLymWc-4&vgUw4mVNX;&*EAbAvv+`*^USGA{|#dFb18fa?_{6Ebm{#+7wGMo z<8Z%SJoG~`@l{dBcjJ1=kADng4s}qWP?c7H_y*4v7>TBu;BPDri5Kc&U&vHynfHyf zYG#vtJj<z5D;p&|sJ4_-|kx-O)D=pT@7C zeP4`Gx1fl;*wY0yN_WZkC+avl{3iSF#sR9--bCBV%_$bzkwrV7kZ|=+G{9XR?^Zvg zD#~1?Z`X4&ZjBdZt73tF!5dc!sElYPX=ygV@zbN|vzl%)B@Ln7z>sc^6rcU{aXR|G zB-Hwl~1N!|3z?V;d-n! z*`AqxK1{q%kEB{w&v-WAB3$xC9^31rxk-<1ky*Ay^z3yH0*B7Rw{P3&`jQx;Jt&I@ zk58wvm6F6@!aVqVLmOTd<#TD$o^+UnkUM@hl5U+bfqw=u;yKhi=}r3vh;e-n@p>Zi zQFb)#<$ohS8y%@Z;W-#T|0W2mg`hX|P8fG_4lVvt1V1Fs(a6J>==9^m=(HurXwJ|Z zG8An|t!7A3kGE^dfqlQ>)JAFQJv0lwkDM0^-B9iv&}^vKSKKS{^! zrF4AWS#oy;;}Sgz%@?m3MhB9I6IXY2oP7HOIKM68vai;14?kQYLgNM8q&-XV zJpUExQ&6Bct0%($44r3Oj$atZ+lh7>lGPrPhI;OEN;D+NXo!}PEro2M(pEx+N@gVu z71499>xm?^tfWL4Atam1>i@rAdDR=AI`=u(_4|FlqC+BfGh{5krL-o&d4@Qw`UFJ$ zYK7SKDO8`opZzF{q8|=hVYo5Jbazdqx56yR4nA_Vu+o#257Q?eYY&3ezys3JmC8Z` zMG!9GM;F9tkM=mVh4-Gk| zp*wjxR~uzOMn^^Sd6+8B`TcY5;87CF6Ga=CuH{EMKVjug}hA~MQ$uQPMW3eL30|dcIfwv$< zHI$l#C(_P=<)Cc1njGFXS-6~=MU@3BiTtFyr18l%n6qpjW`0@&>0dfwSKK(tLL2O+ zeygdPm3#@#DEreQ^;;nS{VH)t?joh<_)eo`5-gsWLp*CTiDS5kbuRDax_5eU>L>N+ z+-_BzuTV#n=In*>ren~$>mzG%DiF%R0aA4=6qj!t+V-y$- zeB_eiv&ku(1P$8A$^P070=EV>d-fs{Y;tZGz<|@0>|r^JD2dn>&%I8s<7yoil2h#hQYx3u{XKG=rT*i& zY0Z-CTCgo7kBR3VT^NUfykAz*BQ6 z|AwDOGKhG9sqltl9v7+ULBidKaglq@3rN2tYR*?C-g;w5e&;%Pf7cgy`Xk)EH&Auk zzE7|$JOVZxo(K*v#i8VRDCaqGy|Cv6|59HX#l9##W`@;C=E;u%kV7w%*A(P2<}s~Xm>eZ%E_5NGnorHO3rV(edfh|CpSW9OU_ux*bfuH|N+ z*kKzY`OSrALXHFN=zKeaL2d56>nV0-+EO;U(vl_qHw*?h~W5T87|zTNU#Tjf1UYr0KV~V_Y83S@L}M zo>Or<$qjzd7yj0bW9w{Zld$nmg!g5WIBTQ*U~5Up!>FAs^YS<0a$!EJ9?D_5%_@*R z`y{u`(1}PG&0%^v&p1Q*{m}P}pSAX$6`lnpNbxGO(=pu5s%z7P55g6RUXC5#$5kMj z*G#y(!g}U=yP4Rs^rP=6OCy5(tu z;Ot5Xx?tsK*y^DUSF^{%z>+wa|2m6wI%PAjzzgtV;Wlnlg*x26d67$U)M55tWtgM2 z0k=I(p1HV|a6XwQgo2fuA@Z6%*xV@t>xScSCN>Z*2dv~|lWid4@F;AuTScNAq@XSC z6A2lv4#BUsvY<{)9J@`#xKENK(cu(#xkUw@bVUgg22YX;Uor(-LO7yyG=&q>l%?HA z`Z$><FO9^n5S$Jd#Yd6I}Xs?j@1~+oT zf?{?icctLUzH6YndpQO?H^8|@##~BGAyJyv20v!sV>xAAoao$0QgwC`00|3nvn>M8g<8Oj=S2`~9wVG( zl0hm2Jd16YA*>G>g|495FA?>lzleUDx?@=ukpxuy(mjg3On9pNy~|2s@q zX5?w(1rji_93!68gJ@_x9vH|3l@(TO{cuIJ9#FxTGdXw{-bmIeCWCyBIEXn6r<=88 zuuG6iZri2{*Zx=!4()&8P5cN9RP2ZKt37Gcv*mohSPrv%#qni@54!Mdr5Yc3IAy#C zzD*8gZ#vVsbC#CiIdp^TU7^dpx#@%-jS!UM+Sp;Q08pqdV6Th+f$OCEFv@Ets&*-& z^zCu5Z^{x_R_>1B*73kO90RrEp6KF}OfD_AAi?v@i9y{y5~(Kvk|(C1adb00sC0t2 z0Z*7%4Z#@`ErgpS#nA6k85|Tmg`J~LgOlht8=5=_r(MTl?#nvz>B4zp(=`fLD%=6v z8{gq|?Ok{5~YvUR&jm>?soz-w9u;Nn=TFvT%536F~U7Io9s0}8HZVU0><11|A zoDSl@e>z=P7RQ!E%>{|C5#ape9GLx@Km_4Uu-2s$(zPtfp*MQa9<5IvZ{WF%6J9d$ zd;EUh!;fWDy#tLqC$YV?95a&n3`Fi8D9HryA0>}TQ8AzfG#zvTA;Cx_sr;vw*ybB}qw^oA*cIaseJiFeh^FtF7cm)vMz4`rnglixr0@h> z%1%R3&q_SF*M*6d?BmupzJg}{E;R;cg1bu{e98$${g5fBHF5}C#w`ck_!>BUwz4m{&&B``0D#u$qWH>X$%WCV)(*1V1(` zB1%(t;ER6=uq3<#QZm+H)eN53aZ49hJ@do4i!4BBkpoBX5OleG2&T=D!cZp>>*e!K zVY%l(Z6%^Xy#)Qq@2wWOEW&9o?BJuaH?DQ;V5@Z3a zaP5K*>Lok#p0XR<4k(0fCX4?G$3kKW!tPO{uwKdrXGrl*zQ{2|9ey&lV+I~^XotPq z-^0|DLHN);1jSA~snukvP`b90y-v%A2f+>OPfHi12&OXM{bii|2^-WozZluOB}w$dbQ_(>Mbx_?#v*mEXd?im!rA#b?1?OB_EgSHM?SEAV1eLPz~MwjQ(WU*_@Tfvmy)y{TE!mR<;_~h$puv8&X zw8$3|DjS9I1}ga{f!0|M^yO0Czr+CKIzzVn~IiJDMPZp%J0fOEOVKbjCxTtjj9^?{E46 z>Pj}Ga=Q*zO1*@<1HstnV}pXrdr{?T2F!LJi%#KxnG-)(4V*EHWhA#jhNCnNP7jAq zyeBg^WEMtWlSL{VhFc#0Wc|M`dMP@o7Ss-DHm@PbF~nveP_=&>MdaZe`tD-SKCc9}ZWz#%5eO56;p;wz5H=T`UbF z0^{%S-sV0md$bh-GuuE^*$qPFvyiel9A~IL0xkE2*k88=M1q^lX>Tz|=Ko_hz8kN| zYz#-!cAkqVSqoPlS)>2EMq%KUG>Gb42Dhs}^G@(dC}lGT_a2Euol|?zt=tNTs8zg3xF#f{}d&9l>EaU)Oj@KlmJTE(C9fsP7&rczf3Sr?8#l5+|K*%raredl4B4bLT!{4H$Em&Z>@);Q0$knf>Zf%-HD zP|Zsa>JI+~FMeucxw0O98ybr*yVam)mZ7LqsuW)Z5pu_1J!ll7#Xerl;CjC#a{FQ>}SI|jN zIQ5C>Ynj66pk54&kie>+Qlfo)Ht4Tq1cuzn#|gF~SUS2>5M=jJC@Y*om;HAhd*3Lr zj(}6dTsjK{3Ok@aoA+$aU4yISmqO0|3f>{Ii$2p2i?yZ;JS6J=#sV;dcWs6 z02+JnU&39y9HWUB?`6Z~?0;CY_!ZO$M zJwhCAb^=F)?L|Q&zvmCFV%o37@nqvQ^vrfa{aLTkVvadon<0kdcOJysp5eIuQ3M92 zb6ER&m9V^SHu*JkBX*>wV*B~=I6PGyhxmAYXu<*fyKs<4&;;R!J8Gf{%?`LmqYO(1 z%|uu7t5NS=9T{*xPIu_;hvQSb;HPsMwy9(ahqZg4y~Ir1Jn#tgO0D@!?Y8#v4QSi7AQ%6$|losIFiQ|1F(e7>V<`_hHYU zFueM-9V)y$Vf49Dk}x6%uJ+aAk*Jd}XcLE0FZ56?coeGtr-BA6N>OC1fLnXbvE%nn zc%pn3H$0ZZmjC)uX31@s#B;gK*kLyD-xSn2as|E3zoNsL*LdjAeqpt@1m*{&l6yPN zFm{Ur%!+w|qhs>za!SlOvEms}<{v{mGmc{Vf*CL|S{-BM{-S~Rbo?)9t|&#P5&KkQ zvA&}gk577w_W6~l^12x!;&=qor6yYH;QH~Zy9 zQ$I|{g-0!M&+%&fvCv$o5z+{zo7zZbj3SOsX(Cf5m2;bQ`nhf0^*G8C@t0RUiYZu# zS{%D^`rSV``kOQ*RnaJdD`fSZJ+#&H1-WReMo(3I1|JO|Bd0yUMn3oJotZ{-UNSgZ znnlXiS)=LcV^`k9D~e*`g_!;E1PTk<(TLAGbF#zXy2A!*SoP~~|oaoX20eYz{&Q4q&w*?e4_XNO^LKQk3`KR7H~ z&wlXPfsYcWVXVP2ruKU(PR^Dga~;J6yW4B<>^5VZAe9IGg(diOvpBY{)LPyG0Zvx zN@knTNpu4JQ?9VnhfVR&msGo@UbEQkvB9Wk-bX6Eia;Z~9ya!m2AeNMSYWC|T1@7W zV;P3{G~@y5hq{TV&u~1c_K)pa{0821HNu)gMN!HAT{u#0g(zI+D4#;n#r9?a3SN%o zjAH|^Ao(j%|NTQaJVOes*>RRA>&aetTornZnS{ESLso5Xc9=**g?Y0c~~^44DOG4kJxuJ zOdaTiDxNVSi{If<-W-g$Vn&Q>l9;8X4!JfHN&l~CGFVtg*xh&*@ydxjulPr93yjh6 zjHR$Z)Ck>mpULmJiFACdGnxHp37n2Fz|4^`oXUAG*0x5m;jD8 zzw0$bqx%}TY5H@)?|V4m*K=$KtcQQ6<2!xp)|8lYj{p@;JHS4^z6P4r#OGVXMgy%iEhv3Y+9n zh0l&94~?awWkIOs>qp4i{v+1oG@YagWb;L(KP=g2Bc^JojoO zckZb&cI!&eTOV85Nas3|RBB0!_08!i{7I;1I+d-9XMfwI>0osw{N$aUb$u5(Tc;s7 z>r@E!UUEd?aumB`8ACQYpJH+wl*srnj(itCL6*-BBFc?s-u^mIZR%=kQIDquHpM_uXJ7Iq7T~hOEfFyt{r~CXO?8c*lT;U2nd$Cdi1jGQW2=H-dechXS0@ER;s zKLcNTGht)KZWtlPyS)tVk+y56h?Ss-jQQ$-4=UA&``&CQ?a+idVpjBo>wGrp!Weq% z$}qYtSrtE$o$yq~A5)W@A@o)a`7!1Tix^ix)CU5IZhjWXIKC&N6Wz&rYZLgs&<$Vb zjljE~-ocZn`?wdqX;6HkiAcvkC;xJ_=yuy2&|WNqBb=h4IAT6q3`a>{>|t2J`Hl3 zL|%*$!;&e7k<@GvB>v+)Fm@T>P<0NhHynfy64pfZQW9uO&w+rtPG((p8um=T$gSFU z1WUgdqhvXPbFUma-}uN@I{MS)X4}avr4z(VbW!NdvqkR5PbQK3)`AD0wG(%hV@_o? z@ZZa|f~b_qR5#%&_(e)_`^9&WhHh`7bl(u#D*|ZsOMVBhWGR#^eJI$VHH=ipjUbVG zCQu`LIdp4y3CmC3N2?w&vghnDXk0T9DrP^&f-WU`bm~33kwr~JO05)Sa*ql`v+rP3 z4}Ufq&&RJ1{v)}@{`8r}c8U*M8P}!)wr5jm?&)b{SCbsQgQ;}!u{o5Pc*D>7d30E9 z7E^opjabyhkza`%Nii75Hb0gl^^2UzN*h^Pen^9ww4Z{@tCET2NlPku^DIQi?z|EdmE#PniP;zy>GZ2 zeh2BhFONj1%mWJvZ)y@UkGg22@UyZ2(BDu>BySgyhUEst+&zxi6vqLR^oI0r%A;YcdWCZ@<eMgfwnbK~aGlHF>_5O~3Vqkctshv3EZ;O8F#Q>uyW7 zty)T}V-*P5f0$^At)!WvWn9_lcVx7DEU6kVL4Eh1A@Vv_bk|#bsIuc7r}?II7R+Qt zOVY`ooGv0&^qTA#9>@LpkU{2uO@z$agS2E(1ox=rGdJwjIKqi)*qzpWq+`wtBD>ju z#67BHmmWPK<3`ByZ&Xi6_GltwLj53p@&)pPCc@6&qv?md@8JEdl3dbWLQcO7CD(cX zZR(9CVmEaHS!1tE^uCmm*g z{6jR8+USJ6YIJbVR_fU<1=}qQ=y$t*W^Y8ed{rIJECxxb(LZ7J*Ct`Rz6@1%Sf{F`gQ+Y-(?U+5SIGIhR=#QnP=Z}(3 zLmzVN`C1t5_>vUQILi#lEV5?nQJPm2!p%JLjq_MHhPK{T27~kAB>DYI@-^i(cPy%y z98llGe=AQAQ_1ivmcwt5tf0r-vw*`S#&{0hGp1Vo%Y?Fzt_D$AeNa>MP z>B55Y1+;$SO73iIp0GNqiE!`D=wbC}GWOX^ZqA=%a=`bi&}EAo3Cap3@BB{_Ia8jD z%9(Mtjx8j;S|9uF7E-?r=c?kDDN?Zqb7*rDCD(sEB^#uaD6@{^a$D7?y=^}?HBW)1 zlsJO^oE~~EM2+8T{o{g1b#V(G4kHfwl*;`*O@8s;R)5WxB;95odH(kVaJw^!^pi(S zbmbtY5Ew{qtQ|uW&!q_r3Z@aqB;MEJs|ptvPM`{d?ZPQLp9s7x2FdfxV7gU3hM1q{ zneC@)$%Qv zl#@p0!#S{9+~fr-xrC;1BqLxX$%^1;;HF4Ctnq?+<7ZBJ@>o*y^>CG2nik!`N#G5a zvG#J0REZ$LkhEB6lPg9%3;N6onmS4oaw>L%$5kD=+SZ@OR+(~l1_Y#5{~b3gAqIAN zxY1EDz-6qoya-rZAD_ zT`puR&&ZQiHWZF_yyOJQF0{wJhy0qZN5@HDrQu9+~_CMshEil_1I!c z&l}$-A0};}M;y~=qQxG5E|w3*I)`AJ&NU2Ch{feDvDoROL|@-~L&W~slX+95X}dn} z!x^;@YRfWdbbJGSTa!TJE0hZ+~5+n`iOxHXzJZ^|6Y7MZ;gl9x$LliAIYKQh0?!$b3Cam74OQFY~ zo}cGHuS$E;tS8B|QYD1kh{@*EcRi%FVOMDF?!_={bTdu(Fq!;1ABMNBEn!8UIE_)! zqD^<5(2(cJuW#E)UaoJTi68A~pt>s^j8LI2n+KthKfmw(c+H8`ET;84hv>@=f9_Y+ zO6u1=-M(qM0@s{ZO6G{#iC$+gUHw{(ww2TLfI`eiTSrKka8~fvF-HkmYVX`?1J!($ra{&%t z)J@hc&8J5Fk@WccA0%PS2-=w5N46}xOtlveG{8}o8mpa!h4+hTxXB@+ zmv@3#@c$oX24PfrYYRdPIQ0@!fniP=)LQX0Ibp6rL-X4~ z)tv7)zurX_>L$_HtX5*BnZWz(4pNoD5NaOUL2jF0=5CJ?(&PHmIO9jNF<8x+*d5|K zU|a2I-3oa++4~W%V$`F93Ylcn9R&>YnM8ZS?~}t8jl?g)kE!+g(2r?BG|e@Dz74-f z`@VWm8aIj>ue(IPZ}%l@#A2yl?i9KtaT{Iuw}+#^aK z6{^^4a-W#hw};eEjFFFG>&dpWp45r&cm=D^px0`h>CUlZ>7hO~{CUubp8a)|?r4z2 z9=$PWwW}Eq9(Mw#Ip4UaRuAcunf3T6vl$omSEI6BDHM8C;x~tMTwwhVR2(Lt_$>Z< zxxA9N2BlVg^a#K$`{T%zJ#N?`6^b(oZ?P-)I!J;_bp1+AXE?tHSQT7kx7lAGE6(mep4lAv1pf?-7y0wd!`L03pvdP04u;)E%yj6EfX^; zR7DfyF5us90@U}p2MMLCAv*LGwB0)o*RCpp#?<)`oP8harb^-(?M_&FupU3ByF=s6 zIaK_(Gwvx4Mper+4B|Q67kVe7p4UY(XHyY;Ib;OWo$YD5=Mu~)EJAJBD+0&n*Vy94 z_d8B6#O^16Lp|p(wPPL}n!gmUSuG&1lvL1vr8D?Wp22mRs)+7rJpmh&k)qknkr3Lx zAEiw_@vd|(7FU+xUhRpZ%p5HYJ8g*fv}ynxk3gIU@L9P{};&!d}!p?SKv{gXUOMf+l;?s*)a>xWBfR7Hwk!bnf^Xh5e0 z%uh!TYcxjU%-YACL3tS%2cO}8@5}Jpxbvmr{Pk1lsok`1+MSaU~TVvnRF18EOp_C_!Yq>^KQ;3O`A3MK{y)B5I>QldrSni>-G9e$9kYGC_GHmAr8t1HD^|7p!e6i^QiT;g5 zDmQ?*XTD}BV>3AZNdPK`)i62jgP`rhOSXICBy#@!1n|G|hqQnH%H3AK!(I42g*@{g z3;u~dFeUm3OQ_2ho}2fP-=~}6V8wMVf4u>j-mQ$`{TIlUE;+c|I6=^N@V8z378P#d zq%IOTB8=J$S_#T5I)rl`S&(D)MmSSgz~m2(XYcv3m+d2GSn+c!Ng5qR-lR&SbgvFZ zjhoM2X~ z*Tj-?U2D!|lO63IFsAl3iWu9k$|CCeh?vGWc59JHn7VctR17LYr0GU-I@kktDvgB8 zC)Jq5M{`)Yq7}AmO$Pt)Ch~EkEY+I-gRFi19t1VTOvx;k74dWAcMI|bx6Ti+V0T?2 zwnRrbLVPXxaJmUPi^WOM-Vv0&zQ7Kc)skpATev>{F(Eb+xtm@IWcX|=>L|Am3YNcQ z=1a67YVS6#BFP#GayJln-D#w2XCCCOT_-3skzlG5vf0A3>$sMYQlKM_@z6x(k{1bje`AC`8E46&j|Xt>TNPR%X)Cyy z7eHqpctTe6FM<2~-fnTe1cW(^#N-g3kLsbuJ-@wy+dbBmi`?xiJVFl==eL)*G>~)b-bzMj!l*$em4Rkds_ohlRQ8|QpL&XQGcrPOaV@4xUq-MrKIwQDan-`!R<5D zXWx&c3j$*N;qZk0OgBQCb_EWSO25xc>y--Wvr}e(aF}@Q5Gm@{5IXBQll-CUoB%M5u;z-V#c?Gv zX5^KkFIm261Uig8NgkH*jF_Wq!2Z-mvR+OM0!@OM3cw&;#lO8aopzx ztI6bGX)?o`pFt>@;qx1lII%h^Tz@f6AS*uvt>Ffw&@G4v$E%Y|RaUfP!w5iG1xDV) zk+m{);HOha=(hw20cEnGSDt#5cSDcRs1^cS)CZIu1 z&~G+!_PJed$XX|J-^#qmujf&m;+|zp>#-~rUzDc-GbiICyHzMXV+6iQvZBW>`_ms% zp8R`Q1GX#WaaYYIqV+pJe6er^W);jO=Jv_hW*9)luHD5geY^1Cgco3`A5H0nSX!%o z8LGY}GXvh8R;?8%+A9&q?{JK9mrMbgEbcJG;1T8rwe4^qdYUF6SlhRQ8* zI*FFeygz=@9pd&~6>e0j;^MXnT)^*9 z#B0K7mE~+sXsQD02NQ7ZnqD}PegHiawvnJ2LUjG%i1Bkr!JMh4SoV1Vt*qf4qC2)h z;eL@|vW}*`C6%^+Dwjg9%~j&Mq=QA1A1vb5{Bp+#B`HzGJ9CiGj^$2?CSg-#54pCI zP)Bn{JfAJb;N2AvwSEzfPhBo_KAI+|GvgU9W7bly_g0jh)TUR2qv?W-d$4)bY})vA z1QpxphXU;;;UnI6&?NhYHr2$_%%n7&RzrumrIp$k~m-hOg;;>oR&=Nwg_r=53?AYDNK(B5yJ@8M4E*R28 zZT{{xf2R?>%8Q9#`^BTl-EK~@Wez%#Yu zWhGe{dHp6{89Dyx@rHy)^9dUXq{y;ROV_dg*$F1|ve_3yZ^Wp0?8k%c(w8hG4WPH$D6qtE0I zlJlKKkk1q`&L@*jxf{o+W>%14M$@ps;0d+-;)joaxzUks6R|-%fJppEAk~Rhw9>~O zox8@1zTNnM5_7)eo;4HgRKYaC^m~!wbl`xEANR~sTa>|a0cbUWHwVc zmtGpPm2t1ixVvv+DEjQRd;YqL*eHDm$4lewI_}Np#NIw2AASeIy0IrXR28yy!>_@d zEiY`9caP)yb>6ICxICA-$d|k^$mXVd594P?QM9CEJ19sI_&#e6-E~lluKevyvR~A* zg2sMM`K5#Kd}=xOR&x?GI=T^q0fBbSdnOSMfp`w|7w9H}a6CRR=vzpYA^Cgp1MP!<~4Q;l0 z$f;Lf7p~fHn*6>Q!-m@Y$-dK;tjW;{+O!W5>!$7G+J7S<%Gw4(mHtrQ!S8HV#(S32 zy$wo@(pX8V8TsIBLPtD(Bs7c=Lk(FGvu^k0`okIb`uPb~gN0R52fQF$Pm}C!b*3Or zASJAiR)|Z|3wqmV#Iigtc5$z8z~hTxll>vu_31r(ZJG}%%HK%FvOqc_o|1jo#Ql)a z;HF7#=eArw%N-X;a_PB|guRZ54=Fj*pd*@n19 zmJyS#1mWPz-)w<<2K-Q%2QyRkVB-~88aiB(EbO-;`44x~TRW|&d(%sjG4C@qt36ND zc`ooRoqoDr{4i_(w~h3?ngO>RR2etpxFF}|kE$Cc8%a&VCqn$5k%$3z!X2c-ym$8W ztIZf1?hy`6-fsl&AM1gRdo($FN}t@?vHv*&^V5E^JI^I2gp=ipWD2Eo1@1l0Bt-H$3C%MVRy3Qzu$5ytH>WsZ(yt0ROh8;^$HUXgxun?G1;qQ4=y9J;yIop0g)8-KX`NdXysmi4 zz+IL!J@$fQ$M^6|mPEMpN>gCevVGVt_)B19A}NahuPH~F64{E9q!rgE(UXe zEU~trn`}GC^(*f755#Vg+bO+dnNG0i+{`948!m&T=iz!= zGrIAr4rY|s(pV>3QD*x)^h|EVsJ#>Dk_%;Yja)X)Ynv}J|IY>S)kx7@XLStE{D*1{ z;^?4kB045*jbhK6antH_{HM5>#$Ry2>UtLJKJN_@V1e(pA zL~S-nxa{_DQI3qasPND?^j+G8{eQ%8d#405xzz!RKU49Lc_H?c8jGR=reKlJ8O)jP zPW>;*qMVP2Za2}ysA$HY_6RjEDj{%XR zMZ%&7BsE+N4P~kzA}WZ^HmM`oTi4UeQ}(05e0Pl2Jw~n{E5Hn0ecZM*gl*^fedeCk znDOg1PR?*ZlTj0J)W#`TZ~YJ->t4YRISu%7$}!Qmp%QfctRng*aAxkuFA~2Wy4c}R zj5bwsal3IZ9(9xx>GD1GQF@QiuVtoa*2z}9uXqZz-0Sf1Bqz}s-6t5ccZ4V;$q(CJ zA4An2PNK~B&Z5Zw%tiN)oJE6>Y<%i}A74m?l1-B5u=8afj4IfV4>uT!1iHs?ubnBL zKg!VU+)13am;d)6e`gLXz>0clk)nM$p0>P&=O5n#+>3DbSP_=bkP)p)7yudmcTsw= zzes3fC|dU{3!9CbVQgq4_w2nTr<|%y;}7pAi@ihXhE+e&Cgnapo05l)(t+GF4;Rp0 zyQ``&vx45Y`U0Nnuc-L=I8nWilxQNKwQu-$5Y0nd@#R@Vka(Yo0ZDn7%I9vyhPIIA zT;O}GxiD2~Djty;E4pm63SaU$o~0g1`0HvprqoJ`LcF)*r|m@|L!%EBXwp~7ver96 zNLUfsnItQ?Dr<)~lA`ddpFg%Nx`(>&ZBfFs5zg5Bgpwgj$JLk8N3&Bw`0Nt2+D(Aw zYgYt!|7(Y}cWq!>>M~fQ+Xh;64;;PWjmC{PK-Fw9wj6j0AC8TIwMSpEsD&@!Wf{*p z+rZdLO(FbyZ-$5doFs`EJ3xK@eR#|BbL@6a0o$9sU}7eZgO8L&DkTy`-eoa*mOO&< zw;%ESi%CRsmL8QmYl^$(Tm`d9ve;;E#H3xc$f<^PctrLC+1qjC!caO{S7H&)}c30`=Lv3V+GBkao{ScwIgP%2vy=4VGKMIob%6?k>lr zdzH{7YXD}nog^;<{fK7ZSy*I00>$=>!jCaWnMuP0)VyBK75g=_?w}qP>39|%UYmmF zN6Ctks3OiiZ-SS?-@}oWxAB*%EauH#j;1oH=v^a24;d_hvr-Q^$=m8U`Q>9gUVIw8 zeU5RBVXuH>UgTyEBmybhK)Nx4tX;Am%eRacg$+K$aeU@y!Gr@aZJ(X+@IrmmIQ1mZR4$=>(v|hPHSF(f6k*7lU|is^1p`zkf|t4t=3khK*XK)-(f7=S0=2Es7xNop zqBxx6Fd7%{oy~0??FMhZR|{J!ZgX^9CfHhM0J=F5#h`d@=$0`Jt$53=z1#y=7Y%WO zfNq#=l@GN-1@6YBAJF^y3Ou&=LR=dL!pRGXZw3tgGbixr#v%;#nTA(o zWHGUXfBfr>!2Xd^WNMrUUkK$z`*UJ&ApQ^dUOEh4T;Jf_;!HgEnC~r|;+-W8pNnrCrOMbHqbveCY;nY+bl$XQ^7 z8$K=JY`yQ|M4k;eZ|e)8*NbsDUUDQp=_rEkq)#v^W|S;Q2`} zfLL}z-A5aeb7F{}*L6W|#W<4ZJB;oADulL#*?7Ej0!*B)4{Li)u&&}^kT&rS?0fu< z>TGW2?;j_bptlYjkF+wUkhM%#F@a9_lTP##a|E^iUpaOs76uL7iB$1#$eFR0ge;B0 z&U(OGFGR3L>maM&d4lxm3_}@T9g>$-OZuNT5C`6Wx?P80=EN3u=0`PN<-cLKF52LV z@J*;DeT?7r@V$hm`%pNu93C$`!bbK;qg%N}mF}n>vg1KCOy`}j7N@+yX#I68+%1jM zmRy0qp4sp+;w9AW84ELJ?T52d<#=!G4`ybnj1_nJpVv2|9*uWM!>yHAn>`D*UtP_2 z_NrKPgB5a5l0f&=1#;g@jQIHVv$U{V+*Edn8E!oV-F?}xZ|DNg3TuG_!5sEHae(-m zc5YZtCRlSSlD^1>T=wR zZlb&1m4>H}!-QihG(FB2h7`Q$m&03NSV1;eNgSc;!~YSn!V^hP_Uzek{5kSEJN5TCi!*K}6=PO&E(bNC;rRfZUClq2-^ze} zk}c$~s)iBdC`hKyRegiT7Fi}OEFnJj2C!6X zCRTXdgUMdvXrmzsZ*9}ScfBz_;@t&ehqi*fj0#MM)WHPXc`z_}3HDu8=V$b#L|(KC zUOTB{{v!ic_9B(rlp0Kmr`{zE+vA{iGi-lvCGcv!$u8!J*abY|8aGL0^pfkoU9F2rM8^nv}U$cC3sIo^Z)YU z&%jE!WEV@~Cq{89)%?7qViw=$IRMY+l|YHBCvi}txUi*%TOw?LX9w;H6EEN5=em-@ z-Gdo$KgJC9d^nCfK4n1NK6emBjz&Yt6zqvfWvRFBlM@L#G;Hiq=Kii01=~)+(&u{U zMk_I8c zK1Kt==Lf@K*Mk&o?hT)h%Cp)%5%izo3KqS?6z5$V!4``SbNshn!my7-Ldqy-juG@| zpMVQrGr_@o9$i`B&e~qz73<}ez{^ol;Im~J+c2RsfoDr>$vL!Rue$=qg3o;hWgS}PyF!j&{s^2$U zD;h$2$uXeZG8=Xb4i=4_R>lW8-=>C-?R=U~1gVcuhDE~-nNm@r=$pS4J0^2VxKF6D z+Ml<1`5X-zc5WZs(^Q0gDZL^K$pJe0@F`~|3|}jS{M5F8LzvxTh|R{?Eu+xAy~Fe>>QDqI@uX zKk^-{YDWlpi34Dk)(=8oKZ3*+8RE$UjWI1|7X-$yh0BHeA;VJ{PwJFm?|^>1cU&8} zhW_S;?r_2>vj;(83#0Hchd9Mg3+VCRZfcAda*+X_VfK)H_*q*U3NJ;Wj^jkQTAU0b zTj98Htd|yr$D>@EP?LehS~|AqzA4>vxd4;5gmSWx_Tgqxy4{J^bmPgzE&Bjo*@g!o9K@1Z6Y=XtVTr#J8Y1*h1Vu2n4O^x(&p2_eg18J=+&#- zB!w)HlB?og|58MwrE>W0OfK{+Ek`=lU&xQ2;}(RRrP#VIPSv{;{=N;tO*9#P%^Zo# z6Z=tdaugb8TA;U516OfD7AAU)fSrr3!@eblxqE7M;PN389{Zhz?)w@d>lsHY{s=py zVIq5QaZ7*(ha6bqxkK<0xWKPt3A7_f2A4_Bi*~+@#MQE=;qFakTt>^l^HVfly>|nO zT(!~fgeMIfEbnY&I}p2T{=s^0MHX=Vgm{rnV` zcX{DJx!b~-@eFkQGzUkyQatypk!y?)STBA-B8l;1am(5>F!^jF{E_p8))IO2oTq@F zVS(Ci*dW#Ae!zG1=S;iP^Kvl?zSZ2==>>oF6|&19O{o4)Fhr* zVusC@4@A}dG7^Dk=?J)XEQ_NHo=};g1Ues9@jv30!J5Gi(A)G8sxvPO z%<~5L<7oyDbLudy@eU5?RKU%}Do}UgCRoL6M#m+ZSZ=6_U0bB^e&9abap@N9D_n?H z&br(sCo61BFr|t9+1PSU@O-6BhrMmz!MHg969c1Qk>)5IKS%?6??iL!gR0QyfI8a8 z4ums>T4)}r44+o~<2`%=alwx==yzZkB+KTV-)=|;NK z?*jMxgAd%!v4rObBcML*9Grjf7wSB%*tivs!Oy`2-VTlg^S{gao@Bun@Z1j%8pXlI zGf$oF75?IOcczJ#kM9Ic{R-HmuZyA(Z`#t7N!PBX!@WWG!R_og4Es_G173~=uLt?i zx-S++>^0ye#hdu@R|8;q|4fV(PsIamp0svUHVoYVMa=4^QHk*@K5wS3|Zr&e+x5}Pz8w6KbiONC#QF{{1%O3^`dScGeAYRnBOo_F3^rLYK0=uzw0_pyG z4P$(Z;NKTNEV_0T=ro5LO|l7agfH z5g4M~V5ev(PmwvLT<1{EF3-vJnc&6k8p4KHEEJzoTL}JNH8Db~0Iu4u!rYG> z**&b`*Dt%szg0QF=lk!0D=#wykI+p~;EE33bJ{${&)&yQ@3W(hNiJMzDi2ndw!`<0 ziuh}RB{a+FgG=rxK0|n47iM~b`bIa9{d6DX@|EC-%^#=8Tn zC1(e7&w7GW>pZk!38ZpO1^4(?(#(+9im5#U2RKQI-&&mvPmY~{htZ`F{9O;Wd_6)7 z1{~x^>LoznRZXt^Yl6T8-2vJA6;Na6R>+t8%1ww|MJHy|iYlG2L7%rRK9E(%1(t4b zuC5N2E9o-7kr9x#LW<&VHq*`d!gpJ3As!nSg^#Wr<4!DU;qIhAfK5$b#hcTmKyU0^ zFe_RQf31EA+0lFK_|k#!ERMMOXV%mB^lO~*ogB!zTh6;wpMzyuM%Z#ZU3@n83zwy_ zh|1cdxRUj)qC+tUV2`f@thXxXy!CrXqsfEqzx$d(9t#YgfI)Z-Z-G>}6or;3aPr%f zLG_>+$;%4--u+RK?p6oeZwg)i^m1;`5Me3Pc%5@^-v%F4-NYxyj6kb*OW}E&D!(DX zojtOQ;Ed}0`3bX@i)O^=LgTL}xDb#A7w5j_?g>zjQ8rV#=_R-MTd6=B<&M*_z;PIV zYdU}IlC0BIukYN)h(_*nXBaoV>NNNM&1vv|Rs&Kip7WFCpYyw+$%#{00pFj7;@3T? zoXOf8?%S?3&fV=S#9vHk;mi9^`U&%?0?zDxG`1LTFY_yjArhgZ=W2?wMKJraXZ#Y+tp=6-` zQj{RvVLLy5p#W55_s{vzFPnG#tO=TY_J#r6TfY>L^Ii%4U61hl#aUJXdLKQg`sqN+R*!)CkFqSQ*@pfp1kqU& z9jD(9_woM5=Segzk6*l}KW%wGn*Z23r2L0|o9NZiNG`$m2`Qbkp;I#E^mL6oRqwve zAJHn}d*$ZCwAu4u)}6!PJUSZse_72}1=ezd+bf|+^^x^hq7KjR(rJ0FZFKAXXKrJ#;K7L&LBcJfQZeA0;zq$p2=N#ik{pjGo z?6rbd^3OP<nk?_(FMUz*zP|3Fl@^)NKANtnQz1_Xy11uJXTzgFR zO5=GukY?6e8sPoa4Ky}Yiss&bD*j_u#bqrl=k)z_d5h-3R9tq9*4N4syqYO~rZf`_ zoMOQ4-9#vVFHL@pYiZ)X2v|Jm9Na8PAm7NTzS1NVr=z!jiy>@XnbA|7rXDo>jzJx zbHy(F@+%V`#`Ms{yaN!h^e(;)zRV`df8pQnPRF)CalotEh-Qb1VSvEfOm@8^>fWD? zW;Oks17d3Mz}!P<;GByND=uNu!r>?tzaH)`J%F`O%rMF10al&739q{^qS}L0ere4L zwx^NsqyJ|3Bl`~4h0egrsA1T6XE+xwEx{d!reF@eqxM}wN4YEkha68tr+#uMS}+lQ zr#{B#_G{48eIIQ1$!2f^*$4Jsj`I;k!v$6c#fZXDhXE+|wNVu~gRi-6bKvu`icr znr4{UsDxb(lklp}3#@kdOmBAW!PlRB@YPynQgZ6&Tr zQ3jOudxjN}p;%BUBsD*Ohlfv3;r1)xT)%mn@w!^7IIH(GcQVrzM(BJ-tFk*_!Y5#C zf*XIbOBPT5Nx~ry4$-MHDfsqV6Z_r}TrrG8`)l18qg)NHgMYL1!~L9#kEmhB=6dQ= zcOl;NCb{H0V!{g#XA7%q=)LA5{+il@NkJRYaAg=yi#kUsSvnYcwLhL2Dnf3N2OT_r z3k(%E!PvdMe5vhxEcm8|St>Pn_{A`^{~+*WVqHaCnKHPovF2SQLU*NBjO8U0a9Y?3 zG?UH3BkG6wD{G`Mh;7HzRU+rK$|UH0lmvs$UZhS}BhGV@CrEAHfC-(Iuq8 zZ8UeDA5JY)8CWvR9bM=J?3*_c3mP`zOV4@G zeQyQ1s*Bj76Su(CqW~gr*zkJ?x-!_JiFF=_v9oX!=K0HFl&vu<@#+HqDWgDs^)_+! zwQ714whh15rK7{A?Kn|ClFT~K<6E%~zDhVtHqI(|?B97vIxvr3t?k71vtRJ)wl1vs z<4NqQ*v5hu_FxtNgkO1m35I=L2}ziMRrOIg@J)m0y`=;`^|KK-wE4oGs%o75&H|I3 zxqL!*GWCDfY?t9E|mMrxjy{u|MbC*usW|thZR2IZfi3 z@9j{S<(kZor&(-KQ5t(NJenN~U&XfOq_9Jg7wJ)>Hs3zJoNjkL5&Xnie1x?;s~i$S zdi$rb-Fmtb=>>vgXUagyy~m%~ujwAFws;9M(U4lH^JzvCA@=5{uyB zY*OAsD*n(-H#bSxwv-(9XVO`wt$znD{aAWe<y z%`0B%o-ups{(%;~&tU(2%)nmYng!?&m+TmnL6(ZU*eV(zsat3+(b#P$(Frt=n55;h z4JW2En>}mT**RTgm%V^Z6MXTOiY4@~YX*rQ30%HeLz&)$pS0eqjrsl2V4D+HvERLA zlxll{j?|tf(|5P1Xo@^M6=~9r|31;Sjb)_Smr0*z9%oBi;+TT!eI{iraKZeRGN(mW zY`vVcrck@d9eOWZ_t?1YRQr~S&2&U`vXJ;&>4d$2rPJfTB;h8xG$Kd2+q$P_Na z@)zOfTt51ZkYk=clF2U^%C_!uWqw&Y{NkBae9K#1mU(L<`=FQ1e%w!F<95iP%G@G> zKR=wkxOx~&XQ$E0g7Iv}h)@@tEz+n0^=K8lJHf8z~roaC+wZ3~GIv+1&XBGa^EWy)R zRfw>9`4ggJ#**rIeI~!Xza-)17)jIySIK}oe;9xE1ADq$RkEmIhGg^MR5r8YC#N10 z%LbHl6xY)%YoHQq*jPetGqHnaSzwiLleHlFJu5`R^#)>L&ZLw~Emd#g6o8fU4Op7anElltcCFEl4NFyJcJmt8^2(pIe3}xb7CxqD5z&26-Y8@690y;H=g zG(Kj>8hI9UUPmH{lcB_dJaU{hl2-dpW`;KgLq`62`aWHab2uNs4SnXw=6hFRUbh1K z`ca1Jtz2+f-yJjx{*JvBtKeE>KJ%OYkDI!rTi~j;u(wMhm~8TAnrah7kGMAe$3rbn zvuqk$T{?{g2I_$6j|lGag2nWF{C{kS`2zBuE5)9iPGeI23t76(Tl%JYls1C`TYSeC z0M3v}Xsg)r?>x@hV+{*&Tg0xO`9^f+BOA76FZ*J)n)CmV&;79Ed8;v1cz>M|+5R(S zJKEg^_G1ZY3GZ6pAXT<{|5nbSEP>XiS+d<7DQu614GQaWfy0wcwyS*Du}K1dv(gQs z^1S(hV!@U5Un-~%DG^!w+#}7CzxbL7hRk6>CD~|Kumh1nY+J%sRXHnpkAREm|7LssxQ zSfkW~?{n9Y=%uA(gRwnYt2|}8RW3~B;09sVI+o4(d4ylow}UL=nux|NXLUiAG=6WO z&=K8_bL_*wRXHE6`66oN{Kb>&#xUD^fuwzjVEsu&W@AU}r1fjQZXHi8rt{cH(@=JH zMYh;+v8%v3GQorCF|?C^PCnrl9iS+Ne~-P}*| zV_L>Eco)I3>^NS){fZ@3x!0EOUj6QV>3f9jTp1m-5)gDLIg0C$k@f`ZK zEx>|>Yp{CpRY=u|W?W?e7cF{6NB^9mM{U1JbmuIbSIuTOgJ!{!wNGi#^au1w( zyoQt}E3ma?o9OS~qwI-RKe*}U1!2FnS;*)pzUGahiS2k`qQdFQdBe0D|24f$bOILIo3y;U2;t}8mR!>0$K?79UkcZv!wpLL4;6Q84^ zfg@k<6^@6OE z{g`1;1=u_}SMkX4HMIO*&vI@E_wUTO7_ITH$; zvzfy!fBa9FFK<0}oX)&lK*8aL5|t}snAQ`XU7Nk0jXR$S-K`ex<=CS z6+ih9m3m=D`xFk{UPL#KeHFN|GnnRcWvW*_f>ZBaVBZG_Zq9=ROmk6=)036zP&jWp z3tZ})qBW&Z!(GRUF73OrBg~hV0M)Wi~X*Hn{6C%*R2xn;EX`gBcDPR`oxVT zHVgub&nhsbR2^0>TLq;P6RG1&Dy6I^tB_2rZRQ*S8fAj3BEgnlT(=dxN7d4 z)kWw{{09{Q+RW^AGewqNq`buO%r|-!K6(6&o&~)Wg~+S30AEw=IITlhmHPPevwx7! zfEIDwO&81*+_0^(^|a8mi+n$Np`P_DT6a&Ex|9|&=i7$dh~KWTA=Vu>>;8pa8CATc zHGq?|938mZB8?OMu{6?O^VuXC2GhO&Nx zmQgCGA|LX)V$9S;h!qH{q1*=i}2OPo;6eG$jf`;d)4ze$?Zx$(P1A8N*xNH-km0s#>42jEt=FER0Man zDohi#Q`YO-v}g8kw76`z^K8ae_ydJa^eO zNn47G*@PJbV9MP0RN3$t|1Q6aaS zoJL85>kTs!$E)BHcqGpL5eWxFesJfzwqTd#CD=LWI^AmbB;6Su@X})d?68vS0{t*+V8(B(_7Xv=S)5x{Ukhayo7g%gdV8x0%WSMvpbQ|+&>Jod1n2`cwhs?v; zOYSUR!vY29cws{ki2BF#a|#!VBi7Co|LIhMB{NpS%~J*NqL*4%}nhpX7a&JmE>`I*$Mx;Te7dbIN)E=htt; z=@T|lS|$&<20X?_zCxK9F|ax79Jh2i!}6)hcxZPuY%zQf>%AY~k+dwQqpBE_GCrY3G+!pT&8*gfstffzw^qgOeX$6qRa(f_=Y2^!{=OK5mnM zZ=()cG>-!II zmYVtlLJibVF?Bc=S1yLP*N?)qPyO&kPber&$p_^#qi|k$2rip@9mm#o;ep<9INi_^ zS3WX@mlxddO5;Hs(v`|pDh1;tXK82a2d^Nsp&j{zK^W8Rfo7*GQ1Q_d^!vA(eQ~`9 zKlIK*)}{xu)dQm)`T15fO%cg34)Zee+I0*3l^;r>ao&ZSX^n>4PHzPr$Cx|_s4 z^zFw;(_|fBDuMTvG8gZDp{F8u--oG zo-_|1=WfR1V!_4QwHG#5W#Dy>tEgFhj_1Oo(V$`jmdww>@M8<{@WBQ=;&d4u1KX)I zEuEe@dg0J)Sy;X!2s^xvKuEuR*t~cl#*Qk+mR-y7t3o24-BpTx8aXJQxw!m@)g^pq zFUNhpl!lQTj^GXZ5L_<34vX(q;G07>cu?a41YX}MIIIEv*2QAt@MwIJVv2JgGd$g* zh--Jw#eO#@JIi_$pwjOEiaK!$kGsWV^5#-pJ9{z~^vy@B5?g$FWeGkO?i61`jzDA5 z5pXV)$E64gJ5oz= z>J0@#rk_i=+SkIlX|LyC%-A+GF-qH23^DSjDh(5 zQXy1!9)h96)$z@uSR9t4^;+HLUS@@Yqvprf~AmGZ=(U%2C}36c04p5<>C%2 zvWG?@yqkF)4i6C1{Jw`UV(oj9I@J$0e_lf>4H8&>XFMM^*p&~tIzy;by0FF4GjY)v zO&Ejc#j86n!Y0jLamiPqdtez2-H+_q{jYZ*{YIA9`oE<(r>2wU92Ph_O-dAU!-D1S ztOpa{SW1d&hN_O)_%u}uJ&b0<=E|pFxwsHTt5?GIa$i=cosPLKBdOZG0G6G51?6rQ z*tKK`zPeu|9`XDj=Pf%H$1aJ4_V03dpr4aSY3f9{eC;{B%zp^eTh$@b!xW#^l;Q8; zvtZZF5m4k{25VyG;+$DI+yM_o*j-ahjg?*0z`nqL?Ri|eQYG0xeG6rZ=R`e+&2YcZ zT0kdFND5huZ^8p`%H#PE+pI^Vv7HO}6^NnIA0cN?JHNr)5zjWIz?i9uXgW0!xuQ!n zKY1jasZ)fxqt;Te+j&y-S;huF*Q7@2Sjco5M7y`e(pj_l(A*jc4>b;?F9gVdla!cH(j9mD?xBJy!=8cF zhYCnoP(isz=U}*!3ii!B1%uxIk<0-nCO%PTAjKs7a81mPWf?PEJ;7bNe$>nZ^sCG#hP2R z>rDXYE&4(eDz!+jaupbj2nT2~LGPMh{4XsI2dFsU1*;5B+E^F2D-HvuYJluVAE+NW zM!L%!@Tfy2r@o>J0vdGj=R{wS`_>0F$wTo()EY8Z8h{7?EvDLWa)_KEsT|(VmxUam z(o?~l?_=a$(+eQM%9V6RlyYwh5Aq#F)nbkFV_~YM87+{Sghr9pxM<{V?%S$`@N-Ne zbfjMgw;i77xYG}I8RT$oQaj;0$M{!a=eU%=)6w;+J3Kwo0=pKJIZb;z2O7-((obg% zbPrg~O`Yz6lVgWbOu+%V_umsvPO^xPJ8_ljE*}HSL_gdBRl{aZn@3%L z`Z1f`A^Z)sQEa8_EavvFk&>^7xzF3n;E7n;Y0<`2@Vw>}Z7n{;mVZ=YZ?a98XOa`G z80E^&eAi-O%Vu$bF{k+VG5tWPVI{lW(8uVK5>p;EgzZ*1!<$^lAg8?=tj;H!dP>Up z3ga?PSAQg1>%EP=Dotj_c?xXqpq;E*eJOX}Ri&w(C&jfTnEhOF?FIqmR2E5-|DT#Wk=mNu^&X8#^VvFpug zr<5;yZ{8uigM0XRuV*x7ZlFk}Unt#C)n`A?Hqf_~vQ)EMiamWef^o}CXu8aGK6qyB zrH&DOqHFGRn6FtfySZ@#GiuRhSEk-1P;-K;>K~$o&otTA$5p%#MrQ~Ai(%#yhA}nr zW{Z=a^Mw;F`1;sz(X|uV6#B!9t$zN4jys-b9bH{iAh3cq`USDAi&MEhGWP7)AuU?^ zE0wk@88FWSwKQs3EcbkM6Aiy{2+r493(gx47TAi+<%$${cep(5#|d=)LpxpS-b;;h zZji3MI#b`clZ{Wy z0O5t9^jY46wzsaKxdD%P<==C}XXKUHyST};#b5XveSXicn*WhJoUVhx-cZV)*~2Xs zvPn62GAN@vkT3pZLpQ#c(SOTlGtV(w7~g%Lf9+c@PT7{vuTXl(U9(?9``>D_Q(I@U z!)2-L)(<_&!p0YD*TzXK-aiQL=|+>MW(HrpA&^}$G^gVW;^{^FQTlvB;DSsRdZdHO z$m3KGIj(vmp6_Cb9a0T^rEm@|b@5^w9_cVG;ZEv4F9E(Z#4@k5I_$=`I5vJj8F_bk zvkzq&yly}SMJQU(m&r4!?QSXG>Q+YQ)m)kDQ!jS2axSU1PvkDuiK!tzlr6pD%jVkn zg3s7KavU29cG*o_sa!7&iPzwDN-dq1WreXk;WN`b&6la(KT8MR%z@;r8SLZEesq8D zaC)_aD6&E<@JP0Dn+x-u(yXr1R_|8+%8|`t?|f(0=rbNvkNeY(*~-jqVh)>pbsX!I z%x7!P=`c4Tw`5>x%Sr|s(c#+>TxIV_xG`uooLI=Hv!j{p`zx~HPhI#Ek4!i_>v0q; zW6O4&l4VYMc5Lr%M}FR(ee~nIK-+IWPcIHEfzMmVu%Z+CBG;_zTtfa1s*EUy$XVg6 zROm^17@TLpy<(a?+>-tMV#QtzTgygF-@;xE{m#n14r0~o6S!x$&hcJZ(y*^(0vYLk zg&EupN*bXCi&851oti_y)@2a$*yF(>ZI{u)>Bj8jsR>|Orb{oTZDcduO4;jTqY7n~ z2>*tBp;>zUnbHEDy4q98Fm@yEteL_7jatPO{S^s(ocEMstw)^A6jrOH#;nYKl3mao zc$fVL_npnbSNtvBEB*vjul>p<-`LFHtu1f%>Kctc6-rP0tF!gL^2ApiJkU??KQh~% z3)-C-yvEsSv{RA_r~8DSSdBNQsohFNmu2`|aTJz}*Z_IQcVUXFAD;4F3lYK@zNhFS zol5?|O{=hA+4CdWK$(l=@g;`ZlXaPO^A+%R_{nztXU2~HiU)t0Q2HIGaBx^uu0( z9?TA)_qOtq6n;9nEGePmNgKtwqsm}mZXweN`9~IE!rAUmBrJa42$k2j!WoM~uq?j7 zcTSdJJJjFO)*&*y_B|iY%jh#GeGC9A_h)?8FE#wQa34k9*O9cZ3T2nCe5KPnRp5hX ze>VL8&Yuz8)Y)*8vJQl?(4?_w`Edr_ZLH(IxaX3N)?7AaR5c$b-NZ{AQ&`QsbMR8J zh^gF0zPhWA^O~eV{KvKYamgj{_FBbj8os5p3}dEh%7NjesjT@-3B|Wj-{3`k~8BVN)og;^}qa4+2Xa1An#)pd9%H%v+exx0 zjNZoopzpDAWV^Qp*G>t>a9brjp<7RTLhn1t>zJ_6?gV~M(jWRkEi_50f|R#)@!<&) zfxWIo(K9Au(TCY=l<6}5?t>zFonJ`tDRa5!#y2^a$^=}yY7GR=aD<>yTS-16oKynF zQOMF7uKUMv8gz0hUvu&x)jE%*0=K703?kn3ow;_r<#;C9=XaxYXbfTd$I zS*K-D$n;^@rM?<}XdK6ksctO8yBX|OUL>y(hiS6VeGXXtU2ycq!@%jM>1|R8dv@Uf z7&;lT=hA;(m)^YK1=2I6V*}_qTDn!Z|SF{uHLuG=h!2Cx)ydHBdB7 zW(TygXls@?8+!CL@1bK%r4>f(R>yMoe7qhTz3>>UtaD>G4>i)d{zKW|IYZgTdTC~J zO-|Sy7SoZ*EuQbu`N@uNlVjK2PT@a@S`)l+tKq5GN4>Pxi?(E;WOU(C@A6r-6AHM8LriJ^a(Vj_P zDR=V;anGr3bf~5oe#{j3D^E{QX|)V%bF2~gse@qe@Ca6aWdWr*PvEbGUnkcoYuL}^ zMXW+!8Qo92Q%P|T>BZ!btdkt88finbT?R9g&~ZG!N1ln-TGNvklh_6O!%+JaXc|yX%l8DZ>UJ&X3RA3iDg(V)nM5gn` zalTsx$LI({`jThCJqz%q5T(XQeL?5fb8&ZpYr>&P^C7{tieh>Ul4 zax1*fkiY#{ih3t-FV-h@#H*oe=MB9(I()kouN)P-%+e-}D zk*xdV)#JstH0Y7%K}#+z_cc|`9L>5XWU$!%Qp|5uDL4G?6`EC_N6XKyXQL(;iI(c! zqBGuqC_=x0oLWDUdcrcg>NSNeYt+ZA?Ki+vrIRjg-47EN$FSw~o#bd~PAhk|@k58K zVw;c2u^IY{*^J50!819GDcv_bZgq zl|K(?S6dBt&^4Dfs<_kA8Lz30DRM0PJ>6e&hXvea6@)dqSj%)Vc33>8wQX(@-8R&9LW?W{^gzcG}?~& ztoum|s8m0o(3E_NuGeAxKiy(etTxjF#bYc$$nJ@frt!5eHTg+>LrG(^A}ewTf@483 z+-k#e$~>?N&V&~T(I9~{Vs(Y@yVgksG>mSGBd9ca6M>V6UcUIq;jKb0&RLahv#x?O zF{!l8*NkQcDl*TrT5S6=Q~pqID*7MlroOM4>`UCN4cFPe8Ie3YW9J_-Ir;etabz_;Ofg5}5X-o5(A}HaRESu|mks97z7ayQ~ zwBXlb`rcyC%I&tZto8B2mPIC z%>G4PCCSj)?9Z0*Flmu19rc;ZbvEWfS4TB;_ zwm{Jn?lk0*{0nXDKdS|rr+nl}^NQfB;3fI`+6f)D_UEeKn4?8h5#49~5x!tUVC-VJ3vi_vZ>+LdCIL&wWU6W7>I0*~WwJ=wxo!viu6;?K?R?b8U<% zwI!d17qBsWH$@MMqLK2usV7hY&&a%klfSdk;Oqp*h}1(-^Ik00^5u^2H~=~=mq1~J zCB?ouM#nbI2kR{lskKuqK0h=AJSH{5kn+*+y3!L*J(6Z8jJBc9i5uV(Tn_aa+5CeY zC0yaYLXzL)D;gHbamoi@@xPsJ!Hcc+aJ1+?es#8{byj1b)I5{YdJLhu55%dzg&ySe zZ`={BS)7s8B$#dVjkoan&8@i7!~GqaJSIIgf`al^LcqIuAQvu$CkOuFR%#A~O@!vjgL^H2EOujV$ z4CJ$UvD{r4JgXZdC2OEMKOXGeqrhJ+6Dt00$E?U}P~(4?+uJu7*G+qd?-zB!_B)#J zU!@-D_6&o~gWXy1pQU8`b1eS*@RJMvuN|7FMnLJ>D%jB8#SNc^aA~&)3IcDGchCGL zN*`$t!El}TQ5p+sfhex)RD;gy*A(V(1i}XkJ(a>}P@SoQ`#-r6y&ERHbKIfE@*ZVg z>*Y?TFC^7%1E9u97QM%0V4kcad;YIR;F8IskIWL7@+$)j?4qFW??Bv^zK?tRD+`Rq z$b;^EJNR0*M(p{}8N3FBp@Xa}rnZ#9JI^1YSpUiJ!9tHt4^sriGsV#8Xu|bQ*@iNg z(_x?c9~jb}4}1O_%958XB^#j+y-vRZ2N-JN{is=VuFMqhYZ)h(`HAZuc7yJxq*N?@ z|A1SamV}Nk9V>k9PR9*;``OjDd(Pto;QYX>Q=Men7A-*dKqiM}YO-2~Zt6f-9Rii@uW( z!NC5ra7#>{yI;@@O}n$8y3C(DdT2VVZLo&9yONrye|QDDFM6U?cbg3+-F z81ke8mR)}YyFOgtXI2xr%<0>>Wsme&^z~L^^*aVvMxCWYx9&ms&}Znsu1PeQ8cv#2 z0B~e6=6w%=ntS53e_%B?L4Q%@p{#bnLhA;4VAbi$sVa#$Sm^=Je%N4O+DbI@GU2v) zeTH;R6ZYrU5kjVG@^i$yH0jG6unIZ>QZ>rN<4Y3+M2DiC{$5-=E`}JKe?ZnhScgGc zc^J~9#68$#2^p)4g^ux(xNE8}>G;o_d#|~GS@G|h)UGw;orM;g*Y%8!?3LlJHC_Y7 zI(PE&XdX>0Eap=EytsDWd0RM<5SPr?bhX-KZc+9~`q|+e6<57Yeiio9If@Cw3dxJm zS5!h%*1V+A$0AVo%||F*GYHrBJ)|enbh+uy=J4X(7yMKDOIXR(!q{dR)Vgm&TK8>- z69*#cE{|N;@urSyQ>t6; z4r^jZqW-QF@UZyQLBYoiTBbUahK_RX#@CoBb9pYGA50l0oal72rEVo z;pNSSa46IeF4nA|ca*oIjiD*buFwbd>X)=CWh6D*X;TYN^cZ$%@1GTr(0s+b3%hl{}siJ?2ovb;ra~L7mVff<-$V`eMJOR$OPi1wsmk}EkMrKeyVC?03GUCblI>Q z!b_tY1sn5{A#k2AWa$>u$UPAdeW4O*u>&5NB+gc!`Hxo4o&}e5u7dx`cv64oI%Zxm z;3gbuvl{GfC$Zaj@AR09+=J7l@WMjCc5jgoIp=KSIn+j6Y*{|4O?pn`JEL*)&-;)U z6;Gtao2dTP&%*!CtVfTOQktBzkJgURfiX9;gfWlHpj^is^gno@rHmOV>DdJur~W~ z&r(z~ZUAG4Fk0rh8XmW9hU3l#FqBv+v@+g=>7&#!Pd5M@#$Bc2`p@9(`1vqHTT}RS zY%ZKM8^xXLx5jTHcEj0dKhXJg1ZE$&LzXVmgt6k=P|Gh3zJHVgg^T*UZ%2mwnXv|a zOrkO8ygqpI4)7DFDj{UcXwIldo|CYBh31LfkSMDIQ#bGqB8lfT$kB;YQ>ljovIOT? zMH8dbws7I}$dN<8R#7UQh-dG=g;`G}8R&Jwx ztuzGBR(!0Sf7~Cnls(C$o26tZ`VLxKJf!YxD9wBv%_`rk3$8s%0PRh3sQldy`|Ri9 zNOm1gO&ZNz*U5*Y@gr!`<5SEosvN)kG2wi^q~qGyv0VGiDl}fm-;1X5T>)NzN|e;D2uNL4S} zv9HE|sr$>R6&D5L!6-!)(|_sXfG9&?vn~rW28Lnhq)5)9Jq11_nsDidZ<20{kJ#>e z5$%omIXaG`iZf==RnF76XlqkEtRsRKFTz>e*+Q&Sx`8Jn7I3K*(>cp5f9?a%=l1Jb z4%QbhfQx4np0z0#2K&4qWa(d`l01b|R4Jk(m;?D*e+m);?}1CEE3}490*Q6*T(D>W zDgsWx^XXEYV!IH=eLW9P=4pY2(*qiF=myze(oeo}nrUmR^I1`j1R;bnJSd~E25uio5(Cmt)|bIfY+Jh&D^qgD$x zlpyUa*pKixg9h==JN2X;*lQAPHL+q3AFegPseQ-s`SaE`wGiR7+9&~k;C8nV6&DZCI$@WtjCITf4voi9mB7aya#{jL+e{; zu{Q#T9)Be&am#i z^ayxHx?dV#!sV5iZFL(jtUrUhat$%LQI1`C6T@e;-MBx;v*87KhQB^>+!nVZT)aq} z-KaT2M@^{~N*x`A7aLv^=g@AU=dl?W?X?!&D%a53r8Q((d>p(C(kC&QZm|5t0e(&x zJpJtJmsnvjoLheEGJHwCMDAbZ`R(!65aN^%p=%DJgj)^Ow#edH4Ab$K^?3F?+mBrD z=tNKc-KqI8mu`+L#aiQ~AVnt#?p#-9VS=-?ckUu~dwK$&aWBOx*X1O#+Xb9*=R?2| zx5~Qv4&2v{ht&J#C-~~D$Z3yxC!9Vx6aKd^=7)-S<@hLkLUovVRW*KY4?;5Y5YA|} z#DtKybn*WE^uWPE$gTH8=`=O=;lW$t-8Pytl`cn1r>D>~yOHeeJA~Qk?}QKbDd9u& zeXMr(aBi!TCvNc`&Z2^b$R3GNFu0~faHwc0`QaZ$rs!WqrCXPX@ZU)6`?CXM9Bd%^ z`cWuf4d5qfCAfZC4+^$a(OXV=5NU4BPT9rd{`dBH>V-CL`gZ{QQv$HT%N(cNaToUP zRb?&Pj`RDCU=sEG5^OIZ+zgjun!MYLiSzDRxzB;nW|G$kQ3=t&>|<}Jz1K-_Upg0h(!1IA>Ek&2B4e)O&pehh zNlf(X*jkhfQNX)ximWanjjT58WE}!Ama{>d`9-$S?wDunuviLKyxdW_KE6sga6}H1 z%4C^PiQ?!Fmyo6Trgn{N{~W9(^WY_Zc}#6ejB{*-o;!{j?N>~f;ACndP8S3VNC|J2!sxLA;x)s0OXCV+OrS2+00 zj0Nfj3c8a@=)R++DF3N}#DCIfi`t}cMQ0N}`M6EEW#w2=6;oyJjmmj9SOlwYjuag7 zzkvQ*u7h5IfXld<0k`_E)3!ld&aGUQn|-na=jAt2yRp}BcSAXr-=2z%CLuz{ZzA@c z&m7A?PsEG4Ua;e~yCBK(7v>Ei?(8;z_x0z|b=eK{zd8aOizjg@-;!A9#w08{oC@QO zM}h9%9q?7Oj7%%k!AD&iAo=l15S+53&0AwwzpM?5{9T0?s#Rd~MT(B%LK;>*g^g)Y zB{7P;U+||d?0a9xr3Q_+df`#Rw4`g0D^`R0Qa|2-o(j9y9%X-?T&X1ACU9FuB*GrI z$NW37kd9BaW#6?b`-0jk7?9H~X=}*OeKC%Zt{??csMP;rJ0f@L&Ca zaK*cLYSS&pa{F@7&7}ml+H}&9P1Q&;fNnUg4#NHxtlx7Smiqmz+?m|~2aB^whesjp z>J>rV+ak;ewV+L__*ta5A~$#A47OrT7FCG5$-=C62=X+e1#?#>lG-F$W@mc~Z-mdq z@9}a(C*UPJ7$1%u&l;IYd;*oW%fpAfa$!=_RZ=`92X5~(1*>_F@kU|PrBH$sWPJaF@SF#){sEwnJnAP3Qphaq8WQXkkwcA!qHFDnDp*pxZB|}H4L(- zXym;)VXJKgO~KdkMy@XoUcAA74iX|q-qYIqCyJSfD~dcmSEHiSeW*04rQ6i4QR(Cv zFj+JQE)NVs@xmm0tG0$*)e^#+Sz7FS_Zj33sO*?(8Y=83IQpJBOy6LD_Iq9F(n*&E zQ&-L*I}VM4o=#Qlo-YSRe0Oaj3}=q!=~y#uI|d8ksVW)%DQ_db36KAH#{N8;_D52;hX7MG-_4Lg>6CGT_`Q7L^Aof2cp z_RL;`f-Ig6cq>*fk4$eVNzX{r$J z+?_#>x;+9Tw}&(+TZ}vvZWY+nwhH*1m(`H*58S0 zeh9rO^ji8AKN1`Kx?BT)^5=1ronygg^*-{*TR>G*fb802B9NHPg za1uN8meB1dh4$i&!m*)eP;mSzYCn=^FZgrJWzGLcn0_=1ua&_hU0-%xS_%^8EF~}E zEvb?^BXzrvSI*n?o)~n7VyL7H$=NI-RY_)e^_&P+&y>eC1#7Kr`|Jr@!83-WGoi5J z8IAwbNvu*T=q+s{Jas1=LgS_Ae_v{G)Ypf^BV#k9&PG_$A_gLU7kkYkl4No3S zU`cQs)~Q%Q+2JKr(gj8&aFx4o99y zaJGgCe8(_d(BVHC)bl+mFSov-&WT#k`DP3#Uq~W%;`_+|IFP<4PE(nEis)yTkDpuG z$eyN|Fi|}Tk|y54Qr-dmc)l5z|93a}-D^R*N7&FiD$i(O@>Xg-`iJmuP91rxdCbb< zQ85N)<^OJ`zw6uKq>yu@=dgibv|1J2x&J-Z zc8`VtpR06dYA2@s(u7w*$G97#W}@n58-eRjF9^|}N8`qB6J9Y^g67E;^mE~A_|+7E zrAv~bQSBUjH<5(K`}suT&?Rc7ISGE6NYUWUv#6o+G_*D}!|CfC;YNNjZLM)6CBef$ z_VH9u{l2)eeD4CmgY_eDUAq{0mRCY23NP{;`~yU0@k-9!a2EGt_y*KZsG&x+x5=GJ z=LBmNO~{14xlr@g$TA=|;2O3df zb}0!Y+H|b841Dul2X8BfQN=DRFfXYf)u~cY_&7sgFfN?hX^s>`n2)2b#)`1>`Tg=? zyQRU}>95dFVd>MGiTor(t0*t#_BGfsnM$Fc)CVM^Cll;pPxaw~N zX|+{^ffwFZ!ID~N_3-3wsfB^6jTGy&7Na@<>_1v5`X zVtndS3@4tCd|51+7Qvq(^@+}w2;mRI6e1j- zjp_qsko4XWzGe3j?~)wyv*kaUC#Qg|6{mzDSLdS4oTK#I=2Z|r!40%#9E2y!GKt=z zFc7Mh3Zqi52{+zYOGc|WSBf1?6c#m4z~r>&AI9lu|7TjCqD%`Qu9Bxl5CN_7aaAWQz zB6;W=&3P9>+qav;jg9H#X8jb%`R5HTk4HhUp%ar<8Yb{O5rdQO&1Q4U(&5&vVVsKe z7;Np*BdUde=u!JTlz1=`@BK1?8a{vghu^In9NGiBUQy_fI7d^qn9`VElOb}_H#icL z4wq&vNAG4eYIHCePF(h*_wS1M9rHXKn7sk_d<+(>DDHzWhi-Js`-d8tJ#?d43~qBC zPA9*qC5}vm>84d;iobzX{r2TdX<<7`yOv?&#Bh8#NtdN~>$1T)_1M~R8ZRYPpz#cG z&VFkQXcmmXy69!t=dH%dmRhpf?^~I$4;3c9p?82M@qOk21ZLv3o&Oh^TfD+R>j zJi-5Noi zJNj^FMlHE`;|y-)S+`Cxhl%v*NznT*82y&1G0*e6}f56=p0{=R~R17~M00`F8DsK#3`M_jwu4EP05ZKQWv!YaMDGIgFi`pW@EQ zo2X!z4|NMq;+Snp&@%NV8M52RA{~=4YSweC_DN!ynf!UaUzzL4=h)oE*XZyGei+0J z0{0`2h?Z1CM_ej#{x%D*@!8#0WqsNqABUxDpAq-NQJ^_}H7>2P2g|EIXq~bZ8&p=K zyz6N8@b^4CEBk|Pu8u>0!^uqPw-BMMhgmbVHJo&wo`KalU zh!3UUxn>F+{=S9!NDSk4%@4zsRi{b5h6^=Qlw`To(25r@LIuylUp3Q9xInuVDt0YF zlR#I*tBvH?y#&0qQkS1oY@^TC90si$&0x+y^S50KWrN)&oY#>La3?*POcKPx-|8#y z)GQbu@;<>O@1K(qFLHRNxDjbN(}QPzf5y|%4^VOnpAQvl!xJHE;6mdYd}QDxuc}_>2~;*|Qo4H(sVEFUX^bVFb2RdlR{?WC-~-9%+8sO2cmwigb<^sqSM-dl!y2P|;Dwj0zZM3DEjpV2nuC#8WWF{^Gev;37T zeEL8e^mD%83(F|d>D({4cy}_f+<5|a^Vzqt3)XWG`kd=}&cXaeaa_U0IdI+kB(rTj z%I;dVvl8oLX!y>B47q+3uKSZLjNLq*x%RIEJF!NbtT&0OD|A4C)Jk^Ka+ru~e#<_u zDZ;lA#jHbBhEwM=&({-&v-YG0TGMw7^{q^au%($?WNISUx7ln-Fh3)ZY(>ZJHEcn| zOwv?dBN(q)g4$A6nD%-&bh`f`cX{uEd3Fk$cA=5%`IEqU_B7%I8Bg|rEobGc{Fp@G zcYM=U4y?EUlh*IR^o1$ptn+5p>~F~23~bqcp4IYsuN*r!?hX7VJP)xwhqGHakDVXV z14A~2JsN(36Z#w?H4E1AEUyf3ch=`*%=u?L&vh90`2-snUT%dpj-~k*Jqsr>T4&mp<9M;=0 zQdIR(lTB}vWs|p$hISP`;~-!|6_CE_m@+YL#9=_VINHgZyJqH=&WRX#r?yI0tWU zmu8h)WZ5GBzclj0T2{*Ywt8k5GnMX}OtN$yJu$M3NqLp95221MGp!NZtihkr?P6}1Q1e_#Mo z4`;C=+ez%;tF>4zb)A3q*AvIRW8ov+hPtZlbo%5nxV3taXe8Wc0S@QzjhPFZr7)Il zH;iB%t9TZvybH}sv;|H75lnrXG&|XCA^1G40urnjuqDkS;7Gp#7@i-^?q+0SiqCj1 zXw`cVyA#iMAPI{d8On4644F+?JX>P+o9)j@!i~X-Tu>=6&v&lOcyTt_p7tIUPAIZ{ zv*Xy|MPbZ^{D41awArMVbQIPJX?XoOHu~={#_dsOwv)eMNVcGrNKC9 z_(l9)e~^3h_&v5JM6vxItvJm+>(3 zG}qw0UyX1={2>X7YvQhWm17=%lxTZG zr)S36A1q}z9=yhs4llNG@p+b2f17??_>+d|UO|w4f{QBuqL1SUs{XbfdAASy=4L{5 zUSy+J?LunetqFbp(pdB>6LWf=pi$^&az8~CcF$18#o{IWe&8T!sFz?qMzgS~b^+?H z`wRy^<=|jT9V+cf26x^=^GVkRCv`8T3-umh+0k6oTw{rHC)JsDT?fYX90cd>Vdz&@ zfUBcI;YKgtw4I%e^MYlB-Z$>UqXXlyC{@VDPxwvSZI{5#P59Y!svpcfRz(CRe?T)uPwcxA2pnT>wi+raluED)^`E#4j+kErVJ79 z(IU*yn@`5Db389C5%0@d!ngpQRlm)ZJ@#`(we=1-_oWi3G=y?%J* zI26b;#rkt0b>|K)=VS``yz?Y??M@wb<oJ0w(j^%A|~y6D!SAHvv1Ij&D!mg&!G z!1s6wFD`wJ_xjbbv0Q+$uRZa-$%+sALc_^w ze0l#1rib#}#YINop!tFNP4%Eh?*uZ{nO1DeFCDhmMH2PGC0Xf}HX6t_pxBxrp^3&k zzOURt|9ktH{?y@JbG4T2lawS=)ZWGCca7N(?HH!`s1XMyJ`iX~t-=|Vw{e1!EPH0% zNhgH5;pQEZu(j$17`I%9$W3`v}%yJYh9>J_}^F%_pfh@i4$1-g}J(=u^M@E`2O zq57j}e8z^#n1J~Q?3Z!Fxj zYQpe0?nM0IVV*0p78Q?+@b>0R+$UF!83#=vvr-qQ9*w4&k#$($A5ANp9)WygFL~N# zM;*8bD_7oOZ&B(Cm-u<4UaUx{Y3Y=N$$dDwHz0;qKDR#95Oq+>pxAf(I2x9K0~-QUhw-cV;mQU&Fpv>hM$7 z8(kD1zyY&8V0f>J+Z#We-5AOi%!3$qDBd4a2k)cBe-Su;=sljvtmnA<|3PKKJ3KXc z5jD7Bf|C=Sn5m&D=W{EDmaEwT1V^I!&L^;91H-7d&bUjBXMTRZg)b7!1hw0bv!z~j z*q#6&F{}iIohiWSTjI&LV?bN9OtAd;L9*>nJJz3iO6L}|k+=!Ruw#W3GqjZ9CP;7R zBuuwr$Mgv1#Lwlz=U8z9l^PfwHHEwHsYKSyxJfl!a?nS~1Z@Y7VS~#C=2ZU*^276R zo`WIV9pVI2-cRGUY?fn#rc*hu)6enk#u?nqy_vASX$r3*9>>}I8H+219YRMVNwza$ zFQJ2D;H6I=X*=L5h>Fk^Zd`nrY@D$cR>XRu?}}14@u2jJDS;JWL$s+<6b4B^{>k)Z|iUzUsBb54^dBS$mXb4qwSN1Q#@JuP_O zeGz_s3&nuo2yUa^An844i7I9uXx5&HK}&{lvR5uZicv4E>aoM$FH~{=reZaiQH1=_7?jJ#`?A z%^{Zu)mcFBKBjDWgAPBh%t?P!<#s;0#uaRv$*^S&t(eWbr+ohqB~uf2;Fc=W>eJzx zHB~r;uvFIBY{tz=j-`#X0IS!`V>jm=V?CiF{Lkq(%xOQuj_!zNuj(4v64QC?;BeFiXDpmO3RKr zax+y|FmK~otWJ#m|MRInP)y_*a|QZ8yhkI&Pgbug#7NHaf9r{HOD*0T7|YIFw_>k){=mcAvst@W1v`9}a3Fmg7N;*` z<2x?lj;vc~?p9DaXQu`fUtUaNy!5Qk^*x~~*HV~N%q>hl!gtrki*rwAZ=$UWBk01( zd>83}CyrXb6U9 zbHfd}BXvuJ=lZb z)zpOYxy&9|R`CTkioXYMtrOttA`98h0CR3Xq#_Eb^v49uOSoL2A zX7L$_$+7&qxxpAMgATFiM{(@pLl16!O&n~H)L@%FHM5%&YuJ|8Zm_lYJgc{Lh59-Z zYxklkE>}!}+c}qaLVlghtw~?NRFnRZQ3GdLK-no4QC^6Pr_aU9&Z9ZC32vNmz&+00 zDT7m6Hi26cv=hH=%4ZtSLfC}gH}Iq1Yn*+doE^8G&4PL&S@VfKT=iEK?&HY0EHx#G z=YJkzJwCjHpuQ-Evz zq&V4%a_*CSHms3or3YCKc3p5Ho~0{TeP}Klzi}9AFEQh&_=9Sm(D?^885}jA~F8j7|FwW>|-_YF%&EJSB_CEY^n8uNXGyrJ`K~ z@2}k5S-JC#C6}~lD%U$cjZ=Hh`yuo_p?xrl)i#B}CEpWVL+uaZ+L**m_LRkn3^s%OP>DuX`Eq{Im;(?)TRHY#IV#&1;7 zZJ`(Tl?beY{pg&h@9B-G66!Qdh1t1e(ek1p;$LG8(po%gRJ#HOovvVbZwwTzyhI#5 zi*fqT4C)wjh&0cY#KlQ`CbIuG^?!dJe_t{L!8r$#czX%?mytuhH>JSvV=rLkw>x0w zpg<$5-w1pE>l12RD5jX#Oq1WACJG13$r7tKG+=n1P`1?qdza|3`7KgRu=SkKTGABq zN{67McZaYtf2h)_WgJ^HLmsbPI6!G%ZslyLSd?@$p;td>lj+@-&_2TyXI5qkOyMS- z)bSHHbf*aAE6-GDt(QkAGG=eK9mD#@GEA6ej${5V#Py+9@PL6B1kB%DJ~wzS_V=#E zbxWmjzS&FRmL7f|20YVjbv3mswt?KHM^t6!6MF3hhp`pmILsi7UfJ=Tn0*;Vbf#Y- zTF-5uO=FZGIVcR45BiYT>CM7yZwa>i=sn@|Kq-`XrN%uA=z-0j(y5iDZIP2^zOy&VEy@d-&^kQe&`1&<@T)Yu&f9|3Irpi!%)rMUTw&UhGeWLlj z6QPpNc0PC;f{xFW@b*D3TE56vP}+2lNY#Xq*gSm@k9bR~1a5-u5!tkD%rZC?eVM$t zH-=t!x{vD%RujABO2Vtk<(Lt;jyf$nL%ovDQD@Bx6v?#_yY&*d+-3>t8yrTdP-}KO zB$f+OeT(CjE;8@qHAv0IfSz-JIyS1IrJsOC|7;N)sJMjQ zrF@s|vo3^YZ^iQ2QN-SSCQiCPM7FSYFlHxcO@5_7%&C);-zZs!O0LFh}vqCX#NtTDUUx996G$#!~H7 zaQEB^ocP8PW%{?!aD!Zo8hwv;X{+GQ_HI&kejds$P{mc!!#G2sHn)9O5FTD|2OsQ~ zAu|>^3CsPe@%HNkGBh?Eu1|VOxBoMS_Q_L7a9;&VSFgia=XCMU{O=_1F-M+_=%lrg zstgBQp!w8X)LPq&im~$KvA7X5o*V)V-nI8-m;|>Y;*r%TYaP61`c$wm?+W7Jdd%Lh zjd8`Y*y1M1wH1FOi*L`Oi+*U}toK41qj(V)xLVME&;4=B4;M5X{(^>$nTSQrFL7Dn zaEM6-Ecj*s--NYvX+S#J^NsKFj$Va5l{GloC|ub2p^6qbUm#Z`6Y*cHhDr8UD~$=t@DNpMBftVlQeO`s-NjN?uP^HILde9{^?Qo*e>E97)TE! zza<;nYH-v}S-f5@&FO4cfS#+8xKvV_*msdRB#s3^ zk}OS$k)KoZpza9&Jp9Q!hs2uEtIrMd?F>laSBkHUW5J-rP2kgYQqZvJ5`8jMPs^(s z;pIDhfN}e9>*z5;7r}DkBpQNM{CYZ2p@^cypKt zikm~R#Lx^plT)y)d_AVT0hlpq17?ll6{=rCL143A&^tqhi_@bZn5071<-Vci`-elj z)EF-R77w5vnFIC*zS4TTh;#Z;uVqAEjrA zhq@Qq#(jcal03hS+i3M6F;)1*=)GW}+ZU?zd%nPP;TFM;-#>tzSp>r`&WD{_|I#S_ z^HCnmqY`f=+4k#;$jlv2X>*h-$ZoZTGu1x?E7yDA5>f!0Whv%WiqH}OQ zUkA>3EQEurU(nYRr{j**NAcWH7J6p=IYrZbme|I^Tx~ma%S~lUaY0~72pUBF) zP!Sl$GN!c z)ek}N9zXaf_)I_8i$FWc6dRXj5qD!V?xA82bWKqpFP9XO#(RM?TE@9{_ zH3?ITPJ?}YJ>3+1i2fzEyfbz#+U=i3=Z249rd@t0^0UFzF$r+l-xYnXiRhQs99%uv z0<$7>$dbePSl9Rg;`Jj)Psj`E6OvCQ#E%f^@6qJ4<|?7%cw5S{rlajhCmhlE3Wu#L z0Ii$a@a~W=4)OUIDEGn0fpl1?(ALk zdCvd6j-8^w6uhgbZOC>(VFrKiyR(-h+ot2q$pZ-Pr z`cJY0V>Yp+nq6$Q`9*g6cLj6f-P*CrI!v|Mo>i=sXY&j%vnBsc##m7$`+N{YD|V}j zj&lm4PY&&D>h%F;+1$yRxS676p^9jw`aAZ)ERAW$da&y*%A)Ii0;YRc7N%Z1&PIie z7JXy7q7&_#ApPYvc63TFe_!-M`{*E%g3m#=a7aV+woO;m>MAYTnybW5sAn@vT?_7I z#5i`hN=o$V$8u)EIkG#o*(|iEpRJ!+&*E{)4R#MzOka zUs&_RvkXe^vTwI?MULxSMGmr0*|;6u$c}-iDqLUG{ltx>&ahw-Pd}5Z23^djP)w9~ zEs7Z&*vE74uaMUdXW-=SFt%4$mbvByF!{L_B3b5b7CEaY`<$WB?Phi8= z{b1FqKZ)C_QnqAQGuxMy!y=m9n82OJ#Y+${b zBSgh-`MuL4YZ#fqGmQGYMGI9K%MDy8+HyWjq^qhdx<8~TijBRFL4rrDaojw1?>;|w z%sI?{tj=Y(65p|z^PBOIi!b}5T8BH|XR`7VX=({p?C#+p=JMqj%YT*0(nr08-)ZvL z@=l*=)>mPvc^(^QxRpIW$NPwkR*6bFF0vU?$n0M%6#dSS799#W$|?RQTUFsKx#u$Wjo8Fis3?m@EZr{B{r-wIEx>n51v@_?C9 z8>%l}W1Qobpd}QSXd$D4i=&z|CD;LzT1%rBQ+o<(SBj-3f z8FUJ}6PA#zJA+xkiL1<1%#sOf*P+y)JbN@}2`d>Cu&AxN?5^P|QIg#yW|UXOy3Cu= zW4kjOjD0tH3G+s@5<}Es?GAEG{=)@4 zojIE;`>f2YC(L7#_Y+vh-lJ@~y1l4XGmS+Jgt7ex2T0&|Gxodq5>t?TRp}BpQ{)>r znSG9w7Ku;P6Isd36s?O3=C<3HVJJOlM@{`xjp! zdgM3=(1@l|?V6CqO9v)jOyyjCr*VGA-MQ~JGr-CA1qk=fgn>mjaq8Q%L^1YG`M2b3 zm~O^FVERlTb#46D+JTS6Y<#}5rSiB>3B2??%Xf~8aOJ+;RPnI`q;?r`EALDJC7yL& zo1-R7dKwHde&5K@oG8$_dj_`h+0b8>H<-a8T{v6k3cEXPsG@BX3<@Sf@xg!ad!i1E znz)e*Z7Yit6n z$`8g`v06wg9Rc>i@|9cL-jT-iKhUBxpR>*%4pv9SxfuiZVfk7c*nTen=IU9og*qGf zj$0#)xZ#FZQ-=ii(=NgF#%DOi@j2AbapW{(Tga`=mT);x4oqLCK$2G^j4N@6OHnP* zYo!NC^0&#JXELPJSOH!x5Wwnt<6u`>5a=J*gNMW>9X5VoOodpE^=;z71P5w)mp*E^MOcH6?1rw35%XfJ8lQUoKbhgn@H{YKX9yu#;W+Cf-w1OCXl!kqKF zDKA5WQOUDl!ABd~UFJ&5W46Q0yMN(=6IfR>sJ?dfXcdxBfl!|7}N(0JReMw1^At@vwWF|u-D#O`pqZCC+gVLa) zNu^nXl=7XQe!cH?Iqy09to1ziZ8aF*bq<1W0i#&+V9wIDM?=yN4OnmIhduLF(&e)Z zX{+x%GSI~yLK5b|&}bvle_$n(?5hwi82v|%Pore4*t@08nteT{=bV}Xv?3y-8jr zu@uD3b%MN%+YoQ3L(>w|xo-a*up#f8Wc0l(cp<3KtZ}wjdEN??)dS&y^-HnKTSiK& zadN-fNjTtlDIS1ORws288oQoK?kAYC$C5ixU2P8%b+Okou$LIElm$z@M(BJ00j$nY z0ae+j&}Gd*?R!2X+^;PU^zO#Uk;+tK@&T}UCEjOV>(N78kq(m25-MJfpek$g$lJpQ zVdJzbaA|WNy6^Q}{4}eAs6FgUJ#M9lJP>(`$9q6-NC#EvqfgDxkEGv83_0w&06LSp zz$d5(n*PMV^sNA^cRnS??l+)xyEJuD3Zz5lzarP~OVe1t6SP-31FVKj1QLIfKHOCX zPi~u0MU`e@(ak_e*xCqppGDJ>ko9o*mgqCJ>4w7WMo9dj2a_$%f|1uyy5`LsR{wng zHNV}54vaeoH}9Q=Pa-?%`1*;EzLd~|S3Z*5&b~N6NtK32u7H9Eli05G(?Iek!TdA2 zv~uZDXeyl$ zYeXJ3Xv5;UeZ}wo2NOF}!O(LH%(VA`)Y*o#-^NIyTRNJ0ZBDY#I`&$qO(C$>`yheea+Kv``%dA$2FF-UCyW;h0_ zrr!Zpn_6xZJA&RBF@?yyEEc}sNR(8gBmHC}OGA!oP#4=XaLD;I$Uf~ND{tDsrRxNq z1*Cw~U@M67e*%8FtJq4b`w~0-XXJ^OnB!LYDCTrMXvLa5DBZG|KFtlIKirqoGxd$& z8`}$E+CDVCq%S>}c#X{6o5Phl+R<6soTh^dpYxO`i`ZphA_F$12R`hn#H{qiBZ>Wg~_s!5N0rz z4*6IKU-}i3W9%q^WC&gVz=)_~t(-zFwcbJaxZv}1~ zLf;igS4=Jsr9%vpdC-_&LbX*aD6p{@^f|d)CO(W`u%5wpi)!b2OWzCEJEoJ2ugBS? z@uT@?$$DP6-keg&Q{wSqAznVUo64O^p>qDsq%~tJpE~6wdp0~4Bd@lC!{J#RHP4a4 z#w}=JtjLFa4h9GRZLqj53&%a04rOEO!NnvJU7Zzp{g4T8=lKZymr#;3&G$8kwPd~C%_@jY{()v+i4=#?CM zF_-s{Pnysnj2d#EnQb}BF|7>#r3rjqc{HB8t%<9Yeu4hYzwGuo3H9l+tCL`|*)TfC_J{B^q7|RHuY-;V9}I2V%zIiMGmE>9Jbg(d zKRn<%?;m^}T!z}hM$ck;YgaiwJN2IQ?oi<)$4TMNOD`nDVrA*??h`nyz82QiZ-k}i zXQ8jr0D9y5Znmqx6}~8+#D5RB4?B&>N&`eGxNcI=YKw#Y~_F zV*BykV!lTAPO8MV!<1*JzQ*f!$77_t4F9h77i!8{h|#{oY{Qo2Ft4bM4fkyaBaI_G zSM3_^9pWN99i)wWeu`{_enWZu$b~StY8Bt#Go8Piy_?UTr9(dyMe^`5;m~9$?hg|N zpnt~^a;r%l-Lk%s{R+=fdd+Zde#rr@ogEA=LzH;N{5R+zvMa{4H=vah#RInuq))xvje)v`r0Hh#kLAs;<0Yy9aFfe1R=<48*nn zCX)l|JNals(dXf%0Iiny;Dr=}=6-dUC;H3OL+n7U;|4lgghHL0m_>5e<~83c@K(N; z@J0J73o6orFX?SKd4&_q{CWcR^o^EG|Dp%OcE03NI)UgQ2_?I?d_Z}nYAyq5!kyJ( zPbG2`m)jvfT;N>vBKZ1v0sqFvFsk>9N%ttB zrcMV~uXx7h zIp15{|4yW1ul!|FVI{aQ`m?xB%7h`u;<35tAUPgFx%2AVO!ue`%rXDPR-Mm6|LYXS z>DFfuZ*6*NbnmcSe(LoKRKjW5T zU%7^qHrLZX$UPO(x#hMM*tu|>XxrEe?~2X}h6bAUJmB|E|P_$?uLW{bo#G6|EJ zHOz=_#qK0eW?Q0!@!F$MYFh#pUG+R@TF$DaVCSzs!*BntZ4Hw%kUN z=++lEhh4^qxtH;#nIME|@} zL*t}sJoe#&P;*p;KJ4=uZ+5HD$}=l5C)*Pr8AbDtS~tYJoD?a(zmBOt74zSZbMe)X z1$cDJG>lz(f%kmW<`L6l;q&W!`0CcfTsFE$WSkD5d6GX4-?WmaS!dufyFzsCIxI0> z^%kuc`|{luXUOcdgD_*op%F_eLJVXsXC# zzK!6zyK-5EQaS9NwVA8mTLYFkx5-rvSzujBr14QOOO{h1LtTDJiflVj-{32rHBE=r z11(8KN;KYtdE6#_aoIGfUBbK9sc0AN1A=Ng^p|52^Wr9^Ab-X3mufxPHg_NSeC#IM zb^A6M{zDPMcb4&zReCh-@?XgX6C_Rj24m5Jt7O)Zf6U)91*4ZIaD{1(@Yl5x=h=F4 zuLbwPI#LFm^vlHj#R!_3p+lc9`bG*wo~QZ(z@uOKz@I<4IK0pSS~VX*_=S_@jZcmd zzlm%3x7Cqs)dw9M&#nB^8M@{w;_R?ixHo4iud?We zhaVuz3R6L|nN#su_!-D9olh48jX{kIJ#68+UnIC@8){@DMD|ysmw(BiZoL7K&2WI) z>2CqjMel%CoDh0Ef*t#|0L@=#vRMvRux9l+2pHWT4=pfd#&=^_QhO!HFP;OZM8o`n zpeX8G=7OsmO0aiEKYHfWG2+;-9OFd)=U(?@ws+hFqP*^gP!01WtJVl8JO*C*_?|`D zND3AslxlZ(5n&mgM;kU%+XzCovO}X1} zb*DMbw>vJ;x*jP^FC2vKQ(D=Ht#-t9iUMr#HWPh(GBB{R5Bd2*%>LP>k!M|%to_7Q zeCPfMqdhv<0x}rpS6sv!cZ^{Ou7Tl_on^yRyC5UrGLFnzfm@{a2qR6Nk(vwNNxR_A zRQ>Og_gSMbY(O`I-*?!M(5>i0*FxZ#2&|ib61&n@N)(cg!N<}DxC2tC_~!=+j?KWE z#~LNc$_L>@)d@0b{~mPo-XN*@@dGbiPGP=2c666fJ05r(NVaU6is7=xAhmiX=4K~L zR$mlnJO<}k^~5p~H$RXS_m6@LbzPEAZzEv8jXrTs-id`TXR*p18ZfhP z?1kJ)cB=UnVY?o)x`@$ak^NPS>zRN%l=QLKc?5|YTOyRal13GgEh6>l2wCB9*Yd8N z5BwOV%u3D&38O53QXcI~CTMhCbyRmz@Cf@t}Td+{6W3FSx zoY1Z;Jdk-8T>1?vKj1AgGEMtn?!8oEV|)P?ZqVn>hBA;^@t#=eKL?-tgK)tUN1{HU zly#4`VVleZ=)JLIOPC?in>U5^r0j*kZ6R0;4#$eR8@`Py~q*7TRg*UR%PTr4%Jj=p+8$2g8%36qd93z+-eP(mzC1kfl z6m!}gNS^MgAy!f&arw4fIAb#l-0R=LsRkJ?37k$E2WeSEZ0b+mNy|$7w`5CH#>(<_ z1wV& z7a}LRpcW>IdrgJY+T6@QmCj4c$BuOw^win;Y|_6fvb47m2f1uUw|-gN+`kzfNJ7wk zRTWtmBugt~p5P3}YD$tBy!*3`uMBg8gPsyd=yro?(=?&7SRLZ!!lB6e0ZdMx1eH7g z2r)JjM5nJTEsefm8TcRq^tNZQRkL#G!TI4d_rq)If4iGq3>gS7H&l@&Ugyd1u$x51 zMUM8~Gvzkd{vkc9M)qdPT0L^^1io9H_J4R(VsU>2RIJUVQ`=;z%HSA!_rqFRK10RI z&mjYv&D5!%<_fB9Qw@P3w)EV#71Z2yAYD30ftKCY=Mlb6Y)ym2a^%+>kp8a1TYmq- zm(O-{tAzeob0$e*m$(z#Z+^CnZXXS8+xxTRkJGrtrz$j1*@>zx^^)xx;k4fs89GOl z$7YA-!1MdYH2AoIFj2)6YYHsIS-w6Ur&op3#(Lt$3&Egkn+FZH`{~ZB8tm~T34i20 z9A2$7#=tFCV95bpe6m!EpG+#o0Ue>>v@{K;Yl!Zaz$SS9*$X;@6riJRHP?wjvj6%W ztohxEM0zwGxJaHJ+qWF=tvvxg5mPaB<0k$kY65?~^ee8JAqQa#H0V;-WprcHU-n|m z9`Lr9NZpPO#QY~mfL(B;3)f{pg?0$G=o(Q=pPO_|ohhteaR_RCmeYMFhmprzWEZ*H zaA8@GApdt8UH`q5jNXhw-NrgF&=#32iaD@#K?@u(PC)q`efW=@1(@Bk5L^buL$1bc zXx#k_wl&LPTXR2}q9OKw^PAvpQhG2yJ*9FO@3#RCpSKqAu`t{(md;n zIP*(?S~O9fB#lwyQBKY<%6pZhJPNp3Vh}(qbCB8IqbtJCcl+ONpeVA(4S3^PV zF!l=F@ygyeBqhp>{?+Zn52V~Dy}9nwHFa~7M@1O40A zLyMsw{Pyn#m!wZbKE)PxrJctUUk;*SG<8( za7poPV5&DI^W^iO>rQHU<-mh*T>3Q8ljNgIhY_Z}Qy>v-YgzgaKfdWlDpcev;`;8b z=xP2G%U?&JMa&GiKgfi;D|k@D36(7S?Fvi#pNSazk6@Y4KIk*1g(w<)fCJ|Bq|47A zjraLbr5S$oOXEc1;UtSbLl*F9t7KuK?`-k%8A{(Q*Mh|AFR*@M0X_O4fbJ9by=4#U zAXP@prxgyOV;^hLb?bo1ulNXYJ*s^5>bUX`-X%hqc-ESH=_nTa(#MvYmr*U9t(BF;qtVVubc57bHtVbVN?DtDw~Nh3QBhkx_=1MHUN^^xr|k z_s+TKeZrgEIs6S@E(#7T@S>-%Govp_3-@FSU%!=) zf3uAoIWdtKJ{`e!?;J>0pNbTQlmtSDeJ(t8sv?Pd{h{4I5!@F4Ce+~)c-0k<8PWa8 zPggzoaNdfHl$L|7HObK1d=J9CGjM_FE4b+P9$rSaurZ4zI6N<1*nT{l%$qx2+;^OZ zXS21*B`eW$?qW(bcRQKqD|&!+o2O)rgN~#sR};2Ar*PCJt9AVeslnYW z=a4fA33tVuL-p8yVQ={#zg@hdRF4c!aRaS|+3f4ZdqUdQ7U9sIy|7H&j}`5w22Q$7{K;mj$ z$?8Gsbl#&BiR0|`BrBzwJx;$y_P5%Rb3Hm_M<|ozJEXEn_f3V6yE1gJV=cN z&v*f=M*0XfC$|dwpL8&@a0RUXuT>ZnM39GAz~v#!K2L=NyD1{?6~O)sPK5l z+STOPfIj_URc{k1*w!g5zpx4f%@E>l^Vc%x!B0@zcC`GTc&=6GvmNvvTovab(;@Vk zDv5u-9!e%IB4f-&Hms>C(X2=ib~IXnX9ST<>yg6kuWbRxRuhwF+3@u@(3csTi23)g z?E0o{%yR8OdSuFSRKB0bw#caC^pI-O+>=bmsYEtaYao_c9B0Kl+9Yn@TbP4d2zeJZ z43dDGI zGJ??I1up+$iN+;y*EJ~;Hx^65)SicMH7%dL-8&cM_-dA#e3ccx^kSp7>v8_79}1eY z(Cv2!24&QOwYm#=_O%U43+_o)r7H4$rwnkpq!3*Ob+LH|R!Mfv-U6gs?R6$Sj}^||6sIaop_$w zg?_j)p_R=F+KVl9m$-gmlBLqO%dmL376^~uu@|dm;-$6X`TdX+CWqa|noI^i78$_- z2U|S!ZwhyE_C%>rF%J4iiL4Ad!h&8u!y9#qsQp&p$Nz`wBlf3hB=9hEKnU|X$s(z|Nm19#dWoRywEgyz1>N@!B z(Btxnw$f-j;{|&*{ve)p{Kn$@>fo?sN015_3ai_%Sy&c#3tK&J;kV*gLTqf%dDL*U zyb{N_R4c>H9c&y+#f=kAW4P2%F65@eJJm$?^pS*mCmM;2&dqo?;UNx$EUdUYA6v`L zqQmQAHaf5f`WQxF?Q28an4SlcH6G};I~pg)-o?Amrf^lIOG5GNe*BY@4rWQUvvMP6 zF{gWiS7pYa!irtksWgl`B#O?%m3L65@C$xz=+E1Nw_?ccnOvvsD9#^o2v|lF7DWr_ z$7k?cZ;#>5*kqi&MO|d_R$-v)2TVD85*1awdF7@6)_G$szIv<#aW&=a=4=}_xL^k4 z-I@p&WpChLtygHfa}vwmu8vlp=0cNATWJVt&~v z0k5T@$l@T-IlEHuxZaBH;}?@B*B-Ief`GR5OIXA?dweU?AD1`VU{hO!<an-98 z*sb1z%DqRh;$ae28kXbAz1o;Etq^_$D#5x_+AwkRdnN@9!oIRP>~LuS%X`&=|F7G` zKj<)iTG1kLENm6DlXFnerh=HJ2X)6p4BV3rqw1IA+3|UBc=Q64EGQRdzE5Vm_BXQ@ z!wTj$a}FjAOk{ODKvxQJh!ypU)lw_Cm|*uaKAo`Y3Kj4ZEaZ9>;6>a;fVEgo97pMTqCf!2MN;L)rj zIR1hR5^D$E9sFFpR~|=)2z&lsoFzvbCphw+0WZr#eC)pzD@3=E_MLNBF>e7q=CE4O zd2h+5eAP#vOCmGCOp2RM9?6Gay@9(9sPY>(Ye?PBe{4)#D4#XtDDB*5gXRB@v3(hP zQPrpnz2tOw#plha?HdGVMCSZN#|LQSz8`lU&O*(Q6kgY*0ZtYc{F~cd+?Mi*ysBwK zzpc^Wu_c<_)jElu{nUBvFH3GXc?NIRnai5C>EVD)k1)h*xiH$u0^|H=;5Ao{`LC_H z^NWG}{h}x2zgvU&0sZ|fajOO%_7%NXLEF)=t%--+%I6#Qneg)t{b<_F%gBCy#jauJ zaK_y@wnWbpx2KzsUt-^X_KTT_aTJwa9LA2Fds%uwA#6K0jm#=$K3CNY|Uc+&vF-kQWn5Nr)i^t>rV0|G7?{lGr-sm2k!sZg?}}v$NkO| z(Q5fnE^c*UQG8#Ps#$_j=Rc#W{uqArWD#!v-hc;v)%Y5}6?|iZ74E&8itFZ8N?-w)dvf9A*hI#-TYXMHz&Ve1bsHn8C*>*IV8AB(xkhG#SQz(>bu>*!f*|If!nb!ix$()8ss zQVMYA2Mg?9JeC{2KZ%!zOhxZuA28<8MaWY(gJYrf=uDD{bE!67P>I7iBSNs!Rb*oo z&f?|{8!+ScU*?_FjdR>Wc|x9=@YCcMzN@juSM%DiPf0H+*13Xhuco52_*+WX6n&9} z0rX4z2^M|o4yqff^Nf8JSmkydtwW+f*|hye~R6mDKDa_`&SbdO z&9maMzHc;o=tkhazX4QF#tQQ_J?Q8m|AFS#2h1(TA8pg@K!QMBt&I5GttSPwo>um3 z+kLj6RvM0$lt>~T$Y68LQ~1sS=#wq<-0vjVvO@*ldQD^>V?UGmksU00jvPFVcnXv3 z_JVTRMI3hZ6+FHqk4g{YU|g;pv^E>yTHStdF|n>JpEr}~S^p$PLyn^RpdV zDu9dI0M2!MBS)eLncH<2+S-H2(0+TMHqDAIFkb?@Ocbfkjj0fqSHyA*RtQZdo;WL8 zg|ByRg!rIXn4M5b8m>y{@&7jC7Ea6oI)4f7liG z2ZD;oQN3~ZEcP1OL#oMU{A}cfcn#n`sVgw?gDNd;7{-H+tifAlJHhsMme~6#f@@1U z+0x^;V6A#TE;V@yyWA(88H;|=-odUgZr%e-oF7QSe580wziOxpasc(8_tCQU7Ibc{ zgmvF9!qR2Mpi(soO%~0iPnH|gpT{F%kA)(=aSouY$ATUm`AMiT2!O-s*<@Ud1I{dM zkbJrV^hI(Jb9M_QO{E`6jNUNpdv7FPRIA7p0u}kNE;;t=$66eq97}9uKEYxfiZ3ho z)3j^JAl)(zPREVMoPXKm(Mf;&W;_NYPyfSzedn?HM%KdfmeZhbbPdAC{=~YfD?)s4 zHAHWE!ftAflN6S}EH&JqLll>qK#GhaA3j8mdC)FN&DDBr{S`%oAKs9nv<%l7Jpzqm z;V||6c~JV(j}}lh`nO>d?s~t1$%*f&bCz>SMZ`gnekH;E;{UV0oR-x3foe z)|eMGkWZg`0IrmW!{K9-Fj4L-sb6j>F<&-XM)-5?lq)& z!&`FoUK*V5IEFPg7BqZ`Kt}r-(`gcsb<{qahP7P+l4^yASENGNh#cs$Er7TAduc(A z1{-uFm6;vC2G4)!K$D{e&Hdv`2^aB@^Eximb8EuxcYx{|w3?~p1?_4@FL>OD! zh=zeH$h>J~Ox|)Yy0t}%J>?rpo@2orgl-C8aq0TRe1MKh5}B z9rT}d3spXP!2_)Xr3>(n*OYLCVVnqgJvP6$Avdi4N3S;#%4cUuJn@?d`#^_Pk~J z;(#Vrn7$l>&%0CM@-W=BwNd!w@&E@^=a(ByAHX)%xH9tO7j}(rh6$yCjKe74=kg$2 z(}m1w|A32DUt@uDb0xFC&O*EXpRic7mrbi11U}h`lCIc=xIS+J9{C={vSJ7Dv=DLD zoqUT8QX({avk^ue*vS($)JXLsOKcFG*N~Ecc7e+=(dIS;tjZ>(kFwBSy_OvEIR|A| zAHX!HQB2Z39?LJ?goVG3;{Ccj7C$xu9fD-2^n_Aq-{%X3zqf*=a{-t=F@~((bHb(a zC|r8P7L|_7qz!dq26fXPQj!r*u4@$Io|$n(Uc869(?Ym{x*&H#54}#khn%LhL~r9p zHlWOfzZZt^KPN}=o@ymtJ?9Q|bbn#Tms-H>DJ7V!VM{!pxV8;9%}Vbnsga&ULBikse2Mdb=?3<^%ZL-2{h=LrG1=a>$vv1fRbU8I<3% zsMh_psM;BUAZ93^50t@WOFh`7gGbo#^Pxz0hr-(qC&};M`Ru}cIg}nY1q?n<$2nUg z#P=ETH@f`@Q)?YAEE#D<7w`TIW9=RJ@C&!_j$$)DzhDH_)5FDgJWG7q6M|7K9nieX zhkP3!i>^B9(EL-4*DCvf`qd>|+Sg8qiLD|T12cp#qYr~H+mV#cNFoa>yRa#6BrffH zLS#dSpn-uVZa+{6j)hjRwsj13y%@zzx;Xj#HV%tdv_YtGIdlzHfJ%ib2&((X=I@!v zZyXwf4)>yP-MBJPsHg#r;vqD`{}Y=Zl+LWK?j*O2%+V??g7Kqs=>5p^X#XY$)F%9q zD2S}-DMt!mr8ra09H3))Tz8=md{G|mZuDYPeU$*Vs-oSOmy)Aj#d+m}E0VWg_u>P8 zCF-vyPuK2wj4B!fVEob_=r*YarX@^-!F?uRh{s}N^A2(aGZ(l^N{PmwQ?TMvC0SFQ zN)E+Wz$S_;@vk<$rLM~(tCo?<%Rg|HI5vE;DX;wPxSeF-j$_#PZ7;iA5hRge*<|~X zwIW;jmgGub4KSQGn7lcwiu!ja;J1A!iM1()m(LGV+0HM-vHvm2n3@q7H98x2YhKDVy9K~-;22g$R-rYF!4Ovp8N82~&fr_qz)z#ih)HFnq4v$>` zHoEG>tobZnu;_tT9fhFS6h$Ldedy%=QB->ypdZ(kDd1xcJvV z%>&@f;ePo2sv-WnRYyjisK$OiN>(w(hv+Bim-K<(7u0DyNQ19lq76STVt`sW$S60# zJ+BQkDMJlA^U_garUUgikD9Jt>23X)e^bm77{*x&jD_J1qF zGVKHn(|^LIl>TGg58tBA*S%2R`oXSDd z@szM4<~uaqe2a5M&qnUOOK|X(ALMNuXhkp1p&R9<(^%yg(0w0|wQhU)j{=c5-FR2B z!B>{=h&~7LKFhJKL6N#HsDqh(BjK~L4D<`IqcW9U(C4Z-JWi9PYRz|mI<=AgMdFUI zWHz1o#D(UyAf#zMWplmqNZpl>KyP(Q%=gWt8^>0OY$^l3IUt{`QIbHCf4eXzBNh4# zT})pl-D6Qw95QkNTn)c~tobytH~$X)Og{mywxx5qU`k7Z2U80z(Ko;97}s8DfRa#k z`f&C`nxWIj%3xWTRnCNwR$IsGTKTT}12}jv^&dZp9;Q>s5PN+%I(r+u+20G!{~T!F z-L@cebtqMu?8D64vtWMGXDqY`rU8b7Y5qSOY++7ijx#^P!Aspx(yWeZMcVw~?qXJK z%W&nsY)pOA!E9p{!Fjwc=H>@U-bOuxwf&0Wa+L~c)Un}*6OO^j(dF=P;d~lP$54%q zcT8?sFDAtepi1xFK;;#0{`-(87TOiEy(JsqZ}&Z-c>X-B{XULrn|x+_541wpD+T7b z_aHN-{8jQ>_=IKKU@W*)>HaTTKpKP59fs2Od26ixgr9PJQOcb~+i+iK~ypz{S_#!{9mm?dH)usq_@po;8m5x2nOU znZaNlQ3C-H9(2Hk%T!iJnfg@O(m5FsaO9mWjr=u@o>^}XzV?7V5>q;Eof&kk{|L@4 z1)!m$LSw>`*@N0=<-0K)rdWQ)uq!b@FTN6eWbO3QLG)=?F2e(^qsy4Z`6T z$SfPrqB@yGxAaV=TaL}(B}L8ndCWOx-mnxe3b}~w?JVb2Dh649W9lppAGXYefVqwA z=k@gG%!^;OjR0(pqo}aF}>ePw!=YY>y{15)dnEV-V=6eFr4mJ!of^0 zt1m9HHX(&y#tRew$s=`=$C|3o;(peFY_1+GxT!xOk918zzfhTeoimx3x#g3UA1(>8 z!?KA|krtU7qt6fQ9toB0D@c(_KRR)g*l+Tl%^fqxVtwN`$;+z;P%Tvt9UrX0RleeL z-BE?R-wZ_!hkeYr(v3;UMzPIqI=Gh$cp$$ot-Egv17u&2pgpI-c7qXA$tyCc4F{Rj z2NNQd@)$4FPL_Pv{{tia<_RJ5M)N5%Zs7{2kLVORlMPcn3zH_Mvy@#Z_SH^6VNrjQ zTj{_|v)UzPn^kD~zpZ5Aq#pRM`U;xA`AlTUBDhs@o8;e6r`N}f!N9%QP;hM!@9p=N zdEBwbaq>Pe?&DP4z04ckGrHKtrZ^$tOeS6)<|WSNW5|^L@-$HZe%XIGcNud8AGRJ~ z&rT$O`KLoLY0Dp2n>Pl%OLV{>I2w1pc`wXXF<_HAMgBol4~tj-!8SEyGo!8@?A9D< zdNjC_-Kkmy?Y#*&rh6QV$Sfm@m5TiG(cNUl0p0RJd@(Z_nQhst5XlP89K!s8Us;~J z2L#;>1SvDhXDaKEx4xQmokb?^pOnG){HNSj%~vv7b}XInpCPX7ks&#sy2;#=J6MI; zEeMFZjK015_(bt6z9`d#-U!}{eP)NE|Kef%a#a<%(V8jQd;7GcatgJyc_E4DIDMFkBak|1fqKQblbAnts43d+?t_(sP*)X?lIo3#2jD_+w_ zaKF{b;*XYMt!Fesas*gPg@IR46;8f6UkDzt0}AX0(OT75$&Yn1kUKsI%$NG&{EL@h zqe&!8aDPl{O+|L=!<$&@oDQ?cp4-g%*KwqlW?YQ0}fxW%uT+Ghd0$hlKF#msehv^P5v_m*DWkX z<76@0ccKS^XFIdfu~*2WbLxiV4l*P|b5DIBp087mqQ# z@SHYHO!wu{gRLRHzK4x%eg;&`7@o>m!y^>Fp^z;{&*aNObcHnt*S?~??FBy0D2F{* zFM*cVPs#Ij66}bXPd%Lvv3mzY_~Y)8@Niu|JiXh+EMl#2zR>}mpeM4zeQ%<0?TRF) zX(=cepT$)sM{v}UYL+1{!KT}%X!*!^c$L$GCI9uXd-+1xVXvLu3pfP-2v@Bn9 zPm6aSdMMnRU=F=0D&$U05x%+~iR~hqmJxoakNSo3bHsrV)>KZm;9}^RvvG_7Cl9k4lybP0_ z_h4pQ6n02Fcv8+HY=83)r7}a==3{b@V=)jFl)qDm^pba z|MuvN#5N;}FUqrnZ;8Ly&Yx@Xbeb);F7x33Ef3{Bi#+(?g(0XtG=_gT@E#s0&k;Ed zuJo0-JIpw>TXb(6gu3=8>?N*()L%=$r1YaOcfomhkbRpBQ8i_+ZkxfDp7$gt|Au5} zfg-;+!<=|*YXqk`BXDL{HHeQ-(Vc?CvU&;LP%go9f4)QfVl8S~oQR&DvT&r@0!DjF z(ZR|0NRYD`yxoz7zJobN+|9?a(c(F>YXRBmf6HP^_Ao-Oti#hmlc2;ap3J`b7<8uZ z0Jfz^?2trozdns{X)NVeFMU7}Da5eoC3i7d%!ilnq%+4KBoPRQi;Bcd-S_9dp1~UJk~8Fv6j^ z;Z#>M0Q%H>QMp!G^2TQ)e%0K9z23=^p#OTnqnUtbLcNf-AcPE+{l>okoC3N|hoEfs z4LsX>2*#Xr7xx_d`RK=;Z1Z3zILYpit17Q?OT;ATw_^-OrpJ?##!5V%5CsALzYv_B z;jUYa*tGmVa41uUG3+}rKbwHL8$@rheUq>~IiGEv>_GqZn&YmI0*eY9j6a%?&6xDd za;nZxSQozx1FEbb{N#ALqM-&Ho`k}go9(3U$p%T>a^`n7ZkE=9;_KCdg;5OaRG5hS#17s*-PaiKae|nG$px|}g+$IC zNUwUr zwz7&v3t{!xJ^1@YrNpgwGd$W83^G2UP#oFGDt!qmuN1Rl9}6wscI-fp!mD5tDGl_^ zey}bNBROXr!Npp@S9wWHacVOO*jfdN@3M&Vh8WNnu91xXD&RJ24=mV{LQ0a$+0n@N z35M`4d$`d0u@lG1 zs_~CB5!ZN6qRDxxaIGwc4f~iTe9g%t2R6&$!~-MH{Ff1ZTwg)nbs56ekA|=$L<6?F zoF#Rg3c|^o8(_}i@5F9Y1xe(}v}V~z92@YNwB3D68nRV!kMAVt!>lp`Lti{bvd=q7xJE$LU#8gC|;QlAKhdy z<7f(gvRlpT#pS}7g`;8U`0?We7S8K>*QY1VWa#-Yk&k~A^KL{M&zPz!|aIEbWxX!{JPG0D^-(5aRgZ}d*$EMM^<**wYu|$onU$X_dIR~-SGz|4;eB}!53qbbvCf;$-Z1EYD z>Ed>wXCJ0LTRh-O8EzB0i(-T5*!wQPVpJ$?|P z-Jjq6u!5{M_Xx9Lk~qg_pSZzxI38=Cgrcrgdgv%e6}NY>i<Lo(fmJeP3`JJE zvCS2RDBZN1rbP6RpcNswA}Rp zg!>-A+k-AZ`5JdrzIz2)k^-S-x)tVK(Z-^U`rJKShs7C3VNhOQs8m?XKFG}h^=W@a zUq6gy{hBWk-q-+p4-8@M0~>H@%~cqZzLY!Yu8mP=+sWGF70sHrQFwQmjepB4;r$&e z{5Z3uc5(0;uzPAOQCpu1fzyph@y|SV?$&3Zl$ofbvzfpDXfv*nf6h65y~H0q^@JPe z+{+*T(Z$VM{fs?4_JxMNxJ%j)%q~Bf$EKF%(c*|9*x8&x&IQppvvPu)1WNFp}56Bhna8IWZ#|{G5z8(EOPuTiT4Iq7^%^RdDc4f zXPk%eGd?b$0YwLJiue%jtl5Jj7p#L<-6PRJNsTEymxa!O|47k4fR*pDfi*edICGmi zR-ZpE9B~Y=-}f_>5txiwaK=IbkxiDdF9OxOO zz?RBnn9!aj%Gjucn?kN(=c7Q7*&K);viFgi)D+ITi;o-K5Dx^~gUJcxX-W4I-T>Pa$fVl-a)dOXre`TvpAF6m}ND6mj7` zcf08v$h_;KkuAdg>?m6{(fP6H&1n}V_sW%f8~5MDJf2VIdE1TiNhLf6K9uyQ&*r1S-t7d}y2^qsbWLN+9Y>P>7G>6d zvKKRKzao+oxc0#`*4n)=ZwEy^Q=qf3q4+ERYaH==EE((LPpsnmwY&I-d zWlej`Ihv;ujk|C+xSeu=#2ahCXXp*OogYjdAq8y2-!U-%d8Fu#Z5TL3#L;t;so*g_ zjH&O?XG6w1Fh|!d>`ykb7n{RD!DI}K)tnBlm!e2IIFEH57VhkArt+CxjzoPrDe6Bx z8h=Vk3409rT}2tf`Cgsr9xdfH@~b)X%}G>d=g&-p?n~iDj)guO0m%w}jG2lY1&B=*H+Lj)A1xw+VJ8 zP_^-B)}X(X;=02{rS1W2q3luk5va-@j8B2UBV8c#s}2_RHwBX)u26RI5BFKP`#avG z4)3;mvi?U5nbLVv_DQS{>s}uv^Onopr4bR5f~WDI_iiezQ%Ita0dF{QbvY;`mx46X z1D?J<#7vtGu?hVpRQmNNaCU-QBRiPIIjgd_0lM^Z#9gi=Vh=U1b)@{Y)A&6yui@iu zUQ*e;gTHT{sNE?kq=-O!N8%}?Rh4^?MNLam_C_c719dighb z&vi$VcPan%;E5BsMuG=8XnwaK5DC(z@+hJIW$_}%&zujp_g64=m`vA(rSeB=W2v^Zo4;@O zjOIrK&L~p=7zHj6K{(Ac;TF^gU@Vpw(4F6KFbj|_aeOa*QVhkO1 z_yfjI2eC;jHF&AQRj&R(DL*<=0%rG&P|{rmW_{(Q_mhs&#FqQqbyx&SN`>%5Z2(C1 zR&y0WD>;*YW>}3sscmqUC|UV5HWB1zqm1+BjC=zc>5t@#*3mv=LGP<)J{Pu_)&on4gfKMG#|;zd_wo^bx# zFLC$%Vz|cwgEi;%ayF{Domc5R%H~jKM8pNST{(oFBTg*P+?tuVjb5zz1V1F%aBl81v>6n>hQX9h53$)C&aWIBLew zs-Z*K!a-jJja z+0D-jIJ=|_7CNJXELIzX`w=hnFZoKEpK|zEA-}W#R~KKe@rcj&Tn>J}EiwH}5mecy zzy@L0B-g zxc0a8VEgu+Xo*25oVOar%KuvluP*pY?oVCIPc=KijcU8XOIk8mx0?-9@bRORAwyZL z&1D+&dkvlc(F#i^u7ExHJHcbXdrI46!MuxnAyoMXT*^~H*I@{@T&=ATSc2GF`#zLQmvg~N!d?Z%}z2 z%C>m7@uGfvvF)-7E9e?1J$v^7o3!98P5Q6`9LFn22kP~eCdhb5jn;kU`C&l6W~Ok% z5-wAp*H37}%ITQATo+@Oo3i?Cb6CPVGpWZk;X7J6RQl&d8X3$^VRtVbU@Jp)rB&HM z(uU(hDT>o(m;c!@RgWxY?K_I;Y#J@~!5*s4ZUO(Eqg=x`Rc3IenJxIICSAFv67#!W zu|6l>vnJgEQvLK75-qg>(&maNDz>v?KNIVi-n|u|(oxJ}fAp|GNheVB6AJMg!D=hq zP^6H~X0Op98{NM`TEmkCbl6Cpu70GGgL!6sMwsctB+PoX3MsTFa!KpMnU$Rvztz){ zO^eWEv*#N~x8E_5((h@~kCQrKt$h}|l;z3N1FG>?#3)*C%?{5XQhK}$ z6E1CGb#>a(W#|1^&A&u;uwgt#Y3yL`6>?JZtESTNJ;wa508g$X^(LfTaA&smUhHy( zytH9i1XBp>f=hzOr_a9|e2J+HJH}4ISW^`y6bM<7hl^CKI*`2@r2r-Ju`J?KC_BA) zABFA6;MaPHnCP@8TluG!rI^*TH7D<}*U4reJr~2csQxU`TnRJV#jtdQp|tw9we(i) z1MbGVJ9PZ^AQrzuUb=C`OX@%SJoMZyVjZtkX>y1vngi`W=;Q8J3{5B$S5ehOkv zg5RmtYAlrMI<=DT#-}7O!#dXZdwJ!@cpTz3jm0`$+5zN7UF}H62LN==XFls0L zW-)L2N%!~ZX1za;P>*o;AYNp}f(D*rQ|EiLxU-?sKOGO)G{+mP_RA=oQWZ(ZZ2C*T z*b6zQrMm2ePCAz+xeKl3?(Fl#JZ`O>9CP$7XSQ|OFek~A&2U@DzN)>Ys+Wt{fiD$o z!=7Y1AxUL3r>%yjg=?|vWG?$RU^nwQD+>jCevrxs!38rb3+8GS(p$|xWFOV@7e-pg0pkO;LF}+7&L~W4H3wQ^4`foa?s7v}TwgoL;n* zHL0Ek%h>aR-~1;1iJ3})^0(>q8e7)pS4R64)Y!x52=82(nF{@`>-EA8AB@h!o#eueosr31wvCJYtM!KudM0#eP&ce@s z;|IUdmZs`7(zR2Qr5{DHQe_1d>8tFMY(=~>Gkv}YWjlm-LwOZ8Rzr>D2kXN)dp+h| z=f^ho%VY8#a?*?LvCQ1foyi_DmgdCiN}0Z%^t^=v^PK!k;8N|7n5XQe9co+HQ>{Jh zikPwf!5&P}EQCE=U(cr9`_1mYYGe<7)pCEr-*V|e16k;VJ@iUwV!i!s%39{Cu#q(v zVcrH`j8GayS)1>%Y0Gbu=)+iO<7csP=XsjHOhvkR#XDNNcd|51GeYv=W-u$iu?s$r z5wS7n3)uJFTCB4?A5MlOvuC=x_@AYeZ5FL&2B)sDkAKryy^`Quh}4&AYRgMMPe`S_ zt+&|%br%?R^?2>5F;f27j6jwoW69=)t4lZQXRu0z(-h;eojo3@$L1_ElZ5Weq&Ca> zl>Iu7N%$6)a4MAPKJ6z(fuo|9Gk^^nFp3+LI-M;a@t5b?O`$tRLb1!9vp-s|nd5qG z8mJsYuZ{L{nRjH^-?r6oQl^}xEOcV^vSzS!U=xHzIfAB^F6qzsLK*L_Fj+eT=En81 zDH|$T+9+?fZDp;{ZHQ&l9!(VZ?J-PzJ%;t$7{?;6on&UxRc!Z-tMuRGvrNsUkX^5K z1>=|#kgGWcH3Ryvw!UxJxKVvsImOfa{3j%G^<#4*PYC<|a`sgjut{E>&G>4}8uvvq z|D(Q4=ZXz`{wnhF3Vi1B*~Z(^|ilE4=0zQuxcpEAt{JlI)u zfzp{X1TToCTp7pEIg_1?7)%RQ8=2MvFEX0Cfn7@MVl$HDq(dK6P}Q!x#67y6pB ztj1DSn_|Td-kr|m)|_G!wWHa;4YQe+&_z<0pTGhYr?76FNi1vkCz4Cs$<)M_>}QQV zD;5o8^d`HVB=xJ zI$Xmj@O=a6sj5pGovyN7UyB(m>`PnzZY0?+*U4l6vexWHv}jnaz`1H+zfxC`>cn1V z={gA(SsAhJ3ZX;XRYOr)>a4x-F-_X(oCN+rm!KHO@138IHKd> z!1}?`!>%P@VY&!LTXgaZ4nE_qJzmInofP^{9}n_HGjeI|tRw#1B6Dt=^oLuf-NIK*+yXf& zm$@VJt>CAT(77^y=CWdFKIgM(Aly^EE;|1+j-S-&hVSPb2fs0HFz|*pEL>L3r#r}i zq-Gee{VScm?n#lf^`7J3&baOL#npnwXPH6$S#=z_=@-9!O#wH(V1>j#-x(Ue$qM&n z*<8h%JCgi2i}@aLnJ8;RU(RvYZD;EXPdI+81zgTn=Dr;L0$LMKfkZ)G)OS-pjr3K3 zVoM!*f82T@Cz%m*ZMhjokKnD;u)AZaMCEH8sv?=R7@s8mjRw}P_)TUQPj3K3kFrJlx+EMopbTr;WT=-5*Qj~ z@W0OL?JK(aFPRIFc5|EW8u3S}M(_p=!h71g3H0#(WbifnNK?jq8wdaJh0E~Y%fUWb z7e3153)z*KH22O606aGk!5WGV|aH zTk`q39RV=Cdk23%#ie#ugFGMjtw=Oh=+`x7&7@qLUhaF)cz$wbH+RKGfn3#sA^u-A z|I^xmGumKEz7_i8v?jCik@}n|m-ImYb?D&m}k?V34j1 z)N!0>wX;1}YxAHsGjKco8Kwp|RDSdE&;N;@8mEiagbx5Mop3N*P{!+S-Ofv7GvV&) z3{i!HDp@zFQE-1%Jb!-$B@B*{{A)QU8KcWUeUKu)c~&ZkKT`l#G>?JD5f@Hf%Nl;D z->SW|BS|#>xjv0HP=lrBUxnX;3nJsSdwE&U7Ewo^fq3S0qhx<*n<&BTktnVBg6RBI zZQSr(6|T#jqF>|8AZV#6+Y+tE64pGXo!*HQRCbcy-pk^`%H!CH^7ZU}TsCV(3mAUS zmg>slL}!;hCatZF(0v4KK4KNi{CRa(9ZI}# z1T1gPX7h?oS@`@YrVzIoUTE|IGov9aU|kh=WbFk}?%VI=w*M(VXaf_@pns`X8qcSh z3}!0kyQpN2Eb}`5kp^@%QupOmEb!|blJ;NC#`Gj{ra$~?!e1?x_gRh6Cw0j*n_@a~ zd^?pa)?ka|<4FBj2A@ABfmW7CNX_1bblXnC7{fT$@A^A3T9L!8T~bfwEm4p=QFwm- z=jq;XUw-!wefH*dF5C>#WLGAgVP|v)OP}UNP4oIfbgePj|2pLxmp6yO(X(!0_ZRK3fHlw$j_PH7D7JLteJB%hUpKT@IV zzMHxCPi~RRh=qK{jXq4RcNvvdW${U)8hC9>clKcN^V)+mwzB6Q2F%2zhxOHoW~+rA zXWIy4c4FyHmpxZ&h%pO^QH3Lm{4DfUA9s@1hjfzl*iL6euIyjvD_(xh4lc8KF#Y)< zFbG%GuoueNOg{ZF&AWcjxoyiM!Rfh^Gwr*8E-xvjt>DYH^_Vfwu1>NjGNuPUslt0e z4Q1`@0o#Pn63OIK^w&a4w4j6z9=D*UvQ2!`##&l5Wf+~`HkB!Nk0q@^f^#p-h4i;R zqa+_SYPP=4&KoHiwNh{}+@#Qy^Sl-UN%(`0e)vQZoIidd4 z&&-4-)Tq+9oeu13h6=mz_#Az!*Jn@R2q{c)WOtpDS#*XRo%(3SB880ip^vIyep$r7 z-RCcy1L|mJcsW!~jukRzE%dYG8FYMELTg`tqH_^aVNOmVy%VZzX`e{i{m`EpevV<= z$2qf^gFE>P=k}0~k||dbu%7HvQb|<_xVG1;!OLy{>sltTM5a2k8!b*O!t^blH|8qr zu@lae&LYv)5#G#hwl|a2ilA>ZUUSoX3dygp6LYdZNZW6@F|YVq_#pHKR~s5J9QBum z{`RB%W0vfG8#Jin6L{|=Je7%0WK6m^rrwQnDc`Ao=Stw-Lm}tv>FU)o`p^QuR_P* z67J~KNSw5{FZrKM!0x0=EI!yB`(KHJhgE~vj)x;@hVghfw8;`|h3=H@i9W=q4@8Y2 z)%^Q2v*AHd0L+Qc08ZzXXm!V2?BCA`$B6dRm}`~XB#XPGsWlyc9(u}Ia30{c*oaEB zCW$P%%;DstLwwUU9nh|P3Wn9=@lcH|?(PkRRfdKveyJ}jI}->m>hHoS(O^iQTPSLr zQwS#tHOQf&iGLo{1?t7`IH$gGuqb5*7Pb1afeYktdCMsFu)81JF$_cPZ4%TFoZM>v z5^!Od1>jc&+?NN};9w&p#*&(n2WhyIcm{}j4QhrLJQ?;KC;ovViW8ydLXWg~^%%uk8*U%WSDyBin>8(&RqY!JVWowm^1P(!% zId@J!7e`D{rXb!59Nrw^6-n4LiVrdGjgGi*tTsL=8DuxHdV`|c{VB68!-UuyLx90;d>pu-I*0oTO!%++mJ}GLDG+^Qj9n8&q z3RC*eC2fzh{I~jBpty87{+F%I-#+CBvd1rS>QmZ4r)voQTz`xY{Fni5kwQN6jx5d{ zl+FL1FajPOy9iU%#!{N9KAt+{Mg1-3!1?Z-c<)rV$k9FmLngY?sos4MIUpgJy`j~$?# zB!OQi=@K%>Ryeiy1FpDc3v8Y>J^G`Ki#+bTOrN)v`qW&cF0}~mQA#GvUA}`?ymN!* zWS-?eK8+yXPs4;>$w-*}DhIkf3qjOS1DeqcFDRM9_9yY6T$TeTdZJ;abKdZ!6I-15m_Lv|n-5w1lHae?2|MboFkb(h$neMq-uv|q ze#5wEN`I6Da_ekx%=2pQY{W#;co)pw{w8GhN11?Swl6KV8Am>Ag*UOvdZ?}$NS7ZK zLf))JIO3qZc<;DBU}@jW?|r`+t>4~})H)BrMK88fqnrwDUp1DGd+o_Yi!14S&O-b= zt`t9Ayai>y|AE4ro5VG(#ihdYK5THE`}Sl4*1T0@E4CbiB!jiknlTjX?v>Dt_03>t zq79X!Y%s3*F3!HUMSzRvgE1-M6@M*Iyc&SRPiJD$_^IN~0Xwk!#2WOz^A%5KJw}tm z1H}Knt&Vn>TqJQ}bXH@phQXYC9c&6Kd_L?ym(`-XS^ncxAF5vaU*7EY=t0HxqM+!7N7 zn~Yw=pP#RB-1BU-C~e`|uF8lr1((f&6 z?8_2-v@9Qk!`2{$y@ieY?Zn%A1kX&?e7yOz8rolvhgm-lamyxLfuKhrpyK!d=G?l^ z9S^j|Fr{7C+La>oEeP}eT!-RhGqI&nHXgn#^aYRVikA<2gwm9Q=xO;Dxif1qZ{rz} z)|h2@+F=itG+w}-HEx(#8w3CCu)%49koVvUiLR0$SP>BlEzPS%_hVQ{o zKL&?5Z^hMDCZOtzvsn4|4^Fz?gkA%(@V2jk*t7K|w#a_Mx9J&tKd%LFdT9n$Mmj*< zLuJ}D>@)neI0R!at%97kT9{%!0&`3~u((W4e8tB^Y(RmSqO%A~&5vR5ZDSmwCXX); z?8cI;1z@v#IG%9wzzokR_;2bVOiz=+jT!e)oYG(XB}|GM+Kw1H{~Gk0aU2)<<-pCV zA$Tor3!Yv20={|A0io%NXYcGr>tmYuN%<>0a(f9KKr5k})6O)*r*qFkn}#hoL`Q&k{akqPK^B?Wdr<8zfw>opvExh}aPq4$%CQq) zE#C>(Q~Xi(?I~=3vWuIU90``EO0a#IC*%f<#L0WZaowoB7;1?a5}b`H5ejItM2hON z325tOjP;L-ur?$H&wKcQ|A3>AcWN>WJhw`4xi1IT^Fy$Ca5l8gJ_n^^U0{aeLR|CO z2Se(!FkyZG{K|d|E*Ac9^4NO1vi?1Ws~ddI9P)Hem2_YrJy54cgrP za*hvG#Wq(uL9Xx+wC&Tz3xVeazNZHC8UzaSG#YyjG~0qY_`w9!nd---6zVt?2*DlD?HT!L{qruxUmXrnWLX z+jbge&0mI-8shPalQ%BaPURFbf58yf*YG{w6O-g#VSUaJ@oJMw$wl5A`>ixX>*dzW zV|Xm~3%un#Zk8v0T9A#c?Mm2eVuNkYkr-0+6LP~mF;DP3CmuZovSS5Ltl$_~BQ{1O zmXG^~C*y7PY@FuzAB{bfgx&^6FgtoD4p%=2vQx@or_oegzNrLEf(GH#72EL0oCA2V z-zZdgqe6!)jQN~ip^!CyCS0<3fkRJsLdvj*oZ|_Bxw~yQ9#z_e4>hLZ`qnczeS;=e znQjGca2r-_+6te2n&5M&@V_o|aFb;ahAeLe?H_wlHcKBL3eWPHK`d6zyMrTUNm21i zBiz1umvb(?4E1+5qJ2#$7Tcym`Xh&x-}e_NpcJ1_jtrUbKITT9+fu*FLaJGs#U!(#XGmoQH83>+BnfZzAxGenQE z#b+nnardo*xUaVd3+>-vneTpF>eWLR69(byO=*(e-zIn`d@o!*djy~R{R5j-r@=F5 zC4SR9Bs`Os^vysPCkI`VaD!JupxI~~u~`@8J-@-f;=5qU`@*|%HMF`yXlZ*)7kbAx z;p?h!o=NoUXVU57*Y=v7s8`TCN;sx$eRW1TRkgn8McfEhPkurAUDm9-QQKPh78l0mqA!v!37K?nPKfzU&NBGtBhn=ySG(zn!1KN6i2n7VqLl6Wpgq?Z$A7O9JUf7vp#{)?iw;gZ zaGX3H)baJUbkGuVB!%a4@ahx5^Q)F)-9A-uhQ}U!y7RS=cUQ*iLx;n?FrGW&JQb^p zU09K&5B`kwg3cY&z^BlcX3SWJr~LvjZN*F2RCyVEw2Prq(+^dh&ZFj4S)@S%`y+ig zR^@+$H-9cc_24`j`70A&nJxjF1qqn7!W=)`h-8N2Rzt>)$vDt*Ic~fff%%#@;oEdC zwA=Xt6m-vWS^L!R+omkRUs;KN)|KN@-T}6I>WUXee1OPP^F?v$3i#)`uJ{I%@^?q< z#MufNaANFk^e)dvSNruiE%_Eooq}=Y^S3mq%$QPbb_>4x{!jqMwFjrEupIYSlHO&f zs87l>e#o?&e9KnheEQRsE|@5>z`beg?%ylC{(^P%bgehJ%Qw>MfNuVFxtI#pn6u1{ z<-FI^QN+b+@H5S-X|VZd7P{&HJK`Gx2g|;Q>Q^oRjmsMJc3v|Hqnh5$@1o1(%n1-v=+Fs#PjPV(`6RWmzAGLCt{<(Fd8cPUerTY`*X>55Jy>m zhSToII+(0u&qo=#aMwmf@Ie{ZVT0fbQoLc!e4U)bRTk^o^cXK zwi3GgpXJ<*^|{}z zre=PuspPsS_FgoFz&$?XS0s&?f0T1kzfQL`0?DOMDw!3y^3zFTTeBqMu2g%$-YxT`m0Yr)wo8tqEjRF`hLj-y)?1!9^K6mcqY$<|CVz z@Tps!Y4!W#l)A8z6X!B=IV%B!&Lm3DJHeN{>jaHKmqhQT4J&Ia66nSpn0zL7xSm#`m|huyzk;uv>w2&zL~CojMKI&V_=ZN|1hFF;%YL%x|2x z3|9QtBr*^_Z)9-GQTi@yu2TBk}g_)*{zHIjE5_d~MF?JP}HeZ&tQ6HWa+Ur5d! z8qL`pYNhbnNA%LPjvvDJA-Bc>@J>me=9xN?+JIJWf9qPvesxl`sjP`)9zUiHW>Hjc zYf6J#k8tUG4^pm&5%an^i%Kh$=w;jm`p|bh{X5&nzt}Vi7OofGA3qJPU3%P;R*h4l zwjpo$0?Q;J|878@>qz1ko;kl#0f zDM)dK0kBhJ(}#^PO0n0GR0AboK9sn8Grl_49$rj zF8v}nr1WTop(*{_JdU3=)DiS6;{+GVcUr&Cit>IYLU)%Z|8)Ko))*j#E6q3I{X=i= zp-w+;u;De5mRr-f#ph{7%vip@`vT>T*W;abPGq(ML-ydQ2h8%floG9igp5L8CVst| zjm&>epUIkr84RIY23|B*cRcO4HKJAp8}=~3jcPNDDQQy&-#KClZSQ))bajK6c#M$G zpBK#!E|sVD`D*OX{mpbgZK0ElI$tx)I=s?=i69@O+>ydk*IhfU4WaroWGSyjgS;{3dn%e5Xf6+b9 zv6R33@NvKBYI1^0=eh$-|KA9O?8*HQG zG2bZshXLEtxs{a9UZK^lcZnt!OWybU!FR@5P`k@|{=cjua+t7@c)U$B{D%KaIQ(MhMUoAl6cxk&tVpBK zr5vnX{f)Yg-X=@ui?lUL1ml)@uo)e9_^x0^(MX}a{x14 zG=&X1&S>T0oBYkyMZE30D}1KqZI|_02~>II1`Qv!h%YZb#UFGor$yO|h2M(_R40DO zhaIgUy;ILpjV^q0eAZ7ntLnG(J@+Ib;)LHeM9~}{mR}T)yey3!_@tO+oKCutX zT%?VLtz)1^#~nx5IN`0N5L|BPhMqq|K>TwIdX7~V_a4Zg&4**r`h5WgIc~)#BNt*g zUjj+VI%GXm2`jYj0mZ+7O+9%~QmcUBQQOgF=LsmwErNm9^f1-DoIA{Curt0%xcT3B zJh0CJBMw#Wt7?Mh%U+vkeFgY;ZudZn9~|;2q%d+HGoe0yW=gd1^iFXZp>{g$AakBI8&sB zwmbb$FKh~4Xqv{IN>svyU3P51NmU#X{RY)io`Q~UGc3QDBhfok1NWAC8n zQ+Eu*Y@zugH?R-7uUEo;ySy>=w+P#13`nJH0UjNCgsDxK%1%6sLQm~d2$r`M&O0m6 zNjIcMP{X0>h9+G#X31tRd#jq)RG5W~|{&PD|;hw`W^~gCG{yGti zO}FFUEID!bd=CEGXo;K6zrwPif%HP94o>#z;SvfpF?4NzY?;v?t*$v^fN3ANbYVQ6 z^WKR)4|m{?^nNh#QyTj9SYu?&0kn9hfo6-E;G)3guZ53hzk9Kaa-XI+~__JVQ(;Amdiq!pWQGy!w(Kjj)(m+`tbJd3wY?~gv%ck zpv7ZZXz9BVHRY4=NbEVZe3Xm_hB@YPy z(3iv6;kD)1XV_`npFRt>=gW#8Jy#T)`^BOWm;havAUTt*gLY4<(7i4h9p-4@j39mV zs;a;f`#(B6wCu*T5_3#kUIAvA7cg_H(1jgnfG)SsLLF}6a#rfYmPguH_qhv?s~135 z>s^Q~*#+M{oVh6x7eS(*j;pM)@$c+Cd`F}ve!nD#XEG6M1J_}-l@EsHJb=KaJ*e;K zf$N6XASiA|o9J-38}Nyr!xf;{g)cO%5#U<>4yd)C$Wk*Zp!2X4mYgbq@huM6X{;id z%2%P*$n!Y4KpEE=<>0XmQFLTZ9A?O!fy1u~aPrekToB#>m%0((?Y#&QSCwF=#Z>W7 zNj};OJ3kcmCyRvdOPX6E9-g!YPNZ+gSK5M8VQMkn6=e(k;FFkYQ31aUD?ldUFSvK= zQ1kbxc*s2v|Mh5#zdU)3+Fg4fCBg}%Avf{v`x3MiI%)aIzXU#CAwE*^0mqV57WX@l zU0PzzzWs;>gEJmDG9i)t1xLD)nh%q`YQ(Pe^=G*Ujm6AC4|}8YV5Ip!eq);v;{9Eu z=&mUJ;F3yEiV;X8bS#dXo;YQ|kry z^!EX5X@?3uj@?X#n%hbH*e6JT*#F*m$3yqBDrb{MRAyI1OABgWP=JC zXld|F-ekAHY&og{@z0D|XV_*uE%diOtJScJ(Q2&Fco@5TOA|ik*5Tgj8Q5p&Pk8cS z7pq@-k-Gb=$64yN?B?CKoPW72&Pq3DXP2y%7)EpeD3_wHCupw7`ihq^9@%xL|$~v!7Uu=Z0)0?^YkS%q)y~ zt?dJQqaVWWBt3}Kxy{x7Ucv+jFxY9nC%4~~{E4$a_#-Cc@T$ris=w`qvo}QIv-$g| zr|k(=pE?cBkMH8v2}&5~_zXKzoLST84Cr8O?0Qus+bFjLFNMlud)HucepE!84Bzut ze$SXvjPYgc+~bs*-5Uq9yMsz3(-vB@tD$RIx(r5k8}V1RY!wO!v4{GHk%-@xPq&LxQ(Y0@Ig*4hTVS0v&lNR*;|n!2V`Toniv1D zQw$@d1L1L034c`0ge*REP`(9^2HuR$BzBU^rWGuB{A3na6v{ro4`+Yk;$e2cDkgtS zlZkcq^4+EiY=z$p&F(6&ZqlIXh^yTeIZRzf3;&Vm{63 zlR?G%YT2rlpXu$%1FUGCA=!+JfhQ}DkbRRLs;ZX5SdT*by~di!uOP|vt=@Qg%RW+^ zWdP$ZSh6#*{=i$N0Yvu0a9j@k+Kzwz1FdZb>H`So#*GI!da0u`dKpn3Cwnt5169mMKAY0 zht6NDG4%3L#{ZxW&D6M1?T=!ar&o8=KTFThA8q4kbHOBf(nCGE_|G*)P`;gAy*nD> za&n+&UNWAV=tQ3&GYk=lrosjVX+z^WR5x{E%i8SW{ae15$PgDfxvB5kn zeAR`^SDnG9g_-mh*>&_bnd{8NMhE8a)X~aAdl%5Hx}E%w4$J8@124X@)im6(S{6eq zZu9N!8tK=k;%Q_00mtXfWxoWJu($eF(K%OrSdZikreJn9KTL^Z7kKW*je%bweYF!a ztYBC1_vjxeCZ>aCc*4Nvi8d>u9ph0&d>O=-SP6`XiQ(e=XNj8ID{Eabcs zpI-h&$(95BAlp6k1=aO9Gw>i?_v#r_wAq{Q;5>%g-{~>xc6*s4Tt20v)|g$CQ^Y0- z-orKC^|(zi5ThpLGLH`jpyK!kjJp1i>0jtXAB@lE*41+SBhO>95}VkitN2W~7KOUk zB*G=ddt13n7dX6eeS(M#kJ|Q%`aJ6NN+Z?M2A~@W{T0Y*A%){REXX+ zzJ*yWxQ%_1d;`@y57YC%wedA>H=(Vh1a4e^lXi-*r57u1XU&_k(C|kEm2EYj^$Ai& zk|sbF6<6YQ$@#R7Suz$IG5iG$+_{Tm;JDuq!dHeC9Fs_lwzRDQsc1d?E!2mUwjpX( ztI_Yq>(I0!4t>tVa-0ioMpnLq`M0x|N!WH58=h{Y7A*H-U6*ImPJL0RvpbyGu~82{ z-*KjMhI&wI-ZWH8Qln456!a{ z_ovV4t!2e|`eUj_sWK52@XqFlG02^jJEd8TJlg`o417{Z9sUg66On zj)pV7^X@Txw|dmRxSeMD;xW@G1dpE7VJbB3=-u*;?1)b+t>+L(zg=g@_Q|echDx`g zap@*j{=qkTvg|ktPfBE>j^=V)a24jWU_WcUsT#K_X0d65D?nc6B!+OE49l};RIGk3 zV!l)|fxfbEMwR{Dbb6z7DLZCl!nkgmZSHWF(|A5avF8I#u+`=SEfF|}IdxHxwribB zvvuLD!+mGWoi)i^+DrnoLPmQ|^- zpxNqZrbsx6Z@yZT*_5cu^e*mXyI6hPSn;r8+@9rpb!*t*nfI|{;W_$PVl1m#{R`&G z&ZCEYM5}IZjlzn897caL=gpCcLt~q4{x?G%I(tU~y`Raz0nr!C-PJ%J(o`ZGZhII}&#XriVE4vs<-L&R{>g@&nC2sY_ti=rQ9q;puz`Pg$r!U}=UjSVij=usb|l?;t%?8R;ab*ruQ!`*yNanY1zK=T zCfmRLAann3IJ-x08Owh#%FOCBW$y>9q92~KqpM`T&`!Q@D>v#W(71Yn@zT|y+a`;! z)15TT)l_2nLo?LbwdpgsbI>$2sqCi5Kiq&9f#G~N=_s~wqbVH`u^N3mR&o1)e^@#4 z7^~-TePCH~Y@6FN)_=n_ruxoxzO;}nzWH;C|GpxWowqWAeYPx!{hh&eb38jn({uOI z5*GRVG^1shFb?CLb{VFw64?utVB9?blliomwvma~jK-fDE)^O=OM(Fr?nr|_O zP02XLiVTFYYTAz&Ww9oBA^U=zI;ue*pUg4l#S+;o96Rs&(srhF$seYuY7+lNWe+1) zTgqxHtir^m2K0OrMVnhc1^pYAw6v2ED1B*UA1}Pl-Z~S(HU#kKWN4%tG~VOG$WqLf zs)UjE;^tcV4{6)UTn3Q-#uj(Ir?>C#VB$GO@?!O+Z1o$CZ9SXIp(S&ikd(P)vbVU| z=lf#D*T|Hcm2}}PgQraE>S?UorV+MCUxpPE7N`n+D_P|y#V`*pY-Otq7-mt)36}SS z>%IJGjH{MfvT~N~tpD-t*st-9{biEC?AyxqqAssx6lWq!oSiD~&5UAC8E3MqQastU zYo0Na(=(Xy!geg3_JoysuY;oMmF#SJ0Tz$XX2*=X_^!UObmy5;ydnLTRu^z%<;LXL ztGY3j&&1}EvhM=ycy1``HR&aDA=U|(hl;TZrtjDa?mw23c#GaP{~#(2eZxbJGwId# zE9v1$rmXL}NOr7v06}g3th`tTUk;1=sX3s-PxF(QTvV`SKNb+Zc^;Y zGCu385l3gMoMdBNQ`wGwH@faY9;Sz>qGR4dW^BPXX3mNjc8ar5)d8nUzJA&i{KC7# zKI?eQ1d5F^g_bmR$UK5CQqE}xv!mJVdv7yS9|f^@Q?D@~X3o~lt7Y%|eTT>9gXsF} zAEVKr$J&0cWLInus%nx8XX|J7vwt}j^u(bs*5Sb^R!15GlO~T;iP1YH?3#$?!-TKX4VtyNvA2ed}cj@zY&e+FdkrsP#b~OL$lqGCobr6%0ILzEA{>eJ6XoTci2exJ2 zPxM_{!fq}MV=cT>n1hbj*^rCr%z<;OadBD*tMXKto%O4OQ7LLHEM?E~<*OQc*Riqw+`9jF zh^dZ=WX^`^u{zQV(5^Ly2_H}9kL+JbVlPaqvK0Kup2<+G`mGVpItRBfNmE6beJ5tI zdUtGCWBz8=>v=5ev?>{!jxAx9PmN?X{c0JR0T)&`>>0bPr;1IFS&gITqquDEQP$5X zlf9g`fKf0IqkTWCR*r-k;AXd(?7|Wy=8{ANyUAA@XC2dJ7tAuI9h)!Hvv2vs618&H zZ!`pS%*>hQ3*Kz7o*eu7^GRlz@k)M9-#GJRUo7kJOohF(MwL~oQ)36|ai*}DW_RjN zV_Bsg%su04jNg>$tOw>Z`bSQY+TrQ6if1sBWE#k(Y89|{@~W_?cL#Igfh_y&KR&x? zWC5%7;4-76xrAkc&ayop0@%Cf&oQ1`7qDiso7sJzRavuuHvUY(lNfL)7wZrYQMxC8&no z#@q{&X|q#@U|?63i`hQ+6GGEbGRF&DuW<^2rr>qQ2)2MPw6E`ufy2zHT=(KFEi(ZCQ}!aOmDZ-rf(gnf`l8YbpGHq zCaNWq>3G?UiVtQ(oVzQdyJZHxWn}3}fd}~8BpnO229TvVdAudz&YN8jvr`%Tj{6y-41`D zeYH6rG>&0x-GyMeUL3Qz!yN_M6)N}_exhw!f-CF*fhI3@D%U#Ak?MmgOi!?2^W`Y?PDo62!^^9b8DHEIT zPfyPugk}F$u~qk%(xHpJFy@>h?Z#e!hS}omk>ZV*;qK48Sa}{5-nufA4ZQGS|4(MM zVjPKm^#o1{{^FN)dm!#T3)P=);V;9*_;A)7#!|hQzjM<(X2xx6(p5#`SlX z@qRZU-J`h*WmLh78ugIoKXl5$2H@7&^~-BXkK;445?Slyu7vem(y5} zCv<>)>>|wB-oP)~y%w#kcJjWo%wc0JmM{*nYawB+KJ!|~x^n-#+xXgD2{-is&i|;u zIH&|M;f8@gS8t+xhc5ELI3NTUQ5V%M?*rJwk-r`Ci~;TK7?;pSQN;eYGt&1>aYb&nDJET76SKAY*OWu@5jP#f9gV_1}4j*)M& zAlTd$&%LQaljoDM;)^h@v=HvSzg|iL{|R{R-{Q6N z@V2>IJr8=$|eMyXM47s}HGP(a)54H`bg84nb5UvMH>`M;T%?U6yk>#5$tk^~N zJ-tB&{nvtnQY8q+g@EesZ}7rI7#|ruA!GJ4$ilNpB>d+|P&=3n>LSVDQI-qq9*UvQ z0t-BysspVg6P&URK*y$WSh{B>R`d^pd&Vbf#g3aqP2U$ZWuB9@JWn_u(+t*Gzo^#! z3NmZ4CA<-POHCNKli7y4u)f2B7iFG8RhDXzg{_@rgGxIo5$Yrkg=&zrA(r^RdQW}% zE&`^(uR$cx2?O^S!*zB6TzqItCGA&+{qkkNixVwl`dK2Aj6ErmUXcfs<_G@O%U zi=UilW5vX5V!Y!I8SdNyekS*!#MKPDN9}R?$XUoL5Xa{Ufd9Rn0ng3X!sfnuXnJ;r ze4n9&GQA>nkMVM@-)TM;Rt>|yJ^{G9XMhAtYQdW8H84?bj|$wnNH^s`&HOPaUulbb zFJwZeOB1N+4?*bmES_4xAUXc~4tS=GLhRcXm}1%s%Q91;{ahX_cy=4M8@(b<@}lJO z`XORgc?NA&kKl}YZG1ZU9Q>HG9ViWLxOgZ5-~Mt$FB2{FePxAgsxGcEQN#X+i%8wg z67sZoJ(y@7#Y(dfv^h@Uqa7>AffjGdK0JbayfzQM#tD#9ZRx~BU^Vu(?ZPe7lJR?# z2C2-Efo*g-d15y{1N#;Bd#@~A;;Lp_8VETRrp4ZXF z`*VH+f3Y^INc@1PizmU$Komv{Hp1@O+b|_A7XH<)0=KO{Bl2jb9N1?$fZ;{~IApRD zPJX{a_R4G_5xWz}qAeF-nEefPChx&5uapXpx2KY;Bgi(H@5CTGib{V|Oh#m{L&20s zuRq15b$5cXj z(B$CSpCsYt1d+PTlJOIhus`q;q>GtB@2<7bnovx{3j3-1f5fmmrvPlPZh}8GV%YK` z6*^p7NRX*B1dG~&tDq`e{+bUN?T)bBZxZw!cA~C}uA}z)&ZpFa?vhanX_(!j1ZMmb zn0s&q5;YSXI57l!J_zGzUmqCBsbi{s7Ht2l4I=b2_%Tof_XDEAf2kV8?k*x+cAE+) zSPO&A31sB{6>8&%B1F;OsC7E&yk+g86|gMo=~`Vr1o8LmX}6VRxlE9!>g4O%}Svvy8pMn{)La*?rapp2uH? zh@C4y__`^C*_gowM|E)N<`Zpx09orC0-v`&2mhTkNGUJoJ+FCWdetYF=cQr;0`d!? zIbtzTE9YYRe^NLvx*9&kb`tfYDn#!|4YhuWF;0&GYRQ3mbc%wN$teQH=Q2cgMp{-1+--#3axsrOY5UJ(7b*U{$%~(vImd_cwtJiK+(kmo(u32X ziPk?uY_$`}Y%yc9a4>|#J@JG3pL*~`zKXau zd6KRplH{dK9yMZfg`&1BplUC-@V0S}BRAU_n8Y6kIhO&Lc-RW!S0CZB4jbI(rhtX^ zZ(vPnA_O|w!R48u#CL24S`CF_^4E)8Zu=BBucXmmM-eB!TVqZQjc-MNfa^&Yh)`uFx;!- zKxmc8L@V$FImYeH=Wcik=kpgq@nSnVp=dJQuGx(5AC$xVJ0iGuni1-H%HRP>PNy;5 zivFsZV0-8`*x-75ub_tikR%D><~Z3H&FUVyd*%^32l5LZ9;!#{?r(KGQ8 zr2K3ndMDd4`eG#B{)Jd=6@u?KW^ioO)A)1JUWkwJhbKj`u)$^&OlJzA_!0?x@Ff?; zwq$VH=MHq*SA?(2uHYf_6Ij$e3Y+if;I&V4kxBA{oeCAOCRqrlE;GPojT$H;{{zaE z96)*aFBP4YNv5Vn!kI7vRE9d7yzvnvqDH|stsFca%HbZ9lh7}HjNJb*K|;1Jf#+-q z+}p_2BY!L=bwASJ<7p8T_ih8CH65H^OXEMwyO6&l0Z!MiK`BouRFoctXrT%C)h~$B zL-$}I3B{t?&1loR9C!AFfV`L%RG%FsL2HD;`QRQ9*mMrQ9#F(2qZ^<#-RnCbWthXJ)wb4e`cWK@zrE~k$X-+WUyA9RUcB87r7F0<9ZaTQ zhHuS*AT#wgxo-TU_9g^Z`wCE>1_;;JUHxUKLk3D4N zz;#&lHx7)8?ZAsO;2vu^O`ZqLBA=(`z@g>_knej1S6}hT0sdyLbMGRQbG^{_)!b!_DxV^ElY6PC}>PDLB>n zIF;sggo>%3Ptsl);-&TfV9JHdFvJ@q-nsx)CecvqcOP~xT8d}fgK^L7E%-MN@QDvM zXWz~BHp^4E&{_@M=2r1$JvZb1lZc~|Ry0B1{!sYbyasLu&cX__e;_uGg}B>%V%j1E z+l*#`Sm!pZeX$88Yz^>6+Th*TU^`SjpGtuBYCfSr~ zOfPvavlGORq~VBE1g_omAD;Cnr&2b$QnHyrRO_uwvh&d-_%>h#y<+8f;d?5!J4T>J z{by40=RAD&5<@9%MPyG{;tQTRLcB8ml2FC^1WBB!beFunDGb~8&wzg(g>d6W28`?v zfckJpd?nNZnPQ@Fzx^E9=Wb5DzQ;HF+U^6IfA*8|vH1PFIh{5{Ac5OXFDWX)(Eom(BqjF4v<0c*X^K`Iqv;7;b&z1rB%H0 z12)urokS8Mun-^5K7@rAtmVXEED8R&>u~x(I`MPuC3%(`$*=lYO786td3Z{R+^mzK@^_?h*`80(=a3B<6`C-Z&;;H0w!{72HnLN= zoI9KOLG`W)lG(Zuwh0%KX<>d;w{R2$E;53|s;i{*XbK5mFGlrrzc!n;L>_8{Q^{{x zaq{;2MwotE0PK}VDW&DF$(0Z((tEFrcl_&bu3g*ij#j)xfB{8g|L7!9PYNc(nH$_B{WN?JM{YwyhIPH*x&C{Caru@DIej zvd03iUf6xe6W6-s<8XIAI<$}CRuu(0e|k8Sr%_e zO7BjQqpK>3?+#J=iTgF|;`B?KH8CJguLpi_2*`EcLFJbmPyg;}G`Ukp+)MMQK>HY~ zmOB?~=&gedI*rix>JJ|H*NGw8{V4M{4G!E^M^6K;yTp7Ad`Pf`YHn}muX!3(b4!rF z>pp6G_ru$j+UV7-gcmRUg53+wfs@B7NZfJ^4R$?-Gxpz!?ecSA)bpJDyO>C%&DP<9 zj%*k*aE6CJ<56{NBTh9}!9`QWP%lIdu5d#$~o9cxSd5*8lPc9|KM& zFb;vaAuJ?RyF-9^DHLzphA_tx3!2oq*{}f45tqmPe^#M{#WdtE;^85OAav+UKo5;d zn0)RlCDgtF1-JBr(Rn*~waXbwOHDywb|o3Odm8c=6_KrKV&wAW3vlG@71+A^1{j3j zgqVvd@Z4SwW@#@XV`3rXK!q$=&C-Hem1YuH6GwQz{lUFD0@8K&z|^sDvggB1>fA*U z-g2c5s2fNHheaD8Vo4!yFPx*|vR4q*7c7bMLU^D&L7pF0htk}`Tqe04^14f@i9Q#q zTj33rb?rQc@nbRl!GE}`I)z+M86sWRj=(RM$7HXoHK8=lP?rurBMv82AZ$x4=+(%< zi*+?*&Tt8}%t#KruG)a#Uv=2wI1@#>G;tA`gc}>w$i5MVI-izEHHwTviu5>?Cv#fI z6G50MDg6y9(4;0dzy%+6a~_G39$5T5bwhD zG2U=%24y?aL-omZ~!5q z4?|#+=V@?!k_q2V-XzNJODUBzQ%J?aR&X2P!+$^GLF$P(=45SvL)Dt(#+OVuBybW0 z^n$_KG7^ql3We@u4-hj72DuIcs4~@pG23pC?5c)q?#N;O+H>_}(laN#lLQxvd>=@@rTd+6i($@}Nc!;GnB4IBXw)tejDBD`YVL)J_*R4Wzw_Xld4=enJx}_}u5p@DHOyYIi%Vn92gRG6Fun)6oZg&mWWi{JG@ybXgKUy8!HjIL>8-5*QX;LksuX$+CgL-L zqd0RWjoW=iu#vwLq9S5Q(63F{_p%j|K1)GSvmkmN?}v`7O>ijV1(bqnfkjU4KySiz$qQPQ}h1fuX53eQcCvhU!k_Vg9jX~2tg_w%vgHf0W>hvpd-H$3b zEHVsx#;(Eblt@r0NQHA#GT_o@NxJRi7yM{djw=!eLCr@9e{>DQnAJ9DUm6J#iksm~ z@?qHcDS`|SrBU9G5r#t9Mk<}}Z6*+~cbt3Up#5#P$`r>!L71**< z9rumWXg}Kle_dXNA~lOi{fHB9o88OC@gSai68oQfVSp1X;5`1T5K|I-sy^EIVLDMFAV2#p7wo{|G;hQ zXmTwkpA^X5B#W%=aM!3928%xj*0GARdl*Z_Sl*)o8d>7{<^@rV`9%au-{Pqy-FPCu z8P&XFL2hy-eE--8Y0_J9f4K@CU0n@Tk!v_z^9(HhkMnz9o&j~*qF{FC66xMI1YKqR z@K}5j*I&L58xPLKe0MS28Z}5g@;=BDa2w)Hy?-3$elH>oS4*hP*WEB@WCQA3I&;q# z6->+SfxzHm2t4x-s#6}rZ?#Mi2=0S^n};B`{3u*_EsIjC`{5y*3vJ^F5@`jH*kFvF z=7E%cFiTlU{-&}QrQ(7Y)#yE1gaMXAl)3sd-lc{Hv#yCuNcR5@a&DS9aq>27n{*f6 z{Jj81UBkpUdM%l^Y#&kH&1vo8$|#>Hg~Rv;Xe)P<+!%#7JKYlz+T} z^p|4zW}O6nsT>BIhU@Upg@&guG%#s@II6i9;5CwnNnP%^V@MV6j5**<4NVNX{t0gR zog$ZPR+5F>GwX$W5CnN7!oPW;uvWL0H+nsecgil1QgUh`Df0bfgUK*CJNl2jrJTsQ zXToHtLmEy$FoES(j_{@UIEjvTB8|_^kq45&hAu*WQLUsInKrIc@qI9!%QZADtDX+J)sJ~W5 z}(A!8d>Dx z#sHFhy@qg^T+nDb2Insy1}WEe>e^xt>Py2iYE_dnmw%wZZ_W%bYY`$}FZ+_g@LaN4 zK@Fyw^1$jc52DJvNr>q@lDzX6={@8^L9vC1M;tE81WIf5e z*-U=g_>jm4NyLZkB72&x$lzpcVwty^?BqF<`LzYaU89Zcm$U<`>iLj&+yG8*+d)1R zN|BSM+^~!XyyzR%Lne_C@lK*muEC#-QCKip z3cH_h+JoCFq8}VW7CV;0)Zs)Jmc0l%KdOmW@E%e+U68ad(Za3bd?+j54mnH$9C&XD z1!t##^%@Om7%3n(zn2r4Kc?7Rpp0A9RM7itH9Ux037xvH$f`f@L8LAptXH3gPD?W? zQ}+ZF{qZ|>u5J%${H;lj4;zs_Q8^s`H3eh)Y|+=91^wj*V2zCyNX?mob<5YmJL?$I z$2BGBT=IZ><_IVoLEYrf5Ekm%oMGjZxEJL@5 zGRoWgDe-#j0mgB?BsgO^H-CKxF1vDw{h18v@YX?UuUR$K_``r$AC?BU+eI+HEE{e- zi-MtgM^ab#j`D{)l=0b8)al9-)WG-yj#q61y2?Tvr#=fr%S0f)BaY%w$vFRb zBG@(VC3eG4!DU4PteC$E@*ouUm{d{{`;>WOdum~`wh&0f%p|)O%)mV#OQ29ZmRPuU z!-zl(G?yqrglv!L%-iW^i4oUK$ERN+&!fF5r4xy~GP7Y~Z+3xtCj5=pZ?~SyVK#x= zqHxm5gknfL3)G~t-rXOIWAkwT_#ENly7i6} z?w27U_hZNjg&9OvWG~sIu%24KJeC(TSDH7c@CCs9Tf|tslX6zL0%4nT$=*mWYL{9m zNHkW%()v!gH2p9(JPgI0zRg%8bs7wrOz_OigsIC@K(78297#S05v~Vee9kFgOlzU5 zI|{AxvT;c6Chi)O#m|q7a8b88=HHowQ^kG(e|`Z>7OaPHk#?}ul1KeJ|6%oswJ5bU z1aH;4KvlN~oEusUUrz2MYiwDfc5Q^n2#k=7ns_2ydX#+KEdi1dqTp)rmYj{gPNq+v z1PXoT;3?)yOy@5n@2Lw!*E|+P7u3M@CyzmJvLf+scc5z84XFerW9TTOVBdj7@W}cI z?&>^?)b}Ekdut8u#e1M!^a_m2#E}FOHzL}(j|}~kz`bR2FgR%kx~>?9Y5kX=>-;>p zV$?|18=ogP%tgRxixj-5w}2$815hTM15bL7K}xP3@Yov=x$-K+dq3ei>FNo4SQ1=& zR)VO6FK_FYSrkXxq_Qn$k`AXuM7}VV=!nii4~wNZ!z>u>qdSOp*DUD&kP0C-Zg4jH zI0)a#hVlL-ct3dp4)-t+d2R{@Y0`K$E(EhpcEQbkp)i*UgS_{WxcrbRhR`-R>X`xN zFE4`Cg*vGCmd{o6q=8sz8rZDo&U3cYKqhe|M3oZQ;V~Jmt@41YbOj`D$cDklEU3sx z2HEUfSaz`yekuuL)~*3iv}@(AF%`@Lydbph4k#w9;QdVB%KK(*Lv@uuA;<0{lD*G& zkqwjc$kG@g=vurVre|!$mg8JUtV|>58$E(+Wumz9%oI$V*GAezDfo5e5Y%r>pkf`2 z;X!B*1m$}{TVM#7d=CNFmqMiB*gcZix(uu}*2DX%2>3K{3J$&tMX$rbI3TBum)>~5 zKk6RTe@lej52XlIQA~ySTp{weHPIc8V2NfTo?7;SG*^a$v3n&%{+ffap&^J{PoY4p zJg#fELKSx>ERt1)7_rmjn>wc#T=-7rw|c^~y2o&1&N(vW*fR1``v>K8SA(RxTqnQR zPKRm!(YSwo6{dRDVQKUgs`1=A%J+*R*K5v$K&v*8-8TcnmTJM5Y+dkmRe_Dmq&c0f z3+h`Y;Yn2ud_5?KQk*~KQ)Lc(mls6YKRTH8Z3T2Z+zl3@C!zWI1q|lJI2#e@+OYyIgr0_&N#~*J z))bTpFv6$O(~O+TMQ*R#wl3SK3D*WoaeMJX1*Dbhyd?2Y*P9N!=EdFI`acIDNI4?ayY=%J#cv76pSl!zS`Ht_%b{N69)If z;WMY8xHtoLhHC&X&=XYq!{MdGYuG(>0k+Ohg$oZ0Amnoz*j1f`y2;N;>}*-!t1JOY zRaNlFPlaZYI*_5Cfb^~o*t_x#w48MX-N&I|&;90Qc3DF8dm#{4DkNen(&5OQV{rMm z3qWKs%xx_MiPCha=Cq>Nt(x$lF0Tnl>70ax zkA`7#n+6_Qsg4(nxzE`76j>s$}=C1ews)F=+h z%`))m({Hlb;xuYU$KaKcQ#hdFPQLt`0{({_!8W8AjJI5dV8b|=yloTc_65O~$!B26 z%r&6$ay~pcE)MHH1b|2JQt)#U2gmn{c$qg2<|e)Hd}S~Q{#pt({5jzNRs$FJj)A_? zQ<$>z7i@X`8&>N7=DN77A+T^U7(K9s`Hm@|BXSygK8FGS;SDHsDuK`WH84p%7an93 z!`bp1P;Pn~DlgYUF4q}!PkkLoD9nabcPWq=&VVyZ9>WKpcMy1>l{_xXBH?cO#BIg| zx96S>>(X4|k?Bmxt)C5CB`>t?)P!jrQ=rIO6%Oq-1r4*MAi?dgqWr_i&k-IG(wjq$ zq#41pBprx7G!K6N2!Tn8XJA!m24ulCaC?^pTP0(mb5#l)ajAirj%M&)Dv56m1@S+l zpK$*8dHD7095{NWzJIIen=$z zZj-NlF(lynF(~QZip$45p#9P$Oz~TVUv1x!+q?dfyn-^)-5yQ}510`vz7Eh&eTd&# z8MyXlI%Gvh@GR~Sl5?sNKzcb`4oJtAVVXWSV;bpfdO?}ED#6d8J5<1CKgvdR8^Ks< z-1jI4J9iGCPn#hc4mNONY!bK~d_j(JT-}=wo$T~sR&4(X(L+Q?PmS*hj{0$jfvXAEzoyh7OcCFLJo9sYi_`C zFyLklCRa@<=b=yJb?*ZhnIVR8B3kIM%obP9y^f0l5s%!s0z~{B)FoJ>>Fivbd-DkA zn~+7JMicaWdIMGqJ92Emd^{*(&y(z#MLe|5!qL~9w{!Rtywr@u`o;I){*1eDmwp94 z&Kn^;u@1Z*aehgfJ2#vS2k+Ktcs@y%Z1`aU<8!Cr`cVnAe0CBCnEQC<_6yME_Rg95 zr!cBV5&D)ogO<&Cc(K(F{&VPsy(Lw!N4vc$#HEXNKFif4|76s)zChib=DRIw_b$#UcS(Hg8P4p9DsjM3}Jb% z3|x?}Bfaj|NR9qCa&wm%1ncqO-@0%J4`xv`wiN3|N>R>U5=)k}fv<51R9dJ*ZOb>} znI!_X)k`po7l_9#((!J+JmxKX0}4-5!8lt1-0iNDPOUiNr>sOi%Z5_1-0Y*mDH-I$ zPs5WXc@U#p3eVT+BDsX z+q=Z&?ORhZ&#(!By%pJEPc>$he8U>QxG%&>Qs{zR=wrjww^G!eI0{ zJgz>1?cSv*7XH#fyA*<3H;%b0#m)I5m$c{<`Pjyi$ZnC6S^^e0Z^DxCi(u+>5c?v= zVbJYTQlSiN?To}BF@JPKHXLuhDq(-eS~}6wgpd3D#k8z=*tMJ&v+sSNxWbBz#QJiO zycJqIDyV&45@c1*LhD2-8Xg{mvA+z$Kh%)4lrwh>+0MEacAO?|vKrLS7HiKN;dOmG z6*O1TB*O$c@asHoXNTaLkczPpJ!x8-0u}7|Ak4OO!l%$3SbTImCJfc!Q7;GZH~(yW z_XH|8jlmGAq<>{Sc*v{YxVo%APn=ys{+2f6H2s&bAvPLPp*?Y6)f1sonu`j@(?zsL|&r9Bl&; zs1u0sPZy9!%>#;t9J)KDW4`WAgtWh;@~tybwtFnn2fw0<>GSEfhcU@}4X2BeIG9Aa z;AUGT6-_%x1^3PA`6Mk`-)jUqFF&HRTPKCbim}3*)i)#_YgbC%ojC;s$L&b;(Z!gh zCA7YF11+k%CIm!O!s%=!?qAEqh^`edlPjV7{ZvV|eG_)?^M~cZ;n;KgIEsA3@pZ#_ zNzq4|Nxj=&I#jDkX1P5Ozc>NjJNk(1mn}?o?10Sfk&>#2PHKJYiGh9r?3-54 z3g&M`hntL$^r};+tT{`ccUVcbt#T8tUW*|G+h=(EvnMYd`~km?bfaaz%7ps68nooT zCn-N|q*QHPbkmOzG{XCXM?At^0}u52qt5Nt)7dw>0nMJf_@YMuJ7$&9qcy5Hx7%I3 zw~wc!F9YyukPj+zLL?5SQ-x}+hxAb8w&1;@kfv-8f?RHQ7}lOZ-E3L02Jg@11;(uA zIfh@g9}$YsCV5peQld~%iu3OyaX4r(w!9n-iOgFnee5g9K0l7~j#@Y@e-7IL=Y$+d zJ{3+M1=~T>(5u@EJT;MK)vZNjb59eBMyIi^?xSEnVmX!H*-Zm)MbQ#hA4G-xg_5Td zMuuF0&j=|FloVp=>PB?*C`XKo73&%~aq7DXyhP3mhoc7Yo1La?wk-=rtsjwlq#f@c zmg4bkW!8H=o&PzX66WffPXUG@&hOuIvA|6F{CzJ0Dl8a?pJl5?UOxsgoQ=-hrp6=nI z`Y4Wzt%dT8YWT0ef<4w^&ilbNo+ca5?cHS1&`Sl~w;G{k;UONp(Sj#OsqmA@Ega?D zz*TN(+@h!j&CEu+rfo%@SI+UWq(|&!P|cs$0S2$@X};!9q3Y>tL|@B7s6!2k=Z-MV zl?*riTHoDto6<>|*Rmd$bMmpiPb15PK#=( zZt-?*>phvfivFKVdCQZG%UI@l65A}8#?>aeJb!8$UiY}lCFk9^JI}Jmc0zw)t@x8hZ_RX=@-aaxv#!gVwTQjyumN^2O?t z<4`m4D-}eY7MeDg3;J&|1^MnZXkTv4zRO}bV|E>R_Vu9z(c^k@;RE#T-Irq`OxRnT zm;2f;#KNP`khnyT)q*FAE|;$u^I;l#c%4UpOC|WL=*r!cM0ft45X!ZDaav&>;%!ZF zAV^i*!M=#)H0JQR3PT?Lr4OIwm(Wx1!?s#aAk}pjseL2D$VjkE!fmhoA1r_!BoQ`tXbi~Z~V--w){4- z_e%4Mb@DvNdq9D<8-ElSG{;8@%OT<9j&_~Ksl6YKp@`0WGaV#QwfrIXyE@{l*L z-N38tg4yd!CMT)Ypn1LlxO;&Bk=@!QUxf0)+%V?z=sl~ zeoNzgaX&=+={2UQ3NPU^^dudNjH6)5H@bT3B!$k|h@GjU`L^0Zex9Pwg_qak{UjCS zb_#gBZ!yk?6@x|Y{_L<#R5)S=6q|;#{N{F8Rrzt?(&HSl?FSzJ4dRH}Z2na~5?hn! z(Bdn4<4Ov#$PuK;+9|idDi|7d^=N? zTgK<3xXujchljx0SqeWiyM&V4F>oDsk!DxfAS>}4G=4up=EhW993FuWy>p@5Ckl!W ze$ZjLXnMN;Ec!hH`bmu zgl#7zA^3nI5BGB7jVA`M<_}MvV{w$NO6G9IB^iFUAqnOYU-3w8Dm#|$<4MsWn7aQR zWUdb4&fD2|k$W1o6)WIcx)oAk>nYtYfYQ9jaFX5-uBvE2z~(ku{7cN4jZY>&u?7|; zGprS?!+(%T<9>S5{4$jF&^qc!w{kj~~mEuG?|#w_kKxb_h)k@sZeRl*6R-GM=@V zVAGa7PR)(y#)qE#RuH7n`?`Mr60+}Z!hBZ$wS`b ztQnqsKQQY916w`19f~Z5c?SEepp{}8VkJ@P3?!tnZEUS$eKZ#oaHf` zD?5K-V74L7i{HMZ%JOV8)@ zj(MIy>6cinUmhVSb*0kDP#sK6*TbPMV`Ru#Bd(hQw0uQ2^)RFOYXd4A<@kw}GOGkV z!p_!#+T`V$%=5%mN5Y?vMCEc};5iMpy0>lh_XZ8!|3=q$_ zYmd<8TtH@f*U{0v7J{BQ58PLpAlPiUK*gFfQ7iU|V@ne-6D8QK{1EDTVMwnP_chq% z;nuHGEO;M7z18}PeM1u*tM{fAFH|UAC>5scKT8o_OE9#q4#8{1ETW4i)~#%VMfd%z zy?Gy-d2Zz8E^qNrwE@off|xTB&Yr%Uqd)E9f>lf5&4)2qWIW#5y%O0`71owm@XNaS zv^TdmCOo&t>y^f!UD?n`e~n)TZPXd1jRvVX*t)Bn?p{wO=li~7n)QP|W^X~dZ#P z`?+OICw`Km#>zUp|8s-O2a7C>trFYcY@U|lesC?{i?%H)&>S^K z$k@I|a#GigGM5CRdCeg9$QooiRx6FF*ZvhU_g|3QZPldS6TNBko2l4RVMgzcjNsp_ zW12E(5WPx#Oowhc;L3psSfW>mBBOYWtM-??k&@7kQGxi=bV!(8+8aOic=5XhRyjC%wn)7p@&o$yc zK2HppS^!(GCe(Zg5l&BS7DkvS3khc1gyON*+2t&}s?|W{&n)WrIGe;1iE5AOLD9xjJS)Y&x0e^; z?>rjrFZLPZR&bBmzeP1;hT#5KmL@$^!RYY0G_+owwnmJl^xr}7^dFDo%S8TnVijqp z#8JNrB?>yKfev{)6m3|A4Q4aAV?=+B+0%jAnLg+}*#z4n4e{YsBhvO~;!l+q-i%db z9#Dn2wLx$W3KE8Q&Jx(}rx2&7j_cB6aIZoYz4d!=;DMj;-T4+1>Ly_H$R||%=#%8I zTr1_x9EW>bWMCd<0P}v+kn+@pTPD&J^;orA1dfxhtoX0CeSGS$wf?eqD4`b@~+nF~{w&jG;lXz>QqcCHM zy2Pit#`xuu9|+y|1J*N|amJ+?C;xWmJ+6iLtEPxI!KY~W(P^X@cbKkZ2y{o%K4uqrTS9k{1;^=}z1l|0NJne&+ z`=W#{47k86Ub%DDCMmWmy^q%k$q0BQ=8*SUaLH^{4itNXRu<;0xa(+AEEq}~+ zxC65xcd$sIjGs&2XJPzlHe0xyh0x^r z*K)D<@2q0VXZo5-+QN9u8#$CqHDIUy2A;HkCsvl+!TBs%?rV4e2QQ{VakL?O_4$S9 z*EbL^Doa0H-@^;)kBEKxS-f$i$V7;l;spiI@Im%=^u7gq|li=C8 z8FBGHX}`%UVfVX#l((i;NRs+1QTzN!tXFz+p;ZAiedYOdx+NCf{~_!yK0%X8tA(aJ z9kj0_1MdG>W6bvHlurtLKVtybYXsro?mBww*-p0(`O(MK6S%9RJ4fh=yz7rViYOgQ zDJ$pD;YTj)ty9VqeE#tC=0ed4(U+s_BpkeF2cH^$0t*#_VE04b^y`{1Q@I* z6Wf%KQ9?GX3)^taDj7R79^@$PYZcOL=Sz}EijO{LL zG20Kt%R}g@PAl9E%z0MuRW><$p1XCf;-rU;{9^VCtZ5eSvLoj+S1VJ6Yy_PbV&R?? zDohqmSdUxS=RWDBHw*WsMEQEY&U%t9E8^gz> zBJN%_&z%v=w4HgSE6L%zLb)yaQ2ie!P_9uJts1 zpK*}CkrKOzIS8{fRkmEYi7##N<{dZHSaQ#je~goG$=Zi#Upj!tZ!98($bP)Zpom|` zhjZ5I-F#SbgmbrlhVI+(B1Ip=8m`JH8We_U+N%5{ass@sq#?$%2J@@**m3H7-t>74 ze;;=TTcvZ*r*8x7-7HC?L>=)fVxb(NjHvddkd%t-Ma^TJ&a6e{fcJP>c@#FwFXEZ{ zH7xis6c46*qorggl#U-0-_PrKS3HahV_#6(1y4GZCr3A?J`ik&T%%tfF z4fA|=gN~1e^d%MXZhDVOl3P&NIG)EWbLIBwrw|=dfMxCfVNu;PVgH0vbhvvxP5Ss9 zV^4^j3Hc*EM3XLic9E~~50d(OUJ@}pmJ&|~i#ym?^QO|{;y&U#{79)E&0m*CQuzb% zeCI+6+A)>VTD|GO5*vIObQ0$(^I@g=lQzzOLyMxfQDenJD$7+zc%QTQ&mosuP33WS zoEy}|Y+bQ*BKp12=f3iJgwHWzkHm!IzL-&yY+t(eO`cx=DHCo!&7h_R3mCi&qX+Mw z2o3Ah>748+D0tc6;%ZGK=k4@EN~94n2Yo7$=d~=)CS3% zY7P1>h68SiZj$i%SyVfti#95~rJBy?G$Yjr=b9R+@c0WF`OzKIRug>nl`%S7%!f=j z$Att#u|IT4_&8{okW#o`xN@qRbgb6Wp?zCu_@SAYwC^{y4!uK43AZRi_5{Tm8_~`2 z&nZypCRx`$qF!HbP}<776nCSHHo2JMY`z4(3ZoD-E)w?^dSTKL2lNgIC&g$xvX)FB zLve4})s;SI>w5sl=cWt4Gi-z*Zf3%q!~Qh4^ElnPyp%fT7_<611rBR#LZh7~DjxQS z^oSbT`9qw&z5hk?e7nKz({9+#Vn|<~04@7f^loQ{&~fX9Fm%~P>Yp`+X80=7?xJz{ zI(;@~jGBP{Lmqy^%cONw8 znPIe)0?s_TNpcxekQ8YTU&Ccs=_oGh1R+9!Xxvo!ol8K&@@CIyE@CsC(|3laUQB}H(+?N0H1>P zV)JIgMDr<}+jj!%>uPe#Ib%FK0d$@njG3R-^5)bLy#HD*Hh&QFfjM@(Z#J>9x&ub7 z*M`T%ao7<$nZNd!#MA7wIPF##Wxi0L4_jskW{*N~M0q9({*#5nGe6cap3N`9WXQF}I-{_9EN5#`6m98;N(3HT4+17hTUzDNqojc3rn)DU5z_&*G4Tg+4P#;|+jI~eH3A?3UelEpk0c9cRU zv=Bqeq;Yj)gQS0_s$i5?PQCu5kji#}R*g!LmM_8KBOCFb%37SM z9DrHv+PLt*0+*L>7PI_&sb5GK^)Qp---kKIBBz?LfhRne z%|Rd4p#FY6#_Wm4N(Uz@9-T&}d+$(n@Kd~BbRP!>pGQ{@J94yLN{S`pNjJ?;80sA> zY^)H!d-{FxC*ukE|I4FoPqu-(EO97AfSg#HO4^>$0I7fEHEafAnipf@ww*Y#zMLGj z14!lTM2hUCNS9;BQtQ5Up?_!|PU$^D@wYeFzAGN{oYGL#B>IptXvRbs4)Do&hOfLgXk zX^&87_3?#=+F>|o`{G^Y3@m!O23xbGObS#DO?IE_Bbm@&Jok>dz-{baL^}V#x+x#w z+n$dcowe{CEB4?Vs_DnD$*kE|oU^1VvRR`IN~L6QI`9lRY~O$fMgxHUdnsXZG8rf} z(W|Rs#=9y-kEPEvxvpwoT?9XP6(&D?%9|bSx!v`R`g|F2zfZ? zqi^XT9#y)Y)je%U`A?=`rB*{DI}Ld5=K#JuBbL=1GjXC<6U-w;q`+H)jVDg!`_;o( z`dA|!wO>x}no5NaZ5lK-`;rj-djR=m-6XrE!y&Fx#lA8*x*N0wHxm81WmYVFYZmiN z^|Nd-Y@6sA>5Hrpk@W816>_}lMgP?1QDs%2(A?f31ZjuSGgcz0#fycI>jiW!DGC3! zAL4Z~y-khU#B6$d9oex64)wThX-rsPQ+{icHBk3Pf()f;I1^$pLP ztkGJ$3sUJv(9!FZWJdUNNm2R}Vei$$WEgk|Q924NGqW0@;mVw8rN(u?7C^yr3S4Ym z@zF(w^5)-`y!x+5&|5o|Ugl@g@9YyaQmrQ|h(9*?IXV&zpB9tX zhbG}@gEf8f^rX|lSE+l;EL{4s9M&HNya?P6n=~2Tlh;#ho?HR`x#DJr7b-=E)$N4(4WSo_LDi$zCRY(+@}|!iz;!=LA2gDjQ^~+Bf3-7 zH2b%XspdLm(@%AWQF!?uJhskf13x8tTQC?gCY9LMdXul}-{h~AXW8`6EW9YTLha!{ z)U6~CecO^@r+pd?qu29-v1@tO%muv6VH#WhR%B`Nh25KUemf|ErR%Ku+n_1jdSNNo z|Fh@N12saoqjFSi6-Z$jTaY{N5_*c6-%sYxIQ{26J{cX%wmLWXpW%Nz@K7y#-FX5} zv2XJGkqSGUn2N@S-lYE_S6I9*m`4ZX@#mmB{CaXe;>C>5_tW1=&(VrMy2`Tsok9ex z5q(F?vSIRBAIS}4c>TImoLQ@Yn(V>IF!0A{mlAaDos7D)Po&?biQ>NMY=2Rob1!T2 z`HNz%<$yky2iWmve>qmy6*qvN*@?TiHnDPbEax1o;-BAo!K|PilYDpZwpG&j8U2hN zmG^-6qym&QIYU;~4pRrJveXM>9^T^5BmWMCRPsQ?s}91e41biIUVt?9eQ^Kv5KVLp zdb(TCc<~j^Wq*LwRYiX38i?L&SL69DCw$*@98T3Iu=8apjK+B2ntTXy*5_lE|>=^e7%QrvAq?Av1*_ep-2`6Frd??46 z-G!arYn*ZJ#+%M`XUCml-9Nb>*F_KK;Dg4zbLTy>wS7$!7mUW&+5(sxWx`1JIl8O* z!{!@d;JKk9cRNnZ&#NQbTn%TsnQ$jJ;;LFW?8o%L8$V-IED<@28yT2#bT_)&EI^~! zgPpZwCpvxIu<(i_BE1$MOeFwXN6ukX<4p{=73aM&1!!G=3UN!%q9UM~p2?~qA=?al zBn23KKOQUkksE7-(vk?=U4{qS4VNw*BRBD&f-Y#IMJt; zkM3nB;rDzN#vPMJ;f_emFLcA_v)f@_KR^=q@qvlv%r_yNs268Xp6UT~=y*S}eby;#o0Cf6yE5r<*b@qx ze~(8h<#OlxQ#@TNP~?4Q;cc=SVqYJCPk}fiSt>FlKbE7|XCQv&U#2ea8MLCu7vc7X zNa5wklQcYOJe@kONI8@4VnojpSX!J!PtnVG`r|rOU2Z3TaVDg`Y$+yZ$s>AqB=(eq zV}QvqNK+2o^7u?jF2k_lUKlwTZKcgKmXb^LbIe%g1Aq0NFt0J^N%kTyF*O~tQf}e4 z!$}O!_l1;Fe$+M302y0&_ z_Q^D`d;T5dTU20UNj6&NN8{jccQlXJMb>&lu|;{mn~Ov0c65g1=`92=X`aq;>E1oR7} z+N-KG+fR$W^$H-R`U>jO&ZBEHx1#o&A2b|A9`WW&D*D_(&!+9d9&LRblDb1S&m!sm zwApa#84P>VIM}?MfqhOhu|9Gp7B^eqwdiUZoE1XJ7xn2}uDg(-!j!2ro31{$ljxob z73TL4cj}crg8k-fq&9?+*%J>zeM^9#r;vtiyB;9DAs61BZ^>rWFB;eI5_?rOpN;S#cO@|zpEN7KUayPUOvaQJU{WfFdUYr zQZY2P2}_E;V|-|)pfa#fz|k+lN)p{?NjKq}qR3N@`r!VDt#EAgz<`D2f?VZ2K{hg2 zI4Sy4zFt^@C~p_YE-IthGNS7^u#n;>8e?#u(a`c&!j({K>Sk9;4RgfYaC;{v4H(IW zWA`w=%Ru|Q1RKH$4|1+hh*Nc#2dj+z`FJi&-QS@`ic#;q4P5+X|@k@uf z{9D7Dy-t|p;s%E51rx|EZ>L1-(Wt$(4oOB1dJ(~g<00kg304k^m9WmM7LV9_C!OjaqYz|sTYOqUli%cj?J{_ zT`x*oa+b1M%jw)?Io=-LgPo0r@EwmuJhz)6N1id_W#Q_Qbzde5hC;6JWwI&Iy&_Il{_>0p!p(kkaKn zDG|%)Pv>B0dzxeKx~W*Mdqi|66botAsuUUKMd^nFC_g`v4(&CStaUms-0Q7J&8uC7 z8H0TU<$|pOMt!FJvIvGS)raWQ z=q|`5W(z%XP6!Ph!NTHQ%3`AInBZXHg4@>|&{uL0%KIk?V|0E=Ot#*ZY#qCrKkiw{ zziQlg>*P7O(_;ywA9-Tw&^e-mXFd$Jt-#W4!+DJ60KOfVh>mIdFg|P^b{oxuQ)wP8 z-PS>#izU$S*NrEh7d@}lvFQ6y4$ebbXo|}x+SRd!D)han=1wv_c*LU9b2zltC6no( zCs3=%7V{w~D0!HUs?1Zc(+R-9{fu3iLon~ICgcteWY?INkWG7xRVOvE{qaD2XdjA+ zh5h)wkLaIT)q#-3!&p~ai5~_G+dfl!QMw{8`qGKVYrnuQktPpY{tJH1 z9ay={0i{ys2OzADaeBJ|m3j|m2chD!8tZi&<&-I-c#;9Sw#v$Z*1R zAuwVgyq!X@W>hSMod4Kzb{98%&EU!V*VFS`pGeI&3L7^$(F(H`%;@qHKJaBL{F5h1aMTdML zY()#btLY@GNG+5cN~X}$(ol4Efm~}Rvey2_hK!5Q8JtN+kL;m6bsmA{dtrC?7ADd_KAJX^|Fk^;Wkf^s*HnyKlM0{p!KhppD!vc%`P5hk zc3nDw*BV^IA?`8$f2#fUz!chhL?px4h|)c3N5<)CMS=x2Ht? zZyDT%9U^z7I^pxyI+Q2%;OJ_RN&dtLo97Rw+lkl`Cie0@9N-w+D&}Sb>E!JM$(673 zgtZ?NF=nq6z1`Md5RORrO41ALlFGoA=q}0r=Kmy?tJ@?YGkS5}($DbQ`Wzdx9#M{- z0ZDrH6k27X$ZCic{f-MD=iVCBwk&{F8|uL}BSvt)*+W?FWGs2;Yl)&~v!H4?27l*w z2$JXbB(H0SNVc9@2GghAP&uiSC|QMjd=_&F(StbK`V(S)MBsPkNX*d?8Py!oVtPCi zAyL76@x(~}m+1&CG1sj1`4klYT*aad&nT#0Jc=%;aR7`>8~PZSPOg5*C&M0#z1%WH zhK^&`+;d`XSDI}{4&h>>mqK^de4)S{P!Pr+2;D(=Vmc?CCqtMb9e+VtR-Ev(z#qsn(bjdYq$=F27` z=wTjJ#>~Z)4f33;-VY}x6=Ra}68@lhfvci#@ZKF4IH~;zJ`_}t!_7JLd7(UiI~WD` zJNadT&;G3%O7WP^RS~l zFeZRsd#~hPv!d}fZ#s(t51xZeoXPrw*Y`}=uxTIJZt6u%*_xDjHXilS6)^NJ#_30Q zDCNjsS}f*gZ7XIGU4ARLJ%1vM7#mHAp+WT6DT~$}cuU2uz0k0HGQ8{)Xycd^nwWc? zuC?DMw{r1Zz26Pl>Pryx%?C^F%tgQ-RpfOWgEt>Wimvu^R5kM$U8_7vCTGkc|9k@4 z3dSMgsJMe-+1D2hjfIr_Wd$|8pGjp`B*4e1 zFnK-~QRj{b&U^F(6U9K`fuB1yyZX}=uM2eYa|YsmmEn@nAI$ERLvn&4Ei^KrhVH}g z{ooS{xKd8Xmdn%JH{EHRLkJE2mrV<8nrZs4pOn!36NNATOu4ro(&y;+bR^~vE&DbS zzs6+KlebqWu z6wy?Bf{)51c(#jdN8wK@KK__|G^;3mnizM9Re=4pd74J==uO>ht_l^mzY3}sBFR4~kOtn(p?yKisUxPl=-KWRTJK0h zRrU#4U2i8<_Ym6QCeDKkr_tc(96CNPlJ>5+OdmymWe?qhR5T}^mVauXuIv5r^RSp7 zy*5?|Ki?(MyW=2im?&W4!v$y={f0@ZItpLuH-d`{c(`5Xr}{xgdu;(2gjsyS=E7{-S8hOlf_ zA0Bu64VuE|Ch`C3dN*Su-$z%qDFPDN*P8g|4~BkSWGd=I)H`ovG*hJ_J4 z`Wr!g>T%o&ufny=0laCT3x(ekZE{3iq495bJ*hWhZIFFSeKuw6WO%_}s!7UROmAXJ(M3wzJq5h_JqBhq>> z|8dacOY;w6dx`^EGn=tfYA!qK>fvJMWCWbh#;v+3WVQJE;HRs^+Ra|Gbm&^M2@ZB53a9~I` zcrHCj*F?^JL4Y>9y|LnqKh$2r;GlW!CT@RNf(d0Eav{_P^xD&cRa zb>?4jPmCdtU2nv_mXF}ka-(qE-w)4pdh*l#QrvHpGWQrXic?Y&5!rvBm2$Np zqf;b||1_RlmT!iU$ddkcn#1+IvV==>FVWF#OPck_AKBmUp*><0Pg^^k`V9;x&nJW6 zKWGN(mEG}pu(Rm0SVltL4w{z~MeO+<-(Eaz7-kRGJPWAmMT=k1krhAqjUOCTZrg1z*5eRE=##j0MsI+n9SX~uP zJKGBB*;^zfhmE;me{K)bV2hnci?F`QRFw_;(h0+%cTbe7 zl;QZ0B$#CE#ix=p=ySOY!wSn#-?<0pEcfAMA17Ep8N-s@eOT^z11x4LbDv}lp0!1s z4;<)*fPE62JUCPAC5tly`xe|3=h+ie=sFrx+b_|*hncj{L}U=Yc?(V} zjHutD(^NXy4`H#P@E9e!OAb{cr)e#A_y0uC8y(RSvj|Qp40Z1>baSmH^molfAD?K1 zRmQ`7zdu5P41|GMal-rZ&9ss&cvqAXuf28;&XYu3qx8XBi@XR7iR`baB!Fj_kSWf z6nA#vOusb59$G1Kg9Alorzcliy+T4av0rm201t{&aPr$7{FCt#nP464m^lEI<(p`) z-T>%SAIH;*$*iPw1GgWqAiG3kJa=CUkNVjt$;=cIJ;zeq=4$c@))Kn)dPal42f-&G z9k&O*L(8T*(OdnTjAksy)uR`1xc&$NODy4~o+a`^1z1y&hQQ7-FwIFtj>~K8I?#x3 zxBDU{wu9aWnc&_?EwuFA2D|KdM5Sj_`bSxKbRUD!drk53mMcPL?m_bOB++|W3-?29 z2$xz)bB^T*oeJy7-R~p*q~&)Hok3lM&|Japn!P@ z9iofc^uGAs?TqNwOT)sY@#uQ!f?IQTVT$!$JootwqjD*>$o&eP^;%dQu@ojr$Dw61 zj6U?1rX!Z-)cjiuGZn^T@Jee;xjh6$^FvS&@D4g*6EI@yX+)3sgJz>(a+58m0ZX2e z!eo66(HILm-vL=Mj)Kw-!r;;7SnYkBiniPlGSbe|-BM|2NNUM3ygAY z6YJ<)wF#X5=`U=J=TV6JNfUw_?S-Al}4A;^+&uHPxq<<8jzMB-L55U~y6mHxU)1MOtq3aUPot1Fu6?ga-IEp9TK5gG%1Bd zgOW-mr9q}LD?{d?L@H6jb@q{F4MK!yrj*hk%~9Rwz1H`Aug~wZ?)(1ZXRW=ibM`)G zXu~(Lb5l@fn}W;oLjtB;4Zbp z1;NL?QONLqqp<)#9h``sa`j-?xCPs5)?kD6U`)!CfOR9bg5UCD2&kP2vBwU}bX_WN+6HIC<04nk?;cC)4n#4( zzNj#{Q!c?rl?0fu=@2ZBNe0O}HAr(f$n2ld1sVclceuU+^jf<@^i_cY`Q#AvWea)O zX|lLa&@SkAaIkFWZ3tHs_=0Dz65lIrO!{_Nh95Q(Dc@U|z; z9gACuG7euXuvv0Pv~7dUH3=Ru?sL-X)>ZyF&xS+2l?<^SdUidm^cs(=lzDp@mcttkH+E0wqV1E zYB+Ik7fddn44*uG(Vcpt>WqK6yWPbF)kOLP%+sP6`%T;jpxonL27 zBZIb8ozO0P4h=+(5FIcTw%zo>$m(V2cwr%SXS!ms7U0>wAt>eYOvp++4-aQX!JmUE zB7gsU^YB{|WJ%o;H2)yUHZHMe-GV1VoS++dm#&17I81ct-f8n^dQ(VFtO7G|8^_F1 zxl4MSZZaOCeW0svhnHK6n8vkUOmI{#89KU;#5P_8Pk#sWO1lN^DJ$@FUn5Rye=6E~ zQw|1Rmcc8J=y8iP}1Ae3Fa48JE3lSd|esU$~Qq=a=!AiNZKH^l>Vn2&Xb87aq*r4V?w{f*_Xn@h>pW#{~DNTeGK+2+YkLMubI=wW|OMfn_%tgTaf!(7I$hb zz@pbL$tkI;#OQ7*(O+c-TFoK|m@yn0Z#%>86PsX9&S6;bN0Vr8Xk_M&Nhd3M;+PZ1 z2NRFmjpWtcWa8W<4P%TB!3f(0q8pCMB1U^MxfpNG*nQ3=ZOSE}>2{D&nqEi>hTVnd ze1@n|QHG?CRR=!r4jJT0$oET=iRoJ%(PgiZ%sHuNL`61N^sdQ_8NEfArw#i=>}n&S z`=J~j+ojC>y(<>&)khNWEK2m-VlngfvJp(T%OVbUb79z-)%Zo;OO$q9id2-igRh`x zaf(sF^OfdU&sjs?(;Qf4^cjAhu7yi|f)1-JpJ)ytaOQp|M|3eS z;V{Wtw3{>!Th1i^VjyU}Gc!3Saa z*y&Iw$_IzJhVV1z5((>>Mov0b6AzjPAFOIfWV4XhG-3f9T=Ru#kp0HgxC&Xd`4J=} zT@gyJA0VwdXNYIn5r`Tt)aq8+p!NbiOqu;&$YJmk^gb~HW0#OceNLqEcP)AA!N9xl zP_XtD>N$5zznj;RgE6J&DZDR7O0@@B0 zk-_qd;a7Q;`GZA`B<*#Bkh4%D)DW@+-9b2b3Yrar&(iQMY7kg-Qs&@|YG(VKXXJTq zGUlF|kMUT#9Iq!Y#7nWEjq@7{VS; zw_v>@6U|$KCzA3(QPBO_f(abIMpSBKO-#a9nLjI^X}(5nrs!Fokn3nQ39dOS;RfZi z<_jDO&7-dEC)Fm`V0)T33LH?>94-MeHFGh(Pm=w!dl%f%z6Xx=_UQT{oEYTkko!W- zCn!mg$r^P-bbQf6#^K&9&}_4T-q*80IesKj{P+q+G*zNk!f#Tpx&g$S3yOt zUC2m$(t-5lQtOV`{%slF(@Y(X!RBb*?>nJU3&uhFcPUA$R7> zqI-FnkriA^iZ^ClNss^AB*yeLS}|^2k9ytPck()vc?S{ zwrwZ+?`*_J76Wm#Fu$62QyHo6Cn1OPCnTlZhOu`B5v#Kb=0B$fGrt~4VbzHDzzEvK z_)C+}{q0^n95NpV+{f zVg7JGF&f?f=3)bh!nx5;MIwPObSKdj<`fs>iqSk~KlQ{OKAZ$ZNimfI!)VyDdQcBh zg~dMp#Cb;@)=w+Mr|-|=@=O^Ld5Z-Dq5f+Z5(5vm6+l#a4Ai`ogGnVmq;+!*AqV}* z9^t!rb9ES0T)Yl@y(fVNlS*Fx;IXA>9y*K_vLtK=V)&fP#QDSlCaQNa&OPG}BQuXM z?ouOg_Ihi~l(fdIDS$;Q|3cf>HDDju4GtS5v1n!@1jN+A=SwoU@0Bbj|Lue1btZUd ztv!~c++ae&f*G4p`Q)sdfw}SxfAdI_Va&UN!H}V3gfJ!xRd-J#nlntyw-?x%M{Tet zhZG~2Rforr$WgaonTj2-$MnJf&n?)#Ne%7}aw8Q7To}ijv*E+1QRtXPma36!|L25C16Z(SF` z*|+;4c)|`i>wlO$4zPnM$+zIbJ0)QcS^~qj3$=k4XJPJ&FED4c5~idI^O95((fBMI zk;KHK;PVJUr1F-0J569%?+kzq_K;szL>6_`5)X5EA#>s^gghvR>q9R=YUdz0Id?Sd z{WKBAk1;{JJPT}iV}@oAet>1~CwTGt4^%Ej=IK)lrpH@_EPSJgIxn^G#|$Ox3!hGw z7=)6IhR$T%ZYwfSE{5dos3lq2`Ivhv9}AAe;snoNQVUfiX`LBJ^t*ztVj%bm-+B2( zw_w%zlhALo7dA-UVVouV7*g#=HmE0)WBQJyxb{19tDpx|bR_Y~?3eH<_Ygqf9{9ey zha@eqg?e8ZNFNeJ3T6+2!r}8_)Tt0qu@n5blLY48?}Kn%PmjDRz0Iu4I?Z&tC&K;T zmmtos4c6_81L+4RL2>bEm_JiS;BnNGy16%qOlUnbcbXitsWwAoz@K543{}}A5UqaTOr;xg!7OpiF32SdL?C{EjvTMTJTT&jcj(h|auKJ{?y@E`M6<8H% zf%tHMHJYUjz^ciSAm$}L__u2rBMFT13nkb!m#gSaf`GJ zE=agXR7YkLlddtsUb-jNFM9>0!aBR@$tPxLz&^4>?J1!}iIDB!4Jz)|;GTUPHKqq) z!ZS;3mmdo~+eZl*Jo@m#=o?;0E5zyiLCj9PLv{uGLBE&-KU-man)v}%$5z7Clao>O z(HI|(8;++*9$7arn6&kWkaEG-eZ77*6szfj6jy=k0#ngLVIMAgYK(<@2VnBq`(T={ z$Y$KRjrI}|_9M>(#;IKb<&VPpv?dhzn{N2&=R&-aI}Mj*tU-rp1B{+3iEC=(V7E(= zpbftXJ@|%96z0`uNd$bD{Zn*df;G9Id6B48c|yEOJhBt1pqal*0ny`{!~z9K0@EV+v}`3ABe1 z4gMFTy)1?LusDdDokQgMZ!0Bwc!Sb)t$oZf7N;Una+xIp-<4HW8z4RfH&^dm^hXs~}} zC$nCsJlQ2%LdnZeJ+jtc7A#)%lZk)d!Q#kZbo*t-Jd1E(-n`n#tSgkk*;Ug~@cNYDUnPXhmhM}o)30Rw`MlYR98cnXL30@8-_|bvzt~-jc z5?DXX&@EsyMg`a}f#g@Bz=cd7P71!6l12|f10NAa9-cCQO@{_!xm^c@=gz`|uQG8( z%^}j#d5rAT6|%5&2jR-$R;YHs0pI`B0yoci5k_%Wx#bLmKVVN5NDt zbEtV7F8DT!;KafKnCS0`%P&pCO}~n8YzmL;xnV$$WI>p@3Z9S`^ir4G!StmP#tK@9 z-k)bdN#GOS*c64TOD;mSeF#ZrGf9u23o7tuv9M%0KAd5IqLzH95yhELP2ofm*XEMd zcO1}j%52PeH4Ni|g*xw2!J{1b{YKmGzKp{=!Za)R9Lywk+|LT2Zf%CQ2Zqa&T-1@ z>@73cHvD7C6JK+|^JcFo@eCf>AH28RT>upmVw40e-NM%pG6{vPyN}yj$>tRt!e1*)#E! z>Ii&yQy+p7c0$(VYPi$X0>_!hz#Q*@(z;-5dntnv(**xg)@{%_v>EMMs<6rHD)=~` zgzokfI5VscVhsfjZLU3x%B+Pc28!tIxd5}r$AjO%CSon*0+kOO3ilJvlfAveNoL3p zEHNC2cdy68v6&H&a9bCitoPyQTpqM@E2qz$Q=>O#lO{FQhSaD-I9SLF=o6hsF3V4BlA;Q-Xuw=vF1XbYM43 zttcRpU5I*x@4=yY3rsnC07i{;fM=4M;Ir!_bnN(yfv2MN0YsZGL-XiSkLV)y>1j+<+FW9=`81*5 z^+7lHDzUY7C1dNaz=ldX$U|KTv<^Xw>m=kJrhQ55{Ix&T*ELcASi zvHj^tbkARe>{Gyy*NAHNQ}LE}ESlalL93-ju=LYPbdqwwdA(C`(M~&1S?@rKh39lF zv4!Tisl+Vi7~`!c?Dg6%hXupJ$^BpB;e+WCxY@rFMvaMsar<<@Z_GQA`{z6OzLmou zwPuLe=K@EYTVQ3W6c(>Mh*!#O@#(v-z@!S9b1mOUud*@N9A*2Gg2zGx> z#Q~0C7$kDQqmy^SIcZJUSa*t>K;SCX0IPgy4V+c-eOE;4k8 z2?_tyM8q5R5=Je8gsRBH50M#^tK?#j|rzG1>7SAlegNIx2 z^`1BQw9*W&u1W`4_sg*8xd}Lr<%!?E0MZ$i3BGyzLCMY$7P(2_qaV#=*TqOCIc zhS}aT$&5jdMV9skIA_8L+}RNq0Y7 zBwqoU$M4|v_pk8v>l5(Q{sY|tM|gjKE=DW9=cZ5mU(f zRSWQL*bNZv{6qFzxj~@V5SA4yL1OO<5CAn`oTC87JvlI^rx=RkFT$z-fvH~l17>yK zfZ92FaICeNoSF269M+zXD$C|$+OFmJw5kBUyf_IK<4?oZg<7Diy8<3}hr#;f`S@GN zrOe-9gE1T3K=#ls$h;Q`Nq3Kv62Sx2bbKJ>2J(zS%TThawu!VfWW$KC^WbCr9xm>Z zgvmp+!Sj$6T-=%uk5xCrp{%iR=29-W8VMdCek<5OEm*I40MBQ2!Zd*e;uSIwIvWQH zy2LnOPFD(S+74JWc_h4W7eR-NFU(YGhuqr|0#jEGhgtlEx6K-ueBKxvNfXh{`b1`n zB_P)O5qymM1*5x!XWP_mWX;hO5~o#9s@oMog&PdtH_1S`!vW0L7>Y7$qp)lJT4K?- zmo(p6K`z~Mhw&fR!SmdmaKv~WjLeCJXK|;2FA0M&ZHu8(CK%*Cc|st#z}H#_cx=!M zn;WHYPq_rnd5{dpFQmXh`xN2*pNt998G!+5ijm!tcC4nkdA5F9XH1P25TqSTa3Pzd7T zjzj<=6{d<*M0x=|KG0p@3DWb`?v4k*MIu{Y5$!6pPIyf z-1E=zzwJM-{m*Ou?f=i~{&W1_{{QR?`j6l7&+)(4{CoVLih5U|JQy0jQ!gr{_EfHpZ))g{X70o|9>C`#|GDS?(}%*A&cFSaKmNpUI~DYhP{iH@ zfrDl#WExoUIO^(K?&G^z^tplsy*d5^jafaEmsj`WwQkQx>DD-L(=|bVdb5B_t~g1> zVp%>(Js-Gl$#m~QA>X1XmP*7dA@l4m)9-2R^w6w_WZ2FC^ccOJUcL=bz-mL`%fE0u zI|hHU2e=2hitM>f1K3mN53o<ObgeSEgL@uuKdsv; zOa~69lKH1e)V_hF)#?byX6M51U!mfIW#=*A$~M$2S%O{Pvgp^(Dr`vBdFuKohx@A3 z%Kbj-0L2f3@JG!p@y)}dsq7|8I-~kFnBDRf^X|vd=I1*uZFT^6>y`vv=%mMoIu4)* zRE=SdqZ)sWx8>c|EauBJ`|04)d(`DgFEzR1fR5`Lgc%>=QmGy0sm3eo@@)P9~`M}^V z>Q|G54Jpg9?W3#UvztuUEWFCCs2;*wblOm}(hO>SauBcF5WsDoUdH8%PtlslIC=}M zd0)#Jyj`dyzTdeAy|0cSvChHVAEEjiFSU>Qm7b@qtQGyn~s zn#Qdl`SS7f&zC*4a%2a$_QD^|=jTQUI+sDSuKt4Z&T;gzsvIZywkpo%9m5aZ9UR$r ziXQqqkZ*o^VDlz4;_Qqw^g+RN-s0gI&Zt|39brpg z6m6y*+sAMng0B2^os*CgF5=mMRM4t_j4{$tV7hi3E}Q3oTl0hI)P1$^SK*$xJb4?4 z0s?UHN(b~vImg+(O@~B>7xdccEAnYEuEbw7_k zudBj$*xaQqt<&%wEdZwv11Onv9kqHlprhdcmKA+Q(+PVy&9|XkZ_Xt28d6W!#cT7Q z6eum&mygFvvdO+JWB9aBTUJiTBz0B%{ z`Om)L+4>?f;nGifPU9Dgx`#N2kFwbrA!%Jai{MJ}7_YqnJMP7v|BL2G%_)5kJ|Lj8gab!`s+mty}YEmv~`X;cm-{2DWin%><*-~a8UX(nf>79j%kry30Ac=2ne?B(}VO7Z32-{36A za-85+i0clFqvwP7(~*BwX%$gr=Nh=MQqQf~DX$!{_8!ok8M3^@DoL0%=bPXqf6A3C z6?}K?`Q%%K48EQ+n4X%WjAwoxLCz(Yo=vO8<=3)U+56J`V4bg=^2Q#nLwP)O+iFtG zrqo};4Om?_&X^J2i{{UCh1n>SxWAYl-B&^bO^s;JQ5`T7vg4?UC116xpDGK!x;s8& z>1{J<)b%doS}zWxU)*lu#-b9j&8oZHj!_C+`{)qvqQ+m)3L1>B78|g0kB(r^PjtfB zN8C}qvKHNG%$QVn8 zKMv_a{`PsiyJr%194*7gYe!>>R0y|ls5Du;PJ``rGT|1qCvo|y1I5J&gZRYNWwf=; zT)cAmWbXMJSL&LohZ`47=I@o)bN5z1#Et}6SlD=g+!I&fZe#}`veqvM%eufm34AQ{1YsH-6=?>a; zP>qit)yT~;OM>|ddbpytfZW+MoLc{^q3ccZu=T+RZ2233Q=)~nre`pg>#e{afjL<1 zlEO`%5X(JxZlF7Lv@pV8F)2B^hM24L!n&KgK{ICoXE)V>Pg~Ku^p4Y*x0Ggly8%BQRYKqs5yqCy zgH!UW2r3^EFAj?*nSSDms%wT^r0|*O-!8$H=)<3RDr9v1}?tQ zCl74=u>M3p?lae8cS`T&#%!E{!%iy!+=}3oxCe0ai2)ifR)#ypLXO^nLhiJc0j^jl z!@pE^;3vwC=MSnDgU+Q6Ojvr0gsSS{nPcza5wxMp9bfum#XQ(=T7r==b|^XGM1_Q) zWwKrNSlo2RgbzO%k6W#~$<_ESW_=tFb=Q+%+@Y1?ENOjSI$;I%{j!mBnlyyM(b1f8 z)J4X`cr6BthM?E$ZCJ8(cZD17LH^}Y>K3{XhsS=W=XP83$J;dcQ`KpFeo`dw6>tVr z&j~f!QEqTF#FKUZ!?0(n?_<>AXg;JRoO-8A67%zqxb3Cm>BrUMxdatSzEVX7Pe)JX zN?3a?H>pJYOe=^w?U2U)#jm*t#|C=r=L6W^^$jDe8lm(=4#%^}RMMi3k9p`r*FVzb zciv3rueC;UK7POHt0S}d+8r6Z*_1r$Yd4iPSgyh^b0l$djtb|Vo=De~$#FF!q}b=F zJMfu90jw$#Iz#_}%e%c<_HY2wpob!duq5Xx3jZrqs9 zwAQqWJ6Rkj{*~LtTe%G5=gc*xZ}(o{4@ea79;z+;_{|^qJF{wN+wMGAb^HzgWu{c+ z@6$Sb_@Yp(VNTGpNq6AVmTTNkm9gThiSxv>>*mlkwr-q)P~@{H@Z#&8qaj&F7QZ%m z(A8(^xD`+8Ie!m*K6ti}8-I5@Ciqf5tJCcXi$c-8&`8 z`|tv8c}X>Jt_FC_^fukT{1dq7jG`MY68N>BSF!6e)H$7&HvZLX7WH=*p^-xnSHJWJ zDP8VI7R=GX!eO<{_Q(6g$4}d!O!Lt5T3hEL4#?V6RB)5md^dsD`U z&jrsp#n;+?;cFDeSE@d~&d)m-%&!QXK$Sb&@Wc$k+bH(J)Jx-#jt;=Lb%|J<3H0Rg z3_7q^j~g;bhUQ$GPG9|y;N|o_R_HD-z~AH6arbNI1lbEG@-echG@LHPyxb}{anuaA zlgsG((wz3~l;(Q}FXINVm2`aH1FYLBgB|1m zjEksw@>K5OS0jATQVIH}PIJ*FH+fS%BYwfiAS$)96i!~+43pJ6xuSSQei65gR&-R* zlRX?wR}!+r{p0b0z=Mc3P@pnHooTW15Zbq?lZ)FCh;@>xG&0Xy@Z0x8psa`3sudHDCsjOOc9^Oe30b6M7M_ZV!0HE!@W-!@;@dtw zRB_rK8ZR@G*7(b_UIKT~rd1IaoSV*#GhT-XM7GqK1x}n3}Z1}os8+=^$1a3vm=gI;<(T2Kv)KT3M6p~Vz+O=KW;^pe} zi;!p6(LWy?S4dM8lXM;}7V>w$|DsN#%IRvEDz0bvb!zXr1{Wp%hL+6n_`diF)qekr z{&`Eeee+-8kp&d`m|?Vi$!Kbks0vXhS7L*yGB2%lmEM)O#Z5l$&Yk;ii{I|L!=xoF z-7F&90?l-~Xp@uJTTg~ITg%a>^XfSB;KSlniF`8LvX`o9T%;5CoTZD8G}2XZw@I6g zz;UiU$CTZ;NOt|G;QXR5h&M?M;Rbw^qZ3wI@ruU+XqL1s`g?ZMY-viX6-ubw@EklF zeI6#L?xfD%60}QUC_iq~7G6E?Bo?jx#Ocd3bb9_#`lzu83ijr4UIW+RD!z!?&g|f9 zA}#n4KJVzu3Z#X};dI7gPim#;gxdzq<+hI15|><0Pxr}OT345Ny!1=5EhQhKI6XGPOWNXVsf>kJpbbCkYy^Gm_?0_n z6b!v@a=9njYUq|LZxQ|UA*&@JWpSo$yjVG?njTyx&sP+rP|MaDZb!mMF7BYB#oYS$ zti{Z0EXbPUYkN;FLamjP(N<$)ogcF=e>bwLH|laP2A>oMOg|w0m?h8J8Z@w*+AP>G z-GTH&&v8hdu^Q`+8j%?uM@in-Om0`15;k#Xx#vp<($iP=b2oM7(9gwwbP&4(9eU5e z70FO25La`(4~*#kp?Nf^G+W&4*;&5yep1EX!-v`ZNbIysb z@(K6%zMr@^#sCJ0oVlTkj#2Hm?ey!zWpv<`X{>dOmW555k?%k?Z zh+Ek8t0LjA8W;CNi56a=Y>VCvCfQesewF7UOGyi#%W56^`Fr%p3JEC?%=KJGq!bXS1sKhshdF zWBNe4cC!&%x8x93^Ng7Hol~yyZC1ta^-2)aQ8u*)i+Y-kPD03`wQAt6Wn$rtWz@^RQIDfsWQuR>$oQz%mK;O;D(%#}ai z4^H1@*pc0@Iiqb^oKbWLS$1>{IxqMDiWcT{akVOS+TP2_?3s)9AG*+uF=91`n6b|v zba6)_^U0ud@o3UfhWc5AUGh4DZJHHFJ@w~NztIO7YB~#j*FS(YQ)TE>2T5!jEJv&I zYVZPoADq8j7UPTec*`W3t;jvcK8xSQ4%AEMHl)tS-HRMpS7Qshu<9|TH5{49yrgmBiQulhux=(D~DTC>t+}qPqtC!H-Jf`EiDH-m5g|O7q3K@$=Y2 z?Q8i>a-RHl<2k%f@=5OJG!=0}=N0aa+)(P6a2unmKH|`j@qBh-7{6+F5MS}+gn0YJ z@oZSIG;L8;psKF|*`@40Ho`^<+OGT*I7=anCSQtvslgRvELq%Jmp~QQyoNL%Io_Hb z&X)S0!FY$0@agJ9?lJ!ZK5bWHHH&>%XYHx1i?I#M{r(JEr)<&8isfcGYtn!vgW1XF z`*56b4!>rl4)3mdgi4eiVz=r2MT;RH#oW4JmVNb*HFUkn3bqrp+UAGbrU@MBg~MrT z>t}B5(loe{3J~JwAQWieU?f;j9aTnDQk?oZ9I* z+|act+=J(J)b1_vqs@a@yPjk0TA3m?f8!!fTYbD34s9V9T#n;thu`?_R1Z8|9?o_) z=CGTbg{(H-kl*u4pSKPZ`~cHA?#R0YI{TqN<*wIp*S8E|k7f*GD;FBG(y3!vU(H;& zKPX&m`6q>V26l3W_96Tjn`&yHaakPjtPc*&Fkls%lyRsio-O64Vwm`B1g`ul#XV_==Pn!lCS3RyoVM2llkFY2E$^}*5zjPqx+ZB{ec_^w>M;-JC4K5WT~ zO(wBEdIMOyQBM))k7Vg7a6`X%vM(G?vA-@yvHPc9qRJjOxf2#0T>3T% z`nf{Ty2fN<)!kthtFn_=sRM+apP|OaJ3PX_%hutEl4w+_eFSPp6zSqq&a8Nw8ms+w z9qRk(QLWHB-0XGf;sO&Xiv@fJOQILEnePKhx=lD|m!HE)jz5b&4g=ULPo}X~pYNp` ztt{xRFG_6g!VltO84=?CEziUTi`7{!bQX1AI8yLFa^ipxhC8*>hZ}!wB^Q}tNQ+PB z(mm<1sI|-&<(}V%=WPYzER|ldm9`W3xNEce-m;kSo#Xry_i;-mzvFWA!Z{l!iY;>n zVhDZ59sH9gc;*M7jlk$`TX>N@)RD&~nlkLhy}Rk{qIO#7kwY_nnX)C9d(bJU0-5O! z+?#!mxL(Z=v`Lb|y!SV-v$GU^PJd?kTS^u|v47dYJ)U%}!W>#^_!esAd%1CrmE@;$ zT!qP{QCx5M2M(pylN0LZXp^3XBDqo;lY1Y}FMEXCL|1q{G@On(=fTM=s^Mz#O1Pj` zn)JHUQSoHvJ8LcdlC^Mq&KmR2=-b=(#oz83u{!zT;$7|@;-Y9fPIaUhQd^QRg&D=pXqgZeE!l_9O&Rl zWL6i#QujEh2ny$HUCO!XRbsAf^L1ABnmwCnsm1oU1hLQb_p#;P2iYivIy!V$Gp$jd zLkIG@;ud!qxIHqD{(ZWE`VX$*7A-o?39)XZ$-Rp^YiNmFxjc6wH<0SS&7?{{r|>6c zt)?+m=~OBvnD!jxsMj5BEs7ks-9SNzi^q=%XZ7+=EuiUsRK^buCfE3utxSo@4S zocS33Y@NW*GgjlZid6Y)GyCX}^=J6n&nNk(E5A_x@TvUohwJ%N#ijVxObbHXqG;Xv zRh++m4OikMc>Eg;VNo1F*v1a=f|gKNuVzm@?A1U%Zy^7sWD}J>w~bbtHq-lNqiMEZ zD3x)q;l|%{6%W=wDqeAY6W=w{hmSp+#QWvA3kb{G^hCm-5%UtUBQwQB#T0EqgzM3XY(*t9vmBtC`|8yJ;N=&D+)yR^Jg<$F4?Q~yO1(lUG;%9tw;@Q#`+Hfq7;*j0E?<5Jn zd6yK*ew`)gv1GVdpDIqaMSO3%t=Ig7$gF!YZ-Sh=Gk}Uq_Hjes4 z&!(}Qf%vzHnJCXZNW9XV5Lj_ z^;clv%2*t9`n!0f<|Rz?Q)VR4(SgueY7xpUm{*J#I*g`+5&?W#ymf$;k?|YHA3bP+vz?G=nkb zat_`;@rjFlwu#ycdux5#J8_^%Jp7vRmgtvMp`RDh1ihW)IGYXGPYa>y@j7(t%0tuJ zKWMiir4iLRT$J%MDA8F#XZcK_M}KUD3DdPWt7~KEBPU;6xxWk+T^NlUh3p;gtMAb9 z`gCgJAkT-5(4bC{D*T6S$I-N^4VP?qfqQb!;h+a{?4q68Y}x&X+{`^2Y2lD@;tKIQ zsPLOe=ZR9qBif3Dy!m0&ry~cXN8jgm`^=zUCmqDmVtWievXJV2ljN+c2k^Au1j(lhS-K74sPil7hQPxWP8)x5k?I1B%|nU4 z%BjRjN>9-|^ci0{>=b{=t&U%!CBg~Mk5|mTWrKUiWZ>&2DR#|48TRT^Q>^<{K^LVb z(8b>)uNwA3mt_V@0A!d?*;}1>9A3EW!Ub?B3gIn04hfv z!qqejFWF17Ti+~3HH~Ss>gyWrPm>Kv)_%fiIFF?dyA)~M1XKRy)lqn`-35!^OyV{u zYl@2kvPfmqN}OC916CXMqJmI&3%^pqP3*kPZTFo=MInR8=g4!AabS&D`M?|8sW1fh zgt@Y>e#x*8yE9?4)=Tbs)&U%C%`_X$s`?B@nuH=swA2GXX|Fbos+bB+zS z#I-W<=;2dL44&43TPw#ojD7@O>wM{wtuYn0D|kA5>3K4sB!~WfEX_NA&g0j6=7P@1 zTKXqb#OoxC;d3`6@PD3^^NXxls#m7VPV{cUhFJPJZwRuBq}I-TinJ@5sERS91zsW!puV zoK_BFeQk+k+HWHE%;CNnS+k$&XR?td+tB;ubS`CXJQq~r18=Ww!=&X2V0~{fwc)3- z`%5D6_?o-e9n~qkZA|Z59i)>-S93or3&hK0?BMd0 z7P@x3GrkB<68H9w<1?mK^3Rlyb!R9YqW?* ztlMWmUmTI;MNi!L$|W2fla$JvPow-1wJy;)s}%Y>Op}}F8-cfe%%F9q!Bnbg0@J86 zm}+?p69m+9Tx)$IbysW?7oJwb&{+xa@qaON9*$UbaU3t%gpA6Tgb<26_nfPuj6%^Q zl_-*?23mT}$j&CRrJ~aC-19p!l2wWl4GpDKDy5W)@;-mUGtRx|H@=^bO+Gm5O4I4F z*&v?Q4x?$>^qIFNRnyuA8VQ>*`{4(YS+oF;1tMA~CqV<=hQsiy{j_tH&TAK?IiUF2P1A19;))SLUM??A|a3ig>DyE~YuSk`TAwJyKh~vzRP<^g2nK;2<{+0cU z(Xqr2#C-NA)crjK;t~&nqErXhkQl=$etkjQJZEvOv-N18bPpVxF3pM@EJP#Ir?5&n z?s(kkB}(~biud;DgU6$@WZtK(DDgivxSruh;^&0HL*7e`nmG=TCCQw{hWkJIK*Ur0RMRtXa>yCe;caXJqc^kiAEA!A$V1 z41bdjtFLO2gD-_>%Kqt8+V>`CR#e00wBK-1$BC@^u?-|T1^v5U7vOJ;J{i}%!0l9N zL^3N|QQX-_sNocUlcUsz)YxY_`>^K(sMDiPwPD7X-uNZ9WwkiYcB9@ zWP`}HY(n~Ej&Xuy2*?B%F!r-2@*CuHn87WHxJqRa+f`VBR|L$a zB8zqDU@(vVSEvedwk}{Yu^ZAK4&pz_tMLIZF=+BOV|J%xA`Rv(+_PS}Z*8}&U)qQeo;8z7!&036oZ)DC6j0_-{!$U^3BS`(F z6=?{1kM2E~N-pmzuq^k_v`ku?VW~Q8HakL2vMJ@k>>YLpTUOiRGfNo!(PtJ>*APMh zg74FA_6+t|-E8)NeGcvm@w2qdTx8jwZEN|T$_S_Py_Z=j?1Oh!-op=TG+5`Ig1+6% z6t=Ig3H?2I0exHk8s8qjgtndHNbdd;Jp7>>Uo;%UaIk?~au*WJani8Ra~oEdcbN6d zlVGnsX~Md#3)vNq+}N?Nam=w3=Mj!oz;pgw#&}E$SIy5y$9C(Z#7~5Y-Y^^0MFq3@ zPd(U2j(gZ9_xm`q*aBw_CF75Ox>>_F1#C`OKHIIqXD=VlXK!V8u->`JIP`J?syU*J zM|7;9VuEXZa{=l4XOru;tlR``Hq?D%QZhiXE59;>`X=py7bqNQkXu z#b(^c>9WtTPRVB6XFZp7TXq94kJ?SjC8uNIpC+ug5uZI2)XZKzV$aS!p-obT*DxuC zw%pnEOW7)=M!f${7M}Iogsm`M%~l;wXYcGV!m|V!$`SoHT)?7dY<0n1wr=wSHhhjD z8+EUYO?~Pk=)2jHzilIoQf2}AVAjOUI^~T;u203;n!)V9u2gnb^(wZ{+MQj+JjR9< zBCO9Vf7ai40^1YlipE2EmP7T9mQQvbx73n8#kE-#L-3)GSn{GEL&ZpNJI&0nafCTJ z^<|jc>#ZZS z3_K*7iI2(_kvCftxDBEPMDM){+pM0=e%G&Kr~TK$?BAP+MJJy{A-~@c*REaM^@S1^ z^IjxF)&Vm#R&R=D^-4nNJPKhJA*f`MCL6Kp0gf6`VpT2Muu|*@id(AGa7aY-s7#TI^Fqg9+g^XO&_d2OT^^&3HS;Ef7a462q7xBsojl}UMmAFk0X(U)=TE~ z@FTQp{%jI(B9nYS2rz$i4twpkI?ObTA#oDBAU1sgHZhmM`&PT+ADx(#x!d7y@;8u* z=4#M8tU+Y2-5@E?ZsX_lBhC@ighjQN`Aa;#xwspr$*0#L_=?|CB)hE?x6ZXgswe!= zt%bJuRe76$b-9hxmu^S9raM5Fcrv-%IZiUV26+$HD)AMXtVoglF%rHxmT5KU;;xvs zlDgN?aE6zMogYWAT^G-@|K0LrzrEOl3$MCjW5vgKnfe|s=tv(Lm|?{&_W!~i8Zd?v zaf!&&YZG4RehP$ol;M=K9&4#~K(Ggt1IK;eKy+CxHWJU{g?MbixlcpU{P%aDN9_n+ za;_Y&zbMB#^0HxJu@EGmyUcUFeI8#e{f?P$POQN?8M5KxP3$rd$%+KBT)lEUnjmk* z?JrS>N&7oU^q3x5r_+rC({r)Yo>JCqQwI2oKOsYP{=~7d8|`UA=#NG=wqCFir|?73 zg@Bp((MBHLWV8dSbPQ2=j4b<2w1y2jE^7HWX9fA&zZWmNEV%y|y~n;4xA3NnQ9P*a zk1cjRzi#PtVue?@EgN#d;>jgqWWSdz*W$y>UJHCQEP>3wuY&cQ zRM<={7qvjQK{*}*8zuCr`=%>M)(qJSN)k8w}=ku0N2|@=d0?>gm1vX@z3H#rx z^{h&aBD=(1-+Y1MEk>a;A1{&miM{)s@wa=W;M@`kwu0aEsLpR*diWHOS2xF7PY&XZ zQOj|#cmnymXbpaJpHF<=<+~4x}YF2#iOPVaZX3-Mr)tlIV7Tz{>~N9eYjLdDR=( zi~i}vT6Pk0G@DG+j#Uuu%K)05xPrHKg$s^ry>R{alQi7%)lbDA4eS=QO8OS}O6G&W#| zpJ!vCdExkJ*;I0jmxhHeuqFFg?$_&o>a-wDH^7Q;9z?=)_w z7fJk$V9?%H4A+}f>CUra)Ge$9%J0?U$dkgb;>Af)5;({$U9u1_wY0+@rv2jn9d==g zyzLS1OD!ncNwYN{B3S>_GO|?P7{Y^mN$yD@2whx@XB~Ejvj(GZT5T-|$th#M276NS zGzzx$%CeVxD)FED>#(#ZChu;(#c7r6*~;{}c=pnhFeUmiJ}jHfsa3~-q013iouvrO zr(}UUThN=DbD8ILF9n}=%|*KyXHb2)9k%^AhJP(pVz)A349{x;$+Aj<89P?uG}*UR zKLwtjm5mD^WoiJ~FdMkiA~~GuG!qA^*urxuX*#>TRe+YcM63c9vNq@vaKf_U#EjPRo;8zIxtk$Pk+(q|I0!B*6ibd@DAJ8Fe!zZFY= zg!s_zk|-#e_YipZl;D=yE9h);pc~(b(VJOMaD4P>RFEVGosmtjAyR~%8Vv(}O*Ac` z^40TUW_A0J7ksjb!IO6vLrd{|ZqHo?pI2!^ag__{l#EmKLUsymuk?rQArq)X;zhW7 z_APvs5(7hFJvx*-6SAtkA({%q6YdN#*?bs<-&CdrSf1%$_6(jn%2BrvWn_JC6;Chy z0^%hbz?pUDxF!9YsOx|wL|k~t_c|ouj4 z9dY?|c$rAGF7pS3Fa9PHdfUk2dcnTMm1VTLU4>dSY=iJmCg2o2iP%s6N{*~>15Y(O zc=AE0+P+7s+FMepIS<-U^1iTwou^Lk*C-8#}Sxe^Sn ze1|VLBY|h7z@4wl5Fm`j=$}mr=y{YsPkTcg3H)9Gx35PLmHn=GdP^$(B~n9QH>A?N zm8#h8>5rSi9};8yV-`r=^@9na>`W2RTZ*?tJ7dUVtGrkCjLLsGO|&WJo;FhZJf zBiSYOnSXxZEfcP0#Z6V5$fss9B@i+XX#-F39ahz&i4TmIR7#E z_4pH5xBftxiFvT}z!Uh7RPlCRXyW$e%0k6`MYvP;k^JWnO;&VHBipz-bi?2R)3)y# zcub!Oyq61*>D&ks&@=_+o(YFn3rBH&MjlZ}E~2lMcafSu-pH*z8)3I zBz=tx`>o(26I29lZn-70L{vO!$8Sc7r*Fa}wrabx{e~&*gqL}o&9AfkV8>c=ZtE?^J@zO37k!4l z>y#y_kDikSXLKM;kh9r$#1?6Y&qCh{J3&^Ew>z`=46P5FPMyM!Qk~m@)GXWx=hR8S zKg~N}W1wGk_*N3x_By`GRe{6n21*%M*^lT^oGs(?bt4oWj$*t`8PIKV0GWL%pxffi zDn}f`|Jn_>fmzGAfMjjr{YaPo_f(vSuZn=u@i@A^E0+7}<%u~bZ|WbE$|cVn<)?nO zBl5Y*XpPffMpa1)Y>T#H-_JT^<-Q=Yzf}Qx*GbT29pThzaRhu`jRAezNR-v;*fk!m zY@O77{8Y6E2PQx zs5|ZlyQv)@G@Qh|>+*zwmRoQ!cLu$AT$sw`89@iT1`7AYA+M$cG;$#Wtjg|_vZg>X zyG9i>c;;{|CKcaAykF zZ|P2qqu!9p^g{U7?gW#hpOTmz;?Q_i88TKz)3p_P)m4h#)$A8@c_%k&F5=*TPq0Q`p&e01d?UA*nSpLHu(ftX!~&q=xpR z2?D3mBfnUfxJ8G4O|+)IiV^VS<6Brr+TdD)D4e=&Pm8u#(KCg8BxZLYl*BcH-jr~p zt$hUWl^u7{~~Z~xcq<}*@AtAhyv2j%wl5|vdPWg>tPyu00KOo!`7(h z;JZ|eHcn3mGI1`dww#Q6#ycTCyA^6DyMaPwJahf$8}j)5W03kZ0CNup!os)XaBbT@ z^80TvFT!|S0Ek_Vo*xW{YftpxtZ@i4l=+0|NZ^xQA!o>HheX_Iau01#IRjWS4uUUd zz=FvIu({zA2>DGwbxQxypQ{IGU-C6-aPStc4w%TklupA}=n_UwZ5@6wW{Bs{iRVI6 z8aRiGiR5;k4bNxhWK4TP@iQ$kwj%r)Zha@14V!Id{Q7%&KSEaG@#RxdmFr64;uMQs zUzyB0A3BHw{M)N;7I{_qL@41SH(Rlxrvw@;495l!1%8lS$~del4`+%0BKPM=!ST$~ z;Br`jTn${veJtQHqDMAjMZbL3X0V*~YMzFpPEUq$`N@!W>@RuyaM86?&4el&-S@nq zHWBvX(RMsF%EiLVw-bN8@*A%)d4ivmxPtJdbWZ-(1gaYshut@3;?h_L3kAWf_SIl4 z(z$;TFAtPspZ(HjCrKKyJz_VAyg@#hcOe|5{x(M48-@fqDkW^c^aXnCm4tgr-ty}T zO!48a0sJS8j~b5J!Pz^dXjW!Cc`#9ngt^Ic*9MfCna(=sXJi4Rb2yHfuc^b%YNYs_ zL;&LyH>TqJ$Lq;ojeImGGL0Dc6HriMXt-t<{3p`V?pRh+^-6PLde15#;7hWy-n){^%JEC~2oK)(|j{RF5s-#Ftd# zD+2#cV6zztR*XX`FDh_C+8I=_YzQriQpU4Gt#Hk0DOAZ0lhzRLCo+7={9+T z=F+QV#;S+tx!fP7$-0ec><`0|BRb&H@dI_OQDJtbIiTJd8EE>Y8I1U79DnHVJMKTH zjqLFkTCCED7MANdg^e~m!fdfJo8|WzMODTN@}=j15x6NQ?Lb)mfGt?)`_qyBjdY{k zIy%?%6*n0AfOsm#5w~Z2;_;ZpdVhql&rmv;)l1XdN$aWKoh0NRb_4GTHDbk7W8txR z4P?fhq>8N=qVAmLM~d46Z`_`<4{v~F%5rp7Te^N6x-F0p(5g&Yi2fElaq^ZcAs zsC!lpodf%6>V#5~cs7CaU2VctXG|q3cAHShnIB~F1_YA2lHg$NOD^qrLxe2mfyTs8 z@<3-k{PmiQV#-}`0$ibfrsZ_T#T1(R#GdYdqe`{cr3&`6d}vr<3N4KHp;KKvxZBP9 zP{Z9-D7Us5dj6e&;=^m9CgvGbcuG>^LsRLxv@q&2N18S+kfq|rcY(e84W4axg&}qt zS$s&~{1wcqpF3p2O|BC5q+Nog&VPw;Ogyw#ii3li6^ILJgPJG3FmTO=DNlY&^7AGj z{_y1L*Ofxm=DTZYtg{0>D5w*!mb(gWWHwl6zz2~=c@T2lPd6xB2h@@VPHq~=szF}h ze{bij?s-6L-iAPgpB;3;2CBAXAAP*<3jHndkBfhPkvs`dVnSu_lPxbJ;mHaej0rXomEoZfn6ISDr{G-nFCl*Tt#a=priCU_b?^0XTLn95$yDcq&^5 z+m9*W1)rpeXrvKI%9o`){mJy&Q6t*-QyaEjZi3f3Okr=JiCtdT;K=<==XqkQv z?zhrlKQtdlOB@}T=!|PTZ_mkiMOqC0w8I*k*G!;}DbA=_Fb02pZyK%|(?ZHwiRhA_ z3~k1&K&n24AHGS=aN9ygK5@ zs}{^y14kcokJ?sp1)iLRl;1r;k9!iH+UdrTO-}fchaPzza`k%FmRy+JB7-J%r=sNl zj?x1mzu{}`d04sKm{xY!(P1)|o>^T1seX4M!&MrthTr8)6*cElWp0y^>J%6`>`Lmk zULt!|Yt!y)FTnNmBImhb8#pJdB8o8w{Jk%KR|8K0&Q@%xCT zT+pQ3Rb*+)s&KrzV-*NiPO3cXhZ#q1EY!A`-L-6uNqp70s{0VVWGekiIv`10fw|aW$|(NM;ms&} z#3|{C!%L)3v~sn;BQOWgF20S2Tvd>o>ddP0eZYu0pMYPF>KTd08vOCj>G=I_Lbit9 zASdcXAwKgtoDkhb!;7E7-m-gK*s3z*Ba?uq4~D~64M{9@;~4$^kViZ2eumn)-FR)p z8vOQn1gQuSMN*R5Xu^?LeBI549djPQoS<(y*XuLA|E+`8{%odRYH~C$;5Gj6S&CIL z%jXRLj&pfT1>^KClZdZX0yr%L8sF~H#zR%K>*ys^(X9j2HIB0&g{VY(2}+d^c>cEO zfnunDkr1&K+&y-CAt4)Y#8)MNb{Y9X6c zLdM!Yhzu9!zRsORY{J9vnY>DR1Xj?Bwe1k$EXa88F~V+UJMmPZF}m&M#OmJfPLx+- zOr{0>v?wc=n2kR37Li@5`%j@YYQ1^y@uxjNXMl{?=eFf7=J84W5`6 z>WDXQuf-31V(`w3c3A0@1G~L!7M>yc#!QMcgV|55K=zLqiRQY<;7~D_mK((N*(HN+ z`vSOHMF_W7L@-~KV5^qDK|TpK`0T(4X)E+Zg$q96^=;93H{OJe*Tt}}WLns}ny=Uj zZ+~vWqiDQ7@gX+Y5ltQR{At8^8`#R$gZ)@B8G0=TnXlX7^pQXskuLZRbE9az-z?hs z*`IFRw-9d;)C~_J-=Vw9bCAo|QL;RF3zGLpLaL39c=pc0a7@w17 z-+uOFZJNH|S6x%tfqSQEtjHyLA=Qm4-#)>L_6^~d83%CaZvo5VP~iKQ%qa&^3)V#BI0 z@Xvk?7$#ToYlc{wC3X)Avi5?fUIo(An@_gx3x}}3KcV!rC_ON%7AC2MFyGs~L2>zS zc$w3O*5(8Vn2a)by!#ueiH#!e;s@Zepr-eDk%re(h>O2X+SHO^?s0LtEZL3rwEUu}=jSu6J9pDV^*3P+|2a8d-~=6F zMx<<2yv1ONHn&khz&Lx_%$)OkfGRv3Sn~}pP<8t+Sj_uD_imkBedwrOHE8U?U+f(4 z$0cuYUg&%p=Ke^qbF+=k(OnONKily7^=jy*z@=FK=_h&?ewnL&dYc6HhY|DD8Ki>f zK+w^4qCTuc3toQ5hW?w;FW<``weu*omwE~c5wX}fcq>vXIEp9Fvck_o?~=7f!btP= z8aV$*(ECXjL0<2ck}bZ`*z;@#-m~x?-Y_u;rcW=!v$iCn#pN0_W1KHIFEaG)>sT06 z(I;j56giJdhCXv#N2lN4N_8bd$n)(rys@2k!HLy_|Gp^0gn$1)_sk?}X7CYhvT$NA zyxlT?S==C8jIM?$J7&X`nX$+;Dhn%LTgUwIG)I!N&AEpIZ;-y{MpoM;2^$)+fNOQ& zhGrXDVLC+s;O`uGPEut16XHydQ`X>nm+f#!f!ptfgX^vT~F05z3xXF>(;oKKN~k`qee4-xeX z%DB9W!Kb^$uo#y{6^W#biuMn4|@lag%4<{%|u_6nW5rdn}C}dg*uP@{#e|pIg+%wXH z)A{9i=dX?lZJt-Zm70O3nx?zGHdUCA*LrqAmouAic;!F?~XT+(oQ{?RiDd`AStkN zk1k}H)I#vPDRj~B2FO2einR1Lfw_e)ZZ5PYMebMmf7UC&!sdF8Q+!SgzgI1uIeFa?l%~VpfcV_bQBIs|Aw=B zzahgDk2#~a)A0+B6h!g{iQ>Nxh&}d}*Vkf4s)ojib%doA9i*JhW1O8W*RbjJ4$Q(9WO`;xL%ODRs$`6xR?Y zW8&TD%65&NWFt)}yYP(? z7dpuchwC|`gpIju)Ye+oV#Rm%g0vKvyQCnQ{g$kwq7IufXBHc2X~3pxGk8JwDxAMB z8kc&;3UZyxnftDH@w}21Y*pDr)*wh18^}5{4-X15V%yAl2aE%WW}F;)v)c!r{!L_a z7EWM;T6*w5`9*m9lb^U=Gy&gv_#5r`>#Pf5EOw$&Ka|d>eqAYR2JiZ)7zO3f9tT_ocs8-Py!y=vy}B1UBJd<8L@H9kK>)Luo8bF**7Q8vy&zT;Js4U`R>zYNx%d(oV0<5d6qNS z$(N_VkM&bvrvI?O4}FyxTULyM&&1;*|4@|fwU(JJQz-C^&1E+I=wh~iMfi1BEjk*e ziVw+VlR2_^aPdO|BoZRpD(c{*dPA5OoMf+mmEXuxWe3+pOCO3EE@Jx&RxM@GUtg${UF(nk6vF%o+y zLiN?U;nHG;_NuQ$F*?2EBm0Y}w=acBm#=aU_|KWs0-lD`&1>M2ya)X5CPL)ux$vMv zgT|jyslKJlt1kGsp;~(^9K`mR!^BJxY}nHPzr{+)ox9$AJt=+qFlie-xFd>Qd)H5< z`<;X0&I_=1OFfM_9Y_6lcG2}$O=$mHeflxB9WIINp~X*@(CRS+;{^ z(Apd19ZZ7K`BBhyGzqo;$|4hb#Hi=W>mVC-4x-lzYI%AA{}w#6XG>c^$e|-tM~x|DHNw=k zxDaG>Q<%2x>7>X)t=i}1IDMc4zQ-mnla6o1EL1XRM^u4u><-2y&0|4|r2 z)%s7@RQp6Zft0W#SoTh#s7HnVy|o2~Uk;Puurl(sb353Z&L%qH%MsHs1Kl2)ONpE3O+B?}JYl32N0yn4#LB>KD{85-nQ z0KfhdzvP$%IlDLtLEk9jJ7k82D{qs_QR`8cqA1=W+X}i5U%;Jpn7hAJl|G*Q06t#J zAcf0a`9BI9$iS;lBQZ)IF?MUwMV(x+R7bN}O24@A!VfV^=s3^gWaq4N~ z^lYY*EYoLF{g;UB^VXtp;g#Ht(P_X+*`RtURh)C{3K`N(=RR+K2JS91 z*g3LK@VTp25bl1B)Qs0M-haEei3g^_-aSc>`+W+f!dg&pVjYC$?&9#vZz!a{7~M_p zBR{My(fYk|*k^_}Ht^kxQuQuk5t6|PpFWL7tlx2!r70FaruL&Mqvhz@wRUoAQyRDC z{&rm8mWo%ukmQ(@Wqg(Pe4gZA9WH4`J&IMlj+VY!K~{UthJ^`h=tq-nFuL+9iQKvl zPkUX6EzUi}50=zo@fj0X+drP zuq%dy(6=mOX!q|!W&I=Q&Gj?rwOu6ouKkO-bV7l*RAV1Any`@jTsnw*#qXeJZm&q@ z#JRLjI*lxjoWLgU%ci68Ik0%U4Zh6&K?iR~!|&2;klb&NHyp0PARCL5?=|BuRg+n? z$%`#A6Cz=aMz}N47N@wD!ISIfp~uVtTD_iNh2=*0a9j*m7IXpMYd(ZjV!l98?+ILb zT#`|Ky%q0U7RSHn{Sy5(zKMF?jN;&<-2@&A!AMXHF+1jtuU~%($>+Z_UNiLBpVRB{ z#YktgzNMb|xAh@e=dOq5t`D1ER3waz_nbmMq5^(wE zy=UO1+#k5yehQgy-HCZFJZ!bl0dGFYu-%Ji<7@x=;E>7~4*YouUp{EX?zd0JF^f*H z?OV>_V!LkA-M;`YNfO2~X63j@!5KbHj)kN5i{VU~JyOY-LaGE_6pbs}xECIek&28H zi659r;v%BqUv?-qm?F+j(>w&VKFZX&`Ziqf*v;K*n?j3pE$Cf;Wk`im=z6%Eo!Fj< zcR4(T4w-hG6Q)D|w&=k7Oa-1%PB=C#BV=UOC2&nuhQ#w``0xQs*1uMSJy3a)E}43R zuH&zu3pfqxn>PSYp1a}FpF&XMeTyiriK$9>a0XN!tKyfT_W0DxPS~95L%S4A!1mZg zwkY(vpuW6`vw5o6T8W38r4mr4AQ$B^Z860{C8$94EjpOuN(UB1(#{Dl*|GRy_D!HL zYnHc;x<$v)HMeEhhG-2C`5sCR9uNnWUy{(ZU0Seb6HN_rCkfoDnsCW{D*ZlcNNYbo z!b$Irv)Se$^x(_=FzfSw=vGHG^ePJ&1fLgxS5_B9qEGNOwG&Gu?m4+ymR$4)Fz9!;FyONNX#8!BJY$B{)Vul_B@`!(WG&*T! zL&kTiL&~eQq}%!%xy%_6u5Ts#l=EVX`&nwcGMhFUOVgwuD{AS#k??M47D}4!mBG zZ8+eB2sWK|S>PTEB=0m{;qi^<@Yd)QTvFxE%F9NwgR@VtO?RTOi~Vi|*RUaW)?5mB!Z#r{J=Mr%1k57%uPgXQfCuEszK z#y2T2#utTgIRr7;A>D{y8NymA=doi>Calyg!F#i}99vJ=f(EARVh~z^GD#W{l?g-I zb?=!SGiQKBf}q}Y+>BoE-tne6^`NbV$r$Pn;_q7&*&APHvc7|3*knaNPyN#cbau%$ z)Ym(i-6W>V>c5%J>b@4xbsmTFf1ZAYx;A;>U0VA1hSX7%=+i@fG+*JgZtvzaZN4zI z+w{4R=e`hsY&BD?oP&Ots$oyTtUG@1A(Eb{3=8kyM=Rt6vzD!u_?>({j?(ypTI=l@ z-iwLoR+}qc?KBCWn(-2;{JlaJ-JXC?eg}4k#R&V*UBt5Joe5iS<;!lj3ui9~_WqMq zB;cA!1FzBhJ{c*VgVhULaQg5qeAGUN2%WAVNADaV3PP2<*syZa)}g?5>Itz5#WrlF zjR+pyTZkSHKSlXj;@IQwEnKo~6o((M6!2c;;mGJMeo$-;^D;pnpQ`9W>OO1n4=X;B z)wTeUYm4!G^PN~k*a6$Ea=_0Pg`(qGLhQ*MibUo{EHm#=E_dwaT86(l9OtC`!%k9m z_;6G!F1U6Z^M6I7-FAn#vI=3?(iV%?4LwDr)7RsVnH3m)T*G?SOR`gV_tEAS2?$*C z6ow!F;(gd`f~5D4BH_3kOkxC{UMo>@Niah`Y6;JvI9A42Cp-U};A!*3xZK~7`0jgGepdPzxA^@aIal3#+9GyO!Z!BLlrH2Gtc6$F%)p0ocfi?8Z;*AIJ6tz82bAyn4fYt?NkaJ3L7AzBDjXdO*^7 zj>twMoCN6h!7=F`tSR8nY|P$B-5T!WuV3xiitY7+XVL&xY*t{;%u>Z~8)fh|^||b+ z8{0Y2y~eP$q>e0f2tnF{y|WdI#PIj;-ywb4A4n>cp%A+moR_+z)}VMStGf~29p4Sp z-TXi)Oqh$Uyo1lE*Px2Tt?bb$cI=<0<@o%KaIRVK-_2K?3nG5Lxbc<;K2SED*{phz z^9U0#H#0fhY$y*F*33cMlG2ODoKkHg z|4`j#=Ec-IsP@ESy!rNL^jvEu{jdBBGJkoClnHjO{ZI9yncphWnXE9%n`T36)P#OS$u{8IbVe0|XRKs@^p{l_sfQ!XCd1@U6J7_{-gVeAMe5(=_!5${!Hr+0O;^ z+5aZ06$yn08e-Jn|0sE!GD1!Xi_mL}>*)G(O|(Oq19^dyh*u!Q3ma8}TMy}k$%*Rv z&Z71A$3e|C6j~;%LieuKGw-ZxxSCu^GWMSp`L!aIIizuqY^`&mR$7Db+N+yzc0Ty? zzyMyRmw;23YtxRNItZB>N=~NBRx7`JPiq&wps7m^K~1X{ebl>^I=-z#l6y{Zn_n+r z4AQi*G%uNZC-4-NoVKBNw%&neb_KK&C%QdqGi{MEq9?o@s&vJB*+It+R907|aDgDN9A+ETESDqm2dt(|L$*c>$y@zn_aX;{L* z#lQI61q(K3gG}{<-Y&XLv4}q7`{BHUd~DpVOVS^9lSebt@Ig_Ab-S{Gem9;?gZo}V z!@y;TI#WdktuJGE8N_P ze3=h!(r;k)-1~|;3x%-a+mEo*F_rN8SCG2CN8FEKbLu*jNLBoN;lt<8aB_SMEIbzB zQtgkZQcnoa8QK70;m6>8n;upQoq;o~c@TJEE_UA^#`J4Hf-QlI=($%a)c8CH_ay|m zy!ae8oHxq8sqSRQ>b4+hRud0D+l`(6Igv^JndqLMFF#E756Ms-W}5BhGbatNk>)>F zU{J8vmp6YoUD);zLe(*K74YtN9=Z(U%NxNZR*U#gbtF?(Lo&O3eP@3VM7BsH2~!OaNXojKu*j-4wP_-l|PAC?187Ll}p zb;whw5uH(-1Dl?T(;b;vL_Sf$V%Ljvu(Nd!C-#0R3AIb$HU-O({vkc=zS#`lb-fLG zr&a0kYjL2S`-1E8k)p$Ut!cJ^SyT7I2x~<}6E0nC^NFRN})kklJT7TbEBoAl(dvIC{YUER9fnHet&!U z!+4!@KlgQA@ArN`e^#*gINPhjn8&j#G;(+%xaTE7^z&IX+uDSG)pUxVRBA>}+WeJ1 zzwwJp7#$APgKj~v(q2f=ROOsn1L^!pPienpYdD_)j%388O;{Ec0(WHRQKP=6@uQIj zC*DGmsM{-P{}MgC=97%}Is>uapxtDHnI20P?@Ob9&BJvSifpgZXtpp%Ae$Zc(vc0F zT<6JfZqcdvI4VG7eal<}&7VUdYDN*5yt2TiM}_$D+$zkz+byj77E8w*&mt~ao>1ss z&uLr;#Jm;SY<&F_Qb4QeK4A>=jjP9oNp)h^?Gkt9VO5Ft?6dSj`!RCm&`|!{rk%X2 zh9Nh6;0;0hNu8wd)@!VKw-*;0KNS)~0)&BEZqcl>rL<*fr=&^T>|fUopxpD9~zZ|+l)_Sd=LGM(_CAH0@T{fPmzVL!e zX{lg0*$JZS_Dfa1T%~Hy&cXVzclpux2HQ4$F}3x6dKRUVeIa+xTiCfVMru3m*hDw) zoT3qarIN6Xy|hUo0Jq;h2JTC5!`C^1{Fr~gVEUS`hoa+8p-*lsFJmy0|1L9(chwoolfA%ST6!98W*p(uCWrC^zD|JiO()>q!C^2i zR*9bR+k$Gd)ma!0BM*+9p>H=^gUVYB(m zDZeQ3?h+kl!-nw<2G;!BTa~|WkO5s0yE#Xr)AK_C*9p_cKD9{~~4g3CP z!{wp{6t*W|-phaJJElhL&j^ro%Zv}Mxd`?ikMS^HiX%JS;KND-e#FLM{9&8n{8IVD z{Dl+>8KRrw#c54G?ov=c*1^nFS?)(|gcD~}*Pkw2~ciyW!Oz^e0B62Iu2^Pm=mYCOk&D4P!i^;fl zR64qj{Y>`^y$03?CvZNy_t5AS6A|~HCfes^$=o@G;Ps{&mMH#(kYl^arr#!TZ*mh+ zUUmi6FGa3CRt4Ai^@r`}{v*)~cfqMia&+;DG|9GlBdNXP4*Wbd7*;*IEV{_^XhNZ- zl)G(1R&@Ro;*?#fFg0JQvhz07ydA=ydVC#ztzUuhEnlcy&TRO$BOUeTN8#aVn&@U9 zjGtzog!o@?Vd~f}c+9lm{*3YDnddovw_7u0F4+Yw>uTtV6}C{*a~jr*S=E|3y=3%p zbGYhq0JjgRr*ahcLZ^bZNFfB@edf%QV)x}^-1lHGTOAog^L~OKrb5@i8+s!aA%I1xG&BZs%m8^ z`6t7Yby9It**-kf`CC%{UZ2J%JfRL#8&RX8nI2l1Nlt9lM$?OTrLozH7%;F3`Y4`) zsQ2@r{>lPO-I7Yhte!5LJplCDvd9X>OLX3)2_iat987b!fPz)_&~bDqgiaT|bGgpK z(t$&1=CfhA{)`3&D8>oX>PO+G;|-F(PsP2}`8e9p7P?uzL zAv(4UqIEBjqk$%*{b?D`xkT_${@S*mf|U4(P1j))if5yNFQMOw2>8xPLMwOx~KEIg~-$|JVX) zQzo-&&%vOEPF(PP2s^X57-IT-m+s{;pl@%AY>W zeoqK_o}<7&s87O7zjUU#L8UA{pcM;tjA1YJZs0@`CDdrn$DyKOB{sB+9L|cvcgq_v zwEQwwiucTHygg60Mc^;ik7dtPW53cf*z2Fi+4cFO*tb;)Xt}=zzAwKC3p>ODyj?mz z6+5+ZEjqY!P8Ck?8OZJ(6uDVhYuNkUcFe2Z2g_Q@n90*n_V)Eg7FXwh2}`F_@A(N( zA+oJbt`ONx+jP)MNrhhiBPHgLNN!CUOy678(?9xBbh)C)&aRVTV}g&N?CO1J<8+ib z{$}*E(J@+OV*wG#e&WxZ&!UF=u@51asg~kE)@)Nn?|Dr@Z5>NIZ#WWV`D7*&9D>6J z+=1mIqv6M;N5s%}2UbcC^UE@l_%j|3{5!9IT;TOK(%gFw62KVdyr>YeGyjowzulSt zi-k<1O2V$LKEke_3S~#ik&RR-v(YkiWncV`Gts0#vwXd&&RZ4iU9b&*sfIBBuz05X zcM~itHbWm4%)S(dviR|~EPu%iHYrLXGG8ljl$RIlX@5dqht@%7eJ@n^oyzZ4y@ftX z6WHIRQ4;IFc^LG&FB_2(K=npHL$4?;c1k-D=6KBlw=PE(@-Z6bUN}cjZsfT;E7!7u z!bQ}k=OOqO1(L*kBXZHx82zWJfVSo(i0^$SWQl$6W3y{fX66^P^bwmc=NwThua3lU9P{+ZDjA!6nVlK45e*;h7sPHXgmeQaP*4*dW1vJ|#i{Bf4l`on4 zhR?I8g%dMZL*&g%B$clf?(I1(ne7n5y?iteXY5~yFILaR?{Pyh)gu(!(ig+a{TFzZ zrY!z~V?2MnEQXBSz8)TyHjsge9-@1r53=EDcC~G;>sBqTr9HHM{xWf?J>O8UO#@;usmK8 za-Prm=>`7+L^rYM!dv3`4D#+@CRfd0Lr(iZ{C0XPP8j}Cdgi5o20F^{nVrVWJ(hTN zLlnwCZlte@EMQi}Md)`?E~^WOP4mUqDIO(w4f%2%X^SY z9|T!L*E_;@s7CSAI(&KMVXC|X_g7>eEAxMT^~bav2XZ;uA1Z!7pi2FXa7x!=v@_aI zN^;AEUwg)JvlGt=itFw`pDCuo>q&3G@%a`&{yMiV|2E%0QPZ{~Bb|TwZiy7*9l&Yd zeR!Xg3)X%^u{%hY?elw0y!9S&K`W=D)tK>oTG&mna<~8+8;fDx=NOT7uEswf+(;(d zgu%ujk^691mH##=8gDgs3cC*c5x2sp`GHMo{Ne$Myx&CzKh>s-yP^SLU>yp1XLE@2 zn<5g8n}GXTMXcuEg~V%B@MUv0M5#2WDqrLdrGlTdw z3wP0dUDM#pz6!B_+h6RWd4Tn(GY~j;C_i-9dXnWa6P|@+5uM9*`~elpm$nA-+rAEh zH9+_w<8JV72j)YdtTA5{Uj!4>bHQ}{53*ce83!L&By}2NL6^AZL$-k`X#c9=p4A({ zfapjX+HVcLtuUPT@0~*bDw|@=>s(wi|1Z(J6%IRdzrd9<|G3Gi139g_10>k8kmgRf zNtPU#OY5G@Bug$tiJ`dz7-ZCd@x1ALR)~mlwLOkLuQh1p&3+J5T*i+v8Ne@@WDG91 zT`C!;a@b_Ufn?YteAjl+mAzt?{%`t zvoHU1!f+V3`wGC+5_sbhK?=5RhTs3Lz#Q@S^lt50VsT?%nC&owlD}P`n@W zZJki*NHNt2xJlaWjLEh2xp3p&H%>v0LF<8FzCyDBPF(%Sw`O(oh3m)i^#OhOyuP)N zAhJIdlWZXOwm7IQ2&&CDyo`qr0cSYPS5;OzmncSVe!5M)N^!^-;jd=B-?;uVrfnwKe?e#U_rE}X?KD)a>(Ge0h9;wS1{ zR)hC?&eLmMK>D70k6RD_LyeL&G!sAbhJ;wbsH)zQiHW*`-DMTDoncA%JJCY)gdE|r ztOHF8xz7!`@`~GPH-V08OQt=fMHpc^M{uc*6_h>Fg)xmwgpTzE)OBt#O*rvP$eHg$ zZBPA{tX=t3QgM3{{u*+Ou8ACk8wQ;wr@!y#!W+kNF=lrGYqw)WUv2PfJw=8|BwWP% z0PcXtMlvZ*i=J(?C(lKetFA`^7TQ{%TdzME{Jm`7auk~T#Y&>>rqHZ2g_!iIliLy{@_3gf(KVmfgHfL&)VtplJV7+DWwnQJcd@xm z-|7yo?(|zarS3l=T+<%D`OYLSislKudWE>-iz|I9NPn zW%3wFz)~-1@5Oln`_oMC?%Pg%W(*`_uU`}HZ5St6d_-AL){mw?b!XBusV^kQ-C9V7 zPdw+{*M=J%&_EtoRFORU@tj|7A^B0gnEUY~j;np@PfEX66RQI|$#4f*Wk9MH@Z>6Vz}x6)rid%hEc|j!7EQw@(|wz6Ne+t$ z!=7bn)VszPO_HNIzBH2hSNMW$#uqv!e=PMZ`y_a^_RvKi%yHEcv3Ge9P@(e^+V2$E zmwjCzMCLQd6nqfd3WQePaRZ&eBXo6ugxx>9o_%YZ%ckcS2;xzd4!pXa)>tM9V=}uX zNq6E%eVaZUYu%1_;sCSn7S>5Hj@~;jniR{kr=&{rIbnv@FnNQx9mc4La-Dlk160>DW%zFUcE1^ZSFd zIwKCv!EpC^JAHV!556an__gvGZkdvd27S6nq+Ap|*D5M01{#3xTP>(PX+^r1Cqaid zp(9@{1O;;`1|FNh25rouqXNgU^@*0OL~l6h_3}a0O$tnF=Wu3xQ@oeb+=sEJTkyEm zXz=z}KrUNugepIg)wUo8=ef+rZly;g%=#tx{HdZ2BgF1^$~p|{ek<(T&oPsnV7zoC zj(g)2gfc%XIpOylvU%Dy{Ng_X19j4cH(T_`$X)Ys@B7b``(^^eRXI4C_72p}^LT7X z2bx-h<*Oq{69LL4^Y#@S6|rq(f%v{?zlw@t#qi~7Lm1}mZRO(m{J zXEv;NGb>b#!f^G&%ttYv$rK!8@zZ^|%`N40UyM5rYEHs8*othN6ql_YMr15J9r{A;jAY~5LYg`^gt0Bl*l^vwbmjFa zIP=jDJa~B`)?^6m)zfqqWRb~K-qw@12Aa&ST@SxTi5#vW3V12&1+~q$mKeuYkc7z> ziFv9TYKi|}M`xSh`LXtFPl!KLZE|6$54Uj3EbVAXxsq@`^cbZ4SV{aBeB@F_MB$&N zv%;i=S@gudtAeoXJExsoDHzJdVBfUesQ5w$NA8?V?;9<_dO1d=cYg?DygD&wUkdh_ z$qE0q%F*$a=U~^uUBZvS2TRAU9!^cys?gzepNaMdEjnXsFpd(Lw;z7j(LeQ~*X3^v z*CbO;bDe^TVtp9+U;o3k>hESfpS;=lg*xn4sypsDnSuTL^O#)e2>FNG$ab$y;P50| z+G2H|TRr|iZqfIA`u(mNG_F#EFNcb7!~D}|sD7Q>_E~|2onDHC6K2yE&p|jA}Kfx$pcXYMiOMAD99ljw}>>_Hg!1YIH*_1TN-QG{q7;q&O1tXx# zET25{P?IKpvyd#jR%vr6|EFXSpUNe($u#Oi9nF|=S2}TlGDLL-;X83J7?h>Tmj5ay zL+%@rwweLdd2t1&!h2D-)ds8a(5Pxt;=Cy_-1DX!A=)cG3lNZ_-^dA4rqK2r%ebhZhIi;kFb#=3o~~4Vud# z^M!zA;{x!4Y&$q_drcblgwsV^d15tYnPhQZ1dUzpD0po;f`wv9dQMukM1M~_O>&=& znj=mSD3|ApP=!w+LDb~Pd~6fDy&UGsq4#;S3w3) zq&5L7r#p~8hx+5;>)v?ZJ(=#&enrCK4x;;%lQeo!C$)Rn#=T$SL%h=(>C(i>IJmuz zu6BAPDKonzM8t$jx04*Q?_Ij!ZES=+iRsW^?l2sN5ZIp+jTh6#VppGik^w=N!TF&p zceOzt>Ze%Iu<|Ub)!6_+6aV3d!Ts4vJ9(I5(tr_D6!GqY2Q*IqKOuJh4tkkXQ@^m8 z&`?xFeveKd2RV+vqp%b5!?I!KYmpN*uNR33rCt<$mfVkMBR@@5h{Awv)V8{t(>gUB z)cPL>w-0rwF*^$WotjLq8IQuo=kKYDnFhbLEg$QoL*c^x>HN&O3uv>*zUi-XmtCL{ z>~6FhE8jPjUhVeA9E*O`!Qu!yimr{QZ>?nVg<#hFQtYnPjmH9N0;&c>VaL@fQYgCW z2Ly})=}vF_*Zc?;mUvQkZB4$Z?+H$MW`C4f7f)RW1hMrBeVB)D3cj_I3KK&H3tnrt zN(KdM!E@0Qwy?>NHr?3)uIhjW9MIs|)nx9fe()t$1>DzIN+xA^3qjA5Y1h_3U=4wE zpy3|;lbl3rcdupc%8l43oP67l#~LZkog5Yjx=xJRassKU4J^oOx5EH3tkg+~V2W~+^` z{nQ=Be|9Y*&28_o%xW%9xqlawd zwN%XcuFLr(@1Y@GxapKoBc5xF4{GoUwG;V%wY8*bPqxt8=d0kVYek|06-illnjjk8 z1;^e`0!%x`?Gd|Z=@+eV#)e?n^GiW`RAgqifBi>(S@l8TtQ__qHxabc!@)2=hY!d^ z{>#5ozM`)Qv^GlN$Gk<{`Gi=Z_lE~{zGo+-$W4P5_ zTXMOV7jD5(`4n(y51_RXEp+bb9QsaV#Ex*Nf)ay%{Dp=~Fv3ip)(mVWvvO^*FPkB5 z?1Ew0=hae=4i~7@^AzM%5=cy|n16iCg!oSj_@&|R;qUs@G&7&|OZ!+mOYdm!!$@KjD<9L=#qu26=LCBe8&{}Q-g9qP%U4e`EeogD3&c%%% zCHBwU0{3%<#qn^&xCM@WE#NM$Or;@wJLwtukj|Uz&P_=E1rfO~A?ZkeE^y}-Z2KU3 z9qKG#^U1}$zM4L6ky}S*ep^bv-OmIUkLOT*WjkqXmIIHB2>84l;N*#CR6E>&9?4M@ zj;&Y6Y1@@(;amgyZp#}u7(J9%Uib_?X2^q{(P=3AGJ#jTAb?@jQxG24;LpaRv}$%0 z^0|M|M5;{X_V*JyMUIj6#a$)AD?SMUITM7QgOi|p=TcmL!vUo=)u1r=A@plJ1g4|E z(*LF?;We#ygy=mdj;C(`|AazP(i<{8=NehOWH*gEOzG*R72xSTpMT?-0{Y`}$k(ar z63t#u(q%3>ATUc3wQmB^+IbQRABFJM$KFD+=vT8Jev=Qr`4xJN26M+guffMPGdWJY zo3V8cq9b^YJik{${+k^`w^=_z2q=UpH)Z(Hd)6Hc!dng{$N7V*9yl$z0QLh6`PDPUF1}A0y`H{A zdhGQy%ujoU3PB==b!{wR1?Ol><}kGPZh%FK<5+3wGj7m>OsY9N4m^LXfEAY`m{RZ? z^!`%MrS`gm{-NU#>Ux{7;smmMRUKXYGl)+8&_cNJNY_TH;rMlo3)**|JpU~Z1AaP@ zt$JT?}RsIhUS1ZBDc8mXlY;sWwL1s_^$-CIl3VF5Mf2SZkmLFO#$A-~VpV@|gSJzb%BO znby(YYLS%A3ZvRZVHoj2jdgFfXQ_YR;H>y|n7Q)+%zSkiH#?o7FSx~2dk{zW{d8w* ze#GFt%4rbOvscVQ9|}?DQ-y6VsW|$~OguQo35)6+`MbsqLg4K`#A0$4{j*dN*3a1s z2mBS-rv88Nb1H+qreSn_;U__1(M{+Pd&gQ^CJJd4UBcu3#az1v#k-F`;Qq#sk_n4< z3Hm1cg}&)ah|X%!leOeEB+nR+8x3-B$V%}Qd0jkDy2-QC59O(G$r$*AeI(1#67%xJ zXKK#4il*96?^~#(ZHn7wg5@B~% zsNf*4&-&fk!;FS(W(Lj!=&G53%0)xrM96rEOz(rgAs4SIaV$E=2W(|bV4996)SpzP zwEPvCzk4e739VUBkN9=1$t>?%Fz?<|1?^h7(na#4>HNrkLUSP#z7}bgZM^V;z20=1 z?X1>hQ9?UiIr22@y;+QMfd*`{B!xw<*Q77f{iube1Cd$0j9PqMNu{wVOhd*4TUtby zn!h%4|Mh~(M@I+-eI}BspGRmt87XFoQ~2+90O?`jpS$9C5>wQ>s52a4buX?8KGhm9u5~|&QZq-(&IstSsHHZV z7ipBpOIlX#$sW=emLct5_Eo51TI!Seo;8LrG2$mJaN5c$T%^qWKCmjM=XCtl2^daQ z@$SfRIC0DqeDHGtYpJbbdEc{{ha0d@=l9`2vp&p!Ng7U`QHdu;mC?o$o(LZ`VMAIY zxR%>Ohesmk@lfQ}ygtBhF;QWAe>G$NNe+KMPoytxEr{BP68O-4TPS!!x$_ez3aMq1 zvgL&aW!bh<%J6Uq?|AwaWO=5cM!gAYd4}?GJ5TVRYG2`5-h{31E@#`{RkNhK9oXpQ z$d9URq3^OyiCxN4(q_TsM|x@yfQ&qSBI;ym`HaVKS}u!8QlA> zvs`4(`BI78GkUe*9Zu!@vW;g4m!0peXUDpen7OY3t1O$u{?4~z4dw5-m-_{7^;s+8 zP%Cn@P88EC>6v8Bzec*K@Du9as%DS%rm%H8@37r#{xC1Woqrl`!UuJZ<8?h_dF$>o zyzRgYUd`SDFND@Z09(NKs~P~c4-4Q$n-ag{`3$_W(wN=XxsA2M$FS0$GEC#~ZZt7< z#zkXBkYz?E;r^oxC|P?Psur5@({>EO#~ux|`OF&Od*?9v_cF(X?=;y#AdNWa<5Uh& zzCXx%)dri!5rL3w^d6<6Bck{EC>&-RL*H$iLk>lu@N;nup8H)v(|*dK@{YMMIa@q; z|1ls3R;3Ud_^ zemW`t9#3xF4k4#H<`SLZD(p#@KX+=w3$oWXk+dG7*uUYKu)<>q#y9^1vwTB%;IoIw zOKo{Y*^O{yq%lA1y)K{Y7DrFyy`&Ep8FGym9-+>-CSs&sN9FcR;iA2V)14_wQ1@^N z(d;H@v~&ZDTdKwO7j9uPAB^$Z_$$6ILamUUf9;)=rYXSF+-t6)W%oPy5lsD;XK|V+Z%oXd%0Gqgcq{!imp0H)8G` z4Um;h5 zC!m7t8gf-+fH`DolM9o}@S5RRcGEJLP5QO~!cPt2xBSS3#SbVBc3VfQs|=aaQC0Z4 z`y}Vr@vqcrtb)+-qd@pJ@;9~an+S>`2ky9DD!0z`1l*SI&wtU_%>NE7B-StflCcL5 z5IfUqVOkrZeo0F7#b##=UgnJNM(zifs57Lhq7tqo?dHQ~+<>D>vV8v}9e&WyOPtz< z9I|)yYNC+f&MO`r%UcJ|g)CbYY}v1X&i5QK`G`DJKF|;@d@G@x+8v0SxD*2`_p`+9 zX7GMnF3j>3_g)_k+8F1oC9WGX#I8*?oS$gXMx0cpSKithdOJ8W+AP)~p-} zr|z_huE+CSd4(1y6O>1SiqZ-1YFfN=axhm>{Dn@~bBhc*K9LNyI7sbnpG!7sUa=1H ziL_c@vI-`SE~N=eW>LHs%DG9+N!$H3bhzhpcpAI`#ydt3lgE+R-%<4JMpe@^cV!IM zt`TM%$KrgsMq0V7h04ec#6LY_=}fzDVRp|7a60-UI%2l3Y4g@TFlz>+ds^P(j?3P5+IFC2O!cCn)WD~%MACO$OOFNUC> z^x$}X?8hm>c29LSVonauw3x)!6`sR6N0y5$p}}aXtHq67wwT<%`(DyEFH4AjGglIs zVPbQ)cNnj)a7%oLs!QfxzXgV?8igx)$3dr0r}R(F4f3Jl3`tRm#Jai5@$p(h-L8ex zh7=!;p@aB>Vvw11%B zrNc~>7qLYnM7(TN8`ZLnhW>?RFxqt(_(^ZV&hn#>ZEA)o=VH+_DGvtBPDegVhJA7_ z#?jncdUg60>GsV-SdyF%4YsJ_#>R&b%Tqq+m!6L+wl83X_eS98#cCMs6^^AJ+l82` z@|=yf4S8@QRQS;U7LmSwN>|vu#AeB1HYvFW`_x=U57)s=MbekqZPG-cgm*0Nv542VvfFT!8b4z94;-LJqSccYU=G}*^x*kO3b?sRBU2*f! zdm8o1b=ePHGdAq>Bz9#~CGK{XXH6r^(FU7^`ADtQPJ^I7%LWGO2f-WJ`}AnsGAx*? zkKZ%{g%J01T;N@ii}&9Y+`GUIu0>u(7o)AL)Li5mF8s-QjQ+BZXQuEK8M=Jrhb~AO zI-RUmRwQz(5l;oz(0}1U7+o}4+zDG@zIPP_v}KWgH^$&v?#q@Skm$FF;?F> z%$h!qVp>6uQQjpJ4=j{r{T2V>77asY`bLf&37tPN&A5l=>A1aD5(}W`71fW{So0ScFw4be+8oMvX)Gmrs`CgYW9% zTrJLKnVXb$*RDi`tFbs)u?&7KyboJ{4Cj9-jOK3$C-?{AHq=49|5E*^LB@YEfnW0r zxuLC>$j5gl;CG`7iPP;TSPveIpDz(OvS%|$M^`}Horf?xx|>9dxCBFDZOLA--#YTz zD)<bb@+%Se7y|3`BLxINwi}=aM*YLZow}S8SuQ0AalP@s7&!xQL3KExgn8T4$p9@4rj=#Z%t$r=IG9-H;z7b32ff>L7lzOd&sfa5sOGE+_Jl z2GqCBpB#Oo0N>25p`3i9+_pDdU1v5V3{Hjme?h=$JtUi#>+xDKI{cOK@gTeIEIpxo zfm-`rqt{b2X@y!BG?v_hrzY}z_3vzP*W*V&(JA0HLkd}rw!G?`L_WDw?A^7rl9w%m z`7ASm*bjCD9aBx%FYqD7riH@SKp9Z7M?^SpoT=1902>u2zcVu z3FVr7$;x@3z;1*g|9hDXAFv~mPnuW5Pxzu{Te$HD`JM2c+ix`sv>W=N?USuo)#ysE z$!lSEyCUAcl8x3^Glg7*NSLs62(0-U2)^f2LCjs?x0)_LBTSi3Q!;`@+Me9)+RJ3> ze>sxMg|*)B<6^PG%vit7T_e!++ReBDTY(!Ffbx*+z17c-E@wb-QeAGNfYfMfGAaLQ~? zcy1@YJ8C=dZn7JDb9Nn?*M1<&V)dAQ!&1TNd=&I^mIuQXGjP|TQ>bfk56yaVVDu5f z8&4eqEY@ndAI+A*7_J@{>ZFuPq@%WSnTg6=CF@_x!0h+H5mRAmL@iMjrm z+Wj6^J0`Lt8}-?z;ZK>Fwr*KoZU=kXP=P0`zoPBNVD|D}6#ef%ISjT*!gEZ6zdUsr ze{9=n*#2xVCikDj%&%!;;txyQ7TQGqd|HDarLm$X?=on`n!)}d2JH9Jt$4Wm5*ke0 zhOaC);HcJ<&~n0qy|*01PT6z|b5CZ%Ps_!;oBJv7-ChW;i-NKCQz0G?eG1Daje<@? z(N8?aidnxIk8-U$@z&L2bnf^L()UoY;MFt$XC564?yfyBP26DpI;w_)vMz&9sSHUz z;fi}6Y(VO%QMPElY}v!JeaeoHDWJM#ZZPlg0vP2m6ykbzqSveMWXC*BuC^qasQ7IM z^OipJL;Y*=l1TO`M*>fn}-^x7Bd1em(PS*&sW1et!?aCvm)Ddd8yc|c|z7Ias0mR%KS#9 zzI;>R1G0O`H^5#-Do31v{dqS0MUk5lJ>wI3ad-e)CI*vRm->>9mMdIcNe0>FZV&QL zHo#ZWtCW2|gHEM-+@W{RNJ{ZR-gb%`zjXZo?oQ?>;p7u_oZ%UPQ#}q5MQ<}IyZtt8 zmJdXYP;+$mpMj|zN9dMa@$}3)Rpx27mhEge#WgL)gbpc$V5^OEVH~jeo_q0Q~jL(6{q1k(^Vh~JzYA_fhL_Q7Rww@|wNC3p_m3Ho33aMYfWY|52IEb+faR5iBZWEIWGS@Y*)x2Fsm zwrA7kTjuzp-3FF?PNMaBo4K)E0poYauoI)Q+5CHQqzZ0=d%+}dV6(Y7rq)7y;>A*@ zLJdKCbrLBC8Q4`loV^;*fR48e+0~&xuxsixVIaaI`$guFEWH{vt-a59tfB0 zI!T5WvCTVXV>^);+b-L`mu>$%m%U#*kZn+218eWzgG;S(B+6$ts4lw<{kjH$^O*>2 zoWv>Xw0?If!3pV(>cTmZ6t%i)vuC%imFnPx8aLbKIPT-W(lTAFAEKRzA2te>I8rJzClsdJN#N;8 zAK~^!6KD^9L6>K{<0Sbjcun-JPxn|s@2Xxw^MLm-O>YhyPCkXL2?1nTyU23!S42b2 zRKBsH1eVLK5j_~oN!->jT+0VZZY5VsmUTtYv$i47W*84G)_xF+vEb#b$G&MPvHSS}>+zf*YC$R2$TZL8kLh+-xC$J6nBHR8uBpf_of)6uK z;;iZ)IA{D_e6AWsWJ;CDMUf@dC$ye8Z$E~}nd2AFU#PqIDjNPM#zZk|&M#h08;mCb zQJcZPF4X70-ueymtf!JJ2NPUB`vjT}Rm7j8CQ8z$EP|j#c{I}ZdTFgm5xsozC|({A zP7-UT;N(b=#o+Fby4~qmkYOX9=b~u97cE@9Q40qRYeChFam?&&D>x=yfse+{5PfP9 z;XnhQR!*cZ?C-&RG8+e{NU74va_Bk{D!MqU@$viqto`{(ENb~7b*&EIbJwc#;_RI- zaqEY(L~fgM@g~~4M-5$_E^sT{Los2*ENmzXl_r#55f;ClfuoD7q>?o|z;l2hKFEs1 zwQ()jXqpRcnXz=GW5hEy8B}_Z zT)r8|X@p#%-4&Z4p#3A~D=osX@}D$Fy!W`j)}K_ax=V`1J%zmWcj~a_IlgyYg3D)Z z6n;fIQtuJsCh~bcTusk%=y;Tn*eiG>yMI$i%#If%!+N^Ewd7N%;fhnD1Br@}qXi33Hi0E5E z93suZ&i5icGW0m^^;KXQgZAL%Z}*W+Z$?M+-_#&@9{l%eV3}FkGq$StA=?wL!ggXO z4w#aMDnnM_iFr>@cU256Rad9o+s!F+NSB_|ScO^}RM4V03>N8TGdI^?taFw^*~Ee4 zV9nIi!j1fe!eHfa4D(sStVZr+o%cOpf5RcFA@*`Y#AjJ;s2O+We4QkC&I>g+61P@x(+3=6!*!LN}RR7-sa>@J>*LTfXGWE03i|M5~P>QJv37+df;NLwTn}M`0}U*BZ;5r{t6RW+!?g z=r%5Q@kLMVAS|-7WObiC*!5kG+#r()=$WUAm39uSTxJkc2)lqW(`Mrp?FlR<*`Ez% z@u2!R8Vq7Hv0D;^+uGjZzc;C9{9PG==Tw8tu_%p1*m=?Ql3^;p(4NS65Akwx~dn4Gyjky*a!CGThMg99$h(MrpQ z1^$R(eRf;I!?8mkdto`*R9Gdf@}5ph!y_G~o=GRDs4-2Rn9{Gl`Ce%LimcBcwkpku^d zo|mHG$E{==sNl}Qt~OOOBbd2i2|Inlo{e7m5dAKRpQVo;pjV|EaBSxoJl$<34ZFkO z_=9$Cx0Vk{YOscsRy(+~{3HYmDySPZ0kfgcHR{)#uRU zLkhTz`pXTr9F1n1M?j>$Cu~_CFK8*o;^->?7Zv5F50X*54gz|4rlbPqA!l2*IY?DSZQ=Na3wSAM&F>9CLz*V{8&g+6; z*YsH$ZsSb1G^h6SOahDw^SumPB^&@N^Y=uYqh2dVa-86i{ zP@7}Jib-h8OvpbIfd36uMJ>HC?8)C=n%MS^?$U6UEM~c4rzxMC`)D0ivRVd0P#9j7 zrsAvUqxergj6NAP6mqtj@i%@SfkfHk$cJ{q)s!WW?)v?I6rFigOZ=pGGiX2EcxwagvOWtk)P14s(QCHY zy*GAuI8C4Ta-ogMhQiDH17w%8O~@ndtE7LMllW!*HFo9qf8ySY2id#2{;=h*w6EJ0 zA)^mg6+b(FTr_?5M=V+02ixm{uy=Dhc8Qrq7amuo+tOMj=GH~5`k9UzE<>355Kj`P zzqR<&tFBa5;RfEz_Oji>^r^vf3%bYI31SuO=zDKB`nl8@f2bN!&6Ojlk)|Wyl7T!p z0{9SJfp_3X0+)R;5tPMUR_W5 zD9@!_sbT=%F!L=g#`l8$ibB#UvDLO5`NCY!OeH_HM~jQ*))65}i;k6C3)nD@R3*X}ti6y%t#sN2PtYREdv$PE0wdgObALO3c154-|2G+Lt!#kx9`U%WPZsWe7X=BL$!yT&LtrrHvS2!TI=i-m)T*yXfe0wU@fQ>ctG`T9m}vzOILlTfBI;C1yB4IzxN)SK~j7 z{CMcJXtqu-pMR}=&UdEdbL|I_;NsrM>Ti}~qQtauiBQA31!d%^V*&gKNydoDU-6=0 zGkL!EzWDdaUov&25oOPt;IRJ|xHM)l$j_z1jvz~j+IU5pDQIGOV;CE46bk?K+=4}Y z)$qGz1lyY#gU3fq;(Ie*vVA57urs9$jVlM1yq{)L(zu85emP6|;bY^_M7sT)Q<1## zlN@Qib}1Aj4q;kOc6jJ>CCmJv#3E^tY`NrmZJVTstMr`Mw$qnc_no)VE$cmIH2Cp8 zt%0mFN||`=TwQDsumw-081i>jcSz359t?Y%vylVuiO%;K)(SCj>uCfucErF+r5o_C zMbDuwzDZ1zTf(C%8~E_-H*?ooKo%zy!8fgJWCf#zyoFkD>_|8MSIRs!O&Y?V`2L2F zv~Dza`YvcT&;U?0VpS$RpqKh7w0^N2LqA<(^JedW)Q=i;nxhqUJE+TJFAwAEGA8gH zyY%EiGklA8+q{tZ54^{cLf^74gJZGJf+@`M^hu<5Q7Zk}R*V}EC{Wvf%=ezt1lV!fUH z`iZ0EhNX&37E(^8+3#V!hwLF68;s<-6Yh$oid~3BwUmG~Sp<>t1QI@bD=Hg@bNXN; zUn$K>a(7fgdBAzG_Ff9UxN{60L(_%pZo6^yqj0hL-d1vTwD)*Ur{Hsa1rP3*MF zHe9UYga$DO@xkvFHmRs94!rt-_2`!inW~v^zI8k)ci)fmEsqPnZozE##u-pwZi(G! zo_POjm1ul?DhZUGhu#(I;MRX@q#WS}u~*#^2wUd{dSMB|x0ye&_b_+fxp6Rm``TIv zcbo{19r#Ibmk}O}S2Kj99j@i+oe=Nv1VI zVmRp>hu-6R@zlPjP;aF!rtC<=&*x8*Wku#N#WtC3pZZIDI8jS(*Y`2m5pxVuTz@#k z6de?f&rcDYXIb!OmDj>D@|GPkeuE)Tv(RLp0xf%-4_AVlK>KS~dZ1YgM(&Qor(F%i z!n2C9qo%ZaxZQ%*LaCzZ|@}%LHnQ{=yK|MRd`>9M-yGmTXqkD{^I> z8FRTMF%OUbD;}=#7!8wy#l7nySeFmsP}C(A&NXJ@!Z%v2P5UjX~vtwZkmY>|)<@Tb_cOv~Z3 z}47$<0Z$EJ?@g+5FZx;^~wH6_AU;^?q!3q`@Lo~cT>PAHTL|>%{kn9 z$TV*HWiD(B7a;7C5sWM8U>7{^k{3&r;q5XC^$JzwcexjAR~bRT@Gu0us)M0dPSU*X zCux|vB|Uf8L71km1HU&u5pD)2vcfldsJy!yDuiFg9re$!_`gT2`=f2x>Ya?A!c0&v zzy?eg#*_0WqQ$X)SK~ELW9DlU%9{ENcG!B}R)iaSWIV)xo@*QoQ|ET#%{^M2y#9gY z_8n5uFP?3?>?P_>jVoF`S!$}3?1FVsA?(uyJ@))>02?ZIWz8SDf|KVPxmL|rmOLz< zne?f`kWsN*{q|$#mQ*3X-}j+>j{RU*KXX0?o_HjC?KVK@n`A|vc=?GcCw=jhN|O+G zpeMQJ{Q#iPGO%8EfLus1=Rbd{@jgLL5|i%c#a?q$i|e}U2&!j0<)?;S5@#mOU_ClL zX}^vne3trKaw`Ty{OV}Y!Yv-gFJ1yE;hMtLLKiOII~rSK9WnEgJ-Qj3CiBXUqQkEi zd|~JX2TLA8-^86T+4~9`d0Jikcp;IT&eJ7TD!uXO%QW^Q-UnT`8)2(cH%#^pV-?H2 z&?d5!EgwDv(%W9b45MH0C#4Tgx}MDboEd~kfnS7SdS4)7nF9T9jR*LJ^(W@k8R{Lw z$UT2eyj4F0XC5qO7yRXn>FF?o-6zSnJ91F)??xuZP2`^w9QjK3Ik@4J#B0qNBaZB1 z0Us4&Sz*v<9F+D`9DFaE{p%7fR`wnShA9^?@NpgnFNtQYPa27VeV%By;VJ$*C*@w> zxM0}C<+!5Co6XP%d>zvjlQv5ooBd|!d42?Jy&$nv;$E>|5~HN9tPtLfT>#StEM)WR zYp`$XApVel!Lh|b%=`r|?~GQ%ET>*@!$k?y4_3(ni>(E3g|YSqt=4#UX^7mt zXaLJ^k&`EHT)--_sc1}RKMXR7CdX7xqs3NR%xnq9i5(H3E4>S7hLPywWQ;c?pW4(H z&&k7Te|A+Kf&*`!B-V4K_h}mn9|PUln2&#i{89R>Q67#i8~>3|aSi!4Z!b7(dq`sU zcLUSM!KAmj2}VR*L-fnU|L%IQsd0I1-|w5`#@LZ!Yx-xrD)F-Zg%E{kF(g3OuT+UnZ4*zhMxK_Fe7t4j%F)(UegcO+n@sKU#W5DyC&Q(;4#~kn;=UZ zU4S0h#c0&%N&;Kwvw^Fv$eQ72$;Qhbvh(KZyz@&W`VS1lioeGEdT9)}rTh^5(`KQI z?+<2NwO6>6XO2V9CWwl0kBPg#rdZzCL;SkJpTvy2#OBW^W9Nds!OCy87^JjMVxkm4 z~BV1SF&EB=#V<~GKnNMvqruUqV6AtuY zAKh!nTyGBLr{y@}{ur+2Y{=oc4R`+_JzxF3@!f!QdqsO6!BNa1^CoAKSr+E3K2V40 zYgxfo$rW(m=tz=q{}TM`l_OxoXZG&&X^3yECSTv(WS@?r&{_RK;tVlaZSyIis@RoG zGrlgC=$s}Ka}q#4Xd5{ec^-PS_6CJa9diBD1XvWikK7zE2o#;W@Nw53p~9L^#Xbpf zaMp$pw@0g)*?C8CO4mNZ`Kk4Sp=lp(eefc@UbBPb9m>G)+Z|#|mvOB2gg#g9S|u*y zeZ{>z0L~63P}}_pDKL%~RKGnIqUT(Nu)B$T?AcIUI46&NE7;F??@_E^krj-Jh;&6|Z%%RWM!sV;eBqa^&>ocwgEmDy(ZJ8^e4hu-&`Z#8_C5C84`Le?^vN7jstx&c{0kW@+ z0=Li_L1wNZ>uWTT-B{Bgt8Xv`IRBn&{;|1W~+%02WJYhC>LVw zokw2vzQ`QbUm)wfUqV<)e|k4LQDPqK6Dm{G>HOcz@$OA`X7yqf7VRA#bJqXUdu4kp)wvoxTOVdGW1Kw}uFBEK@{hy?-*#8NI>F{i<*<{+jsj z*=!zQ*q<$W7ESbz7{I(oqrrK_HgxaiPL96|2J4^8!ROm?-mSkYHxy(T@u*XM@twGtmpX^;b*d1VgNb$vwE1WjVc17&31lAkQwX_hd} zz=D69eF>N7R-(6J1z!Cg#u`UN!t!7a0RzHV>YyRKwby-})7z{_2q3a}9 z9y700d^$m&ccrCh-fybZA-*QNw9!Xc(Vyl1b(k5;CXs9ijuDy+P+nCDdlz-z5a^$aqPP4aK2K2h)UK+jLS8QCj z1RdfwvF!V!S=TEIvB52dov2Wy^><=fWn^z^>$06KxcmxQj9$VI%VD(ZwX+C%!Kk@r zC#(s-Op5wm6x?dQ;7r3a{NMP4{D->=oOc_8RUMt=lkk>(4LmNYC#k}btQ>e?t3=7i z_h2@BF<+2yiLDAS5&l+Og;-A?y63nYxU4^!(Ib|-ntJi+sU;HAC<;U7PlN-KD|6oM z8PdF^6jcj8$uxD|7Q0P~m3u}MVDqS8TydcX?k}ih#y?fiP_0~YpllVJR~kXrlY`j! zW@U8us%1N#9U#Jtk&v~j8~f(ei>NjJBDd2GrVO#^2{P;AT5D{M6QidEC)qfBvq=aP>TXclaHED2Zb*RPxfa zu0p$B-7rq61E01k@|gO3aY6Mh$*CuGYbR6-7t^;hou2POp{hUawqqPm^op0`J^!F3_f z&2W!++g?L#TpWcdv7D!l9>n_|R^U}y+gZp$8`xNpL!6i&|0&;(8BMNut?4G_7N+8^ zm$j(uWGpUyJQd&GKZRARFEZ`psiJP*(ms>Js#HFKe%m!4Nd>=ec84z8y zgR;Xq((jKcV|aG~d;O<@le;N%)~;pXhtd zgPkk;Ui_^nLU2?_fa+0Ny!RmmzP0@d&X}kzoT=|*-$G82P2HEnjha<>>S{0i_D;n2 z%hqsD??L>9=1H#B8NmZ^3g7K~jD63w#)nTP^GLtRJT83~_D~6j(#^lf`f;_a`TiC8 z=T!lG-??|V)xWQ_Ki9!^>eG1I%SGI9{SCg}`2ZVhRD!Ggbn#(CHxd#V%VI_)l1pC~ zLGd6(mUn|P2mc!EHM0=44XK-rvFO zOt~SHsTr`+$d}@g%DbX)+XM!0K_*wy75nc@P9Vy`_ zmM3zx;gUnBX)uYN;>)y`>KE@(pMkW0o-GRVR_n zzt8%Pd?QAz`2}0Q8RB5q<7~p8Ncp?0M5u;r@-K9{LZM+8)^84}*YD-t0 z{so#z6f!m`GunT@P;$C~=&yCeeVy(HTdV^XK(LQKKjtkILimQt34oZ59LlGq>T|<^wp(d^nEE zR%9pK;_;)84mo{g4=UY1i$VLs@K(Cyi|nm}yCP4+hi;{$V3Q`TP|L?bml>UCe z+Z6ms%JJ{B0_>9U7wdHX!+vw^m_fi@HYwu_uJ(=rh4T*BoV=eYKN?Rq`4qFBt)9hX zMn21(a8!^T^+89sYVqxc^@34~JL|PPkvvL>6)sGBjaxiiVN0tyy}2l#x*BTI1AQFm z77GPB_~Zrhq)R4b>*hlsza>okqKx-X^&-b#T2d3!N~k`O11*a*Sq}v_*7+cq^gXek z%nN%gkKP<5*gAd{`ufdBpR5(Qu)GVTpWFup0dHVoKp~x%e3e=rYokV8!q~9Kvsj(^ z9th7qNX8qdlPIe>L__Tg95yo}(`r1y!oLaj_^pPjyj4IFCZO#r27hJ+Q+-Q!`h_&Z z$)Y^T9q^3xH8X{GrW%mIvxL`I-wIFcsz{5QDedv20&KG+H;KeZ>zEiy^i$fHY?H*r zuF|JlReI3wmo~Bl^;n^*dIYU9@}Stfo2E#4$SenQaw69Y7w1T`j9qO)*Vt^qWcGV; z6D_D;{tFj$m%{GSSFAE|EV(y$D<~G61V6Lu)DQW&kYxj>iP3#lX(W}0+ioIba$dQD8bJ%t~b0gq3s(CTl-z?O%j z()eJewIPFL%zHu>8Z^tV@5mARHkiVQnFsL7Dq}wT^atVbv<{;DZ#29$cZE#9A>gx1 zhx!kHGqN({s(f*^k{mrbxvT^W_tE|K5b-M;KvU^_eE-`akZT}@nB3doeg8uD@dE}1wjAZ;s&#y zw>z-btbGmD}rGcVAAvn!}UXPQL1EMc2&MKDvz$!T&^i4L91;MvsY z#PsI{qPHMh9G+oCHpUeTJ53AO-JOfU?9Wtip7#ZpQr5HMiCD2Pi3V%!rLE=rsY7WBezdSnSF}> zn%srXE6QQoBcix=X&DZ-874MM&d~HGEppOofo$E-Wmskzhesk*_+r&q2G&Xt+gm|$ zsvjl4C;t(3gnne#yJT3R@&V4R3Wt8;9$HxAPG_WlfbD@v@WWM^uAUM`>{4FI1`a!b zeaAgQ3;P82WO14p+HQqG-IOIJ+ZSQZzWKrm$0=-KgAF~>-+}f#mCDvVNg;h>WGMc# zXVzOc6W?W3m};;eXSGW_=ET$NVX&oAOBrIKKhv z$IO9_p#!mdpo)XW$3XOJSOB>P@>oW2B33V4gkmG3i#pEJ+N}w6)z;p?wn@A6{CdIO zVw-rs@*8O$Sb=8$o{=$ShlRrGE)J$a7Sv?TAJT5W0`3Ip^9X6*v~t4$Y3H>DkJf)f z<6V*)PJK5HAE$`nYw}sbVOwIdJD4pfLc!*a6BYmNr%kbx9<0BIt5&G<`cZE&%O#gi z%!;99y-aAl-2(cj(3_su%YuYUrR+ueV{yjH9(>b@`Mi0H0j7?LWnYz5i~9^(F1ESW zu!@AmVv^Y#hu+?=gstIyC8v=Zt*`h_(|y;%?GNvW;(u$|oo{+VQb;muC_Bw!Vn&ge z$5y1@5)DX9-iWfrS6FLj3Axian_ZN4rIGXJ@vQILM3aCBL2;tQIFMW}I|9uiTAKA7 ziE3g>+mzAc&|~)Ox1(d&7(K^>e{*T-!MAK=?h-UR)&@()`^r>Z{beV-sxIvF<|KW^ zMMA89%Hso9QSsVqRx((aE|1uM*Q>S9rYh}{; z-JF~JoJ2F0YUAj$Cj8^d^Ee`83)9iM&vtFsLDz;H(xt*!Sl~K@?O3};79S+#m`)Z8 z;h7oa?spy3FfwLtp>B}tdy}-;+k$69C+l1z&GCPnCck}D*^4229jaeTdFCBE@Yt_J z_9mkmWj5DHnh*tR+Z5@R=EHPIRWwZ;(4StM;mJ%-eq$dVhvCYZHNxed&&A|6MkcE@ z!RWJ|sP-tITg)~?A6sbl3^Irja<1)#r=cpK+SFszF19%# z7xHSC)7LR4F!0S`UQwn<%{zhxF=3&6%fvKcVy>6i_qsCvw{@UsUDGKTSmrQ`cWvy8 z$woXhZWR9T)Q68AmL&5-AF==ILojpGe=MBm3UiL179)4Wv73*&fO2gHPLFni`^tA= z^8i;m*fyAt>}AAl>Wt`u_rdU7=L0OWv7lx@M3~ZO3||Zuvh31@2rb?5!{->*a^*OV ztC_{^q_by%aw5b;s({AlL1t}}>UFgXBc#RKf~O(&V~Tpw%l?I`(ofR0^~ z1v5M@2`*KG$c6;T@AK`hkbC?cy?p*7Oz@KUFFI$br}Vs8J>dhX?y?%I+Q6Y`WsLQm@GRKj0yOv|{g%50AWQWuVq_nT) zb-~$1lV^QAg_jL3Li`mC-uSKxdwv>@E5`kE@O1YCrI4vyEJ%=#|_DoX91nxQ{>aXZpQYo1eD5*_~`E;{(G7}zbt6syM>>H z*!eNI^x$epDLBt|t8}najzhRXVifF~7z;tGHjv3?rI<0ff>r)8Ky8-^XuL2-m=O@n zO5I*Cg>QlQz$=c2$7{;lt_+rMSZ_%7s2{}6Eos&mn&!>uM!_}}Y<36?>-G-;eoELNw;#qRr2~yH-iU(E} z@vi~3Ji_@ppI_ad*G)QzS4Pjl<36XDTUePGrO`y9Cfs26O7_E^-Ujf-(gr6a01in= z5H`LXKn~xWjM3rO(cI`5=FMrvjSbrT<(O>Px%Mn{%ZeaZ8xIQG>;GW6xfS+3@NU?}A{w0bJ?LCT?gL(V_^*Lnu^7w=$(gGNdn^l+>% zTm!G~wiAp}B2H-uIQv!}_IYB>trp}W8FL0b^}gV}Zsxq-?soB+x($S=dh-a)WDYee zxmwW_em=*B+m85yBRbT0Y03t^cGrHss8s|1?45%P65r#b6pqWRQZVw2BKc)t&rhyS z=I7qa`OOPk#ctPo;8w35+(3-PabJ7mt$kbM+0T!Xh8cc1^Q9lZ=NG}dmrmy4KXgle z9VK>7(LEGPMztH33>dPI5B)C{ z<3^`pO^7kNk4=DtCmnFk#D!Zh@{S$EeyAZ{C#J%W!&zk4LyB3C_JZAgO}fWh4Tr^~lE+(Qq;|eHneg8* zvMpK(N0utbZ#GyMXpKic5q00edR0tS5EM4S5d zp*ekq!psrB1XZK0;)9@n#IAWd30bPjo%24h4J#wr_oG9g=5{1ecr}vrvsp$rwK~%q zhHey<${=HSDR9l9GXIvJ%wX{c!Q)mGEaxfsVX_%1QP+h?zht&WJp=BA>C#^1v!Li$ zA+Cz}564uWBy$%Zf`(rkA+fOwO)Uxo)#5_Zea>9)P0be;D_$oa!866>rM`mRBH!Y` zhzo=*OJe%%?}eG$ID7JU31lBHfdS8~;6%G4tXVdUZA;R^CHJz3Q+jV0I3|uNsLiK? zj*g>W);mjVi{*IyL@um+oCtYz7i`|)B`EeN7yc@RktaWIIZUd}COU1~1iRxpG<8E8 z4lHwEM}9pd8;>gk2?)QKX7dq3&&G>g8d6y7-b$9xU)tp_7h%oId?+*YrDkVB>5Q3+ zsOCroD4g3xT7;|Mm}soGXKj7iI8c`xyFWYYSMb9HAe4BIwptKD6yj8ocP+ z4u%^IsU=y!s&2g{5y^L1@OLxXrXCAQ*?&mg!S#6K_kEI><-h_uY++?Ok_VyppkV(5 zYA~%QC3Flm8InPw{?j8m&4ZY>G{Y%*#>i){+u$00OIY<|zBofWQGC^OjCijwfmKZ} z#fJGhR7XUzY5g2w`zAYDuNewo!;ZnF-Wf2;axEM^(jxkA_kgeS3P7RDK3FR`U%pl^ zf?FC^^ogSp8l0Sq%U9kJZ7y$yC%V@mqeYz#zi3V_C_IIJbt}NXW+HTrn<%?z4lqEW zfkmxpVh=AhpnYJQxIbDB-UfGJY3yeHY;qzWvfGyra@OS2r5X90qY>nV{4a4G)E5R9 zq!xv?+0w4%b~IYW6vv*mLe}PJV$F5n5!J4&zD`HF$XT|GxZC|D$Mac?fgDiZPnc?uZg&nRyZ!I3m zdT8p?iSehIwW=MRyyG2lkl!GBxAgf^{qEG&Y(Fhy8)V)R@Wry)5&c!J$vaGg$?eVq zq`M!%X_}KE&fo$GpZ5byEF?Zhq3^>4+1F2YIWuDEa52$m5ONdD`Z0|#?5sBYkInm0fX zKe~Cbrj9rWIbSF4ytxF+_nt#OWg)JTSjq4A6|m>ahC{@sS0KM&Ks{z{p(eiq>7tnx zbl%wE@-fL;sFSM(!#mrd^3YCbYC3=?Z_c6#`BsjTT?RT{xDm;1vnO-iG-LkH+lwuJ zMJPUdN{&5W0!N(Vh*DfST>KaXW7peL(^c^_HZWJT*f9WVe^-gc6Z2W3^nO_y>kitA zDzdvyfmEq9AG zl?L6R@`QXk*%!3DMnX{APw?Ek5)Kb4WQUTJ@p~b%9Lo(? z38)`%9HvaY2XTv{$s3;(an6T>Is{pC5FY29Qvp(nD*_GMGJNJL7G||{BoKIarK_y z-=#ORJnn~EUHajmCkmpLk{oiZ_mI>d%O-w%F2*7xsbD_XF_b z9k4N<=aJA2N6Fb@7btNPVB!=aW;Fc=CIj?oLG&Ydu{n(9UA3oLE!OmzkpmrPph+9I z--I(;d%)FcM__B(A=uVk7t?EW*}+mvwsUy~wGGOr6L!9%$1Du##qTOGvd)?ewXrDf z7rzgFy#EV_4vWm&{VTb)qPNh`PLp;OQ|WC>4aY+b_QZhPltIdVHhx$l$c6@>!r|?x z_Pmnpdv~9`eLoyDPt?G-5F~$R^x)1vhX}v3rQfP&e;WTxliGNEg&UHSa+T3Faax+d z-b=i&e+TvORFH?bsmxov@L(&r90`Y4+ZU3;^#jN!l^{0demqv638J%JaO!fafJP>J z%6{r!WnLYj4uhWP!U8=@(ycs+}lqZb5B zSP0&Qdi2i@TR}T!0;tq2p_^5f(BK7K>7wQlOg-@sQBKt&C2zaRCm3#(b>Hzx_G*hC zmJRB{dIjf*m*y!mrGhZj&sM=|c?1&q6jdpiMc;>Imo_FdB?$1XE zwDBD-_#<`Y!=tci|7n)fvyfG}%ZR&;COL}T$)NQTFFj)j%f7S+Uy?}N06}=F&o1%A z>q7DS^N}pKZ;`O_+7sCV?HJTY72eX>6}LXhBzx`OiDyP$Wz!ruTPAf!mya)Fo}FKX z;$SBnxgdv;q*=uWcOE43zD1K^mk#rOrqlRC<5m3C^LJ!{&k4BwJ02zmJi;>LJD6EB z3ZsK>$@lxlv2kAA*`))L8^OF3s(zV4(~)=#QWvn?I6_A^@FI_C&-@5Ny7c@ z%1rmpDAu!Y8~I#zjT8-u1-Gl$M5kGAO`W8MN|TiF=1Of0WGm2OVhfgiSb;h@ zW1uEKo%Mcf$P$|Vvd--5EZzKvL+LeDAmhK_&xS)-KkF!tm-e!Qr1N0xD>bzFv<_za zwX^#8wQr)(YFn%TlV;23 z4?D%ib~wO_*WY-%6gA(tY{R~??M&hmY-%vO)O?cZ|AY*5)#lv@l zF~Rf3SZgUpMP*pVy$`laH&j-czo|6ZkjYrT%%aShDq*ei?e;OIJ=8@R_P*J z5xGFh&KyS1!9)4sgOdC9(HdSE5YGN7NmSk)*aG?B6MWy=b?XbGcMhHur`{=l-x)_C{R!s2#pP zWXwWaRj_fYA)NWLjGY_r&n6en=l5n$)~kBhQftP0?d-yC zbQOs2jR593%N|a+tRiV=UyvIIZ%Fggqv-O|g8O$}2bWT`V9jf3XTK~2d5AV9e>3BL|-K%NSdQaMkich2`{h9tf)O+ zRZbTaImrweu*o_^`B>notVbjPSuF=CHt6rRRVdP#BQ+?-n|&d;;ua<3ZFtrr%D@3`58g#fBhv< z+TnOWRRNPv>yV<%2x08h@9gNDqpT@p6@eX>@tp5|_Q%HwRBJ|nrnfos_nC_N?cwa* zqRE(0utsiNXNs|fhskw48JS$zg;`o$7chJnxqtgCMmIa~s9IHSb9@de&KUqxiX-r- zvJ&?0Uc;uGE@#?9Z81kVgJe$+BAU+jFtyTyX|K+Lznd0`u}ymH?C$^APUmK(x`V=k zW$$r!?=CDp_>I5)U+_-wfj4QYIIkds1mtG&voW>Iug;2E6y@THxWA-t zcor<>nGSSI!YXoAw8?Fh1%OntdNrz%H_Q-?-ZIIKO*VsDPq|OYqZ?# zz*>H<;m&*QXx}+rv_at(bDKJrbRQZ-qTD2M@G)-~x3rPCIVa$fk`Op#bP3M9C}xFY zlUYIZE%s^Fa^On(c+kKXr~WvF?y{}$U{MU5+&B-eF5iuPRO89zQ$cL)lP;p|eJL(- zb`jG!c81&@9>*RPuNS|(T_I#&4T9ZcqHym)eVBbr;`0|wV&iI*;QZBQoYlujdPWwL zmm?(p&?!sY{BS2aFLL4uy}RM^dELM!vjxuC92YXKh$MfW4K8VSA%4llLSn}=Sts5W zhS)tK)+t&fJt&bK-S-&Q%u%C8KBgG+IfdL`S~)nq^d4mR2@Q-0-h#ef{rH|MFT@WTHq3kIJ2}l$gt>ReiAqsxNlj7= zm}x(RKF1|?(7alh_j(`GUg}4B+AG59C(@Z;R*2?N3&qTqYv6h5yG(oCF(y8#U=iPr zkRFA*LAO$q`X~Lu?dIlm%)u&BX1q}RmOlvx3Eg?&#YR@wI2wPl?%XDCi8OEhO$OY* zj4yj!WS`#OXXjgAi3awbeEif{+;gEPUlKSS{9IPU_sWH!_~a;Y*RR4=N2cJchm`5A zen7T*%w}$QO_n#^pAf4HEo`>)Y^+Ka~%r7J)e+~ zD@sVZ*=E+i{s?=VSjpsic`RE@5LfpdO%$?B;p@=<$idLB!fy`^^5X4T;w>u_#13b; zoZmr;4i6^|9X4R)zZWiPG?L=ye)64mMPlvi`8cD$C2G!XV4oD=!kh>Wh%Pc?+o+)*@IR@`RUZ z3y91zL+CU80c#C!A<=DmpjSEthM&1Fd5Mxp*3lLgx8s*+({hrmd-R4(_8lVLwKXP9 zi}lIcbvGTx(Q6{hvxd20lCMbaLaYu-?2!XYgpv*!nNk=`GJCp`h}qQw81Det%B}GH zM;5$b{a{jrAr8%7jgP&;&_2il=CzE4rQva)TX2inI+ZZP=jvqK>p5iS^$jG(Y8KNO z^bP&75$$vO^T4To@a@DvU|yM^Tv{gc*&Qip{bxo}F3n>1|I~}?Lsya~-+@O1 zfn#iS{X1gTC681&`;$wrl$dB$D-T%EE_h7n3OTtZz_9|1LR4sYXCrL?)02jM*8+og z59RkJ6$l%p{f%eWICf7nosDyRMncYBCYA5zK)=i}F!ivCko6%#7!t8anA4+)_@;bk4Nx%R|xrWH-^yV!$CNo2DeXnz_a&HS=s4fd_d}IwCgt( zm%5CDCROPj$kl*$n~B@znPd3sIc)vrJ;eWWj4-Dtn!VhXhndbJ`7(c7-uj; zcCMQ+#41FoBCWLZN^frs{2p0SK37Ni$~LgRhJ=g+zG)j<^>)LK8Xp=6ES;b zS3zrak&re~f%jje1&2OdA(dx!vHI&Lw3p5Zx93%2bk99_KUx*GuS;bPTTe3M{s$en zyDcgn7(zZ4_UDBUeieuK^%YcA>!9zs0y@@fB0by|37r?)#ZHCaV&T+G=+~YJBh@nS zX7@Jf9Wq@&rNQv($TqxW@j!I!x&?1t?=5o(mE+9o9k`Ok;8iX!4$i`>?x3FW7MBuGp8&lk)u!NJUI7ap|=H z?N>h(oPHdEeBIAr8o){JpPOvSmnYD0<2dH5E#_yQ-sQE%&xE#p?I5GlPA_E#M5|sQ zWbXlHbZ9x6=lh_4${^v|((^JiA0O~c51`6VN72U)Z^%`}Fz|a7$Q*ZlAhA9bq;}a0 zLHAs&Fx&UMcx%!NF`DW!9n%F2?WTgIbkA>E)fb)qox+ScRmAK^3Kl;q*l-zTV53AArchjlmI+6YwT|<_Q)skiKsB%)D`!rUeY*n{5uFuE5~e^K&x$>ozpF&vrUt zf;F3Hl`kZ9~2Nl-G=J*Xg}7yuPfKu{4C%oq>@%ANC7{l7Z*-0#(W z_1>=9HQha7L+_f|tJnGk(Es`x@~ka|^;0(qn)j6S=$DX&jYG-oMWtl>*^lIX!5Vg4 znGYKpGKFpN@_9bM=*6C7s=FtL2CEzJtAlsY zn{%yL<3+RCTb9nOY4lVwbJJGbsq+CZ)UJj4i6dBps=4gpg87sWmZiJB4wB5&vuN$Q zHc&o&1M;GKu<^oYsNvvgZiL(|wDs#m_Gf_!yJ@l+dc8%s_f1>jqRky<@N5z~ZPklC z1Run!VJvj|O=aVr4`D~Y{L2mt{L0SB_QC#UrSRA85ZjkKhm^`5g}CdU;N8)Rrt6iM zj{31v;EvtIvBke(@OBK_=-fuEit_1)#zM~cfjM=1nn|a(>_qMfK5Sdn0(O7z2-e0Z z9p2aeAnVv+TJg$VV#h>9aF}aobEV_^X5e3m%B&NF54(#`gMSuUGN&$ERUgE z4?iF=<0ZI)B4e`_&eE*g+t^HR$Xn94$%-ysR6wN@&*J-GhD0}5nXJ04$R5Qe<`# zMcar-g<}-TE;8knqr`cKsXX5ol?R8oWN^(dA;Xk!lKa05h+5KpZUgBh`(*lPcA*m= zeRvApf%W-uHy+c@Xf33Z{TiBwk6^6^1&>wTg;b&`g5OY7MN5~3ft=|%PVuZbzhn1L z64j){=fqVY<2qGxVPYRQZe|QVFvFfy#=G-!d$d6P!Fc!)b{>C>F6E*YrP6(=DfG>3 zWpG(53yOQM(T&&p(T|lXRC($?y1sN4E%_FLjc#qjW)m)Pzjf-lYfC!GVUOKB(_74^ zbtLms_a@`0QI(`S>O46hCd1n3DzKlQ`M_7_Y4k*I1a5nH3Ca5RLwkrfv=D84DKm$D zvwKMPAGt-2YC03ybQAj0AqtOv7RtBv+~otdDpvm1`~p=jwGhjx6KBTju>QMPa?3Ncr$QR8Z{1RWvD)11k#2Pu)gZ zcF%>L7-)gG2U1wASD9GWR*~<=V`%CVBkJohfo4qJ$e)!-<@KTu&>PDmp*!^??(DyS z#zP@()+6-Ys&24btO3>;FVW$;YS`h&+mQmm|oiYo*1=0 zhMui?G=MSXFTQ2@9FucM11XX7uTIi?We>3IA6MM9VL1A}#1d+aB4Mpd1CBheNLOpu z(jxaGMD4mCHh3vbd-l(!b{#wTji>LFe`}v3c)}atmFZuwlldIl_sO1=tP+EqK`W{k ze}?~grj0MKR;^sS_ZxF@xCNG&q|SZ6noYkxy-RxLNaJm)o#fV-IdrH+8;9%9(=TI( z@ve&2{C=J1NPGAnoZi}qyT0DWTPyEkzqprp$+aqw)LRUvyA@&j3vHUJ6hkLmnnmY{ z^gu%O2lvW_!+#IlA*$w`#B@afnUgBcyB}5(@^0aDnXe=-A#F`_B~)nC`zmTRDTPk{ zn(p-yFV&CK}sDE^k$|gN(AG?_9c9#e- zPd$;;FGlL0>+sqXHEPllLE=-isg_(1nKbb`S$XXu>c(%lr0g(!d%^}h=q`q@JbX#e zrCjX(?GBE~ynwd_%%jDJfyCPp>>n3ZQL)++7p(k(^Y^RK{BT!lKGd5mYu`*?+_a&} z5qZR8m%Zp^TXcEklt88+`2>z`mLZRTHkcSUTacGJG%s;mhYAwU? zsy&I2G*R#fO30&o-C~*3KC<|#!*aMHUkXWz!`QQrYEXOiXt-fJ1wJGf(oIQkXvpns zDtS{)s4Yw&dZVwBQ*RYu)6P&ZeGm*{_kIgC*Ap=L>LZ-)=829--{!X7jmOrPAJDC* zFH(E6C^~ch9ojYX61B8j%gJtvhn~VCxNYP}Fmph^7-iy{}%?y*|zg_2j`o>vW`o_Qcq5JX_TNn;}MPgW#0Y&r?+pMY;1EvD_# zTC{iXdz9B13#!H)h{>7(+p91u#$Sd)vwQf%$|LlR*F`!eR z#M$L%b9X1~=LX9(c`2VQbb8luY!jmh@^NiwT^dJ{dOpzJx(0gj;8D78P=ddCK=9TL z%BPaX5ws|49sj{}K7XXon&>o~r?Zu}@VNJ3c^}zqDrdB*Y^ZB4uXop-cbc<>kK9!+ zJUjl9(ca(b4IvM8N68vH1}viYHwJPtr)F`pUXQ2R=Bjkb*at-X(o zL{aZ&6+~*+aVoj`HomfR939=!PBtt5M!om7apv<&Mo+XhGXd2Jr`)QbRT_c{Dqc$vgLz3 z{P}%*_4&6=;k=7PG%r6nng#~?^1Eh7^V2)WlO^>Z$Yw$7XZyA)u2wppsvloOi*=&0 z`4NGOFn(isgvg!$A6FT0b_~tHj%HD5j^ZXHnO&{xmfD2=(PcXrF=>s6u|SW13%4o^it~upEJa!)SIig+@t3rjJst6)hmNejK&?e`yyfOX zy8p8^{SOl=jqMd1Tw^!500Qr7(P!U`&W?gak0A0SHL1y^e5(ar~Yy!G`s zzEIFeD^awBG5`84dX_;iwErMpqc;+hg@%07`Y3*E^&I|j`&BMSX#`nqAcx(;xAL2p z7E)|4PG`)n<%Z&8T-eCd


Al#V}xWu9wO-~4f0>o_l*+T%xEe@Nz~88oYL1E>Y`!&X-fh-x~{ zo)gw&tJ9xScDyH+U#1BA`(_B-r`u%Z)o8L@<2Jc@;Wyn87RHK?e*<{t4X}E(i=8E3 z%g$)-1jAzw;Anm*bn7_7Tf;zBF-8*3$X_RWns*6##nJrF(W6mRK_twLs)zcY-)O*& zRVf>vJRM#){uue1YQ{e3v|;1KJR$s!GW&R90IMxJ$JTW{ zz%_Gt^tZzs*&G-TGvlVi8_P`m62HNX{CL27PLpQT2}a`c*{Wog9V4wv+x1lFqnmMr zYyLx;>x^ls@)~}vcp82!_ZZiH)Z*u#)}?-7a&&yWENh$FKaU}IjeoR}| zjNzq@n(<}V-|)ZEl(~i3Ebf=eg}MY4K45YYNpSZjVId}v&E2N0H&p56wW&C?dW%pm zwg9;|qj2qv-_*pul%D!pMxV~#%xf7f+Si zdr0x${grsl`)>TIy$>ndt;_$FaKXuE+{orb-6ZQk9+q`rpiW&2>}GheI?Dps{voH> zrk8Q}Rc{0=IQI(F*8Pj~xQT}EpTe(lwqx~Thp`S;i)mv~2RU|d7P)=jhwI`tbH59o zaeg=KPy}bso-_(!gN#-1ll!fls_Ou>G&jL4=Td%1e;F@vD}`5#+Q=f8=`5&iXR|jY zGWGYiFgZ_?h5XxWDm6hHeqyg12zcINQyc{L&~Ev&;>;FZ5|CE`tBynpagh`R$!0*xZ*AWHNdZ;g zXiVG0mO+54Jo|i^AsbWZfOpJj1=~d;SQ@wqyPLE~%iS){P~L}YwAzPb zPKAPJ=O!+6OFp{yb^+e`$O>wn=+m0SSB%(qcRb=*0oifF3ajYfW}lCfg)z?&WSS_G ztvRFdl@-UCkW5z+WHy=l7W_ch@5*6D(w$ZE+Qas`Ooi_~25>Gb7(&9zh~(KT^qS5| zr2kZnHrO$A*paPNvF|6R5q=P%f)H~0)haSm#uUsReMQ-d($t|*o=#alj|4hM(&^>_ zEOTWYYxt!G*2iw7_KK3=xmnOibXVlx?!H7PtnQ$iHfAu-^asg9wO9>6N5rJW((*8MtghOsBlYsTIl4b@u>xi~NN9^E%$*b&xeYi%cc0bz9v{Q%$gJ`d(tniQ|Z^`y|6Sb8SbuCgDutVc!!%5cQX43y4)ng zyT4ut*McqalAwEN%=Kc<(C;nO1Uf-o&=h#?(+>|K90MWjN=Zx?5}(_QD@Q5VN`WTD{Wm%M|c`r%@{;f*=`ojVpZ zO=|+{SuG4)$)Y=d#?wRF&eAr4$A6|b4GcF~(C1HX(V2Eq zut&}w`7O~Vo0pG(y$@_rf^-u7N7IN;>m($0CrqorH;u2kPG@*G5 zRZ6=E_J1;9M*;$udv*ATWF6L3(S_pPJbKG8h)(%JY3{ut;ODA^<9D6K6MmUvCF@^= zcFsYHb(QqWdV6})SrwA3j5*#&hKxF5hmWrLLf6?Q0*bE3?u)|d&hIzqd5wPhaR;Iy zejyxx<3~lw@AWwP`z2y_HIHU1xWG~C$_ghFKPIPR95*1_XgcxkRk9;P@B?U5;Jc-{QUkw{olv&5)D{+^z zG~EzlPj{DhRcxFwipx?Kyd`6#Kyu@7coU%_(E+Dpzh;uMKkp#pzYLvWC89DOHomQLHPL_5`l z-|nHA$g8-D_8T?Qp9}KALtUBrrClSHa^E05>oV@rF+kYTm|b^rK7&GCK=PJNhUm_}+)9A}ty8JZV5%BohY}O>_FZj-O!=b+ff9#P5s9zvJ ze&R4_+4KQ^W*p`A44aL19B+Yhg$rr*gKf0Zc{Xu`{)+R}ag3UTI?2)qL62HvxW*r; zKpa%T?@bpHJE$$_a&eRhUY2K5#YsnI6Zi~Ig26arcE`=xyu_tjq@id5+UGPM-S>1V z_jD9$;a->;*)O8=zZZezA5G}0b)mg28q}p%kDq+wA-xt|OV@lKphI89QkPLXu$q9C znfm7vo!;?*YF*BS)W#M@Vq6<}(SHjTM$bnxV#jd($4i-yKZk5vQ)$E zG(KmSMM(Jja#h{6Ap1nnf2s|`Ys<^v{^}R#utqTYT=N&ztDNUjEfI+=oh5J+C*nQU zUy1tc91^WA#VeP7C%YzO!E>*2*xVKa9rAZ80zA?n>g*)8usD*Po7v3Voy&qtB8A?r zC!#S_f+oF;fu^&aG*s~sje8M9^kc4*=w*X6F<-SJ`RF_`+FCw}@7uspKgH4XfPyz&k?|P~2HS$^M^~~U%beIpjmFPY zqbn9a)?|Z&7qP{eLTtCn7mn@R4r!6Ea21{M^YRTzR!=ROemM`{^hlu+ z4L$U`z!}mr(}&gh#c;gNMaW;4<9!7y@CWOwpfBslKa+RiB|=pACnMwWgt%cO=gd~3 z{M(a6O#gzHY!fui&K2;^o8$P=f9iSjLl3B&u03^ZxlRV2@1>_LTA6eARJl>B-q52@ zO6dHP-%*aGB*?fl2=`|lvBt8M10j-?hZ?`}<#IMKd_pDM*!~^&S*@ghk4V$@3D1eX zLpt62XCc+!dyUJJOC@_1rKrK`Qo1gjlGhtY;^_jnxim49G^U9|qsC@lcw@bHpGwJ^_xMEPyU;6*y$2irmkYRVcPsR?PY_8bx28!Yz8=!{z^J6rGTbgs}#h zIN2)_%r(s6C3H*OGcYZV!Z`t^Q1(le{2D7x4xgM$ zd<72Doi<9yV@m==)k;>Gl_`7pXIU4c4|6r+NqAtmR6Ve)cay7Q(K+~G86%rPTO z^d+GB_yriY<0*`LAr56@u5&GePjG(cdrolhM$WJDpgL_7ZYz%y^6jB`VCn<(kdpc6gwR=Re>l`TZ-VYlDjB&L5eR$-29y9kA(A5gbLcJ)A zFKT@indYFs!# z(@)HTRqX&bq{84UzY(3k`ISEFQzR)%N+9w1iE`O};w0n!3c7Lq15VY-1zlcv0B+nE zM&G3Dq-H-KLZj;zXxh(1&Y6+4%a@@RRnZkAvsUnDRhH7TUscJU6cfQIM&O`4dya1U zJx5}pWlZ+=m0-zSM;WW1a-2aGPE;u&=xP>u{^|m<-{y=YgYTk-sUPu`*Q3E-z<<_$ z7zNs~&qZMa>g2O?A*mn!h`b0@LD?PRU@^iMm(2f#dmeAX(>oB-NEyV{E!QeWoIeN8 zfBNIn9d>w7!5tJLpOBoMcQ8^?3rueT^p|_$nfXCPM&VV3jzu%l#X7fm2eX?&M>EC#?q?K+Jauicu`*BBs%+v7+vu; z5}&C1Om6xJ9$!-vY2&thYN*mcgBtf>$tO#({V8!g+ck$)of$`8RRghEzXW>eWu*74 zo|IL)Qnw#pxNlCu=-?-PHoPtcqE83H_k~}$JC`gtqY3vaPWK1V-0IDA@W4r$@azH( z^4vl!qCW|`wz?H}60$^6s#4sx7d;j3j8l1E+X^~l&Khh!eh|y~e8hGMuju@lHv}z! zaGHyK%B@<*lI(Zy@E1`RJayN^-@iUWp(9VzimC&UwrW0zM`=^tfJY>)DIP0J8PgLT zn6BHplUgLtWdk={0H<5Vc;1>AC=xuT!-t-O1$`pgvoD40p1Tueu4&lvDT1_ZtcXO z-vtl8+pEdF!)a*T`V?e-YY>+hs|()R&D40!2rzpo$^JGQ%3e%eh4%NK0namU;b-Dj z_E};jd`nMZb@CHPZ+9uy9Hm4*?|w#>d$!@O*_XILKX2G+vL71-gjIBDtz^FvU6ww1 z4w4=Rsq^ONc*tIJYCAdtu7qy~w`Y_D;WePM<2TxvGJrnneH8p-RB?93Y?yoLA-Os- zf*iNY=B`YoP@8rX-j;8~W`&mU^sy|A6|Z4N-f0K788XoD%LjZbPr?KKCZwu{!s0dh zm^rzgGgrwLX)1Lf->a8V!kv0#b65;7c2uMj7704gdt~VY^I4$X9>~qLOMtYlt57+K zgR(L^7-s9szWH*QedxKJ?7C7Ve0S}5QSUgaX>y8^->%@MygccS4SX-!X#RMEhLDHz_Wc5zWetPri-gEAHw$a$Uuab-t*;rkKaV^q8d({PQn) z4S^WBjC$WE%;qQIG^8TM1N8_^VCopm)!72@80Xdj~6k- zAY?;X@MJCM3fV<>o)A>9KEHx12gQlbb4PIY+YNm;3edeY6FmN|4Yn`;fwwn}f!sP} z#<6P)KPK=RJ&>%#zYqC@G&cp4^$KAV#m?MQHm%#k6#LDrrg8r+$~FlPPgIxZFh#hhR-S+)SCf^G1j# z+!WdCw>&A@9ZeI>9I5Nfi=-|}9g6QWaM)Xm&V9CrM#hKJoJsAZH@X03`NiPzNO96^YODK1=xQ~6+PK4iC>AQ!K_66%6t#w%H9#q zl@%{zL3#zildp+%aYF_SNZG-+msi2+$r8Hc*GoFRKZ&kcx{I7un2bxp1iz-i0L;{| z^s2&G=K9-L_(^0WM5#!z&9XW8$!mw?DH80e`K)F9Z z=OxcS*4odPoogl6q94&6g>wAO8)EpL#2U2H@*KIpZ65xx&>Ut*7sDTKASPNI1T{88 zw{s-;$e)DGRna(QdopUfYYZkY29bZh6o2=+AtYJ&L&X#GN|i;^D;+{7Remcr;?MiV zlcO`eDRLe{MTb&<2Su4ANA-9`eA_d~i(3y_`fBW*-Z%7(0->$FyXb{t|B6wr zMeyN^z}qq}BOQlL@KW0%vbMY(A57YU{Ds^KcYYh#9c<@TizN_zJqM4O_#OT_zQ^m% zZlW;;>#1de7nO7WLh4gXIfLM8Ja}IfVim%$eF??o!zc0+Du2*j?_SW4+E+&7+*Oi_qXx1DaY5yliP1Lv+va$nPqTsU z>HJai!Gd8}=Z_oN-wyA?QpT;^i()RuY zazp19$rSrYB&ThHE^!&yzJ5F$IMM;5i$B4wD~@bSTP-ozISs^r^ut3@BduOQX|};x zS~H{_^jgp0K)b!DT(+6$j1hs9>nwWAi6Iv!t5FLz%)j?!@vHGMI66{`u4vm!T~~|I zb>@!LWp+H4l6D85KZ$tk=94h0s1|--&Jp;dgQzxT2vuJ=hVdM2LT9b}0lQC(0qgLL zij!eeko1oLJU4MXd{>f(t?%MMtjn1V(mnuB>%{q-t?G=}V|BDf;O?~yj-WdZI&n3N zX5(*(qv##kYPwmdVOopdW~Wytun7$_*@Yj}S(#W5n0hk`& zrY=L{iC<$ab0=duOacbEr>5H;&kU-;$lz7kQo2#LaioQ;C~V1Si6Vxo{%U+HZ9%$3FuKiSdvptXB1`*vCx`K}GM*g>-Jzt7-`3}gYjBy@S>2RV-PeQqj zqg~jzP6Y9DRzj4~E%@8C9yGc%xvd#4Y@+5I_H%$Q9MG6zHmQAx+4BqK?3}h<`hCq& znyKK3=N(~BL1Gz3x0i<7C-xP6200}Ose)K(0qp&VkL8i zTD?lAf4!P%%9L67+s}6LOH6@ZiX+)kYQZp~wh(Q%kY@GD#aM-$$8gvql@3!Fg1zfs zmyaFV%q;6Tif?aB!%Ny!>D#qOAt@yfLfbS#IzhsF8qvNSk0YA*Lu zZY|xnUXic2^@g?ni7-$0E~mQvEw}q*Aboeognu*b1gXrQMUt(0D_-xpNXA&okcVPv zRIMU`9$h_+woNml1A{lv(w#f#JMVfLl~;s3<)$)B(5H%sgBgPMhzWghZZBDouErL3 z^6;wdgHYu4;xm#({D+84aBmPeU9;=x9?4~V``SVJmWP!Y8J_XdOM^ zUWrQ24dt=_AA0|%pyQbSp3ai-qhd>g;i1M-dhFU`@@`!x_o{3mIJ`Ryd0FA4|7sX9 zytI}3$;h#bZxBf93+Ftn^Koc;G)U=x#$n^4vB{Pj^H}h(WWjQ@1iO3wWAOrNi0e&xvz_ZQj$ja`kFqmo! z8l}#-&moQ2t>1&EZ>hr8I=h4%j|+^5^M(cGr{Q9<9Lvm9Vt1Y`#RV}}LC3!H)93?gP}6M8e2BrVj%P16VPme)b>IyDdm z9%rz*F$t`?NT|It3~u!tqkFR+Q6Fceiob%-tU_K8-7fzFot!6! zOA-ZaeVH+oUH*i(K5WGEZmWTRhJ#dCKEV{;y4~ybhc{pFyj z=TG8x%%j)64d}0x>mbl!9DDVI7?jTl!p{l-kGiMHzIibo`W_mAq33XRWs@N*_9c}u zxh%uZ`dls4%iYLK&9^W-QjP81BTqi}-C$lDzCo6aGX-rPRsO)Duk_{ICWs$9K&ND8 z(M9f`>CfZ$;ATlCd3tCRU*7BwCq@N9N|p>**9~V6n8bqLtjVxj+lCxI)x!lUE+Ih~ zU2wl>JuA@^#xA@0hczqHHp?Q7tbF?w%A41V8V5#ksS7Q5!?)jPkKjS1S!70Q6|YhA zO9}L#t^u@oD$pR=V0z?k1&ugXPB$(oqvoAC*nGPlPCVpB`hJGfyf#zb%GH(k{q&l1 zZ*V3q84mdKiUai5t1v2aE}dT4bz78nWGZ;;#xiTy3jRK=`BeRdG~Y2qkvB?mb>t)=FKE8PU5_9+7R! zq`~7BPn-(dNzFNFe$vzydQS2bpH^$hOHP*Lw|9ro6%&l8p4P|&Ro-n{ z6t-CRh`McOdFKdA@Eh4dEAK1w2ktM0T8jkSk+K-vZL7$KQ$gUfI0e)w(9}yp9$2Z0 z&KI78A%@$zmg*`}AN&XFM-nc(M4C+6X2KPP>+_ncwfM|^!n$t9E!q-f$&Wi8K{tnO zWKP%5WH*15<0a?W^Zv`X@*R%TXhqH&nqB&yXg@m2ihj>!BWees%s-2pmFoxJes;m* z?is9mVip_peh<6U;uIUfOG97SNp!_}Is4gY8q3scv8zrTW2cv>u^ZwC@mV=B_BM@U z&0Hxv|HBB@=(nI@G1dt3ns<`a!ElmWdJ2y2JIt2bTd>IoblA&dud=!>m)ItCIhO0G zVNbS|vZe}$Sbw?8Z28yw?AH?|Y|!&->>;UV?9=zD5Vunnj*JS02X~BN-s&uvyIX>t zGl#$%fP7< z9_*5HWgtD*6GtyRh3zj%qTiDGpz-NFXVK(GtakZusS&QE$7&l$&TlDieX<2d`s=fC zYjxR~i3zv`Z=%7=2$|;|2*YP)B z+q|NI*J#A|1P+y&ml;+(b&x&&vVkpVKFeA+?*`k;b4X9yA}~1PP9mr>x+-b-du*;WC1Z z1T4g5fBNI3BAwKg0aGW3U`|{WRF;=RrEw=aDW#juvMgr3FI&K|6nSbqC5(<4>(5Tu zWg%#=>;=6M(%9|lASth3FX+rDu)#l?v0GvZ);@d?{thjK;l~(uAEyVA3WIRTT%G-E zx`{1azKzvYiwDcT=XA&91RAll10sjSuxT4J*{>2pPA$F=nr~S`{HtA6t? z@1N|15gq5Dd*^m!mi!cw){SA6A5@_J#Fuz=t_aQ?jl*-F_`uT(w?Nd_grlouVY7)3 z*K*1L#-H~=*N*HFJqkX^s;vJDgU@Z?s7_Y-8>&PK>c)}=hy_WN92g?-HBRTc!1eN4 zxUxJM@?G25D#t4JTk==-qy8ZHIPPRM0}$JU<4q?k4~Q(U?-5;e|A~AQJ*Y`kAC-6f zfz7loleBB`*lD6aOKSJCw^kXk`+j7R^DEuBtyZ(hk{?U)uEn>?GtE?(Q6Ho5+7aWa zROKi7PA3Vv-=)Ke`w85k&QXM$Hv}T}_6XcASwV~V9l7b7Lksrpgvc3_LH2JnKAu?! zE~+h{p3p$lH@_z5>cdGu&09Qlb2~TXnIv>R4#$tpw}a*+O%fiJ%B_~*uzgD_S*uxu zAFn(B!Q_ibe(Gd)wwO4Zw|F8eb!Y@TgBwSl4hw)`tphOQQ4%gc_=&sVynt>vd5SLG z{DJ8HGK46O;DgJY;7G_QSf-f+jb2qSCZP+tt*U?@fpTo@$0O`!&m*w$VwZuN+hF)P?&Q$fsLPS$Tp1H&3;%j8CDl$W4o_Lq~v#@z}L$m+TQiVIx8FWmz|(z zQ67F&<<6$+jAFm^n6NOU8tw&ULuc%6_Vt=3gfiyi53RG{f#A`!Ft-8wpO^uAp5%dc zOFXCkse$Hux8YHf+sUSfB`ALPEU3)QfT4Gjz`?Z)CXZeP$CYhi(adM~!qYwQ_jweo zJ{nrVn!AeS8WPd%2WMlkpq0TSPFkI&^DFe!n_;6r{OUx z6Rc+88{{Q;=CVcFFzU!`?(fwy^R)wMbKka|EaRpME|7s_V+xuVVN-uw{ zZ&!Oy>!1C3DcfcLY3prW^k3u8_1}&Bb5vRBpYea+zRyS*`cHq&xl!J_s{-+ZSQQ*P+jBKssFF`+J8Iu-;eqC_W9*w)&IT! z|F`{P&jXqN+cgINUmWwF*Ld_@<-h0Q%C#6IAD4h;J60m|r8}AL18W%d+l9r~3kRw$5a}??`4;FFs@B=Y=8B+5n{ZAP}vnImrCAt7T*dvzZ0qawzwS zB^on&1B%}^8TD%!A{CvvsPNbk;8*yBn-91M|CgXxSG#O0Zyk+&s~-^M&9Xw4bMS$7Cp zknx(itDb-yUYtSo66vUV{!U~Pu>vit8 zzIQb0%MwEqrEa2z@;sz8kbu@t+QW2aZ)BGB?`Jj_Pi2-;mPv?>V5&WRnP)DPnJ`Bl zxw{9WfYcpZYf8e1G#A2>2xI58;b8V`QxIir?JvlKa^&33Z-0GhdxeAXNG-@ zWc)w5GT-!GGBoKA)9cg8ybs70oq2Unv;sGar1Q6VjhA* zYJJeo@d>DJ{tcA<^9(vzKvAmrMKk!JyX!Hn+UE_n&7r&1s?Ww6^`2JhW{S#XO3%0qd!|2lyTt% zGP`^N?U>__?!>G_>o(XS+u)sOo5o7i+meJT9><_VIz8y8jx4?~R~OG)uZ^xsDWR}d zL*!PM%!D*|GHSM$nU1%XXmRLTB%ZzojUSL?%o0?Xb51hME2G;e5}qIj{~yTV;S=;O zqX9DrUJd%3V=rVpx?Poj6M3r^v{k(Y+L|R1 zTkeP4XTL*KG8!G2wG*w{XM%1!*^8LV?3)9#pfh?O0QAYC)WWFyRt*?ne z_glP?rv4lxK4b)vZa55KyZpgMh|^Dxtway|(^2`(Qsi-GF0<^hKQp{*IST8FN0oK4 zX!kxJG~=Tp)*sc4;u`Bv8XyTCrjQVmdd&~7d`AP6IJIPVm`deV)|E9GgH0tnNv^pGiTMOFpu*e zqfuu$G-dO1M4pX6Ilovm{$?nuNlM3--%D`O=4|You7#3DilGeU2F9t7FlE-`(PLY8 z^y0{P{Qa&3mU!5Q@^&er@ktn^UN=Waa4l9g&B1@qoW>TaV$2*3U`Efs!o+_2Ec&%$ zG#VIEh7Og!LQ0D5Xi2m`T6^g@9x23}ThAt7RY?W(^=l0C{&+a^z<#}`+ifY+4JGgLLdr8AF~7Z~MH{?s zFiy`8qxwNrESlMcTs`ljz0N73hD+y}ZML(J+#_wgcDov;(j%~9{4%srA1UdA_qRGobQM6kUdO9wLk$T|DMC!Z{{m7IOCRmB6vaCvkF$BkD7_h9(K|Y4L)4 zX!`K==%Xx)jFZ;Eqv`WuXr~#x2-io=N^Oj2#6zYz%2br@H;vimG05;%{>Kn6z$;;7VG;7L4$@nD$HucP5_dm?UXG)lh zPeb>#G?C$9Z6?LpocZAUh}jWx1YN&piE8gpMMt}m&><}^3;rN8*XOrG=UC~VO0A|+QH)yR? zu;4p67fo0*6|Fxo9_imrM_r4?!p1W>3Wrm-9qy5lA=2}Gfe~e4pBe`g+9#0 z73Y|isDn)OdNsyFX&nArVTw1Noq;&xiY-LLm>lJ3M(@l})FU>Y$;_}} z#wThqB~43+i6~g-2vYMW zXs&AoQtnSfQ5UO3!G1NOx1Yx{yB5eY`{yiX#^11JhGl+6nTNZPm%cK#*?bG>k9J4L zoOIDfk6AeC)PUgIS%TVACc@zlCSY;V9vmOeVaD0FiVVt)nN!1;A+gVEn3V}4(c#wf z$TZa*X;z7&j;H6Dgaflsd}{~#x+?+A%F9B(TR7A_FN{g6l|g<7tx-8QA1yzAnOSMD zKooxClPHcCLqGF&poQl{kez}n>WQ3%zE2r0l8p>9O)x4E4byzi^eCyKLXR0pibx>a z7#n8JiVV@gNH?VI{*oEiJB10bZo=|cBL%%iAKW2e3gq8ZBCCfdQL$zLV=c`w1)D+{ zF6EZ+r<>D`^ETvyM#OLKF51ZWZu(8Ftg z5uzjT;*y)D67#f05vG#Nl-dorSIl zYqz3y&zp?m%paoEXDV2+>I0Hn@EtwT9c7wSS|s}ZLX6>F{ASAhB%aHJ>cn#wCZ5?nWDg3E2o(MNZo$0XqJ~UwWS|2bfoSA-2_&}R zBy+3s6l3-MAoIOs8Y1qcC|jllMK7vh3fb>Wp{pbc+!M-Vs(%$#%Lka$bYnC^c(%;w zd@j11AdhsL_A+OV9ub|0K7&*qrl1E#K`1N|M7EEQGZQOhP{yyj=;@OWDDLPNr2A9w z*4ptG{qp*P-2N9u*B#E~*N4d#ijoSYvQkFzJ)iqL(m;fUlF?9_RH#Uc%1k098j6w? zqA2a9QlX@1D9SIPr4rifecr$P<9e>^`&{Qa=RWu69{r*u?jOq~iINH$vwawk7(slU z=>)#x-&mShn$2u;&DrsJ;gYVG1L^EQS@b=tgQUez==r|>XnwVZ*`Bmue<#ZkQWtg|n#r9vsjc3)zIPRR4cm)YNgvF&aff@@G>q(4 z41%ddzITIW0wQ(fV1hW)1v1!s2;qr9Gv_(OTgN@%KOe9j~v*yjIa8ebUy zU=Q8|UEZVGjt`8g!nU^u>2v2L4DovbSI;c;U4H_mITdhDPr@U^OcvC7O%jlrVl}!t znLc#%r6=Yu*vU(tuy*yw9aTB3@tz2|e_gO5AglRf9&JJ zBfTE{h(mvVV(@p&F45y1ZZh0xuIO#J)x|U}=#o*G3@O)@k^E01@#=V=9fFQ>Vdr%xXL>H{^dO{=*cbT=0W+_&uf_Vfm+!7Jilgr?ZFG?3Ux(k1O#d zw16K;ILehqMN0PB>9T$gYnjAtCQg0PgllRk4JvE1DyfK(cpqw$%s$^j-|y(*Tagr$ zSN26`<9W!=(Bv`}%DC6>0~t=bMcS1?uq_+R<_D~m><}|D?p1xpSa?9%!3uuyM_~W_ zD}Ii0#-UsrOrPwIGiM5T;-Mz~L;4g?QC)}2{+&?$&Um?z7KXSSqol}*w0Fz~%;`4+ z-;>p06)_6WWhcSE+Y{q6-{H(CZGP^M5#OV}6mnhhxPE0XoRj?M=lSJgy^J~I5|=uZ@qV*SIaWU3Eq%wjuF}#*9m(k2MeI#{F#9qKXa%HD^pLAMehWPKT<#| zytT;j%{z)qlEPKzE*keKn{2KfrSm5RVbdZV;pDg8!ouf;lGc_t>}5@Va(?6ocFG)2 z?=8W+epAr+VjM>5+KcyGS-OyCO6f0*=yl+9OfnsX!H2Zq+g(N7x|!rZ+?sMkKT??3 zQS4%#f%UloG{sd73Vu^DF=w#VKY1TX^?*uN6I4!Pm44BbeVue=@+-R5Ttlx5Q%UdI zcnn=I5iNSonCKaaJqL84z{{xY#!sH5_>`MT-RFx=R={t;UEC}Y&#LBLn10FyGI}fF zwWU9qxOh?z5!+~p>vZY#HzXT01eZ2k<}coy6i43>Yl$2#LCN*^1+gF?LciBp_=Sxv&zRa5YZ5u$k*u|6B=f|4z%A1PpWU(BMd zU*bvV>MPW?=?JO{`ofk2#(c+YN8aG<#vNY{LCO0&G;nbuW%ski?W%Q{-yDx!nL}Zv zG8prFD&hA(eTpf*MY5lsk?zzOKDE`4%l&rZ{aj4R*;AQP-S;v{^+bx(O(VO=3L2Xu zBk5hLOZ7jTY4_9w>NczQnxVk$7k{OlD=jciYAu}1_Myu!7VjlX@uRhpkCVB= z@0p$Aez_YdtZ^V)a9D@D_%ejoNF&-fjT%d)L8m|&tIQ0D?pk79pcM2Dc$2rM3b)yt zh!Y>$sMYTb)7u$ALpK*w{gJ+R9-8vSm^|wn(xhJGC?aJqise&*)NBt#AMKUjTE-Rb_NAVf8$gAA?R)z zE?Q%=NwY~6>lk>>>+vD?n$0zGidm{ z3U*9AjTX;WVLudZl1cV(tbFyE;$qE7%Xq<{l$N z43lwy1|7Xk@@CfXTK5!v_O8MBB@6K2w7=wO=w*q+nM~Gc7z*S4-zd#FxuyAr&n z@yUs8SQR*ka;D|7H(#ew!8l7gmTHThBZI}foQqa4zhbFmswz=fZ-9;amFYwJMJC_c zibW}znD9IwOO9n@p5Io)KZ}5!Z-PX9LKYivY%d*f)|7a^%aXJ;eqz#VL=NP{EPkuC z6Xn;exprJPTFsu&bp5&5eC{yw7_y889X^4EvaR?oe+Y}iwvcyr3MqQdA@}nAtoqSY zw)BS|J!vgt&ZGaMW%rMwVaR`EHYAy9S9sByAT?-wEhkMsRk$a`lGl*=6k97zqas%l zoqkE|mOjejUa|d~jc)A*L@WMtQ9?CYYS0XJK~aJ|&qidS1kT1U%zIW;{Tb{9RNwF2TXGyD{woHws)9Q0+kkMAkiti@R-GoS;!Dze@#NzsXrrC3LW6%I~*8wBDl+~`H1sRLviB?eBE=CyriUH ze^wWMX9`)n-yqgjV;g*K`WyW*L942sr1pbNGOp>X38HplIw2dYYRRd%Oj=-U={^W``y z9#cV#uXc{j$DZj>dn;Lc0Tb5R4HiQh@pvM_86tb@sje<*z= z&5zCCT*h3MTe?4mt@3|_OKCKG*ABY$Fc~aU7LRNX6RKWsk_l% zGI5_47KPrYF_TPbMRx(c$ZBENyRNhF;kUu%&tq+IG;%UN!t}{T963-=Uv&c|rX_>e z_^A7=wMPUD?u~|9ULRt99c|Lpd6c5*kS%^aA-PjHPf- zCq8qBJ)c(GlYes8wR&fkD|ukEjxJ38jvDpH;vD4(bRRyk#7#|loD_~x0kt%E;bnqY zIxuL@b9lRN!alP*)KgW$jZz!nzh*LaZ=5;>%V9enOw4A6JY z`%b%IgtsTWo5o?Y#09d>0s@34`Vw@Sa!R&R<;DZl$1_#@ewR~Rst?YtoR6IesrViG zfrcv>L2_Xd>Zg37;<9LjnBJ!jI~mH1+Cwu=MAP;6Pw-K%5SupaM0;>9UD6m%qsyMK zVMC{i`|}jhlQsb5J}H=L8-aDVmcdc}tZ=NjP&hsNu+V;PFM5{dV9K}?aJ5fkmAAi0 z?(}k#jJN#F;axweZpUZ_2{+23l77c)A!Qrtop+x7NoeCIXt{eN#e89N^VeUED(INKN2hy zVdXQAWDaCdxN$%Hx)>?mZ{x7LqZkL?)bc>O!SsGh$L|ac-3QW}=+CarwIU{4 zOs~(ZWm8RuQk)P;s(LAOspK$y{Q8u1cC^8>LF|4WvIF^JUHH{veXbv-z!%t+Q%adN zmMKTWWm^UdH@m^QhYu%V^)lS{GDh}E4ZQm+PZeE?G-GucySrGCTMg_)*Qr=kR!8tj zVs=FIDX|NBn-M(Yz zvQ<~eU8WEE))^u(=)R;@Yb%rOaH71AUSt=wo4(AbqNy?OnBKGyDt-QzE?3m^x-4y> z`IV<|@l0><5&B4X8w>@VEu#*8^tGHdTvihI%7n%RKB8&TmdKjD7G^yhvHg`7`sdz7 z=^!J%-#DH}AB(`$<_L^)N=3VpG0ls1q>Eziud2#ta*@}9-q7KQZ|a4|@+zpf)nDILhVS&NeeOv4|_`V&5-S%|rLj>i87E;~bT5LDf;w)L0e-@_E%1LAB*3xBE zZEnEy6w=r?DG!Rw)+FDH>!{wZmb!}zknV8`-#x0Z!_XS8iX6X`H{*FEgGs|<+%$FJ z>we~ui@q+x%jP0sWC&|rJCS-{{7lRCmogLmd{*ga&pP93*$4;FV9!hBar`)X5B-6t zZUcV$(*)9MO{Nj$-|4f(0Q~Y8g*iIw@z&e`_U04e6TTTstD2G9@)HlPYx6C>C&>Ny zVC+#G54TT8S=y!sc3@I1>(TeVNPIOxTbv>7t~|qK?l$v~XceK^emq(|+%SEiKh|ei z!F%5XbQiiI_G~{^F_S2-QUSk<@{sO#9X|$sLXWIe%CmOH(NcfZhjLoEu7bAS_rfn( zSID27hH0DUp!A&^MP0F^>!$kTI!7Mu-I|yYG7`Pd?v`W+ZDPf{Z!%k_0k}5qFjD8X z!R}==`OcHVpjHPAK5qtnE0H@IEeG>X2XZqEr`-SI=QeovoH zCeE4&cpZ=bc3whpR1d!H_;YR|_Q-X2zTk$7#Cf?%91==5VEMXmwsDvZ_4yt|vu~!8 zASZw9qXUtM&&~f(a^kn zoJlo=slGjGAH>t&dq@65rTR&*1eKz5OC4CSAj~ z_jf7HOabQ~^oD!7mr$n^DYRY75N?a>^5xiOns>M_GO`>kWhEw7Z#8qQq+WR7YOEnb zS2ohgu4X(QlZjfpRQL|G6e==A&8xSSuX#9-!~G)ZCA+C z+8X8-L6FIfhV!aOv=5twL!}J6l?LHRl>!>nhQQ*mI{0jk)pIy~lxM9{9jsM2aKIKb4Z5 z3-iTwT!NvCufXT80iWcf&Obb?gF#3=jk4L{=!TT2MR*-@1wB6M z^QP|$d1rMfAMqf8pBeIsy3cN=X$h0*OX^X|l2Sy{`H_fS_ZFGE9^u@B7ck236-MoM z7T$V~5po>dc<+P>JZcQ_fz$V6?4e_LUh)L7J$qxIgA)2ZutZw228wSyqszU{lb5{$ zFWm7G^9(M-@=gsdTJ+`P_PBAMt%Y=S*%XR#?V{d;7U0a|*;taY7BjteGS_Jj*!<4( z)HY-s+iY-{oonvHHakV)?!e1vh zpS~TA?PqCX@oM_AF@jiO3Z7i|6@R-uq8kGEVUGxIn-#&E#$V?{(w}pOg$jb1=WWv3 z{eir{c9QC(z3lb999H-^lH89aOUx_0*_vcaGI&#s@R?ijdg3kGYtTi@R(>TTIenx& zWVjhChtDBZlpWcNZabHf@02Q1s2_py!D;9}-VDnJZ-jMX1PrAdus$h{Hcq@i-O`ri z5>-O`?i5mbXfsuhe#(sXN=YNK7bX;>vY`?B#MNUd(aD=1(oN_7vgLfh=eeSeL>WVC zAJeE4!|3fZOJ*m$8hzEK8@}g_c(=?H?*4fWiao@8-|jf-*IB?{7gq9_r|S8ie@(bx zR*8A-Sy))9iSd|;4}-%ob-^GijLl{SmAhF(`yt#N(MB(G-6(MHbo_a-4ohs;;{fTv zf8#(X_vwi#Pd-Q<@7yi9F2f~VdM5l)sUc6fuOVjed}MP&?n(}OkCX%_RIMmK}{0^%7aQ?drg5-KDs@UjkfjA*t$`km|1j@`h0AZw5(B=G-T_sQ9s;Rm6Qw4iFpmBA+2ak zskNG4HbN4zT8iC2oJz9mirK<-MXW0NAZAs}$6w3Cv~bl*%9XEUFY6ju+wW~ONNy^W z_K2K7Wil>Ren)Z7`}i`^hx>fe=KWjJ5${w-iznw%Kxz&>k*JaFl80f%X zjIzb6k}I_TZ8cMF{my!={mHJ+P(x0e3c9cL!Yf-}T*6GP*yj?*|*YdjBdN3+Um^vt!yv}hfiP*R7db}H_4 zt;6-;I1HF*!P3JL*v(6EEa$*)dT}>~PIOP8-KQ)`Mj?W1*OropMsHqXQ6}m;VsPWJ z3Ie=)L-Xt^^q;YWFW6+q)jPyF)QU>vKCHrWwQ4lktcTfMJ$P3=p@UjRIH+ce_L1IL zSTPNr20E}Qdqj=}b9mqjSAJV}5f5FMgR{lwFejuEJzpkaTW~6l%AG=pgCZ-iIlz7d zeqwJH4B`Lc#k>F;fqzRFfvLsQG49nO7)`%Jp8m7RHsk;~pU$Ht3QhDg?k`<(H5X1Q zDGHVQKJyq+ZbheD)_aAqn`gAEA92C*eJ_IYTm}9Wz z7I=EaAj*FoemD>0Q+~Ig&#g9$ns|s+UyEg5^S3aqrPB}}@`@aSR?|S~AbfXDK<~ci zar|@+Tll?+t?Mdbx0`zK8&4Z>eceC!4*iFgULx+)%SVyUIJ$f?hLu|;F=M&2OgdsN zZTb^TA9dF8WaoO!?g&9m+gd#Bok?#V>rp|&3u<`ni$Ut?IQ)JgU)=vN@2IcikCxaW z>6ItSt3`a*znUEzqea`BqiLMH1^wu|nZ~ZUOQR&Zlo@c6-H6`L4hu$DrkH}Q{T}0Y z&~p)w%|Pwj^=MI(VV8@pvFf}n)GX6qGB#p38?X47o%P;E4$?~~vX?$-mY=0TFU&Ew zXd8T`r_rWcpx8{Nmi%D~k&(1CIFI@dk`@wYy9QTQ0eslu4avkh5jVGIj_h~|> zJpZ^=hL^?5^L)e8%qR9H9cgsHHC09ar=u@#nJCR?s0L8l{Seynbtru@e!&KoPNc&# z)>2n`CCUF3@nQCU)I93L3!3Y2!}nmA>t)nwc0eyk52bdG@X5oUA3U>?y+|BG zVeWEhsJ3-~osm|g)u=4}{c&hXHeJbFqJPUKo?qJk_Pgy&2j6EsACW{99^%MVpqd+ z^3MojWe%eLNZSrh$IsDfkI8&RSsyNwQ-*0NE_j{Du&h?>kWz@jpsb6O9O_Aq`A<=& zl#aNDGBUexkbW(`M%gK=#kuAXNa>5(#^gFO)1Qi5nFH{2PeEUuYWaYFoCGC;y z$A!lzX=sAoYX$ymR||D(_D0m-6Kt4~ z-JxNDj`Z2woqqrK3bT)#K%bRGFv{%B`$QDthM5z7OgT$MdJ1Uw>xE|P0~8lOo%9`g zQeJ8jUwJ;98->X8U%&HUo+o~rOSjYV>KM$RarkfPAjHoI=Cx{vxy#8rJgfHw`ZO_< z)X0Xme@cO!a}1t!rJ!%GM0&gK26_MPAZ_(h1lVR^YtO?l5_7nt_r8Y8xJ$U+6i-_g zihQEMKvt`N4`cQ{L0v!v*mx~IJLWmg#GZj~`FyVNR@b0;{uwwXpT%5TE6|3-jDK4J|xY!9VXAT!`eA5)NYs9>pCwWw4 zU_*8H3b0|SC)PRuM+;hzTyzVehG7`Hx`j2)iI+4r++^>{WAHsF9DhbGL9jtM6wh{1 zx4jQpFIt8f0pkCDw1bbOxZju-(U68~LyOdOB1?pxrvItq_{ zY{+W$2%4E4OOtahqU*nLXdZlpYLsqZPx=vjAJ`v1J>T=Mc^CNlz%*VHnu`{ejHfLd zFxl(}9daCq3FijGPO2VkLM^T?EQR~33~qcbnLl~4iTASfgPK7gW{jQ-`ASif^zt*k z-lGW1zhUsaY)rgE!djkuGcmVcQ zbkV^LLvZ`x3*;r`qT-#X(fS*YS9=G-^m{0MSs2T;{I>C&%xK=np^MF_^`M$e8TzVU zMBP!#>7QZ*eM<2V&xg&dwzgLCwe18Iu023!N@7UCs{|`n)nl=pJl~}|1@%MIFd|Gm zD+cVg8mjQws-|XvWY>TCIGa6&BttSJx^kiXZ^9zZ`?&DJB{R^saRJ=qd@$pEGI!3H z#FPF!L`PXM&sW~cYa8u(OouFfDA&{HQ8`pH#vA4v_Q2962iArfP{|q!kIX5s@l(Le z4aV5N#}>al32KHV7ELb8t^ zV_4uT=+On-oMH!WoTdS4jcn?i6g*XQ#`-u@%wJ`WyUY9FaOWSoGD-t$rmVwV$EzqS z@D=9m6Fb|^OcpBCd-E+8zj1MU4PG8xLe8q2DD0y@&0ctcE*y74uOo-?@mB_I^Gv6& zl0!7N`z)WX@PPmE`oyP8nrYQT6-+7~gzGH}=|K5c3aih>o%OTOXF@&Y6wIMYSxq`U zN1OGscr8hMCdFr-EVf!`L2*6kh2~2Tae? zh1cJCth=1Bz~vkF*xj2?T)hMGy4UGj?h;IHPDH}VLzq$#N&1@|XugviSsOc%^ZgT~ z@t}xK+dt-GrGD}rx>7>hI~#gjj&h{Bkmi6peGYdm-lAUuW{v=T#<^0-Fp$*W{oi-#tHD# zXGVq&{NR<1Tp>S}FZePXQyv>(wSyz9C;nz*9xR|&i_$4#Sukx|yPa&@7ZOhoz@Yb| zabeXx5u+7iUG@>gk59nJHAT4F76r#AdWi7NVD-J`P^$mooA-v zyT)#`-hKu(kx$mx8G|tuD!l&IXPi!+55Gg%lv8z_+j{KcKh!+=1z`|vGkil4Pr|YO zl_FE|S0)=5Cwh4049(glO$wUElJiD~c-^64yz%yNn8e8PaYi+;3FtugL{*H*dlO(yw^%Xvao#xu#zvACD1ukpu z$=y`9@*mSi@e#d`!t!9G7%^guhEH=4bvNHKBlTpqIg^8OmyvZIr7tQO}_sxw}9Vmzrw>6(?#vH3N;RB zWy}Aakl3lK(Hn^pEVSb#D>wf~g^&B8W4$+$3~oSiOf!bN2#1KX9(lsBa{=sY!&)x(7Jls{liW6^K3qCMnuKpS0pdPAo`{ zn~e{_YlX8EwDcZrS(Jw8>pw+5(p`2nPn$Z7hM@e*ajcyjLu$eaNpty8tDkYR@keen z@43N#8*k#=n%&%Gt(eE9xK+3}Hb_$J^OC(%btb(X8RBf+mp;v%M0cG0(PQTr7Wr;;@IC@ByZ%z*ja%flqK1Zj zE~2B~&(W#TRU|vgf)u~%QT=Fv`sek+Wo^Jm-W|6JL`;%#n9a(GVXH;o;}+R6I1TND z@rU`;_nxSmc^OB8H;BH4(N*lkhzo4;k`pZKurjmH%VlX-QsB93KIA7`q2GkJ)b&vp zi7h>`O&G>Up8tV|{y*^RU=&@|lBV^p{ivfj9)s=t@yFF3Vdws1d9875>S#6Ay+Z>v zwx&?oJ{Eo>OEEeu7H>>L(e`ctioY7+=SmNp{p|!Jw>j9kBpRCv7}?qmrz>8w*^w3K@ey59& zXm}6!eF027_u^A!S&{QI*5r}K9JAKq^lK;V*)tJ=yAF_&?*aN0A4fx-^x&Af5``0^ zQLUxIU8dcDOMW(t!(*x1y`H*f$52@qq5o)GjIXx9c4vVuj-N-3u2t-Cdm4LnV>~k} z3t&56&Vo(44~*Ytin*=}FlwznmM3Xqv85VpT4nHL!Wo({a~R#7nMgaFL+Q@O-h5Ei zS#-^9Jw|TE9>-OIok!{;vHa`(RNs`SLLgk4Y;P8GHY%#ic*wj zf0I|zph?`T5s@{X6oo8MAsD_<>gi$g|FHE)@u6kezI_I`$|D3f0GdWID|RHy3v>|Kd9x{ zwNWMA;g%LVoFy4eN>E&^g@cBVXmgboic&Oi`MiMhnxanKCy&3XEah>7WDzobA?Y_o zSryHyp`gxtB(=Yt!mR(1TjhB9+emGV1 zBELvf`4*NH+_Tk}KWv?5coS=4yH z+xVG`EXrukW6`_(t`92J3~1gPBj|4XK<~#KCGQLWkgz>~3xD%?okb`zJGR03T{Z@{ z{H9CFgOE3*jkc%Wr;^T{R{Rc}XU+#0T`b&yx)m-3cGZETo14uAeS zVz8!!M)X`svl8}E_da9MSEjLA;%UG`8>BBY8t>l$Ek0NI46>)}t3dJi4`A;ZK$%z3XO)mo>~+#^pY{nZ{^{d66M)>qNbK$|vA+e&@B zDoE;hBBgp)v!W#mY+=(6TJGLWEn7}eMMk-mOXLI=y8IFqn$Bm^DxaB~vNrWtDB)LU zSn~8DLvG}g&+F3Dc=DJy?rZd#e&0S z5>dM{TxLET!h6B4WGeDo#^A^!agNbuK*u$QkmI9#c%OWav${9%Lnnvo4V(~r@hHMq zjKOqiZ~W372o>W)@LW6rm%_f#gT^xIGhshHcypSvN%T^BC!tbhGWxlG!b9l>6mS28 zc`0$iQn@+8$SuagwC9%GXsagw7o^T*Pl@N@H47R!dp*tG^;FVUCPnT~XVQ?%hbe7h z5)GSRPC+&rf+udqrQwD~_>TrzF-wo|Q!XfJ^OT_=tG{^)KRKc+ei5N`J`a zujnSU{AYD=?IWv|b9Pa@X)>KL&Zeh6rZ7r4M_N-2X^ct@JF!HS7WszIw3Jwy|4WUo zdMu(=ABi9vHcaTXNLtXI(~i9N7ZJ5}Ayfn7QM2(nO156Z(@TQH@jqFn9_z&B{m7uL z7Aa(++LdM)D`ejJeN12|5K`@!eTKE8;7ynh`^$pHyjRS0`5ODS%9|6KH86 ze{g;ykJhl|*56Lz#MW^1KmU!YtiFp{%fqC2JBf__%qT5ApY;$KESnchD0w9aR-R*o z5gnuXFT)vp=&`AMSoLxwCCK5t{Q+_{3_*iR7^axpVD^|y8X1^M0pAqKZ*&$7e?5$( z)~M2A(cj$de}T{HDCG{TBxEDbJ8{jD1;4pLon#0r=S^5&=xjCanz5C1t%GIr5Lr|@ z^+L>ub7T`P#n;E}fyHB8?0#j1-H@h;L(K^)b_&~eajD{$BgSpkedW&gNGzZSK$N7(&1r06rpN+C5mSUNTlql zESnzH(5%*Dq;p~)xzy#5?*TX3>^qgxTpzJJmf?gD1FW>F#70&l>aFgf?XCyK$Ei|w z`cJmrH~_o8&qG$FX2Joa_qHZ+uuBa(gQ&8BVXxywT6!*nZPrM{FRKJtDB9t%hXn11KT^-L zMl>aM7>?PWqn{cQQeLi(0Jm>+X68dGlk=8jjkB^^oP5vn!J&uA)JeoxOYP~D>zIk7 zXOH0N-*glYzijpD#w1B^m457jL>e=mG}47vvS_|~jR(!S!UM)%;w_>cHe#|8=BQTF z=G<>=U{57hTz^ke-}oQf{jQWfjXc82mrvo-L=M@1?0wAjD1o2j0d)0T4)>sFRvNaB zg_a#-%|n*J<$VRkh2CRN`>Tl^OJDfSzt6ZhHXCY+%+lh= zA*OAm%0A4zL&d#P*v7-tB?iaU@UQ6)Y1}$QTbA_ahj;Z9{iJ_k7UF^wdwV2tXWUHD z;Qn{o@wVs$4pc6J-w7{N-L%D>*MD(&ZyZ+UHIVViNE+UHlhnT&fWlqzRev&;c_c#d zSur0o;u_!Ec8fO`F@EqP@w^*+*nYKSF@oS$t^Uh0`iF zklj(qwp`oF)WeJ=8ae~Nu;%9<~TQxB%S~mP1qQ-wplX3-gVm;5l>)RC~P=wb@;C%X=Ga z0tY~&b_s3XJwS-wZ75ucv=lCPT|nKhN?f~mOPn*kMlZF;a9Mf*UC#@#;AsMyiU(ka zQWv_?Zee-)avU}O#T6xD4sy6Vcj|10jO$M9eQkdH73}1QtG9UiCgs*DT;$HRDFi+or+%Ixm^NJeJXnl&^Q9AtnxJV3E{7wpq zJ)yGyHaY#Zr_>9F_;b(wJXklI->P^+InL`Tq*;L+`)A_W)!i5>zZ>&+KVq9~PBWXM z)@-8I31S;>(ljLnd~F`j$9gj!Q?Jaum<;FrdhqLMuMlh!!QRXjSnv5J%zNNkQpvPr z`!&lX@-6XvN=qR3360`MPg(FMJ5+e0$$jK}Kfsr#Oy_QA{@|N~vCukBLGUkV;Nezb zOnZC;RnOT@&B1~6_C^QE=Ido9dj8jX-6=vybPn8e<8*&pSHIbQp2N( zBL8@iw2nQcIbHH}tY9Vr?BXG+{?LB2ky<1D@#xwqSbj$!D9v4-IE3HpH>)| zCqOl}0;dA|@Uxdkb8{2M!Z+|0k9HJ5IS91L@(So0$K%9CA_@5H;zx zI9v0^JBK-Fj_V8kR&(sUF#wAHMshn#Grl!x79Vyxo5sKAhCFr!?1o+=eNM1$dV54hn5%d{Fi>K02j?m4|;I z714({&}$n#`Mj1YFKnfI*NRELp&y<+@1g3i>0yf31b#tY$_z=8C@e+ir{Spd|z@c9*fWw+0{m z+JLuh7{XOjnjk%DCy(fPl$U>VC)`#QbWL&^||Pq^AZUO*J%H)Nt9w{LdPZyVyC>W zFju#B8trUB{RSUpC3W`fxbHQL>O6wi%cS_z+I&a~*CKPSK9(UJ=MGKAyes*%_V^O2 z8FqsF{z+k~{~uU2{lmG)t1wTTNL>@B;_;jFcoWjZqK!1D^Zb2@k~rVk@+KI@9k*#r z&yhly`fEN_lFpUI=jw>Mz@eg7E$iABmYZrqW_|Rjy!#id_3q=Maw%j-JS3xWrIc#k z7Xv>;V`)Muy0^sQ&;CX@B;;V7V=DHU?qPq*udzpZB8UhL!(peD(C^V7k*AB<#pXhm zvbTeFjN2#i`*4G8oO6MWMJfM}qVtZ+@omF+5oxL@rI6B6X+GC^T(@LJl3iI@*+R<7 z)-LTqMWi%nX=nQ!zfWn`0;L`c?qzkm9xPkEmEzOM5;zTbmySS`apY`O<^jdU!m ze2k3oGJK`ifAo9PV5;ys!kqifrqan})M!vlJ7$=3A^A8GZHHo7PBMBg9Yd$8q?wc5 z0gO3OhjZ)1d+qmTTy%d2gTsUP)CuL}HO~zS5AMMAsS3rvZD&g@vYB&3Gt>Jvj2=WS zpftmAmn_9o$RC+ZEbWozzVZQY{tXgDNNQRkG%}{X6JveVX|%U zEObRrvd+FiGJ4G$54A@9mG$IhgKd2N)ssf>9&KEci|^khv;jYRi+!AQfj4!0=e-VxqL)kxY6JFT%`_vly858$PdrZ7HnWdnHakz` zE{&!Mn6oSpPv;$k`MNTa43k54#d7#=87w~Y_Tj>fT0A>w&#tSqux&Y;>D*6C{>oP1 zr!?pCuSfflv-E1xyRlPDi{`P9dpNYA1&H_`~)piHbLKylQB>_M68J}L%H2q7^gqQQJ*?2 zj#kIOHDlpmKLbW5`|<}}Cj9eC2cFvyN~`yF(dV3D=s5hBOb2X$>f>e@{t)w!+s@G# zJ0;BQcM==r`tY~ebNMc*5LzqO10hSqzt`aa+vw4j5{}Aj@tT8RcW8-&tbQeXbY~q#_{B2OiPxKdU0KbxR>;z=%V88Zq90t&4oBW+IZSiyM~!oR z$^KF{%~&2o<0b@Ab<=q^v8jO$ed`VNmVvNI3nIq~d3#QB^WUZ)~0EPmWm*p#M( z4M&x+VpFBaIW5M`i3d>k{S=-&8;Jtp0IF6$z!|TnB=6-x!*<_f=UT#`sNY2guMMT` z-Va&);wHBDs~go>AAs5Z(=f<=4m=o(=ofv-X0a!8uov_G3lE`j#~IuTP{hD~Y1B{5 z12m1YK+x4_>Tb4XBQsAy>vAmCZkvFI3UfuTOAg}14AGe_fr5uepip8PEZnH{L4aKe z`kr`%TeYj`g0?Oj(f!HBLp2>b-KU{z_!-BWBG7A!8Pq0)Q-i;%u=}2-Fm&<&;h&ga zPEOy2F;PVrIVT?nUW@Per&lpDRLt>IC($RB4@^09J%yxsP!w;ad2t~ax+M-CLImDg zAEobO3hDCBNAxQp9Kw|lbhj?T&v$D4(d%*i$AzW5+2}kwfB!V=?dHH-PJSha=yv+J z_zwj-3_;b{Sn745l1;x`1m~qm`0{8GTtv3_)NU(8Rm_CtjemUZL3v@-yx|JhL~0{WBAyR4>|f9$BG-Ue1JQet#+YfiOBbSUI@Q6clg^ISGeQ-QZB{9 zP~5c!-f?r#9KBNH3fI7BK_uKy^nvf@ILfOSNLas*ANlOhtw;LupUcH;ib(<9R;8lj zz7(voGGp9%zs$*8QRWoh zM&v_jS0ZlDI47~&{EQ8Jafn8W`|^#;x3K{lmXeMm;ZXEdLW}5bJGe6o%=tXD{d)72 zhhF2pGsE5I;`t)~8lU_=iA$vo<0psYk>=KXvKca6%m4*aeZ^dI-9Lp6b)=F-za*-; z;7V_HSHr`z4T=-`aqIp{yj`jnZ(^UY@xC-<4nCm7PZi|1Y%8Wow!mX(DBK6%!EB)e z70XUztW_$;tzC-a%RW#;MjCs)}ekNtAt&|wE5_@j|!-9gX*T(MJ3`k}50Z<%61R#Ib17qn z^D}`j$b~jfSr2 zB13g4tm_p(g+aHO)zTfzM`bG#gXUmqjX0mOTMYvx16VZ5L95JI()XE|{gn)1x*c6u zD_xAK{Z@gDlUR?g?JV5Jo|Qj%Cb}K}lAe`4hPs5%XlkL18=c~g={dSrpF`fF{dgvN z(p<;rQQopwZ0PI=E_d9I_wwAqU$3(x+r|_L4~gzrow|N-;xt#X6%hL?WTNV_tmn}3MMQu&$ zl=@1GyxPO*qs>RrYcdo8&bBzRY8d8TmBQf*<0O4cjG5ona27Z031(ZDVwqDQj6UoX z{;pmssFjWsGK?2+Df1nC(xp(o-$xg*e|w^SLpklUeu7h4XYqOYL3nTN$^GB-QqFe*O781p_`o$CNuZ6_LiDDi)ulCDELB87_twaiI7tZrQa^diF=U zyj~H>(+ALrPz!Q;WI|()=U{EZAtV_2VYh7|`>FVZNw@2e>(oTLBbG!qt#e|w#eeC{ zw{!F%@&pxo4CfOxthoD5Bi=S|BtL%Hh@Y|;#En-~Li3~Ol6%{LFLl#-P1tz8Mb4Jr zq1!OMpMmqg*5ZMx3aKVlQGx7Cw1~O3XNk#_J6;-lz@3&oGXDM+i_m_N>j-j?CdSo+g1Ks}}jRP-3p=2qZ%@*cxtrwSh=Z5e6 z<}xE>SoVcoX99gty-e3?HIa5~0VI)r=(%H1YO(jDMR z52~^$^LGx$O*)Au|91ndc_7aC#F_B7U*s=y8Rk`Gm_0B6f37MP@ftoTD!gJrNyFT=tzhPaAqrbR!HBUR4he4D^)+vjR;g#Jc>1lNXUTHJvN<&f+C2<9M$DM#7dlMS;z~#}CP0CAS=FvfS5)N;<^nY0wFB88@0N8fR0u zM=WV4pP}^^)7fp?v&{0E=yLq3j_$(!$nwgA_k%zZ%eCp8$x|}SxkqKQr94l~L17HPUXZ#M-h zEF-1UN_6eYGCF;`H_h$PB=eoxd|AjKepvkfeEZx(WUR)LWpfb?4>?R5OrKM7!4f$3 zh-PmOxv<3e=geo)c(Rz&pKPaPlIGqAG-|~#^02$e{@uS$N17U0oN+Pb&3(-5bsWfd zLK<}jSHN>^G=Bb5gyWkjlEASFtn+CVyOf&8>YOFCSaT4)D!9gu{+l&WL@L(GB5(sUvosCb~=<@ zsxjquGUk06g&T5L==>`qNl8~S+jjFfDO^axU-J*-GQE-_>it;v0?`e=N%RF3m*BTd z8%Byb&JPv|2rajR{`B=2T|I$r>7>z$D_`lBu?E%7>j`}p3)${|g5B1U!skFdnj zd&Od;unKq2o~51a}%acm4$R~>7TiUYBOox^}0X*^lu=SN>tj&|6b#G~xXa{p?jk;0 zgy#Kdw|I&JIiu-RW)S^unM!M0VtIX?3yOc^IbhWXx)- zMYQz;iTaUJ#^!b~spxs!Ovjo}+HcIyn1*07{6BBi<4rRuM!u)OcUdTZ*nvCGJ5aMbM&$d_@Fp`vTSB4&hw?ucXgu zV`gVx!{)qIz~`&u5Y;5Id*9PBD2t(M?;ncz(oE(Kv6S(9HO<(hiiMM>V~fKQXk$Kx znVg5?95uc$)=Q|rAl`xhMF?wn3Chho(UGsk4V@I|{H)PTPk)2$*|#y^&zB-$&XPmN?M%Sfm{<`uhSJPJ!F1F9W$*acVUQG_#6RIgh2`C6iLA|XHX8dP{3BT{s zN4scpIRlFAdxUxZ%td}eGv?j62NiuU9NHr-xTCKIIuyU!>uD%6!Ofoc(ok2=-He9j`~O6TyK+S z!bS@Ht;?^i3+692Rr8;2AIWLpNqX>Q4?Sr($?~V(W^B6!ZMBYrs)-w{b~n*ZBOR1C z&%(U6nRq>|FaNYbpD$WElvkOC!6qjZP9agqyA;CGM5j~O$Q4ZPl@2|euR?{v3gj}_ zni9TGrSKoIbS+mC?Qa9|dcaLA|GAF8do-FyWcK3&kA33}mPF%cn?n zVIA#X`H=Lr?y>cFV-db%zEcqP;tjyN}SzAX{pwT4RXQ1n+cd`Q-l6iv-n?=rTm^_0N?iS8t!zT zgXg)IID6L&y?V*wR+lnvZP`br?Xu{-Dg%5G+o-MJ2mhN%kJNV4 z;p=ZHbD0n69neJI{vK#;9YBX}OsDOE3n<`JGR*2PLngKdkI2rZplEZviCBp!eu-_? z_{P@F)1^llh3{PTV`7jjMc=7D^8&3wDxpaEn45PRhsAhlZ5$%aj}oKCoED zc#&bqr!`wmU|O>Z#)nHG?fV-C#n9~RX_;ad#zUO&P9~>_sZ{&LlGb&ppv~PF`NKUi zxJ?4;l!NkcGu%6uifxAr5bS#yO&|V~6u*{{OjPO391;XdA0>+u_oCpw&PlRoL#$*| zz7$=JSb^rbBVg9`k;YD}Ce;)e`>sZtR6S)+lN?O_C)L^s$MDcd^#mnu26&7;r?8jf42Zkzp>u!w zB7U$fTE7P2`5jB%)_XI5JUW%n5M3EvKRod3i78gj$sv`(Mk?$rgPb&D#@es3NlTna zSibwU_)BXW2jf^cn=*vpbT2|3O& zV!zGEc+)oe(-_t$JAhc)RcyW|$8QHZ^T~Z=gkd84QkI}6 zBu_1bh)tII<~l-T*YkQvLEZQuqXyM zgHA!H(iAh@wKS-&xU;alPSrg>(rmrsv}wLNy-}Y{>&IqMuR|MY`Q2}1_Oy-8PyHn6 zZM29@ii~9|URRKciX!3^9TB%6oRW0I$i4dvJ-N3PrOKNj{b&YOiF4A*xnIbsX$aPo z_mpg#I7QN9gA;rE`y>mG)Fxp%X!gb|T>1T8d=43Lmy$T%oqdW6p4WIz*CJ-2`H-D3 zkR`kMYiLl}Gzy-#ie^rc!jVbwbRcvZ?U`c2eVqpI7ZVitq%S@Blgp*J9G!?B7vC}G z!5z$T+c1*;yBN#tWFTc(NHLqd_@8&myj#T|hr^CgYhD%Isp<*g0%sFGMM~@))g%wr zJ;SE^Nf_qfgdcZ?)1%Q3S>L?>*wCwil<$#8*QL&pnU$*`oRJe$s>S)nokWDqJb*Wf znJ7_k1%21R&4zuH(`zssD?if88HzanxexL)I_SxqUv$hWPKl9m;Va z$TX(_qaGx=tixi3YM4wtfViS+C?sjpmu0g^-oS^BmYw1TAq#lZ(!cmQR}oo8ab!8; z9ve4vEjrbW;c}=aUf0jVxLY1j&YOeflC9`>p`Y065>GQ5)P%;87d-30IlgeWt#}W* zO0mk{*x_0!cDl2bm8s4n<8|reJ4J!o%N{bT9a3EX%1tbJvIjNcM2dsNeOsM`9%X!B zJ7;9D$x*i@o1Es5O_~}7cIZ;qi}k|&aC<>a3JKdX7V)yp!}w~0@2K=GA7$<_G0zR+H-eNGt-<|SKdAkB5fyAcKwq}n(#D$0?Bu*?{;S53m#zK?jYp4|MuR4K z1l^QX9X^=8YNV~fa|OfDfm=Scdx;tIqvO)PTzge8R|%-#v*V=ry3NUWx^E%+O-Z6sV+nG^9g1aaphR(SAz4|^ z#B}8qsHyx)g93+P$)sd-8@z&{MsI#DKbp;nmnZMG<)r23j^Rtou{EMa^xQsVw7KFh>{W@zHd4iHP?XWC;ES87KqwMh{ z@w=!;6$)$UkCeP+nkupx3L0mr=bvgi7b3%*w{_uSO)iR7R+9VO zOnNVqPrK5jxQ=l+pE&X||2C(VwS}A}Umfv1<{e6><$R!R{s4`Es(ja|ddzwia{MJRy_Zn+MoWG!(pV^a8gq>*E_tAjz8<#^RpcR>MttqqNpw;4 zDY^T-puP!#)LFj|n+7X$_0@40B<|wx^pX}i>O@l1I!O_uWpHeh$c?2G;#xv7_P~!; zzQ~|+7qTfv+(l?F6xqFhz4(jvZrUqyBpctZ#@dy>blgt#5qqYxUk}%i)Mq_XNs^*- zjlC#ounRUgirpkWL$SQE4juEPx#9kPe0Yr%ukCvm(T~-HwDu3&`N(npaqmuXPOyo> zPdd@pTG1uq-yfYuzIgAxf{q-$L-vkSAX}@2ZGQ5!c!VswsS%3*Qbk7FhNEtv3!W>k zN5zLkL^b$Q?_)p6sP{}<%6^8Yz53v8-eOuYFP2`3?>eK^-)QdM0$k7wLH3okDBUMw z5ijo0)@@=Qcz7BP9IwagF*k9s!w$<`gR$vKG9-_JgdCd?!Q^kKQ262!p3XdvH5Hd( z)i~bPq&7kFxy^_mesU@xyWj)+}>`gSi`QM%p8Jv=?GF`=LRI#hip3OdC8CBX?>; zF?Bf3pX!gl(OPg>ISR&F#}Ti01PyQU;i!_%jB1v#(<{JIrpzUmhFRpLl|#Q~OhJk0 zMC)nYMWMQD$T42zNe!h~lI|jOZT3a5VyNg=Je^IO**G zuTBa6U6_LotB0(oY7y&H(qW2qiO>-<_jZGw@MfD2-z&O?*9TZ|ljM__Kj9>6w4TO!L-1brb5gMTNW{aQ92$R0(*`ebBj`r}m6 zUR<5|AFZ|sr}FpvY3k5^T-&}1afXFZ>|V+PQUm#cH!)&|#1_(si=&3u-qhkpVm{_1 z>oPpRHt(~RsQ*4Dxs%u5w(8~tN_yx`TOHPsTP|#&NAFismenDODIACt$0DkG?khS30%`x*l1@!Ai5{7SVz=U&aCizi>vgXQgbW+UcPr1rt&!flqUHj2&~gpl@6Ly| zi_>K+y7dxcJ4MfUsxj$qyTGofucZ73VHDF{MRDeXFjVglZN4~#@|?fn`0q>jP5Ut` z+h6ec;3F7K-7MT*v4{^TFy$L&J;%k7M=1103Z=RnBE^i6qbsyua`*yOUq}Uc+X(LwQpYAF2vObBd@Lj;w6)d4& zPpY3APsKT@d}4Pl?;3xar*0TYDfTf`Xxu^P^rq9E`*qZRzOFdC4MOShzclr39zD3b zn6IY3TzPyKM!H3du9xLFVmKe~zv$A_mam4!q!29n^A!b0tJ$S%N%T{D zxacOR;z#G7;H#&^@f$NFn7nW!O0BbS+24-ce2QmK1at+W>aU0GSXkphDq@xHqm`0v%7SSeXhR* zJ1r@Gb+{7u*N{eMj2G7RK7oXiR~YOd&W;Yxz*V~&Tw}s(zO}!!Fr!wU(w44aO}?+J ztwRNJ)xE)}v{yEyl9D%pwd>+PCzY)20EW7@w? zOWaE5+g3cWmMm&-u-(;s!gkBmt=vqURiBnG$J(|F^q_MVoq0Wi9OmpG6N|-k{l^0G z*kXZ2S|((>YN6!awn{n|u#%MbbhCXu+9-3%LRvri6#IU^H-9DR3wd=Vkugz*zQQrm z$)8J;3NomxqERxgvajT!Qv)6CbR(0ti)_HXZrU?^AC@Wk(cZ*9H)Tlrb}2~?3>b{X5Bk$zNU_XwZcOLJ1j#YGVB3-m(Z4oA>>A9l zp^TE%^zB@&M9*4A(s|mP*uZa0dDdXozIPfbzkML~$&DHnp0Um6R9QpHP)X|}DZWwU zJLc%FqqX-yZ1P$5-?ZH^^TvDi7QdW-Ruy zT{ff@o2Ro{cSp&YU&C$3z3LCiK5r_N+rY+ck7E~qZIo=(-E8~0^EFLhc$m&b4ky2W zWE8Z%qMX5QloULR-Kw7;>ET{uV-?vG!;6gQh`JV2@OmN{x1rQV3AQ%-wT#${eg`Ez zt*6?4l8ck9o+!m;uXt%YIyZv0*XOX;M>euo$%&FpFN$oJ95=IF77)NbH-uFmYxZFqWw^n?9Dk0g+e~8(vz=%OhghtySfr280@796THu(~%rMJKDb&DS2>g8#sr-~Nef>rdj;k}`U^at{?g{EwYQ6ck%^cYbZ6Mfl`;6 zLcQcE4LkRlMva+-{!P(nwKPXTU=W#a?|`qbJXht~{O*_6e1;$;WKaIaXPGX+@`N=| zW(P3sRXcS&4uS3N68KCP9g1Ra$eFogv3p_&H+WyfU++54wQ_y2G%N`QzFYCHGKPxF z`eSFOC7#Tm#j~vk@{II@(C_RaI0p_E0*C7hb>UqoU8>DX`snihq4`kvJ&uiiy)kRS zABtXPiSfw+7#!lv&+K*OIYqy4ruz~i7N5p&`waMuyN0EfDm*R3iC5QcLipTT@_TZD zWiG9tx6M67XQwocnArzYPO4E!{CA1r2`d^J96`<3{Au&C<5(-YbAM!;bH#6e8O%aR zy~KgLwN4Ai>PxdpFV-IYvv1>k;a>hE zE|^cM-o&3xQ4!v_KIi|or1PM?%h)BWYm(XisgkDBp12j!N@uc)NO5lhf@m8Ibb+4@ z4frKjiH3!bP#ir)^so;Q^Y03{9dno3%rj_Y>;~$tiK0_#g_NxQn8s&Y@&R39-bXhA zIw|w$kdqoY*hH}L{ua1hA_enzrSzcU9+h5Cq~ngG=|=8B6ui}kQs;4cEPC_e3v{3s z-3vWct%R63c|rbO3x8A?h~(<|=#|KEYoazbG}V%~w*|fV@P%pU^`_jPdstIK8=d%= zO}`v1=-K#Lg8o!3!G8EBKHTMp#Msk_X{qV3RZo&JrribwUNRgP`&_LC_S~0_+ zNcJ@zEL};JO}l)UqC(s$xk#B(9*?1-&v#j~{{wb&>3s^bIY`$3nUd^tPh3p5MaBp< ztaE!JDgNCdk@X$UCd%Dsg_`M1sWhAkMWbnEf)?FfdXvR|R>mgLGc`0MfU?FNr8&{N zNJfaFeOs?_E2MCn(xu$%<#q+c>+I!5O?cIJ)GU2|XWMFDD z}e<2AG<+4 z9uDJSZOw@3wI7SEE;G4$TXrU|K=SDAaJWbIhI>>YJ?wf$ZGBEstzSB2_ZTGviKW7+ z9-UmN8+=g@aN|1?F89-!dL1w(E59$SwQ>g*jn@(#eotuQm}e6Eyo2oLC}o;bsSasr ze|oyY#o@k?JMTsEtZZS$CZ5*(dqTU@e2?DL!;o=e2f)O=Y<+cIfM%3)QB znnJ1MdR-J-+#JYCew0JSBp8M>CPK6099HZR-Jg$>;NG0fd~f8!iBUn;2SVK0RqdnLk_B*nz?R^_XUN0WQC4VA5+H8nQId*HmQv z0@RVabst@Isj)qG$&v=>4ur9EA(zW@;eEq?z(o2qRoxO9m2uuUcy$@>Nr@hI?O5EX zACH;ej3HNFPpNzRVWYU?Fs!>n{i-sV$s|{rG`^XZl|WFwtS;Oy|HWh4E9q*QA*`3L z!;6?wGAiCqS|WRVa^D%&PezW^R!pU>@jcLVD~*hv$Wi&~dBT-B>Vnj~8ZIp__RV-d zmuxUlm+X)KfU{=7C>s2lY>i*gvWLAf{P=&QEw-KhcjqhZoTw{iprVkp;4*rTXhKwe zFh-VWi9Kdx=#fu9xW#To;i3eL2ri+(@9K250IWRm4NJRjz{XdbN$zaA%gREWX|hT( z?H66JSGDb7X4_Nj!d!r|Qt@ZzaccXt4ThSs7-cHZv_q5lsn?>fEGvr7E<26x0ml(E zHWE8d``|{hGOiv=5YNBnlD8@+BpGL4NGg^W&{_3pDtl~AlU}OxMZ3-UmhBUG=LfN) zX=Q&nSBm_3PYr0PoTSex<7r-<4E3MjO*`Mm(mv%WJYH=uzvTHF83PYO^>7JHKj&dZ zb_s8V_L0UTw1>{m?X6YV%8~5?2!6O z?Mww`qQ|UQWF(u;x3F7^*|u*)F2qn|^5nBJNj9~Dk3R3gjZ7Y+tk4SWDz0?VWHy;9 zPvlqnTkkI^c#L9V5u*xhB0> z)kEwRu48}ZMT!o!f0P&>4?`U*%#PN%A1eJ03W>+ZwxM-jaJqPpqxb z#n^ZY44!T!B=*x3HXoG{Dy80|I^sX5#pOb&*c@5jdiYXff=9Z8*y|BNY(bwUHfYR5 zdYs>b_L+Ax>GKKP&peH1DChHj`!guptUrRMi5cjF=h*Hq8)*Ccv*KB3Gd{U#;I(=! zwft>hWB$IPA%_#e%tPQbEEqM@W@D1j#a`_cyJ)rZsK;#aK5r6%gFyoR1=rCPcUc_N zDWgwA4oPI|-6i{)T5Tu4o5OI`c1XN!J1L^jo2(o4=xNS38o#!hety49mlRT| zK|D*xR%v0$&o6v$>kV#blg)pL*|LuVOE9*;8+!MvBm+%b*~5cnqLXkrhNV8u1_J^1#n?R>nKE!plJNHSqhS^U{Cw4}R}y|e3*)QK$6&;cp< z{~32dpH*Dz(F9&SWhk%cHwo=i)v;yjbGrKWC384vOX0(mN%=?sEfw9If2SUXc8USF z44Tj5H*VyURTX(uMLVqjYedAL!DwrC}wD6n7Bo z%MY{qkjs=aWGShCHm2ZfiqJUHNT=4Gq|nlO7$he_>bw(9XkMo+V`d_Cm)Jd{vR~5w z$2jKmIgWDs=AEBmit1LrVUcT z%>m^UFy$f{ILgBJlGyoWsELw&1kFE!i zm)#dW9dc~>!oO_Y)JB%F^#J+#YaqEG0L_Kwc=0s=AC9TuP|YYVyX_Pov|COvm8n2) z_2SasJzjcGdT$V|zl$Hs~bH z{1m`q4vu0WUiV0+;yo_8p2E*xuIMek1GS!}qu_z)Q1Ljz4|#3l^$&=D*MCP@{5iSI zH$mivjZ|X)iejc)VuVkBY8IZc0ZzW`amZ)pAzw(}T6B=l_i)}3$vuCCb3eCd>~wS* zO{>v?g;OdWp7xKLW{F+g;u-bH%owuE>0tXpJ|nBlgx`Lr$M5bCIiBqPcqh0aKqiQ3 zScvH9DAzT^r1dXp zZmJ)58za>d`p&li_PuS#3%BMAN%X`a?*$ZqY=pJW+u&vq+5gw*)$HM_AfT>)Cs zCumwvMhkWRvQ@7&@zl^1ePl~0QNAY?$&0*2w-)tZ&@EYD(2wQ@b<>!TJ=p7TkM=f* zXDibH`f*JjxdUBEWzk_~ZTXZ{6b&bz#dB!fdRy`5SK>lTCqmaKVyU|yvj0WnjpH16 zeX&H%)mwOU;~u{qp2d%OFW{AJM)YM{F|$nQNp*j=uny7H-|c8d8~>kgeb$R4y+MnN zgZV(6dHlQRSNNNF9sxWO?e??a>aRg3-X_uE*^yL{sEiv9qB~`56umUJ<9V4Mk>-_x z@#6iV_>B(iRJEYBauzdLc}p@Yxmg-M7VY`9IN%eb+o_&6h z#&acP<@dn%EE_>SqPL)I{E)!oh-QCRNbP9gafk3OnKS zJ6GZAcP-8=-k_sS?06QvLaQV-taQr(HuB$mG)ru;(##tBMBizTGrh&`R&$XJNFn>u zUF60?DWj$dX01`!lDY^V+WJyx^6K!p`MN~_&C-%dpghKRC&OnRU+1FEYnKH&Zkwc;{nXmMw z|N14~s0_f|B4GY~yv6n8g+YN=x@A_Cwn-GS}j;Y8T&IB!Q8{v=j5W#f!PnvXx)8wte z^wT2>tL5bRle2cb|1>|gJuH|VpV`jNB+fB6MraujoZ#WiPYFi zn8_q#=(1Dr`yl$J6Pz%jeFUa2c|sb;JjgJ36_5Qokxv#Qx>fymV!xax6vwtx*BEs) z=1ZZ+x;6?JaF4xu{aMml8ziaqmlwtcn+j^1rU=e|TS;zgH9a3Q7z;EXWB%?8@PYwY z)Y(O6-b@lZq`L6YzXU_p#K9%T9&y^SG-kAc*zaEN$-~zcHtUamAf@#U6r~+ODTDlY`=MQY>#AA2_?kF( z`dp3~wx`kl`=!J)K1nj8XqO~^b0Mo+&)B9B2PLKAd|{o)3O#=wg@qlW>++-~w|KLZ zD+=df5H$odZ|tY!nSNY|u;MGPHespaVZ47GjN2U>FjdDA0kv+Mt_a>7!BWRoFM%>Z~#KEy=v9)6c`;+yK zmROB~oK*q-PAtNoF-iEQRYcpx`>CON3;oM{&aS-NP20?VQtRJd_|&_vm<3bBj48=D zyDteAyVaoiv z=%zMXu?6$jMPZnN2R|NC&D)DV@ORF}NUqPOl@CVJFXLWhDSe)9t{eotmK9jIX$+Jv zDj+HR5j*JB#^xxzVG|p}P$%yHpL(W1>fvd6c0|nNT-2q(lk*X9k&_6Q^nem?xndXhd zd?I!%75&#c3w^S=!>jXr!M$SU+QWlhXUy19__*VN078Y4=PRe)A{-(Ymu>34pildQ}eaSjk0L|`>9A7dyTRdUnK4O!||w!ky(;B zTX!@NcOZ)(H7Xxdd)>l2t1@Io=3@P$cubGbp@3VN@@yM> z2N|%E61tVAM}}uCQKqvEKc4J`n*V%y6%j*`Bl4+i?Px%63f5W}V*i2tOjG-`xXX26 zZ>EJ|Q}YshRQNz{r+(AkxnifHdnl%^tD_yYyQx|I3Hz%N%aXodXMfY*vJr|&WH9P7 z#m;&{N57gv->@$x`z2D}1snfI(Rs%8_`Pww&`zZ^v_z$3w7&Pbu5%X+qX;D>qwpgX z8Ynxor;?-{MTMfJh=fW@!zg8>C?m6Eum0!%$O8|&US4&dbFS<2dB5T9tu^4Ivk2;( zrTG}MuM`cg)6TX&64`d0i0Y;i?uWf;_Jq{}C7l`Mdm3U`lQ2FUDu6)?5fCxfC5L7n z;@tRRx+c7c1{Iaij&wzGZqzJ6y5bbUzHO^;{eMyDSe1nlx(?8t;tiJGabQ+kPMYRe za>w;sO=c+Gf&&3|kfv8pavU?ris=H|V8^%AG3qDf&A!=PuxE)ca~_Ywls7Su`f`w*yjKb;d?j4AjsfeC*Mdo-CAksrZb8t15dU-W82;zg zk95F&G+G=c_-Jr0u9vVQ%V!n{vYXzRUZ4SZBJK=o%4cC`$5-$QT>zJ5{JEZ{G-x{$ z1Rh%3;m$8HTF-ovZ}Z+`#pqewj<>SLE=$H5i~3hVf7T$J|D=GE%iB#zijB$R*T;>I zjrD^P&sXGcNCy`%c|1++dxWyP!|=bcA3#}G5kLJLiwj>wa}iXT!-`SdzHBq@^<^nR zP=~!>`-*ZXOJM(JBVSNV_8k75Qi|^;us7PoI>z#@BI@PY#HHsXtlr&3I4vPkT0TI! ztwLaXDog1Zi4l6<;RIGQ zze(d9wkLOQCfDn=iI=q#*ROpPo_d#q{fgfZwkHs_@9Be9Pce2!SWGuK*wY^LprUi+ z@#LT=y2_hi_4MEHKJP2s7ga>PrUoMHYz&>VD`7R>0i&cZ@bZup3XO@l8 z-bCV@EFb`iK{0gzL=6E~Dad1@y9mDD{uqQN4Uj3D0zq(rb-YqzalGCI9QC&w^@40+ zO#U^9w;TY+o>!0;AceB-lki2N1#wvYh#Rcg$lZS7j@;XL#*nGRX{lLcvg%nl8?C}T zVQHMk(o3A$`(9gk%PldnfT^wHQoAV6kkB}dDAD$u|Itw)+cPi`YWNZyfh2G zNoB#NZCRw@8-UE>A3(-3{`@o^Uv7xUgKes?j_ZcANz+hs_fII!xd)yagRp|RCZqGu zLjFB(c<|Z?mO0D9WEWEa2MMekD~n0J=}^*RLlWAJxeG}z$hzuS*fvHQg_a(~B8eN| zt(gux$Q=mp{s}<|V=+cg5`TtBLx|EgP?I_gbhudky$c zo@9Y#1f;br!o|6M)MC?NTJ&ZXP1}D3wbf7J`*mk==eA-P&K!qwE@3#F>Hx<-mcj(~ z`{Y|t2jbS>;LZ~fJdoRrn{Elw#SXtwVL}Pnq&b&F2$_@LB?SFWk3zRvF`W0}5N9@e z7MKsVvK#{t-<`T3GcXOym24l$JW3c%qglde<=Ee39_0O}>#UvR2h3Vqz zJ4R?Z;EoX%@w|UPGSBI!@+;-e!BdOFX#cwpE2^6T^mkyb2-}fMq{5%qfuu);I>H% zw9FRZhA0cDo;?#JH!pzohrGbRP8>A0w-F=Ht7xm3j>1DbaoM2<_%H4+Dq9&)MQ1OR z72S$0aqKRjb(?N7>Z5+@PgvpMEDi*hVat;nXxr`}unH>_+~~;Wx|W6GyM`uwGEIh7 zi)?_{aQ5AKavE;#97j7>&8DlC_|u7Y)x^756}6N+ z4YcTCNlm)w;aYl6aVg!kcotnK5s%x=WALdC;{iOdf{u|1Q2+b{dD+~C1I!hqR@s8G zS0jkWb597}7YB1pZSmumcI>i3dR0b+F@ct#3u8mNoBF`>Rm@d#uob|v9y7weBS&V@ ziT~LC_j)j6(igxy2OTUj>jkNU6)>j}X=jH%z11s3*Xk%hq1sGH?HdOP>|Xf8Arb~Z zI>541r*MMlMCx$HoXYIm%T4-W$jwN#6TIgLO^vgr?I}idqVsfq)=2@cVZVYuH*X3i zmI%-&Vj>5qF9#G<`(=M(!%Hrf;!9x zws%Yuq%64zj}L~xSdU(q{_ZvW+BS_lBDIK&%yNLcAEfdAF*EGlbr7$Qv4{A(hd`-3 z2WGsUK(^9zq`WE?y!$T`#Y!b0(yL(eM|p^hbq6i6WZ+s}QYkB4{;a4gpCzY=Pxor# z&rvqGYv~GdX#Yj78B_&}858gYw;Xq^jlsG;dn~(p0E4TJ<<1;WzkumRl4x{u&M<5H5p!R$E=$nukAuZ46JXL0d$_jE3fwzZ zfz-!hQk&FFRtn3LjitBnMBNB}Gnqns06iNf^_^gwqGJt3aV zJ)QHH9DY{-Ywz-?r4oUc!i&&$_coB3>ZSK6#K*_s85(8uKY3=V?x(p z#kqAPt6m&j7RiF^A5UCi7=>$}m14%|QT#(ABYv*<7Cvl|HSRHqz^v}m*z|QNEdT5T z1s`)52T_S}`>uhx2!}S|35-2dir*%+ppEEUcoTF6CMJD{%ezK_@5_hKwsITx0rL}N zGM0&P0Ja@o4#P9*;9KQKxOL7Cm!{^R>hF4#xzD71mw84)u~cI3qX$ z17jYM6Jvbgl8+tC-qTJGeG%gO=a1!Ib}Pd2wQ{iS;~;U9^2cGL!>C)wdfuH&;jQ2} z^H}|Zu7W|_esVPZdvg?hIrIlhwPfi6lSlZN`5oI<89}6|I*{%qXr|$YQ)kDq8EXXU zr`clCyCTSo{SKO{qw#>~Q^<=6gXC44ATs4G89hG(SF!H zzGJxOdkpennRu+^8((v|kC#vB;#bR>QJq^z4_u_w)8h;hde*@q8V!u52v7>-Qmk;oFV;OqqJJn&XJvycQBVya`uUi(-gO6YS9t#h2@% zV72cj5?P`P1KPrH+bn^6zrenW(aX`wc`7>mRYa@WP_Ul$j~wrn$0##doN@jq{MGhC z?eU5DbzK5(n4wIDo(*wJqjR`#ha_O$e{wK%!2)8VJQ#D&1YMd>!u35Gxc6Bh-0TxT zMW7fO+!_Ips)O%_AF1N?A$sPo2yZTT1&wdzp_Ni38XCAmNKrOq92CKeCbn3wew6u_ zi!gmSAKkXZVtbA=`d(M!^|$|FE}KX6;kpRsz=*^a(tJBq91)0y zqUR!fJ@*2xHeP`C6HRoN$u(NiTTM?eF=zNQ0g=o!;)r<)9RK4Bo2n1P6b)MmTBV6oD&=vDln1%}=nPry+(wqJeF8h( z;$Y|?4}Qhpn%Vjn`&VSoCmdKned7oqpxEASEV^8CjaqxqGU zvBZbv7M@lvfF9Fac)8^^I97IpP0DA`WxTHo8hN;-vxzw*Z-CnXV_;5wg(s6{(y~U5 zda?P!(oyZ8Bp(NpxhXg_?2Kp3J@M^|v(R~aAG)}<;}?@as9X4!cva7ZjbFpzq(cYe z^*n~k%HL$q?;sF4_yuBTH{-k3I=pn^I(D;sIFmxb=S9O18WGIPbtmy5ZYTJb|FYpu z{z?!%>H%l;qXi9Gr^)Qm2jRqv0_MPp#=}$eFhpcGwl7$XLtYEe>##n5eG26(9Ov;* z^kT6%;~aE%OF{U(n|OELA2j78>E?;}UJ0-6Ck+SqkSjsNgG?$#^_L7w;?>i_)n_>E%mL>0=W~{?m9V ze0F{tZvL_invUacHp97j%!Ab43Emb#RIn_B%4Gyo?RDNbH)B2P z3TWXOacPu{RKQslVwhb!iK=wPV&vG_kZm8&8_ERocRDxm7S_f1y0;Ct%zunEz9RVf z-BqYpy8>0qn{lzTI6c3EWgK?i#=zR+sF}mM`#Ki%X|y3dtUZ9Du{rq6;ytd}40Kk2 zDL9wpk$xv<@ZDJtYO-pmRBer`8=sT3vynvk`F_$l<|~;k>;f5UIhcLMhFU9|(c{hc5+ zdspH6p~bXd?ixDX(~6FNcZBZQyO~bp9q3H$oh+LNI7c`EnkRc=r_&yMleh_21)c)Q z{X%%km*AxvnnW?Fj_4>GfLPUe{4wh^7RD6f_o_+ICKp4t?XKY-x-GzjvW0lC&k3ai z5UjN1FtdlbE+qTO6DbpDa_|5JyIDBZ*BrMP3L%a425rf;@PPe1cW)M>n_Wgz=Oq$! z)o~O2MrUDXxDmb$KEfA`?B}m5uiz!m%HX3#Q}741!DTZ$pl&r|_9;Jxe;Qrfs<$t= z#8>)Up5ZNgb~PC7WH}r+%bg}iU!+T{s%d4kB*rZ$APo-;1uNe~!u{osK!2?&)kxem`olRGM2(5Rn%L^Joe6IxM|1+f3z5ZrQtoi;`RUyqweDj z<5lb&aAY%>7a$(lN2-5Eb5>lUsYGr9iTyhoB*s1>f2M82kQcuAd7&KZqZnXLK3wfExua@}3{@?rum+*gc`KiP7lqn>aBV%Lf7?^bei z!vta(@kCI>zS~E|vrsc-J<7%A4O*Yf8nq7D5{WQhx&@8 zxT>9xMS|L5$G^8!Av`MYlni2dyM*|LbEzee>y*1`%G_pZ-v{IOF(yaFjQveqnF<=4$NY9iNe_FM31 zT`3e-C&9y55r|Q*#I|xFI&a-uw9Ht~1-i{8?PpMs%Y!3hBI;h=L;eJcqU^U$I`}J*M1GxaiVmhV=iK6|04SZVJOe!>0=zY;k zxJ20;-R|nK98eiJ&Ub_!?M~eG_&J6vu4KE6U`*X#2~&I?!8uiSUe(-%s?YT>;i?BN zq*CaVcoP1T?Z7b!Cvh2LuLy@eA=9?xK-dLG49%L0Eqk(YtK~h^REyz%869Ttl+C*brD$IXsLVL0g(|~}>I6TJ; zbNfxPPc2=L{M(Cc`t1Un@3liratIOh8<6XD%lVBh&b-~3b-YpGf7EI7OqOjP#y>%l z5YPA`A-n=!PW+39*7c*}yApimR)Ab+BIx+2fosY$3^!VfCl5@<*`M~)*zOcsa`r4; zpTCaW`#F_tn=VW8$1Z2ysqe&(Hv`8XGJ<4VX+f5T1eb7h6ILkA$Ca%)jM+u#*ESuh z^H!AhgrC6HUH?C~#^Hq1=ES7w3O8^}n}pO|gR3F7aQ$-*>6kPLP2SE%CGjAJp-pGp zlPkDkTO*E{bDZu~;Ar)Wr#So0Y23q!(2+ai=)3T97`A0GcFf&@d9O-9q--vZ4?Ki1 z;=;J6ioLl`T)_2uF`(pljQCF6K_;xUqf;f#sZf+8{h*l);qg|WI;2ea>M8Wm3<>%- zEd)J zI$zw5o~?NT4>!xBS-^B`EU4#K&THkJvT7LT(iU6UJ9nf0b`1TUjR~v=m)F#W{%O*% z?AHj}u^uP&?sByF#8dn-MviV)*5qH`;rTSr#eB>PX_S6Z4$XoT7`T^(%ctMRcQp?% zWh5WX)6b$?;YEz_G{fDuzr*(=7f5)ihz5on?p^;E7CqmNf7jb%yl1_n!xKvR1@+;3(*@-vxaCKpe%zhuB|*;5dm;RH?LZTMk! zBYs(1jhboc_|-TP&HA?^`Yqr?8{PT8iL3Y-YU0%YP)8y|z%dSQHE z(*SqkAAog7KD2E-4PQGCgUCu($h#*&#}rA?|BmU>6$x7ObF?^(oWvX!>>l~RJqOHH zhPk}K9k?xOKBhkS4j$@<=@Ordbl0H@THdgLIT)gG*;Zd16%z5QrW#46f4q{4fw z_(H$TYogA|UbscZ2;DaMKw~(|af&U$sKdK(qi+N0^~nSLB8~ob(O@$6Ja{wS&yO9l z)N-RDUDm2V9oH=bS7J$8Bi3>|)$6c!n<{-RgEZuTH2jU;3UPjGP^~SHbsA1Xergsh zD>p#z#&u8^`&)bzhQA@Jm%9BL|_K(m+Iab%n-%3Lmlnf?ACFmxqXNds4A z9Dxo~dttlYMIzl>46l1~$q(zXWZ8{$ytMx)S~~jUQYR@~qryC=o?l2W%b`ba92Dpr zlrW9A4<)bsM#z!BpGcDUO!V|o8#cn?QK=9pqsC`C znew-_^?Ba+J!;%MiDa)UIy^m&@s+1BE+7(5yckXTPjW=fUy}8^k7IIk629zMfr5D~ z`=HfJ(^lW3w{+NyYP39VzxxVg3b%oft_qk%1d)iK@$_fPOKk5y$Ih*EYPIP&74nFr z!V?D3MY;e_ddxzX@<-%V_AucVa*){f3O+&>(KHGbTzV2oM-C)X)$Z#wxyP1jMa-u- zXh_>0l%jmqVGI)$!Ci%Z$X!Xp+QdX$yX+I*|CfTddno9Ekrn9f6P`P)0(zm~bgM#dd%j-Me7(}GQ9^i~MgMZbg2L@Ab8n~Z|E zg;eE+K9w`Nj|;j^pciK1sgEh>;%U!K8hu4jtbI=KeswcET@g(3B!8K!aXCTsPs@1f{KXc7US&c}-U1?(y_zdNSc==- zLr}#j6dL0j@PPYqd{StKVTKUIq9EQ^&?aR(iS*?zsZ5r$5fkgb;~x6|5;3%nMC z=Z}AdPyQ?64qZ;i54^()&vXpj^^3U8VwqNvyTs}DRcJd`0RK!);YqO&RodTyllNan z_1T$tb|4%!GLQU~VC+ku{7lrg8<> zI&Q#XsDs+%)9}3W8dN0;!HlmT1Q#Ag2xiS+&L`_kV*dDH+SN5fFz%!9as+%5zh5{cEOCFb6WMK4XycuIJkjznFIvJ zXuc+0B6Z};RDI+(ti!1++q$5vgDl_hjnjbf!uhJP5<-Oqi6j-V6Fdo^k_OwENxeFH@96ATz0C( z(_Zt@Z0iWLY)OVT1wS&|N``DVsUq9ipl#~oN+jm1; z#xt^erZEY;^Ag;a2jMOycO1MDgz_^DaiwuTJeXt*i7uC5RM<2WDcXT!7=t}vNCCYc zu0} z5Jw+&W*^klz}-H9aOdx5^w=j!#lOkZOBG2d$Y88uzXS*>X3nqBD(qNSiJgU0XmsNc z_HQi3%Y8S=jdoiuu`$;4Lf$Jly@R>04o^oNHXH5LUX1HChhSs)6cBr(4r0@CVe;CI zY<78r^#SI?Kf_>}bF7YTm-TKX-%$2Uf0iEfjFRAU z0*6>N=MmL?QG_9)Z!l)x2nq=p=c|#iBG^4OQz#8Z3zBeOcM7IEIHS;IJDf3RJ>Hn6 zMO$l_J9P1849=K@nNwLlu5%XsbNeZ2ow*!R4hBK!ZW)kOzX(T1jc47-lib76jA491 z0VkThg3gAq*c7OVmV#V#Sb7Cd4J6{5TdXsc8inN>kD^39a|Fnn!^KC^XcN8TK@Vpt9BHP>ifu;74M8E z+iv3a_qdTI7d@c!*&&j6x}J4^-=fu}1~?YIotqwOgBtdGpzPxj2>q7<{aaK(MPn9- zj#JWH&kx^~@!($4Z{W z*9VYK@2pXZ?d?QU|8i+_bkMw14w|Q$ky}@a;i`8GY5BFC>#i8hr%f!RE0ddPR+jaJ%~qnf~7@)rMQn>uge#)Ghl}+E{M2FroU$W?nlZP?_j}|=|>1;<~s_K z+$y-a!@QtjNS)2WAh#8 z*XPV*XEX&5ok}IOQBMWG@0`M_G0OqZ{e|_l?r0dSLPJj^(SJpQ@H626%nl1D7X}w% z)4Djk()0?oVwS+G&C)otJOv-Vw}*c!6tWdt;kxY+{Px8k{r0UuvupX#-X{eM6E~5y z9))=QTMzbp`iVZ%t#HS_Yj9Fb2II!^IOob3wA3@e#qIT=wKoH%9XtxZizRSp{|ykG zybJbsJVK@0-*8W&IQ8m_#8l0TSh&2NWt`6QQ}>_eg-3GuV8ar`rW<(N`!lYdO^NfF z-CRZBW$txI4)T8kab*efp`1RBqq6<5DOV1kw{raah8g@DH6uPlt&o_jZR9leJ>=T^ z;;}ou7;iW>;!Q`!{~Q$uwv}O^I!6?xcKjsvYlOLSxQ*_D_xSbJXEaiY0L|Pc()e&0 z8O&UOe>bxIj#oMw=w#zc*E#s#c6rQevEux^RRo6CmVzfUj=_{KWAOf1HSA&EENurf zyk~nB8ve<_)|++Qy&Fhub^D?E!Z4JM`wbN%EWfX(#2e-q@c&XG(CEK|=)Kq-$J>|V z%@JX`bV!40IQ%4;mxGB*ej2gr7tk%1lGH%A5S1Dw`4*q=G{d!?@j2S@z2OkE`L|gjoQTPQ^ihEI&Hi%jZBcF>&rf3Y(+0@ zz6N;S{UBDYJA=p7=imnmwhOv^9&a8F#IMsvqy5+ddTxC^Jry}X|NEUpgyx69(yw*w zj&K^r>gPi2HP#zAtIEBJr@8X9_JsL6AbTp=h~)?v%Hntyx+$AR&y{-z|lf zhm%n{cM?={ZX{JAmHZw$hsM8N!KHL@$npFEJFDJ9E$jO?GRMN%04drXIguXgZl((5 zQ8ew5I(2;4Oq*h^P-D*ox^k{Ce6#ap-H1jqvtAA^ELBIxq3w(@^$JqC5&Y^0IlhYr;!jr=2 zt8NRp{10?I)5cTkhjIGh$+&!w-R*B&VZNDgsLgo{GjAzkHro~SezwH?1McL3TN2om zOve>-C-auh2E1sT5pPhw23m@S;lc?kthEr~zWPlipD)>gk+e779C!pDJ+z?W?N}&i z3Ws`m#x_%6>@ua(^u9qR)or>8+xM!VoCDj97Nmmfi7Rlc;web|-iLD?RB+dG85GLS zhnzJ6l71+gBu18?k30LWMNGk~->boLOfyOE)xsZZJXllR13$RCVDP*GSY5OQZZ8Uj zad+gnrL%@f_X#QZKK~e)N_aC*-Et`KV2pER#$Zdh3)@vL!P1wf(N?t>-P1)e#f-72 zckjT!wmf=&@G&)f*g-Sju&#GwB&KNCVe92z;GA6oYa%zmF{`_Htb=uh)K+2SQe7Bu-j6)^)V1T4Q2!#%vI0Dpd3Lb6vQ zs*C(aO=WAk#Yc!9D$7TQOH4L)&Mf|(73zUyO!G_WtI_G05oh^Qs zZt@Jp7ghOKySg5)FVzAI#%8-2^bX2`TTvpf7>~(%V8*gR&`T4?pT-Gjy#GA`$jU@qx@eknmz#26D zJ_e<#w6LXn02B+R;KW($a6x=H?%q6uY;k`fI65iU2<>%Sy+ewq*@;sCQ-Q^5W|B|N)99sbnj zkt5t}^1g+~aTXeAQZ9nCx~))s*=AH#+K$b_uKcbKTX?0yUHtq9yI`$FD@X|#r}`c9 zbV(|MdRG(C+j^Y1wX~CUe%3Ir?H{DtXs~Yo3KZM&gbqDxp(`gA(_8PN;QNltFj8<6 zrhL7Dj-MiN%~dsQV4RKgk|WT}*cm&s#Bron7KbKxLSJzNJm|=PCvFY!G+7w67oWz% z3(n$XV=wsEX%7i%7htYa9P1inqMm*V#@jt52d8Pn9r(lkd_&Zi)esSdZ)A3W8k}d@ z{Vc~XaN8gTcjgzODcyvL(w=y!Jr9>yRABb}O!|3o6kW4CjE0vtlHPX#P|Eeep(VlF4Y#@(Wg#O7xxMc-k#5N{`f;LC z4g#q!>#mLkZzCB1owJ7{J z#hu<)@TO8aUexKvGPtmB9=`HU#^P(Xpb_r~@t5kMzAAVSYfmKRzxVcpUOTUiAz00L>WU?svTV@4KNmH@7ZivYEj{_$cai|Q|!hTB^ zw6zMyyPF(f{Qh#7J%jCUH<)p!{G?#z)&J9`YS!sKy} zsK)KdyhFx1KBhPvmz0O$vva4gdcOpxiMzQ`JMuUQ_d#gj7veSH7;LUlz}AX9taz+I z^CdK?tF;I1j?Scp8pqK$_6EA0d5i0Pe4+Sm4HVXo#giXzaVICug80NV7+t?gpz?WnIy^#Y!Rp0yoIaZYO)xzUBMOa=*nAU0DB8JNr!VPvNPUSv>#$QA9t4_v!?Nd-^&T)+RT!Y1XrO@?gIQyOK z!1KoD)c(aOy4RtPehKo1)I9~@vSI{8lb8oZTLL;pr-A>zrKlVjg>l?T++X3%rL7tos-Uun`@9p3fE8h-zb zFX*sj0$u5`kUqUS1-qZ6GfqN3{t66*vHv}T@8`vE%*jYBs_)0^54GvZ=4kwNxEL2H zT}H)ciCiPqhw!O$a8l<5{O)@bt1s1{nz#(EHJXlJ{>bBoCH|N{Hy3a8bu%WZ6xXb< zoxB~R3;#Xu2ZdXEFllrvc8V|J>~A!1*-bUX=WaSyrXR<921&R%Z3U)F2cc8rF1-BV zF3}LbORf)22MwulPx;b|+(cgfvmO7+b|=5Ml0%KlV)!^^ z3Ot;VikEJ(oRQLTtcf4ORo8{-*5xlz>(W>Iz;r0y=a-~6@ex2HZ-jCu?#-Sb_ ze-(i|mS^qheUA1gGI3Y$1jazIrG}>36dpC9&X^SPXY4HQa%hM-9MJihdw?garSOx@vlfCDwYVZ4En)#f)w8TT|g7pT%}%qIaKFvD`ShP zqNjvAsx8;ShrMdJ(EcF==@{d<1(%^Z{U>=I&$^GP;xw)19PU$>qT78kF?K`=cb=^y z9#>g*JKq?p8s5VE?CZqA+mCEZ^x{t~3FM>XLV2g;h4jj`2{dZI7+q?vi^GifEYTYU z4WHQRx$ZyQ>==SmTQA}P&qY{1r4dx~67W=-K3b>OgG=yvx`n+thP+$Z5WNEo!~T$x zou`PI>sw+TGae@$JCAGsrV;P+-voUz?h2(`C82j9xc)Gk?(+Ohn_saaZd z>+Kq>lcDHXS4Uz}(@EuqBP3 zhL0v(IXMq>!&Ld#K`Z#$jmP+D!9rB?GRs%^wc+;Y1d}I9p{AOL8cZ_%!pYg-C2&T5 z1Wdy}z?6;4F;6iG??(yf$r(rK{3rKmT>1m7nj=Z~rODCxO1AL)T0fYZF2Del`&|E* zr{t#Ceelh@0(oE5R;G(j-gZV_R>Je?Ai{i z(zb$1L_E-=1Hdm0#@>Yr)Xr0zDycfti85V+q8D$;k)WFpec&`WvVP>vo=ogzz!AtM@-}>AVbG%nLHS zM}}tqT}vcZk3g^Pb}6oC%(ZdZAs9u{|pijt*~No5Y9;~ zMbDk3cq^lly93j?)Ym6CrVwwSOi;dgvN>n0$)gSs%^My&B3_ zNn29gSySoFsLAy5=H(b-5Dmk%nPjba6K-KV))`tiaV=ScMs05BbSIlo4nhwJ5Q!&>47-StafSfEP8o5=Lc6drsVFPLCJCZ|82Bwpshv$gZ zTt8B-{GPK*y9$Bz#h|w<9455iMeR?)==Gljo?LMhs+C`mb7zF%rLhg2G4u-`Cq8Ff z|1EfKj5Q7tZMQC#lw^1 zA3;u51#hRCfzW+LD4APEa)R#SE4yRpSgMNI8K?P*Q89d?@L~Ri`UA>!Hc?5l2h=Ew zb#}wILg4c`5A9_!0@QI#Y!`aD63 z-tsw3%;Yx_?(HcqpK;yf4o$@m>bbCy@vM!PkHf{&OQ9qxjeHtDL5yu*lJ-qnIBX}0 z(LdDib!9XLS3d_k85Xx?-_)+kNU+~@6q3E!+s;u8G58bYr=-*3Ju39{mCLwwy*>(M zw!!|;0T@0Lg-br1#b|>9l!}_kzjo8-OWteqDRYj4snZIAx>;-wEkQ2^v6*te$ zBs2!Ks-mGRRGZD2J+bK%>rO3<2bD$Fh+n`GGIQm494a};_Ne#av-NMHWQ) zV(?ECiCu@^y2s-B(rB(deiZrq@g>=Jo^kWSw!-VD7I4qL48{zUk*LEI)k)hj`@Nr zUt@4mJiGIJ8Alb;*Dm&K;v~dCBUNcq@ z>g7$vbGc{WhSdSOFf^F1A6HB_)&`=iIfYV{apg7N8ykg%rbnr1 z!+-p=^iaNGQz~znewfOw;;5$lOT6 zxP~s7{*2Xd1*_`isFm_DszLTqzr z9P{;FW8u47IA`QF6qk=7Divz<8H!Q!r$d;xTb;_xt-=Gv!6*nm2^UK)GM7~b2Cg|z z{>(T?LiV-8#@>~*Ltd2*e;mLXy+Tx`&r#>kFgo6FMbipp5NNf-k#`aF^@ld9=G90e zG}Q5a!F?zgl!CdZ(wHmT63?I2!0*#m;Mu?}=<{hBK1r@3w-a=sdS3~wn|uMEpSysC zDs8yWU=@nXSfarKRXkI<7cW>XLs^!2U+5M}jGq^>^Trx0e?Ny80V#0w!curtI~`}b z7@=>r8Tu{>!_xOhup(hI>v~K@Z)F#pl>Z;je)t$~%D=`H2Rm^`Kmy-Tl*W&W598PU z|J&(x35$a3QMXSThRQ-9z}^-V?<~WMQ_aw<^D%^<&VvA*ZLm{EpK(y{(k1W3_!58M zbsrqV)}dUu8Ze3bx?PENvio4A&Q`qoc^i%{--LgrFc!^y6EsYA!mmrzaq>TH6dJFH zG++pV4X&cK#XiiA*bhys6EP_^429lrz`8~0c*{ElwO+|$+hSL8`Tka}?n#$mwZ4FP*c9IqYNJLHr$Q{JeWH>ZXK3Tb3sYfp%_7NfMfWzlz^kx5DoDRw}VAhUQK@ zMk7kqLGk%T$aydq?>&10VF3;xF}apVf;VoISd43I#-pp;dRo`DgSO1uPJc1?U-^Y_ zR2jb)#n}1$`1DS&pIQmUX2w*~UxXgs6pa&aZQ^rX_waSwJy|LvgT|>9)77g>=|1;< z(CxYeI_bS|Xt*0T?s^aA>e4uCP>$YH{fYCYJ0bB8N4oa{u2RXu{c`p+*zE*$*-=PK z^yWj%KUrdIK9ii?BMCp=u*VJlqr z%bMLc-=JA#BbvR5#J{nNF;(y$rpISNm!ULv%`n7d=D=y0?TRuHJ$R`}gqqpQfwy)Z zh6Zp>@Fx__hhO%>JQp&5U0;hXp(1-D~Zd?#qjxM9~yZ*Mb9N? zn7eH??x?TANq!s9KZ!RwJn97^3n$C{|tMJ7JW%z4#IbQAWMYn?@RB@~{braX7 zA;V?tJ(P@Z8jfK6#q*rfH40~r!akNOP3kmwo<{s zKLfD#K?$C89>$4XD)h#L58NzkUm|hn4|$ub&b2STPAyI* z`4Kdjy~#VfcVd9uax{EN@LH(~{`;uMTTN2ndBwl9-PjXun}@?d!y0I?r~=u1NBkd0 zXZ}#*_qE{$Y1XJwgGzI0diGlTAXJDFl2D;SDHT#7qB)gj4Jst1GDL_{p@~$+B1Fhg zkxU`<;oaZ&hyH=nbDp#JTK9cj^?M8P>%SL_E%1!I99{+L8A&u|yFcBiHJh4$eT_+# zS21ULCA)JMlC@W)K~kZGXg=ADy{nd^_M;tmJAucuPIcpXur2sf5exeKxh~aBm`(S{ zeIbp@E5RX*WjEa=aBZm~W9mAg-^g~HER7XK6DwN;-+ zufBTxoMr-(Zq9|!CrvPsmqb3Od?WKh5XmBf7t<^H!@$M-fa9d8i9hdT`=y@CmOwykC!v}kKzl)#kv`bvyRB)m>nPUGd~5< zu6Hmn`3t-V`~*wAO|Wg$8KuWPP@iQs3q1b=_u6T!w-kmE;jy@UTMGt_)#B7UFJM4e z9sLdLP$fPMCnq*zQD6s_I;F#V>5XtbW&%<8;zy1%E{yoyDP+9b4E!$qAthaWaI&)= zH;tad=gdpD&6B-z|9Ih-g>%qmo6<25I4G9 z!)D7Z` z|CV#fElXfcO(0zQ5&%AylcCR_pKQLBLPVk@A+DH%DPLc~ZSy}kXWDE0RQejH&3Hi9 zT3n)?NnteqL?Wbg$-vi`S)`-AgJ8-_tCN$@TM6MzSaE3|5jI%KxqG{z#=SOZV&~=6 z2QTBHBS~1ox=*Jhve7^73K|T?vYDSXXpYr!ST&D3`ELTw9@zozAq=k)o%7VXEZ=@el!N{OvJP8zrkJR z3T(b5ivE6|fPXd1HnQEoxLzU-Da*mjdpabg$%$-yl8T*PKDdaTEw;L+kvZO*$fDFZ z^54B$a&*mHcr{W6Uz{$G(Eg1u<~Ra7wj{$Gha|kPX&YAinPZEcAinR*hn;@=(07w6X!VJ}tqcd;lLO-z+6`(Ix0m^}PxTmrUH z_MoZj2c@?TL*Gy~xmU4-J5#R2jo-^d`E@PKjc^>l&|Vt;xrR>Mo=rXD&9J^>7`zE6Y|~EG1L`A}A`?MkxdED=sfN@I)@1)HQIM;SfZ~sx;Ma5l^{x*vw?ZoJ zNY%m@ZmlHd-w0VBkchm9F)*MBs62z)DN&$Qa0{>=7n!A_QMu6Bb+tC z8GkaC%3F&cz&mJ)O6D`TSB6oXyKED8EG`D7YkR}Sualv1w-V_My-kD;%>_-nS1>ci z3!gIHTR>VSXE$*T7aPoR&P(+$Cu|-vlNj=?VNO_U0@jlA0D~u&C4tH;Ckchbn9JYF9(*CcNbkMew3aS;umR2W_`Y)K2TT+;#9|PWj zFTpL%h;d?c*iQZ#yhLYR{iC$T)G1?%>rTOVh_-NwgtunXW{nT zdRV*g2Prk0g>9_kF#k6Xdp{qd8y;0tz1l8%?8RQxP)$IGQz=;5UIHbI*QvO#2)*Pt zlW(75Nzwkbq&e&g$X0KI*498MmtffRLu=@D^L6xtkvO*8)xdOd9qhAQkC7`^;5?3X zE?=~wl;j2EuVBBg=1DY`hS8$OE9vc>D{xld7L5L|9gQ6}gQLp~SS+IjYIaZIqkcY! zE@e)EFP-#a%4cf#aU$>X!3pSjV+^(}oPf(*Yw?oT6?FGFh+bBSaGPF&c7dO8BB=-? zA{e81hC8mmp8%_#%fr>#4@q}7J1Zv|(Mz9fshsBme67HTD}!@k=Jj}T#z&a+Zx0}c ztKz}1dW1woN0ZgXqTu8z0bcUMmYUk({Oysr_jm!uocv14|2-uO zb3zH@G)VJau#xeW85I1ND4DJ5JI^$8vtw87icqOi$<4>ljmfU`rD z&}JPCH4gD`T{D&RJ(xm%Rp)R|402$>JQMhFaW%+=h2x^ttuWiR3{r#7keNgmj8|=i zSGz}`y6zJQI18cyV+?FvdW7^r9;Z@}iLU3C;YoFMd{Np1=a%=vvWy>~S`@%Lw{8P3 zBq4|=cr_gVkT^^ZipN0qyDQ$o1JuqT58{igF78>(>2~{Z&mX>oQGb2BGdT@^X1}53 zPx7g>M<}(C`NlPTJ4f6Gl!1#I1$G4j|H&5_=TH-~4PHTf#RDj+n*=@~dho$W1hy)M z!Nu#$J$;xC6!G8nnAlw~guvY=Nu|aE;-^qd zBt)JO=R;@6Z2cOt)Y}=3-d_(PtG&R*{~Ep&xy(G5$8c<(1mh_4VaZoRJTfM&ajgcaoM(SLxOFXr;2BBmzVs8H z`Z3Q)(je-~E8`xsj}SYNacg5V!7O+w9RIc#UMgkbz4AN^)+=PsKsFs!OrpDHC)3J( zd*HrN2UN2;cgKQ@FgNJ{oUkzgYAuBi_3g2-F%?%`IYggk`qSfmk{}iE-oSg2~)Hu$C_kZWl?@Px7sp zfBPc(M%2Kjxall!wT$tMdSKO-3iv%Q7ADG8L9vG*9=!G!re19$=SA<3Z4dQfVum}8 z=q$k94c1s;n}!OhEGzR!6|;tPAl-q0qqz{w)A>zY%dO$9)NHt1WkA<%Ie>>bO>AAV z6a@MkeEn^7JdOc!=DiPQ3jPXL^*@0U##kH7Fyl+ zT?+3-(&5viGH{RB4=*$4!tm4o$a!^H`l4(KExln#$4aCAmx+VRW(Q&U93yyt*B-3g z_kcsmDDhaO4v{?t4N|u2l%RQHKo=wYFN}@6YlOJ-!;0x>iB?N)?j*V;a#3(jZs9#lf@6 zRWK{e5GLe?P~FXO)a`XTl?-J*&N&BgK~W-pmrW(5y5Yooo)J0rdXQAj(1YFg0wA}$ z7g}btLD}Yb@L0iSfW;rk9vOKEn5T&I?BwuNg$&;DT}g=leqyR}hWy-|L#lq9BKuZ8 zC9A7NVSSY=xgwFv)xLhuC9ax8y8qgckF1^Harr$cR{j8^f!E|^Uq7jR&_N0h*N`tK zUy@6fjxb-Xi`(pMPPA6a5?)vcth3q%wk-jmsjv_hxkN!ikWTT;_kCEYH6G{e@)NT5pgG8HRxIo5XBOI4D;af}let{qy1u zrTY|lL@5q?nv=0)Y9_9`tPU3r>wwK`9{j!&0lCL-fkU(j);tBAust4)Eh^F1$P0TK z8!2rPRvU-&iEM z&35Y34SGOfnK7#REWlG^>#_fhEqJ7LvTUU}2K_6-+T>oWoGwGRDhv^;Bq3OS`U5F5 zUe0|wuS7bA&yuUT1K?w!i2AN-_`vlfmPU8u=k{TAI^! zPwm&#k%TWR!1>&Da2{^vTF$X<^WstX#GS@xN@wv`dJ_(su7Y15tDr6WAG9pBz~fc!w*5Q^ zB@da`a+vj+^~_15kr(+g*295#3uZdJ!$rm-)HpSsDn4nZ(j{ML&NCGnFK9#6W4-CX z+h|a!d-1lSvBeYdOq2U}n?U;f`NelTFcmozD zPsd+p7obv{F9!BFqpnm4)Ezhmm#jzFGjSX@tUQ7%tF9vd0Y7q?ag_W+S3t8{8IfHw z3xuct2awf7+Wcc7u0t3*cd23e15>ohRYw)=B-p*W7P6*wLD;)JkUG5;d{(I9PDgF7 z-24&oJ5&MM70%2}P=Spj{4`>LEU)dBHZMei$Lr{Cf@^`dAv(PY^42CpaK%1wuk(f@ zXIOvlSs4-SW{e3%b?^}~0y;Jf?Y$1L+h;1YUYLr)Y!0*`D*&ax-NK!ltT>xTZ>?@P zmvAkfs+OCz9+KB*7sHb+PvJtg1ZC(><<}ovz%nHx`kBzj za!LYD2B?_Vj#vKi(e=9~(>1K~^5;_wY)*;=#bXBGdPf>Y)=UO_ixRS3ri_df`oZmr zUi7JXhmNC1@vUM4dTgx2!gzi<5yasi4>7c~9UI;-r_&o5EH+KW=A0O`sbt>7tZLjf z`v;h-e}jGVAHwRd*N98s1Zb-xaMm{s?O)!)-u-G+NWB!ZZbqTsBPWa&`V4fy0F{4O zqP4jQjE82BvipIY;2vdKmTF3EcPys66Dr}=%~ZyZ6rc`uEyO7099Jg1g;Rg8PTxNB zpvuc)Y4MOWufJ|JPjWPwm$RlE=B=ED_2XA@+Ot$>Y>`AK*3%3*=mq&oKjB*0avaki zfoaYe_%uR<3e8K!Et5rP)POVX;PQ!r!#ARt|BL*rjz{;gd|cb?P0_c@B474aWuBo*>T=EW2ZowQ$GRfjT`H)jWt_?mQo4+mvxlDUZeEk!@7 zXk1~GiKiJ;I?qZU!q@4*jZd?nQ&kG9WDkRfT`+8ZF^T$Kuf+UsO(-W90fnAp(4xBm zzr|gKF#pxioyZtz52aD($7P7LzXUUf1OYcYk|nviBy-wu+`?${O>5)dWNr+N{AxB28> z`a&o!5&^$XSFk_9m{(haVExItD5K_1)|rSA(PtZpwZ>LFa*O$X*56}}ry))_y_p;{ zXM5DJNhl}z24=M!hrZbT+;aB+*it`}1nqu+Vd1*;&Fl#JK*<|~9(h3D3j!w6QuNKj zizqQE5rcoH!aE%o7?cfwReOzTt&=q6pWlG1R^0;QL|r^Jtqjjp3*w2SKA3&w4M;Df zbYAuxdd4e*uJKucy=fA7@5UXNo2ZW05B!9;%GKal7Y#2mZi0=@BvdI-McuEm7);k< zkKh_^Vtp)edm{&LcGiF^-yHaIx`lYZ&4wifF=%mCmUf)?;YB}Q$xFQ&z}w-cPV=f~ z)Ai@JQuEv?&?9{vq+c4M>(#56SyPH%PVC2W=3@xZkVnbfADH4=4=(LNu<@@U?2H(~ zyEY{_IX4m&WnX}~A=_VdT|>nOa&)1qJdJBnqh6i*IB@F#4zv7^`06iKAzNIzrMo*h z4NGSXU2qQ4)*dZ`Q_lu5BQd zcnSLX1TiW?6U*6o;F6C%TCaG@BHkjTaR zPIzmbG-{u}fU`={Q2Oyq+|Y4`S%>Wy+tab%0XGdC zE|J54CL>zotxTi*9$|4qB5Z9sKrlyy^F$gqv}U6H3A?E-W_hL29va>m!G^0?wd6KvJrgps9-vA#(H zm8blJM4_pSiDLk(edIVP868+n#pr*$B zN@Gy7vIQ(xsB!_dO;*>!8pz<|99r`wkgDETNe^6k4)YD8py!w|{MvXJ59{#bBr8vr z9m=PJv)yQ&*$is4RTMAf#-rBfZj`teN=zM>fsxf)mWL{|IP&|rMT6u;tGHD{yhV>0 zgI{MqwLwKJRhWS{RZ7t1TPGpFhHg(aCdbeW%4?W2V3OCc#A4ibNs)9QRW9lgi)qsg^3KQ;L2lpskc zoaYKRcf>&Kk~ph5K~tEYdMimkG!a(Z;DaQoSdcqcjXky+bY$~3`Z(AdO-KPglzNHW z=^{=mCJ`K^_u$SIwq*V*d$O}Y54f4%hdTQ}vl5J{1 zb{xJ(e%r02f`@r@vB5?fRrH9Pyx;&?P$Lh{?vkk7vkt}7i?P(I1eBAdu;6Y4x)OdA z{BDC)GMT8#Mi%^wQ^=lz2sWp=!ub@5lO2!t68_;S`1HeJtSkP9hGCcS#%)!4)#D5` z6}Zeq3z6sGO$6>f!6j!xz80uESSu}=|cM{eRwxSGzHrO#H${@}PFCk6uWnpjP z25_u>L59U$Nl(ga;(eB4+ke^ERLe(a<~c&%o9m#Rsff&UZ`h6^GIzMDKS~gvHUa%kE&>IY3YfA*8=E!{kkoquu#Q^?Zmx3J z*>n-&lV8I|Jr}NHV~17i<^?R5^%(4`l<9HOOrN zKDJwr>T)M>V7wJ2>^somcna*D{t!NCTjKo-(L6Kxc-}{qSYEWzAvAPPM#bORsC?83 z^O=Lw;ZO@)l#aqx9`zV(cO8F?97Ln@Y3MGw8XuJK@#?%jQhxKF)Optpyc)M2!&%Sa z;zb)=(yod6As!f|T!HWEk6`h`So}6O4z0hl+&V;);hTqX@$Z{>RiY8o%O9iX=LlTC zbP#~`erx}w;>kJN(6mgD>cxD;xjWlX@$__<^kh2ce`ubS)jcPO%n}Eg(#bHHbt&R6 zIw5yl6CK9;(RpMp8nd%GQIo_O8G9jf;Yl*zs*RpCAE4n=rFf=oBRCV^6W-R;&r{}-NiT_n&cwjzc{kQ@q z!*3Ws*iV+s zkq0xwDKPZd4kXtVv9r-aH2$oBQ_8o2%Bfp$+T=4>Rkh;!hX0U1YZA@zJ&n2rrMP3T z40k)U5R>~KImflntvX|0L*la|5GAn$v;&vWg)?MnShXZgQ@a2>_I_V&y%__a?4iM< z(bPvbnC?~mMVI*Brg}2Wk-jq%lP+(;kpOdaf4vqTCcEKg6)QZ{!hYZEJ#f}Ee-uax zfg<0xM6Xu>r~h@rCf30ZV;P@7k8C1mr9g_m$`V(99#&P&K!r3@(e6wNS_F`y$ZHs5UJTS3$4&L?c0jP=hZ2k`7^p_uAU*KoI55uHTFb~X*2f-R0OIUDB0>fpj@zXOge0yR&7Jalt zMV1-!d-{qh2M*Dg%q|*Sz8L-M?Xf<>6yxnTb3)CJxyX-qNS1U5C;i`4;u3#=44V80 zil*ss#PTbd&{>B&q7w0b-cr^*-;I245vk@FESr58$*NuG|9u%gEBlD=W?n+miMi+* zokUx*w^1K8N7}cepE$Hgz;SifvGQSlX5&UM`~Cs;g*L(QYX{)&No|x%y+{YI3-N@| zn%7_?LK8HnP>&;Wv5VfQ^eAau7{@hkUS2nDpTSy*mQ7Oet z6;IaTRK$Z?)3H+05WBr!f$nl;w7<-{fO&D`RINVgo$!M6n9GABe;lmodG6Yy2n zI5=7Rz{D&=%(=e_N6X_dNOUEty6uNs0%AD0@D>P7DuJuJs$jE#AI|-}9o;@9VU$if z`9+?QUuoKq|H%+Nu2^9_;i2vl8_b>Pj`7d8;Ee}*JTE0fo@c8j&)~~^Jod#3XAxI) zat+{?WcZRdAR2o@$QFkWtJsuZ@Eo=yE89_ZHEOyK-!*aUP*Ok-zkC5#(pOj6-`+RHW3tfVZo`&NNvSI%Qpj2Lc@6GeIQ5iGV4 ze5tI4dlOe7zwA`}ku@2&sXc`%S4k{%<6%-j5DGovr~Pk4soV`U>Y}fR%I_b7P2*N* zeqanb9@;R`!yERioB}8Ddt^#^lf}tpMKEBuAB4OLVYU$SRLLH~kL4MtbBZ}+>z?75 z@N=xX`2t+56T#GBgnPYaIfU(vhC8R4L8BobKCPlS=Fj|@iX!B-vN9A7UxS^ECvoKM ze!TIc9^d7A(B>vdT5fd;^~SQnRon(jnfH8>Sp&TJ$b*?5l8I5|0r2>l4RT3`LE-NN zq#u@GTR|Nj*7bmIOF5jez5+Sk1bAJ-!5ZaN5U}wO2z)Wd=Zm$F(b6H~$Zgmk9Le&^ zsd%?_8T#|Tg7KhFRu=<`xqqy8-YE5pOYA5m?g^K`J$yPRwS5CO?W7!+HS;9w5=jGL zJyY1PZAz_LJ!t5rX!>$dDn=?XR^p8%5a_*^E7bUIbv#3cd|g|Q50AdXHtpLu_;(6q zZ!=>4RaMCK7pE^AO<6y74~;15L@8PJoXg(`r91XvqF_@nBNLVwDiGo zkrOnBMZ%2x(Wol$1|v^Sq*wmNS+(}hC5z4!lEndP_>ws&FPp2-w-@=S$hAnkEI0&r zPTnNd51#sT(o7qpLF!4BEU)F|v7`E}$1gm^YV``;_5P{W72nl50^ zF$WUYybsmNpY}%F9(#Z zUBS|G4}2#{bGUdupIHGGNq3igmGQvM`=y%Gfb&f)zVoXuM@?#;6{ zIgREiAt)W|ivHm}psSt6S^2-RYFqpjcAw?PqX|;@`$;tzM_QtwO(t%(oXgV+|3X9F zU#7;73So*~KJXWmFgzaXs~ydSRJlbkchWmxt3Vi<5)WP*b;u<@0}xU?1Dh^IkuJ4c zM4*nb92dlbhRg_zJ^T%h-;TiIF-Itq@dby|;y7n~9u{@%$Fol7iSyg%BzcPz$h}r3 zzGtp*Gi+TF+lZMRPXXIIc-~mwPaMv>j#F`oqmG)<;}04)32IWG?Pp&@~oA zt+WEnWPR9?Mcp9kqzHF3l1NzB2eRD6iFrM&P_oe)f4uui(iIXwnt5oqEQ+J|l(Ol& z2A0cbUX;gj=Aa?Z+G8_xp?#7Nm@F~?&9fO`TAu^4k6WSP$Ws)2GK_h8>U7T3F5=z1 zfn{(TfSUgz^PA_A2e*xhw1g=h`xT2jzm(vnn7KUfeqCOKy9)2k@rBf&YzpAJ>!uyT}h}bTGA2V!ltl}F4+-!scztnJvvIZ6lZ@`}em7J?~99NZnf-Jl{NS1WZ z0H+ImQ0r@rPrr(S?LahfsfdRzhwHF&^INcwVPKW-`{5i-0DkQ-dd<%P<7X>?=<(T5 z*d9x=9~6;0uk2xHjxq*lCt=*fe(HBtlxG@h!jt@<0Z02UlV=wb$%<`DA*?ltEZmUA zt$F$tHg8qI?+*=eS^@Kg>oErPh#vJBxQdQ@GtfKE0u@JJf?To;{+2RkcTZvb@hT3K z$Hk!FhBFPGCr>+Ex^c|)6quE51z7_*SX%Ib^IR)Kj>Jp?K_7p-QD=Z_SkknnVKPLg zmXp84i8HI##@{~WFe`g0Sn!>s#!=BUyd{7pEUw2kxn$fUG6!o6m*K9(O;ENx8V=rm zgK6V+c!eF*TV0qp^MM8EPd5Z1v4y-n^EG%f0#ZEldrPs)${YJS3F~T`&ZjeU=~ z0Yy;~kf%ad=}n?9>;K}v5^2&JZ$Ksl@W`kRbD6J`z&_tmaJ{b!zB|Riq`H$BYp+K2 z`+WEndZFX_EC^0igB71bNKv;oUDMBY1CWBgJTur({E$d$z54&2-DzV<>KCI%7sFqy z_U^{;eJ62`L=W8G+zE$zKf~T(Rq9-ojfvZ>QE)&7&$Y?nw%&el?9hXjC3$eMDG=`6 z;3w{lUEG?zW!$f6%-^>A1vFaigr&L0#8mV?r?jb>R9-iM>NXqLD)*hdN=}B_u?aZo zg)5rs#`Ac&?!5PD$eSjuOb3MIsF;*Etzn0rH@jbREe#JjMcc#ZD82>LUF@(?K^=?_ zmy`A}UGQN{+S+yJK|)Cw-Cr8e>8+h8J(!F7kF-G~UX7gnK4kgAdJy;8ze0)b$9UKN z2k;BsgOlm`(DY4?6R=O<%Jmbtbpi%Bw=kKRSq<^ExuEZwHawq>nQ~7U0v7 zzYuQVPmFAnNpZC$Xj-Sx-d{*vJw)lZR%80W_8sf0+`z5xZ@~AZNwC4OjU0;V!=bJ- zcuOQ04-E^U(%4>*OVA_l8UJ5#ktm*e&)AX@FJRno0*o|9Ydw>kCj3@+U{fqnM-Fzw7Pl)4a)mWC-Pt+W$6n%FL=bQtpb6JYu1 zF$lYJfkbr+kiGGSxO|!%2JREaoPde=W!YrxmK%oJ|70+9?sU{(d=Mio1Kxghe%_sZ z?R4*UE!0)E#<+P_`1VjE+!84Vb3f)-?5O~uuX#{DZ6{byo`@wE+kw>W0h6RV^wbf? z|J+$YJ&c!v^14J4_MPP#4wayDZz49+DAZhFfGUo6AmVlwbUVmGK+PTQqHvd$N4O9! z)IAC{ZyU%N1ICBQ?;}aF*9b`SQ2Uf5o)*0Vr+5)4E$f7#p_15~FGI}ucW`BTGdV5E z)8s@`9(iO^Mcy&iZ;+xlOe+~CWeW{3;_gJe@P`i{KImdyi3o`AOovFPTB`l6k|wN3 zrKK(WDDb5K1|OvYsnNsIx*3>YwGhusmBP#Ln-ImiJ*6xc==bFv9Q|+uyj=q5(Fd#P z@&rQ~6MhO#e5-=*|LUNP-PJcXy2BO`HQ-BO&q*R58tI>c!k;3vVH2_dG;SX z{j3I_E6#_tm&1w5qIA?^ENb&tt8t}E624iWjtVo<0K&!bcX0yTIsT6fUr?v|OHJt9 z^V-xTeG*JCNFa60k*c423FaJXh5VmmU=i=aeV=okjCe-F>YY_k6qo?(SYFEABpI)` zd1FEIFVH#_g%xL(@ZPHXSUS>ppr>W4bqG~ka z_CnlpYXvq~`{G5mTRAvsEAHOA4{E;OMF;g$xQ%@Wj!sU6x0!X&byxu}JLHmmmM4jf zp)SlDN#;(C@mrl+{>mzGKJyA0E8r4h2uFi2;MlJR7+(1Z<>IVxU7{B%MlZpLIwwdz zxu5y#=D_mv#|Ty=kOmKRVr2Q0@MYa0KOVY+#$Y%+9hro}7W?r`vl2c%%=Yu6E?Bvx zil`|M!ezb~oTOz3^X~gVwp0#WNlEAVZ92+3o^zQuW@>^n$7*2M#RB@-^n6w90q*LK z?Uu*N58(NDew;U84$WGOJ?7<&{S#_YnqP`%c14aCAELl(yrhAt3mCV>AR6mf?s@tf zAyl2Bg9Xz^n2SsY&5G7ydrb{o*=U19hjyc0k~>T}@`tE*kC11~_F|3a!?5P4n2!((Z(ti zbB6X~ve_=i)xHP+nLmJx>&DpK;{=Vj-C@=;A#e(1_efJ)$XY{*Tc{CTKd^wxo>@hs zuHUBB8%23bn;m!;)3+1uVlUabzn2(__Je3?15vq_No>Eb#Po^F;AMjdC+~O^*BoiY zC|Snov{GY!vx#Uoc{w`nzX&r>NtbHWvFKi*6%25h*x#jRlfLlLI`w}T|& zPwF%fPgyU3P89x)AX|Z89Ejt86{sxhu}G+%gX@f6w6Nb9dyJECyS4{xdzB137REtN zodj>O>OC5JHHj+y7lc3cnA7mgO1#n~1Y%#rAfr( z`(Y&UG}O0=pzpz1I6l%26U-&i_kac(vb^Q3QfEB8vlu?SoPsh1F1UTA4+fv-5&Oq{ z#C_0|^OXBarkjM2MW1F9E3q&*ebo-)k4}M0rPnd2I1`WjaAUa;V_1;=o*0X05=)a* znx)x-kFIRSNQpKa67$A=tdsuc7t7!B`NGGXAxJEnK#j=&cC%gHVZ-~}KiwAM6uTMv zoLHYTP?Kh41<KB$tJ-tL))a~F(o!edd~M&T^0AKtIQS6WnvpkTw%Rc$**2W^8Anr{D=q=jPLW@@aT^ z>ldq?TJ>B|_7fsxlMACyEzonB6I#w-UTD*`)a*+#-NE0@ok}~&y;yyXD3z5$-M{Cs zy<`ybSntkIObl1ulf_3C9_W%Pf+-JP!HExP__S>|x-KildHz5izUU${6K;?*LmSb( z9Xm=!#G22RHK}Rbck?hp_rAc| zu>$;%yBVkNn~(DL(kOBMFHAVFl87Ysligl-N%znI6dde>4W0YJK{6Wbo^OD9c4k*7 zmc;A4`|#v#B~-LLWLfzt){Dr3t%YhjQy`;CL7Wlbf~I@3??=UTy_Dj$rO-6Pju zD8iDEN!W{hpeO8$)qm7*(@u&X)#LH|rvf@GolX_!Gu{r{!?lI2!eVO|mhC!#37+0K z|Ex8h?U4czH%(AwImOmZ(fFUsGOS2a!ec4-=+~u{G~O(m)>WQBHD7?|SbpVJ zzbM|GFb02CHIq$w3m`DyE4Ev z-_DP>cgS-qQe?qZ!4~hy^pcv40n%DqOg>zhOusSjap{B~=uxkXf1M}cc6G*v9TF!u z#BY)Q#AJ|;{6;ol3z_RC3x|YMDewAutSx5evMpEf;-3ukIvR*mn=avlpWZm3zzkJu zR??@h_s}Wf)%3sjfw;EK7Z*)fi+t1&+n6`4E2;&udzCSID|0F;>!PAU4@M~5z{xuI zQOv;ri^fJE<-Rq#2XzwZP42)-HW16Wwqr`)VPkP4Y@7KFUnX8av63jves~UoJTqaZ z+y*eY)DMy6FX84|f&l|cFdlG;oO$z_RQ^rJW&hkUZt*O`m8y?AR3s!;M*6PhFx;zA!;tS$KnZ>FDsqR%PZo&%GJ`5y^#BKIMluXuu!@4dtu z?H8%F>Ny%zd4jq=R>75bcj8Q?Lu_7?%I*%qpb~JOIoRIA?pZx>c8(@0mnVSyttmKi zH6PE*_(AF}XSmtGTmWe~AoETbmu&FCaHB}v)xHeNUKnGjV-(g!_n^VeKiKHI3}P5J zQs84cv}aBtlNQ8~1^iElpaCB!ODMt`Re*(ee=rBFFg=FT=-%wPSR9;zg&#lSwcTEH z%6~}PAT5bsnrZy0F8vA%YkRq+9vkIC@76*YNJ9#Ed__#e<2d<;XTH-Xw6!qd!8 zViWfOI_{kS>7`vD9N7wacXnZyT@6~c-bJ245kg!Q`oFJ7`)9q}`ijF`i*7IHKKms; ze3Xq@k2Yb7+%6ECJrB3tzJ&EB3{i^l$p5Rl1CH6B$z7*MB%^thWF@=8L)Wt~)W?Un z=XtRXG2n1GAFu43#|CA?W(1m(Ydc$=~vd54ym^SV|Apyo9L^xioFhIumhH&FxE z{T9O^JARCtYk@T{=b*jVdVFlV7^e@;z{=FC^tf*|jdkmyq5Rs&4V&UXxDCc$IRQdP zjj$>>6-zHp!`8t0sHm!r2_-`$#4HSU&8&f)mnCpVWj#EyJp|_rn&6+?RS3Bv#vD@@ z&{AqQ9-hdde^?m``CY?Sv154q&sm}+{)`;d%ptLt`QXrcVX*DBgP(O)5V|1{g7lJL z{on;}rheaSSKa9|a#dVa$@Z#qmWU^pI6NEj}DanJk@b;WGrM-?`u^^p-4a z_)C;j8_2Ie2l4jUJZ#u(j8GUt*T*fVt{3M~#g*E)&O{iaoIgYEw|)|K_%vyGpGW3Q z)y3}56eoNLVVvWiw0=l{=Up^Tb$jpN#@!z=;nZV%x>gy*44*)7)HS&K)(DiG>%sb$ z5e5dNVu$8t{8#0P|5Q(qOCiY+8S))sS52hP4Bp_g=a+v z;CZwXie8zDBmSoNe1*%9I8fl8VwmM4}Oqs5F;E6QxW|N}-*Z%XYb@txZzOK*p8JN#u$Iz(teiXY*2kCOlv3i6hp5$(V{-H1%)3_vD8Q2v9N4mR^M4BV}c6`cKX9aq_1Y0lYLOr$+L*RsR!M1%Ycjt@i3`7 z92VGTBahU2v~pcBGE$JjuQ#=zR;Odgou7>M>p3A;<9KvVsFyhYLJE~1lR!}U4wAl$TI_S-(Z z!YUNS4&R)h+kjiP{7Z}i!@0)2D5gjC*3 zf_HQe7T2l8s;0l0`WOo|J8mHwNm9VdJ=*xYU@YzrS%OZMIib+IUCipPUZ%3^1n(ua z=iS&V!i-Ni#rSKDFlKQS-Kh0M`s!BbiGCq^x3?I<<7~7!awj^peG_`MGYKt}E61W@ z1^7{a61G2p$mV%FjgPaaE?R{*669mA=RSLTcrUz6E)%0?@%X2BjK#&dQb^P;Pfamjhe#_%EJldjJ)9Z-UwJ={EDg8kwbs? z2{U)jTA61%O*Z$x>(868dppxQMW}le=J$pPzpZ!nPeaC!B+vqjMEq?aABzrO#gd85 z3@h~3Fg=~Z)Y}&DF3aEHDXT;=cc)HAW6r9e#bak6_kt!~yQ?^(Sl!K|*HamrwVQ=I zU(EEm5N6EEc4k@hBPQurC-3CZr#$ES{Y;f~J8~NtM76dO*gD#nr?<+Gx1QHK*M5uv zlaaTEk+46^EW52BoZZGVwn64XzeheZfw{mu7>Z&VT`)@46M6uJ?Ly`4yKwg9L-?>$ zKJHA6U{r)t>Ai$kOvQJc=V&iHw}q5Z3vh@HC3tdCG5u^gEuu0;BxfoM)oHrjK-OX!X6j#k%tpu6oaP*z4Ky8rk- zS`wqkxDKf>mRn~qg-yv!h?TIv!lg{5VHKlmQpx1W{h{AQD|Msi21>k#)qS4@1 zq$bRT8k#SOdOS`rs}klh4d;cN!Ei0KO@AF~(lkMR7c9}tM~hI5>{@2a9eZZ)KU>C( zCx&18O~B;>Wt^JyfRT=GW`dOxnY!7MOw;)dOs-J3baumK=7xSD^X!GTPyX|hi{nW=N;-ji^rIm&Yojxy2u8Lxc|Q% z!XFb6p3(ogPA|+Q4`eYBOc9WG!}DnN?SJHaPAq4UZv&cs&iGui0jOn{5^H-Cydh>g zSen^$)(=vdpy`ERG6dY`do#(WifDSlL!bGVas}<*sgH{jXJe1ZOcH4DxYqc}=Gu#f z)A2$bdHT@)Hh1$_BHkuj$v2cYBH>MmFbWTBqNiZo>5I(p3F7P0Vi!zvqIA^W$ZIJm(Zd>%G3HPeTgo#%Q`zq1aM)$NDQ zJ}me{22gXbIJ!NX2bQI8IS+LgzU5OjdbV#29ex-<-rY>*|EU$uX%pV?*CgoE^P&5h z+o|`MrWaf2uG5c^Wxvpe?tM6Z+CK$i+r;pqZ--&7sS^I&Hjc_mh{8xw72R@pDmN5X z2HopUk-#inDt}fR@kLf+gLOx7@vsJwoLE9OF8@aJ#3N|Ycuf>;Ehn(mm;!cv*;vF& zpDVlh7aas8;?sGKsO;TE@5xW3!MW?m;j6FdjhS&2{=}2zQ&v;^GldW=dVHqKEGw>_CjOP8(XLS#;nZ>JA& zX8PhV|F<2t;L&OF`t=Mt{%#ZVU3DX0YReC9bL>C9gQpnjb8I5-Z|TzLv?lU$*)2}p zW)rkmM3cG4#M!g+KlAUY5?m44g{|_Hgn5(C=^7b6J+M8O(YU5a*S|BSVdt08Cnry_ z7Io^ZmbVz#hD4C>2j<{D*E?jwywkKS#(~{&yca9bV|bbRV#=n><~*NrggZ4#Hq;#^ zQj(9!*t+Q$Xa#dh;t1!nbQg-ME+WT$;>p%2nWQ;2ABByZ!<#*9$aJsx%*+@&6$?&H zB~kAKxV_S2x$c`P?4Z6OXQ(Moo5o1crDh7?ymSiQ?rcoFvi9<8`u=c#w7PKQtLNOJ zJ}>6l;!ZqNFo5UxCxUUW3cJob4gXB6;=1Q|2uGb}5>>noR=!CktL?Y2=XOYdx#k?Y z;pZ)WYJd?wo^hG9jF=Mtr%Q?Wle6U0mO$pZ-+CnTHwUDLI>`Lq6r!%7N9HPj=Q$T2 zB0=aVcFOoibR6~&(MEf6nm>`M7_1{IvkZvzpfo#DDFekazsRZ&GJ>YgIKj5%d4kg- zuc*eGXmBsw$>cYM)rL)+BuETQ7d(!9AjBm0qpKlx9R6kii;AM?u-XCq@W~gvqxT3M zv|moIL@QCnkaGNVG>re-)q}saScS_U_l#Iuok9D~*3)y&?YMq*J$>ELLiJ*+X|cgW zEVWS-#7L4g8uRA1ao(}3Lf9AHVobB9cN4 zYTtK&W;7L&Q*ZZiXI7g-^|Qc>_kn5h8o|6Dcnr{ zcQf%?2P%$I`VOLmfpg$7B8D@^nF~U%?qv1$NI>#47k*r+H=7_`<|Fq*mUI%CDKiN>!KP3hfv?_Uj5NVJSm)@95;5Lu6T>pB|+0 z{X)F%kORBGDji?gevw?|GxXi%PGUb!i&k}tu~K={sE@iW9qu1XpS(Aw;Z+J$+58Bc zoZ-g*fmBuEW3?a7-`3>qIxEB4G^?;#nTGh`yaamGeE?52R%7qZlg55CGiin60+O-2 z4aA0wX~)Zn@H#-9Ht*{u7w-4az06!1c*m3KHpxRpoImlb*2P`J3_5gq1nxRt07Z3z z+hx1B=YMiIQ9B8uV>p{FC~{{V{WidxrJ-c5tqklu@)n_HN^`!J6R9t9prksF{<&O9 zq!>&u-qL{6V-MEkj>?d4!?rlGbsEdN>dh{=vWHr~GbO5e^~^!}56H9hH!`qEC2BN> z9<2%_US`vvY0d~;U;L8JbXX5&VQy4&N*;Z2@&hzAsY0?&A9)@lMW+o`@D|RPKn&#{ zW7(Gq;4YU!-6H~p=+Zhd{BgS$9+kL5ddKOrHI4#^{n7=A-=@M*QwesW zK^PeRtcR{MH}PL}O}cR0b;i}znM*3@Aa}`1ydnP&H|1nLr#(rBuJ)cwPrDgGI&YRB zOW{3D-Qx{0FA^AT+X?deLOmzn;YaH_uG89q1gf|_jTA=2u^&7f*>;bWWNxVi%;_n` z5%PN6EBC21`YonT^bOwd%ZNVdzD$=FbkpvTaJUv)ivLgj`Jeu0%*H$9+MC1NbNy$; zH%J$^FKwe&)MW*$&e;lT&Q24+lB3jmb{&~CEsyZseqj?|DQYuMh{LXw0ZpS~+N1EA z@NIt4v*KTw-wvLnvb%)XOEwahRaG>opo_Ys_SZfMG9%k(Y-1wcoTr`L&Sck7UvAY} zH!K}rNLKEMBVIuoWZu3@*tGH&Cs~z=mkGI*D({b>{_7!p`MxHi?d?zRed0jD;sgK8 z!Xr4MC!522ayW3j${9Y@rv{f!Q#a@1)T?hLZ4l$4RFzILtilf4<`U&6 zX7FjX9a((9jdq{P;GQ;U;`k9+D0R!EIn_4wUFRS<+8#&$?bM}l4X^3ik!GxRd@B8_ zFrGcVZwM+=K1MG~2zC#cQS_jQ&xRfP-ZRzK3!tBjH5utt41=T4N_+u;k&=3GChriPS_@j1Jw4#%^qYFl*fklyiDW z!^}`>B(XqH`B+u(yy+hOW$HkmJpd5hL_k*~k6Miqto!9NWr`Qk@%G2CtIu`(yY>*B zR}(|GROVBYDS>py`UINyt&}@7MW5?9(@gYI#}K>eA2^K%>&X>;ecHP*hqIb(gX6>x z!Hfq7X^Ex|7wub0MK@V-8{R%4g<8L`{gY~3*&+d|f7D{q21uMM?UHq4$dzW710Wn(T*VueZ|Et1C$F^D0~v z@QMuFK1?TKLlX2cnD#nVQf2)Vx_(R})CT#G1D;1{ZCN#q_enL0Wr$J7wM7shexG;7L!C zplqWF@qd&>vP2D_r{Fgx?_;1ca1#5!`k!b^$H}yNB?oHa?POEhWy*`XM&~BnrgiV9 zL;RU{N4Jf`A@da3yzmKFM0~8^#%m2&T%->NqPCH1-#N;BANROylPzGrQ2SvtF@8LJ!HZT84Tissp@n=xiWsVI*U5( z&7rkFSK&>nk)Zzb5FBcMj_bnyA^Sn0?j+WL)iEoBp}_^%uRD(JoI4rM)U~6pBCf-B zwLN&9gA}zqe*gmgwu9vH6|gl*ML1haa<8BEf$Mf_>UT#&aQD8Q;N9Jc&^q3e-5Hn$ zSDq1Q*EXaF+*Dwr)tanS!wfggtF6BC6F(6zONw2cxF`Mi*iu9kwv z>QTU_-_l&4L0q1(9TMu5>83SCf)qt7!R)O!Y3ZzoRC;0+ZFx`0=2;2!OZ5q|HNS$M zwUH8(n_3C3eo>)IwQV4>rvs!l-f)>>D)>ptRN7O5Ei0i;8U^ zB)*?0=`W(@mHVmMa2XeJ_Y42ha2_dsc9Ll8>>SE-j|q+l5xKHWgYY{dn&`SFm|S%d8|`$_WFQSz*RGhHO9OC0nw;c2oc zx42+0z4>$**;LlX4E)$e76^Nl&&}s>$*?auvnU2`B`4FExVYsv>Maw_eq}Or#ON_ubvcN>{NqFA=AI$H&DKJTMm2f& zXdJEg?j-&@Uy&bcg#L47>CkU8kE%TG0wt#!FiOG1;2YN(Mqf9l+>CEw#yijk~J%_YW01*S1E!X66(1! zWfG}K^;m(ARRy=E?F~)zx=UWzJx0%u{^693q+zGE9Bgw?U|S}1^W%4C6REPpG+HB) z)=f^P8+()mzFKo&c2Yd__|$<(+eCJ+nGa5^LgbTL0gX8=A<%wTMNht!qu&;ffLJYp z1#eu4&rAuDDEX4UQ96m~oJ@Qu$QaF8|B0r*w-YQbtrP@k>T~DC4stFt=5e>i$))(HCqD`D%7LLA*}kF8oW;QI3s;-D9XN0>{T94SGn zr-k^wxF~x!Uy?e;@yOZ1IC!?ynAHzk$DZVCVfnXhSjWT+^`xzVHj!vNwDmFa*WQWe zbjE{s=2G~~pUHQ4vyn63ZiNT>3h0%!3dq8t4}GYLfXh`TxO|wQH9}8@*AR~k-yK1d zd~IO&Vt=|aqa1p=Qb-dO!v*V(qWvf41Lay_k)V10qydzoSQpC%S zIEZV@tzkA%0dw{9xOdKcmNw6ZBRR=1scr{bW6(f7Lhn-xxAAbQdje~HBo6Km7Qsuq z9}xZi0kXMY30tk5*u6KzS(~H`C{-H7-{ekV-OgP2X+*G9l?IX(cjoRH-)0J8kKq+A zEue8;mvy^R0C-QX5Zix88g8Jv>U?r*!QTm!fFo-{bH(fp~pR60Vi#L_?Oguyg(c7te(GvxaqD2)pg3W-Ik** zU$xo626<{yYDS~$_e0owM=;)~No&m8VQ9}u^jypnN1KLWtuvA6>ZW8Er>RI@HoZl4 zqx-RGeJ*|7x{p96ZK=(ADVWsWUYnlunvB|*!>ZN;$Mq5;PQh9 zp`|LFlr3IGbr(pne`}&xlR47`L!tLcM9(Ao;0i|@Z@#O^pS==ZXNhprJoV6@4lDkd z%a^fV%34kd--qb)Dpdbh2rNyX#co(MkwasB@X;yWz<;`1uuA1V>T+p=4MzP~#R9P& z#y+fJ&l`SlpBBnGzl~^k`*PD>nK9NA3i&Rxo}f~#P4tw#0H2?t4RO+jaCV|S%*@MS z&%JpC4|a8PQ+sluc5x^rX)4r1_-%M>6hH*8k6@29ZPr;>OFvG2MvJ}n6N5c}@Q7qC zAj7@F{>2b4+)-(6SrVv&|!!?R7#k4y)^RUd@ks~bzHeT z8l#AIq|Z$zDKn>Y6Z2a*Tg`gB>a{gqAk1sM-ZK`C4oT2_4O!~lFoS5#SkK>Gc!um# zF(V`1#dyv-NpPBK3Lm`J?Xcd zW8hH88BHOjEuoC@yw!t3O{kmR5oGe$78=QUZWHJ!U(w3|s;*o}W`nG!h8 zCsXVn5;>C&EMFpujDtTDc}q{2vnLkzSbPSTG)Z=f(xqBK;1wJd5W!VNRDpQF2p&wG z!yVXo0<_g{!W!3L@$twH4fP4@*bl6^fEady9i@_p+A*oC}F?N zq!#>Y@^wWo_x;W;k~MiOH)H%1;59YjKEGr9B_(HZj^=f)?a>+%V(Ew%kzLrl{3JRq zd=}&#jIg%n27EEk69-DvkwsQJX?5WuS}ea7=WV-&|0VIEuc05;E?f^Yb~;1cr4k|v z2}IsCj23P&7P$E?CLXWsqjI*Z5_)D&6e)mnaKvkw+U2LQu1?|7;)2e7k+0(?~Ie zUll^#h=mT!=a~S>RfX<7`8YdMXz)^=Lqk{nC844bB;7Jh$V~{qx5fS7c)ur>t}e#C z>TUuT_Z0%Y12KXHmvrbs@zb2rohbZ8=LJ4v`5u!S{_wj`1+J{m#lIJ-kuvElcs}bm zJdbsxj^Ag2&eS9Hc;R!BTdhx_6Re5QpxKji^&Ad>!e@v z9zHS78}vn&|L^(}E!(;p;9V7**|m_h2zvlpvQFfR)+5sKx&>K`IN@g+dr8(nI@!@R zK$eDnKyQcK;DFg`nsVX*)a_jZO9pee+{y(|yF(d84Nk(EODk|}VF(!-e-nKYIYhqa zPv!9Ti+JH3CqhiJaFoqgTp6wkmuDA2>5XRaeR&bqw5{ZjXAh32^&nd=fyN27bXV*4 z6UUFU5J?}wZ1z=nwn`hEMpptnehH>m-I+g)5$J1qCiA1W0S8`K%6|N@fL(h#8>_i* zr#gBAL`UfbI<#aFvRn~_evEI%PL=W)YHQJ@GsW=n@p1m!HD)Yz4dEu1ay1I`1Zdb& zjs3LElvR6M2mkS3Shp|$AKG+-)AN<1U&Sps*GZ?iuY>NmVADjQ*EJe4l1YjG=bY)L?BdK zm7V;m85G9KvIF*~!C%9e4J~(ockz-iVN)3=GkYcpO;x0_?-{BD13Z3i{!rQz_)sKL9yoqa17ZG2JX`ArczyKREZ^}nvc+eMtyLV6ej4Z*VJ5W zk7q8P>qb8h2f|~iLqOvyu;__T=&t+_-nnc%n|kOQ?CbM|;Knr|_WK2XK6Mef|LGiB zmT-Y_?LAM@Diq)i8YTx;+R=q)D2N?afW5n?QkT;ysLRL~+xXW}i;id-oHU8u-nkfi z4oR?u?%5e=e~W7AC-4N-3vc32>XqlV+>(*|71|T5s$a@d;s_L z6|j)Ap`}m4=+dWKK;@Vsthd}uieHYB$|)-J?4Jhcm+yveB0<gQWLrE-pP1 zfYbME!5zgX_`I88yL~f6vcUv-H9w&uUUIeq9!E0clwH_<-_B8F8y$Bq< zK7;*GIxMvOfeI@XJ3xNV z&I1v_WY~3cEF14W6*T`i!lV2pq;A1JI^FanU7-<9ZVKNogNLC+V%lO(@sJ9B-!Xyq zKrZ<(A)fjKDbeSik5F9p3`m)12_`=^N!+YXa`Onm>O~KT|D92o8CVKGolZm4HyJo- zc?T>UKf(NI0o1eT7VY&MM-`={=%=LXct9r|PS&KsQ;(DXyZ(r5ypn<2w&!t&S1hNF z`o}@?>oe$0x20-b%^=R(fRgz$OVg!kAH>m^Dd-KE*wgxn}{%+8YZYj2?s6NJ?A z&o8k1Y!%|X<{i_e;ZF{oaKN$i#li5xSUm5rEDb;5OXh@Xa}&k|;pWJrSVepQr@S6d z6m#70sK)(Tr<1Mt<@622t5}yz*0_iUVmpz|&1GovZy>Rgu3@Y9Ke+zOaokXZ2}uf@ zhc~LTcu{g0p8u7PdmReMUiq_JLiAx0aVv@Ii+o3F;;s=19U<2&;WlGimcnfDw8fR? z^Qrb>W4Pd{0b=>?nvK0TgSdJeo5aZUd+59Ivr-NC+6c_<5ax-u$efqeVOl!M>p4GE7d$TpQ zQ`tu*cTiFa!{;Wb} zZ|8ufMjWci5~R_{ksUDi#{nx|!uL9WfzVaBzBvIKxVDoJ zoe~HSD+Q+@6}F?bnD(yCf#%RLw0J`b8u#fM5f8?o>L$yLEcc_D;`EWDg$H~B_nqz==T*e3LKX7robcOh~ICTy=OD6qV zL97}C=sRyBbsIiRq%BT?O#NE=`QTWZ^6@^`&})tSyvktd(kmo!dk`_Z=?LlR<8Z*g z4VlZ*!k(W*VAfy@9<$^(=PNY@X0`_K20S(J(ezS$zx@GOzB+?RxoZOBp1MLz zRS?@0X347mwqesNvLG@e91Kobg3$yGMm?sOoAdhtS95I~e)v$At~c0>=k9t&UA`2M z$thh#Y|{>MZ=VO5mT(Md53~})3pHr!sXg%a%`5WI=@iZUxC>U@f5ge=#i4QC7FcRg z8(Or!i78nvidIFugd({lcvt-exW8RQ?#3KCwJMKR^4m$xiEZTSb6L`~I|9#oVZ=Qs z{6#FS&colS5AoHHezZV1$Cm2Hg1K`jRBU&led3w87wYhd+i|o=t_udzb3u@=it6od za4ET}G|27_J$dROy_IvB{GZwDfAF8BXOv0qUlBUuSxH+DY$P87^{8>_9~5?P92v9^ z=2Sw3`{4b`)OKZj4yT61h zzj%n8vtjsmw0gKRVQHl40w!nVUXjNQxp+n|PcZ+E9liQ1k1Vu)Lb}%F(}kT{6sKc> zyPldLGc)hE*F&(4thTdK?zb?=dYl6sG_V3c1Tyix4*UF`YVSi z)p<#C*9$l!Y9`q7csj)NB_RLoU{V(yfNslna)0|H1Q#Pd6IUCS>W!|ZJkmb{XXLw`%&EHJF$$=Lqkb$jF3__Zq|Bb*r6zZx^fo8@0};|KW)gX)&)d; zgA^@2ae!R4R)9oCgDz{=<+cVM;)F(c^oCTqaCfVL?`vQ>khq&Rm+LUC6UxY%`_bg* zT@AW_n>^*~q%gJ1PLhV8e3H`J2~9^Y<6Y^tWX;%lRN1|V2G|$~Q@B0piL77r_ih#X zUPXpj=3nAI{W(guES?X}P1o^G`&=yOilLuOi|Cf2R{A{tF6kG4MjpnFrQhC(LDQom z&Tv%*zSCdK{}`dpDGqs%S-Ok4zwwyve=&|QzlHNsTQqI7(q-`u19VDHn!7PlgziK~ z3jCB;(o5B8;D2>3iB#7SesjGj@AMIp{8@opZ&-*WzuR)E^ZDF(-zCK0?0CZcsv-9i zs)^;)bb7JMkOp?158*R6iw@E}MRhu9NNO8eL<_9j1W+4vKhZ~FJ4lxv2fK&k*l%^Vq}n%!rasoE!Rl%_XZ<<)pnVgb{Ud`6z1JgtcQeoj z`Qu#c7CBh^ogsPGgm`oQCX%#^M`mlca}Q2RW3yHB@i`&R?;@@w7%D5F+l9HaC4!zNP zo`3tT7fyFB#K*rnL40>AT(`Uc=l5PC^G_SXx+CB5nse8 zCUF!$&-U%d*?n7_H+d?Q~W|1@VP?JVfF?xQ`ERtlnq z^4PzUh1e;%kvpg-+`;%=rQr#cBq&;srl|hHA*OH1@lhdmM)mO`UIpxqh-MEKzOEf? zxy{{4`HJ_9yy8c;eMH*_gYaJaqo^n;kEvk`aqGqo=1#+&gKU`ZIsNpiWtD}#3tra%zM78Um8j9HRq;EHe%J>X%OPO5*(`$ z1Ppt@QDZNjK3|6PPFa;MW@p|8o!gKMa@|1Wst`UWP>31>5&JcXd=-4Oh45R8Rh z6-Rt7;lF+MI9E7xC{B;!qMRMbrn#2%LbI5lcV9mBi9U@kSImMG@i}n5e>Y|xjw8)& z(!|r(M!3^_O}!%Q$Z$a~={OvYMOTMYYuhF4hsynI(DE$y-V`G+x_us>x|<5t4yNER zJ&$^y^M$ywc)IRQ7S$}(McmoNSTkiE>8;o1?@^j0P+${j(p5P@yZ>rzwWNgL%TbW) zdlgQ*h0~pzE@CA$In?!X78;eW#f_`K;pltw=w!V~g7=m>^r*0}jxue5MbD0rbf-zQ zZFe_*e{mWfN_$N4fdyldH4XDJLjj>#v>q{t@(8;eqzJdlYKGZ8bjT^te zmMDqI(`!O+hHvB7v2~hWY{ZFPH1?q;bo|JMk)Bn|txOlB=J|-v9ULMR5)5^Fbe)VF z9VNqGQbDhBAGCZpLjHM7raV&_GV-qv8yMu1;#Ntx6Cy+ADCg5LxgOkYdI-NtPJ+OY z79#cfASoJVaK_^poHo@DmMnS-FRQblD=HDm7P-+u=f$+yYy~dUm_aXP#ZkquKSbtc zDLw#0;3YMk9MU_9XXeF{%(@zKY+V6*_T($48grYQaj_Y1(L9Wp_f073+F82Vx|G-@ z3$rs4q(C7g7JoJ$L@koV+#R1=aH(My$SFR)aX@PmtLKqD$BKIu3Xy>dMY`5-bs`l|c_$ID~a7iHC9S8w3G+@VXf3+{lSb|bgTCWq*qMI`EIA$1HX!CqHupksYJx9?Ff4C)@@ ze(b~edcc0VxLb|x++j*3o(S-Y5ns-4>L}`*QjJoZ^>AurNo|y08Fm^CgRz~q@Mm`x z&i;L)_TG60M$Ly&x6yM>;iD+MWOavZ9vL8_BDLu0SZD6BdpooEWGV{DZ9u&tkD#jO z3K9L9N49sB(8TJEaAUZQ->R?++xA%FA6$!v-Kcy{{jt8;nPpO{MNS+&Ch26*fBBgF1)E6w|^ z%fGAL0_11}Y@1jOL1r9Yv6Y7dcCFx6rnKWomE*{0<3r^8=qYI|=!2QEQlNjh9wu&@ z#=0)v$;RhILeXC(+NQINN^E;XhJEH@lhtPb<^M$gr~V9_*~x19{e|g0V(b$oS$aR{ zIsb(7RML=mjd}^+6FZZ==*{0kB5rIzwvKevHt{z?cuWdOdX|dbyp9sAyskzT-oFYL zJ{OP&VKPEa_B?Q&?0{V=Q*o=0EcV=IN$!63k+!a}n zp*GxxEj-p=Q3+eP+(xxYqPXyaB9ydJ6fE3hSf<&c_0LA|XE95pr*sadJ3Qm7PxWCQ zBYB9m5Z0`^d!w^)SgWXFb2&`eKLwO9Yn_$RFuG3|o4VZ%W^&U`-RAmQ0zkp|J`*BW3Dr))Y zi^J!gBBHXfD8aRk_%q7jT&e&)CxyHsiBvfHBLnUwzd)y|<+!iGL0sQpG<^CfjReyV zlT5WJ$lTyWaw6KXYn~UkWknU~xpkQgB6YMZV*(Dlzm(28`<&}j7{J;qqhVgaHN5iS zE|@>!LVSr8mhYcUuFP)XRy5nPI*GBc!f_@{*RrL-74e`ckKjs-7he7J59%2>hfSTz zaK?sjoR!N$5KSV?p1OI!kz-(UbPl;1-v|r4tH5DGByC$Hm7IyC4@;NZhj=-+mh;8m{lzUoQxJ*SHE z|1QOrKNY~J%3OFACB;TwY=g8AMCB~PIZ?St{-~Ta+Vu1++dfO5sy57~!FiuhtD+2k zKDQBPjG0ebc<<=uOCqRp>syko6@kn-TjDoPNX__dgCjp1kj+kCq4BRbEHv}L6MtRh z@FO#@Oo}1?1IDzyARj*npUxGGErqjg$uRGXJgn2Vpzk#(kzP3-BJPP|)lMPTC&H-q zbha|Qwpa;5p%h)Z(uO81Ukz3PwYVb6hkiMiLM?`jxt>UE$n^h$QJp4Se_4W0|mU~8^9c4ZtI(;*M#?^9Nx)Y#s@RjLa2>7-d|Gz^>+l|XEBylo+wKPbSt5$ z%Mjn#p8{VeMB+4~h2)rn9QF|Umb^$@0IO_`*}^z)f!{4dy18HvXlu)}_EXRBUrZXt zCwE=Oha5xMKiQJ_ig-UnOZ~;`GA3}E8j@_{zmwpQ`wRDr-Xf2q!*S}OPW~#FZ@jG! z45_hMJ+V3Z0h@TevglrlHzw#4Q`s6XvyE+`Faar(6Rgcxc z_WwWp+3$K38+Cp>8#>RCoxb%YjLdBW74KKHeiu4C2zT*vlT?) z_cI91PrzqZ*YIa$v_Q=F<)E{F5$h|)LkIt=;ojp9aEp!~4hwcVtgP;B96CmpO;~#zT;?d@AvYPcXZLjYqj(MVP(?iB zNI5vIn!?We@`7`GGYeE2@8XLyPLm1x`(Sg0I@>T;1#8KlVrMGdU~ArAg(-{k2pc$y z<#|s+%)G~pdEjJ7{`dxe_I!pdGU`C;(LA>D+!`oQQ-@=_O88|1c`k zT?n~%fi?M?ft2Q1LXbfV%xU{gJUV%T_I6QLX5C~6rw-iJ+A0_gy$t4F1~}9AF0_co zK#1lt>S4A4l(lWypN~9XUQQru`0o&V^lAw-wDD2Mk{GOPFazTE&V&08R3XWKK91i{ zn0ZaLaAa{0j4|oOt|cn$ok(+5#Cp zG5xTmBc3`MZey1%b7m#JKE@^&7r-w0SioZU@$IZ}a5qdAJ|za?6{gbcTNiJ3=H+U5 z)*pi>Ox{Gqm;`vVv4nlqi&!ze4!r5ODK^W{;%kmr0$x+c)T>>_dFP*jLAovUnF~Gk zgxt~L4<-C7>Pxs@<36<3=n(vReuRybjbTL+jbS9W874J~u#vsG5OrrO;Jq3ACWTFG z_`wX&o-&Du?|O!R&^Em6xdbbpZ~)bJ8sNdkXZ+H@64HX!!UpxJ;Ap~wNq6l~?B6mx zri=$!C6kH5eKnF7w*)R+YKJZHDez&fHvN;9PQMPr%PcZmqAS6&U9 zW4^#b6))DVN{lUcQ(&D|o`lSFKD5kO0nVVx_BYR<8irr#=^Z>QQZNSmFYJa}LXD5w zzOih7Lo&?$9145dY52(m?g~3ljUxXU61G3Te7jXKJ3CP2mVuk{?q%2{O3BkK!eMZ*+;g` z*P{t?rBr)OiQvJ)G4x9C8~VftQR}=z)OuJrQVngVm!C@uh7~rEgf=r&zhMhq@@YOy z`PPpEK_&)wkjHM0>ix*q?grm8Y)7;asZDJ4ngeOj{GT(FngO^lEoD z`T0fYjbf6?@8!cV=0L%Q*qD;<)L z#AEjDqLsZrsQc}`)E8rV^FasK*|m$#ZR{f(Wi;8L$&&Ps{TqBMPneD6d70LjHPiCm zS}K&{QEkapTwO{v{r{+X^Jprg_x)colSqYx5Ge|Yv-f@~6d{x(DwQ;tpk%gP|5 zxdDe?yMk1kJ#hJ}@$_b2FmZZX59TxVV6XNh;z%^$#S|U7;7JD^K2#1a-iz^p1GDkg z$tN(9d`8{Zm*APpit&nNSGdD&V{U}0HggqojbLo?a>|;7(F8dq&hN$tm>N+-A4f6V zVHxooTbJJAzK37t@YgfYc-;Z2Pl%Gj54>hG+F5Yz{C@6S#2oJQ$Q8I!&`-2g;_$ZS z*LYgk4YosYgNp8v;)Lfu;<`6pL~mo0;BwGCh_8MP6;Ed45#xOF{OLco+b9)A4xgvF z^FCtfOf@<$x{_{`afjx}E_kU`N>!5zVE^|LdZF(gYqQXkjxtGQHTEM|vTX<68Xp3A zy8_^Z{!wBX;DOWr)DjP$5YlT@3eR&2xj~zA@atkHmfd?3@=triEz4Li>C>m1evT%K z=Y$fG@h@Q6UKU>bt)Vv6y0qb5E@Dk@(w|9jIB@oBY%uFL_NlxL8yCxii_vxPmJA_x zxT{qCC<33;VXT2*CVRHslV12?%-Q`c5B9T-&vtE7!#y52bI zzf_d_J3AlGRd?h9+$k7ltKffK*U3m%BN>(IiX9h*;~2wmQlYL3l1Cl6F`8m@MEfGU zd+;Z9x|aiOMiaOdWzQiZa~Rx|0;!(MBFMA$BDKfO;lfT}OXKFzj@oo~S&uk$g4`RBi9(4)=@Y;uS9ar55OiE0zcTg8)5>?=lOLPy~gsgJ0@R{?okzDydkMe*FO zOdLMwPa^XiVaC&9w8SL?-b)t|;rS&iSAJnZ^o&KkdJP4%YH-!JzPk2kAHgYy;A;X9W6tXl(~ z_vG1)i2<}x(jR7S7|j?*WRX-uir0T$CXk=p!*}&e0pr*t_J`a<_Pmr4>EZKDnvc?v z^A82MYsB+szKx(C+s^{oHVK!`v?teRS5h6LV^liG8rMFYNj4u}L{?i^;yYe)#6a%_ z-qLaj{}H&5WtR=GqvxP-@)`w_SfPy-I@Z%kE8JP{*?s8CTpe5{uEMBqG$P9U43&VQ z*tF)`tV^kiplMY(G5^Z%&{tbxCqCu1+EbK-RG8oeD$e9hi7H9?co@#kG{dDgb5PD> zGu%5cnQ@~b%-{4`%+@I)aQ03deY(Ve9JJ+YWb);SlR%t~|LjS&Z*N5nnQ;PDzW~xB z=S&|ht4CvfWAMd1QLKEZ35|`9!zDAf2bfqf~LN^e))U~h)k;BwE^OfYNCeEBkov9Z37|73gN56z}j)@Kackzho< zUcD`!-d~Bs%^tA3)z0DL=mE_+bc>Fpo~Q1^Tk+eH9zK_?N49@ag5?)V&{gMMwDD#c z9HP#oqf-x0FR6lBVKGX4Iv*u4!*CQ=bbwU9@ z2>Oll|Ak?*>;STq-3{l?9~0KqUS};BM4DUE@ugH5bmih+W;no?NlLgtS_DVo{`gX~ z)$xg7ztE?Gg-#6qBQ|lH|;ki+J>^Q+Vr- z4ndb=ui1_h)&KFqk_~4!aMw!@0qg&dTGU^ng_~T^cjq8ntY!;ajbrhKlsh!6Fow9x z*kFgAG}eU)0(dP!@_6pQg5G*M*jbKhMg`%N`56TE94FuDWaf{DG&AXSB#Ak2lZ^Uy z7GKeL3@5tlX{zQ09DZglj=x=lr_JEAqN7LQFT3W_{f?)I?6N{4^J5_ioLfd*4Xuc> z<9LEv9+B>>DNrIdf|Cv;lizQF%(Zz!L>ms0XQguFO9=0E{{Lp5|GekAh;kY;zEk{= zXGhK%!kg=xsO+Yf&~saw8^1skq$cn>W&9tgHE5(qt;GZdFJDmU7xC`zAo= z{fWmEIgZJWfhL0~ykTcPjDH(Mrq{cW>GN9H0H?EDR(Am`-n6;=OmGn_e(eF4Up2{@ zRYByOmLYjB-h!h8cOp*89~-^3Wv?7P#P<1ornbsoX}yUkl=ZyEGiK#uec?np;nylu zq40rPe4osIUO1i4Yv!@PmA}K6ZQD4diPBtD!VsPCB_90y`~+L9fQ>$oLVr!30owBw zX_xOlDm+(DYG+2mw8>g@&d?Dy)?ps_?Qn$KPj6ALtsm$f>r_5FR|(!NW$X%NcUFWu z2NsLm;Bl`85wYG*@BgSq&80u-k<(XkoM#FS;CJkXqifg)7lMdd(MfXdyE7DSP2slI zDsvlGv_O_>4SAPJWr%* zM3VdoQYGtp&q2Q5bGrN1eUJ&aQcXyA4Vv98r*&-t2gjV&_Voz$_A|K%M> zpUH!Qwi>Jz`wdp*OyR=2gJDBm7_!_n8p6s=;F@eXC;I0Umzpz%oBihjxw&ZpiQxM< zydQ6X&32w(6#j!v;`3tXbmowxMi=HhNa82FC)@I>DA(sw1vi!2aC)pK>ynkuh7PNP z(~Ku+zYFoV`3(J!FgoKJWvY24}hAYHH<(2C}!!BR5dN(jX&D()ngsuDi0I2f@Z2Urwb=G zT|>sdf-#!%0RKI#MAkHpNNmYcI!XQ{Sr~7EPiu=qU{D4&e0Z5;rpJ*Kk_Zke zv$)?+9bndbLeO&!(rvPzCahZw`(=02rnM=oInTn9F*pwyg4fXOT!y1Ih;m*AXCPnJ z0}A|YX~kSoDk}esb`KQ68s9EFv<+zWB@diAD;W}b)VbcR9bgHvSMc}*rq*$PIt-Uh!hUqRkw z1Fg!NgAY7hi|3x}`M>#h@6n89;%_pdq{Ea|x#9SSy!NwAl(8tr#M+8Mc41;Tp5OaE zPpPDpr8j74w}l3Lr2+y zBFQJmcpt3X!dkj~g*PtD)yM67wCLB87-p`D+T0WFWy~^7E8-huL6hcdvzl|)@Ojx7 zq$23Yv3~>TQyo!yF}Q@r#l6C~WjnsIct81lV~9>UB}TRn`QSDjOJAKs-0_}L;grX^ zq-aqoDXeiOdT9=LI+do7N3@hQ8P={M$qD~IF zrojB`^v2V_%!Vv9g=l?`rEj0JWS7!AlDqZ*xhJYgUW(_Cq+OfHHe*TV+vqORKfj73 zm7GToH}r7a#tgEsF#}1j>Bp80cEoGrJ(9m9nVg#Qgk6_tomP7$!sQm90}2M!?>cbs)FW*DHh`2yJiwa0y{NfflimJeGJU$h ziB6FXgw&{V8gla!QtB8DkBZ*F3i6vO^ZI+}fy=m9Y!-Lv#Z>Om`P<;Wt%}Z2KFN@iTaK%N-l|O~;cf+0W?S^(uJU z=NVjZ)=n5c8^b89KFg*iY^Mz$b;!lFX1H1M63MDDWmbttLeXa(2nmeGo8Gkv6`z#j zwV7rJlDERO{uQLhzLM1NnVVNjbJ1fjS@QefWQbyw;RmmUiI0z?qxY`^(Vd-G-zWrs z-949#_@|=pQZvW}@n6J5)ROt~s}Z-}TucbRb98nc#^dT)h_gD#J#la3dA8%R+lmBu z*)>QBai_fmR=Qzt-2*> zGsU9n7!+8@%OJxls*y75}|0_d$N)<@? zVmYGlPk}S+oX6do^Axr9?gW(+V;SzLK2tuW03Qes#Xo0$$2a7@5t}pT82_+MOh=Fz z^ank`huQ+zac`tJBfmca6Xzmg-Vo3HX*Xj$YvPE8h%<>ctHdF@7UOi0Xi^`XP2cC) zkPjilf+cpNm`}^&8J+k^|(}^D?LAGC$%)0%CDu{i+DU!nf?SKAz z{}OGhVn3xG#rK*jVCUqMDCF=-i)xT zk6xm%?1ZqRqmz}|rcVN{__E!XM$x|-GAKwqh3#{E$J!@+5=!d+r@zIX+d8{~riUmJ z!BtQ5k2dr0cJXw!L{L-KsxAfFekZ}Ed8#np*BP?XGGVOx1fikMEVSx!GFm;H1hxn3 z1o|40$nd)rYqP1H%{`b$pPZC~fZZ3-JDu%lr_Og&I^{in91@Hd-jEdj)SiGw|DK8O zY`g;tTQlja>1AkDZzxw5wi;>oY(U~7(~!n5aWQ{aVEX|~8?7wID?!(Vi^?HzS zz}*AA4zobrx+Un<`Ut@k*-cpX-DWhy*HmC5dI7bMsIxcxmlB6FRd9bAfhD)q+1HsH z(2Ui!sB!zS`4?GZR51Gjj`-Dw8|B2=`m(ubXiX|Q=QA44M%K`k^Ss%!Zv&CV@kiz7 zl25VL`mJzVvJ_e_Q6#?dE*?|49($He6BMniMja()*lMyFdh~Y{4);65Mz2YvQoanQ zGHNX3D4e4EoXv2{iDYy~QxgTL{u3U3cblgF)Pvxp_juf$9z16E8}qRR_evvf9TpxE z^T1A=s^IQ~NY>|akig%6HulO8#b$FIu~KP1of6(elRnl_W_TA~|9B_PEsMsoi^>Gr zF%ys*nn)dNj?fAxA8Zx?sG09^+PpbLXq~nk4$esvJeQ9)9n*1I*!W-`-Z6g&*^Mkf z>Te&SN~cr8x&zk*pXRaDOS%b-dDDz1JanZ6XB*hov2p16`rqvO*edh3qw4JHFOLKg zx0I;MGnQ^SX~TEkrejs-ZPZLj6}<@;#ibiK_CeZH{N;@$cGx1t7JYQ(u%04!NqPj0 zo{%Z*bWIn2%%6$US7~6)l2D<2y*j>qcPv_|u@cY!ah>&4-d64$u3BF7#sR6bV(?_o zDdEtP@#HBjMe@h{1?ye)*(;dEhOMj^F`OU#<6r|>X)u#Fi*@e%9h81_o}dX-&$pfSxPPIO9^J&R zJCxW>lJD^YcvQagyEXQ`ugp$)Y>7AOjeT`SVk0CSu`rwUaQ$S;uBPedn zf=e^Mveq>kurtt_td&<|sy5$4EkC5l<^TzjUj3Q|oynqYisuFPt|ytdd?s=Eczxhr z%FTJnl^ zP@WyI=Ob;hcfi?|?E=#*epaum!wu^eAl9Xo{9a{5{z=LblYQH8pzeDJStCWu$HkLt zIGH%LMdQ0kqnQ*hjy&p`OZo$+;i3Fu)>YmeqKn<2evBP?E+&gL2ij<}r3KCX+=6u- zc^$Rn5viXOLXPKJQbVg48skupss-_|xd!2{w@*-;`${+;-J+c$duU?L7bFq82_1TN znwX26XH3Jym=~|?$dO69+?7@5VDJ4ebRa{Th+GcGmreVyb$bTBAAE`G{dTUWw|t1^%%M`-o=H>4wOITbBi0j57X=txHq{UqYbc)D(7JlB0Xi@S;tQfHXN(Nd%(rO`)M+9$sKZV}(RYdpaN|Cf-hCJE13mca5tc}#& zboj0}U3yd&r>$&b#a~Yne9;}xR46U7TgbM)@mFRah{xp3!?8{`d*fhE=|FuZx9Fu#n~8wT^szLZbH zW^XO<#)W%e-W=dd#d3xgzL0F(XvU~ zAYE1kBog(x#6DZN^L-rrdpVlBC{+YjJdbO~xK}uB-Z3P`@6YGj7t`<7aRT<m2wQ9-uXZCn3?G%nRJJaFM$`hbmGK1NTBp9dF3dG)V zF6SO}6)c5j;2V1%YY&VuSJdb!FD&#Ta@l)v&(V80cXTnX3KtMJwLyGvc_DLUFpU{4 zx`2euc?F%$7TD=-txdcMC^Q=ue-r)G`OsQh%MqHJrpBnk49ol~cfK(K|YEk0TkD zyg;MY`eThv8^IJ6k`p@1$a(p#ASU~lZc^dN8@w-V$?#|Mtm2heaoaDN`m%(CU$G|l z429(QD?X#{+DqU4wxAuh-`Kpq45Iv83QukdF2Art6uh&RLD-Ai^uyx?fR+a1jeo@O z9lL3?{_Yy$<@SNKvn-)o`Rx6k2OVgzM2t8th=<`5rPOrnB0_(M(3a#jI&r!T7x!6} zzFZay`tk>9=a^RES4J1|x;p5YQMTl6b1=OkqC-CVc9R3|KGSi|BBa2;8+VN@B31*v z=0SVr(ni+~Ds|*C85ds&*Ia+G8HVO$X~k*s@rs@>t?m<@=X-^uS@)982Un?X`!L<| zE1C@-wTt}KvWB*z*;L|d4RTagClXqs#9ybB2&4neo!&iR_iisnhKJtMe3A;Pt1NlO zK_aoV4I!uHB(QYqJs4b80@KU*Goa}m@ff;HNB{1p8%J%x*$z+H`r*YOswD?Cd+vkL zzxiB$uBAY+$%52oTaxC5r)a|Cw={*@Av~-wolI%BC0$3l(ao<-G$huTTYXlIE6Ytr z(q&6YOX^8de}60!`X`C#>J0LX&voRv>I3?=eJTl}cZqAoCsMz&nLR2RN@@+onKut+ zlj6It>9>;w*mt;!O&VWS9`*eIs?ib?+7`^EcJ7g|^tL=$v=-5eXKu0`K|$1~poA1J zi6^z40>(Fb0kP|nq+{Pl)0$O^aOhwg?p^23yqc0h5||dqHm(DM>ajrhEQw3~N1K$(MJXax!xg~73r4Rb?C>C_HcEFYm>)_ce9=843l2LoM z0M;pY5^RzQ>eruP-+k@yJ+~OzUEa`_JCyM9Jw`MmM*(WfG{90x1=kFE5EV%U96CJ} zHk5n=8NXGuNhO;c*b8KV#Wzx%6GDs}?qFGWIbzrUhdv+FhWelO5M~{PIq!P9oo7D= zjax;lUmAel5q-|3VNCgX+ex@{nS(H=HkK+x&8ERGrm`suI_OdRY_BJt;;tamRMsRckyvosFJnINLB0Gd>te=dHf)Gh!OWZs~6 zm1N+}`;+j$*@u4@iC=XL^j6+u*U3(R1367-w@)LT(D$6)O&^7C_ayQi9iu_A=sf+c zYzC>~ z%K2?jBy6{S#?MTaWuRd0;IVE1(nnP}l_lTwM}Sbp{ko=di<>yoC6ad2 z@OLpL_fy?r@$_Qc_xUGloGnQceKYCVo(iF$V=S3u5e@#`$sngZiZtA5MH4smvj;5V zNM!9j!ELW9>K7b=R$0HNbc+Fw`TUfWdmf<8PpgC2DMxp$mZ|SLl$G@%U>cH{536|Z9G@WnhjKB2#M z3&99#6x=h=qK3Nrka5!$cK%HzHcoOD(avuZK8Z`_TJ}_Poy-MJ>4G6l^DIP$>}-&@ zN`)`;zL`^xbXMio0A6u;38`R~gW|3mu;flH4wb(_`rFm%aNI|#YPy7dW^)v_mG7nS z_$rQmdyq6n@S4bM6Kr1TLmUhz^Ze}^db3lP_1F}{sugVzeoC%opWBb;`oA9KHHIZL zKxYBDf2)!xTyw)BA)=(gWibiO5D^Sk44{(-UkK-4m4%Oahk!A@ib@jK;68hne&9V2 zn-{L2naT#lYuQm;|IwJPmG?lK&R%8RcmJWQXO$9NsWC)xbr0zaxI$LU3Haar>%a5Q zim!YY&mxAMvH2G4oHv83GVOx*8@|IfQw6w_xE`V|SAr5X(VZ|g6`TA3D174 zLXK+A`2EIkxIZ`sY7UkOf3#HM*KebtNn3`?z5r0|Fb-A-^8_{^I%$^ z5)`Qo16wV`s}I*wi2BD?EDPn9CZFPdXdLHUKT6TE%sEikFUpB}Il!N?pR`~~23XT6 z^kugxjd*X0@>^SRaMnfQeC;MZp<+&*Jc1yB=LLtoa00&@rVww>2EbSkmq8vgpvfn>3sLIu4YcwDs@Dys^GzvUG~CuuWd9J~Z6HqN8{C$GZn zo^<^0mpDHYnp52_LHT}X8`3p4h;VN`1O`9N;FkM3vhm+;xMAjnmk*2J35WY=$|YZD za6SnEC-za*_lW}i_fb^;&k=fL`!W=S2622t1aljy5eps;U;6TM?Rin9*2#nG8csA9e(3=}M)$@ePoqvQruf9xH$ zU)E2a>=0!he2S$RCMW3uyExq9B4k^2x8SY*Yr*MtKe26I%~%)fGLqUNAgB2Np7oi5 z*7aF*{_IxL6RgN=zGRIzo|%bX*WH6ti<42b=0Q-4SPkChBXmu}X;Kh9nzm^_#B(0l zQaufxnbLaWKi|WHe&nE|S(Bvdi6rcV3pVq4K(|Ls zg_6BB*tS52IOj4L{TYJ!$86}@@&x)@c{J?vET!Qyq+#l@c!7N9AH3)-iznzx!?(Uz z>WI`yV*QA4Y7T**MeFexyI36f)SP^{#SDzv)MD5^+lmfd{iT8LCcxF zwIfXC1Ow=uJ3u_rc9GApgRIn=M(kem-bPtUP5Vxgr72EiN9AiA`syzU>NFvBbr;B% z=F9l&Ys46on~{SL#^DdKr6i{3q`-AU5SCI{L!%E(r}t!wQCM0Z&U?Nd8kj2fQG$Tk zd0!s{F4ou}wMJ%&0T3$t|lpv?wo? zgpcKWt)VQQNuCQw!Fp@z?+;dx{U{u!mpb&GpJ>^pvbcRK}{d?s(O_avQK zxJ&q$u_L!8 z=O2vruE5cI4-n^n&QQ@Zn1de-qTV zQ`(j8MDmX#I2hVZ-_Q)!#_|t4HWKJ`#sPgWHS)BOKvBM|4uj8YI1IWBPih3Or$8PthV$DiZVZhR<;25?B zuW3DsG}EJSo=rD8@>qm>b+ZH%)@P8DX=A8(j~JP{a2gqJ+y=c1{=wQ3GqS!t8t9-R zzB2o(;E$Fv9eu+exCldr`!~#YN}NWgpLv@<-m6bW78#PD=>TPQXtv%p5JZ>Up# z6E)7i2*u5!WUta~3V&D9TTQF@4!Y6A@N*n-{W_1NPl^`&T0b6Q@6Jbxk3|{LLx0GY zIX6k|n59teSV;D=vJ7@9VEoe)7^}8$G*d8#h%4#iraMynPG1B1aXe+;UPo_BU!h~W zJ^9&9o8Pu}y@7_04TeZ_<+9Ffpb@B?Xk6Xq0^=#($ zw!I~{B`sNaV2$?F#^6asjlz<-*M)zLPRq8{mJGea>iAG z6nI`DX_h--V8~5iz0DoZFgr<)EV6{iczqgFT)|4z{GjcxGuUnm2Buhrz>$C$$ciw4 zy^ovO4`mzK#}>QUiic~!{Q8sf$!0`rca23Ki@0*(j6wc9|8MI z2Vv1!MW{TfPG{>Jfy=6{IO@ADsST*XcInS4_|*!|tw^D$ODjzOL!y`lK@AFUc^1rln+2embb*ozcb z&cv^tmazekv(Uc_e&pf%F(hQ&0Burq1eLoJ@t#w1IOww+tVxk1Ge2k$t!`;pnTtU# z^8v3ZtJ2WSGbHcfT5x}4Lxhq+C}2ng#dmiJ&pZys{whXLcIN{4G%SYBm8;0;GbPye z+*R@^VH)wj)}rT4i)F>=ewT4XG7nRi z*iyPhVTh(#twCltw5dpZCM5e!;41IbLE?<-)aA@|RCvCv%-$ z@@ob@*4D$$6`x2Sw%;e?%%9_-J^Bpt;eW5M2rri~fu+*PL?I>@y^|Lw=e2onVBqWK3&DIhuJ^c|s zHjKgDp?lc}vlf%~VM!YAsYW@8+jQLR1Nhr+2c8LfS(tsrgU{z4gTqn7#4CCkypa2c zhHnLum|dYvdjP{oY+cRFNM^}_H7kh0*J-5cb^wvNE+Y(!)ngtemIDsapv_k&LA2o% zTwu_Q_h{FXx#x9Z-ibn@qBx-l@OnuXjQ!BczG%2a8od(8SC6Np-gg()D>9%lAODir$r~v<<2`=Za*$X# z*izLh9dg9K=>PZ+CvQq3BTth_voRujRJ-u8uzYmdfUrAP`jAN1DGcsVCpmTRuprhR z*Ua2S9Jw}P)!0uAHO64QHcwttdO>tu>xuZ7`>^I_J5{MZ$IAUKCb%V(I-egy9@O@+ z+j}bLhg0*(1KG8Libr2r@1LIRzbFfwe73qg)jtOLbj@SNh**>5J%xN9iY7D6bHo$j zEIvDN7JhSOwD8+>RS3d?wEu7+(oe}}SMmJXw?lGFXRZr2(3n8vpQtmseIm@G#o+?U z>L8pO`iC_Xb_oH ziiE#uT}=`>pdCkKOyY>smDQx!N10hz_mF5^I0bkN#e;{^$d*6BL|ZlqwOVUn!IlBa z?a(4Rns>03pB~A7<-zlD`vrRS8;Kq7Nt!h^hOS9nhbEZ{anWI(6Y{AYNi_@#9gSiH zS0^oJ4{B?(JFb5sk1y}RVU<_eqk(qp+EM$E&vQwfTPM%hfA}O6466{!ok>EY^c2B^ zs|*&a^CGiTZE;Whobm&iIn3?*9J9UiJ1GnMA~bU_WM3-@Xu;$p>QXSB<~$pPwiK0< zyXW}M08s&rEWLwM9vhNHNm>lgpTgh8G;tuG;q#grBUB2!!*hu?vsdPw#o5Q7vASQy z1S7{gaeKQv*(5cSw$8pFED67hzunZpxzE)|k5e?hmHJw^_K`NZ_rst0t7gQAF83rZ z5}LT7FAD#5Z7BUS9~V^bx{Shb1Q z)3>8#R0c?$yP!*RCysRdgMWE#z>-%^6IgeLOv?HpJoD5Sk4t+>eiRq8?aLa-@B0(6 zb^ipsG|G|<>^)7YD)PuD>_lGd&c_u)3t&%5DKQ%nXWILmc>hQx9a-$ae!u+7e2H8X zpHgWM{_U5h1G!IyU*clvF&h=4XCfd^pu&yo@AuPaN+FhJ3f(&a965tgbM z#G}{SGAF0~GCOdGA#U5|FwYG_c%RM&YTY%K3SZ12`*anl_wJ>Fm7jf(ucA66`#fb; zdw${{Tep$o`VV+r@&rr=`Q6x-iS&0@0JFhTf;m5H5!te5G})FGOph;rCTRTofGphk zhE&+hrQ0JXuyxsWC{sh4)u|{zm**AZm248qOLxUtOWaXYq8nbQdWm$edqwVCDf&P9 z`?d0DX!y`fcF-qy=-fj*F)<8ti6S7G;)kBC55gfem++5Olj!737m`(e0AB9u248fC zs*i4FEr%|GL9{x!FZ_dx4jrQ>MndTF7E{0zqsXpb)5w^RR(y7^D0jp77yeUcO8WnL z!Ef_cHu`25aoO2`yXRH1X~J>zsPid0%y%zJyeq-qTf&i{Vlir#yGPTGHzJ61Bjxi3 ziJ4!7P=9D9-7vX9xQX8fKU0#%Ex9?gd(L{?rF|Wm9$lu(_cc+c^`WfBS8=kq*Nq++ z#j{_3l<`_W&jEEE1Cy#Fh}w5sIOczz=ahz{u`_;w!t6Qhr7IV(>_G|QYaIg4<$M+~ zU5b;@e97u;zNJXi=U(L=E>*bd8`Md zPJd6stmWXioILbL^1k{n()4n9BQ@q(S4m?M;h4uE5+45+{i-+t&{BEC3DzKKeX5qMOW~%e894WUV=3XH_!{0cF^x16$Q#IQ=#g2D4usM zg6-~q&-V#Ep%Xq#hkP?mm^Z*PCf@S-+vtg8ZG9c7N!~+#o#F7VlF9IPsFp4h6wvEV zeK14rI6ShV&)gTc-aMnOI+D2m3FXd!DJ?CIG=9okHxOD6<|~Mb9zbc zH{LY0fyQsMBqxt6a3)I|al2vy?Mx4VNiS!?+;vazv@Azr*KAE69JeK0H>R}@g9LLF z6WH3fSv0et9!-jn0><{Dd5uQ2*`fEZ&G&hU!oKY#bm@y;9R2zau1MX1918;q*!T6aQE^e6!<4#x-MU~UYakea7s0+l5of>-C{z;+Y9V)#E);xf(}IyY(ULGtakk z7v(N=%pwKXb>I+x4_+&XU%caA^?m%!+ z?u8|1UQvsfYgDZH0J2sn!FDdG0z>8=e#UfGB)UAr|jv!cwO&_NXl@ycuw@eC@wnG&WO!%XJl>RR@qsM$_ zYYeQ%hT8MkYrap>bdqmYIC2e19K6r2+pwYRbBqCTUUh{nJ6=haioazIBTusHd>qKx z4^gDSx=?+UK%DOkg1Zy=zR6AO5|zW~Fn6mw&HfeY{#K8g z?40SXJ)Ugy!DROCs&@9+--z-n6IdpA{{!a0TYF~TkPb20bB1hvdWwYA@T?`<2>iyN zo%%SgWUfl-Gby(hVbdG()WLWeHF+l_Ruu>6O`7 zt~!d99UBOY>lZSvPnNviY4w zWcWvUxp`qbYtg1a57~4GoTLx2?K`~aUU_3O@uwU5v#6e?)J|abZ;@m2LruxX$qLNn zUl+;7Wz(3fsH?d2wu&0#0JHYsD4;X zP{uR8UQ{b_JM2~vzYznxD_)G4#2CTykV^twT*+G|2$Liu6qcssq&g(G;%AEoQCE#7k#7oRJo3})1z1hd{2$?1bnFp` zzfhs0-)w~1+87jZ|0ya_`huUb=Dtk}wK?@q1DSk%_H6nwiG=mWG5mUk%qoR0jQ#uqaD`w*Pjv~od2@q7F21}!7K(ir@JI=Pj z{bS=y=j#^iJY)zW^@_MBiD9fnj!@^T^vO1@K`fpZi7pH0fyuIN*pgcVgIfmB;!Yh- zUA_ZaD!&j<#U954s=Ty^Xu|^EhYz&2!ZLYI*!FhILLCW;1s_Dw5N^Q=E$4 zSjynbDKz>3P<^{4oU)Zb!Z^SF5D7yHXVh@g*$(Et*mo3}w;ueJe!yCnr%-xD0O-r} zfUhVTe^J;=?nf^r+`es4x!e{VY^-HHzfVZav&*>EdIgn}89){8md9V(%Fx1wKxW3K z7UufmnGkyQ7Pai-FG^)i2{pIuH&)ZHz@HSBBk4Sgg~-=_B0ZbB!}dV>A<{+b)Zx3? zYwXNA3Dn^n)iUQKvi)fcZifshrCoC%fbRkv91e%OC4r=vip8}YeVmnV3Z1=XaKNY% z`<*vn`Drgvr>_tkeUyb9q|!ji>ePSlUt#}w23=fbMxyQ8sgEJy^xIN%G@$nr>R#W4 zx$3ico%d(*y#Mj=WDRES7r&;q53WZ6`;(cw-787YsR?{?wj3!edq76V^=SFI8_0$o zolNFL9$b7i2rpPZC@ayVQ?gR=Gn?Cx`9g%2nDYi#ioYd;s_7*8_;qNVWaykUO}MdQ zH>LZq5T<F&=ZCva_Eym?b!jV&a#R84rGtE%oihZ=xL6cN@{9m`pYM$;Xe8ZlPho&I~inh_ES_W$ti@zu-)|OZz@5B% zfxL2V!|&pv@$Rqv*vG8_ZC#^JGxuER60w(9Dp>?w6gba(3rs+Me-9FW*Kl-P=qneZ zjc5<4N*LV6gB^uH$0?RUb>v5qIh`MQy+4Or8rxYfY&u=!B?3P;(>&h!WL|IhYMxYK zDXEP%A^e%oA@uMUB9!D$iw-ORcQ$eScF!%wEvpga;^dLNYc6c&c7Wi0E*5p10hJy% zkij2Ato6PFPewb!-5(LqY3G6z#-bRx<&~)QQ5BNZOVHM@C zMAUNx`41i^1Bn*IPuGK}x>tb8*^|h?g@?mRijk_13T%3x4cYDUNNYngHhR-VPV%3{ zY5yeAr`~2 zGUU2`F4>`EXzs6Nhyzu&(I4yIpvhiy)&pU{^z>ZDS9{-LbjO^$pYa;G>}^F~D%4@q zHW`K^^%C@^-jQ@lkZc&$CQiI>g7Do`MST=>{OV-1pUV^S zOoG>%c#qs{u0!pkm0-Rr47_(05wqjx=|YzMRvuf+K?^jn$9hCWes-~WQ5oh}YV?2i zU;ndzx8_)pQ-iDvZ{)j65>-fDC!I(?5l4^PHJSPiiTJwn z8cy-O3N(2-8^2%2Wumm7k(tQ?c$s54caaj?srk~6#q!qhB$HDZzPrEiakpoj>r1P# zEbDwYbyt~u9Wo@UJ1!9z*#j#QXVa6-KzxT3A+fL;Db%{6drP%=BQdjiL({fUIiYK* z;F3xd6#W?PuDne>eYT(eUDx%PYa^#5stcc*x(s#_bY1lD&lhZ zeEl)C7dUo8h2=bD5((?W;2Zj!$v0gF<7g4lNLa-TejcFy?8i9is~_GNWx?1l?IzP| zCJA*jj`#KGPR9M4Bs6??qM5{WVm+Hn_8&Jz!E0x7rV=l3qYdsLH*^isw#q<>SQuy@ zUCZGe$Y!i!OYoxK8=3qAOA(Pa;@mm@-kfnRs2^H2iJIkPt|e9Cg+fS`=dx%?0TwQCXQP3%-np~;fXjJs z=I;QFZzJT>y4Ac8Cqrh+Pmp|Ro=@~H8PLa7ec=7Q1&A+8kF@g>vdaDf`6g_HGTBazr!p@JA_8IL}UBB zkD#(;5gd@Pfn3gW?lBhiJj>e~M+|D9W1FqWHhIJooKs3>xiq?VeHsd` zYyc-+8~7+1f%>K*iQwU-B*3tTY>uxaBc?mZltcjx_xl1C5}F< z$NuyZ@=Ny&##8Y`kL?r=KH$M_hfao?cAFWMT!vVP3#VQ?1_g*%kODSq;^Xa16mOj6 z2C9XTYqOh}##8AW|Ao_0CT4pB+I^^DgB>1AGDA-j*W)*e-sE`_rx<4; z&cAGABv6B*4cWY6axX#hH;9&hHnXSj9&Y~;jY76z%++Fd9AcEwm3YvZ`U^cC&BR zhA1N9EkN0X@zWibzN67}CGMpSs_6L66FAy~y5a;7pU0Nf6&%f4&>eb=QzHvkW7t_ zqPyN3i2RqIIPFY1&S)k~i(e$?@%t3!LibZ@`%NEg(>zMvN}J%-JwjxS@^QlZ9zeeD zZ^9vgh4{+J37me!6v|!l(QUgWq{87QL(7!deVsY^V5vl& z6zCAgm<7nH&=c)#yh%EIt}(A1)3N^*EzHTG$wkzK4l-x);e2B-KTpBq7fSd`k~bcW z@gj|5Q`lpAD)N-ALT4T|A)SN4+%mNKVzya>bWK+^7To#gzT$yZ2>&9hJS0vFlKo0JyTKTj_y{!#P8OK;Du$1}m$L9luRNkPvPqC!W{3uzd8bq$Y2qYs@wP=gY zG0N-gI+D}A4+-9EB1c!Pm?)27YHCM8%{B+(t4_eSbSZe~X2@LM zR!H^8I76h08m#$y6kZ&(g04U#kbn6QGW!cadaeS!{Ao2fUi^qJ8uVe4k$9vZyc;er zz5&MWRsbv4a%zw7q+}iEpqCa!Xo+AM>hj4)o68N**Dy=kxBWVVZMTDri{6}F`vHRe zUr-wi`Cu-~EtnJM2CL;WP))rop4Kt|KadQaf0@leu{@`~(tRK>5{70u=<=X6o$PVd zM1AMSkz4vSvZie#S$Qy&S$f@XZXkY5GtKLn-+Q9K-49Eq6-TiGwOCXkcX-rIqWlmwp(_=^t!cF=GAVbSt|vq zZOQ1FMGq2bErw^8cW_*d`{3l){dB#l2Hg-ilYW2e2O79}1L->TayXBQp(DkY&F((} zTj^(b%fS{LD6a*z35n2tt_Wmz>(etvhavjoR{EBZ8k85aXJ7?kI@Vp1IlWSnT=jd5 z_$(IFq1Tn^QnHMmUMxnUw#C4e8g;th^g0}=uoO$3yoN8tOyN!WHmoE49p4|z*FlN6yRIYZsplB}ifr1u)0rOB>cD63TYzCy9Pyj~ z2nM~D(Rqe!PWint%s>7Z;af{d!Bu||HuQRvT1MS92ISfHee;3(S@ z4WBOz>rUQhJ@+T*OdB4UZ;;1@zaD}|{7Tq6bPvYXxX{%K)o9m7e(rBqTR2}(OGT?| zqo)=xQK?uGo>?)>EtM_gv}6Y|obGDWwBC{tIm1V5&-sal{CB_+I}5UR_%c=9B*L69 zJ42t%D4-Xm5!ms`n$)N+fI}Mpu&L-m%K6U`*m^M$`0Tc^UMy`GSNV;PSm&UhAGN{I zb{+_OuEXwC8==SH4my-p1*_HV=}BWt+Hy}Sgq3fhD@MFwxyS*~wDKa3e_t_IlU%T8 z<4vrhd!T;(g;w-(Nh7l*yb2VqE@v4D7WC%Xt7-oj1u`99WB8XllJdvpbnfn}bpCQl zI>Vg<>rDON@7PW3wY!p})-EFZo>k-EiE4_^^$5s)5(Y`3R}dtRVQOnWH0ZU1sqsdd zUoM!=3rdED1;g;}mN+f6s1u!S%EKatAHi2a5b;M>BDd}TkoLUo%*%=ktiNG|8`(XJ zG`bkT$CfEH-OLZf%~Rmi%|fVNYDD+NUV$L;kM*|7(qRQv;1w~78-n)oB(prxB6JGB zTy&IV+&W7F$|Y!>I4Anr{hzR*;si|2UQfrD9b)+xEW1HE4OAXk0?$X~|J=VUX12q} z7s?RZ!%tuSa2cz3NY>ls4kO9ZEPSrN5RY6^CYi1F_|o4!ctcGL6L~0)%D8tP`KL-k zwaH{j}{%#mvnnSfdSV(ef zN~nasS`@qAomwF8im)QfYS*1dnymCleZoKdS4{;uZkov~aqOcP(wk`e_fyF5>t$s3 z?KIlwH9{ee^Qby%H;Fm64FVU-!8vPP=-XM3cB*=V7TEx&Z>vMe&J51{ng?*pd>1jh zKbJh9oXHtw4s-R+7(||t0-^86(L(=Q=!j?}3fV{Ft}lF)L-rjEM>Ei^xFWQtO^))= zi6x7x<`a*{;^bt0BXV@U1uDGDENmQS4Df#8rXkOkf(wzB( za!D{JkDrQ>Xnsrt9_%1DD)mX&#y(0pA_{B`m$2;iRiHQa6}5LU*x^$Swec;R!FAOF z70Mr0?Wjj%qmKANSPU$?s6)==8UvMIN)~>KN864Yurug>99>e0Ex#$TyW>r0+s;Yk z`STIny0jWw_=_Mu9XoFBmid%_YcGypd=AY^2_ZF2hNM*j!TZ{qkUDD?boq8+RFuT- z&^MDGzMo(`;SJ=ZbuvvA!YJe4V)Srr6Z%f_A*i?lO*lG2i!F^dzdS}=XlIzSN!`p$ z%}&JFZwI@^#W1EJhGmo=GMd{Bph(pdn(n-ud#sY`@b5vTkIyrD3)M-_>p1SW z1#ieglYDkgu>>)CtP{omEOUPLZ4x&=PF1QtK=zJ9xPoP4zkU1}nS`=?8?GgfCsT^Q zKd&O~S9IZM?R=={lH!?aHIsbLuh8g8Bf(Tz7#|mcspa3HZ_=W5*>3KfzBQ@^D3F zHvWuPcr*%^5vHnI+1c%@R#eT?1B%}shh;}l!-bZNUPT~dJzfl&L+go4f(>NkSE2!P z27l=DBn94YkV&E(lC?+8M(uLUf5tfonHUD8y=rKK@VKlLRR!|)MVE|tsz@D zSCWe`^Y0Y&%9_HG+(@)w=6kH`t&e(6UZ!40TcT^J8boD*Eb6OV0A)M#xKg?i;Qlq1 zPRcyPnHK(#X|1#8inQtynMfyYxrZVVvR42J`^TVocQMT=&|(_9K2f{Z| z_>TM;5K1@&v(Fv{t>kTVLC{QEX7zn^YBx@L9NtK@5p#h4sJ0LjW34Xry4!W=3 z3R=3k^cU$G2e8RYYu zJ?k|^!6k!1bg)JoN)uNwx5nP%Ap14A--;jR_}f!2i*q?6i)hp#@rJ?+uAw!?tRJAn zo8s(O#vW!2rO$j~{=+RcQ%>eO9hr{vKd~8vqF(Hnj_4O7w)6&t9D29zYEJaC>7;Aq zFEjm`4LpBYKo9Nngml*VajedeK5*29UbIVyUX*vl1f2JV_|iShgoH&C=>>xI4l$~* z=1~I98~==s{gtI33m<_7iCs{=Cl1@*$S1>DJ)E0A6j9U9<8av`0a|?63?7$6BXc_atlNrPve>D6ta6D zHCkUD=nV}G%*JC$_^!)Z5M;l<%RYQCpO_FKBA+xVi{u?J(~{kBm;lY0XoaKZe6Tl_ zpSF9OMi)A4q~|O-2|G-?p(j(CzEoU~OC=D<=$T5fqw#R%S-!PWe50C#p z=bueWJmIlOHAuC+hNJ(+!0CxP9roZJ2;@3~czX@jy_Jfqqq>pF%znmxp9?kI+KLW5 zW_O#P-oe9D9!P}E$3LF^8-k}hQa^9hAvN&{s`I5X%eYxWVi(Ks!l`W>W4CxFYgra~ z5Y2~n@4o^JHgsoQ~2Dm0TnpN(WW4Q5`MMfk4j;%;>KLGf7K~u z?LV8OUA;n7vYwN)&4zS&Tm@$;{T4HJ{};ac(~*9dJBt=`zDzyZ*M*V;o53rr7Or_d zWWILWQ(s~Sap<&Xh@6qYzB<03Z8D9n)SN{p?~kJQ$^_%TN*>^$@Dfg$&8I8fWYB@{ zV#G#WnB;lQBK+HD;C`tW$V-OJmc7rwsVf9X^^1R$L!AvV*>?|S5`tRGbJ3SCw$S-- zKCP;?nog0~gw5vkP#Jh z%c`V6cgzIVo=oDn{)$9jqx}f4IT|;ue2rDiUo)KwiOjR9YEJjcV_2vEFgk}0BR%)c zC^sVl?H2DuduQ1}_klQw4A}_5Ih}BO%{S!wJrh~I5yu9j-9)`K4NCQccuJ|V^rr8E z^oqi6kP=8E*W$uSdeu+#v!DztI_86@ZYRuQKRnMntqJ-)ft_M^p*^1!P^ah|y6Ep* zWPkn=skMwF9hpJU_v-~nY4XwKQk!vMMl6o|;6=vzd$=ZFonTeX4`ltq2v0%;I;}N? zPe3gScGu@TUuR3qlXub<2M1w|@Jk4D6oO>ua;mDLgFM~M4^30u5bOCGb*(%NG9d?1 z&DmnqFWZF<9o+{{VIlC$@}aMHF?>-Hgzt0tUlfSPasXJkCL&KEx`Bcz>^yo8hO19vxa)0l~`c z5b0F~GNEz6cOe>bhFehQ!UQsW;5p+hG5x>upa0o^wAL$wVl-5DfiShY% zq$+claOIQ9lS6X6fCCJ!HjSjzyJV5+(z$5Q4{>CRl0 zSf0cYUeeI{Qk-L6JOEM!$*44af}?kI8Q7MTLF278boz7)5;7pfI9)r5#SS-v4Wj6; zYn#By-yHsUUS`(EU1mZS&Sc-~da%Kx8QD9?Q(`@<;MLSuh%7F^;=7zl@ZVBOV$2K{ ztcWKvbJ=~-f&#qh+12eP*;9LjdGX4v+0N7 z`kuv1kBcPiG2Kgqq-4>3HNYR+--A>tKbFFIXztPT@U$Tn;=I!TCu^$f-j3u_hS93P zQ#i;~3A{s`$+=z*mT#!3v)+eaRVK?M4i@$`Iw`zi{OOAD|tU z@V4e5HF{N)cKy`_55Fv;TU_;dZ9W&Nr#tMB_gqVAwZc^pIZ^{B&cwjKkV3N7XfD_K ztt&j(>;=NN*JDA4SR@#7zJ5!(9DJxTh4c;P=q^%XTdNW*;J*4NuCwS?IAv&iDXVF(nz1HC6}sF?yTl-Sxc=FXX$!PvF} z-a1C3QUg)c+YnA(q?KWPH9o3BYcB{qd5pA$9q6b^W7b`zf~*o{xVIAfD9Pk>X2t%i zOxETIur69g-)(Nf$tqqX{^S_*_wsDILr8Inep>z4@8OE1baet0Y>{ zh6)>WhOi($`j0~x_U%20j8rUf%WV#oqW21W|9OX#`4^$st~JQq`#4&gRfUeyEqLR{ zZhW^f7OjZlr@vZ%L&ATx;kf=w+{kyCDV46Ln#D_Db8HUicx*#&OlQD6r+tj;!oRM}X{st^HB|}a7JQ)X{s?bg@R!PLy;lF469|t!6oRS!4k(@T z8}Y@if%qJAv9bT)_&(hBfLQFwq@~@nCm8@IlYI735g(;h)>kd z!>n_1ZX$$_+(czlCgk>>P;j|XORB!*5G@CDe8|g&`6JYiKF#t)9-%&5IW`^@DA+;$ zNozxB^(l6LrvsmExUroNCrXoZw(jMFqa^OdO}s;)l9W#OL7M!INLlAO+Q@e4znMFb z3(M#8Zb!TGO2)6FtIerI-Kv?oxiuIsX+OY4a&kP?fagSOD2c%G(FWS3w%J|&?x>+)f7#_bxwC|dc+w5L0PX9HjoI8Z)XwRU9Pb}jK4gY}= zp&XWFw-GDlhrpA646HpE1+R^MGrxuZFg=YY(DyPOv{Jer`*L=XxvB~zzh0AYzwD^< z6Vt#;&pV*}1q^38JNH^-wFeBUpP^`}QBL)-aCZMQKuOVOLD*~)SwL zZu%I?Y;%c}*Bw|Wn@^|P?xdYcPU02YmDpYXI})kGpy>E%$ew=o-~Q?UoPV+%(1DjPwf z`%i+z+_Pv#^iH&3^EJ+ey<0eD8_sd>r6THS?g^$lID>iJd7qlOd=E$!slbV$IL_R` z4$81J8FH9!=##K9$_T6HEKiE0&PMCt48;@N=w#YTm8t+|tw|?NybZlx@$tnKS z$lbr{EqckSJd^fJM=BS?QU6|j#PaHxpuh64cKtEr;`j=+8MtBV#zojHssut5Oz6-R zM`>~;imF@a4=06ogRqMaqdn|~GH-1`F>TJUgx8BQCwl7zAK7BHtgpBu{yMTSut7F6 zRnhAw=8RRO&M3#r#yPJ1E5PEoMC~GLF<#Oey;JlI`Qxd_eUVYEf%uA}3P64gGQM zW<>VXp^uRhjH>N>CO@BjO`PwH+`R^*JiG=y9o>%9?(PK}buN_?;>is3Rx=OszcP{D zUm1h8T*Q~7h7v{BqaE{Ws0{`V9R4t0^I;RhJgn2h1xj&X6E6rWn(m`|4RfaYr92(s zxE)?h>*Gq7c2V4ra_Wav0d;$kD0IEKM^)#{Vm6z}5tIA3sp_bMSjk`r8G4pcA@97g ziEs~KcRQdZub>ZWGwH3nU%-6jqfnRxT=P$C)ozd<$8M}f zx87S5%ZgwUvDJuZ*zlpf6A{ee4^h~ja}VFo;NY~SEa&OYN+xZCE7KgM&&b-YA*Tz3 z$XF3Sp51eqL}``+D$JR}Dy$U$h*@r^06x7LMKUl^ zqCq{p#4-e{s;RkeSHRU5tBH_?F%`xYXKYe#F?xwVG52~9TvymaPNxSG#MNeeG)^*^ z%5TwcfoI4!OqwvU>Nqs+F8VYQj8-L8a0EN$*m;aP>|2>dxyvGW*!qIw`n?&g`jHMZ zE*EopOujJB$V;>!_ycj-bB_q8UnCFriQ*eu=Tmv2Pr>edEMC*8gm_B&n7UWOD8}s| z*@<=avBx%0kDFJM3-6st)y!>hBOBi77tavX4Cj!`DFvwb?ay2@T82(9j;CC>x}jySuhC;(?1 zHsK9xM)4(WKGYmk$h?wbv#Jjtu$RSrTqN@b6@JNL%9nYgtH&DaQykvo?t9W`SD+vX zU*CqN&+vz(9->%OteTwoWllyGUSu}ePQ!=SIpBlp>JZ<50&338$0<&ZWX^avM)8*T ziW(nup5vgJkh#>=!>ahUyABq$4#k@VG;!f3-Bm@qYv4lz1Ne1}H0+w2f>)QtP>s1I zjH<*p9A{JyQRA&(Eq|DFB^vMqd4EayB##X75YeBngEwU>p*fpWkinTY^u=8YpSKId z_J1N^NazwimgYwIexD=R(t*tSmG$T_PNJS(*Cz{EAK>SYmh>UP2WXML0Ql`V34Ei1 zM4>U9*!ceYKhGcIhSgARBLJH1@gQ^kJc<0NLu%$E;8G)bCaW?IqIzeO-f=eX79yJu`{}O`8|HO_&ZKk z$YFv&hFW}9oqVcIMoaFTLvCzNRHSekIa?Qlh-wsCfo;jL7X&Xi-p3ecCE(w(?JyEM z08QVX;y>>ExcBii(jwQ-d1k0bat6OKXY4hg@5omCE@w7rJ7G+_cj!agZ!ubvdy>?9 zWWgHer$~BdD(n&DSvL(ajs5YIoB_MDzHtgwiPhnJ_w87#TdMxO zp%hu=mjP}Kb7&cx2Hf*F700bEA~i~>_!Z$I0fPu=4avqYTyDeTq3y8q&Ri1eCV}U* zi;)&#C6x2b6r@B$L1vd2#bx)EnRaq$U!^p(9*jg8TeC@K&Szq!^N|GqKrk&jlF%bZ z2rv%dP^QQ^DVYX)c8id^O7>*_LQ`Tx2*}1#%*8Er^~KInXl6?@>W+)VDFL>GGZlr8 zJ1QACbK$&fD0&m5OLmrP=~K zaC8wr;iTWjZD~93oM}C?e)oe|mCP=PLKS)miNBf+I65g(L%!5E0D66TBrVOKNUv|SjdHoKwg z?-g05bP*^|okC;tjAPPtmuL(WAfFgk$I3x^=w&~FUR`&G%u7D+NxCB=L1 zehN%PSw2^xEwS@sZG?8L4^`KX)Q)u$8%RR^_8UM);w4qu*M;ZSX)yauXP_@zHo~>* zT5Kot2@+dR$mgYsq~T~Q<_74~V!5v1(^CYmGR~s1tZXXmZY0q(mtbdZtdr4fHat3u0bX1oB5;bUn{g-z_H@AQY%zwi2*y169R)5s26_Y+4YMOv&m z2J}<>X~mffK<4i_6FEX~{Hcw^T2h_#-qfQXSx3;-yRLJcG+IGXKMy$9PZOQ5Z{h0f zTDnf!4SSg`g{jk9Af-+mb$(q1HzDJ{{jdMrzrwcPc!D47KE>3nOd|Gyh}R+-LAL(X z#zlvcuw?uz=9=1FysK$BUYx~xH62)4l1L*RqgQTQufRqXCBxeU! zli|}ZaXHU`M5sMO+S*CP@oNhK*_X|za6jszsX z!Fft|8S7#*bo@p&zOnQb*?!_KmdHe$@pr>i=wK6bGqnL9Jnci8C8Npt26=*f0>C!g zgeX~YcmwV{<|W1UJ>)f!mhVi+x$%pX=xFn9EU#eqw)*7cY%5;U^LLJFU|4wY% z?{Usv{74BVN%OqPQr;3_5&S^!AaWOz#J=HHJiX~1WC#Cqw9@MjZXZn|>pBA9yWt<) ze4Xvpfj?R&Ww+>dZ=?CutuJv$a4@OAvx1lTWErz4SOuF&Clc?fZsgy63iEmD5^aqh zM(o*HlxKAxk^@eo{>fG>EmuVLeYeEVTz2CAb=yhPy#i)E)*xEJ7nr8dFk-2Z4u^Wh zY0J58c*fmUv};clxvVQmr`8KHf>Vojr~)k1lweJ1~Z zPcLWtCtR$yCJHb1HzZqjv7T1%j-0Dlq96{Zl&3Kf$sW|JkR%+wB_Dm}$m4+GU+D2j z0w*Jtc&V)DAWaNNTi=)lZG>LtOBagO#M?7sdD%O280b6FQ>i=QBG%en~a)7|xyY0d`XeN36``cQ?x&6nc^ zw%lTl7$i{dWcLs?wLU6mZzMWh7sp(k)=6D3=cl|Mv>?lng`o9Sgm)-egiy9|%<9+~ zKNPItUHpzQ6hnHLDLdp{4!Aq)*_iN8QUgwYLJlD6UAo4*f zy6P`Ry4jgimg^aOJTjX6Z~o{1fByVVAwt*VcvpV>;kAvY(LcO))4#t2@B&Xh!99I) zq{+aEy!RTzv%HJ2`Jy{yh|OF@kX*9u#dRj2wH0sMc?2Q!8G>np5s8co+7*_d$Wkv9y#8ux%Ir(@K|0=h#@HyNg)l_ zDSXK<1P9~U*yutD!`mrJL=QY5pYKE>iDehbz7tjCgjXu@RdvDwBca5uD43Yo>vKA; z-$Ev5@8fpFpn0+z$T4XnLhi1|y||aD4Ax~r7kZG}=dH>2s~3pHI`(gDEg*2an8dA< z#q*ug$@teCaybvEIo@1qB!Nfn&detV~H;o#3_$jGBb zq&l>T^*2TG)LsEkNuY~!O63sNUOpGRR@c?*xx6cA^B;Sn@ZdBpe zOIG0K@4z`ZsK;n#_@HgiUSjl;{T|N#L;{AA$pY8k*y(9Kp0lfmad(r#PUW+C4e!*6 z_fmcG%36tMhc)qu*E2|m?~u9A)+khb?l+S!8->Gng<Sn#0wOA!d!|HMgG(Z zY`0{Hms{q{kyDSf5ROGvtkTXSuRTJ4VYYOP|ED}E+*5q~mUWfG??a+6|7WS-r5avWa z#Qw3*vBK*OO8InRntYJ(CYW)DK8r;L?>e~`H&$`}{$H_Rc@7j0G|=8JX*5-avdeL{9!mgYg~s;8Jcj?V+a|`3$w$XQ_xka z0;Lfz;rY`Dn>I^Lj8_yqpw$G#2Ond@kgl?Ss(A670WEk$7s87|)oz0qrOma^m%J=8~l&cX8+_ zlr`sxM$8LZy>ljeLP3LFSulv$e6qzGGwX5l`3)HOO@a3;Xu*x2S978E6?B#UGx|Kb z1)Y|T!DAB1=(@(8yO=6L=RQ8Y@2im12s3@;Z;}y`LwZdE0049uQU8}BxMfY;dJVGEUAf6aH)Ids> zGCMSSE7>u?(D|96_)6LypM4pGgS{ir>0tmpzQTdni8ARsAz$kghH^VAwZL$C4KttD z!dGujqyNZeVn4?dWyh%C@#Vi53BjvX=&_J3U3m&-KeWPep`px-b5;;~BErUI%1-7> z(J?q8En@3sN3nL^A?ypyTF}j%MLwG9kO$?5=m!5p0Th&NQ|nbnN<7U#E$S>3xxeGq zOGPnPh%zZ!Fd031qG-aiIW)Sjpi=SUQs(mh@uKVNr@_w+vmo(O5R}(3NDA&!?9~>Y(|?5rYsr@!;An_x<=CL`_mvZFcVI8K7ehJ zuOQ6pq7B>SPYkLgxzgmlqQ_2Y_)XG;y&JoV%_`Dn`Jj#P`b;*Q89%_ab%es`CCyy= zg`Y6v@Eo|2mxo<9ESRLVW2t=u{LlaV-x}o7JPK})R=`1@hq9~Ua?vd7ApH{?gHK)! zWm+VR z+K5g-7Cqn|$fQm>PBThRiEb`iL%O?HF@_3Ceuj>T5WtemK8$MFy?)HOEm0PT{T& zd7xiCAmVug87<6Rbkh)CAYbu&CN(gp+h^k{nStRhVCCBV zC=xQ$&U@@e@~ISW*=X=RE7bY@RfshwRH1j?7J8#;7<`CS;HErL#?xab;`bCWIelgU z`MG1Vu!CyC6+;GSe*HFTa`kEDlkY)fxMd>U-Ef_rZP*I=hc(!7TfQ-E%RW$>ktsy( z*h17vvmpu^w{YV8N!+5K+X6qugq-#aqmSYqLx!}0O~Jz%yvGXg>W0sx z#ZZjDceqXtC0=BuWGly* zU9j~I&A(kkdr=ldU)R9zaalyW>?sycmqecwr1c50=&v%sow*&uzUkWm_GL#vFM2Dk zQyPY?+wO==rwrzGhs?m?M~^U{W^G4hH;(BpIm_+IsG#%mLu{1dYq*O_UG(&kZ2Cw= zxCfnNM4v33gO8>%*dMTrF*w+YlT~UlFd!C*qY-+#exLw*gFd@zPS z*BKyJ72=8Y(7~WNRFQdPUP^~v%ZI5GG+Db>C%Ba{x0%BS1~c&|oSC5ERm|%2St6Oy zwonym3z=&v_inDRe_rr`e2|bOgM|5`V8&JQc)b#Q`s@SKrYr=W&Nm^;Ul(~8aFMjGv}?2~v$W~_2%;@7?4H2jj8-O+jU&1)H~kCufR(|od0Wgl_! zlBBUyPKouWgi@!q{*1NjG~yJpiA?O6&%GTLN0~wd#g+q;JM>PIke3(f$B$iPa&i^T zx!4S|UdX|ne^P|~CI`kH#oUg;8O+uhZ^?uDPu%R4olLP#FbOsPKwo>@rMDZ>NPzQi zl6mIdkWm9EPuGMua91h{mb;s?`E#Qog@CVL5u$W_J?@GKkms>O%Bs0 zLWb@Y5;S5ux#;apR(~*n6{%mD$V+F4hnyb$H^dnNk3JEbHs&!d0*g96ct2Uv{fOzx zxJVV+)VLAH7vh@CS!i7if?rk=bD!-dx~_7BeXIc*v;<^t{({|>7DV&kBpBqN#OS|h z0jG{mNX!-Tn-)lcbD%VY1f-GWkptx8$9d55JB_TlD17gijuMA?xn#jHNz4!)>qTSz zXtOWNJPy?&I%;Wz#=jzL*^&Smsj_U_=EF4GOA4+$Y=Dh= zD`?&fcT(MZn!6j*OUvvB=$b{^H0AU-Ccpn6cPC7NQoK!iFI13(Yck-YGlHJpt;TsM zZeohg4}-ny+PV1(OTZ|1B!s*PhvH>^P~he-c=(R7f1OI%VjousG7(rhJ$Fe!fFJiD zt%K2eo+9?Hl^`{XelkHTEa?c%O+@8w9q3g>3OTTGP_WVrqURQpDwz}-uG}s3XAwZI z3)ObFWD2mdPW(G{|dOzB${W_}S*3J0ljGtai#6gm!JYql7(bunYuId5j5kCYzu z(tAOj7b}75+6eHxah6;va-FN^KH8>UA_aFby|K=|VE+rlM z$Jmi0GY&$xyB0rOGFRYs|0F|WcaVl%!}xPcC2;+Y!L)I-6juBUqQqnaWmoncG(A0_LqQF#(3zEt`M!hs7$XWzbCmp-w6JVqlUJp z;n&YGBo_sal;3)$VBiV8;QpPtxAZus3c6jDn*{s&)Len}aD`L0sv#GzI>3fD8E!

((F`YMpWH(`3}J+e+o z1~}z7qF+0KnZ!y%y5S)BsuBqL>N->{eF&-0w&vcMPorOknTYk>ui(@7TabOQfL?5s zrOPH>M$_0lT+mVoF;1Oa?t<;qv+g7{?Trpstny;HFvA&8nll9*mp5VD!a?}S{j#uA zRmXX|?$f$Q%i!(rE%5$YBlhf-<56xiijNK@WbG&BkHSLYzdBmnzdntcI9(;>NAgi8 zM<15>U51e{Q6y1mG|d0;hLGb!Vb1#)_+s)AI?P)bwJEh^_Re@LYZdn12YMl%v!fN3 z8Q|L~&F=YU%XTX-f!(o5;4SP{0#?j|Z1ta&@v8f9NK+bJ);5I{yCrZ(j4spi&Z)eXiy z*LI>{R+;SYHXG&3j8NJ}5)0Qap}PCVlId!XXjpJQdE-8XyLB~{Q~Xd2zJCp2(u%KC z{n2^|s}{k^FU^d8E*Q!?_Zsvh5Z#Q0`8~mYpUsvlG~;(tA)i?kFAY{gdoDevj@=?H0&Y@9A-hU((6E`^K(dpfmChAtdEhU7xUuh_O%}A%`HP%6T<7(RJSx4mUT8T@xKlf|5COR5F;rw4N z2H*Y#5IafG!`~dj!OsO>^6+6qfA%HLb%Yw^N4%rI=sZnNQc+w+qFR7Nt?(GBeWxB zI$l|&FK}S`$QX@waQJo-%2me$8vZ6G>r&x{%~6Q=I{?}3b%d>ZLKJ&;)Hp@R2im;u9O+KdFfDUO7eK>Fg|)S||xQ-1>uX zbAlA>pA^kaUND$FpH@oKH&VDKmV>=>qM##47BV(ZK+|PTWU&2gx=_fW80XAD{q_SO zKVcm?a{nqskJts_XS}IXk`cY*uY#Mq`)KvP)pW%E)zr5rjy~K_#Z7iUN=p?q>6YJ# z)Z+U;#wDSTy!sY{cHv@bFk&0SSmkihY8p5;`Vm>AVS}w_$Ks`>Zm6?(7&h#A#jN~X zONyp$0DlQXSmD2?GRNMG&X^xTs!z{E&6^j|ed-i+w^rkx69OluAh_LD=I1D7P`Yg! zy%n2Gj~3LEqdul6GeZWe4#$ABha&Xf*MK!!I>7l^kLXmU47T0g#YIeCO5RLQVZ?Jn zE5F@}rK=76QL){f@()&EM|>dEtW3fCp(VI?;xJ+EC?MO`8S&`_!&nEyYS5p22=rrl zvD1LJc$xKfGV_8N|4|T(Y#Pfo# zy^Ag<61buIAMSUT!)*?QcypvAXa6h(*ZB)vqlqC9z^#DqXR?_DbuZlNF3(>x*~oX@ z&p;8s3idzt1|xMH_CEWI`*i#}I^6lqSdebfoA6Ze^0#U@JxIulh*9|8{Po`+s4TIO zNOtXq`YU3(z%hz?veNvVEz+=VMsC^uHUz1OR24P9f@~62wS`6l6<3s80Q{C zKaD7W@gb%#aabA+-Q>jl8n_H4V~$dTfCy+Cn89V=suNA|`~`23S8c)huj20amSXHIrAZge3O7S)4(^DLNK8BRKb?|`N854xvoA9cy8=9XJM zq?OgH=)LO?Y4EEe`Z?=4ktW5&y1@fad>YP%8|1+Ljc#2yZ zG%{)j@S1THWb4#G;7E6Tz5F}c&FDnI=z{}?kHM0uy6jxFjnI^K7PLk*fstx9*RL`T zPk)w1sf|5YI4u*58}s1%`#>AlJQc3gppjPSn=nl`@@Q!1aIy9S!pQ_Lpd%l?6%SdQ z0!vI4={qY=tjVoLeWC}tDkh-uVhhmV*TRw1fk?2u*5hIG@ROuDArkLCvEyehJcQ$Bx^apH`E=r;vp6803fT*c z_~-5)vElPCy7%81x?L_E+^x4$DOI61bQBU!@ad~9D!_x!CS#42DcN*<8O`<{O3oiK z!M;Dc@VZ4fIxHs;{Q4rewU-cwls59g$VZ+_R|YDtdYOGojRvVu=k_|)|bXq zwqS*sz}OAsKTH%f&acIo$rVwrv&4 zJQxeUf*rP!xV_sIDn{1OcllxXVc1ogWVanl<5S4Wf=k?(3ZY*p-if1+RB+m(RQb2r zkvPzn!(93H2aRX*B9klqG-u2$ELmlark6O>zIF&hH=d-$TMwgaDbWy(+eYve&gQ(b&J-MRLyMoY-wFdlw1kZD zO*pOK7PD>HE~;~M2yF~X{Ez?e|C_&(oM+-SyJp<+F9X~k4km@)7H}i0ZQ=84CET|~ z8$)D&aY3)7V0!QxcyqI#ti3T4=F$-2w=9{AYN{pMdj>Ht&sy=D=B;Q|Cx*#wW{2snUubx;^u9d{L=*#13#Z8c5 zaFlLr90UX2X0ZRb8T9RJV%AJ3WFCgBq1$dhMZ=k{IDP0bH0xhSgJUkxb9W-R;VzD% zyb3C;!d0Pt-)(?8fqnL2H5~mM!5!{UqnULX)b4H~2^pD2V>2FcU)H<9N=F;mADD$w z8XS$;laKzf2XLy8RiHF^5PQ~5lPJn*i3`YnF0^_v9eMsfO$eF*ds@P1+NwAvPtOQM zwMxWgu?g;J4#I6;%5kG@H|hIUNBUN~Q8_;;FmtSd(=vr*l!XeLG~5K%%MXy)>QkgO zE)q8PJhpN6o=R(tW-{})M3L~NzTCnEH%LH{2lI}!a_-q0q~_}^w#PS-95)$>!FTP* z($IdgqUtZ4F**-nFK1Jm8Hx0B*CyB$u^T^1jpj?feL|hJdHBTa0yT)S2J?WM+=tgn zM0Pt5xt{sNeCj-Oo?k%cjg=wurn6K%`!vc&*MfURI>>7-Vh*jz=lnv{$uUhO2wPW= z@++RBUj77Dcf2%sxk|Gx5B`ywZWPPnDdKB+fzI$x#lzPQf!9${oHqgro z>as7v!CqhXU|bR#cIUBZZu$pdAAy+R^`$GeCzA%rijWg%; zJ@4R|<_>bE$d>+M^l+h!CYwBKA)6uZ$0{FO4$miEC6*yKpj}h&VEH)0s`Arpxb{;} zZqdaJqqfm)TVW97~vaw*1KXW4Q-zf2m=11**WPA)3%oqRBIs3qZzL zmK|v%_`@rmFe5dCIsI}pa0Ak8&hn+8>#dIZ8Sds(S+sm7t&1y@A9bzCtywpNmj+*}J2*32ihakY%BVkt^HG!ZjJV@w#<_&@*8 zfA`O7UUewjFGgpB1sFdxmRP$Oal4N-;(?ZwBXxKk7L)&7x=|^6xPfeiaER{H>pG&UYI?g>F9-q2ZqAECv~K$;;@ii ztb(siZgODJNbMf)N0pHm(Xam(8FKbLDe#&H(cLp)BCMsehbB-5cNwU>JQ?Ts{vkS- z5{P7y9F9+afZN74QUP{@7gq1*wDbM{?K{GNg4pGxS` z7q>~*);mnou@HPSwi$JsyF~J9K9E~4b*TLjZ5Ta20mj%*#QtFoWM)D)FIPT`AHCxX za(k+A_C8JUE1bssRjj7IDg}5cy^W0f5`a016N#PFYufl^1Y8~`2fKs`^5h3|;u@pR zKOFR?!X2IHpA8M7ed`^mdaNX`5*dQstF6#p=K<5w97X0`83sokv;_^Nj(9$PNx!Yy zO)qauLZ9T<_|V&iDhX^X?X)vC15Nf2v>*`PA7$8MH;YUe zj$+p5@?1u}B6p`g0w2$QgN^s1kr*AM$J{=1?nh+A2WuVa1HC6q*Lejf=#c~I$p<0X zJ(<}QCEO01rjYjPZ}@3jj>tmk3Eb5_2J?lO%+52zF+wku=3nilGVwrH*Gtl^j+Cw{ zwV)#RLo`(~3=EsIi0$#ksP{$S4~3e8M8bKpHGU;sd%RneBI$@h(Nl26kr=vJXEE}< zFL74?IDF=vh)>>Akno(tEnnD!+iI`lr%r4B+Yl!_Iapxt99u%h1eS<5?Nr9pl!u(o z{a{*nBb%nbS%#(;i>XHPHnPI?JEyd?o;#eNM)W5rW3{C+PY0djel75%At7NP@q0fR zbnGu?9w@{m3iVX3GLAeIuOttH)ad+WPPk!i3cXP>l57*S;|oXUk=t9J6X!1;Se`rz zN+k;5CH25ipE1ax(vudrvu!xJzvPNL*~@}qzX<{7)9O3zM!YEm^y_G(BwTGqPcgf=nzFS z{JV%S`t=H2+U8fHH78AA<0eH|Ba=)$V^nR1ta;4U-}E7iZcEZiUtvdF(o6;JGM-2i zyj3a^{N&YBv9HPnKj_{->NyeRAGzYac|)0V&E?F2Roe8vp>TK9UQZipvhe!S3LLL< zido;V2**|nJm9ORh{;QiPBO^FwA?tcv1S!h-etyoP|v0d=FFlB&OMyMVo4HrScLA| z6zK+?!6NbiQ&4-wsUrN z+o_p*GI8x0Mi13`QS0<;^m_LSd>Y@6L!y7v(K{-+tgxTd-qDDdE}B9r?Prp5OUX+0 z^DO74v=x4&hCHZ;Hgd8jk%2T4qqDZ?-B9~{?OFA3S98`CXqY48%i|;A*-s?#=lRRxo6c&zrT6N zndhl8Hi@y|A$*S)EERYoDQRAN_92Y;`->*}j3@G6N7&Q}93ZFmQp{}Ij-S4EVTqpL z*;}B)%{;Q7y3dz~IT3y3iZeZlYj+Owt2Kr(u#Kg1TNcnhBQ3h%%2@96yT8KX$e@Hy^Th=3&9!}ru*9Z*kqv+c$ zycb8O!b!O~IQ-WL)@tK67QGH5~Hr5_r*LJMV84*@QUIh zp?_56y~Y2nKmX}{{`-5^PzSlc$3WSGJotO}IafS&50~RS4EsDR=*y=@lwVs&nAWW% zJUw)r*xx zikCKHljUgfX7&R)u49VEj=5xPp%x6^cL}=XZg9V5`cj|#3`VP139}jnA9&LO`Xk(e zoPOCUx^T`PUGxuB9-2#uf_*ayxi~{2K`sG2fYt30Jg!j+{q%1m{*)6O9YhAe1}MS$$T!|y3Lv_%>5~9{v{8y z-$}rb8Dg?qxX~UP6H~eGi4WSJR_Ep_iiyHDj)uIi$6X(<(ja#+S3&QZpnh* zXv<>yJK-W}*ZfCxE;MioYkqP2jE)GNCsQ)aPZt{hNf7Bj{oK&fa{5W=)40mhl+Cpw zF-8)i>#xRh2hwElMbS&TnFK=LU0)h;%bae%7*74_oS4;}c}&6gxzIZ1B60t@iDu2# z!Mnf0nSl%Ixwk`hh&`ph3x8`Z#k5uAnc^v0sQi+CG)ZKhDUBtGUnbG@H+13Lb~os0 zZ4jU7tzi5tS29O?*VE`?MZ9=n4yYR)fN9?<$m^={5FC~RHhLRq$RJ(nviugPId#(; zqw;Wa;#ayXw^PWAKW@|TJ(--HBgf1zI4u6RJ&>8@(MBo_dIXP}BpUwtM?NU7XP#eH z$8_!qGf=w$cAba=k3Jtd$8QQb^ZE}hEIfd|O%dELfojmb*N596Hy)zbY7+g+dd$il z@#2!@V?kzlJov7Wr4g4Ok&o{R#MT}oVG$ZLjdFvS&~sLF#Au`+M3wm2h6Y0@1T>Eh?XqUNDg9#DTRR|+r_aq#=OzKJGpfvd2t53T$ z9Y_>7QVGG^bto*8T1;IAM-&K69yx@5?B~RD6eiFa7FtZ8rzP;l5lp~=IgG!{ClYb> z9+~`IlZ4eB1wXwa4ETPWs&)p#eMNx*7I6+V!XzN0MMY>T^s#Bn6R!QWP`fofAjbU= z*MGB|GTz#7`|coSxr2)6%vmSWxSi+n4n8D@FL^`g4@oWC^JnFr@O7* zNiSM(NEO0d{|G&%n3I3-1@H#r>q)jR+I&gPHD)H`71JBqsDz5_CX}Fea!5SUjYG= z<(ZRrl0a;H51w2b;QmbsBc9rqNaqw=&^TJc8Lh5jWTbDA#Nq)_t;;nokxOA-C5MW| z&yA^({d?v@qCA8hBjkB~C{gM&B4N|rnUZo#R(|hRs`$H6l<9Gqnf1KP#`%>7IoT!L zqfb%6O~vYXt2KeSeK3ctybwe;epaJ;Ez-ndRUP9K-OK$kOD3yd1cH3J0!)kgMOMqk zG5cyYh|=yV(dO;>MEjl4uh;G)&x7@-Zzuq_=o<0q`bn&d_M_*s8I0NZR7OTJ3LcEt zfkCF9;dM|g9B^=AXH7WGEV-r!BO-F>_r@!r*fX9Dr~~$Fig1s!ekniAMHPOAv`~wu z62Of=&Fom2Po_AnU6%9{67^u0if_F8Go6-ZQl3_1%-gzZ?Bsm^* zA15%eQfncVDF^q@y^Q;`V~|zn$tt?7ByMKv;C4Qa*ao(Vs601z$jiIc-h+0cV zbBaWwY6f#R-G{@?Gs%=k>b%mWxx7hJEQahnj_=PX;gj+v{L}S>Xnef_*IdonDBHVa z-a#pna6E`wuV-n_T0`>mk`T2$6Z&orZl7Xq(w+0g2HL*}89a!I9%9vIg_ z#&v!oeM*gV<+2Oh(~v&2l`w!LeKYpnuPdZ5KakuV7z3(5%D~uXG)dg41@|&O;NR9> zu!|nXWnhxei-a-63Rh?plFH6n-LFK(pa_oB~ zh?gz~i~9$ecg{+%_3;~~@1lsf_~o#AuN>JE)AHcnQik|U8xZRqILs{1nM8j*3l=v8 zj}pg}XHpNxY8qKQj<~1&U|PR)aWh9HG3KAoF^v!1>8U|+#PO<|2)9dcr{~#{h=>&M zbya2Mw)8)zsaR&YxYlM8ClZXLT1gJ1y{c!qzWGL%{z_R`kZ)nbfPdD zJ+83f750jq9V1A3Vk?NJ-z5SYk~@>s%N~0^pUp~}h>sr~VSg{lU?0!*hTFC*8SgU) z9^_sn4%&5GN4*Dl92|lf%N>~Iyf!%Zg)==jmI%9~Y-XQg7o#&M5*n8Hu-%J`;a1;S zSU5tGOwvx{cHG}jtR`g9;m|AIExHJQB!)2OwndPM(GTeNfJ{2(nJ<}FlEofeZAY_x zomurCcWkUH70K)TMw>^b1?&wMF*|j#^psh~4UA_S57VcxDbq}-W^DWuIfv@0RPdQ}BHF6o&hSWkW0gW7| zb6YMA=c4Ry6QwqF{Px%vy&u>Schl*Ny|XGQqAoIvak0(NA!b~#b{SoTdR7(H+ za3PzoU7!wItD&=10m2WRWeV?PaY+Ue@Y<}%W@G+bQEOZtu06|(x`AgROhmZGavu?- zO&V55>3_2CyQ&^iTir?UWrZZjj(9=dyJkR5??rr1Q5H$>};qn=U$2`OU82tMGtXC-;zawLD@L9{pyc=IBq}*ebwW1nT7^3EodW z5&!W~@HH}l9q*^kGQIEFiR>TA9c#|6#I{^SE-G-m|ln;#I>N6& zaSM@-n+OuBLjb;%kQ?{9;oh4bSbN?LhE@)yRYe3PsxMJb;r?cZ%URNHz%h+%6bX1_ z$l=SUqHz(U=|(GI7MC4O#d|k_jzc0`dC>*oHmU4|j$^>i84Xf1W^&tq4Wjpd>yjy| z60BSOSoYgi9X4&RJiDe{lg$nj$f(mNu+G1Vh@WK^=eZ+`{b(0Jh6$FiFZ)h`cr*c% z(^ll6!zS{5!#CO&QD6E0+y6-H-}|22Uv0*Ick&>UPMjxE!9!uTM=tJ7DkdE!hUD9q z!DQ6X0^l5P6Caz&g3ZhVtIMTuLXR{mtUQl5XLrDsI0I~Zc8C;=cK{#F?bIq0H=-J&hoxTL2G;-Jy|A4TK-m(^89kPYS$QZmMlI4T0C7_w7P zVynqSD!{kN8f{5dN#-?-$oozH2;7AqFPCDhNP!-C?ExcOFO#0P+OR{d2>p*NL4zl< zyi4RO>SU)0wpXvgk|A;+b2FAaKQxi1o?AeQYDQ3AeGXomw;!QOfT|Xf?B?FYau? zn3mmeNKg3RAAgR-zjeTxr?p&5(Qx+fjwEK)w!32^-EYzb!{HOQKQkk}Ddo z-Hp~ozI3N|8Of(sIx+= zm3_=F_1UcY4Fk6Gix#xzM~M3^>TD#+RA6q>Iq=LMfa|%XF#e7;8-KzFUZiY*U!TlUKZb$9>a)gKTIubpeL_K(Y=?4@d2N5(KuQG z_k6g3Uz#0YQl%+t{Hm0UnmL(aDld~Y)Sp*M?Z?MGujoTvOWfONkB6)5=>7+-Fxa<} z^VXke-5^^>uf^x0SX{l!5UkS%!#af` zVlUe*%D;Y+o(~+2O7Cmvq4bTUF3T3V_CYA&a~Y?vYr**k?$OGDqjd7W3taltK>B1- z30>3uotwXL5R9VnR428aSpD6>Hgu`8J~yAi*?X&Cx$PTJT_Fwi*}0U?TaPyje&F=E zGJ;3oEhkV0a8t@hdbLrXakv7cUm}p?e2=0}`>n9fQ~^^bbyJ0R=g8(Ux_p}d5WZb~ z8x?!kk>FAf{$_~^zHM0rI{R`MiRW*iZc7Lpdi#;sp7g~HGpyMkX8zx%l@wZExr zo&nnGN?~o132ga0oEO)eMe{Ye%-jqS4o%Icxf5crB=jWYeb|e!x9!;XO(tx~JPR<$ zYqq{r`krtR!C)Tv6w13-L9oVsnyqg{XUQty`lbRrCio~l<;JofTIT+z|B?8=Gnhn; zTnfd4R7@JxgOjH9;oB5N{$k%SepB{QbQvdu%dYJeZ;p<|dp8rPcL+zXcE~XeP5n%o ziHFG4Lz;)Lg#wFh2$miij|n5U_wGaJ*LB_%*AEDmB^*2+uT-T5crTvaCON=PHVRV?GbjoHU{yuBCAlW zYMxG7p7Ruz#&e_8-B7J4j7kNH$j>bj{ICVe7}a`$O1ep5=ZOXQGHL{tO6zlnpB@$; z-#>#s`ZN{KzLO-{gMTm`jVWY9={IVTu8R|kI)z-jT&8J!EY-*y%kOGxq7E>QDEz*H z!=9hV!V89!du2{TEQ;yq;tHB%ZX;yW4x=eU14aHdPQ+q`4?1ibil&`bBw)x9jA+;) z(yx-l?D5*z7B&R!KmVlqm(!{BM1*_@wn$va;5qHlOieH{NcXdPatYp?t!lC6xNqbM-S z4`EM`9=hj^by%)H>@F*>Ee1HlN85IXmZI>eg}mPnV(G}!)KghKK1@%w5}<@uficD z`g<=Ut*wHGyB~^`()XiA|8crL%z~RM8A|GW%IU<(YdO>1_qhDX-XWCEROzD`DoV({SU_)X7GZ>jz#|v#$~G^n;>!I73p-+6(44)4^6w^MdzBL| zzsjNGsxcU6rA7lo+;MsOZhp$*3Apfx5r4U|4-anpLYtzL@n!gCs!{h81MhvL+EU5X z<8>hB&Q`>{aqp?y?;%Wdhc>o%YGSzIFf4s=0_`R5Gt%}eP;bme+^1fKqLz0wc1{|( z+WJ{+etDnh!Sf7ikrzsX4_>FfbCmF@j2eF_pctu&3~HN?c;4+8 zKJj+PxKClCYtzdQ~sw9HnUsd#)S(oe@Ypy*kMG?EJlfA z1#isOkFlgvDi<;q&xC;y&Ef#X;h;HjJUM1-0q$2OgN3g$`}*rC48EexD~ILaksw!C z_;fITX?7Z(npOh4cAGMlE?V?jz!<2lzX0|lddV1_61aOh6J>5Gk#ol-iOmGz+a7NX zW!1*;=#Ge4SmQ(;w`XEvpe;I&&O^5asia}yJ#MgM74zxPXi^Xm$K3R4g;_QOc& z@YQr>+@&4J6zq5bu~%;~6;m|OXVFUdH}ow$TJ;zVJ8Z!xvXYaXID+^-O2gS2mB@!) z!luJ*T-~1}T+TnkvV*CRQe6p!;-MJa7EU$oHOV%6N0`-9fcx`(xxypW^tkCffu)tj zJdfW&_&hCs$@9qnL)Dvy)%1Sx-_6lnq`_2%2F>Hy_u8f;6`3k!jzmPWS!kjer3ekE zBvev5`(9fj$&dy^A{j!-n9LMN_ z7EQcF7LeElauD8phbU^)0%?&Uei>zSaA6HdzcCZ@ccQs_D^%G7>NRjpUj?t}{-y^L z8@cYg9<;fwosJqOgE|3=@$5<^euLB#{4q0Gc$e03A6#!@nXew-K0l9?yc#l(E&BtV1cCv8BPA1c3a>%$hXDFUXc}bE?MKN^aN~a*9+Y-6zLC>M+w!mLf7zG<%7_OmUw> zpC5{0j4%!)M=Yaq8Skmz4Rc(-Uz500_t7VNo>1}mXVg{d86)vA4jZ4`fP%~u`0Wde z-Bztcajh1_%aoHFqxVzGd3UMaMWhw-*{HDRBTZ;MNSe(T(dv_0e7M{Nv{0Kv))~*l zypfSi_G>vj9&n2491)X%PHPZ(S;K~=4DN)(VS2#$9dXukfs-ZIC<&6}t0nVMC)@-* zZI|$FLvCQ#+ZibDBV;^wX~E1T1u$gU8urrm`~S;d{8xV@JeSJzmd4rmO?(2pd)8su zyH;u_WR=J6TTgSJ9)|UEmEnAV5*XO2K*J782vyEwf=~C6mAmiZW(h~8N|>8u1$pCz zpGS$N`$saQMT(xDQAAfCPlk5Pf_KYCFulv|U{%^ecAfNn&S={PCg7C<@jv_)hW3^* zg*xAuA|`}<`LPi*Hmt(Zu>+ueeI$Er$RW(I+(hgRW)kO{X(UE-EFZP-HrCY5At^h3 zsasVSY5(wq*hblbw7)Vxnla>urx`Qn6Wbs;FM^EGe8gR-E8_lHO@c+Mf5AU%Rdz*8 zKHSs^z`sRJxS?zgAEjUbF^|U*hsUuOv!(=5fx-oujUP$8uQ7PkR)c7YBMWadNd6`* zP^hsc@8*&pWX?)jrxN`R} zeg5wpm+rTd=*yjlXo+4Jo1Ox9_7BBD^;w)##6B_t?t|?5xtqbcp_ry$`A$Av5_V^T-hk9zN#>~eYSv|p342V)BQ$yp zLQ#_xuXR0B$Q4GxwIgqd5+}m%XRK&Z#yPxFx{Vj0j6}X~JxuJbzz>BkIB@F_TJY2O zt9tvGKL#q?_JSoa`{WL0@DjoeU$|IwT-gm4PaVQm{?Y~YZ8oC#9oLDaLKxXP@f9pS zmkbY>D!h5@Emt<1fxqEH;iugw{;sqs96Zp$)wOMcyT;4dvWL4-=UO&MA9R6~_&M-q z^(*iXSWmA097_J{>*not1`oB{@GQHLb4)q|yrKns3Qr;vC1m)vtv9jO@G5+?|4x16 z1s7XrK6x!FpquyI!68NqAahQIutVwwKfC22Exi<$RbQtYx4+XLuDD`0pk0^}a;qgBz*TodWw1xo_e< z@I^6`Fta_$f$OJHrhO*AVL*@B=MUnq9||BN1n!@|{2?a4D-NcYbCAHk zhDCoX;D}@|6R>dveRo5NJ-_r346l~~RpmJ_YnB!J-N1@f(j1G66J2q0U>$9^Yz|X0 zUy@&Agbct^BN*xG53bdwY~)^LZb5D=Rlk0Kv<6>>zMRj{X?F`M@72NAy!9|MbqOAo z*+L#n_XQ6d_^KpdcGQld^t@!IAQJ~ z+>^YPk3)NUk(;*SyS@Lr#>!UDWsGb^4c$@^#tfF73C~EfIR5@AZo!sJE=R~Ix%;g` zD;HN#t1e(>Zb>C=qP^7URkFQI`pb)sBGt;EJz3Cv*Bv*#y+lt{ZxJ%u(=hq=UYuL< z6B9Q*#ny~zSpL+JZ8r%cK{sE}-S2-?J`0N=bHnD6(OPrp$-QHUXY)}m__G>F#}|^b zt8Q~D2Y@O&O@KL*&B)~D0&2X#j{Lnli1UB#$=*3D#a4XNW+(h|AR|rm$)h+CBOe;h z1Ye&=Vs@*r84FFZQ>K`+ZkR~1GmtpGAB+tdCOG#(9DJhIH}1EU))c!!qFqAh1Dm^76FA@4tl3^}Texz`-H zE9yRE8s|rkrkRm79xnTFbd?MJp#n+bI6>7bseU|kh zlxggQV6zrlF0fZ$-abryZF9)?Mbp^z)opO=(LM1Lqxs0{-X_~uL=&e~6GhvWy(i{^ zDePwBEg)Wgz|s4Z&N(Fx-s!I;2wG2d|CXtMvHIUWb>iF?9$Ak#=r}DEE?fI-m zrSmD1Rb#<|>oJOr8c}JJ49=VEPWCMrOlA}>r}Y!$aZdFp%spc!IA2n@q65E38IeNWrgoHf>$J_*9|d@9KQ}_y z)v?^21aGI$Ba+`b$;zt5tZ8%usCI6`rpIk`-=_&!J>3(FRgOU1=#}Kn$5)giwM72d zO*C?GA_vM_Mb|u1nMVg7iiVC;;)ZJv!U^Vs_}b2DyzlalP8&7{>bf(S$7Kt-T`MI- zX+Kh#rTfN^drI-lN0*00uWCGOUK@zk16M?|50=o-33<3&cNb?8!VuJ)3lgpInD{0V zqL!}V26o*c?fSgEjleNEeXkH_Y|Vk4MiX$X+db}|P_MUI`QgT@Ys8{x2YHvIS1Fe6 zq-L;zm@GTZ$SynvlY^AOK2*f*+~QHOa*(cAr!yOmZg0m&Xa6x%HqONzAN7fju>YIC zYc6hc^1>xj4>;L^6x{wZTHx5Prj}D|P{V2!q1!GqWB>d?w`~vch%U!~M0pNyV>g#KdVR9X@L< zb=)$YPFq?-5}$E&Cga5V4(|c6v_4&HwTWo+=jpWC6gU;61#@>##AWpZ^j>5$bAJc%?Fq%5)m3EnxP|2Yn$ygN?13@04?{v}hHOon5aXy6+yoC@aL0#_>FrZWG;m@KaQNL=n8YlErt2-?ekIHAMNq zUyPoRU6#&23!mzwVW205UI>0gyq2Hmj)qSmX7@vJQM@`$x(cX!c@$k~>;a4EB$C+~ z43&|k=u{?f-qk6NAGU|QnIgjvn|~79>y|)-8#KZ(^p-e96@w%3c3Q3I z`V0v+rm+#OZgi%V144dotqgy_)CDG9brrmL@x*cEV5%y{|!ME~I2o~nA{Vu&w zdp4Sm55Ee#V_euCjlWFm!XK0i65)!j9PW%!4vphaQk(W)bf=vymtj7dwiKRaR6W*{ z(cvPhmWC_5M^p3KKQ#1bD;-EUz@-d6%Js~;%Y+T^)T81yEhz`Q+<1{w7gdlqmo`I^ zuLZOC(;;4Oh^?&iPa{{E5$+I~c6l4=Ps?#DcKT;#v$G=;qWWuzLF*71u3=(66qvb?{a)tZNBP)HW5o zv}!cS<)Zl4l@2;d)_45w_Ut`(n@k?aVN!_x9RAtS6pz- zTRKgvl>0J9p1CFMhNG@;p+)bn$a>pxS<6WBC--ilyA;vq|Fq z@*wfZzL^xt%nw0qOf4(9caAM2J`2yhT^;%5G+=LF6T&msXmP`GfcR7 zZZ+h+DaWbpQmXuAHi5p-+fGh4q~gfHOgwzNfV`Mw$PIspbi(8s^3Zez+A1Y85gWDX zquqw^voe`DeVK%taun$==f|SgUIWY~uLLT+QVMVBTS3rxOSqRDLn7qMh~mzAX6MWX zPDgq-Ub&aXH7Xm^vdDT;>ub#{%vrPMINB+SX6iV@=VK$$#^ec8l^f4!C_M)IJ0fmq zuQA*IMU8D!I}bsYk>qz%67@B+qh0|76@SU#E3J6EuT&v^mVA)wKb1~~9-U72oH+gHLt*R}v0J|}>zDtGM6P>5mt0-J9`T7{j~c|W^|cZx z_h2@e-gQvm$H$Ae-m&3Q=I-K>cS@0whhjOMg&W{(ZWK=1T}To(mB2{Z)%4d}YZw`8 zPDekez=YIpa{pMez<^W{b`lk_c-}^?aaFYV;Z@q^kols_glqOk7yGgH z;hvDu{RRx55SkY`g}$+>pfyPrnAJT4FZ~@wJc|REtlmRh*Q^vOu|is(8*%@RS*6!Cj@Pxc7}LTHNdxxQ@rL+x#Zu_xCK!;GBtZfTk#a zjacA53`SF>S!m;#gZzi}sPlU=ZzSww7lx|y(FF~-+9VV2Us#L=_op!<#{Z<^z_)b& zkx~-ZDGw&wwV;(-j!)*!A}f}aCm8w{%Z>S9`*z;Ebyh?<@Yh_6zw~(nr5l~qU9?DCW&h{`u_NWuY5{y#OO%8NUiABm_ac7_a3Y; zk>SrhPs9!F2}tbYv3y+%ss`xct9QFFuZ;26a zgtd(taOK#0A$v4b;Je4-^sUo02Y-U92rH7XMV z$6-m?92}@Lg>(Bk8r^#mRbQv!xMS`(VcIziYkNjsO4U6>rByj@JMEMJHl20fsGvFAl!#G}zXHHjWxbsE^T8Px*x?2GX!qu+6x#Lwg%_UJVv)i zjAIiQ50Jr|H-J?1LT={dgY<8NEB_?>8})Tq30q`>==KQ~{N=hww4rMTCN4F=W98*& zaKHl2d+5`_Lrrk$f>AhG`Y*$`%Av!TSTM7_Kp!?|Gn?Lp;}qM+bk+JrxO%xU7K|z( zUv!3&P3IP%gPJ5W&%u?BXiVqbZ=dDwgedbGTh^h`RZZ+a>4JKxLcxwoLXCOac*MPw zcun%f5Y_8sb>bix%?*dFiR;Njy)V=~sh{TMn$c&4e`ubS@C^B%#g3qNs9-*v-{nU_yX6Y{89&2jo^e+ZNnJSXQ2dQs)dAU^t$94$K4 zjXxBQV}jp!{QMdP{Oi9vSft~DN8+w{E#nyfD?A%SBq-3B63Ui{^ko1pmeaaA)U+&1R&p*u?y_*2R0t+?VMGl;NY}jwJ zRoO>>_X_P6Mba4jkQ|qNO3t0q26MN3IJqhutR_T~D60%|e334wnlGU4aba}*h<`N6 z?hiS2BA2Xt7Yf^!OA!C$&G2oQh!~k$!@vb-!lahZ@4$FGnlH-9i*x>Mou#__-@&I6BQW2x#?f4CiXLg3!Yu!CnVXXl4y zu$tSGX;I-P5;aW;Z~56nZoeU$5cdd{Zc;(Bxzg0T?G)Cut7GjS8wg&$m8lxJ4x0~G z;)2(w@mi}6x2@U(I(pn`BX<{;e%s9`9+qY&58KHqJgZ^T!i?F)I~CZbExz2-VVmf` zQz7ijw-k0ho5md(kxOfp522Nz1%2>OitOt1;?}JyLAh0$#BD=7-6VU0mMC{|r+u}= zSD$#pn;lUi`9qW8`1nW&on0yF{xZO5PU&X8JRc}VOB>3bZL$v1#oUvo4iC`IrK_v>T$--FRZeM^;(r4gedM0rxQy?2to6+%F zIa7RN7}_rw0p0zZV5`q{u3(}Rd@f25a)hTCx$Ot($I(OBw^sYu7UN>Lqdy6@oVo$i z{gv1h=gY9h#DVN&1wPpz87$86pf9hjhvLNbG;sPaX0a!wE%uYBp7T(gB=r!NFc%LcDLCedTq;lITiV`wp=%sM7s8I#Bj$>f_4J%fn zWIF39m&xANdJh^|zhVC4mBP&E7=*=~Bv(g8LO618U1~b}?O8foJZ3*>Jn){Zeq=($ zQDUO^YzHTGFpaBm+5whx-ayayEJ(8OArGEBB9S@DD4p8P2GJXA*x`B@zpIY(J86d- z?F1gmcghUe`~>a=a#sp>m`p@Wa`PMKn3x}BUy}j!V45_{wsQ& zH9+bHAEl!w2uz{1R@5ZT6+^P!adU+aEssgYtNRT3JvH^HGdvSgZv7@+D;06@vV(Na zfnv_8e-Ngw{DDoi=SkviPhuIrhPO+|Q+ldQmNuuHHd*O(}Ya*jIlAk(d6mOY421|99(z1o?x#C}IXz&sT+`0cM zZQ6dGUKkWb?>>fBXLk;4LPW2L|;BrX4*!W z@$3FM@%d@DaQv;aT*yZu|1!CVTe)T@mniTRZqgp|V|o#sSiF^3@+5mJH><1?t`{X!oA2gQkJa`g2FF9eLgAFDwX{KK`Y$uMM z3|0nr2)Ke6+8=y`yB;@(n)$|3Wv><@JPnl&k$nvP8wcCYydiqO0;p-B9+@1Ii1N-V z7${XhCj?8PI-23qjj1Sme*+qh-N&VkuwXWtmoPgLTBzIwSHW#|f-?`8ORshK@dphq zU{iA(NNshYCj_tjlfAQ{;YT#F$??FE>V{D7R>IZZ)W;2NgX#V)N13Y3X+-DWS`uO4 zPPe44BcG}UVT6MOj!XJZyG})lw%=b4mtqP?c-0`j(ZZ6iOv&drJot#gLoE2%{-fAB zXFq3{$I-isd&$wTbX?dcj~4H5(}anQ^mk7_ny(JP#iLhL97ddr=aTH*X{6U6JZpSWU=Iksp7!B1e%UNi%dCJK?jNYk6$AdrbS<15 z_X1Uym2q8FYx(A$k2q<2G~PPtBj5ujAQ$Y+Z}@hbNO(k`>8;&#QGhic?rnhEQiPfI zimK@RUW%{b1Bs_81m!}bv-nNDqCvujlt%8i=F!aeg2X7tPq7K zW=Qgl(x32H;SQlbFU4`Ci*bm0DhTXis9SuTEL-P9Y_#`^<_5AbwWFR}nLD1+G7S=D z?hX^q8o}0PX&$e1en0lQEIKh%?nCGl zf-!G8?IO8t8bCFk%mL#-RqkVFC|#PC!F^U9Pd{%nq|{9r=9o>zW!zJwe|D3qeO2N& z#cMcEy>!x77ejrcwD9Q3i`e&b6MjSTh~gzZid}Izi(#FZ{e{zy00|sx&%R8dt>}MH`V4_8YIq7F8#H`n0JK z=ETwys*7-ozzvR_Hxx@ZN}+jB5{;HMM04{D;y&pNQEu13D*_`c?L-INfBOfgck8Ec^@DNmpcs5Z}@w9MtTM=T0WB$r{$o}NhK7U$I~h7e%g_z z56Nb8aNg{C@~v$<#LSHqDl>?a+zV+eMRuc`DdV1zbkwW$-?K8-lM+UA+k8}44!J4Kr_&S(KkBiGFjZ(PX^})DLa|jkF_>`e6!8 zo-!Hh;!9}HgKhZRdo83!R$_&kD&0IH1(UX#lHXS*VGn;3L+_aL-h0mAPKnDzPGg!d z!-*9*XBs@Um*OXzb%C1h22#1;H*PN3N2^43cz7UiFs@zPSp{A#X-2F4Ne-hTz% zQUoUcrw8a?to2|0iTrKPoh_%WDB7>vW@3;uNK$dRBDZbFaO zM~P)FsgpllbI8lBCCpFnN;2132@29jk)@`#oQkQ1P+x@o_*`GAQ{Y16!yDM)A5XBu zj1<^a_u{c>)JE!55-y%~_d4a~{^zZ5BF!V}>Fh@rIE7w>tGVal{p#7ETd4`ke}9wH zOBLxo%^>#3Ri3Sx^$Ma+Ml!pc&(X?bx5-(PI2a#MN7h`k!9b&2M%6|dp{s!@7(NJ| ziMv5}$pZ1%r|N85XaOAGvmdm|PNTW-ti19M#_IGENCB%r|O9%9H=74#Jg1`cI$GI=gbEhOtABwW~fjaE$PVeFrp z!Q|@Wv~NNKxjIRorW)L)GOkmxQ(-5@_`akrKVJz5?pY)&YBV%QYZ1k76w01AfhhPR z$l+!3W~n=wv*$O75QFrB%@R^)X$Opz3b2*9jbddk0!>UM_vGK)C5HfiT#LgZB9)<(t&hgdo zxk7`w#m=Ika=I|^MA-j)q)&t5_Aq79J_sit!~T@P(CVtgo%y*HoiDL?!1D{2c`}8x z^sa?-3j0O#Bho=?<~Er6>;q@vGmiuthr^7o8swU9F+_Cef>6J~{li5Pwq2j>$<&0+ z^)HFD#1ITOk6@bne$oh?IsE6Q;7k7LxpOos2e;m@y*IYmeIAEc^=in#pZ zIzF*bUHFIJj!Dcxn!K`$Y;?+{I%f_r(`LNJ1p?b>qv#O*J=GXzn(si>fA(Qs z_&&S=3FvY=i%yX77az}SBPZ^3(&WmIWc|Qkc-HWfdvc+H1_%G7T1&1|$E92FUJ>B@ zz81XtVn5bBn1lITX62G}Rl23s67R_f=f#>g_-5}xTr~az-W<0dL(KZf>-PzGQ1Uv4 zd%E$;QkTh5ldlZ-&4pNuwi79gQ^K2XrSZ4EGb35mMR)v-qouPI!Q*WxV?Q&1Yk4e> z+0!~{hqV{&7`2Yh8?A~@qNd`J&+h!zT|cq0If0(yEcivK`S_|bmg8Rt&u_aED)6KD z^~WUnoTW$6^II8JUpikj*wIX6^m9D(CgDHRXb4Z)6vY#Z)35cL<{{LPch?vU!^{mzhjT~GCp-wr&znHfP1=74piM@ zG2eL)20iq_PZd+?+jZ;spdsJzhVMFl!HXtdqI({H>rV(NOID&kZr&0v?q0)pNM!JX zo*u=sVFTQ*>0e3Ao>R=*VNN(;XNozm5wqW}=GHq7&>43Uxa_T|^y|bc%=U*A|DBPB zW1&TK*Yax7&B&4Bs=(7kCDfWcd3jT;aoL0paq>a&zjT33xd=mh3sIr}1DEaniqlT? zq|D(LoK|@jjp-}rB&B9DWe%sA#KJ8&HChYx>um7qbp?9jtQvpPd?d=!P^>#YAojW` zc+K2DU{&-ee$lQ!$mi_AVPoDd zaqyT_|7|Ni|ol7L={xG^@lqUYU^98>g--<(wH!(}1o-o@#8so>m%gMLCf8xl( z6Zrn?W9sLl!XNx?juwl?;nwYiq-r+9B~U+7l|!Jz=K)?$8P5bvN}b)tfkx)%w~db#!=gUb0?p1!=?f=rxA zTBxG)bTYERn>;PsfRC2f;)&|r|C@jPcLs{2H4UUYu9|!{Zl}tZzrx4UN$_s&T^Mdt zN#X_;gImrC64A8}#%{|D%)qGQC{@oQiI=Q=bS@~+}C3lEwlB- zYZUeHZKMXctUZg4LEZQ~+?HwXILK|7v6^m`l^`Foesk~Tt+_i>aw{7i$dg<9-f~AD z+@k9YOu@0w8%hrJLeolfIQ83}xa~jBq*<>e6VCqSAr@jRz0?< z?kh~z@dqbSoVcA&q5ZWDL`M2JdFD|AcjPr$$F54MWhKn^Zl8p_{npU>ri(~lsweM4 zteJb`Bj`$<67KHG5g4x(MQkRorE}g?lE?Uj^l#inH}?M^tbQ}>A2N&`Hbow^+e65Q zdN0oM@*f|P z@AQkwougUk+;o%fd=$c+(k-VRk6$t^Qs%U%MhRC--NV*ljm#dJ58a;g$d{|#bo1!# zGKVk~L8WJo6?p58n(Y zkL=Xh`@*j1=&SREEMq@Ryp+skjFu4X3zjGL;kU`8(nnC_8^wOw)&%^H4(7m>qr~&9 zuE6IHfX|x8QPVygpKI?YKTH=_N|?CWrEx=u;mBB``Lz=Nz52*i-ugi%C*+X=Cu5>F z>Zg5_{a9$+bW&VzT+28Y4TdT9F>riU1nWGmg(x3kxJA=_>2=p+a$Nl_N6)J8mU^X_ z{dzULUor_?t|!3-<0MW$w!~iL`8{&5!+~8d?tsq{VmhP9gc_eQ2DghLq&?G&eQ7(2 zDDPM0Vw{kCE~R3Zi$SK>Y5Gl(_K70J+e2k-Y03%ze2V zg*|q5>`R|XtZVFQC}_!Gw}gHqYZD*Z4{P(}Mw;HFbtALkrqx8aw1ML^_H2dPt8Yca zI7>!L=ab#;8!2Shg7t7cwFqhqW`VL?KKb)Ln#|&?NpAHX(H_GH=H2P-koi=Zy?K;_ zKYpi~z_L@^@mu#vv#+mswrLI)%D2&h@OqfaEM@2XOl8&OMuOtQk>E8rio|4IgW07C za8ptOGl%+uv$hK?V1E+ttrMW)gOvTqHO7?_&rXL|8aE)URf4tZ3=kN>i@9Zip+qab zo{Jf~f{V$w{Ez?qe`?Qvcd!rMg&1JD8Rixj!kOaJ_@1(4Lv^!wTFnNKQA;Bevft2= zf}12MW-OfbI?9Zh9!;u5(?C--8oH95VPo6@*m$jqn%vq*a@`hkJNsU8s^8PeBei@c zzWEnp?W8K?-~MrnyQYfe9p5n3g0IKWy4U{U;G6WvD`j?VMi+GNJx)FqLe~&3H`T^ z#LMn`=$Qx|Ts&O|^#iVOw*4=tr|%!;^&oqAP`Z;kZO!3qJT35_UOYKtGzRX!l7^8R zTR{Hvd5E&evELsw9K(Jmqf7j7`f$p7?n_uL9b3(bomz$7W!Y(Re4ixNoBkj-o;XAO zZ%Nj5Mlih1ETzXQ>$wJhLi0oC(}@;kVlRFr`DzsgCyq~GS5GjffnzO%-O?teC*B%l zA_x&qHDsF1TOw*R!~89exx+epn5A;(iM3iBO*VU{8!_hg`(Bd$d3ECk=&b0g@ zw?4>#+Q01pWr}`j1bfPuIBK!VGiy*EG24KY)So(?B)u}r2$RfFq0O3CJSb@ z(dIRT)?2*du7Mr%Pv$td*HC!<(-TJ}vaN0cXmyPv=l%hl$d z)X2c7`$^pF^`)Hd8Do0aF^`HC>2Nxi^dWAx2VKz;P5&-B#blm#fTwB%M7| z!3#Wa(sC0LysDHW85Tm=j6ZZ?>O!#klEO6Yui{)f=8HRasMGkfXV9W>Gwq)?4foZV zV&n1-dgkhEI>PD{o>w@|ysgv#U1k}XnstT|oUfq2TZyz8Y@rRtPG~xC4(I$GfgGwdPnPZ@#1+F77t_nevKE=g223%t1hul_{Z zgEVM|our;>u8_Yg>rW4A4r5=0 zo`5~ZdUW|j7Mq_a;=I;kaov?v(Uv5hR=`EZbP@5x|xA2q}izts8tnN}#e zEU-EQ_s|cUq-c=CMNVP48qPDXBxBznXCw@N!szlOn(5<;vo?;w_z~~8wf)oK{6z`a zd?pFLgs-5I`9-wgn81UyyFl&El``VxkLZkw955{pU~86}^GbsqAoc1#P~9_x`;fSY zsr-Awp5IkQpMNjFMne-^^~VfUxw))P$UV4Tew;3NmxR}1-38aDKFJ+d!!48kBr=XU zB3?Q>M|3XQ4?0>u(WKFfV2G+N{5>=lyaj&!m-`2~6(!N!jt(ol5*a}!R8PfcjcdV4 z+6M+b7{}~;+KY2X#PP3pJ|tzn7l=unBAt};mt3pNCaDPm+aREY&S{ziXJ=1kK=UBT zZ+Rqem4m4H_V+Z!aXtj8?uD~wJV-`HGihz}<#HCA^G+WEu=(W;=shumKehH4&RKn# zq%WHcQZHg*vbi^mQRt%w`o?3=GbMcS^b(yVJp~tj%^?oY3&AA@ShLjG?Bt>Hv>+sr zO7AfN+n54y2)2bCD`covgunxM6;EOU-*9b@t&u4j!+iETz&RNSoakS=eDrZ2eo@g} zreIn&G%ngN%zTgG-x0r2a;zoZaeK&V$_?jb6jo!-xUbaln>7DvTn`;S;x#?GbPStv zLY9?3briKsTX5|m!R3&WAr70N0hLqF!VvQUQ1OW&v$Gx8>(V-KRgc1^rSC|;W)lUj8byyf+A}CqP-|hL-+?`l> zD1xf`Z6MjI=ka;mKMZ?w4riz}a>YRkT=Vo0I{)WII_b9yMfrS@Y(+kKqozyR`U)ZH zUl*KBC_#5^Lul{6jw7NPm^0U(!*7do_+eWr7(AOy+`Yr-{1s|ADepTLIww-!Ru?)d z^D;S5{fN{a_J*qU$IvEFn!lE4LGCZw$o`jSkX%#%i>hA=97#hOZL(2(^Whb|c9)^u zDZ;GD>J9g1u@f_|;uRF{nu^mN>*9=#AtW$n8+&HoAsD(soz0SxVt!5yW}21O!LjrM z;BO+P)04Zz@Ax>{>ottG>Js+GXX=1Cdk}(7mvP#DY0UTEKk3-=I{HL&8`QSzv8}RE zkQSd!#!emuwWhkPnzc6gsT>5YLvFAvA{-vQVd2%~3vlLADr1@z2_`zVuu;vHG(3ud z$9r4At9%19(bF6b21>$^PhJ@DDFLVcv-_RX8^6w1(@WldV{+j?;dApY@S`y1=$&2}M4n{N>B_0p!G4abB zxvqvF6i1ZcTOq^MnWjf?i`+2gf)*qCT1s|%jK`;~(?Dr#0{1?e6RS!@l8ZZT(~zQ- za6F`%9D0*Q154!4F~*PIq~uKx&QpWk=l!^+dQZ9L=Rr7gt~7L|NV4C;|AJ$NISs!v ziyYX0iT)Beg}oC`67{JpELzb+dPEE1MEx`f{@o1MI2}losG<=^PvDkUCCpU2nII7; z%LS|N6^#msfidk(a9sBa3~V_ADLpquzoVt_RZ}|X#5Hlle-Gh@+BEXr_BLeZ_&K1G zJqOlIT!$;tB*@ObOuA!#DD$YgleuI)63xEeVccg7p(~#bhR5q)(~@7)arqK6+8Q|> zRWu*tw9-!~F>EzHb|}JCvfm8YcYI);8S+6m8U&VhU4~L34Gam32p3lL0!pm zxE_XrQ*l?twoXT%*VyI*nbu}S#eEtwzlqLBYE(tum7e1e+*wKUl|0Mr&lbH;MfT&-~K za@FB^{{bKVv-v_kw{-%#Z5~CoT#{nDQ&syW&_qoY^he=Som=(OZ+l{!0Wh&P`$pAxXx0e zPA1k^@QTGv6I_{QBX3doR0VPVj|zU-nZdlJ-7RL4Mh>?koul#%KQS#}9>fmrCB9+i zq^Gi&tUE9StuAWQXs-&!P4_CZLP8nt^_?b)mBOn9z-u58vQQ1w9c)If$@QLD^tJY>5v zNs9d1bBNJhfo~A?o z?31B4zZ{}7?eEe5hpO`q$1;5XKV@e}8D)zQWj*(G-jp_y&_E@nMH*Tv4I^82Mt1g0 zi08hpr%BR6N+_igLPlIlkZH`2POzw}<0C?&lou^Yx0F5s7*pDe+>T7NO(0 zX2@ZVHoE@n0(E&-AE|e=#+Ur~iCe%g^ri4CcV<89n{P4Wt+4;d7&b0tjyx?Ui~Bje z@s>(bGI4<@{rbu5Hf0@)`)0Gwx>dY0J1R+--~)1GVGmU_I{}H>j1kd=7vcGhE#%Oa zaO%*p6TsKQg^_I$u$;|C&scd0x$9V=J*Vr;U#?$AWFpVt>q|F4+_6odBUX#Y2j?St z=TQ`v;foRiYSBb~DLUmPjZ(JEq!tDYp>5mnAvfgt=0r@hdJDt2J^6#->+?^X_SEN3Z@ ztPgXDm)~u3ZiF%C@nAbOE}lnus)>!GZX*P=Zy5&AJRxXH~k7vP(puMoq=Mu7QK2BSyn$U4; zQ?QxEPP#OQ&^K*8X_pgKyhC{dynFVTC-Y|xTIH}DSqSB!puy9?zbO;)R=>hIA^S+U z={$64ojz5WJHgTaYJrZ-nIId2qL?v*6J$-3Ht&H$0SaR8xvnK^n)g-OVyn%;O#A%? zYUJ%dO7?LJvw7Dr_0eC7zRG_UYF|DgDN=zTWVZ@@)elj+7o^BE(R3XAb}dq`zCs<& z-%T>_=uq^+CNjG39Ud3fBpzm`X?5pybck0eZFIi{Pj#JR?k{L(Ko_zxb4C9k%x0i0qcIdM>`x^!KNW@tL4Vm^Nkt z^O(CRls^<*Y&Sfj2yj&y5}nmgH>0l=NT<1{^dDXcK;u?*&55lHm@dUpLl>qbQ;Wf6NWPcJFwfD zPdKbwkJgqR6&T`1bcMI5+k&uiS-bq~HL|D-QZ3HGuAVGmKHp zfcb)vDYSWH7z?`8AYQ~W_*mosbr%e&bsQ};Ur7?pE$&92!q=g5e)?o6FBk1knuiv8 z7t>~Af@F)i4ax8eLwcfz!M(@`+^fzI&yBA*7pcp5=j#)U=Jo~@rm&aMlq^R-976Hr zX<2M`Z8pq`C3tYU9#7$C915}IBR5v*BF&A3ECWZ9nku(Nw?vDvq+0-Wxp5P{tvHaj zjlO`o(gh$-Cy@-ep8>tSmGtoE0NTbjhwL97hL5$KbcL@m%L^2vH%NuRjsA-`R1iq9 z=Lbl(^`>uYynxKd8uZzT23%P57394nXQrZta7la@!>^jY!{o=NiP)nbaY$@xKUo~t1oTPDO^E5OCY*M8$= zk6KahnFY*Dmvn;IjHJ&gPuP5BC0+hFozZTO!gBG+C@gk0wke2#KGX@Zs*ZSQ;3D>p zR=|%>3f5hXS_Iq5cVQdt>BuBL7k%iWNb}v7RPwtzB!4WN^##Op-iJNrC>rn+MM?$T zxi^LOw2R^};R+-$GM}!U?uMG)hmevnA!yLL4b~O!skcu>h*Z~RkXkLsn_ZkpEQfC5 z*nQgQ*7j)p!C@iw*LD@MUM)rsby;z*oD)Jb{s_=1?XyC4!GIX^+-(T-Lo}l&r2`%YVPkYjTg0(4~_|Ld_kgHCB+J8_K*xSxb2* z)C*`Ut@HHl-x+kUpb;9BoxnGmdFXiWQWCdSf!HO=sk4lelzZSmINVTd7vs%4#K@HB#EhkQSVP^aUTZP z;Ug%l?&bS2qLcb=ZW>T5$rpe#_UtEL4&XEA@kQcc)Mi@n%(t?`IW7M z4xQ-4_wT6^mu+5TdQu6wv5cK*v)?`KqBLwAi^6M?bYMfg4SO?GKt`N2xNc5-)YV!; z#^+xh$IG)4jjH`K4`JU^av|Q-$3kn)LgnkYVZA81FE<}ONnM6x<*y_Ah|{FVObeS- zMerUEZQ$|GDyoYYZ{vIoh(c>rkK>||DpGoKGw*NUUZS0DMqp_$QHpvwk(w=97$w;#8AbVKXG}NEfW9ymMmm% zhwsWgFdyW^l4RjIJfD4Wyi)OS%r!qxJ_IZxS+NVqiy4=gF3ZCtbHpBr&k!P;-vLU@ zz|=X<@09G-4!rl>Q&_!FoaeJ&ji=CFheE2osUvD#RE!H3H)%>n(!=PsOlSqG2Zf=K+@9ON>nM%*(5 ziO*jH>ZPaw>N%ZFoM!;K>u$|!J8Q?Qnq0>lj`HMPaPq(-x(}E;Z^H?w?dJ6lJtTJW zR>Uls%RCK;!)|`U#H({Lyfw?gv&)Z=XOfAy!)O%}TzU_u9ioU+og$g}RTP)o8!?Vj zm{I6KrvzT)=a**>c z#2W>v#gM=)ioEDsL1gy17;>c>6aDp?*tIVot$e5l8M5-gjHHu|7oXrd??`;!@duv5 zzc_n~UobkcN<`2|3th4bF!ywnM&(Jn316oPQEZ<_a;Qe^JSt}vp=ZEzyE;Nbr$=LV z|5;2>h(6P0^&2l87$ujD96^fSi_e}}!Mv?8ru5U)iBFy-`86QO6IhW(`sN?Ojctp$ z=dTWMdJi4KeO=2)_r^9-{#=r`xp4+>W#Swd`|HS~Beh6@%P?70qeS4`>%lH zRgd86a2t5;y2EmfZON5)(|HA-3b0{U49rlNfS2}8;Kq%?hXdM?=;in5{%9mI9gnB9 zgG@-PO+L(c>PMS>1-f-fHfnE|!19Vxq%$KKGB^`x?c@it^ZY zhydC#6p0kJ!l{qXNoZ*yZ_lnW%J-f>7`vsCiUXA#f6e=KZa3EwYu_kneYTaB>?%Tc zv>u@OvlJ+5Y2&ZyvOn?=R^N~)ZC`XlphF067Q<; zNZu~g%lgC)3i{yRB3UqFI_qlAm!M-Tep0SJb7|wVyJ=hPd|H2102pbA!4it1Il;nE zxZ8ldi;IKbzE8R9_8uYf%v=0r_8^3bCqRlyD)Zfgk503eAbWb-@zc8tzE>E!98!RN1M87|mpI%KdqN~$1rm{DE|zhsL41+(d1epF zP_&09-FfXblUY8CM6`sE-QRW~e}_QyV7!O2TrrR76cJ`a_~8_Y zt7t)Kdkcw<4(D0UmY}Vc8qz}5muYPQA#xz&HZijAW)6Q{3#*U+W&74NqBF-GX0dad z8`06Q@?|f6&~=jVMcpD3=9p5HSdTmX3sL!|aqc=#chG%#o0=@pAdP`D$nLxeuyKz- z^QVcSR4N&+t#%@2^(92m{x{y?BuWcCI>oZ6J>l={JQO^^4|jL9Qf7FROdUB%JQ@Od z34ZI5v}7KHzVW9=6GPC6L&9(zG|0=CL$p_eBiKku(kJSou>Wx**t_l$8u;f5(%XK6 z+EWIdHF#(qCm2s63+&0IABiX=b0*9fyM;gGN~5w|J1VPpA)0Rz1Dk#u(dXx`hgrwV zz+d+poEF|d*ESPcuH`5l^yUD){mKZ=VDB#)rZZ5lz8YRg2I4zf`gk78yzUcuL9Xs> zCpA7XaPE07yK@i*5B}XOVnLL*T>lGk7<>hT_zx_!Wj#qQ26SoZa(Hzhn(Sw4;q~$w zP(BnzyUD~+P1O&u#@Wwxctbv1`4~ygyiWyt=^;4rqYv~XtYBxiCrnwyBBO6cSZJCo zne(*-`%qUP*sUKIRsMweInU7Whn;k1#aD=nRG~rh7FJaIg0+sg5U<}4P|0$4(&Agj z`#$lHQSRCgGltgD)>nJs?H(T9QXK#P)jzG{`-sMky+ruJRcMU9MpmL|l3yPT;e}H4 z`6*f2;GrIUuz4xw?9gGX-rPhQ=kqyl&xtd<9xvoFc@rnMD6!vfOTi<@hs(e1HTpuI zL-#A`W&NhLhNy z6cDg@%Na;iCwD0{QL;)ae=^Sd8+&?NSsU1!~T|xdBm>iBF5ef^V&N}}<-XpGw*ECjo*VN@n(L~0`*V7^da zB=4nyE-mGt^PgSFfoM$-*30wI^0|EYI_r90xi<@Gwxuu$tFt(wRqB+{ z?ntUbh98?KhNC5Z6L?%w3f4YVCw~@8f#n)WNV@w4iJ1rE6Q>pEANvE~aq~22*^`b= z__>=eiFHDM;tr8Vo2HZ50_?n7-jwLS9LMQ$wP^6WB>DWQfwJ5B7reH9f-4_lSihzh z?pDe~$(Jv{zPoPBOl@gAE%*~&@^vn0TX+Cl#G8;Y!$#!5-op#(7(=o9UrIZWTNkzG z24xwX!;w!5qPAE+;T{QH#*N?TOudP?$=Ky9k&he`&~A&u=~D5~>=6k+6&2CAPa0*X z-o*S?GG;P`hVa1P1}tx*4T`6gajfVsB#<5jMT(&a8hWW1yDL;uuO0F=7>0rT-L(32 zEAmt^3ZMLZhiTanh!!`qyUflq`2(gN@HBm^N|&l2dokflqAdzn=>s^hE)VFBPZTEyKX+ zhaYVnvY(zPsOAbw^5Z>fcVYk5dsywHCupWbP%X7jQUAvEOo1+s37j>9c;!tX+nAN~ zl%*!i!M})IjxD0j-&z7wd&;mtTLP6%>!K5pPUzHqHF9N-FV|nU7JYAfi;HH9kRg>O z7|IEyvjl3fw~!){o~Dj2x~@ZSawAat2YsY?w;S7Dh=uJhoN4jlcyg{JhkV@sl=PHr zhFM#$BhiIR==5VTL`#VeD%Kgp+ejzkS}#f8Y?h@LWb7g(eQ{jVsZv}aHid>2MUfVR z{xg66@9g}aFP1jO~c`ts|rVqzb+bs4V?Pp7nz0C-^zpRzf z$qeACxT}KWlplQGqJ<=U7sIQpHQ?UPpbYR}xc40r85rt0xKM)IONu6Na^ z-<~*#G%xPsI=I%O_w%!;&VijMczZYwvd-sbGQ#YvnoE5orj+B43(Q$p5vKmT2xalf z0r`sDMlaZUy?;E98xmBB@J~6S(kjKQT3mrV6#1wwVFPB?qhBDEUsl&ayc0zaJVk{KA&_%05i;i?;HCFb4{t|s&)=zM5?&pFmgZ$>Yhg9@_}eUU`^`o2 zNzoNulxt^b$E{>xvJhFAm<2sc9hj=GHOxD$Sn$4(f)oYslLrn;%)4$bCrxpzMszzs z-kBOv&*i6c@?yi7vxQ=KancPcbZjnm<;CL9M<*x&En`y1*$Lm?%W(_d zPODQ9QlnbWEMo6c-B`YAEYh$`N0>L8YEL*vg8Lo7X8Bs+)0hia+m*SZD3IEzP7{9N zXvmv&0qrXjBVkMSb02sk?7G|#1Ukd9ZWLaGAqdaUasEECA(EA?6~V7<7BdGt|$JG2- z!Th{sh`c;#w8)|muh@|UDz~1X&_+R&)Fz5|$*EJJJP~yK-vo8mD;?$Zd$V)UP)J{N zzxsV`2Vq<&E{QI1Q&9d9gfA=!zZyMDn7r4!*sasc&D z$qSD7hfx{$vHA5mlj!_?H>lr~hqGkM(MscUSkc0tu`Elaek6A@S5~;f47IOJ`cWZf zN6#NdZC@C-`&lh|BQH+A{$`jxUrI35VF_^pU#RCc#waW;6QKYH>*37kI zf=`CvWonj;>s|*4S{Fv?<|cA-5BosentXVc+`|=akRgvIO;P;#3`%*|7Z~{353@G? zr3Oxkp%8%vj#=+bDrHoO`^s@4`5bWs?|E&;5%zXwwk989`7=8~S2zHyd_RFgloMn3 zSQzXA74eB+3F_bWW`_T;JM+NV6u%UTNA9vIbvuM+!`{5@IIiddy0K0U8rSUtT`Nnp zF3yNe-|j~Oqw8^vBZ0mCUF1x&7ar#XphngWaAKo6wSKM-(ti7e%h;de{@?eH*|0lj zlqkchOdis2uOu7SilQSH$3f&{0802e7glRMK{cOsaJgVM^6y9i$sab%=(Z+`FHjc7 z(^O~^zX!0Qm5)Bqnn-;(b&bn^yBXc4L(uLO&q)5pR+6cb2D<{M(a&dfLbYoT#nm!~ z_ooVwpS%WcvO*wwJ_$6#>hLa~6FA;V3mHaiWy*C^m@MOTbnVLu5Xmcm!J_Rjf6W+m zt9dh7yla5^<-Z7wV^?uA;;VRTcsj%u??Z}g_WSgtA$M=~C_LY`kp9J24$pWd zFjmow532q^Yh52>6>=U{+lj(?zuO$oa4}S_z7hn>*283#ApJC?1$I3ugsup^U) z)#A-aX>uU)C*{aVn9UpByNUt=J~JOT+@xIFiqVopWk&vZ1$kMz2M^WTqrK62ssf)IhE8iXJm@_}Pn+5W~aX)W$wJz<@@(=$|`URi-N^#K@1spdP zjKAugry^JuvF`2ZsG~8B?A>_A{EA^8WqMc^NA9>z?e(f*0$V2-|HJk`*uMJwd`<$T3_B z{mZwbtOYC3)*a{3-n>s};(0AJ%RfZYI_=c*(D#&B*kayGJyYh?{a|QM6Q`xh#z6Pv zH~df~5-FYD4qsUp*vBIjy(<48y~mCNosH|@R*($6wagHy?-b%S%5-qHk(}q7As_bU%sroQs_QSio;w4ZAwd<6lelDE{bJ z>@k@M`3o%R!S)%z7#QM-(s=TOgCQhBna3&qL)6Ga^0DRwz4V(3TsIAaZzumj$8jD+ zznlTSBgNE>*&@t|nENPvQ79yQiX|cA?yxv^J^ob}j5lwajV%u|uw=3nwhS4Wj<5TQ zjns0G>y0&F{&^mcc~Z-=My80_cWv1J+YvfGwZJ2tC3K2b4?Le{2LT3L;@BDw0|RoX zeD-59SNSNpbaf*MH_n8yt_isRyy-vlpa0ID|5N|u=o#bO!TZEgz8R-^>fmktbFqhX zB3{)Q54yK%0O}iHO};(Y|L6cpEE5HOKTddWMahO8ibV6*INEVZ70Ty4WhM@r;orTl zvE=6j&g{i6!Bl~V0wiZa^SQ5ZacerP6R&|gOHPsaG--B5E=FF@_2sz`l0JtPtr4Uzs{W!J_U7ShFJy4o+bOas?K37fSQczf0^W5igRFR? zK~4|zk;`>Yu};?{c@Zs4?_IkJJjNYi!D}Jrl(Q!m+1-cUR4YS-kOWV350B!rBqaC8 zCwwP;oYZi`P}iCJ5V6<;sN6}I|9b|#<<$?WGKr4}k1QdfX9<+(Fr2n&i^*K$nVgKS zCP+*aN4x5Ku~_0DSmiIp-J7bgky|a;%uGb@?}(6tcO_|WEgo(-l8Wyd?t~Qsi@|(( z30iJvi##6B#Q%CVP^{b{=sTLp-j^3~Ixbm(=8oee`SA;+|0I>f2CASg;hCKLoaLyi zeh=O>k01rrB2d>)Vl%8?5nU=m`ZN+iWk(tWNG>3%`cKIBI%yuCp9b2Dir5~19VW_A z=%ku0PHJHHaT=@-DVd*lGQN#SE*2$Qou81yk|w0?-%%W{r9gr+b+H#$3#wf1Lw$-4 z?EestgB6o7=i6+QN4-PqmY#yH6jA!-bbAQxJ_c4U{19rY3kTIyz|grK>G#Qzcj4o( zs#~7knCbvyj>2@@ju~Xlrgxye))Hj*Uglmc%7Mb`Zy+!8Hi&sP;{@(Tw6t#y;qLs2 z^{@M3BtfB|524@`lSW1hs_|QF2F3iNBtJP7dZKRQ8B4a3NZ)C^!(&1?Fi@3x>}8DN z`Af;yUICm_rBAG@%piM%B$)M7fnQ++b>I32^;FFSULIQlQtX`W@roh{b?C=&Eql?_ zZgE_uc@)Hzw!`WnW%BIa4wQ26HjL#Ku-v}$z+u=M5+PUk5t)QWma7uS>MDHN$C7sO zm?TRY&Xscwg~jLpF|ENtXsim5vLZm!24&TNQTY37;%)*?4e?Y`$ir2KHC6C z4i#a6YS#4@I>o5jgqpX%h{ZTMhbciiAji!iMSm0^;oe5LZKaNS+{96+dLd_dpceX4 zkN{4)C!on?KKyh*z_;9zwARk1BYN`CvdK{VW6lD&my!o_DzeeMxee$DJJ*~DbtZFz z?viO)tC`p4Q{<6zAzTj32a|y~NLEiKF709z6;?`~zZa!B=fl9ydLI&6k%}HEryzgt zcTh2|h_`M_pia;FjxFD^p6ZYFpgdto8iT}$nmWDWb&kNq-|#;UsF}ObI0x_mkRhpF*9h+Ct*ot4Qp5W8N}Gfm&bsnvz_-0M$W~qYZFAw+>dyr7`ew3L8xx#xd?{^nP9{E&3{gPFAf%61B^q zcDn|&FHWZ_=L#cF&SlQIrn@Nb^Dh(|*76n%ztv|!~O{BVUVzJH*A`YB+G)Ee{1A8~p5S__-snJ0l) zyC~udk2Z5{mUt49B7??m^U-oQ_2|>3ouJ>I0(+?NYeUj}Q1VKaXf zUg)WT`DifGYwN}mK9BI9lq~Kp`5OGaEycXKM2S+d*5NEUsLHLhlIG00X;xRhgoip< z7m?^zUpRH4!+hzX^4fwy8T`#8igWOj57jJq6bd4rasu8Ka3U%d@Sw~O*sxs;X4@=) zlLv<2SoLK*^3jd$3%$|qfCI==fdjkxrqflH-(aIpH@$tIJpIF7nod6BMHg6)QJ=FV zkip4#%6HB@rsZ%5Ba+fesXv-WMuQ7T;~5chMbrnt@!$c`5D^u01f#V*Abqbf$~DMcaK(x;FVCOl?r1Z|i}sA3Mm{T4(hErv@ZS6{O5n)KYTuBjpXko zeO!J~h;`ntr6V@4qJL(r2a{0=_I{@gWVNnA&9qDC-oSOV_Uv?!bI~OWE{!Ah6QGrQ z#>kqJzT`|_B6$`l$uexh;i7yDyes$t=k5u`O12dp5C8{lq2?Ycb)ym-KRc|mc`{EzORwo z>RCzfOkNX>_E?f@C!3gas=pX{VHf;M{3km5z5xAvHH{uG+s7QN(jom9W$|g{T<+dQ zB4oXv3z<72P3qh3Q42U8l&2v(%kxp7W`umfcB?e-!P#bb$@n!~wL=6XbJ*J2R}$@- z7la+~W}LR#5UL|)6M3*9n+yA~@|}6uOFoUs(@G|WYYj=Wl@wN+^Ofw+d`)e?-^Dtw zD`0%3I$cmd20=waX#S6X+yc2QT(nOd`#l`SLsHVP!Xki(!1!xpO>=-PfIko0P)(|=RK;q+^ek zGZ#^__9XRjObr^0OR>IUF?PM-Lvp9pGhX@(Q&K()R91=6$={vnUTrseEHDjMaypS~ zlQeK%ZlE2?yCLgO0QM?-4nHP0(e97c>Ba1gN)u0qF_<32u$Fc3s9HcQVz~H2-y^6E zdJhfV@6j6x1$_DygB$D>(7gwyB&(Q*Ymdtkot0+j$73}XIB*wL9!bW<8JaL$)rR-S z7(ij&Oj<^MjIzCCz*LP0k%aSRI9bP;Y0K?H8dYA8MWd$j#8GGzA6F40nVnHYcu6$)y*vQ77gwX3|Kb^E zdJHl=dl~T}_U8HYb*d*=A6bkPAm<~yu=Xtq*JlZnw_)>8)A4h7*8@u`M|vkV_8(^! z?6QUnS1&+ht37JlF5R@q+IMJvIy_T zmL|V(5PR!BQ76UQ>0ry7rk;W2trKDN#U+&V!5`A3*PsWf_6%p`G*Fm%3+rYnzGUiS)2e!uoBQy#?D}&Hlat{5|PZc3_`_- z!_Js$w5XHyA+P#|@4odS<>hJrU;hjT

*n&)ItR|mt9NUIZBcG;V_@e7dn96$8tuq+Ibqj+sA>!hx-L1OO@LevmJW*U;T*hw^GakK2WHeEgsq6#cbgTGo^dZ;iuFiPD%e0PMd-} z@15cf@{GBMCl>0WJ6A1`l|(A)N;2$;hY*4e?kKJzZw<{Msns)yxrr5AD*@nynXmld#hrUqz8 zWa9$1rk@I$kCq1<$9$#lu+*GgNOpM-wSJWl?%5!M#@_PaXO1j=$Lv}no{l7RzJElRx;%|89%wh!JSsr?g1G|5i$w+r6;++L< zB;w#){GequE_1qu(t2yChR4Yij{k!VyLS=U-Jj4dHw)ChJDK6XtlFuKb*yUe zv~BzFVfA&WAl?X{D9_+%P6d-C7j~mHvkGvLpgEBne994c5elwBJ!E2_m`wevCU;dX zlGezj#CP3Z{HLb^tshF@a2_{=>T*@do2t)tB~~zkJSe9%D5l$ zVmPHs-C$P~ADn%ki;utD17mj!K;pVNaXq(;*bGT}_G?@iy^J*ADk5ud@ErWYK8=
667R=vM2EF1Ra#5&?1UO^TVxBHSDyH|UIe$v+*!VH0u7j> zL(e`;puB!l=yT_6S~+q)^8uvcbvntj|Y(WXCEoQ-AvQ0tVu$?6b(knxF4E? zOKr}yv~NBgoaaEMyLVGz(tO%E^%I6^@1qzqKN_^~AUU1diWk2F@wI3J-h5p_wjGw# zqQm`~Ctkq()E~56Q>4#P4oJ{&!~CdKxM9;kV+K5<_==~raJ~{Hh{=)XyZ*!`_aiB_ zk9@B51hxkQSmU8=X1K77B?jiRsDcu<*5n5ZjcyloM}HCIPJI;)@Ox?#_iY~FpD!I% zX`#bWQfT=2nXV5D#7OQ!NsKCnR7nS_zc!-gL<<(1)MIO13*MfNM#XSf@) ztvqTg2i{RE9%a*A|i`Ez`$1InkmVWy5dRhrl7#7VKA*k7`n?a0~92CU#st%b9xLT?dWD4k2q+m&d){2-EO zc!2oMuPjrIvpO0_<8Ea>L|3k2{C|%SE3*> zDt^0$43_Vp4da(lyWcDdx73GEu`ISVOJPep-{W46N0^Y0!dsJh{#TCX4sd7e4!m=r-?ej=JWaUp$io=xNZLhyL$a&+6+BVU~} z1Jb5Zgsq4gZuwF1vaOUWI!en82U5?_fus>RmR^PxiquRbS(}v>%MaQ>FLm8$v$P*o zN#A8NWM8oX`h%fWzd$IDbrt*`xCoPdw5gvB(cN1dpq}-u*7P+pznN=8F)Qs zGoRM6y3<51+jJ?1&!9Dz|0K80KDxfEiaP33aOwIT?D=pND(xdE#D4%?;XIW)ZL6p{ z)RY`gcEaY6J$~xh!D_HRZsxVJqKbCb-|Hpw75|2d6_Vt1QkwSpmSFCv5|kUo!ZB1D zLeD*>nmQO&EhQuzzE1U(8N_yN#yXF+&^@sp`J#z}Z{-AGnxG@_zMAOG^gz)8(>;9b z*on!*T5;Mt2Yc?1;ViYEBI~{4td8#`?zXQ)yu1;P)(>aNiF#~(rY#d6WsiQ*n-MJT z4ZDs!SaK%X;>EiW^C%hPDx=}Z@8f5r_Mylv02a&rq3PWp%RB}nJV6O^!{$NJlW^_O z9=MJ3q2iz*a=5ve-bA=VW`h?7`v#+6aR8=m=N^}a1XyPDGAR#PM9=%ix~?eWa@+{q zaWcj4jeW$n^%JU+#f1GqADI*9t>yd~388f=nrof0!o>sMGaJ~y$Uc^&%fE->t|Hvx zE?TT^V)^6I+@U-cO=>zAUlzlx{7&R`jBOgJ)@ z%&e8kgh?XEMhVI*J~7X#3g-IaIdfU@p4kkrN8B?vTst0wgAKcp7?%Mb%_yAmKLq3c z(a_fS!Lms~u;U#~y5fS2;8Ipvewyw6o6c;0a7LEJR6ch$B_Zk$DcE<@RrUUYaitCo zxTQ;uPp4Ag2@~jyu)&g8&dlMw(06m6urb!|Y}?>I&M#LYuWS8jfpR8mJaLsNK6}D4 za;~AO=L@cUuHxCJW>%We54+0cxC6U}cRp`ooc(<)+F(OhoIU92uAMZp)(S@P{I0Q& z^S>v|#i~uqP-HwDljPU1oBvXoY{403yIBD=8Nr7^JD{xr`w`=h#MA8n}*N1t?ciT`)s{J3wv;}3og5qsqs%Y7S^;N zTe}Qv#^k_2>@0gU^f(KS+{K(Xxg*)w3oF-bz#fxK93EB*vn4OFP<9nw@H>k2&||Eg zJ9O?#r{UGPTr7Te5b3I+ygLyOWuBdj{~m-1L7X%0F&(C-lo9cCI1=8^Md)X5T>KV@ zfZKmjef%D@cfUZWL<(xp#^bSKD*T?^5Uu~QO*Er5TGY8cD~ z_sSzxp_rzyGZgUf60P;Ugsm-=$W@C$y22Iens|}wdd^X{aw5C;c+nCLS{|cDeoP4%@ z;4Hyx;0%GsGz8&{KV52wps&9+&}7L5<$(Uq{3L!(J(wvm^ z3y}s&nHrQzXp$)8YLvJKFq4~ z>)0#57i{FPE(o5lP|0~BZeN~bbTEJ36Lvu2nj(qsHl+LzEi&qAzy@y+F&Emte*|q=q)oocr@-su zvi4uw#Yl$H$!;#{xjS0RAw6F3l4t>vtxzswATo=uTrB|@a*W;P-j1nff zKAWxl@Ji4*)Gp-h`YDLr=6trrQ~77rbX1=^OG^ewh&w)M7`sHAw0DV;nQtHTp-IXs_382DhZwO%jW$-wQh&ZNjdo<0BA8#@C;Y5iOB?6xq^@XY;mAjOhIhdPndk#Q8VLxY@GDML3x4S!c~S9WX| z*x=nrJ?)QqSyI&8t4NJ&RVkoB1tltS+8`FHb z7kEBgf^N6S(&OjeJlEldfe*FdGeed}+|i(nIw`tzIfnGogQ#MB0=b`8B}wj7{KNZ0 zqSN~>}&4Dq^!%zho=JM+pCwTL`ftwt`2NG|kwo zNY!;BBonej*fldq$SB<@6nnL@9a}os*5$Hj{VhuI1!|=7c_E%6F=Qeq_e z&Rq!USs-*iaS#Lper{VVO)oq9p~pIX5+^HG2Fsg64K`cggc0%D9|NQ+;D-u~0+N)uZUu z6*ap2Nr}d*Kf#WuXV`zG2xj+Mu&Mkuns(j5jpaiH%@iHslC+G_z9Nj~td6GHG6A&w zwK;j88bz;98`F(viJari-LaigI{qb8m8(|T;y2lTt0xtUaG?2q&%4YNyPmQ5t_w!0fL}Ni`@Pqf_h-< z`T~F5ishxuX6!AcC*?JCD>IKek2cV-dA0P_?i?+?yMep@ z0{FRjBgN(ct* zkKMC7#07(^mOLPaTH#dLEkpMqU9snD11#nd3j`!hD-r{q%;D49#Dc%Q=?Ccx`n9mAN8#dZvf&??XcNLm0ccGC>Bb|8hdGT9?vQC z1``=f6m5*AsfQ{3Sv+}8*h7OC6p`cLlcf0WJe6$>K!ctqrra^c z<05(17yj9@dz=9i)s&*js~Y6U=kd3{Y+@PZ#w@=)o_$&CA(#ZO63&P(6>54eLba<9 zcO|^>rT(g*6jUU<%*__UHZauI4Z;2Vw=DJ79J+04O>swO(b`rsA-dU2*yN%mY`Zv` zv<@=5ad#Z)U=|HqwThfnxKE}%mdpxc=u{i$@s4?jX}&-BU2h0&n59F1{|=)o7Fv{Z zZ-g*EO<&l$OHHT_2&75d_mf50F{;yh35S<;(7e`&*3CM?tPm~XclKamQAYwzt52Yw zoc&~@qi5Cr%D_tWp^lYNbSTpuTflDAW^y0(J36`VHzfvj)9mF7&|J9=m0eSCxpX`+ z9@f}Y?Lb^1g?-Z8HU3@)k(|?+k+hm_?(iUsp9g56W*Fs*?V~7nU#i)*llJ(Wqbc`N zDBJD;Ej|)Pd2(ause~4-{8FZe>dMq%rAqlO!%0#` zfn?VWpkBFC6nf|)KNls@)F*=|YkDU(Pv-Na)1ow5LYu5QM^KycXbO02LG_YDLt}{^%~c)TkBmhCiyWGeOi#G z*9*-YLn}Xz5mxoC!_9_XtS-fr`y3M1JXHHX(~{=2VdL{c}{lB!@gwgQ$KhXYVzqQ}U%? zxLpay@{n+x-8@maTkI-$zMCto)GQPXADk10G!_Z@fi^Z_m`Z6$}@d|X~w_{b6B8|PLN9$_lkaU^> zrR$BP`Z8ml!@bI`cidr<`meJsOj|hjUQT%NLROgYYZ`Y%OeE{wiqvawjH3TWp)Z6x zySSgj1*2(2_+&cOW6$R>ZnQgvce*V1)6(CGq~jhGQ^y)KJh$)B9G@*Y~cJ zT(_5IM($^D_{Q{pm$I=(6KS_wHJxiJrYEP*!gpddCbP5nHZV}Q;S?<--ajbhIQLSU zmzXeTn5uAhOonjLBugk1vW4t+f8>_z!-u#1Nrpg1%t> zwirvjUt&bI6s=z*M!$gylrVRZt}tnWilA{!R4Db3 z5U$OtrE?`xRv}V-LZRks!L(--lL%VMJgV)OXIC@#3UX)Ip<^gAqy{|rxs`igy~*K+7d6dXMcKwHNW#;T6yz?DZ)_#C?|nrr zhYpifZZIjWTS9C2jQz8Q2yK@9jV3D@zF!(d4=WW&`?-i<81|Vwf1Rbti8rZa_9a@7 zRZ9D+M^Fo&{RQ10Pu1DUFf!ebobH|ImMmsH&rUIs`T};DcZ$YLxF#fDy(rAeSHyuQ zf7y|L(ojl!1*Ib&(ckd^Rg*uX`+E&CxD&?R*PY%JE+PM2F0}KL7A4(KB-<}6UjmIR<)>ZvI26`*fTxG<(uS zLth%OmHX-3X44yGJ0&Y?B5C^vU7G zQBwrH9*G5f9~Uhufx1n@=~b^a9Xx4Cf39;sMbQCLO--b)awRlIIfdRH45QBxn~_nn z9w!QS`!3SDN$KPmy_W`U-A|*tgJ}EEk9b*kAJ1FP!cd#f zMxV<-LvpFOCxh&ko|;Ix^qanaudmonn+W)w>dQ68+Jy{LiC<(IFd96 zHjSg`zza+I_(P9Oo-d=h8H;G~EJi8wZnH|e!)%8ADJEK(L6%mTbhwp!BgRLOr*{A; zC~YS1#ueP@r(37uP^9n8VdeF zqQafWZxQ>XmwT|DV@KUknp!)W?k*TlPFYdZ7BrV`Z25)fxp$acQUZHr6vvwUgHc&I z7f1dZjl)Wd=yIbs9U8ot=4z+FCipzsLJ|-%@-1uMbe|oS{>o~CuTk8%G}?9}f&3Q^ zpdp-jKToU}Pcsh+=dweDE8T%YU|xvj)i38QANOstl&!L&%K1}CjI;6*1~{;hQJ(DF z^9Uvv6OFoA_K3A66#UL%@tan&b45W+^YbUj4|{>{D_-HsHVv9`s2%4k@1uHEE!H%h zh4Rpci1}8EXQ%FC(coqj%ici8r(8txjQ>~LrQCC435$aHD3Crv$xYD|-?f##@H40C ziwAfiR)&G2GBKWaEWLhghEiJ)#P=q^q-YP$pUuFkYo~F&=^;Wxw+Z<#+yw`v*?iva z&631I*^1hOY;|cnZ2POw!rk_jZN_Y0&I-0uZ7O>s*N8~b`_MXDi#-pYvWiwY&V|vz z+Wi~Y?T|_Agp)NJ9(V|IDlW~!S z;V`>uuTQ%?2U6OSci33J4acXQhIof7+8w&Efu94^Jo@ls?t_eOf%|FYOm9m1H=E)P#30Z%3)5nA5qPYZDUaq3qS|h@b<7jCKKKcHmn(~C!5O00 zcOrMV4=M_>gR{;v;@&3 zzcO&1o6~_p_V;QLowtjmKihZFX&*VEZ{a{8#&wX;m61aQv$9FzQ3@3h)lXKLqBR}eT!mTaaRt9g{8^%IHtEvRWPLtyVcHpAcn3onRg z7nO|g$WD**f@LsXrvQ!9PvIngzlk5Lf_U^*oW5^|kk#q@8jeBo)Ho#SYSNv^cgRrh zz~qVFKyNRhOT7X{=8bHW(;ze6F`y%af3}%bx9c5wVN0_^GJy=CEMn9{>y5$Mf5q*FZa{cJ^ z)I4^(A&eC^CNSUFTF9>EoZopZxanKO;=Bf6#nF%K(m+kP&9}nZ22B)Q2*AL#d*y`p|F|x2DhXNyT86c~zZe{uo9U*C&y2^hEHKAHsd*VY@IM zBaYTWh^<3Zn<6e9FoNTnkF4qFb4>AljnNkyVLRakrrnjLMxJjApCC>THH~Olqziw3 zp;5H2vEuci56V&W+11iA<*id>M zm-+mYP1m9ig@z>JzJ%xU=irdzO343N!FpRxvVELszHa17;gc~D+?hnTn z_ZT=xtVX$XFa9i6p}~q8w12A&wjW-Oe+L&}Mp_3>8ua3yMLT?t#$#hM&yzae#Mu~Y z`aWwG={cHF#7799$@~3%?$kWT#E#!)iw9R@pz%E<<`!X`?`r1jeTbdQPGQGiOF^bq2GSFU z!0mgf;G1w;@C@h@&Pi{>36J+nnvo9y_<9&Dn_RQXiU(=Mya%bg128C0gv6cuU=;HQ7YE94Ub7@M%#$NA&RI2dl%e@chmhztU96X1 zfPL^rlmXAqip>lgm`etnoNl8nnKAZ&2 zyQAovianh&b)YcY?bLc=3)L8;Qbb`r-Z=fmwB1#B|8yTTTEbvyv>De*e=r3L&cqdr z;r$=K&fZ_g;w}j~affrS%Hv>kDFpr>hM?VA4*{m?$Z0-}*hy#c=+kdh58y2R8Lyy_ z{sgD2XF}$;38D*svCCCN#7 z02}n=t))S1v}Lo}Pi*Mrd0P+8fY@+{Nsd&63(w3wd!Z<38)yng+_i-b#rCu%Y8I6o zGo=Ii69p3^N8y$68X-hX0)7?4G4z0d2MeyzCKUIg6L02f_cp>fi?-Z#uGE6OT zLrT|f>{O4&AjwRO*cgnl7yFU%;yd)SJMcU`9-h1}Zga^U3tt4`@mZcnSQ(24Ira2? z<_B_E(?W^{ywg6>hWAHg>CoIUIQDf6n&g%s`tK-VX^Mi7KIR`S>R(0Sb(?7H#pUFe z`bbD~sS`?+&I{g0beQ%^O?K?F8}H-I=X+{9vg0h6y|)4XUWWBU1bjc~Mix?YX>o}Q zP5ESp0b-um(O>~}yP-7pks(butxvvIV`$mZ-C=|lWum6O6SkXsaHqKydZHsKC~G1$8TDe+H4DlgHG zr~MV9qtxj_tPZt^%;z2O;cyH}VA@A?*@6qk%uIh3vk%kYK0wa-3$ezir!B%vvxmZk zA!S0kztV;y3Jg0t#~vxosFWoxmQ{D=8HXLL3roV8v~+yjE=G?xc3{b!Pf%#NjwXi&eDOYrl9FCa z`R@x@z)=hKa9tAlZa+rK6*DPu)f`I9n?^bpJ0U6iojHXTvWugOS*~>}+BO!z=VKi1 zHvNLX{ZAxZ|Bg*}bD#r<&Jc0i!0r|WJyJ{qXq9b zA%a4}3}NNc{cx%?MYsu%OLskl_0$JQiFkq(?O5UsP%?<0PU@q+Fh5-t)F+u^*{gFH z%U7kr|3)Bsu5k*q=n)#?kVp|1QYl+> z0NFWwgYofyc$U5v{he3C=dLLSSI5 zAbw^VYo6L+xqL#uWia>MIqv+5#vgw$HBKB;4~QVt!;EK)zv7O0G0M~Ub3$b}&dF`Y z^w!PTdr+UdJKkfG-VMmCUP#?pqi9FC49Rc?$oF&V{jUrr~@djcu1Za~Q9Qy43yMdem<{P)H45ha^= zE;|X9;s3!Uku#G9Tk-zwaJ&wU7CyaSEvUxN5yE@(c)lv0<&=A}YleTYd1e|s6ns$h z#sD(jiy``tK}KAj%r0qBldmql>#*T|vnlAV-Uq)e@l2$!k!_SNV%sj9M2`0z%pG$F zyXTlA65bQ&kS4|h3#ZSVeVSRe6^`)jqR^qMlIi;jftP7OGZ zbHElqq)cNjdRWZ4%dTkr9OV8Mbn1|r|1AnrIn)`uWSd@JiZbrm~e>a zh&GeklMwo@W=4L04ar$UkrvE$fb;Bqi2tw_gFUaY3ht90^-}^<=Uzg*49{@hxrztQ zk=(g<7zY-oau?fd7V3YN**wW(;(X@SG%g7m%Ew`}T7^>F2T<#wF8q^EM0m`0?DAWK z`Mz`M>BnV!F1L)Dr2n9ESRdZKY{z>OeatcxP?0H*iM~Ldb^uxj1mWWCGUN`-#(<`0 zP#PzWOTS!Un85jno8@VYku;Tz8%E8lL{cVgkh<~|n^X*W2e}GG?=rF5-xgU(gvQ_Y z=(ACw*B=JZUGX01ZmOXO$3(LE7DPe3%V1Jo43*PGD0gbHoL}|WGR&O!={|0yE0sej zYpOuKAJ4L7;m_HYK^-h`_iVx3R8uh1Qxc?`Q{dsxxo}aX_%+^GxOmK5XcV^*RLtCP zE3%gzt@zGfmGmG|@eUkbyvI`IERtDzloX~UQ`_}17+Ru&g(Jpbw39qFSUyMLqh_8* zIEli>J8(6*iUX@tv0J=_=b(6gq%9L&Mg=I-zKOXSroxXrEn!BZykNU=A{BUhkVg0# za!)Wo+IL;#3Y;epHb>C(l&hTR*zf6d{pMmST0H~)Iot&m<^zAVW8~hIL2i2jxq~|d6KiAP=)ui=W8dR6 zfBp+D*QfBo!%0p;lf=0@QFZhTo^7(Djn?6$rVvR}yFzKi<2~e1xDS_~#GpF%ChiUu zB?W$7dDkpLpC3IXtEH_pZtXLgmXJ#Ce6P@@F&W$mHG%Z`K2Y3l9F6nge$3@(ARC&7 zppEvdtl=`-(pAR#$7Bk7oG%F5Kc3^@7fwU7P=MJgc`R&+gtc!vro?Ar+_2B+k&~r5 zc{`fyHI;7EdeTPI9c0=f!E*?rq;Z1hv=%&qUC}ob9{-5H&e?QbKZk;>5{N_dNPO}` zZ29>aHZoeIHba*j=UdR2-uraCGK_9 z9pqCIOb-@@;&FE}R1M3adv+7*)XpP9^a9qp4HUvtm4t`NqlMxs7lhbvfn>EI4*cZK zt*R2F|(={hr!=>9Fi`ZAfTpGsHyH0lA>g+zOB6upHT%^t+y9# zHlNA$(>rqP{7e7b-$V3(80~#8L;79SIO)Z&rR19s-*gt)6N^!{>>>&i2a=1E3@I=E z17%AeGCt2}_&s}C(5*v?b5v;RUNf5B5sW9Nd@-JLE}A#a6vmIAAq<}AD2z7TF06m+ zFT@<)FRUufrpt8|bolx`+B0SeWD|VR5d0rZ_BFGN&)V?jJ@k{!-Xt#)N|TQqqtVxQ z(YITvq{+Kj(bst|a8N2)-Z)NIOk**{IvA@Rk0bl#QNA0?AjWeaerwKCK?&c(2Hd7T z?v6WpX9&3rH6odhcadE75MFCqp}STeE`^o|m0yHKt{SwC-|tcz{$b~&lUP^HeM70I z5Ykc)L)8w<{LDFPGY{ds%OhMhlA<4ecTn&60!Q3mLo!_t?|!P_?yB(^KI1y+yWgR? zQSGD>xq;l@>>-~~`$_&2!|uC=@TQrNI5~~B7kJQ*8W$SRnI#_;4QbQGktC;o527EQ zLxbm9UPf8bA%*1>a@K=tuT7+Bhi1}y&W0?jszLLY*I2}vYn@A8;)m=v)O~w`mHq=s zY`Z>Hx{alDktoVCjHc%+(r8(`E}7f}EmfLF$9F`qD|?gKpY$vy{mzp-;v(tyh6r-W zv&MktiD>kmj4O{!goSJ72*3JW1t$+Ty4mJVmyd_jzDehVwTmtY-5c_SuOHXL_0A6H z+a};x(Ju@iDMJ;@{y{^tf$U}5Y5k;5I`H{8Do&d0j$SAkQ^w_2GQk25PsN zO;StuQ)KyFB-Oq`COd+O<0indkMY@&1^0fL&@pps(urYIU86-Wx2jOp2T_Wuki&Y< zpKO$cD8|Uj3PFpcg>v5>%91GH-ncy6Srvtt8D7F6!==LV3J*d3LXYL(=$n?ai!WIY zZAwGv4Mru(YS`9-` ztAB)FADd~M%o?gum`Y(bm$4)7ELQNWxQ4SDT@X>G=MVc(5U~?}{46l8W+@VC<{|u& z6aGzI1)05u!h`MR!oYK*g@U%Za5z2_;J6WqSG;>EbqvdkE+F)v83x*pL&S^8cxq=x z50_Ywc7Prw2KvKgUMxB&6kZQ}IlIY??*?YW@5xh)=I5c2r`ut@nIIjZjkXn9=$~}~ zbp`jKGxs&-_f)WD?dj}qMJPKG`VQ&FkDzqwD87okV&fzgpt)%vzP}zvsh2EhWu_|4 z>Pkn}i_>V;c!;Qb`WRI<1GirgZjQK0wV7w=#qkUh9e4@nZg4(;(lwN1NWx9^4bv@8 zU|XN|Tc%4cV-m}A*a-cFXq(47aA&>wtR-FeoSrUB>`xQ!oz8)2{SCy`W?=iP;gpxI zN<+gs@tfxf7Rtw?`ui%hOx}x#qD0*J<%emlv#_RSHa5F%hE+iZTqc~w8Obd;sO^QP zKK@WRZh*|b8dQ1YW8czqXrAMOpSBCoSUL~;-fyKEr(ucf;IS68BrU-Wz%o5tm zmy%2QdV1*TPVb9XqO&##S2RNqQtQt9GEtcN*AZ8~ucE6LmQjP?OpmgTvb?0LY=M3Y zvmY!@$BqA?r06vUiP%9Z(F4UJol#K1_o_KP*pVhqvz0tZUMrB2uE$Ws_z39z&E#&` z0vLK#V#(@Oz7m$y_~X~tBV2=V9v`yj!C84NMprF|-U~eh^E_+$ zO-Zb~a~4tA+0a&agd2^XSYy2cGGv3)u-nYQr-hY9Rwz(To_ag+}2l8+~^9BwrD#4SqE;8(|K3I-j{-A@eGCSS9X$jKWhT+NHbmnYHKJ>u;9bxY=27R%l)TaLWcZ7@`cLaq8f2E<=$s$#2cO3aJBm)PWc3P1yX1^TJxBXEbJrr8XZ#=UTmi*L&eV|QNZ0hH zQfGoF&C(H}uAA*B__CGUyM5{V^&l$7KUTJ9F!C>cW7{2VX-SbTRk(3J?w?%joW3)AbP<-h9Ym35I2Jd)ArbAn0S@h5BD0g6G(%^i4 zmOu+ko@Zc%aVwG+)IxmnJrv(Qgm&%!U@`OvWMaiAYw}lIw|It!@1$tpgg%tDzrmc3 zved)AL$^?gF677Kdsa49J&uCn-Wt*|DJ0X*GV=2Y7sU1l3vZ4@3KG}T>BaI>beeNd z)?^sN;@)IeHfc$A+1K1qW@@=0MH&xvb{CJ;nl32!oh*YnBcJ=YVLlhnV~{1gEdLhM~6IC+@g7! zcWDsMbtx^s&z5uD0E!lrG)Pqk4P zGa?mUok@uJdm74KF{C!)Kf3s6KJ7>u&RKd|v}?{6JR7SG^$H36xx$|t3;t1!^B2bOrD=na7HyvR9ap;Jgw=IEg4Ng&LbTUnSyg__YYG6GrmN z>;g!-=)uPA8`{4=!S*^uD*Y-2-4acRJL%$P`*~=Z^4*$)2+cDb%bnu==v_FNDhgz2 zpsFZ&X_?T{cRa7sCqpYUuET#vG2*i};r{g{NT`2^wGz#|Z=^&*e-yplbBLZf+fxni z^ELVK{L>jZ`t@WStu8hvQQo;iMj})LBVnhch?Hkj*plFIw&qs}dt2%tM9#Dmt}QSZ z7LO=lPSZqTI)*dtoYgU|c{^%sF2SXqpDhx?P`Tt16B#B;t!`#yF0V;qi>_fq&nv98 zx`5OBb6{s*fwdo(W9Nq>|%*5`ALmE9VS zlmBFK{zDCWp^z!Ci$?^@r(1=9w}@C{dZNDCcP)B^jS?Y7beibeW|c|SO%}dXzX05CQ zart)W7C*+VIbqnarx|HEgD7qfpJzCiLe*v~zP>ub_8lEU(Z1GX5IUSnJQZoK+Go7# zmL*ZD#K{X~XgMT7()K6NIg)2AW-f!ovmBU&WWZU!6e*m`Fsfua{WA)nHLn#YWL*zh zyOc0Lh%?52*0O+=p3GTOokF&#Q-OpF`KpdV$Nib8c{~<<*M5+}t#?!zt}1M?n?cf> zt!d@D2RLRhSLjwY5f0b1(2MJ%;nEGHjh_sy#ukLrYskfQ;>DRbTBn^x$-aKHBToqv zMlL{4%3*YzodJWVeBWe#6>;^-w5$IkB0R+4m6d=}rC|Kpa2&QXreL0=EoQrHM)QSC z68m|U4sH#j;j`6fl7>2|3>2p(&3mj`GK;NBZDuT8MsVLKCsb~e5~gMkqGfmZ{v||; z9RD2``uh?D*>?#-MCm`vCBIHvj$6CRa)9n52n)Z#VCg$J-#p8n)_-NwRqnC+_6C%t zB&}T=w85@ECQIzsAM)Y&Rg&|+#x)&3lPSpoMfkNl``bz zveoC;V142woRgl9rmyob`n@Sut{H>B7pKru`8m7;JCRa4IX`A0=g2f>Kf?K%DQHkMMUh@S#tWZrdqROd3DWtF6P8$)vbiB&n69llZhjd_=PT5xdRH5)c5+_k>>?QD-++yNH!Bq*RLyt6Ka&E| ze0_vJ-yEcgvp8ofY8D23ABS@ZJBIJS(>O7`a~&=w#L+czZd*(<&H-y3(YVEJ|%-MQPW16C6;r#Ppcy z@H4WZ{!V>5EN4g>T5)uF(|_bJ^(2-4QYXU?5>%i!kT&0o!t-SpI4>X)8)inZ3dvXO zi;g6oSsy|0h6GHTa{*(AEQNK!W~e*+<7dnm*!nXBZ8N~lYbD&DREry8Vx+P%7aLSh z;;Dr{OnR2%@~dS$k8jWW;-hHsCqsJApILwDxzYHdndFsgMl+s&Wt&o3nVeMud&?3y zLo@>l<96eAWg)tz6k+O>+gLkF9)5M!IC@VUy>mDN(Z`6c&X(Z!LJ69;@GjC+8)1Dw zo?@O#(8MqqvXeXuKmW_f4dZjhDE^vMOVJy95qclEi-I+GQ#og#cNlje#D^9a(Wkad+Ex%^BdM|(!&u8gA2 zWkkbWXF}(;U~04)1x8Dfg})AcO+QByN-xn~j|{3znZ|PwhLr2bvp=h5l1rEe z)zvJd;>jg2EyzRW;2WrY>4eZECk*m*gx0HM;nSg$!i274f=rDBM&{MCeW8-rl2V6e zm#4V>vX^^b52JS1AxzWGLY724Eq0Bew;_@AL8!;izy^e}+o%h!fmv`H9^AUl`><1} zdFwP1t2d>R@)J1!`Z%Hyk6SWwq_kU!rc@1~uLD}}G4m?wFXUtRqb(G6XfwGqg_FiU z!Wb2QEOfO)@Wpl1q_d7r4RWQOCkN69k%2rnV?fIN$GC&;2*R~aV|}%(;4}ReZ685dy$Fd(cM<2B zjXeW-*M5ODC7ocDIV}gUFLk^C3}4G3%e=*hRsWP&d&VT z!@f=3P3HoGsfzQ^}ZL75XAw`P*0)!%i$z# ze3F)4OQ8qz3Mi$xo}$0V)2ujAlCY7X?MAO*rOBo=ZB0a3R0?+s zeKw}7*Cu3UI)qH751}pno2hhxIvEW%Bo|#V5|eVKcO}DV(ll#o4>Tv4VLIgaaRfmeAacnBAA7Yeqo)9Ay1V-(y|NK14>h37-} z3a&gTIeFw#{FQaaC;r;!S#mDGT`{_!B|+l{wbRE54`>C?cFo;29u3ykNaO?a&}kPi z#Qi>=d}x64&q3sKwilB5{g^V~11=a=AbM96{QR!q+Wqf1WImV<^oj@_K7$2$OF3c9 z&lr;4>QCMobLeB_eC%A3j6%&bP^}Ob#QO#c8VM3YPD2)12VWq|l}VJi^Q&O{O3Z3` zuYy&HbCu<=P-nJlc?6sG=rY|Jmr7N;FHu9naroU#!QjFjSTI159J4iOL1Qn*D;N+x z?ni$Z?+D~@CePeS^t6L}**Z$0v26-%978RWw=M zi`lJCMo{o$NY9)N|Bp>D@OY1nZt_CArKB)we+|9V$-$GMs#M{mLkoF7sAW3$)!uNX z@$&}KWzRkc!}>9S)noP*QOb(>gtuk!LUZyJ!RF5`LFVjrCeinnEgI3pZe1%y&Bq>? zyd6qTudd^+-!o(@RU>4HG+Cr+kl`O)YACJ4S5M9fe$a(4TH7%pa6itqadymZ2ii8x zks1rfkz;8x%sDHoWWhhYZ9YMzuTv?v`5-x_enm|gzix}vNVWPpnMFLHi*H(K#ZU!O zdZtLTT7IK*Y6y*gvX34cEu`Jo``N`m8PM#PqHV^f(PR4nwlU{XY^fo%i3}Ca4Hg7r z3w7F+K8kMN5~KE*aC6S-ulqG*C37G~hnZvrnW#NQMS;R3eI$GB&6*X`Yh?N;C+i)P2rrL?a3*8B!>UGG-`= z-~Ig7`mXi1e*gXMTF<(FKKJgk&p!L?{o1eB8_JxA&v)AvOAibGqz`VH&1etrWA1Cj zGw-J+Gv5r<7<+vq=B?6v=AxG|#lO^>@|}K}syDpPiFdor83`KXoUBQs8yhlcquy}3 zi_ed`_^zG0M@Ff&CnM;WF0r)1!8rQ0>Uyd;&!1|SucaQv2GQu| z()M45Y1^wX_3=)91#^p(&Ns+p&Ir~2_NrIhfE^QMQN>Wed=w0Y;Kd6Nk}cbYSu zdqkH0!NVpk(^H|(pV>&6Tn?f{&BCd+yaY;s$14?=)I}}R_oj7uq-@4NOy~pVcd4aD zrIcS|2sK^pFa1|}j2_YLp@WY|k0hchFNO zn&`or2lUBPam<$3WTyOM7UNRTu&D9n5^Cn>y_A6aGTP_TLAvAq9@<`0ni&|=Vg__n z7>kCxyj*_+I+I>gFrV%&(W9^0oZ}o_RKz*n@tb4mqs`RSYcaq5 zcsvd_oq2VDBeaB#Gkq=l2BU3xm2uF<%&iqq7oB_(v?yzok8|<15&gY?K7BrLKFtrj z9`uH2%I2E~wQkB?O8N#5Q|Qq}{rS9_GCiA1@n@Xo=~b_%bz?Wu+UE|@%9ShVr`9~a z*CT3l-=&rG+|*U{xPKZwV)B)`D8f72Ps`Ih%>g=MjUXKpB1R|XT&MVXGZJd9S13=1 zBHBo!lvba0fxhRUO?OP4OMCw|qHVtO(Sbvv^zL1vv|33Zl}N%V$WEjBGR8RHF5TsX za1^KvnL~`^Q%}Zn_A%z=Nl|KU!fa{=dzx~a>P`83J5n__6VPR>T&jO7KJ?;PzR z-dy0*INI>V5&9sHC#ZX`D4lvqo&J7nIb}6Zo06Jzp>#9H>5kM99>ztKafQ>=iK-iv z*{WI!rgl?1JR7Kj{Z-U*otd19mAV|QvxM_xA4NS%@S?sP;Ne0N2N~)3x6I}_znD7z zWa{YB%hY5!mrA~q&+9|w(+$y?G@p+K)&FrHB{5BgLcF?z`=3(E(HT>FbskY~TskSK zKX)jilu7wdEud6e$|;#&5_D*vFrCTM!Bd*uLhWt7L*-0js+$u^?T}haUHB48eLq!4 z&&?uq9i`rJw^R1tw^Ik!AE6c+=4K5xswX04>eqZiq@eMTaSy&{IDxqrZMLrRTjhqNCM~>6Z=5=$98O=@$y; z>4-l?bnPQuy1Gf1j+UHH(<{|D{_j_DwEG-5{UXO{sSpo(R$nq*;KtJ)Pio+_y0>u* zZPzd!$2KtXS9UN~^G;I3JU)ZZ4|A!Qp&(|{&rl|=ERng<=}Uc@wv94<6-)8Wdrldv zby6=~pHnUiCuvRRNxHK02OYU}Keb+oOGVzPrM_ohqe{8uRITJoYNB>A{jyP)_7Ik* z-94kJkw1~tn#Z2hE;(K5W6Clr`%x&h=<6tb>WDb=BVUZMSpI<$wi={%>viyEV52!p z8;)_dabh`VRp&8twrVl-J5|O-BZ^ibgy!q2pdTL~%(oVn`Ssy8V+sX}_6l+rJx}de z^rg;Xg&2!zN`gnN9E9dr60lSV;Vn-C|ArU>=S8)+-aBUv7kT+*JcxDfkAE8C3 zALXq}xg>qp@+60ZobONPI3uXPLJoooff;THD zuL)ibXv53Oh9#Bd>>!`Pe zJ1D1%2dEcM6FL3znHW)`|Kim%h+{#tT2`yJ5omlIXc~qJmnR1m8MqJ z&{3Whbo@FE>T;suYn_pz4DHsBpz( zYENP+6(h5UI+MJKO8ss^9TfMcTb%Hs1>j>&YKcG^cuNN)<=+N;OAFCC|!Pw3IwGi>RSvj^#! zCHxfTY06uhX;NrZjc#Uj>79E`Xt_%%wAn-vO&Y3rTrzz0(NZCLkF_w}FT~^h8cU)E zm!wdxmMb{+0XsRTf5vc@3uEf}^G{TQ{U7Rl#8PTew+*E?c8F@{)#YP&e7tfe2sM3x zpZ=)cLpgo=M%@nWqJF&Sr(%D7r0(;uTEALOQagFNXScq_(KqX3>Dpy}^d^1>I!AFS zE$~*G&U&LjZBZAe+9!pnx8`M(nEoH?`7<8B=e8v3(#^A!rC2*<;15 zhka1xNcWp@vIlHA!hwsKHyiYrVHcXwoT*3k{#{L_E>)y5#zGlYr&vZ}Y6|1D;S@D8 z!BCzSi>M~Zr+n5HP+7&zl-)uer;=X*{g*$Nwr*cV9Uk?j@@qn===?D1z>|yA9IA$T zm|;oZ2wg@O#EQ`y)`*IYy!>s(r_ay#zh}c5K_NaqF)_aX&qsl837@mK%RaZGdkl9w zA9m)6TT={8EnHw}0esv3|9-9rn<1kCn%3JcjoS7sfR`a}p}k-#ydTJg@v51q^1~U} zSnv!YeM-=#Zi z>6H4Q_hIi~d|Nh5ILCuq$OTwndK4YLnSgd}pN0IDZ^O?Mu~0U#26cGPM|dsX|2@kbTMZlk`=?I6)$32lLL$m35VsOcA?>eX4Ov%i{2RzD58 zt|541;AL#{Llb{L+dvgrHo}+dh9K_I!2I7k|M#s$w76DgNoaI+2D+bU3Ufq*P*`FP z`js3DJ&Z3}R$qY1+*+Y1*c`=qUq;ebUqI8cFf{pi4Z419Cc3!a3SBdHM3?02;2E4o z5g$`f#^2+h@-P4mU)+zDbR@vl@D!97l7|Yt&9IetGV@`{CE9U94w?^NMD8s%Pz%0=kQCYd8maxeYk*f z{8oXtM?8{htw8CQZP9{7GH89G9>nah$C+|bfR}_KoRNZ_sh6Oohbz#9$P{F>tqN+t z#iLyoR_L+3HsZS*kCd|(p=A~d$Xrn!B{b&4?X7Y+o;r*Q_xK`6r(Vkt;vvjeKOcSQ5sr=T-y6<~Y81Y9_J zA90swp|^B3ig=ZQmTIJjJLD1`i~jakpfUMqWJF#d99jvVE|{W>YTk6?fBgTDS~R@wmPDJ1 z&q7+9(UNVCH$G^cW$vS$1gbY6f~xvKGn3 zhoNYL&+zcmQbd0jLW{&?QGLo`bTTXlttyy}a#ouok8pc5Qu-Tw+ZLm(*XAMGxD*0+ z??GpGgraL}wjgcTj(o-vQT(&xsKyS^o~74OLVy&`8*ya^jXa<$#0fc0$Rc?WOSIe7 z2_1C}LsPtGBEv>6V%k%y@-LbqyA{7N4rf9GlC$W}Gf>1G1- zS0gs5cs2-lO+ke_w2;5VOL(bP2gXX`cti96jDZ~J*{s0U3k{j~0YmVouNW*2&qUAa zry_%um5fP=Gn55#kkIVAmCcqaXobHRV0#hd*ZLSf?>9jO-@Dk+OB6HkqXfKGw8O)v z0>~k75l++cMslIc&|-rg(0ATUUmFjAY)L~HyDiN$*xzM8HavjZMQZ3`UnqQd(+Ep$ z^|O_s&K$ie(tf^)Dg#kv4i0R|BU!`>?^`SFGTP zD=?gRm-0jh0SgPEDm7tN#=sH(QjkGER71dgF&9VBmE04*su=r)cYyzk1pe>7dpdgo zvN=5!g=dvO!q{wl?uS13_-$YUZ*(%DdOS|y86}XTi?QE1ZRohR2(r!}h9GfiG|lTd z*Iv{QuMxC|;HAdsm|HOx(hX-#+YW$|U=nBj3`_LWLl=368)1dN>9lL~E_z=^6u!Ul z7=HZD3tHBffqr`=&TN{3#H(!}q)Y{^ef^mUwbX)bmCwPkpAU@nPQgU0He~WkpvP}N zvRg%s5w#>2wCtnUEW={%;(Ieuje!B&87t+Xo@c_E%7sw!IhNTT@sVlDZlE_~X}EZH z7|IXERwQMX5*+zGbLEB$Ro5un#`q z<{*dG4=FNv-u5JWRw5qhKMA2NnOM{=5QgXYv;4)|u=?Sj?7ES7aDE@f`am=6yLN}U zab*-{WqxD6%1?m$LPwAcDWGE>E(Ge2Fx(N%#@DJn&}NMj;1>E3zF%AfCYkdR*gAB`@OY=0GM(BNCU#%0bG}8fdl5U@D7dfR~5}k9AE9((XDL0|qs2l1uM;HUXmp@yH-zi1H$&PaF0)=Sy*ewjBZx4a< z;w(7ZtAOU#YanN7RVMRj5TvG?qUr8?VX7X_mN%rBCbEtl*<%3Kdv@ZXlpHpc;zL7C z>Ue%d7yDoTKjA-{8T%y&I(v(7ez!3Cx~~J=p5}Awo8K{R>vF)iWF8zm_ZC)cT7*lN zeg)B<4(JsSMuRfRu)0VIZH*aZZN;;IKR^fd_EtbU>(GPA(b08;K4tZT!gdW7I z!ru#(43Q6lv%#*|k8eJ1)0c&bqWNeOk5nx)Z$7F#eiw@!`vAAoF2Nd}+6bMhS@|n| z11{0L0orS}V71~5m@%h^8MpnyI>}D*dRh zEbwCrI@u~gh-(BKJs*!x%G`v_-LuftLKXD>s}Wp4b?|4&cI`v2 zqq1T%qW15B8GH7Ea-bf0e(nVM@$4h%6we^LUujh}&(^Er=3ZfxR};K@{ao^Ja}%EZ z@jFhEDI^Y_tH`g*#-vZyfrQ<&#eWOOSOvbDxTaMB^W9&KTkGnl)|b-A8g zXRa}A%aw`SbSok@a-AF}USuNEgdDV+!J2sU<7u5@}qn;!-Jz zyjDwgZLlDDX)c7aY$MhwT*=kxu>xie%b$meX)5G-)r4RYF{Rh49Q%k zo(&})o34_}6PYmH<3vJDy75avE5hXt<6YHtQXlh~e6Bx(@6_s%SsHz~ zz%hi}=T&^RT#O-KZWxd_TYgg9HJkj`53@ECtWw@WN;<@<))5&pZBG(8tWZG|A8jV* zj`CGGj(sCu{NL~|iFQ2sx|f{H3&$T8zh-`0ixAnRE~NN<9NsZ_ktp?@BHtfn5joB+ zvV6pWjw;N>5SK}~qoR0XZZTQ*<^4bJoj4Ou+H}{ERYLcOkD39oJHL}8W%LoFrp2_l z`4tl3@`~(U9E4ZiEg{b$%gLh5QzThPolB}5L1S$o;ddG%nH#?leJwX~=g2IgM^_P< zrTOf1$0)orz?tAx2T9m!PqK-$ky{Ju@psG1#2{z@AOER9gsQKT^thK~ba*ovu8tv2 zhXROclq-mx&?dV&o3QnX7@Q>SiG$xDsg>>F4F?%JagWTrt5MSQLZCHkcCRC}9Qetd z`YaEp7ZRq{VFngGSPRMnm9We4CfrdQ2MC;jf-8a~alV0$MqllbK5Ke4XO&=y_Vp@+t1JY>}KbkTZO z0rW9U5I$i=CfLi6aZMISS{6a5aAih$6v~8|G^fuVANsg_sDy^2O+gJyE-S)7%N`o=I zOG8gsEq6Fx61%-0gpq+uQ13DXVXemSq4pNs7ZF1Fy^_FHd;#C9&qMr=XUxixjri^M zvpDXW8iRZKAeOFUMs_s9oaz<-w-!PFi%BQAhe7944D{{vLJK36@s2_%;pD|T*gxfdK(4oL1{F$btc{xP|B7w)6kU%H(8B&V))pc zZm{FjvXcWFpj7-b_`bP_z5JE&Q-@%#?vEKz|2vmAXV%UPW+yO<_&h;<23VyFN2o~e zUZykWDl_fGLWmub!np>HOza%QOp57I>OFd%lN-To5CIa$1(pk0LDs;f7 zQ>9)=b0~;l~MaO8KgC#U%9qw6fUN>^7w-Mz-~=3Y-rgG-Bq=C z<_96NCh{#EEoKaR1NvaZYZ0S->kb-Q?y@R)}2uQA;6u8&OD_-)cTA9_`UcvW{EQ+(BQcBp5;_X6EHA^xL>_rh%RMYGBa?~BllZ45);p^Wwq>ti6gx|q zV8v|O+3hsDJx&U1)aFCQQ#ttkZWa%?yB!PiG(X(o$_1Y1ZV*0|u70xqc8E@xuPx00vq9wO&?8{g*lCzXq8@gCtpoZ4>) z*BWQDR)*F@;6?yxU+P7ylBN*TJ+n#MflsVOn+5syqaFK=+^17#e871s!Q_M1PQn~K zOm<$;!CzXwG6%{RLYqJzZhdWw%Yp@nQ0xJshP!c&mj-z=JD!wJs*%wPhQ#Sf7OO9L zgSpQgXXfMxk-cpVtas!s#!(}a*(c=+{0-fZ<+O*Mx?KtlkMp4!)$@tH%zeDB=>is5 zd6@i7&?L!!NAWKIlMs@7m@V{Nfd76O#o;Yk*z>D0&bZ`F7UNJWz1Gj3#L zlNTwQ_W+!>{DP9|M-bmc5=-y+cXJm>U^5~qihj>R)(6!{oTMh%{CgGg{}WHt z;%PD}Z$h-TWaHz_s(^#*Au6elmG&%W65p4A*1lHe61{~PIk|wmpSqo7uzDo)?uq}} zBq-2;%RDpE!m)0j@OSMa3!6Off{#)7vhF4_Cs>=wY*!^oo4?`S z7yfvI6_EaPU$QDj2A${O{EvR0!)`n*LaKe!acbskES+w}P>QnfEx-=1P%6VZf~Lg# z;4$L=JCj{)S&IjkEh7c`(q!HCN<96`8*F71&RnaWhGRwbaMUSteBbOV>&pFwdnzPJ z!SH^PX;DNj>8vKl+_H(@`p5WbD<62>O2X~4can;TGu)BK5=5YEF{xQ50&~0a@blZn z_}=P;L~Y?ZQW&w41Uj4`d@)XB^~`rTuH`b=`>B({_*_7KNW+&~tFXl(VdnJMPdv-_ zIoE%mJvP>NX6AqZa$9u=4|_$B!c9fwO}7A(WqzLf*Qb#!tCAwq{{|AH#bNkV`vaIi zzlgEzIfIkaW0}mRNXWJmh0ax4>}89&ApZK`Kl@I>?3`P#%DtS{zze|;zPZuX?uc}}upa^Qq`&c*2U-;fh;B~dpTZlqzfm1 zaVLsa0p!+BIdY(li;K55l5K`{^ z3QO}%Av$J<@F)IWHe!Vss-2PmF>SfHbjp4#m#_~*{yE%zS{d`(Xrjg>X>`&0Jsgcx zM`s??A`n;&&SIHxK?5MIbPztS4C0+fmm%9@8+$8h0(6F!pl3q&z@Wj7-TpbBd;hO3 z_)jkbnR9`S_0aF8jZ!OCAjYPGeWkVvDZi?PM`>b6weJ*!evM?-K9~tLSrfEqYzrH;V+T@u zejYpzy@pHr2I#4|0(yNf5p8yvg)+o?Am;mQB(iWHJ1!-O1e=up$#sj`j>4_8={R#| zG7k3{hsF1b7?J4jjF0nOi10LkDVc9@ohJ(eE-j$B`2uMrHptv@jPLXN|ezA0$1*$w@Q zvVelwoftx-BTV}D~osIX>JbrSc1Q%38W@9)&P&+(!49YR75@_x8eKEli4*J5l&LxFCh0;TZGo`5do{E^ z@CM&6qA*-ijqM%ckm;2bthS>etDG?(Zgs2$j%o%Z`1vEfmupbgr6+JF`5mt4lE8;^ zjqr~mS#qyHSh<>RCmipf{J4D29{bJ-eZ2 zMF;nr*G~|i63xy_eS^tG6Tn4kDc| zkC84DOJd)rNbZcxBjqLvB=}J{j+cobzubp#qvUMjp}!IbWC@WeZe}EU&rFuvpN!vq z3?m=(;+Sw(9on(93D5t$oNT$EOdMh_;Jb|vSg$;1l3*`I499D+)x3EmQRo|vF)ZhX zzHEjT>qba8tVzUq=MYy+o4m<*k0(d9Nl%D2naRy%-&HrVFNBrJF&>ru_f=Dn*u3Na z*t|F^5;iC%Ljz+=ngX8T(Kr0~;fN=`v-}wjJlcTgAXj#zFq*hXU1kqYcgFiqf8lY9 zg|X&8av4iUE$sJLkSu=^gNp=Q$brO_WJvA{b}No%7j&)z>(YJj-pw70Oh|)9Y1M!C zU6AK{ZE)-h=gIb+gZfHO%$2wXUiQgYC;JjB$qX|yB9~yUjVP%*JBgo%KW4&yOQRJF z8!;ssPlhGs$sC)z%oPnA_L_V&J}zF04_Q5e4c?h}pf4FuQK4Y}{A3JHk|>XT3*Cb? z@V8YStqVXfU0@e+U3h}%YN?Rmov)Y?<^q&nzX+{uF>Xfq*6}UugS>ZmUri4Vc$JR7ICuZoCVVk+SHV#14hV*6 zLPxkIe(&1_cbsIwq(%#eiW%UX=i5jq*28V@Tk(`P**M12gnaH@L>AxMO2lVt;pHlI z__1O)7F;q0wo6N(?}9i;m_EVjJI^tPN@Ve}k@xT`(%r(SyFdptamxn~I->>|ZR?A?nt#(1JJ(ccUeGLPgVVYu92f;P<1Mr93~ zQGnzRmhLachh9J6?#%QB<;yd%@$U7g#V`w!@MT_M)?#g9Ls{|VXK_KMVrADeUs&=>2Ypf1L`@1q@N1?T z26`>BsCUiBfA*b^cRTy3(*!3=9%lE>c#B0f?8)((ES&pkHj`Bv1d=XCaC*%EFODjq z;onA}xuh7hbZFS~&;(L9HDHq$99Z#n8g9!gf**gxa8aKI6KuK@qzep@ntl|TRIdR4 zzz%5d{lQIqv>3UFhoaCf8MasT8@yh$8g25s4EvpCq5ZN=khjl*xh;!O%HdN;^oTYj z?;M1=+8xlG`3H=yEkUdNbkW1EPtZ2+1hYdVmOiVLi`9M0U|OCrdLbu-Ml{!QgEuL% zW&`)=L|F}#vuO<~@Gph>nq=_qaAA@ZIWW~j2U)wWL75-r&=pco^C=nQ9SyGk_&@t; zI?QfRVS2P<;Id6Wl%IVFU90D?A+1voGdKxr+aAyx!pzZ3p($umYB5f2nS>`n3J|Z5 zf>ggNGgW_7u*<$)5QhdL!p!r%pR^Fk#%^He;}dZZ3xQd1%F zVhG;KqiS>);3Lw^MK~w5kvw+#hzxWYW7GH*a4|%A`(`_Vy`wYf23(_F*w` zxoAfY8B8ai_URMh_2Oh3M}t({yg_n0SChG3*Rl0!-o1Ld6RGdACHnRrbc~k@sOcQQ z`Me^NW0)T~IPwh78)(Am_2qbO^rwGpzH&DgLsmL{>!>D)bj#)4lRb$3o|7c&;6ajK zaD#hTxC-R;lu2SJAJM$Bo~U#6NF0Z)yvn3viHc&fmsv=LR1TAyLKI0Jy~(`vO6K)w z{@~iGrFiw>3>?>7hV@TtV)c8n|Kz&6Zf_#LHk`(`%eClSF$s=*BnjaX(a z4H9R4nB2sl^!zOsp>MAMcgc!_L^&yf?3^b|4xIkW)@nz>#@X_?TicUJbt>W0h11~0 zsY)`<)QO14@cJdO;Rq+UE!wNq-@kV}UqWk_VOAeZlVP$>73M*q9UtN+A zxsLtXa-7t^kAtS%bXYWa3I;R-uw`-~o(AfKBVvzNSKolpCTmur@d0j*{l;!Sc8qx~ zIfu?z9*5avF7w~~U+5)=s3_mUFJ2pvIZxurVqX1aX6bH{WhF~CODzL84PO#1W>4f| zGuYAH{9q81fd3qxL-c%=iQ|+y@`DN>cR0t$1N%qVeeyDXYv%*I%BGX?nl$3xy@&J< z%pu$&6;iEgLS7tiWFJmF2UXP>SbJUt&QA3vn^#C+Mk$|XlFF)+LlSG|eN&lV+2>$#bcOC3b{)uonvqn&D=PeR$h$dCYx6{_Foi=VmS^pTi#DAnt8g znCy;|+jf!2$YbP1XEMGN`-572ZW`&j=Z>$224XQ+9dzzK7e`O5XG`0KKxC#lDOR*3 zOX6%uVWcVvdYz6}Q|rk#VJ=QubPC^dM5L*^4fCyyB0CRE{^S3qvTIqV`a;+&g2BJ` z05Lk$jq?gBVAY>goKR-Yc-wu0&u{uM|A#@W-WW$BS9akmQ}glTB~`ehU==z2!<5Jt zn-gCJ87$`4g4McC5S`0sai;wuvOMn{xfyi}f1B9LyxcfW1$hCl&o!O+`+mjyEKd-R zGdszvzFp-0Ujb;?^o>#K62zjf)QQ$%Q6hDr8OsQ(kmia|=s33qEI@SEY*zJ)dwyzpz%)(9a zp2kpI`bvWQbX$wx+-<_~15qR?#gm-Ae(0aGSa@>`cBpa2-)(L2(35;@^;4G&>4d<; zz)ozH_!MXBIkUHQf?2b|SD5Lo#nP5~Om4O!50mnTnUeP%40-Xt0I+hWn=4*ZP0HZ zGwQ&7go4Rad)W9(%jp{VuW;Wn5$M`?+$}L}@FQ#p?uK3FdQw%;U}8ujQngs;30?F% z*BeB}|3Z1TB-?gN0y#Y7hs&``L9HVcwB8=Tm%mA(6TV-W-4^q}KKwRZ&FN>`KIGvK z;k)VDjY=GKpXazrRs?;&90EV>>tSC*0=$db4a;Q8*}7vtV6^igT(CV2KG}pT#jgx$ z%f2u_&$qz%9dE?Por;XsAY{^h0oF(qz#<){Af z!Aw{e(b&s>@6q{Fx<{0MXViv-L}`v_9KXP9SRajbfBG8U-xqHTU!AySR$|HitG z+)#GAOD${sXbNdPcLTzI%HqAZ#nF0aYq(SLiCN*dgZ-Iqh}}bV&{63zc6`hd$8}#} ztVgsFzvU8mmbf1k6+MIS)>&+N@f-T+foe!yr2#V!$)b#9k61ZlNqFlXg5?SXN!e~7 z9-9-HlkUHmDY548Q!^LdHEY80W)YCqNrI=FDzUnIJYyVQgue$n;E;`*;p(YfM1JEC zbUcp$&PW4G;tcStC)spHhCGt};DHX_yb1=3MfBZazOC^J`(-eRD*i#CfoOq0XZVP12NhZUaL%iP0lw9WgkJ)HwMJ;uvcMsFK zyqQTT^Z}Qs2(ZdcfsO;o;Kl1V>3-5j(i%lza(f60o;0DZ%b9S=J_7aj2_O?oee~AL z3eDRTffid4bTBapJ?oD}*>?Wu$#NR~&@2P_*q`8SK))k ziVTi0bxY>JfBi7w+e=VcsE%#F$0E@srKo4*3OZ&I0#)XNOkaE&oACWSbdIVc_TYN- zrcMN16+eggKI)^$&xI&p+iSE@;SQpjm!fC#YtheMceG*-Afc6C;7dRWs*2kMV>%u5 zp~Fh(wzKsA)qnn9>~$M|1Uhv08uGQ6f-7wc8`Y86#GU!UKN4vdOpuvyxknUqA)Yp)JJeD}4w}&v=-C2bi%rBy4t;HzB+!R?k z??g6sG?Jbf26Rd&`ca(*A)W5%;InEtlBFgaa+nDL7eg3ffpak9jzohIZ>UXaipv8tzI(mrw9;zOS@Uw#aqn zRkasX&vQXK5p$7M+$cDQ-r$=4m`yR)Z$bB*I;P?IW6*q0BPZ3lab(nJhxsOUvmLdj_{eNc);H`H zRKMr~!;0yMx*vf2nP?Pr%?I1wkHf(`Vxdsm8U*wXLfzUrkbNz{##C8jqSOnILnP?( zC+E3710HO5QW}pdy8|rJ9{^ok4Xk&-}J>Wj^=OGO!#8uBAOr}y~rqXneEaSZg7ZQ+c5R;$!;ZOd$UxVQ~h z)vJ?`n~{J;Rgsf}Fw$A-gUWmj;PUArynEVvaLj)MLrYSaM}mPcm~Rd#zSE#}#U40b z{sdZ!q|jrTcX0IS0r0!E9Qj%7!GV_o*sr?>!EAFOOr=CA*Hr1SZUM7#6?`f{$)n6vCW-5WOts4o*xRs{1R+! ztN=27Ec-v3{J+?T|HY)*_e-619c= zmDVJt0w(y^eF_SR30~IGO09ok%^sipi4{$@U~ zWLG|!wc-`GN=F}?%#z30M38t(-NQ4V+#^2}r$ha=Yw*J4>_2{Z#Z!+M4q4+Ii*Gom z+Ja28O&}Uw^2Ea;jKaB{Gsu;IjY%YnJHH`ne)_I|a$(C7eWS7%4B6)Hy z3D6nE0itK{iGfpuy0M*fx1A)4BU8w5dm!=fy@$UrckupMgX}}q3uLbH2wpd^k&HCo z=TentiEZ>IQn)jNyeim%oxC!L{7XGj)uK!I6y%B7oaZceS}>uTl*!G|FtS-Rl%#F5 zB{nN=6V>5Va!&aiS$y_7*}3=%iGFmISTsbDlYdVUet$huFl0v(Vj^*yUkd5Tn?r{6 zc(v90I`G--ERnseMugmbi5K@UIezvJR+T(SOuC#&l|d58^N1qf9csxpqg1lZ)|;fY zh7+@Cr$OPk8hh7S1x^Nwa-Gg^V#QPb{P!-#$GF9k{fUOei?txC_ayPk-R1ZTTM2Z} zVjO$#D%Lcu#W}s^WaRf@Vm#1JE_s-c-$F}(Wj^B%uO)EIWEX4T@(k=gD5G;3U*QQ4 zt9vhXCi4DfhHmKFq3B2kWu13M!H2}r!KH7w&yxq}s`}gP>-!V%u)_z|ejNk{i-m|U zZ61o)EP_61KY?o1g^We+b+91Y(NLB(%1B>=0`FJA6)hnoS&;-6)byd&A`^DV^K3xGLMIJ7vvkc^Z^0+jf#p5xV z5>~b`3_7j%q0ot?DA!pF*#|ua75P#yJo}r`=n6&JI}OmUUIXMBmyVS*D_FGr-9Pc) z&bgm4b9Ka-_x~`S$7i8umyJ-l$u_jF!x!BhwnpPiKZF0Cg^)I9DQKQ~1zoxhkaK(& zyThF~BlYJkY(G~5i{j;xhvyRX!%hvk_4fj+T@GE@?;*uN5BLWa(8VErv>5%H|CsV- zt-cRjh2~4G;1X$q47}`7+}KIv8oUpZgx5fiO%^HBRm1iw8>1AA;sy0ipUE&^C1`LeBI3 zt##Hq{dvw=tyX{Tr~BE@-uJ%mYhSNznzpaTb#N|R*RfT9{DSkZ>^aBfVKzs~B879I z`X@&@PQkW1V3Y07Vk_IIU!5FzBNUsTfixg7P2k5uWkGA z?`&Jo+cP;|wXJQh=T&h28p_!Aij~{k@LI}QwKm@7pww*6*=Na|IF`XI&x~+lm)>KU zT=zJiH@eyWS<}v0`sEFW-(r#N)Cm> z&ph^UOMM@4KdTLLcRF1wCAk7_^uu`W`b`DgPYr(DGmGZ&wvK3Xf+v2n`S$oYr-a$V z4P973mwiLItJ74tdhLzeg-`C6u0I&h-c51g)@7D(lIQc=NZ%XB6E9QZt=p5yjaVbg z(@kKtC~j=$c5VdsT(Jr7*x%gJse5j6?)b`bcYn_;xflM6TlU$8_r&lR_la-~_d@eD zo~iPE&aSMjocnjLm3Bm@bDu3#_I@IvJkxE+ng+;1^bypQL6O1Iw(;r?7AIA;AHaX-hYvxw!WNJVoj#0oe+W*5@m zVSldi&3W8iXL#I9cK;%$pQYOMou!+bHgUQ9ik!Q73S5i7kGZFEN+4sq<|m28Ti zrgFCJKhH&qELU#6x$We=J8h)nEV+VZCEQmZT5Sx^jnzL$WscgcXU9H~b=cOl>!Qt@ z@By1kds8^~U6yjSvsJkf@4Rhf`@Ffjsdr27ud*mT-R#JnH*FGECful?Pfmnm^6nT{ zaE3W|QTrv1q*p7aF5Q^(>F00G@_z;#-8LECmj$m%Zw$Hf8UzKo+)w57Z|05CvIEK7 zI<-{JI7v~iSG;gZ!wYa7XDlqa{8@u1eexgo?^!`!(ezck!{@^|!pX6;r1*HLm|GKf z;du^s&u&egL#nsU0iihVxcmlgPf#@Xae59{&gCojRQrAIP>wKft$h)<#jBK?ai)eF zDtmzYG^vX7e4vGsX!?tDGF*<=ptheo{zVS=R-Fdd&G00*M9Or``cE>!Hc6`IN{@cZ z=J*;t%TT$LryZKVGT=DB1<@eDFf@(ddO<2@@)<9cVZEooat zUg6+4-b06X-1!}!C}_z+j-c9n&Tc&f-djyWe{pZ>^FYFH69XSN+J2_sBknw~?*; zzCKgIJ;*cQ9@zVWb6Giw(>mLbH(`4)t506Y{TrCfTl=}Pq=A4 z?@F~APcq^fFV)YEr^V{!Inic3Q>Ad;jTyJN^%vZD8ZYyCJq~`nF6jhrtdkYbNx+Sl z$)CjAyDXIVE7zR&b7f2&H>F7oB@~%Yis?YH8cFdi@^BK(HIxDC0#62eS-YYHV zy{mZ0ogtfP6Di=!yR^=O`zY}fxB0d(ueYv>S3d6)FE3M(J7_G!bK9%Qd%Dz)`{CPR z?saVo-c5-|+)m9Kyr)u!c>8r%aZ}@*c=d9Jc$zEE^D5j;cuTUD@h&O-S9*fgT?MGE z;ibC0;)X15;T9H0@$%TU&R};vFL>4k9^K934R@^IS$H>a7r*r68R#h6R*-(ktvJd5=BgZr-)D| zC;IF+mkQP0X~yd|rY)+68K~@Kmit6e^5Rf>y!;T=&VA1~NdIMY|H?w_%rg4DYawm? zy@2-a?_+{Leqg5fiGX>x36`RJV8Z(0Nlpx>x*tIl^nMTRf3kwyfBO@&&Wj@VccW}! zCkz|j;ZffJqw?VgQ|MC7+#j6@B~K?PXlO!UO%hfHT!6CrO>AWu6)jm`ki6|ZIGo9J zepHh#bt=)!BWEbxG?`Y}CsIZ7?Xf4-maHHx1$X+t zeboQ=H|&KNIkXDUP&FU<@lD70=w*nJUyJP*=1^IbEiIg3M^zR3DXwxK*|Y_*-ebSu zEj|S6;RdYk+(>29Hc~qIkm#2=IBo8T)w9eHx~7eJaJQ8yVTKtK*;>dPEXVGD+%cSq zYsUU0BNs@Pvo_HBsD1RG&n~(dxRcgx*+Kq}+v)L!Ewp5d53Ms>8fORer*iOjX&Bwe+)oCZgQ=)`DzQp(QdDvs^DYkD)gU`3B^ntvqT`MinfY;- zn0DoAW{-#(#eAJeF=v%XY^N|;R`Su=>7B56B@clQ(s;i`5vq(Z@jd0E@k{y0B~+8@ zg*BN!|4`V}_z-*H875dG_3 zLE5qjCd=L+#?&3VB$lFq8-~m+^T@%_m2QnXkkw9II^;QtcCo%%-UB6gT=ocop-uQU zuNQd>p5rMRact2t+Uv5MLWkE=iK8_g*fxy>zJWS+Q~Y0ynQkCPXF!NG;)kyohG0!P+2%u|kB3YF7mx9#SSkXLL8_L|0hJvJIcq~0WW*(}u-jagH zU2%tFhl)Qzh#U=LeeYQQs%9}nZ40pRHR}Vc$)QwL4mI~$(@0=6T0ezi$ImdVmp+S- z!jq8mJ&t+nKf?L?PjoyOMzSA|xtcS=WF0WY!Yjw{-t9Ej&x#u}u4>as950QAXG$n$ zT~VRwhB}nSzJKLzT?(nvp=&0Z#5PRmyto3LU9C(Pc(=eTeSo_m&FE$sGJStJ^z!Zu z3b^PDr(_qb@YX@?`!KK$u<%lf!l5)J5;jmE`(#Plx=9`VUFtZYD~p%u`lRvNfND)G zNo9)}ZM2+BB4kcKj!vR05k{=U4U8 z29 zBUx@=0?rtwpshC-8M9@GHzZE!wn}v7vnXxTlc2|QqbC zXlNKaTkI%8+n&6Y7Sq4U7AOd`!pFWv@NcOClXf3cb+uTr|0L>XokOl(F_t}Qg~!@A zXge?h_wmZ~iDjA^scO@)dnb^8>jXaLpMd{`<>2Jb!@;Q4SiI#fI=&Ra!K4lcS)S}b zy#$4hDw6xqQY7sO#*HIKaQVC%o>eNbyd!OVQc$9!ua#+qk^#xuv+GIoRI)LhO4fTa zu|NGJMoZ7bJYSyPG%C=zH{cWF~-Pdj}Z(0X`)7$y57G9r~kg zO0(L`==})$J9!dK=@q4HmXBO3sYp>$Z(udA2ddqVk-t%wN*s*IIb|B{xZH*VMqP+M z*pIldNF+5!g6~NNe7_&Ustu9Qj13<%uG;P!nBrZCkzyWx+_0nqTPw>7B3 z$L62WulOC)9d#&AQ;Py78<7C}Yyz{J;F10oNz0=lJLwoKZXZR^A5~gqGnw^n1HBVr zwU4ZKuE#8AL(4yB2Q~g zvKzEvSp`lsdeWV;t}mnTwN~VjYeywz^U3e|MO=)%0gF4C&>A>}#ZS*+$xt3{sK`;8 z>jbjcqe=7i|AY9v0ChoPdvss>{ zflfcxt`VljUA;(c{D6s?pU^O^1M_0L(0H^3eI^cMsqIR_>GP>o$&xnRNm8f#l1h|gqfH@VLS)Moq3osd)OtvTj;+!m|6wh9T<`@irwGwz=RcUc z?;{h`BnU00n~7U4LMM}?X!b5?;#_g3wA=H^d2~Kq_CAUot53pT`4rrPX4B1x1+<}e zJsnilB-!=KBt2Jm%$mcMdBA)`Gb1prnmHsij2Ev2=|9==wC(6IEMxg>ZJ}A%%5Os# z&pFXMSC-XPZ;5?VtdOQ8i!05usQ#4)-8jC2{00T7Jy?<+ZT zv#Owj5&xJ0+59Y=e3T3qr)>z0iDj)*<8l0|24%%fq{a&xv~Tzje(1#GR#_ZsvnSF` z7b9BUV?~0Gh3I{bFlBT|leEBh2v>fE_~{?eiczCVJ8k-RTbD+csZfEG0WCUgL_Dqp zEn|5Qb0lRcWwHP@R!gzo_Uh#3qf0A)Xi;CLIt7^yF`3&d7~!8hCNW6_Ni9Oim^aD< z${S+AE+e=LPr$bs3M4&Aom7St$V0stOM`BszPTFT&KP3`{}gO+vKce~P3~Psz0MsR zlDYzs!EVf}9Rcr*IN@jt{I6!AaCQOf{Wd-Dp~8OLd73wAp+Xhy7E@ zZ@~t3p}GfLPtvsjkons`pypVy49F|ob|phDnsY-Yq<3> z59gm3f+b9#M{^6bCT_yM*`l~BmEWWqRBn z&cM`~4A^o4@n7~b+?Cjb71H0JbnOe$jU!GFKC_1@&GyG zvDEH2BX1l7yXq4d-oyT7*#SS79!dIN`N^3m!V*1PfeFea^GnM!=mz_ovf8|&YpQRof$`d&ezCknqd1!DT|o!B*Z zJl$WQNPRL2lv&gP*;xYEb@kN%O%yN~mze*3O5>i+l zCqt^oB}tL>;WeDNA2pi}!MiIIrm1brf~NP(lnGtTmpMC8xo9npMuowojP=F0i$%8j zQRvHj!NJOp*!`pjzhr-5@E0EqMgPX1+ZD)4zk`Ina+rpGL9OLCI2sACsG~-vzw|Bh zduIz1I_WEF)xTnGSufUXdWz>;8qxBzaZF6wRxH7>v-43^F&$BhW%2zDA6{ON!K%f* zOr^&^ChOIA=0g5fl(_q${N8LVoYMevp+`9C*N>yO^HAk~7NT9}5gosnc72;iFIKq{ z*WQ};@ECHaHm1(WR&?ykY`PZdPQ{)l!DV^0g3M<8xqcY-+J`YQArwdSC8$$Wl!6T; zXrh!O1RL$qUhR)U;i(k=(~J(4o6*y~EbBaWBdNdjp&QO?a3|9T#;ix@?}^#ynsppq zDo64D{9(%AKTP-M#nLU6ZS-r#7V_rbOygMpQSUkLaf$ZuATGW`1gs}O^PkwkR zv@_{q?TlKJETt7nkh{M)eVkxISM%)2T>!*XuO>d}jpTG@HR()Thlt7BuOr2^DHtlEU9wygyTe$j--OVlK>{dyhT)f-~am z`K5arEn?3p-p=-v|8NOaWw?;g9(&3fwWo0b9MWE^OZ9u#U<$D<^bC6>J11iLzr!f7 zUylzDjmbC0fcE+sQbNgBNM((pb(;{`&>{+6IhRTcXO7u-u1TgqspT-zckM^Z)NTY% zdymr>Tj0o^x%YY7Q~Nr7`VzRE#$Db_iK(8{Y8`=7b`faT+JWD94$;!!gQVuXi~1zu zQ0x_fOUX;|LQ91jyVPmummIwkRHDj!X{z5RL2mX&)bLo3(rnm#a`FzERTxTIe7opE z!V>xrzK}|CUFfs2GtCOKrx`(0$-HF-B+6&uz&U69aQ7fVhdFdGU=AG)W@rjqP7$oq zBblas6tBYaJ-t~b=+11^@Mqw^|Kg$0XhQ;Tr&Cd9<2DO*UvuOag3&m|MM4c=|~IE5BRP-sME+)E!7WL<`|QE7Zlbg&7%QF~!c^14(0ca}CR$vaTv>&KESEa+mp=58OtC3D0>U37aj5bT z5*wQER=5#aYP}f!na*swb((pbk;9BE^+ZhfF6{oZ0n?}ILhq9ZK9z~1GB+7Ef1=SR z5{*?X@5}qV3c6QLM8e=ytYY^^r4(K0`kZFA8AdVJC!JzykF3O_lapb6#|{I#ve2)8 z4wv3tz>b$;5KlRXyw~yUo??n911(7X)WT5|BV#BVdSVxl_=05>?+C?4@oo6I00&&jKXpmgCT# z2T;m;isp)^@HtS$&SO#ZzT-!lsyM2?WWzr#1tC*Y;JSAyUPgFe(%*TAW_dYnrMa;9 zaubIF7NBO_6u2FC!iZlO+GGEN;`>8mzB{T&fQ?i%;-sUncPkG@(S>mP%R^1fZYb~C z0>g7#QNHU8T=+8ZapgJK^*m&3r`%x-10FKXHdU}Hxr?|5m5AGOnPCd9G8#1{%&R1K zoRM1(HJy!M9^JsI;7hozeg;)i?0IYN5u}|v3^V;g=5KpGvwnU#6T0{%_F8nZY>sYZ zv);TL_g5izZX@C}{cx&!29{lMMaY|*xKvn#!i^8n;xq>fj_5#Py&eQuu3q=EQs#t2 zC6n{S6HA1w5XYFq(Kn27ITg+9d!NqijXR6235nR+bQ*5*SJARB7Xtd%adTlTj1msx z{&Is+0L@z_g-WH}-|0*-Dgr{|{Al<7lFRByH=Jr14h%wBcYNX}9d9?+qTbE`1TX z-}9!ECJ`_Qi^BBz$v9I`ja5fFkA* zr!_Q(-LE=V?#4mE2)y$M8x!-)K?_Wib;tURlaRR3ot$60Qs&tOl%MQH!)smWlZZWW z@43)o>6vu*^$d#Xnhqs{`A|MR1-4SsBoHr0sp8tSa3BU#vyMYqH3$kQVM(ZIZp$h`dkt@Xn|jHrcM^ndD3{nH(yu`oTQu9%96AibFoqh8D2g#dWJ)Y5M{v zvMZWF+QC!FaobxQeA|z9mtOG1>TuG#4NpY+5M#fJ(yKPm`AMs3v|$c;{PCdA`721h zM;+ZP8{%Q0CCkM2rDb1z>G4NjS|2|f`t!X|duk~ZWAy2Em?lj!Q>BiL{}8>o2ld4t z;lo+W`p_SM;Lk(op3}t`v(G0RGs((plUMO-YRr&Dy_XSw&NX0jES80%)(VN}r+7R;il*+6r-DueI=y-q z{pM|>^^)5uV8nnHG7ODZ;*fjfA9xjh!&{cwy!f90t-3EmGyOHFM$Cm)1-}iL9$%W&;zJR8*HXgY zwWMvsPlGn&=uwme-F)tkjvs;GUEhHL9Z$S@<%#M&^H6C#g?uE4$}ZcJ{XuJTVtH31 zEw*&?h%2IwJ0kmzIgD8rNzA`U8d(rQPb!?LE6anjHQg!VkSrzmsG?Mp1~?GJ|h@*S{zI+s!ky{P@&B1+&I zVps2Y#2J2MTx%B7RflynD}-fD?Q}=#rp2)03&Q)Ue;5Zw3<5{2;d0av@4d9amh#4Y zci{(#7AQ@nHKWEPs=kmG?^r^=**W`an=zgEH<3>C>r%4e7bx3*Mv(1GT$yG@s`4hZ z!%2@`ea(S(W+o2bxQcZv49RTeWNMAjA}fDude|_NuIzLnH@;ld)#TvR=hLvzkjDK5 zl1SOFjN7`4D0Q1JH7(vm8*g#o*FG1CHyt6{{R7K7`N$xKpL8}~!}*dbe5PV-x^kP5 zICPJhkoJ~27UWI*(w=1aXAx<#->K{)0oHq6k|f2tm>{!3=F_K-jAOVXy%%74WZ_N} zyu21Sq#om5(_K8R^q{*+p0q4%DKUmKDW=thmZ#4d^Da7Cn9vOIjtLMpADw=%7$4#m-8)9|sK9B2%sfEz z_8*~V>yA@z;5hs?XKU;OcKE8DN%I}A(Auah+QQj@9I@TV%0A5^Xt&e3`XK5tSwqg- zJ@NI&JQUoEhU2#YEc0CrCDuFMI4+%KWL%<&?dNEJ(ph@vnMrm!r|GA9B7U7qLFc$c z+;7r^XSE&l4^M&P?rH=$x8M_71K-+^DtLdK<%^!CQU4=kzONi>FWiRbl1OZs zkU+KS>2%Q~p4#W8u{CkFUgU5cSO4Shk*`;FxhW&c~3JWITx; z4ko|oB-+%ULU*lF>6>u^8J2M z)0NR3bmeR!87HOC?u2-fW&h{N&o7e6*-ScdZ4-s)Y#`Mgo@}k=Hx99|O7nX{wEyHe zvdz6jUzcRj0VOdq99E>mvMO}d?jUZ*9z$5#NhnD*GcCoBnfm;9j8wHf9FH%=l3HKf zTXvRMM`xOKJCW+zR?>>LFiJ>|AV2SNsQ;^hL{1%guO(Ajeg>KJq>#KEKYAX{7NGAY0Xt^#e|s!Dv9m1%d! zGvp~Z!Ff|1wscLzf*>2zg?ht=WlR;hi&1ZgIEChah1ao<_YY|uw0ah z!=t#AElB^|rCBDUHEr9^keCFhtaJpy86tFSh5%)_@*w@`CY0vnA@BAz%+$`tlb@$q zo`fbfSLo8RWDWM4*P+QeYLu9wPGV6abXaa2G26wcuY+Y8@OPqR&0G8ycZ11QHw5t$ zJO*-+wfQ>BGi7sa@vrFW{)u(#S#zogw*ON55SLOP;38XRdo#rVd_8g)I%7-8J1yyq zv^jO&{fQy5uSnYQ3)|1CuslO;3fQSf?(%M^>YIyv+c{&_VZZxb%=52znNqJO%%2Ty zP^qp*WqlVWUdx6eB9XYg+bc}3#^N0dBZ_pnBH_Ti1=K``0@gz*%`owU5K!ky-d12Y4uI!SkB{Dx!pOQ(FcVY(H+-_M=ePnv7?zE1|(G z1Fczu{b$9fDoKt8W#njJO+7C8HDF5qLn!~cj;hx;U>?l!WL*WQoEd`g^IRrV_bZ*&3H(8-jehP*k!Oe!6>5w{#73 zYr1jUqZ0wY+TgHm9LXP(q)FXUbnOmX?>eBvo|$c+k|0cHOXTRIuPn{1Q=}nJ1zJ8` zk6i1|F)`J$=cUxp>_d#_!AgV%XS^AXn=qr>ieSrU~Pq-_!8rNO_ zLss`5j9c{z4S8QMZ4y7(sW-znm}TysD8(O(n~YTL6=rSAMW%{HzNq}PV>#ZA^yRu7 zmPn4r)`gRy_fLzoo22Q-XGMBh`5C{;2C(zH0IB*PMWI48Ze5Co^{T}PIlBgzIi?W! zYk>z!K6v^z5P9MJ^me=$xtOthpTpzmyPPEH^ovk}k39)FS*T&-%^E<1_yk;Y=#-y&SPSpKKB*G!`^`-^%<3pe=ui(5M?F*gSea&?T(V8 zmfO-a>2C$L$=*hjR1vVG9Ygz{LGx)fT(1bzu$%zpz7!&{^!-@Xa2R(J!=S%2481)` zSgd~y@}J%z{cb;|?H(R8F7H+|UP(N_{Ny_bxI7C*Y_7z8WC6ZRX+-?5*k${N=`X6q z6#f?Ia_*yA>KgpREAdp}J{CsvQ=pXy`97ASE5VBq@4gyN@jmcYR;Iwqnk4zgj4qi< zV!%%Vii?Hd+B*U{-BE<}h||Vziu5>HnX(GB$Y1Iuj?8R_R_!zF|MUzL{t%9d4G=FV zVots-U}P8FV2-l+OX|XBFr4)aohdA%$ZI0C-k3~J%_K>9?Ressl%YpI?qYgp8G3Ku zL0*^?p>1n88e`J;THw{mIBC@S)OeO%dg-x8`0AOV+xg=Myjz_8J~!9#`{PI)AiMW z`t^)x%Q{Os5@$*O3e0K6)F~8~ID|O^LkP|kCps|)Yxw5ig2zIPvUS|xOqQ!VN064k zszvgoR;;LfgLN}4aW+{O4nK6U=Y#`o-3SU?C{qxU^%5xtae24r7&$^*C?+y8Sq?r8T)H2FiXQDs!1=B?JaS;qQPc8akbeb>SiX&mus3GS@WS%;z%lPS z>#7KSY!RhL?^u0`mI1!Xcj^5&`&Nct%~zqVnObyph8tR! zTSL~@6sGS^G8fe2nMcM6jGDF*#MAYWH>xluCP9z#@WL@03-oW`T(TOaXRFXQLp_@E z=P|0+HNt6TJ+v3zK2e#~JA3r0El!U@rdiM<@pPJfH~q;bbf(9UE-IPP8r#V@-D8CquPG?=$wUV7 zu*|y*2agTHCUqFfEhE_T{WpFYjN-xoJ14c|$a=5 zVpOtSoMwF#Bh|dO5Da>St$lSMpHvJPZy6yRbN;18)^GbSoOLHPQ#b(*{&T zzJpptDuT36W3kH-%+mY=y#pU`cvTNBo(hHUxFj?MCO~J&A4IKXzts=ow3F3A1T7Y$ zj%*3qX*!_aXW`APzT3#_DJ)KO#cvA{G&g$K+Sdi|Z z3=OEs(UxHaI`cZ72~UY(I@&ffoSU-ryhDoKg~-#mpmw}`@fs=ogAk2UrEk89R93D{ z6Otq;D}dF*c*&2+|3x-wpzx~&hP3sdUaSDgbxIiUo{Uo}zfo$#NB5$}v5e9hT&-+^ zTPhE?d_j-iRAcuJ}y?I}Yl_3X2mufGfn8=Zhi7{okZ~!f#5UUjIN#qSdz2_+QWV@-m?)0rCV^(zXujUUy%6SgQEA$ zr2AIxR3jx${+pyp?w2ApSsRmmoF-*Y(4u^YJ2?7*hixm0p_pt-HCr9Xyks`bna5DH z`DEJvX(~PY)el#j&+v}{yhE}GTq_-c`2 zmJwBDnbLoGEN{?4onj`)(Dgi}F)=^-cM_DuCS#E`gWqe!V4 zQSD#}seQYwkjs5FOU{Y#Bmy{Xmx1C9Y(gC=v;|BrPK=g@kXLV4~IAj)t zA_XC8WOIG@+jNWc$N1LvJB(r$l|p;uIz?Ny_~hIO=-_tF*6-S&Oqd(UrX+&X^buI^(ZGas{k>?-`6av#rPdYCXG?5FggVm&!j(RhJa@ug+yzaA&bUsSG0fV&L`i6z&(MW1JC}sSta}ER`BzoYJ2$ zB`-fP8(cpz9&K-Mg8hax-~GbH1K$u<_ZqoBIzau4nXq9uM&9iJ6MSkB4qTXpGqt<1 zFExWP89B=It&1Ns5BcZDVZK-b^aa!4JDuRxWs9gDCu9^ohUaK4;vCu$Sfzr6MH;x_ zGQymCc@7%I39K!gIifc9V>qZo>gvJA#+{6*$KUj`%ewu907(al(_85$FlIS9mEJ}9ko z#r zia1!NfoZHhBurxf6Knf0Z=E7`e;2|2HJVs0(!{K@e#w-We_;4E1~KUP8wxC!+A7r( zP}>0lSi_X`g`|$=S-&+aIv{>v7eZa^j4=`IyK8$(SQkM^4%Jc}l7Yc`j z^KG0}ZAWi(BUXM2!|Un@Y`hwVnB&e+k(q_9lGa$zs|u?5gXBM8yPsVpA73Ae8q0P|i5*zL_QdlRX5!_$aJH6m z0u`-6_?fdCY4ak{T`7-66Xo$GWde#-jv>J0C{~_72A}Q1Fx2J8kMExtf6cd8E8U8T z(_cd~IGb6R63YxPN@jJCebAa;hc@q4wm(n-|4X-!Yxf9;POV4l3rBFi*Z0TDDm@UjhLO;H|D4ea?q{Z^HjSD8yGH9_eYX;(7w#acTnY|lADPXapBc~WD7?`< z49mhJXd4=jTV2Y~efpC*crFXa_T?bJ{W@|EI^m-eu(WCt%6hnvy>uNaA=hwr;5II` zK18BSIUfEqfryR>w*3-7Wl#+c#uee)k}|fw-+`r$&2VNlSdUnzw8(%*_%!bcO6tqe zZCZ@*?}cD$gOE|T5my~nLuK_&*sX|zlYbl*B@N(0_#6CC?8W^PPH%=Q|%uDm=%` z|Ih56zt1EFbIu&bq$U2Sc{CRhPp0ET?;B=8dkwR)`7slGC>|oK&cPt@H0oPi*_yjK zgr>ScXUBHDRM>(SaqI9e3H0j)%TErrXUk}|(0<8aYncf+ZK9+&SDEH1Or&@xd0Nsj zo(g7aP^m{OUP>iF!94@_7FS|HdJY=x>e0x?YD?}{gs}H|wy`?jk(CQ*uzCqO&XA`Q zb0<)?n+jmiY|AfXn`Z?9dxI4dzX^cN=stcyz$*I(bSq? zq^5g8cfAUmYuB+FL{%lWcPc|N^Q9@VMT&g1)#$;0rer*RI-RkcLn{P5 zXy8~cq>grBx9-50n5UQMqTidsg$BtAsPC2H6QK9g_vT`YC<9`$>8Kv zs$FhF8;8Fz0>({@?W1yLxf>{VD~DX;r_#3Ri?AVKHsVa@qh*T{Zmkr-EOw14Kgxl8 z2m@x>j4|V8@6aMtn3BU?9TM%ELVR@`itw{W_EBcT5q$_6NkCuT0G^xJ zyKI|YVsT*;UP`Col;3GA+>2;wo#Xe>4%xJHJ@`wX2Kc9!SdR7oF)PYa4BhD-g!2aq@D0Pd#wXFBd$t*q` z5>mqVAwQJtpNI3e82Ida1ovmH(CY5Ovu}TK>Lwq1$8c;fUL;EvBR)dNeJ+NaWL;Vm zGKB)Xr_#=wZkYPk7;>W15WIOdw!}K4{Ox4u_bAf4b|p$p)Fi1?Thbcf(Cax2P28A- ze61YBMwS3knMg1$!KRiQ(2mH*f2oD&*QrO!S|Oy&7sGW!1>6(u$JC$SaJu9-dp2O} z!RNl9y=WA9-%g>(Ar4Is)9^&G9?HJ2P^>h_?q6qNygnZhSF50+#6^$fQ<(2ReF87-~2RP!3EtQt`f zcM45LG3@<|V=%gK4!6^9AX&5uZB086+r19+IAJ(%6^sS^(YVnOJ7ylHeU51ea=yFmz-tvc|h% z?>=X=2*zXmyX*D3ePeR!o&0|8k({Z?v^`d+_i^dP{kj^-jRw%K)V_0 zj?cp0NsjnkJ;%@Wl-=JGZtl6*HeRi)Gl_qxUNW8N5ep{Bu}!G$Lf< zbG+T!hEG{~}E`)UELtM25 z_J|RT*&IYM;sKoY)}TK674|BNbJClY%x zjkLe(Q;&uoeN57#hUtp*x_3OSl$Rr$d0Hg+S9DRwWt4F|R#!s&|7}52uQ)q{x6^IvWe8CwcBQ}U~N~5nLdZqW?R#Y zb<=6iqFMAvLz8m6^{BSVnBfQQx`Q^l!kLPIFYq#D%>VlF0T`K-4O2L%y@k=pL)dU(U!=8(XV2 zk5dM}_+%`L4aA3xV!RxvVdw2w-SwB!ain@*k3!F=(ywMiy3^0l_s?!XnjJ~~jT^o2nn~C0>C=Db%xIda23_B)Mn!xY z)Z-vW9scIj=x0Tufs^Q|Ob+VT6+wJX6*&6rJ;yX7T6NEi=G|1G_EsITyETD!9-cyM z8cAN$S>5rVD9zrlO4DrRXrGo232!%{hcoqPu9h>I|8S#rR$n;$gPjAdmaKMq3cZ>f zjZo#2?5*`HsH)VUD~|ePENVb^PY+|0y#QIn3(&2N#$=cV(hWVh*3ds@ z{%c$mLy4XuG8&EGux|hX0wXBO{D~2sGDW}CB~R@Mr1~QijKUtA*D-}|)igSC$Au(p zoajfoBSmj?rnBE%>3{d{r+jw%`TyYuO*w6td(&ht6?E)V8VtU zY&LN|)bPM&PS~*FKtE4p<~e;04A!apo``tkjX%Hq)%T~*^vv~tS4{kW{ciyLoA&Si z_y75S{onqN|K(r)`+xTz{@uUiQ}ti|SAYMn|LgzuAO3awAOB+x8)DLlDq2G)EBb+Gm{ zZ!R`qN@Lu`)||Xt14H{|`D*0lv=OZ|#ghxT2F%Y5;{YA#t2G`#D|{wT&kf@8(Ka|W zf3%f@v+m3u#8ZwMk@#_9toNj*f#R_p$Z5~19DVEL_a1Hg_Bqdxa08rJP2XH#bLVQ$ zXf=Dn*~9$Vv+3_=IeVUbCuja({J;Efb?^xSOf|7~F_!OQ-_5{p8}ji76i?c|@}~x> zeDlS7NBE*NT2MLXY~FZo$VIF>Oa1%GJ+Z-O>_;Hq(OB~rVn2KKuVd`Nnjg{H<&KY4 zbCK5v=C10o0dDAwt!sj~?hY=47+XBPd!2jL+QYi4MOSl(t?4`G4Ewl`eiQiG>VbRq z8E~Ql_wxCTBbQeHA-Chkxe`17`Gp|R@w~?B?0D$X1HST>O=g~$ijdMN5 zYBCA1-v0}jAK*`)>lxl~=GBHi$5}mA54G>R?5fRJ^J|}6nh9r*r}cvPynlP1J9BpV zk=IrXcb?_MNzP-f+J?1J9in($DSP@o)sOsM>mCn5jr}~@^0BARXOSB~t*5wHJ9q2^0|aP5o$piK z%w-!kfL7kmSjWGh%UFEn;C%MP+DGfI4Ger^Pu0K~|8b6YWNdAi1IJq9JoBAedpgUw z<<=VGN9>4010Na^&T-nY;FSIwYBbQuLhUI39tXc znyb&byOPr{J$ZW=|M0o$26^i|^?c+a_I$XJdpBo+Nv?wg14Oo}5B$^ThR)BZ`JQ|x zt;P3w10=ra@LKOuU4i-9VT|ug<0~4^%(b8W%m?{^b9cWkt+U4b6)O_%*vgak-kEUzYYX`AVQ!&%!UpgztowR&e_r=|gI8Y~+*7aCdkXUI-2!k%@<#xU{seWO=(JV;xxUUDbv~Eh zV}8;nU$wD#@5lK+frRs@ulvFO=k$qK_xMCJ@YU5lkT(~J9&65h5?sEtZ*v;_jabbK ztInMZ_-6+-)~jpS)MpoUp4th`1Jw6sIi5$IJ*`pux_^&PPv<5YQSX5|Ypv@EC(sD* zYQ~zA@Wyqp)?CcPN37@4?>kq05Sz2N$NG`$9ojcR>F=x8d6q!u@jiO|;d6-JHhSWM zLF|3#`~B?~VC-%Qr_No^ot^KibsxL7p&97^0^Xnrxa%FuiClYY<-Z+#_Y>|hiTw;Y zuX--POR2h_Hf=L_NY}Cpi}b|{0`K&TzNGCe79Qn8L|3x;wHx&kGbpCwSc&T z1RQJKLG4XV>|XHJneTAc>3ct=qvrCtK2w!PQWjpV$DRYd{1blU+M~GB8;zg0fhNG& zxpAsK*ZnwKO@awp0()b70Q&tzEbsF)IDPNuKmEkVzCgeS{1g8Ttn)tXG2j#XcmJ^8 zANiL#c9gGrG}bp_&&)F4?&)vTct)(cZ@r1HiUvjCEhuG+^BC_#9_^ zpov(|Q{S`LXAq0p8Qleq{q6C_mmfPOT{+*$jocG^J};#S1mG*V#0K!|RIcahm00<} zPOSZZRRd}VE&L?!J5x8n75`|G*tzC2zV~w{wbfr9zX^}M$osjrzx<2XK8Ziz2DWK| z-&@W*Hm}ayLEs0PpY6djocA-}edgDF&6$ZMcK#KsZld`G`vCT}CV_pPaQ|f9{d-Q$ zgIK-Vy02_?Vp{|6>W9xY>q5V_bIxzEW(M}mInSm&Jv%kV_8t1nv+r;8IA0&;-S_XA zwf1(-(0bslf8KYFSdFl!*1Fti#0eVl4ed{8I=*@D*8Bv1hX1qAb&m~=z1%I( ziSt|&-;?O|?XC50^y|5?KFis0;tx3Alezo&MtLS-#K<+^hwD)_ix)Zi3iabKaqUx9Dfn z?-kbn1nRTIb6GRd4q(rxIXE@;pZE=gNb2E2jq|8i?=ff11Nhumkg>iGT<@zs(Fpe% z&b|L~U%h-*Igk1PTG2MF>%cQ})dlw4W5&Ac?vCa`z;K@>gKaKHi7Ve!=Po+yrrN-`-Ep)x-B?O#{~Wxv;0)0LEJL!ny9z z@yp_SB7wEB@W;HfZuC^|d(wI`mOrpF*UxP8-jh2#w^bi#?D@R!&lY~lqp{ZKcmIZe z`k9j}p47kiT)ugqJF$LU)_dZ)ysMvq$KLK>&huYbn`@Sz4SNRh6}|;A@RTKA9Ga^= z0??;+NphUxV)5>L;t6*>_0&A6H=o#l_PJ&TnmTa4WA}0w;fxWjYO40Wc<*1`FIeBH zbGkXTBA2btlrC{;f)uc}EH#}t=;Je~3G-m&s`DKsu^6BZocC*8`vlk?)XOC?uwK4S z@@K3W>i+ngWJ-5MJIMvmTHk^1FtNtkqw&2p;vV0>PpawYy^98)So7Ya-&dXCPLbp` zKeRl%y{y-M^U+#=rFQ6(dt%@E+g`D!rfuJTfz}zK0b}cKM*~-V(8?uwd%et)v&ZYs z^TMC~?61bSqZ#Bb#(&-*mNTxNXc}>Bu4CnAtiBIx;e0+h?|NYU+&J4AA6oj^$1{k} zi7%^nX>;jnS9tBfI?tWeI>VmMv|r6Oe_y;-eUSUW)-Ko|LG_Kbcajf$``Y88WygxK z=6!~N)whqf?g@L&i@KI;zO`Z+V>{OGCk?*^GGL#`HTy4WjIr;n31a8guIQ>(8-Qb9 zH9qG^8BVOer_TBAmGI`&&yAaP`0n;Z(}>L_@f%;%T@u^>N$flA0-Z;#nic>)us5tn z-`(r;gXRKXKGxq!d=Gx_Z0H{NVSd-xBXVKD>1)S*X&R^FJG15553gow?JfU-Eghlt z&X-Sc-I|W|83N9GOJfkhUc`0qpK$hgVfB4BYo2Phz2Q$ZZ}V8^3~(W~-wo?=p0muU zd*RzB@oPuyBqzRpp_W|3cdr+)wiC;h&vUtt^*6QH2R6{^Tch6I-+=xCFD5zdKx;mT zofXFT9(#BycgF(H+1FX-MD++PGmP!qu+EU1#Of1H?X(u|hE+S^&HHTDwfMeU;+t3d z1m-#zaP{7yykDQc0eyFM&e8NQYmZKia|W=dCv*E)^M0?7SbFzAGn{|W@YPJ9#uK@G zt-SB0V;iwt3bn?8J-~gymp%!ov%mZMEcQ~b-?1L=)VY3ry5E!Bh}Fkg4L^aL+Mdfg zJ3;k=*E(BHP2itnF5nut!q3>fsCyDSUoL=}A3iZ~x1IA{zVPjD-t#N|Ov0eopBDmM z{T>DHO!FSCIrks2p9O1el54;_^L;h&QcF*H>(1ODZb1(Ceb-uX!9O*|AJ%}j^Ue+Y z({ti`5AtuUv08iESKY+#aCPU_!#9v{gPf;wv1a-Hp7&LI#kzORC2wE*+qd?lU&6UV z$OC%|KLhXk+NlRIeqblKky9^Dt?!|ssUEl^_yewNu3B?XxYDTS9Dfp<%Ut*D&I8}$ z^ISj+kZY%UHvE9AJI2a6v*W9E=Np*#0k0px9v%DycLnN=t@YU+z|WpD&8bfyUplzr zvE~MNp$+VeRl-`&)N9Ou|Kyzl&PVC4GT z`0oDr`>1_A=3nG|{~f=0Yy;*J+{k;6GrsR)ewu?H@b2R2SnDp>jt!u$KHq2`@M*kx z>#k~<-(uy|3w7drmem@2+^K4|J)dWLS!>R_v8J?qb^`u`xcUTEa7LmJdXsyaH+G-< zn&&gBePOTU2flf=)=uEb;14wR^2}JA+Q9z}eZ!7FAn_Y~=IY}N^FjQA4?(P^`oQ}x z)O4)7xx<5egA>-OsRLinLf{>FW~_Xm(YJ@cJ2~sAUR?rm39Q#X(RJ(t>*uTIzOS}? z_YmqD_ytW8Thrw@Gx$T~1{&iaA8_>^v2Xaft=bNJcW-p|tMA8Hb2oTwV4aaBq3=1Q&fY=Xuh4;C$!W z%dbgw54GH}+HXEu--|Ws-)NeE^H`@JxjvI>jO`obW~_I7#Xo;nCpXBSS*u=1C!D?FzIc7-TT?TMs}5bqzmQAp1m*|4`2bamQPYTLE<0k?cC7n_yqtH% zF)!p_yRKu+f5p~`rP25OJ3HY#7d5&2;?^#3S2Xq+$4BECxwPkAY8N=qpvF3T*4HZ5 zdDSyl?Uncjnmk+r^ZKF(z&3sEv9$VkIL^abjvD{2OP$Z+*N}bz<2`x@+PYJ?An$Hk z_c~g4p0U3Uut|Pks|P#iyGw_+=bPG!<%sPQYUtCjgV@@@zSM3_68rp~bAM0sxj);n z_{6G}ZyR#|4@Qql=3430?z~$;~PdSe>^n-jP5`49(54vEliS^VRYP4xjIcxRdTJAbDJ=R^{ z%)!@8Y>SnTgSTFv9fA;CQexG^V$$?4U6Ul7i)8FSzI8W`H z=;`BeUd_O&dBQu#-UFmDw#!)~2j2IktLKK7^E}jWsxO+JOBedyo!Z!Wbf>PNGw*S} zJ9#h8tr^&Q7sed#ss+e>AJ_nLH)_>t2b}$k&2{0N1RQJ3pQB!#_Q{Nib!U$=t+nP2 zXB@Gf$G)`_IctM_$6EUYZSzw;0Q;0?_j9JbCpw?|0qh;b6Knp=ANVi&n3sRxH=w=U zcTKL(Mv}LOJ?)*K!_|GDw{Gs(ee8E?dah@wXNI@OfGZ7Fo%QZ=u@)}G)+PQq9ba9| zfUNm&*34~*>bcJy+q>^xKAZ5^*LhDc@arzI*PWlSe(w*gclL(wO<`QXFun(+;gx=-kr{|zp&nZ?ySZ+kt=qun>?}e>TEl;xwH@JjR)3KJ!9QTV1-(F zbq{(vuXtkT4SDL-yv%dlPwfy15$n#&hwGb@x8EebV68c`>|y?WI5aPPI5qBEdN}*u zh+iN;^?>tE?VaE^bl!{SsV?9_OMFm=d;smgSF27FEkThoqod)V18t)W9-f|7gs#K@9O+r3X1Bl=6ZU@`V35Lhfl2e2I@Yx z9dgvTgF74hyOsMmJI#X*>~XuV{erj=o0He?K;K@*W8-&vd8`@u1N%*9E>`ZFg}Z@N z2fBH1_Oo7J-hI4F{km&%`retf4RkPK&23mu27iDjy2z!OSWm>}cn+WCI`{nJQDe{gJpXm+G8d=zhtKs)6V4u~9-X~n`|2uh4F4qlN9z7h zP`&9t$u)8V>z%7no0~KLfQud5(>XrN4Geg5O>mAbg5bXQVvf2G^bORWT6j+OuQr{QysD<@No3KJouVuNezx*4P@g zb*uv|_RiQVu{~O@oqaarPOLVu12nkWGZxp#)!7>B)m_nzSodRZxe2TnPtXIO+{D_` zTI&*Pto2|d@Nd?_ zPkd|r%)WP>KPx_fI*)xutRD7xle6Y`@C#_PYp;r-p7#~su>buh zM6MpC-7}LOr?1xc_XO@SV%HNVaR>G==e-!K@iQJ+=Tr^6ngs5XZ3(R!{s7a`u!T*bA{a`%XCP{cP3ArHdc} z^}OU`^Y!oBz$bW8FR!*!U!R%nn;-hXXD~O7bKU#NY0Wom*9QSImv5XKF!h|?d+iMG z=VE|*CVA#q<9+x{-c=`dMqo$c1NSgDv0}uUDX%}U>iu`oujs^cTkEQSTc2<1>)EZ@ zxyn;-ojZ2?x$(O(+D2?It-XDQ!~c`dkD3G=`(~_OrRUw;;OhC|J?_#22iV8`zM`4L z?$bdK2k}rt+~mM-VB~jQnx{wC+Sh(t?{nE(-ucyw_`DzEms&O8jc;$C+1OJxvHGYp z|Ihwm_rtH*%=EcMNBhS9LXQEu7F~UPVC~U?`$X*i?3eg+$2V#`_L_+U0t|4T@q$K9 z%@yA`)etw4-+MdDJ!>YsdF$lrF5o7xw{Y`uSSa_5e#A3G0tXSjnsCTNL{oHpRi zXQ08VmoJSr7KZ*l*4W3I2fms&_PmEkIC;*WKG)aY2xOr2xt>>@eG}a7Yxa59>L<9s zyI)7s=~?eV9^bDKcky$61z*9hz&YNZeZ7wx{`vPF-RtyI=L|K0R@mQ}KUo73{0t^J zXIRsKnwG2Hx&bwd+5Y! z2I!#C$InLW*qxl^xooRG@iQUKdk1hqTzc-oyYMsWJ*%zH8=U|Bw%QrrbM~^nftrCe z9=Y=N@I0~d)+BZi-@p|N;sz!NkxSF@3(Rw;+kJgj8`fAW)DN`nbL~();SzMe| zO>0m4{CUl^KH;s^s%Zg;eJ;_BIgdN@K;-w?$NwML!Fu~x-|?#tMmrK~j`IuFz7Yi5 zFRgjG1Rmm<>uboJ_1|zAvP=86XZJ>v1Z=3@AFPL^&)b?S!(4H)bql%9h*yH zXR8Tg;BKDZw0Cm!*RNc{#r|Bun|qOS_b2e{%=aKS&<*t*b?#%oH}J?suDSM9Z``r& zFtPSLwdeCv@0l4_`{b(oP>24Cb;g9ZUaiMmfte#Ah|R0(Ai;y$fHT*U|I>1Ue#R5H zS7~tN{F^HD)ZNTxI z6MLRPz4a{@CN^UAb5_^47ML@xyV4pnQ&bI_>T}1E7oiUZe&+tnV+SyAkHGqMrJcaJ z_UZUFOW$Yvhn4>s?FX6**3XDN)cZZ#SuMW*Yzh3cuY4Eoocnhr>>p{+mNoPCu8bw`*;;#hZYZ{q>#9*Gz9n4Df`jpXd1Io?yhvnYXv+Cv?UES}f~qwO8y7Xs3Bj#Qu!1llx~v0(beg z0~-5v;9Q@1;y3hXMuV$+jrGjtJHB(xsT&FK1OGg?yVx(uHDF!c~c4yQa{FY18&{;E#fqnpa=c`YA^Tr3p&PZ$n0n}>Mdna1w$_+t$ zqeozqxOQ?kj`5Y8a9?}rtMxPV#21~oqxG|I&K~|=(#(PS+{1VHZyM4A&iD$NdHOrk zeHzxj*6MpAzwctKzTPMN(Fe8ajQyTBL~IVP>dH?U>`&mW(uB?S5Uyx^+saN@Cf>&&~0 z$K6hix+D&)yuQyr^x@=vS9O={fBjtVbim79(b`*IZJ_nJ?9+kI;U1HmTKhJvzkiu; z;6`3#E=|C7{08i2ZS@a0;{S_UkD3vyPscvVSyLz0*IFh$HX5AICfC8yPk8r~^I3ei zb8F<9@Wl7ulkIc)LJBT(@4y|%`+V~5T>p;liQ%djc9`Fox#s%b2fqF6;~w6beOfHN zy(fO4QLDXv**ma%y$ka8H|LJW?{XUxTQM5_hBfD%*v~U!&2YY)`DVYPNjUu6I`?t@ zBzA7YdaRe1y8-iO&lm9%yn)a903FoHJu}uWa<|XD6a7BaeOdKA*PK6*_l}(D9FMcU zqLYij=He=k_Cij$zk8T}U|VcphQ243E^~1~-0*!@H)}ZFiJZG#fxNS;AI1~+^>uFk z0y6dt_=&I98Rk2F$M-qjl0E#~O8?q`k1z61C?o-5Wn zudi2ovP0xT)##&fr-?s+v(&kRxlSx+o!nDij?;6!8}pvJr(8R{(Pcmqo42>M6F-2* zelOtHj(6ozXP=23vHo)!bH?svFZcb}Twou>9sWGGI{T>Am#_PzuR81H1AlByVjGCu zJJ^F*tvPpnfYPDS*3S41jq?KT?4aM!{|>%Aj~XCv-|}PaRQusiV`g`3(|2c(x%5wJ zou|%x5Zh+}ZOCC0eu9oK)K0A@3Ca1rUVBp)t4^JH<6rLIpXY^deFtUPQQcpxWj}p$ zT6@TSyDw@3&htW_K_2Vek*n@k{L795&YbuWG@*`}Bd~(FVdWq9#Rajy!&`4YiOtKO z&$%;BzV=4{fSc&1I;?ZNWECkWy$ ztRt>_!TMYiJLEV)Z2o8)zP(Q0f$wMDW36`+vEH%wZ@y#g@Am`i&3h`x{Ni%WweN^M zAFfU;Q~KP`z3Y`-v8M#%nRnvq_=x;GWmnOJdx{}>{az>y4;^vf3>II%le(e zt6FEMtv=?!1kRLe8~Poj@jg9p(8-s3#?iCp)gZsI%Z>VD)CUtI4E2kdipRqGka zAA5%VP9C=EXW(qq^Al>V>w;P4jjLV`$Qj?<2Rn!p^lclQzIUe1Tmu*N^bGtn&mL;j zo3Dd!P3C&WfmL(rv>vS*dv<(nYX7|Ww)5wGsF5puuKKz|xFJv7^NuB_?tD(F2CnMR zm>(d3`kOV(R`=u?^v&gld;<9i=RW!a-}!@F0QG7E&U>+@dSKn{0cXDDf?s!P?BVx+ zYeeO-lU!hHw_MhK>iy?i&kO7K{n{zle)fFse873E>!4|4?L3UpO{{bO_}mSR`is5M z1ip0_eCy8l1)ob@{qr{YhVOj}_k4ib|I6~_?VX?wyypTx6DRgF;Hy3FG`2s;dnZZ$ zpg7|C`9$8jz`A#b?>km!O#|8ie7@qg=0SY&FZ>T+ozI}ozCPEvs~h?E*1X`)Y&8>o z->}xx1E4P+-}e@Y#h0_c+PYU}%GJ+xe4od8p1M<=*mHOf*3MkND-3*d&ppe!3z%@~ zSiwC%w`2T%Vm6oVrPT#aFL( z0{e_y`9bV^7V__1*Wn-d>e`0S)A8;5!1owC-~9vshMm}pHWvTFw_oi|PMt7c??+3| z0XH_6_+*ghL)M(v(=+k2oPy9Nuo;k4%jlBJ;4sF1XjfVA}Du+M8gSdJf zK9^pNr?h`v``9OdR^FW&zIajZbK0wFnBC#^8yMOEoSEx)%@KPC`dI7lpKAW7O|(Aq zoBr0h{|8Vr(RTCH{teFOcPDvIy$j+V1PM6$nd^?Mds_FoKL_|-zv6m-^ zGuF?PmNIKj?U1WhuI@nxeZ+#H{R{9j(1D}TzmRvAUqKQ(S9lj!@D((8XZxSamrv}^ zmWefXXZ!SApYeh9ey0}r4Bms9jJ6I|opTyCh^^7D6S%o*-2V#Py8v(QDTj53p#`VM zfGaJU8+FbV^*)(XmUj`kdTX8Y8?aXz$I?}f(AeV)5wZ5Ge3Bq`W@5b``2jpL*7vhP zf9uXplh~OzFyP&*p6kAPduJ^H$G#r9|L$}3c8|>Od)ixA(}C7_Xu*}*4=~{B8Svdf zo!sTVY{U065dvTB3*SEa=KcQP@2hXBAvdwYem={@hB3IIclYwCOE_y%z<2QaTt92V zoX>_m{*%1gwmExSUtrdc@ z7{9TB_0HT%mOKcnfne$Ht-vAa$DzvyG1Sns9|vHOiXmUlAXALihk738f6 zVtcrYIrZi{z8dr9I=(u&GefRGfA@Ys));(|6LfvG|E!d(k_DIga*4+yft! zj^}c|T+5Yr-oV#hfV_HjA%9r z>zrjDYt8puTw*)1zn28L2WZ6Vg4_jr=IYm-lDEDI5B$vX%2#9l>*fdk%UbsFJn)@S zbMN@M{KUqFchKeO<>zt)Iq=4x+ka78HN;Q++AXon(VzJ0t?R%YCe(2u7lOL-kJ;A5B2!ITXPdUsIgy5K&)2H&v54%oMWF3 z-di7NtkZVz#{cQWyeB!2b@r`T%X^&Ix`$k`ed`{Lsl75oYmKv?#Ll{56YYe55ZgEK zefB!BYD)(Wux_mQuAYwK>jtJ;(NH$Rug8vCiw?Ymd) zMttsJ!j%^5*HL0U);X_%8Q*`O^f_w}_OXZgPX5d>j)dpfE78a2+OPgM?X7{U=OFLw z0qkX+c3>wkZrWJ&^>x1My|*Oi*QoEmsnHU5d@&L$@YTfjCz?=4eACBHQ29Tv=Y68L zE>1!He&4rQ_o~w0yQ6B_Ps^>*01UBqwa;q{XLQEYfQcic<=0q zozbetkcShmJ2loGF5vO6*2xjO=M@Zm?S!{i`PkB^|0}RB+=DpKRS#@~Gq%>8&*jg% z8`g8Vf3G*Lx>$FRxoZ76B|n5Y(5Ux&ko!*fI$DsRi=du-om{ibJKLN+3i$SPm&k=A zw!eC{6Wg%*b#m?FJ<1JG|NRlHac29bcb<9iZ3ccFEqJ1F*ZN%3Q*D3_=WM?gzJ0yH z+h0w?>R+wLs+WIZJMqL1pmnzI?i}kp=7s+pA?K_{JV5=-fb+Q9&3Z5c3!L``<>&ss zd-CvpE&^S^>DM71x$xY){oJ>0B=G?9&WmJ!`qevuGd`a?wLvbh6JA}zzf#w+12|jW zxM|5b%UlDc;rZ$iztGe@>iy!|E5QxUyu4f!x#moKxw<2LdE@P0sdc|Pxjy%AxR2da zeFAOu;CYN63y?bN7mi>I>=H17*6W{Q)fsMrH*_`9M zcmZpj;civ6c`#*p=t^NFG3VU?WfW9XJE^$C^Tkj6aeNLbA zyyL!VMj*f!=iSH&pVMcX_>l`2V*C9KIM00>zHmkle2(_~J4t?jz1|n+J&eRhJHbyt z?G@`ha~|haf9#E%@e7-=?~=Q=mwOMmNnVXeP^}nN!<$tpC2l%f&+ZaD%uBv>)I+vps^i-Z{D?p1F6yKH&z|+IkMO z?j?L)^`f2$>%5wYKY{mU{RO>qu75zn`+GviHgW&nhp6_juh0Aos2zAm>d)7sJ9y65 z!%Td%wRfz!_HA&^)(bsea;R|GzJHSYO@6ZWx3Vy)}o z4Iew-u;%PvKdmdG{~i40~B? zjeNt31Or5Di{I5_?RUB3v#;lyrI*JVXW1jb1jjemh{xuh^f5PqJ6rb^>)iyLchf%i z400`(?@YA~BzU9ix%ve4K35WJeJ9?lI(_W`*Dve)>=kY%IiS7$A^s7K^SvAU`OmLf zkJ^Ye!~DeejNlJ?)ZOBHSLQl+?mj4LcV`EXcdoVWRa)Z6#RZ&Nd-~5Z>L$pIz`(8zHu^xd{`)O9em z+~@7gCm!HwjD6g(xscche2#yYqR!`(o7=y_*ZZM9^x;E4SKK&0tKWyM^JFZ}_yPaJ zCnRv+3Dh@i&BC4;a@Gv2x`E$7$Di98H|nib^MRk(fFD5H66_fRU-W>}$Ljm;e6G?H zPtZVdlW(MR2R#({4y zt-St0=K5zE`3rUKG=MW6Snt7SYUc0jbMIXB+UL%9Vm}*d)Y{)31>%7}fxG#x|9I4N zU{5v9vriD)*Pm5t?Ax%V!?#9lr{5qpU%L`l&bxsNV84mA-e>v=1isJgJy}~b;jAql zPJR$uSF?yK=Un`&eXwe7`06IHzJ1krMl4wC>AC-R)3ENMuJrUher=J!5zzv$Hku-@Z5dt`jyrFO{u_XFgfXuSh9r`|dCx6i;@qxHXc`Zu)xGnel; z;XD!ikt=?}Py7Mb1$UCzfa@TEGp)Jo9%_B=A9rm-)8w$$bbPhe1hF&Jx|ca?%^m&B zbzjyzKr^QXf8YzX1DyNla0#?+BdF20U%g-EP5W_r@<-6a82`Kjxf^wU=G^fExQBcK zXS)Aa;P><^cmngzanE`#?rp%?SKeAr=C&wocskOx%c+}xPyJ{nZSK( zXRK!=e^Ar0S2%ZmfETb|)neVjdil72=grq0k*}ReJ;>ic)hngOzV#dl0o+-QzPZ5K ze}K7duPxR*pVRnftk0P6x2?%U$FZm9f_2}Bwcqo;p22!^)?VQ3-E(md_4qaSuWMHA zgU&wIKGArW)*h|B^wlKRqu;;)>fL=B=hFE1Q9i>Pc${OcntEpR*6Mfo01c?SQ8V;G z#ImQ}oSg6N1@se?p4xz$#OAEGznWjZ;DhtMy}_ORE<5Kx*G>E?pcjrWA>_} z2E0K6=O^$~KX!2k_w3|v^md=NwF4{f`y8>}-O0JLz3Oi0W1V+Au!&ZE2XenrAMiKq zg}A{RmxkITPGULx9p3x&F6zEN*NldCqO&G|J>6gapVZRlM{@R^J4to^+k_3mV?zYo0d6JPFnUz+O8^F8H2 zBX9l0w@$y|U(Ch&{0z7cV)f6R8(8ne=lKeX$GVp@?2}-|(ipp=b^!0tW4<)_?&!=8 zYIkfHQ)kV+J$-_wrojx%;2cx%-^tb=#g^SgG3w~xHfRhloRPTkb&dGksd@byT$rO) zjl8EvncCBQCOY6;cd+(%d_R|u%?)&uoOh;PsBy-*Z~LYo7w9Iwbq!eOjxE=_bf*rl z4Y&fa@5}j#Z$38MedNl?o!Fk%Rt;9)o?^!8b3xPLC#YKbH+*ZJ=<2?R5BIV@mUhT{ za^E>_$4@X~?OV@)?`JIF&HMfn&Uxn5H!wD4EUo#IyMAupKGZjI7kujiOikRx*<;>b)}MV(FLyCF&{}shW|wo+`#W+348HH> zXa})E!<~o?fqVmpCL*BUqUi}NsK3a37A-)=81M5*!_mKO)i-&D+ z59{@T&(y#OEXSQ|UT%%rSbf!WaQ2Wt{muRGxp&W2J59I?_0DVf)jL=Jz`C#2dHOeX z#2Hvly&InUJil6V13Q7cC)ORjpHA-EEPB4+&52Y)zM((w&_2cytEau2xNi-%QD>~C z_9L!Y^tV>Hhr1sCf&T<=xjVbYx(B~bCi>s8K2zYE3uE>_Gdi4kzcx=_wdS>noj}d6 zXgtpJ*xwpaJ+*LEgEjAYtG}1K?gd|6kG&Iq-2-v$ijD0zY~4#@fqQzKchisWea7(< zj$D0Bp}VRBwIjLO2g-gAcSWJfm*Hl>wUd5a}`r-kF!HPAGK8j z-_Y5s;ZGp1e>n5!_w_JGY~O*OK-)lUzpT$f>~r>=)6Zyi>iO~eM$4u5e(G7R1CzWt zk27m0eCz6Y5}q1kPt}|_|L}DY>z{+F|L3}E{kn5(qV>G+XRf&y{05G7egki~>Vu7h z^WgWy0)LKc&BfNZ*NDaGoBtQ=(bWKt)_8*YnhWB;E-vX~{RF<>dQNl$e4CB?L}!nh zf$#G)V(YIUHYZ>4eQS|f=Vv3ZKC`o(`A>Qw8ED+$0y_Cbujc6O(cp#Om*fLqc(fBF zFh6r?oG;gfHN>&~`2~&{?mgiexqo_dxw*{M-~FxoiJFPE_JwZ7%B%PM@VRO$$W>!q zJs0*$U5~cDj}coRay(Dn8$Iit!Pm2sv)-DH^<3auE|b9?^s?s!Bfrnp_b$zO>X{P@ z>|tM@%UG@b9>!Q}YXAG{>$6TY?p<}n)+Ok%?~AnG2hV%g4QYPQ9nYyPJLOB|Ptk^7E~@q@g*g18xD)z;Sv zvAxvQ=LOz;2dQPh3$@m{ub&yU<_=f)L>%yPspZ(`g?{9M_ogrJ--|xUHT(pTYgWbZ z4^a05?@@Dw6ZSB6w~1|P;r!b6eL1JYt9jxlIgfRI4chwxIyvjCuRgdw>)RZ?&--a; z>=i(Lke3S)o8S5;IrHM|SG8z6TI)SpdG$Yn7f5gg&L5zWcOUnteX&1$lCy5}ALx81 z&I)9?=NEK9Q@B z@1}z{P&44K`q+y(tbWShTkXc5HOt8s{FMRt{4>+|C_$d$EOHRGkyU|)_Km7n#bpbdd_#*-N zfi-VDw@vH~{fs}<;mg}Y>rN-1aPDvJ>=tN(*clC9y>~nHW98JO@rHe@k+95wktt+0?1^zq_Vdd1+ zvy$`t+s`#a-)D)~bDSmqynl|sdq(mH=z#%2jn9?9J*=7LsT<_Imrfj@X7Bjx7i!=< zvFk>z?;mPC)~oTE{Lcn|#M-07)xoNr;JR}IUnKZY%RZ-1#LBs|HIw*-uhwIH_SPPK zhi~`Q$KDtFlJEEpjKF}>Y2oA_=+v4M_E1wkp~w9q$}N88`uy9c53kEw5blx?-w6(9A{kN{Lfv?Tc@_e2Qb&+9&mN= zpC^xRO%h+g3)tTt?&q_3oaN_Ljq@VV5;yFbNKXCSW zZtXy$)_eobH1>|3paJ_PxPi8F_7u)dIMKlXo(Fv7(pab8f%ok5S=%>n&i`(pcYN-B zz(q9vSa_qYca)o>hkYBcm->kx;PfF|Gp}&}f&LRs2R{?r&zdK;fi%ax_`GT+@D8pZ zfq#GQ`~m8DV$E_#=Uvd$uGCrE(M?caPsa3?3($xgcEsXKk3Zn$9^};9)7ZWfxR06u zPc?An~wU(ZVX8@_Rm1$7-v;ED8)rO|ip z+0C!#+J$+I+yJ!~8t=_L+-t__SI;U>-a8tg`oUSB=axj@^U2+WMY_#;?J^I^Vh$YsbIg-+<5J zzRvM(hFXxI>w9wDy|F#q=XzhW9{4`98vQ!a+W+(kg_W=V?BH}k1K)|QeZ$>=^;hiAz+EP>dF!7*f5hU0IEXv= z1wXOYCy4v<`Ln*y`0vHqqd?BOpR5V$x%);e=Ux~2?VtDJb4Dz!+zZWs^PhvfBkQg8 zxYwVU=f7X|#6SFqwfm6E^}PCTV86OoYW&YSCtBxNYmb56IS>3!Zh(45=mw|W)A8Tv zKClt%IU-2_dsQ9&6OA=j>{qnzbCJVdiLJR|?S0X5%)RifwRhc(7TDJwY6fr*wdR^J zHi(_q@SPRds=<2P?{JwbwwHZ=2RCaub)Up_?t* zoafc|-qcTU!B;=keObK{U+&DYr+rSWZ=E#*xMvgi{{GYGQ(9bSwOsGt9U|6j??ip> zSdDY!&u76}eUEow&*}$XdUEQYVCMQBr{){X$NG2ksHlD2!5)F$ueZ{oOJa|AP+XuP z?~cZvk*i;W_uV@80`%3`L%*KszUDb+#NyPrkMqx6T+y3%mi5jrJ$@6h`(QoQ1I-Nj zVD0Z6CH~pLxO$RPBX6JA_I-CjJjg%c-oTw6z|(WRFKg^0KjDnEYR?RNc?SuvF3~u1 z5c}EJI^RC#Zg35_i~ZjBl?!~)jC0l4FYupQb6{$7)$|?LK5&PRFVusePW{b#@&mZH zJGj%imw`s!y^2Hk`drUyAD_?O&I_Pl+}6NUom}3Hh?75p7^`nz=MK0!==1}~YZLg) z4Qnq^$M|#girjs$1Abs1=&r!pn&-2TQybU@&hr{S(I)V-S~IY6>eUW>^=b#`x%+R8 z@eO}sr+Mzqci=s{gR#7Qe)#;ZOZ4UrnkN9dz`n8iaz9~xF84Dx5*|3$-TZ#5^<=C) zo$nrV$MQa7?U_s0u>t39=It39jo96cuhxN|LFMlahdeoZ)@(R)*0gQSmmhqF$Unh^ zbMJu-_$lz4IsCu|@zL4)8&BTjjLg-e$LdpaseRE~zmvC*&)3j+B37NdsB?GcmLA{y z$c2gD!Qs?5a`vrRgdX)(PoEYmuTI|H)|)>z;rq@V|5)p%IT7D|@8BTFz1hXjgwH1b zzz+D2`})lJz5EH}YFBJSGqoUrb?$V7Hzx+M`3?fesVR$Jb!zYHGjYW)Ew=QzKKJ&X z_!;Y-y}O!4-4oxL*4ekA=^}RzG;d%Jk28X}rP1Pjrc8Q$k|RIy2e6+xYm8@Xy}jT9 zu7T@)zk9;-`K=srgnz6)7i#q7?ANhA`;YKR>~s0Gr$+yBUo(?fUEKq|+IkPr{tl|% zxo2kS(YW^;)Vsm{4Ai=>xf`e)+zrm254Z`OAp*Yvbsb%UQ|r!w)gSN^bbRxPe|8_l z&hog+%%zW5ej0DCp6SHc3;LvI?MNKJURvi(tTn=U9`|uC`+I6{cSrM(PyAT-_WH7P z>Koh@d!AE`J&Z@TdM@um7~AiKUd;pz&UoNIvHpF9zCC>o^-t_gpPt4ZHP-he;6$vv zeCEesWR19}`ce8gSweXi+V=RoVbx&(<#@M!byxP@$ojEqt*w?>Pv93-o zV?VX^xy3%HeS;DEe&9TrtB1b5+9$2l0q;e97vzNfoELERbjQT^pHIbObKjHKzNMH|7kEOT9XLSCV+whxtc2L)n5cE%=1s=cNMy}o;_=#_=Hh^~O$I=b_ z>W9YOjrf7DHTHR~cYJmdeyt`s^BHj1dOmE`z{%@h(b{9;w*bc55>;)$nTy17BH)>$ zb?({4xg+7>ysw{#U!e5x6Me_3X~55rIs5z0QXYHqH5bl)Kf>3u@Sac(1C)%IA~!R?7=}} z1~Y+!!Z>)rFmzzd3(MIF9PG(%G+KK{Co(&{nJc{E32*4%Z&iKgT-nBf5*P1()>{Ad zU#nKt_jT`c&pq!pti1xRKC_JlH;Ao2lbiSx&YFgwAh!QN(|}r^n|Za)sqMf!Qd3&F zJ?rsK*~|B#Vf6>-+E{U5Pv)-x-?}^T6Lfjj%#A<-I%v62XTZ0ob9mnBLmb4Pfi*$w zs%MvrbI;m``}E!MS=eU=LU1Ft-|@ViOE~XD*SF621|Gn9st-N&8@d4!)I9_-x*%uY zC%$`i^6CbD!v@ya8J$6W5O=w~KD{r!ocHE_*6TY%^&)Oy=9(dQEa%J(_|D$8-aUjp z2XMvK%v{gjoPC_X;ZOWp^oU$_p3&tjBfig~f9Mn6v)RWzT$%i$kJUfnSI^iScEYQ- z-hOKAqyB;KSzHnR^8GB{GL8lN@(f9A*YFIPLo4uZyoe5)tbG zH)3fcmw&Ma7$?@=*1XE?SdINVyvx1a%d?Cty4ddrwlul&bzk@{d)PO?*s#Cc4O{(x zd8e^@+;{)aXcBG0YhBjVzbn8tpr%%?zID&BzN_5J8G@XscMt2{ehx0z@0@+q_^dlt zf5h7RcqZ1D5AQPHse1*9Z|xnw_R8H`UXZhIBR{Ya_}iuyXzTvk(zxF@Zs$CowdziL zud2uQOiwV;4$y!(_tqZ$oK<)~wIBWfNB#jHx!&g^-^t(T)8W+x@%f0EMb3E|`9{vX zd(Q1!t3Tk(sVfcsD|JD9#+vVN`p#!fm)m;`V$sZNbDcM_0cu^BKCqtMIiGUOXx$CW zT;I`0%~3C3XGUuM9XPv7O$YXJ^;ol?h$nvJ_s_%;`+2dZfwH(CaNgCBb0(~3wciDM zxqjzsZ}_ouf%dGHI&JmfZqht;>Jp9pj$9{h8v}mg*L}wK?}vu(OdV^^e%ksyTdvq1 z3DkIISLt%a`VV?|X7})Y99VJmY;d36*XN|ma}IM2-^e47fZX2sk zjeGsLHOBV&yluxGd!EtQPaD)UeCKd~Ij#30Do^Y(&P2=H_O{0`_6+!6w%)wjw)I-| z?Yin)?iwijxt!Of?vC#r3w!-?xu;t9jkopAln?Jd6Pw$YuX=3gGpDuAb_1^BbJZl) zo<|S&YUu6jirBsJXRg|Yb=j}>3IcuX8G8rr@xa<^z#r#wKQ+!{t^JR_ot*P(?N{@X z58}i&ebC8`B;LzDRSR$4y3EyU5Ic|iPkfhJXA5j0)(r9uzxvVBW&OdmTs{2zX&3s* zn?KHR)OiN`b~P-u8B6CJ)(N?v-*B3So zIL~Mgm$O*YV(rzpH5Trfz+OQtp4Je1pPktJ8DD+FS9>yt{jHY%GvKfI%&T7R8;H#x zkPl;cKg0I1|G)-l^l93_2y)(~xe1Ov%-hSH_gOVq^J>Na^8WpOa@Mi=9;=r%6F)$m z+qtP1-lIJlx%R&A!~SUQ*JYDjH=q0u1pWkVTl*YmQ=@+ePoO{H2HIEr22R&?w~2p# zU;V68cbv;}1U0T6>w8%4M_`}GohLVebtm`^|G@gsjn=uBOFw{1t@p2PT`xaIQyT~kPH0ZeZi8qZ@}C8 zC#=4C;qwvI!!yCT>|f7>T*tSk^%rVg)^>b**;8%h%wJbLW1Tb6UgU`FKY$v0T4S&3 zPdtF)H$?xV31zGh7Wz#54j_}x>Vn^^sv$6oH&;Ofu9h$p`1cmkiX z_u^i$efKl>A&0*DQ*$%F*N8RC%+%C~ zTJs(Mv!JFEC%iS{0@tv$u+}(l(=v}*^Bp_3{ji39Y3$7CZ}!F-*NSC~C%)R!Qy1ol z?a{T=j##=goNzAR=enbQ+@gMMaP`zUD)(Z2!00f@a%mvMqd?hI$#lX&N?70&PZFK^5M0@S;-@59=SA4X1v1PTzG_D^_gJ=Y2AEFIrFRhQs_e*tQpv60sfWBP-#9qW1RvC#;Plm&25avpuRdC) zf!o{iTde+Sth4unoOx~S33t*bA7Fx4V%J#<>>Jl*?Q?uq)HiV1I_v5#V3{hO#^p$mpvPh7e3>GodCI^&^(vntDgj zc+Lca{4uxv8u2lcap~oO+UkMte|$(#R)p3eRNXM9)9n(K`2WzDx*bnely^7`)Y z{U$X}Y?G(|E54e5??B&o%5zr_YVB{n0X04Ffm-LuWKOKB^3<3cNe*7F`oUFiY|E9m zkNc<(IAiMqtEMcjd=hIr@C^R^bN>M5Y6kQ4v-UWHHABE%(8{~?BbUC;7)x^}ANZpB zV%47EyL>L4zCDd6aM|bRZ_Zx+e7XS7Ub5L@7~HR`X6*PgxmRn5J)_q9e!$IKGhWOCe{VA28gQP=6+c$blNy%$WvrR) zQTq`GaG6tM5AE32de3W|;4ATnRoAcUtmeCVV$WpFByPm6POQdfUVK8v!oa_g_nw}> zc>+#8K{p0xYHJ^3_|oj0^#fn6=hwfZ_imiYI`71M13j^jEb>AGjse_wmpzxKy=cw=Y%M854fv~Z{0tIla%qJgu<-kn&jcj*2B zJfr*sauX!JHJ;g??q@uxb*U8t*hAl1@5h$+_G&K`m{r-qzcH!nudKL2TZ7 z&vGX|_HrK4&7-x>9^al1JtjHp&AF;?E-UXduxHAF0rVSOV7>1atEYE0<$-%N@)M}H zM@0Mfj=0|qy?s2Bb6$YjPOfV=XWbx9Y`fxXbIt6$Cphm@UhSE>zWqQazC9Cs_xusP zvp+%28fy==4d2=1jh*e4IN?v&|3^)$i6-aGsW+<|+Yugl)F%qOP3sBs_PL4El%HR_YtbJcUipXRrwYS4MV zuDat~8e@6qut$dvYy()Sd< zeGk!FU4swu2`=QFQGL~8D~3CP(gj-JeLU6L2i*<-W5@RF+jlSPY9_36yn@8H=fF0Q zTc}r7;@1>pt{)AI^+*N}!7x1ngt4{7jT}y!8 zbDe>{8m(u{TztR{{F(>8<6C>f_xDEjyT5&Z@5p}Z=M3JFy{z+2zTFG8E^`m~>L2T_ zo#D1Q>ju2N+^gm4)rsX?k+YcFE8&ST#D z`=Sr`#3r`x3g14R*qmnwVzrZ4zG<%@@PT>#iyGrxwd%|z@GSaO_jhgGUoP%uFSygT z>W|nOXBZpCzdX;O8}ynW!9b%8Y69Dd%|QWa<%Z)8EczJu(k4c{KT5K)@48U^laMF8#C|ay8FleJ*PF~>`}e3_MOIPs~&gc zirq)P^7!>R!Jg?;=gKvUdnW!3?HyDd@n?APZINri|G)sw=$Vc?Hz(|W0S}NsuHGYT z0_Pi>3-W#Q-i3K*et?Fr)_(dm|Mp?NJ=;jqz&dA|_+7;Q^JVXny6TN=!Ur&4?*=)0 z>$@`X1zc!j^%?js&tvU*M~6uKZmcJw4a4|8bV2uFiv2-+hG7pkYM^bHlU! zeDuRNcg70$5%zH}wazxsbhtVr{!I-RmtDWQ;e#5_C zzxNM#@4%VNO*HBQ>sg03wuLwL-O*mKF7r)X0n6I6IqdP@<|O*|3`wFmx=<*b`<$9+}|Hw1ipp$47vYwuck{bR5Cb0FfR#%FOh zx8qd1-;o=fbE|9c4I9{rbw!q0_PY0nds1W0`9G|Wm2+SD4&)lvWi0Q~Z`8QzJqaA@ z)S44Vtux1=g`eO;eB7&Xt>9OXxp15*h}}!8w%&E98EYSHT-P3hyqX4_OJDR@v9{@J zbJh6y82Iktopcbf^jcR2zkt#uJ{se*THrGi1J2b0|r zulRRk_!bL^owJvPvaqj9tt#g>0Ky4TU^B3v|e1L8*`sg6KGw>d^6Wvo}*T*XV-`W-+euwR@8p*X$)^Xh=;jtC%&;7_X}(-Z1s#iC$Zm? z)^~jOoLKvo?z(z5bF7ovTshCBK8ZVMvCh}5A%2hxtmm|Ug6a?N{rMTHJ>*@))r-Eq zht||{#Ex8j8um=inVr$t?>u{px@)+at+CX>H{gC9Yixggduea#cD}KE7v$_;3qSD( zxRC3__B-nB(aG7nG~|x^Y3L`p6P!70$+Aj22zkXkSkNx)q(0+g*Zr~NI^@%@r`~}WF_PmgD z@474U>h9*rS?_a{^PQh??wjCDtW^{E#y2@&e}A9h8nM48vF_vftU0W8-5C1>-lKl? zPNW;w*sBo>=k{#Y`b_%fCswW&R^Ob@#u|Hk;NPra7WdRn_)rUKCBE~53>e~uwgoPJ z0Q-d;YbM$a&u>it?VZ>@wTL@7@4DiNT|R>oux5ZJSP|>{p=0%r*^TQSbI)W968K$F zd2$Wtk6-w(-_CRNaHcu~vwp_^b~i_z_YwQ9;0Au3pP60uasG%3Yd`%7>bwcR_elI> z9@ttlS3mD0@Ea&UR{e$A>SGV~Qq%EozqEYgl9ZyT8_cF8P~W!o_+otT9$+ z?Er~?#{Pu1<%0L`8oAy9>#OeX_WMfy0cs|!J=EAIsL{8V_j8vcR{MNg{d_O_)_7Nm ze?zA|dd7|$oPE`Yd3w5sx}gQ7p^tg*e8MGuH)eScr_IN03qei88sFh>KrVr^*LlgQ(RY2TGsv~i57hNo=k&f@o;hRj*4ldz$JV-s{!~NW zdNuY8{D(UDfwj)1l~>cYEe-t17}WcWJza_6eKy{$>%V)al-$gCnQ#l zy?q{MYDCCkYhH5d5^ezVzCWeE!l`55iN?C7g}=c2vrD~qCFiX6Z0L>}`*vVm`Sfzl z+`K`o9s?_CUj1A(O-mfW{p{_Y4am9F_E@#$Ge^zzzPQR!*U{PUMj!7wz_F?~JzXPK z@46%EJg+Na_oIL2{#k^zPhjosn#A?I=y%_>BUep}b)S=1&Yp7CCRWb#d7sYMjkWBl zCV=xK)?CB4hqce^^4+^;r{+O^kZZucH{#=t?Q0+FC$<(gfc@;_JvLwuVa_@^=R0Z# zys*wSbN66|9%~u6wO_9?oUoy3bfO`U&Sd*s%e;6X%)uwXS?$ z0Oxm4^VSG;?yK)iO^dF@sxjw04|2y{shhcepXO|neD%i4`@CZNf%emTt9~8MIMArK zXM=Oen?J+%1f0uWwPMZSUiNOe_DrlbaufSB_XhCoA#dJZ#%FR>gKu2-&wb1t+e6M} z-J!K-m6v=gs(mEC%&9BkL}CVy^yLrJ=9m8 zT&hFw8A?aK=80toR&D+4q0Ziw$G5iXvFiL>c}E|hK4Wm5+{8Ald-&O$`0Ahda`s5# zT5!fLd+4i)So%ghzm#!>?A(DQD;4VgR@8Cdxtl3z&bf? z#JZoZ4{E=*{|s$wt+DSUpRt~`bNjB=zHqJw8u$Lc-$(T%=W@RRYOZZtE>8Y{Jrdr1 z)eiiIZ+*uv$P>>U>nDEfxM7_$&+fR_TX)=P?Fr?|T>b3nJ6>_D`qVOu z)_fDJI!CO}&e7NBa|1`tJ8x(@JwxEzyJjFah->9K+YOCtkn6e5Y%lM`b5-BtUdUH( z; z51_}YRr^Zp4ECsghPn0_*bX0RK;6w13vbP3`+=Wu!o448t+BplBYz`4ZSIUMEqTx6 zz8%zO?b>R2zDRN+imLditNUWFqgS1qUU%T0H7jM#R(`HNwHI~v9>9Iw%efo=Fs9yp zCa8Vj)D?%F`WroWJ#hffGvK~peKrlQg9pCWcmR9(Od2+J{6Rd4~&BK*pMB z=Ivkn$G`gbfA@d<&+Yt=xQ2h!ZvWyx`o!NG`jr)bfBh%zKW~5Y-~U(t&42qp{MY~Z zZ~x|B|II)ClYjcBfAYuw*&kj0`OBZR|KcCInB&i#;a~l$|LuQg<0k&UX85!Jml^)# zkN)_dzxVC94wpYA;->b2n)hgb!g~aueU51I}yR zWox{HJDhPx;~7u-XzXPl>u+d;zGt`qSIuPnwsQ8e)^m4Y{UpASNUev31U1O|9HMHFBb6`g5``zj)R=f6Q;Mx%G*CxThL;>l;?S0q68A zfp7iazAvlBHG^N!B(?+JotgWN;eRGp>)psl5QD_mTH~yd>wJ}iGauOFJU>uBh=={; zbJ>Pf0_}B06gF4^W6Wjo5^(S@!^Umfo z3uEkr>-xZ-wdy)HbIoH--51t1a>0D{Kx^K8?EToap2?na);RO={Jc}=z5q27xa|2= z4*b>2`u%=~t2qM;?r&>;@5A_5#(o#OSI6%`8s9s!M-UI-*{pMa{S)}zm&F}>xxd<$ z>-i6C)dco!efRvE_tn>ZJAC9ikMrHYB$jI+Ck@Fq1F31gFU)TA% zxXg97fo-vB&2@awU_bBk1_rRthdfsOUH^@}b6MXl`OjEqc9tMlI-aMiwD9(xK>I}N zelB~ASkF&Rz3YKS-pEDE>@lAbq4in_5GaHxtPJ40TP%izQ0$W-ghs!3A%l;?&F;_e4kgC z%dKUOLF`#4oNK%-ZpL~yoZoYdTnH^WIr~^!---AqYA?hMKLbnLv7YM;Zx7?g<~`T( z-{!H$B!AdISMMbDNo~WL?;tl`sX39;Ci#H3_rx!qmig^9fV0*8qdC8<+MnU%t@WPl zS7&Z(?#mkvwLidX)lJ~}J(K&?!e*>kPK|xs?|0bC^%MF#h*-tY&Svm7vgyRJR0QD?4WGuN!J zHCN4zwiX)e{f<%Vz5xcl{r!D}XEyi1Cbq%*83}x~9V9T;f7*+A+UBg8!1HM3eMTqZ z-+x*6G2o3SeBuYV5z9S+Gq^^6->(G(v0S35GqCTE`0_JWZ*|_8_h^56`?(1;=02lO zY+x_oIPVjUd6D`3eYV%9cY4BGV;}P_n)QdwNIa z^@AFFdDfn5Ut`x%JClH#4)oopVrGw6x`elvIlueuE$928R^1@(pw0{3)RS{R^9?vh zBszN6!k6!$7S=x29q)V+Tl1t=?Psi91M(tc{~V;R{#P^|zMcbdfa95{KkjCt^GrAO zoT;O0;Q6-V+*-K={xj5F?bbK)&Kkt#2dI^6&pWvbP-8FkUBo))W!Jg4xqug*-MeaF z62D^Qf5euSlrxXSf!BJz)CcbCbMYJln3HqwTG+4GHPabujeQbc8^HA-XZ{5IOwL?; zxp%|=z~0{Q!O_S02E1q1{~7xV2KryWb2f4#xEHK_MJMj$6ZE(J&ciuJn5R$2wyizu z=jvmx1e5p)?_T!v4(rUE-C3N`e)e#=pFN5TEWT&7*5&t(d4C4F)H~~qc;cToKZ2a| zY2mH;ckj$b-v3{Et*v!F^Xf19@cKia6}k3^SZAA9^`*tS%q8%dx_nQ1;OrhfSI>!U zv3k~g=s)&;6W+7bGsXY(vd&a`Vrv65;Cbacs67&k|DayYoxFPUuWI4lFNm#o&VVmV z>~pc++(cUo|Hvo)ahBtLwAG8;Z>SkT3#`9qJI--Z<4XPqItZXve_}tVvF>MZ^!9xA zQ0M2X_*^v&EAD7KQyRzqeTzQpOir%ySoP=KM?Ueb>%bW=#FZzWK#gmVQ}f(*;3x22 zj=TBE9Pxu5JzDL>uk}9I$G-Nc=gqv%c88nfCcOPTuixPVu3_!xQhS`C_RP(Z|Mb3k zyGQkRhTS9K^zC^E#vR*|qk$8?y&Ld(xxaT%E0`nK3}x{@$T>&V;h(nFcl4%@#&=+1 z{d}p_w||Fs|A#tOor75iam4zp=-YD;TUYh?#tjrljr+Tg|LoIp&HPI2ef|VoM}K+G zn7QhTXRZiUtX$gK18t&J<7eSBaRWE-Z(r8;qwl9$>l2utSkd1=et$NeMZMPig?y0r zTw%@_xYv`oqj|-C5I20En|1mRcx$TGyesP0`ZKn~58%C=fpgur?eNFBwUgLk*iNT zEW8@~SXaI9jW6hZj^3B&7;yGEUDvD~PQCA}wa$0O{<9Z!#^Cn0=CDrRzVcr|!+&C} zndH;l+j?g96+TBb_PcDTb(wQMd)~gVq8mG>R z(^&l=cgI&Zu*diOk8f+QIxm_~k4UrbWLT6nE7kp=#_`Yict8eZ`Y`yumeS0SeHRRN5o#_kw z#J*xZr+WtW>)ZMs_I|7v_B5xa#rJ;R!3AFX1fPL>+_C@i1#Pb0kL!9K_BoUD?Cu?C zJDj?v4fZ3~wZsElh%=Yo8gtgzL*BiQo=MIXxt_Uabmr<2dp5X6zENjCt#@KC=eK_Y z_OP$oV;F-8UeQ`RKqoeLqxJ&t>ajk*&MU9hTF>dShq;;w-@RSOdu<`^3-O0|a^dB(@Q&|9oA&rX0RH_wB@YV*Q&I-p_>3#MnMhtZ~HpPShXZ_>3jI z`&U2i&egB{l2~H*vd_hueYf@!YMkeIK0lzTd7uM+khdnW?qhsH^Kai*?7Z%MAHBys`b{+)w^wOwRtUjD?EHjg(#5nd?m6x%)I?pS{nr z`oyZen`1@r3pwlD%Q-)Y<<&QIT^}o#!1-J~_w&J8r`8(zuUK{B6_|6GGfr$^{cf4D zW-JZ7nmc;;dY6lTj$x`zvhB-MtgUBYuwkk z>bGAplr@u__DP+Z&)5OifS;+vX0H95yPg-^9l!QvCiBh_TYn)}9JIYAh+MVy3*hRx zU~E5Q@4V}rQcjbF(&4f>&M(wejs}M_fM*kB~^X>O(+(9Syb8YWm zV^7#djeG$0A8?+}rIo*7PhbQjams^EESJFiuU=Z^j~cZdw7{}|8e?7B^V@pQt*`I- zM2g?VFMRvRyF6zLfE|ftc@Ap^@N?6^1#aegHr7va0erqe?4Ntfy`mjhH7?KPTxYC% z$XnyQ6W`h{h&vj;Pn|tM@#HS}_Ezuu*?g>?^}Uw+`}A$~Ys5i*96UCGcWNK!&<-tp zV4b_extDbV-+KuB{)^Z-D^Je#Kp(MZ#rJIH)&0dg0{jWiyO(Q=4!#6TVt|j#zMZ=j_DRf0#ofTr>D3 zsoxr++M~lg)P@WzR=sZKs`IWMM;+Y3tFg{=CDvMJko)wqY$tBu6*!l@d{1YtKIS@l zXP8*$X@WQcY-^qT6a9cs;29InTqkz#6L1-y$@@KPo&DvkyQ2v>=TUEufH&WOnhsy{ z=HlH$P1Rv1ypXFi;H#O&>}8Jz1~~2bf^}~-7p!}jmowHszJKiHcZfe5+*4oZKe7Me zeVy$`^7eIx7OQv1-iSvoy?skV+z$?WwuV~I*@*2Cax8n;&!u1WXaoP~YmM;)YN{XB zv)FegdEh$ka>U||&EL`4$2Ia7HSqRSXW#0LX5tr5F461Tv;D%}s6C-;;71VT)J+iN z)VjC4obwK>yt8=swVyHTOpV_;J(lhZ`o!g77VvCI2aAUr6gfrRy2W$uaT=6qE@E_pFPc$9aqt3MLz{ZAqdNAWj@dJoP~_lAFl{}J542XKGqaIcQe&yVkCC)Zxq-@9?YAm>b<$+=e}U+)sJy{((^ zJy#lZcX{mb9PQ=a)sJWE%2?0mxq0=@aMX4DK+~}9+wjdzZ1GSJ_}>8S?QNa&m~jv% z`ZLyCr_OnNe%8x-Hhs^12IfWLyMFz$p0U2#2Q?$tx!tGwV265gCy>C^WA*8|W;U+p zOaH&3ukYC}-z7@Eo&8P^pP_nt`b>j9-j8Q>?^j^nc}A>SwE-LEgS+ zIO`_R_x<$TC+r$x%3U+J`#ub$s{w<=(%gPr%#To}S12#D0Ja zaP?R{AJ;txwgG#(hx^&bncCa;J>ms_-)EAy&zZbh`5+$ni8bf`wXi4qi5m3v$1fNQ{9i)B0hGsKhUHgZq@-hC-jB{~;l^Q;jL@%wHNPkK~EB3EN-oL-%Gxg8lPcN|K&OqH6 z-q_Dbi`Bp8!}l%==$z{gTqEGoo3qB{oPMW#1%YpTQWJ}FpT4>31K+QUvxo2R!Bh{Dd=iBCgt8 zXLu!NUidt;_L}6~D*_Jw3C7ODRNsF?fAw`9df)MTtUGpIdH1S5Z+mun4>az32KFD= zzIA{!q7jkHS9~-&+248yOiPpXW=W>6SGmKa|>w;Kr;!pVe z1v-Aq)$6`{Y5T6xcLi48`gZ;P`?s5UVsl-er60i0Y$oxxBhUcz;#ep0XW-Hf;64d_ z4i}*2^=E!2Gh8Ru+S}*m`l1Eap2&#^w8Ux$*7^qI)gC>P z_?38IozMQY2Uh)AE62MU#DVX7<7e@PAMfN@pWhbe-aXfzW7~kUR*m)53+sLjUOTbQ za(`RC`~b(^coc6AjXf^ZIfp&%;eEK*MAs8&(RtPe67GWi=}HWK?`7Exbnrpmn%|JS z6U#sFe*}JKOyc6v4f4)s&5fLXi#5A@Okz2IZxrw!Xq-jfJ_$Oo_5xha;CWm>;6IFW zeRl0RV$J`p2RVD0YoNY=;7-=T2mZX`vu|-&&wb2mpA)e?8h!^sZ5p#ikjrJgtoHGseKSvPh!u0Ay)q&XZ{Y{rvYbb0{?bJ7pu>aPxuM!p|4yn7MJ?e#~!nXK6t`gZ-0A<0iU1=R^;Z; z)Hz~(FFCh4>jyT$uilr|xWmhd2JG8GfCilJ3()VmzE@7Wm%0gD?)d=fAD}d`@*~%K z;xlp&@8pKY*q$Amf!%Os`wimPYpcEWMZo_I_Z6$|2|g3MUjs8&|G++o?P;Akeec1X zcPyO28ke#D87{H*pMf6KX9L@E%{s7?UiNG+t9Ib~%spF>7cMpG&+v(Lu8F;1Ylf{i z=QHkD^Dg(zSam0Imp4da-vReB_rU&iMT-q$@2GUSe(#{B;ft^2My$_l!@Atdvka`W z-N6m~0MsOW$G1=U+|!ddh~=AkIP=aw!sCUkh-@oVNCaC+Sr?dDx z+@ryNA=awXT3fTB@jiZ53qRol>-W{fo~g4}z#BVn(}Lr!?$jQ8sY~?sxnLVs*e3%Y zoX~f;1i!_%k8^9~0!{f?MU=ji&`isRRs**@N}GX*)%Wqs9P-FpUN@WgL$ z?$v-A<2$gniOsv8wXP%IW6k03=ghy756}}Zho6x^<2`)Fw|lZN+j{Ti-TOwI_{N?suJ?--#cAc;{V*AvYuqR@B znQP)5z*+D6Tla@=Uasx|+j70HM{Ui;xf0&`jMckz#Andq+~2#dS+w%h4RZE3UmU)@ z+Y7*2tET!x4Y(WiwJ(14=j?KR9%=^oC;jw4fZB)^PprMX^V$>ZzVh~VKI=Mh%?H;& zhaW;8Skr>KZ3FItKj5r0fB5Fw+xI;}4Kq!6HSXv8>E1`5hkEkPIj(ztq-xHfRx(?K*l+6Zyx+?q$o+DD z@1X-{a?RL1i6fV9U-z-r{B!G_$GcN!pCmWH319T>-{008p8Wv=zF{xdr61TckXNg< zUX8tD=U%aXzO^oA`Vn6={L;O_`+SJ4Zvn_pZT{jtAbyg!*Ka{5t{P+k@BLojJ)`rS z;hy+4r`*fhf8^@po`LVX<1=tC`wx9!uMulD?TNVFA1$@D(3PKShWnw3^)uV~oW*@6 zxRF2Y8up9g&0X-l-ynWqjqN)@;2V1n zjktr`+d8K`zJe1_`xDljy2KyY3vtT5gIx6q?+gz#U7s1W4V}G8PiWi%zzJt?M(;%`cP6<6T6@^bnSa0z(BMjk^_`vp$8tXP<`axqaarqFA6RQL z*1OTL^)6y3aG49xwP;5!ZBkpmZ{d9M?sdmL;Om~aEB8GUn{&pdALNO>=R5Jg#GbFK zKWr`a1Ki+SteQ#O1nWdyCNl!(bq}?#K<-av>_v^s^Vq-kfp<^!_7(1B zO#=G`Rvh1FYCL!`^%XKgE0n|Ff#2$5cwQ5e@w@x3=qUMJE zioPcxXFsh!(;u7n@A*2n{2lA_)Ax>z?bEQeKlKBsZ?S$Rxz{o4fD4eho>^m=w?^LC zKfTbCxM|5v_^()NI(}e%j%T#a@*}?cdvAld&X4AV{{R7;)t=Am{_k>N)n7o1{h7cH z5TL!SeZOFhJ8(|-YYDKY2> z8vX#`G{5ql3oEa$RT%{PuX(M&kCsXbOKC(w9T)g#xu18v2z z=Ct0PkIi@D0bHK3;#lv>F;fyd^8+|*Bw{e(Z`XY`tbH5sp3cDKJ?UF>#}2IhPWa}$ z+oyRr{SSFHz@Jy8A?Iw)Gq*ps&YsR>-vIWtzn_VQm2be@2!OU0zP%c;xj_G0eD_mV zIeGjfr+vkq(YQw=c3<_0?;I|_$IkekMeYJdkigEw2iCf$_jv44CXSk`s%8Q?gAHcgf1N(Xp z7vQqz02koCrQsR7tT_Sucvseb#=7Jhuur`w`1kA5kg1$=$hjiOaZdZoyI+U1hq(qn z^Y7pPr*~p<^_hKHvj&<@ZSmynA8P1rt`@z#??J9s@GGc$fUACT=&URJg14Xl&MdLH zdpyls!x{l{y^Hqv0Cf+$m-|dK!kPr?n+SMo(73M}S0*X=h;>eNr>(hY$v?>ZE+jeg zQzSm{JGFemlRi4NulRL0*dV?c18a_@&OO9WzS-xC9kTvG?vHeTVK)AKpUG>@$n%Ov&&_XXHf zy?sT`_4~LdxgfTO^SF=u$qD-=RzBft#V_8yT=(Co9q=6-b^0IF*!MRt)Zn}Cz`Ck; zu4iq12O;1Fu&%wq?{}-k>U~)I>wEVDdq(R$`n-bJUVrey=i!~bh+6`dbIh%=R$V7{ z^;o|rw3oRb$YriScWrR;1Nh$PJF9u?h8C_CcEsWn`yjTTv*}+zgRkF9WA!`mS=rm? z7R2^Utos~LE3d{He;?xWGvBZS&iugtcw2GR65G3xJMOG}^jd$Ox{muke)j#rJ7lbL z2G)JuN6x+zdm%o9#CMkJLGCz<=aqYITd_LskPmgWr2!^P4}ObH{JgX+2Bg zTdRL&p9?hu`w8njfi49aebivoxIR3ps1mp48&o!`Rhh|2$<5=kxx|d&XmLpV1ju ztKRR=+S8c2V}GrEdjj?j>`;U6Go04J+0Rurxo2YcRR0scaE=aq-vW5Aci;@SH|+M4 zngYCg)!)-1ZLjve_SQ~gV6Qq0JudcOWvsicvk-UUGr6AF7ix&D9pHa_C#H6SB(|6H zR8M3LzXO-&@;lR|b^?2Rz`v^7_u?~eK^}|)+3JkO9p7AFo!>ig4}1Ac)iiwX#JRnz z0Xj%>lU&Ej`_FQxx3#~u_WVlxui>9q-}^ut*iJ6-JFuVi#t*DZ-X4BVtaYXvu&3Nn z>m2G1B))lLd#In-1}gqVZoI8NWY<0KobgD0A+Cka+5N29V*ppZhq3J09Cz+>abEYm zU?1{m8gXJfR{wcj)}NFden<1?>t5BVx3B(;byxK>7|os9;^4Ih-&mE)^|LR? z4N(2)Wq*5qu%}$ZzPdl$XY8kU=D)5x@ZMX1Bo8<76Pv_Au4BFLGx!27+n2ZQ@`iok z{@^E^J>1Xw5%?@?zQWlDS}b%wMajBBf9?47n;cYedS#|2w^!CQNISu_0(_Ch@11L$kr^LWnI zHR7`zxtbOG^L6c;=IPV1wI8~gSsqwp&jj9&v3G8+;oIB2oIBt=SLHC?hqDXw>g;vi zoDy>Ztog&2)n8vOv3_=6-xfc!)jB>mlYD@_F|arEPtfqqIb#Rj+a2EhGuq;r=K=o# zZa{yMbDs;oaPEV%pYxg@z?#H%_(Nk{cLL{}1N)g8*T81LW1YpiI*WX+*#ge{KZ67v zUXAB+KbJiks9a*R1h3vb)Y;o5AG!Lw$Dwck7bkBFD(RNw`JF`t zYGaHvTCm0#X*{EihE1c@+XEwB+)J|z>=#hrCwuC)o9G*7KNiI#-t&91ZWp3^U2+)H0uS}l9XmEOLr z;*2womzhfIKy{%bAAuk9Qc0s`Mv8uTMXivE0@|I zTg~YArn!)NTHmXmv1)|h!;^f1JA7bW=4uA!x}ouVwZpXlSo?>(HN@5ptjq80+CK>J ztySafiLDsFddBLzpUFA=IdTB+p10qx@U7eTu(#j$o`1qyGl6I6ShdFTvH1gi$3JZv zb!vn&`ZLS?0F}d54t>PB!v^Q`aPE)b0&>$?W1la;9%@~a*yYSMgZH^{_Ki*N?9TST zKdDMp&T@QObJnv#TgyDYweCN2#pVKEByi~uaNc@N<5<}5 zr_KSVZP*0$d#8TF1J7j6Gfb@eUgcQM81bD>-dLLp-Prytea~L+2iD&DBh9Y9#*yoD zBIgX&A9ihQdU$mc_-;4sz}~@#bpS`VY_QeCeFE5Tf(Gn&w2w8p?$kaf`;eEn&w+Rln{(B95>6e~IU6=&^(!5I zz|GC-d&W<|{~pkOF5jse_0CcA2NwT5a9?o*>O=o&an2@RcR;S2*XE|j+GomP&6WQr zV$X|xg&*kj?Q4I}n`m7j7dlJRu@ysoBzHmYvVRlL0Iu#0t9B;+z?nFS8>qdop51f0 z+^cVz&)B--SkL*o#d{z7ik1L-q1HIT9p5|K_npKY-uQ$+uFhf_bNct+;!<9-*cwN+2t z@tu8-)n196q2u4wfPwD47S88VmRTop+j5C@*~`7;?J==+4t#S1{LS-vNBrky&upIn z>IX>uXgK^2d)j8OuYQ0VP%}2jS?`?Jr&ddiJz6YlucI6YeB+LGfHR!FP=Ch`^PG2L z-FN2t*)s4u+}`UzoNBRh7ZAh^OklprKcNFB(4JQ=a^>vpto8_Ct$R9$8g2DeL(Y2} z`ta2c&KkAm?KQ|9tru{@n&V^Hx4Gf$Kgic-reQgE8k5^=q3hXLC+FUKk3`qt)oWcd z*1fh|y1=^ZeWBJl)Ys>i_}%mB-($_~&zLi~sH4U;a(lPo?!Y-3I6o~vu?-x-0QPo| ziDqIi#@t=U-ay;pAZH)fr=_K;G#BC^e?j}U2e}c8Ppthq)>`)(#K)KQ z?|W)r?!g}NSMt_Oe3yF~3-i`D?2Oen@aulib$oRl?E2C<_hao5QFA~i zm*6wLz0RQaN^CV6K)J}p~yYds3?q$8VEtu<>J0h9GJ?eb2-lzTTr;&!G17d62Vqz&FA2^HnwEo%xl#>!t;@C%pO&)XrFY zl!lb&xxj}!niF;E%feNEG}o>6fK_L2_qYIibz*({Ya8~64eS6PAM4)iYwW(>mH7^4 zu74*N=sc^MAb-Pu1!)Xwf6lA^2=AJ)-uq5$t@DX9zWuyU|2u?+t`X#j25WS?e0PW^mv4 z^s$%UCoVN7e62e7wO_+J%kS7*?v6jPE@NjiwwLuS*K7fIC08?Gjn$dArf+d^#Esbf z{BB4zdp{&6^x(hH?Kj1#0 zxqwM5XZ;OL6Zm&*$GU1Sc~*~AYcKa`z?(RgO} zGN-+=mwVc)1J8R04Zr2~_Yc$rexNn)Sw7&qU!ePlwbq`VeW)c?e|lOq#tnFulX~X> zN#1w2?hhNp_U}NQGtA9){9zC3+~WwZ-oEy)9`wJ#?HYCaUJbs!!_{xTHAnT3FlpR z;%1Jx_GBLW`Om`g&VRvLH_*h!89QfUr+^>W2GnVDL&u8RkD9s1TfNVB#_H{NySiqK z&KUx~;cK1IWxq2y&wbqVYChm+tpB?hXP4V+i0c0KbE~Tyc2IYDsMwx+u01}%`!2qM zui!IqRUN)F{e&I4?#X#OTrK=Y&K1cYKf$_k) zP7nTw#U1gjQ|oi~J2}W%XM6zhSo;0mI=THG?0rMiutVNiuopP3{na|R*4Vu#xj&)_ zta~*0nu}QOY%Zw5x9+;d_Az(CTDQ-kb?>UdHz!hT{W|^Jqh`X^K3JcD+!VTh%->~k{jRP%pwI9BmeawHv`ghkStTpcUv$LR|@aF9ATrPV$ z-yXlS2l2@DvoFXezTaUD-@Sv_z0w%_-W+S!)ZE1GKVstx``wH&KymkAX$f>i(m7Z7gSkL6XbyslehdMz&--Eg@Eu8xVRtzBLjGpxbDwpuw;|<<@ zJG^!3J=eslv$uJ9=Y99Q)!}!z3$|fLE}rPEHsS5*a_9N$9_Nn>t1J3^Tt7p)tC7$pNc*pkmNlPr8O-;q*23!~T z-rL0f{$=f@FZT)tegp2~42ge$KRzu^oxRj}UhNDFkiWoL6WGhs>iG%#iTL=^YIX9~ z1aMCKOt`Z0_zir4Z(tb5wl8|?I&cr)D`&l8lUS`YdSBMq|1;?D*7!L(1J5&wtqrU? z_ljJ76Yd0efou48>=9gzbN##yeD|}Tn!^)4f&Jv|X+Qmnp&Guc$GX=}?Mw(_`y{@z znYT{@|4#k`oW1X0n0uhfR&@hg?a_u;JeQe-4WkCP9WhfSo+Ps)_L4Zu9H7$=@qf|shIo` zy*(yKa@MGMkKWIWz3t~3xn{X+aggtD4fI&GYHAi~dHPj7dE?4$_GBOUGdAF=2Q&I? zjD1#L@%KJ`)0mgD)*9DH{)n~Lf&2~lZcq4x53I{vyx`>G%}1=*`V;w2zJC(w|Qgxt4VMqKFcQ*wqt7^s1vdI8SA|2Z*T*s zKVm1=z4QZLt;=`g49s81wOBa3(4TWShy8n!B4CXpR;NL|J53N`GH?re9vpIJ)Up@=iR%ktr~0s zb?&8K8sY&?^l&+U&7A9wJWBvSBkI-u@WL~HUOkTF60Qe5$ank(>=AGWY{MqG3Gd%G zUh%EDZkk3M#A-_~PuziL8K9ZR`rq?qzyr0OrEhkiQRnB>p7lO$wox}?)g<{uXRPLv z8hyBiHRsCkj|FGdcRzbK;vl|ROOJ^4zMkPHy!XF9AMb47n@@OA=YzXjXAGP(u%(kne*zh6PkRsZ%G<8z`JkT@_66+!nOFORa210 z@5ZcsJ~{L9Z@znp@3@HV*Rf25o|@8PON-tW zS?28z=W$uva_i-uvBhxK>^;jTF73wuliEGM^+dh;Yw6v|58$d=Exf($`~BMYmG3~C z+HFOCn{VLnv+Vi3Z|%!Gd;VuUd)IllqSC_I&*dJwe&W|Yd(Djr}mBC^3@~0>jnR|2l{EOHUL)rKLI*(Vr)?Jw%`Jo zQ&W3xzJ1^3+j^s_J*}y|=wBLQSM|hI51ZtR>q z_*o6Js#hbQ*j35}jS=*l1{z*+w z^w<#iE@vs9em%Cc<@&pm`p$>9U-hJ)-~G2u>pc4YKJpPbPou|O3ogJJ-0$$T_P5@% zsh#lV)X6*ZAm?7@K4Ztzn(H&x{W|%&H*)TAgtOkdeXkRoJ!-}3W$eE0GvEThV~w57 zGq}v1@ISyc{J{U^Gg@Zb=W_=4QzyOvXS0v>gScbu={~NAHEUwkUVxf2)-&w+^JC3q z-Fx_ko#^&F%=#R_-O$Q6;MqT-*ZO* z<9@7Ov)noMwc&yXPI)L;XIpGc@w)m?B#2+AY)j6|y`#PWXH>@ji#XZ*h zRL_Lm0d3s_{j9yfUE$R4{m9fRU9SDSm$SM6)VSr<&iv)k5QEsKF(^HJ;JF`mgZzd)s%@9MP!vGc>TSnuYj1`u4nLzT&^ai5Bab zpYW|8kL_K^^&Z`c-K*{rPJP4rj5~f==1({^38$~Mznt%E5WBSQ)4&D(idAcT0M4uZ z@_=Lg@WLOt_EM|P9s}I))eo%uoFA)~zI;b>+H&u;Kenp(UMIPUMqL8;bRM6LH3?kq z<@<60`XQg-vF>HS*;jiv$2w!}sn)xj_-a1kA3!Ir8OY0DvF@p_)*8>?Eb1HF5$n6s zV%4inVzmL(e|lQ6`r3ndq7P!{zkw#;)CBhOK#To|)lc|=#+l4}Z#56R?~waUcs0!! z-&*@#zyZjqbH+2d@{OOD=8R_N-S2>AP*XFJOZ))U%R0D5JQ5$gqVL4Vt}h=4_=ex{6MNk3$W`00a+BNu4Zrk%+I}x2e*aj{GVsqY>pRDJ zeO5EqUhX-_TmK5|rDkFe#2p*CxXNv@xr)g-lS^&ps;&FMzo-K)?a??1^dDG!=aIM8 zz5npCeDAC@*e-%va7OEUZy)c`-?gkua`tuUo7;Pw;MG{)L~{R+6$kzV`*-NY-qGHc zKX2?cu$4m_`1UCt>(7b6+EdisVO{R)%srO=-Se`}V7(W#by$6<;f_?rx5r9UF4yK*KNH zWA)MxP|uf~nv}z@d9cRyU1~piH)3aSrUWzbLDgat)H7m_5y-I@bk-b!dG*%&ZtOkl zj$=Ky+K;`gA85N^f9JF=;Xh&bp6;z~V$bm6wCNIFZD7xj)zdwNI(s_T0jRZq&Gf)$ zxv!kdccUeU?RyD#65HPzd-&cB^0kQ7j$AqC)9>JfF7>g2u9MUH&w3Z^UtoQ=tho~V z4mI%ZW%YOVPEMp6awoV*Vr;_}^jY?pvG$z_fuC4$d90r1yF93w$kmz2`#XsJn_zWp zPv*j^8Ti%)@Xn0&?d@#NdjJ5OC(~qpwDOz%_7qTKgSeh!c?4&RBY)s{M$riG+{svy5-t;ho!Mow~@iM?FU< z`}SD-U-8x3GvVZ&O*@R4b!vZH?_2dIzyH0Ld#Y>CfHla~TuIH=rw4n;WA(=tPweSw z_0C+MGdb(6EsnhUqxCH38M$gat4p04`Gz&_n}4wn+y!mNmJS>6ettavuB|Qh**lHFjaYk8 z!?I6;(&egmp1`-xJsQ@U`uT|8@W&Ig0^b*Pp4aE%p01Irr@j0fjaazh*O|?$CvX1% zeFng}~PWA=?SMM!; z&4L||<-7OqpI84QymPp({sm3oYwMZD>Xq2a$#3Tk^#1)@C+D8tSKyzq3FPH68wr3- ztowME?m3AEkarICu2-ymPLCDWpTs@Zv)S*akw1f(W?pL-qV9RatC0}-mc8GUkjsecY8t?*2LN)@h@QJiW}Bi zZI88=pU2tmG>RDHh|};IiTEYz`2~m-VJyLKMU$7ame8()|IieBNrz> z;TpJ6=S;4?+gv^7W>4_;YsAiT0oDg`V%=Z={;_bOY{}^s&dldKUxVJsPpO6Mh6S_Gf1y?&RgIRpTC>*}AJ%4u4{uOYITp zdxnqLUE@7Ww9aNfW0z-iR&(|_;9GOWsttVS{)lhy0p2tZa?~|6XZ*l7KB0N@pUD#^ zymePnJ>%P?6r{*CyK_fT80p7fvP~#rIdVwdG zXw(Kdms+_F>{<7moSmAn@xIF)&bs~khNym*vF2)?oVC?A7nfLT{k}``-eq7rx*Kp- z??l~g^Cx+E^>04;J6zS`cXHMZ{Pw)g={Y>#`^VC~Vq2`4Jb!7?PoPfyAn&tDEl_Vh z!NhktOZ7(YGrD|N>v=s(-7R_RszPljAD-C@t)K6mvxm>jo?oyN z>*~QDu;-d7O# z&gz=6`c{m8gm<5TKk{4u>dU<9?dejZ?~FI_QT}Pw*+)Lm_^c23iSLXZTXUnyY-_;{ za@JXMz`AFn*17NG8u1xx{ngk-?gZRx#Nsdbp26PQ39lxxYJ|PbIY*#3SNmZDm^;BY ze3$<>8!n&uCwkq1e_!Zqk?28bu_K7Vz;EdF%^%R}%N@Yc9Q$|%mwh6(vl3t6)ywPK zFTuVhzEC6YoeX??{KNCQ%f{;P?(K5l8@@WvatHSQ3VgR(uDdwl@4XVOa|XG+k9+O! zq4Re%6MWS7OlY0yjIR1%@8l+&Gni{Z4$j&^>|EZ7{nXWpbq@EE3*xw3wV`#tg|?OC-1(_;GQkMn~zyf=smk< z3G$!tJE-@cc!B{+)0Xc%)f>Lf2!Eo+UasTj+wX%Mah)I6`)Tl1kFO@w$YJ$otohac z^~;+39oD{TJi~jn*c*25UGD|4GrfD>`aRIIeBSo{3TkgOmo1lpAXh(Y2XWs)>>Pu< z%XhKG_PbRsf5_2S-4(0WdV6Zmd-*K2J%Dopds(N(_)h(CtN){V7j`Cj=1lBQ#P+y8*4gVU zn+A2=v2z6)d%CCl*jMe4$J$qGza|GpZl5K5Yj$l{ORviV9wd1&;qM^uwWUYb&F8l} zb)Smi%(>3!J%jt0v+u;Z$F7m)zp={|Z{-16R-WIj;9VAc${j zK;qjkfxKrwZuQQ21nPu)e8&3ES87Krt1qfQ_bwmA^~_D-8*As&;_uk8bM_;iu`r13 zp+2!L_x>NBSL~hlD7wB&_VZigY!hqmfi8g0!t;8rj^D_4@^uG<`T@*mEc+4v0M2sc zJd1wA7jkNaJ?pG+9k`#Ly$c9pQG3UltFipO>PUX>9M=7;8TeOiwsrUTLC!wju{GBE ze7=Cb#qMDbb^d!#*G%F7=t% z{_mdgvFeNmwO6cr$W5&KxZd`Uwg0E5wO1s`%~xwbV`qL3I(p}`-sRbfBPTyWeRsgu zJNc*eUUXu6pVY&7*0E{jPQ>Oru+RQXlQ@YR)|$`O;a@hrvkiRrb*9fiZlZDaAD*cr zzRRPr?<>A%k+;ryVzc&Db0%laM`LgWwHG?)I6e_$eHQi|g1G0p`>*i!?O6M`k7ulN zVH2Hx^}#o9t#iANE3sistp3I>XY(Czu{6$OFLgdg`M-F^U|TL2U&z&SmS=gF2e^|p zx#?qZYHJoW7vSo_ADh;*2Y$i{&+2ks|5+w-?R6$r+ldEgKtF-|9-mjNX3EE^Jrdjh z3eOIqq1`nLzFSe4D)=>nzlQY^;-F|PwWFXi0x(F4eXye^~QmHyq2JUAeHXlInvdpq8ppmK6MAg9si8x zh$hUlysIPFzh9ln*yWxd&}OcFRZDKKt@|T)fpdTR@1L{4-_JL9AB{_1?qChRdw8B6 z>#k~cg|4mUiXBh ziDNySwMXnEJ^-I{{k*`om$hF*b0fA+_!;%t$X~Fp`oKQbH`o4ZBi21l{0`cN8`=Tt zos9L}_UWiB%E4v-qqaTe{bfpM*?S4^AUU1w;#Ou>I>J% ztDV>Y>Jn>Ta|f(-&KlqY(ErUd{Knlkz6BcKZ1%XoAAo&~J8+%g+_MYr;ytms>tpE# zzJ1iWw|BVTMUZp9LH_u>>f{=+y=sy7zU0r?GjYZ`n{^_|bs&Eu-|*F6HhqKl><4pL z_dh)2v+CVQ9CAOAJ7LwG!GKTT{#C0kmd3p%vH5|oHn4B+7rq+Tb?fzt{kPBS8G2q< z&xNLCu0V4F`p!RN^{zX^H`lPP4DJn0amkGM?l0$gI@UW&5Wqbu7t|1!&m8?}{d38H zw|5eE_4>g67jOd!Z(V)v$$4(~y_2*4x>;-eJ9*yPT$Y&8D5Qft>@GFPTJGw zaxjOV@b|}>$!9d-JZIe(Iaj@Rn|=m_I&9>6|LtiH>pQ-@8qab8RRiZ<&SqY}1ATEu z=e}C?=Iqh5Ea$0n!HuVN4$oK51Y73`;#|5Dx*!$@{NtuMVXbNCuJC?l+@t0omJjTJ z>p&a&?C&0RH@Wukvu|z?t95lW))=4h+so?r9Bj8koR#w~$a%iN`VPxGTbHBe3wD1`>~SErhc$osy!C}w+Y$>p&`)qf*J9>z8%<8KLxwA2O&u%YqB>xq7R;@EyYmNJO7BznNxkmss z_Hjinjo$(4tZ85dmSz6kGY)ma_ebsw=R20M;>zO>^3GyxzZ15jZ{|SRT=S_3Ag5-6 z6EzJxzU=PpgKAzFo8SnbB-g$G0AK-wm{-!0~`&=GtzhAtpz2tL) z`XKLJidy9CXY3qrJi12QuzQVXIzABtKhtOZSaaH2t$W$WS)IS**TUAF$NDZxbNJc| z82H|ky&IUspRt~=&iH`e7o2zb9_yYhK=MCq2SZC7z_@19dPc0cQd{?tt?E7F73}j) zH0s^ERxV@D35rK=uZ~SE`i4DHQ)h*@PXL#`dlr`v=suwFUfsLy4~_L|YQbAyF}80x zKXXTN&b!~eb9pa8&YBMV-On14#Kmtj4C0Y%e|_ihv)VV_oM^;na$*4E{qEc=6AGO1 zUHzGPv)uK3bgg4uvGzY;)p%#d_Os8#A6WgsukU?*^!2>qe0KWktak=EVNHj(|BZO! zYu(>m!%u7n2Yh=p;QrQ`v)@HMu{aW&SN{rs1ZQB+4o**NKYP1srdYM!r}+dE&VAia zt^3(qet@cjGY*jWT4#9=)DPesq6^M%f9riGto7^xtT%QKaRJt-wa&Pqv98V!cg0sb z;nlQQdbtkQHf_M`*BtQ1&f%UNzqHumA8W4<ljhJ`?`P!>_8IIUR~mGe-N#6V9Cc^+684Va>VcXE5+TZQKc~PEEjjZ^tKMSFCgQ5BxivdtaWGzG2;OlDDV(cVOQDXE^svU@vE~pY|uv;O%FRBd|wDcf%j# z%p2ET!nrcG_{TNJub!9nc{Fm`SAF#M_V1s2ZfAkN0ejTS_3kHDZ4X%N5jaZ_JBKjm z9HGw{%uW2?ylficpl)inTqp0+dS_J|=s^R%7rw8FZ{BnFE$1F5xT}CybHWBT;N08q z!vmaiIrFZy&OY9Sdk%axYVD=A!MW~e+)MrhG9a)A%G&rAJ%EB$V8`4BV&g`LSvP>v>()ninTH_i*+p;M~jFhIKjL z#T-_?2Pyp82kY|u=0>c(&G<)<->e$@v?nycdh=&E=e~f09I?71xC8g{cRAO{mA9_N z>TTXS-vxU)gSvw?_;Qn6;H$e4o4;cNd&PEg{+;GW@?SxR7uLAc8^0&lV*PA%X7dAN zwz{v0U;p0OSS^r?t*+tU@ViKOpgj@$^WR<_cwT!N5B%zd|L&Q*_e{?BNR5!Y6RTJ2 ze)h4iztbPEQx2_p@8ApYp8c-!Sv7K(r)BNuDw}I&wXPZP=mOl-8z;bemx*;>pP&5B z4S3hSmm1Hh=7e>|z_+iQzMORdIygVCyc)j)2O9e~at;5i9$4o-6JCvab60Q#_O6A! zY_WBo(V0TTcDGphj`oxM)0)LzXJVKA+|T$1PUNibz&V|x%Y(%K&CBXJ@YQy#J!)_5 z#!dK1+=Z@Y%1yIgdT`587P)tg9H_wcq<4%@KG;wGr!_#`aG*-?IttyF50H zxw=#0LC%%2@}9fnPptD)FXC5xH4S(+_p5gbe!>|~d^u-2ZT^H0Aa8GVgPh-^eaqRi zdJ)Sta$z3+6Lw%b_PB8qjXHU??XhN)x6bEOKkvw^ah5C4KVT;|f!`6<_$(W7-3ite zvGg}IEm!Wa>5R>HFo1VdEBQ0vw>yyQxghUsVT}Ju%@KSB#Zfci%>{7g1aihM_qV16 zz}mZk9_w5ky*c;a&pU|KYTd{C^bY*~8^mYh2fiRq5%l=6)!q7>g199XUwg!sCf43^ z_GxN2jpwu0eJA!pJkb5k%ewO<{vCdOKG@58pIb+xtv*=mZ}MmYzw%Fu{|oAkKah8s z5Bw8wuM4^f=XooK@7;7_-{CrATRQuw9U$=ooY0kK_lf0>;QUzUYvg*Y`K@o{>>(OZ zD>{Av_faF~T!XmY{l{iW-|=_enn`R;q7|OeJ5qDR`q`ebp1Jp_@ALH@)L*fum#rsp zC+B**Q|@-4Y3jhI&2P!ShqaD4V(sf3-oXHqytBVUqqfcgZyynI^v+m6U#{e=?Zk4z zbwrzq6YD*>5&NDD;_AU3AFTru&KY`Q@hiqxyZbci<6V}; zt%ddkMi61$BXd0yzXwv@9`HwE>-{|~;7_}+4)pw{^LJg(rQZI{+>anu-EGr*){nqm zfkv(StF?CE%OCNr-RB#`4ZdbpN3Jw*)_Y&RE6$+*8Fcs-%c}LBss`V_=B>5o#5#kT zfgiEHH}ogI>rBphOS{=jwD?z`-(zWJu-2QGbJqRL<|q6h-@q9pabTyMe!}5*{F;M!zd!5UtCO$q zAT-tvtb1wI${oPZ`jc}V;s4Fc@_hy~))_k18e{FqR?GNAyzgu8o&df9bqQ)l_V%p) zJarxMWsYyti3-2 z>-}dJV`n{q{jO>s;y^z^!*8+f<@#8h-`{umgttdv$A;$0HG{pabr$>7eUkGo>~#aK z9_vnn_=qlv<*adzfq%yL^D&6`^VqL6vlDGc>djl@xj<0Tj z0DZHbyPgNJ`?f82gin02?-}Od)eK-?VUMr!tUGqb>Tm794{T{4YcKgPo8PdXsA=fD zdwaR8+3$I;ow4$}zGg?~xt+Ifep#MD=^x8h59+;_v=4Rw&sO&Y7f-F8=Lb0RYTd)y zj`jOQt`YzKdA(l;c<;@5?bWdM_gwlD)N_gTUSH97I9HF=_f1C}z&>>zWAq8+C-zFt z`bqp&2e0ow&iH}&1QJaLvE{65!~^TQDsP_wTqkM+7`v}$ui3HAP_MC1#^5sW3R(zsb1gF=rY&?>+@Et#$NK%9B|#>j^yk)z@Bdbw?Tg5pU`?HYl8TS z^&ai-9JkF^m*iiuXVCB?*FN@gna}(-`-xh6G?3t^CBN5SpVrKsIyC_XzI{HEQy0(V za@~=4`w@*j)L((?6SjIhts3iH5lidmG;;NwaN>ftWB>5HzNfqcv43X`a{Kq|$TioI ze88)7hKe7HYbj*E)VQwf9G<4 zHGbyoXKmGE6P(a`hb;lipMMdHGuQD?)Y{Kp?rB~>fb-T_u+;-Q;ajYKJ}izJX9{ou zoqPc2@*G!StUs~#*Vp!^b*B?wG?DO|*4YNJb#-RubUCM}*#axbU7oSIaHRGXjkEcz z>}juocE=s=y*X-c_-d>_V6Ce?f83rmbphYtC%!Xjjc;o8$zOoh*;DT6n|=@rd;9xj z%hgjZfpz|~hVd26nK~4{}G4_ycPnXWrLk<3>=UE>~t6bDrYicE8Sw&i*H? zu^QLe`q=*4wg-I5Zs*MP{vP3b$g+S%@H6XP4c}Q>tUlhm+9dY# z;omKEe1CSyxrcQ}5V>ZxpLg%G4m1;}wMOj~eBAucSogHX*#B%754A3?eVm7YU+I@*1G>^bTy0mjl1Lj`guDm)_#?v?hCjB z&s9GQ=<$kgzN4FJe_UrYU$e`{`Z;%od;dTq;GD^Otr~0#e7L|*{DFVOKR(vH7wkbF z-ntWTZ}Z0T=5w=Wd~^2iaQm}R<4gf6M-R2~@1I)DrTqZZxkt-&7w%!5eJ|vl!JKhm z?eP&<*FeGztThR4=sVV$hV5t@7{u~gXHl!2#DT9~&bU9Vp8D=@zMcX0K&Gmlodn7q)58$?elG@w%^A2L;lUmmwiF<5kc!Eek zY_Ga|@<%X27T=(govHOjMU(a;+w|M95 zsb1__x!sqq{=OI6DvsGoquti7HQZ}YN+vN}Vy&&*rm6ak zH)mg$8gskdIpclpxgId!Ely%rEUtPcG1xUktN(X%m>ZkM{QkL<-)prITVMQ&H~YQD z+JNuhq4MxG7rc2nmwTv{6MJmlSlhL%%BdIhGjC1|&_HZ<_tm;~y}Cc=)OHXX)QFvH ztcuz5t=1X7Lqor#G{n}ssy2T5G~dTYPL9!oG9*^mD!SWv_kC(#57pMBjrjzK`+u z9`bvxZ+^v%MeG{MAAZtvRf@2m%U=jMBC;3su*ey`;g2Ud;tE8l%u zzYDc9TP@>(KVsE)tlzJGpI*Qu4)P71@aJ47_UA&aSU>mVTtC6rcQ|)^#J6we`WbP4 zAi%$BS&^$(?TyxbKVj9l&z^Ud2zX$BXYOcH?q$v0(7eM>_>QmE`U8G|3z)=5^;m!Y zc%B~X{GP|3Bi2|S@D2atrqTB~$a@ZFI)X-AIcx%d2e<;yegf*{y^~D#!j_)rGKpPk z?c>j~j+Ixd?iFqZJeoo5o%MKpWrS}@Wg8N&xz-@R^DFLRu1d@ z175VvS07-)eR!}9ekLFe(1Cjo&|*6yx`A)pHs3w|_Ib?|@XpZ9gLm+Ik3n3$$lsq< ze!|UM`}!P(I%f~;4V&aUeghNuJF4fofHV04r$!rC=keX^#AD-+#1p^4b!<@gZ=cum z_1`x&IQ8yI^0*GGRMb{5a#{_Y_LzI%B8UG4!L z^dt^&0awsB-)FObZ}?8S|8cYSv%h-=zJ2sv?mc2zb-p+IOgpdEGY58}u|~gY;TyiS z&NJ}qbA#_XKEqp^1lS%}mbKPoY`+hqubfuhpUAVUmEX_S^ntY(a{hNF-bca(I1}Hu z`c7P6FW(vW^zVRuZu&`&fo5#>s0LpfSaHDr)iXTvc)l}evF^dz{j3N14c1L^x5w&b z&ra;~)DCrW^i-GlzAG;KL@c=X$n~9|zJM9AYCS{EOdosbmxlb_*So2^gKOk{Pwe}T zFWWu|zb$LJpl@KEWo&U`_s`43H}9S&G@iR}+`#Jl9=(^(b!P+XDm~WtNItL)T<~k* z8{fA0vl`Ih-app4^yR81*0~2<;#=oEo4$xMBe~@Q=|BTJsqvhhA!CGrNkoOtMxvC$Uh;?5xA&Ju% zXsbp|;u~9^+!N^Vo_PZ2v8H(Q_*b~P58@mC>1k;Ov3{zkRuLw@5ghg z|BCMnbxv}ZE!W9eAHcnRKL*^H*n0bWA6H=Q=Z&j(_fPaDcCQ4d2bu)%{xhnx40wCF z^zGfS>U?k9+cO4SgFE6oXW)O)#}06OUS~A7W1m5d2vB=r?Nf7LJx7oBrct-ga3Xiv z@)N#;12vvSYdoXXx}tMl>pkBq`2f!2_rE>8s~gzwxyiFK%G4d0FaCj*$EqLr>cjz! zGq^@>?=ko4-p-`SG@MFP_&-a?a4<8t_~L&eiQmg}EcJw>_esA^;U~OVOuW`L&kF|G?w!X9A>vKWu-JQ`I zJ5$XGXYYy5Rdu|rl^de$KJb736d3Y?*#O>)Ae&inx+u-1LmdG-&$cVM7%dH42e0Vkhp9{W_S zd&A$+B-{mi1pA6yI-hH;SVpb7SDiNa8_PWRuyN@-W zfzQWx(S48j+5lI$nuogmUK`6HJdfgre=bZdh__+l{#nW#0PS9KJ*QZYvy)F?2WwdYm4vhAKyOi z-_dyQ7yJQ#!cVxwm$z2{b@FP4F*&*NH@;?u_pG0Qdpf88ymEY5_jk7ruIjK!t$qA= zeRZ!}T<;iXy4uGT>mFXv-RM!@A8^O-`X#YJ4YTwI{gRgRo|Vv(FX%uCexl&UlBl zjd)-yhqX_~4tR5F%pIwp_L zd)K-?qP2(fpF!lB!}=bpKCnq_kFFm7ihU(+f>_@^6aPd`d&innW38GI>ptxn; z$Tg!iRUhlKY3_)wp*udPXJ31`GWY{dUie5F&IYi8d~_uuttJ{LMx z{z5(<>+BtW$P;@X&QU8@pAkf8-o327&3j)V59Vf%X!_>6hc$^6fBmwZk-GhkN1)kP zY%^i)=h^PSvq!ET_D^!gF8g+1p9?wr4q~5|I{UcP)R|)Wf%O@Ar&U8N7ujm&i9M0? zK6%4LeI7dtVhn?6f_6|nB^s^0>^q6oB+&O~!(q!cY$G1z zZdi2_TQ$rxiIcct)dc>gm(^e9lvAMsxx*kaR&Fz|A5VC6n)!^sP6Sh?z(Bc z6YJa~zzKi<^sQR@Xa{Zermk?%HvIhxf<1$G&)@X+vep?oxx^PC$G+;_X8?O9)|$E-ax;j5xsE+Q z&j0#9{l)+LfBp3Tv|s--{z~AVwe!FJPu}t0*8KwApRNC_{TJ=;{=5I`AO4&F{=fbg zfB$#?_;>%}xBv2Yzx~}`{mb9}tLMLJ|K)$|a&Bk%cmMAH{J+__iT}SD{_3C3@K?Y6 z?Qj0o^MC#yH^aLB&V79GPC$F%|4$k4@B4oN{EQ!;);;^Y{Q2j=oX9ehm`;0^B2?R^0__?;~j6!M=WGe246P!k_G2I;8vBXUJIR{fPD*dwCx9zHc2n zfU$f6KTGz?E%)`YzHk2WWzApTjcxr7idjF;eaAs9YOQz97V8Z96=P$=VY3rW0PDx5 zS9986FK1IPciqsjpMf^FI(w-az1bO z9Kc$AF+odw!Mo@^`+j*g^Y;29_#cSRz%yNl1D&6pi6((|-?x$XS^0ea4{#zLN}qISNj8bbM9eH0MB;;?&r)t6S)Jhw=g#EepfAh_W6(a-kI-j>CiObJ$Wu? z@*U`K1NfXgM+fewZ=Jq-&&^)3YQH?tf-4AOF@QZUpn;o~6_53t2i7yHJpeT)(1AU_ zzY{+%K3DJSPEM`sAK-i@p5X&I@BR1q$4xV1_0tBp!n;T1(f$!QgVuR`F6OHay=M(@ z1MZ>j1JEXU&vym~;4}CYn!g4A2!28{5*}U0`YerouU^S{51l&uM=%q7!cFiMyd(Dc z+{t(H9Zu`)34THD2jJ(!em>g^aF0ZHLF1aSX7p_C8M$KTJrSE1p5goDECYDwCvr6> z)@L(-XFmdS*8UOn&0aU(ocHBSO)kKn)_D)`6W$s2XN{`P^a<_@u-^SH!1Fis4SNRG zd&lw}-d^rI0vWzFXAs~s=->bvc{Rd%d#dxlb*yj~Eo7``JcV|va;6C~- z06ix7M10@iIXuG^sMR{7XE*|D)${<=zJILmBj@vZxVJO~o5?9O!q35@UL|3A{+ z9@N$>JJ*}w0(YQb1=pZJ!Lq5KgMzy(wfeT8O^;nP7lAuifxyCau!Q@9d{MA4!pLxg zi4jd<6zss5!kFvn#2k#lJFKk6KG6wJSeZX?QsD{9nV)COIrrWzUvAJE^n1qNJAUT; zzO~Li_1B)y_w5U?-+v~R)EQ&WzFL>FXJ9w#ocVH{?L_P>DZLJ-#bCq`zWJZQ z`}%o4|92}d-}g0%`+oRLt!>2XbA)ZJ_pxR7@eJ-4aJ|EyQ2#lo`(Ab5pwI9Q^1^!$ zV4df?fxG&-m7KA^0_Nmq%x5EC-USTcXU_Zd-t6D$?V02aY8u*IeEwI^;B2{)^E}?4 z+>Sng_vmNM`_T7Un``*Oe(^A8`OKW*PEX(roNYdVclX2d-1UNg1D{c!#N0i=kHHz} zKN4>-KQGSxPXCSA%U!%@`w3?qG3`6{_s{e5{cq{#bL?OM?`Qz;BJrQVr=^4XLC^N( zym$L5`~7dv^K9njUditsce%jl^ayU?{tRc~fFH)d*YD^3`PuQg$#JiQ(?GpniRRu8 z{9nP7>)l+acg`TLo#>Mq_Hl`usiU~?md9~j@ED3zkWY|U;Q`eGrqAC_@D6o z&hyTUi#zxLoFo2%oHL$)&m@=jJq+x6-yi0ClFWO1CqBn_a6C^VcY*f^J|}Md55%AF zZ@}4}^~m}AXlPTDU|?Gt*bQ{x-Po({JSR_%y*!V1asr;k`j6lYl3MR>pvAFtkaPZh z@yzXet@nwzv(C@$<^G&!-wsaT3L3EX{>;vxVS85J*S`mM^77St-oUOqBLA1*pr&u+ zwCx+fIcIPK@;u91@&4S?+^6UHt{iAl&+Pro+uvSS;-1s*IA<7B{~X`n-!8!20@V5O zeQv4$?JKN1=ow&u-Rr*M?%Sxb&kZy=h;^U6f4w7X>zzV-Kl*Y$hvOd3{SCICPh&yO zylD8I(a%ugTmJ|;J)Hj${Ey^(r|WF^Rfj&D^E~(SE(Z43pp!cTdG+4pyfb^pg1IXg z;O(85AM|b1{Fa{X*PrK3cXCf4@bBON|2z3#0q;ZZM`G#^53|2}d_(LLa7XJqxB+K1 zkTH8*p67Y|o^n@t-ix2d8@S-h72buP89%QFet?nl+@3k%|C-zl{14zvt+P738~O^2 zeg8ar$$7$d*M@!2!#)9LP}9H(+xJP|9^Um8sFPR+0o+p@*guf-_xUe?egj?b9&WkA zv5)uax!ucIW7$2)VQ0+QmuGn36WCmHo;l$woWz&6revOfHmR##Q*gekJBPYlB{dw-@SsH$)&e{OpseAc;_*wXKa0Ul8-rsMq zect9;P6Ty#`V6%F8o2@2yK#pd|A_7P?g7u8U^d(GiS~j-v5W^`JHvcx6b>%VxPg*_Oah5xPE^8{&$~7 ztezwGf$e_At-S;9$j^#3Z{_FIC$Ru>N03F!9XWpnIpN#8;tTIo_BGOPdV7P=lL$l_nrF;1i0ec{|3i*)X%i<#wX(L?=uyi z&-cLZF892FzXkFeFel%>`tm>5W_|$oa;FYHfQls=+cW0kxx-HueuaK|p6|c*ci8s- z2;Y4Z_zv6S0z9X-dfy3?Yi1*M22Wrx%R9Vd%l!y^1|rCv@DH@;*!p+y`TKbnKHGQr zH*f_5$kYEVxnBa$alt=>1_C7g21smsx}#_H9rYd$avSY==AC=Nz60;np4Q&LzI#8L z#+_sR&x4vP{t4WG_hg;?M||J23-CP*@E2fED}Mm>fqzDOt_I{Q_CM=W z)oee%X4SXC>lBA^1j1=2i!G) zXY$X`zXX29ehv1W)V(A;VErA~`vXt{{Rko^+|yX?gHCV-*7ywe&I7-Lq|QADa4+kG zHOdk5PSlOOTzj5DCnoQ*U|oPmV9pwIigtpAFHif(x${@};tJ${|2*&JFW=Aa+d<6p zn%~?(z1+tYckT_`pXYa<&(|5wJC+W0&iCgVcz$jh@L4>8w~~REbH`4P6R>|hOHya- zkI!?~=k$2Qet9RUJ-{7~dt2A>Kce3ObKj%wnb}JA^Y**IbN>c3@|R~gz-QyVxaSqf ze|$jmzSU=dn0$LD7}#MRxW{?HckK({&c?gkocRm>3HY7#=jMTT5OCyLW4-U_FM+O`}*E`Za=#|1K-!%!`$l;ozw&{ z_5(TJRpqpF`f!}hT z{k-D#kY~NvKGutP-%sEGzoM`A{&V2>r*n*DOjz$3eMjXuKY+cxhcj@8Ut9A$&tZK7 z&e_lT3vj*ug&+rkRpHsE>6u>(hZ_hwTpbx#KJPM@;QI_y=FU?l1y&$&MXelK`Wes7qM#UDW}JU_Q;=UjpBt^551eDlDWf%oE_ zJp!M%XLtm6ko0k9cliE!-oc*#3Ev$CoDSsnY*Wzg<=y=~_yb_?J8+hFSFa60pa1VJ(Y4R=gCijegGZu57M{Ij+ zxSzA`!1?9}2+*|2e|($7ve&w%gK+xP!F>IV1#e5SrbPr&#iXu$mrdVjO* zE4p&@x<1cm;ydH}@&UNd86-7d!10}T{=Wh4BiH&1u$Osf`n;SU)VY&)XkMP^#2av? zyAEt?n!xvtwq(3M)Al8N>Q6wP+VedvXn*xQ_xkVO&+i}aNq7$VS8#xLz}*h|2S{+i z7k3Z~?(R7gcrS0o9&&{7Dq@<(U%~px=StS=OJ>_B(?UaP}kcZoLcR?rqHch`EPz-BIyu&U9aU z{uurCPJW61&%j4=J{RZL8PWPr;Ql;+|E=#{{0q1N=j=QBEj_%CufQYmcffPH=M}i? zhoz6ie`Y!Kh}W6Uy8~mDmxEY)q3^!W%z0k#;Rp!ud~OZ9KHE7zYrY4+0-nddZRtP@ z`(1#0gc^B!xAt>)^UfIf2YA{s><)ZR)e9}(eNVvm^xyUU{{``<=Xp==9rSQFKL-cC zbpzjCkKhO3J$?^7O9%3O|Fz9MfdGSEp81F8d8Ykcr%rO`#q%BW9rTWF#H>HS$LIO1 zePjL;Zyr;-{N6z%V z+{1Hrpe@f`Zwuc2C*b`)5_v4ByZ0H zG-BtK_nx(lKR(ZS-=PPwUGL8JeFX07JM7&CHOB2>O#<`Q1xVmsWik_ShB^Bsc3=J- zU%q%Nx8;wV{1$WO*Uz)>743X`4s!0`U0%TxG5el?&&i#iz@KCL98TB`eW3Nb>LatTp#n*lW-p4mf`#o~TZV&T2!oIHe9 zIG^CSn>#!$d_mlq_C0|gz;E$=#&xGT_Zu;F#}Dk6JfCInubwITOr1TPbpYpF!9k67 zfV+5}^X&C#9PK=LeS^4Xu}+S&pPuJkCY(mhJM&q3A0P3JIm0{jz46?h*ZQlu=h^3q zwzu2^v;||uGtP{kQU5uxU)={WIiAlu>p=b)_{`sm@q~YQp6~a89l-hv=)m5-(;Wo9 zIs3S`@n3I_|^paK)aVadv89gw>xw=b!K?ZcZPKXe7AD; z@OgUQa=t_V2;W)0i`LwLcN1_Qwby;9eU4|~T_*Xnc_8lze9(U;2ST1cXdvJ>f$a{? zmaA<~&m1}L-2V2m=WQX-C*b*o`+Sd|fe*Y7pVtYzBmM~P;A&ss8SV8I_y$~_=ksdR zzQeu(YtO*F4rY`*%s%$pJx?p*%=z6?Hom>j3m5d`GbM2D*ord_uzf0O$3EM0+3Q zv#&evd$+ccv!Bn{?>O)I4D9tC+B$o>Q=;qLf1daCHCnNs^7qvHp4w;c+TeJ$uRwtP z-SA5|#+;w=`tDpG{CDsFum9)%KR0g|{QLgh`~UyH|Nk$3`lmm={1gBE{}Jx+ul|$& z_5b~S{{8=-UH4xJDgXX|tNH)%Z@>O8e!uyDa`{jH?2rEm|NVh~+JFDxXO};@ z{PZV0gWvnVf8b{^llMySZUp)xG3^EhFjns}_K)6A`&e-~N>Ece<~w%e{JmVBV*Tc} zobRTeFVO{g-&u1B&MRh3fSW#X`3G9Q`pNjWGuN<{$XT;LE2U$`&hM#PH6IrLFh`BN zflmBAzj`yP5ew_k!r65v)F{?BV4b-UvyXKNF4odV+gQ&z>x=iZ?<0A8?D&8qH&?77 z-p{~S>?c2h8X(Ua(Xrctef{nj`qUR6`v3tFm{aSwn2_gr&2_MINxb0?w02@2%Xd!4 zJ~3zCxg295up7Fg4`96R7dx>P@8}!*tb1)SXYJ=;AMFF(;5BssWF)`D>Ir8*ALH7^ z#kTJBrnUzu5Wu*$QXF*Ts>eRG?H%*Gn^5h3r>>hLp6G$rt{s>?h-sU>61APQJbSzYkdgTJ& zbDFc4exl_U4^=k0#qaLf@vZaj16$wsWN5Qn#moWkqxxa{F4X?StVtl(UjB^rBKXG5 zTVo&hJ63$4oxR_k#NW2PnuuAy{dC{LB7+&TG4?sD~}HRj`ohsn=2T1e?dPv z+WT`-b;L`|0dq&r`!TO=?Umm3b1J$%Z)ks>@XrG$boIB_!^8+;)YjS2)rZ;g>?`DW zSH){+@@IYYmf{C7V2#>yHf-zF!o9@>>=_ww$khyb26pjLYV&d%I`Q3gfWTKOM?d{K zBig!vqx9gHoS(;S+v94UoV%3PPk`9d^NpNmb!TIVUpcfn`&C>G4q^e$^in4=cT$d+ zd1o8*xwzxj@_jc4JjK2(XAf)5yOT8yUsP=Ij8_fu+R^yK`WciuYn^NVzBor>3%DQ2 z-GG1o^RE$nhwaah2i%ZPeN)3tbbAlrsLOxgnS&r#^K(Aafp*S8T%LWd@EUgH-pg=Op(` z|HGIP6Nw&)y(Dtpll*WWv{H33cQSSW^&O7)*sz_gKd`-<>K(*EEZxantwGDzKCn+{ z-`B(rFyhs=Uf+8vjlFA{JkWB+ocjnmzCDH90128p=IoB%^?@x%C?n@?iEcpOebwHJ zbxP&e4DY)71alwezKN|E??C?U<@b{{{+arY_B-n!@6Lg?###C|V0~LK@8?6=_i~ZL zj*2rfFwoY?cc-dB%QN4J2YM{Kp$}s2Wv#v#Ik}r(eTdsfzJG3da{e!7%-TD1!a6v@ z-y1y3AQth;>%`>MER^~}ZKA7R62rHLJp0^%zwdN7jhGlQwRPTG)vj19Ew6T#gWb^j zCD`(M;w^jVn^&(`_0Aa&V)iUSTkG5mBHGj*+K6`TMBeih2UBe>h(*qy`|RWM{PXwZ zvAws?usd9BdG=~(^WJs!Cf9e4^qv{<`dPE)5#E8;R?OS`E8^b+;{nvw*WJN>FSOln zKhxp`9Qf{%=;9I!ZNzuxh&e~U&VfIG^(AQM4fFvN_u96;yWfD%LG%ZZ>*r;31SGT= z;7MLgPSwC~`1U!-C${gev5wz~5A4wAoR9GBk+aX8nD0=;g#3LUYNhAA|Mma}Ti-kS z!MgYBdn(p`#Fd(zbB;Ckx8KbgeC^%KerMthT@th28s*5z9f7m#+7VlAy!xTFozc(% z)c@E{V~Ba{6m4Y$c4D~a+Ew2jjQNa?zy@dRj;>woqy6oK?@7;fE>iVe-sBrO`|g?( zz4Rl9DM^hz(Q-=AH7gen|LP9#4q~@ucWixSKZlHYHvhb)U43)q**UgNuYn!$yE*g5 zUbK0j7(Y@AiOH#*um>=v4($X>>Q1y{)@XL@Et%8GKx4}fbloj`K5<@i&MGdp`ML*e zYkX!qhgaX_BhbXpKpkKtwqpEYnm&`gqqDZ%lb-m2R`gpepLkn-q6Y|IPj@}A{T_Bs z#}7g5YkjnFKU>Dl-;Hzbq0c#+8h^tbG#g&Tvso91W@WOWA>NlE+tFr z_aGv2+VYIKr{5he%*c z;E|fm>G<-D7kBxI_WsQUwrJS;3F^MkA#(P<(05?#%h5hN3%z@H;`(0^uRe=YJF=_y zVLTR+yk`_0dk|OGGvJxoYro6RC?74yK8ok)F=O^^BG>O7^svska?E)S@3^?w_H1a+ zQnjcgHh_KyZ*8JeV&<%ON70sTZ~gn~-;J|}TyyTMR^&N5V{+7D!0Bkk`UHwGds*i_ z*|Q}06?m7PvE{l?xZb}aw*y69<~-XGv!^wc#}Dj4S1-7JM)eQl*d^FQTVQ*y1KojV zDhYlC)>SXu#2>~$7o5eK(%4NK-0AVSP%&~r+}O$f@;IM$_H(a_Ym-ZXt&Ch8bk$I& zRQ_Sc>Kw%0?%@n!tZY;Z{Dq2XCuyL_jhK3ZPUG?_ESFG!yHTSRgKjywKT+UWI zz`5+RuO4;p@ZH}%?&wEweVF~8jI%?Kv)=%U-0B&6OJdz2?vXQJb;~AP_wbIMEp=w^ z+LIaEA35hGx+IoHnR~etYLc7?V9&ZM&bE8hb0L0are{$i`2~2^I$PjB05#}ayHf&v z^a1>{>fZoAhjo5pBVO&3J&8Fhh>J8&Ox%de5A2;A*or&4Z|%hy+^upeZhiS^?WTs; zV)nwCpD}!M_9|cpu%BFGz9+`C>(0pamAerWgZu$Ji*@p<5B7^-zx2i4{ms|i=K2mr zXApFr8W}Nr8b2Ec{vK#u?ZiAKV!co3mU9Pd8aTiJ@_(S#yUUzC)N%r-i~45Esttz+PAC{@_dIQ0HH}80MHeV)l@KBi_*!Q|rSiFhhNNo}aHI@615g z=XUX|b>0b{yFJNSJa;+B|LDuPN5g+4*N7e1_OsUo$TMH(U2*3d>pT1+)6sW zKO3d~`#Hn2DVeWx26hwVE)VR%nJb3B?R~cqb4KN{J9Wlx_Qm#YL%WlwR36W=AC1` zKkxPBTl?92-Zf%EG1hY3k$sSQrhy^vVfNp=>I=`_)|H?mrm3#;s@XxwSbO3}5}>d3 zow(BR&kM#Aw3x8Rz;|CcyVlr6e@o1Mkpv!AkeK@`#*di#Y7E~V>OCWuZR`Mk7W{iY z`wp%5&!nB4oG0>wn7hd}9}5jV^7XGJIwSDSn^Palex$bIq|6z2M*VY0*w#5Cu@!SY zFdpV}^LyF%{`2>f8`xXi>nFrV5CQf|^nPc4Z4Qpu?@uRxEYvfkrh39T%X^rgvD$-k zsdrEJ+`Suiqfbdrp0>SeU3+Iht~`2nn{yCCkl_AKX@khfP!zW$!B zKXxy<7AuY!=5`#tISt!)K-uRRF>U$sjAx>O=h}N(TR`pIFz4Cg^$ga#OZ8Z1ALQ*V zs+PDsXUKD>`aK-G>gBCk^C$AL^cj0#+iRaiz5BZJ=>ea-wGHfEc=Mi#wODaX<%L6S z<$kg5I%kb};aqz=Z=mfTpz3ke13!%cdmZQiFEzxPK3wwwUh3(quZ$orR!{2{V^_|p zWR{xs*m9j!?_}UpbD-5PCtA?0)>-Kx+e4p>Gd6X|_@wCk~pxxEk%D2zJ4r)A;eLRb{xxV7= zW^IDa*H@05v5vMzj&}7#*L!PE;)>_JgLlAL@@uX>IDWb61Z#(9Hm%{Ui1Ao`JtNctf_7m{W_)Q-K z=)|n+F>Nx{!@pMD$tS4uZyl|A+Si`@%%^4R->?Vy19T8ef0V;{M$Wpj$$8fYu?D;! zXLkJnh&}X;AKdMVOj(LmiZwfgRM#Oy=1W~hzr9zE9?*`pJ0V#%34eunF=a@7Y(>_*(@HT1#DeRFnu zmj92=wmJ38sNH_eq(+G#4r&H!=+kq!M-o7)E|-s1^o3K&U_1U*uEc$u6_?Q)|MQB0aBf_m%EowT?aLXyzg&E2XMxI z4#oy(-~)WcnEg7~eXXfjEIzKg8S|WhwucBXfHQioJ`Eve+`M}!^(^6yTzsCNdAW*l zec^fb-hLLFzVV##hOR#Ju)g-gw8nQ){{rIu{45VR2T+2z=g8Edh92evzw6UOG434u zs1b!%46?yGCApHn^LBH$VeDX1%sDf?vdzzyL>1 zZm2dW`jVn3(4s^xTh46eldA+o$F44t#7WOl{!n#T%z6eQ9iaDdtKK|b4tYQC+Dlh^<1Jv z9TWRG$UT7jj+p&BTGSlQX1>lx{b}`h*_B+t?Pz;e4}AN1FY*ul1c-ey=b!z|+vkNt z&i6>ZxxIV9m*))o@7{9k)v39FIs?A^1GJcZtUEtU-};@?j^$e;-(1bbR`lzxnBSOj z1~<@&*G}#Q>|;#b(8@=6nHcfP#jCt=_piUMOuha98PL%OaGsx;E6}g|Lf2VH+t=Na zn04~+;CnFeeCescRsYwzdRqFZ z{OS*{e0H^G30k}Sm^H?oEjL8Yv#A??2TD(V0eesV>%;67YbWciQM*It>}Sm|hOT<* zEXBvSr~PVY zZ22YVs>zwVS-5mL-`kRY_-x2-V;_I=+)zShdyyU97sffDZlsd?XG#I)PPtm*g{IcRGW*sJ=% z*}dfzn04Acu`33@k+;TY;Ew(tmiR&ZjP}5I(^_Sz6bBXzVCy0UuZddCpw8;;cdGuF>W7a#I%p3j+o~RZQ}BbDFQ;SMcfhF{V#D~3k#C@QXnFQK(XTH%h}V09wrUJtBO-seZ(q_XvC+tUW(KK+91_PEP2P8|dnT&Ljo~$RIA* z^8&B-Chm?^hi$AquV;d(?ca-bhCSWo2FLm%Cd^%k+gDD}&p)Efd1hxP`L%Cp?CM3%`wRhRfCFf|i#_H04mqRYr?H&2@#5ec zKjT{~Pmy;71_-e4OJCc({S%$U{PWQ#;7;52XKmjB{~pr;?*slvQ1$dUsH+{(?&Dd^ z2iz;#@BK%hIPZepW7eFt=^(M*;EKgnANY#7 z0nBR)YwF)u&iQAsM2iE|{RCRue%N0+T01QuttB^IrqR0f$rtx0l0?>ZALnp^=_ zTa0-1%cf2p##gVzZYvgvjr5p(k9_6s+#t4RY~6^--RIYj<-4E#0ysyGdmFR9>dYbf znUkZ=n7p=P)(+$Pz|Tv?VxG%*0O$D3oMDf;qnzB@kvR=N$PKW?+6{Mk*=O0EoiDcPv`;;uAk& z<>+nSSV(!~^_#Jnx~Z4#omU*c?>Om!Ev)ldsNL;ASDaWEK@O?2kncI?-N?zcrp1g| zcYc`lez#nK^#L{qYt|%&dG?nhh8kaP>0GvRi3fghH;1@fpe_#)N5i^EWe$c0OFur=4lHVV+=I6+n{*3d7 z?}-P&1c?7Em%F>C??EQD*r5gyjbDHgbH^+Bp7Un~f0jJtle1>)j=%Y3KfLbQ;Mu$4 zX!qXz)aE1LEHv_o9mLh{>HW*;0(+3NuAVV5x#j|HLyJr_GGg9yPd#`!|C;@sIY$rx z-p3&B-FE!IPPF}f?@Dq$(*s>`YOHe~do}@g#MFuJ-3_!-XUN&x8Px;7H2IR`R}kvm zKV~2MnX`{t-#*#{+~8M?or^k{P#t-y?wx4#@c+Ob6!Z&`iC$W9comeO4dugqE8QVRH zR$O1s3G92fkG6fy`|QtP*Ec2c*~9#SJ{OCVEnRu;)7lZ_M81A^*ZmPQAL!;h;Irxp!?&l6?P!hAI_&Lul zw}G00wnkK6ZDKci4Rk~+QJX)KJh_H#Ph;95a_(QV@UQTaSgc&}@Eb6$_RbaWtS4CW zGp22>V$3pMXI2MTway#FI2;lp10rKsYXfepi zlcU|URX=ln&h33*2T;^|{}!`HCuaYS->~(y6BHLMw?qx1e>l++nd!IV5^~7K7SaXRr?19c@$Fe&;j8z=%JaM--bBw!RfR|aE zsS$5*N^;K4SUlokZuy7#U9)1=U)Bug1imvAXN#&MS9cO5}&KYxlN7p=ZYJK;7*=L<2iCbS9+nwI>hcR%XHqe#hEXJ&}r?Fip z$K5}Bk4AdH1Ma3ivXvyZ^(8?*XI_qHtiIG}n+u?=mScT@mg`Q?3EX9%tyi4eW9o*s zrf0(d4=aveJZkHF_8{K{^-9$dzuGs6i&%KeCw^KSYZ`E$s$t#@1W+=tQv)FG?0@IQ zbKd2E;^*YbSB{uH;dc*YDC}K7W@w+sAov%z-~Ev`gs*EG5Wcq zcVgTf2mHDt>=JYkYrvdy#TgXO*g~S+siWPg=B^pu^FhqJtUhoKU|gQE`&E5RUye0l zT@n-Yw1+y?zaWopz`g9*^^p_)2zcm3%T+((dye|+8^r7tIkorTb90t6joYJv`}5j| zm|{P3qJsbpeD)>hqsKo(XU@Le9JYC9+gltUV#0o#S9e3aYT#HWFP9$Z+I#uiuKUO~ z1#N$0)>nULd<%#hQ|uLqv9Nnro|t@dFEaw$KA>?YrOA8WK`yadJjP_tw5Jx0@rD_%CxCVx0Q%CcgW<%}K33@Qk0>C(-sF z=sG*v*g;Ir?i2VgGl+*dcKOG78pK!iW!di=NZrDjt$>FJpp~ya{W1TbIdDhF_vyQYf{bpl9=yq zM~__h;J(<6*uIC0vzN#XFhK19-&*G@FTJN*e;%{n+xxyh%%7jE>*j!V1KanWPw@7P zT)7+ZGdT1)N6ia-pzW@58op9zz_)$?-w&n7{Mn;oIr~{7&u3Xeo%IouKhWB84s=6{ zyF6xV`_~o}{m6+#_aF#7)6l2yu90VMa{}8M_c)AeqX$s8TwL@36$>v4Ne&lpxM|KYPFK8-Kq`_6Q(Hv0l4s6I&ohk_!ybvGvc) z6?K=y2axYs2imys9dJi`e}_(T=ZEQ2Q5?^jG4~2|^&)nYr=M-^g6(q+eWZAa5z{tq zOnw4u8-0}8haHSneG&wLe}E@=9i;pRkdxb&ZZZFiHt^m5P|GELLkIrb8J<^OCvUuw z4{Z6yhT1r~V(tdy9mMSKz17`e58!jV6SFSh1Tn?9JY!pSugqC%d<5-*_3qqr&Xm)Q zGk@DnA6(%bD<9~F-_Y`kgFa&RB~mp3$2#X&E7$$b#;}F)eGcc?OU?kL$=lb@&lYo$ zc)(qUni>17_A$2OkE>S@kHu|>eH$1cbM_!Idl9qWhH>j#PMp>b#?@qNyn}p6P9@)7x%A#8@WseA1KYlPUPBII^5vBz zzcSZ7ty3zO>vMr_@Xfb)y;Hn;&V!h;=hVBCJPUg=+kE18wC8e$cI~uvoBlyAfbZw# zwZ$pEJ->yFRF9Z-q0Wiama}}HpWm0pw0Av>I?sr*%B~Q;v*fzVOD|#v{z1Nh#Fn?= z9o>L`9jEFRzwY5%_|({IP*-~_zwUnZ31arPUfY_6-O&fSd~9~doE z>F9lK_tUTbIsdsuf3_RIegVcpor`_#S##tdatf)3B!~(7p5YyG1%915rp>SQUL?8C z-s`P*G0U94SH$ZM_EMbZ4yD;i-N5bryw9Nx*Uz^9EM{1j^I2W7%O|8{%$m&EC-57b zj^6Kv_5mKP&BiaY&b)VNZQTiRpZ6UY-}TmL7vQ+t*fIRwzhcDgr}SLsL05nE>T|LG zFXmnmt#)eU-A$ayEAn=%`X>GZJHDU)9Z-AJ=Ye8`rLwakz4h1>@pQ?GM2BHSx#7&a3yl8yrPW zog?Qg=ec7T2cD%7(=M<_@!icBI?20>BBufU4gw??F}ddT8?aB{oAbVv;s?fq&g^5p zX2MTxW}53F=ec9;DYs!45B?c@zzO2s_w8Zo6SjSf?|!wDy~$}?YhLKv=cR_&0ZP!- zN1bza`S!S!B?Rt*7*aLJBz<1O= z_T3v-tTjJ?wc2t<&RyhbH+XesZRE&xruVZZ@dLD&pH1(>yG!&(ZF2SZ-l51%c=q`!dSok+mp~aYdubmzf1ZD`KeHNs2dCxRqs~jpyOrOFRS&eeAm$E({J;+M z6_f-VU>{}2GjOPBIeYkB^lsG?t)ABWg!s)~*y>Eqy6T+OiSK_-zjnr#TLLc<1U%1% zsPFW71ZwlfI@*4@{4+WG9H1q6&RIOUK(2ki-#y4Dz7jce*2`&d_6)rFE)pB&5^Rla zZTTp7Jn(z07}`Bjox~nie{B8Qk-YU6Y6C4swkhb)CMQ3^h;>H7<~(;NCd{i3(6EOZ ziB;dC+}cM@&CD4))WCDAV*#(~Kdv56X#1JB-k!co2mTHJ4(iPKO2phNQ|HYc!W;N! zYCBqYU{40?7Rp%Wb z_R?>;@GzgTy_|nS+e4A7T!|gI{4g>(8MEe4%SC%{fp%_3e?zT%+(2>2590QeFRx)c z*VxOmB3ELJ-uASg+WT#Z4XiDoF4*S3Ll3mJ{fb9^B)`Z(pV3_(F|J+whq;TX4v-%W{@Z_MwA8!_YFOVyD#ryRjA8#(V_z;&)--2rM>_E0?gh}V8C z=REi9!~>Xjw*zgyfNHUFX!jcVTJyqzChu4|;~hWM&iHor_#F=U9S?Y-)8Cr5V68QZ z{Q{ieSSS1*QRI1U>$K%<8|$Tqy>`z?!l3#p*8`vZH^+0yt!JLo9{TTRt@Q&`e>I4hzHud!Ut;pH>N6y-BqrAy-<3#w z?CQau_K;hA@}(Pk-KRWd>&>-?$edg#b%wV5*m1Ptc}HMy_R8VEaEK3Pi@Q6JQ)*^T zooZmtxVHD}4h{PNitmHDjM>YWJUP~;`WN^Cu(vk%zr3G2-W~{(H#XpSZ)alSZGCZ8 zkANfB@2?L)c~ZxmXWbb;(8d(`e(oE3&pNObdmGo+u3EIQ0XnwlIMBIbFZ~-rOvpR5 z5#O<^J0aEcc)EUxx?8qV%~=wXZb9&kDQDhFTMKL^D<8#cl8beSR=2Q1MX$K zb|Iv6eCH?`%O>^&8h#{rz(#6MjJ3V&3Cxsgbw2|`PR7eKD%-AhcRs9<_~S4lui6b+`bX>eNfk3ptYTSwh!|g ze&T!Afi4gm*auLYRnH23&&k=jst-t@&O%;WgfpnkTzlhxvtmKsUJbtl9Z8HCdv`gf z2jXbOUi;ZuXWu#lx`~)Qs<*qosWrCy%j*l;BipP;l}+sAK4?Xq_|AV^+W2+#cLzC} zKeQ79Vol4bd!XSuqiU#c#1d2uy0|&@fi_oXBox^4gt>~dPs3M?C+MJ_m$Q+c#DQYY zJG8cX$RXx=TN1|~=rg|gF!nI(Zff~E9y3p5o(JFg{@yIl8TC$JAH;jkI5+p7|A#;N z>wkIu@7iPIO7;JP&-nL3euVdr8V+u_gt9XtH#Puby5f4?0(r1Ssi|NiBF z{Cm#-$>nF4f5QBFj%?vG+Rw`8>bs!$xvTdp_&&il9%_&|pJxyo&&2SRJNAA*jG6cO zfAOBUJbdfT9bm+K*L?SUCyjSvKZhUK6~oU3^V;r~*y@J=Is68kZ=HLazu|ZE8J@Z8 z${)n!`Ca5b=H!_7y>Gxe>py~lEziE@6W`ep#6|KywD7$4Q$BkS54*=at91ig-O!ce zv!UH{=J4tc?eho@;vejX_HNzNT{6(XS@wHG548Ngpe<*BpvImTdH8btdjZ~C^(AiK zARaN#=lyuzdZxrYzdhVrp1$_Jr?z5m_2wn@#_g>xpn9yC^f7iv`*VhS21sIZ++lOw z$)DG14{~zG;@louJ--A03TN*q*H{Dgvj51L598Pa?M(X{v)0(==?6Uf%1?4_!Tz6C z+?adZ@O#c#XZ$V~{{U5sw$7P#KKjeo9%#AlBL6!%K@1u3`m?k#GJemw0+(41m8V< z*Y03z_mnR@w|wX7EB1?+-!F-6uf$i|uM^wnm#5fYu}8&n@9_f6 zIma7)Q11>m_}b?8ySR55@_~8ps-vCLhy^fbtj^q1c2dW3WxAXk> zqrH~_yqg=if&>1`yT-h#&w2jcXWw~ga?Z6b(4u1(mwN5P81u!z51^Q9z<8ZY%?>B` z$obC*crH19FCKEgopph2yyj6?caZqRlpfQ*J~S_g+1lOI_7tGuAxkJe|11Yt=M-=W;!<&L?jV=PwOztz6o`$~*tW4r2R1i92!i zpw?L*!2ASBF0ld5UATw5wd(9yeVD`V3#XswRNK{{AH);eh#NMsi4Cl?7(0i#nul1e zeLC=Sdk5B-pIGNrqwie~ta-Tra=%bxeG;1&b+_zUI%A+;`{P<0ffk$KyFG{t#FfWR z(7@0J(1O^u>j!nltPy4EE;8Yn!JhKo&lNOKy4${Ua`u~8*T{ty1Oi=4z_KoYeYM6j z*E0=!<0qW!+_l`9Lv6+JsjEAN4s!Nv@9UZPxqE<=$NGK@eEU9$TdbOiO>o`?yvi9z zRJ7*J)yujL25^?d`W&_A`PBKy1^f$pO?>yP8PKWKzk`~exTCRO#}8ny@6^Z}oBIH9 z5}$ifYh9;bz&rai1}?b{2FPf&v{(8E_CQSo_O*u^@6PiLe9^>?JATcO%ce0<6KEQ> z_Vb+PTr>99Z>+j=2DP1d=6Y`C0>6V-a`K+jUJc(F%ng0=^7giF0JSf0DVG}0v0MUI?ZbY4f9wa;)tP|J@4mG^zVqCH%NoyIaYEeIdpPu2_ER%+_l0|+v9GB6BJYY^vDW$->%OZ8 zd$`QKvA22est3TYpX=TipXShLpICp+E*(AmKH~g8pmpyS%i60}uJ_{W-)+xWdj&ar z3hQbmG5ySqzzVsIIv@TyXWbW?u{pI9`0Ttp``V|@w&zvTz(WnzcmVI*z9&w&X+Ge8 zTzx#NTnF|)XSUZXaK8>*oxJ~lnvtLP*YMSpj-FREjk(1{WAuZ(uwP;m-uUz##K-qu68sHb(FeL;@UKeR#_3d@P z?f0Mg?U(Q`@TY$xc8~guU|r|G<8VdSWA%E_$35)f9;uCG)#(S&dROY4SJeG)JF!P+ z&w(zfaXr4RS)A|kj?ewNmzcn~w3|Eops0C_?fFW*?^L3zz0?QaQ{as4@h6&|>wP-= z$lJqZt@-o(JuyQ_kzgPVZyu%luM%Y&?r|)x} z#O^Eqw)w_-;J5g4=3M=q*Za8x`*;`D*Zk=JgtwQR{o6YrR_j~?-@YA$J}YC*<1-M} zPPqDep4caK_PA=nfKwCr?km^H2Y!%KYj1t=7kRbnenC?+|G4(GkE@fbcL-~r7UYw= z#;>!X^X#5O?&}^zAMcW99@O=X3|is4v?F$m-#{n+xI=nZoimZEJpLU%woWVW9DzNx z-feyU?XmSq-r9Cs^`2k2=jrG9ALJ+6z;>+5{9kAjJ3;NYeYM`fzy=z7)IAcr^sn&t zb-q{p+BZLSJ>Xmc60k)`IdG?%7;4@bz{~%W99D}$E&UGPn_2d^3_{^=woV~t-E3o%p;OLy&I<*0`4V-(F zv-X^cY|Y|aFTmAfeWo&(F0`P5418v>UTzS(oW!_X0mRUuPg5@YedZDDmy**QADZR_lAL?;YNOz5ITqPQC$kcWSKr88q|2<@;gH z$c4k3leczY=k}d3K+iQ-W2tif1uZ@6KOqtuV{um_V}InSK`1rvwP`# zABlC34);Wp`1Tue^gGL%imfGAEBSrqPsG$s(BrfAvULxveH&W8FLhA&Cdaa;v&`5& zb`rO5i?^q-x_@J@7wUxHA78=$1=rQHPM;BKp^x*HP|FrI?er~1D$bDB>_4e!Zu|7c%2lh_AobYZ2cI=*4>=VwrdczMefm-V) z_~;F1?F7Di_I-eV_fuOd$%6(`Ki1c^hDQDmY9`k6+e7Y!TDc3F23P0CwKh+@ZD@gd zt$n{+2mS<}tyZo+^^V0l*DLu6T<#IM-igF|j{o%2erNH``#hYpddA}J!~?x~ZNdj| ze|_HvS80A+ek2<5qV4(-ORs&x+t(gysyF<{!he!KpO;1)po5uMP%--*zOkJ9V!t2& znVg4M-#>pb)>o%*|G*A4*sHZzeP_F(i&(z)mH0(Y$7Oj2{(Yt8SH6XHkLlIue-mR-|^jVz@NVMQ8zaiTD0cvrT*>I8nR9?&#FoN1XRX{p;>=-#sc1{~+$z5o?x@@3B7;Glx2t zdo`f-IRth9t+0pog?`nW+OPgE?gQ7*Stnc>)TPe;`)|9x!OMjfltxRw;@iGQdU|K} z&s_Ii?-zX4!gUd?>du{bk1qL}3lwC5R4e3zOx z@_z5hT(cJ+{7Ni5&#(5t3ZF&5>)-K9L+n2C&MN1g*Drq6#`?TC$Jm;V9q{JQ*V3lX znh6i=Ef>bv`Wgop_#HdoE61LtlgAf#_NsS~l*_%(J(T7*`oUReoQeOXr+Xxrp!UJq zZ=i9RHz(JzPkI@x%XNE5&=7#6hkDH4nH8@Q#CAgLBoLZtiESv$~f$ zV{3fJdM-Y(*1MPQklKva{x_OGi37i5?bGt}_gznP-5jvyN&cZ`FZr3PzUc$&>u#`v zxMMx5)?6*@;m$L9*_!Hs=oj`0aL#HkbJhjcRnHd91)M(G$_Lx!n)`4;Jb-(755Cj( zFFm~TjT6V(>-DzIJcwP^chF+(iTCl`KXP!+Z$BYFK?lZX1--w<@*`J2`PhK2^UeQX ztVZ;}f&_IByod6CGebQK_z41Tf(D;hwg18&v1-lvjL*GyHS{{4efM}F?^5Hccb-0- z*jj7c_hKGD0^AlF*1dd&jl8viMxDI&XypGw_W=Lt{d2AXUuSer{6U}Uh423MvF}|T zzx4R_Q!kg;`sX-U^EEfT%Xk7c4fw9i*nQlS-iIIL)zx{R>n_Mg5Wu@eEMBeX%wkW^ z;JzQ9zpYDbpmV2>sX&*4@j1MFT@ec4y@V%8pv36-kGs_m-?C? z>x$exW6f*5oVtm%&R!4t3~be7`_6eT`>8EGzBQGHQ$Mi&UhTXOjoAA!=N(<~J#WAZ z`)L!ndM?e2HTZvFXOIxN_LU#tPR@IAjre@#9l!LucIMs}8{`s9U>|jfl@oOy&xrq| zzLRrC-(mGf>-mDbcU7MyVpoK}tglh_Fg`tpJbHV|+p}ZM%L{uY+|2cL-}m6Zy`y0U z;r_g%N`+eb6B5`yt+CgT*vR&0o>Eyd#VXgccwMYHH*13 zhx#ta$tV8AH~xD6>hJ2XU-@?1`!MfcE)-$Gg_;a6X z|JwW=`yEX39sG@7z3?yk*q)#JoO^t^t+V=dc95_B_-93|XKAc|W*+kBI#yp^TeEp* zzmbPQd#|C(kmvhP2jO?p#-Wj|j_Zgr#dCusb^~|w%G!Njs`qhij z{k?0KT6>7|9tHlB{J+32p!%Y9xwqP%tq=ZZ9?ls=%g-|tRga4rczgC(dV4yPKS%n$ zH+eMnO>BTa@Lz$vTxs_H_EBRG`=)=4X zZW5pRhTqK{4m-f(w(K2Sc{J4n`$7-*vY&O;WA{G$C%OS%(bbAIM_|vrTWb$>9W;=E z|0C9zb)L(0_fc=I_QtozAa2+LIqyC4^D{`cRtDy_&Svkk)VgM@J?-HhGwDMwt@^XS zx&gkM!+xwpAAmg)`vG{LM`AmC$LgPZ4*A`Gke@sD+!uOP9` z>dZT5`Er|Ydrjg2tjTTd(}CQe*4SsOR@D8|`^6l*`3A2(K?kmpP(b}JXw7%xC%)Fa zXKeJCJ8oEWYUMot6&N?P^$hScsh1N8kDdBpeDNb!?$ixpVU2du^Dkf=ftGXGrxVxA z>}~BNR%2f4Ts^3<)|Kn?n6Y?m1ChkQrG9`qN3NdsnQ#FTPNf2svRY)^CE^&mcQjsz}s9W;p`2=JQPM zSA7CY{UA;-(CH8OhK=nTcl?emUdyWcUhe9TAF;FzJArrMTz_ic*E!E~vZre%zXU+z zT$R6VOD@Ru17bCDfuF|Uqet6$_k$0h6&JYr=cBPcGv0;y(_i1`@CNiB+q!oxto-@A zx((l1?K|+(7=L2rdoFi0$(Vcka*{d2N8YBSO#9i4)Y>?3XK6qqY&-S36?)J|iI~ z0;qqJW8JaW_w^nak6d%A@m$6|R-LuR4{PN>^(It56UPXYO|%>prf_ZP^~w=IUARpgggACOMbAM=VXn z_+@jy!&vj~;~KeYBCz=0k#MdFXTFKp7C+d}b2)=BE*+9-K2~jFI~c%KHgN*D>=R2L z*xCnE_e0!s<<%#CZfjmH;H;DHz`2c`&G-2MtaGVtv2t}rVrvE{-dNCEB)sSHxvG~N z@dpZ(ZrPUwz1z9?ji8#QyhJ2Aa#awTISOp4js5K2(IBWGqH)89N2l+-W z#P;3CS%xvo`D?aZeeUF);|Y9b_IF?VRE_>^?NN6?+=IJ;Q>Qr&f^)lGfxjAgC9d9~)8)%p42E2tb{og4kgg>w(y<=MCN^ynZC(07o;_8eGyOrXx^SUvC?@HuGZv;ppLk;LX) zx$dK5-P8R?V3|+fyHZmz+`|}Ms8s{z)~lPKYT(owKhUaaz!_Z;)aKGoxXb&_!42@M z95`pNk9&<^9Q?EowiZ_2I<4nOY~96=)ZweOz7}~oHKOMRzO@Y-Kt92fz4V=3{mFYS zb(7oxr+;b489&vKyI_59wZ^&iGuHih){gIa2by{h@Lle4-iw|R>zw-bN#aJ0yfZcs zpytWdUtZ2$KfI&Co(mH+dkqlUSar^2 zo%w*PGhyA!dokDW6D#NQD;~X^D`WS|^_f@a^10nl4`Szjg7&`N`4+48MK0Ig_8h={ zBH%#fvFi2DjK&vZ&NuJ_KXTPuH$ejR_WgU;e#buE7pKm8Yu`Xa*Wf&32M=J6A;{0% zIcl8Iocbp@&+9(!XTMiaUl+;y*>TyY1Mj7@^ie;Oq}aTD?XO;L<~plqxPTzG=8A9c zj{lVn%&EDnRoSA{d=zYgIIV*=Wd3 z1oo1@+?HQ=!`bcoFZjg%W2WSc*1!*7e|vdfzv8=Qthi&{TX@C*);Po4w{=&K&3S)! z{HjIcQWHSE_ygZ_j3f#46MOE%8Pu7teaJhXbDSE_YK?s9h#Pr(h&yo3hQ{-Y??Ar# z-L*scbK5(x&QLz_Ab!DKfyHcRy>~uf=-KL#qyJ&3E7)#sS{cV~wjH)@5I{SMYpa@7RphJF)h7wU#}t z?ZB1EFYqoo=bgDUK0{}-x7Pg{8dud5JFCyZ-rkG->W*XGg?_~rnS(Fj+b^)>)=@Z2Gnca*It2d?=#CFE)AR-KM(dE zJ5KUv-N{++@8v(vL0o6Ys_lY45p6B($dzmOH3$8*&VQljKX^F%G_1Y!s|Nki{iAp6 z{#f+^oaf9`HE=^dTCEGc?NK`V`fQ(I=HeCok9YFK);WuFj#!`h(!pQQR35(e!!~e7 z-}PDA1lkWxHpsP~TK7J)OmqX7cVBD$n)vtD1iV}auCMoZ{1)raHhccFue^2kbH8(@ zc4G$Ml8;#L!ASN3gIKM#KfJHcfu9@Shj#MB?l-|%Rcj1Fi{88bs&T&;c>8Cpz1*jP zp77`stKWe29c^jwn(K+np?C#;aXs!GJ z+5|n<96n1|1juEpmU?T|yO(pQv8TG?vGuNF-DAY+t)>yH>!234$L>YG=BNM1S;$|l zQv+|U{j8C5MJ&|)I0y0L^hyQmS^4h|)!uvlR7<{QL6i7_=IlLB4xRo$BN~2UJ$wC` zxBcJ(u5|bha{BTw)YtPucZE0Zz;2o`p4NC)mz*`m zJ}-N9Fmippb}zr?ggn|;HRN3Oa@pIZrsck4tbG!wt9pI*IdcsA?%s`@wF4Vi{Q|w6 zzXu`Q9esTb#JAVbGFSB_KC@iVb#gBAcY3I8@IRY_YsRch;@J0Ujy~?0aO$*b)mT5Y zAaZ@}_Z={H)l9K^JB#Om#=E^UAx-Z?MzPtaqXt2D%ZpFaQ154jE|8qYYke?F@XKEMDi z*Io_X#80^AZOzy5C$O$zoxL>J@8Aj|SMIzc7r248?g>AEHTFtiuH~n1uIAFl`d;YI z{sHV6a9aD?^KDyash_>c+wbff_`W0dQt#@?FW~ecs#uNl$vZrB_J^LzS)>*hd%cb?dB-4F2{K9fA_PK`b66~Y>Av|9G^oN797&EOYc ztux!}oP*5AIjlK%s=no#tLDJ3TKE_HV1=`txv%=| zKx^+zbl9VB#O4DOM{Hg7lH<-?-mkO%4vw~Fj9th6Vc*1Vjdh*8dkm0ZZr`)or-2FV zoyNKLR6C4WJs0PEwTSI=md_%H)z?h;=X})@-dq6BqhAYqrPlSrJTCj_YaiO}w;i88 zEmkji-+$*f=e`$W{i7Xde3loW?(Atk$#rmQ?I*k+m;2jKoqX4_BG!E6=bFVH4U`tw zHDleC=bpKBSbH`1z*^VA*uFLYxNG9JIr+?Wj_PS{_X&9Y>Vu!L_s7avJJb^EtG9>u z;yY6sxFG+O!^)|zMQu;GSbFp30={bT%{MT&P3!<67hA`k)q~hM`sU>lsFk1iwXpTQ zpFm9q4TNCt=iVQ-xRz)H`)H3g1OK+_Pk%L)Pk8q9{(SaTZ|utY_4D10wWsgtMXfrx z0M;M>6+e>@sOhoJpICjNX6Djp?bom!Jn`LQfI3Tj-l3W)53KomR=+pPsd?P?JyJV} z@5DaqS1{mSu+9_Y0(<6kwuAe(Ra?2=8p5^T+!m&4FZ4Rg?^Pq#-ZPiJ>fmz6H6!sG z{XJ*(_nyeBvA;b6cs_NZ4nN2XSKpj-_9&{(bow*chG6|UkF`{SN5)*gO+ z@lIXVyRVwSe}#A7Cm0|>C*R1aO`vXygad7B~8F8ix*f>qBg7H6*xTvzn= z%Ul{cdpdK~<7VQ*x&Vn=DGcm5X2oTZx3w}S1wi$dj@E^=CWViz2}gp z#&zyY&YmL4bzzL&KEi!m8K0bW{;cFX+wkpc-2hs-0JS&q1ObfC`j~;esze3v*fR6obRqyF7`gy0TP@&ouTIWan&|*0Va6vdNuY6Y=RDk))}0auW4J$8@Hp)BnyvG=TIfC=>dy{mCVE1fgCSFK#1 zhw6#1CV@-NJqK34Cva{pJqmkI`H&};ci#W_zUqAbwP$0+g0bhaf7Ruh)qMuOtL`V( zJnB84eP>dGb@fksXLb*B)dT+u5}jJ-bgwg$R&6Kez71=g za8COTWB3Ww2KI2)8>>4#jcc}Cws)=o9KYReKQ2w;yV z@Lf+d?ki`HL0rGLMQ6@ha}(QR)nADR{>cw=F87r;E)Ls(ds=($q0UCCZO%D<1@%0W z5_vG-w8r)+;QM{X8GPTmJlxn&<38rCbuVoa+kcQR;JegjEL-y@I^rSl{d>dBTH+>h zY4sCKV(-U3hf^cQwm!pi=DGuTt>=rQ!?&Mx5ByVGI(gQWJ`p&%>+!Ac*qVW}np5X{ z@O3Wt_xzs4RsD66{{d&2*q|=39W*U-UdY+YbGYoK@7I~q5Uc6LzC-4msR>u?9zRu7 zJ!=no+g~Iamv#W>SK~aB_^)rPr}+-wu%6S}TEyz*U*Vm{)nmb#%w=xiTkne4_hAj( zM_>2`_k{DzN!;=6HFEhw4Xek}cu#louiykQ|vI@}wYE3kHwch>TWwbrV!w&ki1;tMD(eC3F(uNmdoXJF$7wBY~!Y4J1H&&&3g zKYP|}=smk@271n(i4O*`XLb+$0cU;Jg2y*}V*Rs!W1YO*zyAq>ynAHevqr9(dKclH zxy8!g={fKx(06}*(Ljh?^VvU%<U=VA`Kva)GULbIuoe^3IdkiS0n^zScZJ zBbIl$kM-6@qK$Pwm4olZ)|*p1h=t2s)nh{+eI5?&D?RPo!BtDXC18!@mv5`LH5dGI z=ITTKAxG}a=pOd7N6mvZ@7X&3z*=L!No-xiw@z61LSEYg%R2mwm8<*N{T|rrlS_9Y zZs>y8?-kBz{T1Jy;uXKn9P6IUmyTS=&mA|ctIp`XuzG?$bK5=befrt!z`Z6I;EqN= zfWM#k=n?y#u}RLotnbEJ_I-nM=97QXCpW-Vy)met=dlgG<>wyfZ0>FC#Ja3kE8juW za<;K8OwDaywG-awIPuRj_AK(wFhG#Y9qZfg^t7+J4(}WdoIX$dx|6MOpNg^59I(dP z392`?G;LYW+2qipfT3@SUOR$UQ;A+cUN{(D-wNd+Xn^sgLfWzx9)x z`&|XRJ!59gfSI9n}y*)w8&2J{F1 z1fdU_h@~;--PN4bntQb;+{8DIZ3kApb^>ea-r-ykVAnd6bNY@YeC>m+XO-|QIcqXk zowXPAFV?|La_&`meD`~hOMLsD`ZK@Uf7~w```ask)*kNj270V~AuVykDsMK&^Fw zAF=8Z>pj?C-d@E)t7%|>kIx5w2l9=Yb6)*`Q|tcj=khbza?f&GdS`s7;U2B?c?~py zO>Bqr=X8B*Pu>~*Z!y(^e<3y(`rE4SCyu=bR*ierv%|NxW+gYlg;>oKo5URiux|tM za;_1p#{3oj4m@W7WBdGp{TJ{o=0$_|zSJbRysv#5eO`#4H#ER{dk*BWV`wiZ{t`4WhT+N5CzV?RmxjeqX zC0h1!7WEw@X!@Kna=n9U<D-|;*8iuK{uG;D)6*XV!zAhz$BQU7-^;a{=#`*p|F zn|Z8#fxChppL`qGBjH-;)XV_qO&kcBX;V z@A!d@Se&+lgu9SySpVLGv->k{0J#~{mq+mPpx(3dl#&FruZ5A-gDiJ zy^n6_lb@h^&{zFg13z-<+urlcW{+C(tlB@>gE^`P*7J2Z_pwj)i?vr{$$95T@3XUh z!q@!l;j@veIF`1PZ}|2f_!Xm(`?q(jdVFgVP3F2k`2pXE{rhQuknp;TSIj{jTTIt`& zsZ(Fi)8~;(tA4^8duHF4(&Ec`4q^YgSM+M?-sq#&pG98fSf0x|XUf2W4r)L5#a9#9 z1lp0U*1&hUpMLE}jn+MVzV6$|C8(a*o~uvDkq_boSM=&SXo7Xl;O{lmsp(+qQ~LyM z`~QM>U9b`B{d*<8VlWv8u+rr)}6Z#dFssB+xjWcC8+bZRbyT4D-XP1*9S*lPF>X< zKR;(MSNq-2fD!B6b3X6GIyHm1`j9UR*WkP>_f8PRJ~L7EZM{R2T-DHPsv%Zi_eR|C zt5zR)Zv#&M+}o+qUdaFbjWl`t2L8aW^}c$Y8ABda9ILOtNAhkbeD%e8-{qs3>friD zX!m)qx9v9;zCA|3akh??`;%VQCpfj%KdnXg!J+po6P>)Z&eh<3A3o+F_Boz@^o-4TAEtMiicS(vwXY2mc7V{H;wEoYp>LF_%b zud__zCUV7hcy(tkG9OEAr*`C`()deX@2Ja>w@1y1t@i{z@PDPhd&v22)NJsc#s2nu z!Eadm`S;DNv#&kBLau(kfG<6<+Q1su^VHAf53KdSe1nU%-&G&ZTmYAxI{5@=9%p@! zvwxipzbw*m;}=$swO9G*U(vjPCpi79FS=S-dyHJ6I`eW@eb(8dg5*~)!MWEFYbMWB z?^pY$m9s~uMy)HuUlyl-KC=nuGg52K9bc|`$GS6r-ZK}(T`sZg5%O~S*rAW^*wVo@ zaPsG!X}LaEz7yU-JtH{xt}|(4#lF86;IruP_H&=p!+WzvO}!uFst2+Ao<7yvy2D}h z)tF1HvCn_#=gReM&D(bnYn}bf-^qn}`fF?DI-~kmw6*ZkMI%X!8psOQGnCcJ07!u7*5yKBTcXC$TgLEb(|+_BcV%tx$V?$Lp> zya98E8?o+cY|hv_@_hOe-aZ}c%uje@;ZhU1&R9L*oT27G{{kAQp6HyR$Er`PIqOQd zd3)B`?L)3@=&Loh$JBz*s&#JG$j|xg|``$qv8<<;V=pP&K#Kj;;?X4E?S1w6!UX%d}%I#!K+t=A^;#BW%0WdjS$&&1z7FZ2_2 zF7sf3SMoLMZ|n2bbKSdr)!fN9Y@=3v>L>pLYR|j-L?;(mb@HWQ4rd5*19--XHQ!@v z)xg=qod506iGKk#8@hxa;Ouw3Ev|Z#Z^D)9b8R2%XRIFb?kBHyfPayf58#SeTCKdx z`|{p0)?I1Mmj>Ge`LenDW1UZ~?qS=S9slgv_0efZu6uv|#(wXVV&xM!-^3d0Ya6g% z#oS-Q58#<5)_b#uzPRH@>>gP63HX8U9iDgKtD~J}o|L>#lRx@Aq_)cMi`{&y4Va@7Gyvst1AJ zV|`yr?3rnjs}-NK4E)$=*iLOrfG$I4l&t$V|_M^EtM zvQNztU0b4AxK3 z$=6|JGENxaK?)DYcmpjg&fb6W;mUCpVnIxazT8z`1%ZjBV|g_;-4YU?1Rm zB`z&{Jn`ktt9S2)?>xD=iFH+9H0E0_1b)2(hH&;ZH)6kIJ3jA(voxUY)WNJe_+QPF zt3BfLZoLP&nP|6l)&cA@;O(tdce!DS16)DE)xwsa`yPo0sQIV~{9mlwy!}4rgiB(V zH8XKQ2LT2!cHfD$#|7Wm**aXUSnrW}dz*^{SpP1_#J+sdt7{_HzKJjQdfPLQveq+n z_*#iP7Ux~Kx9{2?>cAvcuN~k|xE63gE1qhZ?>sx$b^hC$BXZ@eZ@|xn>-4o&jc|sD z1$m#d=bq#`&>vVe9a!V4GbOq(VyTNS`_t=irtRHwYp~Wimpy-{w)(<%Pz$ck7@vDu zZ(i+{y!YTcF^HYhxb7qtcOiak&V7bHS~cFeTq7RXa~^w7P_T%o!H;~`fT%O5#dum^a19(2a{=NY9W$|lvbQ5lXiyXG@ z4;w&j>GpGJt^n}fg?a1znbSVjo43cneEMc;Mbo4 z=Zai?N344LB%Jr=idgqsexmtN66=0y_2GO@ub`gSZS}Rc7@&a;Z|wj{e4eeG2rVlE zEvOl=6L^LI4c~qPo4~tqIfHS4y02tza_Va3XTQLICFlGTU)+i1FIfB3IlnEd&YZR# z4R(M#np(lHi3WQBQ}^EF(0QJ~ubJiSckJG45Rco+IZJ6VE^D2u=i0Zv?%+SDo!IJ) zHSe>K>j95F?@HpkN5h))8MueDh`Mtq@7fwQY6JiDu*Nyuw{qq|)dn1E#?mCVp)+51 zN@%X~?g?M-CtPXG!8OqFMS!7?-leZ5fI63&4idjKh1a*oof*8x0oSor-`3n+ z&bxz8{9nH9f8P@Bsy5bp=Na|uy=i;prya5M>fBT9bNfE8(8s!m+J+VCwAN_N$^C)e zXHs><@*|h6a~qR)9&6=Xk-P)D&idcY!&$5DLTp_lPU40?vF5b(ymHl;x1Y7;8v|=y zJ^2NgFAm#-7^p9v{@$tg;yj_|%!;M~-`ScYv7iHc%YQg{+lKi7-lduWFXC9ZI_GbTx34q(@czla65DrRpCGAiSo^-69GrD> z)en2tW%JgVZ&)?1o@Lg%1ZS)+vF_RN)lT9jXAJL5a@XCP>^eW3%O0+ty9WEbQM2_y z>~kK`>fLpAeR=n=Uh6w{C+Gcn&)R_ZeYdxD)(kl9lrzVS_IQ$a{<|EuYT;ex)mo?C zp7v|Rzhb>N{n{(g2zqommhs3HpWf!&%Rj%qQvZry8v6Te2Q_uC#O2ek!Rcq>qWPUV z`^dSjaL(!Z??7v>NRojLt3L6~{RupW`%n1lLr$&u9o|}J{(Ec5R3Cdqkb}3UdbMH9 z%+}=QraXAwTHVuxzvG{Kv)+6IsUP^@hxcXmFy@y<_TxMeH`DkPNq~r#%dGE#d z>2+Is$S3ylzW9#c;Hnp{`#GzBrst|>OFVMVeJi?V^v^HlT|N258niyI7JDE3Gv^?u zpSd*d6X+(fdf{_5?#B4Ob5CIZx@WBSndU)|v&N;ycwk3Bu(Y1PCBS$8^Q>Q}3;2Zp zSoj^dXJE&Mb&uWWjIbkk(77h|%wsS2G=E3u-WM?7-QxilpW4J)tL_E-4BT^Us4;f`hV`uW5EEF_ zbH(=1_skRf-`>~Pm+C<-;4)gRAhw4y1-9;KYm5hQdB^(JI)i&8ao<{dH!#RscR4kp@a=6+ZNRCUU;yvOo)dnMH+SB$9}~qNc~%|BfH;t6rV4@6LaPtNGBER=r?8|J|)W(7Xbd`y?2# z&h|IFzI&OQ_~yO4z&^?QXFdV%>aqF`Y|HihUeNgO6S!Z8^IpwauRZsv&fI@N`zQ8t z`<&xvy&|Y*jpP^fi9P4M5Ld@Ti><7`+1VTf=2AV zX9e*D>RtAha|Z1o9@w^P@m4PB12V{)1NNQ7aa%Q~*PYne+~2(Ye3v^`K6Cq8;I-ax_+?5Q(%_EcYWa8n*MwDS2SZu<7d+AqL!_qOI$AKTzPt8h7k{YLJ6 ze#3$hORrTkvFe2T$cwt?#DWe&Uk!HXv&OeIS44&DZ1{CHtbJy#+AFpCt{JO`dHbEY zoTtIL|o+!Buc%98UA)mz`O z@~(3~k&CycdeOhf+IwPM6VAOy5a1u{^Rl0PJNbxJXHQr4%yLUN@#{XZ`Y*fwJ2r_u zYs1>hxqGZ04ZrD=3!t{%4Xm6nFJ=;RZtqFWAhzDmg?|0rH~dhijenwMhI`yrU+pAz z1->g|)mQJ7Su@$weszyYLYI)oi_Xz`4ri-Y3xfw&qgv`>BC% z=HNVYPwZa$&epKCZhIf%ueu;sQ!~Wc&z=*U)$uR%ms6j>WsUP&UkL18BRNB?y|m6( zXXP9oCNy}o-KE)xxqyHiuVwLt#lwDvR~v7as0ISya7ZS}&epGR`(y_1Qb z;PiCQ1WmB2{r!EFX}JAKAtIJ@!DE&&TM~m{@%ua?&y1VCug4?>ubtDqp!|A zJMqA`RzJv{HSrUuJN>IyE^Wh#nmv{^Z(V%h2eGRMAa9KtQLz-hyw5H4!3DqiQ&U{- z9$0@)NHp$qKWDKw{)86+#uvZi_k>q}dbYJ{4){U7?#5m$dE=P`*jhJ!4X|zEpWeO? zm6J<+;Qjmw@7FX}^~aASIKVx3fm^vrF2TRhj#zyHTkke@f?A0VO%EcFpzm1Ccl}(w z!W?klx&v%2G%j_A5Bw3t%pDfwdanGUQIjD&4s!%{07ckbxx@EzP|RD`(@AL{6emD zfqz?bOgQ)J*Z{5(pWJurT=`{A=5>Y_a$4)G31BaMSM|$f-S^bosT=Us6Pxd8p zd&T-XALN2M-@T5neGq#`jo5y!(rs?!+QXdlxS!e%KJ@YVdAw40{@%+ukNuo?Vy!ne zR|{MB7er|8#J&@v<0G=cblReAF zzL_@$7ySE!{RBSyE6DACJ-2uA7l@!noIBA+Zx8wEEhpfvx8=td&V60yj;fY( zc#poX#-(?UKnOSB?AfsTF6&*Dqc)QO1kf)%RIOY(&u70^;F;8Gt+Ai`x~zY0-q^ES z>$x+qIU4@2Y2x6wrYktG?&|YPS!WLhW+7v>WdBH z+D|_|zrQ$#HJ6!*~AgK;OCSucp2R-PSqmZS2yYx$MAt2A913?IS;kJMcWhefs7b z%l!Se_VT$UxrCDs{L*e4VZd_!Z7vt_KA0^7jPJC{i;hyxf~Q%ei_ zw$|0x(z6$_v!4CxOzidCy~)<wbbal>{-1M3%pxx z!k6Cu`d?;f#A1Rs_{a_Xg!8QCi^tl}*mLC8`>v_2z2Q8^Ag(&Bwfdc$D|4Ma0(t|w z>dn<1(8Ia`rgd16TK22CaEo`((>w4%5Zim|=X$3nw(1)e{-FlWReS!pYSmd^?^wVw zi|184K-<>-?EqJepU0k`pDVQ!sFl}>0o2#Rce#&x_rBPNd6L-rp^a9n<3Gs{fnN*T zW7QbfyF~1>alZ!qKB1qW_JWr0K>iJF)v>R!cO1FiBX#1`{X#BxzROc{+4Y&+E&d(& z&r4TX@739jt#w&n``uQZ+;{vsUy>KG?!bK}egZY-n!tB2`6u>wpf8f%948UT^sJc=@(qA8Q_1m$5y~+w;{t_Vf(mK|X+W=KXs$ z_D^)KnJcgM-1`g?B3Ip$+zarmwKuu{`zLbb?d5ZF>Hh#9SodqY?#x_#?)xC;45zo& zHG*Ffu{Y$Ltv%BmzWdI=f&_QszU`g7J$)X&H-TUMh}9aqpE|#` zonAlR_B(^-<)`(2HGH*!o#28-Th9ak71+DwdNQbwLiRl21xwU z;ycSA_B`s$UF_q6dw_cOL7V{F$xr+%HjIB;yysJ6Jg~;&w)RvfH?hJQ>yF54&AG1l zJy+ir{7gw~Z3hYNTQ669F0N*V8(?bT8`d+ZyI@uXEm+ocq7v2fm!Eo~!vFNT6k@ zF*aX!AFJ1Z3v037PscvBr+2KHx${~z_SgP#Tlp{sXGQW}&?m7YmS*6~H)6Ts*~eA) z6W9|cGU6*eYA*Z)|ENRboz~xz+4+WE&gI_j*FkZy_VKRmeQc9=U#z^V$Ktf=y#Jcx z$Nm1HuYZ<{ow@Y&xkcBAT_abW_3mHKD)!IXSbMwQ(5Fvg)zvf7XW2vkB9BetniV^N zcWf*s`GB)VZO6K-ZAtv&;?;cZIf*|sv7bMjv#1mH8tgT9>^q|Gx}vKy#_Ch|>Q3*{k(+SU7vH_Qh^0MuHQ*lfHh(nsE(AUCy^~2^G&pqwF7dlQR^Inr z+<|v5KXQFPQ0u*F)obOqK_x}IqJEd~?e}RGD zL@ZQ)T=Q~0msN9tvxhY`SFAbvZOwr4w(5|oIbzfpL4_3+11!X-rC1)?QLB6 zZQX+!XQ_GcU6EzV>Yw+{KK@+NvDO)z^Ih=1_3h)F1GvwXxS>;j$N##&ygdTgqXYG7 zYX+=6J&V@b?^`#`5!W5TmAzz8c>&&PxKVr`%fnR_R{RGra{0@xe1MAGM_5=^`3os`?%rjRbcD4)XaD%wc zhNjMJ44flC2UoS!T9f9{POP*NDWP+y<8^Nw#n^Y-y<30FP;4_9xiBn6SZsyo5k)8x|TnFl>;p)9$PCWyBYs^=_*wOluZ3n(w2PNp4EB8r`@!ADW-SE8w zwc-x;KXc7*_|EsNiZOfY*RPw{*7!W?PKXD-J1my|Gb?sqGl1t>j&t$beuYT_f)z)6NA_j_^j()VvmIP?B{%yN2+39;|(3?iEijbU#Xec?xH=k zfp?f_YlqOjw-UJqI;a@>d}gdMCO^T2J}m*_e!^bH%ro{4?wqaeb>j7G$gyq!dAW|ZHgfru4^^7_44w5%-TwuA2B-G)J7+K_ zf}9PX(bV4j541H$JWqXYAwKqkYq@;u?f2Tc?g(G5lVe}+-_Jn(w{kn?o>!o_Q^uNW zKYcNP;tr7u_PYT6bMNlrJZnmF=f}E-iT+9KH{eWbt#MbMK|M44V-u}s-U1EWJ+=3_ zt@Vx-dpxo09gCvYJCUn)pubUXk8|D$XK(uix}Ft%?C%*8-hLfjXN0eN)i)14cC0jZ z1NKVo_hn<%{mXTb0k6G#nD6fXzO-Y@%eh14ZEgL4H+FJ7Px)|7cw=$-V79^46Q{ zF5ZE<1EWSFCC7 zfCRdH?CL2Wi}Q>X%N0wse)T7>`jgw>svp`OGd}m0_YRb*4`R7?QQNB^=b6v}_n^-2 zs5%?Gw!QStx0kuzfpr7>N!;`6n-_3BR<1jkbL!2Ta>VUxuJNj&UNOh-pqA@Q#_ebS z#2&=v<{G@Hed$+y@#}kr(68iLTRrxE=1s3 z@o(7fUGv+{4Gg62D$wqq>wCbu2CV(ZJ2sNBa*Z83u!FvJK62E>kvl*Wx!&i%79Wyp zzl>Gm{a9?o{f7pceQTfTUQeN%*9tNcJ+a`?qfG>@AwMpK8#^o(}DN) z0&*?ZnX#(3b45!K{EwXKVeaOf8Fzd*b0_})cwhaTSv`Iy2Y%wKo&CqQ>+hMI1E*Z? z!~+97$x)A7dTN83jq;4bTK5-8ynZb}*Y0ptPrj15_}Y!V%<+x`{`BqvPprY!eWQ)v z=w;oI2bF^vvHGMO{TjZue5k2;+qDmL)sp9&Q|s>5*>lL>_w&ZCUb*rcGO+EbWGvsD z4&3qQ?p6Jm6Xe)Wtyt^47w~fuxzMofaq>oxjJ)n^rxt6MmgKoVyME*o)1SuZqm10U zQtv(1eW9K0b4%iF!@pa2z(3)Y9?Q4JchuagF|_+6arbc#duba#wtVEutGoJ64!*N1 zM*hGLYNj^m;LNDA$kC@~`=4s%%rn1%0QRyb(4u_vnCDE-Hpo*lR{cb0uKGql`Lmn4 z1Wu_vHusHQGuI6JH*9ilhU^gXpu)jxC9t7|WHFzfzh9($kn&GQd)epz{?iMzLVB8(^c z>^1R~=Y4-apd;6Pnrm&wo==|hg|P-wja4e zisz2N$hGt1zMd(}=kHJ8GWHyKYRq?*`~VZkKkS?9eD17d@&o(4ug}dJGdE)0vGoP~ z$|Y8J{7b7@9sfvR_gHh%-qhMB!8d&ED|%-8*k^9vKYuCan&;xU#vt;gi zHJq~x#Oj=Iomjn#%y@%;fZCN<5S!)@YiRrFYy0mg`CJE_&&b>cPQG)R@$ZW_&$<9| zsRs32+miD>ee_9qApf8CL1y452HJWdXaB0lPJLjW_>)@y9gwN-SrG3hvdnd_*ge)A z{2+Gn&VHV?5m&4;u1-+1wr{?%4C2mEQ`^zP?|_!rg{E{;@p1&`#9%>lu7YVY%# zBUk6gG$%lA&RV7NnLA>CpI|rak9oSe?oH0Tfgfo3zv0qe`0gZkc28{i^0BQin;9K{ z0(1Pl&e(Uz{)+F!576>+p1lUXoO{RV2mjA^-c$N0Bi0;yMgp`o&etF4L~AR~vc}J2 zqI*K3wViS5PJV~49Abg3c&=~wzXJbWrTC~&{V?ul!M%N^Gob^hfpH~s-vfGp0LIkz zJdiuj_T)@*e0P+(pFOXkJ)85+9;pqiF}I`bT|L#H0ej0=?97C-Zfw3O*!Hl0e$mcu z`1MZ5m;P{x-@bp=nd2N`59?y1a-qCC;}Z>e z$@e)Wy7GS^pLpPRVD8vE*rz1-Z2ch58Ub&PHSz&w?%vo9d!i@(dt z2WTL6j=sG-M}UEC{+X*Eb3QYdTTF0adzJ@Pnp4w{Boc4b*IW@2X#cNoWAWqPvCsTxO3Fz)IFgazEXMd zWxg};Cuq5Bb2|ty;Oz6i{cGL9fKw{YoD7iUp*z06Km7wV;GG59dY{1q`~(d@z>H;E z@9zyJ`bIq92X;pr3%IVvub%j~o$vVud=U57uX;bx6V8}?13$?(&pn>l#vABh=9+6> z-4~oW&hQ*Q?|N?ViaR`j_10KF$+6b{5v%WscjET3pXW2DtC7t0UYYwRackd**;CQ4 zJ>Ukmvz^s)*~Z+>=lYZ<=S`noYA$)kLJc>F`D=UX1NUm$a8n<7z*$r8%A%@o^@(-% zk^3z8bHo1bZa??(-jbO6eaxY+668g!GZ^N-C{j|`^ZnUGq3dUUhR|c_6lr8 zzsJ(XYezWi%SU@3GnP+e{kJUc0_Tn+m+dpGJbmC@*w6Fm2RJq6RbRNS4b1r}#WP!- zNd#SYfQ}9CNBjAG^-JtOwb9NV=q7UQ=??O9bM7$UraTC0%{R~50QOcg{6XG*>4`t^ z&6R(Hx3~8#uiXRNd_=o_-!ki?R7rwPESN)l7elzx+ zIzc?-M6TI4c+cJS;k5_$picXN_Itvf#!nCJ|C9$|opI`_{`lzJoOShkJGg;ujhxR) zsd{^W7N5Pe?NNfZ_6Id;QGEg%H2kVPyTX}sW>hZzndKc$d~1za-+@|AseRDq*{5{0 z_TS&vcaU{|cm{OP?+TQReTG~;uH<-M)_0H~1o<=8Jhk)GCFl;Cp!P4|&NrYA{FbZV zd2TPb1om-;dc^j(w${D@j6KM&U2^ptX#2@ExgZa1FYN}3Lz}0V>prEi&8f3stB+s# za8GmH!x(zRXKfNUKS2kl=VRx!glF#I%`sni{u1n-%dDB!;)@$OSCH5jbOhgL=!Spx z`{)TRr$51iJbin-wr$LFh`_c_fKFV$>!YU}wGI1lb$)r^P7R&0^1P$7Pd#t=LHt*F z=V1p>%(H&Pj^TIgitqfE>kQ_GdHAQUX1BFtBj0ybIXNSDuh{P%+E{>kpV-Ies4uTw zIu_JM(h~bbvLyqG4DXJF0t*W-Ox{bdj@uUU-5x|z2hTC zXRh6>srk&V8hz}xqnODEczxVMslf5U$RGnY300*#nG0@QQfen;CUfVSG4hIZ$P z9{3ad^SzbWKk?1E?Eda>0}Wi^>Mn7g*w+5JwZ3)M%WH=mVh{K2v9$KUzM-AdK<$3t z{d4Ws(cg$!r_>Dm2soZi`S0uf8M$V=YsER^zsYUz0qjvcI`Pfx=)|{=+Fh)95Hsfu z+w*Eq;&pax-&cECqus&44r1o``!ahM2j77bvC#4}SI)DPrnV#(*J3|oNO!AxV)8%n z8*o?s1Vf#z#QS^^3%D-mJ(3Wdz0}T~aNqEaz0g;9@1vt

=x6PUt&Em zbAeug`_z7I_DN!j`J;8@-NzZ7SOapL<34-bo)_>3xfL}0s>k+x@;iBT^&npoGd6NQ zFLm2^-repKtvYihL0w{@_kQx-cTF^?S(-6{S?98+2+)Dz8Lz0#-0EU-jIYMf`|g9B zdjwisW7iqg;7v~wM|W@q?p@Az|8;k_yP?M13&i~$V*M>@oH4N7%X2h%?&P0G%{`BO zkG#1fI>D={^LK)uwFVB@gStTb*{(i0{|w;%*TgU2+0WYNvCqWZ-*eu^KY^7V-i1A0 z49|W+e4zcldIr0bxBortg#T)|=I!G-{2uAV4{H1~xJ}+Wa_$lPX>)>!X-eEMC^U{B8TEHUrS9WKbZvu9H>*csG4koTUPyP~hb3)uJ2_OI_N^aZhf zu6>T??@rFM*a!0Vb@o7uNPM)rY8&fl>y2wSH8ll&#JKYg^i`LSEoTih^(!2Au%{B~ z5B@_;+nSDFXQoHfr>^+e36wR70sX$&ffiNU$OS)ue&w8@4=%7v(DIzWqO~iBKHh0{ zcbW6OKjB{i&(?@_vGq1KVwU*X*}a0UZkyJFb(8^HG>z<#&R^f?(b*8~zd zd-l))Id{K(i46kSdnK13z!iQ6ne%-84sq}tI1%5)Jr2a=nIGT?n%KmgvcCtB^F8b_ zwf&s!dupy>H*^Pzu@n6g-R%o%#uguFa~H&&QG%ZV?-CC^&=&NOfTL0v;zf4VdK9?=)T-Ukpsp8eO2-=kHX&%%299l=W6Ira;* z@e5GT^v?WV^c)S?WB;>VhvO{&9Q87g*w*=-F6X`I2H$f(3tj+E1NXXpIXN$Zcf`IX z?swV+kh4Jx zC6YSOsJDLyXJC!CI)EH&_0MiZ_MWx*v?6H(gP@K%zAgV=Ltm4J-yq8t?y29?V&biZvQ-u zE#~>&d1v;tUx0-73^8>F30GNj+hgO&tYD{43Ad zi})I|=9)U^xQFNSb9N6Lf&09cI(vH_=Ukz^%l88FNuOtk*{|)ue*SsUT{GvOr#+wcJz}pU z=e~Xx4#d^Y@ID6U@LnM1eJk#@21d^N3~b?DI;&$hx#%_L+2k6(Abx;zMqi_ACR$zz zcE=9#_AF5+AQx!gH~ZNCBp)sJVzV<_<@i}OBPPG{)Yj+79`bYUx5CkG=!UKu^4_D* z-JU0H@&U{nv*sqIz4fmgwlxVi=vALvdTBRe`hl)Ju=UU8a>j=oux?2GY*p)B3~ciU z>=W4Ub`bOa&D+bJLQVf8achnHy!M>W!}?BK(QW}PXUzC+2e!}0yKyIZ{qp{^cm7lO z&T;0Nlkc;B3ar3Au7G`Nugm%V2Kx4f=g^+(Szs^ksok0H;R|SS54`v;HRvmF0(GX# zc`v@l?*uP^_T>ECZh!Z6uDe^~ZVCK+cXUG=v*!_bW_Pfkcm`a9d%&{}P;=?m;a=aF zyIE`B0QPF&qz^B_`Eu^*j>2BX?A6G(oX^7f7k6f#13K{)W1d6be)5mjdp5Ow&@1s1 zx&&LUIeF%+x3>70YGVN|tP6ltU-TN=yNb4#yuJU5?T()L4D2&D$UD%_v)WhmXtkq# zv;(Z_oZN}iEhMJ@Ze|U zNWJf`J*}&IMceBhIrE8qjXr_m;75Wx^I6%?-tPS@@J!~N?Yu77!9L@1zGr8&??vDA zN={Co>n^CdxHI2Zb>e%5190DKFs$K>N*wfo{NatnlQ!yZSkEXm>MLS|9srtFhKzYPoSa z&pYrvLkMEl*ms4Kz`XY|u)U|)bSJKux2EpSz;0WY24-E^G5gfKoVMq&=6E@K76W=RqI)a5*`hn0r`vd;Rk0Bab}x*uxJUAHDU4<>AkK@{#rh@4t8+cX<2TzxRh3+{Ev-!y~`T z4sSTV;n7Dw^U2r0zdP&)b%$U6nV7_=Rbb@KtBJxe;<_R|JdUXz2O6U|M%YkJ^d_f z&<1}0`#tXO*2F0Rz{l2f?f#m$oAHZ6F|97-~&&Cjc2RyymBj@iA zdzPT(uOK&0&si(S-44Wav-O?p@0#=7X*XF0lPQ*P~V6;$z#>*-B|S+=*y8=c;#GzB>i675I1Qb_uZU?b*bYeTW%%uJ>r{ z)$(KZaMu-E&OjHiSK{YO!o-?!VsaBL=ZO7Y;Pb6LH`_k){c}dgz93#bj3e5{19a@t z*h;29W^9lD&X~PL&BU}v#d4l`;G459vGv_S&d9lU_1^BU6JLoN3+$5QC)n@!;8|k! z-}{_A7n}iBV6Ar&o8J&?9m2>`uOs6V6jc@*^hqrXRh#G4wzSYwK+2z`r7= zzv7#-UxD0yme@EabP~^vH_(cGj8|Q5bMO^&##_$!RNFt#t@wM~KHBcmvGo&F?GrQS z9q89Pi{9_WcSc#0JoTPszx(3QcfUX1yL!jyrTBh3vucAt0z|*&p-0@G)i&E$iJJOI zIL<*kJJH6xuNCd*d9RhP*#Dq?S#I%1LW}#~R}Ei&66?TR&sMz#_HA048G568csJOhH5`&$13+PkrD^&nP!bjSf0pc|)m-R~vwm`L*k#`CmN{p#=Og$=Ya3*BlSL~0k5OorS`%1 z&H`-@=Lc}sj++>vp8P=M*jMe`q0f2hOxfBO-6e-!fw=~>eQpD7uCT>-+n#bu(B&r{ zwx}6$a@PCL&8$`Any=4M&UsaDj(Eg2n;o?6QT@F zehJh=J6rWr``)O9?@aBYAB`WIk3Fz&o5xDw~{QzuJwnIWs`{YWVs} z&n>r|s|7(~o*|eqfP2|Pm>WQ;J~?L@OW+Kpbj+Ga0-&p7)~))?9GfOn_3E0PQ_IO9 zh;Qw_gT6CDO}wLbd&Tx$DhclRa()+h*RQ)f=VWgG9F2UFi{F7#_rzT1+1uO^&cHs? zYd@oRF2`J=<#}K3T)PoVVoLRyX7PM}?qlAZ7}}tLRUKO$JaYlM{kL-;u!T7#qEzkK zbK}GHKJ8V2zhWmC8#3n{&v6BmH7Ca%1KapOmjwKr=8f+hIM*9LHjXvgRZGriQTJzl z<5-v2yJXH9c{|r!eRgB!9_(o!>myr*D)#CHtx9R%^X#AcV(7+N%JC1Ucl<$K#FEAf4=>iqB&In|T6&)eSi^&UNku%C5}oLGUe;u#0pJ+VQ?>{~J8 zaGj&fpkERTFQ4&2lCVBgF7G!Eo=r{=7!p1En%jl}(WpM~?R zx3@BK?qQ#Z$u&QK?}L3kpS5z%#DbV{Yvtb5x;vcxc_n#!+Utrs`g}ceY9{Rb(*lS>W|L&n}zJUN&#H{nq z2Jqhr`FG4nV&sK29W;=EM%J8h^8t*<1F^PA+?oBW4o>ZfUcrO?@G5qIZ`@ujFgJS; zAK0B()uOe>rW?8hZG6qCt@k^qp%uS#%*#(u&kc9aYwmxibJr8Kp254T_m}&OocB4< z_UY)#p;w^TJ3t2mH2dThqsL0jocF%>ZRDNlj?Qc5k*=MycA$+1IBanu^*i>6pE-AP zZ@B?Fs9u{ce~Zg$VwbJXo$9XGXN}pTo*lmYedoX~QG)@l?GbYic{@k#Uh1dNd%x2A zTyoYb5!+`bxB2+?YxvIaPOsFvXvgGOd$S{U-LtzR&aW9U{|prnpa)*uA5N+WO_8-(8H(Nj8tL@py?S0O0;*ywgbl{uU zR?Lf%oPDbvyMvm?UI+NqJ7(OPm3+$E@&j8mu!4+f8w>1?c8?2cjWu9gyY?j%#+W5f zc*aD_t8=539{hk;acyMhOmA|GiB+AgW;(y=1I74$Uv-C^M~<;hZkPwfdjdOVZJNWj zCbyV9woi6c-#pb6rFfUf@x)*r+{@!2N zYV+)CUGX09>psY}*q$M|213Mqek-|=SJT;{h^)HuX`tEpXcHExwnsbYs~LG8!=@BaS-4L*V+y@ zbJ|zLF_7 zyJlbgUF6CMH6qV``W<-w6m zZtj+R-4$``23o#4K^M_VtTX0wG?n%|`G4E@I^5SSQySrFvN(<<>0o1OJL$8`@{<-PH{I z1oE$mnfE?cw6dS2k*oWlcaNE0bN=^}(;TpexT&F!-%s)r?z7nYJy>f@eghp_UZBtL z%&p*nEytXFZ|>+lZGPUiJYM-1`1Z3#E#F=z$22k9``(9(_YA(KdB$_^QSL{@Sy+>f1bC&Oqeg`FJ>l<1* z+cW3;_%yasTS*rNc~`(Duj)P1u@%s%>MqkXRu{h(BzxZkt#Id`=#uuEe8 z?}zj|P~1~rUITs8?i<<(nCWM;B<8c+ym|&BC%1Y~XCKd0eaR*MidKB)&aC?hH20wN z1k~6|zVi-n&p_K}UmSVB{hVG^-KxJ=W4ySwF9kf z4|7Tnz@F}IKEO&$eu4l!w;pKY9awK4MUM5#8sxOi)7M)81o^xLozV%c{T4v0mH%eOf0T z*v_dP@avrT16vv9V)EpeuezMyRf)EzJ?z!w!wX{aijObPJ`EikwB@&$XE|Mx#vVX! zA-Cs2&KW|H6QKI$?AeTAcd_j%M=`#`T4g`8btUK(m_IWo(EIbQp7_2i#_c7yq3gWp z!zFQ$#N{+F61~ZY`}xcn-0!D$A!MDXo|~>6@_`)Bz$t#{?^Rldpfsld~Kyisj5H70ViO;Le3gz*~^|vPkscNc*Cx_^mIpi8{f4&rfSQkMUG^IO# zPki`m%y`51j*KUk?wR<)*t5ovy6eOM?$XgKs2a4rj&K7zvGuJP_+1~V znwY)!-724RZ|%BEv|_)Ov*(D}$6mDyy*<}H$9~3xym4pst-j7gy!M7ul2iG<1a<3p znUiO3fZ8QY@E_o;x-)I&$jhAltaF~{aBppU?D;w$zCEnF(vI2p8oL2!cwg4MppTZn z5^td6H}D*}6>UC&q8-E)`E^HV^Um(*j<1ZkZ&zP@WA<z`p?;Kc~KDR~xJe;=(h#OWS>N_wZlLth4c$Gh)*# zy13?$HOYejN{{K+U8WIyW}M)wl@Ze}pSZPEf4Rl+OM*CX?*s7cBj?OSH+*+@&)N&F zdhg%l>~cBdyIb{%2Rg`@eFM8;>(_qzh_*E=I5Mw;+rWD_SKPqgB{5>_^8|V9y*C*W zvtPw<*Wd?I?;d>fBQ~G@_S(7jQudiu6VsP7z%%YhAANhvjZF`=GtkE5bYPt#S9|9q zw(zd6K-C0!0asBM=xgl?y1;K}?@6ui^Ac50T&_9qwZ+6h`+ZV5LOtgW*YK@Vj9GUk zCU>7-I_93{Mb*FyY@t*v%B&yOmE=5wXSKHY+QEOoP$%`AMZfCE?|f_I?zO?(l9+e6 z-|HSLALqedd#?Iydt}a;?~}v1*!uS%56629wDthG!I)zF#@^2C4u01Sa@szhn|+MI zv#x=4)8@71pWr)RF_s$wU3&%E9^gXm&NmO5c`&f$Shwyz_UW?K>_*)?t!1X-Su zIcu$Jz_@o4=zVs_u6GfYId?Vh`%|-t9qdEQcmr#0yI_0HnnTRIypHWY@)A@%e%0fT zAkSWMJ-aYB$aVe8S@Sg7*xvtuFW0_n%$m}h-_XLC{gn)CX77H_ttVRB-ikbbC)mGZ z*L~)!bFO`FzC+0GXOgRJeSn&kZT2-r&fhoI-|V4}tTF3O_-n!iHFwK5^7@0gGX|); z+PDqzT;A8-!*{`F=PveZ*oijoJmYm2_-A0=m^1WObQknZ;vF<0H)pSYIkmeSj3dRH zX4zM8_B*%E9{MY~-nT>p_f{gw3(hOgUGxX2`)|;L7=7)f=-W@-a@GYpf&2PA?PK0P z*VOLkIuc)roroPlgQFOeBY!{J2x1^W%lG{~XMZJ=n%H(AR$P6~V_aYDJbQF(?{CF+ zj(PxNeY4%8;TyM)Jo^N`Fu&rK4MJBaBQw80AOZI8s> z@1}auvlFlTKx-f9rC29NFm+*6>VLe<8s>O_dOzrx%Z~Ke0-%iG0#)8bDmSK;+g&L3LnwJ-K=q+72QDz z{xy(q?B*;a1Ha^Q{+;SMInIx5Cv@ia%*Z)o_Z#u*rH-u*TPdHOo@Y=KX!jcEeWteW z``%;3oYRmb*J8T^`*=shYB#j!t@FHR&X6l!d3W;4hu491NzU4#1_87^Ps_j>NFynN0-be2FXeEK)L5ur4n?Cuc%oA8~BX( zyz{KnFJQ~JF0|RH?muRH5F6O$)EzA+wdE>t-_M+WL!0+(zE5%j?ETgDk~hGppv#(leC3*#ZM_nL;U+}`HN89{#Q zpZIlWsB7mti+uHfQ@-5GdCxvmZRgZI$j7$-^T7Yjp#7~^dQ7e}_L%=$DrXrxU2eU$ zK1bs@VPEHuxPK=NTi;xO1N?3u;BM|Th_4&m)gC3p>OA-nGgLX^x%t-GxB3{{PV zKqv1xogp`YJZqHdm0K>1%$)g%`CRR>Vp~6exg8C^zQ@R$+wY|E?CvaQSYJI84FY(l zJ?5Ow*#o<6ekM8V(Z$6ySCTVVGql}xi(Sw&W4^oI@rrHUSt}fSB-&>n*E-QeY+P(} z1HbCo;~*DUTM}?{tJh}6k-WUyz+Q^{szt{o0dq+09%xW|pcUUS|JnNr&pJPkbv|^2 zj9KH^oZ-yKY1``xzv0*2873yL1Uqq>DT_; zOa0D7TkC(PGRfC&jTrofZ-38Nf^F@H`+e*uuLI|IeD~KcSPQJH^J#CKfHSb|v0@8% za-RE|6A7Ni5AduS-X3q*r^^{@h@W|DJF%X#m%m#(Jkh{VCqP!X_A)Mi;A^iyPTeUw zfOT~??7Bnz5^I1tIo4;Q(DKwiw^!et=V*vKAAvLMEe>!p=H0Jo?GD_*cj*A^>y905 zexT)E$w7>J2QB8@*ff#3Phx-obq4lx@2WQjtgU)%pM`n-0hHn-1~BK|XOJ-w=<2`K z$w|0lgSnER{)yScT*o(8HZ{fD;^vw*b9@(^WxfOBzQ5Y;u}kfZW36)4J3hHjrdrju-<#%6T=f;sQN4=j% zTUn#EZ@gP3MYF!XE4Dp7Ul8Bx*2E^JFMr^#%)19ZXYWJo?9Sx6o>|HVDHTwCEUjcK^qf27$V@>9~6MHzf?vt4NUc(>S>}_8Dg&cI92km!a z!qZ>zTW%U#p5lCaoT;^^cA!UKCj@kW12Ok87GMB#^3?;h#8-pbg}K@Ra?>|E;>$NT zfIaoKJG%CwzT-PvUez*hkT2PGYBAmFftg92v(bl3bbw9Pn7QIo|15CUAf6yZPJ6#M z`x)!-8aRXE5PJ&y!Z;$|c=aGw&&D$l+k3Cr73Um1T8TBjYu5Rk8@@H>%{8=hjcM<5 zwB7v#VvAdwKzWc?GlO8xo{8Oo``0rhUi~lUUD*3tn|RgWmn3e?-qsh01@_+inx5YK zN<2Y5w+D31iwE9C?!r9|AmHuyJv{~7x@5!1TeO0qX%%N@s4fXnL2&%W8GqbU-v|+ zo(nq2o#0lTJ@AcL>+jjU=Gq*RG4sadbiuq#d{BMWF?~6nY47Q|d(OXykSSSnyC=L6 z{D3dVvkPPPO=1J=`!;O*`(D)?6~471)$BIV=IybcKg{V9SL8NOG2>g#yfANVFiXAS z58~Fu@33nIv3*{|w0%bVeCtH@CNC$*H?+u{{nfr-#dRlUS06a`U&-$|@5)#ZGw%Cm ze|@!5z0eWW6$-+9&pbB{o&`fX0n?emGYHa0YM%o!`%epN@jZGK0u_z8?zyYs3ZKZz@QZjJZd zQ=Zs)_VKyPALx$nxs?Gbn@dFLIfHR0UVBR~{ubu6I?flB4Gq<_qR_p+Bol*UmS6uW6Y;iexaw|qi zuDPI2e#NyBc@eYciP*qTAU}c9<;L_qW5-tH?)83__j{|lkg@%9rJOuJ^AXhJpY4fG zp!P1z?T}+n&oi)F%pI@bSrf$Nm`ifz_0@syXwRd4mg`;NJ2!xK@iEm0V*!voo-ynE zv)RaLS073E*Tnr@Xl}&Z!{_5}_R_xCIC492=eSqpb8?LNXHfqftv;aT1Xwo?74z<5 zYHJ3zGtSsopn>}5SlXB5a^v71f&GG5#^mp3k>~edY;k#>>mF#tthtTtjE=8dV4pw& z^TL^)UE937HRDKeV!LNJE4K3oT8?>R#iMp?v1{zgn*%HHBeCa~5OU6!H?+Bbb;j(ITg-ottjINIpV(lHJpG0q z=s=4SYR&oC6mlweXZBn5$#=BSZ-Gr;W!J@FzPW=v!3hlfNPymb=TtxV*4tApN69Ev zeXKW^Neo-rWBMIj+?jXW(asv6`e|=@4}%8{D(ByrD?Xnb?dx}v^jLQz*9UW4cR;;albvE5#Y6^NhdmH5uEXJGGW`10&q zJ@9K5T5jKb`Q;Up3C)eCbKE9$Y=L+p}bnnDBCRTOQ7wq?a)5{pxdo(3L zZnM~J)h1?5f`Q$b<7a%}D@Sbe?vOzKe*QWOzO@Z>_1*byx^D(PkRMwgA|=*ZS3QW- zoY;K0*17vTtusy!`FkIG3+D!C)*vGu_&t+GwtUX|;$Lo`Yhr6aKnJ>`2hg`q6Nq~p zs8QU*&(NClPCozc{O^`~Kd;5^XrKAccdrZz#U*()Qs;>>Yu>Z)>}kC20z1gp`LWf` zb+|b(6%1kqt(4z zuW!zt1HWklpVb~GT64|q2yJIkx62w>0@>s8GBzK5VGS|Q?%gMLVD~L7=4LgA1*+$NHSe?L}knha0xtEogJ^kDmyKP@Eux&m77?ZP#&6Z=X=5V*}D$l;2wUHA8WD)~v zMV{}y^$pva;^9Aoy%IBCGtdKgKk75u{^!ftPuu!~abRr#_o`U568k%^3v%kFO?~-W zpR;|;KfuYHXRjJ;{Xwnu4gS6xUVR3QnG3YLXs_TwsXHPNG0$`x?wXT7h;`7wy4&`= zoLt|LmT-XgegZ!Cx>I5qt!nLI&)h(?&Q;oT$E>l}j{WTR^6#VG#{qQwE3gXq9XR(0 z6#4F8&)6VGTkV~sKByUJpS4nT?z8Q*zXNM-F`va|%eUtW?4@=;Yy3O7J1Op?IIG3n zx#hI2^E_v;VyAIqf*msExv%L}&uNRxwO0p0F4Q?ccjl~ju5o?$GOt~|(B8v-XZAkA zFJ8=d>S^)~o;?QI{k7}P(DrL+k{YD}r1HayvPD+h-8-JVCtS+r#tj+?BX|;|+UkI1@{3d)q_4 zoVo*I_V#)I#k(6XCnwM$Z{vphhz&1ZXW47;4kI5xc zTsiyB1CICYS;9DXA~GQpOCUe+PuSH9+j{3z4cb00fTw^l&n~z6k+?Z%>pMEZiC6+_%m+E&_eQSx_{FDwHHMfExXwLTqn{J@Egt>=E2v&7 z>}FW0&NyOuNPb73zaSNm`|=Qg<3Sz~{5z6;JV{;W1<+3)5G-`?g5 zm-Fm_ALLHJ9?mUZ5|@h}=)Ko9oaey`a?iur^Ty-{*zJ|PXR%iR)Nc64%l1CK)|_=gF0ic)IQrJuL$Pn$V2$@A-`EKB4t5yCtd&=HL@u>~ zya2{mxbCBGP2D~5j$g9ne(!P;ziP1qd%rLL8N2u5UDtea_D*;W+uT5}_;p5h zHTH~r;t9+T{F5>A&TGW>T?W3mA|Jr|hA&Uf-dB$Cb&I=`esK~LRA23wvsdf^tl9I% z8nrjOkkdBr>^(k+8&lV==mt)glhYC(@gAS1E~DJ)5!hWn=Y0G02ka5|@6*Y4QXI-{$p4>fgz*e)yfk(#r0aH7dHYza zR6dE3-}f|kz_-`HF79Q+*k+-7PG6p9aL**B-9*4~K6;mt+nuoU3P)c~QwJK+Uz2?= z=bym?JAwpIyKm2V#=UkBw}-nJvt}ixe=qY zcAhb@bFfMxX52b=ID*=PxPABhzK5I@?%CdGdpfJmOO1JFI%fqvAK+~L$jrRmCLe;BH7ogs zE*|y)&i=mHtN7#-$W5>kANT?GXVtaIg}|0;tf3p)m^e{;0nC@6`W_HGx!3%)CA9XRf0M+I`L&w3GOAXwRa3 zjjkRI3G5y0qP>co{c10FL`KZB1zKKv0QPMHk>}pUOSqiF=h>Z$UtBoa_VEs_^`7<| zk*W*g11LR#JAxJa2;bZuGhbpnX1t-_Nq^7o9DgrbqtqS5oNLZIvv0-8yN~rP=UJ?) z`#@KJ&XwkXJad6A+Z~PNGGbJ9XyNB{84P@$UTha_-}fa;`67P-9&J|NEx@1sLS)mv9?6le-O!2U`A$UFRSl zpbPq5kdtpr`wDd69mug(u|9w?|9Reu)^4Es>$8KLs-^c{+leXW)}=o+dI%td5HGNS+}BV zFL>$`sQ+$|L~`=ZTm0%0PEZH8q0Q^3HaNhmn$7R}+6uPc;jzJ;oHb(Z<4%Qy3!LHm zz3RuD(XpLjt^KOjU63Qb_0}FQr{-t%=`nqKY0GmDx%&1A?9`{9xsDyg)x$Wpe^ytU zAJ$_(_*pS`f>V3T$+>^XdjKo(2F!;#TID--NzA#m3wGOj+5bioD<Kzx4|L5-iA6>Y8WcJ+a; zSZ7>b^+DTT+kH=<;n!Ta-m5YB=S_Rxigk4!wLQ_-aEz_k?jRbFo1obPU3;LNX|2z+ z^2A2Yc^zFj()-??&0d~2fg=Ba?VjQ}<1u^0<>dN(JBaNPi3RL$ZpF6u5!kPTlDI#6 zkGqTeuH@X`+I?OUYxr`n$T@Sro2E}(F}8CfD>T0|V$Lfr{7&w?^^q0i?comg)=zzO zQ`5I3_$hL7J6dj_6DU;^7zrNGa_hc5r*Xtu<127Roe4XM%h{g~Ud@o}-J7>IJqvX~ zT&};<2DW1C8o!}?Oun3>F~pdXIcxi-eO~SEjGeKabDNsgygSlER9|8(W{@f;2!pJmlgPbf8jP-u97k540KO?pwv#=9{PU1j1@0tz5T6M2e=WrG5LW# z^pmW5o2wsk@YCGo103VqEZ@Z&4m9J=vi|_~*pb+8y*uw}&AD4c_Xju)8TdUhfHiG{ zwzHh^;=AOSeFX;b%=vk)XCMzwNltqWJb`%E_V*nRbP0Bk$yM8TKjZV*${Fn(`>X={ zK&-g<%E1J=wnvK% zDEWcE^F611eD^D79#|t!?Y!$lIsRsZBodt1qf0ru}1LphbYg zHn(==d~)`=cvXAzolMSWk!b?SZMNE4q;|bT=e(PYIjd@j58^(94g%Y{740t8nm1NH zy6X#ait!8U@hg`1bG3s$a7Lm>wE5kFeInNIJ5YNkksI*er{C`H8t#BwGl=;;;U0>8 zGG?xN8_S7(PqncIw|Zb&vjS_3={F#+cEQy*C-(T>d&mP<_S|@Tt|2wsIJ1@V?S(JD zCAggLrDw30_ohVPZ!L$+?_}P8CO(T5zqoDdy`2?z*5#%HmmU1BQtW44pcT*Nxr|57U5&YmegkVv zKe4^DfxpYh`EJ~{Kaf+;J+_``zYCmcp9D&ed9F2afOXmfv;auYS=-R^^#iSV2lC|C z8L;K~Y}W^3bry0#Os?A6RI``9ecZ3U&+&);4&t3yafm%j?E%PX3DC|+HC$ufQEYK% zSsxqZuH+OsRfje%_XMi%<;)Lr_TA(}Jacld;O;r=R^lCKUw~blQ?hAmj6E#}ZOwj9 z`h$2F!#-l$&zi&Kj5VbA#1fw04`(yRE9@2BM!|{qpD|k39C+!w(%Fd;ATz z9{tQGA8B9k{)Yr`phR^|Nid49X#Rux4->+fB5$? z|DlH;efZJihjRX74VwXGkECX$Ck&qgHL^bp1}uk z{=<(x@~C(4p_|Y659#?Id-(B(-_P@Zklz1C9)9@YH$3(My#F7>=g$tejvvDNf8^mu z9(u$3d;cHQ^FQ+V<2-{8)%*VlZ@}{Y&i|mje~;Jf_xt<*|M=`5S&m=*Gt2+<-9NcJ z+`eXc`LkcQ{O4cv&CBQg>5pGtb$nv^p)dTjWCzHRvfuXxw;55D{D%Xj|q7cKApmp*3swa@*;<(W@^a{0PX{G-c5fB!R^ z#mk!?|BsiC|EPa>>vi{j?bfgS`A3(Z_`z>qzWnZwEWh8Nf(eh_s`C-d1e#H+gmw)y9m#_NK zA6x$4m;ce_Km7BLEMMQgW%`%P+lUIe+rMx%}`Sdj0aQAAQgAIlullmXH1)e|Gt*7ryt_Z(aTD ztxx&UmoH!W34d+*vakNh(uZ z`P)DL)bb-Q`^Dwf;m-15zxa2T*ZrkWTK?4CPhY<37ys1qQ@{BamtXk4cQ60!H-2{c z;g9&iW%%Z{GQiz4UdYn`@BfnTTt4e7er)-> z-}6Juw?Fxl%YXGt-?x1HZ~f46`%Moom!JAs%Qt+}7cQUo?6)nKzwu|6FaD-~fBDvr zKQC|ibC=6QFFh^)?Nd)L{~yEF!BArxhESmB1$D8Djp6DL_dDG?agyCv7F-Z znYBNSxUt2Wu9Hm}@XMGV(oHxuYdohlm2mjQJg!d3q0>}%zB_-4URFwC=5ra=Ql85s z@8$XTP`(^-hVrN47#e#V{*L$Yrn(R-Hn}slzZ=IWyHY#MgZ~{1V&=I>?kF#2$hS_a z%hpg=W(z|%gtA}gKHBJX=lHjMIY@B;xBgtlCi9K-y%fN0JzwHeLo3dGc?gS3QsVnt z9lRAMC;l2(OqGT-CZ=7Ztyvs=v?B(u3zHVKA?q@0_6N{L&aYKi4;I-*zf9Pw*|yXf|JlDOXcFs&U+X}PqRe-{TbDn5pBr+ulm zMVr^PO<1p|!}fN4(xW#^WcyKS$!nxdkl}D!4JO{~CRPUb62F*D6x*FAvf_phd)>2P z_rL05!9Vm4HtJt*o#I@g7~~` zhooM{^5y9awhma#tjRlRX&KJf!#0TFah9Tem$l-dRZ+B9yOFuuT*(d{~-_M!P;WUJ#J+|TaHKF%+3;y+EM#SWyIr4@Bcw(@cB|EN+I%m>Sj*sSs$$8tfxGEZh5 zb)(9ucs|(phksWniMLFLi#x(kQ!liZ<~wdt>)r+)jEiEUMidXH3EXi^q>KW~2S$r431&dF%QA4NlU z)ZgXJ{ttM5N*)U&@4GeQDDMqB$~FCi_;x}ZgR7Id-P3{tr$}>vAWOx?vAit5l3(s@ z;|-I(T>haiEepExaZ3o-t_yI_Xs8;GY`48^Zf_Tn*rC4NoP;m$Gg-2B2+)Z71+ z!pgqlB^_h&!AeK*Lc}C-#}OBC*mq}9h;&}8F5;edLq+$g zcUh5>ML&y+T)yxmFEl68X>A&H4pm`Wd=s{~eu8nI{VXvE=ja!^dF@Xchu$vW#N28w z+`5q&1x~!FHj;OZRmGQe_9ASI#9ga%c(P^-*H87Ni>{AI?LhHtOPDCm%%yU2B{z3I zVbEqWtF$kD@96Ty#t4SAg>rMsF~(?45JwLmBi=EaD~b&@G$~UPlOD*4qpq)E)!{(C z-WJAQWkRSTJU9}PS(G!al79qDhQc!TbC@e6Xhby`n z!k~Ar1dSKtIBkm$ZqKiS#TpeM>h=@sp40DJ=T&R-+RMX&f|>=s>Prczqj2q1H{pO< zvh^UByVfJ(_gEJe{X$vM4Wav+|1kfzpRfl#g{u>03)%x!1XH;j>x#S@>py8<@t;Kz zA|(Cx(qlVZ912+fXN4GNTSQuUK`TBO^*1LX_lOks^cxP}GdG0NH&T(Ec%g7K~~LdKt3!TX7gV53W;szCjMwi6?|Je=PC$_e0@e+-+gy!#hG^?R~+_%TEZJmoE4o z_7pai>I+t7-MA-dFV6k8$FQUxP;jvo)~($t%uKr~#J+YHdN|z`l3f2m{l_khtj~h{ zzFEi_Zh^gtmxSg}4fNhdF-Pbid`ARKINvcA%E+j?DZBSAasi*@kpeCw>d%fjkw z{=(Zo^Neg4aeT z;bqY(K|a0<(>KJRm+L-EIuf$ zhQg1Cv(`#ET7usucVY2BS7D^SvhZelhcM)Et99#FJK>az6#A$>5^7UF314FUpkx*R z=XZxN#VrH=%f0ZlZWh8W#t3Js-dfvdHCRu+ycme~LztcemVY)8ZY;lT?Ji?49Jn(N zuQwmS>5>qP9dlU7TwP^-_l1dI9;uA3TSJ6@vr>gGwO4T=Z#;zQ<6*SCQBXasCKQ(# z3QIFQaI?Q3cD1(&|K*GkY@ECWPge!uczFqm_na4YI!}l3q$fh5r3_+qWpMID1}-~j zB5~k#j7Unr^(_}Mu}OwyQf)|Cb`-lC8&P(r50ck*!`7oGVX0M*7u%#5Ui<=;3g!4z z`y9F3-r;!D5%lm}4VCpFP#>p<$z5LvMH9QCpUf>GpzM;Mux}u~ygiF`Kc#qGH4_s@ z@4~0L;V3DrhWIuKOB2HppPGg#(~{BC;y1?3)1{4d8A^vdM}_$a9^gU*TpI)#kEQ6l zWfxOUtz|~~4XXSK$IKBAp!89eXBspqH>DLBk{|QdX8B zGGPMR_W0xG?lfFKG@c4`leutqKFb!SL9KK@axH3c&Y=L?pN_&kV;$xg0!2n{I8`we zvtEtHps(Yh(LMsvUDk7MpD_NAKSx8|l`y&g4Dm1rkWVF2m+XX#@+VA9JA^S_ z8XV9O&vUAQcyVSTHjiq7@qj?O%1qLq{3=%#b`z#D^ZWGou??vtWDOk9>I}MDMaOIDEto4njo|Ph+Hmaee;5LdR zpUKJY!-ws|_zGTlIA}2I+g5VBelR*+4??BS5;TNOgzuxz!q+qUxO3~LkdXIB&~H~o z%Gf*58rzM74)$bW^$tvPFoC$v6Fw^Dn!uZ zLcDO-;@v+7(Wy}a9sMkHbj!me$z4fVF3YOOTexHpjmHXC@#W)tOj;#bcc3#91|6ex zKm+6I-qNp25_V~qKz>Igvc41x@|cM8y&aMLdLbXvjx(NJWKf(U=gIg}*0q2!t#x?X z`T#$>_F%>xEtU^+px1Ob@aJ(FthSY6h>ay;%l0B!L6PF9GklTwn1#s( zV%^|Y933}ShZkV|QphaYA{e9S-yz2=@vbEU5B<_q^S382AgaUk-C!ZVYF6CUb8@0Q0Ic zIVC%SQQq5GuzUrtE9G$Jf#vKikTP+jsB9O@SL@eu_6A2zski_ee;v*l(}?r=A29x4 zKRPVw&6^7o`S0#Y)*m^`#6n;GI532oCw4H|D3mcBF|@Tz=IvA&8vkj>jhmgAzj_ZR zuiD3!^J#RD3FG}kyZHXhNgj5Q7Eef>X1~BR?r`rRmW?qH6>n*X6}|oV@N6KT76(#h zTrqc=Twvq8S=5-QMIHNJ(AU@C6ZZkM(P~4^a#KcLdx0M%-PlvwoMRV#qm_FX@lyBq zG#(PdlM4b_lIh6zZW^L%cO~(~$R6U=XIi{`_b>KnE3u+HgS!rIrN*5dJi1a*T%)cg zR(WZO!tZ@NzdwXcqw|?~sRa$`nhfgo6xFBdQ8MBs$`=mcXrBN^Kc31GvvB5jzsw^~ zqo}tii$9BwFuT^DvzDCTq_+o|JSms5J&*B{$sn=W+CWV2G!XTY)W!YbJ;d6tHeyrn z-lA0K5K)KDV$tG@yybS9J?|th_Y*Z&%@D^vj-=Jj*}T&k z##GHH$s4;)lM`2HwNX{BpK4;JFC0%LK9O8{HdefjCpdSZJM3=Co980348Gp1E zZ_M)9S22Rl#=(4))|XAyavWmUjD%IkS?qI~a|Rve^y`{px|FimBr7j2J2RL$KFS>I zNc#J!iO<%4m1t;M;?VIjV*F%fam^!XQAumCXjf(;iYFy}dy_w36?-%F(sr(n&=jSY ztBG3cMv8HU(KK$_$6X2+8RvYSO7k-LOre;X4e#NQ{u;*LTkvyEG^M_up~> z&#`ytMMi%+L_g_kTxGJIP7XeNsPD>L6>H9QHR2Sdq1^iV59e>_nTjOdBlZA zjl7V0i@tv=c=q96zB_t>k1wneHRQI5E5DF9S;r_-h++JTJ z!S<0DInM|DklN4MzEg^{7NP zYVK{Ju0+$9yW@ikuD~HXbKFuNiNV{NP!glUDuZuOYgxUe`-<`O$EgM};Qt|FiR~l;RaN{$5#@X#f=zqyb z@p^^TJ2K#9`vT+he@gJFIfITCFvUBVflEDDXrC@{r4RLNuhPQtG!1%X^4Np(czDZ> z6Bm|2Y1|3qrF_Kz<2&$=olyy4VWLP{qo^#Ylf&Ar$21BjuI^ zU(|I#WrV~dFq7lfut-7of*z-AyM@&=A3<+TC=~NjF++kW6D}S@ujo#eC*@L8;&05z zxrs>!db3}41@&CA(OK~r4WpxlI@?}w?y?hIi_T)wpG7#A;*21pzk2QUj(l4wC+aC43?pCg0UB)YxyEjze9-GZ*BqtHu`?9Xlq2Y407$RqtIJGd|1 zdko=??JHRD;~O*H9b?$gV63ex#0`f`I6HfAyNnn0ug~B%v)@=^dLLoUCN$OyLAS>z z;1ZFEfh%ONaa)XF_g4ws4&=bHpc;quJ^>CbSbZiA=?i7}2>R%s8i>b9UyzwHl~W|X zitadRc25pPQq@AVJ$?E=(@rm`lozQFenKZ?+hRZ5Go~N!#nI~7Pz{oB|BgCj zmo^Am+2>*Kr~tBu=OEQV7uTW#5t@5J*b>knyx1Cm;-EyqbLcoMSnyN${M1^)2h{NS z>q=O5^b_7R1qi{jj|$ZWzl7CuZwjpeBXP2QBD`My!#h=D_DY_DCKnOgvqwPxww18b zy+8CbmcmZXUbr7mZt+(*4b!L0%pJl8 zT{ppO>KJ^S7l))aWiER*p9@MqqtEQW=+T;dRQ)pXvymqAn0iT(I_lEmGY*&^)sS z-jlaM>-8nv@K{RE5iyuz^a_cans7g)4Av7W1;1&=Sa7W zW)?;@3*zS7J%aA54_KkM3gY!_d>PrDfti8m@~u(Q@B6S|gn@9q-!&oVb(`?4e78_> zxlM3%K7#0_${5l#4i>sWkm{W&Y@BO~wDme@(3PBl-DBa^f?dc=7>+q<>fH42BC0km z5XKA%#)UzvF?EK&Fes>9a91i9R1ZmN{sCjBXCru@P87`E%@H~ZQ-wFTCPS;cGPYhB zhf!0ban?OwNK(?mSPA##yxoFNwx3}7bgQ5-FGnzx)raisLukvl5>~A97M{I{70Q>U z3o5b0G4W0$;^&UV&l!p^U$CFri!22n|IxzDldA={FkgH>l!;@9U*mPWI_p}r`B77X z+sv*&oHqkwC(H17#ae7}b3w?{SOiW#gpnQ%h}sne*Om3C=`oUSRg1Y{beLd%Vt`N; zeh&WDHo~yZCBlU%--VSQ=0G)m8QQl_M}p0EOk5R!yrWXQH{`A`e*XtS_-=yax&gwR zTOteA+i}p&X2IuuiO_iCvoPgRppf6S8#|1Ax%lKW!A{)-@49criojojpF?*H&~SnI zvk=^<&&RD-!#I8LVqurmLasPrz?TQKg|#*3ktpf0$4@p0<~?!+9o4z$6MIM4Kg$pL z@nf;cVY6^JWuxG>CsQcYpNhvL_u_j_F>?FH!P8uuHoNR-a()oEp*c}!<=Vv*!X!g-bY!(xAviM-tLOv zQI?EG=|OnC-3sfUb_mOK&Io0v0tK12XyiOw3OD=SaK5L)9!Hh{(&wo}bIzKMEB+Tyr3Z7M@+eqXdqJkJJoIL(z+F)ar_`Pb z!5`OS;zD!yto$Nq&cB8lZ@L8=odhD$iJ}Dg)b`?zLt$RJ$ zvB!zet{lV28z0~=6^6gT>sglAgNeQh%ruE%Wn38VZP>uqchd0I>pxtmRE4U0FjQ0a zKy9A~8hUrqBcOmPn&91M-(hfe2vh4j@GpN5E0gB)`uibV6k*1z880z%>vGKdeuW?V ztYGySO@7V_#LA?;@ai8aT-o;*hI5`nbtYkBy@B1c)tDY@&gDbKW2tt7Aa!KFu&Oi} z#&c$4xuAgC3qQeQ*co(bJckS`Ta;9eM4`SO7U3QS?%snhUmPIqvs@TwXN0weZ-w{T z3!qw}grqex@SeLGcixV`?-FawxOxeq@|5#GmwYxaY3KHJbe} zt?vc&k>LKrA1!cZs}c?j)PZ?fzVK?;8A0%f5_C;(3I#Pw`mfs{YTNYOq*+yd(c=-gRRToV`hO8Zgid&hPvFe zeyy&EM%5Qkj6zfEk7PY(lp`SKzqI8?$;wAyb02f|_%&MS`#MZYQ95mpTKMN%QCY zE*x821-FUC@R8YvFJ++!=(7h`vQ?oGUWxN3im|JBC+_c7XA~v4+MyZy&Z%?A=tnpd z94Nv3XYuXBK15a|fKhRnv!NL2kFwET^A@)YzQ9;!J?tZ{^Xr z?uD$nF2DAEgUqp=D8By=pEtildT0@vy(jQp=W@o>XX5$CJxCm!0rTqhFkjV|m$sI{ zB~5`!Ee`B*h@AGO5&Higz3=X`Gw^L4`kMbi-EigycrK!k|lE%lgO@har8ffUx z!pSApxSh40pWIZq!@d(`XI`MsjC)wtFAGsaORy*4F)Wvyg3CU64s9{yk(hD3SQE+d zX0p_@`UPxKr0jS{>SRms-JKxFFRFwBPU7UYZ&)OPMB-~?n(&JCe?0MP5T9rK#Mz*02-((^ZRZp@ z!s`!G&M)JLJ5DSLmSs@QQ*_FC!5!A9y4Mr!eUq`JJP7^Wh9a@&XV~97kEFN*2&{UD zG3_TYPUAGT^(;g7t$)Jk-@8ygwgJsMI z%EOzuf3^^g>XBHx#()>v)u__G0+q3)xLa`w`hUN}Zhn6nuW84V;1WzR{|1MnS#Y&J zijd7#be*HcK|jT z)?qY;ZIfhZBH=S!iCrzUSnzKERUUrA)ov4*mEq3#&C=X(tv3yPUgDfoJi;Ds;MhmL z%2SLBAEZAVg4H@1s=0RJ?umAs+wuofYcg@Vd@=m1$8}Z4a5Q&c!jM;i===gaKJ#MUEN9Gwswan+fPlIq*7zOFd2&|54L{_&hj2x&( zmy<^j8haC)4$5%is7=^k9E~6SBXIPtF0^M2$CBgbxb)B&zdB~XSSJ?s{f?vb%1d0< znL@j&@%Y>pfvBZpPG;mQcFt zfp0I?<4X4#aP8rMO`9j7Z1{AnUKfIsapSRoev}^Q#aTZ?x%zV?5@cpWP2LZYVZnT; z=E)Y}B|8+A(P=UYvqJ{qc~%f6N<5OyT+G*7QaCiUfXX9oaedQ6tSFF3HAhvMdtw0A ze4U6B@s6mL9*l3thvSgUNsKrWhR-eWh&SuU9`~crhohOhVhWW~-{a~EZ8q&~Ld-5h zZa6cPT8CY^Y4BPu(2C-*`|l;X{6Vh0cZ@MAw>cqAL+mm~MqFd?o4brp!0?tnzh-RU z-2HX1?I%N@_BMR7m`d9&$M~>BNmTZd)WNcB92z;0Z=`y1|H}DPcsqt~CLiEZ&rJR( z-^`u&Y^hM3PKj|PRz6eUSKs$2vS`Ehexo?fKb7wW=dr8LEd*qUJiS4mwHzqQoEhCeP(n@9_PF$n`my_;6?oGW<(1YU^{_M}}~X?Nh#P@aL^5 z`Sf_Qms#U(!t7cRcFdl_RVOz4K?|t`S zc<&gj7t6CU%z|2W1L!++3xDb?$DEc$P}=2$p>pS;x$p`K@-E`Y3oH0pPQ(LGYnH#x>MBO$7yHJ&|i|*N&5K&8Fh2G z{Bi)BdWX;^I*QfRx6wQ;6x9kZaC^somb&j|uyYERMTTN(Lp{b_7)<*y{#4IdD8VBG z=v1qSJ|5O6+dmtnXAZ$}Zw54`eL%c~t0<08674m6Z#w}}n#`K-d-t_f7Qor#66!>RY<0{_kV%=ZD6{4QI}I5B`0 zOSVuPAH^Si;>tHDz3~zoTrWZ{;v!}y$Wp zR}k8J7|k!J^6&guhCgdU|H5{xefS#t$3@cSTs#%V`||qhC-8KrNB=W&Y?_$EKa(!7 zr~Y^D=@P@T>8p6h_XxWhbztW|iJq$}%T;$9aiwxE^1p>Z;uQ0tg$C!2H0Qp-p|rX& zkJE!A8UHGjC-2{-{O)rcaQXo@?L3dYo36um*g8&Z&E=JqaV&}n;KE2}X4)O+4%2u1 z)wh;)m(I~qbY*1oG`5tua{C_}x^^|@+xWHoHD87o!^bd8JBZ7sj25+e2Z5zhchg!KNYGbaGuY6y2a0C{ueiv zbys4s{5K39sK8bw30B=Pn>s=3IbeqZr?JC}DRtl=;fKUO%$vpFmk4tjCO zY>k0p^f;=e^rg=)X==J3#(Iqt__!?xCN|D|={1!@tyl2BMSAqE8&CUg3)rrB5=(nj z<8^%;w*1YeC}c21=N`?*4WR5iCw55u=ypk#LF3#8TK7K9nJJ1?yQ;Q!C3zLY>l`A=;H1|xyw2K!cIe`bbZSoH9I!)MVW5|2Y`|{Ag zcp4Pd@L$w*HVsyz&(~Re-a3%EfdXGk@8hlJc;1)k%b-O)X;G-oy9Ea&_h%Vr_ZZBU zxz;Qa#`2w)C-*qL!j8swX!-pdlU$Ox|6CGxZg|Ylh6&6KiKKz|ab7Y~VeaM@4Ete7 zIk&CczIG3{HXh(#oBbGD*M%iTkKm(GfwX;6e81!(EFVngh*zr4t zbxnVXDH9n|qg zER5gkkMPXyKlrmF3*j;?SQECEr$YmHY4jb9u$abo;_ z4%H3hidTM|(NaL4;5LS>krQ)6hKY$?ti+OoDx&P{^)wC&BBN$8_T&LXi2;}^;iPhp z^ANjdFHCIh#nR+v1EwTZyJ79l4Hu z+P3q^OMfKvOu)Z?`>^t{HbV^lVsBV8j$Bn^Lz^wFtCn$)tsBR^pFr(tLF}HohezYL zvRa~PiLaMKz3)k6uYHb%XM2fbJY~e_%gW-F4|N=B`H|;+G(-(MJ8{5gH*v-L(V}_H4G92*aMH^U?+d(XnR-4@mYSrD_I?`#JLU zIZs}nJO?>@{qQ_z2l{__iiff?bU&=hEbk5YdMEu?>;>Fn-1G+ z|MBz21kRrC%V`D&FnB>cMt-;tlbKnV)Aa%79%z!}o+>3+xES*i^1umqu&Lhx1jjev z$)fX+smO+#1na$cvlZ@(VsL4fAHF^AC&s-UDC#|Q6`j7Oqpj;fNL|gtAD5GyyrPbm zq`Hb8S)chK?m9>7ePdGJL85!9rg&hIym+>9BZdZF#Z<3ky#8k(E;0PaIUBo*qx!~E z?sY1qR^4ZJ(_P&3Je2i^Q&{)dA44U&WX9$!=;Z}t`pUy-8k2xso#i}J8^fs11iru+ zLQRzhiw8)sxv}W!XC|tfj1Uj4x2A{5NdCGvlf4cZihV60UK_3|TB}B|epNE{gfLpj zO~<>EEilzTh(#^O80{0sLGGtGAuLi1{4Yp+V;3q~kGjC8XU{Qx?qT+u;fIyw&dBoK zgl)co^ggtSq4U<$a#=kd?L2~_$v2QP-~@Mk%3#+I=cy8?Ebch{fg@kMr`73rj8#oR z-RNx0(j3W+$yU6kH=65iUq`iZ12V2$1?KFevF&`0i<8WiIV*`d{KzG%8kp#Fhb>K+ zY?40C^sd*~tMUnR_1_W>P}A-+k-lviWgj`_6ulm+R$OBj%L1Mf!-10?*k;l&S>4+%nt*H&nkPNCw>_jtZn zl~)#x;`ZD-g35={_-1w)M&t5uZ+kozuN*?(kyo&G=^cq)B*Xg$2OwyigtsYw#fsMM zEX;m~Dck(G_hd4IRF=`nXCL!7YGUMcJsiCii4b*3FDaQQ$v0(j#mHQE`G&*y!F$-{ z2jIxAmw3?Y4~C>k@|Y>{Sah@nDW}$O@ELQO&Jj30Fd6HE|6!@iB!1UggwmjkaLAQ- zC7UI={kuV^{&WgXGwV^i&X9d2vo%|aN;t|Tlo|$ynLA!VReoIQyirEFJl`Oy6AE1i7%MArU#z~NHnVx z%IuaH#Ekt>*n6rHGQV7Ty>vZJ+caauQi(>=_XnNVykz_KN3^*t#g?B1P;nhXh5Hg5 z-rIz?6Z{#Ss>n4{y6~??IX9o4iK!M^2-|%gcIv4(F>WY-Zau)$^RtxII`thQ6lF0uc^MoVZlL#+qtFn?q1;N9 zy=rHppW7lhD*nKNq|dOrR*m)BI?#LUUi4^fMC-YpTx>O)@0&e|<7J%OzKXGu|23*{ z1kLRf(A&iX3s=UWUiktJsVAVeqZj8t&cMg;9MoLYq13D4s2#cn#w!0Xtf39hdn{z2 zF5-+0h1ief1p>X;XT(3xGBJL$%u1kNq&Gq8dWf?GN%ERLv9cms>&dW8lk1a=r_bE(Y-h?|vhmj`H87H1H=C!=;e3NC!#fsS+ zwPcC7#&@{r5^j-~uR>fMNuk<9yhE7DsS6x)Q4urL% z1{NkA;V_&k63zNW9^HFT?#obT-W@K{vGil%3D znqR}aZbBrf5E>ruBBX8vl<+Jvuh)OfYYDpgGfR_Q(ng=oczaiZ0s>$m?PW-Yf zk}<)1IeK3b^DiIcQ<*G=+G;b(OPY^Y*JG{kbT%xT$--evXrZepW(`QD#hwf{H#Rap zth*>JmqPod4NU(N#2eMAMAIm?UVehejo(4ZoZg=CeQAH`59I2Wh}#+^GixW!#nDbK zk~!XGJX*bwS^pj4yRsav-kZp6V=hu}=N+cLOQqovD>3=RTZZ;(U{vlXaz!_Goh;Aq ztM$Z42^Kf~wTk;j=dkak6sD~>%I$t7V*C??d*^nvQ4aH&d=HiWUj-tN2thijo{B%@^=egBO8GPkASG=p{zwk;9ROwH(OFby|06qI( zqTd?@G5PsD7CvcZdYy)7`d&%w`eqRKSx9RBHAO}TltVh}5@u`O#;_<`zMe3gyWD;u zbHF=VOiZK7j06r_zL*hRM{`@P74@gp@bkeO$#?Ch`U54(q&P4)><^;cTex|B74P@j zLq{8bTGo!Fd_+I4K6M95O{FmZ_Y!vgisH`fGN!-#PNoF1Ub-ipwY!rOU0Jh#2v6PX zfK@LIHv9a;UG)ceaG(!2Uo+&>+frOQZvvVoPKMQip|G@1;fX(+Y5QO@<^Fu4w@s&H z##v45d*6)3C+2Z)vkv#~cHwo0I;Y6X(()2il> z#pMSd(_}yv>+GYsYWE?wb_?R5v6`UR}{zL32X^teVc3lXO4&wUF5SQvMfCb?l8@3oGp z$_WgQ4WJqQd8}d<5B?s+=CMYcyHS+z4mr_Zv4+Qe<-}_z4>E0g1y_jWoGZx(XznKm z?-;@o3rW7ZPJ>4py_i0z2QMAb<)d(Yo-tcUEQq4N)C%^zc!j^@TDg8<4Hp&J(*w(y z(lM7==B2zS-^Lyiu9P-)4gY#)GJGRRwa=tvQZGfU{tp57L8 zlp4doSNF52b1N&}jg{yH&YZm2k{Va~P{rGZP9N8>Mm>=F();;gd@2=I$cu`L)x}y# zUf*!|e)jx&f~~8MaF~TUO_$7IRg5mn987us&`j>_ZO0}9PnO-7$$a$Y!l!a9eWOJ` z=&*2fF(3A5LXrw8RI*8l zXrUzSNu{L{Dk4QzB9Uw&qmW%+J4J}>-|P2B|8zR%bI$wI`!$~T-p9S!g;{1R@$kbk z^l@{C)zkejxN{OEH5HI+orpSHZxlTD$JoU4kkq=16E9Dpc1S&bp%i0>xW_Tr)!Iphi8`notcV?nr!nM8Jx)Gthx76M zP%pRuXZ=)Y3ogp2U(0c3av(gXr=s7ZYE)cohqlBuQh0TTwqKXW&E2;!zWN+ahSy`> zI}3OxMnmG#40PsA!<*rAAu(z$zCW);PJIVTSOYw_%5WQP;$r5*rv=SLYRzQ$Wkupp zU%}__ynwuqZ*b(^C+gia8=qx7F{ou32JFhjzHvSrYQ7wzW215uj248}r;NQIpssRTu-mN#zLKao-@h}rJ8rx$ATsm5h zkJ)zutHzzfZL`OyAD4qCUs92D?+kLQEXgssncaM6Mskm>=-T>zG_CRw{jIiQ6R#g+ zt9SJyvx}i@M)MeE=8?;2<2d0fnM{2`ifENWIa{JUfF-BhWaFxCF|BEJ7B(WAnlfZJ zYYbe)4xAoIaJtDl&Rk{(3bJXaS{##VjAKux>(F+#hXpzmvyH)abj7KH#kM=JD!Dc` zeZY36F?=-p+Uq^5|71X8LY>)omv8h?xtR?*R$$SmvD91`qL@~w40qnPUJ~RG^3~BdhQtp?{`rTVcIwCt4{-TN9-x5z3w7xQl zgEN@M?VhC7OO9l`QkipFAx&NQ%i^Ksc8mWMx3ino8Eo%>w=Dd?PP)^Pz?zqvu+Jl; z$?#PZGapjM&e*)5Bqvom{AEA;XXi?_Yx^*dm^&6rQ&QM~ahXiNVFv4yp+la1`qT27 zwQQr4HMQuaFcr_KtlXi5x!Sw4OfJDThGSd0E13N;_j&fl_RslQ*a zsRnzQOh_hksvJa5Cm&*-pW;}1xB;!6|DDwqxv|L6M&z8~KuviKY`vnZy%QH|=)$54mPRHl4(0$Vq606k8B#KwmH$4>25Bhyp1 zOro^K!vALgI}!ATt=HVejDIN5-czyUA+pL}cg4~%DO*IZ>7YClVXi?5)$s>JfV971kHev>C)3jxVrV$p;44v3e z^`mU;DoOS#Qj6;P`imL%8D>~lV9_}=%Ho9bM7E;hKXQuiX1*^ska7^wL2kunBuCM;2nY7y zthw-B4WPGws%gdgZ{+3imD)6QS?S9`%w>Z&8@#kP-CjPEv@f5ak*50eur-KU)gs7s z&Jgzdi4jZvIGd%E1+m}ClUe?YVsIz>@uJ4~v$|fC5GgN0cLrhtz zf+x$kT1tl(e<0W0ziFnFB~wfFVF4%InA^C%*t}mBhTnug%d;0|>Gj18O*PbO3Vk?0 z3RMZNP_W9tg9j%t&L9t2DaYy2ffkZ{D1r4m_OL^~9jwb#mn2hy*jec%Oe*RMJNq() z?F-Ff)ldA{R;~5yv0fTe2`pt7T&>XKzv+-(9)&GP{ZRq6x`C|umJ16U zTf=@S9B0~2y=nK6a0+`RNl%`SCY=J2`?$D+d`9%fM*|Ki-)b^8X`-R$Tgkzg(`)Cc zG|}b{HP17Eeu*F2q$=P%LmNG(t71^zLR>v+jn6r2aQoP140N_4lh+4CF6K7*{7z;4 zHYKwqq3z7@fhx|X+r$1q74Cf26xsO+ky(!M@X;<_(7_ z@i@Qjyzp2Ht?;PugIC1SKa&!wJ8+pgtRyMs@G{bzn?o*(y3wsK$J-@k`Q7%FxHBRi z6He_#_oQDex+#g4>kI#o#d7vylQX-MyPu8hp27B>@@GqJGFi?p4XhfY2xS*Z=#JdR z>_=6y@CIud{{1JLExne)Tbih8fCaN+Q(2tfT;^u3%ido1V-tSwVmp+EGuNpj*l@8v zmMrjKNH%rXd^rUl7(ljKTyyg>4uez?Lh_V^{jMk@4~GWKlO6?%!wPQDYoh zMLu$B-EPbgXNn8y<+ zD<`ujWpr=iKBiUK$)c=vDBHk`wY|tcv1@Ij2cYTRiB@sV%f_ z#tJwE`opv=6?1Dc$-#Opy?a(cQem0U>nw-Wr%F_QD4=1R<vx-0{HshMD-gCLUY1YUADm1BA?TKy_ z*C~`LA5Ct5YpFcY3$FE>kUC}yu5?M@exIJm)>p!~4nL%iaRtOVPw&BM+=`FJ%!6LI zZl{2wpA^uxT^1wPwUWz{C$u50lLAboanC~&ef)dF|6wS`9NvQ+hWXHZX8_w0IiW#( zpy}j`dtv#g&E5d%kRLSn#YcMPs)oy%(cszJaOzVT3O%z?tGWZ0dSf8-SAxI3Am+?v z_t7FhnajT5T&a&e-!^m>5)z}7w9Z4Ad27Jio(R@SkDE{Nde%OWiU~yU&=HGZv>tY|$#ksf0`nv=^ zzm~#MIZf<3ZjaBUi!sRA471we@RTLv^U*BSPCE>rza^O0vI?8Z79z1-Wb89yVf#aq zJ8jbDPtC0O?D#k6GvOAjt?Hnr`)cI4`niJI9zo+H(x< zwD#gtb{g|%*R=RW;!xm&@lbmy%EjF4cVjRd!sBsmgaIFvV#Gt^@51!+2fX*u;cb89 z_}2x#SQWYzWxF$x)6kdK#=nEA<3JvIKxBIqL}v2)a+G)6q`I+JY5h>au*s9dnrTCD z?XU36$&AJwX-^zEz5sRwKghQ18zubi1C7b!@$B4448Em@4#C=5VkOO67Hjg>8(A20 zYbQG2l|f=-0``nugQ6SjvEraSE_d|9>9|oin6jGR+i{r7>RNNp1%0?gp)CJnwv787 z2K>ST@%gDE+QQmMcm7jKHI?<$or4GIEWvd!wq`uJii)wHwyDR*p>!9D_jbn04 zVR?QsU(j<3{}}JiKTLl^Dw==jWsL@!U9KQqFcuSiOT~SB0zZFyBF_nJ!-kO2+RP)H< zVl_Q}FMK}3dgF)QR5&DXylp*zA9o9|Rd`Iox;Z+M`(bQ~3tp6%^4u9(JYQ)!PyRL; zYZb>}vFvcz%p1Yal}zDN`lRw;%>rbMJ`W469k|)=8;vTjrBw$Mk*}PMP|a=l>lh0j zUcEkHUwe%4cc}fyb-?NS8Fim!4x$=u3Ds&mS4hcW8jg1DZ2K z8L=9xprPag%bML#5clcvw_jq~FC}hMdKBMcO0X;VC3O2ggL8KktWCR-7;_OFtG6Ss z<_Zpuh(?N-TeuIH3ai5TIGM8m-yGfGCYG)3WBOz1udzsv61m8YnYcT5E6(?F$D-R& zDAbaI8SDg0Yz5Ze)D(Js9l~yP;Ozcb9C?<6wK|!oY_>zuqC-&1NyNA3voR)M8Y+G` zq4`}m{mP1kRDA@>>q1dGF&3&`v+(Gw$c#?W=Cf2~dDzV!oGti)><{-5DCSh&uXbVH z!x;SeXoi~~k6;3 zse@N>A^H&PE~g>wZa5Yc1!DGgL%z&NmR~+pFEYS}`0{oT?j3$lze;Xn-S%{>K39!# zYF|*#Pr{SViZw19`PFH-Teuw}(g%xaz414o2dqtWv1!eCG?etkq6xw?SR8>@yS>g8v&=Qdi=yW zO+NEX1OEGT5UZoJ1Y>n0x?gHT`OH9g?R_9vVa}-5ng#zpdN^^W3U;f++F5%XFD_1p zVq9OC4>7@3XR(KHZv;fo#d+ImWT{o*b3h_o8cv|gJp(;&EyiovM6}ljps^+hThdpe z>Fj2_+%ZIObE5e4A1-|RDqY^XMT<`%Bkttdipw9x`K&P>Ef2@xTdorpz62a@_ru_I zYapd#j)}1r=pokkr>A1@s?rbt9zKStLOvWmX5gt>9sc`#4p!djkn!|HPpb%&)=!4X z$29adPlVml7-YS3M*W0&P*saWLRJPMw-up6ryPTnC-PbOL%5N&A&;1Dhx%wooO))8 zhEhddV=u)I4R3?tlY?-=Ay~HNz~h2Gm%7-KKfffwHMjX=zI`;-ZJUp27DoKZF&UnA zPJ!G1xC8UW=kT+4DH=cIAVq5@?4~S6_>8Y~%=sdnO)McJ*#l^rcmh?QQt>bJ2mJ5V z;rh_C2rx>*(KLUwxH-XJ?GIE&*TO~d0z6knL*;@K)^w)ee}-@uPe=)>hI%i_yug~@t$e+xxyl5>}F%m|D!`17ii$+JnB-5CMk*0 zGG|elG1T$|aL|lW1iNBY6`IioPgIM&G`%M1%Y6 zfazVPA#zr)v>V7$&6|!$o}#-$^5{UWGCH^9k&E`+~7$MZd+l-zc#C5fwP!bt7{{e?PD5!&akF> zDN;UUOtRL-^f+@U`T8xPp&OFu%#t+PFx8q;!zR(?GrlysRf@}QlIGJkHQ~w87?Q29 zr`#MLnq>Ev4qBPvmgzo1LN^QCqC(e)>r#coZe|}G%S`KbGxzyYyjkfbiUz6h+}ic{ zF}5cTnYv;_Pi<0G&>0UU}ntz|YV%sjGsc#e*$Dscr91#aD6 zl}CQY*f@YmU8VZo9KL+Ce62?!%AzY*8dz+SRThdhM#5*1B|I` zeigG{`kE#7kl=q()cMbST6~V(aY~OoK%N&Os3dtcw4y?>cv?T)^9n#!v<}v1EuhT&UC7U@ByKae?kJQn_$hGo+UoxBg8Ix(8T`qO^L zcGmx85D$Np!^cUR^4U9|z-hz})G7AkO~)6JflDymYgM8qr&z%#u%~x++iA!oGg?1l zE6v!WMN+alNHTDN*BvF?y{b^>zZY9#yXk|~oykPpH*3;Ju+7u>zkTmAT z)1rmb>3ohowQB69;ie0zzBQRlMx18tMJJe&$qCkJ(37un8O~kz-o&3yHwrR0pnNuz zjDu?FT7eQRed$NFxAl40{jYc}&IaZmzu{4QBpm&6;C#1;{@zQa>j&pk)R*_Ly5W zB{)i(MnNU1n=FfMsU~kI@fzVlFn!K8J616%G54J`#)?ZW`ieh%JXh{_5dH0M;Y~pf zj_F8KVB;?K)#DJ$(Tc@ZFrP*0)Y z*9+-r)L(@H%)h17g|W42Tvu5Z@cN{+nrS4TT5Eio^*VD425mj zLMyISP;1!3cU?eS@}&>`cfn!3hj>qFT*BN(!1q!Q+n7W2fQJ zzWs`fC+;DQu~K*)bD#WQ4@aJYCXU2v!86VkrxdEN+dB-t`Bn7jt|KjfH;@usPSTZY zlK8S&WV$Do(|6(1>?vJN-*hslrY4AF78TMS@8x8y;YGW@j-byP3vt(Q60T%#hSDrM zinVAbhcYL+=r86eef#6A>1s5^T2awnkxM$MBfg7ERK4mMIjm5}F0VuAe!Ue3x9&#N z?+|id(gP{eRIx@jf(H0arn%FcC`bJT8*9+;eK(r9|<=dQ;D2DQfJsOmH6rUvq5~-9EmJI>uI$_2Li;yE=_zN3Rxntaxf{ z*P^4329f9QP#V0eFE-kCk=-UwbZnM}V&0;|Q(J?)dsuB1@5m(0dorO6BaqpD|*=!Dt@^f>wkMhhR}-Kbnza{H**uS=*x zLl$f7Mj|_@C$@hYi?F&c^jA3?^Tw&;i+mZ3Ju}d3c8QMM&!+|Znke9tEMhO8!lu60 z@vBV{5wRv{ud;>BxqFl}Xg^sclu=??FY=C+5__mRT|JUSYiG4kcwQ}?es_dAT_o@z zyorJ;x6o;)y_B@Cl2&I44g2Q?G|Wl`YS&Wq!k^UQ(MF-+*@BsV8Jl0tL`lOzig24x z&z82av}Kz}wzYtM3bx7EQP3t)fC6Ya$QY2U! zA1{+Z>n(CS6Himl7E^WA2|5z8lR7M}(Tk^->0|dB3hl858y40PYrIQCk3OYe-%r3J z;vw~RE2Yp1O>DMxra^J3tjbHCjM}5<#4Jx5eDVg}`P4uwUv8(2-5qq$KoUJ1Y%tJc z4t9EXQt1sFY>F9)A?g8i1;ePIRTI5d1R=LO5)OHJIJVvk^X`dW1IZ{H>QYCdnjStc z(?#|+IozB6md;HdjYXwl2(X7BSRsKf`g5P(CtHHLtkfiS% zpjUR!>BP`7{F-|ad5UEi6TBJz*&^q#HybmqpCC(_axxM;NT;H32w zCg``~FPXa%TIOG%cA;U&{pzNoeZ8<~?NIFY>!c(0DQNFkfb-wQ9M)ar2ZaCX^W=W0 z?kV_%&FL6^A`WfCQV@A17bkD$AV56^+s9`@NAojGrrt%!?I-YSxq|nC18Oj1C|KGU zw57ztG;9m}rmn}nKi_e0`F}9JZpG7*6)?VFIF&)7!P)AI#9pV< zl#t!jy7envy84~6uZev5-PKs4l8!y8QSkHYiTnHXQT#{`;kUbKV7TB<%1(t}vI^!; zlf}A8R`@$s5vi-Bv1HLiR4$cAQu2GUGBv<|B3pf2< z{Ba7wq=IxD+545&9dD&uhX%pMcp6Ol7$MVWDkhN@>U~F`n~g-3#xu%Y@`bLv><`P! zTcIkEkFHULA{$hJ@CR9tT33v-+jSu)<0f*KE(o6Pg!4}VAUSgl78hIMSbiWDcE;e? zpaNK%rNT&L8?27jiT;T@c&zab9cN8Z5j+Wp-y386$1RvKeG4pQu4C%RvAD>M@p{}s z=+uT`T)-9#-cXLmzOK-@9|H@u8SuPz=pSx5O6EqU}(6Q3A=?I&hqa0whzZ{Vd~B^D@JL2>UYXe;JpXW~2hedG;|&-qOKgm0-< z+{u?UMImbOb;^joODhij7P-wrl$Pv6y8avSJm;dHa|k5bL^f^F79<``!^*##@mu{m zPA>fqCvU#Rw)w)Fqz2?$#G*XF68=G%hl*KIEN;y@f+X^-^>w&1N}BIY+~ zqN9<+x?&{m?$^U;6KzChCt`!;D7e`MBB3b|-Y$nR?ph-T&5+|dpXOotU!kkqmE${) z3%_={H7*Zxh3p+WRHbdhjwQ!X@%1~sf1-qlHFKeK{RKq~aE9y3@i6o1kMFlk(7kv9 z-VEJ}6%X>UWOfv6mxy`H!FlMJF-zP7&!Njzfmb?6^8K?K#GcX{x!St8b|D_!HAj$h z;{k@-W+N^1EHundAil%_E*BgCr2xE?mc=PEZNxgZ)03KubY}5i+V)x(njVIDv_oWK zr>4Wp<^;ATH6b*`fQLFOAh~ccv=tUYqF*s)ecJ<%Rt26F_ZkcP_2kz+YVpZ2Z_ws1 z$=`n-#Q)6gfM>5RgceNXtMt6kH*gZ(1qkL!+ZecBO2D;Mm3U^=h10gmyg1Q|&l28U z$%G}yI}#)`wF9{Aat>uonVW{K15=KM>ihpNdiH6ADfHllr%d^tHH$IqXDlYS2`=ZG zOWA&t6vJSK~RnXcStQUkBASiyWNLW!CBHB5U^~qzN6WZ|pue*yg|| z)e=gF28(ynRdlI1;naQ~e5u+adU$dXP@D;Dy@53*kvJn*yGu&z@Uj0)bgoi|m@q-L z(HVV1*WzbL4m$D&^8D-m{G9Z9?o?Zk5k*fS{Y#p!?bRDaSEj*z$qL-|F~;@bHpp@q zk5NTE#5y$)K2t0pbIt%tMjmK1PeRY9{s=yD5Y=X-NZV6}VX0;Kx>}ki&+CBo3QXg8m`z%PzQlIfh@BYW!4Reav?W$3FRdxWv`b{PZUJwO1bvW@SkB@dOuH@4Yos z@aW|%)f>3K~w-|vGL|7UTy)g$clO)TDfADOPB@uO@J_D8zmta&xkES}<6k}hBL z+Y!U`BVjwF1TRBB(9-id*q9Ik09%KN`8!IubXKyIE`HOCn`IwtL89C>GvscD}y{^E==shSO z`2$zHJP;=rg@Gi?V{K~@KKMDFJ1X;sDpScuFh3-QhSSAW&2++`54<({V$PIG%3}{G zvE~EK|8SF>3xqE8@eDcmf2Zcx>acjPjG+4KwEVwnnmq6tUD`g6V(&OpT8t0P7Yw9( zMJvg}dMAyX^oq*L+vwia!FYc6FGYs@q-QhN;_xfMRhXDf(LY|&>l_;d?XiaY&Usky zZ~?yNh5;@I@ndofRTQMrUC##UVPe3&-pKJ69k=k_K?A)lb%ov$h5Od4N#)p4S}Svr zx?Kxsol7}YUazIgKbk0bSp#WH?xTrn^%>0@?b#HmgNhM2Y}nJ}4RO9l5Q(9Pr>BroTP7W*9Gc|3{3zQcuXG8U`O6+vCb zvALsZa-lYjzCMvY{Ol#}vW7J3+B*8~=tCp!_^fyqi@I=~OlS@V$4f)X8 z+5G+Fy}YII2>in@qkk`Hu2r8&Gi(l1_!R|MojO5hyLM94r#c$fb%A;{UL#&zLf14* zVau$b7ZZh{8by>F5=B2({2>kPdh>bU)G8wCWQUA6Mn+7mpDMkH^`l7FzRE8bi+X!L1*Q zQ0gnmRX-|n$#OL=mwyMRt2$6KL!GytuA^(iKGDk2KS{ac5e^-!$IGd&a93^s-L5gB zr{A}bg!~5ded7mH;R)4T@P>`nyu}8;>_e(kFVT~vBh)ro3i^?==u}|>y?c9=b`GCM zSA9K1o+p+DyxoiSciOR~{v~=0I89D#T4-aB-dLZi$bU(Ffzw)o$@v+p17ZM=ia3Q_ANAf+C55m-vb*yiEPF$h9MGbK(1+IH9v*C=66wEfIHo6 z+f7I2OJVblYp{EC5CDLm*d8DxsRVU7mY{27cT9;G;4 z9DoyTGohg64Yj%h==yd5K8NnY>zEddc8h&FPY;US`!Mv=8K`}E47KiJ6d6>&BU6ee z=X>H@P&yvpJCBLgt9Z|qp*-w+5`PjCi*G6iF>><}Yz%S45vNe$D=vb%L?8S%7MX*p zK^QC;YrCwpF+}+EXS|U|Nq`JaebdAM>s~y^YXGk{Rp3wb)gX0YvgnF)fu2)uUe!yH z+npQFr^Pf==CI$?>$)tiOjX6p7NI-FTSD{Aa9*A^kC*tk^7JX$C_8or$zM-nWz`>& zpV1dbhpED?$7=c`@>`>%TWR{5yHq*P9J4*+Q8#=5=AIh_+T9;d4(&z*iZL_m0miHt zh_}k7(8&tGnax7Wb+hFAoh`Y`=4NClD{$8gTYkdCoo8z}@k6Km`Gb_Ds9&Cp6UX*r zd|C`_Z>>k*wrHH#SqeRym(W(4%qtvaaJqRJ{AImi@>b}zo}-cLq=i#%7W}~_L#}vz zHIGWV1|IYSjEuQ|R|dS7q$BzMbJR;I@^7Dw`L14L`H=>7bnfY(S;MU`ee`uwo7hD& z;tg>8;0GF3@rd|Qkt=i5L8>_8*!EotN4po~U-*b>UQ1xwlolHC-U2}*qo6F)ftR}a ze8=>j+~iUvE-k)+g7K9I3Cc!RNFnB2$wb4TZu+{g2kusCVbs-po|cioWRFeBcx!wqW(y1+U-42DJD>C^tfc)O(sR{LmSW8NTKE9s9(Kg#gBW-oq3|Ac&r z2R{FcgnQ^={CYGA4IUFP^5HV{HxYglcMaSKFv88*>+m)rM)W|e#kBY4@Eb0TCn2+O z#(5|niL=FN`|Wt*HHAMpw1AHqy%(de+ zUA|l)#heHAv*9N$Me}~%u3&?vp@&Q~4Brjl^B+6#DLdBj(+7r#zBpB(J-MUscs6Pz z3ScAAjM$l?_ zcG{sU54kcG1RvhQx9rU4!8Qea^|A)o7)bJWu{ZJ5Sg;B|&c@L_8{wLB7aBg#ag6;% z%@ZkbW06UFWhL%1HLxr?i4FJO;=AQJJm?nP1JA6vRI)SXHbz6QIR)F=4RLbrI*c-i z#!bm^s2T!n!zR?##G)ZL9>2u-G|ONu!WFk+;qwg0kKB$=`pNisuoSADqF-V|HHOy- zFHn0q9lu>n{(prRbblP>)K<^}&#`20IhvOH44^&NikPyh5~)Y`6a8IQY^|pReV=-j zf)|V-k4;Bd+%7T06l~l*4x=edFq)B4%&fY1(A5K4sC^Sjuex@S!Lw%4suFvx(pc&} z&4JdJM=>SgV}3rwiP9RsFu&Rn^sf3TbvQ{=M@|8qPm8Czu(>3C_9RVwvI3c>^XaLK z1dawC#Z$vUWHxsp)%>178-7)=9hVl7{+ZVlDZ0DD(j-ZBTqH$y&Z4;e?^(mdWU6|4 ziiHK=XAhPHk(|rte&1m!Fn zNVnrMSb)wAQZ|yu^eR_+e9s8O)dmPZ%2;xio=Z{5*|g@t5mtHprr$>mcD33Qc-aLi)}tm6>bM;>wXg& zC4Y+ctX0SF5r=7=-vL@&Ch}|T$~3IZ48==|G2+NBvitUctukCij~=PhFTn_UWPg+B zW+I&*BN(*WrKoDCpeB(Ss9az{|J~1IYFF$i?aN#8IBZQnh4$~QwGftzR77u96@6Rj zi{IzP+*Yu?Po}5RkI)QiQ8`726i9Gy5=fyXtiJs>wYt!yIF~>^Y}@o zq$QO0RZJwmdy=UCCVbZ) zf3TP8zBDv@TFONt!J>olDEWQ}v3(&(%Fi{8n5 z(aY@N6cT1kc{011eD-B#+_;?NyZqS+b4l@xy~$U$56w|@qV&Q0Xv?Qv?0UvqDiGi6 zJ+E2xTtAL9*OXHI4@X*5Jd>iI9i|=9I<#w|FZHteLPtx}X@l|*S}`SyT5WVFvV9gE zFB0?OtjUVd+f$0Q?uk*4GfCfeH_cEUL6a{m zVd_t3vWj(2*zQO9Z05qr)N|-fiq~Uwc3vS_-K?R)B?e@!zlO#nPo*4@^_bBi_Km6I zfX`#$TW?M)Yp0X6>;ks_tu!3Fg7I=jPwMX;Oztw9sn01x(#~Qud{zS4Z68@F@5iJu>>if}2;7x{3li&YM$_$QpD>^uWXclPL4y7Ruc-pLFX!Gym7W zSn!ib@;l^7J!3%fMdxYycEPUJl|`QhdEB~YNiPZ_Df_?+;!)l-;Ef7R3K3ki)lb+k zh2@mL{yN?Q)A|03N=-(StI;kxHvTnz8WK$F=IYb&osQJ> zRDRl0M?D8HOO3a;{r_DOVP;Usbvy3dk56G-;) zOsWbnr=9+SnQD zPQPc+9MQ{{kl;)j?L#rgEd?3Xg3qTdNk@Zp=}6cB()3@0J!>Z8Qb;m}9GF9uHu{2> z=}v{)yGdq}0~Wtug>b*2IHWsO%(+*CJ48}|-*wWy_Kx~)bEN%zEUodGPftptN$r*| z{g`o^EXQu4Mw1)#$*P;a_l(3z?|z`f5~!VQrN>+D(^?$~(P23b?!V1p`M?Y9W7`GC zb(xsA_#;8f63zz(V$Fs`=&mzE^CDL`WksNAO@DeEkWHIC^65dTCY>*Jr^Dj{Xu?f@ zypr`5p6b&u*Gi`d!HX~VdrJ-7rD%@&kD{wN?H+CK(8DgwM)zt%n$gmiB1VSMtrK>H0VkFKh9*Z8uNu4n6G z>B?vt=Xj1V=_yTmDT9g$t#sv+U|Z^_a0|Uo*hTc_62XJuDVYW9(sb-^OC#y3G`c+O z6tOTLdMVhqxt9{C6|a%M3}Jy}XSFC&!!CV`yD$xuu?FSZW~R*y*u$ZxFQ~%|D@pXOBIVEhv7?yAxZj1lk;;Qno>2H zKDe0(=EDJ6+I5dQPu0;})4`ackw;2`;VNzTh^oRPaKR`Wedm2U8gv4 zX1pQJCI{sYLOtO#ozgo)^QNl7ZF3vTHC875|J11I$|M9mnS}>ygJ8mzVcwycFk9z} zUwP-~@h1yR2{pxvbt)LiByk~}u+vCN@UqTQoKA0Cx?Vt?1xE>!pVOS!n-o6bBgqT> z=&OM>cBC(a$%A6d8)!z^^F)V1QXc&h?Btr!sbuLccq~g_~I4_HOI}ER=5qvI?54m@&*M)|KW$HFJ^67 zA(%6M*de_Rl|nnawRt&$&)OiCFGJu^R}2|61p^}1qhm=aq*Z0`Y4l|1&R#4SOV&8C zzy{~?yfAvs9t`a|fWPO?VN+K)F0K`*|#BBssMd1>%k~eu-kerN40b% zCe7T0(A+Gj>L%kI-wPSpVhr*X_rqp={(Y+fpY+#^zp&Nhd;N_0#oZBDFPDV{N`+Xx zTJV4lAH=h`b{J34NBkd0yv_8$w~wZH)*&pJ74J`@07x!wN9iE*}?ERQTDLfn55XJHORu6MA(=KzX4* z)*P)ymEvVwQB~o+8+&mtw_$wYs8KxV@C&rxm*W=~4d)4GcVL%)9)gb@1TVaRJx>*Q z$$#p6d5d64ENVjLy;~@dlSA|qHIW|>J@nxWo|7jc=%_n-v}D3O<`N3@E1_X;h2NLv zVeRtS80uq=U((i?e=-E;+pps9BMF}8rO2B%#K2Hw=KcPN9QJ|z==ZY_cXD>)P0~92 z=M@FBltSFHTf()5rSg8GxA7O3CApq~9?#RZ;>V@z(VDda4-3OFVd*r%hnN7}bK$6O zHH2>EIOrF+2^}dIFFeyRyR!g0*9PE5RXLJ;9ztq=C4x6T!p81j=qa)@W0(8ltJ)ge zj|dlYh*>BY(FDhLHaI1U>n@bf!RCm=a9=BW(1jm#U&s=kzm@R=dtCXVfaxfiISG&T ztP#BGAcl*(lh5q0u)6pg^Bnqd>0Q0~yi#qxR6~QS$_(O{GQHuEvKPU(sxd{}iT^y- zVEewRA7$Gwq8QiASHJapXn;+IBYJ_?v9# zC+r5Ryo1{JF1RWDvf*q2&YT&@ZO7{KkP$|Fcg=XHSI@!TC!5j#Cir2KVSMu&H*OG< zfCIu;U7>OodI6e<%AJCVR1Kg@n;^F(1dD$L0 zKCj;fcy{fPRe`ZWO;fBWEb$a*9x$wEtVJtjX+MZ@3-oEQ_2=VQ~5-baCdDelFO zR2lH<++!$waUEWv50NE(9rhD1;@H(LgdJDLw2(Hs6{!H<7U8q|JQ-D~1eN@ir1MS8 zKxbsp8~u-zb62pYCrHBEY8cG&?GYw93O!3Bq59n$Z(ZVXB<3XP{V1ek>(VKEOe&qv z*g%W+Mp2!*0^;_LL-3IP0#k8?75|c?8y0*3_Jqu zX^Q0#3U@W2e|>J!i2e=Kb7eB=dtQas388s>xs0;o|5($;5%lSj7pa!pVwI~2jAjw? zGiJlpB?JRsEJl%46h6y^U|D4_B-YG^kBc2Xsn5j@D-XOLd>Bzb55YQhB(4Q&V_2jW z+E)yrRNIN9|2v)*J4e&S(`(4`%^v!-J(Uzrr_$f0iPXQxVVLLMgZmQ+K6UpN+WW2Fx2ov1T&Ra@2!KRRirkJQj6d zZ1DZ8FRoY0AjiTOcU#0da4wM2ExwAYalNw9T<`T> z+P%Drmh5PxVl~lmI@$);y`5mcQUT37`{K3ec=G+HLQ!WdX;Nt)NO_9C>jsre zUl3hLL_RY&iyWE_wTWEo_naWQ^l%du%Js*@#acKVqlhNsKG3r52N{F@&=UPTYkM2e zw6jWdc(^~M_Bui>f}7~RNsHVgA}PvnB|RAYgRU6##lId>cq6pNzzf!RC*z2*dLbw~ zya~@VHzG>#neNV-fFhR^M7V2EhP(;ozUxnY<^Lf zac^G=U6e?Uwi9t&(+-2OR-=E%U_^%qe%rwQ7*^>G>6>x*y3GeKL{9b6g#M^n&M`Q| z8|&hQ_hy*jbq*ej_8Eqldt@LSf+TrFZ)yH(ln%dYIS22{5+T1Q2|K<=P}z~yRM|S5 z;yOil-4c6@Qyz(7pFh!t?QxWKtB`(|4Mkj*1#jr8?=<<$IF8{(en4 zj<4wB+#uYM_r#-;SV&fUV-qSAsXkhUR+W3BM9C6zV@*JV=h6vBHJW(IMLfqbG^KJp zrAbbv;aQDPO_1g4esAzCPn}-Pkf*2G>h#|WW%wNZF7EotShMmFt(g3fhVD5=3ZtKr zwW~Cqo!7>^MFo_Waf!YSZX(T8XWs8!7!OZa!>8<%M|PN)FN_Tlosp&pS*wWgWlHci z>l9oq8T{ICpK_uXV!oRWSHdl?L%zIIz`8Mj@aZt=^_E3-4V{!R+f4KeA zL)$|)YHxF<3kN+(cbGBytkpzAvY0D-^?;YiB@8f~1bfGS^m)=Hx-zPe*w}0;cHKcc zhZWL0!Ljf=C&R06_vW=}JO9Vgc?RV8zHhvuA&LeHm4=3fS@(Hd4G~JEC1h2EGP24_ zwD%IFA(gfy5}|}jWbZOkMo3m)G7|Y+|Nl!byzsc6=f1A#s>)j-@te8)ZnlXZImNb#)Pn!5ggoN!Pq&#C1y~@opnZ(jeat;*;gQvxlbh|6P zxpRTKv{D7n1wF#u|6FN*wU5wiC?$j#B?$K~)(R>$D)htJnY?l)(6qRZq$*!QjXMp6 z$1+b%N?#rn3gR8fq;V$=JNJk_@cY|J)3rnrvGi0sf@bjZrl0L^+EsB=Fx~xzV)9GL zNn;2yC?11zBH;qURY}8FSH!&6axPF3Zh-QLiUnwVb(=4I?lVe z(F@*^z3Otx+-giqgH349cN@X(_Z8vv{1d{`y`O~7YRl=_l?&w3``KiJuAZ=}){H(U zhzW76p+fhb;q*0JTlh8aoiJb(|9)Q_MkD49;XU%*bRqLNt+^b+9cr(5zS2m*vFXAI zukqv*dBSAMiAlmR7fl)^%{!-=hlO?v?(JS_EG%jmD?DxU6n^}PppnbAkb0RYj>aVm z1%G}DGk*<3OmC}+M9cu;$NF61Q@0Ui5_|z-YF=$T%d!^B8aM(Pfz6IsI9Mvte1Tgbhd4#9aBOnJhGL<+VW`m`*-9y zH(Bsn+eboLC*2k~&GkEx!YOw#8v4_YPPiPR@R^Ym{JP&n@}H-0#ptJC_$)}sG#*cv z3fIt_zB*yY=o7+IBHDC+FMn^U;+X3@60a^H z%}>{H+k6ZBa@XT#{YXf}h~e?`A&~GNh2m0Kp3mcFL!G@CVG<3Ktp$ktPY3I#IpW%V z1ALnAiJHG=;L%-prSkqKcM-de6>z#bg(ND2Xqo8>Qqtc6DeF*-*mD)JJeQ@hISQ@A zJo)|V1MRSHB)RAE7~m9*-aA`(hI9rDu5YF7&*xFUgF0!>8U$(6P8vV5fhJjKAw*dg zx;y#1I46#KO4jghLOAX3SHKsY`_zzCNgkVb;gEU`7Cb+I`9IQ0sbUiyJjkfU-WNL7 zNhlOOh0SYDpyknN&QS=&SuH-p`lg6OWy5hOGm=`r$C13VHznV?L8tEAp+!}(6!S6{ zc9FbyX38DL62I_uP6?K}?t`%zzyFqeAy1Jql6trR1+SuDEO`P)V%CxGz93q$)qy5| z*ozfgPUFArYk3E11>UbvN2>8~%zTlBQTsW6ZJ;S8a)w~hG#gxbwH6awa`|uE2~ulG zqnjO<==R{JbibpW{1wxYs&Eij62dUuQ3H!t6V7lBqHm@YO6!B@-uq12B(?q(i^EwdYHrk1!+$Me`?Q<+DJID2vZ6P_e-*4ib1 zG+NDp|4dnY$s3C3m9J>qUJG=un2QxPoO|_T0a8m%a6{Ds3m!<}A_IxGa(n8y|N#M|nZihtHGzPe707e3VX?AoH*{`V>52*6jldxrKv; z&r!~^vJ!bA^meQ(-MF`%9?BYFse?7_=h@)U?j$(OlZJumKk{u~i8Wh7u-Qu+ySYR9 znRozBU*wq^(-=$@(s4Q|1-5gO@M2;Ru05WJhdE)mE1QZRmmSgL#&`FFr;<14ft*&^ zPu?FQ==O_PQvDQ<&TCg8m0X9foX*zlr-frlYRHd0M>@0LQ6P7A{frN%g(u_a)$g^m zC%2ks-%pT6!gA6uXn>elIvm{VQ5?s=C&tgHeby5?v%(xb={AtFbiz-mWLRyS3z>Hr za2``fr_IZ$On*Jinp019E;lLS+D%#$#Cx)jLs7IU7c-h#Xx;iIDrnE808cHe)g@uc=%eTnFM&eHC?p@?x#_k$l)QB-R^5q# zV$n(TjLM|xo%Zx%ks0}%uY_Z06;_?!hdJ9_puIm3V{InG+`Wd3ca%}OMk$3~%7U4B zE}m82!0vVXQLecGQ)hAmm-R#XBCdp>*5R1-me1*w2lI@Yfal|RCw4D)<}G)C+Iuy~ zw~WWYN>S8pGJvziXy~O3gqh2JRPRXVe86}(we!yNxnP{pa^?<}sg#nriK_X2$gXJu z29?X=hQ2H&XWXD7zZ*1PzL<_VOrrcx1~h!?55c?T9)^{1z9n}JA38$#S*8PnnHBVW zlLbjUm!b1dw5T{T1z7>naLFh^Qwqc48Dj9{Io_;D1$;iFi9@ABP+B96wSR|0Jo+zf zG_0mUuanT{}nT`{)5AEO(eJt z#Z^A{s+Vv4eYTOnA<6wL&Xx&zlH^JIegI*F(ncZ0Nq_+03mMF>5qI zRaXf^d8W|ml{`G+c_&j%9`8*3(vxswa$j>Uljxu<#HZl0WOv2Ai(%Cl-Bb-kLf;GJQ|0u_XlD@ z80RwjEk;($RJ3bq<3#rfS}MATCU%z6Ah%C+lh5m;WchCV@&c@)Fo>>QiA7G*sIyT32wFwE1WJYC=?<8BqmZnXVNIG;=F0@ z$SA)WiBy$+;NgC{WOteJyEl`0ng&+Sw?gp&0rjKD;BP?_y}Wdbf^P7x{n--;b}Yv2 zuqZ^OA3)2C7`$41lwuy`(5)U{vV1lh#R{|WMR_XvP4Ccf_baq7sFwFd-obqIF}NOW z#I^cu7{zyX6&5+Tt9cYx;;!@TS`PmCi!rM?FHvh(fXln}k>5NMXMc%e;r2*~EI5vd z@xD0N+(jxWx_D&$nJ%}!rSM;7@a^)&+JotoyS#)>aW~1laVzk5s4e`~@%&bh3e)RW zXX`w^U@6~!RW*-5Z`%m`CsPCO7nitmpa@R6+0^W_iTcBC)9qXZJlVth2n(j*qs&MQ zKED`OR`bqDyC%E)NQDX3K5XT{5`?yu;DY&ihT)#w+ zPW2GBWw|QL===ju-aQ&MuY*+X{iL3|HMkQrhAq`>$Jui$@Ne2)>~zmX!!|R{?$t)z zstNpiAJ6a5S7^xeQ)DpT9p!_=vDRZIMpWG69D@o(hE&63bPEy}y+-=8UNouQrDDD# z-m>HYeVe!+L+T&!99AuUMVPTInX}pA1KR9Xv?E)fyO2E?H;&yo=fij4ymPa*3W2$) zh_6q>j>iQkVE3n-#5c|)mqAI_YTzMDfQS(Wp z2Ia%KI~U(Z<$_wHuy$bqCeBWQD(3+yZoWq~`+JGyisMc5GW^h;h70aisM0oPw_J_c zYuRxuEVc$&RqNsY;RuSP)o^{NJvO{F$D+C#8u&*R7h9|7PDUAb)>Ko|%yfFTY#)ZX z^UO~|G@2XSP!@d+GtX7vquWE&o#=<6RtrQqi@D$R5q;vn37a*?VbK>|Xdg7gevK-U z8qrDrNC$b%&!9Nx72hLe!EwTQEWFTxCp(&O_)3U*yMHt+_PS&n1@SLE5F1O2F6 zQ%6~14~?H|1ig#D$c^vpMqM6+=g;_e)_OSeKPk=beN|#lyhpH!W%?}5cr5dA z9*fP&)8Sk`8>=_|fYh1)aM`sI`IDaUS=|V@a85(U$#AsabA|6PJ6x)rhUEh}fB2;Y z?&3DdpB6*X%VF4G7E8ZbHjUnSh9Vx!z^Nfx=uiVHcvt-Tss^aN{f%aEQJgHf&O3D- zB>TV!(+~I}iRTzARThy{V?2F4A52+ZJS)b3D?^nG@w@LKHPwsaXNCd1)?2XOj~B7R z2z^$oEXk_=i8BqwYV<$Ug8!DWxY;FvW#gqFvP>E=4>{{~YBqfySw@bvm9+h4HN~DO zr{~st>HLhnB+L({dru>%B=S9i{r4ik`x(@h7tonyX>`Iki=0;dMQX1$dv{Nf?U+$W z8y{Y$vxj*Xu&4$-r|U2#?kIvaq*&z%Wmfb@hB?m(#g}nY5bC@b$CvM==}Pg`pPND1 z-*@7@k~-wx0;)NM@Kh?t=de>KIbDyuQPns(@+tQHlHgfyPi$0i!(Yw{5!2y47^kzC ze$a?**O6z!FDYixI)cfXGIl@3h^0Lm$7t(Jc0GO^lkrMN;H_ZzpWOrXkFj*zZ$AYd zE+*Qa$TqJ{VG#%NnAP61xW2Lo@yl|cl=TAD6@?hG;5mAFU+_D%9DAU}v!WdytiITa zDaov2(!Q#gnl~29U3ho!mm-_o{TXL0I;@HPg8Jm2(7QGOJ9x*|VNfY;OyJ!A z2sONOWAIv7h#B%1;eK&D@^$aii#LdF83_MF1gorM6yvgJH=9>-pXl}^XI=ir8^TgSr)JD$N{)w6- zTiE$6#ej`L7|#w+?yAMq|NcG&oY_oS@6S;GjS$+g-yHA0nWNli2=AtzCKJwjTB3HH zj*4X=KA;e)ayb~VBZ(sAYUserWX{OTqX&LPwDL{~y;*aBQr@N0ri}zoK$heU~GaLW3neHy(fpdnhNOQ=WXauod$iLF%A9QP8XiN zrUl8vu>O!1WC}l1V9FAtpR!_p@>5tz$TU`zGLc>RVaUcP7_ujAK+~le=-amzssFWe z_cHInByTx6yO^C-}bVHqs>vkT81~Yc7>!H_n$} zDF1B_`1_6i7QLrS%hO>I<<2{+gt+gc;94t&yV32meAHOXy!VBe_fxVeZzPSQC+U@B z7rCWA!)}#Q82mg8cNKot?plZE7Zee+wH+~svLREYi2r7chV_P4inP8?EoHpxJYfm0 z{dhnh#Q(#3p3hHRbr?f0gyYJ_Wk}(l}b!S<$&Q74$Klai_C7y5a zmO|Llu`s&v3UUw2@P5E9oSPGYdlFXgX>viYR3-knCL^VFFOt`{@%L^8l-|a|RN(?j zH233f?=-Bj-wVI@<@obM3%04j&=#MFDw>RrH~U~)?GLGKI(Yt)u?Ih&!~LHao7*=R zgHPEY+N+idGq<7p^=>xZB2-&3d3!O|Ylh(Nn6Ws!M+di> zIQA$D{WLb+qm^ZpWS@Dza<-*>n_E1(uhQ>Yg-Q0kT&w8`WcK5?(#>DSA# zE53)!vab=f*N}aW8YFv$U`T}`?=GH!{kszEWgh4XaK^6t6R?xo>B&ll*2-}hv-$=Z zc<@;x&*Bab7yy~kQ`wp&;%rFABQ(`Z@w47p?t{3FdHr^1br=ox;&QsbTOA>+oz~46 zh%e8@_zqYca=R~4InT0H^FGapvnT1uqZ`=ow+$J80!mFY^aNn*4x#o!BeEXEt`7(2LuPd%QPiGa?*2qjuqXt1`N3f1><4 z?{U|9W8cWrII?6h4$fSTKe^VJ+bD|g5*6gx{705Q3aR8`DV^nA>EHSL5HMgMB&r+u zUNZ=lRVwJd=Ypa=#@KVIfx2$J#gb{?;X1kq%0;?(Gm7u=D(XqDejpq2x)#9__hA*y zyGg~nQI=zegKI=FU2ilttyIFs;R(=u5sDW_l@WW`hP{3{iOo(hWD<9#Rs{E|x!7RLvvL>y;N|Nw_`KbP zjvY~`$8YjaG{f6FV{ouR3D15Eg_L_Vo^_0Y{U$#QePe{f+g{S(z29k3#{g#8`5eK# zE9~J?jn6S<7^5VI_eZDT^$y+-+SNpNjt^nqg1#c;_*aanyM=_&_2`?LgHZ0>3)cw2 zOTEKT4YbG4V_q1&%LiA_^NhJ=8eW_&!thBM{EoK{X^y+GaKK808gAr_vsGy4y$-B> zf`7Uq+#SbrRElNXRdodaHgCX^4NjTCwaY=Im#^6$|(5hko`~jQc#49pf`y(Ovo&=r;#GPWK?b_71k_{Rgf7 zK%C^Ch10B(l@6XwJrtqK4xh|&GDB*oP7wITi+MVa$4Ov$AK!W*~et>1beCRqXN1dAw zPHlVz#jKx5)%b*oUOay_^fwk?8OY8LkYc?bhOtkfKQKgJli7UMU>g7QSn`N12ndPB z?{A^Ni92Y#_7w?6ZIE!>$PRz=WykVuS!q9ai#P0r6YsjmAC_TW)Q|nghqK5V*E!Sl zHF9phhvrCa7CHlLZ-fea_{0}&4^nX6*d3K)pJJ93=dYXd z9?H+|^57Y??9&L5d{96^tp?(h(i}K=+P!%bs&xp-44a4qRh$j zFN&21GN0j=$bRXJO?Im}*LwhKSL@-KG)cDECjuLeB|*MC8VmE*qd|NpzkelSnc*;| z$(@uE<3!lDcY9gk`2?o8D2z=xp91ZTu{h$%bH+xEh+p#n&XOW5wr4Jq7uz7Vnmb%B zl;C$%G%k-ggfSaRF*~K6djTu3cgHyP_OcvXoNCKbMXsVS{~~_t+=*NIKnIoUW+T6SG526E;W^1&cq{uHQC>~l+x8hp=5E3dt!OBj$6|AG8nOfP zV3u|m2TaDGdD~Qku9}X#k}WvYyaSdV+@ZZ>1rBtjAa_j=n)H>}imm!=?RpayIP?Z| z_3zSm&!YtoJIwy~Yc;)O;F$4`d@qbw9*fgJEer0#g+^_m*eb3>LfKZ?8SB{$7H{ zS*dv6upb$3Q=xsc7_T3PDbgI7T6dTAkFt0oK5Djr4P5T zrfXg-q&E@N6NZd|`DhrQ4P#@@r)jB#<&8XC-E|7*e>}!7XtRdI$?Rl2R~?L6jJbaf z!svSt(xoMt&k03#&3+`SU0n^s_vaAka}LH9YRI-!#A)4Gu$r+9$J#ex9nU>w-pa(! zg(vV}*goVLI^cNrYRrAT2ZsI0+}kXp`5O@aBSJgjKT<^3=( zB-<^;Uw#*uwB7~(U3bNS(}C!!n9M3m*RVqszN}{KUku~?O35!GY|iU#*p`_9$1QPu z-!lt}a-0*g&K@Ux_0i^`i7{`NV5`nP{65M5Z&sy2>-J&Jx-CG@>?_E!3&--}Zg<*ik0Rmhkt%nnY2iFH??TFKSVG;4XIg4rAN=4Oy4AGF$dL9Nl5@ zaBz*sSpMh14>eh-2lo&Zfshe`E8+2IsBl82{AMhv^Tc{jU{1OTI{6IYeb;^5Z0Uex^8@TG z9?vT6)!5tPliBYTQT$$V2=(Jm!@_zAmMd>X4)2z1s9J)o-H}*R=>@04G4Nbxh=M>P z=sS;Rmz4|{HJh@1Y#?*nCBnW;8ps|uTCt_Bb6LbY8z#^BX=dvVk$3N3svkKTQd{=H zv2iP6n-}wQdnA^0Zo;YWlMwIsnrs_v;nl$1=6~j*@}~r5arW^ysb-pW^#e@{J4zmQ zBQP_znHGP{#7Es46u6$l%&AM!L;+ZrxB_mMk`OLeh1#wh?5tA8l!W@t439~2^uTx-?mCab>!C|bsMhYvr_xE9rJbdPOb1$O?S+BKC)M#{$hR^1V{E5R*nDL!NH<`ibTpA+zeBAqE8;z-O zgYx1I@<>U+1hsq=4K~GHU0vkFy`_nYny{K`#7rfu+2KhRtXo-*4e3*1n-`B{4O3^a zo8#B94@Vd)mpurFPrOyk@m(`;lbq;bY{EMm2$uuCW9*yx33 ztRX-ZG+7eu3U_FgV*<_nb)3#le?TXi#V~b{F}BLrK^gDEd0hjcTC1 zoQa}6c|8>NIbrXkNGuPOW+rDxFu&Eq*{!E1pl#EL@fMuVVj6^ub%7Yrk&T_%Ur8tQ zE&ci+hfhX1%xXd=8=J6&O}zpZRG`Z$PV2J$E7!t>@4|9<*Z#JOBsvN@XhqLc?qr>c zzaza?UmWxb=V4>56ReF*km2~8=6Bp7DOqjYxUUYId~K`=*1_Nm zcN}g#i2T|(cvfE}e*;UD1gl^lXIW|Rdq~xu>FBxR$TZiCXBq>-*u0IcRC|5^;s?K{ z^kuJc>eC?RBd5<&zbi0j17kL&PK!x3nKHZM^O|ozDcG}${DrLb?$LSu+3-qA!2Gb?JgcvO34e#cDCj>*TKx#WtZqU3dmri@rC7G2D?6$^ zj@{jM5RS78prN0Hw1fQYGtZo#O@d)q5(-&gcMQp&%b(i_SaL_l{ig{S+t2?k{V%bd zceB6yKj&FqQPyENZ@87G-RWO*ZRhuV2AvgS><uF|VUf{M;qHdf)VH;mtlIL);?Z7uHer^~(2yXk;#o-1zwxvrJDgTW zWzyJP5;*O8kLn9LsdM{eT$rehTt`b-<;WmyqX?WL^`Z4`B^|eMBjbpvv_9UMdXjk` z@IfFcj`#wpQ~ZAX@dM{6+9TtbKQeX%pg-v(IYc}lWw{=jC30Sv5^O>K>Vqk^s7~k_ z_fQDEFHeiPFFSmjJh~Odk#Uy^H(cfmZ1PfJ%Q`9Y{97sf?wmxD2YQ9RKa)u`a1`xU z)e^#1X$e>6X$gvdWT_9gh4Py>1kVh0!8OBPaGEhmxat3toVLp1^ol!FsdPlJ`1(tj zcFBO$kM4uCOE3)9MIpM6kXa`N$H7(@n-+{v&PcOX$ly%xal$%nGhvjxwXldChNARY zXzfWrfQt{*GyU;pk~8cR4Ux*XySm8}vD&^-7%@PFlq#+0o6T4uVap7`?z_35sx}uP zgI)0F`eb~x%A(`pS4eyR1G3pX5Oc2$#h?G`NWGN%vdad6g?ylO?IGNIpN@TX-e`I# z`u{nfJwFxjEhJu;wQ{2{Y|{b3xj-E`np5#DF95r%xMTXSGViG>V*ajd&O#UEyMw(H zV7y8Q^Ia!w;%slDv9HPLMmD|OdV%_{h?1Xy2%S;-B22rff{F8dU{w{4>_OaFI6e*a zdY91r&q0uE)EB~QHG~_8rB-_lE^-h}2Iv>tq#We$Pc=1PCvq}plxvgY+`7XJS{y>qO zWi37aIT;$3(sRdsw5TwYW^^*BF z1zegl0UHc=QdH1cl3cxvguBVKF=Z=FyO2yt)kbvuX)JB$Ojb9Ng^}`n6np{?Z5KL|ErnurTZg-xG5#c6j7#YsykJzE(T19$VYsus{@I1h!n7OWNb$Wv zr>{2AN@Jc`ds9X^ovoB@!r2g0MX(@F;O^pD6p6|4XXOSxj<`=MZHACL69oHs4~&h- z$M7#F={ethZe7l^xxzOJN#bsCIXPJKv#zej4E*!3fYxz6+PU40HY~KDvpxf`u`Pu* zOjqPS8Z$iqVZi;;Pbg=dI?Nvo#Hk+(QGSK9Z9PlK*eZuw#|*~~Z!ZM8`q6=9;FyfghI9vi0Iq^A3sBwa26ZO3ag;3jv-oByNV%g5r1ZWgJV6;PS~i1F>ki>qRIVd;#TA0vd4Z-+^H)H%BQK@KMQ=5%7-!4MhU?&p$*TGF?Fp|~t@K^UGfA%lak(Lfpa9_YLQAP;*osTzbe$x6s z2Jl^`1Y6D_;}?qGRRwvr{kV?)!FTRjAvnxuWHm45l3(vFp{OyM`aH+;yTL&!KA{MsuRMQId5a$9=wpjy z0(}cPK^uP^pv^a)kaFX7+Ot3%O_Ht1n*Ib@A0I+W@(!gpAEbtDH^|@UE3KM;gn9%1 zP~)pQT6y^+S-i7Gu_b3&P0J+J1)NzmAd~8MeGtaQU*{Zh38cR^$2!>J>7@Jg`lvao zN)D3S#d$Qj?LN7z6(^a><5c;VXJdSFN&Mpxvi&s(`!sIRuZ*E+sC~|}QPa?O`VrZV z;C-SQgOJZVX4gk}LF1(&f5***nN1Xqb@JR0Kd&_}o`t}i4H$gF6{4LE$T&O>em^GQ z&;e(}&6tfH`{rYR(+>EqUWK}|3h?`6gOy*VBQj+$q9)>GEqa=u?dr)?5K^XXk{4h;Iq@+0hskU4#V=hFeSJX&*S-vV&N<3 zzxj;%Wi^~Lz75Tjyf7v@7@JC(V0NknMJ+$DC?M*6NHP)ksGny+ zA8p}YHyVSOT`73r?Z?sKtN4tF??49spfex9nWBo2wD>_cOGo47ybSbbAI7Mb%UJ0d z3Wxe&Ov&Aez&(yA888vzwZ<5f?7?>vdhj!whZV-_n1t9uW;$jT%NX2>RKKSJ|7CUB-hGViNW}EHdxx*ZM6^VfK1$*v- zvuBk)9!#owCGu}^?$+o%*qIrQ!?He{OXJM8hG&ER(>rxDFaH;x2`jh@m~Ei}V+S(vYKb$-rw9 z?LMXeceh>=GR#oC@Es;q-olQppHZmPN2{l#kd~zZLc$1_m>hpz^*IaBjAV;lsi6EQ zQKcyqt7jpR_bb(ceo_L@vW<6aqUVQ(vu6WES>-Vc=I@*fJ5Jp#2od`Ew~ zoi;~E;;i2gJcw_k4o3#d<~5XiX(O#MK14qA;}CpP4MO)E3|0?C*+Wa7P0m3dpLOd! z=lndyk31`LpD?tSqRPd2W?7t#sJjjo?lZ9*_!v+397NW-7ThxJk{VA3HO=E7Z;d-e0NW#J9@EUTl^c~?pGPX(E}&q6}` zM2w1c#HtO|^li)&x}~2&5BN^Ac~ud<>-9o+dr5P%-%doOL370BHV8(QL>~tK7(iKT~C{c+}^N+~v zyM-LT>+rbx9Jg+Xp*PkAKPN9jDLzw;R1-a`ZQ>o%NSIaZ#g%vK5nb1WX4yKFHOeyA zkjd!bta8Qk;h1%$3?EfOFwOS@WKv||Q#S?kW~@S-}n5&o~x(3PeGmjP|$DP8B$$O!` z<2o9SeM=qT9n-NnwT(3N_C_GBpOws2Dfj4Pkb1JkwdN#y$<+4AsxW zFv~O=e(v({UpELe#U84y|IzN7B4jJnoG=k4e4$rHh zMtb{d`F*pHTx3-s=BNzmjeJ+wk%31eBJpimD74DOU?AE;`ROmoyrP);N7U1p zrh7Ccz>;g5h*35=AS4%H3w zQOn<1(ic}C`S&t-r-yLX=R}wfF~{2^H;j@##xu?*A*?*aGi5UP^X4<1TP(@_j(k71 zg}XJrn?qx88Z68s5LtB;HVs)6(Y%jJHXovy3Jx$IGZ|CnS;E3#69(~prcdwzjNtFg zif?NWbaDp@q}s_+Y#{79?$dWIFXVigi^ZnfAyvdZ8h;I-d)pd=zD3gS_I0%PX*4~l z&E_7R4DwUlPY?>^A*v4jq@&lwA0kXo%H0}Us}su z3U1%#87PAEu3Mdj^s(Y_|I7f&i=FTQU!0Gv zF_Cz$6^fm=tl&!;P}^XK6sw6y8XtvS1uK!#Pgt*Kgmw13gO{g?cPWxcml%wgn&B{z zT1py8x^!)|3kBCr;`b6$T%EoO|6Syr+N<0>v+gH-S(8TTzvAh{pA7n#a1|weC6GHG z0!^E8(s*{9OnKioS9v*Vf9}P{$p_IWE{PMR1Myh<4P7^WMLXFC%JaBQ{$1ugyF3M;R17}{w zi(%MfEo@!ULQ_=Vkx|(t+Bk771c#Y8#<^90ZV$n!e}fR-HU`lnS3>;ECXk^Ay01N^ z!^4+S#HC$y{cDJzSCTITbb1JDbdCvG(PL=AFz)JzG$5Vm--5T7G>J6K7HnJ0gj&a` zf_C0iK`+N#sE?5s@~5{6t0tZj9G)BzM7(Y3vVH({=?b69k@)*5H45y31D+Ld+U4ntqPQmZ|JfZ$Q34WGy zg^w4N1;cqJLcfTykn>Luw*35R&(FXfW2=P6Z;AwoXT^f{b5Xo3ngr<}2UHgp3(01o z!g5_VVY|Bo4XhbMPqi$lZ$SaPvb?c*YC1%obW-ub+w>=-n#P`7F6a$66YdT(6^8B^ zCu}ZUCK!t32}5pn3%{jK3P-MLlJ@zlbl&|Dd7MqB(T|_gst47Sw&pd-{f**X`GfT4 z*J;|xXH7Aq$D@|#Lu5UyVZ;4TZf1ITS#=(AWyy%>+s}J54%BVDlxCgNqf^@_QLOe% z(z~ZXSw0J4K2n_jzVy?(IbFh=4-!=NZUyyy%Aw2K_mM);KALng43V76;&+7aV|%O+b~IB6 zA9r3j`>slGvdR)>@NR6;Fh9z_6hkh3YQhOyJiqgjc-;D3xN*Y@g-jdhvT2eYZ9h{v6*SOr9`bP_7yySUuMfMDFyMtV?;xIg)qD!~Oyp z{c)xy^&py+kU(0StI7PsHX5vxK%#mRgeNL;!jt~V!ix0k!jn_?gylw01&uSNLRFZQ z5U!>xJk~fx>1TJ5l&dp;&Uj9+<|gzP-NvF@%EDS7ZK20cRxsW@9?R5Tkd4<*IyTA^ zNp>UgTsIKjd>?Mk^P+z*Y)6=8mSDIpNr--bSNK!B9%0+ZC?hMSQpm_IL_$=GQfR4^q(V_zMj>fPijb9Elm_CLUC4XC|M`PH`qcM% z?)$pV^EjYU6%VOA2SS_~tr|6&ton0ly<8`04k#!8rf8wJ)>L?Q(?U2iO_cU0yc0Iw z8bzBz!$~j3p8LjBC}6~IAxzJN)^ncjDgEV$=I_zFrUx+cdIC+^u%EJ5m(iTIEO{05IiGaXv*0q90`_%Fq%6Mew?H- z)fiH0mc-(VbD;V7k8s3VnuaVBC5`HJw8wQiEz7j#XWChU=lWejz=CbUPKRg;eA7+F zm&M`c>Pu6#kCL=sGOb?WN6(fYr+ujlsC$Yh6(=N8wck;)T^<6T#y!~3zKMJ3HR<=H~+Z!Ns~ypxQ(-q55-5h{yp6r2Oz z3NIT?VPLL?O07A#>8*n+K3W)WB!N<`CcHN-!at2O@Zb0bpMopVvP_j3el^A2EAOe; zTn@#L#4%_{GfBVyO=Itb(cdl`x_{1?lCm_&DaMR^HfYhbJ>P_7@xO%wh2msz;3QTV zzs15EClL4kuQ2cF1>tI7k07d_hPuNu@cS5dPhZ+hpVyA3N9JynTpf-DrNP*0KMz-? zUBaxzHOP5Y3GD&W7;()EhKE#99yp953VVf5!=GMWcBFaMv*=Q$D%HP|W#8OuaCmDi(gOHyCu#tE zY#T_ip680nSHeAj-yKka$^R82N$);HtdBx-z+r4Zy$W`N^)cYE5x%ynVk*z&2&*Gt z$34cwJ;Kp<%@ud1`eFX+b@bSGK8-Y*M$Y;-Y3IBTl%jfqyqX9bJ%&T~(Ky5{JxIMb zd8cK0A{BIrLObg!ZKGCt$TK{dTT&5yHXS$D*ub<&6m}O!BFVUcHe{+ows`!e>skh8pe=Z-$1&etxDww?Pz%V zC!w@ZiVUP43&(m#^A2|*O+TiJZ!cfd(V_%ut2;uo`(1c;-wuf-dVC%^n4Fi0&<3M+ zq02*o!Y}m+rRK6^uh1o&&8iTJb-gI%`VU(B`XiljlZJTP1MaUqi`11Be5ThcDCj;F z>;}lvqWEA;p6rE^9p-3$e-e+vuVQa^1swb%v20BqCX0sici=N2_VN!Q?1~Z%3rt46 zJa-udCn0-50=!C+a8jG|Mqf_C0qG(5r7#+4`nx!PV>Rc3t)P3Nhp64ngveHs>fda_ z;G$_*`qTl9`VXmYmLzJA7SX%O#dv)94HReJ!-`$ENmHqkMAHs%{;V8}_*R2W5sGYS z+c-$uE5dR92#BtDDU4cHCDa;<(2DJQQS1?kFSfg&9-_?B&RVdep4_FoW)*Iw@}9fN z5uOcgqPSBc2t6$$-(o7Y-AKA|QIwrz0?jHTj0*ls!y8V)`9T6Q z8+jHwT9Hbm%Z2;XCE;iYpJ2 zaNvGA=8p}6R>xuV9-9u|m2%i{w1>}$`YG~s1e|^^!;-K`m?=FQUxOU5BxM!SUj)F{ zoO{e0HXuh|58Dp#8SyPyjQ+A1FBYvrajpaA$cSOv*Nb#owwJ^zl(5rm0NTC!$Xg~6 zH!t4ecdLu|>E?j2QM<8|XIrboN>MxH5=0HEA@KgUzuXO~(|t|9RWBhduoioDo}*N{ z0?k1+$Tn=nl%naZK*^N79Ae4}?mDviwq@*{w+@>L%qLLx=6~x%HEQi9hfPR1pCF}%x9uD8@*=;ccdpn=5;xG zFFc18pPlIXT%lH*d*s$5fvA^GST+74jo+P!0P%2a3pxP31->}GbPJXiFM)}5C~n^^ zz{Ac{cpjk09d&E*Ga(Rp@ovoQlnwK~If51LGi1L?W-<96V4-@ikV!IZ>>YVF%HSAY zERVq6mONaP2|@4JDoo`r_}$NEGwJ>DtmcI(a}6)S(57U#nDTp7{A7&a&m(>Qd~CS! z4{m;7Geo8{Gf{hXWc>oRV+66Gk2uS*DIXrsd6xIHF3aC)&!m$rnfn_(EM7SYiju2v zd%rk$SdN5_XFnzSe4u|nT`*<%0-gb$h-98szs>j@E9W^4+bGBRpHkSwGkn9^Irrm& z5$I_-Y_|_)2ix`8>g063UunaE-1k^{e=+_^ZNRwmvFLozO#P>L&*O;__tD;B=9jNC z-IU9$eY6?d`+has{&6h(%Q+Tzo+rb1MIv;HnA}E!AYRhT5_n!L8JAjB_WRo~1EI)me<>O!mP| zoJ|hx!`}ra?CHcwOjBbao0;OqZk5Kdhk6?@Bg_-;n|&~Lxfc_@xiIyKTmU{MpAMBc$6@d!rk z7{~Xf41cSqVUp%SzJpkc;75Bg#BVbz6$xQd5{ubN;U}DwYEhQo0V6*9_z~)cm&2Cg zPhJ7O%AdlDc?Af)&_G&K6>)x$0*3I8%)DepHdA2;>;G=U`us+)Vs%#*`(Yij%nl;D z`8YNdh2vmZIDTH!Lwx))Jm+(wzc(1pEkBH!O=0lldj;Exx-2hSp4E*sWYL@}JZ!TV zYjc>xnq4#T-LD+MN@q~vt%1Qu-EjZOQK;9(Lo;m$LKNc>J12%J^1G=2#T|;j+D@72 z0+y)T!1eDcTaeA!7s@jJ0@<{?Q2`mc~r(ACpa(i`@^bdLBnx_}`i3Pyx*+^P7A(S>BI8NmcP6>yUnuYB=TQiWq z-wZxE;XBXU2WhOQFB>M4=^$|&UEeJfUjHW;8l{ob057<2NP+0JLbPwRgy9W07%%t6 z)W80?(;tY|OPla)fg{ds9)kS?_F-<63bh_Lq2m0<^tZwX@e5|5Kr{g+@kZ1a5lek1 zipW>yI>}7>MZHU9;WL6ed~2=wOeqGD$0yR^tyv`OFQo6zw@BL3lb#OBj7XxbgLUAX zwjRfgPU7PhGg6(HO*KU`adgEFGF}%!Np6b~)_)3fMQt$UaXO}J-J!L5n$TavGvo)o z5f>hgSv-^Kp5^%5;kze}#&lAQS+ zPv$FcQK$Q9XdFo(->M^Y_1-%wGF?f_Qax#*^mJ%?n8I%D3^Xdm;ZO8o(iwA+HV$c_ z*IQG_a!nJRb>~c*Bm>UL^1*@nbnHB>f}GP@C>M3Z8Z!s7y&6f1kqStiZAihREr<>% z^R5*4PAg5P()q?1q+Lh$|M^mERX@o*x57MSXU-s*hcUeTvm~{U-aU4J>aAlGweKRe z+3`7Yp&bo%UrS;w56JVU4UC(E@gLtOkKt#vR!M(oh6khAU5h;Qg6aF6OVs%=1pb@i zAaXhvP!GnlifGjGIl7(WTnx5bh%Lw0;j8mfT=a>9%!VxP4&II*Wt@RDZVznu&e&#c zDC#pg%jDZVG)M(widz}%ifk~U-woO?R^!i9Khm~aOb3Dme2Sk#J=%_x+@}cHG7*}! zZ3)RnKcz{BJ>Wkv9%b355a;GV?kAQ}LrXM8#7V;S!b~^~Dn^E~1v0L3AC{g54l53& zL$h~~`QFR4J~Rwzg*;1|nvRWz>nU2eN^{%RqW=0a8uHJL?CL(zChO}Y7P6BJJj5_J zQ;Ysy8b>#)!)WDjds>}ZMyb^noI|8QO)h#A^`euSFI&QNwi0yWw_@iqdHg5`RAdk0 z(SjIQTj{XkZMoL5? z7r+_I0<|eQ5!-H#5nm6{h^_yTx(WBJbLRBFvRJs6Wnik*c)FlDjt*6D$HquU*o`>? zi@1a670%Jp`Y`f4{h8*8J)i`8W$aqZ{ZC&Csp9er+V=M-a_-5q&O4mh{HPCymU7o; z&wje#nMQZ=)^Hz<6j|+R7rcK9G*1a+qQiYJD{c$x%eD#Iox_D+zwBuF>~i5_bD=O+ zWdNm4%@<~v_z6UX;m~WvW%pJ zwjnfnPMNS@{D;u@_N*{Uqez&pqARR^&?RWZjG*EjQl#dnNGYCe6XG%^@OK3sUE8*;%atg6Yp&eU>(h$9FL3FGHC2B+nif=S2pWg$=Zb}xY zHBQ*Y4huFSH-#~NoIBXPj~t%F(qR3W^w(93I)8KscT>&br>T$k4^Pm+qBJ4nTZZ5g zwqLmL{0vo(^dxtiQRGv-iatG;;jFs%Lc?k`dgj(UL1e5fnR)mK&2GVhT33+pV&X-i zeQc5-B_&5St=hCMY62ZiZxiH#+vsd!I?3vmkx0!HT9LV%T&`RZvNjaZ%-LO(Hd=%7 zPRYPx>UcCx*TSB}KXmxWV)AeeqA}KW6dLA4M&*yW3tWYm%}9cu87UQPqyev#sL)G< z<__fE2KSllH6t3Q+p4hK#MyG}b%dWtc*$Bv@vH>GrW=Q#R)f{~!H zi7XuKNqkX*;OH`%>{qo5=O3w(Nkkx>zB7Uj|F|q{5S;Pu+Cjv&i(q(95Cw-?k(98R zWIaMD<6A71mDqD-#z!HuK22D1q)8}NlBTmR;llgji9)D*z3_9+O+n_|F=0blDp{}G zL&uLVrY7EdA2l+Xibr21iEMHHIk+wy9dtk_Xi}n*z;dDY;V^2Lqey09uY_%`VZuwh zgT(4MBi1LKf@G^{^i*g1yZ8=SkBbwG4u%MYHE)GWpZ#fV>^Pb~?J$-041~q)3cA%C zP1RMKC`$i};PvMcm8)?Ne9v`ZT>b??-|izVeyv20;x|&;$7I^~aVdGP{VvQOw1c3Q zLROr491?U@I8%}$d{;CVYLn%D%NJ2xS^)XGNYU=qa|FXh zi-n<_1$V&zr!e&KJj#D=j+Fz?;$C0}(H}jszFQ(hsyoy6b!Fr%w~Xd}btm2Lu{3AH zdfIVEpl25HbaL-G;oMq%8veRlIM(4Uq;y--%8Sytc|{pkwa2OWlLAeU_Mq0idL&Y& zO@-YGoJD0q=bwiQDjEangxw%AnQlR@PnXfA&Of z%DD))pF^P>=z#KjX&7RnguD=IB%E7@?wql>9<&#~l*=(|sU{BaJZ9tLEZ(QkVB?hb zu%`X%nCRaWnEzH|$+i}3rluO!+K$Hfi;?)Xq6`%~B$;IIYYcW`Fgvn{GcFGxMaB-U z#mf+VAPo1dXJSI519VO&!-)5*b{|i{`;0dz`ne7BV^h$fb{wqaF|>zrCxWLw+i+hE z%UA28Ak+cPn-bx)S)2{m(_@%^9%tsCLxoKvrp@Inh_4%Q^-Bug)(k;i)HqbEbi(84 zA-E}Ghe0k87_c`A(!fx&-B*us_ zEAJ~f*r>v?qlU2{zx!D8NRG*<&tsvt+i;V6iSwLvScsjEY(_z!h@kGbP~G1Qm1Aipve(nqXtBWV}kS2sYRCIZ?$oHux~882S* z9JF>k`Yo=Zzi|x0TxP<|C?1xP$@saKcU52KV78Yzt21<9+to8!yw*dk>gYu6;4w`9 zfE%W9pZ9a|{YVHjLe;Fz$mol}WvIA{oy^l8H|EJN6Lo8*hM%!D`rf zhNHBdpIevq;eAI7Our3bmf9V0b?N+8w5042Ud)4VcjtVVR96?9BU1|56 z0i-CWL9^zo(Y0PxI%8!+W%g~tr*+=6qh>QD9dx09<{hLz%!&r3+K`Ng1i2qg5u{g2 z(99qmvN3BDF6Hv}-mPnbl)E2odF)0tJKhW02aQNqd?YPXze9Zu+&2+dO=5G-lEs*T z_?MMNT@Nn^w<`__ON4Vm*O-%(ZXZj{6E>53<9)$@*+eo+7)@SbcHGbGN`pR$)A_w8 z1>5T%1+CVhw4^(b_M}VFnx9o9J*H7eTVOy+$BaoTAdTW1w^M1(NitI`6fC9dgcq${ z!k~6@>QAsE-7l}GOJ)$vpXLgN-^bv#rZ#Qfb5F=FW;8COP58LDLl~c-K!zLh1kYqi zN?BMf%-G{hVbcfF%k0rq^C*v0zH$HH^mdw{xq`f|T@wtS>(H^55>gX5TXj$^XK3?G z+5+xr8f!=6jen6`TN+h${iJs1270vM7Iod*gg>JFg6p-v!q-z{=*inz^fqlQm31AV zmBv=cN*aRtCRMBpT}*G2ZAm*(jRyR@Mql*~k(XE+C00g|G2b2SoGOO!PrJ$c(SPK7 zrki}mNRV7D_lo$blGj)xnrwcCg5^T#{Wk9_TG z6XtGzDl`wWqR$)O3ub|}f{o%KdRM)SN+)K}@x>G9RXKNC&E#2E$6s`qd$#tSSwt>R zOK3p;IpKruFJW`zSE29EJ0Vf?tMGQJJ7vfikWBkfs!vrTU4<3p+W>0DZ2 zu#XhJ&Zjb0?jIeOMK<44XxVN_T5>>xYS(@ceA2lC@vtSu*^MHFcMB-~)O0%R*)Ggm zxtE;VElAYy3~hUSNw8^g=S&xQI5>S0I)4wRZ-M^QBi=%ThF+nt6$K=_IiK`y6u5KDho^89Rz2legEc!GZ3yC7X-)(gLa1lASoud_Ar)ie*4dLzn zc>44pN$@c~E3EvHBUEvBW#%q>`luJd^D%>YzHu(CT9HhLH{GCi-bI?XqF#7nEf}^>6y+VuHw@;vrXL<#t?{9@Mu|yK* z3WaOVwe+W~nzg}z$?RjnCKdW($c{%dRUlG#R4tAa_U{ynL7woziuWx8{7I&G|1Ch;Afq@!Le zyc_&cm}7oc@KaYHEqN7sxko{6(!NMDV5KBKf9=Qs=oNB+m0m zKPzX^{)P%tzB-%oHpO5v@A~*lRp5BV1G0JXkZyC2V@9GeCh=X-Ca2*r;CYoPA+p$i zvzJDWsiDsqJv8mhb=t{$25axCBX^JlzK_y|O|UwQxZ7EK+j+=s;hv|IIHWYarL*IC zuBx)07VkfX%a#kVZdn@a4tLYbgzt3x-$M#~+DDBsyYSdRAAbMVqh=Xrq?u0PJCuh= z{I8jF1tghV-U+CC-Gf?RDWp}F!EKZw7H?UKe*-yZ{#hN_Egy`p$Hk#DRUR2l>hKYj zz>M?exHYts`g^|8tE(UB>XB{=b=iZ>ay!6IEI|0svH z$z!Dq^6w2t&(6OzeZ4E{i!AZ>jXfq!xPYbd=TH`Y4a2q=LtnOsyw*&?NyS9W^wN3=|TUkek}ha$)clvVQ=*y);a$(1l-nyDI^G>ymV`saD;xc1jziAs2&7(K#Yg`r)O;v_$u&ck zYpTYct{KP-#dg5b$puAn{Or8!En;IkpuOcj;`W@v0mCaO*?j@0AEnVriPIE#JdTpb zi6gwanKpdn9Jk%Z7~mZN<7xh!eZk%Ar329VTMd`J#$oGmFTC9|AHwNa9C9v3$;MQu z)WpJjq6ZjvR7d-%VYY!9Oq|}(pYhAk`FtXB-8Q1FJeVys7|eoG*R!mXZ!r1LOT-Ll zgx9>~$l7d*mjzlp*co|4isMSSz}g-w7bO%=-`SOh8 z`z~s^5rqDbx!7&#gXkc2+&b$dXF4Kq_DX;Tdm) zu#xIW?`ws#MHf7}N^zw78tpe$z{*{%B>(Ucp1Hi{ynrT5Q{q0b?T2BiTMEgTZ}_%; z5DW0)jyQ`-N_|^Tj^@W`vF1P6MGs~P2c_6Ij|!--IE**fui@q!SM*$)fk!XbAo2HZ zdUH0G+DlGRb?jq`l)6NVj63PwmlKFwatYVpTt{b?5yq4bL*Y*D)z&x%Q`uuUP?L+c zT^9U%cfqGnN7Nqsfqh<*EN;agNTfz$)QLs7Zn6iG0r7NJE1SaCrI1EM4j!bRMdI-~ ze&-p$?0gm3c=OS0%IS|Z)F=XfP1O-FNDccw_0an#s@%!y%FooA&>iW71#@IktD*!Y zUB2Iq%*BJGv+(0QL?s%`p5KvXmFvea=dV}jkwypQww<1X-nnrwJoJto!W<4~O#laYLalCCN^-MI~YGwws3 zw}l>7h0y!`gXpBkcOm%QIEq<5o6LAW@Zg|qdh%}oo%W6p)Cc`19JHQC6YTZr^5bA4 z?zy4%N&)A6h4RA9!q{y;g~z@Zg&B7;sQJ7hWwu7r3=@#n%9|wiT#?#+C&JZd5rjKq z;oUww|C$)=+k$@EO-h46OT1i{mJkI*d>D(L0=2<>lo3zKh$3lBy2 z3GO4*gaYO)d|vE|B6A&>hD^k?`-gewc_nE)9Y;euPGd(aeMw1U3&pr6@||=J1sUEEGLvk`zF;8TPdSOU z#2e_`mw~2vkvOWD&AaUm@JlEZQuF=^KHW>G?^q?3w8fC3NCCB$j-jQNHgwZmlx9i{ z!layf8q0D?&Ow~w#xAF#Ba`V$>T1#%016r@L4`YHFopY>HU$#aMeRr8AVb*Ca>LYd z7sx|?Gv7IT6Qp_0&8rkm&mv%c!I!>tSOA!2JV}zlK>xJ5TA%gMWkyx>L1WLVm2eZJ17VdPQ-Wdh7 ze0&nEIx&|ny|$-gs?(|4`LvKOJy!@ApG}iem(r(l5As*JE^PLT6ddZy1RC{6823}0 zMkEd-JL^CSxXxYL^BaW3giIlA@*!bT)_kEg$QFsYGmydk1qrfeNUh%x>)t7&rS~Or zU+lx?72TYbXhnNPdWGXp*3;nRd@Rp92lZE-m}2#Y-f!`x-`n~8bksc3T)c>9FDz-E z%3eXI+)vos5-jZaB|%%qPou{RSJP!HKI56^MsttbQQj^~+Vahwa(F&6V%#>o;X9$a zTmv+3|3kL3RMBCr3hTP1unOCVB`1_ndEpTS#%0mx-0L)4Vj+#BA40!RgV3%TNNeWK zB-4{qNN&;;TKw>cFeax{NPaJgnEonqS1zWe@aeR`zgw_(RH3MYm9*M3hn{3yA(u&q zg|+{(g()dVg;6JF;I#W#nE$ke*~%esO`Qz&K{oiD>`YZg8|ac%G>JC8q#Mb9$XWe7 zHJ_PIBgc&=mEg&w*|D9D7nxC#_5?b<@`-0OOGw|UuY4X z(XgQ>H(%n(-c%UzPVEu*V_48~2#<;wX7AgMXYO-wX{r%tb4>)-)A3%%Wd;+wd_#*C(89I$8K&rKlDwO9T%;YEimePcB z=WiM*I}~DiWJVR`BXU^%d{of7AXlf&haiBZA-xh?ZNHztyr3*iRcrWs4jg@^8DXr%4(s( zs-9x~k7D%A2?%faO`VzT2$!ov&d3v}I=C0cd|onXpfB{xldx5t=RB^?$38PnCjYPn ze@{2U(?)`gQV?Ok-74U7YBDR29>Jzuv1B#bo{;@E8kvE7x8s(N!#R-F}(KVA+^M<3AU?dx%Gy9sRMWUw%0IBTfBjrt$2@HXlcjK5{Vws$Lc z;UuH`OE6wJI%4Ud*)Ux@357ET;HqyZ%T+04zt6g`eL0q_ZTcUyq$x1VHnjJo;mt2gX5q(w z?h*r7^imz1JlIbS6HH+rBgw+=o=0tPDmrtgu)Tbi^|)4)4GsZ&d`z8LX@5ZF(>yrK z9e{@9UfgvU%X-e}vGBkiEHJ1#Go2~KIkMAx&ckZzWr*$cgXz45c)MsCE`@90k3G9! z*B4RNhAcl(lNE+GBL7@GZup-?z^*Y6@#LK5YTjYHrG$pbGceU68KR0R-1~YMq6oov zsc9JX`8M3w$g-}LjaWCf0B#aF2ngGXW~WeQc+-U$o?gpLOLDOC<2D=;vS9n|9ffgL zK{xkcObJZl-d;aUA2tO|ca`yU6+^(v@lcq3gW3{H=;)Cuaxy%ZYNM9Pt z-|@K}zhnH1KL-1{hoqW&n`(Ynk?7wxTCwse4O8IzH><0Z$M-znQ}QS^V;Zwa(PCTey z##i{tv&3yrBe3pdG7K}iDCYV*IN@}c72Y)tM5rToxha*x7~($jnH7r=9H4_f5*tX zAfN1Y^qE)NN6yOn2b)Suwxd#$y*2DY2WPx)u$#!Fa@z5wx`f8h{z}eQ?~t7cU&UHk zvrU{Go3rWzwk`UJ&Y~t5KdHc&m1mGNv;=c+Rne1=ZL}~Ti{8~QfUDD5NR~U|_EhV8UuN>7Hd#EFT}HC&H4?MXN?%bP9ws?XLOUC&gn zX5zzzTGURwh(NVes#QHhJ|oVPn)X%1ZH&jZLno2?atC`>7sQUO&tsbHZKzUvfh`mB zQ9C*k%~O&f`6mTAXHt;jm<(&)@3=Qp9g7lW@%P&s^6{v{vhXGh^~vGS)OpOPehQm& zh}g;s2|SoE0S^L(pzEp&wz#dvxTUkPeePYVOD>`G|I#U7TjEuH!kRywm;f*I$!9~{Ay^Y=kxtuD4ns=RzAfBHPNH~D|Bk=K`Lvr!N!ciICn=M=f`_c^Hw8r;j>U3qg8yiF$%3| z7Wj5%GY-w1fp-%|V~$-MCf%Ba;hB28^EQ+%Zj@k?^13lB*dC_w7Mu$)9LeuT_>mpJ-y@a$jn31}i2dx@oM7zF<)AXov?9;AAhFc7NoGqte zM>o*e5xc3c;|!K7t%u%(K=hAwr}AB5lvFsLK1>xyb$>R=WXIAGtIs&Y{ZMs@Cy_=| zF+X<-GHwjUf1PU7VD(9`93n&RD)yAOZaV#z79-c9JW_HCqvsL}$l|;Vn!OWgqe~o( z()GpF&N;XhIE&8$)X;x@DgMil!GG?LQ262o_tZq;+T^1UN<$&D&I75NyHK~_K9smu zYtfQPF!CCRwsAZoZm5j4Be)Z+u9h0-l|r$t2!Ef1;9aLEJO{_oxyzGjFKeL!y+O$3 z9qUz&|^f~=u1d?M`4$3xt~7+zxrAaUCdGJATFR_SrSU(qI>k8onP zpXW299DOF!b{qSLHKSU!5~(xAaBf{E9qOu}cQw|uY>_S%F47{6J3%xoPnSCRDt^PW zkDz#}`qy=+QsC%=V%9Z31NucObde-N(z6`$AgyxqZ(PANr!~OLejObmii5T$2?b_t5~+vttkuYl1U(6;QZT3VO=mb^>FJ+(Q=sveB_y&~v(_LC-08-PuVP4Uvv0!5#0lFzdo z+B)nSeO}#9r*0|ZgWNZIlD?8mPWh54?^pI+Pp3nnXK8QnNBS}(3~%+8!+U8Q4iqVJ z-{p4_o%fwK>k}^jHN((;b4*!~N^kU!(${tKNyYaqP2yawBXxYfm!^gVw$tD2C1dOt|}N8r~1w54}^oYo{_1H-r(8%hJIK-m$X3>P=rH zqqvXw3Vps^Ksk5rQoi0Iq`5d^f08pc{Rn`@#d+v(u*dY2hm`512W^WR^!DU7{EFi} z=FojGK9ns<0|TZDc3_8jtBoUKvGJ4i!2=uEx{v_c+{`^e!as2K+~WID|Gv&`}L zTD0G{!S}0U7$yFL_F^&ieHp^4p5<$IytD#|Rfrz)5yOu?Lj8K4lkQoMH4&3wIQcY9*`G}lDw}D^s?B6v ze~uOvq|%afYh2j40WmxFqqHOix(g4(i|0Cb3>$$FPV#uKAdOc?_F;vU0d7S`L**gw zBWwpmgAB3lVFPS35_o6%6q2l*F+@iX8y>hL{{D9g{+U2J?J<-!G8~^uY!E#4AZ9qs zLc^8$_U4o#gO&nF^-xe)rbu9^M)KO7ijg z@S3uZtfaaBRU?VMcb~xmFMKMgbHA47g-K8!wHgZASk7@y&VijpnZUaEjFUd~y=|W`NXBs$$^Iu+x zz<{&bEp?Yb>E!{me$GO|rFfkD%YTki2D~?8fl$u6nrJ={o42ZC!mwU?=p%)o;$P&l zD2`fW-_ihm&d1^G#rFHNabnUgj7!RcZTekwMM<$DR}rL)SAyP*@i@1;6e?3v@zm)W z{IBn)A+OI;wv8y%{pB%HLLWn)&xWqmOVV=F#y6WzI;N?MGyS{^)@O$J)@eBKFa)l{ zc4D$Y1#)lJV-r5$%CF`0Dt9AY*jmIHqrC6Ym4}lj>(Jvv9@R_sL$w3et8Hp&YXct>Js%@OdkeP4jT*xe|LrV0QxNGpRCb$c$Ts zg_C?S^EuxSo+!ZvU+(;Ta*GyKU!t-FB3RSXi-A22XqO$qY)<~B^b=|r*(1Q@YZM|{ z5@D$K7PiUU(Nhu+x2z&WzZrl>{R%K2%Xc}8t+40rG-R|aKojqZl)MQ-gL5JLTeJ{X z&wUhcRPbuseB7Apju{1qA&eZ10-j4ViVj0I-;FNu_CU3A0gg*m(9Nz|x_Lr%}tEI!ySoz7BCwSh15>AQf6g|%IA?(T<61}#kVfxK zIb5*Tz~vPheAX+0^$O-#^kF70=hr|)Dg{H{6r*fs5^m4Bg2DICaz}yzPIFIh{U1~K zuJwgR_BPmY{`@EVA(-+&1Jcd9FynteR{IP>_cY_g$XtplE1@4YSLyNWKXl=jGM?QO zg?a8_f)bh6(58b^CjT_)d)Tl!qH?`4M`Od=ItGV;`Vizxabfn`ck0a`4HFI zKfu1d7@oJ9acXu0@*=~y`+YsNwj@$ZRTcHG7DZl)JPymHVD_NxFpx>XtO$OORZPL= zv<|%YpNSizEnxe0Dr^^jqs!ankUOWHvgK5tIWdJ{Od5-avIw5ZJ0PHjU0s&6mNfnTC*>6zDP06$ooUPd2cnL zB@O8zS5Ww`gnW8x>GZ}cR1nH=@bXy5sLw=gcO>kBlF%&i5YHW6!b41)oheggB;$zm zwloy2O~8@_V@zGDkFSAtc;_<;jo&swvt&9%j*p@@x4fz9ZzOp+asHfIA1=j-vGG9_ z`208zPfk6Cz9XT=UIW#c^DtrL4U#S_r7mPpTbV7YW=}@hc_(bBj3>KOadc&BBK@Cr zv&VTLA~;)QP`)AW?{ojwVZdRlHWYHTaC9yqvu_@Wsz*{<%x-!uuSl2fsPp{9a5`%5 z!TSmNSbZ%3&V}lfv(lbMah5{)y;11w;_rid3fTL61f~c2A!|!KGE;ToKTea+eN7?P z`+v^WeoBg7O{#e^tjco`b6IS_Skg{AqY}QMLE%n?VYvlTa-nEN;;sCs?9Eh zDzkvF<@EGe2yIJ9rgr|hFX*2I7P}6U*I%c@*>^~C_+^gtJ3;lEe8`gaQf2UT5}9F3 zvkd1_SMzF$-tJ2Gq+RIIb91QrSCPZsSESDSlQrMtNJrr!xmo=u%<16$EA2*M{zC31 zc67y$&$}@6+9Oi`S4bK=e^QRlMtJ{cg8O_{X`FJOI1B`z+~jc1U!FzeF!t!H4I2}{ z{Uds(Vfs58&3jGghNd;WbC^THkH65#ZR*g!F%*NNt?>Ez2b$ooh6#&K60W|XTs<`m zU80JCyIf(BZ3NH6ZW^;t1i|w-!^eYnvsph4Y1@v>D~9Og&L-dOt8my%7yECIM4|f> zXo?QNO}{=Wx~&IWbxU?A+?&~|`moBE;`kZlh>o5C`0;8a_Z_S7d!q@he)tO2fAg?s z`3qc+2*vt^E||AG8Wu*@kblWNv<{B&+3k;B3uCNZ?uq{I-tob?`g>T9($_PxpJ+x2U-a4V zQ6F*l_9#{+I+mXJ4CcKZHM(#+3xeZ&WE~&Es@jxMZ1aOudB%1b&mRralR>|=96lv^ z)1Pb6^hPpH)3!J9Wns;N#t#}Q+(b-9f!@aO%3h$q`Ex>FIbDFNn_tisn z>6R;J8u$j{?#Pkc0pJR;ySuTVI|8e=?;<6+C`xMPyzy>7`f$^pvsB|~-Dd$G&yRq& zpaxl)Z)989!TIgwG@tjD^@4Ab{3A7(8-1lI(SJy4SThMdhS+>U9G#aWIWOo74GB*m zZ`V7tr(cIco|=%kjXZfo1#y0BDvA2+rg=|B;G*J4D4dy&&Na(veZpOO@~xhdW(c5M zF|=nWAX|I_WqdZE{M)K$ugYwV|q${*R*b@ayqy z!*Ed%?L`{2(2zpk`#K*<5+y1XNk%FoduNySF6~8&RHPxLWJN-D$Sy)ZTNEY>nAXKC7)c5+M4L(BDp z@a(RD@z7jI>77RWoo>iVyrN$hYUu&*LtJy5g@(KnP-V%8Iow52E=8m~<|NfTNgxdk zUot(FKy_IG^zp)WTDL!h9uH%fF<>-?t7^dHssZV=Y16qQnzXO_Kk^J!#l;Q%5$@PQ z&rU4EP>&Df6|9Z$GlL*&` zDFB6cim}UeA?ij9M5sg`oqawOuZ(yWc)W zcEbr9m3F|ny4(|eJ`}5!Q=)iT_{PQ)PzPE;2Bu%I* z(~owFd@x{<6PijH%x-^!MD7pXHU1A3MUtdHNuMkRJJJJX&M)mqfZOLNjGb=I|F0{s zwS5WN#M2;oECl_FqY-Z)KR5f{rEPjnXrOHQ{8x<+a0HL7s5ZZ12T1-S+MgUwq`t``WhL`^-;sGQ>C;!{4>Q$ ze5Y=n_LJsqJ#<5BU(|^1m^^M*1IP?&r{$=tWbRwL`XXC*fPu$(bzjd+u$k}-? zXPD+t-|EBonsbcbWl|v(k%tP?aLx`1LchPKX{^*ux~6xXvZDC&AUzSEvuyFAa}!29 ziA2t~x%j)z1w$ViV&)aXsLkrIty99wN1U6zSD*9$3^8t3e~7IL#ryg5ky@O{-!(q8 z$&B~d0>k=0rPC;lL48%J~v~!@{xSbQb?JtGdEj&MlIa!Cln<SQb~R_EpW=BhUOM{PDn=a!80fxu8(0AqhWP_2BwTS48!qha4ku}p!cWfvE~KN zqq$7#*_?rxcA6r#)=}*vSwSZACkEdW7Y;t==iABwNKWB77yB~YdKwSUVQElG;+^M( z=_u(~jIJ-`^dlmTlGS#Qdjn@c9Lzv>TQWQk??Yg95DexW!;BsQ{eN}Rzs?>~Qs^Yt zU0qb{e4X|hJ)sKJkoDDzBv+G!dfvy{I*QK>%{nNys)K?qex$GelrU@jU8)N2kGty+ z!jf~8H#sI_C%-4{l2*mPpRx!jj;HLKM@d&Zk9upb5wi2CHM<|eBx>kRW;pF?xl2ZC z)j5lIF!Gm;z^c%-xOstdvSOFwFt-5GEsgk5gasT-t_>B$>mxupp4^u9o?|r)H@*f6`ah&tq&1Z7l2e7*HG96;6xSYKe z!T%!h*;oQOMK|c2wyJO9e>}J%$#F zK+U8;?3=0yySNTImT-~EEyWS-S3n#2PDTFkX>2`m0YRb!IP`eYtZ|zueCujD(-4jq z%|?)uS4E^`3#DIvO$X;*B>(079?$tQ4Z0?%j;w+z5@ElQbLBWo=C+KxkT%>_(9N{L zV%~-Jjcuoe>6KU*REU-}J3;05X{eew*50~I546oVcVrS$+_v!dQXPHU!@K#*c~8M? z5Y+iRG`-_7=?f8P-%*HvJj-`!UNoIH%%}RCAhN#@jzPaYad>tFk`;#{?VBmK+33P2 zd>>v7jKnLx`;YlH3M)4748qK2GP*ZV7+S0-n0rcb$NMR=o$W}G(+99qa`sdbZcX>y zm08CJDT)zKV19=(L{1ml*{XiVOn zePufIaK&@c6YU2U9iFme5-@~aa@%WBs`Q;kCN83sN!6m{x!h;aY)I9e^F)E4l_}y^ z5DPw((6mWZ`{W&)JZCBEwwq2`1BQ~ZO)_(Q zHG#D5WwCDO3l?)Z-%2a^04s2iWcPFVKENZA27ihX4T-tPidODq*P2GLpLIN^F!}{s zZoWt4a&ZwyG8I-E$Z5v&+ZL>DUu}(s>&b1=J*T}P1G?W zt0h(Ju9lX_e@rURc3O~Or8cwuAxZD}$nwBn9nnYU{WNP4XukVH(WR^-WbJE6ho z+H(OA-0Pmfv zsbuRYW+FC;R(afH4_iK3w7Nu+m~$h${dNr#3$NiE#YVRN{xFf#y1{UMpF_VF$*|Km z?C7+|JvN~yK=gV3Ve)w0%VOGxiQ;MxlJ@Q%*0(ZAq!CcVq8=Nug29%eXST)^rZ$Y4 z{w)!$X-%ioJ2XhTb`(2v-IjDG<+6d#7PI2#-eebT#E!P~So|&9Lm|2etS?d)F|@tp-q!*h{y&8b$UOD{0LsZJy<1 zqOZKVKIvpUv(1<&S~hPO&21e`dIfpx%Rg)SxF?UPnB_9tLj&pVav%1!LnK=HF_k8) zwxPmh#K!N*q>?LrY+{HBJEq`34NBE){L|g6r2j~|YjTQheId(KR)D%*4j}i+RCc@i zD1C14M@MS?+5FonG|r-$eX9Ct5f>s)I#CapTdf0oEL%n6W{Xq7u8$TAEu-kMdk*XU z!$gxlr}1}jE%Veh6a{#=(#X{cw0wFrTUa2A_WASa-{0pJ*PoQoTXj_$=u#qT%G9GM z+aR{u^tOeYMhIK4F_78Y=!y=vo}|5RYFNiZb7q>ni{eXEC~u9kXq}5JwQQ?k?Y}Qr zoN7Bn&)f#m%sYcblg=8F^3V&+vi82kS>0PyJLwzCE-@87pFf*!+$v^DzJD$De^aF2 zw`Z^m-(}gG?o>*Bv6r5W8^X>f{-H4;+i6DG2~qJ|5p7kDV{)6XTiiUiktSxPl2G3y zQsjSEL7NKMvO!8BGx;TSDovg8q+-}*4O6N+xsGiQzHPB#v@}jv$I{%9(IR?#gl-MC zrs#g+>~xtD)s0xj9G}?6ge}Cp9{OfqkYT&$@NH;uv zV~Rn$ZE<714~lXZK(%E#+~;k8!q{zmAF>hc@)meL!vu4GICH*`4-|e9a@L(f{{3cT z8RX)})JP1yz6A?sabM&!C+KGSVQw|SF~bHK<2D<8FY~-}~`ZD8owUv-0jBk$V` zy#n>XGAuo)52Ml{Fj{Vik&B-oZBrH2@vdsZ)e?9oq`}?H#pw}`3lnzZl{GM zDaKG~`%a55ia1{fSX-rvAf7Wiy2J-!{Qn!~e;>EKZekECMD@hc*!RT(!GbBgjy&M~ zmk0PRpN$W#g9SU&W^~O<=N{5?d39>dWt8Xt!S^0Uz_cyxY-)1J$CHkq@|f82(-!AV@}*w6hEVmP8Q2&dZQ zarcD>YFt+$B-k1b@m0{6or6Zr1Z?><19?$y$bUH%3$`r5MzhKAYV*K}OPN?}upK-7 zrsD0Q5R`JKL-4(Yxc__#wDRZTh@l={ccws9Cl)7t!(ej!560#y3s18839FBAR#tc& zCfZiOC&m@$(!J49Ylh7&Mi{9z4&!7da<|l-7rk~=9lTdTwL47%PG zAigaGavyWBRI?sR+iH=OwU)Ed>`^m7#B)POu_h)LJATB&bpLFOKkA7MA01%X`w$Vo zj^dd90a#ogD(n}y|JyFaN8;5uHakFy#ZHsS#(EkH>%$9DPwJg7~TL zbnM$ZTFyQ1K_~c~XMF|E{fWew$U5l9*23mZHcqHGaVLu_COUZ`%rP2&tuoOXumHup z&z2fwfSA|QG5vlWRP3%GKC23!e?;*PLo#lM1jDtRyNv^N;T+_OhL7DaPrr-0HJ8yd ziSwqO&Jq6YcN2c*n_>fJ@=X}8g_wiK@o;Yj5}Klsbei8I94xSQ)x{=?;f80ImFp(jgYzSh32%? zNLBE|z3Lv)o0LvcmQN|BO^&X6snCkM{Cn%r#IjAKY4qY$@-q8JpYH9W`G3l2iBcr( zj^w=Q0f%XMZ86;mYar`iw<-5gBPqzW(o$O&%v8ywYpW$uGCYTV87*afUN&rNU;!H^ zy@U*A9OQm8d%E7Qf}Q-nk6HIPQRHV4TR%gWwNz|od1qZPefkE2av!|D0QwPbzHiQX5h zBPN`?K-)RT=l22jsMdvH>_wJtkjwN=*|H6NFWA`(U#kA8MTU~`6y~l>SKlYl-1rC7 z{?m@mnOMRWv2AjR}*XWX_rHd!sQwvXx!;>|=j@ zkJGvzlc{~s3`$z~oDy?8nS6ULOOHupQ-4-+9OHF1)@C}XZSST%5|w1Www_!Ts=_pL z6cltX;K#nHbZ=NDT?)yfiom(Jpv+wgo`+!f>JXC<#WBl~jZC~ZgBE;Cp^n8rsN=*p z7Pq;bt!@3GdjFe;oqtIc6My}q*J29Me8 z`3-D)Qxj`;Or&A2_t5BfNt8s}=+WHyv?kYwnuhwbL4)5gs<_3TR{W&8RjTMo?4mnE zr{n#X$-GBpik5NK^x=*x4N=aZy%M!#ekp@w6O*V_ES~aO^mt|`m)7#WSEuo4o)_Wo z!J{{rhgUM&Uwn@>iAB>2g-H6CJcS(cWa;E`5&a#|K(~ALv94SDm}S`u=KtdqOIljU zUU|mT<(pyT6IM%Vci&S`>u|a-Y!IbBv!bK3w&Bqz2?Rgd33tUc?8u*K%pxR`6>OY9 z`u{|f7a2s>oSnPiP%Pb-=%FEY)zq}^1YwyfXE=mY$*ToqSLH$Rt;wuNCx|r}b+D$g z9Of^V#PC{*WG|($P4+LD`|Za}JF}kErQBgby!-SdS45vz?4=)1SJ3VJF=&zGK5g-h z@KWqzwa$Oo?({$kEKFd3Q|Ggi$VO&rtck}w&o^7(*@_1V%;I$fo1S6GE~q!t3Gtn@ z_04tqEFn*|yjLT2Z6mq7OC;5lqa>j`9urJQ!C9h-q&BqDj{%R_<}CgWk=3M`)`KuT z@BukkN@0GmK8v}c&TKABV$y%Sxw9}CTf@_Ez<{%uZ%E^Pl{&^P3Z^ZouUThuFZ(ei z4WUlcuzAlmSkLo=eOwWe-sGW%`?k}S@6bWj7u0)V95(K@zz-!A^z(_s2e0EeIBx~t zrEqSP@AqIgF*z2w|`gEx4=zX$Ih zO9&AMv=RGEAGM1NaX0h`#O!yXCo&D+j2B~J&TN!L&BW9IGn@={MuwXc{P_-DDTBKv zKMR=7&6t>5WHCV-!y3&%vSH%V6T2iOoZfLw!^>ZnkA0*k==TCOBZ+*u6*?@{!`a zWUzm(9{jEk5-P(~g=(iELX{wegXiTzY5!6Cnd7M1p2Yp$Q;_4m9S?MtVVYbP>IPQO zp6wY_BoRc-d|oFupoIH0-EjS7FN`EV;mTB_-}Oxt$8q4EK(qUc#q{fvhi@V;{B=0>Dbu24XF44bARo1e227K(%nlbOPfDNaJ=ChjO}J%fCmPiP4|O^-etq&XK#xeqiNS3D*{ zGu;_0e7#`dYlm+wQ?RjmI)?t9$sLu}Xx&kOgtOCNu2%??h!wCB_}q0uD5NF$UBQF@ zyp_E3J|q{F)f;i^3qQA|%*2&&cU;^u3-hWo@YN|6+b4Qp%fj)3^H?2W^00A&Vb~-r zj$DSOD0AM;=ljfzHVT;jm5%UC$&031r0IQt&&w7xc3wij$V=GEGc)hBSE00Y9jC|6GhYHVZ4W!eNhO%2*dnmBAax>0cVouc^nFqcON* z@*jx}IY$|no2h8rSxT?DN)E2o)bEL!u*qFnsJYx5BM_bVN=B#zIc ze#l|uy#Yv^nT7D4NGO(NBTJR@5x#ME=CKxXQz^j{@i1ig?#J?&nQ-)Qhte5s}A%~bPdI#NnFdt8ijRr7jjV7VkNmrg-QhY<{OCc{+S2a7E? z(dsMzc$Qd+`**y_d+9+s{rv@14s9dznqa(@*n$yP2V+%m7tOYm!-7#Jgnb>17_-Y|_6-Z;-``5H>TLSsLub?Sc zS~%bGE?K+ja%X8b$^Se-Hv7h5kc1x|Jd?+K-b+xZ$won`H~kK*q*vq3;FWxu%*zb% zz_kDw+$Rt;uZynes$*zQGueF^3i-DtXkPi95<;3uX3{TudPf_h_nTt)a?YamROg-Y zN!V!XiT`IzO^s~PIg7hvwjH7=YcA8NtP6CdZxiKT{z}SL<8dQp3`(8G!RZde#-rI3 zxUH84P1D8My|tvXsGLpcyPNyS@*Obts&EA>Az{F+Elr)c7k^)C4H zUHpc3_bB<&06d)ELua`^-g3?wcn0zFV15oIbn1Pm$`@kGk;D5yu?L!@* z{L_rKtG%Mzs(Ls#+6At=M`QVx9EfkdN%KR4@o-r)`m9FNR*UVl#rq9;_8)}pzeLFK zS%}U2JJGRz1n*Dl@cx50*^XWTn~Rt6zVi#sa`{R59@YpQ;73AEAoqNp!&SB2^!LC3 zoLp^#?aC_DH93R(K+34s$eBLA&8FG>`O&Tx$KO8gC3~|9!(ty()26Fr&A*Mi=Z>VG z+;P;&=lNxh=Quw&2F_*a7<8$@}XOsfWZ+b)VdjXz4(4*~6Mf9e8IMljtb6@)x@(Jfmr6KppSGPZ`J&j>+ zcbW5wyGUaTpQYyw#tP+Nyi^QFOKUGpvziTw7mM-tzst1!q5-a!DnKjcExD}QjH&$- zA+hKw31M1zJlcYDuq459y^$johY3|WX!xg$=yxh;KH5Mr@eMR&$v_PLoJwg)#t3iV zK6kspsESJDO#FO^*9^gK;_u4%Gz2Ujf}^tj(B`|jIlsh^`h6+XH8Zh6d_Qf!qlo3h zR-+;D4ta0O;=8R&)El8jUp;nE$8^qMPnUt+#Z*K+55zx}4YWh!BE7#j5}#zvX!%9n zRSrH)&2>Eg7%Gp9>7$UCHi^{wA0@{F?KDN!9mOmhw#RSa;5|>;Zq!KU)AiwbYZT;_ z$70J<4ftKWNz-`m;=_JD6l?6~T;yf238=w^lQne4M*{19Nkdk}95M#8;Pq)A7Hy57 z)rIFNTIwH#{8vl=&o}1T{i9snLaM1SKvmcxJ}>8;et#c~x}J>EmpYj9em?#z;%wUw zl5~Lk_UvwGqUkPY#p+Fk+uL9q_TtW%(4(}WCZAlQo5{jc0-qkO!sNjjWFR&OG1~5s z<38EuRi>D>&oS6xx?XDo^*&WKBsrF-?j`a3>a%+!J-UL#q(%64Oq=B07toUA9WlwhBh(=V^3)3psi>Q*r!g*eh&9reqA9 zCVP=XCuar?a^@UJeHvcoNZUILsQmdQD2&;Um5S9^_~j?9IW2*rT~^S@>qkGC3DtK@ zr#)YuvKv)@*}FVX>X6(-Lki<)Z1_nU@!FjXKKapwct7$^mxT`Jb66!2{&Mz@Vzm`X z$t@!9u1Gvi-3ZC!m9WiMfvt>ymgxED|5^j`F^%pi{S< zVmbfE+hQjq{sqDI-+MGI@TQw1i>X}uC>bttqKQWx>1wF!$e#A=!c1R4+A#ZpQydZU=Y~>|N{&NK)ht=M{#|A6xx|W8CQ9ic4GsA2VX&( z(|*nYJwppUf6~U%YHW$VhE%_=5DvDI(k&&-oHzoHB>215U_YhFoF(P(<#eWZ81?0c zQc$_7P-UzoY;l|>T+0^VT*Y8B+6SXwRZ`x!G@5N3L@h?I=*JlYlyvE1(4!UP+ck%7 z$~BPkj%(DrN&+eU_u~57K_In-NHLGc=%r7|dZH4({N(q6jug81D4qHzb<*??{&e$h zDv75ar@;_IBhN3N|D%ZAlfIC+9&mq6(gZ0UhmytW^t8)Pw0vzGex6w#5`Li)Yg8?UZ}V#wjys9ir9 z>h~AHInW3t+O0IfKnK4&)G)6#9OK`_q1=5x-t5?g8|%YS@4N|lL%54V=Ne7B(nS64 zd1At7XQas4!twJlnziu^O*YG+JxWI@ZC(f|d`zHIRZkI>)d(j!&cOPo3$>NwfPiT@ zT7HIZ6&$7LJ7KiOD-F)C6R~Un=MM~bMp2|b()V~H-ANsq1q@r7J@C$RBu4BI;jyYa z9IHpd^>#an&wN6I)crfM*PUu_`i92%3Fk*Wc{`7RxGmTFa@j(NRH8YU(csbht z;~wW(0vf-Z;xjxseCc{aL4A!B^t&IN7HOdEmN^19X+rF9BZ&ue(P!5?MCV)(J$n>H z&&J_pTQ;;Flw&ONlQMhqHW11Nqc@ihR3F0(Yb0|tn)-@{a&d4Sc=;>lVDZu z1jF4t8-MyXT@xjf^#ktIJJUvb_KS#PtGTIxqNBQ+A-UY#v6A<-hM|1t{kkjP$LOa3o*~<{P*ovxC?}WCT?U?Q% z!rXywD1PaJb2fyPqSxf=c$4}T*U?U&bo8x1g6Sodc=aM1rab3fxH%f94m%_7?jB6D z(1cf_Hx6Yk!QiY|WDIbHk-Ha4-`Yd=eGaMya3_&iDOS}~kPMWN>?w;FEmNdwx?s^s zW1Kq|0G&WDoO-(!e}Btyc8ml@uQBKR_t|*1%LyJ9j(E3+b3GqMpl3=j7OW1$mFO8T zj1NVO`w_gW$j8KK$RD zzLNsH&XVe!$yhL21r_k%nW`LYIFbPWzr{G-wg_G)%#p>jm~xd*==iD!Ie;=eVB7DLujIhEB5*>!19ewpejW?GIPc$Gf!x09Y$JEG5#hT#ileq zZz(ANn)hLj_X*Pe{GQ$xl+l0E4!G;*f)|5BVe(@KpD_pU_s}}B)!}@YV;?v#trI0l z9=O?(jBWdm;Ye>S9Xwq}MQX?C^T*>f=E`3RmHR?3|0c1iR@F>wraN7_v6$BWa-)As zWU0Jeoy?j>Q=zyn9jh#5gScDntY#NmyCsiJZ_8k>u1ix==UCF+xR~-yvgmcYGTb^p z(3q9|FyXu-T{&h)wp;uto^u?H4n__q+X+i?JpK$!>cgp<$A%UITUKE-JrP3kFp{xDXV@0sXLkDEbo6tC?BQmC74iMhZ?_M9_X;M} zN%!dS5ANR_ahKM<;|z3-cPv0cmTV1lNC~%G%?(L*VJqxU|52ujh z(PVUP87*(U!@h<~&_Ly{ETqGoEc!>%pNo$4gwK$i9m6RxKc9M1q~SWGm41EchwoBFgbPaK`SdxrcPx@g_qzjVk( z3Q2`EH1f9(Op>Bd*fbVTm?lzB^F3F|4jQtumWEDGrL(+G)y40v2HD<_O5_=%O?ODu z{v|2JOeWde3#fnaN-|AdM~>MWsa)Na!oPBFqjooW=ZIrn;6OU}Ooj}1jG^4^_N4BZ zM_Oy;INzy*;@9Sq>x@_Quw)&T+DFhMn{>LUKc1GaRwcu^TC}WEjV4atKyxeu>CJFW zyjYZm`$H43SAy?bO_bq%OdT7;jA@;}62+ZZMn)0G*pWjGtiI?rD|Sy}zxNiiDUngk z>xdx@3oN4%`unKjkTF&qn*eg;*=5mHS~S!evmcDW-mBSc;`IaU>cu+Nu%&`(t^bi+ z$t{{tv4V}yDrAO5sm%J8EF^iBePBK?B*dE3HQ&{qla*zeY>|tjQz9+{@ zS;W?h;mha)G^F_(<>iM!MRF*{xZ7aw)4f=!ewc>eE+#vNkMw6yBsKgpMjh{P75hG; zWB>lppBdbPcjgn5ew<5*W5Rg<_A(p0U7kwTIgml!F_N5SijP_SuvhmO`E0PmBlU0e z=I=CmKcI;Iyh)^}2cBr{{7&=r?$cYFQ|xCS&%N;+l%M6;n1?FqS<9Gl)v>Vc@^uR zb-e<5OMPgC`C77^T0y;;V@YSb9IdTWARpTd+TZYrY&f51^;1clvXtTOD&AXhyFew| z-_U9wP1y7OfyF$Y(Y?HnN>v`P%=s-WalwcJ}KE{~!|ZZLmPg>Q;2 z6a|ls=3s+=r{EhSR6V;*wX_n~=vVNXQvV6bm zt$7GKn;v0NpaoUdohIdZ>16v`4=cuOgiNaeI>LM*7kvh=0{@_Rl_h%DE1}t%Gvix* z=<@qu5?g+Z^lojY^e1_om%4$Z>i1G^G(R`yG}0ZVEY8+BjkQ}!&=68g=6bT+-z?zH zhtWv;5(^`i2!HNw>Kyn8J2ft#agi5yUS;6@`Aa~1EoS<1e$kNzH1Ft#n8sl+c{~gV zK29dm+(R^k_dNM|b@s^u4Dm1E&shnj#9kop#!sY{7=!GBFtqK;M?qx~E^8fujpi+M zTn)qgqnmK`ZxMDh4MgWjC-kbSLAt5|b57sHfpwL*)XLq(_H~fH;mdtbR-{^4K+fEE z8~ibs?&(F6^MeJHs<{e%Z)afafDN2ivxQ!cJx<96H|WA)eh%$BN-NAHP}8sv|8@A| z@{Lg3HGW0jd}et2s(>|Xq@a30nfqa+QJpV=G)XNCdt<^g(pRzl{%S!t$4AKL`ye;z zUTh92MXcO;G7CCE>OBf5bq^+E$x3Q_5l4NzgJf93dqZWBbVGypmgV?)WA%D?neyIm z{UAt{&Vxt&N(2;mV$#DP{CXcq^OSgRdDSvn`TQoSTPM*&ug9co5kUtu^vJ#_h0a*< zyv+Bph)SP_|Hfay<^{K~a`F-MxAY{Lpj3L;u!|JEdJ?)VZz8bn3m7Pnak-| ze!Bw$LhI3Vvm3s-b)@Qjo>E8l(fraY82x2G*7}JH!_JMv<#UEO`!N+?IiDhRx;VZM z_(_p<%lJIml6pc?$VsD@?sdrF^?3~}b>|$96Z~9x)d*K*YG{Q<5nV26qG>nZ;g!ut z9N?^t^3)c3w_O2Kjm(g_=^F3TzQSR-KD1BKg2YW-^m}fGXUkX8W&1rO$$2Vsok|c{ z?u^!;6qtV*Bx$xF`9eW=Nii4UBYmb zO5CAm=~DPCWrzEL~fi&f;fHQ=TL8 z=g#V3a!8Bpr4sv0i02=Ko>D1yy?#Yzvb<2YjfBEscn{<|6u(4lIM@W~ z$1=FqFc=3f@%i;)Tig#{kJOG(Tk1M7 zMhgMw*5O`U2&As>!|USD+^PN(|JtPlg|m*h7CjSTlFM+X|1W9{nGS8G(Wt-wi#n(E z(Be2lY}r~0zph3ESbfJaP2L|IwhfzCWbk|50>NX_Vxj+uWkTE-XS8n$#>qu%Antqx zyA%cq*2m0+vyVTa>DqJHKvl3#KFT}!1B9{<=P+reB5HgUAt!5!Lpjm-`DqcP?GMAd z>N+;ekrp)n3=rPjw-h#aTMKX1A=ImEL*j`jxE|*@^8|H4rN~hj@!Ul4QGSRnKLx>X zhqDlWd!7(HVV-b#Z-h|);Sw5+Ix*m)t{}E|J?xv)q1*2Y@Bg+VB=HX{1`QI%UdR_F zY&|KMDP9*Mx<9}sMM{u)@EXHArG?C~(t^`VBVk#`O{j|0gq$Y|f^Op?A$Hh)!Cxj( zc-&Y9X&&5lK2U?1Z5l%TT_2%K%T)+Yc!iE(x6fw1jH-=30N=ViJUhc@GFY}7VX7S{@p$iVTOT+Cqv9-3PvdZMgGbTl+>9C z{*y1E!}UGiDL%)D1*(WQRm4Mi9mHfW7KWJx3px?&g}VG=>|39QwTTjj3=pwGrwoV}(!sW(o4a#=@%qjD;7M^@XqMbMV%HI}m@zL*0HF z^8Yi1SjH@9CkDaq^BXw)_Z7dLj0LHHNy0oM8(~w*bu|BLg_Khs%C|j4P~2Bo*oX_S zL(e09SsOgga%YJ7MedFKipe>O!kkUdA%09uaI*gjpYo{~^K${F3iik{87$n-wG>ue zSRt?mKjF7Th>*6_M+h2YgP3WBSoL2i7RJa3(i6>uyM5yX&k2S2TYm)IOHwgE-GX;M zm%#9g2M(R<#jFDg&7HZg<);+Xvs-I?}{dP@aKPRav7ww(h+++ z5SG7A;^52-l&v&^U6Kr>ueoApQJUbqCq?K^i4=6yt%SZ^5LC3RgcnyNU_4?x*ulkE z#os}53p_A-+(dkjvA|@x=TvWYpYpjUY`xnrDjj>0rpOG%?~GUU*!~KsI!I&B?lvkP z|AL+br;|hAX6lsNPW6{2pxaQ5_h1HKkoGaOuAGAr`Q9-4s)QBd!*Ju#D|+8K6g}%X z_gVZE$&JqC+-yrKeQinACpojNgYWyhS74Z2ImFAj|D`(<>S4d(9@}3y;1`Ec{;LJo zGnT@$%_hS0a)v|v+?ZYTmQ1635PQ)DacUzG!p}Em{fbFnojZ`1htr>Tg|s|;AzGzGt~CbIU$yGi&U{=$R+A~ z_=Gl`0OnPWf{j=ssfyjhyxoPUZCs0iDmjqMjliMw&2SkP36;rwF0ol3Elt&!>Rt%- zcw?lLpGV-yIS5hQ4DB0Jge_sSgsG3m3g6P7(xKlUNqi&kHWf!=$=_roZS+T40C&qy z{tuBd{e`O!^zm%z82qw-LSCX+IMsLo+I-LY&P>R7YAC2?OA9Tb_fSxsgg(>7m{_zA zZRbM~+|Lyi-REEu9?!cMdieXL9)H(x?w_*@x-PU(@Z#-cAG?Ys%{B(>{7Fh@n`jly z#$HQjC>ZPGxC8er?zuyUJ|)p~V?LKXd4&YNA3b=v3onI7Fuj|JsQ8_b|7eMTLS0xc zPRAX?Fu2P5VuuUg6-W2d@56`5c1)b`LUOM#Yv3+{{g4pGYRC#k-pLr_s|F*9r<8WR zmXx@wat80C^k}UodYVB`f6O7@%WhQOFcaUen&D?3?_|}7Vq~c)ddA2jz$X))3ND-p z9)|vX$Eaaoy%>pVMGre)1=b*J{JHcbOQQ!Mk44hmqP6jokk65XuutuWub4I_*R^ zlLKMws}Bb@4(2k&Sb1p?%D8j1b1Kg!TvmqugfH~5Z3pM$sSCS1hYI<}N8q9UO;Q?p zoZoK^h0F1yh2XZ^7~C%fX0zO2x6cNW>bFQI?>e;}>?HB2c}QtXLBf7Mk5>5zaVy+6y`|mOSU7i1Ekn;LpxX_`E-iUmyRY$1-o}j#(Wgo^(Zr zWD}*1>Lk6&3s|zM3FD?lpzKa0W~{g6oHs2fyFa2~);Fk1-ilJA!7DN;EwxRtl6?2OKR*;rnn4if_P3NOBr)zIp2KfAw0YzqOms_BQCi??V7G| zC0|i!D0qN-mky$A=VTmq7qH`zvS2^-Gk0-Zq-pS{jGi;D|eFJ(Yrx^-sZOlfxMK-U3o~ys!6qE}l)a#5-Rltnc+epyw>eBvoM2 zjbh~dEJ4KzTS4%Cj_G$zgiD)cg)F%ugs;y)L*Z=1KDR+EKN}hLi@`Vdjod%B5EJi5 zA-}~R|Gc*0tM*2uMVsK-A_F{U{JX#V3S3We&cJeUL54d#&L8*#=`PO9d9nev+x)Pd z^Fyx=xB~gtz6iG34Htzd&iUMf4=w!#f0K{6vfu`6_kBQ7oRkoiEF&yQ@k0fFhN1>6 zfRmnvFh}MQO5WwbeSZa}rCfvJ2X`bIy~VBO2T+#_ht-sPEaZKKSO0Cr;$^>(eWeX? z^W*U*V7L$}`i;k9+c4EO2GLD+P)l-z#7B8dog$AA$$zvn_dL&W)ZmeA6^g!nLGgtK z%rGxUkzpx!mpVYqzz?ToK10v$9{NAx{gm1uobNX0{2ECFW%Y3m_+?y`yo@=o_hUe7 z5bix6i8oi5q5FbAE=n6>Lb0qc{Aw#Q=W=Eh3l_T0jS@Om|3zqRDCDd*qG$6?DBJ$V zTKP_><)47eAUpiIXOGy$;>bT$4?VwZ>|7s*)owgy=}6HL z0)te9@%+3JCixTl4MUJLEfObot>y2JK2f>MB=(^(muA>TGpEU_OgzhqMirEaF!{7- zgsd1{yer1^O;y>*;$n8d)tbF^6|wgfezfs}K67)|q%!M88di>SIPNVedKH?xViq!-0GEG@Z#U0-KIWp8$|mc~O&M*c1Pv+xQ#aP|;e5tB~~ z1R1vL=u-B!Ae8zYl3@w`n?=L)U0ADIr|8^_aII~RjDfNYDw|zLPeG zJBn2g4rex!r$ z{Ww#8S;gFMlrWj@Lo8|7QPE^MW6_!pbGGj%80XZn$(DC1)9S70M; zzS=FSk0@uU?_F6(*(CZ@e~IT-2GG1{8=BN*NNP_mh-#*P7H!|_&pz9}5q(mWVCAwS zX-CXG(R_zSQHM?uyEfytsKn5k)yl_El|-m$_2qXWn@68n&vz2_?+F)`U$kX+&kSX7 z^<-JcbLs3%b!IvTYYr+(!#s&M}{yo9yMX0i=B{T~vLm zSS0!WpXh5;GJ82#f<|kOqJS_D_K1};TVD^V&X_F9`)kftj=#c+Y}c_TjmKn72Qv&;B8O<6@-N|rJ61&)%&MYo` zqfq|$HOx_rjxSH5>Z8(Z$)qmk=I=q8B}+w;lU9hvOnAvOgWj`cPUTExa6Mc2XQoIz zW;h$#W=%Vzx|x^TedZgnl;#v7L}eXWX}fmXEF}TtjlmG37`IBrlXZf!^o7B z?rV$wTD}&od6L9zyf(4?)Nia|MhkhEn~AFTz877TPh;8HzM_wcH6khJ6--IWm;OD` zCi(gGwB)Ols7-dc=-$8G?AkhU_A%rPlbN@d^i^p5 z#f8j2rIOv8ahPXP9&zU71q{Dm%+I5@L^*m1qUE)6lzeaos~>)t6{e@qhDjc*@YiqA z5v8RpMk9w=Jdt76wozm?Z@lP;stQw=i)KF+TSWU_%CY}(bl%}qzke7P8PPyXDk7yl zInRBc(xg44r7bNj4V4moL-r~md#gy2GD=gWsfD(7g!abC zV(=GtEPSGiu08)ljJ!8Y?4b>dm(FMz8iCrT6QG;;1zpd!V9$x^LU&QvT>tf$+aoFo|$rmE}=!PFo-Q78_wga}|tCJ1okwy%5kz3r+TX5cijaqxMcR zT6i=_x3vZLPi#xKswrHdvW5*6`OqwpJ(PR8Tp^nZBMy!P>|OdMUCm ztU}bRqv-$qF0O^nW`g5Uu8FwMP0{rn1Pv3b|7_tbOnx15-y&J z;`w6_k=SJz*D_qPO|H9b$LlIZ>EX0>`bN;NJM&D2q<~36=OI>pK^y)~i| zRy?pP6)v5oBW6_zS~lv!*pW*Z-9~a|yFY=2!jNn8x8cwBn=*rK$@;D*MZ>p4V#yUv zl-WpL*Ta+WDr&`_&7KMCfeqsR$S=aZ?mMasxCC7F>W5lOmmBBEChUbVKFBxK%Me8Jn>EwK2-AZRRQKhL2DVwvufmTVFbs zhqLvxiCFwR3(oG+gRy^SXm??vP)k^?;JuS$sDt9SO~p3 z>2e>^l1|O1qOX5A{AKvH~s~rN{FDb~K2e#e)^Su&q!H14oA-#-I{c z^G{;XLo=E=Ovf0Tt=M4hgB}a|p-*iXcHVbE(`N(GvrT`TJ3j`4R<^~OKh`jwH5Mv5 zC0LzThqgPlY0=plX~C!9WE~9k-_n8HcQ#wR_>Z4nG{=_X(rMlAudsdFkF#Y5xchn~ zU%J@v!EHU6iyFttZ^|$w;V-V9(r4OmUpDUbjOV6Zp~S4zG7@tnB=HF56`cpykH#vg>*d zZWC^T#=Y3#=U`47Z!BHH6S?=Z+;vaSW9YE^oU#8hx(%(sYPoNB8uSDo>Zftiq>Y^X zBb#;+eK`DB2o?Rj_^DGkyE(m<{-i!U;WvqWn@nd-sv9pC9hQ#jyL@<0n`3fh&hU!t z8+0FnJMO7y7Vd{F3N6&!Ylg~q+wrDLH%9lHz=r0V`8hxxPKWy;a>D=|)A7U3!_TqJ z&6ttbtQfE`gzbAu*Gyq+{*ZHxsE%H|urvW-y^^rRvnlf=!>0G!csz3th5a*K<@H&5 z%2k~@DBC)R^Te6+R6APAS9_jfboo=b2TPyx-7jKq`XpTFy%c7XTEh77X)(g*5%lazM zFVE+-Rcoa$V;o)EN~ingiyZnbog@1_mYF^2?ps`j8-A)h@b?KVRbHMtsTcxZT9eZSTakSC2`^L9kffV;quX%G??%Wqwn_M zhe4aJ()S;%U0R+u)vElNy2 zA~o#2xM^D@rlb#p{>^oG*K-!e%^u6wl06)BIu1`33x1lnQ(QeCEQ;H0gN9-?`eF*M z*Ce8s?tYnS(t>^TLIgbBDNcM%6hAkY;`5lD*l{BlOHZ`tzxik2n{A3!r+=Wi+NeCl9hVvu+n&M5rRnU{JxGij_EIztn1WABU2##N!L^bV9Qf{oxR=`wK97fS z@M1GKg%7~0;NhIVCq(RQq>GEILQu2US#*6>DyFniW5|y}+~~Xjb7jA#_E5gCT5?xx zJ=`15{u~ffL-eqwAOMH;F2ZwB62$NttnTc9hW_TbGwc#VRpO_QGwPGn|wmnEb&Hj&~1;uD7oX)e`~I z?|2)VCmz9_!P<1WlmMRxyAi>r4BapY?_Y9=-;hzt5A zhUwhHdWr&0~MdJ8?Msy!|4<;JNpfcEz+Utg* z>6>88h;PQQh#4Yk&P8E)aUkw}*ny1M({br_J{JC}5*N;N!P>E3Q1dHLtSPS+cYQlZ zFG6QV7w(etLPy?g7%6Tb>wv66gEMoLG7#+;Sm!~iass|7DQl}bn*?T zc7kGdbB^jcST+v=S^3SI`HuTp!V;YIibR({=_-8x1k0zslRcz&sB3%<<+oI*d8iDw zy|!c6g(q0IT#ak2?C`E%zVs;`$II4#P%!f)y89V1{O@vJ&0WK}3BlBjxQ`y6PeV8V zH0s|6!qapl%7exrebP|8UFwYcc4HCOJq*FquAm`L=D1Z1@x9`nm^`u__L^K|QCJS6 zawY3!_y@%PmL93Eoq2q;0ty`^T70#HgP$eS>K5~Q|KZG8aT+;>2Vj$Z3=5tHW4T=z z+zLqt{$DZ zrCVEe>nFQiU)S+s`dC_d$nKl;N}a0IpnqpCevEhpwYOI=yhQ~L-7p|q{DWiPer&F% zrVM}bmw)Qil&_AO(foQ_S`6yK>Juw@V}1nd>I2y@V7ugDd$Y23k@V{1O9w$ZojN6O zq){-EnkAw*`x=}(Hs_Y2){MMo$henoeEY+jyADNj)tht3yO@t&A4_m>NmKr9{~D35 z{}7ezzz#+uct`dG&Z}2rvDH)ba(yPx2`w4Cq#Gv~OJ4iO6eJbQ#=W*s0CMUQDI`I2L@}%wNTy zaO&{@2F@O<(a=F^yD@0oF@+^dJY`=!MY1ewac)^0 zTep?3XHkU8#57$0@ep<Oz6&x#lH4ANQiqT-PKF* z(PtjI49~*CfOB;Cn9GV0FX>$t#qVCb=zTevPU|A6w@B_!wf6GjlO#^?)}G7mmeb~Y2c@@;+-djJQU1NuQ5ip~yYkKg8?LPH z&wXYK80`}$-FUI|PqU_HsRqZ$3{*=OdB17dgMaqP=g;yM>~q49w^}rk{UR;4uY1Gz zGc}AU`p!LnxAEEP^&H+OKf@%+8E%J~P?+0Ua57m41?v%AQk@3}-#0j+1N zDgPL#D?{>|D{FtR;-(2MY}$4=EoRi?wB{SQyEo#UrE>TDQ^Ak-hVjMC75sG|iQ7|y znf9ZG=5rr#;q-ePvr=-pr);HV_h7!3%;!hy`Bd>b%Y_xy+`3FfId-3gQrA&S>GAtH z<0ds$s;+9S>^((S`C^W~vP?4WAEl)5)!su)S-nl>kvw=dC6FbhGNyFKS{b-zyt1xl zx9s^wv&X_{PCp=j2Tot$-di{1-Sr0>y=kO;@V%*W&TL&}{qZhJt;a)@9%EKA!ANq_ zj|TCoiez;4|I5Cw+A57-=h3fFGAW9p_~+|2Y<(&-5Q7RAcENo z$8yWJ#_W@?`s12S0 zsds6E`(cpl5hUl#1M|&@p8zynZCfRqkpUHpS zZCDNu?(5JncW-&*q8XJ^Z=;i9!(*E}pxai#NUW z#Y$bJSk~4;jGLn-cG^4@in>jw>fp;G4y1Sz9;}}KD>TM$Efs4?e6E2ct#tKJ^ z6GDAQrQ$K0i+T5Vh>7z$h}NCe z^5;4Y@%LkIF>Zvbc&^b@jCKtY6a5`T!SgCb=1eOQYjZ_Re&Z(Ebf^?PT5lG4$IZmm zXXe7AFh`8)HX25AcZq=L9(dq$Nik`ap$L!bAQEacMHium>Zgr`>dp&_#T_(7&uv+X zgS(3q+uAvbE&8TNKivaY+MLI-iX9N6JBr=Q8==?JN8*}$f}$Z}l8AiMRg~5QDvGZ* zD3X%R#CU_j3fJF@#isR%!oAZEbk}6BkV=% z`~k@KzM~lBaU2orM~aTRu4ub!3MO+kS09%%BJ~QcJQgqP=L|t&PzcP!%|yQ6ZE^me zgXpl%Pi$*48&meiz`&=MLi19&BCN{aLxRO^SD(CA9Ppwyq$uGRb zw+-P6#lGg^;XENGe%+}U_sdd@IWSLHeCQ<_dh8UnvpmI$vO2}m*7m~X`yX*AO9v~> zlf|-J**9);RMgdN5s8YSc(_v&XV_WHiwws04oyWz7gzD?S9f6{+C$?)I)?u0!iOW8 z;q;y%Sk#b>HujqoK1MSXw?ZB(%KPXFtNZiBhTxeZu+3b>rLBR&Cf`$de=k%Rwf+yu z*JH#6lUCx?7y)Z;{Z8}uUS(GR~ zP4GmE$`!JY@>!fImuInFmg0bu4XW40;ZOfb;{MD>V$S9`G1;q^u(Vx+j){`9dpJ~4 zy?TbQd$d}V*Cioj(;xJARHMa)k&2tD3l-Itm5RKvI-*(RSB2m6BSL4GLh-_Xx)_vS zB?k1{tT>;%SMjxBnn?ZP`#UgX-Z6n`>RMAExu;)zwRNX%|RlO5{H<`F@-_}NvA*EYqt6XD|b zTTPK=vqP*??<AhWpqW{PS+WrahV ztvI@39(SdN@IplbPuzTum0O!|@BLO>`8t8!)g^;)T>)p5&V^>$OgMCiK*yHOR6pWH z*V+(vHuqzb{3FbbN@VM)_ZhAIf>}4sm7D)Q;f20usOIvR-&MW1TlQ;T#2n$n@$=~_ zy=-Zl{AlrIKYCc7l1$eED4!(2#>)}vvai%V-3vZ8-k2J+8tYc4^I7)_u1dKhSsnIV zad9~NCr^-@oR{3tyOA>Xi?;Hf_7Oa(&}Nst|1kSlB4Z^(e1_^v78?a~x85n~sy)CD zIbke27siw?2dHy98oU1`;-1PbxT?QJ(;ziQdZ<%9$e6b(CATEqk*6??)#*7*Sa6hI z=M1K(>c!bsmK=O!1B>fD_}zFH_mzEPU8TMg=%l!fMC1W)ra4tum@@GTO zO>FgKG<`O2=UOc>4wgheYJ;+c}Vu~8Yi!# zok|F|J7w^&%<|eEIzr!r(Ol6-LpgHaTiVwK(XjJg*33M~63gd&HcVHk=Au+qN6%x{ z`PR}vmcaOLH}TuM90LwXcY>J-dt?}M>Z|^A9N^3#F_K>j&T&quWZsK9RDYB16blvJ z3Hl0)lkHe~|0GL?o}iw`4>lVnom6Z6WWOMfuQI!FS{`_`jXdYr-G)@ivEzfbR2}Q8 zG-?;G>@WP3UDto+wDktcl`m|R{hh&n4}JK|bp|(IUjXaMSiBgv1ctpQ%ROi^?B7cU zz^r(d7^g7JK9jF~e0cFsD2K@&?S$?d>FT|cIW|eGZ+sXJ>$W4>M)LGdPv!LeN?y11 zk#p*J%*|eb{IkVSy&T1Ho~wAWLoUl?ZnpbH$!DE=hKr45ckj^#?6G!$VO9pMU2@oB zVJ&lH-m3e7TsDk*%3qCil{I7gD?22ORc;EE?y}Qz|E)Kh+UMFTZ(M1s)Grsx|Bf7B zyDc|pV)cQxpZao5mGoFjzx9Fb*Ra3aaSV{%_Pw#o*kM)}+cjCv$8&7iD0K&$jrOI- zsKLDbcm$tiI!PYFDW+XIAZKgIG@ZMRMiW*sajiTXo;Oiym$Xwh?=eIf^v0H3_HW~k zmC}n=(pfTI+f(Pz92#0Au;H*LZEobyy6+{<8yw98(gE`qjLLoN+amaR!a=M&@(U^DH?g#pCZE^M=OLTHEDT*ny+{`}N?gZ{#~0I7 zI|*}c-jV(&UsNA%q|B=_P@2zDQKo$AC!G?qvwyq+xxpLxWUypMo)|9~6IMJXnZ)HD zMVPftm)|0ExzDOGZ=9ET?B)LuF`+5*%Nr@(0}gO)^f3-uKZFimWBA%`G++FdIpTxT zEn4*h*Dq#bmU?STi``k8=PMBhCO_>l{q6mQGLxg-ZB1+Qn~NcnfM+N z$)C{vbz|Ois}b`T9T9)W-4#iP2jFM80hHPX(3p`AyMA{tM($kAocuU-mNz@t#4~h% zH(ocBj@VDlsPZq383uOzHq3zq?!Rzsi(~@&Xfe8=w{lR^Day`wdMoYbzk$E@6j5gv<|Y= zCX(y-oMK}edHOGyM`(yPW9y3!iuA~s>_9kriPLFQm7b~l*J(V??{gif= zf7o9f<2t!BZ6&*cYa&lWbOd@|dyi}{O)7ubapSF?k~`6!)BAkHnauWF(sBh0Gg>jH zGK8wlLz%s79Y2~|(?aHKTPm$-s4jbpD?r|Q|K1NBY@`d; z*pI;jgSohSD|A0&j%5o}p_1vsQe_;^xH<9DZdY`!nu(^a!w}j&94$L7!zt}O_>kO8 zIXvz=_2$=0uUa9-OJ{BAu6i`Ptgh5iwp4!mtfACT>d(F5&NONA8!5%z*kpkrBW^Zl ztjwCMcpQm{mC1Nuo2oSZxLJ8GE<^c4XFtb8d2?&pP6lKQ;u5F!9Q3FicMo65^n2;l z`#GIYUi9YaWt#LI(nJ=*&%rZWI!+eVz^BV4xNnHTR{3w(NpCrqk(~5YUuva3MAA4H zsF?-gr?wdzXZB;A9_W(e#pi=3@xfAC9=wu*VUyA^;9U~Tj9joNG6dG!rXzc@JX@@I zM9+!0SY4mRyx263cyOM^IrDg`WF;S;_viV~-*8{6A44{oOCI3>$u3i+XSI%G^w=>f zB|th`$MSw+KPERd;rS0n)UItt)3DYotnb88({{_Ql_SR#hB3o@Da-t(GCa|ZiFMJ? zTpoc_DX~y=X~zSsBrFl5YM6t(&Ul!st%QX+IQwW=Fp8-V}|h1 zSVIQLz3hU@9Fv@HI2Nrc@Ug9Wf6$i3YY$*Rk#!Fs}meT%Q zd!=1o8)Zg`E&rN!~`Su>J*V%?~-)3NSbrG%}E#kO21KKt zvU^^F7CUa>>E8}K`jJ$8bYaNQNS^Kz&3|1(`F!XDtn|8zS(Bcid!%IU5A4FVMSW>{ zDTIHINR~oP8kgr;aZa<2TyVG#s~efI`@SA5FSnvY;uQ1=nTCxY{V^x_3^wRy!|~ER z+>;EbcAjPMY`Y4JyHDW#!Bc2Hbp->Cc46d_b{tr)V8&&2=Cv$ClLS>>&eZ1NLRFR- zHQ;Ts;K|P>GQ-h}9?<0tNan0jYxb{W$S-3@0|(#nio7+%sHy>=sIw-|cWDcs|059yQSKIfW08tfOAG zJQqHl#$%O&=jE=p_p!j18&)6!~yc08@Df-YNJY-Z*W<52c*~8(4U*V zoNz;)TferLhtVtjSXH%?y(^@!F?KNq z{}>LnMd8Y>xeKnl-Hbm~DE}E(9tDL4$N3vk2wTnl>swJ@8I2DFt$HOY!1QBD(g;mOX@vi2z7_+)s>~8l7 z$HVr@TtPBUf1ihh7K1Re{DW}nCi`>?p7Q+j40#q=#CPcjpy;K~K{dm%cZw-Cl)e)m ze#-B|lpb7urXyF}aX?I}1vZW9hJF(avHj%-5ip@I^a6D7u)I0)$J`gH(kHbs)m*xV zCBse8h>fP4#52EFsMytvKj+Ic_>$Quu@6P0%+ybr;)mp(;WDqZQ+Bu#*?CZF$%OHx z7(J6NsWM~LyD1Db=HOu0W|;TF01Y=9V@T7R_+haCz_#k(IzJF0YD2Yi6 z_HvNzHqLI+9@#Crz-?$}T%`@y9W&;jHkQ2K=^Z*rhK8HRQLHsrWulHASNWRJc~d^G z)E#1*^^xqjq6Au6w_(}G8SAh1=egk;9AaR^oFn=4Ny+BOdl$L9sv3H~eDQr%I?O^{ zm?v4+4OchtS*u6}K5}83p62X1eIsAxf$-TdK<37xmqvD0z3)VTumMcyoE24RXKBB6Z+fMVA+bBm@jkgmtqdW zIXw)mcBh~zujA;75WGl?!^2YtP@~)hy&?P1c$Lg$*Cb$s{cEiLnvTz#H8@4R4UJKS z*#&9*eO)?W2fn~o$Aj=()tv1^v>10j7$$ngST*CG2)d~Uhse%2+QJ6@`q?6C$s198 zQv-)~1!JMP1r85tgaw+zxb>KN}utCJ%+j2jhYD8!@VxG5T)!2!pAw z(Xh&x4u^EG_M8<=gzHa%7aFaII`dh?L1D2e!ahmc|Wz#YvUcP zk)DJD=3Ti5PjPp^Zv0)i4_k(}WE1~O$Vxeix&5Lf*Ze$kdfY>yVJ+-8Uw~SVGl*-O zNpIalRBYYO5o15WM`tr07=&P*)pDp_EW(*Bdth=_Fj3y!2KiaB)?h4uUnaZ#n#2OH z7ih>xKuYB)Opgg-eH3|!b7atY(fUl`9<+Y^Svw`v7HUEO_j)`c(xzT zj?aLiOc!qzgRsa(^8RCI3irj!#n^kgxZ$)5Y71jfk#HN^5=?2Yv7MnNJNVabw9HWa z5PI&0Xe0e)7cTw7w};UPSQ;ZX9^54!b-XRSUdChTAy=F&KZ{cj!r*h<6W0#KNEXjQ z2D>HDaKHl&zCVbrjZIlFNr$bk+Tzj5B-r2ZhwI)1aUw2Pqb(APef4<8#I&O;~!W1(k&~H0ohkEZ*vFs?Y4vi^vcjZJ_F+`B@@qUg_xEb zBED8>%J(E0qC<{ieZB1a9B#x>(rM>y+=uth%30EfNG44FkA7h`h*8^%m_8>lVATPb z<<3LXyT(*HZ)bi+3VWq^@V~RWIAB=>^){7o;m=Gw+4LPbtLL30)_y;rlbfvGq z5E*S(ip~zF#5mQX?EI;YV>Yx^Vxb<*lKas$v@dfTG&y2j8-5?;!#jp?f4{(s0~#3A%op!3$(+n#$)mlphVED9a&W6-95zlVdyiwe$T^;mTIGxW-)n`B zba{5_8vxtLb8xVbY^T}(#OtjxQ(ZR!o5nk%|6B(|tKP*<%R4x}_%2-AnzC^24LH58 zgQ1ft$6Y*)y$fowM>2|@n)@Sj_*fj+l#TTj`?FcX;%KS()#{PB-{-UF-n|(H zrggz@%XBfKb-eiKqKe(S`f=8^K{Sk&=LP>raq&^QSkdXaP`+?xj~)N9Z+a#7cKL*7 z(gnM;eS3bDdC9ACujf1?hLa!e7sdy3gz2$6V!LxJX9R3#_?K|1p5D!t(voPDc~s`e zdmwOaFq+Q`LY&h8JU%)Ey(8RlU1gV0Jhm6jic*E&b89}fcjDa0L%cFS4nOu(z~PM= zZ{Kp{)XjbAJG2qsy?BDXcb{NlWjp>otHSZ}SzV>pnXhpY4ozF|)I2?AW3^~e9U$Dy zFN*807I4CseSFX?mSc?ma)5OM`*$s;`ozQd^xPeh@6Y4nfVH9~;*tnztPYRLeVF$< z75g?xMuAnds3{2+_pVfmeiO%Vs`oe^-F28}=bxi+d`qw3%7A3(& zu&zIfDbv2AV`Cv5ecjORqAjeBOhP^u!}a4bG(MirU5Sx2xYEE^-+XxJNEl~*I>x(t z()D&P1X@lzVZCTI(ieuq<@+?W8P%Q}p6l|>oS}3$e+0+Img0QlVzjQEfFB)pz_Nn{ zxl~Kb5~3cFvD&U*3muN+dj1KZf(^-l)-AfZ2wQNEZdzQS}rW<0>%hb0pq8 zxQX%uk5P5}5zKeBV8Ajp-mGu)6x`U zVy4Qx=pMv6jYs5fKWuY!L!sR!xfAonXM~$Mp2CJ^CK|zaShfUNs;q&$*@cusqA8LmJ{2*;-rBoEbKI!Z*2C;c~T*| zC%2Tdr*>?yy&k(QPUGtEtLXC}3I{48k&tu&b6P#cY?)8nbm|;RKKj6*JObwPBC&W} zEIPR-;i}}IO~rk@zgvr;*0bF22Twc#|no{3=3dS4!2xQjyv+w+8#0o8J6)0@)K zEt$YNLDGXWmHWP=s9>@1)j&u(Xl_ z$e*^Hp}vO&*LU&x;v*b5eh^FMe8ErmJZMdo=i_FV&@;*pkFI;+mUKPuoxcJGjWlSv zyA-9fufr_agni1!NcPzr7Br;c={|3C^IL?XDa~2GwkIt^qcM8^UA$kq14{>;!Txid zd8leEgStru)U_05J742K{UqK$w-3#ios}Mlzt|_=SLO6fyzUW=khE?rif_qR@=RMj ztEu!br=aofm#CUlD_OHWr9)PiamnY=bc78}I;>=cr4FBN`HhEeDh!nw>#WsZ#j^rW z^z7%2{3pZlYFSH67!@gIZ_N@~1N<=Ohzjpkv> zG?oaqIw~CUV}+Jyt=RKb1A%SS;PfvGV&Oh5?2&iQ_z_v6c6?i$*2{o_Um%7ant-CYy%?MpkG;L*Gfeug2DOqs z>qU-Ycg}8j$n$FY6m5QxnbcvW=D520hH#6!E;entEqt5b75iV^7a`AjVtj|IBE3nW zn9=!&C~?w5#@9*|I_}4t6WRzlZ-><37sZ58WAN~@CHM8J5>_StA|*LRj6SH4yMb?F zZc;YlFFzKwYHrvvRi7Jv8e-+P#>g^j4xgUa#Bp!QFbyaW2lrbF&*)jgGykU89C20X z_}huf5F2d0PBipXayS*Jgba0nheeR{xtQ4(<-0_c_>VI0Cm{Op)h5KhF9% z60gSjqqE~*Je})_xUTzQYC4bBNhv=MRmRz#5;s2%K)-`zo3b!G(v|Q3`(o~CD&3)Zd17>r z4EP^NLuBzV5;n1o|oVe*B?l?6@agz(^H&usWi+5n<(VsYc#vc

o#L z)LHwX6+eToi%Nyv+lgY`_X=@;V;syj^unAgLC_gsgauklr5Dan@)#5Gs@XUE(3(%3 z;rW=G5rt&?HPS_Hhr9ca;%(tIJl?$$`6f*;_3mlZN@l6$t%qXqkTp>M8X)p4zAJh+ zsu$Ov>2pZlaKw~0;1di{Wpz{3OK14%wgn<_TW6v6xmXO4cayewj&oaWcsw#!EIL;y zZtiFhe!unM+I|Q0wB|AXh!NU!JTK;ozR>6 zo25{MRdM{S2(fk7XHn-KhG)MQVf_A~a9?!-E>kK+&L#~!3=hQpWqt9nOEYLL{U&ZT zS%5P)zlcj3UocKh1F80_VS8#2%|iF#Qv0{~(%cG*p8ga0JN<-p@k$)ubOBvPEyK$* zmRNISg$Nu}Al!odk=rd5U2n$Xolh}l98^R0k*i|OleX+2CgDVy4Jxb)U{DgrmlyWZ zdF^$M8g9j0WfNw(^kUP62Qfo07sI^UF)Qd7Hn0M3%QX0>a0q|BZOV>|G}s{_3X!*x za9rLkf3SWea?t?q0hS^6Ik1GEjL*%<|C&Kv_83;CZ|vE-i9J1 z-uQ{JV_UOq`a~Z5w1qwPKYi}QO~JjmaNJ+amiLUK3C=t{S@y}_=3;Q8IJ7X8o~5&ExLdf< zbihRRdhW$*39^6J;wub}nKNdr4SC=x-W<-xS;;gH9WVjs)u+SG{1~#HPk_T7Uo5w} z1pnAzz0FZABaAK3`6a zWOvT4+sjL>ep7qRZF<<>hT*$-o}ah4vuAo z-ALxTmGMva6du3iCc}lX)Ht7k(kE5WzmkbV@;gfzT7aU-CvZ%%{DTwYXptc;_VPRO z%)LZRyQI#EEp~tjaabW;K#pfeGU3-enm#|lof|bcJ*5lp=GyS|F3Bh#_E)+?8qvh0 zJC!nrsd}On-84>7BT!Y@e_&JP`ns(gd2kn#f7Nr$l3j3nF%PjS*U)e29oVK^#>ad$ zK6bbXo46DV`I?7i7aDMB>tMEeW5Ri7B6;Y35=-}bvrPX8=ifWYminpOQ+Jtemd#v=D&E@7}8Eo78BCS_6&~Noa#I*=O zyH@$=YOTt*iqX6kIg=YLJovNHgOxY-u~FNZJk~3fr_Yv9JIjQJ4Lh>YxTREGIGOq1 z64~1?lbPcua=P3hJgACdnVmILvb%AOnH^UyPexp~g-E>{iP=_bIONe#hok9=IEA%_|jM}OB zI54;p1BSoG3|z$P+q&GI)0kIwFXa0Hk!&q{5AQGbQ2so4mRC>pRVI$Q0e0@d{HwNd zXA?!u-<$Yxpd)XLx`t5Mdr7M(fs6c`|F)ntH%gYU=2ta#SU!uNFH4V-eHkVM$@|iO z(JTy(<-LK6sH5q?S$V0@`y~5C$6Inlw1S@g+BDuLcy8tyXjMFbe|ZOn{8}XQREB&v zcQKC})}!|eeTJ^;NNHAPJMA3al4r$$m?(}|v4I9!Gud26lfS}E*bh=hS%I*AmK_6y6G3dD)k%YEc3;tuc{pRQgSakSDj6t?tC4>nb!n`WQ!l|3>`8i?}74^S+@s`!8P4u9rISVwR2M zHMis_|Hj;6V_Lu1k9!Y_xb0*Hw=eKzK=X~<5Ilwb z^h4ycA%|(}1NnMvd-hr$&Y7pf7*KJX8oK-GExYR_GP|bnNmV&%iH35{*LpgQjKS!x zCopPEG2-3)*g9&ioR=(Q+kHPsxu@E!7tR&>GTV?@2Xm_=tV>Mf_mNJ_*wl)DX6Y(B z9&MxypQEAd5gEbFA_2a4eW>ziE>|5KBbi}xURT*fW@y3?JG>J;clcw}GZ!@f zGn3awujQ?ZNLnSB)A@gtooPH(Z`ihFCS{0_GNcG4QucM-OC=Seq>0iX4S$hJlakDH zq6|eu6BvOeRd#!a}*LfaCm=TV-MwxYRmNjp>)wxPNMh}+3k1~Ha zQ?&`hWX$l5qXuq{+>cKABd~R5JLHy6!L*G%P~BJygX%wNUQLYbQW-#L2}5Kyy#nR( zNGQ7W3U6J1hm#i5Oqj+>OujcCfj%S8&s4!yTWJ*iNrtQQ&Y=IcZCH>o9al0MWR;pR z8B@BDB*}zelpA%8tXhZ1N>X&$z$p`W4d3_Pc}4hNd2;mIL_%Ey1f4ygff3qh<4 zraLT&M-aL41o}EA;K>=Yaqkh@6MrU;U-~uh+1n;K82JPySiOSdOQzx-<_o;|(*m+` zZn$i|CU&M%54v$4w#@LtjN3~vNY57QuF#I$0u2lvR;28#d%%W>qvGraT-M#pyed?|T?;vJJl;T#p+z3`5mzYZUi9jwSO677wjJiEBr(dyf)!7|~tn zMCx)4BDik?^=dEk#{a%HU{s?7d0y9tX7hD0y4Q_n{I22TrgzjI`iRbaJ#qa_O%$!r zz_uV~^!@n~ZrU6|)uJTK-Qf+t4v(UatO!h3{)M~uY7^VPK4ieu9{;N^#+a)!NkBju z>^vrkO%6_&S$+}|eph2j?nmUuQ?F;bEAEn5ih4!;AUI0*Lp}#_(awIjM)M4<9X9CP zx*wNWh?8rrgoFoBzT9eGJX@TJi)jvL_1OP#_Ox#(hO#)rOcTZEUYnDNMF|aga!Jpc zIBsmg;R0E5c*i_aySow3qQ8+N3*R?1m^D zO)jE=mJ)FfQ6WK-D1X+$4F7K1io)Gd>ZkJ~0tFT4!df5RIeNp|7tM0t|4IuRX@ zoWWOrCz6SdW5{E-QN&3&iHH?;fXJBOrgw?dLtlXUoe5Z=UWB_!H;|Ge-o!E3g|x(I zqw7~0{92cX>)$=V^Owhxr}Mfo(kBg@$Na@0KXEcb5QSG?$q-SY3~`z*kIScQqdD3W zsGiK=8>wu(Ce@1PdY|LXyf!+E8Ae}!I^(qwLndzqcA5@Ah#rHQEt~PPXf&}4jG_Lc z1{{;Z;G?y3aO?G5=sR5lL&WS+c+(%({i64#5y@z%#G~F-4cstyF4m}B#FwX)@!gO< zw&Whh1n*@Sq>zaV3`=lO{tikhNfGVdEo7G04eGdig-?HvCu5edAW6MfOos)oA`JRQ zA^sk@8WmPgqJ33uykbq~9()%pbJxY)Jz=<`^bl%C(L5Pnk9z`Raod+{#3V&>fo9c| zZ!vhOWhYuGY{m6gvT*gyOw1ppg;n?apfZs9(dL~($=NlSl9z>(-*sWd`RQcY!X@PL zn;H1jUKtg)s$iSeCA9HO$7L5=aLF$VY@zPe5xMPX6z~Rup7OYQtv0zhtpII3&SG)S z4lFv_4nL_A%E){q$<*k;QK6%W)lzd3-Z=~^ceJPr;{fVY4;4e7H7B3RVdauaY^Bd2 z$>IlSV&RRi{NLg4KrP~8PVbpzh8U2Lfj7VYhXygZSaZUTJdHXIzjdo&##~p-YRSeA zbuUmebp~-A?t%Ya$>3^@qgdNT_eXjTSSr39?c`tM?UjvK-8+>8O_3tA4yq9Gdl$*j zS$#|ug<>Lq54WjSVB_*KJn1P%-bY*k=bx!yRTGG{cL}JLTdtmpk~dmje~E7U;{h!4ccnV%QZAJYp7(lY)LiltC~C z4_!ld-5F%`{a5^jFm38njP2iu>Q} z$2mC>*r~k}xwueNy}BEx7Y1Pk*@?kQ`%r#HApRE_j@F~y>3iN9G(2UGUqp*hg66OH z?^uJ1wMWq0U4oQD4bJeAB&I#PaXV$w?75$zKwMOFoE;~Huu>fUfIpE$J z>XS4%h8+bok8;Tm#hE=AvXc5Wi;c*}9ka>nSVCs((j$(2+T`IGAj$$qblk8Nmmjpp z=*K5;R(1l8G{219Q)o9;#|Kkv-SEQkrPy(HIgUQepoIT2Jf2{L7XGWyVCV;icK*b< z^LkKjVFbz!_~W3rJKmD~gObVmWblF*X&AP~pUeb&AJ5Gl*{IF?>N*R5zL)(u}gr7W*oOVPwN5!-g0K$}p?%`K52 zt?jfMX!Q#lk-oPaj6?B;Sai|9hpjfZv7nUp4b-(sAgf4>A1V;r%`#+7nl{N4sgpK6 zOA_6pOMZ!^n_)EWWj**YJuX z7+v=Xxs6qLblz{&>5?SYHkzdPnJUTNZ%L|T^vTaX=0tn(QQB!eier0erkfi_GQSL> zPk=nRvh^2keQ_6qA3eZU4<(G;I0&+yV^A_;EiPRXgN6=!aB*iQbPDvwad5#2vyS>b$6KIf&;p zX}+lM7>b2n!2cd@BWaCE#Oh%%xiL|bI6sjhcb_Ye5gt1*uWKtR+DG9h)*a9C)DaT1 z9)s2_!!ctWP*FG&3)LxKHYo`gDIdpvn)#fsc@*!^o&D3bLOj=c9W9a@X&-PK-XB>`>}nv55J2P$YHfVcrCRPSHF@V1>dwu0_Dc$`e>4)=VNf7D-v%7)9=6K zXp;S80;$XyPt=O!NqPM!^29}oxYIu1-dUNb^d<;PmXAaykpdRmkER{@5FBg>!Q&Q` zo$b_z^ZWjwsPq%%g?nQ)&9a}kc^J=o(hR%mSYmFcOFlEs*m?2*+MV5mG3I(C)6AR{ zepMol-r4xm>?EEPyNjlcGjLtT0#tg=VC&dlsK0wOIW|Fz>}lIhLxe+^c z#*ir!%!uHF4slY>#M1sSTv8c=fAhlea{D6u_SyyKzYjo(>0$VC27TvF+l$9u_+cV_ z&+D$8O_~>5knqQb-;I=_UOus9I zZaLDJm_hr->mt#uECrKCh|pDw&db^eJ~#J71@1Q(#M$86CQppp{{^Nj6~eh%`tI)A z4uRu7fZF&?_`0tY6e+(%=ON-%-!gotdK+I=i;;p=ixES1Azyxu_GVf^b!`>=E*OS~ zX9Q?WpK;Sig=3#QWunoX=FG7gsPw4_d@0w-b&d+w2Y-Og39ZoJI{{<`#BC_CK80pA$!N7?3R9FcMy zFYeAlKd&}8*V79BdG>*;XbbJH#?pIQAtwLxMa!a7_$%Tp);&nZPPyM0THK1pM>4S5 zqX?CTAL273cgh*0?mm4tw56TyQ|1BG9~_N&t9x)$sWyqYX+&aj86?!ZAwF{tethJD z70owrx>+9nIc9I2rG)EtdYe@~ZTMeYLZU!;HmJ%H!M%(3RTE;X)iLAA$5=o<9_7Wb&4GEcMTdh^lnkU8Er zrkv!tXE5{qP0T8mA*P}jRCu@sE2SAcL_Kj6FPmVhfh*-eSHS7?$vCs_6Rb)RN1rK8 z@H%i9;si42@K_&D`K-kuFLNvj+=a9HlX3O919<%VY5e{E1djSW0N3tJK&x;=EPkhf zy)z(T{fN591c8bR2rS3!g2zjdsOPP`davXqLT&BcBv- zU}hfT1`*z?q|E9S^7wqZC!!cZuV<8%pmH4NPy3AH?=M2@3?pp) zikGBQv09>tdLt{a_In{%tfdaDv@VEp$%QSHq1Q965<(*Mur4qdC(^U>)y^YW;&%z( z80O&dd_7z(z)U4VvHZ7hM^`OU{Sg~UZ!ko zGaD<~pO444an-O&n)V3f4bg|b-{oyUBrevN<#`V4?;XW~!SlGU?>w$u^aL{$$C0`9 zb8ylOOI#N2iGFH^*p<-&iN9p%jCw43_Znd_vBka1a^T}KH7v6pjWwI<(Wkfx4;6et z@9pLIY0+Igxmbj|_P&JVp9k@As5#zyM&})^*D!^C=4NI`@wv}dJk?!=&+cmB137!_ zNp-?6D^*b^U=H4x0BC&DALHr0XhcK`9?@Bf${sFQKIu40`bwi$2tjMb<>``Zj7w%j!AA21GF<YU+h1fl4?U#YhDTlF6?JkPu)L^{) zT>KIrh}JJ+&^*Nk(`g1Q@5Uh!dSz=eNKvr(mAsbxZ zf!##9&wsT9d+Lv1LvhLf> zwTosVxg3SBwrA5h_Y>T=&j2Tn48#kgPGWnv5&lYBg?=8+_|YmHKiy8jC(lpQcNYV4 zGR=>q`zMiatKMMB!pF$(*COf$58&NeEj;UHiO<3g;`y8qOz{lBVs8`N<7WEKJ^JAt>}f8?MyZj+BYiX}CW@rH!2>sL+k{fG@o4LygY~Np zp{2?dv^-LYKF`17`ToD?-q(g{l<$)*uS71{`=A{?6JMnrfaaYk=zXXLv-Z;aRY)wZ zJ6?#DkALApD{nL}4n_O@$8e!A1bjx+!K{91G<5gGtKZV_`sI4e?rnmA2`g}8Z!~UO zT@3n9-$9k!Ol*Jp1=or`;d2XRA|rDN9vzm&5!3^9Z$dxyO^+w>bDW5aXCRueX}H{o z&ayZSoFcIdbAIl@^H28U_Oa)%X?+#)DQ?8X%a2_7ltcy_PSQTs8R|phP`A|wosQi= zUgIH_-HX9VJ@kIDI}1Q<~978Ugz(1jqc-CPNhW@6#&5ZTZ~abQ=<= z1NNj(J}4;-3XVL{fUm!#;PXK{(0!#2khD=?lF=ww+tni&UH?rWlbQ~p%u=vb3V}g^ zF)V@P|H-a{ri?y*Q#HWsy=wR%RS_?Jy$p^?7HCY4Lm$nr4El=%Hi7P#J~R=d)b+7) zEM=Vio&&ZAH-od16jYoY4+_^?1s(oB1atSb3o5cp1*05ZLcO6X&54B49rj8%T(KUc z+LR!*I~MK^_|ZJuG_dO?FtCDwTp|fea<$P!UY9bG=i$~B+PH6r0PW|O!SC1@$kca& z-{q&^=IX5gS5#nf-FrcqgFWnRp9i)>lc86`24rvF6Bs6{fXZ=qR1{l`XL9H{pQQ}x zv58>jngKtz$4Y+TNnloj$7cpoi(Oe+yz^jGp-pp zDHt`=9KJaY2?SCcbdUTA88S2JGmhY-q%mlyuZDXlOSQ6A3#y&9;XCyTc9qV>%u%ax z= zJ(iep*$t+@p9wzVzTh5j0DS!iL87HADE5fs;v*|iQ*s3=TI_}Inl{is+a1Pv1N6#j z!s%K&cstkz>^o^Z`Q8J6J@tj>G~>MT$#nS3SV8`Ad6*mRLg#?jsPpeUcs_0dZ=)2P zd%GTk-qGHf*kqhoV2)?|n<3Bhnn3b(w!rJtbHVne4#AK4Vjyu&AMVJ$6huwefWoVl z0;j7<@K7!ks?>S}C+=wAA`@xM`(ut#4~=ko=1lxpJO<~kxhD8KR3jKT+bAd=r~ri} zt3bwf8(i_Sg9;-*!A+)ApmmzT(S;K+~i}ui*CkL-Sm&3Q)cj1-NNc_wH2eUk4 zphSs>stqzQXOSxX8WG5TlMGG5Z172(hGS;53px%61)FHT{ajcoP783s+j)z@CS?jt z#ud{ z83_aJ`+?DpgAvQbV2dt7n}aSq^=T1k@6mwEV}1&3U2_C+?UXw<*Bgdte{QhN30-79 zLtKM9=o#ifo>4rMHWb0W(shtKlICIP9&w|YC2F<$VP+!DOD(zsQjNV(tn?lx%+|&8 zU)-^Byf`g0lE2=I^c*ZdObHz;Oi?bPiI)$o+M2;6?*nHdMmEJrC&l>JCocJ%WVZ`U6YUhat^W z8Z9Hyq{ z=?>ICZGRISpX7t|o@O|k5C8)TDWEO$0BSRvVE*Utp!eMhOEi@*^`$;;Ei%Pz{^2-W zcnW8B?!;*@&5?7QgN{XO za7*%N^21A)c;$V;(C8vG)%%Oa&056JNDoV+eeloAK$LfWgyfVgaW?#d`lDOmhnfL8 zn$i1RZ90_MM1pwmOZd2PJI&){KyO?zWcVVEc;SqCQ3Te;eJk&P@BF7O3<*4f{fw$AgZQ0$j1lXl$+;imBs5!&Y^ceG=rA!Hl_rgSlVr(QCn>UYeh=y zN;~~~Q9SS@&OTg$W1e2ZZzWC;b-^Cu4sQU>KEP@+3qubj;2d#7T#<1W9qGBSrTI0y z;eNn$O>xXVBTIO1MY4#p&n6wEbAUdATV7MPO4DaFFnx)2PG%%`=WKkTwhYOZdIf(gao{PIpB+P@Ee&x&(@4DD zwh$j}qTS*#bOs>59BZ{V0IazM9%0X*sKy+P)m_kYbvS;x6@i<~)9^Fxlzo|)fJGZ9 zJ1(UHT^qE?*HR;5n?8*+bTkOUJ1PZ7xHkfwy$NuwO%8s}PXVD%BRJ3==F=h(h?m`h zDO>l0zSJ3*7#Ry4zjuS*Z41ykt_btm%^@by0Ftk)fTxs?o?JT*+#IN@%1sO9i)FAt zK9RB{L*e|gqaZjSg^8B{w^`ewOeO*ArN$7kOddAupj{QG%{Z`i3qD*Jf^KFW_%}Kf zi~2WXwtOO5iN~SKLh7m7B}?9J`;IA;V`8eh28HU*Sb>iO6=$0S@7F#TP?-T#q>hHF zD>K1plq@7DnZTW{WiVY*0*sc(fK;3=M5X0o&!Yl#(s_YNN9`c!oh`_l+QSmtqj;(B z3_A5+$2rNGaH>WO_URcw+Q0jh(U1jYA8VQp#6Mqu`58q;nWG0$wSKwu>Rhaa613FweKp72fxZSY|MW3tiLuH%ba@Tvox$jD_ zD!fB5*5s*R{PHFN@0|$7Roj67u@($V(*@bK7X+#D4TA3(O@dVcw*&_(2L-<;lz{f- z4EVL=Jbd0=jhh@_V2;&gbbN1u2e)`*O6W=ym#1?^Qw3bwJr3Dp(h&1m24JZvnI0KeViAYr~1gxO4n z)$<*xTXPw

+gpVVN*_ZorpXeki?S|+$)_EZoUP%22jGz<2&Ffj56?FmM@LFRoU zD4wt$vQuZl$rICQws#S1-tk&6D&?o3=IkrMe{EyPZiz|cR=P6rUTF^UB3Vc|s0$a* z2BDRE7(Q-r#of#Zve=RCaZ2iO@{k6(+57>sOJs_dL61v>Bbdz-r??h%D95EP>&t|3&)DakSbpj5|*P!1bY~~ zC|HE+x2!;i?IR&QR|R%FnnHOP2{d0k3-rkxX#MC+ekPa^>xCQ0jEtigt{s62N-21u za1^MXNf(rVc_x@6n2U4iyw>-EHTF=3wR*cOJp&lSr*ppOIbkV2m-NNR#~1L)+-rDr zVJh`$dVu_US6Gv24bP5!gL12H@I3tujC&CRVDX*TFzs3~1noNnjKZ+sYojZKuh<9A zu7!fbjSG+|@dZ?b7huipQ(&|7KXASt2ZK}G1d1=u!!d{B@T2i0EN!6P%P*Uu!0LzSI6*do-NUC=$H*_6UC6xdQTS z`{2I^^{`8B35bGML-cR}+_`fTX0DBZC67el|8XRQ3|<1wuGt{>J{k6IRKiyQm!Nv@ z8L$eA1WWxUnET-qH0vLQF$&5kt^Gx?vPJ{`bkBhEKhDF^;JcJN-VKgsZ$Rad3a#-ZSunP{5Ajo}+jeq2_`ach(=WU6*F;^~DR7{9$ZA1% zk_f7I-hsz|#Blf33D|!?4maGhM@_3AkR&dT<2RnbXEFZpx%LV~mr)iD(E3knj7K)5>=_Qs3@fpMpxVTmIwJU$-u01rff+84xRRrde zkuXtb4IDL1hL~?@aBf$a;EaD8oYZRu%Pr@igue^ve(|85b`pjyKf$4-?Vx|w9>xcp zgir1H@YmuvG}mT;j!hce+#Ui>Q?udl`~=Ev5y1_v7JjB4!M(pPW6|V9j9qmEzS48w zt$0TW==Ou&_$U~VT5A&If5FjpOy=jB1(j)yR$d>Yhd6@mjj zUr1#S2$mYBfNK19NY;7}3pG{Hd`>qkmmZ5BqTUJQBdP@H^f?w@aR;)8df{usU0A(j zH{|?GgN|k)%yu6S3r5VQ9U=*srY(nW#7!_XSPBErXv4O?Z7^UOA^859vZ_+{0sqVy zjr{zPJ3)u5@hRho;E&kxFF>o+Ok{o~alFi#-L*_baFsd@*TvzW2PPs&izh8kQ z_{2r>+c2ElCA!9)e;CMb&NU`^d-jvJX*W2lS8g1Nsq#}EDl_f3k$JQ%k5rwtV7zmW zG3shZNYbJ6xc|SI+`{`QT&&F>-0?b-srehmyz-1>4!d0=;&m4}tukHySnPXp|IBHk z_g^;Ay1I-cN=6BP#1*q)dbyl-NET<8Kg>v6F}3>ksf(F8-G+Ft@!-}*Xfny)pORP8 z8p*#;O`P14^^DFyICHqpiYxjNhdTQ8++sIXQc4;4p-20!qjy=!HeGJh?meW$Qnp?NY6)mR1s5PF1p@e9?Jb-KGpX#^DgEvQvdaE^VP<3 z$;#a%AT*Q=swJb?;4piboaR;sMUsv)-ORu5$SkCcl6{?<7~kJIRwtHkWK^C$BK%t+lz=UY#w^JjF5`89r5aDPZ4cf41DiF|v3pSEr~ ze_`4$u62>0AcXFglkd54?|goc+V45!uaY_A7}Y0?|1A>cj&tJf-C4s2_FIvh$gRwJ z+hTH1`5O7=ewfdDbdI_1P)DMNiwUt&WyU)lW3FHr+r6rZ=q_K)jF>3T?R!3tJ1AAi zsqNdxRPUU`1;3CX369p>jz2bB;MNh0ruAv^UgZ`zHCl7q?_?0wu4MLoVlsQ|&@XPN z|0nlR`6*Xh5Jv2p3h`WbG)WEUCaz32w`Pfe>okz&_jq07B=^hkj9V<fzs~5TQ&X5_B4dom%EST{7`NT#3lyK^6 zB{I+Smhe>3ICQO^P2As&$}LYNN$Gx6N(cIFyw;q`{mI!Pyz@G}l)rNXUY?5*&QUBR|3#^iF-k>D z$;nMjCHI9q6|dyJW$N*X?<=|D@atTj*=$Z>?*dFUsl?=cZEUYcIKz9d<(aZd?(=d* z=D(J8Tz96dU{{NXd(T9WszX!x$b!}Us<`K*`_QeDrucLY4C=W936nT4voGBEGH2o1 zIThS^ja|%)Q}%p>doMR`i#_+=;~Y1yZ#UcU=NIdB*^+%S?;E$;3Hg)iKHMIi)%?4g z(R_ZbH?jX&%gB!!Vw`-{dD%sB{A|7XT>U-+?j=*to$68In`@Os_RBu=rSn_(pqb|x zeSs~J9&sQ1?ychwBp_ea@&~s3cH}<(vB0J&(tN?zFPyM&9GA246vsPmw8EMzF?MFZh3~^yHd=UpS^-y5VZn<#rLqq+rT8-| zGr8{-W_+TyB>$#8m>)G?SrqcHjXOPIF|Sq_&TqOt2P2(g8)`{#Eb{g1Et$BXVCMJpE2KF0Czq)*na}^7%N=+2=bvqI zNN(qkBQ_ynT$4>PuGfwsD+^{2 zVU;|Wq!0%?3%+rMkCM29VJkV^{PEnixOOsUxiPb@e>dWqeJV5Z|$GP7)SoUrVCA=|fdn$YF63v1PRo^dlD zMO4Po>ZPX}H_^G4JurEI?L8TTD@YkRv%UZyEj>@ftWK7er-;b$6bU|kpqcG+RU-*X zn*3cq5B`r+Ik&aTir+rRkw3YAFaJHRpW6~MnzxEJV#1}@@xI0azO_o9|2IR0RXK14 zm9$-9#{wyIhIef6+XY;7k}=7hWX&z<+CjF?yUIOI&tqrZZziv;K5%Jr)5tU3RgCr` zJxf~<#{h$$!WP9D-29pm+}}0Fg>fsZ@%z_HtYqbV?pxL)Zc4Eh=T>COis$R|=R)K7 z<8hT-$?SF9D3%W6XLV;6!#$Q3~TUnADJf~ z%Mbnz=I5uM=S}4F`EBRs@jtTsxJK=RyhGc3o)F{{*CesMW`rEcP-E6D?%*aS>##Qd z4_Q%{HvjvEEEiv=P5wkBa>a42WV+X4&M#~^Z@)){&o^Jg>z9SIzY^{Fsqf7B+Uq%_ zH+U3TbkB%4m|exo%z4k1DSzU<{WLi%_c(C5Fq4bTNfWZK*Wl^=Zq8f%7}2|;#B^M- zXWscf;T&cN_;cI+xGgiRMW*5tMN%=AB4vvV(d`ya^zT%dXtcjHmnUh(-y)C7q+$Z-jdFJn%#>t#t-nPUdoI5=Y8is?w4Tx8QkZrH^s6?J>qcK zY!YvkJDU%7ujl`ymh)Q!pYdkz%td**P9l9lplEZUJ5JM1fH@y7v70`|ko?AY&g|57 z_LY1*nKUfN#}+*1XJ6aG7mums6X%*TmqyQG*zSKMo&1Z= zVLTAi>NaqBmvJL<^Sxs~(Lt)iDI5hC660OB)om{iUE zg)#enu#>HgIrlgXl5qw_{uBT4zV4g&4?hToO4xCJwGsTG*&~?eKcmRp_1lU3#4{|! ze?zCa7q~$;MW$H%A-QZ)P98xGS9d;-yE1g1+0+oiW({B96f!!w!a7xE)Ld0Q?Z;iB zKdynBrn`f$HVNcyOwKaz9P&upngGW4GQFO^Ugd|8CqHFYiar}Si1OL>qMm{4@}=5s^-hHEB2C;A5Y?lY6<7palyJ~jN-12z18={!t( zX3jtMImbV_@`?|QwP3_7-;img56PZk7yeSXGH-k8B(JgdEzQE-CF2aTN!Q6dk!PhKt*X{hF8DS#JgYhu&IF-*Jc!cWGfC zT#?}&ep>PUSx)@d!7TP>>PqfxDaS?1UngzTeaO5)4{mSi7+#;dC~TBT<{az0aO(Lj zT&k%Q|1#hoz5hjsEK=u+3XJOcrwZz#cVsuOF8iDA%GR+CR;KKvJ40mJuNWM$=@NNy zKofTlnlXRqp4hZ>kLb3!hp4vkps38RlK=g)gdfa2$IG0PVRjrXBnLl@V-BqfLqm!8 z@V7acoi(>Pu8N_K^> z9QPpDhU=odtm$+0IoWL=A#Cq2LA`G%vCJK0H~$caMTf&_uJLKxM%?-DUBt z(AS2g#o0f|P~&Vi@uoOCar0tMw|xZfCOL`Qr7h3eJAmy^b}WH5yo>Q;;UpbMwZp+5C*dHYE3GnDByrv+&e+2|j*q1-E1DVeY-Tv+#^j z09~BUV(WJ-5abUXw=`OKL8v^WE;_(mVhdLeaDh|fxQkx#{F#LpIMZFHSm9pDlB-$o zZd&RlkT9>wtlhq3_Q|DvVAiL_zwfXU+@%@mmbX)&^vGNG+vqdgtz##+nxA)B?pYUi z*LM-y=JAMK@G+BJq4J{izFM~M(6I^ZLT@FudD{i{W5EqrzAz9@M3iwG7Qg3m@8)wF z!~#~RKf>gohcM-q8#}yvC%5LOh|A9K*cELze{Z+q1Q7;$Kl2f$KV%4DPodavSnnxk_6dUhT&S zwm4xYH(~Zh%0}49H=jq|);pFJJ2X?U!lgnm=F36h_1?++zR?ZbflM>j%tyc_@@w(e z+euvZ++xc(RZrGBA(hSR3`42M^H?o!Q-Rr(0?W6L)(ZEhS_xmctl+IaCBQeW7u+eS ze)g>ENlP6A5qn*4CE984=FSqz4oNQJ7r3YKN=g?w&*WNxVaCytumvZ$zu6H&9I0RG zzb}oAw^)yA3G+$^&z}*FohZ-N(?56DR1L0fumxrv^aDGZ1uAjz!#^+1!t%PhQp-h2 z?4AtQ68S|pgk5q{4zTY zVi(T}X2o8DKTqZhlno0@7cZK^t^b(Kp|=`$U?<1TRBd4EH;v^~;!X<}8$ZV7R*%>n zBaU&`{LT5v8&`08Q)4+tjg|1LS&DO6U(H27U&&jYn85jIRbk|gPXdpP+qk=0GPt2I zi7mDM%}L!0A#*a_g`z$s!Potnm=!i&xL(7H>;9v~?OJq5Fy;9!P#X3kYt$u)-t*0@ z-+7K5EVxaSyrh^*ch16*sySq|@DH2VGecN^tVi(o4dvI!r3w}%THsH8P9Wd5Q26IS z264IclssBl%+bDv}!@;L0Sys|w1H_wN#Zo+U{P)*5p~Q|E9- z35}diwj9@2ds=v_WPxzS(g8L*#Ecu$y`Jl%y@||D2XvsEF7Nt>;A11l?sEDA=R`i7 z+;=;itaFwqus?9z8((fyN;GG%uavpt$iZQx|l3@G`vPwq!J5nZ)(B2YB{XwZ4tBx1E6Wx z85gw(*|#eqSw+=mc86V@V7ZbA29Lfcb6S?bt*vgQ3%ptcSB$>0Gbh|-DAm{cn_D!uWDQ{nm5u?JPWP2BzbwiQ7GiV?;0v0gO7zdyi9J%hE4+g#*+xuq} z`7AcbdRnBiTFZ?{gzFga{xE{=wV%kQc~kVxs0(P9R>Lj}mPgC^1`y?;h(4Qhah29G zqOEmK7`xkyJH~3G#o&F+uYHTsTb1!(rLEAZQ->7KQ6bJlQ%e&i2~#vBoE(<8OAI6C z5t-Mqf^Uu5!qOHa*fArEwcD_o+j;FSyOPg@JnAd_<(yZdFTMc_1M}F_&!6z$f=Ohm z+())=dJ7(2z;PFBg1E>cB}TYT65^P)(mgFo!p$+!Xtbe`J+ycoyKA$%aG$rVaO6-b zH#6&^U@+4J{cRPA>3^QYGH4TP|52Q@DXixfO0D6f{sfT{R+)P^)r067sxjRKA92ho zALfH*3(=gb#`!i(WwYi6V)-l?a=+&nfd>p3JGDgMvC@uBS}V_18f{~1uKZ*r`W6uN zwej4aBu)N6`D=Fcy^Cxjw6K>3ONgARJh^e>8Cx9|Pxxa|rTLLL>=^l1q(6T)F_`w7 zjWbPW7H6rls)MS`eY-f`AY2jePM}9j;ca48*utJ(|5u>fsz(-cKgpjDo>)#4xSwHb z@u7?y*|9i|aX9!B!(@#}R9zD>mwyZ%)AGP;*Im3Cq%3sUC(AqRZe(s;-9$o8uN2C! zn*y;`BN^ZMT1?HtGfdj8chp6x#H(&wKyD`{TgE*os#MP+9bXscPXF zk+wdkTc%F5?p+s7J`})O-ll#A$vY6Fn~U;>ecWrh1Jw;pN9Vi7R-c7A=>0E(z1-!> zrazJ3qOR=X7ArE0$9Qw>{C1GjOPtKBH_DMSpZ2j)e*@Y79`0wiSKi|`zU*aIP8lyS zAM=$>%DHY;)HG1y^=+$Q+=B58=e`OvyY{jp%SV&X#=un9XOX^qBPLF}jhi-6hLH(Y z<^J1ugGBqqv8m+=yu#BAbk9;CzUzva^_#UA`I`#dNbMf1GBw~BpJ$~1x;+|v+RD|v zPUZH`4rj(EeJ4xi9pPTr%ko>F4wjx>|CH64pu?NAjbOclFS4G@bjGu#8vjmsgFUq_ zOncmIR_3i9$;Xire%Q1W#38aGv#w z?4v!E!VTIn3_s`!Pw2nxZP_bEm{HA2H~fPziB05TfjaX{|2{SjDsincYfId}zO|fx z+l4t+)*|%V>A=U{WhVYXfM<1*K^lXT+8Ojdfu%^5v|HR|XhD_c!C zH>GLp#>O+unG=%S%V$=?uK!}m(b*T-%au2|j+zK&>e5rh??@YLuaRMU(j1v{!Y@Mo z3N_|OX&9IOM4DYX$AF2GH4sKs1(D=6p3Gs(CSev^%^rD?$sMIz2%Wc{+!6-|#@H=} zQ`;)T&)8hSC`UaYWAAdp4N|W#>#G8WHti%==KZc1>49D&h>q&;n zPUb)Vq*Cnn;ap?)$}SU25h?lN;7W!nCdd26k$y8k=Mecjt34ColasRxA# zr&+CK%Y=E19{ZKsbzjI=jJwYl-MhnQr-k#q%Xcs}8M$~sMV6bO?7($RNo1vHXinW) zhTT6c;AYqVW9Uqysp{e|j*^gh2oW-d%-pm0zR8eC8B$UzLuo)oX{5oFh)ktI#)^;z z(>;6dOQxu#Xd*?@s6kS_l~V8haM!wP-A`w)bIje zE}Z+A%)!~@x&Ke1=`xq>FqwhHT~1;p&v>{lxCyK-lOR^}14!ogY1H=k6X+Unh8Wq8 z5Y3*yWUmZMxD7$X#cVp!dozoEQh5N^C4YcZLZ2}U4kw_1Ta`?fx;OHy68LgkE+Ngf zjW~EXh`5f|;>QhQ3_K)4*WMo|8Y35}^5HsKk`_QeiOazq9YV0LrH-`g3*ME|gShpJ zA2u+3%$ORiMZ!Uv;KgVp(YEalClm)%Y1@ET{D|L_fVEz6*7;3_e^ z2%$3A40dZi1HV`N!(V=f&{~;fas?kol}0R)IQ)auU))U>9(j#D>ZZb32gk8%ml>2X z$tBC{GD)PkIq)#m!znU5SgxHV>2q2k$o`6f|D_t@#uJI4_TV6yb)gIL51z(%;nv62{h4-z4RVoVEe!pkog!trY|Ah;$DzU^y;yl`>4(^#LC^O#0H zew)vH+bIPfJuJfO4IZJvrFY5Pr*5zWys4$J0cqvPwz?C(wU>*$uWC>X8+a^237{*Z#gDJjFE!JuJ;1JXaYiRA#y`q zifApi!prygfP3IRjucv(v(`lFz0YZYioNaIPJHvo?UZ!_x8 zj>2bZ?l8Yk2X$QGkk1Z>iQ0_Cuzb%=(%tz2i~K=EUvPqN-uT1glHO4C&`#_luLjEO zs>v22f#rOTfrjcmTzzN-Nv_z)7Me8RAlylAP1fM2Vh?O}oIb-Vyu-wF zd=!7!EQUE1HKckd3>}*Pj+&dV*N41|<65EV5+V4S04(G>v-chi*T7 z1ONPah&@I`i0$J{yil=$b<{V8PEBb*SaANenwUMPH;r2H$AOIfd-CC-0zYfkM|Rp` zM<8|o8jcD61615Ef`A1P__Ba^C%L_nt-ydx^BASo8@c=hHjf&;d{4%NzLPW6uZhpB zK{8`SD3!H~C1pznfSk2H`SJ2PIr*mthzWY*+B5o88A$P~cO(9km6eqDsD#Rk9ihL3 zQ>kJ{1c|G)r(w=(QHF>RE%CiV8j8D#Yway$C#8Z6=6b^FFKZYXfp+O*iYyeJP2+VsoNB3T4>y9Ll+!^hAU z_io5j$pcrd)zS6wPTKX~AF98;ow}1);O#dEqdddVh{H9o_xVXu7kTLSeVO} z_B%k`uA@+reGWD}nTclZd`ADf;lQ6F;fVUv9{`;;A8?&`pgD5A?BB{Zw)LzEky;5+ zZqgOF^+*bIJK0TBWYy`PQX^tmtID4~ypylF#gISn&x^18wS=#4#N(fOE&+R1T>^fA z3iU%z#Q8lk;`}bFMtVD97e6$thJQJvlb?2MKK<(coyruSr1s`IH0o(FUKk%p)<(a@ zZ<{{U52h%=Nr4S??7t;wS9>R_wBCVQZy3|9fjnxN@`Rr?rqxiOHNC;?q%9G&Y$Pdc z3ekvCq00{F&^V_{#Gda&Pk%W~OJ+CILepueUrgW=3$B5V`fWJuwFi(uBJkoeIlhsk zA7ALLBHzeQ3Hu&-&FRV648!I$Q-c#q{5_&ld^wG;;2WdGbPHbuPa7j(TE!jqj>ZRa zcJXwo5Ez7WUZqj#!oBc}d^TM9Hl8tgWK2X;)=|Gz)A{7v1L!xPPhUR!L{^$=gEN&{ zIPph(-7#RoxBX_%zh-}m|JYvw{X4$~37vL9qgU?nv~JH}v}Xtr?m!(JGdH5*p3|u! zup;>%W#C@@aTunk%1ZY^^!w@#B%U`!j8#|CnrZLh(VvINnVkEiaNq~)7n({}`p46d zOV{YZqt)0w_9$#ND*z`;F9W3ur^zOTK$sGEgYBI9lXk?+=36hl27jm(leoBeI#jKN zA}{=cVrswfC0`*dzej}})l7q7u4>?^_EOm4@exW~RUw_-K}1bF5o;@-g7;S4gmMx- z=--n_I_8^B<$d~y!5v?acFZ51xwscuEoleaZ{@=pK@EMx<0)f#Z8341^@@zAodpMR z6Iin=4*c0QL^k%O(%zqCv?+Zaovyx=w#1*svyPbZC*y4Stp;-ZtXsa!Q#TQEc*SqF zc%=)D(bx~K`rZLI2V+5$-7Xj`S_qRoL%~JgSlHXIhTO3h7|&SBT(bNIqt7p+#)@SG z)h|ZDqW|bx9X-DMVhg@*Z3Pwec|*00HBqTh4x`oIi(dqFkn$g)Fm9le-gG|ADS?aN z2ETXY=e2gK^h^qh`z7FBp+vm=y%*RJYl@?em6F|J$Dx(0z*DHlMFt@;V87RDSX-O| z9u0TF4xd?AZFn|)^6)0=<129A^LxhDN(kviRtFOs>{t$@t+0D5V!VEv5%Xe_%2MbFrbOsz2dx_lINm8;@k zf2NWjAy?o(g?VW2&_*==+#as{-35acQelBwB>Y?c8mcGW2F=UrV9uLmFlnPG{2?_2 ztJ?RGO)l-wb$$lgzStM3u8>F5ZvBG_s!7P)+#ac)K7_WOnvGV=dqU%i7PwyhHq5GN zhucMNqY>q6sL5R(jihZxE1q6KZMPnx^Yi0a*Px@UeuN85b65mZC(P*It;gsKH3J%Z z(uSaA$BBPO51EO!(iK{B>7%3wP^Nni?rRm~lsA~bONVbU(gSW-EIj467pHktvHc=OM4=@IWM*KP zy0sn_9mH)>) z?O%jba`M50)cGjH;V7w|oI+2|+l1B$^lW{jfheR*nteA>fM0HT#F_d|73boO(Dq3w z`2M;C84SmRbhlq*bKfu9>C-KtY&)0Dxl@i#4ZcQaXFP{aYZk-g>;#y_3RpQ$1bm`* z(}1gn9&B`Qg}d-1dnbGpbj%lJ1_YdJ@2P9?*U$)*a_cI3HeLboo;o((c{;p$-wHZc zCK6t#;Nww6z=Cj?xts?>j*j4)i!E_;>J-?NQ@Kn_eC1){d+ zIR*i11Uno(Qh4(P;bv?Hi<*krk)BZKo_K&f)7sOGZdo1$JG(-Q1JOFU%K{TlF z8_wD1*4HzTe+USejW0p< zl?*ngZ5p_*^%9Bv5?uMAT6g*TU8q$(2Fqfsh}+?Ol0RF9v|4bn zwO1W}n6nl6bhsg9aVwN23Sq$uG$ z@>)~`-IV~l?b|~(xI%_w@?kY+&E!KSx1khB37gXHc&q7_aY0(mR!P)}G9=;B596tf3J0-F1i}Q%i$CX%Q&Imm9`4}=V z;o~_n8SrmzD=PIm2p*W%g5k^*&f$X8`h$aNaQKTUqwA&y6aJe6>_!XVz9bdY-(>)O zJJNZDZtA4*%@=ZUp*c;hmcg>aJmNU+K(-%W4r{+uU|GY3M2Mq+NYm0~$oU@w4R-&4{V(NE zj<+Cnt}7fyCpD4d$qI1gc5CwQ_IV`ow~_I6s35~5QbaK)i!;)6oLFC6OJGq0Jb3FF z%;#0Z)YvoNq{w+@dDIK0adR3_vc8IS$BnVP@d?NX<`#bK2hhVZYtk6=9$Kc(gT5Qa zS&Ij>zHGBHGBiHUvw9oJ%)ScY5$$34(c?HSQ*C1tw?x5lxfXKDv;>==gHXGv9dEkc z!5)5&;R>bg*sC>+YI+}nlBF}Dr)NGqe`y=gEj-RCUvkHMk@dW%ND6VcKfa98+#f2xj7Z9KS%{+)d~_C-bT9k zrBD^yk&zA-9VMYIw;L$~sI=LOZ#m$j<2)tlq|ehf_zHdCeM#zlR5>+b)9}+;Xt9?ZR>t#IRopP)123~AvBXmW#YAmz&+FhA)QOeY%n_c9(w(&#>!jJ{94oG_$X zyLD*exmna&W^%BAd+_;hGTH2miSi3`eqIjX9}`>1H*5*w z%Cst=|Fk>d)P_#@@wpn5S^11~*LcuMnN!%haV^pdE+;D{Pr_lIG#bmAAiF-CARTtP zsOnW4nizVG++(-Uq{>Sus70FFsQV4YW}m|Nwtzk2-vQ2)<-?^8`%viot0?YrFST!M zpcc!tuv5W1qP9Ahywf)(W`e6p99~H4K9$hz>kYXeN{+jK%5_0pZ!7MP{R4A*`jJgV z3Er<32*l*Hh<$-1|Ap`iaz$w^U&cj+KP~bj4cscs&(lfZo4nn|e-=?eInL|px!Z$e z{1=}N%Zu`tS}y?mtrBP@w!?lpedN|uK@GviKp$lrdP1X zj;Fw|{W%~ue3w1*U@3D;QNWv{3iL$JMmqb~H_}&AhTi;9LS^e0!?qPuxIWv`(Db<) z{0dNn;_ocszWrRt4ZH41AA}#L)8%fF+{vw6cJv@Oa@3d0G!$^xSp{=*GEeV|D_<#aRo3&R9^gLh9R z(UKD}TnEdU+`;n_+_Z`kx_?AATeLQ4|;cT*GPJO72@x|8tQX>ru3m_ntC z<*9Yi1tNXtAF2J8OpGUIQoE#KYBQEeF3IK4-Q#Oeictd0GcQCBNH{$4JQLZmK1fq$ z0Pm~ZMsD?%kln8l=~OKvAwL4pg7tNHD9;sj*6Smc+A@K@^eB8`cadyLG3Ot%Tta*n zU4VLvJn8u9do*p=1G=pK91f6HM$u;zfrNw_msGOq=*%4`t zk2Mw58*X^blg~2ZK3jAH9l9gW?SGU+bNFJcI*SXF7TNDJgsSmnfe%BPen|&X63yLGjx_hg1h6 zsT%;JMeFESPYwR785(?R%P&M~?knQ`T%0B?kmA3XHJ^WA=^v^Xtpx8JSd9Msn}SNO zDRWmMN$x8|kmj5=a^+w*Ic=Xo0)>akSo4AG*O6Cd7&$7)Mn9Nhf{R|fA^*rlV%74k z-fgNn_02d7zjF z19qX+NsY{g6^88gs+Xw1RR!*vF&}>LOu?5Dzmt(yT_jGxDLgh_3zZMZK=IJCXrJp@ zG-dNW_Q2k0puM@4@sddbO6#?VPh&aJKJo{18m|jj%3JAyPJQSV>4CRC_7(VA=3!ZG zCLEJ;z;}QBf|`B#aE5X}oGm>AIj6XBA2dmGKZ)sc@6T9_8qP@o70WJExV#vK2)J3V zaVV6T%oJp_jxsA9;sidHM`T5=2Q2z|6cxvtbFB+>xEp^k+@>JFHg*@1djC2S(%g#v zY^p?0H)?UK567dmI1gF5`J;30M__3d7fUMjqy7*lw7~ush%y)>&$B2AdgM>%ty_y# z8*NcgB7$v}3&6&#ULq{Vq32zN=#6I_tp9ljI$707!mk{L<2u7YWZOBYe1x#?*UEs2 zICF5}*gM#6^9AmB5JC<&-hkINp2B-eHX+^GMxgP=2Hr}@M8EpWV3*b^a%k@=^yLl* zHml}=@ii_mtfLF93ZKvIe<{OF@n3?-vIXe7)+MlDwlU0m)BtV-nILA%E6_d32Y-)y z!AAil&}PzzeA(qgE*@=T*S@wS*XKN7w_i}>%2Y1nMzdPn;vzRtWtB{Fva+G)UL`D3 zRDm|>-b4Q-93r+dGb!;Fu;@DbvCN7Y#Ckv;ENPhyt(!Ix_p&h?#ez-b(bB7^+Qg7J z#^|A4%XHC9aTOGIR}|gd9!h%FT*BLnwL!y?c)?zzSZyQgvQrxl<-hcBVeSQ|=DnIs9* zCh&5NLi&)~Mv|vJX9pW1kn8mqXt8b{%Xt+*1@$Po`;ipaJI% zV;#Fm)2(~N|F8}U=AK99g^;cdEn#_G^U=z1PinPp9-kEKC(7eTNZa3~WTNyn*nQj$ zoo@b*yqESO=PQh9UG#T0Mq2{Sj|n2Xw|nE*ovA3Kc@1oo*aCgE6{+3MCq!}S1H9_T z7BZlCjD0+^fy|d}Cf!yKNVL&$_@N}8ZaZ#8b5sqe87GH6a_^_Y?Y1~VV=W_A_?r_{ zzk~cK`S0Pl7aWG z@UiB4di-Ds>J>ds{lApcZPP2^sRhSjSNsup-nNs}>Q|D>JtxWjH?3H5wk|sP-IzY; zoktS7?~(4N`Ow|m7KYtPCQ3K*$nC)EP~Q7I*?P!>b{6)Kieta2p2BjlsJR?EYb)~A zZ#&_vVNKdUKa+Mj-UL4Gb5LraE7InN(`64tY22XTJ-I}e0H;`dF53@p>z_}rpDAKL zZo>4K_&K<|ypi!Ax=Ci7Jwd%rd?3AE0krk~Y7~+&OhWYRs9K6Q;WssdZE?P2*Tai6 zpT86NOLo(jOC9(+dyV+Hl3skZ;lni4IR-WzwV^RhH)-SZm-M@{4ZkD%G#xz`OZyll z`eCS-vwpyanZBu<(JwkqPDq~xj}KIlu#d@1c4Qzj@pum2z0{(I|4qQDM`qEA)h#s1 z)PRae-=~>R4Pf7$g@Suh2VLJao4kIt0bNpxgO-(zL`8KJzj^%(oQ~;clG-O3S=Ak2 z{6{`?eR>_P8Ly%d-hMEBhbUFtFStIxVrci)j^^-Mh|TwToJ;=MqkI%;c-skyMEmL$&hx1>M${L zf3abAiXdaE!ZH8q$L>3;39@6wIFA=b+13V5Cl2a$JRN+6y;Qo9bcCGb%q&Y|;`)!` zl1PS?F&2RmRTrU|wHJ{5xdk>eT6$AlWto)H1xsLtZ>(?5Yr z@k;2i(E?_fglOHS=PkSlVPPity#+{{GXuMhE8;s-Qvi7u1mABoVoIv}aPnPU63w5q z6-s``QT%m`bG`2|$7JUS3BH(#|9dLT+k116bLXH1%^MWr11khwaNK4zT~W}3A1d0 zBqtM1Wp3uqW_EGkadvN2Vb&lS=7!vPprddG9JfTcTK^_6@;nJs{Dj!y$AHy{{0FW} z&BVb!wgAx)1S?XUn7h;RIA_JRh}_~KtbI~~`KopbZoRL|>?!O3c_R0CMfqhszd4QI z+*WM?p0j?xM;)eJDm(6FOOx zl7V@xx8X)g;1pGZG z9et{!(sQr}Y1G`CPR^}R1{OTR-xgvIEL_E@-X`4p;tbw2Xw z5artRJj0XL)o_*79o#Y=PQEUeBd#mF$bi>HXxJNsf+t#$mmGv<_P?MfXDYYkUod!4 zZ;4lGI1#hu8_~0OLFh)#9@0G|ismdYM7>&Dxs9KW(5RCwG|MP~COW%vqt4uis}D@D zC#)OkfA+>iC|8||?)IZ^u4khY0R!m8DSu=aWXv7Y;X~);w|GBW)}TEfXQ4Z5FT=vU z*+}&879zEy7C1_BNQHzBjP(2nH*Q48eTo=YZJj=vWuZk{Z>@m~6aG?3nM`sPRMG-{ zG5G6TB-}P=2dn2c0e#6iaAv_A^qcnyZ8ms^EUr3mJcM|l`l3cD@1cW3;)q z%0^I_?{s=ja~-|pXhmCn=A)^Fk?5wi4sxk~L%f|m$W9v>s2ueg=G?KTH3nz!E9ygCAVnGZgr;{E9gXV8a&`!7xs7e`h4Y}0qD!`s7O#2=#-)lZpd8^}4L zvmHzIJjR~!LNr8qUABpi3 z^J7TE${_GBY^R`qsihS=9@65ZpCmA22o~8s#*Z~mBg+k!$x_kz#Li$Ui8DV-4z)^A zwP6`jmwu3c=<5l7b72(!^x|VMAxJPUUHc4MeC{D}Tqm;W87A-klO!2~8RW%uJ+#2u zlmFd(E?ROS3q7Vv)GJVrzVbET|CugFcjlf0%Kalg|u)Yi%-7djv&Pt;r zBT~S77sPAkK7}2QLxNhKPx~T+sm6#ZieD>A4jc7gzXexd$CD5?V);raBPxQ*Oh@6| z^yPT|XJKx0n<@SMBZ+VV9O>xRE_!?MR(kQ@X?krzADd_}ovN9((Sa={WOtQ5^tFgV zr({m>vwt}9PZkf+E~^W4L60jPw``+@Zc^0yLJa+9dYk0gHIS?lMY< z;^iMNqxLn^_;)>=sPofbL^`S!9@?^7mFle-4f z!*eNlnMQ>(I|X@CJbJ$I7>$2&gkCx_2zCAUBUNr0`f-zm+eZY~r&Nkd-_9l&tw)E( zN0H(K5hSrm8F9@-nI4Q~$!b8EYvL9>7aa^=)f5FM;WHk1j` zokRZoN0q(w#{v^tx6>bPcoz?gJkyC^@lB>9(UT)}z64u@8lu2g$uvXr2HiLCf`s%?Q|7x&B5txykU1zd**3k(H4>roij8;2B(zY#L=j*Op4K%NT( z9wDJXNduWX*^9X;QYclRO^a&q z6YQ3&;hwR1^!?IZv~{KstXwMKLcLrK$lYqNH}e26tKiawNd`DPHW#*CyaqK4r=nEn zAym5&BhfSQ=(@%_xIEa7u4;{;-!}>K56;%(cV1h>cbaS_Q`Ntc8o_xwHuCVZwE?8v zL%^It7nsGfDq(U~1?S~JJmsxEPCc8W=;*Ywus&WFT&q_`KFgNz72jI(=Xx0NOU`a3 zrTeDyXFqGA^@6!@`=xd8iWZ0a7^PDCIdN3V+nq{E&4H753gEWiqTFlz5PGI_3yi*8 zPClF4q4X_t(4tg|_B=C(gJbuZM_)bZ+4jxcy__7xSRaLJqZ~lNQA3!pZZ9!iJ`NXh zcc6h!tI(C$ZRonWKANtz8D-A<0Pe3xUzzC~Rk7O2ab-iT6Q35C10uM^@}Yn`WlM%?{7eKEK7Nx;_bhjJSi@khdgm z%NZ&u`VQ@pDnfoLa_I4mi^x3eJ&I_$i8jb>0&MmhR;bmUj2x~-L|g*hla4~tdyLRa zdkggXK@hC&4Moor&ml)0Arxj*2T8vPc&JqZw!3_1{pEvEk(C8HY16}vMY-I(O<%cb zfl`ov#fW;AUZ58yvyuGlcsQ{823)x7CHh{t08J5i@->S3;LP}P$Pwfbe@=@+e}B(K z;$z7upQgjko-*9oAxd*wJz!AqOmy912A%d!7u;X3&-4krPRmOjncnE>=xv@Snd)qV znw;0dPwHZ{3J-ukN%_x-6x%e8dTL1{~F`N`Wz_ufZzHsu$(N_hewUN_4B zn1Ry8vWZ#>CM(8YGY51Ifg69{z=wy;;F|uq^u7n8*FS2Yu{HfL<$MjAoOyw{anX`@ zKsOEa+O2`h_agLua20+dFO9nMB+!W#h!U@ha<>CzRO{)1)^7FZ_AK7Qz1AngmGa{u z*83pxa7#rGC+2Zgo+@+a?x{oQVIR?oNumw~_0&~%E!i`209Xxdgo&aT(4!^Skk!9S z#O^{WTp=_E1)VA)_EAMJMUaX1HwP0dxdg(meL##g>nO`#$qQ?X2m23%lh;4qp&OOO zXsi2FuG_ksSZ#X(RcqNyt>eZzM=S8o82?cdFasdxv=%y}#6wsQ61^tAl@3!;iq%cK4Xhob;iAat<`tqJvlD?$rk zo_ZuXIdO&DNRA`5M!Ts>y%7JRvlX@wHUe_}p2X;N0TMhY(e0opG@4z4VkVa&1yh#GCQFJHMe!dLt65Tu8c83F=e1 z=ZSM9D{PpgpH+i@#&2*Kc~fvtwFYT&4dL#A)yRI{B4lQwizJ&4qkmy%;IY`NWV`t- z9KS=7zoWaEz7;Tt*KCzSdnVG6%sMf2M_U1Tzc5A1On0Nd>MXnY^9wR>b}!xa!Wvu& z2|+EDhVV=BUT{Nm6LdOci01B9K(7wxLCfqabgE1Pn|iD!DGsX0RPzYj)bj)A@AN?( z)d)u-Z4j0DnKR?jEg;|G4*x?%a7ybGWF>MLt@Aj9#3X-06|05Ft-b*!6*|()p$KaH3@uGdy?oX>;yr(KEU#rL=tWM3X`WV;L0U2Xwu^$j7=06{Jftt>CJcHO~G%Q z<~YNmNFLaAnGe>_Ng~ppd~n9J8MN+A4B5->BTpvu1q(zBCj@=&%7b=DD^v>2*T0P( z8K|SQ`Q1eL#}e4rZ4KYOI1is)@kWJDoXFn9g+S%c3Rn^<2W1|Q@V(3*X5@PtNxfUl40ucsUExLsq!$wHjh(orNfyV< zmqwoyxbV?-BNE8`#8!7o1^4$$;PBm9z;xaUEBfXk$pz(zQ}_H+2bGIQuKS{O=to zH(dtJo_L|~Ekazo__^ED9zH&8 z5+1rB!P{}Kf|Y8T3400-+jQkgabyc?nMDo;yvtMc*|eA`P@?r3&uhK?G!UHQZ2Pn-GGvCz47QrerL_C}Ff&k_-vFEQ_seCM+| zUg`}SU!hc>t2boz;KXiQ>62Ypu44mqQFtP2x$rJ{C;XG=5f;p@I+_B$t+WB*)>Ss-$petwl2L!LDBDK4v4)r5&gcD9 z&j&G{zktv~A?))0EhG1LAuOF!Q~%_YF7U+&q&PQ$naADh)D{1M(ybR@!`6TGg};Z` zv7J)nEi~bMPnyoUI&@%l;$+Pnh^{Rue}yFZZ+L=5r;VhD2AX)Y#72DSZ{Cl6*2sPD zI>u#BGpF^k0r3j@|-*1(f_opqaHfcCBPA0L99Gjq@N2^Qbftcr{|ea zry$!pzA)!t)^wiP?G~V#WDXbU9j&{qTxB&)`8kqelX;6G_trJd`UE|DMy(q6Db*KQ z*r1c9p;oh|J5;;*wLq6Qo9lfy$nkdTi6ggFQuxxhHNbLCoNbwbEU(YF6x8G z3v^)5KpF2XQ^#&<4r1q4*#f1iS|IN~#=BFcSFg5!%gFeeAlDzC*woCwJnb%ndbn1$ zHZ(NTs`lBWP5v@-CjR4_TCc11JjY9}yc?}snTBq`j8MMT+BTcm9NhE~xd?bBSz`LE z`J*Jxn9NQP@A%nj#nZhupI?V@O0><*W|60LBac7u$}WAc8SqKvgxz_^GZ<`V*gghz zoO#Q6U|Vi8P3H&LX`9FMjyPdc`Q4;Gr!tIVllGqTVRsVVnR3DEVQf|PI=`b(J-v)g zi_`-KQTt&>NGGfLsUMt690u0!do2T$NzLAc+PucD9;>nzD;uq~;$U9ibk5h@R;-bn&53YX!7QV`t(XS+uJfb>-uBA^5l%-nf=>r(C-7H z_^R4Zke5}>Yh4DwmMOO2o>N(U)3`mb4^)L_mTesOn9925&Ej>jd)n(aN5%tbO;Oma zU{B_Lp*Hh$1z9NjT6q(qN9tb0h}wipS>XDFCZe(ZI2C_YMN?Nx^G8@sep=WUoE_5w zI%~|a=;$HTv+^DeU*9% zA*r#2Na2ApdS+sXqMCM5uH;1mFA4Tc-59Dk8lfl&d-P%PPx$_+G@4uKizIff5!BvF zDAqgyPmXr*v||)Wyw3vERyZG>X;%PF(TAv1X*Tt=PlScJ?nLF%Od!>Q+5MUmFzMwK znq;}YUfjI|!Ae==;uAqSf2JeLqDkbrunf1CX5oNk)5w5%60SKxmonuP z$Wz4y%gO5ZR`lFSO=|T}0+ntYAo77X$dA)|$TTSx@~ua}(OmEedzxfHr-WA6eeD|g zZgm}jSLWP+4N6?svybuF#i_(ezL^~8R6(y)E5YQ$a+nrY2UkT-fy-K)pmB63qq-(b zkbN2eH{7OZW@stW+r0&~JuL@U)s+Ym{(?KZb-@mB^Poo~^RZ!2ist6ju#Lkl#F*bhfDPe7eZA4z(s5)B%5 zgYUjO(u6A+L@=2GZn>T0`=uvjaqA=wy|jfy3^$Psu)oh}6bgsCn68)OLG1k^h^GHpi7Bu_YVu+C7+1wLIE?C6pAL3qjMAJW>DsA3);P z1#0A0h2CbTpusjs;*+$|6lDeKb7c-{?DWOSlgC)BkPojZ6k~n2&qV$5F!)fd4Nv|D zVWWIBS{Rmxu0J?I_NN`?N!hJrcN8e1^TU_X3k?yJan1t1xURsTKDZIA7M#Yg>!FbsYC+M2efR!60$swN&$V|tdFza+U-$lpC z$@SSZNZo^;8%`nj-(H5{J1T+x*(!*ivXDso!iVec!D~NKpsig6l!|mD3wB7et!4$V zHTw}<)NTeJ>m46B(wwjBr$gEp_!fNOS=fVm#TQ~i34 zoKdsDuY``1)v}*?f5g5JZ>h_qagegUv2A#@!*6II*eyv!OQM68Pe@4eK6rE445H7- zGoP)R*^9ToVd4=9oc;h|}b_b_}-7cY^RI{Ar+;t(6LmgzxJV0i* zv1ArLMtlW6b-Dg>7&K{&Lbl$6?E;ocg_JZoyEcl%%NWB-nG;m~UM>achbe8KB-r*a z3F-2u_Jy&q$Wj!!^gmz>?&QM}Aq^5!A%s(8F=!jEB~{H1?8-DHc)|2N?j4&!k7mcv za}Q_I6&|aoxwbA^G4l>cZ7`y&*=E)xQ5KZP%5(1SX(LsAM`-c6OzO~gh~2-i9()qL zMsgoU1F>Lj_#-2X*xAY93n(dFfKS+|;^lr$v(82X}Q@|I2e&$Dv2vvUm zm8=fkMJli~Dt%S~TvY{rBtd>gUDZ;2;Wz^vk4)pV8cXA(E zGRb#8#loH4q(D!M>TVXIx1%P=ei(qwGR?uVK!ER8$Ah4Q*MQa^D_kyi8ymMJvB6RC zQlmv5yq#_ z7h~VOGX(zkrC4*@RBcqz1}rbJ2 zEg4x23K8wCDD9$B$w+B^KKGt;&5D!?**mFJ2#JXN?(a|N(fyoz&-?v)J)ezW;i(Xq zT5=iySuME8$_^0IOmIYQI}_Hmfkm(7p@iv7HddNP$8Cy{O-v-$J>mi)1(TMjbWx}_f`tXqQRvhT+MZahYXYUwik@Ln8WU{sm+5SWg zCfY#wecoC)ce4P$120&Qp#iFMYyw$rUyoNOJpkZkmdNGRb&x&0i!4bQWU?hR;l9$} z*d=5uIGGjBPK!7ps^|3Bkxg&e>JO3l#~%&wYvz4+=A`R*;jDdl{^w%;jJ2TqYwqC# zo6A9_*>k+d!wp+K^`t$<#9)@|6znLYapRY31uFY&gq&NO|N((PP7ez1O^hw#ZIMBOd zI?6byhE#(;<46CjuvpU#yldJfwzPf++f{GJTHgo;lvpx2JNX7MkO{+1j*Zy-a1nl* z{}f2IRpRQc$Hko9ss0Gb#B-Q-c?~66i zIwLd_=Is?H`dL0?tiuVO6X>9JkWEy0qnmt6TW)bK&Qv8 zBP+Uk;O3>0=-Wb;vQwMP9)CMXe`>NuwxauRf8%)eZCwgXPptsHJToxQ=oyPO3q(zR zB5X0H3R5hsAj}p4$)QyEp?nly+}wxdn$nPSxDRTxwMLhDwb;ypYAC5N4Sc;KN8WlW zqRj!WXbJD7zIiSe%G^JNZ+y~(31S8y{Mat4b;Bz>M?xGQx)+U~=ud?5DHy+bQ^cHX zFo4+wdT8R~)hKJ7CbUebgzk&uXmm>xonEC5jb${U^_yZ=@AOG9GiD-ur)!74gmwABp}NL0=$Ep%rL>u{o`SHb2uB;+5&ZwOySnDzwGx!HE_!6INW9>fsTFA zVw)1V`zW#GjN zcdUF;1_l*KBLkg@=)=XIz+}%W)_7Qi*Zz=%SUJx^C2=d5H!T#sSR_Dm_U(jRuZa2Q zdyiezzlCaiWD0-mT7@^xDilOLwZ*})Q`pn3h4_N;Ag+}2Wg{m**04ehO*kKqD&!;K ztLzAH1`n|s?#XP}8Y|lDT{^ZQO1R(Dn!VomLbPUWD5$y{AX@&r5SI6pu&ZWy;V?5R z&>y^(o|zU2>0d{n#)&exHf=J@v?!t%RfYpa9EGEw4A7ekBS33O4xD$T2HswC1}gF# zuPw>PfaB4#7{y%1bI-}3B>}eZ%N-N^PcojF-?f3yDkoVyFCK%dYS$olJ|8k8kOi~- z#jyK|Z|tKRAJ}mJJ0SJGH2B~gMv0Ie98ipbcS5INx0{>cAwJW%i|@=XvW;LDt~v`l z0tfKihYt9-GT$-oN`xvs|B;?&86t_T>G;S6hV^zy1V86!09JDqE;depk%{TBFzz8x z3mRgr|0n=#?1xvjorO+~j_~~kY1l3AiyYVMp*YK>Kw32#ON`$OA7>d+n!EDA(rsF7 z-;rASk?Ujl^7{_>YABIjI#vk2RR3nI^8*;SIjiBvkqML@y_uHZ?!;;qd(uy25~z@h zAky^B1DzP$1&8!kL#3IKuyIO)aP=+`{G4zCI_4Kq&Eo--`1Ol1N<1d8+MNu&L$82l ztwQQ`MXDp2Brz!V9v(^x;$tmyuAC4sQBDOL4lSz zN-yt*wvAiCibwlNXNEi~mMVcW&K!b0Jx}Q1H*cx@Us7Q3d$VZG8!Y^`K_GOb~#ziJGA z-D-fID}}=|Mz>%{$7TF-AP(mB?4WbXbK$L>&qV!-8KQv?D(uwbub@$VIh?G275Z+B zMJ;}*AZ6EmaQI6qT(ne)X2K4__KlZWx20!+^Qtk%_q`1>IeM{0`jZ!6My428Hs6WS zYzl@a!D^sUYzKM_w&P@RX|{Z{fLis!6NX4N!yXFjWv}xYm!W(-XZc&;zoi6f{_&#Y)MvwnhLcR+!g@M&hZ0@M6@d#o zf56Fx*TCkHHd?oIAxv+JgF9YcU=pmI!IO+mK=v#UebCY)5Bcn#sYVricvBM*PhC_s z<16*?P7k|#vIKe=G7dJ5zkpv@ronS(14K)@{9ts}4Y!6KR^PaPt^;By62jHG}koCQ^3863ns1E;p zb+xl?*CJv0v)S;fS2%mb{4_kg#uPm)6~T|b=|F6Ar)a-0hg}+Qj=41;Kp}hiUHn-B zee^P+R?H&c%(=toibi15`edeIWE56g+<~9nM0DSVM)1B+6U_d+oAEd806Fj8G5h7T zY4b@9d-sk&l;!wH;h zGO3#hyfXa|lVe^>UyD|O|G*W-<*yiqT|a3J1&VbOON0{xqG9)^wejINu698y*OZjXqVAEpp`C5k2R+P`)v?f7FAGF56)nJ zAE*O-o}Q*A*U~5No)&E?*B3Sa0_?tUEpV{#Gc_*E1NVNMLAGZTiW;^PweNinHJ3Is zJwdlA|D-g*dUHKWZjupro-&Il^OM}~Gyk&5b`p5c1`i5Nxk7g*He+4a5K3#+qq-m8 zW5LxquSA^J7fN=)Z{U}?UD(3!WbQ4<#bM@3lui76Mj_FJ%bYz|)a{~V(fT)n+E{GG zbCbjAX?2T%TVfs4o$?i~03L9#Pls8XwhLa3(}%yOeH6J(YZIRNlE4}%iP5hx#$Kl4 zM6o!7zSg9HrPpixpD!m;`l*Rk`-d2}WFz7GkNU!YDizd-nkv&$r^#&Eb{x(xOBTMW zxk>rzS^>XP;!Ny=EYY}&8DPHcmb%47)A(7DG`{t+R(RDm3cRhn%Q}Z`rhhgYz%ag3 z__51~SuSh>(O)B3hgMCdea;y=!dDVbI+7x~<}_9G>GMRkvRPSlVotwkTF`7be&u`m zU4A6!v6@Jlmm4x~^e!^ilq!5a_JeUcp#hgaFM!MKt_$OutpJFx1+||O>7p}TOy|oZ zI7hEfl(lQRs5B~vJ*|DFuBp_Qc9)ibp+m8BpgN5sD`VJ}i}6g@f??6cnK5t})Ke33 z?L?ncBcZel@8&CNg8sWVg0ErA*gr$>fa@(2EPcOJbbr}Z(JN;S(2P=<-&} z<}KxTYgBYzo$rTtYtzy$L5z1K!(@l%2;JjGM3FSb=1cCO&vkiVJ;{~K`oCpVAp4MN zv>0LMkAw>}%+9da|DXB)q0UbGnFqJK*x}uW8>rCtmO>dnB-(cQKQ_-BVx3#cMB39x zB!60&-LWNyEh+X9th_1HC zH?OnPSD;PzQ&j(w4)awfheckI`{-C*aYmuI!y;7YteKtrB1*&DlD_wPH}&VV2K}f= z5$MdQqrIHM1)Ue>fGzDiMK@NBtFxSSysqrwdBN~rN#TBvL=y$8pTO+ESqB@>eiiv7LS>=SGjjMABNvo9Wb1Aj&OXB0TAGgI=3)n#tWd zkv{Ja0T;*71+(+$f~Fnd^fxDZXXX966AGK?$fD)6@qjL+{>Fm(o3V~MS+-P!27Xgu z%_&-SkDr##$z(z3m#k4{zB+Fn0T-?$^9D^^s zUu?0+PCOv+-`pdZC#FH2x5}oQj}|az?&MG(myH)aS@}gYuu_F7t2|0S+SvvQe#cl0 zndAt%!*Ydfi)!jRk`4-);SRxiiQ{!KllKd%I*!)GI9?L0X`V0KGQ5x4xb1bV(XVyX z)``yq>fd)WoibYB%j+TLcB&(#y}FPp{$s*!j$13dWUnat{MA_`_He04`9+H8QQKYW z`L=OXw$5J4S}Y3O?5-5Oz^%2)8O6M-SYJpDX4dJn57TYRouZmwi$o497O>xbtJ8bp zkBV;BsIw-^lc|g*NjgSqq3GVZBf=94Rp?!HJA{`X|7THMzE03~y^K2FYG(0$(@gru zg#zjys;i~6c8OF4&D6#QWlG(c77qAOG){O$HRN0r9rI*?Q@#NcwOj}IX34M?c1H!% zGX%_K>uqazDYl2ub=vRy@5+k`oBus|)TGP5=XM?h2+%juw0xZLZsI z#Wh_3t@2d6{&9G>Q=&p@Rv0=dM74-U7K~(9TXCl9X5c+AY zvALS2w#X(ag1Y0igsN{#6Zu=3P*DfuMRT5-fcvJB;IQ0Gi^MK*ei zIxoRw(OUXrUBl!hLU-rsJj-1~ohcI+y7tFYdlf`<#MiI%pjtGzBb7lN4LCx-)f#VM z+UZWqj=f?u2K?EO9|r{?n&P5`RmYgUjd_e-kOlxVsu(PtM`4HSv>;R)4BE>IOCMNL z9C}Jkf13h-Ub2ChE25p%&yffv2jXez34}lbKH1VNDqhR)S4`#LLkUZAefK@Q-|jUV z7S;eam`UP^hH+?7)+Xp}Qp`F%%*8i7o0vC=V$k*9XV4X&MSHsnV9AIANy4VceCa5B zs2_ut3|xYDA0*&gdm>@=>e-0dJQW4S%tAGVt*}8Q85HUAyH}OSp450A^UHf}|_HxZmIrIr3(he27=(Ttu;G z?%`=@eBe`bG<_7l-0g}yGxox;ZMIOX%9<{U{{*7UuaGOTHXM*z#>sunfMpl-(a-mZ z*gRMU$yzQ_U1 zcD)bfr}V*Z2t<5EXtk@c5-$7gFtM?+y*s|i$DoB-#VH(|@>C4^qOl|+PU^DGvA7F&LrG-T}} zd-R`?cTZ=tzmA*%x73B~d45hmZATS~e6)g@mY0J@C+6=Vb>U{0Y*N;Lmni4TkuCi^ALh{rbWk^h##j@h=cn-S1|#xwAP;8- zC!p+8A3fpNntNO)le?6g>hVwa1-7aN*DwM887LFMO}PbQ=9 zU=>PpZva|W*?eD>NF?7*P1pk z;kW_}&@_MtPRWvT$^bu=nt_|Df&4R!@FJV zHI-;QV3*4%?`Z^yzdnIuVn5)qt0_>_7mA&-^tbC9*+b%>)J$E_P~o|eySp`556Po`Df^DUlz8CdJC^xh@sfq%4EEb5?4E@ zf@hmKlciPa*KnoN*;h9^c|!ULv-ROR+}OylNq`kPlAGbprXU7M`QlkY3Brg$+H6&=Di?kTf& zD}IRfY>_1W>PpNra1@wl&u8|YmZz^htY%fr?L>cK;^56Te;il*k=a^ul~VEh3}&Bl zU{mIMKpp8qX60>X;AY|v4!xTMZdt9taL5$P>`xZmcRN6m(-Zk`Pa)>g1?*WO5ev>cE<3diV8NPCJ4ej zr(=)H0dOesJhSLy6?LRKguR&%NlN>t!4e}5&(-WA_?IzTd}KM!zIu)f?I-|G;{cpw zxtQ4UzqKEcGhqUW$7>Hh=KE*Y*_gFrP&wNj^d#6~r&XWGfwGNw^UMr9GC7vh^P28u%~ylXUyJcJ zrAqojt3RG+UBVm?H=yQ+u41NHKW0+=hM0(3Q;*y&Se z;zv`~k@p_`?Cr><_^j4cQOwVqqA_VH9JPK9to(PHeZGQbj>foyah`|{3JT(w>e<^_c_^IVF7 zu5LZk=WRm|jjh4Ju^vCEZNzVkOu@oA9bm@rX;5ozMR9DMQF6A=9N9PkRtV#CJzK&gQ|I8gQh zoM@8A0WE=|zl+M4kIrV$>-#TuQ0pPv_3c0K&|m^Q9Fhn&Sic7j!z!TV${+TxpoL}S z4w8^9wNy%gCN>z!C+f2taiAH`#_=*|-Gx5TRLKsi?lZ?n&h)ZQ>?AfP--a043^H?@ z_b|EdyK$)NBvP7eh1cqMfqHfYo;ag+Df=D&p2=lfGLm6CDt^FB6h#(O%^=Qh1| zTdByT@FlZ+>n!$Z5~0`EnlQmCF~C_o0`3TUhQCVP1alV1lJ2WZi45N}ONpC}4cC|e znZ^+|YKkUV?#+;68)o2po5F$QZyK+C-U*TUtwk)4ddpxU%-K9bcw_s zRc=CS8(H^4pQ|-HL%AM^V~;eP1yioekny$l*!ETptGwU{7cS+?>3jj)#ISkXsT~tJ zFUOB$=e#sHxAHJF*tQeu@_y2P4^nYm-gNTtO9Jp3#F|0-7+e3KjQYvtB{+RByItl|!K`f}Cfd~f!UF)3T= zgMT7rWIbUy_)}C#vct2%8-)ub$<>mWf1J*FREChUGk5VD{S%zDe+0MV`d_@ZtC-ST zA&o(A4zZ}Hrw;SwW1EQ$U>8ur%HNXkPJ1IT?r8^>_|r)GYfr$`?#*Q4!eNqnco$c1 zy@A_RC*+ubT+XkshR;qHu~{cq5u1QEuHp0-uGPJZTky(7xu8||_u^`3&-$gs0;fwx zATcBn`M$dYm-Lx2qJ}iCwoHohdD_=l`ccH?aIcXEY|g?ReU zhpcvR4C9wpO>9<=l2Y%fWFxzV%&odjEOWz%a_dHNZl@VhS>i)Tfi?NJYBw>zdx9M0 zv(jC}i!7*l4}beF1ARJC?CI+-MT=CVxft(xTt$PBJ0hmeroX?!9=-MjuTU%p%atf1K_ME3oSBWjOEdH&9~! zm3ZBY#~w;D#4~w>y(}MvG*{0>H%0sm(K;S=N^V0joyEk!MT!f5sK+HbsB_2uHj|vk z=g~pCDB`h7nyZc0;!?g0;XubG(s`+xjNCV2%06D9+ONw2d$~y@B_$pFt({M{x*aFG zuRkG-vkkehZ!VrVgJ+pf+zr;PDInG-P9yDO#+Dmqd2+1}leu`47|!T@GnNY!W9Hn? z6RmeGBiZOJ89gY=`+Kt3{3+gevYIM3xxkR<&0C52%^K)7V z`5{@6xeuN$oP(Cx-R7kKT;#4?IK>@UBF?p%s&fCHU@SG=o(r0D4Sgt6M~a6LYg_Y& z`F?u{XpPN7yPu0t(d4bIpj8PX;mii}2vt4%`~QgBU$agYWcq zAPq$mWWFn(S+ll^vi#ft?mstSJzti9BV~~9j+}&Fu3p8%dJ9QFzBm`%VeeEi8l2mySbUqy=P7 zq6AU;ARzKG>8z637N~C)!)ohhqZ5Jv7%;Gul*G2MJ{oBtZ0#+capHrf-rR}AhCI-y zSS)^LP!BeF3Sh@fWBBRdO4j>J0-jh}fsg(xB%P6pT%}bpd=q{L32aS(e$WAM=Fw-c zQhhxecI+f%>lfpNQLphiWQ#Jt)(~fx-{jmLacKW=A~9Or2CbGGqs94?;e2@mxxZ0mm=!56xknLjdx~GM5A*Kp~?YC zbYp2cv{xH~y@s7kY)%V&dF4E^_PBs7tY(m1FTOL5mFj3pYZ-m=!(W^`*AV^bTu6?I zYm-Vf4Ps)XfKPO0BB!VT)H}NyE>a=TB<4x8%~;z;FWaGaVAGuB749S0&wYL7E~QScg21##H6X$umw60s-T z=c95WCW=3sNG^^GBk3;Q_%xG1yuKC@sqYnt9j}WXhE&1>yZw+!UM#Azxen8x`rlQ1|!=NM6i4Npb`@Im!!=2l=l61}TXM1Ks=GM(&K6Yv{v9liw#9wQ0Z{Yu zY_uu8o2l^XW^3z)kXX!fq;0qdN>_Qqif}`4r63dxRTSb9Qx1A~8-XAxFx*_Q{nxvvjb9@~ZEJ$o=C=>WBF z8T0vjHzXhSfvFZBPZ~e^fxTJf&^BT!>1MmwuIssQ{475-#m#~QCT+n9_KR7u1}}F0 z-yE0|vmHqk)xnJyPNAD}-S9!KIy&~!5OIat$W*dGF!!zl+_B*<79aPP?Vfc4E&P>) z7EKz!>Fxx)-D$*UM|7~ymydA$6ASWudjc``%xB`NCJPfPJZNpzc67V`D0nROjQ-gE z6n;@|p*CKK7x?|!26Rgzgn^qE(t|#IV9CxzLAjAc-T3pa?Bu;O>M~rO3)Y^WO9u_7 z2^*IVSr}#6qw4nSg1cDd>_wX&3AFH3SRlMUsw*rL{ub%Hm`oQY$bz=2 z2&UxK0s5;}H2r$f4|tNw5>A=0QDF4pvY_+K_qr0>4ODyIDD`vVY-;3o3w1APE3-o~ zS5Q2?!y?YOKrkocxUhH9SxVDd4N6SQ1K}~Z>7mV~@aQ)?`tRQjRP(i1zO#F-ZX~Wr z5F+=7_adCJI5$)xlHAdZvR3H|=IfhNi}(&|Se7Dv2`r-P550y;<{t#>vyKZbiwi)( zvkw;AT!$gI zJkv(g9_sU)94f%k1-)N4T`=Rzs32WThW_j;EuyBKq(_I78I3<#!b>8t%lV}z;fUOI zi)+h_g~idaaQ2|1u(#8$N@Dv%G|StQ`Ur1PD;t}T_|Y75$M(y0Bb$rS)#weizNQ*= z52H7uEuwM4Zs%FT_PbBusvXaTYsDq$+?06cVqdh-tusZSW$upDv&HM)cuEKh_#EFp zjjQy3cfQc>Q*2T1*Jpy`>0xzt9;s;4%!hQQrzTBrU(c@JOH&#kQt0HC0Bvu*r($G+ z0BKL8_ir1aHulO=7fe@+T7a+c*vz|BeOe;Q+HDQq*S(_+qmIMg|I}(XWm9JF1mf%^ z`>DcH0jx;&ixeo>8B}YUlx2Rt>;?ZmctE){U#NR8Gu?8}p)h*gbtSsXty-w_sF2#A zvYPr)ze_NAv9n-d;Y?v{ko91? z;AE!$u>HI%?Qt~5Vyt{K`lPg+O6qhH zDQh|d@7Jp-iN-A9>oY@WpbDBck6$cM`Hu)@8e9~~7MKYR{E|dk(!0PZwCQ)5hV=aC2I{rnAPf-=nDf9NGfC&mb@3qwC~`DHxU8;$S(|i8@akZc zMPhypvm}eA<@-)n?;qZeyjGL!ceJQ#6(DA&*71;CTnzQAwn+ zWp=c@<@rBsL+7y+2-&+<0DyrD7kkc2>uc-^F|wIQWRb;OAtT zOdrXp^G82(Am&41_Dq7g#Umz(oOO^sTzLXk4jde^Xt=ZR8y37la_=!bRkJqa2UvNC%qfk9r&mPE%$J#o} zIENx_uGHOtiyZe5!x3LH;&V@=@NE>$rwcIvE_|NzEV-jn#|8|%CknGh$@-uAWW4Yj zT6t^(KIl*hV{d&2Kq-s$H}vLNiu1t{9}0_`EFuGo=iyJsjvynUBWeGlPkOba(XS#2 zc;jOu+G_V6341&(pUsiCta{c3dzF*no1T5}quT^>vG6;R9Q+DW^<|NLk_e9s$&lc! zZ*cS*3LgG21+TeMh5K#pkx#qSIlUq=6mar1o<9FPyt(NuG3%1zMxSZ8{^26b+e%d?|+%F?mKN$vJS<`oIxMY{)eV&48w)qiKzRXtmVSD z6jyrQk$bB)jhpRp2=2%qVrQH-gThz0u;5fY*>fn1Jv2qi(m^oUQqW>yS$XLSibxob zGJW$w(6T&y>QD|Y^J+m&=d?)V$Z4XO`WV@V>?C0;tB~hcZE|S$7P5WT5d1un4a+7d zFlp5{*ryKT*ex3OSYy{<@*wGb?_|*#qf!b`pdlqC%W!U5)RdVaPx~1#~ zeap#L-;)_)y4+fwQn>2yE-ck(MNZWil8-;Hqo)T<(7Az!=wCVSDtU7s=O|qyZ4a%< zDv9f?dQKs(zN2Vq;q)0T)Vz=4?0Zm@K>qX8xN|6D77WnCT zqP-{MarT`gxbI{ziP_SD59Y{YBUud;GA$CWv~M8}k+CE_rxc#CPhrc%ONd~6JDJqe zf}>Y+WOxB0Et`~)YP|%p`fN^?%zY0FV=lw{^}X1|)emmegHS`UhjBZqgG%)%;JvYs zIbtm&#fNT@1)H}KWnfLxyvCCq1B!UL`)7RPyC=5(N?6;HBto%oS(m6*9J@t}Of}^* z=2e*Nebs{%o=qjElO~X{%a*vlUXrb4ikWJ|PdIH)F4<>sg2>-H!~WQ!Ngg+i@%yx| zIN5qNE?9Yi^v0&4W$z}UYe&|Ar%G{jrkXUn>!Sg1{Tz$&Cx2|V`a7E*x|Lj6a}gf# z(xI}Koh8#RNpm|q9670#&oJm;iI>kfjb+>V>E*S2Ryo0w&RbUQMe1_cR_gzS`lEH)yA7a-=NT>wQUhnoGgQ{oo-9!s$T%Rh^%?~1aK{JTU2YsS1 z!TWKe;z_LYeqwp@IR4bANn}KwSgR9(2L9@3NzN*)*Diy#dqc?? zX`Xrd*b~QWHzJW5OYq;OVj_E1i~IUKoDRH^2E&bJpvWze&)kk9aZG3p}A$9 zP|N27z9IJn_~y4U;R{R2hDsw&g+-j_AwCa8`@`dp3UNVn3VgA41V0()2rYlBk*394 zNT+NA8Jnul1`BE29$rhbpD7U))n7zgOPd6RT)}RQ>BM)z4gfN>(VqHNJZqaKIri@W zi8^J$UYMFl(uc;_DIs(50pI)hxz#yv?SVQm{X<~1of3KGUQG0pf8sld=g8}MvBXJl zF{z*XjcnxqE*X~}cxw>PRFW9RtNUqwCJ}@->}qC|OUm#Yv3k6v>o|F`@FkF6_7(hh zU79HK9BSRXAAFYeBpg3yIy@rF5!0M;2yDB5x}4 zed0*f-Gf9wLzi@ab7ns)&mc4Z@xBqv!tNwZbn3!CY`?_~AN(oL`?!kP2z@2`Pq#QL z*(w3|A1Y#frsd*CTn`Q$&BwW9E^Mld0(H;CUR5qDm?S@ zB5P&ZgTFO;W1FHG#b@qzVu=V^&0k3it1YxwtFM8w}%v!$aFn5)NH5f?falVG;l7*qF_>je?zJ%|E+YRpfFyicK;~VaPp((~p|4&V!TM9q;q-HDY}3ZmY>Iw1 zOrNib%g6X`bN4McFCiQ`IqYV2A3b2EHhg97s7>bb_vvzzJMWY8_mZJv-XrGpg9=)F zZY!I1FoQ_>P9`18nwaE^i}BmaOg6+`g_Zhi2(&IB_^+yw-D|oQKio!R?+?boGTl|4g1Mw|I8~_2 zo=%wxrb&vSd536b)4$o^QS5wn@g+S{(4YnXWvRf>`SI+Qu1MnKI~R{iyW!crnYixT z0(`VK9-1FcXFMKLxa>y>k(X{E!*u~fn;ONjGR>^&b!EJB!)sjr;~*5BZh*_z=^-lR z8OVE50kTVP!!)s8m@jh?C!`~st4PDB6nz-Kb``5Z0!U=kN}{v`lZQWw$&o#LZ)5Q> zSddWwJ?8}T+-pg8`*stw<<0Yk#uoH*Yc3mg=mUHo=fX-q+yTu4 zSHgN%Rp#vs3S?2ktgmDjQCJX9tWT8_e-{UmjaQK{$7{HW<#$lZ4%k`VpA>BELJ8vW z=qijrehao!BX4HlKJkg@tlxH=1H;LbL4VS_XB{WIV-r^c2gv@-V&v)(E3(*m54m?X zn_W^J%o>IlQx4-=uu^~%&h}Z)>S&c>wU*<=@arRdwI&SSiZv!eRX3Q~ewE+_nWR?u zk+>^sa?N3e+>eaQ#OMABxOUB3oI5WR=ZSbn*Wv%jn(j$t&)pZO+gIH3zvv`&~#vW-6xP}5GjkU57m&i#);hM@*m`CWfa4|m9YKoGIm8w z44WDog1Gc0D%6 z)d0=9Ekg|DAF%G->ZsHqfK#_t;JBjMoLkFb(qA-{a9L}JboOR4&AF0YtFsO%e3=Yq zs*Hn|tGvm)-U@=Mb;%k>-Uq5H&dt6$htqL({w9L6HDDZ=pWq(O7 z?ESfq(77wP9+x5VbiFCwFq#Pl_9w%eJ9hKu)m%~=un@UA&B1vRJ>d2%KkS!QgP*vr zgwdo1r%y)Yh;IOp8k&s4b_QNTT~J?vKt3SmyJRUBZ3KuIEcnCcmcMNg^XYhjiH?Z%fAmVh~32iBS z3|B4Wm5<6zbVySz%JYduVda6WR&6Stzey3>&gsU-_H;pq*=ym0{403L0XJ*@ZOIh?dI|eT~(}-sMnq)aZ@8Gd_E79NJ@c1>P5%&IXOS;H}dXEvFR5p>eJ?$T>3s zIy9=GhLe+!Q_*`|)0|HB)+y2R^X1^se-$uj$|E+?U_NU6bp)Mh4#C>r?BEAVgT1-9 z2*}l+gPNz~ack-buE9T$j@&);=kh#y1G$0K>QABiJ_F008eVwnWEePDga4wB_Sx9jG|57}d_%fofLI zgrC*!fo>%coV07YrT+zEf7Tn!A2vrluu+pd@bk}hQ_4@wE>di$w-F~U1i6aZurFpLnQvfnE z+z5Bs2gLF>@NMc(mJ$u(dG|jrM+4o1`JNqB_!&7S#pwkME zhf2bk)P1IOi3a{ebkU!@Ecmz79wnCCfv(3jknO$^S_8PkfTEe`KxHFrdQgPsZsIwb zIcBg~HwCBdxCb$#Lu}0Sh@SaR{Ppm2tkV%pX7^PRoArxuV@eIKkK`3RHKA-`XaLyMsz$@0C2B}?jm{*v z${IB+d5$zEM6o-Q+8JNnTVU<;n{0)AIW|ku$C0i((4;Smu&GBn%m~|pKgY}9w;xzG z>BB7cKkoE-a@_8T}#BAYeImxTYMb}}0F>lrpt8%i~l%$rORJ zac7v*>nxFQNC20Wy~pcu9-eHeLtYp0?zzK1@witV@Ndvk76=lE`cXf&cfwre{6Qo3 z1iSG6eka`9bQ`Z&6oTA)b&-FJ5t8Xh!7&do8=^8Fo~rZ1ZL3!Z7hijYtrHsQABT8G zeTFl<{lEeH+J0jX*XH3fYM;UAcPU)H?=<^AhR(#VrY8==E!r2QUD93@6)pG7%xyy= z5|Qjtb|Fz@O^PTjDlICNQnaGFXJ$^w9wmhm2^A&T{Yr{{=Wn=oK4<3pec$J~wKE_@i}y`vJl)E2%-V$%G^Oy z3lbrJpW3dvLl2q!BU}BnK=Ys_JUI3zI4x0z)ytzmXI=}P*xp40-!RmQ?w zb2LDbPA&eEKNiM6l|%1&53cN+ZuZwSC6qQNhjH-Bpw@d*$d?6Y$&JW|r2n8An)1(+ zs2!LAR?RX)hE&5sGh=6D*uSMY|$qAbC$w@J}=ZJV@LK&2T{YD%IMzB=?tCM0iUhPqIJFtNdu!m<+41m%9om#F3V zLONLCjOMhiqi+WW=t!Ou+1Qp&=Y000mhc6Umd_){Uq2%=7K@>et39dwljY=@r3LCt z_CUcFE!bV1XQ)(O!O2ON0ljsUBr80^!KEijo7FaAx2^|2=2;z|19pM`-X#$==Oilj zq#Q)#-6Xe=3SHr+#wT4B=$dX3t-=1{TmCuw}mDOy^-msYIWOSs|(fILcr!4FQ5hleovU|m9n_dBDB zA8#=?9E{+HYs19NYbJYt;!5o0u!=4e-#{akjDgtoX{h>(IXbVhiw=dHAanD75y{F@ zqJIQ{r_Y1XY}S#wzILQq537mZiaB(rcQ`FlFQI>I_u(10zX5rdy-eSmy|8a!8}-ob zr|0ausan`nBqnj5DHZC_LC1Qy`p8l$w!aPUZAlPq>ivZ0JLf>3{#Pi9pWhyX6NvQO zO3@3oa$p)+i(iH-!GKS(6?s?78Z7|nRsyi_ z_1fx_4tnf&7kAv+bDNcYxd3$KuZK8OlbzS?D%c`_3cUQ)B}g6($7XNF!HCJH@w|u! z0-3s7V5?p`d+gsHvURC5R0;ek`08>BCtIz=Vu9uCV1^F8`}HKqR9O!yZY~0aX-VXz zlQuY#c^|WBiX`aXQJiNujg;j@li``3Y;RBvxR`ShIF_o43T5ZBey!6PE!70HR6QAK zEh}YCwu&Q{*6~oyhQKGsRYhK-Ww7#$9FjVE#z7l{7mOl&I&L<0+_DVA|)6T)! zR|bJ+ur2)O{#~t0Rx4jya zxrkhSosq?^6Hv;23v}?mh60tJu#>&_;;?<*D5-iPlJGZyr2`Jg&fJ(Cd3Ff14?97b z$1X6k;|UY`SQXbz{?2Z!ddLV9#Mx_Y73_qXL14Yu5W1CBvAayqz`cuw0+@A|@oGJR zhqs=G9bvst`?frq`!fX|KAgse@m#F^CtN}LjstLRUN+>kN7>35A3>5=EWDuPCn#v# zMLf5bu@yH=(dXkgKzCgjo@;PJ5Er-w@2T(v-z=Kos`4WKj(ZmWIvolw2YRyOE=&-_ z#k;Wm4QYap4_ko!zo#ftToa)M70jL=4BO!-03pi5>~(1ef%Lfy_E2P^D8_~FO~~(N znw^`O4=?YF)T~ce-(B4-`k+4%d~H{kz|RHv$0&>9XU(i$KV3&) z*Bt>iuA0gi3j}QYYbV^$*T?Lw%EF?v2CUwC4XAZ%HWQlZ!!8cd6b%Tx*xX$U1kUbb z1WTS>VzisxnF9;UK#h76YpF8G8hHbD;_tuA3RN*=SaP2I!`Oo&V?+4+ZaDNkzZH(` zRA6iW@SLT|>)Al93g+4N^Pos62R>crk5z)@t1Xi@fJ00EvOYnDWKzQ(kS)Cm^>jZK zyv*2D>E8R7Jtv7MGpT@tXIv93+bqUxuv!Q@`!(6%%2^_f-WE!iYI8@Xj1rYZC0f1c z9vpKiA5IS}f`>9ap@+^KrcC=dIl8RgVRzd}a8N3lnEL$%ItsCLY_SV{cgT;%&$W{RJ-;U8QaTLv%w(2Mwj!u=nJ2^4j?aGs|ZIQg60I?ndcQ^3@1(7HmPx*eKwz ztp%@Wcfq~!3hY&`i#Bj_oYpUKZX+zBNn5W_uir!bPNte9-~0of-abpZZzPb&q!QB8 zeT|I%&Z1-wzk3=gpgGQx-1PY=T)^K{?(yjjmN5A5)Er>=K^aLa#zPM|Be?0o7ZP0-3w@;LVEqtRD6f)CWVj<_dVV3DGb@=2 zqkQRqB^qc%QI*JcRpNY+4p-ZDl(W&@!`*nY9KqGMc$d)%Xw|t0+3nzWJ)Ab#5+)DZ zTTBRdVhIcKWzhMzzo5DHM0z~Oi4LFCcG$mL8zV42O4h-MRQWj7qB$=My ztBWoEPN8)zJ#?UI7u^^_=(wo2G&xudDh%^`keoL7v8WV`iBce>q!=2;zlAFbl0?&9 z&VjdQ{v($Q_{_e|V&XKCN;jJ%)23;>w{TAoecjwjA{u)L7|I~iwq+ChO}g~`PgV4R z%F|hlIiC&uMMNoMDAQs>>;D#!iw785HoTr5IJu1GIi8?e?k@D@({a>k#u?0S`M{}J zGW1#Nefs8FJe^^3jrxH9$cHWULbn0QEL%er9_dk=Wy1t^^6u{2A>;?6 zNA}b$BYsJ%Noj{6wS25dt_v1((w0ZKJ*^pBQ)~lRtdL1!e2%g6-t#jkcP08&Nkl8% z&d}9*R+-(0+o5t<$B;vofKmV=Va? z{0t=Ng_F&3KiGiBez0O;3VmLAj1=j9C+BN#(}l(7=~JcKbR89tl}F6U`wdaN!(N}8 zE8)OxbyeWXGYaSfqm$(Mjxh4)W+>5}m`bB(YSK_Mem?MTDyqF|hZYq%z{5|C% zki0OStbG^G=lvSMp^Yctk?T1q^|PG7k1EkTpA~4m@^R>?_5*^H61dd*Iv#itL)Ioy zT)fkdR9>D3XGp3t>H!*bPlr34JQzV%o(UxnTGygoU%jBqN_Uj-Vu+wRG3tDE4#bS+KCq%#ZhDv9Hk z^Q1GilW70hPK|e7Maz9A3ai{Bi03JJPSvM^#_yg4Z3mN}tF-Wu9Jhc?+b|C zw+bM(;}iI|#Dr)Ve-k)v-GR1j-GJVd&j&$8t$34`J=`Hb25p_m^C$Gru_-&Zv+0Hc z=HIGRvQ^Co9~sz2f~_p*?HmuZcc!dR>A+^8!r1A;fMxxt@>UzV-P(gj=J1>no+0kt z`~W%!$DKe1MEgqEW4$JtzA3lAcLiK^|@{5zclQBXk7`Xe!?UtzfFAb4dcx zcEfDtCRcbq;|@H0>Kyv@u@!APokhm~9*evCu7de@+p+DFUg*Bh2k90|bMD^9sPqn7 zDqiWrROdzENmmb`^IgMG_Piom5aff_#9X04VHNbrvt@XXr;_MF`_7 z1jV`Vm0#(da0%{A$2AxgUW4XtSb|t?H`8uCm1NIa4<|HEB5GnS;Nv-acGRu}v=^9y z6rP=Su3ejEKjnS7TITdcM+{2saYDfl^r>~2x==P@rZ72qvT$0)MVu>)!XoRVWV+@A zqEli(F1>Su9lvYgN7e77q-h+8IPiva*{nuUpKinSMU#==jT{skKOPOM7m%ujTM#g; zN0N@`sQS+j6xZdFuWjb2^q4)$9d<#TGh%pu;6&6Pu8DqW)zg&OujyYNq2S|i9UHd1 zBKdA_;opk=Nq8C*CGasR!8^+V734r>mw{E&f`&=ts?&zR4uDTn0~V_9F3 z7B+i7A4m=yAp?JZfzje@V7mQ06x?5dRBE5VMK}2!khw3ccvS!uI$L4g>>Kc+PAJ?S zBMxQ?SI#vCN<{i(6y%!^&^NaQUW z{kJ7(pMN)+Hq}^|yzT)!Kqn)Oz=yC}$Bud9$by)HT8GUw5Ll$1r{908&|@i z9A^vxv;7R3{9q4I)ci|gWgdah_CfN_w+bGMpNKZvj77T^M-Zcx_0W??|4ejU1S;Tc ze6{Eg2vRRsB2vcxlg#atStowXEgcu{A^ILW5;Tv9gKoV$;OBq*y{pX*j9usr&)i0EzCj6drOX%wnA{=y6%}YpnFBOg z`5YFQn4!!$>Tt`2e89QrqmacF_6t(;1j+$Tj9X1Kw68e}@72$NHg*`Q={<*0m)^ne zH@|?;5_foENg%e!UkE4d)I*JVFUfi5(|F^uY%-kOP9*J)GJOwKKmS!5`#LMsseJ~N zn77~r(=a@K=6$Fo?*KM#Iz&QSouQKYeKM;c+P@tcJ)lkv&D_CzT+6=~}O|pkt?nD;pB%sNE|(!cu;0N>?`bidv&=Ihi+Vga{By1+;+eemOd>xxOyu%A5*wUb#a=GhK>t*v5vg1b3mR;xdCzV< zj%Q&+H^t!N6O@SF&?H!rww~OGyN1)RO=qu###aWu`HU~}-7JKq!Tt|1RnvD_1C5C% z*aRjY)H*6cGY4s6{xXE^4E}kgv|QSvaD-H6OsEviMe%x!EmXM!5t)e6KdQ6m~Ea|9^WR)A1 z0kSB7sAlD`+uL<%+<)`vK)EtK*dc>`_wQrs^P-6Mep#4#$df&plLYHE#-b6oF?dv9 zh$}yw1e=m~;WN0dL;NdpKs^`Po)Z} zleR1TeV`SGOrFO~J!DTNM99JiJ08(Vfw7o#Q^sGN)XCwb%!! z8jShwp!jNT?<6BG@!V!^;~`53sWhErC`C2zMj&DkyQxNRhaOa3n8h6eoTgytr) zqrnlCXBZ*)#Re5*UZmSx)oHHlBXa4XrV#JR6#Ckp78<-&po1!P%$YL-v^d`hoje>% zuiq{rCVg={L+29Jy`KS(ELul3&4SUEMG7QjlA`cnzb&ek-HM*|#i79#Kf3<230kdw ziR5}*rg_2JNkvl!xu5?82ix#{#q)tkxypjd#9NXNb&u)3J-yW1&ww)?I7iLOy2#WU zG1Sa@4wpJfncKWaooln*hfY|D3mfn^xN6&H;(jz7$=&v*znRapc2tv-0;b$xmcB6m zkOa*uT|}4r{-UAN{xA~#nZzybFmg5+fq7D@Xs+XWq2KGtLfhvvg<-oTxqXSM)N1G> z5lxY=S^ZeChKx?D`I1;j)@vov&}&+r@Zp8O&k z8`UJaE7jp##dK3nLQ0lf{9PU4;iYiL2?<(csmdXpJESr#6NZ=Y-GjzkbeV1mO_S4v zM=~fZ^-Ra*ci+(d<`k4W#hGUBHs=bm-D!l_3VMI)MPX%XwD60Dws4BXBd~U=81-$A zhC0q!NU8WJ8kAN;F2VZTR+&uB!h>+W1+%I8#|Sdxdly`@JDawJRS>HvL`$|_qnC~b zlFRX}I96>x(fcA#MGmj&D^V^tR3Fb3JwMF-T{(`JO&mvT`hLQpmqO%Jatih@6rz&M z`Dk|OCf?Ck%mlwLg`15Gfu7ZI?C%uEb(fyys!yEfTrKQS#jyv7yp=?puNFMuoCn12 zO(EjP^@X`F?1htR?9sj4|IonQXDF#ah?a)0q31NlaYd)5agE~Fp;4kd{aKhpDdHN@k<8vX9Um&vj@-=05q<$q^;7+sH4!0wxzXkP~~-;K)iV6nbka z_2=KJ_Uq$F@Ay(8|kn`=q7+-#q85;O{hDq8oUty%J%fdf&Aba)^PkhHZ9i}8QX{>%g70Azt%4XQqK0;X|Ywd%SCpVT18!~ zJP0A51X86Hf_uVM%%qGyW{pXpsQktxX6|bW&fL1qY>|Bd*J|ja=hvI;z05n{O1FHt z{@5uvSF@d6FfbXUjocJS+;4ZlOXgi0@)k{l$p#?zV;`O(S;`m|TBD2rf7Cre9jr8o z1bg$^7`fv~?6(Mghr>T+f)U@oYF&k~=*QAp*0S-uV2sm2xPUcc2bHz3@c0AxsAiMk z;MHr4N0%eYREYs0JAMeFzMO+&#{PlVkIM^~N=3oJqhIZUF2u0bhdv7)T&iOF_bOxJ zbrA%XE`etbYQVK_b+!-vD*~83#4@FF-H!_pv|!&IAAb zcMKZy`Bq<^KR5OacIXj?vtRU`rq`#G3id9r1A(iKF;7|aOE0n8ZZ;PTFL;Qj+>^2YXi|klV{Osr&lD6K z@y8iv&9L>E4zgG%g`_VUqlzbQs{g(|i)Dk)3g&wi0#AGsEDavV8kdK$j*q5dMdf&& z`P|9`)oXzA+1Bv8(6FkiHy!yt^+3D{O0W>?pvG@f=xS{TYrew~-RW>a^P|<-FpmZ3 zG|ypij=#d_l$L^r%P%k`6?M$Q`VY)ZW(nwAHv?qj0#x&09CGtOZ@f$nox8tK_ z(m1753a3V&29*tq*{!FiivFHVL3-6I&>U@bFmT}>xH9Aq=l_nW_A?j}?e>xYAF_91 zQ>$u5W^^U<>Zk$J-YXC7BIcrY-=oa+i{9w1|4C%sa0Ujfh^f-r9apt(t^yOe=Vw)B z`JAiz{XXoC;aSX0QHj9VXPy0qB5%PKH%qX6Obxg@a*MhD<3GH$IGx=yR|(-;UU2G_ zREH1C-#eUU-&Xon|E^k;vVm2aJ(0ccd6F#-y~}=HG98vijb*%blOecu4tl=$0@rP5 zfc*z$kvC}<*milbM$(O<-6}CvOH}`X!$V`>7PnfS{oBc^IPQSOj*tzsF!HsB%q1(I{e?V zM$vDXEzoOXF+( z==k@%1CK*e8kfh#8G;i)*Xy_IU zPoFS>E-PZe&4>b2yqKV++zmA3TL=uiCqcfoiz@dh!EukP;RoG~B$E^pL-qja|7$|y zpU5CME)8z%%S8KzGmxv@8>~C+A+>*ffh>IBPaB8>bjS_|zC1&%*mV{xJfnmt*JqEjS#@=shEqX()-a^NiA!tGn`tz0{HjlBDEe1RPeY1!e!0i zuGv=hg!X`_KQIBd&*G zkIt3|LZz1=gF!!Zs%a8(c^r;RHZ~)@heCGXMjCj(d_L@Qw6JaOQ5NPueTgI$HHox^ z4^%hkhSPC1nLl+mI8(omo~)69$*b>yOFuq?udSbOjqL_FHnfI0JlPwHuI@sg_v;JI zm+7Iu*2bt&bH4CyaSOHq>)O1uat7frmD;WMya+(wlG}tZI)Z z=>iY(H)R{MH@+Rnk2(UAU1CJ>hZMN$bO{g0F2oYR5NhW?hn^L=P_?`iZtv&Ma>y_< z^sWqAKoOd`GzfplH-~wjnn=o{|KO&$D6~{;Gg5l@3%nK?;&}!;z)c-jq%a?$f0Hht zZQ{zrbJb-inw-Mt&O z@CQ;dRS+im?gDc4Mx^1kGc?=v3hoX%iO#%Tf^Ph`0wvY15H=Pc5!$Su3qv}@z`C6g z0!gX6%n{X>pu-}LTsw(q)I={DfTmE1!-;VFJ|oob*#=JvSU9`@l9GLLs9v%Lh#3HZ z;rctEGE$ZnHyy3anlO*7&WS^_*RtsMt6!+~u`=Q6_zeHyaYFfJ(fH>k0s0}OjnX7m z3e@Z6VAanVXdq@9Q~kgP0dfg?Jc&gH(iDwND@6N3zkr(~;^^6(+q@$*0%qOYjKrWw zX!2SbTJ64xRb%^M==8PtXo(9bUC{_-6`iS#SsxVNsxSPZAdak(R-;3UJP}~zmd&t zcVLj48Jye0!gG%jV9VA+_A4J{Qh zLuZu#+zLIux)MZek0sCYR>HVX;;4@IL=4B2qr7O2XDA**_3ICznOPg)`yO#2=XMT> zc_@O~#U^C=Y&G<7G!eeP98GTL@MpWC0v}m;4}LSyfWc{Hz;YlIO>3Ei%>QmBO$iJe z(_={e`S*63f+`Uo|BKwWRfNkDX2WGtSAk>HYiR12g`PzY!Vf+3fCCJJ%qD-7eQy`K zn-_+By=)lepmauMf+u(`ybQnM0CaO^6Pj6;NmWM<2n-v}voq3o{^|`$JWaF%DTdk! zYpq8~-nesc^RsZUZ0aoh=TR59c{K&y_M%L=%m7onZyTJMZHW&?S;J*wjwpb+Ou~4c zuf55AlID=cK2uMEixh5P@A3DT7X81d-s=wfvd9_D_rs z+<{kY%Ey~p*F)QVG3Zz!gK}42MdmN1(c@G}p|7I3@W7|t5VTl9?}*R%+WI6meANh~ zpTp4!$tWzveB27!12!*xV#wym7p0r-K++b-02l8fQi7@t(r_4y~tS1QUjYmR$z-yr`Vy( ztH4f~Om@TAH`r`mE?`j^u75d~-RJ$1-PAu3IvRZi%dgpyI>|b)@L98{Uyfx`*eRe& zw}FiwsAGR@Q6t^=A2O>}Z^z$?I1u(^9oc(Z44jHmMsq##1-0YXG2nS4OHawM={9!+ zZGNTr{O$_8wZ8$z3}0s-+&d|Xe^v{=uBySZEf<)`4o|Q;XdeFWMu^~5(?yuNa{{cN zk<5raQp5|4yTOzx7ueA0!)(>t4P>6DIM0f@1*fhz5op!=Fden4m{Etjc)IB~!E%0& z^kR)Vw9NGqsV-g&Ul2KHaNmMi@@OSccytezgk5Bv?=N9Hril4 zYxBrOFK=LF?g-CJ(PR{QE?^^f6?omq7j8OJ%)IDUXL8=TFeV2jpr~;>d{w&7?p0hU zh#S%d5>J|$xX{;vHet75+i*GHx^;2v_6GKBm#Ra<@k{`3$-x)?TZr316R@cE1>QY- zIv$D>fr!PnaH-`%GB#ZUdMkF|FYo8!?S2yERbiW8mAn_(eCVhB$gLWIlaC?Tqoqh@ zbd3jgu~qQ8&0=PV)r1@C+F(S61#a1S)FHhO!@;Mvz$st_*xCME^q<*Td}zEFNt|&F zyk#$gS?`~KaV@){yoIPHK z8QBs!6e48`Z2prazn0#DOyM-<&H-oak@T3|lBkPKUB=*>qj%V@2eYx<-xPMg-D!NN zZ!Q*J@e6EPuRuCN^}yxNa`4@{t#EDKcERDWEKnmW30C_1g1cTu0(jeqN$fa|-`?3@ zef^^uuAX29FRSQ)^Rc_YpXc{Lo@p)!B>9ne?n#MWdbu@}1(z^>GOy!WCuiAj-TVq&%Nx7Nw9al&^n)Yb+v zT+iYcMgidcSs8f$>r!wrc|6t2dI-F}G2@R89!!@50(48h-r25Sq=pC4fa`*2? zuUrRiE znBs&qp4O(+K?VEp#;WWls%fpmu@YB|aCaU3!74cBLyjJv>XnF=5? zun2;C*U7)<7vSRDayDTok4)O8LD$OKff-ePBy_NXnPn1>YlAkSgO?Ve(Bd zuU1GeCCAdq@@cqex`pt`nmxF19N)72G=fT%c&1TGC91Icj^ZAkhFYd4kx$AGq^!uh zpC`%^@l+@BIQRn=*OWvj=EuQ}KmPFa;s|tV&tjp?N*L2&U9{@a^^fb71+{C05hUSrj~Bo!o?DWbnF-rL zGxo{YLzcFE1zne0!D!QcRNCwWJ$q+>Lr&qy>qj(N+qMf0EwM#u7mPrEVhd|vW`Msu z%p*g|J#csOG4wX5nXF$n23@GDM;o&*^IJ+&*f{1AxgT(y|3)t)?nx8qgQ?N@)DnI4 z)cF~FVSJLz?{c7l;wD(G&jL9wwS^nX#*lH$O?0?t8>*ja4X5y)8=IOBY_mo*xbprs z*3Wc8o@GJ8C*DdZWzHEiab-C|LRWH7b1QJ}R3`G1;sw@HYXv60Zt&ck48j9(v085? zSjHZLk1s(O-H`>FT@~@2@7d(PSCx=lE@UTcDaUeD!7frA#5-0IHPz2SQ|lz5<9Y}VtNW8*U&P4CpD}38%R2}}*`pjQ6}WHp z5Tl!M9sAPXcz9|(w!S_E2mIECf___KviKeB8#4`s>@`LC8%v30`X6F)f4W2bPJKp3Ip=# zYbbOzH3P+QhsoUsQb?nD0{S_&4b2vWgUK)t9>(m7oIYyn3R(DGM!m7J|H{-hqWP79*o7HKf|G z6)v>SfVzRF80|UN_;<^6q*1h7m>OU!dP>4T)A-G>y0`$omUqD6_XNBfEdZS=xroji zR3QIbC&<%`mrU!=IH=RS56%B^miLx@BcfAHRHN34GnW={kIX9=w~9^Rui{S8ZbhIxyR7;Sxt2Pw&mR3k-40~t3u|q`9~bbb&y79OYZgw2d=c;noAQq z%dN23$rWC);k46@VW|Ef?e9#X7w7&X%Hk%}>4PEN&Uh2cuJ^<|?GZeqXIInOG_U6L z)5$eW=^~;g_kvhO?xcb7*Xdq&HTpjC4UK#`KucFkavu^TxxU#;Sv9_kcjNC2^5VQ2 zH*vlq_pk32ZO}I3BC1|ci}oQp{W+x?n|nywtqD}CJqR%Vk$AQ!Xb)ILzc8tIod8at(cZa$0@|!pf&1CNC zDqk+~fGT81!M_FjXs`cTK6~p)+iWzc|Aq6^d+tSgomHkE zk9pIk(GrT*BvAR~x5<^6+vp{?5^DO&jfCH-B#X36DelRp;~##cG5Ts;eOn8?-ZF)2 zJf_Lj47zcPxFGIR;bHD&Z#k(?nof6yY$N9BVPxJnL)>bmMipNOxr6WiqhIhKof)ac zl|@={jq^WH-Lq?ngo+F|F;>8p&%Fff0#1{}{*T!6^&I*r^&5_DPNv~97l?X>AAK76 zoJzgE$%!RB;DY>mxEGS$)(a=+0@^>geslrqe~y3hlfh8(--{iH_?Ql5BS#HDYu1m z*Q;nKsToHudd$aASDd>mlIQkSXmJm%juVAEZQ83lm+FUY0H?LHVN;JKpQV=)_8UGz zOYN$W7d_4u%ldL(au#s?w_3<%y;{n5TV5Q%VrYX`BP9{}XT6WD=gr?AYwvBFboKhTSVgQ(F6 z3J0$r6lz?O6k17-LpSBdqg3l+3~fFR2vz zyIvLA_o)dBuCVAuZ7&+<@Dy^YwXp4i3H&Fx4!@i=gL}*L;NGDrr1SJwSb*(7fN32% z=X@NwC^n)yna0BL<`gY%P{NgW-hxR#?g4EcGp&_$MX)aVD3jl-ge!PA_wryPvQ~>D zImc(R=Q7;k$wmokx!V zFO;Y=7RpIDqg7HSDC?6FTHW~pDp+j9a9tF5)bE2*k6nWKwu`_UF$d)JHX3D`o<+A# zKw*a>623Dr75+Qq2yb;N;mA8K_|QRVq3H`b;Q>cQ`n^+#Vi$gZLkT%3tVCC6aj6aE z_tv0E2~wc^oDuWu(H_`Pt3b*u-{On^nGny$W*Dhtfn*N!!ZfeF=s|uQdbe^b&&M5L zHrGp|*%A^+@ujRVKfzQuEkZ^(x-ke8yUv8_n-$RMxgvO?xB_nfR*##;x`5mKtYC6? z8B9s)hPf-{&{|VhlykokM%Wl3i@%SNMg27BZ}gt5tiC}!BNs#I?(xuOR|Y=3D+N}h zW<%c}a_F>D1N>3C0uIcwg9F{aas0l^jCHRo$vag54LU=hnVlbgAFB<9e^|r1eQ(I8 z17T>U}T=Vf3T0iG1%1@F+Vu9(f#^5o0dBhP4wmyf4Z@h!&@3cc%4O685 z^0Dacj3ykv!if!>rh`7b`vdo1R!1dLPDr6n1fzb+q7SwA@f|P@{^qWc8P}^A=k-s) zJKZE&yZ8^XlemvA&yf-Ko?0&|3owDf=cj=eP5$W1zA~UY>jnzHvKro7qXyL)HDJTq z2s}2wkA;0g`Ym=d&!aj+_V`!g&X6@oT=_IQ;k^spv$unm?v}9k*&Fb5*F6vtrHH<* zy$X6-mLSjV!%)xr6pC-KM0IEVk@x-e=uhb#*!Oq_(#)QMLGv=<`M0*hgw=15%!gF` zvrQfyt*>QGNEa#?e*mxg6%zA9typQ&Kr_d!{U#aTy zUa;9T0+Nt$q`2h^?z}dM@!nj;;q;uHxPA5~lJ_&03|Gd}wH^EE&0Y`G z_|_Hf+@V4uJ#N9983E8$?K#{OFje^EgSxP8p^?z+o<7kTJB9kaEeG=tSRfmZd}tJ6 z1_#}WVEv>k@R(&c{LU2#R%K5`yNW67;F}ap=lq#c%|T$e!vgL8_+9W*rkU?>JfIAC zMymu1xjoDlm^SeUbUE2f=I)F_M<-lF)11#DMp%gU9^>=Wc@=1(#Bl^Xm4yxQ`>Cf1FYHI!osl8M~CKq48 zfAu?H$AP)1q(u=G`tlwHjnjBZ$r0#&H|7O6Mfl^S_`{ zkU}rCx=BQ1fiQgo$G{2ohpqkhN#BYIkXTtwG=g`NWs3WmH6bA+XC}rYoCECEm~m7k znWM<5$xtt`PU^o@8K#Yi!{ZBesW;k7B@;?XGL(?Szu8cj^ABE{tz{(GqiE#&J;pjo zkxc)6ogTT9O)Uk-=$EF)G+-5_5~~-``UPjGDaYEnzPFp2-P%d_yE>D^#W|#K#xz1I zKVk*hSTf(^EVJqn9GT~Shff)A6f;k zYR6IIs(Ca-$Bs&deWBiL3Dt5B{P@7;`6e{KVZ zp_g!LNEH^0Na9F^cr2qHWTU&a0qnOOg-zCrQPHD2@TYVN`!GnB?N-i4kedeE*7d;Q zlAXeK4@2tMkc{3vwnJr~6OnkiFBxC67~Nbqo!eVS5>qinx_v~9Xx!5WC7u6ZG_?Z_ zBt;OHubJ@V^LY^coyaO~Y~%AE#Zi%o5mA155T#%INkW??DF2ZfYv5D_ZhywoW%0>a zUnh|i*;^sM#56Fq8^f0QIkLS@ZtVVyTX6l|Bz9y=44|oF!8RlndiR`1H3wzrshN)S z&lDbNnj=GHr;WhfmI78k`ztFWtY!1OD`272V=y0PSygXwSXm?qKNEZL>_2hn*_|Z1 zaFPdAvF&4=b2H)*Tp z^FOle#-ehhVe>=CTYee7u?Jb(h;-KFX%g!{N19M2`wIRg2ebAcn(*`U)}0T!GCVT*yDF!pRd+9Gd=U6!nd4QX~D zxn%)-8t8_Kp0Vr)l{Tm@wt#Vl7vNr24u}jDV506Qa@%kfj*Jx{r59QF-qKd`)~gAb zN^+Ug#qW@O<7D`F^C0M_)We8)8GZd#m$b7 zqHu5_D!ns}QLgqN*Z-KY=jS-Fm0m|7DMF8KlQ>Q%>nx)mxf!3lmB3|y)l~lUn(3dxdf+cbzxSh4Ganv!29-LW=~)* z+Ob-bXg^$*Va&!5MD!t_Gr?ns@zZ3rZ=>|H+mhk&q2V7jk@zhsNr}zJb(;o9S z=4xmp`g|;iVhvv!xpoZu)@cT7cxoC8LFpuoFU_02VJ5#?^$}E-ae0q<)lfPd2w9fy zuwnB`bhMmLY(58)RpOmwKqHqq$#FRITyv@9np6Dg6$aqdVojVw>Pdw6V{(#=GUp7w zF%3)WkXm9pG1<@uORMg~!+UY?-0KR|+)V~EdtIp8mrV9(%fQU0Db&;7i;c)w!H$SU zvd_**AqCTLYI7qV-gF9K{`OC>*<>llAkRVT_DMr%gg9^z9uU=DfU##jf+MpZf)mff zy0P2AwbuuadW9Sd(ljwHLSu!`+00pePA{cs5$o$Li zAl4^e;EN%*vCq6kq7s)`q8AdnqGQ`z$$#paJ~P>U5cXEwyX!ptJ7Q)7_>s7&B!}npap)u16`6 z+{!p|H_n1~wg=GJ{kn9wf(Kpv)mxuaTz~Hm_1vC7 z%2ZB?CLGx#+JF4IXsq}yTE+1UI}Syojx#@Tysr%Hoq7Q~uCSte6Jy9}>;NH^4IWV6lNsIROC(;yOT`Dw!WpbYSZIeabItHR8KQl#Z1GW?0Cm-?t=Vo-@uMz&5PfvxZJKBY5p})wd ziy>srmZzjXSDs2t0D3Ykm-apwqKtYQQIb@kUh;=T(|3o825q*Ap8X0Cj^CtBhhu%{ ztAjg5>Mtur*Jj3wSfQTi$n5>}TC{{n`iC7|Yr2$_mQ&O#-djxdqqp?ltf34mog9OOH`q+4xfXKYc_qjEjlKpSfg5dnsh!hqT5;z zque{;qNt}TqD!7Vw5d&xR87x^cO19f;b;!d_{$l)eu|?p*pQ}aNYeK$a%4SzLw~=$ zMsGeV=jI(^WS5o-OpI|OO%wG*Cru}dD$Y+3P5rII)XY%?zl&30@9zur!O>Q#ko$t{ zYnp<5V;>4#-cO?@4XV^B)|YP8Q>4$YB{Iu1EFjqEHT-xP3XdE;rW&7d5R08YUhzEEk|@jc{`e#@`OpOd%%oa8%_;fFfrb8hKzd~ zP3EUQ=FieX3v}XGjs&`XC+c9X>=8-}5?Q*- z^)e|oe1O}(-(=Ku^JrH6C>`dz()0a_v@dZnRkk0X|9bDyU*EWldaD5no8*iRbAGwj zD^B?7;)TM4hCVp_5$82iNhP&C&q-dtGxeA%qAxiQto|K7y)#vsxg&#=%V;d6mO97H6DZru%J@da!Z=T5 zks;NYTZV2-eatw8zGQse_9L0KI`HEohk#_!?RP^#k&r(WAGvNh_*OPcW6N#;XXKZVJa+q=?MjwPe9VF5H8~h;J-!& z0+dGhThjgDqgo7VGjwC+0kL;BuVt;=caqULn-Tv|GVU?W;fMKGF{rwn(F@%Ms)6nB zVNL~jT2M5lY6Ql+^?`$5EfBp-FzU62xu()A>pF!UwkUv|`JPaDwHEvm+9ADd4g0b^ znw4xj!Wthx$WHD$&MxasV9D*vx(%L~w492jRkt9kQ_h-eTCj?h5Pg^u6SbcO@XRp>p{(fB$)a(3Z7=mgKx(OeCg-{ zwpWw8N!;@IgHJoeAf8upf>9Q!R=p4G~z1&~X?JB(DIH1jYT zcfA)Vvj}V!EdmLt>EJZoj<5MQooVW6Wb%$O?3smb>=ut^coxdHCnQOWc0+7zP(5WD{ro<%7L zw@G~f3$CZz^IaeuG1!eB;@9MSZzV`hOoI0-@*)0o8AP4X2KUu<$iFp>RrwRk)|@!N z>P*~C)_wj?40oL4*QQv**WhKK&he2JF$u&X+6gvqJBHj1+?f4EfpF+*1ol)u0O2Q3 zfgG3L5Z|~3ihSK5W+(xkuV_c@(i7m4vKRf_a2V27DX^NiI>GI>GJ7M(khPny&xT%U zfD0dTp)uq&^ju#{!k0@>zkU|!&$ed6Dg>~xs{=j`#xr-Xx1p%GU=n2U77lW~bE(sk ztSy%#eB_z{(V5Xe15!YH##y-EA;XH_(n8k1nwXA+_rmX-bJDh`k&L3zUO=w;(;#Kg6nei; zMFMGO^y7IDDHzTYHrqZ#Cr@W^948mLuU?#%bUz?Rdhe-I?9SHMj039K}UKDd9Zvs%2xdk1+CbFW>mJ~B`Z!dXTDtJ+)BN4LSr7?5xyF| zc-)NFUFgPoo0^dP$L7zGKR6qHV!c{Id?g6fpXIO&)dIg^%0nm2Eul^ahJ*>$?Oxb!%1>o%}oCWXAOpM$H1`9Ktp z5v1aQW|agmvvWGQ3|}&=Xg>+fs&A0Tt|iQy%W=%z&||_&4e#j8CuQ_#Rx^ER`2&$B z%_z`q11jS7cz^uA5xVySxwL-?-Ty&@zM3jQcj#F&wL8u-VwZf8s;m@?bEN+UGkkcIudhEFc?K!MB8O;f z^F5u}?NW(WzZ>8^4MW0^--gI&P>%UI;*2zQ1t5ik_t4=XW2hNVxV-@P+}M)_?@qGJ zR&GxeFnKDjOFc$9zUTA9_C=G;ybSvG+i$W>5JBt-AMx{~8HJ7@VZDADDvsGoPN|(F z8%}RW3t2@H;J28J7#QIqF*TB`aTV|V!8skv)0w&D&xl2m88HqqC9%re$y2xAxHkL^ zdel0d6d7-Zy3sMPLAsYzgdCzvP4AH$?P*X&yV3FB$)MU?1|`aUuq5*Y7-#kfi&7Ox zNlq|Y7&DE`4AlkAm3)w@K8^GhLuhhiKIxcagFCsk{?Ph*Vt-{jaXA}7M(`_QlW?94 zp1VT(9OLMU^{Z)Yk_BDWXhE0Vb|l-)A^!T0fNkBjm)C^ z+nJp+5i>I9E7lEjqX8dF$*+qXqj{`5UMhc>@byN7rBMke!f^u9OiBmEb$(7rBRTDV=A zI(+gV@3v@>qk>boKXVb;?P|aXE{~_~0tvb%FAJO2Y(a+3jl_3{koj#@N?%0HqQb-} zbnTyGjLxY}Vjb&Abzi%p-YZE=`sE`Gt<^yiGb3=+1mG>HFs3{F^w8h9)xyrPDYPs# zmbSdCr8l+|5Oq5#^82SD>GA(S?|Btb{&`EW$;?wL@1ZlL`&Z(AfvF!WMmDOvoG0!cg~uDCfQu!#T%OAoS-_Qao`g1 zl^sK+UTTtQWwIpc?gIR3aQMrOQm|aA!&)SMTUR@!P>(5iy+B9nIJ(<#%-Lrc# zZTSgTE(jmz9!7@6Iebs=NwnpiKB305Q64%@I{oqyNEISH!(Qn3BMLF2F(}C)kF4Vt zQsYIJXou2Ts{Tom-Uo&b6l>8)#sF_mnS-`xfFqaUk!Z~&3jUp`^s93{Lk=>z_i`IVm-CX0W*eJ;>CBg8Hj zZef!}+>9_gm0WdcAisCdpo>EnQ28fHG}JJf?!Kpv^#`)>0b?a{w9x?1j+`?4=s>w3 z{B|&P75yOs6)Rf3;6A^jya$K)IiPc|QgGSS_oQ^9DOuxPL>B*+fF+gJ$(5jW%vC38 zqF(fjkx14QDIHN24PF07RVyUv@5(gNlCz5(G_4@_#`_ZEXH!VUjySw7w4EGjeoppW zR->5Xg-*{F9`@HJ2klJA~_8JF(C^{p4_U&}P& zhJe1l@J6V1tx~xE@_&2_l{B(4Hjdz{N!WoWN5y9aAa&1Mf-@mm$VepzTn`69|F>4* z`1Jc^-j*5kslse7UwVw$o&Ag{$f`s`K0U$}AxWfr|2+D*)tfBLi^jc!%A|F20P?zW zihpRsJyO5r9*I#tOrC6XMbE0Ln&7KsGJ|DZHEH>$xgAx{V%?We?U=6h!E zc^M}0uL+Kc$q@=qKSg^@dQr~imw2+<7`St164EJJh9x3nnd5^-Na}bjITv$;-zTvi zr&Dgi`&5lzs3C^xGXq3Ck1ATs3h3Be9qKPGf*wg5y6~J5vybYbzy^QSsI*XKfnUID%4BFb{NGtc*5#7|kjPCV$ zz#6VYm-yC<*V)s|ptgj_f73*fjqVfr6bgk4o*g3PjU_nU`VRqmoirCTQV71!JTG=Z zi`vA5Y4`Th6yY7DNC)|g0%SziCx>XQ(|#Jd_?z&d%NrsWaf4I_m6P-vn;@Y|g)Q|~ zqRtA<_@Vhc#z`lKbk}^wdU*=M8DGbf!ryc8>fPydj(3}IEXT-7nkR+=orZ+}462wT z5mwB=GA*QfI}0ir;^+m%Afj)h4o%|G#CZB-rabsLG2yzTyOWHG;dNQsvPp%$U#Uco z=O&PbVN0?%^*WJT)`Omm9)%{y8H`%69%|W4$f)>fs(MBNEzS5sj@hR&2ac=I*5+t> zcCjnv`{>e#*01=2gC*p&xe-YkzmdAGNGCCW%SobaKbg{0T76@`w#}@8d)P<29`}!| zCmxU8QEGB6U${RT`(62jXQW`Nx-X8p-ye^T{&|LG#V?1;A8Y8G>-t1WjdQe{2h#Q? z5x?=1IYf2dMuiJf$$H~avS*0~-86d|qq9g8sRbyYT1{(oyZZ#3+BBZ*uUbL+j^|Ws zL~bFbJrY9CRWjH!QVdSHpArU*_~41#CX+?euA#mciHP@Gj`sWBAP?MkL2u7`8ojxW z*~{(4AMZ0G5nDX4-&aqxxbhcr_{=$}+9hb6Z82$j`3c4Unk(#X-pt?od^QQ39||x3 zB++f$O#HXne0u37#>(dx@J;@UBo4ht}$JiDPS5y{Nd8vLGZO*3UdENU~#u@TdAmqv;~VltAWJ2tKPL+N!e+nH@}Vcc#gQY0MpV6cCmK>rr=C}%=}hws zersEzu%Nmccl$nP-YbL>Ti+z&fWF&|*qRX6aebs=b`w_bJVBN&0lZhUkdJ-Rg+kwC z994yxFRNcMejj8QXHN`|-ksFy0r&lq;8B6)n z=?!RjeG^|Wbp?@Bxkht#s))=Ld+E(bJly|jBbHl{fmY9E7^{X{!MPPF{3AzKqpky# z8S5%ZR*uNoZWZ&)Bp+R#8HS!Ei%`qY zbe!k00tfnZ@P9f~VSP4|tiO|suMYWBpBYhfK0l6(_xU3DD%d3)cymD*b?p>?`5Gg9 z%rIPd`v&(pTzxuOKVvZyRz8__G)j`MoR;yha~f$+$sx9z{fVDpFB8ILp&xM`y2Yi% zh<;u{{HIAk^}#N(b5Nc3<2c&-=`kAoqK%ZoZjp#{vExSiX5i~0yyCKe{Qh@^)Ryg{m2I8K z>1hwj@}5oAIA2lb2F@+nYfKl2{6P7+C%O~5il`M0FnOmR;K0{ze6br+Bzku%E`L|T zmoMnV7nKu;_G)GFH19ojcT6O0M@>i_nn~7&j52>-d!YJx@rW-Oif#<03G+_*AT8z& zPH!G0r+qk=Vy~1)>2^2$SJaK@`E}@qnHpNy@)h5t26X_l-PKvjXi|<~eg4rT`Aj5{5+ZYRLYPAHFxvPjsX#~L}bMjYy z0$QSFiQneO6Zu05s8aSQ{vaz~9jGvlx^4J?+kJ(+1Fnute%KTKx{1Zi@;AyvY;P0| z6qBcgoa+OZLsg&uMyPkxQW!oqg5I#^x>XyB ziDRb^?KmDr%V7e2k@be_G)R;7==AD~TMe+!)O3zhxruyi)E3U-oRvlQ_MuV<)ai6g`Jo=In#{J{xNXJV^US3o4z#AlHJhktW;t25CXq_+m`yFe?5F3(ou)2ca`eES zbHv%bo4+jgShYs~06*p4GUUWNfhTX^lc^FW=+drHa2oFk^ROjL-mGCyuPlV<{)NP6 z#Vq)fYrvek;Q=wF_t3v7bJ10=8pv??3cH5nSo4#ciAeo8*#E_7r(SV@Oq2%~*Jl&tW3wCm6{ND--kWKF=qT_W1-P|+=O&$LdY4r6WpY1#8?Hw+( zXw!A_Fh36CTW^K+&o_`$I%cfm^d;<#Nh&O47J~p3*qdfzAn{@xHm#kAe0^W@oz|A4 z=;(gDrqTdx;U=wm-bW^xt|qO4GNLlIVNJqn4B>kVb==J=q^z!#&f^r1V z)!Is24}`_bQuv#65*-L%S>|eJdbX3=zz~yAdC|(sR zZm);m|I6Vc6ZVJxa$pQ2;ZlYV*cL`Y_e>>xAaSCoMWqItH?9EpUl$?S{6DHMyO!!V zETqYgQ#h~84EiJ{m0jg@fi*JAU~6)Z<1*L2>dKElQFG%4dSp9yuIrr*hkMqcWW8Uw zImw^=F8M$YY%8T_?@BYTewvB!uamSGWBMgWR>)5}$ppvsp$97Gk&BHa+Zi*9RXN>H z-|x(#g|1(y^u0(pg`%J!!4UdX$5Q{%<5c1DAM{qNmz;Aoq#tIDkiZ{qpjy;~tlShK zI89%q`l$~;eqTyuCioGFLpt=Hx&;oA+d&?7EXH#$=ioOA%RpX!8f&?;kO;O)z{$YN zs4?yWI;`NyRt3#u&xpT;1^^Vz>!tXb)f zG3Z^hq(7HmC+DrELtENuM0UQ0u0Lmh>|T!=`N^WmntV~* z9dl~)F`PdlBaYvksKrkz`6w%R0If=0j23DY!}Z~0gb#Eu#-{q%q@C-GDSp69D(*9F zff&a%On}x)FA0qsB=s*&((Gaw;KA!#RrpR^(L*UVT zj!l~v1jBvJkFEyog0y2)DvWNe-=y))hE$;>(R`w z_ptMMC05B{0G2kkLW;v2xFS7|-UwBI1GS@gL6<%AZ&Cw^Gi*U|T>gYt*~Y|>So-+b zb@S=5KQ+_%lyoUZ4lH?4N?>?3~3;OTP%Ig%7c$9mkxic`UR%dIfL)bOrwn zn@cRiV#u4OL-dZ5ADZ+1C+gYugA8w16{XL-K%!m#A(VR!J&fr^7v;Z^?Z0lJy&KHn zT}&pr^XNW0^d=oH1p9#0s^h%hUSH9C^?da1Z5lLL=b~3nT4>wdY+(LhTX`$bx$H2)*fXhp#^Zic0=4Hof zLGS9FV7aCM+MPy`(NAXxc&*N`A(c2|=sJ@8xfj1TJH^b*+sOAaR$%TcSfIy$tC@|r zdXe|qQ=q4-Ld>3<&~u|wqQ^!x)V1&^Y&P12v{w$3q!+88QIre+TC-tL--C^bIDw=q zhEd+DyL7Uwq-bVG2z@BBM}r>wndX-t$If6?ZcSjPZsRW8*1!{ef{ue z#|adsk%F%I*puj$n@DJKIC1_n%-dSwfr1t*fnH-1dhC3a9vyp!zA3m&_i5y!U1#%< zy`TeG}O&|7vHqdF!I@+a{tx16!zE`Viba&q2fHCyd0OeegVLDoh#| z4LRo-RKN8SI5QP+Fz5qPc~^}7)Z4RDyvDO1vc|!VA34H}E&GsW);-+)`V*~hN@XO6 z97G=L&Vq^gS?EklhsU-8)WuE$Y+oaMHY0_-cS{H%&kfms|NP*|s41-3sDZ?ePlYnM zY@}-*M?!2l#<4;)*0}zUi5Q#(i*8JS6@FppZ@e>G$m;^tWuvh9UOFl)jKXQJF`1Pa z1K7-# z>npXwE5*0L=|V2LyXG3ldkR2*mM$c6y}HOzI-T_q9AgC!j<6Cr4Wxq~1N~PY)2iM6 z)c-mI(-YT%nABHD0FL)1b{UeU9Hm-2w{mCApZK{}7mA)bhKw}VktT^W^k%geowsO^ z$u84@vqqtGa$*dXp50En4$6xn^OQwD<&t^1;abf7ZcQXJK^zzz3m~V>;CavsCjU`2 zTy^6(oAXo9-(6z-==3yf`|B~T^RdFJ2VBUXBg^ThXXjAZlW=0*oP{1FX2aE|DdZnp zMe@aN^wV+O|}`J7=a$jE>^=7OVG|m!Hp z1vL22ly^D&IRE{@Ym8rGJpa3V9EojGqb^gp`(A1nvK+qd_2=antJ2&SwoY#o;m4-8)$nRX&{tZ%L#^7yW6Fp9y{S%Zh%vT1hwR z3^E#K^SHcE3X$}mN_i7%`EP#@k`2cXgQCY+_Mpam#uRjkoJSn7Vhfo$(>CFp6{_&o z*^YL4%F{8m0i<;4YAi~=h!kSx6Ddb|s(K}gwq;t=_{4EEL%aZL%*r9J^%}YMIZn`a zrH}Eczr=r4ewG|HiE483%p3=hrE=f@n?Ka`^%3~^sdy{Hzm+*k6hZLlcQv%PQu|WPPD}u7IZ|HVJ(V+R0q;-bl;(T0XC%%K4T zj;&(` zFXrDzwrT6oys!ZLrpt(|+oz7zKDQIgSGQ1RKsw((el59B!RNh;QzGwgDB&#|S7TNy z+NLQYQ=p}Dl3%*7li!)W2qhJ3ZfXT-dV=T)3Wy-_9mycIODQ{ItlSN>%3am%r66IcZFpj26mS zAHwX((&Asf*2>RqG3W2mDL`g}+hJb!7-rOKB0oQ%l;QV(;srLT;$AaCYWD0T%Da}J zlittpi!*u5#m<$;c&idRuX2V-zvfGR9d5vfCC`xVH7kh5_>IVEFp6wl#^?Q7Crys` z?IoID3~}%Vj+6ByiSgZ5&U~HU!1tMQ9ZM4t#>caAZCxsAL=~VxP|yEy$20^{-5XW8S9V(#wmuHA^gb)%EbgzB-})FKh5m&T%05 zZ8{vNJcrV69VZSYk<7QX1J%!eMUC76s#y*Wd8N0M7zI^v|I>Rp!o(@N$ z(j-L~S5ptMC)8k1T@hTAzeWyZ7BGu{BtQ)v7A}0Y6^%<90}853;n45H@TRE^mahwk zFS7;so#h3x#D}BE_v(`NWzLA>vBT)!_Z)|+6(V~zVQlRV=qZ{c-0!glnas3?ZXX72 zUh^a~xsIN0z;(R-q(1tr?aHJ|Oec9;TOfSLc$UZIl6ucHfmZ2p^cU{}mx*^!MOHle zr8kJ8R&?Oju~o=NBaT*Sec)^I8<}#SU+A*RM|5C}BUB}I;``GP`rLMph!^QW{)(F@ zj>y0k`)RP`j|aK2Z#*;9ql0|6a=t5_G4!c zzbyEmd;=P9)PTyJeK_zy625He)_)=pRN0 zr=Ny~n#(YD#u&IDw1d|cH*osHzbMSm3BEo14mU^I;8^hm2vtZx`2lN4%J6ioV=$N9 zr|!l^E{b3U4ZcKKJyPh9zk@EYKLCOW6W$t{CA;o$5t^m(B^GE;3p za3P_$?I6_!3M8~JnjtmAsA}sRbSPZ|-H#i?j*jdG?cMQkZdNmqy|?*5gV)8$t!&W|MnS46O7f_**SA*zq ze>DC%`y%h#twCWQTSgW-e`aiLL#f^ro$4igqw7^Y*Ya7A8_oN!fZI7W&(qji(zx- z3nEZoKxZC23K>%Z$cd(8*jg6MaU}ud3wMK&K@7Z{qJXlaY{7Qm9*X_C9$h@@2cQ4_ zhZ1CZI45c=*}qB!&!_X5kPjZr`zARQG~*R>u`e7gZN13#$eb8{ngw%gLaOkRvJwuR zQO>ltr80|1kMQ8bb=WajfmxOhOo4t3QkCmt{11;Ip&n|?Im^pGPbZGASV1oKwDFn0hUmbaDoibxq4+!cuxOS9oLl6MdVfA6ZUrWE>INCwDGbGU zuNF~ks;m|qI8UBNOVZi5|B~GJ{UmwQL-IXSg77xo<42yCNA+(FgwJN@3GE*h@DG(_ zGEY=>(ShR@r03;QBCfcNym?uRv=4qq%bIz3^UO3Rwa=S*H+L*)bX|agf6n47sEjAZ zpVWxiquuz!4QV`K*%Gc9oXop!<4$;sJjs?Psr>e|X*efw2%ij55TeR4{Cz1)P=7OF z>;}hS`Q@t^vzsz_q_%-)+RXze!%l*_pm@RTLJAE8>g=2a)-MJG`e&1tQa1k@^ZxToPGhZ3WLmYz`WHX=Sv@Iu@yIt~{A4b< zJD@~*t;Z1Q@v`Wv*;wY}-b=)ImmHJ3jq75qct?8Qw=i;&W03r^2~fzn}(H&C5 zR2lDQ{FG-B{u!={``-xj&RtdzWnGFClv$+5+rpfk>VS0WW(X(jQ50Uv9FI42>EmB_ z7m=cn2J(F-cUJhfhUm8G(q5GfBriA>?->(_jicOg-GhBdeMK){?azIIdO!iQT{i}q zZMuc7{D~!UoTni1?G{FT-g~C=Sv>z@!&7Ez5sxW!&0;p(FGAC6&adFVeq@isQ86+ZX9Lq~Ldnn%2VC|%2-|&{ zhrZ;F^5*O$h<73vZ@E5!u6DUiE~kuFjT@uU?L)3moc$Ns2F=5o-9ogQ zJAa-HnNG~&p9tkT!|3Rp9^CPB8U7QkOUC!h5VJu6*Of0MVFB~Vk|(S9gDaP^QbFH= zH#m-+vU>%@IH-ZK`%@I!wVr%+En%XUf5rdF&Ll6y-5H74Vy5+jI-A}yp4IH?hf)nS zvd~Bojg!X0FZWZeF;xAL1qAd+AzkMh zRM6&++A_2VPc0j5I+Dnm?z3eRp3Y_iUuL1EtOjKDMFu zMbed%K|6Un`jwa`RGKx&SgU8Ck&IE)PtD+3^l30VJAhPY-DDKrmof`Rw+l1-bOkwT z0(^#Nh3E051Ot0cf#bDJXrLt(O-ajxQ8$YANPcBj47@}tt`~UvewiHeEELj|TbV-; zfCAkY!fOWL%-UpdHkM-xC4Ylk0Ag(=LO?N9k+RpU@V6yC)O-e?>UMo&j8e>?@)irj z1|sA-I)f-a+k>u!@sYL{4-S@PqWlIcGCRBqYdRD$fs>yi*S1LZsF?-pBb*PMjp zm! z^$~h>z=}24Xv$8I-_5=-HK*^F>yVAglhBuqKApqUf`pE@WHMzm!18#8Xb2z$YyNg*^F@--`$RGoU60&H%9W1`yhc@0Gi{_+8uy^WZ zSf|=)Y}oJ@(Eo87EIQLbblVcvOkWGPKMtXRD2f)}o{#D|=Ddl!2nLdWfwWyQOiOw| z8hflj|3N=0St-Vx@cCAK`pz92Ip0_`P5wFaF{9f$j`Kmr-tTAfJ2KG?!*`5tc>pY>*-8;ZqlBiqTD^lDU0HG7d!Uz3>1(eh>bMhJ28pKpE-;ix7ndj%9jLPUe`$N{JUtb zEcc8rGz2u>7-HY4z)Gu?#LjXM&HA;3>}ccq69Rt#%jIyRrUKTPE{8AglOZE}0<4}s z4a`dJ^AnnvpvpK0lA->Z@3vS*_DFYjs@P<4vH!i@7E4HqiOI@}{l8w?Vmz^pL7TVw zhiozP*%-XhjIp+|vazwWu(IUy`Q{58ti;y*-@pE!AN0TLc-qub{lBmO_x}I84zvHS z>tM`h+bnRH@qcq2o#mkv9RH*HW?*ViHA2h$Yv>j0(=>WpElobtOlNPA64mT^P6v0C zP!r)3THpGLo?ECwZ`g7ULxZpMy~ZN?C3q^0Q#?W=A`a0{Q~l|_xqnIOja=gTbqIZY zs||*m93k<~W%O#lD%yUGb1M$NLLIRS(fDyPsMXB`4RGJD)>}m0B}q}|JbBv6TSN!_ zhDd?JMS+9QVz48jvzHaURq3FyXYWlh`-b}Mb6loA? zK&bB7>natQ5}}AFA{xzv(5R3?DQPYmG>V3^*G-6&2GM{L`U<5`rX=M(?>|sK+ot8U~t`kuY|C9aO|WgX8Y1eEuXOKF_=h)~fWv z>3j{o?-d2Ugv&z+ii?8Vsq^5QiUPcQ^@emo33(YINB8=z$LzP!^hDJqx+P$UUb7vg zzH4Rh%T@{aCFKrgJxicpYYS`|N&=q`vGA-+3MQ}TL6db5Q>(WEWH+yc+f^rkiiAKw zw-7A4;s}{7@-SUS71SezL3x=6{cb;jR@}VJ^Q}5fj%y~7?rkAtLzo_Y4D$iDgfQQH zZxQ}j7WvsH!P*~{Vw=8)Fr`kOZPdZ#YKxD&mk zoeiTaRkaz}!E&;xVj5~23g}v+F?#TOCH~nJj`JJ29AUQ$zS*8buX^b~!tgViJ>Lu$ z{{?$MV1U5yg+oLHT$nF}WLyi|-x6$Y&F< zVCi%en)8eOlOBmaKf|!bMhUI@{qX7&9(5>Rh?a*plXcpP5LX8rJ29GM6-AR&Ek_~& z@<6(l;53tX4ARM_v6E%-dDu~UDKH&52ORE?OUAnWiUPj~S;12wDM4UVHQK4k3yhVe z1#UaF1dB4J3EWL`ITlF^?vD(?`8Ph$twZz3*|cD|`;+5?#FRkdgJ5VK7KN*-YiU=; z37oug4+WJ>>~7@a)RR} zEpW#`9k25Dg0Rvn(l<5%tTLh*6GJil@h2QNgx|pfPU)Bx)P_p}Lh;`13wYDO7_AK6 z(IM9-%)UB*T7N|scFSduotu8p#AOeu^Q}Gf^V~8T^244ksg$Bw#c^a-?qa4>vWD{8 z9cYXwQrhH=7Zycht(7~T<$pjuqXU>XXD1dt+le1C0~a z+HMi=03`Fz;ERvrTT zJxTD{D9%4|i3CU{)9%#)xIH2S-83h0xh*NYzQY-%?ObqPl_VBg2huYNDmeCg4reXC z%zio5K$c!VNfMbkFm{oF(6wA2^uz*KZ$C_i53GZ|XY*mPlqCGVsSOsM45?dDLn`-j zpR3zabp3swy_mm;>t+wo9XBfJ=21Rv?@-I9mtmY)*iUk~RGMmk7eK2!o9*JfwvX zE|CL@bH9@tCSzpWHpTLwpE-1H-cR~Gw3^QPz;!v+S>gN1rgZql6Ph~Z9~FNQND3T! z$Wc!oII2p)J9~L(RXPiM%HELKqc4c|i}U1kRweB6w}hw*XPNaH&v=j4<+8Usjc8@H zKD+O5F%u=w#&N1?ly%;K#=T1D=6DAG(jz$fI0iQhanG+*BvsFr#NVlhutst@4#xOU z`}I*|$BH2OpT;POEY2#g>d_**XJwLkO4>wAHID4!*d!&Xam@QUkLalxTiJ*E>Zs$4 zg*bNZ7P=Y^W9hUOOg?lPPu-Np&#NxduoZ12T%?cqF70JQCMA&IJ0j3CXM|YFJ!SJZ zZs)CA98V_R9VDkBltJinEa|))3bLMQF!kq2D0mkG8E^&M9Q46*O%#(6b&vW8ZZjS~ zMX0bvG?^yoB%0f=nU0g&#BPeSBp1KWfO;khqHdf7i`co)yS{}CUVBe^Q;L|iG15d{ zw}Y-f!nm07=W!z9bBoDuAn){Q;Lq-DZ zvlc?2ttgi*${{AD+vuOq_GF9Q9nxVXL0k4Nqz&>Kg!aE8(XXFVAEEu!zIZZS87z#M ziqGkXJXPeyPezL;OYk)WBgPxz>~lM4c1JyZn*WfR-hM+fMPAV>&Z4N^B8|;q1#Hu7 z9bT-}EUKk(on)0NLB*7DAR-lP_TZm4`D7#p*D}Lt_|-_t$SYFY*_-Lwc{}N3BMVxn zv7SDvxW~E;PDMEUmwpL$ppDhr;Am3_w9Zc~JF2P*r8^2iGtnHfufsd;JwC)Xz_9p zcV3sHego&J^QUJ_NrMe~Oz_6ugoWi#w;Q5@<|L|aD~6e^)6l)8o4RgL!2@)ZzMU?K zNlW+eTKp&B?SC1lym2qid2NULD?RXskuxcZh=bj#OW@Si{je}A4!ka=!YhAk@EV%| zZdumkgH$4Q9DfphE>03uJ?O_Rx|{IF2RXqS!WX=BkP}QbP!`zj5#YxKvVvxwl;E$^ zC0w}uAHHZ##Wda;Y&FlpbB;$)Enz9@HwB=i`et0)Zi07oopJfMuk4RO9cuO@i#oo^ zqf6Rju+Yj3`ak$BueuprPHhLl;>9p^ zsXH^M%fP3EC^9!HkRHCfj?UX7LB#cDv$A5_$d0zXuuj7e%v?>Od-y2i@nA(u4%YO;2$wxEtw{Lewtes`FvsS59|`hZj3Rj`YUAk*JuvDj40&e+gHss}X7mq_fU zVif_%K9;~^+a_QqL}K2WLEMwqh4Jgp;zhM|3mmRvhJV&Yw4A9RBMpEcgGMNWU8FC07;Y!0@m?n4LQdxd%4D z#LbJ~fq6IVT=E&_jQa_bqS|0$>16(O$G6bF@+PdZiw7&|yKp7M4|Y2n!Jn5&56zw%L4`~RQL~u^8|)u5Ms7oNzI!^ibGd=MhW+?_QZ0^N zk458WrTC=oAr9T>$8)E$F|IEPHD*WRs{4zvxI`U`N}O2H)sL9*@tep&juR*qGn0zt zH&dsg<@on7$D`{H2cX#~1cC2Z4(7H3*LsgOlZ6@KSb= zY*bRB+FW+&>xup4`~p5EXou1C_67J((-PfeN0@^xkLY9bMB=$5lMy_#BnsCSfc}4_ z@TPhXxU;bwKQV@TjyJ)fc|X8@eIYRA380|G2d~6FV&ks^O$8dLziU2jw_L^Kc65?- z&ABL;$)KB20(H7y#qRo>Nn=(!;FF|xJfF>5m=}|S$cgMycGvvPY}(DQbYRP4Cb3O} zZr-Z}(O*-@{q3AvvBncVRnG?gL~$7RbuQ@$O{Z!X&XTYz157KUM3+2#L%;GpNO$8O zk}ng^mRDp@rR@ud^NWQ{DVHe`yCj7l3p&})cJXDKtTMt|?|XnBt!( zaa73ep$a#4p~=`hY!0GG?w`X0X3=QFF^kUaRzNR%0S&nDfhMx9`0t)5J?dXXE-OtY zdH1tvAwL|I*BwG}-8=Z%zZ^#3L}JSV-W1QbaKDpbt-%sR?HM_=wp`MfjmWYZ)@XGe?%A+#@(d4KGt~o>?OSOBAjl2ECErT$LX-cYjS&R z08}4N;V;;r!uLFO2ZnfYP|M6@#&@3pt(hsrO4f_47+MSdJrltvB9~-M$z~I!zVV76 zmiZ`^Ro=Enf&JxUh&~&|=qIzejOfY%*0l5hSo|}CmlbE=o_-H#Hgv$#oO^J(q8Y@7 ztHAlkOZe9!#D8X02Rg@C2=;4-cTb|=zaPTzY~BjE6e9y#wU@~3f{Ae0MFyKso+0%m z7xCn*HK^^@Obw(`X^j32T9{R5FuxC*3TLhLRIUsnT1+5w$ z@cCmfaa9^1tx8wOyn|7MCTxZW&rd^!e<93oZ6<@lhw0Gn-FRN79{vBNVCLOhsB*Ij zGZt~0o^y4$D)ln1-quW)PuPj&9ktZUUV=SWG?k_$GVJ{GOQ6QOjeINH1wJ$LV7z7# zF)^xOub*pVrrjy#g#|{@t4Z^T%JZ}2$7^@`DJg^vxzs~yAekIEG(wro)#&V3Mn0`S zL{F@qidp0l)%GG}Li$P=ry4}Ikd#dfZRB02k{RgpihzE z&Zm|z;uSzX$j%3`yGn4pK9j7@7lWIZe~_MdFF4^hMob1aVg70nyekuqXR4D>c|tkr z?D4_N`hdGtqnOK+gJ{FTTGQux*U8pJ$x!?FIyfyDfm=cyupxgu|9E#NXwnAGuN4k= zMk*lrKT$~I@?q^<-{U$j4sYnJ#nQTTJodT(=V)f)#^Olq(2rrmJ}koggTg4Qc#+f% z%3$(u4^#^dq`6+YAf&P#))+d1{j>8Rv)2#oEULhJ3D?O8*XK`I`2n1^Y4Z2q%!Q3F zxVv;k9*mhRf#*VI5Z`M8tDEv6aQj5Aeh>|rr?R0l^&yy;8~}5Nb+AL>8;J_>W-B&y z)7hqmcxh!A&W07pQ=fu+d?!M?^(-dym?OUBX0Lb3wK2uO4j0#kVU*}6nq(!7jaBoh zu;F=Le@Y?ypIkWVc8gN1Xrdm0b5O`F27hY%<9ClRdct=cZTnR~4@MhPz6zgy4LC>r z?bY#d%Xjv|#(2{9@CKP|rw31Z&y&#{Kd8@(0$zI%6gX6o`tfIA`8I3F^5R^H<)h4n ze@DUlL@;dOoZp8FR)Dq6Za6$`5+orDjpC+o{LMykyepn$g{8v8(=3>;ON5&y2=YFClueYP9vVclCO{Ww8ej}GAZ)6s7JtIb* z224yLP|2Ocy~qvtW^tta=B?#qR}iJQ9YnC@$X8ku)y1}iWD~s~vmt8Ck9_qNAtQ(C z$*jB8)Jt*~TI}0`viu|1^gI|tgdbtwg4_7_>Ph@;2yoMv{&A(5(HM7~Qqh?TZtci>~lVA^~siRblJdPa@!;_O%qssb38uKLxm)8BE z4`mh5CF(x!;29rR~xIRYZ?M$p@@0sCB?$#-5Pk@i?sroCx9$1i+zx0QnhFWS{0yNZ-~$ZaT=8`#x?b>lP=`PXnRMd4VC+ z+ASx_M`qIP2Yku*_AE9}<1W*Qo5_SVBShBmH<@%cgeK=cq(jGRX?Fh}GC%SL`)g}I z^=V&7bKB=&**Z^l>kTt};qi;6Hc6x1!vGwyFu{+{Wbos;S=jEXgMPZo9ujVKcl z*bnz(hTM38lTj>6e;LBBFEVlaSSU(QX+@cHK^U>@J~o^WWuDio;IW`k?3O)ZJcPZ=iXIKDq}b(%flx>87C7 z7&0az@UZ`b>UqsrSGxfZC0d}jWerx5_esk>g4dJk;?hr8HnQ-G}0z4{lt!;C%Q0Gwbln^pqMJcmpXKrveKN*P5mod6QDO$q z50>9y@-gpeC;pC!#WfSIqtdht60szK3Hq5zx7Tl_Q{z zrGJU>z9d>P<2)1P8Nu{FoC`@`E)vD*U#VTPA0sEM!7NCf3Epd!z=-R8=(zqz%ASvq zQTqwy50`p^-}EFp<(WE8DEmld?A2(?@2y}k(nX$M?O|U2<9JRHuJHDhGW0IUTB^sng48c@`GL1pY+(5&!=Mc;E_x7|~?xcexW zxOHcU@6-Br)M+q)07-8EUG}1lR8rU~81Y39I(p7)NaDUYuVjda5dXDGj z!pd26XS6Adxc{P3&+D3M(r9H>i3d8lUb1?SlHB{p`uOq7Bf>51N zd}bmk2wyvljh{PE?`k);c)Uh)HBX#o8G##j$qL4wq`11m37y-u@sghvKAWD2hj`P` z(8`*&=teNzx`9-yT@+4ysv#j2kz~B)5OHy~qwhPUpw?#_Ti;T~I>t|8XU|9>B3k!o z@Ch}V_AVXLxo*ejv>4dZQ%8r3&f|Hb<8<>52EQ!Xj7D5X(o}2)&J4ZEYWQ9t&K3^v zS*-;+jvIh>@KvDyDq+WIIsWt~qI|>oPr&K@PYBE(1fQy2D7K9SpZr`X+o28{g_Ozl zFSW$oB!Cnx4FdgMZf^8Tievv-z_QU%dia?)Y>Ox&2Wq~WCi^PDkWwj|z40K-3%>>@ zeP+RzU{~<2$Ohv#?}+QBII>ABn!1O!5h-pqY;1c7!b>$l+G9Tvtz8Anx4s|(d0VjE zTu7d%Z-HylJdmxH!EUJ{nx#>KtE9H#sQ*-4GWwWyrJ1ntr+rFqWtzg*`&ux+K^aa? zvj7dv9y0#LeX>X-0;=|CgOnd1V#e>m$_tn9_;M{Uk(>a0)32oP@M_jbEr3WLZDala zyF&wJa=UTI57gJghn_L)r^(ma$mrP_s8tX{J}+TNK%y|+Hg6ZR^SmUu8GNN1OfHiA zxL91HgE#`^}%} z^t2!pdr0v0hD|7^ydFQU9HU!%>>0&Cdpa&r9;Z}_bDY@H@-u$_5&xBvFmaPIG^Iy^ zda47|2%iHzM-#X$)CB2z^87K`JkBc)5Up1L)BiICtFw|2>rg~z8$V{0r^hk@I7s8X zZ_v-|B-$sv4YQxsQk4xeapP-C`drirJ%VEJTkd(Pv27lPCpoqr z!*o;eQH;1ibHG@Tab8aFHE)t2I8aLv6C@+}kkyB8*5AYak7sbgk-KQCUxa-TvDo=F z0`rgjX0oQa;MSTBI?H4_U3guL^jU91W`Zj4j$49@Ll{|MA`S2Ielo{9rQp}aN$@po zF88h;CA2ph^h2z{!r^GS=J+d+x>A9>(3M1=HOJ6pX#)P7Cxwf}T+jrA@L3iQA3Z$4KZ-MF&!T%~ zI78`!m9To%b%<$`Li%_lIIt_JPnqLTY)XMM8S7qjQ#6?9vk{R5vm&J_UBwn3t4%^alnWPv-ka2T6NuQx7Y)y)WTXh*AcQ6!0{es}% z^9Z6jv6*GA3z5j98>w)L6#X|t`Z9jvTkIRPH^n- z=6a&`Rt%QB0SH?9nCCw)4np$Y0^6AawC^I!Jl_R%6P`j`gBE}7t8PdbpAPWp4CJr! zg(a$sh{W9y^8U35B;@q7I4g}hc*>xb$_M)4u`!Y^SN@9zOVPeXEP$O zS0@}DPXEAO?>lJO*^ZvqPvWjG`Di9%i{sob&{?79*$b}MO#kcK&OY9yK|?eH(eT`+4cVD5Jg#|Rv`W!uf*pt~_`IWlF zd}b=wIZ*|_QnFVzy8Mi6J`Feg%zkb>LpJ`YWlqd#BAY+jfsyul7#pNu_BIMKPxpY> zkvnib-~otyI0aJa8Bn+H0o*xL28aCr5gmzS zWPZ|J%GUTmXD9lEPR0On5lmqZ;L1n)up@sBO;o#2OI1dR+3{>x?Z~m-<+$I=wm?{A zrT|}@H$Y9xage+30gqF)!1v}Yl6pcI42S+>D)Ul#fyv=isVV}G4Iaidjse7EDR`0K00o z!N3eVIQ>Eic&s#ib0?6xwVx&Y9g$RRb0nUB@*mQ?B#g-tLXoi`y0pca%~>LfY4ak< zF`X7TvcDHLdH;q}p`!e|mg0Q7oe9wN;uDNGFM`4dj)QSA3R)H1!MBQImV_@Md*r4- z&W;mYa%Kblu09j3Xc(?MAw^EU8el80_R)>Mv#I{)FH~vaPbz)p08O(!Nvqed#C68= z@kYcv++KT^juyLNS6dNnk~7D{V$CR&AAp9pb8yF%eB9Srf-74+@xi!o9OQE1pX?{$ z@1&S=p9!fXO8O&vp(CDVX-3c~cdBXY@I#vNER)O}_lu~0)!=%lp~PWx5gSbOAz)uA z+)!aOIR37%sTn~EBv&gcHY0L}5M(TLIo*Wp|BXzTM zaP)8yIsbk($r=u)D$aT&@)ZxfP1BgT!&h0Q+ECbcKoV@Dn7x4Z`5o6_Mw&l0G*74Ltqd>Uld7#xBQBrbmJeM3I#YN#rde#l8oNO%zC#VydB37T)7N66;$-gsyk-7q zUnP!hPGoD+t@3lXrJ(ECaQXGC+>CPkAfrD*@W$Xfs$H-joA;%n_thfwR=$GGJGbB+ zEmIuItD<@fW?+C$5f#Dt`0wF7^xOQBJczl^n0#Vs-lWB>R(B9h`?nZ<_K4uyq0@Nt zTMe!G=!mLTv(Q;$4_!T@ni%{srD7STBz5;Wv*1Ja&~mDgEDn~2yIfv)hsFri@O?&G zuNR?wei+tzy+mcXDT3e{3Bgnk4Z)8=dx38DV!;S+rC>>(kRaGXQE=tbE7aL`1&{2> zNBJEISgW=IQwvKN``3ziMm)55N8D@X<{~e=dV^r2=y*Ifubyg8FroL(z9$DmB;mj9 zN8rS73mnZZp!dJez?E%L^pa90u9d9E#&6-YC4N4;ymuZobL=cPUEWS?530h|U;9Dt zm=ST5w1%e+oU5ik6a>))u>RHtvQ~Pye7dp^Q|zw6W{G9cV6LaT(5{6{o3U*AiZ?PtEr2%l} zMH^GziKJfLf_#)&NG&a8h?>VL!qhDRnebkc!jVJy)65}nsw1qJrcA_MR??DNT}<$2 zW#VtT3s$^bPhMW`;j(dx_K`_sx~(5H|EI|D_RPR7Q=aJcOu~{Cijdwno~3Iw zh>G_mN)^(n=ZOn+(sxN>-Yr4C>{X`w@hp39YXX_p8V)aS*MOb+O=#bA1@8X>(7JF7 zWYZ?Y>MI(2brE&Gbn0S$oShe6G<6H#c4Z)c{%j9^Ajc)hULeG;R#4%;T>2W0*B*jJ z!usfNwvEfRO%*5{ULYtH|BQ|bBAma(K%jF+RPe2;6E_G7@ye}y`b(FPx}5E>rQZoO zSt)4U835ugwV(}Zq&#n7RxT9<;dgK0 zoSej}I41^{LHuyY5)Aiq|@6*C^8Llt+VE-|wd;Wq*)ZK!RzHVq; z{|Xwd9>VtBS77qaSg6W71M6p;hLspbLS9ZK>dO?U+e>l$c>Mzv3Jk=P7r1-x-wP-+ zSc{4#-*KsSBrZMp2H)*V#s05z zPMOY^OEBQ~_XqJMe7Ezvf(IbrVHjXsF41_DK`S0*q05dZxcBm8K@a8Fp$DY|GW|0I zMNrP4GJO8`PZ1q>v`P2tGAm2`_MRakqzA=6N;uu}U%HhAAUUWD8NHl|tz

0=DR0KMlT8pT?S}r++c4PEh?kRYW9?8AiX-YFQqhbvf_2!- z{=Jk_gGmP)5EEU8&Y)XZai|%gdbe4B(S{M3E^JV2LH4OO^l;jdsB#Y;Av`bZ1r)H_N z!=bVj#_6}veWMX&N9$q1>Z|mOM$Ea>fMYXivA=vP%3c$WzS#=p!!nFfQ^3M4I~X4+ zhm<{XcwoOBW<49x7qb;IWt%W_-d6NxZ${xWN!*a!jCCe5$QwgQ_%4G}pLc-X1I4Pc zFgu_Ko~j&XzuOLvyK=aaF9+F|s#v;f57XAwp(3^q?n)ZSJ+6+3Cn{LMQ9@0L685fE zLerL=kpBf7o3<4>JGP)Pe5)VXT8BtcKU2^r-|Ix;8_tZZjECfb>$b!3i8~7#CIPM{Vo$~Cb@ixp3-iGrI1hsGls17MW zEigHCT!i+v9f0;ZtqY41uEg@ zSOV_L;{FO|1^LN9uyQkU)HlF?|0eu?EQO~Z*FivU6SA=d`Aa2Xc~%CSytiWaSOv6r zDk4Bak$t9>a63+w`5C)$;IcA~I;$a-y9ct1)iF*@9Sf)I#hH5w;BHWcuHX*D(oVcL z-ifjgsz{x*2gkz{p~zK-fBIg=%W5HNiUtO!86frFKCHd0i>fPokeaLkKZ`xkG}(!) znq8+EejgD&_7k2N3?ldaM@;2>#+(<=F_JWZ zi)-I9e)khJHlxAe*F-hh`z<6_CYx54B% zZ~)`g-eHEsC#*Q~9ulKJvHimz^v3={LiH~^Jo*P0js8Mn!zk3Xzrka{7slZI#pTHH zq!{@dCRcugf9)?8ssF>ACBM-gl~f$cyq)PQ8w<- z%JM>HU;qlId0Wlet!PqG22hB(>uCsNCr-b6{iXbf24uR<7Q1&=34%qqQyfGIl4Swid6@e`LVC4xOrWHn_&)X3iH|?Ml!r1(GCb(aC9qr$3VJ%{ZUL70UndgY9zRYuc?F`R& zC!DZ!#4=GQJRN6>Lrta#aIiyLzat{l%yII}RcsyagwJ=bp)<_{GK(4OR&X9NjP=8J zb9|p`hNW}tuye09=0`co)cm-L3rAgXG}#$14m#q7 zj5S_mSmV?8YjCY_!f=BPJ{P$_(8vaT?bom>+8$?m9dK>E6=eHNk?dlQ%gkeabH@q? zB#mqG!wva@Zs$L-sJ&7>6Lp>xUrR#2-I;Bhhst7+SB{cYIM4673_A5gUPl zwNZErE?zOMO)}jdyg3ng(G!I1{9rs$4P*O%F!*|*xXlT{>+|6_m&}843=h*=LKssK zf!(+G*lo_m?r=oD`Py~|BMxbLd z7aMszgo_5@Q(GV&*N4JWF%+^cY>wI-1@E{-aDOD@qkIw;{D^|5eFWM@*gB2FfrfDW z*U3lVF&=~jg7ESlA7>@_cp4jm)$fDRk{-&mvnc#}7>g99$8q}-pph7Z_EH|?7lmWR zai%fOj>YizNH(AH@j*Kf@>zlCJ{yjqB|LN`M#9l17+3FvVLRJzuI&uQ!RNuaLXk*t zion@lk=PX%g_KnxSRx#WV>ePEo)m!+zfcIZM!+nIi}N-Skhcj#hAbC%LWT;R>)ATY!M zCb6z~v&j=fQ#>&CxhI}QyWsC!A55 zLR9>rcq|kvck?iD10O}pLUBGK81aLla2p>0t5iSeZ{=d@Xb={NaMM5QM80At>MFhy7OqA$}qVoVB4?7Z!}8 z8zS-fas)WVe4KA#@85+O{F@nvQ5!xYN5i12#>W8>9vY7D@Z)0?4(PDDayuB`{siD( zLje2^g+R=Ni$u2G)W0E&Q4d2yUJ%5BLoseV7bmv)!*v!Hsh0vFIfZ?G6oQexF&yJU zLs>t~!>g;o(6eRrYD*|CZ3u%ve>k=_@nAK|W4>G*=u`xZ)!6soU=S9M4Z^=nRujJj zL4$pky!q@k#duhDHio_Ld|W;eh4(G7FmH>-ukoqKZcRXOOA?+QjDyjX7<`V1g!B6_ z>@^HR>_`NxE=1zn7A`8%dDwS11Sgp;aCuJ<&T#^8c9iKB4g0b7ix$*1wVAhVfa5a{ zA*MndU!Q5>M4bUty!2oI|RGx z<5+%04`TU7FiKZvGua+Y8moqk=ZB#esf)?#+9>&?ir)EqA^BAa)#LWyTB14@%+kh_ z5l#G8wg*SL_ChsT6H{|lP{dco&&A5vA*l`t-~Cv&QwafE_e1`~9$c2#g}=kHNHSAG z(0zH71*l<~u?C)~t7GyG0|aIshR-fTI9xr1(DLI@iah}N=HnQBe-L|>_3>TT2r|;z zIC4}QZZ~w9AAS%eRoaa0Gk~V`Ui^qZfFYrMShr9Y>K2FaQBDt*Hb!u7JdAlM=dra) zAJ44yu;AVy7<3$h?o%xkolwW7IlI{2sg8+Gn$VTmhv4-GF~9yWDh&+qZlNKLoYumE zJ?hXdJB;P04?-~bIL>|7$Mb|EtoJ&B&OAdDKGsLY#UtpNaTrFXhhfM!KzWKG6oXGe zJw_k*rB34Kp%a+wa1wVeoB(grZu+6KlXh)XAWn(`wf@>k`>ra{_=k#Q;k%1olqk|s zhQRLbm80}K3gn>%(mJ=9di$iQNoONTyxc;{)8r|1k}Qc|QzWmsJLx~qJ!CRgmrCT- z>7VZbT4Se6GgMhV=$|%?i`Ju@WPK_>qeC*?hsfGhhcuQPp~PQUhGc!& zkfPHM(x~}C`Z((l70RBVvtoxSFZ(#n>pMt;uXX8{-vM&lsZ0LhdujSfHG1({nP&do zON&0JP{^vCbX8xLIAgZbCRL(|^LLZ5paxx@qD-eH*<)|2lFLqY@($;!J*nLj4)F7GTsQeq!ql6ss>a@b&&=o-_R!RR~^z> zq)VC_`zXkXWwFe(X?`UK^{pHX>3c&j(;s73`{1RRFEr}>u!$3d%xG2@RsHa1oaog<11F!8h^2X3s#JlVy4`>rJ<_ee*#_80IgHz!YvMA}9DE{JT32P4_|C zLQnK`y5h958_t!vqCe6D=bU`;6rPwq$SQwlTeQ^N8w|-!5y&nJM8MQY)NBqwF)tA7F9ox@#`=_1tUvDw#@;x#|9b@@Z*4H1 zPYr^FUI13H|2xh<5Dgo+Xn4a#vPlr;vYtITIuv!E*z2<%$(z-w-t+)$+ZY6!D%Ri2 zg(0>-6myG$FlrnO(~wZSni+v<+B~>cCg8nP5*CKC`f-fa(c(}n(~rX3U2N?){jmD3 zKZXUpkT$}>*_9lOuMI-U41W|f1Y@0aFpODku{jjLdPg55>t3g9!|U|zkS#q~VMChB z9cg`|9X(xnoeZBbUcShIwwAb1)HD}TPQOkkOsRoZbmurp-TX>4pAP5@73_)niEpD{aX#&XtM{9BJgcJx!2vraHJ%<53rS`rd}> zh3zQ!vkeW_xRAM!1GQRRqgNZOsNc$wzGbpZ^?O(P;^|E~S{&+?btjozFIpw%NsD*7 z(IU7|_aZ0SRO&#sEc>1tZA*L}yB>Q7dfF}Tr<}9l|>`r%PxX=rCXR`A2pzpdaq-*3( zcl9{5!r6loK6+6>g9kla;z>F+ZX{^sL3PC(`Z3O(-XC?R^?N)>ljXj5x_ME5zZdnM z^rWy^ZghIYgVqPKjQKH^-k(3|3?`y+J$!7e}Yr~#NJtz%q!=wi_a8_%A zb!G=XTD0TBmrfL?_Q2fl5u}nkptkNV7LU1u9ipA!`?tW4{olS<^$2>?4t`1p%YZSS zv9lFzrS0f@aT|7HnvtyEil$wSFoCSEPx@4by#i#qU2xe>$CO;Gt)4ZlA%nC8)l#ih65 zRMvsDo(~~A_a1&ThX0S{GZ^LcV(W?fIN#KT&fH#j=X5i#whOXS5Ai~~pRtt>U?21x z@^fCmB>fSjm)*y@!#xnS>c%^t0kUwBqZyJ6J2(jStFwcw#?@J@Err^>rAVm>>Ul@Huqi zUc$So9~+q7v(xb{wEW+qA><97|9X%7paJ-7XCAu1H>|FBh9KKMbS)i14r6TJG0mML z+=`XT-MG+}2_KUTYX zA{{rbW8P8UiwC2bI5j^T35*TdU6#qR{&^5PlLc3=Y*bFkMJi=L^I`@(Ym%Yrn1UJl zF*q?8i;(gJ%)OKXPJaq+S7)McGyw{hX>k3Wijk-kh^)v!`v}WviKHi~0<}Cu{7OaNwgTKZo{QxxGSSV;#!1G%8Omp)^g|vFPcDQ{K_2#rmO?$L z0^V}D7*;F9mV3pp*2{)uK_Rr`a`5GJA&%7NqeD9n-Y&WD-ByZ)?0HKM=Hu>^GWcG) z37d6gcpzI0&$2Q|2IfFbs1PimiNnQt%$qI5Ws3gS#Z8pusRZbmdrvtuFhmXFfCd>DM>BS$3+K|9%8@hAvc8$+RB6pE(J z%^{ArAUMi!f4K_kz$nslhSl7jb#kZmneJ*=R+bp0kY3xaCb#K<7ksHc5gI%CMBZ4J{B{7reNqp zIy-A28GDCQae<$J6>@C$mx{)8{YZ?AghT#$D2xSS(Do|^69yBJa4;HmL78|xw*Zcx zQm}SK5(XwFU_uWcP2sT^{g{mXIc%;njK-l;;TU%3WAbn!v_fJL(VB$)+tR^*nS^af z#us*<3}e~#DsplAek2@fLa=aA7)pwH7}Lf>!i)%nK8{2+o7weplW}fG0jxsPki9zt zJGksR2QnZjk_n+JiCD=?fW*6G^tYv8y;ls5tqF!rTL6Br{%U6yA8Lv*;0ZER)$7yx;J~M8rlbeLqLb2Ft8ixzFqw#lY9HQl7(JGLHFVhq8 za$z(qS$06V$Q~ckoUs0~EnaSS#$}!@1TNacEyxi)qfXc$?Fj#MuGoLe23_ZDu=1u2 z?yq))SC|E|X1Jq)=}i@ft+2Sw4LcXuqvz5!G&F2ekw6ULa? zL%qoXQ(RquYFE7d>4rJ8oFTl#1$KI_a6IV*b}kBbFJoG3t~=Z}xnW1K3u?99pyus{ z>lc_l=Y_$s( zgu5f@h&?_zo8U%}J!-bH+((KHzRtJ7VUKID_qc)&(`{h8)E<_F4*34o4ZA+rW5y9T z_$9lbd%y&dl2>7H$P&M5?9sW)3JQjHI5^D-_aj{ylYb4q$#(4XW(FHu69itpiqdEH zkc?#O5_5!fqctMs9a$dBiRtLVvcP(*7 z?s^ZD2>RmHIZtpcym7_DAJf(RF!c)u8XQlgJoAO1KGXOZPqFEqJI*l%Li?~69t(M5 zNt_Fm^c~oFAa)p$b%*j~4*qWSK_Kh*k9`eo-t)s~NP&oZ)i6*DiSmOMMqvpg| z+PW%+a$iPMy=V-5pC858NIs2fMbe~q5!AOYoTg?6)1^#*I{7b%4oC))p0ht~Ghoc3 zr!PGn@THAU{YmOY0GUgL(2A=ev?eEnJuZZ#)5FMlT^J?GhSL}8NIEqmf^_(iBr3(D z$`3pmyC;ZN{p8a5+kxa}8c5U5`O<Pio_OQM#cIMa=W3fdXGTaf3%QKLt~iawrWl zp7g`RNIE#er;87wNV9-PQ*^^6u$NS^eNq=9UQhd@q!G zio;32A(}$AMAI~ZXnI@7xZ0mlWPFBCXMV-d6oVM*k%*$oyjUs(yPkD?QryP4*hDnr>~5 zAup8(O1B83UHw6{LN0`qi~Q-gQ8Y?jlaQhphtK!(F?@&B&3?v6=CTa+NFh{=6CiJ% z4o{nO{9?b$_DX@8MG}4pry{K&2}$~F4{J>ZcV8M#rlsRcdNu;O@-exi1YY^Y_)t>{ z(=9iU$@t`@Ysz3ew-S>lRKRUkAyx-eqMCmL-EWJ)mB_>c|5R3AlJIM4GV2wiG0im! z0ZUnSI+)FbS3+3!J`$S_@G;Lb6n{(XBZCM7RD}(55>Nm2q|s34yp6-MxOm)BPsQGGsrcNJ#F+B}{GOeSJ^IBoNW$jH@fdp|9na-ck(kAp)LmJa!5Dur`DD1T{eIy` zC#coBp_u&)7k0+RI$u1z=mxX9PEa>?#Ew_2M(BIukFOh+4LG6O)d^-(U10s%5yve& zk-FRw&-z{Ph3(S|wb`DY>4L?%Ob1SMf;D5j`WX{8j_JcwJee_Vh61k&M5ulhJ#kj$7*EP zRO*T2MjqfZPjDZPd2gR;D#Jz#9$fu`#osJQEj+78CPvHLLXp*!AB zaEIekZ}|6nK%CXNtQkHSI_{1130~mUFiriWC%WAOkbQ}Z>LEX982jLhj3+w#{a~)? zhshdVSp3Had+&0vpXupF-ad$Z=mXwPKU~)fMC+(Ob_)37Xkq{cll^gduMa{G`{LF@ zAJ$`g!Tbp0{6_sCCJ=z{a(=k=YZIwDOH=n-Ns4^Ci59%yKrw$LDT&1C;JsCpSh!|GUI_h7)iK+u6NwZsmQl4$5>$^9To%dFnI(IXTT`5b7U*%{}OpbJS68Wmj zkUo=BpYXR*y`l^)RNq47(>BxRqV+VQyn&)?*U=Ux{X_DAFGH-E{1;60QHZhfe?7LCUm?46C=(=1xVr$(N(0-{fif1$j~(Cr{N| zh$f$9uN|_TDs^Q@M@pI^thSJ1i8S5kNK@V|DVoy-5(?T#WsVZ0AF-akn}}0|>}m?1 zC`SKESCjUBG5Yy>70DE?C7WGqsY-Dzt#;T*0kbyHXTf!}!Bw1QX-iW7(~TrOL5hS1 zq{;l~7E;jNMt1eGw9I-3ZCETzMRqF0N|0%M>WHM~SSb?W9v1 zchU(dWqLE9OaX2?XuD7b_C+VbeR&>y{F4}yk&HUVv|H3?V6#vPRx4-VS$8fb#uT7Y zE*0x0WkGtR7}GW|etc#wlBcF(#r!BZK8wdjDRw3Y%K^9~Wnx@;4vzPy;ryE%c)!iV zv*2tbGlre*fVgxckDVbFg^7l_Ff+@>t0`$L^PR%ZVab5Us~qU1r$Kf~K4h%ZG2uu) zyk{0cdv6W~2l8-}ot5GHJQZ~lGSPW18^Ox?&_0pRa?>UFTc3toEtwE-PsjVBOnBc- z#j^9c5PJ}dc-d4m6=&d$D4UB4%FsBo5hS%CZHkP@!K#3?t>G9evj%JQ+TJQEXK zGcc2vfv4dym?@nD&58`RkL94-i)D> zIj;XG#H#5TFcYnTi(ml?CKo{FRtc^=$Yows5f&8{Bfz;FquYuQTU-J`HI^rOR>|`6 zwXj`Yg4xD5P%^6)Um93uPpp9Te?{2Xm4~nOm9TOwL+znDY`s>G&ZK(Wsm#USrhE)V zX0d%M7rlqFFis>J%Vy``>(nv?3umLGEe9exg~+w8K%9CB&c7~)*}i)GdQuPDeeF1M zzZROC85=&m4%Unr)SXp>x1So&F{Kjz+J&fB$;aWh`H+lZwf{E_!EN|w0tpvRGTogu?W1dnaYVNV#!K)l8 zEhX4Iz7la_

uqg5?04g|;K1r4M}B?6hW=6Lzzj9U|?H4{R2~TYu({`{ML> zPb|v~L>J56kMC!5(}*YP;04()?&xJqwnYgCd8*Fnc4Ha(J}2ZSIb-inSA??p^3)LH zAbzsfKjw{>5rMc-?FCWqKx8ug-d8XTA@ltqJsgNO*3U>LbC9GS0u6sZTv_7}6`mJN z&Urw35eMdK9QfSyf}n;E#Ed=A{>>FJZa(Pme8EzV}7&OkezF z-nHg7Hz>t+(e z0&rnM2sn(ZXe$hX*0>l*RtGW-JQ`MynC{aN1M36HIG_><>luun(r12*bR5P^i^gx3 zy=v@Zx|Kd-K6;aJ*(er8k#V>(Cl0bq->KDzhf!iE_}nmbu8xPPTP#N8($TRo21Spf zurnbBQ({x#bvqTM$J4=Cn91^D(J*26eriWL+=pXOq7)79j5K`WGlpb*3f>(|#tQ zXwxZf~oV!=kR>G8Z)0}*3PCI`{&X`l{xgTel|I%&Y`)}=h4x0 z0dgNRmy8$8rgVonByeAV4zQm-I`c^=T7YJ_3Xw|Qf8@S=F>(GZqVU}!o z(P{z8YMV=wf6XOT#{WC!&!z1WGbz%4F4-gt(y={)q&cvFj$B+wkxNDBYsez{pvZbs znN<`zc?HE@T0-k`mr1(S)jn6gKlex;=d{sdb9bq}5C5V*64` z?^!_wCzsLqXG>}BsTE|ByOgXyEus}0mQab%5(-+dkX{vwQ018Ul*iViCOVJ8Q|3^N z%Y6En&(69gK}xk1qW4boXgbRa+;fWg<+YZi^^#RD^0Hg=x;9FbyqTOh%`c zP`<%ZieR~bOtGc(Tzer2-xDF9A`uGYFQQ|-70fqZM$x;MQEmJx%1;rauPfJ6Wy40A zbbc*8i&#e)4&pSZy`J_Dt)-p1Vl+QojOKW+qVYU2s%c(L$^7*+CVxHct6oo+^~LED zebT-`@0)4M`Ar*@|WVx z=xP{#SpmfhqKxSi!9e;$taLZ@$0$H$vNEDN2&cK*kW#CJH~!mk!ecw@ z8&sewu86nSR1nOVljbT_w!S?W+_Vb^a~0rIvuljU%2vkd0Gm4w&B zZD?Z8(aVvNM$>_XG#!MRf3|FBHW@CvFoEU>U0#bTuK%fj{$k#uj7#tJ44XX z8I9(SIALQ8mu?HZ^fpDqtqU-JbPhorrk4(|GZfF-qyMxO#D17!*wPXff3Ks0V~c^C z_IM)aio_@voZ-44_NEmK8tt*vGU)_A_c4n>2GEZ5K0$au+PvXvIyFi@mlQbsK zm=={9khq2xxkzi#wO5+7aE3PN>FQI%@xx>^?g))*K2GwJPEyvv(^U5LB*}^&p-AWB zbjVzv#ML#)?w2klHfoZ(VP<9g6v(O&><|NUu$sOqF$Muf7JI+O11doHS{HJY$Z^^@!trgyym{k3Eu3k#XfY zdO6FK&Yv));@6i+!0i$#EHb5bKMPtbVnWYSOi07Rj6_;ZY0^&%s`S1>2dm6U?c`;0 zGO;FsVG~;GYC{T#%}9TZDc#aBp>oDCO#N|&PF=b{s~4Oh<^D^w=KCr7a?zabax7?G z#We~qHKk-9Q`!(>N*QftwCSojO?YyiE;yf|$Bw6I+t<_dUc-nMY&0gLji<=)?MYf{ zai0FoJWjt?AEV+`hGhKUFsaHI(DP-6R4%~Y2Oq{cJw8RBKb@un4@{{0^cnKFafaOZ z=g5j($Jmn>=yCN$Iw@;TX$EGbq1AzHwVmL7zK8qPkFkSgwOdc$hy3szh)#Qe2iJOG zdG#Jn7T(9j#hqAk-~kr?PtsY3Rh4*O9}q;$Q85t%3%gKR+aMG{QLz=f3sFQd=@B}4^&`~ALu@G$cXbI*-)&)KooX9eY>e^epN?tH@5H)XtcXFlM_ zayUt{hw9A>h$_E;FYnT|m{VJom4iud@{q|M*R^7WcxY7yvBMP*JXMKj#$|9Tt%r+Q zC5+BAV~=1ZHt1C1RmL~mRV~HGfMQ%K|AgIn1(>a!2ZPb?k)fA^i&5_&z3KxZvy0L3 z?g!=zvK~RS9821&&~mR4mQ7y~IHM3E5d}CmsuXwTm*VN19~i>NT;^DXgYE2}F8RtD zxz7k!s=$zqX$i$Cz8xDq}}m56&#kLNSKBlY7qEL>WGp!K!5&hOPdP5$@xQoh<1LA8m0Z}^D1 z@H~7u%q+Wo-?4IK1$yOL5c#+TF(uWI(qbRC60@7WF?a5AHztq!iG<*0Ov$LjVu^C7 z52?ZCZKb%FSc2T*3f_k`q3vcd>UV#`Kl3uk=YBy>!)I76W4`dxYQ%(BL(%6ewmdCD zeS0wk9Dcy~W*IW}m0`8vcf`-NK@IQH3c2SBjGr#h)LVReUb$n&anQB^}_#=HQwjhLVmj)ZgY<}hy4^4+uh*e z?}#HJ4y-M)fqE(bwsFM@o~2e?vxUi6YgqT%<1=g2<8++h-R}Ue&vp=Xb3n{KSFGM{ zixmUhA4b^V*HIhvNLb>ioedt%w}IgiCn&RiOaF)u^h@1Ym+S=t-W{ZqJ5oAaAvS__ z=&8X{4hM*Y^RXw{!m`5-pZGZzYumt~oOh6K__#WDm?7&3)%|?@a@M{dw1@b8 zCp@lqM&CRq)GxN!2`p_yTQW18#1STk;;1kb=D@!ui!pt2KzW>Il@5C z341QMAu^Ibi$QmU?Qq0*XE#jx$)4UiH+T*8Lr$|7Zq;ya*3VwvHcy^Iu%>>9A9e+L zv@)ggVDl+=m1r=4&+)Q3fJeQ@~k3$H}~BBJXzPLKM>te@Y=i|j-_ z|2OTvPG%4FL(hExAKL_}x8pBbPY6;}*Z>T!2vX9lK@5cc;eCQIspdDrVmSM}{(ght zpE?B3YDe6^X8e)E(8im&d%fjQl{fNm83<>zkQjY?+`^70q4@)sGUS5phE) zwg^(6ya@C5#mK5ah%RMHl0^SFD)1B|=K(Q#)i2CHW2qx}94$;AL$kh)AdO?g zX!6O?lxs4Q0``og$7w_9gQh5@4;x1JHFV&mb^`)~I(V}G3asq3;G1y)S4N$Mw3!Ca zQ8nRbbQfzHE@2(lX4eOGvB<+1Ip?*35IsD3VupNMb8Is-g+-?^K7Qs}YOXn|M_)ls zycX_lxX!Z}EzD!x^9~s;tR2us>kH-%mT(3|FY~LH+=cDc+gN0(g#%oJ4&LC$;@x;$ zvk`_QnLsVU2o*2yz%EQ5P8FuuF=&d1y8NAH_F|g(UFJgQqso@|>}C4!QrG9%(p^m0 zuZ{CIx)6~u##qv%p~o6nQWd=0$`_pg^xzwZ)WpSXi@%(pzN!@O-izQn){ zWdFH_9QNd-a{U@tZh$;BeVkf%8%|T1C#kH1S`7m%%`ry?YXMezSRqZ%1a*?SI63t94(Z{96<=G;?%-gdAsiOpf_1D8 z+)v(s!Zv1Cj$oaHKF{SOOmXG18J@9sN#vI?^7Zwg@=q6kNe4lOdg!>W2Y2bKI6Kb( z<%fwGoCNzDvdrOMikO zvy>a37HI^mUkDfN#c+(DkKBiIP|`3LrN`!@IAk7PXs*Iuk(IEqWDl>)A|!DCrq4aw ziuvm>ym%#FQ&+&XP!_S<5zc-JoKdh2x!&usFGn7|Yt}&ORC)B-t;g>eAktF-`!@q4hOWc1k${!dI-KlV&pSs&+|?ko zYRN+X?K<3Ql84n*CDx><;K%W8Q2#0qpDD^P*H=Ul_k6M?>`iW2jkEy%F;T`|@r}6Z zwGp=)H(*e8J)Tvp!}5D_&_BHiQqf!RdGapAnC--|h^??cwFAt4en?hw20%R;2dZ(prwU$cD=|c-0@EzMU}01d&K&=YogIZR z4=6(0z0Y{@pbV+g%F)wRj^4vR&^v`$qx);{Jh~B0KWb4it_cxGoA5KI8fmS}ie6p| zU-fTjm|l*;A7AmX|2tkQHDIS-J+AL*hmYnjT;JUazr+^!K#8(Em?;ngH8BzubJ!E?^ws2rn`TBz~jPq#Pxn>uj@CQAN3tS#|8s=RaCbb>cf(6HFB}}_j8(bp*U}EfKu;*-irq2g z0?%VpyfJgPAGGp(u8T%mrSHR&2$!(Dd6EDJ~YvY+fl zx)aWSaDn@KNA%0PLgcC+hPDSHGujj1SvR-&l`DE{{IFr1CmPFKkz(eI%4`n^Yq(&) zm@|z2I3hBSwetM`2eV(%eu@L`hPuG#p(FIdSwGL(`XgaJP_*&F_DB9m?(#x2&w!QJ zxS?~aFOnv@);MQqukyr&^Z?Fk z@WpafckI&e#LCOA7&@pp(Xre%lVm|h5s`vRF$ z&0IG5DAYZOhU-+;Z9WJ^!0JHEIpvMZEBz6|@3#(ourkFT3)Y&!YSCS^^4!6^|1N}T zOrc%D_pp`5c$j5|<(5Wx^71+^o#t6i4c7=S_?}&00AJUe*tGXHey%jbZ)G!lZ!iJz zy+5_x0`*5N;lAQ`S&u*aj@}(Zfw6pl&63fDnG5@s`z&x-*%*gx_3-kH0i=qUqdbE(A&$mSDB!=< zvB%p3=FF3|L&{C=BiOroGt!pN9ebRTwukpUCsd4afWb{iO#k4F2L68Nu@~F)q#Kln zxgnw60ps`BV;bvVQm)#<;;<#QbUWgtB=<40T;~_D=e64%JKWvibf0~g>Fix8WgkjC z`}{uJLSM_7XEzQovfv)0;Qyc3{^xBWGKIAwx13m?!JMNk8yx#$j|*QdF_FEL2kxM~yf6qB%68AwuquF=gXb(4UXQbA7;8YmF&OAF5aSu3*`_?(UI~-eK4VgL%%s6C&-gMp}p0L5kb}I}HuwJRz7JpMM zQJTYhKHi(n6tKZfMfOwp?uv2=+aWNbsRo_k1*4FQna>wz`m0a*LX6{93v zVSd^M2f1g?_c1}{L?gU@V2p(Cw_#y#f(brG@NG3kDDOVSEq*{mq!A618gOBECyq4# zg7T0?)?u~q_qhc^XyzQFLAZp~!r812t^c)PXWDlt?frmY>2fT;ThDC8Vrgto!#sZJYAky8 z4dT+ZIJm0`lb=^0%)b&HZ@)wCVl6%~ucC4E7pSi)fPGs97S?~ovVY9=tEt4qi!Bgn zX@TOJ2KZd+L|sV_R){iVWK$h9^J_7CWj&5htii@V-|=r@6IRWwK>OHlc*_|ClV6nK z@w_5r4wk`o1m{8cF(z9P@5Z>2>VBu!0dENq5t!8v(H{g(N6MMp2kj#&NrK1)B zf0~fn(TuzE8zCD|jhfc)xV60k^9$?PJJx|+vze91tdrFm%}`!e2Z?L#I5w&kcKHoB zp79GBb@Vj!LbC&t zr?p|j^cGb6w;<&#v(>xeP(3^uavSeM^V9>VU(JA>Tn6_p_n^8j7H0Ap=-l}TK^0G- zQpW5#i)Z+9J{ij?6EdQY5a^$QVv#Jwbw7vVl5C7>$;S1$kMZE-Lp=7%#D&xM;EF#hvuKz5BQ#IAZGn#yhUryJCw}$i2s%_ zt10gpI+(fp$Sn(c%vaved(EkB=~#Fp6K_K9Lq+=qG~RrK=kIa^ZTg7M%*gA?dX6_D z+3;!Nys@mO_`K#R-@_kbH|t_Njxk5~b{e8KW}wtA3!QfNVZVuaxnVDmu_qJq*B)d0 zu!m@2#w2nVl)|4&Pr=^qBKkRa@73B;oTe+>P@HIuv_3d`BkIL{xyXZXYGNDPJ?h(Y0% zNNf`eMcM<_?Y?F1X{bN;j|oMXR0wnJ!%*EHj*{*u7_ALKZDAnFW=CPUKqA&22t{r- zKUQP_Htc5oY%uF%nGK<=6@*s@eUb4h1kK(7%-s%wRBkxN*asnnwPW5I@%YP{1IjNB_{S|{y_8VEfiACjUu6^D_F?TZxU(_SvH!&2YBSNqwF&Mh6tL^(72#H+I zY`PhTsSEva?_xNHZwzMJzg0F_sVs>IYoDAb2`joloB{4W0oPggR(MVt&*^bN@eCFS0 zR>i{CC>qN$k}=gj74>b&7_~DE?dM{ks1kuBmnf)D4#(81vDkSa0y-IqPz_FkUnn!L zV-lgmT7|EYF`TCvjamIM5cZA0iNld}_EH!PL`2a6pLi;C1_DW*EnipKk(B(aUW~>@$xi&&DKD$WEpv=>*D}mqJ;MspS1IoqY3> zN%VF)O)O8LrtJxo(w#u1d^~xl7)pN_MR}P~@zX@z|KYv>VXEefm_l&X5vE8c~DEU7B~^h{m>Z2DOtBMf#es zr{0*gCi=Ae_#Mi>eVcYJ)}|#P*GcBUWg0&F0%dCQzR=(*y_kHJF5YBc@Nqqgjp3}X z_jgJ0f)VZf%$a8KX4DpGOtRXhbnB`aMea7I?o}3aVWR~tQnjF?Kh3Bk%Y-(6Ga>O+ zCd8UBS}Mq|Q{0ZEwH@hzr2{Gaa-z#~oM_2(CtB9+N@p!NN7=@KjLIEpeX1SZNVey^ zYC9_KwV+x0CiG>61x-xh{IE0&n%-na%Vt`TW~v2ES!YYZq2@I7tre~R<3K*CPBf73 zObIoPB-!mwR-?QrR>PTAsJqc{Q&%cp=|aa;oJm5-o3ggfA+*T@-5m3Wa* zi_ovFoQKo^-^~@Mm|h1TyLLQTTMvP2&B%G&i*usCfzfTu<86h|-DaFxRfAg2RMOg7 zg&dg{EP3CA{i~`_@~RPgTHBy;qaD{C{6xXHcBl>gi4`WztmA5Du1+(Smhp223t{1V@eORO2gNPiS(PXq^!uCdVN;F`dO%rQb ze?iTI>rweZ#NHF3S1!z8Q~8C7p9V3f^B?b5f3arl9}bAOqeQY5+uQ%(O!_Yf@Lt5` zViS6w_u!&;8-jd)VPQYldtv|JlHSQ#Nv-&&(SaO?pLpQW3MJkVev$vpxj_GLXX0-( zSpCA>PPdY=pcd3(rtU7&*-bXLh_S3^V zyXfe{tu(rqHSkSJ)F8Z@rkz!#wHvq77{C2A-T4UZmpx2*=?7_s!7;j_bCfQ)9w5=2 z19W%)5mK>XT|?n6+VbHLomh2%O6^stv_qA~2yUg{BAdyud^25GyP4$Ki?B;ekyh6! z(BKv&x^Z$F=^WTag}>EkjMYB+`e-lR^Vmm{%m&F@b(n3Jvj9^IJo_!m~icVMwjAKYYm;Ar&|h3`2dO}Ynr zYPulh*^Yd*zi8?1!lWyGxa3`juUERE7}$YDBl_@NzYF6JccY-D2fuv;C~c)6g&Yzl zWsjkhmm@|s=SAq_j$t(DJdE__i&52(VU!v!NJ{?Qu$a_~(4D`q#E5run!HDZ05ud1 zB6r+Sx>7Gnh1Fsd*(yx_oC#qeLkGr4XeG{o%Rq0IW{*W8AiWjF>Qp z8qQcs4HKm0y#u(S{0A0oeaJ{^Lz6})KFsYzN?tFl#{I&E&MwRm?!$S-K^$HvKwC_O z$?doxjaxp9R_z={eg>ixATLTzXGG}rY%%gXC`?f$!ZgfRgaZ0R=KX2#GVfNThr3>HZ(%Y zo@#G#Ugdf_%J^naVu$VN=Qd}WwaS6))os|5Vo%rh+mg&JOETJRNhdB^QR8Dv@>4V? zw~w4R>TFJXEKF(lbW`eUG9*7Mb8GA_p>WMO=oslM#aNLw8#+Z|fCf}po%<23oYs%=bB8hhv6u;h# z-hH;9SNRU~VWS;6k8>cs9rl!c(SgP=dt*V86=??AkV}Rg9l7K{aSb-4XK7DgtsJNw zZj`y#iH=WpribSpY1bHAnzhu9E}yiaV?()Sa^l)b!IGwjnbDI`mh|5)3mWs-jLZ&L zQp!JbiY+sz`K4wws@Rlf9N{`)hXoBAFrlMamR#dnP!}IN`?xJ#*07=I9%~YdHKS&( zUCJ^oDb3%CF7a~-WN+*{ReL(s=0yKOILpPxg?gW`H@Df1LWa0eO}huxn7Pu0d7SOS zd7w}Dd5`UPAhC8|SW9^${AVD#*^i$t6Nm@jgP>CGjTyrOFj^)I8!ef?`8E(I*2ckz z{rqp$Jj-#~G@$nT=BjXPUloMHwI2Am)ft~u{9$<3m-C6)*J?%24gM5AX^p#iyKi`s}llkvkc7F=M3e85op~WkC{agnE5go?h5|cXv96~ zp=d}7g+ih-5-VcD@osA*UbQBp`3>g-v)+AiaxhLrGv|PPw*kcgaM{hit!s%`%FnN# zVo=LWgwZV_c<__^aH~*Ub__>%Oe9wAOhlkv1`c_#=PEH7(@uoKbVMW^bRw`{i@6%V zqIn)00_)qM*ncnpt0VZmZ}CMxAOEvPA{M+$z_&edcsnl|w?osg-Y@}18&lCN8jtMW zcx+!3%ii@QXbee0Y!!P{=Y(U>B%E`W!r_0NIgFqD5Pc{Fr=LgSH@`mb&mo-S7K258 zaX85Rw&D3`^e{(czIQCjS46N5ArwPZ!ddIYS=2xM@ijgaPnU-xh?xr$B$z!iJqm^U z!f>Ab!0SQ>r}Nx?_%n~jfWK$PF3p79BY9m;uB(_%TR zHU*a-WZ=OZ=4$jLFz-AW`a2SFtAqI?L+|0khAhr!VMatj1ZowdAbC0tg=X$`~t ztw|`aWQN8T_TUn;O716Oy=_=#<4GYI`hVg);D~ zIRRk@<1q17BJVF^d480F*IxHgeJT-W29ja$ISyZBqpaRg7|fR)Qk(l@Z&)!*%HVM@Bo~33B|_Ef!O#w8WZ2e;~SD7$Fm%nMNyDtKVn*M zBt9lYV6SU9OoxU;=20~3z#}pKMlv36O~h{H1jNKL4@)kBdyP=wQy6mp3xnM1Xq<6m z*5ipN=DUVs!-HUyUJ8Lq4}a&{BH^JJf+@pNFn(MN!mdRF6Jt;^KMqN*F}Nd|fQ1^# zSg(@Ao{WLrJUfz�frs*5`3n5qma9Hzp#jI0CxyoR1`%fW6s# z4p=0?l;<;%OEMwhnS|+yDY!W~1;#c>2-}x}#@rY*q$TmbHXUPjBtgO?32`+^DD_Fk zqwN{^bmtx}?#h7rr({_Brm+6&9&2+R^7)wp-IwuDzZ?NM@q3(K!hXw?6qId>!APSx zOqOHMIM3%smU5<>DEk+`#$(3;o-J)o#9YB7_DQ9ppePw13X_qssswXXKBJL$^4N{5K%ujB{#`(>d(Afli?G~)6ZA10+W}eNq zLik8Kvq3u%W!?`t$2J_3X~)&!Eyx|&0KpyK@x<&qM$F>O+oo1*DeS>o*&!sYBt$}u ze_&qRg~6BIcwyLs+cz3nPt=Ar+_%k;tH+k$DtxW`h6yeeSiGwqb{*d!G_n}0S61Tn z>vC*uVa=o451cRjj#dA@BU_~k-SRakIA4!IWzNd9>w-&fE!T*xkbB+=PNl}LV{O>? zqlxpV>Ua*$nW06EcyhP}=Qz)E@48OJOz-5~Vk`Ttt03=I1>YIZw3RC;i5%}=N)^y29*?q3~wFW}nBT9jX?#!|cs#D7#ib*{ud?*FQk> z{#)!h^aA-`USs3yS4g_ZT(sw}5ikA|-#7l0a~_4zBlyU_f!F<9>=Q3X`-TsYKA4M*4(#>R|BQRn%TO}- z7OGQ9SZDAS3lj^F|1%#~P4i*WQiL^9ud%xE8yfdk~5$`G3Yh#b;bi z&&BkF(1d;`xHAK=vb0TT)eP;>JYq!#9* zyNolL5_Bk|R);nmHK6$K#&qwY3AG!TvVPN;62(nuYLYn>Ofja?979UGWk`w-Z_{}v zZAvlGq`)(p1VwGq61_#ThjnP`Yi8&zze}gwZjrV5U0PbGLt|cQlY^5kxgOD{=ufxk z_!twiDK@5;zl~_|Tw_|F#dA3!p8FZ<(AbvibkFkwEf3S6(aX;8KcA-D)(d2)bA(Jb zou=@#GZZd&kre9Bk+StgS|YDa121pV6ph==%ru~4o`HStFr!()ru6lXDQ&xGNw2z0 z>5`oh?b%~WuU7MG*lt4mi|$atS!4Pn%sqh&&liWA&|R_Hw7TIg4gY3HQa1YZW|1My z&o(6Y>vyQ-s43+hF(rvJCRDn>j4tFDQMSY_GK#xR*)BKfWzk)lUTHx7dzdG<_crS{ zb?NHYyQDQto0|V=QnC5h;36z`-%*O-@hAnPjKU!+AzKd;caG1n-@ zNQ-XPUZa+0*U7C*hZZ{BC6P9L%DZPm5k=;--ouQ7B#o(Ife8t$*Qcc6hV(hdfF31= zA+stF+8N&1BIO9{Lw@iw4uVZe2+CdkImaRdTe4y~A3Y3fb>m_7AOOQ=24K6n7khq! zkjTA4@1$e|6!_wbIeSP;{87vI|EGN+|I;2{I2ML;0kQbAlKJ4rB5-kKDi;09fRJA_ zX73Dzkar|BZbhRXAsEIxp`4&txPDH6R6!O@x$c(adeC$a`?&@9-dGxqp?jlY#U9SoD9AsGz@n20a7{|Y)H3D(hzH_b zcqAG$nbYz$6tn(Cv(_#OAWP~MN|#bJbNGBn*2A)C#6o^XGhc(GCNtkyjoDn^P??t*pDE4bVYxD&uYPt}q{usm~zK>bmZNu(@KIjf>M$CvV zSY)(e@dwt!BVi)x7V)}ilJ2ZF9PVy!_3Zs~PmLS75DZmPq(TP>K~!1YH=8^ktt za8`CF_C+^fm`4LL4%XsdOdaZ*ny~OQ*DeDM_$u9tfhA4M+xiLhoBY~*JE44{6HEB- zZ<#mXnMN&4+iG!OTQfE-{ek{dKM*;(76K2eAbzF^we7VqQ2UB?v#N3GSOtRHtKsEa zg*(a>P)V+4rd$=)U;B#N?lrLXZo&_tTDZ#8v8IKyNyNK(PTPs_`M*)=%GoAbJ*bcF zLXk!{)Vv3=)$|YZ)p{_^updvy4RBVqAkADcgnHlgqhf{t70V0Ix1mDx{_Z~rz5fTB zk%B}#@42*m5YoY%y)Ga?0$cmh+SP-a>;14>El66bf|M|A0J~)b$#(k?GT0`zc!$H`tIE|KyE1a{= zn({XnFuGC;!%MHC=HyL?ti6HNV=rR&=3DT3c@^5jZ=!snCZgZ4f9^i(&6eNAiLrO! zzxp=kd>Er?ya{e8n_ySh9n}8E`s{(5=n}bwRf)IpFyl6+j=jx{2whB7XYZi#S-!rU z#|7zgC|5s+QO;Me-~Ad6%+kUM*Gmvf)xd}qI!Fo9M)-Xr_`SY{Jb5kTni%3mq#F4ZO3`gTh@ST!_4dOQKhCn6=&(Cv_3kZ-k#8^Gc65uNKu zv^PYQ<{jJ|V}PC3y0|QO6LZcQV*mC#Fw(z{*#ehg+IR)=M>Nsctbr-21~_4Cf{Uzu z3pTXEb8~CBxSHc4dtychT43aHL!9EwJ(UM$82{IXcP+N~z0wxZubp6K=7b^o4sh6P z1BP%zRhfM*AFa6Gw8oVA<{05&gn6pB__Jl6|4d7sGcZrdQydH|;F@(BR*FqQxYZQA z`aKCP_oty{#!TLI%*Ec`*+?mwijljfVbti^Xq_~R>v2i!SSW=RH>J=KG7Db&r$g@B z3}#77VP3ujq|0YuzKRs~MNGk$i8FBf_ALB~oq%HtrXjX{Du&2SM9bN!*mrC?l3z@R z(6x!2do_i#4<=%irX+jwrQjSq6SuUdK{aOv-eVR%-kHwXViV9KJ_$QUjmNpVDd;>s z4R^jzMds$oNY$K*(DDiJnL7>vGE?E!A_)hJsc6-og-`9X@#MiIXpEbRSEUnR{AL0a z>_#J3MiL{uB%#(n6{7#9;=KMe=(o)PHqFI9-#P4amhguz9O!zo*L7io2ImQnlVM5WtQ?+J-Rpf z4s}erMgPU?(vtKWq||ns3O?!4*jh6>Y;I1&SKCnJavQpqVNLH&*wcbh4rC)}Pv?)@ z(Y3$U6vdprId|+xVwgQShdWSisw2sbvZKGR?CJ9YD{3&cpnwoFI=0N1ZX7Zu-`VEO znKGv>Ja2xdY)dDPIn(ab_O$A%J^gaDrR@e*wBn8xwbt2^n~51+tTiIPt0r`CnHfn% zSsDkJ#!TlW%yO=? zq8oAMbakQy)$wE6q%#M))`a9TO=zAdv#L#uNP34Esil~ZrjRLVjWDB2uT81rq&ZnG zGos@g4asAR4lU`wLq*Ygbh^TT#J?HS7zGRZ`I~vqnl?mo)--{4#!qJ1Qbmp}dDmHx zb(Jw`yfviyDJIlvY(Oz16XDMsqVKOBplo3x{J7?-b&7()on&Y*b2e#c9OrE^6Zqjh zEE$u5JB|tX!!=k^dj$4kmq;!R`F2dCg^&h%ILE=s`M^M3OP0D+US<(a2*iv0!E>+AG5FYH}D% z2ZQjL=OiZ*LhzVtNmt2WBuGRjUrHbfn;-Jr#vm3C)8a5JGZuF$li>Oz5w1aTm^C35?Ul^! zxE}&l#dw?%i^LF~1IAxW;4JMpoKoOfRZ<-0JdDJbohb;&O@fMA8djc3$L4}K%#}?= zbbBUNC?;bWvx{?-Gx1pC0cu&3=C1b;Cd@P{se6jwOP^!1ayGo@y}+W-Cs=kUn|U!W zp`Q8zcmKYH#fVpU`rU!+)FzEZxyKk7nTlb>88K|j9fg|Q(GZ+A8pSI{BWRfz{_PflK&&99 z%ZOr|gfM#}gweHkD17*Nc25+-uhRpv@3n*>ohE{vnj)B%FNDe41u)xW7$i52MC-7j zn5i%X^X?6WdD1A1mlnhhey?U04$7{+G!)e{#qhIWI40i|Mvj391|Nyx_0{2Umk>e4 zE-|!O2_w}_0R4}LAzFAOj!hba-H$lv>pos zg>l$!H5%q689-P0N1kuwx z|DKfupNceyDaT=aTo!!V6JU3Qxz?`Pco>)goA`Ll2#e)hnGCGrYsuq;K*$cp;9^EB zl)uIxF5w=Z8xL@rb&iSI8Au+&%$uAHY%IEm6Z}11>zWA*v&VS6G9I^LGq61W9_~jc zLCGK!vFp7c!PhjK2h6;aW^JFSKbBPnB4Q6Sp&y1oVQvyEm~)pT9fMVSBG5H00ruiS z_+iVl+hd%k-WG#p+DTa6!OSGqVjGN!gk(u1>ZWHQcSk&qIL714wG`YNc!U%2_Yw6Y z6U&FDvj&E*7h@A(@+=%b_C!NsR{{{oTr~ds>evJXmp_8unOVfUc5xs^@q6s_C6N>e1P2N zPk8R1h6;sbJpB0(8v`FgRF(ZXU6#;Z>x|ZBKVqQ_l_sR(!z0Zu|Ld`gHSUz z1xfk=P#=gyxpoLPkMzP>6(9V^xkNoL-8k#a1;$VLN7)TEH-gdE;RBU1oFA6LJj=~? zh>Ewz(;8o_7xKW3S^gMu*$qJgc38Z|3P+D}u5*$%7TWruPTdw_?<}})w}Sf~C+LzF zTB_Wj=EIpAu5CQ_V_}lEsp1Y;V66U4fkz+sL_nY z4)VixcRyH3`(S>N3pPu7;N1pi?7Z#($pOw~&Tz-kELSY~=!Ku&E?BkHAE~O$@u%7R<_Q(S$Kj<<->G=-!OYKxq1Zho3gMBV@Tze^&v{?;Er`d}K+Zl>j^LSNEW83ju$<5L z1qyubbjD%fs7zFJMIg7+9jS`$7^NG6*REd3S(t=y?{pOMc{(XF9FZCx(Bj;xn0>76 zUB~%hk^#_VKT#cf31kYqFkzD?a$k6IUCsGv4o(k}8ul5@cl za>fTYZyY$_g?|Gc*ktL%)0$)ive**j8SHGz$Ql|>AJ&aBIjgr&R9{HCqjmL;Xn3v5wUN^%L!lj`Vip2 zdSqg5+Snj0EAha8b3HMgJzGB_*>6U^cwFFy85Y5~urCH9G-S+oE~r#hC?5yNThUzjqF%+)3u(_eAj_BIR>H{yBE9DsJN%WlfYVV+Ph ztknWgxQDgBPQlRU-J4>_9G=Hb!yKq_(=l|kLSwP;%< zi-+#3@mOgEb9h!F;>%L(Z=8e0)r+xp^J2(#N@GId9BgQw1KFuF@Y`fA`-|se>)iPu zffe{`wjAGguEF8Ot9e(m40>KGSpU2jNB1p6Ufn$8sV;#1Htsc*q~W((8qNKS@O;Q@ ztX;bh^_@$w==lPi9bAM()0g1k+eH|CWDShw5oe1i;mo1Un72(C7xv5HZ0>3ZJzEcT z1x2iOQ9zd*^P}YD@Lz%=UaSGOJe0vvvo&~By^Q(Oi||WgEuM?BmpDNNX?E){PYJLe zy&8E>fH#|DF*rpYkC>aZ;rtf#-B5tk_ss}lwv*yt88qa`V&Dz?f!E1kcsJl>vI^5u z_;F%aW9NGX%-^qo@roM}<;*#~$K`NyojkIul(EKu;kH^z@aa`%-`^G#AC!l=njE%j z69zQpvHim)W@2wb_i<(RRjiL(ym`Mll0TELB%GITvOYu90< z^=90<&c`fNz@loxBfqVDo!o-%TfhUQ&3LwcGY*T(L!ea-u|3Rw(qDxSv<_OkqA-j# z3}G|5c0a~DPu^kcu8e~(>kj@*3&-l!q4*`t83Mecm0HZ#6Rk*mSBqlpP&j_qgdyfQ zYZ&&1AxF(wL1bz!*kn)j@;Ly$f@5K*k}=-wTQLB1YU)`z3IA_6|UV-Yzw z0Z01cV9VFwi0`4e8y<~ZzRz|_MnS|k4l$pSFq`{=ChY(mS{aH7^FyF-?E|_Oj+Ix! zu%LtW5j+9RrcfDBM^R1>JjG)6a^;dH(mU;LH(3V&HWcEJA}(zbPDr zO2H^P<&Vhu0XSYAgmaagop3uE)rGNWsN(vb@4<3W2{3Go2hOCx`&lwB2PC2^IFcX0&8vR2O~bPR)^Q;t^GtaYxCW%Nn`+1c5Y9OgdrHM*Jg`xos z-skuJ`TqA>%UZ}eoxP8<&plk%=TblyBYFJ3X%tTEP(Z8k^4NW!{XA6(S3Q!&>2u_< zC{iAutdPT-BSzujU^$dpCBw94qtN@yC``7J!%w24@c2i0e4L_$rH%4PgbLU^N*?oF z6>!lNCCpV-z$0T6uuNhM4$PLrdu5|=%>j8_c3%z^+2gjWV>&y=%b2Ahfkg^3IP-=S zO0eS^2o}fPcO`MP#7Hz=Dvc7=66l^Lik_ts7&A&5WA=~47j`3Y(?e0r*(`w~!IC(8 zk{CL%e_K0K3Ny+_qNSMxZV8dZeKw4tGG7|UKbOT0CnZ#5kAt#LL8~lAsE)$o8#1^m zP!Zo{OJaJj4F3903O`+z#-w&B++8b$lAI*g+e@I0k`$(~^Ki^j3G@8r(6~_!GioGp zeY-3kkWj$ESqhl_T7l_W;>nW zT;cr^7Z{dxg--$xP;YgBQ+GVTqBsZ^cm=|+&=ydb5*5gJY&g85pWkZ|4|=6-cz zdlygm#j|{1>IQ%R`hc&d8`ELf!+BLtcs%h4JbQNxPUL&TInE1W-tLG0ybprncW3DM za~N`lTw!yk6F5Bffs5@YKv}~N^1V+$zD*!p{}c)-(?S^!BnSeck3kjh$7UifEaTN3 zM#VdWsg)Bfx$FTOYP}do$P1K@yMb@52fY354pL2tAh$Ud{0}C;A zcv2v62J`xlNrIkw=38d<+*i4D7`T%GUANNUvS&J|Bql>lP%0GOj)gsolb~6K`Rt`x zHhEbb1RYNSMXO}kFfI+aM3#BFEeDRPC&SzGNl-GF1Z(D`vFjnLD`V22-a7+c4X41J ztW=hR$7*=yjSrum#I!O=(D^wA)=XseU=^$THmAZ~b`CBvukN|ZL^#HnR2i(+Z(kAv zuX_@qODYj;Sv|gVC=;Hq$%lUNn^_jqO6EW=^Nx)vONPblybCm0Eoq$rIU0p<%is*8 zkSs8?%mYiCQxMAPzI?*!!}-jsS6u+{sVsX-_aewoWm>VvmmrMw@2b*;P_nWR;{Tq7 zbhjc1W?HY@mea82a5`8!Fb$r5CYm3}#9w=I@%G1DRNRt-d6lPeO3f)uxRi}?+4;CT zH5Xks=HShd=@{miiU~9xUv7;a^uk~BN-J+($Kgt54B74Q1N*_2FMrSdiAsDS9==!Hs|BIBLz6e znC0J$$;VGxxft4+kKOEh?8?D>>m2+ok%3};DY!o(8KR<9pj;0V40K$BH_#Rc)(9FFz_@IiY|n~(t=P}EPfK)jUqsJDVA}% z87q2yCOl)B+~CpdwUx?t`*!SF)0hUY*|q31>-}{rPQv{ukznNCf;qfLG?j7em<%Faok$+#l9lNjHDJ+3)xPr`DEa4=+9oevn_dn3DE zt?f#Mhpb=T-IofU28{cWkO*;KqCjK;%OK}7V9szV7-h0t(CNv5jLmlPQ8L^*$eDWJ2^gzY1Cps!#x2ub;N>jR2&5;z=Ux9g0AcN@~-%ZhYZaW)5nMl(-rQ4)M%Gm_V{nO?vs z2i7jmWjZ{T0~5k(L#BuIvM++q9cLgQlIb)0b3tb$%Scbmf_pc!AV!HDTVx8v-%kMv zF~%Hp?FOaHE@;$d-t+q(84s`zj2(Z#r(<2<==~iw?_ypH&0mniSZW*A{DO76dO+=8 z7ffya3W9!l)a<*L;S5RbOFrWe5`824Uw^Q8esjevU^XXp=Jnm(~u$ zImV4sWW2UrUk6}e&H#9v8DN>PKf(6FH>msl9Tb!PGFS2tY&!P?MDKlv?Ir)fYU==e zJJ%0y8AGp5r4Opqe?r|;_VwZ+80**vTZ{VPUh)u(iun#?$}d}W8u%>;aSES()j`{Ixj$V65~4TX@iE)=OC{21@7ju&;4J(QkL=UnAYQZ!B6-! z^b<}z_ypeDyI}se&oI@!AFR&3gATW^;41zatUVsXL*|2N{qzcw6MCRdndvr|X45FU z1@=x`voLr zG{eZBt?-TMOIr4HvRxMQ9jQs7_p*`L@>>KQn4V?9vTmTNU%+Fa2X1T-MT-YtAlGyd z9%(eg)*o-6Ij#YYcQiq6S1WYZGG^h07B~~q1~K-pVSi`~IMviM9>EikTigmoTkD|n z+$+$0^&TdfeuiD{uc0aTIgIUn4qsf}gP(2}6f>>p>GqE>C!`lnZ(_Pug^zI4svQ>X z>jcM3!yx_S4IJ9v3i5@$Y+lm_!B0D&WZ)HO#SQ~~`w4c$|Ao)_B52*;5AhYuonZ49 zWQRuJ+OuN#9tT0Ir4Km6Z$LMRWA|3}x+-9=ry>!I7WoZZ28QALNtQh-eHWrV>S4JTu=<24jO*2 z=eFY!l;_-s$hZeEY*G*9R*mrQLNmmld;vNI5207?8OzVO3k#a=K}=#D@TSdRAb13K z)t`XMqDMgIJOUq=C$P`G8BQO10-f zTj2ip8?fs9Ds=H%jh$*s(7B&OU4sR9`h)@c%j)2|z}fim;2cz!*JFC3`M7duF7xi| z;gHW&-^RwE1}AtO1@eScn>55eGd@@vFv4+!41D zJ;T;A-@ZBemYJd1c@8a)nc~hkV;s?GjGi zpO}kc)_S=3G~&If0!%u&5H+70V}9HcJZ!WWwXT`rz!Xz#IKK$1!j|IuaV+zG({g;I zz6>iHS71-ma`clk$0BQUY-n7Hm)p<-V&Fn&MkaGyDN7(Z$;gS1Oy~ zTIQF1*S!>v=$qk$bnV4RJ8u2t6zI!N+C?mQi~EoWHq2$`xPu`XvxP%Cmlv zX`sx_9l>qD6Pm`KU>Yh(V0XVzQ6>^WT z{(qYbMBH_Pg-pkF#nKy+N?gHLkFhleUD%A$53Zm0hb%T*=UY&w+|kd9)`hJ9$@&H@!uc20X6c5xmz6}$JP<n$?X9hsPZ6Bt+^@dA%-k_Ix3<6qDfPG5{v>F9~!DR=qdUufJ zwHySMdv;L%?-2B{V|c`#$NXzPu&3()jt{V9xt|Ug?Y;|7dst(;!&aQdywV!C8Bcw; zB|iCNg*Drl7x|qvR@^&?8=u~7wd{*_CENJX~5N{xMHP&4?48^;k{o+@zyF| zmSf?DXBGxv0n0>po6qvyqunw6yA$5@amS#gjFaB&giCbn@bbI^=we}wF;92koa9~T zW4;ZKU*3lH0o%~M*9xmuw_+Q&8%GPR(QworJf*e+cP!kD*8EoddUqSXirIsjj1gZw ze=~*z?8LC{?Koq@Ufdda0HsD8#*5DTF|q#uj^o%dPH@0aQEpg1~Q8z zJLY>mgo;h}*k$8{SH3!7$1FEo;^)RZt1ftCf)oB$@xpSA0Ms1givQ(%;Qw^%6+V6# zI@KNH?jOWOQ|+^AEBfiOpPe$0vg*+m*?OMFB=8fus-1_qmb+W6os5&xgry!Y~_JEE!At zD$C%}iU-%fY0#RG1jlbAL*to5xcW8*p6*CvT2Cb}S1DUY8jrBvUKV52>2_tr9Lc!)7P;om2n+~TYrm*W_0!Xpr&t01en~tTy@7N@!2aktRAJ+4UW9FyBCIqovsJ>1X z^y{XxKdI0HnYQgj~%dxvx59`0* z0pr|r@;oW z{q>yfwwj?O?>RKF9_DPqCs?ti4K}dHx$@IX7}wbe8kTJkl-mZ|E!*KtVFz4a*9l5f z+aVyU8(QZ60PWar(7F8ud?$9o#_Vng{oD~}DcVk{NEP8hYi7nEg~XUv%KNhS=i=d%|+1ogm- zRoyJxp&YJExeQ}|lt3QydX|1?KIXI{&^0fDl0>EjoKgb*j+bD$M+rEE6~V8UB~X^Y zG=FKOVA+2K9_=DnW^;8J=zmX-jjmG%A0Vl z{toOlzXkUR%Lp7TWXse_I4{QX{LERePW=*fo#B z_5)?`PK9Sow`<@%R0Xd02z*-2zP9}j8mCsm@;|j~&RYWqN^e19BI}L53fcVSHcZsL z!*<*hu57pgAuB2&Iq?>BTi%2-A1TN#5yI{2TQK|4Er>6!gKfSKq2W{=EOw}a2Q%uR z^viu{RBM2|QO{wOQ9Z~=u)poWS_u5z2vzSNL4ki0R3$%#;urPs+vGmn4tIq+7Vd1f zdJsOdte%B^z7Ut^3VLUqVbXDXNO-~Y?GOOvem>BNZt&FQ2pq2Q1?lwx@O0oP(*`nK zyL%v9J01XjY?mYz9SHYXwmZrn0qbY}z{PvOBYRgEn&A!!d)y${iq&pqK7ji@;n1`L zP;qfT{LnfK-`*d9GI<+Vc*z4on;A0bUtCu-?%R{_4<71z!HhkwAj=rFEk6U`MXDR*2ROkNBVQ;z=M8K9 z*!^Z4gxiZ8pjguxGF^_sog5FCeA*SF>I32X5mu8$y2FqYDHUZn%TlD<{wkvV)&L_QL0^12D~pvFc^r;N(SDrtx6s zv&0SV&kKNtB}d?-jTaoB7z7tnPQcKfK<3#x21l#?VLS;0o2~$uzb*_ezYd4n(ZP@^ z7Xmw~LZPakh)kg;rS~HRM|1!k-Hc2|M%^3&dB^jF;Wy zo{B|N^DX)0ZO21A{!)TFKg*u8uUyRa*@bhi_ZD$c{W9ED(t#`fD`TA;n(ohZ+$DU(&+k5bO;YvaF};2~ZEe~{o0Ub4XG?0JnNb%1Z}&(#)E&!q*WH-;Bp>BM8}O=!A6HS= zM$Feca`WCErY9wqa65J;b64DpxWu@ToL-+lo_Wr+rVkivvNxEk@)}KRURrT|l4?}! z-GJawRwBx7{)nr4{^9$^Xs*UUh4`xkle_(Mx%@0k?(h>fKeqNp^XQp$wBHaMy%EYK z_Lk!b5=wSoU&YyL|Hnyf7=C+jH$|3NRJ*!!Np z{Np)<=nsM*x0rMhMeae=G5q=~n)^5R8vc85g1!kiCaZ(iaYsJ{keg%Q3)AK;MX{m( zsN=Xc)Kv@P8hf0vK1@syZ!wW`?LQ20Pa^SG)l=TgV>&lh&xo_>o5V>UTgdT3C!9EH z3QqofmQze*tj05C_-nQSn8YujFWdHV%~j($rKC1oqh5tAF-TwTokjM9$B~2yvc`YX zy->rVguLm^=jJu$qm*a@&3d(vdi%W~>%;p5RL7o6+^)jywz47{Jq&X~d<45c1pdG>%v?ncM$h0+-#l1B>(alHA)(+-0$D z+_O1?i~n>7zqq^O#1T8Wrs`6z?J39Q%e>`2^t=Y%m-Y4{KR`Zk1U*nSkEszT&yS;*LSi^N%a*&X6GOLI7IIVMT7-XYbYnr7KR2Yf zlzZGI#%c7Or3n!O^k_pneev@bdI+w7h}l1~U3Lo9oS;vID@3TJ)EW3=u@hH~driKD zEu-6CWZ>s_L8NhkBXzORpd+?PSFZjv1oX&TEUEGoxWZO!n0%F_uKtLDMVg$^b~p6? z#Lh}r9iUj z%q3#@)R&a3*+vFecam!>pH~Fb#zMf4cI+tXAS3rg@#{*K*v67Ork)x9g^4mQ?c zi}Fz_zeA+*=;dZ=ts2aU1&!cD=Pl(;!?NfJ|444q>?n>~HH~!HCy?JiCsW}*I}%s= z4?EpXal@_3#<33!Xz`I;V!cq4vl$LUv&GJI$dJ;I=eJ1tD?M7h`3${2rGdL~vyv-` z&*zk%u;*!_3RN;Mqo@8%3{`Qz~L3_t33~&>0F{doJ-e+~BGbftl zUJ)Z6?r+GC_*qmvc?Z=U9YI!#|@$jO);=6WT7ODeM9 zE|O;46}g5Mjqb(W$2Jr12j$FLtjYa-c7@n(b0BTAv&iXjL)ffb#eE#mq>C3;kW9m1 zjs<+sy>a@Sw!$gAl)4*h)V~tp*`NG*{tVVV>k%#%*@Bk_R4YQy)RDCdfd9Mv5WRS> z6K@zr5Vyc5{?GUdg3gqxy6LsgVnU*^Kly1`-4d3OrqUt-}o{a zb!vUTn{Rk2O@HqRq*)W^QW?8hG{`KOJ26p>e)clw>mh>l>OMkqiCB!v4&olkkL8yB zI!wQgZa^6&HLi8!L0p~`!*wU#CMTwaaz+W}Xy2_*`QWKkyZkN3w6Wx0)J6W^31zOV z>MJQ;c9oocWk6N?!}w#Fy=3yV3jQg}`!e)N=QRe6$s^CzxKDGK_T99=Ty+(a zrz1}n`ur0{%Vm=1d^ZtEPoQ6Sd?50rKX80Q1PyqmiZh-ss(dY}&WR}L@b@$rB2z?? zK0o%FeCa9^ejmyt)pnyQ8&o$^uc;j*+rEZ|lvdC|WfMBr$B1IMTID^%(Q@vD(Z}`?#aFHP&R{)V-#?e$_|Jfh*qh14&+B3tf#bN3v311AA(4w&(kf`09!U0I zozJQ3NaKn-j3L|+!6zSDr)>=}mHmFc+Yt14Z-U4sT~=t6musW@fDBa+)2!tdAXB#rSWh^NI2a8r08 z{4%_rM5Ml|xU8?h&r7P{Z*+{Lw(Ity3tdT;CVwSAE>7d8DGcxqiKEENHS@^DHPw96 zOjELHK`DQ(Dv6XkQ$i0 z-Q?Zm#pxK;<4A&SNz@-zQgIDj%#-ul$Rg%mj3D8&8xb zbdoajNTPNlmVB{zjN{e+@MByi;eTgE!T8T?;@WSI7Y{Trt@ds*96O#A7|cK$KQ@1- z{_x9m1+VDWj+GBANc7}#SZ*I8Y{^vzznkxH2{!V@T47gi)UGC)X65A8jqkkOluH;X zFhj@MZxQsaW_q~lSaS5#CO&Vi z2DvSEoV3r;rgyp{$tH~}{LDW!SU4(<=LPBLd*Kw>;-o=(?zsu#=25=K{2Kg@mLYS5 z$;5lcX(W+B#O_!nS#p0exnrnJU9^p`CeiG~d&JJ+K4vdmOe9YFbICK~V8UfF z`eDjix;QeIcc>drl_r^yAm4mIhjT1v$DI4Gv{GQQwbfAlyB;aot|-WlyFk8qZseCb z)$+lgTzThRqGXlD50-0`!gt*p1&%wE_&g#`evH^pV&!8A7e9ly84j$-y>Ockn9#+m zv`3+m%?#nRmU_N9={7&{^;Z&Qo``J???`EWf-vY*1mC!2JmWb|Bi}-1kzhr6(&cuS zAGp4e6h3uDx0wsTeMu;PNZyIp99qO1ebvG@TdZ*X<2_JVp9BXqLs0M1IkKmh3J*H; zVT$n$?EDr)xCtvbSsM;?w;he?&9W03n0-0BWT0dBjo1cO7h?C6Gk>dpo9)qCzRMa-%C*yjPqoZ4BTOkFFz{QK@{Hff(96@h7d{ zP7~9SZ}>@kKkrsM8*7U`;Rt>PQEqvNSARH@vAio@-+2foenylElI0AZ?&l}&TSt<< z^pM*rg&4VvacQz1vfN4w{ML7!bSwJ^4m}!1J#!%eG)29eMUdN(2ODcyK4;uB@5fy$Ce?ZpuO-j-|B&R4Ej7W>`WHyoB4y0@`i!i!iiD~0;@skgQW%md zih(CW$Zx+=GN!+TGqTynH+wE5e}1VETURI4(KQw>N{}YLb8BGIrh2;7BndV5J>d6^ zwj}}2CAhNmCj8NyOVDOMK1dbgGQ~TDRX^{8_by#-O}jnrPcSFq>gCvCCql&!wUA>g zI|Q*eB8mLDYBHkp71>~UfK*OiOFExEhvAx9n3CB~j%F<;UY={X6h$>|Nc9w{Sy3hI zoS{aB+NN2e+hDBVJ;upa*asyUNqusC~*TUNOB^FEkg7JvyO&Cq3~AE(h}#s$B+g~$JGLz6lkuHo5l z;_9>y<(sO>Q)>^t@$5E&?@Y*^!RI7HZWQQboFt2KLy5$UWhkz5kgSV}=Vh-keX;RG zqT9Y0MWv$n*y{5bB-_dv4NoTf4n^R*Qdz=0sQUwySBE&fdct2=Wk^)AjX-WbC6}#F5KZk4 zq}#NhHcJV=H+Pa%R+2>I>LR-1JtZqfjKNnT?U=qWAD=E!pc>m0xF}5rqLk+ZBfUHM zf_QUM-6ld7hmIy6%5$-J(@1L9S2Amgv*7RPH14y^J=9W=<;H$)7mjc+ zAxZpvE@@h@;FQ=efo;xBQhxL__a-uyyT99xJCReyvL`jLcU1`1O`lF@_8@(+busOI zYRJVuA3~jtBLxvkL^x_?2nGzP@_PGY1o;nYVZ~Af{^-Ye@?LBtRK1yibMzK-`~PK< z53wn@y7>t2|2ZB<>W<_JUp^HG<0=K~)Z&CSRrjzo^f7nlTmnh^rGa)fgx79-Mjqt9 zLvO#8*iiZ&UR}FI&Y09df`T+;thh+dB`l&_MwrsVK}k}&YbGflEX2@@DHU=mO9Y8C zFVXe{Rr11Hk5ibifsCCOPKp%sNcRqHuBCG$Mz=2KE{v$f0iSx|B1q#yW9;#m@pZOi zk0!Nk6R1e|?W@UalHsrK5wdOdN+Ky(uE2(VL{VSdD|pr%=ZK2FWj%=T~PQNA;~L#)pbm(r?mdX{EL+9e7kgmS@{=n-<<8 zbqBX_F23FZtICDsez63XxkZ;Q53=P{*WaKw)^6vPj<}5b^$$~1sgY#uj%!4-xr95p z;V@O0Slr4#K0f zaLNwao$!P|oVSxKzVN!@P}6H(#&rztY~D@Zw~is*QRApc z!aV$|rwKv*>D-eqax~@J6=K%&3{UoiVwR^LX)w*DW)pJhWOG$c=zWXG4Vo)#u5y!zx_N+>z8hHIvG0oJikBkETmky`&>u zlDN4Ocd;4PAl8Oha*R1ntgA;t=b#4lnW$j=y~+XpNhqVZuOT)1V@58nyN-OREO+z-SrYSa6i0%0zlO;xzaD zT_Xt)59TDpp3wHOpU`ou8g0xK=hD6?;{*GjxbD(_rzao}siJV0*?{Ty?e;RjA zeRTZ(m?~MZKGckB&;5--Cti_M zjj8m(EIHaeaXB}XH5~_wRj6hR(LcW%c=GMoxk^i>FlGau$s#-ahq+hwp zouBlXJ4DpMcAy8(+vF3FUP5ufOZuPkd8)7TgYG#VN|Rgy=+)M**lhWlv<43FrEE7_ zdy6NoVb{>Tng`wBGF*JF^I zHoxRo1HV99ihDlvkZ-@cl8byf6OXX05=)Uk@NTR;z9j*CZkPjV{R+iDt%q55TN2qA zzL^`Yh$XpNf3Z-d9-rf@i={YvrD4ouB3vT1cUEgN>JuQXrbDKzN=2EcN`i57Po{$>1ALPifC{jHk zhr1=Ejr0F|SP^FuhuS0?ZzfH_$bh-%!`G7mvpmv&VIe+#xP&|RxRzM12u8L0#i$i- zK(lIcVR`-<6p!D*S2Yi#T=XN%TD=3Sf}+T}S(EsUV;9q1;c)^Tu{tp6i^A?RwIHLv zmLid+FRzOUeoT5s%Ijb9B^Gu>OnNujmK;PAj+(ML(I1k3tcyWjogn^p{B@lQRPU1UuoUhSrTC*I-w z6C{o0M@&Zh&5~&BXi83+XF%O9JG8R0BlVZlaFJml&U+ukt(;THMZD6Y8qW>5!C&rF z^LrazUW}eGra)>?J&SVGh@9K8!ZkdNKbSho`S)BeSey zOo23+_C$&5-%_F9KdmJy%@gQMr2^vc#g`=Wx_JBgeiARMf&ZzD;|lHIr^_7-eX$_?LLdx9}BQ2r(#?rZ?UsS+z#cP%47|jSJ+o+mYJz#uLL$ zA8_RK+2qfzJf5vKNt?a`-3G^Dc>Pig-PR`j^W&A!I8+07UkfGMenjz$8t0Mx*UH4P zJe92KUL#z+z6id#8xps|P@z-iIOf+@!B!VVvd{KBF4UR8NsIz++q|=UvEqKL9w;Iy zIs5<0U5N5njZc%m|%Gmw4PKvsp%t@?sHw&y{xG zY^yv)KWW61)q+bC&(LMYy}a6cB-@l4gqI(;lbVmg1SWk0OI1^xS#t_Z-<}sdG`VIJ z`gsY0^Ut5`XiPG+iaGit2k;(A=C_-u-$Ee^6JRj54!8nZhL8 zT#&%;bFk%K=bA%GND3b&(NAi#wMoqlXRK;>BgX#~kdH6rXkN!^qEIA4M;S*Fo#I=3 z*S9YfB^eF;$I{Qj7y6^|K>sL`5Ox-Pr;Cx8W=%fg%o!4K;yX@WZ%beHmGa`^gkL7_ z2tU^z$K-%ke$8=ja^dtIJgd;iAG)SYy{`(1!rc{YUztWS_4km1?nk7l;{<6bW?HxQ z6ZpAe5ODcY;-FKD<)N#|=QT-W-mkaNTDy-PxOWZ~*ys_#I3u#3lR%eQCM11V68}!- z18=`=1^Jm*L=K=K&RJQD76%@X@Fo4ECbj_dc_SKAc34<>=nDy6TT7--_)GSj6rl!d zFOs*r@9@`m7Lkoo>O|@HV_~{vJWkzT2T%0!g{f*XMAe)V?u+pvS$SoA!sP4x@uUPY zGIlq4ptOc;JXeI*JJje}dn=MVxe&yQTS;cgDg3r^CLFY!K*Y17DxzaXkUOo(_@G~w z?p1n6KFs-m7gELPO8a7xY$?g#xkIVS_^S}on9Z`rPYFyD_K+%Z7d#dcf|f0oAlSKp z%td+dGf*W*#G{DD-cS-f-BBpAFN6%7Z<;cBU$*PQS(Ge{0E+fK)h}Hkb7@eLP*Tf|zDzV*KiSyzx#9UHyU~C9oZ4 z^-iYB2^Qp%za+J=3d3hvrnHmH#(JOU|C~x+`C{Zkh!DJuc}^5{{5Xs_PI&W z%;9q+`nWhr__~H$Vd=NRHI4sC#>dTqoRf9A3Z`nnK-UP^`0;?%!z3=GstxJG$OZk9f|yB%>C~^ zwoe^T4lb1;*T%RIlY)EX@iQSwoot4clXXyRU@GarR@GfgX**LeVP^)Vr5&!Uo$e0`_(kt2C@3^y(Rm6R=C@26Dx)KDvzvB4C z<=aV4vo@N=rxDTW5}fj}04IE}z~jH~^3StV=z`;;=|6d6njY0krrzlzzb7Y=pWW9; zLHSY|o^qRvd3=>CFh9d3j>_j2h8I)>`>Mld&s*@Tek?g}5l2!j4&g}$S$c2Adj5CY zMZS6XDppA^CkbxzxGBb(wEyrE0XH^|Zs?&{V|m3l_LK3G8Qzz@#5TtEk}*+k6M7K6?N zaWiNozD*g2%U1e8)8nj z$#h1rDL?6<3|{}a6P>0B2u|6BVo$T^$fck8&*=wA>&nCQV*Wd@WPRh_-kF@v32XkZ zK?vU1wUl1Pd+_StTYlAbSHa$Si%8prJaWBsCh3{fO$;-Oap?ymn*E5q9$p9_^2a!M z`-Wv6o^eK|a-bcRMqpr;hk2nYq(&>0-{YLe$JhQO6C+j;Z?%gsR#F@LRSuzVR0?0& z`Gp^Sl2N?xB1z zIkst+V0Oy@Y%+ffjw2qzkNpwkkiH0YOkLt|;xwJ@T7`4znI2&7%+ z-GX1|l=+Z>X(VsLV+AQZH`I0Iiz1WKEjLL?WIt%I9s~gA^QC+x+wZUi^+t{Q2Us3AkctEkC34Rc*P}{lw>FqQe97q zM<(F)1se3|HmqPws*BUf`wF*O;86SMtCjRT=AnJai8s=UN4km2*N7J`mY7F3-;^d_e)*8P&Pim!viY2<+7@!FB#6uuY$3ybkNK+q zL(zE$`c*(v*sdB&8uVMe05G{bn?jj1bBe zB_t%i$oie%U;W>C&Ux?Wy07bVb&kWD0V&w0w+W6NJIFO1s~3G#EWtag$aT*fsU>Cw3`7#ejME z+FS*ur(fp2eY%N`^Je3|ow;JoT31|ltOT%+aO|0lsLAvx?{aAZKTO_&%~`8ON%=3}R;i4r zqkRLcyQ9q(4knNvO%*g913|2)QxL^F&i_$uWS%|DQNIXHHG4xC-qTbm83*3lhc)H1wlDGZk=UwjP^cKqVD~4BcM=ej|Bfq`k;wA6I zjpswiDCR%7x9t>K3fc7NfbF1vR0YO0!v1$qnG0REJWgpU<=7%4?0FmZLCE3%(0SM) zZs+gk5c?#AAO1y=^cDU={K^Qv@cK@a+p!w6PxxR(Mu6koFa`enyTN?^)l~Y^R?Tni zo=)TKf5DfIGq6kU4p*$Z1&8S9!9MFJXf@`u=vBj2fyug%Lbk~9cSMO$KW-S#;bh4C ztSPhe$%41%m$GH`qxc=hC0yfI4pxVlVxaGKE=oQDvQNK(I3I-VbEj~dI?jT)Gsf@#57t$YH(g!kb?{cVJodcX{H~)HjqTnCf&%S%M^IHd{L+uF>4AQQ_j+phL z=AMJRc$EqF?%oJ6Y7XX%X6VCn+Y-F^bvE2tdluIJP^OzAWp;F;24C)?!_L~>#9!OJ zLESS0op%Oe?~Fj+d1^LNe+~L8bQ0H^Oplf&C6Y#heQ!P)N`q zw*AsN<~lJDHLpw4vzpO-@ybrL=vvDT3CyXwahn`9e?5dewG{rHdp$R&!W+F#hl8Xi znV+R@$6|gt!+i-NxVC ze|!70q8U?haKUf9G5P{rH@Qt3&HDIvyd91y7Faot4uD*bw4mdUpt3M4Y^W&%o1}WI z?lXr8>vAhyt(!1gxC3-e+zkd=PH3!qLbONFz1ByK1L{31k{)5fiM1v$f6rl+>G183@S zBR0f>R^9|`JvJPEYpda#_0vT23wPtG%BR>eCL0bOolkFC7UE&ggLqtK5x;I`0b2h} zrR4lDyuBzz^l!xlIBD{RbGiA7i%^=vPRfr*&z!p$`C>Ga&7RIY)^k)kb~4!vThI3L z9_;(MeiX3ITzolKi4DE@0G8;t!d*~>p`Vo5?v5WA+UZZ#*ndR>K{y99Hr05lS1|(1!L%P}<#s1+z9W%e4bBZcslicz3S2q;56sGDTj} z^${X{71`ML4=OziT*2Eqliy$Sm!mFKd@k$(3ruD4Px5WZUt-AYcmLpq-EM@b9}OT@ z>lN2)bPJwFYO#^mrs75WA(-*Th&ta-#mKlss1Z6MQ)*4g$mTLm``p6|X?dt#aalO1 z2RSx+IIm#RpVYdD8}@uG?`$*zRRZ*Q-MEQ#;i(!;uCbo(6d2dNV(dJB-iH zrt*z5rh?-=Z?VdrH+LdJd{h)o2+wR82%=^LLTe^?C zE!ly?Ds`c&YZliR=qY$l;_>X{ZY*fwaI@BYFo{C+xE}@Olg;=|^8HwPyaMYt^bmjS zlN)Ticv!TzT?wBge#5^D3L)FehgJj(q%($M(V&SJLDW zFS#OoqwUYVn?H^9FOws_WHafnk-Xi$B(y`#>w^4px!nDj$QkO(=tBb z_P#YN<8~43%T7jvq)|9=_b#sBT@JjwW5Q&!=HRHA<)QrIofac>&gdcNdWg!G{K0~DAR2e>1J0ixb*ceug6o4Bnjn?#$R zXz-R}TJTJ0B9}hMh~?<1gS__{@^hDxoP808mf?o@#=V1<^=i^-kr&&rxd8V~3WMy& zZsazl4+hk3qA5+w(0)KT^lz49x@PHYq9~L)k|OBQ;z7t&4TKjWp&Ks2Lagon9~4g3-GB!eeO5udF=;>ha~pXjcNO7GdXoIa5Vk5VRy1tc zV|-Rtgg3&6@h#@X*im5wSC9PTPpu5D9PEBts3iFW6KpSVqT)a<(1mk~-%|x1i#)00 zwjn&s)FIsrHxBK9% z?^#Gbvjmqan6afni|9k@47Moc8^3tM3H-aP4qpZu!id*3uv@AN@8!wziM^rXq}LZQ zxl~T@p9n{5jUQ9%`OJz0y=cky6y_LxfKA(fh_PE2v477fs`8EFJ7#v0&toxNSE!`) z=496WsE9?`KF6HzdMwx>5yCT5S+CfVrCaGSkx7Sm{2~Ex9Ol5()wkj0mq+ORFoC-s zSV>bFTDbTzo@BemiWxZ`X2lB@vid$nR(I|$9KIaM&Ryns*Q^knlX(nt;_av@)|%LN zj-G!X1&;G&>7m39H53{kd2b)5|0Wl+=1gG`=*Mlp+m9I+BtX@u+Elg)O@}a|w5Jq6@l& z)x&CQdA9gMA++zBrq4a zvGb)iGAH{aic~P;Ivb|4`MWQoMt?^LT;dBk2N;#(8+&9FD`3P`t} zT2hehm5rtO!;u{)pB zD;N7z;HfREMVUtjv26(it%7gia?%g}RDB4YQC?vH%N6~2kaPrjaOK%j-wuq z27Uj9q;6*b6IRb)X-5a*J2p%*^_&HT!dIBtF9q(V3ckByzEZ*dC0`=-8>cH((E{}b z++h0_R@h!c(5?S-! zJ!~_-k3IdgiCym2geRM>Iyz5}#kmUyki(E{9JSJqBKiuc#L}IOzGym6l!bf;=%^@qqw0)R9)VOw`6sw+*Rm3^Z1IP_j$1P%OaD# zBgjW@BwX*zLo+jdlpmfCrmsr5H*cO`aesZ9d}bxh+n0{f8=-GJ5guhJ1kx(MNglf z<8?nLv9AI4G@{Ipp2SpRPyS|TKdVL8%EL&#*MizVq(Wz#A{9l+(-Q-4kyo%bX`2*M zvH488SaSvD%uNs}Y^vkp>ptP=<@H$Op2wYETMH8xiKxTyHb(tTraDm!-VQ&Hy%VBo z$H@IC8ux_EE4umO=RN$X{6c>1o>a7|^59(OKEYoWKHPYw3k^S-DAiGen`CE7N_MGW z?BYnWyz{xH<77Tty` zi#IChk!OK4eZ9CAPPi9h;fO*Eve%+>6}Qo*^*2^&u0q%PD*nWy-+YJj8h+Y}1L*nS z1sW$0WV;@#QP8Cm`YHOq}}O&r1n>x#B+W# zuKX)2si{sOl~x1lXy1z_XHH||ixE^eBa(cia~&pLm%vi{6}aCo3`?vY(TW2KR2KV2 z{H3jz70V8gT>AEy+DDg(xb2n{_-qrty({E*_&iZzXbhk8Hw}HB45wFb2l1)}rzq)= z8p#@`QqEcuRC@+fW_XK`=UO0S6oSM7X0Ne(_<6cDL|gK4 zpBLR0cu8WN;iLhV>U4y$FRQ=PJ%7_&v_f z>)^k7?Bc>_ufeoAgJ8=GH~y6LbbiU}btL}mhzk~+f>@_|zWml*oH(SMQx3R5U3xh@ zrDe*!ty42PiZW^|X0#u|SW-so?t_@2xN`kH;3JkDL_Q^PciD+~DLM+cx|n=RiEs z7}b|aCbdL6lTOhdpV#Q!W9iosTe>-UGPkDP6MOzmqycFySfYFnmsih) zBR@p6;9UaKjSGT3cMp+c{&C2Nt>hX@CeZNKL}sM*SL~s<3~hR~>6t|`Cs)0SrhKud zNkPrzF)xMAEAAnKciouiHy;;B6uG~STlujWiYS{p5pp`^m?alX)~XHo?V2BLT`?YA z=8vN%_6r2|-XruDytX3@?$SfOZp@QAK*~A(sKn{h_95XoO)rVoowudu*?B0IUJ4o| z`^85Bd$8-W0aX#d`TZqBXAfpR zlXbX@Cu&h`d=w;okulcLqg&^ewfOOa()+uv9FEJSYDCg681dehAqjctf{R)51$NrADKmIn;OXEo(qoXmx7WB zqiBI}UrAg!kX`lhz`7Na`7JXu@Ysrc`qDU-ra#?7X|F7)NqZl;{Lo;_?N;C#*W)NY zm4ZFfF2Wc26a2(hdonYbjkljh<9GA@pl%vV6ZWnkvaVkJ1s@dl?y3#SrPs2ID%(i*P-u6f9f<{LR;({NTJ=5jt1wDP3CJpUQ>zfUgs>) zaJos2TNabIN-F)Du#Eolv4U6Q9`#G#inh=DiH$mM@k@RhfS#7wEUHl72I!%JT7yU5WAc~bfIP92_ZA#V} zx#)VQ0PRe5X@cH2(A@tRFAY|LfC1VK#uX_F?Gix zGFjlk#vJQ*ymoXo`F^a!duLr}@X(LA&d`iRkB(Aq`YQ~V|HdC)>mYI2H5z{whSMu= zE6LDXBPr{Y8%@^nlg!tzrYkE4)7MyYK7H~Lphdwn*rlA_W?rKAPf}?=ZzYk_JVP2P z5BWVMz%HLcJgXVTKZ^cBAx<8wK~GV#<0Yc!6o0yS@gvfbW~w?>Oe6Fqbi4Kx)g*>f zhFmqenh%lWb!5^GA@^jYCV2DKMBtE|)2N)&Ox=RoVonU`E}*rqf)<29a$Lly1QOOO|GDg5+TM)FKD#vsTO^uK|;t zQ>oqP7!xhc=iJi2LeSO*^wjyp|LnMdb8n1;_$C>rji(hP-Bxp9vtKiwNt0uZ6K1mK z^a{09Bf;E3g?(?h0xOGGGNoNZocMl9EOPE(q0@dUi~6#dhLpIH)X@M^|C7u+_*?|( zc{lL!(MPDZX*4*@F$R@>o$Oe*CQkKV3;pDpajN|}-r4o5c%SP_GD@4rBK7O}BN~f< z-_Z@8n_j}qK`+6j(jTpU1;e-SGpsM`B@B5dM>1x%5Gl75T7~R#<82q><|%?*y&YLs zoP&4rJ`CR+2lwbg`eU?>o>ZJ-8*kskfmT^m9jb*hd^=(Ao9RrqYb>+*BFp#uYR8RN z8o^KS%ui~~VJZvVnUir2pSXA!^$pqz%~owFe`qSWJ&J}=3r=F=jDxINTEadZQ=_SU zTc|7J2yb>>mFlT%T~0t$qB41#DXk*9$&#+($bXAL zTH7J2`mz}FA{E#gmdU=Drh)0X8H@|eVnxXluymIyYk&M72Aq~vcZ?AR1Z~fI^%+s^HOGzo$?a{)uxaZ>CBs10+cbPL;a2h|D9fk05 zBk;blQDtyP8-7?`g-TiRXq|hATb*x8e@FW8ZaoR&T?ySd^lyLMCUl%x_!mLYCM_zx zx0p)^@`h3G%DESbnOu;}eeQGJb#eQ!b5I^8aMp@;VehO+T=sT9G(DV#B@c(=BDrgL zsq8FH9vvadv$w$S$G=xzUEMAEaOMgB{cZ+1f|0T79OURc4x6zF_xIPxkg}eIf zC2BZl$N5UL^55dNst*2}iYEjg(G>gq^28w9-<5@>&iE#~ReVw4Nne;UAGSqZgsidc zxcO^}&=;@_vO8nIEb=5~&O6Inwn+1r0$1{3C&yxcA#nCH`ZhmC*$21BTEo(eNUp=~ zI6rIFRZ+#+cy_RQJj#5D=SE~SfYP3R+=%2vZVpYvxtHoic9H9 z;nomN{e=l?xg3S+#uV7?l!V2hE>I|=1`(sxv0+0@Wvk95XkA^&;qbL6y=fV2qDA;_ zwi_xs)pK@IVOU-`lS>u-0+nrPSo0_i)<*fW5!2n_d7mTLIOK8Dh3*xL*>i+$!h?|3 zqJ@8#SaSjW&%^!>DI8)NhSFWG+>QRkbr(nRo9~CgNtvhOgGrXm zbb~ZLxw{>aX7ab!Agrsb;*(6SaEliW=1U$Q;h9z#N}X#E{FQPZ*TBV|HWf@ zi~@87+Jo#sQ|zjmf{|m@=;FdMEPNCITV?C`=E0k&WX~a(dq=TSyltT4tg0Q@q4@wc zMqc5(&JO@_*liqbDKNp4=CH!*>EZx|z5LW4UihK7Q`lt(^4i)HIoV88h>1?)#3hxy z@?Tr#RxpwO(t4a55zgunHFts zhm(`5`Qw}y^qy(PQ2iC~q2mf)DC|kSGoSO{;*H^s+E0GxZ8N?+&JFtQPrw!5A2|G$ zj>Ip!hoQ8RBJbLH5>A$<^Nac?^RI{(&So#~d@J11tk_q4zxNKd{h0%I7wK|Zx$gXj zWr1*h#&hvvZ&fyL>~Hbz)N$bX6^m9V&G5^t-girF!G z7%U|)a0V=gUkO51&ff&T2<-pm;!=Ln+WQ!c3DBnDjSc74<6TK2CSI!J#w(q`D8C+P zF6cqKsVUsN(FPDX>lg&r9~P_bn#DJS4}y!$vZOh7B`y?FH=Dc!X0t_i<;=q?_#w_q z@pHuN@ny4}9f(MZPI+VM$!junBNZ>X^wS!rJAO18mX5F1{c_Y_J zF#3BqKWShtin3<#C0eV%4U>ix%%u`7)VUicm`X@kd%?LU^-~zaqtFnJ~Q{boI?a$is7Ty{Z;pNse zSS&M%>o;{Tg0T0tn(~zY!&Qow%bVegCN;P@IuEi^w)4OLixRJVFbiG?oxVTw-Pj%7 zv8tn)&O-hvVMZeSGetf$I84JHbNTs27ouUeEQrU z;-X=P#i9B6AhW#?+*d}68v7rDTgUoDt(ziX$8LKz=Y}r3G29x%Tg-*5mmQ1DiAD8S z2{6w^S`_AU-?6#(38cUyp?|?19%Sgl$pUEZz4M$eT+MJ zNs+(nF%*BOh`C^{1izZ6Xr2MnupKG} z?_d}t@Wz+8u7Qj%@~GpppF3nZ6jOiYinDPpds3*z`nDGFZrA2>hs+YWU*2L!sV*0< zZ}fwK)r;^)TMFFX?1V{g)-%TeYUKGWju`~#;n$ZM81F7X)F$kPCGx*3?_Qk-LAIf& zw8j}0YZ@~vsIirZWg?rI2^8M z&w;O#v>|Be0bF)IkB<`zysDA$+}r^INA!UvUnxgWaa#)ukKW@}$;E=az}4%n6112s z298br@TE>4cwS4V6E?dkL+~|na#KW^&6&6(?=sZK?qaeB>_NqLC`+e$6i;15gQe0a zFZmbTXj+L~ny=w_%`E(?F`Z6n4~AC*g0MQ|1aq97M0G=dVuGq7*U>(N-umvs_Rsfu zS>?~X&VHQgXYaT78+1Es0r zzAVg|w;$VbG`RB_C!sjF0C#CBlS3ddop~qe*y>62YuZ}SHn2eNEka*}VFajWr&H3n z7Whyeirtjo9_7hv?_3i&CMR z$$#TeDpz{MjD0gePB$Nq#md zlOhiXQ?4|^m{q-ezSB0|Rp7{s*{M#$?0)icmWyff8ltWA9^Ssvclua0kE)`oIr^>4 z_S}x3TVFmv!DTaw=B{B~h&cpB%ka|IO_=xlN$k|>^X&0`2iADCjQ7%M<9>vjGMg#a z**fJYT(>xgl!MP>XSg1#(N07Wq|v<9ogy=LUB>?>Pk)4&plEs`+}d&l^q+S^#IQ*$ z@XvKRr=QN|q$ER{jEmEUjs5tm-U{@5!dy7&^OYQ}p5vdVBHI7gff^z!aqFfcvMc|L zE~dZm`Q&(TQ%U2urN;BQq(S+D-nY^{pE~zWp`Gid&>!HXv5`2FDkT41`DejNLQ&g*XRy=NlvpV$xd zn-@}}ArA&8pYki@21wF6r_yv-}VBt^c>muS;CeeAQCMvh-CNY28J&_tFt$ObU2zg(Z&cW>T0-WTd z!ujtR!d9i|5(%Cn-$j!_B-4$)sYbM*cqm#)Ug3(im3T2qiw&8Z&AGd4P?F^l80(Tr znTa_xFFTrcZqh^{bcu=QZc|-GDNOp6PbuL!)bdM=rw1;lxhIdX6@zT3#@d;^n3G85 z>=i9D4W(`k!K-z4J#Np6BDkE!y$bQbjL)|8Xa6MjCOeK2%D=(-3R(JmGY`Y~?|k3a zOms0iO{vpG6ls)8yZ-*gUs_e%fiX6c^OrTbXXE2Yz55`2pFNLC>I7zujDqC!+(3HZ zSxUFfXVc$cfs;lNbjZ_=I;R&vZru#JF?TLzzJG&hxy`tHb_Z2IuqCU}iZpgZ8r@x? zM*h%+M+1gpbmee*In@hPK9|vF?lsggOI+1+5S}HS2K`Isc)V*2Rf)Hdr<*ZN^s2=+qZmuW}!ZF3;xW*5C#k&BmA%PH0H4oT0tAL06RVb9&PhjhmWGVdKxWcv3pp5MT8 z+Z-lA(6oLeSuu#NYwn_xFPzC~(1}-1FtI#AI2$ua?0S`24kVHL z={9OjJ51lV-{a%;~Er+Em5IO34OUQ@D44zX>R@%GRinj|KeiFXNd{j|8+rRSl_~#Nrb(y0}n&( zrm>M{9>Vlhf==w7E%1Nkn2EXxN(p!DGr^7A-php$a^p2`_N-Ru{tJMBohz{6g$*0e zPliNSRhaTa!XJ9K8mx_5L`yBciw1n};Y{0#z#}sX$Dem%Tcn>OcX<@FZj|L_j|s-6 zisRUCRTOTuuL7GJ2cY7E9hRpx>>@V0`O&rRZiIm^UT!A4NS__Q8pJQg@oaxlRW+*$-svn^RD;I~HCx zw(?h%idf6fEVeH`9fP0danJhdp`GSH@1M=IR7sf=?;!4DTs+^uv`5@@`!1LG!kK%V z9R-2B8~o|5U@;ebn7l#^6P-A~9H;xjTb_sIiJy7z&y!$PnK^v&C<4idm)wyc2_Gs` zQQ4gv2W~}!p);fo`p>my$$96wnB^zAitUf!``+zLC&Crad8=?HzMio0<{i+z7Q_yZ znXyP(y!k$t}`1$oylviAN-@J*!{4hvn5rO$g%>5&BHu9o9^#w=#% z4=aGKy*ym+E`+qHSx_vQBeL6E1`e|gG4t_nZrF+AAit!Z*+$f{76lc~V}TA>EI9## zU5BtKmL{yXY9Et-6vB?*@Pz<1MfSFDJHPjLEjKHvPMjC_A6zY0QMfhgWjE#h87nHbjy+XIA@C`TF_(}XVe>N0f6S8LmKfzS)9rrh| z5f&62f+b<;@OR%lmTPYVf8JEXf-#<4B872tzU~99^guSk^)~w(rs$;QU5qR2YuTOR zS?sCm9kft81y%1nq0VtGD4m~)e&40}_RJjq(Z*Qt%`Ou)_RVK$2}dC}ZZ=!IMVGVj ze#=EG{Bs;qd= zeT0VrZQPKt8IIePCg3^w$E?60mdTp#2h|u?=(l+$grxZjOoO2)ZMz+o%(*POIA9SA z8GQ*nV+9RAZX_!rLhRas#Fbgnl<9$NQmG4}ZaH*IDDaI(?R5xiH> zNdyLqRuv45F@&Ak7TlB5=VABcNNDbjgVqPa_q}2)20U^QOU*qcHV<75`LRE_&~fv) zk7Y&7WrQ5vx^@oIrj)|Kh}E2Sp#k%*R%M?%Guf7iJZAUBmqqW&<1NPjfqg=EY|p0{ znD8o?&);%@U!7ydr#8P#&^ z)S*#q#0w$!a(O=#?f2u1Yj47=dTr4lvd3qGn}ogEavk!p)+vDWZp?P89sIviIyyQgKY6o zQj%YdVPjpmYt6r*5&m#KO@Fz4Q{wn(A*vYdav78Ue1opXmAwBXC;stG8&=WJiEY?! z02iL7ki+s1nEK@wu9|rllDn>A>7$kWGUGW=yCIjmd1OBSMNtP=TK^V(($pjuwJbQj z47fet7h~fg4!6Ji!C9OZ;kuyF?Dw$<&`mIB0S;I3vDaVGmNhDZK0KY4o!taao=u18 zd7dyi!9AJzBVRvH7YnxxVB>?5P~R*ZJ|7Tzgi>^c-t&CWNLOYrAMS+9T5@pw<|&v}{DEJ! zT8&m3B5xji2g5E2vzAK~UK0%v|2&)u$~To^?5YJA)2_{)Ti?fb3oCH+xkvCZQjct1 znqgqyU%c~@!@~zkq2Z}8_ca+)jl~E$HR~gcJrT+>xo8&iFB<1q)!<=sdqUOymATuO z;b-N6_|q^OPQBD&_B-d|N1ZTy*(PGu)lU5G)hBS?@SFUIgkki=XAG9yIK(~9OGa-p z|#8?D)D7g2UHaJ@EO96uGFv)RIPNu?VvR4q@{F2Q;XmBLFhL+JcG}Utb|SVQ+Vx} zWtD3xq@n6m0QNVur-iOJaJXM5>inF_-Hi$2f1Iy^=jKC%XzD%O^tKs~3cXm_R@`Ag3T$>IyR$hj;?WAKxUh1(l`%!!re*SQP$@Lu=GcSv+uh*f2wjyvjb`hiB$8eJt zsk6IlyJ5dKM@Jk>xdnm_UF7r?+l3C4{ul2#$Q@rzIy)n{cgF^jUh81gDKn%RBNJiQ zTui=qe?U+9I6Bj&<0#`+DX{#4L{UG4JI#OcY}P}2^le{=FU}}KdD1GOdn^(2<)+cD z%cgX8Kg0iasiQ()52v5|L3B*h2bXU+269P4PZpmA+eVmCQj0Eq@UA08%}I28%UgKQ%k~7DvF2Sr(NE}Y z=rRd}vPX|_t$w>myK)b@VlMY<#w!dv+y=AUK4P|x2YS3;Dw=lxB_|o_$BsAbz>In$ zGJa;xtfzQ^ByK;-WLr^a-AYy#@{C_1y_|wKUB^#L4zl&XlDNe=9_*oq6jO7wr70;3 z;Z$fUjOHz18Jg42pD%G@N;tG`{S4!eHFI_bGeq}!CE~v&pyfFQHl^{opk0^4ow0_F z#%15Q<5l;;;npkx)Tqv7pK9hDxjzta{Ta5k-T|?b1L;jML}~sT*k6oc>O1a3M4B$h zo)Gjm!T**Y6pM5A&u5BX)Y+liIrOkGRGeaz1-{p$*}bi4s5UQ{@8_b#npIaA6h^APg1=tM=yP;|dw$kv`a3I}3e!n4i8aNSg)yFp8q4R-h9 zWDcvbi;?>5TzIbd>A>sk2W7xI*LVDb(iLo+;z15xA7Q>S4_JJo3%v0@=B zdI)>=kWo|EZAqoji~SC(a}H40Tp7uiS6i@OQ;4t|c!qq_7&@;U$94_$q?4wd+^5s6 z?CzC`@NvmkF7mh!uHK-@lrlE5ZZ$RH$0xFbxP>)Dy6~5OXyUo{0zCc0fNhdlOKoSR zv8_;!CH#KLrAkNgL8|ki?~FQeGncZl83NC8t`uGWH<(nSR*7zmAA?(Fh)DH%e-=>j znR|6fc-LQt;)}b9WPeDV$zBhJrCXQ4%GSBm!+8UsqOn%RL|el(0d*a>sf zFI8r#hw+0#^hN8pjH2X>gYcTIq8(a;$!<_8lR3SJKFu7>23{IU&z6PZzj2Dt^id1u zkIaWhXB!=Rri|wxHlFJ2*U-&{QmEUkKw)3=u(Ue@2M#$5=Pk6@q15fHTX_QKqI$vbVaVA}ub(7b3@HmklEdZ!KLHrS-Ij%~IyPR0<|Ua#e5wmiWRJ-4{W zn(_F`&yTkM5cngzweg^K3_Bkf#r$3!7yn%O3b$N{qDNa@*z-X%S?iz0u%N{dOh(ND z^8r)oL+}`yK1+s~Xv9GJ#&U>%@C1i_9?q)vi#elrfQR0jW4-98C~VPSW^KHS4U~$8 zPY>kj+CXC#VR@DN)X#uTJ|4g##OuLpy#wp~Jp`(<{&4}!p1z&VgUk0?xht)`;5tbH zp?)di+(}h%;MZw(s^}byT+qnfJvq>+`F<=5x#kRJTB}(6Y9I1=Fq+v!AA?ID%+R4T zjD2%`1T?lEi+dM=$wB_KsZEZVrQ7lT8w{9U%xU0r$Fb$g|D)(U{Hc83IBsNw%xtA1 zAtAzf?pw;YNM$4ym8LYzD3nb$4V96MM9NAUoaeeuB5hPsLXxB*?Jbqx^ZNswm)9B3 zecjjR^Lc;#!RMhHBayKkd?Qt9t3?W#cc7Nu9KKECcFEIIgJ2gKbgj; ztz=iaD4ni20p@3i(5_%{dfMd_jaOPgQ#;O*h7<#SccLzuem19b&t}qK2cxs|2)oYYydsT|Kenk@~_ZedHKZ#Jk4X?ZrE0R%@- z(Jc9M^{FdvNzNLgCYjd_j)XGbxEX6;dklXPkGc^<0X96Ba+fT^{rlDNkxCrVd7r>;TxdmW%H|<$?}febxkP^b zL2xMFL46H^=@45XZHwdufCB^dVikyOu7ZSdO^fz$u|0AAIA&+ zAOSmn2{X-88!+{mH3(7>JEyx~A-xTurjzTE-FH%a81BFu>j zXSi#)2C$UW<1k76vZlk7NMv7&qLhe z5S%^Fo0a1CLQZxgX@%e9ou3JHxvUOOb5_zM?=eO;@&LVhI}0?1I4)5@ED0!gAgj(g z5)~Iwsx>hQid`@85_lFwGQ*NMUb82~S4`;s6f2@$p1^;$-kDxo<3!T~)G>IC8pK82 zB;)O5alUQ}*(!O0C^zZTRdK;IGGIUXIi3Ma`OVDif)^yxWP}kv@Qb)Fn@yvIJ}|mt z+RPCVf0{W@hCa>am`T}Y;Hm43XTEKtGF-O1;#4^8YL3A-r}tr#6-!PYeMY1`btrrN zG#po%O(PXPkeYMBP-%3K1OuN)ozfv%qS5fnK#CfD8jq@#RhSw_c(dP`(eIbdATHr2 z9Pn!7J+WL&bq<7(U)PiwW4Ib3(!lZ+*H6Vva`bhlF`v<6s$$xR6yy41+CHK=3`C&-YqJ*klTItIS~5g{V) zr(^WA1K{bYOdUcp$)3kxM*hDo`OM;HvgGjaQaD%9%BaP}!t)cGh@<## zaFL&`e!SeKKLn>L(befo7ZG)}`Y2v!8j;(ZvBteT7(pk;} zM7?i-M88SE?{i(rr*k|kQUAg0`L+!CzpdbY3f!4d~p_3iw_5fS5J^ByiFJZ%sYQcK(Xz`7^!rix;QenkceI zZ!91q5q3~mEyOc1YXyVqQTTW!pIS~UBw^Fs$b@;ZIAGcWXrRqY zl3LW}UL3QEd+xmbwuPv6o0D9bD0YEVGi+{|N5{C|zoASqj*0tlUJ5^^B6R|0a{5Nx z17YZz;7%OO?|^?)7l`#;BuB1~!_11;yzj2 z*sP#`gKWs4`8fC&PzTpS{_vyIv(WIvRT7tenX#_XX1kX<(;mky^sc=mb&Ql}Wry7B z?)WQ!v#A=_{h&dvco&n&UuVF`jlbajbR71%i%}_C7trPIJ{dQ6(PxiEX{XaRdeiDD z2?>fMo1#rH{P3l%h{qL=${^eO%DXWjW-+if< zUayRE%TJ0EL5Xdm3v!&yKz=o=t;JIg(3dV{mA%IlTI% zOIwY5;Ozct-0~v{gW7Jut;@csazX_)7o8_}IiLCWPrc|Z%k@xl96Y!8`7mYw9ynVw z6Fh>#uz+KKCUov&Eo%2MGq~=#U+&@PI4prl=N9AKVt<%2$q`q&$)cC60ofN+#m1p z!lWZI5tJ57VC0>2{5>-QvY%B!-sNieGxHSxM^+9-c3cC;{xWu7as>)4_zS$+10V0+NstUDW zt$UNuz{D07+;_l@vPR&$rk-cr z`b!VW$J<<#{Ozi2|2liq(*lwqQx{r9;Tz0d0UN$T@O@^|T)BM?*7O=4*jp({uffstY zkg; ziz?$D)YG~0Xq;IE4+8f?TU1NkA3qU1)t(KKvOF#a8jR{M_JQ3&Z!&XkKiN5O3!>Xp zvAOv_^jI^K9I87Em!|UJ+x*{b)HEq=F}kJbO?IZ7&u9e38Jn`}^3uogQ%f*JRjismF8DXole1 z74UhP2Kbq#k;eXU?CP#vL`&r`q_8=lYqgR*-k|_nKc~W$(+%vIW0|BhPm>rrjWOQB zqHz8BCcb*yDR}Ll4UK)FH(gp1I~zs-LB z%<+ZFE<&`sB0KGP9@-t&hw(FQ=(EZJ5@OLy?tJU!P5yh0y`6uGT^GI{>T`QRQ)y{k zb?1Hl=!_m1jBf&3_Y^iRr5GCVky+L9ja@Hv3nn!5L+t@6M#(`9)utP>dOG>c(kf-V zFwGJNbhL1+R!XV3DrIex0L$&us6p%y~!bNiPJGQFmDJcoQi5 z&BRjiRiO35k$ukXRwL|`=>;QYLVg88nQIICZ__9n@0JKE!wdLVV|3u&R0^5ea*U?N zZCKuL3y!umLch;4c3aqH6lq_9E*rW4;@_|}n&$AVgb!&YXTklh6>~V+h6FV%K-U#+ z;C8={wc+vu2PYq4?w$AwM<&*Sr0+y<4L=WWx5vZN<+bcJUJ|?WDyQ@8{tFjBmvh-2 z8H{LJL9E91!%4#_cwcFV5&p<^Z`53ZK#*Z)dG#}sJ>)UhdK-BuD8LfUS$OVd0pEDK zGzxWDVWH_qIC647h&+%5gLiFU{v#ChIZxF`4SUcpOk%nl+Al4&*9CUhHi!&;00Y0f z*^Gufa^hw^TQXG3^aTe(jIA=`yDNg1wbT$rd#kHG|N-O^lNM5Toxp z2J?=|61hpM$;;+eo{nrAG*-%yRev|Z#Nx|DwFsHj6W(y$;_KL(f#s~^MoF3`oySI( z-hpYlGZ}+2fRP>(ydxTqq0*eMY;ptjUm!s>T4y6;UP=D07luV4EaSb;AI7)+;;OG3 z$eG*o86T7fgCJEVKUGWrMs+ov@EPHNrO;+H}4zcL<2UTrx`O!p(6hr zNQsBBN*$bsLN^m?k5WS=MXqWmy*B38bs>(S+ZO{lw`dWMvp#S>Zol=$N!rI zyE0w z`@aHq{_;9@;}uzqJt0P@ryASgGK(~nXi@cEXUzY!l75|01PPlD!YA!QCOlLG)46B% zC4*Lu_j!QNUDx2=H;%F6eTB6S*CM|k23y)}Sf@Dsc|C~@qR+V%x zB8C&m3WZz5y-kfo4{`?yC^)@&f@nysE6g-3ke+Cji*;fVQsA=JbZf@LX*bx zL&cry6AtHr3D>20LggV8#3Yc0wK?!NbS7T#4`sKQl`?I?3n2j&@cqs^Q2EP%edSX^ zHdUmub5<0QbDFs@W$|ayx1tZViob!AZ#E>#=flMlbs)hpOgs4Dcq`DI2+PTWR%IV~ z5O)XKZs)QKm12lcP99O3Rz{OvS%c}?GPxi19xoHPVSbL? z$>odg4HeOt{v%XI%Liu739f%TupHOJc4pH|XO1Pj7ap5W#nSDn^37>>s`PyY$kn@+zorvr_#K($+YyxY8|a|N#(Vqh}uUHgT}3n`_YO8xM0vN`j#$dp!2eL+v@JCm+Wdmy4i%fk!EH-%Ra zm-K_D9=xBVh#Vx93LFz9@F4YGt7#n`;6^bclipBOLu;9dsH;B(<6j7o7d2^c?EExh z{AwpX^5!CwvnC5dR3^~hK0;{OzL2Iw0IgpmZe3KLKrEGy(@{?A+^Uj^-}G-e-WTPUCtGCSb|QaoDp=9G}Lw&?s&vwD@%*EZnQX+$}f(wb8%1 zEdLiqiB(0@$})aZNIR@Kcakl2*hd%MPa$1>(r|9af7G=241|@r(LG6GB)xbw)rk?d zHuS7v=3hTedIULi!n+kjvx_0y2*()T_8)HMx?!4^d?Vw{7t^xD9GY_GH8|AF;FG~I z-rH~AsP^9$tbTWe%vf52%C8SC!86)_i8&%*`11U{9$N+oI_hJuP~z7 zN;owkfxJDVj+=dqaYXDH^=>>yXS$rjzo{`?W`beFxXx$Ko)&8IbOoka4bfK<4ij_d zDa=ZFP3st&pG^N6E%wpYd7KFwsDpZW{p2`OX%J*0bf)7 zAgF!Fq9IehGP~;I=zvQKNx2q9cP86W%4Gl!>-xib)nV+e&jYRbwpMe^Her^Y1e?)0 zna)w$L4IbKp^d5|1T5l%Z~Q(2`aF~pnNFX+E}~Of3aOOmQm`>jA^h~^kU#S#>-ABS zW}8{z1>F$dpF@f$GHDRXCO%`uFK~>TN^Q(Nn8Hr1`~|+BLa?KMIorIkfUgwO$lMAd zC?dsumxo(n@_!OIm^jX=iF=(qTYsEuS;3^g^<=V+D`K^sIBXrP2Q{Z6XtfFQ@KIl%LBJ3Rl8~aox{+)+J2`EBVc=%h9(`y(k|(`Q!q=(Z`*e8T3u{ zVB-dTG5fhHtd}sx*?Y&>)f1OO*Y`^1squV>TW-i^>K}pTDk<<$6vxo=Pw=tjA~c)a zhm{I{S;J`(sN-vbYTC;%+}ITDK7EBNGhN7t>p_lBC5+Aj32X^Dh6%ehNwmZfGTwe6 zin++6S2Mt(+P9?7$p`W*KEmJwKA4Ef;4_&N=*pYLUna8sl9cOAP!f%U;GeE==vW+B zS3QR30n0!vS`rSI-eRMl?1ScPX~z6P2FI;Tg143OD7tVx8k)(o_Aiey2Xi!WwCpPU z+y$_>_5|emEWv4w6WE9&&mcap3!Vn4avYnT@Uij^%o+?t`6c7g|JXv@@a_nN7OsG2 zyDz~RpJ!~qaU-}Ym53Gr4Mga_P|%(?6BYbE!0MP$aLhi2?Jx7$HESO77Zqs3lRN)m zaC9BK-SrxbHsqkuUrwuT&1b$JnvRyv>&WG+4G>1uVQNJ#S}Lx^(<{sH)EhrMX>}X4 zmTW)?<9)E@u@?Cn@gFP=S4M4@Um*2S3{RX&V5)Wsk?)`Sxcmjn4(MHk-r3DicfOnT zdHjPoAJO1hHL2pA#hb8g-Ai^%dq1xW?9s5wj~tvOgI=rNqIKCDwEZv_?{>7{Jmx;k zTAv3`+(dBd+hKP4>waccR0KY@&t%q$OW-S0O^B_$0Zs!`Fyo06vn}on^pC{Cv_JBA z-XjiHjlG428_4V6x@vyz9AYPi%m%fieAu3`AM7nMl2#$Sg#*Y0*z}`O}28>f7Z?FcB)E$7&Thnm- zVRP^dS4D5VY!YfKMyLA9!wGM09`9-e4GVGn_`VeS6~oYB!85*R7D0X9RdiOrg}hU_ zc%pbE5>*v&SG)&J_h+(~mrLVqM^Q?<#mUntvNY057XoVbgJ5wZc*JCIJ5i34e>|9p zo$?f_YPX{{l)#qf3$W%&4dlx|WCy|%*tK`_P-Vy*=byGhsY$9Zl%I{-wiWn8xd4_` z=kc$tnur~t74Xux23oJyLEiO3sOm#_rL}`S{3??qXm^l-T65S}c#XgI<{j|e@6LSn zeZdQC`fZi?ozLu%jbUfxTwyN8dlL81UF3se3$eyXR;D!&Y&aj=f<2;m{o``7VKjzF zCsmP{h865-{mDdD;VZebRhFNoFH9TTxPGSMFstrcM|tF-BlFj%j?GbYhF_OX@?GzE z^Z!&VtCuw0$8+0poxRzb#%$#FS$V+&ym#E4l}j#~dXl*sT5QM^OV)df z784)j!iaddGcAlh8SD+`=?bD*LDgp#|Ggp7<&PQZamMtK^cM2(Tr7DVs7*HQnnr$m z3p2hIYQ(?dM!oVwX>!ocjqH(aC67Nv!70%XBwKq1c~PE18Z_idL*-2N{bpq*c>F2; z4NZ4)ayYqO>rECtd|(UHZ{kE=%!s5u$2E!Nv*i#Qv5H!CTd+s$C!lAB~$^cma4lgdmDQas;}Et|c9X}U6jw0BF8JDY@=>)TRTh5j7&Xg@=i>{-VoG`}ac#u98< z^>NzEYvV_+xX(n$l#sYCO5Q(v%HKM5E{S?Oi~QtulSx{x^m_-V2`C+Ahn&pGqP7WS zip5LTfZHwA%ba0`MkAOq?tMNkvyPGbb&H=aYQWA;DrHm}GT2*d+sM>`Xyy;KX1Z3q zWqv=`CyPyM$cDxtX41YhWX-T5op&RW+?A08i3iun(2_CY5|>OIw>Q?Wl#3^_KgB3` zXVa-gA4%{M1E7s3nauz8vqGMF2;CN>^GPnpJU-1U{65SW?A=O!3}0uad{pMea~$d4 zbz8}(hYz9i8Cq{3OtJ&gNzZ33(jk0<y)tRe=wFvSb}eWp#~2@Y1p#h;nF$&7Cz?yR3VdZ=XBiMvEM3 z%r(Z{GoQkr+*|CU84o~b(>D+p^3cWQHFJbm5bMd|?06c;811|U&%aue^pMH)etRJ~ zE}P2+elCIj{vr}nw2vMF34XVC5}tl5&AhzZ2sP!%3Z>j7VTSqSe&Pbu+b@QCZx+C# z^Cv;{X9j*;?vDlhU$D)K%K?5YA-~!*v3+3zkZ7PpG)DgFi)va&RN z@jH;`_*R8CIfkIHK86wqWro0a&%I9AhRFKF+Tpy6pExW%xOf6zmWjJ96pA3R|L zP1<9uMfwAhIm+ky`w0Fi*5dpjA3HMV zHCkT2(n3P4|K|)#6jETWN)^BeNzn6(e?Ghu6AGiw*u!1--iB;fsI>hW5i`mf}Cx;rK^mDQ0riPK4z zhVFy(<(*_k=W^IG{F}|^{*Nfs3-o-Q_ zcODsWmcy09k)VIa5&i@vLLHCu17Dj&t4vSAajEGv^F#z|CCV})F43^1_6U5|+CglM zwb-@AN8sIwMB=lrfK1SICeP)-s<ctu=4SscCux#A!aWKf6~Ku&syM5rV%F3*1?{!K+wFcgQtE^!j+-o zAm3*|Jzsi*;Os6aD47DWgR_b2qy(7k=|%O2`LNJ#B1FHgU|y`1Wjwt8@Lqc45<2<@ z962t^BAd7D*;Sk1HuucjARLstA>SHaMbG+DEf`3 z3(q7YZ^;k;jp*e2Ck-2@texHjOe6f-Ak#N%I4SUwk|iw;mOz zya9c>eO4J#-NUHuECN?0l$ojG|nleT|fN;eq)$Mht=;vY+v zS`Oke>UA^K6DvpXBf4_Ls-J@RHe7G2I3iadUg^9vztIw8m1`dMkP7l3ed`LkL{! zoS9D4qpvJ;V2aFH`ug5`E*}v_Q`dyy9LG|Q!M&P2Tfdm-W@VA>CXYeB0NM8rnnY^< zG*AlvN=^;`0BP;X97{ZndDA<}h)106V=EMgtE6es&He#P6EgSK}$C*W#AF`3UeHkLBo{3vm7M+LH zLbGwt@G2@4c8xGi2DI&E0lWWw9PdUn$Hlp##6MKG96mZ6ZY0b0Hp4oDaPZr6p80K~4~{21Ns~hdSwEx#s}1T&wCe&W zyIl(#>^$g+XNUOziOj>`Gg^dBX`o385_tK;TC(FR56n+l(!dYLxP8D<{BzX?$~g_;J3(qxtBbtAcV*HjwbpDOe@3jf~u?g*vrxczO=}>=Jl5_Z^uvr2&_F&SGtzx-hZ(tzd)ccse?EnXLNN3(ZE+&@^p?Y}<2y^d|j< z87YSBn~NzR9OcFqZF&TOo30ZB*oD_tl@XPp+n93wDRC%BhT@kRpp(q`<;C+*dT1k= zWtakcI_2n&W(t~5W!ZmcUXg8TkHAmegA7P}6AxuI6zbdu>hHLO!mR=lyR#E`tpzx8 zcmVV&Cz21cod0pdRua5%1~slTpm(|MpNhj@;GSh43Hcz78?-{{Jl9s{`^0k6^zJXs zqt2k4X@|xgo#gW5V$fXHP7?lIz^x}IUfZ>99xXd53a%W&QdsqcMuBa`keG16SrsAK)X~L)puCpY?Jrc)~$!@HA)!8 z`)kPvInDY_$N^u~&k#G!p4q)L9oCr~$9wySVcR4h(xUv5>-SV*j~(?!ZwHR=nzjQD zhd!Wznc9MT4;DaTY#7$0oWiFIhCutLF!}8(E7&tv8>MGDk>U4~&|>coc35~J^H)L} zuW!vm4gP%6W3dfad(0#+3kasH@WBBmCw!}5Lsq&zC25_l_)fAJk4yc-sA;c|zx6Bd z?VBv0rsU`3D}(u#N?J_-skxWg0l|InD6u)rnY`yBDjo9Ot=(=4FZji`3ciX zuVcQ5IZf^4`~=ts^eKy>@_tw(X^$U^i*b?tMrL-t8>qJKN9K<)x0~yq&@j9>Tlm@=dxP&hq?8(IsLU`zG zJf7YtEXWm_KzlXgFzlre-4;`T%dSdMos&DEgX|#6p}(x$ZWJw|`Y0uc4N8T(J#U%qS$jzLuS-O-WF?nF z@}VnnQj(}t#!9@9v>2D2{sf6lqA>Pl zA-?bzqNl|!W6LjJs#rFiyqcW@6Fv^&zy^05xcdnBIv#j-;{r61;W~AyI9AVcN8BC~ zgq>@~3F0=Xk#l)6f~!l8fm9#I?3amvUBO~RY_bh}TF+_MdRtKS-ArtsBTf{iY{p5F z0fLg1`DmAA0eWu_v;WEl;hMMr-@08R6K2F>YF;7;J)1;)sxLrw-V>CON_s!;oiIyMpM#&>FqU#054gR5|u~nw`vHS9)3Z!W+G_(`4_+490hL$SJ*VU7?qR{ zVD(2s{AiuSKDY71*-hc~n}bR~I%x*Jn41SrBj=*1X%;8_$_etuI?-TTE_wzPl82f( z_#Q(sQF0a>-)4eeo+ZILy>b*03x+_sCbr25L zl$504pF=L@>>9!Unj+BQr;^Ww-N34fVVJ#mD?QVbNX2HwQHkl?eJjwAwK)@C|1h8kf6gw% zA4_$Z&ICr_e$Ei0I-K#}PbYHLE0woK55u^pW=@s~s+niD97bOV)1vA=2<_i2Wy~1J1rPV3C&{8TjHNsE>%n zzp*NUzIoiPch)n;_55it4jaV-uK=wt&1II?^uU!vJ5arKGu+$M!M@zjb>b<`z(s9w z0{yc#f+E#kw4Ep|cx(O;S1*i0>2HB>C(4m{Of4en;?jb^$9qxAshp%A4;P%&-9lPQ zmJ;P`M;cK+p0F_r7_=k@QyXLj@iT|50M@P*tW-7^7`g9c=e@|p zXaC;dMe$ASgEOP((iti+RTRRj*_@w$NSfA7{5v%a%;4`=+H31}6D+=)<{Xqjy z9a0oL3_|R+QG&;(7eJR*Dwe+6OYf~3L*w23xcT#ZL5E!h>gfuwcCVhm$a5LpWxN2V zrKF)r*iSfI^c#L`I7qHqPo~-aD%AB$CW)Qjh&>@cA$-Ribcm{g)0Zkp!ui?Y5_J+k zL`6Vy!wbkzl|YXm6BxKK1y{w+pudk#XH?U|_)7h~@M~#3&)u__1Z?0KC#w%pk4RIj zjcCK%sc~5QYAWkut4DW7_452H;$Y%Op1`Y4NZ{2lNg&p)k47fy0wb+9JUzHcV6y!W z4vRjFywOi8jH~K%so6d>m)AZytrvWPqDqkLXcsjC|G{An-#R4fg}0au%|W?%O?9! z%~ff5*k^^{N@kwmPQ)2ORo^Yt3DZZmd>@XVJR6suE+d_-@%(|Ko4K9%Y2095hNAvQ zA;QdtPUpH1+F$L&v)`)G<*S;Ycz(S=Yi}ukSKb~L^M=kvr!h$aE=kCh1}vT-=2eOV1>Xu=LB`#Y)2huGXmLzn=x!# zI4VB2=XCoA3+TEwdeB*%SFWX$m(w~{AGn@!g(Gs-$^dASMRQ4ThH;x`N%@F-ZT^2@_)1Q43@zbv+==_nz`W9-CEhBxTbeNl6AF=G=b$^)Jjb=Pgw_h;z zYbom`j_ky$Y>c`sjw|M^1f}CmAhTTnD_p%mX8&INU3`?wmkS~PmpN`bs)jA+0>Nyt zIMyxE#&#`h^fa`E$Lg^#`B@XQxl4g)%l(Jp)v-9`Z#{44{hK7&M+Pst+LI+Z`=Lka zFF?c~%;I+H@0%a<4#fDv_yx|mCEx=|;t(58;2_p{-hoCE$%>q2=SdsJ|C#?*(^5T`i+u2*K#9H&i~ z9JK|Hyfb6IO-qHX9||$U_Yj;6sYOqlK(fDhHLhp3kwqPgwAjuLK773`l{vCzjZ4MUT!56ct@Ydf8=Y@pS`s zaGJ3%=jXlMco*Af1E}Qs;msg%D^82Kv{9B$zzs|t&V|!rKc@wTS;yNXtEyj0Ui_v@BZy2E0 zxc5SX^}V|ru8p0>XB&UALnYVQN9KiA-w(3&kU0d?x>uoZNFtehu^!a(mFpK7&B5o- zb#Ptf0yJ`+1c7IBK{Mn$<91C6_sm<#XkUo{QRy6LeY+k)3Wk}Cup4msLOOU}QYCV; z50OKKWiZL`I4qRxXB}UCVjE{Iz}9W0IBVWp_I~+9dQC_P-+pL;--Z1kUwEE13d{yK z9V1*kT+ZC)@_@a2e-PW(!r=8m9@l1PL6g!F?9;r3%R-YdariWu7RV#Uv(9icloi&N zEW()w>v8JzFX*ytHVNmS0x#Jtq9Oi-k^Oy%PTO^tJkn_*uWaOLxbaEay3PXEYj=a_ z6BY6}%LW(fKZG5VQ(@P}Q|xtvMaaV>=9$A^I68D0*BxFAP4Syyh1yF@)F|TkVDBAoE#hoW6 zvA@*U@>=4C;f#GDR{OS++-scHxGjhEsGf>av&MiGt7AUQ9EanQOGxN1QH-_B#m$pW zkhk9Zi0zgg*rqTF)kkU|VCZ%o<(^{?)qGLg`ij;14GU0TK@StumSVudL2$m=4}bQ{ z!hT|iAC!&IB0dHZ4juy0w(o3P|1r3AlFOP{P_+J1g?$Lc|tuOrq<;prJhucdd{i6(d{GZnT3h*P+4c{-=c-0}Ro> z;UZRSn@RH>xxK*ff9N&O2K25SLycs0s+WI{|MUDHyL8zG`b}pJbx#Q7IAAwX>WC8B z*-S*4%V$AVIFuawPX{+Y6yRt=15Eq88CTo&AmxW}Sri5&hZ@P6M_d=ArV4C{?11p6 zDzHeY2@Va_;Ba6In@zq!V8v^qU1@-~9%nEE{Xb!qp(zHgKTpqe`LoG)>Tq|56yDi* zjTbM}0;zMV(Ch`5;Sn!|uiG18ma7<+SvG)}U?+-k_uA86GVxHwV!ER&hAp%I59U8O z&mM_AgB1nF_?csRgjPD^^~GCZ_=^vFR$Z0zm1MxiQV*P?lnt|X{2^b9ZE)v~dDL=J zCw%;nh(#tdQM5w~qgSscmJ74l2~k1#)n0@$_X|+$wK{f}9){!A65y#W11l$`;aL#{ z!M1;L0v90lyTqnP8Pg@1<|e}Ke%S&Gsly3oAD02l7EAl;J-a5gyz zG)xyI?uF8LFI$@LD$LXu3D{2uB*@0LA9x`x4^KGEqf=h2CK3@g)Uh;!6n(8hU$s%# zb6-xdGo_awu(gL|B<17!#8wD3Tn?#MA6x$G4}%+z8e#P`XH;Ig5cgTUMzdH7aw}AE z^THgcx%>_0+}w-TWp?8@{n=EUE8$4~ehzxs8qEH8f(-sn!n?1Q;O4(v-?j1_TxFg^ z9vtMlt6n4#C4sm=`*=P*^pwwQ{SZ$U^~B-Ot%sQ9y&N}7xZ(3J`nY=UCa7M(F=Ref zV)MGCbd^>O>$J3iEs7HpR6VJNi(J><>Q8}axU3C7qp0A-)_v4tUn$mc_uUhH+9-c! zgjBz8W;mWK`<-I|mu$EK=gS7!_9o723sqn`c!s-&7NZmC#H%0g6Tdo9x?Q3NKVGqe z+WTU#PIeu1{8x_C+Qjhm(HIodi-+4H3$Vp$2}Y~zpv8ki0`sq}@NR1WX56%=SFTGRIDz*JMkmBU$AAqp-Q*r~ z;CnR6NtQw9yh=I*Jh-_vgE)kW$CChzC`HgYtAM_Sa zj;zO!CC|{lb{~g#{shbJ40HXV`#CS#adPt*;@Ux_c+aY=5 zMMF8*v+M%CQ{n#hRt}!Xu40aEyUoTAdSL5B1{8Ogq7KLF`(_$~hq)PBys{B%Y70Td zv=e_19z|6JOVGDY!TkC_I-fH)@6wVI^o~0M4(EHA6J4F~&btndyYE07_xWhB$s8^p zEF}FlJ{TZ%00nh3Nl?xy+$x!*dGu{G=&uLXuy0p<{SsI?DpzHDy=xsA@pFYWt^bAVSN4H&Q;HMPw zDYJ^4+%Stqw=4#&N2WNgd<$+@jlvzmQheEI79erj1J?Vxk@uo&NK&s678ITWljEaA zQMH)<-kb#IKdm8$4CJWd%VqSu1$WODUdzTUJ;d)iz6l#Zd`Taql5%4iYBoKAmKP&5w{WBN32#qCpPnRI`@W z)nu^FiKn`34w{FIfa=b<_;o=XDrid4_8m7^jg5*pU4(!?g?%{^UE?j=ck40dAVSTQNB{6(H{-N-zdgnBXV{?E!pvwAnK z!$sd?`h`&I>dm9 zCXIri`Trd~5Vn5xQ; zRUXKDpEyp2qclORTL%wjEo5p##uGywma%*!MfJ?2&jHIF8=l9XZ=|7mZ(1Uf% ze1f!NE}cs4z*d^c?8S#*^^qCeZ?7b#6Dq@!O>g0pBz@)~bgG>$<+9ICBr5jrg^QI= z7zsiqQQ(N}de@(w+~LKh{yYRLObbJ@q z3>KIZHEzsiSP^p^5Q4kc+mfN|Xx4FRC_9x@4oOmf&~E=p*ce{UyRCZ-vr`OkOIrY` zCmNFGPGxxH7ES)^gc)DVZrrE+6=tObfknm#X!lQsyHm%p!S>1M7Tb(Tokt^8SJ9ox>v6E>RcJ{c#Fic$&Rj>l#Ff#V=>Pl=j#4m!`YWeI zhS#prVEr0c;J%X%z!RF+5se8WJ-Lq?hOxqtgJ+ZRmwW!uAaqw zo7Zpd>xx*CYv`kwXD-lfOAT>Xrw2RMGnxLb0V~HK;P535 zPc0MrGZ*f`)Ahm3YtdlI5Ze;6{JxL9{OC?%*A?VRhQL&fCa8Y36aE|fl3d3&LeYdX zG=9lYQpCX|9_$E;#}vU~c`q7H+l_xKt;yqL7=*djK&xds7p9hubFSy$gSY`Sp*)56 zSUQFNOZm&V@4&L#zN2OKXxgc$N19zT=$NZ7DH#l<9X7sDT4F_N=NB=z;9W3B(@fHD z^=0ODDTxJTJz_5dpTLRK+bktao0e~^Cl-^r%MXkj{EfBgh?v**ZERJ4eOjgyf^K^a>4LE$?LFQ~${&4&UPCOq z`}+*bUGR*=FG^Uz;@7O{w=0WLlA^1sX(Yeu7N$G)@Y>3WbpDJLlbm&?Eu$D5=HCeC zM&XXpVvlnJi)oz3IKqWO_AZH0pUN%vR{J(<$rwcS<}=ti)7h+9DVMwPLz%5jJ^;h# zY-ZBAW580_ld4;)gVlm`@z>~c5M@;W6Qqkot@j$BaOX%?&?wJJvTum5-t(up)yb^% zf+d@`HHs~HI)qJ6kL25@*TV2HBdoVS1VI7%LJoWwv)|yt>KyLif+P#Z+aRv_DMN+A zbaZ6q5|$)+14nb708|Sz!Z%ZKN^5`H1yQ>6JtB?yzt3TN{SD||;Y^G<9)+~QgTke| zu%)z|o9*!wRPCd=b0uRbbcGx}h)RIX1yx`+D+``ZbO!U{1MG?5v%Hyji2BJ!;-RY7 z{Gm0oFmc{{ky3dcv`!z0cCYNIyLgC5<6I4vv(KEtu!z#Oby_64n=JZQ5Q+QM*3hr4 z3>N<*l6|j;W97jM;IzXuuC6Q``ffzAtpeX-amsa}k6pwjKUTF}@^vUHStnsDgl^^~ zsR^v&n42a zSppKTTp_qA%f=R*<~VPGGh?@lOoU#+Io~qcG@=>xvrbT(%rUY4`OVC#uZiidYhlF~ z_OV6#D#7Ymtl%;m%~q;d(~^i^w9?@TtPr@*0n=1i%ZX21*~|M}grX|s%6ZYt7aL&R z&wunI{SBK|Y;4;&5^Sq0PG70sop-V~pZV z?x~YATl)F5zz6o`CLKQw$6t=;bQ9NM?Bqk@lGz3nv3Mcc|ni9E{i_=9u?wSzV8wO4Z zM`2^j5k7ysAC6BS#;Y8e0~x=g@N?dJestIw)}^1pohk1Wr->)D4+E385BOkux55@{d2b-t+ zli^0KIk;%_a*fu_+#BsBY)i5ibDbK=WKPSmmdEp0mwGJtsC?u`ulxwBs%N5Sn;XP? zRKe1-S^Pc)35<5lmih&gwq|j;8J*yq zV#yTVTnDp@`{0gse|FVwKaA>E0S|Sm*h-~h?r(x1n%`*$?HY|ZhJOeVhKcYqcnbK+ zJO=IhEZ(l-Gd$JSWI=z_=&VI5cWl5^KHS)sTWwk^>Y+6@?w!y1=H3>N`>w_Ue$D~! z;liAwu7v&kzZb0Ck5v|(WeSHB@$lyN;P+YryZnl|L$f2fQEM*oS3~STeM3JS+qQ(s zsP?nnZk!HVMEk)jV?1L~gCHuX2BxjGVm)%fF#TU9aE%`z^6oP*UZ4hx&u;8qLOS_41 zPo&T5ecu2OGZKouhJxymD_m>eV&43mJQ$eGVxD#H*q@vv9CqG=H+8rQ^37skvsWFZ( z;J14;`(-wex$d@srs)qvojV?JZ%U&<<&O=YIU}Awa^eCPtEC77)+nL9!5Qx0tO~Yg zW*KKH%#~){cm-?K>hXm8PyXpqcgp&)hGr}o$!SF0ge_|~i-+irVT->Nh&MRzVxsq5 zcy?hj{+aUubre;A{oM?4^B!a7jyTAE^Ov6`Ck4L;dNJvOD4bz7g;g5vK#gDfSY5n6 zI?5g625(k?W(#>%J7g;HX=C}vE9Fpe_ybI^>c^R-d=xwEw?wUU3;w)|C#=|i6r69L zgE7zdgXPjQF!)w6*qqX4!-{J8JNu57jPaPtEl!Z3;`UcmEHJDi^V8Yh6{GQa)HP;$ z2KbSOISB4kOey zyJb}A8F1GdVms2;m^NLJWfxcf0S8MjSZLS8AN_R=Kdo8FHm6DOyUGFa_QOZe^|!#z zt~*S><4+5{mPyR`pbmQ{JBemJAzZDliQ9YEi1LgyvG!}XNJV1AjDA|-ns+C_V)Hvb z?$#-RwfGh<4U<9l!a-nv>=dXz+KIg;Zfur4hcaiZVPVEII6JijUGvYvDJeCwls%0{ z+J;^A4$k4+Cuow1`5tJ}d&TW_dBH6;|BFFI)6v*ehf7~R5oH9QfR95dc>64*ppgBT zaBU#yZ4|K|VdebiicVzbcEW3$Xo{P273^jV#^$B+?AIkLN?j7pE0?QAtP!4QJTLHXI)tvrbDUyoC2kW}@uNB$ z@ZBXf@!`ryp1VV0m6CXT@=q33yW7C>tH6`rFbBiDZ1IyMi=UWe!iBZ^;3ngGa5>Pz zH|5(>)$RZYjyR4I?R&iX1QpV_TLLTiLtq`G!p7=ou!SauY(ViDHhFH9c>lk_P&Ofw zuV-s{#hCdJIDZ(kKCi}BR4hb=DglI0{g_*`RG5pj1+ZgLVPHLdF@HYxCyZ~F;kl8+ zM2?5Yu<84S^T4Gq5Hmo`d^^9R;QsT zvXk4_F?(-$vY#f$B4d1*Rn==Id()oDIvc=q`8>8p*$bC!EabNar=X_zKW<*~Y^G5t zu>JmKGEL9D>`U7Wep7=W2bQ`8WuZBOGoc0ES5%|!T;YAB>jnD`^n?99qgb0uGMpd& z7garqL8qvbPh2n0nxkZxUeGbf{L%>_yQi|BLz-aR=TI^@smDfg9OgQ;V~3U|j^Dlv z6E4d`d($sGbaxA%noe7K#k`ndo*)HQ-B`VF!p_8H)e~RsnBj5U0#sMp-KgWX={M> z^=Mk&_LA0pzeWb{-@pblZ}iRDPrjvtxbZ7}=~bi&Qy71dBFqmE?NOkHq!Uae{K@yW z~hcN5Yj+oGp0Iw^n z==yL6wmPqzGaEXIzA6hlMT1lL#$*bOy68gziz8W?;P|-Y>m_g~zM^8;1lCe~kt&2C8~!DV*>`_B4ffYe)l?gwpFwR8#T_!Q!@@x~;zJAqH=e-QrNuEtd@GTfL0OSpUg zRp49Yt^AU~j7AL2g9Sdu;2ro5+iotQ?GCeXuFh;WwtEmQ9siX3^01o1PDaqqob#N0 zT{T`!uf+dSIR1z3c=!v=kbmk9>R<9>J}p1#%d3aFnEg!JK{f$240zq+Y`&UEtYoNmXz&SY0!j;J1*j<$`Y=B&Jl<* z4WalesklkXfO$=8q-j&MaV5J5R<3KfyjSn}GY8ZuD)>z4wlEEG?j|p4IO0w3PZ+T6 ztOR>r?Ifqqm$=%vL#*PW1IfuIvKggoS;OWF^z`CleEmv^6*yl+1MU)S_3=g1qJKC~ zRTUZv0!Vk&DDu;a72b#UK&#hWWS}bCGcRwVBfna>`FodwX=w-7v3fl{8zBp$z8K?i z)wyJHM2$v}BXvbtvi+Ua@TjDQn|JO24E@}nIUYUCkGVXB+}<~pKTpUHW5=r-4BPur^;@d z*<%{fa_1E~;i|-jj){Yq(d|@uW+mU{V#1xZbR*H?WatZQnBuxywcv_St}CxzB<- zy$7lA_Xx7?pUKJsVrc7-xzs&s2t}>`0=pk1(m2O?^l;}=tjReq8hJdPH`tj=I;oZ1 zoja4+zK%V@-qW0h3-`3mQ*Ti6*Z^G7XvH0!FaT8Em7-{wHmRsG7sZ%l-J)e09Tn;;mvY$ZRYzC>(IEr*%$In!_ zhXZD5lhW8HoW{paNL}}sr};^AqAQNC9qUg|q~d5av!p#reRS`}SA4ZV$RL*|Gdu5j zBwCY+`YG>4TP}Qu-;bBGtA}T>1c4=D{NG{lbP)V+`(D8}(KU22l%udk&vEq4A(CYi zB58-wG_swZ$r9crL*Akak@Fh`c3o1)i$WNA?4J)u*DS|B|GIg@vpcvI0!Lx+216Lq z6idGUZK3z|3DoPo7<2B(qp8qwUb85J%kHw4+=e#p>s$qvbGVT7r-qQSlrVRT^+U?N zgNjyOY~65UTDUBf_~{CgDH}4VX?Y|~)vtnw2|{N1lCngu@DMIMGl!S@cbR5CXY{FM zJGvM56M98~@MVlK)n2+z|CFRa{%(+v@R@0|AGDaQqM@zm>I=EX%XJF zQlzvIQz>5~3y*XhBd^;6-*mUoSASlO)=LuLL%g(Pwcc~AD_kbzk492+X0-Jrub0@B z6)WaG+=nFB?QmD`n)vPq6)+#Vfko;RQSdVd8WFLBGW+L~>$8!pXi*kyan+^sAAq%$ z+2i2N!X0+zVfuRX76d+jgBOzK)9JVWpxkCU7G6F}TZ+1A_rMWsW5gzWcW4`pbZv)z zg2_KKvumU^Y`pR^N-cs*0f>ZvY~=I(S5E|=HX(N$Bn|^v~kR; z)(Km^>)@5F8d2kReoDS8RJ_wGojo48SJSS+j1@_||BWe--7|tc{WOP}|M24#>IYF! zdk}8S*oiZpO2wA#x6xqfU|Q>@OugyZ@I^Q)#s&YwK>6R0`T#h1tR~zOa&cYnQjD81 zmOeeK$F#lkV0(DGc-*7}n4WnDZA}M=Z3Um4c6|&hnkq%{x2)ltL=huCO~6_4C3v=1 z404s;bld3&Zhln7`@}xC$?TJGUQabJdEyD!TFnx7EcA2#1yzQ*z`A##b>(b(SbmTiUao=3Ss zo-ai)-QD=w{}k?aF(n&cKV~1&2wSJzX4S2ksBx;D56p0(-lK->RMA@2VJ1)6<2Ta1 zRjsAkjmMc;TQHpYu@PJf6&RYdV=GL;$0Ii*xg^2`_fg_Ve|xSnb~qbzem7*@2}I4b z9`0n{WZvq(1{A4V+DJK6LgwL6F4TTDbKf8$jG)Ab zFR)4e8CC}G;cxMtP%>cz{L;+jV=XFhT$K^tTDBTPJOAOcCiwEp7KENoeWo&$e(sXE-d{|J1< z@JeJD61Si=S+( zLV=3uxNJ@^>NuBTd+~Fwcls&lI`;|66PG*gKhG@wT18g^EeOsHiuZ^X&6xmwLjPx9l$rYg+*u*AK?+wGrZ`dD}?#vNiqKXNNC`%wxj~ zI>2zV2u}2>Ku7K_+&{S-yvjU4a~KbasutLH%ZX`pp2u?|U((7mIhfR=%cX?mbA19& zwZ`dYsrsUBPQ!FM?DA1(JLV(KIpxHaRw&?_XY(n|WC+*ZkdC%nqrqy%VLE#x63+^} z+8L6I;!(HnK(cxq8u3%W<<%zqH{V9|#m|ZwMI3j${T?5f9)UZTMZ&&62||Xd3AV>? z=S=3`Mmk_fayviZpM#1dTHcRdX=Fmq?o;AVc{Ax14ycUyY!= zqMkjTun!F-S{$f6!rb`7{Kv%C0z-g+l25?bg>C%n{;~Xj*5gpVVGTPS-(&MSI! zZ1lJjwMg+R6CpGNwK10``pih?k3{*x^nwPCE2~8xWv`^Afj`UEgia!*@G=-`x+D zd_AF`>|=3XjlgD8dV*nPH}UP_pQT~_V%RB}N*XQ6qL@v`x#!D9Fe{b6y!H3d5GY-P z9d}p4)b<|EWltJBRaC<08R4+?>0zyB5-B=n>G=W4?(k7ry>eg(?M9JT4q z2*b6PwD_DG-r$;fl8cjci#B}N4(Vt105r;j;2avMPLFG9 zpTc~CPq7aj5tMbuMARK=1TXJ-!!U*8T;?jCA8|NG*n#}Txo_OrzMxX^&&tJgb@_2& z_7?%4SRzb!`@o-V3qz@EmUMlQJvCphg${#6_{B5+lqKNMdnMDel)8Iu@C3oeUJQjbP#LJ)77ay4t%;J@dX+c^z z`^4E{+Z|b2S}yRO?>pM6Z8Kv1u0Mdpp(327n}UnXEWpU@9v3R~%+LNA&*=$#qfp0j zymrN2_+qmQ<5fdw>6yh8ZMK_vetrvQBOfsD_7^CVtj;;_^oIhA4ltiH9&i4g!$x$r z^UYno;s+aIxQ6aq*evvKEKXX`)rTwj3N68>ZKZ+N+qR;kx*RyXbH|q_w4wfuKJ0I? zgvgXHymv<#-tB9{dBXi?P`5GDp1qM+@C-e~fVDr+@#<-T-SnxnPD|L;Zd!ulzaFFyms4rm2wz$d_<>HebVEUq z2JP3hp-U^3Bx5-}`jGpX6x5n1G&rBGhGgSgXC5uQZG@g>8ihR^N^7F7@v052n7AYj z2lt66=lObCS+NO|$~EYPuOcmz|Br^0=}1-vyrvEBb`(}j1psyD- zqMuQo`UrZF8o}50ef(oZ5(T9cAr1v+R9*;aqj@Nl1V`zwq%Z-Gr?l5s}aS2Ml zCv0CKqEWi}IHWFA=peb%5W8Odu;dVb<@qUea-K-f(mtRlTn}5~cGGkHqkN^1S+R)T zPmyjtq+KB`S=n-(4s$E$UhiW1xlx^3mYS2K{TNMtI*gRQH{b?yU13_1C}GQNCB@bA zBtF;8C0#G%CFS{glCwr5$@z%|&AjDIF{#68o&Pyh74qP0k22lgT#ugOZ`h{)n*6ow zQ5p0oML(VnH7v*Ng%%VUyNvkzL^9L2(CNVf=kC=sl6?Gx^8(&r`NwH=cxfog_i2%3 zXcI0Pkc77!teMo`m1J+;hRx*$5>?|48rM@vM?8lJGa?g8`1%dAuihrPQ5=mtG@X{# z{GvTeH6_#RUeF!miR8?mr#IjFN%*oKXgq7I!2M{%=;iZi;NvhlQLQ1lq#jJWu1ToU zwg4xV-=p;_GRUv$9!b4Q#NP%o6h7!PPQDU_Sra^je1kmQU#2bb`q0B=x%?=t*&^^K zPDfGiD0A}s8co@r{e_I|H8M7`pqxCxgP;^hFEge{z7=It^!`G!8S<2@;5BWXGMrrc zzrZ~=h6LePTR3mFJhk2n#n9ij)Ne&FJsY%-dd9sY=|WYB!)-BSce&!6)mrRD zoiHnXv5x))Wl*NpZ}h89A?a7UskpO)4vt?*HwH&jp0Y0OzLZV7{{Nyy8=5~UI;n)~Z8xN)^;Qt32v!E8*@LKgbVAq}96h=Ww+t!~W}%V1{FGDJIeI-AER@g< zooIN}{~7d#45uEY8j7$QE7|JoEm7UNkUdV_)4Da&chT0Eu_%J<9!(K&`1oJl!e7?Y~Y^zegH0|56oAcgPSI7NnDC(Q|k` zV-Yww-Qa?ycVd6zL9{YOhul}>)9+K(8ZY)lv^Y`2E>VOt1Z#qrA*;zDbqc1tAKj7w$KFz8_Wzg}Ep~UIV zI}9=1NhUgTXxTR<$*tG@Br=ZSpfPnTEuPhZv+K5#kHDYNRg}RA^AiOB71EYuBe-&H zBJKTifDRl~qnQbJ!F}&EI`F*`NiUQ0G?bCZt~6y6n+`y=;w0F3N6fYf8C$7>dJ@GZ zqcA|BJpmGOsaiy(Uj1wfhTF0lS7DY@<{=4fJq2p_XOQXFev+EuYLe#tUrBuO8ve{U zf#;VhliCGU$=SEZ>0QBQN?!Pqi|LJreHEk0pSP1JWtxDi)(sphc*Bm)Erct)A5|Cp z#r&ZA*dXj=_EhFkwJ`H~I`KE%FH)wEwS)MctfgeQI2`{R7yPC_I#6;W307~=p~gp> z>6~~z-WK)=a$o+#9d2uIiKZ)*MSMj!ZxiMh@sD&Cny?cmR7q}aJ|0S_Ctk9IKIcAy z0LKh^ambgsc)uo+KT6^aduUB>9X<7UgQJ2|_*nr%*f*_d%=JB^@cF%H_NyGt_xO`@ zu0H;B84R7$%PIPsBDKZ2({-VLzcD+Bcp>J#eA)-r(xCw%LWZIvu>oI7KM?QSx18$N zO_c;hwSwU zhMli=LhS<+;jufwu(~Q3Qj9LxJ6xTf|SD{}$vg+QJ`AEq2S@8|)7haJ#WT zQ}VCE*+Wj^+tbplXq6MFp4MfBQIjC!ZWzl}zrbdmZvmHx3yf4QW66yKd}0{Lk5PEe z7Oa_q9rHq2%pYq^o4y6*@^sn1WHqkfkp^zq+=o5pzhTwS1aV7Z57_4U!NTmfaQWXN zFnD5$zsK)|(&@k^uRV#)e_G(A?GpH%`kdEIy#?APj_kwb^*F{_xNp?tFrP*l7Or0o zv9d?Gt#f9=dj1LbA4lBA_ZPwDiZUJ3n87;V{^gCwOF(I&A#;uQ17~}CEIB%WO}0@J zzxz;)_opA{btmbuviqOJIaV_4X*kCc=WoFkUov1z+YR*CG!soTM{+0MG(f|;-S9JS zy|_Mi6#mG42s@uE@OQ_wL7Vm_`1-ILSR-VF5c1#BKj81%eVkRed z@)n%kIgp)pw16Su)y(z#OmJ$c1iLXKSaR%YcF+ARH1|Ys8;^CtuICvn!q5zg*RBPR zB}#0OuQMHbIgyU{PeIW}X?8$&Jk8Mk1+wmzL~F{>EocaVwvyN!^&?VASD)a*~ja-m__Qm_^&jBB_lyaIj}!3lz9Ye@_xl5Sw+xbTp8-j>=no0jwxy|?x_-#44lc9IL(KF zX%#T^LOz=-c($kK)MND78{&hz44M9li&(McJ=hk@z^t(f$Z^NmGwbPGiFgO=qj&uF z!#Au{*0;g>@ALR49V(^Gv&V>}Hicu-`&54GV|m*+Z9O(TA{?B1gl_5JD6UfAMy6g; zE4_5si+!tRtaa8?c+j2CUa$9Hfrqu3l{9iLPwc@XZapj3ieT+hn^~IsC^q_h8LUyZ zW)_pDv0vF0;PP~w(A${~=6M6)=PAK!C@`8v=2t*&j3Owd-{G>no^m(ul|c0NOE$h& zeYoNmZ~0ltJ8)uP0sA&ShF|q;#ai7mgA&XQ9H()5c&~SwN zvZ<3>B{;=Jd)3&v&C)FJsWH3RvyUBW)1n=TC*k4uICPA1<6{2YG6}BE+ye{b|ANtFU8WbK#L%IZl_u|F z*A)ub-oY}sTrr;G!)2NF1QYD#x#;p(pmj zTLlTv*_{B}MMe3Uzj?REs^FC_Kk0+gQ|tOt!0}itPnQmS*h58e7BohO{p3k@y~dx|#(R z8)`8Br6X)SHiRoWZp@#vFo$(Hi$O;1Jgo1jhOf^mIH`Ynw&kHb%jq;1v`5J>MkbwY z-kKqc2R;b75K;eC?A z6tZJk|J6^q>BEk2V-p%7XVw)6(JbalpWnbKC&se5w$C~1+zs5;gfDRSAkUSzCWx#W zf5VC(d6pNg!o+I~S@qDnV6*W*zVW~l<~{T)tLh41>%BwRPtjC?2^Pi{?p_FI9u%^* z2f87BPAwc?!a%{i4z3vtXTO|X*?|+0Ag6K^&hM5(g@{&>D<1}fV-B#MfQ#^(^J9|r zPocDbI@1t*_lgV7!NmH55Z?3#e(RaCqQtZCb%G_EJ54b8evLt8YvH?SI1Axp{=#R$ zPhuwA4;t01*wWQ^LC(b%4m95sP5w5TE#C1OcLgS~kYV>Ba`im+HYNxZdTL==x)PgK zt6=+mp9YiX)!4Mw%V0M=o@spj3(AUr;oYYW*j1w?KHhHt`!Vt!=nXG`MN3Ly+l4Lg zOQno^#4msiYcGO-qApXMT>uWRZAdc8nzg)LM|aLk_!pgXxPNCdp!iiFyCb`eNg1zW z%cB&S=Y>=u2Wtvtrt?_1;$bdQjuRX!a-e9O{m9<0nmi5+{c%37_gVXxswaNQEh zevWK_t+xiSq!uSOH!PH$pK8h4NA6@P7lP^T)ClnKXyM~VOrkdd8`-v6AGU69F>7jH z#UiZ)Z){^Etk`%1cdCVe(W)3y+?S2AuLlzws>LF8Pr$z7ah(18VO(Ej6xaD+Kiv{K z)P?AWMWYYV<)B`!P>3Esb(xN*m)WDfcM~sOeS~vO&|v+8`q;N)r`f`by_o+ah_aTc zvR7a3h-$qGx$QeSC_Mg%uYKpnITr51GlvBSOOHBh>aPy=J6w5nr~B~GF^=R;HPh+1x+R?+bYr@?x$%HTRQRqTd{M1?M=^15cOgRvptf}4(yc` z@`zz9%0!1RN@#7Z3^Xdi}J05V#aZi}7y5K@=wPIJ+H&H_BX2>5V zE6EOO<2tse)6`d{G}WP=OC7DkzPmPo_r?#95|D*526p_hly0ySIu3rxGoZQ9gT3FL z0ByCvzR%TTUbhZW&MS9TIPeemA;_5>tsP{m-R%Zuihr)wWRyU|6cIn`^F!(w{tEi7 z9)nqR0dU^wpD6I;J$C0%D07@z!=BzeF0Og_1^rczvWtorajRVu?3v&|USA}5+`No* zr1M2FmBL;&sg1pOxR8Y$iJ|j%OXz%BIE9P|Vp&V9X?0Hst(Xx-K8J-oo%I@&Hi;ss z&thw}*>SXTfCS#p@{!UDtkGxwG}_F|ox?UwLiwp}i_DcswDY;u=F z@bYdp&s^EI&-5x|VR~d4aga0x&Vx&I9E%d|V^8EBP>!|W3heKPTR2V0*qULaSD?V2 z1PQah=G*Lb>_a#s%`-RiVXUEi5&Nd*#OxeaWAWQgyft|N%dVcy^mn&Vh*u8$j9A95 zEq@PMCpM$PtXpiDo*Jt8si5jz4|dc*nvKj#f`tpmaaH#cSw^R_>~@jF%(f%k+p{I5%hnob2^x}gItjkPei|3a7&8;3I-dqjT^+VR#8deNoS z74rP@L`UTO`6Z#ZN`mWea#6kNXfXR2xP9qISEku=rY%*x*RBHiC9lhgwS#bGvNfFU z+s;|^?1P_f(=fox8x>0q!ujiE=xI0&HqA0(rvEmed0QvkTKWzN^Dbm1{TfLg{!;MKy`ATSZU%x{=uH<_{77DZy49fk1i?Vr*t)O zYdmA%Scn!2+q+We`6l2iQ6+b+>m@!68V8G~4ZscZ1L3Z5KDzbP^ItL!fLWC%Z*Q-I zXFYW&@>nO@76mcIe1RQ)bUeSt+m221^#jMt-neK&1oq2nLodytobm&2UYdVoJ+Vcb zTj4FotUQCUV+LFuRQCmXU+<^$lh~FkJ<9+NyFhcnT z7d0>eoxj#$VXZ2~HIETz$fk04x~}uf4-t;iUCMoZZ2+5&^&|7EHGJnMW&C)O@N@VD zTz`B7?ERi4@{Titl%hT^WVjI<*MC1YoL2`QUo~d?Zv?*H|3_>zU>Wqy67rw<9bCV} z_0U~<3i88OW9r%A5V&j%_dU%V-J%&c`pZHLy>^d3m0`gjan;29CtjVXKz&mU(Q{RQCiUAEDw@4`yQbm1XMny<9Fu~- zUte+6w4DETA(J;dy%qLOkH!Ov)8VhBF6}S~#ZrN~yHn4zw8F%K>Xp9DNZ? zaOcj_nMS9ed2IoEC1iovr3(U_&ahL2XVJq%A0fdF#Xw z%yz)M=yFJYwV!=j@)Cyjp67EvAH*pNdN6Q6K3_FtHI8kHWsZ{rSogq1eDsz+I8ob- zM$50k$Crg{rAU_TN$!T{nkSj`HYKJq)EE4xlt5rvD{S~zz%NwwWT)cQnY@tn-uQVe z8y{Z{8t*jW*JxGvGBX6K17?e^FR&7QIW++?URbl~qt)2ms|BDCF5J1>^T1+nB#vvC z$NH2uv#pZNoTJzWp7eIW_D!a+5H7QC+2`0Uy+oGWWynSeXN>pHu7O>2m$rP?oI8Uav{vDRC})K0h6M#5M3tV4WxI8NljZ<$=iO3inD&osIo6%yyfqFPmR~ znhW1#BkbAwQS#OYur%}?S9WO_4&5oinYx47%}+;oSG}Qh>ZS<=-?rd;Us&RQ%d*(< zssgU3pdXv(q{eyg&E>{&4`9sMMDe_5iEQpuW66bs-{9IGGc4Y87nJxtFv#^QQ})SV z{~natY_D@=8?+;Y-#j;#rHwGJZ>OmKR{(om_=+pOF`UK59AuUc#^UffhNNM)?|%%P zhhL3<6vtIswD&S1tAR?o&-vc0jFeSINJdFSB9bi)rCo|>ktCrabf0ssP(*_iSy{id ztjHGX_xuUn*X=&fIp_0vzuEa;=b7FZYc|5$mt8F=hLZPv*Lu(Oq2T*2fPtZj9`5eJP(=s?`ATf^?rAK_eJT8T(*&Oy9^ld;0&n|4IV}8`4FA^mVoUHKYSH?{`TOl=_j=>O zH+e4m;Cqr@s0;+7nDcN}GYqwVAOyL_!dAK4IAm#G*l~6`I66kM!RvzfgvMV`zJ4)Y z5B|*;&-@O)<85KDT@YxzI)s*9d0@@yvstG1uzd9h`-88`n3Gc=bUc5CwtBlU{*rlV z^PdC=$yQ+lZuxLSt+(N^;pXsRwhlvs4cO2vxVqGg+0A2*cD^|k0^QMdT-|)^ z3ST2S7IK&`ZPI`T9x{}px$UnZEskSDNn#43tb13A@EV8wC}!;w3sK|AJpJK*ZZgHEDcS2V4-`I{zgn zz|zM%`R|ESn0<}F`Q7SQI`zNlxGN=(m-%uWJmc2!ozDj2h0L3Lx%?@}OHLQfs~yB$ zaqQ$Phq_8CIBQa)FoS*aS=B8}5(czx1_Z zv!;eYZj~Dx6XuR|X*yHhl>uC7Ay88nSKMC6HU7K<`$K|M=;jQ_}4tX~8d_#u+-Q_h3s-|?&qjhXlOb{G~qhTZA>!=<{M zg0=7cnWC`2w2CO>){a;V$G6@St-qATZ<14Ge(ELsF@Y)(I7yY2p5vi#;b9i}Cyrgw zN`ljS!l0^iHs}^fARzh#7qvT&D>(F8H0xCkRIBN+*nx7TQt@uIUzQ1dx-8H}rV(BH z`U(Auc_K+uBVQFpr5#SrqVKO%C|U5XudV%u)dQ8-mTMoG_gWX0p5g$vcmLs58|@S& zjyuC`t!d;Z2YljEzxQEDhQdI;WHwH19mfVg)eyxwUgsOuKas?-J%bhq@L+|4LoKgDUucQ!^lC*In+s(?RgP;>7d}53rSi4g8MER^Fmx9T#A% zFn;5@{UCjn0;3f7a$A01l0*po?A&rM{(65e@Q;gPzCQ2l)q{V)onN!qt37V4x3@3a z8TxRuD;Du`J1wctpMG@Gd=2g7)WO2~2e0!rlpi+rGP>XN!9&w*;Vc<(a_SAD{aylB zyf+;lp0Bi$ z2v)7#Qab0?E7nlHlyzRJ!k255a9GL#zW$396bvXTU1+flaz_sq{Zm*4D{IDbe@hA_ zLrb5-jbW*L;??22pT9fWhdq)k*FC}UHZQqhGrG9_Yf{-@owqQgR*fwjMR0RV8GB@| z%vvAKVVfg9vA&bG@_Q}rLW#v;zV>hmD(DPkHxI9cHrH~#d&(?4QC9?y7bQq4tKDEt z=V%D-J01X7WW%7N zLZ72+tWfW`A@Tk&9OC+Bu=L|z?Dx>4yzeIuC_4NTT;o3TM!n0hG4m|%-n$K!SX_fo zU*o||?g=*`W*-C(yUh8|H=$d1{)6%%-KF6N?Qq%A4frCh3lFZD4jv`SEb#VeNPS#| z`9l*?G}o5$gOthBZ6&IX3jA#YIF_0 z?x#tsZePaiRVL)AzXEpkJwXQd^@KcNB&n~g%NfKx zCs$orc5KT6nt#!NPH3gm)f;WxF{6WY^1dwvbvu)!CW1`y3)P*TO@1M>sAladV&_iL z{XrKvztw59`&1|L@1MeK>+5hxF^e>pR+FMmIgR+?jUHe9ap)&wI{aFl7IoXx+%`W9 zO#35gv7SO5Rlo4k#sPGv)0g^(4x+K8HgKTsC7+tu&d(dRf(jPSB#(%Ea8zDIm#v3O zJ*XDTiWgIe=@uODVKn@+y@cm122kZ76>1rEAK(0NW*=r+OD)lvtn-%AbM0?*2vz8a ztrwYOnbDS`FL6_j(E0F~L;kiks1jW!0BR)Y$$|u4dphkMSA!d8U&Zj`5Zv4Fl%V?t zp4okY*!&P`Eea=rf``|z7Mp_f*n^+-=++(y>pnM@e(~E+FK^`VIpe=!)``)4Rn14} zNHLelFN~n?!@da|$~4mc6-=KGWQuGA7eZ(fk)FVVbZody#Y0QTcJvR=|=e|5zN3Z%XE}r7g5s^$NC^-9^QSB~(p|Xkx`0 zN>VyU<+&@VGNp=hInYGD^Ezm5p3sk}ZRgBKZHK=1;^=y=J-IKagn@QG%r0>Fn?Y7XqiMpI*W~))9iQLdg*Prb4fEv&(?hEvkasE#qvBJ9 zj&Kr%+g8veA!GFB?^s%y^N)_M9z_#$n(1-vZ<<%Kg1(;YBc-4W+LwL^zqXU$L$jup zo43L0k`gBG{hAH#+{3urgWz6q6z5u{#r`fd5ce5AlnvRhC_Y}D$Qml$!Myq&Y^%7) z{qEe)9I*{1R365Y8|AU;Q54Mn>&B++vt(l*X0rXA7g$KnEmpByP3(aIT-`b;=Dm-G z4ufX)jSpdy%WGNnF&p+%Q;Y4JE`{~ppTO9=3bw8pfEjll<8bRTcI(9+HsyR2dmOAs z&&R|w!wJXP{9aw!EljCb*lhgPj6S zXuy7mi4Wku-W@DzI3&Z^nm}+aJi#7MjADaC73|l}l%qA#+CClt&c``9f#?2c{^UgEpyP-_Cb}hU4{REV+8p-&r z7lnB)g1J4r%Pw3P!DbZ@3p}PI{v7Sh+4cmogqqCUNTR^DL$&nmzy9PrPGJ5gTSQjP)$NgMTk1 zLTkJRGuvv%X_U1x)5q1U{ zSC~?=9@8B-QLMJifO+1#%!(1t z2Kt{Vw4X6{En5E-is$Nrq3*+G@Y|5aRGSRgzNy>r)?I6HOT#d+eq0z!x2yqe;cj2@ z${ya-RI-_6<;>^&3AE6Y5ijt}XS2_2WL0%qux*!;xcfyK`?6ykFDmf{|2=B(y-QBK zSneKM?ley9J*}SY+|ysYT)&28?M-0|xfFI!s>g8ST=u+85wbdh@YInG*ke%4ZkMXE zEqUX`FQUZa+ru})tWiAdQpsWJsS5>nP!8)MMRCW#H%$GkBD0juW(QWAvu~#ciI+dU z!sgCaWJiWvVgb*pp?Azkrv73kZ~AB__fLK@e~=d(&ApRZ>%T0fTwyHMdL1p^@F75a z#eW{l$T|kvCzQaXBZ6NTD{!j{t(b|-N;Yb~;QlDN#CCA=Xx%vxj2$!{(+a*KDB=0#APjP>i9JAeb2|(^Fjt*-df2tE~l`bkS2&${sG@C)4ABfdQfeC1tAhT+?*E@~JCnL{)Gh5dmnSb&Knm-Gdzqo?yY!o7~~1 z7L5Hkny<9Z<^#rzf*)UXXzIPbqHBd2uocKn(0o!d zun={(Mc~j$UcAX*4woM@rjFOfY;K@D+q!<4Xisz$qGmu{N`Vnt>-<0 zbSUGKJ&m*M#Zg+CQupj;tWoYjO=aOZ5^@S}x%}drKB-}Wmpoos76OT9XHdZRk9_Re zk&=?3;Z*qgADV3NqbWjP^M_<4#b`22_-_f#ThfEi-{w%Up@dXgji{-$Kbhniav&ambS8it=dV{e$@CVOp_i(vNV$9ks&s?rI z(lq~ATx_zQCsl|P<7r~XyE68AA2R6oh!6sWzQm|fu z#x|yu?5;=#n)BGfl}Z$Fr5DxCy+Vt7ju`ED_H);Z1TE&s| zo@x;DODL@WbMS1dft%*%(7;!LPWoB#2~)LsPPB^2Z(fcSdJjl8C>C8_3EhVEWB5E3 zzO*+u0%iqP@vl{S`GE7UMLV-G#=d+a zd8KgzR%c(~D*UIis)wy4YoQ=L^eX{T`vd2BHw@EcZ_^BgQtI^gqEzWR8ZX>fYprWZ ztM@G!{98a}_l8sSCKH(AXGPcQE}+JU(^Pf29!d`lr}~#y`J;*JaH7EpjHs4{oBUof zRSBZMTX*vX8vQWRQHV$`^wL?M{*2T`iEv*WtuS2rvNVQrXde5%HJ|)a6LDosIGt;W=VIila8j|2 zXdINGG^ECU>RQ3eU8NxXulh3uuWYBfs9<`xB!;#XKY>k>y|nk=HfmB|PkXn_qMvg# zq3MbywY6+OapN>}9iIs0h3aI#Iv3-g)Z?h0nb@{*2AXY1$IhD$T+@^meyfKqjG0v; zxF~<&<%;7tOKQbFWO{-rTZ<7_G*BU3LO!uwq~>=H3MQ{5sj(7^UZTO4atE-wE06D^ zCks;+6wu6x?O1Lg+(J#4!{0~ARI^T=D*Ybe?TR&6u&|zQ+H4`2nOg^G3W3;h?gx(# zE%`TV-oTb?Gbr(j0rPO*j6;$hqWQvQ{Q2@?2oFU50cQ)&F?Z24NR81MDgV^b1e;j^ z#%46q#UpEB%Z*YHxkrm;ig<4E{>_r|0cOw|QGx68dqCUEnzn7b2(oP^0O#7d=`|c3 z{F%p_DA;lr&F{j=G({GeB4Xn=S+kgPIk0O_A6h$WK7(`aZ2wPXlJ{*B&8&*U+Wud$ z`g0lQ{B#)YJ~0L){Vm~Tm@&ZY#{#1wlj}Yj3Cd@qMSmM~$l!bq7%p0Xy62WKY1(6` zsLIFHB4e)q76Y1E?1BeB7I7K@S+FvD2fFuYvU@v&plUCnXM{7oh!ai$`+F1}z zEkYzv27V6AM-QiGl9U6wta<)bZp+XO{QZy^evq}8#SZS~e}##dintUq{}aK2`Vcs> z;3Va2H~?q8jo3Ebb$suL629wG9h+EN3T<(kG)cpq*ZA|-uDZ+%ZCW<-w@lT*-hMor ztTdF3yO~MSoc&CFa69gQcZkhezY}Y}HSzK>Z(wUgFFb8|$DMY5&V@|=kG~O{z=|Hm z!{c3ka3k~)cilf;l(b|TD_kDJ%`3YCS+*l_wO$3vuR4TfqC8yXvxx;izs7Yb>BIea z!P|NzLg?+!!8z!u-{=Bz0J+T>Xv@I=1Of^-lj-%2gbvTRG|}lybxp7W!Y;D zJI(Ewl8F^PBj9@H4c@sjods{z5b3)6;3(OJEV1f1oLvzJg9^s7h2D!X_jowI9{mVr zm_A}Hp$^PpLJLjtKEe%h`)zmJ(2DeeEm(o_4XAQDifNZtpw-=fBIUOmNKN|~{7MP{ zmDeA!JmMZNcXBcgy*!jwq{@Lxdp+a}T+c}Xsmzn}WMAdY82M>|##!+aqs#>aN2N^Gq0XWVNFG_BOsDpeHqR#zIln$J#mGvDl3QZl7oDxkaMw% z@Ifz&YP|Z=hx*hjvGqG_A*J&>i+!t)*Q%znqh4w3Q{`bvv#1`Oo(!@7Y)e2jG8q)+Ok#g*zQf+e@8Ft{4CitMKFmsWcz#BgDn80mR_tS(e^C{m z4l-xiuCrLF$#GG7WHSFvP6Jdwu8059wK1^wB3^y)0QXQZ%t;)^-kp(QA^y|B>UQc*Rb!gNlHh&(WJ>Nl3`ukT<4^34pvAx4xX-+FT553*K5lcM<#mEv<#oK!ZFB|EwQ=awoQdy;2hofpZ_!+41x-lGM(Oj( zv_Jd+`LCf6tzdjh@{)aw%;zU(xef}#Nt9y#fRVGuc*<|KnsY5ePx6yzj?ijpVmxi49 z%3G)m=e3Lu;C>woQt!WoE;(LD*K+}AG{&4V$3DPjMKzR890&%9ZIT_=uHm?n`?x63 zhzcq_u=~Yq>Yv?$JyIi)a)~!TsOK?1ZN@uPj*AiH&ixM6B>@y7^zH(GrqgrJD8X|j z{GIpK(~Zczs1-1kg5-q0ZBjRS7ObTq#|m&yAAzCLnhBo-7an@NgN;f1sqXeWdbIYG z&^dgM4?}0urj!l%!NLVwG+d~1`vAI_uSK(ujUc7!65%u4j3>`7q30I0@GScuYP4=7 zwF@s`MPoeG&1j`T&v(;4>3(>+>OOW39S5>qyYS>H4K!HDFts!eEl#;$q~NJ%x60_> z_Tg;tolYFyCCC2$MO6CtQQ&h;Mft7uxXvh)93^VxY^O&V{bK2Dp$-l|@DacH8<5+^ zb665MjAZ^xMqd(6NlIk~0 z>HskX*U8cLXG`hL(&@CxBLDqXtN@ zxbIC|sc{*dSKK5=)zQ?pwH-7z_7S`Wukd(%BDTpb6__ul1?E)>jf<$FNk-a&Z+toh zRb^q~oJ(-NES&zAHQhhZVF8A)gJ+my;?>xf40%F z$aLDiUqSltfhq+=oJGSyxk)ptu7%=wYO>HLu-g3sr>?2Sq`DiJ;U%H@Pk&*I2cT#7 zD0XMdZ0dViR=PbRgT4n8klh^>3JO1fqR9foIbr|}nREv`3dfRu@DTWRU=rC?Yop`F zLu5TK6U`-)dEe@tq&$44)Xs~i>I>TZhnnXU>k&+rk8eO$+9l#&e!z7XtfDrS<^8J$myb6~lTKm4^0hs;BO_pxiTu**ITdrj_x`=-wSj%R9pi65qnd2peWJ=_s6wc+57q zxQXZYJHe9pdFCk=krjtk^6YJbz1ZT^S@uCok4?Vp!_2x9V6XBDu>3NMX)XTARopzq z1s|_~;QT0<``4B&`>xFe51cGcj6J~Km;11PMS<)`cRHKby@QQ234`OpJ>NW~FX&`V zfHw8*uy;fsR>Yl$n`XwaB6lA+NL1OqO+vP>U?@urO1 zp5cb4^X%)9P!?op$SkS?L2dF4SZnXe9A_?M#9u$wmxSwv(F^j}fJPW`;a z{2zSdWcz0E6*8(!C&rQ$8ySdvn#{zniYBq(=0v9M>>%3LG6Rg3{DEADi`?3UQm!k& zn^jz20-b>$`1n;a?BdpV2>n^a_F1iD%q1O+-iu&`>3#myxyeHJ(Ft6z7`lHeL-)H6 z>@etyo7Nl!WBClGP!`Lyq?1H*54{Iti&LO*YP{rzgk$F(DYEm=njzqhkZ3HgYGR%p|34lf7gpi{>8DHiFV8-c_uS0b7E5GdWd?~#5a6d z!T1pcpqS#pZYl0yaV3E)C4U7gPy2uY0TSkC{D!qhorN5^zN9eMfL%Q?l6~;V0`$8C zrFGuo2&NE%DjsJkKF68BLT4vk?V0XAC#H7WgB^MH z0+uY!0V9E1sJU+k8^6Vh)eJCV%j!(v-M|XCJ@hElzY1k~#*q+h`VwycatDutfo$WS zT26fbw!}s40>^yKSo=9)pLxHAg+)f53)A5^<*}T7TI|-FI@p;X%bgQHWm{{##moB_Gw(r_Of9OG)xUnr4t9(d zIOK**%XB|;ZIrM_A$4r>a-pYp`2c@FB4PbJhBBx3r=UZ!1kSrGhT1ESVD&y%_GjW? z_U6J?Hht+q=H-yVV9*m-?=u97db|0#M;gG{F@o8?7jgjU0^e@>WcggI+Bhoc^hwkgo(*z|3 zdEyovrafL{UE@WHYQyMOxHriUwIP|feRO%#PW03g?u}+s$l%a2)IZjNI>OH4ic=D; zOrK7=UtUmR$ZKl&UdT|fiL&a?vFB3#y_B)5n0%rXG9a8`x6xQlM3fM zf&ZXL-a;Qe+t!tmmM*3&H(T2CYC65|{}K26TtN3!=hI>4L4MBG(lt$8l=kp0&0ac$ zzD!zy;m4k${p%gHZ2vInn+J#Kc<>p#0B(ZUz=D>ptD`g$Xq{_ zB4v)yHQ_n#X|R>bT}5R7UWsn_UgYDvhti{;OVGfo2<@6)@-qke;f{swq<&RQj?Ojo zY~WZ5s`Di|Nhry-Z6N)Xm6U6%Bi-HfnzV1(Q}g)Wv^%bl3a6_`&ktOTBW+faPJdac z&yK#*150hBnY*^pk1O$vAkQX_CT6LSNV^poDoZbSdrIe=97%iMTS@niyG^fx6{U6C>v4Q(J9)p-lGX{m zU6;^!+P1Nc-plr-0+o&Q!1O6i`rJWkuNKgB*`qXT%WzUvPb7!i*>rM4q_E>Nl&a7f zdR;z_33wJDQHev>Cf54?fyyN}Y`C`+kGl7sofi%GP|P+GaI6@9alC|c;&TSP7q z+*uqK84*M?bHnk-AsbTmeM8>T7c|D{6aDwTj7aeYT~Uw5?a%V)BkrUkb-{Pr=bgiQqid&4q3B<-mI~uO)J09 z(bK~Bz4-`P$L=SKRSlRjFPc6Ns3i9H3Hfw)Vo}`y=`?e7THQLAV*b?9rLOKJlS84?WTnb7`v}((#cWEX_rq1lyjg?#I0ezV*Sh33+MF^pkYVpn1`8 z;|0fxp1AOA0DCU9^X2oW)djcMfaaquCFr0IkD;?99u5YRX@7hanvDp{HO}>Kjk`JQHjKjFCIf1zfIa>`E zBZ|K4$`-jj;s@$2#cP9W_%DwC!PLMCKFQRJl}<3>8a9e?z{7($r9PSFbtuv8Y5r__ z_gU`OGY5!^XyG%q4W#CdTz>BNQ0R8f;Nn|VD0=%#>|3#y6oY0kkD-EJC107z4SFdm zcw1;U)FFfrgD2dAqEGyq()1Ft%;s zUMv&*bfKEk^nZGIxot~n#34ndnmP}%0wtV7-ys4E%!b{OdtCaXV=$(;_;QmUtYWI2 zyIJSWD(IQEho2Rs!d`3{!&LIN3f|{Sq^CR<6rSBin|33r4YY>9J!a5FW#Dq!3bsXe za(C?fXrb5>LVs@(rDdBjm23q}P;uf;hxUhs?fv+!WJ4JK)&be`c2s#?53#v_`E6#V zTw&jCc=hffTs@x;M-t2UI=2BZFWVeOcP8Ov=1qFY+T;3RCec- zY&2JF_LVbp+|7?Vy&tUSN5jai26%k!Qmp*;MPmGH6z8Q@!mHo$7MRif_#zJkSoW{yL$?U8~)@An_GFSX~v}S@3XzdJ3EQ< zh6}uY^+i#I;cuK+xB?9(EakUud@rfCuf|NN0v%LXjn_@v`P%{m)w*#AKXcU`{Os?~ z*Gfn5s%By?X<9HC&6vwCyPL)7>1jZSpAlz~GYyS4PlLL9Qzbu^zr`Ip-tv}v`|#1P zG^nIC2mLB*#wP^Ckio}kIAhm4UiY32+;H)SwEh}6Nhyb`I`oITKh=^Cduj;ZuiBwY ze6i%+`C%~N##m9yU3cDd%`eGY&x@D$Wf2cjTX2xi~wLZG^ zOk*HF=ZU$%W?}p$rNt;2QkpvF@=G_{I;x=|>b{aIl!-r_bO5XTRpB7i_{UYmaij2gXR&FQ~%S z@?P%hId!~n>KlBlGT|GiSBb864#5RRJ7CtnJRY;W(f{aQ(We;T-E{ura*kASbEci) zlg`-kF6Od0%2J1C!{b4cHHsJInqZG&B_DP55YD!^V9)gJaLnZ-j9*vApJeg&lm0G7 zlRzW;e`9{}Hm`|0ayXIi>5h;Ld6$lZ2j%hhrRlt>;X`g}jyK_m2#;)OM5&2>GR$PqRz4JxSe+n^O=$=R=>)UCHq-|_T7E3Bf*dx)oMR}+wF1GnxMdbRfMDF zUj=6OOqDUs?{M|)cdW5!L9-btyywpCbnUbOwSf{@xpSEN2k=uv*~w znyS;?AJ+w*!w|t6HjF8~w}s;FDePI19DC@SfiJyWu&Cue_7&!(L)!z{VAo@`Vv+n?|og>Ig?EHZix3`k9 zr#6&E6?0#Nd?lIN!}#z_;jUi8xl|9~7S?X(zGWBj;$OSjhWj?C8owAX%)P{|C{=<@ zJD0%486(NOXbJP^@x@86c4Fa1SKL-Ujzv6=!~0h?*nk<0F!=Ew7JX_A>~b++-%Ur7 zOdr8n_&F74ELu!|znNp5%p`R6jA6d5Z?M_tBeyMjDB68l$_DExv5mYb6}&h{SKmp| zL`{)=3?uN)gEuhY!F?*eJ)ZetFnsly1HFHf(X4nRPN+_!1C!@4Yvn2~I_5ERIhtR( zdGLLVKaj^JOgH5gU2tXw)7O#8>U6k0?*P;+JWip?QDkkom8^$o)7_P&FhN%voR)q> ze~m2qqo7XSt(x@t{1q;+Cj}n*STdbn6M7`{8r%9k#Ii#LxX)`f*S=MkWry#e+x-PU zShN$lZRpG7`pVN(tsQJcqcWT1{}dEVV)1xsG_x;@6kRA##=&0J>9IWbXc$#R|UAB%z=rh&rG1OwhP-jv)Q1%KF~VU znx18}gARGnll(G>&mBjZp=-#0Y&@N(vDE%OhAu4+A^z$)3M-z^jUE?8uU@FJ-7hAP z$Gua0{EMCR$36m~SRL2T&7;fq8=&Nv615r@vZq7aQSMJTCg{gP(Wg6-{!MrBx=kuw zaSml~I>c1q8jVBLc=q3}F5FeVj&AkUB<1@FsCjZeEH6I`tBObPYs~i1(lQ=1WYZy97viQx6Ny4aH-xXeo4fY{rCkz<8k{59V~kNjbHRei(G#hl3lJH6&pokkEt#1IkHmto+e6O@6pAw z2vvBH^olPERtN2#R9q6Lh?kFQ($m=!=!W2?t%^DgM`kF}#VbegO6qf-1V40T5S#zIFQ(ZTvNV0VGzwI;p7!5ay8%$0$(KMExN_ZxLu zuVK{aRlM)7CN8@*9<9TM;p4lfafxF(*Q=+{AQii^UD}p3%$S zw0VhBO?;R6klWn&i__|uj(m|JWi;%gr$!F=>vA=Jd{-l0NK~LL z!%Q$Nv<1IaKI6aK9gTb5YJ>UeH=<@)#0LFn&=oqy$BoNy@P$eIUDa$%xqSf~kL;!) zPGRV0=ET>GuZFSL&8VZr6aRV3mNw;Hgn_SL^U{{>B6lJ0GWc^7PF64>ON;(g zQFuqRXOTHHKl_1(A^-S{WO-=%Hw3mEQ0HFWkfUIOQrsJ}jFSqtAT7Ac&CT7;pSkpm zACa$srNQnZ-PO}@dO;Mq9E%n8!T%&B9SoNmZpWt|T= zqjz)ogZqoPes|<>?y3o#NpuhPRoM>nrXA)pHl^a2*MYR>M>)owj ziptBSe4N*Fba!LOyXwQv1^IZ-Xe?Lv@g09Q>Kv*KQ=(*>7`(Qx3tC1`$9(wCkCRoV zAHVLSTSN$+HFQOFt#&+ESt2>IROleo_Xu*~+ue4G!$L@UD2=zm=QNV$XEmGVCoG zHr}yUhp*aMO4gbaD6UQvijnhNZxK$5KDBxi}@BRHQ)@~TV*|K+Vb>BdGDQm!@O-JF9W^IGjU2}VF?3SYfOJy>SfIOCNF|@%iPH-^cM8>9LN+bEva4mFZWRG z04!ZOmr6ew@{!rq_|RqrSxpaNqWvm3zI83XNoqu9Q}2T1QNUk=g4xEeh%M@>kme=0 zfSpyL@^LD)9i0T$zC&=_%}91_<5iL6z=3RfL>$YSGntt@8_J|km&je>$Zl9n#UBR; zQ0|@xzI@SAuxNUTkK@M@-yMZ_^V~U;k-wpzQxsRbGms=&#r)B+CqxI<6yX(jM^-#h zjrZI4l3ny)&4SyPGb=|8X3;r~8EPruuGBdgsPO`W7U_Y>IPdYr^KXNk>l%LU^*pWSEL77pm}PtsL9f@B~J^ z3c=pmNjPBQ9He2-E{`(@-owr+3@N_qN2sfX(&Id(pu@|tpVKp~z)?_v-Y9!3N z76)fdotfGcbBbJ<0ov-R?Bkg~(72K2KOh$FpBu=kyv*o^ zvL|yJodE2BGma`T!}C!yafezxynFu|FBfHi`=hBWcM6Xu2Sstu_Rg&%y-JqqcQ<2l*h7CvXJ zFB{r6h<&rRV&Tdg(P+0Ujqx{Se!^^WY^Ne~@(qDS-+E9wPMOWjO9OSE1z@Em2l2{r z5HI@>=K5%{sm8l$_(DZ6e6)#`YfWNZ)*0}}RhQ)tdyCyU#;`A|3iWlPh%b?Fwg+Bo`h@DVOKCQGfChT?uL z6MA7@fD>(QV5Dsd4yy_SwW|-HZLuo-IL=6{9m0#ZJCf=FQ%rc*g~M4l_TLnbrlY+m z$Eg}S$Hkz0h6?Fc+(jJbk7xA*>GtoDR6YDIex*>#T^T{X$s5T`e1t}y;mB-j1l?CY zM_<0nOI_4<R%_YO*Y<4UTJ z9H}W$4R0JRrZ=)m^g(!*%^+j8evPpQB_nq^&(k zY4=T%MD@FWe*rI^`+lx-ozLg}Heu48gPhH*Vydx1oz7QCWWL+}(IPd%auy)>Y}Zw0 z#B+Df?bT<8G{fo3{TFFhv^nc~ca3T&E3=oE^H^7_I4is@!#3eOcKae>zA59`jqUT8 z-bokI`O%jK7%MXauA0gBRI&@HR;=CJ2>x^3Mt!6YGsQ)pNz%s>+7FJ*z2F_Qop&7X zPf2C@qxkW6Y&g5~zLPCp+QjS{?OAfNE49C%$3_cY(3>yQ!O`*&Jy&#yDULqM5`QPN zD0$wKI&%ZOR6Wlw%DS+vuhrPqtb=T06YnbuI?5UPtzdEagr&Wj%Rb+_1gEw+FjX^s zcG`J5J0;&r>*iUoX6Mti$7m9}rkTW~gECpH^KeG<4A}?%p3lQsnx0-hfi@nWN+(a+ z%lsMd^2&ChnI%zlZDS^9F5y5MZuqd_ALg+O{3qi;TyX1?>cHZIgqUzn9KFb z{Ge^0w6X6-8|c}&idM`!&C(Kn2%h%)GAH{3tQ3FH2bZ?7T{656N$f0`j2KHV7X6|! z20HYxRvXP@lJud|L~8w;Qn8$7&hOM9-O+iBE!Gud3Y&gYyQm@f#y_jef4j4qXbV=g zqn>W=ctlLU!AB2Es2}ZyR8D8q2f zy|IDq#Ba}R>{)7si{SRb^ZNxnfRq1)&=Vl_c16KwWLtor) z$i6X>JF;ywm`v^^du)b-^utmZT0Du^lH>PK-zjTO8;>qLqjGpSRD z01Lyaz*D=249;h8IIfr&`S?KgghF`u!yNwnRmWRrrePnSvrTrJ2Sb7$uCQwuXSBo? zE7!VUE$>LOw;vAcOjd9Z>Y;yeGKBhM;v1#q&@|~UXoD5bx?0S2U1}hvSD%aC zi9SL9q7zu|z8?<@GvPKJf}NSk`277Q$WJ#!VbyU63Hl3Jyi3|H{4Z{``zyHWa;v*qAwK|D@9?FMt{HH>2MoZhSQTc^&k!W+ zISTPZ;zF?ybF`G#!wGMNsKdqywNov4c3LZBdDxK=&MVOC)KYx3C=;HXmnP#QhohLj z8SdO!1HZV1sFQ7oWiK{kJMS=B*Hwgm1H<5VTD(9#`yhD9#KDlW4!npOi+4H(KrK`i z+aBeB;Z9eK=lQHIlKv1QP=@}e4`6wzIu0Lk0IgkHz(x=PGcFrJZ?$Ojbi~?s5j0;?AC-9Lg4^tQ@NS+VUUECZ1-Ip5fP_Dt`St+D@w0Q>6AJz$nv$~7 zXn*S(*)uc&6lY$8=^yRIq%569~VQ zkTs4$;6K#{zE3;?|EWm{NB4h$JKbqa7HeSA?kDiB)ria(Aeg8)KxPgp3S;(i^pAHB zL=AsIz?Gj}st!Pl!z!Gl7EiZnT!U#F2H@YhY?L`HFU(8fS;)b~sIw`ER+i_|Uq56q z=6;%0P-7GrPpsfwCZq7g&@sCFxDwhpWPmfDx4LfrL8RmI2sY(d<9y*Z@M$=Lqr6X} zI)*{Hc{rr2-X!8)JGm4XhAWjPptac${Jdg_TUJMa@U<68c}QTf+y&U-vJX`4@?pud zSuD|37u%Kof!F?gQIY5hY>S`E<$M&!vvSHzak4Xp|ByyAwG4F9bS9ax@wiHN4Y=x$ z#2E!|AhlGBOAfQb)dys8NtP9?Gc(3HoBLt(ZUy$3KRqkw=EJpTCBPygaO?6HFv+G3 z6_T$)vHDsh*Tit9&uN^n>l-NZj_|0mNf5t018XPb;I5%NT;Td@dcUra>saUxX}42B zrgT2JX`_H^WwT+cRyDdVa>6Im*JI}S*CaV|1n!CEM_PkXXr<{%Y4>{2A5=x(QI<^l zQW6f`xxhvBN#f!4Zg5UJpO)=AMjORH!UwO_FwO1+)<*mS*YC+V_3a;M`Bwm{2jk(B z=T5JaGH47NUIAaQ$){^bAbGJy46qEp5=3RZ7$=tl(<(ba=ca4L1&iV%7Y&+!gB# zQjqft%siu^?7KD64|&j>wd}5lA&EaOKU%MDDdURUGd`3vY2mgU@sJj9rWu zL;3Sk*Fi37UL0nAorC2E*Wq_-FPyqQ9^=n##N|#_*wI}GJBTv-B>V=$A{Ee1KN^L6 z|I6Q18D>WJLE-5x*f~B0CT$M^#Vxw1sbfU%4&0^1rx)T2w-o;OmW3wcc9_#}mYb(# zh)cGP#VHwc@ZOFkSTuG7boP9OTif{D*v|nPb;S&0iz`vrKo|FV@qMQ^k|@${72Rx* zgB^P^aosm34F9i#l>Jbo9v|ljytE2oXxdeb8aWRCZ5_^%_$=%~o_A{X=@*QXt%3=H z4EQ-ug6RijqH<|0dQFN!Q)OvE{Nj(0;T8>!Cq4*NHz<&Ce}6;sY!7^@sD)lv9)o6s zCf~zTg|!c=quP~!18udSW zVSQ#8PTr)(53|{LuGSHA?+YR9q&H}-yA2w}0dOV%9aJox4ZVM-;fQzQIL#&mCwJ|| zyJb@Rj;03R3@gBK-)!)X(iE(`CWD%{Q?Wcz1^>l$lLNI8SoU`v=$~i>vy&@f{YFJ* zy~+#^xT(MKOe1SWEKQb*wL z#Ydp#=yUM!JxgXk{Q(IDmTbhX(V{WlRy-H2m^(ZF1r50uD3UXs!6iKv!&WfFpJU3w zZdeUg7u3VvNe^*r_7jk`$%9OT2yi>Nk_~^~4^~-M=v?<2qUj>$QYTNVl4&s(@hcB)|9MM>mbBQRZ% zPu%)W(rttBVBeG@iVcse+qUZxQJxn?@9IQ=_RaHnZ-pnmxv7L3Osm0g(M0^w=uFC& zMUiRy`R8m>DX|`}O)(;lYKo14Z#xXA!{%pj{PqvBh0ohtFNh<5wNuH^i)<)(K8O4` zbc*`A8DnbNY`X03F!nqpl5T3Pqz&2pE^k>bjkOmC@xf9OlrX^oM}b=n5QAgnz%U`m%tbApBnK z3Erz4z_me)ritu{^VXZ(oc#;HEN4Akqs6-+W|q-shwEYes5ow^@D0g(iTLj98#?8^ z02K7I=uA6T+SK}*On+yIBlaER9q41Rrh5-+dt`$CyAv?m{vDBU-^wXuG9nt7NberY zgrR;R^*^&m;QV76NUi-zZnsdV5^WYtFm%LWyDG`s%uC$U*KY;aK5Jo@-xa}m7ZEiI zE+S8oQ|YSZOwj*LA9nuq=koW7;PQ2Gi2qN6`e!bn^CAscUur5RbN3$@+*w0|^V+%h zXLaehVskRg+@enL&^uUlvz&AOA;yJ9FzUNtA%yRnOo{^Ei%Lf>p`QH~FeXDEO5;Ys z#p4>#_)-&R$ybohOHou-;y4V5>EOltBhmJdGamLo0?irCqF;`>R@O-;LDN6G_Fh#F z+-2jLyTf6+r}ZuMur(v+l<&czzqXuHRzA3nIY4CH`Pr;=9$Vaa90J9afjaRHhiRhHeh4~S%{jPzqKg0Jlgc6mxT$vs z#x!@)?JslT+fQZqeyW>W-|~d(^SQ$bF07^_<|)zG=Spz*z8x(&mR$G2ynu9M_Comj zo3uqEn~XH%dlGqaaPr4d7-1a@+aHc+X11NA?fqu>t@)B}=ehMFmCZOwrWOP~0gyGD zKP!}lf!Cjxoaa+JE=1=Z&D~K5a!;;8flm|kjJyXfWwW95q&yCd`wppEYB;rC2R0+m zzIc@i3A?+E;E10JH03%jZ3`dW1W5xQ=$cNw4dDpw48vm6?awQ+n!> zBD-$a11{nlXxD;gbX7A)zZ6TdgAOHhZ-^r`i~Gt=9vt9a{gGnPpWA5MCruXDdW5>I zBy@|3BoqGkj(R(HkVDztC?B6mW$b=%_xU`5pK1_{+oOWh6*!V+d4|^iQ{vWt`ODv# zZ)Lk@1meoNE$sP+Z1m7B=HGdHz}xo-3743{v%hY^9n00Cr#DZrOHY~v8;nJ?{*nz{ z+NCV&;90c=k5cJY_kL>kB)QI9MGn)&E{YyUW>Rj@8qITyAV6gv9BsEJHy7n_DlY?A zokKG;9y);WcLwR-%mmuVa;etM^Wb0fiH5p;vGVUa#Omk9WA>G=pf+<3`k(hf35!Ik zbXJ#p8#WGtO0>yN-3sa)Y6>^g-qJC)yt8P|a4vIl9$kJm0{fD~aGd{6mfhMxC&;R? zhf+`I%4%IqsnvzPPYEc^EwH-!(T=vo+@uS98@b?|D=_>?25gESWX@_Q>MpZQ^#1wV z^lpa)6C3iQVzb9Fdyl#7*O>EkljU|ec2R-v)}7^I#^q6Hnn&|5+QP3tj!d3BV@Ji;l~+JKAW&#?pNvNm>jgLsiNJ56q;K!S;#X( zWb)30EK4fyp0MuqPh&96axZz4LNtr|~*SZoZ7;o+qJ0 z=MS*!yF#1KITO1p3Upw!1^XGDLRHVZvx_dLm@e<4?D_96d3C3OSu^SS(Jq%jd^p8q z(-~CoUmsQb-OhFD%%DcA2D#*VC8DP`nrHZHP?z)ikS;01*3Xl`P(d54IGj&yzuU2C zzVFFp$zbll@H@1lZXq2fejiShOJE8==Pizlq%t29;lY$Hx@&PB_g|eeX=?dRJ>?X@ zS!x=$TtkOBP7HxE4M$F^WitEyNDR}1ykOR?lTcDF2Of`)p|_xbrt!Iv!AA&Dpr|N)UCfwGaxwCPV72ne^xGS@^+Ff^E#DwE9*g{l+N> z``okeElI`9-19hQ-WrUonTUI%u3(u;7&}|7B%J!d1kKm-`=^of7?;$Gu5Wy}gvFy- z&vsSxDA<7=NdmOqo5s#u?WYkh+)!(k0ON`}@!`v2Dt6S7IXGLvSZhUmbt#@G@b0;L zS<&nl&+T7*bR^H~a$^c>EBQUmC=^cnNOq4_$ELgoFm+@Z7Vm6_XOC?0VqG>fHvPkl zet#3KbMObH(VD2XjdxW<`=QwXxk|kMYTV}K=>4vnXBwTSK`$q4ab zDXap=k;j<#k_?O(Ylovwtwp6RS~T!v2~B8`78dMkh8qgR_+-2%(31CEIEE8j`&F@z#S-{vq%S% zp8*&WIgy2qtRySlY_o!%M%}BmmJf8K-xWcBAM4Y0v6H7iYG|oPW&Tr46 z<)?@An%#b@TGKJyz`ztbZ{;P-c=!M}jBBKSLS!kik-$-!(P&UTp577}qh8WQmh(E7 zqeu5(=*L(VAw3gJ{~aMOXWxKzcWc=<*8nX1SIyG)&Bb$8a=5_i6#DdplF=X1=r&DL z;eewWR}?&nov&=7N(PqD>hHvAF3b>iH2j1!gM1F(<1B5fyTKXs&W1<-b})-Myia0m zB9=K;z*OClOtyO~-YLE*s9G+D=hT*BhPfNA3AIM$xY6vyUKQcnvs$<#HAj&C=so^% za$sjy<}=ev!MNLS8a*2ojE7ZK(QkP(?9Y}IPTjPVo%$Mu*R=yNeD@&iyk#cr5gM|4 z&hkS0igRfBstKFg1=v8A;gT`$@Z{4mLL0>|IJECGS(rAKyyZCD_3AN9T`E9<{1hyk zH&SS#5=+15w4v#!N$AqNHs|rN=I1n}~Y$^l^{O4zkU0O1$dwps0PqA;#s`(@(xS%x}9H ztQ)+?tw>u0(o38uyF8ld58a>!vi2-)j5chZrNV8>jbiTvt66LBXg0M?1_x$0Qlrx2 zd@dlAoi0Adj#$oO5gupg(#mI?Vy_?7pRiXJ zE-qj{hFa*RfAgq~e=mJKat*s&Bgs+Ut99EGcJTefAa>5-EL}B} zP7IZ0G1Ka4LNU)eT78OE4HQ%J1(TWj;}YuXmrGX}A7k+PiFdvw&TA zX;9bhcZg2=3(nyN^b1)L>c>YgqZBQ|#Z$Ka`XG!^ZpUW&ajUWp}pbGZv@Fg1wi}!|#gOwS}js z%f?Rn%Vm1Kt9V$w@#&@Y)k00?sn^3I4fB}eF+Crb9xmlm(yq8OwYi_VKbTQgK6w&k^{>hrig2MC$Q6U zakYB#w@Jq1H?&gj40lPxhFyv>rlYQ%p@Q1Yw94o-bF5Qfai5DR@#trxmh0C4o_2(K z_K36FVH$MmgmUI$)WW(A=dfwKv-M*_IlCIt4q54??PjGn6q!PBGIO1!!5s2@1-9WP;5B|dGd{rR z`}lg!8eluUb3tY(NwLE ztLHy!){QKlD`S$_%oKbwK*A@K^PRqlo>?be|MJZdKBn`X6BDnZk;PM3*HKGOO{IzD z!vWg)OrGsGd&I8!d@kEO`UlfZ>ZaKn{?gh=byn;t!@!a6-43pRLY{R#YAA@M zKlLIb_k@!6(o(#~RMGqOA(VgA38xmn;l581;hTnCZ1F06_Ve5!bZ|Y5A4_uK_Zcs= z+s5dY3sb4m=~wi^)npVu=7ep!OQF+gF6^wTfCal9@!vf|^0dK=p1d#wJ{upBDO3va zo;;VR>5aDkUehy^q;SLH2coAN`pCZtYlyXqFN~Nmnp|%wWi7k--u_sMk6rKLfJY^c zJs{xzbMEIPW_sXHEhF?-=GnLM;q1gG5q)*%A%qP-fH&Rv{Oz8PG^?);(teb2kq(>b zPsubob50I?cjvvIahhzH_D&pkVg&nV;snnN&cizEpXlGFjkbx|*wa16N|DcFUlQjT z0Zn7csr~}`IItHgvm}LM_nVn`5xru-bAe@i z@W_ou_$NMdc?xM_djC#|PEeckm7~30i=yV(I9X=!_}fQ{nsgboxU2F0@4% zlTAzBf@1h>P?#PA6~gf>{qYn2Z`(sg^*eB{<_^*}e)l(Xz9%t|pH1(j|AcT+3P!C} zVD=;R+1knRuuOg|CcO&gwp+dtTs=MoJ+mX&3?F%VO1zqVcdlk`$|Lc@y%ji*--%cr zdB>&7O=ADFT5vt|0G)q```PNjsy;>1h|LFyR8t1ljr)N{o|{lQ#sVU`l&MWu1bJ$` zfTc^?;v@-1b-yqW51PuVH%ioBFUUj7f5pt~o*R5p(ZP=U>v8J~N{q!`aZK?M3E%My zN>}d0bGo&x`i?$R%&4MYvmNp9@Pl-il_~RmK8l5DPDGz)Ep*hB8L)1dqafjjKVIf@ zW!jG(L+9N~Xwj6<^Wlb}r>QPW+3UfEKe1-|@)d~H(Ktgr0DYYn(Z0+6Bqg?-CC)CQ zp8S1qYo-p>@{?uhMWHbGK^b4ZKPH&3>BAM}zJ_TB8o?wqfVwj!s@GkLMh-tO#%E6Pc0!ST`Lya2_%Ce~cx8Qsq^_?(&L@!J{{IBul;a^(@qoa+|2Fv% zH3w2pg_9(0Asq2^h6klmaH3DIMk%6)tiMaZcA6RJ{d^9g3OVGYXaOu)sf0_U6s!)` zwUJ%UaeOz~h71+%gd4u+In973Zo6$D1pldltBYb_Ue;6MYjqgvX73}D4mpA(?`t{q z_$F9jFl_4f<~nqg;1&OWH@~b2e&c09Lf;PpKZg+N7YAcgI1;mTKNyHcajENT1hH|H z6wG-+{Ea`5(SfJPgFh?4MRg5KZNExt+mFDAJ_oY>lr5KiKLS)|sEWpI@dDH4W29eY zqA1DOkl%^_BHQlY6vgd139A;4hW8GZFkbkLt1oba8nKDsd$Wa8*H4AJSv)`E`F8Hh zY+tBI-a-6@1)273YW&1(w+jkPM zM+_er#c`hXp;iTM>)^=ZwU9nr9m>7SMVxDuAoK2ba`LhR%yK^hTINSZ&+jM^6+I*J zchd^E7=2N2XKAk0x ziMLCL_2@FzgLubewz!t7Q7|XU#X#NLMxm$-ASGmbc0ikCyBD}73dvSg|TlIa*BImL2u7l z-Y;;K%d05jZiR-xtrz>rl;+>Wa91#d9V;Z8loilKNdVs-0rd|#17h}1u!hTngB{Z3 zX}2yO+9U%<(G}k9?SM5s?qHtPDQGSAqKV_Q;e=8>*r5b2Te=jqtka;F?|_dTuUOag zSP{GxC5ckdAFDm;wW8-*x-ePpK2g{HL%78W(7bvtX_9Uc$XMip%GoXu)L!Pg@Efd9 zmgKDNMUuCB)nMzNG~(;{kZ3O$gpEaJ z#(9`**2=veh^1q>L6WL8AZq-3K|pI`i9zZiSe2Iz$DVEk-``RqtzSz;p0u2l-U$WO z`yMdound+5NJIa$WYTx&33p#L8vuha9qP$5jErV}g||OU@!#_~7+ofVGjc{@%YkM1?NlBacxFS!Uw#M!9~B{F z%U5oi(yocyuvT z7EoCJ?J8ty>7ePu6bveg!0|QFF!5|0F0L@anR70}f3L4Y&0l?3+JZvI7D-`E(;NID zqk)FuCs9a0@$aB0+`9V$nP7QK#%E@q0s2gy(fj+;(T@w2C~$_p~e zE)lHD0kF#%>w|~EjPi{rD3aiMucbkPDc|LOr-_eOUZKxdDPq#^aGYKa5b`;jk;d`6wwAo#OKQzhQg=+-rWT+Wagl9c^^3}{T_`J{J>7@32c~k3*^4tg0t3lz~@gblqkqy@Sl@V+ms1* z9cMwscL=VS8eGg5rgeS>^X<06X5+E6 z$0-Aj+w{;=75UH>JRk2b+#)(KUmMQtI|~2ohT*DJeEE1~8P+wZqeXBtUAFZv37ydb zy~^2w_Oc6*y{Z{t#uo0bH;~IG%Hd*R7O8!k%lX~D1rF=dNL1&3tL_F4PVE<{IK{Z6K`HpFKv!Q-42^Ox3g6=UpAw#HC7Xcx6SB!HqXhg?a0;f+ zvcUBAnRufl12#T($C&w>QL&F_m{(Up0;b|zi=}ws-)wxG*9Z@@hQp;`A@;r*gC)*2 z@MJl_w-GA}o?O5*}^m zdGOw*kY|34$mp8k?sqvhFPKt35&p*co@9Z!-ZYo*{DV*sG!`Ek4z{i(7@70>Z#B~yQqj&{G&E5!W zZ>Q7qI{QUaMx~R(=9wUQIu~E=jKNm?Ottf5F(CIj2sbI={ViQ^F|ds4?g|lXZhQhx zWvZyr+d!AoX}FiqLPU8*i(WL=lbs0xu$P;HJ5D=L^$pS>DHlZ#4P=1N?dQ}@(g;GO zHE5%jHmhpNAXd?e)Rs8XjJuj(ZvGH1em+5W9*ltI^hj*m*#?ib)ZyWy`BatPv0eMV zl_`Hy zTXv-3#&1qhDW1LY8O^4BSVjlOB(Y$53HS-AG<123;Qd2_Mw`XhA7LK(F?tC)=Den> z%df+)#^pSFG6L&ty15T6zoCEScb+e@mBt@CLqknx!p!3L>lVHIRl4O~ADt?evZ2A?i2p9-Xr2BzNyjE4}FA z$1=B_61)l*QUyf;n>uYfHGZbeCY$sU+lP96-6g#T#$bc|ZL zm5N#;gq|y+2NP#ek$fTzxwH?4w*4j%H$C7~=5g$KS`5m~0|M>*C?c=s4BaNLK>5oE z7IUGJMDKCt%=b#MsYBQ4mDYALO7aNNef^vJGkBIpWtp+FyJygH_g<>Y^IaEhZDGBY zJJ{^g4V>S}2@o%>&C0TDV9U>Rvgt(?oUtDuk>^LUhq|+==RgE`&pUR6?qT%dx##rf z;tG1RyM=_^oXnl(IgEc7-rz+2VmyTW9KPcDSye$QH1gF37N%%PU*7x=&Ws+0zs7{q z6IOC8y4SDsss@%?d0#iJ2)HtW_tGddepc+oJ~G@40>KD!sCVnq97K6 zT+1E$rtCagy7;1F-4gs~+hFN4aEVOaww;bM`DpbmH;LQxvX1JR9Ahif+Q~krbhO*T za|mClgMQgkoLpx@GA^4ihx`2Z+t_|V|G6-DU$1FJXID}O`7TndIG;Y2TZb#RG?0$* z$sqT2IS4&{=odppTx=0VS7jy90onTy61@*@>>Q#WHh!gnw`5q}hdMI%_DHy_Q319a zqlt@K3w$}*LJCw1;rCa5UVXL#Q$CvU`NJ3zvQ0!5FOtI8TrYI+9*%dIH2$m01G%=5 zTIb!N3V5#~LQgO!G-;8w3P9O$m* z-0Q}}b3Uv5Cfo*}c?MDWt17tmjxJ_CjHe|YeUOy#92$gU!06s7nq|F5lzJ%yChadk ze=3i}B#<0htP9z3pTPFj5@>lC4s{&|2wX9Pllr%a=7$SJZOayNru!*)Mho$uUL*B> za05q{$b-pa-k-ZH8%i?E$(CU?Q0q9Bm|Kj%j6Go_pyxKIyi{bp86mXA>amq}^;9xH z!ybMu9fLAIp1=nuV|eyy2*U5q0iX9BAoXG@@#VP!TPT08?tO}@EZhdMZ~64y`M<i>kc>z@^oQJm!hiJlsi*P+D0-_umL2QHzL=XOijrwXhV!95$Bb3F)NB#U6UJYm2 zmXepVWU1oRH)NHB48&O7p;ME-a8nZnwAn}u7tJ@tIehQ!iDaap?3y8%FR}rX)_!n3 zGy{GlPNtIzRpGSFRYCblMY_Y5KcBAHMEBo03KP4uA?eXeh+FoBME|hFYu&^79%O{o zp6&xMyLUb;ev52cwM|xJO$z6V+e6BqyEsi}IOg zmnxnSdy|1N3!=56v9y10uxPjVZSpm87l=p^iSvwr*w}bM%f>V=WxqX6on{W(%XFaq zK@PFzL}aed1k6|}0`omb=~k06?#~Aq%s*US_qJ;^9Z()a@1BVvyDDbG_ZS_#uu%!# zkEw&>|E`nu7Kz-j*d;W`mY0G{?;#^Ce~FgO91SriPeV|zDoOp?3E{m1Fz31|t$1<^ zzL_m23xXPPKHHB=n>u0bf5@FOYZD!A@P(v{*I-cm3>`Vv7JObT6)0wJA@WL>>kOq@ zKKOzRpFC37KS;yX<_YdueE}~m2Y!D0ZdJ8f4Qss)qSWAj zFt%EWt`lA1oMq$4=$@6-t2Ps3mTG0Wk0(lR#K_fY4wE*lQ*Mqz(?{Qhf_jLKp zfcWt}OFc?P@4(5Yw=Wy8BE{D&=qF}pY z3hrqug{sB&SYsT6`Fq8L)-Sb%SG-q2dz~~+A0E%2yQPFj^VVQx!UL*f8HpPFxvtTf ze@AF^AUqbv`> zV9zONPuPolBr2d{Ycve3sI(0EriyXyOGSQ>6X^8IU!dmQ9jKfxjYaYKINZzvzmo&p zmM<&GqL;E5B`=Hn{hVQV?oIkwG(}=C))0o8M?l|`AQ?ee8OD)CU!+cTy?Ov;$)1w7gB=O})-un!%F{-X1*NoaY^ z3tzJ)xC$mrZ|;72$u*q~_cx%DkruGwY8w?QY0=|yiJ-VplWC>%JIv3sdFRYhEIF8r zbF3tBn9LE*XTP+t*<=HJD^W+MDT~?W&;0je;dbWzdOLBrp$LV!m+3XhcaWyAhLssC zM*H&9T(iPJ-IS0H*dLgVeS?cI;h-)ACLV*7JWUo`x;pHg4{R2FmZi~8La@roQ?3t=^sST_GJ&RmBZf3Otk5bqdfuq&q-d-oUcs7 zcb*2cX!IIB12_b8ZfIkQ>`zXzAp>o{%%OFUZI~~AA0~TeHu}uu;LY83TJ%B+huIro zc5e|4c~Zlj9d?3j-@|ji#t({IgD%i_K|v;OdvppMkvd!;!MwtqKhR&7ZBsG*+BfoEvd_F|0JKxi-R%>FidX8B@0p@^$st3C*{4xB1+Scyb}j*)K=UCUk?6#SAtw$)r68qP8Jb_b=gsuBEbsL|ts#z<2xt zM}X7=ac;WYd3xVsh?>oaMv;CNcXdMsew;$k@X}LyL%W*Y;+Y_~Ml9gU-%i3ePd(X- z2{QHL%g?d8Io`On!iLRM?Bvq(%E2(o3fkMZHB|xxxvLX_xjrSo}zj z4Ik~oVlDSD-(})ZQm28-O;^!5|81o&?zPe{Sxm6MzNPNl_2WPdWpS2UA+2fg1O4s8 zSg!tjvU1@9k=pbMa_ev`Jf66Or7OzQgJ$yu%rDWgpYL1wPD5SjHM%L$UDAUGRGQ z7(S(s2mhP7?30`ZZ7GOEvrJ7E{5zWddQeMeCv?Nj6(iW!Dm|v$(M2_jcW}{TPg(iD z$fjk3PE3VuM)~LwXt1pwrgtY%lXtwIUH$>o*(^cx$WS~h;fKS@irE{NiOj_QJ2&1< z76b2;(<+`Z^*!MYIINan4+S6L!P;jmf5s;KGH*AC@8tyFU(CU28?^D8%2v|2=>m<* z1f2ZoGMgMA$1crTMs)sp;0xzC_VU>RD46zyUX?$fbx&$dX6yfBgFt&rn8 zH9}Stc^h6|Sj|*p&vIM3>X>=C83t>YLEtAhLD=68PT`gdHFi448s1g2i-2B_jdYd$K7y9qwc+!cU({LDTwD(XVv^Zsr&fQSO;T zcC9WW+ZP`t&pYk7lSfXHv6rR@Dt^BZ1-A}}K6gCf7IZ6fp%-FA-uW@4Bm6p7AEiyE z9+HDpe9FBZQNlUKQ@B9_1!W&9xj);I1*h9ya{p<|!y9o45N4%;rRsF>pY=)*c4UL# zmiKG!-g`?bxtz~lC>Ig0&#t6MT>)}R{}Wxl^NuVx?d4)#tOxNgJ%amTV*u|@qq@I0 zfx?p`b#pQb!Q#INQXUivqt#)=SXn)?jSZ>kj7*Qk@8h^ySGbboSqTnJ1)sYBgU z-N`KdY23pK1(M-*pNt(DB09PIChs>B!^GcP;dDU1$nwSoQCJq=iS2qI@_$iJOCl;!E682?fWx{l)UB{Ku7XO2)z!4AEN5^w_Xwm z`7G(y0APUy?S$6%TNS1~C(rIyakb;0SY4NS3}#9*NlX7?;_J0iux;vMxb$9uYk9B< zJeG!$rPCA0z`!Jt{qBXNY3+VNrI-u-7?~n!->f9ao*73Y(*6=`IKb%(evqZqj9V^J z<|f%k(C6zez$Di?2#f^cTkZ|MKGVsg)MNDR)%^45FK8=LBq`xJ*KI|FBn@OFEuleam-?OGzy5kW zuk)O9pX>U3-f!JTHnZas)1=tIXy?A*I~Z8AfimLkNn4Tfo!h&}LGK9eH<5z3{{^vu z|Ek!`4;Eyw?*^~;P%1ktHO931iV=ef8}j7kGv?Z(V#esoM)t$kXqY9J_b#N0P0^W%r%e5Tm=o?4iSH%utyznyV=?-I?1_ zZIK23_|U-oDcg%z?kj*@7ljrNGc;K7o+-cA%kI;7$;yB2W4&vhvs-^?Fvo{unQbKz z?3UQisO`Rno$>PlQyr=Fr7!Su07jvQwBnr$gD*L+)hg*f85YRR9W#e(~-xojkz}lknwYMKbO4 z0`mAk0$+OV4)Rl1hP~RL!8d67jlp|=kv7jGbiacbKH02HKRlR3&-I8=Lw$L0O78^0 z_hO`^WF9;2wggKj3sKJ}_lU38IkINUL*hOEIvb@_MLfd&nd%g2a`GGE)Q#DUUPus> zUtY#;7Un#uBNy0I*Jj*3CjoR{-T_HvuDS92Dz09Z4ubTgs?m7yMiTwxvxsLCFc}?^&+^0-BcaB&97X-bt>%P=5I>oj zc*rCiRzCdCDt5{WEA8-5t6WcY+!+*rG_4F&FUmusy)wL$Vc#*c>;t<&JjA?(#F_A;B z;JB9`onEE@(IJ2Nj>lwRf%hnxCgFoea$kcq=Nztndx;hO9)tFOGC+lUzWU6HV>D+V z2{Y1#TYM+xcwYpYs>x8HAGzekg(H*pZMf= zGwYr<@*WQChd>b(`oYN&Dn>K`L@SuFxGW5#kBP+!HQ4jxD!HZ8izkP*Ne9&f?WKL_ zAlHK~qZ4V~d_|CHw*nuH7$(U+9v2xVm3fcOf@4#P$dARd!Rfpo#GDQS*By%>;r2D2 zf~yGc-ENV=&>iW0RAS9e7%IF0JEfdSlusgg zalRkIk2&G4Tn0{;T}PLi>ts*S0hD$r#fqUt@Z{BYOqw}|ZIro;-QUiD{st==BlHa8 zpOujD(kalI<_Ok0E%1xu>MVNymHF#EigPp*pwOQ{_d|Q~w>%0K2KbYYFXV_JtpF({ z3aWBXh|E9ESuidI)vxb!a~LyRuz4dhv*Ii+xzU3TM=Obk+G1G#)*2pOI{^oO6QbPE z24Zhdkra;?aCj(%{!C=aM>QqfJ3fQft_Xv~cu_iKp$2)oTOT_uipe^&1ibst3T_5O zk%=GQL6Tb+oJxEHJvFj;G1!2b&$_@^FO0$aT?1rG$PaSt$^;UK=I}uzg@qRo+>m?!A0U@=;SUz|Bmdl8k=dVc;cyZF3FmT9X3KC&S6Ni7c5qwU^obei$>~0?t3V1pGJH(&*p% zuy{a-2Jv!8>yq>Ql3aiaSse8Bvr^$0|QUOsmM%iDevmCK9B zXuBXJUTp&Wa{yNQaNY6DL|*dCrTF`)EY@nXBwa#^Zh2x)Lu1u=k5WW1DJ7p-sqTis zU)SM4+$Bu=Qh>L)_kyQ;2Dx6U9z1c?gpTc!bb3cV-T&nb9b2DF(`ta+cvDMG9Pj4| z^?B0s*4b2Onm+Y>CIbmsb% zRADdYbhI_51~x_PA^$ket7(D;;Ro4Wd9CREY86p=Ek-2e%=k_U1`O~-1U|1 zV2WX1hls}D1aha6+xb39!}(6DXv|a2<8ic$7HC$Yz4g_dEUkJgvdL5r z1n!f9D;SfT7439Y#xT)UDI?uW63KnN4l*xgC9(J(igByov%zDz_|b1EOfT)Ckq8k0s@9zMaR+Yka7=B?xsXhjVy+nbjuOZ%dH!Zc# zMd{!_xPkSgIl?z_^^0X_#JwM{UKWQZr_0f%Av>aC)Jhh1$WYU5saT_yg1g*h8H;^E zIQaDsabL$>*VlJazs#9Xef|LN1(!{nbEgLNy|_$yJGUFO4@P~9_4HD}EOwswKlBOH z#=G9>tlFI=WM^v|9shF(jjVps{jN#0YT;qF{On9Rq2n9g*0@^g-)I& zPrbHUphP<7)a;r3e`h!`DV{Kzd!hhUeU`NB%i1!HA`5yt8r+=$$%&r*j~S3YI6}UzZ?Sdo~B>Zn;ZS z6LvyNwE><9)1Yfo&8g$45|y?2k5*sX1*bk-WrkM@Qnabq0Jo_MvxT6wQAAm<@ka$t>(=XzR~uw02o66%kHE*_9!6Ud!k9W{9)>-ErX-S9q1#M$!v1q zKg{7Rp(XjDcz@CxjF)w$#kOvA-5z;5nE0IJ$p+G`A*X5NlmQe9xPbGp1pNkfvU}FH zp}Ef#GD+bc71*W#GR^W-UM~iHZt2oZ!_jp88LlV1U6@)-2%w3k3==3nm3d+2!wA~# zWX3Ji@b%Pke0FmB1`gVFPUfR`!BGPWs@NO-ycQl4x`O=R$Ut;OKCPRE`@DWWbXRx{*ANZ`X z7OS+R99ONsh~HWX=H({hj;@JlE!j*|hP(;+Q-vM~QXc3Mf)fBLzGkW{Tqu zP%|whIz|^!yx$S7bG@QtA(GJlJp#4%`m(BZB5Xm|0DJSf2geGxrr%1!>5z#u{<~mI z+%J35gY_@ijKen6Kzud*cHNa8OR1w5w&&72d-kD_hcF%4QAic$dT?%&7T%cdH(qlJ zP&>Eaxewl#lOLCQQGaADZ*}HbeEn-BZOQq>RBNxL2e?d0n}7}atrf1Y&Uc`eP7^DP zJ|E`_0~Pckvk0rU0T|ylqh1?zY2q4ROw~4k8#~KUU^<@{sjwBVe}Bb1)GT1)VQ32I6s&j8*C!J_fNsv9WIb8*-k2>I#{J%C+u@s zXUS~2K%@g!lH=gO%NyxJoy>kR#Uqatxbcbppg*~I&6Dg1&>}v4VeDU(cf|O*1aa`< zGvm`a7wYF0@HMdmYSThipY7na?o1}y0UUcmk@~3u$un(#)G7TFvth(?21Rl4WljzB1{={?4@XkIAbwjmi+&%n``2K#ymyw=_zIs#04KZsnhdDY-HgSy9!J$qtZJO?Id2=YQ^XmT7H zGj{INNSt$W5)3;|#S6+!%o!1mg)N>)%Bwz*zs)nTz%8G!&y|5&bTK~G#$>`eD{kkZ zg(-J8vC$V!;>9{MEa^DR#(r}rCd-_0QSf!Nb<2b+omW}k{}xi)9XF8Q=tAl~Qg)vI zFLsq>4rBJ3yC*9zA%f=KAn|Mhew`Lb@>=iX0na}ee5W1N&OBjlmD}h8JyG0|p3T~Z zTaf@Gjy2MFjok7JF5lU7k>p7JU^Uj%5yjGpWJS+mvgMcvy0{iIf!{4L+HEpD`q2IMnBq$*md5Uw{*rsYmtKb1| zN0>Kuqzl3J+nPi|PoC(gSduyAvr*`DIrGut2FVk;%^sMQ4_&*(Xi)@9?*7NeyZJR_ z$&886Qlr9n1k6HJbv0rcZU|9N>WNa?WGGQnzyilsRDa9m3-%qrK#qZVu5TXh@E66J z)YVKE$C|5hxWRhmtHU(~j%~%OM3S)vgl_#|bw}i&=&uRrbmp;rx1OPoPY0C^e}YAa zUX#=%mE>*S19H;w29bV!n#M>xpnqmP$7>3XG)JnAz11rM3zw=B`Hcqf`NTJ(@i>BG zH64Lwu5)BQeu~0NIT|V-2$y?2>G-uonsaCZ4)Sc%@_u4u1W$ZMPo_ZV~ z-l2>#oaSu{ylr`NM2bjlTL#?EfnWZinr(Kf2j_4uXCN+uWM4LU$zCEAnNes!Dwtb& zCfwuRpYv+|B=5I4G2=b*#8ht&9x_TGvZuNrP_mX}XEx#5xvOAMOVY~Ra1y;6Kg#=f zekHT{c`zI5?}Mtrv6wbJ7mLq@vau6IkpDv)nf=D3_^BjL=bQ}Q7X{h31vOZZ;6vXp z4Z?v}DpbtQmVDcpfhxX{czaqG1}ae=`nI2LE*HYMVI{1b z_8u3Mi_yGRSJHFCl5@t};5Iq#5xt*#HeTXf@22Z%!>B6Sty+qn{`0AE%U{;=V>>3+ zN>?n-Dy7+`u_SDLAY*)DGS+nnQ{nz9ym9Xg{aJhnE#@_1UQ#KBJnLlExHPf_o6E_> z_*0m-UyOd+9cD55;vEw%r9_tBI*5C&arY&a9Q<$d$MUC9XEr+OF|Xt+>Fh9i9Zcf6M!QV6VOX91^+d; z;qdGr>V9dEZC<2K$FB-Oulf~w*)D@lNu62otu%_PxG7d)@oN#;EVP?e1m#lM)%kcT zKaT#bY{E}}U7231?O3&A4k7jDDwOP&&`js+6>%lCn4WbTuL;=V(#1fJhaIQt%EA?M zCDW-)#ALeaIJa-n(W3e38Wrtq0V#4Zr!)Fe=+XD~RCrP{-SzG%mD16rC$21nkZ@W0 zQMDEgRU8Rh=Z=4Qp*ZWp6AXNAM_q6CA?Nao9yjsEgv+=* zBh&KC;WWBC%ZZJOoQ(bEJs70F4lga0pf7(2P_s83EIR97CIe-y%yfPne^|Q!^~XLk6_)Ez zBk>@-dR9eQJ@JaD_$QRnx=AC#_RymKowU|Okv>L2dUUt|3#U)T`#S#Ur>KIio?9v^ zrcSRo^TD7Z>!&I0EM{n(XDM#^avmN3+R=C6m#LxcH*EeaLajHNkUs;_^ssdV{;G(g zUKYvxRgD_-gi$7a+sZ?kj?L^<#}snv`ZMaG@`-+p3Zeh>Dyh_}S9A^asF>iVMiP>~ zBF~ti3l!(Dtu5w!&3Of^-C8bh=Qjbp%4686^8{&^#YZ;d`d6OerZOricaEB3IbS8z z2GUYT>FI_mbb4A9ipZa5)2x;-4n|X%nbU7$O#dAEs9_};=ilI1&fzRQwwD^d52CAY z=+ZZd$`x)8`PA*oaw?R)xMb30*7RNRD~_ueu6$=9e=%XEyrybb?-t|W}36Sf6@A z%9-6Ib-Zf~qh?7?9C?As^YR(#!{V@Q_c>y-P8F_*S}}fg-+14~qS=5YiTM4*B0PU$ zkjbdDp$#YbWZ|$g*}E`#<(godu0@ zTu&cKGI(;d7XMWFQrG?o)b=Llb)9KJm(O1cKNdYjVW0aHA1LoxB@Q&5gvM z=TYRot^&EVxD6vU1nFMEhxk@Yj($#3f#TA8n2h4AKsDzpT^)@_`%9UxhK1~AF)?h^ zK2LT>81hO>rqUj7W6HvJa`SC1u3UBlvu4JizIZb)zBGW%*V#&QtTeGcMU_2jb&mGW z%Va)3(52;1{*X$eopfgPN;YO&E6%vbdHhz!ac+!8dXbUCjf*nKu(de6j{S?J)9=s) zCzBY1;!yrhy(KV5>nJOKD3I0>O-f~qsZmrJtNww3&9Rpl%~k8kf6uQ{_kdcOTEV%H zcHa8`EDC$-^Kn%AES*&FjVV}liQm3uHlJzErk4*z;>ay89NB(`2aatdYR^*K{KApP zot8+f)on~RIfLr2ed#{)OX!sqh^_Go)P7Agy|iOK(LZ&Y*I>gZIjgRrke@WMziW+y zKRH*u(s9xrS;msS5**kai>iSW>G7-YNk)ApwwBvKuwW*drmlm={}#{#O~X8;Q9`v! zgm9OuBfUH44JkBOgwsBk5Z}!lqrH^Je)nEUg9S!$QlK?f{&9nkb@P}9J#uVKu^e?d zs7HrEgie0!MpnnIVBV}&L!CQkP<443y7u2^FF_&sF+U07tT$qG&J0|C@)Vu-!iw}{ zs9=xOU6gv;NTj(uY3Fefx@kc+_ONHM^)QRYsU9@dQ4T^f!|=h$ND_B$A$_mD61sHu z(5wd`WMj!*^uLjc8~#aibE&B`{Ffm#fBa4^{@O`53|8Q+r~}N2rr8X0t%aXmY)(jF z5dY9(1v)b!2-95&Gu3?x{q{nM^S#BhmWz|=w*Ak^$b)+3^|r}W)bT5sK5;6&qLM|w zckkp69Ox}OV*iIZDmg^X{@XU%ohcRvRgb7{J)OoJA1T-9mn<@nP?A0rRO z&^plwbRF}>6^jPQ!mwI=kSIhIMg6e-dn|5R^BD&$-s=UQDj46_QECm(h0Tel!&pwaot!&f3d+v3VW$8Ec_R z^4~}wa?Nu59Qc);W>QXq7M@`16vgSVST)96Rpj0w>d>Gk826ry#BVbD@P?@%)*rfw zog)Hl)2~S=AgGH85X&BZos3WSpE%w$@7YZjXEeQ@Jsaop|loA}4r!lBbQnGWqB$y(G=S+DND3xIjoZm{cdYjo-MaP(9WpV62o&dEtx`XkUrb8$B)}qI{ zXRO7;axx871xKpu?eG_~d~;yw4550JWRgen^B$URjP$J@ZjYwTX49ROHwn$ME{q4$S1{ zBk@K-yxMDRX#Kf>9Sc59w3mHmPqtJt%`fWMwiJ#5w)zc2AKYSXw2i4$@(YwGI*Z*G zuHboT2ON#A<^7Q#u}s>mj&F1%AyJRd%+07M=NRy?Pdl2l4~x?B#A&QOx5}8$g^EUFSUNf@;ybo7pAhb>P{AE&HF>prlO#mW;e&kJy~V$4WoaB;s!C!@IDTu@Z+xB@%g?s& zW8U&pu~%h`eO%Lq?H|^nX>B$CuveI z2s3o&GM7L3z`{%w+GRbNURztkkJGUuLc8u;1}`k&zmYtG7h7J?2U{sRYTU%@9(DX3 zS4GH|({njbbT#?4^BZnSS%G@lfqyw>2B}|vidm{*j7Odp zl4ONFWI|*XZ7tbCcZ*wMw$E=i^-ePLxc4E`aNC~n=e%bF$82$-Loi!uk)vx&|4r{$s%GU-E>xa%Kru*knV}^;GD#GU0#hKErHZ zuS#Em4$5!PB95N7VBAuaR5qO@yOsRlqG>cdXmBQ$51rsm>Uq#_RfhvxF2D|VF-%qv zhxb+LFfpQ<^*QYg!jpSp@2`d6q&$~rDd7ujcoull3KH|X5!SEFhXY^dL-a2ZtL#Ib z%+=B=&YIVF>JD*4DbF;hb+S_WU5stbL5pH`?Tync=r;T=fr4TUwMYa!LT2|O?SCZmN3uyaKOBxcRQKFu1!NLN8LcfD@ne<0)C zvR00~xiH_;pUGW&84MQ{P|1)MloOr=3HdMiyS_w$(g5mr zI>QG&DKf2%H?jkucr*u23HK6#av7NLI1}dQR}$6U12B7aA(;L+4q8fmP&L{OGq)^*ub!RY z6Jib#`S)?YP$|rAi^9*2k9hBYBtx+J26W&lK%rm`i8h%4pNOc{YmOE8P?hVsSL8$f zI}Pw(z|BV#gMe+U0(z2T)4ZJmWqr2r({dU(D5b&Jj6D!nvK`Dg$CedOnVNoC4#ieg zWbOPR;x71<@T20fG~y9iEgQ*Edgb6<0VS8;y(7Z$o)FBjw%v)sPK|=bBI`Nb?5lns?W)cJ+IJBlaB{_ zR)F!20+@1g0#V&@2PW!U!(XF*T$VHfE_07U`rD(hMmPgHH~fX%FTz#{E9GE*jvIEC z6rynVR8puF!H$!U%)58fKxSMI<_*n+{dGd{_Gc^4Poam5CFH`l4<=RzvzNi>-~m{* zF%l+LkC0bs{*a_L6*jtCTU|6WvpPO^4%{tm^_ujn`S3$v%KL3FZ) zd-e{<)AEPBVq=K9^p-Tdj)0W7>98)Oo@9IX!+|0fSe3Jt*dKWcM>|$R=fzCW`Lv3> z+^qzsr-wu9yh!4(Ar%Im%UI3pI}7(c3Q<@66C=FZ9ENKb!5^1=_~~Ktqf<_v1%)A2{$E^VL7XD%8Z6R2%|r=-Kejm2^BUUL7|2VczmG}O)9s* z_qpOU^2ZYVlDUO0J#I#g^X=$YyMt_d*p!M3pAMk7?srBoSl>R<~J07{}omO`Jp(j^DcU z)h93a3J^nuYGT-P4Ee&dQ8JeYPmYVzZ$o1A{_7#KX|^i;cWEDemLG@KH9lnLK`FGj zwUsI_)~3rXMd-o9vp`?2iW#|MLB?lZ!|ddhZ1BGZy7#tNg+ym04H;U787hY5s>=;& zXs`+l96k;A-6Kf*A}2Ea&^^5T>KxtiLY1A-@SZUl3P5e|(enSEO(6>1pIE`u?^rdl zVCu153nYq$$STXl^lQ@q8=tg?rnk+d6K7JYdqe{-)IaA7e7TQ%rBWHGf5zmbgE?4g zHj%(>&zU1u(P;SEofbvyqL?N~GXhKKwvO%W!WrqVuS-J=_5;%RwZ1;?G+%u~NLza`whbyjJZ-Z*87LQ@5SR zJJqvk{=Y}K%$s6!+aF@|wSi$y7~;rP2F^KYa@m$ge6RO-g7Vb`U5RkJ+BA1eF&jn*DR;47z8?OFMd_@r1(<kva)UZ@rR3yaD)*Zn>CqA>Kr92 zJ`2Eur*`D?MDPYLWF zQ6v@?Kw_h9qmjryKe-!!C6o-4)xx)EH zZ^4Fl1-iw@Sd%tKw%%7lS$4N-=hkaA8D#@ zGI;6#27yOLaD|>CJ`3mZBaT@?x|9U2zRh*RO5cze>uLCqV@X%c*-S;xZi7(46OjBv z22Lk-qtvjOoGBLBRW|B3Z7jo|QB6)U~(x^TQ80>=&2>9sSCG(R;1 z^6j(1J5>(OUM_$$ZdFkF2_RoV63!-nB{fHiiEG>{DkoOQ_$--1&!1RGcV`qqu&WD6 z{>b^`gN{P)kw_jHO_?$ zDQVa;*9PFJI$VFSkseqY1f%hR^wsAJ9J^>eu0mPpXjxAly)=XrpL!CqHXC$KFDFV{ z6*&j^6f3)Tu2x9}3RY)`IDNkRDEsW92Fd<~8ZUe6u!z9+jO1_H(`w4_!6T z%t)sqCxhvByHI#P77YrgjJXWQEV!Bf9>wo$pgqZ=mRI_p(>^mdYIou=9ZoW&t{)9a z-GX$S!ZA-TaCwZqO=l@%ng)@17m1ds5N&x?ikFhw*po*U=!~wV)J~F5GVWf0c7;NG zJFALTws80U^u_ea{IgWy))1LG!5xfX*?7k`qpYCLL3LwNS6Gj%gP^IjTyDw1gKL1Dgixf1xra%X~@>Ks!i3sppX z;NYDb( z#K1D1XnfKlTNiQ}lpAFr{dN{}S5cat`8dq)()o;A_j@oen>SKp(FeF-coVs*cnIFU z;ED^L)^r{HN6h{1;;h(XSSx#$v+t(vhS7=)kMzibh(?l7&MZGd&YF6$ece(V>_){>V#L* zoghIV31Y?az;D(*vRQLk#rp3#WbW)FX1CND_?6mAO6G7m@PS7-HEAa~I=`P--uMsB zB^6Nq_36MK6at0&iS+K6EFDc3r`h^m_#`5RfA8KJYV+QdWiNlmhJSl8;72zqjLyfk zee>Y=9~*8~T8bZ%j?jM<^2B!g0I~DTC7tXVDtExK;z^xwMH7?Gx4X$_P5gvtbk$K( zBjgI5-qD0ZalPG913LQn zDfUFopxwbcXrW5=`-dY7 zKT`W`;V7zX1(UeEU0mKUopSRP*|KIIXuEVWo;P;U4NI&bO~4OZnsbPFivT`(G6Uqp ze2LMLa8z%)R%U4e)cl_$E|LGs_PpjCbg63O+rgh#ES&g_JYF=N?$dIB?wR9QxiSxL34F!a=XzxA z<|tTM?9SB6NrSWV3^+HOLQ*S~7|q@luuWEvhU5g0SsbrRVy*&dIKg!%c6?`3`jbht zeiwfEUPbs^mp@zG6|P0BBCBP%tdOxfjX%>(l#67rVzLfg{V^5zb~~_OxgL>Ow~k!D z-phD(&!Y}$v%q1hH9sdwi&R8zp{o-{XuZlVx}x_k>l3n+#-FPro{u+?m;USUY;_r| za6d;ExQ3%|lLh{>+zOAs_A=QSJIMUVUqoBn0BjtkVdzUX8M_w=c@v+Yb9*+&K2~5K z9s9?|ipxPsz9O{yezEgp~BikYP|iB>QdP7RG?d+fx~Xng~_$mS$Wb-DwO><(Lftqg+G zBhT^3z7!I?eswRIRPR93$$HhR}cIO3lNkp*@U(Z8;zCW{m z$P^}727wQEwyzm^PcEk1L02<18h@dPoKO>?eJ|hhR+;G$`Qf?H5!**>B?3u`(r2>Z ziV)0PWecI*@oYl;0#?;-3#d=t4buO%!e`E3ARdr}37jA3V%1_OKDm_)-r5` zGlUxZPlE(nKsKG<0~T{iNJU2gmDLg_0k2H(NcIBU&M_Zr+-~EW_}3UZo`va}KS8a^GcpyN;t znrVq_`kj;Pc;RkX_LR$ll-6PRw>+@n^4Sqx>3Erq!yLi8u(=@|rY=24HcPAH=nhRX zP_mCC{|RARTHg`1p%VP9Y0tEW9@+f8v;r*^5E1d5U zv4z^?QG6t6d#1sA^Xw8vO6L|09?!f%l`$~<;8cDeBr9j}uSM4)wOG#%xs8)hRVij&mLlEK+rs{x{~zdwd6RGVV#&Md zN8qz(7g;7e8C{$ng86v>TMf9qmAygc8 zDX76nW)wvI@rAs(dNklNAI^3NSxxvc1mb%WAo8vws9dfh$GS?$r&*`LdipKcI`0{f(qhi|OWCKIp;I1OGud~sG^H#z2cjww-Cg|`o_fW3O#VPD1&8I+d*<*@=-X>bVW*YF}vFV?tMYW|G*W1x)krP$E(jPmIHJNOkQw7&dGs z9f5g3>125GFAE+@yF%lxNUAG)iiB}<_=sJ>P#WKe%Lisdv3CMoYb=2zt`n;Du8v$g zu@o09S_b0XmvQsDB#4gaV)r#E@?X1mz^6jkfzde#eUnF2*4Sgki8Dm@ zmo(@{=90o6UL@!D6(V^np9q{^0*{B(P(HVtRNq_#zfQSB8Tm&H4_d=Ziy7Q~9=qOR{kI=M;HYU#RoxOnyfJkgEDHg*a)NmLT?w~inm=m8^&^GV{* zN#Gl@66G>NK`e#veZ2x<{TZLK@<+LBGqaNZQfgw0$y;J}Z8}IeJSIMBh2U549i6lz zu;6(aR85->CmTZ{W-r$(Dyw7Kq?O>|WCjk%Yy}bb49kF&ZQz<92&(AM&h&PMj}kMu z4sRB`=5}M}pR~fZge=(TApzA&@-Ua34gYik!Qi@xRlxS|P!&;*b3b}OpLj4%605{h zZ;`N(Hq7DGE!=K39+pjv1|go2m7)w_=_pYb(7$(cU!nU zmV&-tWT7p2QWOSO2k*c zWkc0}!%^pt5c4mXdHUxZjQ#cl_mL5zwuzE$c?=s7Yf1JC){s!oDsprW!xY9#5YIAC z?ASXO?VE*YSYjko0}V{8TELtO-$L-{q|MBemS(o=XB%d4yflxK-^k`8$4Hcu1#?+< z6a1NA%3s~^3zt27##*imVW$18CL+Jqu%;gjh#$|E^~s1QS8@x8z>p1kyKQG(gSgH} z;6e;~y_Q{BG?xer?1f{CWZ?DKS$fXoDEarnnKI{vL8H@}J;%*|qlC6#-R?$yLPP+n zibSD~%SB@LQ-@shjV8|dy8IsoXGrd7ndQ8Pmc+htB35qs$7m?8qdP6P;f&wOjPS}Y z%=NS&-1w!7H!bTnx!jx0%Lr~`75bWpYeo^dY8nsC)v9Fev0-8wvX`%XQVeDJNAcBs z0mi=G099fiF&>{Rng2!s3!m6vkG};!wBDUWFOXtx9xP*4o=PUMM~29UopHoHM;_cp z4+gbRsU%A-q`U zPvpe+Ub4Q}hvWb6#$c@oQabYr+D*Ps-u?N*q`V7doEMKX)4zO1L$gQB+B^Nsz|dDV zY!~dB~kU+ z$<#L2GfoS+j&0LavcyvX>JpydI({1o3n34^$}SlQa+Xl2yJC&elRAYYC7dl7deo_MM$cL%Q=s+Fzx<% zp09ZI6jN*`je~3~nPxx4*8OcJt(pt@PfqS7g{pkgFfh(^Z|=bcmx++;5Xj3ud>&&H zzmt6ZIjp*YJMMp^P6CXj$nyj-)S9UWC#rR6hEWjb)BegDukau(ORknbD_TyTE!xIx z&%TbE3)FbWE!xpnK$PRhFT@A?^uebphTk(g2i{k6Y~HOII69^aLIb%(NNxt{+x&z1 zq571uNckUWZypWx`^Nv1HEV<-N{d2S#>~84_kCTjN-9LFgtVtc8!9PUEg`Zmp~aFd zvR0U}g!FDtNsIO@Eu^%nB;Wb`{X4(kIp6b}bLQW<@AI0Od%3RX^YKgs`@)qF(R&wu zZ!{5|R4Ib(*J@}Cns=KO%iRNx8zV$nS@vQL$-1I+BOG?TyDr|7&Pn{$`C@N{Ig(6_ zQR2#fn&JrtshBCzc$}J755MmSt9>_K1C8W~)t@%D)6nCyAfmXD=$m1w=*H9u;5S#D9`$9b zGwWB;n52hva9S4nKP`d8+N0p#z7F;}+=XkVYhk`f7pBasfffi4yOR*5=6ImbNG;Bus-i&q%UHa<$Q6A<;_>GF?;z;qVTPRvERwwkaVBzj%wVJF(Mw5Ip!yuN zzbb-c*M4 zEl3p)tnQ$JEetMuHwn|YEm-r-3O+Rkuyt4~XnDkf&-gr$0WJJ?dLrfa z>*DD8D_GVRf>%YCG0jUIAC%bP_qX>YSuf*odxK1MioJn&uT>C?ZQ)}=kA#0;_5@m% z9)diHhx_MlJ*2bw*ixg9@{=W8;;FlEW#a`jSm%cQ`7`m#h*o^u8izS|reeZxhUav+;LLB68Mi@n@$jXZ})_ zlMQji%|-I~YK<(HSa1v89gl?KGfTleS)v~pm4wMU85r+fg}rKF5)OVC-j~(IU1e*r z+&U4;1`Sca>o!z1ZbMtM^41AIw}uiR|H}h zR^XKTozy2SmPpARg_{E%G`p^bft;pTcwK}0oY@L*B=yUlp8c@o zjRFv<7c_3`0Wk7VtPYe*!_wD|=<_2Pk34#X|4fdF?|liuXTyWhzc>-!by=b6bqnkW zKMI$#g3*zij>fr5u&ApIa@9`3fe2d+DAh*o<)^XDMz*^B)JY5r-T^3Z!q|XZF#6s} z^-~J)i)0V!yQYm7cMZap$5F(`@E?S6H^6$sTiEUK7euE#a8BZ46f&)lF)J7c{BL8` zEs5sB+z=g-2SNAb3Gri{-*ifwGU^ro!+8_(aQMV7k~u6)T(f5c=c79v)^sff>B&b# zPb9zn;Am-dbSiu8H4dcWvch)|9heAi zt|B;P?Jm~TvP5Rb0ttUR7DrSs!-S{|n5+jlZm&1m;Q(~Z-;L7yX5qPxpBVW=!e|`s zgpmnxXvwwXDaA(2cpD==dPW^Co|EtyXH3L(8~?$1V_AIL76J2)8lm!S58SRZT=J}? zxc?f9(PoDkS2y7?-o1Pp#m!IIs{>P@wB#++i|kSO+(^8E)`-$2Sr1Z^##N0z7>LO{ z7tv|u>G)>ZMT{J|2wqJ~k}#KU;L?4wxxx8);+kI>z_>|r2f0L0`I82*Z=^9sF$!k+ ziy*;8h|I?%lu4P4FD0(OJvXy)O2&O`Dr^I`{R%Gn5P?P)ri1h7Qs@o2P1?T}3d!LTQyd^!Fyq~=AL4) zMs*6TJ`*S6`Mbirdxmgmy_$Hb? zTn*)8wu@au?}PHD^J1Ikt>Oa>m&7B-@2L9fSA&ZFS(tMmuFC0IwfM}xG-`b(6r~eK ziR*Hk$X45LRXcf&K$cX(Z=-3Fyz6P$RT(69^*E066{E!23iS|t@QZkV&~s53S1bOR zn^WDCpi*6>c}sL&X)DNhd*QXEiq$8sYhy>zR6M*(io2Zr7^Z*wg+;fmIFGZ(aiPOL zQXJAM_N>W9H-|jY+B>7brRXYr4{X7!`X_OGk_zZ$YQcX$#^LuRVW6&)4#CeP{j+`} zIJ)=`_>7R_BAlMXtpbJ%4LpY9R3>xlOWEqen`UU5Iw;!Q@JHf>ETJkn&1iaGOrP0R zi)$vzir@O2rRtkB#8Vu$K+`H4{CDgZ-RpN87D{UV=hsbvMK_CSV{3vq*GQxK@7fyL zXzO4b zm_~HL(!^!r0;70Qtlez<{C0&nzep%n**i-7ILB7pWg1RB{;VY1UaN{5D?WnUA4zol zy}P2I7yV-U*;66VuoIe^6snyYbATyur9*9=;<(L6q2!qi9&$BAr{Ffcwpy7xJ8B-D z9xX|+Di=|sZwUzfGF-xu4t#GOTzz@wrfU1y*j`+RrFy70a4};c=hL)c-E~- zk$0aD-9z5Mf2VV>IYowutIW{in<9?dI2Yr1tE&U@tEyW_bd}D7KG>TYkKx9_D9a5K z*0LG7 zcH7rDOzVcJ=T?dR$FcbM?0B(`t35kVSR;~!D$%5YBh^m~&x%NHVb#?Xiz>Ce5#oRm zBOtnOJEnZRfQIh2;40BJt3_mrLxpEWMH8OFg|bE(T3`&)6ZqnjU+1BEFo=F=M6#qxejyeGOV$p=1-j#onQ?||ftGscCk$P5&f9ahF^ zv3a6jI+2ofdp@?u{3kB`u7sV7uhB_mxar3)<_|N<$T$*`}SUv`9^v@9Qo$&}pSj(Yu;8*%c zb{i;c_C+Z@XyR9T%0x9};Wc?|~&F?9paG9!I-(H~3{S{|J0G7%L&Y=->oI=txE z1>H-}QQVsiS6#KyuJe5b=f5TzByWw zFZM<>QCFwhI#wV~S9=Bliv8gB;j8HN0Yh>)$xYPhJ4fub?P#@AzhcddYt16YF-H^~ z{kd}Hq+{YOE05r-k>%nMzrTu|mmd))ugw76_EAu{W+Vzdvn<`OpU#*hgv9wHJ~*u_5+# z3_xr|NTr^nCw<}gPbgdRP`r2eB-mE&3zvM1;r_8c=x&S0Gkq69;Nc;5Jy9$^J6cw3 zz3_`@s!<7?sxA}V+V_N>Gb)Giw`4_9#VKM>V~MtO=`0bM7B24o6e#v<8c{6_QV_3| z@Z-A9eWqITt3`Y%d$7Nx2;aP7Kyk`f@wT*5IwfX-XuP{67JqDp0Op|R*{l&*fAlAH z+WWCerMsFwN!|c)cIW9D&GC4RI#nE6_lBl5R{~jQ52L|Q{KBJC+!Ry>*Sj9l-*=yj zOhU_Pcg!1c=^YlehI(L_lch-O@N<#DDs7OzTmZA{1I63M7(f|5V%|mx7h=i@I{a9ki1*A3B>C2&c8|x9F_a{_ z9(GK;b*lyX6iPwi`57eWuPu!?QWu>cUMT(+qb5#%I}PLOmgD;?+u#&9f%3>y35u4!OdWpVuzb&krw4%D^${)CF znjn4|aItcljWu=^c+o9Ecfhz#MJzMfA0nRCf}4c%vSFVRs$IJ)=0{wBjq_Sy!G~oi zt?@~uekcwn3qC_*k1gKKy#Z3KOL5Bii{O22iTHERFma>8Ga#vjV7^(S`j3}+)fP(! zjBwdq89mz)T^Lok8WswvAC@9tBMU-Q6jA(a1{}im_{=1i_W?BgIOs+vVZ8#c!jz)E0iWAT4G2Pvr|GygaK)~%W6|9_1y z5HgZy@P9qN$KYMKJ-rSlQ9azD{+d2DuBW$VzN4>x9>n-E9xhNc!`b9IGv)MHHZXBN z+u|LcCY(Lbmv67=_P^Q{U66&*Q1n*{S5~7*|@7Z^>87?Xn739Lg-+)r3T5k@RVq>QvfqAucRaYJftNGKET@S!Nkp;*#3F} z+Rd6K^lvO=Ok1Y0wozZet4ETz$Xf;0`%B?m_#axe*`KnFP67+(Xh!{X4Qo(e&1P0q zv#Vn+k@JxfKaYnL+Pp~td)9z`oUw#8DNJKsJ0H`o{ta|V)Lb%Wx;MEnJB;T z3YaTX1gzR)UDDv5N3&@sEeL&0OP`HGwaF9kkd8FAuM4NYynCqg4RxrfsDsOX-(Z93 zUwG{Lo#a1qgnX_V0{tICN1p;#xJco@keRqCMSyp2FU3V?&OzYlK*;=$K(b;EDV5D7 zO(&L<5y^K14GpK;AUqCzKS83 zthO2rXL{oNg(B8f;VOwXoJ195OrgnB7Q#nfhXjF&(0xNK|KXb9%;g<&aNk!S_UUhi zhD`@BZ+i$18-EZLt=z!(>p}P)a|~jS3y|Puj9C|obB@H5*h7@$Kb}JlEVdyVx+l|= z%`)_&%1*krvzN>h45z~nETR?dhp4VQA)}PIU%`sk=c?pe-oHFz{Z#l|5`M!*+aG zz*edbLE^zzPNYSEZWMo>9fkt`&33wkMOuLJyLs^njRH=&=Fr z3XIWxH(||)WZ?>90nZ|Ji=fe1h10p9$(>xL&As$qMGb^^$?Mw}NUC)bHHmV9kw>DT zrpXAhQ$pbEhduDd<`nty{R&&Myr0d>pN2}(6a#!$AbvUnxfbUk@mD#_m*lQ8opK-> zHW56c597VO7;Ijdi>@0IAcSj$ezk6hN^g{C)+D)wEjRJo7#?vL8pArxI?6t~8w7$z ze~?>s5+*wQqzqL7`;~@pMpg&(|C55@i!Tx`fgDno#&b8GUvs;Kc26=+3>S&Npw+fxfpiyYLXrpL~D@&f?L`gpFhe ze-gdp6-0I4r~pk+hk#)w@FOV}5#qi};i#F2ADT??+V^7WeYb#4%XFnh zi!-rasuV*C(vf!EW$j{OS<{K~tP3}nfj)iqZE6g&LU9kde^Q;8Z2!b)He?aI_(*cZ zZys54@Ch9gTS=!@L{PuAiL`3G3LJ1mhz{9Cew1${@nJ!%Tt+(J8g+Q?yo1g>Uj|Do z^da*&LpRyP!Ij-7z_98lbdKv{6YS&Ip`H%mj-T>)L3j`6H%^F@-W@faRu@MuB1ni+!zHv^qAZ#{I;H~$F?>X-w zKGnE~YJUo__M|#}<=jZ}3^Ul5)%KvV=?FBLVa)KLkiZxp7v7MvB3x$ZT%tY zMhw`l(Z|Bs1eXruW8AjW>=z?v!LMu8!us(P;+LO>|2U04*sG}qps)2)5z)MgF(cKvaXlB`b+LGtLE)JNtVvB$~VyXMhh1f>||0$ z1~5`@vzh%n6PTuk0iogX8lm5m0&-3-jkq}`kmPfHWS+`Rvdl`0C|{UNhYF8T+Oquzjo&B24_Du!lR$yv z&3whw4bKpG^d4c&e-0y+(eqLF*a=NrZBzAr;;L1PYyPfZ}N{)x#9>yeZ_W<~m= zN0ayaYFTbfC~c2hMpvCVO0PeeDXfhe5MKXxg}Ja!f%+djN{wG?!2Va?>4pSrxUZfC z&Rw#!uD_T}yj8_|OW5Y^hq9=G?rW-=H3sre$wK=zJ!tW_W{2G!PF&aMldXly=Dl-^@^_akjbD^a zjD;!WK$0{KuP!FH+mDl)!?Q`)UQH7J&6cEcoy4>wlU=m$CA;mZKKU8BiZSt8!tR?p zmfT4eQmt1<=?N>zeKc$)9j$nY+~Cb8-I@@J_E*>VE z_OLW@WDIqf(@Q7U<@rJYm8(oMT>;>fGFkee$;uKNU8oTWn&Pvj7@ zJBX20v%_NLJutY54+roF=<>9p8T#! ztg7C#C&yZlgVj?w`KHm_zbGXxP0fs3?5WP(9X*UYRvCd?AE%<_>NA)!APc6dHuPGd z9xYXsA={a|Z1=UN?8>Knv~HV<&D$ODYWFUxCt)eXj$TiX_uuBezW9$5CST=>PBZjP zq&q#FX+hs!)g~n!Us=;tn@E6vhTz7$Jc)!lM7X!)Csz0V#PsLwXs&iDBj^ltlrYK=39 z{J$dRP~RHzGtL2)j|&0Ylyl(kwHx<4#Ng_|P#mtli>?=3B8Ts!ktqGLME+3h-pnbt>%IPA{sp5mKB@YBzS0VcpG= zbEFA5alVr7oZ3iTeaC}t@>_hc<~};*T|teFYGh!xHsh>W&kAm2ie z8A45NU8^?e)?0!iTEd9W@8M2oF5G(>2)jkiBO6P>_W!QKGOG%J!Cde@ z6a$myEP|ntI`G}QoK6fbp!XAvapK`YxO4Rb1h{Fl`%hnBi$~93hj``GuWUIz^2m^` z*MGw%9Nb9mTOJ^H6+Y2I?UD59vd?6TcRI{kKtbo%cdENhK!3^C(SNCrXhQI2vNfep z_(ooqZ5S~btP`J8EAAyFR&%J_A#1wF(UFdtaupr_UdE6!7x1+6FgEmygTQC5s?{vr zE;MWZfN$cvF6kFyK2oZhwhV5Tiba%oaDKinIAFI_6<^bd9r-;R-(D3m^||ogI!Moe9n+~ zho6jwsIB{H!TVwK^MO08>CkbuK-!a)_SWKtrH|wcY?Zhge+`=6#iPn+G^ncNyFB4i z3DUNC5dB~Qx6GGt8(b%G2B$Z}9glAMBVh;q_Cd z?6<1>!o2Jl$QaxWu6+^kerFQI>^}*r63qfnqPq>q@kNLK*5F*uhD$za$xZYd#~r`) zmOSe}P5a6|(cIVBU?iy(ZVaoS8>RAu!=DZ#@%w<}&vL?w1&cBF@f-gE?~P%TwhVZm^Mu%q_vnOq4OD7fG!nf|1g<^}V*HVpC_27=iGqsQQL zN$x>d_-wr2WJO9w%_8p({6{=ZTe(*Y+qjO^?Oe?*cPl?TE7B;hLa|^eE60mrciLvL z4-2!c79V_KrLugdAmM@|*AcLqTj;xqE2wx%K21m?tAFTH*ZH;Z++3nPHT(q2{!NB` z$L&D#(-3V}b!EP-bhlbGeIr{?^%WCWe#9f&a0&;SGi1~?>Z2IN~c5XI$zlIZnR+a zk$C2MiyV1c-axN5cG5tr0Xp&4YS8N13hNg6L2tu*wqX8ymh4m~%3GdN&B72me*F~+ z8#kk}^fs*PT8=BN&7p1_Lg3N)l1$5W_~VB+rXAdd+=G{}YVJK48D9jAqh#=lpAJ6X zJ`ok>Pvn;G(Bk@9G`WPQDNrnsu%*i{z);6La{H6sdhtC&^Y-!q`X|WP+rz`eQb(vq`{f_F?<;&12qfR0-HJyjAwnIng%NL*i`HeCG3jxg|Kwv8<0Cu+-r(%%U1>XvXG)})~6_Z&<* zREU@4O7VPcE+)jD!rj4-sNwH2YI6QEwcQ_XerwlP3%SIZQw`td16W>%<=0yv|Mo++ zNvVbnNR^zQv!ghZb5Y#w3sIc%5YJN8UP{n^njt&uZZfD>yT7?)0OJ^VBQWJJFnit{tb=L>Tn4i4wv9K?{nx-@rOUN=Kw$RvoTNiPcDp}DbY^)*Fotx0!Mif zVB?$%;SP4h@!F&6bV2wi%-aUT> zx#)*DwX_|RbhpCx?dTsL-gC zrl-f#;*m|Xa)vYvwP>SGkrD2gIStLtl(?s-HMn9Y4Nk>r7u&KsfbrH;W}-(Kq1~23 z80M@a>4ygKRRUYBruix{3!hk^^kru>b@#&h;j?&dN8|;!M}-R_TsK)=I{BNwG4h+x z``|@h(JmLk;fed0FX`z*HC0o7yOWJje1E@%msy0>qe;O3v)IS#-&Zfmf5L<)ceYqM z+*>EOI`<0Od9H!izs^SZESyhlJ|yz&Mig0&Nvh+0GyTYOYnQV0wHwc06WGL`vNTjE z-5A8$1m1Jqva^X*r!1Bpa!kgi{l5c;b z+-k?0JgXO;y9EB0ih`o8ross!iv{<-m@|_*zLA2T!+Gfg)J&8AfS27MYqfKP1#_i; z7WsB6+0t8gzE#d-p5Vw1D}ipHouDG;yx`g6UcS4d#Ic?ata|Hi{=W_1`G)#S1+IB& z{0#1*|Y-&0q{u6nW7^40#&R&h$#1Y)~!{G%%l3e>k9Aq{^n30&+71kLNinCUCk znT{bn{wjd|}6Ji_aBJXxQ-xNk~1a=52*vbQLVwDS-ZuzL=Jsfxh=8YtOs|%)o*; zVc7C{?E1q|MESvWBBSp|_H`(d1U634ml(*-%bZT`d$h3o>deWioCs=H#-m=}tBA{% zEU0%XgSXCk;G6k{bSGPrir{VR3XK5xp(TJ-hQmOH%Y_$F1Y8=+=yThT^!%5Turv5K zSWY-amg!iN9}Vinh-ZnG5xeGoDl6TZSbOMR@~u>`~S+$J+@8%fO~hJ3c$#to+VakOkdmsq$^ zz-vimi!5!%Z7(}$;&AaqV6aj;WnhT=!EKJ6Z?<$805v}___ z`~#%z`4keLcamzIJ|^_B`j6Q_I@yE+m2|~%9-WtyLJV%K1cNix^zhq$GG)qSviGov zv2ax(flq|MYmo;CEu=}`gF(q70hqCAaHb~>TGQO&;7w()FY(1T;Y_^bp()`CNOIEs zS|PS96|8!$!hN4!_^PRh>C1Dlqx>AkZZ5|8xgV(Ge-j~dxd1-3S+Uy6OG*B%;q?39 zDq0vck{<3=p-HD-(I1c1AiY{23Z6frkt4KV>fE`YAgLYB&{_+0>1K%E6^n5Vd06dU ziWk-{WSKQ%Schgc#(Bq4OpJBJ3FA6oldLp(_9&8UoEk4VW89^-0wKtx?u0o@=i(!Y z9=N2G!OHYvDrH$sy*qp9k&rq}yjO@1))%7L?pk)I_7djR*m-PS!x;MP(hs74UYBl3 zXGv}60DD7nrgpj#MgDwRP7bUTv*&`4mWO4LovVw=}S77!{(AU&p7h%g%aKKL``u0dZ4hY z=bTVnr_FQzK8aWVY!)x$gAFGUCUPfASuQ!Hmi&@CLT-O8Aq_)u@ax_waI-A~wF8#$ z1f{{YvyASs%wj$*n9k@38rhwH260ctAbu3RLEShR2s2RvST_Q?-^YSoY!28CiopBv zG|p3P92a_Ale4K`2z#uL!1RTuVXB!Fr~mB-7D_y5Wxr~LyIa2r7cPougobgk>2I>wr?6){JEReDBnxMb`(=b z(nFv2$id6r3@~%61$~zs(C8Zg4T;tzYULX^-l@+`zOBx^og~ZsyjUaLzEX?KS)59e zf9RuK^dMZG+X3R!1#FqFfN?DR!e+f6g~n@DG1p#3($Bvkye7WL{)tIvtL_dm@hf+; z(~UQ@yi`5XCleFGDWPj~a(*CxIoEN!M zCFN8q)V)Vjs$!V06a5+M)OKd^h8%W+cnR4Xjia^>@$|N&9v>a=42#(gzMx;)swEJGy^}kba)T81{N)h#R_A`f zRINQi!o}de`!wHh*U_qaLxh!!ENiAL5w<^A?lA z<{p`r?H>zhFg)3a)MKrl-WR) zSmNGgPMhDP(0}gRnalH}h-SgJOR`*27|(yY zR^_{w^8dTt#EaZ-#6EvuOe$ZFriY&WBiWgyGgr&Y_4ZoR@ z5j%A#GY~<1e|#pM&o$}hLmljX`5<=C=`EwGFpCDtFQ*v}muSSNSaQ@ohCFh*$!cXh zVeTIcWV@GIv6>RfRpy`}Em-PC<=iWU-)>qmdvE+<218}o)0-bKFBd5?%b!u9y+H#j z);460T8tMg7fs=5Boo2OhbqEfI811_B(7lR| z?6PAq%t^sLVXt!p@jM$${{;*O{akNmQp;9hbmGJzOwj zml%v6#=t@8&5+_T8ai?w;M$K8?$Vti)EmpOJX6XVWhJw=ZJBiXr$oAVw*$R6ass)& z;RP$mb!O*8oP5+8=5H$?mFj^dB*e2nfO>J0kG$)mXG~w;;kb^Vqh9FNIE0MHsrG7LNqo zLtkYnPP6(0-iqr$^&1aSKjjuaA@$fOuK+J5UZCD_5%loCYQa;ZM61H6ulx+_i5Q}V z2;L5Ot{|K4d09u-Mb^=$alg4@scz2a(R0p1i+msC-4AQ~WA`8CFsRO(GQaHaL4L(~N za+~A~xRFvLIoJ4hIFPJ zyoASOYw`WMo2b;F4xO*h(-ms5^!eZh417KYM;|o8wx?e}y>B{PFkC|$HmRUNg&tZs z^Ki`Jk=#pLdG5&2FTCX433-M?5dPpk=)F;)!6SJz`7u(bv=DlB$_3(eViX}$l;FIR zJh*D9L*4LH2$^36>B2gQ3^PSCYN{loa|B+};B%XgP2g@hX>#>^i7!xgJ2h08P48Vw zL$}Ma_<7z&aN;2`sOx827LH=CWZrLKDVT()ir3 z^v!=KVGH*deAf;^YxXEc=yjQou0Fz(w}NO-=`uR%qYHhXqeu^Hh>5=6dfL`)4KH{1 z(c1P0v}0Z=_+_R_yn{KgV6z?To4ryP6#0V<`L)M^ekJv9c?+nV4%$qXt}6@;f$Pw{uGHxSB+Zj+rd|m9+Urv!sKbfz-G6H9c4Pg_*J9Ex9J~6#fpD!H6D9JU-}-53PUD zi*7me@|tq`;G;a==+#2L6^Efq71;-Wv)J6Wx9rrCY^M439P-RyGS%{4LhtNWf{pTy z@JjtB)n1TC=ia+WuTHqZe$@TSrnj#o17l;T=jAXOdZU3V&gvBGS|}~(kcki~AB|qMa#qK72PH?Qc1%-Xga8dUONsen>V$fYK7vM_VwV$__n5@Rv~ zu|DNJaiN36bl58DA$jK30~V}vNh&+fw~K5|Oe0>&B}8?9Jb795ggAx2C8IvxV& zKr{(d3ndL}PSD7RL8|?c4+};LaNK2ERJrGZq3;7E|Em}X-zkQf^Gngtqzy;E=*Nhp zROaaJd|~CuBkYA&c_c$OmSFBHQX^kZ#yF@`>tT!NXww1qXG0%>;2Cs@(PH>qvk#>H zMM8@4c$zknqx8-i`uCSM{W5nVZH}8mU%Iu!+hejQ4wlA|!;e6LLNH8s+6Kc`o&o09 zU5K9d6x^KOlitTuXnx`Xs{7#|sI8F4l0^nMW-uD7o6cjLLK9|qJE8r#P1urUj|OM@u8xQ<|85w4jl_>GYg>AeYFG;i_*Z zbLStsF-mS{i9&t_afsT;`l{8jF306aQctX)rQsw0%-mw3-%Ne(*atJN`L}@cI3wwm zxtt=|(^aYAgNLx^!COdC{{>a=O+YEl9I6*B2c4g}f`{41I6FSUp2{S&}&Mk#RXUx3=ResI55MXi_-pj0Oh{Pa;c@Ky%b zu6+yrUlg#XLK+Pqb2QNt$jGse0g!Fbp*x*Od zv$})_&OO9?p*^^_z=d|UaO6)x5nFax^4tveVY>4H+-%ar&InmY?raYx2l>e~;Ylw& zls5_&xdC%IZjV|!@wU$+B^meF#@Au& zwPqD|Z=f}s(Yuh;OD7QXN%+4ig5c4(T&th4li3SD-jEeb!r7yCJJ{&vXxKGA0c=JD zLtcv>S$WftiJS34I2hXqYt@cJ*BK6)LNp`gxlE@zyVJhY zne6B%d@^F|YPS4d30ilaM-7t#OtRTc_eM(5VHB>^V=*G%_Zxj=U-UkG+^! z0p}{}Vau!r&^cuY6`cvxjJr#YM}(68x(L$VAZB9^YjU|6irk)k189@@8SHC1Aot8I zSgNmqi~Bl3OYaVhGO0$r>|FeKGYf~cr2wCw0{b$vVFy!>j*ewGr!)!oE({f@{4%4nej9^&KA9941%+DDi}?ZBVpZ_7@hl{B!!;G@M>-wya{Z86t5x)x-=Hr3f4hw z;|VHdP(&g6I^|RpP^Pc}b~?sDTm1pDEmDS9IWer#gYkTmBq^rl(plCarH`5U!+@T) zze>Fuz6hKWLdniq@l-9oR(QVs18GQUr`pQL;3*t{bBY^bb!C7x#Ze7ZJf z)we_O>|oMXy^#nF&(bKp6l`8_0&B-yKq@nqT)wJBV$Fr*-8xnN=MXU~mWm>~=Dg<( zHnm%A%{(b^51NEyTU2o79xV(AG$${*ezQ8xpV_t9t#t5w93}Mvy1?I!yoe5DO>O4V z=bJY$UOCqo`A~BWsn6Jl!O1057VzzRdj-e2{FG8{`Xc&Ai~)#wdjdS=@6 znyUu9KPQK>7m~qo)IHcbF$-qiDu?y5DG+)o5_d1NMcI)HQ9jodKV8X$^*?^o%hMw0 znY8PCMa%#)dg3v%@9HQRY8^>Or5z))B8zeO=7adDXC|tfO`%tY#*$&m&eUdeD?OJM zLPLD^g4{hqucw*P-*eq)>MS*opSh5JwunQ!f^WQDL_42x#*o;*{{{~%n&C=yGpNv0 zHWh&(oarNTNGm6NgmW6HvXg~^t9!&gpPTvGH+@ttKY;U#^W&%8*5^zD&A7>x*Z7mc zAIMG15b}LRB>lBl_y#tvr7No};I?`=><)1w(RE+x1f@!-pC4$`=iFnRH>OoQUdKY5 zI5L<#J@lKZs#ucS=9}Q|W_4_JO(IW~YrwoV09SdYaTTqjcrBkIvdSWYxX(OATeCQ@ zn&b{)1GDJ4#&kaWehaPA*#ReVY;cyoE4R>|@+qEUA!KhhsrMKIOK!Y@q`=i&>T(ZJ z+PNN)VN?tUKb^T+aKms8u9qb97Hd^#;LKKfgbom`%+nD+ zSht%SI}NzEE5`BiSx@OhISbCbezoyCfb~$n^5JT=x;DTV}K| zUqW-^$3oc=S133(Su|*OzKy$yp=ii{X|hn!8RjcVK4r7Tt$xT95sEr|$BBS~YP5|6KLWNSVl zOZ|THH%F9^o$LN_Uw%Fo_ZmE*dYjZ}T-`G6smU@}$7Fy>KPE*Y2XLOG4E|Sdg7~sN z6{-`|R$G&}Hk_mC5xKB^+k41&D}k60XLP!hj3u!Vi#d!Q+& zHNTHjdbAhrNbQADRt?}dQP@vyXorNA@!kT#_ zx8So?m?!@im8JRgEZOOByh=2yIA5GDn!0*g(Bx?;^pUHo<&7{kZmH{7{$9pmIJ9~{-1`#+WtTs~^z*vtHD>{iivgGz zumvIw^B_3W4D~PgwW{e*%?2L~vDk%k#>An&>Lot<*;KB&U60F_ z4x>>s&(nPqtw5o{k8aqRO_xt@q=7+~$+ggIeqQ1_GRi)e`@3O=c>muoobUKlh&FAe z(faRcn7X3PuFiGhTVLbEr~m2*Ug>VS^<^lm?UjeU%^&E^;GmLkQtau73Qjg?)mTVa+xvKoi@dsoLqRWJ{v@y83Mof1npz*sG6-IOtR@C&zny2 zvav_V^)WAb#Uw?3H^`B3f5y{}@-bwI{TRAjHWv(!wZO*ZZy{2ojVY!!c;D6ue};cax#X?0Arm`LFG@3D5QHg0jt|cM&&FI$20i$*|eV7?K#J4Ml_>d=Ve@#nT8FUZOF+k1=@YXnA|6!th^$U zEjqTB8LPLE-H#2Z?1r_p>Rf^3OZ8F7;{T3HrnO9i3EUI<*7+J$6YlF_n}^W`EgyPf z-e8;VHCt>>B&-%MZPJmrd+1B-DuzqeuQY`V*Il8cBZ4Zejz&%IBy>$k!zW>1!79BU zMlNWF=u$m;B>x3X`PWKJ{uZ)*2Md^y-w9^5LU14loW+PQbtqqAW%?zoXMdD|9GzDdgQw_z==_?d?p zHOc~?M~!84QA77QWjh!S~iM+_6p>H6E0J{guTn*sekZhdZ?I<{`9)qnxCEKM} z&AuHz!GgowKqbD1&hyfN5u+nm`lNZRsLF@QT8(7Aii4TkyIRz;{{buMj@A1t$Hi#~0Hd!HL93aOvefirc(>1 zFtsubR(NF@-Re<5@1N))13#*;<#&d#l#e6X_TT9kG&uza@rTh|<1`dX9{_)yXz*Qo zntAHwvFhw_=H{KxdE6bupQs)#_M(YYgRY@*D?RDsEKT})Ul)<>e#1}P{*PBbTu&wy z3-g5D2KXga4uiMn!Ud@|^62npI%c*Ayzb8A^Vgl>?@hNL`pMy_dw4Z=oS2Lc*R7-j z|Fuwu+lt`(cLv)WJdWiw{=ug^Pm>p`cG8k1nbhNpCDUF%h`A)cz+L?l*rU}dEIP6d zPgVplorycx=lmGfv3U=@JI8^Rmr7HUSKmahXKb^%zAxV5_7`0?`-T<^FH&cQYWmb@ z|0*i?$C8SdBuO5QOqGmD%aD|N^@+_2O}XsA9>U)qPR1sra1-PU_)>prd;&bNtW>Y4Vxrnpj-NQ`Q=Oasq<{fodriwGcgTSyS&g`Q5kX!uG6iZGBi2wn6=TA z7or0$h1R`S21s@;_{N6V_AywYgbQXofydX*f$WIu_;^_})^@cD&#ERC{Pi3g^tgyk z7n?#*Qy#UyeuR$tRLGX;m#|Cim8|A$2=}n*8;L766Ycrkf&)i<#?CHv_QG&9Gc+_} zC(TuuibEQCB~B+3e7Etj$9&0i<4#Ug6ega4+3-n585@F?P;Gx0ANads;_%E+yRltjm2GQVarr_FD1TCQ*cwlBHdZc{E z$%_lYTj4NRNH@a6o!3F^J_}n%W@9Clqs^X*w75)(CdMDYZ;#{fSoc00yz2#v9o@u6 z<(*0Ac_ZApRR?;YN;PiykWCtLRPl&`_;6RVxH=<_OE$a(=EF+C@Ny>14h+Sw z_9t-%Ey9ALcKmt$Hridjf@#gYY+kFYo?2Y39p73+NFvoYSb>`$LQdnkJc&dMEt+MR_kiS)CThwrfm zZ*MVE`$7^Ta^`&sVu*f34eikgrDxZuQ=`mydhXA0D(0`#)n8B0 zJu(X1Ym#B?$K^QTx)T~y0w!wMk>9nJWR=Y=a=OqKIVVFz-UM4_O#?lHd$ePF0hQYx z!Um6ABkU#Su-kJJMC(@X<6VnF`3Ik?n6JTU7C5AWWl1%$*EWqT+Ta{}uk?wz?Rv`& zs@;^Y9tw39mJ;Jivu061l zX8(z$qQIpfb{+&q7uu=5#S8o%(ugknQQYkRh-@hjCbJxB$?FU0a4OUZ(v!uo>t8Of z{z;CQ&r>F+s0nYUeS%EgM5wi@7GV4mI5{B+<}9gVHl=0kbi#2a+i(ER7>5CuzXxQ! z$FRbm6WKjy2c}%1h==(&xV_&BiZ*DnH!-8w5-lxec3mP-*ENs~RMU{WUvwD*dx~+t zd=Bc5Er6drLD2q^hoXKWfl0tH3`R;oy4j$ zrI?&ig;&pMNqk4BN`~x~mxOdmAj#Joek^VTZm0|LnxA1##ec9ZqnsCcD$$h&#l&Hu zkeOPNicbp;;$y$jO#I!H#m?qfl8{xMvK=r|5`cyCx3HzZmM~SfB`oZ27VoW*#uuM{ z%IO`;C1-1#c+vZNyyf;=d_&wUa&*Td61Dpi>2%bl7CwxOo|Ohetav!4770DGtg*Fz z8qVIAgnj=KsgsaPNm_D~?34`^W_xR}Fm^IVot=x9p4p;~!&3aRZWDg9*F;HG5_oIQ zp)T^?)L~L3J>ON277>#Lo^~|k2G!wiuXOx9a1W+m3}@#91pd|Pqs-)xB~%@j0w>eY z^uGsc)&s8Gx1QP4CGws(g!PEb**}rM49jSuC)O6wO5vT7@KIk<<2FiiV5pVkeVrQL zHp^f19@dbc0B3%0c`4`cbq}YKmL+;RHr!h3z+$ns=rVKoQqCk}PqPbOchW&;Lh1LM zRV1TD8V?Pvgc*V7Vb0qqh?8>ynJ8@-c;cgtz4~8K@p@x%bl+L_b;w25p;E!3Btf`H zD;>uz%*Pj1@33F%KWvS@jd$XMn2P0m_W7U->n`e}_irlGfBmLlqHD``?48JNYfWNZ zNmt14!o4CVuS(J~`wvE*{EKeOo?)(_hz*>cixbSQ3HCVg%|o2u)q^*>1~Nw>5<*ki9BLeJy^F8H&H zo&2(b9XK|F^`D)^FSrRLd-6OYqnwWxpAX=T&6lyFb^<-WWG4NaJ(h1S_l23hJ7G~o zDohu6YBd`3aPXzAm^eZY9!!k@?TX6~)u#XyGGIomB33TcC#9ZE+_c^f@zX0M^k&;> zdcpq={afTl>F#d&ExVi^d#(yjao(`2)E9ym4ady&Q_v*I5;y7}A=PF|(6nSRG&~;% zx{fMvYOypt*^>ju?$pAut)GCOc@x&i_CWlLe{lVHCJnT*fkBWBvYS@ZO8ZPI`{yw= zIVwSG2NP5^Gr$d+$8fLwTU1q2V9gKm!R+>EG)xad^YaSC+B6ydr*Euc?366y3n5iBh%WE-vDv8WrGl3hn1@{a$sxsb_MxO~|N z685x|)E=xQZ}06FliUpPkuR6H59^NLSbT`ntdy8?!Y5L7SDWTuxk8Hk9O)C83-nH^ zHf)xk2^SM$!R1LU{Fz!R8qqIfqh*yK&iXxsT5SA5RUB?X(5~Zjxw8Rm_cVdO$5p9G zUoG7_ZWs(a8HnTl96?*Z4jgc;pT>Te1&<@$L45iM8a%&+ic7xX&(&FEbhs;dHhv5p zWnW3Jx(7-7Pi3m09!X7n6QSzn z1K3jc2O#PIxE7h?!_&+0rt)UsiX*_CT!Wa19$aJd41dYL5?GN>;mJ)|9JyQ%|9Dr? zVW7qPhQMrS7 zj|=A)d9LEt4r?Uwf5Q2{$$8}UKLz^ywjHQyG{AuqI#|1QFP6q$#F%}3s8`|x=B<`+ z?CC5T+Fpxp4T3-8^)H;-s6g#al&S0fP&z$afszePRDGidNd8P``rK+(Y_gl3GUssR zZe47To{E^E z_pmXvMR2g#dpk(&cfj?flhEtgbiDR+D8{Wj2U`EG!`7l3^q%V;daOwrO8eAMbY&|Z z>Pf+o+m3)<`9Q3cUV~rNTY=or#k1c8-%C&y<@T2gUbAzw1@m!qnkHJVaBM~pUFog5pxfBOnRKvwuoiO;| z0O7vuq`5a2!74aSWa;g|JlunoU6zs=cm3Q7H_}X*w+8D zv9fa#6hG|(pGQ%$U1TV^FVdzfnKkWGNv7Gy7r@t=Q@}hp{9T*`4FZ!tzDXBNtgT7+ zt?%Np{7##aN%zrYO%Tdk9mTUr%D7VJJ)GXDh5v?R!k(jl>9E6j)T?g;Tl{)2`<0)^ zPRx7^=3bs?T2+YURVUdbC}T%LYMJe&Og?1qFcPPEoTM7fL7(bRLT;iCDl$6Z?90QT zXQYi*bxL5qb{?$ti-)1p2GiKymt@9*OC(%s6}s}>uywFDr1bjmGc%P%(HDM-9B%%g zuSZ;>I^$wt@AG!xg5=SAf)9?j9#2b`hEgrpTXc8%WK{b#0^M%E07HAiVh=j85W#^~ zbvpRF5s1Rv$x#w%mrVE)}itc|*Y8`Pv(yW~9cEBnH( zM`=sCkL?jSi%v+|d@-V^nr`t(AtB>jh;8M4I;Tqq6n2jl^4vSXYWF!9`sEs&d6xl{ z)Ds0cU6K}Hi_Omhuw`!)E}wIXO?iEXmEHQvj$JQ;o$DvVu!R+LRG1%+9&?O3b*Nb1?kcr@A!T5*$z!hg>CfOSX=8A^{t$xaGN8blc(@n=!|7tqPJqaEm9bhQ$**23yt}jm z{=8R!#QPEuej12+e;$50z7Rdt?&5e2PyWtEyYW*yVl7PH92d97Wbs4CT@-Jb-eJ`~ zshCz8|D}-w3B3BK#W~pp6D6rlG;D9b_{H@l;wiGi?DVB4eKRzgI7iLp?q0|;{Bvrk zh2+y@o;f_>^A*Z?sSSI{8Kv(+55gGYL+?=ETs>YQspr&x`|!`-WYGazLTF<#(2jZO z9IMsg=Q<|weldaMOR^VTc0?8$LNob4X1QG86gewruUEWRL_6Vi$IvoQH!i1Dg>zYv zFJ8LaSd`Nk%o&>wBdz1Fh}IcKa)C2_$$|~j=}#C123OZX>A+7`$F`)47XC{V%M?h9 zip%Y}>t)`wX{m&(%-u`ES5xSdEa8dVEt_Q{>&3q&w9$^eU#QmnMyinZh|^sjMKbS> zqs529xWAU(z+Nft}mB0)*fW5|LtcT-{Y`ca7Eo3a2z*R zce3wocUjTkXYAs}I@;zkl$zDwC-ZLflk?vUY2oHVnkzK;Js1NxZFaF z$~s>JU#kaN&JDqD8D5yDS41ydTu4`j9VLBAS-Wu}w5 zS=Os(EKl^9O}%A`Gad=%=!xAwZ7*!dm(f6;hs})}geM$D@N(u zgZS2D2Rjq(%g}r+>-0NK-kXglno{xHkHSm%L-1H0QSCqn6(jb&dN6BeYHTvU6B@S# zf!nu3P#IVN6LxNZ*^^R$I~0qXhlFGD+)!L`NZ1`LuOrS^XVJyNH<@;6l90($f}ZpG zIPi2LEGjrgZ#LzSmuBA7eakXhnNp5P?_=@&d@Ef4ZacfPCX)GW-NX8yY=X$x5wMR6 zdG!8J*7{TrmCJtd+REM!Avq!QiOCs2Gi!fFm6wkJgzlDiq zb;xyHMbG^Oa4uvP%)h!t*kNS~*Uf=W#c+5RG8}jF197qvkJ?|du{QNMJ}Irj|F|pI zW?YLNGs`jF@)k?Wzs@{|UT49ptzbc01C20dwAd|*y4GFvXV42c4{{&>!s3S6 z@YHP&JzQSNx&MlxJ02HPwNIsVtom)T^TS)JU^@iwCA{VjW_!~cEmJUWF{6*OZ&1~7 z9*`gWkXZa3PJ_-oq2EnR(47rHn{%V_dvhvoC{t%U!WoW`}G%$S6qnJzHRsdJNe9`Bguf- zwRC9eJ95~lfo$+iA?j`F)I&oT6sod7eeO8%=dV>ZVJZXozb&&sW$`=GJ@hC!{H2_( z>z+>?!b_;BPYJPmlP=aZ&m~oglfeDUWmt910nOctsa#kCJrzBe9#OKyoHjeuSRRFk z$J+6aA1c%P((j3F{~hXn!=698V+#2q8%je888_djl0MmN1CQS=f^S28LD61`im-?1 zo(`bX6&pZm`Z>5deh+|0A-Ac1DnBPcRopo2Ek80poe!>#68;0a5Y#&Z3;+rwFL z8xMS18EEzlMX-BLH$E(;F1I4d<6uX$cDacSUD~Y2Xfw12XuyaJ;at?wOHLih<`mr? zk%@!ul0V~3=)M0O=%@jn{4}F4{37fkfvsccyYvPkr>j7YSZw3h+&e)Zdu@Wm_>J(j zeE^nh^TG?))#&wy+GM~4EgHGD2xO z;CKX+6`yI9PbSP)culPO|0DXA0{_fL4?aQ+eCGav(!3&W(910HYM&adnmeC3Ew2+I6%mpIzvFl7}!B(!WE0z@Crx5o$a;s+1(>#V9Qm$s@9Im zF14gtzntiH<#WVsvL@X3dIwozN6M)d!BHBD;RYQf?!a`q$LSCqk?lZrH!IPM>oNkT z(g;J=iqWxx;;U2lapKf|d>8x>w^pTrb;D$E+ItbSQZGZDy9TCdzXYkKwW9x`<3vB> z^0>MG@*(748Z5L)h7Ly_ZkILEi?R;1rg9_;Sn9y~jCL`-cR`|=yZuEMYS-`v9s9`h zXVql;(il=})+t{0E{>b@Y7n=@DjBZr6LS3@4q#$^Bxx>KPXCmCqLtZ^q;_dMKR)fC z_~(B?bg;oZdfhUdzWn#qdcqQsxT&^?8}T@d>Tif3lTFO%v_dN|cGUs@rN3xpoE>k$ zq)RDflP#6B6AxRMP)Omd(|q3*?=N(+d%OgnW zWKfDcCO$T;fGavL#aFKXL!R4xqcajSVDZ&^)a>yV(B3o#3%zsb*X>?(<3|a-(&7j= zlXB^%A9hsHV=7J;JTY989EM-_pz8zUX~CfjG;3H6t_Z3`(JMLTEU*S!|7?Kci?qTaK?54~jNwzW4)M<7$Ro`t@}XRoz;}20sYeEaH{PK>3t}Mhy*aM%^q|W$ z#?!*&K{RyfVQ^33F}S@B3nJ{PRpmk2?YV$nvY_-}6Of&W=SjoW7qn~32&nmMC-6=$ z(gjN^=)UR>s+Rc%&kjC=4-yhF$x)ZaP5esUSv?{*JjQcIAN08iwYL24l7;m6vq3c8 z?Io!YW&|_WI@7ZklgX(XO&D{cl`b>+LDP2LCXb(9Am_5jP%nLH;;*AZf`-i}dphQU z)p~)AII9%C`f0-Tk{arHX$)k0N71F@)u>VFU-Hz%OnmiLA{V@)k#m~9n%y#)&bnHM zGhHuTrf$}WBf^U?QZx=7cPnATRB5c)Na=`VB{D9;kmiqcB2LFI@Uj^q-nF|BNA_mo zy69M(_`?iuPt-+k!JnGsry==o{C9Tsw%~~mSEE<_a){rnIFc7Mg&$cpiJSQN5$CV; zh%r^jp;;kxhE@hWGD<>! z^qAAnLn_F=Wzodu)eHVWtucEQt;JMCvMfURCp`J{7eZcXq1Mo|u(abgY-^B4u#~~u zG6wi&kO<3K5~$vvP&%`!iukS3N0SpDVN97W>Ln+WrGc;c6{iz;m!>c%)z|UVw?TFKKmp$Y)PCrUy<|(9h$;>8|XJbhVrz7FoOkX^UTQ{6#i70nw!Y z^d!>js|t7RZ_t&CM#6e^DH<}UoP?LP)2du068&l)zXnXmqSCgjgW-h>{bVK>xhp`JZGEG58= z_Fq7gQ|oa{@I)NGxR$DPw$faMCMsz^2e$$eq0FHeq+)(?Gd?zn=l)F>y&b54;+zUN z${m2LHBE4|uo<@9X@txf0bsH_7QEDxp=9<{Y*f*}Pe1!1c>iW9$$3GSS&Kk8mc!?) zI%?JtN-t+Ep*xGLXm{iYy4KYLhu_h|?P@Pzb=!JYpRk41tIuKm^Zd|A#tU}_%))&* zls&tu%jU`qW6hXN`E7cXAKXb^w!Nk8Cmg6jQ84|O*-kahlWE9+DKw_+G%?gSrP~S> zz%~0ISiL+2A=6HRgF+fsTujBS>l5(n-ytA5?oM|8RVBBC9kJWeQ0Tw444$|f;)9pb z074$5VXGUSpL`#}{~myVN)f#^xK`}dX3V)bsI%|)U*hX+(rmla22B54h?~Q|p``3A z`|##C8(*Bu0-qGoY%zZ;#fuQmqNcix3&)!lHUQ57;S<}$c&6GA&5|RuurSLR{hhXA zK>RhvkI7?2dqS9Z{}l0(-QK*F(i*Y%qfu;?ka@T>@Cm+HI}nd%orJwRo#0%{JB zsnGUu2C?6xN3)g65YH?&Xa))xO(hTPg&;5Tz0HJW>cehT&$zK3?>g6RwwtJ~qZ z{5-rp@JiKW4A`y6N>(@FuP4d4c8M;roHm?n z-&n(qFEGMQkBzYYe+srk^ zd;YE!YqYHp_y4(Uuu!eh~;54e{}&>)({j3PEKU+=@v5g zxtN^0Gn%-qc}G-j8_5c}Eg~oy&W$cS!z~fM_Z*@P=uxjO(zL9a=UnAURnS%9TG-0j zr|XcSv##WxUncM3GnbdTaf%x#WOHI(iovGU9n2GXw)8NmX58>tiE$3GlCDVqxH>lCRA9Pk@ z68AMknH(ti!#Uoq7AwwMD>f+~%a1D4r~RtUB%IfzhA#7MmeyUdbn1O7`Z-aFY`ay$ z`&^MDrCFn)AbFj@4=8}3A4$B!LMMLLh%5ZbA_d;K_&oQsDUy#rsmbdd)*$)Z0n+#M zA)j)>i+6pqncN}2`9V9DlG!%Jqc?7bQ&EINSX_Ij5{2_GsXHDuHXiSZ!W^q3sZsZP3XyM$P2SCZrL@3xLjjOh% zaVobj@m}}(xK6K8ynFpzGC3=h`sMoZJ<2Y8+|h3S^Uc-7+hhtEnG`@?O+Uyj6!?Lc zr##^_es33rnaNncQmnVRc2Gy$u_TU@4AST7vg^scl0)S8>H+k5&mag))`0Q8Lm+dJ z0XHf^n4PU#O(vDabMjaExui4G_{(p#n4z&FtDfb@b_k2U*ysV+bwC@h&wGbs7JIPO zrs+)TsvYrva+XNvPb5$1K=|*61Xim~gbjHPaATD-v_GE=+efHk*>fda)TN3WDzi~( zWD)jcU&SlK-Rb6oP4wVD4TuQ&4$hm(Kt>!3C$ApIXPQN*(r^~zjvvI4R}Nst!7LoQ z(hEG*1;5XZ0+hZy6I6am!RWc;!9?E%i~g=a@uoupUtkUEiq2yjo$8oDh%hIzI)=Y@ z%Cn0*vd|?Z4)3}PO!cEH#7*scL>jNpa;?#O=&9_(fJ3}c@+uY2uiS~DsUxs%psvLD z?*xfb)>6r@8=l;|YmWQ}oiWtxwlaF9@F$CD^=F$r3q<#_$?zB;?w# zFsNBtjx4JMj_)3g-yTlEm$8*3DJh&Ln9hT?t+lvo_i(2CEP#34o`#P<)Z(Y4PMjC8 zfKBiSWA~*5##jF_^3Y;9Jv5UeFFhN`Eg9(sG)`jt&t>RQ|;*dbhevG6GS{_m38E@$^|6cew z)fOMR1N%5Pg?$`R$s%Ly!0GX5ujdU+q)HCEO|x8v?kD7ZCj{{iZ!lluYLf;LI~<5>rxilUzK--!AJJ^;JLe-Y?p{Rt z-<0uNZ~x^(V=Ymudl)vQ+hBOH4&0g&0-n-wFs5)e^KKf)wt5S0?}Q4nQMhl<`qk6B z!-IKogsOPj6D5A5i3;0LT85==AvmRN6m&nB0HHH@a8q9_v8?0;W~930$&Nn$X;2E$ zTS{roUOyX$gE7`OtwPDhCni<~b26-U{`_ilTmvyZU6_F!Qi2&d!-?L4LaxuWk#E~B z3uis2QGb?0-d!&zdsQ-t;}k+*`AuGJ&3Cbez}4Us8H+nxi#zW7qpHeO+%rcXEk7t> zso>OnvpNH$&V|A>DJ8TXyBbn<@emy02Xn6thuM3dQteKmCne+t*ryM8q_R}l(H-J@ zJ3on?g^so1)=Ahkb~uhWbON#z!hoGIfMdhPh*N(a6os5<=IlQv^Zu8P^4``wRTx=^ z`igsS!iq_F@?a}p^=&3!%8v10=G#E>HAAZ8F@!u*FQ!-bf8&>rS0ZaZ)^VcOntaei zBjWf|gsY<%n%%L)T`R}3JJ(fNphTL9H!#vs;>j0=e&lPSUZKmACs=aq3H})&LsJ&? zkZgegr7%bXH$@b|=28>z2y?{J?S|O-c?32lmtbS{I(%Snj~6c~fK!Q(`8XmC^8HS{ z>4!4D=XDKVB{hVUg=-K!D+#%LV={_DAHd4Oi?DWuDoN}xCAAgtym7}X72|T-ic#`Z;Ph0*t()3jsZ7BlA! z_V)^XDuKz9GBH!|9)w}}gs03!zJskh{DOUXmu{0krcX5cMiVc+EFRW{EQBzh1W@;Q zLAOX*f^B#@DDQ8EfX+)WBK<9d&AyIlFUl|{{yuiuj|2JS*Zd*FXz|}6;qWj-34WOB z07Nn67gVV?M)_8gttw4Kea*w1P<9AvicA(XEs^o&_4mA5#`rbS<3 z>B2Qxoimr{{dAVx3UZO8Pg%)o1}|U}6BaW&$6a`G^e$Yxek*pk8(2TlthLguD&lsB zrQq(HLe6&SQ`|ghDBSEAO(*y7BY_7hCG)IKN#6cFBROk4&*pq~7B4q+I*pZ_CbO>% zrh8<{>98>~#g0uzHk?AX%|aPhNznjK@@tueGAPVCs<=SlGM> ze$DNJ7Xbq#U&aoUBns@+mS%l+d!-gT|M8%}hAv=hR-I&=e&m-4gYw73VYeWDbB zrQu|z122@VA*j3-?R-DtxQ)ilZbdW1I$ea3y@Er)E`bd9cOtEK&r+S{`%ql*5QbdU zz`UvqvAl;q?=o*UIm)v5a^r4Z?phtax9%zU_#B7Aq%JtDpib+*r_i#o<@8rn1RVTV zLXX&#QNGTB1?FvFV+uoA*iv=;6m~~&0ACjPTifwS@O#*xITFS8Pw}pNC?5Jg6^)1u zyZ4_pQ}WYh+SVG9nR|r{;dyh(rk`J+d7M7N&T@dJSG-zIG{44r84Veh1$DXp;A^u_ z;BO(C>MX{DoD3Xhc9d+)yF+V3V?m))lSQi;v!Fs`2{#3YXoICB@RzA%e5V17^h=?O zpDm+TH(9X}-5>Eqcq{g&3G+F(A~L1Fn=JV}g-J5!u*8QonsEWR*Q%2iz9NbEo6@CLb$K67Uajzf}SLvN(Oe&tY|Ix zZ^B>LCphjVE(n9vx`XJqEDQOv188{sGkCe3hceF?DEAM7L4!hJXhkYK725dnYim&X zY!$k93!UFnns{rq2f&j}Y)i*BHbmgh<;qu6x4LdRhW$sS53FTN%(k<>r>^Yi&Sy-1 z*L@~u-o~bKmj#dSL&oMmU`^8VKxxMaNGn)JUBm@e@3yoMo3@9fe|b1w9C!*VV$1OI za49(X-$fe#DuZ4#tCn~_IxiVuQX`pOy^CId_L9ErBJd?|Fmcz3AV+1ElLv!5tQ}r$ z6ct`>2A1HTo3l}0+W~idpN5j% z=WNEbDNwpdLe)OMXI_5Fl0UojCH|U1*6~dytawljk*mVdX_w%*+j1UThGWF-H|OH0fJ4wj?iU2bo5nP}OpJ0$8+IL5Z8;l7<0ap)vV zw)}uSJJ(~wF04og$%O={+Pfbz8;_Cs?So<7OM7t3pM>5q5jZ47@P1k=bMoh9$T!o2 zr1yLV1mCxSTw!l?pvVee84knu4y_P%+J;zPvL*_)p@iEY!XqEdaa*Gn?p>0{25(Gb z$LAbiGXk}7RD_U8S?-9TZ^TS~`7ma(RE@X2esj+`M%U#{V|=dP)|3EtD_C4&eG+Lg*5Xr4?(KmXz;@* z`s$}4{q^QP;ln%#+m%iY-dH>oU3?GU^9aIf%Qsz!;;BW(58 zV|^`7EG#>c+0NIbp=pFfnI>}MmROK`#nCiUa~615jfBU)q^QNgjYP_92B=RL3D5Cr z5dSmAvgYl0D6s%5^eiwsVYY04FW^& z-40x?V+V`gtb%dVrlb6?;V>)WBpq?KktVNE!gF2o;Qq!Q==iA5*39>1tRrH}|outxK?fkv(p(o{Zmh3OVuaP8**?7ey^Ihhh3g1*|HGM6KphSXdJSZu5TA zBVAUKuglydgK)JZrthx!;cy-9;iC_9*&RLM%qGvPXeNrwZT4_oE7Uoe+CcHOl3K<+ z{=!0LXi5J4REMEI}moqSIGYU54u zbKj}RKJaK;d=o#+T7>c^HsOk#1Ppm{3eflr=$%wWIi)ddeTBf|J{Hfi92#I(@)^7s z_5<~bx3L@F1vhEnKKA|LP+Id_osK_V!CmbfDmXqTq0Oo#D1EU3(q+!T8mR*CkbVFw zTE9Zq@slt}c&8c~7|^peqv*uFqcQz&6nr|m92xZl$rif696$-N(x2kjDjP&ru- z+8>VNCX$Rb34&X;se|UV-X}vRj-a*%(=fSy91dt6j1%6jMSJZ{m>3g{p8AS-qt*(^ zw7HmLyqDD-UCPLI2e$NuIlH-+XU>b{SXPM;ofh{0uE9TWW%f*VZ=OGUv2ibZaoQH; z+{R%1)&CZyeW@N^ zUmuNSM2rfxX}C^s#>Pk;U|-vBun+3@shQCLdUjhCFPjv{v}YV+z0WgQ(D8+|H+2&Y z(zBr#-CogIieI4J_5emFJ!1Tg+f01(6uWi(9xbV#NW0(7qKdK4m{x?eWJ|A#B-=Na zE+4mn`s$g}gmMMgpYBA%%Y|s z8A*E^w93U_ruk{QWo{NAxE#>G82_ zNlF6C$zBR7ai{2$!Yj0M-V-MKt%dz0)_ zm2Ay@oS;^$3qJ?s!m45Bq;Oy%xp%pfPjBoMAMSPMN6lHv-}jv@iCy6;IdH*QV$=@c zZXH3tyc`SvN!y`JY&7<4%EdbK!8j~+6zcR3#E&jt=oOkr@3|LI)lZ`&&+U{Y*-pw5 znSm)FTe~0L86Sc(dwNlN^;PT|m5r0z#z^K(9WLqc86-Is-z6}u4A^%&VPkuAVlBerCxrsX>SlGmE%*cHid&%!+OZSE`d#4yCsoxB1CohE;I}bu< zYCSyu4N$)>8AS8j@bs;_7&E#Re?I%ddkTKMCC!`2%*himxOxEAZq>(ozufRovpckz zbi)N9-{^nm8#V9Cqu1T<+9Nj^rb zXZt0{dUGDaw+LLw&E4>5q`;smI{}M>Z==k#N2q(+V zD1C%CY-US(8fQr2gWV*jic6qYqY&u+3=ls)#I!BbSjmV;HX>V(>Yi<&<~M8US`9l^ zX)ul5sGP{&4!ugk-!}20(~NlSpF)EIkCAg@49WGzE;zb#C0u(|PIZ+3k-+E_GC5VA z?$Qb8cDNi8*C`~6pKJ7!gQ7?>DPSqx^YFPaC)h&I$vKeyF1h5hIEKF;^qj0~IzbN( z3#L&%v1H@tHa@4gf>(P`!9Q*)q5AUzh}x@tQ(73|Zp zgoEM^`0J=5w5ZjD!?Fr!lJqsW>{KlDbxDOT{^hVx&;;)$MG75HD+qt))(ZZpMl!E= ztYgyh4~c?~e}O#`POOHd158fMhpqbd(0qv!@MK1y!uO{kKWi;qk&pnKq}#xim;GS+ zrDW)s5)NMEBm!>Id7!n_2fQzF0^aA{fP-HI(2!mVg7+>FoT^F^{Ad^!Y}yiS-;{7l z@UuN#I5(`)-Z9@^sF=1?Bz6=s3innBmyg^pO!vNKADm(-d_FQ-DE(Z*N!Sp__}pZ{ zHotn2@T{!F(R2W8nl2XBU3X=oblRDNVmZOMck={4%F3p$93yQf)D}-&kg6_f`O<0s zXQ)#+H9;g2i={=cKg_a^N09=p&7Q)ouMdh;Z`>3vH@zVE(vTw>^w1Mo+ddWgUFs90 zzYZ4Wq)RiYt%hJx+b2LJeR8`16X6JrO2L3$Lf_cRJ;1QJ1}*F4dYFI$9159o`WU#JQS zT2~0+VkM!+w^pkMv{P#kl6)s9St;7sGC>N{>k`lSt5>b?Tn2_xc5ozx_C|crREV``T zD%n#!$=C}m1T8V+g}Db`+pNDFH)TWr*C~;qR|MUC6{3%_ZK5>)L(Jg=Z6be%BIfLo zT;_vQf+%P8ZIQXj9-*JRgWaX7YP(eSrM<$Sr(HzoBKvc1KG_uNCfExnEQqKaq*uo+Z3kf`!K)%QJQwh9KumA1K)5!qgV_3Qtd06eSH@ z5V@I+6uo$USd{pu$*y>{gI%-Sexb2XmTk)VIy<**n+0*uO~h833KPGs5w=)0+ON}& zxBYFCWcOr}uJEtsW4pN3Cc-WLa#KI=DY5l)jJDmQf<^91CxtT$q`|ifQbLod;etnl z%1p^PTZU-1*$FkC3Wh&#Wxf}BGBb*|G6R*zz#*FukQlNFSYJS}WLygP$N2*b!A;N; zvIP`1o@H94FAG<=tQGW4uM&I<^Mvy=`@vM}Q-a-P>G1M-FWC8^8JN5c0>&%=(E(P> zQngwp%CtZfv_DW3!_8#gp7IecYTPV*z3v5b_1Yh1wT?X4{Z$^GecJ`HXIjA2C!=7J zzcNtEh-QG5jVQ6JNDv#SWGl*l0yn)$f$Pr~z?X*lC}`gY*q!$uw3}oP&s)z0FDeg< zzQ*i_Z63uS^sost^05nO(&=He=Iyf|)$9RJ@34iVXL!KhI`Jseh0CJ0vl8xM^KX2X?H z7sl+#6$lrHqEJ15bfI%IdjG}?_^q#J3UrH@^@)%SY;{D(^0~Fi_`%>RuLfZQb95j7G!tnTsY0i z6U^Qo0?>PNu%KuOSUa|vd3K^(G-3H%9BRH84_hw7PryACR#S@9;)sOX8G}zg568E& zlChpl1+&iG09byV#r&C_2G6Y83%5Lwyw6kaM^K(`1&Rb@O(hRh`l3>@{6% zm8pUUC7d&R!6)!|D+R#@y+E8H$!PdJgo7n-;OF2EC~EajwB-I<S<=`lkH$2D48);Z334ve3=D#o=~HD8JM|rIoQAB2+$P_ zup`^vvw^2Sv99@AaQkj07_9Xi%)F})wEB|4V!0UL!PZS>Z<^S7DJM?NHFaR=HaGVD zupiry8vtrERA7o+I7qqm01fVw!7tBi;k{19lHb!Ygbgy$gsnS7iFYJD3`t(g#ZZP- z-loK+?bc%VS&Cpqi!)fER1Cak6`>5{Z|K5rZCuc>o(=x6pY7zMS$U=iEi!nDa8E0u z?~K^l^_Facz>YPtb`{0$NM{D0Bns7(uA@S?zY=ez9Clu$k0&lN$20bt;(yQ{HRU^^ zg==P^+JhxvZ$}giUZ4kWyQiR^(Rs*nV=6j7d9z5Q`$bTnR4K5f)8X26M)1RE9eBa; z5Q>ZqMrJQZqDKcWFf+$AF^?+JK-TutP_f+ z#$P4JjPVa=jO|SSptw#ueBizgUMVAor?ZRjzYWWA|MRVQ#Xe2;`)W(pX#r**=xs++ zpSU6|K_b%nc@6|dWr9X}4w%3^_(oSC$w(zY;pSNA<-82OP*z58o+he%xCTx7kc?u( z6Yxe!=T@gzj4zLNV-M%AX1BZ#Vf{OD;ibEy;ENn52rQ1{tyVesl5Yy`SM>mL_Z0xE zr2*#lntwb4=>NdU>FLU9+v^#fseu$ z;$jKY>KF4Nv_*K}p(@jJE1c1D2mp+`5lEVLo5|bZ0?i&vcp%&ML42|ZY}CmBCtpp3 zU)Ed0_>*!l^za!luO&+K@>H5ocI8pwW0c0M_VWSh`S~zyuA<}`_brGq9g7y6FaXcw zb}@_G<3wTZ$3&a|(*t2&bD0OO2T{WHROC3c1653ZhQ`AN^t~q&?fd48c1Ktsuj7-E z*SRGycgPNIzwi?7&^#y{S1Bv-k=ZDi`AG|HIH>`$&*I#TSY%Oe%v6Ic@@e5Ga zjBF^By#r=NjesiQN8zW#FJW-bB=F*JAh65O0>8mBRG7I4fz&{x*0~Ls{TXJQg0h)V z#V0Vy_zBD#JqJy7zQ|lTy^axY`^t>_>I^l2DfABz1FM{5CjAm&{O%}%cny{@t$50K zX?|d)nT`j(6T_IP`bc!o!xC*v=HR2=Hh99LaJp{+@OrliXyaG#jM-?E+p-BQl<>MA=oiCu$?nkpZHA1CqdFMj zxe)AAtwBWtuaF{|iSKj;gI#TJfMVh<7%({&I;?sFjvvbh@-luBFZEscNAEu5Gi>47 zwln~bN`W)eO~EvUQPAf|4BV2xo_Wx&2+E#H+-(*~ptW7n3r(s3)&;6?(WPo8^zK9E z(XI8Ma)L3C5xfE3Jz+3MCIMdlmIs$^Pl0T|HMH>*LGS8mD7DiCy-PiX1P`5t(!{=jmgiXZZ^7QNrA>(2u~GuE|(V_ZTKQ<|s8lfR2~Dpw`z3c-5IieEMHJ-W3%)i77elSh?y%go3GKRn09gz!MZ0fZMx7?@s4h+ycXyZKtrGTQ^TS@8 zYw{Z|Dz-)6e(yn-U6Ek^nPI^*<8*=ZtQFWsv;=p_t;AL3D(Ie{C;B>^fbzc|VmAh5 zv1)pi>?>_vxVU+ZNTz%$T-EqUw0mzj1M3b7*ZiAoziYpgP(~uEeMnzFG9bvR!^2fJF)I*U}~b1e*w+q)Z^>|KZpJptb8b`oSN0627OG@5gz z8s|N^iCwI3V{2xEs5|kp$l>#MNnK=(b^Ud*TFfv?dyt7z^|m7EPfJjWb~0nMT#5N| zQ<+h!-oo69HeybkY61NxXQR80lh9(6jSdxm7TFu@X1wp+W{%k{L}~S-(cfJYP>Q=7 zTKaS@8VZR+Gd*_!H7*`#Iz@q5fA^rSm!&Ae!ZjHc&v@#DRVkcsvVRQ^H2cNrRiVod^3<oSvdj?w9BK;DKWTd<{Dhw5r|z%FdQmVgTmTg5>e+L-v{Dolni#-@`>Wzgl?)Y2aY#b1(2UlOa0=?gn=Y{a5!R*+rB zK33QZiDY*zywnh0`61!hxsL(@qwh@Q{t?2RXX^ykbFgS`fD>D1I)~L&a$@=D3D82~ zlirs#3`Pe2K_>3MQHjYHROht_y(qUwQw@Akv-5Y+MCUFhI@}c~9MWZ>v^uM_R)Gx> z{x2uQ9Pc!kiubKCz|ylc@F{Bzyrl099#FoILsCCr`?Xh1GT*{xGWwz znLmRDkvyEzCyjq}Na3m%ZHVSe=B#?|pmmoq=yR8_8}i>kKPfTPof3vN*Jhx2RWGDn z{ZkYobBU>I{mgvVR7AeUDriQSD(bv%hUNc^#ql?FaY7A7zZ%ui?74yHMdkv0WQi|c zxOp>vU~~!Ddo-e*$+bvhvnSrTdoI4+gYmL|hfwUXC8*KM7yWyB9sM(HLc@Fss=sFn zlQsR|ohy5wjq*{j)%Frt!L@?ErfAVThe@L87jH1n{;NT=%PY|U%0lCIJLARHPS|LI z0N=j%5T~AeiS_hf;Ncs^Q1xChZ0H(-#R|SS!!#VrOxTSF+^s;M3}ue)9wW(Rd_kiX zKO@Jrf03!PA~W*|7S!$*Ft5+2iNqsEif$L=34cB+1WO+YL}QSb#&i<8DOqAQCS&53D@ zqQsdUx=v4|C+%uqq>w1sdO1ZD-F}2A4{2ro_(#EMvccfF#Lsi@&mEZZF2CkSOU&!fHLPogQEJ5iIQ1FkSB9XR|qS>l3;2QS~$GJj4DiE=73 zf%N`7@M6JUsFh-EH}S*~(Q8WyKWI}YxT{wH0-}bP3gwGvw?{LQrk9Y-kQoxc&4H$~ z6j79l8IxLXBO0~!tkBAAyy)YRm%`BN!}eY~n}z1Tt_$ZotP#biMnl8%Zg5}VTe$vm z43pk4z*ygss5$>Vmv~Yy2(x}~5In9Ffe!IArm$EAJSkjc8%w(d=8pG8h9j@Qtx=Y+ zE$%X7)ffZ+9aMzxk7>akISqJVK>)C7mSSdYJ_D01JYcsE3s28E%dl3KqPc;Wg$nVq zP|0ixT$6DI(mpFuO6&n)e5azwIP(j5t zdPOJa%VM+7T3ALf2AkiQD4^>WFtf`&;O%N{a7)s+vOxFkLv>$*;RS=>X7Ug4VC`RK zth6s2Y-rmOwE{x)EA-!mD;em$P%!_ zM?^8VWWkFA#f-UK5Yzawgqdk|U*!AqCv$9U6MQAf^7;EPP;KxXNUpsODjjyhC5t7m zl^O%3^qy&(P|VCZ>Vh0kw{?!ingqVNxouu@7`;s@vZ}C(Z9u%jh)Q? zxFKY%&rM}B6cSG%$m6xGPggWD5Vc*Va{7rBFPun&{s)t)v=?@t+6VVvNrhL! zCyBDJzF=gWE-=Gu7K!edeiyy(Qev_$x&!SWJwWJv8&ni$;h}fOaA$cER(hh2zSqm6 z3`Pn4ckUQ?x?1AVmdr?2+DwJAe&;~h8%4NknX|+*Iuq-CGr~qs%~|i1QS3~04fb+u zEXwkVLTAEvq6eo&z&Yd=C`g_F-+5Xy6_YvyJLemSIzAjl89hhQzt(+7ab*{}R@#dW zti6H48l5oSG6&Z?O~40t&SxWn?b(w_M(o;{0pY$OJH~EW0mx1(kbD;MNHIhP#dQ@j zWvY{azrrQPbbk<5f8dK(Rk~sKJs08Am`FHw*$FuF+6G`NYX`2DxPrXCxybGKYiK>T z9olUVW+ahr#<_S7qam4vzCL$X^y|HkDB$U7=5Un($dZ4|q-a?H{iIl+tbGxD%Q6JX zT7jUzTLIFSZ-A|OC#c3>05fNd$o;*INYvuNn3J_&O|Cka@FpI9Ut0vKo6CWtb|Ktw zd?JipEZM1KwnKXF`(n{x%<|*eMtz)zqlC1 zUot?=5fJWAS_&(168y6|)&6-7uwPPoMewp8BM~zfQPVMKQT742&B6s%6otbfc@@_G z+cWHYzZ;)X{4P)$QW7q|c}nzgwG*?s`xxW7SC26rp)9!i%UN*w#X`X=O&Pp0x*v5q z{gAk;eHd9uZ*u2w8#C+T7mzzcAD9YlnJcLy;Pp61km8jIlA3gcYhQ%fk2~vQ?^0TW zZk~2Q9fwcD4;A;|MbFzXval9@{1F45t6u=vClmO*S&za`3efaq8`N^h12T~Y&~0S~ z&^$W{b+3o0$a@v4e5%N#jdEgoObVDQ<90xQZFji3a3oyqv=aOce8kut^#JRAm6<3% zOXk9`Dy#|}7U>FiOKPx8=2re%1Ouz#n@uMn7`;ezqc%>6*mJ_220ul=cg|qWHQod@ z%darYUSV+Ey%>7l=a{%b0T>7^V%{Crff>8>;M-SQ;Udj46e8S=&RUow!KZ%a{H#ep z^Ux+RXr6=iJ->oV4OH>MF9qOXe;|laSq9q`8sLm&>hNfY5uoP{@Fgb${4&@I|LYxr z9!%N}73T?{+^9ORS;8$lm#Yso+;y1}jcLr&Etf<`O!L7bv$tS$mJJN=GloY_8GtT< z0pPUXgS8)RK$(yS9mekHs*E<8#922zaF>h1pP zH-rV+1gze%+6D^j?L{IxJ3Ao3lMCj|vX}Du|KA3>wb{lG-Kb?aop%4{$hT4@engQ2 zpR73%$CWH6K7J0w@kAoEt-4A_8viDRotpfIW4ioqp(_7GV;U39dG^WX};^R_WRuVFRqFVbV{sJTU(9o%rxO3b&QrPMgpLyV{XUwXm#~_@a&-uEe zal#bk+>E6CXS~VDr6wp~N&-;v)}^w#_4KU44EnAj2Yw$Alc7pk5_xAnHJIH%Y|m+O zp2;tGnf!MCS!*hPW-fyXBaai=gG8qFC(-juCl1d4iM}>XV*Mus;NkV&4_HoJa^G; zp-+MBST|ZDyFR6awzc*F!bD!4mg0f}Q zJ5gJ_Zq#Cwy9tRGM|r{3#u1dRQGdUE?agYP*2_s+PdZM;+kDMI`V#eTwMV z%mwJ8Up!heE|RKUVEI1>d312xP3)Vvltkwn)0!)~be_x+6bW>&$z^F;CCLlPzPd|{ z2FFvW%06!Mgb~=|a5qZT)t2Ngd3Z-jgsjpF=~|D~yk>MBb$$Pj3fHOeuC^z+2=^)U z@zqu2#2gQ@S0f7*Wd7vz4~uKFH(S>Foc~cPJ=ue^`|oV+ul;}EHUm>~ZbJ^9a?Ba? z*M`K$&i-Q}zjcFw5J``ysTa3D8H*#$?h?0(BP4x}7CmTOP1TN%r{*-3_Un&Cc?%V= zymcX3KPnM1+aS>k9fj{$F;H&3EPT;Y%@uw~<(eyUp|cZ7^Ql6vLhwZ zw&8|xSToL^8rM0Yf3L#OXES#ck8{zCdA?}%rM2Y5RCVZcFPEEkU6aDUyT#L*r*V6= zjrco{&G}nn75LcoQK)NS%Ek8W$?&dG8rZb=cCG%U@7zApOfO$`;g@7B;J3M}v-6fv z9CcO=r^U_XlsBe=38M~))s3?-UAG_0OgjwfmWLxf4Go&mu1xjqRPo4%Gx3Gj#i*#~ z4>u%~CY1w6xdY@RZtu2)mOA%P$>S$v#)EjMCS0J(!h+ZRmj2ZTxVciBr(}Dt`6s zsrdK~BW&vBz-lln*xceVob$F*Fg`8`H*M-CqGgBa8|^6gd2=;gBRMbl0!w~bnm%iH zBML86I*K2ejzdO0Q`m{)V%fPy(qvvxEuC*^21U&eNh)~_>vCp8fyHw&)}{}2Wy~Ql z;4UsM@j?}sMvIkJ$YLp5Iab#y26Htjq&dnE+Ng}Ch9fSKoMTA5yZAjCGuD`_+0LNK zr6;MDLIIYar^d)^UCmRgf4E{$z}n86&RWI)txeU6V4m#V&VA^-$hq4~a=0sZ<4DId zWa=e5a_MRwetxSESFd_U21i@tYe9ze&wgX>dXR~D+4lzAU66xse;SSF{5B#>zjSbi zma4#|g=*BW#Tf6MC4-j++{9CB>frb3+35MmKpYePjiW1L=ruI~E}H#-e$oMagd}&Z z(I8-Ntf|3{W6N-gxqy{c_GQP6QpH8T6}i8Y4x=fmbHwc{YQ=tYbH!y>81AyftDxzd z56)iABBi(WNpx;1m8#u{vdIu!B=PLwllSoQJTo@?*CG7tL^@dxchO;w0J>-CX;5jf zm$S0H%0+yVrsMK&F`EAMWW~fk)bO?j-4K;WE>*RnJ?lrYx?A$`X8kPsCwU=UwY34Y z|B++K)iC_2z=edE1!7;-5xC-@HU1*NXtt*p7%5nQzkD5!HjNU{32yu7w{7Y8V%|%5 z!0;!uFtVU0>nGtbaXgnWU?@I*sH3*u(p_Bfst%{+_~Q!;Dy5fOaJkP9p}Db=hSlS zcV`h^5MF|nrHsSvYyrpVAyl?L^A5v@BP;NuQU7!f`i zIhO7uu8zLd`nmUk@#R@K-AILXo|=an{%z%6YQ2QlJ@1kAo)KuR#(4Z_c`16=ql9!{ zW|QcT1ijJf!!1e=@$q+KaD~o3G2D294aBe(mJQGN6AN)Ydeq|7SBMsaUtbz@NF}O`uGHBk~jQcGm z|Dx~3=v(<3N_M)Vri*n{_WU|(?-oY|SEOL=C>^rjrU9m_$&m$lP2<|4TnhXZ&^`se{rqa*s4+dMwa{aZ5>#d;-`e(`^%Ac|!w>opk+p%$Q zeE(to_tK;M%Y-9*i2~nHC!(6x=gFI)N#tYA7p^%YhmY|-$1A?A<|oDK^LYU+{N95v`Nqn6PPZ=) z$q*SF|LiuJapo_V{Hcpf7&Da~7$o?V1Em!QRXNu?6Jaku0t{x0$z;WPSORsCrte1n z@$_PTT+L}dyW6UE(Zd;HO;rj zfg7%v%6F&8^U@`5{K;xN)Npt-?UnN9%+@{Q)~N`ge&I$~87G7C?qIaAU=Q*8PjcSs zeueBp)v#C48cx;gI((xt2Dq;2pgT+LDfJGdy3G=9enbd4ldez41nIE5j*Y|a3F+AC z*cQ&W-Qj-Ux~!rK30qH9EZ!~UK$>&%HZ&D|jaa6a};>n*vam~jGTt$~5 zw9idv*39zcU*sEzKc(KmJxnb(N&KF>tU1Ujd)%NWZqJgCNYSY-#Bc4 zGJ~7n=15$e{(*;LeRMisM&gjM;6y;5(0Py;78P-~C#U0?K9_M$gg@4ix`^e*lwq02 zCD`Jg46D`Vf%_v%@I1d%@>{ip^hE9>@^BuePVq!~pAnUxY=bt37~>nqYKU#vc}`Il zlAHfYX1o&Z@?ZFF5$#I7r}Ojd)|x`vany^E>D4M`kxzXEQu3N*jsPZvf*q<7n$%6_C&B0J+; zOs3HJ_fC+RM$drS#s}oh@!OIe${e)FwE~?jd5GpGzYvdjbeoIJ7|AL3*W*VgKB4)Y zyYPp-lhGy_4-}VgjMtW^!QQGmNq+1Mn%_2s|LhUZKhm$J-?mP}^RHV_sfcaVw9$l4 zd*y`9(Qmdfs%Ky_uJ zXs*%-{?C98fB)u3RP*&X*>pjM?$h(2C#QU+@nMnl(C6=5l-VdaPGtkBdHjyVsYzC$ zWFBGVi3WUORg$3_FR5`xQA3G4Hz{be#2KSb z)2H2|-EQ0YNtKtN!9*AG=w=R#^fd`kl`UIFbZ@+kDnIoq%)f(!r)Fq!KjJAhP zVRX~!U5xqmBGA9L2$i)*)237-8f&dgxm}1n(w4#x-MaATYF9BUQvz3~{1w-4SOaqd zx@o4_O!6w_CHYtSnY8ATHsFRG%X`~>bJ=E`~{L7^fVN=Z3-&i zy92#=8;JL-Ib;8g*|=UBAY}=ANYk+dI~@y>)JmyzsFcSJ+kfDzlIt}2hkMY58>Vd{W}z~NMch%_UVOpo9zLX(g(v;fLf?3Kdh5|m@+Za`Yo+Ruf=0A5lZy7#5NZWQM>D@#H%tNf17g=3#((%hbs%=ojwfNG+L-@(5HQ0IbUCIp2;!Ro==t<*5 z7$wN2K^wkNsc6K9mOP-ZgyZqx&3I_CeY9jAF&^GZpUR^ras0W406rII(`?Zse#~8W zUgY(hrhJy=EjusKu~XIf=^Je5l>l>k;!S4Fe(RkVdvElKR!VA<-1_-+Wc3-cYpOnP zdSZZI;zECyQ41pzX{_zOBMNX+b|p5S3vrTmN8!}9f{Jb{rt?cDQJql zA*tMBbP7q2!ZB(%cReZG>PofBT*$pc_rb&Ka-@}>k<7-g zp*5ZcSY1mE%LkXybG3hI)d*MmAkctMu*;-}Ub};j{)a@p)Ay_r6WZ8E-wp3b>N>onwOM&%h0(WnZ4)%EZ~??T>r ztr>q!ilL({B6*J`@9E!PQLxl1joWikp~lRvk96J{!SDT@M7@7+VTXN2#=UVD)M`H)x)niFslm&4o zjb3nR35l>u-GXgjqJ!$3b;;567E<0Nsdu*;h_}rfhdp+F;(mu%@CpTr^wH$qu=%79 zT(qJa@JGMZE)Z#QVohcGXv=Jvx?7!UTUxX0LtC-&I&-)`qXyo4!*EkGJ;c?2dvV&R z24a~dN57QH()=C)E^Dph+NJ)XpWkyi>5w+8_Vhm#_PrR*Z;2yG_7OO}E)FsCN7Sso z{taY47SauQv?N7^p!v2A%Os^&NP{ z`2dWLY{7q-?RfS%X)LSN1f|wDNO}??lCBX+RAZ0QPh&)M1SMf;$#vy;xmcg=RlmTtH|G-`qao-6D^vkNkn#@?811-Jo7?0Sy)v}R~VEKoePRE zpz{E#@X1CF;UVN|S~iNhc?$b}9m(1(YC?A!>NzW}jJwg+L@rDmMWPZ822B)Xc*dovFUo71_^ierYCG8n+rR)lEQr=>^FSF{Xc4xbs1I;e6@iH2%)7 zef$#RmHZAre>iX2Y&>GC63=Rz!HBqKuElCD^jvX==~z8MJf_ZyUM?7c7uU?j`9-yK zgLX68xA7*HyYzx|e9)#}I`d%OnydT*En5lradh3dSEKmFvhLJ2z!`)2Jm0${k?(o= zizrMSqAhPz>A_AX>~DG)1*UGMg-^<8^P6^5992jX^M>JsqS>UPrGq;;wgSEIFyjSy z43!UA$_->R&{1A>wC~Xik~uq-uenvmf3O|Lw|gz-YnD6l^N<-GS-+EZFO#LgLB+Ux zei6E%G#ZDCTv`9SufZ{kQR0;cf1xjT(s_^84SeCV!!&eWFUfGQpgtp$QT9h0>iFb0 ziCpswy}jm-4MOf?u5TeKU;Tm#HIMNv$31z)Q`5kiHL|GXMmSk{;T&CN;Yv%ETcWe; zd&of6XF6fUB;Kmw5Ffg5GS3Z`(x_1x{I<>asipop8g_pU$M$Pti?5kzcE})Y%Jkta zyyJM|k|a*~W;IP$Dj>msYQ;(i;>gOV4lY~OhsgLSla#Y5Vn0$$U7k;%x3@e(AuR=@ zNqs6gy`Tn8oh{2}j?>}``mfXVr<-{lZa44kuFIQAU4k2uCZJW8%TPX^&2Lo-b07O^c*AN^{>%4xofy zd+;M_8i+_p7l&QijP`Z}AbF`;*uVKRQn|hwsk**KT1!8{oWDPq+jkDaEf@M|JM^V> zTbH2W?PBqq)?FmJHkMple~rB8n+#K1p5RiYCES1o3wu6(fPc7fu4tJ&{$6tt8QpRc z|8DN)nmW4R`Jg?>bwNCOEny{lyp5yX)gILNZ5sTwSHM}{I8EHG&Z7RscF4$eA~xOp z8tzpHBu6i1lhWB0=v$WtHMfcTvVo8#_y)j4&0@Q zKLFJ{nuV^Us^Y=w5PbO83fwQa#%Mu67@d(oN5T`6-iY z4_8K`xtl#XTQ>z%|KTyWQ|=IXtFRm`?Dj+2##>>c_jdTRaTd{=98W5^52Uw!4+)IA zC#e+@N%Q#WsHbdQ>+*NMq{eWL5o zi)LrU;lD9`ME^-FwZALI=gL+hv$FBE>0h4Hc3oLMV{{ljs=AIgSZ?I^&L749%n{JR z@_i(9<3V(3jy&C0pvEP2JK~xDFn;QLkUY^6LRc(?s=W^4-H%LId?uNd3t7r1?fy#r z?a$CdPk(d2eox{Kz8y=Fo@Vh&THX1upcH!M=ogf=>MXUK9LyWYj-e~R+@*7LqWSdl z(R6LbS3Ex2gB{wn9ND?vL}}3(cz#O?&6=A{O8ir3w@3v^y?g;jKiE#@j(m(CS&v|? zwJq4a(bs8k&IYRVy$J1f58+1_2GWDfIHHv~k@fVP%Rb$=o=v?w5~nKZVYM+c$>m5J z`h3U?CvwTOHbM(WFF%FWSkFWujf$*W>2h{ntvmanw3YnJ93*?kIMHRFj+2r5&lB6l zE%Z_RCDM0-M|H~=BANLK$Z(V_d!<0Y{+x41{4De^cjl))*>gIWh;H7YQ$J7PvCk!J zDg750*^Xn?ew^f1tzC&W{Mp1u|Ii|BLHCGosFRjV>c`7h>akmc>ybdGjkcfHqS_}9 z(aXUd{4$+8H0AhA)VHgQ$ut&{fwYexdrS3s~Xx^rw=#i3#r_lFx>j&Aj>u&b{36bi@P4Wo;%R#X{l78k&3-nMY1UlSJ;fohe)w`G~T*K$mS#zBGa?k zt%d_AyFd|7{{92KYgxv|%P(gS z9J$5T-%7yCC5ECeZ<1)-@hrSc=#A@+n^ME)*Cf}G%dz*mujqZ4DLXtNhP^gYaoUbb zZKQ1;hn7E#LuD+&V+y3`masN-?sXJ4Z0^AZFNAFP*w;YLdmQ;ube^uYR>s3KvXR9G zS5jN)&EDIV%nmlaVP{nv;cq*`amPJLCOXd+o$@=3E6XhK=jRHjt0)iNaJgF@*t!(d z?L30~p1a|waDA-(fJYa??~)29DfUWIC{{;O_)q(Ew#>7Nt&z8$W@2+5FIB3+?^4y- z>4*F9HkS|hj`LmYwMD{=?8Fk9sC#=i%JcUnj{H~D|J{YYcy*ZOodb00&@McuN|yaT`#vskno1l(=Te0(HNH4E zmrfp|!>{RhK!?AZ(cJQ3T56C)&p+D^7L1n?cUwz~er!?VYYcSxAm^QQAmbU@HV5*{ z$H(((m9lu_6eqmcEdhOdABmSm%Ao!J{b-WcEcQEbX5Yl5lt(U~YH|OdoNGrbva*EvfUkAI6R1Se;Jr--%#&LRS^tnJT6$E}GM-MQJ1< zxC06dlhMCJ!R+3B(rjJh88XAq8K>?V#H#L?JvMPPU+&UQ7N*Qc*Vu{3+42XDUa^oh zoc^UA=X3r%(96^GE#}l^`rlg|iC(@dT*h2m1 zNaNsX=%HqU*L4{1KZ=xTw^Ruw0rmWuTkiZgdl#g)?l%cL>yEQbwJXIG%j0??h3q2$E9lV)z6doq8Xjsy66g&ZySg7EZRw3wLD6&Eu=lG zPVl9xI_b=W9DHYR0)G2H6rFcGR_`0f?Y&pBXHjOx^PKxSN=QXfUzJLOQbZ`Fv}Kkk zMcE-4nVHWy_kD^eDzwoa3hko3f6woa=Rf~E$MJIC*Y)|l-&c-;)}Jcq72cUQdOsn3 zZ8F@qZJSV;?ONvbk20JRz7>gXdI6omx^(jW8Q5xRE6N>|MnNl|k-FqTGI3D=ah_>S zQl;&odshr$m>@bWLjDyPP8b+KP!3I_CYO@YQ2DtHlIb6;|TUDYhdqB zBFI?pB@+|Pko3#4q%EZgdmmUvX}Z}ku7gSZbJ5b=^^h16BlOvXNWLaYN-U`kjwGwL z_K_n?glGNJrQ~ZuKDX)bDEBZ%>fG_%WK#Xx1h!1Ir$ZMX;f(T}xf5CvoaPlvB-p8q zmMou-wSJWILkfKa#m%k!s^ZC%lHFovzJ@7|8*$|Dn%&$qMF%db<}$Tpeh;-~d^5!) zmJ;!=<)nT?AGY7;M4sjEA-{&LNe43Jf2j~UgXZ?~r&S90rF)&YOIv$5UoU$I&o~8T zZTqQZvushTM-N&*Hw||<0yw8n0ojwY!Dan1DtJ!>JXNS*>FHr)devHdPp^P|Ak5ZI z@Hqjki=t8XvrX*#>@L(+xe)JqDCEFBpGFsiY@)N2lVH~DP^OEb;a}Pu{tQjdA5-D|riA=h5XKjzFx*Bk(-q03)e2 z@KbR*n&qMo+B>@OGyevVRr!D(I~l>BAFAMLrjPtG!jPG;8|3@T8x8`Gm@Sve}VeoT#Twd z{RI8&Ld0!&$G>MRgEQ=O;HdKu^5ZMpfHLK-#w-u*6-C_K{Zs<;X-F zuE)lA-LE%2}P%HA}37z0QiOr-}*o!!OP{41FasEnW;62H1*lxm6=G%#A zy6L~Q^hn4#c=u2PwqH)7s)ft}l_3-SdD~ZZ(K$sVa8bZIma9mK%6{zAmMr|gwGWQc zQm|-SDXkh^Nbi`JLbvUGk9Pl=3YHxvB+a|LGRUW#7F^?w?psdMJ0& z_tl!4x!#boOCKeV&p#uRZ|-N0-H_mdA0(3-dv0MaT$%grt<04eFTz7cYDDU-I~!o8 zjZefllfTZZ$S;j+=v1JEz~rGjQ@&%Q<-M>=ZfUIsJqm{)j{Aj*9Cm|j!Vt4I`U?Lg z$z#u0t!8>XcHoAH9>P`}6UNIPi24Ws2~@!C|X{#H5|S@nQK zpe}hcsT=VoJVSNY#c=TbbxdVOJfGY26ZM8|L|T_G2|8eR4K+vhXjOcQO-HzvL3Rh?mSnA-g%M`xkz@S%uBZ+l6nJ|3h9Cf6$YLn@FR6 zHY|TL1GMZZ{+>qx+|i2HL~m4@GqtZNyR_P)p7B4eOUD5A^+H#5T`HcLU-a8H)MZy71=r1%08)lg=dT&BUVvIu$JQ` z1@X%7^Z8nR@zBr5>1LOH(p)@mpY6;pE8jmD0-=e|7i<<{#_6R`&aSI_qUQs zGvrz0Eiq`cxSAauyvufyPHJCNDC`WH4iPVRGN;}LAo4_x{E*_vADx9{@h4OKRH~Rr z%&%g4Mz2vY7c)nV&1tcZ4zzGb0;_jx!ENzGRUnbB~cH7g&bI*W6KMSe@9`xK#C+I_QyXkCyu@)c2Jm-*KTX|Mn zx6$*1(&-U{FgmR77`?DOicVg>j81IRCsRf_D$}o1u+Xs)X-scNjlqG?=-Y~PrX67` z?`Ja;RS)Csx6TP1f^H&Tzp2pvZwo~0p2a^jo>9ws^;tzB6KZvI5;^5+iNiK(I$Ivf z0wqlkNO5B*tMs!d)UFi8Y*WP7NI9j45rjIdL3W}9_Ky;JBG2qF%VrJZ{m|bNR`g__ zXu5csEBz*6J08}iv0B);v-#*O`Yedk`ZhuzJGOxItQ7cVmkJAJede!SYs@w=5PUX!0B*vb~5KYp>7X7-S08D$SOeND?RAlV~(emar+V z{?rYFEKmuKg`wM-a5K=0ySGM%L^PeFzR|1b(p#D|ZwnqeGvmVjqaVh} zp28g1b3C`*mX+S~fO;ttKyBYCJR8TjAt5oYu$b$PYasuG^J?`kX0)wU4cyvZhm=<-;t(l4 zT-CgQGFp_4^$He{EQvVM6txEP;+1h#i^hO9@HzFn$LW`&Wt`IHF09|5c*Dog9Mre{p;8#sgY8Jw4G7Wc+SjXsbeMxU##gOq@3YHQ9?u-H5o4i29~FORfPWuFr9PSX@Y z*=c#^@l{te&>@e`th!BQM^&TB(s$$>tIHXUNOPCQSQNClj5=2}#o3~L8~WKkjMpzJ z#K+cFla;0)iEN%B*J(K)$vLdR`nT3GnO}C2L7zM1_vGi4FrWcuPjx{1h!WhboJ@|{ zt5BDnM|d}M9l%Mr|J>&r3u4i01>4P^WFMQGBz9x#*3M=mMfki%X9D$WQZzt^dwgQp^qK&F&^Y=0bY2-T*)*Pvr{mL% zqH8C*nfVh{X<{}kjgQ?Y7-RouwM19^HP!pDmff;95QXR$umN+&8H>JM*x=(CMs2Ab zobOMCkU(YXT)P0zIHpJvW=%wA79XXGJmO$wO%Sz8--T`1CC85S<)VUrdmvXo24N<);e_bb?5n^pth|gDF5I@9bgW!QZf^U*=4t<+mLG|x!Y`yS zHfLOz3C5!AlG|sn=tPcG-IC@~Y!(1hvj(+1eM>6S3$aV$EdM`hsy*--H_cfnHr24^}U_;t{(GGZ3P!1aF+Th7+8T!ju1DGX!#u`lq zIHzh2B>&sz_`2EM>7{g~!1mcRC$#J~v39mK(_rk3N2sELcfjMGmbDsjC4j->I$G94ad3!NkZOd@jL>saQ1$rY07^pi409mA8f#_jmC0mN47;yM(rQ|BQIw zcurI+*Kvwrdubyf|2aKDj*hDp!<)ab_=1cZx3NlwIn|iL?v`E4KR)h8qBCD$r-ja> z&R?3|niz$4t)Bo5sfK7&?hO0)cNAK2PmaIDwGR1LS5W&BK9V$KN1qf~v5?LfugB8B3QsgXq3wu0_B+Uz^Afxxq)Sl!L#&q6FS|?&Mz2NUm z`doxK9r)OU&KmBA`QCG=R~-!8+3*qOhMGYE{gt%a)|2?<$01^p0(O0C#(mKHN6mVh zNan?j;@4{qpg&%gU^!=V5v)T)@VLB-}qp%-W+1L%hFwHXW?eN zB4h{DK%J;?#$-8$Z2D9*-1V84?L9ON-D)XNVywrF!=UqSE;%Py2m&KC;4|%Gaw*f*mE$9-lQTXt%f=PI3 zgOW<#qm5&RP@y?U9x3j?^<|O}s^|t%rT0+i)lM8!nnvB;H$+~EKZT@~lj)d**I=?+ ziN>@s_h57e=R8sb$Ey@*6P0vW6T1MaBjQo(vo0a`X%>a%*i**Sp0~b|D!^NoKA^UF z@Tp1Rhhd$@C1^WN(~sQkVETfW;1a9=S7q8hmH!$`Q#fmcqE4O_U|Qgb+bs;yDP}< z<`*2Xw2Bq2w&dc9D@d>5XY3-az=>Q-L{h4AsNQQl_j~Jq+Es)*^bZ)V{^`C&I{5!J{xBw%%B40W|D`e3b5+~8@zGCP1axj34Wq5AL_G( zobHGN=m|g;b*?{NohRrvkP^*2t6W08$k2Oetaxn z1xCMyBE<>8l+)?AkX4cb9-G}LyPMxBY;MH+@Jtn##jDbmtG}Y7n^o{d^At!pEKLVX zHA4A=e4g*P4gbvkF{;7o9VM~x0AJ_KGq(MN8QpR1Bgooyp&F6X==TN%xQ8U+XyZW= zpFEwkmZamarG;?vVF~OWEf9t*Y#{9HY&dDB51F^s=?`BpFuYsf1lQmry&j2XuE#c~ z=aMB_r;+ToQ%L0HRXWBvgDx7P=*6?!LBD+&wZ*yvDb4qRh4#}xn%R!@1-Xzl=m#wh zzI2eUC}|EGN4fuYfu?^UDC;eQ5XTBQX{kn6hWrBGq~pjet)n%q>?!kWempaL-G??a z5r(j|uOqi*ru6i()wEyhCot5E!GWKDGNlg3spjJew9@`qT2etZq<9wK3Yz4!^Hh8@DDf+WN2_00QkB+1|pvCbL(0q9UWi>mJT~Nyd-MC`( zXjV9@)h|PTGZ&?)?F#f8-!`hq@(*f|Yye}AKSCc=F2wdHKu+CENUq<&Dm!=Mn~oB6 zU&}F=#~8rHHP&>~$}!Gt zEAOFL??|u`x)Mit67-zgWiY|;H^_ym()MyMq5t4t*eB$Do?o_~JsG&4+K^0xkwq;P zIO84lC()8V^HYW9g**g$iX1((%#J=Tqezck)e2r+!eBkobVeTKe)U?k{10vjsCvc zjP^Kw4^3|4kYec|=+5${dz-as^1hvEsjFkT-$QIu z6%WLocO%f>M9u4wfs_a8aQ1Ww_5peBRKpC;=#M;Sv2+Tj=$=Y$>RAzwphC`}vxytZ z5#gjRE^&^WFT-2ctkC+pcrJIR@-^{_vZv?oUBxYMwfRN2 zCaus_%eIyA;Y0cW*c`IPH@@Y7_RDERd*&ply||AosM9CoGfcolWIFx(X(r0m3TFc^ z+CxCC27NwbF}!=X)A@7ae`L!bLj}HA#GV^GObvA9^ZxoyXVdfx@wQvXK-JTm9DeOj z8tl69!u_|1lgm?Zeq9C~_jXcxJ}mUOh12zGwdjYg4xFq4B2PoEur5uRjQc9#R_Xqx zRv&eFc&OG44=%UCk6F*~z=p*LFMHd18}6V=eKlT{#$9Udh!qGrfDN5+oc%l}2_{UR zMn9PmN`IfcfL~D&#~+V->+F8Sh~6vIA6E%k@Fgl=A>g$<-QPR{&-a|=nr4S`w?eCLh!@$Ow1L}Wd{pKNYf%+E~o7~nc*=R#pxS3hik86_`4?2+qyI8fM>CE zPKzt2GS7{h8hD#b5_(Ybp2zVc-mP}-`8t7n{3w%Cnli}M&wdSCdfQPLLfFqES@3u0 zKGkI>LDi_PCiBbQkjaB0z>z;T!;xq9k6`}UAd>EN z6F)Imgdv|X%EYmYU6h~7&b%dbeCCz#Ho3`Tvv0C^>E^8@R+xd2e5{G*%>2kdA|VFP zCYHm)cS`h(lKIFr*9|CHf4sa(2`d^}66^22K#j~JOS@|@Hcr9z86u?S^F{p1=P#Dm zHX(tpkMU8eDr?}`%rb)=)U<*ixE*^L>c~YP7ZB|;GLfvTtRkZq%JJ{YJTg{wi1c6G zMea|q$8uMnV6oCmlKeu*+pf2-i-;b>|1LO2Hs5&Md}>Zg=pyucJ7?CJ(P6 zT10Sa5;^fUij9uYfoECI*_K&SxMFcC>8^1mAzpa|Tc0H5uQfU6oC?w>yAZy8bpac- zakMFdN6^Fzs7c%(yl*rL%8L&1-_;PFVonp$`KmxB;x^n-@fTI71n@f^`C#1*P2`-a z8i^aS5(W`8iAvUTaw4k|o4Ks!|GjVx8Oc}jTOUf2vhX{CD|#mxJTsJjzC<{u($<6p z|D?E!OY=#0(@9}IRGtW(*;vKy0(!lJ&zZY?BOkR_qYGVuI8~>B9bYIw?_{N5d6zr6 zwV@iPNW|etx=`pjaN)EgEIFsgZ$Tt{IUTFKkiPjUnNCU*r^WuM(EZ0Up*iss1Z=tE zD0s1thpm@6lL?2Y$u9b&Mmzzx-j^gldc~pN#sKbD4pJzthDqBXNsYH{>pFJh+Pl33P0kshw zT1V_S{wb0~3`4Zp&G;kPa?FXFVfKm3xi>rL5sJC(3Uqg zGmnIMpsJ}Z6g9h#T>rcZnmfmF;2&!!Uue%sALt?fnl_W52u0!^l1Kug@?gLG2)H@S z0;6{e*!A-|*aw9rXpX89F?_a=80@daIe(_(r+=5?!naqjzw2VQ!r~U|vr3nYW!xh& z%g2c6?@Pq`;S82rI}7_~A7O7zS?wfUog|okShn?8uQR=k&7f!Ny3^DDog!IJyGgy% zYocRrMTzb;rA+hNcwMCm+&YC0U<}T31h! zPFQ&h-SK1bOV>)aRB0oc^VmiVyc*K3d+SIB&@vIe!KPpAp@6=>ty z0`$@R27W#!^sgDpGq>hsGbN32s3gb?nR*RUN}qU4cD5nwYNCl}?=q#zH_jxYFT{xC zCqueu<0UBbolKXn%)-8r0xWu~6WK)kL>_LR_}Ly=Z0RH63`#{Fk6n6;4>S$qL;5qg zIYZ?nJ=X($m8)Twgz2*nRyH8n-t(w$W*?TAu?Qd7S{M8Nscz8XtByiDEv6TQi=`mhwNIxS6tTxs^{Xc6;GBb)NjaYhnKE?DW?b9Bb)7_tpM zi|4nbl1qQh$fo5FaoL=iM9%yadD3+hUFkD{!3C3{FAG@DXFu3rzYcc7-WoP)yDz!^ zpbIa*-hm{BKd?1z*|1pnDC`&CV$EI&UF0IxpuW=xUddI!!qGUkTcd!r^qWL&7jnJB zO_mUygGb5Puu;BE+|V;UTD(0CH(2CiC4z9> zZd!+q@p2&c%>@v>+z3YAUm%??i8E)$vAdEgaDk;Ox$_~1A2R76ia5Ig`ZO}2BHjzc zJ|5%#o!8~g2YnzT`ifAJJOnuo)^s2p51PL(L#Dz+P!GQZ4@3-TzjtS$u-ciMv}!+B zP<@@uGEt$H=U?HyS|MQAkq6{=AV>B#^RR3C3C?xtZZ7uybm7lR;Iyk^#D*~^Q~S2l z^S#U9)RrreD1c=VeTy#BIIdA;RL@^ z)Ml4ac??eFmV+yX=M3Ieu3`ntGKdbbz6*(zK;CH+NsH=Ve zUZ&xIN;2{J2r05mO&&Gg?*wTT5qetUWucp|i%B!qL2P~rekfWaxOX<6s!T9NC9m8W ztH5h4@4PX%-47-5=QS`4R>J+@*Me!|b^Q8MlhJ#Sqix)$!_DUNNaKA3UR@~0dc2T@ z%&j9hIE==+c>z#9_X=$Pk&5#oE}#zKyCbns2_AnoCD9kf1;Yl{v775#_Jd~g3%PUiEfy2cVYT~~u>Uk>OobS2v{8dN0uZUso)vjSb`g@qeurCMIXcx*s&BRIPuyQ z+Ep@%k$nhde(t7_SHBek%{0i?BsKi-5r$fRFm*L^6A7OGotYS?%!YnHj(25mV`uP> zLfSGJm>LuUYY$36yu&EF@oW>kpVtk%;GZz^vy$cISz_^%y@Vfrf)OTV>7Z?r$Qy8`7O=>;LffNGF!NA4{Y}Qc2E@E7+4S+?TK1O-AFq zh<98LZnacIZ!9|5Qs1{IT*H9`3t2YHFD9T}(;Z0H!3Ct)f0&&Rmxpg3GDQC!N4Rp~ zAbK)p438e)fqSzg=%~}T;Han_`VldV)I-zxBf`H;kG3li`$LLOgG_MkRpXm#?;^wf z?$|%d0i4$zK*_(eAxi!Mn&R>sZ5%l!HgKqdVnsmNLD5@%Jp!dzcPH6 zvZ8u)M#=btczk8xXl-nNQwwvLk zUtW)Vb-s)L?SF)~PF+NjckUwNK@4%8FGr>&Ekob$UBR|zSCdd}HL_bwgRboyLJ3|u zxMD#$e%@cm`W92{dhLB|%%W7BzAl?+Z+eN2*_EJebB9=8u1nZ?RwNsLE+SLx_Yl63 z4mVY&5&kpoV)u9sP-5aWxZ;Ha>}_1nkIc&^4NVisk4MJbit9Is)#1~4$E*ERS>t!X za`krn!_SMnmz6{$V>WelWfIOhXGA6&93V<_Hk9x8unlxw+JEKSr0TEYWZm zknBBWSfRrmJ&C!=TJX0cqM3&+>={aa6`#lix|2r(PasXZ3?xmDz)a;@C|s8X6GRQU zvSUM7#N8h9cBR7HNLzS&S`KB^U*|jgb7rfYZ|bKg{VO} zdZT9}H7@K9tx$Y`^7OXAf%aW27yTA%Ph3WA8h%E73;B;-cu9eyjIHq|(FXi6B8O7j z8vz!v`t(grisAiQ&)$-sjp7csqV-lQ;kw*^*w9NF{pp>87hF3Fft`1-<&X?o*{e^# z3bz;L#tO*l|JGvlJx2&fLfA^ZgM5Yb3Vv%y8txMA+IyU3A*+odF-c)W(Nlua8=uSj z-WVh(5qd#mQz$xOwJsb}2CmyHmixYB0hbf^gI%8_LcA^~u^tUFY)ioSn z@;2?@)_MvI?#r+V#ozc{QctLw&4<{_B?dT4`6*6Nts=8`>vAR9zlfb#19jJJ7hVzk z2>)l;!}1Nc@qHpQSnE^A$^DP_$h<4A+%(w47(Uv_M!x>X-acwi?-IIQLPhRj%S>rl z*lq<;^*8a%);8fY)T3AL&;aO~4;dR&=%IxBpq_LVr39M8&KJ*do{TQpbw!QbnI}z_ znr7lj2;;p*-oOPvqh2oFg|*Xkpr%jA5}Mb|9-Ep8z(j&tls>)2WijpgQkC~+SdsTI z(ZXqyuv@ye{y7}kbON5e6LKJG7SUURj?#W^PXu-|o%zu-YXo(^sbKHEoL+Ky6y(pv zv46+6knC+qc;yy*E_wZRa^}xN(rp>&GDQ ze!81!lXaw4%VeOjjr-A0S{xZpmeo-Y zLukHh)lt%@Uyj?)&P6Xo)FC_sNC5K2Uc2X#jlvGO_t-l!CvGNKc!|Rv`F!|tb{~XI zF9AV(BLsaK11p1C*uGpIIwd91_3bCv?g&|G#>8o@d(4OUi>^AeyI;C8yp|5CXZLQZ zS!@_xJof@BoYm;4qUp%_^D;J|Bm^gH&7u=4CeS{0Q|Z0WMDZQl6GYONBAR8#m_v#F z{6&Y3QEx9gL2ZQ}S-PQtaOrPQQRHp-nzs~~<`uB%R}v(@IYstF&w|3#&*)8uKfU$M z5PLAC4plyC1j|?EX!37NuQuQ`~6jik#DLZkYclkJV@}44nGc$|6 zGsTbo{Bu71$|(eOmxJ`0_JfT6_#eLEVp-?d7ArU&CIxZ>7Ib^mDaezaMAtjcqQjKL zs4qe8==bSt*1Dzy8yOjqs`HQW&EXv&#s7<}@6m80_W+m)DxoOT3zq+wM>jM_(Yf0W z()}$Gus@53JQ82An<~6u+3bDjnyd#l*!K>Ns*a%(0ryew`8l-TQx#gn_AB_CRKY33 zYw#zh2dzb3wCDE;wEgCb(4(qKFK}?8WuFei@2u&dyW=BdYe}Ljfx-;MF%>#Z*vHV{ z(Zbt$^$>0w@$*#yc(@fjdovSa>BdefkZo zLa-*@78(K>Nt1=|{X`fUJB^Oqy$_2!BWPbqBbo}WA`6UZ&QyI1NljJ-y@VfZ$iH8x z+qQ=F|3=s#$5jHiIDZDurdb8WLP2+kJZ$d0izHsIgqwKX8-YWRvzX`C75*V-iisZ!<;9_kDs(d;>&#?H9`>$dlj z9CH!Q#@UA}maHVI@6V#*thw0WXaTl;FHgFrrgNtY(zx~}FHYi;EqddeN7|3&lI<`5 zLl2^2aFL9F-S_1_DQQ1XW;{HK+uV+j?3ip^x}}F-L!QgTLbXE}cTy!U>T4 zISxki#OT8o5hNx>p0inSg{-2TNa!RMpMS4_w~0yN+`|*doC=iXl@fePf{2pE@uT=+*eEJXIgKN<{78|K1~IIQ#VW^x zg#CzZbTLDcDfczdn`pezdXOOKL&-9lc{m_23hy?ggMqU6gaS%sWiUKnl9YSKUF%P zlsz{js;9b9j`KTat$rjv+R@31@AM{5Y860il@iu@K7+A#(S`882gRqC4nfDXJpEx>k#t4 zuFZLe@iL+fjp5tSLEbi8B&&*>Rb=oITt?iNm=mt3kFmR4P7XhOOzo+kLLNn)#4%kj z@Ol--dAHVTe$<`4R99*_)>^ayXT6ca>MLDPtHA?wDq<%3y(bl>#7u^yHb=z!vIrTK zW)cOhOp^6J5q&uzLyV?eAj^wS;*%#l;pdrn(7EJBHZP1PMzs%dOI8d+dB+P3Y+tnG z2gReSa~rVdgsUjK?k;@3E=EsTAqLKAzLee?%sz@(%y<8CnW{B@jUsOsgR$6AY$scc zT_3Bkvf^iOQBVzAlOuw5Rq8?4nFsJy^gKu$yF+O#YQyu7Gx$P%7h0eFLg;2uq%UY0 zQL)}z*)x`pQBZF?XqpN0LT6fFqtk8^9Vck+}v7b z{E{>~Z@`eHqjcE==G|1sQyI$5^F69fBe3W?i$ZrdvtOq>p#PpUAk(ecC^5){ni80T z?Gj32xOs3$pt@eX^-Iwi)^Se>B`6cH*&<2Q(`t8W2W5qS?>Gc4 z^{J3s8w4SDB;fjA31|}L3jHAkM()T#_^UA_&wq)8=V7qzxrXwEnc1x8pZFWTM4}z% z2~3ynf=~q&x~Xm@>^ZgwHXG{@cELRSL1i<^og~4zUg?680!tLY^${N*WpwbgEeXuZ z#3ke*@+op8Z*yhHX}1g5oTV{(dW{OxBKY9A3dwyEMZV8o$BzBXV~d5I+K)qL_-U7p zQU_&sQWbJe=(PSv)Hb^c<;jbK)VZnfkpBuf?pCEVjBMD_p;mtCss~`Em4tkK+sUmc z8#0((Pb6Y5W0mDo@Y>s**wH(Tob=fTQ^%fRvCK|Pi#@{WPxPSankSA^>tyy^wk2oo zq?2VC9NJc|2<}tRH%eFWgXR;eRTo9@p=o(xw_>a!Ld1OerI`#EjYe zSp+MYwqv3dkCx4tMkLC#xW*Y8xVb(HE1vE_*Sa_3S&tW?j(;c6i8qC~?V>v2YM+rW z;{tr~*Lh}zM={d7=Lx@xB;+x!5Aac%91V@EK1;%KpP6hlsdxP!pe#^GH zZDWb>E}zyWLTsP76J}OAr0j}^=rvKuF+>raDX>OP0TF0%WGeM0AfQ$MMH8-wKT1Zm zuah~~9k@=526(wZ3z%^a(2E^LpK`{~EcrKRu5maM*}RC~r}Y5&R%=7W=G*A1uQE~T zd&}25_?(Hnupbo~9AhUosvzBsxvcGM2A(Z$PH^zZ*ik}o!qC97xP^Og6h zyqtyDZ#)N#--P2x??bG`wCnhGyE{quUW(E^yW!~bGzfTCgO+J|0!lmy%AS9TT6rOe z-dlsmRo9U?{yZ{i&SCOzaUI$BHv)N19>fF7$I;6XBWf&j1g#$}pmwZkf@97ea5k_R zx!kTqmSMN4BX-i{#zse~faFtp))Uxv+p93?t~>Z;S+b>*Hp01^(`o(PmbC6pWz-|& ze@yrK&FIhEM;rEt(w}PN>8U&Fkj>FmX!e0|{K?A#@eZYuNiU_if&T6EEOQ>+y-bw0 zzCN3}Qp+N()tiXN>?`P&*?+jWK#W_vIv7tL|Abv)__)y^o{qY11x$zre@%rZdGgCm z$V2!^r1dV+tNSWwi_N+8%+2fgwl|z`(K9j5ucj5-<@`c;WHmRpWg}^{sU=e?jX13! zHGzzbv0xH)SMch05}b=|M~dT9$@3&zq!92$=oY_;?CzJc%HGFNT$(;SHZH;6M2B(C zj&wZnS`LbmRK|Adns7I-lRB9)iAr~{Ky#%g3d=h=WJhEHDLHl)$v@FSCv_yKtXzHi zQGq@Cr?-$jewXFTSee0=qi#?)+X+uQp-=X|*~d*@YDVk(Y^D<%dGuTL#bmXV5n5O; zV7L7+p<0P3_S2W8ByNnc>*QkDz;hA+;nCP6LL8Y6cu-b76RA$e0OZi33@@GIA;PDZ zEnA?-ec9K8R=J(TJ<6k4^+Y+-_e6%$c%dQqrqMyomP|w~+NLm-S&v-272#rj0E|6t zA(TuI44b_Z&VuE*{0{d`ck`+8K0Dpkwj z?vi6LYf=>qEZ#57Y&L?%Xg1Ms5Fr(xB|?xuee22+1Wq$xr)a>I z`h6Uc-!e&9E@I4u?AuRuyy1Eli{V=D%9SP2Rq)g z|%VFd|&Jrr1F%Y*U z?j#YK_V|a(0cL8_686mQPVzgt67#criI4^$+Go|j~AM>Sg?b%I!E$CD_7IxHu8fqa}X8(zFHB8^XE$xHKCd~Nw8cJ|4B z+=O$4J@X8Z_rH%utq%)X)DAe{mm2o>N^{cB zb|!74TVO>kKbSz49Tl?Xw(+>={46*btqhKR3(#Jniy@}u8PPp?63)F?!{;Rell#|2 z&?0o8+}l(^!cI-&#^!3m&Plr|rSUG5S#5#Wi^x*dy9e-Z`+TyRcbqgY!KD7dO?Lle zV`6giC_d>|!7mZ>tm{sx3K@vIl&l2*r&W~E+3ap zu*B0OzOyH@%BZPZtJq71B1Av@FG@Wjjm}(O!CqB7LfNO5Qu2$+@$In9d}?ETJ5JNg=muDyjv3JfGp*~gojUo22PzJ|=oQX--*n5mGg zBS)%tz_H1_RAQDHn(=3xvHTEB#d=DC^I8)|Tj>{*xc58t6E~An%9w165uP&+rPA&y zTJ%i)HsT)Lfi^yn=6dHJ#}9^MXuYBLpnhT^xx3T9HK*reb5nvL@7ewl_<2ViYzyqk zsk{d;eAS$*+P0H)+sv`cn?){IlGYzZh3)^@!X_Pc+Bub@8 z!`XX1>m-#5(L|XkLu4qTGW-oiNi)$znnVMo`Rui}G>{=t2@x_SvrL)wp7-nf?cLXP zKAf+6U+3&+J@#{h!%T7Xgw*M{WjZT7GO^GP1=@=KZ_9XtbK#L@C+A!zT$~v2q zL-qd!nZR0YS@8Hz2kQu4;f+ZuBAXR=M9#s^Y&hY#^R5A0%4liWEB37i%=1IuFH6uC zHVfwn58$U##yI+xc5TzE2ewXZa;?b|4|4Lo4!7H0T~wVN0%PPBz+SUH-H_Kc#ri>KkCG#R9;m&LhaU8wAdr_<^Rkc-m_aJ@0i+Hq4z##~O!0Z?hD?^!Z3g@Jp)i4yoW4DT~<})rL4E zVIH-*pn?@zW}|n52We`ntLVafEiQ1@M6|{S$(N18!|{ z5bF4=Vy*NNd|%y32=~U|g>6E7b6pDFQ1Km|>M&&>Zzt%Bz9K!7f1KW{My|X47=Dm8 z5!}}+()?;G;%^sAm)krfiyjPuOTvF-^@JE&VUY*Ym5Vs@9>C{|j2-43Zxop^K zTn!hLHh@7*+;ugf9hW|T#%(MyhYF2sDBExah0RLf%vN1u=jLT|8xKj45bvKbM&+~c z=J^Z44YSYTckAMXU%m8%elv^k@ly$$-vT}6Tv9O?@nniHa)+_-!KrjSU8pbYIj$qT z?j=hVv)5qto+NCjeIG*Kt|ni)bYQBrBo?iBgtw9ta8Kn9v2VS`y}sKB>jrYMhJLzm zzTst@5$42SwZBhV_Lp-`Sqs2ozO2L8M0JN;i$Y;D!wYwBlXb8;W$f^UKEgXb9^u{( zZ=fX^zQT}!0AaOXjZpsGu&5$vBKcUd1e=!sW4*6Sf=|7eS=qgY43_BNcl01u%e55x zHaQ7*PcIc#?@lDIHpFA&Bx~X8qtduHya=5CxZx=}slwh##X=d`SmDc4ZK9}glEit} z7LX8UHx@lE5h@Q`I84*Y;Jl0L@lho`Vb*FzF<zdpGxzp~^kf~#MST=zlH)Ax_ zR!-&y7hBN33;bwigE0-Ndqi$2Sdp`B8eEt|v1r-%nVj&577mI1ERq_!hnl$@kW(a@nM|5h;%HOz6H>dmNX+xxMlC&bY5dd@F|tI4pQb&T7VQ9fvS$~Wbaf$F zVLqB$W_g@^zc-n4K4K_blHyG6teONeH@D-Um$meH?O*Dtwv$F}dkE_5tZ1y|H6*W^ zz!j7x!bJ~%hB~jHAFZ81Z^l~sw(S_T){`cG9vU&cwhooOx0a~M9p#oDR;8sqt|&b0 zC~O~$qH$lg(1#mm^3f{woZ#s)?viFxoqF?IeBf#l_KbXhobJhsdrijEm%rTU^!9$V z>{kQ)7q%QEu?k+l9bx88K4!M~<8vO-e09Q9zH#SBk-|_UT~+K&z3vPHvv)G{=hhBb z@?#M3OZ3@!JN7f%>ysg=d@MYaIKjRBo(ihAVQ@NQ9lF1+2U!-M6dgx<*?&h$Y3phw zer&BNf9;bho!Z((I-R9y-S=otp>8qi;WZdtw?E7rB{S45KI`dizDd*fM8MR53B-QS zcx>9xj@>mrbAKKj;kVqL!W+LTrXl6YSc#pEwL$! z|3}RI>*&^+`E>KW?_|nheSF{i5UyxIxLNf%n7Jj8ik8hVop> zOOCYud&DxiPpQ?4qcH!Roba%Ts;FU{j>t_Ov8(5tCmWygFn8p8x~?D}RJuJNL!3Xb zJ@o_k#xBAIW1MMDs5=h6bP5z!Ibbz~K5&|ZaKQ7mxYO!5N&5Xv^rOL(^XDBgA9)PQ zX9-E?voPWN-(!S+o{spw{ad1SV5*-$2ZGi z3)cjeTi3=G6<6TTf0U^(*qnYTD2JaD2(*uqXNHP9vC7MLuaQ9 zejbfWhn;XG#BtJ%zA*Y$KfDe!!&A4!&?<`|x^C)4nx-8Jx4tE!sr@c&xRgGg)DjB^ zU+RbkWN!>Ty|% z*L`NlXXMY~pKe)54`jU{`}A0>tLnr}zMPM0Hb}BtH*Ui<#SVl^3didgO7T(S)98DA zStO@zN$L!x$;!*m}MX4ZHk+BuZ4%FR4}-`32(<$Q~kb z9mqV^hvUcrPx-P2AMtudmpN9`@P{tkspSHiJhz3aUObIAzCMe4=Uu{*dg|Q9wFQ_? zOvagmsbN?c>)SwbhWX7L+O} z^S`fm&~G8H=wG$}(5BPQ1h!pf_5QH*-TPkd-|!T=NwpQ5#!3tCnSfA1Mgi~mFg()EnMQ=u6B%-9t9VJswROIT{$Uk+R7nc=CK3tybJejKnNZ z?o$?uTT^Ynh`&|)%XTCqWA;I0f9o#$eFY0P9)H;46gOHjViqmwY$uMDXUKw-4qCcP zn~$6A!hg#i&!^XW;uX8a`J=#D_*uzOa_XTKIdMOaZjhTQTuVOVxn0Leqe2J?x~7kQ zR)1sn49~;wjvfHJDtUG*PKK2BkwoE+9$h%W7f(Ml1(vLrCOH(OkWUAlQ49N?dzRU$<5g2V~8y)VzbT*&Ldp*GFgG<#5`R+x)os z=lI^^NBJWk=kUG#I{ZsjHGbjq5&S>XbF}V77BPwl6#Z43Sd$jO)TKcTf>%CceCA!fld1$yo7eOC5|{Y{du)ZD-%JrY*p0?x$||Uj z=U4InzrL`=JdK`golf0$HdBW+O88BoJ9T`XM!itJ=(O4%?qQ2bx0zm0*CGI=+cRZZ`nMOS3j>xA`hoS%2i8{8(B)SEv|E?m@UL~<`FtG>o8e1 zxQeKDZsvXn#B=Q)1w8QEil23@llF{m#@-Xw^J~kx=*Foaj57JdO+Njv?%C;eqEQ>7 z@Y2A&^{bb95|=mGWZL`3=;Nym_@LO6`Ia|kMx%iM7s@u0#(IJW7`%H(I zZ>u1VGZ0l86ldLfEJ^I2u|&`0G<{e3o4(c1;}c@9(^nhk;O2D|B*oW?E-xQJpL+8A zmz2}|W%pftm_`8o_d)DG`zEA6tbcRJ&z~NMv!F9}Fm%O7J1Q7*r)PKVr;VKkbU=R; zng2$b@3Xu}zdf|1TFURi>2eHw6fUBbl5ap^gBh;ayqRv|Jg8^(Y5M6_0x>L3CaK|- zWal4A+FFN1W5R`O%u-^wtWeakcM?@E5of1XjE19M)^W2IXrkS| z^Qcx(H&-IRhOW$~)LZchaxZNMkN6{?x*oQskGB1fu{GYqrEwO(#y(g5euUmu~3{EmLf9leu z9|q}D#UA=f+LW$GpUI$xi^%3!0qwA9r}G}IqxUsB>)&jY6wj51=;7Rh+&RZrY@T+Y zDB7l-Oo}Wb6+p?mzGrAnr#8+IvuKtnxYL=vsnoJiti^`b(Zw#Iw8io?bxB-IEA{&6 zG_zZ@$8|qv<#UNs*j`C8xfQ?-_Mth=Z|k328pmIdbL6Az-q2frB`9P64^?|9;Ey|1 z`CFw;wBX$>TGO$DHoS{R_p(E1qyKBtRdI}Hx@Hk8)oq|$?MybUjYd>sywLHRlrZ8} zIQ9?N&ATjE&(Cu+;j21E;kQdToTIiGBYp(qh!X6(i-fh#F)SXx&>ZiIuY~JCd34_D z``m8n3assSs$P(Hxc=!q2T@V>DBdd|iGEn`O1ow*Le}?2Qu7cinD$<7FKLLX zd`hS0R|CM@b1Su8rHdtVv$58aXAqEcl-Xo+oJ-g* zM?;@^k`cNht~`9Z==>>LuJvdxIaq$0_CF-l^qUed_46qke(jM+ur*C|-3Z8mwo1Yr zDM&7DK|-+)kumY0hZByGmm0T4?lXB&_>MyMW4|N*G+oS7SyxbN9C(TJ zq^A@9r7b-8F_BuY%jV7cH}FR59fey#+@YALCS3c>8$5IrNYo@&?wv6e zdpBGZ^NEgQg@OQlN81$j>0U?7;WHu+uWdNz$!+X^G8em^N)-;w^c1fBm4*E*($I`? z-N-lR6uW1;H{CboDXGdIfn8dbkQ&DTJo?Txe0ukpdLko@)Pq$;on)#wld^~UtyxK* z#2ey^@+|hTd_+TDEucq-4~XY!KgK z3+1)xg9;gHGhH6cb_z-Pu90A;s|QDWlc>{sH(vX`JFgtJlON%f!Iv=-7xs_2!pr98 z)!TkIK&e%YXt9Zi_p}{zVVcztFSu95Cmue|uT4_ruPoAH#|8x=X$V6-ud7Jpv2bc) zCd*5jkEfY>yK&x~Ji1;-NCTH>^NRU#{AE=GTsC%wXwB3ebiVWy_!XT@me$@zX684@ znSvsCwp)_QB{8(cMC@oaTgN_A^ue#kIg^$N2dIo}B+=g`=H+$#U_JFkpwuRJXD8h}Z*qp|*6KyOTBX}g$L#WZ;GH!k#v^G%!a z9Tg+;`s(!hV)3ljvAKlkoree7JTRKWpVz+PpE2PbRzhYcC9_S?7BC{IDN>vQ-IBkH`n%$Ta3f)){Vl zg8`8^#z4t>TO78>oVV``;3e;B^BErgM9!s}92{t6r8?61D+*S;(~ugP_(2+9xSmR$ zTkj>Y-W_D^n8`HZ{sFRRMJ(F0tcy)H-@qUHk$|r*P~uE3jif0a-;rnYdo-VmLBDSL z3iq!L5Z3SCBed?l!Y$7!KznZ7{bQkdec(u_PBwBPsrm` z#f+DNk=5`**9uE)uDf4w3u~m03Ow6nQZ%NRCKC{EAOv|MHhC%o#sYx zU%mxNxQ5eF^8PSazZ!g<^|8ZNL;A~SH27T}CG=l<1Y7C230KYABph9-CmgOw7jic@ z3N1@ca!+RI!j*9|@n^RZC^bt#-*$f@+heBSJq|yyUHC9oA7fAVipr>R;vg~A^T#Q1 z75IW!ulcc4%ztk+g7Lc!pfOLU!QZMPcGz89s_5m6u$v@VGx#31pbu!Q&{;S!B1O0$ zG8g+y*A~9)+CkGM{YP`E_ldnXahy?}4dlH8tY35+A9Q*INB&6(^Oc71jDuC$@nsUMzB7?#tZn5gf_%m76HQooeKx-FmBUNXEc!rD z3;OnnSS4ow!e-l|CnfJm>9t0mdm^B^l% zWKujt@vqlo$&thtbOqi*%^H)ak);cry{!;s20SHd$5m<3eKDUTmC_i0@op#&;A6o| zB7ZlYe>DC!?=3&TzbKx;SDewI?IiS|0z{r$x)flPJI`;+<~ z(6{Y}xC7F8NV1%#3rzOWk8yd(qg9mz`|qruQa+m~^nXTsRS$|L$6G5Ki7e)y3@L z_j}dhVP_1vVqt`x?l#h%J1c3-aaT_9V7h3vQXDc#%%YD(&ioI_c&e09fS*2n2NC*8 zd}W{wPVsp|X6yc?{#Sx|ub=(2`eQFmonu$4^8QHq&3vP})%!L0+vDx2pSX{qdw2$& z+%=llHVol=tsW9_E`XbFs>n)g%Ed=j1^8vuZd&uVh>p>mO7H6uPO-m`y1#3H6HzI& zn^nSG({JQa-intDA{gZ(i3*>~(KvdJd$lyGUMqJjn>@dSTLcPR>|mM5$aywuR_!GB z#N3wyOe7AJ_$hjyCV>{@Ch>QFdhxTJHTZS2H{gd8_u<9bTEY$H+Vog$7exR3#V(jC z$v@wZargo!_?6aHrSHV&V%keY0-x`fXW`_uFj*Y9awJNT+9IGyp_$bX=h~Kw`XX`-Aa4V^* zswB&3H;S{<1{|;m$f2?h0a`Ue8E)TXAA4if+a;aS1FP1jv}sY>me*3vPtOjd5*9uvywYAD}#;x z_L|WfjD;~n4V<;q9HLO|D(rv1kZvC^<$UgEKybe)+kIG2uX($SdArS(KW<^h&v;PB zXI!3!GgA{_!TKB+DW1pIigyO@H_B3<_c640`dSF9&gL?V4?@I0DVjIGlAa%b30!MN z(4hKUdIc9_<*e77b*}_p*acYK#(-YmXG@D`Nn@3XeRy_?DGBZU3@tiByx`k1kTkYn zL+Lty{;_jE`%0j{h7d+5wX_Uv4vJFL<1vxq!648mQ{ zfo;tvS+l3pKR;#2A=%YbjB+Ax{A1~>_!Xq9?iTkXb`vfRK8mv+<&w*KWAKTiQFv42 za`LWI9`Y3vxVjQ+Vl^#~l+UuDt$TmKD4lV5t9L&Q9==PLiF zTT%Fibs1SK=}YH%Wz!#m3jX|~I^N6Pmk%>CC7JV!h=<}iyvSk%m$5^V`?=vLcgQkcjq(s`nYHa_-29AY|3HL+Z^l@TtNdcSM zUBl6-BZMJ+8}Zfj=kzyt(bmG_T$t=*(OjV!*)vs%#{~{Njyp(Km1gi0Ycu(0XDj(@ znYB2}v>GN&NJ0%ut>Ev)k@UdU&oJr1fIwD~W9l|jG$TF>&QK+i=%*j>EwzlpuZ$vB#;r}xLQ*@KDQF@$$W+T z`-E&=Y@1y!4Wb58=6)7I)_GBQ8}>f;yVYx7NU>z!yk z;=dU%w{Z+#Y2l1ZLHxX335|ZG&Rd+d!Fo5}(CoIKj{M?Jn^4w-B_g?Pv7%ca+mY=EJ*@H22fIEG;w&m>;I2VJ zbavjQ%)xQo+f9vRc==~~CqbPbGogwae0YKaf8E7q7Gs373T^p_E75$*bQ!);YZAR6 z^_^bHp2mCRY`2q4p2MzgpUgFkN+S#Q<#X{NUG;i%viRe3lX&Y%&b;HoEb>BoG|W-F zC|doK#SJ!laQ%d6s&B-ibK@28D~}-RmsZcs_TNV$OrucCXFp#0{T%+8%zEA?dp!s` z%3xnXj3_YgDGBXrAzRgca`X1z6|=sYa7Fbjbg!dd-%iADjFtK5DjPm*MLKI!nSP;(RDAjr6U!&?I&=*GTAX3H@@XW$ z5Lxh#n`C%@r3CtW*LiO4cNV8;TtaOr?qtk09}vcUWxwo9^I1H+&a%KpDsf*={OqJ z)J8tWQ0|$g1i2djn4Ys{9k#Q^!6R1dS%Dz+iLilscVPoZl{&g2iG3aTcE(#Iwc5M)p0^cLsjfM zsK|yzIEb2-B?}h4zksC!XV7I2M+=v`*KyJlm(eBcTeL{a0JnA|5a)$)_=Tkp)x_t(+^uDw!PTbC!uqDB_HyMndt?H)PlQe{^=C1AMuvBFs*_ zP4=G5A;W#=;c@k5n$!IfqVqXoP%ZA7*msOR;%<_ApJo#^{FXcQcs2JnTZjkSbjjVL zw(ym=W(AY?BOj9&FyoyOA7|JFonvH#Gnb5~(VF5MLS!W1{3lL77b zv7zC2wCJaI4@it&IoVfwfF6O@+`V1p1WaLxy1|*@5Kz zD&T6W9J(U=gbNYRWgchSxYF%0thNS6BrgWRNADbRvJQPSJ1I0~C#EaNvSc$!` z!Sqj+GH*Y8ohWp8;VRitv?i42GQx@IYjZzweJv|Y+tr31l?CGg-%0$x`>8Z|_YK;l zt0HvD*o(1-1bttYL^I{9VN?DQ@}fbCn7=s<8#`mD=iV4-^pd~^#cJICJ};2dBJd_M zgibab%li#!@QZ}6XmFDpKU8>tp3zh#;5V1{-SHttAD%PsyetJP%l3*o-z0GhR@q|h zuM?s8cmwX!wdebP57J%1OX!?cH$^UohpF zPjaGrWHj*kNweXIxc}huhTHh+FHQkk7&dlEx|&c0X5oi-j+^JzDhFY z1K)jCpPxT^Kec-#)(Yp{r_%rJ!i;W!d zj2X8eoY%zXb=*NC)(lQ8m&Xd59P#Y9Ik?bOh}Zp=6#8_Cxl8gt0hM^+bjvm|Tjmmy zomLFNiKUcxs%Dqm-N}vLYjRj?w1_EU0(E8t!!GS!fSij(anw;xO6WFjg}b zT<;d(lWOL|3F18HwcI(xuR4PJb;zB&wsN}grj{o?ch7*%oN<~Mi}yXhJ9)UD;)SZJ z*I-`dIJS5yp6#Wp@S}%C@OH{VFkZwXt;$7EI4&J_HyyykUOaTa1@`uBU90WH{=NJP5``WXA2+w|QY z^u|fxrWGk%z1Ri$EL)$wjo}f#owae8 zyN~CIIhP!){d5jaJG_OVy%U6?^Vedh+?`Oebfob4Rt4eU^KJM)dmA+8{z4pN5CI21 zB;)a0PT{Gv>meAVVU-1v2q%o*d0w;C?MYUQT%OynLMS!f1- z$A;0TanoqZEfxG+?1k7-IfYIy)S=Zkm9P|1-()5sMDMdezgP%DO=v{+W;}WWiCwV`@wzIF%&j! zD!>6Ahv3NdkvP$m!~c95*+R(|;3GIghTklvi82xNh*lc8^`Vl^khn==F3!Yx6EtwF zhAhr4PJ)$(b+C(4BKb1ZL4?I2Q1UDfy^`Hbg)z1`&iO8BnKGH)P8!2^H*Mns3YGbc z6@h%={ziQGaJVqsAEt+K2BA0q<#QLEGz_@LiRgqD$;z+O{f$ z8bn)?vpqAo=mkks^Fn)F(hp;<;liNkpv=77XI_T5HDzy754j%h0aP@VE6Sg0gEp{VS6Cv3!c#6 zN%r{p%NT5M$`pHh$c}OjF%FYRm5}(~kGF@Ul!U~nQQ~j^=hBg2Bm%=$tqYD=W3@VP zTc8!Pv$YinY;0_8Y;75Vo3pLN;{X4p{Gkv!RBd7Yg>7iQ^F-{PISJo0n1r`%^@P(# zr_mVoI2R3u-Dc0wD1FXA}1gR-2;P2eaz}+(jhm}VVtms5{tHe6VsuS!f zr88_?g#!C@T9=^gzj)j85BzM3kF?r-Fp5JRt4AXR-CVXI`4;l(D`GDkh-LJ=$1(Y1 z-0h{5&)9~{tHvclU)(m~7xYz(8ksY*Rx#_$mDo`}YZ<4RM;OZo zXPA&VE13QJCkhf5R>6{rO|Ume8Y*LzP~HYDw$L)0*}YGk(NgeW`<^XfmuDq_TlQ+; zPg=wKiUBA&UIkxw{0EMK3s{+p{ft6*9;1Kk8pB?l$n5Jg5X?6$2O0i2*hKG#stPSu z-?x&HnsS;6|N8;C>bs+3ix0Ew<mbR!J37k zFxEp>u+TQ&{?4rRwr@;DtVx_M+Giq%4vrI|!+C4j%r`^K(Ba$c-cf&<)d}+%yB{wh zJH860?TvthU)}KKgCAV!?m|iBHEhgXH#YQ11GBh!GrSm+hm20Qv742%V4tBY#7&b0 zLuVHV-)jOJWNn~QqYHv^h0tO+4dzA7fr+J(;MF{U%#WU9+sFT7ng%X0C#{U|wtY)+ z%Rm%fdASRpyZHbQoqm8vo4MjKul|C~B@yU0sbfCt62yitflP06+pNB)wt{{Io7UTI za9~&-$xtmeb%`XlYRJWY%k;pz^&c7$IRd5C9AlR^IkC^56tTfgV;SYgN0~v_Xl6=N z8ngG}6jpBKH@Fcr01Dd8F!pX4)R#SnBHcll8yt;GHpJoeo44Y*zgNI|vv|mwl?Tk= z9$0eg3cAjCAw}?muM1ZqiBM$}jos}7tB38&PPz#ai!HF^b9t+lPWFOJf;-%`*U& z(a3g%CfgIX41KXWfu0!+AnWiGIPLr7 zb@wai+_u2!coQBl)<<==S;${`nje0KOG z*7*>H=a33Ec)SbB5}iRFC;_5ZBs?-Qm}vEBhsjCyr;|^XD0tEuYY2*SV-G`2owuse!Uu2~v7F1N}&|MN^{2 zpc|)euqSlC+T${1f$`kig313lCi2lBL;Y_sQaP!hAu|)A#2leR#=S@`b|2cKp@uG( zcR@<@9&k$jfcDy5V5>t+SYInCW>c26pwIuQAg?K$*`#ra>GiZ=G?(9CHYjE>qqQ2D z=6n$&W&0Xo^NpxS_?lgMQI?f3>tr6y9K~)taD{!$cCqxrDAd1d8SESF4HLATfE(eC zT=xBDm1c_AMw=hZF=t7Z-O{E)&^$FELnC&Bw@0X8X4)tYal;sC;WD` z1p779piX@x7(8r4Go%jLYj$?qJ&$D>t=?78TfPwfOPmF1Huf;x>I2G>%R!NX82hR- z*X`WDWZP|9$Fo^m_Oiv@Q&{#>DQt9$hiR58;9ssIGH7McN*i-@QSuV=F*=zssuePE zou@(C&kd^DMnJ}7ajy?3BG0Z-XcyXv-#WP91KRp{(}QL3xn(BAZUQK=$V0CqHPNsQ zVw0=`@N=nyxKeT(&ep@IX;hIE<;hQdXFxe{9b^h6UL+2tB)fWtsE4tHXEs0T|sK^ zr@^XY7r>xFiXFT5m0?DH3VI6o3*2j#)&g|#qa zR~;%UZ)E=%d}8m+Y(!fwyTgy8@1gzCH28Do9_+0V;J6?ytQ3)s{m$&h&yQxo6QA?o zZdwCNcIl&u_?i9u%(D#o{MG*VHg$W`*d}{hcNMfMDv2?Ba9A+=@iEjVA%qv1RS>=Z zCL7i;iIxAhToCl_CCI+7fK3ZTFiP_Xayvc^(9YXnA8*1=Rq$e-s0#!#rSF)Ud8qkTDO=sHj-ua|BBLvnDAnb-xU;AjR*AE){whXLuVeCJ$eZr#XGgWqIURtOc%`RO<~*y z6PV)ghz+XAW!)RB*eR`ntn}D;HZZ-3P5kH&ORpP3?A9C<*!_fEJ++seDc0(o9nxWI zsWfGC?E3P@%mEV}c2Y_Svm$i@+dmMCGTJjyvC;^TpOOusqVu3_+5j;( zlwfwKJS19-hk?;+#N9UQVRyuM_+EX(_G7V+z52dE`}rDTe(3ihVZaE??bubXX&0c8qT}bpecKvp>Hqb;7aTV zw7#|$HAy5Pr)kox=b4pkctkeqI=+LMCXivvAB{vmIwjb%N2Hn0F@sE!N;_&$c#l*^ z4}Vb&2{s4ifj{b4Gy=VFmfP7Z1h{fk~LD???!G{Dn-JXE`= zfS&sh>RPGANLY5;`-HNBi+z^xFhUcYt|&s5PBgP7`9tMOzQ{ zm%=6HTMGl~2SY)_eG`lmwZQ3DT`*+y6QuY4L25P5DDCnIHnDLrB#vDJBkym5rb-p~ zHNO+3%6>;~J}>Ng1A^?22h0)}Sl?l_+9lat&-2-z+R>nKFCGq_J_*l)zS{f$k`+iL z*a`-8ra*gB0UA3Np@MC*z+EK}{*H`*8~=2fy`EWurx`hd@CW;0Uj7NNy+Giaz@Ggt zAfA12yo&v{;~J{Zut3Z1y<%-*XTs0GD0E+SEt|wo2HSuBsB@tXy7pHB%~OtMe*7mb zSaf*`D7?YYa&Zk@+mZuJPd#jNxC=x1Bk}RvFAzKM5VmH-1J}L}rfs){Z@aZ^l5%$2 zlxvK%F)cF`GdVY+<{#SZ<@@(wUF`tWy?zGPe_`!2NzneE<3%sJM~F zN}Ud3$ApeR!u3m$rm6USI(iEmw6z~SmZ@V`_%Li#awD@zsK{JAa#E0dcPa$DQGm_E zeF*iAWBr8D0uBEJ!9bY`iW=+1>gqpWQjH%VEsGr}Eyw`9I&cl?cUPi65*N_0>lEfx z|9wI5o%4bPVPnC5iUkVmxWb(9>t^T67_y6(E@O7u*Mj+}lTfw86%PJugy)Kv;g;up zs5VoCtJ7jx_4X1$R_IwsUw0l-CKZFmo+n7g@F03XXP|XwGwdSnE889W_s`Znk!SVy zFJK3jJz_Q&6++bN9I%hhg3aSq(cS4!*~{DN+17~VXzA(-_Rw$_+kepsRFZ_SbiEIh zy7sbGmg%hjz!ElCPaZGtG{J2WPIx?H3A@A@CI4;3DC@8ZvU4&(<@Y<;U;n1!Nt^X> zj&cVyYNnwHtGbXv%NLZ^O;FhFcSx7Ji{Aa5iPT!IB6SZxFx?NV_bg*Hoa_M~f_z|8 z`*j#nRKR`i8E8X+6SRyu35#t)?X`nh`!e5ByQy>v+z#6ZkAo87gt%WWA|x8L7SzD} zvs)Ukm}m$G|wD2EuiBL%@en_+FL+-LFqVjHL>4AF&eA zv6s=vuL?-P@ix1xVjUajya#RMRM6*)8-h^g6x*b9ktKIZ*j)`%VNZH8NC;YCuJ1FZ z^WX;o|GkmjqM(dqLm!~{e+e+{zB&5qlFhEvn!)Vee*&mw6nGep2D4LY*zfl!tar8r z9!=CY5s zJHiWyfw)B)PXy>j1eId)depn0lH_m2Bf+F$~CZWLc4t8njK9nJTKdRV0Kv(?N!qE%4V1nAfTDc3}?ENI3 z@w<_XjUlL(tpl|nZ*=cf1v+6>i0sEr6qKpx2;OBsweQ=ooxRWGv-NHj?8iNy*uHDI z=M z8iU>T+3fzJAI$5a4XjM70e)671y545#4myb;tl~v{9Dl$2Tup=R4@_m`>ckSq$WY5 zn+g2;^AZ`}sIz(cGt1WO?|z#~Cbw};qqMt9+z`E}U(=qjMXjdni|jTb=q zMmzaCX1MLjWc+W+czpTGF)`bIB^%cBA3Oc>W7uk#1=~}jflN1oU~dbE4-JMx%VW^P zWeZW{B|kJcdcSSeYzzCy7(06>|15?f$C-Tn8_c)iv3RSU1WtQD07pIZ;QH-WaJt_Q zS!4F$kcCnB+rL13|DP^|hIxVh{0P|m;+3G})DZ?oNiny?Gy4q9gY2T;rtI_A*O*&z z$;^tCu8dOOPjC`FhQaT*;nP-4X6-9|CU~bF;}xlZ?&^PGZ5mtI;jhth7jgJ<8gQqfX@Gy8H2#qJf%es|z z>1)&NPJCEq`z5}V{bQcVa+B_}+e+o}ylLN{P^%T1^*LtV^~0=`j0y_9xDwubhr!aO z6u34*68kv4f>9RTaOCnw*7RK_YjHW5b@M65fA(ME zRjAq1&8k)&hxI2#aJW<(zjkwgrAevCsyPYmFttZ-6qloAo#L5O;t9xvh&35a6Bt@l z45}3hAh-E48={nqy3YBaX(N>2-VFtD9#4)ryQh>f{iJBGc2L32;f%iBw(GB%++BIh zJo`;ZN39E$HV29`!%I=8*?{2FvJ!z~Utg$u7Ur(oo7sW(Y~4sEm+O zRv}bWMkNW+FiIsM($F5>`}6(<&%Dodo$nlkTLEz(A?r`yM=R1rxtvB`C>o-w5L`J2Y+6ITW`bRLhf~VsZ&WSM#~9`97R|C_i!?y zlAb&Nie~cND8mn@Tg6?-3u_yylm{3g{FUzWz5@2nT=#}|CQ}=KiP&}ABwm_*bRK_@ zTs`AR)3SF$_%;Ui{*!*f zF!jH4!1p?gQ;J`}wyo8)cYimkxelPaZ!PLb>!C{NIJBPm6w~94_cKE3tyS;ET z-kgb8P-K7$4Aa3=ClWS5I3#^+#f8{}_Z6RG%ePfrj+`W}o!$l^8ohY(=?k3xuMhc? z&%s%Ahqfx>is>UDo-!GYE=popfFZi4ucDUZEy>MU zOq@B-YKcIR{ZS^#%5a$3jtReEnt2WQfAYt4eHYB$H61@^p2OycNqCdPEI;^d1OA?s zpky!y6c?A%Q_+)X;g!WC%_f3QJRC=tudk%y+*$v-?F(Ml`hz1@N;to;6#o2L1h>!q z!0M9Ec*y+;wwMZILS!eTEZGNDOIWjBnAAvqR;52VBKS*SBJFGY*w_v({F zA`j87{}%q`vO<1no&ejGq2MCn44PkdgX&*ruxjDK+jG;fZLTi1k80wpo?JY?FdIz+ zB5|-YJAralX0Pgld|eM8hj^%G6vc&Xz)?=hn35%vaO#vi<6*xc)i4Qq_i z-@=yKXl$l!+BS5{r(IZk!2xT80&v+$CuBMo;_cU5o}KLoooD)to_qh7-V~XMqv8iJ z?{77l?)3u6XVW2Y{AqZxSq>J7>cGI=Xjs?a1?DjhAe|@%TZ*K?Jp3bd{n<)$2Bv{Q zU@Xke=01OzHLRvrZ?ovLTo+%-D&ncGM#`&O0a;6)!-lj3dRYG#jn!xaNtHwzohAm` z-yDM%Yr{#-*X7hlR`Dds zjcO;m2PV*+I-2mhaXYxz#6y>B40X0wf_ZsIKqFU)Cfi1V^74&1!ubo<6n}&1+imb( zgEte`^iW{+D2(yR`bN$2#US%aDpYc%^KZ3l=(-~_z=zBPX2&x4zTE;GoU*|5eJwb+ zr^D}TXBxWT679)0fT^FuiArAzxs(5t2w4=-AKN3~;2jIZK5^PP!I(Zd97bhN9i_I1 z9uw2vKH6C*j^%$evBq>ATK<-UW%UL)Yf2r?w_gvU$K7F0`aak#G!MjdH^TeF$*^%# zszC8yx?4jsoqU3)K9j5Nz#w z41OyoLCusoP}S55mivy8Y5kGpZ_0O)eQ_d<%=$%^cYP-VV>T!(#Nkbu*J1v=;O6cMdwcex@g%%0E`=hzS!i<37=o_1z%FJycFeQHg%W9C_+TbT$DF~h z_ES*2AQ!GQu7ZFf7hLl36|UZyMuc24pmT>ewkORM7) z>JY>6I~_PIOA@Xm41>PAHQuZa!JVVIcd~#!$IO`d3&u@U1V%hEy z9Z$V!!}pTfxHtAThAYdmriO{cY}k=Hl+2^qo?M3ce|6~St<9E7TY>2H+0b3K4)Aai zH91p4^NyIoSa~o^d7A<|{^o$yP%&m%|HcKYbXg0j1ezL~Mq^he(cQle!Kmt6&c9=Z zrI*&=mBW!3=9Pgm#&ej8Ll?=Fe<5(p?LKh*Mi}yZ4SN0>&!)I6XDiO_WKA;$zwuX9|~Nf|ME3;S5rTeVEVjtJZxK-NbiM2)4cidbo7lj z@}l0rWu;5tp|Fva(mu%shhJl@ue!nea|3Yv#a84T6L4HK9{$ESL3!p9CRFd9VAj{a z%wqM2D3n}|OAHU=rIXIkd)k7U&FW*`+bOU^o~!Wp%H7oG-6!hiu>nq9l!sjJc(A-3 z3HJ_4L-FcIcGE31TFUWY62@`SJoalF|o20_aDv~+#}d9wBvY!p3D zlUI8RLcO=pvPx}gdM|=3FtNhAcc~aRa28d<=5ckT4{6J#N!T%Mn2O!B<#13jPWyL? zY6qO8o@_R}QS3$!rCQ9E(Z{xDE7)UIcQH$~8C)mq=2>rV<(CzwFw^%0;?9mVbXk2F z#~$-h_hkcoXsx9C76wD21`l!?)adJe8=~qKN?PMK(T<++SWIIej2VQxZDu4yeFu4G zA4~LB@Yv;zW4Qg61ctV@a2!A+*|o=moF44MwBRZ1$iBCjxcVBZB$%+HHoMs25@F7C z=Znc4Pdj~0Dn?#RLWRlE!GF|ThruD+j(8BWj8|4$P$sV$k~ zR%cR!avRz`=8QM&4cN^oMlftO6>OOyT4cNrDsFqj1?O$>g!cjiN9vJ@s>iE)`+!epXt3UPl>HScL)#|Lq$P22 z)P066KGE2Wc9qe{=emhmqiOhn2twtk7;7q;E_+V@k&j09-4-bcljrzc88uk`PzD;RuF;H2bNab#J_!xT!lje1p@m2f zN@sa7(>K)%)~@<0n9f^()irbRI{gn#TW>(kn{<#k9s~-ReW*t2arwzRxap}b`|r3b zYw$RZ-E*!GU88f+jk%5L-4C#2Mib(-2k6~3#Oz@g(#MS_U_odtU-N(}GgKVLcun~y zNIBcd&ujchN)OiIK;HnTMUrIazaB?K&R=Boxq=;0=|K#R{D?1djcX5NH6h6G;f-89A*xDu+di2so=-@D#fv|4)wt5Ndeb`KFx!1MZRu4|; z*XhI0HstZd%Vb%40jd6Wf(E)wf#z~`+~nT|ag7!9>(eEIcEK>!)0j_NQ~Zd^te-Tp zWDQJ~X@=l?3+PR?J9J7`BYdtpMI2_AlYzs<%+=3|*nR8?L{5@~4NL2pvbllGl+z1n zj&~>Sb?JfYTjXK#^DN9<+lZGhwP5cTHC(3Vl zGOmPAX-m;EeH!*$u7q-zPKZZci0%>vu~!!ACTMc3>$kY zq48!BIN}IA%^lB{**rp#K`xJ1Z8r*xyz%Y;!AZwL!9(ETGRw{9yd=D>ehtCw zK1d~%zz*Ii_%{bwDS2ylw++MfN(JJE69H&mFbl^;ZRULlNH-514CGy2nSuw8JK?~| zCMa%C!e#%R!#`X{eZ_${%ty}^jQ%TC=1nb^;ry!u;|4qMBc6r#@^WzfNC{P0bR9EZ zC16#iHy-*}h9{>tV07>uOnc-5m2;NB6}v#VUp0|>9=$FYZzL?ZcRZVxDH@P6<5$cv z~UMaPM+IV@3y?J(rChIVqe?9pW0Rj!8{6_Oz}@HOo> zD5GtKM=*269r)>`1{df2CI*AUjG?HAU=zMYe&l;}^?i>)>&>ZbO(IcG2_!$dUeUI0 zO}g1Xf=n3^;8&Ey36|oxd|NQ?dcOxtY&{WsFOmsO*BOyF@=VP6C@lFVgeK;%X?1KD zD5wuZy@e1OpSsVT!@Wox&)}{Ae|)-kJ=(rojZe($@t@Z_NO#qsCnT zia$@TEz~4K4%Q^kGj-Ww z;6DHFbhLUb%;B+3)VJm&4%Im0in40BqZ%XVN^EEL7di;$uG@{f?eF5%+A(}%VG1K| z5_HzsbHUfTJutK56El=GRYi+U4>V z0-8-QP{Ib4-5ILN0P!H=0^dbb*cHe4Y~^1=Hl3URT_;6copu2G56)t4-KZ7({MNj>a*iG}*#afGi(q(u1J{W&g`N%XqpliR#LV<8m2K_8bHhiw&y;~IXzw)wHBWcvlzjj6Gwobarv#UEp% zj$+zeA2g}yQ|BG%upSi3L#j-dQ~Fw-?iE;^5-%EATV+A?S^}P5ajWB9CrQ zr0o^jpqrvawI=B>?e1MvM{GZ}UT}uY{uG798*bs@!8Vi(4q-fcp3@HXF%Ugk0S21! zXxj7+*H8V+^u!y}(DU82bR!4FY#oM;C9`1e94)r{o zZ05z#0}j398he20s;I@~4bfbqm<415j8iEyisc1f$&vmcdr(*{NFc9Jo$A>cnlNO!e?R=psILGNK zmwd>j3vPtaQr>KQ6XTC3m18mJ%PBZlJw*AA`$*weA+Rt>qV9D%u>6Z2*1U1X7vPPu z=57!j{ERBao+VCm_rhcCNc2uqVYhP_<9LMySZo#q+QOV3DD?aR`#~l8)+&pv@a|&1ZB>F> zK8+9`xDJQfXOj8%BFVvMAAO-lUX;-igf=Dq05t&(sfH!=yGRabgkKr#{xsp zfLA4u&9WwcN_&aavM}luSxWaTG=kIKonW9YkB*hyaN5@dG`(-Ytg;NavV8&7Cwz!3 zc@AH)IF4N-Zx`XcIap7+AN$wbR za7?As-FJeSNF~`KUrgK#MCg2*i)8aZaa4JhhRQ!Kz@w^CSbnw;qVf+C`dtY6)?|R~ z&=SULjVtrY?i6!lb{m;3ag8oB6@tkd>}hw#capSLoXn}006)oWP|%jecLV$BoUGOK z@~b0snifl)o?oFpZG_W@z9coSSLlkrw!pt#%ybpiGS(Ly83)V1AU5tG_HAs(ZioWw zem5@nt)AoK)u=>z7_D-Q2T$`xI6? z*|BS<&t%0+C$fWY4A^xA`t0R@GOX|FJj_q?!m}G^pm1I6tdn2b%&vGFo9%jI2KQZx z=#1SrX&0Bhlh7m1&K~%UNx7TM^PZ-fQ+UGn{t|>PQXwe(vkaZS#nX-gw<5t=0bYYkDZvum6q^T7_wE z&f?&?Tw=MTnegkskR^XN!M*fC=$-u&_PMp9-GyryBzFO~{UpT149KKI0+P8ag0-2p zkM&5N%Q{M}gQY{!5OF6Br0P@2JiF^8ZRbsLXoeh0W^_T`98py6Qi7UHaroP(13{l} zfcU&lSmZncK6etaKIJT4QOv?BhZ0ni`-FEk>acddC$dA<2JEJBQ`rkDoYpIj!Mr*i zMr^swA3d5Q2ws-Tzq#W%Oz3Wg)}n79F&co2yVKD~`3~;QXr|dBOK2tElSpimfOj!* zwCn0?YPaAlPTI;tm)=ScEIo{Tj{nLRU5QyWTXFjFblkJ-1Lpqshn7teg6N&gVN37? zIJUBgrrbP9rEWOV@88^LNnjG~bo_x|IREaQMg>;iXAoMfEzpqL-}$@BaYj@ZhUSX0 z_ak$l=A!~e`i9~4slIqsbRHT`nSxvAZos71Je=t#jGxu*nQWOW{JU*;%ns?7U|Do4 z+RIC_~bPd9lz^6aoqy@3>4f6&UqoFqUbAZT`J5U3fY=-wDQjsjtWe z*_rffVJ!VNR}ORA-C**wSbAi(IaM8dPW`?FfT~y@T5)&juvw9~>FNk77fZ8af7Ds+ zc2(38mBlX3v$smC8xL&ziN+wx>K)3YYD=$E!ICfnhB{2cRAolQ&6{a>W=3|WR*>$- zx#Vhe8$O)$9XILxK>6v)M5Q;J{EU&J21#qt^>8j8TO`d6YhA|?k$cEn^%B!+`kB)6 z5sY5M8F&xp6hts@+;-Hw zbsAruuf|pN_wnJ>J`^+2Vx0n9*yPR2P-$)k)^$j+g>URYsWXrUTAL6Rxs%KX4;Myq zk&3{lWg7H&pP&nuRg~!?6)I7%5%w^M;9nlB!TR1Ep%Ep3I&Fm>_LO^>;=w0 zENa&R<%cThoUv0hy5v8w_^_FFJWwPhQdjxgT*C#AU(RNdtE!1c+7xEN=MT)2qf0SC zqM8dkp9*H{4Cq5)4SKcHk}98kh6W)E*=h+X_T1-*a6d*1DoX9a_pCPdpKOPVr&1yF z=mS98BB*qk2{XzEQ^l7Jr$j83pKvoGF|Se;O$^OwD)W;^om)B{7*xAK8psWH0OzyH4swB<}MtjQRH z?~=0UJSh!Q4_pS7(=9;kDVnCuLN`vkzco4ozWb&^uw^hju4pj#uJGi?jZ+YaE)0iD z&Meq1d`p))p95XJA_#m_3L23IiAL~vVk9d;eZOtRHP(O++;nhpT@b{kTSAypIH)FH zf~T%Y;JdyUlAHC=-b@v_f_Ti`m_QBHu2Jc@S9FWpe~fT@3^_kmO_c7J5!sLB;C zRVbYO0(!_9Sle3!ie~4S?$}Dk^IRh{%VrX;_L9Z7Ib$%%;VQMtYNGpHZD2@mlzK|M zq;LJy00cMT)YHqL7oG*rzw6=daq@WRxEK!hE+zN1rjm-?E6J<6qY!z<04AT0hN-+; z5G{8ZEEk@FIrDn}9mFy4qztZ4(!|=53FuEXP-NsMH9kS8;+tSPP@<0a$El-7H)mjg0XFq{4K^)1Z$%o2@>4p=;FXU(iW>ovlVy19=`}^+MEGhqn{{aT1lh# z*U)_zHIesQ93QOx0t+8b!T0+m@Y5V&T(Ec>y`i{(Hoc!rALPiwHe)XkYvQ`oj+o;U zDFqZ27s2_p=V8VK1A3X$dV1RM*bToPVwOM;n**ER!AvU@n5mAmo6TGz z%e%9r1TXZT!Ok&7l&M^gy*;Iv>MO*)e7=%0p54qp87G0Y$pLgZ?2Ge8BXGQBHNt^^ zFtPnK<(n7d^zaINV3&)2mjuArnPSSH8@SY|frRflK-v$#qyuX&fJw$~8uf1{Gqd9z zU6(e2I=2N9xgYY-&GE`H8|?8^RTWWxei9}(#-rGzLZIr0u`FGeT^Upl+wYp-Mu_A* zX}|FJ%?Jz(w1@SsfAPnd0;=4Iqxt2LBrH^3@aD3IS;7A6V55o{n3st5YPD2+K_ro0 zR?N>ailQsG-(prw+afr$^dePnxyU3+=ThZEk+pT&#%`3H%NjnogA;k) zP*=7NeRCG$g&KQYV>}ba?lsW6o~blixsU&HILmCjnzs3W@f*nVsdoucKF@sMc+%!F zPj=|Td{%q%5AtBQFTeHRRza%370<*6V6Iy-mQSg|Xt|dtH!_C0>s4rmkQJ7aDeMg5t}bShVFKOi*s5UTJCIJ<9?v_I+0zeFULN8S&hpo+UfLu zML1Nw9~>G_@Dqokc~<_9c&-^-=85!QII3WT?3RmIzyB--<*&pSxr@k=O~)9Q*KYhd z-(28GT8V1lhQc9guw@-`zj$YUM5L?RfZ8asf$ z(*T2bx9OFM|2Q4pX0oQ{DOKoyO1(bk(V0+>@rEL-Z>K8T@-`U#M6z(#t4*ks!tI@7 zGgy_sR;=a`?#zB(ju*z&pvUL)Fck3>%5SE@o$>(KsXZ4|sz#{aLVenoHjb*7%90S9 z5iFQFiB+^RXU9KGr8{F6l0cyzzC`h4y7pZ^=TW~z#(y_JGpG3|tfGs=WFzjZ&BZ03 zTTyA$l1A7zla_!rWN6d{{z#j{2d~>S>xw(@lAi)IVvJuE9?_V5TPE*o8*?~3lFPGO zjMwUgadIbMqFM;vS*{=b_T(6OvLgU)4Zf$t4}9ncE(fr?b~1ZC$Bg|f7LBG0N^xl4 zGpw*(4}rPA>Axpx)b8a)d@*r8uAP$rOt2B2HdR71$1dPP({NmTBP1+Rh1$3l+<%aV zKkssxeB)y{zHkIYIqqWq&Fiq?(LWj#oko>Irl866EY#_jV?!s{uzSS09%b)XcHAF- z)|2b)TleP~>U8p0JT1abk~)qDV?LWt$;vWU_!!0C^eY6VS8+3J-ZJ{IKN4qs*T-lz zZcop&AQ4iJn5W5-WbG=3z0);_3U6a^lh+b_pBo8tzw5!4CBROv9KumYKeVy9fV;GH z@x^E)Sa_TT?^s2sYCcAj^m6Iu(;7_w#WX=tOaQqw^Drqkup2 zb1;8!8lM>oGiOq_sI#NuHTb~E2D9S|>9wHAG~GuJWSpmyLGehsselExMGbwg4?*I> zpOC#a2}|}^;=ZX(op-@fE=OJLcV)VcxN(r@>wIToIML67E0znxOFs;^- z|4aS?vuJxXGpX8vx@o1-g80MKcA*IC@y(FkYN5;KE6%6sJ1yvqf7Mjij?1{ceSyf< zy&#)s=CDt;9bkFR8tk*iJIqdn&BSMxJAJrGf)&|d&uU7Cv-*j@;k3327SCCX=dwa! zPl*5?>rRL9S5(wU+YELj3o7+pY3qI1h z-`@y6R=bnKg(g(Y;u_9Wsl$>#SFvZVFxvDj#GBta?WVwyUlw7{=Xx2jFHuHI<$Jg6rAp@M7x%FT;!eR%nqQ1Pe)q+H`LwwLD?+ceO6;FBAfU)(J!_}DAbH53szsU%C5Bbr_OFU`jtu{LT%sIgzFNVoD=}itlcY-zk zYe2Fu95Q<-+rj1Bi2o&QKvW9Wtc*kU@J2jk-bS-cyJ%D82byc|NfZoI>5VBz!0y%p zK>AT~N#Kd|E1aH^}^-n#Z zU1ckEnqxtCJvd3mN6scT9$HLhUm)|}gAzLLg%W5v*T9VcW9I7H3_;^KS-L-0i&?U= zKv2n@?OWE3r=e#eNptiyqEezjbqY0T@UuGl5LXbsw)Vieo3Mb%ysLODn-&eUo)A^L>tC?*-R2#%FyhUJtXj%Gl{>T&BWjB z68xv(Pc4(}=yguVXINc;rt9v&M9JUK3+obsA}vDllXHA%xLLpd;hIG1e^)neanv7!$)ZvbIW`zB>=J`S=y$ zWyz-sKc9g2pc5pmJPSL_GeBJ>mVTDirE?n3)5J@~#N%Ka-IO7ZLGGaxFO<*^KgPia zSAQ}`>IJhQ-j8v$Ux2?3oW`-QXSpn#Hn?MW3mkXN!h(4(F)MF1h7`%6!`^tbmOG56 zcDiG6N*Ij089^skGK3IEw#nu&Ym$+`w#} zQ-MZr(>T4W8UKZ5s+m&sSDx&hQ8d$y#xqx6f%WaJ?7Krr?6!ym_EN+nGH85_v0c*2 z?3mhyPt+%{ZL@6H?@MZN=a2|qOB$n3{4?>$I0F>&Y=e8tl5v{iY1CYtgaaAEKt)c_ zf3J7a;wNe3*_VDna`Yx5Av})`FZL(BlOoCfHCk-LtO|S@xEiHhmt(Emd7LlxmeV~P zW~WXrW8>e}v7Z%YaQB4}u!z^etCy1DX;LL!uzEc$<4kHTU5jtJi4TO$1N z-=xTzt5?FN-$C@h{hm;&6A15aWq|gso9s5TWLDaK9=pb~ z4+Whpt1TA7T01_+1N~wcYIButy?%>5RQQN3>#AWdecJ%(w;n)=hXgL<>j<94N$~w9 zI`Y&)k;eTyO^-;1&^%caHd0{=8x)Ype)=s-72m(5OEfpalj=O2s2_t?g&`OheT??( z`$H402cXB|G2<94MBU}uXjnlVJ8)Xg!ZSwK;@pESc%LVX<2ZSjsc!>puU`z+c9Y;e z>x8XZCRi4`ALa7>;n0H_P+-yonWJw>{pU2QwJ-&&KB}SVl2LF^5=YTUTl~uT2UH&2 zfr7h9sLAC>EILzzyS*MmL0JO`>t?_%yJ2(dv15XNSr-I*tOn?uUCNwC#1qmE1!2>Y z5PW8J7_S&E=i4WW(($snRP*^tHLuVI80`3uO)+$_l17K%G|3Bg2{Iym9#jV(X6 zqLrg0zMnb-U+F41qqYP@;{{+Qvy)E;hUwSEnJ~d4jM(MeX1o*ZXzI0GEaJYKW&Z`E z=Y@1wAe#x5dv?JC)l=xG<%$Kl>u}$s{WO1c5=s0kK_1IRFd-!~Vd#b~o;@fAyGsd1 z-@AnR&#sX=1s=nQKyn@5RDgK+(0Vy}L zz*~&+H;-3g#-I6*NooH^0~SQn!D%1pvcd;2=XN7V%8r5MoYSNyKa$3`M1jy$HPB5^ z0H;O$^l#rxtgvi{t2-OD)S!pw=3Hj1i>Z z#8u_k+?j`6cdF1kwE`NQIgZ}L8JzpC!C~&sW;@3Zn(8JCGVJpB2iBkA4O)AG`Rv_r zTl*5tSyKhFr$sS2RRw)+{iPd6Pti%2$1OzZ=}{t1{N z@eh2ea$ufW96j;2gX|SqPm7*Dpvn=kG`w;rs-49U`4vpQJLCH7Ln+r zo3!qF1~u)^rYVZG^fQMC3~IR3mG-CL?VNn@eIkJy?9S5n#}-qa7^I`lbFe=}8Zj;o zCb%!cyJv;bf3gJpap@z+zO1GFr!?r>`q|LdUQU067!pgHa%g{b999{SJvEN#pn1Gt7Jt?7Z9n`gl#=r4Wh;Xvp5?x!IG{ZwnXfX?;qq?&*Az;ClZ zPJVwDSJ_pf-K{2+_a4Gu`l9SB%IyiVe;~xg8uOg$O1=^4=pLYX5>>S9`A@31-InZoeG9;dYRd7pCFDlh$O8^?YWecMGv7IZSt5OsB>9WmMpJnKB1cY4OW=x>(MR zc+DPQj0=tuZO;~1bxaNmr&?gfRW15_UK}+(xCUa|5XejqXjV@Fm99JF;I<1SKX5*+ ze>(+FxJ#haf+q0mj)$P346r;O3$Ly`=66j@7dXl2lX2Tm!i?XApeghU;70^R2gQSR zK^)jP7=eS7I^?)dgf9&;fHwr-@AUvu{8rH`YMS)-K`okSe~ znYiYk3g*AeX8yg~C#ZVuD_C`~5rq<-;jaFd*c)&SNntroSXIk)fT`14f#c|h?jy|p z4tqSBv=fibK7valcfy!n9@K^P!ORFFMlMB!5i!aZ+)Xni0TF)G@xTBrcvFU%XRC2< zKqJ2U6hNN0$I;szf9So2f6TlsQ;EB@FG>1)82#@=VcPl_ypphjS>LKn-ro|YyQ1G> zQoIZ+rKZ9vB=Lc@3WmB*=fU88IEd}O03AOL!>9f!?0HVdQkSa>ETm1_Sh#DK;rN=IIzK*ipo!HxO z2mgB(#{28zY##slu$k*<`6c{1_C6KA;Y=qgb`jWJk3OAGQDbu(GQAq0<$ab89y|{f*WX~{aZR>I zbP+qvSPebbwZNTQo4H3(MU|6 zP(j+f#OT3D5xTc-DHhd?!bO8wAlGsg`i&S^ZnKuIuWRCVJvEl|*|UT;&b++BM;HfjiN!xk>wqj3Bzyga6k_R#5Bsl7IY~J%7ur z4Dv^88QuIk78O2+p{S@gKD)$qopjHqHxqZ$jARLR<2o*jOMfcDslE95P7roChv8{u zHGz86EOK(7h8Xp|q&p&=(7;QcHFet#7Kigeb#FIZaSX-In=7&9GUqp$_ZUUuv{}Dy z3)V#{0dM}f3K!l`{4XVjGVhMUTA^D|_+|-y^4kT~8As@GTN~6IX@=;>yFpLL7-MH7 zqP(OcTc9fgI^5J=(UprKWnBVsovHlZ;9@g}432A4u*GeuVr=#CKulJNKrhIrt;O;hmuUP*TPjbTm`^`BtJ zK^bD5evD|UC*zamMD#A~!X&|QSgLRW>|CbMsjNPn__PPCYQm`(2;=y>%COD3j*fID za=AOT=*MXwCwMe7suF#K=#8WA5+BhOYbKzE&?a2qwgI>I7-5Z(F1j{L(Y1c@lfXEoIk4kd97Il-3!0^dEP9)=&KFEs)44|Y zOiBf15Bi{zN;&O#{fSO(7lF@zyis4tAIctRz++D9oXK@LjhuK7-iDdz!Ts)G-&8UG zXejHR63eEFX0ZF4E@ENa03=un;RDZ8)HB-}lnN(vnt3Jk=^2NrCtIPj%nifm-hd*d zCc19Raro)932MKDVV%`kvRBQO80?UOJ%+l>Tw`Zu*P_b|)4@H{$t&O($IU;`Rm3_E zTNt@0j%`iy^ysMqdW7=|T@t;_27TJfo|z)de!KP`Z7AG;scvShhdKBEu2Dx7F6(6X zo1cg}?s%x_8YH?ZVyLSIUOM*}nq8LxD!W2!MlHND45GpKnfB@5fyT;gl;-qurhD8l zT~7!98>{K6TQ70K z!3@@UT?Q>$=SQPWR?@|S3|tW~3+G6WLfkubn7`i(eU?4L-p3i#-!BR_ymZBaO160EkD7#q#xVwtgDOihfG7R{bwfqkRasx=p~Z z76b5S#Y>o!!|5O^TCs1sBI~wgF0N}`g?qQMIN_KyNbu!Bd@zm9Qd4Im$BD6%2J3O3 zgA_Y;DR+K%Ey6=b9PlhLLXCK3T*Xs{l9fxTab7h2e(nulC_9aJp)!o8*QP;N&h4Z7 z{BLk~pAk?st_4%GTF92RAp5tDGG{I>AkjIhY~zt;G!Ezb!j?>g^}vXEY#tA*zKDPG@u5X%|FH}iwhs4%ZyK0RcU~pm%R{{hu|6;Y0gJ3g-TjU zIJyZz(<}@!t1m;brZuUF4q)8UTli<6j{6^FXBt)08}{LZ2F=olN~S0p&N+KO_q{_@ zs6;7a(j;U^NEFSp1}RNSg9a5*QJu3ZLqZ}+rbuKek@y=*$h+UK@3;4?Wi2h=j?UT7 zbKk$~`r-M_Ll9AHgue@&puvHgeQ(d8Gkaq}Se^(8->1X&DVwmpcm^)o&++YsMrf^L zgUi2VqF7izy{l$B{Vab5-Q;$Um{ZYYy-N+9-rWVhAH>ntbu~^nI0z0LPyO|Z57CO+ zkCtuv$h6Err<-#?Il=+o4{>+Om*4TQ4aZOQh%wC%)Uj}X5{&&W-Gi+d}rMEx?r8OIVyci-}*Wg6?I` zkRkDdWcDi(rP+40>}D-`Wc)Dd?@BK4R_iwVA6zG8Ip`ZYVz+&pP}={ zTToa2=29xj71W{&u2ip75w-GP86N6AjbqeTV^B*qEi)#MvVI&%8+h9jwR{8Ge2xTt zsYpm~k+@AUvhS%GCc8-7>C?2upBHrHrU=Sw>{MziFM!HUok~k4C(^URz3E$Z1cV)x zM1PDqDOei=(Vdq_aW&FUj1(ZBbtA8LR?|c7&xw}zA~IP2hMLT8q<--(Q?EBE)6HT^ z6h`+5glD6uG2_P44{l}BA8sYkb~99A;G6^5`usKB=RKQj%!*S`JPpsf%0Rh!08GV1IBEU~3_KWuEziz_O5QP;n~(#JzyH$rORC6Jb1N{a z4kowXgb|~&3B*m3D@TaSKw(%QhQ$vORbN>Ew^0)RUnJh0y##%;wXt9KHRQhf4uXgM z(C|?UPVPDa!HgoN>2WO_Zrt;O>t`%`wUgGrevVSue}J;JyogII(=ZV^$Imrmkk+Na zuOFJ?>b%@f%NEGf*7+y78;3Zo zv2%d0&;@di({ZKSMAUsOfo@Z8lI4?9h{M7Zk`o<7Vn3fFO23Uk&aetrhyMqKasGH# zUJEiSv|$@r2FG800Cs&5@TJ|jJl_eyr=({B$peYhpRHd=;D5)+gG){Xi#+K^#=B{l zaFYH#c|R##GfbQ2_0vsVCy|j)!4i)=9J9t8vvp?U*7&J-rDr1X?>|DP4r4PzqC94688WSN&KW0;1S;>mEh zsjw(S5PoK>z~f-LU_qNYR;-l9_<<2{A3TNCkJIo;Lju;&f%NkOQsl1JG}1jM2Ios9 zps9a3b`327F|}?u8>@>OpL(DJ9gGc+xM#jj49Ay5;^c#4#SbbQ#BA223#DfQkCN z3cB^z!h3-X@LCz82V&6pk__={zUPTf7I~zX<}raC=n_)R?i) z0ZW}{f$u;$>^$2F57mOm&|C>PHeVgAIR|W6uLKc zsGL3pd1l#oTE_)@MglSIawu+#jwg*tl5nm_9Olia6bvl2r6x{ZMrp)KP#4qQ(dnZeaW!8@!qs4WbYZ89S)WB8rUNi}>#K(@h%*XxW%=PeI zEOz7EXH5hJ?Hr4zTVga*eZXXqpM>Cp{SP>KH42h57s6sSai(;aJfm%@!@NrRByjG# zP1TImB4hUEW7(&Z8231YySwnoXP0F1~t)C zcpWb-@nKTpwlVp->zVU=Kh3P>durli&R@EKE z>rWP;gHIYGwEaRmFpUV$2o3#R&o`*Xa(H;lb<)#IT4^D=jr9Nc4 zMgmdDs3X5-Z$^>Y9vsm0z%#F}5Wl<#k^#jq*W)~`B`*nI{?vifYH>!WID)P7n{dl^ zN#=kVVGd}xGh_ecphJ#0Gv3#n5lMLBAGNI*@Fft*L{)wm?+{=0=s|(T7cZcHaCvuC zTR1(Y4Y!!~;QBxR;lQ?JlF#MoK6&~OCocOEePtR>xw8R%UE*-mB$RU(?8NaO^l*0j zF*Nw|0o8qT$^G^SZdQK)cd%XbA~kFBB}I+P+^U08^Ab4Hv=?|Q2I+O*w8)m@!6fqG zCn%US3`L`k7&tN?3w$$h!Gky4bL1qBZjQyS}=YR*$}tG2!46Fg971ZWVWU#W$+Cu|0o+- z^Bx+_?Tr_FFGz(A>eBG=LN0OasK>mqap--Y!Xcd=%758wYW$X?lm;uyTu&X(SZ$fa z$op-FdrBg*t5!s6oSs6@yb5rBQwZ!2Z=C3}9eZc)!rq+eRG{hnjCJOVjjT+L9Gdv+1!`B)9_ux#Q08n!(}1&j6Q z@FWbY)DNTmog%oXSWhH8M(8(AOCWAR1Ua1$L;Rjd@{{_Cd8z{|1g^RhKMSjc-rZ(I@LC?PVlzg1A4fGN(fRkk&o+)$TSaMBVGnL0IF`Uk@sgew7 z83QLhUr=ttk10BM9P_#PD>}8bV{rLx!jc?tjn_c--W21FZ~IIlrPlIR<>l}~8y)EJ zl~rVz^Afh0jAPyo-p9S>$%2MVQ%Yv$I;!QTIku3s+}yspb5HOS4g_Lzc=j zO{Er03!x4?T+OWbyp-`1*JZjyONnh=BN@*9K*CK=lf}RK=mqBzNVIuA{+1zlTSo~G zt|%vggLB}a|2&9#z=!)M4Pj{RQHaWofH75*FrMQ`C#Fh~Zz*vwAsKOAUItOH^@9~_ zzkA=L5^lWA(o~K1%F#u!P)ky)RrB}uxgbxuY4*z zGwUK67si2=(M533k;Mxw982@=F*@G5j>DELwnlz|&@*!QVBJg%|Gb`_7?4fP>OD`d z$Y~`*@2-*?uXW-1oCPq!@h``nR=}XmbNbF2U6R^&mN@;4fXc8&IFbJl(&PuI^_S1n zQy!VZA?|rRJsgKFgAuvu zwwMJFyE6r}VdE6?QP+WFIV5v=oer`nv<7edEyBJ3&ft%(dOCiqF3me&N%z`Bps?pU zre77J_6jp}Twa0C{l4M5!FLd|>@Y-Wj>AGJ4K8nPkC*zno`;#?)KdXB=gObyTwA?&WYG8>`6LUB=x&>Xjc0qS?E{y!T2ewOB!jiX2a8EoB zf(}Q(9^-ScrS>1JnwU#`dLEK3!Qv1w+CV3EaGa#WMKrITkBNPO_~3alwyu_A*|e?sfv7B}~PGXVvigsk^YLUJEbmxC?IFd{FqTnU=gghL)#pQ*w&m zD6{xWL_wc(_ZipW;haG7%mZhy^z9a-05Pq-Ppdh1A^ifV`U zZTT?!Gv_S~UWhH&hUa=sm<Fdk=z2jb>}BPhN$ z8^%rZN2hncp{~21?2s>}pPHG`_ji==lDIQ`i%};|^q9ol@Egl@1eww)QdfArI_n^M zy(ft*6{GX*GGVjsKB(*24>b=HLBOs@Qs0BeoeW6Vwji247D~^uT#n9$oUiz%kc_L3 zA->BT;iE+xsOIgZyE7b!8?TVtcU^$3ar5!s&kgv&9hvsW6PWxXs$34Yi!7a=0|jMT zxNxI4Oqn7@zxhB&7_Wo>uCCAc>*td^mFYK$!jni^{kSW=XGJI}^wbmtne2yGh6V zS0<-Pfr(eC5zaoLh%E~zqk-Ni#!l?VrE@ulO;m*7)U31o{AbGiE=q=3EAt6eepvad%!RY;7CIWOwv&{=7!i?wyav z{l{W^@^AQ;vk&X?PGM}zHFT+z!;H0h=py|Y-eejJR{tB}AD%Iee>wLm@$%{=?dHv- z&*vuo_$9>2lT&fh$y)MLP*0oTHEO&{JHBw}#)Ac~@IZAthVIWmsVzR}eJ+UP`G2Cz z^i`;+lXA?|(Lp@6paPAasbSveX}HaCPRhSb@S4U8I6k2f`%Z$9{~p4xiy{jbnh;x9jnj<)^>@giD%Z7g{Z%?HS1!eUl`*IoFG4!+9#-mx zph-;>bYC^>q& ztqCQ2><9kW@Cheabm4`e=d}J*AKK(e8Wr(Po{rJnNKZ4#pnspP0OJ)&pmcK$6ejO6Voh=UUI;g%$>-Ea$5r)t5XX9`Sn(i@!Ke-X!)tK;wZkI-oy3p3m{2)yTM z^Su@0_|x1<$^4XKL|ft@aqCjRhd~q3LRS;-mENFF4F}Uck147=Q;~^QeS@zTJ;7xX z#i$otf)+&=&}GqNT6fZB+BftPwNY~i79@rssw7~GwF#I;d%~Phb;z-ngO0FDP~`fY z{8Z}*W!Oct0n(&tW(}-~AZU{ugZI=Iz@_sZWJ75M?c-z0Oj~ZpOmSSu{A;jcV&goR zjgki$_y5e9pD%0}omo6asl*r08*adZO`MZ^dk*i^k{!nSR&R{I-#9_pMo@Hx=@1nq z`x^bSbeZDsG&2d$k<*6dbZz1j{%q%TZ2a*JW2UWPv?NYr!L2Ij-*SpPTgdUtK6~;0 zbx(9Z90_}^|3XiTHHxQ3lbTtagT@fR-T5N@c6Oy;%j)TZ&MBGnnZ`_dPE#yxwd^gH z4?f1}gO_nr;Zi)#x%y~Bi{TBb;7sct3D3CEQmw}bzO4iFn%E65EF=8rq!!Sj$xAP+je)2kzt>8l%EnZ}hX8C3zF`Lky& zXrvoLb95(J%Sg^8`V1U=)q}Ilxr}<~OB~>GTIUo7$)v3^ka74M(~}`%>L)hN zbbOu^)-Gs(Z%&aM`!zxW6D&Y$r4Ky(!f^m1Pj0uK1z&z-fK;G8`24GfzwXmOePtC1 zvY$^JIM!_H;zf8f%nZLa*c;Oxg<0_96ys?qo3cE-Dl$nA9?u#+GDi(!Y zhCc9t2;Ahw;eh>3ICR||Ts;(^*Nb3dlK{OBQdm6T4u;YU7{?exw~;cr-V{$aIxEpl zf;9Tqy$pJMnJJwrlZYL``*7|}Gn89=5!Q1|#D#C;aIWYPoSr-gN54$Q9H)FRFS`vB z`g`Ebuonzy&4S`Dy~LUJ1yTA5V&ZU(y0oW7a6@$_6}`Kj8hw2ht52ok^GRHX$8j^T z{i+Uk$1f)a^N*8FHb|DQETc>LEwIP21unNKVATC4$bR|>rhihvFRU{ScCqxIDI4g8 zKL3G>{shpKNu+&lj0Ni^L-_tZ0G?{EN7_UTPcIDygS6eacyb`FxnhBO(Z;|t`A%DV z`%zT|$&kM?9FmTtz-ZtjN?k*h++02lHqR;*+^#(?=sO%Om^)xd?|oWB_Y^H8yHzbf z`y&OXJJ6^fi|^=3c=pu=R4UHIqT}1KI{F9IU(rd+YUmTG zJWbT{0Bl&k6oWNgpt$=ul-O*CCo0_j{=yDW=$nIbj`sL?e^x?qw2ga4#EM`6TgctbrZYh~n#}0evCN?pae|++Kh#A91%$X~{$=OzYk}JW`A@Y6LJ+}e>%}v4W_8)Ll_H*{JVhk5DZ_?wt;X9*UcpGm^5 z0arXE@q&)%UqXi$^-&qW*T6d7RFdMKO<7K=rgd`$IQDrf$(*gtlsy~A4776Ro6((& z)E^6GwO}&Smvj+=H64gti3U|U*$*d2cq5kV#-%swz|gv!%eXYddbJAv$@=a5McU~+ z|Ex&jotI8_Wgj4;hQ`dEcFvFD*N&s74e37(i|Oy~SLhvi1Y?hiVRv;KOpLcQbgp_UvcsEKe}zGtwA`(HuCR8xA(Zr=V`^7I+_f2{L5ONN)6h zIGLC+b`{MyIN<O{Ua9?*Nj`;84op}|+?`&NpFnd^nag-r5dCF@PE#={G@dDJI zy$%5-x*-{t&0qsKcO!C6~l$MSK-6n%eb^E zo!cMg(RTwkkgNZKVD9cE@Z5PO>b~9vo6REW9jmhGxD$ccn_mVg=HYNy<^tZ_x{~P^ zp2aGeH?;B+CDO%&PzH(?c*-RNjksK=&YSJz^*Bo?Qo05O3XNd@Xb}{czasoTQ%s~g zxjdIGY&O1x^qaF-RCNnd&gqfL_ES_$>M?4Wngi7RnSq^~Bk+)T5|q5SPo}PNr!TMC zLGL&nguW3J^Jn-fgpce6$@Eqzpq`@zjAxYM{V{i)7w-Ic8gos8IX^V_7U`&?aGM6o zn>`kc{rs2zPpO_iv4@*yDprvYjc3HbJ{pu4{)Nn76*S(u9CSISft{Z`?YHkAuG!{= zeLb!av?dEOCZ*^f5&6y%I`L2ewLK@D>Wg{>YPRXHnr2rCs0nK>FFSR5h0^oPPO+;EUKilKw27 zbgnIe3y;FdK)^pT;pq#QVy8?Vab3RyN5tX5^8@6A<794U-32lFXF!p2>|~xei9yC@ zu+3l;*7+>LP%m-PnP^H|FYF??`#rhvUPRh-xLw1vb42CnGMEfWFwkd9UP+e|zx5Bv zvy)ukvZfz=v}%N)R%^QP`*AAeN;S3Zh!{~#38zW4AE~l%qp~~K(^;3Z$kys_WM`Z< zx^sKQ>#u*oalIH6D&=8<`ZUtgs!3E8>*-iac{23XpJOxE!_FsHU`OXaxRIrZPRikM z;Q2kcx4RamTo=QT`$O<==o5T%P9{GzoC$VKfC~O8&h;NhCQdCOpTrM=aqA|OJXwjB z>yFWC=alGelV4ExCt84vbs6v@)}hS5c|^vq2q;ZUv^#GEdwK)$Z{<65T`Po9kVWr#ETNoWE^2A^nyBIlnky=?)|#eme+CGfh}mTuYkl6r73k(^Da z!hyJ2oaFKlwO{N5{YnCbOC&%~+!1cye?wH>+#~9HcEHMf0X(pl18f&mHunK3-uE3-Q3HY;$hIq;zzBqrS`O^8E8{i}5@neKn`q+r~ z(9h6h@0*{Lhsz8>Nscm~k}M>Ft9;4swaduNKOJP@bZMfRmnIm{IgC4T0rp>&z!N{* z(Ed;X?2NT0;wnnu31i4d$vVpEo(KK+i76$ix|DL|bLUp8R;t`Pm`+=pL=MCn(vtt# zQ%$#52)foxqZ(h2p)=M>V)Sfn%;Fr}D|HAF-ycR6m3y8rc`LgSx+T zFZC|9hz?bJ45X3s6xgq%ub*58yH3k;nWgpg!)^0;{+6@&O~*nga;=ukt^Y~<7f8cf zCLH99ddZ&C|LCEQ#pGw&MRL({knGJXCYr)|a2e`Aq2)29xgtdnxmk*;=qRH+5;xPI zw_c>@?|lU8KVKuaa<ey_NhWF&(_KlOLh!)*@_LWQX|E1b)tpBLf*4B>bxh znY%oSda$yMG8*l`2U`X(W%d~6zvd7UCLKd&eU^sh;3U{Oeih%yT#WxrrG?6m?x&jJ z6?L~_3r2dXV$azc&S@D%q`!=iyPHOcepG6^A!d>I?t2c=0@b({l9eR z*HE%AM@alzZ=$&O4K%oL57T$1fSeYW80dNlosU}}K>ReAo#1-Z>kF_a_Y?f$IzS{f zQsH?b=Z@1@ik4czg1Wow1)g)|s4ky5jOdaVbCaLUr1TwQjz5oP@@9K5={SS&%3v4` z>kZ5s z#en?D%$RrK%wzQmrdH=Y(-oQsGLE~!c=8Cm_`|W6)2u-D#TANL@f0h1_{atsL9`*C zr&5?|^39`-H*29CB+DQ;&rrk;B?(aV=jP10W^}z#sdt&5+E*Dy?iYE~u^eJsN5II< zopbHPxDtJ4amQb1)0aaBRKPb} z-_ESBJ@`9q0#h+ik8Tc_+mTdfGEqomq^?yqS67>j>T-jWB6_J?@@69ez2V zCO3_eNz0uPJf~NIi)Xsyn5BV~bFYv}YkEq_Oaye0_=O2C4=@GhHDKpd4I6&%gWgkZ zP-J@^4Af&df8$Ydf!7Prkc$&!+QF^68Mgh_!O*FaB>SxvndCoS!JbS<8_3`yX>>ku%KetQ$=5vP;a{ zI757?`wA|n-3GfoX7FY-3bkZzp{zMDXxB+hIRS>!7+tDC@Ce!d|UDewTGuL=TBr~sUFvb*K-p-$Pbgc>KmY+tI1on zJc)nmRx>}@ay%>=*iWP@pTh7ONhZWtf|2}Nf-hd#(}(w6r)!EgkyWSjF?8=*rYy^h zF*aNc-TNsh9H)U_R~^77k%~C!wI(k1PC-WW5e;&?(B_R6=)yfXttgG8`YLhd-j`&B zY;e@kCV`FLXtA$r$m6x;nDZkWKU!p?PPYWp7DX|af92vAl_|`DzI}|*vGvU6X{)fJ zvlpqv`*_2@5B!uYu=963s+FrjYo;rhZY#pm6qhHuQA6J?)Zyo?Igi^#Rp{p{#D8gt zH1)`U>`glbV-HRxPV*(`rr}63)58Jn1RF8H@EW}3WkSv(2W+tEzF3ciVDo2d@EBN+9rZ%W%FGF_lrO~G{&INVV^44T_m`a3*Jq}>4B>E53%Wlqgd4LS zLDBj|jP&1$>eY#Ozw#PB-dF;ql6n}IBY~=3H1ANwPQKSzJwfqGA-(YZJR;d}h_suD zF|RM>qs3n*jO;y41Z!Gpu?49_Y{o9!oj!)S=O|!2bSiMnsgt;SRTR!!-b-Sq+Y$C{ z6@9C6C1W>lDq|Y#i)TkK!H|p^cygKOX1^Dtt3^omK6y{RTXmv)?lY_%^BNxn+~wwR z@mwcQ2s{hYgZWFhk=pYg>C~S|@V)Q^RD37|zob$ymL7tIyR5P6Z#~>!ECUfWl0^U1 zEbxiX0>zv8u>WNP-?QX2KfgEGe=@ds^?yY%*)r|+70fGbuAxaR!S3_my^kv z-Xf~}>L478RmB72!!a9zps`60l8-DQ)=P9@ruA)NxGSAB%ACc1&R@P>N`wwqR=|Zs z8;JO|5I!aL5{X>St8~kc{_9#wrY4RjKIQfF@1qNFXoV8aI5G@U3(tc6S39`7b25DT z7>%>!^ATI~@LRVLo{@INr&h5jIKBkWy;_9fs- zIA-)Yf=V3y@Pj|y#t)+vo*2RLC?C?XJdnO)+6PpC0)A6BL(U37JIvLi-%cG%J^Ltw zv$b7Ob!#MUdVdh7dP;HqNwXlnWeL9C?TQuME|}Uo9Vk@=qCaah;8j*IUG1;n^{#FF zZwJSaT7}hAkHJ=Yi-{fRh^51Ixf-bW+eTd8>A}I#e!r;D!{C>@NoSbUa2Whupq5Ypl|BS_0gDcoh;GhOcgL4HglNREy|2Z zHOEm#p1_LRL1_N(D5HEch56hV$GrB+M!AOx+*URcPw0N99`uw^?42ljYj6NK{?dh= zA~}fNAmm$CO{W;KDEi)uDNK}$2IIv{Vy;LXCKd;u65U(fM4>Z=Tv2{U%(S+^nnra@ zjpAeaseSm)UGq1>Y7U zlGI@GwJl`KI9n1wb`%SiT|)yV7kM_bVBDcFvSN%cIcy=t+&lUie{wscS9;G#GWCww zD=Y^;&GCYlzsvZyoaF_&lmGDF#XO~db%zs)$8vax>m>{K$sxZj;z%IZMb=Qi0jLQy zqq%i1v&Ti7;m5qh`M%9qxFjE6W_MDjsDAp!ze3{M`50eChhzRR4fLv>g!Px!<9{F0 zap`gf{{|YPOF6;li)r{yIt>|)EApPXiRQciyH9R^$OhNeuVRUZ^XE82Ye^DjhE%-E>JC-LoMF! z2X|s^xm|S$NbHbg&KdTid3H1Q3kn2x(;8{BRVBo6S}e25?+|1B(2Mcke}wD^kcB?r zywb~0;+peEFxJcuc_&A4mdZ8sIhKx_Id)j?c{+Ry&w$*$DSWG|Wwej&J%Vd5)3cXv zAYxA0BzoFaj6I)>Nfrlj=awQ23tNGo3-TaL(vg`bn8BShK48J4ZJ3cg8HZozLy|=n z{osu`IrxST`qka!*Z$kkUOx$4ObV$*7N=-w#VkLd#hKFBe5Ly&TmBBjs%`O&ZIu78`Dp`meKyV`^ZY&rEtC9fVs%!K23gH z#aX7nnFPx2gNk-ZG zT(9vVm^bw!`M6~@c-)ABbAHF5Lh=qyS>1>6O|ndwQU}JRG@&P#Lp*UQ0^ixBVNm2r zjD2~B|6hhVZ_42+ewWmxkB#zkRsC*rje`0rC26i zf*TLSp#17w{5$s(_&3i4hw3Ys=zWaymAT=sel2OB&932w8xjZ?B!EpBhVrEoAjiIiJhJkn`{JYM^M0w+{5{`LS2&0h z_rJiJd`}oZogz=UGj~C6AE{m0LMGifN)HX*g`g?FA=v*32o;MlIsZHcM~ZM;q7oTd zmqC|Ss?hHTlDOWsA8;3*J4{5GzxUv#Eihc=IkR@m9HzvQ z!R8JMrcqV2u9!TG_fv)|X)EE?)oExrPYHc_V^L@~o>I14B(QE)q%}tYu z%(qLliVfFUY!Zb1+eW+uCH9o#OsNJ z{r5gHXJo)b{BrsmcVzVGEkbhiCo@M1SBxzgYOJC=JWbM{#b=Kd^7K0Ij)PN!Pe;q^5KICbLDj%e4^$?(Y&fW~B)# z6x^x%BZ?@N1vo2s7ml%0C;gLGKyZEyG|SB=G?&j`) z%YSl?F+FB(dpS;f^A?vHD=|Cr&cn2_cYw}Dcs6VzN#`q*C!4l|!X+1e?%o-c*PbiX zIoU4C^VKT)iX`W!unq;wOf!_{@`bJP#Z=kJ9c1ec3XZOiKxjFPq6{v>)-(;1U&W); z*!Nhm+JMe}WJGtX>XX998_21%sD69 z?3T#=^({77Fs3f;@k*$OmBoT+XE3|9M|3ef6eXb;7c5vSdfcZ*Hui56t+E7B)OlIa zuaCb;m5-aqW~;vFzWr2kW6lvgcH}DDi0y~v;<+NTx=TX6&LeEWWG-V~xJmT2!%ArA zYE8OU&SuBbZtUxsbJ%Q0zDRtwr15X9b!by`kcdBfCfse`P4sf=Men)KfZ^huICx_n zF_0Pn|BZ1Fpc@N`VH?2P-3V9KX^X0|e4#DwQ|*ME-oi^gc0#8qw*jxm*G8xLi8hiF z;m@QL(d>O5aNuYYsEtpAYzSf7l$WtGCFeyYZGCX)uMca`J1o2qA1RXjIh|c-w;1MG z5cIzv&uU*!fp>EhMaSO+QQ|u&c7MY*w(7|iQAuTna8UmyIr-wQFzmuqdWpm7+Brkp zFMZk*FM4>-fHk)`jENR6MImwL*}GqPY@Eb;IN!GhHqPaP=G(D|S$}B-mop-r# z{#fCr2Cjqp#bMF0@??=|W4~z8=uLXf?R45`sSSQv*Cl#t@CaUWsw?r!>Hz$3_Kx)q zRxP%cq&F>sP|jvBm~(EKqgrxfMjg?RCJ+x@BBxms*h$+|+0uAB zc5PrVyXefhT3c;Ow7Ae#^xic|*eK6q_Xp&{R)zcI^g}f^aN0fgZ~0huacK%0*4ZR7 zzv0g6$%e9CEoa!ih}$C9ZHA%?Wdc$E3k%WFWtFua3%-z7$BvPEe+PutuO@;94uj8& zY7x(L1*^}@6JB)HtUdEFw5B6)waEYK5|OG}I)?sLCfhB4LSf%Zp z{}zs`x+wZIl0k00G(Z)h7W>786{RcZk(q}IS+Al(I3w{CGG-qU`a7>?qxy#+YxgqN z{e`0_w_H{@e9)`5V!JD3E;Yo0506C8kB?^$_=dr~Vp&$RZ7~MS^ASa%l<0EYY0WOrS<#9lek%H9t?&pP&nkaHQq68SHnbqpXY5BzA08m+)44I??8G3!dBeFcVK4V7gsG zG53oWYZo}X?sDn8I%VE2(Gklmm@>lzr=L?7{e5i3dHTdaV?vSW+;?QJgzjJ;P6`%H zSLhTX1D;3H9lwGXV{8|)!)4@%Z0qDdESdVWt!agM-t)F*~9lAadOas+f zlg?n_yZX6oXmTODPOO}z=KHWmpM9d=+0_#7Q9W99>?eB3k6zJ|fIx6MFo|s-eWI27 z*9$HGY!XGkUIdnlI7h*r5;VyxfD^8BM1S5Xv7|Et#~bS5Sjz-bWoIDkSTru&wihLq>_*|3A+C2kU$h`X2?fVW@cl?On#LW% z>>2sEZG#s6?()D6YcKZL(ktYdniGa7xZ>v@zftHSU>r_=!A4HMx34`;G`_fAWGmCm z5;B>+v&|m-pKXD)6;n~ZvIM;DeGy52KLQ&9L1t|u?c_Ho+OqXAJY4O_-r+{1muhHc zN$U@ktxyuBy%Y&I%1hUJ?&Lhqr!`sgt9#fee93n0JjK@bjlrPWaws5Dc+=1dH51Cw zh93aGU%X()t&G7leciA>RTh29{KwYy_5 z*m;F$?Qwftb?P$Kl1gZ{`VIGNVz4OjfvCgk3tUT8XUcrEF+?Lzq~PJpehiAK&G?lk zoI1mYPCt5CH1v8lY|NO*s0d!e$WKj7vXeyq2-iJT&;m=3#o>e3XIU$MHMUCjN9`}p zceT^!7xML1GNxzhMGqYe@a)P;JUHAy-{)nLLwD~G7n>C9K9I)s?ABsbdnXh)JO`Jk zf6)F^L3rO_lW5s<9o&*ugx{T;*;~RNtW;2$P&lRpAI#naMaPd~SKbgv1j(^8zdMl~ zU!w4^R|Tx=PsfenVR-FpD!X}dEGqv!&0-!S>Nw*8VIMxg^z1gG*1^(t-(o>ytT%k% zdIG(eYLWF(715GcYS_BxyKs@xQZ{(sq1tt|5gZknBU)E;3qOug!LH_y5VyUuM$w;w zx68-jf7icaY3+E%tRL`xvnrG=K854%-$c{-()c@NH)bb30;kp`#E8of#Pp@%NP;0% zb39piUu``~w?8I6E>cLfTcN>dCA42nBqpPMH7*Gi!U?=tW`_n}&q8o8iz9G*thiS+NifT?FV$5Q7- zz>anJ%T=E)*u4smJ9J`uhPG&|$QyJHa?Y0zA4p_OF1|NWVUEN&F_J4(S%KO7I^PWL z^WHtZE_Q+`TRzl|LCe=LmsL(NYmPOF%$zp}-MQ~r^;wA^zRFOfeqlM;n^7$~qF=>s zTy~aya^xG%YTxoF^5#V$-6&u)EtuIBO^f7am0eBny=Cdlr( z%k= zl-Iq&m{09&hm~ZVV}BnzeZdh?9_HbLRVG-sv>d4;V$AJ01<@bLOm_d;1pLLBKkiCv z)zzBv*y?Oow*FZzZGAXkbCyfF?{&GrhT?T?Y(qzIVaJV*5X&3NJOYds_;F`peLUc%Q+AV()&L!rtC!f#WQXQf-gVi99lt*ncI5**kYu)nI8r z&29emnr(sQtaf9GsJ>N|^hSo9fn9(bbH zjyRY)?FaX*nBP}-;)j4%PVfU$j1|6%Z?BIy1 z$Tst{=$tHCgJ-YR&Os-A?bSyFhE#t#4yk7qJ3Xwebo0#`WpgdS%oS^_jTZ zW-+eQPr=((FVS1#e>ghtcr3p+j@wz8i6}E=mQtSkoSTM3(V!41iPBISUy(|g3E5k+ zRb;Q{KGzW{4G|66M7t#I((-#=zyF`-pJ&{!bFS<2dB1y6XM8a>yi$iRy@^Anr+t{G zGxsvyrxfVRsh;%xWp^s`_c{2?nE-Ew8bN!z8(wqLnN57Rm0jX~oU>n`jaElAleart z1bjAOEOFBvq_C^Uxq?B$T1xQWY1htAGyH-UIs^0JF!9 z@oP?=-W&w5!s~$pwh+%uN(a^EG@y6N<}lzWzLe4tYIQ9tVZqzIFluL zAvnx10_ACI;{09>7?^Px@qW&=aJnDDc!@2+DsMEwZ|o|pdr(Ho*QOB7M+az@xhwsU zQ3-FBya%(EP_DNR98(ja@v{a9%5{)$AqPuEw-bkVe|WFPn)xE9oWXHL9V+fkMG03& zIJYcAru2~q=UrY;%zqt$np2|mw5=VP7Uqm?uFVsunF~mNej3VNup3*{eI?g?K9m1y z#P|mG-E>S(i?Vv*v`wUim=t{`h6Ufbnh&zDdY3pJ?Ro-rG7dmXl6ledcVO{|2$1=j z0(->wf=b2H`7b`5<8CrF%oex1c%6?kuJRJW3*wHW;TluCODhp?DqTTyZhwZ8kD?&> z%X?6~zZQlKy5WzB3CW&s)AYe70KGX;L!um4qM0K*NI}s7C|=e}y#Aeo(xtgjdGr+d zxOp7CIPVYl>Jy2O^JOAta|29LuYjbSFm;|H=qQzlvI&vJ0$xrRV^}zjw+_j`>PO3A z{r3Wntnh^0HQt~lZ;mFepNo_JdxfIiOUT@Vt^9v$4nl5vJvXu8G;{p(PvTT>heS5$ zg4S*gE-cHMIISEdKQwv(J63}KJTo|AF$VO1v< zrGzH1!+0W}RmxZ9#b8x7*7``ZqqV;lD zQEt&9K~D8940(1z$3qEPkZ_2+xG4_CZHe$MO$H9SxWJD;a_lx_HOcI{QevE03I9!3 zfImrHL@oarICZpw;#hrwte1WxcHwYD1aQ=CSXqMWOd*c6AYuW z3UE(q3WO<4gXn^K^!%kTe&4+w+a>7Wm>0$5czZJJd78n+Iv<8XyX~O2xdg82(>dlAVEV_x(cg z)syj3^>w+)DmVRRb(9~Yj^I<6vsi*H z2bSQqfWzqbm)61Eyo6H!|x1M;0LK;?(yiMlHg`-_# zdvM)meQbSw92L$4lD)&6GmH3uVjd~t^67%}MYuTI870oj{F}#moCTnexB9Yv29W}}Bk){+p(QdIjZ94iZW4?+$?oGG4$o%JY5Iba6s%}kN>?@;t3 zYB`?u{4zQ{`<_KedKh1J&dm=d*p ztl`Xf*1F4*ZPP7cUDoHZ)3$1{vCY>JnfwX`M*1UVL6#%>#R&J!d@pG2Y{P#9^Q7?C zN>uH-5l#5oM?ZHDQV*{jH04(Wta6<~7t1e#Yf24_wue13F;u1x{YFUAhcc!sUz-*o z8M-!927VNLC6ff6x50zvB(}#Aoi=VjzwJ`drZ;({NvVZ<8e75qr{o1_z6jlYHx_uK zu8h&7zr@+=8>4t6mr6g#qN`(b>A4rtaO-y*Y+5S?_C}Y<`=ZlK|6*TaeD?>jiy22l z3WAJOi6MD$J-g{>#Trgy8K(dJy$|1J9ETh2g70JOAUaV|h#uU;#KwOg_{(^L%Y}HD z=&eoK<)=VW(_BvLaRNHS4syx)a~R1nYfd|EKX*L*qJVWh7khNNVn62tFmN=1tp24) z)Gp>i7Wr*4e!HH*dnhdVnho=9GQfGYMzg1u+@!VvEn?MA94WwY+Ay230@n&u^Ha|U4}PQorB%( z67XrU2|9ahEzQ)krpxgRI1Cwx^{w^91H|1a%fm^%# z;TXns7x3SIQurL*gOyE#@S^33B;Yj*&Ubf{in$il{`e$%Na{P>Xe*-z4SBR!GM>sE zm`^S3j?fFU8|XHPFzQarsLI5Xv|Nu563=?boTCBg!I4C+>)?d>Q=&T1oyj_^cE%{a z)8UBsgd)b&sgyUGJCFQu5U?o4vM^Q_WTY-@;Wbi{SV%ezf7`d5U1U6jn{Teb-wz?; z`gTmvuP()ouNUGSZh0vBd@4>pdXmjvk<8w|A<53hy%x_}JVH%lDI3o0lzg? zKyPj<<9yq*?5XSBtkUEC?7CiQoHL;Z9Wbd!MmF29&)H{K@2?^IX09fdICKH)q;%t! zOLN(&FCuWol5<$9Er`>Soz1mgkLPXwc7lni*5O_ZzeI~%-Pu8jLHsFkGj3d7g43MO zW4T~u{M|c&?;cXYx8L=TYu|mCRc4Q|7f2GDXgH3&{|bp}>QVk(FGp7Qye;b$rOKv;52E|0Dskk5ENol&5$!Y{Mr_eZl-!eu zTB~LtJ$XSEI}J0H9%baq?^$H4@;s(+@2lD7X}wJ^w`Jp4UOkfb5EJBo6!0JYsdzTG z8mE5jBrBGRBJVZ6_-#ZMHc510=SiJlLp0j?k6cdR{>Ss!q^1P+6E6=Z9f;=r=bnU3 z$yY2X+<@%_4C^KbHN5I(CLVY%Lsldeq2-fbVaJclSO>XHY~t<$cI6ov#(ULeCRaqT zGqsz8l7ga{>n_#U`U;CK&})pkptIlZGKq~jT!r}p*J~`JA#khvF!6%yn(LjZBr?#3 zC_D*ex(~-<*L}P3*;;-4cJ(q`BXj~?ygY!e+A_${^E?hCX*h0wBR5*|6s`KLiDUds zn1tMZw9QHaZ;=hf&J%jchKNF>_h>qc&rV`nO1g3Hm#-+NXBS>__A-t%+Cv(L!pxL+^m7TZ7FWNuj^ua(p)4vTfqzLYn8N;LQV6xqR4?Q^V zFZrF7aAA7MBm2|N9X5Cz7=X6DC={|)jnQowAx7A^plRt_zSOrVA zjDebaCCq;ANDtUX!)v1jaCamQS9|l2m-IOtw*MO%>@#B3H`ue@Qd8KS`wr6JU-q

jLDejcr4cMfsN@7$ws1vfHhNOX5CAp)v=5sivRg@Kd>jEPp z$*?8S1-;|5oXD47yp`qq7;}v-=rrzxpZz|x;_hu!Jo26A-0BA>zkesBqaS@x7sYSY z-ErL@AHx1f(U7Pls@409-qn?A9tc#TVm3Wc-!g+tez6s)KcLuL>JmD5Y9Y~!ID&p` z3xPSiXR@J2E!a`L1Um=|Jb2IVH;pC7@E)zdi@5v@=HImic-TRM|Km>0{L$0v(8}Zc zi49cY^txpHRBbz2m~jY2`0n7g#)iYa9$~gKZwCAGngqL8bT+I0w+1gUY(syP^3h12 z6y9#PPi>L*9{FV&=n5`N-M zpVM&e%^{MV`H%}Ld(2f|(_jXb+qgwLJ+b-0h45eDF}Queht&M;LvvDVxxErsN$;`s zD5>3t{ZTxTo%7`a;wj1FtgOXspHmJi=6D$n&ayi8xAQ`@)T?(OeT`_+fc%xfyJ=aMRw zZo7gSuWRE*xl^3thXvTrdmFsa_yv(+i&&$1daS;v0;@JyfrX-mu+o}vQbdB$@uEy9 zl=Yw%k+)HT$O#M!1RUqpr--o37j#3wsQvU_grdz~V6T}TUEi4o+ok5QZ2vYkZi6bm zvSAipB%Og47!#C#s1!c5%z)RI)3E2OQV6wahg{1i(Av8Rvh{~Zup=Tt#(U{n*Kzpa zd>W!(B*52&%LKdV0+Ol4=e>O9%gEaQLzVL)$^3eIVtv~YE3-nZm2E60MQi!-+PiVy z^=**br3ue(%5ok7nt1U3RebZR4Bm(xaG~yUoZOqnI~MkXcm-T0JMwIi?NS@uFl{zi zPTk7UA8w4;9%aUF-46DLK?@t~lqKkd%CJwePO#&~S?u8iMbf?L6k5}M7M=Slha&{; zFl_^Q^etr;`*?{s+dZ_OUAQt9C;3XTZ43RmeLQ};aTyL=Nz&t1fcg!ZDTYdx`L zf)EK^bdxXhUBCeRbCkS{*Wz^-81i;?eIiA396?)j8(81yMv-ox5r45d-ue0gF?xQR z866|2LNNeVb&A5*KpP}nGmHnclyJ<;c%qm1lN*eGOVDbD{9O`F(tX0n?&h= zuuh&%Kc0rmYBInsGzCH&2Vhg&G*ULS0X63~q4YaRoYG|@!R}m_6~2*yoMwDS`fL;O zP!GlVucDdr5(~)UyJyMoMQ%i;Qv~%+n*q+>C^Fo&02@1|H2u2ez(olDa~>CdB6$Nj zHvUBnTl}>Z9@$N$J&#Oi4Ov-H1Z)q$4i!skl@kIJ1PtcU>p74Uc?k3zhqtg zOQFw-gMv=R0(^VIH+0<589$jd8|R2STIhQ5nIl2+`2GTG>{^$M(j80DuL%!0hui;% zy`Z-jf#;GZBSz%ngTq)|xDn+aX+#$T1>O&9PplL33?KQa!n(8yv$u?!@PhXaIKcim zy6*E1CrR6|jy^N+wsmeeTBTobu8BpH1s?S|PB(B&sXW?izk_j(>m~jCM&57y7n!}D zjQ4NhLnIqi!jZ6DytV{M#Y>m!u0 zqKQ)s62_0NOJOAAkKA+=*{fskaHgCrTfF`s<9BB#+BBmMZ=bYg1+(y$i6Q8u8Ncb^n;p#CL3hR~eF8U~)}W@t zM%Y+|hbvbIe#;$dY^hN(>**AU4!pU`tCpBYwnSEQslAn)%R6MDF;SMYalIzsLm$Fh z2mMj#GF4pI`W+vvc3?QAE2uh97=05OL^i>$oQ>!VVzIvXm9lB`n6O;JDe z8s78s37F|3w0dPAUX!KBx>riES|{FMCh7@%KW0l;1)0Lt()EnFlMV{2 z{*P~%x|b}Ow-_hNo8so!!a8c+Rv06drn-DL;B1m*lXhQ_+kt*5|R( zTv66bUJ3{Ac!>9V8-C$4h5h(o9$Qqk7JIDu#ZMeLLGs=hu?f)v{^ryH zJcql2C;2|--5S(l?uuBTfC^g>?fe5;{1(8ePsx{mQAB+8J#=8k4Ww854mo!$hNOZp za-Drf#-dE|jbHC^piu_i+5HeDT12C7!ACe7fwyJdf^TU3MkPEtcRE`g_yULNT;#e0 z4vWo*;w7O1*VPUqwj$*#e$pC%6(_%gq0#^7k>S;JS=}%0)Sv@My(7HZ*i)Q`kq;_A znu#BH)c1| z=L@sK8Fs8lnGmuY)59wwRPg@!8#uGvGJf!9L;TwRAMu3GWL@1EklLw%-KMDEoZ=H$ z<;o)xrM(b$d^E+{^6EI(F`Il+94BQ8=FH6Y(}a0o&of>~;EYcb$=$R+ECg2^dYj9819uyiLice}r z;jzdAIQC!@Ga;^vsS2D3PiI7c(8HVL#W)Z5=#L6|MR~l%=bU+~q>>pa%?Fr&)(Y3Y z4!}*CXSt&_@}yBb3bcQV<2b=)cfI{dNEdWUTSk(YRbnn&N?tQhw7&$3uI4~t*h;wG z^aNjYEyp&)Y0T;&X=G=q%y*T&!iWf13P+=CNMFPXkgKzyzbz3}c6^5TV;anCemHqP z{)t@4QG~WTQ{lVh8Mri@ivMUzuojkL?2^-pY>TKhp83$39cdwW@74uaD;<&A41IS)>tbh9+*D!hYK&&Zhk8Yz*A~sp*=VKFK^S z#(x`R!#KU^5oA8M;@4j~aL0C*wc9Jf|5usYn82I`{~b2G=|@&_McSu`p5g$R?-C3D z4V$sM=M)q1w1do04fjjq!y|BRsDoYrt1}FTDG*7=K@P152%YCCHYR z@)jAb;#S_aVCG*i#|C-rykp5BkQt&uKHFU6Br7kG*`ZGwzwBMf{nSWqDyz@OP^rw$ zzq1XyDo^8rc>Bohe;)YbjCaFZqb!+U`1%@!vE0@rfi$2{KN@Hl3vn;Ua5!GhDeSK_g=qFkAFg(Y4<{EYeS$BAbp_k}K`{5OVLA#ez8lyuG`7c*PwJ zoW6i1zqfHAUgc{B9$Jd9$!Q2oQ`7lm%v@Ac4yFeW%%bkg{Vs1 zz}wO%;D6tQ`rF4#Z!WL|Vc=aqs2|JkUC)xt8Wf z#JeO3JsI6VG-unQoGG@jdENxtefTZ}$J_wt#U6BO=plIRvH(;MM8Sm&Df)Gh5iK-X zPv5MOWh#aT@iE_7^yD2yl9@`Gy+2dXTg4g_qi(}R>OUcGZ!87t#r7~Ir4E;t3;O2b zBWO?H6c`WZ&@}x!_}0$X@cT^%Z2oS8mF6u%!LI-DCq$2NMeZq3u}+r`9~lRxHHIr4 zoEjL01%zbDI5{sy#l z-(8DKt?@OwN}2fkdYLNuoeQJ)Q}1*EwJFup||CWU|6^lluRPf ztbR=xx_uui%a8Cwefur^4~<(qOcL~Tb^anQGa6$M0<*@oocN2R^TgQm zC_DEPG`|dh$SLhBOS7CWdYtPS;fnmm+!jADy`H6^~t*kGCd`kUtHVKruA|GCkb5-sf-0 zXR{xiu|_ue`KlFpgb!MDjK1TFOdc~=2Mu}J5pLK!;vSYBh`?(fZ-Bc)`f%maG*l6S zaM+70_|)DWE~TuEkw{%Yyu;3r{ZBAbR`$T9Wn#GX)C72cZ8OB+CaTtJ!iV7 zzd`T4S1|9JRZ&+V9}2$Yb0<#jgQY{?(a|fl_*-o!{Bp8{TTNvoS?>xNuuS1{+p`b`$c#tOH4?;JdM1Y78FDy4 zlwx~}K4g^nik{44Pmek9`ud>Rf0wh^#ekloGM>kPB1Gl~8CP*`zt zB3xbY0uPTGvud&rad6c(TvT+A6eMUt&&{jEDfJ?k(tnPTuDi!=tQV(mQzui2ACI8# zoE)V8UWrn4<sQQ?qtmr~cNiP+@YK!2` zE?LKWbxDOOELWsmdp|;C;yF@Ln?PO?A(Xkn5m|+aQN=5p(EXq-OHP{K@#@8IRcb>P{T`mulL8ya(e zkje!#(BJ0M*#6IPtZUpJwtkW&jGUVUm8K=6WaB41Y4Hq!({CA05YeJyncDQ3u%_$oj&%HT-0EzWT?eOB@7ayFPMUSqiHh{sWbZ%AoQs z2cy?cko|w9N#fm0$V6BsX>%WHLvkghCOWjk} zf$IV4@B-{M_YrzH6H5zNnhkSUWYo|CJN05fO{5$qF{}9e5@A&HqL(;CT_r8c^l_Vx z8+PG}vBd2EP~>g`ym2>l{`5mMXYp$|wtBwhe`g&n8+;5b|JkK6Znpo3qV{2Uohb#M zXWk;G0;)KdJGbHc^2^YBy&YCc$3m;9_Rf^FI9=zS%>BcBI#2;b|0>MswWR$weJMPOr>-sjH0JDVqWC+DE&cqH-4QT_ zZeFWI=a(r`<+kbYTl@rk*is9VjlVIj1E#z!TtAn$w*xQkIfl2N)}S#wj+@2X2u_RB z(ImC!V7g3*%52^Mm)wG2rMd%=Q9cT&aW(8#KM9{Yf0MViXX!RLO{MQHq#xyr=v4hf zG*!u)egaHS!BT2lo=oe@FYpcZ+OgXcX?Dh&Rcv60I^O4h3Zxk!WRRkOYK@MeeZH4T zQ+PI9{Zs(vx)YIF{8H4d@B@WdxsV4F&%)8m?qEGIMurzI#y=$@QS#yeu4Sb)sr@w* zw`R{I|C(<%eSK<$O@lnhX~}u$fXr%=ae*T$Is-_;EgJ0&&Ee8*6yQkhD^gSWifq?) zBgf`V;EpSBXj1Y*usbISVtR&9aeoOIw5?;FJ@`jDLvtWpYK}#)O9s3)Tn-6+rqD3S z3%9gLg6d33Xlt{ln>I<&jCc-ea!(=8!wPuhwWh@uU3J{A>y2*BSp&qg6#71VBew%1 z$=cnjTyX7u1Znf}rg9-NQKy}oAUl!F*lL8M1a6@4Srst*K_Z?WD~acwKMG4cr{U*e zH;8d$4;cOoMWu!1oQ1Ip|7%P*RA2Z?p4Z8MgjXgR6pd$w_8r54jfSv(z64`ZP|lI@ zDcH|Z54Zh`!3S*X(DX<^{`?sbeW(d(SZ(1P9xULep=8+lc>_fIw&D+`ckuF7A~-Rw z15mpGuY8+}zXe2~C*_eOWQ`mIovA^UA0x5ZokJ*2(0kjVJ(ncSoIo!9%7Hb_mc%5F z0nNmxyp1{Cr0I7GdZo<3G$|48tM@AOMtB5CUaldZBL&%^>_GDL-gdGktQAXqu)(Qs z$A|z{2f4Msnc;jL@@H=;oVodc1Z!<3!vBuI=<#8wkgg=fN!y5EtO(2VQpp9A2!1cN zh5Xh7@X6yS%snVfdo@m?R8mGpUOU44eIY3L&}t$tslz*bJsM59wg`D|)_`f@pU|Hf zgJfC6QMfQ}2i?CXfr;2}W}~JAIeal2G#idUq-8zhkko=46v~(tn^V!3t%is!I0GfG zF-Ta%f%B6tG8h9yv{)G|76@axyejV4=V@U3TOFCLdV#)98bG1z70785O(eD63Qm=@ zk|m8#I0HX3zUTgUB&9x0&{2Jj6#Z@zx!1aAWb;a9V!&=jkuieGqF`w7n?OGun#2s4 z&qdeFYlzatNwh~_n>KoxKv%at{mZ`ydvAOJYeA3X{PNAzs%IiycEf?hGdJP!`YIaV z{D9iZHq-mZWNFa6AbMW)JcO?ag%i~ge1|I&iRhnwfcH)Uoe7p)$-N>b`}7mKGvEom zZ?cz;Dv3j;sSd2a5f0AXUeu=Bh?v`!@tw8^(cGY;c&_w4v@&ub6t7)Q-Hk%1oP-a2 ze>eo6fZtIg|53GLM5Us=!DDu%-W5`yl3|x0232Q&qRh& z>ZAkLcTXYbzR%_Ut6$AIi6^1c(m4VScVE+s8%Fd?MggtRzYGz5-{ga8rw}}w4Y76Yo zH~~-nRA{TZ6J1~|$zQxuiPZTT%`4vC2s)p%=)J*eDl}~=m1~uxhYFNv>Z?ky|D#0j z$JN7%`7KafqD5QJFQ6+TbK$=65_)AerV}l~AVuRV$*V4aRUW6oY;X^;ijoK4Z!G8= z^^)X8YvI1Bc#}Vy)hH`j(Wq|S0TL1$pe$D$8Y=Cnc*;C_`G*L7GIcNBe(5{Zbw$z% zZw@eClch+YwGX#w!8lKB+D`Jp`VxrGiKXY*9Qvp?nzs8G@=6Eak!F)HXx*5BUK>=x z=hdJyLJoV2bUfMwe2k@4PtNbgxLI z%LoDP0J;(>K~An9ilKP?Kasu`xLXxQIYp)ZW3(Evqt7# z?OffJ6cYNb2nCHz=ah}xf!8SL3$A&D0<-r)A2|u*#&=-T#&`T#7n*?+92WWaGD-4g zQDmo7#M@)=1l)5!5I5;DBCr<1zi1@ z+!7h&+-|I8ObnZtng=c1$&Z_f)<_Hp*`9-jutEIWVi3EUKE&ZWrm(TF1@GyPAVC?` z$T0H&HmLuAuj`2+=FD+4-dKqX#e~VCeQM~5a8;vQ>~}PBG6Xg`3=xUXQ?cP?U)X&{ zjvSV4L#kok_>g8lc^~r!?Xz7);`i%-SLqWXzCsIL$Y*lr{;j8BNv zk+PkvQCBaq@A?MYv}BmUoAneh7(lG@xG-Dy2J9-Mo zZfHm5Ra5cmd-XWsYX|dXRwTZ&vjzp}jFG8@2O+y$35-@XL!N6I^Js57ar^X_+y1&8 z@BVX#OiDH-D^R|vXMp~#BwhYStXY0QpEbW5-^kN2&X+pcvDnZ@!TrqvBfbDzMzs4UQS z&4RSF*~tBwICI0`6r(HON5njZaK%G49A&+M9Q_y%i_bd&r#DItU8sd_6B9Zec9O5s+jxP1Y5+#1T2}%uY$t`X6{oyfeQ0&akSC(cK z^;59lU=65|M*iWc_n0k%-Y9-G#;B^5WC}%tspcQpcukmwJ|tKdBQ~AaiS1XZqsjZ@ z$$x9b_@jYqaHE$bTNBcQ+w4~10H@#l_sWxq+0iVt{dzO;if<(D72n9p7cykPEt}g^ z6@Zt!-eppR#DM zVOEJAS-m8&X*1D-zAR)Mpp9kzO~;S_oJAMjOxC|uC_Ktx{}knS8!qWg9~ z{9U(=p7DPHZ&z)GCn@{rpKFuowZV93eE5_!>rX`!;(W>O$wkOpaw+aA2o~J6TdBR+ zaaw{)sKpHvZsS%=rvHu-_+Aq6n;l&w)W?<-xIIAMS3l>)KTZUbg+Fl5<6wMbat^Lu zXpL=V?LyT%wNP#ULG0b)D(L2kf>VSFSPYE9T$B!FJOLA~Z+qvTwkUYApqEn|OS|{y(88Q16j(Wh$$0$P;@bl~ za^F}R&CXIs1+Ohi` z({)8CRNApg5Y|r3&XR&d zWfPdYX$FF>fiL*Abde8_GMFyaIwHI)A0N>=%w5Y_3$LqYbN1il`P{8AM( zfB3KoKfYp*3^1Vi4#lWoC=a$D5QC*YZ5F$whRyZvd}Q8*u4ZnzkK?Je)7fV!&+x{+ zWMugB6gimc#Pk1H$MttU<7Hgm#%Eh&1r7lrhC(?xr_)Pv??^$yuEj7TvbkxVnOEb2 zI=y+klX=*B@dy4LT|Kzlx0GyFru zS{$6`%c=BEMQq<=WFj*kZQZ$^G)=h4*TYYlzLJF~>B|-7tF0MvF3`vQ5=J=AeTesH z;y!%#Zv=Kp8)H^XT*ZHm!brl4P82oyI$t&RBscau0mNPmkqo4SY+aV~29GS`E&MqH z-U`?Pqd}L@{#7ODp&)*fdFBjn#iJFB!{@I|;?DE@h0Xim#_dC}zN;G+m~8~PIhy2< z5^8iC3T+zgi01!XtIC~mvgfwO7~ms5E+DmJD>Nt#qigR>@s^FBNaaCHLT|+LZ_ZUg ze?x0f!AJ{lq+X;+_*4()wd^8FzORI*55GYsv8z$}jT(fenLunF}VIS zBFCNGn0vWX=<}cDRN%p(CSl8)CoHjSp4ib(uiMH{x=oA5#d=Whvi!?%!H0fO-a>M$}RDymv=Th5d78_B`fA6&#s4wo=yKc*8{ zlLjkqPN37X#AwVRb=ubxMT&P;vn*Wc#@gj5ZRI zZ;^?Tp4pIblM1}y@?YkW@OmsVWWialJ;!8tu15a9b5O|7XuKoE7nfR}B_cZ|K{WXz z><@Yay869bcULj-3(7`5Sc8lT`dxRH8bYuB8FE8Xjm8Z7LRZ(cW}z8!&Gr(joAr`X z;5k1POeXDs6&Gsou20H1n4U*hjIybkYYok>TTdr^BG`269z1yQIMU2oEqFG!p@jne zf#XSORN#0L$v)mp+$C3H5v9MxQ!^Lb9qxknoCq3~y_R}uO`^|b-@=_s84y+NNrJ>< z$-&;+7HeANF~3k4nkO!R1g&gjW&MM@ebk)F?~LHMrWEKfC6ID@F{-u6A`W*6DE_u# zi`S{K(*_VGE7bVbW~F#}lm=1#exIz0UIouIl;KYNbrh)LP71h*&=}%JTUcpYIX4)j z`lpbFQE~VsY!2hYZ;`q65#&~>if0aNp&4p^lqWWYnl(t!lt2+`-Iq+&Qm(^zybQ5< zXiC!8?ja?E@+5FL8}(o8rUeZf=^o`=BA3{Nhjp%EuiWQ&rtxCz^R)##?{UFv6|0E0 zbR)R0=>e^G$#6D70M^{CK(#+xBnW3hC8I;I_hBMBcm$UJu-b zs%d!;lWk8V(T6bBbRQnH55gK!2mQ~A_>b3U!1veP-1!^x@i5xJ)OjpJ(xIuS zbHfb&VCF2cTP=YZ9EgYDImaPO0mvRJYb3)LM&@N!q`JKU*3H`ivA@p1pLkU$7Ic-v z&%A<*VS4mM+j59^IzfeH=FnEFm!Orn5S$JvQ?2w0XsXi_a3^+vnwc78?6D&&c0VO= zk8R}lxry+nY;{5L2JZYvV^dL^m_6CKJPxgOFC-5nIAS4SK$Lpgz`EU0u*G>AbO|{7 z8y|*p+TVqczm5=!xz@*N1Z^iB3z|tt^lR{aeiO2j)WLS&O`@gq26^-@B@d&FLE0h_ zdDlh}h5Fxw-(%Gj(Vho|Cvsu(*1N;)PuG8LESk`Nh9UVOMB zY;XkfWTq@V9MA@8CsH9=mt$TOccE4PzHrks7#N$92Y>4dVYcp0^5$n0F>CH&3fCGS z^H+w9=;tJO`dg9ide8w*xB2kr-ZfIcVLeJW=aE{RjKlk)Fpf0GD@WcFQ}uJ)!jmSv zMmtR+U!o3APv}6G+jBH?D4MMB$R>&Ns=#A<0Q7r)WkU0%>BSeqbf(aJI^=l-;7}gN zU-5>UaIh6uDz>6zw|_`%{v;?1KEVVZup=*D1fz_)S7dS3dGbPg7ht)oq-4ky4r^cJ z=a+A5TK1v3A=*C-rCQF!8`E=9$Bj^O{$&Vusy4-eBg!Pd$M;k%@np%Pr2b>q3BL+m=ZYo?GYn&*9LwaJ;w!KaU!`C$T4vZJ@8EoA~Q?`J-#c_+zUHN+#0w7=XgvcoeFX!YAtD8c1{idxowW*A4nk?6Km@E z%bX5dzY=(E)#*+VZ93?uL#rlv!hu2|+Gu(e&b{wP$IZ%+%MD5HL53zx9FB%6p8#gZ zOZjG-pEs!0NFu!d+r_*TGGsJYh0niNFsGsMls*;DPop(w8eyO}SMZGg4~p(vlQr^V zNN{jPB}PKbKE2%tek7#L4_81{L}^M-nPI~e65+E7O_kk?+kMnfU>@>5`EQIyA^3Yu&B9d`ERuT`7hX-`{|QuXa%5=>rOu zLuldqW+MIhGCZ4JK&~Xbf;GVtXyCC^+=<#=GRs1U-rAi-4+{N&ER`pbp?naMa8$)kyJa>(v%9(sNF9a<=>jum3panAGw zI_o_b8!XBtKWiTlk@5(Vb*2jXdW6YHyFFJ`H$rl1)L@DK9N0Un4;YVIWGPkS5_QdJ zw|^`ImOMm%T#NY3+++faZ}DRU^f9N-3n(Kv2<=EQgG@_52rX>j+R}f3MT;*H^2>z_^W2o*>UjxLGk9)Wx^~!&c<~ zL?7~&mvaylPa>}WCAQ&0AXDDQD_ydMjI1?A2IVnm_}Y9E_g@Ux-e9Q$)ru*}(WxdvJQ=#feKwbB)QlWbw31B;oo&{CK_^Yyn`9qm4PBQiDN&J zV;1RfhIzJd0g`xcz#YCCM2wdIV+Jx#6O!qt@hVr6nE@ zxrfc6-Z=fVn0OAcrQhy8CWgwj^xEkiIPIl2PSh-+)r>8qMtQ(df%iScK8==nJfzzf zm(iN@x%ji76rX>&iW{X@qwcF_n(Wj}U0Wl-pf>9T~U!UCE3GBTDXeeMYMn-%%L2iiYlMMcXOGIM-7h)Bcp>;>p>#->Zf8 zyv`s7qhFJ5hc{$%$_=9iC26)WIK$C-FHAQA&q0x;i2*dJn8(S>>i@?kI#y_Y zqPsNIp?>;!_OruJ2qHo7-<&CUp>Hfo`z6DMq-o%!oX%~T<;(hGs}Ij2{ouhJKX;4HTd|TPO&Q=e>eZ3= zr#!fZ;E{C8ns?&#{w8cV{-vbzb_=o1Tt*Y8Zsio_Wrz=+NFZlyy2&z`Xj0{Rn1og~ zkS7+Nq__44tQjWoe!oS*$dW1SYdd#VVq*!kJCs7Pj07tmyB-Sc^>DW16u37}ijLXm zOsb-WkUrTbob9$eGUL1!_OG*piij3b@%e7sjtmoVk=9QTN(|s9nFDIa=CFx7P1s=F zPEbo4gP-ab;q}_(c%a9M_Q|+V`{YvVG9@#%Y)1r~)bM0Y;S{7=x~#|S<4fWul4Wqg*Y zl0PBxG;%_AvlmS>P3Kw#k$MgZy&#dk}%x(JFD4LWI$_zHAa$Kx4ojYJH8tQwG zc$Dh{w|W;?zX)S!&<*l+gTO0qd<5EWWMcOV!jDO%pAR`wztw-q2W4Y+^N$o}wczUz_4tDO*f(U!?ljOaRb#h~o>L;} z@QH{^bBNAHU(RFLRrqJE&-U$zCVn?=(6Zwb>6WBu7^tRWwave$6}qJ4d#_vc_K8C{sv}rRj2m_ePVXeXr#o+xpl8oFC#|*DJN^)v#$t!;)>b~eZInNyc9hYTPwCynMjM2uRY0~VCJ&j;~ zISy*%PLTeNyY#f9E;UxP6&?JqT0FxseTwhpY`n5mcrO14_Z@-bK${VGRe2-WqPU%q za4d>eMV2!PP2Sw?>T(;`n6BawA%`vZWCd&MWy#tuv?OoBgQ49gmP{UXkS1uHBZpt! zfU0?gba(11BJDH+mPCh$A8tVMY*8Rtuz_&uVbMkPekZ2Xzx0BN9bu62WjgGgGY({A z-;-nNwxE7Y6N?{OkwyjY#q6!*D&8ME(kIDBm11|U5 zBwV^HmOg7n!;P=Ta z5~Qe)5fekGf>bK?9p1^!O6{gbFT=>387nwNrz1qAy_eW~wUNNPJ~Y8PoRrVbWxic~ zLAZpS&|oUf&b(U8KD`tNheoB5w%P}zyG3x-^9C##DT0loN3agn(tsO+!8q~}3{*Ff zoX|*a?~h2P6XuJuRx(W5f5lL)$3uPKMR->I3^d{%K>mTfP^qCr{2NL*otFP#QAC!Q z`zS|Vtc#*2q|UOs@6NL`jB{CQ-%z-v+dv$G+!z(bC7@-ULvCz3%Y=mdWcrd+sAqvO zRd#V_Zx@bZ0?8SZ zEaLBqr1YH;++Q3+^1@G30ek~%v#ctAX)&%VIS?utx^Gr?5G5Tj|jd<_R zTTmV731`-{F!#4#ATdi+87ZyPqQIlEBxiw;5t$e-^vopLH&QFuQzO5DyzUpWu3OB$ zT=Rjw7n%w_-Uq;Veio57ctcEgE`gVh{|UUPCSvmPHS>B!y4Xdya~K-nNuoRhnH$&l zvCWN{?8wY@tj6qBu*@igP0oyGU6z%@=9)A(o^^@5_x!{SUonEHFMTf3{}%_Yi4(AA zrQiye4kZ^QQ;F5YLCzq&n2UcOEM8?ejL|zQ3$p#TdWbDg0?#Q=T;d>Oi^b$Wb{~X z-2<>%^Z{z4<=Don$)s{o4|m^e2*!G)!~DhZpf6uf3V!~ho~2JY&9ow#%dSH9=O8_m zHUfTwt#IbjLp`Gm=JkwP8s!;=$+-&r%98h-a#<-Y44g|{ADo~`l}5Pis5vPd$#E<6 zyrWh}KtuGu{d4=;MqF?bpyZn-1~|lCGdb%>&%O zb_V}ib^;&nbP-i_EpgYDb+n~u67H3kpxtY4h}Um5r`prw$?n&wO#HrfF3)^1xqK>+ zHj#3sQ)L`ZU2y~_Rgb|B!H2P6*e)EKuYiHwGNg2lAt~7Zi+TGakG}DhMcZ+8w5(t} zGVl7irlE-?ZZ9I}0fhkSaZiO<&$&%O?IX<{G71dcbf~WR52pC{P0?q09c)a$%{=|4 zfH!Ql_%(v)@?4}oA7vWLjZ(Wts|UPMdd~(@c;6LmzplkyuC~-f`mm6@YvM*ORKwL@ zKT-Q3(%9R09FK1@6xKCSC zN$~Pq@+SHizW;TWX2mMtYr}H-{E#nhdels>nFzUt`d|zg?56D9G0dD1dSuj0J+f=X zR0wm|CDKv|fBJ+wh;li6T7L{J7N(F6ZF%~{=ovF3F&AT18*qbvG`bD1CY8=%kl0}b z-Q$#~^K@(KIzg4b7C4O`q90;jfCAjEQ>L~LwdtO;?UXYaiz9ZV;trQT_?Nj%4Soh= z*YGh^JEw&1mJLN~<#)I_V;9c3Y=cwhZlj)JWi;(1cVTd2@yf4OWOTS>!=iAu!$sR193h0Z2-OT6bVKg+hkoxSI%<#GuyzZ?NJP!r@ zx^8X!dwemP6mwkF;#RE7X~or^iDYkM$cz+TTkfDfGlu0q0T9XCGQEnJH4cnMmhBAa#qZ=AxIM;of~8Ps96; zQM<;8NL{w$`~}WXx@-*A_SbWlBZ|nO<0f4G&fT^~`Knkw?Fgyzl!HqZZ8rC=tC5Lr z%SgV&NQix2%{i@~gr@fuPG2-&O zt3)?D=M$$nEw*E9e3#*^qtnWJ$=0hQc|B$rAGk+=_ZUp!OV8TzYtH%c zE(@<<_RI6w@@@&52EC!qan1DLzjjKrGjaCdRy<%w@Zqa=X4YU2bKO&e+$oDGExbSZzE%)>R6FBG!Fasy!c&xXwZT|nHnhic9&M3*M>~63iPMd1#CPs1 zB$prJ=3QmzyU(64mXP3&98}>YCD)?I^$6TT8o4uNO8iH2vG5yG!{o+aoOd}x0%cj~ zZjJ%_`qDX&_fqtJ)Oegm=Te&w8KhEcK73JEB>J-z>8$)1PHV&~(isJ0>9?6;XX!?T9&eNjzRSE}PF;Y>U)iiN{sU)T$E zG%Z?z#4Rsjl+>4CPxU13Ym2{dpKuE0j-4adTk635=X-G2(@KsGc}<)`|1f{wK{#~-U6()s&OaA`U2F!tUSa%JIkCN1S8WRDoaX0u-)CeM!aN-w9;K6@}P)ele0 z$B-HqM;Jb6#vQ&LSS?~c=+9lf?iHM_|LT?PZ*2MNrh z6)G&eV69R&iT!*Gi<`sfz1X6Tz+P8-b@k}~6Aq+9L>Wj;w_wygsqrPb77y#;JKXiR2h3<2|j zS~y?RO8zJsLud3yYW3R?^hazUy-``5jA{p6r}+RQKj%@iFEM06c>-NGej6-VmH~PP z)5*2+83ecKg8PIfZs(>VGVJ;z(BJSL-fdBVhZ>j2Ur$Y3kg+`ui>)(-Iv2Wz7WDzf47g0)&0c0pu(%{U+kR2e*#O_J4vqR2;(!^(Ue#|c1 zKDbeo(K`Xpu&%`9o(J`}pDVD4LSVs+NJvlYBMa}WfCKxFGo_mD5Nf_s?3AqzA2;iO zjo~rozse3M378Lgb&jlk=tz(+3x@J#(V+VLBwFOgQ`@H=_&7X>E07tWuUh^?k@I8n zp!yQosBHq}#{ZZzHSIzdXgC-*`a#fKGqSYkp}3;%Kb*dXqu1hnh@)dZhI@MR7K=X! zeD7=c?y?R#?Gu>7U)s3fkCQ;!B@T_wsDXEUEuf|%TRq2&bsBLVPPYfZc!3jp-bad* zdkb9KtbcS`#C}S4b&}~x%gBfwM%3S@fe|f_rXMyolch(h=$^kubemir4r(vMrE9|} zpQA$W?Y&(RI%KyfyWJDA_qLIj*#)%6q=L)T?x(8mW~jeY8!me9gTP(typPpatoyp1 z*`=Y$yu);eJ%n&Q;V|@V-2ef{(jYUK!-l(;v09(Qg9qP{b^3cCrXrJEbW~KqRy^Z3`k^C*ZiUx;npu4&%_NW~o88s^Su5u{fg~zdO^cv{z zo^H>c#1c75Jy^PJGG4SU%mM9R?{+`nL$z&5&YeFtufHo~KCwBS#DIIR>iKa&kl(!QiX8j*OA zUfE_tGPb?q^q>DFzPZ7S*3vjKgKWnwB}Xt}v@T9NV2TSbB{7=PL&>y{%9!MQU9@$K zaHpZY19t?)5}m#Y@LS^=arnL+PE%FbK39dbT`B}Ujb7^D@s_MB39 znwWe#MOR(jO*N9NOAd2pBKh2S(SOG&_p3;mZm!RvmromV_G1%pmBdgiyR{Zga)tZy zRZp=i^c?mz%z#CcAJKO0e6(sy#*4SiOHLb3hT7aHa5yAI+-y}S;~7nN*VfY17Y-=( zRe^83Q_RUOPD2Ui8S32rjjM15f9KC)ey7P~TKHl=zWAtv$=B8S{)H>~*kQZ*fY4Q< zb{_*Ma5+x)T0}$9WGA*O#)OrU-%nJ$?TJjB15xexfD8L1`JlUUy!MA^E_{nEYM`d5oQc+&MtU;m!IYK-a9-%8&}J2J>aTdQ z@l7u%ot;U1<;Fph`6kg&-2hH+!xA!a`w-Smn78AR&+xskn!FG^W*3`NnIGme$e8JO z$$i$5K0PZ5SS==P<+8-_`WsP*eLFL5+GO#T4k;Qkdn$9OMU_0U+eb7tHpA7YX;ABO z081`Nqg+!Mow@rUSG;K_T_V_SFORcfZYZ20UQ;H6?$|Mq<@pbMY7JTaS`AiRc?>J! zLfD_PM=~@1blZw{%aF6ocxs%!P2gx*VrgEk%^n93J^EfixUd@f>5Mh?-?sqV2b4%) z>pptI&yb4u-V;gd8IruEVO&P{ROk+sV-IQUWh)kBa1V7>gEh!Q?6_jl*0>9d*TT^x zEoU0Z519;(rv(m%mkb`AuZ15y&J&L%XUHbEK@!%pljN^FN4)YRAmQk5+s;GhxX3sE zxM{62oav@bH2#JYncPyzNK};)j?O1fujZ2CjisO)ng{Dm8_7Mb(`3r_95TvzIV4t( zWgW(cvd`Y_VV$QXQXjHQ)F*gYlJ{1N-@r=JB6K>eSLl$JU(3nOpwA^wW-W%BhLU2< zS^b=rRG)24i-FDClgp=gZOLGIXQYy!7e<1g8zEut1DtY~3>v1+BUyJR(uK|0RONIg zxxcd>?l)>eLGxVh=66YEv!n1{JW+!^ZghmT*m8}vX#7s=qMfPY)Q}RZF{{Lfh7VyP z72n%#`p=qdkA1_@r5mWk@uiI7uDvjudk+s1jp&U_4BaT?C5z^15QX;(MCW{^ap(aN z(<{(pZqJR?S<7`EvK24f?ZzzHHlLaGNt2Ptcc($8weaCf zRm`buVeX84N2D)HaYw?-NoBv_;Efyyu09UTwt|{&3GTD9Q@c7j=E$re`($f{F=`JYNS3K7k@36={?8U+z^;{89rTneNRo$!w;#BHm~cL2W~{JI$newE zr;(F`GSL3p2DK_8fS+E;v~7*Xl&?rMqYlG{Wm4eOyF_$r<0yJXD^MI~;fOHkEEM z)+$vES1gTPt5B+cS`HI(zKS*t@1XIKlfXYr6Ta1H63gH&tZsRaOAZFp)0}BKiGGk}PwXK*t2Wrgqt~E zzbqTNt^EPbat$PU{Y7|sK{e|17{gb&R-!2pK(^Tq!68*oF6IcU=A7m*5?wM3HJ0urK3&JfMw2JQA-`afXpx4=qXi$! zL2YrZ`6V2%mf}+uX5s#)1Jt^~9mgE|N{_Ra%yjoq9JoiY*UyU1Uw4DZG| zw3521NwcdOy)4JJ2aoY>7Ep`+BdEn2j zFO|jnrgyl%i&k-W?|R_wNfLBZxEXuDy&cwvr9j7RQ`CGDh#9;L2FTjbbDx9pMYTFw zRchg?ySeoHhZxd-H3f?5YeaLy{K+(@Wl%gf1oph^f%^+J@T8i(Y<)BJ~hBoLr@$P44&LgvS*y|d%jagBZ8(f#*S~eHH}iF^MM=t-f1T5 zQMrR{nX;b!Vr;~U^_|(!%s6ty=`E4V%AtMhODK2aF6Xs>933`g?9`KmdQ+8-FjGe> zTZmjQS99AY_mZQ|&&XxiOym2fP&I{_aKE;ZH2Oc}2w9m39Dx0!wEn87+kc$2+t{$xdS0XVz+6D`YC zB=b|Kz`e|8?(DawHU;MxJ0B6~S(uTu&i|OwP7Tq%Yf;cfhO!l-zQXmrP0+s8hLul` z0Ot+QVYI0}YcoZPL}7%G%f3#;ts7aYOxWJEPH44C#t@OfYrXK0#_SvkE0jDq!ypPR zA1r8|cPg#c$`Y;B6*^6`y+FN5i5-}}9W1o}aQpgCbJB;J>5t=!$#xGtNS$y&Y|$;n z{b>oJP1+yH8Y@%L7t5z4+sqAnmtCPB47A~-t}*+~?>bEVVgotTCeU9V^0@ZrEU0Kn zrQb?D!6jn|nc@{kIy_6TF5*142V5k+zkhMnO~YyG`UVncBMnO;RIt`zHfQQS12QjG z!OFNbux(BsM(Nn{H!oO0+sTzk6ne0&vjPH}>cQUm09El!BNO7DgT%x+z`C!4`>$_< zgzRP#vOsVQc^hMy;woaH@&_`#*03SWR905wF!Q`b0!MiZ``?Mn(2`sX>K)HGlb==a zVs0DY=pr)JvmFnYXHze67;J9d!XLU?&$~$1@l7xHp+#5=ecBsC){mWxb0 zy4z)F{n7@HGJB}j>|;F4jzGMzAM@(sn8JdOwnxgWnN2e;kySn(D4!p|U#nP-S)KLd z<%NZ~YCMbMbVpFnGC$Oq-z%zE=*#KslSIsF!qMke@;~1SpXd4;C(W`V?Js2c)@T2S zS6~{*KFk4k$6S(VaG9*#bb(W`2`+wl$ALuUO@}nsB{+Fr91e^Rrm!@E8cDB!g6ZXC zSU@Sb+gjk({5DjId4N5x-r~jEk3s5AN!WXKJU8vAG@q$*Qdlp#1a?3d*(_eg=nocC zGu;8(to&UiZG$x&_x&Q{?AZlVE6a#Y;ZB6VeO$sdZ93IGM>N<~g=4K5Ub1;9f2Ym`<#v6bsjf|U(6pA} z!X&AT(i<96HV>8@RAG}W71_)4t$En+1Dg$kaoHtF_9rVz^cOs1TKE6Ovby)^H9>*z z+kOW;c8q5g9M`k|Om=a9BfObYJ1sbm#3Wmh-4)xsrfK3zyDPLtR|~R=gXk>30~ltg ziAVPjfwul&l>70ED_`yj9ann6=O#kmi!WrZj2%j?`^05l`9f5sN~ncT4jF4=P7iL+ z#FeIJ@g}E9t#5bY9NFXey!0(j*Qz3PBIS;yso)cPU+lLSk1-hqWTv_StZh3_0S0JE z#aMjMdAl(o^%|XsqgTs$$uRzGE*_t5t*a<>^n{_1AJpKMQ=;ekor0zzPJ{ zk?bD7A~4J?grM0fZ1Q1OHhZ5w2_Z|wCq7Ej=xK>eWoI-@$c%+~msr9=aS24uS&b*(dZ3cK;8qZi!VmK8;umH4w7>s=$dylKl&>n0BlUlo%vXol zOAV*k%0Mgj_16f&1x%R!0C|jDJdCz{n#u%+sEQ5J7K2ez60vrOCO;p9VXr|d9Eet8 z-%pl-g=5B(Frg2<;?8fbX}vIeecVN#KkMaA2lv_9E9o*X4|Y;#-+MII?+ji3?J-O@ zw`c8!stG>k9IBjDLi^t*l2^_jOWFob(A$;YsG4D#_(R=1FkfZKI>S)(&Hqk+Upzo> zX}x7$zL$mpozXBt@gogevYK7>>Mz?vXX5sFXVP-#K9o<&sV;AJ>{oXPdH)T@?%EHB=hkvskE6*!xrIdL#$D32V=TP( zHv@&T6j~803m&R3!O95mMjVwCSbx?ye$^!Lm;iydaG{v_bRwMY z+S*3z+d_oweu>~3x{XC$(s1ff7xPm=lNgyDCF)*k><#54c7^^ncFhr?%OQD;b_L|X zUNb{9Z9ayV^(Nu7w+Te|i4knevw}^oZ^Tqf;IiMXB#-8eV{_|=vT;kiz(#*KbvUbr zEh$f!{8Pr@Q0qwYz8w~etEJd;)-o_6MaX17HG`A|zTj(~2?^>FWX<9n@?UqcXj1%Q zI(>XOb6a7MwEgmiuT_A|p9-q>If>gGFa+&CqiD#YQ?Qh`122IAZ5HOkTjeV8V{4{B*o=6% zQsE|ewhm*!=q~Dgy6p!qZB6yu1n?j zVGuq)6{k&eC$@ga;ojj0fr;cmQs)i@*)75F;z~6myKy&Xa;lGB$ZtZ={0+F&tObiY zBbk>;M};oMC!6`qJUZ#XDe;7KXSQKk9iH4&h1Hw_UU-FEU4<S287-~f?6Jrq-X^2GNKzb5}$j-s`$mCf=&OKz>?WFbP)Onv!a%ztBv6Nj0J zz3Vs9!5lyC#`Xd3Sq+qAE>kV}^R@x>D$~G!MY!mXRy5t1!*aWVL+IU>r6l&nd(6p< zCp8Nc`1e`iyx+AgeBsd(?DW;I$&w9yG_+ru%V7GsI-}b~8ytZ4+5~Y=V$(!fmp9Y6 z&YQIDOEmS9`CDSL{h{b|i4tA?*Nn)|J;=yM&%t6`*cLI_o=y3Z|owwW3Jc{v$RR%7n6 zeEd-Ri!Sq*;Qz=T!RVw2PSP=w{QXUEuwpwckx2&ChPh;uY#betcZ!Tz8HlwL(#S@Q z3!=b7dd%~JBcvqnhPdi)JY!|4#4XSBylVyCkhbUp__(!p}kHA`WyRZ_-u zoe8A6xSxius1q$xJp`dDM%1`*GV#?5K`rmG=y84y34W@KcY}+ec!Ue?Xf=VimB$6o z$!GdO$`!XxSw>A~jRN~QowQ-?bI#fGH%TebhO-U!==lCU-9Fh3Jv40jSHDK_6W7nd zkZ)18$Df!}SM%QtUXP_&z6r>E7v2-vlerFg8En(-p=R7BDz#df&yd}UlQ)_Wvm2Ro z%E?ahc1 zwyRG)qxU-d=qdFbv}}ha$QVRYoug-I|H}a6ciO@kpIz|Zm<6P!{1%4C3cl^h0&7V8 zmX2_m15SG-sD}4did(8F`>>n-oKVhn-;;;tfS=Uw?@wl@kcS_CVk#*u9s!B0B+>{) z^v}IbduKkN2XD-T&yR)+u92srO@AcGqMGtck>Zzy zSzxz;ienJ#&%%)|SE73##a- zgW=R!X@HKhUry(4Qo{urau~WNg^Uui#a=Q;L~$33xwk$e$+!1I$-509wAXzJ9d^H$ zzUyH4k+JiHXW|{F@j8+@HsbkkZ zn2+R#`jp^eF^`93rjTKlw@Ab#6K2bX6W}61E@MwfLxg)h9d+~yeKXe#b@e1Lk{wAj z-6I&iDjm?_f{E6e2_&jO24-!YPd2;kAWKR+iSwc=F6w>*ef9SU{ui^~RzvA5X^QZD z8`hOl3)LX19d(zUSSv*eKb3MDy86iXTX2)anMao@YW<^ruUXdf)t*BnmM3~&_NlWxbb6TC|boW^+_~mj8 zg&Hkvdp(VO417cby<^3ecj7ozAA#+A-49Z)Eu#;|xl=`%Rj|tFDB1q6mE65|nZ`|j zBAzHg#6&YkoVw*KSy;!wck^OS%X6yG&)$z)U(KUt{mY1U_C)4V9gA-@%&@^%i>rHm zihLUF&RrNg47;z3$)|V?X7NhOOzmF_5sO|En?cI-7(M1HG%e{_y=~-H{CUx!o+`J| ze>}a=G@LXlCQ|9Er|{6Bhcv3r6Jw3Dv06@ojdD&0!;rTyL1qZ-ITr@ZVoz8xB~!eu zzKTpAcnZ^%!|BxKS{gjJkdlKFY3lD&qQPnNiH~F&Y`=aNhQE?#Uwmp4+m_uY)vfkq zSB@;)8-1F79C3xNba}uOx+T+R3yO&D)MbQU;7yzMs6*-QT;jj$HzW0=kvkrIo;i79 z4e=c(uo_QVqv8ZLG-i#c&r^M@ejo)hew0pM*GcC-0EC6r`xt zUoK9ZXp+mUi8%lqKMZ=!hSPhdLbv)}4K1{JKsz4B6YX#}V()HDKKsdGlv=dNRO%Sr zEpZVx-!p|Hwkjl}SOasM7_wm2dAerAP8hH%1;Yw0czeo@s#9ZHd;Bx^@8Jewc9h|) zS51g`SJ(u*JJP-4_l<`dc2)NvSgYk%12NN{%;Z0C04RW<7;|||sQW$?I z+$y-`4Ni*UY-fPpT16OGn@%~sNwCJjjICI{n(c5N##m-eBb`YjFm)tHzAnsx9JMTZ z?z$#Xu26)|$`pLJxQ&i)8BfEm=m>6{D&%U?(V>Fn@PQUqe9a(U4t?B|>mRw;r0M7> z%%B}SOlkejD<%KgGh*K^eZ}CasegBeuUF_(-y6 z!6Nn3)g4LWt2A*YiMemBNL^1Tf(I;(;y z(F2HC?8N-evV?P?J0ZkrGhTZ;4cr?J5c~UGgcxN)`%yV)2|EowC$>VT@l4V0ljd~u zj(Ae#5Djm}-^R$Ip~TNr4xU&Z1oOs7>UrigW7`{yl5^DXcjyd^SZsiXZlQR0s0_~P zze%6;Y$tKqt(@E{D3KFShl7qbFf(PpsKHwVHf#2a21aC%hm9uuu80n-{j11ZWc!0b z>RK48?m$XIjCkXAhOf?<#_J#32`6Hrh{?>oPGUns5nN>xT3#HZbg}XQE;0 z9_TmO2!DT7=1w+@#^asIBvQDyx)84eGv4M9yOea=eqkvTY|3ED+DaH*A?ss*`aGOx zl8N``=P>@uVfdoE4~|{`O1=G-;ocS2nEhi2N}aVLzS807I!A&}Dh}gUwhN3b|88IF*^pJ8CiF}tte9h*Q$|FNC_-_EwH$6ctH@(IWG46ub_B*%V zUf^*|6?*;kwRn!(On=sR6DG@)q-R)h^Ypd3aa&K)p|bmU3u$lOWq}Gm_Q?)9d*4<} za9)5#dUzoAd?Ib9D zm%vZWg{GQ9*wteuwlESi?v0ljQOZ>uC!@lDzVityH}(jC(n2VVJO`|~Jgc?821wO9 zSWTU6gV@#W~FPlxVv7nQ%7!m4(d@PN2tdj^;fYj_Xn+_@25qSfHdq=YFme zxV!7=`Hgm%qSVJt*StaIcHJcgZrO~i;vLb*quJ>7PaD>TKZLO#m$84E`^eY9pWK;1 zRkSHtg_|9ZV&$DV&?t0q=H>l@!Hzb1uU?k9dEq5f%S{D)!(q5Ed_N}4`9>~IR^@L# z(&zT7JV8t4Zk)a3gormfhI`_paAm?!ylrxmHnCIMuf zL#ZnlOh0Xzhrcr9`9)npc*cDVX2&PO$W}e}n#l@?ktqe2nn?J3E|)RgGo1Ub;7$G$ zIx(t0>dCF(o227*2;Xe%${WsmgrhtXQCDR*I_wqLnC~BA)s#(Ge!B$cZ|=j@ZWD3e z!tq4GTniGrE^hYgDigjs+UsHPSOS@P@nZ2t_*>ex#Y zYC_OA*^}N1O2EyVPmu=WQTSAb(hH+|>0{?cvZN;+%Pz*^$4~`)5pKkfh@8Uz3cQ6& z!{5*f`3%&g*XXU0hq&4s^NE?;P|iKa5$=5Ug57JKiG5NBb9cfFI^=gB{_THD4ZX+X zZjC$)3Wz1*5I?%e>L=yyKc*EYKG49OyXn^sBS{b|Q(tbHFh z@ytYQ5BtOYfd`z^IEISr^l|F3;V`LRaDA*80~sO@3^;KI-Op<966vXYaB~VD@GOCM zlU64&ePej(CU^eAsW zq9%X2ISmK@8w;bi*^|G$Q|PEIH>pFswSXt>Nkl8# z4O~nuNJ;(?81!9-$HoTX`3Wy+-R>1s#_=2X;GzfW2Ib@7$JgL?ge9G@P46YLBgh^c~}Ih@=a~jtYab zX0zb_)VX5Y#0&Uyiv?8(u42NRF5xTL{iMqXXng5C?##wNbjhKku>5isaW2s!l{*75 zW=M08X_e{`XlG52g%i%!C|b=qX#A!jfdSOa>3gT6{76#_mLf@8k|ZhUMA ztM%h)(Q*d}IrNRTo0Uxsd`s!oI$`s*H>It4l3~wbR413jTdFWKW)0%I@;g zXIEY^LAU>;*kf!Ia3x5A8FnDs~aXt-( z1wE4F-34X(?5;KIl=T`g<~w-T#gMPJE6A+*!i;>0;MvQ2NRBzy;jUgEG&ov^Vh39~ z@Yz|kIrJ*=YFY)6T6^(}%MhG&JcTSeFab}vs)}B z*x9K{?EV2;_SR>62+*;GwObV-r0|WX^-?l^HcrF8gB3#O#emGrUkheW(#ZJN#n5YQ zKqfCeN%7WMQd@b3tejfHgc_LRjb+C;yFIat&Efm>aM>x0OvptkLt~^dG5ASi6Hd@o z#oIF8FsJn&Z8Dev`;!@d$B#zrn%hlt7ldM;uwI$YWbt{D4wlE(Q+x4eI-)@6_D-z> z`9opsH{C<9;7_PcyW=G|^hIFkD$Wp?LV+-9Rse*AiRkM0hu~hUJUaEUxT?Q^8cy3! zIydLzh|qOH&iXzqm~$M4mc$Ed|9E!HwdXKbr398az9Q?)%h0Nf;C=5_-1aI$;QO8j zy$dSTLoJMG?+b^YaXsW2KbPFp4bT{ZunvTIW0}eV9c8 z;2@*F&4JXVEk^65m&x~$%|zkT4md)76Y=-c;%_@bz$WYz7{}ZsU3=ET))@@gZnsAB zPBr%V?WL@VBd{Jlvf_u^f6^fba^Z9ARrZGLMRwCkNw!d62zv%g@aG@J;X%o736yV+JS&ZD==fv*7J5sz|EEX>|!i~xwsAXX) z!%jIQcAIKQW>jkPfkG`r=2a4M(KqSw?B{T~M}{q}FrZo|hd`EdApBmV#s3Vl=FggF z@PFgqlKw;L)LPC%+_|J0J&zqFGvsGL%~M?x8kKJQM8Sty)9y+yRE$8`(>h4EO$6D^ zm1OJeSGJSnv}uI|;hG1&FiT|n1;$f0rsk*#&x|}4{xye_5A(>btLprz?gnhzFT=N= z%%yI%*T^WzeUP&0Ek04S!5=oIkmvE0EUuL#TVg^PJ+t4q=&6*zb=Ts#$|WfB@L&?qiO|4brlxQW-7iR+wH+us@8g`T2$`8`0sZ0e zc9C<&Rw}&F~@}|Fsn^~;NiR6ys$^Cz5mDDGq(Gh~F$>UHm z*8szZjmLplJ6N(nfvxHdgSux$r0)50ruAq$**s|;X&JtUv&s6xc?-P<$sg}&MsF&8 z_j)l6jeE&$`E3CgyPgw;d(X&@FA2C>=_Q6NXhE+IF|9kWjZR#@i_YjP!lwL%sGfHi z@%k>T?iq@%5fb#H;F#WdWPnLr=}qkJyD~)!B5BZD2PpWSf?*{sbjZ-T_&xgt^*OMZ zE-Ze`jF;8tB{n6YJv)+L@+(Q)H?Eh9-7Z2s*(!{v7|MGGG@=?a4DNKPLU_7AvD#q+ zE5=Ddb7~2TYjdWTWb;xg^ z3GbBc-~8l+IO<)ZTuiX<+B~}6)fmxBioel)8ci2Y#M3jj(wh=lboE%IrTXDIB-?}?h}t`8swaxBWAnHfOQ2-8%}1fSeizlf`-O9T-z!Rfagtal9m836T70T$qj;b(8hw(F z;To+7E^W+blK%K1*YP196)OpiIy9frF+YILewT4#sW*RZT_bJWmrsL&`e^y29DMKU z&d+grhBmDc~NyISp#^Iv5etOm`?) z!ApUueywvjf1+517Fx^TluOD;QniFWjT~saxrDP!RjAB@Xzbpi&zpO%rZX2w(f-B^ zW+3Sy-7oP=@OtdUIR{j|wm#VvN{XZ<4^nh{dw| zwq)!cLpWC%NlX$ukbbq~2OX;M=igzlMl_7m>sx_syyD^2~r}+RP!@HZO2* z_Fr5c=EC3GT}eNUZKO}RNPO?HpR?sxb9M{oa(yv~;cdS7Z}e(%@WpdxyxRum_oV1b zRbiKqSR70LylbabiG3uuxwzxecxs0^XFW$zG_!gHZZ`F% zV>HI$ok1I7yD^s==O02R(iX1EzKRLbzC$X4)2NNmd0loURXqEBBbEid!lbA&>~v28 z=ha8>`>Rt_Cw3NQTOSuX@vbn&gxlIX63I!KSx7M`^NY^K%3J+*;+czX&= z%QoWG+=bYF%mZ)vgp;K=meQ`8@Alr_XNZPcFY~2#I$~@jM4ebgtKL`Q+~5_c*pfs& zoF6lzmOP;`sS)Bi)myoB_m7Dk%9OZE3XcU>+ay@9IGTvYT!Qsd31qzfQ+l|^i$1ww z04qil_&igN9$UDZ+}RRFpPpU7ct^aVNMnQ#7H!{Tm+}rE0Q@& z%eYo?I0VHX6!=Ja_&I1Mxfs8LGZuV3@h*?Jn|Wq*zG@y7xp{(If&+})bC@2!e1|@d zK1VwFM)BW2r$MT=lZ0;jM)haRp#%QmXgMu_l4^PsXlPQond7`-S86jS?DTCi&g^Ef7knd&?O8lTwE2QkJ>^!mul*DZAM z=(hd1&G9(drLd2Q@)wvb(}Kyy!NW|!yD?<7x+5O7MW|!wPaF{WNEh;*|x|rFs2Thl0RiEOek>z7#Zg_dnNP1=r)gXDRcr0BR6Bp+_6Jfc=hYRvS& zbVCa_?dnC*zulXt%i$Pk2WdEcXfv+-*Z|+2<-xWKcOiG#BH}Vj77C(HP`78x*^BX8 z**(@4tg`lSm>KsG7U#&a#a|A=`}+^!Ra_73+$P2LskVWH?^rhFjvY&<1PJ@B<7|+R z7~E}?n5{WZG$l+4xQSIn@4FeeAB|&|Y(2x?(7VpA^{E9PA2W8{FX6KKS)cf-=#$CT zaZshdT$DWBi~4q)CAHiec&0DITB~pnvQ`~lJiY)c7j%H4FsE>>UIL*CHBk0Q4Rgn2 zaGCc*Y23JYI41oI{@%7^<)73F-d`4WPWK~MVH}K={|T3TCD@61F(jcR1>GLMp=yx} z@RQ?ZI{UyK?txeymS0VPSHJC94}VRv`N=Gb>%i65z4=J`#`8#l8jpoME z0LRnJ+_`3C<(lKCu?FL|`1pB2!_+DAGb9d86(O{JiqLE3{*}MJg zSY0JO(CjQ0eV@6Ac*X66R~es~k5aeE_ES3`P5nMM^5tG=&eDTf*UrF!ojovBn5W16 zmknMf<*;HP3pRfm39Tj5Kz@Fc=+4{?u=AfRn1mWI?LL_(*(*3LAB_j&-Eq|7>rFbU zdKW#p*M$A{;{~L?6u34^jKC?shhAMK!Mtg@Pgea{3gQyM{d{;o4vpW6pKJG!1yh#8 z!K9_Y;!0S0_8Q08KNM{^97Bpnm4gEK!TJ87!t7q2G`14*LZBISdzt|2@;B7Bz zn1JUJw9sGMfFAEvh1phdWWCWCcqpaGwPz#)OY6zEeJi-3laip+KOEfTM~E|O%`0Cw z_A`Ggqo~6CE;^BCu}>UBrBBWlxjU?doaGsCa8Mo2>z|}t?p*P;fIOmJSqAMlote~s z(#)3XO3})enc_QMn~Bkumt<|R4O};@B(uh^pvDuD=-ba>)M5HxI@x75@vxX7>UpXK zD`&Qm#tR~H-tizgc*LCjJXwu>@OLWYJUYqUV=Ca|q%`QuHU#O8o8s|l-t9{&vvLu#xpFfPZ_$O0WQ`!&U)sac%NXMZAkTt=_N7ts&PH!=1dIe5_JC$l!F zkjM@9V7L7BBqmN*;coU+NU5|1tuYCl8{-3EfjN-WM8SU6TvR)fO`lJ@Kvr(hq#s)! zl8JpvYI+_jO}n>Z7WWSY=5 zdc{OeaJ1KZg+alTOl)U_-20$B>^U)y#KuR^gFT8vWE|YtW|yHgp&>nf4i&5}TvNq$g`Dc$J@oi$6z!cg;KUtZob0I_DfY zr1Xj|a!jZ0E+0sswLdK0F_B(Py}^9)w&lAI&c^g;Np^&$Jv(ii78^ZIo-J}RL1`Ze zRNSOM6(erb%V)JfYo!Ml^+Szb|Ex*L<{w;!y(%89jKYr>W?)n7a>`YA;K{wZ81Z>L z-{7suC%l`?+kXFurR7d^|BW27{xItN$#CeRgXZQ`4K zuc?7y7{B_=9Nwd827g3$1O6LxjQ>C$H?OBHy|0P% zr$F-7>kwGEuBWT23@~4ab4MLW#%}G8SU&YL*1A66makYxwD`Z|@t_!4*VAY`tQ^Pg z{Y35JQ|M6D6yo+vllN9D!1Izf@Vvg@anVAgvvsh_tc-cuJ_-z!`@m$f48QLEE`!g92_Sa$Ha&5jq#*lXz;>qRc>5^lC zpSgjJM=@%5HqP5VioY1@#y3VRLY44)w9a@q7zy6R^uJ@6>6;434^t~P@uV7TzF$b^ zefmWfpSGrD>1XKOP6N(-t_H3PJc4JJj(|LYrMq3mSGazig9?vRkXy$yi?^{Tw$Q>U zFSGIGC84|hmnEB8>@m-ZJdw(wb_o%qvj`snS~2i~qZP^+~Ty>I^n%_*L2+3W@EGXr_nb!vd< zrp;7rcybW?_fJN>@MMntCQR;6=%VRF6Lv?`5H|RN4(nrK3VU-6iC1hT8XdezeQ(_m z+rQ1HAx}PWg*ybUU!w$W4RdDSP4fY-fKu`?D;2)PWkdA4e9kpKiFuIy!rs9MULyWNQYX+u$mz?aM_=Uad zIAn4K4q+=qe?i8x1P%$kz1dyU>1z*Z{GiiAt_K_fA4{I-e$v8$+)=n7c$&bRB{sDCp!uxpblSQUI>s)UCdY3^1;zRJM`t}Q|C5ZzBtFn^9aHAb`zfqam^<5?avoGi z29Pg?3z!83QAESrlm5yNq)UJArg@82(1dw8G_hl6(J>3<{KjC|j&9oa z;~;+9oQ8`dr7){5kWP80MWnlLi*K+?$U@nn%nh^e+{OV(Ib^aSjNwabt$s@w}w zjem&3@;Bs?-Z`eJR3E?9hhpdbl*&^^|7SshXTBArc{%pC*$&8iPs!J}QOvAeNbhe`!$}e;c;7vS z(X3-|LH8sky0VHp7E~$K68^$9Pl$iSF*gR1@vMI#?&HcYpX|QGJ^7SQf6qum zuOX{26O|liTF^oJrvjHtadqu>KIa(WXa={K|>` zawqQet8uug?jhBog-pGqKVKg-gx}#9K&xd+cr(?x{LO3^UYQ+5;(d?OdVs64599Oo3HU-$jUO8r#7ERh^R0)IMa6n)0sfCy;}`T3PdVvloe@tgVUR?{E^Q_lbTg)uX>XrNMsdSpMXy$NYwWkN8JL&iJG7 zFkgD)6hCv>e2e5SiRe{~PT@7cPK zzvUjq%bF$g!I@F$;*dd{J|_r1Ks~xjGm@U0KZ-5vG-8WJDzW?SltSw-X(->p^2rC! z^KqB!_@+(A=!;Z0JZ*}2+-)Be(xW7Np)P-{y^O@W|D(>b!{`Z<&7zM@dGy7@2Mjsy z!3?TY)2-g&*yMJPOi=YFb5iR_$2NU?txbX0W?YE%(#~*x@-V3LK1KJ;d_+U%Or>oh zlkr)z2ULBD0srLzuyj=zjhwB7qT}bVDO!b0-&T*xM)P_39X+UBI} zZ6VG$5qAIVCKe}q$d&=%9tYFk-*!I7F(Zp`MH zMbm;wU&-qOB~Y|R0-85pBI9olBjeJlImPdn80~vT;`DjRtdnpa?SA|VVh?3;flUv| z`*;=xpL+^j{Us>jdYD#MXMk5t1w_RRWjCsp3mqOSYF2Ok3L-CBwv&ZV|G*ggNVa-i6sz^S2rTYzhdebkuD2xgyAoT@O=I zL@=%~g^aK5AwM)-VT9~$(h@R(oz=05HBQ(F8aldg-eeL~X0D;NKm6&+4+XG$qHs^1 z62#3YNfLF7d%!}!5pvaslaw=|EO#f2b@P*DJ=?>#|r((NJ7A9v%aat?&$g3x=koISh{eRo`koSd|)cFJ}Fx%yEUYL+&vn}Q({ixoiWAbvWzDUbY|2pvg{h#s4(ObI-y4yD8K(t23)yBEkQiCvkfd%T=aVqTH(h;r!ap zY1CEWrc=5+&T-~@Z^aRxUH561>p^0g{0`q;NJFjd-gI(E812a#Ligu%GINd<@jH?; z_!Sy~JQS6oxnB-*&Fwr4|8FtqtV##f$(=B8`vLjqnnBE#hmrs8$n&Y}30&y>lDzIa z$y`=i%9%?YrOvG~s5|E+ObL--`)>VW2K`6jeU(R|@!sX!!+{bKaU~k8ZtjM*ie_lF z&4V}}J^sm>Sl(f06#sr(Dj(!@k{_YEiEr(ih5DMKsA=07Y^e>RM-vT+k>)%&b>%zY zxlY(QT!x*}UQHDTzF|(bEbpySNT+#^WcovQ(8=xf;PwyDZktmiQ*tjkFFB7~6o-*(P999Cd<-*KVM%I_9YM`~k+|rzJ>*@? zKqbjMbaBduUs2K6IpGpcY+QmOwHzKc-sL+K?(z4dN3oWNF2Sw{Wp*Dk6WpDpVS3sw z&STXfJZUmONxd3WcANn^Cz3g1lnLraEVp?44$;YvQOt4uJRIHglWM%!#|<$_rfa|I zaBl|kpk|IGo*c}_z_0e?!SYUwcAw2}Q&8t$NdDxU-y9{U)V?uj6v-!=UF8>rwDCuO zd}lg>){8v-7jctKs>sf*LL^kVg{+&mo!ih|LWA>sz|%or{Lx&N@6C+mwcB6N&CzS9 z+rmvG^n)@qg}B1X3EAZ4>p1$>%LL|a42CSfZ)j%4pzK~{STHP-ySJm3ZaiBnJ}m60 zRnJU?ZH-6iKu-!3r&z+A(^gO;F65TH(h;)HC&~2{IzW%Af?rDpQ4#j<*TZ$;Oq&GM z<{zVRWixS1=mxa>?d~@?({MBm?rKwR1yTA!pVE)QP0ePquV;W{9TnVSI`Pn3 z+Klrq9$-?$A3&^}L%IW&W736aZl8%XemyY_U-Pr+FX&^E?TYBGa8Iga(#-{KF(SKW z%%w;3-;nu*FX;G(X%IB~C1V%BF~)`p$I3?u1qJbo@?p3+&wg- z@|lo_9}SlB69dKyqy zc9V;ITn>}gZ(!I)U0B&{4=W1Z2q=Y2&bd1r|J(YK809aer&b)Ke=XyPb@(x+dEpF* zd?yDZA4-a!`ztb{wGGUIwh=TTScAs(USYnVF|#5`ndYzaqf@j`QJ-{`h^B z#4MWxfo{Tis$F1&hpyw!L|?h=sJn`s2u#LhS>;%{XSwGbu>_;*IVf(wuHI0GKyq%EEiY=E_Q|H ztGQyG1z`KU8$2F#bHj5aAxOQ7{MniUW<685g3<9X?p7(t3#*t+uQ@lVX+Y@oG@N^FGOH++eN&%>lK+v z-f<%TTpDz(mt1#DC0hy;!H5@;k*&frWgSVzNKGP2rta`~%|USb)C))Z=dhF8JXyuF zcVT1l7&vin8Qh4N01tZYp?u5;)DWI`{3_HAe!aOv)tBzUhW*3EAAW5Y zdO5@CjGzC={QGl>+ejhcP;!=#f|5$l-WPP=j9Tuj+y^dc{AbEN=pk#zg^=(40Cyg@ zbFw+>#UHmA60Oz};&D)tJDy?6{A<}wV$=?acHNv0cIKa{s<3}}TFo&kPj3;r;3mnn zUM4nv^qJf8Q;oC^+edUtOD|WZJQF!@*AcB)GRywv)JK<-p572Q)SJVVDIw&}wpZfY z(S#nHnMOyIe&MFw)TB~tZZR$=%(#waqwqv(INg6zhUqkYAd)C{p;z-3(XO5YWPEfp zj9#*c`TJlDH)c!U`#&q`t=4B z{oI3gF98$#gGt3tPyG1vRb`iD4B4`4C3B?s7q|G^1h}~N9(awf1dY|goGv(={;7=? zJGspie~f?19DJsNU#D5)>6YtUvdRx?pfw6F9+jgq_M1RyHV>(Do5h!0t`dpZKg7{n zIA49gF50K&1{2(=sC)ZK?nl)>&imeWW}=M>U3cjj)hPNx%A>xK1J5@Qvq^V}Z91iG zOE2NMkzwGQz7lG$E@s{)9b>v?o!}mJv~X&rnnc+=jYh`&Z>&rNE;{lIg(nP1wL)+EwuS{jFIsUrNPTqpzbj* zlxpS$7Ni;7JvD+_%E&%*dO z?@tp9HPhvP?72g)ZP$Z*ZVZmhmK6AZSIM=x#pu)-&UNwH_-N<_a<90an6Ni04c1QL z9p6mkb#_bhZv_7FRI8C#C}c3srQ70^3*U+H<@aQ>SP3RNE+98zdq~-zGlCb{5+=p? zkadQ;X-1RIq>g=7gz;OAN*#mX(J$YWI zU=lylZjdXAQpJ2o!Q}cdY&rOr*+1N^N^HgCA8Mz5IGn<`Z04Ug|%0czcQD=KcS zAirY_;r(JO((hKvI0YXT4PSg%^vTAUf4yb|KkdT^zQ!{Uoi2h`A+Na`h^6Rr5d2XH6UDvfmAl9R5Hbn)FbMK{?pG zF^PoMj)l6!h4#x6U(wvXx?mVViGO#Y=%cRz72z!Od2N6fGpy0vdK4}Go`}6V!YoR! z6>rLEfk~AsncF%@?r(7Cjx;&bj0{E5)FJnoy*ZN^Hv>7EJxrR97T${khc(Cq!!*2> zc!dNwErcm~DI7Deh&*^AMYJ4vdid#C()}zQj?`E|UPv2wJgNp4iwS5fKFFonBvh7u zj=)!vM%erK7Po$TEdKdE3P&idVutNc1E*D2nHd|BNTc6YSXifxbKZoZTBbRbZXXXn z#xI7mRY$RG;}yp}m%@azw`Cq*7JRBo;T_d4PY9GGR5e91`7h~eo+Za%<3hUAenAPE{c$Fdh_?~6) z`~g=>e#$3TwBg6X*1%)<=uacloh=8!E-Ua{K`5E!{EW<>U`=SM8ke|oJh!I4g#3>7 zB`e?nnbuPTi`-){Npe4I><`1h!G9QWLg4+5W3W}u6CNuRKu4}UwXoVFekIoCS`>$X z%g*aeg}^ABUVE1WEm;Dd<-^!r)uZ{`Xe&(IwhimtZjkIF4;Y`PhSWQ}n`A^)lJEB~ zkQYuP+2|FaIN2{9MiooI%{c(Y($TO#MV8kQPegXBA*5CIQ+R4mr;Zp+)BO5qsjng3 zqx=phR*wd&8?tcey$(^XI!}vM2hwG~qA|0%7$p`8y)KtLJH?}Q|zDP&l z#@5hWU4bF@IF^{N41{w-c6u)%TO+i*S8&HR|tPWMj_Wljg_vYi$SVaA@zbY|BC9P4TinMxuu{$B(} z9xi8w`su)+moGk4GT^PtNAWLi8}Vp9pTBls4FB-%cidZCN3Wz6;K6l6DY)b@X&Oyb zYHJ*m)2>SW@wzzU)e24`UYD*qqsa$^_uw3W=T4c zYdaO7X7vRee!Uq(-WTGfXgm0(nNRll?{RNTCB7z}Z1pm|m?q%!D|sPo-3 zJn$rle{)%l*H$k@R|PG8C^Zq>(kaBYF&)QDJiyIK>EZk{4XB*bOm1_TOx4tn9sEDZ zI)2xs6WAg(5(hGS={!4IBPu~uX4PxDUA0A&wecnW2hO;sYz|e~ zuR+M73~}BjEnZJ+A^w)Pq5p;Da~q;7aKnB{zRzO}UeR?x>vUZ-`PfLjo^GLE?`o2n zKN5+W=}B(mrq?2ez<)$jLQCA}?~B8hULvQ{a)_j8FTHWSnFdZzMwR>&T>L9hbadZ! z5xa33eHpER**{N`n`U-cGwK5_n=qSi(RbiehP}k;7GwA$CdbLbU|(8rK!fgCqD@Ea zRzTHDPq6FCdrVB)2Ho2lKz9BDQ101{>w_II;F1+i&z0u~Z-Rv!KaF2|TPZyK_ zDziu|KZi3{y~cR$R$*G&Eg<~-Lb&1EK}Mc);F6c*k|mWK$)BrEmRMY3lHIq`+_h=& z<5LXyzIqEY%_70^LpZL{AI=+&KMR>2+RzgJi|!n=2z%VE_&dg*FfnxrDyuI8pFdJG zZb2ZusyL6tILngJVJhf!T!H+ws-Sb-eWB^iGYpT`6EfK{n0oCF3H_eQdF0O~x0yte z9{z-EvoYbP$K-PtUP`bP4_3pljjyOi;5h!fw6N!cE>6}+lfCfOjI4EvgpNudAfd9X z>S#mu*MV>F*Iu37_@60kUa$k|-*(ZveK(PF`FlHto4b`v*TTw*np5R`M6VR#5nOX(SD4cO#k$e_WoI zoIxM!-K1IX!eNtVIXS&w6WH0=;J?6z*v}P-Cj2p@I53GkH7lc+FMK3S%V_q8&=fIOQDKOdO4ScvQDZXKvLBGx|Wo+k0K_;ug@}_@aYi$GBa938BQ20s-A1=KG%XY69vS1h4HA-h$hbY2^jMF1p2iAhub-}%- zzZ^ogJP>zz^>G)yRv~9Fkqum_z;0Pn4BHoY;Jv0nbTzET(Dxy5aOMI8#uv2%mx)|Q zui>s8iWkeNTm;k2)^K)92T3}n$PN?pPxi1=tLM<<~)=) zbx5Z-rt%`M>5-&&w%7MC9<$Y3YS;ShA^Su-#OQjUg=OH zFOq)<_s1x>ksA%xPg}9MU+D3uZ6-}hH^g#M{o;EXvP|Lq0XR~=9Y!5Yg-2Vw;2tca zyMm)=<^7kCcZm>}SZDFu(g+wg+XLb)_mTyti`f%54A{HhG}yQEf8ou4%jo?L`eZ^u zAoqNQ2WX6HWSj$Y(756n25;4-e?3?lnpcOl;lJorhjqd}t%b=rz;G!Va*)hy0_{UC zn7_V+*{K^v-(NZa>3$roZ#IVq?HbT<@i;23dyelIWq#>oeST)7EpJ))3Ky!$BlB%7 zoz$@$++jSPvkWIc!+hbMrgxkziLgz?nP?6 za&Nxr52V1d8+l<+u$I&p6OQJPPW%1bJtDtpaJL!^GjK&sYu%AsOce8h5 zMjUIv}NAkrLV*&*Bhk09rDWf96eyUhP7Gz$>R;?jy`Q4ka2t1gzs!_P|I*);`jqn(2h1ydNaR;47 za;~9>`MIk6@zb}F->c5W+-RuGdM4ruM*Lz*>%wT;%|N1axq*ZYYrrY@4DtSz-H1c2 zG4Dw{XLRB!b(dEs=gL+=eV-(&?)ZvGuipz1?UzCClsecrD2h~vP8a*6|71E9%!u;( zK599rgZ~OvP|MMZXgT?-=*_89jECwruC~LCY4akL4~|OGBlBnAdh1NA_PCEA@eA>16D9iB+*;Jt;Ke-qJyEQl^o)CbrxHh7Y4Pti znDW);rywe$i8KsK@uR2T15x5%h7qNZO5JeGXwhJIrmtt+I+kE|LJsCOKIA%?SlY0Y zaK~n7al6%5ki9wT;34S*Z_~8lY3mVia1?&;E(Wn_gJamf(#4>X=}yD9$-u#(Ib?Hy z2j6o0I%jys8sj%VqvT^e^+ikCVBJnn_-S!^YyPXeWvU2iDRpGX^mVxHeLLAy)d(Br z<-w3Q_2_n3gf@Lcu+%`z8Q8ufPf}x{CR&-@xN9_PwGZHk>_5_>CU^^Wsj;n!nc})h zcj065IW~TuE%fgENL#$GQGX~yQLYCTT2A9w8x1a0`Mk(iZ~z+j`H~ES*^oQIpGKc| zfmm2i;)1u3`N8L5nMV#u6-yKS87E-k;WUVU>dby~yvtgc#j^K$ze7raEPKDC3WEI< z*`Y2L?8f){Wbe)6mo%Olv8S{v;O$Bkk)vS&{LQpwr~J!j*)JnOzG5`o{~g4HUI+oZ zt19TA7fv6x{vqMTVQ}$e70gjQO)>`w=#|Pr>>ASAN?H^|B=M91E+y!&k?48rt6U|-hkcpP8 zso@ZC^?k&gE-Ya}z5{ci+KQJtvK_)lXR?n&kAh81KPbnUvW}T5Y_wzzS&2nlj6}SB z?A`qw?=MA`O}Wo{ou%xDgI(-ebHW-#rm-EP-PrP)+r+sp3CzaLVb9!(pn+7C`aO}R zvg6E&!NjLxlKqgL+U!G&I@L(s@{-FH8_z*%$Zck#FdN)=-T>_m-Kf|#@+b3fk|k?< z;1ySX_#3xWyGdYjtDr@_As!l5A+Yj>vj=K*sZNbHjFy`LMtNiD)hS`*;E4u${BJqs z=0yM>uoRl6e+3zxAd#n99^;tnM!oi)0L`_Z$d)ssV2qs`+xFiQ_Q9WSuzht1qZ%I2 za03=^KllrBd%8fs)tpt|xf=G5R$==!b=kTNQn2R93ubxYJmv=?IpuhQbk03UK5c0N zwerR6?hOj;3HNpI{9*>I?2i)i?#cFfX<`u9J%_lGAI!>sCMCPMPtH;XSQ5-3PO50bD#w^xCauZ=OQR-Zbk80rk(DT82VJdJ&69ZwqyoMF_OVXXIAZD?Fr3Se6XhR!4L z#P=a!vA+YhY<>uvN`2suY@Oidxd)Be`$%5nVrF^SQV8DQfguutTlk(a{EoK+|FLQ~ zcajEg({e=cbA{5vEzPv)$!faBQkPk9^cpty=<|x10sNY!hcUW-2;ZVHkFVdN&9~Hd zp^s7?_-GBor7g3ucF834^SA{;*3ZfDr+k1wf-;ak0c4_=(luQA%gtILkz zeMU%dtU?>-ZaNaT*hUeX7v-Sl)C8H5p_I73qhkgB$WEtZI#*Mfe`LNFQwH4W*kX11 zQPNhN;}8c1Pg=PC?1NaPt_yAX^)w*sC9cXih7Hw4cwl!APOyz(Y`nGUvp?sl_g7Q) zy@3wIZ^`2y8IEMXmZgAiWe7cRw+!~*+y|@P)R1DOSTfKk4PMF`Q1h*s*4rE5VuwW0 z*OUX?wY7$HAjd$o$0Q6-P)UCMaS^X6d5sSEb&D*sJK<0FVfeL79?DNYB@KKNN;E6r z^8r^_Lw^zzZ6ke+b;m zmBe_zH+6qdLyg3jNOFigO+M)@UO!z9qVC0k_?s+o*gBWEy6yz&Z;zS9wIR6ZkRBb= zD=?2vt!D+c6}WCXOn(O*qn_^T!Selbuw006)yWX%U6p_lDr2bYsTpW6{-wz7+%x9U zmxJVW?pRuXI*{z`3nJU{OUSjLWkgT51ajx>1O?M%2;Uz{O5TnjALc(IQYxJy^}#y2 z>tK<1VM#a!9@&X~#vUjhA@q$FX9`SBZ`!BQfu@HZpl6_xuq!`AE)ecIYMy$32 zM^v-#h*FDBh~NC{Ao~AJqAfvvdS7|vN|HTG-pZ;M&f?jiNlP{9l zLL_l(KsViQMotb4Lrq{5JvCgxN#S z$!Q|X#A6g+JST4~-ja$|2hgp~FOuCh}w#jTo39$HJ@ z`dz3rZw}-Bbj%Qsd2ycxrhH>I_%Fl^fnEJcV7B?5{eu^+&!JUV7CMAX$-}GY*5qo;G;-FoNi~Wl(Xo5o;Z; zz~+v1V$)0{nT7um$`qYpv_Jg@W8r(U#!!(a$z^i>L2k^Z25Bfww1cKR6E5H6EZwLd z1)bi1X`bIts(m^MZXcFm+f&{_R-P32d&;s))D76$gwf>iPf1uSJ(~J-p1SNDdW5LC zTqL>cW$4eX(o|;K1hGr+1IYNS${I_RLiZ^*+L1Dc_6FIBPF-#Sr`?I*Z!`fe+Y565 zH>5K*{YRf1vn9GU#k79pc`|bCD(X2onRBe(O@N7H$+uJNl4F&umiaXpc~pnhe_IW0 z^3q^xeVGid%OiXKl`>=0o5*jKZ=_~zFMZ6P<-(62W$u5LM6Ce@yf^wiRh99F#;Y=H zl9Mhgy?83A>e@il7kPmF8$%e9cn-`C<=^uI0Nts@8LrG)!3BHGuanuwruFXMo6CI2T{LOVdt0I zknlbgZuW=MTjyCWt2KlCj8h>!3%kI4Xe108^>J&hEV1`QF5Xrhj&>cf@S{$K3wYZ? ztyF!$&D0ytX(hm<8)ul_nF(C5L>YZfgdJ?~QxbJ#ES3LlPwy_Bik5;?##QJ|!9Wbi zx^0JZszc$)s$_85lmjRxOvH*s9FdNphwk@~$=fXG!X;nnUUoWZ_NB~12VvU1@i=XS za&9p23W={eOTm8!bv9S0nppxDbMIR&Py?sr*H+BVeDUXGvf}`5{)kYZ3 zJQDl9TSK?|NuzmtI_y{<%M9UW<7|~YOfWu!8x5Az`|3a^*sde0%e8^se-01o8DjEx zOVMlfELweQ0lj7Of&M;`gQKQ))2J7j;Ps0vr;pXFMcxy8+D%kyv+J}deM^ECZM#EU_O0vY| zg7{~%$!9pau?ytS{7=z&xb^hDaa=>Cw2Mj_ipp$%&U2p#A)}BoDkQQgBvhn5L`qAd z5)Fl-@j1_Z8cI@DM46GDWF>r&`kmiD&~iIlOy*r(@~GAj`P;HEA#*f_4tq`meoZ8SpBl(<3p3G)xTUO3XA-;UQ3~s- zZUvJCu75y=H}*YkqNAKIpx*YY^wn9RYoc?WcIX?E%cYN*svW0oI;)pa+cC8w$>muU z*_vl7-yW=?yUoq1<(p5UzNe<}V)_VDpW%YHUR=Vn7X;3R!BF_~wGhV4mu7X9#;`8; zuF!F#YG_Ec1~v%XzEM*qvdqXR_JDyu6-ilO@XZ&L>$#7e1t;ju)62NOyS9-)<#f7l zZa!^qv89Th?p$Y1DWemfM7O<>#Y_!pJSkPmJk?4N&1r1FzeD7)P(2TKmjm7ppir}B z3>YXp0A0-?)Hm}e7g=YEF&!UiN`p4GiJDN-Jq=7QDTD2@9J)j%A9aphq5n+JFf3!r zcrUT#jGx?L5;pYFVPgiU{^lAwGok`*LtOCr_Aer4V<~o0g*v-#|xeJ#<;S&ciKI;OhFDJk**|!*Wtdkj4Sw{*@ z7n3OqTgltpyFwq*hiJ_IN@lugab~d&e2~_BUhAeCKlaa765VVAE)N(|q<54^r&lwd z&uxc>X>E-7N(_1a{fKzJ&_T5uY=Tkx6Uq64F338d&fa`r#GW`IO)pq<69uC<%>THV ze3;2V%QcQR)tTUoIfv=b>^55SXd*krB@T8dS&-g~4EgzBviO*h29vlaN_5$&jdT|| zlM&W7>=ieC)*|f!Ec6oN+Z->fd7evu9mt_w!g+8*hdMDDHjYjEvW$(eTE|w}&1Elr zvt{+K8MCe~f`d+NG;#OV#cuaE+`&g%nVZ8qsj`$3S?DvC?mHSS-r8$LewPJNsZ1Xs#;0*n61HGZGjPasl)SwO~%ltKj{g|G4}1EFL~EkLmno%Pni$LeR_Azae3x zl=rgH%467UPoB}AE%<$sTm){k5qIai;20Yjfj86^2pl@3_L|`s%X3&hW(#X=9mX~; zI>D-*cujQ_4v`z7Ly6V2tz=_sHLY#bC9(q-IFZUDX6o%NBA4mExT~gu&(&9pByIUb zHw*09u0B(2e)xv@v6+L3(|6GimhouuD4Xm5D$G*rBZT};4|!`c9-QM3(M6q3^y~ge z+Vj(1yk=n}ci~PF(Y#c|o$v{>?eJ)ZO?uAoedIf?LCGCeM;{~KR>pwId=`XJ3UfRx zlT+>0r)BHZxS`1mw^LJt8~@5A%`?H z{A2QDkhULm*L;}dGW8WEas1m2&J!AmC!0@Nh<5+NG-QiGgYor zu`Y23o%NET$DWU+>emYB=10}EX|o0Xu1OX@vQvSRTYr+6$~igzE=i;c?o<#;yTyISNJOFBF?B4FzApHexjr1-?)K!zRayCR0s%r*0|!Kxv$?jHfR1rSQH_1Mwea z0S{8g;T*|@xYzYKXS8fR_Bz<2#rAswU%Zdbdgo5_jd#$&i5=Vxk2>x_@lQI$eg@e% z7(xB2PvZJQIV!uw0#BPs(9CHkNT&W9rd(YCO-D8`DNTl4viN>wovH_JIVz&(99g0B z9SQ1+YoS`dfE>7`$BY;DjZcn-Q}z0IT${3lcZuG~kF=3zCeK+e=3G0e_Lg#bWz26; z0(*$o);r+EJqcKU>>Jj0XMr|D>DLoccaUoJ|&)6U|4GdnFp?d^NeQZZ=@X=EXbpIPC8r(JB56Y=p;if%Hz)K zLPl#{J(C=Ya9CEI{cn03bT|CLu8e59*}sy`RsBxOmRuym8r`9O?*jN$I2-;wY$bCm zFVpa0dN{+d23w?F3f`79YUG0z)KHDoY*;3(tp@F96nxtPAGy}{7c zhs0}_j3v86PD6A|AagKMN|do&kKEcZqjG)zY`kq>k0V28pm*vhK2yjJB)quKXs^#> zPDNESo5o(^qL)8otn{?t+dfH%OwYqN>tEx-wkI~HRj=UBZ&etcI+QQ?a*Ot6H;XRJ z4<`5D11gQm6tZp+-1YOGbaYPvS7zN{V;!l9{$YBw4+S;Ppa(Q0Ho>Zmh7hy-Cay8) z=Jwt*qujQKOdsPzwjLe}zTV?_e6)eD)1JtO`}$DH@U={{%x|(F%81H$O@@xJG!mSB zlag0UxzoWHh}D@(96Rg~sj-rVwemkmhhwftrMrvBC`AfRYz1zSwiI)1M;Cc;%oMB? zq#$6<4bnSJ174XL!Q%OQ(Je-r-(+CN>y+QYp$9eib-Q%Z?!MtUEWd?IuvtFwunmoYPG;$AxZ z-f9@CdPltRyD0>6YOquKAvYyn7Ea3*5VL1lG)erKdAHq{tZjM9D6a3ImL1Vtv8^}h zcq6cQMmLb4j-|9ne>udxS&C}}2d2U(!GBc}iUa+N^I=#?LB8%HM7 z8v@@>v113ltei=uu9(B0*Um(`ES?4^kHFo5C+N=3cldf+Y~@1DHrh3?2v5pH(WV%E z+`RN8oqgsZm&KUiPf|iIl((Xavf!27l#lHN)0jJ>=g@DTtkAzD0>73-a`FXEbk)*s zJaT&mKPi0)-@(+;hS$v`Q(-w?m7GGBUI>OU?%&b;*au=*wF#w)hVW;n89|@rPa=K7 zkI2g~?3Zp4+i~20J%5*_-5rzgZ^tZjs+h#@cv+6w=Bv_iKbWAQ=H@Tk_>tJ zos1400@GK&B3qQ3xD}=h4X_Zrgr5WO9!+exMwyhPc%Z8H~!pckE2X)@qQ?s z?m;CJN7LXbS=7#A7K&voiQGvS`m=TqwQYUGgqjavMMXdCJslmnne{W z3HmQGihudehhIE>IDa6coEB#GanG-LU3>=u~KGHxC8luZqHVeWB}?zTv5y(onW zlg<%|)%_xI+@N^3;Ar_FmZdwcOu{^wM6B85i%QX};fb^ct_c?5p})rX@xWqScA*c? zdyK@|a!Fn(IFIZ&^pQ-Hl|||2=V|+-2Ao>7OyC$raZQ(MXnwT_<$9~pZS{73}SC#PsqBVrxv#R1Dz$i!14@yzPqsQ0)IFPwXfDq1JVe>SVpBt&oy3to`k zTm_t6Fo*fiDV_Y>wxB!vY0ytRoM#7?QW%7xjU3q8clRSII-N1Y|9Q9FrJ{|@@l(~2B0_WOT!4ZQ;m zFjLt18#ewXW!{5a!H6nUaeqgw{=}2ntKG{$ZU5@TqhVV6jGbvyGbwV$*IGpC2)*?6he z8uu)bp|$eiT+8Sd?&pLHqBkk_AbV*W?U;E*yf(g`xttYB3YSMhWbP_BmokKZyD^o! z^rC?K{B$_^V|fV2Hjc;8o3FSzFHcg_%Dd!|eh$es*TUi7t?|stXXLYM7?Jp8#GIS^ zQRM9@aOKxT(LbeO_-@-zI_=4SbWfKS+P(ToDx9~#0jD(TC2^K>+pfdaP2I(;es~lu zuFGPLi3gVN?!)S%<1u}4jKC0+q$93&qJofrjh1^zGGa#3?G_6#`BezXdVi1k^W`^L zYZ6HH=k!xEr4yWsTOt{VtDy>y&1m!BAoFXh3GTi!3rmz`K{eKh&f9wei`LumL93?l zm1BC)vxmVcYkjC?>~?DHUeATWaok3p;pc@?oYIy~oBtkkajLEd_^Tt+d9BZj_~i5& zX3(ORX4U>epQ%^zp-Cy;eA_|yzx3cU6VlO)si7Z^>G3ahr1)2IiI`T90LBuEu>GVd zF>60S3pzhDqhWxXI(s_kuPDOb)iV6>kCvkUm_|l7zl``yF@T`jTpAqTjzMm!xFI`= zDwvotpZXQ)uDh3Leb)o9mM+C(k?r`~q8Kkd$|DyAHtUBiPq_#`2X5IKUL;+TO7ETi zh;nPDVpxyh?6xYVj_2iY)#n`Q{mlSQdhTX+Sq%eq{d1&KJehu$ZNyENj$(t>SsZcV z0zRl6i9NQ{aeUn&+WCUu1X9F(-=dD8Q`AXf8$+i)TY>c7A@Nw{;b^|^2W=f$!>Cjl zK;xY}vZyDGnjM(OOC3<+?=R4%)ox~3nseV;NktA9s9fTt1wX{zj#x2DCv|Uv3T1CGJd8CO#ghH8ovH5`jupk+qag`irz`gvQ-ah?f%D% z(-lWbD$;P3Y72TUdWDBVba|JgLpY(hk@k8dG9!QA;s(c7lK7KWu=<-KJX1~(@&iYZ zUFC=ymzZ$VUN^$@;A>?2bhX+E3^=w?`1_rV z@=h&CEL=%@>{BXrc3vf4Z!adg=f02(?|$-A(}xr+*+E$NvP+@Q7Hl!ErZr`S%ZNU;TzVjc4Jqstx@l?AszgoWy|LGf;n58!S{w z181F1a`e!3Qlgc~ae9W}Pop@QTOO1!vgmxfjl1yChDJO5r9q9R#N&rK$mz#Ir?4;X ztrq&v?diC+yvQpGu`l ztS_nb-$M@0zKut|WYD~|#&l)-b#C>*LXZ#4f}?$Iu(D~-_vM{Le@Q0e zJBwl5hviqk_@V#`dkzxI(jDyhL_0QN;XF9^F@wBJKF_$GmaD9PpG8J~JV*k)HlgY^ zN-lp%CpJfIz-mP(J$6oTp?N$djiUuV*|<3R>DpC#bugLA2k4Mzq4G>km>Rt>H%(w- zW+JC7e8WObq4?P_cv2vI&jW%%^GPfCXP<(EX)$n9av42&?g4qVu|*vBU=_Nv-Joa| z4W)*1tm=i=V#lx%kbGw=RM!{TN-npgEAtL=9zUEgOfm&6L&K=lgEad4K@?T)sirF* zxWX-q4WMQ*hYXwhSls5V&ID?1rx`CavFAI0*5fR)GIS{_P8Xb=%5B&m9Y}w@mWG7C z67cA@h{SF;A?}qkXruN`nAz^kj@WRNja;S$p*MpeV9FJe@@4`mmoLT=*ECw$)Jv7G zN6<@_C$OzLp4w%&z~+Uq@b3pd znZwFQV*%(=%AU4^Qk%{2>e^A_-SdV>r0BxK=8Nz-`3EHY5IE$XTQD`v0}u4%V@~S^ zeAX^UDrJ_zE=vOIH|gMm@Ctf1Ko0i49?vY>vY>K#}O+F|;*1ZryCgm-E) z@hDS(Q@?1iO1@9YxLrfgz7=W5#W~P3v=UB8<&gLr$6?;Z6i5zn2K`eUq#b(?8C5ZK z)`_Du#Bnt~DU^fR6%Dvl{2j{_wAj?En(TiSlI&kuXC|h%hsbF>V-$*C5KqM{MsJBc z1f4ns64s+&aBddyzF9+-wfK>)LTmc0@Csb@%z&D`(WIj+mF!ScBp0XUFiS^2rA6r- zgm+Y-m1}StA|mCyP#^;aMnwe zP@5SIq}wkN7L#=F7qWOJoIeZ@5m>41hw`R=Qd8ATEej`tiLoX7K2D4_-h>Zdt;lbd zF%j;>nRJ5tGU#fF=ekutQS2W_^v(U~4aF5CVBZqPK{y{CXqrv)Rh4P-VOd^nj|Km8 z=qz6I<3j%K%2oVF1oR6r;(va=Oy?bW$S9c)LpNZc>Vl9F9OF*oT&0Qi&i&%1!=_|S ztsl^_njn++gLGRcv&H8k;IHj%44Iol?UYqXtI#1z_%so{-l@U8y&1gcggD+M-HdOW zY{EAdC-R0xXLz@yhvH|C%DC~{uhCms#@Ms&B{3^;VYb=Wp@GL?ym~tcJN;JhJHOiT z5uwjfaYrSm+MvsQcg!HUhC9U$p)O>njuz3ZoxuNkKa(#Xy@n6`6@`hZ74$BRq60PO zY^NvXi2rR^!E7$>#vO+nc;SU#wN!ExU#n%o2R@z7yUU)Yuh%v)XNnETaIN9I=dpL}(`P2K8 z=1-C)-NFuav%_H~Wv&a!^>-8HB!!EgeL5oU42%E^|6*8Jrwie0RQL@ShVo;BdvFzr zz?(q|e3$81-qboAEiYAbe;hAk^j3ivCAdcJ|Et35w={USED=BMt}|ZWGYb1mfiCL5 zLQK}}p@TIkWMNVreQ?1Pm-*chkH7hh{PqjNlX(TW`ROxMkMAL~?qtGq7f+^Ilts3M zEI^;mVp^^@2PcmiiR+_XNK$MZPK$FFIZZfD^i}qgoT_H-?Y^VjKf{Z3+wBr8oMpuC z85W4ABzNMkZeuhwxJmoh4S~{CcCc39w-_9DBhRO$aDy`5j6#bkEjgY>G#zs|e|#hU zuYWAbpBl)#+Pxf`ew#p4?_IpF_m1v9=1dE-dFmPIgQ-hjV|<=7|MO`D@3T0bXG02c z$CD-a+{}{ioqCK@?S4c8^IOT?6Ao}ZYBL>Pc$?V5ebBL-3@>9QL$8YzYwM5(ABJVX zl+pXCs-xg9S-he$IW`X#9p42?N>iZN=o;)uodi;5@kA!j5|wO+z=}B~R5|55tWj!5 z-xmjXKKm2gJarQe%a@`I`2hz`^Sz4M)wg*h{r^Oh3aq47rO zKmQhKE_Wy4t0%#e6-VLEK1b-+l4EoZrP7AN8nirHMASCLadW=+lgm=&BvI`ntv8gT zbJSVJHCAv-ozlTw$@fW+Umcjg>9DCEnFih(+GNdkKx#XImd{KV8NcnK%|sSgIn3s_ zm|Vet9n<*(XTM{2%NHDKF$9Zi>cxNkUNhbnERdZCAR})v9GjJo=CeioM1#fr!@FH# zdCN{>d+`zN%=$}$U1f=1w=TZ0Erl}E2%s%aG%__3p5**w{;mopA)*uTG%f-x-$&7- z+r!E6HaYkbo7Bi%0nmMbt_#ik* zZ)Ax|F2uNsA!CLER0!Ur`vPaJalZ;*@_s(G@0djnuRF@gl}FR|Kty5$u?&gDq7^9{=|Xwlk+_P`(4!s63{feM^W<{yH+xK%Sf|xQX1J61-TY!H;!c zNty(&QO)QonExsd9XCn|j$%tVx#A32Ty=!xYG~l3``h>tStfj-W*6PN>phA1HH(hD za*pQoF;saga?++x1uk|J%nIq^8s}9}S*xWYJ%MQrL%s;?2T90Z`-n7M9z`uzG4QrY z7xZ{nZqZL;XpO%Q(r@-bo$?a?kSK?5NE*rqIArrCS=qea*Mt0w*nik;*p8-uwxNI9 zCQkLODlUmFXR743i>-_kNsodoUO$}(car-sqjWbfuN1*#=6@Bx->7SAtht+9x_p;Y zl->`yhjx)+*}JH$vn)CLq!aslrFfe?Bl%jTxgx=mKz|Ir!u~re`H5|2yoSFtzj)0% z3@RPTmrODSbx#p2s#GQ@>}Ty|9I*1q796zRNY!G-@I_80yt1hkzg5VPPpObbv#^h% zbfG7KUy8WFs8TFYZiKW?%^>No!8Qh#V#e*!{DMy{Xx;h^CkuP9l3UMFC*g}YLGCki z(jBWbD9p*e@8NlW$_xHqwx z$@YlD`KxbX^OzE<6!wPMAFvUoy&VrC=~(JU>ab3xAIrDp(o=15T;6pB7)(FS)g~%( zDK@H*?iCE42dCrQSql8H04rRjv5;!ik3dm-8P&eNbBZa?%NcOJ0R@kW{8KOlB^Q0 z+vh@`to5You8N3eLYB$MmHAXUik#>jMjN8D$=;KaID_SHPt2u(4Di5dQRO{&H@iVcg{R`bcypTT8l%}6j?1}b$I}S%*=bYN)X!ptC zuy4n0=0(A9c%#!#3QCnxBk3}2gbir7DT>$~t%fGsVXRS&6;unEt-XKtu*%1$a8JCQ z$(tFD3^$~Y8YT1)+fx^D^o(50r$cdZ_erQzSq9UJuEOyeNs!$;0&fO%Kzry*xStmV zE!NWP9gPx!Z#NpYgc}GwM-{B<^`=!L@|d}!gE7=%6IZU1M-D1qfj1wO=^9P~rr2Gh zNrEpaCR_stM+{PT$KiDMtahT7I2(eV9Vg|lP7ud~t0DEg6rOb*k8jlN=$iHl;&kCB zNj*)#^ve&5=i*?mQ7bdtjQsyzz%@0jAU{bgzo;5`D^CgjoAgFz^jyJtvy-Fx)gy@G zfpw%Oqn!9<`7vqo3rWbeL2lOWMDAf+Dj5>;gxIy4K*+7bq%$Fd340zQ%scD_7j>6N z=i>x%S;QrH-0lmf2lkM;<5NK8R6O|qbSBopn@LBqIkd^|=bYq<83nTjQuQJgHLtIQ zlz)dA<)AyXK*vz%hS$-}8)U&SHi$`jER7Z6k>th%GtN0pk((&L16tg3feulJ-#!W$ zyvi8H@AHHV&QhF`>rW19s-xFiCAixV4i$lxbl!+m{H#z&gN{$KJ+bv3k?9PFuQFy( zR**)9beD3s%QYeG>M&kK=qzXIRboL~CP$`BL;U-WesZwG*xQOUE&LeKxR=EJiFix; zyhf6BycBNyG7_EdnIk=;M>8AM!SRMQ*!;`lbVT^vG8_V&YGX6kL<3MYEW$=w#8DdP^*yV#a+al7PdLcd=82_e_t(F`990 z2660`ge&!qn0r`E)(d{bmZEU3b*mcW3OnD*jN_c@S$kG>>qT(At^yxQ&1XZPBc*QhL(y0191I^QBmOTUiFTkK;DTmy%D0fdQf%fv zYy6@4^~$2lFZb~qJ~O=0p&zLFLQ3?sK0;LX_oc1w{2^#N%nsH2uF%o9ei22hTGAk} zNE}V0K=FPmvF> zhg8W{aF~9Ls|igJIDPWe+Q^TrVVr14;$g1-)+G4Q-cO=qwaFq@lj=82qz4|Ta)H?) z^txK2!NDZ>b#UxZ)S2WKDu!@_M(@W7F5 zCNjF1y_!Q=saZ?dAqRWldFp(@Yq^w}wNKa;cFZJgzh^`L(KDiLNuxo*{|L0d%Ax!8 z;+Q|~qaiyhlxj4@6OZ~!Owk!H(a3_$C|+PGI1{IFoogLH@~PmU-#C`hG9E&vD!D-4 zc}rOSw-=havLM#!FZT>|sOvfdd>J>Jn#tHg?P4EVW4R0@FPU?yzA5BfQZY1)KLQoA zM?s&RCu}f037M7ip=y&P87zEG*4uogAr+6I{7^Z`t@ojKB~&my=p^McO&~CBGTiP> zp(VFoI|xfd5{gx{em+rp8DxNq)Qh#!nuqn z^7+pbB75r;*?f5x{UAxmjk3jL!OjkWX*P}q=NnOFz2l_m>U#V5G$yp{ zo4|hDiX}(RqHUrj?pt}4eEW2d?zmqJB7-QJUDrt_EiNJD+8W$kdrx9l83k{u?5Td^ zyvoc$X{vF*k*M9A0GjoSfkdPb+8Rpqq|Xz-E4_?g{tUF5_=j2#eF>e4w?RX#972S- zrF<9`LuZ3X!p!oWXwCwewX6}}5zIK6c{$ia_Kx#i;|p)Mel zo)3G;D6|TBWA-`O(zA~7cFGoJ7OU{ol~OUJ+aa0Vj{)jWxJLp1=tyS^G%85t_=%SI zq53G%jnp9i*C*0LCO%?6w*o5jVm6lc9HmZ;x9P1nc~m88hQQ@ZWOi9o>Yge=KbTC! zDHDt^X2(a$D&mOePEu}>NSLSPN2KYWX`aE+X3ET`v1rix!|6uOw{Ekyf`B&pBw|_!49nAAqjSW&R@aj)dc#Nj*%+; zaxyJLjvHR#hXcQjutL~pP0a`P7kEBI^1yTBM@}T$(>C^&v;%Z7OLNsW`w^Z>I)fjkNxE`{Z zZiwso&0QN2!_DZr&x!OD$oBPut6WozstW&+nBvK#c+FwPK+cOUP>W>-62B20r?I$N z@hwyIEr)FUIElWxe4bQ(sf3(V2{u!+f(&ODGXpF1VB3&;@X7TWJgI&Or#{SMt0xAr zH(sQZb#CfF>+3MzPflz5YQaYK#0Hv!nu6E?H|wMs$rQfpGZN_gfM<-mmSWI8;p6m8sd@ywd%G%+(4 z3fD<+J+)Wq^IyR*CBv77eCO!>HT&s^MfP;d^;fpzBIeR($5r9Yl2CBAeG6)3i`fP5 znqi5L;I!VU#ocTFz!XCwEXOw@)BZkDzV}oT5;}+UJ*ngFm=}@Gou8R5-9l#aLqodt z&vcQmz%{vTIY{5%rH|+t8<6gTXw&f-UeD7&g5<6^n-qQ0>swIA{Ae+CEVY zuQ7z2{!oomevHP!)oO$(K8;rz@8EfFmUl2oC3ANip=w<|LI(2=L9qd}|F}mAzmDTg zEjm%vq6vll0EV7wqiG8#Vd}q67`fO1TLydZ-yR)ueMTYKbi$iFcKZwmeSqz|>cf7k zw1im$skrw3S6szk#j=D9)VsP27PfS7ky~uZHPbK>lA>vse^A%%j>aoC@#HXGcY`&* zWS2fHT~yaresinD^)?MlU+eeK^3=eWBI#uIn~- zB{{edrSS5%ae0G(#mDe&Nx0l@H^#8<#F|+WW^WF;ruV=vF`8NFPVeioJMF*8oRKm&nd*OM#J?rHn ziz|#;up`oe|F+?f*v#_{vn2NdeR1&&VgDtQ4O+k8^@{fxuksO9=V|iSZ#{l;26bhAR%9YL&Rzti8Wu42%1x%_r8bV4Vhoc6^-8_hMi}*V5j5Sk z#O3$}9|ujOx%f0Qk~*viuk87+%3Xh!5E|pZ~c-Abm8yC1YP- zW{zx$fW+)3Xb?CI@88!kKT_gJMV&NwndZ=|Pq$-sP&i#5r6X4Ek0*=GO^M6xFriO8 z3bg;ov%#_};Lalx*3N4z+u}8w4Vo;$-uE1Yct*USo{gq~8#wgP~-DlCbMt zPzviSgFya64e?!C3-7mQfQ-~9fxYVp)xue5Q)VlbTH_AVi#L-+uN6V{n;8!2Euyx$ zhpD1r1-NILQ3+iG`sj8)9CR{dlg;mem+oY^_b`?jwstCLcXh(8z00BA{1*B8&=KEQ zUtoga0(VfxfgQIriaq0Xlx@sSVJ`f+N5)ELGs|A?tkjA*K_{fnCJ`hPW*m`agU?E^ z|0#r#d(XCE`q>LuJhPdOeL$(K0wq5i*R#HxYuFbB)$I7_<1l0QbAgjHhfQ$)!c6nx zx!DN|ME!?Qu#X?8hl`Klc<`;#DP`B~^Z@gFIEyO{Rv zR3jhPh~cV`$t>FIM(3Z@hpU7ok8*9v!->EYzj-QhKR=mvSlDu3CR$SNKMh(K@(PNc zo3OhMj$t)Q#})sLk&sZD|j{Z?Fmsd(Oh2buHw8 zr#p=CiKoN9x#IdumqeNt*Gch`!*KdS2)t2OqN-I9OsUd$&NcrxDGJ;P#}(vZxs@KM znb?ZX7o8x|cb<_7fv>~`Ixy!{me4817dfs=mhGCifK^@H2&21Hn7++NNVvKWsI8P_ z$I)9*bEgAj#;$+^52wPW1(on7rJa=4``X?p+rR{-+r!EYm&s#!0%mfaAgNg(?lYN1 zUDIZ=`;-%5Qc5xrR+>}^7LsWTMsd2=ZP|yB*=+ikG&Xwbacbc-op@fD!0gNTh$^n_ znAk9%hV`9c>c-q8+wwe^Fx19<=93^TOoP2uzXG;y(ttI0(jjk;7u}hvO;6f7kWRhj zIAmBYjj7P11`k`AUmxC)<^Ep6*=#^G>v0_V+dTpQQHuPD!6?CHbOSvaPjYuiHW2G< z^75WHEGm3P3<6Vd<)nKkBeuNuce;>qPmn`@(CB5*gVVC&6rU+Pp$^AC&5(!Je#4GV08H$lh>B zcm@YZb3`1uotn;tUYW!%Rd|U%-BXuhZ%Hl*iLRmWF6-{a~(YB6a>5rQCAYBJfVKh1P+mz z#t?d!l#|6C%3R0KK2&?tCr{1cl^<+xNFZTFo0*{Z9gOZzV^;OHG3)%~D5R|sI$rj-VgAZc7(O|b zGPMu6@lw9bsHCw7{|wONqc*kOvcO-T|a3K!8#*KGv4T#T9?BMeY^c+CK$$HYmcskJgx* zYi^tSJCE}_!i!#?*2Yy*e{qOH41vNdGA6a2yKl1}0+WlNCFT?jClHk5Ta7rPq`DuP(P`F%vO{pE0#|a zwO7k?zn@bsT5})YesDDZ!?+7S_U*=zovPTgKNXj)mJzKRu@e8hx`&&4C$R>pe<5nm zNl+a!oW-%Z?AWjXwpwcw(eL>w&U*BRZc;I4b{vSHQr9ey&NF3YSVeXTGm4eJy^Idq zHJ#3}D5t4f6BwNJP!uskiy93S!%K~YF!*Q*X>^`M{{DGPr$juUw&T7);-AYfMo*Gm z*K0^)S6Ehxf8njEE z=EN0Zwy+X72{(})_Y2#Gt72#lG3(A`yr z(c3im$yN=x-g`QA7dv78gi^fwUf@SA>ZPKhD(aCP%BTd^a^9=^Z2kXQvQC+YA-_-` znAAmN=6Nlcxac4ZH1DD__IpzsevmWPh=i{nY;dcd2DKF*g!()B5XF_^-?2vG&6{fJ z>fjKZB4H^uqtnGdilyj?5F17(Z8ZGSaUrLi|8T!Bk(QWO(A=yfuG7w!3HxV5?XMX? z#hsmWQ~N_K+o}WW1t-c~6M+va5zGGmwSe98sv8C#=dseJSJ>f-A~vnaovt1=j>gPQ zLvq23ZcA#TTh-6vOz|_C<=sOKS`Fz?e?2Td5{qH!^8CAyT@acv9DaW&rw%>)gbYqM zx2IvM$lSS}{A!VbPxUHrHj0Oh_TixPFoo{FZ$sZMoQi&~gS6$I3vyql;%mPx^pwPP z>|3q@S2FcQ*EhA{>(+XD=1D$TH7-Xi^?eJvrk7C7d)ugSM3?BU<8Az7dSnr zHvAR-neM!(g8DLNXzxffvPRMa;xCxv;^YEm+|Z#+V?w9UFIJ_s-QV$N@=?A+c{RVz zESr~lIfkF_UyIfqT^+7xhF zZYEs(;3oX_>m|wD;Yu{bbMft<6kgjNkFq;_am1T4D*Hl*vI841;9NYc-_itMm&mcV z$8W{z)59>rCLVrY0BnzSV|J{S!E32DB>4M5F3DB~N--2?n7H8*eHmDnEP-zhts|?Y zFOppeO&H&%$S@vShecGb&^>90E_8|BxUm0xD1ecDS6n}B= zd}hrl!2__X1ring1xCV6x>pf!hg}>!^(u%|EQ^9CC95E8Sq9z-azlwtO?2a=a(Zyx zBV01)61|r66g;h;(g`Q+(Bq;WW^8`PjS;#4%a<}B`Q{V7xpF3VpkJFF(4IqQtu;nX z;SOY^E&{X61|oCan5_A;1|LTx;Fg85m>Z%??2DI!!}b{@M^osbC|o2F0X5VIqnUti zA7)ehOS0)oFTG&=0iDkfK5SbZKD%)fgOas*n|CpwTB(O$AKk^`Ek5MO^9lI;zqx2$ z=}JO2++%JwmQ+Ze8$*l>P3Vn@aYP&~L5f?YnargUOsTI7U)nD4gBJ^Kr`-b2Y1#)a zL4PIMd@Epl{T0c0^+WK^@gNm^z4Y#yzubcnQ=utm6^ttngEcP~@c7h~w=I+9!#$pm zp{G}gL;jnts>>Kfzaw$nOpD9L%}4!a8w)(N=u0VHEeY=q2}`!Ar3p`-KS!h=oR7Z=#jrhWQf@i^m35;(bh)amNhbBqxWQ6ZaDp<;EsFMjREhsK>_JQDCoU?KhMGmlwB3Bm)N7hC&b2_A0=r4_og~Vb z)`EvlIWaD>hN`1xr1pv;HGVdS7mc*w!!>NFPM{3YcUoC-@$yOLPT(Ca%C+D)w;OnHkuIJ{QpWUOV{mP5D3hWwk9o8(hB=x%o<{g8(WbfD+^Li< z`mvyxNtiK{$@sa5_G_FXq7Yls{_g}0Tvf)1tDll%JUDO|)|`XeLGF;tHC+@C14FM1gdKUB_c#vpJ)! zmgMO|P44rFGV;M`74x-b30*Zog%n)3fj4Jjz{>gocdy_+iNE*>e#Bhl+9elryj4WS z`MRrA!F3um7a@=7qHW*XNN7*A{Gq;W0j0d!MG z2R$}KkvyWh__RV)Qb-sKm4^T5<0P9XBzLeqWJ>&beoa&C7wx74vJE zLnF+n>56O|_oWhNKH};8Zo##B{WGaqH=R42_>AuCFU4&^0ZhB{Tl(wOQ<{GGHrKrD zH#7X!I|iyVnXZI7+ppPAL+EN*XRZ6b*L@pj^PdDZ@dR!)tjFmn-5+_HEMfh#=bK4a7tNSky&_r! z&F-@`oLRwNtXIc8*?oXcTeH#6> z`wVV0-o;P*d7XEEi62N@w5T%hY~{VJ4CYVAZO?SQjN}qLqOe?!&RSoR;#cQXP`A^TVxc|xDRZ%5(1)tn2d1iZA zkNX2bf^&{l^evS*#@Z^BX_U_*yP}RU%Yo9Nz;QhDs87f-E5No|- zJbU*hUgKXvn~lDzdTV6B%Zp`m0%O_T3126adnhlEHh3Zk&yL2sMmD6YaUI4x7UFGJ zapu6MhrE%K8uXscVSH$drad+vcQA&K|>ao`Y2@wy||@Ljup*wT|9@ zsgOA!n$Ngz*X40km-A$%&gNS!WU1@rQ|PozUQGpXEIp#{|{?)vl|qAox8L8@Pk?Wwc)-3Wt&y}OOK@mv%H`4KW_fPPinX)@GHp? z+-X!1Xl+FLwyg}qiIbr>I4?rq(8GdhpKJx*(}sD48PfDNNiCjvmZd=b>ODb@jd*oe z#AU&p;~D(Wj4Xc0K{oSA!HH4XexKZUri6{24ZN;{T70v(Au{XSHD1!m2wd{~w4l&@ zsbHBzi=aV5lYd0~6`xvsM<6(sQx&o8XN8!qXJz0>KY#w*S=E;NCkw{u>GU~rqw>h! zaNga}LV-DG2)u*l3!3}%X{uI~Ik>W#xBRd^-|0|{AhfJqV5y$K7jU_JKef5M%f1=B zOPAt^-@5y}ZSz6|5$OSf-Z81_m`}EBcFIB8vu=Yx{hN^B^;tgusb0Q7qfV5!dckIa zeTxmhxaMHhkGC;At(su|j~ouV`3v~34_EVS1A2I!R;GMYX+y!S6Vg2Af3=*l3;k6( z3H^e5MxOZ6D#sUTyi>*awN%D; z*z??^qHwxDS77aZOHh>fK=3o-r9d-BTi|1QTCnDy4d19BN)QUM>aPsxs;xB4P)%d!-{-&1@3f35L+-<~5<-4=8S?bvwaJn>1TDFk8)C&Xti|y z^{d_dVZ+N+KL@w+GkeT=_dKVv-ruL${OCmj13x)IwtgeMpJuJnjJoDf)r{ei8 zW|#SUEaYid(+sj?!rGZtRwJSTqLL%Y2dSt>inpluYA+{ z=a}E;g_z{PD0;8<1$xt=dS;1aGUNU3KL5&fIYHz8HT>J`PV|D@K7Qt>X526JmG5&f zfVVDJPLSSwkw1}@$#?Oo2j8n^X(lyennI=Rf2xogXI%&oX1` zPakD=zsaY6D^F+s!%e)-z)d`>gC}X%C^Nyi9wN)`Ev;&J*gHJ)yld>C4e7%=kb7z5cKlePGY; zs+<5hx@kYVi*h}eAG!J$f9Dq)!NKXLcnywXI8hhHw>W3RSE?)L>+vK7dt0rTs*bz# zo^Sv0JmdrTfj3XkMjnIoA=4V>h_w*@DJ$V?$cR)otr26q_Q-QL`|skG$fa`Ao?PR~ z1`g78_WI1nJXXu6K;geu8(xXaOgb`4l_9Qr%qEPY4@-y%Qa+h5W+rBgmw!03Q13Qf zwP!Q))xnS6q@+(P9<%1(5kJp2lXK(y+Fsij#iwfJS($C$%jrgVEoE32&=prb;9nfLF8XhBRkeKsJvvf8H8*D--r2xRu71x*o2D{LM=Kcl zFjfBKlWDxG`%7?f%OmKL&O}KztBiIE=PI^abF<%bxu#w#82Kw6yf!fdCiuD-!%Td_ zgcpHK(1P2T+46>&v1%rF-GGkKmliNOTe-+6L}eTEqDg@jazcQtd#sEXOw8N@6NdRdj;ceUzmlLUSD@KbK`NGLbr?R>89)pGiX z&Rk|$ZYST+ZW_PGp^432h-b{U+kGk4!p7E!&FnQ zBFjf68u&6QeXqF2J1e=i4g0v?`<8PZSpO&YV?#{+ST$E-x2RF;DPg1bAu-Kcs(YZ!@C$QeD@zm=Q(E0eqA=0&dF#sUULE;Hw|2GM=< zdUWE=p|zEi`FT3gT#1r0ZlCTtt_CxmIWlU?RG+k^IaNx?F=E-AQZcT?bcNMYVZuU#z zDz1IRjeYW&yLtM3Ceyl#YuERMdnKLqOw*mpmEJ*dMP!fDmn`H2Q^E`dXM1Ys^c9hO zM?Yy=-9=OoCtN7l{F2pdTo@DV`r#-r@1Ef4n~2cfL$UOARZUdeY|qg7B1{*nO;mfC z&K!vNLKg+BE^=q+_d&Gd|jb{W`Dxz;k)IG)vGkR zv$}wH@!~pOSyLViDD+n(xvb!y_^!g!I>=^7EIS3Mmac-2%7)d>4`k@T_u_&-vo@gk zI8A%?ouDmdsxT(1alED3EBGd5_nDgX7=cc#2|xa73_oX?R`o9LEZWDq9pr8d@wJUW z;C=2Lf1#8suj`r@eYhZkp6ETt>b$sHWR3{Twy$9QlK1j6?CtpFTE6^@?mR4*9nCDg ztWC$bRr1?6_VVqE+G#zP8FaOc2cuO|MCYAIWZ&JX%oIIo!N}@EyvKjJ%$wgd9bA7I zAMRhw?NPkQ|6zVsu;8;0lbTwYXk8_K?YdyTP{1%fH+@_X zxWP-%6k~YCuq8lvTY3H+>Y46<&*{Ixfx_;+cK?eIycJyh2>)hM?O}BFC zFVsw|8!=_h_I#(?q}~%Z^HE+5_|n{s#R3i2qXK1pA;!4-GC%t7C&8p~H@@e?MS_gI zix^L(yG)jg3&XQN#biyI#&y~!!j(6C$gI$f610AB5|mTNnZJL-m>xaW3pZdn)AB8! z_PJ{>*s>{>v9p$xx|kpGT4qG)u`Ifpx!J%ar^ z+Tq3iPSE=K7PcEuILf!fFY>;qEE|l~RvDNRC=TVR&UIgR(A;BjY=| zNZW(G)beybXL8?Zs?yT|=5-f9L&Zn9Jbe{DUA_tTS{_BG!e1P!eSw$H5&sxT23O{S?lw8}xFCaX=k9~loNI9RWGk#l?jp(GjKQzz z9=xHAL8N>xGOOb7>g{k?Iim`G?u~vZBP>1e0S?wI zC9zX(Q8QQmr2ZQICDVQefurOr(qBId6uw!*JgpRHEzO1-?cD$ulrg2E0A}82fB%Fr z3Tw=vJYK4i?5-t5$Y%ky=kPfyV3jcWeDF4>V)PAhS|fwiS@|TXgMyxDJt*0-9W8^R zafj(C^e`;paLk&h2k+I$*?(qaW$_+ZJ$n)c2Ft*U=XtOSEtE1OmCIV-P5o2&GSURQLgs>wkqXRDl>+GxRZue{1GdW8!OiQt;LNpg zC>~yqQFi~Jd(%hQ;r9oOBJI)g?Gm)A>W85*7qF{a4MMpKAncPhmepC{C-xi#p3Xzr znheyl2u9n)9+GYoN7h@d;?Q~@Kw@1Alw>DC>*XHut`H?FnKR93Y|wEyJ$%7wwX%5;tE5S>o2lILeedp zrEoRIJ{Q8j60Pt~Ck^)1IZ^2~L6q*wVsdoo2R-M82(z(Af=PMc0m@7w?ENADYXw8J z&D?{tjm}{Hxny#oZ3^M0l~Nyn7=d_T3Y_?O8%}myr#_a;khW8^sa)roc*8&**XY#2 zQ-+<%eu<&h-)tC2NG2ak&ywd_->KO~+wgJHCY&F*32i@|0WG8D5MS&EldadnNbn{& zykRcL9Y`S^OR}6LC0tKOK!DWAII#D@ykkV9lw!_^LGvPE6B4u^UmCHINThRdJv`?hfBs zc$fdR9=ujwg0RACm>(!g$JZ~U2md^PhBisuvSJPXl`_V}Y5VcCwI5P}uORq}6uw+B z6We^evHqYnP8)N;it0sVWnLF`bgvzC-`yN9ezZh+!CTOH*g^)XL&!T%RkHev8aO|g z0?Q7Y03$pGoqI+h)wTvs8E8Va#%lPRz8RLDwj|RXHh}Tj8R$|r0%truaPG$@y#Ln$ zG`1fA-GUZ4E*J-E-ew%DD91O!u4p0^g0$uqd~WoW5_=a&ifjAGhbkWVk?93l%Ws3y z^-j*WpNdqxawz4ZwjIp&?tzH#+mLy83veCFK=yPO99CIN%5UB!?+%!Pnsgua)$bgM zo7hN-0|()~mnPc_+2eAfJ9uni0p`yX;HNcmDqBA)>WMCfJg zZZmJA6>Y5+31VrHu%z)OJd9An2D|B4AGi`-Mdx5ei#ocjRzU?hHF&T@4=(=uK|Z`M z0?Sinu;P&rR@t|az@J~qD?fSI7Wn}Du4<#A=vMsfF99aI-jc;vn~BWsgCN9x480LQ zVQX{?X=7UyYk@U7g>CoCW6Z~WYAx05HdgB zB4v7gWS~M5w3p5Y!Np*ZX^sGu7s9Y>HUoabA#mr+5K)!TheF4rAaTI~mh@6EV^tQL zxtd4SwNFrzdqv5zyy+NrGYEs@nvruM1aw7)A=J_kOU|6*^e^%v-f4C~dG7*6rxh^o zgaM?gcyanxrBHJYhZB8_g1qt^7?}D9a!Dnan_q#zGuL4^JS38ZC%`0A3Qb-J$hmbZ zz(Z>QcDgL*MA9;(YjGwUQkhGfrYOVJ&9+dHAVgj0{zSypec{1V8?t_%2N5ehNZMnM z~=h(hI|!?XR9@_>%C8k2Bv_W-B}2HnFqi2JpqSD$}ppTHbmZC z3H$9t(d@}6+#LA{A+H{BOnX%!xOWg#pqDe6hd@*%B2L<4At(OQJVd9;PQoQzD#1 z>fr=ZH7;$$N6P{-rp3UsCP&z^Ba|?oe5e_jg(z*Afm4#sq68HKlNyqs?fP!m5}%KA z`H{Hn-ew&5oJC%hh7uu})kOWue2kZ2>*ekz)F<7G27W3d`gPIW2ER;|=)lF&^vzuo zbPtx}ZIgTqH9e1h%~!}xA7OG=Y=Y`@+)jiyogteQ705KOW2*S9zE8rEkzUWmP3jo{ zO5)HMs(^7)(=e=56v0#heY>v0`b2eD7$Xe|F$zlDU+2ED-gs_Qzfjc)h8qDkW<+D~l@j7K;T zkD0Yp;Ms9fdvG5qI2}!O^#8+1q1p86NlWRi`Y~ACa}i&R-NYFCP|a8#rKc{y z%5($lWB(7!Td|lWu*Qa-r!by-zE~luH_utjX_{9{s5_KDja0WX1EXRNj zKn0&gxHL=;p9X({H%9G1MA`ecAeI=(ij!Mv&1@!yBqkVrg>BBOi2Dmun6aPrr8B8O>i*H3Dj5o zA&kENA&p||t3c_e4z{gTK$Yk=SS~@qrD$uIv}hw- zUV8?jW6p!CTM?|d)I~mPI1x@(IHfZs9tw{=gU&b+96vP&4ovMPPb+RgX^al`f;zgl zOu)wBGnl<)C+2GHz{}U;;i_91%^pAoaZJ(ZH%ti! zVDutjYsh1meoq;zw$8xa+jTH~c@vy}`WpO~_rOZYHL%aS3R-V{gX5=`z*67aMCZeI zk}+p8R=xQOH=P$_kjx=?d}R@+{F@FR{Z(*(s1AOeH4`KKq;QAYeMqpl3oq6zfa{GW zke_J-I?-#vz3L9|LqzdWw*{_njKF}$=kdWmIXwGC8Etet@%h9mpmwf?%|pgelobv_ z&qG0NSPW0>VbHT!3e$tH!@%otIAuHqC5IMc{mtb>LQ@T7x4J=1eL9Ria|H{hmC!%O z9rRD?f%V!E5~rYoo-?Q8MB{uM8c@Z%yV?A&6DHV|7>R$Q7+fxL2i~8K!pj>f;D?bc zobzc0-)ci#axxS}*S{oDe|N$k;nQ%`+>MsWbfo9h&Zj+h2&h>(OCY#C4i=7?bZRZ@5BLEP-h26&GgOx;GjM%o0F(Q8IMptw5Ky#rQWZ}ZeREwt;7)OkP z=hH!Q{NGNH66t{Qpx2Q2FB9q(ZilIX|B;5%p9#zzCdaz};P;v4^f~nadSl>Q2)Z`_ z2YORLr{N1_RlADxeq0Jq+!jIM_Ej*m+6!vig|KjKFFdy|gR0KCgwgXP2D<~vpQP7N zVJCaxmNe26C4?hB}QD$~o%=i9KKhuaA`Dv&JY4d9ROS z?-RjvuMZruRD^TFH{evz74Wihfzjh$DnR7awocVJdR_fJ)y^kDb>{^HZCYw-Y`Br2M8#()R{*qb4MzaW^eu+CMr|E+|wmTRQ)@-N}% zm5G>bo`jB_+Jv?y70f+c28OKw zi)BP%3NMJPH0RUfpKEEp@dvufHx;OYUdVYkAOEZwAfY8KWVm2}oLaLSg}2CIU%D_Z z7rPDN^#{PqW*LOm`jeRXK$VWhQ0b2xaPNZ?Y(6oorG7dL9ml`H3xgt%HPgabRqR@F1jR-L0RSmh6Qxt_5VsRA;2E@E_K3~WKaA$H;ZJ= zze`Mxap2eS01_j+7mPywLiW!za*{THtk0pa__rYpDtD5O^d5(w)D#T)+y{^A@4^Ac)3~*F6|QZZjBh6g!jYHDpeD>3 zl(RH()M^a=d$0wHo*W{aErn!ZXdl^>bQFX}TVd77KX6v+2=qvuhrR0}K;N?rK2?T- z;lufGlIsgs`!&I*eib;S1>ocv?0lPOf_ky95Z4xC*u)V$ldA}4M7|K8{sv-oa2AR< z$)TN!JvhlfKr64S*dlU~^%fVV`MMwR%g<`O(Bp)oLMFI!;Q~B#QH(sg^p9Fvq(l}@ z>w|RP91xi2z^Tk=yp?T*QF*T*^VtmKEgXQ{MbR+XTa!#|t|WNPpXB+<5GCuCBrHh` z-1i?wsmc)4U~$aW!8X|IaR+|<%YtoXn^|4)4cU;Ia+ zC3TIMBz2L~u`fx*l_#WMDwL{xsZB*Redbt&TVQh6KZs1og>~MlP*Li^KDV7vs(usF zscUdqJ_@!?Xu{;Px5(u6n~23yc6JGVPI8TtNuP)_=Y;q(PBZ@i^`XB69W_dDZsld% z^;89wjt+roSQBLKT?s^f8SG~9tmg@F@L1sw1L>H3UHJ?1wB4Huy%(Lt{*kUJ5$ZEG)xi?Y^sL2@r7_%BpEgon!?$W z;&8t9GZ8wr5H>t|35t))VcaMb4Ek+wbg?Dc$8Sa5H_Bif{F)Tr9U~4_ny{(pENq;3 z3I;PoV8c>vI6R^VpP1iZ*Di(u&R@XlV-2MJ-U#9OnvmxB3kS{mG5%^ZG6sWGsFFMx z)J&kJ-@eBAaX*5yaa3h&Egk^7r&uYt0v7+P@pGQPvD(kJsdgz9A115gQ~f@ zP-43ZCj2enh1?Z<_9_O;e=WzHVIg8pw^Act6{u_zBkC1pPagOR!^tXn2#S;fmhXnh zLmTnt&{iBDvOq11dGIj92^3snVC`iq*m!6MJMX8%yFDe?vWw+c?DH{ObUyuer8GUy zN`&5*Y)Qw-E~H`oRJ!X_8}4nrgt`^Ms3Ic-sy4Ssz=d$qIryG)N6Ot`QPwp|Ejf)O z4n{)2+TUQ78_f}Cao85WaH8Pgh=E%qao2wiaB{_NxE+)acOC9Q%8>wgo7e$6gQsAC zOEK8Z^TxZ{@#u9e0e{OFVC?KC&|0NO?-7-z&!rULsc>P=sazdWP;nMC7MDRSokz@q z?F>(vK7;N&AMh7C2wxdt3{)uxuxlW>#2xn(Dr1juC5*q{ftUONk0}Y^P*fYyuh0S) z_C7L*{ELfqr_nRn9W1S?7}pQRXvSdKu~$E)~- zD#2e?Yw@4tC|Po=pUQu^)i9nTMHxxlA+J@MfcfGCU2M;?)3^kz3#VXGQYG~9Tv^z%I>X!WoInNdFK+9xIpM4LS z4{2iKpcj@V4-;SYTzKj<6${wDL}s-vjO`2p>(zg#JDyj`C*N7H+;1~HpS2Xz29xpE zBAWawI!8RWFCeCyHly{Qe(;!5PIeqpr=qeVsguWQDHY{aL`x!`j5zNl(=X-2gC~)& z^!x+~u;mle>QG`mt&56%@QZ{R`-AtsIvD9Frozot;or>%5FPh`lVReQ$&pfkwGqAG?fnx<1Af7gx+CQ7CR-2`PQ}9= z31C~f0bcRDiHGbq_%;v&IkV2f-!ly4MB9L6o&Yq~si5(XX}D*EgVjUwFn+-RiV~wB z`p8N&x~zbymZ2~%qKA==yKqyEKW@(3K&~VYP!GZsp(@xJ`{cKw9P2Z4d{P3e^c; z72q`|69e7>oEotb&H5K$u6;W!Z+`)6#@IVyOF!&<9s-|c?*UN{9>>8bkRu+V%<*~s z4pKk;1jXO-XdYAzA8y)!U+oQYGhP%8ZDXLdeSow*jV4EbT0p~>I7q!(M$17jT{P(r zUDt3I9&Pu62!%H!{ICl=@b!iI^sR8gyo%C~6Js%-6e3x~&W;CmK`A*59Y)kSd3<)-l8{nz;+lFRoCIdj9A=cla2GU{c-!_ zT}YX)z`13&NL_RYscpGH-e|4E(zH`}JwG1rhBS1ORq>_oEc$93xWPC3qrr-uz6 zoeQBdN1jkZ^EjlpZwPu=Y-~ZV7!to>vgV2d6t#LoM~@-YIp;%GM-HUgB*O}oi=cNf z8j=eigTMB1{B@=Q>ts2oFZ~?Xelno_>=VIxN;#Z({s^?b$H0Q?>p@aS27Y+d!jI(l z5N)W5o5E7?Z+I3OMxMhLNgp7`{4UI1lMYL&wvvByUQ@-gHdI*sVlpYX_e=2`rrdAN(s!hRoEt;4$+peyIPB6SX7g z{zwPg=I08?-jGG~pGrZ&We+T_F~Xp^;#lJ`Kr%OJLip#I@Y3iQw6r{i)@PIO-D6|2 zZ*(bH_VN>zsv?97T++bwyaQ|29MJlI2AKhWg9=~`m7V+wKFdysn4sLOQT)n#A#`dyiQ7`W(A4QLs&t0H#kDD*+c^l^mpmpM2Et_21CPF

=A5*RIh4l(-qyT@XZUD1)r@?~lRrd9n zU<@On@P?>3gg7rrR&c&vt5;_GqNH$CiHA&lXl zCg5416NZ>u;uX_HINHQwyTTVC_k}vB_MIUd5m6$!q=TySzei36{Uj=Lv|yK+fb5YM zBfGyQQF@mi5r_0Eu(CoHYmq=-iv~N3Zh#9P+o`NJLr~6Wg>@Iz!7M15IQqLn@9f3k zsW}a=uRDVc%aY(s!eLz4D@BXl=TqyQUs7KtvdEizQm8h65l+=X+|Fd-XvPlY8mz_> z^>^@2NCNIZl!*sG0*|gagGIhgSen!eYt=OH;GzJ`ejZAF?=2)f3ew;t98Eg@RKUOD zFmzZX5BqH=!+dHztk_gSvdiS)=8+p9`*=NNRZ>9Rn);H2-<|;%hr^*Ibq4Mf7lDPx z=Rl>;NwB{@pVGb{%89#}K&9UmBF)2TWPoo7ODcZC_Tt6p{c;`7>a+x{sBrN0dj%7J zLooZL2fDA@g1a{EC$`oWpqSVNAJ!?u$FKV!@beciz9&M7{cPmK7^hJU*`jc0>omxh zI|E~mHUR8%zt&I%ay_}kU`!Ki#PdMm${R59R>t!a>rkky4eVR0a?x^Y#Simazze-kHSQDLN~X&05VNS5fl&LK~CI>0@-en@3C{x(X=FmtLKoKnnz z%bVVkts+2l=SNc7^X#a}&daEPpF!lhXDpRuokZE?_fch`&D2&qN3!V9C-RObiIdY$ z;r-Z7YTNj|vwRzCasWz)?O@%wE*j<-Vy448*swhvlYcm%eOfNm*YmNZJ|3Hl)zDhU z3I}cS(Y<;UC6lg!;+_^F?HfSG8}?vXdkWT8G-2R8Cp?)VfjL!)V6$Kyu9K0$C;HJK zRDKTgb39NrOA1%5cYvk4yukfEpUnL}6&*$%K*>HA5FLCE8@GRlGalLSur~v3v_7HE zpDDE6wbw99?g{93WPtBTIJ%fc;M6{STzz2;(c2|Q*iwx0xSD|09D9%VU0Z)V7#3E6Y)gELL-IYy5ySakil<|t5n<~W2Vlmy#XAtMfj>C!< ziY(`{9XD-1gFnn8A?tw^EbMgwnz>1sVs-d1*@9G^Xyv$n+zCl7ZSeSz z6b2j(gKVak*t@xr_~rR9l4b&$=hu>DhPPmu*+tl!z|loCvOdM}XhD>Jd^tub<6o8jC!0;2FOfDHaxM}E~fqRqnjI3_#|_g>wE z?YmE-0rv_@`MxIV)`8IMppHu&HbZ%eAN>4o3hrK~pyV$Pj%@e|E>Ybi->Qc2hLp*R zXQC*UDvON=8^O4g&Ctuw#CH;3u%V?7s>-{;Yg88x|5AbND=t9t-g=nEf?X@dYGJh9 z2-WXg#Pp0C=<_2B4_&c``W1a-m1r^9%yKQ{wl|9>t|Iwns_=C2Y{=9cB!h|Z=(f}y z4~mb$jBzi#krs?vk{hw5ng$V{>u_MB1g`HbhBw3J*gE4r{*Czr(}NPwaNA7!zgwZ8 zbM!h}uTP+@Xa)>D8-???RMFuW+gB>rl5Jk^IY&ffNZ^zdBDzZr9v)SPqhtqIUUr7a z*k1j(IiCzS-Cj#QG%P1er-cx2?iwPt=n^SB^_hHY5CwxP|B>ibe@MRD1(?ir;h20| z!Wr3M#Q6(qcInxzk0IdYKBjbi);v-EM>Vin{2wAp|pgJBa1?BC>fn zpPH+=9HMHbLhRe!u-mniBWkvkjD{DG&xP8gS;~~;9(+hO4GmLyNgt^f`C_DgP#n(A zS^!-I%Sly!yg|z5zXlJY+NsY4nw-jqHk^428!7!hOUU9!fFNlrB+j9U`6jvOp)Z2XNFUT{#~3|ce1F3vWjG=$%nIwFHAgo^I$}@ zf}H&Gi8ReofEmxN;ijz(2sKQHSKA9Y8p%bJ=CyVb_CAWbOo@>S$zk%7w&qx*cT(>s zJHfwDMZ7GggTduLKyOEeVI}^@K~XbG&guyXn>-164wisY%n&6mq)66J5eDnO${Y)W zb(HX#bm}3iqv#e)f|sO`tS(jM45sx^@UWYlEn<0?`t7i$*8wiQzd-h;MG&pZ)8zJp z2y)|73%TPg1=n81QnP~tNiB=3e|Ti9wph8PFC*TN96e?q_RF36r$^>#CS$BQ&{W*+AO=Ob462Gl}fg zora#a!k9Jb2h4w#2lpP#2Rj)F2owEC^bIMv@=61qsIs&4J8dwzw}@QIcujt*P6Icg zA;?}WVErkj=;LOFv?9lV9vIF5#pW@Rx!@0H$-RTe0Z8#9*DuunP+CXBt1&($N z!>3!)c;ky0o`c`epU@4~Go;Zsdn>L=<6%(qKkOboN)%7!QeyoDRCEXjgA|j|EBh+T z`LjC7l3X(R=of18Un%st(FQx^>cB1E5d>D*(EsHYd^=f3{D-aBTG{}9OM-~fDm@bM z@-yW;^A9!pTAmy{;XrNwlLB|wU4Xc>BIu4r4S1u z2juY1+XxiRwgVI0CbHXD8n$_5fsR2CG(4IKr`Pq7f9n^M*$qR)@`erUd!qoJL7U*^ ztYV@a7EIJigNWbj6hgo4Ams&G2p(R`11Tvl5YfH@179^zYZ0q;zDPmNnIo`h@_x{_35I>nj__vT zG0w!H0#4HyMZGpVL|*b=l7gSQuv-2*SQg!Z9M3~ABjOU6PMVFxU=vd) ziTehP9rC#NMX#Z^{Qr_x{0*}^C#d!v9^`haH{AYD7=AApCgcEK9i_cTaa5z*)%Qd09G$10*PK4FzZ7Ia(TmC_S z$R=#VIkSB$AFX#my~(TqqG$hI1@qdh>GSX677`)`=m=?rY3<-mm+d{I0E;z z`yv&$0Z&A*_t6e6In!A|^$A#QT&q8LUFn0>57pr(_aOMJ;elmnB0lPvi?hpSL0RiR zn3Na}u3w*$F&}?+U-$qVZf*zt3K!UV>;g>x{g>6pey4lx*U&d*BIvOv1RQ*K5+ll< z3LW|hHecVu&=S_CHAEVEw^WfWBGX|=e-{)!3WBEfJQy&nrGD(4OqBO@u{_!eNH^XA z`@&-Yg+^iXYALKp4M0aa6{1#GLT7p~Y^%#8vd@l?qT97ZO+E;`@|Q!>xkzwr-vr$r zH0&PY!Bn}qY`vGqur@7xr?n1RMUF$oV+mNjV3dk%TR>6|su5)tBO@UK9IBC_ZT_yr znJgBUQyh(2i39M@ryg#65yy~~L7*(lCkM`G!kquEvK+u=pkM2P&pj^Dj8h=?iPy=8 zeWyvRY%Ph4TusCeAH%Sj-nhHb6cxPeaAs2rm}uEU=*R$kKIRK9FS@8Os~p(U9tB5V zQ1CBV9i5X?K*)13bZ=cqTrEtAR&ypX6U&7!$GYHxdp!25o~4GAEU5O$k2x{r(@<<8 z2co|V!F-Kg*fje*EDpa6`>hj!SKbS+SAK)M!Wtm$IgoPR8BVkf;`W82w87p_7&a_I zOMZ}~-!;h5ISTjS$GKEEQ*Q+w?=>O%K^7ctaK?d$(WtDn4NuNFjLULmG3~`q=(&>t zAsPf$73>E2;7;)6d?U(JONg#~ERmV96KB139aML!L>DZ@H9adLaG)JoAG3_rSlE?YY$?xx&*z?yMPqeMG@~6BJBOENK5{j zODjJ%rEeTROKrY=gHj8yAo}c__us&I;-KjR(oH;adusTB~C8ogp1@7?k zp9v;qXyY?saeSM15(2)gg>=<ppbmE4hu>YLF|B8wiu<&|x(Quhwax z-RUpb@$MFy?a4q1kLeihUkC${J7M0zGU6;I1}~JQz+xl~_LQ;rt38X6XgR=q4-wcF z#m>BgNET|0QCWh1>g{4}yfEmByXUgF=V&Hg{aT4dEdQzGDvaqli}7&!EX+((Mc33{ zP+0j6PK=!c#h@M-^)te_#x>a2Itxt;enZiH8L)PkNrXJcsF|l_@U+tsJe_Wh2~rv8 ztDl7KYopNMPb4Zdm*Rn*`#5lv#?oBCX=0{$^IkMgjIe(BlM^r#ZjsHN|55SIB2;H> zEu3z?4Q5||LFlrGSh|t}3lj*=m6mGOC#!WhGo@&=EP2U(I7-xGEy3!dtT?0L`G&&Mp8s1 zvQr|WrKLgI38g{E%t&QLA!TGHGV-+&l~Ka){{HBn&*RY_bwBss^FHVKdLCZAb3ISJ z6VAouHLP^*c z^I93{!#h2so2*=s^+yMS_CI=E|h$A>v z<`i8%x`sAX%%x}g$1r2n4cJ6#vb@|247(`D`;3hF)LutEf6Rb$gnWd(nMA>j9l~3K z4Z@_rVbTK?VbXtD7p47281R^k!F)_hgWKf2P(NiZ8lQX=+1@WmJ6MII-|t}CK7W*s zDaVKXbucfD7qoo_3p>WYw9NWx#V-_m`1qe-*4P-1^y#gn;rWDG#2iN0ry^t|9ffVF z5#CRlfgLV>a9qEL=e$|Ty$Vuz*Vj(0vHFYY7M*x!hxeq%3F&SQ1<*1eB2jnOcgIEGg*6jQW$f!z1qN%UH!%ykj&V*Vfwx8J@1 ziF`oKH!URI&x3!mE`NDyk5kFrFemZ~&0JlMP>bt`x*CN;wj#5z-<5yLCG*lR!F2t} zC}g}ijLT}dRI*?kB437K-C+;0u33Z`ZzmwpSBGC259e)f9J#EdjViwGr(-itQ0Dj< z{6ebDcwpIHv8!4!YPRIK83s2dQY23Q7KO8!bOG85!PZpfFK`Z*QE-As!bcy^VvYq1_tse@1ci?F9HH29Zm$$iUWs^-X|4|+- zdUOhJR!-($s~u?i(XEonnT0}K`*Cj3J<2Vz`&n<60!M$DjaJoX^!cYXo9kHObeB{i z&d!lf6raWGwLKwwzXU(;+{M%}`!R1|9xqU@=kaB=+?wivJL^lyp)5r(pZ}Ks`=>13 zkpGvb^%36hPaTnXtMvV770TsT+e#7lZBOG7*2q~Gs#EaW!as(r=j5?6WO@}s@3Y4xTbu5qa?=v_ z-w@5IeN+)q_K`+f0|gqta3)WDCXQQ%zpf^5|2z>Bv%~O0bik|qxXf3~(pjOr1m9m> zrqpTOC?UI)!dtu19;F0IvlV+yBYI$1doI2CGT2fYaMNP%raVbs=U8t1KA4M5O5im| zoLR3M!Q<|^a^=M|9B6Umm*FS5bi)>sx(~*!8!0&dYQ4lu^%M=@T40mdqcV8}+p&|8 zJS&JF`=p|j`6QVrAa zSKO{4FWskUAgvEIod?e3%FPd8sCNJxHY4ieM#ys;1{~1A?s;Z-Hf)SIw;B#bk8xaA1*?0ZV5{T8#p>z%Qrcm#T0 zyGgdvdK#UONF%l<()=2+Z_s82e+^FrsRW7dV0YebqREnd7x8!fU1TU1K#VD&VT3Er zpB;vrybDx4F_a2>-=Y(De{+V{bN0S)n~wx6l*UhSkcNJ+m;M)C%WuD@bM80?{_@@t zD}8n7)}6z`H9?h~<=x;>STFi;CV!E6;NkK)Yq~%7e4)gx3cFzpZ7vlQ)G)8&q^YdzT(63KG^LN zOTBkJ<^y>b`E>G0Hkfb-OI^0$lHpRUwNsI*HX2D+sJKa&ZC^}P7vEC;i=84j=87YU z(@6f}AHh|7Fl-K)QQ?CFG-Pxktyhc^``uTBjWxI6J#ZpaGwus}^FnBsS}$6hAV)PU zNTcqJlrGvaKq~XV8?lvtX^xKqiLq-a@85#6pPrGh#EXXdmr&azeYhkuJIpiSNjH9o zHC2CpDf&S_D^17KW7E*9QWaHQyP-wX6h0#sqs!kG=neiTP1k)bwRL}0B%a_&E`HDskhL`rh z=-|H;`SKMF-!>4sUF6Z)HVfwZb--l56KXBL<;P2s-)qwn#p|@K zL>o>wlc64dmfV~M2{KP#qEnoIyUP}%;$SOOG!8@U)H;lMn2yo+c0%q=G_nnpx%b+> zY_2Ea=h}5_SoMaBzue{tz5VG>_#r{pzqfEqbYdNQD$aku?8AbbKjfwGfYvL0rWNvH zO&F_%tl!6JPiH+)TL2z03<+}v;p~^85C$*93fB$vs-=ZaoUuWmeiHZmcakHYe&YhO zN>-bDmQ^3VVb>$oSf?3-F4xE6vv(OO}ypm{mvCdfn6yC?a)yeFyb>Plw2YK4jx1v>HAmfy{Y<>YH2>@{AjA72T) z;g1#H?YW0z-%aB|UsCX@V;o0av%~NWmnBoy?4XaU%Ta9W<15HF_|OUU`H=0fhexkcEH2Pxm4UmFo#BN9oTWk4bJ_T<4f_{A zf%&&&;nO1ziT5%E$tBC(!q$Q^syF@)!N!gDH9w{aQ7MRc_7!V3-Jym(5tjaQLL@ak zHcM{V+6%4AF0=X1UefY2TLCD)8q2|8-(UiK~RIYV{MAI$8SKgwti6`l` z_&pzV{0#SA@4n{Y36Hs>`eik^ln=v2OB@Bfm;=lwW#ANP<4Pu$Ixog%B}R)uX7 zJ|In3i51mOVt%MBPZ8_l=d1j%b@g(jCRyOyu0^|%@e105)<>Z#;xe@o{*ela9}Z zJF(&1S~Pg6^3(ar+|@*nRE`JrGx4*}5vG;l;TC_2CM!x0q~p7^Y6ObmhFry}ikm_G(`v4LD$p zqvuY*q2w4=IHcq6+&AbqL7tyY{EOP@vOKKcdlbH$g@PYjF;>j9eevv$45c;n*yWp0 zAfErjpI#EC9&V?G+bP()?;1RIm*H2_QQqvcl(oBb5zj9k2$TUZ(E?-c_CT6@9_s!) zMz@*wv3~tAr0=f7-`}Uuw(KSLoZXLA>Xxjo*^2`=e}?153vexz6?++@>En%4ED>v& z#`oK}P-MSs<}JaqyzfxTkE5=BUa5&eoHz&W&Zc(c%1WW-s0>7YmED3>=_$SoZ4#GVOx+OM?xoyPU%DDN*c{XT{T| zbwX?HGM-*yC337yn3JXHq z{2L{mUpGa1r}`l&PfNn%`NR3Acy@@kD5S6d$>L=A1UmdXms(s#;$vOCFx@4bZr5EE zG%UrSx6d{~=1GRoc+H(2?3hV|U5u&hyAvz9+(N%A=J+y6Px5|$x+LPq2w`Kt+d|2= zc){lGZ=vE(8a?lJg=|yhut>cXtFK(ft>L~ny!eSQZqs!^XP-6c6nrN2xAG!q?}1ae z3WM@^#6|4IRK+UkhsvYUb$@)N!aEJp=&%+psQH^Y)ok=6IAYU>Cg$wZy}D-La-&EY3VrLA0_1)~P%|$jn!eO?-}P zhpu5y>nqwZF^&3Er<3pMRGgh=hHisX>1U`rO#0|U(ISSj{&k|j>H*}`GDN25JQlYn zz`|}M_B8t7zeZJbJl#c|k2mw-j8rl2c$asN{|&e%Akbqq9zL+=$xpiRm#rdm@=Qv$ zQTf7;!yARcw|?Wp)rknTX%^mh@uP=h#2MYLV#-tZwT!qHWVy2@SaNKWKObz>1fC zq$k6B;mFHEEPfS$ZlWWi_l+LXgT^;_{CzQlD_tbLc27e({A3pIjULJwJ+5Q>yA>!h zcBU$=JbJQ4WOi+SNXnJtB|h!R!qC6ndDQ|5kL>ylv0m#WL${bw#Q+Kdm*X4^rUI^0UTP;iiL|h>F_yIt~K%Js>7nkV#IB(kiCu)r^iBxX*6EM z&k>yG7g^lEIohDA$s0!<uhBslt z)J=MD@HG=&MW43ps5L>7?GeHJcA8{t&msJ^y9V2h9E3K<9=!BKPp*$P5ObF&DEONR z_6`0>Z$@jdUeZ!-dF9UaNB1B=_(f}~P035Wl83Zz;%|GWLw-?<<*1z5ma`r$mt=W@ zc8&B$O|(4U{N)Mf-EyothxpxBI~+N#g&|woXjAiH$-E?Q;b3|YZ8`FUpN*R@z4D<@ znpL%cvtOIDeC#w#GM`PacAC(y=Z7dUIUMV+il1quQK#E36yME*Y<&!JpO;D0<+X)# z*;c}Tljou)UCgUTP2(vFgJ`G!DSGN@56`da?C>-QN@){F?|uvZZ2E<^K|e6rR#m!7 zWjC*Rkc^LKKa$SY$-;hF8S*?-B#h{-FUY3r2y2Qmgj0(g$Rh1Db+^m0)Y@wz`1wib z!`umU-oTS|f*NS+%V+Qr9YL826WClmjjpfELBlW}bEcp&!jm+0 z#oVlK1x7u~M64u=@|&Mit;q8iR}P_BJ=Me+*&v*pdm7J^zM)H(D;Tw34?2Ohg1yH) zM6IvK;I?FXe|ik!SB`?5kt_QsC-JaNA$(y$B3ge8q8%#zErSNxkk8+A$%^g=C3&Ln zBD0XQ*fW_B-qUU zEz}eqC)F22VOv**IX8X_ZEab?6q7_uzxhF!X4xrZNeczxMHr5|UY3+pPmrt$9*;1q zM0BlugZu+J%nJ*ZSm3;r%w5pFNx6^>hZjp}6nmaD}k zDWkDPPjr!NnTEh|ukh^q2zES{!{<$8=;)_U^z%$5N!MJbmm@7O!8Jns&O}mOS~#{p z|A>e`U8%umoy2fYxa6{uDv$2!g|biW;sTO4q~`gScAD1F__b>U%@$Xo`0Fs*Kd?j4 z=+h|~v&vAAX{o1oN8bwNCH)1ZwSMn(~Qt& zGEUdVpa&K-ySF37iAkQN*Kb(vHx3qD_k1SPkao%P>PjK<*e6n&+J=pj*P`>$K%6sG z5clpnV3$J>!bSeL@bNh)xG1si7F9l4{Rl$>#q-XsQ^Muf$A!wep49)~7V_$|pImdh zL80KfuwmX3+Uih6>L%Awq4*Pix4Uq#$xli&dr#TREwIvQ3O!ZLhKr6nTb6jCSAse; z&MG3Tt_T&)V%_7poLtV4&qcJr!H0YjM$^nm z>4MFux0bFWJlSNsCp$+@W;@YM)F&|sGEWl_<6_8e8^7X;TNZYUvh{g&G4%QD2eO=| zz+GGW@a05B_CvMh;%VP4{cnAh{LD-d{&f3F&z|nWxW5x{H&O?GyLLrzpvWzHzQ*d1 zN(}F{g(`=H(&p)I)a{@HGQY>*kYNbIV(W##RSiO(!$q37%aq5}IdfX??R+QmF}tqr z;Fk^Er2&dc+)*I19+!Lbxcnm&E}IP@CzV-gCLNJB z9%8T(GNaT`KI$3byH9}K^!>um-G7AvZmR|NVf&if^Bq_4>SbM9hRS8wU& zN^|Mg;s;!B_#gRqE5bE*v6j5K1V*R&;Y!9r`gtJ*i`S+jIo=#a_x(6JG?)K8*~h*U z@+EsdF0!-R%Bm@AMG9 zI(`*SRNav-9p1(2tG<#6fsREr*@gN(UD(YJ_lIt@^@QD8XIK?YN zK*v#`v#76NFgb_T$7D)uwQ?I_?eK6ug269MaD#Et}vuV`<4o+So z=7-{W*^5NJ=(dZ~AI7tm`+kb);@w)b%%t8=?fR9ou-=O&UE*yv(V|UpPW~1BS7A>%JE^zS zH_4rSasoL^D$Z6^~R zQJg7k+8o0#JI-*?+#meO>7!&&c`spx|2fNl+efhe3v({iC04ZZ!pEV>)N)RXLY_^< z`y3^-KiNg2oEGxOi0{}jCpoXwp^RMI`50cnjo6NNi+u8U) zCP&^+{pUL7Y{85Ci;R?*Dz)ZOP&TeqLmd*w1#109~? z=hZ^={1S)wq(J^~A%~x)SAWv@Jw*X@#LUrWn$0dffM$BIn5e!q|B+ z_!!#+oTV(@PD_fE z_>6TXE+lE=P0mG7ldh;ILL(}l`c0K3rLvtW8y30lnw z@T!WUZ3)+e&nwF;6<=9ccFpjT4AHdYSPbB2Ss##R7ZPcBDFeU9iEHD$#Q zL0nz0iT2`3@|?Pgrk~t}$Lq3S5a@xN;0uD;+a`(8#9l(*!&?wm-bO7?V}yP`ym|4p zFrF?xGuo>ZIOeLr1}0lM(!3uc-#nq?l>;a)ZVrYdm6F|vT8XN08}`K>guU`O^!&FA zBa2Q@V^2NW5N;&;>B=Q1rk#`2{LK++hUZJ}Bn-CfuGbZntNx>?!KHLrEl=E${ZWXV zk|5{@e4-4iu`pdU8oqDuV&DLg^N)8&e~VT;Nx27AvELS9qK9KUFT!x)DDE*^WFfCw zvf{cv?CE@7I5Ody;P=>!B!QO9J>~eO$5(ituf~fr;y%qqmvL?66y7&x7T;I(Wb3)p z{_jjB-J65q75E-k##e&u&ti>U4y`cgi{(k*C}4084zX*6hT1-KUn7sYjx;L2{GP&w zKPJ_W3*kF%9ilf^^O2ewzW1YwN0j#CYvGD)8n43ouac!Ytzl9V`R$^oSc!dyiMg-4 z&1gFGk$R6B4MFW9QgeI+&E^cM|8tpE_O-%&V{`KURV^r1h~I6eeqs;Xp8l!J;MR0y zXbn|F_Vpz88M272SB~bv3H?cLf`p7?V#xEWJa_R4#XIGHwCbiioto}Qdlm)L?=jJc zc_zB$-e)7R{2{*lw;wZ#HbKz+NB=GSLe;NsP>PH%O1FifZ{IDjsTyo`JE_0bm)-JK z6YVC_`xsx!s|lye8;nRx%n2PBrH_UaW=Og@T;zJYVCCRO9Epp;s=c#dA?9aF`{dIM zuSAOQTgb0}Snx(`1-9RH3|rUMAZ3}zk5&)k@C0M7`ex0-J<-Kg^Z+|AkH^r@cd>Q+ z26XoAgF&7Vi1VBe?@K#yX~QMl(9ae5$XF~`EaqRhj7;XEcwj_aL1;a@K&cKF6D2~uSp zJ6S$7_5vP_%RzF~C|+)=%NNX>u}Jg?)!Uz-@HgpXFy#yt-q{UrpJqgk68}EqNiZE} z2-7XhD&N)lxLsFPHP{P<(Cer#YZ3R!#vtZEDxPkSK}f|lY!UMq(?%qSv%<$5^XdxA zG@j%`>K0P-54zGli9MxDp0x{Ke7jI5-U$g4*GYb5U#6fa9n5VH!HSwtM9XY~dT|_f ziG1#ojx*@zP$!IhDMzPbSy1mc0t15s(K3D+(&xtDieWZv<2o>B$1<#QT!z>{XFN(c zhaKw*aQa6aa$GZMS?E{M1-g-%G%sN7qJ608$i=qS43SgK#OKb#DAb)vr*o6ZI60Pd z`rY6IJMZ(@*ehHi)`PcwCL$^>8V8Tax2b6)Ag z@Sg=rcKi|NEqA0(?Mg*w_)+Oj|3YEW>VYKg0HnJw%b_+{JhRm(@!_)TWTT-0^Vi)l zZp&b#1w~`_sv0zZJVN_&KG2XQvY04x;yxpjaC`Jl9GdhRVYiEMa%ncY`dQH6jhE@& zDY5?RpD&D9s!rAs4}}HR7m+{aCnQ}zVn|^I6y(zJu3{J7KOTtjS5+au?LS)bYZ_{= zu0eK0A*RgF7Z$|57NTa~7PenA;jz=bIqIaiV`oP@8WSEO|3oGkzz}26i$!5 zg7O>nn7Q%}RC*9+6)xb1Dj9t9;65y#@DMgnzM^=24%|lE#qu|;kl*jgvksc^##6d{ z>_aWWRCM`I%@{5owE{Zon~~}B2Fh~|O5MBUNUbc6N;|K(vQE@kE?(@-6_b`@>APq| zuh@z0az|ljz87J^hvAr9PJQnWg5Qr|c$&y_u&0<`__Kk_6VjUE1ye%-nS%@a3KIq(OXKRhO@SrhQ4Fa`b>YpB4Zo}%8?(z5D0 zG`BxRR+IvNy!!zga}{}U1-M2vi&~EiL*wrV#1!9=%DYxe{q&zm3xe07LIjPvjQtxP zVP}msSCrau#CjJ#v2!Xbu9?iukL`K?qH3%ktH^fkdfd;u1wRIsi{6=IIG6gC#&4Fx z)1lh9zV<|*f)&}{!J*kG=dW+8t8w1Fv7GwVIHRyR? ziZ0?cQ*(a_+v+0xH>HBwD!byp@z%($)fW4Sz4?UxKe)@KVVua!HuR2zvcEN+DMdnc z-*HroS`6n|`N;ZP0uOBicJ!SsI;cgD%#lVqQIQDG<^eqQ!7ToMOP|O6lIPsQ2T(P> zhiA@7_$6n8i%CkDV+YaMBF-?3(JTB@Txh@U%8wYW|r}bDF(1u-#|)28jTWtNbWvKynbdrN9OBr z-$m!~w!A+~V@~5!RxFB&-cYgaJ!}q2LS>3EjE4CjYLYHg4%L#L&mc};D>8ju`*N{3 z!@KtDjOYBD2e@;8LokdxG|m5 zwd%;W!~nZTuf=lF?-dyw4jwoQ6-oE$x@#Pa)Y@FD&fzq_gERlCWs; z9Y(w9P0@Y9K2z+gMQ`WDWurMPNsTX-$?@CLbFh3{kGi-sD9U*Th3W__ysyBj)7)8c z<4Hac)4^56eWWu24s-0?YpnI|F^4yZ_w$}^Zr^T*KB9BRyiHTQmlS!NQ6_xaobdkJMapYjPTf-b)A#)bti4>q z0iT@NSYtQ$KbOWwh81!~7qGot15#`jAxh>1PpDnVV{`O)h>H_9$82MXN*s61v4!3X zD})d*Ie6R|%=YG|) z(a6bIeqjbORv1Y`yrxPcJzb?29~9A_53FVqwE!G4D= z9NfdWW}XW#x9!h`Ik!pPVj^`}kYSNnu%2%imGZaCZQQ+VEU(BJ%|AU%SoupOx~%iZ z)^6sAvG*p6H#tI6{tUqp=`Nsw^V~}(K4jy;m zCEW!#7F`;5wJ;-5m6tqfgvW@hut4V3s~Z9$*K`ExQ9zH-{|~;;Af)4`oVN=`0j|YVx3EM>oEUK0KO>qL(vukr2etv z%S}NXsdtLC`x)|viA(wDidgRD`Uls3+pxoTF_XI|4xx@4QF+J$y*KaWo35u>^869c zIFlQI7uYv8T^1duh^(Ae>mdUYuEobCfiB&Z-ix+f;8}qiZPkvr?ed-jf3m$sLr( zElw4D&c8oTS#QO*3+MCoKo*wonQ3|Jh^Az-S&6W(rksvv2Ecw?o#ift(Uv#QPmr{h z_QvIqp|~Xb=+549J$P#CPpmc{h5lkrFh6PqN_~QmJN7f0H)`{ff#*d2?lN~|9pEX+ zf?%<#nC=%EV$=A(NZ6f;LvH;z`D3N9uWl->kU33y)BA{?<6dmiCq&HKJftV`{W1St z6|He!BC*$4B8+Ucgr+VbWyKXpswc45{3=WzD6&`GlzGp3847jTEsR?iEzDa#U39^W z<8@_sxznS8=FI$z)1G$RJIjEk1l^;rO&R#8x0W7URioEB+QP}_5|Y6&Vfov?!dlat zq`UPBMHY{u(LWZGZbBgKa32lXWmfE#bqmMzG_kzLDO!->hT%>Tc(d!C$j}?{#of$# zr8TrpKAFztdDF_0Y*H8JC@Fov)1%+ENa^T>+J^qZtiSP8aC3^JNK=P`md~^JPs!D+ z=b20kqi&aw9i+r2iYg-4z66$wJh5~nV`FA9wVXAj3n^{F=N}{ap(K$r6_Po8*dWQr zq!dX*Vg&V$dy6gYJy@l_2j5KnL=nTzATmefR987js3l!u>b;m$t#YaQZa$sd{F}~% zZxuG?+@z!L3UTL98reP&Fr1G--6>k|iS9!>SI3JrXAU*$_a|c)IialOIj<_(!Lz@g z!S@+;5>GA>hOASgmvi^skp z2e)FxFT&l6tm6-qsQXL zkXGt7b})7ElM{rIrzDe}Xh~*mG(gFwW2E@oQ}FQbiQ9?;>4V;T;jh7UaVAy3?+*KM z!c+^2nNcg;QQaksI-d)D_m$|Ko=RnE0bEbBT14q|`Kb*vHc zv$82NSsjH&-66mBIvI+&Gpl{;Aw-%&t=Yh8zrTi6`&KzCtuld%GIhzvB3}62_6@sk zE`@i(P14b{N7}cIh+DS|!Aq~fbN^wOXcnOAm46j+Vj2p?5$4s|on+jzf zde)j%zK&q!Ydg9d>JMY(jk|gJ=IpvSA5W3lYh6zIp@KsS|C^jF!6?$^ZU zVu%utoN)(+e>7pW@*9b{A+qhbg+m()ki6_1K3SV%mBTP>DiG_(5wr06;|XM^&%&N0WlY@F9ZTJf#PiWkED5WnUjOAvmWwmbWfDbRTK8P^Lbc#r zwK~kK*N{nmwQ&Ad6$QQ+h8mIaT>H9)u53=E@DGh>3$4d{aR)*FAUF1Y;lQ>x$Fp%C zLG(@dKs#8EhgaJQeG4nez5J|L--sTY#$uSU7C**qOb~l|T7i2JB09rd`p@8j;+_|6 zl^HxYVE_j<9K!ozZ3w@`cTC}Zy#FUJ&U)o>e4P%awallGj=rREi6LsQdA$lJaLK7aM#UQL>OU*RX@d{pp9?j9`}be{&^8H`q6Z(K8M!JWP%#9d(r zh3tSf;X_swf0V4^Soh_u+#%2G-2s`>w;1Enh`Xz62Ktci1Z$k@cbr_GEup6}eMzfdA+(%E;&03&Ui2nZEhEC7QLfB`;S1yu@+K= z{rK_s5mHpTVY--0Rj6}?Wqg}ZdhM^!@A*zTdO4Yot@mMb=@hn=D)ZxYU$J#(8v^`> zV%6`7cq%#*6!S~SDq*kC?yN;qx18ZRonsv5euQh3`XWF)#|2chQsb-!JkuY`Tm2(h z);foNl^9W*nIkQXx8>*4=CFCC7k9*pXYsy%7?V+ifbO>NIcb3t8IGW7Yw6lvEwYFo zOl!NV^OAW*NDWR$e)X z+5Jg?w!&pp)bwWg9D6o%I7V&_^AUe17gJ?A@O+NAm#WGiCq=Jm%Hu+q-85mjHElS) zWFbF)AiAk8mC)xE>+yH&L-cWw<7$zo9(p^E8YJ!fv#X-?Y3U`tUm1*^i`&sOZaU{q z%tgbYeEQ_vM8AR~dA!>e?t5`E-|SWcl?}>l=kNn>QpIzbPcOEqOht&B3cuW{&5w-x z@z;fCVZ5{u8spO8-ciaI?pE>4{aVo+RR^VLSWUMybg(g;G^ZG!3D@`QB)@gVd zl8UB>B7c?Pz_R*7dE53Yw06ltnr2uaspVh7+Fu^1&He4 z6Q+FYLeD=Apg~2d0G73@xb~opYoCz!*Fkir zVmw{2-;7(;@#uc`gm~RvNgU7Z6NVYo3U6)yLXy&#C#)OB-5g@@{A(7j4mklG$1vKI z`-Y05z43n9Cf+b2kbeZPWcgzS;(evddva%Tk0cw;b1`AV8RB`xxC@`SY{w*L$7T3N zDYez`YF{bl+jk3pe(AyBr|4c>CCkSj_hhG^`g|d3GWcpJW(})CS6?I1`(uaSS!rm` z8$&uX!$@sF5#@Ln@{3E`IntZh)F++h{CGyWsHZ`B>M%X97U(HD&+G1?|CMh@uoXFx zi!-rtg*#m26L9Zg1HB4x#=4YGXi)zsDKj060ndBzKu=(2_jUvdqd11d9G-1S$#3Nqd)KRXv3$W*N}bHLF{P`;&XDXSaV$;Kl`fT&5C$>vFp9W z;o&<;gMl1vDeUAC@>Lw&l*fi@Ua*}q7t!PTBH1RL-CSZhO!VLOP`D=ai7}!aFMIOe zs|@AnW}4&moLax6QnmI!!Mc7vHP169tB#&<+8;&#TxRhl+wttXYysanYKQ9{{ZX)H z3wfR)PWv-n^yv7qfzu8_f7%#2^<^GuyJ^uP4--M|x2PM_-ex&u^;k($O`+u7m#=*A z`Cp!?dY5OqTJjj0KpH-QH$5^HyS$I0a<4HpMQLtO+|StbM@I*aVF$WqRnz!7e` zocdDS|05rZjx^DgFe`vJ%Q$nFUK4r8+^;m+(^UwJNfb;s*h~K9os}%_a!**=0EBsW zP<+8ra=gEWZ}oEHaHBpv-Ln>xy-p$1HUT%TBw$vM$l~|igWM4;) zJLwE>%qwEggF4cz&E2I(Gk@}hMIxt@AIS3p7x29^&Vo+ZCrRo=9jdzjQIgx_Ku%7! z_$AKxcI!UE6H`l;^i7at?*w^w^udev0^Dpo4g1--VjU1JEJ{2q3|^~6H(mPhKgHFo zRGG$OVH$4&U%^2Hop%Dy=wnQ(Rnyx`L< z70HUMj6#x-6h%WrJMCy|Pmxkdzo@?Z`wzUlp69-=>pYJmLELA1Q0A2K$%^)W z>47KBM_`lr0Z9#Ku&wYNdTqAk+jBIyXi#sywacE3KhEI`+RJ!c@=msTTnZ=d%jrfk zcGn#UE5Un7+ddC9f4e zh$UQ3v-w%)GWyiT5{@~ek>2Y)d2e3?;}`egV!W9m+PEU7TCf_ceelQb9ExAJLT8pO zCePQxvNj5Ma-mo{;4jWD-M(N}UK!=R&Zl#mmo%eR?Ss z-HOGV19g}gAUsW5SZZ+~O%QwS4b_ ziTvo*KV1AH58hVaq1Qv34_*e>n_+S>hkN$?08f&D*Guun<*e#nqY%!%Cy94$25uECw%nh@Gc)`~^9)GY83TyM} z=S?%RZB&xFm@bmW4}K=4%m|XNn0QJ4K&ONKY{x*hKCzrx@ZERJwy-mwzLC1wTo6uM zJ^piDiC-feZI+Af><@iv>M>og@2p90T(-=3V!B-$4Hc2O@5sG7)Zp=H4eDt&lUg(bgo69v34xN2f4%S zlpcm2?TQbo*`(7q4i|OL!=dmImie{7XSoj#3Mk?iOU7_iepjBLa2W4)`_m35OX_|} ziJrRaAj{ncftlI37kd`Im#T3h{tDFGZD4otB?8t6?vH;3xn2npTz} z$X;B%BM~)gHc-H*Wp+c|_QUssIdu&0C^LU$3gaKb^Q(Rs&+W!>`JgNgE~#O)QNcLU z_M`Y&t+-d@8tF^O0(v#p4|7$HQhLXAG+>4@eOh?cZr0^{(zXd9Sm?Tx_IUr_PgjdC z?+%GR>%BY3mer&*4=Ba86efb@*DxVuOnpX7yj7p}x6UqY?jqk!sclmBH-{^IRg;AGZbYDVG zt3ji>rPB23O;|c-yy%{HMux#rI^untI@Xudh=VSCcS<+@bGjLx zlC%<2XD>xRv*Q>%tCFtk6;Z?0J@jCGd-O{(LU%<84solo{?l5Rzg;Sr5FVJSxfzd~ z_Ts^0OPK49gXxMfa2fU*d#7k}m9rV^t!>X{d)n|@T`RshwlBxO6tIufMLckO5)D+m zN}Xq(B5BZA%G!R3u6}4F>2_bF*1F;B*0K2BRy=DvS%~~dF`17|r4PGL)3>p0M5g`- zO9t6IBIi1v-uXo8qZLX&QcUq(@b_c>6Mli#!I-n9JO9+R=3_}>|54eN4n!Mc>eLF< z%)1Wlccm~m*NkBUL+JdO8Px6A2ukhVNPBv>LuRT0uKYKYMm5}|-NU=%%cXE^o|FaK zcY9HxBD$w8x@fs;3+FZF(A4CkO~Qwu;)Gds%iuVD8&7zZ#&r)a@r?BwsQ-x& zdOa)%N#*(SIgsT=sluQ8l|(F|qlCR0Fq7c$>{L%Mh?K{ENWh;G~-$NzH#nE@;3IK6J1nuW#BN05g!2pT zBl)lsP7WN1!SWUw-M13&4GUrOZxfP#wc{$;vCW`O_2=EBixP^;xGRU(v=+!hYzH_HDD_@L9(FRklkL$uZ zzFRrz%U+z*D#x&8ZMjt8KW-eLC--?_BoDLOj8#X@AjC+4e?L@}2TfO($Ec~v%N&2< zOYmVt?uf;sgA4KQZ>z|6YVc>}-C~wkio`H)DE8Wgny$TIz4!`g2%p82={K;pY7HJO zY6H6mX6V;-CT@O<=VYfyuAZ=#RjT&GOh~ zo$UXe?PMRfNsn_~PD96fD-M~s(8je|q;lAn9@d=}-Xw1nid~#x?FYdj_)jWt&Xje( zaDlF^^dPlW+LEzg`mHe1xY_ z(u04DIE0)IGx6zmS18{!qKBqOq=6&pE_V7C z7xNIG^*o@c8+X?H2b&&sP$@Xb#!kW!Gt`I05d&3=9@CFif|YdDjHmAC!&iEhW2FBG zv^vkjiFNDHzm<6xsc~6u1(aW%g-*{3+)q0Vn@M5V9FdAd`5|nwR715*iliC1SZcF> zEG#Ei!R3P%uO=Oy^Kl>l?W!UlHLfdC-@F#=aycD-TZQATZCP0(5seQnLv`If^s7u0 z?xXpl%NETVnbADD+b2}-dX9YycOt#N0)2jUNHS6?kY?VBCFf=9$=Q4fJ(KU|w$JCY z`e_FtzWT%`#$M#C)L0Ikug{9F4dC;Bqoj7SuPn~3Otz_SrL=sUwR}O2l{~V-T0Z-6 z7g!aTV6smX<b|+PD@o<_$v94GkRjo})oCVSRVv8=7xgRHuu08QDa3_1Hhq12929X7>rP8={ zq_V98WnL>mvtjs>&6n=d-=@ao(}~GRNu+ zf$b4j^vPO+8;=VmRik2Qsr;X`Kg-3w;O98|;}6E#AKGyWIac6KT|mH_*^!>w1&U_JO_KTCK@rYm3VPE^=d7qF8wO0 zpZXaZ{c#kX^bLkn429-8o2nKIwe z_2=eq>v)4{7@ug?;N?%0`GxS2Zz+vn`Tl&KwEh%#!FZT#P{iqPd30t}XX@J2lXC2i z&IhMRM*NML_aufN8NOZoEhK9#yQFwPftFGC| zX=lB;vic7tYd(;QJPu0vYVOp}>%KJ9s7z{}c%C*IJ)?H7-;mN0Coa0Tk~Q{5@I$TJ z*iio*!?vig#*V90_q>vvfAvSxgp-t~d!JmCRFSGNkgKG@Y|?QiKQ8m+2f_oWVQ0i6 z4LZOXIoMkD8Xfodq&F*;OKU#GOLyx>^VLu4-0t*u@tiW|TgFXr>DrcE9_GQ$EDjGV z_TXx~oSt;QMehv{k@=+hcw5zV%NCtgMNNH`Nw>`5S1Eqi0GQ0(j7?Mp??b8bT?+!` zYr6Z%7xyqFW5wH~c=-Uexp@rr)|;^B86iz;F7CD$JEPtr%Xp+5E$jEAtG)2_yiFsk zwYTZtS{JO;^TQ!iH?-L~6>D#Y7_ThBXVpTePVK-;x4pqphodNp*hni6MA6dcPIPbh2NaDdLuchQ zbWvSR-;%D;hk2hxhyIo1__Zy4`tC*Vzh?6IiQ(KhZWf!WZsvIr{#-j_I2W`SB6`jv z8vUpp4k$#!ATkcYOU?Kj7?|GlRZK5`6k?1m4!6fx`R4GIp_bIqBTEORKp;oxALeYzK4TZPLb!3Z^mmT+6Q11mePKnLLvo$7AHSErh=vzi6p zh>3&k-xSRGBf4ptL)mNlURJL@!`0Kr@}M=|+-=b~)*lebO%LYs?2#kcDk%%8^U)~+XX=_eVro2RY5*kZy7q#-T;CCz!No2GY@3WkHP%kP*x31f;)Z#qeI*SbS zfmoE>d5+`RN<66Gvpgi|tNdi_H~H<2*Wqxc3e)v|!}8cmTIIBwhHN-U4Mr+Bm7;*7 zCh54HPSAU-jQ3^cm~uq{E;AD8t4tZbq29Q5I0J{mwK(j+YkW((1cx122=TTPj-1)t zR2{$z(q6*oascM<55s}5JNQ-FhrgFDWy^_6@#c0YvId-i_Sl2)iaLS=lhk;xjyQW5 zjOW78M6TLlg*6N9VfddZ{&X9~(a){<-y2Kbw7v^MZ?3?cHT$t?z+QA~@Iq_&L?ml> z=HM+sm}FuG(_AaLex;4v|EaRP_aZmyH$GV8i5t+gd^O3HZE1JKUa~xLnj* zdjyqPVcfxb3oqWI!#k}MscWh^jq9_I{+O-d>#OE*g1#}AsXU~eUx#8#r%&V`@B)(p zJG1A!E<7%a@g%sn=tCT%$q{Ni{a-RKUiN}bEdzMY=O(Tj*jwJCa0@#;SjUIQOyk04 zf6jM~;Ftuk$=QDJEOW!T*Bvm-_ytN&AB1132mB5j@R6_8ShT_q`8&4?NBBwjtxm#R zB@GImb28N}G@|Y$d+E_C$xj z_i($wlUKb_!H2<%F}eC0cH4F4=j}|{c#4c&+~?81qg!dO*&*qqiSVS&AIHWP8@Q!1 z5wiwAgwbqG)*4{KTmNcutXmSERuocE%@8VCI#XKu?2No-=vBFQ+y(iFtOoQ?Q)I6h zVwO405f9zY;*~`=RxLbFF=KAgt*8eyU+alf6&gX>UmUP)Xt-3yh=y9-Ue(V}#$KJzzcy~h{ z0>r(z{@x&r($R&bVj=wv8zfz}amCy2a%>jNFNarsq*cx%B=14mvPtbHV%okuXn0m2 z+3*%#IlafUWnKB}GY`sokU<);1!QfTh)uSIXmY)Z^CB;BeSawO65o?YFGU=3*-x5n zcS%7pUX-P}1g~nVuukyM0)8BT)vW@&Q#eEm(%R$XF>4&`SBRe;ckrmPE&m*2D%pJy z4u+k=8SsCe$Ok2EyS^(oIUJ+1lPRRr5K85%N`%k+2Zg+MqIYpes6^|uaO3Tv4I>lT z>_`en{E6hWML%GA$A+JG9mVD!24b^HIn7XQp&1dH(wN$plHrUc=klC57O4ZJoM>AVbH7>=bI1;gliJKA-AAntwi zqfGZp^v6qGu)%~MBd|RW;diLoa8f+0+oJp31iVmuN<9|3LT!F6-Bxs_){=P2?X#6z zoDQ+ml185O#Q=789^~6c@CT18i!*sFy06m1%Y_}Wrqd+4c_)~Z#)eb7HLvK@-IL@V z)C1nTf0AL_Zluw_hLrlo(bdykrCGx}u`G50@ABQlkMB5=#oJ`69Mq3IE!SgpV;fHR zX3VX>deFc}${6@&4+h+kQS+F&WGZ!nYML@?+R3o4dl>9;lM$B}id8*Apm{sP?(2ds zR2hAmu7BIdCBX++>6e1M?|ljC*(>q>oC0R(Z^nz|TF{bzrL85;rKd}BB*)j2X@5o% zJvv@XFqw$Sxy~}TXL(XirxR3rCR1OJA);e|&l?OQ;U6!Bh?`-Ic&x>Bi|iy=KRpNiteH8RfV zAiG~;BNfWR?V2B5wflJ{13C}8vc>7meDLIa_--vlNr5^Sh_*Dw&~=?>^!LU*8eqO2u@{2G`>2=}kFln zdP&AJl-Vlr50Y9G`9;qQQipUU8kd+NdU@>x^Kb&3jmJXauM*GegBuF9+EHa@GcZ~bQNv-I~)Tw?CEH-SdZU~pbsXy zpzx^_u5aRC(S9*M^qk3Oj|;cvfvp&ra|SCEY7yYF2NyH5FiSA~u0&rZ%TaAs8W&)(@zIEYT3^ z3z|t`$bK|thM-}XH(Zac#N-q|oa(s&_8A}e>+BYGGgOvezq5-DSU;eq44^=8#p-kB? z6yzGs=h&l96lYiUgG#9#Y)(F-Vc`W-w`DbjcHK@nBaJCgA%kkub?{X1vL{^K!V{;? z<}{Wl4 zfewwdrB_QGcwt9n_7l13KZ1k))OROV*7`%%W)SOKHswEkOxbJKIH`5bYMQRqTljm- zh3oMT{zcu!k3Pq7rds&7yHp@-Y#P~5`TvY72Eylul@ zmz8nocw6}8za;q$(e-*$jVsHp!PJ`3{?2H$UR?s^v?SS&GZUmSB2Shak-}eEQdp|b z;<}gNy!G>Fj#{tHK6wxL&ipgH=H_mmaNs-^Clq2@Q5YsAxUkjyF!nnc#|;z1q$!vC z;lZp#kt-4mAK?Nty00bOD()`daKltyvZJ%SkJ)Zs+rF5qiXU-I>Kb?*Q^J+;ayl?e zn?3A5VHrPy%@}W9;^)E(S6lJ(I$bgv^+LLJs7{J3Xd@+WoCvYO!1g44yNwagWj~(1 zlzbQ7CB2%SbT`wL#wrES+cQ>_`{$U{?S6NhebEg&M(?2EMNvHZLn0rWlF6k@d&$eM zS;%MgVtL1g2FCm6oZx;ZA&OHS#LYEp(vovq#~$|3avZj*&u@+hA^C6iW8b zgV?vDQ*mEh$U9GE5k~~lCr4U5(U@b^bU5$0F`F-Ufk&_L7_c%0o2~V+pezq{4Tk*W zhhTlp&c?{m793l!iLamE&s`qparf~@`RJ=^e*Eh_|7%sidG8>xC{SR zatdQci4N)T7;bIv%pNTc+|W3d?0{j$!wO198SyqC@J9K$!TndiZUPKSNg&@uLY^vfd zGT^D4pRxk>orYkxcwZJi7TryedoDPA3Jmu>fl`x?)5c2ZVxaro3o~0H_^R_>ys+e%;SCB-Gdb?>$ zq9DBe6N!9VBo7?E}xE?<&R+v_wkJLf_KKlyXGCC}|;#+6QgDRip^ zmilJnX8a>5t2|Mgqo=6+s58_?kp1kn;I$j)m6bLFtfR!?jo!bwM9n4^1-2tBt zrqeF(1+?pM3>oJ|!fRwT_6<~IgT~vi8PJ5+{Y%kJr?c=2D2v~<=)n%v;0{B5+2VQv z8(k8a!o`BwGw?c2Ip2ht_k465e~CPXj1$fiT|TKAiT*={yZcsGm={%GtA#OirrE(` zNeCQm>ZXEZnKw+& zS<@-E=T!a^;1^w_jVXtuA*q^V)wq|u`|TsImSD6p#?tP3kR0P+la@0<$pq~Dj^1utO^1!`S?3l7iI@(iCM-{Eb&TFAyS60AF z@Cq`L7vXwKGii1nOd)?`aJ18ASS~+|?UDgX7uM5MgM&2YM=<*Ke@H#dT*#--MCzKN z4f$}92R?TZ_InSa;%gC3cajNy!wb?Jzn-$C-&7m=jnpn#30{LIl52&B_s&t2zdw(6 z(+dRWrU#U}sB-emo_ue)oG&+QMC&y({GWHJWwed&WoEFu?QVW~;k(G%kH(%?52>L% z7@w==QEXg)NpJEI(i!rTydU%yXPJvwdS4N{Y=1~kyw_3S``#3G;+-@zvptO)Q$bt$ zx})>lwp=G%ty`WaBii9Ung;C?S%RVBZ2XyOJxs{j2W{z8E}hESK+3|mSnD(a@t?LKyYU1f&5n^>9Mg^+#nRe_R{HeOjFvafltRw_ zB$e|fXuWNOMI(Hmt=^HM2sO8fu(@l1Jv}_+BZf_ttKFXrAHGm8mJS6W{A$%j28|^Yg-S(0elywYM1A6*l;t)0W$+7;yMp2Wgj`_%H8b zD}6qfixK~(<3M^ae*Nr+MP4TGofrhA!UAk4?v1UX|7g>xJ+#_m2aVb5NAWg~&mT(@e0g+du-<@lgSq$lk9GQ6&w75tXYXgZ>UN!LPfEG!1&kFUgzSV!5n zMIEK0))1L(dMKuBxq-?Rwyb8(Lhn7nGanw<0v=>a|t#ba})Bv3q!N{HtsNOI{rMUM+ ze$5qLt(_>pG!>tIsq)Odx}1{Nj>oipjx|adNSP|`vD2+^eyA79);mBsXe}$}+VRQf z>b#(CGA!Ce;pook&`xPX!$L06(A7ssyXSWHoNL9oEj4g=?Itg|*iJ4LHE`nH9O=*2 zy|ll1B!Zp{gx4xVEPii_s^k~q^Y`RYp4R-m$QK`c49}Dd+!En`>g;&#;!Ou?2)7qy{%x3{9abC-QOUJTq ziK@JFYbCeQishyTb@|GnmFyR^mj?#evqzK>Pg?dLJhk+>LS-*jSbZhmdP8#QrH&1) z<6-n5jdR1Z*llw(xBpOs>v5VqHAs)gf48DhYo1C0Ju^uEct2XO)Rv|y4?9NX^r?Zl3AWIXJ zp|NZqHuzg}=#>k6VssJT(H8lS_PJQM@*^&}j9|+Z>QWsapk}EYz4V{a!i5Hi`n3ha z?{>h~mOs*(GGl3!{s$BW?MH5pj+_~ApW0q_g`blbs#SF01uI``c2YtT)}flu0qaG9>j~+sVms z96I)Qf&b841ZBR)t6BYUaQiYi3|S*~ujX8~O5$B1^H_6xne;5AKdtC(3xoal=<2>5 zRQN889xgtKSep&F7$GtT>wWmipd>z$o5HKpcOv0~0fLQ};FR%7Y5PPAna0socKL~y zX{+xr(Vg3bb6Kg`VSA9bDf8EK+k zo(IqMp4{TDC>Lp=x%vPV&(<}jSx zaF4v3`bw8eXK?tm4E~;7$iKfCQ0~AL^z)K2#tnRj{3=7vh@L07A1;Wwqe^Q}*Gch_ zok*kK6dF*pkKXDZq3dbYlsq<{Ru6xV!_L|0d|T`>P90>`+k<%4@+K4%ry+Z47~Hf| z;WRK2XNrFyM7IrV=li0~cCpin>5B7u^VqSyJ$shqqFaTMEVN*jOmXRCN#nNYCA3rt zK2#NVOiyI>nhbv2PH^LQ#$wj~fjBHYj-7TE^(puL9nha&sgcYoo@ zEz`}p-=p*R^0I}3a>~%G*o5SkWTb_P9O5<&+!c9^H(^g{l;T2KHF3P;q;gOee0`L3 z`-KYnyj~9r-R-D;dm63Y+1Roq8SW`wEZwx>m0@jo$_NFyQs47@<69hGXfj}{56dt$ zc^I`58I~Rw3+Q-)gnZL2v^7|T4ro}@__a0o^;$SC%S}1DkjdAzkoJmh=l%LYRHf-a zSHAbg97kU$zkiM{W!;%w*1+s*68!QGU_l=%9+|AmMgCXNQST@>FM7nm1DXW~RfQki z7o5g$4%Sc6MZ41vB^QNwc6gD;_j?}VaUIRMvs(*xuW!$-wE-L$7{(LBcJkhm zO6n8uhIL04i1X25))Bt=8nxA&wn`ta@Qm~f(Y;FMg?|&o%>N>Ki8;#_v13hD$Q7BLGz3IPQsOTwT-fyx_V^5)*&+P7 z=`&9fESB>tlKBh2;^eGjd~;DIulA1Q!1R|`=2Spe$3K>JlYN$@Z|O^e54J(8Za8hr z`bM5@Ja9F35iLyEME{wbrB&M|vQwp;I}dq*&o4#4arYM9=AO#K=Q@x_bQk;@c@Kw* zF46wz#n?()v9NP8YAV&RYDYeG?&U7;uH`PD8(oOmVY;AXKWy7mZI`2{n%GjCQh zSEZ18F9+BX7oItg*?+U zB|hJ z5%C!qSa=wVLw`Vsv(VAUMszA->BYM!%=amS^1>KoM6RYwsWp;aT#Zzn-j)~5$Uu3V z1o@-wbhf`T{kyrABJGCJ_!)+{G%!+ppM7w=Oc#c)bTIP&Gt1p^^l77D3q2K_oWAd2 zrdEJ7-EHJpK90;R`cXxP8~i#mm``2R;fyn*QN89q{fRJu!^T1M+xtFU+n<5;bp_bd zTbJK&kyv$W6J5I-4Ugm6yt&_L;T~Lo2Ul{*EOaUxe-Gyx%Pp*vWyT8|m+;x#VvbMs z#|a&6PH&Cll{}Za2O44P^iIOPa));OS4Jz4L4&P5?7l4gBxMCogKW)NIuUz@8l7`U zzTbz-?K|?IuN5%Y+#^kQ^_GHsqGZnd+e`WBBPnuYIGK0Zio5;^m{K2%!q6vVaPbBG z(T*fHOW}?@x0k1B-Dks|nrQ#+1Xfy1B`W?x#YLj?QMFgu%6qz7{iM-r@2>UzF(rhRpe8rJ!AQ%o#J!up#;l`RLj%xJwP96 zhC^}KMxEU{SK`Wz0BoMH3lX}4={{jGdPdi=1%2lQnL6_E6D3j|n@u~Sbr3F|0kVU$ zdB?t?tZk8uh3}5y(5nYQo-SaV5>Ny(vAu7%k~cdQJ|%=ZM0b3afYQ z&)Xv%x#iCcR`s39PD3a0%M~)-`l(2glBP6w_OA#SWUuH*C^@TxnDA!civLOzg~Pr z!HNgDS#e;!7Aq!dh;y6`Kg|CC^Zv!?P#ccllRL6`r8npMA7Jasd`=xJ-cQ1>sdOzH zNBvSzZCeU|`$U|}AA~QrXJEEZ5jq)r30_|xJec}HHt;~YEM6^M7XRcYyxtEGkAd0L z*2|DD=%{mp(qFv2XThVt_vS8pg)bQ=*_(sBYC;Wd%g859&3rQSnt|VD+9>aG4zDuI$ZGg8^6$_KmMuk4 zv$f*z4k5hZ62oGvBQBdA!$_q{I+J&Z7R^6R58K|sgR{f%Gu)9%HQeD|X@K1k#x!t7 z3`*Z-!dm$tPCjhF>8y(wbhr|8e=zTS(1{a!T63zGB4_?Mhn?leF=F8??r-45>8o8> z$Fm1FI@Mve zN_;Ozmpg_;VaCf+EVwz6_i6Xy`zM!k+2TL=aNUaUKOVscdT+q>k@ontzy}Tk^4YiX zg7Dr~am|pf#-aM`u7i@fY+}mN$m~Ie3HbS zZ)LLmo^q~AXr-tvEi|@x0yGR`ndd~a>xOx(ceIS=J}{@zf1Pn}+aUg;_XRiiPr^#u z@3MXGO=;WoQtC28hu3Wsgbt&%JT`C;ru3MDgf1u1=K5Xa9&ds5gw9-B@trh(+@vW1 zH)&bUS$T)DOY%{(uE~ql>#(HPJ*b*QAV$RwIuV0$`Gq^2^1dPR+7;}hBhZ!KAVvMF zbaK%J8hdcBlr-|A^!j~Un%umK_I)@)fsW!C$fsqmeP`aDrv_~5r*M~mnp z?EBH4$C)>yBEf`jmA^pFleK8yaDf(fnuwe?X$Xl8MPKzi{C@ESKgWk7vwuJ6xwVD6 zjT>#BQz2a_YLZ@Te}k-b*RXh(Dp#^Gsyjp@DCr9JHf_a-VJR5r`Up?2n_;Aw1ON3` z=iJ9ZDDc}1?Z^9Zb($7zLgjdJ)C&!{!x7)`ic(hefbK90;R~|hTKCbMy{47obW||g zWh$nXlw!`Oy9nQ{!KuP03XQ#Fy|Fw|w6Icrsn|@+Q9-y3{9SH>z}A z;h0umZfkvjJ_hclFJZH2h~Xo+N44W0VGr?sP7v-cK7n!S4dU)J8=-5I(uWIlp`?kDmPbcm=aW8+V+`;QwlX%p+F`V_+l6M{J#g|kZ+0LpzZ!ze> z6Xs2o%sQG%zI7YyR6Vo=Z^RSNtrz3kbu($GOO+&iC$#(XLz>?~i87|yNIiPm&@vNi zx-#LG^g(d|DUU6n(`vRT-@F8WMn++I#d&1?8-lBT;dF1I2C?=Viq2RoEx-TEZd$4$ zmdAS|DliKTHQS}x0m0J1qzBRn+w;7!>^Q4DE92s`o)lx&g-nNZAg_iIGM&k1>=th6 zEK|vjvTN*p(oU;yqAacVJQzkt)3fw+sp-UA`25ktm5)8K^f*yO-)U0isx6Yj=L7g& zeHwYL&!CgCl@bz7>EPQ&$>ccm+lRLxx9@}@(@sjhCl*S--epVguN=pMm4fxK_$@}d zo3Kf|9TyL^=C|kiVBs!P=>9emGrZ0+*^*eli>4fTwc}~|&;PE;AI*DA^PO`g z$F;`N8mF7m&!7*|<~?aru$~1SIX$1=WNxCxJJw0lz7$Jmcjrlpq0^-n6uP?+$)f9TIbzA*6P8A`Ljh zbmi_yskHvP>~5U76um7#_Ih)+?CS0llFI83G^zO>S*BFeJJm?Lt+RI7xt{%ES6XI( z%IEpCx4f2=<`DG#f|2nn8@+u_(c*UwXb2QN;;p@)uql~#>{&te>$}j!n>iGGSr;W! zvY`7&>@g2Kpab8}(U9~dQeO{uSwQ9?(z>IJ-osU~CvX+T7{`!C?_>0GqnIJ_c3j#M zgh7I1oINUz#%H)v2j5~W?2b4q_;#HGOyeI8gca|-m4BaGhX6Lcw z`&SG-_!;%Kc{p({0nJ_mIA!4nG#bpsz}8j#TwRmf=o@?JKb#SlWrXpiqHFS^p60&3&4b_7^89^@^4Z(M@v*~J;dxWQ zyXE>^P<$5|f-$#7WiZaUj)R5aLi8N3Bznp5-gKPplm zitYdEXw2_U;{EB0HkwN~zx!Q&eMw1P;CKzgP849wpfZ|w$cUq3TM?yy9=^gWHe!Yu zfnTH)8ggY_OSW@v4 zf7~M}K2BTku#6@5sd=&?Pwq%({R^anzfvgDFM|v;RB8Iej!YJxF=l%#yw(+>oo^+k zi~Z>5ze*g@uOr_Zug^-GhSRy|ebjwRXI$y`06N#(@PbFitOhmS*3_FzMlsKI6?{B} zp;&rvAnyM?1Mf+BP@Z@mW7ofu)I&AJ47*m!Xw*b%>s^A#o9`3Adjn7N(TE!E27_PR&1-sD6O{qDDgVpsErd@ zqbcl@mCmzoUEo&+#!x=m6VbOkuwJtRyZiURg(;h<>1Q)(4{r~Nn`rI5a$dLM6-OD? z@`6J#oTjml3kAFJpMN<#X2rpLfef=eAJO=jUUaL0$bE*)eqvO2`!m{__DV^gFmuNW z+>h5p^UMzXY0gyc*CM(MePl4W+?V1~4N1%OIOWBBr>ybw@cv&c>HI4r`!Qvd(_V?= z(iO$sUx5cLIU!i-JH&o?3^s?AVzS;t%HLl?>ihcek-;yqV4;t2VhzK5BjNt597J25 zZ=vE8WsJF#jk){)ql`##}r!8D}|P?8_9PP7IARF zE}j@?#80doxY~FW-?Q6IS43v^+C~%p_T~g8A2Q|s?FMi!jnQoLLU3W@bh$1h4nc!- z(fYlTlI)xDw{02{-dV!f?JMRfIP-u5d$`wnPd?La6~CIepO3R1`(JnD2aAXC$dB{T zl2L=-e`MTXV#R*zf8*ZX3TTCtOS@xNNlO|7Xra3<;?@nrzd13;6#MDc+O=ZWUPCX_ zPI2YoNo-%c8%c-uv4Lg;Yn}1o@PawG^>aVYiNB9g*KUZ^J%ZIO_Utp_5wA#SC%@?} z$$x2IVfnBOer~y+mqdw-FAYQ&pDKhFOqF`{o-6tGv7%e~J0*L67mN^nXHR=0{uF7< zFEbT+N`w;h#V)xxYbq{$`;EX~yD;fkUu0~&jbECN&}t(*+2*TZwtfpLa;pVLSeu=_ z#VjT=l%4x;;M~1!X+l^aF1>@t!p`PtxLIkyHNO6%`TB zc;%<#JoaKbSI4izmlZp(Wb8O(8pgrywJ}5|5gNX|ssDsSQs(NJQgD_6|Ji;G!{vja z*mPd>#IJGh>~mbEti;8`>rndTpl~jyaZ*(#SBpD9wP!cfpY}ww@e1K?zschsCh@M{ z>P&+ykuorp=KoBUWv|%A1&v}pR&|JPhS>Ak<*v@cQ*l82MMCt+X47 zRgOH^JPO0p+6Y!m8D2P6<4Z^`+Pba4^VVD>4StKwJDccKrx;S&t^&o8s{G%Unfy#J z{d^{QhF=LuIF_@SlV`o)cB}N{(Z@zn zMrnV{4pwF9?^@3JwwfJ!&*FNA|8aEQk6garA15PwuaaFf(D1s?IWH`@Ub-bZTjC9xEuUTlAtAZ}O zGhod#SvDg{!VMFJ$!O68N|v9EI{J*6Zg&NB>^~uIxCdST`-gXy58xs#9_e+mZ@}dw zdf?$le9@&e-bb9gRJYN?ifZ~Ila30-q7*YGk`x)M=(9VU72k3|R|R2yi(4hp^4Zj? zU4z;~6UgO}Ea~b9P@(`wf_DQ+yVZ_FSa;&L

_ah~29)2k=rw6*e3=Maj>5X;y|5 z->|xZo8ZrK|0xG?V}~0RKiy212cjrCbPK+?nt;WOH}|XN1J)llApC1eK7(^{(x@WJ z#Mgs`-%cb2x4RZf2~&sHm-= zGmy@C40w`)#6gN-J9BU4H8l6^LJC;NcBWG&kdwC>HOrZh)_Xp>WD8TCR61Q4S`JEn z&G6@Y3Kq+pLV+L~b~Z=?p9#mfF>!rRVelP|S2|F;fdd^ASWRlr6Dj{yF4_2>AmiLe zu(GZaiup%zs(m*^c#OwqV{$Rf;}uqJ*I@Hfcj_6#FbdnGal?5foKx*g<*XN3 zGmgPnB?a);d1GQeF%sQZ6yvg6aX5Es6LKxu^s{vp?VPup`3;yaYDfnSi^?(k(FT0X zzAYs)R4MCDKYG&u=FV77ck)?}VnCS89*UsQg{iE=as|zk#?$F%OUR;QIeq&Rf%e5R zv_i*=T37C)uXVdf!7P*%x6P!NHnt?Bvy8;poNTPqQ+y#Di%JVGV#o?Jv|_ur18V}P z=a%a?U||DoQF%=T`Te9k zt`S~E%wqG3ztHoqp1MNTP<)jT8LlMwIqU-c8y-Q&b7MSEDvwkAl(C5A1cWYs1gE*9 z@b*qC(&sn0O&#$57+1c-26KM;kS_mJdnl;p=fT3kQLqxqpo>D2NO0{54 ztZR&)9zl-P>XaVtiTUhK+P!}%r=Yr+_xUsH?RO`UyjlmfD0)!3jW-QG_ot&P>rf%W z3X3imgM-#(_;%e3edZ_Qga{2&#bI;P<25d(x|6zaXNm?Lv;2!2ZF{-*!=FJREfDeh z3H+>MjuPJsVTI2xw5)N*J1PNiQtARHaqb;2>%##PkIp`Fb1SE-MF&V(VjF(<<>QZL z#%L5^yoI>)==CxIbL@5Lq@)cU@w6h9FEMoT0Fv^LBCK+`jeRol_$F3>4-b#yipeIM--89-P}r zPL4wSbOjy$m2rnDd|?iaY`Z}pZ!Dr6>APsBZUjmC{X<9LW_-cA3QhCV$?tCti8Z9q zRok2J->vEJX-yGNE+`teN?3u!ky4Y%a#EN#wi7ZQ#=wN6Y1ncg220wKQE%~F@U}Sy zcJm`(M$;q|8Doxj4V_TWN1ZrPBT}DXL*Jt|Qb@)zs`^wxn^`Zdrm2QXd;U_}0s%ZS z|090#jibN9qEw)og#$lSa7EZ+c+b2|? zB6vnd06cPpxt*qB+~E8poRd*CC#(6DX1Cm!8ef1{6Uc%}{5Yxh+TrlUsQ??a3zdp8nz) zb#5llu1Lzh8ciPM>h#@OlEQqYN%(vn8vbxZdeQ?vlfxj?>H_vPFh+UU2*2j-c+dG4M!Yu!CvOS%^RS#lTgd|zoOOpTGHC$R=4!89`Bd%WQ8ZX)D z4o_p=MsD$how!P}4VUlgW=xl5s3qlw{)!qnMQ=VEoo%MDrJKoXV;bof`%~e<1GJob zMA!H!w8K(^5*4H|@KYvdad9hX9e&9<3NY59sU1w+(E^%Le)#_E5WdKM%A4-UdIYa$ zqK5hmF5v1o7$au_^{wTo@1Q`Ao1JKQq!u0guHct$;?|aN`hop$6Yttj-Ae1xZbrf zCceF`pn1$5Kjo$4CjB=QFV3FJ)Uo{DVG*h+5v7zvvXr{i2Wl->;z!3zXyI}doX4Gp zxq3d_CGTty7&wWk=Nu?wVI%WSC6HuGhqY{7L{0L>rsjdoqru6t|289B}xG(NB3aT_Jd-;Z8?85nWkHuC;%rWuZfG<%T%|G{iQ z+S!wbOBQ3+3&)tC^-lnF5+glmoV446?fPBH%(ok$ydKSnV;2EO*Q&ANxR?# zg^uRpk;d^jzj`YC3~1&=_Ey5WgA=jRM9B2WP6^XHM#82Vl}5DMOPETvGO*~C7rmKz zgVb(*qDuAGusECD8C{>iuUIwI%AbvT;}ej#{{`3j=>~6y-o#m+^QFjCVirbfc5@ed zr0^w)Vun#G2))t6A8XIT_Mk?P%u*$(7upnJu0Rzf0KVZx0R2ozwf+QCn- zKlB@Ks$3AK>b?QOyRPG^($8qPrWX~~D)1un4i?o|v~7Dy|9nbWZ;BIIb>M!Rx4$SLjwt@yMa<_^Vjn@gtiCdF2x zPgo>|ezC=_#v(dZMq-h-CSF|<^C1hK0DCb z%5;(pX&}Y2a>nKmf##vzymPZ`c~Pk&+>stFu2O0>?^1jy%&bp`ZD}{a;ZG7pM28Un zi#xeK^e3&@aB}BIlA*(O@UD@?;W-=8+(Me>T(Bjz15qRvoKCYZDDyKG@8qXzoWe7Q zSr2p4aY`GwN2UK1`BOUh{Pky*aqIbf@L%zadoCPG#lqc~#{Tw&sf@Y0*_~V(vdHXK z3Ceb_#`Ra`W3>8hG<@HI&BlLG{E`NqmzfTRDQ4WY+M9HeJtt3obWl<->#DwTq1n}| z=sIJYAH4e-I~NY(u=s0Qwm^h`xm}g-9^*pB>|E9S6UcfC;~XS(qEyZvv>GU-L$ca5 z?`SxdU!O#E-eQ!JKEU2xL5esoP45nC(B)neQf3*?st`}=?ypAOtOmTXvmOr%81P-I zo%yF`M)0})X3TRp1?2M;VO&@r-trm5YX?4IzUVH-S=xcx3&L?t-f!gA_Ts@Z8L~Xp z#*5f{hL=52hNrz*pS*9)Atl*8G~D|*vK}d+q?kbQ2E3iLI_AOLL(kTr-s$nkxR!8Zvmg%0@5cx~G1xm> z0=_&4uK3DMINsp~oxz@v6PJjS#WrD@O)9Q(OU9zENZjD!huPl3bl~Pie50C+mc+?4dl|QaL2=oXm=qBo7c&pw!lgF*EEPjN0`T&F+Bo{ zh0rce6ud01@oc>AVcwqQ=*9YT&o4!QxM>)Ch;HU81FmAnvwlpSs!M58H?WRyC_QsJ zO~!62r1OyZm^1^}JfH<9h*#l#>lS3&b`08(f;o*bxZBT~S~}gRt1ppEzn!5!S}A0x zb)5R-9;0*J5QdbFXWb@ge4ZHtMF)Z)L|vZz9@)}@mr=x5T!MQqeS@OT4EQl?2Z;%K z6TfH|@gIgmvBWBvwTQVw};H$g#P6Hi1kwioXaJX7URa!?YD`s<*SffZ@Q~8Zr_%Vr>vY@u6WOh` zqJ0ZSFmTEleC*mr6+YK!Mr9`Ld#gm3Kcr*V(N*XZ=Lruw(zxB1)^lE(4{`Nl_U{Mz zVoM8SlBrFm>FWCQXyYb&>u*8F$J@{~rH`yj#B3if^b#^nraz&ypd(6Mkgg%$YHiGq!kSF)25AI6cv z)j*1`$i{*Ci}*h2BAQ*j%Sngpf!d>BO*apw{=0lbGhsl^M;( zhR`?A@NXql?sp{t=4&}Q^*K&`p-SbxA2C6`0LNrb=akLYalL*l%einAn^ziAmHi@$ zU)qZ^HjJec&o#*Bl@k`){e%fEzoFQ+5_zMAxO!RyFSr+9)~KVusVsV? zjKIOkd|aC#h-U__!M5*H>3$Sr$e-6Gl}U}5HXGlcNOV?osT?+l)=ypNKFO0>UO2p0EAK-fq(*FW+F zcCWe!V|%PXY^0iYST&GRdN(yC7ctNIaVi)W%e?DWSkHPy{w%Mqv0jFGyB}erksnS{ zy+~H_EED&42L;$iLV}6~ti2ft!AE6a^4Jw1U3~-`)|5ihaX-j1*a;^k=HQcD=D-Rb z##`dykej3jPeW#c%H@MFHzyr#el3DOjMHNH@fdAle1+XrpJ;4AISm|%qN0Tq0cDfjTSfPOtazAjZt^jR0_yvt+dGvGJ z3+yv$!q8ae?IuHLt!m@OJ$}nrGjAYwQWU7V9>z?mbNKRXEH;!KLiK%(=-wbso=@GF zFGY@?YOldW?Xx&eQk&(_GGR($D(*M0z|U-#Us<^aWit`g0>e=xwu4+MI%rQoD;@ao z2VV|;#=*kB$eZGY%UFi~Q20&wHeDD+A1gur=A|HDZ-e=k%W-1kD12_3iw9c9q5M%r z?7nyh9CsE%(r^TDOQ&JZ`FLD=BL&a4OoA)C1TK{~pXYZ@odW+D5kIjB`7%kc&5Zd^ zS@s}fUOMs~+o0qdNffa!#CxoJGQIi?8eZk_jzvABg)+8QTOW-69);OUMKO1ZGym{A zTmHGy_5AWbb1_!S9hb}!!Ng8UxOeXY_jXPMm$W+phTeYxiBr<({k96aw%maIJCDHv z*5xZN7z1vhV(??$1jc}5-5bW-tI1u2Lkn2`zkVivle`qeuVh}MQHU^mMv#) zSa-uxOjjO2ZSjpP-y8{}y}LN)eIwkqC#zsihX`aaM{M7m2-NAFgXg16@Z}6ybQ2cC zy%GvoE3ZX5lT@fLf}>8 z(~G@kVw5$9dCR&tq206vD4gvNqQ~4p>fs!&s(%E|?6rWqoA*JzH)F1)>tlD3DXM6* z+1fX2^pMeJGhZWAxMzje;`w+|QWGV0UGb1X4BkjJ$KwmyAu;G6RLC&4S+74nWqjuj zcEP`-Y|i{j9P>Y<;>3_FT(UkJe+x4|$(nrN2VaL(G0A5S~nU3LmpQ@Yy|SJkDm> zw#JNIYURbeMFE%;>rc-G6R1`%nQk1q0s^yw;ak!_xVh^U8u(77tJ~&MXnZqVTG$3Q z-|xWR)idbKx2052VMP{C7r@>Rny|5`o8!fLz=-=*m^C?pi9b#8nc7iY@Lv&H)1&=X1ujcDzZ+-n>-%ZjvyZz|RU?&ySqQxG05nIMj0nCxujlo8)}>)I14{ zQv;#5;17ITtARql0>Rbd1|&O|gJoa{EO{aWDWB503b8=EA)k*i7B_IcXgPZ6K4i|( zY#g5EgsD5$%%x(VT$in#{3EH3s)>EXDeg_o!zb zfHt2@apA{!yx*gT#rE#ZFL(e~*7(E7ws!Dyu*WH{{McTl0%uxv;0L83^mDbt{6E2P zVR-;9J!FJF(*rQhY96k>=8p|EM=;~S9By8269_toVU_Vxa2XUw!_R9O*Dnh9|Ne=! znx@o~Uh)6m@a;*_V7Q-8UVHTDnCDNtxZ)HS^imZvWrIL_`*Ey|@5CMjC3624g~cw% zSatjUtFW@J7dXf zo+2p@+@;`+$#j$XlEw?1hKt3&kTI%B>%mLhdQ^Wn|O#G6e5K$-Kcq@4m*;EXoJW%l9_v-lGG~U z<`q8nOkq0?A8(vvJH$FE|dy}B1kSZntsgahV3rnF>l;))IIV9Oubxi_|`6bvN04Ue%#54ja$i?=M!f2 zJYm0wzTCMx8_`GpHb#kFN0kb;BkdT5r0sWLXM#KZ*-BLQMu}E1Pop-Q8QvBtrl;|R zD9(y9f0r>H{AO)7v(v)Y7zX;01{lJ;J3C&9aW}2baf|p%p#PUX8W+`aY3EFM^Mtl? zF*?k{*uv%?2~lXeClDoslQFR?5lx$#xY12%b zlO!#tlaw-UJten|rMaW`FwQL>7fnh)r!{A|H?h;u_fZ{=tSiF9kw5Vn^OpD&N29@~ zW-NK4N+VmCZ`}VXsy^;SueYVJKe7n-8uIAXc@_NiBoA6j4Z-K=XY$V2K=a28gX)I) zxN~F!>z;<-nzi*TGp0-Tz1NW!)G?-UFFth|Lk(?L;S2MQYfh*HiJZ9*mVN?e-Chs- z7g+K(%{PG`iS{Vg;KUtV{sw&2GH`a`7*djPChsMgv?!vJ6vPiw`SA{%>Bd~~%a-HK z_aEUlB*FGwueq=4+1%tzqTo!Vh7Ds8p)*m>tu7o^yo}}v4!Nk`WBE=V@c*zV`GRLJjeBNSukylEN+qtz!z_C;S0M@m}mEs`Ign_ zy091hx91(Sbcn-~J+F9jo6gd^0~?8_G?#kcCcuGAMHFvkyEC?fStN5G=d9Bu!yyHH zuyi(r7Y1<^509hA+ilp`?1GJd`M5Mh8JE~zfJb9I@t^f}lvkRMu2;69i=RFM<PL ziIK(PTJ()vi3Z0N@ug2X>`7JvN7(>KTiV2JJedqL_HV`Uc^PQ8GapNr4xyVM^Fs?8 z;^>zev|1!eHsZ{|ow@^ePJaRyXDM<{#)AB(iAAiZvW=#WR-uRC8<10Rh345)anFJ8 zaK}pv7wrtf0eult@YzTi4!5}5rh#zlo;WTm(#FKyiEuh3g5w~p=QIWE4A;=z@!hZ zUYdb_-d}=%0}41t*$HDG`~b^7W1Oj(iRtHWu^GS@NSS4TDbFLx_+2(S`WAzeVjC=T zR>Vp7PvicvCot0XIabEIv)yP1glw4rb8NqYGjn}-?98CAKhM#aJ*wm_u0?Ot7f}5g zdrVi8V?FmaPFckP>JJ3)EK0_63!{~3{tFv=J#!<)2?UXK%t2z-55k}o^lgI;?HPHF zznZM+&FwbkGZ#WU6GKnci)gdKZPJ{kjQIvaD5(7zKF`=i3FbO9J^d4Q#2i6`{!~~Y zT+el!^uu$1{=*OTH{j$ZH*DxFhL4>t+<)TnxaoTYx-X2xuPa;W{*n&*tWiKRA0#mB zumN^{cScI|L(!pfP>NdsicvG6ae5S&sLTO~Z(3jSqa3!q`XL$A3lzHQme$RSb zVv4;i%lD4CxqH5`-?RNTG@;CbEVZjKT-pc=)mMXOsXb@cR0#5hgD~CU7T>e$J3mua z(NsZOnjYwDlaM&`zJ8hvWy3+-%VcTz=o5-#YPVn&k8v{AyK+VK!#p9u920>eJGfu_ zg$r#C;;FkbFQ@A^+O`U4!@K|S?FF%5-DHphRWiNJJh|mc)X)6tu8&z3>18&_E{ns3 zVfR@UrV4q@!sw|Xg%aXcAnx!F8d?+}_S_!c`j|zyH_`>iQ5uF!cLKk8Z{bylCUUz! zfZ=U*t}L>AR`{tvY#8vsowHv>0i|OI-A5AZRt~zH*HcYA>*vC zq@DHy#d~g`)!Gm&e)k7PWCy`dCmi}7){@sJ8Z$t3v z$`Jgj7K{gvFbDmrA^cBHks4K#;EBRpaDR^&#~1<2i&QYOVmd1S&;vigD9-Eec1~vV zZ?5EX6I|to;N>|d@QLgc!oFklWWF`FD=x;Xi%YQI+nT=l>(aZ7V!VAoizJ(TNU%GO zb?b~!sI?E~o*jnXjkaLkA__~#MR1LCe`1;C7@FK(gIOLCv}#2%-D3CrNeYWWF#j(s zuAPjEqY1E1BLl{Ge}GepJ8|8;Qe5zjxfB8`h~wQN$-H7R8SUfBG|z&;Q$x((SBVZ? zLAXkAE;L@hMg?wsZG)!+xiOnPl{lpHdq(~g*0 zsIcvUhQ)8;?5!Cz&&-~@WT#S0-Yjf7B8eSA_RIlZKxP^FG+47lma+65@?n3G{Tt!XJ z7L>gC2Rys~5k>{&p)(?h`U4)2e7pjGhKe~}i9du#^Rw~fC211aCrdZ#W>VgpL^vfs z1A2^<;C$!=60n+0?v6|7r0fWUbzFj9zCN6`mn@$7!t%&+1JH0Y1#`v2P*`{cy0*>4 zT}Esda90k$|G5ORtImS6w-cLL_~UkEVVYy3LyALZu=Gv=+y4+w4rARd;R3XGNWPP#^QckSqPJG?V{rtS26J6qhVJ-zoI%n|4SH{}O9!nb!iBNkb zbIHsZhJ;M^4qmOL1+I_DH~Bd&I`)KKX8fc0TMGOXmf2|9G=}Kccq-p#N%QBMP^+~i z9iQ_Zr3&G2tfjn(239fP4dX&3JLF_whyPNED^Pg0dz0}I^}KqmAW7>lH1ijzM+ zHjTu8y65rCqp_qfOf>D492xA@rkPXJXlC?uG!9&bIwetfuS%8Pj2P3>^-{!3NTVv| z+WfabkIkOaafi=uoN8`Ff{y+WP;eGrj92D_!nLW`U5h$9rqW=ZBG%lu^mY*nBFao|V(Kx!O2v1!q zqKiqw{0+OO@b~F^;i-Y+D97A%LeCeHP|FGuJ!el&cXnVU^WR$MYhvcfa1?)_P4&EB zTGDli^;Uam^v5XeII|I9>uy~3i1C)nwlj8K8fSkhhr3z&k{abS`6r8P`B%A*Xq+fb ziYb3_;+C~^qA`HXb`+A_+soiqriQLIdvJd)oA;jXLdo>0bUN}3NUn5-H?cC@%BkI) zlX?-Sy{nq5WEt_@Vq4G%RZ;UH%QTk^Lg9_mAlsEnT7fyVr6zz%yq$1TRt)}gsm77r zQ*o>)vGdkSv?&OJn0>pU*YOD0Z4rRIPzJxgZp8~9=bO}3NAm9JF5wy5l+l5CpJ_#m z0DtlL3>*U;EYmJUemU#m{tOBBKDgoXh#j=8_dn_t;?aLuYPhuREQS`Tkd1LOesxO6 zmz(5qxy@v1N?|$IGaoT(S0oDgtCEd=5OqBdfz-(?|L4?$r9L+?;qiN%)S^UQcV);( zeJ;iMZJ@r*4Vc_KnSR==An&I+T(Yq+gpK>ivlm`N1-rlFTvHpgNihSDJC+bHe-yMs z?*cEHWzhSV;Ezkpk(oS}p4^wEfiZ2+cUl2g+*Uycoiwm?kHELzf8zvSQ7qH=fD;FT zD0W&NcYT2uxH=o7{WeQFa6Xz|QVosf#!}Jz`SkF774Bp`!1hi{l3C(Hj=SDtYt=(s zR>e3!*(<@j(i&3#`wstVv`}O8FT6jm3_fct$T1LT_?tXg7oH`}XI}Ja4$CgRyiE_h zvnc#h8m&+Xp&6VhO%*J}t%kw4Y@Hog-+I8+$vIK$g(xx#JVtf4R$S_|u|QpD|E*D_$`bq=_4@<6Yd3 zzkA2O`u`>9r za|0Q5E+&oOX_S{Z9^R&L+@~91JjHp($!B>SyO`v}1V%MQ!sZ5z{hMpA(J|%u2JUHVBm;dSbN0MvT|r?N#>U->1a|tnT4GK zvyk1WB^pfz88o z5_`x}{V!R23h+BB|3Q(72j)MygfnuEqSfYXJm#+C3~WZ+ztFkbqlC0xXmW|Yie8JhjGC2IvRajn56%R`tO+ULyhjn$Ju*Oi z*#i7kuSiq+Eogu12HF;W2^-g6L+x$Pah&BBlEsH~%-}BlW@jSTJA=5|X%kJ%4FkV} zO|Zks9Svs`qtL)hEPOka_Aj!B;#2AHZC?-sYJ1V;?ON2&?$N92{QST?gKDhyd!cLN%Kz+}m}O;?(qM;cLc$?mB|P z?_%)HK?4-~%eoZK={T??9{;u0$iI2wp| zKBhvp%~WuR%z=t|XUIf-A05=1%-*LGn%jP!iX&sG>Xs+hZZZzG%PQi7`=9YlodBs% z>%*y0@iZ24u->w54ycD2>;vD zile^r^la=+IF>dJb6N^;{rnP`)BllsdR-PRhGtQ_hc>M~IhN`*uhL)3NLpmQloVa= z(qKz8ZA{}+k7xw__wE42thc2%=h-e>VIS@&NyLc@gwW|{72{`EG2(s|SR|@o!J94U z)n>%ocdD5e*CxuXUHTBO8eGR;K~8wi(ThAZ{HQ8+Gkv^!4mP(tqw#c6Qhk>P2G-1r zvH22|#pt8BgB7k9a>P9j&2)a%X>u$vqsY1x)MjV6+s%16sVM||`bO9tC4+4f@QdgsyRPR_YGQ8!6y_Ery&PCU9y> z)4_A?D*g{GbO%1#AC*c<9 za9A+a5W^?DLc0UexHp_-C9UJwyg8linZ;1$KXtrmQ3?H0WALHJZruK*k9in`>8o@H zmUfz%d{MXLMO}#F{s!)(aWgeZ&3XhM{kMi3*jcS1$b-}jT}dS|i~>xune*Za1Vt!e zZfXKXuzcZz3O{T<{0YgMc$A5gr`;h)TWa+ws!5V`2X9h@i3tCfwkF?yl^BI- zSGFsM&t!vt_f>(#q`TNoC><+XwVA%nfg)5pi zjEW%QqkEpa-h9)gM1-DPHhHf<%&~w`W zLywg>k5B5{NGqRi?)PLl<1~s9%BR(*?oy^$Azj(6!M`}OhC1FXrOR?Y6#F>>zbd$J zo(p!YwvF8_9{6BZiv(t_5yCUitI47~jEW0h;mlTV&`o;_`5&d(*~T5re1h>(krSOW z+C$9+c@%g23FT)MWB4Xh{LbR7FS_MST(T!|5fZi#x7wC}F`|?|{fU67`VSlG2{fWl zC&b9`pb0zcZ3Y#?WjI#n6yum@Vsj|t$7OjjFYqLcOqqdgii@~cS<|?&L7`k^T{FGS z>ZB#Yh14)aXn9DFF<~a6+rSV8cW%L_I)E}8mXPm@gS5mrhti^+)031+ic>vBf2aL{ z#J_XVrm_`uBgCN0*cv*Yc5p>^Zi9GQ5J>nIfW)wy$-*7iO|BbG;==DVV&6-aZIYBB z)gcX%aZ#mXnHTYZNE?ck$dbN-5$U!1qE6OKO3d6yzfZ=Zh@Az^`*(%5S38lCjvn3i z4aDmkrLao*12^lo5UdeCikq+dqRRZk%u6an>Pd?9-f9<3XPFxhmg~S1p7eM?9dBrV z0p9G-Ab$E5I@TGH?-+O5`6-UgheL5fsXpRrX&jl70Kv{7AW@}-f_{HM&%YnUn+~Gk zyiBw|^%@JVx#OnEe~|@fX`*j8%Wmet%ePaZW!@p&d1VfYYZgOzoHc0AF9(Zwmi5xz z29HH8akYChmYwV4RgDc|9NnY1;)@vFi8_s@{v)V{F1oB0T?cyJd7 zlrCfQoWq<)(jqWE}X8^kpaZDnG)WXIG78l9*0@HAji#l!@nS)6Y0@zGr@j+%MINGm54HYeviWy{g*t@iTY!+3=#M0h2 zIa>cP4+Dm@nXCOOx25tdtbF7Kwdvh(?SKUiL~X$ZF0+`^Yb^>4EWuRO{d7}h0Udi_ zPVv#@IPZNny4Pl5atM2W1oA05yPU2y%0r2)2)r-+%9X!uGq^jQiQQ=2br{thaZ>hV_fk-{KoFeql-Ubjz=&a(P)Rg&zQ6Fg9fhuDvGsf zqO_$=f{aB(X`OEj&N#Fa>qpjL&X?Dy&>o43?k%u7XaGdoRdGB!X}fM>bJ46sln!OT z_fmiG>o+0FN!O!Q+05Z^e+v!Qvwg(-ev-WWl;q;B)3gPGWVYlODv8+PyYzJIt!J~o zIS#me=pOen%M_Z#EOFfG0!-7_p(XcX>2Xz>$%bt|xwu1}&?|cj@4Lk_F2{PTFld3( zkL9r4bprlA9z@%xou$(Yh4}_UkMaIDj?|KO(91`R2Wq&3{^^F%uf00_(|J;S;brfb zvv&wg$DPKZIAuB{D1#*{dtkPb1O69yglcBSkj&HTB;S1z^ZHJrp>H@kteZfe9ohHW z$dg1W8fp2mN>Wn1K)d9>WAOgVIKL?t7wr8_)3kN@vDO^_#dtaHxYI{2!7d&`jpZma z%!4-Hj;1}y`ZOb4fsXziOD)xbc-a3v+hq*l^?q^Ay0({dp2e~mOXEyt`-VVX`U+Ha z3B+Tw`taAv8MN;EEBxu$gn|!-QA#6^`+O(^j2v=ME{2Wd)^}0s;8Om3=SobiY{T=1 ze_>Wd1S&C(X5ylFY}(%d`~R%LXS^=Vwrar>fgiDn<)e-WcXD~Z6ChPlpZP4j@J_NK zt$VCVP8RV{-N?RAX-jFH$4nfT9gW9y>rrXVHPBLW#f3^jwEl1}{hjcT&Inwhq>ra@ zN5V0zNj{B*V(&OdM{6(yYc$*R9#`H!hT}yN(eia5uf1(ASLGTA7oJ99YuPtEH#ePp za)-c6X9eEd{ShzN7eEP{dHOlF;kxh+eDY12c+7)sW;703w~j@pkO!F25R6M1zuM+J zW5+ZvN2M3Gm~Y4KP5f?jsZ%EHUrVUGCx9AH#!|d=30*H<$7{b)2x3WMSUqPKoK=y- z)>n)1U#t@K81APw!3g7+=->BZ9uR@>)u(lx>`d!HrbnSF*T zJ54MM_QQX&0@%My0^ddM!E*J<*dyVFi{ekAL2E0XO_X36XmNVhoC_-b{u31Qxd&;Ocs*ASzP2vxpk|zH>Yq4Lmm$Ms)$NBck*rIF5 zb|9(r{bdfuDpzt-?_^_3L=KuZp2HMI<12xa;!PsGt|#-CV` zfV$)BP-plZoU>d8BgL$fa+L8uoh^}zy$IR8n{n{@W;9_l@FQPp@zefRtjp5G*b_6b z)XEf_>@I?E9?SPXnhNhe|K^%SY$5UeY#>L*!0q@1wF}pwL%SP%etrc?9tdJ@7?I}P zr4(hkmmaG!PW}u*Sfj1NT{&?Yk|eZY$-)r``XG)C)~)dT*KJsnv;oJ-@X+Uq8G2<@ z!0PR_P;})W_^w+)5&FCFc6cCn|Fp@sYGaCV(KI9X`ZI{_s+&3NdWdre{Cqffm3hio|6_q%Y@c6FG1Hjd}0 zD#x$b^npsMOGw1J8n)b*#ooX_aD3rQGL@~PYTfIkX{Sg$PK$C~>rsK90oH;+u;{%u z_%tno=?_$4+BRW`C=J8*MoT#F7tfn;?=^Pum@8ap3r_v^2i1PWVC~d%5U%;0&ek`O zWZP-#89D=}r0pQrx}I~{txNxXFGdUZMbKWk5l+1BFxe{qYgW1ccRGH&iUfUj6MtJU z?Ui7C3T**;$GR^EcxiC9?-}=F*#}TxJ_on3Zgh|E2>-G0FMh2_4}aw48gz*6V?dqB zuw#c68OInAWeU+s<0#ylu7{Fk{SdWqEX##=;;r3RaiG}@_Kv>gZS7q=>pp}aS9c5x z!gk{ZH%avOEQ1*)Gr`boG4+?e!Q+0XQFNje<-bcrpXzGx3{zzOs~|YKLj==zJ%O@t zS?t^+gM1NNzx!K-7R(9D`JIvG5P(hMWf zkWG-Sgx}+fWFy+5=R^?UN^PHdu%sXnj}`pIVFo(CfFXatd(3gFz zn>%S4n{mHGBgb8M^j0l6R9=CL*K(k1%nL|!V*DnPm)u!LHJn>>4`hBpcB|a(T1Qj`#Q)tl_I}X8ZVqcEK^Rvw&k<GlfwF7a3@i`Bw=2?-dw4!|AORHq9fm)97tx)>BeYHB z8!YYr1woZ6xc1T%?uV-uxBP}A_gF0)=4)kwR*N{+hMmJq`5zb~DoTEW$zVLW7?!lY zgk+6vRC#a;_215+&wGlYw5lBCESNVuR~!UV3*p7m9$1v!&pRO>0Kqr(F<|lt)EFvY zTEJw?oRa{jpAVMWZCgUuT6R*n$uU}f?F0_5%fZ7ZpJSYCF}TcI3RN9xQ2S#UCWQv#WS>ej z?K#F>)tBJHd&Yux*I6u{&*oHJtmn3jIo3D71{JFwDCo=uCk;Qid!!5e%U06Y4nxMj zWbVM%N!o-6(%7?6 z8SDPFamP;9g6p&dljYlT-h`8{o59VfY_M#OfUwbU z0GNW)=Ne(J3xL;|BDnE241GSwVAuvHGGuY>3D#GRsh!$n>1O_Sxs5Sy54`kfb@K zs5F-{WeOn)Nh;Eup?JozQ0WWSr=hQ z+Dx#$VFU*ssDZ9a0ZI&|%#1~k&?EPDG?66`w`L{`?mh+212@4vivdvka4^)1_t9gk z_kikvH(>bE4DT7$i$YUIK$`hos1v!=4oCLE0QJoxZ~GW5+wBRt_f(MAs)GgObi{dheF~`%Im}O5>AS|Wo1 zwCC+*iL1D$Szw?7SLI95;@KjXWZo(iGd_knF3wL(LPtS${uX2wwGerzhB2SxDwzj8 zedy-6IdJisDNG7JhQ77QP~j3y7#kRZ8wSPU3q9Gmw|PCZmx{ZEGn&ELV;W3e_Y`eB zlz>v*_9_(%8fv>Lp259Le_*az%3Oyp%8Bf4Fr;Wll=k~+!B$m0m?}y}pO0h(apNwA& z4#GasqS4CpA+vu)$I@Lzh;3KtT_#<=Dq;8r|Nj;uqv$jMcjdVUxUX>V)5$g zWNh?t6&~)d2ScK)p}C|H>Ac;5M!YzQ26ttm`A8A=|2m6Cji?v1Y9l4%>_Q~|TJaJO zI0Fm2ULx6Tb+mY)0$%%84tI-lG`FHCY}1^JnQfIg=DsQRpTc103odxaUq8`3RDeA- z-{8PcYS?a~9M;^giRUI1!&H+ysLN0uonKXgH?J+mH}=(F#TD!ES8=&}az!iF5q)Qe zo~^_a+xFrj)uV7=u;?u*3WC2UL|2#e40gU)fdkJI@C5|tqko|}Wyb+e+Xs)HbcpYI z1!dX3R!s;LBMJV0^<$WJ71*xgtZWWPb#1iOs;(t4(o3#sK_xyA=-DPzVp2 zDI=>1ltelm+Pj7ew`lEUN()u(Y@Z6?qIt zR#8bztH~Y^92n42a|BC=AELvpKO3H)#Y*lapy8MHBIO05dw22?*cam_X8A{)K_e z$K6z78=nWKug!r$V&5~c#1t+x6d*}!8ak&xTFj*dfL7CC=n&`S?{6%||8}m#4;Cun zX~o~*&dH4skvSHh+vb7^@x^hSiXtz6G~TJW5myF`5_kQk;1z0SI6C1R+$*X?e2EL1 zwp<4{yLrPe-}R{cUI|(m?~W=vFQ8pZsASB}2+8v0SEII~ohs@+q))hYB`ppAEA98&FQ#Wpw0W2RgrgGCaPq4&rNmAuVw}`q`-r z#Sc(}p?kZ)^OqOS*xrrKnN*7VZU_Y|qBx~e!2L!a!|TUt!1M=$p!nWMkh4gK`!27= zddw6@hmHnbeh4Ia>4QX4j5ZmqMcWJAkiEtg^de~z`f)rK#cx^-j@!i9@Rz&jq&VM7 z`4kAZMNiOd^VtwmFFGGoRj}{Pc&yhPi!aQM!@4DNfcq-CNZ(A5e4ZYK+n$=>0EZ~J za`q5xk&T4QMM=z)fEakFbrHfFb@94+S72=acQE_33$AO~EwV)nVb;+BfXny5x+U!> zXm~0!b@DhE8$pqdSvfkf7VxHcCw%fwk+|db5IcNq!_Ru2vbcrwYTU>4EDemj#-1ws@-5O=AQohb_AT``(sqS0XsZz5_x{tk>R{o zC?!^8M<8()w*3r7dz!(2-B$=}_)p}Qj>Q>Q#^YOE+wtL*o6zA#7L-CCg1EVlv+jzH zAkm+t^SK1A`xJvzb0p|Y-Wu$rR*OgWQ)F|^+{BsPUVL|WHD2I71TWiRfs>R6;a>k( zIQ7j;^rX3?rVwQkH(Y_NOnb}DXdi{Q8)=9c969VUa~2Fw(Zq)}E3wbI`S53IEqYcf zFL~Ow71D#e;rD6-unF{q8%;KFzsneO7ajpi(>JIkO9|b0XAMfrv%ys19e5i30&kH$ zAET9tEH3%uXQfE&KYfLZE0o~AzJM~d-Xdp>D2Vj?1s9H*;xvzN$WQTuqDNw;Fi9O> z$e4&*7tX_a8Fx^d<3{8WEgbwG(_LH05|;$AnTa8V`@AB?=9Yl z^*`*!7k_Bte~U-qen##%(^wPtJvGMjXRz4zK`sc(?i~=(7vp_;NLHP7vp%&n$58QCmDw?;!3Scn%LT#IAz8EWu>+^4}{erAN2Dsw+VEp^y zCdsBFYRp*8nMl6M9z2h5;(KctaJMJJii4Rjf8KxKRA(pl6IJmVF9ob<@f2xIGnc&5 zc9YbVf07ivP(d^7V_`{oBfghajjNoG<8OI6DC?CIn10*@#ksNgd%py{=|d!59u$ur zXHJCyhf_iJ_#N|fix+Y@au#iMD|Xh{@=)aAjf1`7=YIOfa^&%%TJpWr8;9zw#fKd> zV~3@kNH6~fmO8J39)@221_S}Az7jWfQ!SNHX?Yv28Yrq=0@&|8BXB{?wXpU6cp3xUN4_5kf24>yS|qy+pyYbR|( zHrrRAVI9sAPue1RR`5$QBX~1zTR8{smnh*yJ&G{qF#_!&Gl1k30~5Ld7FPR!#mQPQ zaZiA~XE>3A_6Lm^KOb2hC`Lse7oz!T;~-{PBq)biGX?LjqgnNKpjy<4Mh1<8c)e%f zx$ig%5t*G|YHLNe^+0icHB)>S9s;%N>G&U7kKtr*D%JSp(IQHujG*E@tz>&mgYV@jeDOS$NCdS zf!){?=-Z@+rxwq|zgwo^W4A@#>fh^P9$^-!ZYoCGvX)^#ODEh|Pz5u21BoQZfyuem z!JIO%W|IGEGWX-Xm`n9zQRu1~CZa#jL@bGe_7;0kL382G@8d{K;}`SwQxMu!eHd*!I&Mds~kQ}Ft(jpJwg;?UF;xOmA9d{Nm7VqYzW{G3lvGo&20rwsx5 zfrTjP8w_ir-;CmN#^))<7HMQCi@U@-PL0R;6wg)*go(S{O81@$BJF( z=Q?Mw8fgeUS*fU5XC&CnF+ra4+t50lIIuf*5KQkS!N%$cXgybqo}83|!m%VU$D{D! zjZ5&VIk`B{ZzA|~oq%UfM)*kiW>_@uCKR;~!$l_B@n>y2%w2v5ul1AhoEcSksme_} zB0vpmZ@C7OHH%?{I3JTMRmE=3gJH*hW!&eQfNzCA#J}=AU~!d&Sf}%FWaT*=y{!Oe zDel8TbsD%Y$3o-^IpA&G#&{At17+NT)8qH$-pE=uBnos2%`WRxXe}GN18G!i?qL zy^<}esVJ0_q3Pn@Q%9zZ>D(T{IFzoGY&Md|={F|gDPeo?Nez()wf+N|z!$_8zmqaSJnB z?Az-t>w;B=UC?k_g-t%7$0{#YW*1KI$5!hC@vFtbctPO@@SZaeZ>=AL|1HzRZ8`h! z(ZFso6ReA@X6K{hrz=ryvJ4hXo{1kXs>asu7D3qh{qR2i5crDyz{&d+aN_*_*w*YM z^K#5Jw7I7PLQ~^la#$HWEcpdvEK*_Q<6N*g<^_^F9QvTOhq+%LFG=pu!d;FdagEhX zoLOj#+k?cuiOC(jrb6V2=V-y2n`J1@p$RJ(B;m{nYp|u+4rZPCJG6afG)TG4kaXz` zY?|~EKDQS$S0;UDh_}Gx883r=8Rd{)qk#A94?3Bkg2o$ljF$a7YV( zu{#f+U57&I-a^`!-N9R3GB!FT%x233y70=muwFnE&$DF%MTSbN)--;?H+q;(==}inGENnA>?D z>Q;5Yuq!cW@OxD_KRpSaeDQ;=AIFRWa*(|9XFbp}jodQ$0Ga!vS0jKsJ zgd>0L@Y|rtIJn;mXubFklqW35MNtCtr%nOg>#0CqlN8Z}?Tw7nm|SLE-xIVt|2u-w z*O2<_eAI8IJk0-Eg+A?dL#Na^H2>dQG{UYBH+~&$z{OQl!yz3$PP`yOb$PY(TE92l)Z6&kqNgku@`<9ttpTdlg%SF8|i$HH% z7%~*>QL)Av=4(wWW4vY|^DjFG{d4M%?1D8!K0d*w?(dn|3BHVRs{<2sXEIaO?Ftvg zXZ6*ZCU}tg2ugbzVc^<{%+TKRs8%lt>fV-0c9;30j@9=dOJ^?A9@B@`X+DPS;YpJE zn%xr1khvm<&knH2*Sn@(-CiWM-$_LK9Mb-mj(SZB;k~??wCC(oM zm>0n_kR8!MS~jLk!XrgVUe8J9;P@^ST2_cOqwP`Sns}yR%~+`ZA?^{I-vY}$_t9qc zY-r^)@kZ0#65h#!nWg?1E&QXzUiW>1eY8(vBMlYiRV+w`4-JyIIU(kwmj`p_SQ@gP zdy`S9=wseDFJ-LMjx+P^UovA>L@-w0XE^<+*~(1Ps6p2=*Py?e3(@!1U=;LA8)iRG zga`g6_?EgWv*nix^QgST5L5I-1P(@g?;uMlVEqFrGTZ(T zm4*1g!w)BsUvC_`+y9@4ZLVZ44lF}2P0qph8%vqhpvPR8Y0NBg(2~^c3P7r@;qc>3 zfAkD)GU6ys83&inpl zPTe&`QG4$*KeHnkyxVZWm7ea*0`<9?!&#^hEXV zBTz)&Cv-n~C>X`7GNnjavaWj$bJe2?PCFlg?mrsP*>MV7!Xx2oIDxE|8EB-!12k#g zP$<~0fa;&?qle*Jk@B(EV56sxBVUfeGI2L4-LMR7)Te=ma=0WR={-~WMHLdZOaS$% z)#z8AJ7Ze57GEgOz*&Yee0xapv_Utr7~UA*W+HfkDbi~jBmM{Vidj^P>GB%Ze>GEbKeh12qhP%V82&O464 z-JC?Rj$a6Lb&>OL9PPaC!U^YuFME*w@R?AiY>8tkGEnRF-{{BR8^~v24^$Egmx=?y z&u%K3_$>lG&=Qaio{ARS5PMcCkHt*xA{1nBS9EYEOR~B6VPfO$(PzUPq7Idb=k!VBB<*30@+Ud_rB*wwemh8VZet>TE3?EapLM{^aS3RqUMbQt z87;EXQk~OURxolcooLL}C(N)6(Xkh>3@xr)e$J@6F0zpo8jTTI|X zs|TD^Ctx>e6!<h-G-9;S)!43XmE{yV^@XgJJn7hOcpJW$mZ zThQBh5;BRHFB;V!kDnd`{j0w~`dBl(=B_JFxmb>@N8})@MSoE5!5T0S@7GGZhTzdt zzTj@Ho7lB(H`dZy&iE90!I^B)QN6DQWR}n2WON~ zra<|$W56c9faqPLaQ=iEBpei0<6|i zIHo1~XI#g?wJuZE;9@nF+r9!Xo3;)mPAr7q4TTWdcN)2F34wr9A~XC}o%VE_8k9L{D6HLd z4x+3IVA*B#T>@i=SbVPjS0E3 zA0^MvMeA2jfz4hM;YGq|`0lQU>XU||C-;XSntU0p_fv#Q!(K+d@efi7+XR0858-%# z3}#pKz`@poFuJ}8n(C9_&0@-&n>B#ZXk3o=l=vdKW^3eQ5sB1;6`*LdE9$d#MR?F0 zaQp6nJSO|2N9Sr8n@$1`v#oHC+eIc_O%dIjuZnubbNZ#bDqQJlL$3#|gO~qKz^0f3 z5Fk2q%FtUhQ2PxknYR#f8kT~Bd?oB2pN^5pFxav82OJ80fw%q&#iy#8;KcrDlw6sK z45FRD#H<(!4Wl4uNIg{iw896)ye_*z4&UE*0X}a04)WLmw+$?1Dq`!9hlvT4&i;t1 z`<+AALIvccYKTQU6^!c&fR(${Fpkl~@vW-ZzM~8#cs&I3O9OG-$U~4fARSWJNl%AHi)WLeaaH!ANfCMJXKVqa)dKFB1B47XKJcZp0dMl!@LLK4y)i>f~F z!p!)QnDKlD(_Ie0Poe~ciUr8N<{K=nQ^J|u&%l0PI?6ll0dntZVXrKO-Lf>7b!;hM z6>hFYevhKj*+`j~jbPz~dIM@`cSoB}M55_V416d`hj7a-xNz(^#5Rikm&P3u9CAXU zU|G-Xs&~cjul2yTr!vUNqla0c`@hPyfH5>ApL~e8NcG#O%1G`PjfjuaoAdL^A91H!ed@e}!2Y%g+eUt}HxfBzwH(YXq}Qj5lyFDK%? zrAx(J-Z9C}o=iz=M-b9bYhz9;)G$64|1m?p>997Qy*T>QBP=)gIX0j82`d%U;iCss zVcX3JRC7%Q4W}vS!*2!+HC%+gZ}rCVKO^uh@AEh&@dVnq<|qn(Z;dovOeMEO4ZM@x zcXa7XpyW%cvLs*4obfk3kJihk!LE5VP~NS9LI*u1M~MuDBc6M)y22y~9m|UT;#kn- zN3fPz3OGA>C*!|Q1JBCIfWgNcKw0@0zED>Oy_YUB+b_*w9&e3B%)LlBwkHAnk>A*5o@;hh0aX!GAHz7aWwPAK?xGBKY%@+xsoZjl$f8*R$x0^AHVB; zi?*BnV9uOABC-GBD{0?)j)}Zz#GF4jNa7!=22u55#`*MaxNzM7f7o3DU8T+F>~4aF zPOWD0LL}(wGHXbIGic~~FR;9H0X8|_2j68;5Om`XH2k>Fs$Pv|KX&S~6J}gtvf76+ z`f(LZ^xV1N9k`dt*=Q^>qU3SjC0E$|+!Kus-4AWy)8O2$Eik0+mgIxDGnCV_9Uf0r zhxvAwQInbz$OkFmu8WblVQm@i?RCOlY5)88bHL%RPoX)ZrH~R+1B-UaL&D(mh?7r4 z1L~9EZ?X{_8MZ*=b?wAifo7sV;U-MHITzC@vw?dTjar%!%)RiMX&*3#@$#L3^%V}_ z1>J`1@F|0_`lA8(9TEFFYc}EC>6hT`V^_FQGzQh5v=h(8W_YsZH|z_&fG%!!U4%$OK;%WlPCm|znsm~+ z+CGV?el-k)O@-+0mWOEXt?^Kv<%)8>Balt6p7;zOWBP9MC@N|PaJP2jv#RoJo&^Eh z7+;i}H34-u_`*wZYhshqIRq6ukZZs=XwetXTg`XK#(oq;#E$_}%W|k| zd{WMIeB4xt-S2Hew!GOt07jL+(`pY@?J)^U6Tqv#~WGomPoSVeI*GSoJqPb zBCfhf$}P)&#ZmWS?#G7_BvG!2p1^u^k?gu`Ve%~!tuuj3k?rP7=4F+0Kdbny<5f7< zO=h$#HHCgiV4WdPGbR_YtjvVSX}#uF zH~DfgQyb}bD4=G+*QxsPg~DXzBB5#3S6Yi(aHnxBJW#`uW|ZP6dhB zyv>$aWz^8DiXYtAF%P(8R~IfL#*rSWJxE)gXVdO8dgNI4D?0t`BD#dm5)?Y+gzyY| z;rnbWY2ubhsmTNfDY=+IZWaZSHu>vhRZ0WQGwC2fD`p8nDUZ353+deb*&$@Jdk5LD zK9-wr_Ly^d>qJLAY-GJ{0|?H}WTUe*$jVp=(OVlp%V)AeiwDrkL3PBZERuYh-e0=w zLN>3p&x7mH*W?z}mXN8qkFNS!By1`eBLuFkqLJh*zbkweXZYX}k-jWqm;Ce>TD5A0 zDav7jU#PJ(&C^Kw*UgL`8GC@*F1H{(sgeAvts|s<#ml7O<2vI!I%jlcfd30oIB-MCQOO7aXl6k6O+^wUZ*$Kotj_pNbo2a3&gFnE7ud9) z`=Js^PI%AeCUI%N*+XBsnwM;J!jwz28ajJUon?xMj)xuG_^; zl{{syyPFe7hk?ZFx+(1)cvgn@`LdR-$?)Wf0UeN;KsIiiMBH3jWPi*Xxg1A%Y1b$% z=?%qa)KKA!kQ%v~I@li*I=1~G zxwl4<0~(fGqoESFR!&Y3%nXQK&JLk?W)<~Ga~IwP8*^zn=h@uWbb47ik##m$%RTyQ zD>b)L6#k^Q_|ThbT#~<)n6IfPIOomA zYP#SmvTmARdXnhgn`G}ROD=nwm7u%Woop`l;k{5)HZ}drNQ;mo8zQMxE`oH}v7mU{AnD<7~ ze;wzz<14pHP5J#aqSuD5^RcG?)kaGd-?VW>mG091%hQC`tXQEt$d~$e&*v4UJmSXq zYvJa#yXm3nD)gu409l6W{7Sz3Ec>(Imk`q4P8OxH+)Q=`d%81E7!;8#T-0_Iw%bgg zZEk+N>+(aqo0c~ByzHBxJTyl5J|8ic(?O?W0_F0dt zICGRXB(#%%T1oD_`pcPwI7tmo&fvbJo#uZ!#_-F!uJAdRHgg^;^J%F62JWoK z7^<~CO_(dwsBHXuhaEF&2|s?pd2&xTfO=V)2>-gx1kKb%w7u#JZun1!Up^$AUZsnK zvq~yl24~5C&h?;LPnS`Z6V=4&wu|uP*+#*C`A}iejywF7(VFzta7QlDFN1#_wuCRO z5xBhr&v9R3R&rs>tb~>+$4FY*KyJaWO!A=duq>h@jx+6c@7O z9>>lrErc=rPGQT>>vT#}1T}Mtq%PiTgs?h^Fn>-E>Hp7yi{15#JT~s---YiM=B_(T z&#I>rkIvnk+x{CwU41BZJUl~4_--I=U9Q8`?NH&|9Tt<5XMf1XzD=TE<6OAyb8MwU zt@Y^TkB`Wq+Wy>s_Oqoq!PVsJmg!Qft3}j6u9ag8Anv`ocMf zC4w+cmy_>Q5@s5!NT2kyaZb+H`7ezD0Le>aYo}?m&mJ3)DcZ{*V9<_=&{MBjog20! z_s${SP~$y6?FwamP6Y}!h5EwZkzT^<1U-5*Y*>!Tpi4rJ-*EK*}vxMGex{< zV=G^~ET0?cy`1|Vb6M8rzMPGS$|alc+H>>xcy3OHjr6~wTilM@I^5ZjPC`G#Qs=gn ztX!ZsH?DOYCk{JEbb|(`y6hEOGH?!6O)uga(h^C-rnC4_n5#hbD#>UcE8(oHm?Te2 z#Y44*@)mzH$vfAxY+LJF+1d}4oO(27C~_^y{^Zs;b&_>Wi{HI+CYid)gd320nhSYs!_6#s&z~9b zm)L#-?vwWft|TOp^qujh5tfGBw)9f6Ve$=HH}NUy$r{F$pDtiiiZaQlq-a^WT{>?f ztsr;%+f~E`A7&F4w-M!Y7deILmgMa>c}{E0dhVBLAQwD+jcor#j*C?N4wJKNRFp`rW@tvKtdNRqK zFo+8)8^?NhJtLK$u98!Ij@*L>Te%^frNpCZB01Km#9b+CAsa7$C-p8*NO?~(+4klM z8Q15;Sq|*a&D)kpc62;t1KsYENmGm2*l%mtsF6|R#I8te61j-;m!9F8`X!RC201Qt zkvv)E5yH)G9>n>sI>fooh$3cZOKF{JJGuAYkP6+M0qkbK58MeOJx*TXH~LpuPPIa( zbMyWclNF7v=hk56~YeYSxl#HsdWe?7)Ww(q>fgmX#CW#cX}1kGwOogDN;yKNI-KS1o9PK%k7eAw?MJ!GuA903Pls?m9vVb} zIg6`4U1I;5Z{g+g-RY)z*~IeL0n%&k=)IMTUxi6MwZ+o&%Y`0+g|u-&r1s8Tj3*pprgup?TZdfEHq zv;>bkLj?Uzd-#qED%3SMn;qwOjXpSRB>29u<~1vASh<>Gd`d+nJLUdA_MiIzoI8pC%=~K>oLcl=Ivv*zI(;i8p!bDU(5Lmx+-*W_A6QV zrw_6n>!0w+g+AOyoeS)j?4kUvdzOM^aSdG{)?U*TZ7ZCe9+!U``IuE$c#)lJY$(*d z)fRH(pAg;rzvKyL$0bVLS=+3Y<-U6lh~DI1Y?WCP$^P0y)+qX zvz)v6*}ZePx8YA@KchPMr>KoezsuO3$xigj#Pe)Nd?7o|w;%hv{<5%geUr@1L=9^M zou%x8X~KkuDzr04Te?3&LEy^Q)1zlc@Rfs7=!IuqZ1wGEE`OdGTQF)YnQk7>X9YTt z$JS+ZO|3Sy9YFYVR@%hN+l?BnE2KX2SJ9z&ca!U31IXwxV@RQ7JD0h0wNQVKqo)1# zu-yl;*}u&dbVb7=x^mJ3%9(4*hUO{ocSou4=Pd@w(EP^B z{o{CfK(XXMJH$_$T@HHBj)W^9P}5fnAlRLEQ`X5(Xq@zx*H_+!&Rc2Z}wprDgP%EdF~ciVG5JS7XK zPcxN%h%e^4Q3}@?|F6PS%SRY$)=XYJP-m~FY-FFCFtlc33C(>zP;mVN!kibz_!+0f z&*^Jo-#{}+~&)*5O%;Sn8? zIf%O$UO*Kx5-N;ZQ!3E>Sk9}^fPPz#rMA(L{5|J`eDkrD{Aa_9d`0wBL1BA{aBRFx zwsO~7-sq?(;2NxT^_izDU~BgbNDP_l;&XJq@@)DbpB`EZh>E2+n7E3&8jIYee0;SY9iZ_PVqX4YJ%fWQOvZ8IoqdZ;fBay+ z>go-?942s&^skdEe~jsbjv}hRZVB<31G3SN2t9t~FeE7nT%JiTeYo@|NsJBV>NR$A z-|DAuYlmrZe*W&lw)exx#PEF9fe=b|#>!@=8Pl)d)abY^^1_%M7UWdI$jZ|j22}R@ ze3opMA4?X-$q8FtwhIdjO@wdOMFO5_EjSy72t%6J(&1*0Wh-s<`OHE^Qnk@jC_H;p znC~>Q^7tArp_^Qx_dFYf(+ThCkCZ+bV{(Td@idEj=_Dbne*iUnQAuSf3HVBs2{m~8 zkxeSGlP&aF%Bp72JO>5)?hs7b(Wvi6TYU-w}lnPL&g+APTxQe&#V1_w*I((HXy`FTG9W$Y>Er2F|N)*9q{xE!B#c_Cdg%%0L+ zH-yj&HJ}qP?8za-@^sE%8wEHT|GpbU5n|x7JVVIaV)>LR7ZGYt1TRgY~uG-#S6Ed zy`T}t{b|9y<-*&w;q<-ClO}CALT06&A!&Zw*{8=8Dz92N3WJVRQOh^lwCu?!a$=?l zEp09k?k~U`K>3&>@mmk;vwV$v! zN=w*TXGh1C4d7PY8_e~SPbWE}mkXco6bSw9NrmI_GEy^Q20P^FA^!20M>N5}ge9Y# zW%n1q(12Zu@pL& zCjMdXzrVn{FD#=!W|{~#&zrE^{%FRb={(7Ql}D$CeWcLQ$tyn770zX62-RyB@CQ3% z$_K4A2Ezu&%8~qf!SG-#J)as&>c8Yw1jN56pEgR3{#de-x>jz!Sf;v+URE=uSI6F` zXEt}z%i*5Fif!7$vR_fcv#LaXqhul3R`gxS34Jdp+tkr*Y&v~4y_GlWa>rSjt7*Td zU#WJer+As!Cum(=E3A3(Th_KXn$1a2Cb!KuW0zs#e(yIIcK#|&HvHu(;Tu1Q(^+ti ze*KE`cZQue_Y!*~(U-IbVxit(Zso1nXQ^U)x3Jo} zLr~2I`k><$-^V_K#XX$SNr=v})%iA?%4)CG0#xgXijSZ;EvxeadlIb3~Oq zUAIek@@u?|KYfxcJe5HHb7 z50hz;I~LBZ%cj4-pQ5QdZqj`b+2nJf6-oQ>AOG4XPlz3Gg;u;NB0KszK*(BL%;#-rplYd?`4elF(+AtW z^B=48D=a;`XwJr2LfF#&RPVxB>g-oZ?UST*;L;{Kc8rH$W!xe}_8U$k@BJekX-dMy z^|#s6rJ{Q_Nt4}Cw~_@^B?#R^fLL2{AXulR#S-AP05D)Y4^)ZvFaqtR(@v0Oq zd}?65*3A*RnJ>b;n;PumxGXaD-Y9l|#6kX;!&CT~on8LS*&97vl`3o6*ixafvJ(2a zmr;H5nu{%NTAbE04>WP7E>ADLV+XwQqPMe<@bI)3B>KIt(6!d$VQ?2eMWX{-o^%u} zi^@n(SQY+w+Njd~nuU;AbdM&h%@DlzZDu2`D$s5JMbo{1YIwClR>H4UD+E*d4}wqY zWV-P00dlEGLpUh*CH)tjrNMiLlMizcJKSjx>4GhE{H!XzKzS7Lzdx6}eC-%wc<+6w**+d9Ta;Nta7EuQ)bK%#$JN(rpChP`}32f=^J+xi$CjE16$u4X- zK<@4cp`@XcA8KbO&N9Z(9~*wsk@Bxds)-sGXly5(I29*MpRQ85rB+2)S8phs*%m}Q z9@S88-C)+XtxUY5zTyv??d6v^Pb19av-t8chRk<*N6%Up&`s~2$_iHo^Op~oA+;Op z=+uChh(e7b(W0E`~pcr`Wb_3-`)@jJqVijc>2qQ^~EcWN~oxIR~Kb5tN3J3rF3zePYzC=&3^i80$TJe8DSwI!Rj z7;$R?uFyTfC;2h9?$qSS0D8h+%6a;ya97>7@@C=k5ND%E+#03i+psQTq+-BD8CS8z z<|D~8=P7hx^=e*8L@(L2m z>#^z+8OzErh@wDbm4t=xSpOj2rCwmy4Bg@Kmqb`%xg?FC^3+LES?&h(4 zPD}9xxlntGmCrvw?!BlM-e0X{I|ohY4o*oRc`^q+d0G(FGkHrcrdsk5qk&rg*hm$- z{OBW9u~t|#k>0nwM&I9#`5%VP`=Q1!j^mZ~UbGOYR4UQ9_k7Mxq*7M0ziC<#%BGOE zmXWkcgN!tky3aZ1iUyUCh?GrKWMq%%d;fv^%X6RSoX`97dcWT&)}@5TEe)sEchArw zc~$no&4#VayUIpP>88J*jfYB|Jos$o0t@<0S(%nGn;J5WzI30>@2Vw4B8;P-u9m^y zg@@QQ&7*pEZMk}Ysky%0z61qOg+47 zAbyc9cWP`oeXe-`o`yMz_ecr!(VAV%SLzVuB_!c-hNG}at|>FL1<_BzhUt^au6AxGG_zl zuaV1dBt-w|Q}$!aOC}Lo%M`traHXDx^uP*PdfI9Z9zB=^m+y=B&MyzpWw(FP>Xbh; z!+AT)?X_X@l5>#4hEKGCtAwR`e*E`uIa-HI#Iq5gKN^CJ|mc%x58 zC1jJh?43M?&u)}n_{_SO1%lBc@v_0zpKOn-q3OR~VCxI1%rVlHrOM^8;cs1B_KO=l z%`HcG#|Fi^(g7uQXQBa9K;rY$&rEbNV5TVIVJ`eh*bbQ+W|CwjW9C?#jBjUNfr)A3 z>9^8W`cYQI4|FjigYGhL&H6r9_FyyJ`pbwoy`D#U{ARMVq4U@QDS13xYfMj&HSFNS z+br<%37U{l!cw$@+1Bp^aA2^6-FB>GW$pdEj+kP25#da(DSd=5Cp9Sbv`5K%_Or?- z@0n716l*U1j|RazQm=fReN$#Qy!<0}3U>lWeifCPp~c4T(Z-9a4FoxEHrPRRG2FKr zfq9+^P}sNv?XPL4Vef-kLU=hlKaAL{uWs~TmXO8Vzd)x>O2SwBEE}=rRwuiahPdmK9gnQhefbQknj6jbT^mUSzJ`p~=}Yv~ z&M~9WVNC1yUFz#y%#x$B=*jJN>}_lU^Nv)(n|*yL|HepWPRC;3B1w8@%>k4>X$01| zt_<&AorNxy5o~5bIeV_;B=RlYEShqo8m(F)g+HWAg28kRViG--3_QKbeHeWVT6zaz znZy|US7WK@)o*{%wReRgb?3RH`nxY(;lG#7o$E&P>~4xRM)4mfTf{Z;+Nt5Ab+l^o zR#sSkfH}EqiQZQklG$TV(PwTQr2NYjuF7K{^*=O~n6j+It zFZjfke$J+g^^8QvsFY~JtQmOA9SO>E*U&3F)bPesWsELc!i&8@@Nr`tY%6Hs^m~Tz z-;Y1=z1|p%hWcpF&q4OBbPBtWV2h@9rsMEROoN$@XzBg}7QA?j=tS8#rYYgcW-sgm zSTM-OyLhmJBUXw8(^bLVzXsMhdcZ-s_3Vm^wa9vO02^>v%(8-NXx^A%I-{tHeo@e1 zo=fJiv&qK1HAQ}qGX5#`%hwbo&Cg&{R!(ERdKy^u5TduI2J){B3NVs9O7Es-u(=F%gymHWpWVRBGzJuZ`&}RrSoaFbRpI`IEBf4zrgmUnd4vd0q1jm9!tI| z$BY~HQq>2#%;tM22}mDiJ2d_a0;QbNiXp{Sr22 zQVjEJRTpW@e?^75K+o)OA^O~HI-^<$FH9GaO8Gv@70*GR)=80>(~XF?bU9>sw9~Ez zTPFQSQx>V0~lWG&&C^{c#?@n4{_Er zZ`Nc^kOh}VugP6E752_h0SD(CB8|S@w5ukMcVd2|kVM)GL*rwq;S_VW@qHcVq_vZl ztBpcaCzo(tGp9jR-Don$G=ZD;=^HYc`4p_jRl_82@jj(v1fL{dg`8c^v@QA&R-En$ zAbd(On?Ns1`q8-w|8UpQ27KqZA$-fWfDwz-uynU1cJiD~1*5b%r?)O>L*{mpwQU$k z+;aR&`3RD-jKj(s6|tqNHQhNZ$7ZWbu)go7QHV_zqmzD2sA|oGMr;l^@u}Alhu>VXc zr8}RX9Pe>>?+iI;T$#^VX+`1D_GwU&UO?t%y>&=A@JHyhLaf@Mn&EuRO0xH2 zET7va$==+&M<(<*;t;+J{{1hJI1Q$Q{+CK#*4?dim6&;wS)|La%GpelCPv~nueafa zZb|r?(G}Xbcn7UzH(4@Tl1` z?CF|)^rPc(HnlsowgKlWKMfH8I z;V%DZYBi(-52yWt-IfbrSYaNud;N{;o?JoizE9yloS(?VWFos@8pk?9SAli#bl}xr zf@40@*q^KzMm-nMW_~KD8H91u)s~Zg!SgsXn@ybhV@())crHv|>Oj5pcS5F4-#ZwKjUf?;7Yx_q=oPN%ft_(82nn~*9!h^p?CZo@qDlp=L*gAuR5F=p-NRpa>oadUFA~H9 zBzY-&ovxI&pyAVRB6}1~=N|WlPaoPK`mj2LN*xd$AIT%}3-^&9eP_AfejY?Qh`BmdO9E}8-*%*wo##z&8!j}%73Eb$()IJt*vY;?fS zrIba7gn2^Y3^VXb63=KOtC&USMVxy_8~!X&!*fsEA|8($L84y;{Vj5$qw50Uq-P3T zEETicAFXiFeK{zbdj=%5bcoK%F?8x+8J%JoPxsmKP?W`L7|hS2rOWpqk7Y5mcI67} zcmq*<`OM_Il=dv?$Y5G@O?u3bc+91)5ESXSg0@lD6QW zGfU724ISzyh3MP)Q_+re9*%h@U;+0mS?YgRXpC^GtL)7AE>b}322F&e5;lB?6HarBTA_4drV~R6knT~5a z9bIKdjBJhJin0XzVx&*+KD&pd^Eux4{P_rt8jmy@M!*I$6~R!9rl7JDV2vRaGh|Kl z->I$WaQ#VS@-3Z?6W`Icg$L8>@iuf_jyKhBY9Jj!|LD+>E(m_=1%i|PoRT{suDjC6 z()4)N9(I{V^(3*aNgXsKZVlyUnzFF+{bXBJ8d*GhDLP!ZiF(~Erba?z3XVKdoAQ&B z_BhL}+iXnDY+~uD-5M^0gc#@GWFEEx`%sTO9wB@QK z%%}`uWPB*Bwcbm$k7_YLxlz=j{jpH7PL~}F9)%==K2p7@pIprn84P zKt;?$`s~9Ndf{w3ZU5xWrC6SVB@>P!uNOc|4o$+UGi$gVGbTaJCIykJvoVv0s&KE;gr}2P0 z-u;0)eLor<2@;aY_cT~$uL*m<`vn=8dx%QqEGD*%dvNsf?bLBeB3Y7Fi00SbAY-Z& zXi?2!+FaobpTO&QXU{$qN&Vp;#1lzp#x`j zhVYBuj-z9*eW*DxFa`!94^cI%BXCeQ8lBJAV2bQIsPrGA&lfk4qR|g&W#a_AUZN09 zT#yHq;(2#yNQu93jivy*p5T{fE3xJ{8~k^1DzTixW6vCapoilQ!`A&x&{~X`OX+3S z^y3Iivs@&Kf4E!Zl+P2rXw71eV~R;8B%+=fJ6MjVB29gM0eXfk@p6F-UuDZ1Bv)KX zuL~PU(nFp;^lmWe^0rvH6vhkS~l5Py@kKwa1z;0h-Nv!u(`Z2oja*Cu7s$P0h$3(jE9a#MgPt3XDS(LY&K5h%-2 zT_>@*?y0Qv&U3zvLOy9*n@7!bjITTs8_H`-uT47>ft(?L3(UTX4V8?sHA zX82=rz$gk$#fBoy=L2*|q5>_MV+JxA=jp~1H{kBeELutftc zH0h`|8+d_PcWVODuv<%V)nmA&iiPNdp%eDGJPBW(bsvR#=R%%DJ&f*BWbv`DsGY}W z$XxlJ+SMNA`Jw6L*28Le|4dT!`CWo2VaZC?S~rE>IP?e^7ev&2Ge1UDqf}Xo<{p|7 z?+0V|If8!SRT7={n)cYZvI_Smloik+ypViE*cg0(3_c!0&U;@nVO2I;Fg5{rtBh#) zvQ@0{Z2!3;QOCzQX(!6TsAS2}qu%aPNdUj?b0C|LPiGQ=%jJnevQji8a{2KV8`7^k*n$ ztruEhKuL=WVechvnEj#_miR4(4ooz~RCNoSvd z(i+JOk0e$3ifFv~EZeWT$4OQ2_?q84%{aRUYY}q0#jn+}=lN%saZ|o%6D0|4wD|5* zdQ|kFAwx`!#vJUrbL`~ zCPFzl3fhbJQGH1%I$kLk)vpePU+FX9wV)2(XHLf+zk&!qcq4@FcnzQ5j}qqsEXS9Q z$HJxQIe7OrvBxTOn_XvhpnXq<6J+h4i_`Ke;K!>N_$|7KBhM=dY<@Tk?E$w9NlY`{``~Y0Lcq9I& zDJjslEQ2k#X3@uQHqj5qp3o_-ChSU23Dv0_K}B6PXx_!koSiTi9eX(hQ$1dRPQ5zR zUhYRO8Me^4+=#rGQG)*#`r+>r%RoU@mzJ(N$C;{I;c;KxqvRDT-Iu}joLffBuI0chDOdQC zlK_LHHW?{+;q(rETj@|<{zJO4RtHO1 z4&k10M)`w@HDjoOQ&e#H5oV1JkKz6)9?WQ@J>0j z-t&~ho#PsCbs4}I(+M=UN{Gv5n$lS>K9SxhCGaonH&H51WM^VI3@k3&f7 z%m@e%W}rH)fSkq?V5a$EQvSV#xIZ>z7F9>EOH~=R&l7hB`MaW-$*H{c zUbggbuQIY*z6$&yg3?SoTqq*|AJ@g+NgHjxC3Ljs)feyun@OdZV$q_Nf zw=XUOu-{I+aF3zD!Rsv?-D8Sx{gxFJKYWc2=P1#LZAmQTZ4_8msdCrb;z1NJ5mK}t zqW^Z5p)=m85U3c1A}5bvJC2^B=XHE>s!swlo>+v0{q5DO9&ez2b`^XNK@nM%w-lsn z5Y5XwOjqrbgM?9TeEn8E{A8Uq4%YdBE{Zvtv=fJE=1E4vp8N2JJ51@k@>jf*+Pa)X z@Db?!yaI>b_CpcO2=<(r%Kfs~LUg{V0dnv}^XIN~ke&0K*X%ZmhWAAAKJO}l*hEc~ z`yquabUX-KGZDnqwSrX}2R~^9I_Kg`4bM&kQX7MIR?DN3xN0P)RgY8tG~(JNC3s2h zQXCYy6Wh7^;H$qT!D@$C{J1ay%$BR;m}{wc!RjL1yC@Q8%vnIBjYkL;jH-s)n~!1t zJqZw|sD=M@`{E~l$HhCYN{r{$;+kkV!Q!y7Fi=u}PRZTj*AWS<@p2JP*1wKdXPn?6 zcWIRVZY*p_*#y{EUGOH`N#Lfo4*!055?xYs#gA0nsJi_pqFy8inYTNsQf?G?D&LDP zmeD2QKzvWm6X3w^MA-N=9!`cd*QiY}CU3t5;=e<#Sp3m~?VE<-<*pr2(EvEi_%!w_ z)WeQ{jA6&telUBGNjj39k$u<>m{cE17mY{M;h8>^-KZe1G!NiM9iA|%K^3zn%JkN^ zEi~kY6juGVPB_Xk86E4|LPfV%k^Y)Vpmaczwk#ZjE85#|kA%yDpdh%Py zsp^2`fA?d1-wOP6k{b8w@;mUz3xycN_4t;}F}{~D8XGxm!1dd1fMQh^JhwOu>gqpu z@-=<%qf!@I@HVb*6jGwCcQxtoG9nTmk zDDDg3dYzr2d1C{~TRT9WUh^ed_6k^PVmvzkO#!jo>+rJjF`&#=+?N+6X!Z%DA8t;> zk`?(Nb9Ea|TdjmQE&U5FMMbE&+KtS3W{ghH`{`ivG#8>O8n8w0Ji6rTU7|AS2JQSV zgS6}ou)3{3*5s-3bE~A`q<1V@@h6$go-~364|Jj58;3cyO($Vyb0dmOy9`Dy>*;&c z2oBVfy6c;eqGl2^aU|Act9nhFEFk5^O#V_k|lmG zpmGabVAkyzraiNnd~ICF%N_rorf0^Z1!it!>V$vfkB5$6dFCfd<~PG))r)v%R4ZqB z?f~macM`Risfeu1AJEF$9@hW8mff1BhPFxApuYWLk9ENW6egw38T?nz#)V#@7Qa4G zy0H-*`)WtG=S-uYrU#K2_ur&@O*M5;WN5(xinhn;}y9%%{;EKWkz;>{%eVJv)|WCQB|4szi)BgsS6=lp+jwTaikY>4*L z!TJ09NdM%8bmzf2Y?se^7`|r(Q~mYXxU(m)v~2_yc_NVaXoQ!~F3JclB zdB2(eZ4WloWkoyReI{u;D$x5g63p*ApH&RXv$csmO#ksMHZL)m^^@i7$lqBs?~f&e z(9LZ5zjN$H$r~zIIK-^f{b*039v#xlfd9JAb3QId9Y)?^s5>`}syNO@vpZ|hzEc>e zbQ~=6!Z_J)8vG#Vz$BB6Kn79 zDB1`t8v~C+nKGRO*&n2OqHx-!Sk96i`Udfy?^60pU`{;PJ zFtjXuB8@u#go-X6XFrUjSi>77w&0#TCzH}kd&qo@kvC0kc#aQ+O`tQD<*;@~B^vd8 zH(ef(1Wb2`IflNW^Zgo8qE$U4AFhY$XSxElcc$!*`waTbrGhgXNCV-deoo(HJpJ1e z$>g4{XVTjX$nTmqXwlyf`E#P_&d+n1z}tm|A9bU__G3g|ksq0g#Ygr}--I3b5=$>c zc)<0csqC+&1-UAFge4g{3OngZ(*9GMT(v1;&YSF*d3_ai>+YnzH8ynlyKB_0*%C7u zFZ{~K6PDJj!ml? zeCeSgY58{uR?m3JS}b%#w-ICC-RIL~$Hvf24r}qU&K-2)Cx7aHTaVRrdGno~zo5FW ztU)_pJckFJrqhG@H29Jpx;$XWUJor5v&Nn{r`Lm?O;Z>9#iiMFK`MM)rq8})%HY>A zuN@5A&yfndD4G!!&$*q8=6y;u64|%8u_yM6nRd^5Ryt`8ZLF#1EuH7d&OFv;^P_fA zCu2Suc>IiXr#6<8AL+Lbe(VQO6tC>EXzuC{3uy-b&=~i$kCCf=8v0Tb?e04H6neG7?x* zkRrTm&7ce1P1x6-77%}+=(|n};<~Yz!ohUzzgBIWnjSu}o+$MmgHd~~lG7K)iv0>B1vXw?xb@{6Z183| zz0V)ZPIoKN_f0aeS5QJGwDgg+`EG;-7Ls|bitt453m)NR&6l@w6e`iSy0b zL+dNcaDqxVULTpyJGIn^TU`*%Nxsu&U%#l46}_=oZng`=9j)YqZ%kRpzZ0CwpgagW zJcP-;F5J|fKok(!$ito%tl|84h*3ES`zkWoq3~T|f59*8og?;-g__bBIfZWTHRpz z=oDe<%`f~-@?!4$d=I*+kx%3H#L<6iI_UoAGNk*ABHMU&fX1bbWBJe2*@)~eF3m6+ z3YK4@O*vKc^aCjt-S5ff%HE|4PIvLe8Bv1nzle}UpZKjuUXhz6-f%g47oJqB!D6ik z=!Xs-(av*&67NyWY>x}NckczE9#SM~XapUMjz`Exic8Eer3F^+=;!O&^zNSqa-n%4 zkOp0P+C^5-*MAe=dS&3y*sIO7#CrdlC;9v@&)?CMo8HVXUJ<=o(<5}z_Y=y_(WJAl zKVqKqw=qTAgOkT7o|d@Sq)S=k``7G*$#&l6(aqfClDqG9|4 zDZavg2jD@lm;vATAJ*D?0$)M)ICxea5r3SaVNw!oo^X!xUsrM|jbZGrY${2b8-W5_ zb=iV3HFOGeW51>&*!cV;rtPFpgf-FFfm+fFCMkGiojSZ6jKuT(?&GR7CA>Z@ja)4F zOPa+x`Mf`h#K1heCivV)F7Tovt?xO*k2|T!^31l=*|(MGJgZF9zT^iQuRN7h-Zuu> z>Z8~p!d+05PV( z^8{es>ImPP)DY@-16i-9Xve~M*jg0`J3j5fTZ8UG^!F^N+L~#n-W`d~RvE!5LwRzo z<^x*ooJpq?{vnb(ClS%R-w@y+jq{JMAp5)ip_%Q4ylc6?jPH+?w` z&EGv1WuCRgfy>9!s%%Lt*C@iD;?6;O@Csb;YcEwcwS(-v)!2RHO;8E=1)DV~-chz2 z_Ku3gMZb5!pKnRHR_io;o@|9j#}$(0a}QIgQQwf4V+2e-{sI>td5Q0Slch-a2wzRV z73L3zW22diK>y@ftTAYeHq{5g_ERRHd~r9PZqIi!+ zloL3}-+Lm7x?IWu<(+|GlBx}tfrKv`i2(Bt6D?GX!X%HgYEFr1VKW75UhV9kB)zs z$&3&8z`>xA=mRprj+L(+)V`;J=IP`3NPiTx`_96K;c{rM#W^Z{a1-#% zDzL`t?RcfuIoKEMjun>|!7qvTVEX(rzx&sBU~|twW#TY$U#5@eqB*$!>qPt{4q)y! zb)4A#jUVw;26K8|P?7%=%xs&`r14AP;i{)Nyi{8d!r#Oie>$RpITLtYGBa^gwKHxm z9EUF?ti^}x)v&AQ1pM-x3l97|%-?w~7Dv_DS>#dsjZBJ#z&+jH~5)OjF$?3Q=BapvWzY>;+HLexn8GOW4 zabAM&aU3sLfL%gj@P(*IxW3IDFKko5!x3)S*D4IPj2Ht?ew~Gl>-ku&>M(ZB-GV1s zA4am%5zZGm;cdOIQA1uf1nttq?fv4e-v}OrzgP(^8tagCh%0dSH-Z0=FDSJ!Tb$K$ z1M1K0A=+I6I=_9G=!^3^HfSA(d&;4F<*|w4XY-NVP|QRx*(rW>L#g;~I2k^gra;+h zX{zEU{+-VWQ2p`;v}&e-+F8u0U3krZrN02qRY(aAN#6pRGz;!Yrr-;uA8EJEeAX&@ z!<~1~;!HocppeNvENq)3M6OVP0gK7(=jq98PWKz`>R|$f>*fQGRU=uxFE{i)mv+v% zNZuMD&TFHv8Gqh*P5l+miRAoWoVrJ)#{~RX4+q?UU%3_R02(yL@XVpUCDP z$o_;CbCdDp0c*P4`7gfUB!fNYX>(cjA1ZgAEarLTZH0*8*DUIHHp^Rml)hd28#%Rm z^L!1RQB+M8Uji#}?%KyVeZ6}6A#W2YQXj>atZT#hf+e^q#ELG8kwMp2o`XH&dD?oX zKR%UjB}kPE6#R(dWB2XuL~BAC`f}HnsF+mow`dikFgKCVS@t>F;&hw%Cppnm?#t;j z+Jt9J_)8Se9Ayd~0#x9T%1+g71ZBZ!mi5&_B+F}H6F>W~wUL)d!}@0G5}`=0^%{|v zq7dGSTkSM`YzjL3AO&yCdx1XIy@Ztqi(ske7ye7X3HY_N3`Fu&vB9qZx?@={svn`x z7N=wgloqv8te(|9chvd8WX|it}bVgC&HUv(r&a#BF$R*czHUjB!$3 zJ^y^bU2e>gLr6PlG?=={;_OY1Y}$e=R9NW1!mh+JzPNQVdO~zm7zeSFQ25 znRnr)p%=>^Uczp?bf#AVH_}frw;?T~ia?SO;Y(BCTzmz7^QQ>TD_OINM+{I;vL}u7 znSzG4zotnR-Tbo6f6x~BHE>hrH+*;zPwjN(B73K8(C_$!jya}AG*%Cy-X|`cch!Gr zHb?_^bq)-@qx8$$3)JBBJ~}gdDV)Cc+JW~(JnM*#puaaKL+;~VLjQ$h-F`K({+17y zCwa2!^QQQ%n7e3?X@rv2Cj2-v8?P=tOxE6RCc)=y@Cnx~H0z87XsrCqA3S%Jwb@OC z0+9zUx^6(@pMU004YYTd{m7O-bdNcpKo@@4{r#MZ!m}DLrR&_5iL>#j^Jl=ylg-KBDo}-7o>2&%f@x8(}k}Ne^ z2;l>*aO~Sy+~5u%EB_mr^d5oJ{2n~6{Q-?Uy@G_D9mguUSQuf|4Uv3t2Tt*5u%GD6 zsiZ2Cg8>uJg!2}}^7JF#uhpAqgl{Zb^>`{3bfuzOK^M@)JPz#jOo`c#GvZDuUGi1; zETG@Jao)_S?AT8UarV|7;Rk&qwoSE*j=pmlB_2CX=GZDj)$bDM{(#vR|0jH#1Fy*P z?1wOU`C)ji-0kp8fN4+}Tzb8ujT&y3C42P^u|ZlXb*|E6n#E~scuzKKj@U|0FZ++| zj=f3lzZr?;9Zf)$p2CZFi~{%jQ@MS-4tmJik&6$V1DE|1xt~YIF@rfK>{rPpwomRG z>fTzwY5VU%6H?aEk^scNn;c7~N-N?eX%VzV_A~uE^CDd%SxX;nQ(@$82C+JF3Kght zBHyKez6=Q=5#`!g_WUTCDUr!GelI}gJBz6LDluc-Jits`t7+$PZx%RLo4V??P>Iir zVRYgq_T6WQblOd!Z_ZXyUVjjJw$OtL6YYVNo<=gee$fAVs>zYSZ}gjO7Tr+xnpP*> zAjTO^EV8y5&Z_02+MtD;PsV)w;L=HM)d?}5^H=PhUbGG#_N<3TFMTnKZDmUprD4`v z%8nP;k`DjpXp=`JiD|6?Ur%eY#SxKT**@%VRwwmHc10%EQ|R;y6~cDSE)Yi$lR@>< z!ei6AkaM;#3%tFK-3q%+O*@N7LDn>;_q&>UE|d}lXfI%<6-S6+<5u=&=_cY({)sy% zX$ond%7oW-;+Wd~J2dKqF~o=s`WrQe_;W;#+<$^{Q0g-vs{a}B;g}y9Hf;qtoftMV z;5yrU*pRMc)nx7WL^`I!it%&1Y3Hjo;+?<)`c$Yx+mj1;Z>wj+^RN!M=z5x*jyuZE zTZgjvFH=Zz*CL{{={T&K-AZqGeFeGTNd6n;$)J695}Fcw8P>V|p^y2=Ok6h1B=+2= zdLM+OJR}y*6y$;X-zN}uO`hGkV#uaB?q#>0MX}@wG0b)8Ig+BANVji@rz6F4_4t-O z#5C?4u9!2G2K4vR@)sMpD4)%2&#%v9$qE@Znv;ax<4;lN0ZEYf8G{18x6)7(9TIWF znN1tKNLqeo(OG3VG)HeX9aQc^$wxGqyk{l9?O7&$YkC#&I8~hga2wpo=fUgVAb2+V z1=6_woQ<)rV2VDEnaoZ66SbuV$4r-E>wCJe9sL?c@R z!L3gN584VC)}Dk~3N2}X&3|~T$6EM!N(OK1KLpY1#)98sBd*XUiXIiC+jnpE)2EBu zQJB~-f9XG2T4s_?O*0H|Yqu88ux*4bSF({_nLEu%dP(N3l&6Qb3Mv0U7kMt$klqz1 zg7W+GsD5`dEmBlvGC$+#qv}W`Exxn+h7_{)__gfz2gV+)I}csQ@~A?lDjP^X42?R= z@K=vB;7~o8n9LK>@Qqp2OGb*#)#F1( zVpA@bpgFq>$>EiTWcADp`a@$3ZdBPx_cZN9qm#CSTU{L9JZ~!=Z=;NxwnPcPb%o*r z+X?v1KXK;I!g{W;E}d5BJ*Qvi%tV9h4M_NR4}5&_68dY`zgzzf zy**AF496tk6<5az+{SOlef~-~BD0*3LnraxF}YBGNP!7n)RAd}S2?{^Wh8afB4T@I z98B9^4f?`3D)V$Q>|T@t?Y|CEOZ9s+Wlb%0DZhxP+=|DS&MD%Q7yEHt*Ed=u@`4FL z8z46IJ@0STE4UHX28;LTAfqMwX;#%C=*slMKYVP-TJ9v&kGKgBOh@8?`dXMOgR#2S zNVv2wnm?b{#Ies^Bs4b{{hMKe4cA;p^$m&W@2niyJhcT*Sf580J!v5BOvl5kDUUPn z=VrhgWE(0=ipR_*yE4wve*>jdqTnq5eX0t!Y>#GXvPJ+DQIvp3NDxf5XM<96sfpw3#2PNFf^lK9-g52V}l zFg5<^jQ+fC5uS9{5_Fp#pg96(>gwad?$7*0SC)PuQCHVtFBu-*cd-yornxZ_joT1z zHx9yjA5)LcZWR9h3rJ5-r1l4l$*=wb61v`q$&6J+hoy|*nDsh5K_eQ(h76vA^h6L{ z+yS$c#m(3I(=bw(q_?b}k{0uFP_Q=@cg);JNf#sGa%vJbc^(1Rr<$;sv;p5pkwC7kmSRS$E5$qN zPI^={o+|QogXEARKCEJckG*K5;d=L|$!rb<%#+3w^qxTKt08i!Ycz|zqKa!u40nZy`b#E9Q{o@e*)nE^L z!(;JKaxd!C(IP8KUy-xxYq9q9d!X>vl8gN8zA$X@Qm9Z}Mlz7hX9| z5%UUHpe|w}-ZWAFBn1%|IEYG{!bu8g5{^B-nRnCW5PW%U2+MV^5Q&dFuuwA$H+HHr)2x}u zYD0^-k2QsYiwT^FTMIE-Gih^T7CbyVl{|91%-vg3h0^-9$u3VN`n|&h+;9T^%d^Iw z+2^U%u@1U2zKQPmxC}-G&L{ov7LhOF4%ezh2Bd389(X^n5KSN3B=i{_1kuTPIRA$? z{+nV%YtEDsaUuuW{5V6Hk)uZxvMXudM<-a^IFX;n9iwiid`MA#8o1BaAz&88WHC-9)pQGC#FD`v~I>G0tm;;>i9fB(P(?%I6f z-}`ri&s(M~cyP=e`(!?4x(SYgZK5SONNpLtE1*>QzB9VJ)tUBOJtYV#e}>bWnsBL< z37%Us4UY7PS$MS!;#^=wnoQrKN|j9fc8(7WyLv;~&@p^%p&2d-6W}+N@py{z4wzSd z2c1!P0}Vyd+(=0Q+>=g2GwRd@uS;|VeP`A)ucmPLeYu-@sa4@|;`6sF*bf!XEOw9$ z*iTF5n6n$_#F=cTr16Qa3t-x>43@3e(cqLZ{DYq60=Y%y)M>2<863LE?ym?3Ysrf! zt!W)Q*Q>_{KklWEmheP>hGcQwm0V(KK@hoZjD6*!ej z2}F?zSgkJ_7W$Mzlgv0gI#dG|K9<7|+KQpW;uwANs~i_OwV)E2z1Zc|PcZ2^h{9}F z!{mmO5VFDxmJb`?c@|!ORSKGW+Ly@ z*tZTJ>SAzsKoJBOy+d1CEU@Cs!&o74J@$xW@Ph*wy!`>BpEQzx#E32xu0py}5$Msq zRnSm25u#I^aYv~;O4=5QPRD$OcT&0F&y!%|o9Ez`!E@j^RU2Q;7>!L*-EdjqYPvV+ zB^qxSCA>RQ9w4`_`dG6cx})a|TN@jQsH~l<`D)60{~?c`_6_qp(_bO8@vq2rVi)=l?m*X$a+aAhW(+bj+L5*I+P zdoY%2^|d#-s!87&YvV_u^Z4$I6Uer|LiGD;D!=>AWzf#u4qpQXP~PM=Di?m9DSh-{ zN{Y(V^?eOk#7A)#L$z72?FhCjV7eWQD;`lZGs#r5T7o=$S=9^?V^(QDcju-O*W;p#SP!e;59u;QM zE1BI$enqY@q@$N~h%=fiyPl(%VpC2c<_M{4GH2^|pQPbht}H}8ng2o|n@ZhiBEi~w zaiO;it$TX~ZMO}l!nCl`?P@F!$mK^9nl^P{siUPsMm5^-lbA)zxTm$G7yi_^+XVm%6i^5(FrQH z>}%;MmajgVC9h0@0f)2T?R^!EledD}V=;8;=DRR;+fl6b*8qQgrp50K(nb#-FTkxO zzTmq@k1U;Dhn~pi(|( zm=j2*wQ#Xd6VJOL2fN$8Qwv*Dmi+w)#NIrMhXR&Bb6zd+S@8qRYNNq-O(mNbGLfln z3T6H=@_7Cj88I6(7T;JN%=W72k=gS~@%!PeOmq5NoVZjTpX`;yFK#|3C$i?Vu_3qc z&Vx5`mR|^WL#RhjY8%mU&I{O2`w=W)$1-~G$R#vG>mzM6KZ(qdG}4=Y4L*h~M+cs_ z@wc}bvb?@xdMYTF7L)BXZ@w+l`0|U!4csSp3Uhcno?U~ri~6bG7d!4z#yE)kD9hGm zsj|K#aW<2cHr65WxNku*2%g#svsxOdPawcy#VM4lI7?n1IZPYcPO!^a^;fuvk)Q;-e_Ky?e2u}zr2%+rN)35vr9>TK2hiA`=0wd>88^Rk1GzA< zze40K=6x@dLiZLDG7yJq4IUtmg~o(8*B|AtU5yfw_R{Jm4<@-)pISr{Z2N#ujdst# zJS%-Df0#|;Pk-f4AlAH$p`)=`O3dGIHK zkdVrFx=E&srmo3>bU_&5iF(NgvF22q!eFW7N2J>j$k5*w&PgY?CVSsz5=}Gs(N9~@ zQR{8|hlfXUURjg4-Q)k%Jl+?>pBAY|C!bj={63JvdQPO$XB8Jvlt`8dHk_sx>}|m~ zNtZLw_7#2|^M*G#q7R?F5sSyZ`61kV!I%XP93e|IG@$jV9{GN|0_@IsBa@HINu;G5 zxO@qOrdEHdD>()XD{SacQU!D^R;R+n71Z#TqTuD>J=ny40rpinOWHCzkxY6m@2QyC zzO1jwf@Lltt?D`S_F5DCZ^=ctv1}sUEWL=tnBL&87v~2(jlBSRwkL^UcNw(Xo^s!F)0aJaksxM;_qhwG}szm<$NYW!y z5k`KyL4VFHr}ZcDTxI{x^BL!>aYWc%i2Z$-h83mL1qaT$&ahGB8{0H_prQhAy*r&Z z*|vkytcw_3U5($KM&c>i7989082${k0ELNR!kjn7rQNHql9$3} z?y5V9{Uv90``YM;$zmAgW(Et5Pr=p2=UD7ZH-7QlD`M(2iP^gLfaRNNB=Wy|N;kIc zo#3C2|T;L-eY!S;R+>K>TO zRhDV+_K{L(8@(R4K1!gc1FwVZx*H`I@9@%b17VL;p8JU+sui6&G`(FPd`Fh^`ZMB< zddCuYpr}kM2LQ}Wc}u+9EWw+6VuwA?FuTOwbl#U4IBT0DRGu@Um0>=ZE)0jZ$;V;y z%_{Wlxd0~^`lD5!NI3YUH|^Ljz~Ctt(be4(mKQa%RjJjmeeyM2_BxSd9RC6%mDJ&e zpC6ffrWv}I{3hy$!<|(A{W`>kLbH?9a}%Iig}(rNY+h%gYiEvVOPi;c27@}`7Qd))Txwv*46UF zwQ903=RjWCzf_2LcvJ|ueHu3Y9l&GUzd~Aw1>g1Z8PwY&S^3wE7t|2my>ctc*~Wp& z>0II5_(43fzY!LnNP;moXPL2_f~Xa{gwx6m_{O?^pt86gYwmqw4N=EL4yY8I`+N|D zKgP6oOMtUwosQe#+IFKqU1pzv~@?*OSgNl&W-(WtnOf57NN@82R(vm zzxLv_C&MM*W}Jar>Yfy$N;^6x;itn!cAtV01XW{eL@!Z2wtgatL`?>uCqwhC>+!CO| zwlBTq^@@G!s{woK@-VRHK$v{lO)+kYr>Kf3rBA6LDo7lbn$*y!x#?52LGCutg-nmo2QtxF&%!Q+K~5?vc(*fMw~x7+X!mnaW~RVKH{jISTr z38PoS9Ftn|Ch!#*pwN?>?>+#5dbwosyH_MA^O+3pq;^DLDsq)p9U5 z>l_>J*pE-S&;b8sp2xLM8rXYFYxXO}iTAwK$qa7kAt!rqP~10Qp97$#>nj|~y(%cJ zo5M_p04)4Bi4`U?=5RWd+#f1~qAh30_$h-)j^`mx~mN1UndK`f>*IR_%9zU?Y zzb+qfTbWc^BS{lF670O3^(U+}g8w+wv;(1oiF?wpR7Aq9@>ZX7D z@mJU4+0(XjxOC=G;zTdpCGdR#U#63 z9W!RBqov~nl*&B$&KO@Ba{C6xpZQQyv}P!OkuZbD_uIj{Hdf=mwB=yE#fe;dL0P$L z9XVJv6lV-egTVg#xxZrzKAnA^?fkq5!)C~^W8gM^WZV$mf5CekwP`(eEBW!n^Z6KY zh)6sXYlMJZaT5Cr|DgT5vtS>hk5K^!(QA_i76bvR&4|P0Cl%=q@vglk?!1s&whUzF zOUZM+P>F%dEnFPZ0f*^#oW|FIE=H2w*hI|UX<^d+FL)|Dnm_A@Ji}r(FI#_^A3Jn_ zkEhp2o_DLQ%*LpArC@%)Q1f;JolBD3z3DDms{{* zZZpgC@aC6_z}$DL0KxGF4N zct_AR)hyZSzJ&LvJ<7)DPXg0?6~127J%;ulE4*ISEKEFhOBmU7sPt83jHpd~gdJwR z_=N%;cs!ww-M8F|#l@O#A67%JAIp*%Eh74b|C zK@Ee$s4sgWlqc^KmO3a%KJQ$EL9Gs`y?q?FS?9qzR_RLqSlvQ3@!t8c^*Y`S_$@i( zugC-aC&SLI5Bb)UiZZ;Mf#d7@2{FS?V}hC&Ue9ijOr7T?$Yri&*T#!X8M8fLd0-e_ zKN#p~#WAGAY%VW8eG{EaJWxOWo#fCKIoPRe#Xql9r^Xd`VbSC?oby2*Z{2p_vi<%% zDy5ETOpF;m-(f?k$WowpHR@>#uVf}2c97dI5O-DR^Po5TdBGYjI%Y_f5Nuh~JC zKfS~JT+pJ zG4W9bE4p<-upS`6Gk%G%XYByK_-!{7_H2fXAHUh9&QVxbz88CP4Sv1Olz*(!rU$o_ zVtu7D&WJz7dk2{bmzQPnC*D8#Fm;OK)*a`6j{{qvWKD)9sc_|{X11bfF-9%AjuW!J z;M_mneA0ibNN}HKR-6ApIQJ`x*Sn=~{dh<2b3vVr4N8W|j-#-v@E{*B-x*3eABl{C zr9}Pc5p@6fjtn(yljJv!Cw@0I$$`gqXf|jNH@@+YhzBK1Pu9YT`$Rt1#5&xq$4Q_} z0^Rc>@Qb()`S-7yU+0!oB`zFK^iM_A2NJHBUq|+gnE@T$+xfnd)7T_e&!XHf!>|h~ zI9cOHsqu#Gyklvu=nL5fV{JvA`j1B>u1f_T1@`01raJK3qW3eW$7YP0xsMy(`wm0Y zGBI`dU{;y@5PGUwQgxLFxc>SwYHj<7f%u&-@8u&X>$j264|3oYZVxr_QogKRo{tUq z!g9^xsrGs)mh`yIZtksMouTVdDAD3^!-n&Y*bJ~KDiOZ_P9bl1a&qgXIgF`J;H&a; zc#WEj+wpBI@$f}Re`3SV1*Cz}^s~ZlBN2gS@|!$ZdJhKmu7l#QxlsAYio6`!2Mt{> z!Vh z`89_P7_oo~lGk+OfjhL+KvA06SWL&|R*^nmlW1M_Y`QCFBe;Dr20q?KC^J=}8@AMe z*Hwfc>+UewU|peeKnfeUrCL}%beHh>W)Cv`l^hFl?9XyjB&^JNFjZ_iEHTgbBWg2} zf&W|tj-Tg9bkjBI$ky3(e%u23`;0ENBy-VWwIMoh@+aQXpDcMnE@_(b0nFs4kzG@# zvUN6D5H?^PJTRL`UMAne9xnEnm6%6*suaS!WsbDxDG6(Bn@<&Ta@bCzHX$>x8%&P= z6Hcroy!x^MeQnbO?{HfS!v_7T*+NusvJ*8-xF6oAg>jmL-q>PT7C;?}kr_h}|NOsro2u#`0lO`H0 zhqCGK=+I38)TZBX=C3yy@}C)@vfmWh5=&$0FS&7|A8a4h>F7-d-Ofg>X-9b9OY*e7 zR)KBv+)K;~CexAYePqW@naj?cM_QH^Mu()0rO`WvQiH^kWX0_yQv2r{T@^J1fzM0(1mV=ufXPMl4dmg>e z9U5$=($?jr;N@%z3oAym%r=R1v0Dv<9TL4r4}xiTtvLVuu7SC~3F;nS%D4WKgQN$J zxHLox^n5qY?i`ClJNr__f=BQ#{UAN{eGt6-SH|8Iydxu><-u+JTXM|j8QilB#meJW zB+c~?Gga9K|7CcQ!ko*c=}I@)W4fLl9`g}Q(hk9>!%tz)*F?Hy!D+hg=4?`vIfr_@ zuA^_ef6(QJI>_*kdJB5sLv|Mi+dtJN(%go#*@@lj3dX6&mWIjCaO#@8MGox0~7FhA(8OzH! zDw#X<9q8KYL2$+tXqGD=-{#(u*d?b!YCyuN3 zV=<`^(lDH!(YQ(Xn~E7oBYHx#r!MuHa}3J@ZcD~&eFCGOtmK=kYA~c*nYS;C<&6>g zc(IQX80?dSh*|CM(`_?-6LkfjIIaUNhZdNc(1L!JKkbR`@_M}=g% zf8Gdc-nWs^vX_Dox0TO5sm~)9Y=#wo1!#J&06sbwMDIv{EdLN8<}&2Kgx&RI)~Uax z;Z>_dPK6UGY}N~L zKhwvHw*S_m-Pa<)MDz&u_yzD_NzGk^Kd%Q zYZ-}MsLo>y!%58DMbQ7K18tk!0M3@hf`@kk{_8jm*t{R2%~fOvhP{*Izp$XYvqz(U zWjuBIw;RkVdlKGSFL8e-PlL=y$rQHFg*)SJV^MJyE;u!f&p4?d^2W9kwN%+j6UYdwV;Z$Ch-y1IF$-zTtk zZom(382fc@5?1SHp&_xOn=DqM&6Nnus`!T<(+9(By%boa?Lor~jd{31H1I1I@kWZD z@GW%$w-vd+CzpncUX8!#ueF@s6K6mpMQ=JIUz5&XQwYW9qj}Mz_n1(dMEgIzM@kw! z$$*24sK&iWmN!=cmc8eAPwNtTtGuQ*f9Jr)YN?>)kc4ipD`Di^)pVQTQ@Hl;5qYbb z%70j_A+(#*oFUr$+vr0u_hcE&@u(~<)-;t$A&rz0}qf$g)m`(><<{ z#P;h*mM3QdcPkBHaoS$&8yAC)W{HxOi(kT;md)5twHoXGJBvT+7SS0VQIf9rzrZ2n z9;!VslqF1f%(KttgJDM@ojZC%pGb(uLjuDbd?tnBMrn6lN_Myd@s7$@DN;ab%q*42R@Rv(yb+f zuw!iwJ-+!eO?4hkje{-uo{MsHT$m|s9(xcT{n`j&H)qoUInT-c4YjQOjvAYio*`^2 zy2EruoyOXtX!u!q6%RTa(JdFR6F=1^i9L3Lm9IQ*K3Rg#+$ZtVc?SHyMSI!NdHbmL zj~u=&Y9^#i>O&phY+`-I48M`1La2q(BHA|RBMDv60U>+qVcatlJeV8@CyaV>-@M0y zf4Dq+8Jr9rI!9>s!yYv6n<*dA!=2CjHlL@8!o1SnW56))CzOqfg4x65>C=!IaASoJ zO}>5)6^aSq4bAyEBKr`5Ai_Z`-}{H^gWS*)hGOzYQ!~$F2_ZK$6!*ZGBr@IU)w^W5Qv4aRtmd zv0JjtWh9+^ZIWwgp#i*^7fv^g?Lpr>`_7u0JjsX$4v@WjGkj@dq{RO}RyvXwZ(MB$ z?z;{W@7po7w&pgO@yif5K23)q2c4w0Tkp{a>84crv=nTnh0~pxduUo|7aFUm^9UG% zo`tnI|IG$|)k2-8POoDt|J;CrO|u|$>wA*rkpUAf?{(evAdlv(*+tDw>X>PM1gK5< z0@W+7!>=vZsJqiv`oLg1UH0x6=yyz}r$&8+#rC??@O>i>v~f&EgL5b{hb0w$mV{@+)Z` zd>;b4dgJt8pNRhv8!!s}gxa&B(Q(69c=zBe{0zzhrvs0`al#zh)iVgJGADuj>s-9> zN{)VCK7q&SR#E?;&s6=yKN>XnI=$pi=?HPBy?6W$Tx@*}-xOzJ&TflRzRgkRQpL+sE|NYWkwY0`~s%6l#P zt?>!;nl6XybT>$@9O5u4P9A?QI}A>-_tAP~KKz&R09Goz!x?=7S>OBxEKkFlYfYTS zMfRucYiAYjdT#-hF@~^oxiKl(R44Y3PBd|p8QnW&AMaeUT9mV2;TaP|-$`dUe{7~o z|0-Hi^5!p`i%`G^zclH6^(6W=qMT+P--g4TN||P~Hn<$P0$c5)c}IYZTaOQCC9jXO z?TZ~jIdB;-_v*yT$y(^~`72zWdRh25_a*kbz70;7+VJo67}lTq2;GJvH}LjQ`uxOT ze7|D||E4>hebDd2Yqms!?)NLD=fdy8ji%m2XRR?EIZ&VGuNp`fXB-qBs*a_NJ0B6& zS7ej8nb4EjSup45PQF~P5-&yGl5FaCP^d5sgPe`7@O|DnV!2>GxQ{(UghL0Rh4rE* zPAsG*TIE#XXEa=o?ZBMYHng!kN5TdwLyUYH8`-E#j}%-dk>3nJeas=||KAgmaWsqc z+`dFIBE1|A`BQG2wgeBJ5j}WIeR#3QD%$gPJsoQh2xXgNRMl zwDU<0j7dLE2X0-)dnFIWD@IZtTW3b^#_mA-eot{#R48pRJWL;*^1u;3YS5$l3(+mx zkLKn%3DQbmT!fNe>T-TB2#;KaK9zF^xdi5 z*rUK6+guvhzCOydqe(})C-^+fhS_ZXktf*Mt<9H>8pze|o71ksa2h>t82{4kM7NIW zive30WN!Hddykk(WgfqAY459`I(rY@zH2FYxgD{vkATW|`#@CiWs<$?2QhN#P36+A zp^ze@`I?s8F-+9gPE(RTC_78%{cVGy`EIn_=a;Avt)!)oGcjO=BYkEQM88eUroH|Y zkeFD}`|kOJZ0y^YE^=0tj_P@uMl4-JrMQ`O_g4qWmeo{6xt<%T?7>%F^_S6ZoFY$@E#OI`uuAiz7T|@DG#H z`Dpv?kTN8Us^hyWNlDD9;-Rog_=Q-R~wTkDfY^B3Q9@O>Ox$wYWLmJgI zP{Crw*OKP|F`+z(T5b-&pd;P{YuN%~7O4j-bn zZ^K~O<2oU3^IlrycouwveeqU@2K{_4nHFy~HdgK z-Ms*Ue3fWWLjj43c`WHAf04Nj-2uZJD_PGQBO$hrgmiuwhML|=I6H6>ogjHZk1xCp z9!oA#k4atR#me36+V)ql{q-T7ElZ?2%HFiE-f&!d zGv}mXII}dBl%7~ge>eH?FGJ6R>bNn~FmNQT>*x*k8qOGOKT~ph(jt8NdpDn|-;Phu z=#=j2mV+HZ6G}_chJgH%G5EWzUDE4;H7`=o<0;9DaCU_jUFdp8GJ3^aEVIglC+DkJ z_-`HYxecRpG|r;KM;Fwux8$v=&a_U`K&t%sHdxu7hSkd=VD7dhC(Lavo@WrORV7D z8#Q@wj3ys1Q0lj<5f**Kcnv*UQ=o#T*~M8S4V4~#Dlz`w2OAaCu;EUqi?$TBB7zDFBLnmti+_uOLMdS{(v z=!UCM=Clyi(sQA8`2hNK1EF_MYQc|tW{~|N6W6wFC-P6cVeUK)$%_08h|SI;u}M3i zcAgE4AKyfzf7hXfizDv-bAt_En*uLJc+kdockry~IougtKst9%rLVoi>7(c4XmUde zy>T;w?35GTwlm{!xT!boRtbaeuKO@K|0&K)D&U)^u7ZF!!v%h5JvQV&BlDbM_?LI1 zx&6CX9=y)VH8kL%WJ%dP8s$?4>Qn6D%gD1>GxaLE=b6Ax<81bJ@nY)m_bjlqY|wt) zPHx*b3424X(&i#1`o4Auz1Y{2_TP7bh6p7zAaNs|a&R}uy?IAYNb+gu`I}sOVlJQK za2N-=g-|Q7rE*!1@YhNit$DGD9;+Hie`K9v(|TOOj2HXF>>r@CIv2_59TQm@jJW4$N?%E2pM`*V8OIN(^A!lsXp<{X2of-~0!fqmA(N&Jx^b7Y4Dh z*32iUn9b7s!+y4e!q2X`bn2>Q)WLZm@olbR+jfKN!B$1s^}dnSIRA$u)YM??(ta@O z!c;h>RY|mzGPwS}Az1p-0zBgnU`^##R$#n_Rv8bZJC?Yh?%Ne~o>vMuRksQI(=xI8 zps3?^(}dA4IW_?Bv6W%&-OYcZm6+3M{q8@PJ`txF4zKcg%Y*FOJ^QB%_ zuy;TwIa{<<_B?!qY~ff51|L+xsy9W%&sZL7OFZGd=sP(-U5CfbvgWb<*6?9A*F+Y_ z9{xD&5P#q%W(k%m@ntuj;gc&`u*h2#7WUSr@6R70XZw1>K)^8 zI;mLux}4~1s$j3Tg=nb#hfP?q2q%9YOmh>)!=o=dz%-aJkIi+ZR<@UT_L^rHG1mfq zF*nKL=xJDSK)s~TdW zBS0ZZkN>{*nT2242S?(oG0@4G1)EHP)gOJ)alAX^8Q#Y-8VSkUuCRhReVEJZ)#Rje zu=pG7ftLr&;tNY|Nj9sQk{thb(HETt`LFizm$UzJWl@(tX2NACpOA#Qo9$ts#%ySt zp$88qeLz)@N%+akkxt(eNq>#pM*JR`!+76y1db-79|4*iLL~)I$51MX-D3 z00`)%O{EGy;ev&wEdF^D-}Xa;n;$-hQbT+0p;n1IuK!{Qg)MM%xS_~v-5{HM(@Z8U zh(~Vm(j_dYKQ&O9%on%zmMKool_~g-07X&XHm|=A=6L(c>PPmMIiCoZ1w8lSTJFz4 z%lr`RsC9$yCvQOh(mrV1yM^6Mc*5?fjK|yw#(d=CXHfUr2Fey49DY|?heBsFhVR;r z9oF$UeUky2URXlRey=Cd=dVi~JT8y{%9_-E*H#kuM;Q$!USMC<#a!tTb5U|l)CU%Q zWgkTt7@|MNQyR<2Pi#oHoDn z{5)THR7E!avIV)ns*%|RYVpQv+a*J`uEKe@ocSl24PRMh&Q&_!;sY_aH+I?+K{;bM zJm`|(A&)HBvmpw7?j_Lu51c3uYJ)4bW9W}Fd+7VD3G~W^q4f5D4&-6xLTN%BKlUGsXa0oK&e&)yRoTme0{zhL&3eA>v#7m)tR~ZXF;7-M(3)4g zP@$SrRZ05?74APx4=;@^!zv$RXq;q)8x3Q*#?@@utb3jqrDXw2ca1~gg#{k{`H}w> zwO}R5w`jrCPxR{lJy14pB^tI>xL~0zSWfU4wG?Um{7eVPZ!;tZBU6}@-$HJiIS^;E z!{pEJ84&GzkDG@~;IBW4*+}2wuq!ziF;Wo@E%3%E=HqzJ!KF;&uN#%)yiK zYUHkVB)?Xo!>S@XuuAg>a~SF;GcoATBZKDhUMD`Y`+FKB_LHxaj<~u6oBE%?>nEMy zTV5kE(#*if@Leo1|1{5$?!-k=Gx+q8m2AfLuVTIh!7VS8Fm_hEWPWWwET^r4aP2Ew zZ>~?i479ty1Xf?NOp18 zbAIaYLdlA^A6?=i13hp- z$w_hEFy@0Nzo13poFQgZ$zFWABWdn&54RSb#^Zzf(amnJ_>Y-;`GlDn6y(qFuV1I| zb@s(X;h8^~I=_M0&PwLx+NpGpOniS&7xMbZH+ZyHj%#&DKd`M`R;v0)wWPZ(0kS*p zvIIp%Hczdc|3|ZAnbCSO*DdQHc~%y9CY^y=Q#I)6YY!VPycRV2+@K$%Ycc%JIXv%xeVS+~1!!r*%QH>REIdbQ6Yz zJJ2b!QrIm!&X2G6l|8HY!B17ou;PFrH8JhUza*TYdxdP!=u)Ncd{&U$uDLMliX-kY zFQy-?*V3e8M`*>51o~@GEgTX~)5wRe)G7Kc?YD6zz2kEjehp70)*HfT-!X--@x^YK z*mh24bFiB4yX8H+Li-j>lTMVHo2f}1CMrv7fBmGbzYfv^j&Ee6+}dTGyLDuj0|wCC zN2b!cD@&-GVw3EyWw>nLjYl}zEP~7J{s_a*c0%Ll`?8w-i)FtqHIUNzt~BFt5)tnA zr?Ue)@vpLytZH5=-`TwnVtbp?z;Q#V^q02u=+b@C9wS%~1K|bGo z`Ls}I6bFf8uL`4vF2bYRw?p;KT&m>JgIcy7A~E*S&@W^*{Eh7=`C>m@5@oxc-ZihH zbDm3Sx@k}8=}-S?d6Sm(ddfy#_O*$RGV_-mOv;p{&r+7XtIuMN$;yI?ii6;E~=GVeg>9ImTYPwW|fX+PGzA?GN{6kk@#^BX- z?}7#%A3h$L%}nyHu%6h&Pa&(~q1saI}OE?2GOm^0aT{T70JgB~2WoyJW!;6~-UAb7YA^}l|RREIj_&d=+}CXY;FKuTz)|2A0H zTn_K7g0OkICMgVP5`M06rZxAKrMG+8OMh=L!E~YpA?BlD>)<}9J!Ar&Hb0JU#_RDB zmu=~w77hAmVsB~m=n84zUQOwrs$e+bC>73Um$(R18feR36BrS`60RjhQJ;tdw8(N5 zO)tG(s_CmN9B$0^#uX&_p4a+ z3-1I!(=7U3^t^{_okCxU9pv>3#XIuu`18m^0YX(kzHb4uw~l2lRsE#%KK)d%d2d)74$pz2+o8onHn^zP7RMIlwKIa+sZ#RQmA3B4JTOIQmX|#_KDM zsKnWXCft$Hfgv{NP!{Y$HY&iY&GR5ku@=p9PlDazub5qBPCV2nqHH2ZeT|d+_@@fG zIVBTs{p-Om^p=TPhV}5l%1{W1Zsqe%&X-;@xh+*1(+V}hX1ebFK{`RF0Nd?dxtsP- zzA?X8Fl$to+$=o{quZwA*t?1N^z$q1J#{;Y{2TR^&q>f0JRluOC zP;S%pj;`qMBTd@gO%>+try%*b%g)g{?As7-!Wh+L-u!GSE{{~;i>9xVjJC*S z4-4PA+>7_d2R)uKx4Dmr{>rmV#jd&3GETmlOyj+QVI~vR0hS`;>>L^RY57f~sH$vgH`Bf^{d<3@lvIQT{6^vR}O8UkvF0G3U zr$>A&p=L#@5Ik=k$WEz1IN*u@R}%N5oFj^QEXlKOws#oD~{YFhL_aGv7sEwje zIOZp*Qk~E3nlyRwLu0{r7!_-V3GWjUN5_R;gZ|f631bx3Am;SuHQlR)gu~y+`l2X2 zEsDHry}U{J(*nNm#T>34KNstNyP!+|BkbAhQqjBehL+ja(2LJ5kSpuH;$9zJna$ET zzTZ=x&XzOAy`QHuv!Qb-*{H;44i04s0loRYH&HC(U7QdRFW}hERoK0J6lrcfKwCAx z(=q-VsE4-+>of|1Y*`lcyyAxo@7%-ZCH9H|Rehp=vfv*&QVL_B|NvZ0GXGAyU3e zvYe<(_ra{xG;n%)20jndqwn)CLyWNivqO6D=_1Pc-rg+9ihH7`CCyRtp&?Q7?(8(l z+e=R+MX!@2)Bk*DcRyaiF%WQSfSoc9*d=g*K?_EX0NatFmdW~bzL<>%65c7?XrpTK_s%KX|lF@yT_ z8f-n@i!E!p0z(qk()pR$R593r@7?hd3o>P7R;eCl4>M+?F0TZ8FL&a;*o`PSd=dKI zN$0;)ca@&0SqgF`#f*o%WZsI$grBiZ*UfUqxYNM`!N~o2gXsSNwZ;- ziy>+LVMfkv@?*Q+9*YU8u<4>Xe($`>q+9l|z!TQoPkL4ID@hIgOFz)xQBjiTm%g&P7YD#q z4Kb^08%X^|&{OfxVDtDiVr2HzMPsBqTo{xEtERnUMviq%dVIRjHti5B3R@1x6m=<4 zYL&3~X0oE<9+e+928QHZ0HMu-+K0b{(`AKhWU>M|wA71lzg#8R9MOf2{}OS>nZttL zEEUPusWoKkfSF)%O9>Auq(S+sL}tF|3trrL7hhHj65Ym^g8zi?(E4ILYu}p%kI(Ia z@2&sfZk8|TpPT^FzrC2b&L;9M(Th$}FK0DU6P({>M~V~<2*HMia6(`p9JxYrHyz}2 zKN-UstbwZFrO;>oJBamXeQ})@QX!FQ^Awc7F;@fH8eY@(}kh#)cUj`x)5vDT>gmUivQaIGy6!htuN36@8__E zyf6Ch{|+Cub?8*HcQEMlemcB)03I+pBJSfYdBJJ%Vo^ucneO*}nMu zni6z8e@Ys{r_fz5?C6I5Q|a34{asVG^k%zusIq8-F2Wj%Kr%U*5BoC3quqxy*&^z?lN z`fY6iy&bU)udN)6^h`YISJ4B?u1_Z(Hh-Y~;Rh_08vrR673rt`1mZV7!=Ot;@OJrB z)PI}+sqP72bxT`VdA|bQ>Hi@+15#O^F`@AL#za)g%iyoYSwTa^3Ip>)Fi!p>#vfP) zcWR7i%H>(qVbdaL)ybfiAN1&pvLj$WcNh%%@e!8HFriINe$f3?nXl|P!(Mbp(OT7R z@diJfZf`YXyDtnv6@xUkwRr(uWx9`nZae73nMh91&e9{}b`cBj{h|jlT%v9@0If$W zLeKlBSW};Na=2|Ydp4Si8X_gvm$UN8+h6IR_$?_6a+J zzk?~wk@&dh2t`AVl7If+A>o}9S`Yh(Ef=>#=*3WS_r)9G_`@-n=;TB%uWSZqG0foJ z!g6YKuLGj*ZpSaLrTp*pV#y!vIP7_OHB{7BlWAcsSkq3aBHd0K?H|C0j87!ki;;6x zp}1HQ4(}Rtg_px@SeQ|!#MbU8rYU9!2i8iM(aLQAsM{@2%-01|!c-0e{g3QR? z;%VUKeFNq%{YzFQtmZds!-XLE5oi+Rho*%w;9wJgOumGu1RY@){Sxtg%^(=EOdXdT zJA$UAW$ev|X#9Cjkv$r60!LMR$6a%-!qxNd(R4`wT6in*q=U}T^HvImKX#W~+&5yCde{t;}E~Zxc@amqKQWT-yhc<(M`12hL0GNfxZSM`AM+Xdg=j7C6=& zcf1_S9&A?>exEg$7@eQNLS>>3Bsu^$p4@@ge11a7_i*aF%|q1CYQh{8vMg11HMP8oY-dVca0jN%To_!N~9me zY?J1*g5$YK@chq1EY&>-vrqrzz8^FBB(GvPIH8S69r9Qu`FA3AFgkHG|U{ZS)Yn=6wb-bL6lYcj{{&I7ncC!(a{G1P~by5YN05_E0 ziC|;zB$3hj-V#++1+aG*Oc#Ya)7wY%sqik47ulDB$dbe!`@gZJ548BHq)@)uYzID! zbOEDG8N9R0XWn+kT%k`jNaigkCl2?e4FO7c|8^@&_k6~d=v~2IhF4tl?ry_9m!{yB zjS(;?XF6Vz#-piT41~<@g|{zf<0`$4lG?H(cpxK`>3%){jhhufJ1Le73%*X)K6}Js za#M*@Uk{i&eG6)uq_8oj-&v8589U=C4=$xGM87c)!-9tLrJ2WY-&4S4vy}LMVT!yU z{uW=`q9_}avY*ZQ+DwX1>?5~-KOkEw!{LQqFX|js>H6?dG5L}s&wHhrvi~u3p7B_I zVH`IzqO2r^goHwi`K zf4}#_hkKpt`u)D&Y`zz+6R-roamr*j7D*$e;tRqt--c1m>=xLd*2+%kyyDiR1i|#H z(r~e}&|K!sRA?683)A2!k~h9h1HNp9xcOzm?c;XSxe7vh>BLFY>5cJqeirHYY=B0L ziDLfu!&r#nW%jmx1svgesLhHXkkrj#f0JrhsH7yz5gsD{ogGICx%I?&YbQ2J%U~0G z(!ptoNOj2NH1=I)51YO56CChdgU;S}Vb&YwvfWoEGXwA0sIq!DNm_N4^iO_3RCVph zbHg$!B3H>pPc|fbzK`JP+k(I&qXQlo8Z!G;M(|o$6MiXu2ZfD?@OgJ#<{>S`{{CG8 zk9^jkZz}s~p5H`HD*p^RJrc>;Pq*=@ z`B67!fuZItoVHJyB1ZQRO zjB01t8?3>`J{ko*-Qhxw*@fuq)njmbu@QARDZ`)XH!v-c3s@!Ag3LD#g*Ej#z-vO0 zSQ+m(*|Cu+w@NV}hf1j5a$A%8(>1oaj3j~2o`VY!-Fn*vF!xxuA9X!)djPwMHk?N`cpDc{2iR{s6ff13c5Az*WB#An05KY zu%?K+VD!z8*j_n_Z*_Vx%^z-T@MSywTp9{9UxorXrNCA$R$~wNeCv!z0nLrOk8j^d z!TV}Q7 z7Y(ie5I;|;p}_|FGFXXonX(Yam-D%%{cG^&k}#a`miL@VXuzFqW62i1C^mI|9lNS} z2cL3ZTV1f`C#znTCOBK6PLw`ff+N!t(WVUn-1o)qc;{M*4_52L@6+lmBYzHi5nO<~ zm*n7j+o|9!Q2{4@D6n^t<}mVP00vVHK+Dz@XH@OLEq7AE=jKlKDGZZETg-@P(q>#x zP{?jY1cSwzvG86*6gp<9^E0J*s@XY)W!5f%F^87XyQ3mOBkDZ7{^briMpcyJ^dNy~L(~zm_GZ zshEl;2r`kc>VPnj=Wl6BmOygBbck@dYOeU{IvjfS5AV#3;e$h!SB-T0C4qog}j^8kpC?Xg(8WDiL1?||k-gw9WMWDZyx8D=bo^yUI# ziC#ZE+q(cAE!aXvr_92+>%Kui+&fq|c>T zrcdBPANWG?)?BV{&vYs==Mg9dhLay6)A3v<7wWgb0p@p93oVZNKy~;>x;k_-*V$Bn zPE5DO*=6!{L4+R(kY0s6KdfR8<_hRx9&7}P42LID^yKO%`u~Ez7dPFy|^P9*X+`Ugy8s=kDiObl_V>6U& zufzG9-SA@n(QI$bC)^jbPv|1|O(+&UfKFbI;PCFBSo(D;-xE*ZYN$PK>WG1lkHpA> zEmN_0{36m{Z^j;G#ehdx5lHs&Uh(;boI{EuKJQfocZ@77j@DF?8{Vhszg&HoUbPA) zEgC~V+%Jcs)<5`2$0?kg^n$ZXvoex-hjwU+W1@UVU{H-M3c^X(iH|buyOw-$o zK=j%a?&#$JyzWsEowt1vh>1}YXlq3+-xP2n7Z&49DH?bY*M&Ctq~Sd#mf$+u0sOM6 zVd?@0wEa&7ZLhn9Z|h3Z$@7HpOuU}X9V~;w(@j{^u!))MMQlLpop86YI}Gr9+2r<4 zwCkJ!_8Uk?Mi=)WTXi8ilWT*=7(5oN#}d-*dsAmQzB}Db426 zbApkaf~@fFg`1F4$3f5J8hUkII(PJ=Ccfx-3mT(;W93IX;OI~!eVTX?ls~?Of?j*} z`MoMRxZfNf9ni+2+5dzSwzpwVb!)7!*G|}}QiJF~CVp6d8zptD#%uNU(8lIsWYOkN z;u{l?YyB(OeoY*WZQIH8$KOI4pC(ZYoi75d4_m2pSTJ1>^o~;zKZ7(|&U1xj^Uz56 zW}$fMHC+5@8$A~lDoC7AkAJ?YMjDym&-&A4I^ALJusfI|W&qvshJyKf7?S5$64Pf_{}4P)ika6r>VK8|#iju>K=q z_Cn`K3FY1tw(*jWaYDD!O?4bKRHOb6+Gj7R}abU7`3sg*R0;_AQ znO2V?>$oXSt@~2x+~tqCA(s~rSM-cJ#kA8zL&nxxqDAg`;I{LwNQq4Svj!Y%R8AvMjp#KT?%`4FJ@8BNuay? z8V&dQMU85fQ_n?rgt<2qggI}Ls8`4w5Y;V)oL|#UuL^-U*;ItI`tq7 z9T8>iQqS=7b4_%~uOMh?Fh{ZL%#heYF^D`DMC;;bdpQshxDjmz|uT$hKXSmYKo%(F2_I(JCw=$+lW*z=kQ;eL}gZuqqt_3DZSsgfylRZ24|9nMCl{}TmSQ$R-= zUV+hVbNHMbMBg)c8o$+%s^k<2g9_E))r4b0_cCd&O>d3hyiXk8uWmuLR~w)_(U3Xv z+gxjrZ6Izv3^o=EgjpeC$aY2#y;9~1kqTb$so@3e)U^YXH?_iTZ=%dCcIOE;ylUcl z`Tlq@&nLAwUC(Fk{6I_X9B5~0W4-qgu=ijgO&=8E=#pS4s|`l3e$UJjZ>;8;M(3MP z`jJONC;mWX>2qkRTWVE_)=gn<*dv7R&Y_PFuAw#8QiRX;UO@xyv(Qfb2etGE3X;E{ z5R7Ht%^$CFL!<9>2-%XOv~9;8YCi8IbbomUOO&Uwy3H5BSxbyfKlfPpd{rRr_B;)9 z;xtIfRu1+UG{e#OVRY2`Dj53OunZ}TYgTNBH6MhO-vOekZVk3N{J>?lI*+Tq z?(_o>ctz38YiFTb>0ZKB<7|*%OoGYiI(-YlcNBbX|UNj7egArgsl&UW%SK$)r1QH~#3y+`gC)=(^k#e6H zQk#$ie@DmRjQk~-*x3=e&xc6t*>8~8t%m!Jt8l>1*T{O$W@h?f5qmxCjwd`iK&8#j zP=kzSI8v1Y$~~pj{MCQljTiB_%~=W8q%{baw%teBW7ZSzWtHnR6^+T_>lapdTM zYiRV0dvw$4JgOPf1HZ>V!8#Bic-bQ}$ko!s8AZS7q2e?c^YbLV;$lNSG<_f; zqY3KRz8LE#M&iQP?!?P=8oQv-g@+~Aleyc>Vbjz(OmHg>4*FZ;JxLbq`MeZ(Z8L_2 z$x?V^mrOJFn3I#q=A>Ms8*Rubhm|MC!RVnsENSK*+ToKr<^@nyn z)9U-DQ*c_pHp$yC4p!)tBMA!|vb6p-$Q-Yy*-Z)9^}u=vAMz*5*ND?yPYT$}bwh0T zUxn)T);aKmKU;y36nj2!kNaLROwU}s&vADabEn>I0d=J|=p5-4{>#Zgo2as2@mGwJ zUb+yYE<2Ptzm%HXnTnsDks%NMhTucX*06u)Wm%TBezlZOAsYx&1jYT6*?jG2(A7H( z60gUTs zFz_+2-n_d2Yi)^wq}zP{%Q1xc#p+Z)8vll^8rXtg{!2wS^3RedKP(|2b22>rwM;Ox zw}up39s+}!qhK2v&i?(9ht*gLBJxM$&)W~ck+7NAEkFUP8og*7`G`Yo9N3ftv244X zCuxW@AnsdT$z0)c`qO?p&ic%|r^S^Re|8ndq}sC8;d81b%5#K4yF>-c7Hx!W?dhPV zGlQt_OGA_W^Q(r8>*uOlPh;!%htq+r*Xe}>TbSR~0V^bWIqx-3p-U2=6sFu{ z>y}@EOEFa_SYMMlMXK|>M7%oWODL{b>dkv&oXw)2O(d${7FNr=n~8(h zCor*C`^q_jo5-##8$S+ru2$w55OdxlGG)gl+OlXWU7S6@CGPV?k!=~wEwTs&X+>~p zakr`4<`8f*)B~x=PO?3;h3x$Hj#@NI;cS~R_|by`?4}w>#+F|ojXUg7Lx%_aH~2`H zJpZlmW>G$59#SGze&uA}sS(`n`;elMXgF~q*!;~4bIxIJ5eUTIspfSwH~JFnZQjF% zY+is{4c{;KY2(V1yQz+OywG5>DGAY?4K^1mNU8K9GQ>oQT3i7+Z)gO$QvBX^q}e*2FSXW>#O)8KgB`@4?N* zj7+pkCiC+0_<8GnSYrDg0#PZMlsF$5dmG?Iwc>2z$sFc?QnY$7G88A|&0je{oNbHNzISoX6)SrCGLl4Q)oTiXL6pr>3xEF z?Cw#IyCvv+^C7e>SD)0(l!Q}#wz#r#3eIl&#%&3Vg4MmoP#M1tmL?Yqi)%}u5VB-zYfvMQaV;1j!+=-7}-;al{4Gb@2@o-? zO&4|Ffq5OvaM{Kg)a!0Cu2iW)R(I@}-$famenK8Te=|ZA{}sS!(`a^VsS(SaDuHj$ z&^3QBlE8+!QgbiTPERNFac80qumkSLkhjA#8s$^Q4ryAk9pg8Ev8IqcF&km6CAI9T zj0~&U)d`Pf@8SEIM)*VNZ0ca0kLEq-z$-4s;SZGpOl>sbjM;cj;}y?Bt~|}jmc>B4 zv@>qf8b@qYRZ+}3ZQhfy1}bt1Ha2=e;v!GN6vyu z6UOgwo=7!-e`6W0wb=|W`PoT1zbg+J`A1_fugA(swF2?4okIPmu55~|G}zpl2yX`s zVCMl%R(R_s>Uik{j~4T;VkdJlAvue9ie;lXrJtZ^n2%(B2J+_(Bl5g+9A4)(3od#d z0$ayUdSb}{-o4C{RRrp=R)utCnL*(c&x1X@vI6p^CV|A5QkZae3Uk@^97t~rz2#U& z^R9ZM^J=5<$Gu`K-bbGBZEb9EYYSLN2?gfd04n%8mifzS(68PFAn))Ow@2GS?B7%D zobGCRbjB#Wy?F?O-^7MV+*hagsW+A>@nvhAig5XiU^q7Q8t4y%@b`QkJjzpt z)9OyFNIDUpPN(Sp2`3V>Y#Z5ieGvIx7h#V01Ui#njTL&E!DK)LN(~kfBYx+Ws(l!C zSl*zrzIrU>StGnP)CI4<1|V@?g55|^W_9yBLE3JM+58{xY37MlEY#sNJeCLCJ5?P< zR~q2o$zgc>%_wfu_0zCyrvK1`pi-4By_@`Tx?Nmy-+B{zJ;7fQalp)Xw$ z(5I^jU(y`FTUwXC+qw}Q*gg;5&3TUJrv8DQDy!(%7m4VBUIZ>xKf?Rm=Ca>zU*ToN zTlnI%4Zlqjfikm1sGY9F9KMOM3lTOr;$|i%em9REF5U_|U+}%|z^Ux0^EjrOHmW)% zybSDgdFEAw5}qEU4x&v>aLHg69&IEA!xNn`Ziu9&`|{DW&JB3omRF=eQkwa%zlbX5 zh=Y5CHCyy%E^|RgY1Wl6+P=!1Id*-+(JAq$4Q<3{x9p>$Co^#6N@GZ(l`whr4y-1O zzz3d=f+_s@VoaMa?j0FR#EbczkLMe39wkmrHXK2ge6u-MxF4R0m7%=*i`WlIDW)c2 zjeFK)(bF|1=+*8|T*<~0blrqQVE%ppqz2vD20c^0{~^jOUhqs?g)m(3$3RdQ;z5&d?_4p_$Z5D`7~n{ErdN5{{JW{HWHkzsooTCZzP(g{yOA?c&?*Lmu6`6&MS9(eP6Ub$7a0B{35!VVgpu^xm>gDVtA}D0tvh8`Mh`r zyc%l@(m(IBR$U8x&I_A;?9Ap)dn+(~#|-#ZRShPrUnpWAM*`0WVQJSDB+w`d9k)@% zPmNcxLK{8izo3jRDk}h|xq)ywT>@2ojldV$8);V6SEOe>53E#z@S`bX$*kMD%s{h& zZ7UVxtdvkRZ>b0NAJSoOpN_)TJeT?G0a3vt4G$=+E(ZBTeU@9gk6kmX0Hrt5bc47k z);w|-Z0D6z`|DoBL--o?ncs${idS$g;0&|Rs%3I|j%=sRXyy_96FNK0X_=ot+?0KY z$<&$fzWOS@GEudXbY)g}0>Xyst}{ zqSY_7&1(^>UsT9`T4}NSdZ*35=0?NS#`k&E0a zLE`mom>hkAuk4bCJiQWZlNUxTdp0taPDK{=_MWiKl4rla4uH`O{haZp6ZF!-ROo9y zfgjK18G3(;Kqp#)9Wj^<+yDhD(;&!g*5#k&fUeeBK*jcdbdi<>KDGZb?if2(__m-F zi3PZV;GzR+HN1+NLY45?Aaj;-IiDSRR?K?8y~2v2am+{l5sWz<#l|P~;XgLdz`k4u z|C;{?WTfV>tFv>_8i_ABe%@&;ckDXz&naSaI(W8e*j2pWc?0`?&zEWMyDSv9$wF@q z2f^D|RVpHA!!^FES%#huYZ-NhiTMdwMBj4kLv?UlpCrvDHgs!UCY?ln!jk@K6h1?N zwlH73_>l#(pEi~ismx=4Ki+|lpYKBObw?b&w2XRN$bi&{D|K70L!Zy^C#UKQNW;@q zWLZ^&7xd>co&8d*ExZa$_bIT>hB_#{A0=FnJV0xWKUd8u1!T4BHF6J_3FM+8^%i{( zMtj3x>5N!<>bsD>U#Ln$7d@rF>_b3*HaWGlBSDh{mf)2k%%_K?7%bkFen==5L!y#!W0H2VObgu04vx zNLiOGnHY*ij%>sqyG)S7^HLCVU55iBje&SP)jzIiI4tyTOf@92oW>Uv*QD>=D z!t1TiVRH0Y8oA1X}gdz783) zgyA>8^LYQOB~#Y;gCy3?$AfRwIfHZ|6tCJ3GQ{aGuXFJ#$$V-&c$28~Xpyz~ ztBB=-@u1_Ci?wT`=|UqBd^vg{Is2;!-#n3sA1qTPd#6vv4=4MxAMN3c_D8YV)915Q zuQk}D!?H}fNd=pFjY1bP~rc>M6V5?MO_YgNu*5c{_)pDmemn3)TL@t4vSEMTi4 zd$c4G{YmAVEM2oS2dn)Yft8o1K^;}#(m>lsgwyFb&U03&!T#UcuxW5Ab=pa%hab z0Oyn!!iAkj+0wtqVYc38-b2jy@YnO-%I3lFF7+3_64`|cpLbzpgUt=xf23~bF0DscNYn+%X0MfLlK%bcRc9LcVvbqpTgEn zzo>HW#Ah4KFUT{Zu734(lD%&UqpYe;WPg+)=XDz zHRDQKVeGuGY)-E}|GoK@J0FE$9!kJB9M#FQD~a^W*oU0)sVFS}^a%O0_aF9B=|*=B zJj3OCmlA`bMk2RmDUm+(9VP@)I&MIZb&ik0ro|C3YtJV5(a3wbEsl_Le%oQpF(t-4 z`ObHFpHQs#910v$XKr!pEZXM&-h9iaKGgKRoT_do^gzfd&@QxOHueu#)rW~JRc!*h zWHycI&)$qZW{2TlEorFL_8k5jJCAa+TG19k56m>mhQE`I*~gmoEaO}StE*`ePB7X9 z%jJ&4>Hf|5=e+%_?MfIJ&e()Px{XnAOoI8W;7!&$DedY!(Qt~ynNwAqPcAzzTuVzQ}`Z3xu+}GEYM)*3R~Goj3(VTQ@Yyz z{!Zq)^*C_DysLmMVjf9fplxj*ehWC_m z{Vykwau5C*u_F-fB}07%#qlL9&+P0MuuaMKSn`+@EZ?-0C2eS6#e$P$amQbBLItpj z)F$%!Z7@mq7+`kG)?n|(Wyr2}5Jk1*<53n4puCJBEze9kzkIGY+;MpUftG zihCrnP=NH~cwH z_Ki{Izd+am1W@5?TYPm%2RK>lkm`Gy zkYzO$XKm_3d7saN+V9P5**#HWWAPCA>8m60_Y?SYMkzZPpu()IG?;SUEZkBgTCJ8H z&QgE%;Z=XfvkmLg*yx8+Y@~K4s&uOqMp^7bS2m4fR0nfhg5^Z4@i~U&C30YS6 zK@?jjMdRhaQ>opuKjwbT{A|H-AFR~pyF~swQQ{64R+TxG74LR~A2JA*ZCwDalh&h& z31zIfsUGI|@;>I;VBw8Ti|Oh*FIX-fOBVdwMtZRf>tC4&oX%;|ynGIs{_{G>m@cKq zPCS8;x3lPsxm7e{mMrtwJs)>%Im?m`pJx{W!kK(?7R&RO!ucytL*tJo+WRku{TE{h za+dq(n^9BPr^7Re$DJZBzwn}<)nqz5cT^MtMCJ+iZ`cdD@p*8+(UdJ{nuk`+wFmUz z4XwLyjvgrwLV*%|ru}UKdG<1$kU$$&F0Y5>6}`~kJ~ixnNR5s8QVe_j5`}g@C27(D zQFd4AINQ8`2UYUoca&LopkTfr%|}JKpQrjT zLluHIGRtNi?Au|8_ABkh#+oakd{hibKRyU{2X8RNFRd`x;)J3KQn?rrWAOZXikv(_l8)PEz=l1=oX76IQvz13m*~Gp=xbnb+$q-qFCKQq{h7oAW=qtp^ ze! zX;2lj;rEJ3ywe~>O@)0_{|FyGcf(13E_Z&%PU^l)hq)Y&U{5?RvM!!AWZNXhY8*Gv zU*1>ff7SWSL&p(pC(mJb12V`dn<4WWlE6AmKGCa_T!rV$9zpV($;|rrZl+>H@XO-6 zRZwBbd_2~HZdE6|2x?=4mD8(hGi&fYK9dwKp-dD4#o_pH1d~!TVQ+h*1pzPb(f#jV z!0VoS2x*-{`_HJ*Z$Y}S@B9li{ACWhmFNl`F?pO;?rM~>+zX61wLrz6S~_FCDot8u zhIGEzfb2aD7CML_Cr}BV?0*NI{JYcs_!{Bz!B^z5#Vc}G?ilfEn#@*|8-u*hExc-R zCyEoSBHtc9$C;@&$P@ome)cwjEl)oS$73{!Z&(bO5?Vp72Q`zPZC}WwR7>8m(gk6g zQ%FR88awu%EgPLC!UCdIVCI+G_|4|AkfQsMd-zWUoW^{{`>(VS(>sUoG0|d}d}<8- z=3<2PO+S(}!wJONH2@rWM6|o-CmhRnfQv6Q;p3xn@as=^@jttdX#1Bq(rUbw%-_U2 zEg#pDC9BH_x&Y*MXa-K;tl+@LeCRtr4%a+4A&V;a5c|p2)jjc+tawHkyKfXsj{FTG zMPx0>W5z_|vMIe-=18WLzM%KUu7;#rCvd{37eskwIgz#7LRKZK5>MC3xPF{37UREx z1j#kRA}UV~wpkMUy^yCT=>!^#%QB zvV2XPe)%>z^!yjO9WEfpHuCw}N>TP=&n4zeB-X#z5y8ywxXmMoP`mZy&Fe`dK&*x5x9b!0Q4&PHN*#Co zIZ2g&`!Ju6t8k!DpS08n^YEcLdS{A7ZK z`Kwl@vA_X$e7%MbAIKp`J2cq-v=1OTi?EBsHt_ZK1$-_^iX2UN zN5&viQnGIr8|{aQ>FJZCRcSntZ5AWR-%p{9mF0NHe{b=Vwpu*Bv=vtlWD}uT2a4!h zN-nz2BKI|=$XU@^+(4#QzghW+tSO5#zj7x7q}^xY7Z<+6gWqpiu9G`jEw&18@QN1B z7{8WjRfc2xy_<07xVKzPgb;>AW6|r&K2*~om7BIukz{GPv$XF&AyhAbh<>PJ^E7+7 z?*i1vR4-!G0OE-wpcQ2$(PB%98&1+}<~bAKZ3yE?=q zSOL$~ac3u9h)|EY={S8wF@AhClp3tKCETO00GoZKX|39HrXX5|_64gFFL@8)(<|?xbP?PpMfE-iG!+rPw|gdpDOw@L$QWq8>;Sq^BIMw`@vub0mW*zlOE#_wfw1i&%+M`? z{z#bv&a%$%*Ks_IjWQz{XMKsyxedgyjL&W;r?Crt_o+y12wuBajHsDmCMtnQMT!zt zr{QSZkSr1(ct-7O?~pye2|K@YGL_-5s?$r*rf@0vC=u@zM&q23(EH;R|N#sUjqEmMTqlgQ5KI91dykv>9 zGrxaq-AjrTeaW&sQ=%Jd%a+eN$PA{RB7WUokQL*xqQf^1~judk#m2*Eiy-j5#DLnRnEd zyHTINW$?Lr8-1!|4613mc?2RnoWgB|kT7Ml+V*AtJW|N#(r~ys0S)ELXS)Pp7t150%SUBV{)H z6!}fFvX#-F!z51VJp*W@Vf^HciUgYk#`yV-8%%|UY&!&o=h^)@(FHDHRL@MJRk7nRJMJq2b`5$ zNhe-uq8I)g#j1Oa$)x<($n-=L*QRn9>UHIW+7(Xl->9!(X=K1w#Jq+JnfVY?B*Fq` z-3AN(yH##{3e5i4ARO0xO5lFrHM-2Zy+3QF5Fhmz7%c9`e&0TEZkKX|J5F7I3;*iz z;*0gT>rySc>ZAob9*v;}cb+2^=U|!@sfit0PNBC`oR}i7^HOn8WdryL$T@`}m4|ic zO=22Uia0U(T4#{ja0&Kjy@xM8^=$CS5wa)gJbunpSmbC$bU0}m&abOr#ZD&#BXUwE;GG(L7NA9NP;x&LFO_|G29 zB6V$8NtX(qI^YHk4fkn{;zvB*NSAIgkz<8^Zy^2p5T_9tM`|lN>AQUia9%POFPCzI zD@%;X-AQ7kbbUUzOl=gK5Fp8xm0hk{o4g7F_UfU?J5>-`n1dgA^KllVF--ILfUrmU zF&6lKMazbR@!DT`f~ujV*s5P2YisS`xdG}-81VqEehz^bJ}%V@Za1+CzkBR^XaYi5 zlYFfk$I43i90h+qJ$d^E-7wk!^agS{ugi~xu{(xg&8!X_I(-4QDiy*Q$vu2WuK?5w zwy^-;Q1<=c3if)UGT^b3;rNm=P$e-+<0tpvpwQz8hHWl3$7xFE`&z>7<;=^demzpOwRW;5AR zpHLX)dIYHj8;InxBrx13&tT{QT;6B^!lVNHW6OW$1(WW;>Wkx8T={htSFjA{oOuZ& zJwNC{gGT%?p5Gm{k7r5lE7+J%=eWpACn5V<0yB3Nuda~tVM;Qi$=7|^=z%L?!|m6E zn=(S-@9w3bUZT!+Klf$f)3sRh+JE%v1}&yB_!rMJwxV)NB9NZwcYJnYJ{D;Y#i}Z! z*tIv8pwuoAZBsvkZKpm3VVO94XuFy%&WI3JoLfM48tg#x_JpyrbC=mma*u7jl7+&O zl~`|A8$6xUg0%VHQT;lnHzk`6R{sEzhZo>ee;2Ii)?~3$op7Fn5o?&P$!-cX zK=V*7vKOuZS&c?)<*CBHG=*}O+VNEKZ!vz5-bAiFT2AWkoFNl`MYH#o?#w>pG<&6g z9yVspU^4g5!T3=_!h6SUVM?+Jdpa``B!fcXo$4z%>~GK{k)=z-5^INgU zgs0}gjymM~?+Vy>^Dy33CBiJ9d_;i<=ChoK7{oOjxm#!b@R%o0P|Mxfa5XUyjGS$O z2ioC&S3Q2Oqc1G#Y(RS11DHt-U?t@)oc6+-%JfxZQ?cFLs$0`7mR;IP+6SeO^5|<= zX;K9~r7n*bOP(T?^3qIyx&=%06bHAEGiDQdN5dC^CvjJpO+q5ZV~Z+xvT;Hp=~|>n zbgfU4lfGVfe#v;@yw-`R+f{^QHXos{I8zeL_rb%yhvK^7Kg_-99(c`oflpr?M6C+B zINcUAiRX6UZTGv<{J4X#xmL+SvN{LWS_G0YuZmHIY7BNY`iLWnXIT90Nkc6mQ?Sys z-Q-nEA*!EgOiyVo!1q)W$eL~ma>VKh>?{5cBA!>!mEYz-iIW-hEI)%r4Qxk=8+-7= zU#)y@&jQWS&_%tQ?jfb!?zA{$v4zr$)A(t#3i(qgDs;ZK9{n5i7RtrNpw#L(q>-M9 zuRn3YJ64R~x{|fXF#0ICEON&pGwzUsrz#;|=_xF>zX+0?jzy~mCYeh8RH(L^%M@K> zW)$blb+6iqn#TU(s*wv@^i~rO9*%*4dEbNy=72P#TWROlZTRIvZ}#&;4HO?}1_kv_ z&^C;N!cj#eRWudl*=W)F9csd(R2TJh%!2*b;;G(}N#yp;66&z;37^N?O7tgPLD{UG z*4yU6u;DWD&sfsp@b6HvMsXUp-!=+=5>V9C`2}+(Ti808&D5iHFI}H?00rpA3JiP0X%)!ApCK5A1YD<+Gd`QoyPft<{t?xQCvoKxFRU6 z^`{-5dAHKSS`hV05d7*N%f@$|N9!~Mg03@x&~LS!UByB4@3xhw-XtAQ+8{=sk~Fqt zxQ5B@bFl~mDey{^w9s*W&hsbZ$h!l5G&E0z=@j;2xtk9WyKjrmsg%;&#TXv$KgFy9 zG2Nr7%w}KnLjSJpA-hw);_Pfu?tET1r*vJ%&7%OgzuTNFSvUj~eYLt;M^oENd z{izhUVDwm4-|EYvGO9V%>^pEv+Zt|3>j|HGE@2(LYguNX2&;R!12yjF@3p{1#8+o2 zEaP)ak$d-p&b1nR;(ii#M$b{CQ665|-h$rNX`q2!cdC--&lXh2?Bpaq>(O^=GvIw@ zw7Fb)Is^nxVm^r)IK45;EWSn`79LY2{aczjrO-2M*8Db7X1$rXYiz(1m=N!;4<;Tw zH^q0-I+mO<8>FJkQAb)I+Iza1%6$kT8*9SwQU1Ggjl(O@T5%Xv_rAedns;$Q_CzLY zp(#A<62l%;2B5BJPb^cMjaNUuL57P`K(4_EKUsSgepHm<&H)GFp*E5D<@(akI}+)^ zJMrMrZAxzbDI!L1T$sAWIGm9o$qqPqvib3*P<-Y;PG`e4p~Z16T-()xk7QXBZ_QX_ zbZ!!EpSm4O&$J`GYF?s^|gCS<`Lt0>9ZWxwS_Jq)Jj_iliPk1*= zg~jF@fSh$I+Ty`GKg|qS)RJDT#OK!P8eO5t`VSZ5Jq8ABmGOVFW}u@M0~Xo*b2As? zy(bWUHKZb3c;yl)zjKH`L&%f6+lx523=I-B!uJ~oH^D9`K3Dek8*2CW0h5HYgeot< z50mDQ{TE-NW6_sEU%D2TZ+3?o-#Tu=e+n#G#RtFcAI7t<>cYIg4`{Eo9PV=4LKat? zz)_~N8Lnu+-j>QNNBbCju)B#hcn(GJ{Yvwg2cvM%N?CGk;yP^R?hkchi`mn3d6w8_ zOhw~Ng_GOr>1O*sl?o?iq4Sd+`<@xYdylMe`kME`|BkLA6*kQzYs^_}XP1KgTb7a8 zo0p++s#g(lEkoYp`AkyECtMxY1R?)3bRPa#eqk6lva%x*EmTUOGM;lE(xgI3Qb}k~ zT7H#=7Fk&(vm!-G11TfVxsQrULJFm+p`opzz5Jek;r+ay^PF?v*Y*86pCq9&D1Pt> zfyGvtkdhTc&Yih}LAx!v`Edn4;-3*e=1RH#DJ}H7IhPk_AK)o&xlDJ&S$4a72mV|W zCm#K+MJsfB)8I}6TD5rvS@B>uFZ{BAgv}X-&-M3UbKN9%vmulI_gsax%#+v!mEo|r zYYqC>Ncph%SPc4Z$Tj??uydVJWVH7Mc&2cIY`AX^X#s)oL(>Sh#rScT=2iTmlQP{W zWnf%FFTwuu%R=lbH5lStE{<__6HiSYh!6LDM|+5l-+oBx=ZF_UQLoVkRGT)=7UX zy%45}J9CF&r`0}U`Qta!3G}1~tw+$ftc~R8(qVk&pmL11+{+yI_GNzh@?>yoPh|bv z$YH7X(?^1`WZ$p_y>J^zx0Q%9oL8~5N_}_T;hJ6!2@q6)0clMiSeQr zXuL>;mky0!njY>w*4~9saUH($J}gNb^8x9u!eHl}r;B@_X5l+%_=_ zZ*Ozt|4nq_lNK!`iuwN$Eo(Vk^)rq8eqTehVkQ3R^V9sEK_rz~UPlM~8%(u%186rE zL*nXRpr#?myk*AV(#&sY^>a2az1xq>?zW}ckI$2W)gQr7!H~8&^aG12_gMd=Sv;V) zqk@_~#-R&;qSb)CI3hESYo3sj|}tInf{1kFLu-N(IIBSiGT~?K<}o`Z`<&jStU{{|KtX?Nu?vvxhR@yJ7_R%I>IW^_cn#gi!+?~_NWKK z@vlGeT+v>3c}6tEge!@~qX*GVhMRf!0FF-c<-nJ}!LjK(QLDZH-+1O=LY_DG>9v>^ z&_h&xL<5-AxN%)!Jx%buK{Hze`9E=k*zA%tA)B_xB(Th}DAPH(Lz#!ioCV$rrA6L5Yym1e4hI}J-8#0KPlgvk6x8ZZ9LP~Jb#mJyCpda?>1m|sWJ?kUq)gx z^I5mP2;0}Y^4AUH`0vUwmG>Wg;{Psv;BS_1sEl{~&CfnRjI~Nx++SlHU)Xj9Ce_@* zi9LS{4@orNRH;;nN*&xyDUGea=EgSN`;42NrqbQ{Z<$c|5pO*@NGIQQqjM}XsC=#g zy-Xb-WN`wQNga%9OjKZ+v}Y~Y9)i0Dy0X8k?fIT~fnJ%~PM%X&w3)pUQwlP0x9>x8 zV#`JT=<@=8^u7yZZrZ_@p2(8=C-0z7L=;bQh~R~(TcIYt5dVeQ@uVZJXgI2dtc%*h zy_a3(y9)GpQ2Ut5OMegZYm-;NTT=ovAC2M_Td(4r8|CcKeTgr6cQ>(Bp3lGKORU%` zLJdai!0(nK*0%I66w_|9tSFf!=38Rq!Ew0AoSZ#jWJHX%D`zPo7=u=Ke6;Tal(-xg{hl&Bp2y3$#nW4iWxEj;bDKjwvB!` zdz$=Aind#%vV%2O!?i##4^$zK0J%@gC@}il)I$^;aPteOBjQ@<2F>(Dud|nzaUT@vY zqW?6)RxKI2_@WX$wLypdb7~{w3tvI_S_25)_7T@yieRP6jujP)zS0lw6SNgXp=H`*k!^Oe$3;^Nj^ z14*n{$wE!b@!imG!l88$L z1RRsqj~8iciZ@p0!qbQX7VlO=@^`A^sJAzKD4(9#MZX@v@ zfTJ4+adAm3KcPB`hYsH~>W^h4o%bmW&VMZyEfgBWG}U4fJ$kMkU42BDV4g!{&jgSo zd!NGg7sl}UVSgB*e~_Mh8B7{%9LSsFez?;9o?!m*0cs9ej3y~T%(%ZB=)Rak@02X0 zj~lvS%G9IG|7(L-`OneLbYTqHyi}j|ki2!~k-6xVw3r1y2xKL?k=!b-m3&*`0PC~o z(TI#x2wW&a6HQ=+o1&P(ntXQep&1i)nG(y=66Si{hU%*WVa zXkr;rYuqW_VXH_;fCi-h)u2QC-T3v#jwm~~sp97RQxN=kC)_)&3B&ux3NGmb$!aA7 zan=1=yX}z%Y*zYFay80Xm~Ouod#&A0ZXT#0c1i=0D2{@E^|w*5Y-djonL+B;Kp4?8 z8ShSMWHxipi6c{F#knmn#akN!VO5X8WY*u4IOUodG3f8AE$3oa^C?*;w&wgG+l z-->E{9@zomWsh*>xh$~xu!Fhp@6F3~|FP8kBEGXtfoeX`;9-TkaC>eqa?R{gg>sw= z%?VmYOXq#XCvVfxap4}E(dAj`)2hN2EIe6W-Q~g5)4Or-$CLQvj}abrSHapzX~>84 z6(bVHfs))hq5p*o;-9C1VD`z1&YRf`%T7LrfzL9)`A#3~KXWa2zj>9{U+XX2J`Q+( z`(Dg>uf-P%58?W{`P3J7QV% zf(}e}8o?VEOrh!h2U4qr8Z0kN6O`Yi(im0AIXd2+{`gZ0Lru%YnO;tKV(L@atZGYA z3N^X8v76}PHWl~xm*anbd*Je`ljxTLao}c^AbvV)NbfDXL!s!oi72KJsgyxE6qe4_C7s;g|m z>&4KODD63H1L0owYj|Mw2Q;QXB+p%r!gFIKxa@y|&I&Z9zIEy7H6)(Qk}^<_cDT`{ zV=d^&7t5$qa15;Ja}=VF7~3g^hVoWNL&+(%5<5&s3uReXFiiIW!lG#Ka<$?aS9?p& zuVQwrMV)lT*it{i91{JG!kryExzAdO*;&&>R~2T^hu;pv^6D)5_Hz%+%9ttcjp>EH zCXM*1_Zu?u=vWxPC6f6ByMYXiq3y~)Vfm32G+h1-OmVM}a;S`Xw_EehAE|t4?Pjhd zAocQ$MFa9yyn8 zcQ^hBE)GQeCbNT|-yF@IJj%fA$}23}oh#+&l4*Vae*ES8)qLa_sb^Ph$?4K0(8-i| zOgHTK{s0GX=r7>ip#Es(by~{pHL=n$*U(904+c1E(vgn#G(+_hxD67?_S1e~z3Zg7 zanVf@F>WZ5mRV?597MZZyNR{)KDMDb(ss4SByM{k0IHK7h_17wtn@5%`uLvraE*IsBBo?&3}XC)|I=uJ<}mChKHj(BfL*y!vesFE~_$MjQ|q zXjF=IvNs5p?IL%sCkX15BA%|x!sBj_ndD5z6DQkAlTL&1Kym{pcHf6}txLrKF_WjO zY$STt`|-w|(LCt+ZY)e3#L9bo7ng4;M+5s_9Dh7#l{YWIe^O@W{f^0aa_3U`a()sC zb9aIM2W26A;1i~4+!+H4{5I3JVN-8!QBo+G6EnI0V?bx-yBgyN%_wd|-H zLPN@A@yy~1ROpH1UG^FHala1L-yh`v&JN(yV;{0}9|bb8vz6o}_990jK8jzSS}=ZF zgTpONT)X}gd15cIjQ<=Ft#Z%cyC19hi--E;vP&0flzYUC_9x=$%tP#k&m-cdnTelC zD9Ro!=Zn`B@;5R`?8fJ-EO_J%o|EFmJ(Z%^sHc;$u+4>NK5pT!mz?5p9{u>Se5o%M z`Gvi(|BhcQc5#^`!b@~gX?a&Qb7tuTQ0#N zgIhd=h|J;b1YCI05-_q6_5Bv3Yb;z@n7bH@m|azu!|Us9ZxOjzM0*$Pv{jo&+0g6_zNZHoxejH?woBldQF|d~I*UU=aVj{9buCxYK+i90`<1Pod(gbp_+PiiZLhIy40uCW8*zRDw^ zk4?~+{J0xrg1rqd_4!SDZ#QHU4q#MMRL~MAOCZ%E3B&3KV_K9IGhQVJ z_cF%O`Y%JN%ijqgKXnIB_KSiA{~eJ!DamlvEErcbTe0lZ@8EiNC2k2zW;Rj7sbh9u zxK}ub+8AWvo%$avA$ZgS*_x0i?+n@;d0L3XI8`kr*}c?&-d*H#>m0r=bd zEl&0ghWTTaAX5E1J5f^1G|!)bMN5&cOE(6w$3As2I{y@@iW=- zRGH>%PoZ6VLg=o%R`8oS9=;YAL$qr$cI+-CU8(8hfb@o}uu_1fYl2}xN(QX7noVbf z-=WU*9=)M_jb6R9l75)o3CRWVv`tOV{!^O=9B-DgcIH9UF1#CVKi!DiHeF>uW{k#; zm7SRJF#!Ild86+)RkEiloVnLT(U*se>DPQc`;p!u;*f|9&~uUzl$|<9Q(86bcTq=s z%Z_;Rd+Y&v@Z@2#&$%bOxmJO(K2_jSb`O66BZIy*v)a7^aTX=M;IQc=NEktVdb!Yo z{h<}fFP;z=X(oHDc_^&_bzC#gh^09KKDg_J@7Co3_@;9DXCKQ48Qx>hAfC2E_mSww>Li$zN~9vLHpl>!`CVB@3t-5w#M$#+<>9f-gdfa0JdfN3+J4w}m4!9+NS5rqQr~0y^iF z1MQyDLKZTq>oaOOdplH?7b?#qBWp*{J9R2l{*)6v|Gc;Tkbw#Cwp)u@wv6F{5i{Uv zkuxrAw}sz6??{zfICJiG1X2wAF!X&I`x3jB+zp(I)BRF-B9*C(aTr5=+osY($EzW3 zxC0Hgy-VjkQn#O3a7E(HU!WJpxNvK~V(A{ci%fbOCFo7tj+SX3@yjuDHqfsXRvU*? z#bfRCK-^?J~U!gZ8Bi@efmB+-W#KuR1usQiay8M&6nP+YO^dH9utt%|65c zrMadcL(lhHhd`oe*vLW}p4<QA z>7Gfw>A6oCc-N^?Vl#$8adHgrZrTLX*ZI;`%UUqgs}=Ot?1I|t0rdOsMEbzz0~j@n zBzo^1TDC2m4o`RwvKl(nJXVp03{0lV+xEe)_$auOvjXI%ZwK!YCy2a57`^jIm7VzK zOm(jQpzjn4;O6H)L}R@*e-hq?vhtC*I$Ie|DAkkNe1A56+)v{5uMA>NzlV^B+4St1 zcc66XGn=q`A}x`~*l_DAo1c_N?LFFwP3CR5*m)F^6bo?J0&DK-)SIq6AWwob&#;2} zp`ci##k^vJh@&#akGJ2kB^#XugUo19i`@fBIZq+-K@k}LjHYuIY187471Xh~Ll|(M zk*cpc)a-F7xi)z;c)ZqSKW#;<2)e}XJTm9%tDYnM`VXs26rsE`1qRg5rQ@_jVeu>r zcy+~#o;m3SOLJ0~%fJ>Q=9RO}Cp)oEZ5HlcpiT#y9$`J^^}^q6!>OBZC~seMg1xUg zPUTAev4hqbM0Q!UH0zE*4XJNb>!$=aGdtN(u%OyAd)q&&DTM%6Ba*8k`G-Buki}KM zK{n?I?HMA2`2g(0?+LWQ##kFDUY>d6w(~|00r;>S#c0{IeY(4_sR50J7SDzvRL9Ounn4(X$ls88q^cyZN7 zIPupSQA!i9u3QN7>j&|LmC4LCZ5LG=zKxbXk-&fB@4y9XYbbTGB(Fyw!9O1VaK+0Q zlB%c(6W?7TzeDEo?&*8!r(dnKSJxw{|5YWt89Ik8vr@3v9Z^7gyUW;@uPw)YKia4n zi?Hv-KCxeycZ-Ilx4^GVkJhiqhr*e;Si1E%`&p(V)`}U_a%Cd;{@X!c_1Z-fa(>t* zZpl2`YiK*0x!RC!h*YC~*Ot->bBZ<38QXGh7=L!6gk4C@gSqWJ=(c!G>Xl?_U+_W2 z-uQqs_)Xl&t`t4vm-{!-`4dtpFCK0$^Ytb4h_|QKuS;pdS9R7;emb^{$rnwZ?588= zucNCh2H9U~(SWc5DgW8%0Q1JmphV204HHJ$e;k=bLfW3fc*9MCx!VJf8*l~^78lrM z?H)mPjcFtoUc|DFQ^}}neHcDp_NJ46hfyQHyL5_A3oXzcO$!2g(tcN>Xm+(W$Xc3! zhh{9h*w9z(li*J#NREXT-A+^)t;)ZRN`{#G&0yhKMHG){QAIs-+_BPvd#;ru>k@SN zU%fys_b{5MZ7snKb$8g@;h8L`S&{3k)Zy#ez4*<}EWV-d6}u~R^7h%K?V0$qoLzw`HQPk5ZaH2RR>=Kdt;WfU zdN`)@(^nOPji!N4Ds5kI%Z& zT}$(Jvr|H!d$FROPMw{ecR%iwKcq6vWin4KxT z1#N!Z%eNaRN^XKjt1pn7_Ex0mv?D2MzrZF=n8%E}vT@YHM3VaN7m4()MW0DKA>899 z0n6bWZ#N1XUL1q+iaoq;`3d&HrGdDtvE*J;jVt|r2UgmEHmqCx9PJ7&umWQH+#**e?_;z^YnF+IJur_-5y39 z4-+;y$N^n{w6ZteUF`UK1D-y14USn6DwH_8i{QNnZf5I3#*`A~Egr@3+JpJd)l<2- znHDYUzm;aGR#)_xv=17kY4M8n!<)b3K|f7~m@c4?;+8&FfhXAWlrMwzn1 z&p)t~e>-{D-EZVaq#)e5WQt-R9SF7R!Pb82W?C+jFeS2-Y1O#%oP=AvZ;)K2=Iq|Q zQqLbZ8=3OD%@)G9pKl=AvI$nb9>))BR}rSk|fTdq&Lw~wK*9^b`T>ysh#eVlk0lVRHQLHIMT2J6!IKC?d#eq42X7uxPvMgrwCV7ki=sE*yu z9;NOOjl%-SQ{xRVG92;bz9U%aVFQ=`ItwfNB#94NrCC3_49&t%=pCMewKA8vlG#IS z3A;`Xc`jm0hfCgT6Iao^`7+EOxEuPKjYebdRyJdtrnvcr0ydNn;HQ44QJ!Md>Y=S2lU8D`K;2C{G{9z!neN5U@+4N=a z6xVT`<8@rm&H@d5p4#<(%h~2b!Ps2Cn!l+Q(Oj(%$4PgDIftV8eaV5|b6_l={Cy-> zzd0L=R_+9agrSf){0bWp?22>SJ@N8}d3HY()ai@C>xqxdVPI~7K)%S)9;@=`<+6t~ zG02-1*`^>ZSk9lB`tq)Y-HfpN>RI4R)wk3i_9|VcrV|9(#TdE^~V*wu|XZ_PU$}=? zsal~`sSPSyErImd7Q7@m=<4+4dB{>jte!a)itp#q@K`Ha@~nn*jr&D(E}v%U?>1u9 z3Qc}{fj4YCS4kJ87t(YRK|lU!#Am`8e*Td3ZL4*nMSqUc%xP~ytumJ%&ycUwdU{P_ zOWZ++?FS(LKrYP=9|s5Lt|7;sos%9CN7>r@L-E9l8n((ykcBx59U4+Y`km8rXQJ5+iD>gpS(t%iau~(=>YjM1ay8@Z2tuDIGJ|?l>i-nHUEiB;YZM!-Bnu$}&JmM~!h^|`w=$h7uvsN;ZaPpyZ zMX293+x~ez!oQ`iWz)p4GKvvEhHmkfh_M$=>r5L1||kyohe4E8kzI za|c$^?X(#@M!sRHM$zP!T8#K><6Xfr-crn&*T!PM6tM;42ijFjeYxlvg|Jp~!&w#$ z#w&w^$(`uo5D=w8=dO$uv<;%fzoiGUGOQZ2oo3P7yMjs6*GIxKw>Tkej|%QdRTTXb zKiO?~Fb^_RYHfq+>`BYM3Nf+YP%!QxLr0m7V!QntDi*J{5uRIKh1sM0X!OigG+9d& z7f1V1m1~mIeol(0kp8>k)CN^Cc6mBci*;aI+?&Nqb^YlyPelq2M?nb9qPA@Xusbdc z23^dC60@t|_qUI*FIa|O?K2k-Nm=EXsq<)VlLIN+evP#2x1Cip`wYKyD#_T8W=y(X z4w=UelJd}5Y~KbPI$Yzifr|;WbZR33KJX%xw23hW*-@Mmfr1wWv|r4eLer29b|Nme4coe zTuERs@b5j+uH2uXo~vDhPaN}lZ9&4q{B0B8l)Q z(<0|z*Fe|A2M{WgOoxu!4f78sk|(`VMT?BFLcwS3lWCOtF_#Dzv zJ^<#dEr;s1=X~tJ6@2WcDSQjt&9#Lg{Ah`YjY)FQbw=P*=Y&GXJ#GFx?h?-Z{srrs z&3H#nCX~wnx|rR>Er+kd4*Pvv=4K)5ezF$soHm6qW;5xWji>3kh1=-4gfz7JbP2s2 z&cclPOyO`|8p<49#|3>=x>$EE3>v?N+&&(ULB*CBqNsr_Mb~lT*6VOg)(M;xb!bqy zC;xa$n`&Ad!W{kie0#z%To&I4vlI*Q_dhr8IOU_@@?wbCvc-_Ap0VcNzI99a=WV#Q zPc-SbJ&Vm7`vu0$Faz=KMB1^dXJy6rqvXqpyKJ7m0$jYCO&Sir6h>M_av?7qC$4QJ z_gCn%%ElCN$!#tE;@BFyQpcX?-WZEL?IvONlWP!lcs{Y8e}N6PD91m`dh^FEs(fKo zf2bamjC%Q9IAe<|%A{QsLV{|^;U(8mySSBk^__;7BkrT>;SpR+N``Aa-a%&mKvcTa zj0JU`-1X-yUUKCa&M;5KNsfnw+O+<7#lnXlTRVlvmukbAPr110v^!mjV~(zJko+wIPuNF38Y#n;8@T^2lS{U^NKnU907GGWZ+3@q)i z;p=t3;M}-E;jNTg%NV&wc$BBaUwLPNp5adN{Z_k_-~KF+ufzC!)|<*NlqKiu^l7+3 z2WwD^gxJY5;M0FY@acv(m}fbK8){_o$Pmf@YJ3*L4UO>Mu_mm&VZb9?vr+4j9^ZSf z0xNPy@E+l7NYvSA@*{o$t-7<9#x}^>4>@`YV{hfdK00a=msR?P?eg<5R(2PfrUYS3)I|Q`d=0qx z2gH9}&}D|>6Lq=Ku3JRmP8CW2ZbEuaH$ub6 zt>o|WjXd&aKb-S>1vX!6CNCV9^7`ac*qEpZw{D!nH;Gpz->U^USg(g|rvvCZcXxK9 zza7`UxRU4e?omlnq;6SwDE^zVk1SQ%P0F+?#0|VdsK~a4Uk^+0fU_x&u5G|6!M||V zpKXF>=7B@lzW*(3TQ^roci04Sr4jJ&Zaujp zR#ar#=~gr>azW4BP&UHP5jGo3yrkwFc)9)~9J##;Tg;{JyJ93J96LtBLwdl^r!jDF zp%pH>HVBGbkrZ>e6BSYNtSt$hK#c4Z8Dp0Xph zzm?fqX)n3it3$AQBm)7pr@`k}e`>LJGhM&w0DIQe8?#v++we7+j~I3tZIh1R9Da;+ z=5*3{&nEi5qnotdTfovr=z-bKF`%$dm)@UdL1#FPWLl&DkS&b{^h)Y>l$oDR^m!r3 z_f*4(NoNGl)Ht3HY|EagRVeGUc@18hAuPMj1noZCDXncm8C^s>ca zI{mjCJRj^Pypie6V}?{hwbe;DvGSzdAhVlr)<&N`={-~|8RkwcHpbCe5phuH^M^D$ znu?uEPw}Fs+WejOA+jph4E~#RiOstc2Kz2q!vV)uVIwhu-a~NCd$Zu9 zmJ?1)Q06hex`outEav(~Ue8Tp ze`gQ%J*7!b`=!%)b@^DfQR+BINoi-x1Hh6O_Ykl8S9R3lv+dy&dqilEi@59!_HX5itEiO-B`W)!nVdbbxtADwlu zK$<-V$gBk`{Y`Y-n-U!0;z{oQYb7SiI{{1^X|j1DSnn_emN*m5@BSuTqZd*C9^Zs@ zgJrqQmNGD!{sR3r8?lS?EuW{uq0+NxE~t= zl{fQ=?^|=aVX8c6bXiE5AWEkOzJrqvoiIIZH{H5BAFjMQ0Da2M*^T)oqRuyETtI$9 z&j$K@!(wQoT5ZmT^>pkzK^1VZzaMvB_ldVeFSZrbdViY{=*DS?Lba{1U0{P z0p6AvVsO}b!N8;nO72UYk2lNd@a|}Oq-TG${l1@0u-5{kl;x-{y|vD)nQfnZ^t7Ec z@r3}36jq!TM?H_kQ|%xbtjcx8cYQx#;*V4~)?=l_^4SA}1~E7nm_@A@b>YF88d&{j zApN&&DeAc$gA7~V`N3*R#i*v)~gZ+V<=rZ zE1j*Ye2ilUe1HctrelHG8H{_;ht5urUa+P`YpYRI zZD6-w{j;5GG{-$we9eQzgm!+T2H z{GJV##P+QGrgxU6_AcQK=fC0Z<;y0 zZjU!`zmaYDaKjZS^NGO^CvVv~U!Fk5#|+^wK9!NHk8iRXnPS{>{VltB_q(uitPei( zD`Y`7df=gF$+h#-C~>z3uT=%G*kZ4pPi7O_f9Q~Sd69^P>5GNy+Sg&@k%bu4^M)9q z63x!Ok-QfT#w_Q`R!Yy9@+)CaAT3p5>)g2k6{Qhu-N~zXWOXq4&(KI%bLuCVxHbfr zYaA78s)|^jpm|K@%wwoOa2|8l1oIDz7xDWQ{%|AjI)E+!Z$9&l9$7b2MZ=d_^kp5b=%rN<3rggkU2@ z;*+liuDx{mL77_ObaRlc;mr3;K?*873Ms=iGgQ!e_j5KjuTq@i7Qy=bnL-z>ZNl(l zicmNEm~iFnG5B2^jXuLpqNUt6{!$^5-|Vgw{EofC|Fn0J)9O9wN#;O;rRB}JKc zhDOr?Eq*++X&!m_H-tGW9OSuNr0Y-hqrFseSjLpoI5x>l$ocC5dj{l?Njn$ueR9TJ z&+!#jtouX?#4b_eT!En0uNy}7^JLj-XZU~n@1kEXiG>JriN1~GjC>R+a*2`u?eiLK7gmS$8nS#Co5|R#El7qXYSuv)>B1DE-zza z8kD)_`JwpV*(3bgdu{HveI}pat0Nw3O{72S3L(qdom(qSrZ*coDqWB`%GWLFtiIRC z#~v{#w{no()17^&RU3$b{f1O7mUb6AZQ{|Qs+ws9pJ7Kvm#`aj7I!V!&AoSC;0}Az zSm_&yXVm_h%8r~0cNA^-e`BwLX=)Xz_CH?nUCMLhP0*)m9_MjJP_TGnvM#2m*2DAw zBdn2&VtwCq@uaR=-skWWX0C7;RxBI`!%dfqUDphF@|RA$VmgrA*qbE09(J8PY4jF0 zP3**hV@$a1uxJ?Z*^+u#rsLF8$q=C+aU6=xVR>T#2@V1%Z`eS;NdC&_EB3R{2RUGv z)`tHq=Ww6FKXBuPz38y52A?)uU}||WWMhaChFV=@VQNh%KkW|541SK9r3+w1j5a5# zzPwre2eGqn#(}4A;nX43c7Z<|1g~|2r1$@3+Gp`HdR!+T7msRX3sSN%Kj^ZcvL%5| zUG|esf2;vh-YMhd)AHcA?jUBbeakjE{Kgd!O?$ocq(L=BlAA9YQ(I=T+Kd0u5=%w< zjKxi;pOhzTyDKXwz3jnr-n!wMg&stqVkjLQsKzb2p28QyGQoYDneCEaUrFxAMl>u` z<|_LP$U`te)sVe-S@VrBdwMo|o!lvOsIIi@RvnGY55(ix(f({~;b@-S*+xS8yuf2~ zGI9AneXPiIBZeL`;QUj0-abp>*2qUodw~ZuF7*>_z8Hfc-L&-Wk7 zC#T{Q|ldaBcd zmzj*KjEi-K%GBvZcl9Xf=`10CUShO&H(}erFT%daC|ETz z6gDZUSL%4(!)lvczUR(b(x>PEy*PCQoym?v+WE2UEks~sv=Xk48wXwe58$1x6if4G zq127hJv5OoINXO;1drmhBZbkJM12b!rT;UB$=SkP*O2BD`>?R5m5zo{Qpcgcns*?ZBsJDtSyer=BkOf zYm*JVwCyTr`}TvTj(6m0xDAf`+#s$a-W4-!=dr0TAEM8nEUHEmaj>V9lg)@^<|m`E zLmya4mK<-+j3wzYGnv`bENc2%fH5G4CyZN3S}!fyAP3|r%;(>OhI3?YMtDi3?ucF6@1Luch z{C;)V8P$rNk5Ab3^;Tv5deyVqMe>xaJqWSK^U(cW5I<|Rf|r>_RNlUvQh7+>KJ;JN zf?Gl)(Co*9TtQ-H^_pu02VC87O1IRBsW`<#HW-5QS{aP+-V3dZ2BCB1XIy_okuRCl zlgi$vbi>1O^p^_d3zSbU1b*@<=y^IV0_f9<&MQafqYD#TSVAN9}O zW+yMTvpN-PzNW>KjkVLjW5;K}>pqQ;_bnbv+8)4zkqKmN&A$ekMGV3DFe+}b@0BqcZKO?)bpu6;%fo?K+=Uv5J4Sue8Xz;c+l=^`uk^u)b( z6&2V1w2&-H`+{wnb`Zx>Ocv8@{DivoeOo93Gtwg3k;ySOM!bhqH^9arLkn`CJFB(sQRaYPL zV^9C^@3RI~DxNfwSnC_{^+P>+ZTSz-_~Jm$8TS<@?itGp_m`4Q?~b9-MQ50}MwYhC zc?@@cwZrbtO`v+!pL@;zDNV2R%Rk4}C#mqgf4y-MTSCQVt*;*Vb@(~pkF?&M|IYt}7%e2a2} zA@^A1wpf-MsDzc0N8!VnTQKUlG1Z7W$5%a=OC+rtY+LyoPwYB}&Z_zR!Rtr(Zm%|$ z{FqPbUnR4tk2=whY4Jr5*YK9IGBP3lCdMa~3X{f;Mdhj;EWfY|^L0Luk;k53Rq@oy zjn3yl=0_MrUo8Lydt-2oDIs5LUtsUNp?K?r4y#-Ifz|4TkVmo-r}6tuq3U-d?$kek zrAw;Vi?DAn%Q%*O7=?WL`Yl|0)Dv`UbH+~8AO}Btg3c}13gePz{OfSgS@kfA?|nc&==vvYr>T9N9qcRkUHS%8!by8K$)7scm!y z8w61^&Qr_f{b}OcdYClUi*ziT&nq+(D#!13!s$n^!2*?G^pjJP*tSBTUlqsDX`A-* zm#N+S?&wNPh&IA!Lne^3&zr;vAF#rshYjrMR|&8BjTa}(NTQja`tzpl|0y~XcPgJZ zj1#hll08d>3PoAYJaa0tmG&e>T9CBJl7v*Y$SxE~q^xP9IA>-gQX-WkDMdvKMQQt{ z{rCR<1J}7O@60^UeSbcmd4}RHcw@gLKEXGIz)8P|?#iuHqIxh)~}bL-=1v4t}$6K@7)bhD~l&a z1qyiLqipY3!5OFZ$`!A&^$~LH%}ju=OVo8nJ#AOW%Bg zF8(_ar_DW1GmaZ~Q;4z04`Kc!)_?-qSe2s|3EG2S))x=C@`CelD(#!sMUjz?t>P7UiixkHe zn*dSA5L+**MmfVAp2hVauX^sI6R|tqbIFpdSoR3=<%3CGvn*fx>2zrLsmT8)KNilX zbPILN5jysHCH?lnp8cJ|anHoKaQwj&wEMInZoKG8^1n^t{x7EUgmQ0TBs>dkm^YP7 z?$M+YR%1boV;$xvT_fI^=IGwg1>uUO{X|5?3sqYzK--#?L6NM&YsRQ^t~^XXR6L@G z-^fs@rJvw)@DWte#O;H3{ec7*IsS(i1^7~26Z1d+{jERFewzr){sQTg`A4tSCgB6F-Q=6=SS-=HfPedp7Oq>H4JV6Z z;ntN*zLEVHzFcG!J}~JOS{i7Iy)TzwU-xLd%i9X?UUL>KuH8eHM^tfFvKU?wv=v+Y zR{@d}2gsGr=fJ;T7oSuMrnL!|VD6wBKI*;z4F06i5`}s6N7ZTi*2^Bde``fuxhAxn zTJfnw7EGIDjel=DiLyFA^Qy}KL2m=+SNVJdg^Z4()@LUXFEkq}U3EyNnm=wVONYvZ zfN$y8*W_tL&@QDv$Z_0c)|d4WNhsc?A|e^QuW33E&lh3dMeFE?3zx~>^9>~S*iW>o zZa&-^a~I4{XhCY!C3M}Zmh=bg28$J?yzf)WknZf)IAm!mzEmT_|E;kG6uO02EB*ph z{#}E81}ET62Rg9_+Kl@ZUtq_SSvd0E1zw2jYot8&6{Kam2o?ybZm*glAKw6H zlRWJ3EdtCcEY zkFa~7>zM$4iW~9gN1{A43xBH2I1XDt309cD1!t7y;rK)merLHZ|3BNyI5jhfeZIiW z5$!k9nhOQ!?6$>_zFz@Lr8FVfT?60dIpLvC-*NHhP;^=8G&WomiVv7DID$4|)klkQ zY~Tv)t|QAYjjcnSI;z-}Z-9T=FQCh9vxQ^#&qnvXh9GXjB5YQ*05AGoiThe7;JAN9 zATE9wOyXPk98QE z%$kW;AASaVj%u=H2ff%TB^A6uPz1wY%3({8C6;}YLB~mIz^+^|sBA7nOS^@rwcrHm zPF91qAQ9;He2rGGbcbn9VzB+A7C)*{oiC<6gcg~Nr*PaHe>v1=H7Yt8yzlDb0NJDD z<(7E5cTFz6efbvY&vv9Ye!YZwj(xn%N_tRWu7iTwOd3jW;X#>?3q0;Q1`gVV zcx`nSeV18Je~yGOt+IbqN-=bzh z^4=_uLaK`@AS={S_;_dnD?KL;qn@sKvPL;r{{rY2?FPq82k3gS8f-;RfLlhMkh&Dm z427Fyx2-E^d^!bN2aiIYU>-SOTEJ?=lW9@37Ag_nhhm5ISau4xGg>eLUk~TvR|Of^ z`qUZpJz@g3S#kzx+!+hM&z94+Fbx``QIA}I{($BwOL3mS8gKKM#Ewa{5sTnlT6)Bv z85R72`y9ixaY_x%*9Z{`%_4|of&!K+GGsgVZ)a9-R9K0_Ja~1#hn{`gPJJf-Ln=K6 zU@|F|=zXYxu#!zMHOwBbO*3S*e^)Zm^i6D{R2DdR&&I{!tA!hPX27>m#C|@Qg{>V% z(D0}{x;@&9Lie4f&GCN7|9JrWcr=q5XSTw%^le1l?g-heU5+~M6PoUuj+Q)#Wm{D3 z!R>q`C8bovyP@3x|8qLbLnB^<*(sG0*vV%?O#QDt%tJwk9P>N^)`q;Y`P;?a4KWPpX8Fvr#x5U+L`-EjgA$?9SDT$W>H$HSHQ{cy~3Ku z@l@7w2a|cBi{#|Q@RaUVtku08JZ7Z8P8N^m2R}i-p8taD!7^0qfB_!03J3R)c(gJp zmU%P|(rFzhAuHGxR0Crn_}(A*eeez`^Q`6V{v!qjyB0^XRFX)?YTTxYTPR&*lt3049AiBe zYuI|$gO-_UvcT!bI1juvFmnOvnK6MShHKLimD3zJcFhRFxIgb(kSfcn}oblUF$ z@;EeuT&RhpYWHo}0Us}VCR+g>*o5Ga=pJO;ijb4-k_Ta~4 z_|hh%o4Q4@nH$G43kheXJ> zY=*Hu-RrE&fH-|Ru|3%Gb=~=}sULceU(-DR(*fS55k? zG@DvVC9=!Yq?v)vLzZzdgpF>!07l`HU@$-h{oNhO3#yZ2_MGRnGWIL2OFBY^3TLzD zK~uoaZIjS%g*y_UO@xS?_xIalWkfQ#D=$aL6*l_w4h26d$fH=m2>3T$-n8y$@v9cc|H_h zog8HK?tBrJO6GD+uX9+{<6IE0yv7O(rq_looK<^66|;(#tBeCmvc;eUW+8HDZ0Up= zP3=WY?u8>wE0<-f)raBuLlJzV?~sto&yo7jFUa?)rqxB?ar_B`rupMCR&(!)y^8<6jPs)|C>2&wZr!@+5lXof^ct{-HVOAr7&x!%lDH`QP^BV7rUb zwY;sS>|2T@@_lxITz|F{6q`iwqoWE;+U)`_uEG?8zsI6W;_YOwhB=ktAH#D~XTVo) zBlKXiHOKw#rUOf#l70hah&*QwQ~r99@wE@p7mjiL+Pjpwp4~*x-FiU|`ERGWized< zZX65N(FFFI)WWSQV zf7+~>p;#uFUo+YED?{`+&z4nl8L86eH(|ZBE>&r11-)@EAmP|k+Wak_ zCe}x@wy819WNZL&N}mY1K~DJNZ^V+W>$8lUG&ZoKo&KP`bj*(=y4#|HXg-Z*I$X9Z zYH%0Qec(+ts0w(#MIy*=Xd)FYeQKp7bfWmmS8!g|OapD+)3UX9>EH(m7C1pj4>lfW zYH}*{KRQ4^jhEti%;s2;9@p@R8RpE&SU~5U=6074!^wzSK0T|QL3KK!>7JkYgg!9m z-lJooc8wQ(!i%KA+-_nLv%s(S`$2GZ9jx1|&4PX&<$B_>oWskR?k>BHijvcbDdAYF zd-qe%=Lgxg&=l6G(Lx{Z41=LJOTcMC7X4Nsi+?ox5f9%Hnqe)=M%1IwF24gbOBYzb zngg|rY@|u`2arU)6BD0QOAq)1xz6zjk8M_iz8>3^>(0LW z-=OY46JXM4rO@;Ab$YGQk8R>Q#j(v{ILu3p%?ce4&p3ax{qR3@wW$pCa}Kg6)iT7a z(2Rx#o#5?`dkryRDy&C+GW!#i4DI7|>8T5g=vVi1D$%XDPnsxQqi{)ooYJ)P%+^&Td6K!T)(@MJoLuuGEDMn=Gu_I4*(@DpY z*zcVSSzF6NGW}hW)%Qn#=(P=+Y&X|CN!xCU8fr%<^XjDEw=JgSfAeUkbQn!Ab7dCo zGW6A<&-7H%6SPa|C>jh>VrfGPtZw}vD^#~-25$Q)^3!F0?ICz|26vxZ(#kUO57O*b zjtA)2%hr87&la?_Te)m27H(7sg7ow0Oki?`zT5GcI&J8t>#R1Sic>OdoW@1E#50C^ zE1zOE3A)VMQ-MC}NI~NN7DDusV34YOO}DQ6Oyd@PrhXr`vlGe7$;&4IX!ID{gq5Q^nfU}{l}i+3TAN1gvNgrVQ=%c z(xqG;%lCT>i#4$!dtOh#nsxt?=s$2hxgU3Oi19CIx(;az$C5M_S800&-)(#0R^Y2fWQ$mwr0)ylA@A_*r+`gnb+zbF+| zc}Jq(fgVJrQa4%^- zDEeyR2XFu4IeLYl{6Z5fN<`sG=YFVpwiOv(x231{uBLguBJAmnG;qlH1mlI8R)@;- zVe+SJZ1GkNel||UM#^ez<2f(*q4S*XsXR_!hl>jXS6}5h)uiyIzDTCqr3;1k+w@?( z@eDXrD~|8(`3mLDlC()qnflt@gE)_B5N)|2>{WclQy>#SA$}(vQJh58WhAhhqZ3#J z7~q+kMu2{ftzKnxktgk?32Q!Yg-hpJVZwyj9AC?xSGBniZq9r_{llk`I~qrXKgOPg zyS)mqMb8xK#b2U3Z(8wEr{(zF$5`-jD4`b%b5NX~8F`?4h(9=Tgzq%hm|qL&m_IVe z%M8wo(Ym-E-Wn1T0-t;Laa9F>0HL;mQ< zqt^q&yalUT$UB)_9JXgI&RbT3)?Jeq?3G-JZzC7{eQpA3;>DoDc8{%MR&nRu(@40H zv6xM?zYYbX%`ms#8`AeF--^||jbqC3ypOtkH`jOc(4uHq zx4;Ys&Hf0x*F^~a{WuEit}nzg72}{Yq5;nAz6;fpIaff8AB6Lp>9t$iEl<4|fLE7) zpnp$XVPc&p&J^P|W`l>R$>MapJSiUkY>4ArbwN;n%829pokDl7q`_^OV>D{399<@- z1V{ND@cPO!bY!;{Ry6Je5!nZzxIh=qHD^Qi0iccLe00|NF!|>d0~JJs%xuw`inL&Jxpc9Y(ladJW3mm_XK@q9~v10NUi(@g_$| zfZSlKaG+(U(0t7oboS3J=C1=vD4vRA88kEV5HOBcdCyIS&Fe`*>+(f$<<>rW zy+9jX?GmB4HuzJs@zVIVkpOOQn?mCG&M2Ngg7)tah3WH8!2Y+1^q}-h-sS#w)E4dx z<;FAk*P`_JCpYfHcScviY@1vd5WA0$PTvk*IXT?@{|;=QB!FChD^MoEbm1)%oM&mw zlFy%~kMhT0E$3arh|XPf(&&T`GkEiZ_o(os4e8c@gjvRYd~lo7b4X98P9HbrqY5Xi`lBGZ)o0*KV-S-A7O-kF#D7w##1e~q_(*` z*nmM9eKp>M_U|u+^_ffP!cA*&sLyNO$Bduo?MWj#c5wn8Z8V4P`3u^;b5j^XsNr-o>OWMGB;jKZ84lPtZRz zWt2ZbhP^ItCYv+=!)sg=_;)xe<@5_m^p@!mkH!nBt7jTc4k?2po7HgqKqK_~L}G;r zoTD(@+d^}zDtZ$0h2)Rt&U{}7zVuXmel*-csS(_HW|a@e)z(2_cnNKvBS*8(ThUFu z6gfmX!KtSlE4;ptTy!j^l6rAM$xAy?9CwfIyq`qvxAU2QxU7}({3cla+5|QlXyOaH z&q3F?6y4OR5?IY~Jb!;lHj&HGIv;T7FKS=JA8ZIGc>ZRgq3L*P(!g;-W;}<& zn-!?@$Rn1rB^TY$t0BDGm5e{l0~ZxP1HZZR*pD?y?1*9rQ&(vu>2qS)Czl68rRH0B zhJ+G-iX-OxX-z}NVmpOVD_Vui8ilCi=P~LR(oa`54AYR1+0eH70Z&y++v>p5S3{aRaOH|BIT;`%iXRmr2#>&n!Da9_UW^rcrO zN58k2!KDC6JobbLefAZED*lrA)1jNVz)GJl^1G4NP0tV-6H%t$@dW>g=%@SD+u+|f zc_zw#0LGM?yJjg;O0OP&nQ@n(M{F5unLdxEgoLn#<*G~}+JPzk6Cq2?^N9L$ zYtpu5CiA`VmsF?xL2Z-7;o`t{-o8tBN#p!RG;rDnJiL#?&L!sb#^x1hm$r|v#@Lll zZpGo7&hBVrWfuFHo6m0A#<9;$e{g10gYf-EO@35o2RWKQkNiDQOHM5dA}^+H!_pm! zur^i{izOus@9fc~gXUdace|DLel%tt3vc6wY2#_Y=MQYNOE%3<{STcQ+0P^5%u3h! z3|U$?jXph>4=aAi^51Wt!e1m(fj%ahlC`tCQL|PJ=oDtsm6vwF#NMxDMqi#)ZCZ5jsP1lcAH7w zY*|U)dC%qBuF>N^IeiK@=Vg$*`HFZ&#~5}mkL&X<=p$WU?8&oVGR*&;0)9X{;Wmei z2hmX6U8s(}Hd&$b{Ta}cSc;;KOb7G=;f*_cVd3~;w8KLhzgwAyKU^-R-m{+3SKOH> z_hU0r-3+ws-pJMo0DNc^t)pK!K}C~dD9K|N7s@Je$A z3{Nd2I^&A4F!wuhPZ6bI#hu8vIs;zVXbV>V{LC}0+KFm2tWnj6+rmP-UCeHeF1``5 z4c^@h2j1{k_!cjN=a`v-L+1>9Q!fcCS$JWocpu1I=0v_uk;R+M9C7yxQ+8xU4810y zK?h6b@TMHyik3}$B#h|34`s(Ip|B?!qQ>myy6zw0`uJ8Fww2K+3qp*qrBJmZH)Ou= z5sg{firPMI@ONe+TBi0-mQSVv#P);i%?hgWM(<0Sa@)HHa=&_ zW$2AJVtvU3yhWH_qiEF3bayXg$yMCn8G8fDO{*b)rW`(7tbmEh2$X-h01ojYP}zc^ zHZg-uXfDClJ0&$FEJiDz)YlhVK~=(A%O`KDK0 z-IKKtZ*pZqH@yU;UvZM>b*~3KHQY#7KRSVIV-#rZ%}ly;x(16}+9Fg=dI;t-C*g7R zr>T77Q`(l4$2qqOxcj>{KPD!bPq)dDMb}$-D*G;?KW`{{e&dkP{mDwCt~U+Ny%A;J z(drO5^qe}4-o#r~x^RB;EqFR~4}HyE%|f7%R&FJnl&Xet%XC8%HJKCjI#k*R{pU z-=_{tGJLG2*ruUrX}bK0O(Og&796)|U@y+8yNAbaU61=$J;Tyato(ByOYD4&!rXiStgPVnoHy9(#ups? z9q{Sl^R!!zQ2e)my-K{wWg>0Rjij+KYqdP@>ezfb@F$(*Ygb{{@m1KvQ?oYP@(dJ| zjgZE@$H?pasVKg)7)^yv@*7_AD2yRp1t*Eu;#VXszX@XItHVL3V>Dy^YswFI zp!&A2Y2AHyJWkaJChgS1?aR!_YgIQ=&5MRQTU)xrLKO0QzX}c9KckyIhWJ%O1}s?r zQz-gP0_J!rWB>RZa4jrE_6OC$hjS%%Tq0;JpW_nwzk`m|AA}8xwpdE+JcM2O1Fk1O zQ_WolEaTyO+Txu=6L(6Ztbiv2Xt13s{^ZA#!eO4W;{4&HjU_a6SikG8; z5mnN+T^ZlK@DB2uTj@;)ZL*BRYw6EDhboUJ3d;{R!F1bQWcVTnZ(UQ%YZyKbH5`ji z%uf+l)XG^cR6hjGjZ5jop;)=ZZwgCL6HFj-;Hhp-AKqS$XoDX)mD(Y6dCa+y zj>UqN*%=~RA4(5gN#bFqmZgTfyK!Oj;__?h-4x;i|R`6|e>l?!K)y@J)W+;2QJ zlayga-&P>J#$x_$w^CJ1U%5V+f@YYsmIJ8rVdnS=b=2j3#(~hq6nl^oh|P z+L5M&QOjqPP7m^wX2jE>ZD(L&dl?kDNa3o!|6nLA2i@)6g=QF)Ami>Ss7iDdY>;^Cs`GLl)aEFM?5j4dx0@?WPLy^wh9Yksc)m!@o z5+ZKH^uZ&ypW?U~(7;`J+uudD~`8=p(hEqcfP8|y&7 zC`>?hJ?b1QK$a%(KSE4m3OZ|+3wJ-fAnjY$fwk6KwDsmN+`g=W$><3X_q>Gy=GzmI z8PSk`^A2b(%ce6gRDz;xHUyMZ(8d=psayXDtP%;}g}rIC*6-}Z_z6%g(s7JX4N$D^-Q``e;eHHI}72{bBGd>g71U*D7w#-W#w=r zW>q&D3LglbITii!h~&L{kPZtaC8=ZfMUu{RV1{HByQWY=GRnR{XlE2VmOhQ`xp$N% zNiODvgr(A*%dVnnay`hw`=t=lUz5_a7kNeT0(d#`3G@#*L*9FTI6FNOwGzW)^D*tUaB zh&NPlS>`Z~xf-ns?B+6UJ`O~j7vn8_%G5&Pqbk${WYFrxO2X}lJUF}Cf@iZz41tFj zteg}`llaMS?%WjOZEi|s$9d3y9823%d>qGWv%`84vb8Iml9|)R4m>YCmNq@<6K-E! zN)~-K#{Ru)iSuM*og2jNf;`93GX zmB-FBdg*NxRv}jHT>p&iy%fYYEG`!+1&qaiyrIOM1(`>Gok0V*ZK*GQ_2>Y4 zx%m;QZ;PbGUQT3;n>u|zaW6>LyFeP}RBZ~@#V=+I@fyw#5P4&Bx=wWojhjhGO#MOZ zC=!j=lyLLDm5p>JGJ-p5Wl(wT16xv#c?*k61%C|_k>I9+@JCCUV6bey)uFtNg5Cbk zg4Wyebm8j{wEK8G+M+>WsO~rI5TwEV!g7R_r<0>fZqV{4jITB&g5%kXaJiTq-sqhm zYN+ZD%~Nf$x?DWX%AUv!ehhPYph`S9pbU3tD1ozHJRZtc0rKfI*1lJT-+k!D9V)8q zN}3ok-ivvzCSpK(2#Gc@#BORfG&*@ZoHVPUs?xUDD|kGy5~+q3b6q^!Oa~W|q?_in&1Hhcs<%Z+sQQUXsi zena=s<#^w_$$b6gZ*Z1;I3#uJ;b%LW>D6;qG+5t&t@lc#QjvDZByTX9g z#@EAw>PoCRe=B}Gl8({B-FVIBGgjimQ&DTMHM~(eAzZ}qM%9AO5h+QIwfUkAWqFok zx7}HwU{Hc0UzdPu!%2K?|8sh5|3^6Vv4hLn-ow?(W_aG3uZVYSJG#*5NR`I~kS{A% zu?_oYgLP^&y*^VO|MGhPZS!xk4>BHXeA;Rnec?JFpK-L}fjs|{`)oFE#Dv*&JFw(P zZz70ohpIMv=vd%RbH90jnd1sv```x6lAOoZw%bw1Z@pADEQ{ScxQpF6dl4!*pF~pg zF5GcuA1+kc#rl``3En-R*ji>ah^ATMb=?AXdP@l{sFh+eT`_EhQ7`&?c0z2zzFbf3T%rZ-R-?Q5VYjp&DkJ=mIk!9IPpcx0gu z{;cIr9b6J1yJP{WPP8R=x+mjV{ZFuur~|$a_aXJpEV#OI793uzKyND~V%xu?aC827 zd~S(8(`n6sc@N~+HI1`aBm52SJ3hj0tkb6BdsIQ4XN3Gq99Y)@?pv!BL2^I8hp*9v zH13ot6L4&1zV3PU=!aFUw_+V~N9zYICfwYO1^_nac)&@a;2I zymCP3=VjT!dyuE&G(V?o+UllSq1I22wo!cBr9 zt9W@v%Z^>AHBDyBX~YSCP&rB8K0e2qc^mn;oR{dolY7}RpEc})xD)G5egHYC z_h5nU4QOj}!hHR??2uXuy?wtIesAUT_)*y*c@^kwLIJl5s}UgxQXFD@ zQlH>JfdYR~=Rq9oV2Wik*OM79tFc3i962-Z8CoNyj{O&K16vQGWgf$3zo|L65yDA0`D7$q)SBWsnioA zs*GFid+(@Uw6dl&kBxkK1IrI}pa!K3e=b)A#7(8znrDP$^tKLkgFf$#Rasb>QN?40-5n z<~{Qd7j`LDS@kx)1pTc$sC01(l4;VxCnwgxD=$&rjO`)N_x(A~qBn&1ZWDpA;+gdN z_KWm)t{n;-lWBQKAdfz_MZ=;8mc)7}1Fcs!WA3MB(%cOOBxUDs>~X#n>T>g-KROxr zgk8jL87Ht-M=CCCorsG<_u!CkYhm@f0M;jWg&fj&0XsZ$A?~aYCEMLV?YlGyHNBpiCWifYOf3M1Gd zLFn!=!d)A;Tgm3PS|z%^;Z2xdOZPOqwODY*fmGR=;6mkTc&E-O79T6hcLKVwzi zhx{BYnJU2#|Gg78Y&y!X=9mLJ?0=g$V6lUF7fNIZON5K)++)*ro zqUuGNd>oIztX&^3DPDy2UHNpl=mY9?E9L2lV$gY`4A*>*p$z{NvdNX(4IZ7x^c=U5 zMlQqucOVUfy&Mn5GO)q~YyXAgs+1?hj}2Tzzz*Kb

5$Zwk#_|Ax=w-+{2 z3G9MjJ7YY2_#|`xw1a(p_lFv*eWiOo9zmH~{zKc77@adozB=+U%A=(?slG7pKQ=SVF1X6OYQ#iUT4**6k(Esl5LZ4`6!(`OFrUD(g3BcP#a zh5bWZa9cgMHMuXe~EW^cWtvZ)KfVGSVkq#rgE1NHqicVt~^WhX) zP~nf=r+wosZNEnDUz&lpKU|LMPbl*5T}k5J9oNYZnIig)%X@2NN@3+^>CF245E1pS z5uR=;1npfMZ}99@l<>|UmTx8`&37&?F^|E~`!}FIg*~Y8)N!-~bIzwc0s1js6zR@w zN3+MI&<~DoOyAuKCO;U0_4)QNXT>uVp>T@cH?M((dMQxsat8KyFQqy$Q7~?yDH6Nd zOCPwsf|^a9s6yMBtxi~ne|U96#M@c;Qltx-li&a@+{|N@emS}re+qs-^oKv4k|d#Q z7BzPK1CRf^i1g=gKrzkZSY`WDGGLgGh8zzg^+4|IUA+iqyY`{k-$VsF47z!7zhfys zQ5(%8d1yI)0guKF!o!w>NaLs@v`6>xI_hu0vuX?6{rm`uahL!OdumCeVIgm&$Q{xf z#pV0|c%kw;GlePRC!jKU2f_D!M!el){^O}En+AJh-XY=o3B1x_7xZ{Z0g{dSP9B_( zqk*xew6L(BInF!>R=M+_Ta(bVMjxoTeE>~4_0>uR^$SZrO+^+r_~iKUbkM$Zn4Fnz zPR2PFgXH-*IFh>t&NZHf^xLcHyk{b{J1h6_K73;6PE-NfCu)J`yaXmQTa%5~8i#sv zIrn$eAf0jMBq*$JM;oU^v-P*9H98 z*KXAFGWqfN#EC25YMD|^qBK)6u$@nbB;Um>3JIVemMoZ{7t80bw-ZGW_jy}{+P|W4 z$wM0)KWiheC(Z=m>rq_&a0|x!N+I0#JTmyA0lV8Vyt;4{#IEb$YpdQv=-daWa?DCA zDV3dQ`Nl~QtumA8PKYG4YaM9*pWD=CX)1CzlmVo&-L?T@>7RZ1J6jj3NMxk~fCEWqMs(Qw*PkGm6ctUJ0L{uw5M zm$yCc_iCr13p60WC6xsqnF7^r3ixTYF6Ej%sMgjFwP=arCzJgxYuaPL?s+D7O?*#G zc?Y2GR0p)~n}MY+T!6B@uRtmdgDLxsWcoJ?O)p+We@C@IuD=fNYmlR|rk<#vbRORP z+zCeIli>Ts{ZKZ2BOG+_gyG~nsG$EMl3G$Bw0>#^qrVMd|Nc%G@^XS#Dk1d4yLjZH z^$Ydsi*T;J)wE{)1>$tZ5nVOaLu>wUyke7NN_7-y`et>@>DLUX8Mhy6jsHQnUpk6v zYq;Lh83T|DIzg*93|k%BU5>UFrLe~J*6d$MFc_a2$6u$KjI&1$Gx7u@({|1cyh;vM zxE&Rao>9V&3Lc`5`U&vzcnXY~XWS5ci1U&KX7(AT1pOn@81+8c5P}LQU2QSCr z@Sv&q-n~`A9PdQ>CgBIYHO(5I8QTIm$GKj`t9YKw(gsv<$QdQ`Ug1LBYuuduAj~!^ zq02r@!){aZKz=Y6f=6;-|MvlCkPBnW4L{K2uah|D`93<;Bo(^eM}zNUYp`F+z~cb} z)sMyKbTWU=#$4=caPr6eHe~k(p?({X^c2 z&Qw~v&=F=iML<`Q4PLU^7@wQwiSy(hKLdU!}+OC$kU-Wj51H3Ue4R*y}hBEj|(r)7LuSxatyoaQRd4 zz4H`W^LQL1-5mYrtBRjbzKbrey-?$R;Rd`cDThXjP4KTqlw(Dwka=AUjIMkGbt@D2 z_-+=8+Hn)o<_DqoSE?wyvnS&+?~>i+TamihW7@Fpp>U{Igeb?4AZ(}(FIK7IKVKTa zGE@Pn1dV~xg9i9}?Q{HJOfOFCGvxo#TZ;`sxS5I1Pk|sdP;mCV1+=GH;EonIEVECM zetpm9W``YQLQ@vbxS$4(n>Fy$%r4|->;q(6GVmU$V|ffgRsWwpM~ez=UDu_3mW=0ySDF$F454{ro&#r)H*Gc{jqJen$W?78NIy_>OB@` z?T(|r8m~flNEsHrZ^CyJJB+R0-+(utz3H>UUQj)76gmTI=;Jm6Y~6nk-8w1FbVhBh z4n$5DwEj3EEZ&@fO)vXmyLKT6s#JJiuFqlz9xV_i&AyAO9$E1Er4CtoSf7QWCVP$z z$xv;NobaBJJbS(J2rK-%f^&R`GcyNW_U!vxqMV^my+*bO6wmD-j{j-k*?X>7sVvD6 z$SY~V(8EZ$>oXCqTTDaLu2k45UqZB-oGJNgFI4YKqCGp#q5In|5l}EFc{{ zgG~x-k{b8hwjW^n=6l#~gABG{z7!3dx|p$nX-xB<9C!8=3a>nn0->ZQY*amrwKOl` z9UuCTspu_kFJ+3ulKD6{C4hbq?5Af8{DqrqL2%C@M0=c z9z4T-Z|N1B+K~l~#VT0l?QE!)%7Mj?Wzmg`hlNXH9huynUY6P|S6iKUlsvg!h+5Tb zSl5LJ2)k-otMGy2x=hX{{}f)6yQMv(=FVgGw$r0F^^!Dg+aUrcpP!-YBd!WY@=YP% zTb26uUuT;t%h~(D4eZ+ce%f{~2_uD**kt2>cx$#aUAJ#C?(xqTinoe@K(D)cR7+9d z#E&NVVcyksW!yYt@DfgMdk@~i9$181S8z{@<nQehH7yKFb1b+tkkaatp=sVv$vFqHt=PB2&0Mj}6tj!9njs zurf&mexzR|cVd+|F6?2tyWL)Rx222My|lRIOw0=Msw5OXr%U3!Yjp75{PFy=J=^d? z(f8b2;y10PQ)tx3dE~}|iTtj^N!U60JzaiK4(&a`?dwb;c%8Ky@bcZV{JG@`c)xB0 zdp|UtY79GLLl0T}>WezrTdoHwF$v7c)&Z^7odc>B9<}7KDI3)}MX=E|2uaF8xuw(D zi7T4$W9N0=!sBBtSEaq7x5V{p9V$(0H!Qpi5~0V*y%ULuZtJ4|-YN) zxUc(hU+28vuh;W=ct#QPZ0KdPCj`(n(QSNep>T#01wc#tZoF~pU)r~89)2}B3j0~@ zW9PQLVa=B(v&CUY=$%a}bnF`w=J5AFLS;v@d*$QkioI`nteu4aDyf3@(0u-6(L#D| z&2VgSzJWGnr%<=WfJ#5o*)Qex8|Leb@UfIKOeC&Zoz~^Vu)> z*!4u%2}@JekTPPWC&JmwTcBjQGI)MUgH4(yWO>PH7N;;74eK2*Qav-)YU}H}yw|xl zk-lrW=t0tWQC>ww<(_d|)r3{aG=~TwQAGOa`K&!qf&|SqVz}_|)5?V(=*5GUM zd5%5W#l(dod) zWPi$ZROC>HX0JVu_dj>XU!OZOC43CERYs%BPtQVQaUzyF+lKyYJxf=sO~ei{%jwi7 zVeHJ?DC9bE4k!EQ06$vxII+z;Or%dLQ%8qY_@-n#PM^O47fjEi7MrH>3uX(mi-!{- z({l*c1!~}DN=;xHGZUhVUP8*Ur}QZ5q@meU7^m8Y>V&hScJLMywZ#CxZT7=kp2lFa zaGtFY+sjt&?P0IuN^|ddEBX zt80qL`#liSugtctKEXC#%Au#Heiu2WEg@C4Yq(=xU6!x2_wY{V-x2L?+liu`50d;- z05>K|ad{6aiO+%M@HXfjPVJb2d!T}du6-nbvc+jui6J8qpWw_IV|aD*Go-wGiHhFt zpujSrxh);I?aha z-GtIaduVx`0^{D=!-s4g9B_L)eqq@N=fzUFeII2})Zse#q$UCxc^RDM*iYx*%%Ljs zarD`O`)ETzBh_Ca$Ii_1!-jWrY2Azt`hMjR(0!zXzuq2!lVu}urrL1)=W+_R`gIbk zj4;HaLTx-cZyGz+ZN*duyP18iI;d35pzl47pwl}p2z%xT+%10s-=7!5#TQD$dkx;| z$=e{b3Qc5bU1M0y+(5klLmd>_?V=l=9zoldWzx6VO|(vMqeZ3k!Ra(}?31Bt-DG?e zFVGN&Q57KMsAS<($Q!OX(jG4@eFhhe@`awAJNuXs$_iqq!Y-R;&R-!De@n~5!Dp{w zx7r2pO3M?1rbsZ|rBUef?zep09l?WW{}gEqibExv%+4KN3<2sU*daHIwp@r}f3LSu zs`ON}u}Tf58J?gzUHx=o6`>qU0-rE%0Nlg(0@05 z8Qi_Ajc)Axj;farLTO=(mA;Uf{=IH9{u+LpG`d>RhaY!i&j$+l#|<9D3}oQytjQ?S zARG+}{)3esQZ#%*9N0z}K7is6T% z3QgTYAW{4pPE#9;uWFox8GkXpkfeZL$}PmjS*+I7pk- z6xf)M8k}IDX)W@VwEhwMn+!uKU|ZLXRBB35L0U3)S{;mJV=Qpf_jar`eH{MrQ4L2O zWc-vLci`}z^Ju)99)3*{@qs}zyyjg2+P>S1UOplLxz`;bb4eS>)INmn`5D;5#-H#d z8)?_wM>x{AfhdL467kCd@AQ~C$t-Ebfg*A1Xm45T@vU-Lv}Gw*v-%R)I{t!F4%)e`#dcoa3yd<@E~Z@}mDFtE7khh^_dgT|;hG;n$Xm#}#(Z58_k)~m1K z_X|t$otJ+2mT-3)h&l^7J)6j{F`?A#HIF@hIpFkeA8WC#n$~k7R4i-Pen-w4s`%f0 zUEC{`#tF}e;ajm(_Q)T&UM7L>d#uCGtykf=UnkZ6A@FAgJE?tz4z4;TM|O7y!>A7< z;H2zQyx#vRm>z4!3;x`KQ+v+h!Klahn1cz7kFJA#!=!|}UOikpWrUZ6H3B0LR1Y@jinia8@CgjyCOunaLppOec`s6YX$iap-D zx&r^(c@)dXAUt?;6dxNZD_o?R9O2Rnp}Shq6m@Ldn8O9gLFSP*@& zaX+c%9zf36D%jO+iI?(Yp)*$l4!ugmNqc=^vlWBCorO3r?KgNgM1hjMQg!Q^L(Dg3 zE$qj!IOB07elhSxI2YBBzd<{|g!jX4J2LQ06Ao*%OksJ>27I&FHTZf-mv=}ofLd2^ z<}I#Lou&VYi8R(?k5FIs&+sP`XyDbqqT-Z%Mo=R`aiBmx3tEvuMX;%esdx(_lA)iKP!4?sFQZsJ?BTX zp1@mICSWbT40Ch8LTI8w^|)3s{B~_23JExgJU3nBCdBjP&xIq*-{2oVDcTHTX2yZC z{u}P%U0voxE>d5<2}OnqJLY);GsvB~r4c(jKYF0OQfbwPKi>K12s z_~Rs;?3j(4AHQcE&sVca8lH4U?h&N-@-iG!=)>l2d+2cf33$ayP?sm;=~Vykv>`eP zc78a|iX058^->PASvto-w&xJhyGdz5<6*Y)R2}{LM<1(;LaEQrcvLSj6~1+Dpw*b#YVOvlV$tS?&!I2d&YXg4_d&M!Zp~p*I!`7wBbd@t@L5jdYbLCxH|Dy zFMZ*hNjp{?$96+%*uZCuwfa^G>&zKR*o;PK ztfsHN8&*&GnIvS;9#eZ7j_S*!=-|yya4GR9iTt#b^@(OMy+<|3f6EJ0`O<-YowJ5D z(QKCWdkQ0bAE4(2S40%Q2R}Rxc&w0<7M1&A9lvIX zoOyzLYsrT_-)rIEWjpq!F&6*%`~?>LSwlUnG^zFlbI$VN06Krkkv&iDV($IUtl4f9 z_u_gD4t?-|^d61l?V=7d{Y^XQ38%&Ad*m2A;kOK(xuKb~J-5Y5%Zp&6eKzs<`viUd zkU*-AHgcZ(_fZLpN8IpD>2&18323+DBnTT42PT_IHiSGP>)LbK65kef{kJ4OqnwCT z8y7IorykZbDu;0Ei8Hu7ZxQp%x5VoWQn0j5F#d5o6F+5M=-g3nmQje9(&>2iyVVRm zo<9P2Z8pI7jMcHzx6e4>U<2;i_>)%0RFlDhVZo^_!QJ$xl40YL-FFoW|?qk>&t&Uxve}+Qa zQTUT@DUMhwMisWX6XoF@L|rltxyL1t+ugUw{$yD?>AW6QVLoj1(g#pB@eP%JrO$57 zW3)PaJ@-5OEGj8`M(v8-=+QDSHZ!J%8bu4Nlxz{M(=o@J-E8op{8)N>rV{mV@1oJh z7ihhrKZ_Q2X6+#j(EIZk10TUXSi#fRFErU3wfmR`kH9)+)$mtS5fA?pkK=syBKqzY zNnG@I9k~?21Y)*PF%k#GwqN6(E_@b zBwCDuQu_qHX!9^^7keEqv@*1QS@0B%+IktydG3UBzedp80;{fjlNa^=sD*7t$7AB0 zjh*HvqUPllXzq`Pbj&Aj_%fpv9esBctF%NjAr5=EX zalY2d(~9uxJ(RQ6ZV_-Cuy}^6uJI#2}=(bXM)uO$vfvE-H-)vxYmz$ zj!|ZN&io)>7g5;vMw~M%j)QOPEZ&c`sh$37exb`9O3qE7%7X}});ZGdN1uu3wAIig z>xJX?T!7LQBdJ&R5wjoR zRPGfXayfxVhh$L6)^K#za6JC+b}6-Mb7Ye)pT}-Kv%%s7#!2NSEHqe|2D*5$(XugY z`Z-Hh-}8)m7mtBC9s=Y1_$s)6Z4PW4+lb{4O5rn2vY@kfCEfSSmhHZGiPr2AXB&hr ze%-HGsC#_}TCb717i_dXivcvd!R7QJ zp1k@1`rkTHTZk@Ad4HT2dDp`3|6HI;;xQrVM!2(o2F@M+5yYE@e%qzJYSX!qvBFIUHKCovB?=!>=|+!o!HsbWkMl7>Nct*4r%bC2Q%=?8CG)C0}G?H$?Ep zU$9T17Ya+d;V$E7Zi5AHUpEY8PgO*ZTrAM0tx4RiyTx4jqZ~YGd;ks~8V;l8dccfY zZG5*do%jSrpo>N`VZz@+QuLM3oZ^r4vG+X~Tb>Oix;8lWng^&GVo=VXM;w$Eq8Bgi z$P(X|^jqT^eBIR&WVQ611N_$N6v6BL+)W z(!cRR^x?yLdeZ0-dLei*SN%RsPPNMM@f{VovR{k0NVlZFM~1O`bsM0IcVe@5yd}?9 z?ME@CIWT8o8uH&{jzh&)(|vjo{Fb%?Aa~1Pq>c%m)@i_A&3VOdvaY0a??}?_Q`f+8 z%4NJ?za0M&U4nO399|$en7rF4-0sRJ4>y%TQOR)8ICB?TSzrRe)$O!vf3nrSy`x!r zWIR3R7J`-=jG!4&rp#um&`H~tf-+nbu-}@|bnW6N=t+wmu9tD6&lRTPF)d5bfcrc! zb@YIU-=#v&>YNOv2J!iQq^0i1A$5n*#78Ff?hrJ;EH zi!KnwH=v4(Q}B@bB2HD!9p$;nq1g`A{EOEC0y)bPY3#d1*o-@*b-{POyy_%U^g4=? z{uI!fyZ4~C)*r+6RvI<$l<-+a;r6-r@$)uUd{w=WRZEnRs^(-$0-q8+pZzqpZX_<) z6pjCtR?s#>FIo^^LHeyG(Uq0uRW&%D#{YK+*M?kye=>)enFuU`Y@P8@hx1tf`gK-f zHpEPtt|P7YcOY~y9Jh`*j*qmN;h67%41qEds3X7^Ty{$1AyNYq zPjzrNj_Wah&XDcg9FCWc_9pG~kI~<(4A4mv{QBBlJh4Otq+b}rjJ2LPLfs7i>|X9Y4uZ-S>OwOE z^go4~{uOv}ODtWNb{PjX>S6YxgdhK}j=y;NC~|EgoWs?};9v0`9*cHDlW7@BcDYLx zHvsLpF$<52D8#FB%kc>3BszSC3S8LMN|s08!*QMR*1n3vu#Kn^Pg*Di*MIbYfqooJ z_-c+j=EmY8uOhnP!bG;*GML^?se=T0nU z8ov>7zeUb2&?ggCE0e9XOGMLBMT?vQ(QRu#to^?8(bP zVlKv{Crbj)oS-FSvl-BU_^i2|&{8e;^juh9n!kU`Dy+XbE%mv2ne5n0fgywr^b7Mvo!l(HY zu#Z|a5DIjsi8xH!WI*Rgl<<*zri;4grIARklPsO1j~noR z@cm09B-)Ayf5a00ON^merK8x;7eRX_+)EAzTEJ&B7nE&~E{a~UgkJb5F!;a8(qXAF zRPKI06F83S^MPmlgUuPx9lQa)c-LFauII=*Uw2+7*2e0(+-hEJ84Z2&LH5)|@sJK5e;K(5>u(vjr2QeyvVGrQ+;l~WiOL7JZ%TRnR`zEZCL z8cER%XK{Zn(10CJ$tkA^EIuI$C4RGk8AID)F}+ZAJ0JwE^eAJ6C57yZk1iAc5oTp0 zV@EnVE^uoUg;(Nu3A$|iCO)sjhp$;>Nh-elK`)B}ty0V+&@jVfx@lq=%nLH)Z1aPN z)$*gPr*$Oe*A-FS)h6WUDLJTcECspC=`J;GV+PB~_;JP-L?m*Mm#O`@CkXwaX=R@im@4(RAf zgL2n+?z_Si@F<%D1BYf4{=hMAcxM-y%xkYknP9bZl2CJpLfg+=~ zGDLateR8+-sbxPRRl~yn2KV%SRF2YFt3jbf$qAah~|oMUZy)av>mE3tM( zW3B4B=HgO;^L>z&_=a*PHtB%`v6{OgRyi;g6zsDgJpC~H zI1GWx{0nGpMX2SesliY_OonDmTugHnI^f=iFC^*sVN{kdg~{7kG1V?Nw*8hh+jjmd z@*V#Qx>8z5cxfhmWb=?VAM|8D4vvG~Q$>)J?Fb)U9D`@gzF4xf6ViSi7u;?4;F4=C zD_(V%j+gYJ6C+dT>euN^KKLv<6FgeDvlF^A`xf4$(t@wG4o%;PL-lZ2)+6*K+vlKID{XMf#_0B(Chms08vhM` zLg*d9)@x*Vv4#7g@>-Da_>sst7^rIn5-A5A+aMp9kn5%`3uC&;TEhRGBD zkl)%GI5Xuc%p5)p$=41*V^TM$`Y`JCU=IBx^@P9R`G6eJUBZRRWrNMJUYPFi7Is!= zLyxK=yLo$v(0LqvdvhuImG+&MUl>=dea{E%r#cd)1aVliq!;{Z>d-smO1N-C9QYUG zpm}f+JyGevPLJ?r7H@r6_2_IG@Kf+6gh%u13w&AP8#(eOVLz7PtB7Wt4%|G|19nHv z*w*-04g)Ip&;xwtxXYnCw`;($Jbh5@(yzrbN@KHU}uTO?Kp{(rTbB9iv+3|6l0;` zHq0Ksr&;GEsHB<+p)ir2)4WQ0pL?@NpZ(hx2rn(M7oZS_)@Jzo6fGqG<2Z zX^dmMDFO*{^vn-2H9&(Iy83_IV7MttEq*YZ#c^O~udF&4!qf>zLf} zC+N?T;Sif21l=Mxc#Lp9a*GzI$QhF)XMVw3UtG9 zGd*i=NEKE}F$qyHQL0@^VMio$exJ&0^AniIzi2GVnFw~L&(b0x&%Nb{0}Q50@+qd_ zEWQ0J5IuJskPt+sYK4=y&VKr+GzQ9Tn5b1?asAob2RTP5JO(qwiGQ>zJ8jKI&XR$W zSts$_X|>_f1hkC55V*LfA&%80bji>8bcWeJ_9mx}@z%eY(+Dwo zsZfHM_LS0;E}i?31b- z>W%8FdTeh`JvHaS%ZDZ0EA@p)M(HN>FT6`%+MkEs5u;$T?kW0k`xuD(K1kk9>ZbO& zC!m2YhF9b#b@{oE(nJNk`l>zqbTg>|X9=B*ax-swaONHQ=5PXi+-}BpqzJnw#h2vO`?YlPcPmg) z3!+=T=2F`)qu9z?Rj?Cw1fRD*Ldo}ZNo7P9zrA*<$jMJ0{*%$gP3;lf6Ppiog8x@? zMXZ<3)NMeohCiTkni|yN!eTy3>H#%Md`Ol1N@;#+E}XF1Os%K6F_l+oXnX$x{3a%! zO1;y;y-O;o*V%ufr@p4lD!YvZVo8 zU0YT}--KIDbr1)8%9)5a`u>h=8${rcL8g{&^% zR4Z?izQ0XqUQ!ZM=gpbkgl6jaHIi*>xJc_v)Uc@9nJ#&7fbLnUNWTTUBJJy!xT*UN z^Pb!4xWjMEXqvkpEsMjjYW)fLmXuHAlm8K~-RtR_kFgM*rA!iCtZC9c5Bj-8mhAkg zOS_~i`OVIm?2nc^E3f%Ucf^ij=*DThNum*}HcVl+BF2DOaxz}}WrAp(Wi*k*1a@BFmE|7aM@2KvkVo`0|DRzC3byZtkg_xM?wK~K{NE6KH;Tg_ zFE!F#)t_mwL@Yb>Snveg&L$Cavd|@E8ItB2L#+c;vE}DkfJdq0jpKUJBe8a}ai;_{ z|1`(?I}kmlv7GIxb)r%3{giY2h<4g*xwrX8sf@IHQ3@e zUij9)c5I%tf)31AqpraMgFa1x9&la@S|etnhL$7Hwqy_a{wf^SP(9Q-YKTryPkg@BwpBsl`$4|Oo>0gHXbk<|lD{i%EI^OUA3p-#0tFNqQT@ z@q zb(CI}cuv1b#KS|Qh2Y^JP7+sst(wz68Q&W$hRq5&==-%KC`+=#SI5r96(6=)=cs2{ zKQEhWy)mty>g{Q#M-rOoz`Jw*pYO@}P>;V_4sgrH*K?N1KRFe<3+Vl+Dts+#HrCOS zwf>cJ6}>7^#lg+4R8PMHDHTg#MUM}dm;rKq)QSI;)M7et9UDpdREv(K zGS~X!aB#0Clned9I^qWVCKSM;EKhiJ<^ftAyb=n=c~b+Q5Sp|=4j)aO0}b`h;hf(x zx_Co0H8^eq<2!~(V0<~7u4D$fD=siio0W8r;B3|Wxq~*`^Z}#&@wiN-gB#3W14c^y z{QDicqWfX#XzPzC*c3Pp_B$lwuJIc|X4q=h=&HiXsuk(tzIv=#E@AEU{XPD;-kI3H zw}aqW*>u*DTBMsJ&U^-sQ}6%VY3^wUJImY9J6VR_Kjw$CXUEfJD4Ls;KZtIYDq}74 zDO9sz0ZfZl0_&~}$T}EHm$ggtl@V`XbXzQbf35)LS4F~snZz_vrhguT@Flm@Q4SqL+3-6bBQ z{_+p-X?j_!h;s>5vAT4&fFx@kA|Ee}1hF0=_jjz0d79=x#E6kj(;L4sk_?F`Wbl2+zxvpMK#smpn+a=4H(vDk9L;Nb(hh)IC zJJIm@TsAroY>BSJ5CA7rrfEEkoavD^~j|~n~|HWb`oOXyd&)KQ@UA!exgu^DyYW}?aM`@ySrE;^(iKt1e6VB%YgcHVzT7aq^2 z*G)_Dt|ftR@$*EX)3pIF*_26S1=f97MKgOkH-#N|Jcqu^4XC_xb)=Q|!g44sLs)9$ zcRbDPJ6(S<9n96F*u$S{aG9S0viw9#i|3i-$$4YqXt@|EKO0XvR*k27)Yc-8>(MYq zRSMNk?-p%HDIsIM#JB}2961`$OBLn}5!Kl>RBL(~`tQJWnwyqQq!#?+SKK^AomID6 zjjo=BW44~cKbg0+WsDRYi804|@y>Xj1>t}9?Iar(PvWa{t#HP`YFs<}85dw?1Ah;; z!8U=L@hIvp_hx}L)zdhIyRQF3Y8M$C?b-FO2yf{bV}1xSh`^$mWB8np@4%{lN$BGpqj5d%M~@&$HS&YX<~Oo5jkW1QJw; z_+OesK*b-BW|Q;SWyXD$@MjZzwT~dut1|idmtXUx6J8@KxXd}-1maYj!F`q&*rhk* z;lZPLvS4y7?bFny-|ECzYS>gEv*FE)*{^4=w>C1R%cJSp8iQ(sPxaNKj?AmhYzRd& zmzbmP#hLv262Xjpek5BpY8t!ut`vQUxlg`i{^InioS8pQ*tu)Z=yFv>vf+CL|1F@B zJ2=@9FN2SC*1Q8adV*Q?=m}%1{jWM!uTfQlNy!_~1UZi6H~HeZC;Eu3!0O7~bdIQK z%Frur=ZVX=LHcXB6;n7VO*ai4qiYX@QRU`)$o6hHN>z&m=^rw1xXPCN^Y0u0`ftHXsZ56xOgswe<~wS zkNP*ziimSeMOG7LT`r@A3GZp;3T5{B{Ce=38^VHY|8lDnLg}8ny>y|~Z)l6iz&B!T znM~3OF8RR-PC;@mXS~jWgs_)ZGt_<&jooHsXWbdn?7fH3f3Z+hHw!Kusi5AY3|Zek zDR$e}gTMFmpXgKQEYa_jx2*C-0lTc9MP04iVG=z}pDk8n$>;0Y{2e2zH-sv&6zw8X zw5%8uHDXY^mkQ)a9;auI9l(`_yO89?bI7nU6n*>9%AY81=1%;bMEob*VOB{z4eh&# zzM5CjRfX|ZmFntrrGh`*BGEu17pM^DOfw?AQVb_=l_E`s31sm36jEhf!ae!c2W>Hz z(0u7auy&itzQ3G+*YU3ILpvs)>s_Fc^-Ed`x1(lk96P@(5=RIWHl zjvdg4JuV}dwW|_d{?-)FUz3O;TIBh&{qqFhT0Fg{BwL+-WPsJE&ZDh%FVI>?BNnmB zjt$wHp>ZKwvFfH>RvA%Ywxq?3f3Wd-Gc%8s696+X_@4%)Z0^;+1 z$Ol_DBy&cDlkO!!@MR_ZC0UcrFl|ONMP;BjQOMD+*I+XyTXCNT?y!<;(N##-kR5fL zj0WCxkh)ACylq7VJy;SBXE$!AIp#{}NAz*{9G8kVjx55xmM>u-U@AO&;{e91Wth6e z1n$2B1`u-m2o&uPCZ>C1L=R6(kUnpFI#|6FKN~Cp^X(SAv7g{|Ew<iuIU8wa zhZnr7KrfU-nO}GuOv~~h-pO6ujyOtcg?A@!vW(his?xAcuVKg0JS07HJ!l`27L_KI zqDv9YVgjDNkKrY zB~*4vK=rvqlt!;v{ra$+|MWQr(hSsib8{OAuv)`nev4IaUoMAxTW=!SMlJNfAsICV zZU*N_9cJb-mnO?Aw>n^v z`wV8}E&?ucD(hZ*gY@od;XV!KlB3OIL~AsH(d^g1Q17p&)HFhzd2G2vhn-YK-f_(k zWK~Sti=}bh8hJX!mgC#cZ6|#`$H?w|_Ed7aDpw*ukyS6UXS+=`*r~xOu;xxA_;zXv z`S|hp_H-M%Yw=<_Y{vi`wl|}x;>S6s9cv(H*=qLt!&4?b%pa0{N0Nyu~OY9n9%cxS}&IXL3+#4Rb}k?sdFG5E;7c@)b&$l(_jAxt?f{W3zQ$R4n4@%}NUNV-rM}VDEcC`W&{9|k`!`9G zh>M}D{E$@j%l?Z@_Gu`s&veEQ=j_9Kr!=x@$Aj6M&l9SDAJ-G^!3wnR^Hi>}Mi!n$ zS;AHoO>DnP3$_&dg3q>l#5f?G`K~ggrn8ozxU*_7+^z%#hw#`~zn-YgIYiB#e&J6C z6rr;N5mpb3ruz9_oW!uvI7f0fdY~UpO_H}#^=N5a z^{x*!JaE8`p1bMoz8rebMMPUd=F!oqJISDbHoALu4Zp$9g15S3gwE)yp^~Bf{Oq0) z@Js7IlK=7xs_}Y5L$w3hR@)Y~v9XS%-I&fi!f(Qbwq;;lbDtb)d|8#Dn+i&E)acUe zMJy=L0(T!%5!kT0sO#1_NLnzPre$YRWDD3Xo;b;G3wygoj#WIECgkYtSXONm zx##1H>?~b5Ise(#Q#+^OA&*(qVeTisPCMMn{nKJKBqI;hRm||GJ=R!e?Fe4R zZw!7T7ll8d4`_Kt0CqWFLWUId@v+O1e8#vjSU!^{iH~#1{%u!a@h2XShiuB^#O*He45*!H^JfsU!=oRx6uJ7`HsQunGIypE1oTqGr z=us88BYqwF7PywK&pbji%svvY*~aXNlP)`GOL4}&9{z604J&VFHLQK~ww0aJ3)*;1 z=q}|vw93HpsBrB9LSnw4jlWjoMa%l(W70|39$Q9f^gP(Rb_=wJ4WgR;E;ujoA+8&` zfEPrZh6g|O=-Xy9eszvG_I%aG2RW|eUoBO z6QYAoJWX^FXKcS@_1;}=EZMn-48#ZH*E3zLf7Jk8wK@%^)Q9q!|6RfBBU13!WqGi$ z>MUMy?z=Gm^+UlY)X1YY8N4vt2d{Tuj*Yy{@rsx!)XLNwN6OgXGm_4Drd}A^(88md zz&_9ucpgKO5~-yBS_*%q@J=ls9Pva82Ym76-TY4jm#08%hgZ_$m8Ps;fl`Z;`nW>N z4KEXOLUmjhNI%H|Pw`^Zv+)m^`0N!LP|0ppuQVAMPpMe(@ULbLMJ65dD;AiNZ z$7{wcXLsM%5=Z&NRH{RgX-KE?K^pI{NpLwBUie7rb=&BUq$9BAt0y>Jnah&cpv|BhMmnfXXZC>9vfE-erW<>yyAVL>+leO-C<( z1ku61Ua&r_dvpUYyWX7_0+jCIB#o1JH_6JbBeF@whA2G5097CrVV)5irX;fpeo4TtLB3T=2jF(cM*%f4*XEY5$MrL zq?h;fvEZYT_`rUD7(eqfSMV{0%$(K4ulza%nv9abefT82x$py8>-G|=UMWN9jvmzi zWCmGZn8;1qDH3^%8X<~Vlxk%Uru>5DQ7HZNZ7b>OJkjaa)AUxg3a4|#1YGt{Md1#1 zxGX3Omzf5kZMLV--9^pd^{kg94b10UR*WDzwj~sY#Nnsas(8DuFM9lOILxl_1C7^~ z5ES;8DzsmROK%dO)l~3&=iGqum9A|0hfC;rMHRALPz)_cpQ4U~SD1@t0NXL-TK%uY zoXXWT(CUaI{P6S1M51OT(j zh}2q~$QA2Ck$!y@bv>EKR>yn6qq=<9w{s$RY3^e;r(LE=YMJz9OaWUcQ_n`Z<)Yzk z31IrrhLpwxBGtkXY|pVs&R;JNMk+@U?X7x5-uD}s@jeAz@BYf=x{AYAsaNDlPa?%< zZ_^E{mD!(Vnv{K5NsB}qi2Ry&r1jMR7-#5U=dAOT-4uFg0;9auZ6Y?$5hp%ROS!`e zabWHl#4bkxIoF;I)d4BQqo#zdNF9$gO?3ib5aFheIP(NUrGuvRUBmo^qQ< z`$AuV$BZ>})Uz4v^}#p%{*(rK3oqbq*AzouLpS}dzJo2RnnKeCMReuwA{KjkCku5` z!ZH&#%R8j0sUJ26R^baW$@r5*7%u5C1aU1L#vd?3 z^?S!+uZv?@gZu(!vcexG-bkaPRaX#xb}rtv@FupM)dQ_g3+dO8%}9G|KR;|$ACngt zG;W) z`8nhymHnwsWF*I7>8y>Q{zi)R2TozF*LIMy|1Q9?hPzlRm$qklt%xn?Cg-2MwH{`8^Clb=zekyapf zSDJpc`$vk^v(Y^|n*O+Hffnd>peb)c*`4>DOk&X&>QyFDtt)mPshq1r+0)fHvwtFT zXZ#~FNyxe?$=Z@F|ViSJl&@QujsS0HWJo^ zzQGj`bPl2U%~@5S6?R#MNN0n)?PROrjoRp4{5f!Pz$o;_JbXy(3;rkaPk0?jDC;vHe~8u`a3eA=8u{1@gZx*;o2Y8L1cb!6kS+fN|KGUvOtyXu&bOU_8PmgC z%ukSi;u2Uca}bvwABPnTZo(*m6YVY&i#?WD0GDwTuvcN7y4FSELfm(b^p-n4##iNpG=E@t|=SjKI;wjgkvFI=n^ zGU_e8%;Lhw%oRCj#y;XIe$D(!L-S>sk0-Ji^Pvu=wl$lP6LH912doj49#@8`1dhq8ig1yNVgBGz}C6=W?FGVeDf!E5X~O1vq{|34OR{EN9()oH{4* z*dgwvKp>yani|eyRc^@9S$)6wY5kYjoGq2?o|caS;YD|*Idv7A^OFkhUd_Sd%U9zQ zi%o3pXOVx*NQ1Gsk<85WmJrPR@Q+^Z8DN+ZUx$U&dUQ)-BhA&~G0#*Y;27{YN0A3) zKz1{?K5qh7vMQ8$m0+R zmNhfeZ6cZPk8POgkHuMSTUkNri2?SBz5>%WYaVm$MGd|()nl$p-k=+G8>qvCX$(~b zM)mq+CM54WyEa3CoBK!P%v+YuJQIv%jpCq$H&y~qG z+eK@(-K$l#zA50nc5sL|JjS6V-JVez{6o2RS?1?0X{Kp=1d@4>n$)zaM8YPg?RRyXqus5L3exC%mJw&Nd+DB!(@m+46ge^hA;tb4#y_`8spF!)GsDhr91o zFm6sJ{j;7j1xs$w`}}bFx8WFFm2-_9IvGqiZvMhwpvf|CKVM*UKZw)maXtLtFAHd7 zxetAGYFIGx(f}9sRKb^#tJKLLih2Flk+BkGrOyUmLtozr);VuI^Knpuai4RMO}^_U zIKeYzBZpVgk&erF_mv-=S<^!m?Itt+x7}z1w9q+i=FF%G8<=3t%}fT#rOHP|eeFe# zj8U8h)d+NEjwH`uw$F*6(m0w?%qpRt_8lngC(3|Y=F_pcEJPPp@*Ouf^Q#wsqS~pV zXRg{MX1`MidY%m8SBd_2)5Qjvi+>F7%$huw4;-1E7Dp7Id&uUd!5!iEM+*ztlA$=&pj*Gf9e zTZieWD4~kk`*H4eCvr^HUEoj`%W84u?D~I>kl5Bh%}&SCob@fVJYFA7GByhawvR%O zX&&fv|12(EmB~!E`$P3Ql2Nv8HM^H;(kVI{m|ug_sO|WRm!tpN34;nO-Iy?zscO$< z483$v;Ga(AT^BRSb;|U6peb|8Ro&r!<`}-_oR9Erp*S~F>8^ntZ{5AJ>CDFPB4@L=Y{)NiB28C`sM`! zkIS{pfaPCivZ<6qXN!VhU_}i-Bp{4R&ur$;ooUTxMV0U??hOdi4}Pu*vdCn2O0Hw{ z8w;7gx6_${*?Rm6jh)Qf@}o@B4Qcu__AgG{^&eJ#wr7Jk^W~;E0Y3$N)I6mLfL4BsC z!(tT~?#{3O1Y5-B2-b9a(U7Af0`J?iz|U2kQAk-up<0(|H}s`pN;2$tQ(K(&b~5!_ zeH9G{SF)E5%x5|Rn;Ay-H8XHO56ee=r0>nP@{WB1BHDr^g3U(N-lX05VAz!oV7 z`MOK=>Ml2CJ)dFZ-f+yu=SLaqBbykxj?;8O^jJF5pi}d={93JiSriP2U#F&rUqasf z8Fc>6P0RtlAB`)Hr>|^{c`LG?;@gcW%$~1pjAz|K!RrTz2aQ~@N7I3BG`hjMo!ZNe zdgOqKB9{IBzQdIMlj4?;a>4z-XJ{41FfF(21^&irT$|21`X=cCzu)=;m-6%- zq*p3CxXhHmx8ffJo=u}L4Tt%+!=^HyU%NS6u54g>>>BApjuniJSd4365_4Qvld)Mm zfwB21&Ydt1XXkqN@&6@db8DLo9MUMdkzhw@vPsXdv`FK zF5;=4(J&WI&Clm|;6(epv*T($FYKr7_Ih;My~(WkJ4<>YU^SHrzD3Wzlok1*uG7X{ zHujUZ&!QDWP8JrXq?5htufjyj)BJ|l4y@b0OSG!{fZ&|ouwZQ0 zK5RMpfkysF!}gdML9aswJ$GY-8r*7QHZm6liW<_)qPxFoerp<=bb2iNUaFpD+{2Jt zc}*bkI2+5Ox&#V4(}=A3UH03R3}#$!3iEDeIL&pu$4&dJ#=Y)*!EW1LAuzl&3O}n= zGttx49X{EQW^_yp1P^2znTQ={U`y;qc>gtxo%2MW`|0t8(w>KL@@<@OiY;A0DCs1 zPae+f(4adr0XnNJ1&>14<2Z%ujO4gjE_2Z=!Rf#ixK{QZ>zOF&a3(jD$@$w=o24Yr z*R)i%-!{pT8CR^p3|v^uxCRPn{?~i-l1Ty=`r`=e>!ZT$nQ@1%>e0aQa?|+RPB`H2 z)QPlH?JMh~b)UZ{?kVENGQq-Z9%HH(#(zqyQB2w$b2BG9ToL?Y_{&6l{~}Mi#hbFF z7Oq^KqdI*%J%AS5X)bdb35Uy0qQ%Y;LG%ZA>hkY6 zE2iE-rR|ES(t15E=iherS%WrDqd>%jzN7(`xvy(qU7t@kT^WnoCoP%a#s1VgatRh4 zeu<4G!PI=y9;Qcx#OiJ=r>p)Q>A>9tL7OR$D;CW$hyP~NtjHNO;<_o* zwIo&4$x)<_Zx?g>pqmVbo#NK74QJ9leUVAp&ZyUl?)BDZ=#FKpal25Dj*2Q}T&6TL znnlwnClm+o0S%n8;iO>NoWxoS?^{%npo1zS;>&4P(2kGEG=3eATW#77U~0@BkmgwF zhkU`$>GPQ>PcJip$yex(!%Mn8Rh11C;eQiuEqWICMq zqmrX*&xI_rE3r_-ui|TX6Z^`k=lnZ@;QylN94~RkESZm!Pu{HE@m)tCshh!xPo2cQ zad6=JJ}jbJ91_@UxtUn$OSsMAAJ~x2iQK(sW*BjDE>n1`9sEa`v0u)|a4z9m0`;R3 z!Z4#EPMbZ=h1Qrb;+vk*y3qIB(23>DX|ZAaV7-i!`7p%$I?)M>`No0(TYJvo?j7uC zxhk0HIY}7usF^$Sd?9^z+mk&(3k7X^}?34L4C1&7xyysnk|E`%XDbt zA4!nd_=>f@#N*C6%L$idrD5%WC3jh8fl$?0O}KdY9oPEes$kS5J!W4-4=pXe$~sK9 zW71~-pkFWLvgdVfa|Q<`*|_*eB6sEm*6ctmx9YMg=2a#WRB6EUo0=a{H~Q zhXuk-(;ma0s4V>5Q$$;?1PlL8R1?1Bd2{-vK^(7h302Ms6kI$riGB>ejMin-aIR>l z{P^M$Zkukm;Qo?ix+d-h{nc`nEzX+7OgiDfWUl{0t)XcQ>r~<9by{4| zfAg5$>g$4$(4U~9yO&!lvx=?%*~2znILsbS^WkQ^i>1}(&$wghxxC40(P)_KOs7;_ z1l8MFbi9EiqrN%<|H>`p67!U~rbT)9T*PTfcy$wl9W<@70 z5)j$x&xHS&@5*3n^iP%4_I+wMK#Xu5NmnIIvme(`ZVP8oCVYavRi|=q`h$ z%%V_hy4li&RsMX3Rqa)v^SV#4x8Gl<>Y~q{L)l~2y<{I|L@iT8hu$xHAPVl zv6%a;-2OUN-vn8)lLh?b)HBTfK!$Vw?I^6A^;fXB;5xg3yTXO-T}(Z7m!QP05!P2Q z#DDx%3x|J7GddcDAZ;*L5S2KEJ%JCjjf?abiu@4JRbA@9RIPdPi z>?Z?}YgO7n_%{@Uom>C04I9L{sxLWQPUa16i{BNlYULBQ-;GBUUVPxZ4072=lWx%S zy*q^8o+Xlgem48t_#2z9Dd7AdS}<`zr$C3;V9(AReDh!|mltlxW-GC*ZCV1=QYhwp z1p%DQ!`Z^&CsN#QXCp3H<0d=0`C!*}j zN7nnb1huRGj3uk&1iTgx{-OlYK~HI%u=fEGPHA4qjh9qq3*UbdoK;V!vfg!^zFVr3Gj9DO(_}y;pE~)^&DueX1bX&RcM(`Yo-J+rWvrCUKL-I0&s@ zQZ8_xH5)MZ0sBWii}MN61-U`uw=goy+YJY!$Jk z-44CtCRZ@r<79uXP~D9y`Psz@Z54(6gJQyt=}*}Ye>m>mDhuHRb~ElimdH+AEh{{h zd6l|v-ziX?QA#8JvlP~^?&Vx>sR`K=cHF-29|fwnF0*`<1>j*W&4tz3GFOU|g%6+Y z5yW4S_^%y7Tm)->xAFUCvYoQuAs~7op8(HCEUxQDcrm7^Ejz&OFCR0&i+Zu zqq`@JMvHro1uYAC?4$C;tmzj7x8E z^okevY48`fLAi#@xm&|pjjp2Qk0vobyP5^Hl0~$1pE>(V+?h+%a^{5mG$EVhB}`s& zR`{)NH5d4PH7Bu5$bGGxFP!FN&bllq;HFs~6%Gca3r8zd2sMM$IiFqG+&hUoT(v|m zJO1ku!BRsh%9D*@pBByGQqSJt>I&y@7N+JnW&ALWcz&2YroWr(-KQ#i-jgBRezH;6 zbJ$OqYQLG=y66R4tnAMIn$XBLe>W5EKVT<(+7K;VvHXb8c#5BpSF6r?UDjqZ53Xjf zCl3n!#+e95g)Zb)EGgpruSyB`oN?i7b=PoJ&QijdmOa90Pf9qi!bbL6*Hf-BU0#H? zNaZH0<=3V>$`$egQ0RW}sc_Qnx9ql-lN|H8fLnIQg4Oz%OV`ecLy4y*?2OhTft`c9 zpuuR|<(w`pQC46SCv>*t+IeYY)V&8dujq(i?)SrdX%$U=p2APqXsu0`%4Z;z7N?5i zG?}@_SBSFP&)Ex8y}1MHytplg3)q*E>)~~iJ*$2}l#8l=YCtudEZycCTmG{+Gx;@@iniYF-LvzqVo3g-tkj+IV_y zey*VClo(sX)H6dSp-k_^V+`%M!+xA_k_}p{!%l5{&X)Vga0~sUI0Ky?dNwM7`Jy(L z`j=#4nZgFP*)CoXX>^z_oI9BnH2!4I?<%RCFUsA!{3piVe67QY72afzI4OJaUH@06F^$wSF2k98@`LqE+ z**q!s;-}N>yPMUlSCT3FQTibFSM3DnPJG#)FK-DPMESbDPLV5P^pN1oMrTg9;1jFK z6tXvtDsw(r%c%IbR@SSCXLfvQN6s~bJ*BXc zomKptyqDT2upb#>PjqFn(-+&(YExJC*!cL;y1hbtdKQc zca*Ogc?O)62vswxWiqy}WqvL*lP^xw04n`J=g;MFFBtr3EYa{et~>#fWq1dcywl z%Vdp5Zt_>u9~5kDdc|gXjpfY#tLGjZ48UhQ7E*!iLo{+~X7}Vbu}aHb1><~pf>wVu z+Tf-kh}kRUkU!=)BY82NNo=3ON}EmM%;OGo=hPDEbx(J0Y-$8&oG^(Ue@|WT+0&Ly z`w$A|`zZ5!(k1qS#{v2(u!7&87z<1M+&PuOa4u5F537tHQW3HH|`3sS+{1#8XL5mI+e$>wdL8`Pql?P zGpC9z|Np-v?uw(A?qAr{J`?g{?8jT{TFV#om6MJyo zL}&a}Esd*trO`Td0V++=#8uPwK`cL!)FgZa?IrPW;A1oKX1_sn`*#RWT@Jf~V|dBG zPm?K8GoiYqoy@yx4lz&1!r4Gcuzz(N2G>PFZn6-pT8y!$OAAxW{2?>O8Fu-d0*%f* z_%Ehcl#loX-qG1)*S>hTweKewm;5947g8X7j{@E*n?@#Ac#&I!SBR{57+7a7g*z4U z5b6*GZ)Z<~0~Hb&dO4ElYFQIAotb39^QT1pn=GDDUXO{=$H@Z^IgoEEg0jh9Vf~dU zF!Qq-jBcApmjCM}cLRpVvS*Ft%C=Ha-RFwo@eWw6Cx@>LjnGt}hcc1|cv>nFm+QZV zU6XHv@JAuI7b6kHz;PTDu%rNoidB z@frjN8RDz1NYp-@1HGCfKsuM9jcz=A({{$3+JjhrC!gH;DnSw=RKcP}9{;nA$Fljw zIJ~F>rmVdW+0ycu_97bG@~S~b_9m>p5d*B4A1HQ9fPR}3iLW1l_~|dOvg{y98hS>w zUDv|X;6ix&YYPN?kcZvc5^!uVjYxkQAamkNcxL|vgLKm{Tt3T(Cb#>rM==hbx<-JG z76DbqTd?fD4>`Uloj6?E07}K;xGKjGAIN%uaori1VYmhEiij%v!)IaGHc#BR@Gbl` zk;E1Heb790JT{Jvg_MVv;Nf^}e7vy~d=4B1^|n$7VtDxb!XnH%rim?j$&fHj6V_}h zCGJ|~L`>5URQ6SZpOzv{^t=jhJp$o}&skVs%z^)!9*`V=h3qt50k<>Vp*3uTbR>Tw z0-Yez68jAzCoRR2mNmG=J(ujAu@rv#WmE5 zm1Gr&Ey}{W>AtYFFA(;uYapFg6GY4reLU>ph^D9G(COR}{BG3=q1O<+^L~*3x){=8 zWq>0Fv#|1{9+ppD4Nl)x;c4u15FVHess?Vj)-4Fj8r(@}?pa>H`cGcF$6Cm3xZj2ndZA1APGw;~z5 z7(jd!pO6(~2u@qBzyn+;M(p1R@(89^4#<$2;^(~P(dlI2jR<0! zB1t@xy&yQ{0#wd>3G_iY+3-&nt~W1-ERkz`yV4eXGf9+%J?xC@T4ixVfDx|G>IU<2 zXH4c|u-e@ZAH}@^^UFV=Vf0G8+pdF&)$%wJ;)4q=_@eZL@u*)Ii4QKDh@5d&c>R?n zo#EAkUs^AUyfB+Vd(t(M-;)l1neBM`;4FOFs)PwsVo>6_2MV(_M8884epj4^T~V%J zINB78I{t$z{-)@p&64Wh`$^y?DL5DF3cJ!zk>!iRNo%YICayFa;M{E(A@v z`=sL2XyWRuh2wL|;i+*3oDa0d&P>EI@M6 z35V~CQH7 zW2R2TG3PCCr1v9aEh)gp$^x`)>cBNN^JzQx5fx43=(?D*7&dtm`p?M2dcR2U+;0GC zTc478o2!_U*o9jRlCf?`0?$~Dz~&9_-#fr}pY+sYt(xdUsyR6wq)G2`L* zieA4~M%lwL@VYPs;s@7&kCOpzm$?b@zea!&A0W%Co5}U4lW^d{Z49_lDRLJjW5~q< z$lf0f_f-ZVpiYY>YlPt7%L6#BMVFd8Yf+i#uc+d^45j!J(N$RnR(M~)Nb6bn`Q;#Z zyh$N0yDW&WUJQguKZm4@9B{1FhN~x!T8U_+xgOse zMc_q$DJ+$E3$C#|i2hUyo${HWh7vgQ-CfW&)5QYScTgi!1JB~Df%pC~v5>tB=TlnX zO!^`Skx5P>#$U(nkGABMkbU4CZ>)C>)-GXDr;Xw&DnK@gewSk9#Hkg4^~E)m9`vStB-}N{5wSB zk~e+`8iNk`QdqS?2wAQ-;FYvIv^34a;0KLpT~dNZn@(essTP*RF2IL@5;*(CX$&1N za#>|wBD|6ea=+jUVWl3RQE(XAK8_J}@#aDI*wL_RWih$eCxoA$PC|I#3DEf{z~j5U zv39%%X2irGvn>!G$@=5`Z${w$PZcJgJr21C2?W`XgYmDn@E+I-c#VH&^YlGG@mfDj z0jHc^QZvCA4$K>mNe0s~Z;~|5%KHbAN@kd_#1t>TG^Mc{T&RzD3+^1f2{=`IjAJ8l zncij0_Irt6mJ2X0xSIr5T!NY}Y8dyU6;Wgwb+(Mps%4Qm&tF2^L$d&(-j#O9-TXgHjtX_&U|5ONPp@RIIS z^nH309BwVe;_xWccE5o)#;ah*RSDE`Rz>O8YB*L-4=+xb2*G%a`ra-9Up-k-)$saqk4(S*$J?&HJImWeU&(V7C_!OB`p7M z4vy_Vg@-%4aQ|#mT=2vU0<=GaHiqH;kv@DdN(MKZZNX~eEBO4O9TYs$g;Aa}A+YKT z(I4AD44V7FFF6EmPUrxoQ^m0V%5IoG`6ukTKLXDxfAQvz`3m16Pv946TkKNuzzPd( ztd%+sU(6fehVUu~4PJp}&meet{2^N5V{lZhJXYSfz_z)W@Ig`y?{*{Vuj(Ouk-N)g zj|M9A$@1m~DM5qbMLa)cfXG!R!-XyrpqjQQmK%?A?2qD*g8jjz za?p5s35E>q!~dop$6cE4s6Xoyr2kSyD7z1xQR&dvcO1o-Moe;f0umc?VMnnK?|Zu< zJmp59%uW(H-Sx^&EI_h!34xjBA0vBd^6IJpGNdHi<1Ll zqp;970*mF&p}lnt6m)wd{k#-?EY0>E;j%(!McxW3=&@08W zL!0pG9Ci9~(3`$+SWX9fR8Z^EY;20m!pY7oT#PosPPKfPJ3AKE*6Cnd@jOv~>NuL8 z8^CdO+H{p%8<}+|9`+V@LHENKP?CEPGnT)?;H_Ul^4@)b*j@Nvod-lbUxx<%7#` zyQ2&$&q^l?t|md5sMMdZz8U1^EX3vBN6^wg8BJIC8J>WEs#)q6DcF%H&V_A{C=m7=yq%+}N&j@tBIRlrj?}hK8 zbDwwmRp@aT2b;~hV5odH$Sd~|^VANW_#eu9H@u0+{auBV>;{S9CnLz(q=b)*L^{P| ze~5zg84_8m54GRyFfh3qUkJxQ|1VSU3rYbqDRayYOULo9EGihhC5nBY!LFehwoEL5 zcW&cx&FN%(aC9tWUTznC-iu&tb+mot3XZ&+unYfbp5V1<-y#1!tp#T5eGm_RM|N%4 z&1;IBhzDlxgDx>85*cY`e?aemotm*bd71o?98mIutNV=6z3?{j&Ywj+KYjeQat3eb z==Js|q`ug@HBW#sixBcED#JbeTsQ|;l}`bz+rCV1mzHE~c3gF{X^IN7rq z4!w?n2k#V#Pg)D_>$>0YK~I^=ExKsGbEMnO;7AhL^I{A}Sa#twm!)*q{q<<#>5X1y zVwk5{K-9EmK}e($F10!UHy00)^P{B6il^rEVQ2)N`Aoq%{4&X(tOWZm)M3yJ(b+Rw4nzH- zz}7JZ61UIB#0qm<7q%R)Y7-25D@XX*2E33X4n*4NFNjZd!}>Lcak5K231Yv)T*-Po z>^}iKvT7mywhqdqJbcdEQE~<{?t}=&;l6%Bucsf{& z3x=wSII`ieIxhY46WgsdNyPGJq@rpw%=`dO^3(vcAEXN z+?&Hp1(O_|Ge=yS+_0% zi`qTmHJ9eIzbz4D%@~(ff>e6vIUD-lgM`U_t5$# z4`yn6Lk8a*tq+%C>dL$5zcLjb=Go!#ncCm>*1fom~&_65>)%l`l$Cum0a8n$yx_T0GlvL5eCJj6r-0(-yObq#V z7M!l##Jc2m^n(-VI$sraMDNkUAt_X}R3JyZqls^`4TyB`@NB;#z5_S>FIwc*Q&XXG zvrTB{+=FQMGa0iZPeHNy4{&@Sj$a)LVBLx#^lW*L&sy_v)$bBeT5b>f7D$jwm+jCX zFA8r($zYnH7P;~HF!{a0m{(@tLq08EKr-C-lln)Op~!C{D%JJEjhsXjYUiW=$s-tX z-vf?+&4s7?v|!)kYsBb>8fYd7371@e;V+7D>FZ)_Ji|gU+X1rWia6@%OyYj(7?F)r zBG%Vuh`x`L;L6=|V9=-sbE~sp+u}b^t3QVPH8CZ!OSE}rE`MQSgde(Z2*mQ(P*hs1 zhTGvjytdzguV+p~TT!l2op+wN25%)h7MGJ{!Gq-Rw4cz#4@FCLRr>y|EPdGV7AKCL zPJiXjpqp-up#|qZpkj~`Jt64F#eUDwChiO-^~d9}wwvHowh(0BbP>l*8dy>B1P6}H zq8kgA;zC6U{507N!y+!AWxpBvY!icgqj*f7{sr&G58&(8Id}+`!eDwGiF6UoPo`gk zq+O<mQH_kKRAP~ zEY6^hv%k}^b*ZAWni%?um&1un1FUL&3**Cbpl{?336b=IoLVoqOLK9Nm=MK>glNz= z8f7FJ|+`TCMup17I9!1A^y~pQj zc^EKL9N$adhVK3MNN3`AxTL-dYCQjtO}h8+YjqMHT(KP!{#Ij)=~^6W@x#9@E1{PR zK-8?~AUi7v-#Kl^%C+-Fv$kunI>rQlXx@OU^As^_{~_4_eG|m7TR`sGMo@oVO5{$B z$C!32^jVq@wVFFb?j~at^H4%r^Ck>6^v16rY|!OxGTz8uiN#xwqy3e+D4y958juUu zS6zetti@1m;tz>|`8c<9JNmvA!<9Qv!#E{9Q0CtzO>NLB9z<<*iM3%{XSxsk2P z2Y3s4{k*F$&Oz+OZV>-D7x(Z#k{i==$inBF7g&^(u%Wppm(SCGn%C9 zr5P&us?d9Tu3_cib(lIM%f30X5|)*R;;*mmxUEW^hDP*ZM~VyX^t7NUK6h}vMHx`D z7F@9VGky(;$1u3Y3z$_!)C*-GnHvJvj0AaG4R>!J?SWRCMxo6@H(KM2X!CSsK zkQci~1xy_+aLEECSpMW6T&>Q>jsI!G%Is)Ze_eELC@=8h-L+2$Ix33 z?QpGzA?`S3f$q(TAou1HGFmI?q2WL>>^~NCb>E2YPbVN|zYXbf@FTx6vhiWC3wj5a z!n&i9Sni?)5s4M{GgZ}y!z?4%J6aiMd^?QN64$}{xgktk=4$sM&)2?WSDbtMihpF6QGnCep!f=zR@GQ}lEWe&ZR4v6&F5MaSk1_z0 zm-k`Di!Lzj&c&brU+kT(N#`B=4M|DiShtLKNg|f(v;?v+kLQ?+n%=c`E z?hku0DQOsY=ik9KOSa;*w;AMGu_w`T7YFTr6Zp{PO>BZ@V1=I=o%Q$#_yxQm#~cNu zGC@Q1`K=+V0z-Lg$6DB58T-?I-NjU%@Q5*to4$v<)mN}j%-4l6K^u9=)mQBz-&l}2 zn_74Q`isf+oFcMWNy?t5AV zNdFK|Z>Gn~4AH$NMhbUy z#(~RaS&^?O2zyUVLY|&3x?Hit(EUZ^tY|-z_8=8HngStNV;m&;JthruCy0Bp9Jrs? z1{1wO;{H({Oa;fFeGS0Bh$8rWSQ~;T%>u{M*T~3ZTln^P4s12|hlvdVu*>Ecd3tml z6pP)4)s<0@J?0w74Xa>c{Zb;!b0JP*XGr^T8MN&lgws}g!SFMozo3(@_yc?o{JD2-Mb ze~IxPCs59r4-OA^19QI<%H7tYS@CMa8(&uTB>iD(K;l05RBV zmpgfU*-g0;H(DnG-EBiX1k&PxB@x#Z3{O?>t0!KBy zgGA+CvO2;S?p%lkInN(tm!}F8HpjvKN->;gt&I1_xZ%X`R?**thrt=^aIc&X>3L%V zYxgF@k!zh~WM?9@>OFyvSJbh=DH&VzYp~;8BhhOd5B{nXAtA&DN?hB)`I`j3hJAQk z`vH3RS<|7^KHe;y6L8*W8y3Elg1okMzz1dct`QCWCLdrJ0OyUYN11*^f0qE{f4)eb zYuiH2w>mg?WeXZE8iEzK<|Fx74D&21q3K5t&!xeBf;*7Je~dAu#sb9fuy zsX$fe8zPgg2Q!`)Lr_&7T%2?a?j{(*ngg#y`(g>)ctafzEHy;^Dkm_IKZIi?LNI2g z89Kci!3vj$=o8b0E8KU&!m?%v5<3O&6-;rQ%{+Xw*9}{aekSYx{U8Tz9~1pY4J7@e zGfZr8g_PqpyfsHFq0T-PPl|Sgrcy^xy!8c+o}-Jl+Oc?k-f@(lbBfe1se*LXtynv| z5lUV@hMO~4%*!bziY2F{LgB&sH5PmaDPiH$Pk)65i5 zsrJE)5YbG)E*Zp~{2|}TgvedcB7e?2CwVo6r0H-bsZTmcsGw{{-RJuHCG?fuCJ^sa|;<=6*tc<*m21`8foYyoIdnxKN%8bReJV*`WeZD%hd^oRJc9e=;Bu=5bidDoZ|C*k{i#*( zYHBd9Al*q1Hz*qN}8dKi1oKpwh+hN z&PFdb3c_E0g@;nRq3PRHJl*yg!;kggW3vpHn3D&K7s*13j2vvXaDnmaaUfga10Bf{ zU~yCzu60~Mms15e&RB^mUy=k)T@T?pYD| zyM^O&_e|6p`N?Keq0 zBG^ik62fVZ$3|M!QGh$HT!)M$T14mi8eA!*F4_tA5o1>&L^^Lm%bKzHVDK{DQ|ZL^ zgj0A`*Nn;-0o|x&Mtvp&j_ddW4R!t^_ugJ)cbvfPpBHd@|=d8rA}9EC1>1JGKulZJK`G~IO? zRsRNH$2xbM`O=TlCyI1z`Fm{m^NO5H*8w~205YRil=rPsq!%`KVL`xB2+Z_^O>t6i zQZz57J`cb}`ZjS^Z-?APhj6>N4_Yfc#NUVC;k!|{(QeuXRKA*q)c-5=-k`8fN)ui= zP9PU0i^+z0*<`+$B_up@=Pm3zO7zl~k#mpDpvw3+c(LQKf8Hb%bbEmcS%ycWG_do% zEWF=%kk=J74^;b3Lgj)j@Iu=dP~#&n+DMGI$|D6|nWSUR$|MY$eFh|wZi4&WEC~4* zi`QjMP_2G8M23o5r#GZ=;9nHvU_MV?G^5o{A6FcQr&LA^@d%4Um#8u?H0?-Dj{`gIk8Gm1ml(afG->ccGjnO zMdJeO9ghqV=i~R`$&)JZygn9{^_=ie*FTsxuOCk4&j!^-T?o3&fV8s%Xx+R+I;(Eu zzPAmyWlJ8u&OHTI8}eX^(Q^`@Pyr~sMs%Coctg)F6UR&$Xp@)>LsfgohhJ_WJo}40 z+!_g;7M`GRV*%0DZwG#U3`S-fp*N@COp*cf{T~cabsx&s#+YzGW%0_BW68b zX|_b+PAO>CE#hteP-B1Z?G=*9{|*bLpT}puVrbrc1q@5N;mU;HVC-N7CKi8q13y>b zUG|o!XVZ%A=7w17bqZdnJm=Nes^FqM6A3I*lEuG|u3&Au8U3v;awL0RgV~)~F!#+0SgIa~Ut4cM zS8E(xEHHuc#c|}7TQBdexDkZs$3hzZfd>y`K&NN|Z0fm1j*ZhpyGNGzqR|8MZ#7~* z8;go~0M&ZNnZ6u{K59lM&!) z+llpi*Wx@;-)(V)GB}SP4eM(n;OtuysMgQ}F_D|%d7=W|EZ4yc>O4% zXgpwm^WX5pJBww^_?`PqZO$%AoM%rFB_+5-YAH6zM1b*V5ObDa87Gx#;eJ(D?v`pQ zZi`$}G$4_c9kSZ}v6xExU=g^iDA0j(m7dVk*kp{i1gls@T0dCYop1o<@E_ zaVnQ7!QbkxWNY@2vUYpWqEAVv=a`5kTxw!_Jne)L{M<=3M5OPXiYD16Oix#1YEemO~n<;V^0GBHDG}6UQ(e!561PG1YlG z1zDfMd4c1&Iy(<<936ulLbG3QOUFTQ_g4{1U{Ps8H_3W(ZP0jHS*^DDmF}T=>8kHb(K74--w9n6YfeZ(cjr z9i52D&hMco>I}R}slgWtzcBPpCB1&b2QohU@Q7&%-TUDMYwaiC+MCBPenmfXRYnv( z#Mi>&$R3PXXH4R{Q>YH9CYDSiQ~yt3 z__P<@m*ryZ8C4282KYM`fV--P^AAf1KhUgC;onO4Hspq zQ=N$xI~G;K9_|}pb{p=2?VrTRT3VL6+UC$Mzd_8^slX!cDyP72R}AQ}fZBhVsHQWI z^o{&6;;kSC#Y*G$2Z4-+T?O>2b54PRi=cbt2_F4uL=yL;@%}nKDlV@e!}}iKEbtyO zR#jr%5mAO7r82pyOTk1{nd-~>X?C_2d-IqWWwJ9V%SE0ZpB4toCFR&F>%b`spgL|eVVR7N>!GW6c>V$SJ&dMhyWVmZe}&!--D(p0@NaEOo!|; zaZ$V~8h?<&ZQ9Pr_bvvv%Mb;w+5?GKrO;Kl71pfI!48Rk_-p@XOsL`ZdP`?g!ME@9 zAw?A#B}Hx?r%rHK4ZgOo#+N>8=*pROz!ohAx1e*N99K%vA;f-&p3Yj!ZQx zd}x1)ina^TaV}Hsx7UF7=P(#`Oc=XHCZM422y@7u`%b2PqsvtZz(#1Ihwx)s>va&n z%#@(eAsf_rug`3JC5+#0xl`V}XmU)~rQ%DkVg8jVWN$cuRkDp{b&zxL&A85fR5{6t zg**aX6McXSX*~HCMzqDlmd@|*tUEeJYIoy!^iN>u{HF|N1wa}pQ7r$6r7vkh4<`DP;jpvhU?#k0jd2^ zwrLV;>nP18|59SxZRTVAz83s_tsVO;mV>%z3aC!kfigXgUwJJ+53Nga%AsnE*<&s%gu^?c0`7EB0cv+isjaO%tnM zFN5jnyWy#O7FEwIp$nBxlV~0NWX%8Y31ch z_&s7LtFzdF)#?yoJ-!^JY~@L$RVa&55sz@^!B+Gx6-Gbx28x}eN)tbi!7EJ-6izL{ z#|{_KCigNHx|O5z?`7Dwtqud9Jj3KKX&e*f%!Uu2X1@lsu+sHJ;yc1oam@)-xjCKm z5fD)H~sLO~e`T3D+}qS8CL`HdklK^k<@D~$ZlPocd@!)O&gg)Qmsra8HiH2SNE zx%2u8Z<5nQ=Jfpr^7*ih6b|=bxKt1GXr&1ZeCvSc&7xEjG|tV*z47I8E9Twga+sTb z5&KU0V^Q>T?0O_fu9abU0`}vIz9jr`U_S8LHiKTlImZ8h0DD5Ok=~d*%ub4*N zNvr6*_)YBOugA(hZ`@E73S*ub__BNj?)IC8t0PTGyvU7&A{=OQN-632g;IcVQ z53nh$2^P+9#0MGN=c?ESRTi&!|6SAM1y?bserY~B#qwjMjT2L6Imq~#B}0W8Kf9^^ zB$c2$xo!8u=y&b-vM?2!lh@E5{%CaR;#hqy$8lhJ30mCYcA?&N?COu4XFfQLW+h$c zHSB+B>OM4mb~(=u9UXH~+dLBtqe9pnBerb7M;-R~VIAJzSU29KeturP(t7Ikk|w=H z<@kBA0tmfNNwjOTkfUcO6%nMtD?=B!(YE!VWrQ%s$N2N{}_qCn4^`cd=1 zT~K|r7S_u*z>=-P%yKz7-Zt+k?5(&=`r@#Tv@{>!^V~UTJSziB`ef1h&3mSL|0HIY z?-RVZa4vp-cnPF(@1shJ3u@l9$NVb=n3=N_I{zGH?)=~w&97m!e#rvLKOaxGxoq=o z^RFgRV~!s6RTwBn`ZzX&NtR!}~f?_IKjuUi|dO{V2X#znHbr z%V%x#FSAc4T&0hT&r{gpqm-%UK-2H#VD`{-+?u}xykF=u);Gg>+p&fnDOksv$2zmp zHV!njK!{A{W@6OzQ9Q@Y#N8J;FWl12tbZVny~I;u@3Ir<#;NXgIws!eAIqz6R z@9fM?x4zY3wAy&Qj;E6;(%BHxS8jqlbqzK!#*;McyYRZ)1`_JEr-1nzD8@sdslWc6 zQP(^Pu32I9rYVfJPn|>6lfL0z-f{f2%?4Wp7W4dc1epG3`HW)eHF%Me3VQ^rplRqI zIhU^^9n~CsB5#jdrEgX?9*(OfKS(8sD{-6XqTgN&>j5w8Yd(EKf259El26#4oAj zz5E~eUPYWY+y-@vx*6VtD5m$>IcC8J2a@zwN4ND0Agy2~l&Gy{F0GovyWu|!bN^}K zx1k#luzm`0C?-Yh(>1j8x$47qlr??%rQY|IlPW>KdS>R7DY@*y&}`~XDhQ~ z8$VNV{uyulaX;J@ZD&M_?lb6Aj#@7kAX}@9d=)!T?96${a$~@1=nP3F{idpKhgkmG z=kV778QO5*6iwB&qbEKG>1Ulbjqx7Ck|}$z%hv_Z91aG>8R8iK%L-dczr*K%MYyFm z9p!~0!Frtr^#2-U_~v*(uPY!vH=$jJ+R6XF4*HooLi6f!P%QKZeqyp|OsNV-*jXe} zRzME5O&kyS1w$33$ilRZfnVb~Vff=vcHNa$1$?%TKs zTh6_PDa9S&V3iILE6h=DdIIJxK7p&lV(H6;a!4I^y!ScH znY9=gVMp;=Gs)IPi()@M$By)3RPs87b`pQFrDF^2o03WbXD72&2D0Ecr5ENENnoC_ zGWy7|QAKf7OvKs>~W#>;sR8dXhyT!W{^XyDHartk$#@j5-Gh_wJCJCL06X2GoV$6OOs>D~lA=mJws}0joVgrxv*0FX=hvg{ zV_kanMS|*PdC}&5VHBfdg8XflaN#OririhnY}MV1Z^AUm$;TEDBe8c=<^f_Wjp|Ir=-O@R^zk)}O z$dYuEFs*DCChPO_Xu#+)$gR;qDe(sIaN}{`haT!H*60V4x0crx=n!Wv&(yj>Tk?K{J9UB{@q zdkn*>Ho&o|+1&oPgY-npXk?}w&6!TvS27<{lLP6Ucm=8TrjbQS5&ceaB#AHsdVS#= zmEG{ATc?9D!$gS2xY?*^h#>Xvcz|X0@)UpZ2$d$((B$glq_BHA#gux}>eevI&g1r0 zQ7Y^QeH-@f9hUtdDFAUUxy;gsqP(UbF=TV4jEZl^(v%afD8GjfX9+BZQ)zMdqo@ni zBR4WZUFtaa=NOJX9>=h+9A{#Fi5yd}aoL_fwBFK~s^;f`=EH~dtWS;99fVQk{9e3m z$8joaPtxMkez?=^EoRB+((sE3)ERgb=cM(*3qePY$=ATc^_yU4qcGHFPJ<_sO=vZD z9HZOBP+t23^KyI#@@N6Q&)7;=B&O4S%YMwhv;~he)Kf~>Ny-?_p??n~;PN(Y2$zKO zE(7%0X(g@w5H<{1o5&_kw$V8zXvqDNR1Qla{DkkXyG7 zefc|stV+sp_qILsrHS*$*)ON@&?Y$N?@vz5VfKCgN<95r7_GYFK;Y0>nk=16-FepZ zr29SOy{dt_m;NxPP8|O0&&NwQj!^I4?KG9RS*1=Llt}QS<)9DV)m}vJ`vqv(n`9h% zn2pb6>WKg93bN)HxbUO=Y@{SV+iNjImZ`6BxztVE?B$F(zGHMbAd9B|{SW8-IzkIy zCQ<*S82a`j2oqwzgKek^BoBSUb>2r%T%{T<*E`Z}!%V7B-$>S}L8Sln562}whgWkQ zDc|fPUYh+4@~!KrZ{9Py7m`5F7H5)=PYiu{WJGTZ)oJX+YMNmiOLn&(P*_+lO?a5&zC}WN#No(V4N6P?CI{K5gwr<=17fChs~bbE?+jiJ&7$8pi$ zH>mC!Mz?z&;Z?~MM6JG5wc{!layyY7n^r-DzA9cF*a}Bx%3#vB2$GT1CBN45D7v7U zlw~4uzvy@7z`p`)6yxquaNN?D5I0(IAPZKy>0pOqJG5Pu1*yD!Jh`My8lN0PtFO4w zx<5{Izvwh}YJFnr6y>Q~;0LzwXJS@#7K&gFch6OXt(&O9aUP~5HpJBxvmL2iBpb_R zm6<(5wV<}MlnEHj=844~gjWWmOv>#EOjOuENJBw%oDc{8r)OgD94iz#bqlADh`^OM z8N8N1lW@I>Fw}1kUr;IcEXKD{jHHDQ{7i8Akubh0Ms*7C60U0rD3((%so{Smk30 zcutS>`n&t)$xj9t%=ysr?qm z`f+ZoSBWX$Bo=|Y3twT{j>RyXDuzK{52DMG2{?S&Da%@($J#>o*;W675d@E*3v$)UW#oSd?6yZ%-9y93ggWJ&Gpb0K> zGO;Xp5Z})ArLeecIxFgo#k>ok*}nmb?Zlw2lMfzDm4K^5s(50}WYWIvN*POU;6b+~ zIJmG7Hn^!Uf)`>Swr>QFyB_0qDcR&|J%e;x=3v-`-x#ORNUbvtW2BNKHLuFUZ+!pZ zqUf2lAz~H{?%IYs4O21vKYf}I+l-s9_|yFKSLDPefKqvL@gHcR#I?1kUXX%%R!-P| zsui!Ps*(tQ6IDOr{#|l){TeYh>{+Lce^#lW`d}%eTd7V`vnG>Gg8*$8xDR_?{b3>{ zrLgH{3u()|C*i`GY+KhJ%>JiA;2BMqWb@$Q^q0)H@`l<=D0!A-J`gbx ze-6_OyF=s^HH+3oWun)wB5a=Fhvv)Y(-|%UZ*YWryE5mpt6!dE?~4_%uKUAKTJr-M z>mMcEb#fTN1(f{b z4DC0&MwWl}qEf<4`p5A+q6ZG6=qYFP+*$%`a5$~3|Bu$#o*}=P4dBwF4{rbYGkJOA z+$~;7Ht*R3l5u!LQXBj!zEu`Sr@m&Mcuis$?zcGh=r(3^IalVA9q!#60pI2gLideq z*v0L$4WdhF;p$i#fIE24@GIszUBIBwR`}U-9o~#dP-^uixFw$ruVN;^7wi9EaE}Pw zw-iR{RY}y-QH-~rrZO`FcH?HDQRvM-#cZ!^2KD7RaKOX>{5M6zbtOLF_gVywMUj~G zR*TezZ{rJ-uZ)CaFud@wKt-uo5;`zOwR4ZKZN_J@irfFb6@Ns(<|ojXWyi^7o(Vhh z?mdK^Z$fcDK6*U73KM=caasLD^x7#x*?;F!<}gE&A7#+z%|r}d;sqA@#uz!{B6h0Y z#kZT}spH{6%1G*=3bR&tT5}ym)8j}FOpqQ#p}{FET$a| zqqk6w`9Gy;-u6|bJ}V7*KZfzmvc>fI1=mx{-b2X+Q|Upl5Dr)+<6YgERBg3@WS1M$ zSiCvCJo6q+uDVjvKQ6Kswi8DAQkglft0DKGEmpi8ho55cFmL5lV1Ld9d(i_hVjzjj z=WIvqnbSE=bvelyZ=tOmM{?6Lf|-_^3Qts*fY2*rd>{N1R5s1VIQx^3@=zS@F392U zsX|cvISH3&FQc_88I1k&U5uUU9hf?0H#9to=dx)-jGKHNu2z^wpAJ@%s*5MP)Ok6Z zTsoJvgJW=x%O6-iJ5C3G4B{2Rm2_cGCy9xBV%`*Wy713}N2m>NgtFMgzB4a%?}%c)lmJbjoWWU zPZ_j7ycR?M@Ip!22sF5WYSwil71)#NMnAkgQ5WSD^58?}O#H8+gJWR!(PNuN zkoi=HV{_KgqPhI!mnT8xY1$N$rN~hqmh?b;ki1h*qS6L6^5!0g-pSmo#HWKvs(8+* z&5i@$E5L~4jdU!}kY+^%QrfO_l=IF4BNDkwpd;_#y75)|aA$xHrCZV9a0xLt&d{WL zQncx`1Rnl%9r+5m{L!Og%pSRkAH$4D$L|{GtLmZGgAQ!^R*sr4;^_JfSyq)d8^2Dx zg_||J=+;v{kSyATz3UFp7D~ij5))~y&m=l=N`mem-b&V@p~Sz}1^gpApfAr19~W7W zh5Tk(Za0N|4nD)!)5|H&wT)gDT%}9_K2~&QB5TUm#bk{0V}|b_auP%K{uNQSGf9W7 zY`Y8rg{!~|MxmhTBZ@p+KnWb{wx{F@w&@PyhVG+S(0UF7*SM4af*ATMsso!l*Fwau z6Zj!skN9o$DM)N3t-i63q|RkhPXQlW>NSBraUA^f|4b>y{SpJ|i@>ba6<)1B4~;g9 zQLSSpCa&}cIqZO`nr;|p)`cd(Wf9-FP=jnDNSU9)=T{$qedlqLAw&R zZ;uL_^~{sj#r`r&PQA&91`Ggy!32^@GX;}%FBxY;O-ko7LFS9XQTijt8f^(ii#_My zLccEaYFrb}>bAp%m4|qNqw35HFHyYtrjQ9e-N}^xi{xpZx`#vBk!Z31HH_J`!GKy}>O~YVGqC|rp9RF_>-X1J4 z9T2`{nvF5MwuUZR73!I7@IljABV-O+F4BEXIqm# z|FfyyrB}Ox1^BHg1JOz7Kezl?`Um(|N4k$Qr8E9e@{|3`XQO z!C1Bdq%G(C-Yf5r$QNgpcRdGZ&M$-Z3F5$dsWZ9%HZoSrLLs4IA#bXVAIKf&vb=E# z?8nFbWcDtQwvX7+fxi;un|2I4Yg$oZR088aOryQ69qfu0eztMiA6)e~16l{ANVOx0 z`t+iCvJyXeDtjj5?2<7&vn!l3CSRq^?isoAG2g zC^H#$FP~%HaVd+fyA?;my`^|x&K<@7%3*rS6Es-M!^YEtz?&$J=E@l$%-aCF+wEy| zJOT@L9t4wZcY*i921{}|eq1+;4dMdF!u-qWdh-oRNUg{DpM_C=JeBS%jbKIfeauvr z;ClCJoZ798la*?qs@@T>pUVi<-h{eYl2E^25~Ws8WZ!Jc=XxSN`tqcOS6y!e*`qx$ zWx6X?U0en4ubgG>a?s(kWKS@YFJv5-{)JLj4Gv4@oZrSPuZ9nGV^j3Aew=W-tf)`Qy zf3NZR=WF!+b3164lyWzoWZBb~o&X@s)68_z|E$1ucE#%9g==B%FFX;Vf-^CFv`o7?7mlWJuQ-3>oh8u z{~L@88-UZAV!}Bo+Hmhu2pZk1Pn4zJs8XZj+ z!-K{V5c%&2(-#y=mGUmQ_Z3$|9+*gchKk_#S`W3HPVrW3Q-p6nDq%n=jmbZ&0C}&S zK`9}ecT}L6sa3wtZ1-w`k!jpsZe%9&cWNjqu6AJN{?z342K~h1c79qo{~P8OsFIPG zB&o@1qMY>*?wkGyr+Mr|{?(#n$Z@zGuA-dV=Oa9kv&M#}SD|E$C$8ETie+2Qb9*G%zUQU^-Bz6>|td;-aapYVBBAB2}r;@InW{NsBU*S+PZ`!c{W)S9UL z?;@l({D$}X+IVwv2NrwtD9j@rWh3@s^vuiP-kO47T#EVajrgJO4|8$-BnY-ng@dYd z@!E|7^j!D~BZqFIVCV!|(lClrNs;WnIXdjPxhd-uF$o2S641Np6F$s1hSrx==+T79 zlp22tH*v{`QSyc@tx+^b-Gq7!#p(Q~95mj%9viN5_u5rLIOffR4h4;%V4oF~9ZA9` z2Nsk6CU3gFh)2n(UC1}S6g$)8@yo|TP}j&H<@mXzY!gEzeFK;}JchE1h3TZ{LN3eU zj4ze0VSMpREFD*+pUZEex$#w8>+}vMFGf<8>VtBbJ0P{!0tdg?p_5=HrY+lwRt?vo zCaN3`G%v-MbAxePhYPBG)km+1S@_m82PZEQMfZ_DmoH;xC3Rihtcd$J~|2C#wxKB zs92qaNiIKd;I|MeS_ol1-(`5*;!TC8=F-N#Rb;(nKfe4EhXQTk=w~~h-u*1XJdWLV zW@E4`X+5UD<=%3USn3uMBI7hO&OPe{17ijt+8_Y?f4Y#-(@;8Wok-(nPGaanEBqoT z1-Wxr6dXB>V{QHzch3e!RNla%f>b=-{|(>25+&vLi)aql`v>fsgu91r;iI-PMjJ4& z#$5&7(j>uq!3$=2mI7?5lLM_vj%oDF0o!M*Q0MV8@Kesf7pwSj{q1n*G~?W5aGKkx z^Pn`oqyuj*GEaf%9iU_xjgBEc>3`%85=iik*JXi zDFmK{{z5gf6lHIdttXAN}@E!&qB&(pynUdy)gO{@z2p(9KUj zYS&;waSR?kk&azg+hM$R4qkYlgFj3Q@&1v!D4ecMn=aUshpaoj>)A}viyYyn<08`J z{Alry8gVwC2D#-;Bz?gW3@_=!PmxMwWR*(qi@WId=2kj+@hZxjno)br2W(iol+sV^ zArDiaPwM4Z<|;#H|9rtgZ+T{oZnrUd7lfkaXz7bm%4VIJT zP7>F6^}!AK{VesFv8kDFDLh?)HJ7cS7ZYC4cljSGih7e~SvjUC3eu7DMkF|_ z#O`%@#4)#xQ*knSTYtv}+dX}VP!|9;v zH?&8V#;*~cdOnK=PKQ(KeQU(@88k`HfF=(vqtQp<^mefwZDs?R&0aQtNTpez%$W+ZNFI2jQr(?IJ7a62|5ZKPLZd5ty8J$!zy{q-j%^xJ+G#1%oeS8Y4C|hFjL(U;IS_d0m!@%D_xy6%$pI!U&(=#k+q0 z45RxbfsL9Y&jwVNk<5PXw=QczF{sSVTk1fy44@!kFY`Qa3&oz9Lrly|oc{GQ9k4h@ z4IkH$#SUdO5Df#*h)^adJAri$9;Y9__&A~A>e0DGXM7WB_@E4? zONYY+{~XXNNMMgH)?vTyuBOQ|*I{u@2I`G!l06KAS|rlv(a&V7^psrB{UE!pA)2{s z1lK$Z!TDK)uiQ*gCHp5+@N5lJGNp!{xgv^9t(eMo3b@ecTpK$0z6*QL71O$S&TS^t z57yFG*uDihtjIncw$#Csu4--Ias!D>wpl8Pe3d6p0l-eqJGbt^WNa;1g3luUU?`V= zkRB7HmmfZ&?v4|X-tq#jsh8og(B*V4V+D=+g_xJT6g1ye=w<;Q1XMCVz-!{^cfG!-Dxr!l6VsXvR7T~**#m;k_&(67V ziWYFbi=+=TDf(Fv-glVFPTNpJ+3h^~9Gb*N`Z=;kLYruhoh(}0dxX5FXOc%tSWJv7ae#MvIvXsN75Wt;fv;%OtQKU+x`+yz<9`yWV;`#*8v zV;0rDLSem25ojFFg$p;9!!|A(U*)9@{~n)b1cSUNLTMgFdYD39LWrX(kh^plh4e*}da?t}elkYxJvAho)Kyn*mIRS1jPm)DgDlG3;-1n zY&!({XS0C48Ur`_E>qy>ZqCyq21fGX*u0_+j5c3^z+e$__qO4jffMjtPc)eB>SGpS zCiv&QC#@74qGox|D zktbvk##56k!wsbi(C}><6rbUEt5|M++geTZaV^?j+>d@g_u*N`hj3V99+;hrXK0Ns z)mYqx`#0mjNa8By^vYwt#w=KQ)0l+w?MR%D>(_aPtWcgdCD|OHalXxT>a+pto4pvUjO2V#pA5Kq;W)mr zcYrOYlJQ<~75euag{QJAxT+uo_;0)b?KnevZ`_FQRt>-naVxCIT!KEq?xCvT&&|kQ{5BDF2KzS)LQ6lhCamF(lZjj41@W*=2$S701tfUDk(fH_yM z41S)De_y(T%JNO%UmXS$`JACVd+Sy$}&m1@7Rn>Z?U%Z^RF_z06uf75QMagsZTNZvl z6oC3W>+rNW*ZahBweQqIG~6sh_oPm8HRF7|)g?XoR(YrBKvZ4fTw3DCh!^`#eob@Rh}P+WUd8<`(BK zil;&OW#lXGL+ip_+08Em*r=*7k~$Vd$`>Y5+Qwr@zbeU6+J~G!)}aI=%=x!DA6wlQ zNC!TeetQG9lPKI@&te-=zTYdp1&Kx zCBGfes^}P&4j#ZSo+8K#K8FSa!dT*U3oN6JxSYg0G*4cDSNh&#Kd^!a$=RczaVPhC2;npV$<7-C)f11T|FpNQ>&F#Zh|E12EK+ zr_$*WXxV=s*4EAkn7)kLRrbKt3nt`WSBs(=KY;(I7I~VSK*JEg@0I1)Ul0h2!*7_i zrPg$4J)1dITA`UTQ8xV$5}KyJAg|wIWM0>G~@c% zhdFp|l)2HkoHU-a;9st`F5>(Q-Xiy4dojnoN~Ga?Sb#b)(%2LpO=*!_wziQ+uerK- z5sA>+g#*}aFcA_5x|o2>5wMLlV@76bF`cQwOzZwfyj)X#o}R*09@Kt^C2kxWwy_Zo zT3*FYziiH5cLr@XC}G<71bq76aU6X97Ix41XLeUkpJ!jBKsFWbXrHISxa}6Fu25}6 zvnX)A{sK3AS%}f$`#?+Hvw_ z7tUd8O`{98UuO<|)?Jo}}5T8>#1~ zEX^y7!lU~x!jt7qJUA4Cx~6&QZe&j{D`pp0>7!4!0#0u`0fB2JsiHmxZLH3LTbnEA;Q0WbJ-t9rd;{$Z&LN4P zvD9^E0`j#LK?u7Rs>aPIs572g_9fFNE;oDO5$85-97LGO(!%B`RG=qBvr;T6_t6A0 zINS;|nj^5rrdy2Gd(x3q6n1V5MC$(;syiUzw}WFz5B^lRZjx842W4dz41S zbm_75WoSB+#M`vYj6Qvvjn=_Ya4p3T%v2vR6U#&4k3|wGYtNz&hV`iRCZEguZ=u4oE7v~ARV2)fdBhf@p>c05~n*~MXN~M?4a|e8Fc@J(*Qo&-W7@{w}BvbtkJC|FK zFL!&uYn*ep@%3-@*C@IneNX%ar-w<2CnjuGd}_p5>xh%oi7P=>OTy z3|`OVJ?k%ki&HWnr^X3(?b5@D!;9h@6n+bn_8QZl z%wu?IZZkT)Sw^y)N6>e-FqfUajQiKx(?r>Pd>MWkThzEKeM+3uJ12zR95AI_Fo8Cm zO+ZyBLWh%OnC)FmmxiZN;cX*Kvs5C%tTMdh-H6s(gtb%l8Og3)+ruTh3s- z>_=3q;Kvm2T&81IC*vB$hcedLXxzw$SLWN%J0&?1Yu-vTD#lkiJxOvnR?D{eau3Pq@?%p2&m&{;k;t6if?g<_gfSOMXaPiBjSjI|G)y!Gc z5%Uy>boY``&oFh`PGyUer?Acs7Lvxn9?U${fD_%sXr<&8ZXbT0_UoOY>%TVP&%6%k zT0R7Qo~rb{=?Z?0cn=Bgd0_Oj1DA{zk#ZMTFF-y{dpLqNaS9l0Ih(Fu*g{Wjx^Y&> zBJLK+LVVpQkEgqSGd|q{kg6#}vN<2{sVKK!-gF1`Zp2b}X#piTU&9*DUmW+NMRD&| zqT2mLoN}WQ?_Vv#UH{VQdhI!yl8_8b<0Y~B(NeVfW{m25Yrr+B2M>x*!FI7M5VUv& z+o~oIiy1H3$gpA3iGN#b@i9y#z29hOFu=9?NEN|2DzZMr2yY0h;w?3&E{MP`qii+{eT?)rjcuO1qS%uL7lt3puT1phD08tSXmbSQk2Bn z=l1wQc@|wvZ^o%B0%$^aI|`LKWB*lgJa9svMBYf#lADgSpzIyC+zq18>TP74bcOSf zR^rN*J7Al49)kSqKx}g$4!+tqTpl%s>((aHdv#B%P(UOg(j@E@*BQKR*TJ=uP{ z8W0Qnf+c4Dpakpb0fmr$aSlBPerV9qgsW=Yo;dLqc2t;A@nb!@6&Z+!?p=o@k0kVY zI0_dBdU5CZdRk@>g*W03QsxY2UpYf}u&xa5GsdaW zgF>3Ti&wx9(+|l#@o;etqKV-oyyx)@9xFzmwvY=_Y$@WLD6nw8$2_b##rzO$#+wVG zafA1L%(9+>JI*}8{|;+Xn!-A;vaW&CU!|dYZVmR%&*Zqgi=4OU8wh-^gxyPs3!Y58F)`&Kc;^bQDKM+qj)9ALT7|LQImvb*H#LX6?a!2h8wy`%|3z@GkV)BYnHU zd0`c2;MvG#^fc2T!*!dWJt!AM6^?{f8J)Y$ke8kS3Jy-^ z*g#P{Ug}7uUy@19@)$iYS^zS4WT^MaX*wuu4rj!|&U zez-9_6GkMTFw-ml@V*C5gA#dpe84$?v)*N(u2vqlTOX#E&$B2Y{R+7)+y^Gdj-ui8 z2dElU$1He|#6(tQFyh5-WbjFjLX_U%4U=BxkMTZOq}L497bc=b)*_BOs6wx`+sOC( zY?AtNh+f~%p~7|N$+nK4HTxI?BCl1j@ZN1Ody)vD-1q-;^DETQoj?V;M4hP%=-f(v z?B8FBf35$=(0PYr`9^WP$c&IsMjD8=@;>LBhsq|TK_nC^rSVJAlu|}ykIc#lMKVf4 zQbrl2DWOC|n~3)M-G6souFCto&wZcs{eC{3s28I_b#+y!s}+cwcRVF0=4A{1)!c;LyxX7;y6@-k*L0)tlqdm)-xuCPvWqUrAI_Rz%;l2SCWw9+l(R z%ynutad!DsF}{(?W3QmWdX{-fX+{3XYba;monud#(aN2h>H9BFnzUM->P?&|Yfldf zw;}pIE~T%g(R6&DC!Mj2!Z3wJH0^yPU2UHVX5SCtcPzlcbup+pPn~L|w8-$mWU#Oq z50?vG!o((PnpYM~7bnEgmJhXXV|qV??2N@%A576xCJH70V>_@&BP^+wrl=|;+Yc>Z zQ8EP9vuC4DoiKG8Xprr%L-gO#CouU_9M&4L97f1yaJjVx?whQDfnI%BjLqD#UD~Lz zV;}xij>2!uPjdBjFW9Jg<9SJKm=^jHQ}tuGoD5&V$fd`ra^ri9F5QJ42c78FliN7v z=`$`endQDtn$z2la`eT(l-{_DQPPq`9NgSXPCix?mwN@G1_7Naq-eLa4b3fSV@`o@ zbm_Dke=aVTuB;eM3D)V5RK$7*|E;7)8c3Z5&*_@eO-df^LraXNg6VxRF7@`M`R0`? z*-VJ}qgG^t&LKei)pH;$JV9XmGz7vEa=DFbssvBot7GA%VBGoJ7em^Q(#&2*LisKn zYMeq!hArq2Fb2QdFQ=s{A`~;Z0D@|DDCvDIb!V#ah5BXaHR~r%bD2-QQctjctrT3S z@)a!Ga|xn0X0dK&ERNdN#$|5kf{IOl(QNk=GXG_d9~O=0zGmISZ#88kduJM-t3H-w zJPG}5rNCdrhKAd0@i`a9-S{U4zU|GR6EzBke=`P0L^?fAl;josUVz?@bkr9Ag7Sfq zRJ1b`Pb^4<@0O!z@o1i;s}*VCi*wY_6-8QMV<_R7F4gwEgMsOrxdbhM@v+OODg6(1 z&6|j0pZjBSOe$KtWS+HitgfIb_O(! z_(NB`8u^=teuKxQX1wXpi>-P-H0?_!&X7053)OP?R!#w`y;Hc`*^1cpMS&)T>e8g1 zGF0duiPu9@VdRhn=lp#JPQ2Jj)25`*iR%j}sq7%_6a7IdlJdLNp_ zCvTlXxk-<>X%k2BHi8|rCNvuqm{&pX;R7zI`=-Eup*C-Eqn3sfX438Zj9D_%48KNC zgj(fNGEaGr2f8+J&Yw(ZPsajMju4?4zSD5w96Joys6z323rYC)Y+5Ngg6>=zOX7DG z@nMN1TzWj8-C1Ul`r{&cHZ6oMsAQt~x>WAa4I>&Ht&8_%Za{?Z9CX|liql?{p_PUu zm2cHW%XlHqExr+63qm-bz3yDfeZi*Y^b}Ge)BD9ueGoSO>dBHF1Y( zGoCNmL7y(fQgA{M3N9W*HRc+$@@4aq$Oy7j$fXt6)mV;MiJByP@vV6iZ77kV4J*gu zbLt`Q{O@>jmKQpHXeJ{)1@dHjA+KMOI8$dFX)lwe#c83GSmR3uclxlLu^ZP*jN~g$ zas1^&hIsEpEN*RDiDApXVb%i1G)Opu$~paHh6ky5 z)nn4_z0Bjp`d~lVK3%VXaGwl6On z75NFMW9Pst&3Q^GJFCdeAef!uzGBw4&6tt+3ESKB$@f12IaRe%(gF*XgDs_FKN?W> zGvZpeesFykiCqh>!gdxy8z0<$A3Yl7l>BGq?`$CzK`mx6cP?vdkYi2#%N%b`@t_dyUW7=UZV;8vLEs$ZeE93=O$! z!Av@Zbh9&%+apQbr{gHpYl*J=_Ct5+B7W=30NShi6qnU6rAB1gu#>*@PGUcWu8yT& z>x{^hvD|wh2p>lOL(S(?*m)q5HvW^KH`R3%Z0<_#%w;t#+kw`H{RB1dR=8=i1l>Mn zvw4397j}~EEk7J1_cgm{!5T|^9d;3P{8M2vW8)rvyAHX>jN|VdNXIgzX#AXB7_zUS z*Q15_R1-GsDqVq#d2JNdXon=_icOW`um5-hueyK3Zh8x66@ySTFc(XBG3?!b0@q3% z$0G)l>0EXc`B?OV&&OD9qL~<)HY^1TP8xDM&QfrDCrK@-rrEER>8VE?&TCx?<~5V4 zB9*bIv)*#jH*b>Q(tP^tW=DH2yusRlZ8$~iIhx%110D6ZAjP-{rbw*fF5kPsHE$Zl z1rvf~@ZP%^E^k=CP0ZOsnH%JJ z!>LcHVGq&d&lhP|dImL1n|rG8GPzXz)+)^-P}?pRIdCu`C9 zLyK7s`z+h*EM>kej(QV}X~YWw%5=xmqdhmMOgM!|rX8!*Jjsw{TL#Rh)A-s5>QdWH z)1LK_=D6o%y6HF8iB@2;Pa7ob+EDrW$@IDHH%_!>yowRu@Uh%!?$2!IXVt65tQ-wo zpm75$y5*^CNQC}aG8W;Q-K6HemI6QQCxh+E^e{A+J{YCY{I~YhofJ<`!|YM)_Z-ql z+D;Q$M?G5j6Q*w+O(t99F>~B;_`R(I%sN2eN6ZEVD?j~;@3u`34MipEm~`*FiN z73}ED1L+SgbgcX$U586#pKuR%NiW0ujyE8cb*kzQPNx5AUg6gIAlQ;6z=y-*@RO_< zMX%pXBk%o3zXTERMAM8$Vh#y6=#tlTdAe=*1UK0)qW0lqbgljjy^@b4yU<2bt-MBJ zn$~z;aRG){{D4{gl6Y3ug5IBJT1hwkk<9%FtvG__r|qT{ zeLeKj){i;+^yy!uH@hcCQ@TMYZJlaN^J)l7*6>(TvX#1P!kOr!}SQRH#`B1&E4b=(pr5p3Rh|dU*+Yq@IG=PnF4~Ses-P=hDVy_vnRb zKTR@YjEhI}>6wZ-6>hU5(*p9KW;>3;J&3`**)luyQ#)&)Y=dZxbo3a5c&0v$N-Ece)!XNu60d6;=O24`asf z!bE&hRLIU|l`xpO2&OB}g=1?M!0G5RxO&(C4-J{XoX=I*Y^g&_P94V`f{_&XdIg$& z6Q`XIO(;dQ35|`r5KcCsc%l?#3dE^ohY1~enufz^vUE>gpQf}PVa&S=sL%K_ZqIZm z?dW&BF^_rboSxwO^&{}*qaYKYd!IKlzh(tac=HL9vO-{C zz;y`i*aLcX$5_Vx26p5Jqe8e2og0|Vm~!&8EY2NGS>NLKl_hr%5?JyVXp#w*K?sBJ`?I4ly<{Lg6;P3fOuv0$?A8k2? z*_N9ixY-K7e9MPDYhqzafg_$>B1S?t2QX(c^Yi?Wp%LktP&HN;#T%Dlij@fbR>}ak zgp)8bF#slPPeiF#ZD?n>8P30+gkC>fFi_4ek6Gl!OhVt*e;M#36{H9xCy!;Fa{_>wBl-9{)nTuq!D`d?)jW^IU zJO*1LCgF=C&M4NPh^_3I{pjgLTQ8_nbn!y$Sg(u8J^$hU7#+N2u8#wU=VDXdJIXRV zO)oo6Qa!Fj*dYWaUGfC9iKs5x=f4qXl%u>KT%jUn1+Xa?{c#?DB+(_A1o`1CfybBbaw_) zX`v_fhzH~QpilUImpR=%!tOo)&Qg^41N!Oqo(@0iA=UGN^s%g%RO{E%f}%BOQauQV z6z)T$I!6gY*>pChj6BTCN$<;1s%T$Cw>&%1+8`BE3O=HxogDq0yn-@=wooI@q*3!1 zQq1E?w143Y;#w8xNA_Oat?P`lZH_9EMyQt0_XSb$i!Il_r$Sqh6_m`?cfyEd+)A0#it*#(HPLvd0GyX@@75oxpi=WsW z@mIQ3UN8ke)#^~F9M zkUfs8(=L${aCbJHnkqa`&PNYJ|B6BsJ$g^>l$I4JLlvDc%}YS%6_^NGO^Nt4Mj%8@P} zlclv$+cBqm1O`7m21(1Jam_4C3cMFbH@}UfcWl2`W_=bNyp?GnVG)gXTTMH6n!@9s zeqiz>5ZdIIlE2C-8vaj`lqP7KF zC^=i7428~6zt=PJy_iYUzHTDhKR_nug=v}INxYtT9w&@hPv4_IVQok;=5~8h(Szlb zhUfC4s~(%yY$^d$H+ zC?6k%>g@l-Hlh&Aj}kSgn6X*mD*CTHoI=OOQ|H}5<_KWj~B$T~F~-4JdU~5bQbd6{ba%fk0vk?e&~RO2&t$gbhD;wsd*G?G44Ig@yAEbE)^ zA{)l=dAUyuwY-lZHC)0(;Uc;xT0o1g=1{^sasE+oC7aisq#T_HTKm+TWG*D)ywU~O zIo1V(x#`$1Z$`R88>u$C6FD6NYME?8%Yy=F$HOfAdU*)dHPTRf)(}2;BSF#w>uGt& zM%o>)fP4%N(CM#_aao-_^%;4Rdel>H@XJNW+nozf<`&>~MMVnT!8*D#IdbzDz}9{P zl%;R9qcei`GjG?w+x6JkWl#Kn;q+luDt+D)M3$ziWV7`sm@d@?(Zr{0U+YZ9KO^W} z@ka6q>PL;p9_(6t7blinM&Dt?0jqKWMmi z0IhpE*}iKkKCU%~&ivP$_+S~gZM!R{6<`fctS6nVKMEax_kgOEJDv-=%4Sskqx#kq~?ISP2J=Oq?;vJULEP6XF<3>Z+S5$8iOQYi{I34hn~^rimOyH)0-xp5F+`y zTv!{gM|#d0^y^R;w#CL_`>m_!?Z2MQRS#kA&9m??kAqJdA~bc`GZdJ&V{Nt`jBhA} zwaa<#B-`CUc@F61Hge6Egs5^y6Pm6Z!gSY-lt|j)$*V4Ui%2DS5d&OO-*1~`~zykW|ErjdAw%Np&#ou ztQizg=d*3Jeb;r)@OCP=Tyz1ChHScBtVOqPYS8>U5%}^~2+M@6KwZZ-=u#7md+&Y4 zT`R|+WgpAPtlJFJoXjZ0QI~FgnNG&#sdNT+Qs00B&3V^|Tdqft)4mMq5<5(lO^TFo z^(Kl9=1`o7DX}gU&6*^F54?84(l5GPtNChP_MH)L^Z|L}94!j^^9w_lzD06oJa`vf z9KpLo@7h|r)VG!N?F4jxN-$=~PhtJpzwo(nA5?7RQC4^}o;|h`TSf%p1y7bA(|Utq zb=vgWWd_ZP41#0P_u=f$%UH*{{W9(+@BlZGEc9OTTAxqz4Q4@nL)v0oaH|G#1W%x> za}WQ5@AA7P9YHcEqJbVD&<{1>=s6hvkA53ze! zI0_kkpntPI&_<7Hl4vaF`pqZcjDdKRVZF6s`)PFlmkv3fwIp}70MvGGguj72#Y(x6 z^8=1NRlG?0?sT#)oj?+w1Q7e0dF5K>!0y-HB(q{YE!nY*G?V|~IgCQno>EMclc15w zz9`fZ0Y^V*QAB?_YPDU%{QdyCpr%O$S-`LXZ1aZe3Mhh>DsJ%vzi z)B)7f^@o*7E5SJ+2`_GI!b=l;@s-d`nxW-Jk(Dp;&X6)z_}JhLVMXj~yhL}kqUf>6 z3OXxUMMbw)(=88QO8K&rZmneR&ZZM+-}#Tbx1<7Yn}24xhqw6IhwUv!j3D*63^-l4 z4h#HN;kmSf)aI9hrB+MPze2JX9$|-&_@GblDpWs~tq6+#&4Vc^JQM zTZ4B(wPB*=Wo|$viyLWc!)=yu;TEO4!0r{|(0V?HyOv;t2VOh`bC-88cZM%j3X9XP zLM`%9E5&#BoKfjY4f(~;JTbbC$_xq9l;BJl(C;O0t_w(+9f67iUSQWmwmJjieS zGh}pRygg?Du?Q7i*y%+XZO5s2{y8%H`GsyjOrZ@gPm%uBW3*F50Ph#hMge$|#0*io zxGw{r{nH0;PiHDBD51D>8_470R_w_`(g{za%5Pnam!OTiLoVZs9Y(Y%u?ZVT5&m-g zj>Aj;q1(GMOdVH@clh_XCHgrYFfGOXLSE>su@ak7B`H%f2CvC)fytNC2;|Zkm-+-b z%CT&^X)(*;CQ|WQOSJm>7VlVQQID=T)#R|5bYm?}WgXU#7eYAWxg=(edW~wMnX`>O z+Z8t^(x|wH>i&L z{jxA+ttK{A*TRn718`1rCeet!l*@-u;Uf)ty+xYwN6(|m)Oga?a;EDdY!7rZm4*(S z##dTJ+;o#6JeM9!3!>LkUlHrpByd=slnQc7#*(qI4coga(YY6|uy@;2Tw*bYoUdG` z`=4v+zj>pn)nNeBPD;?n`1LLlr;Giy9>n8 zkV`HiwL*w*{b_><8SZF2 z?iU^_CzOcR#lvyt*b+Q}d7I-Pd94X|IM9gH%Y|w7gFGxBdw^nMI7mrWB+(m^)A=I}A=X=;E9{=QVf)sGbc zduM41E4YVJrMvNG$Whv$U{2=+^Xc!XKx#J}px&KdXu&3Z=yHrgi+nLki#-mvMFLRN z`7!o4c~MU7N%|FdiC!0JQ{+M;5wZ# zv-Dm4ES<6qp%%w{Qu2(Xl^3@YWFYDOO2XR$w=PNx5WDPJ#8%8T{1!c2~Cm4Aag9~tbP z^#x>3e5Ly(TX9+v4HNs|%^YD|BU_7J z&5XNbxRgEbQ}DsP6mXli3U0@az-KD?+?Va2^$JG4BBgd!YLwmuCW^XvQzlFBr%j2| zC@rI(+Zt$UQX;C#GcWG4IxLSVLE~l-YJ`3iZy$$>W4F-$_COk0$UN=~Uh<>0pYk)E z@AI9X(=g*y4i~&3Qtw1+J5^b)r?KhABokx9U-R+gH67;g^Q;$g<069uUHQ&jt<)kK z_ezaq9n71rC#CLAr&%5uJrlnhT=;Pb@^zhmxj34a> z5^D?vVR~OMGG!@wAK|EHfaO}&vrJp>OiG)23%cr`!RRX}r8L35v8|!VRl<$cd<-O;5!iB5%~2wNY6@0n7=v{z?(!L z<%fGulF5&?l={?#E{vMRk8RJPC7d6gxw3}ebS#NK_0){-Ncu+wtfRbG_y)wP#^A=r z6rArbjw>>UsDFkI{jnN@6J;!D=ZQMZ0UOjCV(0rfjt*-okZuO^v+bWwPyGt8NO7dT zMVh64ykMq&#oj4oG2$^QIxIx*u8;WJN0eHbx9@6R2sd6TnY-2d305)ArL&7Y)fw$W zr6-Y?|2Z0+Zp^??ol)ea2mCslMSSclQ~u6re}3<@06yTM2md2chu^$gl#jX^$S+*z z$=7@w$-kRB5wj|9aK#72VENKG@?SoZ&BR^t@y99rA!AuSYIh0w?%aw~N32FUwg*%D zD1lW54<*q&#)Ag6{kSJB|rT0-UjYfRd7~w+9{dc1!o00 zUc@1+f)F+##LgT*VSS?)eoP-px*)pAzCvMe}l? z%yTk#&-9I8$>(qq403cZ))-47571@#YZxp&9xW%_q>-I%ba0FY@iRloRQn9(>qNn= zHw$ousvXP2kK=~NH&ZBM`9!5z^1hn;kY^tEj>#IFa*jDp+aZke9T1mfC!(g48VNjq zqq@>Ln0zpTOMT+SoqQk;=IaKzaX(7A_df&pz9lV`>tIM4KZBTquYpU*0njkcfnyux z@y+WtC=~xr-=vypRPSIj zL<}e820+ELwY0qXI1OgY(2|yN`f`63-!UHe);1kf4Y{fpqFFj`b>4Zp&*nbz-rmTa zXWR%O1>Wtw6MyK40(e~bq<3rV7{Rfj4jk{CihGOxVg5!b+L^) zZQJx-{7Jw)N6zBwA`9FbFq2Y`?8CUUmso5OhhNjz;=YrAKq%fFJ&r0t z1(A2ExOeT3IJ0~s-lS_bzoC>dr~bC#w!t@;#@JlvoFuV*M+?54DM{NcM6th?{oHrg zxh01*c&{FDe#)CdQW_nNI-0taEcFBPmUW@Ei6;HE+C>?Uljx+WGd2A%qxOIPG$ry0 zZf*6z;wQgAFJFy&8E3|z@>Miqe-j<+d`aW;a#8di>mfg!OUB=%_`kuGsHfqJg*TV- z2PHlEmnI{4qX{w8v6%7o`ZNUAvx*`1k1=|kQAV31YbZEo5$4H?fRO{T`PBkyaPB~Z zWhbfS`##KG=!S!lQ8+Pm5_3YnMs4)qMV;>RUxH5au}iZk7b`F$RdS^$l<%I0ftT~B z-=v%ln-|carx4Xxc;GU}op|s^BaS_}4L1Ghgd*EAa8nOPcfI$h+?ekJ3paZg~G;zar?wh?z3xU)Is9@-~S4V(8)=g$5(&HXx) z#+|ztj??pn<5q|->Oka^C=t^<|qX(%FP4jYrc!l8wx zdB~aI zg|SG_3WmXE=X3xYGcG@J0Zzf*yzjKx7sgDpJJw6e)NEX7u1(xv5 zP7!8VHgGcD6@shxYPnC=Be{tdMgoD^M1ix`X~Aa$J+Sk=1HO(GP%Ly8v$dqP^EXVuJEU{P=t&E z3)67!Y`6;D+RJ(q{r!TyeQNyE)p?X85=lv}auo0(9@U*%!6p@;du9liyF5hT74sF- z=Gx=jyjB?3Bu%$wr(vCYFn-#02MTwZaBpiA1^X3;=}CwW`8zy95${aUPSJsmIU^uA zWEuwTzXPk5c|o@HAPnv=1C4uaoP2Z!)$LcJ`Y&myVEcmF+S_Q(&U&)`XN_8G%;{uK zE=ifkQ;NJbDY?qh52;R4><^_DPZ`pRkA<7hW^=Kp%>|0vFY(9rT;k6OpXcXCea7*& z{it|o3obukhgRMqxO3!f*t^Y)-%;C54f{h#ErWSnj>f=<05fjc=M$V^yOp4Es8KJb z-yFKaa^buAY0hOl>u?%0VXL1!zB{Lhc~!S?shq0G}l*(RFen$bMcAyWZ}HL+=lOt6e1bdc6)m{GgqR8PBI_-WWDx znhIO`vbo+lqj~ME78HA69tryA@Q(|fc#j%qUh_*Ku1(Oz>AWz;rW_-#QHp$Bl2QJ9 z4welag>ePFf+6Qg)FEYyALV61Q7e#rs~d1^~#x0Olg@=>;{e+rm=5U#f}f6onJ zFg+c>tys(M5t{k%cmEqfqs>Un4(sCjY8MN1{7#|5Zc*I%;}Q3f%O;Jo8Pu~l2CEn5 z!dK@|2o4_uF2`kg&$Uss`lk?$5WYp$M*~SeH;#2Ma%rTvfb#U4P zP&{v#?Ke-7_fj`nn?4mYet+QB@5$!k%sX(1Ylj!}ro&jz05->9e}3y3KGE68xK!Z~ zsqqv1LoC}9zy9C<>{hOIBGsJMsv+jcr58JFca9}zVW3%+nu2MbScT5 zNuXI$CvWO52Ev(>Hd4>r*&IEF`7ALqTlbyKM*A03yLaX)q zAIal!b9v_6kb?F`=CdepC;j(hDaCFk?mu@J3d*NKLv#n4{AQD#ksncga{}s>oWUb< zznEw34vL3g#e|p+EK731h@@83yQWI9qDwJIwi6%zoy+FCVbD4;6`fp}qg^Q%K3hiN zm4{APkduT~h5wO4axsbCH>2>wpK;d>ea8P`S+At)*z&Uq(=>B2?BZk^rBX*?_LK4C z#UzY8bP)Xyw{Vq-S~y!qjDR_G%JvDvy2@8vYKsT1T0I&$xn7WTEQeOdGUg4k2emoM zII?vQ8vFdk%SRa>n0eIA>qgPRtQ*X65{PbGBY15%jtyt!Xvz2x`k0>!?=Rkgfly%- zXM7vCqs*PB7==zk1YRG$!Q8WiMqL*$sihx3F!$7fHgm96x&j&3c448a6o&u1hsq`T zbfNSUMy09Ji&M6Ap)X;#u)ng!-r0+ONX>Sj~G|C5sgCv=$Gdi`V!B$^7n!;^ZFK?o_hx~W<-P6 zvy1Rm-yA1AKFmIE{dh+0JGR|UB;D>4RLb}sTNZ8Nj9LwF=*<|q_qUkVudX9Wvl=S9 z&)DzxDlu&FcDm$PPnILH$or5x?W^02kA0JI^}0)VP)Qh@H1cq6mNT{-EZ}yozY9B@ za`2c~1xzRur*faubgQiaeh)0hrI!=1Fe(ymM}_0L4a^DU91r`Gp2Fag9GKv$fvTnK zcb7L6SA6`3$9B!8;jcSr{oEXGrR80a*!vmIooj-{w#Tu$;4VAQY$d0ZgCu;ZfR^5n zqPfN!=;fcpZi`t;R!Gvz;@ga%@BaBb=p z>`)Ovv|%wcT7`o7za0=hbqem6UPs@He!$`x%h9*L7%OVz@Pp@D{CA`QZ!$i6+eHB` zU!;l}hB7$X>jz3SEoHvsh0ts>32j*pG*qsO9=vU*kk)66Yub!r3s0l}P!7uXh0~qN zpJbKvhHBcc(3myq`~<#$)~;c{hq)W**tzE<PRivmA>!s6=Dbf@SH9@wdX zhk}eb`R#Gw`p*H@yfNtNIqsAed zyV5BK&RfE`dO1bzgorr={8WRuC);t$>QZb?slY)aAxxj~8-AY>=1k%{xR|lUoR#Q3 zkUE}*4)v!|D%~7+Z4$wm2W!B0rY+iKdZEk*OFWQHc-g@nAFejR?jd7rIPnjTS}p_a zg=sM1UIgAapo{-zs9=-HLh8{_rm%bsib*;E*&fy0A)&+E!V%v18UYJ*JHYqG3CQid0xQ?Wau+r}Zw%n9&I`@MEk}B;Qp_>Fm$6k`UsU{$oSQ0vqBdq zt+)iIh5}$Co8eGH2D$8TqP25(Q`ZwSRCzcZmx}%Y+gArr-uE1iSyO=FDYNNUxHMI{ z+{QdHOH@7^jFFM$_`-W9aPFp{`_uvU+CIc&o9Bp!fwXdqaGrD|PI$Q=-M7xeVEqaB z&2B1=|9b|f1_z+iuVom}^?@!9#nbx2BvP3j2cye^L2c0))=?S7IP{sY$t4<|dt8E^ z!%-mqUmk>tyJ6x34eYL-j)zq)fZ#_SZ2BaM$1f{T)WRkBO^x-Cwkgq#KP9;4-B)~b zEEjY7u3_taA)0aN43^0xVCk$7wx44GVu*tw4t+AgrZt=K{6i`HbbS;$tNey_v){sF2gdMUQxB3suQ}mA z>D>Me=J0OANxUBxgc)*wpzBB-82F__&Fx}%=Jk=WaTM`IObgK51iYNT5@&h@;*+X} zcwG>K7HLIjTRnhg+J@s>y(@V)-yL z;v_gfO$Ct`5uhM>5Hxd#L3B|Cp4nN#auxpgUDOZ<>!NV1Y$k3m_JB>hmc!p#C7ix) zF{F<(fu9eg(ZOsp%RNVOa&s!U&fqTY>f{)1$Rz?-$CMxyytd5|j&tGA z`SBTKW^IL-{wCbJy$^*_*w6Rzp`QB*^zqVpn$=p0y4%umB6I7Bo!pJPMxV#@Ob*lY z!r(@t9ZF2d$8SNSX>6%2>FBPa&TJFdxOF33)B48UKPO64D!OrfwlHb75s9Y`V%uRU z;*1|b#y@l1Z?qdCk6r>(#al4qUKllCT@6~Zbpy%uc9B%2GrddqVcnH{# zju2jVpn?Rp>xkXJx@^1fsLW09_nk`GZ-0?mQHjkG7=p_5rs*n9aowuZhHS@D}}_wJ>J5 zBd&h<2j3W{pwZt;`10l$Y-PE*5Ro!6%V!yyI7xEIdW_Zb?PxFa7(dnZp!S@D7?AQ9 zH!lk&i#Zoq#^5rp6$bdaH5=70iqPW95wu5S3#tC+LBg|SY4`A66y4&21NQ@9xRdP{ zH0RQ~MV3^Q=+78P%NfV>IILP8LfIHi1N;>_TOEjzp5HN3rvkS%>ymFAP~VT4^y-ZX zh3}h4mf641TWBFJNGrtW(aevW_Y#zr%Hqk*^I=8pPsm?Zj@@P%jA@}r)y^N#?w}ev zj30w?266ad%t%`Qu9u!JIYXOHGfw^&VcwtdDc-M5BK1=dBs3SOa$XJoeV2&`EM0N( z_2+Qp+YhAK?@@m9R7|?lKUeHHOwWS>QG7_5lDQq5sp-r#JUZd zaLc_4zM8CtJcpZtsvYC8%<=@3oHm3#t3E*LBE~z`8-j@+dSJw4HQZnM8TTFX#)!c% zeBKy{Y29vuAP5vhJT?%B1Z>1}qDqkX{JdbU_(RwfA%kW~86Y&qm((_YMHAzHVB(!l zN9sgreqRr6-yMTnQW`;L<2y)xpG3bp+-dgZpXk1}gChDmC~tW@z0T`jsuh}n(N z>qzgzdbt|ZZ9a@SPl8Eqt~$9h_gIk2H#omf6K7^$hLtk%w68r5dfN_j4uwtvp%dwX zWT#TW6z6zu-nVapq}e*$*9RfABlicQ@*S-ITSQ{aWA_}{EZQI+#g`t0cI|X-PJ$|y z`DJqcRT2=J+)XBj&XC{-W7sX(MQ^w%bQRn1)rJGy)tPR3&nBnST(n)G=sT@~qttNNbY^ll4*`IH!RoD&ZZ z)*s=PsSU&XKxY_IafQK33I3JYSH?X^BApcy4u{F4Sle*1Yl_AYC|qoa;t zUV$DiSj8N+pY!2IDM+)=WqF19VE7=G``I*`gmP!m zoIOC2Rido_GL{ZbDZ##m+{7``hIjVf60L@}6swj23y=_Y<%N)T`3cm~rX zl+beX9Y~HY!`Eu%klvIEpG0lY{Ksyne7g?hO?S{H#r?E|WmA+@gJI>uYf!dz32dBr zfn0Yozth|pioU%P$Da6sqwn%$I@g?fJ|J2A`iuUH&eF88!{`{UkHB_8nTEsYRm&WB zxz2dDToD%h-V0xE=g{UNBWn0Kf-)v{a{BAWfzv)|R=4w{w#S}yqHzhW*ViC7$wDm3 zcEQs48oa9-h{3Ns(xBunAUND8_}3 zg>ZDdUnY2ESU->twq=Zxdy5+wKw9{{)Ko#*z=Q-IM@ zh~CMfuygM=F6CqiXtVi%s#p~q_Ya}>Hr)KT*Cg|vRNVvb$_BU6}TW zvkPLK>l!I2$iIcAy|dB$a6Mcseh%|S&V%OPa*!YQ8*1xjq1MdvFl&W0pQ9B|9hr>V zQm~U^UUuWsR=_F7r?{RUaZnUD3m$*&7c_si6m*x1aD1yAcRyO3GcIft{B*qvym}jZ z$NRvJhEr&Bt_-iWufmtfck%eC1axH?0;9|FplP)nuT{Iz_D3h+yIKj%(Bk0CqYSV= z5&>zmwnE|cSKyMj4uf>qZp7v)%rI31uUq5b*XK&`>6d{cuk$z^Z9g;(aKk@3+W1qs z1s4zgg*UNZVSe5|u4iBha}=B7j8YqzuuKf@zV!r?L4>2P(2S8)!T zoVb0PSZ`(~nZgDy^c2uDM4Ii%Sdm{H=vOBlGR1}1C{ zT3G_EiMRCr)Ql2Xeig-gfqL9>VK0H4!bJ$~E9T0(`#I&)K=Cm6S_m6LY)k5fN44(vDg?o6U4d)2tGa;Nf(k+@Icu( zY&pZ>!G+3&XY3b8O>aWn>mRCBgh?g%2w_HtIG?OHTVJ(ULp z!?G_0n(@E6HLC9g{f0e)LYGy7Ux!b@<{~Jj8 z9>@i1)?l6SVN|o0#Cq>d^tyzn&)aRNCEQQof9$W|du$LlGyg7kTc<~0`S}qy;YB_- zd$3Yazt51n6K4hIqffxY88whM{Ez#x<^)$A)6SK7-Nhe{A?VN`O4H}qpoC`_o;+qr ziyr`5d;Ejgx^yr*wgvL7mU7A%&3UZ|fOY;Sz}9ma)TwvF@Ca2{D0YwAVdR1Zbta(w z;0L#3q7}Lx*b1H{v$GJG{(pwfJ09yU4CAB_$tFpnkSG;?=enP0kOn0xG)RMX z+Nrdxtn5uu3KdZql`>MIB9Wq^r16%9w6y3wfA>c|>N(Fj_kCU8uW)f{y?JT>Y_ln{ z@`6f*r67LXTyV6zBh30Sjct=3gpzVu8n&PsqnxsM@B9*Wjncz+nZ(iI|E@)DJwl-=plWK?*_I;i!xc(d~JdQmoEZ9AlHR)bvsh`Bz zR)^c>F=uRq*vskcldm2-dR`O1N=sSZ#VttF7N?_~3GAZsG?@MOoh1$m71nrnnm@cU zL?~LXOV$65LVxu}QtgNr9^F!A7c)!I#ywd>UY-w3!8;R^O>n~=3gEiW2@~KN*l{IeH$YT;+^%*cTZ5{w}ieqn6RPW@3B8-a}Wy|L2Hw- z`4{&nv!0IKLgzON^H;A`h1SDa!o!^B=H^PXg-<1Sg%M7P=5(>oe1BFF%OAUwS@;(V z(obE9}Gof!`Nw#$HayEGO2kfPcn zM_{8`hm~R)^zorOU0 zbbnAW#3NhHe;%q3_T`-7-$!-mSH1(qUQ-~dx-P{387QngB{22XslpDUG$FC8P4Mp! zc>b09cYAmSeb5!2DPK=#B*xMRUWWg@>`| z<9bY#bC&&1zs~&!$0(1k(37)rwEyB~7#z*V-ut--xR;MxLqA~j)-p=oQ%vg*og&MS zEEsPuLa=2x>U)a$@8L3&*5AeQ$q);S?qj40_XbD}pbc;MZo9yg?#rh^bN3zihlXLP z@m?xaH>Q(8H5gZOkzVRririb4iNfY&!Y*qk=83OBR9PK#q8H#v&=CB(Dvh;D)l9PE z7VF3Jn-@}*$;rDv_c}QsS0f5q-#;MJB#+e~9L1+KDNbq(jcWS`m=|Mwb}d|l-)LV| zh#}h^(uB81$gxbHj&8j!luTaD67Ro9U6VK+S;~34cm6P$hax)Jn}=DAyoXj|gzInf z*=(Qtf{XKUEX?QM^Rp*XXAaLjIo^ZYlpK6eIE&HZ`I!9S8A?Y?gY&JK`0=P2JGF+> z+bz;`ROKJ)j}U#adw`UkCm}wv49iRxLu`@>)XYI!tP7#UdzJNeML0Rr0ar|wa9G@n z1~i_aA-uQrb{C_RCVkrL9*(r~eDN|GpI9eQ+h6oM%pN{0yi_E0y~^vzWzpKQtMf#MX*D ze4F2oE;j{ITjya)Ogchmeodh3o&uR~Uri~se(d142Ih8tJo<~53Pz5{gjrJ)*)mT# zHn&xd%`BP5B(^lOhdYfh`-LmqCk5i*uv|pi<={e&B8KLt;+-nrxujKa-}5e-8}k8< z@w4dIhO;C(pF^i~-06ehi-Bq$?C=z2%u^dghI_N<*vb%Oy-7pZ`eJ---beCM$>ccy z2uY8*#(D#aSXZVc)Q6s9s;SbX`1dl&q+F!zzYAyq=MIir)Q1YA*|g<{F)6)2jJWDc zbm71tQTTm*gvhLAzm=8DfA1JdWs5j6d!8i)XYRzfPd`{|#6t`5Erp5!j?hX*pS*dRu=h1Fmzrg?071To-!x^9>PBX{jOYr4D0;GpDk*oM1QOcZZYQ6N@Jo{7|#~kNU<`R20-gXGNjaC;a#o1uU-R0PvScO5;d-z;Hi+|2J&u;$)s6^B; zSNF3*qLUcT?AE41QYXo{U^NA9mL$Wh1yJ7_#^eV_vYiVDV(`AJSX<^nt3Ibws@i$1 zO|+rRvB6Zmu?K65{KzTiCRM3wQ5b(WrX)9U_ZD~0^IUy~=RwLJ*9Mgp@?_ShM$Z0o zaZ1}B^;bkFn4*9M<>T@9j3f^I)uEU2b0LmhSco7qQmX!t=W-KSXYk$M@yTpzia&~j z{$g_KCb~FoBaIIlOQEk!M0;=Qh^#~YP~6gFy3Boi3m*B=RVz6o>6-g^(% zT(yzQyU~vxa2BpjF2y@Nf%#o6i;BJ+h-QNlR$Ng%z|;^$ZRy^bs6tL~x0J z%RXR$urWJ;UHdD}{$9CYK6Q)&raU^z`4XGiZJWzX%zZny{FJ4>YlqmhtdYVXNd-YJ z!j$ae4T*7<-Gk8n2vDoUzJpP~)^zkg^byt$Qq)~~k^(#8DEDMA9cCe0r|u9jS4p3;9&3kn+F?{NvuC+_S~F&ENCpnO~XN zvvW-8^l_*>4#K9+6M#t%+WzU2l+iqrPYoj5F&ea}Pn!O{I6yZa|3p`=JOwz7q@fQu zN2I}$*4hmv=ei-JV4Dihe^qQ0XWyEQm4)u^9_(<;1h?+t#&{^wDDyW%7&sR?%&8@u0 zhIICXbMv;|@Q2UXddj}}neK|FiU#$%(#%EOn3-@KH4+llv1 zZ_-mY69oCrpcupXbd&G0-)(z``D-{oQ!9XcF72gSzed;_jHN+yO0YQA3ZK`>llFU6 z+P6ZRg1$E6?za=rwmpRa<71*nHv&W&?P;Rjga6_9;zIuB$%E0#n@s6-I6O3XKX+p* z+uv}CGw27ie4jK@s+FJ+Urn0B8RZ^dI{5p>4>#kj=$2AC{;QFoMbDKeBH|`_4mS`j zZm|HHavfLQB}gJ}4nkMbLoSBlJZz zh^iOIFexW17S0(tDX+`v*4$C_$>JDJw2ef3+Y!V^DbqO5!Q_%&4_O^?ay8bZt$aTkf8eZ}c zBb#zyDEAGc`_}PZQ9df>Ta((X!&Gaujr8_T#^hmQG`*!4*+U9>zI_;^^$S>?u`GsZ zw6fhcFU_m(U1SBnyjfJeADjM52K5^*v*o-4Wbl4C-_N$7XjCt>j~SutOa)u->j$;E zt#l#dA$@XJ5YbCT`rNjXV!TY~W|jn9RxCw&-#rvv6sMC7#V~l8kAxDcPd2@z&uP5*Iu17$DAvU(HWV1ZK0#bwMwM-z5Gv~Ko zqxgg)2$&cH$F+6P`l?7wCbXs*c9-*V)IJM`rD6vWR(~wHl7^U1VS&^7KZ{| zAT~^umYz9{X=`#=*vExfw_27g9sTI!Qf*qAIv>p!hQX&~4C&t9L-VX>k%O@Xy;ysc z=KXj~hX&k4YYulY)IDeQFZ7Xk#)3WkEyhY$l)>8Z8a{YdVM_H0>f1S+WSRo8)Lfs6 zcxG)|+Y0Jy9ZZ#}>OB85k~**Gv-xRfS=V1z)bcf}aHSvF+rLE_=ez}4{ldAXyO`B+ z76wBvF?0P+;asU0b_Y#F!w^{vx-yRJvM189Su;pDb`8&J)3JDxHySq8vx3A*)NO7? z%~sKU|@1v7QY$i$6|KJmA!xH``f9>Sl$ zL+RUQC0g5Mia)>On90_ew8&eNn)>UJvRe@O4B@>grN=mHWdLol$z&q_h+t4uJd2ev~#Und*5rYj^zw zxJ2_E$@o*Sp7{h|j*AeV~)qdvKdyPf+*Fou5ImCZ_#NOtPM(^C)$S?My`ztb7M<>tf zNxHHpZ%)x)_b4h0b)mf{qp)3WJl##6PP3fG(6&-j+S+d>b$uF$Wz#qpX!#FJ>NyA} zBNb9ocBIHv8|cRn5sk<@KvkdD;_2dKY;yE~lYG74VNt<$n|a{9@fDb^984#~TTxLk z20uI=A!Ng3>PZPhgti9xTP`GzOS^FKsjLU+FnW?XAG0;txpA=``QB z){yTbby+YTBDDBOrA(IV|q~$LDk|lyQJF z5?a3qw!5NX)IEzTXPJ?Fu{S08kvwlg z4o`+hR0S?IXi}liZ#bPl&3Bd$pft3d*;YGYqoWh#E^TFN#Dd`V`37cN2IAC&hkQ2n z5B*k4(L)7sdL=!8w9BjTPQjQKHTqJul05?RY|!%h5PXAgvr+GYV1A<<&&&V9-C`&0 zZBHR%wgOA!j4*kE5jIEW;S%2;-t7$~McDzOmgN#6y&^r4ou)bN%o$G?f}&`kY7`4f z*}^0eYJ~oweptl$!0QwEJ#Lf(!f*RS+SU->R^{wMgB3pVcR+7_C;Ez{X#ARqv}~I> z`Y9RW&*H`K8(xH`ZtXbY##L5Ye%ymz0JoAZjGR`^%8tBZ-IE`)5JfZEv3U&j0KcA0%NW#?|@E1JW87#?S4i32D#WE~~u zms7+Ge|}!oA$_wr{oW!8Gsz&lO!0&;T8HH>UxKuTp=8^f&*uNQg>&2u^lY{)<&?cZ z%WVa;PC1C6U8-1BK7jL7P73pEehKAKTC6q9hE361z-;y;v$)?LOg=rCjT~SDl|$pu zt8EA^ZEt*EG*HwMqb$0;We<(3%!NkQeZ<_Br3&MRDbS-F*Z5TF~=t1Y#82VIVLX#i#VK_hM4 z1AII=jP#uq=-b40JWVdgy^<;fm2|TUyGvkw-JJ|V*OF(2HXN3(gO#uW$LpeS`<*@| z+zz59$!NA;aSa<3($2cntx@Idh7%{d*l4rM>}jzx64h+5Q{f7FUj<^Q^9Q`&w3uw= zgK7JRHk7^B#%^bKw7cHL&PA%o%NW3!e)@3m?PB8>DPo$>3{-uVqx1vvq;bHX@3$8) z6`Nv~(>w^OlZGMw;A!TuqJ$lL%5wz?^C-dh1%*3mh$7V0L`_ywA~pR0QZ8Fel78uw zxokZ9I%W?>%TJ>xM^*4^lO3jgcZ9)`mAKAb$k*=fqA}bV%~Cj%ctj@LoaN|qs{(mw zFuJcf3y!@L$AIzX^yytbpJ}R7-z+Ce{G3FgpEPZL`mDr*uPU_bt8 zMq@@9)(%&t;zbKs-&0kXw1=Roh&z-BwOrGVlq0?ikDQbtR z=)dK&C^>B+$@A=IL-iCY9-K;F_8q6Kdskrm%u1vxeMFD-7TPjKiHu*4qC zDW9Tw7p;j7FE1wlW82B?QZP~q_2@txf1h{!q}R@}BHx4eC}NolV>XA`{pUk*Vcra$ zllzMm(|1s3XE1AbpMcB%EU{06dtYL7Y2|1OIwK>^PP)ux^KbNrhqx#88d}ruq$t>Z z^`mxaCrV#`jwUwyLzH?QK6uZz$K^7`d)rufY76s#B+_<^;Va)4x^8_aXtm8kvif^m zeDfNg-wYz<^^weB5{V+bnrZeIaWpyxGLMc!Sfa9+j2 zl4;jqNuCR|qsJ@rsVe^}DGVGWT6Fdstvzu9VIPOmSovQV#=9na`uQUDzynCOYEw%W z=$F|Bx---dRj&G^5NJ%}ho#}?&rGsdzECt^e;xce%kZT`7u)J{7KQ1YhuuDdy0({~ z`NTLFUERWpI6L?IhFDAvRuHDN?uA#K8FU?VF<gcT`AQB;DMDDT4~y873QS~*|w{h~Bl?VU>~|ZZD-UJ&k3zD+|rT?1cG|yI26fhXqC-2=|ze%URJw|j(D2Nl*RuCP8t9Ua zplpvoXm;e|S(-dukWeGxzYQdY88k**2FWW1(ZX@ZY4c@ex_)gZgL0EN%LTYPh`+W(~#IHf3zfSogTa6ha>&^8v zN$V#SPPk1gT~#r0RUNx9K!SBDN76KpXLO)skf>wHEfyc-&wdOKX6Mx}_{?|t!&rj zf9%14X0|PKF8QdPqFFT+JbN>QiW~K5N5^(jx_cZ~#>OBcjdM>29u^*$r!aM?N@P|Y z!IP5;IFheR;crYx|Bw@Z_bagK+QW!9oXaj9kzwj@EZCbNi`m}s=B)ZhD0{G?g7v6p zupgxcXgp;FzvsZoNeA#JRTkcDLxs*V&i*LsCyGvKqbcGSc;Cf_u6XUHdDlY;^5GPq z`j&Qg6wuSp`n(%?7`>mDknyc>3Y0iR&CAarb!jX$f9ara@pa6fyLzNni|EL->-04I z2WeFmkmYTne^Gl#-Ezfq72(ig|WJ(ON9S_B+3U*?D`~c0GeCh71txk^D!ztAFxSLhFRpG*3F@R~K0h7aFI7eYo-NWpO& zTvW_PFh^RbBiBaIqD}r3)6q>I zW!_Mp#64Ou@fQV0{-U7ECurl3Zr+azz{v+U*fJ(g=_hzU>HS+6y__amvn5mXK`uko zuPPRe+f#7udkUYKchKHFiButNLlt)u=*_blvU4EHc+9yT=7(tT?Has_e~!MNK@iuN z#2p<+xsuk0{x+VX$kl7`?OHO_y(UoFX$H?k?jckv!xi^@>`Dp8Fx|6QwbDm4eby6^ z%GD0hjsN(pw0Ajp#R^9KdCqp)E)w~39ysqRED<=b>cxH7-YKK*$M5LgFcne124%YY zuo=?_q#=rX*mk~Cq<9Jb!)Z#3cC|Q)#ug3}Md%--ir@a^rV>iwrQLM=-x<1g!h?2{ zw(vV>IE*f4!{XXND!lAQua_s#AC+}9B4-7?&^Sh|2Rae8DHHwORM;$jUpxO`JH67r zOVfuRr?T~S^!xrcx)2qMdIwW1*mRsdZ?>Upb(2WeJBKEo8_yn?tY@u_f7pmnMKV4x zn*OevM_CIGP^pq7ZT)9O15B6DN68)-O*xFJ3%O96Z~=N{fACw-r_I*Byn3DpQAb=kH-{=ps12zl?Dr7wR76#`6fO)P2oT^vBj*l<{-E zXo!P1i8l?Q4wFUHEAW|;aT?D)Un8~8lJwnbBL$Qxiz4TFk&lrny(&{BE>Gm_H&YC0 z8iOS)9s72@M!eH@Ixek9r*{O9R`nT*4!A~pRCiGf=L95o^%F_0uA`cBQ)&OC`^b{n zLSN;Eh^|I1XD2_(vnPD6={{~7ZT479;Wg*6b-y^y1zm-}Jvw4V<4D5t8t+?1!B9u3r|w*@QL@L!-kmSo|`PKYa+Tg+ME_InL}dy{W&b{Dbu>^ zi8nJF@XLfV#vL0lv||U^z28hHJ=ajr^?j5p`4qY3%Jgr|Oq#)GK0V@6biK!t)@`5 zlm2ks;R)>@kC{YRKl1plPjAob)0wtL+_YEayvXe&K2{4l>Nik6(uSryR>J(&iC9u} z1aEF+vIPB~Y{P|#OsOmg_7lr-;8F`5R2=Eau;1wLdW2;z(nxhN#*cav?t98-|JDa` zPVRE*;$6AWb4Srr8IC>M+~M(?B=rWy-7zQA8|&%Is!Y0---r+O71-8(8rG}N z2+n7~#ETRm{%aS$J}igMX};t8bdyqg?oq#;MYQ3*y)YqWAp(z#r*di~*|0|xoHLX1 z(@e0oQz~t9JMFh{#eUpR)EjQu(4tdTt*@{k|Q5A2N#!~MyO>S zQ%i9dKAsZOUdrCYz%RMRaZVu?U7mD>NAMu)J2(tCs+0Vy`!bfLCDk<>7;tOWj@ura7 zI9HA7nR?_A^BIogn$goTkd~jAf_aq-5$nPEb!PK$<`&P5Onru<-}_T~ViyGN->WTk z#mzB^FdujxPMu;1*yo1O+(TenGVwb(4~L9SA=cOp!iQ7n|NIQLW}Zdj)koO8jC+xi zm2oX?dSaP5V-(uU~mM{^C->>8QZ#8n#osG*|3Q@zEqsD7EJGEAp=AM?J zhFJkfC^y4FgK)fESB^$w4a&BhMV6PcQEAkH{;Rt1l<)H9#3e&2sT~%hT;bf^f@akT zl=I{c(torgbbcR-xL-iO+$D%6{^fOU=b_A&bfDC zIa!SMeKW!Zel``!+~fPmShzm)!Qavx$orJQYtLiM8dwL57)SQU;tXWZ^rQ2SLm)b8 zhfKG1*zCNAB^AG8@_$d`KY1mZV>6hXuB@Z!TKmzYcoIK`B_VRT1I>LJK%4DXQTwwv z+Wzt!IXpW~>StPDswYmSVyZOjz+jkcz>U*?iw(@*Kt&0binjV4OnHqd?X6W$Ehw1QF<{2vS`& zkzQHmvMIr-Eak@;w!mODCfcZx@@!oy;k$L0@O`*hwGY>Qc<1BledK@nf{=-Q%;otK z#0~q2AK~dJERVyoDdIHh^g&3tDdV$l5^|2p;IdvVThi->+oNsRkq@7Q7d{RwbnG!U z(%heIlS^jqQ1+G z>o&5ky+rx1X3*8r_ZS`dgxa47qKDk0Fz@(u;vgle*8Pt(YO6@IkQ-EuCz}E= zLMx39RvC-R#^0e5y}6{y`$^Jf8q~HRUIAe5n#HCpSYepdZO!{YyCs0ivLLYiRSoGD;lmLt=5$Y4qw!?Cud^uiiA_ zj-eAvY&%TnL^y`)_4W1^#^-q6`UG{vzA?StI zbEoxo=HfSjHa6}OnXQP#k?lRW^H+u(7GI*dn_B485bl~E6NiCcQZV+p1@A9EU=nM=ZW$*!{-F>GU#Gz5uM+#d_XSDBM$yU-y%>L~fm&XCBGaM_y1PGzPkz3zDUyd- z_zXJ;yWR1Fro&!504^R!sJE?xI)jT*RAWmwHvBhh2u${x+gMt1-H zxPIU(t-SJ$Mz7mUPWL^LZnuW=EHdd%%NOb~^Py?{J3cNd9m|`(p!-iQ{+bVm@}^-Z zKkdh6gdP@=x2kA*ZV+t0*HXg4WE$^$76C#xY@J`=*`tH_>LyJwu8DY|QN%2@#6;}; zTG8Nrdqr=|htsoxK6J0flOpp~NzFczHg0bqf3XyjUXe*+%YT#Ry*BhJ6;KuU&yx9) zJ$5Z;7g?roZDT4+3c7^1rQ0ZMzl6x?aU@+EcM!L-#)({cpUv{qV_~k?* zBK6jRWRkuVdUjRJOYxR@e!p?T?@MCl)89zZ0=}ngcq0bufWhQn6NbhC%h;aK+X!rH z=T5*c*ie{mp7te)B~}YmY9LUfq%Um|k7p2BF`82a7+TYonRSr!1$R&(Vzk&TycUbnx6DVBcjZPf})Hgk(!}+yj5j;|4 z`&x@S~xm(hll^PnsE|%xO18dp*Uy@`#%b#ZC{$h(3OVIHC z^XSaM2&Nu)0Oxn8QFz1mUY-m^lqe}4_tZ__Bx&X(BELpU*~7_+R12Fw^jCvxNH zm1YS^v{i6d;~(s-j78+q5;k((SQ@u2jD~lv#(k4Zh$>a5xVCrzJ6w3S`n?&;3g?rJ`tNafDD;C&dOVtY^I`8O-~@H)a^{ z7Vmm0vAE$fG%XpW4;m$kJ^!7)t4fNxYNW9zH4PH66X|C3D^@U(XK@?_((8E+quLwr zEnx6idkZmdm*M5o{vdZr)Z54M|9f^QwfCbr>!;G;+NX%~8pDP*+9ANJ2yu?b=wZ}y zdTV}+HYn@BrJxSSLI%;!ur)OLMJP!`^`i>I1&Ccf3fBgC;AQG`oF0~n&HasOZAKWj z6>Wfo^<8Y*yiC}t^n$I}T!rejnQUpM4sLJAM?l0`EWDvap==d>eI3q*?L2`D8J=TV zU?PlDtP)H|Pi3KA@8I5POV2tFk>1!HIJUeM8XKN)PhSqJ=8T#AnkYIxVkYfy+e#nR zx?%qN8FnAtMK8M2XzI#N$R0S4O!Fpo^xSu}cj!@NMl6k2zJPmS!^!b%7)`$8kB||Z z8FTF-`|Xj(OoQJd_Ul5L-8YT9Gw$K_+NI>QEC4wdTCuL<1u_!4z$t?CZRkPT*>Z~6 zGf60mPT;y^1f&K$KvlLL*2Rnkb&*tu3Bi(dX-Fy#BRjo!3oLI@_`KY_b}yWrO{fg=4zad(Fr zjT){_i!u+>^PwRmZ}I>aHptSPb5fLb;e_z3HWXr7OK3>77i`PVV3g}JI+A9KZ*fKx zI_d`=Zz*7U`mQ)|U_09H_rcJs0rx+DL-@M>)Mz}P8rJ#JlmY)y`xzNg^GYj`!qP!( z*b93WE>n#GCdQQXbUCR@SCHwT;cVgWsZ7mkAv<_QAPEzFV*91&rduXPX1o5jw?H?j5Zaafs}faqB*cs|<;BJp0z3T~ss z{jPM~I-1P*_tL0)^j{r68> zG;@WKDCf#xQKfk*4)Hg*{*tMfSaOWE1%*(P&khQ_^cm)BEU=>Lx#0cZHHgj0L)gmx zxD%Iw&l^Y6C$Wb}kItmgQ>y5sLkeBweUPoumXsrLgeEVYMlW~pOmA*H6-^jT_tq_- zoXgwDEE(kV)R+>VE~6!41<-i?ft|Jm_p6LT#id%5$u@GYbS5UK%W#(v_arX!L7JNq z0^FCv{FoM%YlV`%Z8+KY>5$5hE;JSQz|8e0x6`uB^a2=^L0{)5%61%z*VuKki@HMk{+~&^6tKR1}y-?t4%K7LsM~Uj-b~m-0qV=isjokt zC$GeYmPnjso3UVxG>sa+n;xYMrS}%flvu;(WjqgXGkqSJiKo%`4F2zgB^Nv8Kq2X3`_Xp#7Z=o8mTV{Zh-J$9zG9_94cGmNgET1%qe z++iQ(kJ8xbFui9)U!xd3D+(sXnou(Ey+jjq#**u%3`9R1M!Gk-2hw&qJrkcqopx&U zNbf4&AsJDC`Bn;ey`8p4t5RH;1ijd_pVlyY`s`y&`S;>Ule>S6ejKKQ_o`tqU@+Be zcO~oWL$r8XHC@(H6p6(S6dCC&igF5!L`ppSrZP#B)^MiAc^ zqe7b?Mpu6-lf^$7lC)n-Ze#0ndhx!rr1xq|8&n zgMlxFngfl(uWu*Wf2(G)DMtiU>#Lz4bUbWzFCxvc4&vM=+CD9UJ4cQqa>+b=_V=NN zgDEV)rOw>vR}+3M&&Jpn!)eRfV>J4Z8F!QR!>c$gk;B|BzRN8jkJtSFi9|Y0f2KhV z?xShOPjkx1y9$}Pr|6b;8a3;N(cn)_bjAJ^MbExO2PRLY6732!SeFROK0M>rUzu|g z-DqLDjz~3BRx~Q?1zGW&%Ep3x3@==0uDM52WD>@`&I>k^-+vx#ue*z|b@?dPS*8pN z89x>_F@q_dOhv_lP*^^@1leORVImvAcY%T2l^adk%NX5Coqz?;jA`@|BXYU#3diGK z=$P7svz*19ncxQro_CvZbV(|S$nr)*;TN=l}>A@JvDqBqs4V+m&A_kkDnjl1@ zm%T~{<>wtH^<5n_S7gMysL?c6? z>1AKquxcI6QXEAlPxXlWxo2ag6op8gCCPgpB7e$^Hpu?WGoD5zEY$lPjPA>EC&?sAY{-V)cx5QwUx4O6f0*r~`MBI|2ZsfY zbnZqUZM4iIOHE^PSvdePZ>EYOAL@$4S01AyoR50g`Z|_&Oc$zo|EF)nCB6^qhb0#p z>25*_Wvw=#WS2H}y`7(U6FBqpXA328p38!|&#?AWB;7Tm=x(GpxgJ_Z9``5HH}7H8 zW;}xeAD%|{wZrH!BB*iK-axTl6io``q3shSCZk3}4&1A(Gili*&iVL05 z?DWJDWRX6EL{dlL8J0w!EA`08X&Vggjuh=wXeO6uW$1ep&F9RYp&6ylZWSD6hce_? zsn;W+dig~9uvrfKIt|&JelC==sRS8azL3b5qeR~lb~*RB0O^k`;nYb9T9C!Qgq1(-yY3j**7QK2Es~!A67(1p0SC?KzdAkbf-7e*9z>!Fjl3{C> zI#7VI2FzEaGlSA-8lss>YrVXvgEJA%L6(eBKtV zb%f@p9wuFWhBr&TM~uZ!BxOj`g5Exyspe<(R9X75N?lZ{H&8Tm&>QahQD*Z?oXsB& zA1ECC<%@kI%AmF-0BhZ+h)yh(5Dk1iNK{y!E?DNZ31@l>1bKsIgnem7h zEJrr(Rh{`rEfe^;vS9icqOh4*wYw`ObG7m3)~&<7#cVvlvw7_5vz}=}2gi zC&__hFmou+o65A4-Io%I4vwI&vYBvKyNi-tUTCYTV7I?KV?U2~vVi)lD0O;^#!*RF zZ|y>rkJU)&`Ab}QD-Ew*olG;ah;5?>H2vUb-pfs+#cveouXr+caQ2ts9J0T4gc|#7OtBw3VwfBv{FloTwFWg zvOE&Mei~EML0P)$u0rQu1`8^_QV5Ib6ZYs2fDfOI#lCDs*xyMQw{RnD22Q})m1n5m znn;>|$%e)>deGtR<*m`7w6Bv8X-ixn1;Lqay96TFeFd5mz473( zF3lRzhMOq#?12#+6Bt+z@RfzkI-6ci&+}i|%40_nNFv6DPmVoR1}AOsoF_ph|nAE%T!Zz*}Cy(Xw~*9^gl)C;gI9ohT(?Nkf@YZh!7R3_qosOZ5Rzv zMxm(8C>bT=R|qN5rX{JQh=xi;Q3@fIGK-=@5g{dee)snuh&<18UDtUY$L02ukj}lw zf{bEWwfrN(a7I6IxjDe6-R7Vz2F9eJ@*yo);!tcm9t`taFQ zDg{G!jiB`GDAF8nO?wO%@gAKjo$fscowNbSo-&VZ8ucH0+^xfuhgu1{M{W=tT}wn2 z`maSz&M_jX;nP^Gdn7YS5wUnDeVppnX1w~!Ejg@`*?Vf5ZXAEK3rIe=oeX{ zb$+$5e{>S@4r1Kxm_)H*zqt!40)Cmt*pQx4m_M})N)?%yWhg;u#gCyeS{W5qtvEW9 zXK1FWQ^CS8oSD_mG9Iobx1kB#-!X@dUkS&I_(de?rB3-ZPAvU|KN5R~Qrf%-Xj}|N zLv=aUjf#O(SrSutpT@LXTUgBeixf2q&IVWP3u|J!r^91Nrz+CF@KLmB z%vHp_ZxCMdj%w)%N%4dbYIQtRpz2r;vamR4Q&WerAGUQ$??%zAR7Uwrflhq^< zec13q*lAwJ;wITcVO}3`tK3-94xU5B9zb{B4{S2`(AGH^Acl7d2BlqMO72s!+}M}> zeNiWP@}AE2j4q}WISId|Z()}7V>}Vf!9?x_6wRB!hEM;(-t$~yZCD~v;|FZGJ{=FX zjwOeC^Jx9|MX=xc6aS{lkbhkTYZ%#v=wt7({8>39BP1v+BpJhaccnAFgK554jhUrc z?Dy8aS($ZY9^zfy-|Ycz7;JzuZA)(eLr`VIEVqGQo&JGw|Sw1HE*q zr;T+NXn~wBGctEZgeiBJN5sNxR5=Vate9f3v5@_E9y_70j;Vc$1R9*p%4AgV*z+1^ z#9o2UDH(dBzLa%sHNl!!pV;H@1@t2K0!Ccs``mVYXqjukaH%o#ov$VS*1CcI?L5u? zRBN;P$NLamybqmxrdX`~j5f^4g7F|PHli$x)vF$2aZBAvNv)KY4oM|9kB2yTeIzCA zKY~5?Zt`!WFm^LnoxR*N2OWcCkn!^$Zd}}m#asDZNjI7f7aT>C+(-CmnqZReV>UN6 zPN=pXAlknvTF|@gDHPhdvsl}!EG+ymJDk5jxT48B0ZTN57e%MoyXQWnocD>9_d6r3 zJlI2{D&@tYxx>X?dgDo_bw5q3yG9>;%SlD@44waXpBz@N!D!Ak|D(H{mg`PN-a=W5 zs&}GXxAAyb%XbK`Vrh-sc-U{A4L!L5Xq*9-tbUX!y^*EUdI8imD~LW`I7CyMwxMyYfVTu9|ygY_)_{I=v? zcc;&=x2?s8^>%Prc@LhqvXK;4gcY4zNElU3j(4YFO1Kkt#gsFfbC=kpfEM_dtI!rH z3zl+1nXOFDKyLaaf#nym^(SN49_y`a%)+xQuW&EEE>C0!xF2AP@kFSeX%_r1BngEc z4lHvkey@LKIx&@a(6gY^fjWW>9Mpp<)(OgQ>eK9{v`3}Kc6sXX9OLzxj{*vM8a-pc=9|3V#49PNL}+4tv^#U>xG;6}qu zMCd2s)$dy@x@j}Yj>%ES=s1jb;_fszX)5{aExh$j#ncPWDZ@OBNzSpspT`^UDO-Zt zOE}9t-hht(>o^huo3TDD3_HD7WB82(+&rX9 zu`he@`N$>;@aGhybVX+MF9_iwZgfg7SD0Hhf%~KSqP@%toBFiF^(LRs?c!){b1O{= z>Y{I)^Zc|s4KsK)BW>Y%EUdAn>r$B%e&7W4T`*BB@p`Rzn!*eUk1eAIfnVtEmt*{< zR}6x>7(c_vD;@!Zq_fle#KLq=URL1!onNo6lu%77y+fE`=E*?^(}ytpP1r#ZtLRAl|NssoiTuU>UhgRX@u=i}rRqp;B)dy5aI-Hq4Jj{9% z1T@Z-VRgIBS$xP1=9&9LXj^iU)t;Bd&g?gsJf;-plVh={`@Xp3aK2c{vltb(T0QvzWAZ43=Py@k~Ck|1dz zO`BKl$0K1lO}AP|0Y-}`-ftWYP;nxCw<@ZuNMYAyvyql=^DjQ)n8mvc#66ho4kr7Wd)ByM>+^X~>P zrneypu~R?d;wl;XdEb=Y_x?v$_v~d_%`0(r&pt@icr%IeUu^w!9u*_=SuO!O&L6Q*g+d^h{Tl#Dxomg zou=FmqO(mSATcNrRenYA>Su}5Y$4f?h@tF1g}7ldpGGV)r@E1mFmsb4`A-k|Eu{c= zA5`F9MG88jgyZPoqcjbu`iE0tH?l z$9qYBbXA`3>bmpj)bqDA?$9LhvxbE%MphC%yPesf%P;8PuqV_xSVR27I+iJ29*DIo zr10^(I*s#5A=7Q?H0)S7OTM-R8mn8;yS4(-j=o+RHY7o6L?2Ny6OGXM`Q^)M-bS z1&MqPl9=!3)+XB_@b)Oo?W)J;zr)GDK!$!8jX?Lf3T%Ebm1=K+?tiuB-RQBTb3u!C z=BZP2$w$2Ys6nbciJW~u~@T<{-FyL*=+O}?@b7l0YbJD{B1 zA1+e;Fr+6ETQ_rlqTVq?o;pNNcZXtcj}-O~O2XFJ!SG%DnLo!=IZ5yf+wo-@_WucG z$#F8IvpNxB?c<2un9nLNnX%DTQ(5fF9~i-N`TYmUh(DCLkl}n#>{?AS4CFcR(BWwD z;~9bZCHQlN(a>!b|CWL#rC)dSf2JO4x~(&yQMh#+k7wBbJ1YCK(V z41ZSa!}@4nytrY7zW?R36(v^OZ-0Qs=Z2B8x;Lh6I|+rDDz>}tbLKoXj-_4-!0N`A ztjXi4a47AKa5bQj`)$kdCVC*f<-2Sf>;B}Ls7=oE&1lq{5~wzGGhOZ%j~;Z7%r~zQ z@0)#|EtYl!b_OD=S%K!Q=}Ui1bf{5pJ3AEdmM!^}$vpGyQ2cTY&j&VP`LrT9ACH5x z`4|{qyaAc7W9d?G32A;yp`>gPjoLATR+lo`;gTYGh*hmMuu`6^y@I3Mp?4>D*NAdonyrF^k;c z6g3*3Z+T&YYX)*wHlgWO9g00%X~ZmjYTM;W!+Zb0Od<+fCJ#aCI|~GwT*ax=J=k7p zhWx-D_RhtWZF)L{v?pK2%8>2UGBfV}T5%sV}S&HMC+EisOw_~ZZ@=cYr_%XiR!q)p!~YEjxSg9dp| zq-+@ts^QL*H;rbrV}>);H-CX^+j$Icz6;rz+iBgj6~x|n)0@c8SbklGRxNkr|84?2 zj~Y|Q+gx&(o`mKS8NSaw07b7$+%a2>hadOBuh@mu?rot3!}n2PTNEsA2jXDIYV7*m zfRM<3b-6hc48o|!> z$z%`a?|{$g2u%NT7k386!{bCG^e^~h?0YAqywb$7WIasn{Ro?&U<%W)q6?#Aep2yC`WWVJ`uupdunvcRkFA=gib%;(#ZPiP~bEoAAwp&V@*ZBE0@ zf1!BU4!r*G9S7`e$)?nceEnunsLddntic@!SEX>kSsN!-=dyb4VgBO16K7*B==qA1 z_%--A-cGK>J(rs>=T0m+{u~M6-x*taa;c~HINceRg;a&hcrLz^4J^M+2$9%FW8ACS)*I{a2I!Wc& zh@VR?MDt+&_ww*KHdfw-_v=VZI+F&8gug5_UJ6Fbl#uzaiOoAc9;(~Rpt-aV^L=mQ z?#>qE)p%l7P8{TPwJC#hX00v|HFQK$QU9;Fw(t?6SEfR4WGyoCRcNWW?*^VGrw_^NPB1I+d45;+c16DfaHCrmF$}`a;NJ3eI^h>4af#XgT zu3L$e&TR)Zmx`@o=yH}ATSr-nc*TsV<`OtMd&>ch{m@c`j;x4zA-JHqh zjWzaF(Ed{c+gr(qTkHz-W?_P-GC8)Lg%Be_Q78CbqI>`ixL$-yvoE4V#~!DGPV#4E zAIf~9LGKL2RJEA*J^sZ|$oT;3yrNB0r|QG~{(?(Z_FHN{-#Bwf2bRd zHfkhUU5B@I525$@7^;q3Veu36VW*MD0vZ+zllCjKngQ;tb*{ihB)wzbxhJ+zULMh( z`0r2OHE^z(fNz(!K=PwDdZ*oEE+0m*!jxLUe|Uz-qT5{9=29$F?+z4MPEQg&F`Z&; zvgE4KVn5E-cs%b3JZ9$GJ`06+ zH1I?168q|t$DrZvYpnm#p4FDsPcHB_xwviQD(njda9M} z>NknWS~vxG5{zDHS zF+P}cZ!WMWhULtnOp(<*vl9Ny9>vry{uJ5zcZz;Loy^SpJP>A-cMCqAxv&~th+W=+ z&{@7raII_*?fLdZRQB#8totk?J*N@$cuzePy3Sz`-;o`AcUv^#MU1h=!%AbR%(ra# z(-mx%tb@>A{Ym&8o6oK{?qkM5vbZ3h$QD{O3(ptGGc2$WB;w>nGa?J{SucjqYM!i$ ze^*Sqrb<$$8?n(Z9#@8r!uv5nXy*KX-M2ODGJn5F)DLBY#Ct@NYcxbN9EalUk_@Ct zr18#KA9ViBVx#4@2sXPTsiDT7+Rb@R$J3U*bPW`|v(E`WMx&YW>wkj%!QY~NQ#9D$ z^^T%(1zpC`xt+p-u6_c^#fqdBr?ct#$;|$ix8SlpjLpw+5hk`=7WGq%6>R%P3g)9r zMGu==*fV>crLgmZ|IchFWuC`|xT7$a8wLHaLKb#mFME)b#NKXjWwtl=u{Ui2Y>juS zP*8neG(}`7x|TiMIJeNkNLT9?6L#)q3jLQe(;aJsxCsWLf4fA+`;N*B_dRrstB;x+ zduR>Cr9-PxEa)RGMURGbDN=$)FgE>DU@d1)i+nRa8;5b1!15b&1%uULA~W}|xWR^y zKli?D%c*C}9k#G_t~H{A4aWtKFBe5y+boO&eviPW-We##G(>%H8~bA=W&>7b3HlBJ zqW0%cjhBW$Hg-}Sj21&fD0|his2iR_w`!bdw=hw3aOGHGLC7;v>)gE}>FiFHZ5zR+ zDpZJGe%U0v*dr?vZHhB4t_%^32VN0XXHPb_X_TaS{vLtZO{@cDC_9pEx6T zk+SIYhhn2G-YbnJkE>*r5&=T7j)Jgyk(cnu^|Wwv%L!r6Aq)DfJ&LBPJCma20bDip z$HfpQoRzG@Y=aCKQ6oaX=ixy^A>yLkv81aO37x^XDQ%3}i}h$|_=W|&{pk3e3mB+( z7%o>&V#z`SG8-vL&PsA*pmY=s8aYVd>>$bOd$IphFfLw)7@}0?g6z(HeDE=V6%bSgh=0fbo@t%+@qm z@xDjm++=Lm2}W$`HgpF_;@`Ea$n@L);E&)tWm(~c{ij&*BkO)ewg5R0wFzDnTPcq=J~yc{Z5sI_sB|Se8?5w zYCo_~tN9Goa+zgz&PAy1ZZt~x;7j%qHffm>Gt9JPhyL~8qERV6;R*)Y@EpH!9V%aU zBW?9Ox<1F921;+DFeiCB#{IK5&t1nX=V*jW#ltWv2@(D8u@hU;narqN%+xNAbYv@O zT<%30Yi0!N$9+xVx zMbLWejZw4=UT=Rer$O_eP%{pvNA|~za|aPKVhP5)I)M+zdB4=!3Hcx0(7jy&o*k1p zQ^*wJ((zdNZ7SUDKC>M^|3Ky&ciu-jV?^Kpn118g=O1y1+c|;GdR)Q%563ZX?L@j? zs7pQ5ROn}142B{C@Aj*}eOiAUEJ$Z(eqLv`EFK?gEs?ylmUR_KQI+{OtmEIjx35pY z8_qzn-Kd7I73-0#{}PI8`q8t#2eExoGju0>L$u^EJhF^N&oyU6Ir5vQkuuhHPsTyz zSX$^TO$CvXyz5edap?`{wCAkJ`@_k3%4vAqu}06HkI+jhLHmb&m@c%UdEQZYXl21` zfCN*Iwh@MV{1irRzR0^{QK;J-kF&E|x&QYt+tHOH)X%ZR_&lCB9jt>6%Zt2c5r_7! zV|KR1n4Q zO;;tw7AOAE>~}5XCN-TI*Hkf+;UCy^&K=DiUWMb55vUNqrtl64vHrMj>aw{m)I2=L zvSh!pU55wIg75ki>pYs4c?StyAz5sRy$j@SWU*gyDlE4=5t9st(#msXm~wU~-QKyL z-f9$6(W$eX@4A?lc7B9XSrXz_ZlpY(WA!?;7O6j4skJ+ZwuJFp+x-(%{c|~$>>opm z=J;aUiO;Oxy5lV3#4l*a^De3RP^2HUr7zvq-7ONAZQ}JOxMG}>sn04yg$o&peh#gEb-@n#q?_OZr%&M zLM63fwCVeFXbqI5AG3Yv)%Qe_x=}zsW1dsB$5}dhGM*-m38g-B2eJ*%jtlA9hQfjQ z@iZj-Cw1_S+oOM0^t5>n)u-)+`{fzX3w+1)E!vpYPiJw?Y(?>6gH{?oQ5M}7%Gi5y z#+2DIC>|FG|6NG1!T@(?ls!*8?mmrc3B!eT?B-1g3vg^)}Eoay7 z*nNSD*jo~eh}IRY#<|NMSj2e=3|>DBnmV2+sF?-z#hf92-Gdeu51?V4?MN8NZ?ik@ zC+uxct;lIrL$hOFWXLT3bQp>1;xGa5${nw+)6xtW^JDC${4yZw`j3bI#c5?rVJZx)*(gc3V5V6-t$n14=A zurJ;uynnb8%w`@|sDNpIlA)gNB?f?y0DM@G7Ch~W78S{Jjk_8^pWrYgv zY>tnv(0}@KLA!K63%WRx?O7lXpYQz7nWskGxmvXE*%RT$cAm+o|00;0ei4SH9>Zg6 z-rF;1L(d@-+O4*k?mm4;%Z>Mu@<%0-@|{K>_XpDVg;yzK`(|`{45jXoU*W1Y5YHZu z#Or%JLpg=B$ewN|vu=5@Y=jG{?quOo#bUmTDS+Vh4~EBbu_E}4P%z;TYx}E+Xz6+k z_m-!3xihHxxi<6b*veu|I6qj!mBRV%cxvxoVe0Hbq;0$l6Xs{L6%)3>d1Drw_bXz| zcLO%IuM!+yYGce@eKv9Qb}DdLB$iQ_#b(`-z`5)ng5QS{Y~BA8+zC#CnG_x9(*?7~ zGGx^#%M5Qe!JYRflg4l3{ee`nPl};h6%`W8^rP&zk8q{Zi24saK>baR@g0snCC5C# zGIWaWZFWPBdN}R|9>$8E5~}4~m=u#s)Nx)5YHR&qZ63--udKi#5B`ntj%U`+h`2+k zoW1T>!BXzrrvx<(aXIIF*?PT!?GJULx#P&DCl9fO?U1=TnrzbDsq$3>oo*gZ%N{5) z^HgbU^KHRQ?X~Pyl^fe|*qZqsOy@W28i?eEP@t(8Z*;gvmRV!`g$`ED88-GtoSUsK zEgm#2k>>pkM_Yp*W%C>I`6PWvW^_TmWf)mK>c;hD=YZfo^lGO(RV;rei1qWCqsd{G zxIt08?dW*%t(slp?)kPjIOIGMUXQ_E$2qjG>M)(zxQ`C}{e^Y0QnclDIVr`pV=Xv; z-nNcxQe#+F;C$9q#NFbeK)N!4Xw9lGY*S}nYCrrCt8|y(XxRfQl~56vm84Nl*F>}p zS0zn1O&X&5hC;8Nq-lEc^g>0F+1Za{R?)lIi>_(R?z9}1CGhUdx>qd9I|jSHo<*Pg z2K4#NEZSjq6Ez2onNsNinD&`O?u7yL=gCC+^u&d{8vF1&Dx*cbm+PAs3FD+>=-hN= zsXTkT<4_3e=)S>zUfjuqU$5Al--D28v4YjB_XV3}4q3w-G&K8TVC^_08B4P@&Z&Zv zt+{b46H!w|5C%BzXZ@u343Xo+%G+KF7WYDE6Bzj)89-&9Cen-01kU4FkHqPRV3#%* zMKeP2{Any*|NIjL<7$vO(nqWj?k1K#VknkZvF81^E@4~eHDfzWrmW1TycawJ#>#K$ z#rj72QRYQ$)0&Mv2bqehjf#wCYBZq5{3~`T>A_BM3|qQ}d){Zpvz(G6s z@+gqaYgcBu&GO>IdGE=~^(Y;!Z$PA@5^ff47uKrCQkzL6^51c9UQ834(q{#0&&U%p zZ+s(dlS2BI%)M4WuF=JcNi_BE30%ukfvNmJQj^-o{mv!S)-;oQL{zBJV=t8mf593q zv#diI?CbYoLX{vKZka($t zw77(4659ql^3Kf_cE{6!HMJHZrd9^5V-V*pSYz>k5;l2{G&7od3;vl|?DcaWrV_jY zvKn)7>g+x|cOOQV=6G|?ry`oe~!Bpo4yLGE~^RS#!GRx<5tXEI2V;)KcXkI5Iz38vDP3Bi6dUJ;>`u@ z+|5Psymx^ujCU6P`b?n7HW{d`od{z`qP6wE;IUva)NCiyj@#T1_Sh7boA=Y|H}!ZP zs)EoP##EA}LBl=2WngFB9jZrGjRDPKL&kcYd4?RGK3HeoRQmmX#dZdkL-XDXx{{sI3! z730#0L^h)(oEh0=GbP_!Wc#lR3NOUWm(S@-Eca1G-4(hr$d_*CFQg?|N6>#YX9FCs zqW9y=s9c;(e?zy>#``rWPWwRbPnJ`|knQwnpb{y(GNP<0-XxiN95rG-^M{FOk%B8N zT5CgwgQUn+I~n8GJ;SLv&J-s4ODDXGXxZuIBvIl-RjvPFOZRt?_Pr-07Zgj`S2ki| z?oz~THbB~pHs)59f$Mcf$gmm7VtHP;kH9@SzvrM4OQF__(Qs5>g%eAC;5=<0%rcX3`C$#y8h;iFo{O-lE`+7K*wB^FJXd%- zkSSc;z)~vr3Ih+wFr|@FoX`Cf-`h&j{jx=HowtoeHVmWgI4wHOd%~AA6i)gXYY+SvyP3g*!R85;P0=)Je+6rpTIzjh)f8uH58Q4~>fJj_K zam#w>gwu8M-4aX+@;7PYn)mdVGoQv!IK`fMln8EF)xyJMZH%n_g)~{dli+(F7TK3n z@(p2pIRPIQuEE$p{`7cr5^bFKg*0b8z?0y8xM07My;-DAnplB*OO^_5jW35SM`L!O#==AT*zM1db z|1M!BQkqPqv4n2SIZE@sohNG(e{?$R!6V+c>x@XHvfWPPsd^d1B@#(bq)g%kZ;>O+eyxy~%;~hc4X-jCB0q>B0%b_*h6RF+! zIZF^#;O(w)~sV@j9W(&42&rH)Hcp{kH_5vOP)tE=5zO4n(NjL=^Q6K(RCM3uuu}WN%+zI9UIAN zYc(4FTfr_l@a+8~o`1~T2@AoCxd;K#;A0j zNh%Dai+TKx{&X6dYpGFm^lNOn#k&sHAr$}OHhtPMoEkKxscw}b71WG?Nc9L)yi&mY zXO_{|ohBr;HXWxv?jZxICm8j&m(7e`O{ouVP`tx7Iz3dL#BMi{xrQ>aOTOFu{rfD^A4x-P_IeUM`%NPaeUbX% zD7t@~z)kCqcz6CQuA1+_Lyxme+Det#E|g+-ga5G8lkcz(R_#Kb{RTGSJ!4)!T-lL! zXJiUin3Xn#ex-1}c!oTaHGC)B-M*8&Orz1VgK%xkM0z~$GA1r4#fA_0DD=+3()>Oc zayfvcyZOy{;1UY42CWQffRu+H?5YE)qA7+pm`tVEB+kIn>O&WtdNJz*XFO-0Aemi- zRP&BI&6Hx`)9WH?x;21=B}2%2{Q#Q&sDtE1ab(Y#FKWJDg_mrPsMh|mus%kWtR8#5sCErR!75pv@sxwf z-V4mBj}Cm970dMF?5Zyr&{3a?HYd)_@a|y=4i0pzd?&rXWJNQ6uR-nK!I)C^g#B^+ z1eK*fapuKC{w*!fq+VGwr#J`pxyOq19X3)9XHz>w*pYJMI&xZDeZ5a5X zjb{X3vD(D*h}iW7y z(QRi2(Jy_7Qpb}`QWeaH@%&Lk27C1566+jh!^}AA?P#?c6{>TdWlKKJ)>>hF!UMLr z!-)nb{6<>s0MbamL1(0u#aU z=^x-wb_)R+k@$B&32hS!Fs;i2RT8t2Qs6*mSB<6!wc(WfG8{`!JVt}wHq4Y(rm@RT z!@-K@DwVhsN|}2Y%(4;F{}oJlzpy#ditPVRrnMg9DR=uNOdQZl&DjS?yI&7HyZIhE zKMQjzlZ4FjG9m3)gmC(Vqfp;)fKpN5UakaxP!(u`wtJxTXrRS^G{w;jR{x|aT&iZ^qW*^;Dt#+p7e0}NYkw8I z>pluz+Iu-GF^G;_n?|3PrQ&AqWK6K_5hVI3W4K-`+p=yOqJ>b-?U+rQD#lSuuq3S= z){M0O)N#qQ8f))8V-xFM2%oDTvFMRX~|{bogeeJ9gLqoHJY zYB@SuoS{|l8c7SdzxA;??X&%go4Y+}f>akI{wkAp;6xfZaz3pwSV7-T29d72I@~Lh zv3KoYQk4xyh`t!^e?|E0vW2c!J5#^Ue9tX$97%b)BxCGL@gga@FZ}_lPp!qY#zlNq zRi&dPTgf8s3A-KT%Mt_nz@_01o8#IStD>H>C42{(ms*7O4TJGZ%@|)z|AU#qVf+e? z!bWL|JDeqWB)eS}z+B}HwAVsjyl||R00O+b_Vdqm5oOIGc%ZE@lS+$InsOe(* z+Lvh8se$pBBy?RmhX-$!Fxn|laIx--mB&k1-r6;I6s3r7Ju)yH&t0q%_l2Sqk z1bSjL8qT!=P^^%q%gHAE2E;w=`;M{@y+!ojMGN|2%R9dn|7h%szjUynnF5w-i}#c^ z@Sv6+Mfo|B?=Vlgf7*%m8lR*wPWcr6!j(=%I0_3Kec6orTOgk-OUpQ8BX{)!tefme z8V%N*eZ7o6rT0a^spELyzL0`(=cvQAmL9nrCieqo6jmY7@;Pc0={ktM=d7Xrwufj$Z!#TAN#y(D$E0*b zPTbxS%}RVtVV;qWnbbBLY>hK)fN!HTy0ke{((kqo1I$$Y9QP=;W3_~bXce|ka)`Fe-*O!J3v!v=3 zSMhIXD|?W#fii+ml6plV#a3KEAn$%-lROnz{>0(`8gY?xcw%!8QnceZ`n2^N`qlKq ztuRHDPX+ezeeAtzS$bgKhzs?~G;p9UeXL$WujWo6g`xzSe*ZLi*toM-cP2t>RW2~h zf!S>vgqQc+ad9I5-kxSk?XNxPlzkLkIp9som-IgVQ5N^a8W5l zF#2hLkhVkg-enc-T``JuHfKS;{W@&^N^<7vO4^p`LT4@uRHq-w{FhlE?#pHrUaLfB zt2}4H|7PprCPBZ~5oYU_5_4*VcLJpGkg`yLglGZLxutt_N9Q7 zo9w?c$a3)UNrx4aRk^6iX4r9CD7E`m|UDEj8LN`0|aqrq$$oZ?%tY_o+ z@24c)<9%#H5oeLx#UcMtf7+#DN1>Nh$^HcI)%gcQ>bws%$K=!Be&%#JJ|Zhc_z zZQh4|Tf$(p*bFf%m&2xfHU+OUp<}*oBq_0;4c6h^ljZ{au2!K{Jv;H0`M`D*cUFi) zpxPUU2U#u{o2EwnOBa)u*G%?B?LYSFVI52SoI*bzt)-Fi%5>NE71}dC!pld7a@H7# zN1j#@Z<-JBa_+}uG=(5tbR z;9H!J`_c8dBzB}72_vWve^#u$rcD|OH^^w_P5NWWZ^r>QF?Z7hYFx3AGfT^1-hVu; zbFAR`!3X%-{RP_-4#VL35Sp-sNcDIfdJkNKRL}&P^0b$2y8MbwFO$MhDOGwY=4|aH z$7$m6*VqtU1-BRKblWZ)^UMia_Q@y;xrV{*!>GP}I87S#hpmsdMB9cyo*}u4^u)^ zHMM-)$L@a$W`XzCvg>u6%RA#BDs0Zds38T@2Sz~RQUrE5)MKa41qArqgzT5CWY@0- z3%?e_O@0);_l;6G2MwGIQ zAA>RfxB~ix>f+EFFO;mX!E)gdw(mUzi~dDe+qFtOI7ME3-$a}HZQbz4PXVu5q!`WK zNVoHP(Hp9WvdzQjL}3^mvI?S!H8F6xtx3U$UqK~qve@vEs_t<4I+JH^pDe zq*rFwDgFRw?USO|rz@FGHb|1YQz=VLDW;jfj*(M)Fum&;Nul2^(TC)PVpaDFcIwYT zQFmstC`tDzUEKSd=J`FLk=~X#G*8U>jlL@!G~7p7&kgC~g#u1p#u0_jpq zJ@qUZK#^~Ul6`_cnV8r^yW0d;6I8Hv#$7?7<}5qf^bKlX)X8I(1Lb}(p($m%*}Kmd zA(14ebN~L3lqK)xF4{oHz9rJnl~*bF>PyN|6(ilYp51)x&Thr|u#&!S(3CDMzUov; zJ64^dpErx3Ixm!-^X~B?VLN3;NFlFZUo!KvrnL=n;;4tGsif)`Z2yXRPH7<1Hz^U- z&*;mhq(4NB|c@X0D;Z!%tC%Y>Vh=a)g=;$xKm zK}2IWE~1%yhavx9Kb=c`jZ1U#S--(5!m)D&U=q*J;nsyj^V{6>#h*)W&eOlC7D$@< z2Ihw(sO$A6axs}I9&x2cTyj1GE^}pQ!MBsR=#fDQ^RwxSc`_-a%F=qC{l1B0=5KZ# zO}Br*)U*>zA6wGEK{ILlA89Hy3P4$DI~%1uoo>Z2ai3^QvD+6Jv2?REW>;Lng3DL1 zFYhJAs+x&+jnEVO8AoD&o+P{XZ@p+>_E7PiXOiN;O9>Q^CL=x=uONQ9@H3qsG9ER` z8xR@IbFQ0{DbqWjCWwP+zo8O(q&o4bR3NQ8D)>420#??m;%($bitk!Vb)UXtnxhCK z=BA)jV+AB9)l>fNL1HUEMX{z@7xVJk4L{2%h&^X2KI|S%pHp06F*llIcm(Yse^$GR zZnG89tr&e|HaY)H7c2cu5Px4bPTXNpOwmT>w9|esPkR0!|)R2FvS@p`8U|%z4k} zzx68clM|xov*Snd?ngRpmwXiwKA06cwGv!7R9ITebHB!CM&=&}^yj-}a!6G!FT%3>0&O&{Y0!$Ms z#RbcK=+xgVtZq%M(gTPTItN$PPQt77{p3Nb z$#u8_fA6U{Y@TIpf>1rDc7;EC9igT1xsz?@oSI{Hz3BJ_SCp%LwTGi8n8{}lECE)?q z1tyTK?M%8`^A}e}3eyy8Q?k`>q*~9}*wmwljmQ4M1cL|UD;!JHvVY=;UqYnutQIq_ z>(IC#(RlHnH_jOG43*1UDC8Goeb&vULXjJoZn6#^TK$EwHf%3veIB(KPm(k}V6G)VrNEuV;VeTCIVOx$z`!s30k_tVtHz#cuq#I!o6w2m3=59K4eX1Aj z*PlRAJ;r2NGMy&&xRIE|ERxVRp{%v@;H}X}oNc@Wr>@eZCv2B9-NcP1o;Aea_$ZV+ zT8~Zr;&iJifO%2_Fz>!59k3K58Kvn|IieP?U;K&Y`9pYw{>6Un7Th)Q61cj1 zp{wB*Ecc9{h6Hb#vpJZiR_{P*<`HUOy{P)i3-qd^jD(~Ql6dV6);o%)khC8tJ;9hp zbu6W=>x`(oJpvaUpN2O-q@dR(MRI!j6Mvt(itD_Y*IsHNt%~xd{n`EC(5#16Jvx~C zEd#RyWN1N&fc#|4>Gnc<(lyninHFw1C>)1o-_r3QqYdvZ7=@N)nPA2o^Y-sJ_Pt`> zSUtA)SW<;sYi6NS$W3}$F^eWx8&lpjb-Hbth-_quN*~tH;tdsKg0bw=Z$aNweXAZA4ncm zh_@;u*(|=F-S5q@PG=T6KX3r);REP)K8I>@<@kjUTOeC5mWwy=2mCJ|U*3|T^6P1| zO~we{r8RQ4lOA%u3)oxz{v^z(&P0bV%!#d4jLiw_sLXUB&5WtQq%&9XrLs3BL`GrI z8)rC^cndT;XJMt{GRm2D8%#x&Fh{I_?M{^G<(m!&+j$$}$Nhw^6H#dVLW17wJOWwU zY-svggYjw-m~}r8g>we+Df4yc=dt`~WCCND$ohTHNscSbS_Z3f+(9 z3@b+Blee;DvLYNVzx)E<88fuIvJivcgyH_XMrczYMW@z=V3uMEj@7tmV3V>{z?VgX zRgpgKX^zKzj<2xFau4fk&8O5o+sQ9sGI!PgoZ$Pva&8Dr@La4i9&#K440_L%TfT+G z0nX@VA`L3_54bywKeb%!BE-6+ftEuc+&Yv)U5Q_@rQsN6jem@%4_`soXYW*?LarmF&Ak28^1sg4`(ZM{Cvt>K&u_0REu78tn{pP`^EfeA0 z7H-FvHRJG_r5T<)5etP@Mj)bSgXSAdD9pbXmOV@rG?f~2zdFug_`bvB>S{0eeR2t> zp?DUOGUdr~qPlAwnlR|=u+1%RQ0&-n+x zcJakODk${Z1n@tW1)t{_V(*t>PBr8ML?^dm+f4~PzRU|-$F9V`M_*CW@i>yUk))!P zeeka)3EOW6kzE{f6<1EA(QD7(RHYO$wb@K*m68;rz;^j6F(|r}anXt*z;)m<7z%0O z!`f=}y!#NV(`Hb4%1c;Vm5-lA>`6`h5-Qzw$9ol*@yoadSfx@7GhUsB&Ic@OSj5p- z$^G>BK4SoJMUcB{7CuQh4gbzYfvE%Yd}ax;Ou;-{^;C{t&6j35$`fF9=nVL1wQ(I! zCxTZ!`%a7>f<;eOgV5_FxHJ%o#Y6A0+ryBkZU%|I89@^Nokaut25O#ML*dg$^I!Oh zoZ@>YY;So@{M<@`iIgh1d`rU6=aG2C<2oMt7mp@sL!7tr5)c+yiZV*^^tdjbO3nq* zr}|+??(&7x>~m`Br9>ayM99-$o>lcwn!ooQ4Kziy))bz9-v8WCVu?0hYmeD zNe^uY>G?r@NEo#p><7-n7{hFEx^@j3rOYt(qY>zgivZ^^#wD7e$+&C&7=7{%7I$tY zlLOmPGdTatS>%}J~{K5ATA6r;S!j3 zMZ+|+MDRB3gT>FAaY!u^Gri7WTSzV{K90xg@1sdKcN4C-%4}6WJ25ruInM1#K^@fq zEEO}Lxo!0*+MkP;U#OG9svkJITAH>?Z>Ed7Sr{~}7ut2w$T8N8JZEUok;rt~lCXu$ zohP$Ag%RbcEhbC*3FKM%5Rb)dLM3+Jc!(h+Ho}VWnpLTHLnby)-GXXtZWwm^9ZdCo z4IZCsz+%-tx_wfZPK=YFnws0(Bn?fN{@xb$>=noQF|)z3#|P9$&ZRvoFXD^bSWuUG zLtSDCv{O`-l6JVCR~x+});6(_Q&u`Y@pGpFtLE?X+uDM#0Q3qX0S6})({2eQ7s z<>l93<0tyM^D3(+(ZsC|^lG^d4J4L;kW&CiyfQ=a(^d3UVh5e<6(-{%c85)2?t$-% z(d}mdsGQ(o$+keo2e}J!nd4ZuLJUG~ucC7*Td6nl2PO~fMO`BWJg2XQW3p<|9>geO zCG!DD%`&WCC}w!ozlhhHdjoE1`9jpXRbaGHls3A=VpXUXy6(Ne9F(K+>9@-u)3A*9 zuao3m^oqz(_Zb5J04I75;mIlcDY$DTJydvz<^>!0gsl_#4g*ErYOWMlE2ayIqzS*Y z-(b;5YkczNJpNpfgH4Nk(D|4Mw(Viwl-nz5zqu^gW)5T0sEhbf%m6=L(}85G4-hI| z3Dc(Bru)Li6cTY6cMJ;1mGxbI(Mi0&aw4trFGLH6o#@h~L+gHu({HoaC{YG9c@Ixb zBgWI3{-1C+NE=lj*y1Dinbe(Mj$f+Yd$k*UP@cpr~n55udYwc%@qChWQX8fF(R z;-~-pKr>3;(5OQj@Z6U{NI$HOFLZVJ4ZU4t^Y19V-#C^pX7`v2`x+?q@EzK-Es}CD zhOG4~Vd*YO7<1DGW@}GnuIrO{HpmvUKL#_e%@kbn?<;)Px`wiU8t@M5RSXZjqzgCp z)4J^flF=GVnuFmO;USD2^{>I@+PGXzJ=3~vY{H-Vb@vFsep|5fG8daRLVH&pQ?8LVs8n|iV7ckK&6>LQln6PsN zxBAR7c-}Z0iXTdW!TKY}Pg24|bA@n`{TbT4ZXIdoo05;Y2TqzSi%kk5kU!-asyN(6 z<#}geyj>R-ZDRiU5yx=vtf{q(Bw-; zGP5w}k23k${lXb1W|Lym9b6^83a%YHONMvM=~dGhs-AKRYEH$%;MCvT3(p9gD_aY{ zUp?o>w;x68qiy&hlKCk!50Zy>3Mpwslc7ow2B_Pj4D0ef|5HRy#zm5(uQGkTHv(l} zUtyl|mEta^aR)0KL(eWX;iGdilkm0q#w>@ zP-CHmF+r7Jc1w;c)K>*v->aZ99E?>XjcLit4OF&14WpMnME~bM*n4C>x`?LY*Vxan zr|mBuT)vwUra3XL?{(Zhm`?`be;E({8D=of_lBZO@^Va~zH`UPEmnx!({iZu;9pW3 zd`)rb{g`QY3&JFiLc{8dROYS2cl@&A52jS&iwGvozJHu9_-vsY*^AI1+LHVEcz`pT zw30cQLU7}O8K~ow2Ga9YaEeqiAF<5#Nv9j+G{Jz6_nXHTh%1uwM*%L7ud%WqTpL;kke)NaA%R7cb>H`5DueEo@&(>9?3B!in< zC}oOGX8W!loUGiAA2;lvin|^(K~fX4vg7b$g$;d5NWyb|Pci(~Ve}SZd#uhXN*m6k z;_@&W7x{;J8@kD5o)Nv4f6n&&@zm7w2E8YZC(oiI^kGgVWiR1qvuF+2HEEEX5o18U zyTr*#vK%tY9LFqg<+@{IdB5G$_~_w4>YJ)cmEW)7?37!ubN@SP?O(`8{hh{N)fT0{ zDsQOK^CP_(HI7eS_?S$TPm#&1NN6&=3|9&qp!)kl^zxdDPAgO};_YZkm|l-s(Po%2 z5J{;`fVBB_#^Z>q&ESKxVO07PFp$3n6JMW%!4`3nXgWimz7ElG{xII1 z^cUAnl%uD4#nkQ~&Zp|i@(uOfAU&9k+9esRZ|p=*b3&n}3p0{;P*uJY zEiF4p5^qW=M&kv}_M1&7$dwkxB;q@no7n3yo@RS!aR;uoz|ajfl4`t#n{8MhXef?; zu;1U9PI2ZJJ5D8D!Eo=B0Zl(zN?SZQau{n%{{^x4uT>$gk)KFLjD;io-~-Am(Wjyu zb=q50g5z(Nqm*4929DgyeowQ|dKvrs!+(G)`}cVAON)L*sgcw7d6d^4M^^jA>5`%p zom6n7t2f-CGyf1ehwW!GywNbDB@c#@g3x68F9=#Dz|V|%mf)Ic)5l z@qljDTVc-x4y~i?QL1?hRx1=ilHmnxX-ucJkJ`}0N0Vg}ho14^D0ObnoO!+dCRKuA6B;W?dq-J7*whyj8 zTY(#I8q?1P#y^}WOrO-CiC!2ELs84Rh*l;4%wQ z`lIv@eS*SK5Mau(nJZ|s6`=c($rRpZO|PwH|3AYZ0!50xT-g>J9Z;;$Z6`2Bl>orY&Zb>Y!(p*ZdLt^~*uG{3> zz75+x?!%s!$?&X-Wf&d{@wehN`GY;X@qx$#w38fxLy~16t@WODMRM@!Wmjwn{)}}t zD``wGJ1bY*!+~T&5-oj!x6N|Ue$yE&VKX8D%iNXhO~yCp%qZH?n(Tk5llM+}Qe+vR z%ca?L!bg_hPow!Rrxm=@)H-nOo{Ot5GdGXVc`_D1MuVfzGHG!nSoMX#5g{*FZ~qlN znQL~_?kIAp(xP(ypiQm^*p(^XZyMNGXi1bsTZ8-VMo}#Cl<^XNx`8E5i~M# z!NfXKbg(bRhl^XWd{r~dlaCOLGm$k==xQYU&qXXlFo=^+upG_%B#eQj=zpP!R^O>3 z$rUWKnr?+muHss?2v3 z>`cXrCgIl0*7#398Q)Jl%>2I7d7tmB^YXw172h8MsgwXg{qHrL_q?wZ8)`(QqNDNl zm(}q5+%azD?P!>_YY*5|J``Bolj4f^zQrG6EMJl_jiUdm@DptR(Wl>+C}rh0#OyYh zrEkH!UAORDj~qHo_kc}Nh4{Hok)G$oQDsmGj+?4QK6_@+pQ~w@U||5wai0xt91B1# ziFmlreD7Y#AnudTuRGzgFdM%xb|=;&D&T)$4U<3w*`ZoqfoxU zEiQ-lh*x6V@iDkXES=k1aEbHIy1?!5-cRlqYRF_DpRP(Xug--=QhJz9($S1@yw(zb z9^Z`l#`b*p4?TJk{XLJk`qF zrPE|MvZ0Q%xcnHdFm`(8jwY;GSBIz5IvCUJ5DreBfKo$Aq}s`H#yK%K`D8n(n_TB> z&Q36_VfUhy#`AH68Dr`(cC}mEMtsV9Z~^nOaQCf`AlzBX#R{aU^Nb9C;fgJP^{gcI z<*tDZ%0=9Zk!o}_JRe2Fk-m=2Wo)~C=F{vY^@}fX zv!O?u3b1K!E2n#haZf!{L9=ZcG|cj+MC)%T?)DFdVvKOy{SX-Cq{g`h_@GhmLA-pp z2v6+0N$1Bdrp3aRI93or3r#9fM`}%TA~qxCSE^X5r(`DfF_@fb{V{JU;a=P-r9Gooquo z+Sj>dj*f!Atu|;pbBMFi@0;r#96?RZK^W-!h3oXNqoZA7G+uZDcK(>d2adAl*N2bf zU&Q@~_Z%*v;hAPQ?WImN5?AS}>sbE0WdZ8E-9}y#PbgPGgTJ~!g3mTSM~|y!qSx~; z5GgmnDR=M16&nwudQLp%ef`3%mzL!o2Nny~ftCR8BMP`sjQ~1wAv^ z!yL9Np)%|a1}w~nO|gxFw+X)7sAdCNqqCT-7k$8oBF4BvuAQ@RT?CG?6KLSjQWU;0 z0`z@-$gE@>-#NOL23_6gey$H)+Ac>gh6-@uF%Rxn<`~YwN`}3I#kst`Ac5iYC@>$- zycSW148F~gMlCj5F_Z6sn#M5L_TODJ$kHdRcluoMu?pzl@&zv?za^8U+vug2D+Z1G zhhy)1()+|n`e2ew<6;9T-Mk66J&k5>-AMNR5#yb#)##piBqcHKN|4!Qswga_3uZiV zSIh8k%m7ZD*o==%D{isH4zNRxTt&3>-JU-!@8_6Iw0tK|Yv zY5WBT^_{WuN+m|IJ{C^=%;x2BEWcHVo7|Pqp>q*l5R(Ak@TG9Fs2dkP{)V9=Gx6c- zUJ!XS8l&@cP=DcE%KjeCcAg%z{8bPoGS2nus3%mH+QX&VtcMl;K@cTe22-sr;L0K+ z66qA8;_v2k==mjz>aV5oU#?N!nF#i~d=}TvI!AL`4fx9rpYU&uBmG@$!uIlGsdM!o zTy8FnXGHS24ug2eD)@jij=jL%wtFZwX#`C;@PPi{Pa4?HdZoHg03FqtKk_}P{1fA^ ze(a!_yNQ%MW&urYnMQ@YKedlk;C+vU(I3Sy9CekkQiSec<@{5)g0a!YgErN#lcMUl z`4l4&N!Okg(y!7O&cR(9H{JV8A8th9e|uB#!&D~RDO=HT)HQ5A zHHMDa2=Vz&6?B*Rmg=-Z_|x^c)x=Q~sM<_Q$L{x8OB3eiBEF-0@R*q-`2>=g(|xy+Az z#uSpO)PH2MX(}x{CPt=Cwz%{K>;2kyq1~??_+`rx68WaZoIz?>{)XMfE?SX&V+TCC zEr;vg=#x0hVI(S~QTS>ly1dX1qDNWaF0DT3mpTtYF;(=SvzmsVpT(~lYRsuLl1>=~ z5d@z~$ z+wb;pC*;pFUd=}sDxC(BDiN?_&n4KSE`$;LW?}nSVf^Q;55LYNVo~fXu14oRcl>Y~ zr+Bzr;3oUS;KW~F`l9?5O(NFdPVEldR$|Wj3H@L`7)&Nh=29qQzVx0vPY-LJqohPW zihYo$_?LI_r+*{Hc(l;t7aK{sw2QGS6sbM3iXtzE(8AMa(6oIU=P*Z8@WIH0R7%9@ z%hF_QD$yg6))UwvT#KT<37CCf7&DYApm&TcZS0wY$^Pph-h3xl8MmE_yP^TRQZK-g zJ@(wePp1U?PRygGP;q*FtrOpSzNPamduhV)9(>@t7ph`yVX=k`=hD%OAxoyV*#xQj9ZQfpub>bHc!{57} z5V!F=ybNySmOM-0Qo@}v$~FxC_?W@{M^^mhL&AKP-eb}_5DR{HWuSPOI7m;lqN}k% zG*59dRV2sLLqjQk7hOAV5JweaRI|)aidBm zC#k=ed!ZDEIZ7Fj={Es>*iHcRXRVz7Rc~&Ot^(ipFPS#E8Bs;z2^19%f#nwKxNhm6 z)bT!@j(^`ki!XkrkGW}--4#P0zU9zwA!Ax$(}g#@`nh2lOU~m3#~GJir^WJF6nQm~ zT;+tIyLmAjR%zs>w@PB+vlD1DQU*`Y7Uz#Q?Vx$G>eOq&zH>|FQQ`Y~id*~;#@}-X zE9221Qe{jwhx)Mo{B}%k{|!~u2suOdVKC?%$|TLle>-Hr;O-bcN{;bgihp8?qY>Gt zWTL?=#vN@Gg4WeZFs*neV5%2qc_NzgU-^{1U!7p7stiY3=LJDF%WzxWDs^=g!1tU-z^fq!2|=CeT&xMJ!qL}0FlTBG zZ((_z->h1}|2=aNs;=FD)Nxg)yip!651)dWe#2m-qyptjmSRk|KCWI9KrdS#p?s4y z+T0O_kJY9Sn$-fv`2(o)X%}Ca z@HUneW9PyCIb-A65u@*&r!+sb)t z%LGahp_lUu$L|sa{wrw`E|Z5<@}{Kih`iaSk2opVfIg3kWK5v@+{f)Uu(R_! zoL}4ulcrEyFYr*tvp0$GuAWLR~mYp z_oirJeK_1Z4^7g2aLgnr?%=opa|cFs*?SDds47=f#$O( z@%=+Ts#)eqS8P(~N}eRYa>GcfjlN9N|7h~>x4gopIvsvcD30HC+z(Euh=Kmy2KYE9 zfYh25`LJbI_>`}?+{r~>puq7q1XO;+?^WW|lA26HV&D)H)x zF4WoQ0Ju8~Y~Aue_+l+CbyB010~w@sGJ%XojOJ&2_)J|PKk&N4Jd)-a>)qv@Am#7^ z@ZD4f3msj#r|UIg_a||Deg3|nFZ?%WHYE`XjnnbdnVqz8X)!4tdW>3(E1bUIHsiw= z6mv3**=oR+u~7fWMfd-8f`E zyMIo{SrL;kFYPYlLu#PJ*a{q6oQ(syGs)}lCv*v_$L!sTq&w#keBLdIBQHu|q8+h* z?F2j%G8wP#P=!>z!=T!@4|F%n<2W{V6MelB4K~cgX)a0-*72HK{?ZMua-lFuc{gW}riVN|y5fl5onJtHC?DQi{epK5YcS0I zB7SXT{`6{5(&_dTSUlJS6)l%w%Lp5~;i*Or^+70fV=K8a{$hh)5PA2$2RUa=tURQS zi=^g~_EkkPyufyZe!&!ZUW=ItHo=jPNtgqzz(pqYCp%hEUE!C)IKoHq;2 z>Vwk+Sr$5v`vV~llX1_r zd-%O83X^<}!kAwXXg^PkW_0hseY>_Wf3XN|Vlzc84|nR#%cI@9Yca83iPo^(dWvBo zjh=RvHZ-53dxPvvXnFw?3vZxv!x6aFF%=zTx1)OVQEa#ni=jJrVkl!4O*dB~S$20G z@_i0$IfGvpsMGQT@t~126FaaJHT}+VMn5mXDZ6Rd{7^vtfq}HLD}XrxLTOc|C&{M; zP~4qG^qB3QUlpk!mYY)T6%ESP3!u)9Z1%n49 zJc{;jrAT(#O!{?bI=PNIf=+vFaK7zNNc(UUHLjdMb6p{N^*aLpNPnHA+LT4eis5p~~E!q1L5;ITpu zYc7Ui@%}noE>nU=HgypEW70Jg3ef~^&P;HSylw(^p!PnJxVi=9#LA9GVM zUqSNCiD2t79lreM2)F8&!}SE=!Hj* zg6LbT0B1MXz^Nh$`ebK|YvV0(zz^wrwgUfky)obYvJsz{i=cn!Nx_`N6IAWMJS>Zj zpv{(7RB=X_f7E@J?530n{*78?aO9+wf%!}y{5dNgy`$&i;nTMb0(AWZ+3WfYejcmi zrWm9Npnbl9?k)*b2nfaNBm6M2T!_r_d)d4H3@(Y6rkI)T_$hERED~3z8O~qfj9!8P zX0_vh;}xJTBN4jN3hO;ZvFh%&7?!B*I8ecHUno+T@L6gT$c!nlgSi zIE`yY&c_w+2DxfiSHZZTOj0mmguiF0CH#N#)FEYnAqirgzm4#Cx*&@>Gmmb{{Ic!{JVR64|7^CvWE{I(g(X zwaAy#%MB~2?5i=B9-IbE_3irl1#f8Sr3JiR?mj;A_DA-aF2Gv{vQcLk=|pD@J{T#2 z86Q>o10HMmZO?T1qXUw3QpFK}Y%=F2NHWLYhbkxuiQ}BkyD;{X7iJcJ;eHjbpvC5w zarpaqJV;MiZdn!GrRA_TXcR9qE}BlsT*T|9`}hN^Yx(&XdwHvUGco+lITSWX$3+Uk zcvkcz<_c+`clInnnX`&jy5V;&ZGL6nXDu94IL+y z(t;PO8Jn{i$bQ#RK0J7%U3x;){kDYNnxBw=}0n19g2&aN^9R# zqTdEN&LQcj!GE*P8fYyPaQ)qKT>i4%f)%~N1_|eE4Nf^c)4wtO4ovYjhnaDP&~Vt5 zG0m6{Fxi0WWdCt59mauEXC@f&Kj4U-8n%ClMlsbEw5qhDt@0=7k7o#8n#ek&MrOFP z<`JG}j;|wSGJMo<7jB)sf}RAOBdK$zaK?@l+*--wX`K=1S8Rkr!oiq1wFV^_ALsFD zH=4uuKv(lQxaYSBRS#c*eQCk4T4)m6+h3=bHFNm5pI-cp!ZYx5fJ5i0{-}Ab2h4gO zp{}tC?U z=4YN`sl$0(w^9!+FHXh%)boMNYkISSw5yl(|KF%fI*Wu2Th=PRHMmTvdACy#P;^KD!G_yX2iVk?ubNk7> z$x12y(Ebtpy^S@nM!;ZpAD+^Uk;~}pXq?#?G7#129Qn zAB1a}+qi!TYNmDLfh!7Bzc3x^5@r#9W*e#bEQS63|G1*P@-VcB14Hv(E`7{jD5;Yn zv7>6t=j}|*l>#nzQx>;w!;z%NFL+l$Kv2fJG$KFM!J>kt-hofXL^pNb+J?EJ6FeMU{`3GNdn+A@4%$yV>U&GIXQRr-V6nlQ|z?Yrg7$mJu zvwk{K@t<`#B-4*}BlJo5%yH`Ic}OMCzEGh7^M)KEx=djxsnJRyTSI6G<4XyPit{m* z?Q|-B4KJE~n_4={D8u17WrziGsp`QXXEPcihw{kVllg10m-<;Z!PtEzRLx!o65Z#} zlQH1K!^R=dUrxU4^yGzmkg#i^g(e0XYAR2I}n?jjOG6 z@#O|hOtT2a{roDNsdxetz85k_RutK}-luQ;Z2pOGEO-R>V|4rl3b?tGVos{Dd5J2m zzVsWmF3kgh+h>@XpGHfWqxrp5HOZPXFX+}u6er4)YyUp9Xjh~o3aX6J{)e0MVGO=;F8%?bn##r^b160M8AEznlj*mEfXF!lN3SWx!c|An zeSQau*Lk7W8|D@nXN#woy~GMVA+mKV$3ZtGZ1&5-u4CcoJ~V;unmi}nhtW7%cstr( zvB#w+l<{!g2Q=^1A?+qx{Gg|S+t*Hj_qYh>vF@Fvq#7N%)rOk7skr0;JBRHjqGiPr z>=5HAe{}-EZ%Du_B!1^5?H=9dM`we0F_;1|Lsx)W|h`@OtH{(-UfT`gZ5ISC9Sk*AX zoGE-_*&P0l%M|{)hBG7>9LDx#izs76Dr&tg#m+b>s?F<%xVsVf@V5h6U1zS4f2sI; zOEG>GPQnF_r|^!=HN5Tp)N_&sfv_)zCS3Q>TNh0pR2e^7U z7E7rRcVAW|>8C1GRlS!y7k1;NxbJBAV>5Z-!^<)&DnZ>wB z!@2lrC>n&k7GwXI8cf|FOrMM7X-01}{p%{n5)0<>sx-wJ8B?g&W&_y@ouS%0 z7jpF(#d{`?z!bYdbmYuv$50vgo}Y;qh%WAOEH)ZWrO8cQj_z!LQgC0_GG$RrqHd?wm-j9VRrsaW|GWB;tO@0nD@B$v)R<_#|-{vp!XTh$`cI z|2AS7sc~5IMGx1zDUwuB9-5C9q6*LD^dMk6Qj`wfobUiw7fpmiT?pY-t6*+#G00!M z0RFC7F!s@PT)$;G<{Aj_)0-je3CzGf;}hZ1k{8s!rWcEg>>$vcvARkHq|knZ2J+KM zPtuBxHG!Ae)L}A%f+MXNOHGSbR0)$_ z?E1BShW|$61Stlp($SA**zEZZix-WgE!sZZRFg(Hpgf%lx?9lu(`=mh zb}Y=FE=H^)N9ECsbNp}@rll&w%WorKnRgs+b$Ja1WqV*=P&~GoM?wDK0M2p4188Eb zJ1ebJaJe5ww-x zt|u2n46WhXtr8r~c#1O0if}LRHiVuz!8yD-#jS7Fz=dYXcs*?w%7vM50@u+f@E4)V zxJj^6N*Pr=(ot@|FwFj84|hhLfpZTLLh=s5^chp(XG$cD`#26yG;PNvnU-ieuOFSq z2UFr}EmD7c0~?C^&}-x_{G{~`Uo0&_?{`FFJTFj|)myS~(Z_X9%c<9-(8xN#TqssH+XvcdoYSUx6^eeqE zrC}{zKbMFvmNP!P`gO+J0GfM4hMwX=+O$-U92GnQD) z!6i8|wC~Y5_IX%=7gHQCW78}g`TZ__xfX$09Yavm)(M_>J3#YU8uvcA6RhRmp(zE> zdso)I7HP)?+8%g9{4t!<3xp1~e^BJKsq>c$>3kO@-LEsy&{PzF0I>2Nr6#}TMTyhu7 zboYYJ`qeB09S?gCYhp>@8|)2|qgiufNQkOwr%*4>oG3xjYpf_-_YkQy9iYt-2T8L) zo!3dy;Cq(V()aUSn16phwT@XwIuV5^BJYYGC;UJ!`3shK?Im6LBJwv&p}V6zXyNA( zwElJy4rp%12uFE5u`?az**@j^E@(bN+YM4qZIN&j;-KA!s(kD7>3>GB{vctH`rtWSmh zqB6XbC`M*R%u&}~f>n}zm}qekr_*E#soaYNi}OLWc{VP1CJGux zIH%hWL87GtcK83`)M^4?hzGo7*?|8!oy1@LN?N#X)+Tz#kw^Gz<3*UZ%*lNgT&a>eLzqbZGReuk3A_mj>6 z#-&s2#{S0}Fd~n!cvyEwd);md{nUn?^-?6wvaAu4zhh3}1q|;_hSAG~X+_L9YQ1Vg zQt^x#2qP%)`7r8!xr+Na3-Z2QPX0~l5UTwUGa~$`(WeYNuRO#^n+Y_iQ_sDy8H6o! z^Y9d#C%7M9gYGMfar%aE*f=r& zJuE23_1lWD?R*;^sj;IqhUs`pKNL60sZ#S2S-5)m1%!WA0sZAGU~9uVTiuB z96j+fqmaL*bfEJotYzoZhv8&Yd#O*iHXNtj|LV!EDTvgbEuyaNl63!50F8|m;}`uW z!4LX{;OD0QF#P92;-s$A20fM!ydlGD%~*%7D<-gAKpIL}>fqV!EQ94CPIBY-;@({& z@bVfnT#Vz$FgJzFwzIr;{9<(fU4+)>!m#4bC{hSlBE1zMWZff)(@e`?-|kP?vuF>V zt-gH5d+R3{yR#FEj)u^i%m1V3 zyyLlg+c=Kw>`F#x7$s8jJ@<7TMF^3SLWv|Tm5f6DCQy=CKKFHfKJWK5PMkbWe${w@c2y%R^h<@Z(~m$#OBN3q2r{V;{*l&A zqp;Pdov3kjgMz6Re$V!UR{J@))?Nf3^p=uqO)7Amx`<94J`JNr39xVXZ4eyM!Ty^8 zs3m&@=Ud6Mx1M>h&%Ui-+2v~ZIsYRZrM#pU#}n|!>>udPS<1EsMG_CqtK^=71YC;! z3$c=_Sau-?1>d{{-Fq}2t{#B@a%y0?dm0pJCqjRX6pFgc!tJ&WV_!ycq1Q8s>?4^ zFIGBatI#~bZ(hv0Zcfj&L^P~3!iS5aAYD%!6th_H-~ur9&p>|hSy-6y2{vXq(A}aC z_?Fm1ddxno4b{U`9)0AlSdC5-2Pwx=8f|}5CSBzQtaw8+p~4p!o>3lHvV0c|cQ3`Q zi<)4Ufe1RkHNbbB^KfgtJeIA}W6n4Q617v?fiLG3JkR(5leD8%d377^KVyhf6g%(6 z#KSj9Nz%0QD`RRW4?`D22sfdQ!BsCHc2hjaRRutRyb|7{euDvtNm$>ZfsvG{+4XP` zbfi?#Q|S*}-KvPKQ;3&?g(wgD4cu$kj{n)yXR*Q2(O)_Y;Q?G@J4>bS9&qSHrxP(PmbLJ%a0J zoS`yV22y@0fpg^$`SUxU?6Km;r<;B7__PnY`7glnL3wm?o`G8jq}i{Lxj0@Jg}n#3 z&|pzN#KzA-v6gBWroX3oBM*FH2Ilni=Miz_;=ndtlD@QOg@x?@_!e= zEc`l22umQ^T*G9fNSVn$bqTCXUV*i`C^@g_LiobEnQV#WDlH%ye&^+_f`q) zi>E$q9ZM8#Q^38p`qX8z0$+z|`U^ zBy(InNS)R;m{-;XjcV4QmYW2z9=@<)+yGLZTm&J{Lnyp-3--*ehNFGiP&}CkZr7*a zpX5F)8kR$2;dEG0NIlt+*uu4l@ObnvwS0dW@1Q-nbe9DdWiLe~ z#rbG^TmqfFdC+S>40TQUQLp723@lnsidv+gN;d;Mym{H3r*2`qMll}pI|~JCW`j#_ z8Hwt+45iFbIN^C2UVot8QHf<3`(qL&7v$5OZWUbYDgd=jp~UCXZvvxP)}i>6|-CvuFa8kCSFs!vJ2T{SGR{{mw{o~&-X4jXawDF%67L7$Q`jNBE84@A~M zAPIyMl@6ekc?O10wN&llL+_Lv9ep+}1H>IdT(>4=$A!h?- zGFve5>?WM^_%!l0%|jlV+u6E#Ki)h)pSrA?=}fj14!&+8Jk&Syx>y8glNqz`rQpU` zM^t`ojjCdAKqf2;)TCx&nHrsSESm!fLYbh^drudk(vc(xuBVy#h16Rgpv~6Aen$6) z>Bw7B0e(&&>G!V`_Gm`XbJH7nGwRU!UKAb?*^LEX6sa$Y@;pv21&MEzy%>`Pe>mOX z-n;@|ToT4H{c`v(R2A+OUV`D(0dP?1GFhgP3?^o`;QHLvAU})l(@U2@P~$vUn9>Ob zUllMxZy637>td#52<}hS!0d`yY>j^l*4$~w+)Nq#Wv_?lR&7M#*Z|^O{EF0a3qm1f zuDIlLD0|ijKky~v<3D9sakdKAGp6uUVj~DWxmGC~4463T0Xik*nbFWF{`{ zS&#d3C`W!~Exf)w1OL9Y#QoYDl*_XWXKY+b4qIq5Yh?E`K3PF9mDU3?8~bS1ju)M1 zy1aLI0d7398T*FU!KIXA#Jov@<@OR~jVY%~ni&MIoJBBLcnI`=6_K^Ze;M-+i%I3r zA=u8OkZ77kuqiOd2X~I)ZpUmCbo>rGy4zuB>J9`RHiLev@6^pFfYk-Z*{(`U_JjBe zcKqxFc=6T4%*}J5fh~iUvsKVLd_l}MNRee$Z3n`Gb9gr_5CAn#%W%oz6tk&t}o zG7g63aA|fZZ4}qus==wdp6Ggd8%}Jt#)U06@b7Heo4eit0>`*mZP%Ts%GV4DF$1uD z$tqmGSOrrSDB+UWy;yFYkN$Ci*zFyOFFtADz0c0XMxu(zEfFD8^;_^m!Aaa_m5)At zDX_*Q5^nbtL0s2g_$pZmJ7wePjL#A)3ypB|r$E$8`UeW=h1*ABaqb~?yk;~GOYZx_ zf}0&ARFGzdv`)i4x?8_Dq>hXH)}g#N{+}RqG zZOO+Up1Yu8bulxrzks>^YB4BGF_16aP9|vH^m*7Kj5Xz@=eatbe7F!-UyZ=GpKE{~0+jK6IAkii<;lc>xjUh{AtX>v8;(0B+fz3!>5A@Ywx8%65^* z#a-dB>PadgmcrzQnivQd>?1*ik2uG9+u_}l%`hSn1%21d+ z1|~gg;i2&gNZ&cgM78%3^L5ur^|kGK5C^;fsz?=%De3RAr;BM%OkJlW-R&OUvuS`HS$6iSMD~g-4?!g7>KUhl{ z4qKm54l(CEE6c;p7R~m@PQQ5O@wG9g^OO(Sb*zpwj>?nhC}rkE?gf&+s({R0oIo~o z#GzZnb+DN~8`2hcGD8-kM8KgOj+`>Y*WEmLf;!g<*CDDnHh@+OosYDol3A5WbVqNC z?*g@mhbkW#Fv}vm9{gmy{V1$AoQtu`d@;9b3~bV@;8wE>e(rw>S4Ro>4b+lHn|WYd zG8w+lh{G7V%gFSV#ajzBm}fSXWF$f#+D!UMglIX*7*Pe+`|0qfewg&_nI=)z zhzfYi93cy8?vcQ~wC6Wcqw9X*v#jL;A_vZz7O)yat5U z{)J>$9g;LAMmZUEP$&|Ok~@HXKRC%$>2Z<5uwNWYRdwiFIz{Ye(%kd>Lb43XiP%#& zaGkUvIThAqIFDv8?-XNM%P6Rv>;~6-6F3u)4L$WfxHbA2c!b@DBJv$xwQIow2M02D zS1$nMGJUgj7@b?LoM_FPB&fWXbm-nBa=*J_e%MSfs+~=OG{(VZ@HPAvJqJ&5DvGK^P>z3x6EZ$FO5u=-1JL z^IMmqaz#D(hb+b_V|xs9+>Mp39dOU~2dMbWLuKt_IAKe_-HHZ4J^LRleyESOH21vU zJRi?-2Jo7DKdP>MLh~9D@cWb~kmN#K_Gk~vW(lF-YcDuf)lR-C7?XFp!BBd=i-g$z zAv;T5;Y&a`NMFAM3wxsQ&yXW}RoP?WuA30nw+;APSdb1s1`-jMz)#~4WFF#0k*o5^ zQupon)eT5Y?2y)B@w$#P6+wO-V!3OyERXj%gsKfoy4VW}A6Sp+`;7XsZSSx1; z%NC3?+w3ne@h$`C)UM8I+bm~Y9uC1{@eL3fcoj^yUqL;22CZ+bh8$mdr+(asDh`v# zw`(Ux(0BY+A6c@=fHJ+)H^R2&HFWpLK_25Ru#IzyER^EHkUm2!zAlb|^Uq=Sw^g`j z=o8!)48cc<3D~e+3ztWv;u>X7w&F+)%hRwN!W%@v!Qwhe-Y13Szp7z$Ng{Cl2!MS# zT+r_Sf@qBNLxY|V{^74fPe}!==l%%wbe~rdR}J46Mq)?QIoz%G3gy11pz@wUl)Jwa zor={_Xes~>^SNL?Wjh@H(*Tp>3D_@s3~zKU!Z(_8@YDIt^c)EQ?mxm9cW3}S&QyTP zFEZowyFNTy& z-1zwNU((_~0fN-iu~Tpc_C9ro9g(x(*`ZSC<)gdX#|kJ>Pg!%xpNK*$FCOH!0->2d z;o7&AI2~cgR_ReMi1sw>o0`O0KX)K~G9@<~E|kY$dPInwHYiZWwZ?YR(t4{XK$@_p#D zI})vZT4**f0Z)JBN9LA00dcuzwa8 z{!-qJ5}^*5K|6*SiXX9T;4}DzDT7o;3Mn>xLMGCR$d*M5&@Gw8msb4vS0xilkMDp5 z+n0c_Yae1Z&(v zBge?Ksxg1e+(`Ac9puIl9?Dp}OSn9n$pWQGQa3cl{0`!WotoXSb<-gt{^J$%jZYTR zWPZTsDGkW_A_Et%q(g+z75LL{4JIwiDGx#k_4+Cy`HVG~3Vwnz;~qHv+Xx4Z0q?|E zp}fR7JV2u>0ei)9^>xa#8Z<_IXG^TPEC*Y=qCsM)6fAd&p}C?e95)vwos-YtdzcD@ z=nBG)Wd*RMpCI=f18f+S1g}-gU`+uR95;+b|3+7=^}h`Zf9J!k1OnSje=}1%{xZ%5 zq0E^Fb&UF>YEWEM4qb6CfEn0?u2%F6KSckXoAl8*GlOzs-;%vKEEbET!mau;qH$;h zGEXX?A}C|M^=iyZJUP?zZ~#f zk_@?GluAYgVuJNBU`z>=pL`{1zT=FP zbP3qG%*U%@p^$5Om5A+DA>#KlAZmFy@KO(M#=AR2SJV(foYsMrh&WD2@uJJ@C-CP~ z1Ele+#c)$;v>OzL_ZLq?Qm_hacMV`u9~@(!?}%h&kG&yR)m{?Qol0+Ga_Dc7sO36;p|cYj1G;!+6N!d zE$t<)7oGtgpQ*Fw&vRJgodWaRf?@uONYJNQ;JI!QAT~|;SN{UAh!w%Z)0$xPLLV<) zT8oLZ#Lz;0CUNUgBME!gkfjqB@aB*`Dm}Hp3lZJW{7xPv_a#A+jVt5?tc5zYJy4NJ zbE@&?Fj3S){_5NSyNBh}jeL=`s+Q-C89j!K^aHR}yUp|Xs96dZEbscZ_ z2jRz+d-1>0)p+Y0?Kz$}2TD96Zr24igh&-a%2?s^k zO}j86b1Z4+MHG(A4F+b=4C=aMAY?=wHf;J1jZ3!E4)q667ZSj4C&r=Zh&Vmp3<%$> z&&-UMZVbP7Y)SpxGPiB|Cvv zNDL&YmBW*8GOp(MNg{?=o1_U$Acm`-WbO?eKmt)-$b$eoeFj~hk#pE4=ie1 z2komHNkeoL>`@7X;Oc4!{*ep)e!+OSRt+Bq6hMjU5%&A4oA`Ur0bH*l$wXE6lkHts zLHK(jNCtI4pEWn;rpvI}jx$+~pffvp?j}j}uZ7CA|L;xikmvucgm5_yKB-G#9NSiq zs`ZXA>b-y&5c|W(>L?L$!^L<@Z5S5Rx`6J~QTX2Yofr)yzz50Mr1nZMX?kr*-IjUG zy5p^6dLj*6kLh7vt|R z1t(!$kv5DuTqNn<3Z&*KA6vOE0LOBo!T;nOR;x#zg<%@GaQ=G)tDDe_C%P<9@JK!C)0}Yr`zd(Q zKhBx0eVbUG_yMLzR9H7zI*aev4X+pnrs_#N{GHJYzI;;*IeUN@^J;?Lu2kSUas$TY zxKQEE3!F;oMa^aP7_aw*gwyx%Z>?2uVMafC8@J#$n+g0G+r)C~-o~SGF0f+M9n#f{ zNXF@Ea*oas?Vq@V?CuCSx<(Lm{^>&NpK{nJwh`f17pNc0Cn3sG&`{D1-m0JBipOj= z_T5IB|JOygJ$A8$98iVWTIrGJCC4Txy`8>6Map0^ptF~5&{gO5lebr;3AzF}n0NUZvt4R>M zzZjcc(m>(+1{mI53*1~U$C>*e`fY5QE6ZvH%EJ@N!x~6UMd721 znCBnA5cohaB)$Kc~lV`OvFYz%%^gO_+hm?;l&l3zGY zZrb&eLoW|Ol65ppvoB%Q=J!ymG#7ghT?FaaOjzWpfOlJ-F)!%+RmFchWL7b(r}i9zJ3Vh@4f#!qM{kKPzX4)0;d2TXg79Yjyn(ahq z$r{?jHioC#gmRsZVe;u{ylU}|jCCo(!jIIWV|x#36F68(-3+0pPq#O;Ix>`9_K0{aq@ z{Z0VawbsBcNTn>m3D~qQ7q&inNA}fe!i}$s$TvyC(U4%F!s9yGZ>++rlm*R=D#6>3F7d`#eu|7PF}w2G;9lRgk#q{;d7fARELC;9_v6%p0Y*fi(McW znMfLve3|OaFF9|<&nfiYevaVQ}*^L6Ly|rJBEl|gk5}Gfcsv<%||kjK0BDvJ&>y7f3Shm;(426;P{bs zk$cLfe@tNq?|mlGVIB|}`;ENZlfw8vLn7kXMl!bPvy8$^+?S$)x-~R|T~^BJ4hmw- z_qoF`-9c!@z9DxOiEsq9XA%{YVaAyI2o%qzEbSj_U^7oQ*lq}fB@)_T@I~nQO?&lT`71zYF&n zCZSuSFz)I3LZ+WZk^^ByjKRVo6q-|l>jhWinel2ixA73=bRk=K-GzvK2qcj&Mu~<- zFse5l1g(1y$5G9?E-ff$FI!s5?U$GS!*$A=9D`35T32{I85c=y& zVAd>vCszs}vZ$Omgg+cr0@PTH2(7!}8}~ zW<|SEKCN(Zh7D#f?S#(QDw0uV$ys+vg^4h=W(u^Mm^hn!W~Jd=hAT6HSXF)_-F+X4 z3QJYu60d-pW^}{&XOTM-1Dpeid7LNzm0;ygFFd{f9+W)l!1q#=A<8X}&IR+akjove zw{IoJJAc8g{o$~uE0g?N(L!e-Ie1F;8oIXkLrInyWRz=?uy8?g`Gf!o6^LM@8=vBm z{XF<3@ju``JO;Z(G+-on6G&g+BPSK+lDwqLOt8@q^cwO|p5ILJ{nj`mckmUXJLtz8 zch4mu#UtcaS~bax=s=O`edukG0e{cC!~DecBwe|ek@S4Yyqvp|QyTY?Q_BI)+rbu{ z?z6m{4JjQQ)vBc&#kNS!QiV@M*5@)&b2BD3d#Nve(1~7jPlJtXmC_?1XH6sn0I*K4HlNWk!h|;;w zC2}kN7fE_lLR|S<$Vcgi%tx{XV*Y$1U$@TzzH>9caKavEanfjB!wp|8o*=WlTtVv? z57=oXk}V3(@bhgb=)`qnYOpFRdDfXV@o%PQpeCyEwt&&5xvaC2G3yke%Pyw-nDcHt ztX-o5yQ9DYRljt=d9Qy&e8E{bZ+!u?E*xNOV=fUh!F_N)hlLkE`Pf$do2VOXhbJpl z$j{bN;1B);5|lC7@?Q~1wB8}IYAfL6J6YJ+{DSP)*T-uSolP<;pe=%IWrWtR%|8u0$_$1{v{NiVkB> zVX5LKu-qt(O5?^*ptcB-jDk>TNP_L0I>y@dQhz#~H~;uj0hiMyabCGUyyD@4=WDu1 zetRMo$9k~7ug;LdECEs_+06NA{|Nrm{s&|4)KQ^colW`4%Q{AJaCPupC>wqbxnXoa z_*fbq`iem%k1%)}rI61?V}$kk%$)C=!ond**6!IAjMH|-6uU4i+i)Ju)>NYHhTXV0 zDiu_evvA9kW#};M2ksdw@RtJhUhi&2o>LM`hT~#_mGj8+`N3H6v>2s=dQfSXF4~#L z;TO-m{Olm=_X1$Pst76884Gcy`0M8(Vk%AP$VOzLE0zhtb||Hr8yq z!PS`aPy}71l|e3kjMy6p5$*MRIXn07hC$9OSba&88kt<+*ClUgp1MsQ7ljb#y^=UZ z-I=>Nu%?EJ;A^ZRxvo);;?+0uh(nC1>!jTeDfKZm*L9gU$| z-r?_y-0bPPSk#z^Mkjhd>N9p>OUj?)Z+aIyJF^1F`zP?DQ~_VdJc5xchcQzl3wfP+ zVAi>7M0#{Dvv!jto?I-A9-(<~?)X0TbYeK0I@gF@zRZ+d=kH}+v=!LI9{M&Bq`*aQKy?;Z{ z%}L0pJPhWc(P%s)9rKq0F!jG^#{CR_Poqq4-&@3O=n`BoT!C)+v++l2A}p-8AX|4x zLuHN}yt}vr9zHWAyK>%aXa=>1TFew&W>!a%-4z_&d(LwH<<*=dJkTcmT~? z)I*tvHL$PcF>U)8Fg)Lgiug5rZbsNk^A55hd7JUxw(lhI(kf1XXAL%*FGS__+aR6# zvevDwA*@{>Gt+1zypNT`zr4nnxK<0Ngw#;;;04g+e~&-04qv=)q^#~mMc@dZ}`k%pY6Vee?$Vv zn&u(W)V7f6d&I?h1{EQXf*9^OV~ss+o6vT28oGG2LB&24tg@CsUS1zusm=$JL{~VX z5srTI2JuzweGI54Ao{KqZhr42d#R)CERO=7JRV6d4%MJm&l6N#QH^7-KA_$GC~Us> z1nOs9LS6ZAe412|xVP zwEM(MwA&8hKukJ(@9$+An-+2I$ZTMiFGyhGgvKn!Oo_kudv?HKzRHYoy{#0yMJANPE#;=FsRAc!piY&c$5pz1=O4d8Ug9 zd8;zVdrV2eD+Ok`uPR3(?*X=7y@H(+wy5bH1G8}*oU{&L^j25H#=i|vMdxq^MN47f z;a#Y(U7F2)SqG<IHyHvA@jPAm$4}@nD3?pe|E;h zAEPd!G;WSk-Lr9(+EHK)8909S24$Ab!>Thjxc4jv9|oCYX7V|7mX5?_5{+m)4M zFdsM4=Si!+A2v)UW9@o*wp?>LDtvkfyKV*(WkG55&visv`NetPxKXTo5KNZSu0-e% z{l=UihNhCRs!kmbPO9Pv-2?ZMFrs@^mg&roWt!Z+&|G&4bEPw!!+ZQP7-iIV{f!HfpmR6aP2z{KJIFu-fsnNi}pZ-%>)y;a-2!Af5I3(9fi#Yt8w>a3s0A%z=zUn z%v~>K$ZmNG`^7zwGiw_(z5GeKH@kqn$q!uFbsl||yoAZ(0$6@O6^WM=yK6KHC8h<@ zQG5=OO$dg#JMTfgp&lB-sDtq%&5L#|$E6R>fobgw9Id4H{G2GXFw4L!Q)g_kR07RE z)GJVQAC6NO^=-BT&#Vbz;TP@0yF3Av%9l{amcW+k9gy&FA#{B-qdn;qP|X#E&;ubP zB*GnjPu_-Zp*Z*&@qkdG3{eo2Bq7)GAi%c?_06=-Q&Cu=2D1y8CG-LcG4_0YkhH8!^+Vi==f#`Uq^e)Zfyt61}<}U->?3@fSW-~$dZ#dSTL|O~lnXAPDOtQ>N{P{c$+xV7aa2;Z0-UuwcRR=;v!zAt2 z8%8+s2Q&TBk}^{0v+nPY8nV=(899I@e5rWQ;t^@|T}0&eDG_Vg zCo*BJ#2I4HGfDDmWgz3dIY~1Qfx7j9c>m5F^xi4NUN_=lHy&#tv8EqEYXjx3d79u< zSP%N`Q)Bmv$L_F+SL2#THMWj$st zL0@JiTJHG-{YeePK2(A<{b`0M4J+Unp9P(qQAqeB&gQ7jVMpBW;l*Do$)0r_NRd@X z|EMpxW=kigl!{{eVR=}!^BJt#F$Eqb60CgKJIr(v!_u{p)Q9y5Rx%10yWt}7=iDMk zW(<;Dv$Wxa;S`+wx(C-koyYhqOGBnu3MyUgBnj=C$l<8XaPdYZyr=xig{CX9Dj*bo z{I&pYeJkkHxJpE{!az5~1oy;k!OyLR=$R#qK|_5o+x#gEepdy_i3w&5F9P?d3EHjQ zgjY2ysc-Qi{&lYfRxA+Buko-CUR0puvdeh=b2g^*@Z$Bi4k-KmKU~0F2Q7&*c)-gO zMM`Vnbb2hHmKT^5Qcj)p8Fc>=hQ{LS;9ABH(7PN?{haz}m8^#6?K$W$3-Mxx5xSad z<06`|=#UkntlU7{S^SBy`=Ae8^{RO6;b-zO_$7SgTZIY1kMPL$8ElJ`B>Ux`IM{mx zK#hSFYTprK=Lt5TY}|7&hI0JR@-F81N}=bXRLmpe=#*ZJeh);@<}){Hiv0wBddGR? z-GWA&*HH)513YJ}#GV^r+33nyxax&3zVmTH^~>{!(!>kM(YlNG_S+IcsbetRIuBQW z-iEi&yP~6L8{N13C4Z+^K!8L$(Vt!g-RC=CS?@y#czXd{k1xVhXD(cyP1w`c^Vx!V zl;@8OY(DA?h0*J=@f8jFTMhksul6y5e>4(_cWPV837&zF32|!iG7FTu$?+>+yxbIxb@iLE`*1F(2v#v(V!_1Z+#5)+OhG@|y;L$fi>f_z zZaz7N<_ERe{?x1Z-w#c!-hC2h)Sbcpim!~FaSfR?n2E!M{kXr*0moMukq);al&5Fe zMZbJFJwQ931H#C^T8&Nj;DyIZi(uzs?ZITPSX*l(91 zaa@MzL@B|0yVH(Dbeoq|+p6|!iv|I9o z;|CwI79h{|vpDoc36ImBj#=qFoDtzb|4t(K^lmr|EvMgTtxF(iqZF+5wE!_6aSZtx zgFWXo=$|nJik#iR7qW_ZaPT{!7Gp>j6@fGRazJHyE!?swg&}{+8u34iwYO6k;e0`| z)F6wqq2M4C(M*-IsXwN@a)Qi_agfF1jO*yPpI-1&D7K;O_Re}nak7=Pdz+$A1in_+jW zFmwKp4;fss3={0VF>Wb!v~dYB(bfVm^wJ4W$ZtU3lZ0$tWk=F85{UD{a%jAU*s~}F z+wadpC#kKdLFe?zf|rQwgb+rp&c%_L!tk?oE4oX~Vw0{0K(CP`d=K3XK3o0qzty|Z z(!~;IE#3qv;gt95EuQdWzB@_gziEyRV} zT5&om6qn6x;hbh(k@ZG#@LPWeGoEV-RT9BCp*cYe5|l_rOg?!ja1A{le?`V70f%== z;_mu5+>w=!?(1KXFU2J!;psUtnO=f!s$bC3Gal31yvS^QLvZ{Oil;o z^M`V&XWJ9G*mJP{^$n2mGQ&9TSbUdf2YkYFu!)t&7ozVNvrtTjYpyAgMl zb-=fr2PA6jBCJSMM*h16sJZk7HoUosNrH*2Hdv;7nr%If3F*??8^j*!*#8QkkE z#+Xj{{6JhPuw= zmWiTZxgr|NBZ}UtB{yi+S5}4xrY}ANpZ;ND=+y~LQ)jS6(Go5`_JWa%B3N4X2@VWh zVY)p{z+v|V;$6VUC|C`XK)rO*p}3VSt9E8?-co{1Bj3p-5dj!6ihxM@3Mk76g9AZ{ zL}bTL#v&_?N$%Q@bKR6t(Ed27;qzlJ3YFrT9htDWO_PYPbRedEj--5g8b9{>vO(vL zvv=jSkzdXc1BBJF@j2ps+6OG-UW@-?lyTT~3`z@VKlYg{d+p>|HgLxyHvQ^-%6UBw z=>e&nMpY#+ir)ltj~*elH;P!%$WN@Hek;4+?P-SJLV$4!DAoyGw1GOcyiwUN6-Qr* z!1*gl#I)@=oL*Z359BWsmG!%IPNhvTnpe`u-Ypi?tE!5&4%%2FvKap~r(kE!06QIF$UjW>X;t+fQ1^raCt62E0z2TRc!+C#I+7A zz1m56?=|>2RfRn3ng#x;TS?<&Eg3NX3Vu0$xO8sjah=5LgT?RAOu!k z)4@BHeC(kMhV14wVr-RA3;HVyV3^c>BpP#I3%4w%WydhHj^?;_K3s!7I0v0ianN?h z7f`b*VgC&(vL_`IAZb%44qVyA+W!t@H*E7^k6xp`yrUl&b9=g9uepNDo><|Q%>UTn z;bP=)wUFjRF)02l1r}dC#ps{1fE8XTASYNxiY|Ue-g}+sMjfT1w0~O_(T(b|vFwPJ z42mB-iGjA!c++e#bp*TPQuCRp;PsMog?A0^Bb>ww`-LAK*&E*t+Zp5=eMlC_I(V>geEu#ehqvCbjSneu{B z=9Jug&X$>ytbfujJWR6$@_M(BOQ#UO|3~M<-m+NUavN&16<`ma5L4dO!l{t>%^5fe zXjNB_k1Su)-O+70{%-)rn}jg!t`OwBk|S03+L_*Gn(Vf|BB)< z<^fH7zC4G{WTSAy;Az&Oc{U4EuE^flhbI-*WBhzGR54Y;DEhw4NMYbZaULqBM&TQA zMVx=YhCRh;!&CO!p#4LGym67?^g7>Kx!3Xpd(>Ni{dy`J8e?8DhxX1SX4ZC0<{UjX zVa7*1ZWs-DC8_{>Gf2*zAaaXl@3X$BAicQaZSgb2DU$ZHMtUKW?f_Tcb_MB=Jdn&~ ziI<&TVZ{)BbXB@P^Lt659XI~DEOj&_-V;TpKOXD&HVRqGiMK=7K3+zkXN#yu= zU~FA0IX6d(2|5(QoFBNt`uHzqW$()3H0@xIQ`R1A;v(-yPXdp?4LDi21Ge^rqP_8Q z-1(CaqiRp!f0Eva=}$mhAORN9xx?N6^J1ybXV#k)(9XS=)VGj;C^qdt>5Y(+xB#z_~P&1Nv^oI@Pj7(24#s4VNzB7pMwZX{mM4=!EX0Iw_8 zksBd_a7s&;x}@UJaEOa`-|mr$31K3TQ3m;xpPfj1lAD)y;j9<@?A={E5HquJ@bn@S zoMD0+DdSM>dNhQ|bAh8~0eMvOiKv&#lHL{;OH$}8?9UJ=ESUt4@Fh6ASOItZ=SMVK z?vrocH{rsAD4aIy!ZFu5$P+e3mgT6DSq;c+?GneN`O-vcI-K!ZupMH>RI&VaE|#C* z!+;}^`20r^F7R571Kj7pZ1)lJ>O&RNDLX(8j8u@E=i~TABnFqc6yeSt{A_~^7yEeq z7o4=)gPWAh@bE)bSeLy99%kSm@7O3 zMTDHOp@SPctsH2FehxF8nTsk1&ZE?YzYx)Lh4~)5ggJSynK8e%p8cJ`&mN<5jsVF4 z5-)U#oV~pT=5*}GuLeVS<-hq;07JTzw0-tNb zNpw{MN-x}l?keK=d099pIPHTg`XL~8as;}L%HaBlLb&C61yAfS#*Ae%v7)9IPk*$+ z)bE!_yrv2Yj7XyPl|8WF1-<*`xZ%@Bzfh>K0F>Xhg7szUP`-Z>JMLUS1%obpUuX-x zKi$CJ=RYX$&S9HB7_zBJBCJLv<(3%SC*6*suwX_XK3!dcWi;db=ExT^_iq;XdUt`P z!6J}&mPia{M-W+>b%-gb1jEtaBLt;#oqkp3%_qm>W-4 zNMQH2RXFLn98KO=Lfns7QXyDHR_~z?;e$BT8l<^i&Q{Rh6OMMgz8HC54LctDliu}9 zLH^Kpa@o)3H#+bXy(9}a>E!DQ(OVp%u-3h z%vB`1?J%)ERDl1Y^YPpOdxq!LZRc0pi_*K}xm`Z0^Xy9O^UdDq(P$;}7a&F+f4epp6)$T!QR(mksHrt; z@!}IpxE&mWzF)-Ibq%x8HTn<~sf#kfWt-7tmkX}_Du%8yW|&kcPL^u%FsD5~f|=V_ zGUvZFME?DGaBH|nycWJCrSc4D*4zM%s|xt-Uo++LHlpgTOLXq@1Ir)RV^m=_J~|am z>?WIF|CtEn4POb~CKh1EtOkK=2VvLRKv1D`(!Iy%9<@ma4BOt3m5DEzhdxe>`JDr# zW0gBv|2&29pZg&u`W|V#A%~-vFX8I}2N1F!gypZhAycXsPW+O`#jo{1583+Vgr9A(blMlD?qy2ev}?0*%+-Q+BMOL9QIo;zrxbsw_= zxY*!wXV%{;9rW2b_%bIBhO+EPp_c*7^qGVH60d2upbV^ewm|QXb$FaIbI(U^#U&0d zaGB=xb2sX+e>NONhdXKL=iLn%X#tS9jt6)82ZDS|DSG@~gWujYL4&G09(~W?Ann^K zHl0F~)FYVxe;)9hkGTDjBD=gyhW#Zd1JRTny2L__9+h@(dbJL$;^p6WDw>A^19fsP-hfo)Gg;ku| z&9)o1VzAt17^eA_mqxbW{dqnUp?nic9>-zoE(PGpZ2~8+Eb#fTnthcc%fiw}=ypXK zSJHRL`*S8}eefwGEM!A`f3jrhq$fhgKM0Dv0=60noDI^0Wb2w$uwLggNLXs(%S|lK zau8&tczQ81TN1Mi?qkA>52*ce2J6wBN8)~;gTF1#uwu>=vUZs;_S2k#N9Q6oTu`08 zd{mCT72XZC=NjO`@Hn{V2cuPz3r5}Qpxw&rtjX7KHo30{jo(_KW#xAqQmnx0K#BiR zcBRo&ePQ2NLdg&^Bts z)ep&RrBGu1Ox&m0ChoEBm7!l&Z-ig1)}T2*pYD8bN&|*XqS7-o=$oK!625RE4G`b{ zn%0o-*GDThZl)Qkv4Y@`EJ0dp6uH#3A39(B6yprfh?1+P zEDWXwadmXJ_HMeZ|1Oy(IbU$HO<$mu6=_pspag5;2I-^sAH^DJV>*4F9nBNtV5a3K z#D2iDgz9CGQtlVoAGwHn?KnjvTBnma(p%`b<}|utK{GKFXC4%e39~5>`*LIcUJ(op zm$Eq?J5n%nMw~$E=V*bj&6XIAlq46H%@b($fRxNrC3XE!`TwAs1fAN_f6#Sr2C|(x12QB9wouuJIT%|F=Wm+XY%LX6!N<^ zf$S(AMH<&yk+>zs#CCQtnLJ#EIDcO&_#(U~Sf8UU5Hy+#sDifOk9f!GKT#{ljm{IS zOEe{p35kLmi;D#(#kww~F=qs&)@uddrLD<~wg-Z}Kjss&bEAmdFiSE!U!7z=1)|a9 zOP&~dk%!ekY(iE&unAh;XCoIV<`Tp?GrO)tlc;n>dbzrp_;g(%2D9~GPm3Oe{iU$L zO_EluYbSO6`^ewGJTmH8JW2o6B`7^APog|N3XU%66PPL9qFHyU>8f@SP5ri=%vIMT z_f9V${g3}+mN4A`{|VnwY2Cy zqwP8!f|osM0@3yuLEkD{vgVc*aroy%RLG#f?&M*C-Ih0kv*q2y>tGcbdS)NFDE7>) z63-3$wFt>uok+gAZXm(iL&?{(#&k-YIZb*uM(opnMJ8vzB;=w2y)EvyUFOs-?nrGU zwyqjv?{Z0EFsW3a9bQ7@3y?JYZ-!B?7~3dZP1e3RM*;@dkOeDpNXnBG^2*79EQlCM zVmn=k&)XQ%ZoG{|&nOkNo*PQq7Tc3GpOwf@fi1CKGK2W-Z4?y8?GyWr{7FN(*yFXl zS0K5_m_%zVCd+*1k@6At+v!#!bIF{zwhP4cB;yfuv&f6wP^`iT-Qea*0qxUA``k~!to4Kw(H@ zN8oXn33%W10MAd(`0e)wr64F%!MeW z{Ds+>sj-Cl^Kq5(QuLhS2)8$B(7l!tbeOwz z3YDH%<4(Ij?8kvC%*QJkf5`c;f4aU@1dNqsFz&Tc9ISrVfVW=%V#jTckUr%q{F-fw z=Dp)sYhM(ulv;qF3Z#ggu{yuuMlxRd8O&}av|@NuAkNiHC*RacMfbm5W^SjOScAeX zPJ_syyFomjlxb%c+q`js#6#BnNf+bW>@c!+F~;3lg)@ddp|6jG0%&XSZ)PmxU#XmA zuc{uhI=`uS^hp{%PyR*6bjHKX8pf>79%pHOi%7dh5gbq*g^j}};{+#9mS*ya1y8kt zIq&kx^9)V2e6SJse4LLj8a2^(SvD1Y zB1~~k&~^;nF&&%D-QmHQF|gK94b``~L6z_bgoH#n2M{^XGK0YrbY{`_k|KW%U1MWcmbtfXknV~cRDQelAf^^dkP90C(t1CI<#_@DG^OhnwLFWu}_c87GW?+sjXkb;I2h-MzZ17X?Us2=da@@x-Y7?h1bkpC0zc6M*|(UJ z(jokMTpIRNlws8K#aL##9y@HZh|!#hP@Aw5vsMmJ`_~fmexfOkpB|6P=ZvF4(hkfl z@;#1zaRIN)A7qBMd)TKgGcdI^h3TVCqM>{;dOb?R@21h*y&rAt--yNBvnq8wwWX0s z6>Bn&*GeeVyvWuZ%I8LBYQd{r7IdbqGxT~%hyssGK}6?pto9Q7YFdoAw<}k$c{*>n zrQHjd=LjA2H9SK;7x=MsuRRb_ybB8_6yv_a?aZk;UIg|r+z`8^wDs>vqW;hg1#7dp znwLp1dfip{ApRUXAdSz4M4@?s2^O(-u&;6!PL*6KRFl8S=|o6DT+tEuBas1zK2?Cd z_7dUM>m9%fEQI}!)!`>+3~I5@$)CTM=|a8BY)5w!oGWUF-EZ}1Wz`S*Ih|f=Q2A!TjGIr(P+eAvYp~O3t_o^OJ*NaHX+uZ%7%G zzXifDPcxzSMOoUqEdr*N zrqfLO6gZgsl1=ED29GPI()<6G!}0&J=!$EGLdPMJLj9Ux*rQ;Gt0G6Q+lD zvO}qw?0)YS&R|;gj zSUgOc`BfOBct=>jUt1WwU>AKq`7f3KZxzHO2MHbfC54)~o8iAns@ate0b@N@$w!PgW%zG6o{Gb;nR2J)m zRwqw{0rBZ6^u5VLH{lK2rG?d z3necof=x#{z5de*WQs=$ryQ6mj8L``F4`wk<}>jDG-U39hQAjmcq+mu_hP6Yz6vIN zE2Z}o-+|?<0`L@Vgt|i=@H^@~Eyz@at)EM1%5i_8%`Y!uk@(%GR~5s!!;{Noj|xCj z_Zd?@s>|dqo`tI&rfrlt5*HG0gd0Lr!y^ps{@}jQ_X?hX2imIG6V{^SKp` zUi=VTa<@RF(_i?>&lMKMU4(k;Pqaa-75!|}NE6N8)1lV&?AVDoxHEDU*p^6u_i-~Y z+t3S3)^w2Us}sO+YcxQ?5lHm=N)FD{gLVgVSl_xHB=(wu=(7k;pC2VWSN)9*jm@T6 zy`~VU1!VQ4e(F{t?z5b!1l~q{&{o>`zXx+c$#$nIE4F+u5e~f$@sQHlKIr0C0$$3(BQNrh$*k6flg}t#Vl<>QmY?p zeEx$e$VE{NG@-_ua#_8+6`Nyd&-Jc7 zb@$NzH!j#zI1#U$IL7RTOrQ-}a_r7^MP}T0nI;Qv(14S3Ao1#UFg`OB0u@8awVm6k z_u*aQvqu(e95Ts!@1+8#h+A~$6EDu|(LoH*9}f<5ke(HLWya)+Gmh_j!OOtABmkD9 z^V>i;)pY^prHsW1Nn6S5j_q8Fl9m9rXu-X|0dTwYKe(`8k($jr{t(SD zUdA;q^F;5(xaDb6;Y!oOpJ4$64a${#%l>9`7Nui z6EY@>?(FwxKf3NR(*T~V`0)x`3sY$3%`aR`hb7jO+~htV(uRp*jxD6R34dI-!c=)j z{AuSb+IQWU$(t#IYfdF+(e8mit@BZJidaM5o(WnT{*VU&Ghk+Su|SR;Wcp8a*w#o7 z2rs$8T83`JSM%c8JogpouumSpOt&FphsrZ|_Y(H=N(}T&d`!AbdA8B#AG^J6E$)oI zg%yAGQC0e+U`WYdn&Ppac8R~M!ml12cC?U9jXhktBzhz9U!?(uRb?<;ZUVgRJ`S^L z&a=*HC2&rCAd1!VWY5~J+FZNynSN{)!Pzc1>U47rSPC+TvVA6f8?TEsS2KIFoBf3kgUJ83su=L~O{ z(>FV|kzy-Lrg(NEswTbVhFrTY<}K>!^U+CEOFoIrtoq6|SV-cv?Bm?J2whI}K$Y9J zNtU+Xab&lwbBSA+8AcoMWmVNHZG7Pe*K9CA=B)WjXDC(stePqPY&lI{3(c4vO&@g2_$o)5xcOF%S^%q9a zou%>Iryf_ReqP2*!b0Ip9!DK_rs3V?L-DcUFj#QSjB8P*{O6?$;nap2D*L4pveaW> z$ntIc+W$_{#PTZEI3t3+T>lZxB3ENvs}`j19D~aFlxWwU-?nt*jHc4i|?-JdhM>K|h@+f0A7u(?QRL zTw@pC{}oAoOyN{7g+bJ43sj4#VYxB;;K-Z@Tz766WR@Hh`}B&y?ZQI-M(!}448B7B zQt|mxBMHlYTGIuiCd20EPwDQRD#9Ia#29beHa@W5ht1}bp+Be^OQa`*WN;cRD|f=^ zmBVmJ{XEw6F_Vm||6P&M#J)dFv71xaW$PBYpK;GgWeb}>&8P4HxSnF->UR4-h zn3l*-xwDb@G^KEP2OfaWnQ5?ar6Kp_iaVyh9nOE<+|5R7Z^OlD=8!+d5KoH=QS?v^ zoHkqnP1$tR>0SVGaFdHGzk=OmRiOFXg!ehvj}mHku(dT{5r9u!@u;!Yk&L{r6RygzFe9F8>OmuRf# zuP1%t-px%DxW|`Zxoa*|9I|Fk6C}u-Ys>M-C3lc+1pe^Y29zAgK?BWVoNeoc#~UWo zb5SAuyvTUoAoLfWPkPS$#fc`;VtT3=`(>@nmuD_R_ACl5ld;DmMRu zEtfrMEmRfTAklJT`+S07pl2p$p??rvj$OfoO=fr_ZxNo#{KKAJ9*?osIaDY2A*=nc z5H|;@;H7cx^i$DqD!*Hb{L*8fHiY7$iaLBx+flmEjK+(5bA9&o3Np`rMfK;gc-QX^ z#?-!Jrtg2@;Y150{#tz1XF0y}91ng5$$VXNDxMly2(klPadv(RU42bRb$Yd7*R30L zyjVNB)p-N<*B0P6uM6yL*b(M&{2zB&EsEXyTF30J|3{(9k=$ESCy1(11j&d-(3{c4 z{wvbQ{CZg&b|aQM)e#Dc8$8iSq(|j#SK`CknS6`GWNh4Hj3@IH1z|t@sVxM8*OEnW z&NdH|Jjbv;!{Inx;D_2jo(Pm~9fdxda$0{#lb$w@qk5xLnf`optS1A+`(hAFY*51y zZF1;wy$Lg2C>@+zPnU!m3%~zr!K88n{GDGz7DktF7cPv(CzH>?4$~|6qG^!rJ=I1h zWbVVkgKc5KPgcbIW>0I~)eMX546T=f26R{TJK?mM31ONWKL z`Oa7>zw9ULd2h}XSM4P2Z*zz+Pn+z{-i4(x+fa4wXWX^6SP(SgG#u^=hTHE`$fpHg z$)?SO4)2@>xjm=Z2S-P|aCt9HsryNH9QLLQ7WQ!Q#`grh{blU!8h>1*p@`>l3UK3g zZ|Z-ih%Jn}%sL9fQSbT#!EMtxs{bsOTIsxH&!0bKp6kQegOF09d;c?M`mc=$w!EW~ zf$Qlxdnr~tGnG9L4MV4SN_f%90)K{$z`gs*>6h8LXm>st4~X%V;<7-IQ27N3cz=}+ zd6voD713hs(!B--#RFD!k0^Qu#=hl^db4jsn(VITvR- zS=si2YW3WrlXH)PWgHLtYPYZn5|#97%Sk%Ww3I9^90t1;^+5Ts31>ew8d7sk&S zo?qJH!zVsHK}!#qbI&ghvYU{LwGwLlrcm+B;uOPfUW>sA@6CAgwj^S{(gY&c%9DMn z=>qBVuLNO-Jee;}g>%f6g_~xuXzx&{`@R8X?|eYTv9|0~R3_f^(d5%joiHLomaWc8uT;pX@}f)NkDD0}Z5oFpjzWdQDpkuh5j~G9=Gl zn%+&&rF(~DvOh!58$9bZ$xXU|*y}JLJ z(b`tlrv6KGr{ti(-giHHzo(v>{D}lNB|Tg*zl+#b+!Pd^iV*Eqy}`AIuYgBm<;9w% zB5@v`G&?_W8RWUPa={$sey;0g-z-j0&!HKtgY7~4j8LZaUigR3;h$i;WQF+@jQS+o{Fr_{PGS(;vfA6Zoo=wrr&_aj# zJ$%a6KdYg^x(}Iqau~PXay2~OJD>hsSqoqH|A5h}RE7G!kEo(ZA14d+QSRe?4SZRIW^u__hf{^?f-Qs&-4lhWcz~Jb zga~FR9cIz4(cJvJ7GgV1jy`Uj2ZnD&u;!-;cpkRG%_WIYpJ~n-o;QP$ z=MbU)vxjuY_fxFp^K@oaY71({nz(hr5Ek|+367K)3G3$&p=-1FY}K9y`5v*N`j;*{Ul^mGPXegH^*C?Rk^?h{Otv>x;g=-6csYw>o}FDFoR|jb7FBo z4c3+Pi+2PASmId^Kh9UM^c!CAAg>hG6ylfUHzys#Yo4-ki%ZC}wiK5BWE8E-9}q+xj-~5YzMu}Z8T9hb5AbMX9UM=( z1hOx;k?#5W`1$t~7?w5}m36Avq`kV-*P)9pR(^K(yyr~>Z!n=misZ*HwzQqEF#mYgm7H_7oB!H z8#I@hqObtqw?!9Sbova)|Dqsg<_H@PJc43PXXv!OP29&;Q~a#LHR%n5h9WJHZqs0| z75vE5h{trII0xfZk`lV^k%yeEH{sr-bEMsR9vhf_3f}37c}(drRyrYoZrbiDs-GbX z^*vEwzf0^fe7=j>y5z7^5-C)#Jpr3FBk)!ADe`Y)GHm#l1Ns4TP$)8>=l?_Ie65Fg zzcFF6TY`vu$7|BmW(0!8;na4kB{Xh00W~FG#n^)eW=s#oQ?^TSW7;^T1-614n-8=7 z>vlrg<_xs*`oO-eEaTqgCNM_@Bjh6P(fU|1z0+fs}7!lJRnV-(G95TBci zud!U081}JyBHOT553d|M#68aKVYBYUi{2Vf$GnCaq^bH7i<__%e}1VY*+0Y?{zbo8 z=!#ja_*E(Uh&-EnBp$EIHFLAh-^G<;k64%aaoV4E5RE1!(1e^T%<8*2n#?{z{o@X@ zr5on3iW%D2H7bp@Sxdo%E3RxXaTsc~pT{YY;{W~b7v#UDyY#j~AsZ5Nlnd!p!2PbX z;9H^_^S6$*V^F3VQ9ga1PGa=JC0&WZm zqx~gSB*R%2|9)*{5_iR%Vrf0Agh7$(Y!{aCd?LDO4`t(ew?f?5Nwliag=K|D;T`u8 z+#IPemYZS6*VUDyzlS7mYY<3t1P8I%xe^!cJWNN)AYFNTC~W+)6f0~(*)MViwmtUX z)(tbjb;nn-ZpR<&VdqwwXfGl5Kx*OF&1E93Bv+f+e_L7qE=tbVKVf$bcC#B{i*bt0 z2D$mRMWIL{yjKlv?XM_#AEuG{la(f0}cu*8Ckd?xlhsBXaq znWH#VN|`Nw(?yKRoS9!#2)8xx0nYbLVjhMD@Uee1dIS`p@r%hA`*a3c8Md>@A`N^Q zPw?}>$@o6gi8-I>W(Nv}lOXzyhPis7yC@9*OjN>tEj_>4zX`Bku04o!zmbPu<$9{@UzAtHc)4RnsF1^w$j76 z!9xp|jf}zV$e`XgSFBOB#4+A3lb^FspqvuY?EAts7X4(}yTy3mwn{4b zc|7JC%*7ifF?7M9T{P9}F$p&-VeWdd?7mhM&eNFAESqv^>hezd(Kd?HoG{3OSA}7b z#v#10%aEPi9>#Xm&%%#4BXH+~Oxzs&54Gh(==U2vf}+YQCU@4AE0NA*1v)Zp-Qr{H zK+YddwMvXn-f-jm;^n}nSDtqM{L7>q@@QOOJzHIUi_PtBXI=I=f}7QMsOs8mnt5pq z_;$=8Z^Yid;RAil+*927J>H5t;4-zW^xI9gxX=mYi3hF?Yv=rTT8o0F4dousY-cxj z#9;M{NqlQb6_xzdM+1J}V!3ITA^*%%c<%BBzDGPFOM`pF{qhpR=W;v1!f7-d4mX8p z??|vx&4B6uIn$+QbRf8{lJ0w@3hbN{=HCB@doZXXY> zep7Iy+IA?MR}M1zmgv&6o{lW=r*`KPU`S&l^gl=gJ4YpyIo(A9Iy~{K@^H91>muEK zcYtno{6aguQkngjgKT$_7rgekE^tfDW6z#B;pT0AP>}hK2xB{Gl6*N>2}TM%$K7S! z6Gu~pC5{j@YYj{@=>pE|8mNEyNmuT$gb`BNFfUSqb4fpjU%rJ?&q?|)a_mKzmKp~t zpihgh>Cu-G3t+aB0hK+u0BQz9Mcy@(Zp~MMGKsUa*zpqFa}E)<{|ynglqRt=8>7Ye zCxKc6DZK0dfoi2&(b-K&*j}YVYf~oE=642^7ZrkSz<*%VT?b#2UE!(ybejy31x zfu5-*RV};0-pr$9w>M z>^{p;Z!KZtg0nO|WHSuY(7@Bmx0pm@1#?LbrRz@1!Q0brpnN0;M!b)O%%Qz>M7oiH zfA|%id@h1^&7nd&Us;&YQ$lZCQJ@cF-ofI762hvzVK8v9ihhddWgX`K33@e*X>y-G z$v3QmmDh$)ce)8o9E4D`*90!TRfWMxqc{y&ITUuR;Sz$fN%4?H*fQ}M%6(R)t^-H8 z85ds)th~gY=Of{C^QsB3dV?jfRB5^SDC(8vQUo zi+ab-h5NNa==@hG-VaBBROMfKX2}LB894&xl}!+KoE;;4I5`;BOb&wDQ)i)i-D`0^ zRWu#r919WilrbUJ31@B>ck*_Rr=GXNaOI^Da9rDqHe3Fok|yWquUF^UNb{}K;jkOH z53gnxV~2yT&mYp-IhP*y-3#k2yI@KFF!=g@7HGfB1g9U)^wvZN+LY%2)v`Be!JHoM z?)ht&`#Kw(c3^uzMdA{Dx%fXQo%=PNKc=LMx)(YsI4zcH=G&^6WfQu!nFBR z)jD078x&UJ{^0p!>f9pM>9>qLme@emq;zqmc`9ot zT!FQz8!^f&7AL>z7JJc+z%EV}e{{`3`x}ednSuTIJu?h*cvJiyF%z#!R&fgVro)(^ zZ-n}}u+ER8aVkF>L-)R6k-K%V;C>Rz*cDE6O&@~!(;B*T+f4RdR?ImirtlaMarf>BL^OC zU<)fEamTn4wz~fZo#T_tZaV}r)!A9}Nq8^2oGiu9@Hv7OnZtxTeSC4;#Y{ZBsgo^O z-NkNS&O_nF5SrF7{xFcjanXtp=S%}B7Ecsazba}%o4^h=hn#DYkq_K^w;QbB` z`v+gJcWWnO(me^h5v&Q3AHCs{R}|~p=8j%%7FeKM%iXDX0?+=QfU$Q%&{n|~UtKfD zo@yCh{q|m%tFsvQ<)6ZXQzOvhg9U20$?<1GUg2et9`CwJ65Atcah_Sd;KlAGXgu4J z*^Hk-R{5>O>hnhI_8-J~tF*Ch_!mY@6WPL!n|L{J56(E?hBt2~ip~@bA>K2yS<|F9 zSTX80IDR)5uJ8@P8mCNrlXL`CRMv=Fw_tj?A9Y)(&C<$j zsNSJ??$2)-oHg0;L*a-?8%CaD7Uo@-Lul@ z=_Yxs+J1)iCTFnNUqZC9ddrq6K8Bo{4(btqh;5#PNH>1xHq@tM){ZzFb0-oGj$X@} z<)d-1vYgo;tzqhKw_dFUH9VPm4u|JkVch12BwYqr$EGv%`QTV2!H-zbO?im-vO$>@ zB{n~(iclzRLZB0}&l0`S29|Gs4hw$M|4L(-J z2AE(gNE>%qpHrxSQ+5Wzj!H*3rn?CO&wheOuf_=Lb`-&oZ%1Lz>@3`>W`=4$8wBq( zYC$&b4Y~Rxu59R3Cr}(EuE*?n&6dya!`YAi&|6Ld*x!1Lq^Qpz`^0%Q!*?0MW2JiT z*kUj28*YJq28rmnOCL54)Pi5$d|X$!5})zeOfxr=jeUCn#wl^sZMqETztD!U{>^Yw zwv(30s*$<#_OKP(BDjxbzvwQ_^H-njT}~Itj^`R$Z!yQ;zV!EFJ9aGE2O1`PWtQ(N ztoZdB+*NxyEL@`lqb7^_zq6jSZ1sIwuCBx!#!0fz`^%{D!-rhjr7bYFW(|v#5~F!0aN@&S=sBZ@waa_C z=pt=6YRo|QKRLSWu^$e0#<0h+;V^!#*t0jU2|j5l3YQNL1T&Kvbd_lV@NTu>w!V!y zEIrPOlUwPV4IXIMlZI7SrejF*cRJGYJAEu{U_mG6!ergE)Z6bR$P5Od+SC#;Ru&Bk zgHkv`cL6Jq<7wMXC1ICNGQ51SUYy-zgF#P9kRR!d-I~@o!fGwq`7|Dt_XOaC9|V8Q zK11?;N|=))LQ&(SDls={-K-28Oao$xZl8&#hbX5 zWs`8VzclYXV>w$jW(4S)Re(%S0a{jlr{N2pLC51xI`&)|rT0ug+BXEA&9)0)@_&PR6mSTX z(tIK0F%Ko*zf*x<7WD2Zq#s^R2SpE4dM0fEB_(u=pZHAE@NeBU-1~;(?d!%EAxqC36xvS4J@t-5c1MG6}~?|6*%4 z&x5Z=FS28Hq3|pv2ZCFZ%bq-b!`${;;hz*qsOyNuJ1Tm7qLu+pHvYtvs~s3`V*uMi z%DAkW8ai*vVOX+GRT$KB4J2Rhx+iUul73U<*i#4hrL5?`J^=ZRv2; zVXXYXY0y8M1&&S}1chY4^5!z8yY@d!{&AkW>)9bvQ)#4ynoro(k5_41gDQm0)x#1O zX~CHG8EBZ?%eMQgiM2qHuv9$?Cp5e#KHXbb#>?$&HLuC68pT@bRgE;dUYskt+>iyl zf64X!lEl0FHTX9FU9@E2I&99B6-G(D6rKH7NA;@Ppy=8o)Q@(~Ko1tvYQY>0`m$dG(hQ?`QA^WdBt<2j()qc7{jqh6a`e-AL z-P~a#V;l~)b(_#kZ8=?0P)5eB5x~SZ=`44ZI-Zxf!4B?v$Z2osXL+{!*~>n8_FIbL zc%30Q<&hB_PIthXrYTq@?hbt3bP*husX4Qq<3S9Q`BKvUX9QBYi!)AXCTzya#`X{OKFG6&A`?sey0|Q4S4GKi?&J-eIyDbHHYcD9 z+r(aPsHB236=Gt9{}0FZ5(puVynvn;W$9(G>ok ztONh~hdp}5$KZ+n_4w&nE<33ah(l^7qtu$)us>@aExNuL-)7{H_)>q|J0*iPMI^yv z&kbhJEN9ebDf&2lCa3L(;GapqX?))+uHgM8aM9ig13_nCp%`bj`AV7A z4G}J^ybB|0INIeL2;Y3I%DRt#<-AS)K`{q?8#PotF}pMrN2|_4=S#ZyQ8fo@f6QR3-xLsyS!pzGw-g_E z_AY)A&;G;znL_OmaV2HjOe~vrldk7(Ld?=b&?Kl~jg!qWc1Ef70pEiHTrHmc)D(nM z$JesXgOpa!cA^+Hnv*Pw2Z<-w!M7|2BlQO`OF94sd+U(f_!Td&sKW1rVEEWwxFgvK z+yCT1<>W}%m^%zNub6^joio?id4#;%8VVt$4siWc51b8&!FSCjd}qLAH1vpLZxm9f ziJJpto^od=C-0=!vKMh4e=guT+w*9htjd;)yI8&@>4Dr1V>r6zFiePHTStF@6gS37iA(hJV24GV$bwnj^gL5$^y)*5dZ#!K`^zHdvRArqS8UFyK`t_w~>x z?ng*Iy?LPsi=4x;S&TjQ?lr^?X%R}cbmPzHS`2btOuuRFg9HDvZDu~%3(=q2X<6_G zXqsLCIq{in({5F|;lJmCb5a_pAWlmwSRR9^TaKZ>=U&VcoaZtoF;w>ylAt$FS;g5z zNLT)IRV$uGF*6O4GUhh2%l6xp_SD);o0yAy6nx@yh1*3=tVa74ZB;)``b1Xr{j4pd_BO-C zj}Ya8%!J14ynu$!Wucm5ajec$+;hd3pT1)eo1-P3BPwNVQWLk}rRbi`LX$U zI_fKZH&V!BNeTA&rem8Zn#k&>uwEl|STeDTy@B^29ihpdxJmPtyC376ms!~UuaCa* zEuyZG(IRuhE)X8Q2XmH{GjI|2l((NDTNerN?tnBXXuiUZ50AJ|lRTL3*+l(TX0d6m zF>F?W3MjkV2@KtAm{mqB44WDbi;@BKitf|lX_;iuH<4^`?E$ZrPU62P1Q&>rGXss0 ze7nyo)QL_d&E<1&X#dJUN6>8yqp=<4T<6=fFj(=(B~Fve8*^hg;yj07J8# z@Xf>k%&V@!@sI0S|2q}_+>rhBL#sP@&TIh}cRO72XbN{z`5W`ezlqOJjppl_GCw=c z3U}EwVN~%Xd=$R|$E|sS+m8*w1;2o8GQDqH`v^D5cxD~T-8#*uEmYSmT#}rZ3%m1?x= zuH~Y3OEBH2C8)Qii!Hu?foq*=%AHs>4f9j(a=Sn7hfUYC(bU2LlaD7c<8?Z8PoO+^ zloiwQJ@YtYw@9Y2;TIXTb2H-~v~oYqbTRj(Hu`ZR#eIFx=;$H;Xwu{?8nMJftT9QY zCC}_I+o=s}G*OJfj=+N*f$Zbj4W%Hzm0dN_!xLj>CW{dN#ejO-beLbJ?vAitb z5io%-*d&7zi9`8C4slH5(K)p7KFfx7n_#%EJTGJN1}_(?^Diarc{7=re6HC6)cbf9 zFVDC^>{A5d=Tyvu8x5)K!7-fB{|GI!2}l2#3s6G)H2Qm8KxdVG3~e5;eAj6>YIiD^ z_eBM#l^$jF=L~SZ{1Ci4kU$T|veql-TtU zznk>p-DH1uKllW?XXK*V5F>W%qBaQk&q5JvVcE9Uc;KWz#&l1?&>Pcn3mb{f_q|a$ z;}$kLhGDVNG4@jY+#|A;F}wUFEx(Vfwzh#i(R;;gClV;LpM^$E5%{R950l#(ah&&F zaHpl1llvI=-_FDX`Oi@&U7Nr5Vvw7&ZZW!pwm80fXDRQwsF*JbScuI{ z$M9yxR5V$NA|;PJ+F5b|<3=;aG!$9OY&B@SZp*f-tFfR5UV@k-fyf_2<~Jt??W_V> zNoO(|*PEj08GkNo$Z6J@rilL{La|RvgO3{gi>3ziQ6(Y^*H-GVrJ}ib#8969;h%!n zx?0$|dMW-H{lg9vkD>;7_wmLkA-@)!`MOo1{G=PhdE+xFq7V5S*p_eWG2mbgHk7FH zecW(NduM>-$X7I8FV7E~7>^FqkKozTjaaIni#yj?GIX6!xp&&^Yk~`Ecf7<+tK@On zYa2ZHNgMC?GW^$}%1>`F#}9I+L|3!rG1PZDebjQ0*&Q|G7n>OIt)CY|=xZ0Ya_`X7eQ#2c$Fio=vK3WboAArwO5y=Pz2jHE)!(4ZNmfiy{(DU`|_ z75ZpYA@4o=64Io3kOo7PR4Qp8N__Vpu$Hynb?!NPKfmV}s>5rZuf>x>FG7XpEc`F@ zFV09bXCA*fO&ge#*t&$n4eweoWKR{vfv=;k8r};P+#~e`YCSgoPqyEyE4Y$fZZCdO!=7@ zR7vc@v1^hs%b3Aq%g>-gts<|PRgPgNi_vyK7x`*sTwJ#RgX`{N-^4iV;ZNdqO`|@7!G-C#RAm^tcCClTn5pDPY!-bKK`=x0 z0J>*{q3M)n8e4XnUf(kyk`VYv)jCsD&MYJ<$G73lq^oG`q>6^SoX|Kl9Tmq(V!cW-mf3j9^XwYVbX74QB=4fBYLi z2pKzsMD{$Soe~kyINlf{^bU!KM9Gm#DG}IcJ?8ZOWYYxcUZ%pnow*<-h0h;n;?Bvc zv~#vRx9OffnRD(ZY5nO<iPv}cp{DDd>jGUMfvpa{ZV*7?+DBtX9){t zou$F&I=N|!~hKcwG%K(@!K7xr<>@v@C16ezsUMWC*pd`bQsbtEC^V zWx|}P(q!6))A;Irk}#L8rG0zFTg*(C=VOEV5ka7MrJ>PhS*O z)w~O%q7Ase%y;horv1zU-#g^ho$o}ftk1d%{KZhA1U7HdPa~dj}Kltfkpm% z@k!lbplC4xcgrsq2k{)8t0r_C6y$>A9Vu2VQi?6+g6WK~9(vnz8IFHe!#!2{O@`Ob zhIP($wA)J(j|*CbGlCh4jhNlf>!DI+!r` z1MS_g2x`4X!%d4679CbeXam6vZ9Gd={`+9?hE_=I|wdg z-Ngn^E{i(ajJQ!jEo6V?A|fTXkqOp{#V<9F>8bTf;MOY*kBp1KHsuCsIxh)%IfqF~ z#w?QbbRG8HoJ`x5o5}0S2h4(RCrJC;JM>S)H)gF3lK^CsusdcM&&y?{>Hwu$a8o z-OF5eImF~vM{sv9w=la3qj6cPEDRqm&6_PQ68aEBxl6tsL}6hoReq34Eehs|UdTM7 z3XV&tYDN>CadQ%m%p1eZJ2Vmk(ngY5m1?l}p+3wxJRO{Dvu##xh@qvDtLdnZPK;Tt zG)(V)PY?Mvkg!k_+PZX*4#Z5tgHNpR>~R@7^t~4KcNK{mml?y$esdsNMnW(tQLrKavE0LWL0?gaZ{5ri#}?5PQ+?8uQb{*p z60SI7h##|K>DkM7Xl+S5ZTGOl?`=D1=}UpdQjsENR8QQtDV2$sd6HO_|DcPTq+wRl zL)vP4nJTCa5^gY;?6{*$MIIVtU-VBB@>rP$&7DRcB^cqBRzE5``2@z73VT>@V)2w+#?C?R_8TOU9 z^%%pEH45On=P>D!QKPbB#8l4oE0Jj#Pl}G$ppKR|9UhxP9xwbrs*av3`zx?4ErX+( z==5la_!I%JHtZl$uP)QMhKKRAzAog1OoX8OYhdWK3|JH|MYJBvLjQt#GF)B}?3Y9e zSr;+=kaCAyKC}h+MJA|vHHxlOmt>YNpM|bA$x!Uz2Ohr~#TOf1(cQc55&uejUeo%0)50b!@4s<#_LiTo*k-eqz5U_kLRCMN(M@!~|$n6M;PS?W;WAZ|%oiuSb4-qcmE~(gWK;KP0NJFolq)Jm`aOTn#_|T(-F5Q2gYK3Gng9;Ow32$1- zxb-zmy`k_s&)~7eIG<+hPZUieBS|s86j#(bk;A=d^s#}!O3PJJ{o|P=Cs8c;bZ!yD zUo5U`pDX_FU^G2*MGX#qtRlntZRFD=ZLnXcF0xBipe}D?n6`FZy0tu=%=~c__Nx*w zPJK*MI?mxMUW?=v0)lkl5_b=E&pk*bd-<5>f4jV{c z&uz#F(_^b*4cL@}GVGsHW!5@>6XkoJ)6`i~y!GvDGVQ4&uHL4Lwt5XP5J+K+!gIQN zvl{gp76UGRlB`plG`nl-LiW8#l2uE(4{wfr6lOR9?A2{KT+W=KxU%mwNg6+bjVEnD zuh_GaA8H7*WF1as3TTUsJ6Dl?lWgd{M>c3JVfTEJWY0F%0B5-qZmNnP)A&EwYl(1O zT@G$1E(GP~0@zV%!p;pF%?4V#GTQeF$lV=Bp~+$snP;jbFo-R{`i4V?L@e#JKTdqy z8_8&4PI6Yr4ut+*Nxb~ah(yPGNbgXCw?!8q-bfc-dR>AnuTr>@FUv+$%pk+|KOjBQ zK~R^ML!J)y+LUV?$0z@l!{)6LAfYeZ4R=MMo~JVAl^wvWn7gdl7!WUa4FA`lr9;dpOd{{>rxj`>UaYRq4m(K)&;NU|Ae9m&q(N$BDhof z051A9L10-l_Ni-Y@2}Y4v#wN6X$vG;$$;aKt*pc6 zWcK#(#ccZ{4>qK2BWq4xv&sXPAaDIJA!l-fZe2W^>^@}&Q(lEa{Mg^L&9nnA%u&YT zwBsNV{{uYV-iKwO8tjS_SIhcL<}ur~h#{Z7K0vCj>0n2(B_WhXmB+xbkKlvO0?yfha zuKx&o0!FevQwzah^)sg8 z*5mNXQh~MYpD51BHluRw&aivrCe$BYfYT!$qv@Yg#$%g5&hXIUUu9ci?vt}rT6!`0 zY#v2*w#jp6k9)&2A9+}&mIFt9azOOr3w)ZijIB%<%3cZrviq1KWWC)>I1O{~oH3kz z_|cG6E1O6r-u3{6C#|4V-^Q3fe+qTJ+aY%Lc$jr;B3%>7lIkRFs`;Uj`Sh5lOHK`k z+%wbJDer}|BPR>|<%U8Zh14sfp!RV!Z4%m*R(?2y z&aQ-$e^w1SwxigsC-ou8QH3=ZX|avd7qJV2wy|5+RI)M&HRPcefcwQlCjH)6w9Y(5 z=6@-LC0T+GMmh$jJc+^A<5q(~)-Uc3Eyk618_0^}Qab5EANjzDY1C;`QIg6ua&hKC zC|Hv!pP~xDo#)au@1D1&-E^)V+TL zkCKltHJ^s#)25lYc}5g5Iz9qcJgcR;(qmz3{CTps@(o!%wSg&oAWORhsrjgeTcGj1 zmE?NgM-RIpxO4q;dMiv0TbhfYdCX9@^P?@dR&70O`JM)us|8;AOE!)g;*LctYlzFo z>3rxxNer+X52t3YL>=jgv{ddeJi79WvtA&NY1fXTy{9dxq+3BY%hKoKiKw0T5~_CA zfW6K*44s(Cd|h{vvUwXBi)DiI;pznV^Me8F$dT-r1SwY1L;=4wHPcmTTj--Jlc02K zJ;Xg304)*=U29)Lvz7^*y4s8t!W?s6hAYW)4yU1V>C}P?f+=}vw4$P#i>r(S>3h%U zQ}Z^irEf0GURnWr6V=!R9Zj?e0dCn5cam-BN$$MvgR&MNzgZ0O(>16Zq_b^r zpTMa_W_{^x&OY^tjZ|}==y2%~TDVIl++^tZPVaWzAR|TUu z0r0sfNzJe85wn%55FR@Z++>~c=C&cQ-G3oncr1uW%$8&m4EmwtqA_EgmN@7Ix)t&L2|F*}&FZX}Gd8%J)~PM~fVjBxVH zJFsZ%81}fa3Olz<4+e&1;Hz&2IN3M?W0VVU%!@+WRPD|!0&SQs_{*Oz_y{Ko%}A&7 zcbMHF%eGC{z}N+s$f|G2kpK1&c`qIZXCF-kN6B71_EHJvA3p-Co_rvQTg7EB-w)$9 zFItXGdxT!h)iK1e@EbSL(FkuSM#8RX-h$Uff=t}44C#aEOhm8_Z9K4>%>3&`ECknC zQHBdDy2X)RdpqdaRtb&r7r{5Em7|d*WRA)`(bkgdGw#O1HUv4FQqHRl#KHvhozPdId;ci7ycNM4QX?8S@dA|---4FnwV)hn#P&F)f%K|$ z^6c+rist7?i{feGUh|g3Mp)7AGZ`3VQcJSF_LE+jHg1hg33M#;hbgX_FzH<{@rpNP zmTC8p$QF5U(6WZPpTCpdg%WJK$E2ro*dC&n7%h=ZzB#VeU+$ciwk-Hm2@DmN}ap&!ia2Aiy`TskC0u`BjJ}w;C0fU z6Rjs9B>zJnF_L>~tu=nO$aI#1(CHz;ierD0CDU!G@)R{3mRH67R#k*6&EXIdZvgra zZZpy;lOZjB33PqG!|fX?&y2q{lk?6z%>4J{F|=!Hv;Er*;E_3${jLwD&*pS75t%(O z_NoK>FMK)cnLdOZsNVwmMMuE)unalc?oFP{s>9wxHuN#8{? zveG1j=@sU?1_m1;XRwdE?GX%*cC?cZi*&J6`4Y*we1oJZNnxq4736yhjQOeO#OLpH zswzK*4q3gI%bK>9yPq+Exol<6{OT5*m&?pxRjLdeDbRz{2Mlqa=q#s2chaI4YtYpx zf;&#s1((lbdVBK>cvVTr!Dae5825$ttk_2v?2{D{q3vkKZBLw$zDu;sYBPD!H^4m| zSwia{&%z*c%6ZXQ!gy0->`#fjAzPVz6G zJ)thiEDabr46KgG@i9L4NtI+MiFN-$7ToW`g33Rba%3%+2zj)%%7sdNQb)xXgR~+( znVdQKmb{-Ti3@yk_{Uy1d8JdM%h#kVoZMgC%2Wz@oULg^bZC4$i3vN%@0Zo* zFO0Cjj`Gp?$kG^9|4l@WvqRlKg}7z?Q~u+PKK`xha(?yJuc&GB2@{sDrjJe?!!!C@ zaq7O`IJ=VYk#XL*@3AX$|FJp$k=@TLO-RExH-#z^=PF!u*pqL&^&j^5q+wA1dD^9M zk&mzL6S(#k+V0yp=5raFsX@G~x#@l|Ds*iyfqsSzBoopYa(QDMgPYMS61s2a%^>G$v^ z^CS2r^U|rfJ%e1Bn89uI^8%%OdxT&qObgn9Qymld)w3na2jgRT8rQ%(9hE6x`)mQP zbK)DFY{v_`Qh^5(oJ8^{gNGk0!v~FgEL=R9Ke+8D>clKX9oIM9DoK4>)U1mVRn2tz zOLe@OBEw5fcjLR4>GG-0<#dmHEp>k-DGICp&Hs3^l+V_h3hq^h`607`$Id5=$D>es z+PjV(kxrvum23E4b`89F(JFqU@^kLsR1;e3y%y9xBl(-29lWREAaBGelMY*g&T)~v za@|7u^FITk*Fw2_9HMt%4ypy&i1v?oOZrB5EGaWvbg0U}RW0vuEygdfi+Om!4aMf6KC{)$lr| zSeqrU%XJ`PVgbFdQkB%qJ4T*d{f5%xwxj&JI41p&KDGNDL(jg`!g2PlaJbl)wjMu- z=Gh7O^U-K*;=iEp@9iS_6;r96jseoHVeqi{8G%nbspO11=+_e^WQ>w&1^7+C&ocL)T&uSKS*g~zBbZ^?T4tk&UU&xybOOg-!3jgNSCk%66#Gf`}XvL77=*i5+wT=2{KiOUI z=Ty^+Eyq!E(inPT(ll<$?Ih;L@TDXucN(q?Q^al7C&`rl-850Y6bHsULu-Wx) zso@OO(l>x5d$yB96BgtBapR!)h!SS*oQ?yZrsAB&7(UM77?m*_OFFNQ4O&(Ov6UmB=aSI^CweVj_Hoq=cUS?v2e3n$t{ zV#n)wT-%ZF=(ooo^KT2Bzqu8gcR&s=PfWpu&pYw;eIYxkc1@`Ka^v~*=~%0j%_XhV zr}VQ4$~mjU*S7JvXWR-LC#g;%WBgG5a}zba=|feV8_2-XU;Mt)5Ba+#-}%g8ZS#%I8?OHxiTiECnN~9?RCqlNcSW{vIyXCQ!q=Xn(Tz*-msvNRnwd%K+F#SKP4)Dz zkqa_>4;DTu$8jIKx!V@kXszxTrga%$OQi~?Usnff;{trJY&O4q%L05D79(WXcjLLd zCd?{IqPIWF^6dlJ_}W4Vr9Ja-McFYNwk(BTp8N@07T4hWrWfJ|bs5|lHC32jFo^FO zjDZG=kQB`0MX%lY6S+3L{TqsHK~FLFkG{b0#$s2nJ$Bp~$v$6w2H&r(eURJz}&(1XFmCX%t{-{Bk7vG3?oR8tyjrY*4Kc1K@GsAf&BZ*ISDn<-yl5Zsm zsF=}#Wy7n@pn#{v_KN8{{IbFs2M0hevc!lq3^w$IBQxWdzPifkL2WL`n3 zX&{1!h#&ZpLru3UCSf!!!;#d`RNooZ2A`_c^c=j}^NHs9cGDWc zOR%S1mFin>Cb5@}(e}6;tQP#yX^qcD)ZskLRCC| zd zS04%iU;dCM7Ly@`duDyN=!x*V_(Fr*W68ftAACK`k$TV2rB6@3#A-cdWaBg41QmTTiNFRz^{9j-=$MZVB<)tcGwFpn-89f4kVx8lc86WmoW2Dk2Q zrZ%SEs8f`;Xuohj8QWlmPet$Pen`U<7y!I#n2n2di+l2I%c+C!3y%S%w_Kw=7BO2wa~+uw%`)R+#HW~V?(*YvTp9N=1{zN zG>=mVxIxeR9A|3UtzpsEaxN{eUo>T6FS%JHN&UYqp_ajO$dQ|J(5!n<$RXyC3v0b; zXVezzb~lex5gvV;ga1-BhuM&4bA;-=jH2IGl+x#>lBDCSHd-uHB9mmS>0sX={g-RY z%^LNPdX|alVb&OW4U1d1d>B6HPc)EE#*}WqI zn*ER9=zIytm9HlkErZCgF^A|w@puyTZLnjJc z__Os_h)tFf96WK$=A?EiG5J0RW*?jh>OJnXrAuJ!W=}<_N&Q^kpLqfYXoOb%ny|BJ zh4@~K6?Ok?#`V{orlnB_nVb3t;PyX`5#JMLBBJFGvR4J9OOA00zOTtfx7Q+F)e=$M z8Y#55nMuz?`k_i|5pHwmx#{Ky=!*FwVmLE^p7O~g_r9*9XOg_Q_PiXXM$HmOTyiA4 zr)P6Md>0cqL5A75-<%!=a3R~K5|CTD`G%|6@P2p_WbPXUtnC1i3#w+u zd9J5A&kSLX!wWKCm`AL$0Unx3pj_KbEXc}ZcK42;ntz%ZQ29g-9+9Js5+(G@$#FO} zZ439_{~~R$+fJr6ZDvLmZ{*yrv1qyS3)Od#A$ALvK%qw;S=BI-{=0G*FX^5}&1?m9 z*^-Oe>4M|4xI*Y+kYG{<$I;1W_A-0iGAPb$;r@0jlHIkVsLQrRv?#2bF4%C4SidpH zfr}4lk(U*hwfj4_Cix;05dK@p(M^SUhrK}A?gqURD2ww#-Dp`+6r}jLP{YtPGWTNv z&OYzUhi7>5E)7UK%GPq_ZO3i49=%83Z$E|md@j-4rf9PK_bj~pN|E33-H;|Mo`fIs zF2Q*rSGvFd7_JNU#Fo<-zb{!RFTjfc}?S6^_+7VcOwU1-2M#D5;f7tj$lm53U z8XLN@&^|>D?by2*`KO51-58D+W*o*=H$80J^@$Oi3wd0Ly}0u8Z>%|zLY2*fx$Qqk zGyCc;lANFm`1e>D-Dxj_*M?rfms*ZwR?`U@)=&sm(c{SR;9PF&!WB3eJ3tkmZV}ue zsa*W=9pv4QbfUfE9C@=OhLK)ZOeOA}LA~>SWc#}iWCJu%pD!j0)JhnK z))_SL`pg}c%_C9N1O7b;goTMx*!_JfcI`YT##aHVSXmDY#pQc zLkidM^0=3Eg4FhX811Q#3oB)5jZy-gd;c+fYM3Fm^t0r?l$3E(63Q@iQY!WOX@Vn_ zws5ic&FH9tNjPKIYr4GRj!ob0Dy}u|3Hf^BrQpW0;p6WT;MQ&cjnCQmHLHSJtAo=dwzO!JU?qu2W9sS zC#!dw;@?s?yt%9d&o}7vq7>ga1m+^gZ zG$yv(qpth?=&{VR2+9i3+G~umpI$-UqZkssx1H#g6%*MjYUH8NU#@r$$k5;>^y*UK zbsPR7*Y}BAep!`>;y2QWrU7~?J&eY94PdswQn;DB68WhX5ch5hnHzVLs>uEz=1))3 z`&!;uTWbY!m#Vl`kBY>%J|~c%-)2;MM5f5xr(xtVk6 zY-8lC3bEUGR;4vI$-BekA-G@rA2D z4K$I96{mk(i+>Kq;M>}V*eLmjsoH)38@Em+Q?F^_d$N)l`2D#|Wx{SO*0&T{eVxqB ztlNHdyTUxMZTrLg6n1Das}!*-)35CM@tagqs}|n=8x23y$K$Y22AzUOk-qJX;qWa>3T4? zX3|rdSN9XbK2fkeAVXwc2`tUVCq&Iy1qY~P4v1v{}&0MtzDjqz8hJ|i8!kouA zV>a4_El75~h1xg!wlLp{+)q zT_V4f&D_(2S8^uPwp1BbS2~dE+dc>G)oHVa18!u6xP=^AG#ACn@wD%QI=*{h$tGnw9X{9v(jFro&qEy;eETl^s{I1SSz8?;T`H$py~OU$xMDh>Mfk5`hixQZ=+W(Tf#(9F?IYmfvfacK~)UYK*_cd%9pFMlbs6ansHKW zKye~lfA2a`ICX+)=ryNT=VdW>bzjn!*l5h%bcv{+m+OC#Le;SYCTT7nNlhTLL=u#>} zws^qQ=vex)#vgy@2u#7VTbxS8P~7QmjMn+n`EPI6@sSyOIgJ&6sOX}r+ zrNrh@4T&H$zxx`)6{hif6q|((Yh#$C;zcSa->1`SbwGoiL3(L9eQ$jg&u%ZqqG&lL zf3=8M)f^Co_O#FgnsMkNTb&(XW%g%W#|yXZvWiMah+WT zj7}0VcE!?s({?H9o|;TeGMCanQR6Xc*8?(IG9AU;1dny)qL*?HR#+oBT&zTnHlJX2 zzjwt*l^k*^!HX;soUbL77pd&n7`lynPBzPppaqVtob7pMK5prE40x)~JPC+LZHGa~^>3Cq457g%r!K?KNcMfQ#~_ z_>8jvBYLyYcftYs8{G-HJq#D2gRBS7jfdYHCjW z^-IXRtqlRkFmqfXY0JL}UJkEfvLP~eJk|B}gi&KH;L_IuGGF33RWmrow0^NeWk(I1 za<>SD&OrXj+WXuHt+D*cM;rLHIy&4kg)h|cL@#|6c^`;lZr#8fh92Q-tvYqv`cf~YNFdxEpPRIa*%RS7C zt5->@k?@un_6qb{B`{{=1B_-RATd$MA%`VlEN743x@YpC594|F$Mb**dCs+H72)Ue z4^eVQB@No$O#(`m;=Zo!IIbX`MrK;`%?G|Qn_cy>^6n=_X-5pT&Z(fUZpJ|LD>dH1 zdq4l2pFocNc?Ff`XKb1l4`qvrySd&?@wgN?J{d@~sMq5FG_flH2uZ??nY$teJPa}alyNFuuGIGYDihgg_<~4^$61S2|v?5p= zyVHbFYOfntyQB&OTmr}&jU4Lp^CcDsN6=?BLNz05oAIA*F8GOLiN0bE6iR49 zj?;KtvA&nIWQ|~?zeZv0gkAV#nFi1;!^-X#oIoqJX|QAAAsF{c2k=KCtzp!0?fp~G zx#|7OGGv#fu=-Sjvv~qru=tX_> z@B9-mGI}CeJZ&}zyF&2K$G69dT&xq#iOgeF22<#ejnIm%oslVwMOw0Wa0YWA| zdgXth8{mg=+l0Nq`3$r!N`|C`$!uY!H=ALSgf!C$IkgVXV~r{?@!Nx!&c7vV&L2gi zzf}s;7xz>JRCvyxVYV ztSvbsxG)xnUBPgp_h>mqhIbX_@Y~#{VujI0c7mHFJ2bQnW0XE)-`&5U>0pj#OYSnd zsimNj>H=w73ZZ;@C{EQpfTI@s;_9pY%>1iGm^!xxht~?8&pKWDAZ9GyK4r}DQ3<#+ zz7Krscd}XIq+w?LA4tg+I>_cq)9yn?M7h3Bcn5ss<_muBv->Ym!8;45)q@~^%y5=G z7hy{8B9!afjj5XjAJKuCY;?EYax0L#>7v0dH_Ktc`!&2Y z`UziEhp;+6nsnpp>!QW-`{7Pn95EOEra|Q`xH>5hBC9Q7!n8tCI=BzN=G7zrfWfQ( z=Hls>D!6z-02no1DjQM_LKd%<1_y_;)#neefjQ|gd-+1_xsk-a-}?eSWO~?GjIBHY`xL!_S&xS0+>MIA)rf4m6ZfJ*+ zd_CwGnFH>Elk_lq7*~gDK-5h!PRWWweW53V-P?nCsxR?S^l|J=U@;&tik7vP#Y*Fd3(GjaNGomTNjKvLLM zcFe8>p+1JXua!ps*$dIV+6oVU%_HBMtl&1fDaW6Wi4>V0_~XS=Z+)s^8^C zx)&P2=I}`7=Wh{LVpznC{F=pd?X!lN!z4kYIGg!TOW^5J8%Sxg4#J^M=7ZTqIP`8D zH!XEKEvtUV{c#n}BF{CzdrcyS{X))e+F@?~w-S0Pzz{ah?li= zkK*ZBwdbO_l4GdZ4po}|?I!N7+lCJ1ktEB*9xrs8qUV3P#D~@~79PUfPj?-?vHO$w zVz?W@=tZn!QI>cs4`zQ~lSq=+}PzTlKxZc*pV5)8Sqmh%Wy!f`@w=8@TD;PEyc z8srQaQaQ}=t|~ZrKMB3=3A5hq2jI%$MDnV~gPSy}pPIIuAfHEEAy-OXkc)jHnl4Ib zHuX+I=Yl14Ote164bP@KtrkGnuFbGAA(76$-H2IfPw15WCy3mwAGAU;5{%RXFd}dd zLZ1!pF0+A~%W_HNI%BTyW(?_AqfX5Am%@#r(J)kcE}R?h0a6wnR?HK9O?Bv$ zI|sVU4gW4NmgUe<@y%vc(MOtU zCkMHOadb@lNz6aE5u+`St*%CNR<`#HpPlqYReax)b zL$D#!kS%;Y8IJ18lh)yRG@;IjyuIc?Tc$4N;*VsJr(d7Yg*|JiQP49oqMcx{t1_;- zdj#ZUyl{M?6F!qZN9N1FAzPY{kvb(iDqnL7pGHhGcW z>&j?rzsJ6)I&KI3cvKdLl)a(rK90hsy$A8C8PC*SoDMIZJs|JKykwqyoCV~tDoyP9 zM(a`v>9VQ17<*C{x}@}gol#Cg1TMnxL=yD9HGr;|bM)1;jZ}a|MEO;d>DXDLG0^=D z*WjRxu+@zQjrc_*)_dTMDq~2o+yhPuN2vDoi{#yvd(0>!H*6kbLWbf9R7&;JvbK4Z8y@|Y0li4xE)NtuPcrCZbS@Zy|}XH1LR(F z1QrQ*`1F#m^dTP)u4DGY?RV4Zj~Rse|4AfIR^1i54=<(49$6$gTj*OBT*c!PsQ8C{ z1M^=}7w4`o=ExZvZclX=p4AT*U$}ajlRm9YN;D^sU&l+yud!2bmh%O=rNEmu3Vl#M zu$`E{ZQvHPI*@c5Ptl?Q8Tx&oM7&_A19klDNF9EJGs^=ym}jzOqKw2joO9hD=Imyk zG57A~T+YXmqQOZ-xN0()@dm^$x{ho=f|Pu zA`4tFA)3q7DPnql{N-Hl&gb@ZkH$fhgXB}Vg+Sz;;+A}mX7>4rm}#MfB+^L-7VO$V zQ+h-s#$!HX)fUct8mP6=+!#c@^UIh|C5_y;{A7|ecQ1V&?M0Ssyi4qQD18(;8f`6w z4zTeN^uy<4;@)|Y+@y08sD5o3Cp~2?$-4K8`RX7N&B%Do75vyhogc9Dl-6DDislva zG$)l@n{Nyq`;Mb%<0bS(bdYFmMHvl((6-9Xv}bJ)Vm{2;EfO zF8zepo&?k4f6`0UEu8f=b+NmTHS>4nRc4{RH3qq33l;)URCemLccq<4G;_xJO5s_~`}uy`Z0d zS#Bj*?W;MR#!;MnpfNq6BaK;MvX~hV!&IEwM+=@r(mQHuaP7`WT%nN5Np8Hw$c{88 z>1!X->NlU5G4YG2!1j}UHy+VP(K_^Ot_Nw}vPqk94Htfy}U|L@X)#mt>Q>u76x z5^b|m$5+bg82-bFUUgEUh6_#T&{5mD$UtxI#HGXJ!vjm^r+hhlQYBrDRPq@PU zPtlo%)%3MtT=Sr!l1fo34N{0|ujfz_Ng`xuFed3AWfm#XoJ?t;luD8YQ`Fhdb3&$+ z5Fr#rrld%jGrar#dOn+CjzuAsbyQ+S*-bPNda#b>IJ1LV>ahI77}z0a3>MRJn6aEZ?3GV|mXtwg zwtp;p_MbA&mUCdvCGp$^k5gPqVj6F9?gj6r|DnQ7)rnaNS-n@S>zHQLU@p`AtT?K$ zj=OjD4_;bjiJxX1UzoG1g)#vggTZ}RAO=L{EAwZm?m zABVX+&+*m5e8p;%5>_~+Gt1Z6*tlMnj;mkb2FN9{1rO`+)~ym8zqUWEHF?W?XM8~0 z`s3txW)FA2F~eT-g{I`ry6;Rbu$2~nTT4rlf3e1gVs_K7oNcL0hR5e(*_Mkvm^da4 zPiAk$=0AUM*FS&B_ZSDsxpnpUbo@-HEm{jY!Lvaz{T!3_u%&+=YUxUdxuk#6UE+Vd zp_P*NT-CfG@S(LoC?^jELnm@!v>Mzh@1AaZ}`%X+gWh&Xa%3=e~=xpE+yY= zCt7A02nj2+l|9W$_*D;ur(yL>eCx_EbpYPm*345T#D45PeI;A$x zrl)%S`K-`6SlSQ;M^tr0PXZXo1#W^rH@CvYRHQ(`-+sin0F7pighz&v6csNcY5eaM z-ME#DbtSbd@b)+inqP&Tk7L!YFor? z&)tNNwhqvdo{pZOKbYdb9KJJVCmqZx!)6sPN$K)E0{3DJwV!@M2QB*vccj0Nf3kw_ z5x5#>=RIO`>kpBR;eX&=TF;suo}sF$onRO~7A}d`k&e?%m@}yYf-g;in6)yj$TOR_ z>yE-JOU)%aT@obzW+9Ttdq*+*Z!Pp&Nmf#Kv%loX7!^t2h$*zFN|CI)i}0+9CpKME z$Gxu~qiN-3(pLUXp_9uc`87u+>|qkSEu1+P4qQo(k1mv)%9fQ>i`68b9Bz>!Es}&R zGM3!*u7a(X&VcETMRdCK0sOK{Wpk?fNM+DIEIRO;g%#|wZ&iOun+8pz%ze+qu3kox z4r?2U?^adGR;w8l)*MAI{A#Gg_#R;GRLS3cagw=z-tl@d)x7W9Fw}Co0B?UM3%cD- zNs+FC$SvKUa)wWq3^h;^^_p8UA45}t5%v;RrJ0L@Cmo|}>6Lt};YdpL_a=?!8)?wF z2eglqlO*2DhL!1^OzlWDe&iCk>*JjyS7xh8uIC-5HJ#&Vb%hsoPR+v(n{8C&{FD89 zKOgR|5yPa~duTiSAk!N;NA#)fEd==I!S_NA?gu~Saz|{EJX|zdGFQKnq)baezHlxm z=(|hu7af3!xiRdw+J2Ii`oxmBZ?N0653+={o1FWMvMa->cxWK|wW64%&>Q^jeS+a^ zSvtqP!ol};vuXL7RAv2^qVfxP%Q$s`cYK1R$FJn<6*4j7ods2FzCwG)2h)($XE-n7 zIolg~4Xl1&V~%yklIuf8N}Atgh%ay1hh0mb;UkB|)Y*Ore&yB+{lk;#sALB$XtEXs zYkp)$yKA|{Pj*1R_!DgS&m7Qjix%ac-z}0@cR|l$1heJ}Jbh;kzgfvlqO)$e#QK;Y z9QJ;I565}Z?Jz0P+mK#d8&FKs5>d$9q|yB^x%}&S4a`maltuhrft6cyKz4o#bBnvj z|Jy&80$vZ3lnPAa84Fw?^IEw$Uw#E$Fp=gR^_9`Szy&SGZX`MXUYtK^JjTu3NE(g@ z=zL)#4u%o9r+7MbopcmB5<)Q}z?mAVM$=`zFT6!?I9g2JKwZDuFevH>CG`Bm;N|7E zimQt;(fA1)ShAcPRx`5BX~NVIO{nJk3Y%Xnp;t$X_^@kA_^WI+1|8dpdqZB}C~F;> zXsd+lAMBuu5m&iDt^P1=e>L41wU-)A29wvSX{6pfmj)HqOq044YS z)bN{Ql~Z`?e4I%IvCi1=`47H&n807Fjl-^z&nOPv&e;zw$9b8@(Ma8h`r3YQPakZ- z(~eKs<;`~7gU27q5lOBCp%px61%PoRmGniQulDOcKNEb-8PNK=jumBg<7 z%g<90Jdr^POk?d+JQ7|)rE`R}Z6Ar#L|aJB?G#;HRmOBxzO$ppFXHDZrtoLMHG8%; zjv5}k;8%SnR#ocZV=XDR)84yimG3bnf9==5CS9DY@c_Bh-pUiLU$A6r3zeU)rkYZ%Qgv6ci| zKclAS)A5O#DkaYUOpEsqB>DB$w4rD*<@rzGR85U&18;@PZ!M!aIq6vbR09rcq*OhXD}JlZtU5Xfm-*6QAp_^$-BVC zqX(#tx%*-$#-9?Kvd5AoP7$sPH%UyyX(d zm-Dq3=J9sw{wQubz}YJo(7&@sv2w~rylZhEb1XO0@YSJc_~#)OO*=>hdHr#{Rv(|e z<{aNR{5|*4*vwv~$W9$cH2?P-daL^_3o0DJ#!%P zZ7MjvQH7wXI$*N*5^SE+4MxVv9DhLQEixYfYRyq#5bOujW)6ndUMFw~837SaXGFvtf>kPn=o*@(O_?D{GE-g>|wGoj&S`;6q9pZ z%{&@8m^(7T{>G2%6&L1ggGa;E=tPDECwqK2_&x}L&6|})#U(-D7x0GN*|!WfpQzy1 zNDIuuO~Y94#N)8nI+wouoP=6aPeR#acdT9m@ZZRL@IZPE%jum0CB9XB`l3smhRs6U zP}>@GE-f&Y~FY1m^2&? zK8=BC!O@`EZBBU=0i-TxOGj6xLNE;lQc?xy2WjYYNrAJx^oe=64}mU!p--=+6=Y3G60V4d*_eMB@TGbACQF&52+}e(G#o zraPOxyn=Nb^%IqD*aL=|=b*v8g}ENcwb#D7foU(Qg#FK6aoQ@z`049$7WP+GBx|iC za?gJRQD^Gmz9PV)(A}V7z@U7%w5T*ph4q>4V^wDI?5go!cJiAylvkYvIJ=BZ-XU~D ztQia`r9XI$G{KJ)a{x^|r-Sn~AzQ95bZjsC!D7wk!0U2FJhg7LXwGs=kwc*sEEF|^ z&zC>2bd1m+Tl1Rb3VmNq7qoDh#}82YvRQ;-fucjBU17DHgvoO;@VfCFKc`ejv@U(F zsNW?^NNUoDxy+Y&-#2Hy4TrE;V=P-H{TjC^TERivS~e)n4t{!{6EFO|6>rTQK-mgc z;hDp3NDBWCHeJhOUsq1S*PJgHEf0r(H-)~Sx23E~Bn{WCOqjLaIrzNP6Eu$5z|>Wt z+@}0r0_%PnXe^T_=^3)Lb{Wr4**XaAHyR7wly&ynuaeoAlYt=G{*+nVzREp19R{Nc zBbZEOHmn)c$odtPz$a^O(i?G^8~8*U9+X~UTdiW5a7$(p*Uo{E-h=cL{Y5VaZly=R zJ~OuihWr&r1Ce2#k!YcnoM=I;wMg-kkw|ydA(%exAe7IDVe3A8WSi=2gVQS8)WfXPB%irU8{`%eRe_vn{6E~C{4UEbqi&oQZfdkofFpB)H_ELyg_@1R7d$AahKAf?DJ zz^wiw&}Pd!K4`;F9Ip^b=d{H*=1nXK;dXquXbYX)U(WBIvVaXWTa}jfwJx;;hu&)Yl_N7PV<4Z#$3H zyr{wHQW5m+w=F&Cd5tML`ZQB9o$lqE(%IR&X|3%V{A;(LTN1UCRK|ypNnbB+9V>&b zzg#Fl@Ke{U`Hl|;reNHacWAZ8MVOWErtFL1^gXOum`z=x?9Ykfq`*en*|?J)FIYly zW$|P-e=)V45YdXKz7%#|kun2riXY7%&s)E~M9pE!H0ebl4eY&!!(9HLmgFJER2E8} zEiIHB)4C;@KP`_R_og4sE>xmE%M?x~$By}Se&YJ#9O(HCW&H2E375P=mu?tJ>~-45 zQo)fAxLy=OpZq)V$IMT-R`Uo~nrKN^b9BhI{W4#fFGD9ylPT!Zeag!&r1|})()L}& z)LIuxy%z-U+Q%$XG}}zGuKJSR`?`vD1yyogvI-l+9Wka(lN^7mNs1$;O9ETwN*)hR z!ksa1F#6+G^wJbObL=W@A6`Uy=~L)rO*hF05zQ5Ma}QMtxJ%XFu&ck&Nd^=87Fjiwn#1dW@opuCAEC|CL}oi0nFhBQ@NI_DIdKf;@i&CjKt(|>94 zplp8XU3bG~K>cOipH;skIxdjNg)I6yJq$5N?U5k46H7qf*KYU5{b zTG!o-*M=Kn+NkT8mY6{%Q~YVjuC)|>R)vBlb>Vc4+347F9MczUq(=2f2YNbNp+x zmmI@$p24J@_J$^oOvlX!m!gBo9kh>{MQ%1aB=_zy84c5-4|PiX%AOKZ^H_qi5_#6z z>yQ5}xrY}&I8ewsZ#v^1EB-!Ffof-7#lwlIDE}jZWCV8b_Wc&T$KNTmdN|_W2TF7& z>LmtOOvWSYOT@-v!3!<0t~0(1{+|b@=*7JyQanbvL*&tI|d$6hZ%3w79u#(Fk{%!1^Ql$v{L*fA%w1YE;;%0_|MZ3_9)m=(IT|9h{Tb}9z)W*!?&fDcAI>=2>tJrw z$I9aK*oy#FQEj?`sAaPT+bPso&fzfgjbo?JIh&xxqFh zMnOtQH9NyvVX3AuJat`-@pGKP=SLS#P3%N>gZ<+1ch|6#NjCH*dpk8c58+zW5-IcP zIZ#YEPbnpf$@8k9-&9Ox8Z+wA;vVDNhx7CysDpWJvZh&B2bMiC5b)3v4rH9ZB%2qvy*v);``+G*##>X2%W5J6-aH;a=Hb(@K8u0j&m@< z!8c_nR=$%h=(~fw$A75hkdN1w8`9lAe+qP$@O#(;Ru~a4_=R`Fzz_Q|&sY{djg7={ zuZo$JXg+=YvxS4fXQ6iMcD%4_3YoNpVW-YS46zjU=f>(ip6mAF?3Do zxcJ?NNUGPhqtQ#6*^!w-&S*uVxcHTGj`#e=6XgRL{kV(2jTcV-T0yOi-I|31O? zq=BgXypPj;mCZbB{&K;K^6C9qDN*jc-?;eOA$CT)KifZ9S@b^nD|b*W2lm;VWcuDW zG5kWNJ@>4fUvloIeaUP^ko{Fk0bi;}`^HInl|6)A^N+lKPd398z_6m z6C7lh9KNyjyC_4DB6zlB2FXDtYPl@48o zAE4B<8g7c8v2FL_Sjy{Txbffsy4R@7CC?s$?lT9|pf*FUEnEvt?Y`Rk7;&_|UK*aQ zFrb~AWa!w^%V1C+5B2%I?2y?9{_W(c{Bkoz2*^Nw;Y2B_SMn3TP&v!5Z{YCtnz2~k z-;S!gedtKPcoNzA(Z0?B@O#cu*!k%ZFXy#|<-FbxPM81MZ*@9|{Kyu&X@=gU_H#Oy zFP4J=!ues&6cG-{G@-(qciFRRBS3blAsATY^M|`4(ONo!>l?BMZvCm}Ml0JgyY7!* zuqX|pt4l$2vWKa)U&N!Ua>#Xr0&|mpEMD|y1WP-sjUk)m$z0chhV*6rU$q|&ECt<^|Zx!~(_Ca&3g6NBZ!0eGSh2p#hoFZu9 zPT_fUN9ihAc}zj=(IL zG?IUE-M5U{pvNO%-F!hoGkeHKYYc+vq4$_#R3_Wudrf@3b`GDkMiX*Z4Z@)86W9ZP z`9Uu(!F!V_yp#MD-da{Otx|9e!XoPn(KCm)1B; zWvXwISod5LXjFVw5m=_r;@)n8m)1u?y=f?WtlGi0jLu_nbGq5h!?*08W?Qi>&9_Nm z;Uio&O&PmGHDTX!oYen76`~B^P;+zH1R*zQTt;cU6^s7;VJM&*GT-xM-;N za^oX{roc00Me6ZOs2G099S?jLH+_;5vkV@^KRLQe8(E zUt0>fwY!)anoE}w(@^(<(5dU4hnMSOS)^k>Q1INs`|NUN%{#&gxl9V~UdiQ`{^pDh zD8Tcj#dt5H8P_lJ#;C_Bu=G|XQ$ChKby=se;Itpi*t`ZluMjwwFHgeMLL+eTv_e^H zSLls#!tVBV)-`CJee`5c&i(fjPC7b-#&3Q@hpT2Io*Kq-gGaDI5&Bf{eW=J_Z3Tvy z-^U~UjM(~m8!Q|kBPsm*p6<@g0MmC;lJ%1e@WzWlXj&D?S9IPYR#!vN7A%a>-_HG`jX2jeMdDN7`fno~(jd?;Ox}cO_2o zspb5O3@FBVEuHwWgu0g}({AC+_xs^THfHKsaZukuD7v1*YNtl(RE}lT$t1jTh5)}4E%1w;qYjf>NJxK z2bn_QSLL}y?W**T0boT->HUH>o}{{mAebcFoj8zJ^eIeY4?OK!_w(4eM4@GZ)M z{haQ|RU2inKdpoLm!S-*9s1G7bO#z^aXi$)d~t&3M4b#r6xS#iDw!|h)ReSxdqiYZxDQDR|QgGLqYOe;MIy!y3Z zWc)w0h|-ljF?q@R73e~NeFSXV*=qM&DFaI)oniI71rwzJ;xOBeEOy^AldIRivj|gzh!LM|Lekp6fm&;f6fF z#y0_cO&AK=?qWlt-WJgZ5lp7D81<0g)0e}g{Tw|rZMYjZE7Y>fvUGrde2ea4_2kj4DX3h?K# zH0GIhl6Bmj$DA`lXmh=g+jRGUapPXIF;9#6wuw^Q(5y3TU5f(7Oc+5WMf2>}3~fcz zqIT-;E8^5+e7UEaHj!N7M4)Aj%#!;eeh_{M>_ws#$qtqF(R3;JA|&mQQs zzlq}(9HrathcQmZnQr`EN8Ct%c(%cubIVo6^QD4rQr5=(G>&Hw+Q%szIBn-!7mNOj zRjGLK7tYkzfui4d!F_>KU*`ClE5AHLGEjS$Ef0S@6ft%y+Uk`~L1A+ZeHo zwZ4;v#tV$zTN5zc@-o)_7fnMsCQ-1;0IlaPtZZKnlm4)Z>$|(D{Oi(2 zena^H*e@qW$>>7dD*8?qZd+4alxYV#+fg zh}-S8xgK&x(;X9Vg;o~jZC8^7k628P_9a5cNflB4c?({$S#}3B@DFLu;+DouV zTLk}Su9wFZLU;H%oRW~tMYk(+_5Sy8gVG(&d)OHqB)I$hE(sYENd|_r4d5c@G{MEQ zdqLIeiokwrfyR>t@G>n5R{s>d=`AL(e)}-dx2e;iz(R{+*N%m$sX6evEFNSAE#fcz zy@

T%+Iv3t)6VHT$pP7!V8B3as=Y0>{7&tUc;jXOt378Fi6*-rvF1GlhH8Qz_8AA`8X~qe>lEXND!gNpeQspD7@N8r0T5;T=Q-xT4Km!l1+zyrD zmG-p`vh*(13Qj3EQN&swiLJ1gpE++zTk4nMEhBqSJ~0UNI;Fj>h9>GL!#%!3_& zvxS9w?Zk~r4J`iDF1oK70c{i7A#F`Mw;=Qc|8;GFSh7ZrTr|5#Uht*|eO!S%Yn4SK z>!#6;vbAKs;TTmrjD=|vzoJt>E59eXjqUwsA@UlbC^~`T;KlM3_PzgE@{E6m^6npy zj~8Z#X7ZqTUsFbE2!7}6IrXsF_RoO7@@Y0k2t6a z-vhhx_1uRP7&wbpUvdtPzqm&G&YUIbLoPV~btA>@Vxl8Ertzz%2V?5obSfIsOAnnC zB*ms%*`wyS_A?_#h^ogf5XH?I3)U&+WVRraDjd#o8>*#n?~N;DH>!vRF1v~CZhNWu zg@hf8W32ARM6yfspw5F|(d>%{i#R$+6!g)H%320V9JdaTOu8{zw8UYwDAh&`(oOp% z_omI3-0fJ-NfyMCkq6^W#%nQ8??3GCuwB#elTZpo1@p$FSM$nE$lr*QKc4-3Y z9vmo2zq=8fjubP6k~Atzbs|%AV^1P_FkD_qw75f+SJ^gG^i^jM3p(AvW!xA3kNwSs zK2aT3a_A1W`tHW36$*4=p-5!*M^7}Q<|=rVy@cF<8lo$E`%!$knWQ!@mqJI%*xLk) z_!x6lvO2FouWo6e%chx7J**Q>>l;ZnhLln2tPHXp-HE@NPN0rQC9_K#NK1qa@t84l zsZPxhr#>%WK7XFz_3C+|t}D6rE96%~mHlwhybqNqU7-zO4cfHz-6_6z>0fqeuyBUc z9V1fy7YAGN<*4hWDgSePJC{+eBN-le1kK~yU|yxrv1npW^6R8&nxj3fOg@8|hZwDR z{S;M)e4`70xTKc1$0yg^7`B9f?Ws5w;`JA%5*Cr+D&>Tn0&GayA zj}M8%E~8((JKYVG!WR46Y^1D?z+zjBwXvE8SYMTAIqOklIqNxFSdsfr2N4rGVM=gQ-)1x8fi@fN?j`g&s z>M67?NhY}g`zXM&7TG`<3cQ_2?BQ@$bZReU>hGWfKZKt9xBKwV@KU-LRE5zyFN(iP z+EFdW7p~82<7G$vWUi(IDEfjIP1q;s%3DGx-&q$=SQJp0i#$a)4CKFFzl`D1MRew- zErn;x5u>>$D3h8mV(+zr=V%+V+c&y=0C zTY(K2wyX)41z%u0b+d&Ygeo{%`ycphQlW1nlPNrIANk0wrpE;dSk>bVHld@qapphp zXL=V7+`m!C!>ULIPTxe$CS`Enaws^*H?Sc|q10)&lyyCM$k1yx6^F@4u7y0HTODWc z?SZen!cv}1TD%fI7|jI3z1Gy8x{UWZ6-yhUR7mzfJ7?cGg2mdn(!l64_FxCY;|oKm z%u!ZS8s#e4@B4x*W573kcyh{C2w3keXs!L2uxnt* zs0?hs6idfG>2Pl<50Z8A2C@|Vnm+B1gwElSlxQ=7UAYhg_xAeJf$sw)weo)C+Agpi zHf}_}U{ki>O9{@1TSz~gn@~Znf!)~qnZKB7%}lqgibWF@{zB_Zdju}7Vz z>_5Wmw<*B-^$~pMGdp&w#7Q!Both+NO+U%1$Z#Q()sJfZ9+J=dFVuA8GR4%}N0(ekFOX@zXyvvq8H zR3(e)UcjcV)C9>}YxuFWiOCLm!lsy9VvEh4nb}4Q9CtvooK$%HF!T^_+&_ye94t#6 z?_N@Eu@`0fUBVfI_rd77UJy|<6S~xjvF3{cX)WKz+BGjx;*p;uG1H+xp?2iq_)6e- z+Cx;Sk>qgyJ2ZD+HYy5vt9s*eB(Oi&eal_6e9cuHtJ=ffpZ!SEHP!SVDX|T*nJld3 zH2Wwt(7jRd10O-#nZEBJ8$BSF$p&gMm#-b-s5V2E_w^aZstJAJfeB!BG?Cpov5KiZ zddPN4-NJR+9pd-{omdi~!fdjV;iq9a-)m+@-Vf$bcIHx4D^_D`eDc{$7ggS{(-ze~ zNI~|wYj}Rk1WKQ~o|Rp-A?*jN}S;-EuD*^7>@@0o!kkIPZ{F2EfT<*Y@| z3N5sfP}Rnm9=@5uqREHt4z@i&H^%h+fZsLTX0c0Kf5?d1P@ekNBn7Da6 zY@ogfPXssFNEKT@tBZQyjS<=0ZsZ^AE`*ZJ zicq}#JWj~eAn&cW#21rPsjoDh6hFM>d$Tkc3uuG0J1n43+0rh?aF6%Yqs$&+xJ-_~Rx+mTFpJO5Z{|E7d9l>S4{&Yi z1z5f04|B52!F85?xV?BY4efsiMPtu_`@KKRzwkTEvyQh5tVp2v9aCvlWeK@uAEna^ zTuCXT7?=H5gME7w;MXlZQNQGkP!KIIk}v2lS}Z#cs?s}Xkmqp8rn*ybPNB zx;2Fj-$NRgRB8FAWo%)lJAGQ6LuYNmY1oD~Zl&K?yyF#57jDM0@-zc<6py15yLDNs z>^6Ra+Xy5XSyAz}7c8o56BTcpEO0>lSj#aUr@P<6-L5IDI=q9kQx1jeaU0;G_a6ER zsi@$6lOL(^h7XIZ#yL+{LQF!t_}Yg}6{m#z&;;lHq|kE`4_W=e8It}quHPi)u5b&L zzC=>pkpb*uSTWu>W{Wz4UwFHqhfjKTn6z4V@rP1Z;=zCq*qwiZjAot{dh-_Jy=6)E zKW;I8jD|Po-k65=oxf4-!5uryYs8*i!}0E942(_0@qKM01_@lCx+|^a@ zaG1dKanp38&xN|AH03&P8&`t;7OtiT@m6FK_z?f`A=DZa!+x$Q=k&gwXLpA9QUBlV zSfam|PvuYG!8b$r7m8}oE^o(LT~Dz;HgE7~tdYoZz9w`>j-j?+!$kwmr?K;&Vo7bj zfZXp*;8niY;Zh{r@JPrJc+6n+3Z49vF(LGB{%zdw#f=KK+R*qr*}UlPMn1j=d6gg! zkp7U1zg6wY?1A7H{4|CiTRj{v>pn$ezKMP9cwx7(IG6QJiDU_l8ZaVEii+0`z+I|; z&_q;=w{N`TZ%!D3b)^Cy%4-3t@7u?IJbB56Ps$VCpLej@<2PQ^+<^b>D#SBGw7EIb z{jr~*$7EjH$q%xb3)-v7Am!2?Xm4r|r_KU$)9*&xKfPelDJu$JbQvzn?%_EDU3eAw z7-~faz6$K;@}jY%#|BbUW_!td<`E%hVtvnxp?7o-mBo_uC5H-yh#-NvSXmh-N@k7=U7 zQv4gWn16Edjosf%OS!ThL%7_(-LB%7D#*?x~fu=`djW&*4uqHgGN|kAL8{$7 zme=kymMFHKr71zT@kY6n#9`uM%-f)YF)sg~Ic325Ybu~T_b-55I1KfeMW?q&(VZoA z+>;Zfv~ArreDk1&^P2pbTeGH|{n7u-=5C7==SJUUC%0Pi2_GG)>W~)=kJqNNFFok! z<_WAhakuz$&l#ptF$zvbY@_c}WT<`jEc}wM44a;7Laf4ERuTH@d|Ub`THW{>m)SO< zr?B7YRhP2`GGTB$d>`}atU!KD02ESeA;rQLVaBbKwlo}dM!Rc`{Xy`anvl&6| zQ|@7&i!rS`{f^!_KBZwp0{C6ef3xIDZFuQ5749rHfoC@aK0$8^tq}C|%tgkcoO|Qg z>lP8H@7I7!J{kCjH^tx#T^Mv%&u(Iv7Fv~6@>KN0RDBTTK8~hlw>va{ z>Sx~W>q%JhX%2*T?t%;19ehyi0fc+=$o4EUQ#cHbXU=O8(Fcj+x&_l|+k zqQmUo{DYvJmH-$2_0pW$wbYy!2PUm6=w*hqWSn)K{Rdw;IB)%rSF)7gny2@0>)m-c zGk6&EhVk&pruqM!m~eKZDS9i_+c%j_1g|Id*d5t`f0sUF?+tfCm1{RQRLDNRl$*vb zybh)o!kryKP68#k*)g|az;e{f(H>*VbmXE;?Pu^ly zTL_zUyO?_uG)EM=;SYFOhEw!y8C1J_3C=!^g6NhY=vS^NS}#=t_oFMIKA{eVTgpP~ zdS%edk!NGCMhdURi?C3555%065zRW$2%FT)ASQY$K9`>fx0b(TxBlIOWwmdhsMt=V zqoW0D(_En6k(;o|`CWPcjX_{z5+b@cI~rcAyTi9_ofSp(uK2UwRLC0Ffm5VET-KHc z7xyG^740EoA!BIrCks{o+X%|(nxd$l-%zEbLyy8jaB0tEh+p=QEfRjm?xUi}S9>H>Wa4V2y1z+y(qfn|mU8lQ-zX~N!L8zKvL&i(~u zU1gD7ehb#b$U^PdG~94Mki~VWp!7^(#zK$z#I1WE{C6Cirg9CORyRY_bbEL(Jq{jS z|6B3l&|X37>tP-LY?;CB5Hjkffx_MJuQoSvV=}zIdK`TW zcaY!EUN%%TgOr}!p`N?BWL4VDx7NPFh~pzM;8ZccV|yz`$?T)Z{279u$c7Y7|Hp?u z{lO^-e3}KR=`i^7Ow7 zes(4n7uq>eNzG2~j1|z@rKK$Jmp2#csVqslA}zVQISD%wEa*&y0oc^kLb6*r#U);+ zabtA_W_J%2XKlncq2;vby}nOTzKU@7XjZBABGCZlwE zEh=1}2rqsNK+(NtSp4@gTVcJ6`T8fK{JcCeD}F(({3M)jSI>?8x`&qBQ6sg+%itP6 z9c@-SiS3P?*}9}a$@cM&Xe`w7vEI5==I=}~7OiNVVfdC%?GdD2nj#kL2a)U{1&Yz&qE8r zVbVb`*p!J&3dhm&IbIm$7DmmlX3>}T$|9c-XZG93kVy$y)Q{UQQ_<~eYOB_lv}c{i z9}9HZ(>^Mo1DipQQ}p-oEBbb-Cy3W7NX6qSo(px;%6Po;VmKFHR;wZhX&ZWr@oXB(PH1>YhQSQ$767;IsO&L>0 z(Ndky)LLSR-!-4%rCnq3^`lrCcKIl8kT8m##>eA`+ckK&b_80qS#tYMWw2*f$7%4j zR2s?;q$7U@^X8LJfK1?48mLhYX54wmJtKH8vR0Dex*$j$UQDY-MbhYnNo;CWDBqlv zLee98>4d;4vF^8!q_SN}#CxJ+(NPTA_Y4m|PsD#NhXoF>BGIm=IOlC6g6l@KJCw&A zYrf0$ergLm7dv_vAJ2t0Y@w)=H~F^);dto$1X`_phZ|VqMeCegX@}i3^tqKqZ612) zaU+(j`(!1p);F1tK*s(NBulE>3vfz{CLRiqqJ+JgY~{8n=AJg2GLL<-Pg;;qD)0t9 z2Rvc+)mO>-R|ox2k&y(HnnIG!cyQX6%^x23l+^aWqtE9i+xHv)lPZ75lV#Xu{B_Gv z^dr4Zyu4kV0?zHmeA^tj>5>V)>pb!A#c5)f{xWF)=mA+pcM9yUiB#B_$#ms4!Gj%O z)NcyTef|d5{5-=l3*_jo?;=Vp3`K+LNf@m#hPAzpg3qJI)4k!^bV#}Z4Q_vA9}ZbE z%f0>;CFgI^?6-oqde0-4{Zx@Q7v4t25(S)eT8d*Uk;cMH{iyoJ zCDfVS0~0Ewp>v|LXidXFq4j1QDOhDHtZ*z~)uI zXId$*_$v+xtW2w(4X9a8^M1bI%c>HnPv{wNJ2RFZw?@FQy#(Wny_oh#cgY`x^Ay|m zi!=^?VxvCVGKuiKd)98H%1eyuq<3S>xL4?%@5z=fxQ^FM7jOe63v7MUL9lT}1Alf# zI^_2%k>m9UQpwNYZEPQ4*VSyM$0!={d=mL) z&A?EvpJZL0LL7I1JvWz@Of()t|9N&{(t9<@tlUm|DVItIe;uIFJ6@Bi)MnUm^#Mg@ zKZlF$;dsI&3|gj6h4ZeXVY$#RtMslDVDU5DG`R`CKOIV0GY&9|C1XkNW)hfx84dQf z6*$RaH7gxw41OOCB_o3BXq!ho(;9h}R%wUPh*g82U&RVsn3T$tE02ngY)a$DwRX@y zp&!Ac>mud}&yu_K0i>Q`lHDOy_*G^YxBTiim?M22O688B`JMk*QeGjP-Y;ZTPnuGS ze;(&>3P?>~hD%;{o1J~UgdJLy16mK`@w8etpZVWzO5SQo3%jyOLAQma-IszhombfW z$>;2@&XJ+{*N)@*(n>xdzMiTIGs$at8yURKqYRnrxXsXs;uf0lal&lA|7d6M|6au0 zb5xj0=4RG9WHtMo?9MVXlxe|uTbgiU9}Ql$l`EWR&JwGP;QbDJT3%QL>4PRSWB+2Z z>M4TsH^O!z~GYkyBgy{L~t5&aFfYZ;YcwRSI}AWjb7ccAK(Yu_!S- zOqemOhu@J$aS!*Bk(JWNW0_48#{Ei0IMQtfmT4+sYGwel_m_yorrh9aDgyA@w7bInv56ad zzt47!%w3YG+D>Foe4?fW+L*pVa3Ni&rrN&y@!)DFR1XxgFDs7JmX3%Tr{{2sDvTlE#Bi~2%_FS!S0-}P4;p227~XYGApy4Lu!emsU*?8E<79iY3+^0`yf zwIJYl0X?;*n0834$N6SAur=c-c%AYj?Gq!g$jKi&LsIde(llI_(@d?df1n1BEl?&r z505R#pl8b$Q?uRsiS+_clA2VH5rVU1mYXIXzGHxY;wu=hA;B=k`WQ{x+QA*ku)w=Y z_ekvVLb^IJ3l{pkAqq@3+)=(&PlTyYLN<9|h7AsJe@!nL>d+q(ZOGEVd=eM;l{S2x#N4`ai{5iUXg$?ORqGVNw>q5( zxOtD;d)uCOTbqR=6*TgEQ%jXnj(yKMXhg-(l>w6oyP!CI`5q^kmgvMyXT- z&Tl$^qZXbelMQ@vb(0#jUp`8kzMO*76SC+`$)D)&89~arbFd{`;GeYn(-gBxdg!tT zy+w!dCf#zwh)nf6LLO~iM8#+R z67P#aRGbx!l2zsOL1#FcmUW=k)(&#v$|qX$CYXDxt%Xh<63ksmB{rj350CZ7au0N^ z#0%Mv^xggqxL3HhyuLk-q}N_y%!gFL;>+Wy-ZDoVYJU+kCzmmIr$^E&GSMXNngm`* z^`|v^u0n&=P0<-kS8`#TEOrOi!&}b@)P-y!BP5k@HgUz}=^>)d=43o?RmnJWLq(=z zx6qAdr^s+7fwTLzn=Z&VXWZY0aK6fx7rYRSHmFM&t+6l(X+!Ty7)AXgDX=Zl__Xp@&zqP`du z8f74&ekcT9DPZ*b9Y6%{(CuR`M0F`(j=v%M<4F|kOutAR*9O9Wy(u)YbS(UrZ$Ko% z6xi1H5>(Dt$Su}CCX)*W(Zi|)M|{;~LvH^BiUrq>KI|~=sS|ND_t-C&@9cr^IOPX zKM=>--PL33ywAYYpm27_jC_cjjL^4xE4VChWw-e$QJH07=mys?Tt}C!YsqC*=7zE3 z+Lic#C-eAp^%7WEeT{uK_?vC9{03SQwPe>m1Hl__2#ME&xl)%&ur^1RKQJ~E#Ili) zWoO6UnmHbZ1mEHYO=jS^%9kizz~Gk;(J-^k9XvFy6a9Vm5Fup+MJ7$yGp?2R-W<*Y zrvWc3PXmSveI?09LQ}wvFq_`M%sxr>tah>Bv*yr7&x7~2)*vGm7;@&V=I}Pph%JbW zVdL||*k@WRh>O&6VEUTr*s&_`N>z>;ZOp`hvTB-^wue0*=Ed$)H(*^-O32Tk@lKqqUnrLyt?~D_PFB%dc15QPLEju5;-dDtdX8Lb^1*jP(6=I?>P)#)&D|O zzAsp~Thd=#5htHnM8+>Kg{mKGV1n!|BBLe8jwusn>bx4;=;+KETDY?%naS|UIFS7! z{QXxCZiGjy?ndNlmOja_?moi~G_*TJS^I z_vx~kev>z6Phb>&onYQdD8sZEJv7#`V6*g- zsg7<6yh{_D?m^!m%9|(R`FSMXtC9)(dyx30NYjq%NXmswd&ZqLs2Ckm+xuJuO0N-RW~_zv~!gTNE;?N-M=Dd=+?Q2~}RREuABKR#S`qcCyB)68{`&rv?fFKc;0n zvoT;Jy)bEz{!DM;n(wWoO%H--)V`@SR$(}qf7O+&KW&8;LyvPy3;bxhLpL)*r5w}S zQ*ri)5#rxtWFa6$6KsD2rVM?;gl-k)2>yjA>vV=2j33R9)sH3De>0kP;aAV;U} zenq?E?ZrKRRmp?K^`b4qbjgcI6S4e`XzDE|!=IdIgzUTfBs*J|>9KlCTYndb90mW2 z?<^JE5Ub3$sy5;DJv!vJ_Es7v`-SfIRKzWPJ!H|C2h{PYC;FAh;DHdqlQof{N176t zmw~Ey>}Vs2-K0b0q)(!LogAKVnMt*6FA?XW~mFbJ58c4FI{MqofOTwlZhkmoA3@gllhw=r-;hb1N7uSY1;DA66_vEb5!CY zjehBCXgy>iQF<|nez_G%4peTebve0{iuE+;OoyM$%ttxQrz^IMb>&VvrS>^Z z-YI0*FB|g%V|DmtL)M=EaaxUJK4ST5g&O3WY{aJF{@n6)Iz)a!34K^TmUm2_%Ex&w zAj#E6P@UF9_~ixkO6LNsX<3LRo=51U-G!t&{|)!<`fO^@)X3E}wvg-cE8wehFOz?> zh?|;nwf5r457l>G>w@Wq1ElNN5ju`3#TlI=aH(GvbN-O9ud2IFqqs;ansxx|=clTKaQoc1`{kXSALuLgilziJkdN5^Whxyj>Hh?WS=y#}-78a{^o1=h!W(IL@(-Zrdt@wi_F1b(9s2|M`!U%=Dyv zU!=$k;}7I^^*pl9_9pkkeglcfb_Od+HDYhHANzvr>4QFRPBDKdmpsy(jEuf-d;aqT;7FvVsbOR`=28GStHmx#vje^j6X8?dDMM${S*1+OglYf!xbTVOFWF60#nLxo5A2qVwxSM)uis?t^SH z9cT-o15-=HZW(IqPp1}`^3DTxUVKcoLfr}E%d#3`Gdduwx{Lf;)D8Iy1s- z65iC%;;wg!@Wj*`wO&(PL2{HfNE9Z~&L<;i{gn=Sxc3abyd#wO%$H+po9@F{@5Rt_ zG@NcWcV#v_a-#-w&x%*P$-?O;rH1c7GOaN$X=UGr|l0X z10~;?tlkhfxMl-hY!GtPG38|CsSAXu3#K8%c0dY$0NhVx!Rg?Cv^w<|shr~te@^Tr z8pCIh;!A~0bL|-7bBlp($_rq0<{A>>EYHj|6r2f@vN5du8`&(s2ijkR!)F^KxLtDw zBEO}=)6wVP=3-^|`Ti)qsaQvH4@vPxAJ5~`Ru+373K@yk^>n1bxRy24ro}!cn71T{ zJf5FO`~mzO`_+oBJT{&!uS*im|59O zcO6QFK!X%cA0=>sz;6z&O|n_brZSz29ATk#12W6P^;wcYO>$*$Zax<*M2N)P(O+7+_} z&hi^79hiU`x73&i?LFLt>qD@ei^r&?d0fDk0eaUmjv9G)aRr|Tndy0sv@mNO{MqFQ z9-`a~>x6o5U9h9?A)T)@6;FhIqeO8W333Z&m=Tj~lQj0yp3gaSZ=oxl*>ewe zJgR{D*Qdb1a3MJ|IgzZ|t4H0{me3y(skE?j7&+)Z$XImQBRrYRZY}?h6FGg>mjvZDn1-> zp5tC_7W&=Nq zy%;Bk$>N#IC+Vm7ei~k>f%;Z^=~=tawbL_XSdYHvV7C1nmWqzyws)@}#P|v0d3YL= zdT|H+>EMi^jwE39agKr)m>3$k32m)VU=S zKPyOZCUM0y_JARp9M(nS<;F~CV+v`g8K4c4*J|t6t%dEfFPLhFpY-E`S>*ZSILyaa76;&W|yoB`Q2T>ROG&(|KcrS=&dl;baxgWoVgQwo>kHhsVpuYF=#u|(SoX5 zgS0mKWw>=s*yjsp$vZ)1ffkGai3$EI$GE zn&XLCr#hd9!ufSpG;3-82NKf5U|A6ZRz20wa3Fz3r^~V9Uns)H&ErArwg=v zJtqr;+L*L0y(BHskN(@ggdRF+0W#VjNS9d~`Q)EP{N6^xPXDX0sVfy@+nz9oR}6@w zb4xMc-ZPSsqYWv3#l&ARl)7$KgI7m;=&Exwg`I~wJ@*j6@tqM2Y)_!`@&vxzfz{{~ zz6vi@9;X^#4d^I;ODg4Y3a-T(v4u8?)uxn^?&MK$ zJ8~3#6;;bwr>nzm(>vgcyWvjKIhr=)4UAde0Oz2dUXF|;dtWll#A&i<|DzYSHjQOZ zw(q13N37ZHA}iL%O$6l^-;guG4#4a0;@->}2I2J`&@f^M%9Zp{yRdDvbxAGN!di0u zjgT=ooX6chT1sq%tZ|X+bna%6AA%a}G{tuW8jQRk{+`@Jw(Gh>uh(R5 zyho(yasORfS2}{Zdfi^EU2~s1diXLKJFcBGsmKDKpi$Mke+JX*v-&t4x0zVO6wweN z7ij0D4nGz?W2UchgjZp%oXv}7W`iF= zJBB&8@dJ0~(|+>aM2_ShUj_q0U)^<186W)GjeL<)gl4{#q$~}mJhvOSk4>e|m;WNN zH_GVPArp~j7<_hV06#`jDuc7GZcu43H7%3guG0>^O-#=c^RM>+-#8(MpD^Vo`e{gU(Uzf9 zu``k8PV4}mHNVL1z$YZXI}Wu%qG;sU6XJ+bdW?InI&ON3_*&5&8++&QqmE|rWf!+I z8E@QBp>8hP2My;^2JGo)lPCBDmH4}o4Sa3QBHlXnnaFp0JxO{+Xl0u@GEFi#e{v`H z`}IL$8F~kYEw17N`gdbb^j6-itDHBL&BEVpWq8B5f#kYKV_!xTElkLuK`mzZPSO#R zOmE`9;jS2y%VFczaFkRpu=)Ar3>{cG4Z~`d_X@!Hqtxci|G_c8umZT*2eq7~tP z(Azzq*NYC~uS`hd{?0bwWtXb+7v3j|yASE(YsXw#YFk9_4%6rB1KW9-y~_Cb;deSe zI+=>5R=~IfCye7m`EQR{VY8o#Qj*VT^O9IbD@&H#%BiIiZw}$?s12y{`!k+j<|Jf& z@~M8>MmpT905wMZM7OMMXm&_ZIHCLFo4r#p<(?HiSCvk@UpJ7|GM0Ea-w6ZN?$JAJ zHkitMqKm&5Q-%AQ*t2#cN-4zKx|k-B{)9gw{(B7x{63U^j=#@XZ;PNt#?~kwt%T~w zgSe45b}&@~OK47(1Kz$K0n}&+o>Nz$_tcI7Cf+1xI!=q@!UECkqb`mOJIFb;wKKZ;~`mUw}bkZKBZaYIEFDwY{MTVHf&uL97@cGC8iFbXi*OJ=8>$- zv2ZqXo(!&=@RnO;p^35;f5@>a8TPNcCY!d=gIs>U9_YGj@m zY*J4Xv({@Ewg;VL{~J%o=m{A{x`fHRK&*T-kko3d@8=J!TjjABK2Pg36q=) ziPJ*hRf{_R*Pky!thIZD$??JYB-JhkG-tf)jxZe+e@LCd2CU_u=_defEfB5!hxWz>y_}thL5Y zc3s^|_RqJEtm4n9q-l0MG`nV!;fbBFKqG+7RB&ZaOQ^uy4<>BNw`lfX*#MRFXR#qP zox3dLSSO~Hk{io)V0C#ZO_?eQC-~u@ydwwNCo93lfD7<2{WHu`kYr66A-kHrgK7WU zOFgu=!O)rKiPlv+wsz4Bc1Ln8{ob5KWqPtG*AY)&t=z`OoNZ$lrY@tW4GDQ4bXdHD14xSl=w6IiS@hJ|Zwm+m9?+r<&{!`I{{W6$v))L=& zrs3X@bY^UNvWULFPa6;11)Y?7n7U4xHLh)gp#~?h#pD3Jpu7pps;n`5_epr?bqs!$ z-@@Z3TJY?A5pC8LGY{vnU`-h~eRLJPeoz2Qgbc^2y${Hun>Ms}K@1(~-Ne{$8K4{g zCJB3l!@}NIn!Tmk3R)KOkhyXrc_Ipc6CZETw4dEzYWM?I*`_gu7h|FJ{Zh6m=>hz> z^&dD~{RcS$gY*9SRO)1Ch8jEbF|cL|IRDtruAO=rpS3PW?L5lG=)I;>Pu-?FV$+Ea z??%3V3qm!4byf8HKl(uKK0T&-9G>>?A}%tI!PDm;jFL|wn!nSaHg6u-WW6AVmX2bL z#+kEgq-K)nz0MHwco{jK@skV8>t|M&uVP%wCWAzA21wf9W_DS55bLE@(A617%3Izs zI&1`+CTeAUx?*b2rh3B2_=9A2&OGk>>~=``cO2wiooDXP9U#gwf|qv$%kDiw;a0jd zxpdtgUEkluY1{sTr@Ov#Nu?@;IJy#b6?-DuR0)pWDr|z-8K25(fS5lC6P|^VIX*_< z_$Z3h*FF~ejrv5=%GSf3G>QtPnym4sY?doIPOUAr;r7Z5TAl0#JJmg4KzBXa{;iev zriPKV2NGa!@)44y)JTknJcctDC(`WQ+3=v#8J-1);Yt1=o~!mjV{;oU^_s%g@5o^- zipC1L=}2 zsU0hRlEEI?n?aLhZ!&|G!f!>xZYWd8BTvsb;J&@uFn{hbSe)MkYi~@4bq_P>s`=Z= zvM)Ri&!f0#=uPtX)H29i{*Y=5yFm-#-r9ZJmQ{&OXCqT{VC2UM?9C@utXtMl+z%3D zsD(OAn<2#coaFg-uP`)!l4+wNF$eWGR?y*Fhtaqfo{R#w8B*>JWwWn^v9Ee|u}`P3 z2mP#j^kdC(syaD?v244+RL)XhOC;jhM(aHGM<-zuAJReE&lOeuKNGf3(rOUP@5 zD%fOHL>CNLv082?S9l5gHX50N`Oi{xm@%%$- zOy87=Y$dE?6V4gX@23uu(U*_V%T^}z?LtR%@=75CI}!zVyfj))K0!(rREQKd>NA@a z-KcJfHygJjm#x;&W5eYeA@1}!HZR4KZR@>C->OTI6JI^y-P?S6SYtVh2kO|$@+k5; zRtH`j^dO^_)L}%R1nbs(kPU7BDtc107h85IW7TMuvmND5w|!_PvA-=KR_hH}l)D7( z?YhKOF1$^i-9JX$Kdr)zCz^=b(Al8v7z6WPn~@jieF1+9-kL=sV5YQ#;hF!)p=-9x zvUW4D8+Mf@?rWwIk(4+F`HL59Q^cXyvbcL0!mNC<6kVURgDRiA#s%FUNBhf;QDUKm z@?lZ1Fue`@183r_FCpag_9u+&Ib$r))`!#Ua=>~^DfPLCcyC$_Js=|quN#{PIz?ey z+kDX4aDf~NNr&Xo!@(kR9UQrOlgvo~NbrcIk=h?1j?>Z1-*_PNiFl z%9=hBo=I0|_iIULu&80)SOR_=n@bbMF(lbn9V((zxaO!Fxb@MWn)}%@38w4tOzKn= zNyNgs-Wce7xRRS4w2pilt^(Tv>Y*WY8TtCT2L6>Q)6;(^!n#)zpm>)9mG#!4U*dF# z?3w~%?E91?D@ihqy)LBvn&7>*n*i~z0c%J7B8icXaBGo}mzyeyx87!vz&DkgLb)-y zx}t;>?mSO5cb_E}la7I!_7j>r={sHT5lF8e5_l!&rC6sI8(5dZT)M0}fs#4VG}UAs z9@wG6D-G!J#kV)&UoM<@P0lAeLJrS#G{)Ri4yB0kRY+-(gK2?6C zfD=#4(3Q_lkmIVheCgs0E~nEA+WK^-Ul4t!mq_j$ zn82?-7=yWum*`}nd$iM`2X93!MB-Y3ZWC?r8L1<0UDk3}ie}J-v0j+@YALP{6`@sN z95i*7~JjY~!1htr(szX)IYrgRvVt_;C-@aAirT zaE9?`j6Zo2JAvbEwqJ_Q|M!5N*w)Ut+#iQM5|(JYxs!;jXNr#Ppl=pw{4xhGHg6?fRfoZP$sM>bTLTPkoPqU@ z2H1B#jvQJxAMCAi$dj3W7CO0bkR@+(qbcvkB-Q~ zzTxAsa8ff2b&UobpGDkKb1kToi{PsseUBO~J4ynQ&%vsZEM_z?kWr$}{r)|S9I|r)!|6vsRy~u{ zc~o;JF9Nx} zn3%$^EmrX8(qz)9_K_jnMCMBHAh~5anc4|#rc5E9srzmWJRXQ5@mH!yCLciLyWf*} zi~fo}v#x13jVCd3F!HsZ**ep9oHk7>K zR0>`bpT1)-OMf3N&iPE^R()jVcpt!@@6TdW&PS5(^$~tb*TFcA4l?NQl#?tO$;~fY z%Y=Qm2CI%BNPP2(99ob=-({M?9;XlBHrj+y>C}Y9eTyN*coF&5Cb&n+55U^D&&5rp zUzo9XlgYoJ7E*bC4y0HrK&j3!+?L}ia6d(m_EM9+&@zQ}1!F*CRV1x_<24T;4b@uP16HK7)3VJQ%C@qaRO{RHti?10>f<@ovlV45uNbjE| zWRuegu~HXH6FVBYpz3mXEAHVWw<^$G_dk%CBTAWHGjw3ZmFGmP(M8t3ZHG_#qj9Hd zEe+cqM+#5N!k@<{>2sZBpz>rk*>Lp+F<+?(a!1z?pO7><{C*WQ%YKJR!$-iQ0DEZ6 z&7@L)PciWilECSu0n{vOq}}4h+zW+K^thoDkV~47V^T>@gbVrkd9LsP=Rn-M2*OM0 zz^(sQG1e<~lW{#xB;%zd{6S^P#};xXbu7f2<JtJf`x{BHkd0KFIt(^R@OiF+v4Mm$fthk;5~(ZJPv_@&VNn$UN1G>xJ{X z>Z!|2b+s%1V7V#pp#C47@!wp!T5~&b-k=Zv3H~ps>1WUe$1>hZf4L(cM#GAkCEy?9 z%D>Ih<83;2BU5~XeDNW4SZy72n zEUXK3rk#3Q7*%~o%(&jpeGqNrWITc)eBwBKE;u#+w93<|so%JQ#Y14Ly#!24DWZ)x ziaCS5@43No`(Q@qMA%jrMqD;+C10cb;bE&ioz{~D?SBt||68GF?ERKRbo`=INE!91 zK8^Fvgo=CI7opO071sH}AtGI7jW2JxW6wE7OxG}?FIQUvF@8uAo$wu#I^%Z_GjRs9XkuvnGl6Y(GI)sQn;^b{a#) zlCAirvYF$)9!9mn1g>ppG$D42$ccttxNoe^ZvEQMUGY0WU$PUp?pMCJ;a41YNhl*K zB;BK_(FRyPG8b|$GxTG+F|$V_lkTZLf`eQ8FxFgnFQ?rmHJ6Wx2VLFqk*y^@(78vS zsw^W0*UP!XQ?&8!`X8h}tq{J{8F3~*%;>8-FPO?Ts(8!0pE0{Cjb44*$(P&85V19$ zRHueun9LS*G3~>PR}bQ(F+AOOHi4e^SHsSoJn7Rfh8m|4T{)z`#e;cKi@C7 z!vYe-?^cJCyM;Am^U0wwCPtF!oL~!c_sPKBeRsH~hET}ePRKl~6qv7EMuw%j0d?7h z%$C*gZ1!fdUYk&BT0$3$4CbubPjH7P)d|`EPBM_%AwKd+8glMtlH|`Iu;mI!>mPM! z6<@{c-b(yoUmkJ!AU?cahmWL%OtQvSrser<@$0+{SZ#>CegaLZ>enPe8?J=&5+}$y{ljQm zo-8m*63EhmuUIIXfNF(C*dN`;Oke(lo;_y`+e6Re2)VBqa;6qjCQ(>A=^@A^OY>fv zX7cN_Lik5rpHb=E7#upInaMk6%ujcSql+eFP?e|@%9|LY?DZ5pzLSNLlP^ih?+{%0 zpBw*eH-PMR9ujrNfJdkV3=Ek9(~_>kXzeYa_e_d+9*_K=xwFCL?h#P=lmM;^W-z0a zGjUDHH!87Nn3D{QfexP8B zeTKf9dH9hocnd_gaDwO|tUR_GoHFa7++7i-rs}~crvrR^?s*!n7|ywGTmhk7!n~x% zh`#L{jxYPvNlwp3zPLD%&vwj*9VOi)udEdtrl^9I=~`UzS(y2aZNJR*7bAqby;eYeW#`^-VvpC_t78z z>-n);l6aes7O+#NpRCcy1FcoxNWRM`{NiWA&l#`JyKRf(t9{b&(UqHIL4PexyFR7n zuUi3_27H1`dX-?{nG9~hw{VCiGz8FKkdhqbJfDaid$A5f(0dB=-@#>)kyw>X5VxLhu#NbyIeU@poyvPiS@OoB)moNd^W5h>X#0a z=)V6zVs9}#m@k~UOm7oS7r~3Nc%isJx(0fh>>$xumb5ATz}P#*^f@ayV|GWuLA%!k zq~DSq|Av9;wivExRvJn^@rA^3tDy2l8~3l%gm;%Z&#RwY%**&l^KP@g@Gqy-@mV*k zI349|s;w|ctFCK-_Sa0Z@6BmUDBS^>LyF)UF&6e)2VvogELfE{gpHI+;3SM^QIAU& zbR^kR<(q94QlFl#r!SE-A+G2W6qOTLJv#L z9}S}MPdGBUZ4pd9bslcdcg2QZYIO0LGUoH72ROJn1HWks8Bu#(vQJA2ZKp*-$e#jO zc1~bzFY$oYvpyo9H-@%(AE)E8l}W;VT>vL{`n$J|KHIIuW+o57iL<+DxyMZ?ype!Q zhZ*oz86TnQpFA7s-c1r-w2;qPzsLyFMex2i4i+7iV!I2!k-z0E6Z1O&hTRFJ&C{ZV zb9o$D|Hy#sYE6Lqian%hQ4A(e$$`zg#!-2b04f{jLf*KI#UH6WRvX(g1>z(qO4wIAvq z%z(Fr=4jg)g{R%}P|I~O((hkM##~*T;W`?_yDh0+zCW&-mrAq8pCG^XRWou?jYQ&V zKeP3w1Efy3AlE0AkPRzU*~@Gz*jKoK-=8BSI?Wk>{MBMR7knna6L!<7M+O*AZ4-Wi zvJsx1COmf!>cA(B5V|Bf2Mktd@(RzQ_}&=}^hkG>ZDg_|zA)_uUG2+|P_qYbEc7P* zmfwh0e+G|{ic(B64e$H{~9f2fBVHV_M445z6LArhPA$wvZUQ(HiIcq$@J#Qtp zwyAL%6NaJrpV?5ma*)Zf-$Rq)9*|Dv1KAyKPe;7^P3>nmWBJD(?)kP(Qj$49eskYB z{RtkTCrm8PnK2dG*$QyEb(8e`Fd%07t+gBW$-#c#D4gqNi(iN=8jThF2Dcr#)-5WS za@rFdRj7?C51=7Lb{<4<@-f!XN@9&@sZ7W!OD}-riFV{&!D148cm#A! zEh0bXc@t}!DbVE;Lob!8(@9U4(0TvOAW;gDq|A6W+&$?VR zX73B^oh5I3sZR?>FZpCMJM`vp6@}%D3v;=)D9M-J?w>#&ya{3&zKiL?6LIu|unBiR z8cODl%%u4SiBvnkhRj@bnbT@r0nKw_NVMN9?m}QEH|oKCywJe#?}GF2%JZ!>dEys( zx@SN<*?ow>go?wt8S?NjXec=Pn~+r-0Hp@A$Yh;+WbMiq^yJ3fxc6Q**RR!xdp&Zf z<$@k@flDr3B_QIP4wmoITD(Fky#>}2+ysLK>80u z(5o6Gn-lMF&HKO>H6!rY%pSCezJSK~nbZr+^-aTaP$4Il8lO`V+^j<};aoUrE11Zs zHSI&wPQkGgqzanHR)WtX8Q`34!Q_Jp4E~yK`!w`1Em$5*Q>NHZk?>ioeu)Gh;hyuy z_$2xLUg$jtd7+#@1zh=U0!=-B706(LkU?>$eOVG@%FzcT^=u?)K5igJDbGo>x76X-MR3gRc7!Yy6x$0Th^#s5)u-f=m9VIL1Il%}>clvSefz0Nr|C0a64$SOZc zMoK6|sI(VNm57FwqC|bqxl1UcBuUChvMFR`gnGWu-_Kvq{d)E9ec#>Rbi;vF>uh|7O_r+RJI*0OhC7nyPv57H zPb#tWsv|hAXuiN8Imo7GP7(R7wGkO>8jH3G>%(m-vV?c&#UZuJ$dfZ;@e0S~cw|EZ zsj!P9_SHt1ybvkJskK?z=3Ske~wXX-S zdT397Dov*M2A-fq$DhbNa=0jhY7r!xBDezW$kmEhxOTq=t_ev+>m01fn?2tJPaG%e z>JVzzYzQd|Od>-pn#j={FJic41X0dOV(JN9#JRhN1RWb9YCACtrjD7-3LF!#)1y!L zhSUslQfoA;?8(HT=BG%W)(ZOiP8YdbT}fg_6Q(>Ae*NVU& zWG0fTYICybg9E9(^bpOIosUs5BJeXG2P|E)?g zOU0i!DeH>TdaOi4+O0&Zjn0$&$%hHrE@bjD50FitYe<_gld0|GMta8iVEKw(+_sq) zeV(yFq@rUkGVPMYNij8iLDf$F%%hp29LYjL25rgVu1iF7`FznV|M{ZYQgixox1r$3 zsKXADElhULW`T2bj_lTT71o#Ol@F@iSdxdl$W^!wPN}S;dcoDiaLo=P^)!`^OB7~7 zA1jdhKu@x_^&4T8`-#!4Dmr4_NBXB!iD(Qzi|1+BisbdeL=A1eqFa;C@&`wBkwo=A zqUdsz)HKJEPovh7_d>MV0Kc2V?3~1S!aB}G3 z6ue+>4)ygJC3>f>BTDHu6-}Qx1RJ+{7oe~T_T_2ePxKf6ct zSLBTE-P9z>Z9Ls@VKjMNB}2|gEfDe{<3-=r$%-cA*ph!$vP9o33G3%4kV(G3;b0(- zqfs)EeQ!>xh38d$yBSFzdY8;pHx_+#+eB)eY{=>N6w8fJz|R*Q!xz_Akq?fS1!ws< z)O@^&zc(u7;;Wz`={EPUcXJpCi6(cU@aq1QGNERJBs!a7B3ODx%V zp^fzaO(S~+cImNI?Zjoy1sEKbK_;zPM>5j`Nhncf_gotJ0gpcH^C6#1Trq*PSY@Gq zQ>T-wBRMiOCYxQJ?m#bU>aw8m5@gB*7jo_O5V9t*oTkQ0lg;6Oi70$KIpA4Ne9g{~ z+@gu()z%79NCH_v>n$WFx zLobzHv05o!fFrax{+!^$h})LLe!WnmCr-7}^omaC`u&4;bw4IQ#zm0%*D-nEehm8_ zHxQf(Lm0Qqj;0HFQXhRD4Bb|e*V@+jPOJ@@9UQ~|nxIIdw2$*U9yeEt_l00i=zU2~ zDZu>WYy8~{o+KvBj*l;`p^Ll@(0BJo;oW!dqIKJVhASRIUkq*|@24L0+>rY;CZZn= zYg$3;sw7!|$q7<6c00Ral1uL=a>Cq^E)5DfOy6&wMpIYxQnO)$xamX(pTEG3cNwqB z+`S{&_=#KbQlnJ*Mm+^jRM4m2l}3^G+MCE|{|#6%bt}5nAH#1Im^jam^`iv)=a^qT z9B=t=E%Ui9!7LqWP{@u{e&yC+UcPh;TPwMpT@X{O@j{gtOp7LOp10$d!knWO+#xap zzGRG=C3bHY=tI*ZG4WN$Km0Udx?%&pRJ(}eXC5Y7Pb?*0M$hEcO_%d`@1zKQpHi&u zKZ}Go$dl7THqUa&RrdEmCY$`rlGZFfNDl~gGDS9k=d1hELl^B>nd~AKFJwo=(zmJS z=`^}y%quFM-awlR=CHp)e|tjbP}0yWkGqU_^7nhrKv#zbleW7=kNi4B_SbGh<*#$F zpK34hIp0Ia>U*=C&k|%vuF(G+UW^idOu^TluVzgy5=`=dKiRha7z^tPC3?1nWE#4P z&n}cguXe1*UqU=-#H`&YTZJQa>f=e7x(AV5CG5lh+b3`o=g{T$XQ_dXFi)#95y$yK3&#E>&vF`u0ED*WrY^KvEeoEsWO^2wc7EAogUzC z&;DRd>)rezsY-tC&3?S{L^E{_N+1ePm1&Jr2VUJ*OB<_vYT$;kPm%OICR$r&PKSa}yzXYE8O27{CjoA0QEMEC@E(!9q z$A{M0^MRgk(XIAn?9}Xd{AiyBdDoYZZ&v`#-@Sq@)_zOjp|P;A0f71Zqb2xoz&f1k}Xb`W12}rN!+w8cwf*>+?HU&q~|@PV``V-9Nl#6)hC0mo+vE#!(bZYKIdb`ktrU=<_9fvlUd!`Pr zwmE@YK8EvC)-Ge+$&LK&dFuFS!F5!tC{KRK2e6MNhv|=f$M9m;KXkS*b8^vNnLlt- zjs~UW<3n^j|H+}5rp!KyGe1iaKYlT<7#nhF)0;9ZwMh}TN35hS+%oL^C!3056iKhD z9)5LBgM?o^fu_ffV69tBN$J}zJA1fNzqBRv_7pV&=nFO6l6`x)&G zJw@+3mvFn3r;)h$ne5~3zifSM9kUwt6{}0mUF=lQv{~Ta>FbUob58Y>f4Qg0 z{ZHetSHehoW|k`ZnIB7QP8H(k#xCfy#hXi0=T@WB2G;zA`NP=4!RbUZrHy9)E?^BI zdStoWLy(aur5^)d^F}*-S+Ij5o9umwTubS|nJQBmER|&=7h96AF%5X>9(j`9HHN&{ z@P-=3&SwWkyrMqcA{Mu17@NG;hnUVCg~N)?**GEdXA?dTZ{7cx9FJ)rQw)CbZBk{- z{EM_WD^OBwU6??m@2w;n@p1%rydoElF64cO2>H~qLe^1h!?uKsXL2p7EHfj44fM8A zv+du}q1uzUx>=gook=58HwTbct_W{+f6okxBUo!SFm$e!KKVQb&oE11e)5L6X@WG- zn!SdQSH{eBmcU<=e~L$~`@r`N3eLN3TcQ>@gP)xdjW^}Xu;Z?msQ#&=!ZWuFtF8+p z)h~d4%?~FTf9~)Bg?dciXB?U6XGWr{*ON6{ZZK|`G)qiXV@?Ni*kt=6@-)SYxs9BS zm9mfG!Q$O+LH%l1CHLDra=%k_S2a zS>Duce&%Ccn$WMs>@1hDBOY@^UBV1v?1mO%vn7t+IlqmKvluOwU-uF#bqU%1WDUAw zSPQf1IKm3`KCqJ@DPCNWhBWKvvT6JF;&Z<*q;p{!qZT1uu4VjC#$?b)eJN~2F zqV`i+Z5P^gK*yCriv_a{6s3h$C023lutM^8Q(hXjlX9M7uk8}v$E_Tl-0Gd z&=n)u8|97cl_ntAuEc#y4g07Xzr#nVf z2%*woQmQtNrRoant@_dA{qR-vf{=}{Ks7E8@!wR*EWM?)tTYafh;`hjgG)ja>7NPCr}fu zrzgIS!#E&{T3-0c+c_LYcfHQyajNrJ+utmH-;)WXGsuf6d(RN-r$mv;_H*#{b`ihS zLY)@t2{ZIRdMbu`JK)*rJF&sZZ34e-16^x+pBib8`N|` zj9|H^|Iybv&vCZ;AFBAgju~1zG9_~tl;+t(djk%kK4C60I#G`1F4n+B2Vcq`ku<7vL_Rj$g+KUD_QE}50y>{LiSr! zNs~pw+~0>>Hl)Le%~H*##-a|oGB<^FP-F4$qxaY&xfzvv?fdvIrS(|v#u{=fZW-IL z=q25G@(R^{+g90*EpX0%M=^LFq~pzIF)xD>8hmsFxgoeRw!FQ|Px6;#K~)RU=OTfh zGbx)opB$j?@?y#Dxvki+SKwaubfBKBGc?1{==DW8bY#?hn56fJ<``YXvwscJ(uEn+z}bcw3{@p< z!NF8w@j|?Js6MlN>_yyXD&a@lJy=~cBB6hz2<;c~^M#(}hQ0tCf+Co~*pIZ}-y9lu zXd%1*HxbRM+Ya6?6WBi0*|g>UBDzuf0XBPmfalshsfU7yhS>kctGiY4oV|>dJqu!S z%|nTk$tQlYmo&>85CYwD)98_VJ89jbwbUXcj;d50caOdf`bNzgWJztCkUaz?*0d*$ z%#V9nv2j5>-WcUV=bV#eb4_+LU1LwSZ2t#=nN^Sa`jdpr_-}sKzUkCFQ-fjg}$2T!bDOfQEN;+2y&lbGBvvisvBzAgxpNAf0k&+6^? zar`8<318w5U)zJF235&M&5Lx8EyLX{+SIXS9qzik3QtNfr6pTuvH6UfI?r zWQT@R*>|NluObRRzwm?iUU(5p%#TB_jICMZmQXlYpMaW8AM%Nf|M=kY2V6>G9X~12 zpY^X(C0k;InSncFsr0gQ{O!tj^zOENIz8tDn*Pz0eyaw|%@{%dC=DgQPp!r!L5jTo zlQe$N`5NtPETl_r?V)j({}RjL)~vp_m7Va7W4D}4@Drb_bi}w`GIGH;(jlygsqz?f zXO#i_9aX_jD#i;WI1h0oEmw(+%j;zM zo)P@1%!&BDm6*=x(qVPWtx3=25F)+gC;IR?gyy*ap>wCcC8j4#MF)+fL=%UH(2slP zVaJkrBxqqe@hCgQ6|JqN5{3rEXLU9ezY1cBCc)^^uz05J=tCD@4ksxqjEHuCP%m72 z6RX`z#U@UBXuzV?q-%r$>2$)SrOnzc*!x<9%?r zFR*DFO4#53E@QLkRaCWE1>bLk2WNN!L2YWZ2MUf}{h{FkUQJ=$9 zQAYD4;+;87G&1fcYL{qXnF0?ww0``O<2S2 zyTa?TyU1dhE?ZNpDylzHMSguZB}>iR&|U2r#M{ac$t+0bt+h{}6{h3K4Q*eXF=aB9 zeB{B-%UjcerrR|1cM3h{zKlIzl}s}p3}+oF!$ryG%gKLlW3hP>BQKWN5Z{qAiO1W; z%%et8yynj{HfeJqj^A#83ob=q&-jlP=JbP-`f^ezG@$PeCYyvT`4IJOxq^Pe6vziRjekmlL=v?eH+k| zkB)4*4y^A`V`Zbny3E3#lZcDsC#wOgo&e~-v!YDPC%%jB_S@z|mCvcN;$ zv}349%jz>3*58Gn{VpfhXS8CquJvqMdodpUPzQzRWY9)4LKg0vPwNFAU;oB-eB7g& zdQ1Cbt6Pte));rXKGchlzGPO2@VpbLhtaUzY2)hV>+l#Uj;Bq@hWZImGS8yXvB-l9LS!e7B4` zMnzR#%yg$AuYv?;gezGZ8;(o@_mGi$4xvB(QpDubCK9Z+g-_Jir?kD06#n{xk9|MR zx+8r^{M3asVO|`@|MijVs$V2zb2zE|QAn%Zj7W{ePI`ZioM@S;1d)i5B;gx%iA-Z6 zohNYjJVs{GNmp8lfsHmvG5(D6*G}T&j~$}73oJyTnL|ZZ!VL4WoZ&QHLtfPV>K2(P zIL4$>^oWv5KHc1DgVXh+g!<=iy7Nc?DNmNBNBKbPF0gz&!dD}e6^6`i02P(I8ndt~4Q0o7CFJ3jc#YYW+NnZI?|bC^7U#;E^M<0^ve6_N?++7~XDql*PG_Z& zqU}8Cdvb)7DEubo6U)iei#LdLrQk_5UPz*+UKDue2Z-C%yW~{fJW=WXxxy@vGg;j_ zQDnQd5C;Se;nK4BwaCg9lA@3lMI!mSkXTooA$X!8iFnh7{mtIt<;tp+PCfs~^Kn;ca-!Fo^hIHdw&V0nITAKAi%cyHCvHEN(M|2iB17RkQTk!T zK1g+7v9T+0?*D*d2NL;;!h}JwmM9wBPQnXo)l+W2tW}KUZ>+-uu zR!s-lGBz4yFG>^7m?T!?)j{r;ChlNa72}FVoVSI{3h-TwIZ#jqmfv zM3ozMivqqaCu#!`ctB=3PZusml4s;al9hKv(3K&YJZ2tQIjxLbyJIY3{hl~6PlNUy z@uPRT|xJoh?zf+hg8aW}Gc*;E_i$=|5K7rYE|HO47 z$#r8%mcYjTtN24;^o%Fr0TN`-3t?v7@G^;W@*;j_vN-YFM>_q9z@_rO!W@P~v7}+f zX!4vJRJ-I3ub3q~4=!4YCf~~@_qGb_4f{MWiwh&M(dndN;3;W*^O{Io)S^R0mCV~n zLSQJ45%y!9I9PKe(VLl%eGbZ#!i({^M#v?PyLF4EU(Fy$@Gd-B`H;l!Y$i&Ni%CKr z7VYpjLwp*9`kG1>NvOP!ty-4i;36xLjd>k8`OKJny)Ybmlq{{>w%!o0JbjV89($D> z-I+&rJSgVP&g)#YbqlF&(WQjEoHPX~0zp-Ma83{c#jYPUjie}Y*Cg0I2Vii4w)cz|WPnX#- z%OQRwaE$}$^^C-~Zg+AA)ztAi(EwiU>4zIH1(SJUX=H1$@Cnw|ZL(^Cr3}t76|EpZi5?Q+5A5^w{Kf9k{ zBVHZ8N&H|#2UCzai>7K^;r}R}MkzD$*x&22VyE}k;L|uSeX^8| zm1v@&3IbQ9MVH0sk732l32e#5IPt3A&FtQ=-T1F;B(n;e!|d|vSyO;K8?`iynR^Sf zA4&4$#uXp3W8XMh^}GgiO5t>7-f0$Mmd9+qSJF|*BBJs6IiFd5hn8qch*w+{Tr5rZ znZe;s7IZd=4KFKT%S~z!yLy)K8Ch&i;x~M{)>6FVN4~f(-;6~hZe;pdVQi0!KkF9e z7t@fCHJ8t2s&X&c)RBjo?b9g1F%XSae+mAspaAx4#vWF8Ae$AIr!cvn8)(_^CYE+@ zthn$%j5zJCn|SFuNijM($j~OBW^caGX@Awk3zCWWPtiuklFQj2A;+_5^w7$0eKKM! z%wZN+6jRTen#e0~JlQ%Tf$a8bV2Zaj#GME1#Ir*!#N4oaHh%ASadhct@nNY~;s*vd znC+{6s{i;JT{7`K4LrMmt$0KQ59dS1H5W0jPagO~@orY%b%FYPIYzy(uUJvOoL$zJ z!9iuQ?8W(mtSdB_Z5T72srg0GX-<@x%8fybZ?9#WhW4=0kp=9u^k=4&_JUQ1b>Q1m z^u;;MP@GmQCEk@4C+=LQEMD{5Tx@yPN!;;QU%W5ojCkkvrQ-YfQ^X?e#ZWy(vow{f=k&}L_T=F{=)Z&L?v+oHv!_XsTNrBlQ)PS$Md z-$0hJBV|!gZTOOU(D0jx&A7cTxW{SgWJy}cV8~VPEqpKc#V=j1; z*sxDcyyeIzYK~LslF(?uiH_Obr^5AQ?jTNzImB*_z08!1GuXQA^TdVckvL0gqIg2k z33}H;jZWKlifk7>q7MW<`*6itroHn5v(}ywYN+SxXSl6@He0b%h1ggE(ceCbHHBods2@j}d1Mpw zteY-QDEh*lF47ZkPA#WYG@czvt6&EWQ(4^5X>9406jl|on`(8WvH<5ZZ2taSzER+z z-W<=72Rrt&Khtlp-fL0J_2)h26DwvZW?$Im-IQIuz8ODoi(_}a?Zn>6G2-3RfL)v_ zVja&V+28nQEMiYFm8fbaD<7os-2qeBwuk9VeBcQ-d~ZWf$VAXvjv-{X;ztxeyo*IH zQD&8MgbdA6FDl!tAh>sQFxh8H_5Fs5|9z1VOSnE`cXzxO7&zb9o3Urv3+3^6^&M~K ze7BOm89A9fj-STtV*Kd23~8!8<0}4TTSU)khcnNVeL^L$i~9Gi62kNJLi#1g8viz5$OGj)Y__N`u-y^!9(?%e#v*PCmT!jBi&xS|BH z)a_MbKiAFto`w!O?{YFPe`|&K)^cs}@@uaJW>W*z+z8@O(_D7&=t;Uv>2AgJJww^f zwB>Xzl(9DVBBpdXOl%pMBz7uN7aw@lb?G9J!dl+C^io3rd+Bb^=Gwb6-|heSwq%}F zpgAmNks`i6kWO3oXV5bHPWEfbJ~oj1iv8SLUb$#UO=VoQj~MMv5PLr?7XQ3f&sL;r zQb~b*5tH9THLh=F3Y8&biQz4L%tg#Tb@+=zEAzz{D-BrMm8s&=rRUhB9Vsjz<|+O6 z=@_akAI56hg)DkO3~PPtjMsnLA+CI%%yz!4q~CRWdDK0b&W@PL{J-vH>D}vSong6n z$c_u*CW~TbZoF9V0Ul>c`i9V{7bl9(Y=1$e9>$9oEF5HCtc?Zl@MvawI*&b{_=GL)$Yw{v z-Pru-B=*=dUHrq5im#kA5WhT7D3-ohEmn?`WIgo;;>)W&#jmgK7t2~5q{j#9sJ^=n zHC4!AZ3i}pi9F61~Bt&%($Y9W}Av99}%<;79S|fH`#U{5|;X&VF`p^*4I#eHyi@G@|hihxoSG za@-`p5?Y;A(zkblN(3Vyr?p~GU5lnsKJEU z>H1JNZzB>fTg0w7WwCe*6VzZ>i2LkXY1GvR*h}P1=Os7M<$tPaTlYP*vZS01)#$F2 zQ2xYr_=Pj~>4kX8`9sWisk-0~PhpN_)~tmKV?)w1#HD8=*|!rGbiubAak{XkKkW3I zp0Ltk&!Ywe{!3-W&q-2baE&T^YMW30m^V`yO$C-|et?GP*|EH*yP1ED3VTZqS9Gsf z#3xOACivacu*1z_GVZE2eQI_ZcOO(Ep4Z}OkPAo7*iNDUd2GZFchADbk~(DLx&3(I zixOD)P>&9uc#WFY$K&i7>3GR=%np>)WAoJWEMS}md$1;n`ixq~D%ZQR@I9C|$R<;_ z?1fY#IvTkzNWk*`r|IAM;Vi4=7yIw;Q1Rs(5j3U;GehC|yi-3z+&rpCyvw6XtZ?cW zD%wAUH7UB1E!vpvdT)$M3nsF`)$jN-o|mcSKsw%(xPfn+Gm-o!C$P}YouS9__R#T` zcPPJbCi)if1`pO|<5x4%sr^3(O76YqTg0{~BufSNs>_feI`hz=vJwv77DkV_E#xoG zIYBS*a`?tWcj~n(va)de0QU0rK(48-`~fd*ym0#yI$e7uRj?NG`^#VBtwBzdFRQ~# zE5i9>P4}tx=u7nCBX?T3`Uiiku#vRy&OuAYOvI-X^lA0{Z}_)&7~3iN4#V0NAeA45 zpJrXfhChaZ^|^CMR5g+fDN9FxryQnIUOBAzLIR$-*abI4O=pQ-IaGW$i56UW!)vT8 ztPBmRz+vki(Qp49#4<03(51%(?nn1&BF_mQh4{~ zZruJ^gI?X3j&AuVD+rv#6n5ub7@KrOPZXqVNRNoj=`=QrCVkDJ@qd;v z*`F8i<)-ymF3}#Jw6e!druw+;!2)`G)(NzDT^xtRSw1^Y30m@$V>Ec7=p28Bk{^d7%yE}~b z)jh*wqnqhEgC*3sVv1Emk`9?%GXs0Sy@7*2j-r3wI-s%(nVg%XA=+ee9gVR$!+#Ze zz4O)^(Otvuk{;MwK}ua$e=U(#i-xpEY+9krO6TZ_<2jLnEkhLLYDV9w*gdO$m{V)equ^x)59R% zXZ8cM@1PSjm@TA1RX^wldj)(q@T0)NUX4|SdQzv)J8U)P1X(y&M)p+g@EZ~)LnQw9 zZ{gfwk`fZKvJ(I2r7ZyxUfaAkY~0~vw$>}i%M4puSa6)VIhuy7EG->oSxBt-|G!Kn zdC+an0q5aS@ZlCloT(dflAMK}wU)x(v_g2cI}b$OWw7jvAIz|^fyatBVQ8Ziay5Ae zTSvdLa=r7~>OxhU)uMzSaP`w|aB+AHHfqJtkevk$LH~hV{Q($#=O_f(MZ=)VFtAk` z;O^%>iVp=Y6@&K$XK0*e4GmuuVA$M7 z?ud&phpYeL;Au4@Y`fB9G8lL zZ>@{rOr<|m)(P*gJRNB|9zer(r6cF0Dj0f$!Nsxy*zMp7I+>Q>5wsD~oDE^g8+A}f zT?8qv4B7*(fc4M}sGnd254&}tVxtlGtT_R_C(gj%gn0OtsED$2RnThRdXQ5YiymvJ zp^Zyr(XP>VVZt_9G^$GmEu9Pxt>DGpwI+=|o22(C2eSe$AS2KDNTA!5MDgmW(L*bU440OEv4l};@LEq->pdTd${(Bw-6vV+V8%=O-8v#{b zuQ=o4NhnhDJ6KjK zR=HjVy>rQE^WSK6I_DhnSRw)DF3-4erls7?k6mE%z7Z_Gy@5UUJ+Rf}6C|wu2}>XN zg7d|dFhR`%wEfONp3^x9eqRng#{0qEbP`a+YGrV$~li68-mPTZwr1nPV8_<$Ky}QStrGw)fcg?rcMf^Fe(^ms-Lx^Q3}de497MorZL?w%o>bYn38QVG}vB!Fx2AJ`@T7dA|M z4he~~VPJVAdam_$KUuuEu0Yd=w4+d0tS={6Kn)6iAvYp)Jm`Xrqn{nz%C_3L;aWsbm!>Zv6?j zV=1WZn+Q5FA~fs4P(%Zo;McgFa3t0i!g6IH*Q^wBj~|Bjv5`RT6oawDS@?PQBHXrE z2k)&?K!VSO^>#P8dyOsJMU^$&ry=j4YTzwg4s8ccO$I6>IXK*53%T4WI6Cq)xCEy| zV2%O&m-Q3|#wep>ZKEKoN*M-CW4Hsix;XzKUEKAj*SJL;p->rd5{wfAV3Kn^1eV3X zZ0R8|dfg%JXW(gW$%|v$dL1R`RhEUG=6Y^=0_AF|DmW|Kz1*_Dd*R>xdGH~3DEOTR zEb)inopvU8a$}#G)S0aCF&q5i&@g1u9Ezp+Hp~JiSgr@qfjD zt1iKT6Sd%G#~|vyKODNP4*UI-V8*sFurym0S_l4cn?2-!*C~UgW@TV|G7aX4&w~mn zfN;k$IO-k;(`9x++_*q6^7saY@z-GD{L|q6V*Go1sk)#Q66@p>z%PzPdwjIh9uONf@SI`q@hGe!`Bkfre(WNme$fk7x1lpOx=cf&v z#Na;2={pFi)zNUg{V-TRDTiINgTdhUW@s&14!61&!+#ITpr5}8a#ew_z4a6X7F>mc ze%as?JPok>WZkxbL z&H~N)G!A*ZltY@IXF+e+GPrs-8Y=bFVeB|1Xg(|pW_69QD827rqq#gTAR(A>fJ{mNn`@8;%}B3zFu6@4K1s>*ES2zPTS|)c7Eu@8ePR z*hEyZ^8l*1jz{k$IB+&L0GW>*+>E;buTIy&XUnIsBJKruxTck>KbyiS;sNN7e*(X+ zUWcW5R;coXXNEA?@i7Lhwbh~C$qdPyG(+i+%~7L_GW10&gHl;O7bjc- zo0ZMby>SFt=_2^8-~<8w0C)PE(X-zLXrL|-jg7twTMAl1b#o1@wYmgRk8)s`O(c9z zor|V?GeUOd`l!Ii1XZNyqH7H@DD;yI6ve8-SFJWK?x`I*It-!H-+zHmWd^LSIRd_0 zlb|!?CP;@AL-J-StPO0?il}Mm%xE}6=d4NmtD42Y(24rJ-m{fWKwkC{0 z``SmMpvV_c_~!^T%N&EgQIYUKcqXAw{aouuWw_lIgfjQIApux{j!bqz@m&^Z`btF< zHRKG0jYx%0FAhNO&bcW5k~`Y4YcU%3?>Q`Y6yE!F1lVuh3}MToVfUvv_+fMdXzE39 z-JAt;pZ$ht|C?Zc=>XjRy$X7cY=nP9yAdgkI=V-f*v-w*A6Veot7BGmZ82Hp5i5BX-cz=S_lU?g7z zy_FMS>}@sh>kH!&yRu=d*FLxtu@~x`MshuuhjXPxy;ifsw2;aUUDRr$j|>djxJltJ zIK_fNZpXocFj?Uc>=+pbO7)9C&T0k7)CNPbTNpHb+W^KczCagQgTZcH$nH^s$(IoP za<&4scO${G_b5C+QV73&%AxD9Jxs`72L{$|;OyWJqqO{l`_&lu>0<-WUUqO^;Z58c zwQJB(p9;|`VGymj3Qp{D0Hd=i(5Scxz|;_WWej2N2uC#ezAaMU=7_3L9=xCL*Z&0GRJPxf$L!Z9hekB9pu zE1^Vw2e?0rfQ}^*@GZ0+%JZJUeEVuxGCl+Rl?p&I;XKe{KS;P91cUt%kTFpMIz{rZ zQAP{gSL#EHjwTF7{Nmn}W>kSmizi?mZX}b-MfU(U?HI=Z-J7T4w<| zI^&@I{ax@=t1n8$f`=5r|@lu<#83Q`}CMx!G`;pUofFc!tYmcPm1otzC$ zdkP^u%nx4ebOx5G2LBXefp{jt3(IVHBVU4=Rr1h&t8Db!Nd^?3D?s7pmz+e{I3ydU zj|Sr>p*cSckb9*Ox@^!IByx-+#B*;ieKvx}=BP_7814vm9T_l60r-A|lj*AyhbYAkB?&_*iP-f{C= zdbyBil2Gw=8d~_*0I6`6XoG-Z>x_1UIn@&(Cv`uPIJ+E)EM?K`1T9E#P=(Ui?;K3u z4%fGCglVHZ;l=vHpm{AEKKL7h3A+mO>MlVKoQ9hl3b>i6r#PLj8C>a01$0_M9zCBs z67A7F1k(poAa&;VG>fA9=ihak!Y!Sm^8 zwd-+VeH{(lK3hmn%LKFh8IbtB46+?n(Lo0l)Y~7r?Ln^Dt;s1n0N;!$zABUMTbOyXhI~6GD(4H6$qg@YoL1;!2Gwv(dXL|$i4pynC0B(Dyb6CmB+zng%il0T>`oC zw(vqz0*MY(!SokV!Y(coX5=J-_+up4bT@D+cD3B6>^zQpWQ;W28o*@UDJUD*4R6J+ zaLURFI?a!P+DCt|?fA%Do^}%?l3Ji+LIu1%G6MY$d;YTt##~_rU)iOxjWmo)b=knPWW+@2!9);XBZ^b}+JK5um)pnmleXc&Ej zu?hCzHd`Kke6Qoumdl_Pn+!0q@P+_8bL2Uwjx=sgKz9yia1q|g+%WAuTv|{)YHBM- zkrq*C@5%8{rLzEb%t{AUDH$9*T?*@0zej%`-G&nV8ZghS09Ci4&=I8onc^1iS(qaF zKGzuO8yTU*32NXz<{KBPeu`UYZGk$qtkJUvI>_|eN-$b0f{dq5u;)iQde^icSuD*! zPgAFGKD#z?N`rH`#)pY8;l@FDIBB1-=WU0lf1ZQyeKjN+o(#8U=EC;vm*CZXo|96l z|V}FD0e`nxcR2ICMEJF7RB#~Ou zAo#iNg?A42V0%Oh>bKkhiL6qXp_~W0qy50?@HXJ?uZ7Vc)1mc2JnVDX1XI7nb4ew~ zxu(h%PH(9;s@kN6tZxj0^jHlri*bV~WG%P`+y@p>19;sf5q-Qhx~)N*^r2Ecs1WpGLN1-D?@D!e#r8LpUUgQEw2aBsGK z<*NGbajB`gutLHY9IX$7g}gl+QgMMzmu5kco-SOC91b&gn8KCex4GiyZZIY-7Iw%V zvQqujWA##F7@1==`9<@{V~1I^~Y$zsJvQkdDm60 z{$DINX618ENzVXwssfa#8(L`%8gl6~uW;}0%!F5oCh*!!SPxA};huy)fbCb2}xzpKR zoSNo6u3S7GE-$U%3O{FY;w8>p<$s#oX*WAgyC{a!7iuRn4c#I7sw2F*6VAo`_{^=n zE(54?2ep$X2sv~BqTg+Rf|?j; z=*fi<56{B;V{fh2Z`sH_sm$iK)W>o=9B*@LU&z5H+dNL8=n|*6_b4apy%|EBcY>_y zM#z`H0b_=phl}Gb!}mKwxIYgaxUNzwZnnjHunv_&=5rL#&t>De_vfy2M-D`A*Z23p z%LqV9QtOdL+gWbXs$sA`bS8v`TXVNpgma?J5u74eaxb2%!-S|%(Dk3iMdD1(gk^Dq z-M&zNG7_eo%!5hiia1%rD_p1R6;AcqZ@5$R1I}zzLFPyASUqk|;I^MR&#g!m&P}0i zebQG0Hc5TuEJZ)K`VKY7wY_gOzj`4zMdKmY@~hD*db%5zo@>N?nY$6*JqZB+OlPP* z^u=oUGB@u2x5FF>u;etpRd6NFkGRA!&#i8`?B_ZbKja$YE4b&^>$z!1s<|oQK2WpD z4;C-C1{={UuC7-J_RpRIN!Ju$fvYl@9@l_I%L8yRJ|1dErbFY^ad6xO!(_{8pp#<> zu}5v;;{pQ+)lA?9rRq65VLi*eyvA*Jd&y~TQU+^TS1Y}Z*;c9ln3aC{5w1f*0p_ls z3+LY7w9=ZfkXz|q%_+X?<_>pub3gaCak`34RQ1p+Uaq4}UlghvV(-dG7nV z&hvL(mlO>D8%H~h_K{}X2`b#LO!Je%DdJ@^J>OnNS`{K#Ek6cox2EChe6C)7q=Jht zO-bd22d&GVK(THiCeO{rliPc{z zTKD56jqfmm*|Qxuvp;yFv)Gn*268~V04D(KO-uPgW!2Y9a>#Q*0$UUmke?oidFa32*3xPil}j+P;g%X_8}Dy zB5oteq>+?N#i1SSggw>gNb9i>a&Ky*_m>=vZ9PD(--N)cT}2HcQPj0T4$=OnP1IX= z(2+bn_$`TrkzX91tcbwrElIeo@irIIQFr;qhffVCBve7`LpMuI`zJ%ztwc;b2Z<1Gdm%kx){+ z9gDov4am7s2m$t(g3l=-D^ni1OH82i#RK*)=i?(+Q@*mBh8@ZZIOH-LD;AzXdC*=6 z73r}HS<6|S`!c3ddz9k1v-e+OAjKTa=6a>;kQk@}(qGcr)js&aX<OxvD^WP(+1T~vT^{=L{ z=MPind?hBXyqV3+^<)ZfVrX=e9(`W#g`U}*=4qqIJ{9a@krEroF8vl|cpKqTkQw<3 zC6g_8MsCg;q-dAJ*tM?!q8+OA`(z8vvFWAY8Oz`eG0(w~E0lV(o0{b0(EG{~$98C9X1F?LR<1^BP_)VUcPYg4F~wnn z`z9tIHObxHh&mLPL*FG3g31A~JfcgoZ;asOwHPl}gyLc2S)}EK;dE!L@pjwgZ@=NY zOFPtN$5HoGF(}oFBdK%)9*RX^AEyEPwTIDU10ie+5Jk+ZGbEtZN3T-GBF4<#q{Khd zx$7Q%RSbQdXnc~651W3i-T4%SAw;g#N4MVu_I-gp%byUzzKZ&zia!=N zP0}DAtj3Z(l-U>Wd2Hv^A{-Q|guZY)+Df9Z;IKCuuV-Uz*;NeXb>QmCCM?YppfFgG zPsv!?SN#Zmg{|n+m1N)Zv=AD#9`=7{BDQ88N~XA>-^7j63T3fqU;{k1hSzoE$NdLa&n%Moa-JTX$it1BGBBT@ z3w!SUQ@ef-vAuUu?-ho>y&?G0nu6-ZXJL7}mhz@{(fgN|NOG$HJ6kq}@!L#e2i~MX z&9xZ8S{3l{J_o~Fv79~1mP3`49uqb%p4u*$RV;yWqw3C{h{NB3n^OoI1s6m7i+x z;`SYwZxdi8MOUaQVU+e>9ws4k8x#gu;VyUQJ8)WW!z^F4HH5)=awc{^eTMp*J!saF z$ALvUaD3tjdDkbz+sH7*Y&AAY-iCB_8G?E`k@aK>o2e_vnvTq3+}3i`3V37cx(P^& zx`2HfVz6y66>r{LhJ^JiB>tPgu?*G-;d)s||8UGZz=xi+324~L(CEYI#d?iUs^%7O z&-6x<>H&OT6N;WiqxA620L@g;K+)$8oQQmgb9Y|hV7VTGHm*R>#o35gDkK@(HsVYF zLzm5Rp>d%Hnq&HLBI7yymp;L%jzlDO2awdS3hI4wiN=mEC!gjw^x#zu-JR7$i<^JZ z2X=>K=k(Hz$r=!g`9#4ce7us*3A`H#vv|$#Pb1yh4;d32kW{>l4eVdWu5zq_&E0A! zj0uIBOAsbJ+=>0oHb_{mk7*s(5WMCbJYCNsXH=YSh8t3~Xs?O2uqyN8`ww@|)Zy?Z zb(kJ7!jg7X7_IUI8@n3UzZt`-!Gv%3gVgBaJsIf~yvQz{lN|!-=n;Mda z=EJ$o97ei2=-BFp`#lcGf4dq!dtZ{%uCw%FLK5X(Uy4(QOc3;HHfXB_#HK1i)~uf% z&ijghxDEuZxP#k=9g*s83w!Ckc(y1Ozt?mAa>zpHT#jQ>>yEOPp0(^=R~s6pS7N7L z7EVh|!niO&ocJJsme=a2$diJk#(3!WXQJ)TF|--3M7r_<=+qnFS&%e-zl(*5e;l}Z zS122Q3rikdMWbyBX1ho->509FYHh*NE0$SHpq4{pM-(hy$6bZ?%s_C{WMf*r&@ z+u)CmHeR)kXO9a;Vc_13F@ut*Twsng7LNySq~JA!`~OafW9441=If4x)0u6E3M}VX zf=p;~XY|z8Y~pvgMiJcnMSOlB z^1HmTs@4_bY}0UmK^!JH1YoCd40}4shh=Hou?&qSa{FCGt-h%g5!R0H_C*L+NkQSX zSxihph<)|wf@*6t)<4*a(}NapS9*;2^*0dt>LRL+PsIMYVhEP1r>`AZ(7qRnd})1n z7K~w|En@6*`!Do%>tUSNU$V_Arzw#!Y;(pj_Q=tW4&2FW13=NG9Z%CLEw zoDXIp%$8}JVqp%0c8eh#gZx4ZpWgYR!S%k#hqCd^me`|`y$BCj_!N__hrdYX!VEbybZYXArMiCdr)F= z0`KLdS#*pj>%Anz4z24Z?R{-@c6T+syQ+)HjnlC=O%$RwS5W*g0iB=yu>0eB^z<*o z1S=iPh#aP-n1^(tKa{52m&f&i$5cP%Hr=_vhqL~jw90q~y>ZdR>o!9?Odk)Ah9$fM z!i?vwro_AUFA{UJ&SHme49foLGu>_O>}CH3b}w-lV{h|wnd^3#JWm0&BtWae8$oFW zFke=U%vtrgZOQfMDiP!_Y)fVJa!hp=KRYwE3;wHQFO; z4Y}!FW1)6U5@$~3V$?1N2IFfv{zeSzrcA=rPA!}-Jb~$XJF%{JJN~d0*mcMc2fv&^ zji)3YF>!qFd`x!D(O57k6tyt{m?zvr9u?0hi0koxJA8ucgFmP}{u-B4Lm|@=2dN_` zp*N(9yQgL%NJs(b(`POV9XZeG5Uab83oq_Ga@|#dReB*uB9kLw#^cDA(7ZT=PK;> z-NTlqb4Y)XidAu0SonAs?`qFMo??I>@A&$BxHmpQ@z3)ZNi0EjbummI-NV9VUueY= zdCcohN2{VfY(LF_;;=b(2Ipd-@dG&Jl;UnnEH3n3gW4aLu*@Zj-JfVZe$xS50Ha>owWwFcBn|iek_~ z03!JXEcwt)7Cc_`oV2-$iR%2W0#!@Nk_H6x)@M zm_88`3ZB5|iZt8QKbe{RB9tYVp?lpls05E?;=>9obLDvE;unR_N0VUtBO1{w^H}Md zWLA@Kj#XZp4CO}P*9gb_?M>!t(IniToQM)padx`B57J%S9*0b4Ov=o{D(AbfUiy!W zB~;+uJwn_54dZXsE$F_8gEyy%{xY=(Z=ow*bG_PbpLl#M^n_<4_r6%3g-z~7TyNb9 zZfGiA>oZtC7hvDjKcn|-5B^xKhMw^$$UNGN;qCR9Ys1ZLnt2C$yQZ_f>eJb&$K%=b zEyu8~YCBB%tza-(h45?F5iL-Jrut?yzUW5Yw`NSu_QO&iA7r@f!SjnZ>Aq(T6)D`M zGa~X=D5u`rWF4s|>?SP!k+2e8iH1^)39SnvamVcGWwyIwfp?6R#mSfPV2F_tK7S%B$p z$3nOA5(%uxqf_1K)F~H>MK+0OwBi_zPjireZx6cuTY>e`Gmt*U0Xqd3U5%_bac4y>Ohq2Tw5bVZcl(jIei$-AlbC|-WcK|CH_*E#2KAdu@NmL4=(Pku zp(_Y;8k6uVLInJ}(%2ZPhj_zlq>%B27QOmTg3H#S?&mqIdwv-@ZG||o{t1U+w82<9 zg4eY%g!dpjfR{MW47!l?b{Fuz~A`hz1BH&a!jxO-S zx42?V9C?Wu-7S#!_J&<_4%S+G!)3`iO4)pwsNfYzv1NS5 zi&*bDd91UajH@4{0C^9#$;_R(YHnahqb*Rdd@Iaa)lhr;GJ859kIfqw$J%bbpHZ2 zxpC~cs3==s`-7`hhoP_g3U6-6v&1VxEb7a9c#kc`a`YfG?G;QKxby1hA|y4>MM?_C zQ^!~!Q^Jb#5be2o{}`@wz1q6ZpCMq&$F^~2bH^Tg=GV2F<*(9auH5@N`?xP$#&i6T zmo5t`QDc`)$1`{PC3xzz4fn6@gwpUbC|25HAa^CIjRVQhIe^@s8B(k1N+$7T0h_Np zk1dG`z^SHasJ`~cGix^*8~U8y-0!29;7$x$3b43yQY?5^EB*K>hO$rnl-2VMZb?sY z_e(vl@^6Q?!6tNhs352#2nQ^a;QcHXUH7%IYT6Xc==wsUc4efm8bhMxK2%u{OEZe% zDCKD^38);w@UQ(SpX7>wBcn_wM4a2-EX?CwkcWKlM5LY6M9(_~w$o0F{d^|LJY+MHkb z9HD^ESBL23VgY>Nbhea1O`LrziJU$CG&$n~1?=F*M0s!8#CaoeFQQ1%^C{Wl*yb&STZKNGw@rjCPLCGB;m`!HXhLZyloCx({S`-JI3gZf2{0u3)pGqoB=F z@Yy&B8KbIr!m)DUrW;W*V;s1tNT`;b4%w)w=)9u^&!ggavs(&Dk^@xCX%`ENj$(Z8 zF0{1l$L~#=kPVy%W4_sN=eR}z&Y%8I{1Mhweuk1sA5_&JBcetTW3El(`n3fd@3j=G zO;$kM#Smk0lg`yd)5xB5I=e-bC1?sW^I5Mj?nyF9Hlz~?R@0;&E8eKE18=F|W}dhE zX8cuk!CJc&7_2;wWvRi~x+xm&)1K0fqf_z7#saCM0(i5di&Uj=)85SOC<FoX4?d%1yNOxIZ2nABM&)S11BpWVQaQVr4F>D8SWR7)wjzq8ABtmDyf=z!uCO_$ z64@qmKemiJbLUH(MEB4ADA>9g{OX*BT`mt7Cd_0n{KA!rhd3-TN)NQ2a2m;dnsn+9 z+FBlAaqk^`j}b!4GBLQ$kbuC;y-a28LAKWJI8(pyjy5{IC0FfK@(@^yTAO*$@>+$Q zoC;{{{fZ@d0?c%6A`FVl@Tckqw)zxO*zq59yfK@u0{l0BCp`ZNmPCjKSCzdgjt z1#!AE1>TGK*>7JlmMwe~fj^c)IZ=w+=Pie_$>XqExtEInEXFY}XPp0R3dg!jsQKv) zn+677lLqoyTt)NGohOOqj&y0A5d~#Q(8r!Lq+%CCr4x=(M;jku%)SlK{6D+;@<<7b72?pu^{lum3J$ z_55lq_*umDvpptv$4Zi|vJhQMuES1t4`b#Gq3TON>5dtoi+WNhJ+l}al9waQ&k6fP zr6F&n0q4JKU^0=*_}nf*A>(|e4sn!fPDHJvGG0{%KreecWEX_s8{cCR9pL($@zs>m zTu1{G%V{LIo}3+5BCXjTy9z>aoqv!DJ#NtZ*}1gRq#joL>+sgN61DAZbpOyBT5dlc zvwx|B-$ep}8~>7Xi!oc{WyfZgak)yjC6M~eabZ&(ar~1AM9U?R+Ny}W8h#udkivkn z4miUDAKg}9s=X2H)dXOzB#5xR!(=z=jC)J>;8gx8yh-gu^TMZaS#=K6WH`olqz(@D z_fSJI_*$QXDXCYHA8d-g3Wn6N^N}LPaoF2Kk=o*qg)_!r+9`f8p%)~s6OWt&mvEq! z)5Ey$SX90UN49xj-I1BRS#svQ1*w~P+6&w1z{hUNGj1o1SvPPtqzIdLRG}cRjf^Uk zvGUp(^vBlF?};yIwDJd)ai8PYsuB#C_+wp;9Gw4{AS7BB=2vQH#f|&a(N|5HTi@XH z%vV@bJc8v*jKCW7P^f6|`mFA${#2WS^%~&$vV~w8^9o z!5ptH8jDaCj)H%~bR)5g!gHR{+vUZ$d5Zfv<{iQENLdzSE5_D-n9BP1-J~k{A}X^! zPb!X^Sf{)_ll^SUo|X4fwz>j3R5v5~^i@_o?G_VuyUI3q9mVcFYoMLrg%kBR>DGe^ z`osU2+QWmdb(L*)4CqNQpjOx>Jd^vVN~KhvOauNq~7FHzJqLQSe>~Kx!m0rBIAPB<$Eysf(P8acEalW7LSq>RC2n%BMte`MVe-MI=uC2@$+KaHfSt$1V}LZ)4bG_NUvwtVy<HbII|8MpF=G+4y73G{!C* zR%B!4RG94tIaWF-z@|IOvz_+KaB?P>qsp?x)?c3>meLLVj7EswlA*bKwCRZIMDj?M zWeXn-!^gH0?jc4Hd~z>EVH)JT^X*hI_{#y^a?|;#OB`tcane{oGko z+Ch7EwUd8P4>f*10^3X5@yXi)kvuNYzRN&`muCk7It3?fSgbwQo78LSgL`$%~P>_Lm4ywQ_Ozm zr!lddBDADs;kDltbcMPBZGzM zhdsWnZGq08I^5v$2GR|Ixc&YJUJ83cYU~(xsalFTrpmDkTXdNG7en^&y*g{Y;ErQ{ zyC5@TF@)o45ua9uiYzXJXwk{NpZRpV^*V{nUB;RN7qBGFX{==~KNgmcgLj4u-c3s9 zJy1{NomdykE0^>`oA?3nZ+7Ntz?X1fWx&gu;B%>>T_VrvdUqSWHmju#$2)1MkN}Rw zoyE8QFnllefgm>@^iBRwRJJd}I4NdB6$kT%Bzd{frqo3tXXTpxos zxxz5<+70u;HBhtM0#9z{UEG}+OvQ(r$=V}~7F}@|fBi$tJQ)iv;j%N9HmpW&1N~4q zOP{Bek-GZ`^St(j?Fo6tO4eP5QfV?&vvV=P{SFO|?nhCy7tR<=VivE&*=!#XcI>PAt@5UKXlsxHFY>L}xw;KMS`sfZA?Nx?Z`To~tY&C7Wt2d{}|CUA2g|y@?a$1#s};m>B+0&TsaIt4|oV zXf(r7{x!1H-r))N`SmvBA!Ky|E+jP3L`QkFCre>QN*@}fS}`@e1xj~cb9~TK@Ski# zu<3OCX0q_f6vQ}Xb9VcxDw}phhV7B}hkM&T9FL!cIe*vCP+JKt+aFI+KEb#tp9JyN zTyCiNEtJi_iYm%QH~5fiD1lt*TO|KhmX4Up(XCcdk{GR_gL2QgT-JT+i@uJV?)iAK zI)Xb>On3)^jd*e=^>~Q}b77|KhzVGm=}zvK8ksuLl*_v$+mx@lDv@>x>Nv z+fmSa9C{r4F!xL<1~oHq#Cka$|87nptu|zJz7e;(n{i<#mw~wVAB`J&P9d8)jpoZX zB<1>IqJbZtW*5^~7EC7`HRx<}AO(yh)|tvYBLA0?eDn`4yr;XnyjJmWMaeY|Hy(sThGj{khn>{VIM} zYT)q(W4QiyLARwWrshjxmwFcsq^)PUcRblurPb`^%}^LB?15LrK{&-~Adq{0hKNX9PjKhTATt+Kn5zgB<;h8jp#IIbe559=MYj5Ejr9;~9E|k1) z;q=pbjHmxxR`O7qJ68d`j$1;PEf*K=ifAk z%kp{pyo7+%M=UPwz?Oz}xT?5|%Qe^|U26mH$JL#@g>Sa-wEW~DGqMQdj&1|r^*H$K zDMhPl9-e5QqAmJ?_rzE)QH9fSHhVJtY{8N3&v4JH>bDzlmXh^a`$uaD| z$V{}@N8o;tPiH!*&Q2m-yHnR*0H+ z0qC@d!re{z)UR1Y>3ucy=vWQ*y{bgoB_BMUr-ORG%~ zaP~oYnjhr%GrWJi00UgTweYV#^wwJ9)7)A3bg&KQ3NkV2P#Q$EH*)N=1Dbx$#J6S< zw&L^vR-JIf2OcJ{*KyhuF!nT{#) z`Sj5xlWLT`>9I&N5)Qq`+LQgj?nH#&aE9y6^>`koi;0~wNIv5?Q}8L*e%4Y zrLw3cqnMiCHqvE|3px8$9t}x-#Ak2}773wn89j}Q^`^k--Dv9Kyyoj2w4w0><=yF~ zpyNE0EC&LI%<)0~6h-xiQP7nP^4)S1sk68&m`wsUc=xa@jW6u%)P7bbltB|(8fm2# zrwMqhKw8cUESNP9CtV7#Cp`|o$2ep7TP!hu3%Y-A3>iFsP10VybUNn;H9AF`ELp$X zWNN&niJZO^8@3z3v?In~u%im2pd$!U0EtbTZ_LD*2Ynt@-GPQO^Q{&YRiv3$cS?c9?vf$kb4cY8JMoVOJszPr#k$gxKq*(AsP`yWm@bR+)>9r<{a<{4E} z7w3;#cPX)B9g588fh>yS=5d)`D~NlS(3$J`^rkkR^7K$V`S@U&WPsFXx)9kQjEU!0L+YFxvfVb})5`a6FgRhc@`+GE?0T)qSo)*nYuWG{`~Q9))cdhnX! zMZ2%rQhbyV`6k{cS?jm-)wzVs?tP`-e>sLmwuT%ebg1|F7|L5KLQZM9_&zln=C_YR z=<8xEJiHP^qI#&7QUia41TMFKqH`)Tuwi46BT_;wigO`w#03APi-Q|J$1aH4u*cq$ zn4;c0lhmbrWOl#8N65>P$L@eEuzV7Vj9)^g;TY@ zCq}qh^Tp(~u+eIvsY@P`UVR~*^e;pkm)&Y3ALQ@bLc_CcXsXpR3iaiD;4yUw(o2B+ z4lVd+8^E%wpKND6qraWc=(OZxTG_V;A5AvlnU4b=)-}`e>LHS`Q%7g+fXVO1_0)6B zf=b?$K~y{mw0gN~@O){?s^O^@tJU!{HHKc|9C>^QAaPM~ z90-ty@M9^QuN6m~u_lgx)PPlK83lbWpa7lwbnA%?{*?w|+MguY$oHD8ZqTCjkNIit z^A3{!&`9PVD=4g63v<=2uwGUbxhbVoxcU+`NjcJ0^{;fnj2}BZizt8Ks)@GM3zN$8 zVJ3SMe5s+{o*w!;&}Jd-+1_uC7guF);$3>bc?b(q7&Jfau=L^&&?6}m52$qUL$qxST?q2EW2;Kiuy}dlc5uj0%{gB zFSq$jF2sSQ>!w2FP8^y?eG#&?3XWUpya+d@I8i zJrlspYDIwK&=v7+8?Jd6N?`F{?Y|;(-yo{`Ag~^`!TL73iE~n*?gayj187D z{Q?p8ZGi?Ga*}4|4JXk2(HDlP;Rv4}3-htz2tFQ#$iD0N&o&hy1$Jmy%hmm+{!&RPU4qD32)fx z2CufgnU}cdCZxR|qIY*09Pag^Mtus~dU!IENYq0Am1HbG5(5`oZ!DRx0Si<8fDthk zIqo&CZ@z^)fBD&o+P`>_*@2&f%J7xo`g2JYOr6Qi{~p6*VzI_7a=$EVNF2{T?viCL zCbg*K81?;oQm}5Y55lFtu;N-DRIbRfF0pPzO)0{EXM}mL6NGq~2dDDt4Hm*dat4kx zSYw}q9k<89mL=Z^VD*2zS(LjS>!@7I4))(dQ9vR(Cq-b;_%1H@wqghP*n@LHP@bv^ zU3-1BUb)X{GTD%{ibQpy1y_?yT91y;wqurIw1nnF4(x!#U_kKN5Yg7}!G!+4gzdooLwbL6-&Z?5lOg{>;%A->?y*C!(>pI|!Gp z#95bNJ*LlVhSi_}dn%*CUe!u5+0k^&G`z;myeWc1qA=cy{-8T|+DXZ52W!_|!G6cD zWT|GNZ1OT?R`EoOrHAR^_v~HBzkUFB3fi&8)|gfHI5Wk$9XMb40O4iDIJ?FTpZ9IV z(C7WA=X;CP`2V1nm5o`Ja*X%*0|GPH2u7XCn>H?6#0UoBlfj?x9KexakC* z7HZ(Ms}VHz*x})^FLcQ5C4KD_gScTP3Lag?E1?vm`*zW;`+TTZ8;f81$FasL1-HDj zu)ydYp619ft$(wacK8V9t{Km+{T5|*HgR}M58f7@!|dgz9KRHe zv_sdCw__Ysoo2yIV-EB(HR1g-79GL^kk9*u*Y=WZRr5HGBeLZETe^|Qd*#JzsN{IO zN@b|L{*UHx{Fp<^3%sfOgpYc~bV0VCs;axlF1ejV)W<<2LmMIcUcmAz_r8vs!0vyH zz`MavRG9~&JaRHdNdYq&WFfFoi4CZkFy8T{Y<;vZR{QwFK%Db(-mQf3=Pfuawj8gv z9OrV`UYKyY0K7MXY}JLRaFjqW|z!_l&YDR(lt_<`-rOAZ;-zOKMQy#%G9{L zL72-AJaZCZ+X}_l(RFtyiK~--{_7_R-PyQNdm4441z6xH4etZG7_(RlenvyM7%++H zd^BapXKY~apRZw&tBx|~;fv(B^d22jJx6nz&#;RQ*{r9slohY!__d!JkgW*B4x2P4 zr1+SLP48m0wjL#Pf0u0K<~&7m{U4VpPidg?srh8l^cMR%T2VOV17xe>$iPqv$2i{q zUXn9ZsK#UfAmAkFVSvw5v%3eP53wtX7y!7x;+)f%iDrc#qa@=X8RV z_V5(s^fn<8)*+(94rQCNq+?5&{-*70i}V^?Pj*Ji7cFc~sO5OBTKo%n1M8guc(#2H zoD7^{rnCS@(hs1DW1lZBIZd$*R-{*yO_A@OpjG@0iaeVUSG^3{zg9!jgog*OHlaFj zJHpe$@plZDHD0_6LY$wujq}wa{|poUJ*3gvE2LpGK+dB>bh9l6bJ=xnZ`FB}ae3zl zjgPUBuMFdMrqX4{N}BeqjE)_eNWQ7AH1cp6Wv>28yDK~BRgC~bx~YLHPPU1Q)L#g~>rVcJG-y+rDB1At{x3&R2-tDYp>sej4NAxO%95 z9G2%*A|x{xX&fi?N_HlRr>vx-P$Q>95^Sy+AFJ;j#(vjk+9Ep+kxByKb2x+V#@?u( zyb3(2B@hc-k2?kuxH+*7Zw8;@Zes!dJuRl%EO`u!%Hqf^e+UgFVBMqDP`xlpjlYGU ze5Qxq?v{t$-``XwF&+i)91yL$6|1i}&uwPcOTeo;8?Q1nXo7j#lC?tSRy(RiqAjN z`t!r|Eno_?3vv*vkc5lpbg}nXC8aJLp#nc;JgxM_+`EZTJbwlX2``cUq6rbgLQJ7l zh%H|z$aFjeSgAuY3Y0Q&#w8bOGdJN!!XgZ;IgW?@-q={5h_lQ0BaWRX;hc%Ef3AZ= z&zy1P>}5Edx&W!_F6l*nLN{S`7KW7#1;K@b2+WeoiI&ln=niE*Qy)0Y(`vcUJ#n}P_85Va)hGohX z(UGitT4tO|uj@BqO!_g1=k9~Uc^O+YZ;@M28j|t|F!^rpfeA zh#+C77@E7klX7A>8_Er0x>KW=%fIm^=hOm8VQ3$+#ILco&QGi)UXVQ4&ozaU;Pg7crynJ_d$A!Z2$dhDui9 z?dz4u_XuV&RpCr-bUjNZizck1MahdAtQ?>@vsSZXO`|h#Um?kT*018-8jbZjUdfi)X-h zzZsSEn$mA4Lz*rqiQ=YKZU(_k(i-G=h#c(1xK|d! z?MCsyKYnXg{+~1ZWWAUvW$#9Y^%i*e?m*j}cc|$o!5*PjnDg_q=23YjvqFMddW@PB zOOEGe2T4+$b_rgW261@`UyN;UKu1gq)cP}^@=YC1HQY>&EoK-SluS_z6DYf_kXHV6 zV&U~C+2R~q)@C>bL5F5ysKXU@zf!UNzm0GSHR8O*HE`xQT5C5ogkB7%<;h{Bp>~E= z_8vs&iDfwCr~!}XYcO?|4XWmPB64UBi$61$73~vb7>mEB zxwFG(BQyRU%9`gSu{gc2P`@h4@@!l2#6<>=lity+5*6&Zmqd{pgNZ-Tp6(5`A++TW z)_Jk7A6O`R27cbj0gYKKr9~R7|_C_Y)+gfTX;s7iS6!&MbHrDKktNkUKl=3 zyNt&2a&+uCiKvTvv9*Z@_nCuvoBqp$D9>=KqY#beoTmuCI zj?Zs5=APAs_-;0q)!h?hjw)4PU2fQP$^)VH9Ak3-CH?&SgZ>DnQoh_KX1~vX<%e-I ztHkx;U!{UwuNCn!!Wu?ntgxTUR@s+-N3WO&!=9@sH`Qbkr3y^rQWLfmIWcE`Gq!iG zJ!^BXp@wx&=+FIdQdP)+glIY>bdzAZCK-?XI2~Y25cCGQ`B+ji*f&=gi|*&(`Qa+u zRl0}urm?8Ie-J}jYEZ#(R=3`ez570eRc5bXGX46jc_oi&jK-1YuLv?a6-fGqO-wxP z3G0i^W5F3glvC11V+UlAxZ*dn+QG+bOBUqayqpEcg}JEO!Yy)sC4|T8h0!K8Kt)^z zcAUi*j4>F3yL%D|xFpgQ<4pROF3A4&NHVopNtUH%fHMm;;kTKajW^=O7M$42X4d;L zh1@ILcYKrhElP!}fhRTsHf`8o^i|YL6wy6EV_Nj;qz8Naq;gwLjm|Z}}FQ zz3wy37q~~^sb?{;`~t$1V{x9-DRy-yVV;8ny7Z5d$CT65EHjHNly-A$K_U6{o~7G1 zO?39~b1EnuAs5@}pdZs=S^k19ZeGfp?74zBEz*Lgd@&l2kcFOQDTutQ&h!n}u%8w# zOgvhV-R5RF@q7KmHHlMbbv^+dnQUYPHlT0PZzSn<pLz z^cC5w#toR?5RBd(TwW%U)2D~|pkKs!r2<;`r#1!~y2O#$`->6{<#24WCZ@<$(tiOd zw2ot@GM(%>PQV6jAGcy|XZZgdop)T$@BhV%25l9kK}3Zp>is&`O<4_zD9Nnwp=5-} zj5IZ*Noj~mNkc{{LN>{0$V?F$B4rdQzSr;fM}OWu?nn2#>waD5b}qTc z%BTL5?~&}h4Dw6u!JXC`TwB(F>TzQDb@nNJY5GX7SE=#(EH$?L-~jfl{t65i6=LY% zS~&CB%xR+`?u6T5kG6XvUmnIxEW@g4rRb1UVrM>R zv3KLNSj2({@VTzU%pQ$l(RUIt!tE$(HYOwe=ve$U+JaSK$+%u62Ad@VP}0MhvJWDm zS<=pVAfuSvj#fIABMZl0nyB@N5Pg1_Ai8$;oT$d(8Cu6bsPN3R6av65u_`cGnyZkaRr#z`#rs3Cf9 ziC{cPfSh|U?0*L0%jRf^mT`~3ml4>bH3V^SJnOwD9}}L3!+^(62YJt6kN>S^uh|G3 z$}xuPP$k??KgOh`6PU}EaF#U0iH6!I(cUng#~CRuiW@B_+B*3MbDwgX-@zZi&!-$e z1ZoSKxu(E(^;m%4)fiCT;l>vI$*`QZDQ>1PjMz0EX$H_bDclLAJZ4@ zf>`WpHn8dylOEH)$AJxJ$E3t(t7v*lyf=b{zIQPQc(=Cg+pY z(FsQxG*5m@8PAlMVev?It5%IIzL|{ZHv^f?@FA?BJOVElxnP0YdW0<;fs+w;F)~b_ z?Yv$Hb+6ly9O4I;c`wO3s-9+i>>!n@E71M;82NnHH_T)J8?M-Z^eJ~SFXK7=H>H!( z?l;g<>G8DFZaz&MZcPocEhM%`4qBCi5i&Ij#iuSp_v9(8YO@s?G%XiRKIAN_igJMK z$?2H)P{g?q0-HG5h`ra==9+;A6dYopcWVoD^xQ$V-q4R+iH)5F7=8H+S}&Jiwu?Pc za<#Car9iM88;WC34c3j;bZw6;dK&ei*rkoO zjV72sYbpXbGuitx_u*~G!}8N9&^r(Vs~5f)%+I&}Y0-H1q!s(#H$l9w3C~szW?rWn zaLBoyKX*1LkdT7uIvwOD9Yt3k*Y@+5VZkm*+$#J;=O;X;*GFX8=(Ru5vFaUct7c;W z*W12H&BOazCy~x9N0H7OOOd|25vpS+a(l_W;DUaC9)v(@Y zj7kYZl>8cu`^!UN;Svw$4;e^5mx|3P2RQ3(3!WTGC&}bS+8lU=vN)grx7;}hUODLG zsK`i<+4z2DIb_vc(03NdJ^GXCCwg$+urue2twDeFbSTOfqVLolm`ZGChu?d!q8keA zO=K|@y{o0!J$Got4tu2T55%%xz2p*b6Qz%Gu|6h|=dQM3cYzg-H3I**CU8lq6rnn& zIDhaL{k}VhXC^f;zk4?9tK4z)FlW`j5n~aDB-#ApI;3bnrE}XdXy>qSaY z6TzeB%6|<=90_N44#18bPqGR$Ylx=)@=SqV^?W%--r}@xmld;8r;jdu~yh~Oc@J32crAPX14pe4@(}gfDNp@ zPfy;SCDn>ZDlQF0dg69$Hsg2jPM%5Rely#s4!GS-;#{`pOmD(I(WevA zc&x;~%g%BRZt6PB=$DS%fxDo(dp|zk@`9`94meBCghlc%q|Rwa-pA)qIO31kbDnsV zwGoHL1fxqa6ir9%aOP(iLWV`aiF=m4PA-GmgweRZY6RSViil}TXvi7$bBWRVsS`lDPV+0 zCtZJ2PZJHcAo;Zwwo2Kdeqtogc9#SE%+0McCDVS9bR=dGJvM13b&G21T3k*07y44U zdMxGcJxWM$!_B=bad7Q2+|bWD;(`_C*EG z$nhLUl>&~0c2IizeB_uc$4t)P_jsg+dL{?MH+rxcc$0oyeNHwv63JAxfbe6c|I~z;vkk0r9nfq(6LvF!d12y^KGi{UyE&_j?=x&hXF^*23CZmK$#*c7QByJpy*B8&?h1vc7)->>} z^Xko9w>m{Chp&>L`*17 zObo-#2|G|0%LS-6J!5fgz8CH;D2L9nQd}8#1pALAva36z*#d=IY^~TmTDtTm314#P(DENh8~zG2 z-YPP++&0uc`h+*d4{^U;4&S~=WAm5(ID2M1_YD9en#aJdvzoN(l4xSzIa;X2^XWf^ zBKeOF1}UjBEdB?Jj~XmI@-+SVbCzzDm6QBYCl-EeE=#YnWGk8u(_KS>hSo+=X1_u< zP^*+xY|Li^CYzwad?n_p?t_kU4EB%RfiK)^c<^y1Xav`g=0At?t%;CW&spIQRH18q z7+MuS-Js+ubogGbP_YB9l)dSi=gQKlRo+wqjuXc zm;|gs+Qqf-keUwo?t_zc0S3BXf^Gg{v=|v+n3)#Zqx>OnwGC>+_F`hCF2ZIn#jB0G z;hcMy{(L@8BhF-yhv8GMFI>l(mIl09Z-NE8t*~~>DAEqnd20)QvC+m$NylGDaQHx43|Z2Q0CstV{gBbn@tzJ z4dQo##0Io?7ols-V_bM^h?`#1;dWjXW}a2_cXc}rl(|o0rkinT{zk3`>)?@nI^SI< zU{gXUB2w<%?SA(H%O;FqjCbPmaj z_D+=)%}@W$PP|Fv8c__-)h6(shX9tTxR)L3@?@{9lMwst3_SOr#)v&tIOP$89Z9j+ z(PhLem(67@-=;9VjB#X9sX-fU{t0G%vh2>qUX1D|%j7p{Bj1qw!e2~-6+2Yf=Xbig_#5u1jxNRweNZE<#(#b-O%_$_-T*mRP4oFPbV&OHfG2LFA z&l_?~r=TAja;X6;w{xHOXc;ysYCBX_55mXrIE;_^<9YjCWNv%}93Kn!BNo^!?usV7 zMA5b$&JZ5Pc5$COABmaNmp<${l`ERIJy}#RC`mLlVlB?h^+We|M^wAs!|t7*vB>WS zg1-&GF_|x9?;8T!89Xa*t%(>%eLTn(XU~U9vNu*`n5=&o*~%$!)QN+uK^vtXtRdsw zVi;8-#qOxeG1VGH*4Y{g=c-e%$-9Q4!owKhe;Nwcxdx%mId-Y}_&0^~L9S@B(Wf<- z|B{cmar^|+^UY-~CrVg+(;#T=)JE5ZdK$^T@y>z%?8w}DoJ;u!oE2juPbsn4ud=Ye z#~-gM0x|ZS6dR{1%T5I;vpS^_D7Z5Qjq{aoM{F=lj{b#Jhfc%6d^bzky?~wAX~zo3 z9H5r18T8;@GEKq-X5Rk-iw`}@d=jsd*WpS^`xj4|mOq$>vbgB?^nWZx_Z*64b20bt zO>|8)gnjORh<<9zwMSdFv}Qhw*05vi2Wg^9oAdNDEYZRl$rFSL?9FY)LYHsEt#fVY!B&K11IuihFsHeJNYD@UNk z_gM|jm!Q1C7K1-|;dt~4nD{)v0gF2rt9BRTGF>sKZXLR+LgA<}gl!()fyrBn@MZmM zc478twr`^*)A_}BXfwia>2?hK2aBaOiWR?+Q z_^w#Q`HSxGHFv>o&LdFUGy~bU9kFYE6fRWpT}IR5^}a3eoF7T&HsteO zkxOLaJ{Q&1MsT$o0|lQ-(pmhB?3TW!OG^Wh-WiAioENeuXoTo*gPv%3yRoQ%@5Yk7 z;<0#w2&um|u;$e(*eAPrjKw_1-t0@zK2wD~!9`H}pp1ie=hMgPB1Ca-hmF^Lp3QnL zOz0UxH-)!RWJ?vInz<&cl#g$R?jZMjG){G>V%shoWPhImRfRFQlRXD=hHH_r zwhfO%)EFDy3pIYX87tmHs`aDrm(QM^8-LUCQF(Oj5cj5cB!MT7@Gw~t8M2&DrN}jm zk9>F8bqag~j73~q%l;&f^*4;MFRUBKbBD0zLlx-s2}S*WFLb@xgmmiz*cQDBzO($% zln}%g%LTEOPGPL-;xCN;@(-PBe&b=ZHYPrig`wUc)QlR6vr=*h3>$^sq#_)bYlHsT zQ#kH#gtap@vG>U^G#MxJ9A-MAwnU(|T#jXK8pR$+J;ENp=lm|!fG^Q`=&#n!dB>dL zuJjJOe!C#_X&`FtIfuB%JLo55caY<3!9f5 z<(#pXwAy#2Xnor}QO#i+QQkE_+#Vf+tq(#_$a6^@P9;2x&O2+uj*wSM6K&TE#l-r4 zn6#Z|;Kjzkuq~T2=S!jTXemzL;k<^WoB3JKnY^Y~X`f;f30w22Z>;&xZ^fAck}UB_^y082O5v;LP;7jC zl?t315I3+EExO!`_6#)a}*egkC=DJ^*U1<$Kka8IMG{f=jtqMw3 zFTjx3)8G&#f^DZWnpWFGd+rU+{Jn_Ytq1UIY#W(;?&d7nMpE~hPQ%`LlYy%*c{qyW zrHlk#-aJX$MsLFC?Pj>eGnCw~EsC9{EHdVuM`G+c6iu!{)%ynOJ$a@=J)W&y7tOMy zxQD1xnf-46g^P}FxzFh-%4YMfzS1WM>5yhA-}tUtF_VHlF3=hAG^)=aCZ9H*T?;T} z1;2_=-+K>dU**9z$rE?G)pKQ#2?T&;#Ee4DTezLu+yt>K5Pr_s0kn>(1*Cr)79#q-#w;ltQP-SI5h&4?|m`GW!HRN1)k27C|j1PPvxVf6PY z_J8_GD??`@aIzCb>Wh$ebTzhX`Qeh`S%^;E<~iJftnu_=MDuxD%(Mn+8ka<8uH6=u z%GHT-T%8hS`M$Xm%{Zo3*M;lTC%W%v22SvO}lT9r#SD zgO+v$az^*!k4+B-|Mw310oAnrLM!cDF&yTZ3z%Ec64q`tg}r(H7e}s0F~jk4tb0;3 z?oX&hqxyB|^Q>g!!MDi46U@HchgC94?Bfs(w##9lsKuv`an(h1&RH4Rhd(!v1?dpM{Oe#4Mf-OW&V~9ffZ=EAKa&-Njkmvow6`NkruN zP+Yqq%_5%w_1TL4Tob%@H8wp8P0y^@>z0~ zEIYGr7qZPbBf8xVs}y)I#TV`od%OUdGw))__e%7X)*#5Z9WTvaW2Ijrs`u|>-9nBngS5B1^hOkM!d~V|1eaiNa z;8OeyTi(>-*Tj$TTK!st8 zr|Di@8sY||V@ds1#BDYgNuA)#zQc~9+S)uU8Wx2O*1;&&78u-|*k_lSY-4XJ`VG&7 zJ)dtr%v*+;vO75w!XMvc9^mEhas)l%`7qaQbTu)Y?;=-{bfGVnZSh5@>RuQh?T>Ts z&bUuS^acS> z*6|svd>74%V_=5wBY=vxgLww8iDs3GBejWhz8=-0 zPW=vU)TeN^PAaO77GS>WDM&R=K%6-5Vk?{iHs6vBmf6ehZd=d({7u2S{9^trOY~M|oePIDv#v)-xl;pI{}0G7COi;1VC!uG3TAr9a3T!Uv`3+C2hZG$XdSZP`Cx>h*C&YoX5KJN!%;$ zkB(D^;Mf?2JcR==v*Vd-w?7#CK!O=;?LyzQ2C{T}Mz>sVQgr(v%xc|>!1_7(-f75w zp3!5L1E#Uh=DAd&oxquLarD&oF>~Bq%3iH0VlRK55JFt`3QmEbScpOsj%^U9o0X{3fbf*j99^M2Vg7My!Ll?{zG zW`(c#t}7`4ruV{;5}n0NLULJx%~iIyHHPHk;z|C-V`?6dhy1*Bj6HY(T>;M^Ew0Lj zj2p*Hes1C1s>|G0!o8BWN~qtUFQjSLOIi^Ru-EN0)IJ`^&Ku9M$mBk%2Ao4eo{Fa^4Z+BY|YD!5L33rPDp}WZN4 z6j|#(6!srV6k=|~3#ZDGXwR!S8kBvAmNhg9+fSVn9Ink5(xzS!6cQr@{W)I+TRlH| zd1@~WZkt7$-IkkrpQtd^J~++vdhJ+B@O~wz$)yX^)@&A@^vw}IB+nL#9;%Vci4i0l z_CgriwnA9GXpw3B(sGkC#&VRO#`9~R{t1VF)d*_~?+a7qo(VT%~g*A6K3aQ>Q!o}k8f>7Uo2 zyVc(Xot}8Xvt*x8-nUQi&|WR5cg_@sUpgoZPtPEwm6ymkt(Gd6?H8i_j|)YYs)V45 z%Ys-$fN(JOfZ#jMOz59CSaA9N&{Vean4mX0QwaN%Ak5AW6XyOqC`^$E6$Yk=)1-}! zg7fNB;gSyTGUziR&*mPXj%-Q)=0xf-?GvQJwvc$DC-rwQCx`Jh!iGobg1V=saMk~e zpw!qZ*q!bcZ05%bUtb&(2CUs9cs$pk?zKj=?(kGf7Mn`GmV>CI_q`xLG7EEMkZG19w2eO~JnoAW`kk*4hPRZP{xp&mJVRoIP0zA~#+60Fo01}-YvgR6 zH5^LmI}?O?7Zj+zLz=ASGzbTpBZam}K}^1agzPX%-4{voZ>tF9 zEfPZfHWSk*4;9jyc2sZ8$%}-kD@@$Tf)wTl|rKRHQ~heEWy3~qws6UZ^5lx zj%H8s6Xujz3%4$+3)9YxA>6DNT2m?n*IipkCsvt$#97g%9x-9l32n#Q1-F%h=;|49QkL!(zL?Dt z3T~MRPmHC6uzo*&T~%@+|S{n`cF@+?8Sev@z~4u+Rcz3DbBdseOG#DdztacQN#rsVAeg+An?qv#<=zSpXFL+r#0GLEi6S%18qI81Ens;ivIyL3jhW4kSh#m7d?tAC zUE)q$4%&_0X(v$`{Dt?OOo4tCCn@T!!SOjy>9*4kS{*2Z=VAgD?RG|Mw>xwfE5Oxc zEJAZ!`bNVt2 zJs5~f<|TM5QA$-(jkM=%I~nDTVDJ2_SjjaDHa%ezlx(u0rh6V{Z=`W>63?a>kHZ%H=rfI*6Bz4v4;d+6knTN=9j#rBZUn);B?*?5-1B+sG+u2@L;GUBXFRkG z`rD7!<9sRkty^mHSRhTLUqeygmGTN7wK58g_Y2a$#+TS$vKQ1dL4m=4VaMpj{1v# zrLA1YpKcJztZc8d^vrx#Z26O9CJcw;(8*A_z&#Hetq^j4H-5_83dJi9OjCu+^1US@&v`TkhO^FT9V z{gK7*ReSlp(09vD^hTY)bd55+oM(Xkf0p2H+#Z~E>ZT*MTHN2L3D>s&aMU>tL+@Ik zhtKhqrK;@ymR8u_SB6BQHhvCs;(k0IB=FtmDCbP<9^nte8HbR2@fzZjr180LEPUUu z!$FH96dztd(=y*vD%bQ6nYUuKjw)MG7k~|GaWTo5P((D#BI*2WjD5NiXCsW@+hq%Pa|aj<4h)Q3p?=gAq5{ z9u_J#ya&PyjSr-7+E)XuRpTLXaw4`x>2Mae8%S&_BoD7Z{h6@{z9or#Ni}57`bF~^ zo$7wnQ99}5Q;NG6qc&NhprMd&*|6ve*Zcj%>@I{myEk?eF3Z`{T z0zI1uWrZ!G6IYx>D|UN`X5ZAoBL14#>o4d{L@_p|WI#uu1gk{%Y2}XZ^!#Qbbl&IF zv>)eah({S|=|^KpMIxkK!!gn%l4s0$=UVp)jMuqKvrnebqmFp09<9QPY9v^}@n(Gd zz5>I|=V9-{xp1Dm9R-|?U?i7C1Ak1zvwdT6ke~Ta0?t4qoHKttE#TX8g01}I#}=)~ z=Gnm0^yES_m1gFV(-<|L&-sJxPbaYL56oEh=21*$?=KiF;%6-96Wy9(1F3{9Na4HR zyziMvagoPd>+iIuv4k2ItcTh%H++j&0=t1?Z2gG}ELi*&8+BxGK;jo|?Cc>yX(tIZzk;PcDkg)GZ^*NIB|c=7v9g8dSf^edn=w2eL)-JA>|cn-BV}RYCW)U;*C~8z z9Zqcfj@WKVW>&6_AD_kWx?dT&>hQq4rV?vRkYNiOH8AW1XU_WmBZVb%*rWF*bb#Y*zU$d}0xBnyF!5V7Ophku&kf%DY~q1m3NbW$?=k8~ z^{2%Kzwq(OJ-nY?h_DU=-2SHu#JsHnGK;TQ?_)y@l- zI_CfZGB`Qs9Neo9z`y@aOc)%4Z5?~yrEd;{)f+LhZ!$dhNpmiq1NkcZ)8?cV=5DbD9ypv6gAs_6fUo@IX`v5l(u^jr+fSM`U^#ue17x`5h;H43FO58#+-DCBn};(O2pOb+d( zuCRU>6XLAVIQcVjxz=}55S4= z3-sCdEdNZ`W833{m~o1~SDgmZ6+NMNUv*Sp5}+9&flY0#w2AA)LFGz}bxSba-<+kU zX3D!qhM~OtBJB#|%$oWZnmcs`1$o(0-@bKJH?}`!9UO*F-aRyx^ON=reopuPJtdE~ zd+yU}%1G15^z5r*p9*6$PV_4)EfF0BKL-+4CC>kn5S8WJX_YFkz zU|kfy)aTrtF^Ky5irDXJ+E^%oq6-<=n{XJX)w%YcvJo2lf-%oG5Mk0%qLTIjA`Q;3 zj_|3Wl8joi(Yi=U$3sx!z%y^3|nh3W>aYF9R=sXs;%CyAl**j22KO+j?V3yP?HM4IQW)7$+HC`#gU z<_6xA8)%OE_bid+Zh%AK>rk;Y6fZ4&@U29R<*!p@BI`-aNB2J(+9RR~drWAhxi`LC zUyo}Fet2}pn=>hWApU(F#*|J%Y!a4WOc!o#o0Kp_~>~|H!hbZZ2V5PlO=KV zelVL*x}KH1{*V3NzC>o}Y}oPp!^-+t68rEUtsQ4hbA7AfIpiwj)ADd}dOqDY3@0(( zC7UtrCmK%HAph8Fywov7fU*kS&XY&mnvv{Y`*5~%zA-y+Lx#j+MD)f^ogNohLvw{b zN+q@7Q@H_MoZ;%%;{)CJ8e%VdX~&NK*gs$feZ6W<(b)zRl)VfIJEkM_zi~Lmv)#w; zX+Sex7W*W5-<_)yzAv=K^WKqICqDu!rVhc@doA?zq9SB}*HNuU4)&I3a5jDtN}XmP zAYl^DMGnH@%_q^@l#ANXUD!0W6$=)%VQj=>%x4Smns<~gx0Z!&VK`XVZYkCD^*B;Efb~I5n>k`lK-GtTMtN7@_ySVixq193grPBY% z(4mD4FHV5UlJS_XpoPhbgOPBZ2~jh0g;}+?h5nY|*tBLDD$o01-6klSrp`bnOyqCP~hHMbkk}rUQ}7YV5BJ;r}&Tp|E_=aCYEAd(n$4964|@l zro(p9Q0p6k2e$^H;(945UCW{xg(N!q^#d&zeWJp9Vo3JnXUOoIRFafRj-n-)@L(~n zCfdQw1Lg`dn z2zA%_Q~AKtR3dwfOjr`#*)EHFI|pE+ycAw0?V`IK`Sg8B0hz=~AvfT0p$M_?0{W|EYNulXyf7tVLR^Q_;OD%lHa-8c}+4dYNuq~$8o-CTI zq>cH@Re07}6C-l`sQK&}QXbJtAx06@vH380Yi3i|Rz0K*(!%wB8c=)AdrRpDMect= zTCJ8;=A%N2QKRXU^jXN;Wno>vY<^yi;?KAwmr(dKY-)n*iCi-nD~+EkVu4M;L}L78mlo+Uo~b^I+BfUilD74!)QX)1X}ta7%3b5xX(Kr!<@n~m20ni z^tSVUumrlUf0?dd$fE?2Ieui=@s5~j7;~hI{Qi{Eqhk^D{`w*aO(t-TpNw|jSPZE3 z=058K*f(eZZJn)7;iIL<^~pH)XWa;Px>1*H>)K64EtwP(DZ#s$=dwO2CssGxiapsd z1KQsfLX(UU|G}BmU+<@e?sfF^S{UCeA4EuY3Pvf#(~_-c=*G|p8WDQ~A+K-aa^nrG zefWtSkAETaQ}XDEX+gXF2Yg)Kk0q2@Q^}#7JjXhN^!#KH&ZJ-z(?ZpX|LFIBir6z+ z2Mv$jlJ)Z8sOP@uh4NaI?=U}_3xe*8trTYKr@i6+u=afJTL8OWLJhCZuwI^6V}UTQw2%G@n>bT%|)eDmiLttXP7x%~U{+j(=^lV2D&2^}z){||t{_Ja# zw-eBh&n|Oam*N)h`=v=?H22#9^5-0WnPv${q~D-;zh~t2_&rrxR+GtFSwubM=eO$@ z+NZgj#+~w{9E0IB#X^r}T~(!JcDy_NhZ=dT&?Y5EeHy*=Eg5o+==W>R$hz#ug7{BB zpAQ?tnXz%a18O?=b8O{zB_FXROzZGuHi`RZPAzNV-Qy2Y_@^CK!J()*b`hPH(^w&& zk4+cpv*X6Ug`MT&=*?$Es#LDVvA4CXv3mj{awK9u13k@q{7O5xL+U-^L>dY@Y*eTEr7gUAPV@#gd6_Y$`)q*=#n2{!5x?|Hnj4d1q{!>)yoF?B=-MDM?%LCO;+r2P@*zZYq4eb`a2 z!L~JOGso1`Od^NxXDq7l_bEEc^Muaj^#6c3_5uSj+y^QGIZ;tj;sm_ zX|rNow`ADS*H2KEy9FU@yz$auDJ0A0!(MM1zMYy4mwx?NBIkpynfwEN-CR4m#(O*d zNHg!)U^czQpN;WPVj4zV0~&LizHT^3uje*H-s=svZ%|_it&NxvR1VYAWtg&ZF1(Gs zAdxl~vd&$Qeb0G0dGf4E%pSYdr^9Z_APkgrW8Dv=+2rK`7#MMo&0VyG1*j8i-?WKF zEDWIYOODbBmlk&VMLoM1c8Xo`bRhi-aisGe_r>8FqVbNyL^Hz`MdRD@@H*%euJg0_ zS9dC$MX`9C=Y;8#beZ;7WA?Lp9BZ;^A?$0VUuXZ)&`YYURab?TH)^tz6YR0{hZCl4 zcEg(DQnuwx4(qGWWGf@H=(ScdDTHT{2EQAmy+4Qt!T~Jq9>e0rN3p=x2`uE^dhFud zgBz=&AeYPY0f|O9&3Qpal7aZObQ@eBM`30jXE49B!Q53#pk!Et!&Ho!mN7`(c@I`v zcX0J6?@@Tdf3H;pM*L2~uv-fdH+%*B+74j*?)99>=7pXf2dYku!_H>I5x$L zsafeTYvb>zk@v;9+15BDu?)Vv3!r3I4jhf5G5c;U;vQBbBC8e;&IRI@Rs!0$N8m`2 z0*W%5Y31X~lyganb;OQf(@W*q)(0mb?O6h^>O3TU4?_R%oE3bD=eQ><#?&>h@bK~v ztf~`Xsiiq=($rB{l}k56D(L?9x76Rh2vrG1*r#w7Urc0K?Cw4^mldI3+i@D4a)=y8 z`B89~1(KVrG4{Cxu0Lp?*wGJY#D@nwE9?sO6yS!396lUaBT~QTCX%gPEPCf)0+SRy zDEWzD;pc7y{*z#+a3ZbTDOy7+hyRr+@mDbTr}_$#GUqfXTpyh^1%pJK?KJuy#QA6ce%@(3y;O|a%;Pwf*NUGH8u6^G1kWYTqUT>R z;^esZCpi^o`5CJFkaxHob%)QIKy2vUi$VLI(26IwseerkEePPg#)Ivc)y=yZ&dh*) z>LlpC;rXA#wxaVgb40eH8KS{EB`~r~0dI1D)1#d$@z8k}_Vh;LZcjBua}G|?NmEpW zyeH*9-)PH*9y*`o4`0p?Fyi_dd&|3eA8taD>Uv1%Z6S5f&GaH$laAymvlKIB)^V{L z^9Brq&IujNaF#~cpYL$l)``inW!V1K0#bK2@jeX$-b28B+guy!9~Fh+o732_#UX4n z{MfbHRN7+{Lq0x>sYOhgIehPf#QDFl+cJ|Kcqy>OrW2Wcup`e&xTAKVIo5K{i0;et zoKtcJVv%aN`a~Lsy`Pbbx;5_2TMnncGoY@g$aDkxGZpbxgl6p|NBKbN(A!O4=Y+72 zexXb&bO&?0C`&(kgXz-T_w>{?p9Q@uVILM1GUpH1Aa=JBbI;e|)*wgvFja<3WPF4T z)@n>?m<~U)6u4hcA0s&vd_}Vs-d>x^S|>ZOI}_%x`LjfjlN}H7OY->eWDVQ$ZZ`AE zn8*Ug>yyp&#q^TzFy}6I#H?d;v7^ccEeGOIqIUw-UEDW5N0Bo|#zH5EK_b2%9gz4d z=K*rZh|#e)eBeJMu9GFSiRj8j;ox`6WgH&IkzJ*9`P!&uuqOpyD4)28!j65Bu$8%NN$ zo&iYxJ`$3$dXWCJ6}4H3Xb9Shwne5m`$!kPYIQW(as<{t7eVWuA~JJ1n@+8ncBGWk zgYCN1e%XP3ebA$&zCU0;=`lXP<6T&tSJ>Z2H7wfbF>8pIgD^&dcQh(N;m%mL`HLni zT-XH}of23TXX3Xl_v7a+z^>_bu%UT~bjiYvlM#@S@!|R3W%OQUH&xCGr%`_bk?igV z6Qlo7w`>?j4<7~J5*_?n_mcXg%jxg+a@sZCj6}7jWKeNcSdv>oCU!6Abyy4ezGkF( z;W(wXpQNL0r%CI}1(N)pN+I!SP`|bd8{_}O&U-<03$E1SG=~;n3`fthAndsB0V&-e zdNkl94QxJ5abih$usIcB&*D*Rr^tK2ltH}$>=vlt@00>oBo|N#^y=mkh)+j z$~CTH`R8Ih@5zJMj}%z$OooD8AU^gyN9ng~nAUy`!JGw}UU?Jh^LV%F1ryZmF~hof zda&mEmABtS@aDYOD(?)cR*s-NuNaC>yMg4j&(Zwp0<_0Jqny-gs^C4M|7NWgCH|N% zQhGgEH2lUgj5eBwKI8G|zm|LImYl@U{}S=vkTay`a+HE>D`~s%op%a;rMR=LCb?QH@s_>ZWsk(}_HIjuoNAaT1P8&%)PO zo?q&ULFg_YWN{ri!yp10QH!vl!4FLqp`5Sai(3bGFv~U^E$@8oN0 z6$R1bx-eEcr_IpW!@wN ztl0whU$)qOcoH0{*P`Py&tzc?!uV|7{XdG%I}ppa4Z|`r$_SYyN@*e0d!6@7=vP!K zmCTZcC{dcy&L)HkMMgwsnMFh;Nkl_CN>fQmi!^-q_kVxn^?IKBzOM5;4&H~b=K1ra zLVa|7o`Rrxysxt2Ax1ddK~+Z~&L2LEtJ4B;%zg_*j|fp&$ldg10@Q!Kg5A9rATM6Q zJ-{(A;hwO2-%s!iWiCE_k4N&M16XZX4vWfM99(h)IxTA0C8h^U8+C{^=29l-QYpz5 z(WtLK@Yc8NjxM=)(sP&X zq~0=*l(w%WAD-#_=yib4)efOlej7&s@*MvoDdBw;DWTs13E`~(8z{$3gZ??*Nl+io zyynX>!Ji>)oP8|MTX1b3?@jHQ1lt)FSm801XNCr0R<|scULAp}ldY*^*;-Q2N})6v{=4-MXLbhV z5OW!eR#*Nk88ivMbr0f`yF5xGWH4n;Jry=~Qc0*7G&BbUrH|hW;#Y`}u74Ds1p6WL zXfm88^W0qWCvwVZqEGer5g2kGl-spl0X- zSfwxwx-uLMcayjuB^&_<{jvU$81y}4AbwPy_gh3*``IOX?x=Y!Tds zO%rD7YYMl`*AvD)+l<#OQIKl4#%-QEp4gku@61K$uIJpu_OG-uYa6Kv!N~j(k2ki7klG!O`NNLl?X}kwG4>~=>wG0~g&;_k_~Q456Bs<{GmU8- z4D*w>$v#P(ZMDe4jCb7YEvCiZ?0tmam(D?FbR7NQ-<^F9*T}OWlIf2MW)BW&vQ2WJ zd|h?=pkj;-icguskVow9k1BT8BNY_~IcrMd0B3Gb!ZKHLe2|^ZeUT6Gw5tRY!!E(Q zwUidV$|Du)7bIy=jKfI7oYb>;wsaISazr?HI+4;(>|~3KXS0tTn^~OJ4$9{Di6<-5 z>4D*13_ma%Uzb~;zxE{#wHBcuy2lozL>h;C3#Q!NNRjm654eQd%2&?gNk6u9oE-v)wb$)~`ItWF2u z%C2bi{<1;w`)W+HE5(hgDLezJz`ZX^DA(ST;;tPb2gfWjew{*Rho+F*-6K>u;xZL< znRC925aufPNnoX zYF=bv?+AT7e7hGR*P2jd_zyNbgB!bFn{F;%K`}=b)6L>*Sl`2QG=fOH_CE~y^$Cc) zZw<%oBk=u`JH+f(;C_BDEq+x-YkMw}%6kch2(~N=Yg)8WZX_Kxx80{Q3}xC#ecp)NTZWY|cT} zlwf`*%b_537*D3}gwP6@86Am>XZT%z?RJ!Uy0bTy>sWfPCsQ##4U5h)luj+cA-%cy z^m80{5XggBw9>?CK4YBqm@b&d!}-!>Nbv0byLbcCH*xmHBYix5l?>TSHBerc21z!A zRkaObzF%|jGVV9#Et6*b_O}tWx((UOKcM(b4<5|(h4~#%m@W|_E^{?12m8Qk<}uVS zBiJ4Bgu!Y%hE`Y~F9~}Kp zh(W#CxX_pkuj%(DV(jDdT<(7AZENNzESX_&qGQco&j= z%u(R-m8PzINQ00>BVRD4#~ptUuC!x^mLFn7_oC{VGk@Ol zgiS%!$!Eyd{S3ERQ3y9pWvYA@8u}rS`Fe$r!m4^Y;u=cUHn;KkXE~bXQ}MH_38Tti zV?}u-}oQ$#x6#(Trf0_B*DSt3e;+NrsS<1nge?1<6Aiv6gr!NybJU}^H#$n=|otSj)DVwL#&X%Y?V#S`B z7<~0OZjC8Le4rslsr?5N?rzy!)(&5{ZmfKv$dVI8IpdEztS<5$QMCvY-7<`s>>9&z zY&N6m(<%(@9FHTHqFJVw3tM9`jVtNTIo7GzRl(It4soZ{Wl{g_g$i<>Dx{_`a`R}%z-jFY$_ z>yCiRP-q-lj?Umsn6YRkI>m>f@U}R1cXVR9{txVpZpE0IvG`c)iD5^AF)x*W@5;(h zypx|%+br2g`@t+M|1?Tx@_W(aVAQDSqolA1^`eoeSNaeC;=^H}a15vjgcUSiQ>F*a3|_e&m3!A6^@*y(49o0C%T%jFPy^uqYLqysLN zH?g&{4eM2VX=u3~A|;38jsj=yN&F(Cvz@e>pFO26x}v^K8U9-@K_uf6azz~QJ1dNY z{5*GgWEgFqyoLgEkJ0ySxnz7Qj+AQll5_7S8tpq3y9Z6g9OtiO9wHRV*y#$b)s=+N z9~aY`1&%aiP9T+d4#D_G8nFH|0){35Bp-5;bZ3;(&{t8kO4Ns(L~qdHrg9ptb&Wzq zn`r(=S%Kd$J;9wcb%Ex>EHWRTPgO?u>5HQ|y>xM*E0?&V<->4#q)Rlz#+-ayZVK{N zeHBP8H=*ysL-fYAfriY#K&I!83H-K9Yr&S`On%t(Aun1TW|Qy9PXqq4IRv_UPAR*NZ7bJRE*V)s|@<5r`f zN@EaBY*VBka~4qN9a|Ffa-zuxDg`U!q$zHZ7EN&9DnlY|rJ1*C8{Q6tYi2=9=~ zhyrTqucwCJwRB-eHeHUqNFfq8X!4M$6mjZ5TKsVa+3f8Xj66Gw)csbF*7~ag&v|k* zF;<>NfAuE`%`7TdSV3!NmD9qB<#cFf6m=gP$rN>T7<>(x#v*N6czGJ#=v+zd(krPX z&Y#Xqj-{tVD+F(s+!naVhmpcxF|rOBO{uPWWNdPVWD|18olcObsud*`&L`vP*7SXJ zj$r4qTY{f=+5}^Z{?bpYZ`68O3I0?3>7{NsEq3IYp{e$y7P*sp7cQfi`DtVybdjF4 zB~f;20j)`@C-=&e?JtxTq1ulO?D?B0)9@$LA0|F}S>zEyB>bFScu$U1Z^j>Xy` z<(MocOMTPa>A=iTdaFBz{^-pm%|QpqF1VO~$8M8??0Z@rb4w6U4+Nd96$0bzXmS#3 zrr@OKbT3VbWDOO_&SnOE_~VF`QTw6l6^+0r8wG2poe@Zez7QOHBSTl?mQrNU4qB&w zgDis!X-Y^F8P3k2ine29Ul2|80T%_ghgS;*c{~v~GzQa^AF*^h>>RnqkE5grGg|28 zNo_;@NRLI6XygV`vd*OUPXegYKAI%XEhhJkHl*<~kbDL_=&6P?eQ9=}3q6kovfkt9 z?nK62Ql+GSFo85R^GV^j9DMC1u`Ea#2MtVVZrfZ!^-}saE1aUA9ibI6r6hUIk=|5n zqUC#{=~VL%!Aom>>iEx^X5Kn3Fs%<0O!W*D^j~h|zU5|WpZ$uS`2^80BR8t@38nat zzXi94iqR`$C2H*BeWd#h

~|49(`yMP@?B7FqK-#4UljgeJ*e97eJY^GNrNEh$a$ zAmbND$@XI@{pH-v1BWw`d&8`qHN^9 z4DJ<;LdMiIycl%_0fw(opLZAAC8KG2SPNxew5GpGKk?`1KdfsJWkaI{Sh&#+le2}G z*SU(bqh0Z=!yR>-BzPB50hziZb+-iLnZ4&eWdggtI?4V8!{xu(&0TsprDEQcA zh>c$^Wbxx>aOa3W>bGo1bzUrHKTm_x{bGcP%tGlIJ^Wg)!8@}V7(O@|UpDW8slOfP zbqbi=O=GrlQ6rtx?x4Fv#nI|=hJ~z4Wh28P*q(~Pg71c+G&TGna&;ovGRb!KH~BNu zf6+%y`k8oMSdTY*4kGq^IEL8!!pEcxt1UFxp~E_CqL?VKMTqUmUwFPwnswRpj_-{z ztY`d8D8`QA4wunT9(0-2FDhhfnhMy9-T?}JG8mWi#Nox=GvYk|l$VnX^TIxOFPCEt zE!r$pQUn`*A3?*Hn{aoEqC+{ERK>|t;d(>Z%A+qZIH(3QTRqX1ehA(b0enxB0lSsa zpm}TYYex-P=5Is8@JqbwkPNfu`>^+N08Xe!fJe56;vSgC`f}{E@^F~7#UMbw5$4$~WaNK^x+Ff)6vcV8 zLco1f&f-i~(hzbo&e&xk#0wD#XclzQxXt(J1OFaU!g3yW|}apnM&nrXo5;BclpIa`cNfq+9yIOu!7Y%iV8c^l!Z-m zAJX3$JjayMNsd|)tpBJAtKK+op`@8A8oW5`SEb(EVc> zQr#!fr%A&|yiS}(rOso!HZr#6jWj!OFP;)Ow{7eGOp0+_jL$uG7`N60&su|E9lrpJ zj*rB+&T@L^@|a$#b&^(Y7llc`qH%$u__{5T{!L9HrFXd`_dEkLE!*&vJHftQ2|~c) z74Tj-3Hy_*5Vn6I!i-12O1hT*JdL8~EupkjWG621bM)acd=@=j0hKp35hDMZ)VS;2 zM(QSQtS+GG_Xji4E(7-BtB^@F^wW(+gK&S228OZcboXBwS#{=8?AHA_6S5XV4sjoL zrW8JJ7z;LA3Z)Ynw%#;HZUvM(cyMNxkc-QpHo%}pSYINg$ws+d<*~E<=NcPoXNI=^L+n&m;lu_ zD?GRCqiM!`{>FKaPwsr+=bQDo_h%V4TL~~*N)I0gYr|ea6?Sfz-srLS+0H)iS;w9Z1Dp19EV^TC@$S`B zxV^iJ`OhB0tN0l*?Do+Mjl(p@J&>Bc{E;@&jx*ggk(Qbc=kt*$nz#mHhtufl&tNj# z6;3{dek|q6einCQ1A7?aPM6EVDLBLe8_J8A;k66woqZ;2vHJ|^kU>l~+9;rY9gN zx_~x6=9#a{TWHk0$t*EmlT|Nyg~8Lgt1vhLdmmRq%auPrr$6GZttv=Zn__a;I=tZw zN&;f~=CK&vf&x8++foSFu=xxiw#-Pvee^)~G zJGHUwoF=pv9>xiiO3uqTgv*=c@wU2)R`btm#i^O-`o0J0?sNxkH4>wGU=Rd5Q?j8IIxJJ*3!U4rMJfSTA1&wdJzd7Cr(UfvUKl?Sw=PccguH zhFz6B979InVS^4Lw;7{khc(Z1@4~KZZ~S`4{gZ-~yyufg^EPlF=+29jxPmj?xkG7; zWD5RjY^GUJ_sMkmO|lg5XYrI|B^G?I4>P+g5wv3|Ow;zGa_S0U zu8gO!;<=x&VuC8(WEw!tcr2!EN`d#cB+T#U`TE$ubSm#R-97GxTn{;Xcp!&4oI7B# zAppwfkK^HqT)c}~kN3y6;ArFyT+s|8vnAP7aVwM510|S8*ie?bScWYh$>;2Qcj5Fd zfBfdTc*$?P&(F`-=k~A0`PBUoWCUQ2dj>RKpGC#)Yq;?5B)k4EoAn?Ne{6{A=kTXu=INVMh!G@Yhh_4@k^~wflcQnG@?{%DmcL)B< z?!v6Q6>Xd=a&h1P9pkHERI(jiqxWM$$~)}&(TnM2(#*3e20{JNh*IVMhP(5=i0K#A)fz>+V{xdqS`l_bfuU26Mj%Wch=9!c82fD z<(U3*3l>fZ!LLDKIH!IX`LBi`>cTLjPLsvzT{kHGd=+sh7cKpEoE8R0(`dy|nwV9A z%_bR8Ov-}Q6EWd7A90~h+7GsFL>{Dc&Lg)v4jX!$*xA!w%*V%xb;KAlDGM_e;%36y z>jp8OH*fGV(+hfr@@#z17}n8a&OW+7;P>&DG*e{|MtPdEr<;lK8YcU(R|PHKyNCMPMqUs&OcU)?CYl;@Xp#g#73A1i33+y9 z+Z{N)?M39oKj=_2`V~G)eSz%tyEJj>7+lm`hT6CPaIkR@ zORcO!_0&ChKa6uMoA#k2{tB%eVt}#TThUt-!Uk6rvyHp!*eTxgJhaXq8-6ZAMgez@ zE%C$H`zJVmWf~;S2`4TrMD6}CjC4PNX(#<)nX>?)fm?9JDFo6{#HMYV#>`5lu)*s3 z?DVPGY_~m_P}ZC&IIdxNGnTVWS%X-loVaK~AxVU)> zj5Nj##kL5F6eB=5C7ellb&Oh525RWtXZ4vjqW- zXj)p2nS0{!Wr71UxA$YKSMZ-RT|`*z`kOUf9V$GUJQ=_DaNlFobWALEXZP(_GilWj z_O5mp`?}wcX@1_v=i7$2eK;EY^@@DKFL8 z&I3HpyEc_&rN*<)PaaIx?>f~5pC{-3LJB?jk~K`|VMcLnEL^dQwr#7YOE(?ie&Z)o zJFF?ROBpX@1Dw~=#@S77FL0$y45y-JVEnrv{N-NM;k@(s?mu7VledU7kLKZd^cHxA zY+<$Phu8(B2=?b=6lyYJVf7%EXEN`x%yqS_yZ!{*|8*i>{5Ko3>KH6%*Wy>-ee_Db zfKv7={!VX4%=DxDZdZupJkG(K&s}lL+$qPXh)x_YCO=6__I|s7nW_E91`2Y}{iFi! z!FdScjQ_J%S9vDuI>NW);=##$)XvX_L|YK@GyPEAvJ^(sD`4q*2Aew0LuSn-NZ$K{ zw(ajBpCit8&@iT2Scb_1hgjhhe>Q33Y*xNTg~^Q2U^YgY>_e82&DmwjZjTwuqNksP zTWcW<6N_Q6E*CzZd6v}qD{@Y*q9-n)+~qTzwqBjc>O00TnO~DxPg&aoQwYg|Cawm@thWX z=H6bL6mj-dRfP=+vN)ggPb5;CtTQbfBgTHdmt+I& z-_S8gT3E1tk}x=Oif}@|9~>TU!`+!1aItwQ3ohHlE;PEZeA(B?>FK~qv#&^+>k69-;L-7ZpSx&WvtSc!VnD+oEH~kiTv*UA^sj( zKQ4h*w-F>pPQK020#)*NTw;xlmdaU4!R$>sd;)tGIy0|^V} zqho^&3Knigdb}-nbFAlg&y9%r3S6E!3tE4yPA*e;&!QPk*^nYhyHy1O;Hl#T8JE;yu;HN$_Twh*4XBHwy$< zaNC6Q^3yS|z8|5h#93f<2FB+GBVINdue|((Ug3v?bG0Ic)8miBwUEyZ&hkD$bUmgd zK0&;56KW#+sce%B4o)z|&8^_PUGDPEb%M`C5vH5cjC-GQQ1G$?O9mR@>39RPij1(J zP743UHB;s%Wmfe;hKV2Jy#VEO6zB1McjH#1@H{}w_#kY3u?ct5b|IqC6V0RII15&e zO-O%%<|nmS`LB^Z4Xa}(7G7cAX`1N1uZ+m=BAD5&$^v6lnV*Ud8}m$;xq1-Weql0G zFbT(B>0@}bCk&wmK~Nu>1gFcfu(>)BKl4=Jt09fMVJGm2_s7LOqETisoJD9Uu=ZWr zoOh{;$D>qmbGt@<5xO;W7>`5>mI;)O2b z^-wgl#_D@VS={o&tT*c@+r=|Q{plNV-f$t-Ja5F;GnXK^avD?ScA)P3)`K&j;= z7I!>{Q_oxc>3Bo8U;ZU4$86fX_Y1BDHREwaH4esDqBv*+9QWEo7CBhVJJPAPdvI`| z7S|7-$LVKTSgFl9v8#qa{lqZ5lSsn-tF<^bJ{2pn7ju^d?>F%-d+yjd@Y~Ow?`P(q zWyvU(RXT?KH&lmxdmV{!D?@Oee6f7%bBN{Mz$T?S&I!qd^!_XGT~>5zGS0VeHl z5y(4Ae}8fBTRYE)?&zY2NhOp}TSv3_-p9)%4|l9iqjE0ql@v4}=v@em<#yovKp*w) z{zaAItt2;I98W_;QTw8f42~z#$&6eING~8Y^5+?j!&vYr2)Enhgmq@ag_lmL3WLrE zbC>B+n5;R9y*n8z7O`O?KF?;TszRh^A?=-F|r-2 zb7qlr|1y%ZwWUptYE0Hxl|7hDtaGaw?|+(L@@st@e8{_qrZ=HgTZcVYZ7|qvJH~}a zAY1(@T8FmdUv>iyO|wJw!7#Yk1fgnHCMwGduyJNCR;|^>nyGU!-_HQyB!?NTLs0QS z1^Y7j|3jw&-^{M#(coG9T|N$z?oWr5`w#47e^6I*8v}p-!>B=K$h2RJ-+FG?r7;=e z%6zxGdIytE4rVgmkxZbl1vy30aPJJqO`hLY-p8G}I~QWs6o&h4{2cypIlLzvg~W57 ztur_TS1*ovu~Y|_KI5uL9VYClhoa97sNH*y0{L%HT-XZnfziynOO3r)rNeG&58&)+ z2{!tZ9P91##Mq$q80r?u9Vt;*-I9xkTRH#k*IcANn}?9*#Te)t%sPDJ*n5e=tY_1G zjOo0G;b=tawP4|%zIfr8&WpmdpXw0xoPzJX^BtMgiD%jk*!24@Y;^b?F~u8ZYOhhR z{)hVcj$W1fuPVNavBJIo(5?FoZujc&LQns9gf)Tuor>HJdpp-9XsUm za8S7zpWiOV-khTQ9XLG?&CyvirkQHUPBfMpG`VgBYjnR&u>5}kRSoL48(>?JAO znRtpB#3pf1p)}t7J%mGN>S4D1EA!p?hn;l!%PK2!G3-eOR9BqA(6c-@m#KuXEtY6k ze*gtVbtb09ovt~$xV%Oct9%vUQ2ZM&ZKavH#Yh(F=mgB?dF0oeWe|IuC8ZUzB5p2 zdf_mb#eI`v8y;!0vQ6c*oA+_=EgXsI{b!ILmyW~-DV)n40@b;0c>f>w6iOMQ^QIgA zEZ>2{c6>K*b~qbUC(9O(-iM>Hp>Tc@2#GZhvE*|(c#aikbLO*^{Cu8b#CrtmPh!su z&QLKr0nfO*xcL4HXPCc&VQe72O^<{0iX+?+aTGs1xa(_ZEXJQ-M6DYf=~2^c>M8z& zib)d8e$!y~q0t7@m&``aa9gatnv7dVq9A-6g_Kq25%;zO^Dhr!L5`A`tR;$B3uYCkRIj7zii1dEl>dD6R(Xz~mFr=$H_X z?X@X*^Jf~`XD)!)6<^qwM?%_mBYs>s#xu(XaJW1JwcPO)w&|fj;CY<7yf<~)OgKDQJ_Cg-8zBd zt=vWb{v_=i$RM*fGO!CDf`vT-&Wds+@jW|9JM1tWF4`{8j`9}_PIDKuG!Dh=WNAE` zCBQ)+CD@$nC8N|#+ErDCRP!cOs@376!A`hDad*+CA_Ve0NLhdg#w`*-W%y6H{`rb0 zX`-xW#ecZ&ybA2y8if9NOdIUGN$amPdRwyvE-yu>xpE}kty9LaJVV?)#WR*Rg*4gW zF5Tegyx^2z{%$#j*Teb!Ml&2)a_Pt}OXhryc!a#nz^|j3c&oDvTi@=-P!T`WCrDyO z_i!ldkHL!B@_4>j3oa%L;G$nh^Z%U;9lj?^3KR@ScmHkcDTXano4W7;@_QK z<|`av2`&o4W2dBGwo5 z+kGr5zI~vQXfLet4#a=g{2}si1e^VC0!t_#$1d6G(5XUI%6p?k0s}Kh4-p^#GJO)YE_?;%3 zvnsL-D8_R&Nvxhj8S^*5Vpj-G%{~K53GRq}2|RwndEQ>GtkGpBo2<8=$t}~TSJL`q z;wet9v7=yon)kV~1c*v(NAK2lw2ys+hE_ki%(+@&p9&~7HIO3bB+$oA1r&c~9ln2e zfWjtcJ{KNGhs{K&FHnoVBpoCzSyvM6x1i}U3b=h$9G^Hd!HCbs+RFp6s!SQ$rmrad z&PO_ZL=Hv2TghTdEB$H_!=#EGe4e}=xEO_2{Q}etFT~$}4al46jsuT3qwx{vbIk`v zmXAS#!esoEw8x^d#b}F-!iRwe^kJVgetwbTd+Sn~r1qE&7kkqLBYp0ZGsdX06_{7@ zi8A#*)55cTByaeLZl7tTqZSG{(zl;@u!e3pOTzN{Zq5u@MPDcHr5AfV1##XHf`C`+ z1zux=uwrIDZjKItRnuhfRy)KT0CzsqxB4$x5ZhWL$miLcyMe8O#5a|K&kkX9E<2W% zRqUckXGKsqPaHZ@&*{Cv5BBixZ}#7-iH#LXsdu}9KZM04x=N7#i`YG5WPx^!ni4MFlhGUyYu1r zWu%U5t4}mw?g+7yHZXqvAATI~C0~OYI#p~#`<_Rk`9dVl+OEX-y+VBVACG%g40Yoz z@!_5*k~a6!yJ_i&+jtQogVS)^sf{XKCGhof6z!`Gf@DJ&w3cthoZ-zhWWy)g(ceIq z&tHS2%|m?qkc!ZEJ7Covh^Z4pG4|+38nIwBS`&1+3u_iTDR5=>hc>f;*Y9W*??Y~u z3gW!{@o3^M&(&|HBh()DS4$A%U zIC1zm*2(kTrM4KBe)>i8t)*DmTp9LvdM{en^0P(bK4^DxcUZ5vurF+(u=ij0xm~q*T6G?NJ8$B#LK__jFD2=Nhv-6`KXzPR zhdEC6llELg>Q6TfEacAd91$NW*k&>B$hABy!+TjOh2;?FQ3dtx#>t82&#tc@d9*?NXKMxFZ@tU zz!3Q}@K0QX^J;E*=raXF5_Z!)o+tdz^{b#ox&t3ml-QJc8tl`l^_1zvJ#;afQTSaK z_qB|mA+3Qq72A<}?JRck^TD?l^(4k~1HDGdShrLQqW>(hX`>M`jxOf!eoM5yeMKo_ z6VW?_do|YcJWhTU#rEAKm1QYpoqdwb)}_%2w`l6SdyLG8JLZn_tm6bf2W zM{t-fV$kv^kpo@?28(R+j!WmR-^eW@RgzkaGGYMEDN&i8V0j2hXmL{Y|+V z+i-Gh05*qsW3(>!evA&q`n;tONbSHa&NkQ@vl4eyi}*7j3d0tD1T{yh$GYZwSQ2dMUR)f+oHik0_>Q2LcZ3qs(9E3&moRt1P zq3lg{*nMsotEeBrw!NJQi`5g*I7AI8Y9T^;9WE@C3>My&mciKdKWTnX8+YP<;W@_d z@Y~dY=l55GYw{4ZZ8Ws1hGMsKH|1qkk-1wcu1`6LvA=wIZ!{JQS|gD2(*a@KMf7f0 z0J-q4s?n-(EOUqwb3F7C?eTmsaIFgCW@K|NP%O6Y@ddrI!nk8$@N@S-h4Na=PkV&h zN4W1|TM|s}-Qo`W^UV0o8CKJgK))r9Q_j!zBrQ6V&6}&kIxlImPTK&crVzlMZn0&$ z-(5M|)(D@p6|g@&3`_16K}01J%X+w<*d&eA3{!}`%7Vb@6w3J-U|-KmJhG_6=H*v# zb<0xxG@gr{|0EF<;Kf2)tyt?UuzY_7d@uJ$#K?89`1g}t+T6>!7QSUN=0n+$OCl_2 z-y7r=NSWOXl1zUqlqSz~ATt9!&Xw+;dN{{#+bB z?>&tOxg`9arifdDPRim8y7BMi*(6vV zfkJRGUPfeML5@4BV!5}fY82M%KSsc}W_%fa2@c&U$bVl0`_yyz)G`B?n+)*#-%Z{@ zF2tZ&xhP2thwH;*$l-gL@#BTO^Li0C4ur!tdJP^+eZ@tyP6(o&;QISgC`s`Al}P|% zM~h%Z;}^;kUZs`=^{_C%fS^AoFh^%Q=zRcZxH#gDn;nWCOoV8NE!_EDA)Ro>w;GmZ%Y)w~Ry9*4c>X4%P4m z4Rqz4ptG;e&~&K~vR&Fm&Jv}h9iKrrD~qxFcNT`p213VYxUlb-lF;;mjPUw|Ks<}) zouVZx@oyZlxN9Sr^D7-isaA3MRbVP{jE>Iy_}k{aBWjzjQWON zq*i+ZeKpp6*UFzU2gW0_`5)zv>Z2uzoOd|q7c67%<8PoMidys_(`5*&MN4pD`*aK( zmXKBK?t&d@pOjkKW$k?zIapizqKOj93sO0=Q>zsFe0XJ{FsHJEEt ziz(w9`8o9wB?k}2oDM16uXaO|(mMPa;Q7B4Z6qww!=5A4u;2VK0-rsC`|M8;T;v}0 zY3HSaI_jCY4^~-I-CCo2i4%N-FRakD${{$@IZ0fY!T7vKi;aSeDce z+*rt&8PZ-T;4`S&(bq}d@i~>W_0k^s<8ZK!fNfVIBB%mW8(zTm^+S9e>cE_R&6wn` z)$F9gZrYKxlC1dXna8`$B;_1U8OM&(oJ)Qz z+B$|Y$1rwXLkR{}%h9W61I5P`Y^zxrOHD0iEzdbeLOC-He+*w0)kIy_+Gr^409+r1r+L*)_Y(@9p} zU(hv~BCHr+f?p5v(XnSd8c(Sr#mNl&Qa)nn_fEuR$gqD(s_c=i414%piw$V$V4su# z#WAu79+(TMi3Ye(HxolGW}-}H8v^b+BiZaN34=7CYoLxfpKR#L=k?^~;zXIN>=KP|>aHz(^(>Li*&SoF)bVXs8m8SkjkL&eXjNy>zhA$|*{6}7mq^mN z25T}@@~33qNl=+;ftfQj(3hWww*CYh5D$jStjhvTRV|9%;ZA}%s-$d2bZNB)bv1|3 z=l!8{Io^+&XS}BoAuq{f=rh_@FDIN$Qox#evVow}v$6h5K1^)>!s+U5++27BiBA0_AwC#a z`v;@fbU$3Ch2V3>9sG;n^K}PRo)MmaxoJPCG@y^ZTpA6L0RF6`H%Q`n;c69GSlk!~ z3&RQcdSC*Rv>C=EM*KtHrc?|IE`dFtG1}UDq9A%Vy06VcvE)fM&!&Ja@rY*EKXl{z zsArs&asu(k_&n&*3^?%f!{FNa*ycJDE0ujQs=5pRe&nIZWfnG7kL3MV6LcpqOn6X* zh?-OglPd9m&rl2Q@_X~@t59xNX31va?9wtRw*HGY^O4qJtCF->Zu|~7e6WR8&MvGg z<~#(6+lZZ*jy?0oqD{pdWA-XyiMI+ne9w?g9MEJZEgs|V^(+)Ll;Eyslu+d631Osf zsjw>23!>sy*#2-GF3lUtj@{Q_w?~OE`OQzM%;FR2ek_H>A$8C@RYdbUi1d;+ggoFG z_Y3#n^C^;IY6; zSy)yYqg$;UTik1)?|B{{5AZBuQ6~bOy0OaeG-ssLG1~(&Lh-3*DgIRwg)J>3MJZ8s zN}1;pMy+Jp#UiXLuN$?2GVIZr8oIpcG&$fXwGVdWJfwY?chLuutzL*cFc!Mo=Hj8p z6I6$0px2`r)!Y%6o^Ha@PlvNfF`cB?&7Ws4UeJW(6xI~ zb1}-od2>aDJDdKp+ja49+gpN93TH9h>=-^uhvFB{6g6dCgWj%M482lHA??9;$nv#lrC(B_~8g#dyvnX)o-yI55$DJX<>-j=ZR}$R)9?m2XTJD zr;Q1C!Z}R+1A6R{EqJXTYmG0NCu0h7f-U7IPxL&0br!OLR?HMFWbH=mu zAB4+2#G|vr7~n&_kz46bkUw3u&?AvF6*Ac&Mru}W5S+E(eOm&vp5$gA|H?A>T%#s6*Rt9;nMyZ zc)v4eU)K`be0CJmFuo<2;VDT9`xR)6bq)M`3Lq(d3x@h@;qrVgQtYQdOz|Q7TAspZ zW;F)$*@ci9z+Ytpj!Mm8vFpaLYc4zk!S5_;{OozdD-K>p{jl$@Mz_{BG#Wd=a{mrw zjtPOYm;^Rui$nd#D9FmYVvVO4T$7S9!7&(dSDYaF8mLYf#`+hGWF!04nT+^MoO|Gm zp}$YTad`&T)g?obGXZ~kA7!;2M_F0QF}BL61#il3aVB30{xMTDn(6cGo;NC-c2U80 z-t+3bK|40rV9J|ftetTjd9pJ2Q1F%(Nsq(~sVFE^WT3qC42FNuX1SRv?5?6FE97p* z9;Z?41J7ic&3?r9Tp!^ke1lq^wQKOUL52MW$cNuS^5R;s_m{9QIvz>2A<#_Q0%`ef zB*p*11@5?B{mp`J_olFaQxut&(oW&;L?7Y$;27cXSVOR-vtVPn7{^ZX{t^1|!t^)f z+7DBG)<^mp#F;)k!;y5=4BeJih?*zM>fR}`L-~9LR3DH0NwF9yx(91gm2leS2OT|^ zP4rcpaUBUeHhdNfix36dJ{-+6^wHa8`#+A(I~?ow{o^7d6p~0rGRtbX&-Zmx$!L$q zNLy1XskB6s>{TH%LXkv9MiHe&TF6LKC8Z^0B(2}|{r%f>=s2F-_vdq6=lOoWUdRpG z1#jm-1pm4O(T}$=W9A!}r|YsqBBR;dm$FR#aV!fgI>EA|9cV@g@Y)O(_erXZbUc3ov>j$LO9L+jbv`6ZO_^DUjP zn8%%0Qo?`|3E{=(BEmP~m$B<#D(;3{K+S?W(&3&tv$eXY5tV1@*VI`^tO~1q8AlU@ zm6Xw#OM{2?W6QWftj}-+Gh4A3!#}y8cGVtiOwMA_pE$2?#52~CZiBkd{Cj2V#Ipo< z@Ki4!Yhw~2{M3t?(qc?w*myQ|&Txb|4u^OO&v0wCQBSuB)LLrDs`et(#cx8X{2f-r z%0XGo7DeXsa5DA++^$~WOt5l{9m2VM<%=+MjW7B`Cu8~%U+BghL(sZ*EI4r&7uS~| zYHI^6XR>IpQb1Q4&x)w&uxCCdY<$Hg_Dp6IJF>xp%@16PGO49lyl)|{Bwxq7?)#{X zzlLG0`q1}VgLPRaaLn`o+E%ZDE$@J8W);x1{qN{i=40yHJ%DYYBUpWh7F%KdkMcWJ z(OqPW9?thJOXqn}-6aTK;*WQkk+^y(pYxFIXJ$6P zAoL#%m+T;yV+(~zJDh|M%$5jKADlyxLmEPc1i*Z-537mW%&hiKXFZ%z(Y!Pg-lr1r zpwA!bY5TBY`5q|ywIgWs0|dxtqda0LX`dWQZ{5Y{PC**7C&h!Uj>P0=3{R#^g~Y+7 zI5a8+t7b=F5qB*lMUH}6*e@Dk%=-gkbx=<^fxCxK;6+se&K~^_ufC`uZ~aOb51EWH zWunMzjl+c8NSM8w4wrg$s1Ex_pB@y^xsKWRYO@>%r6W0)JrAAPi!in44{hJ_5ib1s zK*>L_`3ld+BMeeUgE%{^94kD1VZ%9yBJa+z!5(33e3J)Tb@C?WX`I0j%X2uiPz5PI z@91oj4%QFbO4IG;4bI$#Ckacq3o!&q=L}F{rjD%vgYj<9X{_`(fe9w(Fitrf zJFHU>Xpss9?*2~V9eM4KjW7%y#FkzA0NaE0NcU5~$0h2V!!;6Cac7{k(F;{K-QY1f z0xhS7m~+Dn?Kb0?-=6ny_qc`ovOADx?8`GsrbyfqFBG9;LcQ}5Lem>SbLB2^%*k9 z91Z4WJDzDv2arR6D`lDa(dgJ5Hr6teRUSFYuuluk#$$QbR~|vbWQ0YYwd~^2VzyoR z7eww{=&8@+x-6sFPj2*MxXvB=~3}N+kcPT8nf*R9WNO!wE>yw(yqz%Wg z`DK9^+`0u(=a+ClOD5a!BaszP%wbDrDxpG}`x&(F(>G^6|Is{x8sFI%kmMfHh0<(t zyBu>8Erd;C9Q=BdxGRJ+2->#O%25&AVf+lI`LckKj z;9)feJ4CtTv|Ng%e!&=Hb_;!Ve~{N*^Fd*&XcxvtRH*bLdn z`Z#lP3_6bl)8nQ%iu{vG@BZqu`Ac+}iCQl~yQ6C25)4S3EPRgS~Y#%y-U&c9|PA+^1vMk=59;bO*K-Idfm921N7} z;p#CPLFFf4{W=~d1Bak&V1S8r0-n1x;CDf5K+_C*0V}b87SH89_2Zl#b9~v+NSmL( zq8hP!GOxJ@>6(1Z{&fr*uHBr;f0_!ecaZaqpQw&$#-I8U40o-@eZ70I%`(dTvw%B_PTUp*0V zJ5D3OQ4Gl|tdQTU0*@1A@cLPTcVZ`@9To?zB}q8A{0ub0&(PlS{?xaNbA9FBflcj0 z%klwyuaIKqoQJ#W_6O*+?Z?3(D{*MXlA$K1C2MY=bYtWQVsov!wX*Hx~&K^ZTbu`&fD93{w}V#+2a7uK!yI3V4?%84twLVH7u>b?(+?(dz2#PgE@J6LLQ$_Xdt% z9m72PPO!@tj|ri}%}rW1`+Xm+X(}UotRxRy1<+ zT#?s)7Fpv(*}5n>He_D|eLM4;_8xDfm+zipp!OG%14LP^s0U&{yP=Z%#dA7Ovw`qq zZ2$fL*r%{w3V1XKCoDhE?WLZm{k0KBE?#*3=n=BD9-;6`A@&6v!!e^$^m2F1%X#A5 z+a-pjHLu9fq8g#GC2)ED1o~rMQTRr0&Ntf!3-$wY3tnJaZ$4bOZ@%W-5yX0MR-Stx z3^s%#pydnQY3##so+a*amSQ`aAJDi=5!CbdX3UvKpb^b*)2PJ21AW$ZPK7lb(qj&z z7T}x;-;)&Y!IPK^$i9$+SMM%k%8+|ln_2;x6;U|w{0V)^`9P;`D)SsuJVk`8qtxkJ z1p62F;9aya=LSl$j!<=siCc{*_`u744L&BUgK}*U7EAXbgU>zd9Cg{Aq*z2*UgW&N zRJgq9;J+iD*D3r>QyaHnQIRX|W)hAy^IlB86iYbRguUxdaUV@2EgbDcdOG)Mo>&b{ zig%|w?Spa0MgqIPAEvV01e%?iMU@}Vkic7BxU)x2xb>H$aHfMl-)|k@bH_0JN3+<4 zWHUB>?I`w1q7*kPuJUfzBbc_#hix?9fy-})t(pg>SBGu>QacrP;$OMl6&Gu zGfM*ncJGfmyUhDI@&Cr7aza0OMbCkh3ov_{Dav!W3&e0NCWIP6Pg|F>yp}^-+yyUh zKcuMvH>jfZHLXq`hwGoDu;I@FXuM0rmltk0|NW#LvCE_M-982I$S0f+4n(ku!8Q zc6m7<%_ALwo^jZ4ARe`gA~1Si4$J}~VQr$wl6kIu!GdAz)$(Wf?(z<=3ZLQXx*+UJ zjl~qs@c!|HqhYj`BAw@(2WHD5_MZrLSc%|Q@dfUWx{VQ|3*Z%~j(a!wGiSCsB!cz| z?{v5cC3fr-F7jW1yR}+atgeW$+zqK&+lj&6PdMXJ46R|ZSPnO=Kfzg*5xShWp#{bL zd+^kD8|M$LgZ?(oYki#xxuj4&E9A5ExMJ$CJwr+E-|_A2duSDWg(<)HzE=#!h7>nk zS?mJmIw$U_w1v!`eVA5!0^YTW(6cYYxuIoHOXG8~FQ?d&7isL_i4yj8V9R|7tb>VoURG3?$%29uWpTv-orBm zL)dAz3MY& z@m@*2!xmI6vB%{wYv=~L-WUqihH48hmli=N+lcaQwK$7|Oy6t^%Sl_p`k-y> z5q}Pyn;^rQ`&(hDHk_3n;O|U6I}MyvOHWejNSh5t(ZF1G^{N}w8wg`xU)bRGnjo0* zGnAaKKNP+Mp{z0)gGMcecUvSDX{TT_-=!7uJoQRDBQ~*iD?+5Lu&3M&K1Ekx68;)9 zfBb@V$2}ZQxdpo`Wf*uQh4uu_TX9~-J#S5DQtO4uQ(30_UBEL1Rv4abi=l#FIHM@W zvN-#H(k2bWW@tlsq%G3jg7Bkb9~?Ul;mN~dR`Ir!-Ac=2U6DL{KkGiGP!1#-cjKvk z7>18c;@tEZkess)4dMKZXq^dH|4O(B-lBeS98`ISR_{hE4th;w6X)5o7b_iDcZw++ ztGJjsg&$zI>n1U0vE^+49B)>xG=e=Iroq z_-tVUc?U~~bpSm-E!fjN)7a7@lbNU1A8d&q&IbP($5e)06*@Q82w~bL{A_272O+lb z<9(Wek%QTS^lnf@H}Y1k#~w|8{H{yJquhDe$N5k$+m>S6vKn0R9L`SkjbjBL&S9f? z8gh?cfw=`|=Ks>}7aoD9IE~)neo3Uf9jw`u~0Z@Bgw9m^YkN%rIf0 z=aiZM?Jz!EILm4#WwNk{Ut}r$m5wS7LGHsk)*L**?CuQ}UOX@nk_-6m@#-10l`07z zOdBn%*=#Iad{v39^p<2I{QI}d`8|!>(Mh^9HMqlK9gDV@$qw_m<3iB`$l$Ye8Shv; z&fCc@DFm?cJ#lO|XJhU6)yhRUKW>dMI-+4@S?sD7-%I zi}NN&Fh9%(J{Ll<`h^~A?-;`deN<#;r)JQF=V$o+?m5lYbAjvE>G-vlGa*+NqrmMp zd@Nrer!Ez(tOCXBMcAK8S0rxmz_-TZc-!%poJ>BD_S=5a;(3bEeLpBLNS0^m6xicG zS}f?K6*Kq!PZ+swiEwPvBH`%6D>1NXBQ&KKBEMaM*|#gQjq)mNi2q@1H*v!`m%XU4 z%tH0DG}L)qz%|7ixU{wsng8l|7e$)RS{c*ibEYKWUxAw$`JA16A2Sws;aI~Qtf^QG zCGYJhJhBx=Nv<#*(aL#s->6792?0KB_?N)@N5wt(ebg93xZ`GSivYiV@qWN5&XR1Z zga>Ec_n%V3*tRcJwkVbYeG|$4S1J|pyx}=T9W+>Iqieh!bRC8GQ>=-Y_jy?Aa~ZmM zH_&k9B1zWgQ?pMlxqasuiMh5YbX$lI^1kez!Y1CYn8U`ti$~fWFJ$Wn!u9~z{!FfY1J?zApeJiARI;2b+%sn;KNHtcNsy?0aT)7*-(tYIBaQS9`4KiYKb5Vcx3(D?WQ$iBJ{%`#~wGrkQk z17>t;x{q_g^u(|` z*&W?0SHh=jIqo!zQ%qr>K#((v(v?HeKjtXL^Si@sg);Ko^oXp3ZjnW}7R-X)QmXn# z3bjhb(2+4XB5@qD79&W`K$~1-hf4n$ab z;Hmy5oe$u6ruUm-sJrGEUC!EP(kBSI-god z!(;xV-oRek);bxZubRT_3ZL;jKZ#rW<%Jz`8p2C;mck2<(&$=aKBb17qD5kjwEa>W z9Z3_xdX;v%+$f8>&P0fpMNyykMd}@yN)JEoM&aKuczMQvhCU**w86M%Gz`6RCCDs%=+7W7S#|tH;2uWeP5(P2_ApcUW4fVgv8q?qAM*d3;AH zdXIC>jCWvNEMxykE@!7NN3&Ywl5lY*Wh@S(F%4CyyI+eBt?Eq4s1WWy?qhEL3y8cDC9xvYBnlj4XMjpi^kNCC*^-h(6Z#`>OyHIQk8`bn=0UJ+6{eu z1-kG>kyLv|(6*xn?9?AKHY#`q^V0~U_Ydmm<+%6Mo}kX|?-Q`I7W3KH>q0Dk$#agT zo~TZ~K?$5CXwsifxe0~5PjU}ckDsF2EP$jvLTN&E0_7S!hiTno?xp+(uNxM0$^QiT z-#bmWgL|Mi`ww<(mt~EH-sBQ>h(;FdAn6Nl=!ruEjcse8El-Uxdz}dqHjYQGv@+k# znW5BtCTzzy2yS?YQ%Uwuf%uLXJYE-vt42?;VY>kLGVEb-cq)FjbH4rDFg$4vLe44P z>y#zV)tHSxOCOTu>2DMAz1QTq1u+@Ad#W_k&9`tUm%OD(P6M_zvG^ za5k328w_NNuydQ^Xi^EE|7ctzqal0-JJuP+aEwJX%tiKqiitC~oBs zno@9C5MXgh&>Wd9NQrO6vy=B>TznlPk1L_0#|ZwMEADk_C?d&O|@>}g51Fuw~wLtf(Y zrQd?}-$oEbb?Ir=Y^InwgUv~q%0_;cN4lvhR{WO5oMHL+XrBR@?^(Q$0nKhbOtVC`~+vF&UiGw{tc3{MsOL4Zjl2 zn~aTViQEO4#&bJ)SY>{fX5TqW^BRii$&)H^?zKbvCs1j9{?S z6xUiTVAnPn!%~N0`#4izM<9M^_#&yr6F<)vVNv;241akCYc6%-=JYmv)@?(;>Ry!T zJVqApEC@7?;iTee+*_B86+R4$cg%sG%@oYI$9V)BWT1RShUap=V%Y*dSAExwXzx=f z=L~w?N}kb-xF+1JP$sm9E*0K7?vA|v-MEnCg2HPT5V9=~&Kq(ubo5N_OWOu7qYkKD zm<^dFoS!p%5?=e3!ayqzPT~m&mcN8A$cF2fG!(y1pwOg6G_iaQH7!zOSEj1492K6c zfBOWhvzsvgQa?_7hlwTnjF~JM7EO<8mbPA3=Kguh?^=xkY1Jy zRJ!1o{EGe^y-rL2wo_0^F`V z`zxsmv4g|Wy{d;&?g!!z@5gjCc;eN(UX1Pf4UzAkA=oL(&ieLZHZm7r?wN#rcz4v^C?5OBK~R-DV=oc0Mb&woSL{#ML+wicIP z1#ref1cCL z0nZLw@c-v^kKyK8!MkiFs4q^$bca*e!PyCN_Z)Gs*#VV&SEy7KMFxtQB)^vLuIzN! zp8_=|96gj>UHcIdQcob^kcX}x@i6WS!Ku9maDNxX+3%d zj^pszBS>1d9iK9Np&h#hV@v!IFLfOYcI9Hy#A3MJ(#FKZA=t$`iK3HrFyX`yGzW;F zYCnglTK78MvSeq>Cf`p5ULnlHSMdpqzipNVi!Zn>19hp0nS& z`)>+%h#J$(s|Trf>MlIZ351_w07R+UOpfvUN5Xy_Tz%1?&cr+(ZB z9?5u3uj>`)pNA`po;sTlJ8=U@i%7|0QMK)f%FvO*;#nXx!6xl&CL(hePEt>zXPo zT_=SrpUFsP3nY#H!>R~VjJ@(7oAyz_bmvTBCcAV=rGE@f zE0QIfv>+HcdZE65B69qB4sCb?F0CuY?%Y>ovUz}B@BBx%F6Yxu_a1sx^Oa8eog`nU zNK(A*PfDx5ljDD1Xq!ti>E7j8kyaz5H%8HVgVXeAcQ)zGX{IN`wAsq1YD|7(C!XKz z6I|hW!2;BCn5`Av0pgCLryfFhkS{Ie5{r8i>{ z!F@fyr3axdb_7n_9YW2XVEo{@^nbf#vFg)f8e@5b3hjn7V^3+8TsMGSb*CsfIE(fi z&Y(`~X*jfK30C~&8Kyn~u2u;!*nB?ro|uoB zEIh1Rjrx;Y(UGgjQRf6C{hSpikrH z)9i}Xl&8J{ztzle*mpE$7z>%b@@}T@a+Gxkh$3~W46YmV+5F34?9VO(*43lOlFsMj zt5ynkvnHWWGm&?G%ds>67i^Q?U~fS$Zqq#|d0wM`n;NOY0q#Ok?+djO6^s0UG7|h2+)iF^bQ>BQ?|T z<#H^79aAtojDOEtda*A11%9-D#jb+G%t&M(o4oEcGhN(ID-S)UBHuDPwnB^zefbLm zmZR8}w_scMYcW;HQA|8%BV=6tQDm?T!%n}%SH2TpWZaI}fJPdvjj)wQ&h4dE3zFvP-OI_i5H4NM)X|)s?7I5+w=e|F|x2Lxu*udE*7n8_H){#s!XOMc61lzcD5>wtXo>?<q|8m1)1D2A%yPN-lTa@Qh78*4=szqi@y_ZuY_2 zeeQv?_TdcbOsbc$@mu{EcmN_@>+s^h z3IvW^!}B`(U^8hg2B*3}RdE5Sc5Z;9gfp=js z5;FO2Q6dbMT84Ny{2OgP^M~$R%VT-{By93{LrdR}Ci7SJWWK_LKAw7r**WL1Z%ia4 ze15YfHGOQ1={MGPf!`-DTERJ47H8cwnCEy?_Hv^h%PM-teS?{hs2(7d?%CM(?G5>E z9YNps#$&T>8M23Gb1ut!flQM%9SV6Y7-pr#N)!KrotX(;Cq)?Z?_ZyCDJ`o!g{RYs z5uq5*-_6T0h=(m)Sz)rB4qnZl!#l9PFq@zUPX$#{**+K6 zgZY_l_hRI!TjFctG^p@r)>!Xt7?B{3sxeHkV#F}ye2kzZKEqI~Xuv7in^^e44H` zsUrj7!)Dt1TZ^nqCt+B#B}zlVor6PBJa;fG3fm~fBOI6ab8c;$1m z@P;Ily*r5gkF$RyC=1`WmqUCXcX}`J!Vl9KQ1wrxdVNhiyJd~e-v5yENDS46(m2~O z7=qMbD2)n5Q^Fba{0PTrX+IRpextAZ)(Ds9TL{P8lNGMp;RX4;gE+8-kka=QJKeZT z)Z-Wb`2>=R5m2=94pO@Em-a{~VO9B144GRAy_qFw`ZfuAv#OC3ScRBTC*T&_O+v4F zQtVctR?c{xXjg~AH!3i7OUA1%8|c3}O$)+z;-=~fbQm2W8=WBhdb$oXIJ4BpCKqzG z+=r8_0sGQaHpt49$?epH_B~0?HIav=Y$fSM@O@;+RjA9npy#VK*v|$f*1s+bdTDcU za$qZF>|O=g-(k>Qkb&VU8JM;A31u{A)0kn0Y09D~WL_EzhsgtifeZs?d%O{r{yMn+ zVleFRiMIFNr1J1nEaJxrc2|R#TwE9l#xn9UFr%k7wai8R2D@Qaz^+Y6!KAEWv?i9| z#XoyA*chUJa5L|&_MopvhAmdqV>2or(k=Naiqu?5N`@El?$Jv$I+Y{#t`zR~C}6Jr zR`NAF$!fA!vD&>unDt`^=&f0Xmn(G8bm|o5FXJ8-2}j6r&#LYraV8hq38T8Bya#-Y z|IYbmq>Y~+L&Z_H=RVyX;QYq>SE2unpMR6oU{~vo=dZ)CQ>m1*g_==P5r`E1V>p`@ z2@juQ*v*}bD^oftN$L)%$PGelZ3Q)NlcA#b9+dy=JDr>PmJV31!q>+pSnuOQjRG0g z>Qs;G^=kN4I1;CmMKR0Vj=n|mzGr$II$D)c+uDGHL8Zt{3BmB~t1uyJ97HSI>9d^# zrt5H5)xvnXvRRtF{4$oMa7MP?Kqil)sqIlE z_V_kqTtqMGcXW`)KN+-ffAUSIjhJ=O4Byupj-4KUoPgWY`gt}?I|FZ*AS%+*d(xHSZ` zv`v|K+jLg*awZ!lp^RWHbC4va z1>GVkTzfhg5_`u$;B*O!_ite;?+t1OedXTkVN5(zk6F%o0A|>R>GhwGQ_Xp0{)ISU ze+JJj+%a*N4{C1(VU0>4luq$y-2tA@7axkUicwG=F%rRn>8KHuqiWh+ysEf{t$(lJ zL*>dUJlN^TJB3Riv(_243+|HAq*J8!Jc;5*jAVD4A=K^0p3cNSCd2+OYifoJDKeEh|^?-JH5;H*1yo4$o6oH$8|7thi2lEZ9-ODOx* zkin)Oc|{hQgYd(39h|jt+0KJEnE$~lrgd)xJ~!>bu8Z;Lnv+JwmMJ84Hkw?*s$fu7 z1*gv~I2(7D?vHs&N{8BM?t>oAdl6xySIDt4eIbk<2~fVm9=|F!uqGnQJMc;{H%%$PLD!E)^I(73ZCjanO#R$9WjXapU%R zEF1L@MF9`sWm1fpuFDZyx(RdU?niIVEj&Kg2~Ue0&Z?DXcUyFr^cDrS{l74n6~$uI zrBiq_`T{h0|0L_cRT!tt#kH55v3BS%pY8JO^02)KxEX}C{gpXbXcNL09vK}siayWkw84sI5vmuwft zT-+nfw~fIb^F##le;diZtN375f;*Svuq$>wE#D|h>Ys;W+1`=RpFAE?gC-(|Gnu}H z6hL)jD~KNwl~X;juI3F=Ris3$UR_DW-2)5YX%Lc=_ts}g!z{nnD2ZIrN&cWX(9`ez|mOO zD8(u(deO|e3}eUJFpWVo*&|H{mijiFmW&Ff(ltEKFy$7TH{ueDD7?xVwK@gm!!8Kw zXVnM<%R1TcL!Vil=S_C8G!ctb9wOP`F7H@*B1qo{1F_+-8!ONDj@D$`E^4!=bAL%Q zWH2_04TI$4$*ggz6^keZ+ctso89C=ysv;N}SFW-pb{AN)R3f|Nmqo3@Dt<0FORb#! z-|80yck=)Q{@0J#aAo%7^B?%R@j1%X7RtI#yUFW7$qWSTsDR(+hP_#OFVWcT})u z{vn)o4296$8v!z$8C&!SOC}Cwzan+nzZ(YJ0pP@*sl}rI{C@r%-H%r92N1@TKv9P? zk%wfWM4}Sw*VH3;7H2VjTg_Q9X2_gePPR1_^unuCFwEG5&2I#oW@XN7rO(4F=`-%> zG$DHAdzzcYnbH@GaAM149G)%14$T~KiIzHHe{?n zW0{vP;>v2ts_z$!3KiI_AVcOGeu{?j_tm$EleBS!Iy3ImW8n(oJnK9_H+&`G{ZI=Igj;GB^<&bFBp3r~p@ z>l z;gu@OZv9ecB0Rg|qEN-uz4F+-T03^s;T666PZIAMWMHA7%eFi*Wr|)k$b2}Co%0tm zr6EFgjn9QTnj)e2AQVH#l)@u-5R*ASf~`~kMx`PTX+_uzJ0SC@1=F{@CGDBY+#hcOA2AW;t6|6z)?2gXMk=5?{7l`w9CCI_ELvO0UWZt- z@x15#@|Z8){GN#^ug|i^sysGsxeE(9*iBnjDWb~Wmd}AtbHD5<)Qzpc*^r-zdGQP; z`6*a(*8}rBA|Ns%4;D-EXwNLZ)4HgDp{YFk9;eB67Cym^Md0j)X!KuA#roh!m@jO{ z&b<$j@4gFXUihH;a2!@U9|MCo_If|0 z8|#OHi;)Q28c1UE<7nUA@njw+&Q43~v)M(HS=tp{ly2m1jr*1;%Tqv>J@+(zkbubX zWBBjQb9fc-J7{PnO?~rKQRzWrTT}B79>uO`dU)8Y&XVygdLC*5UB6kB9el72&LA6=9(2 z7@=N|DvZ{uqH-+3Xau*q5#do%NfLY-5E#Q#n)l z_C~nt+!o|Lyex3ra!w$Z(T_mM7m$(V?^2_ocr(}-XODDKR!A#-aaV?V&|oBYDxu0t zA3D>_5$fd){}0Duo4pbiKP5@#cQs7_=TrXGhO^;dRIQ$ZzDy&u){o@xE&lw_DyFW) zD7v&ej4G;^V)T55uBwGtnD$*@?hsE-;(ShWI2-5Ba{kzR&X8@ZCX1B26s{nKMX~Nw zeo`F)c}jTVQp$ZD#%%u-UH16SF6367gw38<}E`4Yyh|oK@GOCyi}1fAvGE-C@pm6ncEuYJ~2l0?g&T1xJ}soJG3sc7__~-Q>r1uR%roseFk5`1pz9;nN`XBoD;|Ddh7{KlN7U-4) zpl81W#wBy6O|lZ23gm?WUE_pipXUin6FMMts zO`a!HyBEjYYfh6_Vjew@-%2l&zQQZ~4Q%*%JUm{MZCNbJ=BWS0=Bu}fjcp;N@=U6q zV1&tqOL2g+)#S$JQRCEd(s&V16S-?|OH~wqAD_l3iza;2zRCL{Z*fvy3WbUT^moHg z8X>lfoep(iaXSszT>Ec=%PoE6H#!O*Zf#>1JWsIAH&WTE05iWo_*Yhm)p~^(K#D<5Jvy)IYN5>t)|9BB{&#WhVl^Zy`6K53|;fdP^*iiyY-pQ zQgt?1!km4-!C(`njQ%*zB_8>Jgxh5h)~&{S^pkP;dN8cT3*fZ1nc7okqfwLpKDOD? zcg=Nl<=Iry-&6=4&WrfbdWGLXIFoaSDEbq>)A})H=r>S-qI4Rq-ai=nh7PE@o=tC^NF1Dd#JvR^L)FfBQj(qtZzNQejC1?=wA{Z8YpU9nbm2BMB~mO#Z8{Qb5HG5faTdanZBA2Gm}Ap~~YiH@kolU@F4 zx;88nX-iLGwbcPkx3a~c2m4X8n!B;&d#SR3=jMKnLC|i#gM5$zpV`-NE2EEo4C)}& zgsW6My`2h!CD8r$E9IEZh3d4m*nQO={}xV1V~`NbHV?(NDiJsu{-j#3B61J4AnSrn zWV&8RBVv+i(e?~lQg?;AhNV-42;YmI_n}_5<7|0B0V^6;#V&pCC+|sJw5&%SajRkx zFg*cv5{Y=yZV9WPMVRxKXYOpf$>Gp#`WF0xT&lGo!(9%^E9PN=^d*6eTDV~6F<-$? zp0ynxrp8$hQdm<^Nij!0lPPB~r^Sa*`}}xX7aByCz7r@yuT@a|R#xCQ*MKfAJxIE8 ztLb0qcftA<-vts2Cy~bhf4z?Pi4J#rcl_;S|Mn)o(@w>mje}Db%>vdoE?{m*N z&+~cS^LaQ=J9;y(swpt&y->k>=9tadChvf7rt?_A%yTZ`?e2)@iHC0DoqnVa4%!nT`-3T{ zj1KTzcB?Tq6dH?I51ZS1o47;Sk;*w|JhQxQzFzYm{fO zUzE8xvzZsRQbl;%cdTZHduB4)iKDzxfi4pnFrD%8 zFk|Eri+DckQ+PZD8=l-7VD#gInKgIbkWuEO0esKX_@fhZjv?JsJ@>q7p9h-p`d1 z>O2gYc9&wFqWfOnB(o*F0PTOgEg?g^XV1%dD^xBpxqAp>_D_oK&|J+sAqeK_HyiWv zeCrwgHQmfDlbei1LIJPyjnaWDN6`GQt=pp5Q$2BF)$G&V^Vp`D#0bID-%K&$NO!#WoU;XhASJUy0OTD?Ko>v@dl?61L#vl_!hTmIptCW|v2avWo8E6k($XUvYjE$rY4 zS7gSP=`bFzLV0QbR`5p01@rt~U*$D-r}Bm?$Me*}Ht;rIyu{0R*vJz#c);^sKFaG` z^O(mBxiOA~3x)Z8|1tcJ6B&g8U&dLO`?yE@3iDp=M2X44y{MexB>( zv%E^nV&0z#*}OCT-n?65gx+<_r));rKX$p3l(cD*o|=fMsL20*mph4xi-<@|i~OI@ zL=i^BYv1N=+YfCu+v0V=%M51En#JQ;Sgc?Vc|sGMxGBtmQ&wSjZW$)+OPgWKaY@UMTSACg#009Wn1+9R^w}p;y5#lF zT*y!^CTf?(;BB%e9I8yi1GRqGJa8VDPfo$;pv7o)XdPDGI!ElbokB;?*SMh&@LI_( z^tbcF+#PM?jIo4e;JIV4cH}tLSL~z61!1*#DQb14w-9Xo16bbHJX!kUoz_Czetb{ZN$uSV#KW zCQ5o(t|GQuCxOuW!CY*7eC$`-DHm^l6L8YVBa0Fh2^<-zEyx}oCU^9-on~Z zTl}@=0tvlTi^2DUQB*INm<}C5t-B-S?HEbRg}sJsz>zRC43>eQt;6K{KXp2H)n-nu z*NE7CdWkA4@8TrmTK4+BV<6Lc7EI*3a6>{ccDkrzF|EaN-Dp&t_<_ukRRh2MwfuKH zPtx`1E-8&mLSY^y+idBAv%BMQ@#I3Ryq}LoS;sK(Q##VpbohDP8XDV9gZXS%d=(%E zfxflqyFQQh_8i9bc3xO?tBY*#)53B4w!mS}vCznxaVBv>z08-DG$1zzTgLig?_EjD z`#$nGM_d*4^bhHLJi>XuWooc|310rKZ`CM6AP)sCS=`kg%Gp9 z%buwTz?>mX%k4KDaca2}nEF!?UDX9aTLQshnK>pYxnZMoAFzkcz~$e+Kuq6*Y-{@t z#TSRrb+r!Nct{ctAC7^!%VVJ|V;pQ~F2~&;f5DHP?eM5S59eB1;H>)+7}NC&zC6!m z``6b%rs5KuXjuY0qe`3-e+M!^MfI^Vh>t~iZ*^6%p6vT$4~o{POHCvpFk z81zr!GF+7Xg#$86g@?KWwr!t<{FcD+Y-a)X& zB9!(Zc<=QOR;+Uzu5=ksREDKv7C?6DdX^p>uGdpKaZb33ulPV;YKJ)uv*CBY6MWy70bN@!!C_|;d|0TA_zqdkZ3*P|mP9B~P{04Ob#=Q#yS4R^#IzNQHH<-m}SuZ3nmQ@g&a0#-mJezz!(L?skj^@6mT;tYu ziqi1Bi)8Ps0diyi3VOywnoio%Pkh*Z*jjB4UN0ipYYKsEgGm54lzNkVS=7r!%AE&6 z#XIN=Fk!dqtY$6-t|n}MJ^7(mLYAz*%N`R`q7SAR(wr0WG__)sOFh!Os`FqjU*;ffhr;?GW-$|_*OK1Dq(eR#ABZ10&QWZX z_$2Ou*A2+{kVKd%H@VAeufe1556CVbce=_>vnq2>Kihbm;p`8VkO$T>G%8(#4k)iD zL;ioDM9GC!ZLZ{uhr+pUOPk2`NoBB3`;ob0miQjgyXE1{n_uk?z2G3(7^v6&IgAC&C z-#{8Ry<-w861lr(;V{c~8J*h|!rD2+LelK$>eP92$$J$wFwjfo9D=rxxj|)I=sq=4 z=Pd>SLLZR$U>Q+#9$}w1#*piK6sfYPD7l|S!SO{D@jItMJ2^et_+JA=SBYZ8x=eC< zB!n#d`;{!2a*)?f%7}h6wqN$ob-WJVvOJW#b!&o@$PY zF7h~Vo<%9Q|Ilq>06L&MR7l*$L!}$YSKTMhFw`lRvWiA@Cny$zEUlEFXnYhdD5s zyASj2RB@X1X*_W*OjyIAgc%Xvpmt^}_i{`slsucn%}CIN>sEJ}?jIJoBf$?t4jO^H zwITXfTVci;UA)#Y88wFjVSP(B-0fbEj+Kw#?i4Hhw?H1x{+YlQSgVuAf;mvKtR6H4 zrl_4}fd?)7xDD;wab`v}*o8F1f4oB=H$M>?!sv~ftW<0*1+YfJZ8{o}VQH$KJ6Hm3kB5uJSZktvL#s)E>Rk-EgMYB24&qnEUr$ zfs5@|;$A3t^E~KVd$(RjWu_1z17*1OWwasxtKl&BkzhVv^zH8@R zmxyD``#0EuVQ0r=*7lKK@b?3q_CFst<*C#&fSmMt&2ye(Rw zkaCWU*;hb9v^t3LjW}4s%Z5}h7Allo`Q`E-xb9g`*;kt_VX|#3mw9_3IdCYDTR46a zxAmw6R4D1O*4x@ylfa`)kGLM*xH<`zS)78^%}-fcn#UyVZfDbqq##Df5w3o2#zbh? zk(uMQ!9`3Sx8IjXM){SnQ|Jr-K%rRq3Bg?n>+>OSX-$G1^neWg1Sr8qY(BFn0pBE6~)v zUc$Lr0Ce|d(AN0Pjhl3x^$_OQ4ORh^M8!gMM-n8ySpkRT)45w#ufe`98FoZPa}N)n zBRBjqiSq^V5R^AikRFJVVF7?bBJ zTi}<%9aj0NE|Giku)21PFZ4xhCTAiXP%0u&AP43? zk7Du-BSHU11FIue%OyiJ$zRn3N-B%ENml2`;pZycPs3B>One#fvs(|MzP+$)$$fC& z=?7aERgm3vwrDh29^YmU@pHz95cSv~xKKG6>W#+(*H#635LqgzN8nJNz*3H;hKe(j5|Cp?iZGTAWhC zykC>COKug`^E04tTpgr{)WI@iO`MC#TtTTKr&J3<1n>6SBZyr6jy?s6&ookK$ofasz&k7FS%m(S1 zGqH8TNv0vPib$SFM8{fT@{Q+l*gLQS0*l`Az@K-NqF)l1q$GTKIoaQqNr@ESQXz5oGC59HM}!2Fbel(L`)onuXDM^F4(j)Bq4L+d1PZyCL&96e}-dt2Uh_pIxr7=9#LncMFeya@=fk zdQSnX6cSEmt(*fIi{`-S4S$f8+QLoqyv%;;{LNZ^RpZtuSVHXu12}$J1-|_ucM_#?PAZ)R+_m|d5Lq2fQlr|qu}TOD9ct`~dk0t$cEU!M>XTRJ zy}53QH1_m?Y-XhQ8h1N%IV?JAOf<)CBX1{OV+^PCFk_!s5<$T{@USW4FV(Svw8~A8 zyQ~I4@_U&3$l`)2UP{-v2gm8|fFFAd)5cbFU zQ2wPSr%A$H9r*fEf{QB~;hVlG=04glg_bx2a5bC?Tc3@vN0<%d{>>szO_;-??Xa9E zEI$K}dnOU?!YC_up$;^jc#!gwxhywsJ%7Gp43{!n_)H!Bz}PsOLx_wVskgXBX8IIz zAJP^=liWq_nAkedkWhu^FQlMbcz-k6ve~433D_n&mB_1|=XXk}!nb*~toauac&4}n ztdd;VPAp~P zFO0W@!{Ks#+;)z&GKnXB(Vjq!UsorY*a*AVv*|J2A-W;#Ce@PM3RA;9$!7yUDz*GM zb$fZ87Bou>#u(<4W8J27xq>rYHkd~HEMn=Y0deYMEybVpvxi(eR6(qMt|4E}^^!;O zFNnMDQTlr8Sz0pf8V&QjN7sKEqItjM1Yx29G^$)m5NF&*11B)lbz3=2NN=X|%w+`c zvg~QP!cCgK*PK={8)$lT6%8v*XCH4KOJB{-Cu6w>WQ&dur2Yw@KfZL+Vx7g*d*@=h z?fxgONkx=4jK9fFm~(?ZoEb#(-n7s=heZV+(m(WuV=9|>(WeG9$!g*&DclptUW^)sGgv| zm449p96bTs{+GrjcT%h5c=|N>7mbUzrmOK2m1}xP`$AG#7hN9o_!6qjPo~e>%ILK0 zcse_n<6i$ZrYA&#sn*^)8gJZ7#q6_b!@U5ypyUy4>I|Y4D@{nxup6k9ko$pzTiPk8&;3d2MFb{KuIPkl)9tRnYMv~mj_+29R(G~PmPyc{WOV$icOip>6C39HPK$Y_mlEts7TM-ORWu)8y! z7JmSz)eG>9R4RCG7UElf3^7VxxUPE|(Wovt+%hm8RkQQ)(x#Ibq#;3@4a{Jr*8-GH zeGFHRgy1KqoA|eT0E`-HaoX`BTrlM~*!pPT#wBO@3l$346Sj6}n)a35^)Cl67aHTa zaRQb*R|-96n^D<16IZ?~CH+}G$j;5fHFslB)i4rWZ0zvI^J!S3TMqIMJaK)`e#|>L z7uA!Z(N^XMWIf#v+ju9rSwF(z?gJkviaZ07@d6w+OM>uK3V8EM7744Y<&~Y0g_CMQ zkdb{7iq2Rf-{BO8gJIlJ|6H~;MHRJf2k;+m7=hT|)9EL95e#-bi)&&c;N<8WRBU{N zf6C^-sUv6L$rW>aY_}HH#7dw^OD_Mo%uCjeW`VVx9&QS8CFa*K{GDWn9Rh83 z+4^!QG2f5cYI)@6-&*ct#v$yDPeGOSKS6D?D4q3gR?{|aM_?Aj&HWXifL<* zs};p^_c?S~vnyR>{+2#Y78hK;q9s^1t&;1#tqLbRei4WI9=LQ_5%@Lc0+CDpsG(Vo zBlkV1VT=eWbak;?OBtLUKN)9Bhl6XE4J(z?NBbX13HEi(5jd_1qqfug>7<FuHubWfj+FiN73ja@bsrbdhrT==3aSR5}dFud;yYB9gawX+o@bIebw z?>C?R*ZzneUSCbqyV~iWs0(z=4tarnfu7*j=vcwO?f|-R!7-|B)J>O`sdFyLt+4%J zH27o=LgCu8P#`!9!6G-gS+R299({!-AJ@R0#SAFNHCLI=P!q)QWCW&Z#jrWuP_X&L zTiW^O5`8>uDCj#oiF_Va6@=MG;<4a7*j%P9Xnwm3BNsQ)bBbdGC%xO*;t_ShqVy^< z?PV%;Gb^BMNH0j4o~L~6aRMqE&xfmvg?N-UUaQ$mM<$FwYSMKu%+|uq#Q`L1eJ1s} zS^x>mFg5fx6y5iPNB6g^l zwg<0)_^o<$X4ok@#Td}b>nW@|d>zu3XoI2jGg@`IoTk-pAPLv~py7@NWTndp7O1q- ziD9>B#Nul-Yqq*T-Xxc(_LPtr>PaMJjyLs_*vdM^gi-5LqBx=b0&PB!MhD*%(1y!} zloKtare`$-S98Y0!Qm=sHF*uP2JLi*wFo6XN9gQ%tMG!!9eUcmfn)C7$|N`mbXVZ@*-i%wm0pAN^~pcUL@#!$3? zoH%)wOP%qaQEXXA8Xrwy{cH}v&52p?iKhS;y590deI9U4ze2(7s4i=~?F@TVzlXIy zrVL-6E`wW{)1hO?hb{ZGjoZ8IABk?-%cgKs;O<5T_;9U4AFI?elFGmwI*<0+5{2k1)8dc)lf0az|%w^5a7L(qPF7j@#3#>UBM1EB~<`%n3 zakdvu5bfE%TyG+e8(wO|8hFS;i0WdPFv%Uh{BYwW8m_Y)Gj6g!Uz%`f${W^Qd-t~%P_Q#db*Q3Ut!S=B3mtXKM zbd*C~wFp~q{VsdU!-QPR0aE>18jjdf?&PuC?6mt6;n0I0+}Mqh-1(RJMmj8 z$onVr*KOgGxZMs!G0+6;|8k@t#heU>Ny2^U_pDlN7`gVamp5_AU2fr+5$2~=C%e1S zoQz91A*NbBWb<2nVtY1_OR6tob<1qP?e$&$-JWC!3hsxU7XEN)bsqC@w;T8xmcUMS z0(*fQAg>OpkUWbWWY3CG?x}4Ud<`szPV@QPpTT|PYLo=GS>-T0J;8{qEzN=onGSA% zb>}9<&gNFWc7!bwnqZ$20N-Axa=(6ulIr*O5G;GVB#C0 z-#&=6^_^tL@&j1=FbQ(+>kLl0Pw12XDFBb?4aDJorl;ZPx;~ZqTya#5xj}bgPh%F z@SyiH@H~V)-{LpfamE^~L#sHHUtI-}YlU8}=XPxR^XE)|Q6+bH|7`faCyKoK97Du+ ziBuWHmX?9I?5p!n7&+1)OL9oQXjotM^ne;YCa=w$^5i&*$JM07Zu1| z9w5A3(NyR0SV2aU7A4(ru&zpuXx+=Ev$wCH)o+yqfjdX&&4~ka(D)0D&mN{%-rlF5 zQfg^z_A8KF@PT~OxXYQA->1HR`BZ<6s-WJ}j_euYliT@+NYa2AeSg1?>PcLqNmHc+ ze#gJlRe5h{V{aEkO*lo)eej|KcTZEJwE5IYw~~t9mE@j@E~VkUhiSv~EA(N#vcM|2 zk9$IN^K-}Q%<#nUQJ1(Q=i(>l(0Pd zvV1q?J&$6=UzAc$@7+W>$%<@jy+Ye=ucXD-(`e9#>vUv%CwcvSGW}!~K?8I+x;Uzd z7G0AUm<6=a8Pm(D)XRE0Prr;9?7dCC`>mi3@_(pxL=1hXdX8>z%%{c2uh50d+o{qR zX}Y{Jk!tohQ+MNd`aWGmAfc`<2r3m9tWh5$7<4PAe-?R>(S1IUe$SD9RzF3^l4Pii zKT2LNtEFb=L8dHgMy-qhYrziLF#jYb3X&$|&?=6Zeh z$8{Z`FY=HCpYen;F)KK{a2Ih1j-v*LCQ|z`IaJiKihh#Hr9Dr|XvNXzMDn%-9V48L zB|H1cD!n1nK9WauBWvhPds#u_n%7im_#=JXKTP-S8>BhHYj1B*5-9f?3b+YTlvcLT z>EZ1(X^ybd>zN{bSp1#%+5}TorI&QDQAaQ=lTTXn7LuSNyXozj>VlHPh4g8{E$S-C zLb@gc(^uqy+_NBhw>6gL-}**hKF_9Y$scI-FX6Sm-$qYIv2=s06a5e=%=(VF#m}_N zBnBDUP$1LESUt)HyW{^s5hF{?tw+e%s5bVIK!{nD{D9-v)ycl*mvB*O5#G6^LLw5T zvoq2VBbziZ_wiHqfrkjz3R21P$%M@ObeU0Fq9^?C!{M^Ju{505kUlgFlK)Mv2f2IyLJV^=kv?q{Qdc`*-swM8tCGEbp>+f#xK^#eFgWyRvGeYhPc>J zA&1s?f=gSyyh3P{fa!OIq(lYK2T$h?Xg0UN^PiA zyv&}Fljmt@=MnAfB%*$-l1)+`B>SxjA@w~6l5wbZvOny4uUl5GYl+<&{IF(JyU`tkeK z`20c)jOjiAKMuSg3f{*_?Yd&3&s#v8Vjn|)k3PH0A%)l&j6<&%-oP8ug{3EjT}$cN zoX+|P4mveJc6UBm)iFS9UjzuV%TANH(j@lPCS&sGy*)X@yTa;DbR*lVY{}7&ePm)` z3j1)?AFeYjgL82*#JV{q@Xl8h%D=l~$IrEJDE}e+_^XY)v!(2Myv5n(O@c+oJ4j~j z4mMg42-}~{lUBt~Bg*OcIhV?KLA$I7*!tvVJ5F$;Fe?}h1C!cfyf4$bNxgP%_tw_xKVw($E- z7&k!%HjMiZEZw8Q;X)+T-$8uw;Vm>sOvV2WsN%EnbrAS>B?jd&D0l2D98XIIndwvT z@r!E^eo_uae)hmG9TwM#ctF^Ggt@j1s4E#ky0io5u*wR?x>gAj9u2}CpEr^z7ogG*W1XQU6uTTweP^ObP(2+D&W4%yUdPmBbHl* zcxX)`d(E2cuATYIZulwW3OYPCLXj=|+!_n`HM2Q2oB0J(!PVE?w%S7P0qg|aF8 zv&*@QUu+>MzyR(oo{OJGr;*5dNj%{ggs-$mxrX)CoU-R0Sk|`_B-5i{`!qMmY;NJ7 z@u>oP2^UN zoN~7vU`|;93G=?enD)(Qm-p*HQ-C|H6lbYKhbZS`m(8VY{K37Q;YzQcQlL+qhB?pR z<@B_z4Bh$ZCAqHVMrM4@gDMzc%L;CjaZ% zDN47$IY1j!*U}|Z$CG+f5jxM$fbRPcK`${wY~xTCd@OO`Le6K9lt0bH^Kw4hYG6dv zOS4J5aQ`uYe6F}i7ucwWgfw*#%o!$a_r_6k8DHw&+(sTYEv4&&w^P_~gf^PrAvOmH zthy%#-!Dp2ov}w@;Tu^oWoMtKEYMXb{1MM%#?eXz3eF|H5z89wV=97~*V!6djO{w(^8#?<}5Rw0Ogq}J5izr^QrYYqLH2kL`|Ixx* zQL+uV@Ksvu_=LTGCyu1^9Woj@d>koMy9Y%@@9q5DM8{G5<@?<@uLkpDM zsP%(xvPIm)Sjm7cm(OCT-3)dC@ zu?+$xe!#L}cAfK9NN~<%&mH=YOY$>?M@@bFe=Xf)%BIgCyZsIDu9m^vIS)a`Llzrv z&OzzlW;otdAGUO+!NX5R>=e7-oNLH^u$zAx6>dL)jWa~h`)x5B_Nf($#~gu~H`+i= z{sX9nt{{?vgV@~Vf$`2W@XRU~&Mz{UjD7r?Gi_0V1|3J7u3Q8Iib*i->`c(_tS4xk z4ZoddM@*1Me}pFSvHS;h)y0tx^J^Q#e9hS5Cko+CZN`;JWzdSjdh$D zMurZ-Jm*k$hx;d3v}ZEVreA=By?bz8Vkpk{G(^eCiPdr2q}i%zZJgYdbL=9a z*Yf+s3B>W?8&E&#h)#aym}c1m2UV&dYSkT}3E}YZNGGdxbTgS0*6I0nT%fVno4 zL0qhc3mTpQHLEwVw&jzF#p=oQfRHS_)KkQkZp>pZ^^L$yrWa1kJrAP64q$oh3LA8E z985X&6z1w)gZ@_&AVAn!eJ;QVHw70$#9~qWGNT8|4$E>Wah`a#cLs(`mBH3pAsyW> zhWsaM$?8#aep+@7E4yfukjst(D-}^3^T3_!?Uuoqd0(L7$!5^OnYbjl5#B6K;BUV3 zggY4$%02t~rTXu=@i@ejz_Qp%@E$)8bK+)T(&8DIZuc0(mRK`mTAQHB-VU3M9}?58 z(qN{f0hj&GfZoMbV8dp>-XG`SO1T3bI~GRtmz*c+CvHIQ>i=N;EK?HtCWJolbcR92 zQPLY!N#hQ7lMNyk)GEA_z4T!m%62Q_-+`;3HzExh4#}LzhC?J@g1|)~*Sas{2KU?P z63qE+OzblZsOi8e@>Q<|bf->)Uy^rbpVt^mWsU`tLqhtM;g!R<>QC%r@ux&dsP?1eNR~4 zyaXyXT0>qAm%-8GYA!{sj!de}f)l%z(lSPYEOJ}KMiyNleN)>=k!=~LXBP~!YAKi2 z;lnva&WA|hJ*{g}!IphSl+|(tiOR_&>E0}ww)`Ra;@JR?E`@VP3oljI|4W2t7lXOg ze--%ezTE}Es!q~J-;t^af7WASHPL$Uh(G&)0?|njY9n4`vDbzBz}-B{BH55=_(Y-P=N_{{%-;pd;i*UnauwE@}LVebo~SHE2N97$M6r9;>vz zv(9&hiL){T@pqlcxB1F&-0BtkJ+Od%-6c!ocAe&CS3M`&-pr#v>$b450mU#^ycmv& z#Iv%yRfT$6A4+x?bH>{WN!?F<`hN06yp`Mpx2ET?C3CysRGBVnrDk#uO@?82*%DyT z6y%qTz`IO4ygbtp6!s*N?0fd)1D+;{wekEGhYnKIkxM+PHWH_)4&=|>F7_)mg6B`i zvh#cu*mdL6i1;<2H>4eC^ZvQe-m{l45z!9su9bjg)mf4-!IK!j*b5siFTpy6e_Uh6 zQ|{Z3m#{rn3Y0dEN3*zoFgg5@h~H~wy`$8qt5Ym(cvT2e&6e==83S_%otab1J0LhN zgbdVAM5j**M5qantTv@W5jWhaRvN1{RG~AvWAk3KDOJxh1nVR5K7mSkUZrG z@EN+s;XOC-`#A##21Rkb*g{+%y%awy*1?b9ZdhThj@Q2VLaS31oNXI{UDBKQYukO` zYfLuk+^|K_d(XHYjDlq64}9gIQU1lckFe*EIw@?h#Q>`;EN@?oQsPD=uzf5nm6!w5 zDnGJ%cl1e!oe;@LP$MV4jKd9rOz;$)iKo00;n$dI_Qt0Du(K%}f;$DQwP7^ev{uDg z>fd3jmJ7_E>fcK&paDA-|N4ofMeOd{zcy=0c?VrQbwF}v0 z`=*0d{sng0tf^S=ZV6Vs2*%p>J5X*|%bxJL3db}q!N>4Y=niwoZO=|XSIct3$l9EP2d5?I46hOV8hoLreHnAgXG7m)`WuLcmG77RaT?8WJM zQJ8Hq9hD+FK(eC`PP^8DS9>V$U~2-r_q8J$FDGM2e-8Vpb{mK$$fM07FZ@_(k4vOq zgRRR2xE_57l%!Q~qG&ZdaM=NBl|~R)uo{N+MA0+v7WAn~gX;Z8_EGviob0TQCYs_X z(=r#8|4PH6_+)T6a-DTi&Sc9fJ>b83J*?P22x`|-LFPg$oWI!&tK3$wo+<>^s20H8 z=1_=yEs2{quEsIm^D()^9bTQYK;tpS`1yh(dvXlF`o6tA>~(7MFHvLSoMXa4(f2ffkm~(X^w(_kJB!C+A5qD%9?X+rBZaT_yZZYVkO_%J(OtL zeUhwlUZ`nIAo9XvzFFnuc2lGKe;sy7pd1S8|v2}MXhfB<%&<;BeN~MNk)7o zeZTZ94VKC!2VM7(W9!4n%Bgx(&zyz2Iwuk=JTKf)^&{+kY3iw-MM^u)kXzz6_%Z!HwPGVl_?s zfP(EaaZXh$!-&hl)0G>k`4U}{{-YMEhcxM6gFLC(9Y-fgXwjH*pgw^=NI*pqS+)28 z^`7ZYm45XTy_2h{#-VPuQGO=-w``D;{m+DcGclkG-_M|%o4sg`jvoE=XA%u`QKd!q zLg>`OBxp={mv9?qs zqnjMj^?>nrLrK0uIf)Llpt$M_$x2Hg%L1>H*Y-C^htzqZGu4bnKrPu?6Utpn{zxuK zPNtduzI1;1b{dfoKo?cT^XeSzIDfCN%+&dzz&RyDL46BJw&f(uW(NA3l!*4^!aid zpIo_xN5Uo1V4erw-?I#NEWVCPmu=B^@>kq*qXphHl|uimZ{W!L4O?0xaIai2?$20* zL9fSSanX5ef>z5c&)c%&oyiN5jxdsToDTUdOMeOVO<>9IYn~qxPQL$V#oix0hw` zb(sL=LW0@+c~haTV;Ow$PXfR1ftclY4%skmbn>r*L9Br2DPJ+VLEciupa-v-Za`6$ zIF#IS4d1QK!7#N+7??Z>FJ5fNBM^y|Ij!h)(GRCzz6QduNGq&onn&)z^PV2u{rxj8d2;_kLV3@E;Xmo=Ku+w3pqM8+W9kcaO-Qk=L^19?vyp}yw`t_nGd4`yD$ ztWV2PwK@~ICJ{>-`~;b=ZO=y1m2il1K#+}xNn53_be#U!ILZJYC58X%C!?*E0 zuK~vuS7ZCQNGz$3!*7R6(PFj{1~!M%s}-lI?raNcUm;IvdlD&|FGhO1UUOZB64d0D zDHWNWK+nqO(%7w~w6^dxf930ZQt9`GtZ$MRa@I$v!RSfyy=NBvVswrutoEeJMGRf< zM@iU+HH0s3M!AS);vzbe?!5GaNIo?p9X8d(rX`k6TaZfM1@5LR!>`Z)>3o_Sxtn@D zWNCH(A`)l6j7GiuNfO@4P^pua6svsbo#$~>r!0?_Md{K$BQN^q?q=FuG=@e@)}ZoF z-;-kpkQy4+$>l$4e#cls4@&kjQnvqiIzrdlwm*2^o>uPLYO@kt9U+$mnSl*|HT`nNfs{ ztnfQeuh;AS{r~ejx7(w$9_MkL>$-ZL>zv1VoX34$xU5p$Qr6<$tHNISePq|qZIMN; z43u5pW+L-QKi0{sQA#J(pqVn4C?naxM@EH1)%zB@w$1GnW!O$;Gt5|)_oTK^?9a(& z2F;LtJ)SK)m=P))+C@v|f9zM`R{t%s!!@R|O|D%FzlndNPgdHpmXjLGmajcf7^2%$ zc4uK);o0aoS>#dwLh5zxJo4kzV>;^DvdCWBWQ(F!%c3)#W$4tu)1TG13T0+L3bO`j z%YNEu%9d^YRk*}q)UjX6#|z(Wx^&FRTar0#N-gYOv`v=My02`We}=5)j(^AS`RRo} z4k^mUKV4b)qtI1mQZYnU=Du3?ZRcEB&&x5g6=l0*(u;Mn>9?23va;sN?zA+MU5QeX zxjxI3b$hc%_TW{xEaYBj;T?y>qoLQY78<4}$(|KVk{x_JO*Z0FysUNaESYnmu)7Rw zCi9MsmfhK~TK45fj?8CthOEkNg{)e5JvRE9UHC3RSJp!NYhho@3xz-D`WMbwG^)^` z!Cf|E%5vGqwpy~j7t^S8B=%ToMR!^KIUQMPk0!E-Ga?H&=FKjg#Wn3F`*iF~q5S;)ErWbT1ut>nUeN#H{Cy|kCZc1< z|Ks`l;or`i#0mF#5lKYir@KRGmS@b7q%?K5kJekgS#gTlKJCk!YQmA_{l~^MD+gUbd zJB8mEm-rqNG&a%umB7o1%4FuO0Cc;x1TTE!p%Fwqa>1%}t*Y)nqt#)SYV*NwvV4;Oqw=}WduL8g01SVzr z8~A?{zGB*MVb$slJlcC6pWTe1sn9o1J+y-_XYAquX%~kautmuOGc5RIjWg<7aj11M zj(x4h1l3ktnxw+A9be$xEmlg+nIlykj+b0~?_-A6L&WYmjT=wKbFfGU{;SkrHPtGURb18;%zKTP;3HG@GEuQJKwHxU-qS4mb$6hucxMCj~bb ztzw|tYVKH*D*Qw|l72?~lFADEpp*GsEN*ihKhIV}O`JtuH|hsI^tz6nP3*Zgs0EYe z=`!hR9)IQMvaH2=hOgYfHy78iG%u53{Ys?MVqzfc@m|XGTgELdmN6h=5nrEe!QCNN z{Fvpy?vEQWchobaf4+-F-MUlB%Ze_Wb-2$bhKT_o467NzRjtbK$NnX%r`6)s`15$! zs2n@cgu$BKsMu`)^(OS?aDmCVt8X*BxYG?82dd4 z%fE!T0Ip{5~875vkEF}+i_#O53kSi zV1%M6AN4Wiq?Be17Czx+BX;qr!vQ+A@@3q@UaWO+=T!Aul3&0FDg2r-0^Eha>)a0P zV&lp5=_^n=IvaYO*W=CNBe=5VFmgQiV8-P=ICAd*KKW(Cbj1K<4fhq;7d;T3eG6fy zUSnRY4rQj_A@K)vwCkX__6n*#)?lRYFCVd~3v+`5Sly`){m+~5X|^*@3cq76dv#dZ z)rcoLnDTDnE4(RH;gp+yFh+TX@O3wV7kdl8jc--Ba)Sw{ZO~@WOIP@wZIAX9CJ4(_ zX1CXk>8ha3QS&=c_kkO?P9lf6<-uy_Mr^fRg*?@LlB)U^>GIKQQq_faOzGW*H{Z5n znMG5+TcX7^R;pAfjfQ3CB&@Mqf%^$Tc#-XgLLX~vncjx#PF@UFYsra4+N|!aMePND zP}5GEwzAoMUBsD=sQK>NX|yYw_k;@rzdcvR)e!<+t6C($aSf0dAqb5kMH(m|9h7ZJEjzl zUGoG+izbe2crT5;{#6?F!-ZX&w&V3C0$X5-D`U)C(X*$Ez#e*!H-mrS-RK|iTzwCH zaT)vd3$e&+0op`O!}Y!4sG8})S-Sq*;4_qULrQQ<>oUH_ork_XSw2YOk##n_>)TOa z-t?v7#{QhIe#;)5UH*cM9c zS*lG4f09eg9^uUK&*-16$n2$yQL-@>8U~9I*=sP5Y?;aS9~Sa{tHn6y9}fH0Yp~?_ zSnNN&3@vBGq0^uHu&);W?SqfwdV{CrTQ^rKQOuFrO?U~@1GVTBVe}Z_}2I8}*>|S$9^24rKS|L9}(4PRRxg+6#Vb;mSmpMR@i)3|^Y4 zc;IG*X?N6sOC7Q5g*O+c`_n9WFwIiJ@nYss>?-Pw{DIvOkv$OZ4;?X5;LZGYkYUB7 zLcELmDeW=QN884I@zn7=eki`ce)q4ae%XxG21cB{vN6|aSa9V8 z%x*Q}u+QpLOxC2~B3IrLxEGzq2wau5E;zZ$9$&q>Atm2{W6VtGG}D0}!*a2(O(r~6 z@5HG|>eSKI=Y8S#C(x=A)&jR9!mt(v!Y|Oy$9;I!tPktMt@(ItGgAD1JX_4b{IW4H zJT(Q$0)tIIq8Z(tjD#OCdv5h=!>SYQdEmidX5Sx7&m!TYZEdkM+5L*txbBQ}ud60I zD(Z8Rg*p3vbmhXrj+71TMydaLwAi=~8io51;aq^!l?5=sk-4Iu^i$_#i62 zD95&l$JllE9js5Ck*+0tk3z;rd`AX$45YqM`+~n5XY?FqVrh= zP8jh_y1neWWK{WBnz=-YzmFL6V5B+iL&m_V_iRi#ABfuBBRR4*l#cpyx%pHR9=+3y zQBAced~q;9V83P+8}iooK(wlyBm6^dK)9YAeRjFgeY_V}#kE1yD_!LMyf3|;a0;P{ z$KmyC8_w?*0HAXMdArvbTGj?ab=V}-KMBM29obmKR2+J{9F8B&8M)7arS3M2&p3^S z3uiH9W--QhA5G0mQ~2tABzG6}V#BmSY;t=97i|711vk>c=E7>JbV)MX3{K*q#i=|z zUXOECEf_aln=@)3!T-TSRQJ6O`_X!=o6(Z7sSb>Fc#Ay&A7PpF1DC2>G2PRZIyw$C zx%XYNn{0!|4%YbX_Z|km6$ttG7^T~X^TE1t!bf+g@Ig8fN^J*W#olqK5xy*2KODmA zJBCnwp_nuHIkLQ+In%cpaM5Zbo_ept30{u)Ri}vqLv)~#ya@|BT)=wEvpBWw6}Yb) zr*~H&sbwB!b}qoyoWm%8-ICA7InaKuBhAl%otpX5^^^xq8e*|``UAFIe&M&V1Mm2>XId*K;iq>5{SJ@dXS<;cb#KCm zl@(|{T7l^`eX#1x0IWYh7-{}q!dLuIT3z?&!noDAcVz{(`Yl68-^o1Ld?t6rs3@$|{W8cU+;jc`M0ha{!PumaDF0Vh*tATY=!N93Z zwOqgo?RXx)^%5NiJwdzecTjpr73S-e;OD7^H!~xUZk2?RyV)2JZ3@*~6O43dinF3lMIV)%V6;$ z8d1GF(d(eVFjVfsR;O2B;rESLQn?2e%D1JX6Ru0ce2z%pH(rty?^j4`k0`*e#2(%H zcp_Uss*0=D5`-x>msn&yg8fgN>7Jpr6Ad_}D$e^pUGk zvr_mG@3s?mvFoAblZ~a1Qc)!4sg;wuaGXOgT9+P%zwqA>c&G$BWv8UCUoS|{Z=aCT z>;*P=D=k)gD)Gjh7ASqLhv=i4u)JrBZVRpPrnd`*ZB9eJaymA=+l9>iXE00n;yb-Y z25wIje&ko85=idA%zUe-L&flZxGI=TODwokYxCYIq z*5h$lBTl|z&QCK3+E*x~J096at;pD}2xN6oB7ls65q~iz-P;lkZGi_Lt??AI< zy1Z|u6Wd!{%w0R-pE*Bahp(etuqlRuxL$!`RT({fm7X{iW9oB zd6p9&-hGQYpU3E=eHHbY!|1xPAEy=ia;@7p=vRG)v-lov7@3bKv*XAeauU&(*CPDR zGQ1r%5$`t4KY}V@`>h^%r=Z|(SeudwxY+r;T-%~VANEQJ#1&O zP5Dd?ez2H*n~Wg`&SSZ13{`}$>0x7qU){EGv^XYwQQuAAjL;bNdl$}*{TH!=MGU8E zbmpkpBiL2=l2&WI0{JT0*fL@pdUbA2{d3MNd1b{rJ+lM`>~`Mq+{Gz|lVP|s0y7d~ zab@5-%>8`?wqpIJ={1Yn)#mVqK_J&U7~;@p;!eIlq8Ep7cJ&x`(;Ug0$3Efi5gn@Z zHX}~9Vp^RqG1HUUo8rWLGl7S7qv&R`o(6SU)L6WUN7huJ!?sF{Hx~16u~z7HVJ2Sm zPsHTFr&65DMajLhK0>x?ap1S6R1@!Hf8}XOW6ViuW5h;j+H4JUF4utOxo=X7$?v2) zJr7IAGPg@B?$3qy<8Zv~n~7D^N5XGqByxHtV#PJFu2MUPCf84+%dTru+2&Ho#^I8r z*&`P}Hg7~)?pEmQw?Txx100oFV_~iW40G$HCpo{QGv*0sX&H{spJronNG$rUUx?kG z7h{3$Rk-@yM)|>OXj=M0+IF!8l(PFax_o?ry{Q8GK~E9$l(o@T%u%$4#7hej zV|gL2W6EoAdyo>IRvN!o^=O`nnnyZzbY& z$3%futALE|x=7uq4~wxL2sdy?hNnAF5#i@)Hrq{(W5 zFE^@K=*9-@g}?m?)ORd~#k+KAqjW&pV4E$4$x1OY>mp*4&!H?Z5tFJ`V^eZ6X4OZ- zssB=h4qk!!Glr5~A6u#CH!Epm&mFMpE$~?ak3e==9NH>L$CSq#P!!pivL~vH($wRh z^kKa5J%ClZJ~a6uu*lwY#m?%<`0+G^dutZ3l|?jtJ{hrfjWLtNx~a=i1!gZ3^Un^C zq3o%O^yBxXxwXe6pZg=R-fAkE^-REJak&0|fE9m@ap0?y;nbBy^NZnpK725mZ8|RI zwLF0*YHiKET`YOKUvruWd{+G}QyHx@k9L;rg^sivifqTizZxE?YJwS6bysc!(v@3o?wrwN5gknzv0&H?z8$Sj z$KIOklhuT6-)2asIu%OOH+_|!Uks(6{d8((Phfdw1BwEEK~Lo;3Kl3a+((UnhU(I( z_6=^Gsz%95K&{&Qna;Va^{EKUCp^^ky8gV?4(lnnYWVLA)a7EaRtHv$MGsyB^qy8AT_t za**&R{JsyzT^P?#jb}0=GM@GNOL<~W0>k62_Vw~5SdLLY9qu81%Bdw^VXU5N~u3+ZtA{ah9i$3>P zv0_vXkKW6r-+)foJ)kS$d{6IhTOpNTAkEx6I zAz1AMw7Z@`PLI=w$xq@SSqj&5&E)*%$6(sI1g+xBFt~>iZp{}C4L+*i+w>52>@t?w ztpdgO@V#^`UK>w@2WH)~o#|^U-a}4zM#gVIe9u!*JW~PJy~?!dphq(=9lm?u!Dgww zse+DtwK)i!(dw`yv zFOhg!U`Op)0L=~KVVK$nNu#1reJBN2Q&KRr^*fZetb2%N(-(!);8OuZh8NO*)G_uyP$C7l*&%g(xmvoiV?K_$34Me`JdSQu<~l!pju(F8 zS6KIxhEcS>l*oJIqB!oO2UV*^ag*Uh?rXM`0V`KA_h%*>tiljkm4Zj% zt8jLHOC0;=38^FiP`y+8J~HN*U52p(1< z{_o#nQ~6lhAByA2$_x&f*MjE7F4TR}l?ol2QYBO3iVzo8S;gaGMj|RhHy}Io5L%TV zMzZc6#0lNuhw{0s*croaA3mY)>}I^EY{1jqX9#_292SWE5$z4mXeaB9+((_!{a`o? z9T)R#>uDTmw3@l;Sv=NvEmt^%(kD5BKHfrqZ>z%fE_y6BH{e;NE{y3glym3zW5;1B zXzm_^y4}f`P&l9WIwrAuW(xPEPN%Prz(;nD<*q%RJT=pcb9zXeeOTPfXBYmt-igM4 zT2i?4XLQG|R6c0T@Pp2@D{jx*#u`ZJZG$SaUbsFmlD=;f7#*HMr7g3$C^~@&B`f*! z@i*iN-SI*{6PllL!}Nqskk-3n<_s4GMvL|Bjxo$%?$7=YM{!-)WPUj8E!{2dD(N4F zw5iEq^gC1lgk|CCrVn`TT7#53pYd*x9eXHA)Y{?4)%D$YB5w=>|4ioPIsLFPts~Z7 z=?C{MhV&iMmLC&b_u!tdr6nOu8NdgRa` z1>Wq5klw%XEK-LX+t@N-jXq=a#5enm@c-SXAD-2B$D$rSaP9nFTB24Z&Dj1!8eedM zPsiQmh-Q_%WORVj5{r2(=n~&911(EBai73wJ9qdwzr20Re8=bPZ95-t#F${T`6O=7 zspeAM_xyIPj!p84nXuy`8y6g9ZQqZq%CBa@&NmE=GLR;0)|YJ8>PZzHeE7-HpP3&V zS#dKSs;zSHdh;&ioe$@OVX^GmK8?Su@6)QGngQAmcxlHeZhmr09MCIa_tRDUQ}LDi zwtr^P<=@ivk|yZ0!UW1My|JJq37ad9pyf16c6zQyL;LsGd+wJM@Bd6{sZ}lMt#jgl z#%($Ei4&XO$mHRKolNS#pGB2hm|%IDt43U*XP_VZ^$X#gx;UQh*Ph9B1DH5+FqMCM zux|EXPG2*Gt>%70qw+8K-r*LGSKj3Ma}POpO*wDOu!EcSXsA!$griN;kuq%+!c)Y! zs8ff7F||k#_}z84HuFg9?cC5|6^+KcVVM0lmPEbg*Oj|jH>rTl)|{YsdC1?f!MRg9 zJRfbOd9Ncho1aVL2Tpw2UE&7oj-0$EkTIbXdFjh2Zhx_szKLr&GbdJjLsIz1eg)4r z5@UR?K~%gpmr+}y2-lCOuTW#(Cy25_9?P^OmNlHMMFC7C9Emn4r{%)0f2 z+qK^az2QObdUBcX*IZ@foM!yt;7r+a2m1S6KtbhYtZ+MomocTBA6>~n!T&OJD`z%8 zL6a7T*>-HM^g801q^n;mX-1~<@u>BTIli2)#okH6fT>&@Ce|WOH}KlwA>M{6b7W^* zfyL^{6rdNRt5<4f*?l0l(!c;eO03>7HhhRG&~T1v@Ii zE2dIXzWxRS_P@e#@ox5&8zSk22AZad`DQ@|Qj62kylfR5mt?`DPd0XL+m0vQ>+r-# zu9Hz-FyiWg)|9PnPL(p5q;%@C6j*#kG7lWV7ZZJXTgQj{)@{WG zpM3(iWGhTPt0f!VdTITl2a@;qV<;YR4zAsbF>8b>mt0Y&f0ZWtEl}W0;~!|3@fK5a z>ZLF4+VHTo!=dGBbh)U&pBEdmAx51llbdpWx`DuAGC)>r3v91$hNnsj{IW!uxz|5I zOXZ=!a}fA84_{*BS{3$4R^x?3jj3Kf0aq^hVO#tNjE~Xe;~E1_+HAXHUY!G&)#Mj$n$;k0zC1X(FgMb z4G%39Hd?1g%N8cQc;Y1L`W;2x(F<7np(`SyJ<#N$FH}~wM2|JzxK`CndU! zX)G*6YOr#`4_vQnK&aItcpe@AjoH3XOdNx-7TQdBXi3#E_8jA+$t5#&d8LOAgYrRd zo&39Gh&)h{r}e`}7;yf2%>V@jaa4 z9tKNui#o%JsypAH}G&x`4NNm+<9kWA2;ZSYT>2<6kj*RvOZM zg1|hfG(-P6?NR)v55nueBj?X&1Y4;oOt;Jrrr{MNSx ztuI^g)$R?rthfzbhMz!ns3jGPY?$oqLT9g^P#oWgC!ZH7-m7O(l%`Ckh?e?qF{z)xB$*k) z^L{fpRyvPq0{5o+%PAy^`S$usEjH`bgl`9EF=4Sj$B5mIyAI7b_e(3rD2n~UvIeYO z_zP|o3j85+=LZdU-W%47N@wEm{o`6_FW-(EYIAA!eU6xuC9#X2BiglghF?%C>^iPM zYflYN$?%DtwCk2xtL$Kq5VO3PH6QJ zm-N44t{4l4FRnz^t#>%9`Wmj9I^5)F%Wf~*aiH%>$wGaT6um4)8o1PjyXSZ0aLfMu zs2`8=Pb=VJy9WEGsxY$YFU(x<9eeL+a$bNATP@R|QkV7|xIkikbT@i>NZ6qY%*k(w zv_optStr(8Gu!cXiXJv?H-cHHIlkn6LP_m6w6pzzL0e8pCy$?$KB^p*t_h5>jL}XU z=-7st={GQGNhM^RU&4D^GuHhuW_)E^HcJw7?E-g(nflVarV;NwY)qXLT{?z7#G&Ue zalz#Sev9iVZkD)ha%;9+Xu;?VV}5L5#eNqJ*wDEd_xf7#;e>chtlNgeosVJFjdSQW z>khOZy+MnLo;>2~!;_B&^Vbmte!Kr0s#`t^y-ZJmzd4Mnn~q>AELxti80y%yxBj9({g5Wz2XBnO?iiot3RXf*``=prGVb46;g&*Ix3%TLF2R% zs1$f{vaMK$+74ud>S(bZ3ljLi7H@kc~h*^dgaW(F`(BY`CYrG*h zRqONA$kW){={5?CuEJzVEJm(R#L@gsc+#m}dfwVZ*oZYoL9wuZs!`>h^{Py(l`!bK z5A+_#z+|93g_{{BJolxM#|=E3atTfkj$-VxSg6lR!KA-T_<_+wUSN=dCcoM><+$})%nz+W!Kx-4^GA(m%2c@^QJtO* zKTvtwqsSC#Lnk_LG7h3b90(=viWV8pb9U8#r^yVfZ>W@ z?47a*U6$DKO|31jU$ti+(>B8Yv=f(h=uEqy(b&^8S?s%+Jk5$hecJGWjV0SD4Q2PVk(3S&X8DTkh>TB#`S~4Kd+ZV>o_da&@@l+$ zStDJyYK#;eO*~hqz)gpzSiS!VT&8MpOh+Y}TQ}yF{%!egk_TS}dvJ@e1@dt}h=oC$ z;NsJQ?jcTMJti?nPm8%;0$=XFEmN-e;rZnLIBk^x*&#Q$zn+R00$*^9(9>z92>qXu z8y|eP=emv(lS+N~)u;|ve>P_GC7RUkAnbsOpTSL4k$r}?A3kPS=t8+oJ-$yC`o}~Crp5bWv+gV?M$SN3m=))3Y{PM0?Ra?7Pq?2{ zXKP8D?Y}B;QnCskLXX+WI=q)+&DJ}OxX+@qu(T+W-E<75z+8GnBSqR)$bIv?x%QX#0aQ|un>Si6p#mzpvx6+fNS9f8U_7TLVpGR%r zWoR7GrE75$v45|`cNc8gE83bBHm+RM&k&9LTjSDWV+{1LW90!$noTq2qDw!d8R?DS zo%3Dt(0hmc6P57kd>6AXHbuRSDpDi0pzocE4hMz4aK=uososGjY#YTJ?gVLN<{EGY?F6ZetM0&`r4wF}GyeyR(9y!e7P z?;3EqjlNzcH0p)5p>5>_{%&P>WT+Z=V=pzOg1gX> z3vAJtTYglC^5nzVR%{s3jx&yQX4&(0TvpV9MRPo8{-O-0{c17qRy{i1sFM0!*Td&@ zU91?Cjv>WK2x%UJ2p@fZ+-gF-@1~sJ>jcz0RbsQ^4Sea+9qvb*QS!nMiKf%J?_CHF zD$Qf_)9pBXm50h%kCQBaAT$t@8<`uu1bGOPYtW*ny&Cd`Fo@`Ga zS!)LFCY8#RxhFw`v2Gev-FXd#&F|pBu)A;yH^TMwPSD75LhHr7xM^^I_RS5X>wXV> zzR(M9;|5_u_j|nk&}21?QX1ghzM+o~vhIWKKG)#_mGFo?QpL)&8d~<+n8hU zV^4f+*%7IYdgGUY9tPVvBh;`5;^+3pgV7$C@Tdds9=M6E=Edl@QMt#6`cR~4G-s_^)cxzde*2&v1_P12#rE6{rV z5IZ-#f_F@Ryz&|d6QyZr9k~ME`>w;E&)0BUYpS%rL#pJp>6G+&+YT5nSSj{sw?eaD zGFJPpK}4f0WX)}mbk}LaOhpOjr|-npVFj4o_Z*t*-IjJQIVQQbzanXxd!XXuM7-9F zz~lbAVN`uo*pb~pV8(fLQhA4>HT95VYD&G!p+E6Hjz7vm$Y>&Naq*{+Au*h|U@#L?P$|7y-2+m=1y8zhw|NA7Ft%<*%bxovSa9)5_&vH>fQm$w-G4H8jUBXElwf0kB1 zR)g!E-$68?@H*bvs7Z16GQy1@`Q2PVsYbWBn!BQ-FxE_y^gni(;xA^w59z%CH z!q2!PifscC^z|GTn?J|4@#1{r`$I5IzJTQ2&*8>5LKoAB)%%35$yXm%E!}amu@h!S ziIapoCgaY8X;}8E8Yg`kF>I*{7kVo*!O4`)LR91e5uOs6`G728YgKFPLf{SJtFzoTt_#bDx`dEK={-+v|PRn z$%&hBxR)Wet`&Q##Ur@%;iF`!X8zx_*-Nm)i-Z2Rn zX&()_Nch?nxqfMy(;WDW)7&7?p?hmJ)3z) z`gGL?=ahS(^q~jdovoIRUAixgxZDVjRWHC>tm7g~WZ0e996uXOa9XbgOtVg++wo#J zKD~x>OLk(Z{eBpCI*Jn|IXE^tAO6+5p%S|avF9?7G+3-p5*sn6NRdf)0%zRu2daJF zqThv|*!s%@sw*8(8rT+R+9#l^S1O)<+K8{dN$6>jit38xm^Sn+y0-g(H{0v+U}-EW zgiS+H_6+DuS7m~jALe^#(d6kUY5LHy(iz)eX{)epnAX!2sb?MFyZaX&c`EaxvpU_A zyJoDvhmMt@M(hpTH!Z}6vJ=?jRD|{7o82MM z0xingA=^WnWk$Lj^+$u%vo52~QNC~g-CLDK!y6D0F$$I! zdPA|et2mqBilp65v9+v1`f#EJx1P7*A9p7<84!xp)}dI~unG&*t1;?qEuKbH!PxNy zW){6h@v2&!@y*4<&>T$PdHF_v5?Egjz z+t!PBPxjli>>+v?|3Hq7z#{m~VYQ4{@Iqb!d5X3l+U%s3~gGcCS)FsVs zK7Lw@w8(piU;G?zm)*uQ)?mXD6se->z3EsHp!;fpxD|0g% zj8(16&yjgD1g9|H_xnaKv6BS!BbiKfaFX>KG+YYo_l89b|qLAPcie0Pb z@@7aBZ*@pu$DWRG2yKU(`&&V!TG->nsc}Ndd$d^ChkFl=;-Ts8!%>y}S`Uuu7Hs=ppYfiDSV@Lf%c;a~kiy`#!-$rAQUldlRCF9KXwYb)E3!>*< zfV#dj67$W(JVU~srH%MBUf3|1m~!(bD|+9uqmuwSyBXU?I;1B_>kRCqqT#LC_(&VZ zL>kh?;f1v8kpf~;v|zfo8{^V_`1+SOr)C|+oz4}A`A~;dD+7=_a3NA>rl4n^Fl6_S zLj3$Fe7`=4w}WF?Y8=Jf4MR~iIT%wm4#bZMy7>KD1rG}B#QJ@+C@;m5e(&+=u?mk_ zG~$Jmx=b0r50}GFA^z%C_zcNI!R$luTeuaSwLVJ2^VUeZW7bJ_8*4Gx;v?QgXws#i zF}wZLW2vnn=QVA_Ufq-#5~fb;U>)3RqJh3mn_=e06S!Bq8*6u7#8g)u^jNBd*8 zsKbd54M;IYMeqck96txOKu63{?}*F@#r2kedn!gstk{TxX zG{TxzW(a+*!S*>iEX`G=dD(7EzqTA>jMrn}>EjqtTLFth4{)8QrI+SUr2coSq{d?m zdEtl&dkJjuPX(sTxo=G$H;EM;TH<__Dc(1CMsn2v+&Slkh+=!x9yy4gX6IqDtPCv@ z%8*<75W6~k5NG6Sp>$M<%LDcJPMq;d9&1K@mlpI*z9YU_FR|_XE5xm|hu)D6xY?u! z8rOcotOzx}ZKOo&AWOC_vZHxkOGnZ#okNYTjUE}MB$*ueDe4+3U!P*T6`;(9T)#2fp@Sa_aVA3 zxewp!{+ux@nED$A)3?7Sb5o3XX{I?Xdh4_Bk}bV=deXRCDb8KGg*@>cDe7R%VY98+ zI=C6tiZbwN)(Th@haxj0mX;fYjZ^d@jy?MT&Pmsic1h@eW;J51Ml-spi#evcBkP^W zWBpsRbdwHuFVW_dA~hPVp2gj*qd2T;F&Bkg!{*Ny;P1Xq*nfbNo03PLTX4)qS5B?& z!XUfu47u2!Uxtt75w~CxwjAu=n#q2hsk^ud7Zzyn#o1=;Q~3ce8;g0-UIpgrII(yI zIQ~utTJDwD*02Nn>a=3^X|kcKH*c(U=g#9ZY3ntcw+1g}`<3oe~4 zxx7sW>NV@kpGM~}^wmWK=Iz7H*yh~mV@{hVwv?a0?_^$a_y5oNd)JQbi2vjHdvWSt zgq&hQ-XbBV;(y{OxkmtR1qFF5uh;D&s{YLvi+q%z8K*@oNFIBO{1bV7kw}YJo|jXO zNIwy=oD_r%%YS)APC-|s6^j2K$$1szB*I@g#UdfvM2Yw>{o5zGEP4HZef+PDx18Vm zpML(=w`OPl;gy%mWyr5D5AycK-Xc+~px`3Ph<#N5fMOwuF zU3b6WSu5yondq0CM_%{D``RCK;-{>hsAi-%1Pb_c~Cy{PkXs6Im!Dam-m;~Tj8I3miI-J4}bYZzS#SpXX7nmdEI|<{>my= zP}m{b{%!Mb-QVXe(o02ou^<;g@;355jUtW=q5_x%A-q-&< z50RG3eh#B%!gIm+e8%l-xt7YmXHd2A`JFYnj? z`Y+P|>!-Y2P_evB1UXNRw@3(*$A6!fD2o#5VnP4)U0yH0=3f$Nd7ZbQ96@p(dHO&4 zdtFf`=k*pO55?a9kN&>jV$rVXpXf$kl_hxp zyN|zoqE23(BSl>*a6(_ zJd4lgyAr3?cpF!udZxEZtf%_}+J>La9jEPutjVaC?NnddNLIBv?UdOLK_|N^wdro-| z_;ptD`u116a6rnV&^{1|J~Ph7ji02FFEd|$hnXce*yaP zMxycE#opGK6Yg~O5Bz2vj<)%#&E*@DUmR@x0pxqHLGG>w-j8P;z}PF8x#CU1{tR>a z_A6j_@SyHuk1wtANi1W{mOrtDHTu@LN7?%vRZGl!DT%emzxYZWv5^b)OsSdp#U*F& zItQ`37qrrMugJAakJYns+TeOyHO}|m>?_WDZNGuvK<$WmcE}(H0@$lGciMBooZRLf z#LOS)8@Fd*R}J|qzULai-;vx&3^)lY2XA5wM?Do=2%Oj_3{1^1oiZo=m0f~@X?t##7opk z4oqO&UUJS}`h&RQ_vJ`vpn3!f6!&P(U`@-VYd2!@FPurP^%K8?l3cNuotNjF1l?S$ z9n?*bPcU=kh^(2h{Mb2pE!Rw}bx->27bm{`8LcGoom2OPu6p^{q6wtV?75JUd=XpD z`6Cx^1_^e#fI6dbdUkZ>$ajIC8}$yf{e5%8h{bvR+=KZ8et_B+n%W-n1GHT4ll_&R zi*Nq3)XoOzKh6wxw7z<)-5CUS#Xsfk>t}9&8EAUQk^2O(L9S*KH$Ok=#P)nuOCP`Y zwj(qRk zGugZ0S3Tw%G4JUL<%VKZF+p(tNa~vhCJ`&%(Kr^Vq4^m zSFIY9#P0pam1{{mA@13#3KxR@Rzsy?A^|+k-J7s zf9oq3TjN*U9hBTVpldHepYIFz8JFu$f%YA9?-$VHqsH84*sJ)?+dQn2nnPZxy9jiy zXHhY7at84*hg}-mUP|p8Tc3T>%&Kd--glp|wJ*07uRGZNC-DK4&%2LYeT~%|^k=)_ ziymu+-&f|1sh@TbQxXJx<|g_L{zQ-b=-;tR&{ZFJ*w!@o-oIzKfJt26-4c5Q9CS&n zzZ3EqMFRYTjxTSh>7$%J+BGXz-njMhMdbqf6P#Q3$sWqpopSd_TQ@)e>xF%d^+03$ zz1{KaIpJUIvHP8we*?!qbG?9Hfj!J6b=t<~?xBtiXNXk8Z@K$hOKfa3w7&V0*nAv& zCGa!a-Q3eZv!5NTk>k!0OSjKLPO03doll*;?e_-iOtE*l;gfT>S8{6M4rdQ{8OC$R zsX4xVt+(eDp8LxWVlQZWY2UF2`t7#n%Krsh$y_<>Y8HNM-k;Gu){O7TcY#0A*4f{= z*Bdjy#Bcb4{sCXk3tCQq5@K`1edXJ;-Us6Lw2$8d=l#X1emz$o`weXSdzKMve&w<2 zIZ^LUCAs+4J%|OqGuw_^=d29k=mF0jN&E%d9SbOHBGyd%-FNPS9y=$;`iC*l@!ioL z?yL4*oHs=h1J?YBxcPy9LCbe%bA?!at*PDcwFA!2@aK?`|J$$2Z}`?V zP|p&xbgVhCwcqd7DbI7oo{8>BLJ4eOPw)BN8jxSs{o}Ro@x8MtN5tAsZ6D+Apl`mR z8``-($DhE!uNb_O!+Jg|^K}liw!H>OV4v!Vl4IOmmB_{K=mPnP?R`DItog>wO<g&*xyg1EqS?f98UH#EB<%_05$}^vPRu-Mbqb z_;nUy6~}KPc5mX=4EzAj2z)u}3~Yg1Z!2%?oaOLGu09?8IXC12CG%s>>G|fa;1241 z+>tqJ?Bjd>E+=C+Pe`O=Ge*Eio6M)+I>0i*~jOl?enwd#B*(cYQQ;J){0vE~H1(@UN;C)TzkHS$}c)L^KQ7jiz|6YsIk>K)Bo zpLc3&yV_o8@56V|KJ~n?)pFb?i4W`t@$=5AC-ESz_<3wBfOlTKO(5Q*+?l&Sy5S4w zBsxImn$r;NCXm{N9`+fz`i$6IW*PGg6K$Wswno2+zz5d2;{_NK35>b3&!gw6_4hGr zjhTOdz)xU(?Y;ep79GUaJr>8EYai~>9V<@Gcmt*U7hs^MbhZ8mLXjaXzY5<*c01$pvT75I|Am6b0))WtY|FZhYJu{u_^9!I|f;MKa0HyJJf**H3!uszw3{X6LdjxQof$p)+ zZd|+iV%ML!>E+!G;?A#~+UkpX@+%I<+M2(2ur9&T+Wy(g`HI| zb1!UV#5%A2m5AlfTr(TKHTE&q4lti*Bj(-6@%#D}yaD+g7;j(zeRXJ4Bb<@o>`tV9 zA1w~ETdumP-uxcRmNS57oWufjeB(;pL+;ts8lRwe(J(fbn6|ZwJfS#q+9-th=Jt*m*XGU-wJgTnKv3Z0mPAe5LLt)-FvAio0WpwqFMq>SNna zE-`<9u*P2Y7y88^=3WorndH<=IBNSfZSmteb7C9!?=7rxuKBTH{Crw-`p49Ta0fT;6nGb&^A610+dH>s3u5#zCa%O2E%eO= zT3A~%V{806hZ@*Rk)wb&$;LRPjj9j_uLrh&T{^KIm9@G@)Ge5`^_l%y^ zna(@AtBqTq#J!WY^PO17e*ok5P~>?RZO8xdtHCwf?}OTnz7t*v`c*#Kp4JU@0Q3Yw zKG8~@K_ft-z>(tMk01ihtpDvRRQ#Z~=kd?n-ev#9x#PQc7|Xo_dII~7Xzl;}y3dCH zFR_i=tD%*e$Nk&q^r}Rzp2fM*i~KwA&U$>*1$GI#&WWDDIrYAvojEKh=nN%iUbf?D5`AXYWl`ZLz^tr)g50+h~=ed?UVuhf{opeOke z%;By&`e)+rz?i)oI>G<(l|1K|*!Ap5Df$E6J8;i|-v!(fZ1)OKT=mA~%i?wX&wTr| z0K|P?-PNsh67CE1UhJpT+j(E;>(tlfU4?g5;CK{s-W z`JR9~3gYG_{x6$jTwjjzXX4(`75MK_?P-sZ>wkMw6`-k&ft)Jur4B);U?Y|dsw>NZ2@HJPhyNX}HE2#P8 zJ%X4$Uf>UtbV?-N1cI~aUoY^z}xm4{GaS2i06(EeD7Pn zdCywU6s>>Phi^=&p3ck_x6e3;)w}kAKk-j(92>5)`X}8K{1!{s-D3M3ymxo1p6JX~ zYiwXUV*qv4qb~B5{O$Kl#^QPw`~Kb6^*Qp6U1uh*F6=V~vFd9M41MLcAlEtW>~79> zFKzjXIeUEOR!tHH<_o)5)f(S9IZYtdYnjho?X8?#%?Ww~>+bBT-SxMA^~||oOsR9n zdY319B;{%_u>(BlRsGwNQ&Y3atL3!=Ol!O^dp1xtoV8v9dJ63 zdnaG@@;-?@+@W^Vb`A)6v3eYieWo#>T|DdnT~K%8J$HQfd;0L?E0LdSe#8bX0CMid zI&q|C*;B3(3;k$va+KO97WeGt%$kd~Z`t(v6X$*Lj*R1<`HA1ezBl^_<6#^*e0jCk z)^dE7-|t%a=8es0CH364?HNx0S88T1{y_VE(v86x8^!`uj#%vh_x*+qo}ZJ%`o^;t zdz1uUbDd|RAM_LTzHmbJULTQgV)m0~pW4s<*s)>$^Uj^PdFzea|HRbpb#{C>gP1U`L;~X06_*~y z&Ws+k2aQ<8oxgW5mf&!P_OAEL@~o-)hHoAiQ?v)LW(3;4Ac>{*#KSn+Sb*XggJ*5S zE{!Sg^uS$n?2|YfftD{OkZ%vAzTa*bxp4a7Egtb13xN*&4hHaUt*^bz0sD3}awVr! zPxib!Csq&Z5?is49D6Hu9{kG3ntje#yC)pb)bQ-(J`tbu{P0fS>OuJM)m5WL zX3z)R>1#an|A?In&K$--Z0i!7J3G4`-yToF z*1u3!yub%dD7k^E`h5m#oaOn{r?)*uKnB_=4`_ep*jK-1Q7^B!*fUnmkOKmg1YdzM zYtM|cclBqE`!wKg!uy%j2&KpBQFg9gr{5rF-}-DJ7ufO}_Bn%j|!tXiP$mX4O`fIsdV@>Da4H5 z;W^(uUa{?KkG}Kj9&mu1@;Se;SL?Km%R!u5c9#2`t>+nAu9|@s6TA^0vFZZf&#V^^ z#2Wso`+=Cf<+yVJJ7c|vwa4B?p7Jj5wm8NnkmnhlYk#HA5o@OTBQ;AWR-{wY6I&?y zKkgmv7b`ZCIJUj)At%uTgf-v-DhDrtv1uOL^Ed1|6Sej#$;D~p?b%}Si^nX*`s#68 zeFlC#7x{?ynYsHI!0sS}I%EZr!1l9PiA$b^}9AOmyta*5>3*wC};4xH~4k^TRwCz*t7B zcc{*QZLM+lkL}l4u$w@RueQ!PCD>J$X!?xpx9y)?`NYhJ*m-rs_e}0>4gVXORc=3os+Td zJh7dxwvTfLwlkC&>ulDa^UPeeSL?_Js2zy^nV9(iY9)y2pZn2v|89(3>obmBAH>?` z9c}MUy!PJtM$CQW6c68+J5=uW+a2t2p84g}5zAaVm`~s{(H^;Gx{rPr#17w`?PE?P z_CPCD8yFyR{k{1l-!|;gfO;Ms_~31G)b^}r+z@Qv zIU_&=^Dlc>zbR^?|N? z#2z5QD`%B=;+>f9hVVTZ$A(k)jkZr@TY~KjbM`Y{IdsLy%Xdcc;3^|Opf9%(Yq9F( zT4${KVV-<(^<&kUPhj4hehK>OIRX!KAn(b+H-7Tw{M=SsC-24c;|EacdoovE-|r{y zZU*N8_Hy^s&OHxZIrQ(ou6NSjW9J)U-8XvOfpr};V)m-%fp2Vr0nT3bO`r~7+#QXn z>wU%k0Xvc&iTKdgD-XG#X5jlfw!PftZ@&A8TYLGk&&YYsxs6p{YP=io^o}0D zndaT2Q|}(3?K$BoulSK5aIoFuCpp9vYvkz58?k-{%5$Foo;!(Mu}hM#=)pOvHEis=ZIMEOHdcY-_X84nXm=YWA*eIIM?`qYi$D)zxKlJ#0UNx zx}y)rJv;fAoey&Q16XU^KR34>Gv9!H68%7r9h3XI{dyMI@_$8Nf!q^!#>G53jH4&8 zx0t|M<08~NGkd&)hA&2;Fh@Me+xO_| z3onRU=b4V2--G2lccP0QFtpKrmf71m=KNfF1*bNNDel=qYxQ}7BYgh;E-|pZy9Vsr z!N|3bv+u;~o{1^W@Uz&O3<5XmfO|P>5Su`;PlspC1IT%mi!HD2lvrp}JF(>?&~}EO zW9s4%3+x_nz}UK>>z>qtUIlE=YQGKw-`(vg&pKmrOXs>zYwo+Q5era)9>9H$jG+F< z)$ej!eaubb?rN>i$ez{?wEXIceZC&mi;QPE1|LpbHF?-mn{@#o}4Z9PwR~PQ}W2^YYw`T*^)g3XXIDt>k zitk+2X(Jha0chiaE*tIH6#LAC20DnXo3ibRp0Vy|U>gs0tlc`c&_9y7IcRGGsFjiH zJ**n+0F{fjdKkm*>=>M{;g{gMcb$=dySG@odk;@uk|VCzdjR{LeAVIpwbaaj1{$zd zXrEaFyL|4`cc$a}ZWuFHyj(Ng$#>wqdp^4<2e?b$2z)s|fHX#L^8tKcub{<>kF5`f zOmS7tF2pC>SZ)xvPxa8w6;HJBlH9waE6*-3w{`C?>=DE}P@5MC^v~I!&*z*YKKSzP z^tpigUToCC3pv0Z6YX84K5)Kz=BgX$Z`7Z8jhte>0r#x;L5;|}VjGXZ&wa)lNidBG z_kd%6dubn$BUjGDc(&2yJDhBhaHT(c%WX=C`F-C8+4)@Q~W z_WjGsTlWT>8|bdyaE@ng=sMeN^|8kI1B5wtndnB|KKifZd_HRXo*MJl?HfycW7<{w zgnWdihafctqfM?Vm+wbHQ>oYNc`%m->Iqm9;Upe;GcW-Nq$t{j5R-Yht z+4aWcs5^KNuinJ`o|QMSJ80l>LvOJra@ESQZU80G`d5&-Y<+7K`FH#VCic;<__aT0 zQ!9}lSb^0y;66U_t9Pt(XxIHv+l6;R&qmx?#_W5&;e!l(P&3uJ@(sD18T7D5f1rgu z51KxlsouGo7p=s)Bz|FDU$`-t_tPxro>L#!=+;uu5g`#vRF zyKilcvC3`kFqdm?L*y&!%oRWPSNCI#y%n|R=>mJ?f;(upSbd6vJ;4A<%^)V^mc;5=XG?6j1@rhJVm+g8 z>~`uV@mG9x#pHpK;A<`nwDs;`Uarr>_u}nSzitl5k(+8IS6qJvw)>&w`Lox4?&0^X zxpQCkZ?}~naIFn6fIMrOv0VSHbVr+4GM25c?Tn6orC;WPJv>Lle?gnC-Q*IhJ0U%Y zE9SLN+&hnj=q61KO|fypw_L9b9%D zjFQ^tlkHw|A=g zaUb{-+kKqfV#V*+zNf~`+b_`ib+6b1^k`M{yeC}6{|!|2TyE7y;3Pm--L{QeUwYd< zg9+rHXIAs@)pB3=P7Tp7Bo0XMxo3+tGu8XOXo|id=A64Tl_Ui^7_s`jkgJ}=tUKom zY;)7Nh}Em~w(TX%_kasd-k7-#D&~ILLdiAPy1*6_BxqubJp(8`7f;*RJNtbDRYPp% zo~P#A1Kon>d|UDJyyf{`sQvu6R|m#gf>`xf&*xNknfRO|&_f$2)+_qvgm-tfUjQW+ z&w8bvzjo|iyRS9Y)(+$*u%G-SFL=9IHXB#uBv zu6PCxw&LE&%;Ig|9hIJZ3F^QE_EhBl#>THZM)~(%B78t;qnZ$q`W9F+T`}kQ__egGl5ORQWw6P1^hSqNR0sD4f z+`js9jcJdqUx)?vnJ@p;SSL@kjjoPWW2|;IWNgRf41r%j zX6zYgdpYmaRS#o3XK%%RWuv>GCSvL8q5b1RLuQia9tQBUJn`#X=m&m%U&524*mG>u zF0uG$#@R98*sI;B1-`HHCo#G1BIoRKq5r^dJBGE+%l>jpa`Am`3A*qs#H$zF+TA!Y zds+Jp{|bJ(5yJOOKZA*V>VMONxOTddh#AIuGVjM&c(t?x6CAKSm+TUY*%*kIrr`;3P4 z_^7Kh;h!1mgX*97^wMtF1?)Qar*qlUx(hMC=e0|6+10}UzA}O@K>qK(62>;xft(4< zg}sRDj<@AmANe-u(^UR;%8Qt5y6TE%F)(1@}v5{!xoX_vik)%P}x)ZCsIzC|U z2eB)VFK3{Y;>Xt8HzwSp^2AGG>#aPS;w-tvCwAiI6nkVa6Z8%3^2t4C5IbkVddj)O z>YFQUP1lZ9Up=<%J=R^2_q@5$6m0v{e?xOy-!HH5l|a7{^XIU87S}n%&DqPHPke4) zzQ1E!K@w|l&)M$Vw${0R&s$&b5x)DECpn=#i~;p$yITX)Z?~1JeYaibBPZwdYV-*3PON7u@fi*LN!(bX zm633Pu?zW!HL>a&wr6?@wp@FAH`Y#lPdg2==_|kEzZ%zO zUh%QzIyZn)T0ho(_`8F9+b?qM<4#T77H^QN8T54*`EtWNXxP8|Zr`1Gd}A4l`%cXM zZ)&9OiC#T$sPDkvhkgdXf=uFH*c*L8-@%o*{5m6Pdrb0j>K@2dKm3wd_vgNHyi553 zN|X1QyO%Spy`r_%;|7R*kCsz{U9-q54|K-r8R*0}pSkZsO^tmTwtT-&%O^C58P~7# zaVGs{A9S5to1V1?DRVyocQ>E#UeMmh<+l2}i#-phs&m_C}|hJ%(?(A7p$}18guq=XXTZc7_s*AZ2B)CbG^fH+kY2D zopL2NW6zL>ZBO~uIY+w-Yw+v6i&fWdE9PDSr-9ft6T5bc6;FJ{nF-uqzW4IDaW0_k z40Ar?K#P%kenYT*M)Li?U*eh09%{KGh@l&Lg4$i3tKWY`-P;!^@ZW&%#Sg&$Zmz?# z$CUe}l|-yQ-sxk*gj{(g#O+@`_61ny9o)_BJ2!8g_xroAaPaME??GJrg5A&oTJjH& zXV2<`eV)6y8LQ{(4W4~I<>fhR5^EdoV}EPS4eY+{h79aU+}h`JCsy^S%Wc(I7sRU` z+xhmXx~(TV$r-!SQ-1E;8~(}r?3^{s0lB_^`kwn0*rVYq&TJrne0iesw-qyA@=2$5 zl4q8+advbXcMoP z`bey2E3dQ9^M33hFBeYS`e6*|S?YP0a)Q|X)43AAc0^mJ-^f=S`vu#boAHg>HP+6? zPG4s$wbwGAxthWJ;^)$V9^kL}_DkXsOE(2u3}Dl%4>4#HyKS?8vuXzR zAm6~tZE=k?FhRygUt=}H`$XiqgZuZtn&JEHXM%VD`zUhXzpS5~k<2@}APHW`)sEZm zx%;Et(fJcweC}<}tF`3qWADJfgNAKS^U4Elp9JzV*Z16oxb^_ns0Xq7xn>NBT%5MK zRTCtL3;PDSdOz@;>vO32w^i#NN$v$Le_$Wcx6eTPJgk%J9A}m!U$L!oj(K%S@D*5R zFLRYg%b&nI>CsA@^eGuv)V>b_Y?5mPK{@^XM(=b zklH)7mW)8>vLDuaF0?ZSm|M#^wf1YD5abe+psOd2yL9{leKQE+4-mI?9=Y<2%c~qZ z^od8b^`eoE9ryJbgr7)uXGwI(aXs>1b=bvx#jyz}Z3Dn0MhlR2(yjHDGV^ zA?WAa+N&+@=_B`@vpr-y#qNc8b9?uzJl5#YA=tls3$obu)D zxpg1-_EXgL{q#qi;2$uEHQ=2F`qZ2rWS@5|zv9*_vG^6s)!%*{*vq_rN$?fe|8pO0 zzlVJ%)%H44Jopt$qQKtvv2WS9%9FeK&l+-vXU&Yw-P&Ht!}G^q<&p$3k>J99>d5wt zKZu$Cy!XJDQ+H{Ool0`4&aC=*iMCdeGXocNVEoi;x7@iOmu)VIbufVM^Le)VVGPV| zyPth)Pc@jpI}fyI0$(}xJ-1D4#5%usmBj5OuX^vj8h*vtuVdSDs6p|J%SD_i>g-8? zz5x!8$g{sZ-%&Z{e?|WSDwf1S)de=FyTdFSek3u0ZLae1|9+3mmyql9zW{q%@6X00 z?v8$r*l*yUclUZD<}7k;=dJB<&bw_Nq5SUa`aSyf@Y%~AEmn=Zs@uM^0@R;Z$t9>> zXy@v?ue%u^;0?RZgx$c;`mtc%m^uR+Of_Qc3fJAIkEbbhF5dc^mRvT%f>x& zX315|&qN1&=6TMlA(o(l<7>-xN2PjYnRT9s{Tzu6p%3g`^N9UsANvGxu(v^1dtX*LUvR z)pH(aa+Y~{ikvzt^E&l+Ag6$GUf-zP&YxH#7v?{mU0;dZbI`s6GnRIL`wuWdkHtOr zSv%41i1(fO(XV4$+hXswxpiK`&N}CkH-WzQ<7b<7RZHD-7O?=QuI1u9dDPiQ-<}8d zOkyoNR|ya#!aANOHT@k+dSDr|daEKR6# z&b3Dv&yVk!fh~OJW-K)Ps^{JsG2z`fXMoy0*Sl)XPt=~j&mZ_D*_!|N+x?kQpEGbe zxPl1m4ej!s1@|YQ9>moTV7)t8bHVp4i9hkJ^<3JXLEoB0doKMzyZ_M>{{>Dn4vgQa z9q0)Z&**&593XSy0?+==v*$p|Q|x1p2Ha@?cdc1L95HX*M9VunyKm^jPqcfRpFn-Y zULD`L?x^pbcKnK=$w>-~LCxRmLLraAd)(ZJA-Ro~z^!(I&-AHX{><_^Z4 zZ>{ljKU1H4SI2&i=)_O7GIHP5uG4pB)mRIBKJxABT{SuQqV`Bk;-}X765`qyePGWB zfVNg$cZjxDJBfQY1AWe+7AIC`BA>~<+LyXBr)r4z4dc$5pm^K%IeMC7t_#68Oq^BFsiu6=UNuvhF@5OW7*2FO*D=mrk= z0Y|Q}vGutN`Wdlv)z6;f8!_wa4AiL4oi_5%=fwPOPJL?I z@_)JUPJhQcD(K}750KOgV|OsHh0=5Fc-_2Lw6ldX>|vdd+qO2>c%D zInL0IPaHDqEU`Y1`i%fMHRjKp+KKw-xwEphgU>v^TmMctV{_PXXv1@-Kx>N@I|nDQ zt&wY7Uzy350Eo7C;2B<+bp<_E+}_&Xi2aN2*6)Fb8`MpCAb_*$4zL^c*xLT7jXdlJ zUe(@LRJeALV@&?(C9eVN0{FX1u%KeK0kTq`*0>U zko!FU=DJ7qVzzvgD{zUh!t-WA(b6-}kz`wqSV_r2c8qut+L?%+>^ES}S<-j& z{S5d9Mz*?VYZYViTCQgh*gp{a+pp_c$@jkHxJ#<>|u?yay?&ZV&3~8R)Tgv`98yUa0lm}PR|a{S#p&4vUG7{-Nl*Lwr@}Oe*@7}p z?%L3GFS&LY_{My`?pl1bT>D0>p2chXy(4yopF1b##d;;{Yt@Bw_gUqid%kGLI{!p_ z7xt=oxz5#)H{wbrd}_@5UWqr*f`8P+C$PVJ+cWUZJ4@gAJNBP|_tJOHbDO)Xa|8eM zakm#xy{Q?ymopRiOynxHiw3>y=ME7|%O3)NqV0F5)_3{=x-k%-ZCj4{*UcH`zkKWp zbA8Xm;`u&!t_EJ<$?w2^CuYyYH=aQptr)jnp0SDcT*d=`0CTzR>W^>CoOg5uo~0z# z4DZ{W?O`wLg4_TlxfS}%>#<^uUUG$VtuJ8fD;X=+&@BP|uV)6nbBzymtQlBkmru`0 zyy{W*ANqTC?TnqoGWYx+$s_g*;LOTt2SCoA?xHPEegodq9Ym~qo9M%bQR1D+@4%XR z$9BwJDu=eV0q0~ce&A1R(V62sYwQuklpcHm0~i}XzTBFHZM=irx*_^iho0I>0O~@W zt6sYc{2=$T^Ty-^P%>6sJ+m)6p1TM7Ld+iZZm|Ekhoh?xd3y@y4s?PGGbg?{_UY4z z1)PC4W=-XcZM!M4cyc>Pd}E&1=iar^Rg)i_ij%Ug?YMRNa;=x6pJ;RXihI25p2fwU zxq5fBcjV z+!LpVd24Ggd~06eo_^MLVBB-MleXuoI%+d^Jv{puchBkpeUSI=kNoFf$>plE- z9mL(;n#8t`^9DFSljKeO9;=5O`vkG8e8e+|7vk1>U+2#3Ew=-;NYDZxN!>+$tljK6 za^1UJa}Ut5Yc4$Jc&<~U9@eRW{sh+inRo~O3`k;4pL_G?Z|zU4e(yJzl*+wf%ft7+ z5-kHS8ds*aK}$-#zqepWC_@xdk}(tNS42vu|*$Ggjv!Z?B&~y?5jyKF*o=_Hyq; zw*=T#k8f-O=RfG-e%1uBQ!}uIbMDkO?4D?3CN44WJFwLmE8g%E*xUOsuRY1j3DAkT zi`sX>oaY$yk^hRm6L+`6b4HVo9`K#t$T##PZe175s+w4R{oa+|K!^Wg4ai*Dx&h1; z4?AM@(l>s+t^RN52XN;LuudeHz`n-y<@oFZ$T__|*Zl@J0NQ@~a{7+hE5EFG5WA3f zhw8hz5AvS1!*!wQHlBUhgpINCsN$!+y|rH^ng30iN$1_D^uWfmd*PHaYmtn!Dx& z+qw34j%PP7&zyEx2i!A2zkT@j59&OZy9D|TB(^p0?$^zCkI2=#a`+GM-|jw>++W}Z zvHHBlx0h$>*w!R)&Yjo;1b)A*yKUH>&AjmmoY~ubk&bwMi?k8<9 zv6UBWduD8%>lx`!>JF?h?t}5zAP7dtZTi5>saK z1wTR82JY(5;g=n=#+dzv9CF&;iG8#c@5y^P{9nlTPn>sbbM`p*EZ_PG+{1X&2EY5d zx!8}BBQ;{x1UiA-0292z`#b97yXOO3w8_bF=c*wj{{?>lbBeLRcTf8_ePGS|jt~5X z{eEL#U|z1;I|$(1#~1R%+yC zI`-T4eO`+32y)wZKX))zntb)czYy2%!~)%b{5w$0H&C-;X*uQZ=N9JW-o=BMd&xI9z@M<4X>KNQ&S4Ml#r-Fo6A$9PZS4f+8`{~%uH=m$PS20r3%M&jC$S6K*&VJmDG%GYXxITpuD<36 zeggR&PHE@L0aZghu&uREhbyn`m^r!6UU1&YdoJU?2O+Ruh}m;upIPQ!iQmm3zEk${ zj;(jEZ=ivJZx3@5jKXk^^jlh80VF1Ty$WMi5B%7(9YC$ zX7OUr+&T9PY|rgksuqs>)_&MYUNmfb$pv2Du>a_kex$A827OeR+y`aRHxW8Ot9D z30w38?5EgYU!6diN&nm44d1vk?L7ibuKLnLo^WP|=kw@r2DWt#|Md$R+n?R`D30@E z)t%2$`&3_hfk8f~tGmSZ8QITXp4B;T`beEQ*XNzN;-V+9At(3;ZBP5DE0@cEC3kuZ zV)B&?zpS0*cj12Y&`xk=PD2mR)kVz+5NS$=O$4 z>D!7YesSE1-Sj6=*I9FYW?5srgOd+pO~i`JFUd9ATyA)$Bi4*SpIj&JeDC=cEhaI0 zTB|*=l^%c@#vAw^uJ4w9{T z)Y$Xo%g3{_zRO{M<0a@BYfpdX+1I?aa@7;~EUMnzCmFf&hCGY>5V7n=oputFtL|G* zv^^*Ao~&0tfO{Bk`d`qwcP8fj*|&X1nEV|mJywlnYx=-F?Q4$^#(-So6X*wefDVkQovA%Gs#hb&T+j9f1~~l^ zUU6dWs6DYC=$eOZFGc_QW&7yov!!?SL-zoQPyQ7IwmY5Oo#&b4H*GMm2U@=aWA0&% z@tPAWH}OT_%k$p!2PmGsw{5HtE9Y)OY=Q*(Vglo-k3M;0ihbQBV!casU+mZL&$)c= z-mluTk6d?UP4Ut8o#T;>wEw=*YpY{+TBKev^W8~AVVu0T6?-t(Wi^6Y1A$1d*XKj4Wguy%qH{Nl&z z>%GYt_};7b(dLwyYj^ua0ORPAT>Ds8pKY-?*3_P8?bv8Yk2TMjbCsU^_w}zhAOrnn z?LD!@a4zt^0{`ILUnh1UpJ;iX`-eX90ptn&2{QK{=q8*ItCxBb8=#&$+P(DUWuTFW z_1{4K?(2GP)|or+!r4)t_p5x)|3c5SFDOZBIp+v8c7O*tcj(0IGl9Jl-N4z=oW412 zcc?uK6Ek-~Pqcl_CHAjh*6-|pL3jLjIIqCIL98@(f+NO+yVjgoeXEYznoF#5aGsKX zm5Y^euON3tmtgC=+t0w?W4{4^errE&3k}`n67wwfG3FV4Cf54@Q?-QH3%TMD82E?V zvE9`(U8o6j*!s>3^c1o8$F^nwdnneJ&kaJg^E>q&G~>B?dT%~c`y021yBW8K+IZbx zuAYHk?*VmJV9xsra0h+!f8sN{=4&Yql9T=Zz zXAiV6)<9r4VBdl7UDnRn6(?7ctA3&@PcQGtJ{hfg)!DIMjiWth=Hd)} zyc27Ce(Y$@-=C~;*4R1q55&~gDDpD*%vf=EyP79%Uj75$S?4{uV9S~0uGp_2u=URw zjahqvr?{75T%PBI1&*Dz*P;q_wPdNHl;`J_K2ljdH zbq=`3lAL|2C$=2x?BPCb+uvgkZ2O(F$a@f%WBd&}^^xjF+;cQyJ=bT7JkQ}>H)4U- z?!+!TW}iED?GLAVsfkZ`?)yq!&MWq*jm?pt@SGdyH+*X@z&(D#A82tUmS}g7E2rU4 z;BFmVGtkC;cGfskPVuo{;JUxDPJV!@2?Q)j+~A zXFp}ex`(s>i#%eHVAt4l>Vpe))~cOtFZt%xsRjY;>Auc2SMSn~*a>!Q7zd%|tPlA5 zGuOX6)%JZWV3)wj#1i!JY=P}w87t;)a$eEya0d-M_jV7&oOQ-$5C8%CZQn7vBv(Ik zN)NsW@)vLg=B6>AZO;ZeaFi|u>X z7dS0efA3D)-uCL~6SJ55-{k{2#{{VpQAs3uFvx= zjV;eQ_wSU|{P{kU(F^e*@ReX~wXvv%j+&cz|EEZve6X^IgQZUY@m? z*sesbjcd3xV*_iiFr&RQkNzW{rft24tJ_}bQJi#xV+j_+MX!ZC(^koU|j03!bt zd;=3$XMHBvF`s)7cg_{aajyNwZJnW`len|wIM1`nZNRzj;KX9b)nb6^sot>HnRBPM zIP{#vossa`hTIoo`T@=y_n7$V{FF0qjeURf4s4Enlo4wmYbOYB?4}$)SHFSw{_dcj z4f#e)vHtJ}_KO^2h}?+P+aB7U@d7%KTVfpBdS&LSwU6fyeK7H#_sLm}C9v911K|BJjk7%=CmhT=>c-pq&B(4_c>VqLT%f=y3TuB z`##9Ybr0(Xb^vFb=W0)F=NPwMZTx8adX8=!+j|gg`^M~VUa?2Uo|9YSKX--e9^TzR z2RZL^psNS5E4d3eedU}z+u<1l)>hr8xcmg?JnmaF6E0A!nLw`3z~@stlCw6bX}Ny? zY53Z5m6?2r-3vXyZ#d?7Mv=s~WURir+Lc3FH-KmMPL${Q@n;> znGG$s=gK?h+@tpqIrg9AlavB&qu_$=g^*S5#c+H%N;9DHYc$L>C~ft)+~ep@^Fo?Nk~we+&)fqzBYXM*$p?=;}p zPktl*fa9*te*+EFPVnAt>#hd2z18k-t=tP3z*)6#qVc`A#CDh8eQovr-r@W|T)nN9 zBuJ8_si44u0S5K~3^W)pz`*hn%hL~Fk6>`1!G((U6j~D?p26z!q;=s4AzBh8j z@_rV62KG@S=jq8W=titQed`*r2(2d;GFN@w9c?4l_l#J5ouhB<;{yv4|Auw%%ICU= zd;jD6`kC(@=I74&{Wcc%^7u9*;O<}jylcVFpPbq&`3LYM_2L~kqfYEM!rITi^;zM2 zr^GtriQYR8(DBPM@3H#?yz}Zi({>MYcd$>-^*sr=9{VnE&U4Ol^Y)nVZRaLAfA+ig z8%TLH-p&43IOBl-+yVbzn>Xjrw!cAu8*rZWaud|Q;S&#d_j$?VyTcP-eG>_Yo%KTh zkKfQ?-A%lK(vY`L^&lUyeio|n_j^mn`jhv?w{{_y!Z`h~B>QS0RLoCMTJDkVQ-~p~+=ALVU z*kiB%)(q2Q$7<-v{W-wRb0H zZ&9<@%^s0ovoG{W^5+fIc}{F?fS#*Pz7t>fYp17Tt@S zzsG#bgs;7~U-RH*Ke-S;yGWq4X=4kDaxpoMw`~)=vYwq)|);cFz?OIxVZ3lkd&TRPZ zZ|?^D4m^RMdF1Lbwd4|%9^>f=cC8xwsPVI?d&0>-;ME5FIYF#e-d(O}JkC9{iORuK(%%zvN zr@Y5{_o%z#C$Ptfd*E-*x%v}SKGtqedJZ^uxNg7US1p>VXMgK!XKcMYnv|m^@L$*q zP_H(yXE%Gfml|iOnQ}l~!`7jXy9ssHn4kLbRb#G+0P3;sG_^o|>^|kgxy!kuwFxSR z^}hc??_IoOz;&R_SogZZyU*CU#tCvQSDkYc-x~dnHD}$+S~<{wyN$%cd(OK!f9C2T zug2KAu~D`7_N|jkW4_)P-aeklHA`!s8EAMXYX`BX=hFC&a`tIJ-387%;{+3^f4~QR zeZCDJu0AU`^_~8=-`1bSPqZCgOfbx2t@ZDS=kEgUY@Y9pHI~C$A=f{jTt_ zxie>&1HY5|6Zro(#y`M>^PTM1fN=otme^q&Yj$Gw<BV2I{?I`R1&#f5%r}I_x>d&h#D33-$gTp<$of60J9b z^L?J**Z)5p4c|TeJbe>)=~`gD@Vjk1@clO-{~eT0i%O#?K)03(jjB?&n*&Qv^ORZx7E^ORfhg?8KUPFMG&q1D#0x(xX*p zTv~DupkB^-{y9{KxB+*cSZC-DeNgX=9q4Qqe>o~yStVGQ4~?xoH?NzNmG z+p)Tl=mh)#6R7b`jjz<#z5cy-%w@gDsjuGf>O0oY=Cx;>{&M#Gf4;Bzp%$c;oWA{9 zu0HPAKow0Q{biR{&IlE@Ug+9GzEZ+Ev&bJz`>&B=e; z`oK5u_eb76{FyS~Y=eQm9rv+NP-u;Oq0AM!y@V*5;NZ2xiJ{~4!| z+jn(_TJLBt=ZiYIddTN8^5hP8+B+z709@N<`UUQ2Yq9rT^|Q$7}o$G^gN?8_X- z9y1mI3WFVYD5+njjv=Xx&48f_H7M-*bW0f5Ej_zE=M1)`{K0 z-=o(%eaIyU___=It?^jbfY!HGQy@-!b84-Vx9@;c-|#!uoI3MIXN}y+P2$S`n|gNo z0epu#!LP~hrOlrnwLAJG7r>s{+sE#!CGV-;?2)nh_To9KQ&V9XKald`t@y4$1tbU@mZ?_Kkv-=5B-oDs~mG4;Z{{#bmf;!m8 z=1*ORbN-C=*)=V3y93VO*a<>pQEQi2vAyjTAL|aKAokRLK@QlrfN@{%HsR_sFqexT zT3}E2x@{PjcIUlYkJWo%?RllA_YO70o-46^M#6y!5;U~`tsd^wJji*@e0$ce^pq3! zvPWQ@;coscd;w!m^~-WU|KDpgV&^tsKG9A%VK4a;Yv0Ltfz~?j7}&P!o zo}J1kexSdBCnyc6nosVJ`}#cswYbEB*mh!tT#xnH1=e_A1Dn8p?&S=vv3YCMU*KMw zmplCu|Lsl;raG+O59{>R%EoVV+@*z)vM7{s6Bli^tFdjS0`8AeiPrjPkej%9{${m9a~d7 zXn}Xj?Ps>Cao*8uyF5B!Jb*nC+qOTk)>XW@2XhZcru0uZdEejJ=0){_ub)-1`a8q7 zy@Ehfxqq&|FE(P2oPGq>dVTQv*t55taOUdWupW2zZQSQGPrpobao`)`3&+ zo0O&Ao$WW^?L9!lH!mhR?^Hag4mUs?o9i9bih9eoW~h1I34py5v;Z9a5$n73jsAXq z4OBgReXp=b-xHrc4IBEL2l_W)>^pbxyO#S|r~Q-o9e4-(pMKV6uJ`R~@e|vyYW#d& zXzV%Q0{dpH59~eA$*Ft%lB@i`msU;gn(KXi7rkS^Uw)~bXhhQn79_BD)nmPX&sC@P z>|jmhV(nzz0M79qp2&sL68Bj3$3lI+uX|c=O<7|5I8Uy8^7*mgoE!cm*WmqL-_(OE ze8No-;JNvq@U?%osgVQK5_5W#`Mi7P(iBh5KGy%B|KDJOkOvd=TyyoG z1J1m=`(1nkez&_^?i{Xr4DbiJ3uv)+w9ekf zzVq9TLksu#SUcBz;tAxdCf9!M_yF&~8Rk9S^~~$|=A38mV_)#@J;^!y8%SV}E4E?J z-Sr3D51@@)bpfy6fbVBdXZv3E^ZQ}%MAw0*=lTrP*xw#?CoP=4{5}2#?Ahee74XN7 zZ{(bPyRUb(HqBvA?+coN#vaz$+mpGp-ZKIMnAaEfbcgypu^zb$dMClL+c);^&YHmf z?rVR)pYPx&sC$z0*xR>KZ{6qK`0gO=+j3jYeeEyrO4!oGTiUxj1X8xAtSb#krHcoAAWi!<_qQ z&4oPHV_$W}VU54hGrBme0SG6=LSA?Xnqn8 z@D4h9XU4~Z_e|m^)_&G?(9mAs8oo2^8~78r%Q-Xgh}HU?x6kvA7T(yONA|b(3%@=K z;>XrFOJ2?OW7XO-wlD9`%Ru8h2eEqR=v!ZV;{SyI8@}UvX7UTMcd^!*`tHSgXLqyq z9oX}Rt)BR0+1a^CAA5E1P(!`P9k1qNy^%AzF}$2T?Ct&o-p}7!;oI1&8Ee7F75@w0 zzsFkZdpY-}uZ1s;*ctZdSnu@{-@dg6bsqIIzdl2Kd(GIKdi)08u+H`0UERmK4EO*I zKd~LW;meI!pV`2_pnt;KLydngHlO&l57v9yvn+2FMhCZ+uNDm=K*||0cSt^IBQ~Gz@1O-h41OHKfh!iu&!x2?`Xo9^T1c@4Es252)I*| z+n@PWdqFed2pgZvw?rgx{r@;}u9>pI+ZTX&ws10=e*x7x+M^o6+wwCd~K z;BOzRS8?Q=r_N)}xihgK@B`bi1$=V{FMHWr4YS*Q<zw6#40v&T zbHCyZjr|*N_j(63wJSZ0t+%hgpVhpyz@5(T8{fql<{R2NXgoKpKbxGBx$g)+g8rat z;Qfxd$HW))4%mqGPC@Ph&ddp)U;t-Wdwlr`uJF$6g8Qh^H{bDp8q32Cpx!%sH|LqN zkGljm(9|KPzO>{XV8q%_T_?6jqA{LW{fQq~xdyz8J!&q#oOkh_9(i|exqfE5)4;FK z2jBdNrMLf!*#6e5d1%Q8y!+_ezoR)v{fyPW^u!&_Q~lm;%F*uvZ=Xi&?A*M4)dyDo z^l8Kcjn*Byz7}{Q%d9)$TYvjlynA~04rfjIxpr{Ygnxs-;tzal-|+4EfDe3osB?#o zeG=EfRxaVWW5(K9-PnFJho4x_jQxhZ!@GTQ&8+W;KDfYZJMj4V+Rr}T%X?e}{&`!i z8vFZh_O;K%TBq-Qtx4jBmA}F3XRg>i%!^P%-}*ZN`GB*=9-bLr{p@3p2KU0ZwqpE& zZ=Xiqw{5`v)eQ9hy)ekx`>}a*_Nn`lbJi2io*mxUqqR1$3Dk6~{V(_)V|_IZyyUUY z8oBC{xcXxQzUGj-;JbtGV$Umn&D;F(?NfK8z8}A#RWB##ar}lx4EQ%R=5F-t`f%R8 z4!)Y1OOwWE8hD~fc>B1Udm4MM3x40c+JSGM2foKyGuQj+|JWM)1pMcL_t?`~d)5wE zefJ(%X9r09D^|R~6I?*V(zbi6-L0A6jo9y32hVNo@3Eg6dx$q+FK3t=_x0VeuEAM% z#XtE|Zx8>T>^(0)+djT@Rf}c-zl#Y@Z~we{fDYVK%vgQ2*N+8f`1h;_==jd|_pQAC zO&+xSnt#PQuf@_k#~Du$avK91NBgn**|))Y4|9#!nf6QI8+`+wCsw_mQ_IzRV%_r! z)SsNZ-(l;Eig8usKTEJTbLiF8fpb6k&st(@ob~pL^$tDOtVr}6=UC?s zSFHDcg40vZT)6>Vs zWp9rdC!>C-C^4aE7%W{Wo$~^SSaJ?%Y-1cQQUZ zSu=od`mX# z`@Xa<;z6Ce1e!PEhIQV|-EaHOk^7JD>)&RKeHXR<-tNSn5$j#t)p^eS2`=CX-0y$V z!@16zpi$#q-s^W_HSVPU4y;w{8zoxb+B?LqQxoVLIG^_<=j^|Ly{&stV_)C$2`2uP zd}3c%Yvr%NGqmvP8o1&209f_Dv$f`Ge`0+<7w32UC)~tZt94$Bb;o!7nnf(Fi&}FXK_G@T-tE$1ySiRlB{y#S7JN+PbA8X9{reaXztd9MgxPh91^&J9Uc+U<- zuGl-IIW2f2C+Ca+=Im>&-_eG(ckOGKXox$!`>EGD(;e$GAaCE6YiH+MSNrbw7_sW? zA4lKTyIzT(!1;bR-A&Ftr!l_q0|c?VJ%BSVaL#?l2C&9=b&l59&pn76*6*0T(mcD_ z$M==%aGB_Ub=3o_&YAY>#P-tH+B1N=>3;*x^X}HGdBgYjp0jQrtG9EUQ9bGJiCmv^ z+xCg}NzI?YTKNPW7~6Z`PprPV0B;{?K!cMP7h<)Ix+Lz{fqh}cZNoj^!1!2qY2Q~} z)uXr8({uFZkAb)Bl|bZHJHh+UTi?7zSb{JyR4-0t)YEbu<&#RF82Tsi;uH{AV-wQiy}Kkmz~-SHow%Y%+y z|B4mv{06LRz*swg@9+lf8^9gH9GveXmjdoWuIh-@IMZAQukAO^C(pjVPu&&Ye*XMa z`wr9zdtSEfAA1+#4)2^6tM;izp7_)RvAlgU*_&Jfe|EL+%ihTOU2?X0dk$c1y|KP$ zCUG$1pKFha-!^{z{oy-3LCwikcO|yo9i8cnQ}ZIHZ;gNc|AqY%sP6&r^S<68iS2&@ z0UDYKH-LTuXISHR%RXw&J+K`sN`sB~+9mGfggkhmH-Ck@;@ihP6RwfhH?Kd8d6$8< zzIVhPwf6F;ztQ6x_%C}2IP2SeeNHoW4c2#h0ecT(PmeO|7kyxVwbh5%UiJ}jY_9t| z-+P!Fz|&*R^p3*#iS?+pznpv2-q?q{fUi5;S3fm%V%1nX@S8Dr?Zgw*TVbuQ9KB9$ zy&v=>r+@nUIp}9n0{VtMN@I*)b!a{M6;l`Z)sLL@)x({f0R}rv_@7vF9(jL%IJ?2O z+!lYt;_YkBTJ;aK+L{MfHL>4;uzI6;QKPTc{jK*+Z@_o9uE#=Rg+1%DzOQ?Y%?0rV zm_PQ^`@b9V7ybPGVBWb;(Dd(Xw)2H~zn||L)`$yGZ@%Z6?H+^J@0-V*C-SS;jI~d~ zsqH|$?^!;)dy0s4$D>VRHA5}CT|ZWjK^P2fP}6&&>6{&KbbHj>ejZ|BfUUe#h+Fa?Nv&y*!a{ zv1Y67!~-O#&kU=^yH{N_2#|X={*|0@S>m&ESyG-6{E}ZiJN}bBN<+^3c#LPRd(0rl zSLYqf1v>NYmRL1GuHnZ<$C{UOuKB9P*P0t}e}U?Y?)I^F#GcQvk+;?viN=0aPwdh6 zvsI_>ah`g2a93l$FSSowe!w-jhV8)kNvs{YH16;wM{KVI4a`{9-t`8=em27I!xea} zaj$Q{zFl~C__FvDKS6*I+s8Iz>JrHRj(q}q)LqCMKY;J2ro$z+iE}>IUUd(Ab!tYU z|M%KUsdw@Io|)?oa`m>cdIx;qdp~DY4{V3~yca#qozKEqb8FnS3AnPD^A2+MbFd1O@%#GuHPANv-?7fT-j_~aG{Lb?12=RNYi!@#{@8KTGUwD~;?wiF`%d!a z0?3CvecA_r>tF)&zwTXg(5f-7-Z}c-;k9FR3HKy+S5fzd4|wCVW0$*seNLrC<6O1& zlDlj)&V;ir;eLSn*_BJ%u#KFW2f1^c;fY{ZU~hf(zPml-TI|`_NzFfggTr3I4X7EQ zlea#>^`ljvnX7lline2UHDU%i`gnqPWV!1zfOn34{f-52rZc>M^+X=YFF=jGjC-u! z6aR_U-ufL3a6wmnuq~GUd>6x*&(Sxp9nt&d);DZ@#%Mev)*O4ZTnOw>biSi+Zr$na z{?@3i?-+J~245$3yu<1%2YPPu_`APsz}Zuz){~2`+>X6ZkG1pM_Q~wp3;pTwg8R8) z{n@qp8*k1!cUL2PAOAN5{x9HvXQsZlJ2%a9d<*}x$pctFVqqrZ@w-6dt_i1q_NpA8 zi#2KkST{iBuxf+8??4;$sXTq0XN^5Ypq-#^s7?I#3pensIeg;FiH6^>9jG}v-&Fpp zg}-3c_-Bv*`wIS>pZ_`7yl>G#l5hH0zc&}~0vQV(Yi?|)ZEyql&gV{(SmZ{1cKE0F z>2L1=d}n)x8vMl9x`TYx-B%nvw45INlB;iGC$KJ< zb!IyE0&c**3983^)rE1c8uM!H6~K3V0B!Xp?@^ba1Aj*PISgP;gY%ddFKX>yAokSV z5{I0h<%~TK&K^Nr-zV$<=B-y-dF;*FkEQd^_C%9t?CD+0Csgw?ChwaxY~0t~)r2wj zX^t~<*9X1vi$^aXXw9a{LhBRYFIzj~7E zTi?`p=wr^=EkLgN#J8_^vEIJ*J!l;Knag(kz`yN2SM;TWw{L;G`Uw~4)M%|sP&;6a zoingcHP|Kspw|YxcQ>E8c9ieL>N}`fIWUO5N8nF>;lBnQ-`i zZ><-nr+WMKTt73cwMJg+{jGJzOAa{KxS$F2#P;v_6WP z#O5#L2iCm&AeQ&o)0}bG2PTz45-}@#01Ouq=!2HC@ zRV}`!Cz9PqUERk%@Rh?(Frrnj#1H)Qj@OTUKX^40B&a@c>Kmxt(AHb+H*&*VY;G`+nw~F@e3U8T$LoTv}@q)QP2WzBASL+m-?;8>9?Y#dZ}%kcJT*bS zlfT_pJdEMx%*nripLy*FZ_XVP+Y%p?rEbR3T3dPQ2Ap~0>TQ4O)tGyrId`tT;U;K6 zO~II>-qU0Ez=l4UU~II_S?h^dn(`AJ@EsDHvEt7;#HYTS1D|`dd&uRo7j{=`t=7|X z_578eaMt*_JwOBJ4eVjh3F;kw@iSMy4j0z#oO+M^1opTB=hTT6Cw2h)1lHd}p%0FB zY~LO)aNC;Pe2?~dRqyJ4@+rB^3DpR+ZMSgZbt zrr~==^2-OnMl4&q=!5EIza$nO>ND5R1@CeLzMH8XP^lRiq2kNZJ_^S1t zt$BeCZ=IULKfnCyLG1Z8f2B^0Bq2V?tJCNFLHrZAzdP3atkl{Ezh+uPUf&am_D&5L zAc;q;eCc=WoB=wyCUT)`W91(7{rv3ZfcgLzdYk(ljr;lcdmkdv;_P3J{^RXzUO`HZ*3v>Tx`vR);_7Z0C%3|zz9l!bscyw zXS{(cP!pitmoCu^VC?MB=hW`G^wzWuYmcV-ka8FO+RZ-tvACljaQe<^@%8h@{TuiT z_}@j*TC1id{++u4|L-^4zxHI_!1vx-b@q-sK5^}iD?ehzp>%F_|^u{I%jU*H}z+QkoU8&r~i)l1`?h9+QzTn347T`PMz@W zXD<7s-uX9RUwPxiR$Z)}3;rg#Eqbv3`IiI1%ek>Z^`BG=2E?9W6cm zebtRQnM%dU69QWYow>-xUx?ko@A3$E@LB6l|E{X~Mc+EL*4cXk|D2WgIO~rOc;FlOZ|_ea z=bqllUH#tpe(ECD9OwQ6?SBCO?=u>`8o7o>`@BN|XE~=VHS_rFH_>(ApGD5__kp`M zk!$xid^G_EoG@ps?)9<0H}6<^_fjM8nYrd`y~Bugr>e)-c57DSoiTA#5n`r zoV{-J`~&<1e#Y+P4xN1HqiOioB>of5T2I9CtskHft1l3L0~av!D`)>3a7OXwfO`sO znjdf@R(p9juRxf_!Gq&8Pg<%v77cNO-j`_NPSxbN=(Y42p8P{WMs1z&dazOmoy z;;;i=-@a9+C05t4-cQZxt=99Y$@PvCYrS;?D|~cfOu);VJ z(8u>|_`ZRCug3R{So!njjo9P+j9Bw3#<#}1*uUoI*0ATqKl7aBoCmP4@#$}0dz?A) zPXRY$eZ~WQf&iWud3Qg(>}Sqi9&PzaUEs6Nm4505n83Kl+Sl0uD6=XGvVATumj%DDtE5-#h<`F=kxHhGUt293ujbs z!hvs`K_ECi`6bqVft{dJZ;gG`Phjo|=UqE~1J24^`C*Ly`mx>xMV%Y^^ZgmQd~1Yr z3iN5j9yR*j*E(lR;sn~_;_FV{HGugJ+_PbQi@?eq+~Des;9mFxd!@Ev-954X?Df3l z@tqgIT!V98e-_PH+#qM)Ag`w0hhFygjr=pz|NK<()^vD#H)7|TtDpJs_UM8h=W`y! zHN%=%@8G-s4&=>;KG46wx$_&a&jb4dG;+>)>VtDfeYu|?(5SPfKHpe9?dxpw?)(I5 zQXc;geEW=8`nn@I|NA&Mc!-a*Zb)z4aeXWK8qi<$uSxf83GKRRdrmHR=?-Re!?%-x8s3!v_tqp?R9e6IS|luzhvYf6i6?&KdIOPBIsqBA$~ zub}>pXwMIH@Ev;tem3sZ;Do%p4bbJWPdM#JEU5YF$WP*Tth*(?do=uxZ{Ge_U@m~i z@AV6)5d(PKN599~e`<*XG{K2jxhvWR>hsO*lXI3kxbq->Q^V(Ioi!cXfV=toT;Ces z;0cB~;M|(Ix#|O-K;C@y)1pbNwU^@)Up|&jt)~aQbM<$3Yn=7MZ}{qbFZXgkkMoS3 zrQXx=doI2Cfgf;w#y9W;zVY9n12s+L%3JGkKi@{o-1m{gf5Qe=jXllV<7kiXo}Ycx z>^gPd$h`xtZzN~l{sUA`V(Xi6ED+Uq$hYWlY7^^cG_cmW)73iS0O$7z%(}0&Ne~%FIT#GXg&Z0p`YhqrgVHNNwk)*RZI>-(YJWAA!LtbNp7K=lX} z9AnFTo}7sSrC*4!JK#1`PsZWGQOv}2=TYwtTIzUG?`IKNkKV4NI$ z0{2jF{i(a5`wP6ljl4a&8e;1!hO^#1I`+J?wHMIfCJ3O$T60%WZv$^lVQUgA253PJ z1hwwqo-J4HgC338@1MU{+$n)`&mPWrps60znXi*@!0*fd_>Eeuzi$%ESoNPf$-}wN z{Ae}vPpp5IxlfRrd}H)W{w}5jG?^y3a?ES~qCVF}AXOHRy_YM8o=Z5cW_rHRggEsC)Tk?&zGOW^AbQ zZuNb`TI-Q>&mOB*U4Sc`J$*N=dTY%m7{DESE)dl{2XTN2%8yp{1FduXp88JW4K#G> z-J=En2h?YJU+;8kWwg?zqQOz3&27??D{E8SmgP(1KndHtaj`op}YuwI|whPi^VpoHc+N z>k`zxiLGte8yMz@)m?$sKAxGo-!E3)UZRc`s2zfO`x$rqH{)n^PxtnChZ$dQXTJ1F z65o1vwbs*f?WB!ZC=GjR?PX3osd4U~pz80fdIs?WXoY&w^kez%HSjOkEBp<-z(A+3 ztvdD0c7~tl-yq5<`bO0Jy(76#J(%qpVWFEd-x8{41DiWCs&Ud`?JEHS%aKA)%Ql<&&9kT zuD7I5#pHZn`<{Ean|<9mfO)m{ouGl*>jPiJdL!qkdBOP^yo0|$^+yxrthKj0JnqDx z^6&xg95wcOpy~KSPH<-K9jxWJbH;k3(&g&!Uf$#6th28(&pf$^)k7 zoQu{yD@K#CeB&2!CvLzQ=4P%L*YE2yvUh37+b7U?E{DgLb5DCF{_o`6&+n-3>={W8 z%viPd>iFkG{_^t&UtPTgwt)w|oH^tVcq{9Q!`G%cXQB=KdJ}R@p8dqc_r8bw*X|2b zzmMGA|NGUKe9Ce^{4NsgS@w!pcej@_hZeZkJQm;G-+=EauP#8#^)B+(crw?!J?;b` z$hnLC)wqkA0qV`ME!Ow9J_q)5-h}u2@eZu*#GW(Pdijhm&VHWy9^+?zy@CC#@x7kF zxMTgiADauTd1nN9&mZ^o^I~9IQWNX`-c7#&dwH*pEzN%0&wbE%rzC%@x|bZe02A25 z9SY>_-GToW<@Yhc<-YiUKlw;}a&>RsZyLwi$Deio=WqO304Cf3&bH=!2YyofZarv$ z1*OA!N|#uAcP;l*XCHUIYy?~ad)Gc#ef#^a58!92Z%yv}>0^KUPWXWp)*lRNp5O-D z;b^)Vv{m1hkE!o#pmpAxHNg6oOEZI?Lw7>p-~D{!AHW@7paE;=o!?Wxt$$bhNo)^y z>eMH8=z|Gb0P+ttzg zeC!=NKe5NJ?+!WhLmL~I^0i5hUVd*L|NoxkdMr)N^7Df8t=@p|Rwq{fN!$_1{GcXRuMX#%$zR}aSdW^S>-X+A{1&UH-;vvQLacDs z<-YpaL!Ec9=f8mcuUL`z-s3O*kM(!HhPHNrALRWmUxD9Y``EhycMI~?cX**@=(B@0 z(>z*d-F9sK+#36C@+ah95Qa*6N$ zT6GV6b#K5s7&mf3?yA2zf6m`Php$>v?&{q04*I7*>3WM1Yp+i3LcP6(zL>^X=biU; z?(+_(4*U5g-{&!@boQJI6lmU~lhUJ>k8(vnE*xk;XG|q&xcL$!##mR?2Kd=+t zx`zL7Hnx#-XYa19clNEx*>7S4Yu>vI;z1l+-^^{!``BMw-wpgLegkjYT5He5Uhm{M z_Pm+LCNLkl?iui%SnY%heD5zj>g2i>?G0Y8!PUV&;Kj(LO*rpu{FL9ZJIHmcdV9o% zymdlu5;rs#Ylyvf)ev`b@^Wul@(oV@e2U&81`h&+3iJNedk%zL2Rx4C%%2uH4yh@A6WNrH*>${tB*aY_1wriL*6~L z>I2(}PmjR&&RY3*G;iPnz5)A9;B2{pf5ZBIRYzR?$ZN&9SKvG6C$NV(_qSHwdUqV~ z4QS;nCie!syYWLlmkr_ug1CC%C+G+8=dEi`YO4m`-4k1SY>S1CuVxB-`7YS6<(h5X zfb+cXKJM4>os+TN*gE-_y!TTB*S3|nmzw%{L2Z3^vGzCTo;BMV;?tj~dez(SKGyVH zJik`0HTECai@kybHn8XY&AF4a)azf5PZ)3+I;&=Gzk0~!>TP}idHu|FFS#o{otya1 zu&4G0Uvp!{{91dNQ#*lL;l9QlxT9S4#M=LS))DK?y|4WrSUG*UfxZF#7OT(cug+Wt z6P$PizJ1^l-=7C^{^yVze#d{{+tc`kU;ChUZ@Gc*4!-RZ_;)7t&N*+c_5z>Q#hUew zCg9$H^8?;qT4z{O-;uU%cwr8b5##K_Vd&m#OgD1 z@z$7kR>D1Z&qjR1_pbKyys&B`=>-yh=wq9fJtqDGo$qjTa^_;^f}EP#n>zFMe%;}S z6AUIZ1#J?FixxxoDi62AlWf1H?k^1iqA_OBCbXTMwW-d9`uVOs!d2L8l1=f0&Q zmivp?zDZ7;cegfye_q+Ik@ItK_YN;S5&Q!G2Co(B<(?OKb2G8kG6K zYc9ah(yI&oSh?yKjFmE+gX zpLNbT_tLkIv&6*zL9K7&tbj|nU-i}m+~Je>4XtlF@x6HB#fHGX$qzM2*s3F*K<74Eq~8W>>a^yc)TwFP4lR73^*dPevCf~s8Ny>vIW?L59--Dg6JK<2LsL5jmg9Tc%UZ49 zt(L2PV6DFb{RF@7`0c)CU9tMss|~E*L-$Z$JxDeD*l?!$dW%@PhCRGBVgmU;>03VW zMISDK);`XBgHQ5L?2FhP)O2jqC-#q0RNi?p~iAT*3)^yU&~2ZPh*Qt6%j( z*Eg)OXZ3AZdEVRD+{q{L1orOm^;w_`{9&xm3~fWB&K}-JJJsN8eIK>w^dKjf>pk@C z;k&5+zrT}5)9~%v^uZI~H?Lh1>b}sh>a8t{@B6w-&4suBlX!qGzqJExC+Dohx4(1T zS=52I-^iu0=0ZHNS2f@Ql6->`>g)Ro@4n8n-aq^P_xJT@f4w{0ZP(hX6U$p;&q-W8 z@lVeNXD{p2gc{GCx>(=sft^6D^=i&%XiaIE@rM7fhxWc|?2$n09DlEmp9n7nN{4QZ&>%he{NrG)iiXuzH8^*lGq)rZ}`?~Pp$n2u{N-SUhd^A`G)nF z>#=xiWA_Ptcy(8N=X%$k>vMsrHyp9wQVo8?`Y!Uu{=9Zi=h(Av-adgpuv6e$YiwP8 zmiXsh4PSj7bPszPXE93$@5x-~v38x<0UB^OZBGn-#}9n_B>n^gUrp$PI=S|g`<1`3 zhkH+Wwf58|@V(4A>zoL3x%RI)YT(Xi=c-$BF|tZi`CXe)>H4nM&AFLmbCAC32LX4RoH_Rjv9n)TJr&3aBtunaIXR9ZsiA- z*gJYM*(3BgvF4mTtTW$01UYa{PrLxw0X*g>7$9@;?knGbI^X9H5Wqd3-~w`M)V_bL ze^c+zixv3E9y&YCVq z-!4brv(wQg`9{uo=yTMT9|ZVns-7^2yD{^;ZvfxQd0O-5%vkSgy?uVaum2uU`;8dR zy(YZ(9me!@N3{uz<%M^?0q4k<7XC%-&yb%$oyVRnc28{GAM453wdim69mADA*Pk_k z9UEloyU}BP_p~hmwrh8f#24l#*0-4W9`|;ZxwDHliLdaHOYgs7s2RjV3m&^xoiza` zj;7_}8{7oGk2zy=YDa zTrl672DGtJKU;J6Y3uFOn4j?W@h-+4q%kt_w(s!tkoChPfvHTXT{W3pN1aaoLYDAJs#NS=G9N&n|QZG7jjs8 z_?Z|7amRYz;N?cHUJc*4-ny-M)(*Hj$*=UWzZ!eH%N1C6YIE})>vubVGvw_T*d7gH z``qiqWyx97fjf3|&h^-%#aF-61GakjMg=tPnOJKER;~U7YU0>fTwpt#nscWcJ$zGV zx>MDt!M}mW=1{D@=&WyGf)0$WwPyqFbapha&OYWYz`pKMI^P-JJ$!@uxsAPo{;_JF z*dQLerX@N2`W)qm2e40oj_-+Fv9@Zl&N;f;1Fpq_we=Rvn+b=t&b^$~@%_BLi#q=u z;R5WV-rUhswLW7%FKX>AtPPP0)|<&)Hs`xd_yo?)Tyx#KxRb;BCTiTX=bCMO9rzIo z*12b3-O=Ox)33+ZdwL>6-uDeuk{07w8-@9cjzv|fI%nf+~=iQ&c9h~Q!7q(-C$K4w6ZuXtA zYUNr2&P;v@>dC)=cWxlztg+Y8$OWiccY+!GCGx zWB0)~76VQ^@2z^Oss4oL-oA_dwZ3iTv1+X~E-k+K$kn@Ho&5#^4EboKao4Id7WuW$ zB(D2n?QeXgK6CZ3R=p=;p=%R@Q#I)Q-y!wS;uaw3dFtTQo*86orx|b{fHmqLXubCh zR6Qv*bzkzQ=ed8qO(Nn{|NGiMgO>OM>*={_6YDJZJMW=(5R1?f*Slf^n7`7?**zEM zXFc(awQ4-KJ>MP^7)T49W4gLY_6F^;Q@ekM0*~_8S!$l(omjkqP!CR>ea|~~c#qaubx&gZU#T6y`)J*7qSFrY6Et{t z@H_GowC}8eRsT6-zn%K4e0<$~68{PO{O#Gd=0)tC&i>Z8_wR5Wd;{*1;0Eu!2Cm=% ztaGQ9duMd+eU85SiMls-pt0VX8}@nhYWE!)8fUs^gAe>G)_1qg9`>C0#_DSh)?;3+ zb@hG5TA#UQ)Eqc_TW?%BdOqcV|K@gj&sg`fXYD|IK0{{*eBYXfK7D*2-|HQCZtx>; zk{)m!o5Zp0=iS?S{}&oR2lE5WSa44D!CGVfhR!#wzHn+9alMJL_~*K#pI7nJ1-hg0 z4&GbN<7fW_?os{7wb*&=lh`{Ye)0J3GKqbo0j}VWkM+Nq=J(J44#^wVcTwkTPXvDA zz&CvN_&kA^>wE8ebdbdV!a8q6tGbE*xi2|C2j}Ous7jYX58lI`PjVjbVXb|A!-E^X zGdfW3?AkfL?q{vX+PVjKvVIcFJ@8+^Jv*?Ex)*g5dveVDrb0zk(wPtL-ajZ4p;S)TG-+*s3V`=Sm0}l{rJnDKZjj?_06L5YGlejqIdIPMn z@K`&5?=`V!p4tEjcfqQwH^pzpf|}xU^&s>44&#FWL@Z6^(qFeW*IQiCxbFxo{k)%V z_yWJ~7+XKF?soR9dN^m*E`e{JxaI1VaQ3oq-2wm1Qg;FN)eo%bK+e0j032)7*5|~| zYOJZ<@#XzFCj7ZHV||uYkKfRiuLaIA_Ra%qU4X>*es93LzJL5Ohg?f$VIT0R&s_Ty zFi-f>9G=iU<=o#IXNphV&W%|8oa;M!=K}e;a~(g7wOi9X_5(FP$N`Vq%EkH&0!;$% z*um-HY|jI(#e&+AYu0o7_H#e$t+`+mYmIMWKJe9aY<+I{6F-60b3Vt~FITR7a&PqM zKx_Z%6Kgm7HgL4|^A2+MZwa&>xSs0{BX%D@7moGy+2Olm?F;9fUm#-L`+{$uNzOMg z-^mFzSK>h~@jca-epU%^Lmy>W9Fe*oHRoy{}qz^5?$a$iJ%rpLZo+wNT#S zMXfdx3+y2Vs6FHBbFofM$4+d`X~>CB|Ip7>?_2r5x#OQz{%pGechP^`*B)nwNu9XB zd*nx~c@Ol)_SKiK9R1asb1&cHztn<26Z-Jh_^ty?cx$is;-_%*k|!`Z)K&3m`0 zWwv$ZtewE#`u1$t^2zrk*19kC_OizS5rja_Txsl0?D6dmaa;YWKGyvw);f>9Jl5CS zWQFxTv40Xf%RbiW%lAZ2|9U5U->kvg&slz#oc{*m21m~O$yxIRwG(vhKyI8Gd3Tw_ zDaV;vrZmp2UD0{=XR1ax%N~jE`wcl@ubPP;pmK2K6YqJxh1x4SR3ChMdYt8bGS^)u zw#AzJVpa!o=Dnji?_Krg>FpcXdyqS@PhhOJKF7r8M6Mp=ed(NSjeN%scxxtbc3^vc zXw-c_HSVzM9{3OVdW+Lf>jb_!-^7=r1i;sVa_zm7({+y^+Tkpv%<2Pc@9sTd8AFeRh^R93M-#b`q|6ONpnuB+*pTDe{-LJm$P8t`Gb>duG5T4se$I2txe8HejuJ zwcqi*(?v_K4(EN`ZRm6UpD+9pE9V^(KXbivm#2OZ%WEIh`fm2KSM?`0@7y|CHCi+g zAMac`a`qbV=9(6+YVf;Y?v>h!pFq7jkDvFxPvA!ati11U{efQ^_ADQ+>fu8lo4LNT z>h0eJ`cBTz^v6Bncm4vO=recu!GGM~*}28Ki`IHi@wu!u2YTP9`UTb#^=s9P{P<_njxW^v7Iord?ypnue5tPBgrMU4%!(j_Aci3eVki9w&mK#@3MWZ@4!Jw4XW?mzI& zO|1Ro4z020>~lp^JiTj1uKv}_dBo05tUUvL>9FpWa1%6OpQl`|pB;6ck*lXOAHY6x zKki43oWI9b4!?tW;z~`=vo{)jcT!V;Q|n#qQ91Gl{sjz>)co>J4={jr=BgLD0G+tO zzsT22{Fdv_k~g?-!2N#(fAa+?);cv)8~o?C_89oa33`IRuGqhS1h`Y{ev`blo!A*# z_Xs#o1TlB#givx!d}?hWj}G4UPBrbEv(ssTONbeS7@@x?HUHbH7H; zneK1@4&>DgV4T*)%ON_pilJQ;eQ4LSm!?K#TD$+ zW9?uc_bbh9*}aoFXS#dKwW~YTUEtNdtjE7&-N%~J|VeZLoQj(x?C17E!$BGzpG9{Np9t{%?$4g3NQ;5^}e z6HTX1ji=w%F7nRt{jKT1IN&|<&NXN6E3q2w3+vmPw~t8tCpm5PK-2KmdAAFg#Jk^v zyt)B`oX5O1{$1Xl&dHOL_uR?Z%f6p;J&1GlaNZ!c#y8b^>{$of!2nmZ&r@&gehq6s zHTr5q;QKB!;IV^vg12{K5vy)s6R6*|e*Vm}cbz03A949`_M7lx&q{sz`}z8vF*YB- z3 zT>v$AptWvWxc~#MVV$Ag8FJnDw5iRL!&;{<@%P(WTYRt+F#ml2=-J>z?e`~qj(tb2 zKAW#O)Y>nBz0~dZ({H)n*Wa7``R#Ba;0M;a{SIP>RR z-a$=SH2b}N!yoW>tTQk8|LpTL7UZsI%m*l4;=_d&9PpkC+)E$KSbrXyPkcFhxxf06 zYkzwuzI`XQX2JdR1wEfr)xaO>;+!uJ+`(FZ&ac3F`6k1lAZ|)f0~(=JY@+ za>00@Nvyrpdn(4?GwkVJ4O{OJ_{8>>6Zf-r`w#NgT;T1azcn+~eHuCQS1{<|+!wxi z{Q&xPA9!b~7k7MfZ}`r1|GT~?P9$^S602U`{nhTzV3G@9Pknn1pg-UeSetNl4{U%m z52_dUd!n({9XdA7oP8U<$9%;G_*ZK79>Z8IR&FNWDf9GpXYaoGno0aR`2#w$ZFd>7wy#L@;{PXdl_Hem+!r-@4$SJceZiEpV&M4hBdEs zm$De&c;?v){ugs+e#EM;H^FaceP{J@+g6^m`5x=L=o|kI)SFLu^>4(^==g^g{*D#i zMLs|SLmz%(-P3%;y06xrYSgL`{{Cbyd&{fwn5%oy$F~&5=Iu4`-A#K14bFOVRf}(2 z+$aBqHi(Cu{%!p`aX(ji-^84|&%gqC>$*HhIZ!%my#x6qS9hjI!#}VW_<^RhysLE+ zBxonxU#Rs&^2=@Y`Vq(CeOG&xhNPcW%|G+Izx}+Q$9Z?^uAn7);sIX99BU4=TG4ZT z=I-+ZUCtc(+7Z72@0W=Wl6ZpJh1ecy15T|vbL!k#ew=*cm~+1inDFlY9SqQk)%qrN zV%1k4@}cH@hpm|jhyG`BL!bWMv7P?zSH0ogRnEEu>fAwuKBr~`4%qw3sU2X#2k;L2 zE=lZ>AHW^%;0fGy!d2fS7H2;3ozVmO!~?ysk2PxCyAHm+tInP9{n@&L2WWA=>kSue z2X)BpUI)2)e{%M*RvTF741DdLqt-cU2L2uF-R|)EB7pCqpC_DUZ6p@X-r9+^pY_f) zZya%I@U6991Lpnx$(^bvTt~alNW570t>!W}v`fcG9) zdwA;e^Rx2ZbM;fNR-Lw;zI#7_yFP)vdXo3V`5EP!*WRH(?Fu)68fWb9M#XT>yP$oZ zHEO+=^OF1k0oQ@Y`WaZD*4@=tp4gdpAU8l$2RtMA3OgUv;I+^$zZ4O}!m9(AExT&F)zo zH3Ljx{=~kU@A9OdHO^HdD$aGsH#C7?Gw}!cI#_4O`!lMhfdMujL7rGE@87xJfV1kY zsq^!3o^UU99d2T+yTWy>yCz%%{;WK~1C*W}%w4~%|9>rV#bfPdpRSMnPTYV$KiUaD z;TzW29rpWPPfe2Zy+ci`@72!9TeIs9YV4WlI(A^~zpc73CfBf?_ztwr_hhd7+Q(f2 z8{{(9do+%H)OGO0S2K*!`!?0f8u=i`iXIr?-ji5dseb`KmqBdrjxEs7yt#y%U?2G~ z4r<;pSba5W?IZ91Y{8!y-%MB|toPr|diNe{uLJ+_-l|W14$L3qJo~LPR((&|wR21I zXlo`tteL=?gMNYU9rlq||5@iw&Q=>(d)D`a*jWQu*T`viJmLKr*ga9ufSCS~_@H(} z+S6b;beje18T?Pwswdeml8=bzUcD?HlpP^*^iVSaCnO z2i85z*PZaEF)-GT8!>SAFa{oHWUlzlJlH!~+sGwceJ0d4eXM=lxy8zr4z2u_@-N#ClzH&|;Y=)!x#)#qW1|1^%(^R5p+ zh!bl(zzbY}-wl2HO*J6!?P*Qy7|L13E!Y3dyDd~w?42F9Mrj=Zz?CO>u{OtZJHc#7xQWp8^peYoHh1RCm`G-S@QCnrlFtV1T{b&evV( zlSvM!+4_b*;Y7#x^RnL9c>}99jIjf#@h)nFv3YgAgU5KrxewM_`zF?0-DBIfbZY93 z=!U*0)(mT%BR`UQu;+GW%HPk~`?K5cTDw?7506%T0(;C{cJ~Xo2e<=k-%h{bR~@zk zbH-i#b=`O3e?$KSYK)zy#y3+p@P%jO>EGxVYv!Ii$?bEc*1h&SY+kN@U(Yquxp5;) zyx-E=y1RTZ5BmW2uie;5-rWLD-d#M_jy!XxZ|?ZS`|fX`!w>Mp&r_4RdTKq+JIGHF zAM?!J-`8CtKR#>wczUdzE@;e$F?L|>>EEFe=zqgLPv6)cVtnEo#~u^7oBodd{i(rC zU{1e_Sp6H;_w&s7_|B`5@4!8iT!8(qWyuAwhr4ZzTzWEf_TF#w@iX?^vHtH=?A|@r zjt#!{%Vm$-n&IcNf6tWr{p2UQ>pR-NqQiH`c5;8f5B%CM&;U>E8Oz#h;@f9yk<{n< zS)XX!EwDo0B4J6xC!v<^pYR>u z8P2+3x8LM}wc=2Nt-1K_V1Mg3H}D%8PsGa0sV$zIGwRPJD0>caYJDGdkso(7e!|xb zct4B#ZS5mxou|Ib@Wz3@VcjWnX_I*1iw3UtB>xrr2C5GJV(o3^@YcPw@V-Ih(zt^& z-CzBPHS4MdzZnBHg_Cox^E+_QK<~+f3u0`8uUhuAzdeO@>UYl(r=C>c5ArYchguK* z1^;f0@9}Jtw~W1HY4VC-niw$v3QT>M0v5 zAG!L}JtMo{7T;s-j5Rw_;^X&iYvk0sN2AZ@bA{iU9;=7>f!~00<%4p3S?D+GZ|SjT-0tf;Hd%ym;sTa$EfyTIWpiaq8V&-_Op!J161zMQ`=!asp?j`y|m zAohDv`yhAE2CS9q-~t|}b*^`}zT>x8T79*5(D0ofSTVHn;Bi|y|5@Sz_GkKre~@z* z=gIqj3(GgW0(*3F);vKYmeb!oeAgc2uwvH*x`ECb>z(15_&>{oe%cH6xRC?q0$4Xe zideqAe?@bc=UnhxEPeU8?h&WGs|Wjg{WE%V4Vbgm{_deh-{U?J>)*RR)aCl;dHE-B zSNjfl?*b&%Wk7 z8RyRQ9N++V;Ab?<-4^G}N!)V%86WuW;OA}MZ`h9Y)DE%gMiP^Ig16h!UQTV0b6x{A zOP)G=H(-BJcObUEwQB7*i67*rIqu}UUWkACvSwY0n;NVLki_!({kHCUH-^{N+fdet zC$QfIzD}Hm*gIJ>upM51zuCYyZUL~y&I;lQ+&SSIwMpz7>I=1Mv71r8@ z9l#zbPOfQzaezCyW<76Wj{w>RZ>@P}K7sdb<~SqxB?;jYKY%)qJK1}J1oqZ$zsKSN zt^xH|YCHZ5U)_t?o@xhieFpZC_vht;-n!Z!e%Gk8?qZ)$`2lZkg6d1o9!(D4 zQ~SnxS3j?>^hnTi{d1t-UH_c=3s!%>U-e^${ap!q1U2>^pbmALcYohcG+>>*0{ABC ztaG2=$ot=eAK1*LtvAsp->}sS>t|)XwGFtJd1v?@+5@OFZ{25o!flQF)cx4ae(J1w zqvi#?zx{WO_E5vo58&y!=G1(BP@fsQTR)=JjvZS(^XxOv_=2DKnX9M$uFQ51eb27d z-*pe_BG(Lc&Y#!^_67V52eFzK^kA-#vxlhr5O@3p+IB+@GFLy}tHIl6V$HjAapc@p z?beK-53zQhXv7`Aosf6-M(qZd<~iDaLXG)@Ib!!VZ|)8kx%O}u{o0S5pQ*Y4`r1i; zU>lH|*pbBV09!t;pH=y>>h9K~vE~8H8(U{T^Sf6k_Q;Dcm+PGpeccDnb0K~M{v1TC zKHf`x1NAmo?VnO`_s0nfn9K`mW9xbt;q1!u{pJ0H@ zR+`#jb69Kk+=)MczvorYw({=t0`BlUEvX;xYR!+i0goo(oZqm<-mih;PTm=ly!&-D zeQW--xF(OU&bJM;70csK>>*FxA@Hj|{x^K<%?-5X)O$Dkgn7;toAEK@Zl1W^RfjjjLYtC9{7#=9Wd^Ywre%#LQ z23GwW+;1T8_vhS_9GGB$1fSoe69*V^*3qleL!JHYIlv_Uhd#Cgd-%_)Gl~7aV12{i z;eRm)eg^H!$LtpCJre6X+D9Dt<_`3`-xY4aMJ_JLnIFVzJ9+OI`f&BR;@eAK-4nez zdj`I%w2+CY<+?s~kPF=JVw2cLD9TcJ~eB63ty6 z`^37RH3Qf~zvhrL_WkPZZfobd&nI2Sultg}0PkSE_WHi=u2mE8<{sqi?a`K>SP*Cv zPVLKjeBUXs*2>xI3D@zhan`{1Wctgp)_CWM)&CV=t^FEt2LoK-A0U8n?W#>UwCbPa z>}!qM#MWG_HR{X-P(O3+(9qhygEwHGol~E{@2va*2Dseds5fuD#~L;Ef5;IJd^HI? z`XjNt&&2QQiS>W{GyiSPvA%;Kr?t0S0FS)K+B-FgePRQhsP6>UnoiEQP@`>N;Lk*h z@BNJ3tDQEq@Z$2m=DTklY-zDy^%EATvuDNlzO%E;*(ZbV2k{+PZ`^^uH#~v6nX4Lj z`$aBo!}n)IxSOzc>StRuw`Jwc3-j)=@7Cmron@W<{N6o?53IXZKm8rYSx>G$2l)Lq z6_cOj zTW^N{)0b8AcJ_N-)jJyczUeg zyDso8*Jm}c?xV(@!amNLa2|8|BUcZoAOHQ7x*M@IYC3SX{q50Waq1H8{(jEG@1V2| z|D0QQz_%`dcI1Nfo!GwWu3!?&RUh-fQyh8w`gReJK%MXG+cZ!;od>_)MgD3&_|I+q zyxPyX1M7_HOI;AFGq$(zlqS}WrOBnO8T8)yln2$9Jv|vqt8d@NT=(ek1$5TAr?m~( zSAQCN)T67t;^fg}F0S?`zxxK7;+*Hnk2kK^Ut)a@&h5G8#c9dbTgXqa_cX7cSnKs$ z0?t7n-&surIriC_is2FrFmvhZ=it1D8qY{-LBO{_3%cMgi5;NcJxQN_y$!x`2ks#< zSAEL+7B|3&EK~iRRk>KR%~e14A7}2M&fZCUr?1C34a~%H%v-BQYu>*4(-?ok`QGlY zt^oI>eR9)qCyW~3TkaREb=vBU z@1KJ&F!8M|%?%w$cs)oI0K&3W8p#Wj>dX*em$H2GJ6fAqcdYuh^3D`$?bE>pUws4i z+(-W)XAkx6as@AZ|1MVV9Q23t$a|hxd;bM%y|vD?-`$wpi`ble2c8G^0;m4Kw}+Yr zthMhvxo`6L*1q(???8R#+S55#Vt2k{y+_r+cM$OA^VIoW8M$VKzB=y?7x*6c@Oxw| z-cFr;F2H@N2f5P1X>-Tc603azwU4t#{|-$_A>se zP4xKIdDL9-?a^X=zwU6E=XVYNNlwime<9w-ITv{MGG}jhm{{wz4RjG}M|C04Ia{5+ z*4};>?qKV`;}3Ez0LMM8Q|Ig}m~q~bJsPn%w8?+8o~U2ze^=Fh&d>(D_fT8!rcchD z%q6hbBriPb0n-~Y!ytV`H68`uEcxKswJ18#nNb?5oUNJoo)P>4tPS|g<-hPd@U1$z24DALcl-G6=Jl=f zJ)J4+Y5WBCd%mN^I>%lKCJ1Bt$=TbtjBItR+cO&VzSRYcSoPLS5J1kHdw8t%sHvUl z-OP)`mj_dybF~g^y*lR|AVUQu;V&Py<=XF{&e_gU>+Bbt`2xPR-56heoOocp;|0j8 z50KhuwX?n5#du?L6E08a3ww0z?%`gsd1g4voVw6*>h0nEJ$K}a?eY1j-ETQ!y}7pf zliPXwd#6e6ack>fgSaLC!QP3hme_gnS86&|yzs}4qt&tAH?sepX`Qj3lW*jANnID* z`$;ZK!Hu%pu zKeb@M{n($_nq6N#;10F9;C%JIzjy)lyUu>@Gk~945btBH^XrYtwOI2XaQ5gMm~sl4{N`J_GSIgE8g(yy9cG_M;+lve&Jl{?cWG?8FTh<&ORYPI2fVeSgYU$D z0QYeRzq9T<$t5^$Yi{7XYXQ#Q<`PcWOWUwQ{eCBR4{`GM)9 zvzENQ{J%-;-oFCx`UL9CcQDbaH}4zQ9{QZCapG(QJ_u0nNx184zC3X5PhZwQ)BC?e z+)h4oy|w?lNk7xupJ#Ic63o-TkaK1yenV4v;x~A42V1WVa-N=x_YNc0j@G)*#QK@a zkHi-obES>tKZ!eV2m1x6TC|?)srU1oxqkn6;rCecvX?9{BaSWAC>$N8aA{>gHqJO+WB`d-M998D-Wl_}+cw zLSmh{^~Ua8-$TE1#<$hqK7&}!@6#Z5H}`75*q!u665rp^0P7}jcjxVQs_z+mngjA( z#QOO=(7pk!weBA1I(WzhK08ce=kHjXK#lz>hgI9*ypYJ$Yt{;D zcFR@k{W3@Zzwi3auxAI;7}&$_%9|G5fD6!o`baE0`IhEV9=y;`bNCGe`3JZF->T($ zo9p{}Yy0S1*TDtI%~(IP_8Oo*U#y>_J)2yt8uRKB9A}R`&$qO%IZ;O&>)o7VALkDc z`O&N64{~>N_<`T>ouSU0cPbt6JURFCjhZpgI%9%n4p{dBO)eJ_8^FGIa_T#X_fPLi z-^;Fpcekp?1lk03XEeKqJFBs`@96$dPpH&igZO6<-~h&*e8aYPkSmt6r|;kn2RY|n@$2MyA8Y{o z|8nMbojsj7sN0%|MzCf#a@Hj%9xE5%$IlDzWzN{Y?|h}#AHe%}_z4Dj?=rF0Cg`BW z(rD#{x4R@kus}yqVoR)J=R-!5B;jgwp=>r`Df2} zAZI_{D)IMc;s4FFe**LUd~Y-zjs2XH*hbD1@iA}rv)2Xv0RDdG_vP`vW(;!b>YW42 zz3h8O6KDp$J>@)Q@j7?};_{O8?KtJ%XJ1gWUcul^)J{-tjw_AaZdXtA3uEdJ{gs z>H+6lCf2x}$i#<}m$SyR-}_08b!vu4w7{MI)0g%CyE&<=9QlVmz*DOZ&Yr0S0rc&0 zN8{f!jO}aQxxTTuklWuSt^KyPq4iYnT=nYZ53GIM{SE9M&It0xZ)p5H)qMp)u46m- z39j#Oz@H2I33syQ0_^3y#13L>8h*pK*1U7%?!?Z0V1LoSt-02`h@D{{f1he!*5ARe zXzKS}H0Jc5plVJ($$8vK-+k&cPv#S=cfK_pO#A>1U0M8Le8PMA9u;FER!t|bA6jaw z9_u~b^T+2)+`$0`TIX7$#&=0#@9Occ#vS+`--zFU`Ca$Icc*$sYV6x`?WCrO1ZcD) z&i%3ObUF3%N#2^N1s#82rd?4ZT0u(Q*ENPm-9Ms=Kk#Wdzn}F3;HYhCvZ>s7ru3O?7&tZ`*FIOzpk7+N3MGH zp+3$X(FVTzOmgnfi1%D)Ta&==h5i-oz}ny61Lo;FQ=Ro{JZkK3u7Su^qwev(@?`3b zykpHJm*l+b3}(S!@!jnL0#tu=_UptQ`@lMTVkfb8>c;p9Jie7Z+6i-kF0dYV^~BcU z+v9>pZ3k-9%ZtP}XJ7B@J=8c$`yjsRqfy)7hQQxC?_)f{H?;0od&Tmt>2UT;{H@nN zZfL-hT*tb*bJVD}mwRdLuW!DC8Osm+t6J>ywsui(-vKAwVF-NtcC6=u=0f}(L=?3? zCwxCo`}ljceP=Fh?LRi> zh}}DYoP7qEz<1kkq|W|QhY zy9M09Ugg2a)ze%T#4q@Vc|R}VF67kMbK5{G+#_P~7xMaY>g=z-_b00PrHN1dok4BF z*|P&R-nC{l4t)~7#nNs2Kr`W;Vcy@kxO;{;bugz0m6${vSZy40yHh1KWt@9>j^Y-WvTz zjmJLTZTF5?``NDn`*ct-l>G)hLLPs=g?%=sHt1PU4<_6NPQ5)!%RaGv>eYESwF7kG zCwg}?XN}e#b$N4Zii}8@da4oSb{r{^WcM^9Q)gGh3}aS}xw1 z_NsZ%?%ZQ>>cj*+;jqF!TD8tFH+D`RR3CJC#*L#cW9=~UJ(25;8hwO$dk=W6HQph} z4PxI*K0)d8L+jn`cVIp?K$wZ73-s!Sl>w9NI0xLY6pDn zl|CMQzXR&^Bk+MWk@%bu`~m_^^+V{0zE#8S{nXXHd25|>7*FHtYnpZx#y`MchIr~4Y z!KU%;oRjObyb^oAAXek&;>?b3uha((jef+^O?)+;&1b-&ds25`?OW%x!#w>j#`uw| z_v5zu)}7&;W⪙-(})kBRY6P({trvfBd|;i+lzo$36{MV{Ty0H*9?t_ybgoslz_| zM6B7?UEn)bZ36q;*~6Uo5`pir=Yf7=1N3Nh>i+V?`iUK&&Z+mla`yKv^zGyOOnmd^ zcAcEF{45820Q0pIId#@1HsK$;&&f}?hHu_C(N?{&CqH^Q+d1at#Q+CA<@8_p_LIBF z;|qQHnb?yL{0UcYf$wgKHP-L?*e6iijDh|*(Kr!{uNlTSVtSO`_Y%a~Oe`?xalU)i zPDxH5O#QQ_>fnBLCwlu`*!S_xc7N-BM`Q07eD$QqfUk4f+6`Zux$;Ro@HD6H z+fQPT`VrV%aE`wEz?uu-pB={jeb#d)F72PrTKc%#NMe2Lu5Gb;XsZuByo-GcYC-9+ z*7t4keV2M8Exa{qoe^N@i(IjL$nU%9{|2X~-tf%H*L(iyqsQmjm8a%vEvNeB`W~7) zZs$Ju8STeW7sREp7M_h*wU=|3k?Y^P^f0sfd8f?f*LOJ9&$0X3KfpY#^&Q^3RS)5^t?{V2GQ&5Raq`&eWlY|>vhMf3w1M6p*6AO>nuPP*iAS6> ziTB-{z0c>F-o?*m)(xQV&&AAD8*;qShHq4FCa>nS);UAY z`vrDJ`}i{zZoj#F0Q-zwyI#(j@=Xx$&mf8Y_l^yy_t-z=V)fCgv!`$N1o~R>IAiPT zZ5ny>#@05h`)$2F?!Z5PY7SOzpq*%aqX6zH?Bfh~5YF)XqqU}y6EETpu3GkUW7<|e3FL5Odkz&B^#z!pcnyVVYG1DLC$jm244Jh7a$o}T=i zojVlI{1NN(HkbI`we-Z>2An$kKB%|H6HK@Zxd$jM)_QdX%s!E5m}h+eYejs9ntgY8 z^}DW}x`8I~#Q+Dr^s5fv?}4Ac^AD_=#25PVYRgAI)`j^7kqQhEK6J`K0XKccb+vpm+zbUGo-zO(vTbK*ugoTfyTOulQDtMnYpwP zYwpAjxLqsnai;_w$T_QF?^y5ak#~=}3x1vVkNxEnUi|~V?vAaU*;TG3K%?$J=iH8e zJ@u7?8}dM%GY5F0X;?Y)3DgKR&U2P?B39h+?LE<0TW^Fv$%}!1g%|eiK%IRW{((QR zFYFy;E;OvQYP50c^Ni&~kV~L#U;yvYgFWdtvHraxu?@^z8siK6z{=Y{lRDnfJ50E( z(Yn8yJ8+J7_8!F(+rt{?%Uk2D0P-HK_m%gUbBB&?0>5(j5ex1!%n|D+zWdep3))%c z9@J_hv2Yg<=7>l9^F6C4+^%&8>wfg6zn_J7wNFPo$Ol&4o?V}HJKqFrz3);r`0719 z$ptaJ2Db9tWx_XLzk@pOuU5?jSM=`I&bZ?TzQEk3L}sJEXrzKM04M_unY;O*T(OKf0)`_E54Y`ulwNzWbqo|$pp7w#dC zCg7|edF~6}@dsSfXRof$u}_B1`DZBc#DAlYdmQMs>f|RjuR^Hk>d?#;>bKTFm9XsG^ zUJ?sx)J*({rESHhT5f1<7Lrg}6Fv?Ac`;jF}OSmUA32{d&_da9kE z2Y~mTon>Ehp3lDf{DMYZ%e8ZZn?VfceBb1PU-j_rAK*Co0atI8YbSTxd2`jf!QoHp zdV-TzSM%V8yze0wOEcy1r!l^<`nnf>#@g9Bci6`o_4PU9J14N-*B&A8)u{C@FEkDR ziSL`#X9RWKJ~=gsb%)Ez8;@L^eLB2Hf8s}cyj9B;JLf{)KJI4UpT6Le_kHDmXyELr zuBjpCPHNrNe-@ZQj#$mas&~I{U=V-iLR~a|S=P5R@44ULa@~Dk^Al=DkVEf32e{)P zmiI0q&;0SeboOwMfcM<-Tap7Wd7$Qc^5%rS>}$U|SZjv%uf?o?4~0%HK<0v3e_!x3 zbVu)Gzeapu?deX=@hu0wy<4n(+~p1?u{GZJz^d=Y@ZxcU!$vNB-N8Ju`78OvdhBPf zCwKrg7cF=B{QePZ?(2r0W1pH&9O$hdi4Sl8iM?YRw&v1rCUb$jnhvZv-q)`Bp2(F~ zcLBbIeumlAd&KHHbM32jpN6guwkKe&M|+q9&W>39tn&>9)^~7L;`;{b2lfS=Z|n}{ z9zfq(dj>V$!`i_9ct129&fUDrm0W%P>d0-_OWi!JezPC3vxC?kTKAa7Ac6DjJMgVJ z^n*Nie&Va!yFX6d<+Op$`asif>u1(l`!sCnuzR*YUjrC#&U?)yrY5n%zT0<)JLvFN zY+$Wzw{urG=M8XZ$y=*V`+yVrFRZl_f5x)bbo_7FufX>W;7s@M9qgt59qw0f0oM2~ z?$OCR!@dnWup`#LM;yd&plX;cZ?Ab`f7UI}IavE0XP>w&{jM{Yz}Q-M_zLVF@cunv zVtv;R8h!?MH)g&^`+Q$Bdagcd0_?foqk}sbXq@L!XHQWaR<8QvTfg5ph<%?2+zU9< zx(KW%(1JUffi=gt8q?41HWiM_M?2Js*sSof(Dr^U#t_kB9Pz2rp4 zm-BmI?FAS&5a_)}^)!x?gYWRxyHnts^Ii>WZG)@cT0aN44kkztz}osgaK}cRf%U}F zY)(zhQBN=DxSOYDeF9+f&hm+*89kt zm-p1}flUJN9gQ-QPws~1V`PHfG@Z=g8(?6-ie|GC5;==|?S2D$2i^({ssAm=XS6aPw1UBu$u@c`=8 z+S|8i-~u`@?`QA3*vndTrNdV%I=*NE?Z8@VF6D$LR!-Od|2e|{{={7^x#wvE>%O%c zDSPSvxSKi<)YQS&Jq;2ay9&>UPu~Y?@~p$Ewa*MJsQ1z`OTOc~<9`AE_fmscsPp#( zKPUgYBNHy=sT)|28sF&x+~572?fd)=zJcFBj|F!L{KV?FFPKxCK@3JNP34V2?FwH$ zGdrC9=LzeanOJku0iN5MjYpuEJhfZss zA2V{zthXi?^z~=oIdwO*WyvM*RPM{t*>3>*xkqWSop^%!4iT$!hOzK_P_<}0Z)dH) zCq#U}8vEbjH=K1)pICi0`+Jo^EnMZ$SSz~q#C7ne9DCTa87Fn*JG?pj*GYN>9{3qf z-~X=Mz&h8Xe^sBW-oE3uwC-i^W}c(pL5q(!PU4zPzp6Da=rOP};Hgindrdg=S~+Vv zHi*5WymJFiYfhc<-uVhYa((Ch{+I4kpU16Hu+P{>_1&p@5KlA%_+7h_Q`6M~>l+9y$G&y2_L;eUzjwC%)ZVe`w7zS? ziG#RyBa4q{HI}!ZeayL6hc{mu=!Ad32Yh4B8~z15 z;TwKwiJ$oX*<`MR(j$2q+n;<;WB-8_YK$kobyb7zj{oCM_zle!-rsBe=Q#6?nhF2H z54a#sVi90~ZX7Guv2qhDANIid^Xq>54YUdGKGu5Y%Z)s5a>oWad)dRhv3GMnwdUFh zzw<5EzEy`mzyuwrTRrdp{=fa3fB5<9e`;U;1%LYdi+2C_|IJVQUEpuf{8#6HX#Z3D z$N%#`|L_0D|MkE9oB#5U|MZXl<{$q4@BZ%J{@cI(+kg1`zrX(7-~KOu{=4=c|NkF` zS?KpzbIdilKdoHI=bACKpmrhuLZ|)?)Cl)GK*zUMyL&f$VeEY;*4g@X zFYE(YH_>_2j65yAwcg1cwNK#R>s}}4K7Rohkib4&FvnxQ?#uq)^z;0AARd9>#(ek& zZ=4|TjpOvazdLVDhd<8vPVGQjb@(0Mx;!~++sT&(zQxkWUx7V32y**f%mo-|?>A~V z-ow83_h;9B&a+3untxcM&s={Vo#i?3qbsxq+|Fz8?E#;;4=FUG(oTSNJ;c z7hsQ?2k&>bp$R!Kz#Yz>6W`sP6~sy0u>(BOsIQa!Dm>^*$9>cVXtCxx+wa2uKA1PB zM&Et|)IE*q5yTH_;)K4(ySDt;!JRv?z1+dNsSj#@xHn+WAR{E$ ztFg~PpDXnZ&YV9J&h$9T*ms)2KNJ3$H@~gAAogvZXWrc|VCKTWUdRu0YW219?bKRp z{44f52vFyC?hsfwH4k&3<9m)j^GmGV4taORx`XE~7wfJw*LzhhzW1)r20zi1f9ieL zz_(@qXNXAf$**44z(1+6PPmJ@uJ-g3-}?QtZ@jl>5~w+Fv))Yltv z$Ew46{4=xz>(mPGYd>`f8qlAoHTLvekoPV9S+3di?r;OBvDfZ*Fw0z>#OE!7IaK^?Ri3+^v=y)~uB+GE$&o#7gsvk&Vy>g!&_euiNTMxNG~`&1o%z%WutWS|4?cEH(J3^;p*{B89yudZWXz`Q-1 zK6C74oWyG7t@X3=xR0~y-`T_Yj=##~$KOBdP0(~WweI0(a0l)lz`6GS_4FG@T_+yk z3GCTU+sP&I#5U~jo5%NE?cHu`<^}5x^__q&o_QBE_Su+8D`Mvi?7+UEv48urc<L%vI{Xv=ss@-3xWu}@ew#r8Rz#zk|nZeUHrF@e|w7UC4F( zC)S)bft5Gju%~_s{J_5GQ*TM!@I8Bn5%@rzn4p0H{64so$GQ3u6;@B*Qmv=-~f}yVx(_<*Gk<`?^!Y`*)#%b^albZk~Gi5v`83e*di*r!}EgU4v7b zVE4JgnGd+Y+QZMx9o0FjW5?;YPd08V_d*+6!+ZI@&bKbrCwXG$)c2urV%7Nla_$2= z;Y0`Ssy}kQU&D5|1DN{?^33-eK6(3itP9}oPjmxVYfb%(VLjg6-JiErYi~LGtM?v$ z4({gf0cuCAw^Q=~jW|wwVEydH@Axy^#Vgq!(}1PHX&91A44I{r5e;P#eVd@jcAh zf1=kW)}5@c-F8now57i--6W2aQ+u}`bybVqZ*YP4b4avWcP|Zbbk4PxKR6Zx@=a}Rnr*E{_V zoZ~TW3CR0?9ULcTj|r~C>PCF}xxqJQt+Ty@=-{RHw*LKkwD-H?a7fCs+EU5wq$H@Xu? ze)?|3sfADcI=Oan_5|*7kawRJAHI2a`oF)dyV^gA#s3TE+(h#mwzT-}Yp*w8Kld8& zLXC05wl9nOmw#S=$Cz;5spo><&wqs`jG61nSnzw<$bSdkf51D-&m-`=F&g>W1K#{3 zme>FNCtpJ40l6+mrq$!Dk* z>s|VpSNAXmLGF&m8vlMLr+)zr+=2DJnZNh_aqB=<1J>ersm z1Qnk-cd*7D-lt|k)rs{kU&&p_+2aA+$vWrN&h|v>G5&mvXRtEb!1i2w1=b$k!{bhV zr;P3E9v5J}+5s9lPmgs+dro{|z4!k0tn)73OWZ-AX}~^eCwRcu{ToLOoHG-sQ`5kR z#p}yeFKmY&SRwzkA6(c=pZoP2IXSF3zZXA~^Bvvw|0ClKVI*6Yb-e@w4tSuz1qK%s z*x-T#3Or!&q#d<&zySsYE_mR90tXlte83=OaJ6MUP_is*1ItQT1|y?}o1K_OgRF)F z6Am1>p}#LOSI?a5)cHFi{`ey@GS{l^o|%1br_Q?-gM8a@`Q)|s@E)8^d<3P#2Ku|P zKH3ImVqe$yH#uuAYT%70IG&TWPix6p=WO=6fsSr$R4w_VzQOAs??vKwk$`inAH?rG z^LKhacJK4+_y1qc4W08i%b?Es#6JW3e9;1DQX@(i>nv)XSI!7JToB6zeqyZ?lenp| zzd6uz&3T45f5Q%}I8hVWE~s%C&)lB@eD#TK;03+Q*nORUVC^&Y;T!SX)^l`X&#q2= znBQpT& z<`cdXn`>~|y2pcuvfqR^fAqSL*MDMt#;$=aK30ztvG0!e?HQ{-w)zvkkn6cLgth99 z5;s;a^NH_y0y}^eq21~)+?wnlyN_-f2`5ZhYo9yfCGXKcd?HTHR4mz5hp z%?U&<|LD<(LoFKjQe$uVW7S>47l+n!_gGoaB*Hk??Dm?*f;w|O$$>-L;k@&Jo50_% zd?yCJI=MS}d&tWfo1bv@HP`jE+@m?qX|DDm9y`BI z<~VC&-J@VX_V(@`!0!>&KNfGl8{8jYY|ea#cb%`xziDGXQ~Cw|Z`62ppGoQACK$Qm z6E*gI0)KDlppkc3@7@F3H`FJ-_ukawU$AN}pVoOjv$_VSA6U;kvBKEC>gyfFcQ$jz ziQn*>)|HDJ*dOcVd#pYKYtC5?y}c?A_h2vY;)XwS?c+xJUN3OI3*KSM<|tUbC^%^uAckvd9D?% z->-1px5d)g_hNsz4rg!Y$&GokNBi8p8YdnK{5B#=c>&$s4=7ZQC_bV~JGr0HE2i6L;!a6laU6N}7 zSPy#m>|5^N6JqH+t38vrv|A&uw!_=A0rv`U$9F&TH+*aBPKbwEY{xpgdpeijtIn_R z~+BBxGb*|#wC-A}Lil3`LUg@ooA6WYz{k%i#wBLZw zL{0sD9{U~7yQ@j00>A8z)w^mAYK((kY9@Z&2{u97v3_j0=NWx1xm>-}9JSWn(cRJg z<#j!;`uDD_vBu|rL(}jlzRNuZ(Dq#Yv;$0IvCpZd5u0=0+8avFRWW_lP0;bJA3IL0 z+N$*)^nsjbtcCCD!584()(82SD?auPcx$x@CNLhL?i%~DXQ)}|l}C>G5A6v}1IG-< ze7*;+nd|{BXg;fD9hw&FJ&V_+QR^&qXZY%j>%OUdfZsrp8}JRQ);fJxk23Yl_0`6P z{ajk>&1=^JM{=!*;8hvN!pvBT?h4t1{4r-wN72JWj;~lWa zKo?*nLM_k=>qY4Qz17`$Z_PURgdgAw@x*%0nQJeidT(k^YL8w?zQyXPwa%Krb};aN ztWoo%?_iHc&iV_!dUMt|k8{c2fbp?>S3PIsnl*MlW7XDOFyAC!J-4>*>F_z#W`&KVBF8kNK*rCtz8R$FXU45>({`|UnH*)@a&&1lx^SJCivfSbxyVm~J z2)~CNnn4`Ne%J3kyaRdfBgt9YiQQx3s}JH%?5gLj6AN{=om1ztPOQBD9@%r%pW)Bw z)EsE|gM4D$tL_oaB-VGizddSSYy;lWz>ZjX&uuU7+&TXM4}3Z2vA)4s<2mI$cM!|} z%P*^T!h3Fi54FY_N&|1rF;mAsqf0qZxm?eWs8;66Id{Z5pNYCJIO`{x3wurSevflj zIp=pS@66f`S9Q!I=j;RCGpg^z0UDgVy{*qAk3IU>OSsQ*KVQffk9D8AA8exYyj{z@ z1J1fh?2;eA8kgFci;I8&#sS>F(?9WDax?bxqPMGRa`hPjfj>Y{r*Gal8a9Ee=YpKt zW1sUYN%kyYMxXbc{mJm<2H_2t7g&^Kn_lRtl2eHylO$h6+G zKP&3Ksc|o#UoCv&_r8HoY+k*4nh($f%Y2K)StrzAK*B$<&g9I2O>Ghb;|>~_xqi=n zpudBVIlM#r+fS&o&a+PZ6Zwz>?$@#QNMP-4TXoi-f%oA|4ZK_WRXw%^bAl%|<^wy4 zkE_|8!Kz*&3v_OIuJZQw=i zb@gue37qK;yazuIbsuOmRv+t~rTWT&k&BzyM(m8*Cw+WJYo7h=?Csu@{NX#F7;wVg zKxm>;f zxbE+R!UUJBUbGb{{$}ZU$Ft-a?PQ}Io{;4J}2ip6Z=ke@&%q%CvUB>dp&^5 zzV%K-YyZz{YyS=S_cYF+&Ypf2)J!xZSC582;r|J(XD%&Vf^YD5Z~ z;F@*#<~|cwU*`k~H)GBCP>X-Uo`L%Nb!qIcwmuKYoV&xh&iKx4UB=pLV2^jR=hk8; z@EO_H^#?W9I_JpML(L%m2%g}7puc_1f5d--JLYNd=j*;_>?H2!8t{Cd;U3g@{{J28 zf0rAd*1iKy+=zoZYyL0t4d2*49V?$`e1Fc^MvWT#JEvN4($6)IHQu##f1&OJc;L%T za^{cs!J7M(et%zk$`81+T&$iqeD_e>uupUDkN@S<(g)VFSktq$e|x0|sQX24&q>aA zqhsyW)eH7^{swPO?Z5_*zX0{8>q5f+&#!3Z$hn7{aE(}SZfEr@Z)m)`PVOsyCO&8Q z53lg}?})X(Ix&*Hshjxr@thq_G<;+G8OP=>S8}oLy53phLC*cuBycV{wWYzgW`LHv z7df@URnLgKiXZ@LyMBUCE3Yk| zZCT&7Nxgdo7{LAPQ9fsNnNt_wL5;oD-0>4`fUfU)fncXG!J-CPo56Ex}vwtAxTerJ*cfnQquiAKL|-}$W%>@kmd&(^U)PTqap zLv1bc_7e6ugOU69baMJfUFrkxzCp^1Yy;K^^Xi;oVkcaH26|uxOI~CU zV2l5>YW$g(aNbMD*8QsiGq9|Bp5{5P+7rHrFuUfEH*byBC1;N~Cw@b+(kRRadT&_9E& z51dapo3U%gxUD)@5x2>gDSmu>RcBmvfFLSTol>wCn0uePYcu z$OV2ato1b`zV^5qeRYR@QJ3tsxf8t`aU;IyV~_K@?@=dj>~bz^2C&Znfv?8D8+iMK z59}@PnI^V@vF-8xIH#O@%c%>{qt)v8_P3XL=b6N3>YOLd19c7D;9S;pYtYME`=jnR za>76J1$i~XKi7Zzw6qzkkNe2w*3>L;)e}ts_vu*AZJ*L$)jZJ}Uv>?Onx*Pu?J>}L zcS-$#JLa}7$a#l1;2bZ&Std~TuBXmE9jI61{%vzJa6#Q2d3om@*dFW8p1zANHSTAB zeS3Yus`Wl5w)Qq=xsQ8}tt}E=^>BvR_gslp-*WZyjP`Oi?af$_S0nG7hwnVD8LQ`s z_%oOZ3E~RKIa4R@_^$W6K>IszUh}o^YsJ=K?e~DUmwVi=>vQE^gWP~mFhT9Ze%6?i zQ|CL@@%??t+zrlE=gk#6N8r~zVD(3=-ui)GE5FZXy>+f1>wGry=DjEH!qt;6_!(G` z35T_|`V-4r*YK++=czly9!srt6L^*rIM#8$_AzH4_w>D)`1bxt>^pXTU40_ZkP}@S z%TBDh$W`ZAI*cf~_WbPUOard&1m7NQW5TO-F8##!JofD7K){)M zzdKxeCS3Il2v+F>hZ#CHa@96`wcf{Zr|ywp!g~+K`f4W_pu>fh*+#C|-WO2!Oit|? zUB{LNop<0qLcM)M#L^jeki^cE_~z|tzdJSNtrgaZiZLOV|E`wUK6Nj#e($NC`18o& zn;+XYR>wDPxq5!4_rU+@74Fmb&z{fM`!Z*bhz0$I?;SkA1vqDE^84NsjWyPLFXa>0 zUBVe#qpt3bKE9*JTtl9CCfqB)_wPH+_n@BmSGnt+@#odkzuWx??`-Bf`i9=0G4|7c z+j;$tZ?DMp_ksSsYQTTn-0iyV!MPKDfDdRU_J6_J%Y8D}tY>uY^91_-ZhyzO|G(k; z?z#S^jn&hiK@E+2{TaXZO8VkE&&0RhUN>~UpWdB&yin6)?f(Ve?|p;VJ2>Gt>g;EK zpVJ`j*bA|`n?5iXS~#D5&$Wm9o{3%c_oc6^&iRjX-|=6pVJ>UWSHxOy$9LAm*M7yG zz)Uo;e*Om5eSH2m(D46CPkT7u4c~j|vF`p4HG-Vycz`E+grVJ zcs~IKn$Or{{pFSUfPK9Gf6+NQ@4}y7>V{f;=MH?~EcQBsZ@@dTSBL)zj9Ab00T=i_Bk$gM{v*D9 z-M3@w`yIKz!TW!QnE2k0XV2Js|Ga#Ai%zX~@Pf{H8os@q-MQ><-=}$aH4$)}VPXT` z_u1SReET%`%BiE)8D_FSR(Rraw zle)J3fvr8aB~$0Fn)rUspx=cVCS3ag)(+w>nDvCt<^33EuKVg(wXO%b8+h5)8IB%- zKf#Gy1NBZ2H^I7H*Isum{nT4)jnMyU9q;%X z4ED8`*1d(kaH%`KKf6x`InZDq9SMF{eSli(1nUb^CWxpr+kqc)v zaMcrAEB6^=&x^I6x(3t=zpp#bf5ry|CtuHs=T`GEo4p?B?5FQOH~iamJx}W< zymi9+RByd})zDAQ{pt*xH>X~l!5vtmtz2S1tuwg4^M57h3<0cb*eAIL)F!br9p^me za9;PW*$y4r6aK%xuHP^Gy~gO8n=t3KD>o~6UZb!jeGYt_k} zH=I*A(;#P!&=(U#wpxLo_`b`&^EDT-bEzBjc>wn^eqx<<;QLNBaO_+6k@O~)z+U!t zFXyqBeGlDne;utl_fR9R&OZIq`h9rlv#)qVqyFd__zyVig|&Ww*ri zKCvCG_3kNet$dLG25zA8>^tGl#2>-HcNY7cfOFn}&nUJvwl;yhy(h4D6Pt5i_jmbO z>fkeQuLE*-;5;v%*1Pn;R{vPJ>VuzXoHfA{Kd=|9J+%Sc`@I%lZO?Tc=l1^Iz<{&n zgzpUh;g@yxCI_bW)9NSp&#$Yt&O@#C2CubOi-iYV#njq&z`MV5O>D>deWAmvHK*P? zw}&(OeftmKdmC`(?PCvT31Xks8B`B)4{~Y~bfE6I`@kRME8hGa>$+ggziD&r<6ipe zJ9dI6d3$S}BjL{2JLqsvtmjfI@7|3U0>Z{LZdhjTl#=L+;M*b7{X)u&@KSN;xv0!Pn4>$z^&P*3kWzDtdBOf4AT zf1|(4y&HA*w&uUWn;*pgfs@Z}4`Qv{G!N9)U1Fc)CiaBxjGsI2490#Ieb?oUA0Y7U zX`SzwInjun=YM)-7W{5ZzQuZm-eU*$P$%5STJ_Fu-@t$3-@%F4a|YI>-kI$i*L9{F zwk3emny-cLULTmJ^2B~W9dIKSe5bu*IrslhzqIy00{6Mep*vxPJpvnW?me)=`TCKA z{{zi&znyrZxxl&noQ&KvZqDzu_OZr&Cizb6pLuF7V{-1}jPBjYt7|}eqwWp(=aD29 zYCHA;?;1fF{%3XM5^GQSGk(j3z}Eb+^1q{1e@Ek8shw)cC-Bdc?w@ex>pb{=t~%D) zzrlC3&g47;-`Jk&y%+av#0~!FZ+*Z$$XOG{Sl0#K9yf9>ZDWBYh*naBdthG)X-~nfick+UNHuiawv(9tX-vz+8Sbx@j zzz6=XYO&Vc;p9(X5Ste**K_Y!pS3@SgS@?+%{y`bN!~p_kiUFdnkQDS?iXvn#5O@r zoqPQ=H9eO9Mr`i*KB8V-l0RRu_*%J#z&988^%=#tr}eIXqW8=8)f=n*2GrR9zk|Eu z5B$U)@9HFO!1-GMR!_bF_4={ly}3{Ai=FV7A2sCd6R~CvWBlW88d`JK6^Ebr4V`<- zH}C~r5)XWBU|pU~?uI>s0P5?`u;%J}3)Z>RX~h8Qt?O!I>$kNw@Le6v9b^(**Ijvb z`%SnfR=sh@I-}>ASod-{liDU?`{Zu$C+ufnKENPv&*Po0e(;~sc5H%(yPb11lFI*7xuU z*Eeq0)!SLlXtefL@4Z?3zzX|zV6Hy@Xq}}6|D#`G6Kjo}Guf+RZSENBb3DU2lN#r- z@5Bo0PsH}WY-=3Xg**O0=e!NhJq~^Kf*bT0@P2lj#lBr%OKjalbKYFQ*Mj%FgIKM7 z?BDPo*c-T$e*yJ;;ZOK-@oD+iikcbjBbdYqUhuW%GM06gJF&ezQ;@e;0{zQ%_3h{f zT*C&i{;&<+IgC4Sc~0X7oHem$>sih)!I>KK9p8Pz7~f^CN4e{+bM+W->da{ed_4=S zwa#I${)u=mwOuVX^%D!!cy8}VzJX(Idrmao)r7xazw156Aog8+%A@Png0I9c-~k%^ z0QRgIp&}NS@Q+UrSh)sLn@evzz%)i@kAZ)(@5cMN$tO8$0;ubtwEEPV(Wd?GrZ+g}mvfKQQhSCo_Rd^#_Hd0@xt6Q9{av;9a_ejDOZ^@6T=m|gv(~(^`nZq1 zk9zF|&2eUHV{0dVqhG@&s6J|`nX#UcT9?`<+5x_5WA)Lx?61Z;xo>FfQ**%A-DA(z zVk6d_z3+>}9yJ5LeX4d_dt1|i%h>xY&V7h$KWy8OQ{#L32j~JW!HBJ=R{b%jGYsMh z)VSn=TF);|AOkK*;&=Q8v~ox7*qrgiS6lA_d#baqt8zN|9p;=#UGcHv)RK!GSC4G} zu&f!N2hi8jt6}RoFvsJ%W~n~JE_JmR#xquLbk=6BJ>>&mUB>#I%)UN1zh515qo_Np zn7lc4J_mcrKk)rs)qD93*!yzbzgM7j5AVynSMR9 z-t}D>@*wbC-fhSC8E9va0}WVrhN~EAzMCE>jT;+aRSdCS2VHq%v?W<-6!zn?d$$>U(t6sXE$~}m(R2Gx%PFg5)3pi(sT-JLW`10Bo{0~f^FPET+pwCf{FE4_=?p3pTh5*<>u6T^G z_vQ=<)C6#r3$eY`A9YeoHLL9ROYa@iw+yl_8f zaQ^~Ed)!gg5nD5{wFiEH(&-02Yvj^dD@N?z@ZOC&IrkODp5gP(bz-6R&}6Q@1KWWs zk}tqo`7_pB)pH(~z10kCaS|B@$iiy7H})L*okfX3ra% z59p7+El9$pd3v>6&;CTLFBe*BD-SI{0|Pp~`2ukOYwP{M_qo@zXzPsrovT_uBZ)O< zFLheybY;*3G@#xYCprBAF8IF!b#nG|-$`tp`37$spaEyP!#{w}(i%0wnm^zseh6aE zD{ovY_zEhg&pPg>{M^5H(#Lx~dJOn3V)r6f_sKl=OVDyb-2}b^gFeQd#eL*`m#htZ zYX>NQ$7<_6xKbyGJ*WNys86)^8`v}XhLtm?#=hDcIdw_w>RIOfY;WWPuE)C5^3iJ> zIoF*&<~>X1>gUH_;a^&MYc*weGt%9RT(J!h=<`bOSb@4)-gciFSYvJLCpzoAt( z$eFJdYnIE-$%_E4r@UvN-aGeu(#W-s^VKZy-rob%du*PZnE0-Y^-OyjtU0O{-+m`* zrgn&wnQ$@Xjq7aoqR=soBXMhCi+v`4Stk}ANu4CQ5{*Ie7c#Z;`Gq}$Ks57=Mf&I197~egA1@Z%~ zdN50)#@SqFY!4uh?|r-g`~LY=1>cYH1I}J8*S}rZ(|!Je#(O-GyP1d6x9`ALf5De? zkEa%>`-&aLXnvy4#H#V%MiSq8_r9qkzQa3*IW_hYx9j&mlU#QGg#E8*zF_U+UHflf z7qn-6c(ron!~ehd{u$8tci?&5>l;?x#5d=Rp3nG>RaY9UvwQ<;W~|?j{kN2X_5t6} zbbOzIwIkNgAN6&=@N%yINS$Xk=kM~>C)U5oyPv%-zP~`P@g{2RQc+dVD25lS|{f*lH9yB;-@mV%> zNo;@Zz`D<$enE!qpd~%7YhU+vU*EmN5A2IN&t;6R1^+G$e~~BtFZk{)-^|17yT3i^ zF0k(FzyG;sf)lXs9ZsER?^tWSH_!4azFY4vDm2B`CCbIo!_)9^iGy&r^X1Jo?osqek8GFF{?IfHc*U%k4f%~h+$ zc?O)Q9&p+dbsfJ6G}fufT)m8A$DaRzUaht2;s>A1+5@g?0v)rdH5L_bu6#K68o;>v z;WzMLzbAMD#^$sg3^2ECdp()W=i@#fvFe2H$1%qsmvG|Pzioe#)AxCqYsAH2>(1Z< zUl^OKd&pMW&`mP>)5xK1IW#S*t4R&eZ0%WY%8T93>f1^%+0^f(+zIK2P;WglWdjK`oPB1|AgPUkvN3O^6BbQa%fcv;#7tG_GJKK?0fB5Eu8hh)H>&_8B z@Dtmpw}U0QKr63{f*+2kh%*P@SWIasuc}eL#a$$7Rx+=-p< za$Wi)c3ET2z3N`^8yMTRT>iu+=+v99IrqH<@$np!KCYg8iPamg;z=9{u+Gu3YDVJk zd-s0L*IBUMpLy58dB?wl&s#f#B+wq`aewPX-6OU~tGm8fw|RBOzirslnki!G+K%7I zeFI7C-R8~*R;~m6F7TZ_unnjcwURx^wb**@QOzV4jocu19q0GF=1$lK$`X6t8+z-l znIJ{9;sU=L1N*kvdl>I!-vNTYf$heL4}8wvx3TA+!2Q(tZrRhC2I}`QIW!NTy?~iK zuPZai;j2$z&Dd7nvxFM7<~-xk=f~%&ul`Bi{`TquUf-GR@z>Y&oI08b?>s)|%r!%U zGiRUM=Bu~bD`3qm^ZxzE$Yko$GZN1Ry{p4NamUyua?}=4gU-#`~wVd0zF&ZgEIy7E^VeB zE7!j6X}!5S_Dt>3TUq&NCi$6bMzzPR_UXjh!%zL#F}02E{@kN;9cLT4a*6eOu5kkq7!dd!JAtbwUqmcUjY~d3Ev#{9-9NtPI`f(H zT;{|eUq04n&;=TMS3SPVni)9suy^3AuYC?nzgom+a6HdU>SN7eUF}D|20x6k_BnEL zz8}V};&WT;0Xyu&o4OBnk0X&0!s>Yc0-Y1ve z0}W8~yS>2NT%9*tE&I4fi?zSj8gcBAx%`HG*WnL&&lOl;ALpnQ_{7!Yy3S>v&+dh_ zk5;bsH;gr7pr7z+2I$lb;Is96Q^TLY-d#_P~JDjx_XCn3)zUSdRk2N>)7yQ7t=fHRG7yRl$*>OJm9CMV9 z&ixwN6X^7?_K9y#*TDC^%_N8KKIY~1Yb9TIJcAstk3DN{(#9R$Gu7T&V2|S4(i@8b z)CMp=uoK+q*@)lIL0@y~CusPm>(V~7a2?JSv3NBd-+LL@Gd4cKIU~?_b-CCU{!aWb zcU^UjTqZu8pKEc^UeSUe=PZHFJ}&1zKJSTEex@0G_OPd10_QxIb>8QID;}hbK+WB^Pl#e5c_8$zt36k&x<$E z;f(_{!7^^K_hQe$pKuAZ=4x;HyJr%MC;dcZtL5(j9e>2`g*{tCPgev!aL$99#3`36 zUp}Gnp&fzGz7wlo^~B{9x@W}o-0Q*YXQ1`jxktc{SkK5F zH*)S{Z|mv|`0Df@)QAhH83QYl{I(U7S08ZxPJ8)84C?_x?)r&JXVo|-Y&+GSjIuMf2a*O z&q#c}u6`Zg&ru_<#(A}#=Y8Qj&xF6o$2#W|?hoL6XZT|u?E}s`u$TG{?5+P1>uf_l zSM38X@SV?Il_z%JhBbGBZ^RSdK1pm(@5FlR2D#Fa_j}ME2ak>Gnvv@DJsyp{oi(uo z%^kS(?P)*Z%olU9_V>;%S7LHE?8N?t{R$swoc}l{s(PmWTyTEz1RcaDv-r8{>?3E~ zWAEAEjOp3QO?c>yw7 zIs3W4>xmUrm&eK^l&d~1VT2kxb&VUu{^*Ie5_ z!K*to>K>a5wBp!npi}RD*1YVT`_?lcwqAddcdt7*0eik%{95pd?t&G8?;YI8hroA# z&(=Z9@9)D6{&(;I>O9lH3bpRtMFQq&aP~JR{QS4q@Aq{)>%)?LpCNPaL%)fC^lJ9^ zIRqSQ#_Fx!KF;oPUpbdr<1V(}ScAC5I+J|M^*q&kH>E#%8spdA==8k@KfAvB@_yEn z+UnKzp4NbXpVZya)V|m{3)Wn}60@o{aRe&^jn@9!0Bz&&?V&E=J2>w?9qtX^IX!O>3w7$Po5rBU z+TYrP8*pl!-S^Et=B$gIyU1;gGdA)8x@eV=IE&mME<0^a9w0`dv( zUf$JR3wkt2@WUFp&TIZxkg<}16%8(c`6-8O*aY@=HfK0&$M-C12Jn8hE%8C%%ejs_ ztz4`+_i_G_Kh&YO#{IOOv7sp~exP-EmjiwV0Z?bonuNEfdE>S2yub^gC+aa|s6@z|$gOf(?C5V2~k^Q;Xn;7<6L9ZxuQ2h8hNzOk+#7niZ- zu%?63z&+3ev9+Rup=Dp6@fT39%pZ2}Lt8q5pqZjaF&!e9;_HQ8Zy;tYx z`1TeqYa6IIaKRw9uAxu-j;;8*eL`~s`2EHAy7Ia8_K)?fpF5w#_Bn&wMpq9XLVw52 zg|BYJ;(aa;@B(VVH(-skOzg>=@s+)>?iJVpdbG-=_EYDeQ!MysrPA*8C05-VNVg!d^*S z9CN8T>f6=_`3cmC$W=SES7IRFK?Cw?iqnDyWe2IhmWS= zH*3&%KkDT!Sf82i-Hh*j{mJB7x+$GE7l!& zN8jX$Yo=WH^x+i_-@GtqfA^c%W()$~drhqSn77W@8uJs|;Ez7_Oo<1%pw>QO63f}g zdvjUud=Fsn&p_Y1`Ip!A{i-E3@U=g@I?<>-o=KpwR|2hC{Y>_HUHhs1aoe|kotwZy~e!PC)+n%D{8|;VI z)$6jkPEDN+`?&5o$u;~D>;3ph?0=iki5nQi4-nuhXrT5;&VOBVtNUh-W&aVYzj=H6 z-+j+q&n3vIwVzsRr!n~fCh>(DHOHOTJjUe3NuSz@?>q(k9_!g&;NrUac%Cqa{&?

pu3fciIDPV8`aH_Z-fp#@~;AMWc3tT3Gvb&=OwDQsX}E(eR)6?%ja*)8X7_ z0y*P^AJ_nXegc^Do%X-ux0m<}zI6$@_<7CKcfE1I&sej2{xf}UzjK+i zy`K|1>3IeTZ%%!{zhFGsvO*VW5;?&$qDg8?6C{)YdA6OCAn zdp4}Pswd}u6aJrnA;*eXcWR%SfqV#J&m?a@pV0^UB=H;h22Q|!hxT#({!FGdXcOx! zPjvd;r?v^!QF|t@&VK%$GJ<5TS#NMbd}zBqoM&@S@5%SVynU>5UTxI`eL!(qczye* znLsVC*i{2{mZ8@Xyb_E`2(No5=g@|JHiCs7STMA? z`UiG^TCr-ZOE5RoJlxkiaZch6v@I6bvAz%5j@ETIrl!saSNdFik3JFX1?PKYuLLK2 z^Dg`SxQ}{uBQeQI4cnalfDf#FwM|^tY!Tn*@9=&P z{|fFvU6NPpJx+WzPja`-Ifp$4Ts;%wGZ^SZBPaa5xMLG&hkmyDS*pGf-{?_$anIh} z8*o2oc|Rk$s=KbwnRV`S)S7d?hV>j5;F%uR-2*$x*-y;??(ef{aL%PQe~~Bl_o%=I zocr`#2=ZrR?#BEH{{^_`(YwiC(bzwiPQ6RL%fF*K-i?{7XXOI^05x~@#O^h6-BH!S zC$Y61KeS*3mb$U?)~R{yKHjnCl)I@R-+;cH`T>Ht5qEu1d(zi?bIvq}^_&f>=7yhY z(b;R@*Iw8L5}j7=Y)!8I_PJkIt@F9BtH6a0Pst~KFa;Oy;k zUi}C5D}2W4VXbxctUZaJ_ybP9y{>b9z`EZBOmjg$e7S+|GS*j9^_;&E+b6N}x@zm( z_~z7Ebz~M?ALn2ABBu7S8M0N)J_fV)f_Obsz3+O@IlG z_r&wc4Ux;*vkT6fSmEli`6LiQfBLF(&-yH4J%`W8o|o&Y8M*9>dH4k0$%Ho+k?Zs2 zc^Y-r+INyOugzTh4Se@?pRP|Y>x_d~xMym?Lmu2gBksAZInSUK_Eh6O6>qMwoaM2# ze|ud$C%S;sI)f|YyWgns&vX9WfqVnrfzQJ;p70YmPt}pHm|madnWxJEm!GkSRd0=Z zH|zlFT+ShmyDt)*b_6C1E=SJ+Z zwt<=zYwdB@FK8~iW(Io^R}NcxIQxYb*i$R#DsApK_8GA5X}wzSVql%2Z7c8DM8bRj z=09WY?My*D1FJ?$Jb*o(>i)j^H}c-WfcHGcH>^J|UDnEVtouCh-B;gzVE9u%y(ugus5s;2V^Nx<$?)dUPM|0j?BksWc z?0LX^+BcToafVb9{bikk?L^vxB+zoSUcg> z+si(QJ$_z#wj|d;0CUBmdjs<3t+i(0H~cfc`_|u05__(JRd&5k`s?Ms~Gj(=BoU-h@0f5SSn+6PE-)lciY zLe9NPb6w|hrXWAn0OvTYx&ap(4f_XuzkwT&?;wHaoyPVj=lZ}c)c}{3< z)M*oq&%v|0sz1Ix12k~It{!(a#50%PS~1{U_HZBfkW0{W&Ek1Iqc-402Q8Og?HB5& zF_^@Obw>MzyuIPgi<))Eli2giH8|^C{+)sE=0iT$OapuLiunHA?0N9V8Ja8j-*s7Q z&-j6r_h(n)8(ZVOmKOe3VE+V-d?VJnhZ=eFwI{XjXN9w;>q3qG1m5q!+Q-@njP0R+ zqUYFL+no6X&U~Cr{jsOLCixD2!`I%m%+uttYOS|tz|9SRHa&p*s;j>0$=lahz4iJ& znrNF6h_+(m!(QeZy!qP| zi|;zY`^%-hS_ve#8R)zA+8LigOb)A9OcO{9B`bmv_T<$Rf2TjX$R?pxa*!Kq1iU1unt#!Hflh;>kpSrJ~ z*Ng!-$*ULk&LhTu0yUj_XSY`~Pra)LU&veMd~aAa@^|#E8S9Mop0URuPN1fK-cjS; z{tWbc(>Htl0?0q?fwi}q!0#K*c2NuDZ`;3P-++1dJZ-+}5-qsEoz3IhJGX7I_OYk? z$!jP4fD7a6?um6@@5QA?sIOUbo!goTj`Len+RdBueQ@Tl!1uxLlg`-C9(P?aTH}GW zuV+xB-@y45AI|us1&M#hig&F(t7FB;mA5Wq^|j7f%-KtwGXyX`{=D(0U&v9T*1LED zYh688&K{m|VjD=HUwyf+h*s+jjXC>VfHuH=;~(z!(|UgfdS`CL?xD4Z^R(c9_*yj` z-}$xP@i2zB#$|lq#d=q+Gfb?x8S8uX3pI6CV&_Y8_8Oq$H<0=}CqLmT&ok`iz<>*^ z>x{N)V{ws?pz?47RE<7-Vy&5AEwT9a zt~{~J8sWShILDOxdG(S%^m1xv?3s)~ZoTJpe=!8St4Eo2K3sq9he&G80rmAf=;v}D zm;Gn%=ZCeITF-9Xq4QpEz_@0|Hjuz)bpiD}>1VCHI(w)Kpf5l4S=!Rj*V)|HIJM-| z+p9EKmvgjS`Q!bGSp5Tjn)`cu-)oYqwd?JtEKdCzC;>RIOfo{nc=uW%Oj zw`T|TJMPT7Ay2Qe=sb75kK`I!>uy-?#R67be{zFFAm=293$2{>7KQS?WL}7H0t8;(M|HM;|}ccKqDHydgrQo zE&JFrz|9=i{EW4ab6H~_q29c!^!U{)_Z;MN`z=-vd;7es3%FXj^+BW${QuqE(do}v zXX*G)_V2{%9{A2Qh-(h~(vzDzUo$d?e4Pc`V)fM8t7@?cVne@yzT@NE2T!)1{m50X zt^17C$Jy+g@b2yVVI1&s>PNsa+p)iK1MX4zTyfWeZcpq4W25S0@#Yek58!e?dy2C& zfDf;$Uaj@cHt>IfFQ1wLXRY~3t^Iso@8CqPf#2|N`sC#XHud4vzn^Vu!I#G9o zEuLHl*85!5Kd);x6g9_#xWkJUSmtxpIjC`edya(E2Q9I??lYR>Oy&dpSS#nVORPQ2 zrxpx+^DcAp>N=3uZ@Fs9(q|G6cyTQEv$oy`=BkB_*gmEwUw}R3&s*pHcj5!yTlw^~ zMkE;U)_ZR2)%d(>PyAs{a95M~r&oOP0cR`{7#}NQox_~Z$R552>O7OZd`GPD`E{(k zn%Y~={otMNX>6TQkS4W?+uEk@W>DgfrZf`8ka#HY7rQX z(V`XyA&RsSLtunSOfwueqMbC235+pmV@7XW8hn|BxdvkrFYIIR;;ziat6X^D=Q-#5 zD$8SLVAKtB=wH77e(#_2eqY4BS*od_SD+8z@7UhO`8~(m`CX@;G4IlqU`K4+1-%M- zSl^horzL*?KQD!v!;a~z_~nkBq9>4NP3Iaj;~TY<)R#Xv=tt{eThr%L!8pjr#a?%9 z4Rh@&U#V{^kA5oxpGR)7o|0ou3;f6eyt^^;6Fblc2fYHhPsH^3+_mM)9kZfaBUWG= zuS$GXIM)?$SC6^co}TkFl0h22qrBxF+CjX4@l4<}ZKNFi;~*p6-pSa3+#NXF&NV#2WOeJEivacuY&@-ukwJuKx<38lDXPAm-np$NWhO zy0wM5f$tlskNd4PJ28)3kA8hZ{hOMx*x#z>+`?nevA0CxOXLO!g#}PDO zp1cCzIlj3Iw)ON|%fCaFoU^Zd<;bUYTViW{dtHFN)K8$stTpn;8THBYIHPupg@Y|W zfi*f(jVdR}@dI_n-{Glwp#I-@IFJ+)Iu59koTbUIRz{%n-!H$j6`P*7cYt#@%!BMjUJnV{Ok4&Iy@ZV3luyzOBfk9qnF(wZA;IJ5q7PvqbRayeu6oc-2&t+huD z^4H&CTFg26BIS7^{A2QZqLQN&2XyNMx*WM@eA*2h)q%6h*OyVo9TGjjnx|cXd+CAB zt_r&QbzkFo7!11&K;#C@*5|070T$yp=fZC`6T_dO{16Wd)o zo3(^-bMz;$o=2|pRHXX)Zk*tJ^Ze1RTM}@pcFyOh-#&qD&Ji4pbDZ5>WDIlEAI}fl z{_@PLg55aBx<7rRC;7%UaL0~y6BE`5FmkcCOFi@CcYL1AIoD2K%$$Or=;OZnZkJj) z^UQTVVO>ukYoOdBnYH9%qee`ij_=bmT<`vbd;s~TbHZgFrTXAOO?fcf^b z_Tb_3T^M!PiGJfXXMF4j+dBFNk2$yG75vs{$cpq{ZQ&?8exH{)iLI}KqNGJkcTF&SZCdvb&d6$IlIMPpNn^=O1D!x$* zd0zb18+F_zK6fGC*$Xl26@2Ecxi7KxcYNj^;Km}IOI`yfbdS263sCMr&k65jJ@XTc zx!|vNIS?Blrw}j1<+@Mr=PaoY$W?F2`${a4Ywg0ZzO^GJ z$DEqWb4B`&bbM_=!Drk)a!<$(lzxwFz_|Oums|D#f8q^E&a72e+d8%+@~U zHQ+s+!>J@L!kVtZ$2GakUXI{pU6I|y&o#qPxAc`kBd z=60mBoPjmXKX0i%0};zYQrm81XQ@bz9agIfcpsi*)+cQ@LA89$G=aAZ7*vd@j1&8t~J)@V67Ya+C#3KiqG?nB)`U- zEy(Tp_I>pG(`P>N@tb2EZMpVUcLrmDZB7OEpy3;SIp_%ra6i@^y9|6B>l$z1iP#oc zdm|6{JgvF<>?PNF6=**gSC8eLyPf{T*T9bW=*u4E=s)7C=wfilaewO8O#Fi@2cI!f zh{bCEe)2~Q-Fr3oN!&dZq;2vMkcM8jzz;zB9NN zYw7R$z?OJJdNPL?wp{1G2Ne`!0{whT9tdoC8A~pDLDy&99ncM>~Ns@Q?aP zPYwP^a;*c@MFSUdTTK49pY;y^0en-RJn!kVJ%CDXkXL~HN6ovrck6E9IDZlAf%i>+ zbaU5T$T!b-$vKjk@fF_NJ;+=8RpSXbJ)q{>zHi_;r*U&U8>BtFQ^c&}jOreH`_5cI z(ZA}64eV!+1m@YZ+|FmNo(t?_zB}|Tf#2g@YtDJ(u69AUzV~YC<}1Olk>lCq?A?*_ zl?{wMWR15tSp%Q2MpgGLwdIb_2R+rj$KWwzBtGUANB9Tu93=h`gHwX|%0-$x04LFX zUd9WyIp$h#wO3zTVrxC^i1|Ba(9;>^u5+vVY_07x(C%QpuXe?6OpdV%oXwdEx(LwJ z1K>~OcwXd~H)b`)>B%2ponL9Lb5@|9pn@Wnqkpr^q2IW+y&BNA*EcqG1K$H2;XIxQ zctG9`zZ_Ap*BZ`bEqCU9JQKssb5=}N={D@lT(ltd)>*|`kWd7{ctyb z2XELbSKWEW{BTAtK4-VjsHbG!HMZsso0v1m8+>`xQ};cn@fL}0Ke5iXzRN`(c`G8z z$nQw`iL{^R8&04vJx;d-a^c|HYI7p^BPXm=fhUqbz>tAXj>o*3GcS?ry&L+NBmTcN z3;uCp3i(@Vtu@+GSDv$k60H52cD(IV(2ZNin6e}F85_0nxxWhhtRH~SYwU%XXVlE0 zj=9$w@)Ca)G4+BTz+;|$TF&oE`x{Gq6>0xOHtZvMK{uxCNPQdqx3j+a1%CxC(U%;w z@y?IA@wI^S``KUKny25|o(|7{=Q)rs!4pUjIiH(c-{W-;6?=eqqT5@JKevaCzu|L- zN8lN`G3SsYtRKh%1`oS~D)6ndjyrRQrVl8`=c;h%Dc>{V?j`5k5$M7ha@KQ>jxLOk z9HhK}S98W1((?kJ9PN%A_iz^Mik(Q0@3&|0a;S6uCBNV^W}gJg$a#+~9Bq&HGVZe; zJbl&x_6}m}4DRP8e0z(*C!Psk9`Zt7LFzN!fw8lGq_xKEiB3Ip-h@0(*9FTJ*sw9P*D~ovY;ZHKaM#kZTWjWZd}zX|5bmlsWkw zDfbDC9g*+e63)tVF0la>zcYA7-I%kE{o$)QckVr2$)}#Nac|<*bf%IsHtvS*vA@Te z)JJ{v9sCy9z&vYoplzQM*e&t_nOO#>DzW8{lk<0d>m7i8?=(0FYg;FfW3Gi$W7@{n z_sJc2r!&$y)_a&M8orPNI=aU?gPTiV?7_oUI!palEYYo1zU3Qxq^3S|I*_xXVpnYU z;{H9(rfrV9_n(VeCj%d7_|?}M@`fLuu{Y#eqhottV;=c(^{w%-hnV#%v1aaK%O8D_ zekR?gF=uP|tQ+dca=X~-9P=*tggYHInQ4{gX3l))eE`FU-(E!|d|=-mgte&~>R`mN z<=)&)-w8bv-r9?}HRPBtJl5IVBIB*!9P{OGl`99C@Cq>2)sgB-Po{s&9JwQ=1Z|OJ zJ^B&@$W;$e>FF`Y*@VyEcm>Y%4G)g~j&%3-*bpCj&K>&hABZKk@pZ3(KNs5?fnEjo zJj3_*f){YsXKvjiK5M8uLtwAFvWB^ZxN?18k8q2~`TK(NSM(iHc&wSA0*}vNo%0}P z;jVhqcRTw#$LNVv?=gR#tnc*hTRh_ptTSw=BRW3$fu88@tjdjLE@H;5ZN9N_S7Z?! zYevraD)l*Y;!mV=TC)oKA%!z|A8U0W#~SKKVhwpl%B$$w9_-Y{O z`VLM39o-t{o2%bCvFPTRQ^6MB*d_di-{-3Idul~u>+?j;+<#UdHOTdO$hZCxIP2IY za^zm;y^}f>p0Pl_#_zmx3aIGE#nyLT;tlB;`Q(}}&wIAyKDYA@9lb1YCgWRBV$Rl( z2kpE0&l=^J>n!?5K6*hPcSHJnU4j$ven~uXa;&$x&OKS90_%)iD8}4{b^Y9R;Q!7q z*S_)+ee4QqB|dWX5f5+#a?I(VfSO27i>W)0dXQhxg}%`rU&Jyey1pZ*_&Tt6hnwhY z-if%MmmToP)i=24HD}F&Z!70^_P4G%4UV66>zd=9$I%86SYqLNXYYFezJF(6Z1|0V zHD~lAHxmxBD8b5ghBemk%Maw3Kjusueds<>%e*Bs>|PMBLi5s zB;Z)5#FUBL5zCysKsv{e^01vvS%LgY-LglbH+%uCx$dkG4`S|k#K=!M_&Rds>)R01 zmhaIwYAwC3+3^+7;%$DSulxGx9d&>^b5`pW;@;nSzR&tZ$KTb7738RauLAw!4&KH7 z_{gjpc|8C z&TgrZ*v9MPU-M(h@j1vhUwu3u^a3)70q@bU&24aAfsXBbiL{q7^E=YJ$yx76bk-$i zeQSyY@=wc76dp_ZRf6w0w(cQ@wJK=AXa4PWK?nvibeeTeiiyUUL&yd&w^n1rM z{+qkShi(5xtb#(!84gJE?b|_U6YofO_2PEU@2ql^&U6IN!3N0J@7;Y5^;fv+o!Ayz zs2AjT9!qR@%S`x+JoNQBjrSQ}0|7jdv!-&?iiJ_opH z8}KgLV;{sT@Lkb&Mi%VNBJI)eSKw?)`5idtk+^!t-;hNPb5#5rU|hc({YvWvctuQ} z^B7Ax>LAb$$isD+)i^oUH&)TjlT)zQdF5Eg<2&UYhYwwj`dZUxVO+m;PS^ST#w_ss zJm?Gbbyp3&CH6YcqC4|A@a=D0uDKn%d`qr#$lJj$NaNeJ7q zu;w=W&Qo&k!`+I{@2ExW3EkO7f9xUQTF<-td+-DCkLBYR1+U6+E5KMP$#?hu{}oU`pgWs)dj*vCY;Y^s?dN=3`~`c| zP4rvh$N^&Voh^{^dLltbS6XvRTzmD!;$Q!FHuovl8OH9tH+F*8`0ha$*zQo>`v!K6 z`J9Zc`?6=jpTN_S=XX10qMN7mZU_7xd+Tq&eB;W9#fNmZaes1ZPX9WOd7kHTV$Nj! z0EY!@y-~yY;Wgl~-VvU=68ZyQRRZre?uD;_dH=)T|GUrr;lKIc{;+KS4rza<9RB{l zcEj%m&*1zs=kJw2E`RzT{JTH@_y5Cx?~lIt(_j4QAN}y-?|ttFKltwVcHjBqAAWcD zC;RV}f9GHR&?KDUH~;v>zfI#J{zWr<@1HlrC%fi@U%{(W!v{U7ih@UQIi_r33Y?+@Jn-^%^J z^Zh^m-go}6{crYsSMz=C&p>4fWR~)e@8-`{^@jd)@GEeF)8QPE{_J_UmhxBlJJNse z+n>*8^aA|;zrW@8RZ#Zyz~}dVCx84M4u66ZIa}~42y8KaZ(pSP2H3NJ|9Ij461mRp z&wKNCz_>qi{hj4M1D7S&d}nU-x5s($SEMue`;ou5o%i#em#g2;bN;~RKU?u<{a-4R zS-=U-dKYK0N2R{L1F`Wl9eIykz-w?rzIpyWDj`ue$TmY^GnQm&9`5Icd6mecz5xDZJj6Z2FP~?pPO;HXKe5MY(Kf)$J)-~?=n88 zpAqveiEX?BYkx`X1MnGeZuet9=ke%shaUmC*1dp7&=Pvg9gMr;v$nfQV!lsX>;te~ z0^{!Y9cuaP&foyN#}^>T>(ul<=5DXE_X+(B%y&;Of%|wtj{A54oJ(E>zsB$TWIg#u z^iS}ekd^wc@D=Qvg;%ic>1^KFdL0DZ3-TB6+U>l@6VkfQ=I$Guiv5KC*?!(L;5ei5 zIe3PDhdhJbb@nRAmq2NMIrb>ja#rhkw+HmEfjim~Z!zar-+&7^TO&K=k% z;P;Q;&&EGkq&xSS1$BH!-hj`SxUr7k8GIhjCa;2fY4*Jb@F~fzS6eL;5~+xOg@ zEqX)BNu)JHne!cR4)3_V((`WCe2(3L()+x#f17I^=M>N3TdNEFH}HuN+~NDZXTN7a`D@_w`V7c-4+ppfZXoU3bLKkx2|JPQL4S=h&eog-AMNMt-Y;_Y zY%$*#=W#~wcm|t=^KhL%SFK|YV=qCF=Z-(#&!4sK<4)AxQOD%3RRw*}@q{{`HGpTgM!clP1_?dQvPM)*FyfS2F{;+47u-M)VQ zeUH4`8EAh7>|1iq;rp}&J77=e5YF7-Y|-D@&(DJMWN?@35?9A}0$Vudd%x%8ehK~) z@MB>A{|3mtT&a)#(SFWoP6In&ufw8icj}kLK3~y61@5Qj)U|!Le4dHVdpO_o+xcAd z{Uc&>4xoXT;0L=hGU*FEp@z?m*Jst6#7yJ^x&m!SJfd}Aoy+`^kIKQ@jpH0TR zk9W`~=Y;NekvlknpMw$m(VW3K+~I$T-)D9K&Shy>rK$_!5r! z?$_rY`}r)_DH0sp?^W_b70-y0q69&eE>$?h2z~?E_Xwou;u**{6C<=t3gqa#C^wd(e*j& zd;2+8C0Bj`-!tdx z`LwrS1I*F?tGoF*JR`e4^1M^ztUYpw-5AGj-Z(LPSj*a<5%-P-|08w-1vpd5`8hwM zo1buAF1qp_G<+h_Phh*B&(2t&e{eV77k6dNEl~O#8}L0Z47T`=Ai*QBzP;U{pJm@s`=SwcH*hO50?Ez~#=v++ zNay((HQpdAI0B!+OW>VW-5KIKdC%ed+5Zy!uRxx?Dm~5%&u6;X&wKt7UB7cAaE3GZ z3V&eh-@<)`d=1R=d%+y)yBL$(a_;C|INHwcJuh$y@bmQ!_zdib zHS#t{cjj(B+fTdVH|Bji@;&^17kHh0W>c^lwy_UGs?fzQ*rah<#J zv)f{+Ma<)8BH&)YGko69xt-H@sH23OonD1)uJ@ z3*0jp&mDe`IpYT3=ue)rnqP_O7uN87`T%Uf``|CZf!G%8K!fkQRFL|tcLdH`K?kSB zb}!z?d;6SzOs$42Nc)RV!TENXZ$F(b9zzQFey-xeH~N02~X0s8(4`JVuHc)rfMrs8NJJMR-L%jiK zkh`JI7I@ddC&#$89>F{Nd3XE2#8#Fo0$6j4?+9AF-L3D;&w(>OC*~gB-Osq(Gq`|C zE${1YmCo|C*k7UBzff2B88`xW6vVy+XOM6@@dWOC2MTyvu-nhy{d_ln1DwBsa$TMs zJae4E&z3WcdoKQ8W82qff5K;N=RRR4@&(db&h6i^{G5IYz6Skv-u(f24=%tN?Xv~u z>2pU)b3X^(Ym2=Bb&=Zg4p-{ZH)4L*eGK;id=?#_yDr4-am4@be%|MVzkvYi$_gHV ze)R*mkmDTd&%l@0dH)T*ME9Mu-!pL1N3TlXirkXF)_y?WfKJT!-}&6}8DGG4Z_eso z4&Vb~M|_=_c7j*HIn494_2F`c7oxPrr&LGOSz)&+z4TU|sbDB>?gOyx+rh z=J<^L>^rl(qc+?}Y&mDJ#TWR-Ge>WrfPO_%%NgVt4`hRTL>6S^toIt*-o}*;T)?wE z-vz)~6aM-hd~!GEk!$YhN)FKfgtYG$pb{(Od58ZF`!nzk@ZRq5IlAwTxfk$&KjCkX zzKaRofSo%2lJn>8FUk1|6#NyR{tLQuzO$cq{|Ryb-msN-#TT>w5zZO-96bkhIIppN z=GJU6InH2RA;&uAKLGE2v!AhmUqy0VWCzyj;G?^_!_9ua`@VNO{J!fe-x=JCITiTa z8#UaMy7zm)@BQ=_@GksU_xo$NJv& zEBq(qZh^c;uFuN*yuQx&>Vfz;#181@Si>FK*Bv`c&-vM~cYr7G8W^*Ny}kPf$X@`T znV%!~SCBsge&*es+%x_y@V_&?x%J_Y{VaR=sGGl^{%>2uINV7s3{FW325+hKdR z-;i&w{~jd#g0${o&u)mzKY^d(`yas*dVn3Uu4jX8&lkWw2y=VPXWFh)`aKr--rbW+ zE#E71?J>B-&bRZuDsX%UPQ+d>K5II&Ipcjpe~E7W3IhHUvTO4`I$wev*Z}MN7LMD_#1?eR7G&%tUjs^-l-{pwVeMRJOTG>PrnZf`VkoO{s~`u^tw_Cp7|&62ws5gb$&-)kn-NWBNv`r zWm|IO_n6Pwy#EcJ-&rq#oFsO$FJoTkcm*D>)Bpv}FTk&WzYCo3+0U6PdBU2`*3n-g z{cM;Y$k*8RKkSL&djU2;ZZ7;r+|PN*`CObKh@FueW8iq5y`9Osd3Wpm_HKTMdUxM} z&%s&T&K}n7@SNMe)^c9ohXb~Ae}EqJvwuZ4{5N;=z3bT4aPA9Sb^9tCG2=V@o*n*( z$@RN8(SukTCuaYnI@o{{_zIrfzm|WUcle*tozY!BfOVf+;?9^0yo2?-ZAbyMgzKJEy(gfD_mO@A{X-z6Soj{8M1xUx7j_Km$kMtj_TYm>=|hj=B0pZ4WF@nDhO=* z>OX=KGoHrK-Hp=v9p70la(A82?+x~Pe|LArpTHTn3+|Eb)7|+V9+WYk`72`9_zJ(D z-2&_(68Qc18}LtndCpUk{bPRsez{0<1K4|u&-l$A?(LkvBgdS+{{Iwcd*1*)dw*||w*?8TU$67sF2sDNd^eu$=R0wO+`+N0T%X|y`1w3y zzXS!2cWZ)KX*zeC;wowz-H7yS&_;|t*D&i7Zoby6PoXN!NcpEXOq{kwe-b5|Sq6$CNg z$6=$}=Or+f;1T%QNqpYHeR)2@*RV5Y4|mkzz69=3X^kgv1}E@m*;%dYdvfKeBu8L>E7MJ1H2tEYj)thFW?L|3$#T7->36+?$|rX31+FFp%>ym z2Hr_t;&&hR^52Dc1*d|elIOO*^RJ2d^U9xHPsE<>CvLDW*Lkn^@OA20_Y?f~_nErS z?d^PSjkx>v9sNi?vH|CD26J}EuYmj8VEf#Dir&bz&I3qvYa6rP89aiU{hYx%^2-vt zM|wB=dZ$3X1noNavqS&%Zax?L``o?5Z^0X4M_{c6UM%vU&mO?LyW3BJ`>@u34c?%C zi1eQ|y+HRJ^G?oczPs`J&Sy~ZJL3j8-%s$l%M$au(s#}}N$dplKOwE*JJpCiBFEnU-mr%IYIt&|eZx-PTq=RP~W570k{_YC-~-G}?SK|03;*vmWE>j@+^?PE{lyX%Da zI0AExAAo(mcK~-Nbj)$&xpT7u5h0o{fS@RJs2~` zw7r+_^Dlt4&fo^VeXSGV9dxDp{;%MD1^mown%FXuXU-_U=K z&*xe2JvWx~?hR?(kHCk(`xboe)jhdC-`&7>K>ry1GvIFJ?120hbB>=76W)6V9)X{6 zb6*4RoA-8pjt`@aPCwU+*IH>Ah8yidl|w^um$Jo1Jg>G$ao{0{gT zcm8*ga?ao$xPzMoUm)%I3vhz_3Vep{U24AlT(y{Y=*U71XLiTV@d%Ef=d5ABj{O00 zyPxm8c|Mnf^Wt{?GeMuJKNBmq_uTafxTj1x}`GPtXc#jKsZ$CdvA0YKP zgT0+czR%kE-hh9E?tAqJXd9{B>tN?ed zui&fb6}iO@V145Uc&Ei@tO56-&zj0c-dfY1XM9EEjLCZdO8vfTT^;#8xS;o#u}}i# zr+4$+@N@7r*l1tpZYyznf4HBY_k!R0N3a3D7dPFEDxQDv~-l4~=xgM`?pSLyLi^umz_}sSmehfYX?%@mIUEhGie!kmBc8~onA zp?mj&46sB0>~7Yu-flm0>~jX*#o1mV&D#Kb$@RIfy9spf^fmA~uCW)_`HYSGjAK9N zYV>xV%w@OuliK$4uHN|p-5hIu3D$dea-GQ;o#zaWU|5w=K`5>+J9{2$4h(*jE&UXU81J?Kq{59~@q|SBD>^z(O z+}#1)8MojFj9J&7J?78H-%{fv;2j^q1GoTZ(@uEfS)$uFuhbLRufeDL`R<(11N;ik zr@$S)!j@;Prz>&hJRrq*4%pty=T-2%yoGDO<@putBdEl@Ux9l>UVyVWr#OI)&%MiY zmmBag7&VE%0T0BU7BqT0*ZceFlm9+=j?X<+(3vHGbHo+N33PXOk38zn`CN@B_6}@8 z1={bxb$`xy2JY+x?Eh@3lQH*bOnwyx4SAHfd10w3<*K6mGnm+<7e zo9%)`7J=V)-~0~TtGEZ+?qUNn;R0)I_jBhyBkwH#H8=q4ZExq^eFo1#;7j1~bK^7d z{qU~NT;%U(zj2T2{J#AN&Ui0OOn~hjoUO*JyF>TbGh@zrL_Y(M`A0Z)u^-UQ(O+{u zfB6?6&$~W=62y6C+8v&;6L zM8fg;8uOfy=A`!ON48u($N0@D+8}^+);#^j(-?W#{5nlO;Y3uyT^-t=*TxH0)XC**Ig05;Ji(D?jRFF` z@o_I?)8-B*q%rTZyPf?qXT8AIr>#6+XZ>y+W->nJLvF#ygX`zX+Sb)4$3C8#^D~z* zYdBlMXZ;5JoQ~S~&2PpQ+3+2fctO5e{DsruB?!=oZGd_a9=I>@nXkVAd)n)SJYOYY zGq*Y3QEo+6?5%Nh>o`xxw;}!nyt_Q5-{0n2Gw@m8{8;RxasA-wI_r2(a|53KGx`n; zp7R*HF13hzUSN+}$c(8gt+9b`-viQF)yE-b{1KnAqK-74=p9sI_8rd<={~mj0$u-6 zJtrT}c-NGpXHNdeMQ@FPP_Szk^=jy=WXS<^gg>NnTBn&%li zkL70&@i)Z0r+s~P9mu!7bpqKyMeo34?ab*LGhpw?@pqW@zJbq}@&PHf#I0rhE%sVB z;h4V#f&Yjc2miWPrGD!-{PH{-WlU)uPtQ54vVaSD5tGxA6<B4U>9UVc6{1_Tsith;%i{c zfi%b7`p!!}-f?8IcIEVyxC8VddF|?l5=M775KyvztSD3%c zlRIX+5@6^_$LwXyyIALen0>92pv0W* z+qnnxe?MU@`vpAn8@5OLh;AKg?BJMpLRu^FJuE(Ba?LMbgT3l_#`~quqhGWjM;`Yu zU%Lxp6&XrqOk}yv-I%A{^>HF*U+YB9-4uN5PN97BIWS%Y{j9&fPu6nw#8;?u0-hd| z>*rLzxpEpX=FCcupCfluVtL2VM}1t+7r);v=9UCVbJU$@^h7`3&fNv^3Y1N`p@Oy+f9TwR~_tzAK15Xb}m z5AVu&Mmu!(VomcJx^>jOqtf|(p5`W~`mmkrtRLGt&g${*-dq1#x1d|c8U-|?@-9;joE$9KUy+UFH{9ceAO z-r1TB*l*;Zm%s-~&RL9WJJ)*Tns43)8Q5j93wnacS^EUXT5^16&FknDI9pSvwzkr^ zHHYs<W2jwy9hy3IH~XcY z9Aa{X-z)a&=pJXWrzoJs+hTP>Fm0E{~Z0KyN_bVfEpcZ>);Nwr^SL?ySA?^}$PE-SIrp*YAVT zo4)3c8kC8g{lBRJ-9TJ5uwkAL%@2q`icn9b@C&puYXR_mITxXT7eTbKj%x z!qL}(IR$ACYr5Z?{p>Broi^b8@09^O==L_(y05Xfz?$Yc#{v7fzQ6|xKBc?|bY(>b zKC#A~-8zk!ea7yvt>>JlMG_r53Sxop0klOrQw9X=Z+3*P&t6A#bGD1m_kW$e5Z8ae z*K+2_KVv8MxHtNk3;)X5q3e@V&<79=eREyzi0)1*IO+o)^CD-h7W0lxcYN~4 zy|9frv%MN9pab`#PuqDSrmnAo0P9@F8vHFhF;3!(`CTV>tuiDlvg_IofMn8*M-aJ!yabsg=@GH7?eV*!a{n_oj>%N1$ z29lUZ{Y-p|o#@)u>qzs~Gu{vE9y8{=fo)DDw<1sS>8G!N1_Ic72S>m29f7&_aYlFT zeeKcBBUX@dS`Y)?)qKy`DKYEHKN1sT57_1;D8dsU=L`kj3-T)H_ydgjk@B3uTJkz* z`L=$!o$rr19{ZbXUd?%zy66p=_zuLK-FiNkj!(aKAVm)(sOWi%1Vir5tu<;O$ANby&m(8u!4~o^NM|+HfXBUANA8H>tLPc? z`|_UH;3MS~Vij45hdTAP*z0#e;*-C-B7wW}c?CYXzMCzV{=_@_8R@_G;2q5^_{L1= zL9R8;G5?5d+`1XlC&v>x>%LDd^$0*sx!M)GEU->Tcg_({d~AKz^o)2Q&B5pUr|igr-`urkpcni*W1P`9kO>u(>k=dGe&iWDAkEt< zv0D(oPR`KvbDYokI-BqkE>@x*7nBOGgwn%*6K#C|mE z$O$+tCY-s-N9r$k^IcSTCSy$mF}aJ+{cW_ZM=Xu!ystSG+q@16wdBefF=`}!pPy20 zSUp3C>kx8hs;&&e&stlx>r zS5M^nomTK4u@nCZzB?HA!?w2h=Dsn8F5kFyoKd-hr>t_wFX%^gcN6pdw3hYF4=~;l zq$d*}@s4yx<7c=PX+85QvC#*gyao~&*RFzoh5Q}-6LS5I(6+z43?#O^Ez-M>xIW^e z4jg0FHhvGRtzPb?3}nFfSoZ?on$Fnc?S2Bi0_z!XIrGOqgTUu3_7Tolh|4$6?@^C> z=JYrC&O3O_)bTgsYfhq{$xAqaTu;ZpK94}R#))_&3=r6(W-Rrny~U^BJ2{{4lfDgT z$}4_y-HZK20>8KQo2wi#@*Aj8}AfT2tLVMIJu$62J55D=~8dX+7&VWG;3`I+u3WFVN);d(i{VZsDDfp&xk! zTi~4L`aE}_5cfpRc#9b~W?g4B$6DoX&TujY&v`0xeMex-_ZHs)j(tncdwb6W+JTIi z`gT8S6u3!joJCe-1J>^XeSc+x7Zw?vuAM7?}730+;YzG9NT^#ds@?eJ#{S4 z8=tYnS8tckDEbq6;FW*=wLAK$@#`-}(+n=UQVO-V~I>FXds z0_!-x9DV0oe8BtZS2u5qlv`uY*NEBMJ(*L1IZY1yQ4hPoH+T5a%`sL5ak(cY9BUQ) z=5!DOpShk4c;stu$SFX*;Sc2C>qnQH$niNvKT*eC9iM&WYI|DF8f%{QoJF3;Jsg1k z4m?U{vEHbO?mYUo$ooZ4r27eY4c!_8_=>=Oi5&gN9rwaFY-G))7t$RK+dPh3=W-U$ z4n1R$=pK7jpbX&Yf$vB2_lDE$j4PdW9Hf1A_)749Nblu$hO**Qcm6HtiO%ufpY0bu z^^=%=GH355Hx@oz>v#ucfFtn&%yFKAuYnUBWy_hr?$H?nUj?HFye>yy5F_rqiCzS8 zbF}3&f$rSCCyDLnyCSzra(y<=rO*5t^LdVa>Era^kA>57=2}m_y&Ap%p3LdnEc};n z)P2VqFyFbYX@7g~KwuYQ4V0J|e(Z+r&H~+@?$lFb#yWD<^vv=WE9Y#Eeq=|k@0E9I*v_NhoIq}|)s1ze{IY-RlW*@rT+TSb&hc&ZqhFAL zY@jc)>c3rb(dF9D9eR3B-P{Ab_1RYZCFUH?tZvQBMOQAz9tqst&E3*Fr|olc_AN4y zC72HxNb7cF7(=Sdt;z(*1jCp3L1?QZPvq_7`#vD$c47_q?=zh7*{kC(z`qadZYN(~ z0~I*;iI{bL?-D6reLP-4!O$T#@S z*Vn)q{Q>yz0Ia8Bd)G)j>c|Kreadyl{Jr>*m>laI)Ip+)4Z1akB(K3U$9wzF`9>{y z*d6Jq!5?5q)Pu2{oWO5v^uT9dcT;cYUbaZ{Dt>428Q9Z&QA7YvO~7n^7kSh(7C*Dv zV@JNH#k7Y^IQsF8IC9j@S;HBuBe$UY9ppVv3xzXU;tb`U$`Ng*)*3GtrMo>zQlbsExhOSU5fAF3sKUDUGba*_?Cqek)D@t*RIK%VcKF?~BFzLMksZF7s@$lokH`z5w>I&VeFYl2vVXDqP&?jE}+ zNauqe_~d8Kn%310?5#2I2oCu5HQ=c+eFeYJ_A_p-|9sVZ$km^qplkb#t!ZEP75MBk zWw2GVoIYJXP_M|c_(wnN z&=-JXeAJBPo=8s%{s24b?#K`D8dz)CIcMxh@8GlWz4PA{9q$&s6TkIGKV%Y5q;ptb z|CkBgUVcYx$kA`Sf`T4MYa81l#X7Sw@0m*AbMkXG>>uGnKf@pU$teTdym1mP_B}lN zmi^?P@HyN1UJXe-^PHn2Pe?gJ+uqjBoIA32N8c`fYsjk$UcvP))^p|~y7|r_*ShZ9 z&(eC#Pmz;X(5+v<@L^l85;ta@a-DS=w#OdUIG6*ByF;I0V7Hw6->JhJ^^xWTu^l{n z6<`fH+8t?*v$#)BRbD5@JDL-L02N)^Gv-9z?B^`{tQ9$VL&nrE_~8SLG#i(Gu#Tm0(YLFtU<7WBcPS4S@~KNE>uYX`db%Piw~pR%7m=k;#mEPiX& zc$+`$+hr#FKKCPd1ob-aw!RD2G517l-How^&pY_fsl1nYLVqBg$$Qo$4i2|VjQG0u z^*)u@(HvwDZ}1CtAVm*K0B2ZpiwB>!@tSjA>gGD5@0501^oBmT*!HxZzF0WcP`77M zf|qbE*EyT-fIed)bM`NA0(jrmFRv5#9GAGhl2DZR#DkJ^K)SOQ)B)~rox?suhv(h4 za625i2W*VC@_D3%33fad)*v8n>=B8?l1kLD{1!bIxi`#kT*M z9QpRrc1BU*uKl*e98QUzx)=4T%3B^;qYu&v{fUi?xT+{te&AC0wx8D<}evaqB#Sj?aEOZ~@kHF6Xe%bL0W6eT{p} zQ9mFr#5;P#{2Avy6I_8vS6y4)~^xIdt#qRN~Ihn(I_zt(*xj$LS+a5>YEY6hr!7J>@8GDKDsWD^bJ7dLf ztmX9k9PC#X^q6|bz9*)C^g#N&-Nlps+xas`+y3@ZA9c_pXTEzD6+40c(Gy==Y<&^a zU(lUXpYtUPw2*-Rmi5zu^2XO9!-{Z4+4)l*cfeu;%{72By z3vdVa4}9Z3%%(o#F*ym(S|?EP9g&`zLyUK(kTc#1V%BuF13o$C9Po`E`nZerjdfxT zgf@Q9*c1E`=+yB<_{aP#W}bJtg~kGFFU-@);Y zAs>ywKcPGSm7Rw)o_` zQ{ONDj%SW@1~8{!%RdwE0(*nJCzfCweD(}w&S&F0vcCKBFJL`>=2~Fk^K+_gKmUEM ziqBfc59lr7AceD9$9c!SDDnSt@xI({Cri3!f&vX12s9XQfk1-`8VEGJYn|A8H5?#t z!GHn!00Z<95)3FPSTLcXpxFr-8E$Bp(9qCuo+mPYcJ<`yw<6+=H~wTwd%I56>GKoL z{hW7#@6P6nU_Bh~defv~jYM%HfY@=2<*PZJwSZCE+ zBbTtFco@e;lmA#XAf(NyCP@fI)T@ARwS!-Pf&M}sEj{1njIcLB(^3E`)#@Jf> zyK`XeU3(hi8(UvJu#MOr7vT579s%Slj^*cT^-kiD<6PiouAS{G%(;)bB-gO+(ctx; zZ(Ec54dvnNbJX66?IEA!<%N5ytNd%t^}BcMpmlCXYrQ*r{LY`YuH}9c^6uM-ON+IC z0j?(NB3nyN$dO&<82x39+;L3|+|Nemj?zkJ!>4`Tbsd#}JYpuJNkKfsgvo3Woe zzUTN3s5=3FjvnOnzhd3XTJ?U$K|F;V)_#HSTs7*A^`GR#e|#ZN?5sgtcSCZ4?qHAeYiT;Zyzy1NC)ekaTKFR$`1TpV{j60xz>H=8cwbz_aOML{`xtwum%D5` zV%4fW+gq?x!fQ*9AE3Rq7WV_c!HwL1PmJ$5de@Hhs?RIe9p!3n?Edi16({;RV*pR) zdMmA*xru+L=dsr(YL54r@Q21bsEfbplX`L8y4%h-PU8EW8A*h8*%!**g% z1U`LJ3vchR9;+sevEHxp*hcK%f3;k7)|itY;HnlYciK98nwR$t+#~QK*LTUCzX9J# z{)~Ucw`QV|QxB-!$mi}&d?R+=v9B6G zXX^(39sesZZ*HR1cJdbx#P&PkdwTE}e68QN{#x-9Ywd)$mw$fff5$rG0+PIX^VJ9c z1O|NN;60IL*6*@dzWX}&tn~yv{#o97H)8vEtaon3vEu59@4O`61m}h>V`=M+$h)g^ z++W>M>r88}55 zqp|(ib&d11HW!TTQM2XW*Bt+TKjGEgsrA_R1K#-!-*@|W^dp(atRQwTPehq|7yS;* zYnvJ^#~BTrUc8e$e#csK#yZRX)@YA=orn{SIQm)Reiv#U!0)Q}Hok#DURX18eJ=Iy z``Y@>2AoHIk3H{g^{NEEJKV{)_hRZE{<&9s#`<}5R>RL+_j13!xf_1fe_1n5@ZRMD zv_ahjYVO2W;641^>d9Dd&~n)z*YS;=XI-E@gX3q%cU9xt)okM9^KGxy(>v(*SbePZ z_nX`R|Ndpw!&DDv9jH3=#@?ZR)`C6k$IJ`15qo!gO>|8kXpPlALF}A*ed|Ba{JIO8 z`mT}xp-;|tx`BY}f;#)kd9NgXU_I8>A+A{jH+<*J*gepUBmv~Slf6Cm_gI^;W_5iu zS784nKI5NXyNiXVhC^^&7m_KKd8zj9p8tKd=q{0aD+C-*df5;FHRh6i@U4++x$i=5;(Jc& zV%29N!=IqRyXU8E&AFd*JH97!ar({;;M_Xsywk|lKZxtZdZ#P73tT65hw6#-)IG@A zU$g+&3G~ee_Q>hGuXg0xF|nQ4Q$Kg?KIqwU>00c5#B#L@eLMbL9sQhBpjN)YJEQMD z15Vz)$1L@pP^%93J41iOs;_*m+=TlzQy*luny(FF>l&Eg`wKVMKZhJx-_G8hT>VVx zv#)1nne`cL-#!7(=+4w?{hXcUKbQM#3~xXCS1;|3C$?=^Gtis!{RVZ;?AV8~KJVcS z{Q&wIOJ5dkZk>IDd^6?_N1eXk!ysSpi{=LGH{qPy@$1ib?+#}VPX;+Ka@|v_Hn;We z_V^#~t=@dX%ZY%u*6*tEz@Kut{@%O62fmsMzPZ{3+kiXX;TzU`VC9Vy7+=7U1K#aH zY<-!?Xt#M+@Cv-pS zJUyum9JKaOGeK_K&;|JQGl0hZd^`Kd+oR)qjPLT?^H1z}p#}2Jk=M7rVtn_kqYd)8 zEPt(?D!=uW-#sQeXUQe-4p;C1=gqfVe@|x4KF;lEZ|1+OS^haT;Rf)yZwvkpYfa#f zTr&b|Z*2!Z@o#9XvBw4L9ZOH%dG2%Qou$^kA0Vkw+dv1#?QQi^XYP)*=ftn~(^`uz z%*C$5>JNDPSnto+g*b>Cemm?tey;uG&DUq?+X8p6?l|tDUR~P9LtcJ@05x~}pV2(P zcaUx+D5I19Rw#T+3_q$*{rNhqH{jr|w?%8Ag**dTZ z)W^2&X5M&$`W&%0=W)#WCeHM`?NMuQW8Zra2flsJ#P-p*m)4`M{`|-a>pODPPPo#> zny141)@Ngmw{nNWmR=qV__qB&zTg{sav@{wNAyc~LgS9wh7C}C2;HfRB;Yu=zAIRD z_PvvLKXH?XOLA&2uQl(1Rcrne@PFf?q4@#to&(<*+6mGa)IoE64)P6*Tz}uH4P)TB z)BA~DT^GF76Rv@|m+k1}k9qFYfIa+k=!~^SYj1sjhugzl>@)CvqZ9lY zE2mWxpuRWQ1l1>(cA&ARHr|%6zLU3JzxwFI88>i6>pM@l>Va({s0AMT)ydWOEBeIu z96g&{K;-J@p1y-Ss_nqsomgv5|Mprv+}-oodH0G~^*8*9KXSc~ys&3ix3%{4pHr(3 za`9RpuAQYoMb& zfsU`nedGiD@xFAIT_=~ZdJOzKyhpzS`yX28oQTh_^|R4~yW!{7KFC!s_ETehfN4Hj z?a;6joY9*1E)Cz`G0quq)~It{-Ff@gc{h)_8~S<+tTRSz9!=N(z9PK`Z%y8si;JrmCTkIp;Ir?Uz2i|XDuUP+V zJY(JKE4IN0kgF5x9$^d~-~`-1$<@yYwr1w4P3(ko#_^6fYUBp+sPAAT0Xg$>zMuVl zhckHU!*^``cL~_jy5e{v-^OFj8O$I6VQdcHyVc318?m@!KXZSo<@8v!J@M(U4LGec z#^wX-F}}h-sF(9yoY(Px;NOflf5lJy24dUxM&3C=FK{u|*P^%f*rV=a4DT-H?%3)@ zjdgPN@QhslZqB!8_|A$KmUF_|r-3`Ty{+2J_0NOUKjDwxf$r|PVEuEz?~NLJ1lB$i zI6LC^_hjM=&rJS8z6EvMS)Fs`oHG)hw<{gF1Px9*1%3eUeg@v%|K5b(+k~%of9>BL z=-o+OC+A+;(vkDdLF~P}ll*{p7yF#yZb0Au@^bz@arcH+{RC>zUPE> zmdCjfwL2jtN(y|5StJDZ&+*V9{8o>&K+(TzmW&VL0kv_g#X`SA zI`3eg3;uw=nit!*=J+n%w%!`QE7p7lBUbO?UTYWYJnnUcvvT5+AL(lQqWpbK8M^?jCB~si+GNrS-d&tkC)Zx(2Yrd1b3*4HLF{~c*w(tq^dbIWDg}$?^2fBgApVa|x zze!wo!aw%8zpd|&d?WAYTRq{e58xa4?=@R4q?VJh?me|&pi|TF3-|-yNPCnP?Xx~PB$A&(9-;h6}_b&R*a9+nx(C`O#=IRBrJ_Gdj zd+z)LtA3!HSo^t$=c)~m>pOL5{pZ#7eIlp*iOyZ>v&P!1lXHi_H@4@*dMEE9=e*+# z@-3+oUxddIf&KSRAtsnTQ&x}TVM$^>cAKI^*Q|Fr}Irs8+jGSn| z9?r3kHn5`SvTDq!Pi$btSuMSs`+7w_G{%iGQQM?yKFLw!K#D-tN%gj(hZ6^CnilwCo)8 z_=@%fyhHVY(?2-%)_3@+59Ch7kpTaqZyh^(zxsaeILX(~8d~cn&<- z?b`)=JJa6PA1=WJ_ADJ5dxenK?*8WeEKby)!42MB@7StE>rSF}!X`BV>@TlQe^S@* z-O0~iow4>m-`Drc*=n^vvFbe5%*28}t;Jd==W(_>MQ+{O%8yvJ_Wa*p)}PCboO=bi z%2RW^kA2`AZ?WAw5r{;QZ>h!Pj zv3hl^^Qs^IFeZOPoAC7+!8Ls6PJM7kW8QjypLpz{?x@YcN9P&XGl*F4Gjn~`#>e~p zQ1e1e&K`|i;CKAe(LdqbNlgdF?G0cvSL_}Gf5HXUJ5Q{02At>f(IW^%uKnCIfb|_- zKd_zHowR;GJ(;U#y{A6f8SAe8j9Y8`z(3xrp&J|C%YN>4Mk}1zwB#C4Yk&I$e*I1% zR-16<16|-Zee9Lk+CXD3HFbAv>EJHrUh8e_<&mpe{F(bMSofa+3Dlg3^&{6FM8#d; zt3SCUXAgHfX8XCwmj>IT5wIUUtCw10;hdjf8dK*z?!aTe>hoIryHCXGk=T|?b5ald z**c@q{|@Xw(R}FBtKNaUv+R3=x6i=0m%ce`?Bi}7KY)A#-qZVkBRAl!srmS&fw$gy z=j{(P?$o#a`daT;dbIZU)IPcTSd;j5g1=G|*su@2{e?R;;I8Ep%3sx~iS_%-x3j0a z`_6v9+%Kq2czydgBeC|XImB|_vFxkWU5>hb`6O*9!YsZ>==}=MH{nYS$gZ)|T|#baMr-eyT6>f zyRY`>rN&$XH)aH0-_2u%y%XEO0CF?2 z>^ac+nLWT2*v}rmQ{W5t`Sm%#>nFVZI@aB~{Pz96I>SGs-OaaYZ50y`JQp;a%kP?PGk@yZ6~R_j_R7CD5t8lj~S# z4eTH{@e}JCSgZEed$iTdxTEvk?9=ckHbGEh{@|Lu;BVN<;eSUHSoP-pbMXoO^9wtW za|d_3zOOlIyqEl-reo`~-LbRm>7Rcde;y;s#aZ)Hn{en78^BriIXJZwzV?UKKckUX zTPOJ|a9{1jegoAXJLEY%*5?&i`<>O~LcFc{53Dl=niF|_>pK2%Kkdy}9$1^(Uhx}# z#aMmc>4ARa>Zvu~!2rKzz?UAaCv(*~OWwM`Hmtbsp6zY-W3LP}US~o4y}D=u70rfvuV9=)vbB$@3uRlt2yfM z-S1B{1Hbyj>gDg=iEocdeIu{ddwnBk%?UJOdEddA_EGzN^MT)EX$DrER-JWMVr{_P zsj)7}C9%g{Z@^jq@@3WeF7{Q|V%^QydwOag{6S6apU8-d$NzwLo_(Fu;oQSJ_1DTd zcfy;W!2a4N`HpQsYtM>v?O%P=5Q`^$&Y*rr!Uui__P!Cj-<`M#PwYDqA3wK0(T!Z+ zhp+h7HlVG~*cjf=O#Xs@!jD*1jehNlKk(ft(dO1PPREz`9Qhe*wmSQq>Fr$SIjey2 zGx<)fJv=oN?gH0>_#bNr@2PK(>Par~?Gxz37@Y7Yw(9Y%v*v+q@KcCg;T%rDT*UR((|6Ff=+PxZF=9*>i zv3+%icEUL$4jr}T61_Q3-G_aBcRBl?@O_t>1?N5d`SG}y+L7x%Ki`OB?P6Wu`Qtld z?)c|KPqdwQf|mFow~biyK2a~{QEQF*`Wa&Pfp6U2K`!kBjXTy}*yE1w`lm7XyI?2! zcw5#z|H~f49p1b-dk8sa1-5)_19x&08(8mXu4b^CytOwV?_7OnG-6>~|NZO0%445! zfvsA&k*n6Xkh6aPdnWe!hR?jSHvaY|#7R6seXeL4zC8+w1$9?A_pe^;={XTsPjV3u z#7|JYa`l!AeCzZt+5r2w;{fLuby#&Z@8HJofbWUiFOC$2$hReNN=A;Ml9f`JKz827aQgSv!{Z{t0ic z<(g$2=J?{YaZm?N4;~LXH5X>-Tz;|P-l#lPk3|JxAj@hB+tD< zp7YrJ|9=0ym-|{bLG?y=%#~}o_O#D&dZ5jU<3H`b_d_kx}QA zI|RHxQ)6qc<_;fy+rBeS#Lj&HW8XC4hOrh-{dIHo{Uh$cocH>V_qBt4&iD=MnelsW zZ2wHIC&|~s6MO9M{L*3V>r6SnzZd;jGo5=OFWh&)C+NgM&fgi%@!Yni$GYpx^-kJ= zbB}*}U!RqGOt=I+*Uv`NhaXsb572=7+(3ILA4}uDcR9`(dwTm_7qNW%oXFXyga7%Z z)q5pc@2u{D?P%Q18AG5o7g+E81c@&i@V@p{{{z3vOAf=1BaZj&fFRQPyFgh{zTld@>lr8Hy6Ys z*Ph-v-#zf#=0>g_{v7IG z<+mUBH)D=9{tom!urq$&!CZaDytTV^;)XqKo%Zy$=G?K?s&kIIdh=Lz63exqKTxAq z&4pU`QplGL$gA;=_3pgoNgr!J{pyP`?;YxF-kzWFovCl{`p)A|{7%kZzDb=} zeR>j+e*#Ygz`udWbysU9{2;!-JLBKKtUI_%gKu!^?R7@)(H{F}uFuZA+NlOE@%`-F z_YCYe-~vCe9&4>r>x_;)`rLQT)WQ$Y-&X&?Kfj~JRu6dVCvkmWh&z60;V(O$_~+M} zr#0vMo3~dZPN465BsPfe-`>d+`?Gcffz~^WiGM_zs8LdVwUg65dp(E9yxmn84$vHHtdb3xPK>Mrcq z&=2_H$oYM1Cgm`0O3{Cvw(K;Q3}8tA{&OEyjKWn2TN04vxO|^!Po<1hMXNMr-b{`t~p$Xg`2` zZrBS5{LZ;^Xp)@0 z1}J}X4cp;-N3Hh@IKKn#Wsgog!9@=1k@LRR-L$}c)Y!jao0j7{+pGG=dW$4qJgLa_ zXT<}JZ)g1{to*O{#*bKUd8c;Z7tl3wPpq-WUH$?+7FTm}eHPxwU4oo_+`r+gb)LD) z=H%VU`X?}-An>nX;0tTKue#&j58{@9lToJr-n8_DGntA0VGU=*fA1C4_gphCTjTF} zcd8!v-ZRmRZG)Wi{pW$cn{^55dyTGv6ZtD}-bD-km-^TG{dWB9qprXItGiL>clZkY ze5|)u^&_^wv+wW|JIVPoW=}QdwB}Co_#OX&@4i3eh!fr(+S6j_<-}s&D?h(HrIit z=Ev%FroVT}Tt8dxUGGNhv9E9D9Zw*LeQ!0!&e3<*rVrHI(X_mfb>@k@oX78sb$@(;i?3}w@><52FW36xS0N(HZw%&Pc{n$F+N{xN)Sl_t;zat^gS);8tC+#lk zooAm023qUV7(1~&mc0_IwZ=WQ_By|>_jthFK&NJs^F*$l&8aS*&hKVy-hQ4tw)&Af19x?{eXhp%4LtCjRiBrB@^-KF-?2DjYYtoh z_o<&#`~mLjW9^{M{ta&GgX#P?y4?;pSx(_wE6mMd2f4vhw~lleX)}o-!1|WYd^3ZcxPvQ!#1q_%sb1xXT;L9 zT)Q||POUWws(&u)dk=hJ-rWPg<7>~m?$<0oFLDE3o8Tus8up6yP6OY*A#&~U#83PQ zF8JC%zF=eZ$hTHKI=ngOm3?1-ub}QChYetV@1edCAA8==47dSz*!nKOCGhz3b4AxS z%&W~Yd(o`0joOopdpE);;7O*p79#+vWn_{aP-_`!=-$l(}QyiSMZ%x#r7t z!5$5Je_Otqs>8}3cOUrf>z(hQ;m_Rfv#s9&^%wH?4Qhg%oZnk}d-qgN&iW5>`Uzj! zf3!Ms$C>!f9=U#gjP2R9;K(Jus5ih~)zDW>eHVz2yG=N?LEfK7ck)i|r}ZwM^x?hp z#J1SqJK`XA&g~5zKSeBFzvgn6p%&i0fmVI(fUS2Tcfp@n>-uZ;QR7^#b-t&*eTF)D z?4VA+VL#O2o4=xKz*xQsVrSWJqUl)ssyDCpgiZ6Y`rluRcUHnVX8`*=fc3dG@C}u-^EDU;XgQlDfbh?^E|9 zw#NKO1ZXE#&VI&?*gc%x-l5>s*xxhYXRKWUE)xy*F0T*#P7SPmZt!yUwr9s5_#bGh zKk@N%JMsPQ+2470km}&tUCp8nT^Yu%I?~y;FY1lhlVukfB$f?8rg$+2h-+?`RAM4etyUKHBtR0VeNvtL| ze9wU|f84hb58|)%kyD$%JIMdUPGaXaV((?WbsejAfB+MCW|XP;_C^l4r+Rmd_w^Yx zV)HFmZM`L&@qqt8>%BX%yLsxaS#I^ShdKMZV~}%>e1n_V;?UIl<%<2R!W!Ze);D#I z^VAP;!5^`8THqUYtbA^35083#I5&Yh^~N`H)(B&F_Ghu~Mb4SNt-b8?Am_11=I?p2 zwc48+^3JgS*v)!t%z0+4JE|G@N6vez)wmJpYyfwb@?F~Oy z&&0aJhxPhE&SUI*+t<5!9@rjyPq5V&-x{^vL9I3Rbk^TLJL?*{1hMPo)OgPiFo+v) zM|(OW!Gx<`r0>M`A6T{Ou5b9nXK=^w@M_%K+L8Di`(5CysXTVX${Rbw{86jke1ZqQ zck``3fSmWZ;yb73vi!CB%vgQh$^C-(6IR~icc4eP%jUkP8vF##*BW~|xt43L{j9h5 zz~0do#~sdasTNMy!|%`yd;`Y%>hrZ~{qL&CU5EqJd+;Vd@cr*t*vI!7)OPB8r-?0{ z9NLM!q6=(`H3zTeoX}Sfc#l1LuKn8=2f{{0Vjgb^ne z68nMv_&gha?Ubv>iQe|9|Gi^M>a$?~4rh-EYF4hlC$#>Ueb~?RsDu4~-q-hT#M=9K zi=L}a-JKqR_5rT=_SEm#KzGb-`1Q^Khwr=9J=C&e!l@si-f72;yxb?PHE8|JCtN!a z)B-;vzl#Z9f2ODx)CAo48ye?E&ZFL=buVYyuMywz-LvD{+qigqd(WURxWGIA3>Ta4 z@JH>;r8REklX!pt-d8K?En-Pa7WJ%ih;)y^0EiRLTT87-DJ%yTmL ztiXpGz?s$@dc3;pfU~a}?{)&m+;~GoTs0W)Z)_j^0-A|$%^CmrjF0e z8u-6zsI`an^4|Bkds%;ep~WUSefN>~&-BC&thHhQ=h(Yu;JdTOzUu4eh5QZP8sAeJ zv-T@YvU!@1~WvxA*XK*H3tBgV z>fBY|eogKTi}t!PumOL;)?3Hg`5}*g##&$Vh}G!Vp4cGYfx9QMKab9IN9&ze_cz9u zQ)^%6=C%#%?+5GDKjlI7#QF~IVeV!Qzil{Au1+kSdHF_cZ~I<>{SMrT)pYF4)!Uwd z#`_(2?QrG;4A5iOVdWlZ-Amr%tQ(wj%s=s)F}`=#4&crm`~R=4->b3usk7dlti6JP zZ-3v!+B4SJTmrRC;ExwJ(Q(f>d~(*Bw|phL#t2p>$cY8JBnn7O8F{f``V5>hlX9Z}vc9~f9A8-$Jfv?uN!yNkz zI5qlrH1!_DeqTK^7Uv#vXEks)Y$iUa{a;I`pX39mZN&P2yyJ76{T1F>@(FyejXqLvEKhq&zWm(eNXZ2;TucEyyt@rJt@mt6 z4wMFad`INHudt8D_pP@HC}8h^)B48ZsPDT@UG-+q_QoCHr+VU_#20nBzkBe_Jh43o z)*ADNJ?5O@&uHy0r$)Wp&6@aqKj#HGHQEPxf8OMMTWy`B7jefXh*;L0yqEkP8(`oY z`=$*W`tTDwV)Y#I~RAp zaCuugXWCP(wc4sre0L*XJCU>AdnC~B0ggJfzxt3n1CP5}Am``* z0C#e}bL28tr*AHRy*|O2lh4h`_YHk_IH4K%KeWL54nC8!&f}Yi5V7Cx4IlU( z9r$x?zdu1k<2~$KpWS{V-?hO_;9bm(fO}gwZ%ed())FW7d!P;Tn@g;Hguc2HJwu-} z(Do!BYp?Q&o$2g=Z&=S)weCpnzB&6?Q};nL@ZD2h*uztCtR3p}&(+8NB0+~6*oL)- zZ?3lXrH{D@%w50(?v9mDpx!fcze~WwC2@eh{U%SWUfy_OZ!a|X^;vMosn7g6XnQWb zceP&FL;j3)7iR>q->dQZt&eDR)IUH=@>pjyFp0}2G`8QsdNQG6Yw(TzOswg#-od_y z#y9uPte0zvj{Ept9p638RXzNLIN;^XCDt0hhc{w<^Al7Yt54$u`2lZk1_ajKuE4ls z&3n|in{(9F9O9pEdrx@#TidXe$DW9-vsd+`*16v42KNL3S3Tji>VDu4{D!YK^+BMY zSo^BeAHdJd&(`0u<`S;Hf5aVKCzfy6hSfij`-B~_v}y*h?-f4qkMDcHHIQIzf9ASF z!zNH;-aRg$!+mJkzr#Dz&+CXgeuKAWif$zy!_qTrH8$0hrJb?O{yEnc)$E(%pXg$sgU|;q6)+F#g9Zo+2U%+`T*dONE zJJ5Us*3DS+8@_v~YuGE+T!;5AcX|BpSUGzqP@BYKBjC<(4`5E;`#aa(gS_{0cWcx) zVD2V|9|;E%n3wY&)#u=`9X`#`PyGNVblxrSLmNwHt~jhU4SOQ*&&D+7xU0UqdFRp* z|M9-~5BQ6@1lT(W^4~YF@BAd+@wL?p-~0Prn)v?xgIbTgJ9sz$tk$01;B)!zp~iVy zdp1y9tk21}6G<%e)mU@IpZZw)KC#yNyR*mdoLu!pQ{QFm&TzY@kOnme1Jn%2*2fq3=k5!*owG+-* zUTxs}&K-DnKZ}I(ef<17zP;+Z)RynPPGFGpxTCwtwUk+R@3HFSui(el*xx(b-yJ4s zJLg>*b;ka){d#}iH*(#x`jfZE35~n`r6ty%SnGWodvu`oEA|G?K#hGLul4?kUf9pm zlYQ|6+kkxsxS&tshA-cUC)WFgxxl_rk1yvA*41}{J2+E~_0F=FxzM7um+0HhXr&FT zb>xv2Vj*vDQHT7MygMY^|D<+c4?N&K?%uQ9U%yvoth!0g9^bTY zYli&>P}APm?`QY8gAON3lj}2ZSNE&$19i?zGz~66vyM0ZMDB!dPv=GaZ8v=P7}zt` zUH%02_`u$`@7#cE*wW)`&7XmDdaT{;BR|x`RS&E))cNyPHFEgQ`i!0C!3E#DO?Y$q zvHh_vnfiN_%WLKT*Zc2%>{WU+Py8$R2_7%B)F=3h*d0zlO~m^1&06wsv_}{Cp`M(2;qP9#%2VsC zj{ORs*dT5ofjZ}pTsDd8$j9no-1V7tg+HVBxR3MHwJ+<W8o1oq`&<6V`h50dKv#Ppn5@O$QafmhOsGYcEgc!ZFM5sJ=g=?r!fuqqS$> z`Npxonwkmk9ZQp|r=QJ4XHW-gZ#CcL-qtQt zj@|9k1MbU?;f~rQcj#}_+22@9eET}@y5q!mM|HWm8(agO*m>@D19EEZ;T-c?IeQ;{ z%2yBnz)r05ypOeKe1E5w&N`5A);sHorpxcxI_m=@y!&@@*5BY7_Jo~q*801{oN)gK zapK30JNB&ZZT(E!dziz@+vm{!Nv!^aO|1Jl+rGYsa~r&Ss2i~mSoc(O-nqM$yEJlI z`;27X&Koy;Yn(NR?I}W^nbw(e&l@my2Yrt{j$N&vAi;40zxrU^ORXBuh;?u02k_5x z`^&5I?&gMkQU@xIwcmey*}dS+Ti38R^bJ3#e**0lI6uiZtY-#)Nouz}-p|_7U}x@~ zuw(mmpV#h(W&mgS=as$AiZ{+J^70o*rvP zVrw>h%}_q2yvHJ*qyzmwDV-qv~Qzw4v}=brY^s&&=?37jD& zd};9;ekWIRvFbA6sDIG!0qte=0v2fn-f@xJ&@ zJmIS!8vER^1I~Jn^*w;I$(@0HPH07#!&V-@83S|I7qPj2Vsp*9z*%oEdE+OV5pcl% z_8LhHZ?B23PK|w&xK7{`Ppt24z51(u?m6c5;pGOfzZ$Lmp4%U25~!cp4xFp?SZj|H zT*RuqTZ?b*tk0dS@A#gcWok~n1N^b`h}EmXIj8P`zItK} z4#NKC)R-TE57c;W*Z}tR4Ezin?EST@J=CADBjJF3?Ax`#ytWy`yR*K1YG?c#^`4pQ z&D6V7V)dQl&c`mtozzV7#|-yv#2x;z{hM`g?rgm?25|uE%^BBQVC&teIr8o~&^0yG z+xJSZC)RsU__IE^0PBuD{aLA#EAM_I$jifZK~MLvf9;6%^w^)1KJVtN3($_Ym6x~v zMts1Yp6ZpW)}1^@pONdk^cQvd?(V$W1>4{zcFNPg;Rl@eafbo^{eAsRBpP!Q_&sl5 zR?jn~|7BU?4_O;%>QFcW3L= z*t26jBg@o`i9cfB1K-|q+Q25ZQ5#ri*x%pXp0Ds%;_9DRkbi@_j^MkhW?MgRbpt%~ zfxk2S&j5As>-Qdab5?ynNjUogW~|RWu^rgw_QFT+JoE1GMC+V0wgdmZ+&#@%GlBK? z5e-ECw(hCsY8^;?xsHukJBT5|*3;XUd{mW$x`ebvI+uzv}T_b`d`LzF^T7=s(DvsO#G{jycbf^F1T? z&l5e|NhDm0^}8d8Pw491vDS7F`07UNH)W|PV!Ud3S(1l*Opxk20oGm`lHw;qA7Mqb~4e{shi zy{)bL!#P`B^~090%~t2vJznc~wEV#TL|u=ibGCOp!^?SmA2lOZ@3Q)E<`Q46wfgR1 zJac_l)dc?W=Yza?|9y%(xPK5gbT@oCcWrRLdJ_7(K#jEnlqOd!|7&M*9&3B9Kl^7e z$9iJ@z*=8#hcz~rvA#D!Z391mIr)yYM+e7AxCx4@~@sx9^UBy%53KQ(m3EKR3sB@{a!im)H7!`g_nh&aHQ$Py4doO->AY9QA&_ zzjV}Hhy#uL1im|26U5%Z`3cvt=K2A&?&r^$b@Bo1-LNC~_m_RH@e`7>zbtj%1#JF&ZZYM;ajd}dtWr!h9*+~ebom}B0#1J2&QLB^`F-lIlN>+XR@ zYd(N=#_r|6cT}fu-nrEuUw*#UyZHOq&-8*l-rv4~{owfG1kK7q_ ze7OV-*w5HL>S|A{?-1Atoa=X5YmJ)vUXU|R;4bFnwQ^#}MXNKg?qSb*_gtEeZEzQI z&)5F>;r+f52f6wzu+Ffj^X1$2AMnTCjhy<6Hr6cf=-eOR1Uhx%`u7;EJJ|F7T62fm>X`R+%fAiiUw45Gge!piD1Mgwq zJGKEoANe6-{r#4`#%tf1y!t2c8C@_ANZcY+Iwbe|L>Re&pw=SN8{`cuiwFS#|^G#81GruH{upt6>Hysa|dU7^v&B# zeZ33u|M|Xl418xjs5S4;y!|J9BR-Ss`1Vb#oICxHXNLLf8#KH<{C%pv0r?IuR|nr* z#n{Jd)v2}DB;VwK8hc)_?yPpA3%C>hz`w)G>3_hfw>HeOy}dty0Y5?FoBs~%Q|CopF1~)& zW6#){PCUsUJv+R|8gmt6?eBY>z{ushL-~Xe%m+X1i*4Yt{W~>1;fTGLKUZ3NKY@MR zxuL0AIehCr1Gz~&u*VMO#Tjj4Jy*0DYp$^~tZRZ(y<_oejr*Y?e|uZ+@EM*T4H+#aR);XH*8`Lm{&is&hm3feKf@pS6$=4wIIj*t|eE1Ke0oAlRdD5y!v`K zUE*VJ`&C7>?E1rIz&7BLXU_^t?{39ozp9I^ z$0mqu)m#4gTCx2S-a6~urDnj{W1_Xzw{w>XFWkvK0eoxg8}?6l{Q`RPE!Xb_>+a+S z+}RrJm3U&Eec4>&$hn7l``&fA&4JHa=(g2?qoiLKqR?yp`>O^dacamTk`0&B%* zH1ZQ@TLSFB+9&YM8Fz4pv#)pc^S$VkKRz?H3H;v|a)0&a55D2s>xiwZ-e?oPV+-)c zZNqsPYYuGX2Juxr+7ol^=|1|OfbUd1b|yM{Jn=tv&faq4wSGqQ1I}|+!|Aa;H?_Xi z1=zC_JKxzJb@5s|IM+CVd9ArCeCDeAfj{vp#(I3SmJ4c~_9{ZlVpIY}C^4JgJGsr}%57+Ul_qKBW zy@9poFY68&%i5=r_xL-_s=`n&5ve&p}(bBlE!XP(HLvyVTAB8cm*@*Cb~qA_>W z*XIpqt^AGJj!j~B@%Op)-tPiB{3P~!?@qt=g)_c_iN=3#@15+UbtmhbXFhzJh$cqxpZV@0xS; zC-#KK&!XXvgvT~wbH+FF_EA5~0Y4{m-{6C|?uFmrI;i*AzJ2VkJ!an3dXjozjXCWJ zt(<*7;n!|$<*l<%`+`m0g!~}3&KfyCW9M~|~TI47*1tH)mUai;GRxq8%&@b>Ch z-^OFFuK$jY&G_3s2~U222JgRHoLF~wz}efr->7TbH}~)F>pSD!Z@|yo_p_I}+FKut zSh1Y*-6!G9Wv<#zEN*a3A8XHFz1|jQedWnndq)@O?LDw^{;s^D^Rw_RPQX58ajmx| z;3L<4-KFE7u=Z-?tkqZB!GqWy18xv|5BUW0T5CM@z0LLe-kzT7i#^`T9?q=a9awj1 zXzd{%vA&D2^=H?W+-)PtTmK#H6S&J)?6LR2caFU$TKBtxsw0+nFFE_z!{4FqUAx1% zgZd}jz&igkuvW+g=)l=>ft_foFZ^9S*1bI^@(DuTKL1$XiI!`J`pjeL8di-pXX}Xl zOgnJ*yFT^eLF_$d{^<;3>Z}(7_+|+=V!;|` z1=fDYT=O?;u)@#%0utZ)ANVJ51=SONf=*7oz5HxE>W>-L{|WCtfvwL8?uoBH@O@)@ zR3GfzoIM+U80XS;eEVO_9S7h3A6Rn(G;Lr(C+E%+|BgmJvX$o0`<@-%_=)dZ*uy&~ ze8W1U;g4MXx|ZF2D|JDv##z?7qw^wG?t!21#sh1OT-Cx^`vf<*JJy``H9 zV*NAJ|4fwl?kMj)t<`VA{}E4+_#S)ofCm@+I^>OSTc_WEv-F?v4S(Q&;CFmycVanr zz8W)gl1sP%=B*2Gyb)fVpytr}=dpUV1KY6;JAw90{x6WZ@5VdTcNFdh>Rr(N`k6-; z;Mk{mUMug<$p8WDt?rEP9Z$d+7jqovdbfyWojLIB>;9gM^*z^DZ{K=*;*krV;O%Rl zhHZxi-eXT|C$<4+s_(`e_d9wuYhulH&&pu~ZsywEx&%+OS8~&M=d6h@Yo@s%XV0;1 z$9F&Xxq^fj#!Wwe_w+rSU++OK@PoWi=czZ##TlEEyU|Y!bJz!nSifrzjXFPfHSW@d zGqJZIYoJdjF2CIc>#fG_mvU>1=yiFMoo2$aAjPo@;;mW-OSipF`%FzftR)wte?f zYg|08`XnB)-df(74I7D1-dqAXXV=N~S<6*FdEn2JoO@MmE`M%YpBI|01}@+JKg!-8 z)RHF6(~Cs~`iwTxXyXHok;Z4V(J-m$X0`)pmbm)BB*W_|G{!SMO>gS6=6n;~YL~ zc^UY-g!MlnvUIFJ`(F+Lbn4{15&05$MOy(eS9JxsM zD^LGU{0hwdgmg}8zIz%z{XXjjzq3|iy(Thaeoiz*p+Xi_5r&4Zk(* zKjZ8D@fDEp>Cb}RD{=XKA3bvIV)T_`P4{E}ieH(L1z0a33$okj`mEb?$v6HHJCF_O zOmghoK%hI1ISt)j#?QLSMBI1WbCK`g)Ab48_Z7c$$#oCb$=C(yYC*|*_>tymyYCg< zyX(9+?2KPHXMiFi^_+~fPY{|=fn^aDFNzU!Zj z)5H0U$*FbZC3NS|mLJA|dp2i7_B`SR-5%n1kCEpY&G)ao8IJpOAJ;?5=`&I{;p@FG zW0pj%j4u;}Lbo*qW&pFHsu%IVouN7nNem2bOTHyDZ*fqxe?q^o<<16^yJ;uAF z9`cbV?9;GaHT;7Z@S7~RpvzJ2*xsjv-jL_BD=~5sQf^;JpT+Gm_Ne>}Jl`Gk8gL@U zjMkX+v%0SiTi+F&_w)5mTpl>*lW+X2ncz6T(iyBB6&rcQ_FU96oQyoryYsMT{MPdS z-FG?o4Lbw#oS~psbo;uWz!q|>V{hXDxsVV&;vOci!p0Mb{?|NN4yYUso^^wKyJgY)Ng-h z%=a;O!M!1*48Jl$B-^(ZgupGAcmd< zOMfBPefAnLdUz(Y`PkN49_$DE3fz0HU+acCoCeY%pW#^3|E@&cxVD^9j~No{GWt$P zXHi-d#E7 z=s&*?a{Kc~8t?PKaap_16+`OF_4xMVv-a(%Wsi(){lI6PhHi}&#G(5Ps+<@+`@0`= z0oDlT8kG zLSB`1?A<;Eu6BS8+d3<}bI->0oqM;ReRjC=8hSlAL^{{H8DCNUfp|s2HAnx;8OLAI zozHobz}FXY)n-k|T*s`=5;5u&eAX}6>H#Xcb6L-rdp1{T9i@GG3^mm;!?V5^d43xc zy7&y=yEX$ku8h>SZi{iBtsD45iLXm-botgO=t^yKonc1T-+*)XS&8YZz!kwCmHz-& z&#i~#JPLVo1K*W;=Q7Wl$_@W^=o#DjtZ6Od4OC!_gnTtljxpy{7Ipg7I^dn!LB+4H zpDDI0l)$^WAnoHi?~=28!aswv#)h8oyIPF1^1OO)WS;|BE}S{C_e8cz4E}< zXYQ%Xzh3Yodkj4euIJKqV&*qw)URjfw<_>is~~4M=Q+EVia+6B(cg%r1N}VlYhU%L zU*Ic2fQ%i$de(RK+S)OC6=bg^7hV5?uHSm<`<-;~vj@IOem; z?EUbWyWYp1ob&trz^|YIYx!MXh+9jJOaB7wZD08X1dwAd`TF|z30Y$BH)1m}x!y7P znV7Rzq%mhq=&1s9efd7lu;Obm&QZQx(d)pmjx+hbFF=mZ#HH@+3CME~8DG>xqIG#hyanA_O$k@&EY9IFKUYA$b>%x#;7{mRcsFcw zoMBdHSx3l|U(xL&PuX?Mw+nxO1!Orv5A@J~AG0QWsgC{kJZ)f3L7F4lA?0=-a<>kv z#HiId)Lnr6uEdzRzXK36rX~fLA;d9kE z>T#|?{4+RbwN9^D$L|TAk8=ffCgysNZB6fFt@wfeZtYpSH)~en<`ks01Gylz05x$+PU<#I^tij6ESl)WI?W=!Eq+J&TU>jeCB3!YwA~bu7)g!p6ZAt{K6dj>_EP@ zy`9ZDUFRKk4tv(X3CFxzN9}~qnDaVk5L7?KhQm|t1_0dr{Z# zvyt!j-1+pmuZAyyK(EMtKag+O3B2f!Q4i1eoM-&L-`4z$Y;ar)`g-`a?Ip+W*oNPl zKYjXI&wkUvuSk2j%;_})E`Fb%-1*?j@%e1}>wtACx?JnnJGAl3k#Efazj_*XF{2mY zUd$H_dqY;>GOj&^HBj{J(r=E?Z;n0em(g9%>eTGD>zJ|c3BTMMh{xFLE53rA;90}{_j^shiqAFS z>p$n<^}SSbV(g=#XW}!l4QWmWDMp?7vi|e`_?!A?|ME-v@jv0eBKaq2{pbJDJHBt< z!1;fz|0Mm>^jH7&fAQD<)xZ2N|Ma(i^}E0N)Bo(JKl#a@|M{Q&PhkrriB>w+q_{sn041fMl|LmVV{po+`44ebcB5^jJS*gSJOjqD} z$Avh4pY1CMd<)W;e$UcepOvwUz7lW93F(R$9QktIz-~<7E5I3gKJ#RK!8Zt>HLQ09 zE}yZ8itqlluK9i57&F|kjh}Zq;eW>VJH7zt^WQ40?+o_x8~hu5-j_@~fajxpkI(!U zkbyPK>*t5sZGD$xKg2w{0vhOfNc+5_hpVA$_j>vQz;XZ9b++9&dd255$G*?R0)0dJ z{@>7N@QQ6*{*3OP?EP!%FQA~iCS=r6$J~4wb4LtWhy}j)#H^uQvHkmdGbRt;D`@DR zv+@e3FZ3NQji<`wW+33o@jTJXnZaBCg;-o=J)eb1`TcKDJ$5az+@&jp!cMC zMR#87HFbUH1x&!X<<8VgNOP=dPN~GXr-3ccx(QTZ9rFuP-C5-=N^H+z11rdNydT<) znEkA0orZ3W&Li%g)Mt?ICDr4PD&O6LLFbuX({0qrP{2 z!OqyuB-fbxwvNlV{vDaYyT>>?_qg20K0K2E+6C_Gg`605C-j1TL-sk~*sCbL8*fq=pXQVPj5)~?TkJ{&%)dt`3$V@99^&D-JkHyz`P*lU9y%x zTYeAx*{eUOvmxa<+g}@#3tqsc4#(af&@1rwlple58Q-%^-hysULYJGcJ&$rA#hCdD zr}MCP^n!HH8Tkd<9@cNzA=g^03H^FOK^CxqkLs283VsAJ*2&XvzAM*}^LiO)o3JZ5 z*Z)j@0-voHqh>~W{_ew{56)pt&$!9~H*{^^PiOWFR^#_m&Sk6HlPZ#qZBC;|Um>u?sR{%+TjYuf(0P>i>(>mmFuUb@)Ap zzs2^h^jg?EIXkk!HP3qXR5#BV6EXYx{qO5(>_cug4kqBaMPLj4l{j#Z3n)Pd{|p*> zA@;+k(Mw+63o?NBWrNq>l}8%y0-_PKwy`Tn*b#{Pi22;ylivp8f$d!8WiX)^qi6K(0Nk-RqMZ=%0XczaKRyli+1wZDkyGhS!&V z^e;%4v%O+}#(o3u(XHnUo|UuttlY0NegrM3!Au!FlV7pjS9)NdI?~y!V_YdW)XBSQ zbEXA7$a_%|v)>KupuwLngO`a_U?20ntNN_D!7;yc@wo~qrRO=rKWoTIz#0{4 zEWkO3{k7#=$8)l#d~@Yb;Fph)V_rJq8(i&o#C)a+UiTr-{=GJ_1)pej#Q3xCNxpYM znLvplGcvIKrY3ZAYaQpWw7#|E6{ItM0GS+@u^J;!PM;5-`?1$E*wC*)p8HS+NQdk? z@qhcEKA6ezUKnqn-jO;OL zyY%~v5{MY`e5ZZ(?%z4hnUF~iYe6^n`oOr1Gt1cKTE}m=ISH6A$K}s=zyF``sayX> z%pAGyWd#-69@_eQ9du{N*zRpZza6&q^?QaDU+8y__|$z~4Sl(c_b0{C7vE0I^@d!5 zb$efY*0Rp@N76dZtIztLyFKOu1zXOB-0@vOV9RYW?xbAp3cRoG$36JoT5n+=H+1vN z-@pVKwm#1&BYSS7@qH$4Y$iA33s8Y|%>M$az^C1x)n&|6DtvU8XOv=`SwF*rYs{MI zY25dW-s@PSQWxorGjI-TxF3B5z2eK4(chSGulA_uz7rev6@CH>u+A$!{qFS>@Yyu@ z`9990nENz(=wM(2d#4 zW$zM5edxXJW$f8Ht{k9J7tVI@CZzAIuy!S$;JK$aumWcdB}==+c=ioxy}obshVI=; z!1L7Z9L_-9p64EOzyqFn#{U6+fwQ2y)SuBkXMN^R#x9%>&>#4$laUE{ZvNh@J)=8A zim@l}M8WTKdIkIaI0K}d75NF+E8w}OiocP+ zclpik=Y;<5L2qy+$9fYnm%T)ak-w_f@gDfBPJb9@)(M|;$n$J|1zz!0;-33CN0-cD zo-l47pFu*mM&}@B{KktuWZyfgYbTb->36NdiE(Edy6?H?;hfsm{sb2M75m-O;5Tx# zI|tbpa)o=njQ8h{??s8Rzc}RIKgRFx@`Jqh$fmBuenDnrhJT)qw%<1SVu3SL#~O0n zv&)?ROp&+4mnuEb``*z#w*s0vaAxg9J@;EczemLE)9WA799O=KdbLu6Scy@$A``wu zd_FKCy+b>^RUMAA=(n$T$)28rb)4gTe|r5&Nn-ZN=+^FcGSES*{uTK@P$SUg z%dO=6FZlW$Ow0!73oy1Jwf`0IOinTfY0ZXoF6S+WPyX*7V}HgsVxH5CbbUvxR(#lf zPNeJ(mJm| zUx62w@vMGe4eVZ*JZoqt;wx}1$S=Sg@pj05Pwx8h1KV&r-Tapb$7EQrb7iMbE& zre_fPL4}i$?k6K%)~qp}m7FUaVQu}!{n?lA$G7R2$GQA2d8Y2^2I7&@KY-tWXY2&F z%a|uq#~N~7=IaZPW1MG0dcXWRRSsY0pgX%U?FKS-!moYa=RPxb)kl4K6TFSQ`N(yy zchpz57(BV3E{R0ID(DOG8QFneD(zzARAeF+pc4Cg>Lq=w9Qc6JXTE?Hjy=Oz9Z0Ml zzq1)PrX9eX3F-TL?xSHBI8g`3S*>-&FV{2hxw*V=a`ji_dt|ADtDTAe>@l8A#&-3( zfh4wrK%bH6K592?-y`R9*<%AYZBXKP|Il+C=j}eM(XiEBMR_0RYwwD_0&~1S&Q-9@ z>oXzcB;w)=dPBN5zdZ|5-iEv$DA@X}pMZSNz&=9V-wR%l3HXietI{H3)0&Cyzy(>SDBH|NPC!RIhVRaNCl+IN``-~bZuwL*mBKx&lTN$UqRmw(zvt8_umQ2}lR-|X}0o^;6K)H|ewNEDQ3_H5ZeWfSP61dM) ziN~nr4DQ4F?$PtOBHhzRq;od7@&j&0COFUdUx4p}J~81Jt&V(uk6Gb3hqKyO+qlbm z_O#!x;WlK&7fA2lf;7io8R;w$ywk>j&!Xa+mH4ffmDm}-+!yo}7}Kvb_q@j$-F?oW z5VKAK=Jq>})Mt(y|GTx%NYCAR{wyijZ@|4-Kf&*-avt~!9QWY75rgAQm3YPeK+HUC z<$LTz+&T?ABdybr&hZ63&}Ve{#@vs-4LyO0SU+=o*05$p7J)A#6L^JZuKcbOlj9!c znCHF&{kzBL>1|TQ`O;(T@rG_a*MwecS9|>&W5$5{$($oD-&#Ai zGc3qcZ%6DIpS*r%$OT>?ulR4k^H_mB&V5wkesdCjeZNG1Mp`ogd)U`=-O!y!WNi1D zh;>wa*ss8z_OYIIUG`D0ahxa8`fvC$vLbJwplkbkh57bfzzSYK7)MT*p$EEr=XQ>Y z?j0^lIL_}gEyR32)^oKOzHiCvwcxE_0j-h;&;15GpQFAm(Klk&?{^C6d{_L=;+z#< zpB10AXMC@~Z$L)cI~{Vqpo1S1^X|*Hjx+lltTiF!?7%&&$o|a8`S+l|1Mn97)^>(K zenwhD{uSsmW*_&X&py`l%FICIT@e5`pd*L%RWRYpNb5V(H*lTP^J?_5$9w#Nt!{3`?=0@C^YOXN zGv*$gqX_Kp9>X@b>8p^Dfo#OJ-9sGlgl%3!TEC*lQ|ixVy^P*+KBH%1*3}Mld#n2# zy>r_3bPv``G5&3;svASzkm?29bIiyM={%m(uc%`WbN(JZ$N2A2r_0Fi_Y=RlJ)d*1 zp79{(hChQ9+uGLJ(Qn#te@U+A721CoJikd9nBzX>G;+mXqHp){?wGTpyT2T%@4X_8 zS7P4FLXLI<_B7|W)R33JjNdsj@>QJ~QU&mW_!oTc%RXYezyao4>k8!l{xSA!Zrs1l zWe?W^rv&|w*74nnI_eaB)^b*#d7tT+L!b4X@0%wi_Gff+D!R4oC(pb-$@t~c^?jJWgc*w&oT`}>D3-@dO%=Th%|$ul=SjTvWRo}IA-7Hstl zjP+H=H|!n#8@l-$aracPona>C^Plh;vrY!a&hLu#;(k0E_E(OW9R02deL;^pdU?(T z`+Gle^^DIw*=GgwVSAQ8B*%Gv4ic#NoIN2^j5GC}xq%-%js4iqdOv{IkggWT-9#D- z_}{|6BF*=m%KDMs6Q7T{)+odax_hoMdNh22&p8Udif*p8eBSP>;WsaX6~E6@u1NS2 z@;yFlOhB&w8C~QHlt6MXKVZuf_A%}=YVZTNJior`R~hm_jxF8Fh z4Y^&$tPL6XT;@L?G0!An>wAM!fzM5T-w)xJ`_Ky4}Mp6g1yR1{=I<`{cu*ekHE=N@t{0O>c)oDyU9EC*e#@dg$+`WAfl@84GB4fJOT zrxM#hM&FU=GS}KsheugK#5j)xw`iLq_pp5r`p*jN4ZqJMk(0rD^j5K1U(mfX=HAeA z1)PV^ti(9uiavvi?H<&fQF}w0^NR=k%a~JpMo&l)hkrw!->Fac!L^6JBA0cQQ~z-< z?HpsjD|yy-h6x$;4dS9I@yT`0-WOdvW1HvP{caw<9bfN{ZS4ZCpkb#vV&)}$8QVTe z*I&}xI~d>%ysBRa;NQ`2VAE#xpMBWouEtN!Fiqu zeFJjr+mP4$h_6U(b6<(=;IvyEzmF{VGMs)l;Ptxr;_z2u&ZT@NE?+%~I`&|{&-f?c zzT8U#g}A-tMGUCB`u#^%_|9ctYvww54SmJm&yO1BH)1`Gt*po-2dMkJtS@Gtzj`rwsxp>kGa$GKjzflo@zNQOCX0-+9QLT+i6LKYB|2x&QV-4*N>=$VVn( zuW%B`G45>9&zkhtF}tYW zn8P}2C<~Ay|KwQD`o;o15idx24c`p+hHUs=k@ovNPwCwOOnb?X%{!+-Z6hF*cs^ncYx%CUyL ztOVBbXL`mbr$19va}$u`TnSl0J|+A0=k#VSx;1+osV|V$Ycb*rx@S_fiD|p+Cr4Yq z`j^W%bNTM2;df~Vn8aoD`;6Z2B=$E?@WFz91&vq+E70~_<>;$1co|vi&?|mtuztgz zkiz}w^VzQG=5L^=Gk+l`;QM`X)##%CUY^klu&&S2{Qe$2;NTBz@6>|KpyIQq{24p2 zcVnEfz4~5=cMh>_B`rXTa^Y}WKedXx?2zHQ&{~Gx7c)E~> z&;IHz`#PU_AK;kh%$t&)iI}>1a?BO>Ux}sQKR_w}Grl)?{Z1iM9k`;uCzg<}hkr(H z_&iVb8}RJS?S06#_Hr5XIlY=_$pj2q6ySEGiz1GxM zMgNYRE8L8(9q8T>^`8=((LE#kxvrPN{X1grrPcAAau5X0T|q+^?p1k11~P#!=oOhk z0TX;@^Y?c5?&>qfIG;i+(CwAT4Q%V}@Eh_Y=r(@yXQcTB^trKp9-_wZ*{`X?S1;%R zKcRa!)Vqy83wRrT>(wLX+}5*BAy$Fsqy6qFjakt@;p^Or4?QCbe(O8?1gvde?f>g> zocS@EJ&k*AL0%zG-K9PsdP83>II-33nFAX*SH|Z)J`+#$$jDDXe*ihkT_3Qnw)INj zV-Dr3HvWFL*z&xa$_32m@)DdYb_NaId!&pw&XQSN?%Cf_18H7EZooaZ%XqdEdNl@~ z@#T8VmWIz+>@COso@D_UJ;1IGDtXqmr!^ANxdPp|^A>b%;kxOIgMUGyS9~9Ub?vtv zx-n~-YaO45KJzQE#|D1%824SSKKpt<`#VnF|7jeb@dd2J+@n6{P1N$8wok?0(dDTZ zuwk!A`JaIC0<2^1&!{hd26EIhex+yU-ZGdj<6c-#uJU5WWKCm~m`fkIxYqnC9mx^@2qUc)xmoEBrh z@|{DePu<`BtmnDhzzaUlK7jznTymemM_>>8?&$iR>6eeOPv>$DWA+L36+Q5qb45x+u?pTB|w)WI7 z-~EaSEWkPJ;hudSr4qlxB~JwUXOQq4b3XHWj(*lF_DA$w$2kPHy{<@CtNdX+##!0F z^Qk|fTQ3o_M?`{NlZ*P}pu#@|$4{ibxqo+BY zPdry*$ZoT*Nj|oHt##h9hTpv1IR0J(j>~zzfCTiv;P3fJeeTP1Ea;VZ)G-Uu@htFv z0?wENSx3$@IodO^Ko*eEZ$Q0E^j60`a_;_YkoxT7892`Z3buLM9SIWp3)0_nGP-B4 zuL5iPPRX0Vd9K#fpWw<1kU^^BzV$tDKJM9P(BBi}2O!5BYks;g$J01p_wMgT#;tqB zZs_(k=6wCV!>8@{Xw!EY{Y#KXtRX#v-w^u(FAID#(s~6f=&u+2=oQ$*yc_brCC|ON z*A0$mHiOT={is|2Z=c5RyrKPvQMaKd>ZCf(yP=D6NNZ;FbRRI!`YvZFB1Vr1{|fH| zxT5bsP9olro}GL-?#((YP|9&n1@yDW{+}LNc1-^0B#i$3&^=>7~W_3JyFiOu+Z2Qqo?E8%PC)_4QP{kilb zV4XtD=kmKlcTVTZ_=U4f;0C&%nDv9$hHSu`{%j6uPq_&w{pQ$XI^uyYUukW5?(G-w z{JV?2D%dY$Z^mYH_qQE-?}6?dzBfC1rjAQ{1vT)21RC+8jdb5~3wVWh1=bHB&)B3- z9&_&K_SRSKUwcMi<4gG5$BN$1;V*`L!(Q=QPrv)J&lm6poTY=9b?U*hN29hqLmfD? z&rI7|_H&;F-i}@menQ%_*CK9j>v@JQZDY$3%Q5W0H$9E>@R^H`z~8~_<6h*NcLm#l zhR@o54;G}g-QzR5GwPeL&DH+?=N@_n^1SE2r_T-DzUBlzYdV*(-;6&YzaT#V>)CHb z?!cTB(m>ZwV z|2sL}W%FEqo2=zL_EC2aGmtZZmp{@|j6S7~_6{1usy!o8FjJ(b-;0cNrVQU2U7{EK34iZ_-}~b?(QnGh2{_&bIj$?58Q9ZVtl{1ByJr6aUh#X6 zcJM3wf$Yz}QUKkRFXQ>AAPzheV_nCt6&u+1FRAfL%-$=IR|5&W!!s!Og!$f=0LDFA z=jittz1PL>UMsL(!S=isumNY!_2@y!_-FjqP~V6HcoS z|NUcd&N}8~IO%EZv7x)CroSHZV9R&e`wgtXUJZOizg~>8>fl=E2h=n!Dt4$N3;Y0C zA8@X&Q{x3>copPBPjIvoXy`lq1R5yV75T3ogA>?s8NYj2_XGYJX?{Tl(s$CBcg8dK zTs^A#F3BhrcHi^=d-p)LwCj%EZ_^UMg`|Rb0&M8XVl+2d^z9BgNNz91`o@LONJdm85v zk!U>bd%$yfX3pIE#gOJ#;u-t5k8wBDXM7#-ox@z~WhL-Tykjowc|O(&@Buy7ai;>A zK$kIU;^}+0PLCb35^o?^{7Cor8SDq#r{}T5Hzq1PbOHZeKI zUqRq^Kc9d;Yie8beD3Ko&fpC>6U$&iFG#^;cV7&wi*W)pYK6nTj%`#$SL5A^<48= z^7QvPk*(@w)zQa1Iagv8IUxfX_}ybdH%Hr;_onX~>B^NqKsjRSV!IRb*&-{pdG2om z)^XXd)B*j@`3m&^_Az+Y2sqARt%Z2TwqJj?Jhw)y9=5eTSL-Ga#N^NTuSZPZPak7{ z?as4CuW)=`Nuc){>M{Cz))jq2X5@OoPu#o$^f})O@*%CagG5dt{tYt2wMRpmZ#>XF zYxM?dz+uU;hH<5F@1^oR$T7Zk{>+$9R|geJ#TYAm5ntdA9xT-p74-#V5z-Zx8D&!~)%Y z1bKbG8{l z%ol+_kmWLZd9VHM`|Yr{wQ7vM#;oUZ)`Ts*2Me}6w3Q8P_+EkYzJY|V|2-(iUi1gP z4EijG-*dBm#qaN>u2RQ53$dA)ITLaL{wxUmDfka4paMD8X-Ip02KwFG4&9z2_2b$d-9v_yhV8#u@(IatT7=Mu!Bs_d4{*kuzj}Hdm|q3t#h7t#UI!W zxxu%F%V#mE*HL>z&w-7!j(Q=c-(2guXZibu+{gI&?z0Vi3o*Iw-?;CA`TCv3-ZOB2 zHIS_L_;ZX{K{o6eB=kN5`^t69+4n=Pz2r{t+`oDty^|}j&-Fh14X&{aN8MRm#{1mJ z8ejeLLe2#~y62iI^J5pJXC}wFCw$g1zw?O~bzmR)3mkcI_^p+XoX#ayiBGl6Y$+L*5@Q{9hdel@35_7 zp3hxfUiYcjF_X`);LmV81826k{ssSMY-iu}lc#@1--soo-;IX98UxPo33wNsJtO5? zb45>Z3UC&=)-|`HXLRQ>ui;y@fpcZ-9jV`*-#w^%$6w(f<$gx~9)H!x>hVCS;~QAf z15N|>a(`by)jrO_?~Xj{nZF&ez_!kebRXtfCxL!P$O)isUchbG-qUC79bAEZGBJ6s zR*#=WF>1fSmy@ua!Cnh+J!5+Yem9)o8nuoZ3F(=eC$HB>&-niP$M6?y&q!`V&*0RR z=?Reg5xl^Y@BI1_`h*nSe;>7T9r3_UkMVCX&YR&Yn{igH*cZyg@c-xuC-y7Y;JpXm zkuP6N*bQ!=XHd}fuW~Pk4?BT=rudaL#?QZiBe(zlAP-$`RkEykMYm=FXYctkdIdiB zu3VAMlaVet!d~5fJO_KsPkbZ5^P^Myd3`Yt$n_byUv=+Ay?{D?-|w~1BgPEoWYDlZ zyNYy=%RTk|_>7G0NYCU3tf%eVg_y{|{DgE}iC?j;;|#5KcdxH*ZhDN`1>OHm>pS4@ z5e2@z{2jr+SEd+q8JqA|?7lzrForD1xQ`hE+da(KNq@x#p36-RSV6^StN?57myy@d zXLR?|i1pu_(DfO6yASDo@*U{+SDjdbvw+jBV{5BoM&AXE?j7bJ&+WIrGuS ze7Vm31!+&$GrHUaW^{8xJH{-=Jf}Z-8oA#hjma;_A3V^>vA$B@8Mod7uAo)i$kz)! zfSii`31oEV_YBY2NAwI{?jyHzpjk)VeNM*EUFyq)pM6hbhVsA}qsK}th*{tK4e3l9 zsOS}51|P6p5!l?pT)*yTA!e?%#Cssm{9X(B?;hzNJ(FIaoSB$wMfUG5dJX;%_n{ke zerx+pDwQz0aF&3ZK>;6#dw)Cw?OI1{G4}!&`-U{8&slH49>yB}M1Io7FO2U%ZU*Lg z_CALM|B95CfX{5huK4|VsL#JIHDs#;@31}349{Ho#wMi8`ub;Ns$YM9m$wM~0jxP= zml*GA|Bg`G_tE!Ge*fNcN6m3I_ZaYZVi{em_$T11m3qYFB&0K$bN-#rne44>#AbSV z*5-L8ufTf#-_(x_HPAP(fE;61=PTIeBzz5cCaLlt`~?}{i+YT>^Zg3Wj`W$oA+4E+ z8|(W-H|9LKj@g{STAA32-yC&&hzfRma*X%6Vm#~QI`|Er-1(vtJw_jGbG0*4-TrU* z<-7X5spDPv(c|lR6gUn44BtBaE+4w}dTng$W~9sK(RHND^Vcr*_+Ig>S7O$+enKYV zSM&)f>@y>cHKe@)oE-fD3cjL_wC@fw_LZC+tiU=q&<_9KJVyV(XPz~zrJgS2Bbz!) zpSc_U9%HVbJ;rnYzmdj!ZXI!FwXWX)`&Hr#xxzEvKoajj-X)jHn8{qT1Vz8^nEed7gw0M{pYKYtp}H$`{#y4Kr( zeb0AGU(b^VoYii&0j6T0)X`|$B+uV}0W zSD@_uk;cTk2leXM+l+q&@A36M#I2jKmkT=e&F}LXBd;JmpZB2Y1E~((SBgZ{V7^*fLPu>4up;@D#y+|Tf>?dOMIP2Z)ec=-GjR5i$M1*_XDkt z=bW&uyt{HD^gO+z*0xTT18R)(be}$>D@gF4mBbpld%j(ez*-r*BF(kF%et$6I1_jS z-G7`%t|Pa_$a8M<3;s9sIBe_qyNG`4`17=wgB`%LY{)ZiPkABm{}R5>YQgWlN=SR{ zz`5O{xz^DC(}QcwGtWT|G4le*vzF(T(XC^ib9#Qt{x0G(&sm&*#xB_2InUnuH;|5g z+lBfu&cHs_);2c|y}z&M75JUCp1c~=$F{eA;|Un+a|O1jDuk}tRC1VZj(NdIo+SvUOsyeheJfU}>zKu`Eq)<(C_Y=zcsrQPs=Zsu|^}{?M$2qOzIeL$kt{memJ)Z>Z@fqnn)=Nmw$R2_J z2J}^Yy{0BU`v~*pdPWH}eZZdP^*aK;;!9vcZ}RzVhWY{6P(W zXL!S2zzeWP-#@zVk9q!YTRgkpB0Vc-)3)c1G`BBmE|<|eqr0r<{v*cu=+D>{B=Tlp zEU-Pl6UOb8p2qWQ=nbFGLCy+lj97n%h}qk<5wqTlc7VfwMSdWbfO~&aXW7ehpRm>4 zzxi@lwr6xq03by8EzxgZINH z&RIOj$LDhH9AjSVh*=-!Z0r+Qwd;WkN$@i=@CADR_NWW|zVrG&-p9TZws+3j1&;Es zAF1!hKJKg6L)x>DyI#h-pWwS6XE2tC^>f5OVF$?Q4ZqUd3@W<6D|rt30$9hrS=%#o zj%rHhj)}0}0#u_VGNf$lpK43^R7E zqeeo0)`l;yf^w%ncg`C2kT35y(}Zr$StrmvC+BxIXEblsMt6<|Cu8^e=-Pqb zncs{X!*@k`PEkkyN=(1@H;<8%(Gz~_3eQLE;0Ek*=F7cerx^3ech24i`B58qkA!>I zPUw{!?~=9Ozy@6UZ-1nl_Y5{*?N?CXsK4At%z4kaJ!fLI9(BIb%0`@{P5_e?@n-7<=sdLBHO|Ihvb6!58K!fwi1z0U6tx z1tjzd>CXV)g;(O9SKzN8)lpCT9r1)z*Ke&rPRJYZ%y!W5`ON)Uw_dF#Af0hW`s{iR{;uQ8@P!=D?+d{q?PvQLhEP z(GxO*UKh^4g?p9Dvfl!y&y6o)yu&y24S%UA{T+P!sOztA`gfXm#!lc0JY#w9L4#|p z30-c!9Dh&3@BZ~G?XCQTy#V*+zo&=|{R*0OF4SOF_gAn#;Fs_3x}S*Y3vA1ovSxjEAd(z83`p7r-<4BweHIdH9Wy^MYh-Fp4`6L-$eLHFNbe8%p}d*b}Y z%H?>z*zVt+=G(*lR_sj80Q%ft0_HoHvu@Oz@SSVI=Xs?%dMW(|$hDTUyF3SHvTlh{ z+uqJ-ey-ylUdgS*<^2MEyNub=p%-{JeD-m!2G?^|I+ryQdcOmIKxfrD_M{XE&LS6_ z^?e5Qu-#KJ2l)zaz&vy8;cOZE8NYk--vCRDJz8(Y_XgaT^PJoaXTFT{+3`1ORD6N- zoa~pdS2)TKNcBKB_65w3(aU*H{uRIf{}ZJ;@Pb{*>3c#h*w%2C0_1(X9R5?{Z*W8* zZvKs!`xOP-K8;vLW_)rz3#Dtt?@UoKh<`xO=o8X<>h7WHC(oMwEY*n_ufX}8^OF+J zyT_>A`{D~Y`(?~$-38ql&DZby6W~Q3u*P;DvnO?Y<~8`v;@mrMPV=>c_zm56>1V*b zB=`;4_eI=33$}Z3M*9o%KY~Cn;@~87d&u9B*2=(o@-ncVy^}Gdh&pC=kDlviVlVXb zXQR4v%9)WV#vZJBJ#0DF@jlGx_VNr0K7Y14lXWXlceX1y?~gz4^qXs*y@Wjda;=l_ zId=f{g0x0KUoS||_m13%>GzJSTPNuYIQY#E^m7F~{hg=8R+rO2f7bMJul@T++*;y? zPw=oa`b@kYGAcfD^!MI>Wbg0_uty_4Uyf%3$6Eak{Kd#G#0!4Ewbl!4ysX zupD|qcOMOzkokV(qPq`gHNFG6=33vgc~8vM`&DWHIiBwdU;hA*v&U)IVl1H5#u63S4BBt(K z@(Q}&O4o|?yv$8?)RyaPH{j2co)bqLx!^n36^>kIEiv}y`NhF8x91agZs)O|{6@@q z?dN$L_pCEfz}<8u1CParF|| z==M+830r>#o~v=~ARhXV<~xfB;&SA_d!+Br{q-_p34bHrdlJ5qvjO|hAQ8JEdku7- zK}Ie}^Zb3nIpjKryk~S*je*xY{yO4*N95Tfh-Y+T&ST9RNKeCWFPCxemd~RF4p8yQ zZ}_HzXKnQbpU*wPapm9-uoAPz3eLOgep?-N&;4508s=B>&ifGBKz|Sy1-od*l_4SPuk#xe54pioVd@bB?cfxz_QsU0Hzq z2qZXt_Ac1D4$k}g=xu*{WMVtA>&)M&QxPiUOfp1Ogn=8Lz`|Y=%^@W@i0Qtv!Yy9*<9-MyG($M8iM@|GWb>Q8w zmp*yl=izVI*8YxqZ*VGbZdZx1PrnV;?KUxM?ASN(j6EZ-$PCicSME&B1b4zO;*QM< z^cG$1wXC<`D@r)-DWhB8eD8$sj5+%4UC{mhH0%ocx#J6T&oR(fq;a{%6Vm-@JHrNc z;PQ-89kW|sZUbM?-#x~CmFsL1KI=Xo(m4XIXPwZE+1tC9@ylO;zCI`M3i4&_Ti=Rq zUWNY&UL3l)1>MzR^!Kb1x-|>(ik$K3w{AfOetii{paFU2oqKTJrd`J~^mj3PInQTe z!g_HToGO?7@Ay|>u5}ZfFG^Mp{y=v>a$oPGw$EPOGqaA*^olNb2l^&#XZ;N^=kWK( z!0$X&4%omi@jI9G-|pkz7p!GZXDQ?u_qpBhZ@@YUD9>+l+JmcqcatJ8$5H%zNI4DI@tg4RWkiu=RO|C*WR1!e@;Yyn>3| z*Jbci9rxnf$2hBx`2D@wo-45D^-it82drgI?}_eysz84M_Vg^gH*yxFbsF#tJ^*XU zGiIMFm_Vg&BW8W|j8DFM@=Rxt@xLJ>*w1CusObA+)YR|$-Rr?C@U5}o^P3dt+Sa$X z_=w&2MGa#G-zTK!kkNA-z42A_o>xb|gwOfJ&){1>h*@X7j5GJW=>72fu59?_*mnZv z$|=?a&f?iM^s72&mC*ZLMmOej(r5oSIWcDITx0r-vs#QX-dybf~Z3eP$py3eM?n13VRv$wuJo7mtzoGVyUeR5ig?fM%<4oL#diNbMdlq$amCpYU)Jo{WZ^1^) zUJ+k^Hp(-1A-*CL*z{e#)|20;@%?**?OX*;#_v8qV>kSP?b&AZ`DxU&?&}e+#Q!yM z=dj1wr`I!I9vsi7-?=pEdM5U_*8;~?V%!t?8GiulCUs)w`#jgfPVjbYd-UgwKf!aq z3H1HJQ8xVYjTPXWfwZoDuArhfq~8K_?6U#$J|pG0a>WH)=5`%vuC{ZwI#7_t-`|G} z^eg(WA7c)mSB??O#2UK3*URvCNj>wNJ%RO*?p?TiUUC}f?{bW1s(tQrl8fy=t&`F1 zr(d7nl7!T6uRvz(26~MPHs^9BCeLr@4PQlHkluawZ|zOV`hf0vShIj$6Q4bU{J#WM z8|nXjvWzZoMv90b_4l*IwzmCT&f~LL!FNw%p19-3UXaeZfm%l$@A6sK=R5=Ju5hfG zV?3*m*ohqLyX091=0T`R2PkFWKcUwRUjhAI z;hT?GKif(vpkBZPPxKy0>o`**&${}}&-lC>30t4DdY*y5fHU6zj*I`~{Xd<(ZLDNl zde^rmZV%VyCaypmUos=KlQhxDy5QaK-NMQ|b#i4nc!;*d7*E0=m0_$5WJF5N*cfD@ z1&L52IY|XNQb$8|If45+O?}rk1;#+@vP5f&JNvkvu6ZYfAAbwmquMJ$$AnAcr+8WCelG^VeU=v##-=%d-P>UA_yE z8NaZne)n^V?(8Xg4>|M?Nd197Bkfs$=QkQV?=k0k4V=f=g_8V?ugw$Pjt~3sd5?_; z-OpSsw%-l<+nt=j0p#`>s^qSSeVv@<$BqijL+Xp@v)X?Pu7KxYk1g_>*0S7Bj(!dk zx-mJ_3v##a{bq6wOWi$g$QhS>`{=V~(8ro<|Lk}c&YQ7~dv`SwL!yf%rq6q*wZ8-Q z=+?L&W7f7szV9*b{1QDQ)eq8MsrG!&^9=~>3Z9u4T_5oK?v>k_w%C~*_pWb)G(UiR z#@vfM?lrBE(a)fu-#+R4O!yj%*WNpjD{8gnZ+C4hI;VZAaqPp4)4D%@O8xH3*a&tY zfvmJ<-Ff9p6+iR2r*zWKav;zC?$NmQa(V_c@)B9`yC;8B%)|rfJ4Jm0S3q7lf|`uf zzI8(F!eP$|zxIawb}yAYbZ7UuPDuHT*>41LI?Hg@=P*Cu_yYO96Gv>ZMKADGV%C{I zgTN=}9+)>1H_v}}Ng!kEH|O&w{kdMD2mZ!E?q=QVX7m+$5Ko|zBb>1ka}U;>f-5i+ zyFl*Oy~hr{fDQ0GJflL6z2|j$e~sAgAsIu;=e$aN-l;RmvBH-tepW@7YX`h5^S9su zx-z1lwVWe%#a3FcUO>inK5GVKiz9>BUP*2OtHo#Sgzh~067mx1-sb3a1HMxrU&2?x z9{m*^UfFK3eP9ZLAQ^48&3M?-nzHX zdwh%EoGrctHb8Fux5#gOh`Zz*scUyZUIEs+g8Vs}`u=BVAnqRAtC-OfGLdKOr9O1$ zJIt{`m#b9zagGCeCZ}OJ*Gep{0kL_}tvftZ<|g2o$m5#KXUTVoz5)8&i+Qh+*2!1# zUx2`Vja(VSU(`XxukHNK9{3VIIgMw0&XUO4X%jQwelFkBKMQu`Jpr1JbqE4+_W?hkoGMPhho3XOeG=t*o_wr;^>yW=q^UdwCAtyK}w( zBM8v;M?3@hoF#xO6Tbv+zywNNo-6iY&l;V>zQcLXHD^Vvaif>@!b$FXCm;0Pkuw5w z&9Uc9&V-!s8T0eoz6!ZuH^;L> zx6cb;p8lx)44tL6Z3q1_T<@t4Z0l8?YsK| zpIp{E&kS5qn`n1wZp8BJQ|t3M0r%%jEl!Ty;}T0JeKva)Vl(yTTH`sCv;N;lWnwGr zXV{sX#)-5}+Zy?kh<=XdY{(yhdG5`aasnA&>skBEjT=1^v&PyjIgOXPm)O1^?kw12 z55V7b^2le6tJNH&zS_^snI`gFa#dmx{ohkPx52NA8U6oEcoM z$paOi=j46Ad(xkg0o#4pSB``XC6LeBboPPmb5)7C7xMj&@YrDvek{oCEtYh&<^nJX+XFucm%vD|?ok@Gv4>B;e#n;XV z-FcmKZuhYR2~0qa zufpdY2mE_|)UA*e>5LVBth;T!b>7h_(%%LC|1IWv{Fatdmo&JqYwyXZ`Ngea`((cR8nXXv?AA z`jb;Zdv=gB{tdDOdpzkJ+E1v9^)d&x`%mQhISTLwct+M`P>8vw)(gMq==V@U?vMjg zy^y};+W zS+Aae(p=Bo9_g&Z8fSK%Q)G@Fza(!21^*5i_}z!|&gk-6Q}9*r3cMtikOgU;Yec#q zF(_+ux99&Ac8cD^J?w$`nH=k%fdQX&^4>lnm;Cl@BCC1$1F0^@mR$GbUIRU8qdVJS z|HdPz0q)&?*2`E-wF7xc^5eQjC~8;`Toe_ z3v%UeXGvO{6<-pyKYt_nr{s*l{)b#=3v8cf_hsC9CFM91}-EIBO z>Fliz+c}*vgF?Pb85IR-uJ#`3UC5u8Siu*>e9w8OTkUoK{BO`JG2`aV$Xa{7_6X#0 zuhwPa6LNFj&nmH2*w&d-k;y#H#hC-!Uiy7b#SYjb;Y-BK$)bMHdn9MZHxNI}Bd7X6 z%wAW>#z9^I#y?y)5%;q!$LEomoLN+OtYwYFJD&! zB7pN0pcEOjoRei=+=F{;G3Im46+6h4hgd={Kxu784#t4F=D4?4*nuASw7p~V#G%GM zD{?0A9$eP=zu$6$Jp-TB0mx&Gatls@wF$^Sg9GoYC3kh!^F1d&TjteRW-AldU%)#j%WTAr z732)|X&3jA+?b*iDOTwo=z#QaSua4Zq?HOpBGmtO0&%CpE7U=}A+^_X>?fLG*b9({2W9u{V0@~Ss zG4)k@O(xH0AW?UTeu``hU5--wza8F!E1{?8=d>bEPTxBjSwWy*;Fl}cUTfSnSA6Qk z>=8gdXO(M%l;@|k(*1f0v^32%QAs0*#Z{DS3IZ8-M-aH-WRcFF9TLLvn}S`XZf0o)GBr z6=X$f8$Xdh5NNk2JebnjvWd+_Ja`?F$=@7}M2g#SIH$dw-0*1E3((z^Fu zEz%hh*bp<%USFct9_H$wm-rj>h;AEe3;V!P7QKCTYrmVryg`1ue{x*yjB4MroWf;4 z|Mx%1nW^zS?_dwe3Y;qwt3fil?<#%nZx3F9ARhRC7Wljl_@05oGUo;UIl8A{+xG>i z`jB(&&$)d!K>kZ0w?3Dgo|n1C?NfpPanJVxBx1&`TM@6w0y4VacQ29l-LF$ya&PhP z@y(zB&r_}^_~qH)6P}@Rj;?!F&SQ@R0>Amr6XdH8JzE+E^O!(|?jvVgE9@=tw zzAx}u>lqgFqVG2$eb!dQ8YfbHweDv%le@)s`K;MX&Mk0%-l6?G1J{5&+)G5~9Erbm z()SbSwUS88wISbr(}Gm{xyo;jJ>{Q}&S!2WXOArStiMF}E{uQUyw6pr^*M7!mvh>; z0Oyz6Id}N2-vav#)T=wEeau^-e+hp^no|U^XY1ZGlXHj}cg@7Cb;h?((2>^J%enmf zz}yQklb0$IOxSX}hIQx7wbxw{%hAumoPwT!_xDZUXX9N^$=iVuBv8nywZpw{Kw!T{ z-Xh-_E`cjY6YXpQ-Or74r`j=~o3qFNN_~0$0^5FDuv`54?7PDDtmMnsp1U#q_Alsi zZjgm|fI%N|=l9t)KShUo81>bD7F%pM+s7Z9PODwnkRQg`fmcK z2+JmzAE7F-fC+p0W z+g_egChojGH!FM_q)Yu2e~QipRThsWk zeGZ(@e6d2mBJSB1?6049j(Le6_F#>CvG@aB-U{qxo@b(-gFJP72|d*g@PJp+H+2l}iI3i_7#9XMC$E_pWS#$DcvIW_wEdf(b#wFc?uybyB*cC7nZ z_}P~~0$KaM)cu~ZZV#N}3+Ca^=3={E5p%!JFvs#PKvv*vsrK(cXS2?;s`zGbiS8Lq z$j}$$a1PFPc;3ipTtUpU@jJ+wPN#GxF*4yj>OBd2{rA7&FgjXaui8|<>vXMk_nN!d!K3AaoSvuUY+|IDa9|5q9?eSS-U!TQ<-x*!T z)#bK!K%dbE^bEYq71#nfGO??()XuY@PfOgo4Z15w?>%Gt&ayV)6LK5>-t#`E^T_EO zr)T|6{H+({RB(u8V(zPwn~=VL%DU%e@~+UGcR1-i1K(jUau@7U69nX;+xIPU-ymmv zan|=~ZpL0=oA25G$@4y2tHmeRjym~$Zk*#5Tobmum*DV>%Wa-~@@j9;2Xegof;}Sb zVkjXL?2cJJieLoQe33 zyq`QrNBXYw-QYZOX5u4xGt&EUufqNn+ulL!;J0tZUK!I5%x!bhKcVZhzt8SWUak1? zP55SXYuoJDJ7CP&J(nQ<3@L|mW_0ZXuX%FnQxEc`5|4NH`wO;3=QyXkoFJI7CCYy zCB6~3uO0C04tW#)8D!u)XCQ9>2_(=sPzyXcmU+-U2k(1h9`ghHo&fI2_fw93-fdsh zdWH#~d6W70)>q)1=E)-tv7Pq1+scH!)kX&W&Gmm*_70z1 zBlZB)?XT{=+0QfCtot)!Y=!;?6ms>4AkW%**16<=iT@SS_lD1`pI!6h%g8-4sL9BJ zoPpfN!WeSEexnYmab!a3H*bS2e?n>-+Y|eKe1|?Q7k?$EE;*OzfzO)u%%W@mMe6jO z5+C77_~hQ97yNSiEC%{ZWF}sL(ztLY_tEN+@AoB^u+4FKKGu28GrDtr{-mEn!Ef#c z_-rKXS7*I1R^!aIpUv0K@e}tvuYkOVdO2PIdF<=?DaC;9tOdAVbA|5{&qU6w4_PY` z-Orn6H)G3{k$%4_Ua=q2C<@(=5j{%8DKWW^_^^#idz2yA&mpewEQedK#O$eDos zX8hJ>WGv9GKsznEbvC6=+Ror!to+LuV{(-L!bp9N== zJFSU>Al^7QiwOKcuDs~x8J~f?8*OCEp=Pz%zL%VJJWGw+XT-dW4D+MUz~GnIf`D|~@;K4bThyD_zAm*8=%Gyc^YAO2e>y-xcIdqh^`C20BNR$#rg8{!#ziwyms)c$O@ z8Fk?c>RT=~<~yHfmC4;AN1$wHguGn$KJ7UpU(O09Vy+Us)_V8kyxyPpwKoU9oICu1 zuDnCcy*P_!-fEqPdh-*$8R^{KozGCLmpFEfzK51ep3hlZ$V=kZZPBfj%k@)oME~ug z+!I$G?oO`B9At{l?>QB0eOK^Q{1rXxLw8@pNopK^#}@Ku{Px@epI7%gSL(2RueW)~ z@i_>jTyM}5x-%D~ycLJ}pj7gK@eQ_p zYjlnk`t6f$TX%sy6O;QAJpQt9LSf`-UWMyo`C+Uj?}kBZh$M0)<^XBhP};s1yGw^$)>G*a_jqiumQl~V}T6xX&{bR2EDj3KgPKd#gh*@L4 zYen2WxTh4Lb{;=}F6Wb1|A=%yF2A2k;0E&hnX=#Ec}V!>6xO_TLJ@hMjroCe?jUc6 zF26Hv&Ccx-~u<37DsEzxWe&CGH%rf&Jvswod`=y@tQwZ`?@F$9tZMXW)!MPF-yG z>5TfXKw5HMlC#5>{{jSl`B!Ju)_xAobb&nqIomwQEs#fE>&@48Z}#&UGS_8%1U34- z`v(&VlSU!VDxwfE?Z(nnLo;*`~I;U%gJnXwf?;pq+xJP9{ z*2QPPD{N!!bAvqeurAdP`1PkC_q^ZJmfU&GUXg)pAm04sW^`xy!jj+qX9E`BYiqzB z$SXGJVp>oae>?9(Y~Angn{et6`22l7lY2#ucUtfzWWtxx16dcj!Ec>C%-@1NSgrfJ za)n+&>xs|0JyL&rZ=nz9haOFm;~w3UzuyBp0sHzrJp%os9Qfs~N&%m9CUoO|CrnEjMX0Z9k3(r-|tF$xmM(CK#tze+CA|@-&#L-7Z_`G#4~W?)(q_KkDW<-BQ?Ft0CGuUeYo~!u*)SboEc!}whLJVaZb_1PSxy@DKh8)ZbW+A8DC402q z=Pd!}e2YBi@ttnIai0fg9l!>?E&S@vBIm4*&ze0*=oJLIz8w9t>icVpU5qdO-Ma6~ zn%Nrc3iQjJ&NzFJu%jYb8C@PZJktqTzz*0;`|#{}E(NT>j9)(Q_d=aq{qkC4Y>S^Z^)O0c+=2?vl9gI_EK0K4Z=**Jyt2Tyg|>yc%k*yd00o^%L?X^ntl?B+} zoDt}s@!dM<=WG9p&t7r{G4J3F_Jpl&%?`collV-m0GGTjIn0&MyEPvAIQNRKO!{l@ z;Y?}C38XU(NPF0~MCbB-a)qxw@A!SjzXt5J#Xpmm7Uj$7T*SGO|1L87seM3 z&(EIDmGE0@-0w>{^&RpHHLG>cR~lal^qJTW>C*2U`d=bva0|Xy*g5)n2htgY9CA;{ zE%5F}bJ>GFc|B9}oaZZYAwADiq_xgngSdHXW&jg#&w*dt+_tc-k{+Rm11{YjW{fKY}@$yUo(xJ++^q_uyVCwo(rBl`CvH5(xF^pR0tN ziR+hN96rQC(#%E0aEPU4RS*I_M6ByUMF7R3B znJ0nY8sk4pKbQKYvD$NY=+;zVOn*g|0O)cI`hmUUpc@mc>5N$I?;hodEsVR@8QIR0 z+70kN{cIYa&?``Pj)HVW?Y}4VWpl98BIoF{XnT&CSm4`}JENZ>|CWCC?m5V%Udi#y z^jYIfllpp@kC5|j@!8jNbk;49qt(H4iS2Wf10T!S273VYtn=AV9%~YD`Sh7Hs7Iei zu1Y*FIp!vO!Tj%nAh#{#h<|0Q@-8~zvw&|#4}1mP{m65Hlw(4U>YzmbOtoi^m|PqD z&SkBA4}8w)-mOWXto!q$ZSMl)3G6q-W@PhGSB)X}a_R?dk=Ir%GZVM(1gtUFy|sGm zz&4ga3KW~Y-=Yg2>UZH1oC77wzFK7EbL=XMW;5^2Cj(q>Rt@Doe;UH%rf&RB^LW(f41c_5T0%kzPLp|6sQ z9D%d775G8c2J)44&&}wr0l)Ple(O%bbl&ImT&vo3S(J-xSo1pJ(P*7DfLv)JOV)cGviW5Vw~U2`RmxW7+*e&(PCpL_+m zCr_Vxu?9VW(%&c>eBPZsoM{Ix*Th-1_8F90{NBBA24^hM;$o#^;#?@l^ZFTWp{2SHO9^gCLd``xf<{i}QIdCHj2@dM3~LJY#2)HxqYW_av`< zT0i^+{cHH$gMNF|=sYue;PW|1#7AI`TorkWT|n#0UQ+bA-1Cl{8JWo&(F?F{1>`!+ z=NY(+*~`5fpTGdDS2|NdZ(K-wxfkoqv1ic_tTivRWAUH&ne@4@8Ci(6dS)otd;HeA z-vO*nQmgkKUgOW?STCHh-8p)VexECRuDLF;=x4T~J_BcJ`^7(l*VxwjoL-_^C-k|t z`0b^%mpxk?pECxw`%c&e9JrMwmcNn6^ZYK5?qdS-X94FGcBa;RXOwG3KSj5%`S#I2 zBE<@rKb*jeZhkuN(q5qr=GyZTK5b{+gDY*8xt>7)=kWK0GqiowmeQWHw@&ae>tGDO zdv7_UIOhxCobOR@Ph&nC@|IYhdGy_y81{(Vs)Gqse5(n?CvJ!Jyw*|M(dyYLK zI+uB=_Vu|4mCJ|Z*r1#$xt zF!l_XZ_GS-SJ;7Ffwuo<;C}}x9 zX`Ni60^fbs2YP#M$#buHi5FtpQG4GFwlh0}oEsodC4TFi82$+q{OU8Y@5S%F%(K?H z%x`-)7r*oB^Bk-V{N9tlzXs$C?D-Pu3^O%5bh)gZ$lIeIcxSkr$G-!t)h}0|%Qv9g zE5WltwmqV^=N)Oiy!N!mh#vSwT5NL)c0tOiE&q&fbxu!oIg;^p|2+7-R^l1h|4?Jy zTf}xqpL6FB&OB>#ZWqX~7QcMAVvk7Tjw9hGfO2p;b5_4s2pGi6O z2Y#itXHjeSd4KlV5zoLp-y!l;@*fbF%h)&2-ABTIfvw-#EpkQ%{?I2_ehzAoSNs0l z2sKk2=+_v`MTiO=W-S&*0KohPvUK2)yA7ZdQz{XF{abfyHH^8owib&t6x z-*G-S8}eKw`dQ0up8LE4<^(XFkk-0SZ82fH?Bm*=bvft5fJ_N3S5C7$t#J$60^ zoVPvYOdx|DwzYwrk!@ev#46JL?ui|GwOr%)vbDhTX+6>9IDBVc;rH|R3Rq{}1ngmL z!Y7>Hyg7OwKkJvk+58L@{N@d3NLH==84qmx?m^&_`vO?6Z9nI0EBZaimBAIU5y-Vg zz6R&#O3-l{&scY=fMU1sV8_T>$46@+p%b$9-;zduDRB{YRgx<=`&Z?(a1@139(!eOjmA zcik)W4e;LOxj5^4wU?YT9P%jT9m&yOu_yGRj~dS_1MMAhJn#Md3~YhVuRJSc5HnBP z{)gF(Ifp%_v()B!g&cwBGk~DZzFXkly_Xem_T6G1@@Du0zp@5q~g`R|_hyiCmHS-kJ7 z{Ve2KE%AcX_FOmAzC_A93UqnxWncH^Y|c^eDG&Qv5v%0P;N_Cz4Duv$jms;~fRv}8 zZ}7KOyWhCym$BRVqWhh`CEuD05aihR&{OG*Z6;#w;T3od;x<880R z68XZ<+>Ctz+@s4p>k_^!(CyjQn*S8FzoXPAkU^_~Pg$e$?e(L-gWtS_zB=oB3}oZy zyv=hqW7clay$5q#&M-&+92l$kJ;PJ<5ez_G{^<;fE{{ET*v91;(JS)n)EbxHvn<4m zHfXuri~C56PmW5C`31P_cLBUZ^9KCZJ_9TC74q;by}`Cl`|We`1vQzR8H^{(-YahX zCLcF0n@FC(3CCQptT`}@g}KY;?QTahEjfrse(zXwY6v-i9+dacMW z_PsjqjBQSZBS1peHrKenxmQT<#QU?Cea&&UFc;hXe}mkN&n1_+_WN+H@fRTQduP7K zJpY{wUiA+^=`4F<${n`9KX0ET9DN@Xu|w?&J>yU0ZPD%B zII%1Kx2dc6H`v~bxt?3X*XCbhl{&c!dCuk>=49|1I9tZwY6;EJ&(0catqFWSOFMk7 zx$+<2%lJxA%l>1Dxvu~dwmlNMKKZ@l*XT3yr>K2}++cgY1-^u|M&0~~5BBKX-bXc$ zS~*5wA7`;vzt4cndcRN1cTQ&<$Z@}?$cmhUL}J!v>>$@XeV6!L?VW>eo&FK&-0z{w zJ%bYcS@XgrP+T#xSHG1EokN6({7HKRK^F5G| z^3MX@c`wNGjC=-GNNe5O9%)bIb7Db$n-RGsZte!i;h7q@_e*^0ejeQKCAR#Tn6Pfb zAL~A&y_{`}-@UX&UVt6Ggp|j8RWv7-&vG}nE5kuCZ5E)?_ok} zx3fdtV#_sB*Lt8EZ*|0c-nFaoFZR1i=yJOXInI7|Psq>zzJz^)~6n#Pr$IcVOEiwtT`p z8e5$q@gJTStah$NA8316bk8A)+I!6C=2$x;1AoRRzx5mB6)?Z$A@!TD?F_GspMi=G zxQ79|qTA2&O3`iiVcv}GZ{t7~Z28PD2Kr3Z!_w}}O@ z6>$Ea!FEqRXZEpfr!Ky@^hb~A`Sx*s=SYjR&iX=*e)o~pf%l&9*~{9B&zu9dckU{+ z*Z3Pn-V|M0^8&k2fA|bJd&a-PZ*Tb``gvHl$L}*`t(fs;Y|qqMXLZ%upO3>iDn~F8 zS8r!Rj`ch2cc~wc);UX760;_d)*07svk|vnLVkr>>rS!V;~R23ryU6VZQkhmYhojQ zbIo!8a=b*(NTFTzgHW>O;6G~rEZEoI?}u}9o|nLVIp15v(i$%IVNA{m+n8s%HHKcI z&s4GXH9v8459gzA1kRnwUm<7s5_(Y*+bur%26X%E@Y_G|$!CqRK-cFS=2d;fE|D|- z4RK}bVH~}HFkZX#%sk^2h$U{U%?z*HJ7WLkd4C_5Lk!=m^#`A$bYMlNs)`R*+()XM`I0kX43tatJqugB+#vsXNPUQH51a@tY67l z&FJkO(ue*&AlGI6VQ(Aqo!{po(6!y`76f_rDU~<~bbB?Mm^}vaT;{cVivH}m@2EZh zkSBj4uFOb3wdn@pf3x~A>p@oAmwq+{aL@0qCbeU3!(Gf;mfO+WB0w|FPLMeJkibLnyV;UE27U;o`d{v+S>wIBZG4}Z_o z2lwwkdUWsp{U?v^K6vnbC->8T^y!yPoZ<8T$9w-KjgvS&!*`@FzCXi{|HzO2XTSSD z{k`9J@|*r;`W_~||Lu1FrJwnafAC+SvzUJ1Gm7`m;Fo{yyZ`Nf_#09ZD+3={A3l-a zxcyJQ@$u@v|G)p0Kks3R&wZrw-fv9r@8!eiIX-&&;Nhe3U!3U0ClEb)_~_$A-*NS~ zeto)N6!6iB-ua%JC%QY{dvN#OlaCYqp5OC7{OQA$eUxbT%g2ZwJbC!^-p7fKpV93% z$A}(|4<9`E7bkjgoapYo2cKeV@BH(l6FqwJ=-Na-^)oMi=;kwe{s}}+ubp%`N6c~BLY5p*w2p>J-++!+LL{JqSb%(nHxX-cgIJMA3nJ@(ecB6ew^sZ z(@*oEz5QPwnTQX_y$9EB?f8fG`EjC0_r_}v>;{jOsZ-M#nt z`b0nahyLhyANi&|O2l)0{X=_vqSgO(Y@(<4KHc})F``Fz#~VJh$0vH{|2{I&qq{dv zbe!n$mn#|J|SYp_^YLyO(d>{6Idwd(&6; zpZV&4e{%Cgt3UK@$A}(3ynpSTef)>k?%(^BM~Lp-z5A(t)xLL}=!O^DKmI4b^gE6` z*~f`)`or(!_x?LSdNa}LZ{Ix8y?gwXb^Tl9_Ag)kg`0_XKl1jCXLRpAf8AgIru`p( z?!WkZH^10c|Lhpiqw&f0KeK-DTYmnp-AuImHUHYp6Fs>9@P-%LU;C@S|9?6196kPy zn~5HdPaj^t{`>#_@B8gH6Yc)_>Sm%x_a5@C{qgU$pZnjw?JuWy`4Is=_TB#rzv^b9 z#}Dq_z5Zms```Ur-*q$5?)9&}ndr&gCpY|Y{5QVyfBFpne3Hb+Ci>3rJVL~$|N0*_ zPk-It{W~`k?Y?}3=;_^ikFI~y{-yo5|Jj=#$kp>>ME9TEz5YOc!l(bvueo_^Pwzju z;emYm;^M0}U;pi2f0XE^pEW=Bn}7c4kth5A{H`Mt@h8OnYd>p_|GB<;>(|~$bbq}2 z_*4D4{`@%6lkx7Sc-7wiwqq0B8*lh4`}o$LA1AtZ!x~vCKG7$f(c8x+y3Y^)Yk$*z z!bHc3?%wc*{4L-2V_&-YHFEo3Ju=bVhmY^y{S;f!pKX>HTe#=p! zr#F1l)_Xtx=H_?y^P@zM#v7ia;~$aLANV(J{`B)NY!9yg;rHQx;{WK&|M@2n-SB7D z@#|mxiO=0UQUC7;KK?*{_@CUr<0Jnt{`^Bk%Rll@Kk(VlzPlv-x^Lkh;+OSXKPI1l lS9koXpYEHa?pNsfQ-8MGd>`}g```LmeZQ6eV}8&5{{gew%nJYj diff --git a/trained/ai87-kws20_v3-qat8.log b/trained/ai87-kws20_v3-qat8.log index 39afdae8..0201e87f 100644 --- a/trained/ai87-kws20_v3-qat8.log +++ b/trained/ai87-kws20_v3-qat8.log @@ -1,33452 +1,33653 @@ -2023-02-13 17:19:54,547 - Log file for this run: /data/ml/afshin/ai/kws20-enhancement/ai8x-training/logs/2023.02.13-171954/2023.02.13-171954.log -2023-02-13 17:19:56,333 - Optimizer Type: -2023-02-13 17:19:56,333 - Optimizer Args: {'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0.0, 'amsgrad': False} -2023-02-13 17:20:00,323 - Dataset sizes: - training=308808 - validation=34311 - test=36234 -2023-02-13 17:20:00,324 - Reading compression schedule from: policies/schedule_kws20.yaml -2023-02-13 17:20:00,330 - - -2023-02-13 17:20:00,330 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:20:01,422 - Epoch: [0][ 10/ 1207] Overall Loss 3.044520 Objective Loss 3.044520 LR 0.001000 Time 0.109147 -2023-02-13 17:20:01,572 - Epoch: [0][ 20/ 1207] Overall Loss 3.043369 Objective Loss 3.043369 LR 0.001000 Time 0.062053 -2023-02-13 17:20:01,704 - Epoch: [0][ 30/ 1207] Overall Loss 3.042306 Objective Loss 3.042306 LR 0.001000 Time 0.045711 -2023-02-13 17:20:01,836 - Epoch: [0][ 40/ 1207] Overall Loss 3.040471 Objective Loss 3.040471 LR 0.001000 Time 0.037565 -2023-02-13 17:20:01,968 - Epoch: [0][ 50/ 1207] Overall Loss 3.037665 Objective Loss 3.037665 LR 0.001000 Time 0.032669 -2023-02-13 17:20:02,103 - Epoch: [0][ 60/ 1207] Overall Loss 3.033926 Objective Loss 3.033926 LR 0.001000 Time 0.029462 -2023-02-13 17:20:02,235 - Epoch: [0][ 70/ 1207] Overall Loss 3.028481 Objective Loss 3.028481 LR 0.001000 Time 0.027136 -2023-02-13 17:20:02,366 - Epoch: [0][ 80/ 1207] Overall Loss 3.019905 Objective Loss 3.019905 LR 0.001000 Time 0.025379 -2023-02-13 17:20:02,500 - Epoch: [0][ 90/ 1207] Overall Loss 3.008199 Objective Loss 3.008199 LR 0.001000 Time 0.024037 -2023-02-13 17:20:02,631 - Epoch: [0][ 100/ 1207] Overall Loss 2.996023 Objective Loss 2.996023 LR 0.001000 Time 0.022938 -2023-02-13 17:20:02,765 - Epoch: [0][ 110/ 1207] Overall Loss 2.983982 Objective Loss 2.983982 LR 0.001000 Time 0.022065 -2023-02-13 17:20:02,899 - Epoch: [0][ 120/ 1207] Overall Loss 2.970293 Objective Loss 2.970293 LR 0.001000 Time 0.021332 -2023-02-13 17:20:03,033 - Epoch: [0][ 130/ 1207] Overall Loss 2.957644 Objective Loss 2.957644 LR 0.001000 Time 0.020722 -2023-02-13 17:20:03,166 - Epoch: [0][ 140/ 1207] Overall Loss 2.946931 Objective Loss 2.946931 LR 0.001000 Time 0.020189 -2023-02-13 17:20:03,298 - Epoch: [0][ 150/ 1207] Overall Loss 2.935690 Objective Loss 2.935690 LR 0.001000 Time 0.019720 -2023-02-13 17:20:03,430 - Epoch: [0][ 160/ 1207] Overall Loss 2.924452 Objective Loss 2.924452 LR 0.001000 Time 0.019305 -2023-02-13 17:20:03,562 - Epoch: [0][ 170/ 1207] Overall Loss 2.918008 Objective Loss 2.918008 LR 0.001000 Time 0.018945 -2023-02-13 17:20:03,695 - Epoch: [0][ 180/ 1207] Overall Loss 2.910881 Objective Loss 2.910881 LR 0.001000 Time 0.018623 -2023-02-13 17:20:03,826 - Epoch: [0][ 190/ 1207] Overall Loss 2.903198 Objective Loss 2.903198 LR 0.001000 Time 0.018333 -2023-02-13 17:20:03,957 - Epoch: [0][ 200/ 1207] Overall Loss 2.894299 Objective Loss 2.894299 LR 0.001000 Time 0.018069 -2023-02-13 17:20:04,090 - Epoch: [0][ 210/ 1207] Overall Loss 2.887220 Objective Loss 2.887220 LR 0.001000 Time 0.017839 -2023-02-13 17:20:04,222 - Epoch: [0][ 220/ 1207] Overall Loss 2.878602 Objective Loss 2.878602 LR 0.001000 Time 0.017625 -2023-02-13 17:20:04,354 - Epoch: [0][ 230/ 1207] Overall Loss 2.871992 Objective Loss 2.871992 LR 0.001000 Time 0.017433 -2023-02-13 17:20:04,485 - Epoch: [0][ 240/ 1207] Overall Loss 2.864465 Objective Loss 2.864465 LR 0.001000 Time 0.017250 -2023-02-13 17:20:04,617 - Epoch: [0][ 250/ 1207] Overall Loss 2.856470 Objective Loss 2.856470 LR 0.001000 Time 0.017084 -2023-02-13 17:20:04,748 - Epoch: [0][ 260/ 1207] Overall Loss 2.848727 Objective Loss 2.848727 LR 0.001000 Time 0.016931 -2023-02-13 17:20:04,880 - Epoch: [0][ 270/ 1207] Overall Loss 2.839212 Objective Loss 2.839212 LR 0.001000 Time 0.016791 -2023-02-13 17:20:05,012 - Epoch: [0][ 280/ 1207] Overall Loss 2.830332 Objective Loss 2.830332 LR 0.001000 Time 0.016661 -2023-02-13 17:20:05,144 - Epoch: [0][ 290/ 1207] Overall Loss 2.819914 Objective Loss 2.819914 LR 0.001000 Time 0.016540 -2023-02-13 17:20:05,276 - Epoch: [0][ 300/ 1207] Overall Loss 2.809721 Objective Loss 2.809721 LR 0.001000 Time 0.016428 -2023-02-13 17:20:05,406 - Epoch: [0][ 310/ 1207] Overall Loss 2.800575 Objective Loss 2.800575 LR 0.001000 Time 0.016317 -2023-02-13 17:20:05,538 - Epoch: [0][ 320/ 1207] Overall Loss 2.790453 Objective Loss 2.790453 LR 0.001000 Time 0.016216 -2023-02-13 17:20:05,670 - Epoch: [0][ 330/ 1207] Overall Loss 2.780350 Objective Loss 2.780350 LR 0.001000 Time 0.016121 -2023-02-13 17:20:05,802 - Epoch: [0][ 340/ 1207] Overall Loss 2.770281 Objective Loss 2.770281 LR 0.001000 Time 0.016035 -2023-02-13 17:20:05,937 - Epoch: [0][ 350/ 1207] Overall Loss 2.759806 Objective Loss 2.759806 LR 0.001000 Time 0.015956 -2023-02-13 17:20:06,070 - Epoch: [0][ 360/ 1207] Overall Loss 2.749453 Objective Loss 2.749453 LR 0.001000 Time 0.015879 -2023-02-13 17:20:06,206 - Epoch: [0][ 370/ 1207] Overall Loss 2.739962 Objective Loss 2.739962 LR 0.001000 Time 0.015816 -2023-02-13 17:20:06,341 - Epoch: [0][ 380/ 1207] Overall Loss 2.730206 Objective Loss 2.730206 LR 0.001000 Time 0.015752 -2023-02-13 17:20:06,479 - Epoch: [0][ 390/ 1207] Overall Loss 2.719031 Objective Loss 2.719031 LR 0.001000 Time 0.015701 -2023-02-13 17:20:06,614 - Epoch: [0][ 400/ 1207] Overall Loss 2.708406 Objective Loss 2.708406 LR 0.001000 Time 0.015644 -2023-02-13 17:20:06,750 - Epoch: [0][ 410/ 1207] Overall Loss 2.696055 Objective Loss 2.696055 LR 0.001000 Time 0.015593 -2023-02-13 17:20:06,885 - Epoch: [0][ 420/ 1207] Overall Loss 2.683248 Objective Loss 2.683248 LR 0.001000 Time 0.015541 -2023-02-13 17:20:07,023 - Epoch: [0][ 430/ 1207] Overall Loss 2.672050 Objective Loss 2.672050 LR 0.001000 Time 0.015498 -2023-02-13 17:20:07,157 - Epoch: [0][ 440/ 1207] Overall Loss 2.660074 Objective Loss 2.660074 LR 0.001000 Time 0.015450 -2023-02-13 17:20:07,294 - Epoch: [0][ 450/ 1207] Overall Loss 2.647764 Objective Loss 2.647764 LR 0.001000 Time 0.015410 -2023-02-13 17:20:07,427 - Epoch: [0][ 460/ 1207] Overall Loss 2.634924 Objective Loss 2.634924 LR 0.001000 Time 0.015361 -2023-02-13 17:20:07,565 - Epoch: [0][ 470/ 1207] Overall Loss 2.623203 Objective Loss 2.623203 LR 0.001000 Time 0.015326 -2023-02-13 17:20:07,700 - Epoch: [0][ 480/ 1207] Overall Loss 2.610259 Objective Loss 2.610259 LR 0.001000 Time 0.015286 -2023-02-13 17:20:07,837 - Epoch: [0][ 490/ 1207] Overall Loss 2.597768 Objective Loss 2.597768 LR 0.001000 Time 0.015252 -2023-02-13 17:20:07,973 - Epoch: [0][ 500/ 1207] Overall Loss 2.585146 Objective Loss 2.585146 LR 0.001000 Time 0.015215 -2023-02-13 17:20:08,108 - Epoch: [0][ 510/ 1207] Overall Loss 2.571819 Objective Loss 2.571819 LR 0.001000 Time 0.015182 -2023-02-13 17:20:08,244 - Epoch: [0][ 520/ 1207] Overall Loss 2.559869 Objective Loss 2.559869 LR 0.001000 Time 0.015147 -2023-02-13 17:20:08,381 - Epoch: [0][ 530/ 1207] Overall Loss 2.547360 Objective Loss 2.547360 LR 0.001000 Time 0.015119 -2023-02-13 17:20:08,515 - Epoch: [0][ 540/ 1207] Overall Loss 2.535129 Objective Loss 2.535129 LR 0.001000 Time 0.015086 -2023-02-13 17:20:08,652 - Epoch: [0][ 550/ 1207] Overall Loss 2.522200 Objective Loss 2.522200 LR 0.001000 Time 0.015059 -2023-02-13 17:20:08,789 - Epoch: [0][ 560/ 1207] Overall Loss 2.510737 Objective Loss 2.510737 LR 0.001000 Time 0.015031 -2023-02-13 17:20:08,926 - Epoch: [0][ 570/ 1207] Overall Loss 2.498713 Objective Loss 2.498713 LR 0.001000 Time 0.015007 -2023-02-13 17:20:09,062 - Epoch: [0][ 580/ 1207] Overall Loss 2.486259 Objective Loss 2.486259 LR 0.001000 Time 0.014981 -2023-02-13 17:20:09,199 - Epoch: [0][ 590/ 1207] Overall Loss 2.474184 Objective Loss 2.474184 LR 0.001000 Time 0.014958 -2023-02-13 17:20:09,335 - Epoch: [0][ 600/ 1207] Overall Loss 2.461908 Objective Loss 2.461908 LR 0.001000 Time 0.014933 -2023-02-13 17:20:09,468 - Epoch: [0][ 610/ 1207] Overall Loss 2.449907 Objective Loss 2.449907 LR 0.001000 Time 0.014905 -2023-02-13 17:20:09,604 - Epoch: [0][ 620/ 1207] Overall Loss 2.438741 Objective Loss 2.438741 LR 0.001000 Time 0.014884 -2023-02-13 17:20:09,739 - Epoch: [0][ 630/ 1207] Overall Loss 2.428077 Objective Loss 2.428077 LR 0.001000 Time 0.014862 -2023-02-13 17:20:09,874 - Epoch: [0][ 640/ 1207] Overall Loss 2.416668 Objective Loss 2.416668 LR 0.001000 Time 0.014838 -2023-02-13 17:20:10,010 - Epoch: [0][ 650/ 1207] Overall Loss 2.405313 Objective Loss 2.405313 LR 0.001000 Time 0.014818 -2023-02-13 17:20:10,146 - Epoch: [0][ 660/ 1207] Overall Loss 2.394817 Objective Loss 2.394817 LR 0.001000 Time 0.014797 -2023-02-13 17:20:10,284 - Epoch: [0][ 670/ 1207] Overall Loss 2.384120 Objective Loss 2.384120 LR 0.001000 Time 0.014782 -2023-02-13 17:20:10,420 - Epoch: [0][ 680/ 1207] Overall Loss 2.373380 Objective Loss 2.373380 LR 0.001000 Time 0.014762 -2023-02-13 17:20:10,557 - Epoch: [0][ 690/ 1207] Overall Loss 2.362931 Objective Loss 2.362931 LR 0.001000 Time 0.014746 -2023-02-13 17:20:10,693 - Epoch: [0][ 700/ 1207] Overall Loss 2.352576 Objective Loss 2.352576 LR 0.001000 Time 0.014727 -2023-02-13 17:20:10,830 - Epoch: [0][ 710/ 1207] Overall Loss 2.342617 Objective Loss 2.342617 LR 0.001000 Time 0.014713 -2023-02-13 17:20:10,966 - Epoch: [0][ 720/ 1207] Overall Loss 2.332349 Objective Loss 2.332349 LR 0.001000 Time 0.014694 -2023-02-13 17:20:11,104 - Epoch: [0][ 730/ 1207] Overall Loss 2.322504 Objective Loss 2.322504 LR 0.001000 Time 0.014682 -2023-02-13 17:20:11,240 - Epoch: [0][ 740/ 1207] Overall Loss 2.312421 Objective Loss 2.312421 LR 0.001000 Time 0.014665 -2023-02-13 17:20:11,377 - Epoch: [0][ 750/ 1207] Overall Loss 2.303121 Objective Loss 2.303121 LR 0.001000 Time 0.014651 -2023-02-13 17:20:11,511 - Epoch: [0][ 760/ 1207] Overall Loss 2.293603 Objective Loss 2.293603 LR 0.001000 Time 0.014634 -2023-02-13 17:20:11,647 - Epoch: [0][ 770/ 1207] Overall Loss 2.284004 Objective Loss 2.284004 LR 0.001000 Time 0.014621 -2023-02-13 17:20:11,783 - Epoch: [0][ 780/ 1207] Overall Loss 2.274823 Objective Loss 2.274823 LR 0.001000 Time 0.014606 -2023-02-13 17:20:11,921 - Epoch: [0][ 790/ 1207] Overall Loss 2.265775 Objective Loss 2.265775 LR 0.001000 Time 0.014597 -2023-02-13 17:20:12,056 - Epoch: [0][ 800/ 1207] Overall Loss 2.256423 Objective Loss 2.256423 LR 0.001000 Time 0.014582 -2023-02-13 17:20:12,188 - Epoch: [0][ 810/ 1207] Overall Loss 2.247844 Objective Loss 2.247844 LR 0.001000 Time 0.014564 -2023-02-13 17:20:12,319 - Epoch: [0][ 820/ 1207] Overall Loss 2.239060 Objective Loss 2.239060 LR 0.001000 Time 0.014546 -2023-02-13 17:20:12,452 - Epoch: [0][ 830/ 1207] Overall Loss 2.230484 Objective Loss 2.230484 LR 0.001000 Time 0.014531 -2023-02-13 17:20:12,583 - Epoch: [0][ 840/ 1207] Overall Loss 2.222230 Objective Loss 2.222230 LR 0.001000 Time 0.014513 -2023-02-13 17:20:12,715 - Epoch: [0][ 850/ 1207] Overall Loss 2.214388 Objective Loss 2.214388 LR 0.001000 Time 0.014498 -2023-02-13 17:20:12,848 - Epoch: [0][ 860/ 1207] Overall Loss 2.206306 Objective Loss 2.206306 LR 0.001000 Time 0.014483 -2023-02-13 17:20:12,980 - Epoch: [0][ 870/ 1207] Overall Loss 2.198585 Objective Loss 2.198585 LR 0.001000 Time 0.014468 -2023-02-13 17:20:13,113 - Epoch: [0][ 880/ 1207] Overall Loss 2.191301 Objective Loss 2.191301 LR 0.001000 Time 0.014454 -2023-02-13 17:20:13,244 - Epoch: [0][ 890/ 1207] Overall Loss 2.183410 Objective Loss 2.183410 LR 0.001000 Time 0.014439 -2023-02-13 17:20:13,375 - Epoch: [0][ 900/ 1207] Overall Loss 2.175272 Objective Loss 2.175272 LR 0.001000 Time 0.014423 -2023-02-13 17:20:13,507 - Epoch: [0][ 910/ 1207] Overall Loss 2.167709 Objective Loss 2.167709 LR 0.001000 Time 0.014410 -2023-02-13 17:20:13,639 - Epoch: [0][ 920/ 1207] Overall Loss 2.160365 Objective Loss 2.160365 LR 0.001000 Time 0.014395 -2023-02-13 17:20:13,771 - Epoch: [0][ 930/ 1207] Overall Loss 2.153270 Objective Loss 2.153270 LR 0.001000 Time 0.014383 -2023-02-13 17:20:13,902 - Epoch: [0][ 940/ 1207] Overall Loss 2.146161 Objective Loss 2.146161 LR 0.001000 Time 0.014369 -2023-02-13 17:20:14,035 - Epoch: [0][ 950/ 1207] Overall Loss 2.138543 Objective Loss 2.138543 LR 0.001000 Time 0.014357 -2023-02-13 17:20:14,166 - Epoch: [0][ 960/ 1207] Overall Loss 2.131426 Objective Loss 2.131426 LR 0.001000 Time 0.014343 -2023-02-13 17:20:14,298 - Epoch: [0][ 970/ 1207] Overall Loss 2.124332 Objective Loss 2.124332 LR 0.001000 Time 0.014331 -2023-02-13 17:20:14,431 - Epoch: [0][ 980/ 1207] Overall Loss 2.117875 Objective Loss 2.117875 LR 0.001000 Time 0.014320 -2023-02-13 17:20:14,563 - Epoch: [0][ 990/ 1207] Overall Loss 2.110386 Objective Loss 2.110386 LR 0.001000 Time 0.014308 -2023-02-13 17:20:14,694 - Epoch: [0][ 1000/ 1207] Overall Loss 2.103728 Objective Loss 2.103728 LR 0.001000 Time 0.014296 -2023-02-13 17:20:14,827 - Epoch: [0][ 1010/ 1207] Overall Loss 2.097149 Objective Loss 2.097149 LR 0.001000 Time 0.014286 -2023-02-13 17:20:14,961 - Epoch: [0][ 1020/ 1207] Overall Loss 2.090171 Objective Loss 2.090171 LR 0.001000 Time 0.014276 -2023-02-13 17:20:15,093 - Epoch: [0][ 1030/ 1207] Overall Loss 2.083695 Objective Loss 2.083695 LR 0.001000 Time 0.014266 -2023-02-13 17:20:15,225 - Epoch: [0][ 1040/ 1207] Overall Loss 2.077071 Objective Loss 2.077071 LR 0.001000 Time 0.014255 -2023-02-13 17:20:15,358 - Epoch: [0][ 1050/ 1207] Overall Loss 2.070575 Objective Loss 2.070575 LR 0.001000 Time 0.014246 -2023-02-13 17:20:15,489 - Epoch: [0][ 1060/ 1207] Overall Loss 2.064187 Objective Loss 2.064187 LR 0.001000 Time 0.014234 -2023-02-13 17:20:15,620 - Epoch: [0][ 1070/ 1207] Overall Loss 2.057697 Objective Loss 2.057697 LR 0.001000 Time 0.014224 -2023-02-13 17:20:15,752 - Epoch: [0][ 1080/ 1207] Overall Loss 2.051377 Objective Loss 2.051377 LR 0.001000 Time 0.014214 -2023-02-13 17:20:15,886 - Epoch: [0][ 1090/ 1207] Overall Loss 2.044542 Objective Loss 2.044542 LR 0.001000 Time 0.014206 -2023-02-13 17:20:16,018 - Epoch: [0][ 1100/ 1207] Overall Loss 2.037964 Objective Loss 2.037964 LR 0.001000 Time 0.014196 -2023-02-13 17:20:16,149 - Epoch: [0][ 1110/ 1207] Overall Loss 2.031666 Objective Loss 2.031666 LR 0.001000 Time 0.014186 -2023-02-13 17:20:16,280 - Epoch: [0][ 1120/ 1207] Overall Loss 2.025887 Objective Loss 2.025887 LR 0.001000 Time 0.014176 -2023-02-13 17:20:16,412 - Epoch: [0][ 1130/ 1207] Overall Loss 2.019809 Objective Loss 2.019809 LR 0.001000 Time 0.014167 -2023-02-13 17:20:16,543 - Epoch: [0][ 1140/ 1207] Overall Loss 2.014099 Objective Loss 2.014099 LR 0.001000 Time 0.014158 -2023-02-13 17:20:16,677 - Epoch: [0][ 1150/ 1207] Overall Loss 2.008132 Objective Loss 2.008132 LR 0.001000 Time 0.014151 -2023-02-13 17:20:16,810 - Epoch: [0][ 1160/ 1207] Overall Loss 2.002599 Objective Loss 2.002599 LR 0.001000 Time 0.014143 -2023-02-13 17:20:16,942 - Epoch: [0][ 1170/ 1207] Overall Loss 1.996910 Objective Loss 1.996910 LR 0.001000 Time 0.014134 -2023-02-13 17:20:17,074 - Epoch: [0][ 1180/ 1207] Overall Loss 1.991193 Objective Loss 1.991193 LR 0.001000 Time 0.014126 -2023-02-13 17:20:17,205 - Epoch: [0][ 1190/ 1207] Overall Loss 1.985212 Objective Loss 1.985212 LR 0.001000 Time 0.014117 -2023-02-13 17:20:17,383 - Epoch: [0][ 1200/ 1207] Overall Loss 1.979679 Objective Loss 1.979679 LR 0.001000 Time 0.014148 -2023-02-13 17:20:17,469 - Epoch: [0][ 1207/ 1207] Overall Loss 1.975597 Objective Loss 1.975597 Top1 51.219512 Top5 85.060976 LR 0.001000 Time 0.014137 -2023-02-13 17:20:17,533 - --- validate (epoch=0)----------- -2023-02-13 17:20:17,533 - 34311 samples (256 per mini-batch) -2023-02-13 17:20:17,862 - Epoch: [0][ 10/ 135] Loss 1.319929 Top1 51.289063 Top5 85.703125 -2023-02-13 17:20:17,953 - Epoch: [0][ 20/ 135] Loss 1.318839 Top1 51.132812 Top5 85.605469 -2023-02-13 17:20:18,034 - Epoch: [0][ 30/ 135] Loss 1.324167 Top1 51.640625 Top5 86.263021 -2023-02-13 17:20:18,126 - Epoch: [0][ 40/ 135] Loss 1.332396 Top1 51.289063 Top5 86.025391 -2023-02-13 17:20:18,219 - Epoch: [0][ 50/ 135] Loss 1.334588 Top1 51.320312 Top5 85.976562 -2023-02-13 17:20:18,313 - Epoch: [0][ 60/ 135] Loss 1.335165 Top1 51.399740 Top5 85.833333 -2023-02-13 17:20:18,408 - Epoch: [0][ 70/ 135] Loss 1.329984 Top1 51.462054 Top5 85.876116 -2023-02-13 17:20:18,502 - Epoch: [0][ 80/ 135] Loss 1.325210 Top1 51.372070 Top5 85.795898 -2023-02-13 17:20:18,596 - Epoch: [0][ 90/ 135] Loss 1.322262 Top1 51.393229 Top5 85.833333 -2023-02-13 17:20:18,678 - Epoch: [0][ 100/ 135] Loss 1.313590 Top1 51.539062 Top5 85.835938 -2023-02-13 17:20:18,761 - Epoch: [0][ 110/ 135] Loss 1.313228 Top1 51.509233 Top5 85.862926 -2023-02-13 17:20:18,847 - Epoch: [0][ 120/ 135] Loss 1.305607 Top1 51.569010 Top5 85.901693 -2023-02-13 17:20:18,933 - Epoch: [0][ 130/ 135] Loss 1.305810 Top1 51.577524 Top5 85.943510 -2023-02-13 17:20:18,957 - Epoch: [0][ 135/ 135] Loss 1.317483 Top1 51.619014 Top5 85.914138 -2023-02-13 17:20:19,033 - ==> Top1: 51.619 Top5: 85.914 Loss: 1.317 - -2023-02-13 17:20:19,034 - ==> Confusion: -[[ 687 11 6 2 17 2 0 1 16 162 5 2 0 12 4 4 5 8 8 2 13] - [ 5 646 25 15 36 20 2 19 46 8 73 2 2 15 45 0 21 3 40 1 9] - [ 30 18 664 10 36 19 93 89 1 13 21 3 2 9 2 6 4 3 10 12 13] - [ 18 49 65 484 12 37 15 48 17 12 122 1 2 6 39 7 4 16 30 3 29] - [ 38 17 15 0 886 5 0 0 1 25 7 1 0 4 10 13 20 2 6 0 16] - [ 11 158 71 45 21 318 7 88 25 13 57 22 20 103 28 9 7 10 18 21 18] - [ 7 12 183 8 6 10 742 70 0 0 5 6 0 2 0 12 0 6 6 13 11] - [ 1 30 105 30 2 30 15 571 0 2 48 6 9 2 0 0 0 2 141 20 10] - [ 57 36 2 2 7 1 0 0 721 57 14 2 8 13 65 0 1 3 17 0 3] - [ 370 9 15 3 17 2 6 0 41 471 0 0 0 20 30 6 3 2 4 1 12] - [ 9 121 59 26 10 18 2 59 43 4 560 1 2 4 4 1 0 1 116 2 9] - [ 2 16 5 1 6 20 4 8 2 0 5 573 151 34 3 21 37 64 3 31 19] - [ 0 8 1 12 3 23 1 5 28 0 8 258 389 30 8 8 21 129 3 13 11] - [ 43 51 28 0 28 55 1 12 30 57 7 17 17 550 36 7 31 11 5 18 20] - [ 45 63 2 16 32 2 2 4 199 51 7 1 3 7 620 1 10 5 9 0 13] - [ 20 8 15 2 53 6 13 1 0 0 0 41 2 16 1 785 29 25 5 1 23] - [ 23 36 3 3 130 7 0 1 22 1 2 10 6 10 19 18 733 5 2 2 28] - [ 6 7 2 13 3 6 4 3 35 6 1 70 104 22 5 51 6 686 3 1 17] - [ 4 86 13 20 7 7 0 111 20 6 110 1 11 2 10 0 3 2 660 6 7] - [ 0 12 34 4 7 15 35 93 1 0 4 65 11 17 0 12 13 1 13 781 30] - [ 582 917 442 291 533 308 168 402 288 439 286 310 272 470 511 330 565 235 372 529 5184]] - -2023-02-13 17:20:19,035 - ==> Best [Top1: 51.619 Top5: 85.914 Sparsity:0.00 Params: 148928 on epoch: 0] -2023-02-13 17:20:19,035 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:20:19,041 - - -2023-02-13 17:20:19,041 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:20:19,892 - Epoch: [1][ 10/ 1207] Overall Loss 1.311825 Objective Loss 1.311825 LR 0.001000 Time 0.084962 -2023-02-13 17:20:20,056 - Epoch: [1][ 20/ 1207] Overall Loss 1.322403 Objective Loss 1.322403 LR 0.001000 Time 0.050653 -2023-02-13 17:20:20,203 - Epoch: [1][ 30/ 1207] Overall Loss 1.311033 Objective Loss 1.311033 LR 0.001000 Time 0.038664 -2023-02-13 17:20:20,347 - Epoch: [1][ 40/ 1207] Overall Loss 1.311915 Objective Loss 1.311915 LR 0.001000 Time 0.032588 -2023-02-13 17:20:20,492 - Epoch: [1][ 50/ 1207] Overall Loss 1.313613 Objective Loss 1.313613 LR 0.001000 Time 0.028959 -2023-02-13 17:20:20,634 - Epoch: [1][ 60/ 1207] Overall Loss 1.307472 Objective Loss 1.307472 LR 0.001000 Time 0.026500 -2023-02-13 17:20:20,776 - Epoch: [1][ 70/ 1207] Overall Loss 1.304048 Objective Loss 1.304048 LR 0.001000 Time 0.024738 -2023-02-13 17:20:20,919 - Epoch: [1][ 80/ 1207] Overall Loss 1.295890 Objective Loss 1.295890 LR 0.001000 Time 0.023433 -2023-02-13 17:20:21,061 - Epoch: [1][ 90/ 1207] Overall Loss 1.290842 Objective Loss 1.290842 LR 0.001000 Time 0.022403 -2023-02-13 17:20:21,205 - Epoch: [1][ 100/ 1207] Overall Loss 1.291919 Objective Loss 1.291919 LR 0.001000 Time 0.021592 -2023-02-13 17:20:21,346 - Epoch: [1][ 110/ 1207] Overall Loss 1.293761 Objective Loss 1.293761 LR 0.001000 Time 0.020913 -2023-02-13 17:20:21,490 - Epoch: [1][ 120/ 1207] Overall Loss 1.293723 Objective Loss 1.293723 LR 0.001000 Time 0.020362 -2023-02-13 17:20:21,631 - Epoch: [1][ 130/ 1207] Overall Loss 1.293515 Objective Loss 1.293515 LR 0.001000 Time 0.019880 -2023-02-13 17:20:21,775 - Epoch: [1][ 140/ 1207] Overall Loss 1.289413 Objective Loss 1.289413 LR 0.001000 Time 0.019485 -2023-02-13 17:20:21,916 - Epoch: [1][ 150/ 1207] Overall Loss 1.290009 Objective Loss 1.290009 LR 0.001000 Time 0.019126 -2023-02-13 17:20:22,059 - Epoch: [1][ 160/ 1207] Overall Loss 1.288625 Objective Loss 1.288625 LR 0.001000 Time 0.018823 -2023-02-13 17:20:22,191 - Epoch: [1][ 170/ 1207] Overall Loss 1.285403 Objective Loss 1.285403 LR 0.001000 Time 0.018489 -2023-02-13 17:20:22,322 - Epoch: [1][ 180/ 1207] Overall Loss 1.282045 Objective Loss 1.282045 LR 0.001000 Time 0.018189 -2023-02-13 17:20:22,453 - Epoch: [1][ 190/ 1207] Overall Loss 1.280599 Objective Loss 1.280599 LR 0.001000 Time 0.017918 -2023-02-13 17:20:22,587 - Epoch: [1][ 200/ 1207] Overall Loss 1.277161 Objective Loss 1.277161 LR 0.001000 Time 0.017689 -2023-02-13 17:20:22,720 - Epoch: [1][ 210/ 1207] Overall Loss 1.278772 Objective Loss 1.278772 LR 0.001000 Time 0.017478 -2023-02-13 17:20:22,852 - Epoch: [1][ 220/ 1207] Overall Loss 1.275684 Objective Loss 1.275684 LR 0.001000 Time 0.017282 -2023-02-13 17:20:22,986 - Epoch: [1][ 230/ 1207] Overall Loss 1.273778 Objective Loss 1.273778 LR 0.001000 Time 0.017113 -2023-02-13 17:20:23,121 - Epoch: [1][ 240/ 1207] Overall Loss 1.275859 Objective Loss 1.275859 LR 0.001000 Time 0.016963 -2023-02-13 17:20:23,257 - Epoch: [1][ 250/ 1207] Overall Loss 1.273424 Objective Loss 1.273424 LR 0.001000 Time 0.016826 -2023-02-13 17:20:23,391 - Epoch: [1][ 260/ 1207] Overall Loss 1.272528 Objective Loss 1.272528 LR 0.001000 Time 0.016693 -2023-02-13 17:20:23,526 - Epoch: [1][ 270/ 1207] Overall Loss 1.270483 Objective Loss 1.270483 LR 0.001000 Time 0.016572 -2023-02-13 17:20:23,660 - Epoch: [1][ 280/ 1207] Overall Loss 1.269494 Objective Loss 1.269494 LR 0.001000 Time 0.016460 -2023-02-13 17:20:23,795 - Epoch: [1][ 290/ 1207] Overall Loss 1.267527 Objective Loss 1.267527 LR 0.001000 Time 0.016356 -2023-02-13 17:20:23,935 - Epoch: [1][ 300/ 1207] Overall Loss 1.265849 Objective Loss 1.265849 LR 0.001000 Time 0.016277 -2023-02-13 17:20:24,076 - Epoch: [1][ 310/ 1207] Overall Loss 1.264898 Objective Loss 1.264898 LR 0.001000 Time 0.016205 -2023-02-13 17:20:24,212 - Epoch: [1][ 320/ 1207] Overall Loss 1.264083 Objective Loss 1.264083 LR 0.001000 Time 0.016124 -2023-02-13 17:20:24,347 - Epoch: [1][ 330/ 1207] Overall Loss 1.263360 Objective Loss 1.263360 LR 0.001000 Time 0.016044 -2023-02-13 17:20:24,481 - Epoch: [1][ 340/ 1207] Overall Loss 1.261790 Objective Loss 1.261790 LR 0.001000 Time 0.015964 -2023-02-13 17:20:24,614 - Epoch: [1][ 350/ 1207] Overall Loss 1.259375 Objective Loss 1.259375 LR 0.001000 Time 0.015885 -2023-02-13 17:20:24,745 - Epoch: [1][ 360/ 1207] Overall Loss 1.255784 Objective Loss 1.255784 LR 0.001000 Time 0.015809 -2023-02-13 17:20:24,877 - Epoch: [1][ 370/ 1207] Overall Loss 1.254641 Objective Loss 1.254641 LR 0.001000 Time 0.015736 -2023-02-13 17:20:25,010 - Epoch: [1][ 380/ 1207] Overall Loss 1.252159 Objective Loss 1.252159 LR 0.001000 Time 0.015673 -2023-02-13 17:20:25,142 - Epoch: [1][ 390/ 1207] Overall Loss 1.250964 Objective Loss 1.250964 LR 0.001000 Time 0.015608 -2023-02-13 17:20:25,275 - Epoch: [1][ 400/ 1207] Overall Loss 1.249639 Objective Loss 1.249639 LR 0.001000 Time 0.015548 -2023-02-13 17:20:25,406 - Epoch: [1][ 410/ 1207] Overall Loss 1.248144 Objective Loss 1.248144 LR 0.001000 Time 0.015489 -2023-02-13 17:20:25,538 - Epoch: [1][ 420/ 1207] Overall Loss 1.246317 Objective Loss 1.246317 LR 0.001000 Time 0.015432 -2023-02-13 17:20:25,670 - Epoch: [1][ 430/ 1207] Overall Loss 1.244544 Objective Loss 1.244544 LR 0.001000 Time 0.015380 -2023-02-13 17:20:25,804 - Epoch: [1][ 440/ 1207] Overall Loss 1.243130 Objective Loss 1.243130 LR 0.001000 Time 0.015334 -2023-02-13 17:20:25,936 - Epoch: [1][ 450/ 1207] Overall Loss 1.241321 Objective Loss 1.241321 LR 0.001000 Time 0.015286 -2023-02-13 17:20:26,069 - Epoch: [1][ 460/ 1207] Overall Loss 1.239942 Objective Loss 1.239942 LR 0.001000 Time 0.015238 -2023-02-13 17:20:26,200 - Epoch: [1][ 470/ 1207] Overall Loss 1.238987 Objective Loss 1.238987 LR 0.001000 Time 0.015193 -2023-02-13 17:20:26,332 - Epoch: [1][ 480/ 1207] Overall Loss 1.237130 Objective Loss 1.237130 LR 0.001000 Time 0.015151 -2023-02-13 17:20:26,465 - Epoch: [1][ 490/ 1207] Overall Loss 1.235312 Objective Loss 1.235312 LR 0.001000 Time 0.015111 -2023-02-13 17:20:26,599 - Epoch: [1][ 500/ 1207] Overall Loss 1.233057 Objective Loss 1.233057 LR 0.001000 Time 0.015077 -2023-02-13 17:20:26,733 - Epoch: [1][ 510/ 1207] Overall Loss 1.231546 Objective Loss 1.231546 LR 0.001000 Time 0.015043 -2023-02-13 17:20:26,865 - Epoch: [1][ 520/ 1207] Overall Loss 1.231419 Objective Loss 1.231419 LR 0.001000 Time 0.015008 -2023-02-13 17:20:26,997 - Epoch: [1][ 530/ 1207] Overall Loss 1.229982 Objective Loss 1.229982 LR 0.001000 Time 0.014972 -2023-02-13 17:20:27,131 - Epoch: [1][ 540/ 1207] Overall Loss 1.228180 Objective Loss 1.228180 LR 0.001000 Time 0.014942 -2023-02-13 17:20:27,262 - Epoch: [1][ 550/ 1207] Overall Loss 1.226203 Objective Loss 1.226203 LR 0.001000 Time 0.014908 -2023-02-13 17:20:27,397 - Epoch: [1][ 560/ 1207] Overall Loss 1.224944 Objective Loss 1.224944 LR 0.001000 Time 0.014882 -2023-02-13 17:20:27,530 - Epoch: [1][ 570/ 1207] Overall Loss 1.224213 Objective Loss 1.224213 LR 0.001000 Time 0.014854 -2023-02-13 17:20:27,662 - Epoch: [1][ 580/ 1207] Overall Loss 1.222358 Objective Loss 1.222358 LR 0.001000 Time 0.014823 -2023-02-13 17:20:27,794 - Epoch: [1][ 590/ 1207] Overall Loss 1.220378 Objective Loss 1.220378 LR 0.001000 Time 0.014795 -2023-02-13 17:20:27,928 - Epoch: [1][ 600/ 1207] Overall Loss 1.218692 Objective Loss 1.218692 LR 0.001000 Time 0.014769 -2023-02-13 17:20:28,059 - Epoch: [1][ 610/ 1207] Overall Loss 1.216928 Objective Loss 1.216928 LR 0.001000 Time 0.014741 -2023-02-13 17:20:28,191 - Epoch: [1][ 620/ 1207] Overall Loss 1.215514 Objective Loss 1.215514 LR 0.001000 Time 0.014715 -2023-02-13 17:20:28,323 - Epoch: [1][ 630/ 1207] Overall Loss 1.214001 Objective Loss 1.214001 LR 0.001000 Time 0.014691 -2023-02-13 17:20:28,455 - Epoch: [1][ 640/ 1207] Overall Loss 1.212829 Objective Loss 1.212829 LR 0.001000 Time 0.014667 -2023-02-13 17:20:28,587 - Epoch: [1][ 650/ 1207] Overall Loss 1.211430 Objective Loss 1.211430 LR 0.001000 Time 0.014643 -2023-02-13 17:20:28,719 - Epoch: [1][ 660/ 1207] Overall Loss 1.209532 Objective Loss 1.209532 LR 0.001000 Time 0.014621 -2023-02-13 17:20:28,850 - Epoch: [1][ 670/ 1207] Overall Loss 1.207284 Objective Loss 1.207284 LR 0.001000 Time 0.014598 -2023-02-13 17:20:28,983 - Epoch: [1][ 680/ 1207] Overall Loss 1.205822 Objective Loss 1.205822 LR 0.001000 Time 0.014579 -2023-02-13 17:20:29,116 - Epoch: [1][ 690/ 1207] Overall Loss 1.204630 Objective Loss 1.204630 LR 0.001000 Time 0.014558 -2023-02-13 17:20:29,247 - Epoch: [1][ 700/ 1207] Overall Loss 1.203679 Objective Loss 1.203679 LR 0.001000 Time 0.014538 -2023-02-13 17:20:29,379 - Epoch: [1][ 710/ 1207] Overall Loss 1.202713 Objective Loss 1.202713 LR 0.001000 Time 0.014518 -2023-02-13 17:20:29,510 - Epoch: [1][ 720/ 1207] Overall Loss 1.201876 Objective Loss 1.201876 LR 0.001000 Time 0.014499 -2023-02-13 17:20:29,642 - Epoch: [1][ 730/ 1207] Overall Loss 1.200680 Objective Loss 1.200680 LR 0.001000 Time 0.014480 -2023-02-13 17:20:29,776 - Epoch: [1][ 740/ 1207] Overall Loss 1.198829 Objective Loss 1.198829 LR 0.001000 Time 0.014465 -2023-02-13 17:20:29,908 - Epoch: [1][ 750/ 1207] Overall Loss 1.197374 Objective Loss 1.197374 LR 0.001000 Time 0.014447 -2023-02-13 17:20:30,040 - Epoch: [1][ 760/ 1207] Overall Loss 1.196280 Objective Loss 1.196280 LR 0.001000 Time 0.014431 -2023-02-13 17:20:30,172 - Epoch: [1][ 770/ 1207] Overall Loss 1.194957 Objective Loss 1.194957 LR 0.001000 Time 0.014414 -2023-02-13 17:20:30,306 - Epoch: [1][ 780/ 1207] Overall Loss 1.194230 Objective Loss 1.194230 LR 0.001000 Time 0.014400 -2023-02-13 17:20:30,438 - Epoch: [1][ 790/ 1207] Overall Loss 1.193325 Objective Loss 1.193325 LR 0.001000 Time 0.014384 -2023-02-13 17:20:30,570 - Epoch: [1][ 800/ 1207] Overall Loss 1.192395 Objective Loss 1.192395 LR 0.001000 Time 0.014368 -2023-02-13 17:20:30,702 - Epoch: [1][ 810/ 1207] Overall Loss 1.191315 Objective Loss 1.191315 LR 0.001000 Time 0.014354 -2023-02-13 17:20:30,836 - Epoch: [1][ 820/ 1207] Overall Loss 1.190253 Objective Loss 1.190253 LR 0.001000 Time 0.014342 -2023-02-13 17:20:30,968 - Epoch: [1][ 830/ 1207] Overall Loss 1.188596 Objective Loss 1.188596 LR 0.001000 Time 0.014327 -2023-02-13 17:20:31,104 - Epoch: [1][ 840/ 1207] Overall Loss 1.187650 Objective Loss 1.187650 LR 0.001000 Time 0.014317 -2023-02-13 17:20:31,235 - Epoch: [1][ 850/ 1207] Overall Loss 1.186341 Objective Loss 1.186341 LR 0.001000 Time 0.014303 -2023-02-13 17:20:31,368 - Epoch: [1][ 860/ 1207] Overall Loss 1.185725 Objective Loss 1.185725 LR 0.001000 Time 0.014291 -2023-02-13 17:20:31,500 - Epoch: [1][ 870/ 1207] Overall Loss 1.184315 Objective Loss 1.184315 LR 0.001000 Time 0.014278 -2023-02-13 17:20:31,633 - Epoch: [1][ 880/ 1207] Overall Loss 1.182998 Objective Loss 1.182998 LR 0.001000 Time 0.014266 -2023-02-13 17:20:31,764 - Epoch: [1][ 890/ 1207] Overall Loss 1.181856 Objective Loss 1.181856 LR 0.001000 Time 0.014253 -2023-02-13 17:20:31,897 - Epoch: [1][ 900/ 1207] Overall Loss 1.180356 Objective Loss 1.180356 LR 0.001000 Time 0.014242 -2023-02-13 17:20:32,030 - Epoch: [1][ 910/ 1207] Overall Loss 1.179056 Objective Loss 1.179056 LR 0.001000 Time 0.014231 -2023-02-13 17:20:32,162 - Epoch: [1][ 920/ 1207] Overall Loss 1.178183 Objective Loss 1.178183 LR 0.001000 Time 0.014219 -2023-02-13 17:20:32,295 - Epoch: [1][ 930/ 1207] Overall Loss 1.176764 Objective Loss 1.176764 LR 0.001000 Time 0.014209 -2023-02-13 17:20:32,427 - Epoch: [1][ 940/ 1207] Overall Loss 1.175408 Objective Loss 1.175408 LR 0.001000 Time 0.014197 -2023-02-13 17:20:32,559 - Epoch: [1][ 950/ 1207] Overall Loss 1.174019 Objective Loss 1.174019 LR 0.001000 Time 0.014186 -2023-02-13 17:20:32,691 - Epoch: [1][ 960/ 1207] Overall Loss 1.172648 Objective Loss 1.172648 LR 0.001000 Time 0.014175 -2023-02-13 17:20:32,823 - Epoch: [1][ 970/ 1207] Overall Loss 1.171492 Objective Loss 1.171492 LR 0.001000 Time 0.014164 -2023-02-13 17:20:32,954 - Epoch: [1][ 980/ 1207] Overall Loss 1.170797 Objective Loss 1.170797 LR 0.001000 Time 0.014153 -2023-02-13 17:20:33,086 - Epoch: [1][ 990/ 1207] Overall Loss 1.169670 Objective Loss 1.169670 LR 0.001000 Time 0.014143 -2023-02-13 17:20:33,219 - Epoch: [1][ 1000/ 1207] Overall Loss 1.168214 Objective Loss 1.168214 LR 0.001000 Time 0.014135 -2023-02-13 17:20:33,353 - Epoch: [1][ 1010/ 1207] Overall Loss 1.167189 Objective Loss 1.167189 LR 0.001000 Time 0.014127 -2023-02-13 17:20:33,485 - Epoch: [1][ 1020/ 1207] Overall Loss 1.165754 Objective Loss 1.165754 LR 0.001000 Time 0.014118 -2023-02-13 17:20:33,617 - Epoch: [1][ 1030/ 1207] Overall Loss 1.164498 Objective Loss 1.164498 LR 0.001000 Time 0.014108 -2023-02-13 17:20:33,750 - Epoch: [1][ 1040/ 1207] Overall Loss 1.163456 Objective Loss 1.163456 LR 0.001000 Time 0.014099 -2023-02-13 17:20:33,882 - Epoch: [1][ 1050/ 1207] Overall Loss 1.162699 Objective Loss 1.162699 LR 0.001000 Time 0.014091 -2023-02-13 17:20:34,015 - Epoch: [1][ 1060/ 1207] Overall Loss 1.161385 Objective Loss 1.161385 LR 0.001000 Time 0.014082 -2023-02-13 17:20:34,147 - Epoch: [1][ 1070/ 1207] Overall Loss 1.159818 Objective Loss 1.159818 LR 0.001000 Time 0.014073 -2023-02-13 17:20:34,281 - Epoch: [1][ 1080/ 1207] Overall Loss 1.159117 Objective Loss 1.159117 LR 0.001000 Time 0.014067 -2023-02-13 17:20:34,414 - Epoch: [1][ 1090/ 1207] Overall Loss 1.158056 Objective Loss 1.158056 LR 0.001000 Time 0.014060 -2023-02-13 17:20:34,548 - Epoch: [1][ 1100/ 1207] Overall Loss 1.156515 Objective Loss 1.156515 LR 0.001000 Time 0.014053 -2023-02-13 17:20:34,680 - Epoch: [1][ 1110/ 1207] Overall Loss 1.155379 Objective Loss 1.155379 LR 0.001000 Time 0.014045 -2023-02-13 17:20:34,811 - Epoch: [1][ 1120/ 1207] Overall Loss 1.153830 Objective Loss 1.153830 LR 0.001000 Time 0.014037 -2023-02-13 17:20:34,943 - Epoch: [1][ 1130/ 1207] Overall Loss 1.152667 Objective Loss 1.152667 LR 0.001000 Time 0.014029 -2023-02-13 17:20:35,076 - Epoch: [1][ 1140/ 1207] Overall Loss 1.151514 Objective Loss 1.151514 LR 0.001000 Time 0.014021 -2023-02-13 17:20:35,208 - Epoch: [1][ 1150/ 1207] Overall Loss 1.150204 Objective Loss 1.150204 LR 0.001000 Time 0.014013 -2023-02-13 17:20:35,342 - Epoch: [1][ 1160/ 1207] Overall Loss 1.148741 Objective Loss 1.148741 LR 0.001000 Time 0.014007 -2023-02-13 17:20:35,473 - Epoch: [1][ 1170/ 1207] Overall Loss 1.147158 Objective Loss 1.147158 LR 0.001000 Time 0.014000 -2023-02-13 17:20:35,605 - Epoch: [1][ 1180/ 1207] Overall Loss 1.145965 Objective Loss 1.145965 LR 0.001000 Time 0.013993 -2023-02-13 17:20:35,738 - Epoch: [1][ 1190/ 1207] Overall Loss 1.144822 Objective Loss 1.144822 LR 0.001000 Time 0.013987 -2023-02-13 17:20:35,920 - Epoch: [1][ 1200/ 1207] Overall Loss 1.143738 Objective Loss 1.143738 LR 0.001000 Time 0.014021 -2023-02-13 17:20:36,005 - Epoch: [1][ 1207/ 1207] Overall Loss 1.142871 Objective Loss 1.142871 Top1 57.317073 Top5 89.939024 LR 0.001000 Time 0.014010 -2023-02-13 17:20:36,075 - --- validate (epoch=1)----------- -2023-02-13 17:20:36,075 - 34311 samples (256 per mini-batch) -2023-02-13 17:20:36,425 - Epoch: [1][ 10/ 135] Loss 0.956458 Top1 59.453125 Top5 90.507812 -2023-02-13 17:20:36,515 - Epoch: [1][ 20/ 135] Loss 0.978940 Top1 59.863281 Top5 90.898438 -2023-02-13 17:20:36,604 - Epoch: [1][ 30/ 135] Loss 0.980901 Top1 59.752604 Top5 90.742188 -2023-02-13 17:20:36,693 - Epoch: [1][ 40/ 135] Loss 0.972179 Top1 59.716797 Top5 90.986328 -2023-02-13 17:20:36,781 - Epoch: [1][ 50/ 135] Loss 0.984990 Top1 59.429687 Top5 90.820312 -2023-02-13 17:20:36,869 - Epoch: [1][ 60/ 135] Loss 0.992210 Top1 59.238281 Top5 90.664062 -2023-02-13 17:20:36,955 - Epoch: [1][ 70/ 135] Loss 0.984601 Top1 59.408482 Top5 90.569196 -2023-02-13 17:20:37,043 - Epoch: [1][ 80/ 135] Loss 0.984374 Top1 59.536133 Top5 90.556641 -2023-02-13 17:20:37,130 - Epoch: [1][ 90/ 135] Loss 0.985604 Top1 59.448785 Top5 90.525174 -2023-02-13 17:20:37,223 - Epoch: [1][ 100/ 135] Loss 0.989464 Top1 59.304688 Top5 90.457031 -2023-02-13 17:20:37,310 - Epoch: [1][ 110/ 135] Loss 0.982914 Top1 59.549006 Top5 90.553977 -2023-02-13 17:20:37,399 - Epoch: [1][ 120/ 135] Loss 0.984597 Top1 59.475911 Top5 90.413411 -2023-02-13 17:20:37,495 - Epoch: [1][ 130/ 135] Loss 0.980060 Top1 59.612380 Top5 90.477764 -2023-02-13 17:20:37,521 - Epoch: [1][ 135/ 135] Loss 0.977907 Top1 59.566903 Top5 90.475358 -2023-02-13 17:20:37,592 - ==> Top1: 59.567 Top5: 90.475 Loss: 0.978 - -2023-02-13 17:20:37,592 - ==> Confusion: -[[ 785 5 5 2 21 2 0 1 8 80 3 1 5 11 4 13 4 2 2 4 9] - [ 6 716 5 4 42 57 0 9 19 4 36 0 6 10 39 2 27 1 33 4 13] - [ 21 8 705 6 29 24 109 70 0 11 14 4 4 4 1 10 2 4 8 11 13] - [ 20 10 64 618 8 37 17 26 8 8 57 2 3 1 44 4 9 20 36 3 21] - [ 24 9 8 0 936 6 0 0 1 6 1 7 3 5 8 18 15 2 0 2 15] - [ 8 89 22 16 36 610 7 58 12 10 11 25 10 56 13 9 8 6 17 34 13] - [ 8 4 90 2 6 17 883 32 0 0 4 2 1 1 0 10 1 2 1 26 9] - [ 1 14 76 7 2 65 14 615 0 2 23 4 6 1 0 1 2 3 142 38 8] - [ 61 5 1 2 3 4 0 2 779 54 14 2 5 16 36 0 5 1 13 2 4] - [ 346 0 12 2 19 5 7 1 28 522 0 1 0 25 9 8 7 4 2 1 13] - [ 6 57 54 28 9 32 12 29 24 5 669 1 4 3 11 0 3 1 89 4 10] - [ 2 6 2 0 6 20 2 7 1 1 6 709 107 18 2 20 16 22 6 43 9] - [ 4 2 2 6 2 16 2 0 3 0 1 205 550 8 9 16 14 96 2 8 13] - [ 27 12 19 0 36 68 2 4 9 47 0 40 13 645 14 15 19 6 5 25 18] - [ 61 27 1 19 38 7 0 1 76 27 2 3 3 3 758 2 12 16 17 0 19] - [ 9 1 7 2 23 5 9 3 0 0 1 29 4 5 0 882 17 14 3 11 21] - [ 15 12 1 1 46 7 1 1 3 0 3 10 5 7 13 25 876 3 2 10 20] - [ 7 1 2 4 4 5 2 2 12 1 2 47 119 8 6 39 3 768 0 7 12] - [ 5 25 7 30 2 11 1 72 13 5 35 6 8 0 31 0 1 3 823 3 5] - [ 1 3 12 0 3 16 32 22 0 0 2 26 7 2 0 6 5 1 5 992 13] - [ 529 489 303 210 600 538 238 287 184 329 213 350 342 353 438 309 678 195 489 763 5597]] - -2023-02-13 17:20:37,594 - ==> Best [Top1: 59.567 Top5: 90.475 Sparsity:0.00 Params: 148928 on epoch: 1] -2023-02-13 17:20:37,594 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:20:37,600 - - -2023-02-13 17:20:37,600 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:20:38,410 - Epoch: [2][ 10/ 1207] Overall Loss 0.952229 Objective Loss 0.952229 LR 0.001000 Time 0.080931 -2023-02-13 17:20:38,551 - Epoch: [2][ 20/ 1207] Overall Loss 0.978783 Objective Loss 0.978783 LR 0.001000 Time 0.047506 -2023-02-13 17:20:38,683 - Epoch: [2][ 30/ 1207] Overall Loss 0.971890 Objective Loss 0.971890 LR 0.001000 Time 0.036053 -2023-02-13 17:20:38,817 - Epoch: [2][ 40/ 1207] Overall Loss 0.974864 Objective Loss 0.974864 LR 0.001000 Time 0.030381 -2023-02-13 17:20:38,950 - Epoch: [2][ 50/ 1207] Overall Loss 0.987448 Objective Loss 0.987448 LR 0.001000 Time 0.026940 -2023-02-13 17:20:39,083 - Epoch: [2][ 60/ 1207] Overall Loss 0.990025 Objective Loss 0.990025 LR 0.001000 Time 0.024665 -2023-02-13 17:20:39,215 - Epoch: [2][ 70/ 1207] Overall Loss 0.994118 Objective Loss 0.994118 LR 0.001000 Time 0.023008 -2023-02-13 17:20:39,348 - Epoch: [2][ 80/ 1207] Overall Loss 0.989431 Objective Loss 0.989431 LR 0.001000 Time 0.021765 -2023-02-13 17:20:39,480 - Epoch: [2][ 90/ 1207] Overall Loss 0.993333 Objective Loss 0.993333 LR 0.001000 Time 0.020797 -2023-02-13 17:20:39,614 - Epoch: [2][ 100/ 1207] Overall Loss 0.990622 Objective Loss 0.990622 LR 0.001000 Time 0.020051 -2023-02-13 17:20:39,745 - Epoch: [2][ 110/ 1207] Overall Loss 0.984186 Objective Loss 0.984186 LR 0.001000 Time 0.019419 -2023-02-13 17:20:39,877 - Epoch: [2][ 120/ 1207] Overall Loss 0.987966 Objective Loss 0.987966 LR 0.001000 Time 0.018897 -2023-02-13 17:20:40,009 - Epoch: [2][ 130/ 1207] Overall Loss 0.987943 Objective Loss 0.987943 LR 0.001000 Time 0.018449 -2023-02-13 17:20:40,142 - Epoch: [2][ 140/ 1207] Overall Loss 0.986885 Objective Loss 0.986885 LR 0.001000 Time 0.018080 -2023-02-13 17:20:40,273 - Epoch: [2][ 150/ 1207] Overall Loss 0.985700 Objective Loss 0.985700 LR 0.001000 Time 0.017745 -2023-02-13 17:20:40,406 - Epoch: [2][ 160/ 1207] Overall Loss 0.984400 Objective Loss 0.984400 LR 0.001000 Time 0.017463 -2023-02-13 17:20:40,537 - Epoch: [2][ 170/ 1207] Overall Loss 0.982574 Objective Loss 0.982574 LR 0.001000 Time 0.017202 -2023-02-13 17:20:40,669 - Epoch: [2][ 180/ 1207] Overall Loss 0.980580 Objective Loss 0.980580 LR 0.001000 Time 0.016974 -2023-02-13 17:20:40,802 - Epoch: [2][ 190/ 1207] Overall Loss 0.976639 Objective Loss 0.976639 LR 0.001000 Time 0.016776 -2023-02-13 17:20:40,934 - Epoch: [2][ 200/ 1207] Overall Loss 0.976520 Objective Loss 0.976520 LR 0.001000 Time 0.016592 -2023-02-13 17:20:41,065 - Epoch: [2][ 210/ 1207] Overall Loss 0.973924 Objective Loss 0.973924 LR 0.001000 Time 0.016421 -2023-02-13 17:20:41,198 - Epoch: [2][ 220/ 1207] Overall Loss 0.972415 Objective Loss 0.972415 LR 0.001000 Time 0.016279 -2023-02-13 17:20:41,329 - Epoch: [2][ 230/ 1207] Overall Loss 0.973203 Objective Loss 0.973203 LR 0.001000 Time 0.016135 -2023-02-13 17:20:41,463 - Epoch: [2][ 240/ 1207] Overall Loss 0.972161 Objective Loss 0.972161 LR 0.001000 Time 0.016018 -2023-02-13 17:20:41,597 - Epoch: [2][ 250/ 1207] Overall Loss 0.974558 Objective Loss 0.974558 LR 0.001000 Time 0.015906 -2023-02-13 17:20:41,729 - Epoch: [2][ 260/ 1207] Overall Loss 0.973238 Objective Loss 0.973238 LR 0.001000 Time 0.015800 -2023-02-13 17:20:41,861 - Epoch: [2][ 270/ 1207] Overall Loss 0.972403 Objective Loss 0.972403 LR 0.001000 Time 0.015703 -2023-02-13 17:20:41,994 - Epoch: [2][ 280/ 1207] Overall Loss 0.970700 Objective Loss 0.970700 LR 0.001000 Time 0.015615 -2023-02-13 17:20:42,126 - Epoch: [2][ 290/ 1207] Overall Loss 0.969924 Objective Loss 0.969924 LR 0.001000 Time 0.015530 -2023-02-13 17:20:42,259 - Epoch: [2][ 300/ 1207] Overall Loss 0.968590 Objective Loss 0.968590 LR 0.001000 Time 0.015455 -2023-02-13 17:20:42,393 - Epoch: [2][ 310/ 1207] Overall Loss 0.967239 Objective Loss 0.967239 LR 0.001000 Time 0.015388 -2023-02-13 17:20:42,525 - Epoch: [2][ 320/ 1207] Overall Loss 0.966671 Objective Loss 0.966671 LR 0.001000 Time 0.015318 -2023-02-13 17:20:42,657 - Epoch: [2][ 330/ 1207] Overall Loss 0.966010 Objective Loss 0.966010 LR 0.001000 Time 0.015251 -2023-02-13 17:20:42,790 - Epoch: [2][ 340/ 1207] Overall Loss 0.965156 Objective Loss 0.965156 LR 0.001000 Time 0.015194 -2023-02-13 17:20:42,921 - Epoch: [2][ 350/ 1207] Overall Loss 0.964199 Objective Loss 0.964199 LR 0.001000 Time 0.015131 -2023-02-13 17:20:43,054 - Epoch: [2][ 360/ 1207] Overall Loss 0.962138 Objective Loss 0.962138 LR 0.001000 Time 0.015079 -2023-02-13 17:20:43,189 - Epoch: [2][ 370/ 1207] Overall Loss 0.961938 Objective Loss 0.961938 LR 0.001000 Time 0.015034 -2023-02-13 17:20:43,320 - Epoch: [2][ 380/ 1207] Overall Loss 0.961748 Objective Loss 0.961748 LR 0.001000 Time 0.014982 -2023-02-13 17:20:43,452 - Epoch: [2][ 390/ 1207] Overall Loss 0.960047 Objective Loss 0.960047 LR 0.001000 Time 0.014935 -2023-02-13 17:20:43,584 - Epoch: [2][ 400/ 1207] Overall Loss 0.959828 Objective Loss 0.959828 LR 0.001000 Time 0.014892 -2023-02-13 17:20:43,717 - Epoch: [2][ 410/ 1207] Overall Loss 0.959109 Objective Loss 0.959109 LR 0.001000 Time 0.014850 -2023-02-13 17:20:43,850 - Epoch: [2][ 420/ 1207] Overall Loss 0.957762 Objective Loss 0.957762 LR 0.001000 Time 0.014813 -2023-02-13 17:20:43,983 - Epoch: [2][ 430/ 1207] Overall Loss 0.957616 Objective Loss 0.957616 LR 0.001000 Time 0.014778 -2023-02-13 17:20:44,116 - Epoch: [2][ 440/ 1207] Overall Loss 0.956749 Objective Loss 0.956749 LR 0.001000 Time 0.014742 -2023-02-13 17:20:44,249 - Epoch: [2][ 450/ 1207] Overall Loss 0.956698 Objective Loss 0.956698 LR 0.001000 Time 0.014710 -2023-02-13 17:20:44,381 - Epoch: [2][ 460/ 1207] Overall Loss 0.956301 Objective Loss 0.956301 LR 0.001000 Time 0.014675 -2023-02-13 17:20:44,512 - Epoch: [2][ 470/ 1207] Overall Loss 0.955932 Objective Loss 0.955932 LR 0.001000 Time 0.014642 -2023-02-13 17:20:44,644 - Epoch: [2][ 480/ 1207] Overall Loss 0.956125 Objective Loss 0.956125 LR 0.001000 Time 0.014611 -2023-02-13 17:20:44,779 - Epoch: [2][ 490/ 1207] Overall Loss 0.955561 Objective Loss 0.955561 LR 0.001000 Time 0.014586 -2023-02-13 17:20:44,910 - Epoch: [2][ 500/ 1207] Overall Loss 0.954947 Objective Loss 0.954947 LR 0.001000 Time 0.014556 -2023-02-13 17:20:45,041 - Epoch: [2][ 510/ 1207] Overall Loss 0.954927 Objective Loss 0.954927 LR 0.001000 Time 0.014528 -2023-02-13 17:20:45,175 - Epoch: [2][ 520/ 1207] Overall Loss 0.954181 Objective Loss 0.954181 LR 0.001000 Time 0.014504 -2023-02-13 17:20:45,306 - Epoch: [2][ 530/ 1207] Overall Loss 0.953066 Objective Loss 0.953066 LR 0.001000 Time 0.014476 -2023-02-13 17:20:45,440 - Epoch: [2][ 540/ 1207] Overall Loss 0.953128 Objective Loss 0.953128 LR 0.001000 Time 0.014454 -2023-02-13 17:20:45,571 - Epoch: [2][ 550/ 1207] Overall Loss 0.952642 Objective Loss 0.952642 LR 0.001000 Time 0.014429 -2023-02-13 17:20:45,705 - Epoch: [2][ 560/ 1207] Overall Loss 0.951782 Objective Loss 0.951782 LR 0.001000 Time 0.014407 -2023-02-13 17:20:45,839 - Epoch: [2][ 570/ 1207] Overall Loss 0.951971 Objective Loss 0.951971 LR 0.001000 Time 0.014389 -2023-02-13 17:20:45,972 - Epoch: [2][ 580/ 1207] Overall Loss 0.951336 Objective Loss 0.951336 LR 0.001000 Time 0.014369 -2023-02-13 17:20:46,105 - Epoch: [2][ 590/ 1207] Overall Loss 0.950769 Objective Loss 0.950769 LR 0.001000 Time 0.014350 -2023-02-13 17:20:46,237 - Epoch: [2][ 600/ 1207] Overall Loss 0.949996 Objective Loss 0.949996 LR 0.001000 Time 0.014330 -2023-02-13 17:20:46,369 - Epoch: [2][ 610/ 1207] Overall Loss 0.949233 Objective Loss 0.949233 LR 0.001000 Time 0.014311 -2023-02-13 17:20:46,502 - Epoch: [2][ 620/ 1207] Overall Loss 0.948381 Objective Loss 0.948381 LR 0.001000 Time 0.014294 -2023-02-13 17:20:46,635 - Epoch: [2][ 630/ 1207] Overall Loss 0.947774 Objective Loss 0.947774 LR 0.001000 Time 0.014278 -2023-02-13 17:20:46,770 - Epoch: [2][ 640/ 1207] Overall Loss 0.947914 Objective Loss 0.947914 LR 0.001000 Time 0.014262 -2023-02-13 17:20:46,901 - Epoch: [2][ 650/ 1207] Overall Loss 0.947347 Objective Loss 0.947347 LR 0.001000 Time 0.014243 -2023-02-13 17:20:47,033 - Epoch: [2][ 660/ 1207] Overall Loss 0.946581 Objective Loss 0.946581 LR 0.001000 Time 0.014227 -2023-02-13 17:20:47,168 - Epoch: [2][ 670/ 1207] Overall Loss 0.945718 Objective Loss 0.945718 LR 0.001000 Time 0.014215 -2023-02-13 17:20:47,300 - Epoch: [2][ 680/ 1207] Overall Loss 0.944661 Objective Loss 0.944661 LR 0.001000 Time 0.014201 -2023-02-13 17:20:47,431 - Epoch: [2][ 690/ 1207] Overall Loss 0.943273 Objective Loss 0.943273 LR 0.001000 Time 0.014184 -2023-02-13 17:20:47,573 - Epoch: [2][ 700/ 1207] Overall Loss 0.942574 Objective Loss 0.942574 LR 0.001000 Time 0.014183 -2023-02-13 17:20:47,718 - Epoch: [2][ 710/ 1207] Overall Loss 0.941690 Objective Loss 0.941690 LR 0.001000 Time 0.014188 -2023-02-13 17:20:47,862 - Epoch: [2][ 720/ 1207] Overall Loss 0.940905 Objective Loss 0.940905 LR 0.001000 Time 0.014190 -2023-02-13 17:20:48,004 - Epoch: [2][ 730/ 1207] Overall Loss 0.940181 Objective Loss 0.940181 LR 0.001000 Time 0.014189 -2023-02-13 17:20:48,147 - Epoch: [2][ 740/ 1207] Overall Loss 0.939027 Objective Loss 0.939027 LR 0.001000 Time 0.014190 -2023-02-13 17:20:48,290 - Epoch: [2][ 750/ 1207] Overall Loss 0.937776 Objective Loss 0.937776 LR 0.001000 Time 0.014191 -2023-02-13 17:20:48,434 - Epoch: [2][ 760/ 1207] Overall Loss 0.937356 Objective Loss 0.937356 LR 0.001000 Time 0.014193 -2023-02-13 17:20:48,576 - Epoch: [2][ 770/ 1207] Overall Loss 0.936940 Objective Loss 0.936940 LR 0.001000 Time 0.014193 -2023-02-13 17:20:48,720 - Epoch: [2][ 780/ 1207] Overall Loss 0.936287 Objective Loss 0.936287 LR 0.001000 Time 0.014195 -2023-02-13 17:20:48,862 - Epoch: [2][ 790/ 1207] Overall Loss 0.935058 Objective Loss 0.935058 LR 0.001000 Time 0.014195 -2023-02-13 17:20:49,005 - Epoch: [2][ 800/ 1207] Overall Loss 0.933976 Objective Loss 0.933976 LR 0.001000 Time 0.014196 -2023-02-13 17:20:49,147 - Epoch: [2][ 810/ 1207] Overall Loss 0.933262 Objective Loss 0.933262 LR 0.001000 Time 0.014195 -2023-02-13 17:20:49,291 - Epoch: [2][ 820/ 1207] Overall Loss 0.932744 Objective Loss 0.932744 LR 0.001000 Time 0.014197 -2023-02-13 17:20:49,432 - Epoch: [2][ 830/ 1207] Overall Loss 0.931594 Objective Loss 0.931594 LR 0.001000 Time 0.014196 -2023-02-13 17:20:49,575 - Epoch: [2][ 840/ 1207] Overall Loss 0.930436 Objective Loss 0.930436 LR 0.001000 Time 0.014197 -2023-02-13 17:20:49,718 - Epoch: [2][ 850/ 1207] Overall Loss 0.929675 Objective Loss 0.929675 LR 0.001000 Time 0.014197 -2023-02-13 17:20:49,862 - Epoch: [2][ 860/ 1207] Overall Loss 0.928424 Objective Loss 0.928424 LR 0.001000 Time 0.014199 -2023-02-13 17:20:50,005 - Epoch: [2][ 870/ 1207] Overall Loss 0.927671 Objective Loss 0.927671 LR 0.001000 Time 0.014199 -2023-02-13 17:20:50,149 - Epoch: [2][ 880/ 1207] Overall Loss 0.927029 Objective Loss 0.927029 LR 0.001000 Time 0.014202 -2023-02-13 17:20:50,291 - Epoch: [2][ 890/ 1207] Overall Loss 0.926179 Objective Loss 0.926179 LR 0.001000 Time 0.014201 -2023-02-13 17:20:50,435 - Epoch: [2][ 900/ 1207] Overall Loss 0.925458 Objective Loss 0.925458 LR 0.001000 Time 0.014203 -2023-02-13 17:20:50,578 - Epoch: [2][ 910/ 1207] Overall Loss 0.924531 Objective Loss 0.924531 LR 0.001000 Time 0.014203 -2023-02-13 17:20:50,721 - Epoch: [2][ 920/ 1207] Overall Loss 0.923948 Objective Loss 0.923948 LR 0.001000 Time 0.014204 -2023-02-13 17:20:50,864 - Epoch: [2][ 930/ 1207] Overall Loss 0.922898 Objective Loss 0.922898 LR 0.001000 Time 0.014204 -2023-02-13 17:20:51,007 - Epoch: [2][ 940/ 1207] Overall Loss 0.922246 Objective Loss 0.922246 LR 0.001000 Time 0.014206 -2023-02-13 17:20:51,150 - Epoch: [2][ 950/ 1207] Overall Loss 0.921893 Objective Loss 0.921893 LR 0.001000 Time 0.014206 -2023-02-13 17:20:51,294 - Epoch: [2][ 960/ 1207] Overall Loss 0.921831 Objective Loss 0.921831 LR 0.001000 Time 0.014208 -2023-02-13 17:20:51,435 - Epoch: [2][ 970/ 1207] Overall Loss 0.921394 Objective Loss 0.921394 LR 0.001000 Time 0.014206 -2023-02-13 17:20:51,568 - Epoch: [2][ 980/ 1207] Overall Loss 0.921228 Objective Loss 0.921228 LR 0.001000 Time 0.014196 -2023-02-13 17:20:51,699 - Epoch: [2][ 990/ 1207] Overall Loss 0.920134 Objective Loss 0.920134 LR 0.001000 Time 0.014184 -2023-02-13 17:20:51,833 - Epoch: [2][ 1000/ 1207] Overall Loss 0.919129 Objective Loss 0.919129 LR 0.001000 Time 0.014175 -2023-02-13 17:20:51,964 - Epoch: [2][ 1010/ 1207] Overall Loss 0.918347 Objective Loss 0.918347 LR 0.001000 Time 0.014164 -2023-02-13 17:20:52,097 - Epoch: [2][ 1020/ 1207] Overall Loss 0.917680 Objective Loss 0.917680 LR 0.001000 Time 0.014155 -2023-02-13 17:20:52,228 - Epoch: [2][ 1030/ 1207] Overall Loss 0.916673 Objective Loss 0.916673 LR 0.001000 Time 0.014145 -2023-02-13 17:20:52,360 - Epoch: [2][ 1040/ 1207] Overall Loss 0.915444 Objective Loss 0.915444 LR 0.001000 Time 0.014136 -2023-02-13 17:20:52,492 - Epoch: [2][ 1050/ 1207] Overall Loss 0.914905 Objective Loss 0.914905 LR 0.001000 Time 0.014126 -2023-02-13 17:20:52,624 - Epoch: [2][ 1060/ 1207] Overall Loss 0.914683 Objective Loss 0.914683 LR 0.001000 Time 0.014117 -2023-02-13 17:20:52,755 - Epoch: [2][ 1070/ 1207] Overall Loss 0.913754 Objective Loss 0.913754 LR 0.001000 Time 0.014107 -2023-02-13 17:20:52,887 - Epoch: [2][ 1080/ 1207] Overall Loss 0.913552 Objective Loss 0.913552 LR 0.001000 Time 0.014099 -2023-02-13 17:20:53,019 - Epoch: [2][ 1090/ 1207] Overall Loss 0.912489 Objective Loss 0.912489 LR 0.001000 Time 0.014090 -2023-02-13 17:20:53,150 - Epoch: [2][ 1100/ 1207] Overall Loss 0.911483 Objective Loss 0.911483 LR 0.001000 Time 0.014080 -2023-02-13 17:20:53,282 - Epoch: [2][ 1110/ 1207] Overall Loss 0.911087 Objective Loss 0.911087 LR 0.001000 Time 0.014072 -2023-02-13 17:20:53,415 - Epoch: [2][ 1120/ 1207] Overall Loss 0.910438 Objective Loss 0.910438 LR 0.001000 Time 0.014064 -2023-02-13 17:20:53,546 - Epoch: [2][ 1130/ 1207] Overall Loss 0.910298 Objective Loss 0.910298 LR 0.001000 Time 0.014055 -2023-02-13 17:20:53,679 - Epoch: [2][ 1140/ 1207] Overall Loss 0.909800 Objective Loss 0.909800 LR 0.001000 Time 0.014048 -2023-02-13 17:20:53,810 - Epoch: [2][ 1150/ 1207] Overall Loss 0.908806 Objective Loss 0.908806 LR 0.001000 Time 0.014039 -2023-02-13 17:20:53,942 - Epoch: [2][ 1160/ 1207] Overall Loss 0.908133 Objective Loss 0.908133 LR 0.001000 Time 0.014031 -2023-02-13 17:20:54,075 - Epoch: [2][ 1170/ 1207] Overall Loss 0.907234 Objective Loss 0.907234 LR 0.001000 Time 0.014023 -2023-02-13 17:20:54,207 - Epoch: [2][ 1180/ 1207] Overall Loss 0.906956 Objective Loss 0.906956 LR 0.001000 Time 0.014017 -2023-02-13 17:20:54,339 - Epoch: [2][ 1190/ 1207] Overall Loss 0.906504 Objective Loss 0.906504 LR 0.001000 Time 0.014009 -2023-02-13 17:20:54,526 - Epoch: [2][ 1200/ 1207] Overall Loss 0.905746 Objective Loss 0.905746 LR 0.001000 Time 0.014047 -2023-02-13 17:20:54,610 - Epoch: [2][ 1207/ 1207] Overall Loss 0.905436 Objective Loss 0.905436 Top1 68.292683 Top5 92.987805 LR 0.001000 Time 0.014035 -2023-02-13 17:20:54,680 - --- validate (epoch=2)----------- -2023-02-13 17:20:54,680 - 34311 samples (256 per mini-batch) -2023-02-13 17:20:55,034 - Epoch: [2][ 10/ 135] Loss 0.810649 Top1 65.703125 Top5 92.890625 -2023-02-13 17:20:55,121 - Epoch: [2][ 20/ 135] Loss 0.806583 Top1 65.527344 Top5 92.988281 -2023-02-13 17:20:55,209 - Epoch: [2][ 30/ 135] Loss 0.802405 Top1 65.091146 Top5 93.046875 -2023-02-13 17:20:55,297 - Epoch: [2][ 40/ 135] Loss 0.795023 Top1 65.136719 Top5 93.046875 -2023-02-13 17:20:55,383 - Epoch: [2][ 50/ 135] Loss 0.794655 Top1 65.273438 Top5 93.109375 -2023-02-13 17:20:55,470 - Epoch: [2][ 60/ 135] Loss 0.796113 Top1 65.110677 Top5 93.105469 -2023-02-13 17:20:55,561 - Epoch: [2][ 70/ 135] Loss 0.803817 Top1 64.927455 Top5 93.091518 -2023-02-13 17:20:55,652 - Epoch: [2][ 80/ 135] Loss 0.800840 Top1 65.014648 Top5 93.232422 -2023-02-13 17:20:55,745 - Epoch: [2][ 90/ 135] Loss 0.800341 Top1 65.021701 Top5 93.181424 -2023-02-13 17:20:55,835 - Epoch: [2][ 100/ 135] Loss 0.802389 Top1 65.085938 Top5 93.132812 -2023-02-13 17:20:55,927 - Epoch: [2][ 110/ 135] Loss 0.800533 Top1 65.113636 Top5 93.160511 -2023-02-13 17:20:56,018 - Epoch: [2][ 120/ 135] Loss 0.800564 Top1 65.039062 Top5 93.138021 -2023-02-13 17:20:56,117 - Epoch: [2][ 130/ 135] Loss 0.801286 Top1 65.006010 Top5 93.137019 -2023-02-13 17:20:56,144 - Epoch: [2][ 135/ 135] Loss 0.799719 Top1 65.002477 Top5 93.156714 -2023-02-13 17:20:56,215 - ==> Top1: 65.002 Top5: 93.157 Loss: 0.800 - -2023-02-13 17:20:56,215 - ==> Confusion: -[[ 822 3 8 3 5 6 0 2 9 36 2 8 1 11 8 12 11 6 0 3 11] - [ 3 868 2 1 7 49 2 7 9 2 15 6 5 2 10 0 24 2 7 4 8] - [ 23 17 731 18 16 11 90 48 1 3 34 4 0 4 1 14 2 7 5 12 17] - [ 9 9 29 748 3 17 2 5 2 4 66 2 8 2 31 3 8 22 27 1 18] - [ 26 26 9 0 887 22 0 0 1 6 0 10 1 5 11 10 32 1 2 1 16] - [ 11 139 6 9 11 741 6 29 5 3 13 23 5 23 6 4 16 2 3 9 6] - [ 8 6 52 2 3 19 928 14 0 0 7 6 2 1 0 10 4 3 0 27 7] - [ 4 35 37 9 0 99 16 672 2 0 23 8 3 2 0 1 2 2 74 21 14] - [ 40 16 1 2 4 1 0 4 805 23 23 5 5 20 47 0 3 1 5 0 4] - [ 344 8 11 0 18 12 4 0 56 472 1 1 0 35 15 3 4 5 2 1 20] - [ 5 44 19 23 0 24 5 9 16 2 824 6 5 2 9 0 4 2 42 2 8] - [ 8 3 3 0 2 17 4 5 3 0 0 761 93 17 2 14 24 24 0 17 8] - [ 3 1 0 3 2 2 2 2 1 0 2 82 715 6 4 11 15 95 0 7 6] - [ 14 24 6 0 16 87 3 2 21 21 11 43 9 679 15 3 36 5 0 19 10] - [ 39 32 1 31 15 12 0 3 34 3 1 2 13 2 833 1 22 16 14 1 17] - [ 7 8 8 0 7 5 12 0 0 0 0 13 11 1 0 865 62 23 0 6 18] - [ 3 11 1 0 10 6 3 0 3 0 3 7 6 4 3 13 972 2 1 3 10] - [ 3 1 3 7 2 1 0 0 2 0 1 24 73 3 3 27 5 887 0 3 6] - [ 6 36 2 32 0 9 1 54 18 1 46 4 14 1 32 0 2 2 816 2 8] - [ 0 10 7 0 0 18 32 22 0 0 3 33 8 4 0 3 14 2 2 976 14] - [ 396 679 251 218 261 564 147 191 136 131 314 310 463 340 352 224 962 213 329 652 6301]] - -2023-02-13 17:20:56,217 - ==> Best [Top1: 65.002 Top5: 93.157 Sparsity:0.00 Params: 148928 on epoch: 2] -2023-02-13 17:20:56,217 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:20:56,223 - - -2023-02-13 17:20:56,224 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:20:57,147 - Epoch: [3][ 10/ 1207] Overall Loss 0.795943 Objective Loss 0.795943 LR 0.001000 Time 0.092323 -2023-02-13 17:20:57,308 - Epoch: [3][ 20/ 1207] Overall Loss 0.799422 Objective Loss 0.799422 LR 0.001000 Time 0.054187 -2023-02-13 17:20:57,441 - Epoch: [3][ 30/ 1207] Overall Loss 0.794822 Objective Loss 0.794822 LR 0.001000 Time 0.040520 -2023-02-13 17:20:57,573 - Epoch: [3][ 40/ 1207] Overall Loss 0.793537 Objective Loss 0.793537 LR 0.001000 Time 0.033655 -2023-02-13 17:20:57,705 - Epoch: [3][ 50/ 1207] Overall Loss 0.795885 Objective Loss 0.795885 LR 0.001000 Time 0.029553 -2023-02-13 17:20:57,836 - Epoch: [3][ 60/ 1207] Overall Loss 0.796005 Objective Loss 0.796005 LR 0.001000 Time 0.026814 -2023-02-13 17:20:57,967 - Epoch: [3][ 70/ 1207] Overall Loss 0.795110 Objective Loss 0.795110 LR 0.001000 Time 0.024852 -2023-02-13 17:20:58,099 - Epoch: [3][ 80/ 1207] Overall Loss 0.796040 Objective Loss 0.796040 LR 0.001000 Time 0.023371 -2023-02-13 17:20:58,232 - Epoch: [3][ 90/ 1207] Overall Loss 0.798253 Objective Loss 0.798253 LR 0.001000 Time 0.022248 -2023-02-13 17:20:58,362 - Epoch: [3][ 100/ 1207] Overall Loss 0.794358 Objective Loss 0.794358 LR 0.001000 Time 0.021323 -2023-02-13 17:20:58,494 - Epoch: [3][ 110/ 1207] Overall Loss 0.792229 Objective Loss 0.792229 LR 0.001000 Time 0.020569 -2023-02-13 17:20:58,625 - Epoch: [3][ 120/ 1207] Overall Loss 0.787049 Objective Loss 0.787049 LR 0.001000 Time 0.019941 -2023-02-13 17:20:58,757 - Epoch: [3][ 130/ 1207] Overall Loss 0.787522 Objective Loss 0.787522 LR 0.001000 Time 0.019418 -2023-02-13 17:20:58,888 - Epoch: [3][ 140/ 1207] Overall Loss 0.784605 Objective Loss 0.784605 LR 0.001000 Time 0.018958 -2023-02-13 17:20:59,020 - Epoch: [3][ 150/ 1207] Overall Loss 0.784873 Objective Loss 0.784873 LR 0.001000 Time 0.018565 -2023-02-13 17:20:59,150 - Epoch: [3][ 160/ 1207] Overall Loss 0.782845 Objective Loss 0.782845 LR 0.001000 Time 0.018212 -2023-02-13 17:20:59,282 - Epoch: [3][ 170/ 1207] Overall Loss 0.786419 Objective Loss 0.786419 LR 0.001000 Time 0.017915 -2023-02-13 17:20:59,413 - Epoch: [3][ 180/ 1207] Overall Loss 0.786169 Objective Loss 0.786169 LR 0.001000 Time 0.017639 -2023-02-13 17:20:59,545 - Epoch: [3][ 190/ 1207] Overall Loss 0.784346 Objective Loss 0.784346 LR 0.001000 Time 0.017406 -2023-02-13 17:20:59,675 - Epoch: [3][ 200/ 1207] Overall Loss 0.783460 Objective Loss 0.783460 LR 0.001000 Time 0.017184 -2023-02-13 17:20:59,808 - Epoch: [3][ 210/ 1207] Overall Loss 0.782330 Objective Loss 0.782330 LR 0.001000 Time 0.016996 -2023-02-13 17:20:59,938 - Epoch: [3][ 220/ 1207] Overall Loss 0.781482 Objective Loss 0.781482 LR 0.001000 Time 0.016813 -2023-02-13 17:21:00,069 - Epoch: [3][ 230/ 1207] Overall Loss 0.781760 Objective Loss 0.781760 LR 0.001000 Time 0.016646 -2023-02-13 17:21:00,201 - Epoch: [3][ 240/ 1207] Overall Loss 0.782572 Objective Loss 0.782572 LR 0.001000 Time 0.016494 -2023-02-13 17:21:00,332 - Epoch: [3][ 250/ 1207] Overall Loss 0.783986 Objective Loss 0.783986 LR 0.001000 Time 0.016356 -2023-02-13 17:21:00,464 - Epoch: [3][ 260/ 1207] Overall Loss 0.785342 Objective Loss 0.785342 LR 0.001000 Time 0.016228 -2023-02-13 17:21:00,597 - Epoch: [3][ 270/ 1207] Overall Loss 0.784712 Objective Loss 0.784712 LR 0.001000 Time 0.016121 -2023-02-13 17:21:00,731 - Epoch: [3][ 280/ 1207] Overall Loss 0.782990 Objective Loss 0.782990 LR 0.001000 Time 0.016020 -2023-02-13 17:21:00,863 - Epoch: [3][ 290/ 1207] Overall Loss 0.782077 Objective Loss 0.782077 LR 0.001000 Time 0.015921 -2023-02-13 17:21:00,994 - Epoch: [3][ 300/ 1207] Overall Loss 0.781143 Objective Loss 0.781143 LR 0.001000 Time 0.015826 -2023-02-13 17:21:01,126 - Epoch: [3][ 310/ 1207] Overall Loss 0.780937 Objective Loss 0.780937 LR 0.001000 Time 0.015738 -2023-02-13 17:21:01,259 - Epoch: [3][ 320/ 1207] Overall Loss 0.780807 Objective Loss 0.780807 LR 0.001000 Time 0.015660 -2023-02-13 17:21:01,392 - Epoch: [3][ 330/ 1207] Overall Loss 0.782389 Objective Loss 0.782389 LR 0.001000 Time 0.015587 -2023-02-13 17:21:01,524 - Epoch: [3][ 340/ 1207] Overall Loss 0.783939 Objective Loss 0.783939 LR 0.001000 Time 0.015517 -2023-02-13 17:21:01,658 - Epoch: [3][ 350/ 1207] Overall Loss 0.782748 Objective Loss 0.782748 LR 0.001000 Time 0.015455 -2023-02-13 17:21:01,791 - Epoch: [3][ 360/ 1207] Overall Loss 0.782717 Objective Loss 0.782717 LR 0.001000 Time 0.015394 -2023-02-13 17:21:01,924 - Epoch: [3][ 370/ 1207] Overall Loss 0.782847 Objective Loss 0.782847 LR 0.001000 Time 0.015333 -2023-02-13 17:21:02,057 - Epoch: [3][ 380/ 1207] Overall Loss 0.782744 Objective Loss 0.782744 LR 0.001000 Time 0.015279 -2023-02-13 17:21:02,191 - Epoch: [3][ 390/ 1207] Overall Loss 0.782768 Objective Loss 0.782768 LR 0.001000 Time 0.015229 -2023-02-13 17:21:02,324 - Epoch: [3][ 400/ 1207] Overall Loss 0.782149 Objective Loss 0.782149 LR 0.001000 Time 0.015178 -2023-02-13 17:21:02,456 - Epoch: [3][ 410/ 1207] Overall Loss 0.782846 Objective Loss 0.782846 LR 0.001000 Time 0.015130 -2023-02-13 17:21:02,589 - Epoch: [3][ 420/ 1207] Overall Loss 0.782491 Objective Loss 0.782491 LR 0.001000 Time 0.015086 -2023-02-13 17:21:02,724 - Epoch: [3][ 430/ 1207] Overall Loss 0.782414 Objective Loss 0.782414 LR 0.001000 Time 0.015047 -2023-02-13 17:21:02,857 - Epoch: [3][ 440/ 1207] Overall Loss 0.781903 Objective Loss 0.781903 LR 0.001000 Time 0.015007 -2023-02-13 17:21:02,990 - Epoch: [3][ 450/ 1207] Overall Loss 0.781196 Objective Loss 0.781196 LR 0.001000 Time 0.014968 -2023-02-13 17:21:03,123 - Epoch: [3][ 460/ 1207] Overall Loss 0.780172 Objective Loss 0.780172 LR 0.001000 Time 0.014930 -2023-02-13 17:21:03,257 - Epoch: [3][ 470/ 1207] Overall Loss 0.779662 Objective Loss 0.779662 LR 0.001000 Time 0.014897 -2023-02-13 17:21:03,390 - Epoch: [3][ 480/ 1207] Overall Loss 0.778951 Objective Loss 0.778951 LR 0.001000 Time 0.014863 -2023-02-13 17:21:03,523 - Epoch: [3][ 490/ 1207] Overall Loss 0.779258 Objective Loss 0.779258 LR 0.001000 Time 0.014830 -2023-02-13 17:21:03,656 - Epoch: [3][ 500/ 1207] Overall Loss 0.779799 Objective Loss 0.779799 LR 0.001000 Time 0.014799 -2023-02-13 17:21:03,789 - Epoch: [3][ 510/ 1207] Overall Loss 0.779866 Objective Loss 0.779866 LR 0.001000 Time 0.014770 -2023-02-13 17:21:03,922 - Epoch: [3][ 520/ 1207] Overall Loss 0.779989 Objective Loss 0.779989 LR 0.001000 Time 0.014739 -2023-02-13 17:21:04,056 - Epoch: [3][ 530/ 1207] Overall Loss 0.778889 Objective Loss 0.778889 LR 0.001000 Time 0.014711 -2023-02-13 17:21:04,189 - Epoch: [3][ 540/ 1207] Overall Loss 0.778200 Objective Loss 0.778200 LR 0.001000 Time 0.014682 -2023-02-13 17:21:04,322 - Epoch: [3][ 550/ 1207] Overall Loss 0.778770 Objective Loss 0.778770 LR 0.001000 Time 0.014658 -2023-02-13 17:21:04,455 - Epoch: [3][ 560/ 1207] Overall Loss 0.777819 Objective Loss 0.777819 LR 0.001000 Time 0.014632 -2023-02-13 17:21:04,588 - Epoch: [3][ 570/ 1207] Overall Loss 0.778206 Objective Loss 0.778206 LR 0.001000 Time 0.014608 -2023-02-13 17:21:04,721 - Epoch: [3][ 580/ 1207] Overall Loss 0.777667 Objective Loss 0.777667 LR 0.001000 Time 0.014585 -2023-02-13 17:21:04,856 - Epoch: [3][ 590/ 1207] Overall Loss 0.777695 Objective Loss 0.777695 LR 0.001000 Time 0.014565 -2023-02-13 17:21:04,988 - Epoch: [3][ 600/ 1207] Overall Loss 0.777573 Objective Loss 0.777573 LR 0.001000 Time 0.014542 -2023-02-13 17:21:05,121 - Epoch: [3][ 610/ 1207] Overall Loss 0.778303 Objective Loss 0.778303 LR 0.001000 Time 0.014521 -2023-02-13 17:21:05,254 - Epoch: [3][ 620/ 1207] Overall Loss 0.777879 Objective Loss 0.777879 LR 0.001000 Time 0.014502 -2023-02-13 17:21:05,388 - Epoch: [3][ 630/ 1207] Overall Loss 0.777739 Objective Loss 0.777739 LR 0.001000 Time 0.014483 -2023-02-13 17:21:05,520 - Epoch: [3][ 640/ 1207] Overall Loss 0.777210 Objective Loss 0.777210 LR 0.001000 Time 0.014463 -2023-02-13 17:21:05,653 - Epoch: [3][ 650/ 1207] Overall Loss 0.777304 Objective Loss 0.777304 LR 0.001000 Time 0.014444 -2023-02-13 17:21:05,787 - Epoch: [3][ 660/ 1207] Overall Loss 0.776521 Objective Loss 0.776521 LR 0.001000 Time 0.014427 -2023-02-13 17:21:05,920 - Epoch: [3][ 670/ 1207] Overall Loss 0.776882 Objective Loss 0.776882 LR 0.001000 Time 0.014410 -2023-02-13 17:21:06,053 - Epoch: [3][ 680/ 1207] Overall Loss 0.776022 Objective Loss 0.776022 LR 0.001000 Time 0.014392 -2023-02-13 17:21:06,185 - Epoch: [3][ 690/ 1207] Overall Loss 0.775708 Objective Loss 0.775708 LR 0.001000 Time 0.014375 -2023-02-13 17:21:06,319 - Epoch: [3][ 700/ 1207] Overall Loss 0.775417 Objective Loss 0.775417 LR 0.001000 Time 0.014358 -2023-02-13 17:21:06,452 - Epoch: [3][ 710/ 1207] Overall Loss 0.775165 Objective Loss 0.775165 LR 0.001000 Time 0.014344 -2023-02-13 17:21:06,584 - Epoch: [3][ 720/ 1207] Overall Loss 0.774874 Objective Loss 0.774874 LR 0.001000 Time 0.014327 -2023-02-13 17:21:06,730 - Epoch: [3][ 730/ 1207] Overall Loss 0.774259 Objective Loss 0.774259 LR 0.001000 Time 0.014329 -2023-02-13 17:21:06,874 - Epoch: [3][ 740/ 1207] Overall Loss 0.773605 Objective Loss 0.773605 LR 0.001000 Time 0.014329 -2023-02-13 17:21:07,019 - Epoch: [3][ 750/ 1207] Overall Loss 0.773135 Objective Loss 0.773135 LR 0.001000 Time 0.014330 -2023-02-13 17:21:07,160 - Epoch: [3][ 760/ 1207] Overall Loss 0.772581 Objective Loss 0.772581 LR 0.001000 Time 0.014327 -2023-02-13 17:21:07,305 - Epoch: [3][ 770/ 1207] Overall Loss 0.771702 Objective Loss 0.771702 LR 0.001000 Time 0.014330 -2023-02-13 17:21:07,447 - Epoch: [3][ 780/ 1207] Overall Loss 0.771273 Objective Loss 0.771273 LR 0.001000 Time 0.014327 -2023-02-13 17:21:07,592 - Epoch: [3][ 790/ 1207] Overall Loss 0.771561 Objective Loss 0.771561 LR 0.001000 Time 0.014328 -2023-02-13 17:21:07,734 - Epoch: [3][ 800/ 1207] Overall Loss 0.771303 Objective Loss 0.771303 LR 0.001000 Time 0.014326 -2023-02-13 17:21:07,878 - Epoch: [3][ 810/ 1207] Overall Loss 0.770557 Objective Loss 0.770557 LR 0.001000 Time 0.014327 -2023-02-13 17:21:08,019 - Epoch: [3][ 820/ 1207] Overall Loss 0.769944 Objective Loss 0.769944 LR 0.001000 Time 0.014324 -2023-02-13 17:21:08,164 - Epoch: [3][ 830/ 1207] Overall Loss 0.770362 Objective Loss 0.770362 LR 0.001000 Time 0.014325 -2023-02-13 17:21:08,306 - Epoch: [3][ 840/ 1207] Overall Loss 0.769478 Objective Loss 0.769478 LR 0.001000 Time 0.014323 -2023-02-13 17:21:08,450 - Epoch: [3][ 850/ 1207] Overall Loss 0.769146 Objective Loss 0.769146 LR 0.001000 Time 0.014324 -2023-02-13 17:21:08,592 - Epoch: [3][ 860/ 1207] Overall Loss 0.768861 Objective Loss 0.768861 LR 0.001000 Time 0.014322 -2023-02-13 17:21:08,737 - Epoch: [3][ 870/ 1207] Overall Loss 0.768481 Objective Loss 0.768481 LR 0.001000 Time 0.014323 -2023-02-13 17:21:08,879 - Epoch: [3][ 880/ 1207] Overall Loss 0.767973 Objective Loss 0.767973 LR 0.001000 Time 0.014322 -2023-02-13 17:21:09,023 - Epoch: [3][ 890/ 1207] Overall Loss 0.767941 Objective Loss 0.767941 LR 0.001000 Time 0.014322 -2023-02-13 17:21:09,166 - Epoch: [3][ 900/ 1207] Overall Loss 0.767423 Objective Loss 0.767423 LR 0.001000 Time 0.014321 -2023-02-13 17:21:09,311 - Epoch: [3][ 910/ 1207] Overall Loss 0.767601 Objective Loss 0.767601 LR 0.001000 Time 0.014323 -2023-02-13 17:21:09,453 - Epoch: [3][ 920/ 1207] Overall Loss 0.767351 Objective Loss 0.767351 LR 0.001000 Time 0.014321 -2023-02-13 17:21:09,598 - Epoch: [3][ 930/ 1207] Overall Loss 0.766799 Objective Loss 0.766799 LR 0.001000 Time 0.014322 -2023-02-13 17:21:09,740 - Epoch: [3][ 940/ 1207] Overall Loss 0.766630 Objective Loss 0.766630 LR 0.001000 Time 0.014321 -2023-02-13 17:21:09,884 - Epoch: [3][ 950/ 1207] Overall Loss 0.765942 Objective Loss 0.765942 LR 0.001000 Time 0.014322 -2023-02-13 17:21:10,026 - Epoch: [3][ 960/ 1207] Overall Loss 0.765681 Objective Loss 0.765681 LR 0.001000 Time 0.014320 -2023-02-13 17:21:10,171 - Epoch: [3][ 970/ 1207] Overall Loss 0.765455 Objective Loss 0.765455 LR 0.001000 Time 0.014321 -2023-02-13 17:21:10,314 - Epoch: [3][ 980/ 1207] Overall Loss 0.764503 Objective Loss 0.764503 LR 0.001000 Time 0.014321 -2023-02-13 17:21:10,458 - Epoch: [3][ 990/ 1207] Overall Loss 0.764105 Objective Loss 0.764105 LR 0.001000 Time 0.014321 -2023-02-13 17:21:10,600 - Epoch: [3][ 1000/ 1207] Overall Loss 0.763675 Objective Loss 0.763675 LR 0.001000 Time 0.014319 -2023-02-13 17:21:10,746 - Epoch: [3][ 1010/ 1207] Overall Loss 0.763147 Objective Loss 0.763147 LR 0.001000 Time 0.014322 -2023-02-13 17:21:10,888 - Epoch: [3][ 1020/ 1207] Overall Loss 0.762737 Objective Loss 0.762737 LR 0.001000 Time 0.014320 -2023-02-13 17:21:11,033 - Epoch: [3][ 1030/ 1207] Overall Loss 0.762418 Objective Loss 0.762418 LR 0.001000 Time 0.014322 -2023-02-13 17:21:11,176 - Epoch: [3][ 1040/ 1207] Overall Loss 0.761594 Objective Loss 0.761594 LR 0.001000 Time 0.014321 -2023-02-13 17:21:11,321 - Epoch: [3][ 1050/ 1207] Overall Loss 0.761569 Objective Loss 0.761569 LR 0.001000 Time 0.014322 -2023-02-13 17:21:11,463 - Epoch: [3][ 1060/ 1207] Overall Loss 0.761009 Objective Loss 0.761009 LR 0.001000 Time 0.014321 -2023-02-13 17:21:11,607 - Epoch: [3][ 1070/ 1207] Overall Loss 0.760685 Objective Loss 0.760685 LR 0.001000 Time 0.014322 -2023-02-13 17:21:11,744 - Epoch: [3][ 1080/ 1207] Overall Loss 0.760231 Objective Loss 0.760231 LR 0.001000 Time 0.014314 -2023-02-13 17:21:11,879 - Epoch: [3][ 1090/ 1207] Overall Loss 0.759700 Objective Loss 0.759700 LR 0.001000 Time 0.014306 -2023-02-13 17:21:12,016 - Epoch: [3][ 1100/ 1207] Overall Loss 0.759053 Objective Loss 0.759053 LR 0.001000 Time 0.014300 -2023-02-13 17:21:12,150 - Epoch: [3][ 1110/ 1207] Overall Loss 0.758892 Objective Loss 0.758892 LR 0.001000 Time 0.014292 -2023-02-13 17:21:12,286 - Epoch: [3][ 1120/ 1207] Overall Loss 0.758471 Objective Loss 0.758471 LR 0.001000 Time 0.014286 -2023-02-13 17:21:12,420 - Epoch: [3][ 1130/ 1207] Overall Loss 0.757854 Objective Loss 0.757854 LR 0.001000 Time 0.014277 -2023-02-13 17:21:12,554 - Epoch: [3][ 1140/ 1207] Overall Loss 0.757175 Objective Loss 0.757175 LR 0.001000 Time 0.014269 -2023-02-13 17:21:12,688 - Epoch: [3][ 1150/ 1207] Overall Loss 0.756973 Objective Loss 0.756973 LR 0.001000 Time 0.014262 -2023-02-13 17:21:12,822 - Epoch: [3][ 1160/ 1207] Overall Loss 0.756508 Objective Loss 0.756508 LR 0.001000 Time 0.014254 -2023-02-13 17:21:12,956 - Epoch: [3][ 1170/ 1207] Overall Loss 0.756059 Objective Loss 0.756059 LR 0.001000 Time 0.014246 -2023-02-13 17:21:13,089 - Epoch: [3][ 1180/ 1207] Overall Loss 0.755413 Objective Loss 0.755413 LR 0.001000 Time 0.014238 -2023-02-13 17:21:13,224 - Epoch: [3][ 1190/ 1207] Overall Loss 0.755036 Objective Loss 0.755036 LR 0.001000 Time 0.014230 -2023-02-13 17:21:13,414 - Epoch: [3][ 1200/ 1207] Overall Loss 0.754351 Objective Loss 0.754351 LR 0.001000 Time 0.014270 -2023-02-13 17:21:13,500 - Epoch: [3][ 1207/ 1207] Overall Loss 0.753664 Objective Loss 0.753664 Top1 64.634146 Top5 96.036585 LR 0.001000 Time 0.014258 -2023-02-13 17:21:13,569 - --- validate (epoch=3)----------- -2023-02-13 17:21:13,570 - 34311 samples (256 per mini-batch) -2023-02-13 17:21:13,923 - Epoch: [3][ 10/ 135] Loss 0.644988 Top1 69.648438 Top5 94.687500 -2023-02-13 17:21:14,011 - Epoch: [3][ 20/ 135] Loss 0.664917 Top1 69.082031 Top5 94.570312 -2023-02-13 17:21:14,105 - Epoch: [3][ 30/ 135] Loss 0.666611 Top1 69.179688 Top5 94.778646 -2023-02-13 17:21:14,205 - Epoch: [3][ 40/ 135] Loss 0.671384 Top1 68.916016 Top5 94.667969 -2023-02-13 17:21:14,307 - Epoch: [3][ 50/ 135] Loss 0.666139 Top1 69.093750 Top5 94.734375 -2023-02-13 17:21:14,396 - Epoch: [3][ 60/ 135] Loss 0.666009 Top1 69.173177 Top5 94.615885 -2023-02-13 17:21:14,481 - Epoch: [3][ 70/ 135] Loss 0.661888 Top1 69.268973 Top5 94.676339 -2023-02-13 17:21:14,566 - Epoch: [3][ 80/ 135] Loss 0.657634 Top1 69.477539 Top5 94.702148 -2023-02-13 17:21:14,652 - Epoch: [3][ 90/ 135] Loss 0.655877 Top1 69.678819 Top5 94.717882 -2023-02-13 17:21:14,739 - Epoch: [3][ 100/ 135] Loss 0.658825 Top1 69.585938 Top5 94.691406 -2023-02-13 17:21:14,824 - Epoch: [3][ 110/ 135] Loss 0.662201 Top1 69.641335 Top5 94.666193 -2023-02-13 17:21:14,911 - Epoch: [3][ 120/ 135] Loss 0.665656 Top1 69.537760 Top5 94.619141 -2023-02-13 17:21:15,001 - Epoch: [3][ 130/ 135] Loss 0.664649 Top1 69.618389 Top5 94.603365 -2023-02-13 17:21:15,025 - Epoch: [3][ 135/ 135] Loss 0.673693 Top1 69.616158 Top5 94.631459 -2023-02-13 17:21:15,091 - ==> Top1: 69.616 Top5: 94.631 Loss: 0.674 - -2023-02-13 17:21:15,092 - ==> Confusion: -[[ 764 1 11 4 5 3 0 4 17 81 1 9 0 7 26 5 3 7 3 0 16] - [ 1 842 2 1 7 81 3 19 10 1 5 11 3 3 16 0 4 2 13 2 7] - [ 13 8 839 34 3 7 36 41 4 3 16 5 0 5 6 7 1 8 6 3 13] - [ 9 4 20 827 2 10 2 7 5 4 27 4 6 1 24 1 1 14 32 1 15] - [ 17 28 7 0 895 21 1 1 1 9 2 8 1 6 31 9 5 4 1 1 18] - [ 6 67 7 9 6 840 7 37 5 4 5 25 3 20 6 2 0 4 7 2 8] - [ 10 3 52 3 1 12 957 18 0 0 5 8 1 2 0 3 1 5 1 12 5] - [ 2 12 24 6 0 63 9 791 4 1 6 4 4 4 0 0 1 4 73 9 7] - [ 19 3 1 2 2 2 0 3 846 20 13 5 2 16 54 0 0 3 12 0 6] - [ 197 4 12 0 4 6 3 1 85 621 0 0 0 32 28 0 0 2 2 0 15] - [ 4 17 15 39 2 19 3 8 20 3 839 4 2 7 9 0 0 2 50 4 4] - [ 5 4 0 0 0 22 2 9 4 0 0 854 35 13 2 7 4 24 2 13 5] - [ 3 0 1 7 1 6 1 3 5 0 0 137 670 1 6 4 2 92 7 5 8] - [ 12 7 5 2 8 45 2 6 35 14 9 33 6 799 15 2 4 4 2 7 7] - [ 16 6 2 51 10 2 0 1 35 4 2 4 9 0 913 2 0 6 16 0 13] - [ 9 5 12 4 2 4 12 1 0 0 0 26 1 2 0 899 11 28 0 7 23] - [ 3 20 3 2 21 7 1 0 3 1 4 16 3 2 9 16 904 4 4 9 29] - [ 5 1 5 6 1 1 0 1 5 0 0 34 35 2 4 17 0 923 1 3 7] - [ 5 10 6 30 0 4 0 57 7 0 14 4 6 2 19 0 0 3 914 1 4] - [ 0 7 3 0 0 22 21 36 0 0 2 44 4 8 0 4 3 4 3 979 8] - [ 302 469 343 317 186 561 123 298 213 124 243 441 378 358 428 172 244 172 495 597 6970]] - -2023-02-13 17:21:15,093 - ==> Best [Top1: 69.616 Top5: 94.631 Sparsity:0.00 Params: 148928 on epoch: 3] -2023-02-13 17:21:15,093 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:21:15,100 - - -2023-02-13 17:21:15,100 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:21:15,933 - Epoch: [4][ 10/ 1207] Overall Loss 0.650820 Objective Loss 0.650820 LR 0.001000 Time 0.083208 -2023-02-13 17:21:16,074 - Epoch: [4][ 20/ 1207] Overall Loss 0.672388 Objective Loss 0.672388 LR 0.001000 Time 0.048628 -2023-02-13 17:21:16,205 - Epoch: [4][ 30/ 1207] Overall Loss 0.675982 Objective Loss 0.675982 LR 0.001000 Time 0.036788 -2023-02-13 17:21:16,340 - Epoch: [4][ 40/ 1207] Overall Loss 0.673995 Objective Loss 0.673995 LR 0.001000 Time 0.030941 -2023-02-13 17:21:16,472 - Epoch: [4][ 50/ 1207] Overall Loss 0.666987 Objective Loss 0.666987 LR 0.001000 Time 0.027388 -2023-02-13 17:21:16,606 - Epoch: [4][ 60/ 1207] Overall Loss 0.662452 Objective Loss 0.662452 LR 0.001000 Time 0.025060 -2023-02-13 17:21:16,739 - Epoch: [4][ 70/ 1207] Overall Loss 0.656290 Objective Loss 0.656290 LR 0.001000 Time 0.023369 -2023-02-13 17:21:16,874 - Epoch: [4][ 80/ 1207] Overall Loss 0.659106 Objective Loss 0.659106 LR 0.001000 Time 0.022118 -2023-02-13 17:21:17,007 - Epoch: [4][ 90/ 1207] Overall Loss 0.657175 Objective Loss 0.657175 LR 0.001000 Time 0.021125 -2023-02-13 17:21:17,141 - Epoch: [4][ 100/ 1207] Overall Loss 0.659122 Objective Loss 0.659122 LR 0.001000 Time 0.020356 -2023-02-13 17:21:17,274 - Epoch: [4][ 110/ 1207] Overall Loss 0.660001 Objective Loss 0.660001 LR 0.001000 Time 0.019709 -2023-02-13 17:21:17,409 - Epoch: [4][ 120/ 1207] Overall Loss 0.664118 Objective Loss 0.664118 LR 0.001000 Time 0.019186 -2023-02-13 17:21:17,542 - Epoch: [4][ 130/ 1207] Overall Loss 0.662202 Objective Loss 0.662202 LR 0.001000 Time 0.018728 -2023-02-13 17:21:17,676 - Epoch: [4][ 140/ 1207] Overall Loss 0.664249 Objective Loss 0.664249 LR 0.001000 Time 0.018342 -2023-02-13 17:21:17,809 - Epoch: [4][ 150/ 1207] Overall Loss 0.667606 Objective Loss 0.667606 LR 0.001000 Time 0.018003 -2023-02-13 17:21:17,947 - Epoch: [4][ 160/ 1207] Overall Loss 0.669986 Objective Loss 0.669986 LR 0.001000 Time 0.017736 -2023-02-13 17:21:18,089 - Epoch: [4][ 170/ 1207] Overall Loss 0.669616 Objective Loss 0.669616 LR 0.001000 Time 0.017529 -2023-02-13 17:21:18,234 - Epoch: [4][ 180/ 1207] Overall Loss 0.668821 Objective Loss 0.668821 LR 0.001000 Time 0.017354 -2023-02-13 17:21:18,376 - Epoch: [4][ 190/ 1207] Overall Loss 0.669987 Objective Loss 0.669987 LR 0.001000 Time 0.017189 -2023-02-13 17:21:18,521 - Epoch: [4][ 200/ 1207] Overall Loss 0.672148 Objective Loss 0.672148 LR 0.001000 Time 0.017052 -2023-02-13 17:21:18,664 - Epoch: [4][ 210/ 1207] Overall Loss 0.672524 Objective Loss 0.672524 LR 0.001000 Time 0.016917 -2023-02-13 17:21:18,808 - Epoch: [4][ 220/ 1207] Overall Loss 0.673040 Objective Loss 0.673040 LR 0.001000 Time 0.016801 -2023-02-13 17:21:18,950 - Epoch: [4][ 230/ 1207] Overall Loss 0.674614 Objective Loss 0.674614 LR 0.001000 Time 0.016688 -2023-02-13 17:21:19,094 - Epoch: [4][ 240/ 1207] Overall Loss 0.674279 Objective Loss 0.674279 LR 0.001000 Time 0.016591 -2023-02-13 17:21:19,237 - Epoch: [4][ 250/ 1207] Overall Loss 0.674564 Objective Loss 0.674564 LR 0.001000 Time 0.016497 -2023-02-13 17:21:19,382 - Epoch: [4][ 260/ 1207] Overall Loss 0.674949 Objective Loss 0.674949 LR 0.001000 Time 0.016417 -2023-02-13 17:21:19,523 - Epoch: [4][ 270/ 1207] Overall Loss 0.674895 Objective Loss 0.674895 LR 0.001000 Time 0.016333 -2023-02-13 17:21:19,667 - Epoch: [4][ 280/ 1207] Overall Loss 0.677175 Objective Loss 0.677175 LR 0.001000 Time 0.016263 -2023-02-13 17:21:19,810 - Epoch: [4][ 290/ 1207] Overall Loss 0.676832 Objective Loss 0.676832 LR 0.001000 Time 0.016192 -2023-02-13 17:21:19,954 - Epoch: [4][ 300/ 1207] Overall Loss 0.676751 Objective Loss 0.676751 LR 0.001000 Time 0.016130 -2023-02-13 17:21:20,096 - Epoch: [4][ 310/ 1207] Overall Loss 0.676030 Objective Loss 0.676030 LR 0.001000 Time 0.016067 -2023-02-13 17:21:20,240 - Epoch: [4][ 320/ 1207] Overall Loss 0.675343 Objective Loss 0.675343 LR 0.001000 Time 0.016015 -2023-02-13 17:21:20,382 - Epoch: [4][ 330/ 1207] Overall Loss 0.676514 Objective Loss 0.676514 LR 0.001000 Time 0.015959 -2023-02-13 17:21:20,526 - Epoch: [4][ 340/ 1207] Overall Loss 0.677096 Objective Loss 0.677096 LR 0.001000 Time 0.015912 -2023-02-13 17:21:20,668 - Epoch: [4][ 350/ 1207] Overall Loss 0.678326 Objective Loss 0.678326 LR 0.001000 Time 0.015862 -2023-02-13 17:21:20,814 - Epoch: [4][ 360/ 1207] Overall Loss 0.677463 Objective Loss 0.677463 LR 0.001000 Time 0.015823 -2023-02-13 17:21:20,956 - Epoch: [4][ 370/ 1207] Overall Loss 0.676903 Objective Loss 0.676903 LR 0.001000 Time 0.015779 -2023-02-13 17:21:21,100 - Epoch: [4][ 380/ 1207] Overall Loss 0.675719 Objective Loss 0.675719 LR 0.001000 Time 0.015742 -2023-02-13 17:21:21,242 - Epoch: [4][ 390/ 1207] Overall Loss 0.676791 Objective Loss 0.676791 LR 0.001000 Time 0.015702 -2023-02-13 17:21:21,386 - Epoch: [4][ 400/ 1207] Overall Loss 0.677204 Objective Loss 0.677204 LR 0.001000 Time 0.015669 -2023-02-13 17:21:21,528 - Epoch: [4][ 410/ 1207] Overall Loss 0.678262 Objective Loss 0.678262 LR 0.001000 Time 0.015633 -2023-02-13 17:21:21,673 - Epoch: [4][ 420/ 1207] Overall Loss 0.678258 Objective Loss 0.678258 LR 0.001000 Time 0.015603 -2023-02-13 17:21:21,815 - Epoch: [4][ 430/ 1207] Overall Loss 0.677482 Objective Loss 0.677482 LR 0.001000 Time 0.015571 -2023-02-13 17:21:21,960 - Epoch: [4][ 440/ 1207] Overall Loss 0.677321 Objective Loss 0.677321 LR 0.001000 Time 0.015544 -2023-02-13 17:21:22,102 - Epoch: [4][ 450/ 1207] Overall Loss 0.676644 Objective Loss 0.676644 LR 0.001000 Time 0.015515 -2023-02-13 17:21:22,247 - Epoch: [4][ 460/ 1207] Overall Loss 0.675936 Objective Loss 0.675936 LR 0.001000 Time 0.015490 -2023-02-13 17:21:22,389 - Epoch: [4][ 470/ 1207] Overall Loss 0.675107 Objective Loss 0.675107 LR 0.001000 Time 0.015462 -2023-02-13 17:21:22,533 - Epoch: [4][ 480/ 1207] Overall Loss 0.675067 Objective Loss 0.675067 LR 0.001000 Time 0.015440 -2023-02-13 17:21:22,675 - Epoch: [4][ 490/ 1207] Overall Loss 0.675048 Objective Loss 0.675048 LR 0.001000 Time 0.015414 -2023-02-13 17:21:22,820 - Epoch: [4][ 500/ 1207] Overall Loss 0.675180 Objective Loss 0.675180 LR 0.001000 Time 0.015394 -2023-02-13 17:21:22,962 - Epoch: [4][ 510/ 1207] Overall Loss 0.674109 Objective Loss 0.674109 LR 0.001000 Time 0.015370 -2023-02-13 17:21:23,106 - Epoch: [4][ 520/ 1207] Overall Loss 0.673185 Objective Loss 0.673185 LR 0.001000 Time 0.015351 -2023-02-13 17:21:23,248 - Epoch: [4][ 530/ 1207] Overall Loss 0.673340 Objective Loss 0.673340 LR 0.001000 Time 0.015329 -2023-02-13 17:21:23,393 - Epoch: [4][ 540/ 1207] Overall Loss 0.672782 Objective Loss 0.672782 LR 0.001000 Time 0.015312 -2023-02-13 17:21:23,535 - Epoch: [4][ 550/ 1207] Overall Loss 0.672167 Objective Loss 0.672167 LR 0.001000 Time 0.015291 -2023-02-13 17:21:23,680 - Epoch: [4][ 560/ 1207] Overall Loss 0.672394 Objective Loss 0.672394 LR 0.001000 Time 0.015276 -2023-02-13 17:21:23,822 - Epoch: [4][ 570/ 1207] Overall Loss 0.672240 Objective Loss 0.672240 LR 0.001000 Time 0.015257 -2023-02-13 17:21:23,967 - Epoch: [4][ 580/ 1207] Overall Loss 0.672317 Objective Loss 0.672317 LR 0.001000 Time 0.015243 -2023-02-13 17:21:24,107 - Epoch: [4][ 590/ 1207] Overall Loss 0.671061 Objective Loss 0.671061 LR 0.001000 Time 0.015222 -2023-02-13 17:21:24,249 - Epoch: [4][ 600/ 1207] Overall Loss 0.670833 Objective Loss 0.670833 LR 0.001000 Time 0.015204 -2023-02-13 17:21:24,390 - Epoch: [4][ 610/ 1207] Overall Loss 0.671076 Objective Loss 0.671076 LR 0.001000 Time 0.015185 -2023-02-13 17:21:24,531 - Epoch: [4][ 620/ 1207] Overall Loss 0.671068 Objective Loss 0.671068 LR 0.001000 Time 0.015168 -2023-02-13 17:21:24,671 - Epoch: [4][ 630/ 1207] Overall Loss 0.670999 Objective Loss 0.670999 LR 0.001000 Time 0.015148 -2023-02-13 17:21:24,814 - Epoch: [4][ 640/ 1207] Overall Loss 0.670320 Objective Loss 0.670320 LR 0.001000 Time 0.015134 -2023-02-13 17:21:24,953 - Epoch: [4][ 650/ 1207] Overall Loss 0.669648 Objective Loss 0.669648 LR 0.001000 Time 0.015115 -2023-02-13 17:21:25,095 - Epoch: [4][ 660/ 1207] Overall Loss 0.668805 Objective Loss 0.668805 LR 0.001000 Time 0.015101 -2023-02-13 17:21:25,235 - Epoch: [4][ 670/ 1207] Overall Loss 0.668498 Objective Loss 0.668498 LR 0.001000 Time 0.015084 -2023-02-13 17:21:25,378 - Epoch: [4][ 680/ 1207] Overall Loss 0.668465 Objective Loss 0.668465 LR 0.001000 Time 0.015071 -2023-02-13 17:21:25,518 - Epoch: [4][ 690/ 1207] Overall Loss 0.668624 Objective Loss 0.668624 LR 0.001000 Time 0.015055 -2023-02-13 17:21:25,659 - Epoch: [4][ 700/ 1207] Overall Loss 0.668018 Objective Loss 0.668018 LR 0.001000 Time 0.015042 -2023-02-13 17:21:25,801 - Epoch: [4][ 710/ 1207] Overall Loss 0.667592 Objective Loss 0.667592 LR 0.001000 Time 0.015029 -2023-02-13 17:21:25,943 - Epoch: [4][ 720/ 1207] Overall Loss 0.667752 Objective Loss 0.667752 LR 0.001000 Time 0.015017 -2023-02-13 17:21:26,083 - Epoch: [4][ 730/ 1207] Overall Loss 0.667732 Objective Loss 0.667732 LR 0.001000 Time 0.015003 -2023-02-13 17:21:26,224 - Epoch: [4][ 740/ 1207] Overall Loss 0.668356 Objective Loss 0.668356 LR 0.001000 Time 0.014990 -2023-02-13 17:21:26,365 - Epoch: [4][ 750/ 1207] Overall Loss 0.668387 Objective Loss 0.668387 LR 0.001000 Time 0.014977 -2023-02-13 17:21:26,506 - Epoch: [4][ 760/ 1207] Overall Loss 0.668085 Objective Loss 0.668085 LR 0.001000 Time 0.014966 -2023-02-13 17:21:26,646 - Epoch: [4][ 770/ 1207] Overall Loss 0.668162 Objective Loss 0.668162 LR 0.001000 Time 0.014952 -2023-02-13 17:21:26,788 - Epoch: [4][ 780/ 1207] Overall Loss 0.667296 Objective Loss 0.667296 LR 0.001000 Time 0.014942 -2023-02-13 17:21:26,929 - Epoch: [4][ 790/ 1207] Overall Loss 0.666304 Objective Loss 0.666304 LR 0.001000 Time 0.014930 -2023-02-13 17:21:27,071 - Epoch: [4][ 800/ 1207] Overall Loss 0.666756 Objective Loss 0.666756 LR 0.001000 Time 0.014921 -2023-02-13 17:21:27,211 - Epoch: [4][ 810/ 1207] Overall Loss 0.666253 Objective Loss 0.666253 LR 0.001000 Time 0.014909 -2023-02-13 17:21:27,353 - Epoch: [4][ 820/ 1207] Overall Loss 0.666748 Objective Loss 0.666748 LR 0.001000 Time 0.014901 -2023-02-13 17:21:27,494 - Epoch: [4][ 830/ 1207] Overall Loss 0.666850 Objective Loss 0.666850 LR 0.001000 Time 0.014890 -2023-02-13 17:21:27,636 - Epoch: [4][ 840/ 1207] Overall Loss 0.666109 Objective Loss 0.666109 LR 0.001000 Time 0.014881 -2023-02-13 17:21:27,776 - Epoch: [4][ 850/ 1207] Overall Loss 0.665835 Objective Loss 0.665835 LR 0.001000 Time 0.014871 -2023-02-13 17:21:27,918 - Epoch: [4][ 860/ 1207] Overall Loss 0.665538 Objective Loss 0.665538 LR 0.001000 Time 0.014862 -2023-02-13 17:21:28,058 - Epoch: [4][ 870/ 1207] Overall Loss 0.665157 Objective Loss 0.665157 LR 0.001000 Time 0.014852 -2023-02-13 17:21:28,199 - Epoch: [4][ 880/ 1207] Overall Loss 0.664699 Objective Loss 0.664699 LR 0.001000 Time 0.014843 -2023-02-13 17:21:28,341 - Epoch: [4][ 890/ 1207] Overall Loss 0.664425 Objective Loss 0.664425 LR 0.001000 Time 0.014835 -2023-02-13 17:21:28,485 - Epoch: [4][ 900/ 1207] Overall Loss 0.664049 Objective Loss 0.664049 LR 0.001000 Time 0.014830 -2023-02-13 17:21:28,625 - Epoch: [4][ 910/ 1207] Overall Loss 0.663675 Objective Loss 0.663675 LR 0.001000 Time 0.014821 -2023-02-13 17:21:28,763 - Epoch: [4][ 920/ 1207] Overall Loss 0.663362 Objective Loss 0.663362 LR 0.001000 Time 0.014808 -2023-02-13 17:21:28,901 - Epoch: [4][ 930/ 1207] Overall Loss 0.663562 Objective Loss 0.663562 LR 0.001000 Time 0.014796 -2023-02-13 17:21:29,039 - Epoch: [4][ 940/ 1207] Overall Loss 0.663521 Objective Loss 0.663521 LR 0.001000 Time 0.014784 -2023-02-13 17:21:29,176 - Epoch: [4][ 950/ 1207] Overall Loss 0.663112 Objective Loss 0.663112 LR 0.001000 Time 0.014772 -2023-02-13 17:21:29,310 - Epoch: [4][ 960/ 1207] Overall Loss 0.662736 Objective Loss 0.662736 LR 0.001000 Time 0.014757 -2023-02-13 17:21:29,444 - Epoch: [4][ 970/ 1207] Overall Loss 0.662501 Objective Loss 0.662501 LR 0.001000 Time 0.014743 -2023-02-13 17:21:29,575 - Epoch: [4][ 980/ 1207] Overall Loss 0.662271 Objective Loss 0.662271 LR 0.001000 Time 0.014725 -2023-02-13 17:21:29,708 - Epoch: [4][ 990/ 1207] Overall Loss 0.661831 Objective Loss 0.661831 LR 0.001000 Time 0.014710 -2023-02-13 17:21:29,840 - Epoch: [4][ 1000/ 1207] Overall Loss 0.662092 Objective Loss 0.662092 LR 0.001000 Time 0.014694 -2023-02-13 17:21:29,972 - Epoch: [4][ 1010/ 1207] Overall Loss 0.661970 Objective Loss 0.661970 LR 0.001000 Time 0.014679 -2023-02-13 17:21:30,104 - Epoch: [4][ 1020/ 1207] Overall Loss 0.661890 Objective Loss 0.661890 LR 0.001000 Time 0.014663 -2023-02-13 17:21:30,238 - Epoch: [4][ 1030/ 1207] Overall Loss 0.661930 Objective Loss 0.661930 LR 0.001000 Time 0.014649 -2023-02-13 17:21:30,369 - Epoch: [4][ 1040/ 1207] Overall Loss 0.661504 Objective Loss 0.661504 LR 0.001000 Time 0.014634 -2023-02-13 17:21:30,501 - Epoch: [4][ 1050/ 1207] Overall Loss 0.661610 Objective Loss 0.661610 LR 0.001000 Time 0.014620 -2023-02-13 17:21:30,633 - Epoch: [4][ 1060/ 1207] Overall Loss 0.662076 Objective Loss 0.662076 LR 0.001000 Time 0.014605 -2023-02-13 17:21:30,768 - Epoch: [4][ 1070/ 1207] Overall Loss 0.661473 Objective Loss 0.661473 LR 0.001000 Time 0.014594 -2023-02-13 17:21:30,899 - Epoch: [4][ 1080/ 1207] Overall Loss 0.661466 Objective Loss 0.661466 LR 0.001000 Time 0.014580 -2023-02-13 17:21:31,032 - Epoch: [4][ 1090/ 1207] Overall Loss 0.661326 Objective Loss 0.661326 LR 0.001000 Time 0.014568 -2023-02-13 17:21:31,164 - Epoch: [4][ 1100/ 1207] Overall Loss 0.661027 Objective Loss 0.661027 LR 0.001000 Time 0.014554 -2023-02-13 17:21:31,298 - Epoch: [4][ 1110/ 1207] Overall Loss 0.660953 Objective Loss 0.660953 LR 0.001000 Time 0.014542 -2023-02-13 17:21:31,429 - Epoch: [4][ 1120/ 1207] Overall Loss 0.660503 Objective Loss 0.660503 LR 0.001000 Time 0.014529 -2023-02-13 17:21:31,560 - Epoch: [4][ 1130/ 1207] Overall Loss 0.660314 Objective Loss 0.660314 LR 0.001000 Time 0.014515 -2023-02-13 17:21:31,693 - Epoch: [4][ 1140/ 1207] Overall Loss 0.660024 Objective Loss 0.660024 LR 0.001000 Time 0.014504 -2023-02-13 17:21:31,825 - Epoch: [4][ 1150/ 1207] Overall Loss 0.659478 Objective Loss 0.659478 LR 0.001000 Time 0.014492 -2023-02-13 17:21:31,958 - Epoch: [4][ 1160/ 1207] Overall Loss 0.659385 Objective Loss 0.659385 LR 0.001000 Time 0.014480 -2023-02-13 17:21:32,089 - Epoch: [4][ 1170/ 1207] Overall Loss 0.658959 Objective Loss 0.658959 LR 0.001000 Time 0.014467 -2023-02-13 17:21:32,220 - Epoch: [4][ 1180/ 1207] Overall Loss 0.658853 Objective Loss 0.658853 LR 0.001000 Time 0.014456 -2023-02-13 17:21:32,353 - Epoch: [4][ 1190/ 1207] Overall Loss 0.658530 Objective Loss 0.658530 LR 0.001000 Time 0.014445 -2023-02-13 17:21:32,546 - Epoch: [4][ 1200/ 1207] Overall Loss 0.658635 Objective Loss 0.658635 LR 0.001000 Time 0.014484 -2023-02-13 17:21:32,631 - Epoch: [4][ 1207/ 1207] Overall Loss 0.658724 Objective Loss 0.658724 Top1 64.024390 Top5 90.548780 LR 0.001000 Time 0.014471 -2023-02-13 17:21:32,702 - --- validate (epoch=4)----------- -2023-02-13 17:21:32,702 - 34311 samples (256 per mini-batch) -2023-02-13 17:21:33,060 - Epoch: [4][ 10/ 135] Loss 0.686811 Top1 68.984375 Top5 94.296875 -2023-02-13 17:21:33,144 - Epoch: [4][ 20/ 135] Loss 0.686474 Top1 69.511719 Top5 94.257812 -2023-02-13 17:21:33,233 - Epoch: [4][ 30/ 135] Loss 0.668549 Top1 69.843750 Top5 94.622396 -2023-02-13 17:21:33,321 - Epoch: [4][ 40/ 135] Loss 0.663893 Top1 70.097656 Top5 94.707031 -2023-02-13 17:21:33,406 - Epoch: [4][ 50/ 135] Loss 0.665366 Top1 70.367188 Top5 94.570312 -2023-02-13 17:21:33,493 - Epoch: [4][ 60/ 135] Loss 0.655131 Top1 70.442708 Top5 94.661458 -2023-02-13 17:21:33,577 - Epoch: [4][ 70/ 135] Loss 0.649947 Top1 70.479911 Top5 94.654018 -2023-02-13 17:21:33,666 - Epoch: [4][ 80/ 135] Loss 0.651816 Top1 70.307617 Top5 94.609375 -2023-02-13 17:21:33,754 - Epoch: [4][ 90/ 135] Loss 0.653768 Top1 70.169271 Top5 94.665799 -2023-02-13 17:21:33,842 - Epoch: [4][ 100/ 135] Loss 0.652861 Top1 70.089844 Top5 94.679688 -2023-02-13 17:21:33,931 - Epoch: [4][ 110/ 135] Loss 0.650714 Top1 70.152699 Top5 94.584517 -2023-02-13 17:21:34,019 - Epoch: [4][ 120/ 135] Loss 0.645223 Top1 70.234375 Top5 94.661458 -2023-02-13 17:21:34,112 - Epoch: [4][ 130/ 135] Loss 0.644245 Top1 70.195312 Top5 94.672476 -2023-02-13 17:21:34,137 - Epoch: [4][ 135/ 135] Loss 0.647669 Top1 70.213634 Top5 94.707237 -2023-02-13 17:21:34,214 - ==> Top1: 70.214 Top5: 94.707 Loss: 0.648 - -2023-02-13 17:21:34,215 - ==> Confusion: -[[ 669 5 9 0 11 5 0 4 17 201 2 9 0 12 4 1 5 1 2 0 10] - [ 4 852 3 2 8 52 1 36 10 3 6 9 3 5 6 0 7 0 5 9 12] - [ 15 6 861 11 8 3 23 67 3 5 11 4 1 9 3 1 0 5 4 9 9] - [ 4 3 41 779 1 8 0 18 4 7 40 5 11 12 22 0 5 9 32 1 14] - [ 9 23 7 3 930 9 0 1 2 12 1 12 1 19 11 4 7 2 0 2 11] - [ 3 87 8 2 6 774 5 74 6 8 6 28 2 31 4 0 2 0 3 14 7] - [ 7 3 56 2 1 9 940 28 0 0 4 7 0 3 0 3 2 1 2 26 5] - [ 2 8 20 1 1 34 5 880 3 3 6 5 0 4 0 0 0 1 27 17 7] - [ 13 7 0 1 2 1 0 4 888 40 14 2 1 15 8 0 0 0 5 0 8] - [ 89 0 7 0 4 4 1 4 52 791 1 2 0 28 12 0 2 1 2 2 10] - [ 5 16 18 9 0 4 2 12 23 2 907 3 2 15 5 0 0 0 16 1 11] - [ 4 2 1 0 4 17 4 9 2 5 0 858 28 34 1 1 2 8 1 19 5] - [ 2 0 2 7 0 7 1 6 9 0 1 153 706 13 5 4 5 23 2 7 6] - [ 12 10 1 0 3 23 0 10 21 29 7 21 1 857 5 0 4 1 0 15 4] - [ 15 17 1 14 25 6 0 4 71 14 2 7 5 6 865 0 1 6 14 0 19] - [ 11 4 11 1 9 4 8 1 0 0 0 28 11 18 0 863 33 15 0 9 20] - [ 5 21 2 0 19 5 1 0 5 1 3 9 2 17 1 10 929 2 2 8 19] - [ 9 4 6 13 0 3 1 1 6 1 1 69 74 10 6 12 1 822 0 3 9] - [ 2 20 10 12 0 2 0 111 15 5 18 5 3 2 13 1 0 1 863 1 2] - [ 0 1 1 1 1 8 12 37 0 0 1 33 2 15 0 0 4 0 0 1026 6] - [ 264 530 342 123 306 405 83 482 216 258 266 383 398 572 221 107 413 62 220 752 7031]] - -2023-02-13 17:21:34,216 - ==> Best [Top1: 70.214 Top5: 94.707 Sparsity:0.00 Params: 148928 on epoch: 4] -2023-02-13 17:21:34,216 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:21:34,222 - - -2023-02-13 17:21:34,223 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:21:35,036 - Epoch: [5][ 10/ 1207] Overall Loss 0.602652 Objective Loss 0.602652 LR 0.001000 Time 0.081239 -2023-02-13 17:21:35,180 - Epoch: [5][ 20/ 1207] Overall Loss 0.570614 Objective Loss 0.570614 LR 0.001000 Time 0.047805 -2023-02-13 17:21:35,320 - Epoch: [5][ 30/ 1207] Overall Loss 0.583782 Objective Loss 0.583782 LR 0.001000 Time 0.036524 -2023-02-13 17:21:35,470 - Epoch: [5][ 40/ 1207] Overall Loss 0.597921 Objective Loss 0.597921 LR 0.001000 Time 0.031144 -2023-02-13 17:21:35,621 - Epoch: [5][ 50/ 1207] Overall Loss 0.600793 Objective Loss 0.600793 LR 0.001000 Time 0.027916 -2023-02-13 17:21:35,761 - Epoch: [5][ 60/ 1207] Overall Loss 0.597946 Objective Loss 0.597946 LR 0.001000 Time 0.025601 -2023-02-13 17:21:35,898 - Epoch: [5][ 70/ 1207] Overall Loss 0.599950 Objective Loss 0.599950 LR 0.001000 Time 0.023869 -2023-02-13 17:21:36,030 - Epoch: [5][ 80/ 1207] Overall Loss 0.596622 Objective Loss 0.596622 LR 0.001000 Time 0.022527 -2023-02-13 17:21:36,162 - Epoch: [5][ 90/ 1207] Overall Loss 0.595676 Objective Loss 0.595676 LR 0.001000 Time 0.021472 -2023-02-13 17:21:36,293 - Epoch: [5][ 100/ 1207] Overall Loss 0.599249 Objective Loss 0.599249 LR 0.001000 Time 0.020625 -2023-02-13 17:21:36,426 - Epoch: [5][ 110/ 1207] Overall Loss 0.602235 Objective Loss 0.602235 LR 0.001000 Time 0.019938 -2023-02-13 17:21:36,558 - Epoch: [5][ 120/ 1207] Overall Loss 0.600700 Objective Loss 0.600700 LR 0.001000 Time 0.019380 -2023-02-13 17:21:36,689 - Epoch: [5][ 130/ 1207] Overall Loss 0.599381 Objective Loss 0.599381 LR 0.001000 Time 0.018891 -2023-02-13 17:21:36,823 - Epoch: [5][ 140/ 1207] Overall Loss 0.599066 Objective Loss 0.599066 LR 0.001000 Time 0.018492 -2023-02-13 17:21:36,953 - Epoch: [5][ 150/ 1207] Overall Loss 0.600942 Objective Loss 0.600942 LR 0.001000 Time 0.018127 -2023-02-13 17:21:37,085 - Epoch: [5][ 160/ 1207] Overall Loss 0.601998 Objective Loss 0.601998 LR 0.001000 Time 0.017814 -2023-02-13 17:21:37,216 - Epoch: [5][ 170/ 1207] Overall Loss 0.600584 Objective Loss 0.600584 LR 0.001000 Time 0.017531 -2023-02-13 17:21:37,349 - Epoch: [5][ 180/ 1207] Overall Loss 0.600394 Objective Loss 0.600394 LR 0.001000 Time 0.017290 -2023-02-13 17:21:37,481 - Epoch: [5][ 190/ 1207] Overall Loss 0.600388 Objective Loss 0.600388 LR 0.001000 Time 0.017073 -2023-02-13 17:21:37,613 - Epoch: [5][ 200/ 1207] Overall Loss 0.601641 Objective Loss 0.601641 LR 0.001000 Time 0.016880 -2023-02-13 17:21:37,745 - Epoch: [5][ 210/ 1207] Overall Loss 0.602450 Objective Loss 0.602450 LR 0.001000 Time 0.016697 -2023-02-13 17:21:37,879 - Epoch: [5][ 220/ 1207] Overall Loss 0.601527 Objective Loss 0.601527 LR 0.001000 Time 0.016543 -2023-02-13 17:21:38,010 - Epoch: [5][ 230/ 1207] Overall Loss 0.603233 Objective Loss 0.603233 LR 0.001000 Time 0.016390 -2023-02-13 17:21:38,142 - Epoch: [5][ 240/ 1207] Overall Loss 0.603076 Objective Loss 0.603076 LR 0.001000 Time 0.016254 -2023-02-13 17:21:38,273 - Epoch: [5][ 250/ 1207] Overall Loss 0.603790 Objective Loss 0.603790 LR 0.001000 Time 0.016127 -2023-02-13 17:21:38,406 - Epoch: [5][ 260/ 1207] Overall Loss 0.602461 Objective Loss 0.602461 LR 0.001000 Time 0.016008 -2023-02-13 17:21:38,538 - Epoch: [5][ 270/ 1207] Overall Loss 0.601680 Objective Loss 0.601680 LR 0.001000 Time 0.015898 -2023-02-13 17:21:38,673 - Epoch: [5][ 280/ 1207] Overall Loss 0.602685 Objective Loss 0.602685 LR 0.001000 Time 0.015807 -2023-02-13 17:21:38,805 - Epoch: [5][ 290/ 1207] Overall Loss 0.601991 Objective Loss 0.601991 LR 0.001000 Time 0.015718 -2023-02-13 17:21:38,938 - Epoch: [5][ 300/ 1207] Overall Loss 0.602618 Objective Loss 0.602618 LR 0.001000 Time 0.015631 -2023-02-13 17:21:39,073 - Epoch: [5][ 310/ 1207] Overall Loss 0.603054 Objective Loss 0.603054 LR 0.001000 Time 0.015555 -2023-02-13 17:21:39,204 - Epoch: [5][ 320/ 1207] Overall Loss 0.603590 Objective Loss 0.603590 LR 0.001000 Time 0.015478 -2023-02-13 17:21:39,337 - Epoch: [5][ 330/ 1207] Overall Loss 0.603785 Objective Loss 0.603785 LR 0.001000 Time 0.015412 -2023-02-13 17:21:39,469 - Epoch: [5][ 340/ 1207] Overall Loss 0.604344 Objective Loss 0.604344 LR 0.001000 Time 0.015345 -2023-02-13 17:21:39,602 - Epoch: [5][ 350/ 1207] Overall Loss 0.604050 Objective Loss 0.604050 LR 0.001000 Time 0.015286 -2023-02-13 17:21:39,735 - Epoch: [5][ 360/ 1207] Overall Loss 0.604449 Objective Loss 0.604449 LR 0.001000 Time 0.015229 -2023-02-13 17:21:39,868 - Epoch: [5][ 370/ 1207] Overall Loss 0.603598 Objective Loss 0.603598 LR 0.001000 Time 0.015175 -2023-02-13 17:21:40,000 - Epoch: [5][ 380/ 1207] Overall Loss 0.603514 Objective Loss 0.603514 LR 0.001000 Time 0.015122 -2023-02-13 17:21:40,133 - Epoch: [5][ 390/ 1207] Overall Loss 0.602293 Objective Loss 0.602293 LR 0.001000 Time 0.015070 -2023-02-13 17:21:40,264 - Epoch: [5][ 400/ 1207] Overall Loss 0.602374 Objective Loss 0.602374 LR 0.001000 Time 0.015020 -2023-02-13 17:21:40,396 - Epoch: [5][ 410/ 1207] Overall Loss 0.602380 Objective Loss 0.602380 LR 0.001000 Time 0.014971 -2023-02-13 17:21:40,528 - Epoch: [5][ 420/ 1207] Overall Loss 0.602870 Objective Loss 0.602870 LR 0.001000 Time 0.014925 -2023-02-13 17:21:40,658 - Epoch: [5][ 430/ 1207] Overall Loss 0.602313 Objective Loss 0.602313 LR 0.001000 Time 0.014880 -2023-02-13 17:21:40,793 - Epoch: [5][ 440/ 1207] Overall Loss 0.602080 Objective Loss 0.602080 LR 0.001000 Time 0.014844 -2023-02-13 17:21:40,924 - Epoch: [5][ 450/ 1207] Overall Loss 0.601365 Objective Loss 0.601365 LR 0.001000 Time 0.014805 -2023-02-13 17:21:41,059 - Epoch: [5][ 460/ 1207] Overall Loss 0.601756 Objective Loss 0.601756 LR 0.001000 Time 0.014775 -2023-02-13 17:21:41,190 - Epoch: [5][ 470/ 1207] Overall Loss 0.602637 Objective Loss 0.602637 LR 0.001000 Time 0.014739 -2023-02-13 17:21:41,323 - Epoch: [5][ 480/ 1207] Overall Loss 0.602649 Objective Loss 0.602649 LR 0.001000 Time 0.014708 -2023-02-13 17:21:41,457 - Epoch: [5][ 490/ 1207] Overall Loss 0.602618 Objective Loss 0.602618 LR 0.001000 Time 0.014676 -2023-02-13 17:21:41,591 - Epoch: [5][ 500/ 1207] Overall Loss 0.602008 Objective Loss 0.602008 LR 0.001000 Time 0.014650 -2023-02-13 17:21:41,723 - Epoch: [5][ 510/ 1207] Overall Loss 0.601963 Objective Loss 0.601963 LR 0.001000 Time 0.014620 -2023-02-13 17:21:41,857 - Epoch: [5][ 520/ 1207] Overall Loss 0.601801 Objective Loss 0.601801 LR 0.001000 Time 0.014597 -2023-02-13 17:21:41,988 - Epoch: [5][ 530/ 1207] Overall Loss 0.601694 Objective Loss 0.601694 LR 0.001000 Time 0.014567 -2023-02-13 17:21:42,120 - Epoch: [5][ 540/ 1207] Overall Loss 0.601687 Objective Loss 0.601687 LR 0.001000 Time 0.014542 -2023-02-13 17:21:42,253 - Epoch: [5][ 550/ 1207] Overall Loss 0.600787 Objective Loss 0.600787 LR 0.001000 Time 0.014516 -2023-02-13 17:21:42,388 - Epoch: [5][ 560/ 1207] Overall Loss 0.601189 Objective Loss 0.601189 LR 0.001000 Time 0.014496 -2023-02-13 17:21:42,520 - Epoch: [5][ 570/ 1207] Overall Loss 0.601702 Objective Loss 0.601702 LR 0.001000 Time 0.014473 -2023-02-13 17:21:42,652 - Epoch: [5][ 580/ 1207] Overall Loss 0.601741 Objective Loss 0.601741 LR 0.001000 Time 0.014450 -2023-02-13 17:21:42,786 - Epoch: [5][ 590/ 1207] Overall Loss 0.601932 Objective Loss 0.601932 LR 0.001000 Time 0.014432 -2023-02-13 17:21:42,919 - Epoch: [5][ 600/ 1207] Overall Loss 0.601592 Objective Loss 0.601592 LR 0.001000 Time 0.014410 -2023-02-13 17:21:43,051 - Epoch: [5][ 610/ 1207] Overall Loss 0.601333 Objective Loss 0.601333 LR 0.001000 Time 0.014387 -2023-02-13 17:21:43,185 - Epoch: [5][ 620/ 1207] Overall Loss 0.601162 Objective Loss 0.601162 LR 0.001000 Time 0.014368 -2023-02-13 17:21:43,316 - Epoch: [5][ 630/ 1207] Overall Loss 0.601030 Objective Loss 0.601030 LR 0.001000 Time 0.014347 -2023-02-13 17:21:43,451 - Epoch: [5][ 640/ 1207] Overall Loss 0.600951 Objective Loss 0.600951 LR 0.001000 Time 0.014334 -2023-02-13 17:21:43,582 - Epoch: [5][ 650/ 1207] Overall Loss 0.600992 Objective Loss 0.600992 LR 0.001000 Time 0.014314 -2023-02-13 17:21:43,716 - Epoch: [5][ 660/ 1207] Overall Loss 0.600949 Objective Loss 0.600949 LR 0.001000 Time 0.014298 -2023-02-13 17:21:43,847 - Epoch: [5][ 670/ 1207] Overall Loss 0.600410 Objective Loss 0.600410 LR 0.001000 Time 0.014280 -2023-02-13 17:21:43,980 - Epoch: [5][ 680/ 1207] Overall Loss 0.600717 Objective Loss 0.600717 LR 0.001000 Time 0.014264 -2023-02-13 17:21:44,113 - Epoch: [5][ 690/ 1207] Overall Loss 0.599864 Objective Loss 0.599864 LR 0.001000 Time 0.014247 -2023-02-13 17:21:44,245 - Epoch: [5][ 700/ 1207] Overall Loss 0.600267 Objective Loss 0.600267 LR 0.001000 Time 0.014231 -2023-02-13 17:21:44,379 - Epoch: [5][ 710/ 1207] Overall Loss 0.600111 Objective Loss 0.600111 LR 0.001000 Time 0.014220 -2023-02-13 17:21:44,511 - Epoch: [5][ 720/ 1207] Overall Loss 0.599820 Objective Loss 0.599820 LR 0.001000 Time 0.014205 -2023-02-13 17:21:44,643 - Epoch: [5][ 730/ 1207] Overall Loss 0.600141 Objective Loss 0.600141 LR 0.001000 Time 0.014190 -2023-02-13 17:21:44,775 - Epoch: [5][ 740/ 1207] Overall Loss 0.599531 Objective Loss 0.599531 LR 0.001000 Time 0.014177 -2023-02-13 17:21:44,905 - Epoch: [5][ 750/ 1207] Overall Loss 0.599272 Objective Loss 0.599272 LR 0.001000 Time 0.014161 -2023-02-13 17:21:45,037 - Epoch: [5][ 760/ 1207] Overall Loss 0.599308 Objective Loss 0.599308 LR 0.001000 Time 0.014146 -2023-02-13 17:21:45,168 - Epoch: [5][ 770/ 1207] Overall Loss 0.598838 Objective Loss 0.598838 LR 0.001000 Time 0.014130 -2023-02-13 17:21:45,299 - Epoch: [5][ 780/ 1207] Overall Loss 0.598352 Objective Loss 0.598352 LR 0.001000 Time 0.014116 -2023-02-13 17:21:45,432 - Epoch: [5][ 790/ 1207] Overall Loss 0.598609 Objective Loss 0.598609 LR 0.001000 Time 0.014103 -2023-02-13 17:21:45,565 - Epoch: [5][ 800/ 1207] Overall Loss 0.599196 Objective Loss 0.599196 LR 0.001000 Time 0.014092 -2023-02-13 17:21:45,697 - Epoch: [5][ 810/ 1207] Overall Loss 0.598827 Objective Loss 0.598827 LR 0.001000 Time 0.014080 -2023-02-13 17:21:45,831 - Epoch: [5][ 820/ 1207] Overall Loss 0.598580 Objective Loss 0.598580 LR 0.001000 Time 0.014071 -2023-02-13 17:21:45,964 - Epoch: [5][ 830/ 1207] Overall Loss 0.598152 Objective Loss 0.598152 LR 0.001000 Time 0.014061 -2023-02-13 17:21:46,095 - Epoch: [5][ 840/ 1207] Overall Loss 0.597982 Objective Loss 0.597982 LR 0.001000 Time 0.014050 -2023-02-13 17:21:46,227 - Epoch: [5][ 850/ 1207] Overall Loss 0.597648 Objective Loss 0.597648 LR 0.001000 Time 0.014039 -2023-02-13 17:21:46,359 - Epoch: [5][ 860/ 1207] Overall Loss 0.597323 Objective Loss 0.597323 LR 0.001000 Time 0.014029 -2023-02-13 17:21:46,491 - Epoch: [5][ 870/ 1207] Overall Loss 0.597424 Objective Loss 0.597424 LR 0.001000 Time 0.014017 -2023-02-13 17:21:46,624 - Epoch: [5][ 880/ 1207] Overall Loss 0.597417 Objective Loss 0.597417 LR 0.001000 Time 0.014009 -2023-02-13 17:21:46,755 - Epoch: [5][ 890/ 1207] Overall Loss 0.597852 Objective Loss 0.597852 LR 0.001000 Time 0.013999 -2023-02-13 17:21:46,887 - Epoch: [5][ 900/ 1207] Overall Loss 0.597749 Objective Loss 0.597749 LR 0.001000 Time 0.013989 -2023-02-13 17:21:47,019 - Epoch: [5][ 910/ 1207] Overall Loss 0.597341 Objective Loss 0.597341 LR 0.001000 Time 0.013978 -2023-02-13 17:21:47,152 - Epoch: [5][ 920/ 1207] Overall Loss 0.598128 Objective Loss 0.598128 LR 0.001000 Time 0.013970 -2023-02-13 17:21:47,283 - Epoch: [5][ 930/ 1207] Overall Loss 0.598346 Objective Loss 0.598346 LR 0.001000 Time 0.013960 -2023-02-13 17:21:47,416 - Epoch: [5][ 940/ 1207] Overall Loss 0.598130 Objective Loss 0.598130 LR 0.001000 Time 0.013953 -2023-02-13 17:21:47,547 - Epoch: [5][ 950/ 1207] Overall Loss 0.597912 Objective Loss 0.597912 LR 0.001000 Time 0.013943 -2023-02-13 17:21:47,679 - Epoch: [5][ 960/ 1207] Overall Loss 0.597572 Objective Loss 0.597572 LR 0.001000 Time 0.013935 -2023-02-13 17:21:47,811 - Epoch: [5][ 970/ 1207] Overall Loss 0.597642 Objective Loss 0.597642 LR 0.001000 Time 0.013926 -2023-02-13 17:21:47,943 - Epoch: [5][ 980/ 1207] Overall Loss 0.597458 Objective Loss 0.597458 LR 0.001000 Time 0.013918 -2023-02-13 17:21:48,075 - Epoch: [5][ 990/ 1207] Overall Loss 0.597692 Objective Loss 0.597692 LR 0.001000 Time 0.013910 -2023-02-13 17:21:48,207 - Epoch: [5][ 1000/ 1207] Overall Loss 0.597370 Objective Loss 0.597370 LR 0.001000 Time 0.013902 -2023-02-13 17:21:48,339 - Epoch: [5][ 1010/ 1207] Overall Loss 0.597470 Objective Loss 0.597470 LR 0.001000 Time 0.013893 -2023-02-13 17:21:48,472 - Epoch: [5][ 1020/ 1207] Overall Loss 0.597779 Objective Loss 0.597779 LR 0.001000 Time 0.013887 -2023-02-13 17:21:48,604 - Epoch: [5][ 1030/ 1207] Overall Loss 0.597228 Objective Loss 0.597228 LR 0.001000 Time 0.013881 -2023-02-13 17:21:48,736 - Epoch: [5][ 1040/ 1207] Overall Loss 0.596865 Objective Loss 0.596865 LR 0.001000 Time 0.013874 -2023-02-13 17:21:48,868 - Epoch: [5][ 1050/ 1207] Overall Loss 0.596990 Objective Loss 0.596990 LR 0.001000 Time 0.013866 -2023-02-13 17:21:48,999 - Epoch: [5][ 1060/ 1207] Overall Loss 0.596402 Objective Loss 0.596402 LR 0.001000 Time 0.013859 -2023-02-13 17:21:49,131 - Epoch: [5][ 1070/ 1207] Overall Loss 0.596700 Objective Loss 0.596700 LR 0.001000 Time 0.013851 -2023-02-13 17:21:49,264 - Epoch: [5][ 1080/ 1207] Overall Loss 0.596563 Objective Loss 0.596563 LR 0.001000 Time 0.013846 -2023-02-13 17:21:49,395 - Epoch: [5][ 1090/ 1207] Overall Loss 0.596170 Objective Loss 0.596170 LR 0.001000 Time 0.013839 -2023-02-13 17:21:49,528 - Epoch: [5][ 1100/ 1207] Overall Loss 0.596209 Objective Loss 0.596209 LR 0.001000 Time 0.013833 -2023-02-13 17:21:49,659 - Epoch: [5][ 1110/ 1207] Overall Loss 0.596360 Objective Loss 0.596360 LR 0.001000 Time 0.013826 -2023-02-13 17:21:49,791 - Epoch: [5][ 1120/ 1207] Overall Loss 0.596142 Objective Loss 0.596142 LR 0.001000 Time 0.013821 -2023-02-13 17:21:49,923 - Epoch: [5][ 1130/ 1207] Overall Loss 0.596010 Objective Loss 0.596010 LR 0.001000 Time 0.013815 -2023-02-13 17:21:50,056 - Epoch: [5][ 1140/ 1207] Overall Loss 0.595659 Objective Loss 0.595659 LR 0.001000 Time 0.013810 -2023-02-13 17:21:50,186 - Epoch: [5][ 1150/ 1207] Overall Loss 0.595731 Objective Loss 0.595731 LR 0.001000 Time 0.013802 -2023-02-13 17:21:50,318 - Epoch: [5][ 1160/ 1207] Overall Loss 0.595177 Objective Loss 0.595177 LR 0.001000 Time 0.013797 -2023-02-13 17:21:50,450 - Epoch: [5][ 1170/ 1207] Overall Loss 0.594989 Objective Loss 0.594989 LR 0.001000 Time 0.013791 -2023-02-13 17:21:50,581 - Epoch: [5][ 1180/ 1207] Overall Loss 0.594785 Objective Loss 0.594785 LR 0.001000 Time 0.013785 -2023-02-13 17:21:50,714 - Epoch: [5][ 1190/ 1207] Overall Loss 0.594543 Objective Loss 0.594543 LR 0.001000 Time 0.013780 -2023-02-13 17:21:50,897 - Epoch: [5][ 1200/ 1207] Overall Loss 0.594589 Objective Loss 0.594589 LR 0.001000 Time 0.013818 -2023-02-13 17:21:50,983 - Epoch: [5][ 1207/ 1207] Overall Loss 0.594191 Objective Loss 0.594191 Top1 71.951220 Top5 96.646341 LR 0.001000 Time 0.013809 -2023-02-13 17:21:51,063 - --- validate (epoch=5)----------- -2023-02-13 17:21:51,063 - 34311 samples (256 per mini-batch) -2023-02-13 17:21:51,507 - Epoch: [5][ 10/ 135] Loss 0.562528 Top1 71.796875 Top5 94.609375 -2023-02-13 17:21:51,604 - Epoch: [5][ 20/ 135] Loss 0.563505 Top1 71.503906 Top5 95.312500 -2023-02-13 17:21:51,693 - Epoch: [5][ 30/ 135] Loss 0.563720 Top1 71.432292 Top5 95.312500 -2023-02-13 17:21:51,780 - Epoch: [5][ 40/ 135] Loss 0.552428 Top1 72.060547 Top5 95.449219 -2023-02-13 17:21:51,866 - Epoch: [5][ 50/ 135] Loss 0.559791 Top1 72.210938 Top5 95.617188 -2023-02-13 17:21:51,977 - Epoch: [5][ 60/ 135] Loss 0.549682 Top1 72.402344 Top5 95.709635 -2023-02-13 17:21:52,064 - Epoch: [5][ 70/ 135] Loss 0.547632 Top1 72.181920 Top5 95.731027 -2023-02-13 17:21:52,149 - Epoch: [5][ 80/ 135] Loss 0.551983 Top1 72.084961 Top5 95.639648 -2023-02-13 17:21:52,237 - Epoch: [5][ 90/ 135] Loss 0.550370 Top1 72.135417 Top5 95.655382 -2023-02-13 17:21:52,327 - Epoch: [5][ 100/ 135] Loss 0.551444 Top1 72.187500 Top5 95.625000 -2023-02-13 17:21:52,422 - Epoch: [5][ 110/ 135] Loss 0.550591 Top1 72.116477 Top5 95.600142 -2023-02-13 17:21:52,518 - Epoch: [5][ 120/ 135] Loss 0.550331 Top1 72.216797 Top5 95.556641 -2023-02-13 17:21:52,611 - Epoch: [5][ 130/ 135] Loss 0.553105 Top1 72.100361 Top5 95.546875 -2023-02-13 17:21:52,635 - Epoch: [5][ 135/ 135] Loss 0.554331 Top1 72.067267 Top5 95.520387 -2023-02-13 17:21:52,718 - ==> Top1: 72.067 Top5: 95.520 Loss: 0.554 - -2023-02-13 17:21:52,719 - ==> Confusion: -[[ 830 0 7 1 2 4 0 1 4 70 0 12 3 10 6 9 0 1 2 0 5] - [ 9 838 4 2 12 68 1 29 10 3 4 8 2 4 15 1 4 0 4 5 10] - [ 31 3 882 23 2 5 14 29 0 2 13 5 1 4 5 11 3 5 6 8 6] - [ 8 3 28 822 0 6 1 2 5 4 25 5 8 1 53 5 4 6 21 1 8] - [ 27 8 10 2 931 8 0 0 1 9 2 13 1 9 19 9 4 1 1 1 10] - [ 7 45 3 5 9 863 0 39 1 4 3 29 4 23 4 6 5 1 3 11 5] - [ 9 2 57 2 2 13 933 7 1 1 10 9 3 1 0 11 3 4 1 26 4] - [ 5 8 18 0 2 59 2 832 2 1 10 5 3 4 0 0 0 1 46 19 7] - [ 27 1 2 2 1 0 0 3 830 57 8 6 3 19 35 2 1 1 7 1 3] - [ 203 0 3 1 4 4 0 2 44 708 0 4 0 24 6 2 1 1 0 0 5] - [ 5 6 14 15 1 7 1 4 28 1 898 2 5 10 14 1 1 1 28 3 6] - [ 7 2 0 0 1 14 2 5 0 1 1 878 45 15 1 10 4 3 2 11 3] - [ 6 0 1 4 0 9 0 0 1 0 0 90 790 2 7 12 4 19 4 3 7] - [ 15 1 3 0 7 26 1 3 21 28 4 27 4 856 6 4 5 0 1 9 3] - [ 30 1 2 12 8 2 0 2 22 6 2 3 7 4 965 5 2 2 8 0 9] - [ 7 4 5 1 5 4 2 0 0 0 1 19 1 5 1 960 7 11 0 4 9] - [ 10 13 0 3 16 5 0 1 3 2 0 8 3 4 4 22 942 1 2 7 15] - [ 9 0 1 5 0 0 0 0 5 1 0 52 69 7 9 33 0 850 1 3 6] - [ 6 6 8 18 0 2 0 50 6 0 9 7 8 2 38 0 0 1 920 0 5] - [ 2 1 1 1 0 16 9 24 0 0 2 32 10 9 0 5 3 0 3 1025 5] - [ 444 305 346 163 217 445 46 281 148 247 239 375 439 471 440 311 381 80 290 592 7174]] - -2023-02-13 17:21:52,720 - ==> Best [Top1: 72.067 Top5: 95.520 Sparsity:0.00 Params: 148928 on epoch: 5] -2023-02-13 17:21:52,720 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:21:52,727 - - -2023-02-13 17:21:52,727 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:21:53,549 - Epoch: [6][ 10/ 1207] Overall Loss 0.605539 Objective Loss 0.605539 LR 0.001000 Time 0.082076 -2023-02-13 17:21:53,714 - Epoch: [6][ 20/ 1207] Overall Loss 0.562883 Objective Loss 0.562883 LR 0.001000 Time 0.049284 -2023-02-13 17:21:53,859 - Epoch: [6][ 30/ 1207] Overall Loss 0.558301 Objective Loss 0.558301 LR 0.001000 Time 0.037675 -2023-02-13 17:21:54,004 - Epoch: [6][ 40/ 1207] Overall Loss 0.557224 Objective Loss 0.557224 LR 0.001000 Time 0.031886 -2023-02-13 17:21:54,150 - Epoch: [6][ 50/ 1207] Overall Loss 0.560728 Objective Loss 0.560728 LR 0.001000 Time 0.028415 -2023-02-13 17:21:54,297 - Epoch: [6][ 60/ 1207] Overall Loss 0.559557 Objective Loss 0.559557 LR 0.001000 Time 0.026115 -2023-02-13 17:21:54,440 - Epoch: [6][ 70/ 1207] Overall Loss 0.560587 Objective Loss 0.560587 LR 0.001000 Time 0.024432 -2023-02-13 17:21:54,586 - Epoch: [6][ 80/ 1207] Overall Loss 0.558497 Objective Loss 0.558497 LR 0.001000 Time 0.023191 -2023-02-13 17:21:54,730 - Epoch: [6][ 90/ 1207] Overall Loss 0.560279 Objective Loss 0.560279 LR 0.001000 Time 0.022208 -2023-02-13 17:21:54,875 - Epoch: [6][ 100/ 1207] Overall Loss 0.556263 Objective Loss 0.556263 LR 0.001000 Time 0.021439 -2023-02-13 17:21:55,018 - Epoch: [6][ 110/ 1207] Overall Loss 0.556971 Objective Loss 0.556971 LR 0.001000 Time 0.020787 -2023-02-13 17:21:55,163 - Epoch: [6][ 120/ 1207] Overall Loss 0.558537 Objective Loss 0.558537 LR 0.001000 Time 0.020261 -2023-02-13 17:21:55,307 - Epoch: [6][ 130/ 1207] Overall Loss 0.557164 Objective Loss 0.557164 LR 0.001000 Time 0.019802 -2023-02-13 17:21:55,453 - Epoch: [6][ 140/ 1207] Overall Loss 0.557708 Objective Loss 0.557708 LR 0.001000 Time 0.019427 -2023-02-13 17:21:55,596 - Epoch: [6][ 150/ 1207] Overall Loss 0.559375 Objective Loss 0.559375 LR 0.001000 Time 0.019087 -2023-02-13 17:21:55,735 - Epoch: [6][ 160/ 1207] Overall Loss 0.558182 Objective Loss 0.558182 LR 0.001000 Time 0.018758 -2023-02-13 17:21:55,867 - Epoch: [6][ 170/ 1207] Overall Loss 0.559091 Objective Loss 0.559091 LR 0.001000 Time 0.018432 -2023-02-13 17:21:56,002 - Epoch: [6][ 180/ 1207] Overall Loss 0.559831 Objective Loss 0.559831 LR 0.001000 Time 0.018145 -2023-02-13 17:21:56,137 - Epoch: [6][ 190/ 1207] Overall Loss 0.560879 Objective Loss 0.560879 LR 0.001000 Time 0.017894 -2023-02-13 17:21:56,271 - Epoch: [6][ 200/ 1207] Overall Loss 0.561422 Objective Loss 0.561422 LR 0.001000 Time 0.017669 -2023-02-13 17:21:56,404 - Epoch: [6][ 210/ 1207] Overall Loss 0.560869 Objective Loss 0.560869 LR 0.001000 Time 0.017456 -2023-02-13 17:21:56,537 - Epoch: [6][ 220/ 1207] Overall Loss 0.558732 Objective Loss 0.558732 LR 0.001000 Time 0.017266 -2023-02-13 17:21:56,670 - Epoch: [6][ 230/ 1207] Overall Loss 0.556859 Objective Loss 0.556859 LR 0.001000 Time 0.017087 -2023-02-13 17:21:56,804 - Epoch: [6][ 240/ 1207] Overall Loss 0.558101 Objective Loss 0.558101 LR 0.001000 Time 0.016928 -2023-02-13 17:21:56,937 - Epoch: [6][ 250/ 1207] Overall Loss 0.556145 Objective Loss 0.556145 LR 0.001000 Time 0.016781 -2023-02-13 17:21:57,074 - Epoch: [6][ 260/ 1207] Overall Loss 0.555666 Objective Loss 0.555666 LR 0.001000 Time 0.016655 -2023-02-13 17:21:57,206 - Epoch: [6][ 270/ 1207] Overall Loss 0.555188 Objective Loss 0.555188 LR 0.001000 Time 0.016524 -2023-02-13 17:21:57,337 - Epoch: [6][ 280/ 1207] Overall Loss 0.555845 Objective Loss 0.555845 LR 0.001000 Time 0.016403 -2023-02-13 17:21:57,469 - Epoch: [6][ 290/ 1207] Overall Loss 0.553733 Objective Loss 0.553733 LR 0.001000 Time 0.016290 -2023-02-13 17:21:57,600 - Epoch: [6][ 300/ 1207] Overall Loss 0.553333 Objective Loss 0.553333 LR 0.001000 Time 0.016182 -2023-02-13 17:21:57,731 - Epoch: [6][ 310/ 1207] Overall Loss 0.554653 Objective Loss 0.554653 LR 0.001000 Time 0.016083 -2023-02-13 17:21:57,863 - Epoch: [6][ 320/ 1207] Overall Loss 0.554003 Objective Loss 0.554003 LR 0.001000 Time 0.015986 -2023-02-13 17:21:57,994 - Epoch: [6][ 330/ 1207] Overall Loss 0.555181 Objective Loss 0.555181 LR 0.001000 Time 0.015896 -2023-02-13 17:21:58,126 - Epoch: [6][ 340/ 1207] Overall Loss 0.555087 Objective Loss 0.555087 LR 0.001000 Time 0.015811 -2023-02-13 17:21:58,258 - Epoch: [6][ 350/ 1207] Overall Loss 0.554085 Objective Loss 0.554085 LR 0.001000 Time 0.015731 -2023-02-13 17:21:58,390 - Epoch: [6][ 360/ 1207] Overall Loss 0.554628 Objective Loss 0.554628 LR 0.001000 Time 0.015658 -2023-02-13 17:21:58,523 - Epoch: [6][ 370/ 1207] Overall Loss 0.556102 Objective Loss 0.556102 LR 0.001000 Time 0.015591 -2023-02-13 17:21:58,654 - Epoch: [6][ 380/ 1207] Overall Loss 0.556328 Objective Loss 0.556328 LR 0.001000 Time 0.015522 -2023-02-13 17:21:58,787 - Epoch: [6][ 390/ 1207] Overall Loss 0.557530 Objective Loss 0.557530 LR 0.001000 Time 0.015465 -2023-02-13 17:21:58,918 - Epoch: [6][ 400/ 1207] Overall Loss 0.557691 Objective Loss 0.557691 LR 0.001000 Time 0.015405 -2023-02-13 17:21:59,051 - Epoch: [6][ 410/ 1207] Overall Loss 0.557126 Objective Loss 0.557126 LR 0.001000 Time 0.015350 -2023-02-13 17:21:59,184 - Epoch: [6][ 420/ 1207] Overall Loss 0.556627 Objective Loss 0.556627 LR 0.001000 Time 0.015299 -2023-02-13 17:21:59,315 - Epoch: [6][ 430/ 1207] Overall Loss 0.555818 Objective Loss 0.555818 LR 0.001000 Time 0.015247 -2023-02-13 17:21:59,447 - Epoch: [6][ 440/ 1207] Overall Loss 0.556628 Objective Loss 0.556628 LR 0.001000 Time 0.015197 -2023-02-13 17:21:59,580 - Epoch: [6][ 450/ 1207] Overall Loss 0.556413 Objective Loss 0.556413 LR 0.001000 Time 0.015153 -2023-02-13 17:21:59,711 - Epoch: [6][ 460/ 1207] Overall Loss 0.556780 Objective Loss 0.556780 LR 0.001000 Time 0.015107 -2023-02-13 17:21:59,844 - Epoch: [6][ 470/ 1207] Overall Loss 0.557216 Objective Loss 0.557216 LR 0.001000 Time 0.015066 -2023-02-13 17:21:59,975 - Epoch: [6][ 480/ 1207] Overall Loss 0.557033 Objective Loss 0.557033 LR 0.001000 Time 0.015026 -2023-02-13 17:22:00,110 - Epoch: [6][ 490/ 1207] Overall Loss 0.556273 Objective Loss 0.556273 LR 0.001000 Time 0.014993 -2023-02-13 17:22:00,241 - Epoch: [6][ 500/ 1207] Overall Loss 0.555517 Objective Loss 0.555517 LR 0.001000 Time 0.014954 -2023-02-13 17:22:00,374 - Epoch: [6][ 510/ 1207] Overall Loss 0.554636 Objective Loss 0.554636 LR 0.001000 Time 0.014921 -2023-02-13 17:22:00,505 - Epoch: [6][ 520/ 1207] Overall Loss 0.554532 Objective Loss 0.554532 LR 0.001000 Time 0.014886 -2023-02-13 17:22:00,638 - Epoch: [6][ 530/ 1207] Overall Loss 0.555042 Objective Loss 0.555042 LR 0.001000 Time 0.014855 -2023-02-13 17:22:00,772 - Epoch: [6][ 540/ 1207] Overall Loss 0.554229 Objective Loss 0.554229 LR 0.001000 Time 0.014828 -2023-02-13 17:22:00,904 - Epoch: [6][ 550/ 1207] Overall Loss 0.554586 Objective Loss 0.554586 LR 0.001000 Time 0.014797 -2023-02-13 17:22:01,036 - Epoch: [6][ 560/ 1207] Overall Loss 0.554987 Objective Loss 0.554987 LR 0.001000 Time 0.014767 -2023-02-13 17:22:01,168 - Epoch: [6][ 570/ 1207] Overall Loss 0.555584 Objective Loss 0.555584 LR 0.001000 Time 0.014739 -2023-02-13 17:22:01,300 - Epoch: [6][ 580/ 1207] Overall Loss 0.555555 Objective Loss 0.555555 LR 0.001000 Time 0.014710 -2023-02-13 17:22:01,432 - Epoch: [6][ 590/ 1207] Overall Loss 0.556205 Objective Loss 0.556205 LR 0.001000 Time 0.014684 -2023-02-13 17:22:01,564 - Epoch: [6][ 600/ 1207] Overall Loss 0.555423 Objective Loss 0.555423 LR 0.001000 Time 0.014657 -2023-02-13 17:22:01,697 - Epoch: [6][ 610/ 1207] Overall Loss 0.555514 Objective Loss 0.555514 LR 0.001000 Time 0.014634 -2023-02-13 17:22:01,828 - Epoch: [6][ 620/ 1207] Overall Loss 0.555515 Objective Loss 0.555515 LR 0.001000 Time 0.014609 -2023-02-13 17:22:01,961 - Epoch: [6][ 630/ 1207] Overall Loss 0.555721 Objective Loss 0.555721 LR 0.001000 Time 0.014587 -2023-02-13 17:22:02,092 - Epoch: [6][ 640/ 1207] Overall Loss 0.555547 Objective Loss 0.555547 LR 0.001000 Time 0.014563 -2023-02-13 17:22:02,225 - Epoch: [6][ 650/ 1207] Overall Loss 0.555773 Objective Loss 0.555773 LR 0.001000 Time 0.014541 -2023-02-13 17:22:02,356 - Epoch: [6][ 660/ 1207] Overall Loss 0.556416 Objective Loss 0.556416 LR 0.001000 Time 0.014519 -2023-02-13 17:22:02,488 - Epoch: [6][ 670/ 1207] Overall Loss 0.556364 Objective Loss 0.556364 LR 0.001000 Time 0.014499 -2023-02-13 17:22:02,620 - Epoch: [6][ 680/ 1207] Overall Loss 0.556181 Objective Loss 0.556181 LR 0.001000 Time 0.014477 -2023-02-13 17:22:02,757 - Epoch: [6][ 690/ 1207] Overall Loss 0.555951 Objective Loss 0.555951 LR 0.001000 Time 0.014465 -2023-02-13 17:22:02,890 - Epoch: [6][ 700/ 1207] Overall Loss 0.555618 Objective Loss 0.555618 LR 0.001000 Time 0.014448 -2023-02-13 17:22:03,023 - Epoch: [6][ 710/ 1207] Overall Loss 0.555464 Objective Loss 0.555464 LR 0.001000 Time 0.014429 -2023-02-13 17:22:03,153 - Epoch: [6][ 720/ 1207] Overall Loss 0.556076 Objective Loss 0.556076 LR 0.001000 Time 0.014410 -2023-02-13 17:22:03,285 - Epoch: [6][ 730/ 1207] Overall Loss 0.555657 Objective Loss 0.555657 LR 0.001000 Time 0.014392 -2023-02-13 17:22:03,417 - Epoch: [6][ 740/ 1207] Overall Loss 0.555007 Objective Loss 0.555007 LR 0.001000 Time 0.014373 -2023-02-13 17:22:03,549 - Epoch: [6][ 750/ 1207] Overall Loss 0.555010 Objective Loss 0.555010 LR 0.001000 Time 0.014358 -2023-02-13 17:22:03,680 - Epoch: [6][ 760/ 1207] Overall Loss 0.555255 Objective Loss 0.555255 LR 0.001000 Time 0.014341 -2023-02-13 17:22:03,813 - Epoch: [6][ 770/ 1207] Overall Loss 0.555310 Objective Loss 0.555310 LR 0.001000 Time 0.014326 -2023-02-13 17:22:03,944 - Epoch: [6][ 780/ 1207] Overall Loss 0.555121 Objective Loss 0.555121 LR 0.001000 Time 0.014310 -2023-02-13 17:22:04,076 - Epoch: [6][ 790/ 1207] Overall Loss 0.555418 Objective Loss 0.555418 LR 0.001000 Time 0.014296 -2023-02-13 17:22:04,207 - Epoch: [6][ 800/ 1207] Overall Loss 0.555072 Objective Loss 0.555072 LR 0.001000 Time 0.014281 -2023-02-13 17:22:04,340 - Epoch: [6][ 810/ 1207] Overall Loss 0.554959 Objective Loss 0.554959 LR 0.001000 Time 0.014268 -2023-02-13 17:22:04,472 - Epoch: [6][ 820/ 1207] Overall Loss 0.554707 Objective Loss 0.554707 LR 0.001000 Time 0.014254 -2023-02-13 17:22:04,604 - Epoch: [6][ 830/ 1207] Overall Loss 0.554558 Objective Loss 0.554558 LR 0.001000 Time 0.014240 -2023-02-13 17:22:04,736 - Epoch: [6][ 840/ 1207] Overall Loss 0.554252 Objective Loss 0.554252 LR 0.001000 Time 0.014226 -2023-02-13 17:22:04,868 - Epoch: [6][ 850/ 1207] Overall Loss 0.553759 Objective Loss 0.553759 LR 0.001000 Time 0.014214 -2023-02-13 17:22:05,000 - Epoch: [6][ 860/ 1207] Overall Loss 0.553723 Objective Loss 0.553723 LR 0.001000 Time 0.014201 -2023-02-13 17:22:05,132 - Epoch: [6][ 870/ 1207] Overall Loss 0.553685 Objective Loss 0.553685 LR 0.001000 Time 0.014190 -2023-02-13 17:22:05,263 - Epoch: [6][ 880/ 1207] Overall Loss 0.553763 Objective Loss 0.553763 LR 0.001000 Time 0.014177 -2023-02-13 17:22:05,395 - Epoch: [6][ 890/ 1207] Overall Loss 0.553494 Objective Loss 0.553494 LR 0.001000 Time 0.014166 -2023-02-13 17:22:05,527 - Epoch: [6][ 900/ 1207] Overall Loss 0.553007 Objective Loss 0.553007 LR 0.001000 Time 0.014154 -2023-02-13 17:22:05,659 - Epoch: [6][ 910/ 1207] Overall Loss 0.552955 Objective Loss 0.552955 LR 0.001000 Time 0.014144 -2023-02-13 17:22:05,793 - Epoch: [6][ 920/ 1207] Overall Loss 0.553279 Objective Loss 0.553279 LR 0.001000 Time 0.014135 -2023-02-13 17:22:05,927 - Epoch: [6][ 930/ 1207] Overall Loss 0.553019 Objective Loss 0.553019 LR 0.001000 Time 0.014127 -2023-02-13 17:22:06,059 - Epoch: [6][ 940/ 1207] Overall Loss 0.553478 Objective Loss 0.553478 LR 0.001000 Time 0.014117 -2023-02-13 17:22:06,193 - Epoch: [6][ 950/ 1207] Overall Loss 0.553316 Objective Loss 0.553316 LR 0.001000 Time 0.014108 -2023-02-13 17:22:06,324 - Epoch: [6][ 960/ 1207] Overall Loss 0.553418 Objective Loss 0.553418 LR 0.001000 Time 0.014097 -2023-02-13 17:22:06,459 - Epoch: [6][ 970/ 1207] Overall Loss 0.553739 Objective Loss 0.553739 LR 0.001000 Time 0.014091 -2023-02-13 17:22:06,592 - Epoch: [6][ 980/ 1207] Overall Loss 0.553608 Objective Loss 0.553608 LR 0.001000 Time 0.014083 -2023-02-13 17:22:06,725 - Epoch: [6][ 990/ 1207] Overall Loss 0.553363 Objective Loss 0.553363 LR 0.001000 Time 0.014074 -2023-02-13 17:22:06,859 - Epoch: [6][ 1000/ 1207] Overall Loss 0.553336 Objective Loss 0.553336 LR 0.001000 Time 0.014067 -2023-02-13 17:22:06,992 - Epoch: [6][ 1010/ 1207] Overall Loss 0.553644 Objective Loss 0.553644 LR 0.001000 Time 0.014059 -2023-02-13 17:22:07,126 - Epoch: [6][ 1020/ 1207] Overall Loss 0.553698 Objective Loss 0.553698 LR 0.001000 Time 0.014052 -2023-02-13 17:22:07,260 - Epoch: [6][ 1030/ 1207] Overall Loss 0.553431 Objective Loss 0.553431 LR 0.001000 Time 0.014044 -2023-02-13 17:22:07,392 - Epoch: [6][ 1040/ 1207] Overall Loss 0.552713 Objective Loss 0.552713 LR 0.001000 Time 0.014036 -2023-02-13 17:22:07,524 - Epoch: [6][ 1050/ 1207] Overall Loss 0.552855 Objective Loss 0.552855 LR 0.001000 Time 0.014028 -2023-02-13 17:22:07,659 - Epoch: [6][ 1060/ 1207] Overall Loss 0.552856 Objective Loss 0.552856 LR 0.001000 Time 0.014022 -2023-02-13 17:22:07,793 - Epoch: [6][ 1070/ 1207] Overall Loss 0.552922 Objective Loss 0.552922 LR 0.001000 Time 0.014016 -2023-02-13 17:22:07,924 - Epoch: [6][ 1080/ 1207] Overall Loss 0.553283 Objective Loss 0.553283 LR 0.001000 Time 0.014007 -2023-02-13 17:22:08,059 - Epoch: [6][ 1090/ 1207] Overall Loss 0.553229 Objective Loss 0.553229 LR 0.001000 Time 0.014002 -2023-02-13 17:22:08,192 - Epoch: [6][ 1100/ 1207] Overall Loss 0.553266 Objective Loss 0.553266 LR 0.001000 Time 0.013995 -2023-02-13 17:22:08,326 - Epoch: [6][ 1110/ 1207] Overall Loss 0.553278 Objective Loss 0.553278 LR 0.001000 Time 0.013989 -2023-02-13 17:22:08,458 - Epoch: [6][ 1120/ 1207] Overall Loss 0.553421 Objective Loss 0.553421 LR 0.001000 Time 0.013982 -2023-02-13 17:22:08,592 - Epoch: [6][ 1130/ 1207] Overall Loss 0.553102 Objective Loss 0.553102 LR 0.001000 Time 0.013977 -2023-02-13 17:22:08,725 - Epoch: [6][ 1140/ 1207] Overall Loss 0.552910 Objective Loss 0.552910 LR 0.001000 Time 0.013970 -2023-02-13 17:22:08,857 - Epoch: [6][ 1150/ 1207] Overall Loss 0.552871 Objective Loss 0.552871 LR 0.001000 Time 0.013963 -2023-02-13 17:22:08,991 - Epoch: [6][ 1160/ 1207] Overall Loss 0.552566 Objective Loss 0.552566 LR 0.001000 Time 0.013958 -2023-02-13 17:22:09,124 - Epoch: [6][ 1170/ 1207] Overall Loss 0.552524 Objective Loss 0.552524 LR 0.001000 Time 0.013952 -2023-02-13 17:22:09,258 - Epoch: [6][ 1180/ 1207] Overall Loss 0.552450 Objective Loss 0.552450 LR 0.001000 Time 0.013947 -2023-02-13 17:22:09,392 - Epoch: [6][ 1190/ 1207] Overall Loss 0.552644 Objective Loss 0.552644 LR 0.001000 Time 0.013942 -2023-02-13 17:22:09,576 - Epoch: [6][ 1200/ 1207] Overall Loss 0.552187 Objective Loss 0.552187 LR 0.001000 Time 0.013979 -2023-02-13 17:22:09,662 - Epoch: [6][ 1207/ 1207] Overall Loss 0.551913 Objective Loss 0.551913 Top1 76.829268 Top5 96.951220 LR 0.001000 Time 0.013969 -2023-02-13 17:22:09,732 - --- validate (epoch=6)----------- -2023-02-13 17:22:09,733 - 34311 samples (256 per mini-batch) -2023-02-13 17:22:10,088 - Epoch: [6][ 10/ 135] Loss 0.544307 Top1 72.226562 Top5 96.132812 -2023-02-13 17:22:10,188 - Epoch: [6][ 20/ 135] Loss 0.535147 Top1 72.109375 Top5 95.781250 -2023-02-13 17:22:10,281 - Epoch: [6][ 30/ 135] Loss 0.528230 Top1 72.343750 Top5 95.546875 -2023-02-13 17:22:10,375 - Epoch: [6][ 40/ 135] Loss 0.542757 Top1 72.431641 Top5 95.546875 -2023-02-13 17:22:10,472 - Epoch: [6][ 50/ 135] Loss 0.541903 Top1 72.453125 Top5 95.640625 -2023-02-13 17:22:10,563 - Epoch: [6][ 60/ 135] Loss 0.544845 Top1 72.220052 Top5 95.605469 -2023-02-13 17:22:10,660 - Epoch: [6][ 70/ 135] Loss 0.540999 Top1 72.343750 Top5 95.513393 -2023-02-13 17:22:10,753 - Epoch: [6][ 80/ 135] Loss 0.543868 Top1 72.294922 Top5 95.458984 -2023-02-13 17:22:10,843 - Epoch: [6][ 90/ 135] Loss 0.545237 Top1 72.256944 Top5 95.494792 -2023-02-13 17:22:10,932 - Epoch: [6][ 100/ 135] Loss 0.549046 Top1 72.074219 Top5 95.476562 -2023-02-13 17:22:11,021 - Epoch: [6][ 110/ 135] Loss 0.544081 Top1 72.244318 Top5 95.493608 -2023-02-13 17:22:11,112 - Epoch: [6][ 120/ 135] Loss 0.545793 Top1 72.125651 Top5 95.546875 -2023-02-13 17:22:11,206 - Epoch: [6][ 130/ 135] Loss 0.545782 Top1 72.175481 Top5 95.600962 -2023-02-13 17:22:11,232 - Epoch: [6][ 135/ 135] Loss 0.544316 Top1 72.151788 Top5 95.587421 -2023-02-13 17:22:11,302 - ==> Top1: 72.152 Top5: 95.587 Loss: 0.544 - -2023-02-13 17:22:11,303 - ==> Confusion: -[[ 726 2 13 0 29 1 0 2 16 137 0 5 1 7 7 9 3 0 2 0 7] - [ 2 924 2 1 18 28 3 24 8 2 4 2 1 0 2 0 2 0 4 3 3] - [ 10 9 872 4 10 2 66 31 1 1 11 1 0 7 1 5 2 3 8 7 7] - [ 7 4 59 767 4 6 6 7 4 5 51 0 8 2 36 5 7 2 19 1 16] - [ 8 19 4 1 987 1 0 0 3 8 3 3 1 3 3 6 5 1 2 2 6] - [ 3 105 4 2 17 780 10 67 2 7 4 14 3 28 5 0 4 1 3 7 4] - [ 1 5 18 2 5 3 1023 11 1 2 2 1 0 1 0 4 1 2 1 15 1] - [ 1 24 12 1 1 23 13 894 2 2 6 3 1 3 1 0 1 1 22 8 5] - [ 13 11 0 1 6 0 0 0 856 59 16 3 0 17 16 1 2 1 6 0 1] - [ 84 3 5 0 11 1 2 4 33 831 0 0 0 24 4 2 1 1 1 1 4] - [ 3 13 6 5 2 0 7 8 23 3 938 3 0 17 2 0 0 0 16 1 4] - [ 6 10 1 0 7 17 3 8 2 2 1 806 51 16 1 17 6 4 1 42 4] - [ 2 2 1 11 4 4 3 1 5 0 1 58 785 3 7 16 6 32 4 7 7] - [ 9 4 0 0 14 18 1 3 16 35 9 13 1 875 6 3 7 0 1 8 1] - [ 20 9 1 17 25 1 0 1 50 10 14 4 3 3 895 2 7 2 14 1 13] - [ 4 5 7 0 7 1 24 0 0 0 1 8 5 5 0 935 19 6 0 9 10] - [ 3 17 4 0 21 3 1 0 3 1 3 2 0 1 3 12 964 2 2 8 11] - [ 11 2 4 11 1 2 1 2 7 1 2 24 53 7 4 93 2 811 1 5 7] - [ 4 21 4 14 2 0 0 61 11 1 23 2 3 3 16 0 0 1 916 3 1] - [ 0 8 2 0 1 6 20 23 0 0 3 10 4 11 1 3 1 0 1 1050 4] - [ 239 690 288 75 473 270 224 339 185 226 348 164 376 503 191 232 487 48 259 696 7121]] - -2023-02-13 17:22:11,304 - ==> Best [Top1: 72.152 Top5: 95.587 Sparsity:0.00 Params: 148928 on epoch: 6] -2023-02-13 17:22:11,304 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:22:11,311 - - -2023-02-13 17:22:11,311 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:22:12,232 - Epoch: [7][ 10/ 1207] Overall Loss 0.535331 Objective Loss 0.535331 LR 0.001000 Time 0.092100 -2023-02-13 17:22:12,394 - Epoch: [7][ 20/ 1207] Overall Loss 0.529143 Objective Loss 0.529143 LR 0.001000 Time 0.054130 -2023-02-13 17:22:12,548 - Epoch: [7][ 30/ 1207] Overall Loss 0.525494 Objective Loss 0.525494 LR 0.001000 Time 0.041177 -2023-02-13 17:22:12,700 - Epoch: [7][ 40/ 1207] Overall Loss 0.521831 Objective Loss 0.521831 LR 0.001000 Time 0.034689 -2023-02-13 17:22:12,850 - Epoch: [7][ 50/ 1207] Overall Loss 0.514802 Objective Loss 0.514802 LR 0.001000 Time 0.030747 -2023-02-13 17:22:12,993 - Epoch: [7][ 60/ 1207] Overall Loss 0.513798 Objective Loss 0.513798 LR 0.001000 Time 0.027995 -2023-02-13 17:22:13,139 - Epoch: [7][ 70/ 1207] Overall Loss 0.513610 Objective Loss 0.513610 LR 0.001000 Time 0.026077 -2023-02-13 17:22:13,283 - Epoch: [7][ 80/ 1207] Overall Loss 0.518613 Objective Loss 0.518613 LR 0.001000 Time 0.024608 -2023-02-13 17:22:13,429 - Epoch: [7][ 90/ 1207] Overall Loss 0.518612 Objective Loss 0.518612 LR 0.001000 Time 0.023489 -2023-02-13 17:22:13,573 - Epoch: [7][ 100/ 1207] Overall Loss 0.521936 Objective Loss 0.521936 LR 0.001000 Time 0.022577 -2023-02-13 17:22:13,719 - Epoch: [7][ 110/ 1207] Overall Loss 0.523860 Objective Loss 0.523860 LR 0.001000 Time 0.021851 -2023-02-13 17:22:13,863 - Epoch: [7][ 120/ 1207] Overall Loss 0.524981 Objective Loss 0.524981 LR 0.001000 Time 0.021228 -2023-02-13 17:22:14,009 - Epoch: [7][ 130/ 1207] Overall Loss 0.527666 Objective Loss 0.527666 LR 0.001000 Time 0.020717 -2023-02-13 17:22:14,152 - Epoch: [7][ 140/ 1207] Overall Loss 0.526468 Objective Loss 0.526468 LR 0.001000 Time 0.020257 -2023-02-13 17:22:14,298 - Epoch: [7][ 150/ 1207] Overall Loss 0.522408 Objective Loss 0.522408 LR 0.001000 Time 0.019875 -2023-02-13 17:22:14,442 - Epoch: [7][ 160/ 1207] Overall Loss 0.522877 Objective Loss 0.522877 LR 0.001000 Time 0.019528 -2023-02-13 17:22:14,588 - Epoch: [7][ 170/ 1207] Overall Loss 0.522347 Objective Loss 0.522347 LR 0.001000 Time 0.019239 -2023-02-13 17:22:14,732 - Epoch: [7][ 180/ 1207] Overall Loss 0.520629 Objective Loss 0.520629 LR 0.001000 Time 0.018969 -2023-02-13 17:22:14,880 - Epoch: [7][ 190/ 1207] Overall Loss 0.518497 Objective Loss 0.518497 LR 0.001000 Time 0.018743 -2023-02-13 17:22:15,023 - Epoch: [7][ 200/ 1207] Overall Loss 0.517840 Objective Loss 0.517840 LR 0.001000 Time 0.018523 -2023-02-13 17:22:15,170 - Epoch: [7][ 210/ 1207] Overall Loss 0.519162 Objective Loss 0.519162 LR 0.001000 Time 0.018338 -2023-02-13 17:22:15,314 - Epoch: [7][ 220/ 1207] Overall Loss 0.520235 Objective Loss 0.520235 LR 0.001000 Time 0.018156 -2023-02-13 17:22:15,460 - Epoch: [7][ 230/ 1207] Overall Loss 0.519121 Objective Loss 0.519121 LR 0.001000 Time 0.018001 -2023-02-13 17:22:15,604 - Epoch: [7][ 240/ 1207] Overall Loss 0.518745 Objective Loss 0.518745 LR 0.001000 Time 0.017849 -2023-02-13 17:22:15,751 - Epoch: [7][ 250/ 1207] Overall Loss 0.519112 Objective Loss 0.519112 LR 0.001000 Time 0.017721 -2023-02-13 17:22:15,895 - Epoch: [7][ 260/ 1207] Overall Loss 0.517597 Objective Loss 0.517597 LR 0.001000 Time 0.017591 -2023-02-13 17:22:16,041 - Epoch: [7][ 270/ 1207] Overall Loss 0.519236 Objective Loss 0.519236 LR 0.001000 Time 0.017479 -2023-02-13 17:22:16,184 - Epoch: [7][ 280/ 1207] Overall Loss 0.520657 Objective Loss 0.520657 LR 0.001000 Time 0.017364 -2023-02-13 17:22:16,330 - Epoch: [7][ 290/ 1207] Overall Loss 0.521106 Objective Loss 0.521106 LR 0.001000 Time 0.017269 -2023-02-13 17:22:16,475 - Epoch: [7][ 300/ 1207] Overall Loss 0.519938 Objective Loss 0.519938 LR 0.001000 Time 0.017173 -2023-02-13 17:22:16,621 - Epoch: [7][ 310/ 1207] Overall Loss 0.520747 Objective Loss 0.520747 LR 0.001000 Time 0.017090 -2023-02-13 17:22:16,765 - Epoch: [7][ 320/ 1207] Overall Loss 0.521078 Objective Loss 0.521078 LR 0.001000 Time 0.017004 -2023-02-13 17:22:16,911 - Epoch: [7][ 330/ 1207] Overall Loss 0.521831 Objective Loss 0.521831 LR 0.001000 Time 0.016932 -2023-02-13 17:22:17,054 - Epoch: [7][ 340/ 1207] Overall Loss 0.520470 Objective Loss 0.520470 LR 0.001000 Time 0.016854 -2023-02-13 17:22:17,201 - Epoch: [7][ 350/ 1207] Overall Loss 0.520943 Objective Loss 0.520943 LR 0.001000 Time 0.016789 -2023-02-13 17:22:17,344 - Epoch: [7][ 360/ 1207] Overall Loss 0.521355 Objective Loss 0.521355 LR 0.001000 Time 0.016720 -2023-02-13 17:22:17,490 - Epoch: [7][ 370/ 1207] Overall Loss 0.521387 Objective Loss 0.521387 LR 0.001000 Time 0.016662 -2023-02-13 17:22:17,634 - Epoch: [7][ 380/ 1207] Overall Loss 0.521794 Objective Loss 0.521794 LR 0.001000 Time 0.016603 -2023-02-13 17:22:17,781 - Epoch: [7][ 390/ 1207] Overall Loss 0.521265 Objective Loss 0.521265 LR 0.001000 Time 0.016552 -2023-02-13 17:22:17,924 - Epoch: [7][ 400/ 1207] Overall Loss 0.520583 Objective Loss 0.520583 LR 0.001000 Time 0.016494 -2023-02-13 17:22:18,070 - Epoch: [7][ 410/ 1207] Overall Loss 0.521047 Objective Loss 0.521047 LR 0.001000 Time 0.016448 -2023-02-13 17:22:18,215 - Epoch: [7][ 420/ 1207] Overall Loss 0.521524 Objective Loss 0.521524 LR 0.001000 Time 0.016399 -2023-02-13 17:22:18,361 - Epoch: [7][ 430/ 1207] Overall Loss 0.521304 Objective Loss 0.521304 LR 0.001000 Time 0.016358 -2023-02-13 17:22:18,505 - Epoch: [7][ 440/ 1207] Overall Loss 0.520598 Objective Loss 0.520598 LR 0.001000 Time 0.016313 -2023-02-13 17:22:18,652 - Epoch: [7][ 450/ 1207] Overall Loss 0.520962 Objective Loss 0.520962 LR 0.001000 Time 0.016276 -2023-02-13 17:22:18,797 - Epoch: [7][ 460/ 1207] Overall Loss 0.521461 Objective Loss 0.521461 LR 0.001000 Time 0.016236 -2023-02-13 17:22:18,944 - Epoch: [7][ 470/ 1207] Overall Loss 0.521600 Objective Loss 0.521600 LR 0.001000 Time 0.016201 -2023-02-13 17:22:19,088 - Epoch: [7][ 480/ 1207] Overall Loss 0.522890 Objective Loss 0.522890 LR 0.001000 Time 0.016164 -2023-02-13 17:22:19,234 - Epoch: [7][ 490/ 1207] Overall Loss 0.523274 Objective Loss 0.523274 LR 0.001000 Time 0.016131 -2023-02-13 17:22:19,378 - Epoch: [7][ 500/ 1207] Overall Loss 0.523180 Objective Loss 0.523180 LR 0.001000 Time 0.016096 -2023-02-13 17:22:19,524 - Epoch: [7][ 510/ 1207] Overall Loss 0.523405 Objective Loss 0.523405 LR 0.001000 Time 0.016066 -2023-02-13 17:22:19,670 - Epoch: [7][ 520/ 1207] Overall Loss 0.522773 Objective Loss 0.522773 LR 0.001000 Time 0.016037 -2023-02-13 17:22:19,817 - Epoch: [7][ 530/ 1207] Overall Loss 0.522193 Objective Loss 0.522193 LR 0.001000 Time 0.016012 -2023-02-13 17:22:19,961 - Epoch: [7][ 540/ 1207] Overall Loss 0.521678 Objective Loss 0.521678 LR 0.001000 Time 0.015981 -2023-02-13 17:22:20,108 - Epoch: [7][ 550/ 1207] Overall Loss 0.521360 Objective Loss 0.521360 LR 0.001000 Time 0.015956 -2023-02-13 17:22:20,252 - Epoch: [7][ 560/ 1207] Overall Loss 0.521385 Objective Loss 0.521385 LR 0.001000 Time 0.015927 -2023-02-13 17:22:20,398 - Epoch: [7][ 570/ 1207] Overall Loss 0.521322 Objective Loss 0.521322 LR 0.001000 Time 0.015905 -2023-02-13 17:22:20,543 - Epoch: [7][ 580/ 1207] Overall Loss 0.520432 Objective Loss 0.520432 LR 0.001000 Time 0.015878 -2023-02-13 17:22:20,690 - Epoch: [7][ 590/ 1207] Overall Loss 0.520195 Objective Loss 0.520195 LR 0.001000 Time 0.015858 -2023-02-13 17:22:20,834 - Epoch: [7][ 600/ 1207] Overall Loss 0.520443 Objective Loss 0.520443 LR 0.001000 Time 0.015834 -2023-02-13 17:22:20,981 - Epoch: [7][ 610/ 1207] Overall Loss 0.520561 Objective Loss 0.520561 LR 0.001000 Time 0.015814 -2023-02-13 17:22:21,124 - Epoch: [7][ 620/ 1207] Overall Loss 0.520065 Objective Loss 0.520065 LR 0.001000 Time 0.015790 -2023-02-13 17:22:21,271 - Epoch: [7][ 630/ 1207] Overall Loss 0.519700 Objective Loss 0.519700 LR 0.001000 Time 0.015772 -2023-02-13 17:22:21,415 - Epoch: [7][ 640/ 1207] Overall Loss 0.519629 Objective Loss 0.519629 LR 0.001000 Time 0.015749 -2023-02-13 17:22:21,557 - Epoch: [7][ 650/ 1207] Overall Loss 0.519655 Objective Loss 0.519655 LR 0.001000 Time 0.015725 -2023-02-13 17:22:21,690 - Epoch: [7][ 660/ 1207] Overall Loss 0.519105 Objective Loss 0.519105 LR 0.001000 Time 0.015688 -2023-02-13 17:22:21,824 - Epoch: [7][ 670/ 1207] Overall Loss 0.518920 Objective Loss 0.518920 LR 0.001000 Time 0.015653 -2023-02-13 17:22:21,956 - Epoch: [7][ 680/ 1207] Overall Loss 0.518946 Objective Loss 0.518946 LR 0.001000 Time 0.015617 -2023-02-13 17:22:22,090 - Epoch: [7][ 690/ 1207] Overall Loss 0.518750 Objective Loss 0.518750 LR 0.001000 Time 0.015582 -2023-02-13 17:22:22,222 - Epoch: [7][ 700/ 1207] Overall Loss 0.518024 Objective Loss 0.518024 LR 0.001000 Time 0.015546 -2023-02-13 17:22:22,357 - Epoch: [7][ 710/ 1207] Overall Loss 0.518018 Objective Loss 0.518018 LR 0.001000 Time 0.015514 -2023-02-13 17:22:22,488 - Epoch: [7][ 720/ 1207] Overall Loss 0.517758 Objective Loss 0.517758 LR 0.001000 Time 0.015481 -2023-02-13 17:22:22,621 - Epoch: [7][ 730/ 1207] Overall Loss 0.517229 Objective Loss 0.517229 LR 0.001000 Time 0.015450 -2023-02-13 17:22:22,754 - Epoch: [7][ 740/ 1207] Overall Loss 0.517100 Objective Loss 0.517100 LR 0.001000 Time 0.015419 -2023-02-13 17:22:22,887 - Epoch: [7][ 750/ 1207] Overall Loss 0.517058 Objective Loss 0.517058 LR 0.001000 Time 0.015389 -2023-02-13 17:22:23,019 - Epoch: [7][ 760/ 1207] Overall Loss 0.516896 Objective Loss 0.516896 LR 0.001000 Time 0.015359 -2023-02-13 17:22:23,152 - Epoch: [7][ 770/ 1207] Overall Loss 0.516582 Objective Loss 0.516582 LR 0.001000 Time 0.015332 -2023-02-13 17:22:23,284 - Epoch: [7][ 780/ 1207] Overall Loss 0.516368 Objective Loss 0.516368 LR 0.001000 Time 0.015303 -2023-02-13 17:22:23,417 - Epoch: [7][ 790/ 1207] Overall Loss 0.515945 Objective Loss 0.515945 LR 0.001000 Time 0.015275 -2023-02-13 17:22:23,550 - Epoch: [7][ 800/ 1207] Overall Loss 0.516091 Objective Loss 0.516091 LR 0.001000 Time 0.015249 -2023-02-13 17:22:23,685 - Epoch: [7][ 810/ 1207] Overall Loss 0.516044 Objective Loss 0.516044 LR 0.001000 Time 0.015224 -2023-02-13 17:22:23,816 - Epoch: [7][ 820/ 1207] Overall Loss 0.516688 Objective Loss 0.516688 LR 0.001000 Time 0.015199 -2023-02-13 17:22:23,951 - Epoch: [7][ 830/ 1207] Overall Loss 0.516670 Objective Loss 0.516670 LR 0.001000 Time 0.015177 -2023-02-13 17:22:24,084 - Epoch: [7][ 840/ 1207] Overall Loss 0.516411 Objective Loss 0.516411 LR 0.001000 Time 0.015155 -2023-02-13 17:22:24,217 - Epoch: [7][ 850/ 1207] Overall Loss 0.516715 Objective Loss 0.516715 LR 0.001000 Time 0.015132 -2023-02-13 17:22:24,349 - Epoch: [7][ 860/ 1207] Overall Loss 0.516906 Objective Loss 0.516906 LR 0.001000 Time 0.015109 -2023-02-13 17:22:24,484 - Epoch: [7][ 870/ 1207] Overall Loss 0.517309 Objective Loss 0.517309 LR 0.001000 Time 0.015088 -2023-02-13 17:22:24,616 - Epoch: [7][ 880/ 1207] Overall Loss 0.517442 Objective Loss 0.517442 LR 0.001000 Time 0.015066 -2023-02-13 17:22:24,750 - Epoch: [7][ 890/ 1207] Overall Loss 0.517721 Objective Loss 0.517721 LR 0.001000 Time 0.015047 -2023-02-13 17:22:24,882 - Epoch: [7][ 900/ 1207] Overall Loss 0.517484 Objective Loss 0.517484 LR 0.001000 Time 0.015026 -2023-02-13 17:22:25,016 - Epoch: [7][ 910/ 1207] Overall Loss 0.517334 Objective Loss 0.517334 LR 0.001000 Time 0.015006 -2023-02-13 17:22:25,148 - Epoch: [7][ 920/ 1207] Overall Loss 0.517727 Objective Loss 0.517727 LR 0.001000 Time 0.014986 -2023-02-13 17:22:25,280 - Epoch: [7][ 930/ 1207] Overall Loss 0.517699 Objective Loss 0.517699 LR 0.001000 Time 0.014967 -2023-02-13 17:22:25,413 - Epoch: [7][ 940/ 1207] Overall Loss 0.517381 Objective Loss 0.517381 LR 0.001000 Time 0.014947 -2023-02-13 17:22:25,548 - Epoch: [7][ 950/ 1207] Overall Loss 0.517355 Objective Loss 0.517355 LR 0.001000 Time 0.014931 -2023-02-13 17:22:25,680 - Epoch: [7][ 960/ 1207] Overall Loss 0.517449 Objective Loss 0.517449 LR 0.001000 Time 0.014913 -2023-02-13 17:22:25,817 - Epoch: [7][ 970/ 1207] Overall Loss 0.517359 Objective Loss 0.517359 LR 0.001000 Time 0.014900 -2023-02-13 17:22:25,951 - Epoch: [7][ 980/ 1207] Overall Loss 0.517394 Objective Loss 0.517394 LR 0.001000 Time 0.014885 -2023-02-13 17:22:26,086 - Epoch: [7][ 990/ 1207] Overall Loss 0.517508 Objective Loss 0.517508 LR 0.001000 Time 0.014871 -2023-02-13 17:22:26,220 - Epoch: [7][ 1000/ 1207] Overall Loss 0.517870 Objective Loss 0.517870 LR 0.001000 Time 0.014855 -2023-02-13 17:22:26,355 - Epoch: [7][ 1010/ 1207] Overall Loss 0.517972 Objective Loss 0.517972 LR 0.001000 Time 0.014841 -2023-02-13 17:22:26,488 - Epoch: [7][ 1020/ 1207] Overall Loss 0.517781 Objective Loss 0.517781 LR 0.001000 Time 0.014826 -2023-02-13 17:22:26,624 - Epoch: [7][ 1030/ 1207] Overall Loss 0.517976 Objective Loss 0.517976 LR 0.001000 Time 0.014813 -2023-02-13 17:22:26,757 - Epoch: [7][ 1040/ 1207] Overall Loss 0.517630 Objective Loss 0.517630 LR 0.001000 Time 0.014799 -2023-02-13 17:22:26,893 - Epoch: [7][ 1050/ 1207] Overall Loss 0.517922 Objective Loss 0.517922 LR 0.001000 Time 0.014785 -2023-02-13 17:22:27,026 - Epoch: [7][ 1060/ 1207] Overall Loss 0.518345 Objective Loss 0.518345 LR 0.001000 Time 0.014770 -2023-02-13 17:22:27,163 - Epoch: [7][ 1070/ 1207] Overall Loss 0.518206 Objective Loss 0.518206 LR 0.001000 Time 0.014759 -2023-02-13 17:22:27,296 - Epoch: [7][ 1080/ 1207] Overall Loss 0.517974 Objective Loss 0.517974 LR 0.001000 Time 0.014744 -2023-02-13 17:22:27,431 - Epoch: [7][ 1090/ 1207] Overall Loss 0.517836 Objective Loss 0.517836 LR 0.001000 Time 0.014733 -2023-02-13 17:22:27,565 - Epoch: [7][ 1100/ 1207] Overall Loss 0.517699 Objective Loss 0.517699 LR 0.001000 Time 0.014720 -2023-02-13 17:22:27,699 - Epoch: [7][ 1110/ 1207] Overall Loss 0.517767 Objective Loss 0.517767 LR 0.001000 Time 0.014708 -2023-02-13 17:22:27,833 - Epoch: [7][ 1120/ 1207] Overall Loss 0.517678 Objective Loss 0.517678 LR 0.001000 Time 0.014696 -2023-02-13 17:22:27,967 - Epoch: [7][ 1130/ 1207] Overall Loss 0.518133 Objective Loss 0.518133 LR 0.001000 Time 0.014684 -2023-02-13 17:22:28,102 - Epoch: [7][ 1140/ 1207] Overall Loss 0.518176 Objective Loss 0.518176 LR 0.001000 Time 0.014673 -2023-02-13 17:22:28,235 - Epoch: [7][ 1150/ 1207] Overall Loss 0.517973 Objective Loss 0.517973 LR 0.001000 Time 0.014661 -2023-02-13 17:22:28,366 - Epoch: [7][ 1160/ 1207] Overall Loss 0.517971 Objective Loss 0.517971 LR 0.001000 Time 0.014647 -2023-02-13 17:22:28,499 - Epoch: [7][ 1170/ 1207] Overall Loss 0.518038 Objective Loss 0.518038 LR 0.001000 Time 0.014634 -2023-02-13 17:22:28,632 - Epoch: [7][ 1180/ 1207] Overall Loss 0.518078 Objective Loss 0.518078 LR 0.001000 Time 0.014622 -2023-02-13 17:22:28,766 - Epoch: [7][ 1190/ 1207] Overall Loss 0.518154 Objective Loss 0.518154 LR 0.001000 Time 0.014610 -2023-02-13 17:22:28,951 - Epoch: [7][ 1200/ 1207] Overall Loss 0.517882 Objective Loss 0.517882 LR 0.001000 Time 0.014642 -2023-02-13 17:22:29,037 - Epoch: [7][ 1207/ 1207] Overall Loss 0.517940 Objective Loss 0.517940 Top1 74.695122 Top5 97.865854 LR 0.001000 Time 0.014628 -2023-02-13 17:22:29,108 - --- validate (epoch=7)----------- -2023-02-13 17:22:29,108 - 34311 samples (256 per mini-batch) -2023-02-13 17:22:29,464 - Epoch: [7][ 10/ 135] Loss 0.498146 Top1 72.656250 Top5 96.210938 -2023-02-13 17:22:29,553 - Epoch: [7][ 20/ 135] Loss 0.518399 Top1 72.714844 Top5 95.937500 -2023-02-13 17:22:29,637 - Epoch: [7][ 30/ 135] Loss 0.499665 Top1 72.656250 Top5 95.872396 -2023-02-13 17:22:29,722 - Epoch: [7][ 40/ 135] Loss 0.499196 Top1 72.734375 Top5 95.888672 -2023-02-13 17:22:29,806 - Epoch: [7][ 50/ 135] Loss 0.504362 Top1 72.367188 Top5 95.687500 -2023-02-13 17:22:29,891 - Epoch: [7][ 60/ 135] Loss 0.499813 Top1 72.656250 Top5 95.774740 -2023-02-13 17:22:29,974 - Epoch: [7][ 70/ 135] Loss 0.493964 Top1 72.896205 Top5 95.842634 -2023-02-13 17:22:30,062 - Epoch: [7][ 80/ 135] Loss 0.492308 Top1 72.968750 Top5 95.795898 -2023-02-13 17:22:30,145 - Epoch: [7][ 90/ 135] Loss 0.492268 Top1 73.046875 Top5 95.655382 -2023-02-13 17:22:30,230 - Epoch: [7][ 100/ 135] Loss 0.499729 Top1 73.015625 Top5 95.601562 -2023-02-13 17:22:30,315 - Epoch: [7][ 110/ 135] Loss 0.499919 Top1 73.071733 Top5 95.539773 -2023-02-13 17:22:30,399 - Epoch: [7][ 120/ 135] Loss 0.502196 Top1 73.030599 Top5 95.504557 -2023-02-13 17:22:30,490 - Epoch: [7][ 130/ 135] Loss 0.501145 Top1 73.007812 Top5 95.522837 -2023-02-13 17:22:30,513 - Epoch: [7][ 135/ 135] Loss 0.496241 Top1 73.069861 Top5 95.520387 -2023-02-13 17:22:30,580 - ==> Top1: 73.070 Top5: 95.520 Loss: 0.496 - -2023-02-13 17:22:30,581 - ==> Confusion: -[[ 751 1 12 2 13 4 0 5 8 110 1 9 1 9 8 7 9 6 4 0 7] - [ 3 882 4 3 9 55 5 30 8 1 10 2 1 1 0 0 6 2 5 4 2] - [ 10 1 873 29 4 5 38 32 4 1 10 1 2 6 2 11 5 7 7 8 2] - [ 2 2 19 875 1 5 1 8 3 2 27 1 9 1 15 4 3 18 12 0 8] - [ 7 16 0 2 954 5 0 3 1 6 5 5 2 16 7 2 18 1 3 5 8] - [ 2 43 5 4 9 869 7 46 4 7 6 14 3 35 0 1 3 2 3 5 2] - [ 2 7 19 2 2 8 994 19 0 1 8 1 3 1 0 7 1 6 1 14 3] - [ 0 9 9 1 3 33 9 896 5 2 9 5 0 4 0 0 0 1 23 11 4] - [ 13 8 0 2 2 1 0 4 857 25 17 5 1 21 34 0 1 3 8 0 7] - [ 82 1 4 0 8 2 1 3 67 764 1 2 0 49 7 3 1 6 1 1 9] - [ 1 4 7 10 1 2 2 6 15 3 962 2 1 11 2 0 1 0 12 0 9] - [ 4 4 0 1 5 26 1 9 1 4 0 829 45 22 0 7 5 22 1 18 1] - [ 0 1 1 4 1 8 0 2 4 2 0 63 810 1 1 6 4 45 1 4 1] - [ 10 4 0 1 6 12 1 7 17 6 10 12 0 920 3 3 5 1 1 5 0] - [ 12 9 2 27 21 2 0 3 22 5 3 3 11 5 922 2 8 15 14 0 6] - [ 9 4 5 3 7 4 8 2 0 1 2 8 8 8 0 923 10 29 1 7 7] - [ 4 11 2 2 10 7 1 1 4 2 1 1 0 8 1 18 968 5 2 6 7] - [ 5 1 1 6 0 1 1 2 2 0 0 13 30 3 0 13 0 963 1 4 5] - [ 2 16 7 21 0 3 0 54 3 0 12 2 4 1 19 0 0 2 938 0 2] - [ 0 1 0 1 0 12 12 30 0 0 4 21 4 17 0 7 3 1 2 1031 2] - [ 212 500 266 249 282 395 154 336 143 201 282 207 499 572 195 222 487 201 325 616 7090]] - -2023-02-13 17:22:30,582 - ==> Best [Top1: 73.070 Top5: 95.520 Sparsity:0.00 Params: 148928 on epoch: 7] -2023-02-13 17:22:30,582 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:22:30,589 - - -2023-02-13 17:22:30,589 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:22:31,539 - Epoch: [8][ 10/ 1207] Overall Loss 0.519428 Objective Loss 0.519428 LR 0.001000 Time 0.094917 -2023-02-13 17:22:31,694 - Epoch: [8][ 20/ 1207] Overall Loss 0.520584 Objective Loss 0.520584 LR 0.001000 Time 0.055219 -2023-02-13 17:22:31,841 - Epoch: [8][ 30/ 1207] Overall Loss 0.516553 Objective Loss 0.516553 LR 0.001000 Time 0.041680 -2023-02-13 17:22:31,983 - Epoch: [8][ 40/ 1207] Overall Loss 0.512451 Objective Loss 0.512451 LR 0.001000 Time 0.034816 -2023-02-13 17:22:32,128 - Epoch: [8][ 50/ 1207] Overall Loss 0.517476 Objective Loss 0.517476 LR 0.001000 Time 0.030734 -2023-02-13 17:22:32,271 - Epoch: [8][ 60/ 1207] Overall Loss 0.517089 Objective Loss 0.517089 LR 0.001000 Time 0.027999 -2023-02-13 17:22:32,417 - Epoch: [8][ 70/ 1207] Overall Loss 0.510653 Objective Loss 0.510653 LR 0.001000 Time 0.026074 -2023-02-13 17:22:32,561 - Epoch: [8][ 80/ 1207] Overall Loss 0.503434 Objective Loss 0.503434 LR 0.001000 Time 0.024607 -2023-02-13 17:22:32,706 - Epoch: [8][ 90/ 1207] Overall Loss 0.502813 Objective Loss 0.502813 LR 0.001000 Time 0.023480 -2023-02-13 17:22:32,850 - Epoch: [8][ 100/ 1207] Overall Loss 0.500104 Objective Loss 0.500104 LR 0.001000 Time 0.022566 -2023-02-13 17:22:32,994 - Epoch: [8][ 110/ 1207] Overall Loss 0.499579 Objective Loss 0.499579 LR 0.001000 Time 0.021826 -2023-02-13 17:22:33,138 - Epoch: [8][ 120/ 1207] Overall Loss 0.500116 Objective Loss 0.500116 LR 0.001000 Time 0.021200 -2023-02-13 17:22:33,283 - Epoch: [8][ 130/ 1207] Overall Loss 0.498406 Objective Loss 0.498406 LR 0.001000 Time 0.020684 -2023-02-13 17:22:33,426 - Epoch: [8][ 140/ 1207] Overall Loss 0.498005 Objective Loss 0.498005 LR 0.001000 Time 0.020225 -2023-02-13 17:22:33,572 - Epoch: [8][ 150/ 1207] Overall Loss 0.495615 Objective Loss 0.495615 LR 0.001000 Time 0.019847 -2023-02-13 17:22:33,716 - Epoch: [8][ 160/ 1207] Overall Loss 0.495230 Objective Loss 0.495230 LR 0.001000 Time 0.019502 -2023-02-13 17:22:33,861 - Epoch: [8][ 170/ 1207] Overall Loss 0.495496 Objective Loss 0.495496 LR 0.001000 Time 0.019209 -2023-02-13 17:22:34,005 - Epoch: [8][ 180/ 1207] Overall Loss 0.494697 Objective Loss 0.494697 LR 0.001000 Time 0.018941 -2023-02-13 17:22:34,151 - Epoch: [8][ 190/ 1207] Overall Loss 0.495726 Objective Loss 0.495726 LR 0.001000 Time 0.018707 -2023-02-13 17:22:34,294 - Epoch: [8][ 200/ 1207] Overall Loss 0.496629 Objective Loss 0.496629 LR 0.001000 Time 0.018487 -2023-02-13 17:22:34,440 - Epoch: [8][ 210/ 1207] Overall Loss 0.494760 Objective Loss 0.494760 LR 0.001000 Time 0.018297 -2023-02-13 17:22:34,584 - Epoch: [8][ 220/ 1207] Overall Loss 0.494018 Objective Loss 0.494018 LR 0.001000 Time 0.018120 -2023-02-13 17:22:34,729 - Epoch: [8][ 230/ 1207] Overall Loss 0.494370 Objective Loss 0.494370 LR 0.001000 Time 0.017962 -2023-02-13 17:22:34,873 - Epoch: [8][ 240/ 1207] Overall Loss 0.494378 Objective Loss 0.494378 LR 0.001000 Time 0.017811 -2023-02-13 17:22:35,018 - Epoch: [8][ 250/ 1207] Overall Loss 0.493984 Objective Loss 0.493984 LR 0.001000 Time 0.017678 -2023-02-13 17:22:35,161 - Epoch: [8][ 260/ 1207] Overall Loss 0.494966 Objective Loss 0.494966 LR 0.001000 Time 0.017546 -2023-02-13 17:22:35,306 - Epoch: [8][ 270/ 1207] Overall Loss 0.495265 Objective Loss 0.495265 LR 0.001000 Time 0.017431 -2023-02-13 17:22:35,449 - Epoch: [8][ 280/ 1207] Overall Loss 0.494426 Objective Loss 0.494426 LR 0.001000 Time 0.017319 -2023-02-13 17:22:35,595 - Epoch: [8][ 290/ 1207] Overall Loss 0.494573 Objective Loss 0.494573 LR 0.001000 Time 0.017224 -2023-02-13 17:22:35,740 - Epoch: [8][ 300/ 1207] Overall Loss 0.496490 Objective Loss 0.496490 LR 0.001000 Time 0.017131 -2023-02-13 17:22:35,886 - Epoch: [8][ 310/ 1207] Overall Loss 0.497105 Objective Loss 0.497105 LR 0.001000 Time 0.017049 -2023-02-13 17:22:36,029 - Epoch: [8][ 320/ 1207] Overall Loss 0.497169 Objective Loss 0.497169 LR 0.001000 Time 0.016962 -2023-02-13 17:22:36,175 - Epoch: [8][ 330/ 1207] Overall Loss 0.497440 Objective Loss 0.497440 LR 0.001000 Time 0.016888 -2023-02-13 17:22:36,318 - Epoch: [8][ 340/ 1207] Overall Loss 0.496321 Objective Loss 0.496321 LR 0.001000 Time 0.016811 -2023-02-13 17:22:36,463 - Epoch: [8][ 350/ 1207] Overall Loss 0.496696 Objective Loss 0.496696 LR 0.001000 Time 0.016743 -2023-02-13 17:22:36,606 - Epoch: [8][ 360/ 1207] Overall Loss 0.494925 Objective Loss 0.494925 LR 0.001000 Time 0.016676 -2023-02-13 17:22:36,752 - Epoch: [8][ 370/ 1207] Overall Loss 0.495253 Objective Loss 0.495253 LR 0.001000 Time 0.016618 -2023-02-13 17:22:36,896 - Epoch: [8][ 380/ 1207] Overall Loss 0.495145 Objective Loss 0.495145 LR 0.001000 Time 0.016559 -2023-02-13 17:22:37,041 - Epoch: [8][ 390/ 1207] Overall Loss 0.495798 Objective Loss 0.495798 LR 0.001000 Time 0.016505 -2023-02-13 17:22:37,184 - Epoch: [8][ 400/ 1207] Overall Loss 0.495373 Objective Loss 0.495373 LR 0.001000 Time 0.016450 -2023-02-13 17:22:37,329 - Epoch: [8][ 410/ 1207] Overall Loss 0.496633 Objective Loss 0.496633 LR 0.001000 Time 0.016402 -2023-02-13 17:22:37,473 - Epoch: [8][ 420/ 1207] Overall Loss 0.497054 Objective Loss 0.497054 LR 0.001000 Time 0.016352 -2023-02-13 17:22:37,619 - Epoch: [8][ 430/ 1207] Overall Loss 0.496505 Objective Loss 0.496505 LR 0.001000 Time 0.016310 -2023-02-13 17:22:37,763 - Epoch: [8][ 440/ 1207] Overall Loss 0.495688 Objective Loss 0.495688 LR 0.001000 Time 0.016265 -2023-02-13 17:22:37,908 - Epoch: [8][ 450/ 1207] Overall Loss 0.495595 Objective Loss 0.495595 LR 0.001000 Time 0.016226 -2023-02-13 17:22:38,056 - Epoch: [8][ 460/ 1207] Overall Loss 0.496374 Objective Loss 0.496374 LR 0.001000 Time 0.016193 -2023-02-13 17:22:38,204 - Epoch: [8][ 470/ 1207] Overall Loss 0.496429 Objective Loss 0.496429 LR 0.001000 Time 0.016163 -2023-02-13 17:22:38,352 - Epoch: [8][ 480/ 1207] Overall Loss 0.496202 Objective Loss 0.496202 LR 0.001000 Time 0.016134 -2023-02-13 17:22:38,500 - Epoch: [8][ 490/ 1207] Overall Loss 0.496297 Objective Loss 0.496297 LR 0.001000 Time 0.016106 -2023-02-13 17:22:38,650 - Epoch: [8][ 500/ 1207] Overall Loss 0.495784 Objective Loss 0.495784 LR 0.001000 Time 0.016083 -2023-02-13 17:22:38,798 - Epoch: [8][ 510/ 1207] Overall Loss 0.495590 Objective Loss 0.495590 LR 0.001000 Time 0.016057 -2023-02-13 17:22:38,947 - Epoch: [8][ 520/ 1207] Overall Loss 0.495265 Objective Loss 0.495265 LR 0.001000 Time 0.016034 -2023-02-13 17:22:39,095 - Epoch: [8][ 530/ 1207] Overall Loss 0.495051 Objective Loss 0.495051 LR 0.001000 Time 0.016010 -2023-02-13 17:22:39,244 - Epoch: [8][ 540/ 1207] Overall Loss 0.495025 Objective Loss 0.495025 LR 0.001000 Time 0.015990 -2023-02-13 17:22:39,393 - Epoch: [8][ 550/ 1207] Overall Loss 0.495706 Objective Loss 0.495706 LR 0.001000 Time 0.015968 -2023-02-13 17:22:39,535 - Epoch: [8][ 560/ 1207] Overall Loss 0.496222 Objective Loss 0.496222 LR 0.001000 Time 0.015937 -2023-02-13 17:22:39,668 - Epoch: [8][ 570/ 1207] Overall Loss 0.496808 Objective Loss 0.496808 LR 0.001000 Time 0.015889 -2023-02-13 17:22:39,800 - Epoch: [8][ 580/ 1207] Overall Loss 0.496987 Objective Loss 0.496987 LR 0.001000 Time 0.015840 -2023-02-13 17:22:39,932 - Epoch: [8][ 590/ 1207] Overall Loss 0.496923 Objective Loss 0.496923 LR 0.001000 Time 0.015793 -2023-02-13 17:22:40,065 - Epoch: [8][ 600/ 1207] Overall Loss 0.496662 Objective Loss 0.496662 LR 0.001000 Time 0.015749 -2023-02-13 17:22:40,198 - Epoch: [8][ 610/ 1207] Overall Loss 0.497384 Objective Loss 0.497384 LR 0.001000 Time 0.015705 -2023-02-13 17:22:40,331 - Epoch: [8][ 620/ 1207] Overall Loss 0.496472 Objective Loss 0.496472 LR 0.001000 Time 0.015665 -2023-02-13 17:22:40,464 - Epoch: [8][ 630/ 1207] Overall Loss 0.496171 Objective Loss 0.496171 LR 0.001000 Time 0.015626 -2023-02-13 17:22:40,595 - Epoch: [8][ 640/ 1207] Overall Loss 0.496220 Objective Loss 0.496220 LR 0.001000 Time 0.015587 -2023-02-13 17:22:40,728 - Epoch: [8][ 650/ 1207] Overall Loss 0.496054 Objective Loss 0.496054 LR 0.001000 Time 0.015549 -2023-02-13 17:22:40,860 - Epoch: [8][ 660/ 1207] Overall Loss 0.495948 Objective Loss 0.495948 LR 0.001000 Time 0.015511 -2023-02-13 17:22:40,992 - Epoch: [8][ 670/ 1207] Overall Loss 0.496187 Objective Loss 0.496187 LR 0.001000 Time 0.015473 -2023-02-13 17:22:41,124 - Epoch: [8][ 680/ 1207] Overall Loss 0.496304 Objective Loss 0.496304 LR 0.001000 Time 0.015437 -2023-02-13 17:22:41,255 - Epoch: [8][ 690/ 1207] Overall Loss 0.495743 Objective Loss 0.495743 LR 0.001000 Time 0.015402 -2023-02-13 17:22:41,387 - Epoch: [8][ 700/ 1207] Overall Loss 0.494730 Objective Loss 0.494730 LR 0.001000 Time 0.015367 -2023-02-13 17:22:41,518 - Epoch: [8][ 710/ 1207] Overall Loss 0.494960 Objective Loss 0.494960 LR 0.001000 Time 0.015333 -2023-02-13 17:22:41,650 - Epoch: [8][ 720/ 1207] Overall Loss 0.494731 Objective Loss 0.494731 LR 0.001000 Time 0.015302 -2023-02-13 17:22:41,782 - Epoch: [8][ 730/ 1207] Overall Loss 0.494014 Objective Loss 0.494014 LR 0.001000 Time 0.015271 -2023-02-13 17:22:41,913 - Epoch: [8][ 740/ 1207] Overall Loss 0.493691 Objective Loss 0.493691 LR 0.001000 Time 0.015241 -2023-02-13 17:22:42,045 - Epoch: [8][ 750/ 1207] Overall Loss 0.493329 Objective Loss 0.493329 LR 0.001000 Time 0.015211 -2023-02-13 17:22:42,177 - Epoch: [8][ 760/ 1207] Overall Loss 0.493094 Objective Loss 0.493094 LR 0.001000 Time 0.015185 -2023-02-13 17:22:42,308 - Epoch: [8][ 770/ 1207] Overall Loss 0.493084 Objective Loss 0.493084 LR 0.001000 Time 0.015157 -2023-02-13 17:22:42,440 - Epoch: [8][ 780/ 1207] Overall Loss 0.493194 Objective Loss 0.493194 LR 0.001000 Time 0.015130 -2023-02-13 17:22:42,572 - Epoch: [8][ 790/ 1207] Overall Loss 0.493036 Objective Loss 0.493036 LR 0.001000 Time 0.015104 -2023-02-13 17:22:42,704 - Epoch: [8][ 800/ 1207] Overall Loss 0.493056 Objective Loss 0.493056 LR 0.001000 Time 0.015079 -2023-02-13 17:22:42,836 - Epoch: [8][ 810/ 1207] Overall Loss 0.493032 Objective Loss 0.493032 LR 0.001000 Time 0.015054 -2023-02-13 17:22:42,967 - Epoch: [8][ 820/ 1207] Overall Loss 0.493070 Objective Loss 0.493070 LR 0.001000 Time 0.015028 -2023-02-13 17:22:43,099 - Epoch: [8][ 830/ 1207] Overall Loss 0.493209 Objective Loss 0.493209 LR 0.001000 Time 0.015004 -2023-02-13 17:22:43,230 - Epoch: [8][ 840/ 1207] Overall Loss 0.493006 Objective Loss 0.493006 LR 0.001000 Time 0.014980 -2023-02-13 17:22:43,362 - Epoch: [8][ 850/ 1207] Overall Loss 0.493299 Objective Loss 0.493299 LR 0.001000 Time 0.014957 -2023-02-13 17:22:43,493 - Epoch: [8][ 860/ 1207] Overall Loss 0.493212 Objective Loss 0.493212 LR 0.001000 Time 0.014935 -2023-02-13 17:22:43,625 - Epoch: [8][ 870/ 1207] Overall Loss 0.493335 Objective Loss 0.493335 LR 0.001000 Time 0.014913 -2023-02-13 17:22:43,757 - Epoch: [8][ 880/ 1207] Overall Loss 0.493172 Objective Loss 0.493172 LR 0.001000 Time 0.014891 -2023-02-13 17:22:43,888 - Epoch: [8][ 890/ 1207] Overall Loss 0.493080 Objective Loss 0.493080 LR 0.001000 Time 0.014870 -2023-02-13 17:22:44,020 - Epoch: [8][ 900/ 1207] Overall Loss 0.493327 Objective Loss 0.493327 LR 0.001000 Time 0.014849 -2023-02-13 17:22:44,151 - Epoch: [8][ 910/ 1207] Overall Loss 0.493625 Objective Loss 0.493625 LR 0.001000 Time 0.014828 -2023-02-13 17:22:44,284 - Epoch: [8][ 920/ 1207] Overall Loss 0.493552 Objective Loss 0.493552 LR 0.001000 Time 0.014809 -2023-02-13 17:22:44,416 - Epoch: [8][ 930/ 1207] Overall Loss 0.493926 Objective Loss 0.493926 LR 0.001000 Time 0.014791 -2023-02-13 17:22:44,547 - Epoch: [8][ 940/ 1207] Overall Loss 0.493665 Objective Loss 0.493665 LR 0.001000 Time 0.014772 -2023-02-13 17:22:44,679 - Epoch: [8][ 950/ 1207] Overall Loss 0.493427 Objective Loss 0.493427 LR 0.001000 Time 0.014754 -2023-02-13 17:22:44,811 - Epoch: [8][ 960/ 1207] Overall Loss 0.493290 Objective Loss 0.493290 LR 0.001000 Time 0.014736 -2023-02-13 17:22:44,942 - Epoch: [8][ 970/ 1207] Overall Loss 0.493427 Objective Loss 0.493427 LR 0.001000 Time 0.014718 -2023-02-13 17:22:45,074 - Epoch: [8][ 980/ 1207] Overall Loss 0.493164 Objective Loss 0.493164 LR 0.001000 Time 0.014700 -2023-02-13 17:22:45,209 - Epoch: [8][ 990/ 1207] Overall Loss 0.492650 Objective Loss 0.492650 LR 0.001000 Time 0.014686 -2023-02-13 17:22:45,339 - Epoch: [8][ 1000/ 1207] Overall Loss 0.492733 Objective Loss 0.492733 LR 0.001000 Time 0.014670 -2023-02-13 17:22:45,471 - Epoch: [8][ 1010/ 1207] Overall Loss 0.492791 Objective Loss 0.492791 LR 0.001000 Time 0.014654 -2023-02-13 17:22:45,604 - Epoch: [8][ 1020/ 1207] Overall Loss 0.492877 Objective Loss 0.492877 LR 0.001000 Time 0.014639 -2023-02-13 17:22:45,736 - Epoch: [8][ 1030/ 1207] Overall Loss 0.492726 Objective Loss 0.492726 LR 0.001000 Time 0.014624 -2023-02-13 17:22:45,868 - Epoch: [8][ 1040/ 1207] Overall Loss 0.492305 Objective Loss 0.492305 LR 0.001000 Time 0.014610 -2023-02-13 17:22:46,000 - Epoch: [8][ 1050/ 1207] Overall Loss 0.492161 Objective Loss 0.492161 LR 0.001000 Time 0.014595 -2023-02-13 17:22:46,132 - Epoch: [8][ 1060/ 1207] Overall Loss 0.491947 Objective Loss 0.491947 LR 0.001000 Time 0.014581 -2023-02-13 17:22:46,264 - Epoch: [8][ 1070/ 1207] Overall Loss 0.491486 Objective Loss 0.491486 LR 0.001000 Time 0.014568 -2023-02-13 17:22:46,395 - Epoch: [8][ 1080/ 1207] Overall Loss 0.491536 Objective Loss 0.491536 LR 0.001000 Time 0.014553 -2023-02-13 17:22:46,527 - Epoch: [8][ 1090/ 1207] Overall Loss 0.491717 Objective Loss 0.491717 LR 0.001000 Time 0.014539 -2023-02-13 17:22:46,660 - Epoch: [8][ 1100/ 1207] Overall Loss 0.491624 Objective Loss 0.491624 LR 0.001000 Time 0.014527 -2023-02-13 17:22:46,794 - Epoch: [8][ 1110/ 1207] Overall Loss 0.491389 Objective Loss 0.491389 LR 0.001000 Time 0.014515 -2023-02-13 17:22:46,926 - Epoch: [8][ 1120/ 1207] Overall Loss 0.491506 Objective Loss 0.491506 LR 0.001000 Time 0.014502 -2023-02-13 17:22:47,060 - Epoch: [8][ 1130/ 1207] Overall Loss 0.491662 Objective Loss 0.491662 LR 0.001000 Time 0.014490 -2023-02-13 17:22:47,191 - Epoch: [8][ 1140/ 1207] Overall Loss 0.491579 Objective Loss 0.491579 LR 0.001000 Time 0.014478 -2023-02-13 17:22:47,324 - Epoch: [8][ 1150/ 1207] Overall Loss 0.491835 Objective Loss 0.491835 LR 0.001000 Time 0.014467 -2023-02-13 17:22:47,458 - Epoch: [8][ 1160/ 1207] Overall Loss 0.491994 Objective Loss 0.491994 LR 0.001000 Time 0.014456 -2023-02-13 17:22:47,589 - Epoch: [8][ 1170/ 1207] Overall Loss 0.492598 Objective Loss 0.492598 LR 0.001000 Time 0.014445 -2023-02-13 17:22:47,722 - Epoch: [8][ 1180/ 1207] Overall Loss 0.492263 Objective Loss 0.492263 LR 0.001000 Time 0.014435 -2023-02-13 17:22:47,856 - Epoch: [8][ 1190/ 1207] Overall Loss 0.492283 Objective Loss 0.492283 LR 0.001000 Time 0.014425 -2023-02-13 17:22:48,048 - Epoch: [8][ 1200/ 1207] Overall Loss 0.492224 Objective Loss 0.492224 LR 0.001000 Time 0.014464 -2023-02-13 17:22:48,134 - Epoch: [8][ 1207/ 1207] Overall Loss 0.492251 Objective Loss 0.492251 Top1 71.036585 Top5 94.512195 LR 0.001000 Time 0.014452 -2023-02-13 17:22:48,221 - --- validate (epoch=8)----------- -2023-02-13 17:22:48,221 - 34311 samples (256 per mini-batch) -2023-02-13 17:22:48,570 - Epoch: [8][ 10/ 135] Loss 0.489092 Top1 72.656250 Top5 96.250000 -2023-02-13 17:22:48,665 - Epoch: [8][ 20/ 135] Loss 0.507630 Top1 73.066406 Top5 96.054688 -2023-02-13 17:22:48,762 - Epoch: [8][ 30/ 135] Loss 0.496990 Top1 73.906250 Top5 95.924479 -2023-02-13 17:22:48,864 - Epoch: [8][ 40/ 135] Loss 0.492692 Top1 74.160156 Top5 96.054688 -2023-02-13 17:22:48,957 - Epoch: [8][ 50/ 135] Loss 0.492121 Top1 74.109375 Top5 96.007812 -2023-02-13 17:22:49,046 - Epoch: [8][ 60/ 135] Loss 0.490605 Top1 73.990885 Top5 95.963542 -2023-02-13 17:22:49,135 - Epoch: [8][ 70/ 135] Loss 0.490643 Top1 73.833705 Top5 95.920759 -2023-02-13 17:22:49,224 - Epoch: [8][ 80/ 135] Loss 0.485183 Top1 73.945312 Top5 95.947266 -2023-02-13 17:22:49,312 - Epoch: [8][ 90/ 135] Loss 0.485285 Top1 73.936632 Top5 95.993924 -2023-02-13 17:22:49,401 - Epoch: [8][ 100/ 135] Loss 0.485786 Top1 73.933594 Top5 96.011719 -2023-02-13 17:22:49,490 - Epoch: [8][ 110/ 135] Loss 0.483916 Top1 73.948864 Top5 95.951705 -2023-02-13 17:22:49,576 - Epoch: [8][ 120/ 135] Loss 0.487016 Top1 73.899740 Top5 95.888672 -2023-02-13 17:22:49,667 - Epoch: [8][ 130/ 135] Loss 0.488875 Top1 73.864183 Top5 95.865385 -2023-02-13 17:22:49,691 - Epoch: [8][ 135/ 135] Loss 0.488468 Top1 73.909242 Top5 95.858471 -2023-02-13 17:22:49,762 - ==> Top1: 73.909 Top5: 95.858 Loss: 0.488 - -2023-02-13 17:22:49,762 - ==> Confusion: -[[ 768 2 8 0 23 5 0 1 13 106 3 8 1 5 11 1 2 1 3 1 5] - [ 2 923 1 2 11 39 1 11 10 0 6 3 1 2 5 1 4 0 7 1 3] - [ 8 11 929 23 8 6 5 17 1 1 9 4 1 2 3 4 2 4 12 4 4] - [ 7 8 14 875 1 6 0 1 4 1 27 2 4 3 23 0 2 11 23 0 4] - [ 4 19 1 1 989 7 0 1 2 4 0 3 4 4 11 2 2 0 3 2 7] - [ 2 83 5 5 6 889 1 26 4 8 4 13 2 9 4 0 2 1 3 2 1] - [ 5 5 62 6 4 10 952 12 1 0 10 3 1 0 0 6 2 6 4 10 0] - [ 1 37 14 1 3 36 3 826 6 1 6 5 3 1 0 0 2 0 65 9 5] - [ 12 6 0 1 1 2 0 0 886 35 18 2 0 8 27 0 1 2 7 0 1] - [ 97 2 7 1 8 4 1 0 62 792 1 1 0 18 7 0 0 1 3 0 7] - [ 3 6 10 12 3 2 0 3 12 1 968 1 2 7 3 0 0 0 16 0 2] - [ 6 7 0 1 5 34 2 4 3 2 0 864 38 11 0 3 4 5 4 11 1] - [ 2 1 1 7 3 9 0 1 4 0 2 84 778 1 5 4 6 37 3 3 8] - [ 7 6 1 1 15 22 0 2 33 20 18 23 1 854 8 0 3 0 4 2 4] - [ 16 6 1 20 18 5 0 0 38 6 5 3 2 1 932 2 1 7 20 0 9] - [ 7 5 10 1 22 7 4 1 0 0 1 21 2 6 1 911 14 17 0 9 7] - [ 2 23 1 1 32 4 1 0 4 0 3 5 1 2 4 11 952 3 2 1 9] - [ 6 3 1 7 1 3 0 0 6 4 1 29 22 4 7 14 1 934 1 4 3] - [ 4 4 4 12 3 1 0 26 5 0 6 5 5 1 13 0 0 0 992 0 5] - [ 0 5 1 1 1 23 9 23 0 0 5 31 6 6 0 3 5 2 2 1017 8] - [ 260 625 356 240 412 421 90 195 189 181 305 247 405 373 282 142 411 137 421 414 7328]] - -2023-02-13 17:22:49,764 - ==> Best [Top1: 73.909 Top5: 95.858 Sparsity:0.00 Params: 148928 on epoch: 8] -2023-02-13 17:22:49,764 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:22:49,770 - - -2023-02-13 17:22:49,771 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:22:50,605 - Epoch: [9][ 10/ 1207] Overall Loss 0.447025 Objective Loss 0.447025 LR 0.001000 Time 0.083426 -2023-02-13 17:22:50,749 - Epoch: [9][ 20/ 1207] Overall Loss 0.466957 Objective Loss 0.466957 LR 0.001000 Time 0.048867 -2023-02-13 17:22:50,884 - Epoch: [9][ 30/ 1207] Overall Loss 0.455684 Objective Loss 0.455684 LR 0.001000 Time 0.037052 -2023-02-13 17:22:51,018 - Epoch: [9][ 40/ 1207] Overall Loss 0.451500 Objective Loss 0.451500 LR 0.001000 Time 0.031129 -2023-02-13 17:22:51,151 - Epoch: [9][ 50/ 1207] Overall Loss 0.448004 Objective Loss 0.448004 LR 0.001000 Time 0.027554 -2023-02-13 17:22:51,284 - Epoch: [9][ 60/ 1207] Overall Loss 0.449652 Objective Loss 0.449652 LR 0.001000 Time 0.025185 -2023-02-13 17:22:51,417 - Epoch: [9][ 70/ 1207] Overall Loss 0.450822 Objective Loss 0.450822 LR 0.001000 Time 0.023478 -2023-02-13 17:22:51,551 - Epoch: [9][ 80/ 1207] Overall Loss 0.450287 Objective Loss 0.450287 LR 0.001000 Time 0.022215 -2023-02-13 17:22:51,685 - Epoch: [9][ 90/ 1207] Overall Loss 0.454108 Objective Loss 0.454108 LR 0.001000 Time 0.021223 -2023-02-13 17:22:51,819 - Epoch: [9][ 100/ 1207] Overall Loss 0.455356 Objective Loss 0.455356 LR 0.001000 Time 0.020440 -2023-02-13 17:22:51,951 - Epoch: [9][ 110/ 1207] Overall Loss 0.455339 Objective Loss 0.455339 LR 0.001000 Time 0.019782 -2023-02-13 17:22:52,086 - Epoch: [9][ 120/ 1207] Overall Loss 0.456792 Objective Loss 0.456792 LR 0.001000 Time 0.019248 -2023-02-13 17:22:52,218 - Epoch: [9][ 130/ 1207] Overall Loss 0.455608 Objective Loss 0.455608 LR 0.001000 Time 0.018784 -2023-02-13 17:22:52,352 - Epoch: [9][ 140/ 1207] Overall Loss 0.455566 Objective Loss 0.455566 LR 0.001000 Time 0.018396 -2023-02-13 17:22:52,485 - Epoch: [9][ 150/ 1207] Overall Loss 0.453575 Objective Loss 0.453575 LR 0.001000 Time 0.018051 -2023-02-13 17:22:52,618 - Epoch: [9][ 160/ 1207] Overall Loss 0.452074 Objective Loss 0.452074 LR 0.001000 Time 0.017753 -2023-02-13 17:22:52,752 - Epoch: [9][ 170/ 1207] Overall Loss 0.452139 Objective Loss 0.452139 LR 0.001000 Time 0.017487 -2023-02-13 17:22:52,886 - Epoch: [9][ 180/ 1207] Overall Loss 0.454468 Objective Loss 0.454468 LR 0.001000 Time 0.017259 -2023-02-13 17:22:53,019 - Epoch: [9][ 190/ 1207] Overall Loss 0.454446 Objective Loss 0.454446 LR 0.001000 Time 0.017046 -2023-02-13 17:22:53,153 - Epoch: [9][ 200/ 1207] Overall Loss 0.453666 Objective Loss 0.453666 LR 0.001000 Time 0.016862 -2023-02-13 17:22:53,285 - Epoch: [9][ 210/ 1207] Overall Loss 0.454081 Objective Loss 0.454081 LR 0.001000 Time 0.016689 -2023-02-13 17:22:53,420 - Epoch: [9][ 220/ 1207] Overall Loss 0.454490 Objective Loss 0.454490 LR 0.001000 Time 0.016541 -2023-02-13 17:22:53,553 - Epoch: [9][ 230/ 1207] Overall Loss 0.454824 Objective Loss 0.454824 LR 0.001000 Time 0.016396 -2023-02-13 17:22:53,688 - Epoch: [9][ 240/ 1207] Overall Loss 0.455713 Objective Loss 0.455713 LR 0.001000 Time 0.016275 -2023-02-13 17:22:53,821 - Epoch: [9][ 250/ 1207] Overall Loss 0.456358 Objective Loss 0.456358 LR 0.001000 Time 0.016154 -2023-02-13 17:22:53,955 - Epoch: [9][ 260/ 1207] Overall Loss 0.456828 Objective Loss 0.456828 LR 0.001000 Time 0.016047 -2023-02-13 17:22:54,086 - Epoch: [9][ 270/ 1207] Overall Loss 0.456782 Objective Loss 0.456782 LR 0.001000 Time 0.015937 -2023-02-13 17:22:54,222 - Epoch: [9][ 280/ 1207] Overall Loss 0.458306 Objective Loss 0.458306 LR 0.001000 Time 0.015851 -2023-02-13 17:22:54,363 - Epoch: [9][ 290/ 1207] Overall Loss 0.459382 Objective Loss 0.459382 LR 0.001000 Time 0.015784 -2023-02-13 17:22:54,502 - Epoch: [9][ 300/ 1207] Overall Loss 0.460485 Objective Loss 0.460485 LR 0.001000 Time 0.015720 -2023-02-13 17:22:54,641 - Epoch: [9][ 310/ 1207] Overall Loss 0.460477 Objective Loss 0.460477 LR 0.001000 Time 0.015661 -2023-02-13 17:22:54,780 - Epoch: [9][ 320/ 1207] Overall Loss 0.460436 Objective Loss 0.460436 LR 0.001000 Time 0.015601 -2023-02-13 17:22:54,920 - Epoch: [9][ 330/ 1207] Overall Loss 0.462021 Objective Loss 0.462021 LR 0.001000 Time 0.015550 -2023-02-13 17:22:55,057 - Epoch: [9][ 340/ 1207] Overall Loss 0.463387 Objective Loss 0.463387 LR 0.001000 Time 0.015494 -2023-02-13 17:22:55,197 - Epoch: [9][ 350/ 1207] Overall Loss 0.463136 Objective Loss 0.463136 LR 0.001000 Time 0.015451 -2023-02-13 17:22:55,335 - Epoch: [9][ 360/ 1207] Overall Loss 0.463283 Objective Loss 0.463283 LR 0.001000 Time 0.015402 -2023-02-13 17:22:55,475 - Epoch: [9][ 370/ 1207] Overall Loss 0.463832 Objective Loss 0.463832 LR 0.001000 Time 0.015363 -2023-02-13 17:22:55,615 - Epoch: [9][ 380/ 1207] Overall Loss 0.464261 Objective Loss 0.464261 LR 0.001000 Time 0.015324 -2023-02-13 17:22:55,755 - Epoch: [9][ 390/ 1207] Overall Loss 0.463056 Objective Loss 0.463056 LR 0.001000 Time 0.015290 -2023-02-13 17:22:55,894 - Epoch: [9][ 400/ 1207] Overall Loss 0.462876 Objective Loss 0.462876 LR 0.001000 Time 0.015255 -2023-02-13 17:22:56,034 - Epoch: [9][ 410/ 1207] Overall Loss 0.463488 Objective Loss 0.463488 LR 0.001000 Time 0.015224 -2023-02-13 17:22:56,174 - Epoch: [9][ 420/ 1207] Overall Loss 0.464439 Objective Loss 0.464439 LR 0.001000 Time 0.015192 -2023-02-13 17:22:56,309 - Epoch: [9][ 430/ 1207] Overall Loss 0.465420 Objective Loss 0.465420 LR 0.001000 Time 0.015152 -2023-02-13 17:22:56,442 - Epoch: [9][ 440/ 1207] Overall Loss 0.466193 Objective Loss 0.466193 LR 0.001000 Time 0.015107 -2023-02-13 17:22:56,576 - Epoch: [9][ 450/ 1207] Overall Loss 0.466266 Objective Loss 0.466266 LR 0.001000 Time 0.015068 -2023-02-13 17:22:56,709 - Epoch: [9][ 460/ 1207] Overall Loss 0.466994 Objective Loss 0.466994 LR 0.001000 Time 0.015029 -2023-02-13 17:22:56,844 - Epoch: [9][ 470/ 1207] Overall Loss 0.465528 Objective Loss 0.465528 LR 0.001000 Time 0.014995 -2023-02-13 17:22:56,977 - Epoch: [9][ 480/ 1207] Overall Loss 0.465349 Objective Loss 0.465349 LR 0.001000 Time 0.014958 -2023-02-13 17:22:57,110 - Epoch: [9][ 490/ 1207] Overall Loss 0.465472 Objective Loss 0.465472 LR 0.001000 Time 0.014925 -2023-02-13 17:22:57,243 - Epoch: [9][ 500/ 1207] Overall Loss 0.465993 Objective Loss 0.465993 LR 0.001000 Time 0.014890 -2023-02-13 17:22:57,376 - Epoch: [9][ 510/ 1207] Overall Loss 0.466207 Objective Loss 0.466207 LR 0.001000 Time 0.014860 -2023-02-13 17:22:57,509 - Epoch: [9][ 520/ 1207] Overall Loss 0.466014 Objective Loss 0.466014 LR 0.001000 Time 0.014828 -2023-02-13 17:22:57,658 - Epoch: [9][ 530/ 1207] Overall Loss 0.466360 Objective Loss 0.466360 LR 0.001000 Time 0.014829 -2023-02-13 17:22:57,800 - Epoch: [9][ 540/ 1207] Overall Loss 0.466183 Objective Loss 0.466183 LR 0.001000 Time 0.014817 -2023-02-13 17:22:57,942 - Epoch: [9][ 550/ 1207] Overall Loss 0.465885 Objective Loss 0.465885 LR 0.001000 Time 0.014804 -2023-02-13 17:22:58,083 - Epoch: [9][ 560/ 1207] Overall Loss 0.466109 Objective Loss 0.466109 LR 0.001000 Time 0.014792 -2023-02-13 17:22:58,223 - Epoch: [9][ 570/ 1207] Overall Loss 0.465280 Objective Loss 0.465280 LR 0.001000 Time 0.014777 -2023-02-13 17:22:58,364 - Epoch: [9][ 580/ 1207] Overall Loss 0.465583 Objective Loss 0.465583 LR 0.001000 Time 0.014764 -2023-02-13 17:22:58,503 - Epoch: [9][ 590/ 1207] Overall Loss 0.466365 Objective Loss 0.466365 LR 0.001000 Time 0.014750 -2023-02-13 17:22:58,645 - Epoch: [9][ 600/ 1207] Overall Loss 0.466879 Objective Loss 0.466879 LR 0.001000 Time 0.014739 -2023-02-13 17:22:58,779 - Epoch: [9][ 610/ 1207] Overall Loss 0.467631 Objective Loss 0.467631 LR 0.001000 Time 0.014717 -2023-02-13 17:22:58,912 - Epoch: [9][ 620/ 1207] Overall Loss 0.467922 Objective Loss 0.467922 LR 0.001000 Time 0.014694 -2023-02-13 17:22:59,047 - Epoch: [9][ 630/ 1207] Overall Loss 0.468039 Objective Loss 0.468039 LR 0.001000 Time 0.014674 -2023-02-13 17:22:59,180 - Epoch: [9][ 640/ 1207] Overall Loss 0.467975 Objective Loss 0.467975 LR 0.001000 Time 0.014651 -2023-02-13 17:22:59,314 - Epoch: [9][ 650/ 1207] Overall Loss 0.468324 Objective Loss 0.468324 LR 0.001000 Time 0.014632 -2023-02-13 17:22:59,447 - Epoch: [9][ 660/ 1207] Overall Loss 0.467947 Objective Loss 0.467947 LR 0.001000 Time 0.014610 -2023-02-13 17:22:59,582 - Epoch: [9][ 670/ 1207] Overall Loss 0.468492 Objective Loss 0.468492 LR 0.001000 Time 0.014593 -2023-02-13 17:22:59,716 - Epoch: [9][ 680/ 1207] Overall Loss 0.468920 Objective Loss 0.468920 LR 0.001000 Time 0.014575 -2023-02-13 17:22:59,851 - Epoch: [9][ 690/ 1207] Overall Loss 0.469616 Objective Loss 0.469616 LR 0.001000 Time 0.014560 -2023-02-13 17:22:59,984 - Epoch: [9][ 700/ 1207] Overall Loss 0.469878 Objective Loss 0.469878 LR 0.001000 Time 0.014541 -2023-02-13 17:23:00,120 - Epoch: [9][ 710/ 1207] Overall Loss 0.469488 Objective Loss 0.469488 LR 0.001000 Time 0.014526 -2023-02-13 17:23:00,253 - Epoch: [9][ 720/ 1207] Overall Loss 0.469577 Objective Loss 0.469577 LR 0.001000 Time 0.014509 -2023-02-13 17:23:00,387 - Epoch: [9][ 730/ 1207] Overall Loss 0.469780 Objective Loss 0.469780 LR 0.001000 Time 0.014494 -2023-02-13 17:23:00,521 - Epoch: [9][ 740/ 1207] Overall Loss 0.469619 Objective Loss 0.469619 LR 0.001000 Time 0.014478 -2023-02-13 17:23:00,656 - Epoch: [9][ 750/ 1207] Overall Loss 0.470509 Objective Loss 0.470509 LR 0.001000 Time 0.014465 -2023-02-13 17:23:00,790 - Epoch: [9][ 760/ 1207] Overall Loss 0.470421 Objective Loss 0.470421 LR 0.001000 Time 0.014450 -2023-02-13 17:23:00,921 - Epoch: [9][ 770/ 1207] Overall Loss 0.470101 Objective Loss 0.470101 LR 0.001000 Time 0.014432 -2023-02-13 17:23:01,054 - Epoch: [9][ 780/ 1207] Overall Loss 0.470541 Objective Loss 0.470541 LR 0.001000 Time 0.014417 -2023-02-13 17:23:01,186 - Epoch: [9][ 790/ 1207] Overall Loss 0.471000 Objective Loss 0.471000 LR 0.001000 Time 0.014400 -2023-02-13 17:23:01,320 - Epoch: [9][ 800/ 1207] Overall Loss 0.471311 Objective Loss 0.471311 LR 0.001000 Time 0.014386 -2023-02-13 17:23:01,452 - Epoch: [9][ 810/ 1207] Overall Loss 0.471452 Objective Loss 0.471452 LR 0.001000 Time 0.014372 -2023-02-13 17:23:01,584 - Epoch: [9][ 820/ 1207] Overall Loss 0.471439 Objective Loss 0.471439 LR 0.001000 Time 0.014357 -2023-02-13 17:23:01,717 - Epoch: [9][ 830/ 1207] Overall Loss 0.471609 Objective Loss 0.471609 LR 0.001000 Time 0.014343 -2023-02-13 17:23:01,851 - Epoch: [9][ 840/ 1207] Overall Loss 0.471474 Objective Loss 0.471474 LR 0.001000 Time 0.014331 -2023-02-13 17:23:01,982 - Epoch: [9][ 850/ 1207] Overall Loss 0.471760 Objective Loss 0.471760 LR 0.001000 Time 0.014316 -2023-02-13 17:23:02,114 - Epoch: [9][ 860/ 1207] Overall Loss 0.471792 Objective Loss 0.471792 LR 0.001000 Time 0.014303 -2023-02-13 17:23:02,246 - Epoch: [9][ 870/ 1207] Overall Loss 0.471660 Objective Loss 0.471660 LR 0.001000 Time 0.014290 -2023-02-13 17:23:02,378 - Epoch: [9][ 880/ 1207] Overall Loss 0.471720 Objective Loss 0.471720 LR 0.001000 Time 0.014278 -2023-02-13 17:23:02,511 - Epoch: [9][ 890/ 1207] Overall Loss 0.471915 Objective Loss 0.471915 LR 0.001000 Time 0.014264 -2023-02-13 17:23:02,644 - Epoch: [9][ 900/ 1207] Overall Loss 0.471866 Objective Loss 0.471866 LR 0.001000 Time 0.014253 -2023-02-13 17:23:02,776 - Epoch: [9][ 910/ 1207] Overall Loss 0.471918 Objective Loss 0.471918 LR 0.001000 Time 0.014241 -2023-02-13 17:23:02,909 - Epoch: [9][ 920/ 1207] Overall Loss 0.472102 Objective Loss 0.472102 LR 0.001000 Time 0.014230 -2023-02-13 17:23:03,041 - Epoch: [9][ 930/ 1207] Overall Loss 0.472204 Objective Loss 0.472204 LR 0.001000 Time 0.014219 -2023-02-13 17:23:03,173 - Epoch: [9][ 940/ 1207] Overall Loss 0.471994 Objective Loss 0.471994 LR 0.001000 Time 0.014208 -2023-02-13 17:23:03,305 - Epoch: [9][ 950/ 1207] Overall Loss 0.472001 Objective Loss 0.472001 LR 0.001000 Time 0.014196 -2023-02-13 17:23:03,437 - Epoch: [9][ 960/ 1207] Overall Loss 0.471945 Objective Loss 0.471945 LR 0.001000 Time 0.014186 -2023-02-13 17:23:03,569 - Epoch: [9][ 970/ 1207] Overall Loss 0.472037 Objective Loss 0.472037 LR 0.001000 Time 0.014176 -2023-02-13 17:23:03,703 - Epoch: [9][ 980/ 1207] Overall Loss 0.472247 Objective Loss 0.472247 LR 0.001000 Time 0.014167 -2023-02-13 17:23:03,834 - Epoch: [9][ 990/ 1207] Overall Loss 0.472186 Objective Loss 0.472186 LR 0.001000 Time 0.014156 -2023-02-13 17:23:03,968 - Epoch: [9][ 1000/ 1207] Overall Loss 0.472153 Objective Loss 0.472153 LR 0.001000 Time 0.014148 -2023-02-13 17:23:04,099 - Epoch: [9][ 1010/ 1207] Overall Loss 0.472268 Objective Loss 0.472268 LR 0.001000 Time 0.014137 -2023-02-13 17:23:04,232 - Epoch: [9][ 1020/ 1207] Overall Loss 0.472590 Objective Loss 0.472590 LR 0.001000 Time 0.014129 -2023-02-13 17:23:04,364 - Epoch: [9][ 1030/ 1207] Overall Loss 0.472305 Objective Loss 0.472305 LR 0.001000 Time 0.014120 -2023-02-13 17:23:04,497 - Epoch: [9][ 1040/ 1207] Overall Loss 0.472367 Objective Loss 0.472367 LR 0.001000 Time 0.014111 -2023-02-13 17:23:04,629 - Epoch: [9][ 1050/ 1207] Overall Loss 0.471930 Objective Loss 0.471930 LR 0.001000 Time 0.014102 -2023-02-13 17:23:04,760 - Epoch: [9][ 1060/ 1207] Overall Loss 0.471438 Objective Loss 0.471438 LR 0.001000 Time 0.014092 -2023-02-13 17:23:04,893 - Epoch: [9][ 1070/ 1207] Overall Loss 0.471356 Objective Loss 0.471356 LR 0.001000 Time 0.014084 -2023-02-13 17:23:05,024 - Epoch: [9][ 1080/ 1207] Overall Loss 0.471133 Objective Loss 0.471133 LR 0.001000 Time 0.014075 -2023-02-13 17:23:05,156 - Epoch: [9][ 1090/ 1207] Overall Loss 0.471282 Objective Loss 0.471282 LR 0.001000 Time 0.014067 -2023-02-13 17:23:05,290 - Epoch: [9][ 1100/ 1207] Overall Loss 0.471089 Objective Loss 0.471089 LR 0.001000 Time 0.014060 -2023-02-13 17:23:05,421 - Epoch: [9][ 1110/ 1207] Overall Loss 0.471113 Objective Loss 0.471113 LR 0.001000 Time 0.014051 -2023-02-13 17:23:05,554 - Epoch: [9][ 1120/ 1207] Overall Loss 0.470842 Objective Loss 0.470842 LR 0.001000 Time 0.014043 -2023-02-13 17:23:05,686 - Epoch: [9][ 1130/ 1207] Overall Loss 0.470963 Objective Loss 0.470963 LR 0.001000 Time 0.014035 -2023-02-13 17:23:05,821 - Epoch: [9][ 1140/ 1207] Overall Loss 0.470668 Objective Loss 0.470668 LR 0.001000 Time 0.014029 -2023-02-13 17:23:05,952 - Epoch: [9][ 1150/ 1207] Overall Loss 0.470460 Objective Loss 0.470460 LR 0.001000 Time 0.014020 -2023-02-13 17:23:06,084 - Epoch: [9][ 1160/ 1207] Overall Loss 0.470505 Objective Loss 0.470505 LR 0.001000 Time 0.014013 -2023-02-13 17:23:06,217 - Epoch: [9][ 1170/ 1207] Overall Loss 0.470251 Objective Loss 0.470251 LR 0.001000 Time 0.014006 -2023-02-13 17:23:06,349 - Epoch: [9][ 1180/ 1207] Overall Loss 0.470239 Objective Loss 0.470239 LR 0.001000 Time 0.013999 -2023-02-13 17:23:06,481 - Epoch: [9][ 1190/ 1207] Overall Loss 0.470064 Objective Loss 0.470064 LR 0.001000 Time 0.013991 -2023-02-13 17:23:06,667 - Epoch: [9][ 1200/ 1207] Overall Loss 0.470320 Objective Loss 0.470320 LR 0.001000 Time 0.014029 -2023-02-13 17:23:06,752 - Epoch: [9][ 1207/ 1207] Overall Loss 0.470742 Objective Loss 0.470742 Top1 75.609756 Top5 96.951220 LR 0.001000 Time 0.014019 -2023-02-13 17:23:06,823 - --- validate (epoch=9)----------- -2023-02-13 17:23:06,823 - 34311 samples (256 per mini-batch) -2023-02-13 17:23:07,177 - Epoch: [9][ 10/ 135] Loss 0.523720 Top1 72.734375 Top5 95.273438 -2023-02-13 17:23:07,271 - Epoch: [9][ 20/ 135] Loss 0.492193 Top1 73.281250 Top5 95.605469 -2023-02-13 17:23:07,355 - Epoch: [9][ 30/ 135] Loss 0.475242 Top1 73.177083 Top5 95.546875 -2023-02-13 17:23:07,444 - Epoch: [9][ 40/ 135] Loss 0.480439 Top1 73.574219 Top5 95.625000 -2023-02-13 17:23:07,530 - Epoch: [9][ 50/ 135] Loss 0.475807 Top1 73.664062 Top5 95.796875 -2023-02-13 17:23:07,616 - Epoch: [9][ 60/ 135] Loss 0.468645 Top1 73.990885 Top5 95.787760 -2023-02-13 17:23:07,702 - Epoch: [9][ 70/ 135] Loss 0.469276 Top1 73.950893 Top5 95.680804 -2023-02-13 17:23:07,789 - Epoch: [9][ 80/ 135] Loss 0.470520 Top1 74.116211 Top5 95.683594 -2023-02-13 17:23:07,874 - Epoch: [9][ 90/ 135] Loss 0.467194 Top1 74.101562 Top5 95.785590 -2023-02-13 17:23:07,960 - Epoch: [9][ 100/ 135] Loss 0.467776 Top1 74.011719 Top5 95.812500 -2023-02-13 17:23:08,046 - Epoch: [9][ 110/ 135] Loss 0.470022 Top1 74.034091 Top5 95.813210 -2023-02-13 17:23:08,132 - Epoch: [9][ 120/ 135] Loss 0.468698 Top1 74.059245 Top5 95.852865 -2023-02-13 17:23:08,224 - Epoch: [9][ 130/ 135] Loss 0.471432 Top1 74.068510 Top5 95.841346 -2023-02-13 17:23:08,249 - Epoch: [9][ 135/ 135] Loss 0.470749 Top1 74.046224 Top5 95.817668 -2023-02-13 17:23:08,316 - ==> Top1: 74.046 Top5: 95.818 Loss: 0.471 - -2023-02-13 17:23:08,317 - ==> Confusion: -[[ 835 1 9 2 13 6 1 1 0 54 0 10 3 4 6 7 3 4 1 2 5] - [ 4 880 4 3 15 58 2 21 4 1 8 7 2 0 3 1 5 0 5 4 6] - [ 12 4 942 11 5 4 25 19 0 3 4 3 0 1 1 3 3 2 7 7 2] - [ 3 4 38 870 0 7 4 4 2 3 16 3 10 1 21 4 1 6 14 0 5] - [ 16 9 2 3 989 9 1 0 1 7 0 6 3 1 9 5 2 0 0 0 3] - [ 3 35 5 6 9 909 12 27 2 2 0 15 4 12 2 4 3 1 5 14 0] - [ 3 6 36 2 0 1 1018 4 0 0 4 2 3 2 0 3 2 0 1 12 0] - [ 2 8 23 3 2 32 10 845 2 2 6 4 1 1 0 3 2 1 46 27 4] - [ 20 6 1 2 2 1 1 2 790 73 24 3 9 23 40 0 1 1 8 0 2] - [ 153 0 8 1 7 5 3 1 18 773 1 3 2 19 5 1 2 3 2 0 5] - [ 1 6 15 20 1 3 6 3 5 3 945 1 0 14 7 3 1 1 12 2 2] - [ 4 5 2 1 4 25 5 4 1 2 0 844 28 6 1 9 2 5 4 51 2] - [ 2 1 2 5 2 8 3 1 0 0 0 69 813 0 2 8 5 20 2 12 4] - [ 13 5 6 1 7 18 0 3 6 28 10 22 2 866 10 5 4 1 0 11 6] - [ 19 5 3 29 15 2 1 1 16 6 5 3 7 2 947 1 3 8 10 0 9] - [ 2 4 8 0 13 8 11 1 0 1 0 10 5 1 1 940 17 5 0 14 5] - [ 6 14 1 2 25 3 0 0 0 1 1 7 3 3 6 10 962 2 1 8 6] - [ 5 1 1 7 2 1 1 1 4 0 1 25 48 1 2 43 0 890 0 11 7] - [ 2 2 9 18 0 4 2 29 4 1 11 3 11 0 19 1 0 1 958 6 5] - [ 1 2 1 0 1 9 17 12 0 0 1 11 5 3 0 5 3 2 0 1072 3] - [ 346 327 474 296 426 393 206 183 74 160 213 204 428 435 292 241 435 79 231 673 7318]] - -2023-02-13 17:23:08,319 - ==> Best [Top1: 74.046 Top5: 95.818 Sparsity:0.00 Params: 148928 on epoch: 9] -2023-02-13 17:23:08,319 - Saving checkpoint to: logs/2023.02.13-171954/checkpoint.pth.tar -2023-02-13 17:23:08,339 - - -2023-02-13 17:23:08,339 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:23:09,216 - Epoch: [10][ 10/ 1207] Overall Loss 0.665844 Objective Loss 0.665844 LR 0.001000 Time 0.087636 -2023-02-13 17:23:09,415 - Epoch: [10][ 20/ 1207] Overall Loss 0.629228 Objective Loss 0.629228 LR 0.001000 Time 0.053738 -2023-02-13 17:23:09,603 - Epoch: [10][ 30/ 1207] Overall Loss 0.614855 Objective Loss 0.614855 LR 0.001000 Time 0.042092 -2023-02-13 17:23:09,792 - Epoch: [10][ 40/ 1207] Overall Loss 0.602103 Objective Loss 0.602103 LR 0.001000 Time 0.036287 -2023-02-13 17:23:09,980 - Epoch: [10][ 50/ 1207] Overall Loss 0.592191 Objective Loss 0.592191 LR 0.001000 Time 0.032780 -2023-02-13 17:23:10,168 - Epoch: [10][ 60/ 1207] Overall Loss 0.584972 Objective Loss 0.584972 LR 0.001000 Time 0.030449 -2023-02-13 17:23:10,357 - Epoch: [10][ 70/ 1207] Overall Loss 0.577081 Objective Loss 0.577081 LR 0.001000 Time 0.028788 -2023-02-13 17:23:10,546 - Epoch: [10][ 80/ 1207] Overall Loss 0.568451 Objective Loss 0.568451 LR 0.001000 Time 0.027549 -2023-02-13 17:23:10,735 - Epoch: [10][ 90/ 1207] Overall Loss 0.564376 Objective Loss 0.564376 LR 0.001000 Time 0.026587 -2023-02-13 17:23:10,925 - Epoch: [10][ 100/ 1207] Overall Loss 0.560653 Objective Loss 0.560653 LR 0.001000 Time 0.025816 -2023-02-13 17:23:11,113 - Epoch: [10][ 110/ 1207] Overall Loss 0.556231 Objective Loss 0.556231 LR 0.001000 Time 0.025182 -2023-02-13 17:23:11,302 - Epoch: [10][ 120/ 1207] Overall Loss 0.552884 Objective Loss 0.552884 LR 0.001000 Time 0.024651 -2023-02-13 17:23:11,491 - Epoch: [10][ 130/ 1207] Overall Loss 0.551224 Objective Loss 0.551224 LR 0.001000 Time 0.024206 -2023-02-13 17:23:11,680 - Epoch: [10][ 140/ 1207] Overall Loss 0.549520 Objective Loss 0.549520 LR 0.001000 Time 0.023825 -2023-02-13 17:23:11,869 - Epoch: [10][ 150/ 1207] Overall Loss 0.546352 Objective Loss 0.546352 LR 0.001000 Time 0.023492 -2023-02-13 17:23:12,057 - Epoch: [10][ 160/ 1207] Overall Loss 0.545977 Objective Loss 0.545977 LR 0.001000 Time 0.023199 -2023-02-13 17:23:12,245 - Epoch: [10][ 170/ 1207] Overall Loss 0.545231 Objective Loss 0.545231 LR 0.001000 Time 0.022938 -2023-02-13 17:23:12,433 - Epoch: [10][ 180/ 1207] Overall Loss 0.542972 Objective Loss 0.542972 LR 0.001000 Time 0.022708 -2023-02-13 17:23:12,622 - Epoch: [10][ 190/ 1207] Overall Loss 0.540849 Objective Loss 0.540849 LR 0.001000 Time 0.022504 -2023-02-13 17:23:12,811 - Epoch: [10][ 200/ 1207] Overall Loss 0.538109 Objective Loss 0.538109 LR 0.001000 Time 0.022321 -2023-02-13 17:23:12,999 - Epoch: [10][ 210/ 1207] Overall Loss 0.537133 Objective Loss 0.537133 LR 0.001000 Time 0.022154 -2023-02-13 17:23:13,188 - Epoch: [10][ 220/ 1207] Overall Loss 0.537090 Objective Loss 0.537090 LR 0.001000 Time 0.022002 -2023-02-13 17:23:13,376 - Epoch: [10][ 230/ 1207] Overall Loss 0.535436 Objective Loss 0.535436 LR 0.001000 Time 0.021863 -2023-02-13 17:23:13,565 - Epoch: [10][ 240/ 1207] Overall Loss 0.534496 Objective Loss 0.534496 LR 0.001000 Time 0.021735 -2023-02-13 17:23:13,753 - Epoch: [10][ 250/ 1207] Overall Loss 0.532537 Objective Loss 0.532537 LR 0.001000 Time 0.021618 -2023-02-13 17:23:13,942 - Epoch: [10][ 260/ 1207] Overall Loss 0.529988 Objective Loss 0.529988 LR 0.001000 Time 0.021510 -2023-02-13 17:23:14,129 - Epoch: [10][ 270/ 1207] Overall Loss 0.528830 Objective Loss 0.528830 LR 0.001000 Time 0.021407 -2023-02-13 17:23:14,317 - Epoch: [10][ 280/ 1207] Overall Loss 0.532597 Objective Loss 0.532597 LR 0.001000 Time 0.021312 -2023-02-13 17:23:14,505 - Epoch: [10][ 290/ 1207] Overall Loss 0.543786 Objective Loss 0.543786 LR 0.001000 Time 0.021225 -2023-02-13 17:23:14,694 - Epoch: [10][ 300/ 1207] Overall Loss 0.551345 Objective Loss 0.551345 LR 0.001000 Time 0.021144 -2023-02-13 17:23:14,882 - Epoch: [10][ 310/ 1207] Overall Loss 0.558324 Objective Loss 0.558324 LR 0.001000 Time 0.021067 -2023-02-13 17:23:15,070 - Epoch: [10][ 320/ 1207] Overall Loss 0.563391 Objective Loss 0.563391 LR 0.001000 Time 0.020996 -2023-02-13 17:23:15,258 - Epoch: [10][ 330/ 1207] Overall Loss 0.568194 Objective Loss 0.568194 LR 0.001000 Time 0.020929 -2023-02-13 17:23:15,446 - Epoch: [10][ 340/ 1207] Overall Loss 0.571361 Objective Loss 0.571361 LR 0.001000 Time 0.020864 -2023-02-13 17:23:15,633 - Epoch: [10][ 350/ 1207] Overall Loss 0.574937 Objective Loss 0.574937 LR 0.001000 Time 0.020803 -2023-02-13 17:23:15,823 - Epoch: [10][ 360/ 1207] Overall Loss 0.577913 Objective Loss 0.577913 LR 0.001000 Time 0.020751 -2023-02-13 17:23:16,012 - Epoch: [10][ 370/ 1207] Overall Loss 0.579413 Objective Loss 0.579413 LR 0.001000 Time 0.020701 -2023-02-13 17:23:16,200 - Epoch: [10][ 380/ 1207] Overall Loss 0.580839 Objective Loss 0.580839 LR 0.001000 Time 0.020649 -2023-02-13 17:23:16,388 - Epoch: [10][ 390/ 1207] Overall Loss 0.582010 Objective Loss 0.582010 LR 0.001000 Time 0.020600 -2023-02-13 17:23:16,576 - Epoch: [10][ 400/ 1207] Overall Loss 0.582880 Objective Loss 0.582880 LR 0.001000 Time 0.020555 -2023-02-13 17:23:16,765 - Epoch: [10][ 410/ 1207] Overall Loss 0.583415 Objective Loss 0.583415 LR 0.001000 Time 0.020512 -2023-02-13 17:23:16,953 - Epoch: [10][ 420/ 1207] Overall Loss 0.584638 Objective Loss 0.584638 LR 0.001000 Time 0.020472 -2023-02-13 17:23:17,141 - Epoch: [10][ 430/ 1207] Overall Loss 0.585220 Objective Loss 0.585220 LR 0.001000 Time 0.020431 -2023-02-13 17:23:17,330 - Epoch: [10][ 440/ 1207] Overall Loss 0.585921 Objective Loss 0.585921 LR 0.001000 Time 0.020395 -2023-02-13 17:23:17,517 - Epoch: [10][ 450/ 1207] Overall Loss 0.586342 Objective Loss 0.586342 LR 0.001000 Time 0.020358 -2023-02-13 17:23:17,706 - Epoch: [10][ 460/ 1207] Overall Loss 0.586290 Objective Loss 0.586290 LR 0.001000 Time 0.020325 -2023-02-13 17:23:17,895 - Epoch: [10][ 470/ 1207] Overall Loss 0.586771 Objective Loss 0.586771 LR 0.001000 Time 0.020293 -2023-02-13 17:23:18,083 - Epoch: [10][ 480/ 1207] Overall Loss 0.588133 Objective Loss 0.588133 LR 0.001000 Time 0.020262 -2023-02-13 17:23:18,271 - Epoch: [10][ 490/ 1207] Overall Loss 0.588145 Objective Loss 0.588145 LR 0.001000 Time 0.020231 -2023-02-13 17:23:18,459 - Epoch: [10][ 500/ 1207] Overall Loss 0.588134 Objective Loss 0.588134 LR 0.001000 Time 0.020202 -2023-02-13 17:23:18,647 - Epoch: [10][ 510/ 1207] Overall Loss 0.588343 Objective Loss 0.588343 LR 0.001000 Time 0.020174 -2023-02-13 17:23:18,837 - Epoch: [10][ 520/ 1207] Overall Loss 0.588526 Objective Loss 0.588526 LR 0.001000 Time 0.020150 -2023-02-13 17:23:19,025 - Epoch: [10][ 530/ 1207] Overall Loss 0.587744 Objective Loss 0.587744 LR 0.001000 Time 0.020124 -2023-02-13 17:23:19,214 - Epoch: [10][ 540/ 1207] Overall Loss 0.587468 Objective Loss 0.587468 LR 0.001000 Time 0.020101 -2023-02-13 17:23:19,402 - Epoch: [10][ 550/ 1207] Overall Loss 0.587354 Objective Loss 0.587354 LR 0.001000 Time 0.020077 -2023-02-13 17:23:19,590 - Epoch: [10][ 560/ 1207] Overall Loss 0.587182 Objective Loss 0.587182 LR 0.001000 Time 0.020054 -2023-02-13 17:23:19,779 - Epoch: [10][ 570/ 1207] Overall Loss 0.586863 Objective Loss 0.586863 LR 0.001000 Time 0.020032 -2023-02-13 17:23:19,967 - Epoch: [10][ 580/ 1207] Overall Loss 0.585717 Objective Loss 0.585717 LR 0.001000 Time 0.020011 -2023-02-13 17:23:20,155 - Epoch: [10][ 590/ 1207] Overall Loss 0.585120 Objective Loss 0.585120 LR 0.001000 Time 0.019990 -2023-02-13 17:23:20,344 - Epoch: [10][ 600/ 1207] Overall Loss 0.584797 Objective Loss 0.584797 LR 0.001000 Time 0.019970 -2023-02-13 17:23:20,532 - Epoch: [10][ 610/ 1207] Overall Loss 0.584114 Objective Loss 0.584114 LR 0.001000 Time 0.019952 -2023-02-13 17:23:20,721 - Epoch: [10][ 620/ 1207] Overall Loss 0.583656 Objective Loss 0.583656 LR 0.001000 Time 0.019933 -2023-02-13 17:23:20,911 - Epoch: [10][ 630/ 1207] Overall Loss 0.582993 Objective Loss 0.582993 LR 0.001000 Time 0.019917 -2023-02-13 17:23:21,099 - Epoch: [10][ 640/ 1207] Overall Loss 0.581952 Objective Loss 0.581952 LR 0.001000 Time 0.019901 -2023-02-13 17:23:21,287 - Epoch: [10][ 650/ 1207] Overall Loss 0.580896 Objective Loss 0.580896 LR 0.001000 Time 0.019883 -2023-02-13 17:23:21,476 - Epoch: [10][ 660/ 1207] Overall Loss 0.580845 Objective Loss 0.580845 LR 0.001000 Time 0.019867 -2023-02-13 17:23:21,664 - Epoch: [10][ 670/ 1207] Overall Loss 0.580179 Objective Loss 0.580179 LR 0.001000 Time 0.019851 -2023-02-13 17:23:21,854 - Epoch: [10][ 680/ 1207] Overall Loss 0.579192 Objective Loss 0.579192 LR 0.001000 Time 0.019837 -2023-02-13 17:23:22,042 - Epoch: [10][ 690/ 1207] Overall Loss 0.578597 Objective Loss 0.578597 LR 0.001000 Time 0.019822 -2023-02-13 17:23:22,230 - Epoch: [10][ 700/ 1207] Overall Loss 0.577958 Objective Loss 0.577958 LR 0.001000 Time 0.019807 -2023-02-13 17:23:22,418 - Epoch: [10][ 710/ 1207] Overall Loss 0.577633 Objective Loss 0.577633 LR 0.001000 Time 0.019793 -2023-02-13 17:23:22,607 - Epoch: [10][ 720/ 1207] Overall Loss 0.576848 Objective Loss 0.576848 LR 0.001000 Time 0.019779 -2023-02-13 17:23:22,796 - Epoch: [10][ 730/ 1207] Overall Loss 0.576251 Objective Loss 0.576251 LR 0.001000 Time 0.019766 -2023-02-13 17:23:22,984 - Epoch: [10][ 740/ 1207] Overall Loss 0.575254 Objective Loss 0.575254 LR 0.001000 Time 0.019753 -2023-02-13 17:23:23,172 - Epoch: [10][ 750/ 1207] Overall Loss 0.574755 Objective Loss 0.574755 LR 0.001000 Time 0.019740 -2023-02-13 17:23:23,360 - Epoch: [10][ 760/ 1207] Overall Loss 0.574311 Objective Loss 0.574311 LR 0.001000 Time 0.019728 -2023-02-13 17:23:23,549 - Epoch: [10][ 770/ 1207] Overall Loss 0.574217 Objective Loss 0.574217 LR 0.001000 Time 0.019715 -2023-02-13 17:23:23,738 - Epoch: [10][ 780/ 1207] Overall Loss 0.573569 Objective Loss 0.573569 LR 0.001000 Time 0.019705 -2023-02-13 17:23:23,926 - Epoch: [10][ 790/ 1207] Overall Loss 0.572798 Objective Loss 0.572798 LR 0.001000 Time 0.019693 -2023-02-13 17:23:24,114 - Epoch: [10][ 800/ 1207] Overall Loss 0.572636 Objective Loss 0.572636 LR 0.001000 Time 0.019682 -2023-02-13 17:23:24,303 - Epoch: [10][ 810/ 1207] Overall Loss 0.572076 Objective Loss 0.572076 LR 0.001000 Time 0.019671 -2023-02-13 17:23:24,491 - Epoch: [10][ 820/ 1207] Overall Loss 0.571367 Objective Loss 0.571367 LR 0.001000 Time 0.019661 -2023-02-13 17:23:24,680 - Epoch: [10][ 830/ 1207] Overall Loss 0.570650 Objective Loss 0.570650 LR 0.001000 Time 0.019651 -2023-02-13 17:23:24,869 - Epoch: [10][ 840/ 1207] Overall Loss 0.569755 Objective Loss 0.569755 LR 0.001000 Time 0.019641 -2023-02-13 17:23:25,057 - Epoch: [10][ 850/ 1207] Overall Loss 0.569271 Objective Loss 0.569271 LR 0.001000 Time 0.019631 -2023-02-13 17:23:25,246 - Epoch: [10][ 860/ 1207] Overall Loss 0.568343 Objective Loss 0.568343 LR 0.001000 Time 0.019622 -2023-02-13 17:23:25,434 - Epoch: [10][ 870/ 1207] Overall Loss 0.568064 Objective Loss 0.568064 LR 0.001000 Time 0.019612 -2023-02-13 17:23:25,622 - Epoch: [10][ 880/ 1207] Overall Loss 0.567367 Objective Loss 0.567367 LR 0.001000 Time 0.019603 -2023-02-13 17:23:25,812 - Epoch: [10][ 890/ 1207] Overall Loss 0.567303 Objective Loss 0.567303 LR 0.001000 Time 0.019595 -2023-02-13 17:23:26,000 - Epoch: [10][ 900/ 1207] Overall Loss 0.567095 Objective Loss 0.567095 LR 0.001000 Time 0.019586 -2023-02-13 17:23:26,188 - Epoch: [10][ 910/ 1207] Overall Loss 0.566680 Objective Loss 0.566680 LR 0.001000 Time 0.019577 -2023-02-13 17:23:26,377 - Epoch: [10][ 920/ 1207] Overall Loss 0.565977 Objective Loss 0.565977 LR 0.001000 Time 0.019569 -2023-02-13 17:23:26,565 - Epoch: [10][ 930/ 1207] Overall Loss 0.565625 Objective Loss 0.565625 LR 0.001000 Time 0.019561 -2023-02-13 17:23:26,754 - Epoch: [10][ 940/ 1207] Overall Loss 0.565187 Objective Loss 0.565187 LR 0.001000 Time 0.019553 -2023-02-13 17:23:26,942 - Epoch: [10][ 950/ 1207] Overall Loss 0.564993 Objective Loss 0.564993 LR 0.001000 Time 0.019545 -2023-02-13 17:23:27,130 - Epoch: [10][ 960/ 1207] Overall Loss 0.564689 Objective Loss 0.564689 LR 0.001000 Time 0.019536 -2023-02-13 17:23:27,318 - Epoch: [10][ 970/ 1207] Overall Loss 0.564323 Objective Loss 0.564323 LR 0.001000 Time 0.019529 -2023-02-13 17:23:27,506 - Epoch: [10][ 980/ 1207] Overall Loss 0.563925 Objective Loss 0.563925 LR 0.001000 Time 0.019521 -2023-02-13 17:23:27,694 - Epoch: [10][ 990/ 1207] Overall Loss 0.563247 Objective Loss 0.563247 LR 0.001000 Time 0.019514 -2023-02-13 17:23:27,884 - Epoch: [10][ 1000/ 1207] Overall Loss 0.562826 Objective Loss 0.562826 LR 0.001000 Time 0.019508 -2023-02-13 17:23:28,072 - Epoch: [10][ 1010/ 1207] Overall Loss 0.562511 Objective Loss 0.562511 LR 0.001000 Time 0.019501 -2023-02-13 17:23:28,261 - Epoch: [10][ 1020/ 1207] Overall Loss 0.561877 Objective Loss 0.561877 LR 0.001000 Time 0.019494 -2023-02-13 17:23:28,449 - Epoch: [10][ 1030/ 1207] Overall Loss 0.561500 Objective Loss 0.561500 LR 0.001000 Time 0.019487 -2023-02-13 17:23:28,637 - Epoch: [10][ 1040/ 1207] Overall Loss 0.561076 Objective Loss 0.561076 LR 0.001000 Time 0.019480 -2023-02-13 17:23:28,826 - Epoch: [10][ 1050/ 1207] Overall Loss 0.560953 Objective Loss 0.560953 LR 0.001000 Time 0.019474 -2023-02-13 17:23:29,014 - Epoch: [10][ 1060/ 1207] Overall Loss 0.559989 Objective Loss 0.559989 LR 0.001000 Time 0.019467 -2023-02-13 17:23:29,202 - Epoch: [10][ 1070/ 1207] Overall Loss 0.559795 Objective Loss 0.559795 LR 0.001000 Time 0.019461 -2023-02-13 17:23:29,391 - Epoch: [10][ 1080/ 1207] Overall Loss 0.559281 Objective Loss 0.559281 LR 0.001000 Time 0.019455 -2023-02-13 17:23:29,579 - Epoch: [10][ 1090/ 1207] Overall Loss 0.559088 Objective Loss 0.559088 LR 0.001000 Time 0.019449 -2023-02-13 17:23:29,769 - Epoch: [10][ 1100/ 1207] Overall Loss 0.558705 Objective Loss 0.558705 LR 0.001000 Time 0.019444 -2023-02-13 17:23:29,957 - Epoch: [10][ 1110/ 1207] Overall Loss 0.558155 Objective Loss 0.558155 LR 0.001000 Time 0.019439 -2023-02-13 17:23:30,145 - Epoch: [10][ 1120/ 1207] Overall Loss 0.557837 Objective Loss 0.557837 LR 0.001000 Time 0.019433 -2023-02-13 17:23:30,335 - Epoch: [10][ 1130/ 1207] Overall Loss 0.557300 Objective Loss 0.557300 LR 0.001000 Time 0.019428 -2023-02-13 17:23:30,523 - Epoch: [10][ 1140/ 1207] Overall Loss 0.556934 Objective Loss 0.556934 LR 0.001000 Time 0.019422 -2023-02-13 17:23:30,712 - Epoch: [10][ 1150/ 1207] Overall Loss 0.556350 Objective Loss 0.556350 LR 0.001000 Time 0.019417 -2023-02-13 17:23:30,902 - Epoch: [10][ 1160/ 1207] Overall Loss 0.556117 Objective Loss 0.556117 LR 0.001000 Time 0.019414 -2023-02-13 17:23:31,092 - Epoch: [10][ 1170/ 1207] Overall Loss 0.555700 Objective Loss 0.555700 LR 0.001000 Time 0.019410 -2023-02-13 17:23:31,282 - Epoch: [10][ 1180/ 1207] Overall Loss 0.555565 Objective Loss 0.555565 LR 0.001000 Time 0.019406 -2023-02-13 17:23:31,473 - Epoch: [10][ 1190/ 1207] Overall Loss 0.555148 Objective Loss 0.555148 LR 0.001000 Time 0.019403 -2023-02-13 17:23:31,715 - Epoch: [10][ 1200/ 1207] Overall Loss 0.554663 Objective Loss 0.554663 LR 0.001000 Time 0.019443 -2023-02-13 17:23:31,832 - Epoch: [10][ 1207/ 1207] Overall Loss 0.554161 Objective Loss 0.554161 Top1 78.963415 Top5 96.951220 LR 0.001000 Time 0.019427 -2023-02-13 17:23:31,903 - --- validate (epoch=10)----------- -2023-02-13 17:23:31,903 - 34311 samples (256 per mini-batch) -2023-02-13 17:23:32,301 - Epoch: [10][ 10/ 135] Loss 0.483539 Top1 76.757812 Top5 96.367188 -2023-02-13 17:23:32,423 - Epoch: [10][ 20/ 135] Loss 0.476341 Top1 76.757812 Top5 96.816406 -2023-02-13 17:23:32,548 - Epoch: [10][ 30/ 135] Loss 0.488785 Top1 76.276042 Top5 96.627604 -2023-02-13 17:23:32,678 - Epoch: [10][ 40/ 135] Loss 0.494924 Top1 76.259766 Top5 96.591797 -2023-02-13 17:23:32,806 - Epoch: [10][ 50/ 135] Loss 0.497609 Top1 76.453125 Top5 96.523438 -2023-02-13 17:23:32,936 - Epoch: [10][ 60/ 135] Loss 0.500698 Top1 76.302083 Top5 96.451823 -2023-02-13 17:23:33,064 - Epoch: [10][ 70/ 135] Loss 0.502037 Top1 76.294643 Top5 96.445312 -2023-02-13 17:23:33,193 - Epoch: [10][ 80/ 135] Loss 0.500735 Top1 76.372070 Top5 96.367188 -2023-02-13 17:23:33,321 - Epoch: [10][ 90/ 135] Loss 0.494895 Top1 76.440972 Top5 96.406250 -2023-02-13 17:23:33,449 - Epoch: [10][ 100/ 135] Loss 0.498220 Top1 76.250000 Top5 96.253906 -2023-02-13 17:23:33,574 - Epoch: [10][ 110/ 135] Loss 0.500727 Top1 76.168324 Top5 96.267756 -2023-02-13 17:23:33,704 - Epoch: [10][ 120/ 135] Loss 0.503301 Top1 76.035156 Top5 96.269531 -2023-02-13 17:23:33,834 - Epoch: [10][ 130/ 135] Loss 0.502750 Top1 76.009615 Top5 96.298077 -2023-02-13 17:23:33,877 - Epoch: [10][ 135/ 135] Loss 0.504537 Top1 75.963977 Top5 96.301478 -2023-02-13 17:23:33,947 - ==> Top1: 75.964 Top5: 96.301 Loss: 0.505 - -2023-02-13 17:23:33,948 - ==> Confusion: -[[ 756 2 21 1 14 3 0 2 6 107 0 13 1 6 7 2 5 2 6 2 11] - [ 3 920 2 1 11 18 3 25 8 2 6 2 0 1 0 0 5 2 9 9 6] - [ 6 4 908 4 4 0 48 23 2 1 5 5 4 2 1 6 4 3 10 5 13] - [ 1 2 34 821 1 6 6 6 3 1 16 5 9 2 20 6 6 9 45 1 16] - [ 14 16 3 1 964 3 1 1 2 6 1 9 4 6 7 4 9 2 2 2 9] - [ 3 73 2 5 7 836 9 35 2 5 3 36 7 24 0 1 3 2 1 8 8] - [ 2 5 14 0 0 3 1031 7 0 3 3 1 3 1 0 3 2 2 5 13 1] - [ 0 9 12 0 3 23 8 885 1 2 5 7 6 1 0 0 0 2 42 13 5] - [ 17 5 0 1 3 0 0 2 854 57 13 2 2 25 19 0 0 1 5 0 3] - [ 83 0 7 0 4 3 2 0 33 807 0 6 0 47 3 0 2 2 2 1 10] - [ 2 9 3 7 0 1 7 6 11 3 947 7 1 7 3 0 1 1 27 3 5] - [ 3 5 0 0 1 6 3 1 2 3 0 916 24 8 0 3 4 6 3 15 2] - [ 4 1 1 5 2 4 0 2 1 0 0 84 807 0 6 5 4 22 4 2 5] - [ 5 5 2 1 7 8 2 2 10 15 9 22 1 908 3 3 4 0 4 11 2] - [ 17 11 1 24 14 2 0 0 26 7 0 2 8 9 908 1 7 6 28 1 20] - [ 2 3 4 1 6 0 11 1 0 1 0 32 12 3 1 927 11 13 0 8 10] - [ 3 10 2 1 11 2 1 1 3 1 1 5 3 2 2 18 968 3 3 4 17] - [ 5 2 4 2 1 0 0 1 3 0 1 24 40 1 2 12 0 940 1 4 8] - [ 2 9 3 7 0 1 0 45 6 1 4 4 6 1 14 1 2 1 975 3 1] - [ 0 5 0 0 2 9 15 20 0 0 1 30 7 4 0 4 2 1 3 1037 8] - [ 180 459 333 106 176 218 190 285 131 153 242 320 468 486 206 162 351 105 374 540 7949]] - -2023-02-13 17:23:33,949 - ==> Best [Top1: 75.964 Top5: 96.301 Sparsity:0.00 Params: 148928 on epoch: 10] -2023-02-13 17:23:33,949 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:23:33,955 - - -2023-02-13 17:23:33,956 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:23:34,934 - Epoch: [11][ 10/ 1207] Overall Loss 0.518027 Objective Loss 0.518027 LR 0.001000 Time 0.097805 -2023-02-13 17:23:35,130 - Epoch: [11][ 20/ 1207] Overall Loss 0.511971 Objective Loss 0.511971 LR 0.001000 Time 0.058681 -2023-02-13 17:23:35,319 - Epoch: [11][ 30/ 1207] Overall Loss 0.489671 Objective Loss 0.489671 LR 0.001000 Time 0.045398 -2023-02-13 17:23:35,506 - Epoch: [11][ 40/ 1207] Overall Loss 0.493497 Objective Loss 0.493497 LR 0.001000 Time 0.038727 -2023-02-13 17:23:35,695 - Epoch: [11][ 50/ 1207] Overall Loss 0.481797 Objective Loss 0.481797 LR 0.001000 Time 0.034744 -2023-02-13 17:23:35,884 - Epoch: [11][ 60/ 1207] Overall Loss 0.486211 Objective Loss 0.486211 LR 0.001000 Time 0.032103 -2023-02-13 17:23:36,073 - Epoch: [11][ 70/ 1207] Overall Loss 0.481351 Objective Loss 0.481351 LR 0.001000 Time 0.030201 -2023-02-13 17:23:36,261 - Epoch: [11][ 80/ 1207] Overall Loss 0.478768 Objective Loss 0.478768 LR 0.001000 Time 0.028771 -2023-02-13 17:23:36,449 - Epoch: [11][ 90/ 1207] Overall Loss 0.480523 Objective Loss 0.480523 LR 0.001000 Time 0.027668 -2023-02-13 17:23:36,638 - Epoch: [11][ 100/ 1207] Overall Loss 0.479765 Objective Loss 0.479765 LR 0.001000 Time 0.026782 -2023-02-13 17:23:36,828 - Epoch: [11][ 110/ 1207] Overall Loss 0.477203 Objective Loss 0.477203 LR 0.001000 Time 0.026069 -2023-02-13 17:23:37,016 - Epoch: [11][ 120/ 1207] Overall Loss 0.478770 Objective Loss 0.478770 LR 0.001000 Time 0.025463 -2023-02-13 17:23:37,205 - Epoch: [11][ 130/ 1207] Overall Loss 0.478170 Objective Loss 0.478170 LR 0.001000 Time 0.024955 -2023-02-13 17:23:37,393 - Epoch: [11][ 140/ 1207] Overall Loss 0.481107 Objective Loss 0.481107 LR 0.001000 Time 0.024513 -2023-02-13 17:23:37,582 - Epoch: [11][ 150/ 1207] Overall Loss 0.480526 Objective Loss 0.480526 LR 0.001000 Time 0.024136 -2023-02-13 17:23:37,771 - Epoch: [11][ 160/ 1207] Overall Loss 0.479609 Objective Loss 0.479609 LR 0.001000 Time 0.023806 -2023-02-13 17:23:37,960 - Epoch: [11][ 170/ 1207] Overall Loss 0.476868 Objective Loss 0.476868 LR 0.001000 Time 0.023514 -2023-02-13 17:23:38,148 - Epoch: [11][ 180/ 1207] Overall Loss 0.477071 Objective Loss 0.477071 LR 0.001000 Time 0.023254 -2023-02-13 17:23:38,337 - Epoch: [11][ 190/ 1207] Overall Loss 0.476511 Objective Loss 0.476511 LR 0.001000 Time 0.023023 -2023-02-13 17:23:38,526 - Epoch: [11][ 200/ 1207] Overall Loss 0.479227 Objective Loss 0.479227 LR 0.001000 Time 0.022813 -2023-02-13 17:23:38,715 - Epoch: [11][ 210/ 1207] Overall Loss 0.479815 Objective Loss 0.479815 LR 0.001000 Time 0.022623 -2023-02-13 17:23:38,904 - Epoch: [11][ 220/ 1207] Overall Loss 0.479715 Objective Loss 0.479715 LR 0.001000 Time 0.022453 -2023-02-13 17:23:39,093 - Epoch: [11][ 230/ 1207] Overall Loss 0.477975 Objective Loss 0.477975 LR 0.001000 Time 0.022297 -2023-02-13 17:23:39,281 - Epoch: [11][ 240/ 1207] Overall Loss 0.478012 Objective Loss 0.478012 LR 0.001000 Time 0.022150 -2023-02-13 17:23:39,470 - Epoch: [11][ 250/ 1207] Overall Loss 0.478182 Objective Loss 0.478182 LR 0.001000 Time 0.022019 -2023-02-13 17:23:39,659 - Epoch: [11][ 260/ 1207] Overall Loss 0.477943 Objective Loss 0.477943 LR 0.001000 Time 0.021896 -2023-02-13 17:23:39,848 - Epoch: [11][ 270/ 1207] Overall Loss 0.479436 Objective Loss 0.479436 LR 0.001000 Time 0.021787 -2023-02-13 17:23:40,037 - Epoch: [11][ 280/ 1207] Overall Loss 0.478625 Objective Loss 0.478625 LR 0.001000 Time 0.021681 -2023-02-13 17:23:40,226 - Epoch: [11][ 290/ 1207] Overall Loss 0.479171 Objective Loss 0.479171 LR 0.001000 Time 0.021583 -2023-02-13 17:23:40,414 - Epoch: [11][ 300/ 1207] Overall Loss 0.478582 Objective Loss 0.478582 LR 0.001000 Time 0.021491 -2023-02-13 17:23:40,603 - Epoch: [11][ 310/ 1207] Overall Loss 0.478189 Objective Loss 0.478189 LR 0.001000 Time 0.021405 -2023-02-13 17:23:40,792 - Epoch: [11][ 320/ 1207] Overall Loss 0.478692 Objective Loss 0.478692 LR 0.001000 Time 0.021326 -2023-02-13 17:23:40,981 - Epoch: [11][ 330/ 1207] Overall Loss 0.478913 Objective Loss 0.478913 LR 0.001000 Time 0.021252 -2023-02-13 17:23:41,170 - Epoch: [11][ 340/ 1207] Overall Loss 0.478686 Objective Loss 0.478686 LR 0.001000 Time 0.021181 -2023-02-13 17:23:41,358 - Epoch: [11][ 350/ 1207] Overall Loss 0.479822 Objective Loss 0.479822 LR 0.001000 Time 0.021113 -2023-02-13 17:23:41,551 - Epoch: [11][ 360/ 1207] Overall Loss 0.480265 Objective Loss 0.480265 LR 0.001000 Time 0.021062 -2023-02-13 17:23:41,750 - Epoch: [11][ 370/ 1207] Overall Loss 0.480071 Objective Loss 0.480071 LR 0.001000 Time 0.021029 -2023-02-13 17:23:41,954 - Epoch: [11][ 380/ 1207] Overall Loss 0.480062 Objective Loss 0.480062 LR 0.001000 Time 0.021011 -2023-02-13 17:23:42,154 - Epoch: [11][ 390/ 1207] Overall Loss 0.479299 Objective Loss 0.479299 LR 0.001000 Time 0.020982 -2023-02-13 17:23:42,357 - Epoch: [11][ 400/ 1207] Overall Loss 0.479309 Objective Loss 0.479309 LR 0.001000 Time 0.020965 -2023-02-13 17:23:42,556 - Epoch: [11][ 410/ 1207] Overall Loss 0.479192 Objective Loss 0.479192 LR 0.001000 Time 0.020938 -2023-02-13 17:23:42,760 - Epoch: [11][ 420/ 1207] Overall Loss 0.479542 Objective Loss 0.479542 LR 0.001000 Time 0.020924 -2023-02-13 17:23:42,959 - Epoch: [11][ 430/ 1207] Overall Loss 0.479527 Objective Loss 0.479527 LR 0.001000 Time 0.020900 -2023-02-13 17:23:43,163 - Epoch: [11][ 440/ 1207] Overall Loss 0.479865 Objective Loss 0.479865 LR 0.001000 Time 0.020887 -2023-02-13 17:23:43,363 - Epoch: [11][ 450/ 1207] Overall Loss 0.480258 Objective Loss 0.480258 LR 0.001000 Time 0.020867 -2023-02-13 17:23:43,566 - Epoch: [11][ 460/ 1207] Overall Loss 0.480594 Objective Loss 0.480594 LR 0.001000 Time 0.020853 -2023-02-13 17:23:43,764 - Epoch: [11][ 470/ 1207] Overall Loss 0.480572 Objective Loss 0.480572 LR 0.001000 Time 0.020831 -2023-02-13 17:23:43,968 - Epoch: [11][ 480/ 1207] Overall Loss 0.481238 Objective Loss 0.481238 LR 0.001000 Time 0.020820 -2023-02-13 17:23:44,167 - Epoch: [11][ 490/ 1207] Overall Loss 0.481596 Objective Loss 0.481596 LR 0.001000 Time 0.020800 -2023-02-13 17:23:44,370 - Epoch: [11][ 500/ 1207] Overall Loss 0.482125 Objective Loss 0.482125 LR 0.001000 Time 0.020790 -2023-02-13 17:23:44,569 - Epoch: [11][ 510/ 1207] Overall Loss 0.482560 Objective Loss 0.482560 LR 0.001000 Time 0.020771 -2023-02-13 17:23:44,772 - Epoch: [11][ 520/ 1207] Overall Loss 0.482673 Objective Loss 0.482673 LR 0.001000 Time 0.020761 -2023-02-13 17:23:44,972 - Epoch: [11][ 530/ 1207] Overall Loss 0.483341 Objective Loss 0.483341 LR 0.001000 Time 0.020747 -2023-02-13 17:23:45,176 - Epoch: [11][ 540/ 1207] Overall Loss 0.484577 Objective Loss 0.484577 LR 0.001000 Time 0.020739 -2023-02-13 17:23:45,375 - Epoch: [11][ 550/ 1207] Overall Loss 0.484569 Objective Loss 0.484569 LR 0.001000 Time 0.020725 -2023-02-13 17:23:45,578 - Epoch: [11][ 560/ 1207] Overall Loss 0.484445 Objective Loss 0.484445 LR 0.001000 Time 0.020716 -2023-02-13 17:23:45,775 - Epoch: [11][ 570/ 1207] Overall Loss 0.484143 Objective Loss 0.484143 LR 0.001000 Time 0.020697 -2023-02-13 17:23:45,964 - Epoch: [11][ 580/ 1207] Overall Loss 0.484107 Objective Loss 0.484107 LR 0.001000 Time 0.020666 -2023-02-13 17:23:46,154 - Epoch: [11][ 590/ 1207] Overall Loss 0.484010 Objective Loss 0.484010 LR 0.001000 Time 0.020637 -2023-02-13 17:23:46,344 - Epoch: [11][ 600/ 1207] Overall Loss 0.484680 Objective Loss 0.484680 LR 0.001000 Time 0.020608 -2023-02-13 17:23:46,533 - Epoch: [11][ 610/ 1207] Overall Loss 0.484856 Objective Loss 0.484856 LR 0.001000 Time 0.020580 -2023-02-13 17:23:46,723 - Epoch: [11][ 620/ 1207] Overall Loss 0.485211 Objective Loss 0.485211 LR 0.001000 Time 0.020554 -2023-02-13 17:23:46,914 - Epoch: [11][ 630/ 1207] Overall Loss 0.485532 Objective Loss 0.485532 LR 0.001000 Time 0.020530 -2023-02-13 17:23:47,103 - Epoch: [11][ 640/ 1207] Overall Loss 0.484850 Objective Loss 0.484850 LR 0.001000 Time 0.020505 -2023-02-13 17:23:47,294 - Epoch: [11][ 650/ 1207] Overall Loss 0.484965 Objective Loss 0.484965 LR 0.001000 Time 0.020482 -2023-02-13 17:23:47,484 - Epoch: [11][ 660/ 1207] Overall Loss 0.485911 Objective Loss 0.485911 LR 0.001000 Time 0.020459 -2023-02-13 17:23:47,673 - Epoch: [11][ 670/ 1207] Overall Loss 0.486214 Objective Loss 0.486214 LR 0.001000 Time 0.020435 -2023-02-13 17:23:47,863 - Epoch: [11][ 680/ 1207] Overall Loss 0.485555 Objective Loss 0.485555 LR 0.001000 Time 0.020414 -2023-02-13 17:23:48,053 - Epoch: [11][ 690/ 1207] Overall Loss 0.485082 Objective Loss 0.485082 LR 0.001000 Time 0.020393 -2023-02-13 17:23:48,243 - Epoch: [11][ 700/ 1207] Overall Loss 0.485205 Objective Loss 0.485205 LR 0.001000 Time 0.020372 -2023-02-13 17:23:48,432 - Epoch: [11][ 710/ 1207] Overall Loss 0.485452 Objective Loss 0.485452 LR 0.001000 Time 0.020351 -2023-02-13 17:23:48,622 - Epoch: [11][ 720/ 1207] Overall Loss 0.485763 Objective Loss 0.485763 LR 0.001000 Time 0.020332 -2023-02-13 17:23:48,812 - Epoch: [11][ 730/ 1207] Overall Loss 0.485835 Objective Loss 0.485835 LR 0.001000 Time 0.020313 -2023-02-13 17:23:49,002 - Epoch: [11][ 740/ 1207] Overall Loss 0.486281 Objective Loss 0.486281 LR 0.001000 Time 0.020295 -2023-02-13 17:23:49,192 - Epoch: [11][ 750/ 1207] Overall Loss 0.486061 Objective Loss 0.486061 LR 0.001000 Time 0.020277 -2023-02-13 17:23:49,385 - Epoch: [11][ 760/ 1207] Overall Loss 0.485535 Objective Loss 0.485535 LR 0.001000 Time 0.020263 -2023-02-13 17:23:49,576 - Epoch: [11][ 770/ 1207] Overall Loss 0.485510 Objective Loss 0.485510 LR 0.001000 Time 0.020248 -2023-02-13 17:23:49,768 - Epoch: [11][ 780/ 1207] Overall Loss 0.485633 Objective Loss 0.485633 LR 0.001000 Time 0.020234 -2023-02-13 17:23:49,961 - Epoch: [11][ 790/ 1207] Overall Loss 0.485837 Objective Loss 0.485837 LR 0.001000 Time 0.020222 -2023-02-13 17:23:50,153 - Epoch: [11][ 800/ 1207] Overall Loss 0.485977 Objective Loss 0.485977 LR 0.001000 Time 0.020209 -2023-02-13 17:23:50,345 - Epoch: [11][ 810/ 1207] Overall Loss 0.485885 Objective Loss 0.485885 LR 0.001000 Time 0.020195 -2023-02-13 17:23:50,537 - Epoch: [11][ 820/ 1207] Overall Loss 0.485844 Objective Loss 0.485844 LR 0.001000 Time 0.020183 -2023-02-13 17:23:50,729 - Epoch: [11][ 830/ 1207] Overall Loss 0.485645 Objective Loss 0.485645 LR 0.001000 Time 0.020171 -2023-02-13 17:23:50,923 - Epoch: [11][ 840/ 1207] Overall Loss 0.485875 Objective Loss 0.485875 LR 0.001000 Time 0.020161 -2023-02-13 17:23:51,114 - Epoch: [11][ 850/ 1207] Overall Loss 0.486147 Objective Loss 0.486147 LR 0.001000 Time 0.020148 -2023-02-13 17:23:51,307 - Epoch: [11][ 860/ 1207] Overall Loss 0.485866 Objective Loss 0.485866 LR 0.001000 Time 0.020138 -2023-02-13 17:23:51,499 - Epoch: [11][ 870/ 1207] Overall Loss 0.485420 Objective Loss 0.485420 LR 0.001000 Time 0.020127 -2023-02-13 17:23:51,692 - Epoch: [11][ 880/ 1207] Overall Loss 0.485098 Objective Loss 0.485098 LR 0.001000 Time 0.020116 -2023-02-13 17:23:51,885 - Epoch: [11][ 890/ 1207] Overall Loss 0.484577 Objective Loss 0.484577 LR 0.001000 Time 0.020107 -2023-02-13 17:23:52,078 - Epoch: [11][ 900/ 1207] Overall Loss 0.483984 Objective Loss 0.483984 LR 0.001000 Time 0.020097 -2023-02-13 17:23:52,269 - Epoch: [11][ 910/ 1207] Overall Loss 0.483910 Objective Loss 0.483910 LR 0.001000 Time 0.020086 -2023-02-13 17:23:52,462 - Epoch: [11][ 920/ 1207] Overall Loss 0.483903 Objective Loss 0.483903 LR 0.001000 Time 0.020077 -2023-02-13 17:23:52,653 - Epoch: [11][ 930/ 1207] Overall Loss 0.484406 Objective Loss 0.484406 LR 0.001000 Time 0.020066 -2023-02-13 17:23:52,845 - Epoch: [11][ 940/ 1207] Overall Loss 0.484323 Objective Loss 0.484323 LR 0.001000 Time 0.020057 -2023-02-13 17:23:53,037 - Epoch: [11][ 950/ 1207] Overall Loss 0.484400 Objective Loss 0.484400 LR 0.001000 Time 0.020047 -2023-02-13 17:23:53,229 - Epoch: [11][ 960/ 1207] Overall Loss 0.484171 Objective Loss 0.484171 LR 0.001000 Time 0.020038 -2023-02-13 17:23:53,421 - Epoch: [11][ 970/ 1207] Overall Loss 0.483813 Objective Loss 0.483813 LR 0.001000 Time 0.020029 -2023-02-13 17:23:53,613 - Epoch: [11][ 980/ 1207] Overall Loss 0.483823 Objective Loss 0.483823 LR 0.001000 Time 0.020020 -2023-02-13 17:23:53,805 - Epoch: [11][ 990/ 1207] Overall Loss 0.483454 Objective Loss 0.483454 LR 0.001000 Time 0.020011 -2023-02-13 17:23:53,998 - Epoch: [11][ 1000/ 1207] Overall Loss 0.483320 Objective Loss 0.483320 LR 0.001000 Time 0.020004 -2023-02-13 17:23:54,189 - Epoch: [11][ 1010/ 1207] Overall Loss 0.483557 Objective Loss 0.483557 LR 0.001000 Time 0.019995 -2023-02-13 17:23:54,382 - Epoch: [11][ 1020/ 1207] Overall Loss 0.483266 Objective Loss 0.483266 LR 0.001000 Time 0.019987 -2023-02-13 17:23:54,573 - Epoch: [11][ 1030/ 1207] Overall Loss 0.482735 Objective Loss 0.482735 LR 0.001000 Time 0.019979 -2023-02-13 17:23:54,766 - Epoch: [11][ 1040/ 1207] Overall Loss 0.482833 Objective Loss 0.482833 LR 0.001000 Time 0.019972 -2023-02-13 17:23:54,959 - Epoch: [11][ 1050/ 1207] Overall Loss 0.482879 Objective Loss 0.482879 LR 0.001000 Time 0.019964 -2023-02-13 17:23:55,151 - Epoch: [11][ 1060/ 1207] Overall Loss 0.482619 Objective Loss 0.482619 LR 0.001000 Time 0.019957 -2023-02-13 17:23:55,343 - Epoch: [11][ 1070/ 1207] Overall Loss 0.482138 Objective Loss 0.482138 LR 0.001000 Time 0.019950 -2023-02-13 17:23:55,535 - Epoch: [11][ 1080/ 1207] Overall Loss 0.482343 Objective Loss 0.482343 LR 0.001000 Time 0.019943 -2023-02-13 17:23:55,728 - Epoch: [11][ 1090/ 1207] Overall Loss 0.482283 Objective Loss 0.482283 LR 0.001000 Time 0.019936 -2023-02-13 17:23:55,921 - Epoch: [11][ 1100/ 1207] Overall Loss 0.482253 Objective Loss 0.482253 LR 0.001000 Time 0.019930 -2023-02-13 17:23:56,113 - Epoch: [11][ 1110/ 1207] Overall Loss 0.482381 Objective Loss 0.482381 LR 0.001000 Time 0.019923 -2023-02-13 17:23:56,305 - Epoch: [11][ 1120/ 1207] Overall Loss 0.482617 Objective Loss 0.482617 LR 0.001000 Time 0.019916 -2023-02-13 17:23:56,497 - Epoch: [11][ 1130/ 1207] Overall Loss 0.482571 Objective Loss 0.482571 LR 0.001000 Time 0.019910 -2023-02-13 17:23:56,689 - Epoch: [11][ 1140/ 1207] Overall Loss 0.482545 Objective Loss 0.482545 LR 0.001000 Time 0.019903 -2023-02-13 17:23:56,882 - Epoch: [11][ 1150/ 1207] Overall Loss 0.482696 Objective Loss 0.482696 LR 0.001000 Time 0.019898 -2023-02-13 17:23:57,074 - Epoch: [11][ 1160/ 1207] Overall Loss 0.482845 Objective Loss 0.482845 LR 0.001000 Time 0.019892 -2023-02-13 17:23:57,267 - Epoch: [11][ 1170/ 1207] Overall Loss 0.482939 Objective Loss 0.482939 LR 0.001000 Time 0.019886 -2023-02-13 17:23:57,458 - Epoch: [11][ 1180/ 1207] Overall Loss 0.482895 Objective Loss 0.482895 LR 0.001000 Time 0.019879 -2023-02-13 17:23:57,651 - Epoch: [11][ 1190/ 1207] Overall Loss 0.482733 Objective Loss 0.482733 LR 0.001000 Time 0.019874 -2023-02-13 17:23:57,896 - Epoch: [11][ 1200/ 1207] Overall Loss 0.482321 Objective Loss 0.482321 LR 0.001000 Time 0.019912 -2023-02-13 17:23:58,011 - Epoch: [11][ 1207/ 1207] Overall Loss 0.482241 Objective Loss 0.482241 Top1 75.609756 Top5 98.170732 LR 0.001000 Time 0.019892 -2023-02-13 17:23:58,080 - --- validate (epoch=11)----------- -2023-02-13 17:23:58,081 - 34311 samples (256 per mini-batch) -2023-02-13 17:23:58,495 - Epoch: [11][ 10/ 135] Loss 0.503071 Top1 74.960938 Top5 95.820312 -2023-02-13 17:23:58,629 - Epoch: [11][ 20/ 135] Loss 0.501217 Top1 74.941406 Top5 95.957031 -2023-02-13 17:23:58,757 - Epoch: [11][ 30/ 135] Loss 0.488234 Top1 74.817708 Top5 96.080729 -2023-02-13 17:23:58,890 - Epoch: [11][ 40/ 135] Loss 0.474415 Top1 75.087891 Top5 96.220703 -2023-02-13 17:23:59,020 - Epoch: [11][ 50/ 135] Loss 0.476909 Top1 74.976562 Top5 96.242188 -2023-02-13 17:23:59,146 - Epoch: [11][ 60/ 135] Loss 0.477684 Top1 74.986979 Top5 96.250000 -2023-02-13 17:23:59,285 - Epoch: [11][ 70/ 135] Loss 0.483443 Top1 74.860491 Top5 96.166295 -2023-02-13 17:23:59,422 - Epoch: [11][ 80/ 135] Loss 0.485519 Top1 74.750977 Top5 96.118164 -2023-02-13 17:23:59,549 - Epoch: [11][ 90/ 135] Loss 0.487460 Top1 74.687500 Top5 96.111111 -2023-02-13 17:23:59,675 - Epoch: [11][ 100/ 135] Loss 0.487366 Top1 74.761719 Top5 96.097656 -2023-02-13 17:23:59,804 - Epoch: [11][ 110/ 135] Loss 0.485396 Top1 74.872159 Top5 96.111506 -2023-02-13 17:23:59,934 - Epoch: [11][ 120/ 135] Loss 0.482886 Top1 74.863281 Top5 96.139323 -2023-02-13 17:24:00,067 - Epoch: [11][ 130/ 135] Loss 0.481190 Top1 74.858774 Top5 96.099760 -2023-02-13 17:24:00,114 - Epoch: [11][ 135/ 135] Loss 0.480333 Top1 74.862289 Top5 96.100376 -2023-02-13 17:24:00,181 - ==> Top1: 74.862 Top5: 96.100 Loss: 0.480 - -2023-02-13 17:24:00,182 - ==> Confusion: -[[ 849 2 7 2 7 2 0 6 3 60 0 8 0 3 6 2 2 1 1 0 6] - [ 6 884 2 5 16 47 0 32 4 1 3 2 1 1 3 0 8 0 7 3 8] - [ 19 3 938 21 3 5 3 22 1 2 3 4 2 1 1 6 4 3 8 3 6] - [ 8 0 24 856 0 7 0 4 5 1 16 3 10 1 27 1 8 6 30 1 8] - [ 28 12 3 2 974 9 1 4 1 5 0 5 0 1 6 1 4 1 2 1 6] - [ 5 27 6 7 6 938 2 34 1 9 1 12 1 5 2 0 3 1 2 5 3] - [ 5 3 84 3 2 13 918 18 0 0 3 5 3 1 0 6 5 2 1 23 4] - [ 5 13 8 5 3 24 1 915 3 1 2 4 6 0 1 0 0 0 23 8 2] - [ 25 3 1 1 0 0 0 1 874 55 9 2 2 7 22 0 2 1 3 0 1] - [ 151 2 2 0 7 3 0 3 35 774 1 3 0 13 8 0 0 3 1 0 6] - [ 5 3 10 18 2 3 0 6 23 3 935 3 1 9 5 0 1 0 20 0 4] - [ 4 3 1 0 1 27 0 8 2 5 0 873 40 3 2 3 3 5 3 20 2] - [ 3 0 1 2 1 4 0 0 2 4 0 67 832 0 2 2 5 19 3 2 10] - [ 14 2 4 2 4 47 0 6 22 26 7 20 5 834 13 0 8 0 0 7 3] - [ 26 3 2 14 9 2 0 1 39 4 4 7 2 0 947 0 6 3 15 0 8] - [ 8 2 12 0 10 7 5 0 0 4 0 11 14 2 1 914 25 13 0 8 10] - [ 4 8 1 2 12 4 1 0 3 2 1 7 0 3 3 6 992 2 3 3 4] - [ 14 1 3 4 1 2 0 1 4 5 2 41 53 1 2 15 0 885 2 9 6] - [ 9 9 11 11 0 2 0 43 3 1 3 5 7 0 18 0 1 1 958 1 3] - [ 1 2 0 3 2 14 2 19 0 0 0 20 8 4 0 1 7 2 0 1057 6] - [ 407 321 334 205 262 397 54 292 152 171 156 221 459 355 291 115 739 86 327 551 7539]] - -2023-02-13 17:24:00,184 - ==> Best [Top1: 75.964 Top5: 96.301 Sparsity:0.00 Params: 148928 on epoch: 10] -2023-02-13 17:24:00,184 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:24:00,190 - - -2023-02-13 17:24:00,190 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:24:01,075 - Epoch: [12][ 10/ 1207] Overall Loss 0.484985 Objective Loss 0.484985 LR 0.001000 Time 0.088451 -2023-02-13 17:24:01,268 - Epoch: [12][ 20/ 1207] Overall Loss 0.464647 Objective Loss 0.464647 LR 0.001000 Time 0.053836 -2023-02-13 17:24:01,457 - Epoch: [12][ 30/ 1207] Overall Loss 0.483227 Objective Loss 0.483227 LR 0.001000 Time 0.042185 -2023-02-13 17:24:01,647 - Epoch: [12][ 40/ 1207] Overall Loss 0.489985 Objective Loss 0.489985 LR 0.001000 Time 0.036372 -2023-02-13 17:24:01,836 - Epoch: [12][ 50/ 1207] Overall Loss 0.494358 Objective Loss 0.494358 LR 0.001000 Time 0.032886 -2023-02-13 17:24:02,026 - Epoch: [12][ 60/ 1207] Overall Loss 0.489686 Objective Loss 0.489686 LR 0.001000 Time 0.030561 -2023-02-13 17:24:02,215 - Epoch: [12][ 70/ 1207] Overall Loss 0.487296 Objective Loss 0.487296 LR 0.001000 Time 0.028888 -2023-02-13 17:24:02,403 - Epoch: [12][ 80/ 1207] Overall Loss 0.483237 Objective Loss 0.483237 LR 0.001000 Time 0.027621 -2023-02-13 17:24:02,592 - Epoch: [12][ 90/ 1207] Overall Loss 0.478326 Objective Loss 0.478326 LR 0.001000 Time 0.026649 -2023-02-13 17:24:02,782 - Epoch: [12][ 100/ 1207] Overall Loss 0.478221 Objective Loss 0.478221 LR 0.001000 Time 0.025875 -2023-02-13 17:24:02,972 - Epoch: [12][ 110/ 1207] Overall Loss 0.473572 Objective Loss 0.473572 LR 0.001000 Time 0.025249 -2023-02-13 17:24:03,160 - Epoch: [12][ 120/ 1207] Overall Loss 0.475060 Objective Loss 0.475060 LR 0.001000 Time 0.024712 -2023-02-13 17:24:03,350 - Epoch: [12][ 130/ 1207] Overall Loss 0.470896 Objective Loss 0.470896 LR 0.001000 Time 0.024264 -2023-02-13 17:24:03,539 - Epoch: [12][ 140/ 1207] Overall Loss 0.468300 Objective Loss 0.468300 LR 0.001000 Time 0.023878 -2023-02-13 17:24:03,728 - Epoch: [12][ 150/ 1207] Overall Loss 0.468332 Objective Loss 0.468332 LR 0.001000 Time 0.023548 -2023-02-13 17:24:03,917 - Epoch: [12][ 160/ 1207] Overall Loss 0.467093 Objective Loss 0.467093 LR 0.001000 Time 0.023252 -2023-02-13 17:24:04,105 - Epoch: [12][ 170/ 1207] Overall Loss 0.467199 Objective Loss 0.467199 LR 0.001000 Time 0.022988 -2023-02-13 17:24:04,292 - Epoch: [12][ 180/ 1207] Overall Loss 0.468393 Objective Loss 0.468393 LR 0.001000 Time 0.022750 -2023-02-13 17:24:04,481 - Epoch: [12][ 190/ 1207] Overall Loss 0.466375 Objective Loss 0.466375 LR 0.001000 Time 0.022544 -2023-02-13 17:24:04,668 - Epoch: [12][ 200/ 1207] Overall Loss 0.466099 Objective Loss 0.466099 LR 0.001000 Time 0.022352 -2023-02-13 17:24:04,856 - Epoch: [12][ 210/ 1207] Overall Loss 0.467961 Objective Loss 0.467961 LR 0.001000 Time 0.022183 -2023-02-13 17:24:05,046 - Epoch: [12][ 220/ 1207] Overall Loss 0.467753 Objective Loss 0.467753 LR 0.001000 Time 0.022032 -2023-02-13 17:24:05,233 - Epoch: [12][ 230/ 1207] Overall Loss 0.466204 Objective Loss 0.466204 LR 0.001000 Time 0.021889 -2023-02-13 17:24:05,421 - Epoch: [12][ 240/ 1207] Overall Loss 0.465356 Objective Loss 0.465356 LR 0.001000 Time 0.021758 -2023-02-13 17:24:05,610 - Epoch: [12][ 250/ 1207] Overall Loss 0.466790 Objective Loss 0.466790 LR 0.001000 Time 0.021641 -2023-02-13 17:24:05,799 - Epoch: [12][ 260/ 1207] Overall Loss 0.466095 Objective Loss 0.466095 LR 0.001000 Time 0.021536 -2023-02-13 17:24:05,988 - Epoch: [12][ 270/ 1207] Overall Loss 0.465441 Objective Loss 0.465441 LR 0.001000 Time 0.021436 -2023-02-13 17:24:06,175 - Epoch: [12][ 280/ 1207] Overall Loss 0.464519 Objective Loss 0.464519 LR 0.001000 Time 0.021337 -2023-02-13 17:24:06,364 - Epoch: [12][ 290/ 1207] Overall Loss 0.464716 Objective Loss 0.464716 LR 0.001000 Time 0.021252 -2023-02-13 17:24:06,552 - Epoch: [12][ 300/ 1207] Overall Loss 0.464211 Objective Loss 0.464211 LR 0.001000 Time 0.021167 -2023-02-13 17:24:06,740 - Epoch: [12][ 310/ 1207] Overall Loss 0.464275 Objective Loss 0.464275 LR 0.001000 Time 0.021090 -2023-02-13 17:24:06,929 - Epoch: [12][ 320/ 1207] Overall Loss 0.463319 Objective Loss 0.463319 LR 0.001000 Time 0.021022 -2023-02-13 17:24:07,117 - Epoch: [12][ 330/ 1207] Overall Loss 0.462646 Objective Loss 0.462646 LR 0.001000 Time 0.020954 -2023-02-13 17:24:07,305 - Epoch: [12][ 340/ 1207] Overall Loss 0.463153 Objective Loss 0.463153 LR 0.001000 Time 0.020888 -2023-02-13 17:24:07,493 - Epoch: [12][ 350/ 1207] Overall Loss 0.463057 Objective Loss 0.463057 LR 0.001000 Time 0.020827 -2023-02-13 17:24:07,679 - Epoch: [12][ 360/ 1207] Overall Loss 0.462529 Objective Loss 0.462529 LR 0.001000 Time 0.020766 -2023-02-13 17:24:07,869 - Epoch: [12][ 370/ 1207] Overall Loss 0.461986 Objective Loss 0.461986 LR 0.001000 Time 0.020715 -2023-02-13 17:24:08,057 - Epoch: [12][ 380/ 1207] Overall Loss 0.461858 Objective Loss 0.461858 LR 0.001000 Time 0.020663 -2023-02-13 17:24:08,244 - Epoch: [12][ 390/ 1207] Overall Loss 0.461234 Objective Loss 0.461234 LR 0.001000 Time 0.020612 -2023-02-13 17:24:08,431 - Epoch: [12][ 400/ 1207] Overall Loss 0.461401 Objective Loss 0.461401 LR 0.001000 Time 0.020564 -2023-02-13 17:24:08,619 - Epoch: [12][ 410/ 1207] Overall Loss 0.461300 Objective Loss 0.461300 LR 0.001000 Time 0.020519 -2023-02-13 17:24:08,806 - Epoch: [12][ 420/ 1207] Overall Loss 0.461959 Objective Loss 0.461959 LR 0.001000 Time 0.020476 -2023-02-13 17:24:08,994 - Epoch: [12][ 430/ 1207] Overall Loss 0.461123 Objective Loss 0.461123 LR 0.001000 Time 0.020437 -2023-02-13 17:24:09,182 - Epoch: [12][ 440/ 1207] Overall Loss 0.461652 Objective Loss 0.461652 LR 0.001000 Time 0.020399 -2023-02-13 17:24:09,371 - Epoch: [12][ 450/ 1207] Overall Loss 0.461760 Objective Loss 0.461760 LR 0.001000 Time 0.020365 -2023-02-13 17:24:09,559 - Epoch: [12][ 460/ 1207] Overall Loss 0.461268 Objective Loss 0.461268 LR 0.001000 Time 0.020328 -2023-02-13 17:24:09,747 - Epoch: [12][ 470/ 1207] Overall Loss 0.460391 Objective Loss 0.460391 LR 0.001000 Time 0.020295 -2023-02-13 17:24:09,936 - Epoch: [12][ 480/ 1207] Overall Loss 0.459887 Objective Loss 0.459887 LR 0.001000 Time 0.020266 -2023-02-13 17:24:10,124 - Epoch: [12][ 490/ 1207] Overall Loss 0.459905 Objective Loss 0.459905 LR 0.001000 Time 0.020235 -2023-02-13 17:24:10,312 - Epoch: [12][ 500/ 1207] Overall Loss 0.460490 Objective Loss 0.460490 LR 0.001000 Time 0.020206 -2023-02-13 17:24:10,501 - Epoch: [12][ 510/ 1207] Overall Loss 0.461000 Objective Loss 0.461000 LR 0.001000 Time 0.020179 -2023-02-13 17:24:10,688 - Epoch: [12][ 520/ 1207] Overall Loss 0.461276 Objective Loss 0.461276 LR 0.001000 Time 0.020150 -2023-02-13 17:24:10,878 - Epoch: [12][ 530/ 1207] Overall Loss 0.460916 Objective Loss 0.460916 LR 0.001000 Time 0.020128 -2023-02-13 17:24:11,065 - Epoch: [12][ 540/ 1207] Overall Loss 0.460771 Objective Loss 0.460771 LR 0.001000 Time 0.020101 -2023-02-13 17:24:11,254 - Epoch: [12][ 550/ 1207] Overall Loss 0.461334 Objective Loss 0.461334 LR 0.001000 Time 0.020078 -2023-02-13 17:24:11,443 - Epoch: [12][ 560/ 1207] Overall Loss 0.461702 Objective Loss 0.461702 LR 0.001000 Time 0.020056 -2023-02-13 17:24:11,632 - Epoch: [12][ 570/ 1207] Overall Loss 0.462201 Objective Loss 0.462201 LR 0.001000 Time 0.020035 -2023-02-13 17:24:11,820 - Epoch: [12][ 580/ 1207] Overall Loss 0.462675 Objective Loss 0.462675 LR 0.001000 Time 0.020014 -2023-02-13 17:24:12,009 - Epoch: [12][ 590/ 1207] Overall Loss 0.462634 Objective Loss 0.462634 LR 0.001000 Time 0.019994 -2023-02-13 17:24:12,195 - Epoch: [12][ 600/ 1207] Overall Loss 0.463168 Objective Loss 0.463168 LR 0.001000 Time 0.019971 -2023-02-13 17:24:12,385 - Epoch: [12][ 610/ 1207] Overall Loss 0.464070 Objective Loss 0.464070 LR 0.001000 Time 0.019953 -2023-02-13 17:24:12,572 - Epoch: [12][ 620/ 1207] Overall Loss 0.464331 Objective Loss 0.464331 LR 0.001000 Time 0.019933 -2023-02-13 17:24:12,761 - Epoch: [12][ 630/ 1207] Overall Loss 0.464303 Objective Loss 0.464303 LR 0.001000 Time 0.019916 -2023-02-13 17:24:12,949 - Epoch: [12][ 640/ 1207] Overall Loss 0.464189 Objective Loss 0.464189 LR 0.001000 Time 0.019898 -2023-02-13 17:24:13,137 - Epoch: [12][ 650/ 1207] Overall Loss 0.463884 Objective Loss 0.463884 LR 0.001000 Time 0.019881 -2023-02-13 17:24:13,325 - Epoch: [12][ 660/ 1207] Overall Loss 0.463594 Objective Loss 0.463594 LR 0.001000 Time 0.019864 -2023-02-13 17:24:13,516 - Epoch: [12][ 670/ 1207] Overall Loss 0.463262 Objective Loss 0.463262 LR 0.001000 Time 0.019852 -2023-02-13 17:24:13,705 - Epoch: [12][ 680/ 1207] Overall Loss 0.463464 Objective Loss 0.463464 LR 0.001000 Time 0.019838 -2023-02-13 17:24:13,896 - Epoch: [12][ 690/ 1207] Overall Loss 0.463381 Objective Loss 0.463381 LR 0.001000 Time 0.019826 -2023-02-13 17:24:14,085 - Epoch: [12][ 700/ 1207] Overall Loss 0.463309 Objective Loss 0.463309 LR 0.001000 Time 0.019813 -2023-02-13 17:24:14,276 - Epoch: [12][ 710/ 1207] Overall Loss 0.463588 Objective Loss 0.463588 LR 0.001000 Time 0.019801 -2023-02-13 17:24:14,465 - Epoch: [12][ 720/ 1207] Overall Loss 0.463946 Objective Loss 0.463946 LR 0.001000 Time 0.019788 -2023-02-13 17:24:14,655 - Epoch: [12][ 730/ 1207] Overall Loss 0.463677 Objective Loss 0.463677 LR 0.001000 Time 0.019777 -2023-02-13 17:24:14,843 - Epoch: [12][ 740/ 1207] Overall Loss 0.463762 Objective Loss 0.463762 LR 0.001000 Time 0.019764 -2023-02-13 17:24:15,032 - Epoch: [12][ 750/ 1207] Overall Loss 0.463886 Objective Loss 0.463886 LR 0.001000 Time 0.019751 -2023-02-13 17:24:15,220 - Epoch: [12][ 760/ 1207] Overall Loss 0.463494 Objective Loss 0.463494 LR 0.001000 Time 0.019738 -2023-02-13 17:24:15,409 - Epoch: [12][ 770/ 1207] Overall Loss 0.463328 Objective Loss 0.463328 LR 0.001000 Time 0.019728 -2023-02-13 17:24:15,597 - Epoch: [12][ 780/ 1207] Overall Loss 0.463769 Objective Loss 0.463769 LR 0.001000 Time 0.019715 -2023-02-13 17:24:15,786 - Epoch: [12][ 790/ 1207] Overall Loss 0.464243 Objective Loss 0.464243 LR 0.001000 Time 0.019704 -2023-02-13 17:24:15,975 - Epoch: [12][ 800/ 1207] Overall Loss 0.463287 Objective Loss 0.463287 LR 0.001000 Time 0.019693 -2023-02-13 17:24:16,162 - Epoch: [12][ 810/ 1207] Overall Loss 0.462770 Objective Loss 0.462770 LR 0.001000 Time 0.019681 -2023-02-13 17:24:16,351 - Epoch: [12][ 820/ 1207] Overall Loss 0.462675 Objective Loss 0.462675 LR 0.001000 Time 0.019671 -2023-02-13 17:24:16,539 - Epoch: [12][ 830/ 1207] Overall Loss 0.462270 Objective Loss 0.462270 LR 0.001000 Time 0.019660 -2023-02-13 17:24:16,727 - Epoch: [12][ 840/ 1207] Overall Loss 0.462264 Objective Loss 0.462264 LR 0.001000 Time 0.019649 -2023-02-13 17:24:16,917 - Epoch: [12][ 850/ 1207] Overall Loss 0.461824 Objective Loss 0.461824 LR 0.001000 Time 0.019641 -2023-02-13 17:24:17,107 - Epoch: [12][ 860/ 1207] Overall Loss 0.461994 Objective Loss 0.461994 LR 0.001000 Time 0.019633 -2023-02-13 17:24:17,297 - Epoch: [12][ 870/ 1207] Overall Loss 0.462048 Objective Loss 0.462048 LR 0.001000 Time 0.019626 -2023-02-13 17:24:17,487 - Epoch: [12][ 880/ 1207] Overall Loss 0.461932 Objective Loss 0.461932 LR 0.001000 Time 0.019618 -2023-02-13 17:24:17,677 - Epoch: [12][ 890/ 1207] Overall Loss 0.461770 Objective Loss 0.461770 LR 0.001000 Time 0.019611 -2023-02-13 17:24:17,867 - Epoch: [12][ 900/ 1207] Overall Loss 0.461676 Objective Loss 0.461676 LR 0.001000 Time 0.019603 -2023-02-13 17:24:18,058 - Epoch: [12][ 910/ 1207] Overall Loss 0.461464 Objective Loss 0.461464 LR 0.001000 Time 0.019598 -2023-02-13 17:24:18,247 - Epoch: [12][ 920/ 1207] Overall Loss 0.461283 Objective Loss 0.461283 LR 0.001000 Time 0.019590 -2023-02-13 17:24:18,439 - Epoch: [12][ 930/ 1207] Overall Loss 0.461381 Objective Loss 0.461381 LR 0.001000 Time 0.019584 -2023-02-13 17:24:18,627 - Epoch: [12][ 940/ 1207] Overall Loss 0.461253 Objective Loss 0.461253 LR 0.001000 Time 0.019576 -2023-02-13 17:24:18,815 - Epoch: [12][ 950/ 1207] Overall Loss 0.461275 Objective Loss 0.461275 LR 0.001000 Time 0.019568 -2023-02-13 17:24:19,005 - Epoch: [12][ 960/ 1207] Overall Loss 0.461076 Objective Loss 0.461076 LR 0.001000 Time 0.019561 -2023-02-13 17:24:19,193 - Epoch: [12][ 970/ 1207] Overall Loss 0.461160 Objective Loss 0.461160 LR 0.001000 Time 0.019553 -2023-02-13 17:24:19,381 - Epoch: [12][ 980/ 1207] Overall Loss 0.461535 Objective Loss 0.461535 LR 0.001000 Time 0.019545 -2023-02-13 17:24:19,570 - Epoch: [12][ 990/ 1207] Overall Loss 0.461290 Objective Loss 0.461290 LR 0.001000 Time 0.019537 -2023-02-13 17:24:19,758 - Epoch: [12][ 1000/ 1207] Overall Loss 0.461335 Objective Loss 0.461335 LR 0.001000 Time 0.019530 -2023-02-13 17:24:19,947 - Epoch: [12][ 1010/ 1207] Overall Loss 0.461062 Objective Loss 0.461062 LR 0.001000 Time 0.019523 -2023-02-13 17:24:20,135 - Epoch: [12][ 1020/ 1207] Overall Loss 0.461073 Objective Loss 0.461073 LR 0.001000 Time 0.019516 -2023-02-13 17:24:20,323 - Epoch: [12][ 1030/ 1207] Overall Loss 0.461092 Objective Loss 0.461092 LR 0.001000 Time 0.019509 -2023-02-13 17:24:20,511 - Epoch: [12][ 1040/ 1207] Overall Loss 0.460824 Objective Loss 0.460824 LR 0.001000 Time 0.019502 -2023-02-13 17:24:20,700 - Epoch: [12][ 1050/ 1207] Overall Loss 0.460631 Objective Loss 0.460631 LR 0.001000 Time 0.019495 -2023-02-13 17:24:20,889 - Epoch: [12][ 1060/ 1207] Overall Loss 0.460668 Objective Loss 0.460668 LR 0.001000 Time 0.019490 -2023-02-13 17:24:21,079 - Epoch: [12][ 1070/ 1207] Overall Loss 0.460541 Objective Loss 0.460541 LR 0.001000 Time 0.019484 -2023-02-13 17:24:21,267 - Epoch: [12][ 1080/ 1207] Overall Loss 0.460496 Objective Loss 0.460496 LR 0.001000 Time 0.019478 -2023-02-13 17:24:21,457 - Epoch: [12][ 1090/ 1207] Overall Loss 0.460521 Objective Loss 0.460521 LR 0.001000 Time 0.019473 -2023-02-13 17:24:21,645 - Epoch: [12][ 1100/ 1207] Overall Loss 0.460204 Objective Loss 0.460204 LR 0.001000 Time 0.019467 -2023-02-13 17:24:21,835 - Epoch: [12][ 1110/ 1207] Overall Loss 0.460103 Objective Loss 0.460103 LR 0.001000 Time 0.019462 -2023-02-13 17:24:22,025 - Epoch: [12][ 1120/ 1207] Overall Loss 0.460091 Objective Loss 0.460091 LR 0.001000 Time 0.019457 -2023-02-13 17:24:22,213 - Epoch: [12][ 1130/ 1207] Overall Loss 0.459742 Objective Loss 0.459742 LR 0.001000 Time 0.019452 -2023-02-13 17:24:22,402 - Epoch: [12][ 1140/ 1207] Overall Loss 0.459473 Objective Loss 0.459473 LR 0.001000 Time 0.019446 -2023-02-13 17:24:22,591 - Epoch: [12][ 1150/ 1207] Overall Loss 0.458763 Objective Loss 0.458763 LR 0.001000 Time 0.019441 -2023-02-13 17:24:22,779 - Epoch: [12][ 1160/ 1207] Overall Loss 0.458277 Objective Loss 0.458277 LR 0.001000 Time 0.019435 -2023-02-13 17:24:22,969 - Epoch: [12][ 1170/ 1207] Overall Loss 0.458212 Objective Loss 0.458212 LR 0.001000 Time 0.019431 -2023-02-13 17:24:23,157 - Epoch: [12][ 1180/ 1207] Overall Loss 0.457994 Objective Loss 0.457994 LR 0.001000 Time 0.019426 -2023-02-13 17:24:23,346 - Epoch: [12][ 1190/ 1207] Overall Loss 0.458054 Objective Loss 0.458054 LR 0.001000 Time 0.019421 -2023-02-13 17:24:23,591 - Epoch: [12][ 1200/ 1207] Overall Loss 0.457682 Objective Loss 0.457682 LR 0.001000 Time 0.019463 -2023-02-13 17:24:23,707 - Epoch: [12][ 1207/ 1207] Overall Loss 0.457635 Objective Loss 0.457635 Top1 74.390244 Top5 95.426829 LR 0.001000 Time 0.019446 -2023-02-13 17:24:23,778 - --- validate (epoch=12)----------- -2023-02-13 17:24:23,778 - 34311 samples (256 per mini-batch) -2023-02-13 17:24:24,182 - Epoch: [12][ 10/ 135] Loss 0.419881 Top1 79.179688 Top5 96.875000 -2023-02-13 17:24:24,308 - Epoch: [12][ 20/ 135] Loss 0.434279 Top1 79.492188 Top5 96.738281 -2023-02-13 17:24:24,441 - Epoch: [12][ 30/ 135] Loss 0.430985 Top1 79.114583 Top5 96.796875 -2023-02-13 17:24:24,567 - Epoch: [12][ 40/ 135] Loss 0.437265 Top1 78.955078 Top5 96.767578 -2023-02-13 17:24:24,693 - Epoch: [12][ 50/ 135] Loss 0.438765 Top1 79.031250 Top5 96.726562 -2023-02-13 17:24:24,817 - Epoch: [12][ 60/ 135] Loss 0.436798 Top1 79.042969 Top5 96.842448 -2023-02-13 17:24:24,947 - Epoch: [12][ 70/ 135] Loss 0.434814 Top1 79.135045 Top5 96.886161 -2023-02-13 17:24:25,089 - Epoch: [12][ 80/ 135] Loss 0.436592 Top1 79.062500 Top5 96.772461 -2023-02-13 17:24:25,232 - Epoch: [12][ 90/ 135] Loss 0.438757 Top1 79.149306 Top5 96.775174 -2023-02-13 17:24:25,373 - Epoch: [12][ 100/ 135] Loss 0.435977 Top1 79.140625 Top5 96.824219 -2023-02-13 17:24:25,511 - Epoch: [12][ 110/ 135] Loss 0.438762 Top1 79.083807 Top5 96.818182 -2023-02-13 17:24:25,639 - Epoch: [12][ 120/ 135] Loss 0.438944 Top1 79.085286 Top5 96.832682 -2023-02-13 17:24:25,766 - Epoch: [12][ 130/ 135] Loss 0.440140 Top1 79.167668 Top5 96.820913 -2023-02-13 17:24:25,810 - Epoch: [12][ 135/ 135] Loss 0.444203 Top1 79.140800 Top5 96.837749 -2023-02-13 17:24:25,882 - ==> Top1: 79.141 Top5: 96.838 Loss: 0.444 - -2023-02-13 17:24:25,883 - ==> Confusion: -[[ 862 0 9 4 4 3 0 2 2 44 1 5 1 3 4 3 3 5 1 1 10] - [ 3 874 2 2 16 65 3 16 7 1 1 4 5 2 1 1 3 0 9 4 14] - [ 11 3 931 10 3 3 19 22 1 1 4 2 5 2 1 7 2 6 3 6 16] - [ 7 0 35 865 0 4 2 1 3 3 12 0 9 2 19 2 6 13 18 1 14] - [ 24 11 0 3 968 10 0 2 2 7 0 5 3 1 5 8 2 2 1 3 9] - [ 2 28 1 3 2 950 3 16 2 7 2 12 8 5 2 4 2 2 2 9 8] - [ 6 3 22 2 1 8 1017 5 0 0 3 2 4 0 0 3 2 5 1 14 1] - [ 1 9 14 2 0 50 5 880 1 1 3 2 8 0 0 2 0 2 24 14 6] - [ 32 0 0 1 2 1 0 1 838 75 12 4 6 8 17 0 0 4 5 0 3] - [ 140 1 3 1 3 3 0 1 25 800 0 1 1 12 5 1 0 5 0 1 9] - [ 2 4 9 8 0 1 7 3 21 3 934 4 6 7 5 0 0 1 22 1 13] - [ 3 1 1 0 0 17 1 2 1 4 0 833 81 3 0 11 1 14 2 26 4] - [ 2 1 1 2 1 3 0 0 1 0 0 29 847 1 3 9 3 40 3 2 11] - [ 9 1 4 1 9 38 0 0 21 33 8 13 5 843 10 5 7 2 0 7 8] - [ 29 3 4 22 12 3 0 1 24 6 2 3 7 2 936 0 2 13 9 0 14] - [ 4 2 6 2 7 4 5 0 0 1 0 8 6 0 2 944 16 24 0 7 8] - [ 4 5 2 0 5 5 0 0 2 0 0 5 3 1 4 18 979 4 2 6 16] - [ 10 0 2 3 0 1 1 1 0 0 0 9 24 0 2 17 0 977 0 2 2] - [ 4 5 10 16 1 5 0 36 4 2 5 3 9 1 20 0 1 2 956 2 4] - [ 0 1 0 1 0 8 9 13 0 1 1 18 9 5 0 6 3 2 0 1064 7] - [ 303 264 285 145 159 355 110 202 105 153 170 149 472 300 165 156 253 161 213 458 8856]] - -2023-02-13 17:24:25,884 - ==> Best [Top1: 79.141 Top5: 96.838 Sparsity:0.00 Params: 148928 on epoch: 12] -2023-02-13 17:24:25,885 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:24:25,891 - - -2023-02-13 17:24:25,891 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:24:26,753 - Epoch: [13][ 10/ 1207] Overall Loss 0.438975 Objective Loss 0.438975 LR 0.001000 Time 0.086153 -2023-02-13 17:24:26,947 - Epoch: [13][ 20/ 1207] Overall Loss 0.445171 Objective Loss 0.445171 LR 0.001000 Time 0.052732 -2023-02-13 17:24:27,135 - Epoch: [13][ 30/ 1207] Overall Loss 0.445895 Objective Loss 0.445895 LR 0.001000 Time 0.041412 -2023-02-13 17:24:27,323 - Epoch: [13][ 40/ 1207] Overall Loss 0.439567 Objective Loss 0.439567 LR 0.001000 Time 0.035746 -2023-02-13 17:24:27,511 - Epoch: [13][ 50/ 1207] Overall Loss 0.436077 Objective Loss 0.436077 LR 0.001000 Time 0.032354 -2023-02-13 17:24:27,699 - Epoch: [13][ 60/ 1207] Overall Loss 0.433662 Objective Loss 0.433662 LR 0.001000 Time 0.030085 -2023-02-13 17:24:27,887 - Epoch: [13][ 70/ 1207] Overall Loss 0.435221 Objective Loss 0.435221 LR 0.001000 Time 0.028469 -2023-02-13 17:24:28,075 - Epoch: [13][ 80/ 1207] Overall Loss 0.436027 Objective Loss 0.436027 LR 0.001000 Time 0.027263 -2023-02-13 17:24:28,263 - Epoch: [13][ 90/ 1207] Overall Loss 0.436324 Objective Loss 0.436324 LR 0.001000 Time 0.026315 -2023-02-13 17:24:28,451 - Epoch: [13][ 100/ 1207] Overall Loss 0.441198 Objective Loss 0.441198 LR 0.001000 Time 0.025560 -2023-02-13 17:24:28,639 - Epoch: [13][ 110/ 1207] Overall Loss 0.442139 Objective Loss 0.442139 LR 0.001000 Time 0.024936 -2023-02-13 17:24:28,826 - Epoch: [13][ 120/ 1207] Overall Loss 0.441195 Objective Loss 0.441195 LR 0.001000 Time 0.024420 -2023-02-13 17:24:29,014 - Epoch: [13][ 130/ 1207] Overall Loss 0.438990 Objective Loss 0.438990 LR 0.001000 Time 0.023985 -2023-02-13 17:24:29,202 - Epoch: [13][ 140/ 1207] Overall Loss 0.439842 Objective Loss 0.439842 LR 0.001000 Time 0.023612 -2023-02-13 17:24:29,390 - Epoch: [13][ 150/ 1207] Overall Loss 0.438793 Objective Loss 0.438793 LR 0.001000 Time 0.023289 -2023-02-13 17:24:29,578 - Epoch: [13][ 160/ 1207] Overall Loss 0.438769 Objective Loss 0.438769 LR 0.001000 Time 0.023005 -2023-02-13 17:24:29,767 - Epoch: [13][ 170/ 1207] Overall Loss 0.440240 Objective Loss 0.440240 LR 0.001000 Time 0.022759 -2023-02-13 17:24:29,957 - Epoch: [13][ 180/ 1207] Overall Loss 0.440849 Objective Loss 0.440849 LR 0.001000 Time 0.022552 -2023-02-13 17:24:30,147 - Epoch: [13][ 190/ 1207] Overall Loss 0.440672 Objective Loss 0.440672 LR 0.001000 Time 0.022362 -2023-02-13 17:24:30,336 - Epoch: [13][ 200/ 1207] Overall Loss 0.439939 Objective Loss 0.439939 LR 0.001000 Time 0.022188 -2023-02-13 17:24:30,526 - Epoch: [13][ 210/ 1207] Overall Loss 0.439945 Objective Loss 0.439945 LR 0.001000 Time 0.022033 -2023-02-13 17:24:30,716 - Epoch: [13][ 220/ 1207] Overall Loss 0.438143 Objective Loss 0.438143 LR 0.001000 Time 0.021893 -2023-02-13 17:24:30,909 - Epoch: [13][ 230/ 1207] Overall Loss 0.438076 Objective Loss 0.438076 LR 0.001000 Time 0.021778 -2023-02-13 17:24:31,099 - Epoch: [13][ 240/ 1207] Overall Loss 0.438261 Objective Loss 0.438261 LR 0.001000 Time 0.021663 -2023-02-13 17:24:31,288 - Epoch: [13][ 250/ 1207] Overall Loss 0.437876 Objective Loss 0.437876 LR 0.001000 Time 0.021551 -2023-02-13 17:24:31,478 - Epoch: [13][ 260/ 1207] Overall Loss 0.437709 Objective Loss 0.437709 LR 0.001000 Time 0.021451 -2023-02-13 17:24:31,668 - Epoch: [13][ 270/ 1207] Overall Loss 0.437665 Objective Loss 0.437665 LR 0.001000 Time 0.021356 -2023-02-13 17:24:31,858 - Epoch: [13][ 280/ 1207] Overall Loss 0.438114 Objective Loss 0.438114 LR 0.001000 Time 0.021273 -2023-02-13 17:24:32,049 - Epoch: [13][ 290/ 1207] Overall Loss 0.439291 Objective Loss 0.439291 LR 0.001000 Time 0.021195 -2023-02-13 17:24:32,238 - Epoch: [13][ 300/ 1207] Overall Loss 0.441229 Objective Loss 0.441229 LR 0.001000 Time 0.021120 -2023-02-13 17:24:32,428 - Epoch: [13][ 310/ 1207] Overall Loss 0.441889 Objective Loss 0.441889 LR 0.001000 Time 0.021049 -2023-02-13 17:24:32,618 - Epoch: [13][ 320/ 1207] Overall Loss 0.442104 Objective Loss 0.442104 LR 0.001000 Time 0.020985 -2023-02-13 17:24:32,808 - Epoch: [13][ 330/ 1207] Overall Loss 0.441684 Objective Loss 0.441684 LR 0.001000 Time 0.020923 -2023-02-13 17:24:33,000 - Epoch: [13][ 340/ 1207] Overall Loss 0.441558 Objective Loss 0.441558 LR 0.001000 Time 0.020870 -2023-02-13 17:24:33,190 - Epoch: [13][ 350/ 1207] Overall Loss 0.442397 Objective Loss 0.442397 LR 0.001000 Time 0.020816 -2023-02-13 17:24:33,380 - Epoch: [13][ 360/ 1207] Overall Loss 0.442003 Objective Loss 0.442003 LR 0.001000 Time 0.020766 -2023-02-13 17:24:33,571 - Epoch: [13][ 370/ 1207] Overall Loss 0.441391 Objective Loss 0.441391 LR 0.001000 Time 0.020720 -2023-02-13 17:24:33,762 - Epoch: [13][ 380/ 1207] Overall Loss 0.441337 Objective Loss 0.441337 LR 0.001000 Time 0.020676 -2023-02-13 17:24:33,952 - Epoch: [13][ 390/ 1207] Overall Loss 0.440784 Objective Loss 0.440784 LR 0.001000 Time 0.020633 -2023-02-13 17:24:34,143 - Epoch: [13][ 400/ 1207] Overall Loss 0.440786 Objective Loss 0.440786 LR 0.001000 Time 0.020593 -2023-02-13 17:24:34,333 - Epoch: [13][ 410/ 1207] Overall Loss 0.440906 Objective Loss 0.440906 LR 0.001000 Time 0.020553 -2023-02-13 17:24:34,524 - Epoch: [13][ 420/ 1207] Overall Loss 0.441307 Objective Loss 0.441307 LR 0.001000 Time 0.020517 -2023-02-13 17:24:34,714 - Epoch: [13][ 430/ 1207] Overall Loss 0.441707 Objective Loss 0.441707 LR 0.001000 Time 0.020482 -2023-02-13 17:24:34,905 - Epoch: [13][ 440/ 1207] Overall Loss 0.442264 Objective Loss 0.442264 LR 0.001000 Time 0.020450 -2023-02-13 17:24:35,096 - Epoch: [13][ 450/ 1207] Overall Loss 0.441973 Objective Loss 0.441973 LR 0.001000 Time 0.020418 -2023-02-13 17:24:35,287 - Epoch: [13][ 460/ 1207] Overall Loss 0.441774 Objective Loss 0.441774 LR 0.001000 Time 0.020388 -2023-02-13 17:24:35,477 - Epoch: [13][ 470/ 1207] Overall Loss 0.441182 Objective Loss 0.441182 LR 0.001000 Time 0.020358 -2023-02-13 17:24:35,668 - Epoch: [13][ 480/ 1207] Overall Loss 0.440174 Objective Loss 0.440174 LR 0.001000 Time 0.020331 -2023-02-13 17:24:35,859 - Epoch: [13][ 490/ 1207] Overall Loss 0.439930 Objective Loss 0.439930 LR 0.001000 Time 0.020305 -2023-02-13 17:24:36,050 - Epoch: [13][ 500/ 1207] Overall Loss 0.440487 Objective Loss 0.440487 LR 0.001000 Time 0.020281 -2023-02-13 17:24:36,241 - Epoch: [13][ 510/ 1207] Overall Loss 0.439295 Objective Loss 0.439295 LR 0.001000 Time 0.020256 -2023-02-13 17:24:36,432 - Epoch: [13][ 520/ 1207] Overall Loss 0.439680 Objective Loss 0.439680 LR 0.001000 Time 0.020233 -2023-02-13 17:24:36,622 - Epoch: [13][ 530/ 1207] Overall Loss 0.439494 Objective Loss 0.439494 LR 0.001000 Time 0.020210 -2023-02-13 17:24:36,813 - Epoch: [13][ 540/ 1207] Overall Loss 0.439222 Objective Loss 0.439222 LR 0.001000 Time 0.020189 -2023-02-13 17:24:37,004 - Epoch: [13][ 550/ 1207] Overall Loss 0.440237 Objective Loss 0.440237 LR 0.001000 Time 0.020167 -2023-02-13 17:24:37,195 - Epoch: [13][ 560/ 1207] Overall Loss 0.440496 Objective Loss 0.440496 LR 0.001000 Time 0.020149 -2023-02-13 17:24:37,385 - Epoch: [13][ 570/ 1207] Overall Loss 0.440076 Objective Loss 0.440076 LR 0.001000 Time 0.020127 -2023-02-13 17:24:37,576 - Epoch: [13][ 580/ 1207] Overall Loss 0.439375 Objective Loss 0.439375 LR 0.001000 Time 0.020109 -2023-02-13 17:24:37,766 - Epoch: [13][ 590/ 1207] Overall Loss 0.439376 Objective Loss 0.439376 LR 0.001000 Time 0.020090 -2023-02-13 17:24:37,957 - Epoch: [13][ 600/ 1207] Overall Loss 0.439230 Objective Loss 0.439230 LR 0.001000 Time 0.020072 -2023-02-13 17:24:38,147 - Epoch: [13][ 610/ 1207] Overall Loss 0.439209 Objective Loss 0.439209 LR 0.001000 Time 0.020054 -2023-02-13 17:24:38,337 - Epoch: [13][ 620/ 1207] Overall Loss 0.438974 Objective Loss 0.438974 LR 0.001000 Time 0.020037 -2023-02-13 17:24:38,528 - Epoch: [13][ 630/ 1207] Overall Loss 0.439424 Objective Loss 0.439424 LR 0.001000 Time 0.020021 -2023-02-13 17:24:38,719 - Epoch: [13][ 640/ 1207] Overall Loss 0.439489 Objective Loss 0.439489 LR 0.001000 Time 0.020007 -2023-02-13 17:24:38,911 - Epoch: [13][ 650/ 1207] Overall Loss 0.439812 Objective Loss 0.439812 LR 0.001000 Time 0.019993 -2023-02-13 17:24:39,103 - Epoch: [13][ 660/ 1207] Overall Loss 0.439872 Objective Loss 0.439872 LR 0.001000 Time 0.019981 -2023-02-13 17:24:39,295 - Epoch: [13][ 670/ 1207] Overall Loss 0.439992 Objective Loss 0.439992 LR 0.001000 Time 0.019968 -2023-02-13 17:24:39,487 - Epoch: [13][ 680/ 1207] Overall Loss 0.439673 Objective Loss 0.439673 LR 0.001000 Time 0.019956 -2023-02-13 17:24:39,678 - Epoch: [13][ 690/ 1207] Overall Loss 0.439803 Objective Loss 0.439803 LR 0.001000 Time 0.019944 -2023-02-13 17:24:39,870 - Epoch: [13][ 700/ 1207] Overall Loss 0.439703 Objective Loss 0.439703 LR 0.001000 Time 0.019933 -2023-02-13 17:24:40,063 - Epoch: [13][ 710/ 1207] Overall Loss 0.439765 Objective Loss 0.439765 LR 0.001000 Time 0.019923 -2023-02-13 17:24:40,254 - Epoch: [13][ 720/ 1207] Overall Loss 0.439874 Objective Loss 0.439874 LR 0.001000 Time 0.019911 -2023-02-13 17:24:40,445 - Epoch: [13][ 730/ 1207] Overall Loss 0.439816 Objective Loss 0.439816 LR 0.001000 Time 0.019900 -2023-02-13 17:24:40,637 - Epoch: [13][ 740/ 1207] Overall Loss 0.439786 Objective Loss 0.439786 LR 0.001000 Time 0.019890 -2023-02-13 17:24:40,830 - Epoch: [13][ 750/ 1207] Overall Loss 0.439672 Objective Loss 0.439672 LR 0.001000 Time 0.019881 -2023-02-13 17:24:41,021 - Epoch: [13][ 760/ 1207] Overall Loss 0.439486 Objective Loss 0.439486 LR 0.001000 Time 0.019871 -2023-02-13 17:24:41,213 - Epoch: [13][ 770/ 1207] Overall Loss 0.439694 Objective Loss 0.439694 LR 0.001000 Time 0.019862 -2023-02-13 17:24:41,405 - Epoch: [13][ 780/ 1207] Overall Loss 0.439833 Objective Loss 0.439833 LR 0.001000 Time 0.019852 -2023-02-13 17:24:41,596 - Epoch: [13][ 790/ 1207] Overall Loss 0.439838 Objective Loss 0.439838 LR 0.001000 Time 0.019843 -2023-02-13 17:24:41,788 - Epoch: [13][ 800/ 1207] Overall Loss 0.439684 Objective Loss 0.439684 LR 0.001000 Time 0.019834 -2023-02-13 17:24:41,980 - Epoch: [13][ 810/ 1207] Overall Loss 0.440121 Objective Loss 0.440121 LR 0.001000 Time 0.019827 -2023-02-13 17:24:42,173 - Epoch: [13][ 820/ 1207] Overall Loss 0.440300 Objective Loss 0.440300 LR 0.001000 Time 0.019819 -2023-02-13 17:24:42,365 - Epoch: [13][ 830/ 1207] Overall Loss 0.440215 Objective Loss 0.440215 LR 0.001000 Time 0.019811 -2023-02-13 17:24:42,557 - Epoch: [13][ 840/ 1207] Overall Loss 0.440782 Objective Loss 0.440782 LR 0.001000 Time 0.019803 -2023-02-13 17:24:42,748 - Epoch: [13][ 850/ 1207] Overall Loss 0.441790 Objective Loss 0.441790 LR 0.001000 Time 0.019795 -2023-02-13 17:24:42,940 - Epoch: [13][ 860/ 1207] Overall Loss 0.441823 Objective Loss 0.441823 LR 0.001000 Time 0.019787 -2023-02-13 17:24:43,132 - Epoch: [13][ 870/ 1207] Overall Loss 0.441629 Objective Loss 0.441629 LR 0.001000 Time 0.019780 -2023-02-13 17:24:43,324 - Epoch: [13][ 880/ 1207] Overall Loss 0.441627 Objective Loss 0.441627 LR 0.001000 Time 0.019773 -2023-02-13 17:24:43,515 - Epoch: [13][ 890/ 1207] Overall Loss 0.441870 Objective Loss 0.441870 LR 0.001000 Time 0.019766 -2023-02-13 17:24:43,707 - Epoch: [13][ 900/ 1207] Overall Loss 0.441952 Objective Loss 0.441952 LR 0.001000 Time 0.019759 -2023-02-13 17:24:43,899 - Epoch: [13][ 910/ 1207] Overall Loss 0.441913 Objective Loss 0.441913 LR 0.001000 Time 0.019752 -2023-02-13 17:24:44,091 - Epoch: [13][ 920/ 1207] Overall Loss 0.441994 Objective Loss 0.441994 LR 0.001000 Time 0.019746 -2023-02-13 17:24:44,283 - Epoch: [13][ 930/ 1207] Overall Loss 0.442088 Objective Loss 0.442088 LR 0.001000 Time 0.019739 -2023-02-13 17:24:44,475 - Epoch: [13][ 940/ 1207] Overall Loss 0.442025 Objective Loss 0.442025 LR 0.001000 Time 0.019733 -2023-02-13 17:24:44,666 - Epoch: [13][ 950/ 1207] Overall Loss 0.441715 Objective Loss 0.441715 LR 0.001000 Time 0.019727 -2023-02-13 17:24:44,858 - Epoch: [13][ 960/ 1207] Overall Loss 0.441850 Objective Loss 0.441850 LR 0.001000 Time 0.019721 -2023-02-13 17:24:45,051 - Epoch: [13][ 970/ 1207] Overall Loss 0.441791 Objective Loss 0.441791 LR 0.001000 Time 0.019716 -2023-02-13 17:24:45,243 - Epoch: [13][ 980/ 1207] Overall Loss 0.441728 Objective Loss 0.441728 LR 0.001000 Time 0.019710 -2023-02-13 17:24:45,435 - Epoch: [13][ 990/ 1207] Overall Loss 0.441800 Objective Loss 0.441800 LR 0.001000 Time 0.019705 -2023-02-13 17:24:45,627 - Epoch: [13][ 1000/ 1207] Overall Loss 0.441723 Objective Loss 0.441723 LR 0.001000 Time 0.019699 -2023-02-13 17:24:45,820 - Epoch: [13][ 1010/ 1207] Overall Loss 0.441573 Objective Loss 0.441573 LR 0.001000 Time 0.019694 -2023-02-13 17:24:46,012 - Epoch: [13][ 1020/ 1207] Overall Loss 0.441235 Objective Loss 0.441235 LR 0.001000 Time 0.019690 -2023-02-13 17:24:46,204 - Epoch: [13][ 1030/ 1207] Overall Loss 0.441199 Objective Loss 0.441199 LR 0.001000 Time 0.019684 -2023-02-13 17:24:46,396 - Epoch: [13][ 1040/ 1207] Overall Loss 0.441191 Objective Loss 0.441191 LR 0.001000 Time 0.019679 -2023-02-13 17:24:46,588 - Epoch: [13][ 1050/ 1207] Overall Loss 0.441231 Objective Loss 0.441231 LR 0.001000 Time 0.019674 -2023-02-13 17:24:46,780 - Epoch: [13][ 1060/ 1207] Overall Loss 0.441050 Objective Loss 0.441050 LR 0.001000 Time 0.019669 -2023-02-13 17:24:46,972 - Epoch: [13][ 1070/ 1207] Overall Loss 0.441013 Objective Loss 0.441013 LR 0.001000 Time 0.019665 -2023-02-13 17:24:47,164 - Epoch: [13][ 1080/ 1207] Overall Loss 0.441382 Objective Loss 0.441382 LR 0.001000 Time 0.019660 -2023-02-13 17:24:47,356 - Epoch: [13][ 1090/ 1207] Overall Loss 0.441093 Objective Loss 0.441093 LR 0.001000 Time 0.019655 -2023-02-13 17:24:47,547 - Epoch: [13][ 1100/ 1207] Overall Loss 0.440995 Objective Loss 0.440995 LR 0.001000 Time 0.019650 -2023-02-13 17:24:47,739 - Epoch: [13][ 1110/ 1207] Overall Loss 0.440929 Objective Loss 0.440929 LR 0.001000 Time 0.019646 -2023-02-13 17:24:47,930 - Epoch: [13][ 1120/ 1207] Overall Loss 0.441128 Objective Loss 0.441128 LR 0.001000 Time 0.019641 -2023-02-13 17:24:48,123 - Epoch: [13][ 1130/ 1207] Overall Loss 0.441250 Objective Loss 0.441250 LR 0.001000 Time 0.019637 -2023-02-13 17:24:48,314 - Epoch: [13][ 1140/ 1207] Overall Loss 0.441359 Objective Loss 0.441359 LR 0.001000 Time 0.019632 -2023-02-13 17:24:48,506 - Epoch: [13][ 1150/ 1207] Overall Loss 0.441379 Objective Loss 0.441379 LR 0.001000 Time 0.019628 -2023-02-13 17:24:48,698 - Epoch: [13][ 1160/ 1207] Overall Loss 0.441434 Objective Loss 0.441434 LR 0.001000 Time 0.019624 -2023-02-13 17:24:48,891 - Epoch: [13][ 1170/ 1207] Overall Loss 0.441616 Objective Loss 0.441616 LR 0.001000 Time 0.019621 -2023-02-13 17:24:49,083 - Epoch: [13][ 1180/ 1207] Overall Loss 0.441443 Objective Loss 0.441443 LR 0.001000 Time 0.019617 -2023-02-13 17:24:49,275 - Epoch: [13][ 1190/ 1207] Overall Loss 0.441838 Objective Loss 0.441838 LR 0.001000 Time 0.019613 -2023-02-13 17:24:49,523 - Epoch: [13][ 1200/ 1207] Overall Loss 0.441581 Objective Loss 0.441581 LR 0.001000 Time 0.019656 -2023-02-13 17:24:49,637 - Epoch: [13][ 1207/ 1207] Overall Loss 0.441283 Objective Loss 0.441283 Top1 83.536585 Top5 96.646341 LR 0.001000 Time 0.019637 -2023-02-13 17:24:49,708 - --- validate (epoch=13)----------- -2023-02-13 17:24:49,709 - 34311 samples (256 per mini-batch) -2023-02-13 17:24:50,196 - Epoch: [13][ 10/ 135] Loss 0.400041 Top1 78.593750 Top5 96.796875 -2023-02-13 17:24:50,322 - Epoch: [13][ 20/ 135] Loss 0.402808 Top1 79.101562 Top5 96.562500 -2023-02-13 17:24:50,448 - Epoch: [13][ 30/ 135] Loss 0.415250 Top1 78.125000 Top5 96.536458 -2023-02-13 17:24:50,571 - Epoch: [13][ 40/ 135] Loss 0.413497 Top1 78.369141 Top5 96.591797 -2023-02-13 17:24:50,695 - Epoch: [13][ 50/ 135] Loss 0.416657 Top1 78.234375 Top5 96.523438 -2023-02-13 17:24:50,820 - Epoch: [13][ 60/ 135] Loss 0.421249 Top1 78.170573 Top5 96.516927 -2023-02-13 17:24:50,944 - Epoch: [13][ 70/ 135] Loss 0.426442 Top1 78.247768 Top5 96.534598 -2023-02-13 17:24:51,070 - Epoch: [13][ 80/ 135] Loss 0.425608 Top1 78.261719 Top5 96.567383 -2023-02-13 17:24:51,201 - Epoch: [13][ 90/ 135] Loss 0.430573 Top1 78.255208 Top5 96.540799 -2023-02-13 17:24:51,326 - Epoch: [13][ 100/ 135] Loss 0.430657 Top1 78.121094 Top5 96.523438 -2023-02-13 17:24:51,449 - Epoch: [13][ 110/ 135] Loss 0.430347 Top1 78.132102 Top5 96.491477 -2023-02-13 17:24:51,575 - Epoch: [13][ 120/ 135] Loss 0.428204 Top1 78.108724 Top5 96.481120 -2023-02-13 17:24:51,698 - Epoch: [13][ 130/ 135] Loss 0.427653 Top1 78.091947 Top5 96.436298 -2023-02-13 17:24:51,742 - Epoch: [13][ 135/ 135] Loss 0.432461 Top1 78.071173 Top5 96.432631 -2023-02-13 17:24:51,817 - ==> Top1: 78.071 Top5: 96.433 Loss: 0.432 - -2023-02-13 17:24:51,817 - ==> Confusion: -[[ 815 1 19 1 13 5 0 1 3 63 1 5 1 8 12 4 4 0 1 0 10] - [ 5 901 2 2 13 56 3 13 5 1 4 4 2 4 0 0 3 0 4 3 8] - [ 11 5 914 8 3 1 47 12 1 2 7 2 1 5 3 9 1 4 5 9 8] - [ 4 2 41 853 0 4 3 1 3 3 14 0 17 6 18 4 7 11 12 2 11] - [ 14 10 4 1 981 7 1 0 2 6 0 3 2 7 9 4 4 1 0 2 8] - [ 2 23 4 3 7 949 5 7 5 7 2 13 9 12 1 2 1 1 1 13 3] - [ 3 6 10 0 2 3 1023 7 0 1 0 2 5 2 1 5 0 1 0 27 1] - [ 1 17 13 1 6 51 10 849 2 4 7 9 3 3 0 0 0 2 19 20 7] - [ 21 7 1 1 3 0 0 0 851 52 12 1 2 27 21 0 1 3 5 1 0] - [ 87 0 3 0 7 5 1 0 38 815 1 1 0 34 6 1 1 2 0 2 8] - [ 4 2 5 6 0 2 9 1 16 3 964 3 1 16 1 0 2 0 7 3 6] - [ 6 1 0 0 5 10 0 1 4 1 0 892 33 10 0 7 4 12 2 16 1] - [ 4 0 2 2 1 3 0 0 1 0 0 50 840 1 1 7 3 37 1 2 4] - [ 11 4 1 1 7 9 1 0 7 8 3 13 2 939 4 1 3 0 2 7 1] - [ 10 6 4 19 18 3 0 0 22 5 3 1 7 4 958 2 3 5 10 0 12] - [ 3 2 3 2 10 1 12 0 2 0 0 15 8 4 1 950 3 16 0 8 6] - [ 2 8 1 1 7 6 1 0 1 0 0 5 5 4 5 22 968 4 1 10 10] - [ 7 0 2 5 2 1 0 1 0 1 1 13 17 2 1 21 0 969 0 4 4] - [ 5 4 7 14 2 4 0 32 8 2 14 7 11 0 19 0 0 0 949 4 4] - [ 1 3 4 1 0 9 10 6 0 0 0 27 3 3 0 6 4 1 1 1064 5] - [ 206 337 337 90 294 332 202 171 116 144 240 187 470 545 205 198 245 123 171 478 8343]] - -2023-02-13 17:24:51,819 - ==> Best [Top1: 79.141 Top5: 96.838 Sparsity:0.00 Params: 148928 on epoch: 12] -2023-02-13 17:24:51,819 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:24:51,825 - - -2023-02-13 17:24:51,825 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:24:52,709 - Epoch: [14][ 10/ 1207] Overall Loss 0.447318 Objective Loss 0.447318 LR 0.001000 Time 0.088289 -2023-02-13 17:24:52,906 - Epoch: [14][ 20/ 1207] Overall Loss 0.441355 Objective Loss 0.441355 LR 0.001000 Time 0.054004 -2023-02-13 17:24:53,105 - Epoch: [14][ 30/ 1207] Overall Loss 0.437408 Objective Loss 0.437408 LR 0.001000 Time 0.042608 -2023-02-13 17:24:53,300 - Epoch: [14][ 40/ 1207] Overall Loss 0.426382 Objective Loss 0.426382 LR 0.001000 Time 0.036821 -2023-02-13 17:24:53,498 - Epoch: [14][ 50/ 1207] Overall Loss 0.434147 Objective Loss 0.434147 LR 0.001000 Time 0.033410 -2023-02-13 17:24:53,693 - Epoch: [14][ 60/ 1207] Overall Loss 0.432577 Objective Loss 0.432577 LR 0.001000 Time 0.031093 -2023-02-13 17:24:53,891 - Epoch: [14][ 70/ 1207] Overall Loss 0.427224 Objective Loss 0.427224 LR 0.001000 Time 0.029468 -2023-02-13 17:24:54,087 - Epoch: [14][ 80/ 1207] Overall Loss 0.426575 Objective Loss 0.426575 LR 0.001000 Time 0.028226 -2023-02-13 17:24:54,284 - Epoch: [14][ 90/ 1207] Overall Loss 0.423294 Objective Loss 0.423294 LR 0.001000 Time 0.027283 -2023-02-13 17:24:54,480 - Epoch: [14][ 100/ 1207] Overall Loss 0.421009 Objective Loss 0.421009 LR 0.001000 Time 0.026509 -2023-02-13 17:24:54,678 - Epoch: [14][ 110/ 1207] Overall Loss 0.422783 Objective Loss 0.422783 LR 0.001000 Time 0.025891 -2023-02-13 17:24:54,873 - Epoch: [14][ 120/ 1207] Overall Loss 0.423807 Objective Loss 0.423807 LR 0.001000 Time 0.025356 -2023-02-13 17:24:55,070 - Epoch: [14][ 130/ 1207] Overall Loss 0.424240 Objective Loss 0.424240 LR 0.001000 Time 0.024924 -2023-02-13 17:24:55,265 - Epoch: [14][ 140/ 1207] Overall Loss 0.423412 Objective Loss 0.423412 LR 0.001000 Time 0.024533 -2023-02-13 17:24:55,464 - Epoch: [14][ 150/ 1207] Overall Loss 0.426095 Objective Loss 0.426095 LR 0.001000 Time 0.024221 -2023-02-13 17:24:55,659 - Epoch: [14][ 160/ 1207] Overall Loss 0.424969 Objective Loss 0.424969 LR 0.001000 Time 0.023920 -2023-02-13 17:24:55,858 - Epoch: [14][ 170/ 1207] Overall Loss 0.428691 Objective Loss 0.428691 LR 0.001000 Time 0.023683 -2023-02-13 17:24:56,052 - Epoch: [14][ 180/ 1207] Overall Loss 0.429778 Objective Loss 0.429778 LR 0.001000 Time 0.023445 -2023-02-13 17:24:56,250 - Epoch: [14][ 190/ 1207] Overall Loss 0.431973 Objective Loss 0.431973 LR 0.001000 Time 0.023251 -2023-02-13 17:24:56,445 - Epoch: [14][ 200/ 1207] Overall Loss 0.432743 Objective Loss 0.432743 LR 0.001000 Time 0.023062 -2023-02-13 17:24:56,643 - Epoch: [14][ 210/ 1207] Overall Loss 0.431541 Objective Loss 0.431541 LR 0.001000 Time 0.022906 -2023-02-13 17:24:56,838 - Epoch: [14][ 220/ 1207] Overall Loss 0.431780 Objective Loss 0.431780 LR 0.001000 Time 0.022749 -2023-02-13 17:24:57,036 - Epoch: [14][ 230/ 1207] Overall Loss 0.433263 Objective Loss 0.433263 LR 0.001000 Time 0.022620 -2023-02-13 17:24:57,232 - Epoch: [14][ 240/ 1207] Overall Loss 0.433593 Objective Loss 0.433593 LR 0.001000 Time 0.022489 -2023-02-13 17:24:57,430 - Epoch: [14][ 250/ 1207] Overall Loss 0.434111 Objective Loss 0.434111 LR 0.001000 Time 0.022383 -2023-02-13 17:24:57,625 - Epoch: [14][ 260/ 1207] Overall Loss 0.435069 Objective Loss 0.435069 LR 0.001000 Time 0.022269 -2023-02-13 17:24:57,824 - Epoch: [14][ 270/ 1207] Overall Loss 0.435231 Objective Loss 0.435231 LR 0.001000 Time 0.022180 -2023-02-13 17:24:58,018 - Epoch: [14][ 280/ 1207] Overall Loss 0.435746 Objective Loss 0.435746 LR 0.001000 Time 0.022079 -2023-02-13 17:24:58,216 - Epoch: [14][ 290/ 1207] Overall Loss 0.437192 Objective Loss 0.437192 LR 0.001000 Time 0.022001 -2023-02-13 17:24:58,412 - Epoch: [14][ 300/ 1207] Overall Loss 0.438918 Objective Loss 0.438918 LR 0.001000 Time 0.021917 -2023-02-13 17:24:58,612 - Epoch: [14][ 310/ 1207] Overall Loss 0.439092 Objective Loss 0.439092 LR 0.001000 Time 0.021853 -2023-02-13 17:24:58,806 - Epoch: [14][ 320/ 1207] Overall Loss 0.438852 Objective Loss 0.438852 LR 0.001000 Time 0.021776 -2023-02-13 17:24:59,004 - Epoch: [14][ 330/ 1207] Overall Loss 0.439397 Objective Loss 0.439397 LR 0.001000 Time 0.021717 -2023-02-13 17:24:59,199 - Epoch: [14][ 340/ 1207] Overall Loss 0.438328 Objective Loss 0.438328 LR 0.001000 Time 0.021649 -2023-02-13 17:24:59,397 - Epoch: [14][ 350/ 1207] Overall Loss 0.437970 Objective Loss 0.437970 LR 0.001000 Time 0.021595 -2023-02-13 17:24:59,592 - Epoch: [14][ 360/ 1207] Overall Loss 0.437840 Objective Loss 0.437840 LR 0.001000 Time 0.021537 -2023-02-13 17:24:59,790 - Epoch: [14][ 370/ 1207] Overall Loss 0.438534 Objective Loss 0.438534 LR 0.001000 Time 0.021490 -2023-02-13 17:24:59,985 - Epoch: [14][ 380/ 1207] Overall Loss 0.438076 Objective Loss 0.438076 LR 0.001000 Time 0.021435 -2023-02-13 17:25:00,183 - Epoch: [14][ 390/ 1207] Overall Loss 0.437170 Objective Loss 0.437170 LR 0.001000 Time 0.021392 -2023-02-13 17:25:00,377 - Epoch: [14][ 400/ 1207] Overall Loss 0.437393 Objective Loss 0.437393 LR 0.001000 Time 0.021342 -2023-02-13 17:25:00,575 - Epoch: [14][ 410/ 1207] Overall Loss 0.437171 Objective Loss 0.437171 LR 0.001000 Time 0.021304 -2023-02-13 17:25:00,771 - Epoch: [14][ 420/ 1207] Overall Loss 0.437508 Objective Loss 0.437508 LR 0.001000 Time 0.021261 -2023-02-13 17:25:00,969 - Epoch: [14][ 430/ 1207] Overall Loss 0.437849 Objective Loss 0.437849 LR 0.001000 Time 0.021228 -2023-02-13 17:25:01,165 - Epoch: [14][ 440/ 1207] Overall Loss 0.438033 Objective Loss 0.438033 LR 0.001000 Time 0.021188 -2023-02-13 17:25:01,362 - Epoch: [14][ 450/ 1207] Overall Loss 0.437608 Objective Loss 0.437608 LR 0.001000 Time 0.021156 -2023-02-13 17:25:01,558 - Epoch: [14][ 460/ 1207] Overall Loss 0.436525 Objective Loss 0.436525 LR 0.001000 Time 0.021120 -2023-02-13 17:25:01,757 - Epoch: [14][ 470/ 1207] Overall Loss 0.435990 Objective Loss 0.435990 LR 0.001000 Time 0.021093 -2023-02-13 17:25:01,951 - Epoch: [14][ 480/ 1207] Overall Loss 0.436576 Objective Loss 0.436576 LR 0.001000 Time 0.021058 -2023-02-13 17:25:02,149 - Epoch: [14][ 490/ 1207] Overall Loss 0.436374 Objective Loss 0.436374 LR 0.001000 Time 0.021032 -2023-02-13 17:25:02,344 - Epoch: [14][ 500/ 1207] Overall Loss 0.436367 Objective Loss 0.436367 LR 0.001000 Time 0.021000 -2023-02-13 17:25:02,542 - Epoch: [14][ 510/ 1207] Overall Loss 0.436878 Objective Loss 0.436878 LR 0.001000 Time 0.020976 -2023-02-13 17:25:02,737 - Epoch: [14][ 520/ 1207] Overall Loss 0.437927 Objective Loss 0.437927 LR 0.001000 Time 0.020946 -2023-02-13 17:25:02,936 - Epoch: [14][ 530/ 1207] Overall Loss 0.438353 Objective Loss 0.438353 LR 0.001000 Time 0.020926 -2023-02-13 17:25:03,130 - Epoch: [14][ 540/ 1207] Overall Loss 0.437994 Objective Loss 0.437994 LR 0.001000 Time 0.020897 -2023-02-13 17:25:03,329 - Epoch: [14][ 550/ 1207] Overall Loss 0.437275 Objective Loss 0.437275 LR 0.001000 Time 0.020878 -2023-02-13 17:25:03,524 - Epoch: [14][ 560/ 1207] Overall Loss 0.437152 Objective Loss 0.437152 LR 0.001000 Time 0.020854 -2023-02-13 17:25:03,722 - Epoch: [14][ 570/ 1207] Overall Loss 0.436395 Objective Loss 0.436395 LR 0.001000 Time 0.020834 -2023-02-13 17:25:03,918 - Epoch: [14][ 580/ 1207] Overall Loss 0.436211 Objective Loss 0.436211 LR 0.001000 Time 0.020811 -2023-02-13 17:25:04,117 - Epoch: [14][ 590/ 1207] Overall Loss 0.436336 Objective Loss 0.436336 LR 0.001000 Time 0.020795 -2023-02-13 17:25:04,311 - Epoch: [14][ 600/ 1207] Overall Loss 0.436630 Objective Loss 0.436630 LR 0.001000 Time 0.020772 -2023-02-13 17:25:04,509 - Epoch: [14][ 610/ 1207] Overall Loss 0.437662 Objective Loss 0.437662 LR 0.001000 Time 0.020756 -2023-02-13 17:25:04,705 - Epoch: [14][ 620/ 1207] Overall Loss 0.438095 Objective Loss 0.438095 LR 0.001000 Time 0.020736 -2023-02-13 17:25:04,904 - Epoch: [14][ 630/ 1207] Overall Loss 0.438702 Objective Loss 0.438702 LR 0.001000 Time 0.020722 -2023-02-13 17:25:05,100 - Epoch: [14][ 640/ 1207] Overall Loss 0.438801 Objective Loss 0.438801 LR 0.001000 Time 0.020704 -2023-02-13 17:25:05,297 - Epoch: [14][ 650/ 1207] Overall Loss 0.438926 Objective Loss 0.438926 LR 0.001000 Time 0.020689 -2023-02-13 17:25:05,493 - Epoch: [14][ 660/ 1207] Overall Loss 0.438816 Objective Loss 0.438816 LR 0.001000 Time 0.020671 -2023-02-13 17:25:05,692 - Epoch: [14][ 670/ 1207] Overall Loss 0.438694 Objective Loss 0.438694 LR 0.001000 Time 0.020659 -2023-02-13 17:25:05,889 - Epoch: [14][ 680/ 1207] Overall Loss 0.438390 Objective Loss 0.438390 LR 0.001000 Time 0.020644 -2023-02-13 17:25:06,087 - Epoch: [14][ 690/ 1207] Overall Loss 0.438105 Objective Loss 0.438105 LR 0.001000 Time 0.020631 -2023-02-13 17:25:06,282 - Epoch: [14][ 700/ 1207] Overall Loss 0.438166 Objective Loss 0.438166 LR 0.001000 Time 0.020616 -2023-02-13 17:25:06,481 - Epoch: [14][ 710/ 1207] Overall Loss 0.438340 Objective Loss 0.438340 LR 0.001000 Time 0.020605 -2023-02-13 17:25:06,672 - Epoch: [14][ 720/ 1207] Overall Loss 0.438128 Objective Loss 0.438128 LR 0.001000 Time 0.020583 -2023-02-13 17:25:06,861 - Epoch: [14][ 730/ 1207] Overall Loss 0.438087 Objective Loss 0.438087 LR 0.001000 Time 0.020560 -2023-02-13 17:25:07,050 - Epoch: [14][ 740/ 1207] Overall Loss 0.438441 Objective Loss 0.438441 LR 0.001000 Time 0.020536 -2023-02-13 17:25:07,239 - Epoch: [14][ 750/ 1207] Overall Loss 0.438940 Objective Loss 0.438940 LR 0.001000 Time 0.020514 -2023-02-13 17:25:07,428 - Epoch: [14][ 760/ 1207] Overall Loss 0.439182 Objective Loss 0.439182 LR 0.001000 Time 0.020492 -2023-02-13 17:25:07,617 - Epoch: [14][ 770/ 1207] Overall Loss 0.439589 Objective Loss 0.439589 LR 0.001000 Time 0.020471 -2023-02-13 17:25:07,806 - Epoch: [14][ 780/ 1207] Overall Loss 0.439461 Objective Loss 0.439461 LR 0.001000 Time 0.020450 -2023-02-13 17:25:07,995 - Epoch: [14][ 790/ 1207] Overall Loss 0.439611 Objective Loss 0.439611 LR 0.001000 Time 0.020430 -2023-02-13 17:25:08,184 - Epoch: [14][ 800/ 1207] Overall Loss 0.439853 Objective Loss 0.439853 LR 0.001000 Time 0.020411 -2023-02-13 17:25:08,373 - Epoch: [14][ 810/ 1207] Overall Loss 0.439636 Objective Loss 0.439636 LR 0.001000 Time 0.020391 -2023-02-13 17:25:08,562 - Epoch: [14][ 820/ 1207] Overall Loss 0.439783 Objective Loss 0.439783 LR 0.001000 Time 0.020373 -2023-02-13 17:25:08,753 - Epoch: [14][ 830/ 1207] Overall Loss 0.439938 Objective Loss 0.439938 LR 0.001000 Time 0.020357 -2023-02-13 17:25:08,944 - Epoch: [14][ 840/ 1207] Overall Loss 0.440372 Objective Loss 0.440372 LR 0.001000 Time 0.020342 -2023-02-13 17:25:09,136 - Epoch: [14][ 850/ 1207] Overall Loss 0.440939 Objective Loss 0.440939 LR 0.001000 Time 0.020328 -2023-02-13 17:25:09,327 - Epoch: [14][ 860/ 1207] Overall Loss 0.440993 Objective Loss 0.440993 LR 0.001000 Time 0.020313 -2023-02-13 17:25:09,518 - Epoch: [14][ 870/ 1207] Overall Loss 0.441059 Objective Loss 0.441059 LR 0.001000 Time 0.020299 -2023-02-13 17:25:09,708 - Epoch: [14][ 880/ 1207] Overall Loss 0.440845 Objective Loss 0.440845 LR 0.001000 Time 0.020283 -2023-02-13 17:25:09,900 - Epoch: [14][ 890/ 1207] Overall Loss 0.441179 Objective Loss 0.441179 LR 0.001000 Time 0.020271 -2023-02-13 17:25:10,091 - Epoch: [14][ 900/ 1207] Overall Loss 0.441036 Objective Loss 0.441036 LR 0.001000 Time 0.020257 -2023-02-13 17:25:10,282 - Epoch: [14][ 910/ 1207] Overall Loss 0.441249 Objective Loss 0.441249 LR 0.001000 Time 0.020245 -2023-02-13 17:25:10,472 - Epoch: [14][ 920/ 1207] Overall Loss 0.440881 Objective Loss 0.440881 LR 0.001000 Time 0.020231 -2023-02-13 17:25:10,663 - Epoch: [14][ 930/ 1207] Overall Loss 0.441322 Objective Loss 0.441322 LR 0.001000 Time 0.020218 -2023-02-13 17:25:10,856 - Epoch: [14][ 940/ 1207] Overall Loss 0.440828 Objective Loss 0.440828 LR 0.001000 Time 0.020208 -2023-02-13 17:25:11,047 - Epoch: [14][ 950/ 1207] Overall Loss 0.440837 Objective Loss 0.440837 LR 0.001000 Time 0.020196 -2023-02-13 17:25:11,238 - Epoch: [14][ 960/ 1207] Overall Loss 0.440598 Objective Loss 0.440598 LR 0.001000 Time 0.020184 -2023-02-13 17:25:11,429 - Epoch: [14][ 970/ 1207] Overall Loss 0.440801 Objective Loss 0.440801 LR 0.001000 Time 0.020173 -2023-02-13 17:25:11,619 - Epoch: [14][ 980/ 1207] Overall Loss 0.440630 Objective Loss 0.440630 LR 0.001000 Time 0.020161 -2023-02-13 17:25:11,812 - Epoch: [14][ 990/ 1207] Overall Loss 0.440527 Objective Loss 0.440527 LR 0.001000 Time 0.020151 -2023-02-13 17:25:12,002 - Epoch: [14][ 1000/ 1207] Overall Loss 0.440450 Objective Loss 0.440450 LR 0.001000 Time 0.020139 -2023-02-13 17:25:12,194 - Epoch: [14][ 1010/ 1207] Overall Loss 0.440721 Objective Loss 0.440721 LR 0.001000 Time 0.020129 -2023-02-13 17:25:12,384 - Epoch: [14][ 1020/ 1207] Overall Loss 0.440472 Objective Loss 0.440472 LR 0.001000 Time 0.020118 -2023-02-13 17:25:12,575 - Epoch: [14][ 1030/ 1207] Overall Loss 0.440126 Objective Loss 0.440126 LR 0.001000 Time 0.020108 -2023-02-13 17:25:12,765 - Epoch: [14][ 1040/ 1207] Overall Loss 0.440311 Objective Loss 0.440311 LR 0.001000 Time 0.020097 -2023-02-13 17:25:12,957 - Epoch: [14][ 1050/ 1207] Overall Loss 0.440253 Objective Loss 0.440253 LR 0.001000 Time 0.020088 -2023-02-13 17:25:13,147 - Epoch: [14][ 1060/ 1207] Overall Loss 0.440226 Objective Loss 0.440226 LR 0.001000 Time 0.020078 -2023-02-13 17:25:13,339 - Epoch: [14][ 1070/ 1207] Overall Loss 0.440392 Objective Loss 0.440392 LR 0.001000 Time 0.020069 -2023-02-13 17:25:13,529 - Epoch: [14][ 1080/ 1207] Overall Loss 0.440391 Objective Loss 0.440391 LR 0.001000 Time 0.020058 -2023-02-13 17:25:13,721 - Epoch: [14][ 1090/ 1207] Overall Loss 0.440214 Objective Loss 0.440214 LR 0.001000 Time 0.020050 -2023-02-13 17:25:13,911 - Epoch: [14][ 1100/ 1207] Overall Loss 0.440154 Objective Loss 0.440154 LR 0.001000 Time 0.020041 -2023-02-13 17:25:14,103 - Epoch: [14][ 1110/ 1207] Overall Loss 0.440169 Objective Loss 0.440169 LR 0.001000 Time 0.020033 -2023-02-13 17:25:14,294 - Epoch: [14][ 1120/ 1207] Overall Loss 0.440309 Objective Loss 0.440309 LR 0.001000 Time 0.020024 -2023-02-13 17:25:14,486 - Epoch: [14][ 1130/ 1207] Overall Loss 0.440471 Objective Loss 0.440471 LR 0.001000 Time 0.020016 -2023-02-13 17:25:14,677 - Epoch: [14][ 1140/ 1207] Overall Loss 0.440541 Objective Loss 0.440541 LR 0.001000 Time 0.020007 -2023-02-13 17:25:14,869 - Epoch: [14][ 1150/ 1207] Overall Loss 0.440659 Objective Loss 0.440659 LR 0.001000 Time 0.020000 -2023-02-13 17:25:15,060 - Epoch: [14][ 1160/ 1207] Overall Loss 0.440395 Objective Loss 0.440395 LR 0.001000 Time 0.019992 -2023-02-13 17:25:15,252 - Epoch: [14][ 1170/ 1207] Overall Loss 0.440341 Objective Loss 0.440341 LR 0.001000 Time 0.019985 -2023-02-13 17:25:15,444 - Epoch: [14][ 1180/ 1207] Overall Loss 0.440395 Objective Loss 0.440395 LR 0.001000 Time 0.019978 -2023-02-13 17:25:15,636 - Epoch: [14][ 1190/ 1207] Overall Loss 0.440642 Objective Loss 0.440642 LR 0.001000 Time 0.019971 -2023-02-13 17:25:15,887 - Epoch: [14][ 1200/ 1207] Overall Loss 0.440599 Objective Loss 0.440599 LR 0.001000 Time 0.020014 -2023-02-13 17:25:16,003 - Epoch: [14][ 1207/ 1207] Overall Loss 0.440267 Objective Loss 0.440267 Top1 83.841463 Top5 96.341463 LR 0.001000 Time 0.019993 -2023-02-13 17:25:16,074 - --- validate (epoch=14)----------- -2023-02-13 17:25:16,074 - 34311 samples (256 per mini-batch) -2023-02-13 17:25:16,466 - Epoch: [14][ 10/ 135] Loss 0.415657 Top1 77.382812 Top5 96.992188 -2023-02-13 17:25:16,594 - Epoch: [14][ 20/ 135] Loss 0.443199 Top1 77.167969 Top5 96.660156 -2023-02-13 17:25:16,726 - Epoch: [14][ 30/ 135] Loss 0.431570 Top1 77.005208 Top5 96.744792 -2023-02-13 17:25:16,853 - Epoch: [14][ 40/ 135] Loss 0.447027 Top1 77.031250 Top5 96.679688 -2023-02-13 17:25:16,982 - Epoch: [14][ 50/ 135] Loss 0.442987 Top1 77.312500 Top5 96.664062 -2023-02-13 17:25:17,113 - Epoch: [14][ 60/ 135] Loss 0.435739 Top1 77.389323 Top5 96.608073 -2023-02-13 17:25:17,241 - Epoch: [14][ 70/ 135] Loss 0.435886 Top1 77.522321 Top5 96.562500 -2023-02-13 17:25:17,368 - Epoch: [14][ 80/ 135] Loss 0.441446 Top1 77.529297 Top5 96.494141 -2023-02-13 17:25:17,498 - Epoch: [14][ 90/ 135] Loss 0.443129 Top1 77.391493 Top5 96.493056 -2023-02-13 17:25:17,626 - Epoch: [14][ 100/ 135] Loss 0.440064 Top1 77.355469 Top5 96.464844 -2023-02-13 17:25:17,755 - Epoch: [14][ 110/ 135] Loss 0.438036 Top1 77.428977 Top5 96.480824 -2023-02-13 17:25:17,884 - Epoch: [14][ 120/ 135] Loss 0.436501 Top1 77.441406 Top5 96.481120 -2023-02-13 17:25:18,017 - Epoch: [14][ 130/ 135] Loss 0.437505 Top1 77.466947 Top5 96.481370 -2023-02-13 17:25:18,063 - Epoch: [14][ 135/ 135] Loss 0.435818 Top1 77.482440 Top5 96.488007 -2023-02-13 17:25:18,131 - ==> Top1: 77.482 Top5: 96.488 Loss: 0.436 - -2023-02-13 17:25:18,131 - ==> Confusion: -[[ 843 5 4 1 7 2 0 1 1 68 0 4 1 4 7 3 3 2 1 1 9] - [ 5 949 1 2 12 15 3 20 7 2 0 4 1 1 0 0 3 0 4 2 2] - [ 13 7 917 6 8 3 14 30 1 5 6 2 2 2 2 8 3 2 12 4 11] - [ 6 5 31 807 0 2 1 4 4 5 19 2 12 4 36 2 5 4 52 0 15] - [ 19 13 1 1 984 3 1 1 1 9 2 6 0 3 3 2 5 1 0 6 5] - [ 3 96 1 2 9 845 3 46 2 11 2 12 5 12 1 2 4 1 3 6 4] - [ 6 9 24 0 0 4 1016 13 0 0 2 0 2 1 0 7 3 1 1 6 4] - [ 0 25 7 0 6 19 1 905 1 1 1 4 4 1 0 0 0 1 33 11 4] - [ 20 5 0 1 2 0 0 2 883 59 6 3 0 3 16 0 2 1 5 0 1] - [ 108 3 1 0 4 1 0 3 31 832 0 2 0 14 4 0 1 1 0 1 6] - [ 4 8 6 2 1 2 4 3 28 4 950 1 2 8 3 0 3 0 18 1 3] - [ 9 4 2 0 3 10 1 6 0 4 0 909 26 4 0 2 5 3 1 15 1] - [ 4 3 1 0 2 3 2 0 6 1 0 53 837 1 3 6 3 15 4 5 10] - [ 11 10 3 0 11 13 2 1 28 31 7 16 1 860 8 2 5 1 0 11 3] - [ 19 8 1 8 11 5 0 1 43 12 2 4 6 1 940 1 1 2 15 1 11] - [ 11 2 3 0 9 4 11 1 0 2 0 10 7 1 0 946 16 11 0 4 8] - [ 9 20 1 1 18 2 1 0 4 1 0 2 3 2 0 5 980 1 3 6 2] - [ 11 2 1 4 0 1 2 1 1 1 1 23 50 1 4 20 1 912 0 5 10] - [ 6 13 6 2 1 1 1 41 5 2 2 1 4 0 13 0 1 1 982 1 3] - [ 1 7 1 0 0 5 9 27 0 0 2 22 4 3 0 7 5 0 3 1045 7] - [ 385 533 215 64 291 178 123 246 215 220 244 185 369 334 190 112 479 84 271 454 8242]] - -2023-02-13 17:25:18,133 - ==> Best [Top1: 79.141 Top5: 96.838 Sparsity:0.00 Params: 148928 on epoch: 12] -2023-02-13 17:25:18,133 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:25:18,138 - - -2023-02-13 17:25:18,139 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:25:19,109 - Epoch: [15][ 10/ 1207] Overall Loss 0.404286 Objective Loss 0.404286 LR 0.001000 Time 0.096926 -2023-02-13 17:25:19,305 - Epoch: [15][ 20/ 1207] Overall Loss 0.414864 Objective Loss 0.414864 LR 0.001000 Time 0.058267 -2023-02-13 17:25:19,496 - Epoch: [15][ 30/ 1207] Overall Loss 0.410501 Objective Loss 0.410501 LR 0.001000 Time 0.045199 -2023-02-13 17:25:19,687 - Epoch: [15][ 40/ 1207] Overall Loss 0.412710 Objective Loss 0.412710 LR 0.001000 Time 0.038668 -2023-02-13 17:25:19,877 - Epoch: [15][ 50/ 1207] Overall Loss 0.409802 Objective Loss 0.409802 LR 0.001000 Time 0.034730 -2023-02-13 17:25:20,068 - Epoch: [15][ 60/ 1207] Overall Loss 0.411683 Objective Loss 0.411683 LR 0.001000 Time 0.032108 -2023-02-13 17:25:20,259 - Epoch: [15][ 70/ 1207] Overall Loss 0.406547 Objective Loss 0.406547 LR 0.001000 Time 0.030242 -2023-02-13 17:25:20,449 - Epoch: [15][ 80/ 1207] Overall Loss 0.405926 Objective Loss 0.405926 LR 0.001000 Time 0.028841 -2023-02-13 17:25:20,640 - Epoch: [15][ 90/ 1207] Overall Loss 0.410529 Objective Loss 0.410529 LR 0.001000 Time 0.027748 -2023-02-13 17:25:20,832 - Epoch: [15][ 100/ 1207] Overall Loss 0.410571 Objective Loss 0.410571 LR 0.001000 Time 0.026893 -2023-02-13 17:25:21,023 - Epoch: [15][ 110/ 1207] Overall Loss 0.414297 Objective Loss 0.414297 LR 0.001000 Time 0.026182 -2023-02-13 17:25:21,214 - Epoch: [15][ 120/ 1207] Overall Loss 0.412722 Objective Loss 0.412722 LR 0.001000 Time 0.025591 -2023-02-13 17:25:21,405 - Epoch: [15][ 130/ 1207] Overall Loss 0.413836 Objective Loss 0.413836 LR 0.001000 Time 0.025084 -2023-02-13 17:25:21,596 - Epoch: [15][ 140/ 1207] Overall Loss 0.412689 Objective Loss 0.412689 LR 0.001000 Time 0.024657 -2023-02-13 17:25:21,787 - Epoch: [15][ 150/ 1207] Overall Loss 0.412403 Objective Loss 0.412403 LR 0.001000 Time 0.024284 -2023-02-13 17:25:21,979 - Epoch: [15][ 160/ 1207] Overall Loss 0.411465 Objective Loss 0.411465 LR 0.001000 Time 0.023964 -2023-02-13 17:25:22,170 - Epoch: [15][ 170/ 1207] Overall Loss 0.414810 Objective Loss 0.414810 LR 0.001000 Time 0.023676 -2023-02-13 17:25:22,362 - Epoch: [15][ 180/ 1207] Overall Loss 0.415410 Objective Loss 0.415410 LR 0.001000 Time 0.023422 -2023-02-13 17:25:22,552 - Epoch: [15][ 190/ 1207] Overall Loss 0.413582 Objective Loss 0.413582 LR 0.001000 Time 0.023190 -2023-02-13 17:25:22,743 - Epoch: [15][ 200/ 1207] Overall Loss 0.412456 Objective Loss 0.412456 LR 0.001000 Time 0.022982 -2023-02-13 17:25:22,934 - Epoch: [15][ 210/ 1207] Overall Loss 0.413813 Objective Loss 0.413813 LR 0.001000 Time 0.022796 -2023-02-13 17:25:23,124 - Epoch: [15][ 220/ 1207] Overall Loss 0.414044 Objective Loss 0.414044 LR 0.001000 Time 0.022623 -2023-02-13 17:25:23,316 - Epoch: [15][ 230/ 1207] Overall Loss 0.413653 Objective Loss 0.413653 LR 0.001000 Time 0.022471 -2023-02-13 17:25:23,506 - Epoch: [15][ 240/ 1207] Overall Loss 0.413805 Objective Loss 0.413805 LR 0.001000 Time 0.022325 -2023-02-13 17:25:23,697 - Epoch: [15][ 250/ 1207] Overall Loss 0.412587 Objective Loss 0.412587 LR 0.001000 Time 0.022194 -2023-02-13 17:25:23,888 - Epoch: [15][ 260/ 1207] Overall Loss 0.413889 Objective Loss 0.413889 LR 0.001000 Time 0.022075 -2023-02-13 17:25:24,079 - Epoch: [15][ 270/ 1207] Overall Loss 0.413234 Objective Loss 0.413234 LR 0.001000 Time 0.021963 -2023-02-13 17:25:24,279 - Epoch: [15][ 280/ 1207] Overall Loss 0.412716 Objective Loss 0.412716 LR 0.001000 Time 0.021890 -2023-02-13 17:25:24,479 - Epoch: [15][ 290/ 1207] Overall Loss 0.412976 Objective Loss 0.412976 LR 0.001000 Time 0.021824 -2023-02-13 17:25:24,682 - Epoch: [15][ 300/ 1207] Overall Loss 0.413537 Objective Loss 0.413537 LR 0.001000 Time 0.021773 -2023-02-13 17:25:24,883 - Epoch: [15][ 310/ 1207] Overall Loss 0.415249 Objective Loss 0.415249 LR 0.001000 Time 0.021717 -2023-02-13 17:25:25,087 - Epoch: [15][ 320/ 1207] Overall Loss 0.416224 Objective Loss 0.416224 LR 0.001000 Time 0.021674 -2023-02-13 17:25:25,287 - Epoch: [15][ 330/ 1207] Overall Loss 0.416047 Objective Loss 0.416047 LR 0.001000 Time 0.021623 -2023-02-13 17:25:25,491 - Epoch: [15][ 340/ 1207] Overall Loss 0.415694 Objective Loss 0.415694 LR 0.001000 Time 0.021587 -2023-02-13 17:25:25,691 - Epoch: [15][ 350/ 1207] Overall Loss 0.415618 Objective Loss 0.415618 LR 0.001000 Time 0.021538 -2023-02-13 17:25:25,896 - Epoch: [15][ 360/ 1207] Overall Loss 0.415232 Objective Loss 0.415232 LR 0.001000 Time 0.021508 -2023-02-13 17:25:26,095 - Epoch: [15][ 370/ 1207] Overall Loss 0.415020 Objective Loss 0.415020 LR 0.001000 Time 0.021465 -2023-02-13 17:25:26,300 - Epoch: [15][ 380/ 1207] Overall Loss 0.414657 Objective Loss 0.414657 LR 0.001000 Time 0.021439 -2023-02-13 17:25:26,500 - Epoch: [15][ 390/ 1207] Overall Loss 0.415242 Objective Loss 0.415242 LR 0.001000 Time 0.021401 -2023-02-13 17:25:26,704 - Epoch: [15][ 400/ 1207] Overall Loss 0.417294 Objective Loss 0.417294 LR 0.001000 Time 0.021374 -2023-02-13 17:25:26,904 - Epoch: [15][ 410/ 1207] Overall Loss 0.417490 Objective Loss 0.417490 LR 0.001000 Time 0.021340 -2023-02-13 17:25:27,108 - Epoch: [15][ 420/ 1207] Overall Loss 0.416986 Objective Loss 0.416986 LR 0.001000 Time 0.021317 -2023-02-13 17:25:27,309 - Epoch: [15][ 430/ 1207] Overall Loss 0.417095 Objective Loss 0.417095 LR 0.001000 Time 0.021287 -2023-02-13 17:25:27,514 - Epoch: [15][ 440/ 1207] Overall Loss 0.416942 Objective Loss 0.416942 LR 0.001000 Time 0.021267 -2023-02-13 17:25:27,714 - Epoch: [15][ 450/ 1207] Overall Loss 0.418421 Objective Loss 0.418421 LR 0.001000 Time 0.021239 -2023-02-13 17:25:27,918 - Epoch: [15][ 460/ 1207] Overall Loss 0.418058 Objective Loss 0.418058 LR 0.001000 Time 0.021221 -2023-02-13 17:25:28,119 - Epoch: [15][ 470/ 1207] Overall Loss 0.418008 Objective Loss 0.418008 LR 0.001000 Time 0.021195 -2023-02-13 17:25:28,323 - Epoch: [15][ 480/ 1207] Overall Loss 0.417719 Objective Loss 0.417719 LR 0.001000 Time 0.021179 -2023-02-13 17:25:28,523 - Epoch: [15][ 490/ 1207] Overall Loss 0.417341 Objective Loss 0.417341 LR 0.001000 Time 0.021154 -2023-02-13 17:25:28,727 - Epoch: [15][ 500/ 1207] Overall Loss 0.416837 Objective Loss 0.416837 LR 0.001000 Time 0.021138 -2023-02-13 17:25:28,928 - Epoch: [15][ 510/ 1207] Overall Loss 0.416641 Objective Loss 0.416641 LR 0.001000 Time 0.021116 -2023-02-13 17:25:29,131 - Epoch: [15][ 520/ 1207] Overall Loss 0.416996 Objective Loss 0.416996 LR 0.001000 Time 0.021101 -2023-02-13 17:25:29,332 - Epoch: [15][ 530/ 1207] Overall Loss 0.417277 Objective Loss 0.417277 LR 0.001000 Time 0.021081 -2023-02-13 17:25:29,537 - Epoch: [15][ 540/ 1207] Overall Loss 0.418293 Objective Loss 0.418293 LR 0.001000 Time 0.021069 -2023-02-13 17:25:29,737 - Epoch: [15][ 550/ 1207] Overall Loss 0.419037 Objective Loss 0.419037 LR 0.001000 Time 0.021049 -2023-02-13 17:25:29,942 - Epoch: [15][ 560/ 1207] Overall Loss 0.419122 Objective Loss 0.419122 LR 0.001000 Time 0.021038 -2023-02-13 17:25:30,141 - Epoch: [15][ 570/ 1207] Overall Loss 0.418901 Objective Loss 0.418901 LR 0.001000 Time 0.021018 -2023-02-13 17:25:30,346 - Epoch: [15][ 580/ 1207] Overall Loss 0.418687 Objective Loss 0.418687 LR 0.001000 Time 0.021007 -2023-02-13 17:25:30,546 - Epoch: [15][ 590/ 1207] Overall Loss 0.418185 Objective Loss 0.418185 LR 0.001000 Time 0.020990 -2023-02-13 17:25:30,751 - Epoch: [15][ 600/ 1207] Overall Loss 0.418581 Objective Loss 0.418581 LR 0.001000 Time 0.020981 -2023-02-13 17:25:30,951 - Epoch: [15][ 610/ 1207] Overall Loss 0.419346 Objective Loss 0.419346 LR 0.001000 Time 0.020965 -2023-02-13 17:25:31,156 - Epoch: [15][ 620/ 1207] Overall Loss 0.419302 Objective Loss 0.419302 LR 0.001000 Time 0.020956 -2023-02-13 17:25:31,356 - Epoch: [15][ 630/ 1207] Overall Loss 0.419138 Objective Loss 0.419138 LR 0.001000 Time 0.020941 -2023-02-13 17:25:31,556 - Epoch: [15][ 640/ 1207] Overall Loss 0.419263 Objective Loss 0.419263 LR 0.001000 Time 0.020925 -2023-02-13 17:25:31,748 - Epoch: [15][ 650/ 1207] Overall Loss 0.419390 Objective Loss 0.419390 LR 0.001000 Time 0.020898 -2023-02-13 17:25:31,938 - Epoch: [15][ 660/ 1207] Overall Loss 0.419542 Objective Loss 0.419542 LR 0.001000 Time 0.020869 -2023-02-13 17:25:32,129 - Epoch: [15][ 670/ 1207] Overall Loss 0.420138 Objective Loss 0.420138 LR 0.001000 Time 0.020841 -2023-02-13 17:25:32,320 - Epoch: [15][ 680/ 1207] Overall Loss 0.420049 Objective Loss 0.420049 LR 0.001000 Time 0.020816 -2023-02-13 17:25:32,511 - Epoch: [15][ 690/ 1207] Overall Loss 0.419880 Objective Loss 0.419880 LR 0.001000 Time 0.020790 -2023-02-13 17:25:32,701 - Epoch: [15][ 700/ 1207] Overall Loss 0.419721 Objective Loss 0.419721 LR 0.001000 Time 0.020764 -2023-02-13 17:25:32,892 - Epoch: [15][ 710/ 1207] Overall Loss 0.419988 Objective Loss 0.419988 LR 0.001000 Time 0.020740 -2023-02-13 17:25:33,083 - Epoch: [15][ 720/ 1207] Overall Loss 0.420462 Objective Loss 0.420462 LR 0.001000 Time 0.020717 -2023-02-13 17:25:33,274 - Epoch: [15][ 730/ 1207] Overall Loss 0.420385 Objective Loss 0.420385 LR 0.001000 Time 0.020694 -2023-02-13 17:25:33,464 - Epoch: [15][ 740/ 1207] Overall Loss 0.420172 Objective Loss 0.420172 LR 0.001000 Time 0.020671 -2023-02-13 17:25:33,655 - Epoch: [15][ 750/ 1207] Overall Loss 0.419950 Objective Loss 0.419950 LR 0.001000 Time 0.020650 -2023-02-13 17:25:33,846 - Epoch: [15][ 760/ 1207] Overall Loss 0.420227 Objective Loss 0.420227 LR 0.001000 Time 0.020629 -2023-02-13 17:25:34,037 - Epoch: [15][ 770/ 1207] Overall Loss 0.420992 Objective Loss 0.420992 LR 0.001000 Time 0.020609 -2023-02-13 17:25:34,228 - Epoch: [15][ 780/ 1207] Overall Loss 0.421183 Objective Loss 0.421183 LR 0.001000 Time 0.020589 -2023-02-13 17:25:34,417 - Epoch: [15][ 790/ 1207] Overall Loss 0.421034 Objective Loss 0.421034 LR 0.001000 Time 0.020567 -2023-02-13 17:25:34,607 - Epoch: [15][ 800/ 1207] Overall Loss 0.420511 Objective Loss 0.420511 LR 0.001000 Time 0.020546 -2023-02-13 17:25:34,796 - Epoch: [15][ 810/ 1207] Overall Loss 0.421173 Objective Loss 0.421173 LR 0.001000 Time 0.020525 -2023-02-13 17:25:34,985 - Epoch: [15][ 820/ 1207] Overall Loss 0.421434 Objective Loss 0.421434 LR 0.001000 Time 0.020506 -2023-02-13 17:25:35,174 - Epoch: [15][ 830/ 1207] Overall Loss 0.421317 Objective Loss 0.421317 LR 0.001000 Time 0.020486 -2023-02-13 17:25:35,364 - Epoch: [15][ 840/ 1207] Overall Loss 0.421119 Objective Loss 0.421119 LR 0.001000 Time 0.020467 -2023-02-13 17:25:35,552 - Epoch: [15][ 850/ 1207] Overall Loss 0.421059 Objective Loss 0.421059 LR 0.001000 Time 0.020448 -2023-02-13 17:25:35,742 - Epoch: [15][ 860/ 1207] Overall Loss 0.421145 Objective Loss 0.421145 LR 0.001000 Time 0.020431 -2023-02-13 17:25:35,931 - Epoch: [15][ 870/ 1207] Overall Loss 0.421305 Objective Loss 0.421305 LR 0.001000 Time 0.020413 -2023-02-13 17:25:36,121 - Epoch: [15][ 880/ 1207] Overall Loss 0.421097 Objective Loss 0.421097 LR 0.001000 Time 0.020396 -2023-02-13 17:25:36,311 - Epoch: [15][ 890/ 1207] Overall Loss 0.420802 Objective Loss 0.420802 LR 0.001000 Time 0.020379 -2023-02-13 17:25:36,500 - Epoch: [15][ 900/ 1207] Overall Loss 0.420916 Objective Loss 0.420916 LR 0.001000 Time 0.020363 -2023-02-13 17:25:36,690 - Epoch: [15][ 910/ 1207] Overall Loss 0.420774 Objective Loss 0.420774 LR 0.001000 Time 0.020347 -2023-02-13 17:25:36,880 - Epoch: [15][ 920/ 1207] Overall Loss 0.420762 Objective Loss 0.420762 LR 0.001000 Time 0.020333 -2023-02-13 17:25:37,070 - Epoch: [15][ 930/ 1207] Overall Loss 0.420841 Objective Loss 0.420841 LR 0.001000 Time 0.020317 -2023-02-13 17:25:37,260 - Epoch: [15][ 940/ 1207] Overall Loss 0.420897 Objective Loss 0.420897 LR 0.001000 Time 0.020303 -2023-02-13 17:25:37,449 - Epoch: [15][ 950/ 1207] Overall Loss 0.420892 Objective Loss 0.420892 LR 0.001000 Time 0.020288 -2023-02-13 17:25:37,639 - Epoch: [15][ 960/ 1207] Overall Loss 0.420821 Objective Loss 0.420821 LR 0.001000 Time 0.020274 -2023-02-13 17:25:37,828 - Epoch: [15][ 970/ 1207] Overall Loss 0.420338 Objective Loss 0.420338 LR 0.001000 Time 0.020260 -2023-02-13 17:25:38,018 - Epoch: [15][ 980/ 1207] Overall Loss 0.420663 Objective Loss 0.420663 LR 0.001000 Time 0.020247 -2023-02-13 17:25:38,208 - Epoch: [15][ 990/ 1207] Overall Loss 0.420481 Objective Loss 0.420481 LR 0.001000 Time 0.020233 -2023-02-13 17:25:38,397 - Epoch: [15][ 1000/ 1207] Overall Loss 0.420396 Objective Loss 0.420396 LR 0.001000 Time 0.020220 -2023-02-13 17:25:38,587 - Epoch: [15][ 1010/ 1207] Overall Loss 0.420019 Objective Loss 0.420019 LR 0.001000 Time 0.020207 -2023-02-13 17:25:38,776 - Epoch: [15][ 1020/ 1207] Overall Loss 0.420269 Objective Loss 0.420269 LR 0.001000 Time 0.020194 -2023-02-13 17:25:38,965 - Epoch: [15][ 1030/ 1207] Overall Loss 0.420598 Objective Loss 0.420598 LR 0.001000 Time 0.020182 -2023-02-13 17:25:39,154 - Epoch: [15][ 1040/ 1207] Overall Loss 0.421116 Objective Loss 0.421116 LR 0.001000 Time 0.020169 -2023-02-13 17:25:39,344 - Epoch: [15][ 1050/ 1207] Overall Loss 0.421235 Objective Loss 0.421235 LR 0.001000 Time 0.020157 -2023-02-13 17:25:39,534 - Epoch: [15][ 1060/ 1207] Overall Loss 0.421485 Objective Loss 0.421485 LR 0.001000 Time 0.020146 -2023-02-13 17:25:39,723 - Epoch: [15][ 1070/ 1207] Overall Loss 0.421556 Objective Loss 0.421556 LR 0.001000 Time 0.020134 -2023-02-13 17:25:39,913 - Epoch: [15][ 1080/ 1207] Overall Loss 0.421704 Objective Loss 0.421704 LR 0.001000 Time 0.020123 -2023-02-13 17:25:40,102 - Epoch: [15][ 1090/ 1207] Overall Loss 0.422051 Objective Loss 0.422051 LR 0.001000 Time 0.020112 -2023-02-13 17:25:40,293 - Epoch: [15][ 1100/ 1207] Overall Loss 0.422139 Objective Loss 0.422139 LR 0.001000 Time 0.020102 -2023-02-13 17:25:40,483 - Epoch: [15][ 1110/ 1207] Overall Loss 0.421905 Objective Loss 0.421905 LR 0.001000 Time 0.020092 -2023-02-13 17:25:40,674 - Epoch: [15][ 1120/ 1207] Overall Loss 0.421811 Objective Loss 0.421811 LR 0.001000 Time 0.020083 -2023-02-13 17:25:40,866 - Epoch: [15][ 1130/ 1207] Overall Loss 0.421961 Objective Loss 0.421961 LR 0.001000 Time 0.020075 -2023-02-13 17:25:41,058 - Epoch: [15][ 1140/ 1207] Overall Loss 0.422183 Objective Loss 0.422183 LR 0.001000 Time 0.020067 -2023-02-13 17:25:41,249 - Epoch: [15][ 1150/ 1207] Overall Loss 0.421833 Objective Loss 0.421833 LR 0.001000 Time 0.020058 -2023-02-13 17:25:41,440 - Epoch: [15][ 1160/ 1207] Overall Loss 0.421799 Objective Loss 0.421799 LR 0.001000 Time 0.020049 -2023-02-13 17:25:41,631 - Epoch: [15][ 1170/ 1207] Overall Loss 0.421490 Objective Loss 0.421490 LR 0.001000 Time 0.020040 -2023-02-13 17:25:41,821 - Epoch: [15][ 1180/ 1207] Overall Loss 0.421672 Objective Loss 0.421672 LR 0.001000 Time 0.020032 -2023-02-13 17:25:42,013 - Epoch: [15][ 1190/ 1207] Overall Loss 0.421404 Objective Loss 0.421404 LR 0.001000 Time 0.020024 -2023-02-13 17:25:42,261 - Epoch: [15][ 1200/ 1207] Overall Loss 0.421488 Objective Loss 0.421488 LR 0.001000 Time 0.020064 -2023-02-13 17:25:42,378 - Epoch: [15][ 1207/ 1207] Overall Loss 0.421374 Objective Loss 0.421374 Top1 83.536585 Top5 97.865854 LR 0.001000 Time 0.020044 -2023-02-13 17:25:42,449 - --- validate (epoch=15)----------- -2023-02-13 17:25:42,450 - 34311 samples (256 per mini-batch) -2023-02-13 17:25:42,852 - Epoch: [15][ 10/ 135] Loss 0.401315 Top1 80.234375 Top5 97.070312 -2023-02-13 17:25:42,979 - Epoch: [15][ 20/ 135] Loss 0.408884 Top1 79.726562 Top5 96.855469 -2023-02-13 17:25:43,106 - Epoch: [15][ 30/ 135] Loss 0.413129 Top1 79.791667 Top5 96.757812 -2023-02-13 17:25:43,238 - Epoch: [15][ 40/ 135] Loss 0.409148 Top1 80.000000 Top5 96.835938 -2023-02-13 17:25:43,367 - Epoch: [15][ 50/ 135] Loss 0.406442 Top1 79.945312 Top5 96.882812 -2023-02-13 17:25:43,495 - Epoch: [15][ 60/ 135] Loss 0.407339 Top1 79.882812 Top5 96.933594 -2023-02-13 17:25:43,625 - Epoch: [15][ 70/ 135] Loss 0.415499 Top1 79.732143 Top5 96.869420 -2023-02-13 17:25:43,753 - Epoch: [15][ 80/ 135] Loss 0.414230 Top1 79.560547 Top5 96.791992 -2023-02-13 17:25:43,879 - Epoch: [15][ 90/ 135] Loss 0.416741 Top1 79.466146 Top5 96.740451 -2023-02-13 17:25:44,010 - Epoch: [15][ 100/ 135] Loss 0.413598 Top1 79.441406 Top5 96.808594 -2023-02-13 17:25:44,137 - Epoch: [15][ 110/ 135] Loss 0.413716 Top1 79.556108 Top5 96.843040 -2023-02-13 17:25:44,267 - Epoch: [15][ 120/ 135] Loss 0.410137 Top1 79.501953 Top5 96.868490 -2023-02-13 17:25:44,399 - Epoch: [15][ 130/ 135] Loss 0.410677 Top1 79.447115 Top5 96.862981 -2023-02-13 17:25:44,446 - Epoch: [15][ 135/ 135] Loss 0.411221 Top1 79.382705 Top5 96.834834 -2023-02-13 17:25:44,513 - ==> Top1: 79.383 Top5: 96.835 Loss: 0.411 - -2023-02-13 17:25:44,514 - ==> Confusion: -[[ 849 1 10 2 10 2 0 1 9 46 0 3 2 3 10 3 6 0 4 0 6] - [ 4 896 1 5 14 34 2 29 6 2 7 5 2 1 0 0 7 1 5 3 9] - [ 10 4 959 15 2 0 8 20 1 1 3 5 0 1 2 5 0 3 11 0 8] - [ 2 1 30 871 1 2 1 3 4 2 14 1 14 2 24 5 2 5 22 4 6] - [ 23 8 3 1 985 8 1 2 2 3 0 8 0 1 7 3 4 0 0 1 6] - [ 5 31 7 10 15 897 4 40 2 7 2 15 5 14 1 3 1 1 2 3 5] - [ 4 1 39 6 2 4 994 6 0 1 3 3 2 1 1 8 1 4 5 9 5] - [ 1 15 13 2 1 20 5 907 2 1 0 6 5 0 0 1 0 0 36 6 3] - [ 16 2 0 1 1 0 0 2 911 28 8 0 1 11 18 1 2 3 2 0 2] - [ 148 2 3 3 6 1 0 0 56 755 1 4 0 16 5 0 1 1 1 0 9] - [ 1 2 10 9 1 1 2 4 20 1 958 1 2 7 3 0 0 0 22 2 5] - [ 4 3 2 1 3 13 1 3 3 2 1 869 54 7 2 7 2 18 2 6 2] - [ 2 1 2 3 2 3 0 0 4 0 2 38 862 0 2 4 3 24 1 0 6] - [ 11 3 4 1 8 11 1 3 26 14 15 9 4 895 6 1 3 1 0 1 7] - [ 15 1 4 20 7 1 0 1 29 3 5 1 8 4 960 1 1 6 15 0 10] - [ 9 2 7 1 8 1 3 1 1 0 0 5 12 2 1 959 6 17 3 3 5] - [ 4 7 3 0 11 2 1 0 4 1 2 10 5 2 5 16 973 2 2 1 10] - [ 8 0 3 6 1 0 1 2 2 0 1 14 27 1 1 20 0 959 0 3 2] - [ 2 8 8 11 3 2 1 27 4 0 3 3 8 1 13 0 1 0 986 1 4] - [ 1 1 5 0 2 5 8 31 0 0 0 31 4 5 0 11 4 2 5 1018 15] - [ 257 283 392 172 219 184 94 204 179 106 243 196 477 319 223 147 277 130 311 247 8774]] - -2023-02-13 17:25:44,515 - ==> Best [Top1: 79.383 Top5: 96.835 Sparsity:0.00 Params: 148928 on epoch: 15] -2023-02-13 17:25:44,515 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:25:44,522 - - -2023-02-13 17:25:44,522 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:25:45,506 - Epoch: [16][ 10/ 1207] Overall Loss 0.372845 Objective Loss 0.372845 LR 0.001000 Time 0.098291 -2023-02-13 17:25:45,707 - Epoch: [16][ 20/ 1207] Overall Loss 0.377152 Objective Loss 0.377152 LR 0.001000 Time 0.059190 -2023-02-13 17:25:45,900 - Epoch: [16][ 30/ 1207] Overall Loss 0.373015 Objective Loss 0.373015 LR 0.001000 Time 0.045894 -2023-02-13 17:25:46,094 - Epoch: [16][ 40/ 1207] Overall Loss 0.366664 Objective Loss 0.366664 LR 0.001000 Time 0.039248 -2023-02-13 17:25:46,285 - Epoch: [16][ 50/ 1207] Overall Loss 0.361539 Objective Loss 0.361539 LR 0.001000 Time 0.035215 -2023-02-13 17:25:46,478 - Epoch: [16][ 60/ 1207] Overall Loss 0.369630 Objective Loss 0.369630 LR 0.001000 Time 0.032556 -2023-02-13 17:25:46,669 - Epoch: [16][ 70/ 1207] Overall Loss 0.377405 Objective Loss 0.377405 LR 0.001000 Time 0.030635 -2023-02-13 17:25:46,863 - Epoch: [16][ 80/ 1207] Overall Loss 0.382322 Objective Loss 0.382322 LR 0.001000 Time 0.029219 -2023-02-13 17:25:47,055 - Epoch: [16][ 90/ 1207] Overall Loss 0.385943 Objective Loss 0.385943 LR 0.001000 Time 0.028108 -2023-02-13 17:25:47,249 - Epoch: [16][ 100/ 1207] Overall Loss 0.386311 Objective Loss 0.386311 LR 0.001000 Time 0.027228 -2023-02-13 17:25:47,440 - Epoch: [16][ 110/ 1207] Overall Loss 0.385015 Objective Loss 0.385015 LR 0.001000 Time 0.026490 -2023-02-13 17:25:47,634 - Epoch: [16][ 120/ 1207] Overall Loss 0.384402 Objective Loss 0.384402 LR 0.001000 Time 0.025889 -2023-02-13 17:25:47,825 - Epoch: [16][ 130/ 1207] Overall Loss 0.387944 Objective Loss 0.387944 LR 0.001000 Time 0.025367 -2023-02-13 17:25:48,019 - Epoch: [16][ 140/ 1207] Overall Loss 0.387831 Objective Loss 0.387831 LR 0.001000 Time 0.024936 -2023-02-13 17:25:48,211 - Epoch: [16][ 150/ 1207] Overall Loss 0.388552 Objective Loss 0.388552 LR 0.001000 Time 0.024551 -2023-02-13 17:25:48,404 - Epoch: [16][ 160/ 1207] Overall Loss 0.391352 Objective Loss 0.391352 LR 0.001000 Time 0.024220 -2023-02-13 17:25:48,596 - Epoch: [16][ 170/ 1207] Overall Loss 0.391175 Objective Loss 0.391175 LR 0.001000 Time 0.023924 -2023-02-13 17:25:48,789 - Epoch: [16][ 180/ 1207] Overall Loss 0.391729 Objective Loss 0.391729 LR 0.001000 Time 0.023667 -2023-02-13 17:25:48,982 - Epoch: [16][ 190/ 1207] Overall Loss 0.391405 Objective Loss 0.391405 LR 0.001000 Time 0.023433 -2023-02-13 17:25:49,175 - Epoch: [16][ 200/ 1207] Overall Loss 0.392101 Objective Loss 0.392101 LR 0.001000 Time 0.023226 -2023-02-13 17:25:49,367 - Epoch: [16][ 210/ 1207] Overall Loss 0.395891 Objective Loss 0.395891 LR 0.001000 Time 0.023032 -2023-02-13 17:25:49,560 - Epoch: [16][ 220/ 1207] Overall Loss 0.396361 Objective Loss 0.396361 LR 0.001000 Time 0.022863 -2023-02-13 17:25:49,752 - Epoch: [16][ 230/ 1207] Overall Loss 0.398183 Objective Loss 0.398183 LR 0.001000 Time 0.022699 -2023-02-13 17:25:49,945 - Epoch: [16][ 240/ 1207] Overall Loss 0.398396 Objective Loss 0.398396 LR 0.001000 Time 0.022558 -2023-02-13 17:25:50,137 - Epoch: [16][ 250/ 1207] Overall Loss 0.399770 Objective Loss 0.399770 LR 0.001000 Time 0.022423 -2023-02-13 17:25:50,331 - Epoch: [16][ 260/ 1207] Overall Loss 0.400195 Objective Loss 0.400195 LR 0.001000 Time 0.022304 -2023-02-13 17:25:50,523 - Epoch: [16][ 270/ 1207] Overall Loss 0.399765 Objective Loss 0.399765 LR 0.001000 Time 0.022187 -2023-02-13 17:25:50,716 - Epoch: [16][ 280/ 1207] Overall Loss 0.398948 Objective Loss 0.398948 LR 0.001000 Time 0.022083 -2023-02-13 17:25:50,909 - Epoch: [16][ 290/ 1207] Overall Loss 0.399248 Objective Loss 0.399248 LR 0.001000 Time 0.021984 -2023-02-13 17:25:51,102 - Epoch: [16][ 300/ 1207] Overall Loss 0.400456 Objective Loss 0.400456 LR 0.001000 Time 0.021894 -2023-02-13 17:25:51,294 - Epoch: [16][ 310/ 1207] Overall Loss 0.400237 Objective Loss 0.400237 LR 0.001000 Time 0.021805 -2023-02-13 17:25:51,487 - Epoch: [16][ 320/ 1207] Overall Loss 0.400829 Objective Loss 0.400829 LR 0.001000 Time 0.021727 -2023-02-13 17:25:51,679 - Epoch: [16][ 330/ 1207] Overall Loss 0.400889 Objective Loss 0.400889 LR 0.001000 Time 0.021650 -2023-02-13 17:25:51,873 - Epoch: [16][ 340/ 1207] Overall Loss 0.400069 Objective Loss 0.400069 LR 0.001000 Time 0.021582 -2023-02-13 17:25:52,065 - Epoch: [16][ 350/ 1207] Overall Loss 0.400170 Objective Loss 0.400170 LR 0.001000 Time 0.021513 -2023-02-13 17:25:52,258 - Epoch: [16][ 360/ 1207] Overall Loss 0.400510 Objective Loss 0.400510 LR 0.001000 Time 0.021452 -2023-02-13 17:25:52,450 - Epoch: [16][ 370/ 1207] Overall Loss 0.400032 Objective Loss 0.400032 LR 0.001000 Time 0.021389 -2023-02-13 17:25:52,643 - Epoch: [16][ 380/ 1207] Overall Loss 0.401364 Objective Loss 0.401364 LR 0.001000 Time 0.021333 -2023-02-13 17:25:52,834 - Epoch: [16][ 390/ 1207] Overall Loss 0.402428 Objective Loss 0.402428 LR 0.001000 Time 0.021275 -2023-02-13 17:25:53,027 - Epoch: [16][ 400/ 1207] Overall Loss 0.402526 Objective Loss 0.402526 LR 0.001000 Time 0.021225 -2023-02-13 17:25:53,219 - Epoch: [16][ 410/ 1207] Overall Loss 0.402753 Objective Loss 0.402753 LR 0.001000 Time 0.021174 -2023-02-13 17:25:53,413 - Epoch: [16][ 420/ 1207] Overall Loss 0.402644 Objective Loss 0.402644 LR 0.001000 Time 0.021130 -2023-02-13 17:25:53,604 - Epoch: [16][ 430/ 1207] Overall Loss 0.403275 Objective Loss 0.403275 LR 0.001000 Time 0.021084 -2023-02-13 17:25:53,797 - Epoch: [16][ 440/ 1207] Overall Loss 0.403765 Objective Loss 0.403765 LR 0.001000 Time 0.021042 -2023-02-13 17:25:53,989 - Epoch: [16][ 450/ 1207] Overall Loss 0.402884 Objective Loss 0.402884 LR 0.001000 Time 0.021000 -2023-02-13 17:25:54,182 - Epoch: [16][ 460/ 1207] Overall Loss 0.403126 Objective Loss 0.403126 LR 0.001000 Time 0.020963 -2023-02-13 17:25:54,375 - Epoch: [16][ 470/ 1207] Overall Loss 0.402919 Objective Loss 0.402919 LR 0.001000 Time 0.020925 -2023-02-13 17:25:54,568 - Epoch: [16][ 480/ 1207] Overall Loss 0.403258 Objective Loss 0.403258 LR 0.001000 Time 0.020892 -2023-02-13 17:25:54,760 - Epoch: [16][ 490/ 1207] Overall Loss 0.403551 Objective Loss 0.403551 LR 0.001000 Time 0.020855 -2023-02-13 17:25:54,954 - Epoch: [16][ 500/ 1207] Overall Loss 0.403353 Objective Loss 0.403353 LR 0.001000 Time 0.020825 -2023-02-13 17:25:55,145 - Epoch: [16][ 510/ 1207] Overall Loss 0.403932 Objective Loss 0.403932 LR 0.001000 Time 0.020792 -2023-02-13 17:25:55,339 - Epoch: [16][ 520/ 1207] Overall Loss 0.404243 Objective Loss 0.404243 LR 0.001000 Time 0.020764 -2023-02-13 17:25:55,531 - Epoch: [16][ 530/ 1207] Overall Loss 0.404775 Objective Loss 0.404775 LR 0.001000 Time 0.020734 -2023-02-13 17:25:55,724 - Epoch: [16][ 540/ 1207] Overall Loss 0.404846 Objective Loss 0.404846 LR 0.001000 Time 0.020707 -2023-02-13 17:25:55,917 - Epoch: [16][ 550/ 1207] Overall Loss 0.405296 Objective Loss 0.405296 LR 0.001000 Time 0.020681 -2023-02-13 17:25:56,111 - Epoch: [16][ 560/ 1207] Overall Loss 0.405333 Objective Loss 0.405333 LR 0.001000 Time 0.020656 -2023-02-13 17:25:56,303 - Epoch: [16][ 570/ 1207] Overall Loss 0.405668 Objective Loss 0.405668 LR 0.001000 Time 0.020630 -2023-02-13 17:25:56,496 - Epoch: [16][ 580/ 1207] Overall Loss 0.405679 Objective Loss 0.405679 LR 0.001000 Time 0.020608 -2023-02-13 17:25:56,688 - Epoch: [16][ 590/ 1207] Overall Loss 0.405844 Objective Loss 0.405844 LR 0.001000 Time 0.020583 -2023-02-13 17:25:56,882 - Epoch: [16][ 600/ 1207] Overall Loss 0.406647 Objective Loss 0.406647 LR 0.001000 Time 0.020563 -2023-02-13 17:25:57,074 - Epoch: [16][ 610/ 1207] Overall Loss 0.406865 Objective Loss 0.406865 LR 0.001000 Time 0.020540 -2023-02-13 17:25:57,268 - Epoch: [16][ 620/ 1207] Overall Loss 0.407264 Objective Loss 0.407264 LR 0.001000 Time 0.020520 -2023-02-13 17:25:57,461 - Epoch: [16][ 630/ 1207] Overall Loss 0.406813 Objective Loss 0.406813 LR 0.001000 Time 0.020499 -2023-02-13 17:25:57,654 - Epoch: [16][ 640/ 1207] Overall Loss 0.406769 Objective Loss 0.406769 LR 0.001000 Time 0.020481 -2023-02-13 17:25:57,846 - Epoch: [16][ 650/ 1207] Overall Loss 0.406741 Objective Loss 0.406741 LR 0.001000 Time 0.020461 -2023-02-13 17:25:58,039 - Epoch: [16][ 660/ 1207] Overall Loss 0.407093 Objective Loss 0.407093 LR 0.001000 Time 0.020443 -2023-02-13 17:25:58,231 - Epoch: [16][ 670/ 1207] Overall Loss 0.407377 Objective Loss 0.407377 LR 0.001000 Time 0.020424 -2023-02-13 17:25:58,425 - Epoch: [16][ 680/ 1207] Overall Loss 0.407571 Objective Loss 0.407571 LR 0.001000 Time 0.020408 -2023-02-13 17:25:58,617 - Epoch: [16][ 690/ 1207] Overall Loss 0.407413 Objective Loss 0.407413 LR 0.001000 Time 0.020390 -2023-02-13 17:25:58,811 - Epoch: [16][ 700/ 1207] Overall Loss 0.407533 Objective Loss 0.407533 LR 0.001000 Time 0.020375 -2023-02-13 17:25:59,004 - Epoch: [16][ 710/ 1207] Overall Loss 0.407438 Objective Loss 0.407438 LR 0.001000 Time 0.020359 -2023-02-13 17:25:59,197 - Epoch: [16][ 720/ 1207] Overall Loss 0.407686 Objective Loss 0.407686 LR 0.001000 Time 0.020344 -2023-02-13 17:25:59,389 - Epoch: [16][ 730/ 1207] Overall Loss 0.407551 Objective Loss 0.407551 LR 0.001000 Time 0.020329 -2023-02-13 17:25:59,583 - Epoch: [16][ 740/ 1207] Overall Loss 0.407119 Objective Loss 0.407119 LR 0.001000 Time 0.020315 -2023-02-13 17:25:59,775 - Epoch: [16][ 750/ 1207] Overall Loss 0.407192 Objective Loss 0.407192 LR 0.001000 Time 0.020300 -2023-02-13 17:25:59,969 - Epoch: [16][ 760/ 1207] Overall Loss 0.407278 Objective Loss 0.407278 LR 0.001000 Time 0.020287 -2023-02-13 17:26:00,161 - Epoch: [16][ 770/ 1207] Overall Loss 0.407226 Objective Loss 0.407226 LR 0.001000 Time 0.020273 -2023-02-13 17:26:00,355 - Epoch: [16][ 780/ 1207] Overall Loss 0.407730 Objective Loss 0.407730 LR 0.001000 Time 0.020261 -2023-02-13 17:26:00,547 - Epoch: [16][ 790/ 1207] Overall Loss 0.407571 Objective Loss 0.407571 LR 0.001000 Time 0.020247 -2023-02-13 17:26:00,741 - Epoch: [16][ 800/ 1207] Overall Loss 0.407525 Objective Loss 0.407525 LR 0.001000 Time 0.020236 -2023-02-13 17:26:00,933 - Epoch: [16][ 810/ 1207] Overall Loss 0.407463 Objective Loss 0.407463 LR 0.001000 Time 0.020223 -2023-02-13 17:26:01,127 - Epoch: [16][ 820/ 1207] Overall Loss 0.407941 Objective Loss 0.407941 LR 0.001000 Time 0.020212 -2023-02-13 17:26:01,320 - Epoch: [16][ 830/ 1207] Overall Loss 0.407990 Objective Loss 0.407990 LR 0.001000 Time 0.020201 -2023-02-13 17:26:01,514 - Epoch: [16][ 840/ 1207] Overall Loss 0.408380 Objective Loss 0.408380 LR 0.001000 Time 0.020190 -2023-02-13 17:26:01,706 - Epoch: [16][ 850/ 1207] Overall Loss 0.408988 Objective Loss 0.408988 LR 0.001000 Time 0.020179 -2023-02-13 17:26:01,899 - Epoch: [16][ 860/ 1207] Overall Loss 0.409411 Objective Loss 0.409411 LR 0.001000 Time 0.020168 -2023-02-13 17:26:02,091 - Epoch: [16][ 870/ 1207] Overall Loss 0.409687 Objective Loss 0.409687 LR 0.001000 Time 0.020157 -2023-02-13 17:26:02,285 - Epoch: [16][ 880/ 1207] Overall Loss 0.409881 Objective Loss 0.409881 LR 0.001000 Time 0.020147 -2023-02-13 17:26:02,477 - Epoch: [16][ 890/ 1207] Overall Loss 0.409789 Objective Loss 0.409789 LR 0.001000 Time 0.020136 -2023-02-13 17:26:02,670 - Epoch: [16][ 900/ 1207] Overall Loss 0.410007 Objective Loss 0.410007 LR 0.001000 Time 0.020127 -2023-02-13 17:26:02,861 - Epoch: [16][ 910/ 1207] Overall Loss 0.409657 Objective Loss 0.409657 LR 0.001000 Time 0.020116 -2023-02-13 17:26:03,055 - Epoch: [16][ 920/ 1207] Overall Loss 0.410066 Objective Loss 0.410066 LR 0.001000 Time 0.020107 -2023-02-13 17:26:03,247 - Epoch: [16][ 930/ 1207] Overall Loss 0.409803 Objective Loss 0.409803 LR 0.001000 Time 0.020096 -2023-02-13 17:26:03,441 - Epoch: [16][ 940/ 1207] Overall Loss 0.409627 Objective Loss 0.409627 LR 0.001000 Time 0.020089 -2023-02-13 17:26:03,634 - Epoch: [16][ 950/ 1207] Overall Loss 0.409502 Objective Loss 0.409502 LR 0.001000 Time 0.020080 -2023-02-13 17:26:03,827 - Epoch: [16][ 960/ 1207] Overall Loss 0.409535 Objective Loss 0.409535 LR 0.001000 Time 0.020072 -2023-02-13 17:26:04,019 - Epoch: [16][ 970/ 1207] Overall Loss 0.409638 Objective Loss 0.409638 LR 0.001000 Time 0.020063 -2023-02-13 17:26:04,212 - Epoch: [16][ 980/ 1207] Overall Loss 0.409630 Objective Loss 0.409630 LR 0.001000 Time 0.020054 -2023-02-13 17:26:04,405 - Epoch: [16][ 990/ 1207] Overall Loss 0.409495 Objective Loss 0.409495 LR 0.001000 Time 0.020046 -2023-02-13 17:26:04,598 - Epoch: [16][ 1000/ 1207] Overall Loss 0.409530 Objective Loss 0.409530 LR 0.001000 Time 0.020039 -2023-02-13 17:26:04,790 - Epoch: [16][ 1010/ 1207] Overall Loss 0.409634 Objective Loss 0.409634 LR 0.001000 Time 0.020030 -2023-02-13 17:26:04,983 - Epoch: [16][ 1020/ 1207] Overall Loss 0.409916 Objective Loss 0.409916 LR 0.001000 Time 0.020023 -2023-02-13 17:26:05,175 - Epoch: [16][ 1030/ 1207] Overall Loss 0.410269 Objective Loss 0.410269 LR 0.001000 Time 0.020014 -2023-02-13 17:26:05,370 - Epoch: [16][ 1040/ 1207] Overall Loss 0.410412 Objective Loss 0.410412 LR 0.001000 Time 0.020009 -2023-02-13 17:26:05,562 - Epoch: [16][ 1050/ 1207] Overall Loss 0.410501 Objective Loss 0.410501 LR 0.001000 Time 0.020000 -2023-02-13 17:26:05,756 - Epoch: [16][ 1060/ 1207] Overall Loss 0.410720 Objective Loss 0.410720 LR 0.001000 Time 0.019995 -2023-02-13 17:26:05,948 - Epoch: [16][ 1070/ 1207] Overall Loss 0.410993 Objective Loss 0.410993 LR 0.001000 Time 0.019987 -2023-02-13 17:26:06,142 - Epoch: [16][ 1080/ 1207] Overall Loss 0.410983 Objective Loss 0.410983 LR 0.001000 Time 0.019981 -2023-02-13 17:26:06,334 - Epoch: [16][ 1090/ 1207] Overall Loss 0.410872 Objective Loss 0.410872 LR 0.001000 Time 0.019973 -2023-02-13 17:26:06,527 - Epoch: [16][ 1100/ 1207] Overall Loss 0.411126 Objective Loss 0.411126 LR 0.001000 Time 0.019967 -2023-02-13 17:26:06,719 - Epoch: [16][ 1110/ 1207] Overall Loss 0.411345 Objective Loss 0.411345 LR 0.001000 Time 0.019960 -2023-02-13 17:26:06,914 - Epoch: [16][ 1120/ 1207] Overall Loss 0.411147 Objective Loss 0.411147 LR 0.001000 Time 0.019955 -2023-02-13 17:26:07,108 - Epoch: [16][ 1130/ 1207] Overall Loss 0.411256 Objective Loss 0.411256 LR 0.001000 Time 0.019950 -2023-02-13 17:26:07,303 - Epoch: [16][ 1140/ 1207] Overall Loss 0.411403 Objective Loss 0.411403 LR 0.001000 Time 0.019946 -2023-02-13 17:26:07,496 - Epoch: [16][ 1150/ 1207] Overall Loss 0.411396 Objective Loss 0.411396 LR 0.001000 Time 0.019940 -2023-02-13 17:26:07,691 - Epoch: [16][ 1160/ 1207] Overall Loss 0.411548 Objective Loss 0.411548 LR 0.001000 Time 0.019936 -2023-02-13 17:26:07,886 - Epoch: [16][ 1170/ 1207] Overall Loss 0.411569 Objective Loss 0.411569 LR 0.001000 Time 0.019932 -2023-02-13 17:26:08,081 - Epoch: [16][ 1180/ 1207] Overall Loss 0.411410 Objective Loss 0.411410 LR 0.001000 Time 0.019927 -2023-02-13 17:26:08,274 - Epoch: [16][ 1190/ 1207] Overall Loss 0.411547 Objective Loss 0.411547 LR 0.001000 Time 0.019922 -2023-02-13 17:26:08,519 - Epoch: [16][ 1200/ 1207] Overall Loss 0.411567 Objective Loss 0.411567 LR 0.001000 Time 0.019960 -2023-02-13 17:26:08,635 - Epoch: [16][ 1207/ 1207] Overall Loss 0.411600 Objective Loss 0.411600 Top1 78.658537 Top5 96.951220 LR 0.001000 Time 0.019940 -2023-02-13 17:26:08,705 - --- validate (epoch=16)----------- -2023-02-13 17:26:08,706 - 34311 samples (256 per mini-batch) -2023-02-13 17:26:09,098 - Epoch: [16][ 10/ 135] Loss 0.385202 Top1 80.195312 Top5 96.914062 -2023-02-13 17:26:09,222 - Epoch: [16][ 20/ 135] Loss 0.408038 Top1 80.078125 Top5 96.914062 -2023-02-13 17:26:09,346 - Epoch: [16][ 30/ 135] Loss 0.426270 Top1 79.596354 Top5 96.757812 -2023-02-13 17:26:09,468 - Epoch: [16][ 40/ 135] Loss 0.424608 Top1 79.902344 Top5 96.904297 -2023-02-13 17:26:09,591 - Epoch: [16][ 50/ 135] Loss 0.430891 Top1 79.617188 Top5 96.851562 -2023-02-13 17:26:09,714 - Epoch: [16][ 60/ 135] Loss 0.433119 Top1 79.544271 Top5 96.790365 -2023-02-13 17:26:09,837 - Epoch: [16][ 70/ 135] Loss 0.437031 Top1 79.441964 Top5 96.774554 -2023-02-13 17:26:09,961 - Epoch: [16][ 80/ 135] Loss 0.437659 Top1 79.433594 Top5 96.791992 -2023-02-13 17:26:10,084 - Epoch: [16][ 90/ 135] Loss 0.435704 Top1 79.492188 Top5 96.792535 -2023-02-13 17:26:10,205 - Epoch: [16][ 100/ 135] Loss 0.434782 Top1 79.589844 Top5 96.769531 -2023-02-13 17:26:10,329 - Epoch: [16][ 110/ 135] Loss 0.436272 Top1 79.612926 Top5 96.732955 -2023-02-13 17:26:10,452 - Epoch: [16][ 120/ 135] Loss 0.431265 Top1 79.687500 Top5 96.764323 -2023-02-13 17:26:10,574 - Epoch: [16][ 130/ 135] Loss 0.429206 Top1 79.675481 Top5 96.775841 -2023-02-13 17:26:10,618 - Epoch: [16][ 135/ 135] Loss 0.429319 Top1 79.668328 Top5 96.785288 -2023-02-13 17:26:10,701 - ==> Top1: 79.668 Top5: 96.785 Loss: 0.429 - -2023-02-13 17:26:10,702 - ==> Confusion: -[[ 784 6 11 1 12 5 0 0 5 101 0 5 2 3 9 9 4 0 1 0 9] - [ 5 902 3 2 16 55 3 13 2 1 7 4 1 0 1 0 3 0 3 3 9] - [ 8 6 936 13 3 3 30 11 0 2 4 6 5 1 2 9 0 1 6 4 8] - [ 2 2 27 888 1 7 1 1 3 4 13 2 5 5 24 4 5 3 11 1 7] - [ 14 9 1 1 992 12 1 0 0 8 0 3 1 5 5 6 2 0 1 2 3] - [ 3 29 2 8 7 958 6 7 2 8 2 3 4 13 0 4 3 2 1 5 3] - [ 5 4 14 7 0 4 1038 1 1 2 3 1 2 2 0 7 1 0 1 5 1] - [ 2 19 19 1 5 66 9 834 2 1 4 5 5 2 0 1 0 1 25 11 12] - [ 21 6 1 1 4 0 0 1 825 68 15 3 1 22 27 2 1 3 3 0 5] - [ 74 0 9 0 11 2 1 0 16 858 1 2 0 22 4 3 0 1 0 0 8] - [ 3 1 10 7 0 1 7 1 14 4 967 0 2 11 1 1 0 1 10 1 9] - [ 5 3 0 1 3 36 1 2 0 3 0 838 54 8 0 15 5 4 2 20 5] - [ 7 0 1 6 2 3 0 1 2 0 1 40 836 2 2 16 3 20 1 2 14] - [ 7 1 5 2 9 20 0 1 13 16 6 11 4 903 3 7 3 0 1 2 10] - [ 15 5 1 25 13 2 0 0 21 10 6 1 5 2 946 5 2 6 9 1 17] - [ 6 3 4 0 7 5 9 0 0 2 0 8 6 4 1 973 6 6 1 2 3] - [ 1 11 3 0 14 5 1 0 2 1 1 7 1 3 3 21 973 2 1 4 7] - [ 10 0 1 4 1 2 4 1 2 2 0 21 45 3 3 51 0 888 0 5 8] - [ 4 9 13 22 2 3 1 31 2 1 18 2 9 0 23 4 0 0 927 1 14] - [ 2 2 2 1 1 16 13 9 0 1 1 27 3 5 0 8 3 0 1 1043 10] - [ 217 298 333 132 276 326 149 106 86 170 242 161 363 380 134 245 270 70 124 326 9026]] - -2023-02-13 17:26:10,703 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:26:10,703 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:26:10,710 - - -2023-02-13 17:26:10,710 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:26:11,601 - Epoch: [17][ 10/ 1207] Overall Loss 0.435783 Objective Loss 0.435783 LR 0.001000 Time 0.089012 -2023-02-13 17:26:11,803 - Epoch: [17][ 20/ 1207] Overall Loss 0.423571 Objective Loss 0.423571 LR 0.001000 Time 0.054604 -2023-02-13 17:26:11,992 - Epoch: [17][ 30/ 1207] Overall Loss 0.412322 Objective Loss 0.412322 LR 0.001000 Time 0.042694 -2023-02-13 17:26:12,179 - Epoch: [17][ 40/ 1207] Overall Loss 0.416997 Objective Loss 0.416997 LR 0.001000 Time 0.036684 -2023-02-13 17:26:12,368 - Epoch: [17][ 50/ 1207] Overall Loss 0.415663 Objective Loss 0.415663 LR 0.001000 Time 0.033110 -2023-02-13 17:26:12,555 - Epoch: [17][ 60/ 1207] Overall Loss 0.420945 Objective Loss 0.420945 LR 0.001000 Time 0.030716 -2023-02-13 17:26:12,743 - Epoch: [17][ 70/ 1207] Overall Loss 0.418926 Objective Loss 0.418926 LR 0.001000 Time 0.029005 -2023-02-13 17:26:12,931 - Epoch: [17][ 80/ 1207] Overall Loss 0.421960 Objective Loss 0.421960 LR 0.001000 Time 0.027722 -2023-02-13 17:26:13,120 - Epoch: [17][ 90/ 1207] Overall Loss 0.418867 Objective Loss 0.418867 LR 0.001000 Time 0.026735 -2023-02-13 17:26:13,308 - Epoch: [17][ 100/ 1207] Overall Loss 0.415748 Objective Loss 0.415748 LR 0.001000 Time 0.025943 -2023-02-13 17:26:13,497 - Epoch: [17][ 110/ 1207] Overall Loss 0.415872 Objective Loss 0.415872 LR 0.001000 Time 0.025294 -2023-02-13 17:26:13,685 - Epoch: [17][ 120/ 1207] Overall Loss 0.416833 Objective Loss 0.416833 LR 0.001000 Time 0.024751 -2023-02-13 17:26:13,873 - Epoch: [17][ 130/ 1207] Overall Loss 0.415551 Objective Loss 0.415551 LR 0.001000 Time 0.024292 -2023-02-13 17:26:14,061 - Epoch: [17][ 140/ 1207] Overall Loss 0.415681 Objective Loss 0.415681 LR 0.001000 Time 0.023900 -2023-02-13 17:26:14,249 - Epoch: [17][ 150/ 1207] Overall Loss 0.414483 Objective Loss 0.414483 LR 0.001000 Time 0.023553 -2023-02-13 17:26:14,438 - Epoch: [17][ 160/ 1207] Overall Loss 0.413604 Objective Loss 0.413604 LR 0.001000 Time 0.023261 -2023-02-13 17:26:14,626 - Epoch: [17][ 170/ 1207] Overall Loss 0.411735 Objective Loss 0.411735 LR 0.001000 Time 0.022995 -2023-02-13 17:26:14,813 - Epoch: [17][ 180/ 1207] Overall Loss 0.410596 Objective Loss 0.410596 LR 0.001000 Time 0.022758 -2023-02-13 17:26:15,002 - Epoch: [17][ 190/ 1207] Overall Loss 0.410813 Objective Loss 0.410813 LR 0.001000 Time 0.022550 -2023-02-13 17:26:15,190 - Epoch: [17][ 200/ 1207] Overall Loss 0.410336 Objective Loss 0.410336 LR 0.001000 Time 0.022362 -2023-02-13 17:26:15,379 - Epoch: [17][ 210/ 1207] Overall Loss 0.409883 Objective Loss 0.409883 LR 0.001000 Time 0.022194 -2023-02-13 17:26:15,568 - Epoch: [17][ 220/ 1207] Overall Loss 0.409945 Objective Loss 0.409945 LR 0.001000 Time 0.022043 -2023-02-13 17:26:15,757 - Epoch: [17][ 230/ 1207] Overall Loss 0.409454 Objective Loss 0.409454 LR 0.001000 Time 0.021905 -2023-02-13 17:26:15,946 - Epoch: [17][ 240/ 1207] Overall Loss 0.409851 Objective Loss 0.409851 LR 0.001000 Time 0.021779 -2023-02-13 17:26:16,135 - Epoch: [17][ 250/ 1207] Overall Loss 0.409021 Objective Loss 0.409021 LR 0.001000 Time 0.021662 -2023-02-13 17:26:16,323 - Epoch: [17][ 260/ 1207] Overall Loss 0.408527 Objective Loss 0.408527 LR 0.001000 Time 0.021552 -2023-02-13 17:26:16,512 - Epoch: [17][ 270/ 1207] Overall Loss 0.408219 Objective Loss 0.408219 LR 0.001000 Time 0.021453 -2023-02-13 17:26:16,701 - Epoch: [17][ 280/ 1207] Overall Loss 0.408189 Objective Loss 0.408189 LR 0.001000 Time 0.021360 -2023-02-13 17:26:16,891 - Epoch: [17][ 290/ 1207] Overall Loss 0.407452 Objective Loss 0.407452 LR 0.001000 Time 0.021276 -2023-02-13 17:26:17,079 - Epoch: [17][ 300/ 1207] Overall Loss 0.407749 Objective Loss 0.407749 LR 0.001000 Time 0.021193 -2023-02-13 17:26:17,268 - Epoch: [17][ 310/ 1207] Overall Loss 0.407880 Objective Loss 0.407880 LR 0.001000 Time 0.021117 -2023-02-13 17:26:17,456 - Epoch: [17][ 320/ 1207] Overall Loss 0.408740 Objective Loss 0.408740 LR 0.001000 Time 0.021045 -2023-02-13 17:26:17,645 - Epoch: [17][ 330/ 1207] Overall Loss 0.408218 Objective Loss 0.408218 LR 0.001000 Time 0.020978 -2023-02-13 17:26:17,834 - Epoch: [17][ 340/ 1207] Overall Loss 0.408128 Objective Loss 0.408128 LR 0.001000 Time 0.020915 -2023-02-13 17:26:18,023 - Epoch: [17][ 350/ 1207] Overall Loss 0.408158 Objective Loss 0.408158 LR 0.001000 Time 0.020856 -2023-02-13 17:26:18,211 - Epoch: [17][ 360/ 1207] Overall Loss 0.407388 Objective Loss 0.407388 LR 0.001000 Time 0.020798 -2023-02-13 17:26:18,400 - Epoch: [17][ 370/ 1207] Overall Loss 0.407766 Objective Loss 0.407766 LR 0.001000 Time 0.020746 -2023-02-13 17:26:18,589 - Epoch: [17][ 380/ 1207] Overall Loss 0.406315 Objective Loss 0.406315 LR 0.001000 Time 0.020697 -2023-02-13 17:26:18,778 - Epoch: [17][ 390/ 1207] Overall Loss 0.405936 Objective Loss 0.405936 LR 0.001000 Time 0.020650 -2023-02-13 17:26:18,966 - Epoch: [17][ 400/ 1207] Overall Loss 0.405950 Objective Loss 0.405950 LR 0.001000 Time 0.020603 -2023-02-13 17:26:19,155 - Epoch: [17][ 410/ 1207] Overall Loss 0.406636 Objective Loss 0.406636 LR 0.001000 Time 0.020561 -2023-02-13 17:26:19,344 - Epoch: [17][ 420/ 1207] Overall Loss 0.406089 Objective Loss 0.406089 LR 0.001000 Time 0.020519 -2023-02-13 17:26:19,532 - Epoch: [17][ 430/ 1207] Overall Loss 0.406528 Objective Loss 0.406528 LR 0.001000 Time 0.020479 -2023-02-13 17:26:19,721 - Epoch: [17][ 440/ 1207] Overall Loss 0.406846 Objective Loss 0.406846 LR 0.001000 Time 0.020443 -2023-02-13 17:26:19,912 - Epoch: [17][ 450/ 1207] Overall Loss 0.406970 Objective Loss 0.406970 LR 0.001000 Time 0.020411 -2023-02-13 17:26:20,101 - Epoch: [17][ 460/ 1207] Overall Loss 0.407275 Objective Loss 0.407275 LR 0.001000 Time 0.020378 -2023-02-13 17:26:20,291 - Epoch: [17][ 470/ 1207] Overall Loss 0.407226 Objective Loss 0.407226 LR 0.001000 Time 0.020348 -2023-02-13 17:26:20,481 - Epoch: [17][ 480/ 1207] Overall Loss 0.408176 Objective Loss 0.408176 LR 0.001000 Time 0.020320 -2023-02-13 17:26:20,672 - Epoch: [17][ 490/ 1207] Overall Loss 0.408951 Objective Loss 0.408951 LR 0.001000 Time 0.020293 -2023-02-13 17:26:20,863 - Epoch: [17][ 500/ 1207] Overall Loss 0.408652 Objective Loss 0.408652 LR 0.001000 Time 0.020268 -2023-02-13 17:26:21,053 - Epoch: [17][ 510/ 1207] Overall Loss 0.408545 Objective Loss 0.408545 LR 0.001000 Time 0.020244 -2023-02-13 17:26:21,242 - Epoch: [17][ 520/ 1207] Overall Loss 0.408307 Objective Loss 0.408307 LR 0.001000 Time 0.020218 -2023-02-13 17:26:21,433 - Epoch: [17][ 530/ 1207] Overall Loss 0.408651 Objective Loss 0.408651 LR 0.001000 Time 0.020196 -2023-02-13 17:26:21,623 - Epoch: [17][ 540/ 1207] Overall Loss 0.408674 Objective Loss 0.408674 LR 0.001000 Time 0.020173 -2023-02-13 17:26:21,814 - Epoch: [17][ 550/ 1207] Overall Loss 0.408742 Objective Loss 0.408742 LR 0.001000 Time 0.020152 -2023-02-13 17:26:22,004 - Epoch: [17][ 560/ 1207] Overall Loss 0.409138 Objective Loss 0.409138 LR 0.001000 Time 0.020130 -2023-02-13 17:26:22,193 - Epoch: [17][ 570/ 1207] Overall Loss 0.409218 Objective Loss 0.409218 LR 0.001000 Time 0.020109 -2023-02-13 17:26:22,383 - Epoch: [17][ 580/ 1207] Overall Loss 0.409199 Objective Loss 0.409199 LR 0.001000 Time 0.020089 -2023-02-13 17:26:22,573 - Epoch: [17][ 590/ 1207] Overall Loss 0.409802 Objective Loss 0.409802 LR 0.001000 Time 0.020069 -2023-02-13 17:26:22,762 - Epoch: [17][ 600/ 1207] Overall Loss 0.409320 Objective Loss 0.409320 LR 0.001000 Time 0.020050 -2023-02-13 17:26:22,953 - Epoch: [17][ 610/ 1207] Overall Loss 0.408970 Objective Loss 0.408970 LR 0.001000 Time 0.020033 -2023-02-13 17:26:23,142 - Epoch: [17][ 620/ 1207] Overall Loss 0.409766 Objective Loss 0.409766 LR 0.001000 Time 0.020015 -2023-02-13 17:26:23,332 - Epoch: [17][ 630/ 1207] Overall Loss 0.409667 Objective Loss 0.409667 LR 0.001000 Time 0.019998 -2023-02-13 17:26:23,523 - Epoch: [17][ 640/ 1207] Overall Loss 0.409636 Objective Loss 0.409636 LR 0.001000 Time 0.019983 -2023-02-13 17:26:23,714 - Epoch: [17][ 650/ 1207] Overall Loss 0.409607 Objective Loss 0.409607 LR 0.001000 Time 0.019969 -2023-02-13 17:26:23,904 - Epoch: [17][ 660/ 1207] Overall Loss 0.409645 Objective Loss 0.409645 LR 0.001000 Time 0.019954 -2023-02-13 17:26:24,094 - Epoch: [17][ 670/ 1207] Overall Loss 0.409550 Objective Loss 0.409550 LR 0.001000 Time 0.019940 -2023-02-13 17:26:24,285 - Epoch: [17][ 680/ 1207] Overall Loss 0.409622 Objective Loss 0.409622 LR 0.001000 Time 0.019926 -2023-02-13 17:26:24,476 - Epoch: [17][ 690/ 1207] Overall Loss 0.409692 Objective Loss 0.409692 LR 0.001000 Time 0.019913 -2023-02-13 17:26:24,665 - Epoch: [17][ 700/ 1207] Overall Loss 0.409842 Objective Loss 0.409842 LR 0.001000 Time 0.019900 -2023-02-13 17:26:24,856 - Epoch: [17][ 710/ 1207] Overall Loss 0.410120 Objective Loss 0.410120 LR 0.001000 Time 0.019887 -2023-02-13 17:26:25,047 - Epoch: [17][ 720/ 1207] Overall Loss 0.409545 Objective Loss 0.409545 LR 0.001000 Time 0.019875 -2023-02-13 17:26:25,237 - Epoch: [17][ 730/ 1207] Overall Loss 0.409562 Objective Loss 0.409562 LR 0.001000 Time 0.019863 -2023-02-13 17:26:25,427 - Epoch: [17][ 740/ 1207] Overall Loss 0.409862 Objective Loss 0.409862 LR 0.001000 Time 0.019851 -2023-02-13 17:26:25,617 - Epoch: [17][ 750/ 1207] Overall Loss 0.410068 Objective Loss 0.410068 LR 0.001000 Time 0.019839 -2023-02-13 17:26:25,808 - Epoch: [17][ 760/ 1207] Overall Loss 0.410052 Objective Loss 0.410052 LR 0.001000 Time 0.019829 -2023-02-13 17:26:25,999 - Epoch: [17][ 770/ 1207] Overall Loss 0.409872 Objective Loss 0.409872 LR 0.001000 Time 0.019819 -2023-02-13 17:26:26,188 - Epoch: [17][ 780/ 1207] Overall Loss 0.409589 Objective Loss 0.409589 LR 0.001000 Time 0.019807 -2023-02-13 17:26:26,379 - Epoch: [17][ 790/ 1207] Overall Loss 0.408799 Objective Loss 0.408799 LR 0.001000 Time 0.019797 -2023-02-13 17:26:26,569 - Epoch: [17][ 800/ 1207] Overall Loss 0.408784 Objective Loss 0.408784 LR 0.001000 Time 0.019786 -2023-02-13 17:26:26,759 - Epoch: [17][ 810/ 1207] Overall Loss 0.408813 Objective Loss 0.408813 LR 0.001000 Time 0.019777 -2023-02-13 17:26:26,949 - Epoch: [17][ 820/ 1207] Overall Loss 0.408606 Objective Loss 0.408606 LR 0.001000 Time 0.019767 -2023-02-13 17:26:27,139 - Epoch: [17][ 830/ 1207] Overall Loss 0.408534 Objective Loss 0.408534 LR 0.001000 Time 0.019758 -2023-02-13 17:26:27,329 - Epoch: [17][ 840/ 1207] Overall Loss 0.408093 Objective Loss 0.408093 LR 0.001000 Time 0.019748 -2023-02-13 17:26:27,520 - Epoch: [17][ 850/ 1207] Overall Loss 0.407764 Objective Loss 0.407764 LR 0.001000 Time 0.019739 -2023-02-13 17:26:27,710 - Epoch: [17][ 860/ 1207] Overall Loss 0.407766 Objective Loss 0.407766 LR 0.001000 Time 0.019731 -2023-02-13 17:26:27,900 - Epoch: [17][ 870/ 1207] Overall Loss 0.407907 Objective Loss 0.407907 LR 0.001000 Time 0.019722 -2023-02-13 17:26:28,090 - Epoch: [17][ 880/ 1207] Overall Loss 0.407956 Objective Loss 0.407956 LR 0.001000 Time 0.019714 -2023-02-13 17:26:28,281 - Epoch: [17][ 890/ 1207] Overall Loss 0.407787 Objective Loss 0.407787 LR 0.001000 Time 0.019705 -2023-02-13 17:26:28,471 - Epoch: [17][ 900/ 1207] Overall Loss 0.407750 Objective Loss 0.407750 LR 0.001000 Time 0.019698 -2023-02-13 17:26:28,661 - Epoch: [17][ 910/ 1207] Overall Loss 0.408270 Objective Loss 0.408270 LR 0.001000 Time 0.019689 -2023-02-13 17:26:28,850 - Epoch: [17][ 920/ 1207] Overall Loss 0.408254 Objective Loss 0.408254 LR 0.001000 Time 0.019680 -2023-02-13 17:26:29,039 - Epoch: [17][ 930/ 1207] Overall Loss 0.408453 Objective Loss 0.408453 LR 0.001000 Time 0.019672 -2023-02-13 17:26:29,228 - Epoch: [17][ 940/ 1207] Overall Loss 0.408671 Objective Loss 0.408671 LR 0.001000 Time 0.019663 -2023-02-13 17:26:29,417 - Epoch: [17][ 950/ 1207] Overall Loss 0.409131 Objective Loss 0.409131 LR 0.001000 Time 0.019655 -2023-02-13 17:26:29,606 - Epoch: [17][ 960/ 1207] Overall Loss 0.409111 Objective Loss 0.409111 LR 0.001000 Time 0.019647 -2023-02-13 17:26:29,795 - Epoch: [17][ 970/ 1207] Overall Loss 0.408792 Objective Loss 0.408792 LR 0.001000 Time 0.019638 -2023-02-13 17:26:29,984 - Epoch: [17][ 980/ 1207] Overall Loss 0.409057 Objective Loss 0.409057 LR 0.001000 Time 0.019630 -2023-02-13 17:26:30,173 - Epoch: [17][ 990/ 1207] Overall Loss 0.408898 Objective Loss 0.408898 LR 0.001000 Time 0.019623 -2023-02-13 17:26:30,361 - Epoch: [17][ 1000/ 1207] Overall Loss 0.408736 Objective Loss 0.408736 LR 0.001000 Time 0.019615 -2023-02-13 17:26:30,559 - Epoch: [17][ 1010/ 1207] Overall Loss 0.408244 Objective Loss 0.408244 LR 0.001000 Time 0.019616 -2023-02-13 17:26:30,763 - Epoch: [17][ 1020/ 1207] Overall Loss 0.408225 Objective Loss 0.408225 LR 0.001000 Time 0.019623 -2023-02-13 17:26:30,961 - Epoch: [17][ 1030/ 1207] Overall Loss 0.408193 Objective Loss 0.408193 LR 0.001000 Time 0.019625 -2023-02-13 17:26:31,164 - Epoch: [17][ 1040/ 1207] Overall Loss 0.408525 Objective Loss 0.408525 LR 0.001000 Time 0.019631 -2023-02-13 17:26:31,362 - Epoch: [17][ 1050/ 1207] Overall Loss 0.408711 Objective Loss 0.408711 LR 0.001000 Time 0.019631 -2023-02-13 17:26:31,564 - Epoch: [17][ 1060/ 1207] Overall Loss 0.408659 Objective Loss 0.408659 LR 0.001000 Time 0.019636 -2023-02-13 17:26:31,761 - Epoch: [17][ 1070/ 1207] Overall Loss 0.408620 Objective Loss 0.408620 LR 0.001000 Time 0.019637 -2023-02-13 17:26:31,963 - Epoch: [17][ 1080/ 1207] Overall Loss 0.408693 Objective Loss 0.408693 LR 0.001000 Time 0.019642 -2023-02-13 17:26:32,160 - Epoch: [17][ 1090/ 1207] Overall Loss 0.408700 Objective Loss 0.408700 LR 0.001000 Time 0.019642 -2023-02-13 17:26:32,363 - Epoch: [17][ 1100/ 1207] Overall Loss 0.408539 Objective Loss 0.408539 LR 0.001000 Time 0.019648 -2023-02-13 17:26:32,561 - Epoch: [17][ 1110/ 1207] Overall Loss 0.408315 Objective Loss 0.408315 LR 0.001000 Time 0.019648 -2023-02-13 17:26:32,763 - Epoch: [17][ 1120/ 1207] Overall Loss 0.408350 Objective Loss 0.408350 LR 0.001000 Time 0.019654 -2023-02-13 17:26:32,961 - Epoch: [17][ 1130/ 1207] Overall Loss 0.408412 Objective Loss 0.408412 LR 0.001000 Time 0.019654 -2023-02-13 17:26:33,163 - Epoch: [17][ 1140/ 1207] Overall Loss 0.408650 Objective Loss 0.408650 LR 0.001000 Time 0.019659 -2023-02-13 17:26:33,361 - Epoch: [17][ 1150/ 1207] Overall Loss 0.408690 Objective Loss 0.408690 LR 0.001000 Time 0.019660 -2023-02-13 17:26:33,565 - Epoch: [17][ 1160/ 1207] Overall Loss 0.408765 Objective Loss 0.408765 LR 0.001000 Time 0.019666 -2023-02-13 17:26:33,763 - Epoch: [17][ 1170/ 1207] Overall Loss 0.408696 Objective Loss 0.408696 LR 0.001000 Time 0.019667 -2023-02-13 17:26:33,965 - Epoch: [17][ 1180/ 1207] Overall Loss 0.408544 Objective Loss 0.408544 LR 0.001000 Time 0.019671 -2023-02-13 17:26:34,163 - Epoch: [17][ 1190/ 1207] Overall Loss 0.408244 Objective Loss 0.408244 LR 0.001000 Time 0.019672 -2023-02-13 17:26:34,418 - Epoch: [17][ 1200/ 1207] Overall Loss 0.408157 Objective Loss 0.408157 LR 0.001000 Time 0.019720 -2023-02-13 17:26:34,534 - Epoch: [17][ 1207/ 1207] Overall Loss 0.407872 Objective Loss 0.407872 Top1 84.756098 Top5 96.341463 LR 0.001000 Time 0.019702 -2023-02-13 17:26:34,605 - --- validate (epoch=17)----------- -2023-02-13 17:26:34,605 - 34311 samples (256 per mini-batch) -2023-02-13 17:26:34,998 - Epoch: [17][ 10/ 135] Loss 0.439428 Top1 77.617188 Top5 96.250000 -2023-02-13 17:26:35,127 - Epoch: [17][ 20/ 135] Loss 0.408381 Top1 78.339844 Top5 96.523438 -2023-02-13 17:26:35,254 - Epoch: [17][ 30/ 135] Loss 0.396540 Top1 78.619792 Top5 96.770833 -2023-02-13 17:26:35,383 - Epoch: [17][ 40/ 135] Loss 0.396754 Top1 78.583984 Top5 96.806641 -2023-02-13 17:26:35,507 - Epoch: [17][ 50/ 135] Loss 0.401852 Top1 78.625000 Top5 96.773438 -2023-02-13 17:26:35,632 - Epoch: [17][ 60/ 135] Loss 0.398505 Top1 78.671875 Top5 96.770833 -2023-02-13 17:26:35,761 - Epoch: [17][ 70/ 135] Loss 0.405413 Top1 78.420759 Top5 96.690848 -2023-02-13 17:26:35,886 - Epoch: [17][ 80/ 135] Loss 0.410352 Top1 78.344727 Top5 96.655273 -2023-02-13 17:26:36,011 - Epoch: [17][ 90/ 135] Loss 0.409889 Top1 78.511285 Top5 96.697049 -2023-02-13 17:26:36,140 - Epoch: [17][ 100/ 135] Loss 0.408834 Top1 78.492188 Top5 96.632812 -2023-02-13 17:26:36,265 - Epoch: [17][ 110/ 135] Loss 0.404376 Top1 78.579545 Top5 96.679688 -2023-02-13 17:26:36,391 - Epoch: [17][ 120/ 135] Loss 0.405454 Top1 78.346354 Top5 96.643880 -2023-02-13 17:26:36,520 - Epoch: [17][ 130/ 135] Loss 0.406663 Top1 78.287260 Top5 96.586538 -2023-02-13 17:26:36,565 - Epoch: [17][ 135/ 135] Loss 0.406593 Top1 78.295590 Top5 96.566699 -2023-02-13 17:26:36,641 - ==> Top1: 78.296 Top5: 96.567 Loss: 0.407 - -2023-02-13 17:26:36,642 - ==> Confusion: -[[ 862 4 5 1 8 6 0 1 4 41 1 4 1 6 5 4 6 2 1 0 5] - [ 8 929 2 4 7 32 3 19 3 1 2 0 1 2 2 0 3 0 5 3 7] - [ 15 4 942 18 1 3 13 22 0 0 7 1 3 5 5 7 2 1 7 0 2] - [ 9 3 25 890 0 0 0 2 1 2 17 0 7 2 25 1 3 4 19 0 6] - [ 27 11 1 2 971 10 1 3 1 2 1 2 0 6 5 5 8 2 2 2 4] - [ 8 45 7 3 5 921 3 22 2 7 2 8 8 8 1 3 4 2 2 6 3] - [ 5 4 34 5 0 3 1013 13 0 0 5 1 3 1 1 2 4 1 0 3 1] - [ 4 15 16 0 1 30 0 908 0 2 1 4 6 1 0 1 1 0 25 5 4] - [ 32 2 0 1 0 1 0 1 874 32 16 1 1 13 24 1 2 2 4 1 1] - [ 156 0 3 0 7 2 0 1 71 729 0 0 0 25 8 2 1 3 0 0 4] - [ 4 2 7 7 0 0 4 4 12 1 973 1 1 13 3 0 1 2 14 0 2] - [ 5 5 1 0 2 17 0 5 2 3 1 858 64 7 0 5 4 8 4 9 5] - [ 5 1 1 4 1 3 0 0 1 0 0 35 877 0 3 1 5 16 1 1 4] - [ 14 4 5 0 6 19 2 3 12 6 9 9 4 913 4 4 6 1 0 1 2] - [ 17 3 2 24 6 4 0 0 24 4 6 1 4 2 967 0 3 6 11 0 8] - [ 9 3 5 5 4 3 9 3 0 1 0 5 13 0 1 947 9 16 0 3 10] - [ 6 11 2 2 6 1 1 0 1 1 2 3 1 4 3 19 985 5 2 1 5] - [ 14 1 1 5 0 1 0 2 1 0 1 5 36 2 3 11 0 962 0 0 6] - [ 7 6 8 13 1 3 0 33 3 1 4 2 10 0 15 0 4 0 972 2 2] - [ 2 6 1 0 0 12 7 26 0 0 3 29 11 6 0 4 8 4 2 1021 6] - [ 344 385 341 180 183 281 102 228 121 86 302 125 499 375 261 103 511 118 251 288 8350]] - -2023-02-13 17:26:36,643 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:26:36,643 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:26:36,649 - - -2023-02-13 17:26:36,649 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:26:37,537 - Epoch: [18][ 10/ 1207] Overall Loss 0.419147 Objective Loss 0.419147 LR 0.001000 Time 0.088739 -2023-02-13 17:26:37,734 - Epoch: [18][ 20/ 1207] Overall Loss 0.411570 Objective Loss 0.411570 LR 0.001000 Time 0.054200 -2023-02-13 17:26:37,923 - Epoch: [18][ 30/ 1207] Overall Loss 0.399135 Objective Loss 0.399135 LR 0.001000 Time 0.042425 -2023-02-13 17:26:38,112 - Epoch: [18][ 40/ 1207] Overall Loss 0.387390 Objective Loss 0.387390 LR 0.001000 Time 0.036538 -2023-02-13 17:26:38,301 - Epoch: [18][ 50/ 1207] Overall Loss 0.381659 Objective Loss 0.381659 LR 0.001000 Time 0.032995 -2023-02-13 17:26:38,490 - Epoch: [18][ 60/ 1207] Overall Loss 0.378056 Objective Loss 0.378056 LR 0.001000 Time 0.030650 -2023-02-13 17:26:38,680 - Epoch: [18][ 70/ 1207] Overall Loss 0.373729 Objective Loss 0.373729 LR 0.001000 Time 0.028972 -2023-02-13 17:26:38,869 - Epoch: [18][ 80/ 1207] Overall Loss 0.373696 Objective Loss 0.373696 LR 0.001000 Time 0.027705 -2023-02-13 17:26:39,058 - Epoch: [18][ 90/ 1207] Overall Loss 0.376080 Objective Loss 0.376080 LR 0.001000 Time 0.026722 -2023-02-13 17:26:39,246 - Epoch: [18][ 100/ 1207] Overall Loss 0.378687 Objective Loss 0.378687 LR 0.001000 Time 0.025932 -2023-02-13 17:26:39,435 - Epoch: [18][ 110/ 1207] Overall Loss 0.380672 Objective Loss 0.380672 LR 0.001000 Time 0.025286 -2023-02-13 17:26:39,624 - Epoch: [18][ 120/ 1207] Overall Loss 0.381662 Objective Loss 0.381662 LR 0.001000 Time 0.024754 -2023-02-13 17:26:39,814 - Epoch: [18][ 130/ 1207] Overall Loss 0.379748 Objective Loss 0.379748 LR 0.001000 Time 0.024304 -2023-02-13 17:26:40,004 - Epoch: [18][ 140/ 1207] Overall Loss 0.383982 Objective Loss 0.383982 LR 0.001000 Time 0.023924 -2023-02-13 17:26:40,193 - Epoch: [18][ 150/ 1207] Overall Loss 0.383579 Objective Loss 0.383579 LR 0.001000 Time 0.023588 -2023-02-13 17:26:40,382 - Epoch: [18][ 160/ 1207] Overall Loss 0.385433 Objective Loss 0.385433 LR 0.001000 Time 0.023291 -2023-02-13 17:26:40,572 - Epoch: [18][ 170/ 1207] Overall Loss 0.385979 Objective Loss 0.385979 LR 0.001000 Time 0.023035 -2023-02-13 17:26:40,761 - Epoch: [18][ 180/ 1207] Overall Loss 0.385163 Objective Loss 0.385163 LR 0.001000 Time 0.022808 -2023-02-13 17:26:40,951 - Epoch: [18][ 190/ 1207] Overall Loss 0.384549 Objective Loss 0.384549 LR 0.001000 Time 0.022603 -2023-02-13 17:26:41,140 - Epoch: [18][ 200/ 1207] Overall Loss 0.385013 Objective Loss 0.385013 LR 0.001000 Time 0.022417 -2023-02-13 17:26:41,329 - Epoch: [18][ 210/ 1207] Overall Loss 0.384063 Objective Loss 0.384063 LR 0.001000 Time 0.022249 -2023-02-13 17:26:41,520 - Epoch: [18][ 220/ 1207] Overall Loss 0.385066 Objective Loss 0.385066 LR 0.001000 Time 0.022103 -2023-02-13 17:26:41,711 - Epoch: [18][ 230/ 1207] Overall Loss 0.385461 Objective Loss 0.385461 LR 0.001000 Time 0.021971 -2023-02-13 17:26:41,903 - Epoch: [18][ 240/ 1207] Overall Loss 0.386554 Objective Loss 0.386554 LR 0.001000 Time 0.021855 -2023-02-13 17:26:42,095 - Epoch: [18][ 250/ 1207] Overall Loss 0.389040 Objective Loss 0.389040 LR 0.001000 Time 0.021745 -2023-02-13 17:26:42,286 - Epoch: [18][ 260/ 1207] Overall Loss 0.388945 Objective Loss 0.388945 LR 0.001000 Time 0.021644 -2023-02-13 17:26:42,478 - Epoch: [18][ 270/ 1207] Overall Loss 0.389365 Objective Loss 0.389365 LR 0.001000 Time 0.021551 -2023-02-13 17:26:42,670 - Epoch: [18][ 280/ 1207] Overall Loss 0.389326 Objective Loss 0.389326 LR 0.001000 Time 0.021467 -2023-02-13 17:26:42,862 - Epoch: [18][ 290/ 1207] Overall Loss 0.389893 Objective Loss 0.389893 LR 0.001000 Time 0.021386 -2023-02-13 17:26:43,055 - Epoch: [18][ 300/ 1207] Overall Loss 0.390454 Objective Loss 0.390454 LR 0.001000 Time 0.021315 -2023-02-13 17:26:43,246 - Epoch: [18][ 310/ 1207] Overall Loss 0.391660 Objective Loss 0.391660 LR 0.001000 Time 0.021243 -2023-02-13 17:26:43,439 - Epoch: [18][ 320/ 1207] Overall Loss 0.391633 Objective Loss 0.391633 LR 0.001000 Time 0.021180 -2023-02-13 17:26:43,631 - Epoch: [18][ 330/ 1207] Overall Loss 0.391579 Objective Loss 0.391579 LR 0.001000 Time 0.021120 -2023-02-13 17:26:43,823 - Epoch: [18][ 340/ 1207] Overall Loss 0.391821 Objective Loss 0.391821 LR 0.001000 Time 0.021063 -2023-02-13 17:26:44,016 - Epoch: [18][ 350/ 1207] Overall Loss 0.391399 Objective Loss 0.391399 LR 0.001000 Time 0.021010 -2023-02-13 17:26:44,204 - Epoch: [18][ 360/ 1207] Overall Loss 0.392947 Objective Loss 0.392947 LR 0.001000 Time 0.020950 -2023-02-13 17:26:44,393 - Epoch: [18][ 370/ 1207] Overall Loss 0.392272 Objective Loss 0.392272 LR 0.001000 Time 0.020893 -2023-02-13 17:26:44,583 - Epoch: [18][ 380/ 1207] Overall Loss 0.393144 Objective Loss 0.393144 LR 0.001000 Time 0.020842 -2023-02-13 17:26:44,773 - Epoch: [18][ 390/ 1207] Overall Loss 0.393223 Objective Loss 0.393223 LR 0.001000 Time 0.020792 -2023-02-13 17:26:44,962 - Epoch: [18][ 400/ 1207] Overall Loss 0.393696 Objective Loss 0.393696 LR 0.001000 Time 0.020745 -2023-02-13 17:26:45,152 - Epoch: [18][ 410/ 1207] Overall Loss 0.393167 Objective Loss 0.393167 LR 0.001000 Time 0.020701 -2023-02-13 17:26:45,341 - Epoch: [18][ 420/ 1207] Overall Loss 0.393136 Objective Loss 0.393136 LR 0.001000 Time 0.020658 -2023-02-13 17:26:45,531 - Epoch: [18][ 430/ 1207] Overall Loss 0.393619 Objective Loss 0.393619 LR 0.001000 Time 0.020618 -2023-02-13 17:26:45,721 - Epoch: [18][ 440/ 1207] Overall Loss 0.394547 Objective Loss 0.394547 LR 0.001000 Time 0.020582 -2023-02-13 17:26:45,911 - Epoch: [18][ 450/ 1207] Overall Loss 0.394507 Objective Loss 0.394507 LR 0.001000 Time 0.020545 -2023-02-13 17:26:46,101 - Epoch: [18][ 460/ 1207] Overall Loss 0.395194 Objective Loss 0.395194 LR 0.001000 Time 0.020510 -2023-02-13 17:26:46,289 - Epoch: [18][ 470/ 1207] Overall Loss 0.395858 Objective Loss 0.395858 LR 0.001000 Time 0.020474 -2023-02-13 17:26:46,478 - Epoch: [18][ 480/ 1207] Overall Loss 0.396622 Objective Loss 0.396622 LR 0.001000 Time 0.020441 -2023-02-13 17:26:46,667 - Epoch: [18][ 490/ 1207] Overall Loss 0.396093 Objective Loss 0.396093 LR 0.001000 Time 0.020408 -2023-02-13 17:26:46,857 - Epoch: [18][ 500/ 1207] Overall Loss 0.395403 Objective Loss 0.395403 LR 0.001000 Time 0.020379 -2023-02-13 17:26:47,047 - Epoch: [18][ 510/ 1207] Overall Loss 0.395627 Objective Loss 0.395627 LR 0.001000 Time 0.020350 -2023-02-13 17:26:47,235 - Epoch: [18][ 520/ 1207] Overall Loss 0.395013 Objective Loss 0.395013 LR 0.001000 Time 0.020321 -2023-02-13 17:26:47,424 - Epoch: [18][ 530/ 1207] Overall Loss 0.395575 Objective Loss 0.395575 LR 0.001000 Time 0.020293 -2023-02-13 17:26:47,614 - Epoch: [18][ 540/ 1207] Overall Loss 0.396103 Objective Loss 0.396103 LR 0.001000 Time 0.020268 -2023-02-13 17:26:47,804 - Epoch: [18][ 550/ 1207] Overall Loss 0.395628 Objective Loss 0.395628 LR 0.001000 Time 0.020244 -2023-02-13 17:26:47,993 - Epoch: [18][ 560/ 1207] Overall Loss 0.395477 Objective Loss 0.395477 LR 0.001000 Time 0.020221 -2023-02-13 17:26:48,182 - Epoch: [18][ 570/ 1207] Overall Loss 0.395788 Objective Loss 0.395788 LR 0.001000 Time 0.020196 -2023-02-13 17:26:48,371 - Epoch: [18][ 580/ 1207] Overall Loss 0.395562 Objective Loss 0.395562 LR 0.001000 Time 0.020173 -2023-02-13 17:26:48,561 - Epoch: [18][ 590/ 1207] Overall Loss 0.396533 Objective Loss 0.396533 LR 0.001000 Time 0.020152 -2023-02-13 17:26:48,750 - Epoch: [18][ 600/ 1207] Overall Loss 0.396249 Objective Loss 0.396249 LR 0.001000 Time 0.020131 -2023-02-13 17:26:48,939 - Epoch: [18][ 610/ 1207] Overall Loss 0.396999 Objective Loss 0.396999 LR 0.001000 Time 0.020110 -2023-02-13 17:26:49,128 - Epoch: [18][ 620/ 1207] Overall Loss 0.397580 Objective Loss 0.397580 LR 0.001000 Time 0.020090 -2023-02-13 17:26:49,317 - Epoch: [18][ 630/ 1207] Overall Loss 0.398020 Objective Loss 0.398020 LR 0.001000 Time 0.020070 -2023-02-13 17:26:49,507 - Epoch: [18][ 640/ 1207] Overall Loss 0.397908 Objective Loss 0.397908 LR 0.001000 Time 0.020053 -2023-02-13 17:26:49,695 - Epoch: [18][ 650/ 1207] Overall Loss 0.397475 Objective Loss 0.397475 LR 0.001000 Time 0.020034 -2023-02-13 17:26:49,884 - Epoch: [18][ 660/ 1207] Overall Loss 0.397801 Objective Loss 0.397801 LR 0.001000 Time 0.020016 -2023-02-13 17:26:50,074 - Epoch: [18][ 670/ 1207] Overall Loss 0.398215 Objective Loss 0.398215 LR 0.001000 Time 0.020000 -2023-02-13 17:26:50,263 - Epoch: [18][ 680/ 1207] Overall Loss 0.398895 Objective Loss 0.398895 LR 0.001000 Time 0.019983 -2023-02-13 17:26:50,451 - Epoch: [18][ 690/ 1207] Overall Loss 0.398628 Objective Loss 0.398628 LR 0.001000 Time 0.019966 -2023-02-13 17:26:50,641 - Epoch: [18][ 700/ 1207] Overall Loss 0.398374 Objective Loss 0.398374 LR 0.001000 Time 0.019951 -2023-02-13 17:26:50,831 - Epoch: [18][ 710/ 1207] Overall Loss 0.397947 Objective Loss 0.397947 LR 0.001000 Time 0.019937 -2023-02-13 17:26:51,021 - Epoch: [18][ 720/ 1207] Overall Loss 0.398229 Objective Loss 0.398229 LR 0.001000 Time 0.019924 -2023-02-13 17:26:51,211 - Epoch: [18][ 730/ 1207] Overall Loss 0.398491 Objective Loss 0.398491 LR 0.001000 Time 0.019910 -2023-02-13 17:26:51,400 - Epoch: [18][ 740/ 1207] Overall Loss 0.398799 Objective Loss 0.398799 LR 0.001000 Time 0.019896 -2023-02-13 17:26:51,589 - Epoch: [18][ 750/ 1207] Overall Loss 0.398392 Objective Loss 0.398392 LR 0.001000 Time 0.019883 -2023-02-13 17:26:51,779 - Epoch: [18][ 760/ 1207] Overall Loss 0.398620 Objective Loss 0.398620 LR 0.001000 Time 0.019871 -2023-02-13 17:26:51,969 - Epoch: [18][ 770/ 1207] Overall Loss 0.398775 Objective Loss 0.398775 LR 0.001000 Time 0.019859 -2023-02-13 17:26:52,159 - Epoch: [18][ 780/ 1207] Overall Loss 0.398479 Objective Loss 0.398479 LR 0.001000 Time 0.019848 -2023-02-13 17:26:52,348 - Epoch: [18][ 790/ 1207] Overall Loss 0.398466 Objective Loss 0.398466 LR 0.001000 Time 0.019835 -2023-02-13 17:26:52,538 - Epoch: [18][ 800/ 1207] Overall Loss 0.397927 Objective Loss 0.397927 LR 0.001000 Time 0.019824 -2023-02-13 17:26:52,726 - Epoch: [18][ 810/ 1207] Overall Loss 0.397479 Objective Loss 0.397479 LR 0.001000 Time 0.019811 -2023-02-13 17:26:52,915 - Epoch: [18][ 820/ 1207] Overall Loss 0.397466 Objective Loss 0.397466 LR 0.001000 Time 0.019800 -2023-02-13 17:26:53,105 - Epoch: [18][ 830/ 1207] Overall Loss 0.396864 Objective Loss 0.396864 LR 0.001000 Time 0.019789 -2023-02-13 17:26:53,294 - Epoch: [18][ 840/ 1207] Overall Loss 0.396520 Objective Loss 0.396520 LR 0.001000 Time 0.019778 -2023-02-13 17:26:53,483 - Epoch: [18][ 850/ 1207] Overall Loss 0.397029 Objective Loss 0.397029 LR 0.001000 Time 0.019767 -2023-02-13 17:26:53,673 - Epoch: [18][ 860/ 1207] Overall Loss 0.396734 Objective Loss 0.396734 LR 0.001000 Time 0.019757 -2023-02-13 17:26:53,861 - Epoch: [18][ 870/ 1207] Overall Loss 0.396983 Objective Loss 0.396983 LR 0.001000 Time 0.019747 -2023-02-13 17:26:54,051 - Epoch: [18][ 880/ 1207] Overall Loss 0.397326 Objective Loss 0.397326 LR 0.001000 Time 0.019737 -2023-02-13 17:26:54,239 - Epoch: [18][ 890/ 1207] Overall Loss 0.397772 Objective Loss 0.397772 LR 0.001000 Time 0.019727 -2023-02-13 17:26:54,428 - Epoch: [18][ 900/ 1207] Overall Loss 0.398164 Objective Loss 0.398164 LR 0.001000 Time 0.019717 -2023-02-13 17:26:54,618 - Epoch: [18][ 910/ 1207] Overall Loss 0.398162 Objective Loss 0.398162 LR 0.001000 Time 0.019709 -2023-02-13 17:26:54,807 - Epoch: [18][ 920/ 1207] Overall Loss 0.398100 Objective Loss 0.398100 LR 0.001000 Time 0.019700 -2023-02-13 17:26:54,996 - Epoch: [18][ 930/ 1207] Overall Loss 0.398206 Objective Loss 0.398206 LR 0.001000 Time 0.019691 -2023-02-13 17:26:55,186 - Epoch: [18][ 940/ 1207] Overall Loss 0.398606 Objective Loss 0.398606 LR 0.001000 Time 0.019683 -2023-02-13 17:26:55,374 - Epoch: [18][ 950/ 1207] Overall Loss 0.398780 Objective Loss 0.398780 LR 0.001000 Time 0.019673 -2023-02-13 17:26:55,564 - Epoch: [18][ 960/ 1207] Overall Loss 0.398882 Objective Loss 0.398882 LR 0.001000 Time 0.019665 -2023-02-13 17:26:55,754 - Epoch: [18][ 970/ 1207] Overall Loss 0.399694 Objective Loss 0.399694 LR 0.001000 Time 0.019658 -2023-02-13 17:26:55,943 - Epoch: [18][ 980/ 1207] Overall Loss 0.399714 Objective Loss 0.399714 LR 0.001000 Time 0.019651 -2023-02-13 17:26:56,132 - Epoch: [18][ 990/ 1207] Overall Loss 0.399726 Objective Loss 0.399726 LR 0.001000 Time 0.019643 -2023-02-13 17:26:56,322 - Epoch: [18][ 1000/ 1207] Overall Loss 0.399757 Objective Loss 0.399757 LR 0.001000 Time 0.019635 -2023-02-13 17:26:56,511 - Epoch: [18][ 1010/ 1207] Overall Loss 0.400256 Objective Loss 0.400256 LR 0.001000 Time 0.019628 -2023-02-13 17:26:56,701 - Epoch: [18][ 1020/ 1207] Overall Loss 0.400095 Objective Loss 0.400095 LR 0.001000 Time 0.019621 -2023-02-13 17:26:56,891 - Epoch: [18][ 1030/ 1207] Overall Loss 0.400598 Objective Loss 0.400598 LR 0.001000 Time 0.019615 -2023-02-13 17:26:57,081 - Epoch: [18][ 1040/ 1207] Overall Loss 0.400456 Objective Loss 0.400456 LR 0.001000 Time 0.019609 -2023-02-13 17:26:57,270 - Epoch: [18][ 1050/ 1207] Overall Loss 0.400323 Objective Loss 0.400323 LR 0.001000 Time 0.019602 -2023-02-13 17:26:57,459 - Epoch: [18][ 1060/ 1207] Overall Loss 0.400149 Objective Loss 0.400149 LR 0.001000 Time 0.019595 -2023-02-13 17:26:57,649 - Epoch: [18][ 1070/ 1207] Overall Loss 0.400011 Objective Loss 0.400011 LR 0.001000 Time 0.019589 -2023-02-13 17:26:57,838 - Epoch: [18][ 1080/ 1207] Overall Loss 0.400076 Objective Loss 0.400076 LR 0.001000 Time 0.019582 -2023-02-13 17:26:58,027 - Epoch: [18][ 1090/ 1207] Overall Loss 0.399826 Objective Loss 0.399826 LR 0.001000 Time 0.019576 -2023-02-13 17:26:58,217 - Epoch: [18][ 1100/ 1207] Overall Loss 0.399733 Objective Loss 0.399733 LR 0.001000 Time 0.019570 -2023-02-13 17:26:58,406 - Epoch: [18][ 1110/ 1207] Overall Loss 0.399814 Objective Loss 0.399814 LR 0.001000 Time 0.019563 -2023-02-13 17:26:58,595 - Epoch: [18][ 1120/ 1207] Overall Loss 0.400135 Objective Loss 0.400135 LR 0.001000 Time 0.019558 -2023-02-13 17:26:58,785 - Epoch: [18][ 1130/ 1207] Overall Loss 0.400146 Objective Loss 0.400146 LR 0.001000 Time 0.019552 -2023-02-13 17:26:58,975 - Epoch: [18][ 1140/ 1207] Overall Loss 0.400509 Objective Loss 0.400509 LR 0.001000 Time 0.019546 -2023-02-13 17:26:59,164 - Epoch: [18][ 1150/ 1207] Overall Loss 0.400528 Objective Loss 0.400528 LR 0.001000 Time 0.019541 -2023-02-13 17:26:59,353 - Epoch: [18][ 1160/ 1207] Overall Loss 0.400322 Objective Loss 0.400322 LR 0.001000 Time 0.019535 -2023-02-13 17:26:59,543 - Epoch: [18][ 1170/ 1207] Overall Loss 0.400023 Objective Loss 0.400023 LR 0.001000 Time 0.019530 -2023-02-13 17:26:59,731 - Epoch: [18][ 1180/ 1207] Overall Loss 0.399878 Objective Loss 0.399878 LR 0.001000 Time 0.019524 -2023-02-13 17:26:59,921 - Epoch: [18][ 1190/ 1207] Overall Loss 0.400083 Objective Loss 0.400083 LR 0.001000 Time 0.019519 -2023-02-13 17:27:00,167 - Epoch: [18][ 1200/ 1207] Overall Loss 0.399848 Objective Loss 0.399848 LR 0.001000 Time 0.019561 -2023-02-13 17:27:00,282 - Epoch: [18][ 1207/ 1207] Overall Loss 0.399920 Objective Loss 0.399920 Top1 81.402439 Top5 96.341463 LR 0.001000 Time 0.019543 -2023-02-13 17:27:00,362 - --- validate (epoch=18)----------- -2023-02-13 17:27:00,362 - 34311 samples (256 per mini-batch) -2023-02-13 17:27:00,766 - Epoch: [18][ 10/ 135] Loss 0.408805 Top1 79.257812 Top5 97.109375 -2023-02-13 17:27:00,890 - Epoch: [18][ 20/ 135] Loss 0.418226 Top1 78.320312 Top5 96.816406 -2023-02-13 17:27:01,016 - Epoch: [18][ 30/ 135] Loss 0.407948 Top1 78.710938 Top5 96.940104 -2023-02-13 17:27:01,137 - Epoch: [18][ 40/ 135] Loss 0.411845 Top1 79.082031 Top5 96.914062 -2023-02-13 17:27:01,258 - Epoch: [18][ 50/ 135] Loss 0.416409 Top1 79.187500 Top5 96.929688 -2023-02-13 17:27:01,380 - Epoch: [18][ 60/ 135] Loss 0.409321 Top1 79.290365 Top5 96.927083 -2023-02-13 17:27:01,504 - Epoch: [18][ 70/ 135] Loss 0.408641 Top1 79.414062 Top5 96.908482 -2023-02-13 17:27:01,626 - Epoch: [18][ 80/ 135] Loss 0.410634 Top1 79.301758 Top5 96.826172 -2023-02-13 17:27:01,754 - Epoch: [18][ 90/ 135] Loss 0.411004 Top1 79.414062 Top5 96.796875 -2023-02-13 17:27:01,881 - Epoch: [18][ 100/ 135] Loss 0.412047 Top1 79.433594 Top5 96.812500 -2023-02-13 17:27:02,010 - Epoch: [18][ 110/ 135] Loss 0.412116 Top1 79.499290 Top5 96.839489 -2023-02-13 17:27:02,138 - Epoch: [18][ 120/ 135] Loss 0.411370 Top1 79.495443 Top5 96.822917 -2023-02-13 17:27:02,268 - Epoch: [18][ 130/ 135] Loss 0.409619 Top1 79.546274 Top5 96.814904 -2023-02-13 17:27:02,315 - Epoch: [18][ 135/ 135] Loss 0.415322 Top1 79.545918 Top5 96.814433 -2023-02-13 17:27:02,382 - ==> Top1: 79.546 Top5: 96.814 Loss: 0.415 - -2023-02-13 17:27:02,382 - ==> Confusion: -[[ 814 5 4 0 10 2 0 1 6 85 0 8 1 2 11 3 2 0 4 1 8] - [ 3 938 3 1 12 30 2 14 7 3 2 2 0 1 1 1 5 0 2 1 5] - [ 15 6 917 13 6 5 13 26 0 2 1 5 2 4 3 13 3 2 10 3 9] - [ 9 4 36 839 2 3 0 1 3 1 12 1 8 1 42 3 3 8 28 3 9] - [ 21 13 0 2 980 5 1 1 0 10 0 3 1 5 9 3 4 0 2 2 4] - [ 4 39 3 7 10 927 3 16 3 6 1 12 4 17 1 4 4 1 3 2 3] - [ 8 4 25 0 0 6 1012 7 1 0 3 3 1 1 1 10 2 2 1 9 3] - [ 3 18 11 0 1 34 5 906 2 1 2 6 3 1 1 0 0 2 18 8 2] - [ 29 6 0 1 2 0 0 0 868 53 5 2 0 5 26 0 1 1 7 0 3] - [ 81 2 1 0 4 4 0 2 47 846 0 0 0 9 8 1 0 1 1 1 4] - [ 3 8 8 5 1 1 3 6 28 2 938 2 0 13 5 0 2 1 19 0 6] - [ 6 4 1 0 4 21 0 3 4 2 0 917 7 5 1 5 5 7 4 9 0] - [ 3 5 2 5 3 3 0 0 8 0 0 93 761 2 6 12 7 38 3 2 6] - [ 11 3 2 0 9 14 0 1 25 30 5 9 1 885 9 4 3 1 0 4 8] - [ 12 3 0 8 10 2 0 0 24 8 1 4 1 0 988 1 4 5 9 1 11] - [ 6 6 6 1 10 3 2 1 1 2 0 9 2 3 0 957 7 19 1 4 6] - [ 5 10 1 0 15 3 0 0 1 1 0 6 0 2 1 15 987 4 1 4 5] - [ 8 1 0 2 2 0 0 2 5 1 0 20 10 3 3 17 0 971 0 0 6] - [ 5 14 7 5 2 1 1 39 5 1 3 1 5 1 24 2 2 1 963 0 4] - [ 1 5 3 0 0 11 6 25 0 0 1 27 4 6 0 8 3 2 2 1036 8] - [ 303 362 238 81 325 214 80 239 151 164 141 266 257 348 241 194 359 126 213 289 8843]] - -2023-02-13 17:27:02,384 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:27:02,384 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:27:02,390 - - -2023-02-13 17:27:02,390 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:27:03,366 - Epoch: [19][ 10/ 1207] Overall Loss 0.402493 Objective Loss 0.402493 LR 0.001000 Time 0.097563 -2023-02-13 17:27:03,571 - Epoch: [19][ 20/ 1207] Overall Loss 0.385056 Objective Loss 0.385056 LR 0.001000 Time 0.059021 -2023-02-13 17:27:03,765 - Epoch: [19][ 30/ 1207] Overall Loss 0.371650 Objective Loss 0.371650 LR 0.001000 Time 0.045780 -2023-02-13 17:27:03,960 - Epoch: [19][ 40/ 1207] Overall Loss 0.365165 Objective Loss 0.365165 LR 0.001000 Time 0.039198 -2023-02-13 17:27:04,155 - Epoch: [19][ 50/ 1207] Overall Loss 0.365663 Objective Loss 0.365663 LR 0.001000 Time 0.035252 -2023-02-13 17:27:04,350 - Epoch: [19][ 60/ 1207] Overall Loss 0.372292 Objective Loss 0.372292 LR 0.001000 Time 0.032630 -2023-02-13 17:27:04,545 - Epoch: [19][ 70/ 1207] Overall Loss 0.371279 Objective Loss 0.371279 LR 0.001000 Time 0.030748 -2023-02-13 17:27:04,741 - Epoch: [19][ 80/ 1207] Overall Loss 0.369929 Objective Loss 0.369929 LR 0.001000 Time 0.029352 -2023-02-13 17:27:04,936 - Epoch: [19][ 90/ 1207] Overall Loss 0.370679 Objective Loss 0.370679 LR 0.001000 Time 0.028251 -2023-02-13 17:27:05,132 - Epoch: [19][ 100/ 1207] Overall Loss 0.368957 Objective Loss 0.368957 LR 0.001000 Time 0.027379 -2023-02-13 17:27:05,326 - Epoch: [19][ 110/ 1207] Overall Loss 0.370708 Objective Loss 0.370708 LR 0.001000 Time 0.026656 -2023-02-13 17:27:05,522 - Epoch: [19][ 120/ 1207] Overall Loss 0.373621 Objective Loss 0.373621 LR 0.001000 Time 0.026065 -2023-02-13 17:27:05,718 - Epoch: [19][ 130/ 1207] Overall Loss 0.375420 Objective Loss 0.375420 LR 0.001000 Time 0.025562 -2023-02-13 17:27:05,915 - Epoch: [19][ 140/ 1207] Overall Loss 0.374855 Objective Loss 0.374855 LR 0.001000 Time 0.025141 -2023-02-13 17:27:06,110 - Epoch: [19][ 150/ 1207] Overall Loss 0.373084 Objective Loss 0.373084 LR 0.001000 Time 0.024764 -2023-02-13 17:27:06,306 - Epoch: [19][ 160/ 1207] Overall Loss 0.373358 Objective Loss 0.373358 LR 0.001000 Time 0.024436 -2023-02-13 17:27:06,501 - Epoch: [19][ 170/ 1207] Overall Loss 0.376044 Objective Loss 0.376044 LR 0.001000 Time 0.024143 -2023-02-13 17:27:06,698 - Epoch: [19][ 180/ 1207] Overall Loss 0.377166 Objective Loss 0.377166 LR 0.001000 Time 0.023895 -2023-02-13 17:27:06,894 - Epoch: [19][ 190/ 1207] Overall Loss 0.375563 Objective Loss 0.375563 LR 0.001000 Time 0.023666 -2023-02-13 17:27:07,090 - Epoch: [19][ 200/ 1207] Overall Loss 0.376463 Objective Loss 0.376463 LR 0.001000 Time 0.023464 -2023-02-13 17:27:07,285 - Epoch: [19][ 210/ 1207] Overall Loss 0.376556 Objective Loss 0.376556 LR 0.001000 Time 0.023272 -2023-02-13 17:27:07,480 - Epoch: [19][ 220/ 1207] Overall Loss 0.374752 Objective Loss 0.374752 LR 0.001000 Time 0.023099 -2023-02-13 17:27:07,676 - Epoch: [19][ 230/ 1207] Overall Loss 0.376715 Objective Loss 0.376715 LR 0.001000 Time 0.022946 -2023-02-13 17:27:07,871 - Epoch: [19][ 240/ 1207] Overall Loss 0.378364 Objective Loss 0.378364 LR 0.001000 Time 0.022802 -2023-02-13 17:27:08,067 - Epoch: [19][ 250/ 1207] Overall Loss 0.378010 Objective Loss 0.378010 LR 0.001000 Time 0.022671 -2023-02-13 17:27:08,263 - Epoch: [19][ 260/ 1207] Overall Loss 0.380109 Objective Loss 0.380109 LR 0.001000 Time 0.022551 -2023-02-13 17:27:08,458 - Epoch: [19][ 270/ 1207] Overall Loss 0.381945 Objective Loss 0.381945 LR 0.001000 Time 0.022436 -2023-02-13 17:27:08,654 - Epoch: [19][ 280/ 1207] Overall Loss 0.384030 Objective Loss 0.384030 LR 0.001000 Time 0.022335 -2023-02-13 17:27:08,849 - Epoch: [19][ 290/ 1207] Overall Loss 0.382713 Objective Loss 0.382713 LR 0.001000 Time 0.022236 -2023-02-13 17:27:09,046 - Epoch: [19][ 300/ 1207] Overall Loss 0.382598 Objective Loss 0.382598 LR 0.001000 Time 0.022149 -2023-02-13 17:27:09,241 - Epoch: [19][ 310/ 1207] Overall Loss 0.382697 Objective Loss 0.382697 LR 0.001000 Time 0.022063 -2023-02-13 17:27:09,437 - Epoch: [19][ 320/ 1207] Overall Loss 0.381328 Objective Loss 0.381328 LR 0.001000 Time 0.021985 -2023-02-13 17:27:09,634 - Epoch: [19][ 330/ 1207] Overall Loss 0.381733 Objective Loss 0.381733 LR 0.001000 Time 0.021913 -2023-02-13 17:27:09,829 - Epoch: [19][ 340/ 1207] Overall Loss 0.381839 Objective Loss 0.381839 LR 0.001000 Time 0.021842 -2023-02-13 17:27:10,025 - Epoch: [19][ 350/ 1207] Overall Loss 0.381471 Objective Loss 0.381471 LR 0.001000 Time 0.021776 -2023-02-13 17:27:10,221 - Epoch: [19][ 360/ 1207] Overall Loss 0.381447 Objective Loss 0.381447 LR 0.001000 Time 0.021715 -2023-02-13 17:27:10,415 - Epoch: [19][ 370/ 1207] Overall Loss 0.382010 Objective Loss 0.382010 LR 0.001000 Time 0.021653 -2023-02-13 17:27:10,613 - Epoch: [19][ 380/ 1207] Overall Loss 0.381690 Objective Loss 0.381690 LR 0.001000 Time 0.021601 -2023-02-13 17:27:10,809 - Epoch: [19][ 390/ 1207] Overall Loss 0.382370 Objective Loss 0.382370 LR 0.001000 Time 0.021550 -2023-02-13 17:27:11,006 - Epoch: [19][ 400/ 1207] Overall Loss 0.382177 Objective Loss 0.382177 LR 0.001000 Time 0.021502 -2023-02-13 17:27:11,200 - Epoch: [19][ 410/ 1207] Overall Loss 0.381850 Objective Loss 0.381850 LR 0.001000 Time 0.021451 -2023-02-13 17:27:11,396 - Epoch: [19][ 420/ 1207] Overall Loss 0.382414 Objective Loss 0.382414 LR 0.001000 Time 0.021406 -2023-02-13 17:27:11,592 - Epoch: [19][ 430/ 1207] Overall Loss 0.382630 Objective Loss 0.382630 LR 0.001000 Time 0.021362 -2023-02-13 17:27:11,788 - Epoch: [19][ 440/ 1207] Overall Loss 0.382831 Objective Loss 0.382831 LR 0.001000 Time 0.021321 -2023-02-13 17:27:11,984 - Epoch: [19][ 450/ 1207] Overall Loss 0.383499 Objective Loss 0.383499 LR 0.001000 Time 0.021282 -2023-02-13 17:27:12,180 - Epoch: [19][ 460/ 1207] Overall Loss 0.384212 Objective Loss 0.384212 LR 0.001000 Time 0.021245 -2023-02-13 17:27:12,375 - Epoch: [19][ 470/ 1207] Overall Loss 0.383940 Objective Loss 0.383940 LR 0.001000 Time 0.021207 -2023-02-13 17:27:12,571 - Epoch: [19][ 480/ 1207] Overall Loss 0.384711 Objective Loss 0.384711 LR 0.001000 Time 0.021173 -2023-02-13 17:27:12,767 - Epoch: [19][ 490/ 1207] Overall Loss 0.384497 Objective Loss 0.384497 LR 0.001000 Time 0.021140 -2023-02-13 17:27:12,963 - Epoch: [19][ 500/ 1207] Overall Loss 0.385015 Objective Loss 0.385015 LR 0.001000 Time 0.021108 -2023-02-13 17:27:13,159 - Epoch: [19][ 510/ 1207] Overall Loss 0.384560 Objective Loss 0.384560 LR 0.001000 Time 0.021077 -2023-02-13 17:27:13,355 - Epoch: [19][ 520/ 1207] Overall Loss 0.384702 Objective Loss 0.384702 LR 0.001000 Time 0.021049 -2023-02-13 17:27:13,550 - Epoch: [19][ 530/ 1207] Overall Loss 0.384736 Objective Loss 0.384736 LR 0.001000 Time 0.021018 -2023-02-13 17:27:13,746 - Epoch: [19][ 540/ 1207] Overall Loss 0.384894 Objective Loss 0.384894 LR 0.001000 Time 0.020993 -2023-02-13 17:27:13,942 - Epoch: [19][ 550/ 1207] Overall Loss 0.385303 Objective Loss 0.385303 LR 0.001000 Time 0.020966 -2023-02-13 17:27:14,139 - Epoch: [19][ 560/ 1207] Overall Loss 0.385693 Objective Loss 0.385693 LR 0.001000 Time 0.020942 -2023-02-13 17:27:14,334 - Epoch: [19][ 570/ 1207] Overall Loss 0.386145 Objective Loss 0.386145 LR 0.001000 Time 0.020916 -2023-02-13 17:27:14,530 - Epoch: [19][ 580/ 1207] Overall Loss 0.386174 Objective Loss 0.386174 LR 0.001000 Time 0.020893 -2023-02-13 17:27:14,726 - Epoch: [19][ 590/ 1207] Overall Loss 0.386170 Objective Loss 0.386170 LR 0.001000 Time 0.020871 -2023-02-13 17:27:14,922 - Epoch: [19][ 600/ 1207] Overall Loss 0.386472 Objective Loss 0.386472 LR 0.001000 Time 0.020849 -2023-02-13 17:27:15,117 - Epoch: [19][ 610/ 1207] Overall Loss 0.386314 Objective Loss 0.386314 LR 0.001000 Time 0.020826 -2023-02-13 17:27:15,314 - Epoch: [19][ 620/ 1207] Overall Loss 0.386909 Objective Loss 0.386909 LR 0.001000 Time 0.020808 -2023-02-13 17:27:15,508 - Epoch: [19][ 630/ 1207] Overall Loss 0.386737 Objective Loss 0.386737 LR 0.001000 Time 0.020785 -2023-02-13 17:27:15,707 - Epoch: [19][ 640/ 1207] Overall Loss 0.387136 Objective Loss 0.387136 LR 0.001000 Time 0.020770 -2023-02-13 17:27:15,902 - Epoch: [19][ 650/ 1207] Overall Loss 0.387616 Objective Loss 0.387616 LR 0.001000 Time 0.020750 -2023-02-13 17:27:16,099 - Epoch: [19][ 660/ 1207] Overall Loss 0.387914 Objective Loss 0.387914 LR 0.001000 Time 0.020734 -2023-02-13 17:27:16,295 - Epoch: [19][ 670/ 1207] Overall Loss 0.388175 Objective Loss 0.388175 LR 0.001000 Time 0.020716 -2023-02-13 17:27:16,491 - Epoch: [19][ 680/ 1207] Overall Loss 0.387927 Objective Loss 0.387927 LR 0.001000 Time 0.020700 -2023-02-13 17:27:16,686 - Epoch: [19][ 690/ 1207] Overall Loss 0.387951 Objective Loss 0.387951 LR 0.001000 Time 0.020682 -2023-02-13 17:27:16,884 - Epoch: [19][ 700/ 1207] Overall Loss 0.387955 Objective Loss 0.387955 LR 0.001000 Time 0.020668 -2023-02-13 17:27:17,079 - Epoch: [19][ 710/ 1207] Overall Loss 0.387919 Objective Loss 0.387919 LR 0.001000 Time 0.020651 -2023-02-13 17:27:17,276 - Epoch: [19][ 720/ 1207] Overall Loss 0.388437 Objective Loss 0.388437 LR 0.001000 Time 0.020637 -2023-02-13 17:27:17,471 - Epoch: [19][ 730/ 1207] Overall Loss 0.388334 Objective Loss 0.388334 LR 0.001000 Time 0.020622 -2023-02-13 17:27:17,669 - Epoch: [19][ 740/ 1207] Overall Loss 0.388202 Objective Loss 0.388202 LR 0.001000 Time 0.020610 -2023-02-13 17:27:17,864 - Epoch: [19][ 750/ 1207] Overall Loss 0.388489 Objective Loss 0.388489 LR 0.001000 Time 0.020595 -2023-02-13 17:27:18,061 - Epoch: [19][ 760/ 1207] Overall Loss 0.388647 Objective Loss 0.388647 LR 0.001000 Time 0.020582 -2023-02-13 17:27:18,255 - Epoch: [19][ 770/ 1207] Overall Loss 0.388762 Objective Loss 0.388762 LR 0.001000 Time 0.020566 -2023-02-13 17:27:18,452 - Epoch: [19][ 780/ 1207] Overall Loss 0.389404 Objective Loss 0.389404 LR 0.001000 Time 0.020555 -2023-02-13 17:27:18,647 - Epoch: [19][ 790/ 1207] Overall Loss 0.389322 Objective Loss 0.389322 LR 0.001000 Time 0.020541 -2023-02-13 17:27:18,843 - Epoch: [19][ 800/ 1207] Overall Loss 0.389415 Objective Loss 0.389415 LR 0.001000 Time 0.020529 -2023-02-13 17:27:19,037 - Epoch: [19][ 810/ 1207] Overall Loss 0.389647 Objective Loss 0.389647 LR 0.001000 Time 0.020514 -2023-02-13 17:27:19,232 - Epoch: [19][ 820/ 1207] Overall Loss 0.389605 Objective Loss 0.389605 LR 0.001000 Time 0.020501 -2023-02-13 17:27:19,426 - Epoch: [19][ 830/ 1207] Overall Loss 0.390108 Objective Loss 0.390108 LR 0.001000 Time 0.020487 -2023-02-13 17:27:19,621 - Epoch: [19][ 840/ 1207] Overall Loss 0.389737 Objective Loss 0.389737 LR 0.001000 Time 0.020476 -2023-02-13 17:27:19,815 - Epoch: [19][ 850/ 1207] Overall Loss 0.389497 Objective Loss 0.389497 LR 0.001000 Time 0.020463 -2023-02-13 17:27:20,011 - Epoch: [19][ 860/ 1207] Overall Loss 0.389345 Objective Loss 0.389345 LR 0.001000 Time 0.020452 -2023-02-13 17:27:20,205 - Epoch: [19][ 870/ 1207] Overall Loss 0.389164 Objective Loss 0.389164 LR 0.001000 Time 0.020439 -2023-02-13 17:27:20,399 - Epoch: [19][ 880/ 1207] Overall Loss 0.389559 Objective Loss 0.389559 LR 0.001000 Time 0.020428 -2023-02-13 17:27:20,594 - Epoch: [19][ 890/ 1207] Overall Loss 0.389699 Objective Loss 0.389699 LR 0.001000 Time 0.020416 -2023-02-13 17:27:20,792 - Epoch: [19][ 900/ 1207] Overall Loss 0.389599 Objective Loss 0.389599 LR 0.001000 Time 0.020409 -2023-02-13 17:27:20,988 - Epoch: [19][ 910/ 1207] Overall Loss 0.389827 Objective Loss 0.389827 LR 0.001000 Time 0.020400 -2023-02-13 17:27:21,185 - Epoch: [19][ 920/ 1207] Overall Loss 0.390059 Objective Loss 0.390059 LR 0.001000 Time 0.020392 -2023-02-13 17:27:21,380 - Epoch: [19][ 930/ 1207] Overall Loss 0.390299 Objective Loss 0.390299 LR 0.001000 Time 0.020382 -2023-02-13 17:27:21,578 - Epoch: [19][ 940/ 1207] Overall Loss 0.390266 Objective Loss 0.390266 LR 0.001000 Time 0.020375 -2023-02-13 17:27:21,773 - Epoch: [19][ 950/ 1207] Overall Loss 0.390025 Objective Loss 0.390025 LR 0.001000 Time 0.020365 -2023-02-13 17:27:21,969 - Epoch: [19][ 960/ 1207] Overall Loss 0.389941 Objective Loss 0.389941 LR 0.001000 Time 0.020357 -2023-02-13 17:27:22,164 - Epoch: [19][ 970/ 1207] Overall Loss 0.390042 Objective Loss 0.390042 LR 0.001000 Time 0.020348 -2023-02-13 17:27:22,360 - Epoch: [19][ 980/ 1207] Overall Loss 0.390040 Objective Loss 0.390040 LR 0.001000 Time 0.020339 -2023-02-13 17:27:22,554 - Epoch: [19][ 990/ 1207] Overall Loss 0.389945 Objective Loss 0.389945 LR 0.001000 Time 0.020330 -2023-02-13 17:27:22,751 - Epoch: [19][ 1000/ 1207] Overall Loss 0.389831 Objective Loss 0.389831 LR 0.001000 Time 0.020323 -2023-02-13 17:27:22,946 - Epoch: [19][ 1010/ 1207] Overall Loss 0.390166 Objective Loss 0.390166 LR 0.001000 Time 0.020315 -2023-02-13 17:27:23,142 - Epoch: [19][ 1020/ 1207] Overall Loss 0.389996 Objective Loss 0.389996 LR 0.001000 Time 0.020307 -2023-02-13 17:27:23,336 - Epoch: [19][ 1030/ 1207] Overall Loss 0.390085 Objective Loss 0.390085 LR 0.001000 Time 0.020299 -2023-02-13 17:27:23,533 - Epoch: [19][ 1040/ 1207] Overall Loss 0.389814 Objective Loss 0.389814 LR 0.001000 Time 0.020292 -2023-02-13 17:27:23,729 - Epoch: [19][ 1050/ 1207] Overall Loss 0.390041 Objective Loss 0.390041 LR 0.001000 Time 0.020285 -2023-02-13 17:27:23,925 - Epoch: [19][ 1060/ 1207] Overall Loss 0.390375 Objective Loss 0.390375 LR 0.001000 Time 0.020279 -2023-02-13 17:27:24,121 - Epoch: [19][ 1070/ 1207] Overall Loss 0.390261 Objective Loss 0.390261 LR 0.001000 Time 0.020271 -2023-02-13 17:27:24,318 - Epoch: [19][ 1080/ 1207] Overall Loss 0.390378 Objective Loss 0.390378 LR 0.001000 Time 0.020266 -2023-02-13 17:27:24,513 - Epoch: [19][ 1090/ 1207] Overall Loss 0.390444 Objective Loss 0.390444 LR 0.001000 Time 0.020258 -2023-02-13 17:27:24,710 - Epoch: [19][ 1100/ 1207] Overall Loss 0.390327 Objective Loss 0.390327 LR 0.001000 Time 0.020253 -2023-02-13 17:27:24,904 - Epoch: [19][ 1110/ 1207] Overall Loss 0.389971 Objective Loss 0.389971 LR 0.001000 Time 0.020245 -2023-02-13 17:27:25,102 - Epoch: [19][ 1120/ 1207] Overall Loss 0.389855 Objective Loss 0.389855 LR 0.001000 Time 0.020241 -2023-02-13 17:27:25,296 - Epoch: [19][ 1130/ 1207] Overall Loss 0.389845 Objective Loss 0.389845 LR 0.001000 Time 0.020233 -2023-02-13 17:27:25,494 - Epoch: [19][ 1140/ 1207] Overall Loss 0.389771 Objective Loss 0.389771 LR 0.001000 Time 0.020229 -2023-02-13 17:27:25,689 - Epoch: [19][ 1150/ 1207] Overall Loss 0.390013 Objective Loss 0.390013 LR 0.001000 Time 0.020222 -2023-02-13 17:27:25,887 - Epoch: [19][ 1160/ 1207] Overall Loss 0.390206 Objective Loss 0.390206 LR 0.001000 Time 0.020219 -2023-02-13 17:27:26,082 - Epoch: [19][ 1170/ 1207] Overall Loss 0.390397 Objective Loss 0.390397 LR 0.001000 Time 0.020212 -2023-02-13 17:27:26,279 - Epoch: [19][ 1180/ 1207] Overall Loss 0.390262 Objective Loss 0.390262 LR 0.001000 Time 0.020207 -2023-02-13 17:27:26,474 - Epoch: [19][ 1190/ 1207] Overall Loss 0.390259 Objective Loss 0.390259 LR 0.001000 Time 0.020201 -2023-02-13 17:27:26,720 - Epoch: [19][ 1200/ 1207] Overall Loss 0.390450 Objective Loss 0.390450 LR 0.001000 Time 0.020237 -2023-02-13 17:27:26,836 - Epoch: [19][ 1207/ 1207] Overall Loss 0.390494 Objective Loss 0.390494 Top1 80.182927 Top5 97.256098 LR 0.001000 Time 0.020216 -2023-02-13 17:27:26,907 - --- validate (epoch=19)----------- -2023-02-13 17:27:26,907 - 34311 samples (256 per mini-batch) -2023-02-13 17:27:27,306 - Epoch: [19][ 10/ 135] Loss 0.373128 Top1 78.515625 Top5 96.953125 -2023-02-13 17:27:27,434 - Epoch: [19][ 20/ 135] Loss 0.400718 Top1 78.417969 Top5 96.640625 -2023-02-13 17:27:27,563 - Epoch: [19][ 30/ 135] Loss 0.397462 Top1 78.346354 Top5 96.549479 -2023-02-13 17:27:27,692 - Epoch: [19][ 40/ 135] Loss 0.404220 Top1 78.564453 Top5 96.611328 -2023-02-13 17:27:27,821 - Epoch: [19][ 50/ 135] Loss 0.399259 Top1 78.640625 Top5 96.617188 -2023-02-13 17:27:27,946 - Epoch: [19][ 60/ 135] Loss 0.397457 Top1 78.678385 Top5 96.699219 -2023-02-13 17:27:28,072 - Epoch: [19][ 70/ 135] Loss 0.398092 Top1 78.643973 Top5 96.746652 -2023-02-13 17:27:28,199 - Epoch: [19][ 80/ 135] Loss 0.401501 Top1 78.544922 Top5 96.743164 -2023-02-13 17:27:28,325 - Epoch: [19][ 90/ 135] Loss 0.401616 Top1 78.428819 Top5 96.744792 -2023-02-13 17:27:28,451 - Epoch: [19][ 100/ 135] Loss 0.401483 Top1 78.289062 Top5 96.726562 -2023-02-13 17:27:28,578 - Epoch: [19][ 110/ 135] Loss 0.400853 Top1 78.263494 Top5 96.690341 -2023-02-13 17:27:28,706 - Epoch: [19][ 120/ 135] Loss 0.401924 Top1 78.330078 Top5 96.712240 -2023-02-13 17:27:28,836 - Epoch: [19][ 130/ 135] Loss 0.402280 Top1 78.335337 Top5 96.694712 -2023-02-13 17:27:28,883 - Epoch: [19][ 135/ 135] Loss 0.399695 Top1 78.342223 Top5 96.697852 -2023-02-13 17:27:28,966 - ==> Top1: 78.342 Top5: 96.698 Loss: 0.400 - -2023-02-13 17:27:28,967 - ==> Confusion: -[[ 817 6 8 3 18 4 0 0 5 63 0 4 2 8 10 3 6 2 1 1 6] - [ 1 903 0 3 12 63 2 12 4 2 9 1 0 1 0 2 5 0 4 1 8] - [ 8 9 922 14 4 4 27 16 1 2 5 1 2 6 1 4 6 4 14 0 8] - [ 3 4 23 886 1 3 2 1 2 2 23 0 3 2 28 2 6 6 14 1 4] - [ 21 8 2 1 973 6 0 1 1 8 0 2 1 8 8 5 12 1 0 3 5] - [ 4 20 0 3 5 956 3 11 3 5 2 15 2 20 2 4 5 1 2 2 5] - [ 3 6 15 3 1 5 1025 3 0 1 9 0 3 2 0 5 7 1 2 5 3] - [ 3 12 13 3 2 41 5 878 0 2 4 4 3 3 0 1 1 0 37 8 4] - [ 19 5 0 0 3 0 0 1 892 26 15 1 1 11 25 2 1 4 2 0 1] - [ 99 4 2 0 4 1 2 2 55 797 1 2 0 27 7 1 1 1 0 0 6] - [ 2 6 6 4 0 3 1 2 21 1 984 1 1 10 1 0 2 0 6 0 0] - [ 5 4 1 0 0 16 3 6 2 3 0 886 20 18 0 7 7 10 2 14 1] - [ 2 1 2 5 1 4 0 0 3 1 1 61 817 5 5 6 3 28 4 4 6] - [ 9 2 2 1 5 9 2 0 13 14 8 6 2 927 7 4 6 1 2 1 3] - [ 13 6 1 15 5 1 0 1 30 3 4 3 2 0 973 2 6 6 14 0 7] - [ 8 4 4 2 6 2 17 2 0 1 0 7 4 9 1 929 24 14 0 7 5] - [ 4 8 0 0 8 6 1 1 3 0 1 1 4 2 3 12 1000 1 0 1 5] - [ 7 3 1 7 0 2 2 0 2 0 1 11 18 4 2 13 1 968 1 2 6] - [ 5 11 3 12 1 8 0 26 5 0 11 2 2 0 15 1 0 1 979 0 4] - [ 1 5 2 1 0 23 11 18 2 1 3 13 3 9 0 6 6 1 1 1036 6] - [ 243 331 295 138 195 340 154 168 141 142 312 150 342 471 235 107 662 112 216 348 8332]] - -2023-02-13 17:27:28,969 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:27:28,969 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:27:28,975 - - -2023-02-13 17:27:28,975 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:27:29,850 - Epoch: [20][ 10/ 1207] Overall Loss 0.357792 Objective Loss 0.357792 LR 0.001000 Time 0.087467 -2023-02-13 17:27:30,050 - Epoch: [20][ 20/ 1207] Overall Loss 0.346528 Objective Loss 0.346528 LR 0.001000 Time 0.053698 -2023-02-13 17:27:30,239 - Epoch: [20][ 30/ 1207] Overall Loss 0.346938 Objective Loss 0.346938 LR 0.001000 Time 0.042096 -2023-02-13 17:27:30,427 - Epoch: [20][ 40/ 1207] Overall Loss 0.355535 Objective Loss 0.355535 LR 0.001000 Time 0.036266 -2023-02-13 17:27:30,616 - Epoch: [20][ 50/ 1207] Overall Loss 0.355425 Objective Loss 0.355425 LR 0.001000 Time 0.032778 -2023-02-13 17:27:30,807 - Epoch: [20][ 60/ 1207] Overall Loss 0.359137 Objective Loss 0.359137 LR 0.001000 Time 0.030493 -2023-02-13 17:27:30,998 - Epoch: [20][ 70/ 1207] Overall Loss 0.364278 Objective Loss 0.364278 LR 0.001000 Time 0.028863 -2023-02-13 17:27:31,188 - Epoch: [20][ 80/ 1207] Overall Loss 0.362673 Objective Loss 0.362673 LR 0.001000 Time 0.027626 -2023-02-13 17:27:31,379 - Epoch: [20][ 90/ 1207] Overall Loss 0.364714 Objective Loss 0.364714 LR 0.001000 Time 0.026677 -2023-02-13 17:27:31,568 - Epoch: [20][ 100/ 1207] Overall Loss 0.366286 Objective Loss 0.366286 LR 0.001000 Time 0.025894 -2023-02-13 17:27:31,758 - Epoch: [20][ 110/ 1207] Overall Loss 0.367089 Objective Loss 0.367089 LR 0.001000 Time 0.025261 -2023-02-13 17:27:31,946 - Epoch: [20][ 120/ 1207] Overall Loss 0.372277 Objective Loss 0.372277 LR 0.001000 Time 0.024721 -2023-02-13 17:27:32,137 - Epoch: [20][ 130/ 1207] Overall Loss 0.378061 Objective Loss 0.378061 LR 0.001000 Time 0.024288 -2023-02-13 17:27:32,327 - Epoch: [20][ 140/ 1207] Overall Loss 0.379023 Objective Loss 0.379023 LR 0.001000 Time 0.023906 -2023-02-13 17:27:32,518 - Epoch: [20][ 150/ 1207] Overall Loss 0.379017 Objective Loss 0.379017 LR 0.001000 Time 0.023586 -2023-02-13 17:27:32,709 - Epoch: [20][ 160/ 1207] Overall Loss 0.378383 Objective Loss 0.378383 LR 0.001000 Time 0.023301 -2023-02-13 17:27:32,900 - Epoch: [20][ 170/ 1207] Overall Loss 0.378324 Objective Loss 0.378324 LR 0.001000 Time 0.023052 -2023-02-13 17:27:33,090 - Epoch: [20][ 180/ 1207] Overall Loss 0.379932 Objective Loss 0.379932 LR 0.001000 Time 0.022827 -2023-02-13 17:27:33,282 - Epoch: [20][ 190/ 1207] Overall Loss 0.381726 Objective Loss 0.381726 LR 0.001000 Time 0.022629 -2023-02-13 17:27:33,471 - Epoch: [20][ 200/ 1207] Overall Loss 0.379764 Objective Loss 0.379764 LR 0.001000 Time 0.022444 -2023-02-13 17:27:33,662 - Epoch: [20][ 210/ 1207] Overall Loss 0.379023 Objective Loss 0.379023 LR 0.001000 Time 0.022284 -2023-02-13 17:27:33,852 - Epoch: [20][ 220/ 1207] Overall Loss 0.384838 Objective Loss 0.384838 LR 0.001000 Time 0.022133 -2023-02-13 17:27:34,044 - Epoch: [20][ 230/ 1207] Overall Loss 0.391272 Objective Loss 0.391272 LR 0.001000 Time 0.022001 -2023-02-13 17:27:34,235 - Epoch: [20][ 240/ 1207] Overall Loss 0.397799 Objective Loss 0.397799 LR 0.001000 Time 0.021882 -2023-02-13 17:27:34,426 - Epoch: [20][ 250/ 1207] Overall Loss 0.402964 Objective Loss 0.402964 LR 0.001000 Time 0.021768 -2023-02-13 17:27:34,617 - Epoch: [20][ 260/ 1207] Overall Loss 0.407381 Objective Loss 0.407381 LR 0.001000 Time 0.021662 -2023-02-13 17:27:34,808 - Epoch: [20][ 270/ 1207] Overall Loss 0.410665 Objective Loss 0.410665 LR 0.001000 Time 0.021568 -2023-02-13 17:27:34,999 - Epoch: [20][ 280/ 1207] Overall Loss 0.413087 Objective Loss 0.413087 LR 0.001000 Time 0.021477 -2023-02-13 17:27:35,190 - Epoch: [20][ 290/ 1207] Overall Loss 0.414630 Objective Loss 0.414630 LR 0.001000 Time 0.021393 -2023-02-13 17:27:35,380 - Epoch: [20][ 300/ 1207] Overall Loss 0.417750 Objective Loss 0.417750 LR 0.001000 Time 0.021312 -2023-02-13 17:27:35,570 - Epoch: [20][ 310/ 1207] Overall Loss 0.420181 Objective Loss 0.420181 LR 0.001000 Time 0.021239 -2023-02-13 17:27:35,762 - Epoch: [20][ 320/ 1207] Overall Loss 0.421311 Objective Loss 0.421311 LR 0.001000 Time 0.021173 -2023-02-13 17:27:35,952 - Epoch: [20][ 330/ 1207] Overall Loss 0.423254 Objective Loss 0.423254 LR 0.001000 Time 0.021106 -2023-02-13 17:27:36,140 - Epoch: [20][ 340/ 1207] Overall Loss 0.424360 Objective Loss 0.424360 LR 0.001000 Time 0.021038 -2023-02-13 17:27:36,330 - Epoch: [20][ 350/ 1207] Overall Loss 0.425469 Objective Loss 0.425469 LR 0.001000 Time 0.020976 -2023-02-13 17:27:36,519 - Epoch: [20][ 360/ 1207] Overall Loss 0.426160 Objective Loss 0.426160 LR 0.001000 Time 0.020918 -2023-02-13 17:27:36,708 - Epoch: [20][ 370/ 1207] Overall Loss 0.427640 Objective Loss 0.427640 LR 0.001000 Time 0.020864 -2023-02-13 17:27:36,898 - Epoch: [20][ 380/ 1207] Overall Loss 0.427732 Objective Loss 0.427732 LR 0.001000 Time 0.020812 -2023-02-13 17:27:37,087 - Epoch: [20][ 390/ 1207] Overall Loss 0.428526 Objective Loss 0.428526 LR 0.001000 Time 0.020764 -2023-02-13 17:27:37,276 - Epoch: [20][ 400/ 1207] Overall Loss 0.428743 Objective Loss 0.428743 LR 0.001000 Time 0.020714 -2023-02-13 17:27:37,465 - Epoch: [20][ 410/ 1207] Overall Loss 0.428794 Objective Loss 0.428794 LR 0.001000 Time 0.020671 -2023-02-13 17:27:37,655 - Epoch: [20][ 420/ 1207] Overall Loss 0.429594 Objective Loss 0.429594 LR 0.001000 Time 0.020629 -2023-02-13 17:27:37,843 - Epoch: [20][ 430/ 1207] Overall Loss 0.430717 Objective Loss 0.430717 LR 0.001000 Time 0.020587 -2023-02-13 17:27:38,032 - Epoch: [20][ 440/ 1207] Overall Loss 0.430797 Objective Loss 0.430797 LR 0.001000 Time 0.020548 -2023-02-13 17:27:38,221 - Epoch: [20][ 450/ 1207] Overall Loss 0.430612 Objective Loss 0.430612 LR 0.001000 Time 0.020510 -2023-02-13 17:27:38,410 - Epoch: [20][ 460/ 1207] Overall Loss 0.431109 Objective Loss 0.431109 LR 0.001000 Time 0.020473 -2023-02-13 17:27:38,598 - Epoch: [20][ 470/ 1207] Overall Loss 0.431473 Objective Loss 0.431473 LR 0.001000 Time 0.020438 -2023-02-13 17:27:38,788 - Epoch: [20][ 480/ 1207] Overall Loss 0.431329 Objective Loss 0.431329 LR 0.001000 Time 0.020406 -2023-02-13 17:27:38,977 - Epoch: [20][ 490/ 1207] Overall Loss 0.431456 Objective Loss 0.431456 LR 0.001000 Time 0.020374 -2023-02-13 17:27:39,165 - Epoch: [20][ 500/ 1207] Overall Loss 0.431458 Objective Loss 0.431458 LR 0.001000 Time 0.020342 -2023-02-13 17:27:39,354 - Epoch: [20][ 510/ 1207] Overall Loss 0.431662 Objective Loss 0.431662 LR 0.001000 Time 0.020313 -2023-02-13 17:27:39,542 - Epoch: [20][ 520/ 1207] Overall Loss 0.431901 Objective Loss 0.431901 LR 0.001000 Time 0.020285 -2023-02-13 17:27:39,732 - Epoch: [20][ 530/ 1207] Overall Loss 0.432114 Objective Loss 0.432114 LR 0.001000 Time 0.020259 -2023-02-13 17:27:39,920 - Epoch: [20][ 540/ 1207] Overall Loss 0.432812 Objective Loss 0.432812 LR 0.001000 Time 0.020232 -2023-02-13 17:27:40,109 - Epoch: [20][ 550/ 1207] Overall Loss 0.432572 Objective Loss 0.432572 LR 0.001000 Time 0.020207 -2023-02-13 17:27:40,298 - Epoch: [20][ 560/ 1207] Overall Loss 0.431971 Objective Loss 0.431971 LR 0.001000 Time 0.020183 -2023-02-13 17:27:40,486 - Epoch: [20][ 570/ 1207] Overall Loss 0.431406 Objective Loss 0.431406 LR 0.001000 Time 0.020158 -2023-02-13 17:27:40,675 - Epoch: [20][ 580/ 1207] Overall Loss 0.431471 Objective Loss 0.431471 LR 0.001000 Time 0.020135 -2023-02-13 17:27:40,865 - Epoch: [20][ 590/ 1207] Overall Loss 0.431769 Objective Loss 0.431769 LR 0.001000 Time 0.020116 -2023-02-13 17:27:41,054 - Epoch: [20][ 600/ 1207] Overall Loss 0.431946 Objective Loss 0.431946 LR 0.001000 Time 0.020095 -2023-02-13 17:27:41,243 - Epoch: [20][ 610/ 1207] Overall Loss 0.431265 Objective Loss 0.431265 LR 0.001000 Time 0.020075 -2023-02-13 17:27:41,431 - Epoch: [20][ 620/ 1207] Overall Loss 0.431295 Objective Loss 0.431295 LR 0.001000 Time 0.020054 -2023-02-13 17:27:41,620 - Epoch: [20][ 630/ 1207] Overall Loss 0.431241 Objective Loss 0.431241 LR 0.001000 Time 0.020034 -2023-02-13 17:27:41,809 - Epoch: [20][ 640/ 1207] Overall Loss 0.430688 Objective Loss 0.430688 LR 0.001000 Time 0.020016 -2023-02-13 17:27:41,998 - Epoch: [20][ 650/ 1207] Overall Loss 0.430449 Objective Loss 0.430449 LR 0.001000 Time 0.019999 -2023-02-13 17:27:42,187 - Epoch: [20][ 660/ 1207] Overall Loss 0.430333 Objective Loss 0.430333 LR 0.001000 Time 0.019982 -2023-02-13 17:27:42,376 - Epoch: [20][ 670/ 1207] Overall Loss 0.430358 Objective Loss 0.430358 LR 0.001000 Time 0.019965 -2023-02-13 17:27:42,565 - Epoch: [20][ 680/ 1207] Overall Loss 0.429821 Objective Loss 0.429821 LR 0.001000 Time 0.019948 -2023-02-13 17:27:42,754 - Epoch: [20][ 690/ 1207] Overall Loss 0.429362 Objective Loss 0.429362 LR 0.001000 Time 0.019932 -2023-02-13 17:27:42,943 - Epoch: [20][ 700/ 1207] Overall Loss 0.429436 Objective Loss 0.429436 LR 0.001000 Time 0.019917 -2023-02-13 17:27:43,132 - Epoch: [20][ 710/ 1207] Overall Loss 0.429522 Objective Loss 0.429522 LR 0.001000 Time 0.019902 -2023-02-13 17:27:43,320 - Epoch: [20][ 720/ 1207] Overall Loss 0.429372 Objective Loss 0.429372 LR 0.001000 Time 0.019886 -2023-02-13 17:27:43,509 - Epoch: [20][ 730/ 1207] Overall Loss 0.428976 Objective Loss 0.428976 LR 0.001000 Time 0.019872 -2023-02-13 17:27:43,698 - Epoch: [20][ 740/ 1207] Overall Loss 0.429122 Objective Loss 0.429122 LR 0.001000 Time 0.019859 -2023-02-13 17:27:43,887 - Epoch: [20][ 750/ 1207] Overall Loss 0.428827 Objective Loss 0.428827 LR 0.001000 Time 0.019846 -2023-02-13 17:27:44,076 - Epoch: [20][ 760/ 1207] Overall Loss 0.428078 Objective Loss 0.428078 LR 0.001000 Time 0.019832 -2023-02-13 17:27:44,265 - Epoch: [20][ 770/ 1207] Overall Loss 0.427799 Objective Loss 0.427799 LR 0.001000 Time 0.019820 -2023-02-13 17:27:44,454 - Epoch: [20][ 780/ 1207] Overall Loss 0.427685 Objective Loss 0.427685 LR 0.001000 Time 0.019808 -2023-02-13 17:27:44,642 - Epoch: [20][ 790/ 1207] Overall Loss 0.428008 Objective Loss 0.428008 LR 0.001000 Time 0.019795 -2023-02-13 17:27:44,832 - Epoch: [20][ 800/ 1207] Overall Loss 0.428006 Objective Loss 0.428006 LR 0.001000 Time 0.019784 -2023-02-13 17:27:45,021 - Epoch: [20][ 810/ 1207] Overall Loss 0.427565 Objective Loss 0.427565 LR 0.001000 Time 0.019773 -2023-02-13 17:27:45,211 - Epoch: [20][ 820/ 1207] Overall Loss 0.427986 Objective Loss 0.427986 LR 0.001000 Time 0.019763 -2023-02-13 17:27:45,400 - Epoch: [20][ 830/ 1207] Overall Loss 0.427955 Objective Loss 0.427955 LR 0.001000 Time 0.019752 -2023-02-13 17:27:45,588 - Epoch: [20][ 840/ 1207] Overall Loss 0.427656 Objective Loss 0.427656 LR 0.001000 Time 0.019741 -2023-02-13 17:27:45,779 - Epoch: [20][ 850/ 1207] Overall Loss 0.427598 Objective Loss 0.427598 LR 0.001000 Time 0.019732 -2023-02-13 17:27:45,967 - Epoch: [20][ 860/ 1207] Overall Loss 0.427233 Objective Loss 0.427233 LR 0.001000 Time 0.019721 -2023-02-13 17:27:46,157 - Epoch: [20][ 870/ 1207] Overall Loss 0.427196 Objective Loss 0.427196 LR 0.001000 Time 0.019712 -2023-02-13 17:27:46,345 - Epoch: [20][ 880/ 1207] Overall Loss 0.426858 Objective Loss 0.426858 LR 0.001000 Time 0.019702 -2023-02-13 17:27:46,534 - Epoch: [20][ 890/ 1207] Overall Loss 0.427070 Objective Loss 0.427070 LR 0.001000 Time 0.019693 -2023-02-13 17:27:46,724 - Epoch: [20][ 900/ 1207] Overall Loss 0.427101 Objective Loss 0.427101 LR 0.001000 Time 0.019684 -2023-02-13 17:27:46,914 - Epoch: [20][ 910/ 1207] Overall Loss 0.427036 Objective Loss 0.427036 LR 0.001000 Time 0.019676 -2023-02-13 17:27:47,103 - Epoch: [20][ 920/ 1207] Overall Loss 0.426821 Objective Loss 0.426821 LR 0.001000 Time 0.019667 -2023-02-13 17:27:47,292 - Epoch: [20][ 930/ 1207] Overall Loss 0.426773 Objective Loss 0.426773 LR 0.001000 Time 0.019659 -2023-02-13 17:27:47,480 - Epoch: [20][ 940/ 1207] Overall Loss 0.426920 Objective Loss 0.426920 LR 0.001000 Time 0.019649 -2023-02-13 17:27:47,670 - Epoch: [20][ 950/ 1207] Overall Loss 0.426771 Objective Loss 0.426771 LR 0.001000 Time 0.019641 -2023-02-13 17:27:47,858 - Epoch: [20][ 960/ 1207] Overall Loss 0.427061 Objective Loss 0.427061 LR 0.001000 Time 0.019633 -2023-02-13 17:27:48,048 - Epoch: [20][ 970/ 1207] Overall Loss 0.426911 Objective Loss 0.426911 LR 0.001000 Time 0.019626 -2023-02-13 17:27:48,237 - Epoch: [20][ 980/ 1207] Overall Loss 0.426876 Objective Loss 0.426876 LR 0.001000 Time 0.019618 -2023-02-13 17:27:48,425 - Epoch: [20][ 990/ 1207] Overall Loss 0.426686 Objective Loss 0.426686 LR 0.001000 Time 0.019610 -2023-02-13 17:27:48,615 - Epoch: [20][ 1000/ 1207] Overall Loss 0.427106 Objective Loss 0.427106 LR 0.001000 Time 0.019603 -2023-02-13 17:27:48,805 - Epoch: [20][ 1010/ 1207] Overall Loss 0.427256 Objective Loss 0.427256 LR 0.001000 Time 0.019597 -2023-02-13 17:27:48,994 - Epoch: [20][ 1020/ 1207] Overall Loss 0.427001 Objective Loss 0.427001 LR 0.001000 Time 0.019589 -2023-02-13 17:27:49,184 - Epoch: [20][ 1030/ 1207] Overall Loss 0.426897 Objective Loss 0.426897 LR 0.001000 Time 0.019583 -2023-02-13 17:27:49,371 - Epoch: [20][ 1040/ 1207] Overall Loss 0.426770 Objective Loss 0.426770 LR 0.001000 Time 0.019575 -2023-02-13 17:27:49,561 - Epoch: [20][ 1050/ 1207] Overall Loss 0.426781 Objective Loss 0.426781 LR 0.001000 Time 0.019569 -2023-02-13 17:27:49,751 - Epoch: [20][ 1060/ 1207] Overall Loss 0.426682 Objective Loss 0.426682 LR 0.001000 Time 0.019563 -2023-02-13 17:27:49,941 - Epoch: [20][ 1070/ 1207] Overall Loss 0.426753 Objective Loss 0.426753 LR 0.001000 Time 0.019557 -2023-02-13 17:27:50,129 - Epoch: [20][ 1080/ 1207] Overall Loss 0.426690 Objective Loss 0.426690 LR 0.001000 Time 0.019550 -2023-02-13 17:27:50,317 - Epoch: [20][ 1090/ 1207] Overall Loss 0.426534 Objective Loss 0.426534 LR 0.001000 Time 0.019543 -2023-02-13 17:27:50,507 - Epoch: [20][ 1100/ 1207] Overall Loss 0.426608 Objective Loss 0.426608 LR 0.001000 Time 0.019538 -2023-02-13 17:27:50,696 - Epoch: [20][ 1110/ 1207] Overall Loss 0.426718 Objective Loss 0.426718 LR 0.001000 Time 0.019531 -2023-02-13 17:27:50,886 - Epoch: [20][ 1120/ 1207] Overall Loss 0.426736 Objective Loss 0.426736 LR 0.001000 Time 0.019526 -2023-02-13 17:27:51,075 - Epoch: [20][ 1130/ 1207] Overall Loss 0.426637 Objective Loss 0.426637 LR 0.001000 Time 0.019520 -2023-02-13 17:27:51,264 - Epoch: [20][ 1140/ 1207] Overall Loss 0.426371 Objective Loss 0.426371 LR 0.001000 Time 0.019514 -2023-02-13 17:27:51,452 - Epoch: [20][ 1150/ 1207] Overall Loss 0.426482 Objective Loss 0.426482 LR 0.001000 Time 0.019508 -2023-02-13 17:27:51,640 - Epoch: [20][ 1160/ 1207] Overall Loss 0.426258 Objective Loss 0.426258 LR 0.001000 Time 0.019502 -2023-02-13 17:27:51,830 - Epoch: [20][ 1170/ 1207] Overall Loss 0.426142 Objective Loss 0.426142 LR 0.001000 Time 0.019497 -2023-02-13 17:27:52,018 - Epoch: [20][ 1180/ 1207] Overall Loss 0.426130 Objective Loss 0.426130 LR 0.001000 Time 0.019491 -2023-02-13 17:27:52,209 - Epoch: [20][ 1190/ 1207] Overall Loss 0.425997 Objective Loss 0.425997 LR 0.001000 Time 0.019487 -2023-02-13 17:27:52,448 - Epoch: [20][ 1200/ 1207] Overall Loss 0.426139 Objective Loss 0.426139 LR 0.001000 Time 0.019524 -2023-02-13 17:27:52,564 - Epoch: [20][ 1207/ 1207] Overall Loss 0.426017 Objective Loss 0.426017 Top1 83.841463 Top5 98.170732 LR 0.001000 Time 0.019506 -2023-02-13 17:27:52,643 - --- validate (epoch=20)----------- -2023-02-13 17:27:52,644 - 34311 samples (256 per mini-batch) -2023-02-13 17:27:53,058 - Epoch: [20][ 10/ 135] Loss 0.426087 Top1 77.656250 Top5 96.445312 -2023-02-13 17:27:53,198 - Epoch: [20][ 20/ 135] Loss 0.432772 Top1 78.417969 Top5 96.679688 -2023-02-13 17:27:53,338 - Epoch: [20][ 30/ 135] Loss 0.432781 Top1 78.437500 Top5 96.601562 -2023-02-13 17:27:53,475 - Epoch: [20][ 40/ 135] Loss 0.431895 Top1 78.652344 Top5 96.738281 -2023-02-13 17:27:53,611 - Epoch: [20][ 50/ 135] Loss 0.424627 Top1 78.656250 Top5 96.781250 -2023-02-13 17:27:53,743 - Epoch: [20][ 60/ 135] Loss 0.419106 Top1 78.860677 Top5 96.757812 -2023-02-13 17:27:53,881 - Epoch: [20][ 70/ 135] Loss 0.418473 Top1 78.822545 Top5 96.752232 -2023-02-13 17:27:54,026 - Epoch: [20][ 80/ 135] Loss 0.418995 Top1 78.808594 Top5 96.757812 -2023-02-13 17:27:54,164 - Epoch: [20][ 90/ 135] Loss 0.418374 Top1 78.854167 Top5 96.775174 -2023-02-13 17:27:54,308 - Epoch: [20][ 100/ 135] Loss 0.415999 Top1 79.019531 Top5 96.820312 -2023-02-13 17:27:54,445 - Epoch: [20][ 110/ 135] Loss 0.418684 Top1 78.945312 Top5 96.807528 -2023-02-13 17:27:54,582 - Epoch: [20][ 120/ 135] Loss 0.417308 Top1 78.990885 Top5 96.754557 -2023-02-13 17:27:54,719 - Epoch: [20][ 130/ 135] Loss 0.418285 Top1 78.888221 Top5 96.796875 -2023-02-13 17:27:54,764 - Epoch: [20][ 135/ 135] Loss 0.418688 Top1 78.925126 Top5 96.785288 -2023-02-13 17:27:54,846 - ==> Top1: 78.925 Top5: 96.785 Loss: 0.419 - -2023-02-13 17:27:54,847 - ==> Confusion: -[[ 848 6 8 2 7 8 0 3 6 44 0 4 1 7 7 1 4 1 1 0 9] - [ 3 933 1 1 5 43 2 16 2 1 2 2 1 2 2 0 6 1 7 2 1] - [ 12 11 940 13 3 4 11 17 0 1 5 2 2 3 1 5 1 3 9 4 11] - [ 3 2 28 866 0 9 1 1 3 3 14 0 9 4 25 2 6 6 27 1 6] - [ 22 14 0 0 958 23 2 1 2 5 0 3 1 4 7 5 9 0 1 2 7] - [ 2 32 1 5 1 954 3 20 1 5 1 12 6 14 2 0 4 1 1 3 2] - [ 4 6 30 2 0 2 1014 7 0 0 3 0 3 1 0 3 4 3 1 10 6] - [ 1 16 15 1 1 32 3 898 1 1 1 6 2 1 0 1 0 2 32 7 3] - [ 23 6 0 1 2 3 0 1 852 42 11 1 1 21 34 0 2 2 6 1 0] - [ 120 2 2 0 4 8 0 2 36 789 0 1 0 26 6 0 1 2 0 2 11] - [ 3 6 7 8 1 2 6 4 16 3 953 2 0 14 4 0 0 0 17 1 4] - [ 3 3 0 0 0 17 1 2 2 2 0 905 25 12 2 7 3 6 3 11 1] - [ 4 2 1 3 1 6 0 0 1 0 0 42 856 0 6 4 1 21 0 4 7] - [ 7 5 3 2 5 18 1 3 7 10 4 8 4 926 5 5 5 0 1 2 3] - [ 16 3 1 18 5 4 0 2 18 5 2 2 3 3 971 1 5 5 15 0 13] - [ 9 4 6 1 7 4 8 0 0 1 0 7 5 2 1 948 21 10 0 7 5] - [ 4 12 1 1 7 4 1 0 2 1 1 3 2 2 1 14 997 1 1 3 3] - [ 8 5 0 6 1 2 0 2 0 1 1 17 30 1 0 7 0 963 0 0 7] - [ 3 5 2 6 1 1 1 32 2 1 3 0 4 0 17 0 1 1 1000 3 3] - [ 0 5 2 2 0 13 10 22 0 0 0 17 4 9 0 5 6 5 1 1040 7] - [ 239 435 308 152 112 402 105 225 100 88 195 171 436 457 227 109 453 118 272 361 8469]] - -2023-02-13 17:27:54,849 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:27:54,849 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:27:54,855 - - -2023-02-13 17:27:54,855 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:27:55,726 - Epoch: [21][ 10/ 1207] Overall Loss 0.426433 Objective Loss 0.426433 LR 0.001000 Time 0.087052 -2023-02-13 17:27:55,929 - Epoch: [21][ 20/ 1207] Overall Loss 0.407254 Objective Loss 0.407254 LR 0.001000 Time 0.053682 -2023-02-13 17:27:56,126 - Epoch: [21][ 30/ 1207] Overall Loss 0.397648 Objective Loss 0.397648 LR 0.001000 Time 0.042317 -2023-02-13 17:27:56,319 - Epoch: [21][ 40/ 1207] Overall Loss 0.398423 Objective Loss 0.398423 LR 0.001000 Time 0.036555 -2023-02-13 17:27:56,514 - Epoch: [21][ 50/ 1207] Overall Loss 0.399810 Objective Loss 0.399810 LR 0.001000 Time 0.033143 -2023-02-13 17:27:56,707 - Epoch: [21][ 60/ 1207] Overall Loss 0.398332 Objective Loss 0.398332 LR 0.001000 Time 0.030831 -2023-02-13 17:27:56,903 - Epoch: [21][ 70/ 1207] Overall Loss 0.398055 Objective Loss 0.398055 LR 0.001000 Time 0.029226 -2023-02-13 17:27:57,097 - Epoch: [21][ 80/ 1207] Overall Loss 0.397114 Objective Loss 0.397114 LR 0.001000 Time 0.027986 -2023-02-13 17:27:57,293 - Epoch: [21][ 90/ 1207] Overall Loss 0.399191 Objective Loss 0.399191 LR 0.001000 Time 0.027046 -2023-02-13 17:27:57,485 - Epoch: [21][ 100/ 1207] Overall Loss 0.402152 Objective Loss 0.402152 LR 0.001000 Time 0.026263 -2023-02-13 17:27:57,681 - Epoch: [21][ 110/ 1207] Overall Loss 0.401545 Objective Loss 0.401545 LR 0.001000 Time 0.025654 -2023-02-13 17:27:57,876 - Epoch: [21][ 120/ 1207] Overall Loss 0.401644 Objective Loss 0.401644 LR 0.001000 Time 0.025140 -2023-02-13 17:27:58,073 - Epoch: [21][ 130/ 1207] Overall Loss 0.402403 Objective Loss 0.402403 LR 0.001000 Time 0.024715 -2023-02-13 17:27:58,270 - Epoch: [21][ 140/ 1207] Overall Loss 0.404085 Objective Loss 0.404085 LR 0.001000 Time 0.024354 -2023-02-13 17:27:58,458 - Epoch: [21][ 150/ 1207] Overall Loss 0.403132 Objective Loss 0.403132 LR 0.001000 Time 0.023982 -2023-02-13 17:27:58,646 - Epoch: [21][ 160/ 1207] Overall Loss 0.407025 Objective Loss 0.407025 LR 0.001000 Time 0.023654 -2023-02-13 17:27:58,834 - Epoch: [21][ 170/ 1207] Overall Loss 0.406950 Objective Loss 0.406950 LR 0.001000 Time 0.023370 -2023-02-13 17:27:59,023 - Epoch: [21][ 180/ 1207] Overall Loss 0.405557 Objective Loss 0.405557 LR 0.001000 Time 0.023116 -2023-02-13 17:27:59,210 - Epoch: [21][ 190/ 1207] Overall Loss 0.405759 Objective Loss 0.405759 LR 0.001000 Time 0.022885 -2023-02-13 17:27:59,398 - Epoch: [21][ 200/ 1207] Overall Loss 0.405233 Objective Loss 0.405233 LR 0.001000 Time 0.022677 -2023-02-13 17:27:59,586 - Epoch: [21][ 210/ 1207] Overall Loss 0.405704 Objective Loss 0.405704 LR 0.001000 Time 0.022492 -2023-02-13 17:27:59,776 - Epoch: [21][ 220/ 1207] Overall Loss 0.404271 Objective Loss 0.404271 LR 0.001000 Time 0.022330 -2023-02-13 17:27:59,965 - Epoch: [21][ 230/ 1207] Overall Loss 0.402889 Objective Loss 0.402889 LR 0.001000 Time 0.022180 -2023-02-13 17:28:00,154 - Epoch: [21][ 240/ 1207] Overall Loss 0.403292 Objective Loss 0.403292 LR 0.001000 Time 0.022042 -2023-02-13 17:28:00,343 - Epoch: [21][ 250/ 1207] Overall Loss 0.403555 Objective Loss 0.403555 LR 0.001000 Time 0.021915 -2023-02-13 17:28:00,532 - Epoch: [21][ 260/ 1207] Overall Loss 0.403288 Objective Loss 0.403288 LR 0.001000 Time 0.021795 -2023-02-13 17:28:00,721 - Epoch: [21][ 270/ 1207] Overall Loss 0.402378 Objective Loss 0.402378 LR 0.001000 Time 0.021687 -2023-02-13 17:28:00,910 - Epoch: [21][ 280/ 1207] Overall Loss 0.402326 Objective Loss 0.402326 LR 0.001000 Time 0.021587 -2023-02-13 17:28:01,099 - Epoch: [21][ 290/ 1207] Overall Loss 0.402053 Objective Loss 0.402053 LR 0.001000 Time 0.021493 -2023-02-13 17:28:01,287 - Epoch: [21][ 300/ 1207] Overall Loss 0.402038 Objective Loss 0.402038 LR 0.001000 Time 0.021403 -2023-02-13 17:28:01,475 - Epoch: [21][ 310/ 1207] Overall Loss 0.402494 Objective Loss 0.402494 LR 0.001000 Time 0.021318 -2023-02-13 17:28:01,664 - Epoch: [21][ 320/ 1207] Overall Loss 0.403384 Objective Loss 0.403384 LR 0.001000 Time 0.021240 -2023-02-13 17:28:01,854 - Epoch: [21][ 330/ 1207] Overall Loss 0.403879 Objective Loss 0.403879 LR 0.001000 Time 0.021171 -2023-02-13 17:28:02,042 - Epoch: [21][ 340/ 1207] Overall Loss 0.404525 Objective Loss 0.404525 LR 0.001000 Time 0.021101 -2023-02-13 17:28:02,231 - Epoch: [21][ 350/ 1207] Overall Loss 0.404937 Objective Loss 0.404937 LR 0.001000 Time 0.021038 -2023-02-13 17:28:02,420 - Epoch: [21][ 360/ 1207] Overall Loss 0.405165 Objective Loss 0.405165 LR 0.001000 Time 0.020976 -2023-02-13 17:28:02,608 - Epoch: [21][ 370/ 1207] Overall Loss 0.405259 Objective Loss 0.405259 LR 0.001000 Time 0.020917 -2023-02-13 17:28:02,798 - Epoch: [21][ 380/ 1207] Overall Loss 0.405885 Objective Loss 0.405885 LR 0.001000 Time 0.020865 -2023-02-13 17:28:02,987 - Epoch: [21][ 390/ 1207] Overall Loss 0.405270 Objective Loss 0.405270 LR 0.001000 Time 0.020813 -2023-02-13 17:28:03,176 - Epoch: [21][ 400/ 1207] Overall Loss 0.405248 Objective Loss 0.405248 LR 0.001000 Time 0.020764 -2023-02-13 17:28:03,364 - Epoch: [21][ 410/ 1207] Overall Loss 0.404598 Objective Loss 0.404598 LR 0.001000 Time 0.020716 -2023-02-13 17:28:03,552 - Epoch: [21][ 420/ 1207] Overall Loss 0.404345 Objective Loss 0.404345 LR 0.001000 Time 0.020670 -2023-02-13 17:28:03,742 - Epoch: [21][ 430/ 1207] Overall Loss 0.404095 Objective Loss 0.404095 LR 0.001000 Time 0.020630 -2023-02-13 17:28:03,931 - Epoch: [21][ 440/ 1207] Overall Loss 0.403970 Objective Loss 0.403970 LR 0.001000 Time 0.020589 -2023-02-13 17:28:04,120 - Epoch: [21][ 450/ 1207] Overall Loss 0.403406 Objective Loss 0.403406 LR 0.001000 Time 0.020551 -2023-02-13 17:28:04,309 - Epoch: [21][ 460/ 1207] Overall Loss 0.402661 Objective Loss 0.402661 LR 0.001000 Time 0.020514 -2023-02-13 17:28:04,497 - Epoch: [21][ 470/ 1207] Overall Loss 0.402240 Objective Loss 0.402240 LR 0.001000 Time 0.020477 -2023-02-13 17:28:04,685 - Epoch: [21][ 480/ 1207] Overall Loss 0.402356 Objective Loss 0.402356 LR 0.001000 Time 0.020442 -2023-02-13 17:28:04,874 - Epoch: [21][ 490/ 1207] Overall Loss 0.402393 Objective Loss 0.402393 LR 0.001000 Time 0.020409 -2023-02-13 17:28:05,063 - Epoch: [21][ 500/ 1207] Overall Loss 0.401858 Objective Loss 0.401858 LR 0.001000 Time 0.020378 -2023-02-13 17:28:05,251 - Epoch: [21][ 510/ 1207] Overall Loss 0.402199 Objective Loss 0.402199 LR 0.001000 Time 0.020347 -2023-02-13 17:28:05,440 - Epoch: [21][ 520/ 1207] Overall Loss 0.402104 Objective Loss 0.402104 LR 0.001000 Time 0.020318 -2023-02-13 17:28:05,628 - Epoch: [21][ 530/ 1207] Overall Loss 0.402212 Objective Loss 0.402212 LR 0.001000 Time 0.020289 -2023-02-13 17:28:05,817 - Epoch: [21][ 540/ 1207] Overall Loss 0.401467 Objective Loss 0.401467 LR 0.001000 Time 0.020263 -2023-02-13 17:28:06,007 - Epoch: [21][ 550/ 1207] Overall Loss 0.401540 Objective Loss 0.401540 LR 0.001000 Time 0.020238 -2023-02-13 17:28:06,195 - Epoch: [21][ 560/ 1207] Overall Loss 0.401920 Objective Loss 0.401920 LR 0.001000 Time 0.020212 -2023-02-13 17:28:06,384 - Epoch: [21][ 570/ 1207] Overall Loss 0.401797 Objective Loss 0.401797 LR 0.001000 Time 0.020188 -2023-02-13 17:28:06,572 - Epoch: [21][ 580/ 1207] Overall Loss 0.401663 Objective Loss 0.401663 LR 0.001000 Time 0.020164 -2023-02-13 17:28:06,762 - Epoch: [21][ 590/ 1207] Overall Loss 0.401413 Objective Loss 0.401413 LR 0.001000 Time 0.020143 -2023-02-13 17:28:06,951 - Epoch: [21][ 600/ 1207] Overall Loss 0.401498 Objective Loss 0.401498 LR 0.001000 Time 0.020122 -2023-02-13 17:28:07,139 - Epoch: [21][ 610/ 1207] Overall Loss 0.401874 Objective Loss 0.401874 LR 0.001000 Time 0.020100 -2023-02-13 17:28:07,328 - Epoch: [21][ 620/ 1207] Overall Loss 0.401807 Objective Loss 0.401807 LR 0.001000 Time 0.020081 -2023-02-13 17:28:07,517 - Epoch: [21][ 630/ 1207] Overall Loss 0.402096 Objective Loss 0.402096 LR 0.001000 Time 0.020061 -2023-02-13 17:28:07,706 - Epoch: [21][ 640/ 1207] Overall Loss 0.402062 Objective Loss 0.402062 LR 0.001000 Time 0.020042 -2023-02-13 17:28:07,895 - Epoch: [21][ 650/ 1207] Overall Loss 0.401678 Objective Loss 0.401678 LR 0.001000 Time 0.020025 -2023-02-13 17:28:08,085 - Epoch: [21][ 660/ 1207] Overall Loss 0.402397 Objective Loss 0.402397 LR 0.001000 Time 0.020007 -2023-02-13 17:28:08,274 - Epoch: [21][ 670/ 1207] Overall Loss 0.402413 Objective Loss 0.402413 LR 0.001000 Time 0.019990 -2023-02-13 17:28:08,462 - Epoch: [21][ 680/ 1207] Overall Loss 0.402855 Objective Loss 0.402855 LR 0.001000 Time 0.019973 -2023-02-13 17:28:08,651 - Epoch: [21][ 690/ 1207] Overall Loss 0.402791 Objective Loss 0.402791 LR 0.001000 Time 0.019957 -2023-02-13 17:28:08,841 - Epoch: [21][ 700/ 1207] Overall Loss 0.403061 Objective Loss 0.403061 LR 0.001000 Time 0.019942 -2023-02-13 17:28:09,030 - Epoch: [21][ 710/ 1207] Overall Loss 0.402977 Objective Loss 0.402977 LR 0.001000 Time 0.019927 -2023-02-13 17:28:09,219 - Epoch: [21][ 720/ 1207] Overall Loss 0.403169 Objective Loss 0.403169 LR 0.001000 Time 0.019912 -2023-02-13 17:28:09,408 - Epoch: [21][ 730/ 1207] Overall Loss 0.402729 Objective Loss 0.402729 LR 0.001000 Time 0.019897 -2023-02-13 17:28:09,598 - Epoch: [21][ 740/ 1207] Overall Loss 0.402466 Objective Loss 0.402466 LR 0.001000 Time 0.019885 -2023-02-13 17:28:09,787 - Epoch: [21][ 750/ 1207] Overall Loss 0.402391 Objective Loss 0.402391 LR 0.001000 Time 0.019872 -2023-02-13 17:28:09,976 - Epoch: [21][ 760/ 1207] Overall Loss 0.402531 Objective Loss 0.402531 LR 0.001000 Time 0.019858 -2023-02-13 17:28:10,165 - Epoch: [21][ 770/ 1207] Overall Loss 0.402425 Objective Loss 0.402425 LR 0.001000 Time 0.019845 -2023-02-13 17:28:10,353 - Epoch: [21][ 780/ 1207] Overall Loss 0.402701 Objective Loss 0.402701 LR 0.001000 Time 0.019832 -2023-02-13 17:28:10,542 - Epoch: [21][ 790/ 1207] Overall Loss 0.402724 Objective Loss 0.402724 LR 0.001000 Time 0.019819 -2023-02-13 17:28:10,730 - Epoch: [21][ 800/ 1207] Overall Loss 0.402790 Objective Loss 0.402790 LR 0.001000 Time 0.019806 -2023-02-13 17:28:10,921 - Epoch: [21][ 810/ 1207] Overall Loss 0.402919 Objective Loss 0.402919 LR 0.001000 Time 0.019797 -2023-02-13 17:28:11,110 - Epoch: [21][ 820/ 1207] Overall Loss 0.402967 Objective Loss 0.402967 LR 0.001000 Time 0.019785 -2023-02-13 17:28:11,299 - Epoch: [21][ 830/ 1207] Overall Loss 0.402445 Objective Loss 0.402445 LR 0.001000 Time 0.019774 -2023-02-13 17:28:11,488 - Epoch: [21][ 840/ 1207] Overall Loss 0.402354 Objective Loss 0.402354 LR 0.001000 Time 0.019763 -2023-02-13 17:28:11,682 - Epoch: [21][ 850/ 1207] Overall Loss 0.402385 Objective Loss 0.402385 LR 0.001000 Time 0.019758 -2023-02-13 17:28:11,887 - Epoch: [21][ 860/ 1207] Overall Loss 0.402506 Objective Loss 0.402506 LR 0.001000 Time 0.019767 -2023-02-13 17:28:12,086 - Epoch: [21][ 870/ 1207] Overall Loss 0.402547 Objective Loss 0.402547 LR 0.001000 Time 0.019768 -2023-02-13 17:28:12,289 - Epoch: [21][ 880/ 1207] Overall Loss 0.402712 Objective Loss 0.402712 LR 0.001000 Time 0.019774 -2023-02-13 17:28:12,488 - Epoch: [21][ 890/ 1207] Overall Loss 0.402864 Objective Loss 0.402864 LR 0.001000 Time 0.019775 -2023-02-13 17:28:12,693 - Epoch: [21][ 900/ 1207] Overall Loss 0.403372 Objective Loss 0.403372 LR 0.001000 Time 0.019782 -2023-02-13 17:28:12,893 - Epoch: [21][ 910/ 1207] Overall Loss 0.403722 Objective Loss 0.403722 LR 0.001000 Time 0.019784 -2023-02-13 17:28:13,097 - Epoch: [21][ 920/ 1207] Overall Loss 0.403899 Objective Loss 0.403899 LR 0.001000 Time 0.019790 -2023-02-13 17:28:13,297 - Epoch: [21][ 930/ 1207] Overall Loss 0.404093 Objective Loss 0.404093 LR 0.001000 Time 0.019792 -2023-02-13 17:28:13,501 - Epoch: [21][ 940/ 1207] Overall Loss 0.404509 Objective Loss 0.404509 LR 0.001000 Time 0.019798 -2023-02-13 17:28:13,700 - Epoch: [21][ 950/ 1207] Overall Loss 0.404732 Objective Loss 0.404732 LR 0.001000 Time 0.019799 -2023-02-13 17:28:13,905 - Epoch: [21][ 960/ 1207] Overall Loss 0.405086 Objective Loss 0.405086 LR 0.001000 Time 0.019806 -2023-02-13 17:28:14,104 - Epoch: [21][ 970/ 1207] Overall Loss 0.405034 Objective Loss 0.405034 LR 0.001000 Time 0.019806 -2023-02-13 17:28:14,308 - Epoch: [21][ 980/ 1207] Overall Loss 0.405247 Objective Loss 0.405247 LR 0.001000 Time 0.019812 -2023-02-13 17:28:14,507 - Epoch: [21][ 990/ 1207] Overall Loss 0.405185 Objective Loss 0.405185 LR 0.001000 Time 0.019813 -2023-02-13 17:28:14,711 - Epoch: [21][ 1000/ 1207] Overall Loss 0.405359 Objective Loss 0.405359 LR 0.001000 Time 0.019818 -2023-02-13 17:28:14,912 - Epoch: [21][ 1010/ 1207] Overall Loss 0.405224 Objective Loss 0.405224 LR 0.001000 Time 0.019820 -2023-02-13 17:28:15,117 - Epoch: [21][ 1020/ 1207] Overall Loss 0.405103 Objective Loss 0.405103 LR 0.001000 Time 0.019826 -2023-02-13 17:28:15,315 - Epoch: [21][ 1030/ 1207] Overall Loss 0.405123 Objective Loss 0.405123 LR 0.001000 Time 0.019826 -2023-02-13 17:28:15,519 - Epoch: [21][ 1040/ 1207] Overall Loss 0.404997 Objective Loss 0.404997 LR 0.001000 Time 0.019831 -2023-02-13 17:28:15,719 - Epoch: [21][ 1050/ 1207] Overall Loss 0.404927 Objective Loss 0.404927 LR 0.001000 Time 0.019832 -2023-02-13 17:28:15,925 - Epoch: [21][ 1060/ 1207] Overall Loss 0.405163 Objective Loss 0.405163 LR 0.001000 Time 0.019839 -2023-02-13 17:28:16,125 - Epoch: [21][ 1070/ 1207] Overall Loss 0.405006 Objective Loss 0.405006 LR 0.001000 Time 0.019840 -2023-02-13 17:28:16,329 - Epoch: [21][ 1080/ 1207] Overall Loss 0.405040 Objective Loss 0.405040 LR 0.001000 Time 0.019845 -2023-02-13 17:28:16,529 - Epoch: [21][ 1090/ 1207] Overall Loss 0.404866 Objective Loss 0.404866 LR 0.001000 Time 0.019846 -2023-02-13 17:28:16,733 - Epoch: [21][ 1100/ 1207] Overall Loss 0.404577 Objective Loss 0.404577 LR 0.001000 Time 0.019851 -2023-02-13 17:28:16,933 - Epoch: [21][ 1110/ 1207] Overall Loss 0.404618 Objective Loss 0.404618 LR 0.001000 Time 0.019852 -2023-02-13 17:28:17,138 - Epoch: [21][ 1120/ 1207] Overall Loss 0.404970 Objective Loss 0.404970 LR 0.001000 Time 0.019857 -2023-02-13 17:28:17,338 - Epoch: [21][ 1130/ 1207] Overall Loss 0.405073 Objective Loss 0.405073 LR 0.001000 Time 0.019858 -2023-02-13 17:28:17,542 - Epoch: [21][ 1140/ 1207] Overall Loss 0.405029 Objective Loss 0.405029 LR 0.001000 Time 0.019863 -2023-02-13 17:28:17,742 - Epoch: [21][ 1150/ 1207] Overall Loss 0.404782 Objective Loss 0.404782 LR 0.001000 Time 0.019864 -2023-02-13 17:28:17,948 - Epoch: [21][ 1160/ 1207] Overall Loss 0.404634 Objective Loss 0.404634 LR 0.001000 Time 0.019869 -2023-02-13 17:28:18,147 - Epoch: [21][ 1170/ 1207] Overall Loss 0.404741 Objective Loss 0.404741 LR 0.001000 Time 0.019869 -2023-02-13 17:28:18,351 - Epoch: [21][ 1180/ 1207] Overall Loss 0.404695 Objective Loss 0.404695 LR 0.001000 Time 0.019874 -2023-02-13 17:28:18,551 - Epoch: [21][ 1190/ 1207] Overall Loss 0.404860 Objective Loss 0.404860 LR 0.001000 Time 0.019874 -2023-02-13 17:28:18,802 - Epoch: [21][ 1200/ 1207] Overall Loss 0.405125 Objective Loss 0.405125 LR 0.001000 Time 0.019917 -2023-02-13 17:28:18,919 - Epoch: [21][ 1207/ 1207] Overall Loss 0.404957 Objective Loss 0.404957 Top1 82.926829 Top5 95.121951 LR 0.001000 Time 0.019899 -2023-02-13 17:28:18,999 - --- validate (epoch=21)----------- -2023-02-13 17:28:18,999 - 34311 samples (256 per mini-batch) -2023-02-13 17:28:19,495 - Epoch: [21][ 10/ 135] Loss 0.453275 Top1 78.515625 Top5 96.171875 -2023-02-13 17:28:19,628 - Epoch: [21][ 20/ 135] Loss 0.418159 Top1 79.082031 Top5 96.640625 -2023-02-13 17:28:19,754 - Epoch: [21][ 30/ 135] Loss 0.412002 Top1 79.231771 Top5 96.796875 -2023-02-13 17:28:19,899 - Epoch: [21][ 40/ 135] Loss 0.415001 Top1 79.443359 Top5 96.835938 -2023-02-13 17:28:20,036 - Epoch: [21][ 50/ 135] Loss 0.417007 Top1 79.414062 Top5 96.789062 -2023-02-13 17:28:20,172 - Epoch: [21][ 60/ 135] Loss 0.413713 Top1 79.433594 Top5 96.861979 -2023-02-13 17:28:20,316 - Epoch: [21][ 70/ 135] Loss 0.412724 Top1 79.436384 Top5 96.891741 -2023-02-13 17:28:20,452 - Epoch: [21][ 80/ 135] Loss 0.418445 Top1 79.287109 Top5 96.811523 -2023-02-13 17:28:20,595 - Epoch: [21][ 90/ 135] Loss 0.415721 Top1 79.318576 Top5 96.736111 -2023-02-13 17:28:20,731 - Epoch: [21][ 100/ 135] Loss 0.414375 Top1 79.421875 Top5 96.742188 -2023-02-13 17:28:20,877 - Epoch: [21][ 110/ 135] Loss 0.414269 Top1 79.382102 Top5 96.722301 -2023-02-13 17:28:21,006 - Epoch: [21][ 120/ 135] Loss 0.413475 Top1 79.329427 Top5 96.705729 -2023-02-13 17:28:21,138 - Epoch: [21][ 130/ 135] Loss 0.412764 Top1 79.332933 Top5 96.703726 -2023-02-13 17:28:21,187 - Epoch: [21][ 135/ 135] Loss 0.417901 Top1 79.286526 Top5 96.694937 -2023-02-13 17:28:21,257 - ==> Top1: 79.287 Top5: 96.695 Loss: 0.418 - -2023-02-13 17:28:21,258 - ==> Confusion: -[[ 774 3 13 1 7 3 0 3 10 120 1 6 0 4 3 2 1 3 3 0 10] - [ 4 914 2 3 3 41 1 26 4 4 3 3 3 0 1 0 2 0 7 7 5] - [ 10 5 939 13 1 3 19 22 0 2 9 3 1 5 2 3 0 2 9 4 6] - [ 4 3 30 855 2 4 1 2 4 4 20 0 11 1 22 0 2 7 37 0 7] - [ 19 23 4 0 952 19 1 1 1 10 1 5 2 1 5 6 5 0 1 3 7] - [ 4 26 3 4 2 946 2 33 1 10 5 11 4 5 1 1 0 3 1 3 5] - [ 4 1 28 4 0 8 1018 7 0 3 4 2 3 0 0 3 0 2 1 8 3] - [ 1 12 8 2 1 23 2 897 1 1 7 5 4 1 1 1 0 1 36 14 6] - [ 14 1 1 1 0 0 0 0 901 44 8 4 1 9 12 0 1 2 10 0 0] - [ 51 2 1 1 1 3 0 2 40 875 1 3 0 21 4 1 0 1 0 1 4] - [ 1 2 6 5 0 1 4 4 24 3 970 3 0 8 0 1 1 0 15 1 2] - [ 4 2 1 0 1 15 1 3 1 3 0 886 57 1 0 5 0 13 4 6 2] - [ 3 2 2 3 1 7 0 0 2 0 0 30 874 0 3 1 2 19 1 4 5] - [ 9 2 3 0 8 27 0 2 18 24 6 13 10 887 3 1 3 1 2 2 3] - [ 11 3 2 16 8 1 0 1 38 8 3 3 7 1 944 2 3 3 25 0 13] - [ 1 3 10 1 6 5 10 2 0 1 0 5 12 2 0 939 6 27 0 7 9] - [ 3 15 3 4 9 6 2 0 7 0 2 8 7 3 0 17 950 1 3 6 15] - [ 6 3 2 8 0 2 0 1 0 1 0 11 38 1 1 8 0 963 0 2 4] - [ 6 3 7 6 0 1 0 37 3 1 7 1 8 0 7 0 0 1 996 1 1] - [ 1 3 3 0 1 12 4 18 1 0 0 21 8 2 0 4 2 4 0 1056 8] - [ 208 383 372 119 95 298 133 272 185 175 272 149 482 388 146 88 215 115 277 394 8668]] - -2023-02-13 17:28:21,259 - ==> Best [Top1: 79.668 Top5: 96.785 Sparsity:0.00 Params: 148928 on epoch: 16] -2023-02-13 17:28:21,259 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:28:21,265 - - -2023-02-13 17:28:21,265 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:28:22,150 - Epoch: [22][ 10/ 1207] Overall Loss 0.385411 Objective Loss 0.385411 LR 0.001000 Time 0.088424 -2023-02-13 17:28:22,346 - Epoch: [22][ 20/ 1207] Overall Loss 0.386013 Objective Loss 0.386013 LR 0.001000 Time 0.053961 -2023-02-13 17:28:22,539 - Epoch: [22][ 30/ 1207] Overall Loss 0.388963 Objective Loss 0.388963 LR 0.001000 Time 0.042413 -2023-02-13 17:28:22,734 - Epoch: [22][ 40/ 1207] Overall Loss 0.390181 Objective Loss 0.390181 LR 0.001000 Time 0.036666 -2023-02-13 17:28:22,928 - Epoch: [22][ 50/ 1207] Overall Loss 0.390281 Objective Loss 0.390281 LR 0.001000 Time 0.033206 -2023-02-13 17:28:23,123 - Epoch: [22][ 60/ 1207] Overall Loss 0.395185 Objective Loss 0.395185 LR 0.001000 Time 0.030927 -2023-02-13 17:28:23,317 - Epoch: [22][ 70/ 1207] Overall Loss 0.395133 Objective Loss 0.395133 LR 0.001000 Time 0.029271 -2023-02-13 17:28:23,512 - Epoch: [22][ 80/ 1207] Overall Loss 0.396422 Objective Loss 0.396422 LR 0.001000 Time 0.028042 -2023-02-13 17:28:23,706 - Epoch: [22][ 90/ 1207] Overall Loss 0.396401 Objective Loss 0.396401 LR 0.001000 Time 0.027073 -2023-02-13 17:28:23,901 - Epoch: [22][ 100/ 1207] Overall Loss 0.396575 Objective Loss 0.396575 LR 0.001000 Time 0.026316 -2023-02-13 17:28:24,094 - Epoch: [22][ 110/ 1207] Overall Loss 0.394256 Objective Loss 0.394256 LR 0.001000 Time 0.025679 -2023-02-13 17:28:24,290 - Epoch: [22][ 120/ 1207] Overall Loss 0.392581 Objective Loss 0.392581 LR 0.001000 Time 0.025163 -2023-02-13 17:28:24,483 - Epoch: [22][ 130/ 1207] Overall Loss 0.391443 Objective Loss 0.391443 LR 0.001000 Time 0.024714 -2023-02-13 17:28:24,679 - Epoch: [22][ 140/ 1207] Overall Loss 0.389462 Objective Loss 0.389462 LR 0.001000 Time 0.024342 -2023-02-13 17:28:24,872 - Epoch: [22][ 150/ 1207] Overall Loss 0.388779 Objective Loss 0.388779 LR 0.001000 Time 0.024007 -2023-02-13 17:28:25,068 - Epoch: [22][ 160/ 1207] Overall Loss 0.388214 Objective Loss 0.388214 LR 0.001000 Time 0.023727 -2023-02-13 17:28:25,261 - Epoch: [22][ 170/ 1207] Overall Loss 0.388294 Objective Loss 0.388294 LR 0.001000 Time 0.023466 -2023-02-13 17:28:25,456 - Epoch: [22][ 180/ 1207] Overall Loss 0.388216 Objective Loss 0.388216 LR 0.001000 Time 0.023241 -2023-02-13 17:28:25,649 - Epoch: [22][ 190/ 1207] Overall Loss 0.387747 Objective Loss 0.387747 LR 0.001000 Time 0.023033 -2023-02-13 17:28:25,846 - Epoch: [22][ 200/ 1207] Overall Loss 0.387139 Objective Loss 0.387139 LR 0.001000 Time 0.022863 -2023-02-13 17:28:26,039 - Epoch: [22][ 210/ 1207] Overall Loss 0.387300 Objective Loss 0.387300 LR 0.001000 Time 0.022692 -2023-02-13 17:28:26,234 - Epoch: [22][ 220/ 1207] Overall Loss 0.387674 Objective Loss 0.387674 LR 0.001000 Time 0.022548 -2023-02-13 17:28:26,429 - Epoch: [22][ 230/ 1207] Overall Loss 0.386720 Objective Loss 0.386720 LR 0.001000 Time 0.022411 -2023-02-13 17:28:26,625 - Epoch: [22][ 240/ 1207] Overall Loss 0.388020 Objective Loss 0.388020 LR 0.001000 Time 0.022293 -2023-02-13 17:28:26,820 - Epoch: [22][ 250/ 1207] Overall Loss 0.388245 Objective Loss 0.388245 LR 0.001000 Time 0.022180 -2023-02-13 17:28:27,016 - Epoch: [22][ 260/ 1207] Overall Loss 0.388125 Objective Loss 0.388125 LR 0.001000 Time 0.022077 -2023-02-13 17:28:27,209 - Epoch: [22][ 270/ 1207] Overall Loss 0.388407 Objective Loss 0.388407 LR 0.001000 Time 0.021976 -2023-02-13 17:28:27,404 - Epoch: [22][ 280/ 1207] Overall Loss 0.387613 Objective Loss 0.387613 LR 0.001000 Time 0.021885 -2023-02-13 17:28:27,597 - Epoch: [22][ 290/ 1207] Overall Loss 0.388043 Objective Loss 0.388043 LR 0.001000 Time 0.021796 -2023-02-13 17:28:27,794 - Epoch: [22][ 300/ 1207] Overall Loss 0.388087 Objective Loss 0.388087 LR 0.001000 Time 0.021722 -2023-02-13 17:28:27,989 - Epoch: [22][ 310/ 1207] Overall Loss 0.388682 Objective Loss 0.388682 LR 0.001000 Time 0.021652 -2023-02-13 17:28:28,185 - Epoch: [22][ 320/ 1207] Overall Loss 0.389279 Objective Loss 0.389279 LR 0.001000 Time 0.021586 -2023-02-13 17:28:28,380 - Epoch: [22][ 330/ 1207] Overall Loss 0.389728 Objective Loss 0.389728 LR 0.001000 Time 0.021520 -2023-02-13 17:28:28,575 - Epoch: [22][ 340/ 1207] Overall Loss 0.391221 Objective Loss 0.391221 LR 0.001000 Time 0.021461 -2023-02-13 17:28:28,770 - Epoch: [22][ 350/ 1207] Overall Loss 0.390795 Objective Loss 0.390795 LR 0.001000 Time 0.021402 -2023-02-13 17:28:28,966 - Epoch: [22][ 360/ 1207] Overall Loss 0.391655 Objective Loss 0.391655 LR 0.001000 Time 0.021351 -2023-02-13 17:28:29,160 - Epoch: [22][ 370/ 1207] Overall Loss 0.392233 Objective Loss 0.392233 LR 0.001000 Time 0.021298 -2023-02-13 17:28:29,356 - Epoch: [22][ 380/ 1207] Overall Loss 0.392836 Objective Loss 0.392836 LR 0.001000 Time 0.021252 -2023-02-13 17:28:29,550 - Epoch: [22][ 390/ 1207] Overall Loss 0.393528 Objective Loss 0.393528 LR 0.001000 Time 0.021203 -2023-02-13 17:28:29,746 - Epoch: [22][ 400/ 1207] Overall Loss 0.394022 Objective Loss 0.394022 LR 0.001000 Time 0.021162 -2023-02-13 17:28:29,940 - Epoch: [22][ 410/ 1207] Overall Loss 0.394018 Objective Loss 0.394018 LR 0.001000 Time 0.021119 -2023-02-13 17:28:30,136 - Epoch: [22][ 420/ 1207] Overall Loss 0.394970 Objective Loss 0.394970 LR 0.001000 Time 0.021082 -2023-02-13 17:28:30,331 - Epoch: [22][ 430/ 1207] Overall Loss 0.394361 Objective Loss 0.394361 LR 0.001000 Time 0.021044 -2023-02-13 17:28:30,527 - Epoch: [22][ 440/ 1207] Overall Loss 0.395379 Objective Loss 0.395379 LR 0.001000 Time 0.021010 -2023-02-13 17:28:30,720 - Epoch: [22][ 450/ 1207] Overall Loss 0.395430 Objective Loss 0.395430 LR 0.001000 Time 0.020972 -2023-02-13 17:28:30,918 - Epoch: [22][ 460/ 1207] Overall Loss 0.394552 Objective Loss 0.394552 LR 0.001000 Time 0.020946 -2023-02-13 17:28:31,113 - Epoch: [22][ 470/ 1207] Overall Loss 0.394528 Objective Loss 0.394528 LR 0.001000 Time 0.020914 -2023-02-13 17:28:31,309 - Epoch: [22][ 480/ 1207] Overall Loss 0.394809 Objective Loss 0.394809 LR 0.001000 Time 0.020885 -2023-02-13 17:28:31,503 - Epoch: [22][ 490/ 1207] Overall Loss 0.394674 Objective Loss 0.394674 LR 0.001000 Time 0.020855 -2023-02-13 17:28:31,699 - Epoch: [22][ 500/ 1207] Overall Loss 0.394594 Objective Loss 0.394594 LR 0.001000 Time 0.020828 -2023-02-13 17:28:31,893 - Epoch: [22][ 510/ 1207] Overall Loss 0.393257 Objective Loss 0.393257 LR 0.001000 Time 0.020801 -2023-02-13 17:28:32,089 - Epoch: [22][ 520/ 1207] Overall Loss 0.393394 Objective Loss 0.393394 LR 0.001000 Time 0.020776 -2023-02-13 17:28:32,283 - Epoch: [22][ 530/ 1207] Overall Loss 0.393285 Objective Loss 0.393285 LR 0.001000 Time 0.020750 -2023-02-13 17:28:32,479 - Epoch: [22][ 540/ 1207] Overall Loss 0.392600 Objective Loss 0.392600 LR 0.001000 Time 0.020728 -2023-02-13 17:28:32,674 - Epoch: [22][ 550/ 1207] Overall Loss 0.392637 Objective Loss 0.392637 LR 0.001000 Time 0.020704 -2023-02-13 17:28:32,870 - Epoch: [22][ 560/ 1207] Overall Loss 0.393225 Objective Loss 0.393225 LR 0.001000 Time 0.020684 -2023-02-13 17:28:33,065 - Epoch: [22][ 570/ 1207] Overall Loss 0.393026 Objective Loss 0.393026 LR 0.001000 Time 0.020662 -2023-02-13 17:28:33,259 - Epoch: [22][ 580/ 1207] Overall Loss 0.392567 Objective Loss 0.392567 LR 0.001000 Time 0.020641 -2023-02-13 17:28:33,453 - Epoch: [22][ 590/ 1207] Overall Loss 0.392907 Objective Loss 0.392907 LR 0.001000 Time 0.020618 -2023-02-13 17:28:33,649 - Epoch: [22][ 600/ 1207] Overall Loss 0.392780 Objective Loss 0.392780 LR 0.001000 Time 0.020601 -2023-02-13 17:28:33,844 - Epoch: [22][ 610/ 1207] Overall Loss 0.392278 Objective Loss 0.392278 LR 0.001000 Time 0.020582 -2023-02-13 17:28:34,041 - Epoch: [22][ 620/ 1207] Overall Loss 0.392524 Objective Loss 0.392524 LR 0.001000 Time 0.020567 -2023-02-13 17:28:34,235 - Epoch: [22][ 630/ 1207] Overall Loss 0.392683 Objective Loss 0.392683 LR 0.001000 Time 0.020549 -2023-02-13 17:28:34,431 - Epoch: [22][ 640/ 1207] Overall Loss 0.393638 Objective Loss 0.393638 LR 0.001000 Time 0.020532 -2023-02-13 17:28:34,625 - Epoch: [22][ 650/ 1207] Overall Loss 0.393190 Objective Loss 0.393190 LR 0.001000 Time 0.020514 -2023-02-13 17:28:34,821 - Epoch: [22][ 660/ 1207] Overall Loss 0.393142 Objective Loss 0.393142 LR 0.001000 Time 0.020500 -2023-02-13 17:28:35,015 - Epoch: [22][ 670/ 1207] Overall Loss 0.393052 Objective Loss 0.393052 LR 0.001000 Time 0.020484 -2023-02-13 17:28:35,211 - Epoch: [22][ 680/ 1207] Overall Loss 0.393335 Objective Loss 0.393335 LR 0.001000 Time 0.020470 -2023-02-13 17:28:35,405 - Epoch: [22][ 690/ 1207] Overall Loss 0.393719 Objective Loss 0.393719 LR 0.001000 Time 0.020454 -2023-02-13 17:28:35,601 - Epoch: [22][ 700/ 1207] Overall Loss 0.393637 Objective Loss 0.393637 LR 0.001000 Time 0.020440 -2023-02-13 17:28:35,795 - Epoch: [22][ 710/ 1207] Overall Loss 0.393988 Objective Loss 0.393988 LR 0.001000 Time 0.020426 -2023-02-13 17:28:35,992 - Epoch: [22][ 720/ 1207] Overall Loss 0.394196 Objective Loss 0.394196 LR 0.001000 Time 0.020415 -2023-02-13 17:28:36,185 - Epoch: [22][ 730/ 1207] Overall Loss 0.393973 Objective Loss 0.393973 LR 0.001000 Time 0.020399 -2023-02-13 17:28:36,381 - Epoch: [22][ 740/ 1207] Overall Loss 0.393949 Objective Loss 0.393949 LR 0.001000 Time 0.020388 -2023-02-13 17:28:36,575 - Epoch: [22][ 750/ 1207] Overall Loss 0.394102 Objective Loss 0.394102 LR 0.001000 Time 0.020375 -2023-02-13 17:28:36,771 - Epoch: [22][ 760/ 1207] Overall Loss 0.394260 Objective Loss 0.394260 LR 0.001000 Time 0.020363 -2023-02-13 17:28:36,965 - Epoch: [22][ 770/ 1207] Overall Loss 0.394252 Objective Loss 0.394252 LR 0.001000 Time 0.020351 -2023-02-13 17:28:37,161 - Epoch: [22][ 780/ 1207] Overall Loss 0.394383 Objective Loss 0.394383 LR 0.001000 Time 0.020341 -2023-02-13 17:28:37,356 - Epoch: [22][ 790/ 1207] Overall Loss 0.394483 Objective Loss 0.394483 LR 0.001000 Time 0.020329 -2023-02-13 17:28:37,551 - Epoch: [22][ 800/ 1207] Overall Loss 0.394620 Objective Loss 0.394620 LR 0.001000 Time 0.020319 -2023-02-13 17:28:37,745 - Epoch: [22][ 810/ 1207] Overall Loss 0.394385 Objective Loss 0.394385 LR 0.001000 Time 0.020307 -2023-02-13 17:28:37,941 - Epoch: [22][ 820/ 1207] Overall Loss 0.394510 Objective Loss 0.394510 LR 0.001000 Time 0.020298 -2023-02-13 17:28:38,136 - Epoch: [22][ 830/ 1207] Overall Loss 0.394158 Objective Loss 0.394158 LR 0.001000 Time 0.020287 -2023-02-13 17:28:38,330 - Epoch: [22][ 840/ 1207] Overall Loss 0.394081 Objective Loss 0.394081 LR 0.001000 Time 0.020277 -2023-02-13 17:28:38,525 - Epoch: [22][ 850/ 1207] Overall Loss 0.394441 Objective Loss 0.394441 LR 0.001000 Time 0.020267 -2023-02-13 17:28:38,721 - Epoch: [22][ 860/ 1207] Overall Loss 0.394024 Objective Loss 0.394024 LR 0.001000 Time 0.020259 -2023-02-13 17:28:38,916 - Epoch: [22][ 870/ 1207] Overall Loss 0.393914 Objective Loss 0.393914 LR 0.001000 Time 0.020250 -2023-02-13 17:28:39,112 - Epoch: [22][ 880/ 1207] Overall Loss 0.393902 Objective Loss 0.393902 LR 0.001000 Time 0.020242 -2023-02-13 17:28:39,306 - Epoch: [22][ 890/ 1207] Overall Loss 0.393665 Objective Loss 0.393665 LR 0.001000 Time 0.020232 -2023-02-13 17:28:39,501 - Epoch: [22][ 900/ 1207] Overall Loss 0.393936 Objective Loss 0.393936 LR 0.001000 Time 0.020223 -2023-02-13 17:28:39,696 - Epoch: [22][ 910/ 1207] Overall Loss 0.394177 Objective Loss 0.394177 LR 0.001000 Time 0.020215 -2023-02-13 17:28:39,894 - Epoch: [22][ 920/ 1207] Overall Loss 0.394106 Objective Loss 0.394106 LR 0.001000 Time 0.020210 -2023-02-13 17:28:40,089 - Epoch: [22][ 930/ 1207] Overall Loss 0.394440 Objective Loss 0.394440 LR 0.001000 Time 0.020202 -2023-02-13 17:28:40,287 - Epoch: [22][ 940/ 1207] Overall Loss 0.394157 Objective Loss 0.394157 LR 0.001000 Time 0.020198 -2023-02-13 17:28:40,483 - Epoch: [22][ 950/ 1207] Overall Loss 0.393949 Objective Loss 0.393949 LR 0.001000 Time 0.020190 -2023-02-13 17:28:40,681 - Epoch: [22][ 960/ 1207] Overall Loss 0.393823 Objective Loss 0.393823 LR 0.001000 Time 0.020186 -2023-02-13 17:28:40,879 - Epoch: [22][ 970/ 1207] Overall Loss 0.394036 Objective Loss 0.394036 LR 0.001000 Time 0.020182 -2023-02-13 17:28:41,078 - Epoch: [22][ 980/ 1207] Overall Loss 0.394163 Objective Loss 0.394163 LR 0.001000 Time 0.020179 -2023-02-13 17:28:41,273 - Epoch: [22][ 990/ 1207] Overall Loss 0.394270 Objective Loss 0.394270 LR 0.001000 Time 0.020171 -2023-02-13 17:28:41,472 - Epoch: [22][ 1000/ 1207] Overall Loss 0.394339 Objective Loss 0.394339 LR 0.001000 Time 0.020168 -2023-02-13 17:28:41,668 - Epoch: [22][ 1010/ 1207] Overall Loss 0.394384 Objective Loss 0.394384 LR 0.001000 Time 0.020162 -2023-02-13 17:28:41,867 - Epoch: [22][ 1020/ 1207] Overall Loss 0.394236 Objective Loss 0.394236 LR 0.001000 Time 0.020159 -2023-02-13 17:28:42,064 - Epoch: [22][ 1030/ 1207] Overall Loss 0.394116 Objective Loss 0.394116 LR 0.001000 Time 0.020154 -2023-02-13 17:28:42,261 - Epoch: [22][ 1040/ 1207] Overall Loss 0.394104 Objective Loss 0.394104 LR 0.001000 Time 0.020150 -2023-02-13 17:28:42,457 - Epoch: [22][ 1050/ 1207] Overall Loss 0.394113 Objective Loss 0.394113 LR 0.001000 Time 0.020144 -2023-02-13 17:28:42,656 - Epoch: [22][ 1060/ 1207] Overall Loss 0.394363 Objective Loss 0.394363 LR 0.001000 Time 0.020141 -2023-02-13 17:28:42,851 - Epoch: [22][ 1070/ 1207] Overall Loss 0.394463 Objective Loss 0.394463 LR 0.001000 Time 0.020135 -2023-02-13 17:28:43,050 - Epoch: [22][ 1080/ 1207] Overall Loss 0.394759 Objective Loss 0.394759 LR 0.001000 Time 0.020133 -2023-02-13 17:28:43,246 - Epoch: [22][ 1090/ 1207] Overall Loss 0.394956 Objective Loss 0.394956 LR 0.001000 Time 0.020127 -2023-02-13 17:28:43,444 - Epoch: [22][ 1100/ 1207] Overall Loss 0.395164 Objective Loss 0.395164 LR 0.001000 Time 0.020124 -2023-02-13 17:28:43,640 - Epoch: [22][ 1110/ 1207] Overall Loss 0.395149 Objective Loss 0.395149 LR 0.001000 Time 0.020119 -2023-02-13 17:28:43,839 - Epoch: [22][ 1120/ 1207] Overall Loss 0.395245 Objective Loss 0.395245 LR 0.001000 Time 0.020116 -2023-02-13 17:28:44,036 - Epoch: [22][ 1130/ 1207] Overall Loss 0.395212 Objective Loss 0.395212 LR 0.001000 Time 0.020113 -2023-02-13 17:28:44,234 - Epoch: [22][ 1140/ 1207] Overall Loss 0.394716 Objective Loss 0.394716 LR 0.001000 Time 0.020109 -2023-02-13 17:28:44,430 - Epoch: [22][ 1150/ 1207] Overall Loss 0.394679 Objective Loss 0.394679 LR 0.001000 Time 0.020105 -2023-02-13 17:28:44,628 - Epoch: [22][ 1160/ 1207] Overall Loss 0.394705 Objective Loss 0.394705 LR 0.001000 Time 0.020102 -2023-02-13 17:28:44,825 - Epoch: [22][ 1170/ 1207] Overall Loss 0.394491 Objective Loss 0.394491 LR 0.001000 Time 0.020098 -2023-02-13 17:28:45,023 - Epoch: [22][ 1180/ 1207] Overall Loss 0.394821 Objective Loss 0.394821 LR 0.001000 Time 0.020096 -2023-02-13 17:28:45,219 - Epoch: [22][ 1190/ 1207] Overall Loss 0.394827 Objective Loss 0.394827 LR 0.001000 Time 0.020091 -2023-02-13 17:28:45,467 - Epoch: [22][ 1200/ 1207] Overall Loss 0.394967 Objective Loss 0.394967 LR 0.001000 Time 0.020130 -2023-02-13 17:28:45,583 - Epoch: [22][ 1207/ 1207] Overall Loss 0.394840 Objective Loss 0.394840 Top1 81.402439 Top5 96.341463 LR 0.001000 Time 0.020109 -2023-02-13 17:28:45,656 - --- validate (epoch=22)----------- -2023-02-13 17:28:45,656 - 34311 samples (256 per mini-batch) -2023-02-13 17:28:46,057 - Epoch: [22][ 10/ 135] Loss 0.390369 Top1 80.000000 Top5 96.796875 -2023-02-13 17:28:46,188 - Epoch: [22][ 20/ 135] Loss 0.375777 Top1 80.781250 Top5 96.914062 -2023-02-13 17:28:46,318 - Epoch: [22][ 30/ 135] Loss 0.384732 Top1 80.742188 Top5 97.031250 -2023-02-13 17:28:46,446 - Epoch: [22][ 40/ 135] Loss 0.397803 Top1 80.117188 Top5 96.777344 -2023-02-13 17:28:46,577 - Epoch: [22][ 50/ 135] Loss 0.396288 Top1 80.234375 Top5 96.765625 -2023-02-13 17:28:46,706 - Epoch: [22][ 60/ 135] Loss 0.401335 Top1 80.175781 Top5 96.848958 -2023-02-13 17:28:46,835 - Epoch: [22][ 70/ 135] Loss 0.400386 Top1 80.044643 Top5 96.768973 -2023-02-13 17:28:46,962 - Epoch: [22][ 80/ 135] Loss 0.401741 Top1 80.039062 Top5 96.718750 -2023-02-13 17:28:47,092 - Epoch: [22][ 90/ 135] Loss 0.403017 Top1 79.978299 Top5 96.749132 -2023-02-13 17:28:47,220 - Epoch: [22][ 100/ 135] Loss 0.400689 Top1 79.968750 Top5 96.750000 -2023-02-13 17:28:47,349 - Epoch: [22][ 110/ 135] Loss 0.402082 Top1 79.872159 Top5 96.789773 -2023-02-13 17:28:47,479 - Epoch: [22][ 120/ 135] Loss 0.403511 Top1 79.759115 Top5 96.722005 -2023-02-13 17:28:47,611 - Epoch: [22][ 130/ 135] Loss 0.402853 Top1 79.723558 Top5 96.754808 -2023-02-13 17:28:47,657 - Epoch: [22][ 135/ 135] Loss 0.404073 Top1 79.717875 Top5 96.756142 -2023-02-13 17:28:47,724 - ==> Top1: 79.718 Top5: 96.756 Loss: 0.404 - -2023-02-13 17:28:47,725 - ==> Confusion: -[[ 855 0 6 2 7 2 0 0 3 53 1 5 1 5 7 4 1 3 1 1 10] - [ 6 867 4 8 15 59 4 26 4 2 1 0 2 1 5 1 6 1 10 3 8] - [ 12 3 965 10 0 2 15 11 0 0 3 0 1 4 3 10 0 5 3 0 11] - [ 7 1 25 895 1 1 2 2 1 3 9 1 4 1 19 4 3 11 14 2 10] - [ 34 8 2 0 971 13 1 1 0 5 0 3 0 3 4 6 5 1 3 2 4] - [ 2 20 4 6 5 950 2 17 0 9 1 3 5 15 1 5 5 2 3 8 7] - [ 6 2 26 4 0 4 1015 5 0 1 5 0 2 0 0 11 2 1 1 10 4] - [ 4 5 20 4 2 30 4 896 2 1 5 2 1 3 1 2 0 1 28 9 4] - [ 35 3 0 2 1 0 0 0 850 52 9 1 0 9 35 0 1 5 4 0 2] - [ 106 2 4 0 3 1 0 2 17 848 0 1 0 13 4 1 0 4 0 1 5] - [ 5 4 6 14 2 2 4 2 20 2 950 0 1 11 5 0 1 0 15 3 4] - [ 7 2 3 0 2 25 2 3 3 2 0 870 29 8 3 14 3 14 1 14 0] - [ 5 1 1 6 2 4 0 0 3 1 0 32 828 1 3 12 4 45 0 3 8] - [ 14 1 5 0 6 17 1 2 23 33 5 4 0 890 6 3 5 1 2 2 4] - [ 19 3 5 20 4 1 0 0 15 8 3 1 3 2 978 0 4 9 10 0 7] - [ 9 1 4 1 5 2 7 1 0 0 0 7 6 5 0 966 6 16 0 5 5] - [ 7 2 1 0 10 6 1 0 2 2 1 2 2 2 0 27 974 6 0 3 13] - [ 9 0 2 4 1 0 0 0 0 2 0 8 8 1 0 20 1 985 1 2 7] - [ 4 2 15 23 0 4 1 30 5 2 8 2 7 0 20 1 1 1 954 1 5] - [ 0 1 1 1 2 5 13 16 0 0 0 18 4 5 0 8 8 5 0 1056 5] - [ 337 241 368 169 180 246 97 178 118 149 192 135 393 364 238 178 305 175 201 381 8789]] - -2023-02-13 17:28:47,726 - ==> Best [Top1: 79.718 Top5: 96.756 Sparsity:0.00 Params: 148928 on epoch: 22] -2023-02-13 17:28:47,727 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:28:47,733 - - -2023-02-13 17:28:47,733 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:28:48,688 - Epoch: [23][ 10/ 1207] Overall Loss 0.367724 Objective Loss 0.367724 LR 0.001000 Time 0.095422 -2023-02-13 17:28:48,888 - Epoch: [23][ 20/ 1207] Overall Loss 0.373509 Objective Loss 0.373509 LR 0.001000 Time 0.057668 -2023-02-13 17:28:49,076 - Epoch: [23][ 30/ 1207] Overall Loss 0.367777 Objective Loss 0.367777 LR 0.001000 Time 0.044699 -2023-02-13 17:28:49,263 - Epoch: [23][ 40/ 1207] Overall Loss 0.371151 Objective Loss 0.371151 LR 0.001000 Time 0.038197 -2023-02-13 17:28:49,451 - Epoch: [23][ 50/ 1207] Overall Loss 0.367939 Objective Loss 0.367939 LR 0.001000 Time 0.034316 -2023-02-13 17:28:49,638 - Epoch: [23][ 60/ 1207] Overall Loss 0.375949 Objective Loss 0.375949 LR 0.001000 Time 0.031709 -2023-02-13 17:28:49,825 - Epoch: [23][ 70/ 1207] Overall Loss 0.375938 Objective Loss 0.375938 LR 0.001000 Time 0.029839 -2023-02-13 17:28:50,013 - Epoch: [23][ 80/ 1207] Overall Loss 0.376606 Objective Loss 0.376606 LR 0.001000 Time 0.028450 -2023-02-13 17:28:50,199 - Epoch: [23][ 90/ 1207] Overall Loss 0.379220 Objective Loss 0.379220 LR 0.001000 Time 0.027358 -2023-02-13 17:28:50,386 - Epoch: [23][ 100/ 1207] Overall Loss 0.378370 Objective Loss 0.378370 LR 0.001000 Time 0.026488 -2023-02-13 17:28:50,573 - Epoch: [23][ 110/ 1207] Overall Loss 0.383674 Objective Loss 0.383674 LR 0.001000 Time 0.025777 -2023-02-13 17:28:50,761 - Epoch: [23][ 120/ 1207] Overall Loss 0.385024 Objective Loss 0.385024 LR 0.001000 Time 0.025190 -2023-02-13 17:28:50,949 - Epoch: [23][ 130/ 1207] Overall Loss 0.387866 Objective Loss 0.387866 LR 0.001000 Time 0.024694 -2023-02-13 17:28:51,137 - Epoch: [23][ 140/ 1207] Overall Loss 0.388572 Objective Loss 0.388572 LR 0.001000 Time 0.024275 -2023-02-13 17:28:51,325 - Epoch: [23][ 150/ 1207] Overall Loss 0.387062 Objective Loss 0.387062 LR 0.001000 Time 0.023906 -2023-02-13 17:28:51,514 - Epoch: [23][ 160/ 1207] Overall Loss 0.390313 Objective Loss 0.390313 LR 0.001000 Time 0.023590 -2023-02-13 17:28:51,703 - Epoch: [23][ 170/ 1207] Overall Loss 0.390465 Objective Loss 0.390465 LR 0.001000 Time 0.023311 -2023-02-13 17:28:51,893 - Epoch: [23][ 180/ 1207] Overall Loss 0.390847 Objective Loss 0.390847 LR 0.001000 Time 0.023069 -2023-02-13 17:28:52,082 - Epoch: [23][ 190/ 1207] Overall Loss 0.390843 Objective Loss 0.390843 LR 0.001000 Time 0.022847 -2023-02-13 17:28:52,269 - Epoch: [23][ 200/ 1207] Overall Loss 0.390332 Objective Loss 0.390332 LR 0.001000 Time 0.022641 -2023-02-13 17:28:52,457 - Epoch: [23][ 210/ 1207] Overall Loss 0.391190 Objective Loss 0.391190 LR 0.001000 Time 0.022458 -2023-02-13 17:28:52,646 - Epoch: [23][ 220/ 1207] Overall Loss 0.391690 Objective Loss 0.391690 LR 0.001000 Time 0.022292 -2023-02-13 17:28:52,834 - Epoch: [23][ 230/ 1207] Overall Loss 0.391789 Objective Loss 0.391789 LR 0.001000 Time 0.022138 -2023-02-13 17:28:53,022 - Epoch: [23][ 240/ 1207] Overall Loss 0.391097 Objective Loss 0.391097 LR 0.001000 Time 0.022000 -2023-02-13 17:28:53,211 - Epoch: [23][ 250/ 1207] Overall Loss 0.390327 Objective Loss 0.390327 LR 0.001000 Time 0.021872 -2023-02-13 17:28:53,399 - Epoch: [23][ 260/ 1207] Overall Loss 0.392194 Objective Loss 0.392194 LR 0.001000 Time 0.021753 -2023-02-13 17:28:53,588 - Epoch: [23][ 270/ 1207] Overall Loss 0.392443 Objective Loss 0.392443 LR 0.001000 Time 0.021645 -2023-02-13 17:28:53,776 - Epoch: [23][ 280/ 1207] Overall Loss 0.391791 Objective Loss 0.391791 LR 0.001000 Time 0.021542 -2023-02-13 17:28:53,965 - Epoch: [23][ 290/ 1207] Overall Loss 0.391042 Objective Loss 0.391042 LR 0.001000 Time 0.021451 -2023-02-13 17:28:54,155 - Epoch: [23][ 300/ 1207] Overall Loss 0.391660 Objective Loss 0.391660 LR 0.001000 Time 0.021366 -2023-02-13 17:28:54,342 - Epoch: [23][ 310/ 1207] Overall Loss 0.391434 Objective Loss 0.391434 LR 0.001000 Time 0.021282 -2023-02-13 17:28:54,531 - Epoch: [23][ 320/ 1207] Overall Loss 0.391691 Objective Loss 0.391691 LR 0.001000 Time 0.021205 -2023-02-13 17:28:54,719 - Epoch: [23][ 330/ 1207] Overall Loss 0.391431 Objective Loss 0.391431 LR 0.001000 Time 0.021130 -2023-02-13 17:28:54,908 - Epoch: [23][ 340/ 1207] Overall Loss 0.389865 Objective Loss 0.389865 LR 0.001000 Time 0.021065 -2023-02-13 17:28:55,098 - Epoch: [23][ 350/ 1207] Overall Loss 0.389760 Objective Loss 0.389760 LR 0.001000 Time 0.021004 -2023-02-13 17:28:55,286 - Epoch: [23][ 360/ 1207] Overall Loss 0.389844 Objective Loss 0.389844 LR 0.001000 Time 0.020943 -2023-02-13 17:28:55,475 - Epoch: [23][ 370/ 1207] Overall Loss 0.389241 Objective Loss 0.389241 LR 0.001000 Time 0.020885 -2023-02-13 17:28:55,664 - Epoch: [23][ 380/ 1207] Overall Loss 0.388490 Objective Loss 0.388490 LR 0.001000 Time 0.020832 -2023-02-13 17:28:55,853 - Epoch: [23][ 390/ 1207] Overall Loss 0.389142 Objective Loss 0.389142 LR 0.001000 Time 0.020782 -2023-02-13 17:28:56,043 - Epoch: [23][ 400/ 1207] Overall Loss 0.388754 Objective Loss 0.388754 LR 0.001000 Time 0.020736 -2023-02-13 17:28:56,231 - Epoch: [23][ 410/ 1207] Overall Loss 0.388712 Objective Loss 0.388712 LR 0.001000 Time 0.020690 -2023-02-13 17:28:56,420 - Epoch: [23][ 420/ 1207] Overall Loss 0.388572 Objective Loss 0.388572 LR 0.001000 Time 0.020646 -2023-02-13 17:28:56,609 - Epoch: [23][ 430/ 1207] Overall Loss 0.387892 Objective Loss 0.387892 LR 0.001000 Time 0.020603 -2023-02-13 17:28:56,798 - Epoch: [23][ 440/ 1207] Overall Loss 0.389195 Objective Loss 0.389195 LR 0.001000 Time 0.020563 -2023-02-13 17:28:56,987 - Epoch: [23][ 450/ 1207] Overall Loss 0.389451 Objective Loss 0.389451 LR 0.001000 Time 0.020526 -2023-02-13 17:28:57,176 - Epoch: [23][ 460/ 1207] Overall Loss 0.390169 Objective Loss 0.390169 LR 0.001000 Time 0.020491 -2023-02-13 17:28:57,365 - Epoch: [23][ 470/ 1207] Overall Loss 0.389398 Objective Loss 0.389398 LR 0.001000 Time 0.020456 -2023-02-13 17:28:57,554 - Epoch: [23][ 480/ 1207] Overall Loss 0.390254 Objective Loss 0.390254 LR 0.001000 Time 0.020422 -2023-02-13 17:28:57,743 - Epoch: [23][ 490/ 1207] Overall Loss 0.389880 Objective Loss 0.389880 LR 0.001000 Time 0.020390 -2023-02-13 17:28:57,932 - Epoch: [23][ 500/ 1207] Overall Loss 0.390402 Objective Loss 0.390402 LR 0.001000 Time 0.020361 -2023-02-13 17:28:58,121 - Epoch: [23][ 510/ 1207] Overall Loss 0.391077 Objective Loss 0.391077 LR 0.001000 Time 0.020332 -2023-02-13 17:28:58,310 - Epoch: [23][ 520/ 1207] Overall Loss 0.390543 Objective Loss 0.390543 LR 0.001000 Time 0.020303 -2023-02-13 17:28:58,499 - Epoch: [23][ 530/ 1207] Overall Loss 0.390500 Objective Loss 0.390500 LR 0.001000 Time 0.020275 -2023-02-13 17:28:58,687 - Epoch: [23][ 540/ 1207] Overall Loss 0.390208 Objective Loss 0.390208 LR 0.001000 Time 0.020248 -2023-02-13 17:28:58,876 - Epoch: [23][ 550/ 1207] Overall Loss 0.389780 Objective Loss 0.389780 LR 0.001000 Time 0.020222 -2023-02-13 17:28:59,065 - Epoch: [23][ 560/ 1207] Overall Loss 0.389480 Objective Loss 0.389480 LR 0.001000 Time 0.020198 -2023-02-13 17:28:59,255 - Epoch: [23][ 570/ 1207] Overall Loss 0.389248 Objective Loss 0.389248 LR 0.001000 Time 0.020175 -2023-02-13 17:28:59,444 - Epoch: [23][ 580/ 1207] Overall Loss 0.388273 Objective Loss 0.388273 LR 0.001000 Time 0.020153 -2023-02-13 17:28:59,632 - Epoch: [23][ 590/ 1207] Overall Loss 0.388439 Objective Loss 0.388439 LR 0.001000 Time 0.020130 -2023-02-13 17:28:59,821 - Epoch: [23][ 600/ 1207] Overall Loss 0.388290 Objective Loss 0.388290 LR 0.001000 Time 0.020109 -2023-02-13 17:29:00,009 - Epoch: [23][ 610/ 1207] Overall Loss 0.388285 Objective Loss 0.388285 LR 0.001000 Time 0.020088 -2023-02-13 17:29:00,198 - Epoch: [23][ 620/ 1207] Overall Loss 0.388144 Objective Loss 0.388144 LR 0.001000 Time 0.020067 -2023-02-13 17:29:00,387 - Epoch: [23][ 630/ 1207] Overall Loss 0.388116 Objective Loss 0.388116 LR 0.001000 Time 0.020047 -2023-02-13 17:29:00,574 - Epoch: [23][ 640/ 1207] Overall Loss 0.388316 Objective Loss 0.388316 LR 0.001000 Time 0.020027 -2023-02-13 17:29:00,764 - Epoch: [23][ 650/ 1207] Overall Loss 0.388104 Objective Loss 0.388104 LR 0.001000 Time 0.020009 -2023-02-13 17:29:00,964 - Epoch: [23][ 660/ 1207] Overall Loss 0.387922 Objective Loss 0.387922 LR 0.001000 Time 0.020010 -2023-02-13 17:29:01,169 - Epoch: [23][ 670/ 1207] Overall Loss 0.388224 Objective Loss 0.388224 LR 0.001000 Time 0.020016 -2023-02-13 17:29:01,369 - Epoch: [23][ 680/ 1207] Overall Loss 0.388550 Objective Loss 0.388550 LR 0.001000 Time 0.020015 -2023-02-13 17:29:01,573 - Epoch: [23][ 690/ 1207] Overall Loss 0.388001 Objective Loss 0.388001 LR 0.001000 Time 0.020020 -2023-02-13 17:29:01,773 - Epoch: [23][ 700/ 1207] Overall Loss 0.387617 Objective Loss 0.387617 LR 0.001000 Time 0.020020 -2023-02-13 17:29:01,979 - Epoch: [23][ 710/ 1207] Overall Loss 0.387794 Objective Loss 0.387794 LR 0.001000 Time 0.020027 -2023-02-13 17:29:02,178 - Epoch: [23][ 720/ 1207] Overall Loss 0.388216 Objective Loss 0.388216 LR 0.001000 Time 0.020025 -2023-02-13 17:29:02,382 - Epoch: [23][ 730/ 1207] Overall Loss 0.388108 Objective Loss 0.388108 LR 0.001000 Time 0.020030 -2023-02-13 17:29:02,582 - Epoch: [23][ 740/ 1207] Overall Loss 0.388129 Objective Loss 0.388129 LR 0.001000 Time 0.020029 -2023-02-13 17:29:02,787 - Epoch: [23][ 750/ 1207] Overall Loss 0.388138 Objective Loss 0.388138 LR 0.001000 Time 0.020034 -2023-02-13 17:29:02,987 - Epoch: [23][ 760/ 1207] Overall Loss 0.388846 Objective Loss 0.388846 LR 0.001000 Time 0.020033 -2023-02-13 17:29:03,191 - Epoch: [23][ 770/ 1207] Overall Loss 0.389028 Objective Loss 0.389028 LR 0.001000 Time 0.020037 -2023-02-13 17:29:03,392 - Epoch: [23][ 780/ 1207] Overall Loss 0.389188 Objective Loss 0.389188 LR 0.001000 Time 0.020037 -2023-02-13 17:29:03,596 - Epoch: [23][ 790/ 1207] Overall Loss 0.388948 Objective Loss 0.388948 LR 0.001000 Time 0.020041 -2023-02-13 17:29:03,796 - Epoch: [23][ 800/ 1207] Overall Loss 0.388985 Objective Loss 0.388985 LR 0.001000 Time 0.020040 -2023-02-13 17:29:04,001 - Epoch: [23][ 810/ 1207] Overall Loss 0.388790 Objective Loss 0.388790 LR 0.001000 Time 0.020046 -2023-02-13 17:29:04,201 - Epoch: [23][ 820/ 1207] Overall Loss 0.388639 Objective Loss 0.388639 LR 0.001000 Time 0.020045 -2023-02-13 17:29:04,406 - Epoch: [23][ 830/ 1207] Overall Loss 0.388789 Objective Loss 0.388789 LR 0.001000 Time 0.020049 -2023-02-13 17:29:04,605 - Epoch: [23][ 840/ 1207] Overall Loss 0.388922 Objective Loss 0.388922 LR 0.001000 Time 0.020047 -2023-02-13 17:29:04,810 - Epoch: [23][ 850/ 1207] Overall Loss 0.389442 Objective Loss 0.389442 LR 0.001000 Time 0.020053 -2023-02-13 17:29:05,011 - Epoch: [23][ 860/ 1207] Overall Loss 0.389847 Objective Loss 0.389847 LR 0.001000 Time 0.020053 -2023-02-13 17:29:05,216 - Epoch: [23][ 870/ 1207] Overall Loss 0.389536 Objective Loss 0.389536 LR 0.001000 Time 0.020057 -2023-02-13 17:29:05,405 - Epoch: [23][ 880/ 1207] Overall Loss 0.389605 Objective Loss 0.389605 LR 0.001000 Time 0.020043 -2023-02-13 17:29:05,594 - Epoch: [23][ 890/ 1207] Overall Loss 0.389922 Objective Loss 0.389922 LR 0.001000 Time 0.020030 -2023-02-13 17:29:05,783 - Epoch: [23][ 900/ 1207] Overall Loss 0.389857 Objective Loss 0.389857 LR 0.001000 Time 0.020017 -2023-02-13 17:29:05,972 - Epoch: [23][ 910/ 1207] Overall Loss 0.389872 Objective Loss 0.389872 LR 0.001000 Time 0.020005 -2023-02-13 17:29:06,162 - Epoch: [23][ 920/ 1207] Overall Loss 0.389820 Objective Loss 0.389820 LR 0.001000 Time 0.019993 -2023-02-13 17:29:06,350 - Epoch: [23][ 930/ 1207] Overall Loss 0.389804 Objective Loss 0.389804 LR 0.001000 Time 0.019980 -2023-02-13 17:29:06,538 - Epoch: [23][ 940/ 1207] Overall Loss 0.389869 Objective Loss 0.389869 LR 0.001000 Time 0.019968 -2023-02-13 17:29:06,728 - Epoch: [23][ 950/ 1207] Overall Loss 0.390110 Objective Loss 0.390110 LR 0.001000 Time 0.019956 -2023-02-13 17:29:06,918 - Epoch: [23][ 960/ 1207] Overall Loss 0.390145 Objective Loss 0.390145 LR 0.001000 Time 0.019946 -2023-02-13 17:29:07,107 - Epoch: [23][ 970/ 1207] Overall Loss 0.390204 Objective Loss 0.390204 LR 0.001000 Time 0.019936 -2023-02-13 17:29:07,296 - Epoch: [23][ 980/ 1207] Overall Loss 0.390310 Objective Loss 0.390310 LR 0.001000 Time 0.019924 -2023-02-13 17:29:07,485 - Epoch: [23][ 990/ 1207] Overall Loss 0.390549 Objective Loss 0.390549 LR 0.001000 Time 0.019913 -2023-02-13 17:29:07,674 - Epoch: [23][ 1000/ 1207] Overall Loss 0.390553 Objective Loss 0.390553 LR 0.001000 Time 0.019903 -2023-02-13 17:29:07,862 - Epoch: [23][ 1010/ 1207] Overall Loss 0.390776 Objective Loss 0.390776 LR 0.001000 Time 0.019892 -2023-02-13 17:29:08,052 - Epoch: [23][ 1020/ 1207] Overall Loss 0.390818 Objective Loss 0.390818 LR 0.001000 Time 0.019883 -2023-02-13 17:29:08,241 - Epoch: [23][ 1030/ 1207] Overall Loss 0.390818 Objective Loss 0.390818 LR 0.001000 Time 0.019872 -2023-02-13 17:29:08,429 - Epoch: [23][ 1040/ 1207] Overall Loss 0.390867 Objective Loss 0.390867 LR 0.001000 Time 0.019862 -2023-02-13 17:29:08,618 - Epoch: [23][ 1050/ 1207] Overall Loss 0.391020 Objective Loss 0.391020 LR 0.001000 Time 0.019852 -2023-02-13 17:29:08,806 - Epoch: [23][ 1060/ 1207] Overall Loss 0.391116 Objective Loss 0.391116 LR 0.001000 Time 0.019842 -2023-02-13 17:29:08,995 - Epoch: [23][ 1070/ 1207] Overall Loss 0.390802 Objective Loss 0.390802 LR 0.001000 Time 0.019834 -2023-02-13 17:29:09,185 - Epoch: [23][ 1080/ 1207] Overall Loss 0.390672 Objective Loss 0.390672 LR 0.001000 Time 0.019825 -2023-02-13 17:29:09,374 - Epoch: [23][ 1090/ 1207] Overall Loss 0.390501 Objective Loss 0.390501 LR 0.001000 Time 0.019816 -2023-02-13 17:29:09,563 - Epoch: [23][ 1100/ 1207] Overall Loss 0.390756 Objective Loss 0.390756 LR 0.001000 Time 0.019807 -2023-02-13 17:29:09,751 - Epoch: [23][ 1110/ 1207] Overall Loss 0.391102 Objective Loss 0.391102 LR 0.001000 Time 0.019799 -2023-02-13 17:29:09,942 - Epoch: [23][ 1120/ 1207] Overall Loss 0.391189 Objective Loss 0.391189 LR 0.001000 Time 0.019792 -2023-02-13 17:29:10,132 - Epoch: [23][ 1130/ 1207] Overall Loss 0.391558 Objective Loss 0.391558 LR 0.001000 Time 0.019784 -2023-02-13 17:29:10,321 - Epoch: [23][ 1140/ 1207] Overall Loss 0.391662 Objective Loss 0.391662 LR 0.001000 Time 0.019776 -2023-02-13 17:29:10,509 - Epoch: [23][ 1150/ 1207] Overall Loss 0.391890 Objective Loss 0.391890 LR 0.001000 Time 0.019768 -2023-02-13 17:29:10,698 - Epoch: [23][ 1160/ 1207] Overall Loss 0.391828 Objective Loss 0.391828 LR 0.001000 Time 0.019760 -2023-02-13 17:29:10,889 - Epoch: [23][ 1170/ 1207] Overall Loss 0.391439 Objective Loss 0.391439 LR 0.001000 Time 0.019753 -2023-02-13 17:29:11,079 - Epoch: [23][ 1180/ 1207] Overall Loss 0.391494 Objective Loss 0.391494 LR 0.001000 Time 0.019747 -2023-02-13 17:29:11,267 - Epoch: [23][ 1190/ 1207] Overall Loss 0.391293 Objective Loss 0.391293 LR 0.001000 Time 0.019739 -2023-02-13 17:29:11,507 - Epoch: [23][ 1200/ 1207] Overall Loss 0.391235 Objective Loss 0.391235 LR 0.001000 Time 0.019774 -2023-02-13 17:29:11,623 - Epoch: [23][ 1207/ 1207] Overall Loss 0.391205 Objective Loss 0.391205 Top1 83.536585 Top5 97.560976 LR 0.001000 Time 0.019755 -2023-02-13 17:29:11,694 - --- validate (epoch=23)----------- -2023-02-13 17:29:11,694 - 34311 samples (256 per mini-batch) -2023-02-13 17:29:12,084 - Epoch: [23][ 10/ 135] Loss 0.422561 Top1 80.546875 Top5 96.328125 -2023-02-13 17:29:12,210 - Epoch: [23][ 20/ 135] Loss 0.411261 Top1 80.039062 Top5 96.484375 -2023-02-13 17:29:12,334 - Epoch: [23][ 30/ 135] Loss 0.411644 Top1 79.791667 Top5 96.497396 -2023-02-13 17:29:12,454 - Epoch: [23][ 40/ 135] Loss 0.416743 Top1 79.794922 Top5 96.435547 -2023-02-13 17:29:12,588 - Epoch: [23][ 50/ 135] Loss 0.423092 Top1 79.382812 Top5 96.265625 -2023-02-13 17:29:12,714 - Epoch: [23][ 60/ 135] Loss 0.421610 Top1 79.505208 Top5 96.295573 -2023-02-13 17:29:12,837 - Epoch: [23][ 70/ 135] Loss 0.432112 Top1 79.391741 Top5 96.188616 -2023-02-13 17:29:12,959 - Epoch: [23][ 80/ 135] Loss 0.430489 Top1 79.389648 Top5 96.215820 -2023-02-13 17:29:13,085 - Epoch: [23][ 90/ 135] Loss 0.426658 Top1 79.401042 Top5 96.271701 -2023-02-13 17:29:13,213 - Epoch: [23][ 100/ 135] Loss 0.431347 Top1 79.371094 Top5 96.234375 -2023-02-13 17:29:13,342 - Epoch: [23][ 110/ 135] Loss 0.429712 Top1 79.524148 Top5 96.281960 -2023-02-13 17:29:13,473 - Epoch: [23][ 120/ 135] Loss 0.427711 Top1 79.599609 Top5 96.302083 -2023-02-13 17:29:13,605 - Epoch: [23][ 130/ 135] Loss 0.428005 Top1 79.609375 Top5 96.301082 -2023-02-13 17:29:13,654 - Epoch: [23][ 135/ 135] Loss 0.424011 Top1 79.604209 Top5 96.310221 -2023-02-13 17:29:13,726 - ==> Top1: 79.604 Top5: 96.310 Loss: 0.424 - -2023-02-13 17:29:13,726 - ==> Confusion: -[[ 795 6 6 0 5 3 0 3 8 103 2 2 2 6 7 2 1 2 3 1 10] - [ 6 878 1 1 7 78 0 29 8 1 1 0 1 2 2 0 5 1 3 1 8] - [ 16 7 890 22 2 8 10 29 2 6 6 2 2 6 4 14 1 2 14 3 12] - [ 2 5 15 853 3 5 2 1 7 3 18 1 8 3 42 1 2 6 26 1 12] - [ 27 9 0 1 963 14 1 0 4 16 0 3 0 6 6 5 3 1 0 2 5] - [ 6 19 0 1 3 970 4 18 0 9 1 7 4 16 1 2 2 1 1 2 3] - [ 4 6 16 4 0 14 1002 12 2 2 8 1 1 2 0 9 0 0 2 9 5] - [ 3 12 4 2 2 37 5 908 2 6 2 3 2 3 1 0 0 0 22 7 3] - [ 14 3 0 2 2 0 0 0 901 52 2 2 2 7 15 1 0 1 5 0 0] - [ 57 3 2 0 3 0 0 1 38 878 1 0 0 20 3 3 0 1 0 0 2] - [ 4 8 2 3 3 2 2 4 35 3 943 0 0 14 5 1 1 1 14 1 5] - [ 3 4 0 0 5 31 1 4 7 1 0 891 21 9 2 4 1 7 2 9 3] - [ 3 1 1 1 3 11 0 1 9 0 0 61 822 3 5 3 1 19 3 3 9] - [ 8 4 2 1 6 19 1 0 29 23 3 7 1 906 5 4 0 0 1 1 3] - [ 13 0 1 9 7 6 0 0 38 9 0 3 3 3 976 2 1 3 10 0 8] - [ 9 4 2 2 12 6 7 2 0 0 0 4 9 3 1 952 8 11 0 6 8] - [ 8 14 1 0 16 10 0 0 5 1 0 5 0 5 6 17 961 2 1 2 7] - [ 10 6 0 4 1 3 2 2 4 1 0 20 19 4 3 14 0 951 0 3 4] - [ 4 9 2 7 1 5 0 39 8 1 2 1 2 1 18 1 0 2 982 0 1] - [ 2 4 0 0 1 31 8 29 1 0 1 22 3 8 0 9 1 3 1 1016 8] - [ 238 318 147 123 162 415 61 272 177 189 204 159 369 486 284 121 217 86 241 290 8875]] - -2023-02-13 17:29:13,728 - ==> Best [Top1: 79.718 Top5: 96.756 Sparsity:0.00 Params: 148928 on epoch: 22] -2023-02-13 17:29:13,728 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:29:13,734 - - -2023-02-13 17:29:13,734 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:29:14,714 - Epoch: [24][ 10/ 1207] Overall Loss 0.380273 Objective Loss 0.380273 LR 0.001000 Time 0.097926 -2023-02-13 17:29:14,904 - Epoch: [24][ 20/ 1207] Overall Loss 0.357243 Objective Loss 0.357243 LR 0.001000 Time 0.058429 -2023-02-13 17:29:15,094 - Epoch: [24][ 30/ 1207] Overall Loss 0.362348 Objective Loss 0.362348 LR 0.001000 Time 0.045279 -2023-02-13 17:29:15,282 - Epoch: [24][ 40/ 1207] Overall Loss 0.350637 Objective Loss 0.350637 LR 0.001000 Time 0.038657 -2023-02-13 17:29:15,470 - Epoch: [24][ 50/ 1207] Overall Loss 0.356674 Objective Loss 0.356674 LR 0.001000 Time 0.034682 -2023-02-13 17:29:15,658 - Epoch: [24][ 60/ 1207] Overall Loss 0.355979 Objective Loss 0.355979 LR 0.001000 Time 0.032028 -2023-02-13 17:29:15,847 - Epoch: [24][ 70/ 1207] Overall Loss 0.357166 Objective Loss 0.357166 LR 0.001000 Time 0.030149 -2023-02-13 17:29:16,037 - Epoch: [24][ 80/ 1207] Overall Loss 0.361513 Objective Loss 0.361513 LR 0.001000 Time 0.028747 -2023-02-13 17:29:16,225 - Epoch: [24][ 90/ 1207] Overall Loss 0.360722 Objective Loss 0.360722 LR 0.001000 Time 0.027641 -2023-02-13 17:29:16,414 - Epoch: [24][ 100/ 1207] Overall Loss 0.363081 Objective Loss 0.363081 LR 0.001000 Time 0.026759 -2023-02-13 17:29:16,602 - Epoch: [24][ 110/ 1207] Overall Loss 0.367336 Objective Loss 0.367336 LR 0.001000 Time 0.026031 -2023-02-13 17:29:16,791 - Epoch: [24][ 120/ 1207] Overall Loss 0.365459 Objective Loss 0.365459 LR 0.001000 Time 0.025438 -2023-02-13 17:29:16,980 - Epoch: [24][ 130/ 1207] Overall Loss 0.365010 Objective Loss 0.365010 LR 0.001000 Time 0.024929 -2023-02-13 17:29:17,169 - Epoch: [24][ 140/ 1207] Overall Loss 0.366360 Objective Loss 0.366360 LR 0.001000 Time 0.024493 -2023-02-13 17:29:17,357 - Epoch: [24][ 150/ 1207] Overall Loss 0.368509 Objective Loss 0.368509 LR 0.001000 Time 0.024112 -2023-02-13 17:29:17,545 - Epoch: [24][ 160/ 1207] Overall Loss 0.367362 Objective Loss 0.367362 LR 0.001000 Time 0.023781 -2023-02-13 17:29:17,733 - Epoch: [24][ 170/ 1207] Overall Loss 0.367028 Objective Loss 0.367028 LR 0.001000 Time 0.023484 -2023-02-13 17:29:17,921 - Epoch: [24][ 180/ 1207] Overall Loss 0.368455 Objective Loss 0.368455 LR 0.001000 Time 0.023223 -2023-02-13 17:29:18,110 - Epoch: [24][ 190/ 1207] Overall Loss 0.369219 Objective Loss 0.369219 LR 0.001000 Time 0.022994 -2023-02-13 17:29:18,298 - Epoch: [24][ 200/ 1207] Overall Loss 0.368529 Objective Loss 0.368529 LR 0.001000 Time 0.022782 -2023-02-13 17:29:18,485 - Epoch: [24][ 210/ 1207] Overall Loss 0.366623 Objective Loss 0.366623 LR 0.001000 Time 0.022587 -2023-02-13 17:29:18,673 - Epoch: [24][ 220/ 1207] Overall Loss 0.367398 Objective Loss 0.367398 LR 0.001000 Time 0.022413 -2023-02-13 17:29:18,861 - Epoch: [24][ 230/ 1207] Overall Loss 0.368968 Objective Loss 0.368968 LR 0.001000 Time 0.022254 -2023-02-13 17:29:19,051 - Epoch: [24][ 240/ 1207] Overall Loss 0.368976 Objective Loss 0.368976 LR 0.001000 Time 0.022115 -2023-02-13 17:29:19,238 - Epoch: [24][ 250/ 1207] Overall Loss 0.369939 Objective Loss 0.369939 LR 0.001000 Time 0.021979 -2023-02-13 17:29:19,427 - Epoch: [24][ 260/ 1207] Overall Loss 0.369646 Objective Loss 0.369646 LR 0.001000 Time 0.021857 -2023-02-13 17:29:19,615 - Epoch: [24][ 270/ 1207] Overall Loss 0.370053 Objective Loss 0.370053 LR 0.001000 Time 0.021742 -2023-02-13 17:29:19,803 - Epoch: [24][ 280/ 1207] Overall Loss 0.370985 Objective Loss 0.370985 LR 0.001000 Time 0.021638 -2023-02-13 17:29:19,992 - Epoch: [24][ 290/ 1207] Overall Loss 0.371118 Objective Loss 0.371118 LR 0.001000 Time 0.021541 -2023-02-13 17:29:20,181 - Epoch: [24][ 300/ 1207] Overall Loss 0.371026 Objective Loss 0.371026 LR 0.001000 Time 0.021451 -2023-02-13 17:29:20,369 - Epoch: [24][ 310/ 1207] Overall Loss 0.371048 Objective Loss 0.371048 LR 0.001000 Time 0.021364 -2023-02-13 17:29:20,557 - Epoch: [24][ 320/ 1207] Overall Loss 0.370702 Objective Loss 0.370702 LR 0.001000 Time 0.021284 -2023-02-13 17:29:20,745 - Epoch: [24][ 330/ 1207] Overall Loss 0.370148 Objective Loss 0.370148 LR 0.001000 Time 0.021208 -2023-02-13 17:29:20,934 - Epoch: [24][ 340/ 1207] Overall Loss 0.370484 Objective Loss 0.370484 LR 0.001000 Time 0.021139 -2023-02-13 17:29:21,122 - Epoch: [24][ 350/ 1207] Overall Loss 0.370595 Objective Loss 0.370595 LR 0.001000 Time 0.021072 -2023-02-13 17:29:21,311 - Epoch: [24][ 360/ 1207] Overall Loss 0.370594 Objective Loss 0.370594 LR 0.001000 Time 0.021009 -2023-02-13 17:29:21,499 - Epoch: [24][ 370/ 1207] Overall Loss 0.369704 Objective Loss 0.369704 LR 0.001000 Time 0.020947 -2023-02-13 17:29:21,687 - Epoch: [24][ 380/ 1207] Overall Loss 0.369948 Objective Loss 0.369948 LR 0.001000 Time 0.020891 -2023-02-13 17:29:21,875 - Epoch: [24][ 390/ 1207] Overall Loss 0.369858 Objective Loss 0.369858 LR 0.001000 Time 0.020837 -2023-02-13 17:29:22,065 - Epoch: [24][ 400/ 1207] Overall Loss 0.370921 Objective Loss 0.370921 LR 0.001000 Time 0.020789 -2023-02-13 17:29:22,253 - Epoch: [24][ 410/ 1207] Overall Loss 0.371632 Objective Loss 0.371632 LR 0.001000 Time 0.020741 -2023-02-13 17:29:22,442 - Epoch: [24][ 420/ 1207] Overall Loss 0.371939 Objective Loss 0.371939 LR 0.001000 Time 0.020695 -2023-02-13 17:29:22,630 - Epoch: [24][ 430/ 1207] Overall Loss 0.372898 Objective Loss 0.372898 LR 0.001000 Time 0.020651 -2023-02-13 17:29:22,818 - Epoch: [24][ 440/ 1207] Overall Loss 0.373261 Objective Loss 0.373261 LR 0.001000 Time 0.020608 -2023-02-13 17:29:23,007 - Epoch: [24][ 450/ 1207] Overall Loss 0.373310 Objective Loss 0.373310 LR 0.001000 Time 0.020568 -2023-02-13 17:29:23,196 - Epoch: [24][ 460/ 1207] Overall Loss 0.372926 Objective Loss 0.372926 LR 0.001000 Time 0.020531 -2023-02-13 17:29:23,383 - Epoch: [24][ 470/ 1207] Overall Loss 0.372796 Objective Loss 0.372796 LR 0.001000 Time 0.020492 -2023-02-13 17:29:23,571 - Epoch: [24][ 480/ 1207] Overall Loss 0.372702 Objective Loss 0.372702 LR 0.001000 Time 0.020456 -2023-02-13 17:29:23,759 - Epoch: [24][ 490/ 1207] Overall Loss 0.372871 Objective Loss 0.372871 LR 0.001000 Time 0.020422 -2023-02-13 17:29:23,948 - Epoch: [24][ 500/ 1207] Overall Loss 0.372956 Objective Loss 0.372956 LR 0.001000 Time 0.020389 -2023-02-13 17:29:24,136 - Epoch: [24][ 510/ 1207] Overall Loss 0.372665 Objective Loss 0.372665 LR 0.001000 Time 0.020358 -2023-02-13 17:29:24,325 - Epoch: [24][ 520/ 1207] Overall Loss 0.373271 Objective Loss 0.373271 LR 0.001000 Time 0.020329 -2023-02-13 17:29:24,512 - Epoch: [24][ 530/ 1207] Overall Loss 0.373579 Objective Loss 0.373579 LR 0.001000 Time 0.020299 -2023-02-13 17:29:24,701 - Epoch: [24][ 540/ 1207] Overall Loss 0.373942 Objective Loss 0.373942 LR 0.001000 Time 0.020271 -2023-02-13 17:29:24,889 - Epoch: [24][ 550/ 1207] Overall Loss 0.373808 Objective Loss 0.373808 LR 0.001000 Time 0.020244 -2023-02-13 17:29:25,078 - Epoch: [24][ 560/ 1207] Overall Loss 0.374159 Objective Loss 0.374159 LR 0.001000 Time 0.020219 -2023-02-13 17:29:25,265 - Epoch: [24][ 570/ 1207] Overall Loss 0.373840 Objective Loss 0.373840 LR 0.001000 Time 0.020193 -2023-02-13 17:29:25,454 - Epoch: [24][ 580/ 1207] Overall Loss 0.373991 Objective Loss 0.373991 LR 0.001000 Time 0.020169 -2023-02-13 17:29:25,642 - Epoch: [24][ 590/ 1207] Overall Loss 0.373846 Objective Loss 0.373846 LR 0.001000 Time 0.020145 -2023-02-13 17:29:25,831 - Epoch: [24][ 600/ 1207] Overall Loss 0.373611 Objective Loss 0.373611 LR 0.001000 Time 0.020124 -2023-02-13 17:29:26,019 - Epoch: [24][ 610/ 1207] Overall Loss 0.373418 Objective Loss 0.373418 LR 0.001000 Time 0.020102 -2023-02-13 17:29:26,208 - Epoch: [24][ 620/ 1207] Overall Loss 0.373539 Objective Loss 0.373539 LR 0.001000 Time 0.020082 -2023-02-13 17:29:26,396 - Epoch: [24][ 630/ 1207] Overall Loss 0.374286 Objective Loss 0.374286 LR 0.001000 Time 0.020061 -2023-02-13 17:29:26,584 - Epoch: [24][ 640/ 1207] Overall Loss 0.374511 Objective Loss 0.374511 LR 0.001000 Time 0.020041 -2023-02-13 17:29:26,772 - Epoch: [24][ 650/ 1207] Overall Loss 0.374521 Objective Loss 0.374521 LR 0.001000 Time 0.020021 -2023-02-13 17:29:26,961 - Epoch: [24][ 660/ 1207] Overall Loss 0.374354 Objective Loss 0.374354 LR 0.001000 Time 0.020004 -2023-02-13 17:29:27,150 - Epoch: [24][ 670/ 1207] Overall Loss 0.374321 Objective Loss 0.374321 LR 0.001000 Time 0.019986 -2023-02-13 17:29:27,338 - Epoch: [24][ 680/ 1207] Overall Loss 0.374789 Objective Loss 0.374789 LR 0.001000 Time 0.019969 -2023-02-13 17:29:27,527 - Epoch: [24][ 690/ 1207] Overall Loss 0.374381 Objective Loss 0.374381 LR 0.001000 Time 0.019952 -2023-02-13 17:29:27,716 - Epoch: [24][ 700/ 1207] Overall Loss 0.375063 Objective Loss 0.375063 LR 0.001000 Time 0.019936 -2023-02-13 17:29:27,905 - Epoch: [24][ 710/ 1207] Overall Loss 0.375385 Objective Loss 0.375385 LR 0.001000 Time 0.019921 -2023-02-13 17:29:28,094 - Epoch: [24][ 720/ 1207] Overall Loss 0.375592 Objective Loss 0.375592 LR 0.001000 Time 0.019907 -2023-02-13 17:29:28,282 - Epoch: [24][ 730/ 1207] Overall Loss 0.375450 Objective Loss 0.375450 LR 0.001000 Time 0.019891 -2023-02-13 17:29:28,471 - Epoch: [24][ 740/ 1207] Overall Loss 0.374997 Objective Loss 0.374997 LR 0.001000 Time 0.019877 -2023-02-13 17:29:28,659 - Epoch: [24][ 750/ 1207] Overall Loss 0.374445 Objective Loss 0.374445 LR 0.001000 Time 0.019863 -2023-02-13 17:29:28,847 - Epoch: [24][ 760/ 1207] Overall Loss 0.374739 Objective Loss 0.374739 LR 0.001000 Time 0.019848 -2023-02-13 17:29:29,036 - Epoch: [24][ 770/ 1207] Overall Loss 0.375062 Objective Loss 0.375062 LR 0.001000 Time 0.019835 -2023-02-13 17:29:29,226 - Epoch: [24][ 780/ 1207] Overall Loss 0.375040 Objective Loss 0.375040 LR 0.001000 Time 0.019824 -2023-02-13 17:29:29,414 - Epoch: [24][ 790/ 1207] Overall Loss 0.375211 Objective Loss 0.375211 LR 0.001000 Time 0.019811 -2023-02-13 17:29:29,603 - Epoch: [24][ 800/ 1207] Overall Loss 0.374889 Objective Loss 0.374889 LR 0.001000 Time 0.019798 -2023-02-13 17:29:29,791 - Epoch: [24][ 810/ 1207] Overall Loss 0.374731 Objective Loss 0.374731 LR 0.001000 Time 0.019785 -2023-02-13 17:29:29,979 - Epoch: [24][ 820/ 1207] Overall Loss 0.374528 Objective Loss 0.374528 LR 0.001000 Time 0.019773 -2023-02-13 17:29:30,168 - Epoch: [24][ 830/ 1207] Overall Loss 0.374408 Objective Loss 0.374408 LR 0.001000 Time 0.019763 -2023-02-13 17:29:30,357 - Epoch: [24][ 840/ 1207] Overall Loss 0.374558 Objective Loss 0.374558 LR 0.001000 Time 0.019751 -2023-02-13 17:29:30,545 - Epoch: [24][ 850/ 1207] Overall Loss 0.374909 Objective Loss 0.374909 LR 0.001000 Time 0.019740 -2023-02-13 17:29:30,733 - Epoch: [24][ 860/ 1207] Overall Loss 0.375021 Objective Loss 0.375021 LR 0.001000 Time 0.019729 -2023-02-13 17:29:30,923 - Epoch: [24][ 870/ 1207] Overall Loss 0.375095 Objective Loss 0.375095 LR 0.001000 Time 0.019720 -2023-02-13 17:29:31,113 - Epoch: [24][ 880/ 1207] Overall Loss 0.375255 Objective Loss 0.375255 LR 0.001000 Time 0.019712 -2023-02-13 17:29:31,301 - Epoch: [24][ 890/ 1207] Overall Loss 0.375172 Objective Loss 0.375172 LR 0.001000 Time 0.019701 -2023-02-13 17:29:31,490 - Epoch: [24][ 900/ 1207] Overall Loss 0.375308 Objective Loss 0.375308 LR 0.001000 Time 0.019691 -2023-02-13 17:29:31,679 - Epoch: [24][ 910/ 1207] Overall Loss 0.375663 Objective Loss 0.375663 LR 0.001000 Time 0.019682 -2023-02-13 17:29:31,868 - Epoch: [24][ 920/ 1207] Overall Loss 0.375918 Objective Loss 0.375918 LR 0.001000 Time 0.019673 -2023-02-13 17:29:32,057 - Epoch: [24][ 930/ 1207] Overall Loss 0.375966 Objective Loss 0.375966 LR 0.001000 Time 0.019665 -2023-02-13 17:29:32,246 - Epoch: [24][ 940/ 1207] Overall Loss 0.376222 Objective Loss 0.376222 LR 0.001000 Time 0.019656 -2023-02-13 17:29:32,436 - Epoch: [24][ 950/ 1207] Overall Loss 0.376238 Objective Loss 0.376238 LR 0.001000 Time 0.019649 -2023-02-13 17:29:32,625 - Epoch: [24][ 960/ 1207] Overall Loss 0.376303 Objective Loss 0.376303 LR 0.001000 Time 0.019641 -2023-02-13 17:29:32,813 - Epoch: [24][ 970/ 1207] Overall Loss 0.376157 Objective Loss 0.376157 LR 0.001000 Time 0.019632 -2023-02-13 17:29:33,001 - Epoch: [24][ 980/ 1207] Overall Loss 0.376315 Objective Loss 0.376315 LR 0.001000 Time 0.019623 -2023-02-13 17:29:33,191 - Epoch: [24][ 990/ 1207] Overall Loss 0.376434 Objective Loss 0.376434 LR 0.001000 Time 0.019616 -2023-02-13 17:29:33,379 - Epoch: [24][ 1000/ 1207] Overall Loss 0.376253 Objective Loss 0.376253 LR 0.001000 Time 0.019608 -2023-02-13 17:29:33,567 - Epoch: [24][ 1010/ 1207] Overall Loss 0.376197 Objective Loss 0.376197 LR 0.001000 Time 0.019599 -2023-02-13 17:29:33,756 - Epoch: [24][ 1020/ 1207] Overall Loss 0.376148 Objective Loss 0.376148 LR 0.001000 Time 0.019591 -2023-02-13 17:29:33,945 - Epoch: [24][ 1030/ 1207] Overall Loss 0.376136 Objective Loss 0.376136 LR 0.001000 Time 0.019585 -2023-02-13 17:29:34,134 - Epoch: [24][ 1040/ 1207] Overall Loss 0.376548 Objective Loss 0.376548 LR 0.001000 Time 0.019578 -2023-02-13 17:29:34,323 - Epoch: [24][ 1050/ 1207] Overall Loss 0.376543 Objective Loss 0.376543 LR 0.001000 Time 0.019571 -2023-02-13 17:29:34,511 - Epoch: [24][ 1060/ 1207] Overall Loss 0.376713 Objective Loss 0.376713 LR 0.001000 Time 0.019564 -2023-02-13 17:29:34,700 - Epoch: [24][ 1070/ 1207] Overall Loss 0.377417 Objective Loss 0.377417 LR 0.001000 Time 0.019556 -2023-02-13 17:29:34,888 - Epoch: [24][ 1080/ 1207] Overall Loss 0.377674 Objective Loss 0.377674 LR 0.001000 Time 0.019550 -2023-02-13 17:29:35,077 - Epoch: [24][ 1090/ 1207] Overall Loss 0.377636 Objective Loss 0.377636 LR 0.001000 Time 0.019543 -2023-02-13 17:29:35,266 - Epoch: [24][ 1100/ 1207] Overall Loss 0.377698 Objective Loss 0.377698 LR 0.001000 Time 0.019537 -2023-02-13 17:29:35,455 - Epoch: [24][ 1110/ 1207] Overall Loss 0.377588 Objective Loss 0.377588 LR 0.001000 Time 0.019531 -2023-02-13 17:29:35,644 - Epoch: [24][ 1120/ 1207] Overall Loss 0.377638 Objective Loss 0.377638 LR 0.001000 Time 0.019525 -2023-02-13 17:29:35,834 - Epoch: [24][ 1130/ 1207] Overall Loss 0.378001 Objective Loss 0.378001 LR 0.001000 Time 0.019520 -2023-02-13 17:29:36,022 - Epoch: [24][ 1140/ 1207] Overall Loss 0.378229 Objective Loss 0.378229 LR 0.001000 Time 0.019514 -2023-02-13 17:29:36,212 - Epoch: [24][ 1150/ 1207] Overall Loss 0.378700 Objective Loss 0.378700 LR 0.001000 Time 0.019509 -2023-02-13 17:29:36,402 - Epoch: [24][ 1160/ 1207] Overall Loss 0.378760 Objective Loss 0.378760 LR 0.001000 Time 0.019504 -2023-02-13 17:29:36,591 - Epoch: [24][ 1170/ 1207] Overall Loss 0.379146 Objective Loss 0.379146 LR 0.001000 Time 0.019498 -2023-02-13 17:29:36,780 - Epoch: [24][ 1180/ 1207] Overall Loss 0.379166 Objective Loss 0.379166 LR 0.001000 Time 0.019493 -2023-02-13 17:29:36,969 - Epoch: [24][ 1190/ 1207] Overall Loss 0.379293 Objective Loss 0.379293 LR 0.001000 Time 0.019488 -2023-02-13 17:29:37,212 - Epoch: [24][ 1200/ 1207] Overall Loss 0.379156 Objective Loss 0.379156 LR 0.001000 Time 0.019527 -2023-02-13 17:29:37,328 - Epoch: [24][ 1207/ 1207] Overall Loss 0.379466 Objective Loss 0.379466 Top1 77.439024 Top5 96.341463 LR 0.001000 Time 0.019510 -2023-02-13 17:29:37,411 - --- validate (epoch=24)----------- -2023-02-13 17:29:37,411 - 34311 samples (256 per mini-batch) -2023-02-13 17:29:37,806 - Epoch: [24][ 10/ 135] Loss 0.416127 Top1 78.867188 Top5 96.054688 -2023-02-13 17:29:37,935 - Epoch: [24][ 20/ 135] Loss 0.412034 Top1 79.062500 Top5 96.542969 -2023-02-13 17:29:38,063 - Epoch: [24][ 30/ 135] Loss 0.406952 Top1 79.296875 Top5 96.627604 -2023-02-13 17:29:38,191 - Epoch: [24][ 40/ 135] Loss 0.403526 Top1 79.023438 Top5 96.699219 -2023-02-13 17:29:38,318 - Epoch: [24][ 50/ 135] Loss 0.402301 Top1 79.117188 Top5 96.632812 -2023-02-13 17:29:38,449 - Epoch: [24][ 60/ 135] Loss 0.397126 Top1 79.153646 Top5 96.640625 -2023-02-13 17:29:38,580 - Epoch: [24][ 70/ 135] Loss 0.398632 Top1 79.174107 Top5 96.629464 -2023-02-13 17:29:38,711 - Epoch: [24][ 80/ 135] Loss 0.397719 Top1 79.331055 Top5 96.679688 -2023-02-13 17:29:38,838 - Epoch: [24][ 90/ 135] Loss 0.400664 Top1 79.205729 Top5 96.705729 -2023-02-13 17:29:38,965 - Epoch: [24][ 100/ 135] Loss 0.400112 Top1 79.093750 Top5 96.656250 -2023-02-13 17:29:39,092 - Epoch: [24][ 110/ 135] Loss 0.401945 Top1 79.073153 Top5 96.686790 -2023-02-13 17:29:39,219 - Epoch: [24][ 120/ 135] Loss 0.401875 Top1 79.153646 Top5 96.708984 -2023-02-13 17:29:39,349 - Epoch: [24][ 130/ 135] Loss 0.403515 Top1 79.242788 Top5 96.736779 -2023-02-13 17:29:39,395 - Epoch: [24][ 135/ 135] Loss 0.400523 Top1 79.304013 Top5 96.756142 -2023-02-13 17:29:39,463 - ==> Top1: 79.304 Top5: 96.756 Loss: 0.401 - -2023-02-13 17:29:39,463 - ==> Confusion: -[[ 878 4 7 2 11 4 0 3 2 24 0 3 0 4 11 2 3 2 1 1 5] - [ 4 961 0 2 5 20 0 13 3 2 1 5 0 2 0 0 3 0 5 2 5] - [ 6 15 956 10 1 2 12 14 0 0 1 5 0 1 3 8 1 4 9 2 8] - [ 5 3 32 865 0 5 0 1 2 2 15 3 5 2 25 2 5 6 27 1 10] - [ 24 12 2 0 975 10 0 0 4 4 0 4 1 3 3 8 5 3 0 3 5] - [ 3 47 5 4 7 920 2 31 2 6 1 17 1 6 1 2 1 1 3 6 4] - [ 4 7 33 2 0 6 1003 6 0 0 6 3 5 0 0 9 1 1 0 8 5] - [ 3 20 13 1 2 29 3 907 0 1 3 8 2 0 1 1 0 1 17 5 7] - [ 32 8 0 2 1 0 1 0 855 46 7 1 0 9 32 0 2 2 9 1 1] - [ 133 4 3 1 6 2 0 1 23 807 0 0 2 15 7 2 0 2 1 0 3] - [ 3 9 7 5 1 2 6 7 16 2 943 3 2 9 3 0 3 1 22 0 7] - [ 5 3 0 1 2 16 1 3 1 2 0 921 20 1 0 5 2 11 2 9 0] - [ 3 5 1 3 0 3 0 1 0 0 0 58 829 0 6 6 3 32 1 0 8] - [ 11 6 4 0 9 25 2 1 15 29 6 19 6 858 8 6 7 1 0 5 6] - [ 24 11 2 20 5 2 0 1 21 2 2 2 4 0 959 0 6 6 17 1 7] - [ 8 5 8 2 8 0 2 2 0 0 0 13 9 0 1 949 6 20 0 4 9] - [ 6 19 0 1 13 5 0 0 2 0 0 8 5 0 4 20 968 1 1 3 5] - [ 10 3 3 4 0 2 1 2 0 0 0 13 18 0 0 8 0 982 0 1 4] - [ 4 7 11 5 2 1 1 35 2 1 1 3 6 0 13 1 0 1 988 3 1] - [ 1 3 2 0 2 11 9 21 0 0 0 25 4 0 0 8 3 5 0 1039 15] - [ 270 532 337 101 202 254 102 217 124 113 154 219 399 293 216 131 388 144 240 351 8647]] - -2023-02-13 17:29:39,465 - ==> Best [Top1: 79.718 Top5: 96.756 Sparsity:0.00 Params: 148928 on epoch: 22] -2023-02-13 17:29:39,465 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:29:39,470 - - -2023-02-13 17:29:39,470 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:29:40,379 - Epoch: [25][ 10/ 1207] Overall Loss 0.373478 Objective Loss 0.373478 LR 0.001000 Time 0.090785 -2023-02-13 17:29:40,569 - Epoch: [25][ 20/ 1207] Overall Loss 0.368273 Objective Loss 0.368273 LR 0.001000 Time 0.054856 -2023-02-13 17:29:40,757 - Epoch: [25][ 30/ 1207] Overall Loss 0.366193 Objective Loss 0.366193 LR 0.001000 Time 0.042831 -2023-02-13 17:29:40,944 - Epoch: [25][ 40/ 1207] Overall Loss 0.370060 Objective Loss 0.370060 LR 0.001000 Time 0.036794 -2023-02-13 17:29:41,132 - Epoch: [25][ 50/ 1207] Overall Loss 0.372965 Objective Loss 0.372965 LR 0.001000 Time 0.033189 -2023-02-13 17:29:41,319 - Epoch: [25][ 60/ 1207] Overall Loss 0.372169 Objective Loss 0.372169 LR 0.001000 Time 0.030771 -2023-02-13 17:29:41,506 - Epoch: [25][ 70/ 1207] Overall Loss 0.372784 Objective Loss 0.372784 LR 0.001000 Time 0.029040 -2023-02-13 17:29:41,693 - Epoch: [25][ 80/ 1207] Overall Loss 0.371989 Objective Loss 0.371989 LR 0.001000 Time 0.027745 -2023-02-13 17:29:41,880 - Epoch: [25][ 90/ 1207] Overall Loss 0.368752 Objective Loss 0.368752 LR 0.001000 Time 0.026738 -2023-02-13 17:29:42,068 - Epoch: [25][ 100/ 1207] Overall Loss 0.368748 Objective Loss 0.368748 LR 0.001000 Time 0.025938 -2023-02-13 17:29:42,255 - Epoch: [25][ 110/ 1207] Overall Loss 0.368479 Objective Loss 0.368479 LR 0.001000 Time 0.025279 -2023-02-13 17:29:42,443 - Epoch: [25][ 120/ 1207] Overall Loss 0.369801 Objective Loss 0.369801 LR 0.001000 Time 0.024730 -2023-02-13 17:29:42,630 - Epoch: [25][ 130/ 1207] Overall Loss 0.368535 Objective Loss 0.368535 LR 0.001000 Time 0.024266 -2023-02-13 17:29:42,817 - Epoch: [25][ 140/ 1207] Overall Loss 0.369032 Objective Loss 0.369032 LR 0.001000 Time 0.023868 -2023-02-13 17:29:43,005 - Epoch: [25][ 150/ 1207] Overall Loss 0.368904 Objective Loss 0.368904 LR 0.001000 Time 0.023524 -2023-02-13 17:29:43,193 - Epoch: [25][ 160/ 1207] Overall Loss 0.366798 Objective Loss 0.366798 LR 0.001000 Time 0.023225 -2023-02-13 17:29:43,380 - Epoch: [25][ 170/ 1207] Overall Loss 0.368676 Objective Loss 0.368676 LR 0.001000 Time 0.022957 -2023-02-13 17:29:43,567 - Epoch: [25][ 180/ 1207] Overall Loss 0.369123 Objective Loss 0.369123 LR 0.001000 Time 0.022721 -2023-02-13 17:29:43,755 - Epoch: [25][ 190/ 1207] Overall Loss 0.370831 Objective Loss 0.370831 LR 0.001000 Time 0.022509 -2023-02-13 17:29:43,941 - Epoch: [25][ 200/ 1207] Overall Loss 0.371056 Objective Loss 0.371056 LR 0.001000 Time 0.022313 -2023-02-13 17:29:44,128 - Epoch: [25][ 210/ 1207] Overall Loss 0.370243 Objective Loss 0.370243 LR 0.001000 Time 0.022141 -2023-02-13 17:29:44,315 - Epoch: [25][ 220/ 1207] Overall Loss 0.370276 Objective Loss 0.370276 LR 0.001000 Time 0.021983 -2023-02-13 17:29:44,502 - Epoch: [25][ 230/ 1207] Overall Loss 0.369540 Objective Loss 0.369540 LR 0.001000 Time 0.021836 -2023-02-13 17:29:44,688 - Epoch: [25][ 240/ 1207] Overall Loss 0.370340 Objective Loss 0.370340 LR 0.001000 Time 0.021704 -2023-02-13 17:29:44,875 - Epoch: [25][ 250/ 1207] Overall Loss 0.370252 Objective Loss 0.370252 LR 0.001000 Time 0.021579 -2023-02-13 17:29:45,062 - Epoch: [25][ 260/ 1207] Overall Loss 0.371164 Objective Loss 0.371164 LR 0.001000 Time 0.021468 -2023-02-13 17:29:45,250 - Epoch: [25][ 270/ 1207] Overall Loss 0.371032 Objective Loss 0.371032 LR 0.001000 Time 0.021366 -2023-02-13 17:29:45,436 - Epoch: [25][ 280/ 1207] Overall Loss 0.370992 Objective Loss 0.370992 LR 0.001000 Time 0.021266 -2023-02-13 17:29:45,622 - Epoch: [25][ 290/ 1207] Overall Loss 0.370950 Objective Loss 0.370950 LR 0.001000 Time 0.021174 -2023-02-13 17:29:45,810 - Epoch: [25][ 300/ 1207] Overall Loss 0.370577 Objective Loss 0.370577 LR 0.001000 Time 0.021093 -2023-02-13 17:29:45,997 - Epoch: [25][ 310/ 1207] Overall Loss 0.369449 Objective Loss 0.369449 LR 0.001000 Time 0.021015 -2023-02-13 17:29:46,185 - Epoch: [25][ 320/ 1207] Overall Loss 0.368662 Objective Loss 0.368662 LR 0.001000 Time 0.020944 -2023-02-13 17:29:46,371 - Epoch: [25][ 330/ 1207] Overall Loss 0.367822 Objective Loss 0.367822 LR 0.001000 Time 0.020873 -2023-02-13 17:29:46,558 - Epoch: [25][ 340/ 1207] Overall Loss 0.366688 Objective Loss 0.366688 LR 0.001000 Time 0.020808 -2023-02-13 17:29:46,745 - Epoch: [25][ 350/ 1207] Overall Loss 0.366277 Objective Loss 0.366277 LR 0.001000 Time 0.020746 -2023-02-13 17:29:46,932 - Epoch: [25][ 360/ 1207] Overall Loss 0.365590 Objective Loss 0.365590 LR 0.001000 Time 0.020689 -2023-02-13 17:29:47,119 - Epoch: [25][ 370/ 1207] Overall Loss 0.366831 Objective Loss 0.366831 LR 0.001000 Time 0.020635 -2023-02-13 17:29:47,306 - Epoch: [25][ 380/ 1207] Overall Loss 0.367085 Objective Loss 0.367085 LR 0.001000 Time 0.020582 -2023-02-13 17:29:47,493 - Epoch: [25][ 390/ 1207] Overall Loss 0.367309 Objective Loss 0.367309 LR 0.001000 Time 0.020532 -2023-02-13 17:29:47,679 - Epoch: [25][ 400/ 1207] Overall Loss 0.367288 Objective Loss 0.367288 LR 0.001000 Time 0.020485 -2023-02-13 17:29:47,866 - Epoch: [25][ 410/ 1207] Overall Loss 0.367480 Objective Loss 0.367480 LR 0.001000 Time 0.020439 -2023-02-13 17:29:48,053 - Epoch: [25][ 420/ 1207] Overall Loss 0.367509 Objective Loss 0.367509 LR 0.001000 Time 0.020397 -2023-02-13 17:29:48,240 - Epoch: [25][ 430/ 1207] Overall Loss 0.367919 Objective Loss 0.367919 LR 0.001000 Time 0.020357 -2023-02-13 17:29:48,427 - Epoch: [25][ 440/ 1207] Overall Loss 0.368162 Objective Loss 0.368162 LR 0.001000 Time 0.020318 -2023-02-13 17:29:48,613 - Epoch: [25][ 450/ 1207] Overall Loss 0.368272 Objective Loss 0.368272 LR 0.001000 Time 0.020280 -2023-02-13 17:29:48,801 - Epoch: [25][ 460/ 1207] Overall Loss 0.368646 Objective Loss 0.368646 LR 0.001000 Time 0.020245 -2023-02-13 17:29:48,988 - Epoch: [25][ 470/ 1207] Overall Loss 0.369593 Objective Loss 0.369593 LR 0.001000 Time 0.020211 -2023-02-13 17:29:49,176 - Epoch: [25][ 480/ 1207] Overall Loss 0.370637 Objective Loss 0.370637 LR 0.001000 Time 0.020183 -2023-02-13 17:29:49,363 - Epoch: [25][ 490/ 1207] Overall Loss 0.370554 Objective Loss 0.370554 LR 0.001000 Time 0.020151 -2023-02-13 17:29:49,551 - Epoch: [25][ 500/ 1207] Overall Loss 0.371035 Objective Loss 0.371035 LR 0.001000 Time 0.020122 -2023-02-13 17:29:49,737 - Epoch: [25][ 510/ 1207] Overall Loss 0.371534 Objective Loss 0.371534 LR 0.001000 Time 0.020092 -2023-02-13 17:29:49,923 - Epoch: [25][ 520/ 1207] Overall Loss 0.372045 Objective Loss 0.372045 LR 0.001000 Time 0.020063 -2023-02-13 17:29:50,110 - Epoch: [25][ 530/ 1207] Overall Loss 0.371963 Objective Loss 0.371963 LR 0.001000 Time 0.020037 -2023-02-13 17:29:50,297 - Epoch: [25][ 540/ 1207] Overall Loss 0.372643 Objective Loss 0.372643 LR 0.001000 Time 0.020011 -2023-02-13 17:29:50,483 - Epoch: [25][ 550/ 1207] Overall Loss 0.372277 Objective Loss 0.372277 LR 0.001000 Time 0.019984 -2023-02-13 17:29:50,669 - Epoch: [25][ 560/ 1207] Overall Loss 0.372632 Objective Loss 0.372632 LR 0.001000 Time 0.019960 -2023-02-13 17:29:50,857 - Epoch: [25][ 570/ 1207] Overall Loss 0.373453 Objective Loss 0.373453 LR 0.001000 Time 0.019939 -2023-02-13 17:29:51,044 - Epoch: [25][ 580/ 1207] Overall Loss 0.373526 Objective Loss 0.373526 LR 0.001000 Time 0.019917 -2023-02-13 17:29:51,232 - Epoch: [25][ 590/ 1207] Overall Loss 0.373207 Objective Loss 0.373207 LR 0.001000 Time 0.019896 -2023-02-13 17:29:51,418 - Epoch: [25][ 600/ 1207] Overall Loss 0.373340 Objective Loss 0.373340 LR 0.001000 Time 0.019875 -2023-02-13 17:29:51,605 - Epoch: [25][ 610/ 1207] Overall Loss 0.373614 Objective Loss 0.373614 LR 0.001000 Time 0.019854 -2023-02-13 17:29:51,791 - Epoch: [25][ 620/ 1207] Overall Loss 0.373751 Objective Loss 0.373751 LR 0.001000 Time 0.019835 -2023-02-13 17:29:51,980 - Epoch: [25][ 630/ 1207] Overall Loss 0.374396 Objective Loss 0.374396 LR 0.001000 Time 0.019818 -2023-02-13 17:29:52,168 - Epoch: [25][ 640/ 1207] Overall Loss 0.374101 Objective Loss 0.374101 LR 0.001000 Time 0.019802 -2023-02-13 17:29:52,355 - Epoch: [25][ 650/ 1207] Overall Loss 0.373800 Objective Loss 0.373800 LR 0.001000 Time 0.019784 -2023-02-13 17:29:52,542 - Epoch: [25][ 660/ 1207] Overall Loss 0.373714 Objective Loss 0.373714 LR 0.001000 Time 0.019768 -2023-02-13 17:29:52,729 - Epoch: [25][ 670/ 1207] Overall Loss 0.373409 Objective Loss 0.373409 LR 0.001000 Time 0.019750 -2023-02-13 17:29:52,916 - Epoch: [25][ 680/ 1207] Overall Loss 0.373336 Objective Loss 0.373336 LR 0.001000 Time 0.019734 -2023-02-13 17:29:53,103 - Epoch: [25][ 690/ 1207] Overall Loss 0.373602 Objective Loss 0.373602 LR 0.001000 Time 0.019720 -2023-02-13 17:29:53,295 - Epoch: [25][ 700/ 1207] Overall Loss 0.373878 Objective Loss 0.373878 LR 0.001000 Time 0.019711 -2023-02-13 17:29:53,485 - Epoch: [25][ 710/ 1207] Overall Loss 0.373871 Objective Loss 0.373871 LR 0.001000 Time 0.019701 -2023-02-13 17:29:53,674 - Epoch: [25][ 720/ 1207] Overall Loss 0.373969 Objective Loss 0.373969 LR 0.001000 Time 0.019690 -2023-02-13 17:29:53,865 - Epoch: [25][ 730/ 1207] Overall Loss 0.374282 Objective Loss 0.374282 LR 0.001000 Time 0.019681 -2023-02-13 17:29:54,054 - Epoch: [25][ 740/ 1207] Overall Loss 0.374648 Objective Loss 0.374648 LR 0.001000 Time 0.019670 -2023-02-13 17:29:54,245 - Epoch: [25][ 750/ 1207] Overall Loss 0.374589 Objective Loss 0.374589 LR 0.001000 Time 0.019662 -2023-02-13 17:29:54,434 - Epoch: [25][ 760/ 1207] Overall Loss 0.374807 Objective Loss 0.374807 LR 0.001000 Time 0.019651 -2023-02-13 17:29:54,624 - Epoch: [25][ 770/ 1207] Overall Loss 0.375132 Objective Loss 0.375132 LR 0.001000 Time 0.019642 -2023-02-13 17:29:54,813 - Epoch: [25][ 780/ 1207] Overall Loss 0.375207 Objective Loss 0.375207 LR 0.001000 Time 0.019632 -2023-02-13 17:29:55,003 - Epoch: [25][ 790/ 1207] Overall Loss 0.375477 Objective Loss 0.375477 LR 0.001000 Time 0.019623 -2023-02-13 17:29:55,193 - Epoch: [25][ 800/ 1207] Overall Loss 0.375811 Objective Loss 0.375811 LR 0.001000 Time 0.019616 -2023-02-13 17:29:55,384 - Epoch: [25][ 810/ 1207] Overall Loss 0.375618 Objective Loss 0.375618 LR 0.001000 Time 0.019608 -2023-02-13 17:29:55,574 - Epoch: [25][ 820/ 1207] Overall Loss 0.375702 Objective Loss 0.375702 LR 0.001000 Time 0.019600 -2023-02-13 17:29:55,764 - Epoch: [25][ 830/ 1207] Overall Loss 0.375479 Objective Loss 0.375479 LR 0.001000 Time 0.019593 -2023-02-13 17:29:55,952 - Epoch: [25][ 840/ 1207] Overall Loss 0.375696 Objective Loss 0.375696 LR 0.001000 Time 0.019583 -2023-02-13 17:29:56,139 - Epoch: [25][ 850/ 1207] Overall Loss 0.376374 Objective Loss 0.376374 LR 0.001000 Time 0.019572 -2023-02-13 17:29:56,327 - Epoch: [25][ 860/ 1207] Overall Loss 0.376623 Objective Loss 0.376623 LR 0.001000 Time 0.019562 -2023-02-13 17:29:56,513 - Epoch: [25][ 870/ 1207] Overall Loss 0.376594 Objective Loss 0.376594 LR 0.001000 Time 0.019551 -2023-02-13 17:29:56,700 - Epoch: [25][ 880/ 1207] Overall Loss 0.377168 Objective Loss 0.377168 LR 0.001000 Time 0.019541 -2023-02-13 17:29:56,887 - Epoch: [25][ 890/ 1207] Overall Loss 0.377062 Objective Loss 0.377062 LR 0.001000 Time 0.019532 -2023-02-13 17:29:57,075 - Epoch: [25][ 900/ 1207] Overall Loss 0.376659 Objective Loss 0.376659 LR 0.001000 Time 0.019523 -2023-02-13 17:29:57,263 - Epoch: [25][ 910/ 1207] Overall Loss 0.376831 Objective Loss 0.376831 LR 0.001000 Time 0.019515 -2023-02-13 17:29:57,451 - Epoch: [25][ 920/ 1207] Overall Loss 0.376897 Objective Loss 0.376897 LR 0.001000 Time 0.019506 -2023-02-13 17:29:57,638 - Epoch: [25][ 930/ 1207] Overall Loss 0.377067 Objective Loss 0.377067 LR 0.001000 Time 0.019497 -2023-02-13 17:29:57,825 - Epoch: [25][ 940/ 1207] Overall Loss 0.376880 Objective Loss 0.376880 LR 0.001000 Time 0.019489 -2023-02-13 17:29:58,013 - Epoch: [25][ 950/ 1207] Overall Loss 0.376999 Objective Loss 0.376999 LR 0.001000 Time 0.019480 -2023-02-13 17:29:58,201 - Epoch: [25][ 960/ 1207] Overall Loss 0.377004 Objective Loss 0.377004 LR 0.001000 Time 0.019473 -2023-02-13 17:29:58,388 - Epoch: [25][ 970/ 1207] Overall Loss 0.377371 Objective Loss 0.377371 LR 0.001000 Time 0.019465 -2023-02-13 17:29:58,576 - Epoch: [25][ 980/ 1207] Overall Loss 0.377495 Objective Loss 0.377495 LR 0.001000 Time 0.019457 -2023-02-13 17:29:58,762 - Epoch: [25][ 990/ 1207] Overall Loss 0.377671 Objective Loss 0.377671 LR 0.001000 Time 0.019449 -2023-02-13 17:29:58,950 - Epoch: [25][ 1000/ 1207] Overall Loss 0.377899 Objective Loss 0.377899 LR 0.001000 Time 0.019442 -2023-02-13 17:29:59,138 - Epoch: [25][ 1010/ 1207] Overall Loss 0.377846 Objective Loss 0.377846 LR 0.001000 Time 0.019435 -2023-02-13 17:29:59,327 - Epoch: [25][ 1020/ 1207] Overall Loss 0.377994 Objective Loss 0.377994 LR 0.001000 Time 0.019429 -2023-02-13 17:29:59,514 - Epoch: [25][ 1030/ 1207] Overall Loss 0.378176 Objective Loss 0.378176 LR 0.001000 Time 0.019422 -2023-02-13 17:29:59,701 - Epoch: [25][ 1040/ 1207] Overall Loss 0.378429 Objective Loss 0.378429 LR 0.001000 Time 0.019415 -2023-02-13 17:29:59,888 - Epoch: [25][ 1050/ 1207] Overall Loss 0.378576 Objective Loss 0.378576 LR 0.001000 Time 0.019408 -2023-02-13 17:30:00,076 - Epoch: [25][ 1060/ 1207] Overall Loss 0.378661 Objective Loss 0.378661 LR 0.001000 Time 0.019401 -2023-02-13 17:30:00,264 - Epoch: [25][ 1070/ 1207] Overall Loss 0.378612 Objective Loss 0.378612 LR 0.001000 Time 0.019395 -2023-02-13 17:30:00,451 - Epoch: [25][ 1080/ 1207] Overall Loss 0.378299 Objective Loss 0.378299 LR 0.001000 Time 0.019389 -2023-02-13 17:30:00,639 - Epoch: [25][ 1090/ 1207] Overall Loss 0.378659 Objective Loss 0.378659 LR 0.001000 Time 0.019383 -2023-02-13 17:30:00,827 - Epoch: [25][ 1100/ 1207] Overall Loss 0.378440 Objective Loss 0.378440 LR 0.001000 Time 0.019377 -2023-02-13 17:30:01,015 - Epoch: [25][ 1110/ 1207] Overall Loss 0.378546 Objective Loss 0.378546 LR 0.001000 Time 0.019371 -2023-02-13 17:30:01,205 - Epoch: [25][ 1120/ 1207] Overall Loss 0.378696 Objective Loss 0.378696 LR 0.001000 Time 0.019368 -2023-02-13 17:30:01,392 - Epoch: [25][ 1130/ 1207] Overall Loss 0.378715 Objective Loss 0.378715 LR 0.001000 Time 0.019362 -2023-02-13 17:30:01,580 - Epoch: [25][ 1140/ 1207] Overall Loss 0.378541 Objective Loss 0.378541 LR 0.001000 Time 0.019357 -2023-02-13 17:30:01,768 - Epoch: [25][ 1150/ 1207] Overall Loss 0.378622 Objective Loss 0.378622 LR 0.001000 Time 0.019352 -2023-02-13 17:30:01,955 - Epoch: [25][ 1160/ 1207] Overall Loss 0.378549 Objective Loss 0.378549 LR 0.001000 Time 0.019346 -2023-02-13 17:30:02,143 - Epoch: [25][ 1170/ 1207] Overall Loss 0.378845 Objective Loss 0.378845 LR 0.001000 Time 0.019341 -2023-02-13 17:30:02,330 - Epoch: [25][ 1180/ 1207] Overall Loss 0.378942 Objective Loss 0.378942 LR 0.001000 Time 0.019335 -2023-02-13 17:30:02,518 - Epoch: [25][ 1190/ 1207] Overall Loss 0.378951 Objective Loss 0.378951 LR 0.001000 Time 0.019330 -2023-02-13 17:30:02,761 - Epoch: [25][ 1200/ 1207] Overall Loss 0.378955 Objective Loss 0.378955 LR 0.001000 Time 0.019371 -2023-02-13 17:30:02,876 - Epoch: [25][ 1207/ 1207] Overall Loss 0.378806 Objective Loss 0.378806 Top1 82.621951 Top5 98.475610 LR 0.001000 Time 0.019354 -2023-02-13 17:30:02,947 - --- validate (epoch=25)----------- -2023-02-13 17:30:02,947 - 34311 samples (256 per mini-batch) -2023-02-13 17:30:03,352 - Epoch: [25][ 10/ 135] Loss 0.381751 Top1 78.320312 Top5 96.875000 -2023-02-13 17:30:03,479 - Epoch: [25][ 20/ 135] Loss 0.383831 Top1 78.515625 Top5 96.757812 -2023-02-13 17:30:03,606 - Epoch: [25][ 30/ 135] Loss 0.395912 Top1 78.281250 Top5 96.653646 -2023-02-13 17:30:03,732 - Epoch: [25][ 40/ 135] Loss 0.395559 Top1 78.535156 Top5 96.650391 -2023-02-13 17:30:03,858 - Epoch: [25][ 50/ 135] Loss 0.397519 Top1 78.632812 Top5 96.601562 -2023-02-13 17:30:03,986 - Epoch: [25][ 60/ 135] Loss 0.393643 Top1 78.626302 Top5 96.588542 -2023-02-13 17:30:04,112 - Epoch: [25][ 70/ 135] Loss 0.396025 Top1 78.610491 Top5 96.612723 -2023-02-13 17:30:04,239 - Epoch: [25][ 80/ 135] Loss 0.391779 Top1 78.710938 Top5 96.674805 -2023-02-13 17:30:04,366 - Epoch: [25][ 90/ 135] Loss 0.392276 Top1 78.771701 Top5 96.697049 -2023-02-13 17:30:04,497 - Epoch: [25][ 100/ 135] Loss 0.394702 Top1 78.703125 Top5 96.628906 -2023-02-13 17:30:04,625 - Epoch: [25][ 110/ 135] Loss 0.394646 Top1 78.686080 Top5 96.629972 -2023-02-13 17:30:04,756 - Epoch: [25][ 120/ 135] Loss 0.394061 Top1 78.613281 Top5 96.608073 -2023-02-13 17:30:04,887 - Epoch: [25][ 130/ 135] Loss 0.395362 Top1 78.566707 Top5 96.601562 -2023-02-13 17:30:04,935 - Epoch: [25][ 135/ 135] Loss 0.395471 Top1 78.610358 Top5 96.607502 -2023-02-13 17:30:05,002 - ==> Top1: 78.610 Top5: 96.608 Loss: 0.395 - -2023-02-13 17:30:05,003 - ==> Confusion: -[[ 848 6 3 0 16 5 0 2 4 51 0 6 1 4 7 1 7 3 1 0 2] - [ 3 912 2 5 12 34 1 32 5 1 1 2 1 1 1 1 4 1 8 2 4] - [ 12 8 945 9 4 2 17 27 1 0 2 2 2 1 3 5 3 1 11 1 2] - [ 8 3 20 889 0 2 1 4 2 1 7 0 5 1 22 3 4 5 34 0 5] - [ 28 9 0 0 979 9 0 0 0 4 0 3 3 2 4 7 10 2 0 2 4] - [ 6 23 1 5 9 935 3 29 1 6 1 11 8 12 1 4 2 2 4 4 3] - [ 6 6 29 5 0 4 1002 14 1 0 0 3 1 0 0 9 2 5 2 9 1] - [ 2 8 6 3 2 15 1 940 0 1 0 3 5 0 1 1 0 2 25 8 1] - [ 29 3 0 1 1 0 0 0 896 34 8 1 0 8 17 2 0 3 6 0 0] - [ 129 4 1 0 9 1 0 1 41 789 1 4 2 16 8 1 0 1 1 1 2] - [ 4 4 11 8 0 4 6 5 24 2 938 4 2 6 3 0 3 0 24 0 3] - [ 5 3 0 0 1 10 1 9 2 2 0 889 33 8 2 4 3 11 4 17 1] - [ 3 0 2 5 3 3 0 3 1 0 0 36 858 0 5 3 2 24 1 2 8] - [ 11 2 3 0 5 20 2 2 23 19 13 13 4 879 7 4 6 2 1 4 4] - [ 18 4 3 18 11 1 0 0 31 4 0 1 4 0 954 3 4 6 24 0 6] - [ 7 4 6 2 6 2 7 1 0 0 0 6 12 5 2 941 19 17 0 6 3] - [ 5 14 0 3 8 1 1 1 1 1 0 4 1 1 0 16 995 2 1 2 4] - [ 11 2 0 7 2 0 3 2 1 1 0 9 30 1 4 17 0 957 0 0 4] - [ 3 2 5 5 2 1 0 45 8 1 1 3 7 0 9 2 1 1 990 0 0] - [ 1 5 0 0 1 14 10 35 0 0 1 18 6 4 0 5 3 4 3 1030 8] - [ 297 315 329 174 211 260 106 318 168 121 154 148 414 327 231 152 476 128 341 358 8406]] - -2023-02-13 17:30:05,004 - ==> Best [Top1: 79.718 Top5: 96.756 Sparsity:0.00 Params: 148928 on epoch: 22] -2023-02-13 17:30:05,004 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:30:05,010 - - -2023-02-13 17:30:05,010 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:30:05,895 - Epoch: [26][ 10/ 1207] Overall Loss 0.379697 Objective Loss 0.379697 LR 0.001000 Time 0.088385 -2023-02-13 17:30:06,092 - Epoch: [26][ 20/ 1207] Overall Loss 0.371745 Objective Loss 0.371745 LR 0.001000 Time 0.054054 -2023-02-13 17:30:06,286 - Epoch: [26][ 30/ 1207] Overall Loss 0.365783 Objective Loss 0.365783 LR 0.001000 Time 0.042472 -2023-02-13 17:30:06,480 - Epoch: [26][ 40/ 1207] Overall Loss 0.363156 Objective Loss 0.363156 LR 0.001000 Time 0.036702 -2023-02-13 17:30:06,673 - Epoch: [26][ 50/ 1207] Overall Loss 0.361394 Objective Loss 0.361394 LR 0.001000 Time 0.033206 -2023-02-13 17:30:06,867 - Epoch: [26][ 60/ 1207] Overall Loss 0.367643 Objective Loss 0.367643 LR 0.001000 Time 0.030907 -2023-02-13 17:30:07,060 - Epoch: [26][ 70/ 1207] Overall Loss 0.374213 Objective Loss 0.374213 LR 0.001000 Time 0.029241 -2023-02-13 17:30:07,254 - Epoch: [26][ 80/ 1207] Overall Loss 0.376104 Objective Loss 0.376104 LR 0.001000 Time 0.028008 -2023-02-13 17:30:07,446 - Epoch: [26][ 90/ 1207] Overall Loss 0.373580 Objective Loss 0.373580 LR 0.001000 Time 0.027027 -2023-02-13 17:30:07,640 - Epoch: [26][ 100/ 1207] Overall Loss 0.372930 Objective Loss 0.372930 LR 0.001000 Time 0.026258 -2023-02-13 17:30:07,832 - Epoch: [26][ 110/ 1207] Overall Loss 0.371823 Objective Loss 0.371823 LR 0.001000 Time 0.025613 -2023-02-13 17:30:08,027 - Epoch: [26][ 120/ 1207] Overall Loss 0.371176 Objective Loss 0.371176 LR 0.001000 Time 0.025099 -2023-02-13 17:30:08,221 - Epoch: [26][ 130/ 1207] Overall Loss 0.369510 Objective Loss 0.369510 LR 0.001000 Time 0.024656 -2023-02-13 17:30:08,415 - Epoch: [26][ 140/ 1207] Overall Loss 0.367996 Objective Loss 0.367996 LR 0.001000 Time 0.024280 -2023-02-13 17:30:08,609 - Epoch: [26][ 150/ 1207] Overall Loss 0.367570 Objective Loss 0.367570 LR 0.001000 Time 0.023952 -2023-02-13 17:30:08,804 - Epoch: [26][ 160/ 1207] Overall Loss 0.368545 Objective Loss 0.368545 LR 0.001000 Time 0.023672 -2023-02-13 17:30:08,997 - Epoch: [26][ 170/ 1207] Overall Loss 0.366563 Objective Loss 0.366563 LR 0.001000 Time 0.023415 -2023-02-13 17:30:09,193 - Epoch: [26][ 180/ 1207] Overall Loss 0.369317 Objective Loss 0.369317 LR 0.001000 Time 0.023196 -2023-02-13 17:30:09,386 - Epoch: [26][ 190/ 1207] Overall Loss 0.369656 Objective Loss 0.369656 LR 0.001000 Time 0.022993 -2023-02-13 17:30:09,581 - Epoch: [26][ 200/ 1207] Overall Loss 0.371114 Objective Loss 0.371114 LR 0.001000 Time 0.022814 -2023-02-13 17:30:09,773 - Epoch: [26][ 210/ 1207] Overall Loss 0.371867 Objective Loss 0.371867 LR 0.001000 Time 0.022642 -2023-02-13 17:30:09,968 - Epoch: [26][ 220/ 1207] Overall Loss 0.372157 Objective Loss 0.372157 LR 0.001000 Time 0.022497 -2023-02-13 17:30:10,162 - Epoch: [26][ 230/ 1207] Overall Loss 0.371697 Objective Loss 0.371697 LR 0.001000 Time 0.022359 -2023-02-13 17:30:10,358 - Epoch: [26][ 240/ 1207] Overall Loss 0.371052 Objective Loss 0.371052 LR 0.001000 Time 0.022243 -2023-02-13 17:30:10,550 - Epoch: [26][ 250/ 1207] Overall Loss 0.371377 Objective Loss 0.371377 LR 0.001000 Time 0.022120 -2023-02-13 17:30:10,745 - Epoch: [26][ 260/ 1207] Overall Loss 0.371249 Objective Loss 0.371249 LR 0.001000 Time 0.022020 -2023-02-13 17:30:10,939 - Epoch: [26][ 270/ 1207] Overall Loss 0.371849 Objective Loss 0.371849 LR 0.001000 Time 0.021919 -2023-02-13 17:30:11,134 - Epoch: [26][ 280/ 1207] Overall Loss 0.370298 Objective Loss 0.370298 LR 0.001000 Time 0.021833 -2023-02-13 17:30:11,327 - Epoch: [26][ 290/ 1207] Overall Loss 0.369131 Objective Loss 0.369131 LR 0.001000 Time 0.021744 -2023-02-13 17:30:11,522 - Epoch: [26][ 300/ 1207] Overall Loss 0.369736 Objective Loss 0.369736 LR 0.001000 Time 0.021670 -2023-02-13 17:30:11,715 - Epoch: [26][ 310/ 1207] Overall Loss 0.369617 Objective Loss 0.369617 LR 0.001000 Time 0.021590 -2023-02-13 17:30:11,911 - Epoch: [26][ 320/ 1207] Overall Loss 0.369100 Objective Loss 0.369100 LR 0.001000 Time 0.021526 -2023-02-13 17:30:12,103 - Epoch: [26][ 330/ 1207] Overall Loss 0.369445 Objective Loss 0.369445 LR 0.001000 Time 0.021456 -2023-02-13 17:30:12,299 - Epoch: [26][ 340/ 1207] Overall Loss 0.370311 Objective Loss 0.370311 LR 0.001000 Time 0.021399 -2023-02-13 17:30:12,491 - Epoch: [26][ 350/ 1207] Overall Loss 0.371381 Objective Loss 0.371381 LR 0.001000 Time 0.021335 -2023-02-13 17:30:12,685 - Epoch: [26][ 360/ 1207] Overall Loss 0.372042 Objective Loss 0.372042 LR 0.001000 Time 0.021282 -2023-02-13 17:30:12,877 - Epoch: [26][ 370/ 1207] Overall Loss 0.372081 Objective Loss 0.372081 LR 0.001000 Time 0.021225 -2023-02-13 17:30:13,073 - Epoch: [26][ 380/ 1207] Overall Loss 0.372586 Objective Loss 0.372586 LR 0.001000 Time 0.021181 -2023-02-13 17:30:13,266 - Epoch: [26][ 390/ 1207] Overall Loss 0.372812 Objective Loss 0.372812 LR 0.001000 Time 0.021132 -2023-02-13 17:30:13,462 - Epoch: [26][ 400/ 1207] Overall Loss 0.372834 Objective Loss 0.372834 LR 0.001000 Time 0.021092 -2023-02-13 17:30:13,654 - Epoch: [26][ 410/ 1207] Overall Loss 0.372383 Objective Loss 0.372383 LR 0.001000 Time 0.021044 -2023-02-13 17:30:13,848 - Epoch: [26][ 420/ 1207] Overall Loss 0.372573 Objective Loss 0.372573 LR 0.001000 Time 0.021005 -2023-02-13 17:30:14,040 - Epoch: [26][ 430/ 1207] Overall Loss 0.373023 Objective Loss 0.373023 LR 0.001000 Time 0.020962 -2023-02-13 17:30:14,237 - Epoch: [26][ 440/ 1207] Overall Loss 0.373393 Objective Loss 0.373393 LR 0.001000 Time 0.020932 -2023-02-13 17:30:14,429 - Epoch: [26][ 450/ 1207] Overall Loss 0.373812 Objective Loss 0.373812 LR 0.001000 Time 0.020894 -2023-02-13 17:30:14,625 - Epoch: [26][ 460/ 1207] Overall Loss 0.373516 Objective Loss 0.373516 LR 0.001000 Time 0.020864 -2023-02-13 17:30:14,817 - Epoch: [26][ 470/ 1207] Overall Loss 0.373572 Objective Loss 0.373572 LR 0.001000 Time 0.020828 -2023-02-13 17:30:15,012 - Epoch: [26][ 480/ 1207] Overall Loss 0.373122 Objective Loss 0.373122 LR 0.001000 Time 0.020800 -2023-02-13 17:30:15,205 - Epoch: [26][ 490/ 1207] Overall Loss 0.372934 Objective Loss 0.372934 LR 0.001000 Time 0.020768 -2023-02-13 17:30:15,401 - Epoch: [26][ 500/ 1207] Overall Loss 0.373025 Objective Loss 0.373025 LR 0.001000 Time 0.020744 -2023-02-13 17:30:15,593 - Epoch: [26][ 510/ 1207] Overall Loss 0.373292 Objective Loss 0.373292 LR 0.001000 Time 0.020712 -2023-02-13 17:30:15,789 - Epoch: [26][ 520/ 1207] Overall Loss 0.372880 Objective Loss 0.372880 LR 0.001000 Time 0.020691 -2023-02-13 17:30:15,981 - Epoch: [26][ 530/ 1207] Overall Loss 0.372947 Objective Loss 0.372947 LR 0.001000 Time 0.020662 -2023-02-13 17:30:16,178 - Epoch: [26][ 540/ 1207] Overall Loss 0.373126 Objective Loss 0.373126 LR 0.001000 Time 0.020644 -2023-02-13 17:30:16,373 - Epoch: [26][ 550/ 1207] Overall Loss 0.372936 Objective Loss 0.372936 LR 0.001000 Time 0.020623 -2023-02-13 17:30:16,570 - Epoch: [26][ 560/ 1207] Overall Loss 0.372625 Objective Loss 0.372625 LR 0.001000 Time 0.020606 -2023-02-13 17:30:16,766 - Epoch: [26][ 570/ 1207] Overall Loss 0.372911 Objective Loss 0.372911 LR 0.001000 Time 0.020586 -2023-02-13 17:30:16,963 - Epoch: [26][ 580/ 1207] Overall Loss 0.372435 Objective Loss 0.372435 LR 0.001000 Time 0.020571 -2023-02-13 17:30:17,158 - Epoch: [26][ 590/ 1207] Overall Loss 0.372444 Objective Loss 0.372444 LR 0.001000 Time 0.020552 -2023-02-13 17:30:17,354 - Epoch: [26][ 600/ 1207] Overall Loss 0.372466 Objective Loss 0.372466 LR 0.001000 Time 0.020536 -2023-02-13 17:30:17,546 - Epoch: [26][ 610/ 1207] Overall Loss 0.372468 Objective Loss 0.372468 LR 0.001000 Time 0.020513 -2023-02-13 17:30:17,742 - Epoch: [26][ 620/ 1207] Overall Loss 0.372747 Objective Loss 0.372747 LR 0.001000 Time 0.020497 -2023-02-13 17:30:17,934 - Epoch: [26][ 630/ 1207] Overall Loss 0.373037 Objective Loss 0.373037 LR 0.001000 Time 0.020476 -2023-02-13 17:30:18,129 - Epoch: [26][ 640/ 1207] Overall Loss 0.372719 Objective Loss 0.372719 LR 0.001000 Time 0.020461 -2023-02-13 17:30:18,322 - Epoch: [26][ 650/ 1207] Overall Loss 0.372582 Objective Loss 0.372582 LR 0.001000 Time 0.020443 -2023-02-13 17:30:18,518 - Epoch: [26][ 660/ 1207] Overall Loss 0.372962 Objective Loss 0.372962 LR 0.001000 Time 0.020429 -2023-02-13 17:30:18,711 - Epoch: [26][ 670/ 1207] Overall Loss 0.372820 Objective Loss 0.372820 LR 0.001000 Time 0.020411 -2023-02-13 17:30:18,906 - Epoch: [26][ 680/ 1207] Overall Loss 0.373306 Objective Loss 0.373306 LR 0.001000 Time 0.020398 -2023-02-13 17:30:19,099 - Epoch: [26][ 690/ 1207] Overall Loss 0.373197 Objective Loss 0.373197 LR 0.001000 Time 0.020382 -2023-02-13 17:30:19,296 - Epoch: [26][ 700/ 1207] Overall Loss 0.373397 Objective Loss 0.373397 LR 0.001000 Time 0.020370 -2023-02-13 17:30:19,488 - Epoch: [26][ 710/ 1207] Overall Loss 0.373544 Objective Loss 0.373544 LR 0.001000 Time 0.020353 -2023-02-13 17:30:19,682 - Epoch: [26][ 720/ 1207] Overall Loss 0.373748 Objective Loss 0.373748 LR 0.001000 Time 0.020340 -2023-02-13 17:30:19,874 - Epoch: [26][ 730/ 1207] Overall Loss 0.374262 Objective Loss 0.374262 LR 0.001000 Time 0.020324 -2023-02-13 17:30:20,070 - Epoch: [26][ 740/ 1207] Overall Loss 0.374462 Objective Loss 0.374462 LR 0.001000 Time 0.020313 -2023-02-13 17:30:20,262 - Epoch: [26][ 750/ 1207] Overall Loss 0.374683 Objective Loss 0.374683 LR 0.001000 Time 0.020298 -2023-02-13 17:30:20,457 - Epoch: [26][ 760/ 1207] Overall Loss 0.374880 Objective Loss 0.374880 LR 0.001000 Time 0.020288 -2023-02-13 17:30:20,649 - Epoch: [26][ 770/ 1207] Overall Loss 0.375072 Objective Loss 0.375072 LR 0.001000 Time 0.020273 -2023-02-13 17:30:20,846 - Epoch: [26][ 780/ 1207] Overall Loss 0.375074 Objective Loss 0.375074 LR 0.001000 Time 0.020265 -2023-02-13 17:30:21,039 - Epoch: [26][ 790/ 1207] Overall Loss 0.374743 Objective Loss 0.374743 LR 0.001000 Time 0.020252 -2023-02-13 17:30:21,235 - Epoch: [26][ 800/ 1207] Overall Loss 0.375142 Objective Loss 0.375142 LR 0.001000 Time 0.020243 -2023-02-13 17:30:21,427 - Epoch: [26][ 810/ 1207] Overall Loss 0.374978 Objective Loss 0.374978 LR 0.001000 Time 0.020231 -2023-02-13 17:30:21,623 - Epoch: [26][ 820/ 1207] Overall Loss 0.374999 Objective Loss 0.374999 LR 0.001000 Time 0.020221 -2023-02-13 17:30:21,815 - Epoch: [26][ 830/ 1207] Overall Loss 0.375010 Objective Loss 0.375010 LR 0.001000 Time 0.020209 -2023-02-13 17:30:22,010 - Epoch: [26][ 840/ 1207] Overall Loss 0.374530 Objective Loss 0.374530 LR 0.001000 Time 0.020201 -2023-02-13 17:30:22,204 - Epoch: [26][ 850/ 1207] Overall Loss 0.374239 Objective Loss 0.374239 LR 0.001000 Time 0.020190 -2023-02-13 17:30:22,400 - Epoch: [26][ 860/ 1207] Overall Loss 0.374143 Objective Loss 0.374143 LR 0.001000 Time 0.020184 -2023-02-13 17:30:22,593 - Epoch: [26][ 870/ 1207] Overall Loss 0.374020 Objective Loss 0.374020 LR 0.001000 Time 0.020173 -2023-02-13 17:30:22,788 - Epoch: [26][ 880/ 1207] Overall Loss 0.374241 Objective Loss 0.374241 LR 0.001000 Time 0.020165 -2023-02-13 17:30:22,980 - Epoch: [26][ 890/ 1207] Overall Loss 0.374006 Objective Loss 0.374006 LR 0.001000 Time 0.020153 -2023-02-13 17:30:23,176 - Epoch: [26][ 900/ 1207] Overall Loss 0.373776 Objective Loss 0.373776 LR 0.001000 Time 0.020146 -2023-02-13 17:30:23,368 - Epoch: [26][ 910/ 1207] Overall Loss 0.373759 Objective Loss 0.373759 LR 0.001000 Time 0.020136 -2023-02-13 17:30:23,563 - Epoch: [26][ 920/ 1207] Overall Loss 0.373874 Objective Loss 0.373874 LR 0.001000 Time 0.020129 -2023-02-13 17:30:23,756 - Epoch: [26][ 930/ 1207] Overall Loss 0.373949 Objective Loss 0.373949 LR 0.001000 Time 0.020119 -2023-02-13 17:30:23,951 - Epoch: [26][ 940/ 1207] Overall Loss 0.374383 Objective Loss 0.374383 LR 0.001000 Time 0.020113 -2023-02-13 17:30:24,143 - Epoch: [26][ 950/ 1207] Overall Loss 0.374178 Objective Loss 0.374178 LR 0.001000 Time 0.020103 -2023-02-13 17:30:24,340 - Epoch: [26][ 960/ 1207] Overall Loss 0.373964 Objective Loss 0.373964 LR 0.001000 Time 0.020098 -2023-02-13 17:30:24,532 - Epoch: [26][ 970/ 1207] Overall Loss 0.373957 Objective Loss 0.373957 LR 0.001000 Time 0.020088 -2023-02-13 17:30:24,728 - Epoch: [26][ 980/ 1207] Overall Loss 0.374222 Objective Loss 0.374222 LR 0.001000 Time 0.020082 -2023-02-13 17:30:24,920 - Epoch: [26][ 990/ 1207] Overall Loss 0.374167 Objective Loss 0.374167 LR 0.001000 Time 0.020073 -2023-02-13 17:30:25,116 - Epoch: [26][ 1000/ 1207] Overall Loss 0.374191 Objective Loss 0.374191 LR 0.001000 Time 0.020069 -2023-02-13 17:30:25,310 - Epoch: [26][ 1010/ 1207] Overall Loss 0.373894 Objective Loss 0.373894 LR 0.001000 Time 0.020062 -2023-02-13 17:30:25,506 - Epoch: [26][ 1020/ 1207] Overall Loss 0.373944 Objective Loss 0.373944 LR 0.001000 Time 0.020056 -2023-02-13 17:30:25,697 - Epoch: [26][ 1030/ 1207] Overall Loss 0.373816 Objective Loss 0.373816 LR 0.001000 Time 0.020047 -2023-02-13 17:30:25,894 - Epoch: [26][ 1040/ 1207] Overall Loss 0.373529 Objective Loss 0.373529 LR 0.001000 Time 0.020044 -2023-02-13 17:30:26,087 - Epoch: [26][ 1050/ 1207] Overall Loss 0.373290 Objective Loss 0.373290 LR 0.001000 Time 0.020036 -2023-02-13 17:30:26,283 - Epoch: [26][ 1060/ 1207] Overall Loss 0.373070 Objective Loss 0.373070 LR 0.001000 Time 0.020031 -2023-02-13 17:30:26,475 - Epoch: [26][ 1070/ 1207] Overall Loss 0.373225 Objective Loss 0.373225 LR 0.001000 Time 0.020024 -2023-02-13 17:30:26,672 - Epoch: [26][ 1080/ 1207] Overall Loss 0.373351 Objective Loss 0.373351 LR 0.001000 Time 0.020020 -2023-02-13 17:30:26,865 - Epoch: [26][ 1090/ 1207] Overall Loss 0.373343 Objective Loss 0.373343 LR 0.001000 Time 0.020013 -2023-02-13 17:30:27,061 - Epoch: [26][ 1100/ 1207] Overall Loss 0.373220 Objective Loss 0.373220 LR 0.001000 Time 0.020009 -2023-02-13 17:30:27,254 - Epoch: [26][ 1110/ 1207] Overall Loss 0.373180 Objective Loss 0.373180 LR 0.001000 Time 0.020002 -2023-02-13 17:30:27,449 - Epoch: [26][ 1120/ 1207] Overall Loss 0.373139 Objective Loss 0.373139 LR 0.001000 Time 0.019998 -2023-02-13 17:30:27,641 - Epoch: [26][ 1130/ 1207] Overall Loss 0.373196 Objective Loss 0.373196 LR 0.001000 Time 0.019990 -2023-02-13 17:30:27,837 - Epoch: [26][ 1140/ 1207] Overall Loss 0.373334 Objective Loss 0.373334 LR 0.001000 Time 0.019986 -2023-02-13 17:30:28,029 - Epoch: [26][ 1150/ 1207] Overall Loss 0.373314 Objective Loss 0.373314 LR 0.001000 Time 0.019979 -2023-02-13 17:30:28,226 - Epoch: [26][ 1160/ 1207] Overall Loss 0.373160 Objective Loss 0.373160 LR 0.001000 Time 0.019976 -2023-02-13 17:30:28,419 - Epoch: [26][ 1170/ 1207] Overall Loss 0.373219 Objective Loss 0.373219 LR 0.001000 Time 0.019970 -2023-02-13 17:30:28,615 - Epoch: [26][ 1180/ 1207] Overall Loss 0.373210 Objective Loss 0.373210 LR 0.001000 Time 0.019966 -2023-02-13 17:30:28,808 - Epoch: [26][ 1190/ 1207] Overall Loss 0.373373 Objective Loss 0.373373 LR 0.001000 Time 0.019961 -2023-02-13 17:30:29,055 - Epoch: [26][ 1200/ 1207] Overall Loss 0.373439 Objective Loss 0.373439 LR 0.001000 Time 0.020000 -2023-02-13 17:30:29,171 - Epoch: [26][ 1207/ 1207] Overall Loss 0.373576 Objective Loss 0.373576 Top1 82.012195 Top5 97.256098 LR 0.001000 Time 0.019980 -2023-02-13 17:30:29,242 - --- validate (epoch=26)----------- -2023-02-13 17:30:29,243 - 34311 samples (256 per mini-batch) -2023-02-13 17:30:29,633 - Epoch: [26][ 10/ 135] Loss 0.398395 Top1 80.156250 Top5 96.914062 -2023-02-13 17:30:29,761 - Epoch: [26][ 20/ 135] Loss 0.383098 Top1 80.429688 Top5 96.914062 -2023-02-13 17:30:29,893 - Epoch: [26][ 30/ 135] Loss 0.386885 Top1 80.195312 Top5 96.966146 -2023-02-13 17:30:30,024 - Epoch: [26][ 40/ 135] Loss 0.391687 Top1 80.107422 Top5 96.943359 -2023-02-13 17:30:30,151 - Epoch: [26][ 50/ 135] Loss 0.391873 Top1 80.156250 Top5 96.875000 -2023-02-13 17:30:30,276 - Epoch: [26][ 60/ 135] Loss 0.389471 Top1 80.351562 Top5 96.979167 -2023-02-13 17:30:30,406 - Epoch: [26][ 70/ 135] Loss 0.387897 Top1 80.256696 Top5 96.925223 -2023-02-13 17:30:30,533 - Epoch: [26][ 80/ 135] Loss 0.391116 Top1 80.375977 Top5 96.909180 -2023-02-13 17:30:30,657 - Epoch: [26][ 90/ 135] Loss 0.392217 Top1 80.295139 Top5 96.931424 -2023-02-13 17:30:30,786 - Epoch: [26][ 100/ 135] Loss 0.391191 Top1 80.375000 Top5 96.953125 -2023-02-13 17:30:30,913 - Epoch: [26][ 110/ 135] Loss 0.391425 Top1 80.280540 Top5 96.910511 -2023-02-13 17:30:31,039 - Epoch: [26][ 120/ 135] Loss 0.390750 Top1 80.309245 Top5 96.904297 -2023-02-13 17:30:31,170 - Epoch: [26][ 130/ 135] Loss 0.391579 Top1 80.198317 Top5 96.850962 -2023-02-13 17:30:31,217 - Epoch: [26][ 135/ 135] Loss 0.389581 Top1 80.192941 Top5 96.846492 -2023-02-13 17:30:31,285 - ==> Top1: 80.193 Top5: 96.846 Loss: 0.390 - -2023-02-13 17:30:31,286 - ==> Confusion: -[[ 859 1 7 1 10 3 0 2 1 57 1 5 0 5 1 2 4 1 3 0 4] - [ 8 938 2 1 18 21 3 15 5 2 0 3 0 2 0 1 3 1 3 4 3] - [ 13 3 957 5 3 1 14 18 0 0 3 2 1 4 2 12 3 4 8 1 4] - [ 8 3 39 848 3 3 1 2 3 3 25 0 8 5 11 3 3 8 29 0 11] - [ 25 8 1 0 992 4 1 0 1 3 1 5 0 3 4 6 3 2 1 4 2] - [ 9 46 1 5 10 890 4 30 2 10 1 23 3 13 1 3 2 2 3 6 6] - [ 5 6 20 2 0 2 1021 7 0 3 7 2 3 0 0 6 1 2 2 8 2] - [ 3 16 12 2 6 21 1 920 3 2 2 6 1 1 0 2 0 2 17 2 5] - [ 34 4 0 0 3 0 0 0 853 78 2 1 1 14 10 1 1 3 3 0 1] - [ 100 1 3 0 3 0 0 0 12 866 1 1 0 14 2 3 0 1 1 1 3] - [ 2 4 9 1 1 0 4 1 25 2 968 2 0 12 1 3 2 2 10 0 2] - [ 6 1 4 0 4 4 0 6 2 3 0 902 20 6 0 7 1 9 3 24 3] - [ 4 1 0 2 2 3 1 4 5 0 1 54 835 0 2 6 3 22 2 2 10] - [ 11 2 1 0 11 8 2 2 9 32 8 5 1 902 5 7 2 4 1 5 6] - [ 29 10 3 27 11 1 0 0 29 15 2 4 6 2 922 2 5 4 13 0 7] - [ 7 2 4 0 11 1 4 1 0 3 0 10 11 1 0 943 9 21 0 10 8] - [ 9 7 1 0 19 1 0 0 1 3 1 6 1 3 1 13 977 1 3 4 10] - [ 9 1 0 6 2 0 0 2 1 2 0 14 19 1 1 10 0 974 1 4 4] - [ 6 11 6 5 1 1 1 38 10 4 5 3 6 1 12 1 0 0 970 1 4] - [ 1 2 1 0 3 6 10 29 1 0 1 23 1 2 0 3 2 2 2 1050 9] - [ 362 339 383 76 297 158 96 228 107 165 188 180 351 336 114 145 297 136 194 354 8928]] - -2023-02-13 17:30:31,288 - ==> Best [Top1: 80.193 Top5: 96.846 Sparsity:0.00 Params: 148928 on epoch: 26] -2023-02-13 17:30:31,288 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:30:31,294 - - -2023-02-13 17:30:31,294 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:30:32,285 - Epoch: [27][ 10/ 1207] Overall Loss 0.385423 Objective Loss 0.385423 LR 0.001000 Time 0.098988 -2023-02-13 17:30:32,485 - Epoch: [27][ 20/ 1207] Overall Loss 0.381096 Objective Loss 0.381096 LR 0.001000 Time 0.059460 -2023-02-13 17:30:32,681 - Epoch: [27][ 30/ 1207] Overall Loss 0.371485 Objective Loss 0.371485 LR 0.001000 Time 0.046174 -2023-02-13 17:30:32,877 - Epoch: [27][ 40/ 1207] Overall Loss 0.365909 Objective Loss 0.365909 LR 0.001000 Time 0.039509 -2023-02-13 17:30:33,072 - Epoch: [27][ 50/ 1207] Overall Loss 0.366851 Objective Loss 0.366851 LR 0.001000 Time 0.035511 -2023-02-13 17:30:33,268 - Epoch: [27][ 60/ 1207] Overall Loss 0.366797 Objective Loss 0.366797 LR 0.001000 Time 0.032858 -2023-02-13 17:30:33,465 - Epoch: [27][ 70/ 1207] Overall Loss 0.365793 Objective Loss 0.365793 LR 0.001000 Time 0.030959 -2023-02-13 17:30:33,660 - Epoch: [27][ 80/ 1207] Overall Loss 0.369478 Objective Loss 0.369478 LR 0.001000 Time 0.029527 -2023-02-13 17:30:33,856 - Epoch: [27][ 90/ 1207] Overall Loss 0.371178 Objective Loss 0.371178 LR 0.001000 Time 0.028417 -2023-02-13 17:30:34,051 - Epoch: [27][ 100/ 1207] Overall Loss 0.370899 Objective Loss 0.370899 LR 0.001000 Time 0.027527 -2023-02-13 17:30:34,248 - Epoch: [27][ 110/ 1207] Overall Loss 0.370389 Objective Loss 0.370389 LR 0.001000 Time 0.026807 -2023-02-13 17:30:34,443 - Epoch: [27][ 120/ 1207] Overall Loss 0.371620 Objective Loss 0.371620 LR 0.001000 Time 0.026202 -2023-02-13 17:30:34,639 - Epoch: [27][ 130/ 1207] Overall Loss 0.371225 Objective Loss 0.371225 LR 0.001000 Time 0.025688 -2023-02-13 17:30:34,834 - Epoch: [27][ 140/ 1207] Overall Loss 0.371371 Objective Loss 0.371371 LR 0.001000 Time 0.025245 -2023-02-13 17:30:35,030 - Epoch: [27][ 150/ 1207] Overall Loss 0.370661 Objective Loss 0.370661 LR 0.001000 Time 0.024864 -2023-02-13 17:30:35,224 - Epoch: [27][ 160/ 1207] Overall Loss 0.371092 Objective Loss 0.371092 LR 0.001000 Time 0.024523 -2023-02-13 17:30:35,416 - Epoch: [27][ 170/ 1207] Overall Loss 0.371712 Objective Loss 0.371712 LR 0.001000 Time 0.024206 -2023-02-13 17:30:35,608 - Epoch: [27][ 180/ 1207] Overall Loss 0.371781 Objective Loss 0.371781 LR 0.001000 Time 0.023924 -2023-02-13 17:30:35,801 - Epoch: [27][ 190/ 1207] Overall Loss 0.369311 Objective Loss 0.369311 LR 0.001000 Time 0.023678 -2023-02-13 17:30:35,993 - Epoch: [27][ 200/ 1207] Overall Loss 0.366972 Objective Loss 0.366972 LR 0.001000 Time 0.023452 -2023-02-13 17:30:36,185 - Epoch: [27][ 210/ 1207] Overall Loss 0.366524 Objective Loss 0.366524 LR 0.001000 Time 0.023251 -2023-02-13 17:30:36,377 - Epoch: [27][ 220/ 1207] Overall Loss 0.365706 Objective Loss 0.365706 LR 0.001000 Time 0.023063 -2023-02-13 17:30:36,568 - Epoch: [27][ 230/ 1207] Overall Loss 0.365559 Objective Loss 0.365559 LR 0.001000 Time 0.022892 -2023-02-13 17:30:36,761 - Epoch: [27][ 240/ 1207] Overall Loss 0.365884 Objective Loss 0.365884 LR 0.001000 Time 0.022740 -2023-02-13 17:30:36,953 - Epoch: [27][ 250/ 1207] Overall Loss 0.365201 Objective Loss 0.365201 LR 0.001000 Time 0.022596 -2023-02-13 17:30:37,145 - Epoch: [27][ 260/ 1207] Overall Loss 0.365573 Objective Loss 0.365573 LR 0.001000 Time 0.022463 -2023-02-13 17:30:37,337 - Epoch: [27][ 270/ 1207] Overall Loss 0.365652 Objective Loss 0.365652 LR 0.001000 Time 0.022343 -2023-02-13 17:30:37,530 - Epoch: [27][ 280/ 1207] Overall Loss 0.365528 Objective Loss 0.365528 LR 0.001000 Time 0.022231 -2023-02-13 17:30:37,722 - Epoch: [27][ 290/ 1207] Overall Loss 0.365532 Objective Loss 0.365532 LR 0.001000 Time 0.022126 -2023-02-13 17:30:37,914 - Epoch: [27][ 300/ 1207] Overall Loss 0.366516 Objective Loss 0.366516 LR 0.001000 Time 0.022026 -2023-02-13 17:30:38,106 - Epoch: [27][ 310/ 1207] Overall Loss 0.367056 Objective Loss 0.367056 LR 0.001000 Time 0.021934 -2023-02-13 17:30:38,298 - Epoch: [27][ 320/ 1207] Overall Loss 0.368068 Objective Loss 0.368068 LR 0.001000 Time 0.021847 -2023-02-13 17:30:38,490 - Epoch: [27][ 330/ 1207] Overall Loss 0.367048 Objective Loss 0.367048 LR 0.001000 Time 0.021767 -2023-02-13 17:30:38,682 - Epoch: [27][ 340/ 1207] Overall Loss 0.367821 Objective Loss 0.367821 LR 0.001000 Time 0.021690 -2023-02-13 17:30:38,874 - Epoch: [27][ 350/ 1207] Overall Loss 0.368279 Objective Loss 0.368279 LR 0.001000 Time 0.021617 -2023-02-13 17:30:39,066 - Epoch: [27][ 360/ 1207] Overall Loss 0.368332 Objective Loss 0.368332 LR 0.001000 Time 0.021549 -2023-02-13 17:30:39,257 - Epoch: [27][ 370/ 1207] Overall Loss 0.368267 Objective Loss 0.368267 LR 0.001000 Time 0.021483 -2023-02-13 17:30:39,447 - Epoch: [27][ 380/ 1207] Overall Loss 0.369087 Objective Loss 0.369087 LR 0.001000 Time 0.021417 -2023-02-13 17:30:39,639 - Epoch: [27][ 390/ 1207] Overall Loss 0.368473 Objective Loss 0.368473 LR 0.001000 Time 0.021357 -2023-02-13 17:30:39,829 - Epoch: [27][ 400/ 1207] Overall Loss 0.368092 Objective Loss 0.368092 LR 0.001000 Time 0.021299 -2023-02-13 17:30:40,019 - Epoch: [27][ 410/ 1207] Overall Loss 0.367851 Objective Loss 0.367851 LR 0.001000 Time 0.021241 -2023-02-13 17:30:40,209 - Epoch: [27][ 420/ 1207] Overall Loss 0.368278 Objective Loss 0.368278 LR 0.001000 Time 0.021187 -2023-02-13 17:30:40,400 - Epoch: [27][ 430/ 1207] Overall Loss 0.367989 Objective Loss 0.367989 LR 0.001000 Time 0.021138 -2023-02-13 17:30:40,590 - Epoch: [27][ 440/ 1207] Overall Loss 0.367558 Objective Loss 0.367558 LR 0.001000 Time 0.021089 -2023-02-13 17:30:40,782 - Epoch: [27][ 450/ 1207] Overall Loss 0.367513 Objective Loss 0.367513 LR 0.001000 Time 0.021045 -2023-02-13 17:30:40,972 - Epoch: [27][ 460/ 1207] Overall Loss 0.367003 Objective Loss 0.367003 LR 0.001000 Time 0.021000 -2023-02-13 17:30:41,162 - Epoch: [27][ 470/ 1207] Overall Loss 0.366680 Objective Loss 0.366680 LR 0.001000 Time 0.020956 -2023-02-13 17:30:41,353 - Epoch: [27][ 480/ 1207] Overall Loss 0.366244 Objective Loss 0.366244 LR 0.001000 Time 0.020918 -2023-02-13 17:30:41,543 - Epoch: [27][ 490/ 1207] Overall Loss 0.366468 Objective Loss 0.366468 LR 0.001000 Time 0.020879 -2023-02-13 17:30:41,734 - Epoch: [27][ 500/ 1207] Overall Loss 0.366764 Objective Loss 0.366764 LR 0.001000 Time 0.020841 -2023-02-13 17:30:41,924 - Epoch: [27][ 510/ 1207] Overall Loss 0.366532 Objective Loss 0.366532 LR 0.001000 Time 0.020805 -2023-02-13 17:30:42,115 - Epoch: [27][ 520/ 1207] Overall Loss 0.366703 Objective Loss 0.366703 LR 0.001000 Time 0.020771 -2023-02-13 17:30:42,306 - Epoch: [27][ 530/ 1207] Overall Loss 0.366289 Objective Loss 0.366289 LR 0.001000 Time 0.020738 -2023-02-13 17:30:42,496 - Epoch: [27][ 540/ 1207] Overall Loss 0.366265 Objective Loss 0.366265 LR 0.001000 Time 0.020706 -2023-02-13 17:30:42,685 - Epoch: [27][ 550/ 1207] Overall Loss 0.366106 Objective Loss 0.366106 LR 0.001000 Time 0.020673 -2023-02-13 17:30:42,875 - Epoch: [27][ 560/ 1207] Overall Loss 0.366213 Objective Loss 0.366213 LR 0.001000 Time 0.020643 -2023-02-13 17:30:43,066 - Epoch: [27][ 570/ 1207] Overall Loss 0.365955 Objective Loss 0.365955 LR 0.001000 Time 0.020615 -2023-02-13 17:30:43,257 - Epoch: [27][ 580/ 1207] Overall Loss 0.365646 Objective Loss 0.365646 LR 0.001000 Time 0.020588 -2023-02-13 17:30:43,448 - Epoch: [27][ 590/ 1207] Overall Loss 0.365567 Objective Loss 0.365567 LR 0.001000 Time 0.020562 -2023-02-13 17:30:43,638 - Epoch: [27][ 600/ 1207] Overall Loss 0.366098 Objective Loss 0.366098 LR 0.001000 Time 0.020536 -2023-02-13 17:30:43,829 - Epoch: [27][ 610/ 1207] Overall Loss 0.365606 Objective Loss 0.365606 LR 0.001000 Time 0.020511 -2023-02-13 17:30:44,019 - Epoch: [27][ 620/ 1207] Overall Loss 0.365499 Objective Loss 0.365499 LR 0.001000 Time 0.020486 -2023-02-13 17:30:44,210 - Epoch: [27][ 630/ 1207] Overall Loss 0.365802 Objective Loss 0.365802 LR 0.001000 Time 0.020464 -2023-02-13 17:30:44,402 - Epoch: [27][ 640/ 1207] Overall Loss 0.366121 Objective Loss 0.366121 LR 0.001000 Time 0.020443 -2023-02-13 17:30:44,592 - Epoch: [27][ 650/ 1207] Overall Loss 0.366586 Objective Loss 0.366586 LR 0.001000 Time 0.020421 -2023-02-13 17:30:44,783 - Epoch: [27][ 660/ 1207] Overall Loss 0.366315 Objective Loss 0.366315 LR 0.001000 Time 0.020399 -2023-02-13 17:30:44,974 - Epoch: [27][ 670/ 1207] Overall Loss 0.366676 Objective Loss 0.366676 LR 0.001000 Time 0.020380 -2023-02-13 17:30:45,166 - Epoch: [27][ 680/ 1207] Overall Loss 0.366763 Objective Loss 0.366763 LR 0.001000 Time 0.020362 -2023-02-13 17:30:45,357 - Epoch: [27][ 690/ 1207] Overall Loss 0.366855 Objective Loss 0.366855 LR 0.001000 Time 0.020343 -2023-02-13 17:30:45,548 - Epoch: [27][ 700/ 1207] Overall Loss 0.366609 Objective Loss 0.366609 LR 0.001000 Time 0.020325 -2023-02-13 17:30:45,739 - Epoch: [27][ 710/ 1207] Overall Loss 0.366201 Objective Loss 0.366201 LR 0.001000 Time 0.020308 -2023-02-13 17:30:45,931 - Epoch: [27][ 720/ 1207] Overall Loss 0.366014 Objective Loss 0.366014 LR 0.001000 Time 0.020292 -2023-02-13 17:30:46,122 - Epoch: [27][ 730/ 1207] Overall Loss 0.365999 Objective Loss 0.365999 LR 0.001000 Time 0.020275 -2023-02-13 17:30:46,312 - Epoch: [27][ 740/ 1207] Overall Loss 0.365916 Objective Loss 0.365916 LR 0.001000 Time 0.020257 -2023-02-13 17:30:46,501 - Epoch: [27][ 750/ 1207] Overall Loss 0.365776 Objective Loss 0.365776 LR 0.001000 Time 0.020239 -2023-02-13 17:30:46,690 - Epoch: [27][ 760/ 1207] Overall Loss 0.365778 Objective Loss 0.365778 LR 0.001000 Time 0.020221 -2023-02-13 17:30:46,880 - Epoch: [27][ 770/ 1207] Overall Loss 0.365935 Objective Loss 0.365935 LR 0.001000 Time 0.020204 -2023-02-13 17:30:47,070 - Epoch: [27][ 780/ 1207] Overall Loss 0.366146 Objective Loss 0.366146 LR 0.001000 Time 0.020187 -2023-02-13 17:30:47,259 - Epoch: [27][ 790/ 1207] Overall Loss 0.366388 Objective Loss 0.366388 LR 0.001000 Time 0.020171 -2023-02-13 17:30:47,449 - Epoch: [27][ 800/ 1207] Overall Loss 0.366934 Objective Loss 0.366934 LR 0.001000 Time 0.020156 -2023-02-13 17:30:47,639 - Epoch: [27][ 810/ 1207] Overall Loss 0.367651 Objective Loss 0.367651 LR 0.001000 Time 0.020141 -2023-02-13 17:30:47,829 - Epoch: [27][ 820/ 1207] Overall Loss 0.367868 Objective Loss 0.367868 LR 0.001000 Time 0.020127 -2023-02-13 17:30:48,020 - Epoch: [27][ 830/ 1207] Overall Loss 0.367839 Objective Loss 0.367839 LR 0.001000 Time 0.020114 -2023-02-13 17:30:48,209 - Epoch: [27][ 840/ 1207] Overall Loss 0.368016 Objective Loss 0.368016 LR 0.001000 Time 0.020099 -2023-02-13 17:30:48,399 - Epoch: [27][ 850/ 1207] Overall Loss 0.367940 Objective Loss 0.367940 LR 0.001000 Time 0.020086 -2023-02-13 17:30:48,589 - Epoch: [27][ 860/ 1207] Overall Loss 0.367832 Objective Loss 0.367832 LR 0.001000 Time 0.020072 -2023-02-13 17:30:48,779 - Epoch: [27][ 870/ 1207] Overall Loss 0.368266 Objective Loss 0.368266 LR 0.001000 Time 0.020060 -2023-02-13 17:30:48,969 - Epoch: [27][ 880/ 1207] Overall Loss 0.368321 Objective Loss 0.368321 LR 0.001000 Time 0.020047 -2023-02-13 17:30:49,159 - Epoch: [27][ 890/ 1207] Overall Loss 0.368340 Objective Loss 0.368340 LR 0.001000 Time 0.020035 -2023-02-13 17:30:49,349 - Epoch: [27][ 900/ 1207] Overall Loss 0.368012 Objective Loss 0.368012 LR 0.001000 Time 0.020024 -2023-02-13 17:30:49,540 - Epoch: [27][ 910/ 1207] Overall Loss 0.368209 Objective Loss 0.368209 LR 0.001000 Time 0.020013 -2023-02-13 17:30:49,730 - Epoch: [27][ 920/ 1207] Overall Loss 0.368424 Objective Loss 0.368424 LR 0.001000 Time 0.020001 -2023-02-13 17:30:49,920 - Epoch: [27][ 930/ 1207] Overall Loss 0.368352 Objective Loss 0.368352 LR 0.001000 Time 0.019990 -2023-02-13 17:30:50,110 - Epoch: [27][ 940/ 1207] Overall Loss 0.368396 Objective Loss 0.368396 LR 0.001000 Time 0.019979 -2023-02-13 17:30:50,300 - Epoch: [27][ 950/ 1207] Overall Loss 0.368141 Objective Loss 0.368141 LR 0.001000 Time 0.019968 -2023-02-13 17:30:50,490 - Epoch: [27][ 960/ 1207] Overall Loss 0.368501 Objective Loss 0.368501 LR 0.001000 Time 0.019958 -2023-02-13 17:30:50,680 - Epoch: [27][ 970/ 1207] Overall Loss 0.368401 Objective Loss 0.368401 LR 0.001000 Time 0.019948 -2023-02-13 17:30:50,871 - Epoch: [27][ 980/ 1207] Overall Loss 0.368069 Objective Loss 0.368069 LR 0.001000 Time 0.019939 -2023-02-13 17:30:51,061 - Epoch: [27][ 990/ 1207] Overall Loss 0.368220 Objective Loss 0.368220 LR 0.001000 Time 0.019929 -2023-02-13 17:30:51,251 - Epoch: [27][ 1000/ 1207] Overall Loss 0.368172 Objective Loss 0.368172 LR 0.001000 Time 0.019920 -2023-02-13 17:30:51,442 - Epoch: [27][ 1010/ 1207] Overall Loss 0.368088 Objective Loss 0.368088 LR 0.001000 Time 0.019911 -2023-02-13 17:30:51,632 - Epoch: [27][ 1020/ 1207] Overall Loss 0.368051 Objective Loss 0.368051 LR 0.001000 Time 0.019901 -2023-02-13 17:30:51,822 - Epoch: [27][ 1030/ 1207] Overall Loss 0.368426 Objective Loss 0.368426 LR 0.001000 Time 0.019893 -2023-02-13 17:30:52,012 - Epoch: [27][ 1040/ 1207] Overall Loss 0.368259 Objective Loss 0.368259 LR 0.001000 Time 0.019883 -2023-02-13 17:30:52,203 - Epoch: [27][ 1050/ 1207] Overall Loss 0.368029 Objective Loss 0.368029 LR 0.001000 Time 0.019875 -2023-02-13 17:30:52,393 - Epoch: [27][ 1060/ 1207] Overall Loss 0.367934 Objective Loss 0.367934 LR 0.001000 Time 0.019867 -2023-02-13 17:30:52,583 - Epoch: [27][ 1070/ 1207] Overall Loss 0.367908 Objective Loss 0.367908 LR 0.001000 Time 0.019859 -2023-02-13 17:30:52,773 - Epoch: [27][ 1080/ 1207] Overall Loss 0.368247 Objective Loss 0.368247 LR 0.001000 Time 0.019851 -2023-02-13 17:30:52,963 - Epoch: [27][ 1090/ 1207] Overall Loss 0.368238 Objective Loss 0.368238 LR 0.001000 Time 0.019842 -2023-02-13 17:30:53,153 - Epoch: [27][ 1100/ 1207] Overall Loss 0.368404 Objective Loss 0.368404 LR 0.001000 Time 0.019834 -2023-02-13 17:30:53,343 - Epoch: [27][ 1110/ 1207] Overall Loss 0.368297 Objective Loss 0.368297 LR 0.001000 Time 0.019826 -2023-02-13 17:30:53,532 - Epoch: [27][ 1120/ 1207] Overall Loss 0.368568 Objective Loss 0.368568 LR 0.001000 Time 0.019818 -2023-02-13 17:30:53,722 - Epoch: [27][ 1130/ 1207] Overall Loss 0.368576 Objective Loss 0.368576 LR 0.001000 Time 0.019810 -2023-02-13 17:30:53,912 - Epoch: [27][ 1140/ 1207] Overall Loss 0.368420 Objective Loss 0.368420 LR 0.001000 Time 0.019803 -2023-02-13 17:30:54,103 - Epoch: [27][ 1150/ 1207] Overall Loss 0.368629 Objective Loss 0.368629 LR 0.001000 Time 0.019796 -2023-02-13 17:30:54,292 - Epoch: [27][ 1160/ 1207] Overall Loss 0.368414 Objective Loss 0.368414 LR 0.001000 Time 0.019788 -2023-02-13 17:30:54,483 - Epoch: [27][ 1170/ 1207] Overall Loss 0.368442 Objective Loss 0.368442 LR 0.001000 Time 0.019782 -2023-02-13 17:30:54,672 - Epoch: [27][ 1180/ 1207] Overall Loss 0.368450 Objective Loss 0.368450 LR 0.001000 Time 0.019774 -2023-02-13 17:30:54,862 - Epoch: [27][ 1190/ 1207] Overall Loss 0.368730 Objective Loss 0.368730 LR 0.001000 Time 0.019768 -2023-02-13 17:30:55,109 - Epoch: [27][ 1200/ 1207] Overall Loss 0.368508 Objective Loss 0.368508 LR 0.001000 Time 0.019809 -2023-02-13 17:30:55,227 - Epoch: [27][ 1207/ 1207] Overall Loss 0.368485 Objective Loss 0.368485 Top1 77.439024 Top5 96.951220 LR 0.001000 Time 0.019791 -2023-02-13 17:30:55,304 - --- validate (epoch=27)----------- -2023-02-13 17:30:55,305 - 34311 samples (256 per mini-batch) -2023-02-13 17:30:55,709 - Epoch: [27][ 10/ 135] Loss 0.422336 Top1 80.351562 Top5 96.171875 -2023-02-13 17:30:55,838 - Epoch: [27][ 20/ 135] Loss 0.408389 Top1 80.312500 Top5 96.699219 -2023-02-13 17:30:55,963 - Epoch: [27][ 30/ 135] Loss 0.400544 Top1 80.468750 Top5 96.718750 -2023-02-13 17:30:56,090 - Epoch: [27][ 40/ 135] Loss 0.391622 Top1 80.693359 Top5 96.738281 -2023-02-13 17:30:56,219 - Epoch: [27][ 50/ 135] Loss 0.382061 Top1 80.734375 Top5 96.742188 -2023-02-13 17:30:56,345 - Epoch: [27][ 60/ 135] Loss 0.380543 Top1 80.768229 Top5 96.842448 -2023-02-13 17:30:56,472 - Epoch: [27][ 70/ 135] Loss 0.388276 Top1 80.641741 Top5 96.852679 -2023-02-13 17:30:56,602 - Epoch: [27][ 80/ 135] Loss 0.389860 Top1 80.546875 Top5 96.806641 -2023-02-13 17:30:56,730 - Epoch: [27][ 90/ 135] Loss 0.391694 Top1 80.403646 Top5 96.822917 -2023-02-13 17:30:56,858 - Epoch: [27][ 100/ 135] Loss 0.391865 Top1 80.398438 Top5 96.812500 -2023-02-13 17:30:56,987 - Epoch: [27][ 110/ 135] Loss 0.392680 Top1 80.340909 Top5 96.800426 -2023-02-13 17:30:57,115 - Epoch: [27][ 120/ 135] Loss 0.389576 Top1 80.429688 Top5 96.803385 -2023-02-13 17:30:57,245 - Epoch: [27][ 130/ 135] Loss 0.388055 Top1 80.450721 Top5 96.787861 -2023-02-13 17:30:57,292 - Epoch: [27][ 135/ 135] Loss 0.384655 Top1 80.487307 Top5 96.814433 -2023-02-13 17:30:57,363 - ==> Top1: 80.487 Top5: 96.814 Loss: 0.385 - -2023-02-13 17:30:57,364 - ==> Confusion: -[[ 856 5 6 2 6 6 0 3 2 46 0 6 2 6 6 2 1 4 0 0 8] - [ 2 914 1 1 10 44 6 17 4 2 2 5 4 2 1 1 2 2 7 0 6] - [ 10 4 928 10 3 6 22 20 3 0 5 2 2 3 1 7 5 6 7 2 12] - [ 7 4 14 867 0 4 2 2 5 2 21 3 10 3 26 1 2 11 23 1 8] - [ 27 10 1 0 966 10 2 3 3 7 1 6 2 6 5 6 1 3 0 2 5] - [ 5 24 2 2 3 952 2 26 1 4 3 9 7 12 1 1 0 4 4 2 6] - [ 4 1 23 3 0 8 1018 6 0 1 7 3 3 0 0 6 3 3 0 6 4] - [ 3 7 15 1 2 22 8 904 1 2 4 7 4 1 0 0 1 2 28 8 4] - [ 21 3 0 1 0 2 0 0 884 42 6 1 1 18 17 1 0 3 7 1 1] - [ 104 2 1 0 3 1 0 3 45 820 1 1 0 19 5 1 0 2 0 0 4] - [ 2 4 3 3 0 1 3 3 21 2 970 1 2 15 5 0 0 2 10 0 4] - [ 2 4 2 0 1 8 1 4 4 0 0 886 57 5 1 4 0 16 4 6 0] - [ 3 2 1 1 0 2 0 0 1 0 0 29 882 0 2 1 2 27 0 0 6] - [ 6 6 4 0 5 14 2 1 15 20 5 11 3 914 5 6 1 0 0 2 4] - [ 18 5 2 18 5 4 0 0 27 6 3 3 11 3 960 1 3 8 8 0 7] - [ 4 4 6 1 7 4 3 1 0 1 1 14 12 5 0 937 4 28 0 8 6] - [ 5 5 0 3 9 4 0 0 3 1 0 7 5 4 0 31 955 4 2 9 14] - [ 5 3 0 4 0 0 0 2 0 1 1 3 28 1 0 5 0 994 1 0 3] - [ 4 3 9 8 2 2 0 31 6 1 6 1 13 0 21 1 2 3 970 2 1] - [ 0 0 2 0 0 10 8 19 1 0 1 30 6 4 0 7 0 4 1 1042 13] - [ 223 285 246 138 174 295 103 206 139 121 287 175 498 421 167 92 189 211 180 287 8997]] - -2023-02-13 17:30:57,365 - ==> Best [Top1: 80.487 Top5: 96.814 Sparsity:0.00 Params: 148928 on epoch: 27] -2023-02-13 17:30:57,365 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:30:57,372 - - -2023-02-13 17:30:57,372 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:30:58,251 - Epoch: [28][ 10/ 1207] Overall Loss 0.391595 Objective Loss 0.391595 LR 0.001000 Time 0.087845 -2023-02-13 17:30:58,449 - Epoch: [28][ 20/ 1207] Overall Loss 0.383857 Objective Loss 0.383857 LR 0.001000 Time 0.053811 -2023-02-13 17:30:58,643 - Epoch: [28][ 30/ 1207] Overall Loss 0.373436 Objective Loss 0.373436 LR 0.001000 Time 0.042309 -2023-02-13 17:30:58,837 - Epoch: [28][ 40/ 1207] Overall Loss 0.369322 Objective Loss 0.369322 LR 0.001000 Time 0.036569 -2023-02-13 17:30:59,028 - Epoch: [28][ 50/ 1207] Overall Loss 0.366869 Objective Loss 0.366869 LR 0.001000 Time 0.033085 -2023-02-13 17:30:59,223 - Epoch: [28][ 60/ 1207] Overall Loss 0.366551 Objective Loss 0.366551 LR 0.001000 Time 0.030806 -2023-02-13 17:30:59,417 - Epoch: [28][ 70/ 1207] Overall Loss 0.364654 Objective Loss 0.364654 LR 0.001000 Time 0.029168 -2023-02-13 17:30:59,610 - Epoch: [28][ 80/ 1207] Overall Loss 0.369475 Objective Loss 0.369475 LR 0.001000 Time 0.027936 -2023-02-13 17:30:59,803 - Epoch: [28][ 90/ 1207] Overall Loss 0.368167 Objective Loss 0.368167 LR 0.001000 Time 0.026969 -2023-02-13 17:30:59,997 - Epoch: [28][ 100/ 1207] Overall Loss 0.370284 Objective Loss 0.370284 LR 0.001000 Time 0.026212 -2023-02-13 17:31:00,191 - Epoch: [28][ 110/ 1207] Overall Loss 0.369362 Objective Loss 0.369362 LR 0.001000 Time 0.025583 -2023-02-13 17:31:00,385 - Epoch: [28][ 120/ 1207] Overall Loss 0.366803 Objective Loss 0.366803 LR 0.001000 Time 0.025068 -2023-02-13 17:31:00,578 - Epoch: [28][ 130/ 1207] Overall Loss 0.365656 Objective Loss 0.365656 LR 0.001000 Time 0.024621 -2023-02-13 17:31:00,773 - Epoch: [28][ 140/ 1207] Overall Loss 0.365772 Objective Loss 0.365772 LR 0.001000 Time 0.024250 -2023-02-13 17:31:00,966 - Epoch: [28][ 150/ 1207] Overall Loss 0.365070 Objective Loss 0.365070 LR 0.001000 Time 0.023921 -2023-02-13 17:31:01,161 - Epoch: [28][ 160/ 1207] Overall Loss 0.364599 Objective Loss 0.364599 LR 0.001000 Time 0.023642 -2023-02-13 17:31:01,355 - Epoch: [28][ 170/ 1207] Overall Loss 0.365210 Objective Loss 0.365210 LR 0.001000 Time 0.023388 -2023-02-13 17:31:01,550 - Epoch: [28][ 180/ 1207] Overall Loss 0.365377 Objective Loss 0.365377 LR 0.001000 Time 0.023172 -2023-02-13 17:31:01,743 - Epoch: [28][ 190/ 1207] Overall Loss 0.363786 Objective Loss 0.363786 LR 0.001000 Time 0.022964 -2023-02-13 17:31:01,937 - Epoch: [28][ 200/ 1207] Overall Loss 0.362996 Objective Loss 0.362996 LR 0.001000 Time 0.022788 -2023-02-13 17:31:02,131 - Epoch: [28][ 210/ 1207] Overall Loss 0.362995 Objective Loss 0.362995 LR 0.001000 Time 0.022621 -2023-02-13 17:31:02,328 - Epoch: [28][ 220/ 1207] Overall Loss 0.362974 Objective Loss 0.362974 LR 0.001000 Time 0.022486 -2023-02-13 17:31:02,523 - Epoch: [28][ 230/ 1207] Overall Loss 0.363204 Objective Loss 0.363204 LR 0.001000 Time 0.022358 -2023-02-13 17:31:02,720 - Epoch: [28][ 240/ 1207] Overall Loss 0.363184 Objective Loss 0.363184 LR 0.001000 Time 0.022244 -2023-02-13 17:31:02,915 - Epoch: [28][ 250/ 1207] Overall Loss 0.363285 Objective Loss 0.363285 LR 0.001000 Time 0.022132 -2023-02-13 17:31:03,112 - Epoch: [28][ 260/ 1207] Overall Loss 0.363452 Objective Loss 0.363452 LR 0.001000 Time 0.022037 -2023-02-13 17:31:03,306 - Epoch: [28][ 270/ 1207] Overall Loss 0.362667 Objective Loss 0.362667 LR 0.001000 Time 0.021939 -2023-02-13 17:31:03,504 - Epoch: [28][ 280/ 1207] Overall Loss 0.362451 Objective Loss 0.362451 LR 0.001000 Time 0.021860 -2023-02-13 17:31:03,698 - Epoch: [28][ 290/ 1207] Overall Loss 0.363175 Objective Loss 0.363175 LR 0.001000 Time 0.021774 -2023-02-13 17:31:03,894 - Epoch: [28][ 300/ 1207] Overall Loss 0.363171 Objective Loss 0.363171 LR 0.001000 Time 0.021702 -2023-02-13 17:31:04,089 - Epoch: [28][ 310/ 1207] Overall Loss 0.362854 Objective Loss 0.362854 LR 0.001000 Time 0.021631 -2023-02-13 17:31:04,287 - Epoch: [28][ 320/ 1207] Overall Loss 0.362043 Objective Loss 0.362043 LR 0.001000 Time 0.021572 -2023-02-13 17:31:04,482 - Epoch: [28][ 330/ 1207] Overall Loss 0.362812 Objective Loss 0.362812 LR 0.001000 Time 0.021508 -2023-02-13 17:31:04,679 - Epoch: [28][ 340/ 1207] Overall Loss 0.363546 Objective Loss 0.363546 LR 0.001000 Time 0.021454 -2023-02-13 17:31:04,874 - Epoch: [28][ 350/ 1207] Overall Loss 0.363099 Objective Loss 0.363099 LR 0.001000 Time 0.021397 -2023-02-13 17:31:05,072 - Epoch: [28][ 360/ 1207] Overall Loss 0.363956 Objective Loss 0.363956 LR 0.001000 Time 0.021351 -2023-02-13 17:31:05,266 - Epoch: [28][ 370/ 1207] Overall Loss 0.364503 Objective Loss 0.364503 LR 0.001000 Time 0.021298 -2023-02-13 17:31:05,464 - Epoch: [28][ 380/ 1207] Overall Loss 0.364869 Objective Loss 0.364869 LR 0.001000 Time 0.021256 -2023-02-13 17:31:05,658 - Epoch: [28][ 390/ 1207] Overall Loss 0.365181 Objective Loss 0.365181 LR 0.001000 Time 0.021208 -2023-02-13 17:31:05,857 - Epoch: [28][ 400/ 1207] Overall Loss 0.365465 Objective Loss 0.365465 LR 0.001000 Time 0.021174 -2023-02-13 17:31:06,051 - Epoch: [28][ 410/ 1207] Overall Loss 0.365893 Objective Loss 0.365893 LR 0.001000 Time 0.021130 -2023-02-13 17:31:06,249 - Epoch: [28][ 420/ 1207] Overall Loss 0.365473 Objective Loss 0.365473 LR 0.001000 Time 0.021098 -2023-02-13 17:31:06,444 - Epoch: [28][ 430/ 1207] Overall Loss 0.365661 Objective Loss 0.365661 LR 0.001000 Time 0.021060 -2023-02-13 17:31:06,641 - Epoch: [28][ 440/ 1207] Overall Loss 0.365551 Objective Loss 0.365551 LR 0.001000 Time 0.021028 -2023-02-13 17:31:06,836 - Epoch: [28][ 450/ 1207] Overall Loss 0.365616 Objective Loss 0.365616 LR 0.001000 Time 0.020992 -2023-02-13 17:31:07,032 - Epoch: [28][ 460/ 1207] Overall Loss 0.365396 Objective Loss 0.365396 LR 0.001000 Time 0.020962 -2023-02-13 17:31:07,227 - Epoch: [28][ 470/ 1207] Overall Loss 0.364479 Objective Loss 0.364479 LR 0.001000 Time 0.020930 -2023-02-13 17:31:07,424 - Epoch: [28][ 480/ 1207] Overall Loss 0.365227 Objective Loss 0.365227 LR 0.001000 Time 0.020905 -2023-02-13 17:31:07,619 - Epoch: [28][ 490/ 1207] Overall Loss 0.364792 Objective Loss 0.364792 LR 0.001000 Time 0.020875 -2023-02-13 17:31:07,817 - Epoch: [28][ 500/ 1207] Overall Loss 0.364657 Objective Loss 0.364657 LR 0.001000 Time 0.020851 -2023-02-13 17:31:08,012 - Epoch: [28][ 510/ 1207] Overall Loss 0.365226 Objective Loss 0.365226 LR 0.001000 Time 0.020824 -2023-02-13 17:31:08,209 - Epoch: [28][ 520/ 1207] Overall Loss 0.365031 Objective Loss 0.365031 LR 0.001000 Time 0.020803 -2023-02-13 17:31:08,399 - Epoch: [28][ 530/ 1207] Overall Loss 0.364601 Objective Loss 0.364601 LR 0.001000 Time 0.020767 -2023-02-13 17:31:08,588 - Epoch: [28][ 540/ 1207] Overall Loss 0.364517 Objective Loss 0.364517 LR 0.001000 Time 0.020732 -2023-02-13 17:31:08,776 - Epoch: [28][ 550/ 1207] Overall Loss 0.365170 Objective Loss 0.365170 LR 0.001000 Time 0.020697 -2023-02-13 17:31:08,964 - Epoch: [28][ 560/ 1207] Overall Loss 0.364480 Objective Loss 0.364480 LR 0.001000 Time 0.020663 -2023-02-13 17:31:09,153 - Epoch: [28][ 570/ 1207] Overall Loss 0.363731 Objective Loss 0.363731 LR 0.001000 Time 0.020630 -2023-02-13 17:31:09,341 - Epoch: [28][ 580/ 1207] Overall Loss 0.364257 Objective Loss 0.364257 LR 0.001000 Time 0.020598 -2023-02-13 17:31:09,530 - Epoch: [28][ 590/ 1207] Overall Loss 0.364358 Objective Loss 0.364358 LR 0.001000 Time 0.020568 -2023-02-13 17:31:09,718 - Epoch: [28][ 600/ 1207] Overall Loss 0.364777 Objective Loss 0.364777 LR 0.001000 Time 0.020539 -2023-02-13 17:31:09,906 - Epoch: [28][ 610/ 1207] Overall Loss 0.364740 Objective Loss 0.364740 LR 0.001000 Time 0.020510 -2023-02-13 17:31:10,095 - Epoch: [28][ 620/ 1207] Overall Loss 0.364746 Objective Loss 0.364746 LR 0.001000 Time 0.020483 -2023-02-13 17:31:10,284 - Epoch: [28][ 630/ 1207] Overall Loss 0.364747 Objective Loss 0.364747 LR 0.001000 Time 0.020458 -2023-02-13 17:31:10,473 - Epoch: [28][ 640/ 1207] Overall Loss 0.364940 Objective Loss 0.364940 LR 0.001000 Time 0.020433 -2023-02-13 17:31:10,662 - Epoch: [28][ 650/ 1207] Overall Loss 0.364506 Objective Loss 0.364506 LR 0.001000 Time 0.020408 -2023-02-13 17:31:10,852 - Epoch: [28][ 660/ 1207] Overall Loss 0.364287 Objective Loss 0.364287 LR 0.001000 Time 0.020386 -2023-02-13 17:31:11,040 - Epoch: [28][ 670/ 1207] Overall Loss 0.364404 Objective Loss 0.364404 LR 0.001000 Time 0.020362 -2023-02-13 17:31:11,229 - Epoch: [28][ 680/ 1207] Overall Loss 0.364523 Objective Loss 0.364523 LR 0.001000 Time 0.020341 -2023-02-13 17:31:11,418 - Epoch: [28][ 690/ 1207] Overall Loss 0.364447 Objective Loss 0.364447 LR 0.001000 Time 0.020319 -2023-02-13 17:31:11,607 - Epoch: [28][ 700/ 1207] Overall Loss 0.364797 Objective Loss 0.364797 LR 0.001000 Time 0.020298 -2023-02-13 17:31:11,797 - Epoch: [28][ 710/ 1207] Overall Loss 0.364939 Objective Loss 0.364939 LR 0.001000 Time 0.020279 -2023-02-13 17:31:11,985 - Epoch: [28][ 720/ 1207] Overall Loss 0.365352 Objective Loss 0.365352 LR 0.001000 Time 0.020259 -2023-02-13 17:31:12,174 - Epoch: [28][ 730/ 1207] Overall Loss 0.365506 Objective Loss 0.365506 LR 0.001000 Time 0.020239 -2023-02-13 17:31:12,363 - Epoch: [28][ 740/ 1207] Overall Loss 0.365603 Objective Loss 0.365603 LR 0.001000 Time 0.020221 -2023-02-13 17:31:12,553 - Epoch: [28][ 750/ 1207] Overall Loss 0.365308 Objective Loss 0.365308 LR 0.001000 Time 0.020203 -2023-02-13 17:31:12,742 - Epoch: [28][ 760/ 1207] Overall Loss 0.365265 Objective Loss 0.365265 LR 0.001000 Time 0.020186 -2023-02-13 17:31:12,931 - Epoch: [28][ 770/ 1207] Overall Loss 0.365390 Objective Loss 0.365390 LR 0.001000 Time 0.020169 -2023-02-13 17:31:13,120 - Epoch: [28][ 780/ 1207] Overall Loss 0.365681 Objective Loss 0.365681 LR 0.001000 Time 0.020152 -2023-02-13 17:31:13,309 - Epoch: [28][ 790/ 1207] Overall Loss 0.365362 Objective Loss 0.365362 LR 0.001000 Time 0.020136 -2023-02-13 17:31:13,498 - Epoch: [28][ 800/ 1207] Overall Loss 0.364977 Objective Loss 0.364977 LR 0.001000 Time 0.020120 -2023-02-13 17:31:13,688 - Epoch: [28][ 810/ 1207] Overall Loss 0.364634 Objective Loss 0.364634 LR 0.001000 Time 0.020105 -2023-02-13 17:31:13,877 - Epoch: [28][ 820/ 1207] Overall Loss 0.364334 Objective Loss 0.364334 LR 0.001000 Time 0.020090 -2023-02-13 17:31:14,065 - Epoch: [28][ 830/ 1207] Overall Loss 0.364510 Objective Loss 0.364510 LR 0.001000 Time 0.020075 -2023-02-13 17:31:14,254 - Epoch: [28][ 840/ 1207] Overall Loss 0.364556 Objective Loss 0.364556 LR 0.001000 Time 0.020060 -2023-02-13 17:31:14,443 - Epoch: [28][ 850/ 1207] Overall Loss 0.364226 Objective Loss 0.364226 LR 0.001000 Time 0.020046 -2023-02-13 17:31:14,632 - Epoch: [28][ 860/ 1207] Overall Loss 0.364122 Objective Loss 0.364122 LR 0.001000 Time 0.020032 -2023-02-13 17:31:14,821 - Epoch: [28][ 870/ 1207] Overall Loss 0.364248 Objective Loss 0.364248 LR 0.001000 Time 0.020019 -2023-02-13 17:31:15,009 - Epoch: [28][ 880/ 1207] Overall Loss 0.364114 Objective Loss 0.364114 LR 0.001000 Time 0.020005 -2023-02-13 17:31:15,198 - Epoch: [28][ 890/ 1207] Overall Loss 0.364513 Objective Loss 0.364513 LR 0.001000 Time 0.019992 -2023-02-13 17:31:15,387 - Epoch: [28][ 900/ 1207] Overall Loss 0.364455 Objective Loss 0.364455 LR 0.001000 Time 0.019979 -2023-02-13 17:31:15,577 - Epoch: [28][ 910/ 1207] Overall Loss 0.365067 Objective Loss 0.365067 LR 0.001000 Time 0.019967 -2023-02-13 17:31:15,766 - Epoch: [28][ 920/ 1207] Overall Loss 0.365220 Objective Loss 0.365220 LR 0.001000 Time 0.019955 -2023-02-13 17:31:15,955 - Epoch: [28][ 930/ 1207] Overall Loss 0.365193 Objective Loss 0.365193 LR 0.001000 Time 0.019944 -2023-02-13 17:31:16,144 - Epoch: [28][ 940/ 1207] Overall Loss 0.365397 Objective Loss 0.365397 LR 0.001000 Time 0.019933 -2023-02-13 17:31:16,333 - Epoch: [28][ 950/ 1207] Overall Loss 0.365970 Objective Loss 0.365970 LR 0.001000 Time 0.019921 -2023-02-13 17:31:16,522 - Epoch: [28][ 960/ 1207] Overall Loss 0.366008 Objective Loss 0.366008 LR 0.001000 Time 0.019911 -2023-02-13 17:31:16,711 - Epoch: [28][ 970/ 1207] Overall Loss 0.365777 Objective Loss 0.365777 LR 0.001000 Time 0.019900 -2023-02-13 17:31:16,900 - Epoch: [28][ 980/ 1207] Overall Loss 0.365682 Objective Loss 0.365682 LR 0.001000 Time 0.019889 -2023-02-13 17:31:17,089 - Epoch: [28][ 990/ 1207] Overall Loss 0.365912 Objective Loss 0.365912 LR 0.001000 Time 0.019879 -2023-02-13 17:31:17,278 - Epoch: [28][ 1000/ 1207] Overall Loss 0.365728 Objective Loss 0.365728 LR 0.001000 Time 0.019869 -2023-02-13 17:31:17,467 - Epoch: [28][ 1010/ 1207] Overall Loss 0.366020 Objective Loss 0.366020 LR 0.001000 Time 0.019859 -2023-02-13 17:31:17,656 - Epoch: [28][ 1020/ 1207] Overall Loss 0.366520 Objective Loss 0.366520 LR 0.001000 Time 0.019849 -2023-02-13 17:31:17,846 - Epoch: [28][ 1030/ 1207] Overall Loss 0.366607 Objective Loss 0.366607 LR 0.001000 Time 0.019840 -2023-02-13 17:31:18,035 - Epoch: [28][ 1040/ 1207] Overall Loss 0.366567 Objective Loss 0.366567 LR 0.001000 Time 0.019830 -2023-02-13 17:31:18,224 - Epoch: [28][ 1050/ 1207] Overall Loss 0.366613 Objective Loss 0.366613 LR 0.001000 Time 0.019822 -2023-02-13 17:31:18,414 - Epoch: [28][ 1060/ 1207] Overall Loss 0.366508 Objective Loss 0.366508 LR 0.001000 Time 0.019813 -2023-02-13 17:31:18,602 - Epoch: [28][ 1070/ 1207] Overall Loss 0.366591 Objective Loss 0.366591 LR 0.001000 Time 0.019804 -2023-02-13 17:31:18,791 - Epoch: [28][ 1080/ 1207] Overall Loss 0.366700 Objective Loss 0.366700 LR 0.001000 Time 0.019795 -2023-02-13 17:31:18,980 - Epoch: [28][ 1090/ 1207] Overall Loss 0.366859 Objective Loss 0.366859 LR 0.001000 Time 0.019786 -2023-02-13 17:31:19,169 - Epoch: [28][ 1100/ 1207] Overall Loss 0.366757 Objective Loss 0.366757 LR 0.001000 Time 0.019778 -2023-02-13 17:31:19,358 - Epoch: [28][ 1110/ 1207] Overall Loss 0.366872 Objective Loss 0.366872 LR 0.001000 Time 0.019770 -2023-02-13 17:31:19,548 - Epoch: [28][ 1120/ 1207] Overall Loss 0.366837 Objective Loss 0.366837 LR 0.001000 Time 0.019762 -2023-02-13 17:31:19,737 - Epoch: [28][ 1130/ 1207] Overall Loss 0.366785 Objective Loss 0.366785 LR 0.001000 Time 0.019755 -2023-02-13 17:31:19,926 - Epoch: [28][ 1140/ 1207] Overall Loss 0.366789 Objective Loss 0.366789 LR 0.001000 Time 0.019747 -2023-02-13 17:31:20,115 - Epoch: [28][ 1150/ 1207] Overall Loss 0.366663 Objective Loss 0.366663 LR 0.001000 Time 0.019739 -2023-02-13 17:31:20,305 - Epoch: [28][ 1160/ 1207] Overall Loss 0.366921 Objective Loss 0.366921 LR 0.001000 Time 0.019732 -2023-02-13 17:31:20,494 - Epoch: [28][ 1170/ 1207] Overall Loss 0.367001 Objective Loss 0.367001 LR 0.001000 Time 0.019725 -2023-02-13 17:31:20,683 - Epoch: [28][ 1180/ 1207] Overall Loss 0.366775 Objective Loss 0.366775 LR 0.001000 Time 0.019718 -2023-02-13 17:31:20,873 - Epoch: [28][ 1190/ 1207] Overall Loss 0.366746 Objective Loss 0.366746 LR 0.001000 Time 0.019712 -2023-02-13 17:31:21,119 - Epoch: [28][ 1200/ 1207] Overall Loss 0.366744 Objective Loss 0.366744 LR 0.001000 Time 0.019752 -2023-02-13 17:31:21,235 - Epoch: [28][ 1207/ 1207] Overall Loss 0.366653 Objective Loss 0.366653 Top1 79.878049 Top5 98.170732 LR 0.001000 Time 0.019733 -2023-02-13 17:31:21,306 - --- validate (epoch=28)----------- -2023-02-13 17:31:21,306 - 34311 samples (256 per mini-batch) -2023-02-13 17:31:21,698 - Epoch: [28][ 10/ 135] Loss 0.362934 Top1 79.492188 Top5 96.640625 -2023-02-13 17:31:21,826 - Epoch: [28][ 20/ 135] Loss 0.366542 Top1 79.550781 Top5 96.679688 -2023-02-13 17:31:21,953 - Epoch: [28][ 30/ 135] Loss 0.364917 Top1 79.778646 Top5 96.861979 -2023-02-13 17:31:22,089 - Epoch: [28][ 40/ 135] Loss 0.376213 Top1 79.912109 Top5 96.894531 -2023-02-13 17:31:22,209 - Epoch: [28][ 50/ 135] Loss 0.375472 Top1 79.828125 Top5 96.875000 -2023-02-13 17:31:22,331 - Epoch: [28][ 60/ 135] Loss 0.380641 Top1 79.772135 Top5 96.888021 -2023-02-13 17:31:22,455 - Epoch: [28][ 70/ 135] Loss 0.381172 Top1 79.687500 Top5 96.852679 -2023-02-13 17:31:22,575 - Epoch: [28][ 80/ 135] Loss 0.379876 Top1 79.589844 Top5 96.816406 -2023-02-13 17:31:22,697 - Epoch: [28][ 90/ 135] Loss 0.384319 Top1 79.370660 Top5 96.822917 -2023-02-13 17:31:22,818 - Epoch: [28][ 100/ 135] Loss 0.386851 Top1 79.382812 Top5 96.789062 -2023-02-13 17:31:22,941 - Epoch: [28][ 110/ 135] Loss 0.388284 Top1 79.360795 Top5 96.796875 -2023-02-13 17:31:23,062 - Epoch: [28][ 120/ 135] Loss 0.386582 Top1 79.388021 Top5 96.803385 -2023-02-13 17:31:23,188 - Epoch: [28][ 130/ 135] Loss 0.389499 Top1 79.435096 Top5 96.790865 -2023-02-13 17:31:23,232 - Epoch: [28][ 135/ 135] Loss 0.387042 Top1 79.513859 Top5 96.808604 -2023-02-13 17:31:23,299 - ==> Top1: 79.514 Top5: 96.809 Loss: 0.387 - -2023-02-13 17:31:23,300 - ==> Confusion: -[[ 865 3 5 2 14 2 0 3 3 36 2 6 0 5 5 3 1 3 4 1 4] - [ 6 932 2 3 6 29 3 22 6 2 2 4 1 1 1 1 3 0 3 3 3] - [ 11 5 923 12 8 2 29 18 0 0 3 3 4 3 0 7 1 3 13 4 9] - [ 5 4 21 863 2 5 1 2 3 2 20 1 6 2 24 1 4 7 33 1 9] - [ 23 9 0 0 980 10 2 1 1 2 1 8 1 7 8 4 3 0 0 3 3] - [ 4 30 0 5 3 905 3 34 1 5 3 23 5 17 4 1 5 3 4 13 2] - [ 6 6 10 1 1 4 1028 5 0 1 4 4 3 0 0 7 0 7 0 9 3] - [ 2 7 7 1 3 19 10 907 1 1 6 6 3 3 0 0 1 3 29 10 5] - [ 18 2 0 1 3 1 0 0 910 37 10 2 1 11 4 2 1 3 3 0 0] - [ 124 3 1 0 6 2 0 0 46 795 1 1 0 21 2 1 1 2 0 0 6] - [ 4 5 4 5 1 0 1 0 24 2 967 1 0 12 1 0 0 1 19 1 3] - [ 5 3 0 0 2 6 1 6 2 3 0 864 66 8 1 10 3 14 4 7 0] - [ 4 2 2 3 1 2 0 0 2 0 0 24 871 0 4 5 1 27 4 2 5] - [ 10 3 1 0 7 17 0 1 12 14 7 16 3 907 7 7 5 1 2 2 2] - [ 17 4 0 20 10 1 0 0 47 8 3 4 4 0 938 1 1 5 21 0 8] - [ 6 0 2 0 7 6 5 2 0 0 0 10 9 1 2 947 9 27 1 5 7] - [ 5 11 1 4 13 5 0 1 5 0 0 5 2 1 1 18 975 3 2 4 5] - [ 6 1 0 5 3 1 0 1 2 0 1 4 18 0 4 10 1 991 0 1 2] - [ 2 5 6 5 1 4 0 29 12 3 9 0 4 0 8 1 0 2 994 1 0] - [ 0 3 0 1 3 4 10 17 0 0 0 20 6 2 0 7 0 1 2 1065 7] - [ 299 353 216 162 217 233 102 250 164 143 266 131 387 386 195 146 325 151 255 398 8655]] - -2023-02-13 17:31:23,301 - ==> Best [Top1: 80.487 Top5: 96.814 Sparsity:0.00 Params: 148928 on epoch: 27] -2023-02-13 17:31:23,301 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:31:23,307 - - -2023-02-13 17:31:23,307 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:31:24,166 - Epoch: [29][ 10/ 1207] Overall Loss 0.337032 Objective Loss 0.337032 LR 0.001000 Time 0.085869 -2023-02-13 17:31:24,365 - Epoch: [29][ 20/ 1207] Overall Loss 0.340698 Objective Loss 0.340698 LR 0.001000 Time 0.052847 -2023-02-13 17:31:24,553 - Epoch: [29][ 30/ 1207] Overall Loss 0.334071 Objective Loss 0.334071 LR 0.001000 Time 0.041476 -2023-02-13 17:31:24,740 - Epoch: [29][ 40/ 1207] Overall Loss 0.336441 Objective Loss 0.336441 LR 0.001000 Time 0.035778 -2023-02-13 17:31:24,927 - Epoch: [29][ 50/ 1207] Overall Loss 0.337503 Objective Loss 0.337503 LR 0.001000 Time 0.032354 -2023-02-13 17:31:25,114 - Epoch: [29][ 60/ 1207] Overall Loss 0.332693 Objective Loss 0.332693 LR 0.001000 Time 0.030071 -2023-02-13 17:31:25,301 - Epoch: [29][ 70/ 1207] Overall Loss 0.332102 Objective Loss 0.332102 LR 0.001000 Time 0.028446 -2023-02-13 17:31:25,490 - Epoch: [29][ 80/ 1207] Overall Loss 0.335404 Objective Loss 0.335404 LR 0.001000 Time 0.027243 -2023-02-13 17:31:25,677 - Epoch: [29][ 90/ 1207] Overall Loss 0.335357 Objective Loss 0.335357 LR 0.001000 Time 0.026288 -2023-02-13 17:31:25,867 - Epoch: [29][ 100/ 1207] Overall Loss 0.335791 Objective Loss 0.335791 LR 0.001000 Time 0.025559 -2023-02-13 17:31:26,055 - Epoch: [29][ 110/ 1207] Overall Loss 0.336687 Objective Loss 0.336687 LR 0.001000 Time 0.024939 -2023-02-13 17:31:26,242 - Epoch: [29][ 120/ 1207] Overall Loss 0.336891 Objective Loss 0.336891 LR 0.001000 Time 0.024423 -2023-02-13 17:31:26,430 - Epoch: [29][ 130/ 1207] Overall Loss 0.337221 Objective Loss 0.337221 LR 0.001000 Time 0.023983 -2023-02-13 17:31:26,618 - Epoch: [29][ 140/ 1207] Overall Loss 0.340817 Objective Loss 0.340817 LR 0.001000 Time 0.023612 -2023-02-13 17:31:26,806 - Epoch: [29][ 150/ 1207] Overall Loss 0.342797 Objective Loss 0.342797 LR 0.001000 Time 0.023287 -2023-02-13 17:31:26,994 - Epoch: [29][ 160/ 1207] Overall Loss 0.342206 Objective Loss 0.342206 LR 0.001000 Time 0.023005 -2023-02-13 17:31:27,181 - Epoch: [29][ 170/ 1207] Overall Loss 0.343363 Objective Loss 0.343363 LR 0.001000 Time 0.022749 -2023-02-13 17:31:27,368 - Epoch: [29][ 180/ 1207] Overall Loss 0.345455 Objective Loss 0.345455 LR 0.001000 Time 0.022525 -2023-02-13 17:31:27,556 - Epoch: [29][ 190/ 1207] Overall Loss 0.346595 Objective Loss 0.346595 LR 0.001000 Time 0.022326 -2023-02-13 17:31:27,744 - Epoch: [29][ 200/ 1207] Overall Loss 0.347554 Objective Loss 0.347554 LR 0.001000 Time 0.022145 -2023-02-13 17:31:27,931 - Epoch: [29][ 210/ 1207] Overall Loss 0.347165 Objective Loss 0.347165 LR 0.001000 Time 0.021980 -2023-02-13 17:31:28,119 - Epoch: [29][ 220/ 1207] Overall Loss 0.346346 Objective Loss 0.346346 LR 0.001000 Time 0.021832 -2023-02-13 17:31:28,306 - Epoch: [29][ 230/ 1207] Overall Loss 0.346322 Objective Loss 0.346322 LR 0.001000 Time 0.021697 -2023-02-13 17:31:28,495 - Epoch: [29][ 240/ 1207] Overall Loss 0.347021 Objective Loss 0.347021 LR 0.001000 Time 0.021577 -2023-02-13 17:31:28,682 - Epoch: [29][ 250/ 1207] Overall Loss 0.347885 Objective Loss 0.347885 LR 0.001000 Time 0.021463 -2023-02-13 17:31:28,870 - Epoch: [29][ 260/ 1207] Overall Loss 0.349377 Objective Loss 0.349377 LR 0.001000 Time 0.021357 -2023-02-13 17:31:29,056 - Epoch: [29][ 270/ 1207] Overall Loss 0.351176 Objective Loss 0.351176 LR 0.001000 Time 0.021256 -2023-02-13 17:31:29,245 - Epoch: [29][ 280/ 1207] Overall Loss 0.351787 Objective Loss 0.351787 LR 0.001000 Time 0.021168 -2023-02-13 17:31:29,433 - Epoch: [29][ 290/ 1207] Overall Loss 0.350613 Objective Loss 0.350613 LR 0.001000 Time 0.021086 -2023-02-13 17:31:29,621 - Epoch: [29][ 300/ 1207] Overall Loss 0.350311 Objective Loss 0.350311 LR 0.001000 Time 0.021008 -2023-02-13 17:31:29,808 - Epoch: [29][ 310/ 1207] Overall Loss 0.351567 Objective Loss 0.351567 LR 0.001000 Time 0.020934 -2023-02-13 17:31:29,996 - Epoch: [29][ 320/ 1207] Overall Loss 0.352074 Objective Loss 0.352074 LR 0.001000 Time 0.020865 -2023-02-13 17:31:30,183 - Epoch: [29][ 330/ 1207] Overall Loss 0.352954 Objective Loss 0.352954 LR 0.001000 Time 0.020800 -2023-02-13 17:31:30,371 - Epoch: [29][ 340/ 1207] Overall Loss 0.353563 Objective Loss 0.353563 LR 0.001000 Time 0.020740 -2023-02-13 17:31:30,559 - Epoch: [29][ 350/ 1207] Overall Loss 0.353073 Objective Loss 0.353073 LR 0.001000 Time 0.020683 -2023-02-13 17:31:30,747 - Epoch: [29][ 360/ 1207] Overall Loss 0.352781 Objective Loss 0.352781 LR 0.001000 Time 0.020629 -2023-02-13 17:31:30,935 - Epoch: [29][ 370/ 1207] Overall Loss 0.351990 Objective Loss 0.351990 LR 0.001000 Time 0.020578 -2023-02-13 17:31:31,123 - Epoch: [29][ 380/ 1207] Overall Loss 0.352826 Objective Loss 0.352826 LR 0.001000 Time 0.020530 -2023-02-13 17:31:31,312 - Epoch: [29][ 390/ 1207] Overall Loss 0.352674 Objective Loss 0.352674 LR 0.001000 Time 0.020487 -2023-02-13 17:31:31,500 - Epoch: [29][ 400/ 1207] Overall Loss 0.352899 Objective Loss 0.352899 LR 0.001000 Time 0.020445 -2023-02-13 17:31:31,688 - Epoch: [29][ 410/ 1207] Overall Loss 0.352645 Objective Loss 0.352645 LR 0.001000 Time 0.020405 -2023-02-13 17:31:31,877 - Epoch: [29][ 420/ 1207] Overall Loss 0.353004 Objective Loss 0.353004 LR 0.001000 Time 0.020366 -2023-02-13 17:31:32,065 - Epoch: [29][ 430/ 1207] Overall Loss 0.353013 Objective Loss 0.353013 LR 0.001000 Time 0.020329 -2023-02-13 17:31:32,253 - Epoch: [29][ 440/ 1207] Overall Loss 0.353428 Objective Loss 0.353428 LR 0.001000 Time 0.020294 -2023-02-13 17:31:32,441 - Epoch: [29][ 450/ 1207] Overall Loss 0.353753 Objective Loss 0.353753 LR 0.001000 Time 0.020260 -2023-02-13 17:31:32,630 - Epoch: [29][ 460/ 1207] Overall Loss 0.353113 Objective Loss 0.353113 LR 0.001000 Time 0.020230 -2023-02-13 17:31:32,818 - Epoch: [29][ 470/ 1207] Overall Loss 0.353405 Objective Loss 0.353405 LR 0.001000 Time 0.020198 -2023-02-13 17:31:33,006 - Epoch: [29][ 480/ 1207] Overall Loss 0.353003 Objective Loss 0.353003 LR 0.001000 Time 0.020168 -2023-02-13 17:31:33,194 - Epoch: [29][ 490/ 1207] Overall Loss 0.353968 Objective Loss 0.353968 LR 0.001000 Time 0.020139 -2023-02-13 17:31:33,382 - Epoch: [29][ 500/ 1207] Overall Loss 0.353234 Objective Loss 0.353234 LR 0.001000 Time 0.020112 -2023-02-13 17:31:33,581 - Epoch: [29][ 510/ 1207] Overall Loss 0.352998 Objective Loss 0.352998 LR 0.001000 Time 0.020108 -2023-02-13 17:31:33,781 - Epoch: [29][ 520/ 1207] Overall Loss 0.352688 Objective Loss 0.352688 LR 0.001000 Time 0.020105 -2023-02-13 17:31:33,986 - Epoch: [29][ 530/ 1207] Overall Loss 0.353290 Objective Loss 0.353290 LR 0.001000 Time 0.020112 -2023-02-13 17:31:34,186 - Epoch: [29][ 540/ 1207] Overall Loss 0.354085 Objective Loss 0.354085 LR 0.001000 Time 0.020108 -2023-02-13 17:31:34,390 - Epoch: [29][ 550/ 1207] Overall Loss 0.354080 Objective Loss 0.354080 LR 0.001000 Time 0.020114 -2023-02-13 17:31:34,591 - Epoch: [29][ 560/ 1207] Overall Loss 0.353866 Objective Loss 0.353866 LR 0.001000 Time 0.020111 -2023-02-13 17:31:34,796 - Epoch: [29][ 570/ 1207] Overall Loss 0.354274 Objective Loss 0.354274 LR 0.001000 Time 0.020118 -2023-02-13 17:31:34,995 - Epoch: [29][ 580/ 1207] Overall Loss 0.355021 Objective Loss 0.355021 LR 0.001000 Time 0.020114 -2023-02-13 17:31:35,200 - Epoch: [29][ 590/ 1207] Overall Loss 0.354955 Objective Loss 0.354955 LR 0.001000 Time 0.020120 -2023-02-13 17:31:35,399 - Epoch: [29][ 600/ 1207] Overall Loss 0.355139 Objective Loss 0.355139 LR 0.001000 Time 0.020116 -2023-02-13 17:31:35,605 - Epoch: [29][ 610/ 1207] Overall Loss 0.355575 Objective Loss 0.355575 LR 0.001000 Time 0.020123 -2023-02-13 17:31:35,806 - Epoch: [29][ 620/ 1207] Overall Loss 0.355979 Objective Loss 0.355979 LR 0.001000 Time 0.020121 -2023-02-13 17:31:36,011 - Epoch: [29][ 630/ 1207] Overall Loss 0.355874 Objective Loss 0.355874 LR 0.001000 Time 0.020127 -2023-02-13 17:31:36,211 - Epoch: [29][ 640/ 1207] Overall Loss 0.356276 Objective Loss 0.356276 LR 0.001000 Time 0.020124 -2023-02-13 17:31:36,416 - Epoch: [29][ 650/ 1207] Overall Loss 0.356693 Objective Loss 0.356693 LR 0.001000 Time 0.020129 -2023-02-13 17:31:36,616 - Epoch: [29][ 660/ 1207] Overall Loss 0.357150 Objective Loss 0.357150 LR 0.001000 Time 0.020127 -2023-02-13 17:31:36,822 - Epoch: [29][ 670/ 1207] Overall Loss 0.356972 Objective Loss 0.356972 LR 0.001000 Time 0.020133 -2023-02-13 17:31:37,022 - Epoch: [29][ 680/ 1207] Overall Loss 0.356827 Objective Loss 0.356827 LR 0.001000 Time 0.020130 -2023-02-13 17:31:37,227 - Epoch: [29][ 690/ 1207] Overall Loss 0.356771 Objective Loss 0.356771 LR 0.001000 Time 0.020136 -2023-02-13 17:31:37,427 - Epoch: [29][ 700/ 1207] Overall Loss 0.357131 Objective Loss 0.357131 LR 0.001000 Time 0.020132 -2023-02-13 17:31:37,633 - Epoch: [29][ 710/ 1207] Overall Loss 0.357442 Objective Loss 0.357442 LR 0.001000 Time 0.020138 -2023-02-13 17:31:37,832 - Epoch: [29][ 720/ 1207] Overall Loss 0.357773 Objective Loss 0.357773 LR 0.001000 Time 0.020135 -2023-02-13 17:31:38,037 - Epoch: [29][ 730/ 1207] Overall Loss 0.358276 Objective Loss 0.358276 LR 0.001000 Time 0.020139 -2023-02-13 17:31:38,237 - Epoch: [29][ 740/ 1207] Overall Loss 0.358705 Objective Loss 0.358705 LR 0.001000 Time 0.020136 -2023-02-13 17:31:38,440 - Epoch: [29][ 750/ 1207] Overall Loss 0.359345 Objective Loss 0.359345 LR 0.001000 Time 0.020138 -2023-02-13 17:31:38,640 - Epoch: [29][ 760/ 1207] Overall Loss 0.359322 Objective Loss 0.359322 LR 0.001000 Time 0.020136 -2023-02-13 17:31:38,844 - Epoch: [29][ 770/ 1207] Overall Loss 0.359652 Objective Loss 0.359652 LR 0.001000 Time 0.020139 -2023-02-13 17:31:39,043 - Epoch: [29][ 780/ 1207] Overall Loss 0.360002 Objective Loss 0.360002 LR 0.001000 Time 0.020136 -2023-02-13 17:31:39,248 - Epoch: [29][ 790/ 1207] Overall Loss 0.360547 Objective Loss 0.360547 LR 0.001000 Time 0.020140 -2023-02-13 17:31:39,447 - Epoch: [29][ 800/ 1207] Overall Loss 0.360992 Objective Loss 0.360992 LR 0.001000 Time 0.020137 -2023-02-13 17:31:39,652 - Epoch: [29][ 810/ 1207] Overall Loss 0.361113 Objective Loss 0.361113 LR 0.001000 Time 0.020140 -2023-02-13 17:31:39,851 - Epoch: [29][ 820/ 1207] Overall Loss 0.361621 Objective Loss 0.361621 LR 0.001000 Time 0.020137 -2023-02-13 17:31:40,056 - Epoch: [29][ 830/ 1207] Overall Loss 0.361484 Objective Loss 0.361484 LR 0.001000 Time 0.020140 -2023-02-13 17:31:40,257 - Epoch: [29][ 840/ 1207] Overall Loss 0.362007 Objective Loss 0.362007 LR 0.001000 Time 0.020140 -2023-02-13 17:31:40,461 - Epoch: [29][ 850/ 1207] Overall Loss 0.362654 Objective Loss 0.362654 LR 0.001000 Time 0.020142 -2023-02-13 17:31:40,661 - Epoch: [29][ 860/ 1207] Overall Loss 0.362304 Objective Loss 0.362304 LR 0.001000 Time 0.020140 -2023-02-13 17:31:40,866 - Epoch: [29][ 870/ 1207] Overall Loss 0.362319 Objective Loss 0.362319 LR 0.001000 Time 0.020144 -2023-02-13 17:31:41,066 - Epoch: [29][ 880/ 1207] Overall Loss 0.362487 Objective Loss 0.362487 LR 0.001000 Time 0.020142 -2023-02-13 17:31:41,270 - Epoch: [29][ 890/ 1207] Overall Loss 0.362356 Objective Loss 0.362356 LR 0.001000 Time 0.020144 -2023-02-13 17:31:41,470 - Epoch: [29][ 900/ 1207] Overall Loss 0.362420 Objective Loss 0.362420 LR 0.001000 Time 0.020142 -2023-02-13 17:31:41,675 - Epoch: [29][ 910/ 1207] Overall Loss 0.362473 Objective Loss 0.362473 LR 0.001000 Time 0.020146 -2023-02-13 17:31:41,875 - Epoch: [29][ 920/ 1207] Overall Loss 0.362507 Objective Loss 0.362507 LR 0.001000 Time 0.020144 -2023-02-13 17:31:42,080 - Epoch: [29][ 930/ 1207] Overall Loss 0.362453 Objective Loss 0.362453 LR 0.001000 Time 0.020147 -2023-02-13 17:31:42,282 - Epoch: [29][ 940/ 1207] Overall Loss 0.362372 Objective Loss 0.362372 LR 0.001000 Time 0.020147 -2023-02-13 17:31:42,486 - Epoch: [29][ 950/ 1207] Overall Loss 0.362047 Objective Loss 0.362047 LR 0.001000 Time 0.020149 -2023-02-13 17:31:42,687 - Epoch: [29][ 960/ 1207] Overall Loss 0.361655 Objective Loss 0.361655 LR 0.001000 Time 0.020148 -2023-02-13 17:31:42,890 - Epoch: [29][ 970/ 1207] Overall Loss 0.361715 Objective Loss 0.361715 LR 0.001000 Time 0.020150 -2023-02-13 17:31:43,089 - Epoch: [29][ 980/ 1207] Overall Loss 0.361563 Objective Loss 0.361563 LR 0.001000 Time 0.020147 -2023-02-13 17:31:43,295 - Epoch: [29][ 990/ 1207] Overall Loss 0.361696 Objective Loss 0.361696 LR 0.001000 Time 0.020151 -2023-02-13 17:31:43,496 - Epoch: [29][ 1000/ 1207] Overall Loss 0.361420 Objective Loss 0.361420 LR 0.001000 Time 0.020149 -2023-02-13 17:31:43,701 - Epoch: [29][ 1010/ 1207] Overall Loss 0.361511 Objective Loss 0.361511 LR 0.001000 Time 0.020153 -2023-02-13 17:31:43,900 - Epoch: [29][ 1020/ 1207] Overall Loss 0.361448 Objective Loss 0.361448 LR 0.001000 Time 0.020150 -2023-02-13 17:31:44,106 - Epoch: [29][ 1030/ 1207] Overall Loss 0.361297 Objective Loss 0.361297 LR 0.001000 Time 0.020154 -2023-02-13 17:31:44,305 - Epoch: [29][ 1040/ 1207] Overall Loss 0.360981 Objective Loss 0.360981 LR 0.001000 Time 0.020151 -2023-02-13 17:31:44,509 - Epoch: [29][ 1050/ 1207] Overall Loss 0.360937 Objective Loss 0.360937 LR 0.001000 Time 0.020153 -2023-02-13 17:31:44,709 - Epoch: [29][ 1060/ 1207] Overall Loss 0.360696 Objective Loss 0.360696 LR 0.001000 Time 0.020151 -2023-02-13 17:31:44,913 - Epoch: [29][ 1070/ 1207] Overall Loss 0.360485 Objective Loss 0.360485 LR 0.001000 Time 0.020153 -2023-02-13 17:31:45,113 - Epoch: [29][ 1080/ 1207] Overall Loss 0.360514 Objective Loss 0.360514 LR 0.001000 Time 0.020151 -2023-02-13 17:31:45,317 - Epoch: [29][ 1090/ 1207] Overall Loss 0.360743 Objective Loss 0.360743 LR 0.001000 Time 0.020153 -2023-02-13 17:31:45,516 - Epoch: [29][ 1100/ 1207] Overall Loss 0.360684 Objective Loss 0.360684 LR 0.001000 Time 0.020151 -2023-02-13 17:31:45,720 - Epoch: [29][ 1110/ 1207] Overall Loss 0.360677 Objective Loss 0.360677 LR 0.001000 Time 0.020153 -2023-02-13 17:31:45,920 - Epoch: [29][ 1120/ 1207] Overall Loss 0.360896 Objective Loss 0.360896 LR 0.001000 Time 0.020151 -2023-02-13 17:31:46,126 - Epoch: [29][ 1130/ 1207] Overall Loss 0.360707 Objective Loss 0.360707 LR 0.001000 Time 0.020154 -2023-02-13 17:31:46,326 - Epoch: [29][ 1140/ 1207] Overall Loss 0.360757 Objective Loss 0.360757 LR 0.001000 Time 0.020153 -2023-02-13 17:31:46,530 - Epoch: [29][ 1150/ 1207] Overall Loss 0.360656 Objective Loss 0.360656 LR 0.001000 Time 0.020155 -2023-02-13 17:31:46,730 - Epoch: [29][ 1160/ 1207] Overall Loss 0.360365 Objective Loss 0.360365 LR 0.001000 Time 0.020153 -2023-02-13 17:31:46,935 - Epoch: [29][ 1170/ 1207] Overall Loss 0.360374 Objective Loss 0.360374 LR 0.001000 Time 0.020156 -2023-02-13 17:31:47,135 - Epoch: [29][ 1180/ 1207] Overall Loss 0.360514 Objective Loss 0.360514 LR 0.001000 Time 0.020154 -2023-02-13 17:31:47,340 - Epoch: [29][ 1190/ 1207] Overall Loss 0.360724 Objective Loss 0.360724 LR 0.001000 Time 0.020156 -2023-02-13 17:31:47,593 - Epoch: [29][ 1200/ 1207] Overall Loss 0.360816 Objective Loss 0.360816 LR 0.001000 Time 0.020199 -2023-02-13 17:31:47,709 - Epoch: [29][ 1207/ 1207] Overall Loss 0.360765 Objective Loss 0.360765 Top1 82.621951 Top5 96.951220 LR 0.001000 Time 0.020178 -2023-02-13 17:31:47,780 - --- validate (epoch=29)----------- -2023-02-13 17:31:47,780 - 34311 samples (256 per mini-batch) -2023-02-13 17:31:48,274 - Epoch: [29][ 10/ 135] Loss 0.361430 Top1 81.328125 Top5 96.914062 -2023-02-13 17:31:48,419 - Epoch: [29][ 20/ 135] Loss 0.375732 Top1 81.132812 Top5 96.992188 -2023-02-13 17:31:48,559 - Epoch: [29][ 30/ 135] Loss 0.383435 Top1 80.625000 Top5 96.757812 -2023-02-13 17:31:48,695 - Epoch: [29][ 40/ 135] Loss 0.381489 Top1 80.380859 Top5 96.757812 -2023-02-13 17:31:48,832 - Epoch: [29][ 50/ 135] Loss 0.387441 Top1 80.101562 Top5 96.812500 -2023-02-13 17:31:48,972 - Epoch: [29][ 60/ 135] Loss 0.392181 Top1 80.045573 Top5 96.751302 -2023-02-13 17:31:49,099 - Epoch: [29][ 70/ 135] Loss 0.392088 Top1 80.217634 Top5 96.796875 -2023-02-13 17:31:49,226 - Epoch: [29][ 80/ 135] Loss 0.393410 Top1 80.253906 Top5 96.782227 -2023-02-13 17:31:49,353 - Epoch: [29][ 90/ 135] Loss 0.388619 Top1 80.507812 Top5 96.840278 -2023-02-13 17:31:49,476 - Epoch: [29][ 100/ 135] Loss 0.384965 Top1 80.460938 Top5 96.839844 -2023-02-13 17:31:49,603 - Epoch: [29][ 110/ 135] Loss 0.382944 Top1 80.525568 Top5 96.906960 -2023-02-13 17:31:49,728 - Epoch: [29][ 120/ 135] Loss 0.381777 Top1 80.491536 Top5 96.910807 -2023-02-13 17:31:49,858 - Epoch: [29][ 130/ 135] Loss 0.382045 Top1 80.549880 Top5 96.856971 -2023-02-13 17:31:49,905 - Epoch: [29][ 135/ 135] Loss 0.382703 Top1 80.557256 Top5 96.849407 -2023-02-13 17:31:49,984 - ==> Top1: 80.557 Top5: 96.849 Loss: 0.383 - -2023-02-13 17:31:49,985 - ==> Confusion: -[[ 848 3 7 3 7 1 0 3 7 59 0 3 0 3 6 2 2 3 5 2 3] - [ 7 942 4 3 8 19 2 16 6 1 2 1 0 1 0 1 6 2 7 1 4] - [ 11 3 954 6 4 0 9 19 0 1 2 2 3 6 5 6 3 5 12 1 6] - [ 7 1 28 872 3 1 0 1 2 2 15 0 4 1 19 0 3 7 41 1 8] - [ 30 13 0 1 972 4 0 0 3 3 1 5 1 3 7 9 7 1 1 2 3] - [ 5 53 2 5 11 883 3 30 2 12 1 12 5 19 2 4 6 5 2 5 3] - [ 4 1 33 3 0 1 1012 7 0 3 5 0 5 0 0 5 1 4 3 7 5] - [ 3 19 14 1 3 17 6 892 1 3 1 9 3 1 0 0 0 1 40 5 5] - [ 21 1 0 1 1 0 0 0 889 48 8 3 0 7 16 1 0 6 7 0 0] - [ 84 1 2 0 0 0 0 0 34 857 1 0 2 11 5 3 0 6 0 0 6] - [ 3 2 11 3 2 0 1 2 27 2 965 0 0 9 2 0 0 1 17 1 3] - [ 5 3 2 0 2 9 1 4 1 4 0 882 33 7 2 8 3 25 4 9 1] - [ 2 1 2 6 0 1 0 2 4 1 0 32 843 2 10 4 4 33 0 1 11] - [ 10 2 4 0 5 7 0 4 20 33 12 7 3 885 6 9 5 1 1 2 8] - [ 13 3 2 17 8 1 1 0 26 6 1 3 4 1 981 2 1 4 11 0 7] - [ 10 1 4 1 4 0 3 0 0 1 0 3 8 3 2 961 11 23 1 5 5] - [ 4 9 0 0 8 2 0 0 2 1 1 2 0 3 1 23 988 3 2 1 11] - [ 5 0 1 7 0 1 0 2 1 1 1 3 13 0 2 11 0 997 1 1 4] - [ 5 4 7 3 0 0 0 20 5 1 1 0 4 0 16 1 0 3 1015 1 0] - [ 1 3 2 1 1 11 5 32 2 0 1 22 3 8 0 5 9 7 2 1019 14] - [ 274 346 327 146 168 141 80 214 139 173 187 119 333 343 280 157 301 168 306 249 8983]] - -2023-02-13 17:31:49,987 - ==> Best [Top1: 80.557 Top5: 96.849 Sparsity:0.00 Params: 148928 on epoch: 29] -2023-02-13 17:31:49,987 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:31:49,993 - - -2023-02-13 17:31:49,994 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:31:50,879 - Epoch: [30][ 10/ 1207] Overall Loss 0.360211 Objective Loss 0.360211 LR 0.001000 Time 0.088482 -2023-02-13 17:31:51,080 - Epoch: [30][ 20/ 1207] Overall Loss 0.350454 Objective Loss 0.350454 LR 0.001000 Time 0.054275 -2023-02-13 17:31:51,272 - Epoch: [30][ 30/ 1207] Overall Loss 0.351047 Objective Loss 0.351047 LR 0.001000 Time 0.042573 -2023-02-13 17:31:51,466 - Epoch: [30][ 40/ 1207] Overall Loss 0.347168 Objective Loss 0.347168 LR 0.001000 Time 0.036758 -2023-02-13 17:31:51,659 - Epoch: [30][ 50/ 1207] Overall Loss 0.352608 Objective Loss 0.352608 LR 0.001000 Time 0.033261 -2023-02-13 17:31:51,852 - Epoch: [30][ 60/ 1207] Overall Loss 0.347901 Objective Loss 0.347901 LR 0.001000 Time 0.030939 -2023-02-13 17:31:52,045 - Epoch: [30][ 70/ 1207] Overall Loss 0.353096 Objective Loss 0.353096 LR 0.001000 Time 0.029259 -2023-02-13 17:31:52,240 - Epoch: [30][ 80/ 1207] Overall Loss 0.351495 Objective Loss 0.351495 LR 0.001000 Time 0.028038 -2023-02-13 17:31:52,432 - Epoch: [30][ 90/ 1207] Overall Loss 0.350184 Objective Loss 0.350184 LR 0.001000 Time 0.027051 -2023-02-13 17:31:52,626 - Epoch: [30][ 100/ 1207] Overall Loss 0.350735 Objective Loss 0.350735 LR 0.001000 Time 0.026280 -2023-02-13 17:31:52,817 - Epoch: [30][ 110/ 1207] Overall Loss 0.348400 Objective Loss 0.348400 LR 0.001000 Time 0.025624 -2023-02-13 17:31:53,010 - Epoch: [30][ 120/ 1207] Overall Loss 0.347006 Objective Loss 0.347006 LR 0.001000 Time 0.025095 -2023-02-13 17:31:53,201 - Epoch: [30][ 130/ 1207] Overall Loss 0.346516 Objective Loss 0.346516 LR 0.001000 Time 0.024632 -2023-02-13 17:31:53,393 - Epoch: [30][ 140/ 1207] Overall Loss 0.346051 Objective Loss 0.346051 LR 0.001000 Time 0.024245 -2023-02-13 17:31:53,586 - Epoch: [30][ 150/ 1207] Overall Loss 0.346261 Objective Loss 0.346261 LR 0.001000 Time 0.023907 -2023-02-13 17:31:53,779 - Epoch: [30][ 160/ 1207] Overall Loss 0.347144 Objective Loss 0.347144 LR 0.001000 Time 0.023622 -2023-02-13 17:31:53,970 - Epoch: [30][ 170/ 1207] Overall Loss 0.348014 Objective Loss 0.348014 LR 0.001000 Time 0.023354 -2023-02-13 17:31:54,164 - Epoch: [30][ 180/ 1207] Overall Loss 0.349366 Objective Loss 0.349366 LR 0.001000 Time 0.023130 -2023-02-13 17:31:54,355 - Epoch: [30][ 190/ 1207] Overall Loss 0.349041 Objective Loss 0.349041 LR 0.001000 Time 0.022919 -2023-02-13 17:31:54,549 - Epoch: [30][ 200/ 1207] Overall Loss 0.349546 Objective Loss 0.349546 LR 0.001000 Time 0.022739 -2023-02-13 17:31:54,742 - Epoch: [30][ 210/ 1207] Overall Loss 0.350485 Objective Loss 0.350485 LR 0.001000 Time 0.022571 -2023-02-13 17:31:54,935 - Epoch: [30][ 220/ 1207] Overall Loss 0.349403 Objective Loss 0.349403 LR 0.001000 Time 0.022420 -2023-02-13 17:31:55,126 - Epoch: [30][ 230/ 1207] Overall Loss 0.350382 Objective Loss 0.350382 LR 0.001000 Time 0.022277 -2023-02-13 17:31:55,320 - Epoch: [30][ 240/ 1207] Overall Loss 0.352211 Objective Loss 0.352211 LR 0.001000 Time 0.022157 -2023-02-13 17:31:55,512 - Epoch: [30][ 250/ 1207] Overall Loss 0.353139 Objective Loss 0.353139 LR 0.001000 Time 0.022035 -2023-02-13 17:31:55,706 - Epoch: [30][ 260/ 1207] Overall Loss 0.352276 Objective Loss 0.352276 LR 0.001000 Time 0.021934 -2023-02-13 17:31:55,899 - Epoch: [30][ 270/ 1207] Overall Loss 0.352263 Objective Loss 0.352263 LR 0.001000 Time 0.021834 -2023-02-13 17:31:56,092 - Epoch: [30][ 280/ 1207] Overall Loss 0.352542 Objective Loss 0.352542 LR 0.001000 Time 0.021743 -2023-02-13 17:31:56,284 - Epoch: [30][ 290/ 1207] Overall Loss 0.352561 Objective Loss 0.352561 LR 0.001000 Time 0.021654 -2023-02-13 17:31:56,478 - Epoch: [30][ 300/ 1207] Overall Loss 0.351903 Objective Loss 0.351903 LR 0.001000 Time 0.021575 -2023-02-13 17:31:56,670 - Epoch: [30][ 310/ 1207] Overall Loss 0.351942 Objective Loss 0.351942 LR 0.001000 Time 0.021498 -2023-02-13 17:31:56,864 - Epoch: [30][ 320/ 1207] Overall Loss 0.352578 Objective Loss 0.352578 LR 0.001000 Time 0.021431 -2023-02-13 17:31:57,055 - Epoch: [30][ 330/ 1207] Overall Loss 0.353430 Objective Loss 0.353430 LR 0.001000 Time 0.021361 -2023-02-13 17:31:57,249 - Epoch: [30][ 340/ 1207] Overall Loss 0.353763 Objective Loss 0.353763 LR 0.001000 Time 0.021302 -2023-02-13 17:31:57,441 - Epoch: [30][ 350/ 1207] Overall Loss 0.354138 Objective Loss 0.354138 LR 0.001000 Time 0.021240 -2023-02-13 17:31:57,635 - Epoch: [30][ 360/ 1207] Overall Loss 0.354802 Objective Loss 0.354802 LR 0.001000 Time 0.021188 -2023-02-13 17:31:57,827 - Epoch: [30][ 370/ 1207] Overall Loss 0.354884 Objective Loss 0.354884 LR 0.001000 Time 0.021134 -2023-02-13 17:31:58,021 - Epoch: [30][ 380/ 1207] Overall Loss 0.354715 Objective Loss 0.354715 LR 0.001000 Time 0.021086 -2023-02-13 17:31:58,214 - Epoch: [30][ 390/ 1207] Overall Loss 0.354015 Objective Loss 0.354015 LR 0.001000 Time 0.021039 -2023-02-13 17:31:58,407 - Epoch: [30][ 400/ 1207] Overall Loss 0.354989 Objective Loss 0.354989 LR 0.001000 Time 0.020996 -2023-02-13 17:31:58,599 - Epoch: [30][ 410/ 1207] Overall Loss 0.355080 Objective Loss 0.355080 LR 0.001000 Time 0.020951 -2023-02-13 17:31:58,793 - Epoch: [30][ 420/ 1207] Overall Loss 0.355410 Objective Loss 0.355410 LR 0.001000 Time 0.020913 -2023-02-13 17:31:58,985 - Epoch: [30][ 430/ 1207] Overall Loss 0.354958 Objective Loss 0.354958 LR 0.001000 Time 0.020872 -2023-02-13 17:31:59,179 - Epoch: [30][ 440/ 1207] Overall Loss 0.355522 Objective Loss 0.355522 LR 0.001000 Time 0.020837 -2023-02-13 17:31:59,371 - Epoch: [30][ 450/ 1207] Overall Loss 0.355043 Objective Loss 0.355043 LR 0.001000 Time 0.020800 -2023-02-13 17:31:59,564 - Epoch: [30][ 460/ 1207] Overall Loss 0.355315 Objective Loss 0.355315 LR 0.001000 Time 0.020768 -2023-02-13 17:31:59,756 - Epoch: [30][ 470/ 1207] Overall Loss 0.355297 Objective Loss 0.355297 LR 0.001000 Time 0.020734 -2023-02-13 17:31:59,952 - Epoch: [30][ 480/ 1207] Overall Loss 0.354770 Objective Loss 0.354770 LR 0.001000 Time 0.020708 -2023-02-13 17:32:00,147 - Epoch: [30][ 490/ 1207] Overall Loss 0.354373 Objective Loss 0.354373 LR 0.001000 Time 0.020683 -2023-02-13 17:32:00,343 - Epoch: [30][ 500/ 1207] Overall Loss 0.354591 Objective Loss 0.354591 LR 0.001000 Time 0.020661 -2023-02-13 17:32:00,539 - Epoch: [30][ 510/ 1207] Overall Loss 0.354209 Objective Loss 0.354209 LR 0.001000 Time 0.020639 -2023-02-13 17:32:00,736 - Epoch: [30][ 520/ 1207] Overall Loss 0.354271 Objective Loss 0.354271 LR 0.001000 Time 0.020620 -2023-02-13 17:32:00,932 - Epoch: [30][ 530/ 1207] Overall Loss 0.354161 Objective Loss 0.354161 LR 0.001000 Time 0.020601 -2023-02-13 17:32:01,128 - Epoch: [30][ 540/ 1207] Overall Loss 0.354789 Objective Loss 0.354789 LR 0.001000 Time 0.020582 -2023-02-13 17:32:01,324 - Epoch: [30][ 550/ 1207] Overall Loss 0.355147 Objective Loss 0.355147 LR 0.001000 Time 0.020562 -2023-02-13 17:32:01,520 - Epoch: [30][ 560/ 1207] Overall Loss 0.355440 Objective Loss 0.355440 LR 0.001000 Time 0.020544 -2023-02-13 17:32:01,716 - Epoch: [30][ 570/ 1207] Overall Loss 0.355535 Objective Loss 0.355535 LR 0.001000 Time 0.020528 -2023-02-13 17:32:01,912 - Epoch: [30][ 580/ 1207] Overall Loss 0.355611 Objective Loss 0.355611 LR 0.001000 Time 0.020511 -2023-02-13 17:32:02,108 - Epoch: [30][ 590/ 1207] Overall Loss 0.355302 Objective Loss 0.355302 LR 0.001000 Time 0.020494 -2023-02-13 17:32:02,303 - Epoch: [30][ 600/ 1207] Overall Loss 0.355159 Objective Loss 0.355159 LR 0.001000 Time 0.020478 -2023-02-13 17:32:02,499 - Epoch: [30][ 610/ 1207] Overall Loss 0.355024 Objective Loss 0.355024 LR 0.001000 Time 0.020463 -2023-02-13 17:32:02,696 - Epoch: [30][ 620/ 1207] Overall Loss 0.355360 Objective Loss 0.355360 LR 0.001000 Time 0.020450 -2023-02-13 17:32:02,891 - Epoch: [30][ 630/ 1207] Overall Loss 0.355905 Objective Loss 0.355905 LR 0.001000 Time 0.020434 -2023-02-13 17:32:03,087 - Epoch: [30][ 640/ 1207] Overall Loss 0.355583 Objective Loss 0.355583 LR 0.001000 Time 0.020421 -2023-02-13 17:32:03,283 - Epoch: [30][ 650/ 1207] Overall Loss 0.356273 Objective Loss 0.356273 LR 0.001000 Time 0.020407 -2023-02-13 17:32:03,478 - Epoch: [30][ 660/ 1207] Overall Loss 0.356337 Objective Loss 0.356337 LR 0.001000 Time 0.020393 -2023-02-13 17:32:03,673 - Epoch: [30][ 670/ 1207] Overall Loss 0.356644 Objective Loss 0.356644 LR 0.001000 Time 0.020379 -2023-02-13 17:32:03,868 - Epoch: [30][ 680/ 1207] Overall Loss 0.356988 Objective Loss 0.356988 LR 0.001000 Time 0.020366 -2023-02-13 17:32:04,064 - Epoch: [30][ 690/ 1207] Overall Loss 0.357173 Objective Loss 0.357173 LR 0.001000 Time 0.020354 -2023-02-13 17:32:04,261 - Epoch: [30][ 700/ 1207] Overall Loss 0.357412 Objective Loss 0.357412 LR 0.001000 Time 0.020344 -2023-02-13 17:32:04,457 - Epoch: [30][ 710/ 1207] Overall Loss 0.357633 Objective Loss 0.357633 LR 0.001000 Time 0.020332 -2023-02-13 17:32:04,653 - Epoch: [30][ 720/ 1207] Overall Loss 0.357472 Objective Loss 0.357472 LR 0.001000 Time 0.020322 -2023-02-13 17:32:04,848 - Epoch: [30][ 730/ 1207] Overall Loss 0.357249 Objective Loss 0.357249 LR 0.001000 Time 0.020311 -2023-02-13 17:32:05,044 - Epoch: [30][ 740/ 1207] Overall Loss 0.356711 Objective Loss 0.356711 LR 0.001000 Time 0.020300 -2023-02-13 17:32:05,240 - Epoch: [30][ 750/ 1207] Overall Loss 0.356864 Objective Loss 0.356864 LR 0.001000 Time 0.020290 -2023-02-13 17:32:05,435 - Epoch: [30][ 760/ 1207] Overall Loss 0.357439 Objective Loss 0.357439 LR 0.001000 Time 0.020280 -2023-02-13 17:32:05,631 - Epoch: [30][ 770/ 1207] Overall Loss 0.357446 Objective Loss 0.357446 LR 0.001000 Time 0.020270 -2023-02-13 17:32:05,828 - Epoch: [30][ 780/ 1207] Overall Loss 0.357254 Objective Loss 0.357254 LR 0.001000 Time 0.020262 -2023-02-13 17:32:06,023 - Epoch: [30][ 790/ 1207] Overall Loss 0.357567 Objective Loss 0.357567 LR 0.001000 Time 0.020253 -2023-02-13 17:32:06,219 - Epoch: [30][ 800/ 1207] Overall Loss 0.357325 Objective Loss 0.357325 LR 0.001000 Time 0.020244 -2023-02-13 17:32:06,415 - Epoch: [30][ 810/ 1207] Overall Loss 0.357402 Objective Loss 0.357402 LR 0.001000 Time 0.020235 -2023-02-13 17:32:06,611 - Epoch: [30][ 820/ 1207] Overall Loss 0.357455 Objective Loss 0.357455 LR 0.001000 Time 0.020227 -2023-02-13 17:32:06,807 - Epoch: [30][ 830/ 1207] Overall Loss 0.357362 Objective Loss 0.357362 LR 0.001000 Time 0.020219 -2023-02-13 17:32:07,002 - Epoch: [30][ 840/ 1207] Overall Loss 0.357767 Objective Loss 0.357767 LR 0.001000 Time 0.020211 -2023-02-13 17:32:07,197 - Epoch: [30][ 850/ 1207] Overall Loss 0.358182 Objective Loss 0.358182 LR 0.001000 Time 0.020202 -2023-02-13 17:32:07,393 - Epoch: [30][ 860/ 1207] Overall Loss 0.358527 Objective Loss 0.358527 LR 0.001000 Time 0.020194 -2023-02-13 17:32:07,589 - Epoch: [30][ 870/ 1207] Overall Loss 0.358577 Objective Loss 0.358577 LR 0.001000 Time 0.020187 -2023-02-13 17:32:07,786 - Epoch: [30][ 880/ 1207] Overall Loss 0.358834 Objective Loss 0.358834 LR 0.001000 Time 0.020181 -2023-02-13 17:32:07,981 - Epoch: [30][ 890/ 1207] Overall Loss 0.359057 Objective Loss 0.359057 LR 0.001000 Time 0.020173 -2023-02-13 17:32:08,178 - Epoch: [30][ 900/ 1207] Overall Loss 0.359429 Objective Loss 0.359429 LR 0.001000 Time 0.020167 -2023-02-13 17:32:08,374 - Epoch: [30][ 910/ 1207] Overall Loss 0.359681 Objective Loss 0.359681 LR 0.001000 Time 0.020160 -2023-02-13 17:32:08,570 - Epoch: [30][ 920/ 1207] Overall Loss 0.359449 Objective Loss 0.359449 LR 0.001000 Time 0.020154 -2023-02-13 17:32:08,766 - Epoch: [30][ 930/ 1207] Overall Loss 0.359528 Objective Loss 0.359528 LR 0.001000 Time 0.020147 -2023-02-13 17:32:08,961 - Epoch: [30][ 940/ 1207] Overall Loss 0.359754 Objective Loss 0.359754 LR 0.001000 Time 0.020140 -2023-02-13 17:32:09,156 - Epoch: [30][ 950/ 1207] Overall Loss 0.359824 Objective Loss 0.359824 LR 0.001000 Time 0.020133 -2023-02-13 17:32:09,352 - Epoch: [30][ 960/ 1207] Overall Loss 0.360239 Objective Loss 0.360239 LR 0.001000 Time 0.020127 -2023-02-13 17:32:09,546 - Epoch: [30][ 970/ 1207] Overall Loss 0.360144 Objective Loss 0.360144 LR 0.001000 Time 0.020119 -2023-02-13 17:32:09,740 - Epoch: [30][ 980/ 1207] Overall Loss 0.360159 Objective Loss 0.360159 LR 0.001000 Time 0.020112 -2023-02-13 17:32:09,932 - Epoch: [30][ 990/ 1207] Overall Loss 0.359881 Objective Loss 0.359881 LR 0.001000 Time 0.020102 -2023-02-13 17:32:10,126 - Epoch: [30][ 1000/ 1207] Overall Loss 0.359641 Objective Loss 0.359641 LR 0.001000 Time 0.020095 -2023-02-13 17:32:10,319 - Epoch: [30][ 1010/ 1207] Overall Loss 0.359444 Objective Loss 0.359444 LR 0.001000 Time 0.020086 -2023-02-13 17:32:10,512 - Epoch: [30][ 1020/ 1207] Overall Loss 0.359285 Objective Loss 0.359285 LR 0.001000 Time 0.020078 -2023-02-13 17:32:10,706 - Epoch: [30][ 1030/ 1207] Overall Loss 0.359103 Objective Loss 0.359103 LR 0.001000 Time 0.020071 -2023-02-13 17:32:10,901 - Epoch: [30][ 1040/ 1207] Overall Loss 0.359131 Objective Loss 0.359131 LR 0.001000 Time 0.020065 -2023-02-13 17:32:11,093 - Epoch: [30][ 1050/ 1207] Overall Loss 0.359374 Objective Loss 0.359374 LR 0.001000 Time 0.020057 -2023-02-13 17:32:11,287 - Epoch: [30][ 1060/ 1207] Overall Loss 0.359679 Objective Loss 0.359679 LR 0.001000 Time 0.020050 -2023-02-13 17:32:11,479 - Epoch: [30][ 1070/ 1207] Overall Loss 0.359346 Objective Loss 0.359346 LR 0.001000 Time 0.020042 -2023-02-13 17:32:11,674 - Epoch: [30][ 1080/ 1207] Overall Loss 0.359250 Objective Loss 0.359250 LR 0.001000 Time 0.020036 -2023-02-13 17:32:11,867 - Epoch: [30][ 1090/ 1207] Overall Loss 0.359171 Objective Loss 0.359171 LR 0.001000 Time 0.020029 -2023-02-13 17:32:12,060 - Epoch: [30][ 1100/ 1207] Overall Loss 0.359254 Objective Loss 0.359254 LR 0.001000 Time 0.020023 -2023-02-13 17:32:12,253 - Epoch: [30][ 1110/ 1207] Overall Loss 0.359334 Objective Loss 0.359334 LR 0.001000 Time 0.020015 -2023-02-13 17:32:12,447 - Epoch: [30][ 1120/ 1207] Overall Loss 0.359308 Objective Loss 0.359308 LR 0.001000 Time 0.020010 -2023-02-13 17:32:12,640 - Epoch: [30][ 1130/ 1207] Overall Loss 0.359415 Objective Loss 0.359415 LR 0.001000 Time 0.020003 -2023-02-13 17:32:12,833 - Epoch: [30][ 1140/ 1207] Overall Loss 0.359597 Objective Loss 0.359597 LR 0.001000 Time 0.019997 -2023-02-13 17:32:13,025 - Epoch: [30][ 1150/ 1207] Overall Loss 0.359894 Objective Loss 0.359894 LR 0.001000 Time 0.019989 -2023-02-13 17:32:13,219 - Epoch: [30][ 1160/ 1207] Overall Loss 0.359694 Objective Loss 0.359694 LR 0.001000 Time 0.019984 -2023-02-13 17:32:13,411 - Epoch: [30][ 1170/ 1207] Overall Loss 0.359758 Objective Loss 0.359758 LR 0.001000 Time 0.019977 -2023-02-13 17:32:13,605 - Epoch: [30][ 1180/ 1207] Overall Loss 0.359725 Objective Loss 0.359725 LR 0.001000 Time 0.019972 -2023-02-13 17:32:13,799 - Epoch: [30][ 1190/ 1207] Overall Loss 0.359795 Objective Loss 0.359795 LR 0.001000 Time 0.019966 -2023-02-13 17:32:14,046 - Epoch: [30][ 1200/ 1207] Overall Loss 0.359937 Objective Loss 0.359937 LR 0.001000 Time 0.020006 -2023-02-13 17:32:14,161 - Epoch: [30][ 1207/ 1207] Overall Loss 0.360078 Objective Loss 0.360078 Top1 78.963415 Top5 97.560976 LR 0.001000 Time 0.019985 -2023-02-13 17:32:14,233 - --- validate (epoch=30)----------- -2023-02-13 17:32:14,233 - 34311 samples (256 per mini-batch) -2023-02-13 17:32:14,631 - Epoch: [30][ 10/ 135] Loss 0.408910 Top1 79.414062 Top5 96.796875 -2023-02-13 17:32:14,759 - Epoch: [30][ 20/ 135] Loss 0.404922 Top1 79.316406 Top5 96.679688 -2023-02-13 17:32:14,887 - Epoch: [30][ 30/ 135] Loss 0.399974 Top1 79.361979 Top5 96.614583 -2023-02-13 17:32:15,015 - Epoch: [30][ 40/ 135] Loss 0.403606 Top1 79.218750 Top5 96.572266 -2023-02-13 17:32:15,142 - Epoch: [30][ 50/ 135] Loss 0.393595 Top1 79.562500 Top5 96.664062 -2023-02-13 17:32:15,267 - Epoch: [30][ 60/ 135] Loss 0.388684 Top1 79.713542 Top5 96.640625 -2023-02-13 17:32:15,393 - Epoch: [30][ 70/ 135] Loss 0.383325 Top1 79.681920 Top5 96.724330 -2023-02-13 17:32:15,521 - Epoch: [30][ 80/ 135] Loss 0.381150 Top1 79.755859 Top5 96.772461 -2023-02-13 17:32:15,658 - Epoch: [30][ 90/ 135] Loss 0.383100 Top1 79.887153 Top5 96.731771 -2023-02-13 17:32:15,798 - Epoch: [30][ 100/ 135] Loss 0.380401 Top1 79.976562 Top5 96.773438 -2023-02-13 17:32:15,921 - Epoch: [30][ 110/ 135] Loss 0.383573 Top1 79.875710 Top5 96.743608 -2023-02-13 17:32:16,054 - Epoch: [30][ 120/ 135] Loss 0.379398 Top1 79.941406 Top5 96.787109 -2023-02-13 17:32:16,191 - Epoch: [30][ 130/ 135] Loss 0.383190 Top1 79.906851 Top5 96.754808 -2023-02-13 17:32:16,235 - Epoch: [30][ 135/ 135] Loss 0.381370 Top1 79.901489 Top5 96.767800 -2023-02-13 17:32:16,307 - ==> Top1: 79.901 Top5: 96.768 Loss: 0.381 - -2023-02-13 17:32:16,308 - ==> Confusion: -[[ 835 6 7 0 10 3 0 5 6 54 0 6 1 6 10 2 7 2 2 0 5] - [ 4 905 2 5 4 56 5 15 5 2 1 5 1 2 4 1 8 0 3 1 4] - [ 13 6 928 24 3 1 13 16 1 0 3 3 2 2 6 9 5 4 6 5 8] - [ 5 2 16 917 0 3 1 1 2 1 14 1 5 1 21 0 7 5 11 0 3] - [ 27 17 2 0 952 6 2 0 1 6 0 8 0 2 15 5 13 0 1 4 5] - [ 4 16 0 5 3 953 3 17 4 4 3 17 5 12 1 1 10 2 2 3 5] - [ 5 4 20 5 0 5 1013 6 0 1 4 4 3 2 0 4 3 3 1 13 3] - [ 2 10 12 4 3 45 3 892 0 1 5 9 2 2 0 0 2 1 19 8 4] - [ 17 1 0 1 0 0 0 1 884 42 6 2 2 6 32 0 3 5 4 1 2] - [ 103 2 2 0 4 1 0 2 48 811 1 2 0 15 6 3 1 4 0 0 7] - [ 2 4 4 8 0 2 1 1 22 1 962 4 3 11 3 0 2 1 14 1 5] - [ 5 3 2 0 1 6 1 2 0 1 0 920 28 6 1 3 3 11 3 8 1] - [ 2 2 1 4 1 4 0 2 5 0 0 37 864 1 2 4 6 17 0 0 7] - [ 11 3 1 0 6 13 0 4 22 27 10 11 3 877 9 3 7 0 1 4 12] - [ 11 4 0 22 5 2 0 3 20 6 2 1 3 1 989 1 5 5 6 0 6] - [ 7 2 5 1 7 1 4 2 1 1 0 5 13 1 1 939 17 26 0 6 7] - [ 2 6 0 0 9 3 0 1 3 0 0 4 0 3 1 10 1006 4 2 4 3] - [ 5 3 0 8 1 1 0 2 2 0 0 12 30 2 1 9 1 968 1 1 4] - [ 4 9 6 16 2 2 0 35 3 1 3 2 10 0 24 0 1 2 964 1 1] - [ 2 1 0 1 0 15 6 14 2 0 2 26 7 6 0 5 5 4 0 1042 10] - [ 211 333 275 201 145 273 65 202 143 88 171 193 429 259 283 123 548 158 200 340 8794]] - -2023-02-13 17:32:16,309 - ==> Best [Top1: 80.557 Top5: 96.849 Sparsity:0.00 Params: 148928 on epoch: 29] -2023-02-13 17:32:16,309 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:32:16,315 - - -2023-02-13 17:32:16,315 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:32:17,275 - Epoch: [31][ 10/ 1207] Overall Loss 0.340695 Objective Loss 0.340695 LR 0.001000 Time 0.095932 -2023-02-13 17:32:17,464 - Epoch: [31][ 20/ 1207] Overall Loss 0.347091 Objective Loss 0.347091 LR 0.001000 Time 0.057417 -2023-02-13 17:32:17,654 - Epoch: [31][ 30/ 1207] Overall Loss 0.347130 Objective Loss 0.347130 LR 0.001000 Time 0.044571 -2023-02-13 17:32:17,842 - Epoch: [31][ 40/ 1207] Overall Loss 0.351675 Objective Loss 0.351675 LR 0.001000 Time 0.038118 -2023-02-13 17:32:18,029 - Epoch: [31][ 50/ 1207] Overall Loss 0.351224 Objective Loss 0.351224 LR 0.001000 Time 0.034233 -2023-02-13 17:32:18,216 - Epoch: [31][ 60/ 1207] Overall Loss 0.349860 Objective Loss 0.349860 LR 0.001000 Time 0.031646 -2023-02-13 17:32:18,404 - Epoch: [31][ 70/ 1207] Overall Loss 0.342689 Objective Loss 0.342689 LR 0.001000 Time 0.029795 -2023-02-13 17:32:18,592 - Epoch: [31][ 80/ 1207] Overall Loss 0.343546 Objective Loss 0.343546 LR 0.001000 Time 0.028416 -2023-02-13 17:32:18,781 - Epoch: [31][ 90/ 1207] Overall Loss 0.342366 Objective Loss 0.342366 LR 0.001000 Time 0.027355 -2023-02-13 17:32:18,969 - Epoch: [31][ 100/ 1207] Overall Loss 0.341425 Objective Loss 0.341425 LR 0.001000 Time 0.026497 -2023-02-13 17:32:19,157 - Epoch: [31][ 110/ 1207] Overall Loss 0.340291 Objective Loss 0.340291 LR 0.001000 Time 0.025796 -2023-02-13 17:32:19,348 - Epoch: [31][ 120/ 1207] Overall Loss 0.339052 Objective Loss 0.339052 LR 0.001000 Time 0.025238 -2023-02-13 17:32:19,544 - Epoch: [31][ 130/ 1207] Overall Loss 0.339002 Objective Loss 0.339002 LR 0.001000 Time 0.024795 -2023-02-13 17:32:19,738 - Epoch: [31][ 140/ 1207] Overall Loss 0.337568 Objective Loss 0.337568 LR 0.001000 Time 0.024411 -2023-02-13 17:32:19,934 - Epoch: [31][ 150/ 1207] Overall Loss 0.338408 Objective Loss 0.338408 LR 0.001000 Time 0.024086 -2023-02-13 17:32:20,128 - Epoch: [31][ 160/ 1207] Overall Loss 0.340813 Objective Loss 0.340813 LR 0.001000 Time 0.023788 -2023-02-13 17:32:20,324 - Epoch: [31][ 170/ 1207] Overall Loss 0.343836 Objective Loss 0.343836 LR 0.001000 Time 0.023544 -2023-02-13 17:32:20,518 - Epoch: [31][ 180/ 1207] Overall Loss 0.345554 Objective Loss 0.345554 LR 0.001000 Time 0.023308 -2023-02-13 17:32:20,714 - Epoch: [31][ 190/ 1207] Overall Loss 0.347056 Objective Loss 0.347056 LR 0.001000 Time 0.023112 -2023-02-13 17:32:20,908 - Epoch: [31][ 200/ 1207] Overall Loss 0.347964 Objective Loss 0.347964 LR 0.001000 Time 0.022927 -2023-02-13 17:32:21,104 - Epoch: [31][ 210/ 1207] Overall Loss 0.350633 Objective Loss 0.350633 LR 0.001000 Time 0.022766 -2023-02-13 17:32:21,298 - Epoch: [31][ 220/ 1207] Overall Loss 0.350660 Objective Loss 0.350660 LR 0.001000 Time 0.022610 -2023-02-13 17:32:21,494 - Epoch: [31][ 230/ 1207] Overall Loss 0.349942 Objective Loss 0.349942 LR 0.001000 Time 0.022477 -2023-02-13 17:32:21,689 - Epoch: [31][ 240/ 1207] Overall Loss 0.349552 Objective Loss 0.349552 LR 0.001000 Time 0.022351 -2023-02-13 17:32:21,885 - Epoch: [31][ 250/ 1207] Overall Loss 0.349231 Objective Loss 0.349231 LR 0.001000 Time 0.022241 -2023-02-13 17:32:22,079 - Epoch: [31][ 260/ 1207] Overall Loss 0.349151 Objective Loss 0.349151 LR 0.001000 Time 0.022129 -2023-02-13 17:32:22,274 - Epoch: [31][ 270/ 1207] Overall Loss 0.350186 Objective Loss 0.350186 LR 0.001000 Time 0.022033 -2023-02-13 17:32:22,469 - Epoch: [31][ 280/ 1207] Overall Loss 0.349503 Objective Loss 0.349503 LR 0.001000 Time 0.021939 -2023-02-13 17:32:22,665 - Epoch: [31][ 290/ 1207] Overall Loss 0.351102 Objective Loss 0.351102 LR 0.001000 Time 0.021858 -2023-02-13 17:32:22,859 - Epoch: [31][ 300/ 1207] Overall Loss 0.351054 Objective Loss 0.351054 LR 0.001000 Time 0.021775 -2023-02-13 17:32:23,055 - Epoch: [31][ 310/ 1207] Overall Loss 0.350809 Objective Loss 0.350809 LR 0.001000 Time 0.021703 -2023-02-13 17:32:23,249 - Epoch: [31][ 320/ 1207] Overall Loss 0.349762 Objective Loss 0.349762 LR 0.001000 Time 0.021630 -2023-02-13 17:32:23,445 - Epoch: [31][ 330/ 1207] Overall Loss 0.351449 Objective Loss 0.351449 LR 0.001000 Time 0.021568 -2023-02-13 17:32:23,639 - Epoch: [31][ 340/ 1207] Overall Loss 0.351738 Objective Loss 0.351738 LR 0.001000 Time 0.021501 -2023-02-13 17:32:23,836 - Epoch: [31][ 350/ 1207] Overall Loss 0.352609 Objective Loss 0.352609 LR 0.001000 Time 0.021449 -2023-02-13 17:32:24,030 - Epoch: [31][ 360/ 1207] Overall Loss 0.352332 Objective Loss 0.352332 LR 0.001000 Time 0.021393 -2023-02-13 17:32:24,226 - Epoch: [31][ 370/ 1207] Overall Loss 0.352905 Objective Loss 0.352905 LR 0.001000 Time 0.021342 -2023-02-13 17:32:24,420 - Epoch: [31][ 380/ 1207] Overall Loss 0.352384 Objective Loss 0.352384 LR 0.001000 Time 0.021291 -2023-02-13 17:32:24,616 - Epoch: [31][ 390/ 1207] Overall Loss 0.353107 Objective Loss 0.353107 LR 0.001000 Time 0.021246 -2023-02-13 17:32:24,810 - Epoch: [31][ 400/ 1207] Overall Loss 0.353289 Objective Loss 0.353289 LR 0.001000 Time 0.021199 -2023-02-13 17:32:25,006 - Epoch: [31][ 410/ 1207] Overall Loss 0.353484 Objective Loss 0.353484 LR 0.001000 Time 0.021160 -2023-02-13 17:32:25,201 - Epoch: [31][ 420/ 1207] Overall Loss 0.353831 Objective Loss 0.353831 LR 0.001000 Time 0.021118 -2023-02-13 17:32:25,397 - Epoch: [31][ 430/ 1207] Overall Loss 0.354907 Objective Loss 0.354907 LR 0.001000 Time 0.021082 -2023-02-13 17:32:25,590 - Epoch: [31][ 440/ 1207] Overall Loss 0.355218 Objective Loss 0.355218 LR 0.001000 Time 0.021041 -2023-02-13 17:32:25,788 - Epoch: [31][ 450/ 1207] Overall Loss 0.355857 Objective Loss 0.355857 LR 0.001000 Time 0.021013 -2023-02-13 17:32:25,982 - Epoch: [31][ 460/ 1207] Overall Loss 0.355728 Objective Loss 0.355728 LR 0.001000 Time 0.020977 -2023-02-13 17:32:26,179 - Epoch: [31][ 470/ 1207] Overall Loss 0.355418 Objective Loss 0.355418 LR 0.001000 Time 0.020948 -2023-02-13 17:32:26,373 - Epoch: [31][ 480/ 1207] Overall Loss 0.355491 Objective Loss 0.355491 LR 0.001000 Time 0.020915 -2023-02-13 17:32:26,568 - Epoch: [31][ 490/ 1207] Overall Loss 0.355916 Objective Loss 0.355916 LR 0.001000 Time 0.020887 -2023-02-13 17:32:26,763 - Epoch: [31][ 500/ 1207] Overall Loss 0.356002 Objective Loss 0.356002 LR 0.001000 Time 0.020858 -2023-02-13 17:32:26,959 - Epoch: [31][ 510/ 1207] Overall Loss 0.355706 Objective Loss 0.355706 LR 0.001000 Time 0.020832 -2023-02-13 17:32:27,153 - Epoch: [31][ 520/ 1207] Overall Loss 0.356140 Objective Loss 0.356140 LR 0.001000 Time 0.020803 -2023-02-13 17:32:27,349 - Epoch: [31][ 530/ 1207] Overall Loss 0.356666 Objective Loss 0.356666 LR 0.001000 Time 0.020780 -2023-02-13 17:32:27,543 - Epoch: [31][ 540/ 1207] Overall Loss 0.357272 Objective Loss 0.357272 LR 0.001000 Time 0.020753 -2023-02-13 17:32:27,739 - Epoch: [31][ 550/ 1207] Overall Loss 0.357183 Objective Loss 0.357183 LR 0.001000 Time 0.020733 -2023-02-13 17:32:27,934 - Epoch: [31][ 560/ 1207] Overall Loss 0.357381 Objective Loss 0.357381 LR 0.001000 Time 0.020709 -2023-02-13 17:32:28,130 - Epoch: [31][ 570/ 1207] Overall Loss 0.358097 Objective Loss 0.358097 LR 0.001000 Time 0.020690 -2023-02-13 17:32:28,325 - Epoch: [31][ 580/ 1207] Overall Loss 0.358844 Objective Loss 0.358844 LR 0.001000 Time 0.020669 -2023-02-13 17:32:28,521 - Epoch: [31][ 590/ 1207] Overall Loss 0.358846 Objective Loss 0.358846 LR 0.001000 Time 0.020650 -2023-02-13 17:32:28,716 - Epoch: [31][ 600/ 1207] Overall Loss 0.359830 Objective Loss 0.359830 LR 0.001000 Time 0.020629 -2023-02-13 17:32:28,912 - Epoch: [31][ 610/ 1207] Overall Loss 0.359986 Objective Loss 0.359986 LR 0.001000 Time 0.020612 -2023-02-13 17:32:29,106 - Epoch: [31][ 620/ 1207] Overall Loss 0.360351 Objective Loss 0.360351 LR 0.001000 Time 0.020592 -2023-02-13 17:32:29,302 - Epoch: [31][ 630/ 1207] Overall Loss 0.360384 Objective Loss 0.360384 LR 0.001000 Time 0.020575 -2023-02-13 17:32:29,495 - Epoch: [31][ 640/ 1207] Overall Loss 0.360877 Objective Loss 0.360877 LR 0.001000 Time 0.020556 -2023-02-13 17:32:29,691 - Epoch: [31][ 650/ 1207] Overall Loss 0.361158 Objective Loss 0.361158 LR 0.001000 Time 0.020541 -2023-02-13 17:32:29,885 - Epoch: [31][ 660/ 1207] Overall Loss 0.361538 Objective Loss 0.361538 LR 0.001000 Time 0.020523 -2023-02-13 17:32:30,081 - Epoch: [31][ 670/ 1207] Overall Loss 0.361555 Objective Loss 0.361555 LR 0.001000 Time 0.020508 -2023-02-13 17:32:30,275 - Epoch: [31][ 680/ 1207] Overall Loss 0.362200 Objective Loss 0.362200 LR 0.001000 Time 0.020491 -2023-02-13 17:32:30,471 - Epoch: [31][ 690/ 1207] Overall Loss 0.362702 Objective Loss 0.362702 LR 0.001000 Time 0.020478 -2023-02-13 17:32:30,665 - Epoch: [31][ 700/ 1207] Overall Loss 0.362797 Objective Loss 0.362797 LR 0.001000 Time 0.020462 -2023-02-13 17:32:30,863 - Epoch: [31][ 710/ 1207] Overall Loss 0.362617 Objective Loss 0.362617 LR 0.001000 Time 0.020451 -2023-02-13 17:32:31,056 - Epoch: [31][ 720/ 1207] Overall Loss 0.362448 Objective Loss 0.362448 LR 0.001000 Time 0.020436 -2023-02-13 17:32:31,253 - Epoch: [31][ 730/ 1207] Overall Loss 0.362909 Objective Loss 0.362909 LR 0.001000 Time 0.020424 -2023-02-13 17:32:31,447 - Epoch: [31][ 740/ 1207] Overall Loss 0.363083 Objective Loss 0.363083 LR 0.001000 Time 0.020410 -2023-02-13 17:32:31,643 - Epoch: [31][ 750/ 1207] Overall Loss 0.363195 Objective Loss 0.363195 LR 0.001000 Time 0.020400 -2023-02-13 17:32:31,839 - Epoch: [31][ 760/ 1207] Overall Loss 0.363246 Objective Loss 0.363246 LR 0.001000 Time 0.020388 -2023-02-13 17:32:32,035 - Epoch: [31][ 770/ 1207] Overall Loss 0.363610 Objective Loss 0.363610 LR 0.001000 Time 0.020377 -2023-02-13 17:32:32,229 - Epoch: [31][ 780/ 1207] Overall Loss 0.363636 Objective Loss 0.363636 LR 0.001000 Time 0.020365 -2023-02-13 17:32:32,426 - Epoch: [31][ 790/ 1207] Overall Loss 0.363353 Objective Loss 0.363353 LR 0.001000 Time 0.020355 -2023-02-13 17:32:32,620 - Epoch: [31][ 800/ 1207] Overall Loss 0.363527 Objective Loss 0.363527 LR 0.001000 Time 0.020342 -2023-02-13 17:32:32,816 - Epoch: [31][ 810/ 1207] Overall Loss 0.363313 Objective Loss 0.363313 LR 0.001000 Time 0.020333 -2023-02-13 17:32:33,010 - Epoch: [31][ 820/ 1207] Overall Loss 0.363417 Objective Loss 0.363417 LR 0.001000 Time 0.020321 -2023-02-13 17:32:33,206 - Epoch: [31][ 830/ 1207] Overall Loss 0.363595 Objective Loss 0.363595 LR 0.001000 Time 0.020312 -2023-02-13 17:32:33,400 - Epoch: [31][ 840/ 1207] Overall Loss 0.363559 Objective Loss 0.363559 LR 0.001000 Time 0.020302 -2023-02-13 17:32:33,597 - Epoch: [31][ 850/ 1207] Overall Loss 0.363342 Objective Loss 0.363342 LR 0.001000 Time 0.020293 -2023-02-13 17:32:33,792 - Epoch: [31][ 860/ 1207] Overall Loss 0.363356 Objective Loss 0.363356 LR 0.001000 Time 0.020283 -2023-02-13 17:32:33,987 - Epoch: [31][ 870/ 1207] Overall Loss 0.363503 Objective Loss 0.363503 LR 0.001000 Time 0.020274 -2023-02-13 17:32:34,181 - Epoch: [31][ 880/ 1207] Overall Loss 0.363368 Objective Loss 0.363368 LR 0.001000 Time 0.020264 -2023-02-13 17:32:34,378 - Epoch: [31][ 890/ 1207] Overall Loss 0.362749 Objective Loss 0.362749 LR 0.001000 Time 0.020257 -2023-02-13 17:32:34,571 - Epoch: [31][ 900/ 1207] Overall Loss 0.362615 Objective Loss 0.362615 LR 0.001000 Time 0.020247 -2023-02-13 17:32:34,768 - Epoch: [31][ 910/ 1207] Overall Loss 0.362483 Objective Loss 0.362483 LR 0.001000 Time 0.020240 -2023-02-13 17:32:34,962 - Epoch: [31][ 920/ 1207] Overall Loss 0.362325 Objective Loss 0.362325 LR 0.001000 Time 0.020230 -2023-02-13 17:32:35,159 - Epoch: [31][ 930/ 1207] Overall Loss 0.361856 Objective Loss 0.361856 LR 0.001000 Time 0.020224 -2023-02-13 17:32:35,353 - Epoch: [31][ 940/ 1207] Overall Loss 0.361775 Objective Loss 0.361775 LR 0.001000 Time 0.020215 -2023-02-13 17:32:35,549 - Epoch: [31][ 950/ 1207] Overall Loss 0.361775 Objective Loss 0.361775 LR 0.001000 Time 0.020208 -2023-02-13 17:32:35,743 - Epoch: [31][ 960/ 1207] Overall Loss 0.361892 Objective Loss 0.361892 LR 0.001000 Time 0.020199 -2023-02-13 17:32:35,941 - Epoch: [31][ 970/ 1207] Overall Loss 0.361775 Objective Loss 0.361775 LR 0.001000 Time 0.020194 -2023-02-13 17:32:36,134 - Epoch: [31][ 980/ 1207] Overall Loss 0.361743 Objective Loss 0.361743 LR 0.001000 Time 0.020185 -2023-02-13 17:32:36,330 - Epoch: [31][ 990/ 1207] Overall Loss 0.361559 Objective Loss 0.361559 LR 0.001000 Time 0.020179 -2023-02-13 17:32:36,525 - Epoch: [31][ 1000/ 1207] Overall Loss 0.361445 Objective Loss 0.361445 LR 0.001000 Time 0.020171 -2023-02-13 17:32:36,722 - Epoch: [31][ 1010/ 1207] Overall Loss 0.361427 Objective Loss 0.361427 LR 0.001000 Time 0.020167 -2023-02-13 17:32:36,919 - Epoch: [31][ 1020/ 1207] Overall Loss 0.361311 Objective Loss 0.361311 LR 0.001000 Time 0.020161 -2023-02-13 17:32:37,111 - Epoch: [31][ 1030/ 1207] Overall Loss 0.360989 Objective Loss 0.360989 LR 0.001000 Time 0.020152 -2023-02-13 17:32:37,309 - Epoch: [31][ 1040/ 1207] Overall Loss 0.360989 Objective Loss 0.360989 LR 0.001000 Time 0.020149 -2023-02-13 17:32:37,503 - Epoch: [31][ 1050/ 1207] Overall Loss 0.360877 Objective Loss 0.360877 LR 0.001000 Time 0.020141 -2023-02-13 17:32:37,700 - Epoch: [31][ 1060/ 1207] Overall Loss 0.360622 Objective Loss 0.360622 LR 0.001000 Time 0.020136 -2023-02-13 17:32:37,894 - Epoch: [31][ 1070/ 1207] Overall Loss 0.360766 Objective Loss 0.360766 LR 0.001000 Time 0.020129 -2023-02-13 17:32:38,090 - Epoch: [31][ 1080/ 1207] Overall Loss 0.360743 Objective Loss 0.360743 LR 0.001000 Time 0.020124 -2023-02-13 17:32:38,284 - Epoch: [31][ 1090/ 1207] Overall Loss 0.360495 Objective Loss 0.360495 LR 0.001000 Time 0.020117 -2023-02-13 17:32:38,480 - Epoch: [31][ 1100/ 1207] Overall Loss 0.360723 Objective Loss 0.360723 LR 0.001000 Time 0.020112 -2023-02-13 17:32:38,673 - Epoch: [31][ 1110/ 1207] Overall Loss 0.360703 Objective Loss 0.360703 LR 0.001000 Time 0.020104 -2023-02-13 17:32:38,870 - Epoch: [31][ 1120/ 1207] Overall Loss 0.360781 Objective Loss 0.360781 LR 0.001000 Time 0.020100 -2023-02-13 17:32:39,064 - Epoch: [31][ 1130/ 1207] Overall Loss 0.360660 Objective Loss 0.360660 LR 0.001000 Time 0.020093 -2023-02-13 17:32:39,260 - Epoch: [31][ 1140/ 1207] Overall Loss 0.360712 Objective Loss 0.360712 LR 0.001000 Time 0.020089 -2023-02-13 17:32:39,454 - Epoch: [31][ 1150/ 1207] Overall Loss 0.360679 Objective Loss 0.360679 LR 0.001000 Time 0.020082 -2023-02-13 17:32:39,650 - Epoch: [31][ 1160/ 1207] Overall Loss 0.360299 Objective Loss 0.360299 LR 0.001000 Time 0.020078 -2023-02-13 17:32:39,844 - Epoch: [31][ 1170/ 1207] Overall Loss 0.360559 Objective Loss 0.360559 LR 0.001000 Time 0.020072 -2023-02-13 17:32:40,040 - Epoch: [31][ 1180/ 1207] Overall Loss 0.360668 Objective Loss 0.360668 LR 0.001000 Time 0.020068 -2023-02-13 17:32:40,234 - Epoch: [31][ 1190/ 1207] Overall Loss 0.360564 Objective Loss 0.360564 LR 0.001000 Time 0.020062 -2023-02-13 17:32:40,479 - Epoch: [31][ 1200/ 1207] Overall Loss 0.360907 Objective Loss 0.360907 LR 0.001000 Time 0.020099 -2023-02-13 17:32:40,594 - Epoch: [31][ 1207/ 1207] Overall Loss 0.360822 Objective Loss 0.360822 Top1 78.048780 Top5 97.865854 LR 0.001000 Time 0.020077 -2023-02-13 17:32:40,666 - --- validate (epoch=31)----------- -2023-02-13 17:32:40,666 - 34311 samples (256 per mini-batch) -2023-02-13 17:32:41,055 - Epoch: [31][ 10/ 135] Loss 0.400737 Top1 81.835938 Top5 97.304688 -2023-02-13 17:32:41,181 - Epoch: [31][ 20/ 135] Loss 0.391626 Top1 81.035156 Top5 97.167969 -2023-02-13 17:32:41,303 - Epoch: [31][ 30/ 135] Loss 0.388103 Top1 80.950521 Top5 97.057292 -2023-02-13 17:32:41,429 - Epoch: [31][ 40/ 135] Loss 0.385094 Top1 81.357422 Top5 97.050781 -2023-02-13 17:32:41,554 - Epoch: [31][ 50/ 135] Loss 0.379934 Top1 81.515625 Top5 97.070312 -2023-02-13 17:32:41,682 - Epoch: [31][ 60/ 135] Loss 0.373850 Top1 81.562500 Top5 97.070312 -2023-02-13 17:32:41,812 - Epoch: [31][ 70/ 135] Loss 0.370344 Top1 81.456473 Top5 97.170759 -2023-02-13 17:32:41,937 - Epoch: [31][ 80/ 135] Loss 0.370104 Top1 81.352539 Top5 97.143555 -2023-02-13 17:32:42,060 - Epoch: [31][ 90/ 135] Loss 0.376685 Top1 81.219618 Top5 97.096354 -2023-02-13 17:32:42,184 - Epoch: [31][ 100/ 135] Loss 0.374653 Top1 81.253906 Top5 97.062500 -2023-02-13 17:32:42,313 - Epoch: [31][ 110/ 135] Loss 0.376201 Top1 81.210938 Top5 97.038352 -2023-02-13 17:32:42,440 - Epoch: [31][ 120/ 135] Loss 0.376878 Top1 81.266276 Top5 97.073568 -2023-02-13 17:32:42,570 - Epoch: [31][ 130/ 135] Loss 0.377443 Top1 81.144832 Top5 97.025240 -2023-02-13 17:32:42,613 - Epoch: [31][ 135/ 135] Loss 0.375078 Top1 81.140159 Top5 97.038851 -2023-02-13 17:32:42,686 - ==> Top1: 81.140 Top5: 97.039 Loss: 0.375 - -2023-02-13 17:32:42,687 - ==> Confusion: -[[ 817 6 3 0 8 4 0 2 4 90 0 7 2 5 3 1 4 4 0 0 7] - [ 4 893 0 1 4 65 1 26 7 2 7 5 0 1 2 1 4 1 5 1 3] - [ 12 8 899 20 4 4 25 19 1 5 6 3 1 6 2 11 6 5 4 4 13] - [ 6 2 7 880 0 9 2 1 3 3 17 4 19 3 17 1 4 12 17 2 7] - [ 27 14 1 0 950 17 1 1 2 11 2 5 1 6 4 6 9 0 0 4 5] - [ 4 9 0 3 2 967 3 18 0 5 3 14 4 16 2 1 5 2 1 7 4] - [ 5 3 15 3 0 7 1022 6 0 1 6 2 3 1 0 4 1 3 2 13 2] - [ 1 6 7 3 2 35 5 906 0 2 7 11 2 1 0 0 0 2 16 10 8] - [ 12 3 0 1 0 1 0 1 906 46 5 2 1 13 8 1 0 6 2 0 1] - [ 71 3 1 0 0 2 0 1 36 865 1 0 1 21 1 2 0 3 0 0 4] - [ 1 4 2 7 0 4 5 2 26 2 964 2 1 10 3 1 2 1 9 2 3] - [ 2 2 0 0 2 17 1 0 0 3 0 905 40 7 0 4 2 8 2 9 1] - [ 3 1 0 2 1 4 0 1 1 0 0 36 866 0 0 4 3 31 0 3 3] - [ 8 2 0 0 2 20 1 1 25 14 5 9 3 916 3 4 0 3 0 2 6] - [ 20 1 0 20 2 6 0 2 38 8 3 1 7 6 945 2 2 7 9 0 13] - [ 10 1 2 1 4 1 5 0 0 0 0 11 12 3 0 952 8 22 0 8 6] - [ 4 7 0 0 9 5 0 0 2 2 1 9 1 3 0 13 980 8 2 4 11] - [ 8 4 0 4 0 3 0 0 1 0 0 15 11 1 1 12 0 986 0 1 4] - [ 5 9 4 11 0 2 1 50 9 3 10 1 11 0 10 1 0 2 955 0 2] - [ 2 1 0 0 1 11 8 14 0 0 0 32 2 5 0 3 1 4 0 1057 7] - [ 245 226 161 134 84 314 69 221 170 141 257 194 450 362 135 139 253 172 134 364 9209]] - -2023-02-13 17:32:42,688 - ==> Best [Top1: 81.140 Top5: 97.039 Sparsity:0.00 Params: 148928 on epoch: 31] -2023-02-13 17:32:42,688 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:32:42,695 - - -2023-02-13 17:32:42,695 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:32:43,683 - Epoch: [32][ 10/ 1207] Overall Loss 0.323417 Objective Loss 0.323417 LR 0.001000 Time 0.098807 -2023-02-13 17:32:43,885 - Epoch: [32][ 20/ 1207] Overall Loss 0.337307 Objective Loss 0.337307 LR 0.001000 Time 0.059449 -2023-02-13 17:32:44,077 - Epoch: [32][ 30/ 1207] Overall Loss 0.343086 Objective Loss 0.343086 LR 0.001000 Time 0.046014 -2023-02-13 17:32:44,272 - Epoch: [32][ 40/ 1207] Overall Loss 0.344095 Objective Loss 0.344095 LR 0.001000 Time 0.039374 -2023-02-13 17:32:44,462 - Epoch: [32][ 50/ 1207] Overall Loss 0.346759 Objective Loss 0.346759 LR 0.001000 Time 0.035299 -2023-02-13 17:32:44,656 - Epoch: [32][ 60/ 1207] Overall Loss 0.350055 Objective Loss 0.350055 LR 0.001000 Time 0.032650 -2023-02-13 17:32:44,848 - Epoch: [32][ 70/ 1207] Overall Loss 0.349976 Objective Loss 0.349976 LR 0.001000 Time 0.030716 -2023-02-13 17:32:45,042 - Epoch: [32][ 80/ 1207] Overall Loss 0.346994 Objective Loss 0.346994 LR 0.001000 Time 0.029301 -2023-02-13 17:32:45,233 - Epoch: [32][ 90/ 1207] Overall Loss 0.348957 Objective Loss 0.348957 LR 0.001000 Time 0.028164 -2023-02-13 17:32:45,427 - Epoch: [32][ 100/ 1207] Overall Loss 0.350078 Objective Loss 0.350078 LR 0.001000 Time 0.027284 -2023-02-13 17:32:45,618 - Epoch: [32][ 110/ 1207] Overall Loss 0.349723 Objective Loss 0.349723 LR 0.001000 Time 0.026538 -2023-02-13 17:32:45,814 - Epoch: [32][ 120/ 1207] Overall Loss 0.350384 Objective Loss 0.350384 LR 0.001000 Time 0.025952 -2023-02-13 17:32:46,005 - Epoch: [32][ 130/ 1207] Overall Loss 0.349340 Objective Loss 0.349340 LR 0.001000 Time 0.025424 -2023-02-13 17:32:46,199 - Epoch: [32][ 140/ 1207] Overall Loss 0.349765 Objective Loss 0.349765 LR 0.001000 Time 0.024993 -2023-02-13 17:32:46,392 - Epoch: [32][ 150/ 1207] Overall Loss 0.348924 Objective Loss 0.348924 LR 0.001000 Time 0.024606 -2023-02-13 17:32:46,586 - Epoch: [32][ 160/ 1207] Overall Loss 0.346574 Objective Loss 0.346574 LR 0.001000 Time 0.024281 -2023-02-13 17:32:46,778 - Epoch: [32][ 170/ 1207] Overall Loss 0.346726 Objective Loss 0.346726 LR 0.001000 Time 0.023981 -2023-02-13 17:32:46,973 - Epoch: [32][ 180/ 1207] Overall Loss 0.346897 Objective Loss 0.346897 LR 0.001000 Time 0.023728 -2023-02-13 17:32:47,164 - Epoch: [32][ 190/ 1207] Overall Loss 0.347485 Objective Loss 0.347485 LR 0.001000 Time 0.023486 -2023-02-13 17:32:47,359 - Epoch: [32][ 200/ 1207] Overall Loss 0.350155 Objective Loss 0.350155 LR 0.001000 Time 0.023283 -2023-02-13 17:32:47,550 - Epoch: [32][ 210/ 1207] Overall Loss 0.350587 Objective Loss 0.350587 LR 0.001000 Time 0.023083 -2023-02-13 17:32:47,745 - Epoch: [32][ 220/ 1207] Overall Loss 0.351653 Objective Loss 0.351653 LR 0.001000 Time 0.022918 -2023-02-13 17:32:47,936 - Epoch: [32][ 230/ 1207] Overall Loss 0.352344 Objective Loss 0.352344 LR 0.001000 Time 0.022750 -2023-02-13 17:32:48,131 - Epoch: [32][ 240/ 1207] Overall Loss 0.354027 Objective Loss 0.354027 LR 0.001000 Time 0.022613 -2023-02-13 17:32:48,323 - Epoch: [32][ 250/ 1207] Overall Loss 0.353596 Objective Loss 0.353596 LR 0.001000 Time 0.022473 -2023-02-13 17:32:48,517 - Epoch: [32][ 260/ 1207] Overall Loss 0.352742 Objective Loss 0.352742 LR 0.001000 Time 0.022356 -2023-02-13 17:32:48,709 - Epoch: [32][ 270/ 1207] Overall Loss 0.354384 Objective Loss 0.354384 LR 0.001000 Time 0.022235 -2023-02-13 17:32:48,904 - Epoch: [32][ 280/ 1207] Overall Loss 0.355336 Objective Loss 0.355336 LR 0.001000 Time 0.022137 -2023-02-13 17:32:49,095 - Epoch: [32][ 290/ 1207] Overall Loss 0.354535 Objective Loss 0.354535 LR 0.001000 Time 0.022032 -2023-02-13 17:32:49,290 - Epoch: [32][ 300/ 1207] Overall Loss 0.355061 Objective Loss 0.355061 LR 0.001000 Time 0.021946 -2023-02-13 17:32:49,481 - Epoch: [32][ 310/ 1207] Overall Loss 0.354555 Objective Loss 0.354555 LR 0.001000 Time 0.021853 -2023-02-13 17:32:49,676 - Epoch: [32][ 320/ 1207] Overall Loss 0.354645 Objective Loss 0.354645 LR 0.001000 Time 0.021777 -2023-02-13 17:32:49,867 - Epoch: [32][ 330/ 1207] Overall Loss 0.353628 Objective Loss 0.353628 LR 0.001000 Time 0.021696 -2023-02-13 17:32:50,062 - Epoch: [32][ 340/ 1207] Overall Loss 0.353798 Objective Loss 0.353798 LR 0.001000 Time 0.021629 -2023-02-13 17:32:50,253 - Epoch: [32][ 350/ 1207] Overall Loss 0.355165 Objective Loss 0.355165 LR 0.001000 Time 0.021556 -2023-02-13 17:32:50,447 - Epoch: [32][ 360/ 1207] Overall Loss 0.355067 Objective Loss 0.355067 LR 0.001000 Time 0.021496 -2023-02-13 17:32:50,639 - Epoch: [32][ 370/ 1207] Overall Loss 0.355223 Objective Loss 0.355223 LR 0.001000 Time 0.021431 -2023-02-13 17:32:50,834 - Epoch: [32][ 380/ 1207] Overall Loss 0.354598 Objective Loss 0.354598 LR 0.001000 Time 0.021381 -2023-02-13 17:32:51,026 - Epoch: [32][ 390/ 1207] Overall Loss 0.354609 Objective Loss 0.354609 LR 0.001000 Time 0.021323 -2023-02-13 17:32:51,221 - Epoch: [32][ 400/ 1207] Overall Loss 0.355000 Objective Loss 0.355000 LR 0.001000 Time 0.021277 -2023-02-13 17:32:51,412 - Epoch: [32][ 410/ 1207] Overall Loss 0.356040 Objective Loss 0.356040 LR 0.001000 Time 0.021224 -2023-02-13 17:32:51,607 - Epoch: [32][ 420/ 1207] Overall Loss 0.356670 Objective Loss 0.356670 LR 0.001000 Time 0.021181 -2023-02-13 17:32:51,800 - Epoch: [32][ 430/ 1207] Overall Loss 0.357482 Objective Loss 0.357482 LR 0.001000 Time 0.021136 -2023-02-13 17:32:51,995 - Epoch: [32][ 440/ 1207] Overall Loss 0.357347 Objective Loss 0.357347 LR 0.001000 Time 0.021099 -2023-02-13 17:32:52,187 - Epoch: [32][ 450/ 1207] Overall Loss 0.357635 Objective Loss 0.357635 LR 0.001000 Time 0.021055 -2023-02-13 17:32:52,382 - Epoch: [32][ 460/ 1207] Overall Loss 0.357298 Objective Loss 0.357298 LR 0.001000 Time 0.021021 -2023-02-13 17:32:52,574 - Epoch: [32][ 470/ 1207] Overall Loss 0.356768 Objective Loss 0.356768 LR 0.001000 Time 0.020981 -2023-02-13 17:32:52,769 - Epoch: [32][ 480/ 1207] Overall Loss 0.356983 Objective Loss 0.356983 LR 0.001000 Time 0.020949 -2023-02-13 17:32:52,959 - Epoch: [32][ 490/ 1207] Overall Loss 0.356628 Objective Loss 0.356628 LR 0.001000 Time 0.020910 -2023-02-13 17:32:53,154 - Epoch: [32][ 500/ 1207] Overall Loss 0.355937 Objective Loss 0.355937 LR 0.001000 Time 0.020879 -2023-02-13 17:32:53,345 - Epoch: [32][ 510/ 1207] Overall Loss 0.356456 Objective Loss 0.356456 LR 0.001000 Time 0.020844 -2023-02-13 17:32:53,539 - Epoch: [32][ 520/ 1207] Overall Loss 0.356312 Objective Loss 0.356312 LR 0.001000 Time 0.020817 -2023-02-13 17:32:53,730 - Epoch: [32][ 530/ 1207] Overall Loss 0.356681 Objective Loss 0.356681 LR 0.001000 Time 0.020784 -2023-02-13 17:32:53,926 - Epoch: [32][ 540/ 1207] Overall Loss 0.356839 Objective Loss 0.356839 LR 0.001000 Time 0.020760 -2023-02-13 17:32:54,117 - Epoch: [32][ 550/ 1207] Overall Loss 0.356748 Objective Loss 0.356748 LR 0.001000 Time 0.020729 -2023-02-13 17:32:54,311 - Epoch: [32][ 560/ 1207] Overall Loss 0.356213 Objective Loss 0.356213 LR 0.001000 Time 0.020705 -2023-02-13 17:32:54,502 - Epoch: [32][ 570/ 1207] Overall Loss 0.355894 Objective Loss 0.355894 LR 0.001000 Time 0.020677 -2023-02-13 17:32:54,697 - Epoch: [32][ 580/ 1207] Overall Loss 0.355693 Objective Loss 0.355693 LR 0.001000 Time 0.020656 -2023-02-13 17:32:54,889 - Epoch: [32][ 590/ 1207] Overall Loss 0.355603 Objective Loss 0.355603 LR 0.001000 Time 0.020631 -2023-02-13 17:32:55,084 - Epoch: [32][ 600/ 1207] Overall Loss 0.355303 Objective Loss 0.355303 LR 0.001000 Time 0.020611 -2023-02-13 17:32:55,275 - Epoch: [32][ 610/ 1207] Overall Loss 0.355427 Objective Loss 0.355427 LR 0.001000 Time 0.020585 -2023-02-13 17:32:55,470 - Epoch: [32][ 620/ 1207] Overall Loss 0.355568 Objective Loss 0.355568 LR 0.001000 Time 0.020567 -2023-02-13 17:32:55,661 - Epoch: [32][ 630/ 1207] Overall Loss 0.355654 Objective Loss 0.355654 LR 0.001000 Time 0.020542 -2023-02-13 17:32:55,858 - Epoch: [32][ 640/ 1207] Overall Loss 0.355721 Objective Loss 0.355721 LR 0.001000 Time 0.020529 -2023-02-13 17:32:56,049 - Epoch: [32][ 650/ 1207] Overall Loss 0.355887 Objective Loss 0.355887 LR 0.001000 Time 0.020507 -2023-02-13 17:32:56,244 - Epoch: [32][ 660/ 1207] Overall Loss 0.355808 Objective Loss 0.355808 LR 0.001000 Time 0.020491 -2023-02-13 17:32:56,435 - Epoch: [32][ 670/ 1207] Overall Loss 0.355884 Objective Loss 0.355884 LR 0.001000 Time 0.020469 -2023-02-13 17:32:56,630 - Epoch: [32][ 680/ 1207] Overall Loss 0.356037 Objective Loss 0.356037 LR 0.001000 Time 0.020454 -2023-02-13 17:32:56,822 - Epoch: [32][ 690/ 1207] Overall Loss 0.355807 Objective Loss 0.355807 LR 0.001000 Time 0.020436 -2023-02-13 17:32:57,016 - Epoch: [32][ 700/ 1207] Overall Loss 0.355909 Objective Loss 0.355909 LR 0.001000 Time 0.020421 -2023-02-13 17:32:57,207 - Epoch: [32][ 710/ 1207] Overall Loss 0.355891 Objective Loss 0.355891 LR 0.001000 Time 0.020402 -2023-02-13 17:32:57,401 - Epoch: [32][ 720/ 1207] Overall Loss 0.355995 Objective Loss 0.355995 LR 0.001000 Time 0.020388 -2023-02-13 17:32:57,594 - Epoch: [32][ 730/ 1207] Overall Loss 0.356267 Objective Loss 0.356267 LR 0.001000 Time 0.020371 -2023-02-13 17:32:57,789 - Epoch: [32][ 740/ 1207] Overall Loss 0.356798 Objective Loss 0.356798 LR 0.001000 Time 0.020359 -2023-02-13 17:32:57,979 - Epoch: [32][ 750/ 1207] Overall Loss 0.357004 Objective Loss 0.357004 LR 0.001000 Time 0.020341 -2023-02-13 17:32:58,173 - Epoch: [32][ 760/ 1207] Overall Loss 0.357606 Objective Loss 0.357606 LR 0.001000 Time 0.020329 -2023-02-13 17:32:58,365 - Epoch: [32][ 770/ 1207] Overall Loss 0.357406 Objective Loss 0.357406 LR 0.001000 Time 0.020313 -2023-02-13 17:32:58,564 - Epoch: [32][ 780/ 1207] Overall Loss 0.357217 Objective Loss 0.357217 LR 0.001000 Time 0.020307 -2023-02-13 17:32:58,758 - Epoch: [32][ 790/ 1207] Overall Loss 0.357145 Objective Loss 0.357145 LR 0.001000 Time 0.020295 -2023-02-13 17:32:58,953 - Epoch: [32][ 800/ 1207] Overall Loss 0.357341 Objective Loss 0.357341 LR 0.001000 Time 0.020285 -2023-02-13 17:32:59,144 - Epoch: [32][ 810/ 1207] Overall Loss 0.357165 Objective Loss 0.357165 LR 0.001000 Time 0.020269 -2023-02-13 17:32:59,338 - Epoch: [32][ 820/ 1207] Overall Loss 0.357176 Objective Loss 0.357176 LR 0.001000 Time 0.020258 -2023-02-13 17:32:59,528 - Epoch: [32][ 830/ 1207] Overall Loss 0.356883 Objective Loss 0.356883 LR 0.001000 Time 0.020243 -2023-02-13 17:32:59,723 - Epoch: [32][ 840/ 1207] Overall Loss 0.357085 Objective Loss 0.357085 LR 0.001000 Time 0.020233 -2023-02-13 17:32:59,914 - Epoch: [32][ 850/ 1207] Overall Loss 0.357152 Objective Loss 0.357152 LR 0.001000 Time 0.020220 -2023-02-13 17:33:00,108 - Epoch: [32][ 860/ 1207] Overall Loss 0.356912 Objective Loss 0.356912 LR 0.001000 Time 0.020210 -2023-02-13 17:33:00,299 - Epoch: [32][ 870/ 1207] Overall Loss 0.356910 Objective Loss 0.356910 LR 0.001000 Time 0.020197 -2023-02-13 17:33:00,494 - Epoch: [32][ 880/ 1207] Overall Loss 0.357442 Objective Loss 0.357442 LR 0.001000 Time 0.020188 -2023-02-13 17:33:00,685 - Epoch: [32][ 890/ 1207] Overall Loss 0.357565 Objective Loss 0.357565 LR 0.001000 Time 0.020176 -2023-02-13 17:33:00,882 - Epoch: [32][ 900/ 1207] Overall Loss 0.357757 Objective Loss 0.357757 LR 0.001000 Time 0.020170 -2023-02-13 17:33:01,072 - Epoch: [32][ 910/ 1207] Overall Loss 0.357912 Objective Loss 0.357912 LR 0.001000 Time 0.020157 -2023-02-13 17:33:01,267 - Epoch: [32][ 920/ 1207] Overall Loss 0.358080 Objective Loss 0.358080 LR 0.001000 Time 0.020149 -2023-02-13 17:33:01,459 - Epoch: [32][ 930/ 1207] Overall Loss 0.358010 Objective Loss 0.358010 LR 0.001000 Time 0.020138 -2023-02-13 17:33:01,654 - Epoch: [32][ 940/ 1207] Overall Loss 0.358130 Objective Loss 0.358130 LR 0.001000 Time 0.020131 -2023-02-13 17:33:01,845 - Epoch: [32][ 950/ 1207] Overall Loss 0.358169 Objective Loss 0.358169 LR 0.001000 Time 0.020121 -2023-02-13 17:33:02,041 - Epoch: [32][ 960/ 1207] Overall Loss 0.358403 Objective Loss 0.358403 LR 0.001000 Time 0.020114 -2023-02-13 17:33:02,232 - Epoch: [32][ 970/ 1207] Overall Loss 0.358508 Objective Loss 0.358508 LR 0.001000 Time 0.020103 -2023-02-13 17:33:02,427 - Epoch: [32][ 980/ 1207] Overall Loss 0.358528 Objective Loss 0.358528 LR 0.001000 Time 0.020097 -2023-02-13 17:33:02,618 - Epoch: [32][ 990/ 1207] Overall Loss 0.358621 Objective Loss 0.358621 LR 0.001000 Time 0.020087 -2023-02-13 17:33:02,813 - Epoch: [32][ 1000/ 1207] Overall Loss 0.358638 Objective Loss 0.358638 LR 0.001000 Time 0.020081 -2023-02-13 17:33:03,005 - Epoch: [32][ 1010/ 1207] Overall Loss 0.358541 Objective Loss 0.358541 LR 0.001000 Time 0.020071 -2023-02-13 17:33:03,199 - Epoch: [32][ 1020/ 1207] Overall Loss 0.358473 Objective Loss 0.358473 LR 0.001000 Time 0.020064 -2023-02-13 17:33:03,391 - Epoch: [32][ 1030/ 1207] Overall Loss 0.358757 Objective Loss 0.358757 LR 0.001000 Time 0.020056 -2023-02-13 17:33:03,586 - Epoch: [32][ 1040/ 1207] Overall Loss 0.358771 Objective Loss 0.358771 LR 0.001000 Time 0.020049 -2023-02-13 17:33:03,777 - Epoch: [32][ 1050/ 1207] Overall Loss 0.358422 Objective Loss 0.358422 LR 0.001000 Time 0.020040 -2023-02-13 17:33:03,972 - Epoch: [32][ 1060/ 1207] Overall Loss 0.358330 Objective Loss 0.358330 LR 0.001000 Time 0.020035 -2023-02-13 17:33:04,163 - Epoch: [32][ 1070/ 1207] Overall Loss 0.358217 Objective Loss 0.358217 LR 0.001000 Time 0.020026 -2023-02-13 17:33:04,359 - Epoch: [32][ 1080/ 1207] Overall Loss 0.358915 Objective Loss 0.358915 LR 0.001000 Time 0.020021 -2023-02-13 17:33:04,550 - Epoch: [32][ 1090/ 1207] Overall Loss 0.359043 Objective Loss 0.359043 LR 0.001000 Time 0.020013 -2023-02-13 17:33:04,745 - Epoch: [32][ 1100/ 1207] Overall Loss 0.358938 Objective Loss 0.358938 LR 0.001000 Time 0.020008 -2023-02-13 17:33:04,937 - Epoch: [32][ 1110/ 1207] Overall Loss 0.359060 Objective Loss 0.359060 LR 0.001000 Time 0.020000 -2023-02-13 17:33:05,131 - Epoch: [32][ 1120/ 1207] Overall Loss 0.358984 Objective Loss 0.358984 LR 0.001000 Time 0.019995 -2023-02-13 17:33:05,323 - Epoch: [32][ 1130/ 1207] Overall Loss 0.358994 Objective Loss 0.358994 LR 0.001000 Time 0.019987 -2023-02-13 17:33:05,517 - Epoch: [32][ 1140/ 1207] Overall Loss 0.359203 Objective Loss 0.359203 LR 0.001000 Time 0.019982 -2023-02-13 17:33:05,709 - Epoch: [32][ 1150/ 1207] Overall Loss 0.359214 Objective Loss 0.359214 LR 0.001000 Time 0.019974 -2023-02-13 17:33:05,908 - Epoch: [32][ 1160/ 1207] Overall Loss 0.359367 Objective Loss 0.359367 LR 0.001000 Time 0.019974 -2023-02-13 17:33:06,103 - Epoch: [32][ 1170/ 1207] Overall Loss 0.359405 Objective Loss 0.359405 LR 0.001000 Time 0.019969 -2023-02-13 17:33:06,299 - Epoch: [32][ 1180/ 1207] Overall Loss 0.359638 Objective Loss 0.359638 LR 0.001000 Time 0.019966 -2023-02-13 17:33:06,494 - Epoch: [32][ 1190/ 1207] Overall Loss 0.360045 Objective Loss 0.360045 LR 0.001000 Time 0.019961 -2023-02-13 17:33:06,740 - Epoch: [32][ 1200/ 1207] Overall Loss 0.360052 Objective Loss 0.360052 LR 0.001000 Time 0.020000 -2023-02-13 17:33:06,858 - Epoch: [32][ 1207/ 1207] Overall Loss 0.359876 Objective Loss 0.359876 Top1 80.487805 Top5 95.731707 LR 0.001000 Time 0.019981 -2023-02-13 17:33:06,929 - --- validate (epoch=32)----------- -2023-02-13 17:33:06,930 - 34311 samples (256 per mini-batch) -2023-02-13 17:33:07,331 - Epoch: [32][ 10/ 135] Loss 0.344080 Top1 80.742188 Top5 97.148438 -2023-02-13 17:33:07,459 - Epoch: [32][ 20/ 135] Loss 0.359418 Top1 81.445312 Top5 97.304688 -2023-02-13 17:33:07,584 - Epoch: [32][ 30/ 135] Loss 0.365575 Top1 81.289062 Top5 97.356771 -2023-02-13 17:33:07,711 - Epoch: [32][ 40/ 135] Loss 0.362033 Top1 81.562500 Top5 97.412109 -2023-02-13 17:33:07,841 - Epoch: [32][ 50/ 135] Loss 0.360209 Top1 81.367188 Top5 97.281250 -2023-02-13 17:33:07,968 - Epoch: [32][ 60/ 135] Loss 0.366425 Top1 81.152344 Top5 97.226562 -2023-02-13 17:33:08,099 - Epoch: [32][ 70/ 135] Loss 0.368486 Top1 80.915179 Top5 97.193080 -2023-02-13 17:33:08,225 - Epoch: [32][ 80/ 135] Loss 0.362924 Top1 81.005859 Top5 97.221680 -2023-02-13 17:33:08,356 - Epoch: [32][ 90/ 135] Loss 0.368427 Top1 80.933160 Top5 97.213542 -2023-02-13 17:33:08,483 - Epoch: [32][ 100/ 135] Loss 0.368749 Top1 80.976562 Top5 97.179688 -2023-02-13 17:33:08,611 - Epoch: [32][ 110/ 135] Loss 0.372629 Top1 81.001420 Top5 97.155540 -2023-02-13 17:33:08,741 - Epoch: [32][ 120/ 135] Loss 0.373590 Top1 81.093750 Top5 97.115885 -2023-02-13 17:33:08,875 - Epoch: [32][ 130/ 135] Loss 0.376179 Top1 81.057692 Top5 97.130409 -2023-02-13 17:33:08,924 - Epoch: [32][ 135/ 135] Loss 0.374048 Top1 81.070211 Top5 97.143773 -2023-02-13 17:33:08,993 - ==> Top1: 81.070 Top5: 97.144 Loss: 0.374 - -2023-02-13 17:33:08,993 - ==> Confusion: -[[ 866 4 7 2 9 2 0 0 6 37 0 7 0 6 4 2 0 5 5 0 5] - [ 3 933 1 3 14 27 3 11 6 2 3 4 1 1 0 3 9 0 6 1 2] - [ 10 3 938 8 5 0 8 23 0 0 3 2 3 4 4 13 7 3 8 3 13] - [ 9 1 12 883 2 5 0 3 3 2 18 0 13 1 20 2 4 8 20 0 10] - [ 24 7 0 0 976 7 0 0 0 5 0 4 0 4 9 7 10 1 1 8 3] - [ 3 29 2 2 8 942 3 19 3 6 2 14 4 10 1 5 6 1 1 7 2] - [ 4 3 28 4 1 5 1003 9 1 0 6 4 3 2 0 7 3 0 2 11 3] - [ 4 14 11 0 1 42 7 884 2 1 1 5 5 1 1 0 0 1 29 10 5] - [ 22 2 0 1 3 0 0 1 902 36 6 1 1 8 15 3 1 1 5 0 1] - [ 106 3 3 1 4 0 0 1 41 821 1 0 0 21 1 1 0 2 0 0 6] - [ 3 5 3 7 1 0 3 5 24 1 964 2 1 10 1 0 1 2 13 0 5] - [ 6 1 2 0 2 11 0 6 3 3 0 900 37 3 1 5 3 6 4 11 1] - [ 3 1 1 3 1 4 0 2 1 0 0 51 856 1 1 3 3 15 5 4 4] - [ 9 1 2 0 5 14 0 4 16 20 6 9 8 897 6 3 7 5 3 1 8] - [ 22 3 3 21 6 2 0 2 29 4 4 0 6 0 954 3 3 3 15 1 11] - [ 7 0 4 1 10 2 1 0 0 0 0 7 12 3 0 966 10 11 1 5 6] - [ 3 5 0 0 12 1 0 0 4 0 1 3 2 5 1 18 982 1 5 6 12] - [ 12 1 0 5 0 4 1 2 2 0 0 19 37 3 0 18 0 942 1 1 3] - [ 4 7 6 7 0 1 1 25 5 2 6 3 6 0 10 1 2 0 998 1 1] - [ 3 3 1 1 0 5 9 16 0 1 1 25 3 5 0 5 7 3 2 1049 9] - [ 249 305 226 136 227 244 67 192 150 111 215 163 426 332 159 132 300 80 243 317 9160]] - -2023-02-13 17:33:08,995 - ==> Best [Top1: 81.140 Top5: 97.039 Sparsity:0.00 Params: 148928 on epoch: 31] -2023-02-13 17:33:08,995 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:33:09,000 - - -2023-02-13 17:33:09,001 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:33:09,901 - Epoch: [33][ 10/ 1207] Overall Loss 0.331834 Objective Loss 0.331834 LR 0.001000 Time 0.090022 -2023-02-13 17:33:10,118 - Epoch: [33][ 20/ 1207] Overall Loss 0.327452 Objective Loss 0.327452 LR 0.001000 Time 0.055839 -2023-02-13 17:33:10,321 - Epoch: [33][ 30/ 1207] Overall Loss 0.340028 Objective Loss 0.340028 LR 0.001000 Time 0.043963 -2023-02-13 17:33:10,529 - Epoch: [33][ 40/ 1207] Overall Loss 0.336749 Objective Loss 0.336749 LR 0.001000 Time 0.038159 -2023-02-13 17:33:10,729 - Epoch: [33][ 50/ 1207] Overall Loss 0.341194 Objective Loss 0.341194 LR 0.001000 Time 0.034533 -2023-02-13 17:33:10,940 - Epoch: [33][ 60/ 1207] Overall Loss 0.338059 Objective Loss 0.338059 LR 0.001000 Time 0.032273 -2023-02-13 17:33:11,141 - Epoch: [33][ 70/ 1207] Overall Loss 0.340829 Objective Loss 0.340829 LR 0.001000 Time 0.030527 -2023-02-13 17:33:11,348 - Epoch: [33][ 80/ 1207] Overall Loss 0.337219 Objective Loss 0.337219 LR 0.001000 Time 0.029305 -2023-02-13 17:33:11,550 - Epoch: [33][ 90/ 1207] Overall Loss 0.339335 Objective Loss 0.339335 LR 0.001000 Time 0.028281 -2023-02-13 17:33:11,758 - Epoch: [33][ 100/ 1207] Overall Loss 0.340395 Objective Loss 0.340395 LR 0.001000 Time 0.027528 -2023-02-13 17:33:11,961 - Epoch: [33][ 110/ 1207] Overall Loss 0.339759 Objective Loss 0.339759 LR 0.001000 Time 0.026870 -2023-02-13 17:33:12,169 - Epoch: [33][ 120/ 1207] Overall Loss 0.338515 Objective Loss 0.338515 LR 0.001000 Time 0.026359 -2023-02-13 17:33:12,371 - Epoch: [33][ 130/ 1207] Overall Loss 0.338265 Objective Loss 0.338265 LR 0.001000 Time 0.025884 -2023-02-13 17:33:12,578 - Epoch: [33][ 140/ 1207] Overall Loss 0.338997 Objective Loss 0.338997 LR 0.001000 Time 0.025514 -2023-02-13 17:33:12,780 - Epoch: [33][ 150/ 1207] Overall Loss 0.341247 Objective Loss 0.341247 LR 0.001000 Time 0.025152 -2023-02-13 17:33:12,987 - Epoch: [33][ 160/ 1207] Overall Loss 0.341711 Objective Loss 0.341711 LR 0.001000 Time 0.024872 -2023-02-13 17:33:13,189 - Epoch: [33][ 170/ 1207] Overall Loss 0.341630 Objective Loss 0.341630 LR 0.001000 Time 0.024594 -2023-02-13 17:33:13,397 - Epoch: [33][ 180/ 1207] Overall Loss 0.341002 Objective Loss 0.341002 LR 0.001000 Time 0.024384 -2023-02-13 17:33:13,591 - Epoch: [33][ 190/ 1207] Overall Loss 0.341093 Objective Loss 0.341093 LR 0.001000 Time 0.024120 -2023-02-13 17:33:13,788 - Epoch: [33][ 200/ 1207] Overall Loss 0.342100 Objective Loss 0.342100 LR 0.001000 Time 0.023894 -2023-02-13 17:33:13,983 - Epoch: [33][ 210/ 1207] Overall Loss 0.343934 Objective Loss 0.343934 LR 0.001000 Time 0.023686 -2023-02-13 17:33:14,179 - Epoch: [33][ 220/ 1207] Overall Loss 0.344386 Objective Loss 0.344386 LR 0.001000 Time 0.023497 -2023-02-13 17:33:14,374 - Epoch: [33][ 230/ 1207] Overall Loss 0.344950 Objective Loss 0.344950 LR 0.001000 Time 0.023322 -2023-02-13 17:33:14,571 - Epoch: [33][ 240/ 1207] Overall Loss 0.344322 Objective Loss 0.344322 LR 0.001000 Time 0.023168 -2023-02-13 17:33:14,766 - Epoch: [33][ 250/ 1207] Overall Loss 0.343878 Objective Loss 0.343878 LR 0.001000 Time 0.023020 -2023-02-13 17:33:14,962 - Epoch: [33][ 260/ 1207] Overall Loss 0.344924 Objective Loss 0.344924 LR 0.001000 Time 0.022889 -2023-02-13 17:33:15,157 - Epoch: [33][ 270/ 1207] Overall Loss 0.345346 Objective Loss 0.345346 LR 0.001000 Time 0.022761 -2023-02-13 17:33:15,354 - Epoch: [33][ 280/ 1207] Overall Loss 0.345907 Objective Loss 0.345907 LR 0.001000 Time 0.022649 -2023-02-13 17:33:15,547 - Epoch: [33][ 290/ 1207] Overall Loss 0.344907 Objective Loss 0.344907 LR 0.001000 Time 0.022535 -2023-02-13 17:33:15,743 - Epoch: [33][ 300/ 1207] Overall Loss 0.345470 Objective Loss 0.345470 LR 0.001000 Time 0.022435 -2023-02-13 17:33:15,938 - Epoch: [33][ 310/ 1207] Overall Loss 0.345608 Objective Loss 0.345608 LR 0.001000 Time 0.022340 -2023-02-13 17:33:16,134 - Epoch: [33][ 320/ 1207] Overall Loss 0.345868 Objective Loss 0.345868 LR 0.001000 Time 0.022252 -2023-02-13 17:33:16,328 - Epoch: [33][ 330/ 1207] Overall Loss 0.346224 Objective Loss 0.346224 LR 0.001000 Time 0.022165 -2023-02-13 17:33:16,525 - Epoch: [33][ 340/ 1207] Overall Loss 0.345471 Objective Loss 0.345471 LR 0.001000 Time 0.022090 -2023-02-13 17:33:16,719 - Epoch: [33][ 350/ 1207] Overall Loss 0.344778 Objective Loss 0.344778 LR 0.001000 Time 0.022012 -2023-02-13 17:33:16,916 - Epoch: [33][ 360/ 1207] Overall Loss 0.345278 Objective Loss 0.345278 LR 0.001000 Time 0.021949 -2023-02-13 17:33:17,111 - Epoch: [33][ 370/ 1207] Overall Loss 0.346152 Objective Loss 0.346152 LR 0.001000 Time 0.021880 -2023-02-13 17:33:17,307 - Epoch: [33][ 380/ 1207] Overall Loss 0.345460 Objective Loss 0.345460 LR 0.001000 Time 0.021821 -2023-02-13 17:33:17,502 - Epoch: [33][ 390/ 1207] Overall Loss 0.345107 Objective Loss 0.345107 LR 0.001000 Time 0.021760 -2023-02-13 17:33:17,699 - Epoch: [33][ 400/ 1207] Overall Loss 0.345613 Objective Loss 0.345613 LR 0.001000 Time 0.021706 -2023-02-13 17:33:17,894 - Epoch: [33][ 410/ 1207] Overall Loss 0.346351 Objective Loss 0.346351 LR 0.001000 Time 0.021653 -2023-02-13 17:33:18,091 - Epoch: [33][ 420/ 1207] Overall Loss 0.348200 Objective Loss 0.348200 LR 0.001000 Time 0.021605 -2023-02-13 17:33:18,285 - Epoch: [33][ 430/ 1207] Overall Loss 0.348431 Objective Loss 0.348431 LR 0.001000 Time 0.021553 -2023-02-13 17:33:18,482 - Epoch: [33][ 440/ 1207] Overall Loss 0.348877 Objective Loss 0.348877 LR 0.001000 Time 0.021509 -2023-02-13 17:33:18,676 - Epoch: [33][ 450/ 1207] Overall Loss 0.348998 Objective Loss 0.348998 LR 0.001000 Time 0.021462 -2023-02-13 17:33:18,872 - Epoch: [33][ 460/ 1207] Overall Loss 0.348891 Objective Loss 0.348891 LR 0.001000 Time 0.021421 -2023-02-13 17:33:19,067 - Epoch: [33][ 470/ 1207] Overall Loss 0.348331 Objective Loss 0.348331 LR 0.001000 Time 0.021379 -2023-02-13 17:33:19,264 - Epoch: [33][ 480/ 1207] Overall Loss 0.349227 Objective Loss 0.349227 LR 0.001000 Time 0.021343 -2023-02-13 17:33:19,459 - Epoch: [33][ 490/ 1207] Overall Loss 0.348973 Objective Loss 0.348973 LR 0.001000 Time 0.021304 -2023-02-13 17:33:19,655 - Epoch: [33][ 500/ 1207] Overall Loss 0.348907 Objective Loss 0.348907 LR 0.001000 Time 0.021271 -2023-02-13 17:33:19,850 - Epoch: [33][ 510/ 1207] Overall Loss 0.349550 Objective Loss 0.349550 LR 0.001000 Time 0.021235 -2023-02-13 17:33:20,047 - Epoch: [33][ 520/ 1207] Overall Loss 0.349957 Objective Loss 0.349957 LR 0.001000 Time 0.021204 -2023-02-13 17:33:20,242 - Epoch: [33][ 530/ 1207] Overall Loss 0.350499 Objective Loss 0.350499 LR 0.001000 Time 0.021171 -2023-02-13 17:33:20,439 - Epoch: [33][ 540/ 1207] Overall Loss 0.350508 Objective Loss 0.350508 LR 0.001000 Time 0.021144 -2023-02-13 17:33:20,634 - Epoch: [33][ 550/ 1207] Overall Loss 0.350781 Objective Loss 0.350781 LR 0.001000 Time 0.021113 -2023-02-13 17:33:20,832 - Epoch: [33][ 560/ 1207] Overall Loss 0.350382 Objective Loss 0.350382 LR 0.001000 Time 0.021088 -2023-02-13 17:33:21,026 - Epoch: [33][ 570/ 1207] Overall Loss 0.350139 Objective Loss 0.350139 LR 0.001000 Time 0.021059 -2023-02-13 17:33:21,222 - Epoch: [33][ 580/ 1207] Overall Loss 0.350330 Objective Loss 0.350330 LR 0.001000 Time 0.021033 -2023-02-13 17:33:21,418 - Epoch: [33][ 590/ 1207] Overall Loss 0.350765 Objective Loss 0.350765 LR 0.001000 Time 0.021007 -2023-02-13 17:33:21,614 - Epoch: [33][ 600/ 1207] Overall Loss 0.351010 Objective Loss 0.351010 LR 0.001000 Time 0.020984 -2023-02-13 17:33:21,811 - Epoch: [33][ 610/ 1207] Overall Loss 0.350923 Objective Loss 0.350923 LR 0.001000 Time 0.020963 -2023-02-13 17:33:22,008 - Epoch: [33][ 620/ 1207] Overall Loss 0.351236 Objective Loss 0.351236 LR 0.001000 Time 0.020942 -2023-02-13 17:33:22,203 - Epoch: [33][ 630/ 1207] Overall Loss 0.351381 Objective Loss 0.351381 LR 0.001000 Time 0.020919 -2023-02-13 17:33:22,401 - Epoch: [33][ 640/ 1207] Overall Loss 0.351382 Objective Loss 0.351382 LR 0.001000 Time 0.020899 -2023-02-13 17:33:22,597 - Epoch: [33][ 650/ 1207] Overall Loss 0.351246 Objective Loss 0.351246 LR 0.001000 Time 0.020879 -2023-02-13 17:33:22,794 - Epoch: [33][ 660/ 1207] Overall Loss 0.351193 Objective Loss 0.351193 LR 0.001000 Time 0.020860 -2023-02-13 17:33:22,990 - Epoch: [33][ 670/ 1207] Overall Loss 0.351322 Objective Loss 0.351322 LR 0.001000 Time 0.020841 -2023-02-13 17:33:23,186 - Epoch: [33][ 680/ 1207] Overall Loss 0.351335 Objective Loss 0.351335 LR 0.001000 Time 0.020823 -2023-02-13 17:33:23,382 - Epoch: [33][ 690/ 1207] Overall Loss 0.350759 Objective Loss 0.350759 LR 0.001000 Time 0.020804 -2023-02-13 17:33:23,579 - Epoch: [33][ 700/ 1207] Overall Loss 0.350687 Objective Loss 0.350687 LR 0.001000 Time 0.020788 -2023-02-13 17:33:23,774 - Epoch: [33][ 710/ 1207] Overall Loss 0.351415 Objective Loss 0.351415 LR 0.001000 Time 0.020769 -2023-02-13 17:33:23,971 - Epoch: [33][ 720/ 1207] Overall Loss 0.352012 Objective Loss 0.352012 LR 0.001000 Time 0.020754 -2023-02-13 17:33:24,165 - Epoch: [33][ 730/ 1207] Overall Loss 0.351965 Objective Loss 0.351965 LR 0.001000 Time 0.020735 -2023-02-13 17:33:24,362 - Epoch: [33][ 740/ 1207] Overall Loss 0.352436 Objective Loss 0.352436 LR 0.001000 Time 0.020720 -2023-02-13 17:33:24,556 - Epoch: [33][ 750/ 1207] Overall Loss 0.352529 Objective Loss 0.352529 LR 0.001000 Time 0.020702 -2023-02-13 17:33:24,752 - Epoch: [33][ 760/ 1207] Overall Loss 0.352578 Objective Loss 0.352578 LR 0.001000 Time 0.020687 -2023-02-13 17:33:24,947 - Epoch: [33][ 770/ 1207] Overall Loss 0.352593 Objective Loss 0.352593 LR 0.001000 Time 0.020671 -2023-02-13 17:33:25,143 - Epoch: [33][ 780/ 1207] Overall Loss 0.352669 Objective Loss 0.352669 LR 0.001000 Time 0.020658 -2023-02-13 17:33:25,339 - Epoch: [33][ 790/ 1207] Overall Loss 0.352694 Objective Loss 0.352694 LR 0.001000 Time 0.020643 -2023-02-13 17:33:25,536 - Epoch: [33][ 800/ 1207] Overall Loss 0.352484 Objective Loss 0.352484 LR 0.001000 Time 0.020631 -2023-02-13 17:33:25,731 - Epoch: [33][ 810/ 1207] Overall Loss 0.352527 Objective Loss 0.352527 LR 0.001000 Time 0.020617 -2023-02-13 17:33:25,929 - Epoch: [33][ 820/ 1207] Overall Loss 0.352669 Objective Loss 0.352669 LR 0.001000 Time 0.020607 -2023-02-13 17:33:26,125 - Epoch: [33][ 830/ 1207] Overall Loss 0.352688 Objective Loss 0.352688 LR 0.001000 Time 0.020593 -2023-02-13 17:33:26,322 - Epoch: [33][ 840/ 1207] Overall Loss 0.352857 Objective Loss 0.352857 LR 0.001000 Time 0.020583 -2023-02-13 17:33:26,517 - Epoch: [33][ 850/ 1207] Overall Loss 0.353074 Objective Loss 0.353074 LR 0.001000 Time 0.020570 -2023-02-13 17:33:26,715 - Epoch: [33][ 860/ 1207] Overall Loss 0.353191 Objective Loss 0.353191 LR 0.001000 Time 0.020560 -2023-02-13 17:33:26,911 - Epoch: [33][ 870/ 1207] Overall Loss 0.353274 Objective Loss 0.353274 LR 0.001000 Time 0.020549 -2023-02-13 17:33:27,108 - Epoch: [33][ 880/ 1207] Overall Loss 0.353360 Objective Loss 0.353360 LR 0.001000 Time 0.020539 -2023-02-13 17:33:27,304 - Epoch: [33][ 890/ 1207] Overall Loss 0.353557 Objective Loss 0.353557 LR 0.001000 Time 0.020527 -2023-02-13 17:33:27,500 - Epoch: [33][ 900/ 1207] Overall Loss 0.353890 Objective Loss 0.353890 LR 0.001000 Time 0.020517 -2023-02-13 17:33:27,693 - Epoch: [33][ 910/ 1207] Overall Loss 0.354143 Objective Loss 0.354143 LR 0.001000 Time 0.020503 -2023-02-13 17:33:27,887 - Epoch: [33][ 920/ 1207] Overall Loss 0.353987 Objective Loss 0.353987 LR 0.001000 Time 0.020491 -2023-02-13 17:33:28,081 - Epoch: [33][ 930/ 1207] Overall Loss 0.354056 Objective Loss 0.354056 LR 0.001000 Time 0.020478 -2023-02-13 17:33:28,275 - Epoch: [33][ 940/ 1207] Overall Loss 0.354137 Objective Loss 0.354137 LR 0.001000 Time 0.020467 -2023-02-13 17:33:28,468 - Epoch: [33][ 950/ 1207] Overall Loss 0.353899 Objective Loss 0.353899 LR 0.001000 Time 0.020454 -2023-02-13 17:33:28,663 - Epoch: [33][ 960/ 1207] Overall Loss 0.353951 Objective Loss 0.353951 LR 0.001000 Time 0.020444 -2023-02-13 17:33:28,856 - Epoch: [33][ 970/ 1207] Overall Loss 0.354150 Objective Loss 0.354150 LR 0.001000 Time 0.020431 -2023-02-13 17:33:29,050 - Epoch: [33][ 980/ 1207] Overall Loss 0.354316 Objective Loss 0.354316 LR 0.001000 Time 0.020421 -2023-02-13 17:33:29,245 - Epoch: [33][ 990/ 1207] Overall Loss 0.354648 Objective Loss 0.354648 LR 0.001000 Time 0.020410 -2023-02-13 17:33:29,439 - Epoch: [33][ 1000/ 1207] Overall Loss 0.354723 Objective Loss 0.354723 LR 0.001000 Time 0.020400 -2023-02-13 17:33:29,633 - Epoch: [33][ 1010/ 1207] Overall Loss 0.354519 Objective Loss 0.354519 LR 0.001000 Time 0.020390 -2023-02-13 17:33:29,827 - Epoch: [33][ 1020/ 1207] Overall Loss 0.354236 Objective Loss 0.354236 LR 0.001000 Time 0.020380 -2023-02-13 17:33:30,020 - Epoch: [33][ 1030/ 1207] Overall Loss 0.354052 Objective Loss 0.354052 LR 0.001000 Time 0.020369 -2023-02-13 17:33:30,214 - Epoch: [33][ 1040/ 1207] Overall Loss 0.353798 Objective Loss 0.353798 LR 0.001000 Time 0.020359 -2023-02-13 17:33:30,407 - Epoch: [33][ 1050/ 1207] Overall Loss 0.353783 Objective Loss 0.353783 LR 0.001000 Time 0.020349 -2023-02-13 17:33:30,604 - Epoch: [33][ 1060/ 1207] Overall Loss 0.354009 Objective Loss 0.354009 LR 0.001000 Time 0.020342 -2023-02-13 17:33:30,798 - Epoch: [33][ 1070/ 1207] Overall Loss 0.353974 Objective Loss 0.353974 LR 0.001000 Time 0.020334 -2023-02-13 17:33:30,996 - Epoch: [33][ 1080/ 1207] Overall Loss 0.354259 Objective Loss 0.354259 LR 0.001000 Time 0.020328 -2023-02-13 17:33:31,189 - Epoch: [33][ 1090/ 1207] Overall Loss 0.354355 Objective Loss 0.354355 LR 0.001000 Time 0.020319 -2023-02-13 17:33:31,386 - Epoch: [33][ 1100/ 1207] Overall Loss 0.354392 Objective Loss 0.354392 LR 0.001000 Time 0.020313 -2023-02-13 17:33:31,581 - Epoch: [33][ 1110/ 1207] Overall Loss 0.354515 Objective Loss 0.354515 LR 0.001000 Time 0.020305 -2023-02-13 17:33:31,778 - Epoch: [33][ 1120/ 1207] Overall Loss 0.354526 Objective Loss 0.354526 LR 0.001000 Time 0.020299 -2023-02-13 17:33:31,973 - Epoch: [33][ 1130/ 1207] Overall Loss 0.354570 Objective Loss 0.354570 LR 0.001000 Time 0.020292 -2023-02-13 17:33:32,170 - Epoch: [33][ 1140/ 1207] Overall Loss 0.354315 Objective Loss 0.354315 LR 0.001000 Time 0.020286 -2023-02-13 17:33:32,364 - Epoch: [33][ 1150/ 1207] Overall Loss 0.354336 Objective Loss 0.354336 LR 0.001000 Time 0.020278 -2023-02-13 17:33:32,561 - Epoch: [33][ 1160/ 1207] Overall Loss 0.354449 Objective Loss 0.354449 LR 0.001000 Time 0.020272 -2023-02-13 17:33:32,756 - Epoch: [33][ 1170/ 1207] Overall Loss 0.354545 Objective Loss 0.354545 LR 0.001000 Time 0.020266 -2023-02-13 17:33:32,953 - Epoch: [33][ 1180/ 1207] Overall Loss 0.354528 Objective Loss 0.354528 LR 0.001000 Time 0.020261 -2023-02-13 17:33:33,147 - Epoch: [33][ 1190/ 1207] Overall Loss 0.354233 Objective Loss 0.354233 LR 0.001000 Time 0.020254 -2023-02-13 17:33:33,392 - Epoch: [33][ 1200/ 1207] Overall Loss 0.354419 Objective Loss 0.354419 LR 0.001000 Time 0.020289 -2023-02-13 17:33:33,508 - Epoch: [33][ 1207/ 1207] Overall Loss 0.354175 Objective Loss 0.354175 Top1 84.756098 Top5 97.865854 LR 0.001000 Time 0.020267 -2023-02-13 17:33:33,579 - --- validate (epoch=33)----------- -2023-02-13 17:33:33,579 - 34311 samples (256 per mini-batch) -2023-02-13 17:33:33,983 - Epoch: [33][ 10/ 135] Loss 0.390349 Top1 81.796875 Top5 96.992188 -2023-02-13 17:33:34,113 - Epoch: [33][ 20/ 135] Loss 0.375890 Top1 81.738281 Top5 97.285156 -2023-02-13 17:33:34,255 - Epoch: [33][ 30/ 135] Loss 0.386874 Top1 81.497396 Top5 96.744792 -2023-02-13 17:33:34,382 - Epoch: [33][ 40/ 135] Loss 0.378572 Top1 81.425781 Top5 96.787109 -2023-02-13 17:33:34,514 - Epoch: [33][ 50/ 135] Loss 0.383404 Top1 81.187500 Top5 96.765625 -2023-02-13 17:33:34,643 - Epoch: [33][ 60/ 135] Loss 0.377201 Top1 81.217448 Top5 96.829427 -2023-02-13 17:33:34,774 - Epoch: [33][ 70/ 135] Loss 0.374058 Top1 81.250000 Top5 96.863839 -2023-02-13 17:33:34,904 - Epoch: [33][ 80/ 135] Loss 0.370722 Top1 81.376953 Top5 96.918945 -2023-02-13 17:33:35,035 - Epoch: [33][ 90/ 135] Loss 0.369959 Top1 81.449653 Top5 96.905382 -2023-02-13 17:33:35,166 - Epoch: [33][ 100/ 135] Loss 0.367509 Top1 81.566406 Top5 96.988281 -2023-02-13 17:33:35,297 - Epoch: [33][ 110/ 135] Loss 0.366442 Top1 81.519886 Top5 96.992188 -2023-02-13 17:33:35,427 - Epoch: [33][ 120/ 135] Loss 0.366542 Top1 81.520182 Top5 96.975911 -2023-02-13 17:33:35,560 - Epoch: [33][ 130/ 135] Loss 0.367231 Top1 81.595553 Top5 96.995192 -2023-02-13 17:33:35,608 - Epoch: [33][ 135/ 135] Loss 0.370299 Top1 81.618140 Top5 97.012620 -2023-02-13 17:33:35,691 - ==> Top1: 81.618 Top5: 97.013 Loss: 0.370 - -2023-02-13 17:33:35,692 - ==> Confusion: -[[ 830 5 6 5 12 3 0 2 7 73 0 4 2 4 3 1 2 1 1 0 6] - [ 7 904 2 3 7 41 5 19 5 1 4 0 2 3 2 1 4 1 7 2 13] - [ 9 2 946 23 3 4 14 13 1 0 4 1 3 2 2 3 4 6 7 3 8] - [ 5 1 14 901 1 6 1 2 4 2 15 2 6 3 14 0 1 9 20 0 9] - [ 24 9 4 1 973 10 3 0 2 3 0 3 1 6 6 5 9 0 0 2 5] - [ 6 13 1 6 4 977 1 18 1 2 3 2 4 15 0 1 3 4 3 4 2] - [ 4 2 25 6 0 5 1020 8 0 1 8 2 2 1 0 1 0 1 1 6 6] - [ 2 11 16 3 3 39 6 883 2 1 9 5 2 3 0 1 0 3 20 11 4] - [ 17 2 0 1 1 5 0 0 876 45 10 1 1 22 15 1 0 4 6 0 2] - [ 69 2 5 1 7 1 0 0 43 849 0 0 0 23 3 1 1 1 0 0 6] - [ 4 3 4 11 2 0 1 2 26 2 956 0 1 13 1 0 0 1 15 1 8] - [ 5 2 1 0 2 25 1 6 4 2 1 878 25 14 2 7 2 11 2 14 1] - [ 2 1 2 9 1 6 0 3 1 0 1 42 842 2 2 5 1 29 2 1 7] - [ 9 2 2 0 7 16 0 2 15 18 8 4 1 926 1 2 3 4 1 0 3] - [ 13 4 1 30 5 2 0 1 28 7 5 1 3 2 962 1 0 7 8 0 12] - [ 8 1 6 2 8 2 2 2 0 2 0 10 10 3 1 945 9 20 0 3 12] - [ 4 5 1 2 12 6 1 0 4 0 0 3 0 3 1 12 979 5 4 3 16] - [ 8 0 0 5 3 5 1 2 2 0 2 14 15 3 0 9 0 975 0 1 6] - [ 5 3 11 17 0 2 0 31 6 0 5 1 5 0 13 0 1 3 976 2 5] - [ 3 1 0 0 2 9 8 12 1 0 3 18 0 3 0 7 2 4 2 1061 12] - [ 258 227 291 162 126 294 73 190 145 137 203 153 357 420 146 79 199 141 187 301 9345]] - -2023-02-13 17:33:35,693 - ==> Best [Top1: 81.618 Top5: 97.013 Sparsity:0.00 Params: 148928 on epoch: 33] -2023-02-13 17:33:35,693 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:33:35,700 - - -2023-02-13 17:33:35,700 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:33:36,589 - Epoch: [34][ 10/ 1207] Overall Loss 0.331455 Objective Loss 0.331455 LR 0.001000 Time 0.088846 -2023-02-13 17:33:36,802 - Epoch: [34][ 20/ 1207] Overall Loss 0.328972 Objective Loss 0.328972 LR 0.001000 Time 0.055041 -2023-02-13 17:33:37,000 - Epoch: [34][ 30/ 1207] Overall Loss 0.344468 Objective Loss 0.344468 LR 0.001000 Time 0.043290 -2023-02-13 17:33:37,203 - Epoch: [34][ 40/ 1207] Overall Loss 0.348850 Objective Loss 0.348850 LR 0.001000 Time 0.037520 -2023-02-13 17:33:37,400 - Epoch: [34][ 50/ 1207] Overall Loss 0.351282 Objective Loss 0.351282 LR 0.001000 Time 0.033954 -2023-02-13 17:33:37,603 - Epoch: [34][ 60/ 1207] Overall Loss 0.347695 Objective Loss 0.347695 LR 0.001000 Time 0.031675 -2023-02-13 17:33:37,800 - Epoch: [34][ 70/ 1207] Overall Loss 0.349052 Objective Loss 0.349052 LR 0.001000 Time 0.029953 -2023-02-13 17:33:38,004 - Epoch: [34][ 80/ 1207] Overall Loss 0.345354 Objective Loss 0.345354 LR 0.001000 Time 0.028750 -2023-02-13 17:33:38,201 - Epoch: [34][ 90/ 1207] Overall Loss 0.344405 Objective Loss 0.344405 LR 0.001000 Time 0.027744 -2023-02-13 17:33:38,404 - Epoch: [34][ 100/ 1207] Overall Loss 0.347193 Objective Loss 0.347193 LR 0.001000 Time 0.026992 -2023-02-13 17:33:38,601 - Epoch: [34][ 110/ 1207] Overall Loss 0.347303 Objective Loss 0.347303 LR 0.001000 Time 0.026325 -2023-02-13 17:33:38,803 - Epoch: [34][ 120/ 1207] Overall Loss 0.350220 Objective Loss 0.350220 LR 0.001000 Time 0.025815 -2023-02-13 17:33:39,001 - Epoch: [34][ 130/ 1207] Overall Loss 0.352436 Objective Loss 0.352436 LR 0.001000 Time 0.025347 -2023-02-13 17:33:39,203 - Epoch: [34][ 140/ 1207] Overall Loss 0.350925 Objective Loss 0.350925 LR 0.001000 Time 0.024982 -2023-02-13 17:33:39,401 - Epoch: [34][ 150/ 1207] Overall Loss 0.352113 Objective Loss 0.352113 LR 0.001000 Time 0.024632 -2023-02-13 17:33:39,604 - Epoch: [34][ 160/ 1207] Overall Loss 0.351788 Objective Loss 0.351788 LR 0.001000 Time 0.024358 -2023-02-13 17:33:39,802 - Epoch: [34][ 170/ 1207] Overall Loss 0.353172 Objective Loss 0.353172 LR 0.001000 Time 0.024085 -2023-02-13 17:33:40,004 - Epoch: [34][ 180/ 1207] Overall Loss 0.353030 Objective Loss 0.353030 LR 0.001000 Time 0.023869 -2023-02-13 17:33:40,202 - Epoch: [34][ 190/ 1207] Overall Loss 0.353298 Objective Loss 0.353298 LR 0.001000 Time 0.023652 -2023-02-13 17:33:40,405 - Epoch: [34][ 200/ 1207] Overall Loss 0.353168 Objective Loss 0.353168 LR 0.001000 Time 0.023483 -2023-02-13 17:33:40,603 - Epoch: [34][ 210/ 1207] Overall Loss 0.352776 Objective Loss 0.352776 LR 0.001000 Time 0.023305 -2023-02-13 17:33:40,808 - Epoch: [34][ 220/ 1207] Overall Loss 0.353450 Objective Loss 0.353450 LR 0.001000 Time 0.023177 -2023-02-13 17:33:41,007 - Epoch: [34][ 230/ 1207] Overall Loss 0.352066 Objective Loss 0.352066 LR 0.001000 Time 0.023032 -2023-02-13 17:33:41,210 - Epoch: [34][ 240/ 1207] Overall Loss 0.351946 Objective Loss 0.351946 LR 0.001000 Time 0.022917 -2023-02-13 17:33:41,407 - Epoch: [34][ 250/ 1207] Overall Loss 0.350859 Objective Loss 0.350859 LR 0.001000 Time 0.022788 -2023-02-13 17:33:41,611 - Epoch: [34][ 260/ 1207] Overall Loss 0.350354 Objective Loss 0.350354 LR 0.001000 Time 0.022693 -2023-02-13 17:33:41,810 - Epoch: [34][ 270/ 1207] Overall Loss 0.349150 Objective Loss 0.349150 LR 0.001000 Time 0.022589 -2023-02-13 17:33:42,014 - Epoch: [34][ 280/ 1207] Overall Loss 0.348327 Objective Loss 0.348327 LR 0.001000 Time 0.022509 -2023-02-13 17:33:42,212 - Epoch: [34][ 290/ 1207] Overall Loss 0.348020 Objective Loss 0.348020 LR 0.001000 Time 0.022414 -2023-02-13 17:33:42,415 - Epoch: [34][ 300/ 1207] Overall Loss 0.348074 Objective Loss 0.348074 LR 0.001000 Time 0.022342 -2023-02-13 17:33:42,613 - Epoch: [34][ 310/ 1207] Overall Loss 0.348594 Objective Loss 0.348594 LR 0.001000 Time 0.022259 -2023-02-13 17:33:42,816 - Epoch: [34][ 320/ 1207] Overall Loss 0.348742 Objective Loss 0.348742 LR 0.001000 Time 0.022197 -2023-02-13 17:33:43,014 - Epoch: [34][ 330/ 1207] Overall Loss 0.348958 Objective Loss 0.348958 LR 0.001000 Time 0.022123 -2023-02-13 17:33:43,217 - Epoch: [34][ 340/ 1207] Overall Loss 0.348423 Objective Loss 0.348423 LR 0.001000 Time 0.022070 -2023-02-13 17:33:43,416 - Epoch: [34][ 350/ 1207] Overall Loss 0.348553 Objective Loss 0.348553 LR 0.001000 Time 0.022005 -2023-02-13 17:33:43,619 - Epoch: [34][ 360/ 1207] Overall Loss 0.348379 Objective Loss 0.348379 LR 0.001000 Time 0.021958 -2023-02-13 17:33:43,818 - Epoch: [34][ 370/ 1207] Overall Loss 0.348050 Objective Loss 0.348050 LR 0.001000 Time 0.021899 -2023-02-13 17:33:44,023 - Epoch: [34][ 380/ 1207] Overall Loss 0.347849 Objective Loss 0.347849 LR 0.001000 Time 0.021863 -2023-02-13 17:33:44,223 - Epoch: [34][ 390/ 1207] Overall Loss 0.347459 Objective Loss 0.347459 LR 0.001000 Time 0.021813 -2023-02-13 17:33:44,426 - Epoch: [34][ 400/ 1207] Overall Loss 0.347813 Objective Loss 0.347813 LR 0.001000 Time 0.021775 -2023-02-13 17:33:44,626 - Epoch: [34][ 410/ 1207] Overall Loss 0.348024 Objective Loss 0.348024 LR 0.001000 Time 0.021730 -2023-02-13 17:33:44,829 - Epoch: [34][ 420/ 1207] Overall Loss 0.347225 Objective Loss 0.347225 LR 0.001000 Time 0.021696 -2023-02-13 17:33:45,029 - Epoch: [34][ 430/ 1207] Overall Loss 0.347156 Objective Loss 0.347156 LR 0.001000 Time 0.021656 -2023-02-13 17:33:45,234 - Epoch: [34][ 440/ 1207] Overall Loss 0.346339 Objective Loss 0.346339 LR 0.001000 Time 0.021627 -2023-02-13 17:33:45,433 - Epoch: [34][ 450/ 1207] Overall Loss 0.347356 Objective Loss 0.347356 LR 0.001000 Time 0.021588 -2023-02-13 17:33:45,637 - Epoch: [34][ 460/ 1207] Overall Loss 0.348099 Objective Loss 0.348099 LR 0.001000 Time 0.021562 -2023-02-13 17:33:45,837 - Epoch: [34][ 470/ 1207] Overall Loss 0.347974 Objective Loss 0.347974 LR 0.001000 Time 0.021528 -2023-02-13 17:33:46,042 - Epoch: [34][ 480/ 1207] Overall Loss 0.347773 Objective Loss 0.347773 LR 0.001000 Time 0.021507 -2023-02-13 17:33:46,242 - Epoch: [34][ 490/ 1207] Overall Loss 0.347966 Objective Loss 0.347966 LR 0.001000 Time 0.021475 -2023-02-13 17:33:46,446 - Epoch: [34][ 500/ 1207] Overall Loss 0.347767 Objective Loss 0.347767 LR 0.001000 Time 0.021453 -2023-02-13 17:33:46,645 - Epoch: [34][ 510/ 1207] Overall Loss 0.347589 Objective Loss 0.347589 LR 0.001000 Time 0.021422 -2023-02-13 17:33:46,850 - Epoch: [34][ 520/ 1207] Overall Loss 0.347731 Objective Loss 0.347731 LR 0.001000 Time 0.021402 -2023-02-13 17:33:47,050 - Epoch: [34][ 530/ 1207] Overall Loss 0.347633 Objective Loss 0.347633 LR 0.001000 Time 0.021376 -2023-02-13 17:33:47,254 - Epoch: [34][ 540/ 1207] Overall Loss 0.347756 Objective Loss 0.347756 LR 0.001000 Time 0.021356 -2023-02-13 17:33:47,453 - Epoch: [34][ 550/ 1207] Overall Loss 0.348064 Objective Loss 0.348064 LR 0.001000 Time 0.021330 -2023-02-13 17:33:47,658 - Epoch: [34][ 560/ 1207] Overall Loss 0.347723 Objective Loss 0.347723 LR 0.001000 Time 0.021314 -2023-02-13 17:33:47,857 - Epoch: [34][ 570/ 1207] Overall Loss 0.347990 Objective Loss 0.347990 LR 0.001000 Time 0.021289 -2023-02-13 17:33:48,062 - Epoch: [34][ 580/ 1207] Overall Loss 0.347975 Objective Loss 0.347975 LR 0.001000 Time 0.021274 -2023-02-13 17:33:48,261 - Epoch: [34][ 590/ 1207] Overall Loss 0.347771 Objective Loss 0.347771 LR 0.001000 Time 0.021250 -2023-02-13 17:33:48,465 - Epoch: [34][ 600/ 1207] Overall Loss 0.347484 Objective Loss 0.347484 LR 0.001000 Time 0.021236 -2023-02-13 17:33:48,664 - Epoch: [34][ 610/ 1207] Overall Loss 0.347734 Objective Loss 0.347734 LR 0.001000 Time 0.021213 -2023-02-13 17:33:48,868 - Epoch: [34][ 620/ 1207] Overall Loss 0.347615 Objective Loss 0.347615 LR 0.001000 Time 0.021199 -2023-02-13 17:33:49,069 - Epoch: [34][ 630/ 1207] Overall Loss 0.347468 Objective Loss 0.347468 LR 0.001000 Time 0.021181 -2023-02-13 17:33:49,273 - Epoch: [34][ 640/ 1207] Overall Loss 0.347652 Objective Loss 0.347652 LR 0.001000 Time 0.021168 -2023-02-13 17:33:49,472 - Epoch: [34][ 650/ 1207] Overall Loss 0.348339 Objective Loss 0.348339 LR 0.001000 Time 0.021148 -2023-02-13 17:33:49,677 - Epoch: [34][ 660/ 1207] Overall Loss 0.348793 Objective Loss 0.348793 LR 0.001000 Time 0.021138 -2023-02-13 17:33:49,868 - Epoch: [34][ 670/ 1207] Overall Loss 0.349274 Objective Loss 0.349274 LR 0.001000 Time 0.021106 -2023-02-13 17:33:50,058 - Epoch: [34][ 680/ 1207] Overall Loss 0.349743 Objective Loss 0.349743 LR 0.001000 Time 0.021075 -2023-02-13 17:33:50,247 - Epoch: [34][ 690/ 1207] Overall Loss 0.349658 Objective Loss 0.349658 LR 0.001000 Time 0.021043 -2023-02-13 17:33:50,435 - Epoch: [34][ 700/ 1207] Overall Loss 0.349672 Objective Loss 0.349672 LR 0.001000 Time 0.021011 -2023-02-13 17:33:50,624 - Epoch: [34][ 710/ 1207] Overall Loss 0.349542 Objective Loss 0.349542 LR 0.001000 Time 0.020980 -2023-02-13 17:33:50,814 - Epoch: [34][ 720/ 1207] Overall Loss 0.349869 Objective Loss 0.349869 LR 0.001000 Time 0.020952 -2023-02-13 17:33:51,003 - Epoch: [34][ 730/ 1207] Overall Loss 0.350170 Objective Loss 0.350170 LR 0.001000 Time 0.020924 -2023-02-13 17:33:51,191 - Epoch: [34][ 740/ 1207] Overall Loss 0.350112 Objective Loss 0.350112 LR 0.001000 Time 0.020895 -2023-02-13 17:33:51,380 - Epoch: [34][ 750/ 1207] Overall Loss 0.350084 Objective Loss 0.350084 LR 0.001000 Time 0.020867 -2023-02-13 17:33:51,568 - Epoch: [34][ 760/ 1207] Overall Loss 0.350083 Objective Loss 0.350083 LR 0.001000 Time 0.020840 -2023-02-13 17:33:51,757 - Epoch: [34][ 770/ 1207] Overall Loss 0.350404 Objective Loss 0.350404 LR 0.001000 Time 0.020814 -2023-02-13 17:33:51,946 - Epoch: [34][ 780/ 1207] Overall Loss 0.350084 Objective Loss 0.350084 LR 0.001000 Time 0.020789 -2023-02-13 17:33:52,135 - Epoch: [34][ 790/ 1207] Overall Loss 0.350451 Objective Loss 0.350451 LR 0.001000 Time 0.020764 -2023-02-13 17:33:52,323 - Epoch: [34][ 800/ 1207] Overall Loss 0.350368 Objective Loss 0.350368 LR 0.001000 Time 0.020740 -2023-02-13 17:33:52,512 - Epoch: [34][ 810/ 1207] Overall Loss 0.350297 Objective Loss 0.350297 LR 0.001000 Time 0.020716 -2023-02-13 17:33:52,700 - Epoch: [34][ 820/ 1207] Overall Loss 0.350330 Objective Loss 0.350330 LR 0.001000 Time 0.020692 -2023-02-13 17:33:52,888 - Epoch: [34][ 830/ 1207] Overall Loss 0.351220 Objective Loss 0.351220 LR 0.001000 Time 0.020670 -2023-02-13 17:33:53,077 - Epoch: [34][ 840/ 1207] Overall Loss 0.351503 Objective Loss 0.351503 LR 0.001000 Time 0.020648 -2023-02-13 17:33:53,269 - Epoch: [34][ 850/ 1207] Overall Loss 0.351371 Objective Loss 0.351371 LR 0.001000 Time 0.020630 -2023-02-13 17:33:53,458 - Epoch: [34][ 860/ 1207] Overall Loss 0.351353 Objective Loss 0.351353 LR 0.001000 Time 0.020610 -2023-02-13 17:33:53,647 - Epoch: [34][ 870/ 1207] Overall Loss 0.351247 Objective Loss 0.351247 LR 0.001000 Time 0.020590 -2023-02-13 17:33:53,836 - Epoch: [34][ 880/ 1207] Overall Loss 0.351088 Objective Loss 0.351088 LR 0.001000 Time 0.020570 -2023-02-13 17:33:54,026 - Epoch: [34][ 890/ 1207] Overall Loss 0.350827 Objective Loss 0.350827 LR 0.001000 Time 0.020552 -2023-02-13 17:33:54,216 - Epoch: [34][ 900/ 1207] Overall Loss 0.350844 Objective Loss 0.350844 LR 0.001000 Time 0.020534 -2023-02-13 17:33:54,405 - Epoch: [34][ 910/ 1207] Overall Loss 0.350891 Objective Loss 0.350891 LR 0.001000 Time 0.020516 -2023-02-13 17:33:54,593 - Epoch: [34][ 920/ 1207] Overall Loss 0.350925 Objective Loss 0.350925 LR 0.001000 Time 0.020497 -2023-02-13 17:33:54,782 - Epoch: [34][ 930/ 1207] Overall Loss 0.351284 Objective Loss 0.351284 LR 0.001000 Time 0.020479 -2023-02-13 17:33:54,970 - Epoch: [34][ 940/ 1207] Overall Loss 0.351100 Objective Loss 0.351100 LR 0.001000 Time 0.020462 -2023-02-13 17:33:55,160 - Epoch: [34][ 950/ 1207] Overall Loss 0.351364 Objective Loss 0.351364 LR 0.001000 Time 0.020445 -2023-02-13 17:33:55,350 - Epoch: [34][ 960/ 1207] Overall Loss 0.351459 Objective Loss 0.351459 LR 0.001000 Time 0.020429 -2023-02-13 17:33:55,539 - Epoch: [34][ 970/ 1207] Overall Loss 0.351265 Objective Loss 0.351265 LR 0.001000 Time 0.020413 -2023-02-13 17:33:55,727 - Epoch: [34][ 980/ 1207] Overall Loss 0.350961 Objective Loss 0.350961 LR 0.001000 Time 0.020397 -2023-02-13 17:33:55,918 - Epoch: [34][ 990/ 1207] Overall Loss 0.350977 Objective Loss 0.350977 LR 0.001000 Time 0.020383 -2023-02-13 17:33:56,107 - Epoch: [34][ 1000/ 1207] Overall Loss 0.351119 Objective Loss 0.351119 LR 0.001000 Time 0.020368 -2023-02-13 17:33:56,296 - Epoch: [34][ 1010/ 1207] Overall Loss 0.351441 Objective Loss 0.351441 LR 0.001000 Time 0.020354 -2023-02-13 17:33:56,486 - Epoch: [34][ 1020/ 1207] Overall Loss 0.351171 Objective Loss 0.351171 LR 0.001000 Time 0.020339 -2023-02-13 17:33:56,675 - Epoch: [34][ 1030/ 1207] Overall Loss 0.351144 Objective Loss 0.351144 LR 0.001000 Time 0.020325 -2023-02-13 17:33:56,865 - Epoch: [34][ 1040/ 1207] Overall Loss 0.351410 Objective Loss 0.351410 LR 0.001000 Time 0.020312 -2023-02-13 17:33:57,054 - Epoch: [34][ 1050/ 1207] Overall Loss 0.351612 Objective Loss 0.351612 LR 0.001000 Time 0.020299 -2023-02-13 17:33:57,243 - Epoch: [34][ 1060/ 1207] Overall Loss 0.351632 Objective Loss 0.351632 LR 0.001000 Time 0.020285 -2023-02-13 17:33:57,432 - Epoch: [34][ 1070/ 1207] Overall Loss 0.351482 Objective Loss 0.351482 LR 0.001000 Time 0.020272 -2023-02-13 17:33:57,621 - Epoch: [34][ 1080/ 1207] Overall Loss 0.351104 Objective Loss 0.351104 LR 0.001000 Time 0.020258 -2023-02-13 17:33:57,810 - Epoch: [34][ 1090/ 1207] Overall Loss 0.351167 Objective Loss 0.351167 LR 0.001000 Time 0.020246 -2023-02-13 17:33:57,999 - Epoch: [34][ 1100/ 1207] Overall Loss 0.351168 Objective Loss 0.351168 LR 0.001000 Time 0.020233 -2023-02-13 17:33:58,188 - Epoch: [34][ 1110/ 1207] Overall Loss 0.351381 Objective Loss 0.351381 LR 0.001000 Time 0.020221 -2023-02-13 17:33:58,377 - Epoch: [34][ 1120/ 1207] Overall Loss 0.351499 Objective Loss 0.351499 LR 0.001000 Time 0.020208 -2023-02-13 17:33:58,566 - Epoch: [34][ 1130/ 1207] Overall Loss 0.351511 Objective Loss 0.351511 LR 0.001000 Time 0.020196 -2023-02-13 17:33:58,754 - Epoch: [34][ 1140/ 1207] Overall Loss 0.351697 Objective Loss 0.351697 LR 0.001000 Time 0.020184 -2023-02-13 17:33:58,944 - Epoch: [34][ 1150/ 1207] Overall Loss 0.351889 Objective Loss 0.351889 LR 0.001000 Time 0.020173 -2023-02-13 17:33:59,134 - Epoch: [34][ 1160/ 1207] Overall Loss 0.351770 Objective Loss 0.351770 LR 0.001000 Time 0.020163 -2023-02-13 17:33:59,323 - Epoch: [34][ 1170/ 1207] Overall Loss 0.351903 Objective Loss 0.351903 LR 0.001000 Time 0.020152 -2023-02-13 17:33:59,513 - Epoch: [34][ 1180/ 1207] Overall Loss 0.351931 Objective Loss 0.351931 LR 0.001000 Time 0.020142 -2023-02-13 17:33:59,701 - Epoch: [34][ 1190/ 1207] Overall Loss 0.352124 Objective Loss 0.352124 LR 0.001000 Time 0.020131 -2023-02-13 17:33:59,947 - Epoch: [34][ 1200/ 1207] Overall Loss 0.352060 Objective Loss 0.352060 LR 0.001000 Time 0.020167 -2023-02-13 17:34:00,062 - Epoch: [34][ 1207/ 1207] Overall Loss 0.352252 Objective Loss 0.352252 Top1 78.658537 Top5 96.341463 LR 0.001000 Time 0.020146 -2023-02-13 17:34:00,133 - --- validate (epoch=34)----------- -2023-02-13 17:34:00,133 - 34311 samples (256 per mini-batch) -2023-02-13 17:34:00,537 - Epoch: [34][ 10/ 135] Loss 0.371258 Top1 81.250000 Top5 97.578125 -2023-02-13 17:34:00,661 - Epoch: [34][ 20/ 135] Loss 0.369508 Top1 80.546875 Top5 97.246094 -2023-02-13 17:34:00,792 - Epoch: [34][ 30/ 135] Loss 0.372758 Top1 80.533854 Top5 97.083333 -2023-02-13 17:34:00,927 - Epoch: [34][ 40/ 135] Loss 0.378401 Top1 80.234375 Top5 96.962891 -2023-02-13 17:34:01,056 - Epoch: [34][ 50/ 135] Loss 0.378551 Top1 80.390625 Top5 96.890625 -2023-02-13 17:34:01,180 - Epoch: [34][ 60/ 135] Loss 0.368849 Top1 80.540365 Top5 96.946615 -2023-02-13 17:34:01,310 - Epoch: [34][ 70/ 135] Loss 0.366533 Top1 80.368304 Top5 96.897321 -2023-02-13 17:34:01,447 - Epoch: [34][ 80/ 135] Loss 0.365704 Top1 80.434570 Top5 96.972656 -2023-02-13 17:34:01,592 - Epoch: [34][ 90/ 135] Loss 0.371476 Top1 80.316840 Top5 96.957465 -2023-02-13 17:34:01,729 - Epoch: [34][ 100/ 135] Loss 0.369915 Top1 80.332031 Top5 96.992188 -2023-02-13 17:34:01,873 - Epoch: [34][ 110/ 135] Loss 0.367567 Top1 80.525568 Top5 96.999290 -2023-02-13 17:34:02,013 - Epoch: [34][ 120/ 135] Loss 0.367425 Top1 80.481771 Top5 96.992188 -2023-02-13 17:34:02,152 - Epoch: [34][ 130/ 135] Loss 0.365265 Top1 80.597957 Top5 97.013221 -2023-02-13 17:34:02,197 - Epoch: [34][ 135/ 135] Loss 0.372399 Top1 80.600973 Top5 97.027192 -2023-02-13 17:34:02,270 - ==> Top1: 80.601 Top5: 97.027 Loss: 0.372 - -2023-02-13 17:34:02,271 - ==> Confusion: -[[ 874 4 2 1 13 1 0 2 3 45 0 4 1 5 2 1 1 3 1 0 4] - [ 3 926 2 4 13 34 5 13 7 0 2 4 2 1 1 3 1 0 6 3 3] - [ 10 4 955 10 6 1 17 15 1 0 3 2 3 4 2 6 4 1 6 2 6] - [ 7 3 17 894 1 3 1 1 1 2 17 2 12 0 23 2 2 4 16 1 7] - [ 20 11 1 0 991 8 2 0 2 3 1 5 1 5 6 3 1 1 0 2 3] - [ 6 21 3 1 9 953 2 8 2 3 3 13 4 22 2 3 1 2 3 8 1] - [ 5 4 22 3 0 6 1018 5 0 0 4 2 4 3 1 4 1 2 1 10 4] - [ 3 14 14 4 2 46 3 869 0 3 2 8 2 2 1 0 0 1 31 15 4] - [ 29 1 0 1 3 0 0 1 871 43 11 1 2 9 25 4 1 2 4 0 1] - [ 108 4 3 1 5 2 0 1 32 833 0 1 0 12 3 1 0 2 0 1 3] - [ 3 1 7 6 1 1 4 2 22 0 963 2 1 10 4 0 0 4 16 1 3] - [ 7 5 1 0 5 10 0 3 1 3 1 892 35 7 0 3 3 14 1 12 2] - [ 5 1 2 5 3 3 1 0 0 0 0 34 866 0 4 3 1 22 0 0 9] - [ 9 3 0 0 4 8 0 1 25 25 7 8 3 902 5 6 3 2 1 3 9] - [ 16 3 2 15 10 1 0 1 18 8 3 2 7 3 982 1 3 5 7 0 5] - [ 7 2 5 0 12 1 2 0 0 0 0 7 11 2 2 950 10 24 0 6 5] - [ 5 5 3 0 26 3 0 0 1 2 0 4 3 2 2 12 976 3 1 5 8] - [ 10 1 2 4 4 0 0 1 0 1 0 8 21 0 0 12 0 981 0 1 5] - [ 6 3 3 14 1 2 0 22 4 1 2 0 10 1 22 0 0 1 991 1 2] - [ 2 4 0 0 1 10 5 14 1 0 1 25 2 6 0 3 3 1 0 1063 7] - [ 309 306 243 166 233 294 66 159 111 130 194 132 437 363 255 117 293 142 237 342 8905]] - -2023-02-13 17:34:02,272 - ==> Best [Top1: 81.618 Top5: 97.013 Sparsity:0.00 Params: 148928 on epoch: 33] -2023-02-13 17:34:02,272 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:34:02,278 - - -2023-02-13 17:34:02,278 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:34:03,273 - Epoch: [35][ 10/ 1207] Overall Loss 0.336989 Objective Loss 0.336989 LR 0.001000 Time 0.099403 -2023-02-13 17:34:03,483 - Epoch: [35][ 20/ 1207] Overall Loss 0.340618 Objective Loss 0.340618 LR 0.001000 Time 0.060185 -2023-02-13 17:34:03,687 - Epoch: [35][ 30/ 1207] Overall Loss 0.330681 Objective Loss 0.330681 LR 0.001000 Time 0.046910 -2023-02-13 17:34:03,887 - Epoch: [35][ 40/ 1207] Overall Loss 0.337193 Objective Loss 0.337193 LR 0.001000 Time 0.040162 -2023-02-13 17:34:04,092 - Epoch: [35][ 50/ 1207] Overall Loss 0.339462 Objective Loss 0.339462 LR 0.001000 Time 0.036227 -2023-02-13 17:34:04,291 - Epoch: [35][ 60/ 1207] Overall Loss 0.341955 Objective Loss 0.341955 LR 0.001000 Time 0.033502 -2023-02-13 17:34:04,495 - Epoch: [35][ 70/ 1207] Overall Loss 0.346891 Objective Loss 0.346891 LR 0.001000 Time 0.031626 -2023-02-13 17:34:04,684 - Epoch: [35][ 80/ 1207] Overall Loss 0.347575 Objective Loss 0.347575 LR 0.001000 Time 0.030032 -2023-02-13 17:34:04,872 - Epoch: [35][ 90/ 1207] Overall Loss 0.347668 Objective Loss 0.347668 LR 0.001000 Time 0.028781 -2023-02-13 17:34:05,062 - Epoch: [35][ 100/ 1207] Overall Loss 0.347953 Objective Loss 0.347953 LR 0.001000 Time 0.027794 -2023-02-13 17:34:05,251 - Epoch: [35][ 110/ 1207] Overall Loss 0.347753 Objective Loss 0.347753 LR 0.001000 Time 0.026979 -2023-02-13 17:34:05,439 - Epoch: [35][ 120/ 1207] Overall Loss 0.345679 Objective Loss 0.345679 LR 0.001000 Time 0.026297 -2023-02-13 17:34:05,628 - Epoch: [35][ 130/ 1207] Overall Loss 0.345590 Objective Loss 0.345590 LR 0.001000 Time 0.025722 -2023-02-13 17:34:05,817 - Epoch: [35][ 140/ 1207] Overall Loss 0.343940 Objective Loss 0.343940 LR 0.001000 Time 0.025233 -2023-02-13 17:34:06,006 - Epoch: [35][ 150/ 1207] Overall Loss 0.341859 Objective Loss 0.341859 LR 0.001000 Time 0.024808 -2023-02-13 17:34:06,195 - Epoch: [35][ 160/ 1207] Overall Loss 0.342616 Objective Loss 0.342616 LR 0.001000 Time 0.024438 -2023-02-13 17:34:06,384 - Epoch: [35][ 170/ 1207] Overall Loss 0.344273 Objective Loss 0.344273 LR 0.001000 Time 0.024110 -2023-02-13 17:34:06,573 - Epoch: [35][ 180/ 1207] Overall Loss 0.344489 Objective Loss 0.344489 LR 0.001000 Time 0.023818 -2023-02-13 17:34:06,762 - Epoch: [35][ 190/ 1207] Overall Loss 0.344970 Objective Loss 0.344970 LR 0.001000 Time 0.023558 -2023-02-13 17:34:06,951 - Epoch: [35][ 200/ 1207] Overall Loss 0.343729 Objective Loss 0.343729 LR 0.001000 Time 0.023323 -2023-02-13 17:34:07,140 - Epoch: [35][ 210/ 1207] Overall Loss 0.345462 Objective Loss 0.345462 LR 0.001000 Time 0.023112 -2023-02-13 17:34:07,329 - Epoch: [35][ 220/ 1207] Overall Loss 0.345973 Objective Loss 0.345973 LR 0.001000 Time 0.022917 -2023-02-13 17:34:07,518 - Epoch: [35][ 230/ 1207] Overall Loss 0.348053 Objective Loss 0.348053 LR 0.001000 Time 0.022741 -2023-02-13 17:34:07,707 - Epoch: [35][ 240/ 1207] Overall Loss 0.347766 Objective Loss 0.347766 LR 0.001000 Time 0.022578 -2023-02-13 17:34:07,895 - Epoch: [35][ 250/ 1207] Overall Loss 0.347442 Objective Loss 0.347442 LR 0.001000 Time 0.022428 -2023-02-13 17:34:08,085 - Epoch: [35][ 260/ 1207] Overall Loss 0.346728 Objective Loss 0.346728 LR 0.001000 Time 0.022293 -2023-02-13 17:34:08,274 - Epoch: [35][ 270/ 1207] Overall Loss 0.346089 Objective Loss 0.346089 LR 0.001000 Time 0.022166 -2023-02-13 17:34:08,463 - Epoch: [35][ 280/ 1207] Overall Loss 0.345498 Objective Loss 0.345498 LR 0.001000 Time 0.022047 -2023-02-13 17:34:08,651 - Epoch: [35][ 290/ 1207] Overall Loss 0.344691 Objective Loss 0.344691 LR 0.001000 Time 0.021936 -2023-02-13 17:34:08,839 - Epoch: [35][ 300/ 1207] Overall Loss 0.345524 Objective Loss 0.345524 LR 0.001000 Time 0.021831 -2023-02-13 17:34:09,028 - Epoch: [35][ 310/ 1207] Overall Loss 0.346212 Objective Loss 0.346212 LR 0.001000 Time 0.021734 -2023-02-13 17:34:09,218 - Epoch: [35][ 320/ 1207] Overall Loss 0.346275 Objective Loss 0.346275 LR 0.001000 Time 0.021646 -2023-02-13 17:34:09,407 - Epoch: [35][ 330/ 1207] Overall Loss 0.346495 Objective Loss 0.346495 LR 0.001000 Time 0.021561 -2023-02-13 17:34:09,595 - Epoch: [35][ 340/ 1207] Overall Loss 0.345230 Objective Loss 0.345230 LR 0.001000 Time 0.021482 -2023-02-13 17:34:09,784 - Epoch: [35][ 350/ 1207] Overall Loss 0.344959 Objective Loss 0.344959 LR 0.001000 Time 0.021406 -2023-02-13 17:34:09,973 - Epoch: [35][ 360/ 1207] Overall Loss 0.344185 Objective Loss 0.344185 LR 0.001000 Time 0.021335 -2023-02-13 17:34:10,162 - Epoch: [35][ 370/ 1207] Overall Loss 0.343684 Objective Loss 0.343684 LR 0.001000 Time 0.021268 -2023-02-13 17:34:10,351 - Epoch: [35][ 380/ 1207] Overall Loss 0.343782 Objective Loss 0.343782 LR 0.001000 Time 0.021205 -2023-02-13 17:34:10,539 - Epoch: [35][ 390/ 1207] Overall Loss 0.344099 Objective Loss 0.344099 LR 0.001000 Time 0.021143 -2023-02-13 17:34:10,729 - Epoch: [35][ 400/ 1207] Overall Loss 0.344681 Objective Loss 0.344681 LR 0.001000 Time 0.021087 -2023-02-13 17:34:10,918 - Epoch: [35][ 410/ 1207] Overall Loss 0.343530 Objective Loss 0.343530 LR 0.001000 Time 0.021034 -2023-02-13 17:34:11,108 - Epoch: [35][ 420/ 1207] Overall Loss 0.343749 Objective Loss 0.343749 LR 0.001000 Time 0.020984 -2023-02-13 17:34:11,297 - Epoch: [35][ 430/ 1207] Overall Loss 0.343246 Objective Loss 0.343246 LR 0.001000 Time 0.020934 -2023-02-13 17:34:11,485 - Epoch: [35][ 440/ 1207] Overall Loss 0.343547 Objective Loss 0.343547 LR 0.001000 Time 0.020887 -2023-02-13 17:34:11,674 - Epoch: [35][ 450/ 1207] Overall Loss 0.343461 Objective Loss 0.343461 LR 0.001000 Time 0.020842 -2023-02-13 17:34:11,864 - Epoch: [35][ 460/ 1207] Overall Loss 0.343114 Objective Loss 0.343114 LR 0.001000 Time 0.020799 -2023-02-13 17:34:12,053 - Epoch: [35][ 470/ 1207] Overall Loss 0.343332 Objective Loss 0.343332 LR 0.001000 Time 0.020759 -2023-02-13 17:34:12,243 - Epoch: [35][ 480/ 1207] Overall Loss 0.343357 Objective Loss 0.343357 LR 0.001000 Time 0.020721 -2023-02-13 17:34:12,431 - Epoch: [35][ 490/ 1207] Overall Loss 0.342823 Objective Loss 0.342823 LR 0.001000 Time 0.020682 -2023-02-13 17:34:12,621 - Epoch: [35][ 500/ 1207] Overall Loss 0.343607 Objective Loss 0.343607 LR 0.001000 Time 0.020646 -2023-02-13 17:34:12,810 - Epoch: [35][ 510/ 1207] Overall Loss 0.343826 Objective Loss 0.343826 LR 0.001000 Time 0.020611 -2023-02-13 17:34:12,998 - Epoch: [35][ 520/ 1207] Overall Loss 0.343471 Objective Loss 0.343471 LR 0.001000 Time 0.020577 -2023-02-13 17:34:13,187 - Epoch: [35][ 530/ 1207] Overall Loss 0.343848 Objective Loss 0.343848 LR 0.001000 Time 0.020545 -2023-02-13 17:34:13,377 - Epoch: [35][ 540/ 1207] Overall Loss 0.343797 Objective Loss 0.343797 LR 0.001000 Time 0.020514 -2023-02-13 17:34:13,566 - Epoch: [35][ 550/ 1207] Overall Loss 0.344706 Objective Loss 0.344706 LR 0.001000 Time 0.020485 -2023-02-13 17:34:13,756 - Epoch: [35][ 560/ 1207] Overall Loss 0.345307 Objective Loss 0.345307 LR 0.001000 Time 0.020457 -2023-02-13 17:34:13,945 - Epoch: [35][ 570/ 1207] Overall Loss 0.346057 Objective Loss 0.346057 LR 0.001000 Time 0.020429 -2023-02-13 17:34:14,134 - Epoch: [35][ 580/ 1207] Overall Loss 0.346710 Objective Loss 0.346710 LR 0.001000 Time 0.020403 -2023-02-13 17:34:14,323 - Epoch: [35][ 590/ 1207] Overall Loss 0.347116 Objective Loss 0.347116 LR 0.001000 Time 0.020377 -2023-02-13 17:34:14,512 - Epoch: [35][ 600/ 1207] Overall Loss 0.347591 Objective Loss 0.347591 LR 0.001000 Time 0.020352 -2023-02-13 17:34:14,701 - Epoch: [35][ 610/ 1207] Overall Loss 0.347270 Objective Loss 0.347270 LR 0.001000 Time 0.020327 -2023-02-13 17:34:14,891 - Epoch: [35][ 620/ 1207] Overall Loss 0.346797 Objective Loss 0.346797 LR 0.001000 Time 0.020304 -2023-02-13 17:34:15,082 - Epoch: [35][ 630/ 1207] Overall Loss 0.346585 Objective Loss 0.346585 LR 0.001000 Time 0.020286 -2023-02-13 17:34:15,276 - Epoch: [35][ 640/ 1207] Overall Loss 0.346367 Objective Loss 0.346367 LR 0.001000 Time 0.020271 -2023-02-13 17:34:15,473 - Epoch: [35][ 650/ 1207] Overall Loss 0.345490 Objective Loss 0.345490 LR 0.001000 Time 0.020261 -2023-02-13 17:34:15,666 - Epoch: [35][ 660/ 1207] Overall Loss 0.345448 Objective Loss 0.345448 LR 0.001000 Time 0.020246 -2023-02-13 17:34:15,864 - Epoch: [35][ 670/ 1207] Overall Loss 0.345716 Objective Loss 0.345716 LR 0.001000 Time 0.020238 -2023-02-13 17:34:16,057 - Epoch: [35][ 680/ 1207] Overall Loss 0.345784 Objective Loss 0.345784 LR 0.001000 Time 0.020225 -2023-02-13 17:34:16,254 - Epoch: [35][ 690/ 1207] Overall Loss 0.345555 Objective Loss 0.345555 LR 0.001000 Time 0.020217 -2023-02-13 17:34:16,448 - Epoch: [35][ 700/ 1207] Overall Loss 0.346312 Objective Loss 0.346312 LR 0.001000 Time 0.020204 -2023-02-13 17:34:16,645 - Epoch: [35][ 710/ 1207] Overall Loss 0.346629 Objective Loss 0.346629 LR 0.001000 Time 0.020197 -2023-02-13 17:34:16,839 - Epoch: [35][ 720/ 1207] Overall Loss 0.346961 Objective Loss 0.346961 LR 0.001000 Time 0.020185 -2023-02-13 17:34:17,036 - Epoch: [35][ 730/ 1207] Overall Loss 0.347020 Objective Loss 0.347020 LR 0.001000 Time 0.020177 -2023-02-13 17:34:17,231 - Epoch: [35][ 740/ 1207] Overall Loss 0.346601 Objective Loss 0.346601 LR 0.001000 Time 0.020168 -2023-02-13 17:34:17,428 - Epoch: [35][ 750/ 1207] Overall Loss 0.346633 Objective Loss 0.346633 LR 0.001000 Time 0.020162 -2023-02-13 17:34:17,623 - Epoch: [35][ 760/ 1207] Overall Loss 0.346812 Objective Loss 0.346812 LR 0.001000 Time 0.020152 -2023-02-13 17:34:17,820 - Epoch: [35][ 770/ 1207] Overall Loss 0.346752 Objective Loss 0.346752 LR 0.001000 Time 0.020145 -2023-02-13 17:34:18,014 - Epoch: [35][ 780/ 1207] Overall Loss 0.346612 Objective Loss 0.346612 LR 0.001000 Time 0.020136 -2023-02-13 17:34:18,212 - Epoch: [35][ 790/ 1207] Overall Loss 0.346873 Objective Loss 0.346873 LR 0.001000 Time 0.020131 -2023-02-13 17:34:18,407 - Epoch: [35][ 800/ 1207] Overall Loss 0.347112 Objective Loss 0.347112 LR 0.001000 Time 0.020122 -2023-02-13 17:34:18,604 - Epoch: [35][ 810/ 1207] Overall Loss 0.347394 Objective Loss 0.347394 LR 0.001000 Time 0.020117 -2023-02-13 17:34:18,798 - Epoch: [35][ 820/ 1207] Overall Loss 0.347585 Objective Loss 0.347585 LR 0.001000 Time 0.020108 -2023-02-13 17:34:18,995 - Epoch: [35][ 830/ 1207] Overall Loss 0.347912 Objective Loss 0.347912 LR 0.001000 Time 0.020103 -2023-02-13 17:34:19,190 - Epoch: [35][ 840/ 1207] Overall Loss 0.348100 Objective Loss 0.348100 LR 0.001000 Time 0.020094 -2023-02-13 17:34:19,387 - Epoch: [35][ 850/ 1207] Overall Loss 0.348504 Objective Loss 0.348504 LR 0.001000 Time 0.020090 -2023-02-13 17:34:19,582 - Epoch: [35][ 860/ 1207] Overall Loss 0.348506 Objective Loss 0.348506 LR 0.001000 Time 0.020082 -2023-02-13 17:34:19,779 - Epoch: [35][ 870/ 1207] Overall Loss 0.348447 Objective Loss 0.348447 LR 0.001000 Time 0.020077 -2023-02-13 17:34:19,973 - Epoch: [35][ 880/ 1207] Overall Loss 0.348297 Objective Loss 0.348297 LR 0.001000 Time 0.020069 -2023-02-13 17:34:20,171 - Epoch: [35][ 890/ 1207] Overall Loss 0.348583 Objective Loss 0.348583 LR 0.001000 Time 0.020066 -2023-02-13 17:34:20,366 - Epoch: [35][ 900/ 1207] Overall Loss 0.348760 Objective Loss 0.348760 LR 0.001000 Time 0.020059 -2023-02-13 17:34:20,563 - Epoch: [35][ 910/ 1207] Overall Loss 0.348802 Objective Loss 0.348802 LR 0.001000 Time 0.020054 -2023-02-13 17:34:20,757 - Epoch: [35][ 920/ 1207] Overall Loss 0.348647 Objective Loss 0.348647 LR 0.001000 Time 0.020047 -2023-02-13 17:34:20,954 - Epoch: [35][ 930/ 1207] Overall Loss 0.348654 Objective Loss 0.348654 LR 0.001000 Time 0.020043 -2023-02-13 17:34:21,149 - Epoch: [35][ 940/ 1207] Overall Loss 0.348713 Objective Loss 0.348713 LR 0.001000 Time 0.020037 -2023-02-13 17:34:21,347 - Epoch: [35][ 950/ 1207] Overall Loss 0.348480 Objective Loss 0.348480 LR 0.001000 Time 0.020034 -2023-02-13 17:34:21,541 - Epoch: [35][ 960/ 1207] Overall Loss 0.348452 Objective Loss 0.348452 LR 0.001000 Time 0.020027 -2023-02-13 17:34:21,739 - Epoch: [35][ 970/ 1207] Overall Loss 0.348633 Objective Loss 0.348633 LR 0.001000 Time 0.020024 -2023-02-13 17:34:21,933 - Epoch: [35][ 980/ 1207] Overall Loss 0.348929 Objective Loss 0.348929 LR 0.001000 Time 0.020018 -2023-02-13 17:34:22,131 - Epoch: [35][ 990/ 1207] Overall Loss 0.349185 Objective Loss 0.349185 LR 0.001000 Time 0.020015 -2023-02-13 17:34:22,326 - Epoch: [35][ 1000/ 1207] Overall Loss 0.349277 Objective Loss 0.349277 LR 0.001000 Time 0.020009 -2023-02-13 17:34:22,523 - Epoch: [35][ 1010/ 1207] Overall Loss 0.349387 Objective Loss 0.349387 LR 0.001000 Time 0.020006 -2023-02-13 17:34:22,717 - Epoch: [35][ 1020/ 1207] Overall Loss 0.348985 Objective Loss 0.348985 LR 0.001000 Time 0.020000 -2023-02-13 17:34:22,914 - Epoch: [35][ 1030/ 1207] Overall Loss 0.349030 Objective Loss 0.349030 LR 0.001000 Time 0.019996 -2023-02-13 17:34:23,109 - Epoch: [35][ 1040/ 1207] Overall Loss 0.349237 Objective Loss 0.349237 LR 0.001000 Time 0.019991 -2023-02-13 17:34:23,306 - Epoch: [35][ 1050/ 1207] Overall Loss 0.348966 Objective Loss 0.348966 LR 0.001000 Time 0.019988 -2023-02-13 17:34:23,501 - Epoch: [35][ 1060/ 1207] Overall Loss 0.348961 Objective Loss 0.348961 LR 0.001000 Time 0.019983 -2023-02-13 17:34:23,698 - Epoch: [35][ 1070/ 1207] Overall Loss 0.349257 Objective Loss 0.349257 LR 0.001000 Time 0.019980 -2023-02-13 17:34:23,893 - Epoch: [35][ 1080/ 1207] Overall Loss 0.349392 Objective Loss 0.349392 LR 0.001000 Time 0.019975 -2023-02-13 17:34:24,090 - Epoch: [35][ 1090/ 1207] Overall Loss 0.349230 Objective Loss 0.349230 LR 0.001000 Time 0.019972 -2023-02-13 17:34:24,284 - Epoch: [35][ 1100/ 1207] Overall Loss 0.349171 Objective Loss 0.349171 LR 0.001000 Time 0.019967 -2023-02-13 17:34:24,481 - Epoch: [35][ 1110/ 1207] Overall Loss 0.349123 Objective Loss 0.349123 LR 0.001000 Time 0.019964 -2023-02-13 17:34:24,675 - Epoch: [35][ 1120/ 1207] Overall Loss 0.348903 Objective Loss 0.348903 LR 0.001000 Time 0.019959 -2023-02-13 17:34:24,872 - Epoch: [35][ 1130/ 1207] Overall Loss 0.349157 Objective Loss 0.349157 LR 0.001000 Time 0.019956 -2023-02-13 17:34:25,066 - Epoch: [35][ 1140/ 1207] Overall Loss 0.349313 Objective Loss 0.349313 LR 0.001000 Time 0.019951 -2023-02-13 17:34:25,264 - Epoch: [35][ 1150/ 1207] Overall Loss 0.349411 Objective Loss 0.349411 LR 0.001000 Time 0.019949 -2023-02-13 17:34:25,458 - Epoch: [35][ 1160/ 1207] Overall Loss 0.349573 Objective Loss 0.349573 LR 0.001000 Time 0.019944 -2023-02-13 17:34:25,650 - Epoch: [35][ 1170/ 1207] Overall Loss 0.350099 Objective Loss 0.350099 LR 0.001000 Time 0.019937 -2023-02-13 17:34:25,840 - Epoch: [35][ 1180/ 1207] Overall Loss 0.350030 Objective Loss 0.350030 LR 0.001000 Time 0.019929 -2023-02-13 17:34:26,031 - Epoch: [35][ 1190/ 1207] Overall Loss 0.350043 Objective Loss 0.350043 LR 0.001000 Time 0.019922 -2023-02-13 17:34:26,278 - Epoch: [35][ 1200/ 1207] Overall Loss 0.350113 Objective Loss 0.350113 LR 0.001000 Time 0.019962 -2023-02-13 17:34:26,395 - Epoch: [35][ 1207/ 1207] Overall Loss 0.350592 Objective Loss 0.350592 Top1 77.439024 Top5 97.865854 LR 0.001000 Time 0.019943 -2023-02-13 17:34:26,465 - --- validate (epoch=35)----------- -2023-02-13 17:34:26,465 - 34311 samples (256 per mini-batch) -2023-02-13 17:34:26,860 - Epoch: [35][ 10/ 135] Loss 0.367235 Top1 80.507812 Top5 97.070312 -2023-02-13 17:34:26,985 - Epoch: [35][ 20/ 135] Loss 0.389735 Top1 80.410156 Top5 96.738281 -2023-02-13 17:34:27,107 - Epoch: [35][ 30/ 135] Loss 0.377764 Top1 80.312500 Top5 96.744792 -2023-02-13 17:34:27,230 - Epoch: [35][ 40/ 135] Loss 0.368227 Top1 80.390625 Top5 96.816406 -2023-02-13 17:34:27,353 - Epoch: [35][ 50/ 135] Loss 0.369271 Top1 80.296875 Top5 96.703125 -2023-02-13 17:34:27,481 - Epoch: [35][ 60/ 135] Loss 0.365515 Top1 80.592448 Top5 96.829427 -2023-02-13 17:34:27,607 - Epoch: [35][ 70/ 135] Loss 0.365455 Top1 80.625000 Top5 96.741071 -2023-02-13 17:34:27,735 - Epoch: [35][ 80/ 135] Loss 0.368497 Top1 80.571289 Top5 96.787109 -2023-02-13 17:34:27,861 - Epoch: [35][ 90/ 135] Loss 0.370417 Top1 80.434028 Top5 96.801215 -2023-02-13 17:34:27,988 - Epoch: [35][ 100/ 135] Loss 0.368822 Top1 80.503906 Top5 96.808594 -2023-02-13 17:34:28,115 - Epoch: [35][ 110/ 135] Loss 0.367026 Top1 80.553977 Top5 96.867898 -2023-02-13 17:34:28,242 - Epoch: [35][ 120/ 135] Loss 0.363786 Top1 80.683594 Top5 96.861979 -2023-02-13 17:34:28,372 - Epoch: [35][ 130/ 135] Loss 0.364410 Top1 80.600962 Top5 96.868990 -2023-02-13 17:34:28,419 - Epoch: [35][ 135/ 135] Loss 0.361584 Top1 80.615546 Top5 96.881467 -2023-02-13 17:34:28,491 - ==> Top1: 80.616 Top5: 96.881 Loss: 0.362 - -2023-02-13 17:34:28,492 - ==> Confusion: -[[ 862 4 4 0 7 2 0 5 5 49 0 5 1 3 5 1 4 2 4 0 4] - [ 4 903 0 4 7 39 5 22 5 0 9 0 2 1 2 4 6 0 5 7 8] - [ 11 2 941 13 5 2 11 21 0 0 6 4 1 3 2 12 7 4 6 5 2] - [ 6 1 14 910 0 5 0 2 2 2 17 1 14 1 11 2 4 9 13 0 2] - [ 35 11 2 0 962 7 2 1 1 3 0 7 2 6 6 3 9 2 1 3 3] - [ 2 10 1 7 5 960 3 21 2 2 1 16 3 15 2 3 3 3 2 8 1] - [ 5 4 26 5 0 9 1011 6 0 0 6 4 3 1 0 4 2 2 0 10 1] - [ 4 4 10 3 2 35 4 910 1 1 5 10 4 1 0 0 0 1 17 10 2] - [ 25 5 0 1 1 0 0 0 880 29 18 2 1 13 21 2 1 2 7 0 1] - [ 103 2 4 1 2 3 0 2 40 814 1 1 1 26 5 0 0 1 1 1 4] - [ 3 1 3 4 0 1 1 3 17 1 988 2 2 8 2 0 1 0 9 1 4] - [ 3 1 1 0 1 12 1 2 0 1 0 892 52 6 0 1 2 14 3 12 1] - [ 3 2 1 5 2 2 0 1 2 0 0 28 866 1 2 4 0 23 5 3 9] - [ 10 3 2 0 4 13 0 1 15 13 13 10 6 903 5 2 6 2 1 10 5] - [ 16 1 1 30 3 3 0 1 16 4 7 2 4 1 972 3 1 8 12 0 7] - [ 9 1 5 2 4 1 6 1 0 0 0 7 11 1 1 952 7 18 2 13 5] - [ 4 4 0 2 6 3 1 0 3 1 0 3 2 3 2 14 987 3 1 12 10] - [ 10 1 2 5 2 0 0 3 0 0 0 7 21 3 2 17 0 971 1 2 4] - [ 3 2 4 18 1 4 0 34 2 0 8 0 6 0 18 1 0 2 981 1 1] - [ 0 0 1 1 1 10 5 16 0 0 2 23 5 2 0 4 5 3 0 1061 9] - [ 298 189 290 183 131 286 77 232 112 101 299 189 364 392 165 142 362 154 171 363 8934]] - -2023-02-13 17:34:28,493 - ==> Best [Top1: 81.618 Top5: 97.013 Sparsity:0.00 Params: 148928 on epoch: 33] -2023-02-13 17:34:28,493 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:34:28,499 - - -2023-02-13 17:34:28,499 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:34:29,379 - Epoch: [36][ 10/ 1207] Overall Loss 0.365617 Objective Loss 0.365617 LR 0.001000 Time 0.087876 -2023-02-13 17:34:29,575 - Epoch: [36][ 20/ 1207] Overall Loss 0.365183 Objective Loss 0.365183 LR 0.001000 Time 0.053736 -2023-02-13 17:34:29,763 - Epoch: [36][ 30/ 1207] Overall Loss 0.350955 Objective Loss 0.350955 LR 0.001000 Time 0.042082 -2023-02-13 17:34:29,952 - Epoch: [36][ 40/ 1207] Overall Loss 0.351218 Objective Loss 0.351218 LR 0.001000 Time 0.036262 -2023-02-13 17:34:30,141 - Epoch: [36][ 50/ 1207] Overall Loss 0.340827 Objective Loss 0.340827 LR 0.001000 Time 0.032794 -2023-02-13 17:34:30,330 - Epoch: [36][ 60/ 1207] Overall Loss 0.340721 Objective Loss 0.340721 LR 0.001000 Time 0.030467 -2023-02-13 17:34:30,518 - Epoch: [36][ 70/ 1207] Overall Loss 0.337608 Objective Loss 0.337608 LR 0.001000 Time 0.028796 -2023-02-13 17:34:30,706 - Epoch: [36][ 80/ 1207] Overall Loss 0.339733 Objective Loss 0.339733 LR 0.001000 Time 0.027546 -2023-02-13 17:34:30,896 - Epoch: [36][ 90/ 1207] Overall Loss 0.340053 Objective Loss 0.340053 LR 0.001000 Time 0.026592 -2023-02-13 17:34:31,084 - Epoch: [36][ 100/ 1207] Overall Loss 0.340143 Objective Loss 0.340143 LR 0.001000 Time 0.025811 -2023-02-13 17:34:31,273 - Epoch: [36][ 110/ 1207] Overall Loss 0.335726 Objective Loss 0.335726 LR 0.001000 Time 0.025180 -2023-02-13 17:34:31,462 - Epoch: [36][ 120/ 1207] Overall Loss 0.337471 Objective Loss 0.337471 LR 0.001000 Time 0.024652 -2023-02-13 17:34:31,651 - Epoch: [36][ 130/ 1207] Overall Loss 0.339438 Objective Loss 0.339438 LR 0.001000 Time 0.024203 -2023-02-13 17:34:31,840 - Epoch: [36][ 140/ 1207] Overall Loss 0.340007 Objective Loss 0.340007 LR 0.001000 Time 0.023821 -2023-02-13 17:34:32,028 - Epoch: [36][ 150/ 1207] Overall Loss 0.340766 Objective Loss 0.340766 LR 0.001000 Time 0.023487 -2023-02-13 17:34:32,217 - Epoch: [36][ 160/ 1207] Overall Loss 0.342570 Objective Loss 0.342570 LR 0.001000 Time 0.023194 -2023-02-13 17:34:32,405 - Epoch: [36][ 170/ 1207] Overall Loss 0.344761 Objective Loss 0.344761 LR 0.001000 Time 0.022933 -2023-02-13 17:34:32,593 - Epoch: [36][ 180/ 1207] Overall Loss 0.344457 Objective Loss 0.344457 LR 0.001000 Time 0.022701 -2023-02-13 17:34:32,781 - Epoch: [36][ 190/ 1207] Overall Loss 0.346036 Objective Loss 0.346036 LR 0.001000 Time 0.022494 -2023-02-13 17:34:32,969 - Epoch: [36][ 200/ 1207] Overall Loss 0.345667 Objective Loss 0.345667 LR 0.001000 Time 0.022307 -2023-02-13 17:34:33,158 - Epoch: [36][ 210/ 1207] Overall Loss 0.345590 Objective Loss 0.345590 LR 0.001000 Time 0.022144 -2023-02-13 17:34:33,348 - Epoch: [36][ 220/ 1207] Overall Loss 0.346174 Objective Loss 0.346174 LR 0.001000 Time 0.022003 -2023-02-13 17:34:33,540 - Epoch: [36][ 230/ 1207] Overall Loss 0.346177 Objective Loss 0.346177 LR 0.001000 Time 0.021876 -2023-02-13 17:34:33,731 - Epoch: [36][ 240/ 1207] Overall Loss 0.346248 Objective Loss 0.346248 LR 0.001000 Time 0.021760 -2023-02-13 17:34:33,921 - Epoch: [36][ 250/ 1207] Overall Loss 0.346058 Objective Loss 0.346058 LR 0.001000 Time 0.021650 -2023-02-13 17:34:34,113 - Epoch: [36][ 260/ 1207] Overall Loss 0.344548 Objective Loss 0.344548 LR 0.001000 Time 0.021553 -2023-02-13 17:34:34,305 - Epoch: [36][ 270/ 1207] Overall Loss 0.345248 Objective Loss 0.345248 LR 0.001000 Time 0.021463 -2023-02-13 17:34:34,496 - Epoch: [36][ 280/ 1207] Overall Loss 0.344697 Objective Loss 0.344697 LR 0.001000 Time 0.021378 -2023-02-13 17:34:34,685 - Epoch: [36][ 290/ 1207] Overall Loss 0.344530 Objective Loss 0.344530 LR 0.001000 Time 0.021290 -2023-02-13 17:34:34,873 - Epoch: [36][ 300/ 1207] Overall Loss 0.346202 Objective Loss 0.346202 LR 0.001000 Time 0.021208 -2023-02-13 17:34:35,062 - Epoch: [36][ 310/ 1207] Overall Loss 0.345817 Objective Loss 0.345817 LR 0.001000 Time 0.021132 -2023-02-13 17:34:35,252 - Epoch: [36][ 320/ 1207] Overall Loss 0.346387 Objective Loss 0.346387 LR 0.001000 Time 0.021063 -2023-02-13 17:34:35,440 - Epoch: [36][ 330/ 1207] Overall Loss 0.346176 Objective Loss 0.346176 LR 0.001000 Time 0.020995 -2023-02-13 17:34:35,629 - Epoch: [36][ 340/ 1207] Overall Loss 0.346504 Objective Loss 0.346504 LR 0.001000 Time 0.020932 -2023-02-13 17:34:35,819 - Epoch: [36][ 350/ 1207] Overall Loss 0.346714 Objective Loss 0.346714 LR 0.001000 Time 0.020874 -2023-02-13 17:34:36,008 - Epoch: [36][ 360/ 1207] Overall Loss 0.346257 Objective Loss 0.346257 LR 0.001000 Time 0.020819 -2023-02-13 17:34:36,198 - Epoch: [36][ 370/ 1207] Overall Loss 0.345830 Objective Loss 0.345830 LR 0.001000 Time 0.020768 -2023-02-13 17:34:36,387 - Epoch: [36][ 380/ 1207] Overall Loss 0.346334 Objective Loss 0.346334 LR 0.001000 Time 0.020718 -2023-02-13 17:34:36,576 - Epoch: [36][ 390/ 1207] Overall Loss 0.346513 Objective Loss 0.346513 LR 0.001000 Time 0.020671 -2023-02-13 17:34:36,765 - Epoch: [36][ 400/ 1207] Overall Loss 0.346314 Objective Loss 0.346314 LR 0.001000 Time 0.020627 -2023-02-13 17:34:36,955 - Epoch: [36][ 410/ 1207] Overall Loss 0.346359 Objective Loss 0.346359 LR 0.001000 Time 0.020584 -2023-02-13 17:34:37,143 - Epoch: [36][ 420/ 1207] Overall Loss 0.346181 Objective Loss 0.346181 LR 0.001000 Time 0.020543 -2023-02-13 17:34:37,333 - Epoch: [36][ 430/ 1207] Overall Loss 0.345637 Objective Loss 0.345637 LR 0.001000 Time 0.020505 -2023-02-13 17:34:37,522 - Epoch: [36][ 440/ 1207] Overall Loss 0.345533 Objective Loss 0.345533 LR 0.001000 Time 0.020469 -2023-02-13 17:34:37,712 - Epoch: [36][ 450/ 1207] Overall Loss 0.346453 Objective Loss 0.346453 LR 0.001000 Time 0.020434 -2023-02-13 17:34:37,900 - Epoch: [36][ 460/ 1207] Overall Loss 0.346991 Objective Loss 0.346991 LR 0.001000 Time 0.020398 -2023-02-13 17:34:38,089 - Epoch: [36][ 470/ 1207] Overall Loss 0.347101 Objective Loss 0.347101 LR 0.001000 Time 0.020366 -2023-02-13 17:34:38,279 - Epoch: [36][ 480/ 1207] Overall Loss 0.347292 Objective Loss 0.347292 LR 0.001000 Time 0.020336 -2023-02-13 17:34:38,468 - Epoch: [36][ 490/ 1207] Overall Loss 0.347228 Objective Loss 0.347228 LR 0.001000 Time 0.020306 -2023-02-13 17:34:38,657 - Epoch: [36][ 500/ 1207] Overall Loss 0.347277 Objective Loss 0.347277 LR 0.001000 Time 0.020277 -2023-02-13 17:34:38,845 - Epoch: [36][ 510/ 1207] Overall Loss 0.346762 Objective Loss 0.346762 LR 0.001000 Time 0.020248 -2023-02-13 17:34:39,034 - Epoch: [36][ 520/ 1207] Overall Loss 0.346825 Objective Loss 0.346825 LR 0.001000 Time 0.020221 -2023-02-13 17:34:39,224 - Epoch: [36][ 530/ 1207] Overall Loss 0.346173 Objective Loss 0.346173 LR 0.001000 Time 0.020196 -2023-02-13 17:34:39,413 - Epoch: [36][ 540/ 1207] Overall Loss 0.346536 Objective Loss 0.346536 LR 0.001000 Time 0.020172 -2023-02-13 17:34:39,602 - Epoch: [36][ 550/ 1207] Overall Loss 0.347334 Objective Loss 0.347334 LR 0.001000 Time 0.020148 -2023-02-13 17:34:39,791 - Epoch: [36][ 560/ 1207] Overall Loss 0.347690 Objective Loss 0.347690 LR 0.001000 Time 0.020125 -2023-02-13 17:34:39,979 - Epoch: [36][ 570/ 1207] Overall Loss 0.347582 Objective Loss 0.347582 LR 0.001000 Time 0.020101 -2023-02-13 17:34:40,168 - Epoch: [36][ 580/ 1207] Overall Loss 0.347583 Objective Loss 0.347583 LR 0.001000 Time 0.020080 -2023-02-13 17:34:40,357 - Epoch: [36][ 590/ 1207] Overall Loss 0.347586 Objective Loss 0.347586 LR 0.001000 Time 0.020060 -2023-02-13 17:34:40,546 - Epoch: [36][ 600/ 1207] Overall Loss 0.347511 Objective Loss 0.347511 LR 0.001000 Time 0.020040 -2023-02-13 17:34:40,735 - Epoch: [36][ 610/ 1207] Overall Loss 0.347088 Objective Loss 0.347088 LR 0.001000 Time 0.020020 -2023-02-13 17:34:40,925 - Epoch: [36][ 620/ 1207] Overall Loss 0.347028 Objective Loss 0.347028 LR 0.001000 Time 0.020003 -2023-02-13 17:34:41,113 - Epoch: [36][ 630/ 1207] Overall Loss 0.347607 Objective Loss 0.347607 LR 0.001000 Time 0.019984 -2023-02-13 17:34:41,302 - Epoch: [36][ 640/ 1207] Overall Loss 0.347696 Objective Loss 0.347696 LR 0.001000 Time 0.019967 -2023-02-13 17:34:41,491 - Epoch: [36][ 650/ 1207] Overall Loss 0.348210 Objective Loss 0.348210 LR 0.001000 Time 0.019949 -2023-02-13 17:34:41,680 - Epoch: [36][ 660/ 1207] Overall Loss 0.348446 Objective Loss 0.348446 LR 0.001000 Time 0.019933 -2023-02-13 17:34:41,869 - Epoch: [36][ 670/ 1207] Overall Loss 0.348000 Objective Loss 0.348000 LR 0.001000 Time 0.019917 -2023-02-13 17:34:42,058 - Epoch: [36][ 680/ 1207] Overall Loss 0.348757 Objective Loss 0.348757 LR 0.001000 Time 0.019901 -2023-02-13 17:34:42,248 - Epoch: [36][ 690/ 1207] Overall Loss 0.348163 Objective Loss 0.348163 LR 0.001000 Time 0.019887 -2023-02-13 17:34:42,437 - Epoch: [36][ 700/ 1207] Overall Loss 0.349242 Objective Loss 0.349242 LR 0.001000 Time 0.019873 -2023-02-13 17:34:42,626 - Epoch: [36][ 710/ 1207] Overall Loss 0.348836 Objective Loss 0.348836 LR 0.001000 Time 0.019858 -2023-02-13 17:34:42,815 - Epoch: [36][ 720/ 1207] Overall Loss 0.348716 Objective Loss 0.348716 LR 0.001000 Time 0.019844 -2023-02-13 17:34:43,003 - Epoch: [36][ 730/ 1207] Overall Loss 0.348661 Objective Loss 0.348661 LR 0.001000 Time 0.019830 -2023-02-13 17:34:43,193 - Epoch: [36][ 740/ 1207] Overall Loss 0.348916 Objective Loss 0.348916 LR 0.001000 Time 0.019818 -2023-02-13 17:34:43,381 - Epoch: [36][ 750/ 1207] Overall Loss 0.348719 Objective Loss 0.348719 LR 0.001000 Time 0.019805 -2023-02-13 17:34:43,570 - Epoch: [36][ 760/ 1207] Overall Loss 0.348329 Objective Loss 0.348329 LR 0.001000 Time 0.019792 -2023-02-13 17:34:43,758 - Epoch: [36][ 770/ 1207] Overall Loss 0.348752 Objective Loss 0.348752 LR 0.001000 Time 0.019779 -2023-02-13 17:34:43,947 - Epoch: [36][ 780/ 1207] Overall Loss 0.348932 Objective Loss 0.348932 LR 0.001000 Time 0.019766 -2023-02-13 17:34:44,136 - Epoch: [36][ 790/ 1207] Overall Loss 0.349164 Objective Loss 0.349164 LR 0.001000 Time 0.019755 -2023-02-13 17:34:44,325 - Epoch: [36][ 800/ 1207] Overall Loss 0.348759 Objective Loss 0.348759 LR 0.001000 Time 0.019744 -2023-02-13 17:34:44,514 - Epoch: [36][ 810/ 1207] Overall Loss 0.348486 Objective Loss 0.348486 LR 0.001000 Time 0.019733 -2023-02-13 17:34:44,702 - Epoch: [36][ 820/ 1207] Overall Loss 0.348698 Objective Loss 0.348698 LR 0.001000 Time 0.019722 -2023-02-13 17:34:44,891 - Epoch: [36][ 830/ 1207] Overall Loss 0.348952 Objective Loss 0.348952 LR 0.001000 Time 0.019710 -2023-02-13 17:34:45,079 - Epoch: [36][ 840/ 1207] Overall Loss 0.349449 Objective Loss 0.349449 LR 0.001000 Time 0.019700 -2023-02-13 17:34:45,269 - Epoch: [36][ 850/ 1207] Overall Loss 0.349250 Objective Loss 0.349250 LR 0.001000 Time 0.019690 -2023-02-13 17:34:45,458 - Epoch: [36][ 860/ 1207] Overall Loss 0.349140 Objective Loss 0.349140 LR 0.001000 Time 0.019681 -2023-02-13 17:34:45,647 - Epoch: [36][ 870/ 1207] Overall Loss 0.349104 Objective Loss 0.349104 LR 0.001000 Time 0.019671 -2023-02-13 17:34:45,836 - Epoch: [36][ 880/ 1207] Overall Loss 0.349465 Objective Loss 0.349465 LR 0.001000 Time 0.019663 -2023-02-13 17:34:46,024 - Epoch: [36][ 890/ 1207] Overall Loss 0.349396 Objective Loss 0.349396 LR 0.001000 Time 0.019653 -2023-02-13 17:34:46,215 - Epoch: [36][ 900/ 1207] Overall Loss 0.349407 Objective Loss 0.349407 LR 0.001000 Time 0.019646 -2023-02-13 17:34:46,404 - Epoch: [36][ 910/ 1207] Overall Loss 0.349429 Objective Loss 0.349429 LR 0.001000 Time 0.019637 -2023-02-13 17:34:46,595 - Epoch: [36][ 920/ 1207] Overall Loss 0.349146 Objective Loss 0.349146 LR 0.001000 Time 0.019631 -2023-02-13 17:34:46,785 - Epoch: [36][ 930/ 1207] Overall Loss 0.348857 Objective Loss 0.348857 LR 0.001000 Time 0.019624 -2023-02-13 17:34:46,975 - Epoch: [36][ 940/ 1207] Overall Loss 0.349055 Objective Loss 0.349055 LR 0.001000 Time 0.019616 -2023-02-13 17:34:47,164 - Epoch: [36][ 950/ 1207] Overall Loss 0.348911 Objective Loss 0.348911 LR 0.001000 Time 0.019608 -2023-02-13 17:34:47,354 - Epoch: [36][ 960/ 1207] Overall Loss 0.348620 Objective Loss 0.348620 LR 0.001000 Time 0.019602 -2023-02-13 17:34:47,543 - Epoch: [36][ 970/ 1207] Overall Loss 0.348407 Objective Loss 0.348407 LR 0.001000 Time 0.019594 -2023-02-13 17:34:47,731 - Epoch: [36][ 980/ 1207] Overall Loss 0.348407 Objective Loss 0.348407 LR 0.001000 Time 0.019586 -2023-02-13 17:34:47,921 - Epoch: [36][ 990/ 1207] Overall Loss 0.348162 Objective Loss 0.348162 LR 0.001000 Time 0.019579 -2023-02-13 17:34:48,109 - Epoch: [36][ 1000/ 1207] Overall Loss 0.347899 Objective Loss 0.347899 LR 0.001000 Time 0.019572 -2023-02-13 17:34:48,299 - Epoch: [36][ 1010/ 1207] Overall Loss 0.347905 Objective Loss 0.347905 LR 0.001000 Time 0.019566 -2023-02-13 17:34:48,488 - Epoch: [36][ 1020/ 1207] Overall Loss 0.347791 Objective Loss 0.347791 LR 0.001000 Time 0.019559 -2023-02-13 17:34:48,677 - Epoch: [36][ 1030/ 1207] Overall Loss 0.348089 Objective Loss 0.348089 LR 0.001000 Time 0.019552 -2023-02-13 17:34:48,866 - Epoch: [36][ 1040/ 1207] Overall Loss 0.347794 Objective Loss 0.347794 LR 0.001000 Time 0.019545 -2023-02-13 17:34:49,055 - Epoch: [36][ 1050/ 1207] Overall Loss 0.347856 Objective Loss 0.347856 LR 0.001000 Time 0.019538 -2023-02-13 17:34:49,244 - Epoch: [36][ 1060/ 1207] Overall Loss 0.347683 Objective Loss 0.347683 LR 0.001000 Time 0.019532 -2023-02-13 17:34:49,433 - Epoch: [36][ 1070/ 1207] Overall Loss 0.347794 Objective Loss 0.347794 LR 0.001000 Time 0.019526 -2023-02-13 17:34:49,623 - Epoch: [36][ 1080/ 1207] Overall Loss 0.348009 Objective Loss 0.348009 LR 0.001000 Time 0.019521 -2023-02-13 17:34:49,813 - Epoch: [36][ 1090/ 1207] Overall Loss 0.348247 Objective Loss 0.348247 LR 0.001000 Time 0.019516 -2023-02-13 17:34:50,003 - Epoch: [36][ 1100/ 1207] Overall Loss 0.348210 Objective Loss 0.348210 LR 0.001000 Time 0.019510 -2023-02-13 17:34:50,193 - Epoch: [36][ 1110/ 1207] Overall Loss 0.348400 Objective Loss 0.348400 LR 0.001000 Time 0.019505 -2023-02-13 17:34:50,383 - Epoch: [36][ 1120/ 1207] Overall Loss 0.348362 Objective Loss 0.348362 LR 0.001000 Time 0.019501 -2023-02-13 17:34:50,572 - Epoch: [36][ 1130/ 1207] Overall Loss 0.348514 Objective Loss 0.348514 LR 0.001000 Time 0.019495 -2023-02-13 17:34:50,762 - Epoch: [36][ 1140/ 1207] Overall Loss 0.348799 Objective Loss 0.348799 LR 0.001000 Time 0.019490 -2023-02-13 17:34:50,952 - Epoch: [36][ 1150/ 1207] Overall Loss 0.348775 Objective Loss 0.348775 LR 0.001000 Time 0.019486 -2023-02-13 17:34:51,141 - Epoch: [36][ 1160/ 1207] Overall Loss 0.348609 Objective Loss 0.348609 LR 0.001000 Time 0.019480 -2023-02-13 17:34:51,331 - Epoch: [36][ 1170/ 1207] Overall Loss 0.348660 Objective Loss 0.348660 LR 0.001000 Time 0.019476 -2023-02-13 17:34:51,520 - Epoch: [36][ 1180/ 1207] Overall Loss 0.348506 Objective Loss 0.348506 LR 0.001000 Time 0.019471 -2023-02-13 17:34:51,709 - Epoch: [36][ 1190/ 1207] Overall Loss 0.348267 Objective Loss 0.348267 LR 0.001000 Time 0.019466 -2023-02-13 17:34:51,950 - Epoch: [36][ 1200/ 1207] Overall Loss 0.348349 Objective Loss 0.348349 LR 0.001000 Time 0.019504 -2023-02-13 17:34:52,067 - Epoch: [36][ 1207/ 1207] Overall Loss 0.348320 Objective Loss 0.348320 Top1 84.146341 Top5 96.341463 LR 0.001000 Time 0.019488 -2023-02-13 17:34:52,149 - --- validate (epoch=36)----------- -2023-02-13 17:34:52,150 - 34311 samples (256 per mini-batch) -2023-02-13 17:34:52,560 - Epoch: [36][ 10/ 135] Loss 0.358714 Top1 82.109375 Top5 97.109375 -2023-02-13 17:34:52,701 - Epoch: [36][ 20/ 135] Loss 0.374069 Top1 80.800781 Top5 96.992188 -2023-02-13 17:34:52,825 - Epoch: [36][ 30/ 135] Loss 0.391861 Top1 79.934896 Top5 96.796875 -2023-02-13 17:34:52,952 - Epoch: [36][ 40/ 135] Loss 0.387180 Top1 79.892578 Top5 96.865234 -2023-02-13 17:34:53,079 - Epoch: [36][ 50/ 135] Loss 0.386233 Top1 80.000000 Top5 96.992188 -2023-02-13 17:34:53,201 - Epoch: [36][ 60/ 135] Loss 0.385044 Top1 80.130208 Top5 96.992188 -2023-02-13 17:34:53,323 - Epoch: [36][ 70/ 135] Loss 0.383373 Top1 80.256696 Top5 97.014509 -2023-02-13 17:34:53,444 - Epoch: [36][ 80/ 135] Loss 0.383223 Top1 80.190430 Top5 97.021484 -2023-02-13 17:34:53,568 - Epoch: [36][ 90/ 135] Loss 0.385813 Top1 80.112847 Top5 97.013889 -2023-02-13 17:34:53,688 - Epoch: [36][ 100/ 135] Loss 0.385794 Top1 80.031250 Top5 97.003906 -2023-02-13 17:34:53,810 - Epoch: [36][ 110/ 135] Loss 0.384724 Top1 80.088778 Top5 97.041903 -2023-02-13 17:34:53,933 - Epoch: [36][ 120/ 135] Loss 0.385950 Top1 80.149740 Top5 97.005208 -2023-02-13 17:34:54,060 - Epoch: [36][ 130/ 135] Loss 0.385551 Top1 80.213341 Top5 96.935096 -2023-02-13 17:34:54,105 - Epoch: [36][ 135/ 135] Loss 0.386665 Top1 80.096762 Top5 96.922270 -2023-02-13 17:34:54,172 - ==> Top1: 80.097 Top5: 96.922 Loss: 0.387 - -2023-02-13 17:34:54,172 - ==> Confusion: -[[ 862 4 3 1 6 6 1 1 4 48 1 5 2 5 3 1 3 2 2 1 6] - [ 2 917 1 2 5 26 3 35 4 3 4 4 1 1 0 2 4 0 9 5 5] - [ 17 7 922 9 6 2 12 31 0 2 5 2 1 6 2 5 2 1 15 2 9] - [ 9 3 15 869 0 1 1 4 3 2 17 2 10 1 25 2 2 8 33 3 6] - [ 34 9 0 0 956 7 2 3 1 7 0 8 0 3 10 5 9 1 1 7 3] - [ 3 35 1 2 4 920 2 30 2 8 4 19 4 12 1 2 3 1 4 6 7] - [ 1 4 16 5 1 4 1016 13 2 1 4 2 2 1 0 3 4 0 2 16 2] - [ 1 4 7 0 3 22 4 929 1 3 5 5 3 3 0 0 0 0 21 9 4] - [ 26 0 0 1 1 1 0 1 902 40 8 2 0 9 9 1 0 2 6 0 0] - [ 103 2 1 0 1 1 0 2 51 827 0 0 1 14 5 0 0 1 1 0 2] - [ 3 4 1 8 0 2 2 4 26 1 965 2 0 10 2 0 1 1 16 1 2] - [ 1 2 1 0 2 16 0 5 1 2 0 908 22 9 3 4 1 7 3 17 1] - [ 3 1 3 4 1 1 1 3 3 0 0 62 832 2 7 2 2 15 6 6 5] - [ 7 4 3 0 4 13 0 3 31 38 8 5 2 880 8 4 3 0 1 8 2] - [ 20 2 1 10 3 1 0 3 37 5 3 2 5 4 966 0 1 4 17 0 8] - [ 7 1 9 2 7 2 6 2 0 1 1 9 12 5 1 929 20 15 0 13 4] - [ 5 7 3 1 13 3 1 1 7 1 1 6 3 6 1 8 976 1 2 6 9] - [ 7 3 1 3 2 2 3 3 2 4 0 20 23 3 1 6 0 962 1 3 2] - [ 4 5 2 8 0 2 1 41 6 1 3 1 4 0 8 0 1 1 994 2 2] - [ 1 2 1 1 1 9 7 16 1 0 3 20 1 2 0 4 4 1 1 1067 6] - [ 268 307 203 122 154 212 81 307 186 137 224 187 344 399 214 94 254 128 284 446 8883]] - -2023-02-13 17:34:54,174 - ==> Best [Top1: 81.618 Top5: 97.013 Sparsity:0.00 Params: 148928 on epoch: 33] -2023-02-13 17:34:54,174 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:34:54,179 - - -2023-02-13 17:34:54,180 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:34:55,047 - Epoch: [37][ 10/ 1207] Overall Loss 0.355922 Objective Loss 0.355922 LR 0.001000 Time 0.086642 -2023-02-13 17:34:55,246 - Epoch: [37][ 20/ 1207] Overall Loss 0.380101 Objective Loss 0.380101 LR 0.001000 Time 0.053274 -2023-02-13 17:34:55,440 - Epoch: [37][ 30/ 1207] Overall Loss 0.369971 Objective Loss 0.369971 LR 0.001000 Time 0.041979 -2023-02-13 17:34:55,632 - Epoch: [37][ 40/ 1207] Overall Loss 0.362115 Objective Loss 0.362115 LR 0.001000 Time 0.036253 -2023-02-13 17:34:55,825 - Epoch: [37][ 50/ 1207] Overall Loss 0.354986 Objective Loss 0.354986 LR 0.001000 Time 0.032874 -2023-02-13 17:34:56,016 - Epoch: [37][ 60/ 1207] Overall Loss 0.353549 Objective Loss 0.353549 LR 0.001000 Time 0.030572 -2023-02-13 17:34:56,210 - Epoch: [37][ 70/ 1207] Overall Loss 0.352334 Objective Loss 0.352334 LR 0.001000 Time 0.028966 -2023-02-13 17:34:56,401 - Epoch: [37][ 80/ 1207] Overall Loss 0.353354 Objective Loss 0.353354 LR 0.001000 Time 0.027727 -2023-02-13 17:34:56,595 - Epoch: [37][ 90/ 1207] Overall Loss 0.353974 Objective Loss 0.353974 LR 0.001000 Time 0.026794 -2023-02-13 17:34:56,786 - Epoch: [37][ 100/ 1207] Overall Loss 0.349732 Objective Loss 0.349732 LR 0.001000 Time 0.026021 -2023-02-13 17:34:56,979 - Epoch: [37][ 110/ 1207] Overall Loss 0.347327 Objective Loss 0.347327 LR 0.001000 Time 0.025407 -2023-02-13 17:34:57,169 - Epoch: [37][ 120/ 1207] Overall Loss 0.348261 Objective Loss 0.348261 LR 0.001000 Time 0.024873 -2023-02-13 17:34:57,362 - Epoch: [37][ 130/ 1207] Overall Loss 0.346499 Objective Loss 0.346499 LR 0.001000 Time 0.024442 -2023-02-13 17:34:57,553 - Epoch: [37][ 140/ 1207] Overall Loss 0.347446 Objective Loss 0.347446 LR 0.001000 Time 0.024056 -2023-02-13 17:34:57,746 - Epoch: [37][ 150/ 1207] Overall Loss 0.346438 Objective Loss 0.346438 LR 0.001000 Time 0.023735 -2023-02-13 17:34:57,936 - Epoch: [37][ 160/ 1207] Overall Loss 0.345972 Objective Loss 0.345972 LR 0.001000 Time 0.023440 -2023-02-13 17:34:58,129 - Epoch: [37][ 170/ 1207] Overall Loss 0.345944 Objective Loss 0.345944 LR 0.001000 Time 0.023195 -2023-02-13 17:34:58,320 - Epoch: [37][ 180/ 1207] Overall Loss 0.347136 Objective Loss 0.347136 LR 0.001000 Time 0.022965 -2023-02-13 17:34:58,513 - Epoch: [37][ 190/ 1207] Overall Loss 0.345656 Objective Loss 0.345656 LR 0.001000 Time 0.022770 -2023-02-13 17:34:58,704 - Epoch: [37][ 200/ 1207] Overall Loss 0.347104 Objective Loss 0.347104 LR 0.001000 Time 0.022584 -2023-02-13 17:34:58,897 - Epoch: [37][ 210/ 1207] Overall Loss 0.346876 Objective Loss 0.346876 LR 0.001000 Time 0.022427 -2023-02-13 17:34:59,087 - Epoch: [37][ 220/ 1207] Overall Loss 0.346872 Objective Loss 0.346872 LR 0.001000 Time 0.022270 -2023-02-13 17:34:59,281 - Epoch: [37][ 230/ 1207] Overall Loss 0.345224 Objective Loss 0.345224 LR 0.001000 Time 0.022143 -2023-02-13 17:34:59,471 - Epoch: [37][ 240/ 1207] Overall Loss 0.344026 Objective Loss 0.344026 LR 0.001000 Time 0.022010 -2023-02-13 17:34:59,664 - Epoch: [37][ 250/ 1207] Overall Loss 0.343322 Objective Loss 0.343322 LR 0.001000 Time 0.021899 -2023-02-13 17:34:59,855 - Epoch: [37][ 260/ 1207] Overall Loss 0.344945 Objective Loss 0.344945 LR 0.001000 Time 0.021789 -2023-02-13 17:35:00,048 - Epoch: [37][ 270/ 1207] Overall Loss 0.346712 Objective Loss 0.346712 LR 0.001000 Time 0.021696 -2023-02-13 17:35:00,238 - Epoch: [37][ 280/ 1207] Overall Loss 0.346422 Objective Loss 0.346422 LR 0.001000 Time 0.021600 -2023-02-13 17:35:00,432 - Epoch: [37][ 290/ 1207] Overall Loss 0.346809 Objective Loss 0.346809 LR 0.001000 Time 0.021523 -2023-02-13 17:35:00,623 - Epoch: [37][ 300/ 1207] Overall Loss 0.346409 Objective Loss 0.346409 LR 0.001000 Time 0.021440 -2023-02-13 17:35:00,817 - Epoch: [37][ 310/ 1207] Overall Loss 0.347648 Objective Loss 0.347648 LR 0.001000 Time 0.021373 -2023-02-13 17:35:01,008 - Epoch: [37][ 320/ 1207] Overall Loss 0.347773 Objective Loss 0.347773 LR 0.001000 Time 0.021301 -2023-02-13 17:35:01,202 - Epoch: [37][ 330/ 1207] Overall Loss 0.347522 Objective Loss 0.347522 LR 0.001000 Time 0.021241 -2023-02-13 17:35:01,393 - Epoch: [37][ 340/ 1207] Overall Loss 0.346697 Objective Loss 0.346697 LR 0.001000 Time 0.021178 -2023-02-13 17:35:01,587 - Epoch: [37][ 350/ 1207] Overall Loss 0.345994 Objective Loss 0.345994 LR 0.001000 Time 0.021125 -2023-02-13 17:35:01,778 - Epoch: [37][ 360/ 1207] Overall Loss 0.345760 Objective Loss 0.345760 LR 0.001000 Time 0.021068 -2023-02-13 17:35:01,971 - Epoch: [37][ 370/ 1207] Overall Loss 0.345833 Objective Loss 0.345833 LR 0.001000 Time 0.021020 -2023-02-13 17:35:02,161 - Epoch: [37][ 380/ 1207] Overall Loss 0.345774 Objective Loss 0.345774 LR 0.001000 Time 0.020967 -2023-02-13 17:35:02,355 - Epoch: [37][ 390/ 1207] Overall Loss 0.345455 Objective Loss 0.345455 LR 0.001000 Time 0.020926 -2023-02-13 17:35:02,549 - Epoch: [37][ 400/ 1207] Overall Loss 0.345365 Objective Loss 0.345365 LR 0.001000 Time 0.020887 -2023-02-13 17:35:02,744 - Epoch: [37][ 410/ 1207] Overall Loss 0.345052 Objective Loss 0.345052 LR 0.001000 Time 0.020852 -2023-02-13 17:35:02,936 - Epoch: [37][ 420/ 1207] Overall Loss 0.345654 Objective Loss 0.345654 LR 0.001000 Time 0.020812 -2023-02-13 17:35:03,132 - Epoch: [37][ 430/ 1207] Overall Loss 0.345732 Objective Loss 0.345732 LR 0.001000 Time 0.020781 -2023-02-13 17:35:03,324 - Epoch: [37][ 440/ 1207] Overall Loss 0.345775 Objective Loss 0.345775 LR 0.001000 Time 0.020746 -2023-02-13 17:35:03,520 - Epoch: [37][ 450/ 1207] Overall Loss 0.345925 Objective Loss 0.345925 LR 0.001000 Time 0.020719 -2023-02-13 17:35:03,712 - Epoch: [37][ 460/ 1207] Overall Loss 0.346262 Objective Loss 0.346262 LR 0.001000 Time 0.020685 -2023-02-13 17:35:03,907 - Epoch: [37][ 470/ 1207] Overall Loss 0.346735 Objective Loss 0.346735 LR 0.001000 Time 0.020660 -2023-02-13 17:35:04,099 - Epoch: [37][ 480/ 1207] Overall Loss 0.346415 Objective Loss 0.346415 LR 0.001000 Time 0.020629 -2023-02-13 17:35:04,295 - Epoch: [37][ 490/ 1207] Overall Loss 0.346945 Objective Loss 0.346945 LR 0.001000 Time 0.020607 -2023-02-13 17:35:04,488 - Epoch: [37][ 500/ 1207] Overall Loss 0.346408 Objective Loss 0.346408 LR 0.001000 Time 0.020579 -2023-02-13 17:35:04,683 - Epoch: [37][ 510/ 1207] Overall Loss 0.346543 Objective Loss 0.346543 LR 0.001000 Time 0.020558 -2023-02-13 17:35:04,875 - Epoch: [37][ 520/ 1207] Overall Loss 0.346765 Objective Loss 0.346765 LR 0.001000 Time 0.020531 -2023-02-13 17:35:05,071 - Epoch: [37][ 530/ 1207] Overall Loss 0.347017 Objective Loss 0.347017 LR 0.001000 Time 0.020512 -2023-02-13 17:35:05,263 - Epoch: [37][ 540/ 1207] Overall Loss 0.346955 Objective Loss 0.346955 LR 0.001000 Time 0.020488 -2023-02-13 17:35:05,459 - Epoch: [37][ 550/ 1207] Overall Loss 0.348059 Objective Loss 0.348059 LR 0.001000 Time 0.020471 -2023-02-13 17:35:05,651 - Epoch: [37][ 560/ 1207] Overall Loss 0.347949 Objective Loss 0.347949 LR 0.001000 Time 0.020448 -2023-02-13 17:35:05,848 - Epoch: [37][ 570/ 1207] Overall Loss 0.348111 Objective Loss 0.348111 LR 0.001000 Time 0.020433 -2023-02-13 17:35:06,039 - Epoch: [37][ 580/ 1207] Overall Loss 0.348292 Objective Loss 0.348292 LR 0.001000 Time 0.020409 -2023-02-13 17:35:06,235 - Epoch: [37][ 590/ 1207] Overall Loss 0.347992 Objective Loss 0.347992 LR 0.001000 Time 0.020395 -2023-02-13 17:35:06,427 - Epoch: [37][ 600/ 1207] Overall Loss 0.347775 Objective Loss 0.347775 LR 0.001000 Time 0.020375 -2023-02-13 17:35:06,622 - Epoch: [37][ 610/ 1207] Overall Loss 0.347872 Objective Loss 0.347872 LR 0.001000 Time 0.020360 -2023-02-13 17:35:06,814 - Epoch: [37][ 620/ 1207] Overall Loss 0.347536 Objective Loss 0.347536 LR 0.001000 Time 0.020341 -2023-02-13 17:35:07,009 - Epoch: [37][ 630/ 1207] Overall Loss 0.347343 Objective Loss 0.347343 LR 0.001000 Time 0.020326 -2023-02-13 17:35:07,201 - Epoch: [37][ 640/ 1207] Overall Loss 0.347512 Objective Loss 0.347512 LR 0.001000 Time 0.020308 -2023-02-13 17:35:07,396 - Epoch: [37][ 650/ 1207] Overall Loss 0.347866 Objective Loss 0.347866 LR 0.001000 Time 0.020296 -2023-02-13 17:35:07,589 - Epoch: [37][ 660/ 1207] Overall Loss 0.347800 Objective Loss 0.347800 LR 0.001000 Time 0.020279 -2023-02-13 17:35:07,783 - Epoch: [37][ 670/ 1207] Overall Loss 0.348109 Objective Loss 0.348109 LR 0.001000 Time 0.020267 -2023-02-13 17:35:07,976 - Epoch: [37][ 680/ 1207] Overall Loss 0.348338 Objective Loss 0.348338 LR 0.001000 Time 0.020251 -2023-02-13 17:35:08,172 - Epoch: [37][ 690/ 1207] Overall Loss 0.348495 Objective Loss 0.348495 LR 0.001000 Time 0.020241 -2023-02-13 17:35:08,364 - Epoch: [37][ 700/ 1207] Overall Loss 0.348628 Objective Loss 0.348628 LR 0.001000 Time 0.020226 -2023-02-13 17:35:08,559 - Epoch: [37][ 710/ 1207] Overall Loss 0.348837 Objective Loss 0.348837 LR 0.001000 Time 0.020215 -2023-02-13 17:35:08,751 - Epoch: [37][ 720/ 1207] Overall Loss 0.349020 Objective Loss 0.349020 LR 0.001000 Time 0.020201 -2023-02-13 17:35:08,946 - Epoch: [37][ 730/ 1207] Overall Loss 0.348826 Objective Loss 0.348826 LR 0.001000 Time 0.020191 -2023-02-13 17:35:09,137 - Epoch: [37][ 740/ 1207] Overall Loss 0.348862 Objective Loss 0.348862 LR 0.001000 Time 0.020176 -2023-02-13 17:35:09,334 - Epoch: [37][ 750/ 1207] Overall Loss 0.348895 Objective Loss 0.348895 LR 0.001000 Time 0.020169 -2023-02-13 17:35:09,526 - Epoch: [37][ 760/ 1207] Overall Loss 0.349252 Objective Loss 0.349252 LR 0.001000 Time 0.020156 -2023-02-13 17:35:09,722 - Epoch: [37][ 770/ 1207] Overall Loss 0.349290 Objective Loss 0.349290 LR 0.001000 Time 0.020148 -2023-02-13 17:35:09,913 - Epoch: [37][ 780/ 1207] Overall Loss 0.349413 Objective Loss 0.349413 LR 0.001000 Time 0.020134 -2023-02-13 17:35:10,109 - Epoch: [37][ 790/ 1207] Overall Loss 0.349655 Objective Loss 0.349655 LR 0.001000 Time 0.020126 -2023-02-13 17:35:10,302 - Epoch: [37][ 800/ 1207] Overall Loss 0.350129 Objective Loss 0.350129 LR 0.001000 Time 0.020116 -2023-02-13 17:35:10,498 - Epoch: [37][ 810/ 1207] Overall Loss 0.349819 Objective Loss 0.349819 LR 0.001000 Time 0.020109 -2023-02-13 17:35:10,690 - Epoch: [37][ 820/ 1207] Overall Loss 0.349985 Objective Loss 0.349985 LR 0.001000 Time 0.020097 -2023-02-13 17:35:10,886 - Epoch: [37][ 830/ 1207] Overall Loss 0.350006 Objective Loss 0.350006 LR 0.001000 Time 0.020091 -2023-02-13 17:35:11,078 - Epoch: [37][ 840/ 1207] Overall Loss 0.349578 Objective Loss 0.349578 LR 0.001000 Time 0.020080 -2023-02-13 17:35:11,274 - Epoch: [37][ 850/ 1207] Overall Loss 0.349776 Objective Loss 0.349776 LR 0.001000 Time 0.020074 -2023-02-13 17:35:11,466 - Epoch: [37][ 860/ 1207] Overall Loss 0.349878 Objective Loss 0.349878 LR 0.001000 Time 0.020063 -2023-02-13 17:35:11,661 - Epoch: [37][ 870/ 1207] Overall Loss 0.349588 Objective Loss 0.349588 LR 0.001000 Time 0.020057 -2023-02-13 17:35:11,854 - Epoch: [37][ 880/ 1207] Overall Loss 0.349413 Objective Loss 0.349413 LR 0.001000 Time 0.020047 -2023-02-13 17:35:12,049 - Epoch: [37][ 890/ 1207] Overall Loss 0.349065 Objective Loss 0.349065 LR 0.001000 Time 0.020040 -2023-02-13 17:35:12,240 - Epoch: [37][ 900/ 1207] Overall Loss 0.349054 Objective Loss 0.349054 LR 0.001000 Time 0.020030 -2023-02-13 17:35:12,436 - Epoch: [37][ 910/ 1207] Overall Loss 0.349093 Objective Loss 0.349093 LR 0.001000 Time 0.020025 -2023-02-13 17:35:12,627 - Epoch: [37][ 920/ 1207] Overall Loss 0.349408 Objective Loss 0.349408 LR 0.001000 Time 0.020015 -2023-02-13 17:35:12,822 - Epoch: [37][ 930/ 1207] Overall Loss 0.349721 Objective Loss 0.349721 LR 0.001000 Time 0.020009 -2023-02-13 17:35:13,014 - Epoch: [37][ 940/ 1207] Overall Loss 0.349562 Objective Loss 0.349562 LR 0.001000 Time 0.019999 -2023-02-13 17:35:13,209 - Epoch: [37][ 950/ 1207] Overall Loss 0.349821 Objective Loss 0.349821 LR 0.001000 Time 0.019994 -2023-02-13 17:35:13,402 - Epoch: [37][ 960/ 1207] Overall Loss 0.349813 Objective Loss 0.349813 LR 0.001000 Time 0.019986 -2023-02-13 17:35:13,597 - Epoch: [37][ 970/ 1207] Overall Loss 0.349665 Objective Loss 0.349665 LR 0.001000 Time 0.019981 -2023-02-13 17:35:13,789 - Epoch: [37][ 980/ 1207] Overall Loss 0.349506 Objective Loss 0.349506 LR 0.001000 Time 0.019973 -2023-02-13 17:35:13,985 - Epoch: [37][ 990/ 1207] Overall Loss 0.349285 Objective Loss 0.349285 LR 0.001000 Time 0.019969 -2023-02-13 17:35:14,178 - Epoch: [37][ 1000/ 1207] Overall Loss 0.349224 Objective Loss 0.349224 LR 0.001000 Time 0.019962 -2023-02-13 17:35:14,375 - Epoch: [37][ 1010/ 1207] Overall Loss 0.349014 Objective Loss 0.349014 LR 0.001000 Time 0.019958 -2023-02-13 17:35:14,567 - Epoch: [37][ 1020/ 1207] Overall Loss 0.348982 Objective Loss 0.348982 LR 0.001000 Time 0.019951 -2023-02-13 17:35:14,763 - Epoch: [37][ 1030/ 1207] Overall Loss 0.349265 Objective Loss 0.349265 LR 0.001000 Time 0.019947 -2023-02-13 17:35:14,956 - Epoch: [37][ 1040/ 1207] Overall Loss 0.349310 Objective Loss 0.349310 LR 0.001000 Time 0.019940 -2023-02-13 17:35:15,152 - Epoch: [37][ 1050/ 1207] Overall Loss 0.349324 Objective Loss 0.349324 LR 0.001000 Time 0.019936 -2023-02-13 17:35:15,345 - Epoch: [37][ 1060/ 1207] Overall Loss 0.349229 Objective Loss 0.349229 LR 0.001000 Time 0.019930 -2023-02-13 17:35:15,540 - Epoch: [37][ 1070/ 1207] Overall Loss 0.349337 Objective Loss 0.349337 LR 0.001000 Time 0.019926 -2023-02-13 17:35:15,732 - Epoch: [37][ 1080/ 1207] Overall Loss 0.349436 Objective Loss 0.349436 LR 0.001000 Time 0.019919 -2023-02-13 17:35:15,928 - Epoch: [37][ 1090/ 1207] Overall Loss 0.348998 Objective Loss 0.348998 LR 0.001000 Time 0.019915 -2023-02-13 17:35:16,120 - Epoch: [37][ 1100/ 1207] Overall Loss 0.348870 Objective Loss 0.348870 LR 0.001000 Time 0.019909 -2023-02-13 17:35:16,314 - Epoch: [37][ 1110/ 1207] Overall Loss 0.349010 Objective Loss 0.349010 LR 0.001000 Time 0.019904 -2023-02-13 17:35:16,507 - Epoch: [37][ 1120/ 1207] Overall Loss 0.349135 Objective Loss 0.349135 LR 0.001000 Time 0.019898 -2023-02-13 17:35:16,702 - Epoch: [37][ 1130/ 1207] Overall Loss 0.348994 Objective Loss 0.348994 LR 0.001000 Time 0.019894 -2023-02-13 17:35:16,896 - Epoch: [37][ 1140/ 1207] Overall Loss 0.348921 Objective Loss 0.348921 LR 0.001000 Time 0.019889 -2023-02-13 17:35:17,090 - Epoch: [37][ 1150/ 1207] Overall Loss 0.348905 Objective Loss 0.348905 LR 0.001000 Time 0.019885 -2023-02-13 17:35:17,283 - Epoch: [37][ 1160/ 1207] Overall Loss 0.348666 Objective Loss 0.348666 LR 0.001000 Time 0.019880 -2023-02-13 17:35:17,478 - Epoch: [37][ 1170/ 1207] Overall Loss 0.348924 Objective Loss 0.348924 LR 0.001000 Time 0.019876 -2023-02-13 17:35:17,671 - Epoch: [37][ 1180/ 1207] Overall Loss 0.348831 Objective Loss 0.348831 LR 0.001000 Time 0.019871 -2023-02-13 17:35:17,866 - Epoch: [37][ 1190/ 1207] Overall Loss 0.349250 Objective Loss 0.349250 LR 0.001000 Time 0.019867 -2023-02-13 17:35:18,115 - Epoch: [37][ 1200/ 1207] Overall Loss 0.349162 Objective Loss 0.349162 LR 0.001000 Time 0.019909 -2023-02-13 17:35:18,229 - Epoch: [37][ 1207/ 1207] Overall Loss 0.349166 Objective Loss 0.349166 Top1 83.841463 Top5 97.560976 LR 0.001000 Time 0.019888 -2023-02-13 17:35:18,301 - --- validate (epoch=37)----------- -2023-02-13 17:35:18,301 - 34311 samples (256 per mini-batch) -2023-02-13 17:35:18,808 - Epoch: [37][ 10/ 135] Loss 0.368735 Top1 81.445312 Top5 97.500000 -2023-02-13 17:35:18,932 - Epoch: [37][ 20/ 135] Loss 0.370743 Top1 82.011719 Top5 97.324219 -2023-02-13 17:35:19,061 - Epoch: [37][ 30/ 135] Loss 0.370748 Top1 82.252604 Top5 97.382812 -2023-02-13 17:35:19,186 - Epoch: [37][ 40/ 135] Loss 0.373754 Top1 82.363281 Top5 97.236328 -2023-02-13 17:35:19,313 - Epoch: [37][ 50/ 135] Loss 0.369509 Top1 82.335938 Top5 97.265625 -2023-02-13 17:35:19,441 - Epoch: [37][ 60/ 135] Loss 0.368183 Top1 82.115885 Top5 97.194010 -2023-02-13 17:35:19,567 - Epoch: [37][ 70/ 135] Loss 0.366693 Top1 82.232143 Top5 97.226562 -2023-02-13 17:35:19,695 - Epoch: [37][ 80/ 135] Loss 0.367090 Top1 82.119141 Top5 97.246094 -2023-02-13 17:35:19,825 - Epoch: [37][ 90/ 135] Loss 0.365723 Top1 82.391493 Top5 97.287326 -2023-02-13 17:35:19,957 - Epoch: [37][ 100/ 135] Loss 0.364931 Top1 82.375000 Top5 97.242188 -2023-02-13 17:35:20,090 - Epoch: [37][ 110/ 135] Loss 0.367198 Top1 82.219460 Top5 97.215909 -2023-02-13 17:35:20,215 - Epoch: [37][ 120/ 135] Loss 0.365818 Top1 82.242839 Top5 97.252604 -2023-02-13 17:35:20,346 - Epoch: [37][ 130/ 135] Loss 0.365576 Top1 82.244591 Top5 97.241587 -2023-02-13 17:35:20,394 - Epoch: [37][ 135/ 135] Loss 0.363303 Top1 82.253505 Top5 97.260354 -2023-02-13 17:35:20,461 - ==> Top1: 82.254 Top5: 97.260 Loss: 0.363 - -2023-02-13 17:35:20,462 - ==> Confusion: -[[ 869 5 10 4 11 4 0 2 3 23 0 4 0 4 7 2 6 2 1 1 9] - [ 3 896 1 6 8 47 8 25 4 3 2 2 2 2 1 0 3 2 8 0 10] - [ 5 2 946 21 4 3 17 13 0 1 3 2 1 0 6 6 3 2 10 3 10] - [ 6 3 8 897 2 2 3 1 1 2 12 0 9 0 25 2 1 6 25 1 10] - [ 24 6 1 0 987 8 0 4 1 4 2 2 2 2 7 7 5 0 0 1 3] - [ 3 14 0 6 8 942 5 29 2 3 2 6 3 16 4 2 2 2 4 9 8] - [ 3 4 21 5 1 4 1029 7 0 1 3 2 1 2 0 7 2 2 1 3 1] - [ 4 9 12 4 2 23 0 918 2 1 1 3 2 1 0 0 1 2 27 6 6] - [ 27 2 0 3 0 2 0 3 872 34 17 4 0 10 19 1 2 3 7 0 3] - [ 127 1 6 1 6 1 0 2 33 796 2 1 0 17 4 2 1 5 1 0 6] - [ 2 3 4 4 0 2 4 2 14 1 976 0 1 10 5 0 1 2 17 0 3] - [ 1 1 1 1 0 19 2 5 1 3 0 881 45 5 2 5 1 12 4 14 2] - [ 1 1 1 6 3 1 0 1 1 0 2 27 858 0 4 5 1 21 8 2 16] - [ 8 3 3 0 7 10 2 0 10 12 13 4 3 919 5 4 3 3 2 6 7] - [ 10 1 1 15 6 0 1 2 18 6 3 1 6 1 988 0 1 6 19 0 7] - [ 4 1 5 2 7 0 8 3 0 1 1 3 7 2 0 957 12 20 0 4 9] - [ 3 7 0 2 13 2 0 1 2 0 0 4 1 1 3 10 990 2 4 1 15] - [ 5 3 0 4 2 2 1 2 0 1 1 5 24 0 2 16 0 976 0 1 6] - [ 2 6 5 11 0 1 1 21 5 1 2 5 3 0 10 0 1 1 1008 1 2] - [ 1 3 2 1 1 9 12 26 0 1 3 17 6 3 0 8 7 3 2 1034 9] - [ 220 188 259 212 165 226 75 226 84 90 231 117 304 332 221 114 237 129 287 234 9483]] - -2023-02-13 17:35:20,463 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:35:20,463 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:35:20,470 - - -2023-02-13 17:35:20,470 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:35:21,354 - Epoch: [38][ 10/ 1207] Overall Loss 0.318240 Objective Loss 0.318240 LR 0.001000 Time 0.088314 -2023-02-13 17:35:21,561 - Epoch: [38][ 20/ 1207] Overall Loss 0.335587 Objective Loss 0.335587 LR 0.001000 Time 0.054474 -2023-02-13 17:35:21,767 - Epoch: [38][ 30/ 1207] Overall Loss 0.326471 Objective Loss 0.326471 LR 0.001000 Time 0.043174 -2023-02-13 17:35:21,974 - Epoch: [38][ 40/ 1207] Overall Loss 0.328689 Objective Loss 0.328689 LR 0.001000 Time 0.037561 -2023-02-13 17:35:22,180 - Epoch: [38][ 50/ 1207] Overall Loss 0.325054 Objective Loss 0.325054 LR 0.001000 Time 0.034162 -2023-02-13 17:35:22,388 - Epoch: [38][ 60/ 1207] Overall Loss 0.329812 Objective Loss 0.329812 LR 0.001000 Time 0.031926 -2023-02-13 17:35:22,595 - Epoch: [38][ 70/ 1207] Overall Loss 0.326094 Objective Loss 0.326094 LR 0.001000 Time 0.030311 -2023-02-13 17:35:22,803 - Epoch: [38][ 80/ 1207] Overall Loss 0.328010 Objective Loss 0.328010 LR 0.001000 Time 0.029124 -2023-02-13 17:35:23,010 - Epoch: [38][ 90/ 1207] Overall Loss 0.328131 Objective Loss 0.328131 LR 0.001000 Time 0.028175 -2023-02-13 17:35:23,218 - Epoch: [38][ 100/ 1207] Overall Loss 0.331040 Objective Loss 0.331040 LR 0.001000 Time 0.027442 -2023-02-13 17:35:23,425 - Epoch: [38][ 110/ 1207] Overall Loss 0.332613 Objective Loss 0.332613 LR 0.001000 Time 0.026826 -2023-02-13 17:35:23,634 - Epoch: [38][ 120/ 1207] Overall Loss 0.333968 Objective Loss 0.333968 LR 0.001000 Time 0.026324 -2023-02-13 17:35:23,841 - Epoch: [38][ 130/ 1207] Overall Loss 0.335760 Objective Loss 0.335760 LR 0.001000 Time 0.025888 -2023-02-13 17:35:24,049 - Epoch: [38][ 140/ 1207] Overall Loss 0.336145 Objective Loss 0.336145 LR 0.001000 Time 0.025521 -2023-02-13 17:35:24,255 - Epoch: [38][ 150/ 1207] Overall Loss 0.336640 Objective Loss 0.336640 LR 0.001000 Time 0.025191 -2023-02-13 17:35:24,464 - Epoch: [38][ 160/ 1207] Overall Loss 0.337070 Objective Loss 0.337070 LR 0.001000 Time 0.024921 -2023-02-13 17:35:24,670 - Epoch: [38][ 170/ 1207] Overall Loss 0.340790 Objective Loss 0.340790 LR 0.001000 Time 0.024667 -2023-02-13 17:35:24,878 - Epoch: [38][ 180/ 1207] Overall Loss 0.341484 Objective Loss 0.341484 LR 0.001000 Time 0.024450 -2023-02-13 17:35:25,085 - Epoch: [38][ 190/ 1207] Overall Loss 0.342766 Objective Loss 0.342766 LR 0.001000 Time 0.024252 -2023-02-13 17:35:25,294 - Epoch: [38][ 200/ 1207] Overall Loss 0.343302 Objective Loss 0.343302 LR 0.001000 Time 0.024078 -2023-02-13 17:35:25,501 - Epoch: [38][ 210/ 1207] Overall Loss 0.343562 Objective Loss 0.343562 LR 0.001000 Time 0.023914 -2023-02-13 17:35:25,708 - Epoch: [38][ 220/ 1207] Overall Loss 0.344489 Objective Loss 0.344489 LR 0.001000 Time 0.023767 -2023-02-13 17:35:25,904 - Epoch: [38][ 230/ 1207] Overall Loss 0.345059 Objective Loss 0.345059 LR 0.001000 Time 0.023587 -2023-02-13 17:35:26,105 - Epoch: [38][ 240/ 1207] Overall Loss 0.344601 Objective Loss 0.344601 LR 0.001000 Time 0.023440 -2023-02-13 17:35:26,302 - Epoch: [38][ 250/ 1207] Overall Loss 0.344950 Objective Loss 0.344950 LR 0.001000 Time 0.023290 -2023-02-13 17:35:26,505 - Epoch: [38][ 260/ 1207] Overall Loss 0.344242 Objective Loss 0.344242 LR 0.001000 Time 0.023171 -2023-02-13 17:35:26,702 - Epoch: [38][ 270/ 1207] Overall Loss 0.344394 Objective Loss 0.344394 LR 0.001000 Time 0.023040 -2023-02-13 17:35:26,903 - Epoch: [38][ 280/ 1207] Overall Loss 0.344975 Objective Loss 0.344975 LR 0.001000 Time 0.022934 -2023-02-13 17:35:27,099 - Epoch: [38][ 290/ 1207] Overall Loss 0.345321 Objective Loss 0.345321 LR 0.001000 Time 0.022820 -2023-02-13 17:35:27,301 - Epoch: [38][ 300/ 1207] Overall Loss 0.346423 Objective Loss 0.346423 LR 0.001000 Time 0.022729 -2023-02-13 17:35:27,499 - Epoch: [38][ 310/ 1207] Overall Loss 0.346863 Objective Loss 0.346863 LR 0.001000 Time 0.022634 -2023-02-13 17:35:27,700 - Epoch: [38][ 320/ 1207] Overall Loss 0.346884 Objective Loss 0.346884 LR 0.001000 Time 0.022554 -2023-02-13 17:35:27,898 - Epoch: [38][ 330/ 1207] Overall Loss 0.345072 Objective Loss 0.345072 LR 0.001000 Time 0.022468 -2023-02-13 17:35:28,099 - Epoch: [38][ 340/ 1207] Overall Loss 0.344819 Objective Loss 0.344819 LR 0.001000 Time 0.022397 -2023-02-13 17:35:28,295 - Epoch: [38][ 350/ 1207] Overall Loss 0.344695 Objective Loss 0.344695 LR 0.001000 Time 0.022318 -2023-02-13 17:35:28,497 - Epoch: [38][ 360/ 1207] Overall Loss 0.345647 Objective Loss 0.345647 LR 0.001000 Time 0.022258 -2023-02-13 17:35:28,694 - Epoch: [38][ 370/ 1207] Overall Loss 0.346837 Objective Loss 0.346837 LR 0.001000 Time 0.022188 -2023-02-13 17:35:28,896 - Epoch: [38][ 380/ 1207] Overall Loss 0.346422 Objective Loss 0.346422 LR 0.001000 Time 0.022133 -2023-02-13 17:35:29,092 - Epoch: [38][ 390/ 1207] Overall Loss 0.346105 Objective Loss 0.346105 LR 0.001000 Time 0.022068 -2023-02-13 17:35:29,294 - Epoch: [38][ 400/ 1207] Overall Loss 0.346200 Objective Loss 0.346200 LR 0.001000 Time 0.022020 -2023-02-13 17:35:29,491 - Epoch: [38][ 410/ 1207] Overall Loss 0.346191 Objective Loss 0.346191 LR 0.001000 Time 0.021964 -2023-02-13 17:35:29,693 - Epoch: [38][ 420/ 1207] Overall Loss 0.346955 Objective Loss 0.346955 LR 0.001000 Time 0.021920 -2023-02-13 17:35:29,890 - Epoch: [38][ 430/ 1207] Overall Loss 0.346723 Objective Loss 0.346723 LR 0.001000 Time 0.021866 -2023-02-13 17:35:30,090 - Epoch: [38][ 440/ 1207] Overall Loss 0.346861 Objective Loss 0.346861 LR 0.001000 Time 0.021825 -2023-02-13 17:35:30,287 - Epoch: [38][ 450/ 1207] Overall Loss 0.347231 Objective Loss 0.347231 LR 0.001000 Time 0.021776 -2023-02-13 17:35:30,490 - Epoch: [38][ 460/ 1207] Overall Loss 0.347431 Objective Loss 0.347431 LR 0.001000 Time 0.021743 -2023-02-13 17:35:30,687 - Epoch: [38][ 470/ 1207] Overall Loss 0.347607 Objective Loss 0.347607 LR 0.001000 Time 0.021699 -2023-02-13 17:35:30,889 - Epoch: [38][ 480/ 1207] Overall Loss 0.348076 Objective Loss 0.348076 LR 0.001000 Time 0.021667 -2023-02-13 17:35:31,087 - Epoch: [38][ 490/ 1207] Overall Loss 0.348034 Objective Loss 0.348034 LR 0.001000 Time 0.021626 -2023-02-13 17:35:31,288 - Epoch: [38][ 500/ 1207] Overall Loss 0.347920 Objective Loss 0.347920 LR 0.001000 Time 0.021596 -2023-02-13 17:35:31,486 - Epoch: [38][ 510/ 1207] Overall Loss 0.347862 Objective Loss 0.347862 LR 0.001000 Time 0.021561 -2023-02-13 17:35:31,688 - Epoch: [38][ 520/ 1207] Overall Loss 0.348484 Objective Loss 0.348484 LR 0.001000 Time 0.021534 -2023-02-13 17:35:31,886 - Epoch: [38][ 530/ 1207] Overall Loss 0.348245 Objective Loss 0.348245 LR 0.001000 Time 0.021500 -2023-02-13 17:35:32,087 - Epoch: [38][ 540/ 1207] Overall Loss 0.348061 Objective Loss 0.348061 LR 0.001000 Time 0.021474 -2023-02-13 17:35:32,284 - Epoch: [38][ 550/ 1207] Overall Loss 0.347821 Objective Loss 0.347821 LR 0.001000 Time 0.021441 -2023-02-13 17:35:32,486 - Epoch: [38][ 560/ 1207] Overall Loss 0.347703 Objective Loss 0.347703 LR 0.001000 Time 0.021417 -2023-02-13 17:35:32,683 - Epoch: [38][ 570/ 1207] Overall Loss 0.347531 Objective Loss 0.347531 LR 0.001000 Time 0.021386 -2023-02-13 17:35:32,884 - Epoch: [38][ 580/ 1207] Overall Loss 0.347741 Objective Loss 0.347741 LR 0.001000 Time 0.021364 -2023-02-13 17:35:33,081 - Epoch: [38][ 590/ 1207] Overall Loss 0.347805 Objective Loss 0.347805 LR 0.001000 Time 0.021335 -2023-02-13 17:35:33,282 - Epoch: [38][ 600/ 1207] Overall Loss 0.347918 Objective Loss 0.347918 LR 0.001000 Time 0.021314 -2023-02-13 17:35:33,480 - Epoch: [38][ 610/ 1207] Overall Loss 0.347596 Objective Loss 0.347596 LR 0.001000 Time 0.021288 -2023-02-13 17:35:33,682 - Epoch: [38][ 620/ 1207] Overall Loss 0.347553 Objective Loss 0.347553 LR 0.001000 Time 0.021270 -2023-02-13 17:35:33,878 - Epoch: [38][ 630/ 1207] Overall Loss 0.347148 Objective Loss 0.347148 LR 0.001000 Time 0.021243 -2023-02-13 17:35:34,080 - Epoch: [38][ 640/ 1207] Overall Loss 0.347068 Objective Loss 0.347068 LR 0.001000 Time 0.021226 -2023-02-13 17:35:34,276 - Epoch: [38][ 650/ 1207] Overall Loss 0.347105 Objective Loss 0.347105 LR 0.001000 Time 0.021200 -2023-02-13 17:35:34,478 - Epoch: [38][ 660/ 1207] Overall Loss 0.347783 Objective Loss 0.347783 LR 0.001000 Time 0.021185 -2023-02-13 17:35:34,675 - Epoch: [38][ 670/ 1207] Overall Loss 0.348208 Objective Loss 0.348208 LR 0.001000 Time 0.021162 -2023-02-13 17:35:34,877 - Epoch: [38][ 680/ 1207] Overall Loss 0.348979 Objective Loss 0.348979 LR 0.001000 Time 0.021147 -2023-02-13 17:35:35,074 - Epoch: [38][ 690/ 1207] Overall Loss 0.348796 Objective Loss 0.348796 LR 0.001000 Time 0.021125 -2023-02-13 17:35:35,275 - Epoch: [38][ 700/ 1207] Overall Loss 0.348366 Objective Loss 0.348366 LR 0.001000 Time 0.021110 -2023-02-13 17:35:35,472 - Epoch: [38][ 710/ 1207] Overall Loss 0.348303 Objective Loss 0.348303 LR 0.001000 Time 0.021090 -2023-02-13 17:35:35,674 - Epoch: [38][ 720/ 1207] Overall Loss 0.348250 Objective Loss 0.348250 LR 0.001000 Time 0.021076 -2023-02-13 17:35:35,871 - Epoch: [38][ 730/ 1207] Overall Loss 0.348244 Objective Loss 0.348244 LR 0.001000 Time 0.021058 -2023-02-13 17:35:36,073 - Epoch: [38][ 740/ 1207] Overall Loss 0.348149 Objective Loss 0.348149 LR 0.001000 Time 0.021046 -2023-02-13 17:35:36,271 - Epoch: [38][ 750/ 1207] Overall Loss 0.348244 Objective Loss 0.348244 LR 0.001000 Time 0.021028 -2023-02-13 17:35:36,473 - Epoch: [38][ 760/ 1207] Overall Loss 0.348087 Objective Loss 0.348087 LR 0.001000 Time 0.021016 -2023-02-13 17:35:36,670 - Epoch: [38][ 770/ 1207] Overall Loss 0.348290 Objective Loss 0.348290 LR 0.001000 Time 0.020999 -2023-02-13 17:35:36,872 - Epoch: [38][ 780/ 1207] Overall Loss 0.348059 Objective Loss 0.348059 LR 0.001000 Time 0.020988 -2023-02-13 17:35:37,069 - Epoch: [38][ 790/ 1207] Overall Loss 0.348280 Objective Loss 0.348280 LR 0.001000 Time 0.020971 -2023-02-13 17:35:37,269 - Epoch: [38][ 800/ 1207] Overall Loss 0.348360 Objective Loss 0.348360 LR 0.001000 Time 0.020960 -2023-02-13 17:35:37,467 - Epoch: [38][ 810/ 1207] Overall Loss 0.348042 Objective Loss 0.348042 LR 0.001000 Time 0.020944 -2023-02-13 17:35:37,669 - Epoch: [38][ 820/ 1207] Overall Loss 0.348029 Objective Loss 0.348029 LR 0.001000 Time 0.020934 -2023-02-13 17:35:37,866 - Epoch: [38][ 830/ 1207] Overall Loss 0.347957 Objective Loss 0.347957 LR 0.001000 Time 0.020919 -2023-02-13 17:35:38,067 - Epoch: [38][ 840/ 1207] Overall Loss 0.347309 Objective Loss 0.347309 LR 0.001000 Time 0.020909 -2023-02-13 17:35:38,264 - Epoch: [38][ 850/ 1207] Overall Loss 0.347525 Objective Loss 0.347525 LR 0.001000 Time 0.020895 -2023-02-13 17:35:38,467 - Epoch: [38][ 860/ 1207] Overall Loss 0.347883 Objective Loss 0.347883 LR 0.001000 Time 0.020887 -2023-02-13 17:35:38,663 - Epoch: [38][ 870/ 1207] Overall Loss 0.347677 Objective Loss 0.347677 LR 0.001000 Time 0.020872 -2023-02-13 17:35:38,865 - Epoch: [38][ 880/ 1207] Overall Loss 0.347936 Objective Loss 0.347936 LR 0.001000 Time 0.020864 -2023-02-13 17:35:39,063 - Epoch: [38][ 890/ 1207] Overall Loss 0.347658 Objective Loss 0.347658 LR 0.001000 Time 0.020851 -2023-02-13 17:35:39,265 - Epoch: [38][ 900/ 1207] Overall Loss 0.347100 Objective Loss 0.347100 LR 0.001000 Time 0.020844 -2023-02-13 17:35:39,463 - Epoch: [38][ 910/ 1207] Overall Loss 0.347027 Objective Loss 0.347027 LR 0.001000 Time 0.020832 -2023-02-13 17:35:39,664 - Epoch: [38][ 920/ 1207] Overall Loss 0.347535 Objective Loss 0.347535 LR 0.001000 Time 0.020823 -2023-02-13 17:35:39,860 - Epoch: [38][ 930/ 1207] Overall Loss 0.347664 Objective Loss 0.347664 LR 0.001000 Time 0.020810 -2023-02-13 17:35:40,062 - Epoch: [38][ 940/ 1207] Overall Loss 0.347808 Objective Loss 0.347808 LR 0.001000 Time 0.020803 -2023-02-13 17:35:40,259 - Epoch: [38][ 950/ 1207] Overall Loss 0.347958 Objective Loss 0.347958 LR 0.001000 Time 0.020791 -2023-02-13 17:35:40,462 - Epoch: [38][ 960/ 1207] Overall Loss 0.348231 Objective Loss 0.348231 LR 0.001000 Time 0.020785 -2023-02-13 17:35:40,659 - Epoch: [38][ 970/ 1207] Overall Loss 0.348382 Objective Loss 0.348382 LR 0.001000 Time 0.020774 -2023-02-13 17:35:40,862 - Epoch: [38][ 980/ 1207] Overall Loss 0.348808 Objective Loss 0.348808 LR 0.001000 Time 0.020768 -2023-02-13 17:35:41,059 - Epoch: [38][ 990/ 1207] Overall Loss 0.348544 Objective Loss 0.348544 LR 0.001000 Time 0.020757 -2023-02-13 17:35:41,261 - Epoch: [38][ 1000/ 1207] Overall Loss 0.348579 Objective Loss 0.348579 LR 0.001000 Time 0.020752 -2023-02-13 17:35:41,459 - Epoch: [38][ 1010/ 1207] Overall Loss 0.348994 Objective Loss 0.348994 LR 0.001000 Time 0.020742 -2023-02-13 17:35:41,661 - Epoch: [38][ 1020/ 1207] Overall Loss 0.349217 Objective Loss 0.349217 LR 0.001000 Time 0.020736 -2023-02-13 17:35:41,859 - Epoch: [38][ 1030/ 1207] Overall Loss 0.349051 Objective Loss 0.349051 LR 0.001000 Time 0.020726 -2023-02-13 17:35:42,061 - Epoch: [38][ 1040/ 1207] Overall Loss 0.349059 Objective Loss 0.349059 LR 0.001000 Time 0.020721 -2023-02-13 17:35:42,258 - Epoch: [38][ 1050/ 1207] Overall Loss 0.349085 Objective Loss 0.349085 LR 0.001000 Time 0.020711 -2023-02-13 17:35:42,461 - Epoch: [38][ 1060/ 1207] Overall Loss 0.348976 Objective Loss 0.348976 LR 0.001000 Time 0.020707 -2023-02-13 17:35:42,658 - Epoch: [38][ 1070/ 1207] Overall Loss 0.349114 Objective Loss 0.349114 LR 0.001000 Time 0.020697 -2023-02-13 17:35:42,860 - Epoch: [38][ 1080/ 1207] Overall Loss 0.349268 Objective Loss 0.349268 LR 0.001000 Time 0.020692 -2023-02-13 17:35:43,057 - Epoch: [38][ 1090/ 1207] Overall Loss 0.349496 Objective Loss 0.349496 LR 0.001000 Time 0.020682 -2023-02-13 17:35:43,258 - Epoch: [38][ 1100/ 1207] Overall Loss 0.349583 Objective Loss 0.349583 LR 0.001000 Time 0.020677 -2023-02-13 17:35:43,456 - Epoch: [38][ 1110/ 1207] Overall Loss 0.349405 Objective Loss 0.349405 LR 0.001000 Time 0.020669 -2023-02-13 17:35:43,658 - Epoch: [38][ 1120/ 1207] Overall Loss 0.349186 Objective Loss 0.349186 LR 0.001000 Time 0.020664 -2023-02-13 17:35:43,854 - Epoch: [38][ 1130/ 1207] Overall Loss 0.349152 Objective Loss 0.349152 LR 0.001000 Time 0.020655 -2023-02-13 17:35:44,056 - Epoch: [38][ 1140/ 1207] Overall Loss 0.349198 Objective Loss 0.349198 LR 0.001000 Time 0.020650 -2023-02-13 17:35:44,253 - Epoch: [38][ 1150/ 1207] Overall Loss 0.349145 Objective Loss 0.349145 LR 0.001000 Time 0.020641 -2023-02-13 17:35:44,455 - Epoch: [38][ 1160/ 1207] Overall Loss 0.349070 Objective Loss 0.349070 LR 0.001000 Time 0.020637 -2023-02-13 17:35:44,652 - Epoch: [38][ 1170/ 1207] Overall Loss 0.349029 Objective Loss 0.349029 LR 0.001000 Time 0.020629 -2023-02-13 17:35:44,854 - Epoch: [38][ 1180/ 1207] Overall Loss 0.348961 Objective Loss 0.348961 LR 0.001000 Time 0.020625 -2023-02-13 17:35:45,051 - Epoch: [38][ 1190/ 1207] Overall Loss 0.348789 Objective Loss 0.348789 LR 0.001000 Time 0.020617 -2023-02-13 17:35:45,304 - Epoch: [38][ 1200/ 1207] Overall Loss 0.348893 Objective Loss 0.348893 LR 0.001000 Time 0.020656 -2023-02-13 17:35:45,419 - Epoch: [38][ 1207/ 1207] Overall Loss 0.348849 Objective Loss 0.348849 Top1 82.317073 Top5 97.560976 LR 0.001000 Time 0.020631 -2023-02-13 17:35:45,490 - --- validate (epoch=38)----------- -2023-02-13 17:35:45,491 - 34311 samples (256 per mini-batch) -2023-02-13 17:35:45,882 - Epoch: [38][ 10/ 135] Loss 0.389630 Top1 82.343750 Top5 97.265625 -2023-02-13 17:35:46,006 - Epoch: [38][ 20/ 135] Loss 0.385350 Top1 81.933594 Top5 97.187500 -2023-02-13 17:35:46,132 - Epoch: [38][ 30/ 135] Loss 0.371011 Top1 81.601562 Top5 97.252604 -2023-02-13 17:35:46,256 - Epoch: [38][ 40/ 135] Loss 0.373548 Top1 81.103516 Top5 97.128906 -2023-02-13 17:35:46,380 - Epoch: [38][ 50/ 135] Loss 0.371453 Top1 81.140625 Top5 97.148438 -2023-02-13 17:35:46,504 - Epoch: [38][ 60/ 135] Loss 0.367910 Top1 81.184896 Top5 97.031250 -2023-02-13 17:35:46,629 - Epoch: [38][ 70/ 135] Loss 0.374792 Top1 81.160714 Top5 97.003348 -2023-02-13 17:35:46,754 - Epoch: [38][ 80/ 135] Loss 0.380602 Top1 80.991211 Top5 97.001953 -2023-02-13 17:35:46,880 - Epoch: [38][ 90/ 135] Loss 0.378851 Top1 81.054688 Top5 96.940104 -2023-02-13 17:35:47,004 - Epoch: [38][ 100/ 135] Loss 0.380059 Top1 81.113281 Top5 96.933594 -2023-02-13 17:35:47,129 - Epoch: [38][ 110/ 135] Loss 0.377482 Top1 81.168324 Top5 96.931818 -2023-02-13 17:35:47,253 - Epoch: [38][ 120/ 135] Loss 0.374852 Top1 81.207682 Top5 96.930339 -2023-02-13 17:35:47,379 - Epoch: [38][ 130/ 135] Loss 0.373944 Top1 81.210938 Top5 96.947115 -2023-02-13 17:35:47,423 - Epoch: [38][ 135/ 135] Loss 0.373463 Top1 81.183877 Top5 96.933928 -2023-02-13 17:35:47,491 - ==> Top1: 81.184 Top5: 96.934 Loss: 0.373 - -2023-02-13 17:35:47,492 - ==> Confusion: -[[ 819 4 5 1 11 1 0 3 6 88 0 6 2 2 5 4 2 0 0 0 8] - [ 6 915 0 1 7 39 6 16 2 1 5 9 3 0 1 5 2 0 4 3 8] - [ 10 3 943 10 7 3 28 12 0 3 2 2 2 2 1 6 1 7 6 2 8] - [ 7 1 27 888 2 5 3 1 2 2 17 1 7 2 19 2 3 8 9 0 10] - [ 18 9 0 0 984 6 2 2 2 8 1 11 2 2 7 5 2 2 0 0 3] - [ 1 18 1 6 8 964 4 10 1 2 3 17 6 10 2 3 1 3 1 4 5] - [ 3 3 15 4 0 3 1030 7 0 0 3 4 1 1 0 5 1 2 2 12 3] - [ 1 14 9 1 3 44 6 885 2 1 5 13 5 0 1 0 2 2 18 7 5] - [ 24 6 1 2 1 1 0 0 887 48 6 2 2 7 12 2 1 2 3 0 2] - [ 58 2 5 0 4 1 0 1 39 876 0 3 0 8 3 2 1 3 1 0 5] - [ 3 2 5 1 3 1 5 2 23 4 963 3 2 8 3 1 2 0 14 2 4] - [ 1 2 1 0 3 10 0 2 1 3 0 919 26 2 1 5 1 12 1 13 2] - [ 1 0 2 5 1 5 1 1 1 0 0 49 836 0 3 10 0 33 2 3 6] - [ 8 2 1 0 6 19 1 1 23 30 11 21 6 872 3 5 3 1 0 8 3] - [ 14 3 3 14 7 1 0 0 27 7 8 2 2 2 970 3 2 8 6 0 13] - [ 6 5 6 1 12 3 7 1 0 0 1 11 8 1 0 944 5 20 1 11 3] - [ 4 4 1 0 14 4 1 2 4 1 1 13 3 1 1 17 971 2 1 6 10] - [ 5 0 2 5 4 1 1 0 0 2 1 17 7 0 0 10 0 993 0 2 1] - [ 5 4 3 18 1 1 1 30 9 1 10 6 9 0 21 1 0 1 956 4 5] - [ 1 3 4 0 2 10 9 17 0 0 0 30 4 4 0 5 2 2 0 1049 6] - [ 257 261 294 140 198 271 76 161 153 146 222 233 372 359 161 124 164 159 168 324 9191]] - -2023-02-13 17:35:47,493 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:35:47,493 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:35:47,499 - - -2023-02-13 17:35:47,499 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:35:48,456 - Epoch: [39][ 10/ 1207] Overall Loss 0.322769 Objective Loss 0.322769 LR 0.001000 Time 0.095663 -2023-02-13 17:35:48,651 - Epoch: [39][ 20/ 1207] Overall Loss 0.316000 Objective Loss 0.316000 LR 0.001000 Time 0.057577 -2023-02-13 17:35:48,838 - Epoch: [39][ 30/ 1207] Overall Loss 0.321747 Objective Loss 0.321747 LR 0.001000 Time 0.044605 -2023-02-13 17:35:49,026 - Epoch: [39][ 40/ 1207] Overall Loss 0.330467 Objective Loss 0.330467 LR 0.001000 Time 0.038140 -2023-02-13 17:35:49,214 - Epoch: [39][ 50/ 1207] Overall Loss 0.334856 Objective Loss 0.334856 LR 0.001000 Time 0.034255 -2023-02-13 17:35:49,402 - Epoch: [39][ 60/ 1207] Overall Loss 0.330062 Objective Loss 0.330062 LR 0.001000 Time 0.031673 -2023-02-13 17:35:49,590 - Epoch: [39][ 70/ 1207] Overall Loss 0.327844 Objective Loss 0.327844 LR 0.001000 Time 0.029832 -2023-02-13 17:35:49,777 - Epoch: [39][ 80/ 1207] Overall Loss 0.330446 Objective Loss 0.330446 LR 0.001000 Time 0.028444 -2023-02-13 17:35:49,965 - Epoch: [39][ 90/ 1207] Overall Loss 0.332740 Objective Loss 0.332740 LR 0.001000 Time 0.027359 -2023-02-13 17:35:50,152 - Epoch: [39][ 100/ 1207] Overall Loss 0.334006 Objective Loss 0.334006 LR 0.001000 Time 0.026498 -2023-02-13 17:35:50,340 - Epoch: [39][ 110/ 1207] Overall Loss 0.336497 Objective Loss 0.336497 LR 0.001000 Time 0.025790 -2023-02-13 17:35:50,528 - Epoch: [39][ 120/ 1207] Overall Loss 0.337317 Objective Loss 0.337317 LR 0.001000 Time 0.025208 -2023-02-13 17:35:50,716 - Epoch: [39][ 130/ 1207] Overall Loss 0.337841 Objective Loss 0.337841 LR 0.001000 Time 0.024711 -2023-02-13 17:35:50,904 - Epoch: [39][ 140/ 1207] Overall Loss 0.340470 Objective Loss 0.340470 LR 0.001000 Time 0.024287 -2023-02-13 17:35:51,092 - Epoch: [39][ 150/ 1207] Overall Loss 0.340736 Objective Loss 0.340736 LR 0.001000 Time 0.023913 -2023-02-13 17:35:51,279 - Epoch: [39][ 160/ 1207] Overall Loss 0.342087 Objective Loss 0.342087 LR 0.001000 Time 0.023590 -2023-02-13 17:35:51,468 - Epoch: [39][ 170/ 1207] Overall Loss 0.343248 Objective Loss 0.343248 LR 0.001000 Time 0.023308 -2023-02-13 17:35:51,655 - Epoch: [39][ 180/ 1207] Overall Loss 0.343032 Objective Loss 0.343032 LR 0.001000 Time 0.023053 -2023-02-13 17:35:51,843 - Epoch: [39][ 190/ 1207] Overall Loss 0.343405 Objective Loss 0.343405 LR 0.001000 Time 0.022826 -2023-02-13 17:35:52,030 - Epoch: [39][ 200/ 1207] Overall Loss 0.342337 Objective Loss 0.342337 LR 0.001000 Time 0.022619 -2023-02-13 17:35:52,217 - Epoch: [39][ 210/ 1207] Overall Loss 0.342584 Objective Loss 0.342584 LR 0.001000 Time 0.022432 -2023-02-13 17:35:52,406 - Epoch: [39][ 220/ 1207] Overall Loss 0.341965 Objective Loss 0.341965 LR 0.001000 Time 0.022266 -2023-02-13 17:35:52,593 - Epoch: [39][ 230/ 1207] Overall Loss 0.340283 Objective Loss 0.340283 LR 0.001000 Time 0.022113 -2023-02-13 17:35:52,781 - Epoch: [39][ 240/ 1207] Overall Loss 0.339272 Objective Loss 0.339272 LR 0.001000 Time 0.021972 -2023-02-13 17:35:52,968 - Epoch: [39][ 250/ 1207] Overall Loss 0.339232 Objective Loss 0.339232 LR 0.001000 Time 0.021841 -2023-02-13 17:35:53,156 - Epoch: [39][ 260/ 1207] Overall Loss 0.338311 Objective Loss 0.338311 LR 0.001000 Time 0.021720 -2023-02-13 17:35:53,343 - Epoch: [39][ 270/ 1207] Overall Loss 0.339536 Objective Loss 0.339536 LR 0.001000 Time 0.021609 -2023-02-13 17:35:53,532 - Epoch: [39][ 280/ 1207] Overall Loss 0.339953 Objective Loss 0.339953 LR 0.001000 Time 0.021510 -2023-02-13 17:35:53,720 - Epoch: [39][ 290/ 1207] Overall Loss 0.340161 Objective Loss 0.340161 LR 0.001000 Time 0.021414 -2023-02-13 17:35:53,907 - Epoch: [39][ 300/ 1207] Overall Loss 0.339879 Objective Loss 0.339879 LR 0.001000 Time 0.021324 -2023-02-13 17:35:54,094 - Epoch: [39][ 310/ 1207] Overall Loss 0.339428 Objective Loss 0.339428 LR 0.001000 Time 0.021238 -2023-02-13 17:35:54,282 - Epoch: [39][ 320/ 1207] Overall Loss 0.338996 Objective Loss 0.338996 LR 0.001000 Time 0.021159 -2023-02-13 17:35:54,470 - Epoch: [39][ 330/ 1207] Overall Loss 0.339203 Objective Loss 0.339203 LR 0.001000 Time 0.021086 -2023-02-13 17:35:54,658 - Epoch: [39][ 340/ 1207] Overall Loss 0.338676 Objective Loss 0.338676 LR 0.001000 Time 0.021018 -2023-02-13 17:35:54,845 - Epoch: [39][ 350/ 1207] Overall Loss 0.338937 Objective Loss 0.338937 LR 0.001000 Time 0.020952 -2023-02-13 17:35:55,033 - Epoch: [39][ 360/ 1207] Overall Loss 0.340108 Objective Loss 0.340108 LR 0.001000 Time 0.020890 -2023-02-13 17:35:55,221 - Epoch: [39][ 370/ 1207] Overall Loss 0.340675 Objective Loss 0.340675 LR 0.001000 Time 0.020833 -2023-02-13 17:35:55,409 - Epoch: [39][ 380/ 1207] Overall Loss 0.340254 Objective Loss 0.340254 LR 0.001000 Time 0.020778 -2023-02-13 17:35:55,596 - Epoch: [39][ 390/ 1207] Overall Loss 0.341602 Objective Loss 0.341602 LR 0.001000 Time 0.020726 -2023-02-13 17:35:55,785 - Epoch: [39][ 400/ 1207] Overall Loss 0.340198 Objective Loss 0.340198 LR 0.001000 Time 0.020678 -2023-02-13 17:35:55,973 - Epoch: [39][ 410/ 1207] Overall Loss 0.340265 Objective Loss 0.340265 LR 0.001000 Time 0.020633 -2023-02-13 17:35:56,162 - Epoch: [39][ 420/ 1207] Overall Loss 0.340249 Objective Loss 0.340249 LR 0.001000 Time 0.020589 -2023-02-13 17:35:56,349 - Epoch: [39][ 430/ 1207] Overall Loss 0.340310 Objective Loss 0.340310 LR 0.001000 Time 0.020545 -2023-02-13 17:35:56,538 - Epoch: [39][ 440/ 1207] Overall Loss 0.340192 Objective Loss 0.340192 LR 0.001000 Time 0.020506 -2023-02-13 17:35:56,726 - Epoch: [39][ 450/ 1207] Overall Loss 0.339825 Objective Loss 0.339825 LR 0.001000 Time 0.020467 -2023-02-13 17:35:56,915 - Epoch: [39][ 460/ 1207] Overall Loss 0.340730 Objective Loss 0.340730 LR 0.001000 Time 0.020432 -2023-02-13 17:35:57,103 - Epoch: [39][ 470/ 1207] Overall Loss 0.340151 Objective Loss 0.340151 LR 0.001000 Time 0.020397 -2023-02-13 17:35:57,290 - Epoch: [39][ 480/ 1207] Overall Loss 0.340872 Objective Loss 0.340872 LR 0.001000 Time 0.020362 -2023-02-13 17:35:57,479 - Epoch: [39][ 490/ 1207] Overall Loss 0.340991 Objective Loss 0.340991 LR 0.001000 Time 0.020331 -2023-02-13 17:35:57,667 - Epoch: [39][ 500/ 1207] Overall Loss 0.341280 Objective Loss 0.341280 LR 0.001000 Time 0.020300 -2023-02-13 17:35:57,855 - Epoch: [39][ 510/ 1207] Overall Loss 0.340654 Objective Loss 0.340654 LR 0.001000 Time 0.020270 -2023-02-13 17:35:58,043 - Epoch: [39][ 520/ 1207] Overall Loss 0.340874 Objective Loss 0.340874 LR 0.001000 Time 0.020240 -2023-02-13 17:35:58,230 - Epoch: [39][ 530/ 1207] Overall Loss 0.341772 Objective Loss 0.341772 LR 0.001000 Time 0.020211 -2023-02-13 17:35:58,419 - Epoch: [39][ 540/ 1207] Overall Loss 0.341958 Objective Loss 0.341958 LR 0.001000 Time 0.020185 -2023-02-13 17:35:58,607 - Epoch: [39][ 550/ 1207] Overall Loss 0.342823 Objective Loss 0.342823 LR 0.001000 Time 0.020161 -2023-02-13 17:35:58,796 - Epoch: [39][ 560/ 1207] Overall Loss 0.342587 Objective Loss 0.342587 LR 0.001000 Time 0.020137 -2023-02-13 17:35:58,984 - Epoch: [39][ 570/ 1207] Overall Loss 0.342943 Objective Loss 0.342943 LR 0.001000 Time 0.020113 -2023-02-13 17:35:59,172 - Epoch: [39][ 580/ 1207] Overall Loss 0.342766 Objective Loss 0.342766 LR 0.001000 Time 0.020090 -2023-02-13 17:35:59,360 - Epoch: [39][ 590/ 1207] Overall Loss 0.342928 Objective Loss 0.342928 LR 0.001000 Time 0.020066 -2023-02-13 17:35:59,548 - Epoch: [39][ 600/ 1207] Overall Loss 0.343004 Objective Loss 0.343004 LR 0.001000 Time 0.020045 -2023-02-13 17:35:59,736 - Epoch: [39][ 610/ 1207] Overall Loss 0.342515 Objective Loss 0.342515 LR 0.001000 Time 0.020025 -2023-02-13 17:35:59,925 - Epoch: [39][ 620/ 1207] Overall Loss 0.342576 Objective Loss 0.342576 LR 0.001000 Time 0.020005 -2023-02-13 17:36:00,113 - Epoch: [39][ 630/ 1207] Overall Loss 0.342565 Objective Loss 0.342565 LR 0.001000 Time 0.019985 -2023-02-13 17:36:00,300 - Epoch: [39][ 640/ 1207] Overall Loss 0.342629 Objective Loss 0.342629 LR 0.001000 Time 0.019966 -2023-02-13 17:36:00,488 - Epoch: [39][ 650/ 1207] Overall Loss 0.342813 Objective Loss 0.342813 LR 0.001000 Time 0.019947 -2023-02-13 17:36:00,676 - Epoch: [39][ 660/ 1207] Overall Loss 0.342976 Objective Loss 0.342976 LR 0.001000 Time 0.019929 -2023-02-13 17:36:00,865 - Epoch: [39][ 670/ 1207] Overall Loss 0.342779 Objective Loss 0.342779 LR 0.001000 Time 0.019913 -2023-02-13 17:36:01,052 - Epoch: [39][ 680/ 1207] Overall Loss 0.342705 Objective Loss 0.342705 LR 0.001000 Time 0.019895 -2023-02-13 17:36:01,241 - Epoch: [39][ 690/ 1207] Overall Loss 0.342946 Objective Loss 0.342946 LR 0.001000 Time 0.019879 -2023-02-13 17:36:01,429 - Epoch: [39][ 700/ 1207] Overall Loss 0.342776 Objective Loss 0.342776 LR 0.001000 Time 0.019864 -2023-02-13 17:36:01,618 - Epoch: [39][ 710/ 1207] Overall Loss 0.342579 Objective Loss 0.342579 LR 0.001000 Time 0.019849 -2023-02-13 17:36:01,807 - Epoch: [39][ 720/ 1207] Overall Loss 0.342283 Objective Loss 0.342283 LR 0.001000 Time 0.019836 -2023-02-13 17:36:01,995 - Epoch: [39][ 730/ 1207] Overall Loss 0.342383 Objective Loss 0.342383 LR 0.001000 Time 0.019821 -2023-02-13 17:36:02,183 - Epoch: [39][ 740/ 1207] Overall Loss 0.342312 Objective Loss 0.342312 LR 0.001000 Time 0.019807 -2023-02-13 17:36:02,373 - Epoch: [39][ 750/ 1207] Overall Loss 0.342458 Objective Loss 0.342458 LR 0.001000 Time 0.019796 -2023-02-13 17:36:02,564 - Epoch: [39][ 760/ 1207] Overall Loss 0.342915 Objective Loss 0.342915 LR 0.001000 Time 0.019786 -2023-02-13 17:36:02,753 - Epoch: [39][ 770/ 1207] Overall Loss 0.343055 Objective Loss 0.343055 LR 0.001000 Time 0.019774 -2023-02-13 17:36:02,944 - Epoch: [39][ 780/ 1207] Overall Loss 0.343071 Objective Loss 0.343071 LR 0.001000 Time 0.019765 -2023-02-13 17:36:03,134 - Epoch: [39][ 790/ 1207] Overall Loss 0.342939 Objective Loss 0.342939 LR 0.001000 Time 0.019754 -2023-02-13 17:36:03,324 - Epoch: [39][ 800/ 1207] Overall Loss 0.343292 Objective Loss 0.343292 LR 0.001000 Time 0.019745 -2023-02-13 17:36:03,513 - Epoch: [39][ 810/ 1207] Overall Loss 0.343577 Objective Loss 0.343577 LR 0.001000 Time 0.019734 -2023-02-13 17:36:03,704 - Epoch: [39][ 820/ 1207] Overall Loss 0.343954 Objective Loss 0.343954 LR 0.001000 Time 0.019725 -2023-02-13 17:36:03,894 - Epoch: [39][ 830/ 1207] Overall Loss 0.344357 Objective Loss 0.344357 LR 0.001000 Time 0.019716 -2023-02-13 17:36:04,084 - Epoch: [39][ 840/ 1207] Overall Loss 0.344159 Objective Loss 0.344159 LR 0.001000 Time 0.019708 -2023-02-13 17:36:04,273 - Epoch: [39][ 850/ 1207] Overall Loss 0.344289 Objective Loss 0.344289 LR 0.001000 Time 0.019698 -2023-02-13 17:36:04,464 - Epoch: [39][ 860/ 1207] Overall Loss 0.344612 Objective Loss 0.344612 LR 0.001000 Time 0.019690 -2023-02-13 17:36:04,654 - Epoch: [39][ 870/ 1207] Overall Loss 0.344346 Objective Loss 0.344346 LR 0.001000 Time 0.019682 -2023-02-13 17:36:04,845 - Epoch: [39][ 880/ 1207] Overall Loss 0.344231 Objective Loss 0.344231 LR 0.001000 Time 0.019675 -2023-02-13 17:36:05,035 - Epoch: [39][ 890/ 1207] Overall Loss 0.344453 Objective Loss 0.344453 LR 0.001000 Time 0.019667 -2023-02-13 17:36:05,226 - Epoch: [39][ 900/ 1207] Overall Loss 0.344778 Objective Loss 0.344778 LR 0.001000 Time 0.019660 -2023-02-13 17:36:05,416 - Epoch: [39][ 910/ 1207] Overall Loss 0.345010 Objective Loss 0.345010 LR 0.001000 Time 0.019652 -2023-02-13 17:36:05,607 - Epoch: [39][ 920/ 1207] Overall Loss 0.344949 Objective Loss 0.344949 LR 0.001000 Time 0.019646 -2023-02-13 17:36:05,797 - Epoch: [39][ 930/ 1207] Overall Loss 0.345376 Objective Loss 0.345376 LR 0.001000 Time 0.019639 -2023-02-13 17:36:05,988 - Epoch: [39][ 940/ 1207] Overall Loss 0.345571 Objective Loss 0.345571 LR 0.001000 Time 0.019632 -2023-02-13 17:36:06,177 - Epoch: [39][ 950/ 1207] Overall Loss 0.345781 Objective Loss 0.345781 LR 0.001000 Time 0.019625 -2023-02-13 17:36:06,368 - Epoch: [39][ 960/ 1207] Overall Loss 0.345998 Objective Loss 0.345998 LR 0.001000 Time 0.019618 -2023-02-13 17:36:06,557 - Epoch: [39][ 970/ 1207] Overall Loss 0.346223 Objective Loss 0.346223 LR 0.001000 Time 0.019612 -2023-02-13 17:36:06,748 - Epoch: [39][ 980/ 1207] Overall Loss 0.346312 Objective Loss 0.346312 LR 0.001000 Time 0.019605 -2023-02-13 17:36:06,939 - Epoch: [39][ 990/ 1207] Overall Loss 0.346313 Objective Loss 0.346313 LR 0.001000 Time 0.019600 -2023-02-13 17:36:07,129 - Epoch: [39][ 1000/ 1207] Overall Loss 0.346444 Objective Loss 0.346444 LR 0.001000 Time 0.019593 -2023-02-13 17:36:07,317 - Epoch: [39][ 1010/ 1207] Overall Loss 0.346116 Objective Loss 0.346116 LR 0.001000 Time 0.019586 -2023-02-13 17:36:07,508 - Epoch: [39][ 1020/ 1207] Overall Loss 0.346039 Objective Loss 0.346039 LR 0.001000 Time 0.019580 -2023-02-13 17:36:07,698 - Epoch: [39][ 1030/ 1207] Overall Loss 0.346023 Objective Loss 0.346023 LR 0.001000 Time 0.019574 -2023-02-13 17:36:07,888 - Epoch: [39][ 1040/ 1207] Overall Loss 0.345914 Objective Loss 0.345914 LR 0.001000 Time 0.019569 -2023-02-13 17:36:08,078 - Epoch: [39][ 1050/ 1207] Overall Loss 0.345793 Objective Loss 0.345793 LR 0.001000 Time 0.019562 -2023-02-13 17:36:08,268 - Epoch: [39][ 1060/ 1207] Overall Loss 0.345802 Objective Loss 0.345802 LR 0.001000 Time 0.019557 -2023-02-13 17:36:08,458 - Epoch: [39][ 1070/ 1207] Overall Loss 0.345806 Objective Loss 0.345806 LR 0.001000 Time 0.019552 -2023-02-13 17:36:08,649 - Epoch: [39][ 1080/ 1207] Overall Loss 0.346139 Objective Loss 0.346139 LR 0.001000 Time 0.019547 -2023-02-13 17:36:08,839 - Epoch: [39][ 1090/ 1207] Overall Loss 0.346025 Objective Loss 0.346025 LR 0.001000 Time 0.019542 -2023-02-13 17:36:09,029 - Epoch: [39][ 1100/ 1207] Overall Loss 0.346046 Objective Loss 0.346046 LR 0.001000 Time 0.019536 -2023-02-13 17:36:09,219 - Epoch: [39][ 1110/ 1207] Overall Loss 0.346188 Objective Loss 0.346188 LR 0.001000 Time 0.019531 -2023-02-13 17:36:09,409 - Epoch: [39][ 1120/ 1207] Overall Loss 0.346013 Objective Loss 0.346013 LR 0.001000 Time 0.019526 -2023-02-13 17:36:09,599 - Epoch: [39][ 1130/ 1207] Overall Loss 0.346036 Objective Loss 0.346036 LR 0.001000 Time 0.019521 -2023-02-13 17:36:09,790 - Epoch: [39][ 1140/ 1207] Overall Loss 0.345907 Objective Loss 0.345907 LR 0.001000 Time 0.019517 -2023-02-13 17:36:09,979 - Epoch: [39][ 1150/ 1207] Overall Loss 0.345998 Objective Loss 0.345998 LR 0.001000 Time 0.019512 -2023-02-13 17:36:10,170 - Epoch: [39][ 1160/ 1207] Overall Loss 0.346030 Objective Loss 0.346030 LR 0.001000 Time 0.019508 -2023-02-13 17:36:10,360 - Epoch: [39][ 1170/ 1207] Overall Loss 0.346140 Objective Loss 0.346140 LR 0.001000 Time 0.019503 -2023-02-13 17:36:10,551 - Epoch: [39][ 1180/ 1207] Overall Loss 0.346103 Objective Loss 0.346103 LR 0.001000 Time 0.019499 -2023-02-13 17:36:10,740 - Epoch: [39][ 1190/ 1207] Overall Loss 0.345865 Objective Loss 0.345865 LR 0.001000 Time 0.019494 -2023-02-13 17:36:10,981 - Epoch: [39][ 1200/ 1207] Overall Loss 0.346255 Objective Loss 0.346255 LR 0.001000 Time 0.019532 -2023-02-13 17:36:11,097 - Epoch: [39][ 1207/ 1207] Overall Loss 0.346502 Objective Loss 0.346502 Top1 82.012195 Top5 96.951220 LR 0.001000 Time 0.019514 -2023-02-13 17:36:11,168 - --- validate (epoch=39)----------- -2023-02-13 17:36:11,168 - 34311 samples (256 per mini-batch) -2023-02-13 17:36:11,564 - Epoch: [39][ 10/ 135] Loss 0.365855 Top1 81.132812 Top5 97.148438 -2023-02-13 17:36:11,693 - Epoch: [39][ 20/ 135] Loss 0.378810 Top1 81.015625 Top5 96.914062 -2023-02-13 17:36:11,821 - Epoch: [39][ 30/ 135] Loss 0.378035 Top1 80.611979 Top5 96.966146 -2023-02-13 17:36:11,949 - Epoch: [39][ 40/ 135] Loss 0.378769 Top1 80.683594 Top5 97.060547 -2023-02-13 17:36:12,077 - Epoch: [39][ 50/ 135] Loss 0.385462 Top1 80.570312 Top5 97.046875 -2023-02-13 17:36:12,204 - Epoch: [39][ 60/ 135] Loss 0.384659 Top1 80.625000 Top5 97.044271 -2023-02-13 17:36:12,332 - Epoch: [39][ 70/ 135] Loss 0.382591 Top1 80.563616 Top5 97.059152 -2023-02-13 17:36:12,458 - Epoch: [39][ 80/ 135] Loss 0.377359 Top1 80.659180 Top5 97.094727 -2023-02-13 17:36:12,584 - Epoch: [39][ 90/ 135] Loss 0.376276 Top1 80.750868 Top5 97.144097 -2023-02-13 17:36:12,710 - Epoch: [39][ 100/ 135] Loss 0.380655 Top1 80.781250 Top5 97.093750 -2023-02-13 17:36:12,836 - Epoch: [39][ 110/ 135] Loss 0.380173 Top1 80.873580 Top5 97.123580 -2023-02-13 17:36:12,964 - Epoch: [39][ 120/ 135] Loss 0.381164 Top1 80.895182 Top5 97.080078 -2023-02-13 17:36:13,095 - Epoch: [39][ 130/ 135] Loss 0.378007 Top1 80.895433 Top5 97.094351 -2023-02-13 17:36:13,142 - Epoch: [39][ 135/ 135] Loss 0.380525 Top1 80.944886 Top5 97.088397 -2023-02-13 17:36:13,210 - ==> Top1: 80.945 Top5: 97.088 Loss: 0.381 - -2023-02-13 17:36:13,211 - ==> Confusion: -[[ 827 4 11 2 7 4 0 1 2 74 0 2 1 4 4 5 4 4 1 1 9] - [ 3 915 3 3 7 49 4 18 2 1 2 0 2 1 2 3 3 0 7 1 7] - [ 5 4 945 11 7 2 26 14 0 1 5 0 1 2 3 6 2 4 6 6 8] - [ 6 1 24 886 2 7 1 1 2 2 15 1 8 1 22 2 3 7 17 0 8] - [ 14 10 2 0 982 14 2 2 1 4 0 5 0 3 8 7 3 1 0 4 4] - [ 2 17 4 8 4 960 4 18 1 3 2 6 5 12 1 4 2 2 3 7 5] - [ 4 5 17 4 1 4 1029 7 0 0 2 1 2 1 0 7 2 0 2 8 3] - [ 3 8 14 4 3 38 9 887 2 1 3 3 3 1 0 0 0 2 27 13 3] - [ 17 3 1 1 2 1 0 2 867 48 16 0 1 10 23 1 2 2 8 0 4] - [ 72 5 3 0 4 4 0 3 38 853 1 0 0 19 1 2 3 1 0 0 3] - [ 2 2 7 7 2 4 7 6 22 2 951 1 2 7 2 1 2 1 16 0 7] - [ 5 3 4 1 4 38 3 8 2 1 0 796 68 7 1 7 5 14 3 33 2] - [ 3 1 2 8 4 3 0 2 2 0 0 17 870 0 2 6 1 22 3 6 7] - [ 11 3 5 2 5 19 2 0 17 21 16 5 4 882 4 4 2 2 1 9 10] - [ 11 2 4 23 8 1 0 2 24 6 2 2 8 0 968 2 1 2 15 0 11] - [ 7 4 4 2 7 2 3 0 0 0 0 1 9 3 1 946 17 19 0 11 10] - [ 1 6 1 1 10 4 0 1 5 1 2 4 0 2 1 13 993 3 2 3 8] - [ 6 1 1 6 2 3 2 1 0 0 0 10 20 2 2 12 2 972 2 4 3] - [ 3 3 10 23 1 1 3 30 5 0 1 1 3 1 14 0 0 1 979 3 4] - [ 0 1 1 2 0 9 9 16 1 0 1 8 4 4 0 4 7 2 1 1074 4] - [ 186 270 300 156 154 296 111 187 87 117 173 81 368 308 154 148 367 125 216 439 9191]] - -2023-02-13 17:36:13,212 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:36:13,212 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:36:13,218 - - -2023-02-13 17:36:13,218 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:36:14,219 - Epoch: [40][ 10/ 1207] Overall Loss 0.328410 Objective Loss 0.328410 LR 0.001000 Time 0.100035 -2023-02-13 17:36:14,409 - Epoch: [40][ 20/ 1207] Overall Loss 0.350729 Objective Loss 0.350729 LR 0.001000 Time 0.059467 -2023-02-13 17:36:14,598 - Epoch: [40][ 30/ 1207] Overall Loss 0.347916 Objective Loss 0.347916 LR 0.001000 Time 0.045945 -2023-02-13 17:36:14,787 - Epoch: [40][ 40/ 1207] Overall Loss 0.334668 Objective Loss 0.334668 LR 0.001000 Time 0.039174 -2023-02-13 17:36:14,975 - Epoch: [40][ 50/ 1207] Overall Loss 0.336880 Objective Loss 0.336880 LR 0.001000 Time 0.035098 -2023-02-13 17:36:15,163 - Epoch: [40][ 60/ 1207] Overall Loss 0.329945 Objective Loss 0.329945 LR 0.001000 Time 0.032369 -2023-02-13 17:36:15,351 - Epoch: [40][ 70/ 1207] Overall Loss 0.332557 Objective Loss 0.332557 LR 0.001000 Time 0.030428 -2023-02-13 17:36:15,539 - Epoch: [40][ 80/ 1207] Overall Loss 0.329426 Objective Loss 0.329426 LR 0.001000 Time 0.028972 -2023-02-13 17:36:15,728 - Epoch: [40][ 90/ 1207] Overall Loss 0.330227 Objective Loss 0.330227 LR 0.001000 Time 0.027844 -2023-02-13 17:36:15,917 - Epoch: [40][ 100/ 1207] Overall Loss 0.333620 Objective Loss 0.333620 LR 0.001000 Time 0.026948 -2023-02-13 17:36:16,105 - Epoch: [40][ 110/ 1207] Overall Loss 0.332865 Objective Loss 0.332865 LR 0.001000 Time 0.026201 -2023-02-13 17:36:16,294 - Epoch: [40][ 120/ 1207] Overall Loss 0.333265 Objective Loss 0.333265 LR 0.001000 Time 0.025592 -2023-02-13 17:36:16,483 - Epoch: [40][ 130/ 1207] Overall Loss 0.334030 Objective Loss 0.334030 LR 0.001000 Time 0.025071 -2023-02-13 17:36:16,672 - Epoch: [40][ 140/ 1207] Overall Loss 0.333774 Objective Loss 0.333774 LR 0.001000 Time 0.024631 -2023-02-13 17:36:16,861 - Epoch: [40][ 150/ 1207] Overall Loss 0.335242 Objective Loss 0.335242 LR 0.001000 Time 0.024248 -2023-02-13 17:36:17,049 - Epoch: [40][ 160/ 1207] Overall Loss 0.336673 Objective Loss 0.336673 LR 0.001000 Time 0.023903 -2023-02-13 17:36:17,238 - Epoch: [40][ 170/ 1207] Overall Loss 0.335910 Objective Loss 0.335910 LR 0.001000 Time 0.023605 -2023-02-13 17:36:17,426 - Epoch: [40][ 180/ 1207] Overall Loss 0.336477 Objective Loss 0.336477 LR 0.001000 Time 0.023338 -2023-02-13 17:36:17,615 - Epoch: [40][ 190/ 1207] Overall Loss 0.336131 Objective Loss 0.336131 LR 0.001000 Time 0.023101 -2023-02-13 17:36:17,803 - Epoch: [40][ 200/ 1207] Overall Loss 0.335828 Objective Loss 0.335828 LR 0.001000 Time 0.022885 -2023-02-13 17:36:17,991 - Epoch: [40][ 210/ 1207] Overall Loss 0.334866 Objective Loss 0.334866 LR 0.001000 Time 0.022688 -2023-02-13 17:36:18,179 - Epoch: [40][ 220/ 1207] Overall Loss 0.334350 Objective Loss 0.334350 LR 0.001000 Time 0.022512 -2023-02-13 17:36:18,368 - Epoch: [40][ 230/ 1207] Overall Loss 0.334004 Objective Loss 0.334004 LR 0.001000 Time 0.022352 -2023-02-13 17:36:18,557 - Epoch: [40][ 240/ 1207] Overall Loss 0.333518 Objective Loss 0.333518 LR 0.001000 Time 0.022206 -2023-02-13 17:36:18,748 - Epoch: [40][ 250/ 1207] Overall Loss 0.333515 Objective Loss 0.333515 LR 0.001000 Time 0.022081 -2023-02-13 17:36:18,940 - Epoch: [40][ 260/ 1207] Overall Loss 0.334512 Objective Loss 0.334512 LR 0.001000 Time 0.021968 -2023-02-13 17:36:19,131 - Epoch: [40][ 270/ 1207] Overall Loss 0.334698 Objective Loss 0.334698 LR 0.001000 Time 0.021860 -2023-02-13 17:36:19,322 - Epoch: [40][ 280/ 1207] Overall Loss 0.334338 Objective Loss 0.334338 LR 0.001000 Time 0.021760 -2023-02-13 17:36:19,514 - Epoch: [40][ 290/ 1207] Overall Loss 0.334645 Objective Loss 0.334645 LR 0.001000 Time 0.021670 -2023-02-13 17:36:19,705 - Epoch: [40][ 300/ 1207] Overall Loss 0.334285 Objective Loss 0.334285 LR 0.001000 Time 0.021586 -2023-02-13 17:36:19,897 - Epoch: [40][ 310/ 1207] Overall Loss 0.334399 Objective Loss 0.334399 LR 0.001000 Time 0.021507 -2023-02-13 17:36:20,089 - Epoch: [40][ 320/ 1207] Overall Loss 0.335152 Objective Loss 0.335152 LR 0.001000 Time 0.021432 -2023-02-13 17:36:20,281 - Epoch: [40][ 330/ 1207] Overall Loss 0.334727 Objective Loss 0.334727 LR 0.001000 Time 0.021364 -2023-02-13 17:36:20,472 - Epoch: [40][ 340/ 1207] Overall Loss 0.335645 Objective Loss 0.335645 LR 0.001000 Time 0.021298 -2023-02-13 17:36:20,665 - Epoch: [40][ 350/ 1207] Overall Loss 0.336048 Objective Loss 0.336048 LR 0.001000 Time 0.021238 -2023-02-13 17:36:20,857 - Epoch: [40][ 360/ 1207] Overall Loss 0.335312 Objective Loss 0.335312 LR 0.001000 Time 0.021181 -2023-02-13 17:36:21,048 - Epoch: [40][ 370/ 1207] Overall Loss 0.335691 Objective Loss 0.335691 LR 0.001000 Time 0.021124 -2023-02-13 17:36:21,240 - Epoch: [40][ 380/ 1207] Overall Loss 0.335891 Objective Loss 0.335891 LR 0.001000 Time 0.021071 -2023-02-13 17:36:21,431 - Epoch: [40][ 390/ 1207] Overall Loss 0.336391 Objective Loss 0.336391 LR 0.001000 Time 0.021021 -2023-02-13 17:36:21,623 - Epoch: [40][ 400/ 1207] Overall Loss 0.336255 Objective Loss 0.336255 LR 0.001000 Time 0.020975 -2023-02-13 17:36:21,815 - Epoch: [40][ 410/ 1207] Overall Loss 0.335891 Objective Loss 0.335891 LR 0.001000 Time 0.020930 -2023-02-13 17:36:22,007 - Epoch: [40][ 420/ 1207] Overall Loss 0.335388 Objective Loss 0.335388 LR 0.001000 Time 0.020887 -2023-02-13 17:36:22,198 - Epoch: [40][ 430/ 1207] Overall Loss 0.336345 Objective Loss 0.336345 LR 0.001000 Time 0.020846 -2023-02-13 17:36:22,390 - Epoch: [40][ 440/ 1207] Overall Loss 0.336164 Objective Loss 0.336164 LR 0.001000 Time 0.020806 -2023-02-13 17:36:22,582 - Epoch: [40][ 450/ 1207] Overall Loss 0.336020 Objective Loss 0.336020 LR 0.001000 Time 0.020771 -2023-02-13 17:36:22,775 - Epoch: [40][ 460/ 1207] Overall Loss 0.335740 Objective Loss 0.335740 LR 0.001000 Time 0.020737 -2023-02-13 17:36:22,967 - Epoch: [40][ 470/ 1207] Overall Loss 0.335541 Objective Loss 0.335541 LR 0.001000 Time 0.020703 -2023-02-13 17:36:23,158 - Epoch: [40][ 480/ 1207] Overall Loss 0.335032 Objective Loss 0.335032 LR 0.001000 Time 0.020670 -2023-02-13 17:36:23,349 - Epoch: [40][ 490/ 1207] Overall Loss 0.335243 Objective Loss 0.335243 LR 0.001000 Time 0.020637 -2023-02-13 17:36:23,541 - Epoch: [40][ 500/ 1207] Overall Loss 0.335174 Objective Loss 0.335174 LR 0.001000 Time 0.020608 -2023-02-13 17:36:23,733 - Epoch: [40][ 510/ 1207] Overall Loss 0.334851 Objective Loss 0.334851 LR 0.001000 Time 0.020579 -2023-02-13 17:36:23,924 - Epoch: [40][ 520/ 1207] Overall Loss 0.335025 Objective Loss 0.335025 LR 0.001000 Time 0.020551 -2023-02-13 17:36:24,115 - Epoch: [40][ 530/ 1207] Overall Loss 0.334900 Objective Loss 0.334900 LR 0.001000 Time 0.020522 -2023-02-13 17:36:24,307 - Epoch: [40][ 540/ 1207] Overall Loss 0.335658 Objective Loss 0.335658 LR 0.001000 Time 0.020496 -2023-02-13 17:36:24,498 - Epoch: [40][ 550/ 1207] Overall Loss 0.335389 Objective Loss 0.335389 LR 0.001000 Time 0.020471 -2023-02-13 17:36:24,691 - Epoch: [40][ 560/ 1207] Overall Loss 0.335808 Objective Loss 0.335808 LR 0.001000 Time 0.020449 -2023-02-13 17:36:24,883 - Epoch: [40][ 570/ 1207] Overall Loss 0.336143 Objective Loss 0.336143 LR 0.001000 Time 0.020426 -2023-02-13 17:36:25,075 - Epoch: [40][ 580/ 1207] Overall Loss 0.336717 Objective Loss 0.336717 LR 0.001000 Time 0.020404 -2023-02-13 17:36:25,266 - Epoch: [40][ 590/ 1207] Overall Loss 0.337030 Objective Loss 0.337030 LR 0.001000 Time 0.020383 -2023-02-13 17:36:25,459 - Epoch: [40][ 600/ 1207] Overall Loss 0.336905 Objective Loss 0.336905 LR 0.001000 Time 0.020363 -2023-02-13 17:36:25,651 - Epoch: [40][ 610/ 1207] Overall Loss 0.337161 Objective Loss 0.337161 LR 0.001000 Time 0.020344 -2023-02-13 17:36:25,846 - Epoch: [40][ 620/ 1207] Overall Loss 0.337637 Objective Loss 0.337637 LR 0.001000 Time 0.020330 -2023-02-13 17:36:26,038 - Epoch: [40][ 630/ 1207] Overall Loss 0.338189 Objective Loss 0.338189 LR 0.001000 Time 0.020311 -2023-02-13 17:36:26,231 - Epoch: [40][ 640/ 1207] Overall Loss 0.338507 Objective Loss 0.338507 LR 0.001000 Time 0.020295 -2023-02-13 17:36:26,423 - Epoch: [40][ 650/ 1207] Overall Loss 0.338515 Objective Loss 0.338515 LR 0.001000 Time 0.020276 -2023-02-13 17:36:26,615 - Epoch: [40][ 660/ 1207] Overall Loss 0.338552 Objective Loss 0.338552 LR 0.001000 Time 0.020261 -2023-02-13 17:36:26,806 - Epoch: [40][ 670/ 1207] Overall Loss 0.338699 Objective Loss 0.338699 LR 0.001000 Time 0.020243 -2023-02-13 17:36:26,998 - Epoch: [40][ 680/ 1207] Overall Loss 0.338526 Objective Loss 0.338526 LR 0.001000 Time 0.020227 -2023-02-13 17:36:27,190 - Epoch: [40][ 690/ 1207] Overall Loss 0.338584 Objective Loss 0.338584 LR 0.001000 Time 0.020211 -2023-02-13 17:36:27,381 - Epoch: [40][ 700/ 1207] Overall Loss 0.338910 Objective Loss 0.338910 LR 0.001000 Time 0.020194 -2023-02-13 17:36:27,574 - Epoch: [40][ 710/ 1207] Overall Loss 0.339301 Objective Loss 0.339301 LR 0.001000 Time 0.020181 -2023-02-13 17:36:27,765 - Epoch: [40][ 720/ 1207] Overall Loss 0.339466 Objective Loss 0.339466 LR 0.001000 Time 0.020166 -2023-02-13 17:36:27,957 - Epoch: [40][ 730/ 1207] Overall Loss 0.339278 Objective Loss 0.339278 LR 0.001000 Time 0.020153 -2023-02-13 17:36:28,150 - Epoch: [40][ 740/ 1207] Overall Loss 0.339211 Objective Loss 0.339211 LR 0.001000 Time 0.020140 -2023-02-13 17:36:28,342 - Epoch: [40][ 750/ 1207] Overall Loss 0.339221 Objective Loss 0.339221 LR 0.001000 Time 0.020127 -2023-02-13 17:36:28,535 - Epoch: [40][ 760/ 1207] Overall Loss 0.339173 Objective Loss 0.339173 LR 0.001000 Time 0.020115 -2023-02-13 17:36:28,727 - Epoch: [40][ 770/ 1207] Overall Loss 0.338979 Objective Loss 0.338979 LR 0.001000 Time 0.020103 -2023-02-13 17:36:28,919 - Epoch: [40][ 780/ 1207] Overall Loss 0.339354 Objective Loss 0.339354 LR 0.001000 Time 0.020092 -2023-02-13 17:36:29,110 - Epoch: [40][ 790/ 1207] Overall Loss 0.339477 Objective Loss 0.339477 LR 0.001000 Time 0.020079 -2023-02-13 17:36:29,304 - Epoch: [40][ 800/ 1207] Overall Loss 0.339737 Objective Loss 0.339737 LR 0.001000 Time 0.020070 -2023-02-13 17:36:29,497 - Epoch: [40][ 810/ 1207] Overall Loss 0.339853 Objective Loss 0.339853 LR 0.001000 Time 0.020059 -2023-02-13 17:36:29,688 - Epoch: [40][ 820/ 1207] Overall Loss 0.339747 Objective Loss 0.339747 LR 0.001000 Time 0.020047 -2023-02-13 17:36:29,881 - Epoch: [40][ 830/ 1207] Overall Loss 0.339882 Objective Loss 0.339882 LR 0.001000 Time 0.020037 -2023-02-13 17:36:30,074 - Epoch: [40][ 840/ 1207] Overall Loss 0.340125 Objective Loss 0.340125 LR 0.001000 Time 0.020028 -2023-02-13 17:36:30,266 - Epoch: [40][ 850/ 1207] Overall Loss 0.340207 Objective Loss 0.340207 LR 0.001000 Time 0.020019 -2023-02-13 17:36:30,460 - Epoch: [40][ 860/ 1207] Overall Loss 0.340163 Objective Loss 0.340163 LR 0.001000 Time 0.020010 -2023-02-13 17:36:30,651 - Epoch: [40][ 870/ 1207] Overall Loss 0.340483 Objective Loss 0.340483 LR 0.001000 Time 0.020000 -2023-02-13 17:36:30,843 - Epoch: [40][ 880/ 1207] Overall Loss 0.340561 Objective Loss 0.340561 LR 0.001000 Time 0.019991 -2023-02-13 17:36:31,035 - Epoch: [40][ 890/ 1207] Overall Loss 0.340587 Objective Loss 0.340587 LR 0.001000 Time 0.019981 -2023-02-13 17:36:31,226 - Epoch: [40][ 900/ 1207] Overall Loss 0.340050 Objective Loss 0.340050 LR 0.001000 Time 0.019971 -2023-02-13 17:36:31,418 - Epoch: [40][ 910/ 1207] Overall Loss 0.340638 Objective Loss 0.340638 LR 0.001000 Time 0.019962 -2023-02-13 17:36:31,611 - Epoch: [40][ 920/ 1207] Overall Loss 0.340693 Objective Loss 0.340693 LR 0.001000 Time 0.019954 -2023-02-13 17:36:31,803 - Epoch: [40][ 930/ 1207] Overall Loss 0.340424 Objective Loss 0.340424 LR 0.001000 Time 0.019946 -2023-02-13 17:36:31,996 - Epoch: [40][ 940/ 1207] Overall Loss 0.340461 Objective Loss 0.340461 LR 0.001000 Time 0.019938 -2023-02-13 17:36:32,188 - Epoch: [40][ 950/ 1207] Overall Loss 0.340253 Objective Loss 0.340253 LR 0.001000 Time 0.019930 -2023-02-13 17:36:32,382 - Epoch: [40][ 960/ 1207] Overall Loss 0.340494 Objective Loss 0.340494 LR 0.001000 Time 0.019924 -2023-02-13 17:36:32,575 - Epoch: [40][ 970/ 1207] Overall Loss 0.340650 Objective Loss 0.340650 LR 0.001000 Time 0.019917 -2023-02-13 17:36:32,767 - Epoch: [40][ 980/ 1207] Overall Loss 0.340995 Objective Loss 0.340995 LR 0.001000 Time 0.019910 -2023-02-13 17:36:32,959 - Epoch: [40][ 990/ 1207] Overall Loss 0.341112 Objective Loss 0.341112 LR 0.001000 Time 0.019903 -2023-02-13 17:36:33,151 - Epoch: [40][ 1000/ 1207] Overall Loss 0.341302 Objective Loss 0.341302 LR 0.001000 Time 0.019895 -2023-02-13 17:36:33,342 - Epoch: [40][ 1010/ 1207] Overall Loss 0.341482 Objective Loss 0.341482 LR 0.001000 Time 0.019887 -2023-02-13 17:36:33,535 - Epoch: [40][ 1020/ 1207] Overall Loss 0.341587 Objective Loss 0.341587 LR 0.001000 Time 0.019881 -2023-02-13 17:36:33,727 - Epoch: [40][ 1030/ 1207] Overall Loss 0.341241 Objective Loss 0.341241 LR 0.001000 Time 0.019874 -2023-02-13 17:36:33,920 - Epoch: [40][ 1040/ 1207] Overall Loss 0.341130 Objective Loss 0.341130 LR 0.001000 Time 0.019868 -2023-02-13 17:36:34,111 - Epoch: [40][ 1050/ 1207] Overall Loss 0.341425 Objective Loss 0.341425 LR 0.001000 Time 0.019860 -2023-02-13 17:36:34,303 - Epoch: [40][ 1060/ 1207] Overall Loss 0.341426 Objective Loss 0.341426 LR 0.001000 Time 0.019854 -2023-02-13 17:36:34,495 - Epoch: [40][ 1070/ 1207] Overall Loss 0.341699 Objective Loss 0.341699 LR 0.001000 Time 0.019847 -2023-02-13 17:36:34,689 - Epoch: [40][ 1080/ 1207] Overall Loss 0.341801 Objective Loss 0.341801 LR 0.001000 Time 0.019842 -2023-02-13 17:36:34,881 - Epoch: [40][ 1090/ 1207] Overall Loss 0.341753 Objective Loss 0.341753 LR 0.001000 Time 0.019836 -2023-02-13 17:36:35,072 - Epoch: [40][ 1100/ 1207] Overall Loss 0.342080 Objective Loss 0.342080 LR 0.001000 Time 0.019829 -2023-02-13 17:36:35,265 - Epoch: [40][ 1110/ 1207] Overall Loss 0.341963 Objective Loss 0.341963 LR 0.001000 Time 0.019824 -2023-02-13 17:36:35,457 - Epoch: [40][ 1120/ 1207] Overall Loss 0.342172 Objective Loss 0.342172 LR 0.001000 Time 0.019819 -2023-02-13 17:36:35,650 - Epoch: [40][ 1130/ 1207] Overall Loss 0.342430 Objective Loss 0.342430 LR 0.001000 Time 0.019813 -2023-02-13 17:36:35,843 - Epoch: [40][ 1140/ 1207] Overall Loss 0.342491 Objective Loss 0.342491 LR 0.001000 Time 0.019809 -2023-02-13 17:36:36,035 - Epoch: [40][ 1150/ 1207] Overall Loss 0.342450 Objective Loss 0.342450 LR 0.001000 Time 0.019802 -2023-02-13 17:36:36,227 - Epoch: [40][ 1160/ 1207] Overall Loss 0.341976 Objective Loss 0.341976 LR 0.001000 Time 0.019797 -2023-02-13 17:36:36,418 - Epoch: [40][ 1170/ 1207] Overall Loss 0.341912 Objective Loss 0.341912 LR 0.001000 Time 0.019791 -2023-02-13 17:36:36,612 - Epoch: [40][ 1180/ 1207] Overall Loss 0.342153 Objective Loss 0.342153 LR 0.001000 Time 0.019787 -2023-02-13 17:36:36,805 - Epoch: [40][ 1190/ 1207] Overall Loss 0.342357 Objective Loss 0.342357 LR 0.001000 Time 0.019783 -2023-02-13 17:36:37,054 - Epoch: [40][ 1200/ 1207] Overall Loss 0.342498 Objective Loss 0.342498 LR 0.001000 Time 0.019825 -2023-02-13 17:36:37,170 - Epoch: [40][ 1207/ 1207] Overall Loss 0.342992 Objective Loss 0.342992 Top1 79.878049 Top5 95.121951 LR 0.001000 Time 0.019806 -2023-02-13 17:36:37,241 - --- validate (epoch=40)----------- -2023-02-13 17:36:37,241 - 34311 samples (256 per mini-batch) -2023-02-13 17:36:37,637 - Epoch: [40][ 10/ 135] Loss 0.365072 Top1 79.804688 Top5 96.210938 -2023-02-13 17:36:37,769 - Epoch: [40][ 20/ 135] Loss 0.385737 Top1 78.906250 Top5 96.386719 -2023-02-13 17:36:37,901 - Epoch: [40][ 30/ 135] Loss 0.373876 Top1 78.906250 Top5 96.510417 -2023-02-13 17:36:38,027 - Epoch: [40][ 40/ 135] Loss 0.368786 Top1 79.169922 Top5 96.523438 -2023-02-13 17:36:38,157 - Epoch: [40][ 50/ 135] Loss 0.368334 Top1 79.296875 Top5 96.554688 -2023-02-13 17:36:38,286 - Epoch: [40][ 60/ 135] Loss 0.372871 Top1 79.042969 Top5 96.425781 -2023-02-13 17:36:38,415 - Epoch: [40][ 70/ 135] Loss 0.370914 Top1 79.051339 Top5 96.456473 -2023-02-13 17:36:38,547 - Epoch: [40][ 80/ 135] Loss 0.371709 Top1 78.881836 Top5 96.489258 -2023-02-13 17:36:38,675 - Epoch: [40][ 90/ 135] Loss 0.372751 Top1 78.901910 Top5 96.510417 -2023-02-13 17:36:38,806 - Epoch: [40][ 100/ 135] Loss 0.376234 Top1 78.785156 Top5 96.464844 -2023-02-13 17:36:38,934 - Epoch: [40][ 110/ 135] Loss 0.377484 Top1 78.792614 Top5 96.487926 -2023-02-13 17:36:39,065 - Epoch: [40][ 120/ 135] Loss 0.375235 Top1 78.730469 Top5 96.529948 -2023-02-13 17:36:39,196 - Epoch: [40][ 130/ 135] Loss 0.377076 Top1 78.650841 Top5 96.565505 -2023-02-13 17:36:39,241 - Epoch: [40][ 135/ 135] Loss 0.374354 Top1 78.668649 Top5 96.572528 -2023-02-13 17:36:39,313 - ==> Top1: 78.669 Top5: 96.573 Loss: 0.374 - -2023-02-13 17:36:39,314 - ==> Confusion: -[[ 848 5 11 1 10 1 0 0 5 53 2 3 0 2 5 2 7 2 6 0 4] - [ 2 935 3 4 6 21 1 27 2 0 3 0 2 0 3 2 3 0 14 4 1] - [ 6 5 974 11 2 1 11 15 0 0 2 2 0 1 2 6 3 3 11 1 2] - [ 6 2 28 897 2 4 1 2 1 2 12 1 8 0 9 1 4 8 23 1 4] - [ 18 10 3 1 988 9 3 2 1 4 1 5 0 2 4 4 4 2 2 2 1] - [ 5 36 5 5 8 914 4 34 2 5 3 6 4 17 2 1 3 0 5 7 4] - [ 2 9 30 3 0 7 1024 6 1 0 2 0 1 0 0 4 0 2 2 5 1] - [ 2 9 20 2 3 24 1 893 1 1 1 3 2 1 1 0 0 3 43 12 2] - [ 22 1 1 2 1 0 0 2 882 48 16 1 0 3 14 2 0 2 12 0 0] - [ 83 2 5 0 10 3 0 1 37 849 1 0 2 7 3 3 0 2 1 1 2] - [ 4 1 13 8 2 3 4 3 18 1 959 0 0 10 3 0 0 0 20 1 1] - [ 4 5 4 0 2 17 2 12 1 2 1 877 18 8 2 1 3 13 4 29 0] - [ 0 1 4 7 1 2 0 6 5 0 1 33 843 0 4 8 3 29 3 4 5] - [ 8 2 7 0 5 10 0 2 25 22 12 3 5 893 6 3 7 3 2 5 4] - [ 17 2 7 34 10 3 0 3 27 5 2 2 1 1 940 1 0 8 21 0 8] - [ 4 2 10 3 8 4 4 1 1 0 0 5 14 3 0 946 10 15 1 11 4] - [ 3 10 3 1 10 4 0 0 2 3 0 6 2 1 1 8 981 2 6 11 7] - [ 5 4 0 5 2 0 2 1 1 2 1 7 19 1 1 9 1 979 2 6 3] - [ 3 5 9 9 2 1 1 20 3 0 2 1 5 0 8 0 0 0 1016 1 0] - [ 0 4 2 1 1 5 8 28 0 0 2 13 1 3 0 3 5 2 4 1062 4] - [ 243 392 528 253 181 194 140 219 134 150 218 129 420 340 180 140 329 155 361 436 8292]] - -2023-02-13 17:36:39,315 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:36:39,315 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:36:39,321 - - -2023-02-13 17:36:39,321 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:36:40,222 - Epoch: [41][ 10/ 1207] Overall Loss 0.349569 Objective Loss 0.349569 LR 0.001000 Time 0.090071 -2023-02-13 17:36:40,419 - Epoch: [41][ 20/ 1207] Overall Loss 0.349342 Objective Loss 0.349342 LR 0.001000 Time 0.054838 -2023-02-13 17:36:40,608 - Epoch: [41][ 30/ 1207] Overall Loss 0.351326 Objective Loss 0.351326 LR 0.001000 Time 0.042854 -2023-02-13 17:36:40,799 - Epoch: [41][ 40/ 1207] Overall Loss 0.343842 Objective Loss 0.343842 LR 0.001000 Time 0.036892 -2023-02-13 17:36:40,987 - Epoch: [41][ 50/ 1207] Overall Loss 0.338780 Objective Loss 0.338780 LR 0.001000 Time 0.033277 -2023-02-13 17:36:41,175 - Epoch: [41][ 60/ 1207] Overall Loss 0.334206 Objective Loss 0.334206 LR 0.001000 Time 0.030848 -2023-02-13 17:36:41,363 - Epoch: [41][ 70/ 1207] Overall Loss 0.334707 Objective Loss 0.334707 LR 0.001000 Time 0.029125 -2023-02-13 17:36:41,550 - Epoch: [41][ 80/ 1207] Overall Loss 0.337833 Objective Loss 0.337833 LR 0.001000 Time 0.027825 -2023-02-13 17:36:41,740 - Epoch: [41][ 90/ 1207] Overall Loss 0.339651 Objective Loss 0.339651 LR 0.001000 Time 0.026833 -2023-02-13 17:36:41,928 - Epoch: [41][ 100/ 1207] Overall Loss 0.334987 Objective Loss 0.334987 LR 0.001000 Time 0.026027 -2023-02-13 17:36:42,116 - Epoch: [41][ 110/ 1207] Overall Loss 0.337158 Objective Loss 0.337158 LR 0.001000 Time 0.025372 -2023-02-13 17:36:42,304 - Epoch: [41][ 120/ 1207] Overall Loss 0.337939 Objective Loss 0.337939 LR 0.001000 Time 0.024817 -2023-02-13 17:36:42,492 - Epoch: [41][ 130/ 1207] Overall Loss 0.336597 Objective Loss 0.336597 LR 0.001000 Time 0.024355 -2023-02-13 17:36:42,681 - Epoch: [41][ 140/ 1207] Overall Loss 0.337200 Objective Loss 0.337200 LR 0.001000 Time 0.023960 -2023-02-13 17:36:42,869 - Epoch: [41][ 150/ 1207] Overall Loss 0.337096 Objective Loss 0.337096 LR 0.001000 Time 0.023615 -2023-02-13 17:36:43,057 - Epoch: [41][ 160/ 1207] Overall Loss 0.336918 Objective Loss 0.336918 LR 0.001000 Time 0.023313 -2023-02-13 17:36:43,246 - Epoch: [41][ 170/ 1207] Overall Loss 0.337587 Objective Loss 0.337587 LR 0.001000 Time 0.023051 -2023-02-13 17:36:43,435 - Epoch: [41][ 180/ 1207] Overall Loss 0.338934 Objective Loss 0.338934 LR 0.001000 Time 0.022815 -2023-02-13 17:36:43,623 - Epoch: [41][ 190/ 1207] Overall Loss 0.338482 Objective Loss 0.338482 LR 0.001000 Time 0.022603 -2023-02-13 17:36:43,812 - Epoch: [41][ 200/ 1207] Overall Loss 0.338982 Objective Loss 0.338982 LR 0.001000 Time 0.022414 -2023-02-13 17:36:44,001 - Epoch: [41][ 210/ 1207] Overall Loss 0.337680 Objective Loss 0.337680 LR 0.001000 Time 0.022245 -2023-02-13 17:36:44,189 - Epoch: [41][ 220/ 1207] Overall Loss 0.338000 Objective Loss 0.338000 LR 0.001000 Time 0.022087 -2023-02-13 17:36:44,378 - Epoch: [41][ 230/ 1207] Overall Loss 0.338353 Objective Loss 0.338353 LR 0.001000 Time 0.021947 -2023-02-13 17:36:44,566 - Epoch: [41][ 240/ 1207] Overall Loss 0.339304 Objective Loss 0.339304 LR 0.001000 Time 0.021814 -2023-02-13 17:36:44,755 - Epoch: [41][ 250/ 1207] Overall Loss 0.340489 Objective Loss 0.340489 LR 0.001000 Time 0.021699 -2023-02-13 17:36:44,944 - Epoch: [41][ 260/ 1207] Overall Loss 0.340065 Objective Loss 0.340065 LR 0.001000 Time 0.021587 -2023-02-13 17:36:45,132 - Epoch: [41][ 270/ 1207] Overall Loss 0.339132 Objective Loss 0.339132 LR 0.001000 Time 0.021486 -2023-02-13 17:36:45,321 - Epoch: [41][ 280/ 1207] Overall Loss 0.339309 Objective Loss 0.339309 LR 0.001000 Time 0.021389 -2023-02-13 17:36:45,509 - Epoch: [41][ 290/ 1207] Overall Loss 0.338764 Objective Loss 0.338764 LR 0.001000 Time 0.021301 -2023-02-13 17:36:45,698 - Epoch: [41][ 300/ 1207] Overall Loss 0.338048 Objective Loss 0.338048 LR 0.001000 Time 0.021219 -2023-02-13 17:36:45,888 - Epoch: [41][ 310/ 1207] Overall Loss 0.337344 Objective Loss 0.337344 LR 0.001000 Time 0.021145 -2023-02-13 17:36:46,076 - Epoch: [41][ 320/ 1207] Overall Loss 0.336768 Objective Loss 0.336768 LR 0.001000 Time 0.021070 -2023-02-13 17:36:46,264 - Epoch: [41][ 330/ 1207] Overall Loss 0.336727 Objective Loss 0.336727 LR 0.001000 Time 0.021002 -2023-02-13 17:36:46,452 - Epoch: [41][ 340/ 1207] Overall Loss 0.336687 Objective Loss 0.336687 LR 0.001000 Time 0.020937 -2023-02-13 17:36:46,641 - Epoch: [41][ 350/ 1207] Overall Loss 0.336450 Objective Loss 0.336450 LR 0.001000 Time 0.020876 -2023-02-13 17:36:46,831 - Epoch: [41][ 360/ 1207] Overall Loss 0.336645 Objective Loss 0.336645 LR 0.001000 Time 0.020824 -2023-02-13 17:36:47,020 - Epoch: [41][ 370/ 1207] Overall Loss 0.336835 Objective Loss 0.336835 LR 0.001000 Time 0.020769 -2023-02-13 17:36:47,208 - Epoch: [41][ 380/ 1207] Overall Loss 0.337022 Objective Loss 0.337022 LR 0.001000 Time 0.020716 -2023-02-13 17:36:47,396 - Epoch: [41][ 390/ 1207] Overall Loss 0.337448 Objective Loss 0.337448 LR 0.001000 Time 0.020667 -2023-02-13 17:36:47,585 - Epoch: [41][ 400/ 1207] Overall Loss 0.338560 Objective Loss 0.338560 LR 0.001000 Time 0.020621 -2023-02-13 17:36:47,774 - Epoch: [41][ 410/ 1207] Overall Loss 0.338516 Objective Loss 0.338516 LR 0.001000 Time 0.020579 -2023-02-13 17:36:47,962 - Epoch: [41][ 420/ 1207] Overall Loss 0.339121 Objective Loss 0.339121 LR 0.001000 Time 0.020535 -2023-02-13 17:36:48,150 - Epoch: [41][ 430/ 1207] Overall Loss 0.338742 Objective Loss 0.338742 LR 0.001000 Time 0.020494 -2023-02-13 17:36:48,338 - Epoch: [41][ 440/ 1207] Overall Loss 0.338160 Objective Loss 0.338160 LR 0.001000 Time 0.020456 -2023-02-13 17:36:48,527 - Epoch: [41][ 450/ 1207] Overall Loss 0.338051 Objective Loss 0.338051 LR 0.001000 Time 0.020419 -2023-02-13 17:36:48,716 - Epoch: [41][ 460/ 1207] Overall Loss 0.336783 Objective Loss 0.336783 LR 0.001000 Time 0.020385 -2023-02-13 17:36:48,904 - Epoch: [41][ 470/ 1207] Overall Loss 0.336823 Objective Loss 0.336823 LR 0.001000 Time 0.020352 -2023-02-13 17:36:49,093 - Epoch: [41][ 480/ 1207] Overall Loss 0.336961 Objective Loss 0.336961 LR 0.001000 Time 0.020320 -2023-02-13 17:36:49,282 - Epoch: [41][ 490/ 1207] Overall Loss 0.336480 Objective Loss 0.336480 LR 0.001000 Time 0.020290 -2023-02-13 17:36:49,470 - Epoch: [41][ 500/ 1207] Overall Loss 0.336090 Objective Loss 0.336090 LR 0.001000 Time 0.020260 -2023-02-13 17:36:49,659 - Epoch: [41][ 510/ 1207] Overall Loss 0.336053 Objective Loss 0.336053 LR 0.001000 Time 0.020233 -2023-02-13 17:36:49,849 - Epoch: [41][ 520/ 1207] Overall Loss 0.335879 Objective Loss 0.335879 LR 0.001000 Time 0.020209 -2023-02-13 17:36:50,040 - Epoch: [41][ 530/ 1207] Overall Loss 0.336485 Objective Loss 0.336485 LR 0.001000 Time 0.020188 -2023-02-13 17:36:50,231 - Epoch: [41][ 540/ 1207] Overall Loss 0.336813 Objective Loss 0.336813 LR 0.001000 Time 0.020165 -2023-02-13 17:36:50,421 - Epoch: [41][ 550/ 1207] Overall Loss 0.337558 Objective Loss 0.337558 LR 0.001000 Time 0.020144 -2023-02-13 17:36:50,613 - Epoch: [41][ 560/ 1207] Overall Loss 0.337233 Objective Loss 0.337233 LR 0.001000 Time 0.020126 -2023-02-13 17:36:50,804 - Epoch: [41][ 570/ 1207] Overall Loss 0.336972 Objective Loss 0.336972 LR 0.001000 Time 0.020108 -2023-02-13 17:36:50,995 - Epoch: [41][ 580/ 1207] Overall Loss 0.336346 Objective Loss 0.336346 LR 0.001000 Time 0.020090 -2023-02-13 17:36:51,185 - Epoch: [41][ 590/ 1207] Overall Loss 0.336028 Objective Loss 0.336028 LR 0.001000 Time 0.020071 -2023-02-13 17:36:51,375 - Epoch: [41][ 600/ 1207] Overall Loss 0.335871 Objective Loss 0.335871 LR 0.001000 Time 0.020052 -2023-02-13 17:36:51,564 - Epoch: [41][ 610/ 1207] Overall Loss 0.335920 Objective Loss 0.335920 LR 0.001000 Time 0.020034 -2023-02-13 17:36:51,756 - Epoch: [41][ 620/ 1207] Overall Loss 0.336098 Objective Loss 0.336098 LR 0.001000 Time 0.020019 -2023-02-13 17:36:51,947 - Epoch: [41][ 630/ 1207] Overall Loss 0.336471 Objective Loss 0.336471 LR 0.001000 Time 0.020004 -2023-02-13 17:36:52,138 - Epoch: [41][ 640/ 1207] Overall Loss 0.335990 Objective Loss 0.335990 LR 0.001000 Time 0.019990 -2023-02-13 17:36:52,329 - Epoch: [41][ 650/ 1207] Overall Loss 0.336383 Objective Loss 0.336383 LR 0.001000 Time 0.019975 -2023-02-13 17:36:52,522 - Epoch: [41][ 660/ 1207] Overall Loss 0.336724 Objective Loss 0.336724 LR 0.001000 Time 0.019963 -2023-02-13 17:36:52,713 - Epoch: [41][ 670/ 1207] Overall Loss 0.337676 Objective Loss 0.337676 LR 0.001000 Time 0.019951 -2023-02-13 17:36:52,905 - Epoch: [41][ 680/ 1207] Overall Loss 0.337449 Objective Loss 0.337449 LR 0.001000 Time 0.019940 -2023-02-13 17:36:53,096 - Epoch: [41][ 690/ 1207] Overall Loss 0.337433 Objective Loss 0.337433 LR 0.001000 Time 0.019927 -2023-02-13 17:36:53,287 - Epoch: [41][ 700/ 1207] Overall Loss 0.337547 Objective Loss 0.337547 LR 0.001000 Time 0.019914 -2023-02-13 17:36:53,478 - Epoch: [41][ 710/ 1207] Overall Loss 0.337512 Objective Loss 0.337512 LR 0.001000 Time 0.019903 -2023-02-13 17:36:53,670 - Epoch: [41][ 720/ 1207] Overall Loss 0.337486 Objective Loss 0.337486 LR 0.001000 Time 0.019892 -2023-02-13 17:36:53,860 - Epoch: [41][ 730/ 1207] Overall Loss 0.337971 Objective Loss 0.337971 LR 0.001000 Time 0.019880 -2023-02-13 17:36:54,050 - Epoch: [41][ 740/ 1207] Overall Loss 0.337715 Objective Loss 0.337715 LR 0.001000 Time 0.019866 -2023-02-13 17:36:54,239 - Epoch: [41][ 750/ 1207] Overall Loss 0.337983 Objective Loss 0.337983 LR 0.001000 Time 0.019853 -2023-02-13 17:36:54,429 - Epoch: [41][ 760/ 1207] Overall Loss 0.338082 Objective Loss 0.338082 LR 0.001000 Time 0.019841 -2023-02-13 17:36:54,619 - Epoch: [41][ 770/ 1207] Overall Loss 0.338463 Objective Loss 0.338463 LR 0.001000 Time 0.019830 -2023-02-13 17:36:54,809 - Epoch: [41][ 780/ 1207] Overall Loss 0.338763 Objective Loss 0.338763 LR 0.001000 Time 0.019819 -2023-02-13 17:36:54,999 - Epoch: [41][ 790/ 1207] Overall Loss 0.338721 Objective Loss 0.338721 LR 0.001000 Time 0.019808 -2023-02-13 17:36:55,189 - Epoch: [41][ 800/ 1207] Overall Loss 0.338888 Objective Loss 0.338888 LR 0.001000 Time 0.019798 -2023-02-13 17:36:55,380 - Epoch: [41][ 810/ 1207] Overall Loss 0.338962 Objective Loss 0.338962 LR 0.001000 Time 0.019789 -2023-02-13 17:36:55,571 - Epoch: [41][ 820/ 1207] Overall Loss 0.339285 Objective Loss 0.339285 LR 0.001000 Time 0.019780 -2023-02-13 17:36:55,763 - Epoch: [41][ 830/ 1207] Overall Loss 0.339229 Objective Loss 0.339229 LR 0.001000 Time 0.019773 -2023-02-13 17:36:55,955 - Epoch: [41][ 840/ 1207] Overall Loss 0.339084 Objective Loss 0.339084 LR 0.001000 Time 0.019765 -2023-02-13 17:36:56,145 - Epoch: [41][ 850/ 1207] Overall Loss 0.339567 Objective Loss 0.339567 LR 0.001000 Time 0.019756 -2023-02-13 17:36:56,338 - Epoch: [41][ 860/ 1207] Overall Loss 0.339999 Objective Loss 0.339999 LR 0.001000 Time 0.019750 -2023-02-13 17:36:56,529 - Epoch: [41][ 870/ 1207] Overall Loss 0.339847 Objective Loss 0.339847 LR 0.001000 Time 0.019742 -2023-02-13 17:36:56,720 - Epoch: [41][ 880/ 1207] Overall Loss 0.339805 Objective Loss 0.339805 LR 0.001000 Time 0.019735 -2023-02-13 17:36:56,912 - Epoch: [41][ 890/ 1207] Overall Loss 0.339690 Objective Loss 0.339690 LR 0.001000 Time 0.019728 -2023-02-13 17:36:57,103 - Epoch: [41][ 900/ 1207] Overall Loss 0.339544 Objective Loss 0.339544 LR 0.001000 Time 0.019721 -2023-02-13 17:36:57,293 - Epoch: [41][ 910/ 1207] Overall Loss 0.339349 Objective Loss 0.339349 LR 0.001000 Time 0.019712 -2023-02-13 17:36:57,484 - Epoch: [41][ 920/ 1207] Overall Loss 0.339243 Objective Loss 0.339243 LR 0.001000 Time 0.019706 -2023-02-13 17:36:57,676 - Epoch: [41][ 930/ 1207] Overall Loss 0.339567 Objective Loss 0.339567 LR 0.001000 Time 0.019700 -2023-02-13 17:36:57,867 - Epoch: [41][ 940/ 1207] Overall Loss 0.339429 Objective Loss 0.339429 LR 0.001000 Time 0.019693 -2023-02-13 17:36:58,058 - Epoch: [41][ 950/ 1207] Overall Loss 0.339378 Objective Loss 0.339378 LR 0.001000 Time 0.019686 -2023-02-13 17:36:58,249 - Epoch: [41][ 960/ 1207] Overall Loss 0.339151 Objective Loss 0.339151 LR 0.001000 Time 0.019679 -2023-02-13 17:36:58,440 - Epoch: [41][ 970/ 1207] Overall Loss 0.339350 Objective Loss 0.339350 LR 0.001000 Time 0.019673 -2023-02-13 17:36:58,631 - Epoch: [41][ 980/ 1207] Overall Loss 0.339461 Objective Loss 0.339461 LR 0.001000 Time 0.019667 -2023-02-13 17:36:58,824 - Epoch: [41][ 990/ 1207] Overall Loss 0.339496 Objective Loss 0.339496 LR 0.001000 Time 0.019663 -2023-02-13 17:36:59,015 - Epoch: [41][ 1000/ 1207] Overall Loss 0.339567 Objective Loss 0.339567 LR 0.001000 Time 0.019657 -2023-02-13 17:36:59,206 - Epoch: [41][ 1010/ 1207] Overall Loss 0.339767 Objective Loss 0.339767 LR 0.001000 Time 0.019651 -2023-02-13 17:36:59,397 - Epoch: [41][ 1020/ 1207] Overall Loss 0.339673 Objective Loss 0.339673 LR 0.001000 Time 0.019646 -2023-02-13 17:36:59,588 - Epoch: [41][ 1030/ 1207] Overall Loss 0.339942 Objective Loss 0.339942 LR 0.001000 Time 0.019640 -2023-02-13 17:36:59,781 - Epoch: [41][ 1040/ 1207] Overall Loss 0.339967 Objective Loss 0.339967 LR 0.001000 Time 0.019636 -2023-02-13 17:36:59,972 - Epoch: [41][ 1050/ 1207] Overall Loss 0.340189 Objective Loss 0.340189 LR 0.001000 Time 0.019630 -2023-02-13 17:37:00,163 - Epoch: [41][ 1060/ 1207] Overall Loss 0.340233 Objective Loss 0.340233 LR 0.001000 Time 0.019625 -2023-02-13 17:37:00,353 - Epoch: [41][ 1070/ 1207] Overall Loss 0.340055 Objective Loss 0.340055 LR 0.001000 Time 0.019619 -2023-02-13 17:37:00,545 - Epoch: [41][ 1080/ 1207] Overall Loss 0.339939 Objective Loss 0.339939 LR 0.001000 Time 0.019615 -2023-02-13 17:37:00,737 - Epoch: [41][ 1090/ 1207] Overall Loss 0.339639 Objective Loss 0.339639 LR 0.001000 Time 0.019611 -2023-02-13 17:37:00,930 - Epoch: [41][ 1100/ 1207] Overall Loss 0.339898 Objective Loss 0.339898 LR 0.001000 Time 0.019607 -2023-02-13 17:37:01,121 - Epoch: [41][ 1110/ 1207] Overall Loss 0.339852 Objective Loss 0.339852 LR 0.001000 Time 0.019603 -2023-02-13 17:37:01,313 - Epoch: [41][ 1120/ 1207] Overall Loss 0.339752 Objective Loss 0.339752 LR 0.001000 Time 0.019599 -2023-02-13 17:37:01,503 - Epoch: [41][ 1130/ 1207] Overall Loss 0.340069 Objective Loss 0.340069 LR 0.001000 Time 0.019594 -2023-02-13 17:37:01,695 - Epoch: [41][ 1140/ 1207] Overall Loss 0.340044 Objective Loss 0.340044 LR 0.001000 Time 0.019590 -2023-02-13 17:37:01,888 - Epoch: [41][ 1150/ 1207] Overall Loss 0.339972 Objective Loss 0.339972 LR 0.001000 Time 0.019587 -2023-02-13 17:37:02,079 - Epoch: [41][ 1160/ 1207] Overall Loss 0.340423 Objective Loss 0.340423 LR 0.001000 Time 0.019582 -2023-02-13 17:37:02,269 - Epoch: [41][ 1170/ 1207] Overall Loss 0.340441 Objective Loss 0.340441 LR 0.001000 Time 0.019576 -2023-02-13 17:37:02,459 - Epoch: [41][ 1180/ 1207] Overall Loss 0.340867 Objective Loss 0.340867 LR 0.001000 Time 0.019572 -2023-02-13 17:37:02,650 - Epoch: [41][ 1190/ 1207] Overall Loss 0.341040 Objective Loss 0.341040 LR 0.001000 Time 0.019567 -2023-02-13 17:37:02,891 - Epoch: [41][ 1200/ 1207] Overall Loss 0.340951 Objective Loss 0.340951 LR 0.001000 Time 0.019605 -2023-02-13 17:37:03,008 - Epoch: [41][ 1207/ 1207] Overall Loss 0.340923 Objective Loss 0.340923 Top1 82.012195 Top5 98.475610 LR 0.001000 Time 0.019588 -2023-02-13 17:37:03,088 - --- validate (epoch=41)----------- -2023-02-13 17:37:03,088 - 34311 samples (256 per mini-batch) -2023-02-13 17:37:03,502 - Epoch: [41][ 10/ 135] Loss 0.349447 Top1 83.046875 Top5 97.773438 -2023-02-13 17:37:03,632 - Epoch: [41][ 20/ 135] Loss 0.353694 Top1 82.675781 Top5 97.441406 -2023-02-13 17:37:03,757 - Epoch: [41][ 30/ 135] Loss 0.360275 Top1 82.083333 Top5 97.395833 -2023-02-13 17:37:03,879 - Epoch: [41][ 40/ 135] Loss 0.366907 Top1 82.177734 Top5 97.294922 -2023-02-13 17:37:04,003 - Epoch: [41][ 50/ 135] Loss 0.365393 Top1 82.132812 Top5 97.414062 -2023-02-13 17:37:04,129 - Epoch: [41][ 60/ 135] Loss 0.363886 Top1 82.141927 Top5 97.363281 -2023-02-13 17:37:04,252 - Epoch: [41][ 70/ 135] Loss 0.363267 Top1 81.908482 Top5 97.287946 -2023-02-13 17:37:04,375 - Epoch: [41][ 80/ 135] Loss 0.366326 Top1 81.777344 Top5 97.236328 -2023-02-13 17:37:04,499 - Epoch: [41][ 90/ 135] Loss 0.364565 Top1 81.814236 Top5 97.265625 -2023-02-13 17:37:04,623 - Epoch: [41][ 100/ 135] Loss 0.364778 Top1 81.816406 Top5 97.265625 -2023-02-13 17:37:04,748 - Epoch: [41][ 110/ 135] Loss 0.365809 Top1 81.761364 Top5 97.262074 -2023-02-13 17:37:04,875 - Epoch: [41][ 120/ 135] Loss 0.365698 Top1 81.832682 Top5 97.236328 -2023-02-13 17:37:05,005 - Epoch: [41][ 130/ 135] Loss 0.366675 Top1 81.790865 Top5 97.214543 -2023-02-13 17:37:05,052 - Epoch: [41][ 135/ 135] Loss 0.364080 Top1 81.795926 Top5 97.199149 -2023-02-13 17:37:05,132 - ==> Top1: 81.796 Top5: 97.199 Loss: 0.364 - -2023-02-13 17:37:05,133 - ==> Confusion: -[[ 834 6 7 0 21 4 0 3 4 46 1 5 3 3 3 8 6 2 1 0 10] - [ 2 923 1 2 16 34 3 20 3 0 4 0 2 1 2 4 5 0 2 3 6] - [ 4 3 931 18 5 2 20 23 0 0 4 1 3 2 2 12 3 5 8 6 6] - [ 7 2 15 896 3 1 2 3 1 3 16 1 8 2 17 3 3 6 18 0 9] - [ 11 11 3 0 993 5 3 0 0 1 1 5 2 1 9 5 6 0 2 4 4] - [ 0 26 0 5 7 951 5 17 2 5 3 8 6 13 4 4 4 1 1 5 3] - [ 3 4 19 3 0 4 1033 7 1 0 4 3 2 2 0 3 1 2 1 5 2] - [ 1 14 8 2 5 34 9 903 0 2 7 2 4 2 0 0 1 0 12 12 6] - [ 23 1 0 1 2 2 0 1 868 39 14 1 2 10 28 3 1 3 8 0 2] - [ 82 2 4 0 10 5 0 3 26 834 2 0 3 21 6 3 1 2 0 1 7] - [ 2 4 4 4 3 2 6 4 22 1 965 1 0 12 4 0 2 2 8 2 3] - [ 4 4 1 0 1 15 0 2 0 0 0 882 52 4 3 9 5 3 3 15 2] - [ 1 2 1 2 2 4 0 0 0 0 1 31 870 0 7 5 0 18 3 6 6] - [ 5 3 3 0 9 20 2 1 14 15 6 7 5 895 8 11 4 0 0 10 6] - [ 12 4 1 24 8 3 0 2 18 3 4 1 4 0 982 1 5 5 9 0 6] - [ 4 3 5 1 5 3 4 1 0 0 0 7 10 2 0 970 3 11 0 10 7] - [ 4 7 0 3 14 2 0 0 2 1 0 2 2 1 1 25 974 2 4 7 10] - [ 3 3 2 6 0 0 0 1 1 0 0 11 40 3 4 16 0 949 0 5 7] - [ 2 3 6 10 2 1 0 36 4 1 9 0 9 0 10 2 2 4 976 4 5] - [ 0 1 0 0 1 6 6 14 1 0 0 20 5 5 0 9 1 3 0 1068 8] - [ 156 329 203 148 222 270 99 215 83 85 206 155 384 330 171 144 252 79 183 352 9368]] - -2023-02-13 17:37:05,134 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:37:05,135 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:37:05,140 - - -2023-02-13 17:37:05,141 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:37:06,031 - Epoch: [42][ 10/ 1207] Overall Loss 0.339684 Objective Loss 0.339684 LR 0.001000 Time 0.088964 -2023-02-13 17:37:06,239 - Epoch: [42][ 20/ 1207] Overall Loss 0.322750 Objective Loss 0.322750 LR 0.001000 Time 0.054861 -2023-02-13 17:37:06,436 - Epoch: [42][ 30/ 1207] Overall Loss 0.322374 Objective Loss 0.322374 LR 0.001000 Time 0.043148 -2023-02-13 17:37:06,638 - Epoch: [42][ 40/ 1207] Overall Loss 0.324434 Objective Loss 0.324434 LR 0.001000 Time 0.037380 -2023-02-13 17:37:06,836 - Epoch: [42][ 50/ 1207] Overall Loss 0.325609 Objective Loss 0.325609 LR 0.001000 Time 0.033860 -2023-02-13 17:37:07,037 - Epoch: [42][ 60/ 1207] Overall Loss 0.330504 Objective Loss 0.330504 LR 0.001000 Time 0.031559 -2023-02-13 17:37:07,234 - Epoch: [42][ 70/ 1207] Overall Loss 0.328150 Objective Loss 0.328150 LR 0.001000 Time 0.029866 -2023-02-13 17:37:07,436 - Epoch: [42][ 80/ 1207] Overall Loss 0.328083 Objective Loss 0.328083 LR 0.001000 Time 0.028652 -2023-02-13 17:37:07,633 - Epoch: [42][ 90/ 1207] Overall Loss 0.329610 Objective Loss 0.329610 LR 0.001000 Time 0.027655 -2023-02-13 17:37:07,835 - Epoch: [42][ 100/ 1207] Overall Loss 0.327629 Objective Loss 0.327629 LR 0.001000 Time 0.026905 -2023-02-13 17:37:08,032 - Epoch: [42][ 110/ 1207] Overall Loss 0.326420 Objective Loss 0.326420 LR 0.001000 Time 0.026246 -2023-02-13 17:37:08,233 - Epoch: [42][ 120/ 1207] Overall Loss 0.324530 Objective Loss 0.324530 LR 0.001000 Time 0.025727 -2023-02-13 17:37:08,430 - Epoch: [42][ 130/ 1207] Overall Loss 0.324250 Objective Loss 0.324250 LR 0.001000 Time 0.025264 -2023-02-13 17:37:08,633 - Epoch: [42][ 140/ 1207] Overall Loss 0.325951 Objective Loss 0.325951 LR 0.001000 Time 0.024901 -2023-02-13 17:37:08,830 - Epoch: [42][ 150/ 1207] Overall Loss 0.326416 Objective Loss 0.326416 LR 0.001000 Time 0.024557 -2023-02-13 17:37:09,032 - Epoch: [42][ 160/ 1207] Overall Loss 0.327508 Objective Loss 0.327508 LR 0.001000 Time 0.024280 -2023-02-13 17:37:09,228 - Epoch: [42][ 170/ 1207] Overall Loss 0.329452 Objective Loss 0.329452 LR 0.001000 Time 0.024005 -2023-02-13 17:37:09,430 - Epoch: [42][ 180/ 1207] Overall Loss 0.328556 Objective Loss 0.328556 LR 0.001000 Time 0.023788 -2023-02-13 17:37:09,627 - Epoch: [42][ 190/ 1207] Overall Loss 0.328053 Objective Loss 0.328053 LR 0.001000 Time 0.023573 -2023-02-13 17:37:09,830 - Epoch: [42][ 200/ 1207] Overall Loss 0.328967 Objective Loss 0.328967 LR 0.001000 Time 0.023404 -2023-02-13 17:37:10,027 - Epoch: [42][ 210/ 1207] Overall Loss 0.330248 Objective Loss 0.330248 LR 0.001000 Time 0.023226 -2023-02-13 17:37:10,228 - Epoch: [42][ 220/ 1207] Overall Loss 0.331742 Objective Loss 0.331742 LR 0.001000 Time 0.023083 -2023-02-13 17:37:10,425 - Epoch: [42][ 230/ 1207] Overall Loss 0.331498 Objective Loss 0.331498 LR 0.001000 Time 0.022934 -2023-02-13 17:37:10,626 - Epoch: [42][ 240/ 1207] Overall Loss 0.331815 Objective Loss 0.331815 LR 0.001000 Time 0.022817 -2023-02-13 17:37:10,824 - Epoch: [42][ 250/ 1207] Overall Loss 0.333444 Objective Loss 0.333444 LR 0.001000 Time 0.022693 -2023-02-13 17:37:11,025 - Epoch: [42][ 260/ 1207] Overall Loss 0.332639 Objective Loss 0.332639 LR 0.001000 Time 0.022593 -2023-02-13 17:37:11,222 - Epoch: [42][ 270/ 1207] Overall Loss 0.332726 Objective Loss 0.332726 LR 0.001000 Time 0.022485 -2023-02-13 17:37:11,424 - Epoch: [42][ 280/ 1207] Overall Loss 0.333090 Objective Loss 0.333090 LR 0.001000 Time 0.022402 -2023-02-13 17:37:11,622 - Epoch: [42][ 290/ 1207] Overall Loss 0.332808 Objective Loss 0.332808 LR 0.001000 Time 0.022309 -2023-02-13 17:37:11,825 - Epoch: [42][ 300/ 1207] Overall Loss 0.333283 Objective Loss 0.333283 LR 0.001000 Time 0.022240 -2023-02-13 17:37:12,022 - Epoch: [42][ 310/ 1207] Overall Loss 0.333279 Objective Loss 0.333279 LR 0.001000 Time 0.022159 -2023-02-13 17:37:12,225 - Epoch: [42][ 320/ 1207] Overall Loss 0.333146 Objective Loss 0.333146 LR 0.001000 Time 0.022098 -2023-02-13 17:37:12,423 - Epoch: [42][ 330/ 1207] Overall Loss 0.333372 Objective Loss 0.333372 LR 0.001000 Time 0.022028 -2023-02-13 17:37:12,625 - Epoch: [42][ 340/ 1207] Overall Loss 0.333948 Objective Loss 0.333948 LR 0.001000 Time 0.021972 -2023-02-13 17:37:12,823 - Epoch: [42][ 350/ 1207] Overall Loss 0.333178 Objective Loss 0.333178 LR 0.001000 Time 0.021909 -2023-02-13 17:37:13,025 - Epoch: [42][ 360/ 1207] Overall Loss 0.333266 Objective Loss 0.333266 LR 0.001000 Time 0.021861 -2023-02-13 17:37:13,222 - Epoch: [42][ 370/ 1207] Overall Loss 0.333031 Objective Loss 0.333031 LR 0.001000 Time 0.021803 -2023-02-13 17:37:13,425 - Epoch: [42][ 380/ 1207] Overall Loss 0.332547 Objective Loss 0.332547 LR 0.001000 Time 0.021761 -2023-02-13 17:37:13,623 - Epoch: [42][ 390/ 1207] Overall Loss 0.331986 Objective Loss 0.331986 LR 0.001000 Time 0.021709 -2023-02-13 17:37:13,826 - Epoch: [42][ 400/ 1207] Overall Loss 0.332453 Objective Loss 0.332453 LR 0.001000 Time 0.021673 -2023-02-13 17:37:14,023 - Epoch: [42][ 410/ 1207] Overall Loss 0.332282 Objective Loss 0.332282 LR 0.001000 Time 0.021625 -2023-02-13 17:37:14,225 - Epoch: [42][ 420/ 1207] Overall Loss 0.332407 Objective Loss 0.332407 LR 0.001000 Time 0.021589 -2023-02-13 17:37:14,423 - Epoch: [42][ 430/ 1207] Overall Loss 0.331743 Objective Loss 0.331743 LR 0.001000 Time 0.021546 -2023-02-13 17:37:14,625 - Epoch: [42][ 440/ 1207] Overall Loss 0.331457 Objective Loss 0.331457 LR 0.001000 Time 0.021515 -2023-02-13 17:37:14,823 - Epoch: [42][ 450/ 1207] Overall Loss 0.330800 Objective Loss 0.330800 LR 0.001000 Time 0.021476 -2023-02-13 17:37:15,024 - Epoch: [42][ 460/ 1207] Overall Loss 0.330214 Objective Loss 0.330214 LR 0.001000 Time 0.021447 -2023-02-13 17:37:15,222 - Epoch: [42][ 470/ 1207] Overall Loss 0.330047 Objective Loss 0.330047 LR 0.001000 Time 0.021411 -2023-02-13 17:37:15,425 - Epoch: [42][ 480/ 1207] Overall Loss 0.330352 Objective Loss 0.330352 LR 0.001000 Time 0.021386 -2023-02-13 17:37:15,623 - Epoch: [42][ 490/ 1207] Overall Loss 0.330333 Objective Loss 0.330333 LR 0.001000 Time 0.021352 -2023-02-13 17:37:15,826 - Epoch: [42][ 500/ 1207] Overall Loss 0.331192 Objective Loss 0.331192 LR 0.001000 Time 0.021331 -2023-02-13 17:37:16,024 - Epoch: [42][ 510/ 1207] Overall Loss 0.331176 Objective Loss 0.331176 LR 0.001000 Time 0.021300 -2023-02-13 17:37:16,226 - Epoch: [42][ 520/ 1207] Overall Loss 0.331383 Objective Loss 0.331383 LR 0.001000 Time 0.021278 -2023-02-13 17:37:16,424 - Epoch: [42][ 530/ 1207] Overall Loss 0.332015 Objective Loss 0.332015 LR 0.001000 Time 0.021249 -2023-02-13 17:37:16,626 - Epoch: [42][ 540/ 1207] Overall Loss 0.332564 Objective Loss 0.332564 LR 0.001000 Time 0.021229 -2023-02-13 17:37:16,824 - Epoch: [42][ 550/ 1207] Overall Loss 0.333205 Objective Loss 0.333205 LR 0.001000 Time 0.021203 -2023-02-13 17:37:17,026 - Epoch: [42][ 560/ 1207] Overall Loss 0.333404 Objective Loss 0.333404 LR 0.001000 Time 0.021184 -2023-02-13 17:37:17,223 - Epoch: [42][ 570/ 1207] Overall Loss 0.333701 Objective Loss 0.333701 LR 0.001000 Time 0.021158 -2023-02-13 17:37:17,425 - Epoch: [42][ 580/ 1207] Overall Loss 0.333554 Objective Loss 0.333554 LR 0.001000 Time 0.021140 -2023-02-13 17:37:17,623 - Epoch: [42][ 590/ 1207] Overall Loss 0.334312 Objective Loss 0.334312 LR 0.001000 Time 0.021116 -2023-02-13 17:37:17,816 - Epoch: [42][ 600/ 1207] Overall Loss 0.334690 Objective Loss 0.334690 LR 0.001000 Time 0.021086 -2023-02-13 17:37:18,004 - Epoch: [42][ 610/ 1207] Overall Loss 0.334611 Objective Loss 0.334611 LR 0.001000 Time 0.021047 -2023-02-13 17:37:18,191 - Epoch: [42][ 620/ 1207] Overall Loss 0.334592 Objective Loss 0.334592 LR 0.001000 Time 0.021009 -2023-02-13 17:37:18,379 - Epoch: [42][ 630/ 1207] Overall Loss 0.334596 Objective Loss 0.334596 LR 0.001000 Time 0.020973 -2023-02-13 17:37:18,566 - Epoch: [42][ 640/ 1207] Overall Loss 0.334519 Objective Loss 0.334519 LR 0.001000 Time 0.020938 -2023-02-13 17:37:18,755 - Epoch: [42][ 650/ 1207] Overall Loss 0.334723 Objective Loss 0.334723 LR 0.001000 Time 0.020906 -2023-02-13 17:37:18,945 - Epoch: [42][ 660/ 1207] Overall Loss 0.335000 Objective Loss 0.335000 LR 0.001000 Time 0.020875 -2023-02-13 17:37:19,135 - Epoch: [42][ 670/ 1207] Overall Loss 0.335603 Objective Loss 0.335603 LR 0.001000 Time 0.020847 -2023-02-13 17:37:19,324 - Epoch: [42][ 680/ 1207] Overall Loss 0.336051 Objective Loss 0.336051 LR 0.001000 Time 0.020819 -2023-02-13 17:37:19,516 - Epoch: [42][ 690/ 1207] Overall Loss 0.336814 Objective Loss 0.336814 LR 0.001000 Time 0.020794 -2023-02-13 17:37:19,706 - Epoch: [42][ 700/ 1207] Overall Loss 0.336842 Objective Loss 0.336842 LR 0.001000 Time 0.020768 -2023-02-13 17:37:19,898 - Epoch: [42][ 710/ 1207] Overall Loss 0.336858 Objective Loss 0.336858 LR 0.001000 Time 0.020745 -2023-02-13 17:37:20,087 - Epoch: [42][ 720/ 1207] Overall Loss 0.336674 Objective Loss 0.336674 LR 0.001000 Time 0.020720 -2023-02-13 17:37:20,278 - Epoch: [42][ 730/ 1207] Overall Loss 0.336584 Objective Loss 0.336584 LR 0.001000 Time 0.020697 -2023-02-13 17:37:20,468 - Epoch: [42][ 740/ 1207] Overall Loss 0.336366 Objective Loss 0.336366 LR 0.001000 Time 0.020674 -2023-02-13 17:37:20,659 - Epoch: [42][ 750/ 1207] Overall Loss 0.336886 Objective Loss 0.336886 LR 0.001000 Time 0.020652 -2023-02-13 17:37:20,850 - Epoch: [42][ 760/ 1207] Overall Loss 0.337125 Objective Loss 0.337125 LR 0.001000 Time 0.020632 -2023-02-13 17:37:21,041 - Epoch: [42][ 770/ 1207] Overall Loss 0.336924 Objective Loss 0.336924 LR 0.001000 Time 0.020611 -2023-02-13 17:37:21,231 - Epoch: [42][ 780/ 1207] Overall Loss 0.337077 Objective Loss 0.337077 LR 0.001000 Time 0.020590 -2023-02-13 17:37:21,423 - Epoch: [42][ 790/ 1207] Overall Loss 0.337197 Objective Loss 0.337197 LR 0.001000 Time 0.020572 -2023-02-13 17:37:21,613 - Epoch: [42][ 800/ 1207] Overall Loss 0.337510 Objective Loss 0.337510 LR 0.001000 Time 0.020552 -2023-02-13 17:37:21,806 - Epoch: [42][ 810/ 1207] Overall Loss 0.337642 Objective Loss 0.337642 LR 0.001000 Time 0.020536 -2023-02-13 17:37:21,997 - Epoch: [42][ 820/ 1207] Overall Loss 0.337595 Objective Loss 0.337595 LR 0.001000 Time 0.020517 -2023-02-13 17:37:22,187 - Epoch: [42][ 830/ 1207] Overall Loss 0.338049 Objective Loss 0.338049 LR 0.001000 Time 0.020499 -2023-02-13 17:37:22,377 - Epoch: [42][ 840/ 1207] Overall Loss 0.338187 Objective Loss 0.338187 LR 0.001000 Time 0.020480 -2023-02-13 17:37:22,567 - Epoch: [42][ 850/ 1207] Overall Loss 0.338117 Objective Loss 0.338117 LR 0.001000 Time 0.020463 -2023-02-13 17:37:22,758 - Epoch: [42][ 860/ 1207] Overall Loss 0.338343 Objective Loss 0.338343 LR 0.001000 Time 0.020446 -2023-02-13 17:37:22,950 - Epoch: [42][ 870/ 1207] Overall Loss 0.338846 Objective Loss 0.338846 LR 0.001000 Time 0.020431 -2023-02-13 17:37:23,139 - Epoch: [42][ 880/ 1207] Overall Loss 0.338994 Objective Loss 0.338994 LR 0.001000 Time 0.020414 -2023-02-13 17:37:23,330 - Epoch: [42][ 890/ 1207] Overall Loss 0.339227 Objective Loss 0.339227 LR 0.001000 Time 0.020398 -2023-02-13 17:37:23,518 - Epoch: [42][ 900/ 1207] Overall Loss 0.339669 Objective Loss 0.339669 LR 0.001000 Time 0.020381 -2023-02-13 17:37:23,707 - Epoch: [42][ 910/ 1207] Overall Loss 0.339439 Objective Loss 0.339439 LR 0.001000 Time 0.020364 -2023-02-13 17:37:23,896 - Epoch: [42][ 920/ 1207] Overall Loss 0.339575 Objective Loss 0.339575 LR 0.001000 Time 0.020347 -2023-02-13 17:37:24,085 - Epoch: [42][ 930/ 1207] Overall Loss 0.339411 Objective Loss 0.339411 LR 0.001000 Time 0.020331 -2023-02-13 17:37:24,272 - Epoch: [42][ 940/ 1207] Overall Loss 0.339593 Objective Loss 0.339593 LR 0.001000 Time 0.020314 -2023-02-13 17:37:24,461 - Epoch: [42][ 950/ 1207] Overall Loss 0.339676 Objective Loss 0.339676 LR 0.001000 Time 0.020299 -2023-02-13 17:37:24,649 - Epoch: [42][ 960/ 1207] Overall Loss 0.339864 Objective Loss 0.339864 LR 0.001000 Time 0.020283 -2023-02-13 17:37:24,838 - Epoch: [42][ 970/ 1207] Overall Loss 0.339830 Objective Loss 0.339830 LR 0.001000 Time 0.020267 -2023-02-13 17:37:25,026 - Epoch: [42][ 980/ 1207] Overall Loss 0.339452 Objective Loss 0.339452 LR 0.001000 Time 0.020252 -2023-02-13 17:37:25,213 - Epoch: [42][ 990/ 1207] Overall Loss 0.339629 Objective Loss 0.339629 LR 0.001000 Time 0.020237 -2023-02-13 17:37:25,401 - Epoch: [42][ 1000/ 1207] Overall Loss 0.339543 Objective Loss 0.339543 LR 0.001000 Time 0.020222 -2023-02-13 17:37:25,589 - Epoch: [42][ 1010/ 1207] Overall Loss 0.339584 Objective Loss 0.339584 LR 0.001000 Time 0.020208 -2023-02-13 17:37:25,778 - Epoch: [42][ 1020/ 1207] Overall Loss 0.339471 Objective Loss 0.339471 LR 0.001000 Time 0.020194 -2023-02-13 17:37:25,968 - Epoch: [42][ 1030/ 1207] Overall Loss 0.339877 Objective Loss 0.339877 LR 0.001000 Time 0.020182 -2023-02-13 17:37:26,155 - Epoch: [42][ 1040/ 1207] Overall Loss 0.339642 Objective Loss 0.339642 LR 0.001000 Time 0.020168 -2023-02-13 17:37:26,344 - Epoch: [42][ 1050/ 1207] Overall Loss 0.339336 Objective Loss 0.339336 LR 0.001000 Time 0.020155 -2023-02-13 17:37:26,533 - Epoch: [42][ 1060/ 1207] Overall Loss 0.339161 Objective Loss 0.339161 LR 0.001000 Time 0.020143 -2023-02-13 17:37:26,721 - Epoch: [42][ 1070/ 1207] Overall Loss 0.339411 Objective Loss 0.339411 LR 0.001000 Time 0.020130 -2023-02-13 17:37:26,912 - Epoch: [42][ 1080/ 1207] Overall Loss 0.339400 Objective Loss 0.339400 LR 0.001000 Time 0.020120 -2023-02-13 17:37:27,101 - Epoch: [42][ 1090/ 1207] Overall Loss 0.339392 Objective Loss 0.339392 LR 0.001000 Time 0.020109 -2023-02-13 17:37:27,291 - Epoch: [42][ 1100/ 1207] Overall Loss 0.339501 Objective Loss 0.339501 LR 0.001000 Time 0.020099 -2023-02-13 17:37:27,481 - Epoch: [42][ 1110/ 1207] Overall Loss 0.339415 Objective Loss 0.339415 LR 0.001000 Time 0.020088 -2023-02-13 17:37:27,671 - Epoch: [42][ 1120/ 1207] Overall Loss 0.339557 Objective Loss 0.339557 LR 0.001000 Time 0.020078 -2023-02-13 17:37:27,862 - Epoch: [42][ 1130/ 1207] Overall Loss 0.339562 Objective Loss 0.339562 LR 0.001000 Time 0.020069 -2023-02-13 17:37:28,053 - Epoch: [42][ 1140/ 1207] Overall Loss 0.339606 Objective Loss 0.339606 LR 0.001000 Time 0.020060 -2023-02-13 17:37:28,242 - Epoch: [42][ 1150/ 1207] Overall Loss 0.339912 Objective Loss 0.339912 LR 0.001000 Time 0.020050 -2023-02-13 17:37:28,433 - Epoch: [42][ 1160/ 1207] Overall Loss 0.339918 Objective Loss 0.339918 LR 0.001000 Time 0.020041 -2023-02-13 17:37:28,622 - Epoch: [42][ 1170/ 1207] Overall Loss 0.339749 Objective Loss 0.339749 LR 0.001000 Time 0.020031 -2023-02-13 17:37:28,813 - Epoch: [42][ 1180/ 1207] Overall Loss 0.339641 Objective Loss 0.339641 LR 0.001000 Time 0.020023 -2023-02-13 17:37:29,004 - Epoch: [42][ 1190/ 1207] Overall Loss 0.339775 Objective Loss 0.339775 LR 0.001000 Time 0.020015 -2023-02-13 17:37:29,242 - Epoch: [42][ 1200/ 1207] Overall Loss 0.340096 Objective Loss 0.340096 LR 0.001000 Time 0.020046 -2023-02-13 17:37:29,358 - Epoch: [42][ 1207/ 1207] Overall Loss 0.340169 Objective Loss 0.340169 Top1 85.365854 Top5 97.560976 LR 0.001000 Time 0.020026 -2023-02-13 17:37:29,430 - --- validate (epoch=42)----------- -2023-02-13 17:37:29,430 - 34311 samples (256 per mini-batch) -2023-02-13 17:37:29,839 - Epoch: [42][ 10/ 135] Loss 0.341252 Top1 82.500000 Top5 97.265625 -2023-02-13 17:37:29,974 - Epoch: [42][ 20/ 135] Loss 0.343662 Top1 82.382812 Top5 97.324219 -2023-02-13 17:37:30,107 - Epoch: [42][ 30/ 135] Loss 0.355263 Top1 82.252604 Top5 97.330729 -2023-02-13 17:37:30,241 - Epoch: [42][ 40/ 135] Loss 0.357145 Top1 82.197266 Top5 97.373047 -2023-02-13 17:37:30,373 - Epoch: [42][ 50/ 135] Loss 0.360766 Top1 82.156250 Top5 97.328125 -2023-02-13 17:37:30,505 - Epoch: [42][ 60/ 135] Loss 0.358386 Top1 82.128906 Top5 97.369792 -2023-02-13 17:37:30,638 - Epoch: [42][ 70/ 135] Loss 0.362260 Top1 82.053571 Top5 97.354911 -2023-02-13 17:37:30,771 - Epoch: [42][ 80/ 135] Loss 0.362984 Top1 81.938477 Top5 97.319336 -2023-02-13 17:37:30,900 - Epoch: [42][ 90/ 135] Loss 0.363133 Top1 81.974826 Top5 97.309028 -2023-02-13 17:37:31,033 - Epoch: [42][ 100/ 135] Loss 0.361682 Top1 81.871094 Top5 97.375000 -2023-02-13 17:37:31,173 - Epoch: [42][ 110/ 135] Loss 0.361596 Top1 81.917614 Top5 97.372159 -2023-02-13 17:37:31,298 - Epoch: [42][ 120/ 135] Loss 0.359433 Top1 82.037760 Top5 97.415365 -2023-02-13 17:37:31,430 - Epoch: [42][ 130/ 135] Loss 0.359948 Top1 82.046274 Top5 97.412861 -2023-02-13 17:37:31,477 - Epoch: [42][ 135/ 135] Loss 0.360290 Top1 82.046574 Top5 97.435225 -2023-02-13 17:37:31,556 - ==> Top1: 82.047 Top5: 97.435 Loss: 0.360 - -2023-02-13 17:37:31,557 - ==> Confusion: -[[ 819 7 9 2 12 1 0 3 8 66 1 4 0 3 10 2 3 2 4 1 10] - [ 0 947 1 3 8 18 5 16 4 2 3 1 2 0 1 3 1 2 7 1 8] - [ 3 6 962 7 3 2 15 11 0 1 5 1 2 2 3 8 1 6 9 5 6] - [ 3 4 32 895 2 2 2 1 0 2 13 0 3 2 18 2 3 4 20 0 8] - [ 17 15 3 1 969 8 0 1 2 5 1 5 2 2 9 7 6 3 1 3 6] - [ 4 39 2 6 4 919 4 31 2 6 3 8 4 10 3 2 3 2 6 5 7] - [ 3 7 29 3 0 3 1024 2 0 0 2 2 2 1 0 5 1 2 2 7 4] - [ 0 18 18 6 3 20 1 905 0 1 1 6 3 1 0 0 0 2 25 10 4] - [ 18 3 1 2 1 1 0 2 880 34 11 1 0 6 30 2 0 4 12 0 1] - [ 61 3 2 2 5 2 0 0 49 859 1 0 0 10 7 1 0 2 0 1 7] - [ 2 5 2 6 1 2 3 6 16 0 958 0 1 10 6 0 2 1 27 0 3] - [ 0 6 2 0 1 17 0 5 3 2 0 881 40 2 2 9 4 12 7 8 4] - [ 0 0 1 6 0 3 0 3 2 0 1 23 863 0 8 8 2 23 3 1 12] - [ 8 6 2 0 4 7 1 0 14 15 8 6 5 919 9 5 4 4 1 4 2] - [ 10 2 3 22 3 3 0 0 18 4 5 2 1 0 984 1 1 7 18 0 8] - [ 2 2 7 2 6 0 4 1 0 0 0 0 9 2 1 965 7 19 0 9 10] - [ 1 12 1 3 9 4 0 0 4 1 0 2 2 2 6 21 965 3 6 3 16] - [ 4 1 3 7 0 0 1 0 1 2 1 3 15 1 4 21 0 975 3 1 8] - [ 3 7 11 11 1 1 0 20 3 2 3 1 4 0 9 1 0 1 1007 0 1] - [ 0 6 1 1 0 5 7 21 0 0 1 25 5 3 0 6 6 3 1 1043 14] - [ 169 306 321 194 139 191 85 172 119 120 209 114 340 348 218 120 198 120 291 248 9412]] - -2023-02-13 17:37:31,558 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:37:31,559 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:37:31,564 - - -2023-02-13 17:37:31,565 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:37:32,546 - Epoch: [43][ 10/ 1207] Overall Loss 0.350947 Objective Loss 0.350947 LR 0.001000 Time 0.098073 -2023-02-13 17:37:32,744 - Epoch: [43][ 20/ 1207] Overall Loss 0.358699 Objective Loss 0.358699 LR 0.001000 Time 0.058929 -2023-02-13 17:37:32,932 - Epoch: [43][ 30/ 1207] Overall Loss 0.348321 Objective Loss 0.348321 LR 0.001000 Time 0.045541 -2023-02-13 17:37:33,119 - Epoch: [43][ 40/ 1207] Overall Loss 0.341925 Objective Loss 0.341925 LR 0.001000 Time 0.038825 -2023-02-13 17:37:33,306 - Epoch: [43][ 50/ 1207] Overall Loss 0.341381 Objective Loss 0.341381 LR 0.001000 Time 0.034787 -2023-02-13 17:37:33,494 - Epoch: [43][ 60/ 1207] Overall Loss 0.338313 Objective Loss 0.338313 LR 0.001000 Time 0.032121 -2023-02-13 17:37:33,681 - Epoch: [43][ 70/ 1207] Overall Loss 0.337223 Objective Loss 0.337223 LR 0.001000 Time 0.030192 -2023-02-13 17:37:33,869 - Epoch: [43][ 80/ 1207] Overall Loss 0.337728 Objective Loss 0.337728 LR 0.001000 Time 0.028767 -2023-02-13 17:37:34,056 - Epoch: [43][ 90/ 1207] Overall Loss 0.339716 Objective Loss 0.339716 LR 0.001000 Time 0.027644 -2023-02-13 17:37:34,243 - Epoch: [43][ 100/ 1207] Overall Loss 0.340458 Objective Loss 0.340458 LR 0.001000 Time 0.026749 -2023-02-13 17:37:34,430 - Epoch: [43][ 110/ 1207] Overall Loss 0.337125 Objective Loss 0.337125 LR 0.001000 Time 0.026013 -2023-02-13 17:37:34,618 - Epoch: [43][ 120/ 1207] Overall Loss 0.336303 Objective Loss 0.336303 LR 0.001000 Time 0.025403 -2023-02-13 17:37:34,805 - Epoch: [43][ 130/ 1207] Overall Loss 0.336149 Objective Loss 0.336149 LR 0.001000 Time 0.024886 -2023-02-13 17:37:34,993 - Epoch: [43][ 140/ 1207] Overall Loss 0.334624 Objective Loss 0.334624 LR 0.001000 Time 0.024450 -2023-02-13 17:37:35,180 - Epoch: [43][ 150/ 1207] Overall Loss 0.335384 Objective Loss 0.335384 LR 0.001000 Time 0.024065 -2023-02-13 17:37:35,368 - Epoch: [43][ 160/ 1207] Overall Loss 0.335938 Objective Loss 0.335938 LR 0.001000 Time 0.023732 -2023-02-13 17:37:35,555 - Epoch: [43][ 170/ 1207] Overall Loss 0.335237 Objective Loss 0.335237 LR 0.001000 Time 0.023437 -2023-02-13 17:37:35,743 - Epoch: [43][ 180/ 1207] Overall Loss 0.334628 Objective Loss 0.334628 LR 0.001000 Time 0.023173 -2023-02-13 17:37:35,931 - Epoch: [43][ 190/ 1207] Overall Loss 0.334205 Objective Loss 0.334205 LR 0.001000 Time 0.022944 -2023-02-13 17:37:36,119 - Epoch: [43][ 200/ 1207] Overall Loss 0.333506 Objective Loss 0.333506 LR 0.001000 Time 0.022733 -2023-02-13 17:37:36,306 - Epoch: [43][ 210/ 1207] Overall Loss 0.332502 Objective Loss 0.332502 LR 0.001000 Time 0.022540 -2023-02-13 17:37:36,494 - Epoch: [43][ 220/ 1207] Overall Loss 0.333068 Objective Loss 0.333068 LR 0.001000 Time 0.022368 -2023-02-13 17:37:36,681 - Epoch: [43][ 230/ 1207] Overall Loss 0.333345 Objective Loss 0.333345 LR 0.001000 Time 0.022208 -2023-02-13 17:37:36,870 - Epoch: [43][ 240/ 1207] Overall Loss 0.333208 Objective Loss 0.333208 LR 0.001000 Time 0.022070 -2023-02-13 17:37:37,057 - Epoch: [43][ 250/ 1207] Overall Loss 0.333505 Objective Loss 0.333505 LR 0.001000 Time 0.021934 -2023-02-13 17:37:37,245 - Epoch: [43][ 260/ 1207] Overall Loss 0.333892 Objective Loss 0.333892 LR 0.001000 Time 0.021811 -2023-02-13 17:37:37,432 - Epoch: [43][ 270/ 1207] Overall Loss 0.332869 Objective Loss 0.332869 LR 0.001000 Time 0.021694 -2023-02-13 17:37:37,620 - Epoch: [43][ 280/ 1207] Overall Loss 0.332364 Objective Loss 0.332364 LR 0.001000 Time 0.021588 -2023-02-13 17:37:37,807 - Epoch: [43][ 290/ 1207] Overall Loss 0.332081 Objective Loss 0.332081 LR 0.001000 Time 0.021489 -2023-02-13 17:37:37,995 - Epoch: [43][ 300/ 1207] Overall Loss 0.332062 Objective Loss 0.332062 LR 0.001000 Time 0.021399 -2023-02-13 17:37:38,183 - Epoch: [43][ 310/ 1207] Overall Loss 0.331439 Objective Loss 0.331439 LR 0.001000 Time 0.021312 -2023-02-13 17:37:38,370 - Epoch: [43][ 320/ 1207] Overall Loss 0.331274 Objective Loss 0.331274 LR 0.001000 Time 0.021231 -2023-02-13 17:37:38,557 - Epoch: [43][ 330/ 1207] Overall Loss 0.331545 Objective Loss 0.331545 LR 0.001000 Time 0.021153 -2023-02-13 17:37:38,747 - Epoch: [43][ 340/ 1207] Overall Loss 0.330955 Objective Loss 0.330955 LR 0.001000 Time 0.021087 -2023-02-13 17:37:38,936 - Epoch: [43][ 350/ 1207] Overall Loss 0.330782 Objective Loss 0.330782 LR 0.001000 Time 0.021024 -2023-02-13 17:37:39,125 - Epoch: [43][ 360/ 1207] Overall Loss 0.331298 Objective Loss 0.331298 LR 0.001000 Time 0.020963 -2023-02-13 17:37:39,313 - Epoch: [43][ 370/ 1207] Overall Loss 0.330800 Objective Loss 0.330800 LR 0.001000 Time 0.020904 -2023-02-13 17:37:39,503 - Epoch: [43][ 380/ 1207] Overall Loss 0.330805 Objective Loss 0.330805 LR 0.001000 Time 0.020852 -2023-02-13 17:37:39,692 - Epoch: [43][ 390/ 1207] Overall Loss 0.330785 Objective Loss 0.330785 LR 0.001000 Time 0.020803 -2023-02-13 17:37:39,884 - Epoch: [43][ 400/ 1207] Overall Loss 0.330930 Objective Loss 0.330930 LR 0.001000 Time 0.020762 -2023-02-13 17:37:40,074 - Epoch: [43][ 410/ 1207] Overall Loss 0.330608 Objective Loss 0.330608 LR 0.001000 Time 0.020716 -2023-02-13 17:37:40,265 - Epoch: [43][ 420/ 1207] Overall Loss 0.330280 Objective Loss 0.330280 LR 0.001000 Time 0.020677 -2023-02-13 17:37:40,454 - Epoch: [43][ 430/ 1207] Overall Loss 0.330411 Objective Loss 0.330411 LR 0.001000 Time 0.020635 -2023-02-13 17:37:40,645 - Epoch: [43][ 440/ 1207] Overall Loss 0.330699 Objective Loss 0.330699 LR 0.001000 Time 0.020600 -2023-02-13 17:37:40,836 - Epoch: [43][ 450/ 1207] Overall Loss 0.330917 Objective Loss 0.330917 LR 0.001000 Time 0.020565 -2023-02-13 17:37:41,028 - Epoch: [43][ 460/ 1207] Overall Loss 0.330692 Objective Loss 0.330692 LR 0.001000 Time 0.020535 -2023-02-13 17:37:41,218 - Epoch: [43][ 470/ 1207] Overall Loss 0.331309 Objective Loss 0.331309 LR 0.001000 Time 0.020502 -2023-02-13 17:37:41,410 - Epoch: [43][ 480/ 1207] Overall Loss 0.331218 Objective Loss 0.331218 LR 0.001000 Time 0.020473 -2023-02-13 17:37:41,599 - Epoch: [43][ 490/ 1207] Overall Loss 0.330985 Objective Loss 0.330985 LR 0.001000 Time 0.020442 -2023-02-13 17:37:41,791 - Epoch: [43][ 500/ 1207] Overall Loss 0.331933 Objective Loss 0.331933 LR 0.001000 Time 0.020417 -2023-02-13 17:37:41,982 - Epoch: [43][ 510/ 1207] Overall Loss 0.332260 Objective Loss 0.332260 LR 0.001000 Time 0.020390 -2023-02-13 17:37:42,175 - Epoch: [43][ 520/ 1207] Overall Loss 0.332610 Objective Loss 0.332610 LR 0.001000 Time 0.020367 -2023-02-13 17:37:42,365 - Epoch: [43][ 530/ 1207] Overall Loss 0.332270 Objective Loss 0.332270 LR 0.001000 Time 0.020342 -2023-02-13 17:37:42,558 - Epoch: [43][ 540/ 1207] Overall Loss 0.332318 Objective Loss 0.332318 LR 0.001000 Time 0.020320 -2023-02-13 17:37:42,748 - Epoch: [43][ 550/ 1207] Overall Loss 0.333358 Objective Loss 0.333358 LR 0.001000 Time 0.020296 -2023-02-13 17:37:42,940 - Epoch: [43][ 560/ 1207] Overall Loss 0.334092 Objective Loss 0.334092 LR 0.001000 Time 0.020276 -2023-02-13 17:37:43,130 - Epoch: [43][ 570/ 1207] Overall Loss 0.333838 Objective Loss 0.333838 LR 0.001000 Time 0.020254 -2023-02-13 17:37:43,322 - Epoch: [43][ 580/ 1207] Overall Loss 0.333489 Objective Loss 0.333489 LR 0.001000 Time 0.020234 -2023-02-13 17:37:43,513 - Epoch: [43][ 590/ 1207] Overall Loss 0.333913 Objective Loss 0.333913 LR 0.001000 Time 0.020214 -2023-02-13 17:37:43,705 - Epoch: [43][ 600/ 1207] Overall Loss 0.334746 Objective Loss 0.334746 LR 0.001000 Time 0.020196 -2023-02-13 17:37:43,896 - Epoch: [43][ 610/ 1207] Overall Loss 0.334906 Objective Loss 0.334906 LR 0.001000 Time 0.020178 -2023-02-13 17:37:44,087 - Epoch: [43][ 620/ 1207] Overall Loss 0.335245 Objective Loss 0.335245 LR 0.001000 Time 0.020161 -2023-02-13 17:37:44,277 - Epoch: [43][ 630/ 1207] Overall Loss 0.335759 Objective Loss 0.335759 LR 0.001000 Time 0.020142 -2023-02-13 17:37:44,469 - Epoch: [43][ 640/ 1207] Overall Loss 0.335992 Objective Loss 0.335992 LR 0.001000 Time 0.020126 -2023-02-13 17:37:44,660 - Epoch: [43][ 650/ 1207] Overall Loss 0.335860 Objective Loss 0.335860 LR 0.001000 Time 0.020110 -2023-02-13 17:37:44,851 - Epoch: [43][ 660/ 1207] Overall Loss 0.335898 Objective Loss 0.335898 LR 0.001000 Time 0.020094 -2023-02-13 17:37:45,042 - Epoch: [43][ 670/ 1207] Overall Loss 0.336130 Objective Loss 0.336130 LR 0.001000 Time 0.020078 -2023-02-13 17:37:45,234 - Epoch: [43][ 680/ 1207] Overall Loss 0.336473 Objective Loss 0.336473 LR 0.001000 Time 0.020065 -2023-02-13 17:37:45,425 - Epoch: [43][ 690/ 1207] Overall Loss 0.337125 Objective Loss 0.337125 LR 0.001000 Time 0.020051 -2023-02-13 17:37:45,616 - Epoch: [43][ 700/ 1207] Overall Loss 0.337497 Objective Loss 0.337497 LR 0.001000 Time 0.020036 -2023-02-13 17:37:45,809 - Epoch: [43][ 710/ 1207] Overall Loss 0.337722 Objective Loss 0.337722 LR 0.001000 Time 0.020025 -2023-02-13 17:37:46,000 - Epoch: [43][ 720/ 1207] Overall Loss 0.337694 Objective Loss 0.337694 LR 0.001000 Time 0.020012 -2023-02-13 17:37:46,191 - Epoch: [43][ 730/ 1207] Overall Loss 0.338085 Objective Loss 0.338085 LR 0.001000 Time 0.019999 -2023-02-13 17:37:46,382 - Epoch: [43][ 740/ 1207] Overall Loss 0.338093 Objective Loss 0.338093 LR 0.001000 Time 0.019986 -2023-02-13 17:37:46,573 - Epoch: [43][ 750/ 1207] Overall Loss 0.338092 Objective Loss 0.338092 LR 0.001000 Time 0.019973 -2023-02-13 17:37:46,763 - Epoch: [43][ 760/ 1207] Overall Loss 0.338167 Objective Loss 0.338167 LR 0.001000 Time 0.019961 -2023-02-13 17:37:46,956 - Epoch: [43][ 770/ 1207] Overall Loss 0.338256 Objective Loss 0.338256 LR 0.001000 Time 0.019951 -2023-02-13 17:37:47,147 - Epoch: [43][ 780/ 1207] Overall Loss 0.338146 Objective Loss 0.338146 LR 0.001000 Time 0.019940 -2023-02-13 17:37:47,337 - Epoch: [43][ 790/ 1207] Overall Loss 0.338357 Objective Loss 0.338357 LR 0.001000 Time 0.019928 -2023-02-13 17:37:47,527 - Epoch: [43][ 800/ 1207] Overall Loss 0.338429 Objective Loss 0.338429 LR 0.001000 Time 0.019916 -2023-02-13 17:37:47,718 - Epoch: [43][ 810/ 1207] Overall Loss 0.338297 Objective Loss 0.338297 LR 0.001000 Time 0.019906 -2023-02-13 17:37:47,910 - Epoch: [43][ 820/ 1207] Overall Loss 0.338000 Objective Loss 0.338000 LR 0.001000 Time 0.019896 -2023-02-13 17:37:48,101 - Epoch: [43][ 830/ 1207] Overall Loss 0.338340 Objective Loss 0.338340 LR 0.001000 Time 0.019886 -2023-02-13 17:37:48,291 - Epoch: [43][ 840/ 1207] Overall Loss 0.338476 Objective Loss 0.338476 LR 0.001000 Time 0.019875 -2023-02-13 17:37:48,482 - Epoch: [43][ 850/ 1207] Overall Loss 0.338287 Objective Loss 0.338287 LR 0.001000 Time 0.019866 -2023-02-13 17:37:48,673 - Epoch: [43][ 860/ 1207] Overall Loss 0.338165 Objective Loss 0.338165 LR 0.001000 Time 0.019856 -2023-02-13 17:37:48,864 - Epoch: [43][ 870/ 1207] Overall Loss 0.338116 Objective Loss 0.338116 LR 0.001000 Time 0.019847 -2023-02-13 17:37:49,054 - Epoch: [43][ 880/ 1207] Overall Loss 0.337800 Objective Loss 0.337800 LR 0.001000 Time 0.019837 -2023-02-13 17:37:49,246 - Epoch: [43][ 890/ 1207] Overall Loss 0.337590 Objective Loss 0.337590 LR 0.001000 Time 0.019829 -2023-02-13 17:37:49,436 - Epoch: [43][ 900/ 1207] Overall Loss 0.337446 Objective Loss 0.337446 LR 0.001000 Time 0.019819 -2023-02-13 17:37:49,626 - Epoch: [43][ 910/ 1207] Overall Loss 0.337905 Objective Loss 0.337905 LR 0.001000 Time 0.019811 -2023-02-13 17:37:49,816 - Epoch: [43][ 920/ 1207] Overall Loss 0.338128 Objective Loss 0.338128 LR 0.001000 Time 0.019802 -2023-02-13 17:37:50,008 - Epoch: [43][ 930/ 1207] Overall Loss 0.338174 Objective Loss 0.338174 LR 0.001000 Time 0.019795 -2023-02-13 17:37:50,199 - Epoch: [43][ 940/ 1207] Overall Loss 0.338283 Objective Loss 0.338283 LR 0.001000 Time 0.019787 -2023-02-13 17:37:50,390 - Epoch: [43][ 950/ 1207] Overall Loss 0.338458 Objective Loss 0.338458 LR 0.001000 Time 0.019779 -2023-02-13 17:37:50,581 - Epoch: [43][ 960/ 1207] Overall Loss 0.338729 Objective Loss 0.338729 LR 0.001000 Time 0.019772 -2023-02-13 17:37:50,774 - Epoch: [43][ 970/ 1207] Overall Loss 0.338432 Objective Loss 0.338432 LR 0.001000 Time 0.019766 -2023-02-13 17:37:50,965 - Epoch: [43][ 980/ 1207] Overall Loss 0.338717 Objective Loss 0.338717 LR 0.001000 Time 0.019760 -2023-02-13 17:37:51,156 - Epoch: [43][ 990/ 1207] Overall Loss 0.338652 Objective Loss 0.338652 LR 0.001000 Time 0.019753 -2023-02-13 17:37:51,347 - Epoch: [43][ 1000/ 1207] Overall Loss 0.338778 Objective Loss 0.338778 LR 0.001000 Time 0.019745 -2023-02-13 17:37:51,538 - Epoch: [43][ 1010/ 1207] Overall Loss 0.338873 Objective Loss 0.338873 LR 0.001000 Time 0.019738 -2023-02-13 17:37:51,729 - Epoch: [43][ 1020/ 1207] Overall Loss 0.338857 Objective Loss 0.338857 LR 0.001000 Time 0.019732 -2023-02-13 17:37:51,921 - Epoch: [43][ 1030/ 1207] Overall Loss 0.338910 Objective Loss 0.338910 LR 0.001000 Time 0.019726 -2023-02-13 17:37:52,112 - Epoch: [43][ 1040/ 1207] Overall Loss 0.339093 Objective Loss 0.339093 LR 0.001000 Time 0.019720 -2023-02-13 17:37:52,303 - Epoch: [43][ 1050/ 1207] Overall Loss 0.339155 Objective Loss 0.339155 LR 0.001000 Time 0.019714 -2023-02-13 17:37:52,494 - Epoch: [43][ 1060/ 1207] Overall Loss 0.339426 Objective Loss 0.339426 LR 0.001000 Time 0.019708 -2023-02-13 17:37:52,685 - Epoch: [43][ 1070/ 1207] Overall Loss 0.339381 Objective Loss 0.339381 LR 0.001000 Time 0.019702 -2023-02-13 17:37:52,876 - Epoch: [43][ 1080/ 1207] Overall Loss 0.338937 Objective Loss 0.338937 LR 0.001000 Time 0.019696 -2023-02-13 17:37:53,067 - Epoch: [43][ 1090/ 1207] Overall Loss 0.338896 Objective Loss 0.338896 LR 0.001000 Time 0.019690 -2023-02-13 17:37:53,257 - Epoch: [43][ 1100/ 1207] Overall Loss 0.338970 Objective Loss 0.338970 LR 0.001000 Time 0.019683 -2023-02-13 17:37:53,448 - Epoch: [43][ 1110/ 1207] Overall Loss 0.338878 Objective Loss 0.338878 LR 0.001000 Time 0.019678 -2023-02-13 17:37:53,640 - Epoch: [43][ 1120/ 1207] Overall Loss 0.338791 Objective Loss 0.338791 LR 0.001000 Time 0.019673 -2023-02-13 17:37:53,830 - Epoch: [43][ 1130/ 1207] Overall Loss 0.338710 Objective Loss 0.338710 LR 0.001000 Time 0.019667 -2023-02-13 17:37:54,022 - Epoch: [43][ 1140/ 1207] Overall Loss 0.338632 Objective Loss 0.338632 LR 0.001000 Time 0.019662 -2023-02-13 17:37:54,213 - Epoch: [43][ 1150/ 1207] Overall Loss 0.338855 Objective Loss 0.338855 LR 0.001000 Time 0.019657 -2023-02-13 17:37:54,404 - Epoch: [43][ 1160/ 1207] Overall Loss 0.338826 Objective Loss 0.338826 LR 0.001000 Time 0.019652 -2023-02-13 17:37:54,595 - Epoch: [43][ 1170/ 1207] Overall Loss 0.339102 Objective Loss 0.339102 LR 0.001000 Time 0.019647 -2023-02-13 17:37:54,785 - Epoch: [43][ 1180/ 1207] Overall Loss 0.339382 Objective Loss 0.339382 LR 0.001000 Time 0.019641 -2023-02-13 17:37:54,977 - Epoch: [43][ 1190/ 1207] Overall Loss 0.339512 Objective Loss 0.339512 LR 0.001000 Time 0.019637 -2023-02-13 17:37:55,224 - Epoch: [43][ 1200/ 1207] Overall Loss 0.339680 Objective Loss 0.339680 LR 0.001000 Time 0.019679 -2023-02-13 17:37:55,340 - Epoch: [43][ 1207/ 1207] Overall Loss 0.339830 Objective Loss 0.339830 Top1 74.695122 Top5 96.951220 LR 0.001000 Time 0.019661 -2023-02-13 17:37:55,423 - --- validate (epoch=43)----------- -2023-02-13 17:37:55,423 - 34311 samples (256 per mini-batch) -2023-02-13 17:37:55,843 - Epoch: [43][ 10/ 135] Loss 0.365628 Top1 80.820312 Top5 96.640625 -2023-02-13 17:37:55,973 - Epoch: [43][ 20/ 135] Loss 0.370717 Top1 80.488281 Top5 97.011719 -2023-02-13 17:37:56,101 - Epoch: [43][ 30/ 135] Loss 0.383411 Top1 80.013021 Top5 96.731771 -2023-02-13 17:37:56,233 - Epoch: [43][ 40/ 135] Loss 0.377939 Top1 80.205078 Top5 96.865234 -2023-02-13 17:37:56,361 - Epoch: [43][ 50/ 135] Loss 0.373316 Top1 80.203125 Top5 96.953125 -2023-02-13 17:37:56,491 - Epoch: [43][ 60/ 135] Loss 0.369611 Top1 80.390625 Top5 97.031250 -2023-02-13 17:37:56,619 - Epoch: [43][ 70/ 135] Loss 0.369654 Top1 80.530134 Top5 97.008929 -2023-02-13 17:37:56,748 - Epoch: [43][ 80/ 135] Loss 0.367125 Top1 80.566406 Top5 97.045898 -2023-02-13 17:37:56,877 - Epoch: [43][ 90/ 135] Loss 0.365105 Top1 80.451389 Top5 97.031250 -2023-02-13 17:37:57,006 - Epoch: [43][ 100/ 135] Loss 0.368859 Top1 80.292969 Top5 97.015625 -2023-02-13 17:37:57,133 - Epoch: [43][ 110/ 135] Loss 0.365083 Top1 80.436790 Top5 97.070312 -2023-02-13 17:37:57,264 - Epoch: [43][ 120/ 135] Loss 0.368554 Top1 80.309245 Top5 97.070312 -2023-02-13 17:37:57,397 - Epoch: [43][ 130/ 135] Loss 0.369714 Top1 80.168269 Top5 97.061298 -2023-02-13 17:37:57,442 - Epoch: [43][ 135/ 135] Loss 0.368683 Top1 80.146309 Top5 97.041765 -2023-02-13 17:37:57,511 - ==> Top1: 80.146 Top5: 97.042 Loss: 0.369 - -2023-02-13 17:37:57,512 - ==> Confusion: -[[ 776 5 4 0 13 3 0 2 6 114 2 6 3 3 6 5 5 3 3 1 7] - [ 5 912 1 2 12 40 2 25 5 2 3 2 2 0 0 1 7 0 5 3 4] - [ 8 0 905 32 5 3 24 20 0 1 5 0 2 4 2 10 4 4 15 9 5] - [ 8 1 11 895 1 3 1 4 1 1 13 1 13 1 23 4 3 5 23 1 3] - [ 14 9 3 2 981 11 1 1 1 6 0 7 0 2 9 6 7 0 1 3 2] - [ 5 26 0 6 3 951 3 31 0 1 2 5 1 10 1 3 7 0 6 3 6] - [ 2 5 16 5 0 8 1028 7 0 2 2 4 4 0 0 7 2 0 2 4 1] - [ 3 5 11 5 2 34 3 909 1 1 3 5 3 0 1 0 0 0 25 10 3] - [ 12 1 1 2 1 1 0 1 908 34 9 3 1 7 15 2 0 2 7 1 1] - [ 39 3 2 0 2 3 1 5 46 878 0 2 1 15 6 0 1 2 2 1 3] - [ 2 3 2 8 1 5 3 5 16 1 970 2 2 8 6 0 2 1 11 0 3] - [ 3 4 3 0 1 21 2 7 3 0 1 865 39 6 2 11 1 2 3 28 3] - [ 4 0 1 3 0 0 0 2 3 0 2 31 886 1 2 4 0 8 3 2 7] - [ 6 5 2 0 6 17 1 2 22 16 13 4 1 900 6 8 6 1 0 3 5] - [ 10 1 1 13 3 3 0 2 26 8 3 1 6 0 979 2 3 4 17 0 10] - [ 4 0 3 3 4 1 5 1 0 0 0 4 13 2 0 966 9 15 1 10 5] - [ 0 6 1 0 8 3 0 0 2 1 0 2 2 1 3 20 995 2 3 5 7] - [ 5 2 0 3 0 1 2 3 0 1 0 12 31 1 0 25 0 957 0 2 6] - [ 5 3 4 9 2 4 1 32 9 0 4 0 5 0 11 1 0 1 993 1 1] - [ 0 2 2 0 2 12 12 22 1 0 1 10 2 2 0 5 0 3 2 1067 3] - [ 159 248 178 210 152 311 115 244 134 130 237 138 446 365 208 209 351 129 283 409 8778]] - -2023-02-13 17:37:57,513 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:37:57,513 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:37:57,519 - - -2023-02-13 17:37:57,519 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:37:58,404 - Epoch: [44][ 10/ 1207] Overall Loss 0.321648 Objective Loss 0.321648 LR 0.001000 Time 0.088443 -2023-02-13 17:37:58,603 - Epoch: [44][ 20/ 1207] Overall Loss 0.321904 Objective Loss 0.321904 LR 0.001000 Time 0.054124 -2023-02-13 17:37:58,793 - Epoch: [44][ 30/ 1207] Overall Loss 0.319821 Objective Loss 0.319821 LR 0.001000 Time 0.042402 -2023-02-13 17:37:58,982 - Epoch: [44][ 40/ 1207] Overall Loss 0.332401 Objective Loss 0.332401 LR 0.001000 Time 0.036522 -2023-02-13 17:37:59,190 - Epoch: [44][ 50/ 1207] Overall Loss 0.331226 Objective Loss 0.331226 LR 0.001000 Time 0.033373 -2023-02-13 17:37:59,387 - Epoch: [44][ 60/ 1207] Overall Loss 0.338134 Objective Loss 0.338134 LR 0.001000 Time 0.031089 -2023-02-13 17:37:59,583 - Epoch: [44][ 70/ 1207] Overall Loss 0.342868 Objective Loss 0.342868 LR 0.001000 Time 0.029437 -2023-02-13 17:37:59,780 - Epoch: [44][ 80/ 1207] Overall Loss 0.346772 Objective Loss 0.346772 LR 0.001000 Time 0.028218 -2023-02-13 17:37:59,975 - Epoch: [44][ 90/ 1207] Overall Loss 0.345904 Objective Loss 0.345904 LR 0.001000 Time 0.027250 -2023-02-13 17:38:00,172 - Epoch: [44][ 100/ 1207] Overall Loss 0.346088 Objective Loss 0.346088 LR 0.001000 Time 0.026492 -2023-02-13 17:38:00,367 - Epoch: [44][ 110/ 1207] Overall Loss 0.345761 Objective Loss 0.345761 LR 0.001000 Time 0.025851 -2023-02-13 17:38:00,565 - Epoch: [44][ 120/ 1207] Overall Loss 0.343338 Objective Loss 0.343338 LR 0.001000 Time 0.025340 -2023-02-13 17:38:00,759 - Epoch: [44][ 130/ 1207] Overall Loss 0.341763 Objective Loss 0.341763 LR 0.001000 Time 0.024884 -2023-02-13 17:38:00,959 - Epoch: [44][ 140/ 1207] Overall Loss 0.343148 Objective Loss 0.343148 LR 0.001000 Time 0.024532 -2023-02-13 17:38:01,154 - Epoch: [44][ 150/ 1207] Overall Loss 0.343528 Objective Loss 0.343528 LR 0.001000 Time 0.024191 -2023-02-13 17:38:01,351 - Epoch: [44][ 160/ 1207] Overall Loss 0.344658 Objective Loss 0.344658 LR 0.001000 Time 0.023908 -2023-02-13 17:38:01,546 - Epoch: [44][ 170/ 1207] Overall Loss 0.343605 Objective Loss 0.343605 LR 0.001000 Time 0.023648 -2023-02-13 17:38:01,743 - Epoch: [44][ 180/ 1207] Overall Loss 0.343525 Objective Loss 0.343525 LR 0.001000 Time 0.023429 -2023-02-13 17:38:01,939 - Epoch: [44][ 190/ 1207] Overall Loss 0.342815 Objective Loss 0.342815 LR 0.001000 Time 0.023225 -2023-02-13 17:38:02,137 - Epoch: [44][ 200/ 1207] Overall Loss 0.341616 Objective Loss 0.341616 LR 0.001000 Time 0.023049 -2023-02-13 17:38:02,331 - Epoch: [44][ 210/ 1207] Overall Loss 0.341396 Objective Loss 0.341396 LR 0.001000 Time 0.022877 -2023-02-13 17:38:02,529 - Epoch: [44][ 220/ 1207] Overall Loss 0.340685 Objective Loss 0.340685 LR 0.001000 Time 0.022735 -2023-02-13 17:38:02,725 - Epoch: [44][ 230/ 1207] Overall Loss 0.340045 Objective Loss 0.340045 LR 0.001000 Time 0.022595 -2023-02-13 17:38:02,923 - Epoch: [44][ 240/ 1207] Overall Loss 0.339111 Objective Loss 0.339111 LR 0.001000 Time 0.022477 -2023-02-13 17:38:03,118 - Epoch: [44][ 250/ 1207] Overall Loss 0.338688 Objective Loss 0.338688 LR 0.001000 Time 0.022356 -2023-02-13 17:38:03,315 - Epoch: [44][ 260/ 1207] Overall Loss 0.337680 Objective Loss 0.337680 LR 0.001000 Time 0.022254 -2023-02-13 17:38:03,511 - Epoch: [44][ 270/ 1207] Overall Loss 0.337402 Objective Loss 0.337402 LR 0.001000 Time 0.022154 -2023-02-13 17:38:03,708 - Epoch: [44][ 280/ 1207] Overall Loss 0.337400 Objective Loss 0.337400 LR 0.001000 Time 0.022065 -2023-02-13 17:38:03,903 - Epoch: [44][ 290/ 1207] Overall Loss 0.338080 Objective Loss 0.338080 LR 0.001000 Time 0.021975 -2023-02-13 17:38:04,101 - Epoch: [44][ 300/ 1207] Overall Loss 0.338024 Objective Loss 0.338024 LR 0.001000 Time 0.021902 -2023-02-13 17:38:04,296 - Epoch: [44][ 310/ 1207] Overall Loss 0.337153 Objective Loss 0.337153 LR 0.001000 Time 0.021823 -2023-02-13 17:38:04,494 - Epoch: [44][ 320/ 1207] Overall Loss 0.337303 Objective Loss 0.337303 LR 0.001000 Time 0.021758 -2023-02-13 17:38:04,689 - Epoch: [44][ 330/ 1207] Overall Loss 0.336873 Objective Loss 0.336873 LR 0.001000 Time 0.021689 -2023-02-13 17:38:04,887 - Epoch: [44][ 340/ 1207] Overall Loss 0.337228 Objective Loss 0.337228 LR 0.001000 Time 0.021633 -2023-02-13 17:38:05,082 - Epoch: [44][ 350/ 1207] Overall Loss 0.336910 Objective Loss 0.336910 LR 0.001000 Time 0.021571 -2023-02-13 17:38:05,280 - Epoch: [44][ 360/ 1207] Overall Loss 0.336482 Objective Loss 0.336482 LR 0.001000 Time 0.021520 -2023-02-13 17:38:05,475 - Epoch: [44][ 370/ 1207] Overall Loss 0.337612 Objective Loss 0.337612 LR 0.001000 Time 0.021465 -2023-02-13 17:38:05,672 - Epoch: [44][ 380/ 1207] Overall Loss 0.338227 Objective Loss 0.338227 LR 0.001000 Time 0.021418 -2023-02-13 17:38:05,869 - Epoch: [44][ 390/ 1207] Overall Loss 0.337698 Objective Loss 0.337698 LR 0.001000 Time 0.021372 -2023-02-13 17:38:06,067 - Epoch: [44][ 400/ 1207] Overall Loss 0.336848 Objective Loss 0.336848 LR 0.001000 Time 0.021332 -2023-02-13 17:38:06,262 - Epoch: [44][ 410/ 1207] Overall Loss 0.335880 Objective Loss 0.335880 LR 0.001000 Time 0.021286 -2023-02-13 17:38:06,459 - Epoch: [44][ 420/ 1207] Overall Loss 0.336086 Objective Loss 0.336086 LR 0.001000 Time 0.021248 -2023-02-13 17:38:06,654 - Epoch: [44][ 430/ 1207] Overall Loss 0.335868 Objective Loss 0.335868 LR 0.001000 Time 0.021207 -2023-02-13 17:38:06,853 - Epoch: [44][ 440/ 1207] Overall Loss 0.336210 Objective Loss 0.336210 LR 0.001000 Time 0.021176 -2023-02-13 17:38:07,048 - Epoch: [44][ 450/ 1207] Overall Loss 0.336521 Objective Loss 0.336521 LR 0.001000 Time 0.021139 -2023-02-13 17:38:07,246 - Epoch: [44][ 460/ 1207] Overall Loss 0.335913 Objective Loss 0.335913 LR 0.001000 Time 0.021108 -2023-02-13 17:38:07,441 - Epoch: [44][ 470/ 1207] Overall Loss 0.336247 Objective Loss 0.336247 LR 0.001000 Time 0.021074 -2023-02-13 17:38:07,639 - Epoch: [44][ 480/ 1207] Overall Loss 0.336121 Objective Loss 0.336121 LR 0.001000 Time 0.021045 -2023-02-13 17:38:07,834 - Epoch: [44][ 490/ 1207] Overall Loss 0.336206 Objective Loss 0.336206 LR 0.001000 Time 0.021013 -2023-02-13 17:38:08,032 - Epoch: [44][ 500/ 1207] Overall Loss 0.336850 Objective Loss 0.336850 LR 0.001000 Time 0.020989 -2023-02-13 17:38:08,227 - Epoch: [44][ 510/ 1207] Overall Loss 0.336982 Objective Loss 0.336982 LR 0.001000 Time 0.020958 -2023-02-13 17:38:08,424 - Epoch: [44][ 520/ 1207] Overall Loss 0.336685 Objective Loss 0.336685 LR 0.001000 Time 0.020933 -2023-02-13 17:38:08,619 - Epoch: [44][ 530/ 1207] Overall Loss 0.336180 Objective Loss 0.336180 LR 0.001000 Time 0.020906 -2023-02-13 17:38:08,817 - Epoch: [44][ 540/ 1207] Overall Loss 0.335924 Objective Loss 0.335924 LR 0.001000 Time 0.020885 -2023-02-13 17:38:09,013 - Epoch: [44][ 550/ 1207] Overall Loss 0.335810 Objective Loss 0.335810 LR 0.001000 Time 0.020861 -2023-02-13 17:38:09,211 - Epoch: [44][ 560/ 1207] Overall Loss 0.335899 Objective Loss 0.335899 LR 0.001000 Time 0.020840 -2023-02-13 17:38:09,406 - Epoch: [44][ 570/ 1207] Overall Loss 0.336438 Objective Loss 0.336438 LR 0.001000 Time 0.020817 -2023-02-13 17:38:09,604 - Epoch: [44][ 580/ 1207] Overall Loss 0.336262 Objective Loss 0.336262 LR 0.001000 Time 0.020798 -2023-02-13 17:38:09,799 - Epoch: [44][ 590/ 1207] Overall Loss 0.336514 Objective Loss 0.336514 LR 0.001000 Time 0.020776 -2023-02-13 17:38:09,997 - Epoch: [44][ 600/ 1207] Overall Loss 0.336769 Objective Loss 0.336769 LR 0.001000 Time 0.020759 -2023-02-13 17:38:10,193 - Epoch: [44][ 610/ 1207] Overall Loss 0.337134 Objective Loss 0.337134 LR 0.001000 Time 0.020739 -2023-02-13 17:38:10,391 - Epoch: [44][ 620/ 1207] Overall Loss 0.336904 Objective Loss 0.336904 LR 0.001000 Time 0.020724 -2023-02-13 17:38:10,586 - Epoch: [44][ 630/ 1207] Overall Loss 0.336706 Objective Loss 0.336706 LR 0.001000 Time 0.020704 -2023-02-13 17:38:10,784 - Epoch: [44][ 640/ 1207] Overall Loss 0.336568 Objective Loss 0.336568 LR 0.001000 Time 0.020690 -2023-02-13 17:38:10,981 - Epoch: [44][ 650/ 1207] Overall Loss 0.336264 Objective Loss 0.336264 LR 0.001000 Time 0.020673 -2023-02-13 17:38:11,179 - Epoch: [44][ 660/ 1207] Overall Loss 0.336106 Objective Loss 0.336106 LR 0.001000 Time 0.020660 -2023-02-13 17:38:11,374 - Epoch: [44][ 670/ 1207] Overall Loss 0.335780 Objective Loss 0.335780 LR 0.001000 Time 0.020642 -2023-02-13 17:38:11,573 - Epoch: [44][ 680/ 1207] Overall Loss 0.336159 Objective Loss 0.336159 LR 0.001000 Time 0.020630 -2023-02-13 17:38:11,769 - Epoch: [44][ 690/ 1207] Overall Loss 0.336008 Objective Loss 0.336008 LR 0.001000 Time 0.020614 -2023-02-13 17:38:11,967 - Epoch: [44][ 700/ 1207] Overall Loss 0.335928 Objective Loss 0.335928 LR 0.001000 Time 0.020602 -2023-02-13 17:38:12,162 - Epoch: [44][ 710/ 1207] Overall Loss 0.335925 Objective Loss 0.335925 LR 0.001000 Time 0.020587 -2023-02-13 17:38:12,360 - Epoch: [44][ 720/ 1207] Overall Loss 0.335974 Objective Loss 0.335974 LR 0.001000 Time 0.020575 -2023-02-13 17:38:12,556 - Epoch: [44][ 730/ 1207] Overall Loss 0.336139 Objective Loss 0.336139 LR 0.001000 Time 0.020562 -2023-02-13 17:38:12,754 - Epoch: [44][ 740/ 1207] Overall Loss 0.336091 Objective Loss 0.336091 LR 0.001000 Time 0.020550 -2023-02-13 17:38:12,949 - Epoch: [44][ 750/ 1207] Overall Loss 0.336169 Objective Loss 0.336169 LR 0.001000 Time 0.020536 -2023-02-13 17:38:13,147 - Epoch: [44][ 760/ 1207] Overall Loss 0.336095 Objective Loss 0.336095 LR 0.001000 Time 0.020526 -2023-02-13 17:38:13,342 - Epoch: [44][ 770/ 1207] Overall Loss 0.335844 Objective Loss 0.335844 LR 0.001000 Time 0.020512 -2023-02-13 17:38:13,540 - Epoch: [44][ 780/ 1207] Overall Loss 0.335695 Objective Loss 0.335695 LR 0.001000 Time 0.020503 -2023-02-13 17:38:13,736 - Epoch: [44][ 790/ 1207] Overall Loss 0.335986 Objective Loss 0.335986 LR 0.001000 Time 0.020490 -2023-02-13 17:38:13,934 - Epoch: [44][ 800/ 1207] Overall Loss 0.336216 Objective Loss 0.336216 LR 0.001000 Time 0.020481 -2023-02-13 17:38:14,131 - Epoch: [44][ 810/ 1207] Overall Loss 0.336152 Objective Loss 0.336152 LR 0.001000 Time 0.020471 -2023-02-13 17:38:14,329 - Epoch: [44][ 820/ 1207] Overall Loss 0.335915 Objective Loss 0.335915 LR 0.001000 Time 0.020463 -2023-02-13 17:38:14,525 - Epoch: [44][ 830/ 1207] Overall Loss 0.335814 Objective Loss 0.335814 LR 0.001000 Time 0.020452 -2023-02-13 17:38:14,723 - Epoch: [44][ 840/ 1207] Overall Loss 0.335668 Objective Loss 0.335668 LR 0.001000 Time 0.020444 -2023-02-13 17:38:14,919 - Epoch: [44][ 850/ 1207] Overall Loss 0.335873 Objective Loss 0.335873 LR 0.001000 Time 0.020433 -2023-02-13 17:38:15,117 - Epoch: [44][ 860/ 1207] Overall Loss 0.336156 Objective Loss 0.336156 LR 0.001000 Time 0.020425 -2023-02-13 17:38:15,312 - Epoch: [44][ 870/ 1207] Overall Loss 0.336366 Objective Loss 0.336366 LR 0.001000 Time 0.020415 -2023-02-13 17:38:15,510 - Epoch: [44][ 880/ 1207] Overall Loss 0.336821 Objective Loss 0.336821 LR 0.001000 Time 0.020408 -2023-02-13 17:38:15,705 - Epoch: [44][ 890/ 1207] Overall Loss 0.336912 Objective Loss 0.336912 LR 0.001000 Time 0.020397 -2023-02-13 17:38:15,904 - Epoch: [44][ 900/ 1207] Overall Loss 0.337154 Objective Loss 0.337154 LR 0.001000 Time 0.020391 -2023-02-13 17:38:16,099 - Epoch: [44][ 910/ 1207] Overall Loss 0.337688 Objective Loss 0.337688 LR 0.001000 Time 0.020381 -2023-02-13 17:38:16,298 - Epoch: [44][ 920/ 1207] Overall Loss 0.337868 Objective Loss 0.337868 LR 0.001000 Time 0.020375 -2023-02-13 17:38:16,494 - Epoch: [44][ 930/ 1207] Overall Loss 0.337819 Objective Loss 0.337819 LR 0.001000 Time 0.020366 -2023-02-13 17:38:16,692 - Epoch: [44][ 940/ 1207] Overall Loss 0.337909 Objective Loss 0.337909 LR 0.001000 Time 0.020360 -2023-02-13 17:38:16,888 - Epoch: [44][ 950/ 1207] Overall Loss 0.337987 Objective Loss 0.337987 LR 0.001000 Time 0.020351 -2023-02-13 17:38:17,086 - Epoch: [44][ 960/ 1207] Overall Loss 0.338236 Objective Loss 0.338236 LR 0.001000 Time 0.020346 -2023-02-13 17:38:17,282 - Epoch: [44][ 970/ 1207] Overall Loss 0.338202 Objective Loss 0.338202 LR 0.001000 Time 0.020337 -2023-02-13 17:38:17,480 - Epoch: [44][ 980/ 1207] Overall Loss 0.337930 Objective Loss 0.337930 LR 0.001000 Time 0.020331 -2023-02-13 17:38:17,675 - Epoch: [44][ 990/ 1207] Overall Loss 0.337811 Objective Loss 0.337811 LR 0.001000 Time 0.020323 -2023-02-13 17:38:17,873 - Epoch: [44][ 1000/ 1207] Overall Loss 0.337973 Objective Loss 0.337973 LR 0.001000 Time 0.020317 -2023-02-13 17:38:18,069 - Epoch: [44][ 1010/ 1207] Overall Loss 0.338197 Objective Loss 0.338197 LR 0.001000 Time 0.020310 -2023-02-13 17:38:18,267 - Epoch: [44][ 1020/ 1207] Overall Loss 0.338167 Objective Loss 0.338167 LR 0.001000 Time 0.020304 -2023-02-13 17:38:18,462 - Epoch: [44][ 1030/ 1207] Overall Loss 0.337934 Objective Loss 0.337934 LR 0.001000 Time 0.020296 -2023-02-13 17:38:18,660 - Epoch: [44][ 1040/ 1207] Overall Loss 0.337829 Objective Loss 0.337829 LR 0.001000 Time 0.020291 -2023-02-13 17:38:18,855 - Epoch: [44][ 1050/ 1207] Overall Loss 0.337861 Objective Loss 0.337861 LR 0.001000 Time 0.020283 -2023-02-13 17:38:19,053 - Epoch: [44][ 1060/ 1207] Overall Loss 0.338023 Objective Loss 0.338023 LR 0.001000 Time 0.020278 -2023-02-13 17:38:19,249 - Epoch: [44][ 1070/ 1207] Overall Loss 0.338116 Objective Loss 0.338116 LR 0.001000 Time 0.020271 -2023-02-13 17:38:19,447 - Epoch: [44][ 1080/ 1207] Overall Loss 0.338184 Objective Loss 0.338184 LR 0.001000 Time 0.020267 -2023-02-13 17:38:19,642 - Epoch: [44][ 1090/ 1207] Overall Loss 0.338328 Objective Loss 0.338328 LR 0.001000 Time 0.020260 -2023-02-13 17:38:19,839 - Epoch: [44][ 1100/ 1207] Overall Loss 0.338290 Objective Loss 0.338290 LR 0.001000 Time 0.020254 -2023-02-13 17:38:20,035 - Epoch: [44][ 1110/ 1207] Overall Loss 0.338562 Objective Loss 0.338562 LR 0.001000 Time 0.020248 -2023-02-13 17:38:20,233 - Epoch: [44][ 1120/ 1207] Overall Loss 0.338468 Objective Loss 0.338468 LR 0.001000 Time 0.020244 -2023-02-13 17:38:20,429 - Epoch: [44][ 1130/ 1207] Overall Loss 0.338724 Objective Loss 0.338724 LR 0.001000 Time 0.020237 -2023-02-13 17:38:20,627 - Epoch: [44][ 1140/ 1207] Overall Loss 0.338812 Objective Loss 0.338812 LR 0.001000 Time 0.020233 -2023-02-13 17:38:20,823 - Epoch: [44][ 1150/ 1207] Overall Loss 0.338942 Objective Loss 0.338942 LR 0.001000 Time 0.020228 -2023-02-13 17:38:21,023 - Epoch: [44][ 1160/ 1207] Overall Loss 0.338991 Objective Loss 0.338991 LR 0.001000 Time 0.020225 -2023-02-13 17:38:21,218 - Epoch: [44][ 1170/ 1207] Overall Loss 0.338859 Objective Loss 0.338859 LR 0.001000 Time 0.020219 -2023-02-13 17:38:21,415 - Epoch: [44][ 1180/ 1207] Overall Loss 0.339214 Objective Loss 0.339214 LR 0.001000 Time 0.020214 -2023-02-13 17:38:21,611 - Epoch: [44][ 1190/ 1207] Overall Loss 0.339457 Objective Loss 0.339457 LR 0.001000 Time 0.020209 -2023-02-13 17:38:21,860 - Epoch: [44][ 1200/ 1207] Overall Loss 0.339550 Objective Loss 0.339550 LR 0.001000 Time 0.020248 -2023-02-13 17:38:21,976 - Epoch: [44][ 1207/ 1207] Overall Loss 0.339490 Objective Loss 0.339490 Top1 82.621951 Top5 97.865854 LR 0.001000 Time 0.020226 -2023-02-13 17:38:22,047 - --- validate (epoch=44)----------- -2023-02-13 17:38:22,048 - 34311 samples (256 per mini-batch) -2023-02-13 17:38:22,455 - Epoch: [44][ 10/ 135] Loss 0.390408 Top1 81.289062 Top5 97.578125 -2023-02-13 17:38:22,588 - Epoch: [44][ 20/ 135] Loss 0.382174 Top1 81.796875 Top5 97.343750 -2023-02-13 17:38:22,719 - Epoch: [44][ 30/ 135] Loss 0.387802 Top1 81.432292 Top5 97.161458 -2023-02-13 17:38:22,851 - Epoch: [44][ 40/ 135] Loss 0.393047 Top1 81.425781 Top5 97.109375 -2023-02-13 17:38:22,984 - Epoch: [44][ 50/ 135] Loss 0.385974 Top1 81.453125 Top5 97.210938 -2023-02-13 17:38:23,118 - Epoch: [44][ 60/ 135] Loss 0.386005 Top1 81.523438 Top5 97.233073 -2023-02-13 17:38:23,245 - Epoch: [44][ 70/ 135] Loss 0.387915 Top1 81.439732 Top5 97.198661 -2023-02-13 17:38:23,381 - Epoch: [44][ 80/ 135] Loss 0.385187 Top1 81.660156 Top5 97.192383 -2023-02-13 17:38:23,512 - Epoch: [44][ 90/ 135] Loss 0.382787 Top1 81.592882 Top5 97.187500 -2023-02-13 17:38:23,656 - Epoch: [44][ 100/ 135] Loss 0.387322 Top1 81.597656 Top5 97.128906 -2023-02-13 17:38:23,795 - Epoch: [44][ 110/ 135] Loss 0.385523 Top1 81.612216 Top5 97.173295 -2023-02-13 17:38:23,925 - Epoch: [44][ 120/ 135] Loss 0.386103 Top1 81.539714 Top5 97.141927 -2023-02-13 17:38:24,055 - Epoch: [44][ 130/ 135] Loss 0.384635 Top1 81.580529 Top5 97.157452 -2023-02-13 17:38:24,101 - Epoch: [44][ 135/ 135] Loss 0.383338 Top1 81.597738 Top5 97.132115 -2023-02-13 17:38:24,172 - ==> Top1: 81.598 Top5: 97.132 Loss: 0.383 - -2023-02-13 17:38:24,173 - ==> Confusion: -[[ 871 8 7 3 5 2 0 0 2 35 0 6 1 1 3 4 4 9 1 1 4] - [ 4 878 1 5 11 50 5 30 3 1 4 7 3 1 1 2 3 2 6 1 15] - [ 8 5 937 17 2 1 17 18 0 0 6 2 3 0 4 12 4 5 4 6 7] - [ 6 1 24 898 0 1 2 3 2 1 12 1 11 0 20 0 6 14 6 0 8] - [ 28 7 2 2 976 8 1 1 1 1 1 3 3 3 9 6 5 6 0 0 3] - [ 3 15 2 4 5 944 2 20 3 6 3 16 7 13 3 1 3 1 0 11 8] - [ 2 6 21 5 1 5 1017 5 0 0 7 2 2 1 0 3 2 5 1 11 3] - [ 3 4 14 2 2 37 3 900 2 1 7 10 3 2 0 1 0 2 10 12 9] - [ 27 3 1 2 2 1 0 0 859 60 9 2 3 6 22 1 0 3 5 0 3] - [ 164 1 3 2 9 4 2 2 26 778 2 1 0 7 6 0 0 4 0 0 1] - [ 4 2 5 10 1 3 2 7 16 2 962 1 2 10 7 1 0 0 10 1 5] - [ 2 1 1 1 0 12 0 5 5 2 0 893 37 5 1 3 2 18 1 13 3] - [ 0 2 1 1 0 2 0 2 1 0 4 43 857 1 1 7 2 26 0 0 9] - [ 8 6 2 1 3 9 2 1 19 23 13 13 4 884 9 7 5 3 0 3 9] - [ 13 1 3 34 6 1 0 2 19 2 3 1 7 2 970 1 5 6 7 0 9] - [ 8 0 3 3 10 1 3 1 0 1 0 9 12 0 0 937 5 34 0 10 9] - [ 3 7 1 1 9 3 1 0 3 1 1 11 2 1 0 16 975 4 0 5 17] - [ 9 5 1 5 0 2 2 1 0 0 0 7 14 1 1 9 0 989 0 1 4] - [ 6 6 7 26 1 1 0 40 3 0 7 2 14 0 17 1 0 6 944 1 4] - [ 1 1 1 1 1 7 6 16 1 0 1 20 2 3 0 3 3 6 0 1067 8] - [ 262 214 237 196 171 222 63 213 103 89 229 183 388 249 165 97 219 213 139 321 9461]] - -2023-02-13 17:38:24,174 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:38:24,174 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:38:24,180 - - -2023-02-13 17:38:24,180 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:38:25,043 - Epoch: [45][ 10/ 1207] Overall Loss 0.346066 Objective Loss 0.346066 LR 0.001000 Time 0.086265 -2023-02-13 17:38:25,233 - Epoch: [45][ 20/ 1207] Overall Loss 0.338961 Objective Loss 0.338961 LR 0.001000 Time 0.052621 -2023-02-13 17:38:25,423 - Epoch: [45][ 30/ 1207] Overall Loss 0.345894 Objective Loss 0.345894 LR 0.001000 Time 0.041386 -2023-02-13 17:38:25,612 - Epoch: [45][ 40/ 1207] Overall Loss 0.341659 Objective Loss 0.341659 LR 0.001000 Time 0.035756 -2023-02-13 17:38:25,801 - Epoch: [45][ 50/ 1207] Overall Loss 0.346435 Objective Loss 0.346435 LR 0.001000 Time 0.032386 -2023-02-13 17:38:25,992 - Epoch: [45][ 60/ 1207] Overall Loss 0.348274 Objective Loss 0.348274 LR 0.001000 Time 0.030155 -2023-02-13 17:38:26,181 - Epoch: [45][ 70/ 1207] Overall Loss 0.347216 Objective Loss 0.347216 LR 0.001000 Time 0.028547 -2023-02-13 17:38:26,369 - Epoch: [45][ 80/ 1207] Overall Loss 0.347770 Objective Loss 0.347770 LR 0.001000 Time 0.027325 -2023-02-13 17:38:26,558 - Epoch: [45][ 90/ 1207] Overall Loss 0.350458 Objective Loss 0.350458 LR 0.001000 Time 0.026387 -2023-02-13 17:38:26,747 - Epoch: [45][ 100/ 1207] Overall Loss 0.349242 Objective Loss 0.349242 LR 0.001000 Time 0.025630 -2023-02-13 17:38:26,936 - Epoch: [45][ 110/ 1207] Overall Loss 0.345972 Objective Loss 0.345972 LR 0.001000 Time 0.025017 -2023-02-13 17:38:27,126 - Epoch: [45][ 120/ 1207] Overall Loss 0.343650 Objective Loss 0.343650 LR 0.001000 Time 0.024508 -2023-02-13 17:38:27,315 - Epoch: [45][ 130/ 1207] Overall Loss 0.344738 Objective Loss 0.344738 LR 0.001000 Time 0.024073 -2023-02-13 17:38:27,503 - Epoch: [45][ 140/ 1207] Overall Loss 0.342198 Objective Loss 0.342198 LR 0.001000 Time 0.023699 -2023-02-13 17:38:27,692 - Epoch: [45][ 150/ 1207] Overall Loss 0.343508 Objective Loss 0.343508 LR 0.001000 Time 0.023377 -2023-02-13 17:38:27,881 - Epoch: [45][ 160/ 1207] Overall Loss 0.340681 Objective Loss 0.340681 LR 0.001000 Time 0.023093 -2023-02-13 17:38:28,071 - Epoch: [45][ 170/ 1207] Overall Loss 0.340806 Objective Loss 0.340806 LR 0.001000 Time 0.022848 -2023-02-13 17:38:28,259 - Epoch: [45][ 180/ 1207] Overall Loss 0.341915 Objective Loss 0.341915 LR 0.001000 Time 0.022621 -2023-02-13 17:38:28,447 - Epoch: [45][ 190/ 1207] Overall Loss 0.343174 Objective Loss 0.343174 LR 0.001000 Time 0.022420 -2023-02-13 17:38:28,635 - Epoch: [45][ 200/ 1207] Overall Loss 0.342632 Objective Loss 0.342632 LR 0.001000 Time 0.022239 -2023-02-13 17:38:28,824 - Epoch: [45][ 210/ 1207] Overall Loss 0.342523 Objective Loss 0.342523 LR 0.001000 Time 0.022076 -2023-02-13 17:38:29,015 - Epoch: [45][ 220/ 1207] Overall Loss 0.341592 Objective Loss 0.341592 LR 0.001000 Time 0.021938 -2023-02-13 17:38:29,203 - Epoch: [45][ 230/ 1207] Overall Loss 0.340081 Objective Loss 0.340081 LR 0.001000 Time 0.021802 -2023-02-13 17:38:29,391 - Epoch: [45][ 240/ 1207] Overall Loss 0.340368 Objective Loss 0.340368 LR 0.001000 Time 0.021676 -2023-02-13 17:38:29,581 - Epoch: [45][ 250/ 1207] Overall Loss 0.339520 Objective Loss 0.339520 LR 0.001000 Time 0.021567 -2023-02-13 17:38:29,769 - Epoch: [45][ 260/ 1207] Overall Loss 0.339134 Objective Loss 0.339134 LR 0.001000 Time 0.021459 -2023-02-13 17:38:29,958 - Epoch: [45][ 270/ 1207] Overall Loss 0.339211 Objective Loss 0.339211 LR 0.001000 Time 0.021363 -2023-02-13 17:38:30,147 - Epoch: [45][ 280/ 1207] Overall Loss 0.339464 Objective Loss 0.339464 LR 0.001000 Time 0.021274 -2023-02-13 17:38:30,336 - Epoch: [45][ 290/ 1207] Overall Loss 0.339607 Objective Loss 0.339607 LR 0.001000 Time 0.021188 -2023-02-13 17:38:30,525 - Epoch: [45][ 300/ 1207] Overall Loss 0.339507 Objective Loss 0.339507 LR 0.001000 Time 0.021112 -2023-02-13 17:38:30,713 - Epoch: [45][ 310/ 1207] Overall Loss 0.338771 Objective Loss 0.338771 LR 0.001000 Time 0.021037 -2023-02-13 17:38:30,902 - Epoch: [45][ 320/ 1207] Overall Loss 0.339274 Objective Loss 0.339274 LR 0.001000 Time 0.020970 -2023-02-13 17:38:31,092 - Epoch: [45][ 330/ 1207] Overall Loss 0.339897 Objective Loss 0.339897 LR 0.001000 Time 0.020908 -2023-02-13 17:38:31,280 - Epoch: [45][ 340/ 1207] Overall Loss 0.339768 Objective Loss 0.339768 LR 0.001000 Time 0.020845 -2023-02-13 17:38:31,469 - Epoch: [45][ 350/ 1207] Overall Loss 0.339482 Objective Loss 0.339482 LR 0.001000 Time 0.020788 -2023-02-13 17:38:31,657 - Epoch: [45][ 360/ 1207] Overall Loss 0.339477 Objective Loss 0.339477 LR 0.001000 Time 0.020732 -2023-02-13 17:38:31,847 - Epoch: [45][ 370/ 1207] Overall Loss 0.339332 Objective Loss 0.339332 LR 0.001000 Time 0.020683 -2023-02-13 17:38:32,036 - Epoch: [45][ 380/ 1207] Overall Loss 0.338998 Objective Loss 0.338998 LR 0.001000 Time 0.020635 -2023-02-13 17:38:32,224 - Epoch: [45][ 390/ 1207] Overall Loss 0.338647 Objective Loss 0.338647 LR 0.001000 Time 0.020589 -2023-02-13 17:38:32,414 - Epoch: [45][ 400/ 1207] Overall Loss 0.338511 Objective Loss 0.338511 LR 0.001000 Time 0.020547 -2023-02-13 17:38:32,603 - Epoch: [45][ 410/ 1207] Overall Loss 0.338779 Objective Loss 0.338779 LR 0.001000 Time 0.020506 -2023-02-13 17:38:32,791 - Epoch: [45][ 420/ 1207] Overall Loss 0.339161 Objective Loss 0.339161 LR 0.001000 Time 0.020466 -2023-02-13 17:38:32,981 - Epoch: [45][ 430/ 1207] Overall Loss 0.338759 Objective Loss 0.338759 LR 0.001000 Time 0.020430 -2023-02-13 17:38:33,185 - Epoch: [45][ 440/ 1207] Overall Loss 0.338775 Objective Loss 0.338775 LR 0.001000 Time 0.020428 -2023-02-13 17:38:33,383 - Epoch: [45][ 450/ 1207] Overall Loss 0.338931 Objective Loss 0.338931 LR 0.001000 Time 0.020414 -2023-02-13 17:38:33,585 - Epoch: [45][ 460/ 1207] Overall Loss 0.339346 Objective Loss 0.339346 LR 0.001000 Time 0.020409 -2023-02-13 17:38:33,783 - Epoch: [45][ 470/ 1207] Overall Loss 0.339514 Objective Loss 0.339514 LR 0.001000 Time 0.020393 -2023-02-13 17:38:33,984 - Epoch: [45][ 480/ 1207] Overall Loss 0.339875 Objective Loss 0.339875 LR 0.001000 Time 0.020388 -2023-02-13 17:38:34,183 - Epoch: [45][ 490/ 1207] Overall Loss 0.340029 Objective Loss 0.340029 LR 0.001000 Time 0.020376 -2023-02-13 17:38:34,385 - Epoch: [45][ 500/ 1207] Overall Loss 0.340217 Objective Loss 0.340217 LR 0.001000 Time 0.020373 -2023-02-13 17:38:34,583 - Epoch: [45][ 510/ 1207] Overall Loss 0.340015 Objective Loss 0.340015 LR 0.001000 Time 0.020361 -2023-02-13 17:38:34,785 - Epoch: [45][ 520/ 1207] Overall Loss 0.339794 Objective Loss 0.339794 LR 0.001000 Time 0.020357 -2023-02-13 17:38:34,982 - Epoch: [45][ 530/ 1207] Overall Loss 0.340192 Objective Loss 0.340192 LR 0.001000 Time 0.020344 -2023-02-13 17:38:35,185 - Epoch: [45][ 540/ 1207] Overall Loss 0.340450 Objective Loss 0.340450 LR 0.001000 Time 0.020342 -2023-02-13 17:38:35,382 - Epoch: [45][ 550/ 1207] Overall Loss 0.340265 Objective Loss 0.340265 LR 0.001000 Time 0.020329 -2023-02-13 17:38:35,584 - Epoch: [45][ 560/ 1207] Overall Loss 0.339913 Objective Loss 0.339913 LR 0.001000 Time 0.020327 -2023-02-13 17:38:35,783 - Epoch: [45][ 570/ 1207] Overall Loss 0.340177 Objective Loss 0.340177 LR 0.001000 Time 0.020318 -2023-02-13 17:38:35,985 - Epoch: [45][ 580/ 1207] Overall Loss 0.340927 Objective Loss 0.340927 LR 0.001000 Time 0.020316 -2023-02-13 17:38:36,183 - Epoch: [45][ 590/ 1207] Overall Loss 0.341387 Objective Loss 0.341387 LR 0.001000 Time 0.020306 -2023-02-13 17:38:36,385 - Epoch: [45][ 600/ 1207] Overall Loss 0.341889 Objective Loss 0.341889 LR 0.001000 Time 0.020304 -2023-02-13 17:38:36,584 - Epoch: [45][ 610/ 1207] Overall Loss 0.341752 Objective Loss 0.341752 LR 0.001000 Time 0.020296 -2023-02-13 17:38:36,787 - Epoch: [45][ 620/ 1207] Overall Loss 0.341623 Objective Loss 0.341623 LR 0.001000 Time 0.020295 -2023-02-13 17:38:36,984 - Epoch: [45][ 630/ 1207] Overall Loss 0.341442 Objective Loss 0.341442 LR 0.001000 Time 0.020286 -2023-02-13 17:38:37,187 - Epoch: [45][ 640/ 1207] Overall Loss 0.341131 Objective Loss 0.341131 LR 0.001000 Time 0.020285 -2023-02-13 17:38:37,384 - Epoch: [45][ 650/ 1207] Overall Loss 0.341535 Objective Loss 0.341535 LR 0.001000 Time 0.020276 -2023-02-13 17:38:37,586 - Epoch: [45][ 660/ 1207] Overall Loss 0.341598 Objective Loss 0.341598 LR 0.001000 Time 0.020275 -2023-02-13 17:38:37,784 - Epoch: [45][ 670/ 1207] Overall Loss 0.341954 Objective Loss 0.341954 LR 0.001000 Time 0.020266 -2023-02-13 17:38:37,986 - Epoch: [45][ 680/ 1207] Overall Loss 0.342279 Objective Loss 0.342279 LR 0.001000 Time 0.020264 -2023-02-13 17:38:38,180 - Epoch: [45][ 690/ 1207] Overall Loss 0.342388 Objective Loss 0.342388 LR 0.001000 Time 0.020252 -2023-02-13 17:38:38,369 - Epoch: [45][ 700/ 1207] Overall Loss 0.341917 Objective Loss 0.341917 LR 0.001000 Time 0.020232 -2023-02-13 17:38:38,557 - Epoch: [45][ 710/ 1207] Overall Loss 0.341747 Objective Loss 0.341747 LR 0.001000 Time 0.020211 -2023-02-13 17:38:38,745 - Epoch: [45][ 720/ 1207] Overall Loss 0.341849 Objective Loss 0.341849 LR 0.001000 Time 0.020191 -2023-02-13 17:38:38,933 - Epoch: [45][ 730/ 1207] Overall Loss 0.341628 Objective Loss 0.341628 LR 0.001000 Time 0.020171 -2023-02-13 17:38:39,122 - Epoch: [45][ 740/ 1207] Overall Loss 0.341532 Objective Loss 0.341532 LR 0.001000 Time 0.020153 -2023-02-13 17:38:39,309 - Epoch: [45][ 750/ 1207] Overall Loss 0.341807 Objective Loss 0.341807 LR 0.001000 Time 0.020134 -2023-02-13 17:38:39,498 - Epoch: [45][ 760/ 1207] Overall Loss 0.341433 Objective Loss 0.341433 LR 0.001000 Time 0.020117 -2023-02-13 17:38:39,686 - Epoch: [45][ 770/ 1207] Overall Loss 0.341111 Objective Loss 0.341111 LR 0.001000 Time 0.020099 -2023-02-13 17:38:39,873 - Epoch: [45][ 780/ 1207] Overall Loss 0.341097 Objective Loss 0.341097 LR 0.001000 Time 0.020082 -2023-02-13 17:38:40,062 - Epoch: [45][ 790/ 1207] Overall Loss 0.341406 Objective Loss 0.341406 LR 0.001000 Time 0.020066 -2023-02-13 17:38:40,250 - Epoch: [45][ 800/ 1207] Overall Loss 0.341125 Objective Loss 0.341125 LR 0.001000 Time 0.020050 -2023-02-13 17:38:40,438 - Epoch: [45][ 810/ 1207] Overall Loss 0.341102 Objective Loss 0.341102 LR 0.001000 Time 0.020034 -2023-02-13 17:38:40,626 - Epoch: [45][ 820/ 1207] Overall Loss 0.341080 Objective Loss 0.341080 LR 0.001000 Time 0.020018 -2023-02-13 17:38:40,816 - Epoch: [45][ 830/ 1207] Overall Loss 0.341049 Objective Loss 0.341049 LR 0.001000 Time 0.020005 -2023-02-13 17:38:41,004 - Epoch: [45][ 840/ 1207] Overall Loss 0.340957 Objective Loss 0.340957 LR 0.001000 Time 0.019990 -2023-02-13 17:38:41,192 - Epoch: [45][ 850/ 1207] Overall Loss 0.341160 Objective Loss 0.341160 LR 0.001000 Time 0.019977 -2023-02-13 17:38:41,380 - Epoch: [45][ 860/ 1207] Overall Loss 0.340904 Objective Loss 0.340904 LR 0.001000 Time 0.019962 -2023-02-13 17:38:41,568 - Epoch: [45][ 870/ 1207] Overall Loss 0.340727 Objective Loss 0.340727 LR 0.001000 Time 0.019948 -2023-02-13 17:38:41,756 - Epoch: [45][ 880/ 1207] Overall Loss 0.340564 Objective Loss 0.340564 LR 0.001000 Time 0.019935 -2023-02-13 17:38:41,944 - Epoch: [45][ 890/ 1207] Overall Loss 0.340524 Objective Loss 0.340524 LR 0.001000 Time 0.019922 -2023-02-13 17:38:42,132 - Epoch: [45][ 900/ 1207] Overall Loss 0.340291 Objective Loss 0.340291 LR 0.001000 Time 0.019909 -2023-02-13 17:38:42,320 - Epoch: [45][ 910/ 1207] Overall Loss 0.340307 Objective Loss 0.340307 LR 0.001000 Time 0.019896 -2023-02-13 17:38:42,508 - Epoch: [45][ 920/ 1207] Overall Loss 0.340313 Objective Loss 0.340313 LR 0.001000 Time 0.019884 -2023-02-13 17:38:42,696 - Epoch: [45][ 930/ 1207] Overall Loss 0.340358 Objective Loss 0.340358 LR 0.001000 Time 0.019871 -2023-02-13 17:38:42,884 - Epoch: [45][ 940/ 1207] Overall Loss 0.340542 Objective Loss 0.340542 LR 0.001000 Time 0.019860 -2023-02-13 17:38:43,071 - Epoch: [45][ 950/ 1207] Overall Loss 0.340345 Objective Loss 0.340345 LR 0.001000 Time 0.019848 -2023-02-13 17:38:43,259 - Epoch: [45][ 960/ 1207] Overall Loss 0.340520 Objective Loss 0.340520 LR 0.001000 Time 0.019836 -2023-02-13 17:38:43,446 - Epoch: [45][ 970/ 1207] Overall Loss 0.340848 Objective Loss 0.340848 LR 0.001000 Time 0.019824 -2023-02-13 17:38:43,635 - Epoch: [45][ 980/ 1207] Overall Loss 0.340988 Objective Loss 0.340988 LR 0.001000 Time 0.019815 -2023-02-13 17:38:43,823 - Epoch: [45][ 990/ 1207] Overall Loss 0.341092 Objective Loss 0.341092 LR 0.001000 Time 0.019803 -2023-02-13 17:38:44,011 - Epoch: [45][ 1000/ 1207] Overall Loss 0.341006 Objective Loss 0.341006 LR 0.001000 Time 0.019793 -2023-02-13 17:38:44,199 - Epoch: [45][ 1010/ 1207] Overall Loss 0.340911 Objective Loss 0.340911 LR 0.001000 Time 0.019783 -2023-02-13 17:38:44,387 - Epoch: [45][ 1020/ 1207] Overall Loss 0.340871 Objective Loss 0.340871 LR 0.001000 Time 0.019773 -2023-02-13 17:38:44,575 - Epoch: [45][ 1030/ 1207] Overall Loss 0.340604 Objective Loss 0.340604 LR 0.001000 Time 0.019763 -2023-02-13 17:38:44,763 - Epoch: [45][ 1040/ 1207] Overall Loss 0.340436 Objective Loss 0.340436 LR 0.001000 Time 0.019753 -2023-02-13 17:38:44,951 - Epoch: [45][ 1050/ 1207] Overall Loss 0.340487 Objective Loss 0.340487 LR 0.001000 Time 0.019744 -2023-02-13 17:38:45,139 - Epoch: [45][ 1060/ 1207] Overall Loss 0.340470 Objective Loss 0.340470 LR 0.001000 Time 0.019735 -2023-02-13 17:38:45,326 - Epoch: [45][ 1070/ 1207] Overall Loss 0.340354 Objective Loss 0.340354 LR 0.001000 Time 0.019724 -2023-02-13 17:38:45,513 - Epoch: [45][ 1080/ 1207] Overall Loss 0.340283 Objective Loss 0.340283 LR 0.001000 Time 0.019715 -2023-02-13 17:38:45,702 - Epoch: [45][ 1090/ 1207] Overall Loss 0.340458 Objective Loss 0.340458 LR 0.001000 Time 0.019707 -2023-02-13 17:38:45,891 - Epoch: [45][ 1100/ 1207] Overall Loss 0.340108 Objective Loss 0.340108 LR 0.001000 Time 0.019699 -2023-02-13 17:38:46,079 - Epoch: [45][ 1110/ 1207] Overall Loss 0.340207 Objective Loss 0.340207 LR 0.001000 Time 0.019691 -2023-02-13 17:38:46,267 - Epoch: [45][ 1120/ 1207] Overall Loss 0.340360 Objective Loss 0.340360 LR 0.001000 Time 0.019682 -2023-02-13 17:38:46,454 - Epoch: [45][ 1130/ 1207] Overall Loss 0.340410 Objective Loss 0.340410 LR 0.001000 Time 0.019674 -2023-02-13 17:38:46,643 - Epoch: [45][ 1140/ 1207] Overall Loss 0.340458 Objective Loss 0.340458 LR 0.001000 Time 0.019667 -2023-02-13 17:38:46,831 - Epoch: [45][ 1150/ 1207] Overall Loss 0.340402 Objective Loss 0.340402 LR 0.001000 Time 0.019659 -2023-02-13 17:38:47,019 - Epoch: [45][ 1160/ 1207] Overall Loss 0.340307 Objective Loss 0.340307 LR 0.001000 Time 0.019651 -2023-02-13 17:38:47,207 - Epoch: [45][ 1170/ 1207] Overall Loss 0.340780 Objective Loss 0.340780 LR 0.001000 Time 0.019644 -2023-02-13 17:38:47,395 - Epoch: [45][ 1180/ 1207] Overall Loss 0.340643 Objective Loss 0.340643 LR 0.001000 Time 0.019636 -2023-02-13 17:38:47,583 - Epoch: [45][ 1190/ 1207] Overall Loss 0.340552 Objective Loss 0.340552 LR 0.001000 Time 0.019629 -2023-02-13 17:38:47,823 - Epoch: [45][ 1200/ 1207] Overall Loss 0.340711 Objective Loss 0.340711 LR 0.001000 Time 0.019665 -2023-02-13 17:38:47,938 - Epoch: [45][ 1207/ 1207] Overall Loss 0.340878 Objective Loss 0.340878 Top1 85.365854 Top5 97.560976 LR 0.001000 Time 0.019646 -2023-02-13 17:38:48,021 - --- validate (epoch=45)----------- -2023-02-13 17:38:48,021 - 34311 samples (256 per mini-batch) -2023-02-13 17:38:48,515 - Epoch: [45][ 10/ 135] Loss 0.364180 Top1 82.578125 Top5 97.382812 -2023-02-13 17:38:48,646 - Epoch: [45][ 20/ 135] Loss 0.361813 Top1 82.089844 Top5 97.207031 -2023-02-13 17:38:48,778 - Epoch: [45][ 30/ 135] Loss 0.357088 Top1 82.135417 Top5 97.226562 -2023-02-13 17:38:48,904 - Epoch: [45][ 40/ 135] Loss 0.354130 Top1 82.402344 Top5 97.226562 -2023-02-13 17:38:49,032 - Epoch: [45][ 50/ 135] Loss 0.365605 Top1 82.226562 Top5 97.171875 -2023-02-13 17:38:49,158 - Epoch: [45][ 60/ 135] Loss 0.365378 Top1 82.083333 Top5 97.115885 -2023-02-13 17:38:49,287 - Epoch: [45][ 70/ 135] Loss 0.370360 Top1 81.975446 Top5 97.075893 -2023-02-13 17:38:49,415 - Epoch: [45][ 80/ 135] Loss 0.366509 Top1 82.128906 Top5 97.084961 -2023-02-13 17:38:49,543 - Epoch: [45][ 90/ 135] Loss 0.369115 Top1 82.052951 Top5 97.148438 -2023-02-13 17:38:49,669 - Epoch: [45][ 100/ 135] Loss 0.367225 Top1 82.105469 Top5 97.164062 -2023-02-13 17:38:49,797 - Epoch: [45][ 110/ 135] Loss 0.364837 Top1 82.162642 Top5 97.183949 -2023-02-13 17:38:49,927 - Epoch: [45][ 120/ 135] Loss 0.363624 Top1 82.207031 Top5 97.164714 -2023-02-13 17:38:50,057 - Epoch: [45][ 130/ 135] Loss 0.363341 Top1 82.223558 Top5 97.154447 -2023-02-13 17:38:50,104 - Epoch: [45][ 135/ 135] Loss 0.360794 Top1 82.145668 Top5 97.161260 -2023-02-13 17:38:50,176 - ==> Top1: 82.146 Top5: 97.161 Loss: 0.361 - -2023-02-13 17:38:50,176 - ==> Confusion: -[[ 834 7 16 2 4 3 0 1 3 62 0 5 1 3 3 5 1 4 2 2 9] - [ 2 927 2 5 7 30 7 15 5 1 3 1 2 0 1 3 3 0 9 5 5] - [ 9 4 957 14 1 2 24 15 0 0 1 1 2 0 3 9 1 3 5 2 5] - [ 10 1 28 903 0 3 0 1 1 1 18 3 5 1 13 2 2 6 14 1 3] - [ 21 9 3 2 975 11 1 1 0 4 1 3 2 4 10 6 5 2 1 3 2] - [ 3 23 3 9 4 958 5 19 1 6 3 7 2 6 1 5 1 2 3 6 3] - [ 2 4 13 3 2 1 1042 3 1 1 5 1 4 0 1 4 0 1 0 10 1] - [ 3 8 17 4 0 29 8 901 1 2 6 3 3 0 1 0 0 2 22 8 6] - [ 20 4 0 2 1 0 0 2 869 59 12 2 1 6 17 4 0 0 7 0 3] - [ 82 3 5 1 4 2 0 2 44 849 1 1 0 5 5 0 0 2 1 0 5] - [ 3 3 11 7 1 2 4 4 14 0 973 0 1 5 1 1 1 1 10 1 8] - [ 5 2 3 0 5 19 1 3 3 0 1 880 33 6 2 4 2 10 3 23 0] - [ 2 1 3 7 0 1 0 1 1 1 0 41 861 0 1 4 1 22 0 5 7] - [ 6 3 5 1 6 9 3 7 24 32 23 5 4 869 4 6 2 3 0 5 7] - [ 11 4 2 36 4 3 0 3 17 4 7 1 7 0 959 1 1 3 11 1 17] - [ 6 4 8 2 8 1 7 2 0 0 0 10 9 4 1 951 6 6 0 11 10] - [ 5 9 2 1 10 4 2 1 2 1 1 5 2 3 1 11 970 2 3 8 18] - [ 5 2 1 7 2 0 3 2 0 1 0 13 11 2 0 14 0 981 0 1 6] - [ 6 4 11 21 1 1 1 24 2 0 3 2 5 0 11 0 0 1 987 2 4] - [ 0 3 2 0 0 3 15 19 1 0 1 19 1 3 0 4 1 1 1 1068 6] - [ 202 280 346 172 124 218 114 179 128 155 226 138 325 287 151 110 189 115 171 333 9471]] - -2023-02-13 17:38:50,178 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:38:50,178 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:38:50,183 - - -2023-02-13 17:38:50,183 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:38:51,069 - Epoch: [46][ 10/ 1207] Overall Loss 0.328136 Objective Loss 0.328136 LR 0.001000 Time 0.088529 -2023-02-13 17:38:51,265 - Epoch: [46][ 20/ 1207] Overall Loss 0.323199 Objective Loss 0.323199 LR 0.001000 Time 0.054011 -2023-02-13 17:38:51,455 - Epoch: [46][ 30/ 1207] Overall Loss 0.327854 Objective Loss 0.327854 LR 0.001000 Time 0.042352 -2023-02-13 17:38:51,647 - Epoch: [46][ 40/ 1207] Overall Loss 0.333096 Objective Loss 0.333096 LR 0.001000 Time 0.036534 -2023-02-13 17:38:51,837 - Epoch: [46][ 50/ 1207] Overall Loss 0.340563 Objective Loss 0.340563 LR 0.001000 Time 0.033030 -2023-02-13 17:38:52,028 - Epoch: [46][ 60/ 1207] Overall Loss 0.343537 Objective Loss 0.343537 LR 0.001000 Time 0.030695 -2023-02-13 17:38:52,219 - Epoch: [46][ 70/ 1207] Overall Loss 0.340426 Objective Loss 0.340426 LR 0.001000 Time 0.029042 -2023-02-13 17:38:52,410 - Epoch: [46][ 80/ 1207] Overall Loss 0.338457 Objective Loss 0.338457 LR 0.001000 Time 0.027794 -2023-02-13 17:38:52,601 - Epoch: [46][ 90/ 1207] Overall Loss 0.339011 Objective Loss 0.339011 LR 0.001000 Time 0.026819 -2023-02-13 17:38:52,792 - Epoch: [46][ 100/ 1207] Overall Loss 0.336225 Objective Loss 0.336225 LR 0.001000 Time 0.026048 -2023-02-13 17:38:52,983 - Epoch: [46][ 110/ 1207] Overall Loss 0.336685 Objective Loss 0.336685 LR 0.001000 Time 0.025410 -2023-02-13 17:38:53,175 - Epoch: [46][ 120/ 1207] Overall Loss 0.335872 Objective Loss 0.335872 LR 0.001000 Time 0.024888 -2023-02-13 17:38:53,364 - Epoch: [46][ 130/ 1207] Overall Loss 0.339894 Objective Loss 0.339894 LR 0.001000 Time 0.024423 -2023-02-13 17:38:53,553 - Epoch: [46][ 140/ 1207] Overall Loss 0.337897 Objective Loss 0.337897 LR 0.001000 Time 0.024030 -2023-02-13 17:38:53,742 - Epoch: [46][ 150/ 1207] Overall Loss 0.336017 Objective Loss 0.336017 LR 0.001000 Time 0.023685 -2023-02-13 17:38:53,931 - Epoch: [46][ 160/ 1207] Overall Loss 0.335222 Objective Loss 0.335222 LR 0.001000 Time 0.023385 -2023-02-13 17:38:54,121 - Epoch: [46][ 170/ 1207] Overall Loss 0.333664 Objective Loss 0.333664 LR 0.001000 Time 0.023123 -2023-02-13 17:38:54,310 - Epoch: [46][ 180/ 1207] Overall Loss 0.333932 Objective Loss 0.333932 LR 0.001000 Time 0.022885 -2023-02-13 17:38:54,499 - Epoch: [46][ 190/ 1207] Overall Loss 0.332986 Objective Loss 0.332986 LR 0.001000 Time 0.022674 -2023-02-13 17:38:54,688 - Epoch: [46][ 200/ 1207] Overall Loss 0.331520 Objective Loss 0.331520 LR 0.001000 Time 0.022484 -2023-02-13 17:38:54,878 - Epoch: [46][ 210/ 1207] Overall Loss 0.331362 Objective Loss 0.331362 LR 0.001000 Time 0.022315 -2023-02-13 17:38:55,067 - Epoch: [46][ 220/ 1207] Overall Loss 0.332451 Objective Loss 0.332451 LR 0.001000 Time 0.022161 -2023-02-13 17:38:55,257 - Epoch: [46][ 230/ 1207] Overall Loss 0.331221 Objective Loss 0.331221 LR 0.001000 Time 0.022023 -2023-02-13 17:38:55,447 - Epoch: [46][ 240/ 1207] Overall Loss 0.332642 Objective Loss 0.332642 LR 0.001000 Time 0.021894 -2023-02-13 17:38:55,636 - Epoch: [46][ 250/ 1207] Overall Loss 0.333636 Objective Loss 0.333636 LR 0.001000 Time 0.021773 -2023-02-13 17:38:55,826 - Epoch: [46][ 260/ 1207] Overall Loss 0.334678 Objective Loss 0.334678 LR 0.001000 Time 0.021665 -2023-02-13 17:38:56,015 - Epoch: [46][ 270/ 1207] Overall Loss 0.334132 Objective Loss 0.334132 LR 0.001000 Time 0.021562 -2023-02-13 17:38:56,206 - Epoch: [46][ 280/ 1207] Overall Loss 0.334341 Objective Loss 0.334341 LR 0.001000 Time 0.021472 -2023-02-13 17:38:56,395 - Epoch: [46][ 290/ 1207] Overall Loss 0.334888 Objective Loss 0.334888 LR 0.001000 Time 0.021381 -2023-02-13 17:38:56,585 - Epoch: [46][ 300/ 1207] Overall Loss 0.335351 Objective Loss 0.335351 LR 0.001000 Time 0.021302 -2023-02-13 17:38:56,777 - Epoch: [46][ 310/ 1207] Overall Loss 0.336281 Objective Loss 0.336281 LR 0.001000 Time 0.021231 -2023-02-13 17:38:56,968 - Epoch: [46][ 320/ 1207] Overall Loss 0.337421 Objective Loss 0.337421 LR 0.001000 Time 0.021164 -2023-02-13 17:38:57,161 - Epoch: [46][ 330/ 1207] Overall Loss 0.337332 Objective Loss 0.337332 LR 0.001000 Time 0.021105 -2023-02-13 17:38:57,352 - Epoch: [46][ 340/ 1207] Overall Loss 0.338344 Objective Loss 0.338344 LR 0.001000 Time 0.021047 -2023-02-13 17:38:57,544 - Epoch: [46][ 350/ 1207] Overall Loss 0.339262 Objective Loss 0.339262 LR 0.001000 Time 0.020993 -2023-02-13 17:38:57,735 - Epoch: [46][ 360/ 1207] Overall Loss 0.340010 Objective Loss 0.340010 LR 0.001000 Time 0.020939 -2023-02-13 17:38:57,926 - Epoch: [46][ 370/ 1207] Overall Loss 0.339947 Objective Loss 0.339947 LR 0.001000 Time 0.020889 -2023-02-13 17:38:58,118 - Epoch: [46][ 380/ 1207] Overall Loss 0.340805 Objective Loss 0.340805 LR 0.001000 Time 0.020844 -2023-02-13 17:38:58,310 - Epoch: [46][ 390/ 1207] Overall Loss 0.340838 Objective Loss 0.340838 LR 0.001000 Time 0.020799 -2023-02-13 17:38:58,501 - Epoch: [46][ 400/ 1207] Overall Loss 0.341838 Objective Loss 0.341838 LR 0.001000 Time 0.020758 -2023-02-13 17:38:58,693 - Epoch: [46][ 410/ 1207] Overall Loss 0.342287 Objective Loss 0.342287 LR 0.001000 Time 0.020717 -2023-02-13 17:38:58,884 - Epoch: [46][ 420/ 1207] Overall Loss 0.342478 Objective Loss 0.342478 LR 0.001000 Time 0.020678 -2023-02-13 17:38:59,076 - Epoch: [46][ 430/ 1207] Overall Loss 0.342793 Objective Loss 0.342793 LR 0.001000 Time 0.020644 -2023-02-13 17:38:59,268 - Epoch: [46][ 440/ 1207] Overall Loss 0.343159 Objective Loss 0.343159 LR 0.001000 Time 0.020610 -2023-02-13 17:38:59,460 - Epoch: [46][ 450/ 1207] Overall Loss 0.342587 Objective Loss 0.342587 LR 0.001000 Time 0.020577 -2023-02-13 17:38:59,652 - Epoch: [46][ 460/ 1207] Overall Loss 0.342451 Objective Loss 0.342451 LR 0.001000 Time 0.020547 -2023-02-13 17:38:59,844 - Epoch: [46][ 470/ 1207] Overall Loss 0.342201 Objective Loss 0.342201 LR 0.001000 Time 0.020516 -2023-02-13 17:39:00,035 - Epoch: [46][ 480/ 1207] Overall Loss 0.342183 Objective Loss 0.342183 LR 0.001000 Time 0.020486 -2023-02-13 17:39:00,226 - Epoch: [46][ 490/ 1207] Overall Loss 0.342434 Objective Loss 0.342434 LR 0.001000 Time 0.020458 -2023-02-13 17:39:00,418 - Epoch: [46][ 500/ 1207] Overall Loss 0.342888 Objective Loss 0.342888 LR 0.001000 Time 0.020432 -2023-02-13 17:39:00,609 - Epoch: [46][ 510/ 1207] Overall Loss 0.342714 Objective Loss 0.342714 LR 0.001000 Time 0.020406 -2023-02-13 17:39:00,803 - Epoch: [46][ 520/ 1207] Overall Loss 0.342975 Objective Loss 0.342975 LR 0.001000 Time 0.020385 -2023-02-13 17:39:00,994 - Epoch: [46][ 530/ 1207] Overall Loss 0.343211 Objective Loss 0.343211 LR 0.001000 Time 0.020360 -2023-02-13 17:39:01,186 - Epoch: [46][ 540/ 1207] Overall Loss 0.342977 Objective Loss 0.342977 LR 0.001000 Time 0.020339 -2023-02-13 17:39:01,378 - Epoch: [46][ 550/ 1207] Overall Loss 0.342859 Objective Loss 0.342859 LR 0.001000 Time 0.020317 -2023-02-13 17:39:01,570 - Epoch: [46][ 560/ 1207] Overall Loss 0.342140 Objective Loss 0.342140 LR 0.001000 Time 0.020296 -2023-02-13 17:39:01,760 - Epoch: [46][ 570/ 1207] Overall Loss 0.341951 Objective Loss 0.341951 LR 0.001000 Time 0.020273 -2023-02-13 17:39:01,950 - Epoch: [46][ 580/ 1207] Overall Loss 0.342122 Objective Loss 0.342122 LR 0.001000 Time 0.020250 -2023-02-13 17:39:02,140 - Epoch: [46][ 590/ 1207] Overall Loss 0.342307 Objective Loss 0.342307 LR 0.001000 Time 0.020229 -2023-02-13 17:39:02,330 - Epoch: [46][ 600/ 1207] Overall Loss 0.342092 Objective Loss 0.342092 LR 0.001000 Time 0.020207 -2023-02-13 17:39:02,520 - Epoch: [46][ 610/ 1207] Overall Loss 0.342328 Objective Loss 0.342328 LR 0.001000 Time 0.020186 -2023-02-13 17:39:02,710 - Epoch: [46][ 620/ 1207] Overall Loss 0.341869 Objective Loss 0.341869 LR 0.001000 Time 0.020167 -2023-02-13 17:39:02,899 - Epoch: [46][ 630/ 1207] Overall Loss 0.341548 Objective Loss 0.341548 LR 0.001000 Time 0.020147 -2023-02-13 17:39:03,088 - Epoch: [46][ 640/ 1207] Overall Loss 0.341740 Objective Loss 0.341740 LR 0.001000 Time 0.020126 -2023-02-13 17:39:03,278 - Epoch: [46][ 650/ 1207] Overall Loss 0.341601 Objective Loss 0.341601 LR 0.001000 Time 0.020109 -2023-02-13 17:39:03,467 - Epoch: [46][ 660/ 1207] Overall Loss 0.341359 Objective Loss 0.341359 LR 0.001000 Time 0.020090 -2023-02-13 17:39:03,657 - Epoch: [46][ 670/ 1207] Overall Loss 0.341164 Objective Loss 0.341164 LR 0.001000 Time 0.020073 -2023-02-13 17:39:03,846 - Epoch: [46][ 680/ 1207] Overall Loss 0.340783 Objective Loss 0.340783 LR 0.001000 Time 0.020055 -2023-02-13 17:39:04,036 - Epoch: [46][ 690/ 1207] Overall Loss 0.340566 Objective Loss 0.340566 LR 0.001000 Time 0.020038 -2023-02-13 17:39:04,226 - Epoch: [46][ 700/ 1207] Overall Loss 0.340074 Objective Loss 0.340074 LR 0.001000 Time 0.020024 -2023-02-13 17:39:04,416 - Epoch: [46][ 710/ 1207] Overall Loss 0.340096 Objective Loss 0.340096 LR 0.001000 Time 0.020008 -2023-02-13 17:39:04,606 - Epoch: [46][ 720/ 1207] Overall Loss 0.340036 Objective Loss 0.340036 LR 0.001000 Time 0.019993 -2023-02-13 17:39:04,795 - Epoch: [46][ 730/ 1207] Overall Loss 0.340326 Objective Loss 0.340326 LR 0.001000 Time 0.019978 -2023-02-13 17:39:04,985 - Epoch: [46][ 740/ 1207] Overall Loss 0.339821 Objective Loss 0.339821 LR 0.001000 Time 0.019964 -2023-02-13 17:39:05,175 - Epoch: [46][ 750/ 1207] Overall Loss 0.339847 Objective Loss 0.339847 LR 0.001000 Time 0.019951 -2023-02-13 17:39:05,364 - Epoch: [46][ 760/ 1207] Overall Loss 0.339543 Objective Loss 0.339543 LR 0.001000 Time 0.019936 -2023-02-13 17:39:05,554 - Epoch: [46][ 770/ 1207] Overall Loss 0.339815 Objective Loss 0.339815 LR 0.001000 Time 0.019924 -2023-02-13 17:39:05,743 - Epoch: [46][ 780/ 1207] Overall Loss 0.339533 Objective Loss 0.339533 LR 0.001000 Time 0.019910 -2023-02-13 17:39:05,933 - Epoch: [46][ 790/ 1207] Overall Loss 0.339244 Objective Loss 0.339244 LR 0.001000 Time 0.019899 -2023-02-13 17:39:06,123 - Epoch: [46][ 800/ 1207] Overall Loss 0.339086 Objective Loss 0.339086 LR 0.001000 Time 0.019887 -2023-02-13 17:39:06,312 - Epoch: [46][ 810/ 1207] Overall Loss 0.339134 Objective Loss 0.339134 LR 0.001000 Time 0.019875 -2023-02-13 17:39:06,502 - Epoch: [46][ 820/ 1207] Overall Loss 0.339084 Objective Loss 0.339084 LR 0.001000 Time 0.019863 -2023-02-13 17:39:06,692 - Epoch: [46][ 830/ 1207] Overall Loss 0.339135 Objective Loss 0.339135 LR 0.001000 Time 0.019852 -2023-02-13 17:39:06,883 - Epoch: [46][ 840/ 1207] Overall Loss 0.338931 Objective Loss 0.338931 LR 0.001000 Time 0.019843 -2023-02-13 17:39:07,072 - Epoch: [46][ 850/ 1207] Overall Loss 0.338424 Objective Loss 0.338424 LR 0.001000 Time 0.019832 -2023-02-13 17:39:07,262 - Epoch: [46][ 860/ 1207] Overall Loss 0.338027 Objective Loss 0.338027 LR 0.001000 Time 0.019822 -2023-02-13 17:39:07,452 - Epoch: [46][ 870/ 1207] Overall Loss 0.337827 Objective Loss 0.337827 LR 0.001000 Time 0.019811 -2023-02-13 17:39:07,642 - Epoch: [46][ 880/ 1207] Overall Loss 0.337373 Objective Loss 0.337373 LR 0.001000 Time 0.019802 -2023-02-13 17:39:07,831 - Epoch: [46][ 890/ 1207] Overall Loss 0.337255 Objective Loss 0.337255 LR 0.001000 Time 0.019792 -2023-02-13 17:39:08,021 - Epoch: [46][ 900/ 1207] Overall Loss 0.337295 Objective Loss 0.337295 LR 0.001000 Time 0.019782 -2023-02-13 17:39:08,211 - Epoch: [46][ 910/ 1207] Overall Loss 0.337223 Objective Loss 0.337223 LR 0.001000 Time 0.019774 -2023-02-13 17:39:08,401 - Epoch: [46][ 920/ 1207] Overall Loss 0.337426 Objective Loss 0.337426 LR 0.001000 Time 0.019765 -2023-02-13 17:39:08,591 - Epoch: [46][ 930/ 1207] Overall Loss 0.337298 Objective Loss 0.337298 LR 0.001000 Time 0.019756 -2023-02-13 17:39:08,781 - Epoch: [46][ 940/ 1207] Overall Loss 0.337782 Objective Loss 0.337782 LR 0.001000 Time 0.019748 -2023-02-13 17:39:08,971 - Epoch: [46][ 950/ 1207] Overall Loss 0.337829 Objective Loss 0.337829 LR 0.001000 Time 0.019739 -2023-02-13 17:39:09,161 - Epoch: [46][ 960/ 1207] Overall Loss 0.338099 Objective Loss 0.338099 LR 0.001000 Time 0.019731 -2023-02-13 17:39:09,350 - Epoch: [46][ 970/ 1207] Overall Loss 0.338513 Objective Loss 0.338513 LR 0.001000 Time 0.019722 -2023-02-13 17:39:09,541 - Epoch: [46][ 980/ 1207] Overall Loss 0.338678 Objective Loss 0.338678 LR 0.001000 Time 0.019715 -2023-02-13 17:39:09,731 - Epoch: [46][ 990/ 1207] Overall Loss 0.338626 Objective Loss 0.338626 LR 0.001000 Time 0.019707 -2023-02-13 17:39:09,920 - Epoch: [46][ 1000/ 1207] Overall Loss 0.338422 Objective Loss 0.338422 LR 0.001000 Time 0.019699 -2023-02-13 17:39:10,110 - Epoch: [46][ 1010/ 1207] Overall Loss 0.338550 Objective Loss 0.338550 LR 0.001000 Time 0.019692 -2023-02-13 17:39:10,300 - Epoch: [46][ 1020/ 1207] Overall Loss 0.338532 Objective Loss 0.338532 LR 0.001000 Time 0.019685 -2023-02-13 17:39:10,490 - Epoch: [46][ 1030/ 1207] Overall Loss 0.338465 Objective Loss 0.338465 LR 0.001000 Time 0.019677 -2023-02-13 17:39:10,678 - Epoch: [46][ 1040/ 1207] Overall Loss 0.338567 Objective Loss 0.338567 LR 0.001000 Time 0.019669 -2023-02-13 17:39:10,869 - Epoch: [46][ 1050/ 1207] Overall Loss 0.338413 Objective Loss 0.338413 LR 0.001000 Time 0.019663 -2023-02-13 17:39:11,057 - Epoch: [46][ 1060/ 1207] Overall Loss 0.338156 Objective Loss 0.338156 LR 0.001000 Time 0.019655 -2023-02-13 17:39:11,249 - Epoch: [46][ 1070/ 1207] Overall Loss 0.338247 Objective Loss 0.338247 LR 0.001000 Time 0.019650 -2023-02-13 17:39:11,437 - Epoch: [46][ 1080/ 1207] Overall Loss 0.337903 Objective Loss 0.337903 LR 0.001000 Time 0.019642 -2023-02-13 17:39:11,627 - Epoch: [46][ 1090/ 1207] Overall Loss 0.337784 Objective Loss 0.337784 LR 0.001000 Time 0.019636 -2023-02-13 17:39:11,816 - Epoch: [46][ 1100/ 1207] Overall Loss 0.337688 Objective Loss 0.337688 LR 0.001000 Time 0.019629 -2023-02-13 17:39:12,006 - Epoch: [46][ 1110/ 1207] Overall Loss 0.337388 Objective Loss 0.337388 LR 0.001000 Time 0.019622 -2023-02-13 17:39:12,196 - Epoch: [46][ 1120/ 1207] Overall Loss 0.337404 Objective Loss 0.337404 LR 0.001000 Time 0.019616 -2023-02-13 17:39:12,384 - Epoch: [46][ 1130/ 1207] Overall Loss 0.337198 Objective Loss 0.337198 LR 0.001000 Time 0.019609 -2023-02-13 17:39:12,573 - Epoch: [46][ 1140/ 1207] Overall Loss 0.337073 Objective Loss 0.337073 LR 0.001000 Time 0.019603 -2023-02-13 17:39:12,763 - Epoch: [46][ 1150/ 1207] Overall Loss 0.337349 Objective Loss 0.337349 LR 0.001000 Time 0.019597 -2023-02-13 17:39:12,953 - Epoch: [46][ 1160/ 1207] Overall Loss 0.337174 Objective Loss 0.337174 LR 0.001000 Time 0.019591 -2023-02-13 17:39:13,142 - Epoch: [46][ 1170/ 1207] Overall Loss 0.337190 Objective Loss 0.337190 LR 0.001000 Time 0.019586 -2023-02-13 17:39:13,332 - Epoch: [46][ 1180/ 1207] Overall Loss 0.336910 Objective Loss 0.336910 LR 0.001000 Time 0.019580 -2023-02-13 17:39:13,522 - Epoch: [46][ 1190/ 1207] Overall Loss 0.336943 Objective Loss 0.336943 LR 0.001000 Time 0.019575 -2023-02-13 17:39:13,762 - Epoch: [46][ 1200/ 1207] Overall Loss 0.337352 Objective Loss 0.337352 LR 0.001000 Time 0.019611 -2023-02-13 17:39:13,878 - Epoch: [46][ 1207/ 1207] Overall Loss 0.337355 Objective Loss 0.337355 Top1 82.317073 Top5 97.560976 LR 0.001000 Time 0.019594 -2023-02-13 17:39:13,956 - --- validate (epoch=46)----------- -2023-02-13 17:39:13,956 - 34311 samples (256 per mini-batch) -2023-02-13 17:39:14,354 - Epoch: [46][ 10/ 135] Loss 0.385678 Top1 82.187500 Top5 97.460938 -2023-02-13 17:39:14,483 - Epoch: [46][ 20/ 135] Loss 0.363275 Top1 82.851562 Top5 97.441406 -2023-02-13 17:39:14,613 - Epoch: [46][ 30/ 135] Loss 0.366081 Top1 82.200521 Top5 97.317708 -2023-02-13 17:39:14,744 - Epoch: [46][ 40/ 135] Loss 0.372075 Top1 82.197266 Top5 97.197266 -2023-02-13 17:39:14,875 - Epoch: [46][ 50/ 135] Loss 0.364320 Top1 82.281250 Top5 97.273438 -2023-02-13 17:39:15,004 - Epoch: [46][ 60/ 135] Loss 0.358963 Top1 82.096354 Top5 97.135417 -2023-02-13 17:39:15,132 - Epoch: [46][ 70/ 135] Loss 0.356385 Top1 82.226562 Top5 97.181920 -2023-02-13 17:39:15,261 - Epoch: [46][ 80/ 135] Loss 0.355308 Top1 82.246094 Top5 97.236328 -2023-02-13 17:39:15,391 - Epoch: [46][ 90/ 135] Loss 0.357888 Top1 82.170139 Top5 97.135417 -2023-02-13 17:39:15,519 - Epoch: [46][ 100/ 135] Loss 0.359639 Top1 82.160156 Top5 97.128906 -2023-02-13 17:39:15,649 - Epoch: [46][ 110/ 135] Loss 0.355545 Top1 82.223011 Top5 97.187500 -2023-02-13 17:39:15,781 - Epoch: [46][ 120/ 135] Loss 0.356283 Top1 82.304688 Top5 97.187500 -2023-02-13 17:39:15,915 - Epoch: [46][ 130/ 135] Loss 0.357499 Top1 82.277644 Top5 97.190505 -2023-02-13 17:39:15,962 - Epoch: [46][ 135/ 135] Loss 0.362065 Top1 82.195214 Top5 97.152517 -2023-02-13 17:39:16,034 - ==> Top1: 82.195 Top5: 97.153 Loss: 0.362 - -2023-02-13 17:39:16,035 - ==> Confusion: -[[ 849 6 6 4 4 4 1 3 4 51 0 2 1 6 2 6 2 4 1 1 10] - [ 2 895 2 5 6 58 4 25 5 3 4 2 2 0 1 4 2 0 4 0 9] - [ 12 5 938 12 3 2 20 18 1 1 3 2 1 3 3 6 5 4 8 3 8] - [ 3 1 29 887 0 4 1 3 2 2 19 0 7 1 19 3 2 8 15 2 8] - [ 25 8 3 0 958 19 0 2 0 3 1 9 1 2 7 7 12 1 2 1 5] - [ 2 11 2 4 5 975 5 18 3 3 4 6 6 12 2 0 1 1 3 4 3] - [ 1 6 22 2 2 1 1031 6 0 1 6 1 1 0 0 5 2 1 2 7 2] - [ 3 3 12 0 1 49 1 902 2 1 5 4 4 1 0 0 0 2 17 13 4] - [ 18 4 0 1 1 5 0 0 864 51 13 3 3 17 16 2 2 2 3 0 4] - [ 91 4 3 2 3 4 0 3 39 829 0 1 2 16 4 1 0 2 2 2 4] - [ 1 3 8 9 1 0 4 8 20 0 967 2 1 9 2 1 2 0 12 0 1] - [ 0 2 4 0 1 28 0 6 1 0 1 885 25 7 2 6 5 13 4 14 1] - [ 0 0 3 7 1 4 0 0 3 0 3 29 864 1 5 4 2 21 2 6 4] - [ 3 5 6 0 3 22 2 3 11 14 11 6 1 915 1 7 2 2 2 1 7] - [ 7 6 2 18 4 2 0 1 20 8 5 1 4 2 982 1 1 5 8 1 14] - [ 4 1 3 1 6 5 3 2 0 2 0 1 10 1 1 966 16 13 0 4 7] - [ 3 7 0 1 6 4 2 0 4 1 0 1 4 2 1 7 1001 1 1 3 12] - [ 6 4 0 5 1 3 2 1 0 1 1 10 12 0 3 9 1 983 0 4 5] - [ 3 4 4 12 1 0 0 32 7 0 3 1 4 0 11 2 1 1 996 2 2] - [ 0 1 1 2 1 11 9 21 0 1 0 16 5 3 0 4 6 1 3 1059 4] - [ 178 215 284 143 111 300 82 218 95 107 200 116 353 346 155 121 313 134 185 322 9456]] - -2023-02-13 17:39:16,036 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:39:16,036 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:39:16,042 - - -2023-02-13 17:39:16,042 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:39:17,014 - Epoch: [47][ 10/ 1207] Overall Loss 0.356805 Objective Loss 0.356805 LR 0.001000 Time 0.097111 -2023-02-13 17:39:17,212 - Epoch: [47][ 20/ 1207] Overall Loss 0.331663 Objective Loss 0.331663 LR 0.001000 Time 0.058475 -2023-02-13 17:39:17,406 - Epoch: [47][ 30/ 1207] Overall Loss 0.326122 Objective Loss 0.326122 LR 0.001000 Time 0.045424 -2023-02-13 17:39:17,603 - Epoch: [47][ 40/ 1207] Overall Loss 0.322209 Objective Loss 0.322209 LR 0.001000 Time 0.038994 -2023-02-13 17:39:17,797 - Epoch: [47][ 50/ 1207] Overall Loss 0.319551 Objective Loss 0.319551 LR 0.001000 Time 0.035064 -2023-02-13 17:39:17,994 - Epoch: [47][ 60/ 1207] Overall Loss 0.325089 Objective Loss 0.325089 LR 0.001000 Time 0.032496 -2023-02-13 17:39:18,189 - Epoch: [47][ 70/ 1207] Overall Loss 0.324885 Objective Loss 0.324885 LR 0.001000 Time 0.030634 -2023-02-13 17:39:18,387 - Epoch: [47][ 80/ 1207] Overall Loss 0.323385 Objective Loss 0.323385 LR 0.001000 Time 0.029273 -2023-02-13 17:39:18,581 - Epoch: [47][ 90/ 1207] Overall Loss 0.320327 Objective Loss 0.320327 LR 0.001000 Time 0.028174 -2023-02-13 17:39:18,778 - Epoch: [47][ 100/ 1207] Overall Loss 0.322768 Objective Loss 0.322768 LR 0.001000 Time 0.027318 -2023-02-13 17:39:18,973 - Epoch: [47][ 110/ 1207] Overall Loss 0.320142 Objective Loss 0.320142 LR 0.001000 Time 0.026606 -2023-02-13 17:39:19,171 - Epoch: [47][ 120/ 1207] Overall Loss 0.318042 Objective Loss 0.318042 LR 0.001000 Time 0.026038 -2023-02-13 17:39:19,366 - Epoch: [47][ 130/ 1207] Overall Loss 0.319002 Objective Loss 0.319002 LR 0.001000 Time 0.025532 -2023-02-13 17:39:19,564 - Epoch: [47][ 140/ 1207] Overall Loss 0.320666 Objective Loss 0.320666 LR 0.001000 Time 0.025117 -2023-02-13 17:39:19,758 - Epoch: [47][ 150/ 1207] Overall Loss 0.321279 Objective Loss 0.321279 LR 0.001000 Time 0.024732 -2023-02-13 17:39:19,955 - Epoch: [47][ 160/ 1207] Overall Loss 0.319385 Objective Loss 0.319385 LR 0.001000 Time 0.024416 -2023-02-13 17:39:20,149 - Epoch: [47][ 170/ 1207] Overall Loss 0.318448 Objective Loss 0.318448 LR 0.001000 Time 0.024123 -2023-02-13 17:39:20,347 - Epoch: [47][ 180/ 1207] Overall Loss 0.319783 Objective Loss 0.319783 LR 0.001000 Time 0.023880 -2023-02-13 17:39:20,542 - Epoch: [47][ 190/ 1207] Overall Loss 0.319492 Objective Loss 0.319492 LR 0.001000 Time 0.023644 -2023-02-13 17:39:20,739 - Epoch: [47][ 200/ 1207] Overall Loss 0.320829 Objective Loss 0.320829 LR 0.001000 Time 0.023448 -2023-02-13 17:39:20,934 - Epoch: [47][ 210/ 1207] Overall Loss 0.321812 Objective Loss 0.321812 LR 0.001000 Time 0.023259 -2023-02-13 17:39:21,132 - Epoch: [47][ 220/ 1207] Overall Loss 0.323451 Objective Loss 0.323451 LR 0.001000 Time 0.023098 -2023-02-13 17:39:21,327 - Epoch: [47][ 230/ 1207] Overall Loss 0.323058 Objective Loss 0.323058 LR 0.001000 Time 0.022940 -2023-02-13 17:39:21,526 - Epoch: [47][ 240/ 1207] Overall Loss 0.325550 Objective Loss 0.325550 LR 0.001000 Time 0.022812 -2023-02-13 17:39:21,723 - Epoch: [47][ 250/ 1207] Overall Loss 0.324601 Objective Loss 0.324601 LR 0.001000 Time 0.022686 -2023-02-13 17:39:21,923 - Epoch: [47][ 260/ 1207] Overall Loss 0.325409 Objective Loss 0.325409 LR 0.001000 Time 0.022580 -2023-02-13 17:39:22,119 - Epoch: [47][ 270/ 1207] Overall Loss 0.326428 Objective Loss 0.326428 LR 0.001000 Time 0.022470 -2023-02-13 17:39:22,319 - Epoch: [47][ 280/ 1207] Overall Loss 0.326832 Objective Loss 0.326832 LR 0.001000 Time 0.022380 -2023-02-13 17:39:22,516 - Epoch: [47][ 290/ 1207] Overall Loss 0.326933 Objective Loss 0.326933 LR 0.001000 Time 0.022285 -2023-02-13 17:39:22,716 - Epoch: [47][ 300/ 1207] Overall Loss 0.326530 Objective Loss 0.326530 LR 0.001000 Time 0.022208 -2023-02-13 17:39:22,914 - Epoch: [47][ 310/ 1207] Overall Loss 0.327172 Objective Loss 0.327172 LR 0.001000 Time 0.022129 -2023-02-13 17:39:23,114 - Epoch: [47][ 320/ 1207] Overall Loss 0.327793 Objective Loss 0.327793 LR 0.001000 Time 0.022061 -2023-02-13 17:39:23,312 - Epoch: [47][ 330/ 1207] Overall Loss 0.326688 Objective Loss 0.326688 LR 0.001000 Time 0.021992 -2023-02-13 17:39:23,511 - Epoch: [47][ 340/ 1207] Overall Loss 0.327954 Objective Loss 0.327954 LR 0.001000 Time 0.021930 -2023-02-13 17:39:23,709 - Epoch: [47][ 350/ 1207] Overall Loss 0.327452 Objective Loss 0.327452 LR 0.001000 Time 0.021867 -2023-02-13 17:39:23,908 - Epoch: [47][ 360/ 1207] Overall Loss 0.327512 Objective Loss 0.327512 LR 0.001000 Time 0.021810 -2023-02-13 17:39:24,105 - Epoch: [47][ 370/ 1207] Overall Loss 0.327432 Objective Loss 0.327432 LR 0.001000 Time 0.021754 -2023-02-13 17:39:24,305 - Epoch: [47][ 380/ 1207] Overall Loss 0.327591 Objective Loss 0.327591 LR 0.001000 Time 0.021707 -2023-02-13 17:39:24,503 - Epoch: [47][ 390/ 1207] Overall Loss 0.328051 Objective Loss 0.328051 LR 0.001000 Time 0.021656 -2023-02-13 17:39:24,701 - Epoch: [47][ 400/ 1207] Overall Loss 0.328160 Objective Loss 0.328160 LR 0.001000 Time 0.021610 -2023-02-13 17:39:24,899 - Epoch: [47][ 410/ 1207] Overall Loss 0.328470 Objective Loss 0.328470 LR 0.001000 Time 0.021563 -2023-02-13 17:39:25,099 - Epoch: [47][ 420/ 1207] Overall Loss 0.328001 Objective Loss 0.328001 LR 0.001000 Time 0.021525 -2023-02-13 17:39:25,291 - Epoch: [47][ 430/ 1207] Overall Loss 0.327762 Objective Loss 0.327762 LR 0.001000 Time 0.021472 -2023-02-13 17:39:25,484 - Epoch: [47][ 440/ 1207] Overall Loss 0.328801 Objective Loss 0.328801 LR 0.001000 Time 0.021422 -2023-02-13 17:39:25,677 - Epoch: [47][ 450/ 1207] Overall Loss 0.328203 Objective Loss 0.328203 LR 0.001000 Time 0.021372 -2023-02-13 17:39:25,871 - Epoch: [47][ 460/ 1207] Overall Loss 0.328330 Objective Loss 0.328330 LR 0.001000 Time 0.021329 -2023-02-13 17:39:26,064 - Epoch: [47][ 470/ 1207] Overall Loss 0.328099 Objective Loss 0.328099 LR 0.001000 Time 0.021285 -2023-02-13 17:39:26,257 - Epoch: [47][ 480/ 1207] Overall Loss 0.329107 Objective Loss 0.329107 LR 0.001000 Time 0.021244 -2023-02-13 17:39:26,451 - Epoch: [47][ 490/ 1207] Overall Loss 0.329793 Objective Loss 0.329793 LR 0.001000 Time 0.021204 -2023-02-13 17:39:26,644 - Epoch: [47][ 500/ 1207] Overall Loss 0.330124 Objective Loss 0.330124 LR 0.001000 Time 0.021165 -2023-02-13 17:39:26,837 - Epoch: [47][ 510/ 1207] Overall Loss 0.331079 Objective Loss 0.331079 LR 0.001000 Time 0.021129 -2023-02-13 17:39:27,030 - Epoch: [47][ 520/ 1207] Overall Loss 0.331077 Objective Loss 0.331077 LR 0.001000 Time 0.021092 -2023-02-13 17:39:27,223 - Epoch: [47][ 530/ 1207] Overall Loss 0.330782 Objective Loss 0.330782 LR 0.001000 Time 0.021059 -2023-02-13 17:39:27,416 - Epoch: [47][ 540/ 1207] Overall Loss 0.330262 Objective Loss 0.330262 LR 0.001000 Time 0.021025 -2023-02-13 17:39:27,610 - Epoch: [47][ 550/ 1207] Overall Loss 0.330277 Objective Loss 0.330277 LR 0.001000 Time 0.020994 -2023-02-13 17:39:27,802 - Epoch: [47][ 560/ 1207] Overall Loss 0.330363 Objective Loss 0.330363 LR 0.001000 Time 0.020962 -2023-02-13 17:39:27,995 - Epoch: [47][ 570/ 1207] Overall Loss 0.330367 Objective Loss 0.330367 LR 0.001000 Time 0.020933 -2023-02-13 17:39:28,188 - Epoch: [47][ 580/ 1207] Overall Loss 0.331008 Objective Loss 0.331008 LR 0.001000 Time 0.020904 -2023-02-13 17:39:28,381 - Epoch: [47][ 590/ 1207] Overall Loss 0.330722 Objective Loss 0.330722 LR 0.001000 Time 0.020876 -2023-02-13 17:39:28,574 - Epoch: [47][ 600/ 1207] Overall Loss 0.330561 Objective Loss 0.330561 LR 0.001000 Time 0.020849 -2023-02-13 17:39:28,767 - Epoch: [47][ 610/ 1207] Overall Loss 0.330675 Objective Loss 0.330675 LR 0.001000 Time 0.020823 -2023-02-13 17:39:28,959 - Epoch: [47][ 620/ 1207] Overall Loss 0.330410 Objective Loss 0.330410 LR 0.001000 Time 0.020796 -2023-02-13 17:39:29,152 - Epoch: [47][ 630/ 1207] Overall Loss 0.330667 Objective Loss 0.330667 LR 0.001000 Time 0.020772 -2023-02-13 17:39:29,345 - Epoch: [47][ 640/ 1207] Overall Loss 0.330739 Objective Loss 0.330739 LR 0.001000 Time 0.020749 -2023-02-13 17:39:29,539 - Epoch: [47][ 650/ 1207] Overall Loss 0.330612 Objective Loss 0.330612 LR 0.001000 Time 0.020726 -2023-02-13 17:39:29,731 - Epoch: [47][ 660/ 1207] Overall Loss 0.330487 Objective Loss 0.330487 LR 0.001000 Time 0.020703 -2023-02-13 17:39:29,924 - Epoch: [47][ 670/ 1207] Overall Loss 0.330366 Objective Loss 0.330366 LR 0.001000 Time 0.020682 -2023-02-13 17:39:30,117 - Epoch: [47][ 680/ 1207] Overall Loss 0.330658 Objective Loss 0.330658 LR 0.001000 Time 0.020661 -2023-02-13 17:39:30,311 - Epoch: [47][ 690/ 1207] Overall Loss 0.330985 Objective Loss 0.330985 LR 0.001000 Time 0.020642 -2023-02-13 17:39:30,503 - Epoch: [47][ 700/ 1207] Overall Loss 0.331119 Objective Loss 0.331119 LR 0.001000 Time 0.020621 -2023-02-13 17:39:30,696 - Epoch: [47][ 710/ 1207] Overall Loss 0.330757 Objective Loss 0.330757 LR 0.001000 Time 0.020602 -2023-02-13 17:39:30,890 - Epoch: [47][ 720/ 1207] Overall Loss 0.330941 Objective Loss 0.330941 LR 0.001000 Time 0.020584 -2023-02-13 17:39:31,083 - Epoch: [47][ 730/ 1207] Overall Loss 0.331172 Objective Loss 0.331172 LR 0.001000 Time 0.020566 -2023-02-13 17:39:31,277 - Epoch: [47][ 740/ 1207] Overall Loss 0.331478 Objective Loss 0.331478 LR 0.001000 Time 0.020549 -2023-02-13 17:39:31,474 - Epoch: [47][ 750/ 1207] Overall Loss 0.331854 Objective Loss 0.331854 LR 0.001000 Time 0.020538 -2023-02-13 17:39:31,669 - Epoch: [47][ 760/ 1207] Overall Loss 0.332220 Objective Loss 0.332220 LR 0.001000 Time 0.020524 -2023-02-13 17:39:31,869 - Epoch: [47][ 770/ 1207] Overall Loss 0.331850 Objective Loss 0.331850 LR 0.001000 Time 0.020517 -2023-02-13 17:39:32,065 - Epoch: [47][ 780/ 1207] Overall Loss 0.332060 Objective Loss 0.332060 LR 0.001000 Time 0.020504 -2023-02-13 17:39:32,265 - Epoch: [47][ 790/ 1207] Overall Loss 0.332175 Objective Loss 0.332175 LR 0.001000 Time 0.020498 -2023-02-13 17:39:32,461 - Epoch: [47][ 800/ 1207] Overall Loss 0.332345 Objective Loss 0.332345 LR 0.001000 Time 0.020486 -2023-02-13 17:39:32,660 - Epoch: [47][ 810/ 1207] Overall Loss 0.332554 Objective Loss 0.332554 LR 0.001000 Time 0.020479 -2023-02-13 17:39:32,856 - Epoch: [47][ 820/ 1207] Overall Loss 0.332335 Objective Loss 0.332335 LR 0.001000 Time 0.020467 -2023-02-13 17:39:33,056 - Epoch: [47][ 830/ 1207] Overall Loss 0.332571 Objective Loss 0.332571 LR 0.001000 Time 0.020461 -2023-02-13 17:39:33,252 - Epoch: [47][ 840/ 1207] Overall Loss 0.332740 Objective Loss 0.332740 LR 0.001000 Time 0.020450 -2023-02-13 17:39:33,452 - Epoch: [47][ 850/ 1207] Overall Loss 0.333085 Objective Loss 0.333085 LR 0.001000 Time 0.020444 -2023-02-13 17:39:33,648 - Epoch: [47][ 860/ 1207] Overall Loss 0.333541 Objective Loss 0.333541 LR 0.001000 Time 0.020434 -2023-02-13 17:39:33,848 - Epoch: [47][ 870/ 1207] Overall Loss 0.333591 Objective Loss 0.333591 LR 0.001000 Time 0.020429 -2023-02-13 17:39:34,044 - Epoch: [47][ 880/ 1207] Overall Loss 0.333660 Objective Loss 0.333660 LR 0.001000 Time 0.020419 -2023-02-13 17:39:34,243 - Epoch: [47][ 890/ 1207] Overall Loss 0.333851 Objective Loss 0.333851 LR 0.001000 Time 0.020413 -2023-02-13 17:39:34,440 - Epoch: [47][ 900/ 1207] Overall Loss 0.334043 Objective Loss 0.334043 LR 0.001000 Time 0.020404 -2023-02-13 17:39:34,639 - Epoch: [47][ 910/ 1207] Overall Loss 0.334234 Objective Loss 0.334234 LR 0.001000 Time 0.020399 -2023-02-13 17:39:34,836 - Epoch: [47][ 920/ 1207] Overall Loss 0.334567 Objective Loss 0.334567 LR 0.001000 Time 0.020390 -2023-02-13 17:39:35,035 - Epoch: [47][ 930/ 1207] Overall Loss 0.334392 Objective Loss 0.334392 LR 0.001000 Time 0.020384 -2023-02-13 17:39:35,232 - Epoch: [47][ 940/ 1207] Overall Loss 0.334320 Objective Loss 0.334320 LR 0.001000 Time 0.020377 -2023-02-13 17:39:35,429 - Epoch: [47][ 950/ 1207] Overall Loss 0.334122 Objective Loss 0.334122 LR 0.001000 Time 0.020370 -2023-02-13 17:39:35,623 - Epoch: [47][ 960/ 1207] Overall Loss 0.334278 Objective Loss 0.334278 LR 0.001000 Time 0.020358 -2023-02-13 17:39:35,820 - Epoch: [47][ 970/ 1207] Overall Loss 0.334219 Objective Loss 0.334219 LR 0.001000 Time 0.020352 -2023-02-13 17:39:36,015 - Epoch: [47][ 980/ 1207] Overall Loss 0.334341 Objective Loss 0.334341 LR 0.001000 Time 0.020342 -2023-02-13 17:39:36,212 - Epoch: [47][ 990/ 1207] Overall Loss 0.334779 Objective Loss 0.334779 LR 0.001000 Time 0.020336 -2023-02-13 17:39:36,405 - Epoch: [47][ 1000/ 1207] Overall Loss 0.335372 Objective Loss 0.335372 LR 0.001000 Time 0.020325 -2023-02-13 17:39:36,602 - Epoch: [47][ 1010/ 1207] Overall Loss 0.335863 Objective Loss 0.335863 LR 0.001000 Time 0.020319 -2023-02-13 17:39:36,796 - Epoch: [47][ 1020/ 1207] Overall Loss 0.335831 Objective Loss 0.335831 LR 0.001000 Time 0.020309 -2023-02-13 17:39:36,994 - Epoch: [47][ 1030/ 1207] Overall Loss 0.335940 Objective Loss 0.335940 LR 0.001000 Time 0.020304 -2023-02-13 17:39:37,187 - Epoch: [47][ 1040/ 1207] Overall Loss 0.336007 Objective Loss 0.336007 LR 0.001000 Time 0.020294 -2023-02-13 17:39:37,385 - Epoch: [47][ 1050/ 1207] Overall Loss 0.335973 Objective Loss 0.335973 LR 0.001000 Time 0.020289 -2023-02-13 17:39:37,578 - Epoch: [47][ 1060/ 1207] Overall Loss 0.336159 Objective Loss 0.336159 LR 0.001000 Time 0.020279 -2023-02-13 17:39:37,775 - Epoch: [47][ 1070/ 1207] Overall Loss 0.335938 Objective Loss 0.335938 LR 0.001000 Time 0.020274 -2023-02-13 17:39:37,969 - Epoch: [47][ 1080/ 1207] Overall Loss 0.335741 Objective Loss 0.335741 LR 0.001000 Time 0.020265 -2023-02-13 17:39:38,166 - Epoch: [47][ 1090/ 1207] Overall Loss 0.335912 Objective Loss 0.335912 LR 0.001000 Time 0.020260 -2023-02-13 17:39:38,361 - Epoch: [47][ 1100/ 1207] Overall Loss 0.335995 Objective Loss 0.335995 LR 0.001000 Time 0.020252 -2023-02-13 17:39:38,558 - Epoch: [47][ 1110/ 1207] Overall Loss 0.335901 Objective Loss 0.335901 LR 0.001000 Time 0.020247 -2023-02-13 17:39:38,753 - Epoch: [47][ 1120/ 1207] Overall Loss 0.335995 Objective Loss 0.335995 LR 0.001000 Time 0.020240 -2023-02-13 17:39:38,949 - Epoch: [47][ 1130/ 1207] Overall Loss 0.336222 Objective Loss 0.336222 LR 0.001000 Time 0.020234 -2023-02-13 17:39:39,143 - Epoch: [47][ 1140/ 1207] Overall Loss 0.336211 Objective Loss 0.336211 LR 0.001000 Time 0.020226 -2023-02-13 17:39:39,341 - Epoch: [47][ 1150/ 1207] Overall Loss 0.336339 Objective Loss 0.336339 LR 0.001000 Time 0.020222 -2023-02-13 17:39:39,535 - Epoch: [47][ 1160/ 1207] Overall Loss 0.336638 Objective Loss 0.336638 LR 0.001000 Time 0.020214 -2023-02-13 17:39:39,732 - Epoch: [47][ 1170/ 1207] Overall Loss 0.336739 Objective Loss 0.336739 LR 0.001000 Time 0.020210 -2023-02-13 17:39:39,926 - Epoch: [47][ 1180/ 1207] Overall Loss 0.337004 Objective Loss 0.337004 LR 0.001000 Time 0.020203 -2023-02-13 17:39:40,125 - Epoch: [47][ 1190/ 1207] Overall Loss 0.337135 Objective Loss 0.337135 LR 0.001000 Time 0.020200 -2023-02-13 17:39:40,373 - Epoch: [47][ 1200/ 1207] Overall Loss 0.337098 Objective Loss 0.337098 LR 0.001000 Time 0.020238 -2023-02-13 17:39:40,489 - Epoch: [47][ 1207/ 1207] Overall Loss 0.337162 Objective Loss 0.337162 Top1 78.048780 Top5 97.865854 LR 0.001000 Time 0.020217 -2023-02-13 17:39:40,561 - --- validate (epoch=47)----------- -2023-02-13 17:39:40,561 - 34311 samples (256 per mini-batch) -2023-02-13 17:39:40,957 - Epoch: [47][ 10/ 135] Loss 0.406963 Top1 78.945312 Top5 96.757812 -2023-02-13 17:39:41,082 - Epoch: [47][ 20/ 135] Loss 0.382595 Top1 80.058594 Top5 96.875000 -2023-02-13 17:39:41,208 - Epoch: [47][ 30/ 135] Loss 0.381018 Top1 80.104167 Top5 96.796875 -2023-02-13 17:39:41,335 - Epoch: [47][ 40/ 135] Loss 0.383792 Top1 80.253906 Top5 96.738281 -2023-02-13 17:39:41,463 - Epoch: [47][ 50/ 135] Loss 0.386760 Top1 80.117188 Top5 96.812500 -2023-02-13 17:39:41,591 - Epoch: [47][ 60/ 135] Loss 0.387632 Top1 80.175781 Top5 96.835938 -2023-02-13 17:39:41,723 - Epoch: [47][ 70/ 135] Loss 0.382159 Top1 80.468750 Top5 96.919643 -2023-02-13 17:39:41,847 - Epoch: [47][ 80/ 135] Loss 0.381746 Top1 80.400391 Top5 96.884766 -2023-02-13 17:39:41,972 - Epoch: [47][ 90/ 135] Loss 0.380637 Top1 80.451389 Top5 96.931424 -2023-02-13 17:39:42,101 - Epoch: [47][ 100/ 135] Loss 0.382558 Top1 80.417969 Top5 96.894531 -2023-02-13 17:39:42,230 - Epoch: [47][ 110/ 135] Loss 0.379561 Top1 80.461648 Top5 96.906960 -2023-02-13 17:39:42,360 - Epoch: [47][ 120/ 135] Loss 0.380442 Top1 80.449219 Top5 96.946615 -2023-02-13 17:39:42,488 - Epoch: [47][ 130/ 135] Loss 0.379750 Top1 80.477764 Top5 96.932091 -2023-02-13 17:39:42,535 - Epoch: [47][ 135/ 135] Loss 0.388564 Top1 80.417359 Top5 96.916441 -2023-02-13 17:39:42,603 - ==> Top1: 80.417 Top5: 96.916 Loss: 0.389 - -2023-02-13 17:39:42,604 - ==> Confusion: -[[ 808 4 5 1 10 2 0 1 6 95 0 3 2 3 3 5 2 4 4 1 8] - [ 1 948 2 3 8 21 4 14 5 2 1 2 0 0 5 3 2 0 9 0 3] - [ 7 6 949 20 7 1 16 8 1 1 2 1 2 6 3 6 3 5 6 5 3] - [ 8 0 26 896 0 3 2 2 4 1 11 0 1 2 24 0 2 4 21 1 8] - [ 15 10 1 0 982 9 1 1 0 6 0 6 1 2 10 9 3 6 0 2 2] - [ 4 49 2 3 9 921 2 21 4 8 2 14 4 12 1 2 1 3 1 1 6] - [ 1 6 29 4 1 7 1009 12 0 1 5 2 4 1 0 4 0 3 1 7 2] - [ 3 18 16 1 1 33 5 904 0 2 0 5 2 0 0 0 0 4 19 5 6] - [ 12 4 0 2 1 2 0 0 906 42 4 2 1 5 13 4 0 2 9 0 0] - [ 62 3 1 1 2 4 0 3 53 855 0 1 0 10 4 2 0 5 0 0 6] - [ 1 5 5 10 1 2 1 10 37 0 935 2 2 10 4 1 0 2 16 1 6] - [ 0 3 2 0 0 13 1 7 2 4 0 908 23 9 2 2 2 7 4 16 0] - [ 1 4 0 15 1 3 0 2 4 1 0 46 833 3 7 5 1 24 1 3 5] - [ 6 4 2 0 6 18 0 1 35 30 11 8 3 877 7 6 1 1 1 3 4] - [ 10 5 2 13 7 3 0 0 32 5 2 3 4 0 982 1 0 4 13 0 6] - [ 5 4 7 1 10 3 4 2 1 1 1 7 8 2 1 950 6 18 1 8 6] - [ 3 17 1 1 9 3 0 1 5 0 1 7 3 4 3 18 960 1 1 4 19] - [ 5 1 1 8 1 0 2 2 3 2 1 18 17 2 1 9 0 971 2 1 4] - [ 2 2 10 18 0 3 0 29 5 2 4 4 2 0 10 2 1 3 987 0 2] - [ 0 3 1 1 1 9 8 31 1 1 0 25 6 6 0 3 3 4 0 1044 1] - [ 224 391 307 161 180 243 78 256 194 161 193 178 322 332 236 107 176 139 273 316 8967]] - -2023-02-13 17:39:42,605 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:39:42,605 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:39:42,611 - - -2023-02-13 17:39:42,611 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:39:43,600 - Epoch: [48][ 10/ 1207] Overall Loss 0.380465 Objective Loss 0.380465 LR 0.001000 Time 0.098833 -2023-02-13 17:39:43,795 - Epoch: [48][ 20/ 1207] Overall Loss 0.357933 Objective Loss 0.357933 LR 0.001000 Time 0.059181 -2023-02-13 17:39:43,984 - Epoch: [48][ 30/ 1207] Overall Loss 0.357778 Objective Loss 0.357778 LR 0.001000 Time 0.045738 -2023-02-13 17:39:44,174 - Epoch: [48][ 40/ 1207] Overall Loss 0.348189 Objective Loss 0.348189 LR 0.001000 Time 0.039032 -2023-02-13 17:39:44,363 - Epoch: [48][ 50/ 1207] Overall Loss 0.339019 Objective Loss 0.339019 LR 0.001000 Time 0.034996 -2023-02-13 17:39:44,551 - Epoch: [48][ 60/ 1207] Overall Loss 0.342627 Objective Loss 0.342627 LR 0.001000 Time 0.032300 -2023-02-13 17:39:44,741 - Epoch: [48][ 70/ 1207] Overall Loss 0.341694 Objective Loss 0.341694 LR 0.001000 Time 0.030389 -2023-02-13 17:39:44,930 - Epoch: [48][ 80/ 1207] Overall Loss 0.338405 Objective Loss 0.338405 LR 0.001000 Time 0.028956 -2023-02-13 17:39:45,120 - Epoch: [48][ 90/ 1207] Overall Loss 0.337336 Objective Loss 0.337336 LR 0.001000 Time 0.027842 -2023-02-13 17:39:45,309 - Epoch: [48][ 100/ 1207] Overall Loss 0.337721 Objective Loss 0.337721 LR 0.001000 Time 0.026943 -2023-02-13 17:39:45,498 - Epoch: [48][ 110/ 1207] Overall Loss 0.338652 Objective Loss 0.338652 LR 0.001000 Time 0.026206 -2023-02-13 17:39:45,687 - Epoch: [48][ 120/ 1207] Overall Loss 0.338801 Objective Loss 0.338801 LR 0.001000 Time 0.025598 -2023-02-13 17:39:45,878 - Epoch: [48][ 130/ 1207] Overall Loss 0.339648 Objective Loss 0.339648 LR 0.001000 Time 0.025093 -2023-02-13 17:39:46,067 - Epoch: [48][ 140/ 1207] Overall Loss 0.338809 Objective Loss 0.338809 LR 0.001000 Time 0.024650 -2023-02-13 17:39:46,256 - Epoch: [48][ 150/ 1207] Overall Loss 0.337188 Objective Loss 0.337188 LR 0.001000 Time 0.024264 -2023-02-13 17:39:46,445 - Epoch: [48][ 160/ 1207] Overall Loss 0.336467 Objective Loss 0.336467 LR 0.001000 Time 0.023926 -2023-02-13 17:39:46,634 - Epoch: [48][ 170/ 1207] Overall Loss 0.337772 Objective Loss 0.337772 LR 0.001000 Time 0.023628 -2023-02-13 17:39:46,823 - Epoch: [48][ 180/ 1207] Overall Loss 0.335781 Objective Loss 0.335781 LR 0.001000 Time 0.023362 -2023-02-13 17:39:47,012 - Epoch: [48][ 190/ 1207] Overall Loss 0.336311 Objective Loss 0.336311 LR 0.001000 Time 0.023129 -2023-02-13 17:39:47,201 - Epoch: [48][ 200/ 1207] Overall Loss 0.336924 Objective Loss 0.336924 LR 0.001000 Time 0.022914 -2023-02-13 17:39:47,392 - Epoch: [48][ 210/ 1207] Overall Loss 0.336764 Objective Loss 0.336764 LR 0.001000 Time 0.022729 -2023-02-13 17:39:47,580 - Epoch: [48][ 220/ 1207] Overall Loss 0.337275 Objective Loss 0.337275 LR 0.001000 Time 0.022548 -2023-02-13 17:39:47,770 - Epoch: [48][ 230/ 1207] Overall Loss 0.339658 Objective Loss 0.339658 LR 0.001000 Time 0.022394 -2023-02-13 17:39:47,958 - Epoch: [48][ 240/ 1207] Overall Loss 0.338987 Objective Loss 0.338987 LR 0.001000 Time 0.022243 -2023-02-13 17:39:48,148 - Epoch: [48][ 250/ 1207] Overall Loss 0.336947 Objective Loss 0.336947 LR 0.001000 Time 0.022110 -2023-02-13 17:39:48,337 - Epoch: [48][ 260/ 1207] Overall Loss 0.336034 Objective Loss 0.336034 LR 0.001000 Time 0.021984 -2023-02-13 17:39:48,527 - Epoch: [48][ 270/ 1207] Overall Loss 0.335257 Objective Loss 0.335257 LR 0.001000 Time 0.021873 -2023-02-13 17:39:48,716 - Epoch: [48][ 280/ 1207] Overall Loss 0.335636 Objective Loss 0.335636 LR 0.001000 Time 0.021765 -2023-02-13 17:39:48,905 - Epoch: [48][ 290/ 1207] Overall Loss 0.334465 Objective Loss 0.334465 LR 0.001000 Time 0.021668 -2023-02-13 17:39:49,095 - Epoch: [48][ 300/ 1207] Overall Loss 0.335773 Objective Loss 0.335773 LR 0.001000 Time 0.021575 -2023-02-13 17:39:49,285 - Epoch: [48][ 310/ 1207] Overall Loss 0.335092 Objective Loss 0.335092 LR 0.001000 Time 0.021491 -2023-02-13 17:39:49,474 - Epoch: [48][ 320/ 1207] Overall Loss 0.334902 Objective Loss 0.334902 LR 0.001000 Time 0.021410 -2023-02-13 17:39:49,664 - Epoch: [48][ 330/ 1207] Overall Loss 0.334349 Objective Loss 0.334349 LR 0.001000 Time 0.021334 -2023-02-13 17:39:49,852 - Epoch: [48][ 340/ 1207] Overall Loss 0.334577 Objective Loss 0.334577 LR 0.001000 Time 0.021259 -2023-02-13 17:39:50,041 - Epoch: [48][ 350/ 1207] Overall Loss 0.335245 Objective Loss 0.335245 LR 0.001000 Time 0.021193 -2023-02-13 17:39:50,231 - Epoch: [48][ 360/ 1207] Overall Loss 0.335171 Objective Loss 0.335171 LR 0.001000 Time 0.021129 -2023-02-13 17:39:50,421 - Epoch: [48][ 370/ 1207] Overall Loss 0.334725 Objective Loss 0.334725 LR 0.001000 Time 0.021072 -2023-02-13 17:39:50,610 - Epoch: [48][ 380/ 1207] Overall Loss 0.334927 Objective Loss 0.334927 LR 0.001000 Time 0.021013 -2023-02-13 17:39:50,801 - Epoch: [48][ 390/ 1207] Overall Loss 0.334486 Objective Loss 0.334486 LR 0.001000 Time 0.020963 -2023-02-13 17:39:50,990 - Epoch: [48][ 400/ 1207] Overall Loss 0.334908 Objective Loss 0.334908 LR 0.001000 Time 0.020910 -2023-02-13 17:39:51,180 - Epoch: [48][ 410/ 1207] Overall Loss 0.335189 Objective Loss 0.335189 LR 0.001000 Time 0.020864 -2023-02-13 17:39:51,370 - Epoch: [48][ 420/ 1207] Overall Loss 0.336143 Objective Loss 0.336143 LR 0.001000 Time 0.020817 -2023-02-13 17:39:51,560 - Epoch: [48][ 430/ 1207] Overall Loss 0.335956 Objective Loss 0.335956 LR 0.001000 Time 0.020776 -2023-02-13 17:39:51,750 - Epoch: [48][ 440/ 1207] Overall Loss 0.336140 Objective Loss 0.336140 LR 0.001000 Time 0.020733 -2023-02-13 17:39:51,941 - Epoch: [48][ 450/ 1207] Overall Loss 0.335850 Objective Loss 0.335850 LR 0.001000 Time 0.020696 -2023-02-13 17:39:52,129 - Epoch: [48][ 460/ 1207] Overall Loss 0.336099 Objective Loss 0.336099 LR 0.001000 Time 0.020655 -2023-02-13 17:39:52,321 - Epoch: [48][ 470/ 1207] Overall Loss 0.335768 Objective Loss 0.335768 LR 0.001000 Time 0.020622 -2023-02-13 17:39:52,510 - Epoch: [48][ 480/ 1207] Overall Loss 0.336231 Objective Loss 0.336231 LR 0.001000 Time 0.020586 -2023-02-13 17:39:52,700 - Epoch: [48][ 490/ 1207] Overall Loss 0.335403 Objective Loss 0.335403 LR 0.001000 Time 0.020554 -2023-02-13 17:39:52,889 - Epoch: [48][ 500/ 1207] Overall Loss 0.335701 Objective Loss 0.335701 LR 0.001000 Time 0.020520 -2023-02-13 17:39:53,080 - Epoch: [48][ 510/ 1207] Overall Loss 0.336259 Objective Loss 0.336259 LR 0.001000 Time 0.020490 -2023-02-13 17:39:53,269 - Epoch: [48][ 520/ 1207] Overall Loss 0.335921 Objective Loss 0.335921 LR 0.001000 Time 0.020459 -2023-02-13 17:39:53,456 - Epoch: [48][ 530/ 1207] Overall Loss 0.335562 Objective Loss 0.335562 LR 0.001000 Time 0.020426 -2023-02-13 17:39:53,643 - Epoch: [48][ 540/ 1207] Overall Loss 0.335347 Objective Loss 0.335347 LR 0.001000 Time 0.020393 -2023-02-13 17:39:53,830 - Epoch: [48][ 550/ 1207] Overall Loss 0.334939 Objective Loss 0.334939 LR 0.001000 Time 0.020361 -2023-02-13 17:39:54,016 - Epoch: [48][ 560/ 1207] Overall Loss 0.335282 Objective Loss 0.335282 LR 0.001000 Time 0.020330 -2023-02-13 17:39:54,203 - Epoch: [48][ 570/ 1207] Overall Loss 0.335431 Objective Loss 0.335431 LR 0.001000 Time 0.020300 -2023-02-13 17:39:54,390 - Epoch: [48][ 580/ 1207] Overall Loss 0.335984 Objective Loss 0.335984 LR 0.001000 Time 0.020272 -2023-02-13 17:39:54,577 - Epoch: [48][ 590/ 1207] Overall Loss 0.336157 Objective Loss 0.336157 LR 0.001000 Time 0.020244 -2023-02-13 17:39:54,763 - Epoch: [48][ 600/ 1207] Overall Loss 0.335906 Objective Loss 0.335906 LR 0.001000 Time 0.020216 -2023-02-13 17:39:54,950 - Epoch: [48][ 610/ 1207] Overall Loss 0.335862 Objective Loss 0.335862 LR 0.001000 Time 0.020191 -2023-02-13 17:39:55,136 - Epoch: [48][ 620/ 1207] Overall Loss 0.336157 Objective Loss 0.336157 LR 0.001000 Time 0.020165 -2023-02-13 17:39:55,324 - Epoch: [48][ 630/ 1207] Overall Loss 0.336485 Objective Loss 0.336485 LR 0.001000 Time 0.020142 -2023-02-13 17:39:55,512 - Epoch: [48][ 640/ 1207] Overall Loss 0.336741 Objective Loss 0.336741 LR 0.001000 Time 0.020121 -2023-02-13 17:39:55,699 - Epoch: [48][ 650/ 1207] Overall Loss 0.336499 Objective Loss 0.336499 LR 0.001000 Time 0.020099 -2023-02-13 17:39:55,887 - Epoch: [48][ 660/ 1207] Overall Loss 0.336555 Objective Loss 0.336555 LR 0.001000 Time 0.020079 -2023-02-13 17:39:56,074 - Epoch: [48][ 670/ 1207] Overall Loss 0.337235 Objective Loss 0.337235 LR 0.001000 Time 0.020057 -2023-02-13 17:39:56,261 - Epoch: [48][ 680/ 1207] Overall Loss 0.337498 Objective Loss 0.337498 LR 0.001000 Time 0.020037 -2023-02-13 17:39:56,450 - Epoch: [48][ 690/ 1207] Overall Loss 0.338128 Objective Loss 0.338128 LR 0.001000 Time 0.020019 -2023-02-13 17:39:56,638 - Epoch: [48][ 700/ 1207] Overall Loss 0.338072 Objective Loss 0.338072 LR 0.001000 Time 0.020001 -2023-02-13 17:39:56,826 - Epoch: [48][ 710/ 1207] Overall Loss 0.338624 Objective Loss 0.338624 LR 0.001000 Time 0.019984 -2023-02-13 17:39:57,014 - Epoch: [48][ 720/ 1207] Overall Loss 0.338648 Objective Loss 0.338648 LR 0.001000 Time 0.019967 -2023-02-13 17:39:57,201 - Epoch: [48][ 730/ 1207] Overall Loss 0.338424 Objective Loss 0.338424 LR 0.001000 Time 0.019949 -2023-02-13 17:39:57,389 - Epoch: [48][ 740/ 1207] Overall Loss 0.338866 Objective Loss 0.338866 LR 0.001000 Time 0.019933 -2023-02-13 17:39:57,577 - Epoch: [48][ 750/ 1207] Overall Loss 0.339022 Objective Loss 0.339022 LR 0.001000 Time 0.019918 -2023-02-13 17:39:57,764 - Epoch: [48][ 760/ 1207] Overall Loss 0.338965 Objective Loss 0.338965 LR 0.001000 Time 0.019902 -2023-02-13 17:39:57,952 - Epoch: [48][ 770/ 1207] Overall Loss 0.338623 Objective Loss 0.338623 LR 0.001000 Time 0.019886 -2023-02-13 17:39:58,139 - Epoch: [48][ 780/ 1207] Overall Loss 0.339032 Objective Loss 0.339032 LR 0.001000 Time 0.019870 -2023-02-13 17:39:58,326 - Epoch: [48][ 790/ 1207] Overall Loss 0.338988 Objective Loss 0.338988 LR 0.001000 Time 0.019856 -2023-02-13 17:39:58,514 - Epoch: [48][ 800/ 1207] Overall Loss 0.338832 Objective Loss 0.338832 LR 0.001000 Time 0.019841 -2023-02-13 17:39:58,702 - Epoch: [48][ 810/ 1207] Overall Loss 0.338747 Objective Loss 0.338747 LR 0.001000 Time 0.019829 -2023-02-13 17:39:58,890 - Epoch: [48][ 820/ 1207] Overall Loss 0.339118 Objective Loss 0.339118 LR 0.001000 Time 0.019815 -2023-02-13 17:39:59,077 - Epoch: [48][ 830/ 1207] Overall Loss 0.339722 Objective Loss 0.339722 LR 0.001000 Time 0.019802 -2023-02-13 17:39:59,265 - Epoch: [48][ 840/ 1207] Overall Loss 0.340070 Objective Loss 0.340070 LR 0.001000 Time 0.019789 -2023-02-13 17:39:59,453 - Epoch: [48][ 850/ 1207] Overall Loss 0.340495 Objective Loss 0.340495 LR 0.001000 Time 0.019778 -2023-02-13 17:39:59,641 - Epoch: [48][ 860/ 1207] Overall Loss 0.340541 Objective Loss 0.340541 LR 0.001000 Time 0.019765 -2023-02-13 17:39:59,827 - Epoch: [48][ 870/ 1207] Overall Loss 0.340813 Objective Loss 0.340813 LR 0.001000 Time 0.019752 -2023-02-13 17:40:00,014 - Epoch: [48][ 880/ 1207] Overall Loss 0.340934 Objective Loss 0.340934 LR 0.001000 Time 0.019739 -2023-02-13 17:40:00,201 - Epoch: [48][ 890/ 1207] Overall Loss 0.341184 Objective Loss 0.341184 LR 0.001000 Time 0.019727 -2023-02-13 17:40:00,389 - Epoch: [48][ 900/ 1207] Overall Loss 0.341349 Objective Loss 0.341349 LR 0.001000 Time 0.019716 -2023-02-13 17:40:00,576 - Epoch: [48][ 910/ 1207] Overall Loss 0.341118 Objective Loss 0.341118 LR 0.001000 Time 0.019704 -2023-02-13 17:40:00,762 - Epoch: [48][ 920/ 1207] Overall Loss 0.341358 Objective Loss 0.341358 LR 0.001000 Time 0.019693 -2023-02-13 17:40:00,951 - Epoch: [48][ 930/ 1207] Overall Loss 0.341114 Objective Loss 0.341114 LR 0.001000 Time 0.019683 -2023-02-13 17:40:01,138 - Epoch: [48][ 940/ 1207] Overall Loss 0.341157 Objective Loss 0.341157 LR 0.001000 Time 0.019672 -2023-02-13 17:40:01,326 - Epoch: [48][ 950/ 1207] Overall Loss 0.341568 Objective Loss 0.341568 LR 0.001000 Time 0.019663 -2023-02-13 17:40:01,513 - Epoch: [48][ 960/ 1207] Overall Loss 0.341655 Objective Loss 0.341655 LR 0.001000 Time 0.019652 -2023-02-13 17:40:01,701 - Epoch: [48][ 970/ 1207] Overall Loss 0.341311 Objective Loss 0.341311 LR 0.001000 Time 0.019643 -2023-02-13 17:40:01,889 - Epoch: [48][ 980/ 1207] Overall Loss 0.341006 Objective Loss 0.341006 LR 0.001000 Time 0.019634 -2023-02-13 17:40:02,076 - Epoch: [48][ 990/ 1207] Overall Loss 0.340969 Objective Loss 0.340969 LR 0.001000 Time 0.019625 -2023-02-13 17:40:02,264 - Epoch: [48][ 1000/ 1207] Overall Loss 0.340827 Objective Loss 0.340827 LR 0.001000 Time 0.019616 -2023-02-13 17:40:02,452 - Epoch: [48][ 1010/ 1207] Overall Loss 0.340498 Objective Loss 0.340498 LR 0.001000 Time 0.019607 -2023-02-13 17:40:02,640 - Epoch: [48][ 1020/ 1207] Overall Loss 0.340656 Objective Loss 0.340656 LR 0.001000 Time 0.019599 -2023-02-13 17:40:02,827 - Epoch: [48][ 1030/ 1207] Overall Loss 0.340649 Objective Loss 0.340649 LR 0.001000 Time 0.019590 -2023-02-13 17:40:03,015 - Epoch: [48][ 1040/ 1207] Overall Loss 0.340792 Objective Loss 0.340792 LR 0.001000 Time 0.019582 -2023-02-13 17:40:03,202 - Epoch: [48][ 1050/ 1207] Overall Loss 0.340763 Objective Loss 0.340763 LR 0.001000 Time 0.019574 -2023-02-13 17:40:03,390 - Epoch: [48][ 1060/ 1207] Overall Loss 0.340954 Objective Loss 0.340954 LR 0.001000 Time 0.019565 -2023-02-13 17:40:03,577 - Epoch: [48][ 1070/ 1207] Overall Loss 0.340927 Objective Loss 0.340927 LR 0.001000 Time 0.019557 -2023-02-13 17:40:03,765 - Epoch: [48][ 1080/ 1207] Overall Loss 0.340962 Objective Loss 0.340962 LR 0.001000 Time 0.019550 -2023-02-13 17:40:03,951 - Epoch: [48][ 1090/ 1207] Overall Loss 0.341260 Objective Loss 0.341260 LR 0.001000 Time 0.019541 -2023-02-13 17:40:04,138 - Epoch: [48][ 1100/ 1207] Overall Loss 0.341270 Objective Loss 0.341270 LR 0.001000 Time 0.019533 -2023-02-13 17:40:04,326 - Epoch: [48][ 1110/ 1207] Overall Loss 0.341644 Objective Loss 0.341644 LR 0.001000 Time 0.019526 -2023-02-13 17:40:04,514 - Epoch: [48][ 1120/ 1207] Overall Loss 0.341834 Objective Loss 0.341834 LR 0.001000 Time 0.019519 -2023-02-13 17:40:04,701 - Epoch: [48][ 1130/ 1207] Overall Loss 0.341685 Objective Loss 0.341685 LR 0.001000 Time 0.019511 -2023-02-13 17:40:04,888 - Epoch: [48][ 1140/ 1207] Overall Loss 0.341711 Objective Loss 0.341711 LR 0.001000 Time 0.019504 -2023-02-13 17:40:05,075 - Epoch: [48][ 1150/ 1207] Overall Loss 0.341799 Objective Loss 0.341799 LR 0.001000 Time 0.019497 -2023-02-13 17:40:05,262 - Epoch: [48][ 1160/ 1207] Overall Loss 0.341700 Objective Loss 0.341700 LR 0.001000 Time 0.019490 -2023-02-13 17:40:05,451 - Epoch: [48][ 1170/ 1207] Overall Loss 0.341773 Objective Loss 0.341773 LR 0.001000 Time 0.019484 -2023-02-13 17:40:05,638 - Epoch: [48][ 1180/ 1207] Overall Loss 0.341781 Objective Loss 0.341781 LR 0.001000 Time 0.019477 -2023-02-13 17:40:05,826 - Epoch: [48][ 1190/ 1207] Overall Loss 0.341633 Objective Loss 0.341633 LR 0.001000 Time 0.019471 -2023-02-13 17:40:06,069 - Epoch: [48][ 1200/ 1207] Overall Loss 0.341521 Objective Loss 0.341521 LR 0.001000 Time 0.019511 -2023-02-13 17:40:06,183 - Epoch: [48][ 1207/ 1207] Overall Loss 0.341231 Objective Loss 0.341231 Top1 84.451220 Top5 99.085366 LR 0.001000 Time 0.019493 -2023-02-13 17:40:06,255 - --- validate (epoch=48)----------- -2023-02-13 17:40:06,255 - 34311 samples (256 per mini-batch) -2023-02-13 17:40:06,658 - Epoch: [48][ 10/ 135] Loss 0.330297 Top1 82.812500 Top5 97.539062 -2023-02-13 17:40:06,798 - Epoch: [48][ 20/ 135] Loss 0.354015 Top1 82.597656 Top5 97.500000 -2023-02-13 17:40:06,926 - Epoch: [48][ 30/ 135] Loss 0.354944 Top1 82.395833 Top5 97.565104 -2023-02-13 17:40:07,056 - Epoch: [48][ 40/ 135] Loss 0.360669 Top1 82.402344 Top5 97.666016 -2023-02-13 17:40:07,182 - Epoch: [48][ 50/ 135] Loss 0.366798 Top1 82.140625 Top5 97.484375 -2023-02-13 17:40:07,312 - Epoch: [48][ 60/ 135] Loss 0.362086 Top1 82.252604 Top5 97.506510 -2023-02-13 17:40:07,439 - Epoch: [48][ 70/ 135] Loss 0.360930 Top1 82.405134 Top5 97.494420 -2023-02-13 17:40:07,568 - Epoch: [48][ 80/ 135] Loss 0.361863 Top1 82.285156 Top5 97.441406 -2023-02-13 17:40:07,694 - Epoch: [48][ 90/ 135] Loss 0.363271 Top1 82.248264 Top5 97.408854 -2023-02-13 17:40:07,824 - Epoch: [48][ 100/ 135] Loss 0.363473 Top1 82.179688 Top5 97.382812 -2023-02-13 17:40:07,951 - Epoch: [48][ 110/ 135] Loss 0.362952 Top1 82.254972 Top5 97.421875 -2023-02-13 17:40:08,079 - Epoch: [48][ 120/ 135] Loss 0.365028 Top1 82.291667 Top5 97.438151 -2023-02-13 17:40:08,210 - Epoch: [48][ 130/ 135] Loss 0.364693 Top1 82.220553 Top5 97.412861 -2023-02-13 17:40:08,254 - Epoch: [48][ 135/ 135] Loss 0.362778 Top1 82.186471 Top5 97.414823 -2023-02-13 17:40:08,325 - ==> Top1: 82.186 Top5: 97.415 Loss: 0.363 - -2023-02-13 17:40:08,326 - ==> Confusion: -[[ 805 5 3 1 12 3 0 2 8 90 1 2 1 5 8 1 5 2 2 0 11] - [ 2 927 0 3 7 34 5 17 6 2 3 4 2 0 4 1 1 0 8 0 7] - [ 9 3 946 18 7 2 16 10 0 0 3 2 3 3 3 7 2 4 8 2 10] - [ 7 2 25 884 1 4 1 4 1 2 12 0 5 3 32 0 2 4 16 1 10] - [ 16 15 1 0 965 15 1 2 2 6 1 11 0 2 8 6 5 1 3 0 6] - [ 6 25 2 8 4 948 2 18 0 4 3 18 4 12 5 0 1 1 3 2 4] - [ 0 9 23 5 1 4 1017 5 1 1 5 1 5 2 0 1 1 3 3 10 2] - [ 4 7 9 2 2 39 6 892 2 2 3 5 7 1 0 0 0 2 24 12 5] - [ 17 4 1 1 1 0 0 1 878 45 10 3 1 9 30 2 1 0 4 0 1] - [ 59 1 4 0 6 3 0 3 36 866 0 0 0 17 7 1 1 1 1 0 6] - [ 3 4 9 11 1 1 1 2 24 0 949 0 3 12 6 0 1 2 14 1 7] - [ 3 3 2 0 8 10 1 6 3 1 1 891 38 6 2 4 4 13 2 3 4] - [ 5 0 1 7 1 1 0 2 0 0 0 22 869 2 8 3 3 23 6 0 6] - [ 4 4 1 0 8 6 2 3 16 22 8 11 4 912 5 7 2 1 1 0 7] - [ 15 2 1 14 2 1 0 1 18 4 1 4 4 2 999 1 1 3 10 0 9] - [ 5 3 7 1 11 2 8 2 0 1 0 10 13 4 0 945 5 16 0 3 10] - [ 2 10 1 2 9 6 0 0 4 0 0 3 1 3 3 13 984 2 3 4 11] - [ 9 3 1 4 2 1 1 0 1 0 0 9 24 0 5 11 1 970 2 0 7] - [ 6 7 5 14 0 2 0 27 4 0 3 3 6 0 19 1 1 0 984 0 4] - [ 0 1 1 1 1 8 6 13 1 0 2 36 2 8 0 4 6 3 2 1040 13] - [ 176 261 230 165 128 214 78 176 126 114 219 150 359 354 266 76 255 125 213 221 9528]] - -2023-02-13 17:40:08,327 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:40:08,327 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:40:08,333 - - -2023-02-13 17:40:08,333 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:40:09,241 - Epoch: [49][ 10/ 1207] Overall Loss 0.326847 Objective Loss 0.326847 LR 0.001000 Time 0.090690 -2023-02-13 17:40:09,447 - Epoch: [49][ 20/ 1207] Overall Loss 0.334340 Objective Loss 0.334340 LR 0.001000 Time 0.055622 -2023-02-13 17:40:09,642 - Epoch: [49][ 30/ 1207] Overall Loss 0.312870 Objective Loss 0.312870 LR 0.001000 Time 0.043574 -2023-02-13 17:40:09,838 - Epoch: [49][ 40/ 1207] Overall Loss 0.312155 Objective Loss 0.312155 LR 0.001000 Time 0.037578 -2023-02-13 17:40:10,034 - Epoch: [49][ 50/ 1207] Overall Loss 0.320434 Objective Loss 0.320434 LR 0.001000 Time 0.033972 -2023-02-13 17:40:10,230 - Epoch: [49][ 60/ 1207] Overall Loss 0.321466 Objective Loss 0.321466 LR 0.001000 Time 0.031576 -2023-02-13 17:40:10,426 - Epoch: [49][ 70/ 1207] Overall Loss 0.323061 Objective Loss 0.323061 LR 0.001000 Time 0.029850 -2023-02-13 17:40:10,622 - Epoch: [49][ 80/ 1207] Overall Loss 0.325100 Objective Loss 0.325100 LR 0.001000 Time 0.028575 -2023-02-13 17:40:10,818 - Epoch: [49][ 90/ 1207] Overall Loss 0.324007 Objective Loss 0.324007 LR 0.001000 Time 0.027565 -2023-02-13 17:40:11,014 - Epoch: [49][ 100/ 1207] Overall Loss 0.322555 Objective Loss 0.322555 LR 0.001000 Time 0.026769 -2023-02-13 17:40:11,207 - Epoch: [49][ 110/ 1207] Overall Loss 0.324518 Objective Loss 0.324518 LR 0.001000 Time 0.026089 -2023-02-13 17:40:11,398 - Epoch: [49][ 120/ 1207] Overall Loss 0.326423 Objective Loss 0.326423 LR 0.001000 Time 0.025499 -2023-02-13 17:40:11,588 - Epoch: [49][ 130/ 1207] Overall Loss 0.325578 Objective Loss 0.325578 LR 0.001000 Time 0.024996 -2023-02-13 17:40:11,778 - Epoch: [49][ 140/ 1207] Overall Loss 0.324189 Objective Loss 0.324189 LR 0.001000 Time 0.024565 -2023-02-13 17:40:11,968 - Epoch: [49][ 150/ 1207] Overall Loss 0.323082 Objective Loss 0.323082 LR 0.001000 Time 0.024196 -2023-02-13 17:40:12,158 - Epoch: [49][ 160/ 1207] Overall Loss 0.323203 Objective Loss 0.323203 LR 0.001000 Time 0.023868 -2023-02-13 17:40:12,348 - Epoch: [49][ 170/ 1207] Overall Loss 0.324921 Objective Loss 0.324921 LR 0.001000 Time 0.023577 -2023-02-13 17:40:12,538 - Epoch: [49][ 180/ 1207] Overall Loss 0.325680 Objective Loss 0.325680 LR 0.001000 Time 0.023324 -2023-02-13 17:40:12,728 - Epoch: [49][ 190/ 1207] Overall Loss 0.324228 Objective Loss 0.324228 LR 0.001000 Time 0.023094 -2023-02-13 17:40:12,918 - Epoch: [49][ 200/ 1207] Overall Loss 0.324600 Objective Loss 0.324600 LR 0.001000 Time 0.022887 -2023-02-13 17:40:13,108 - Epoch: [49][ 210/ 1207] Overall Loss 0.324666 Objective Loss 0.324666 LR 0.001000 Time 0.022697 -2023-02-13 17:40:13,297 - Epoch: [49][ 220/ 1207] Overall Loss 0.324255 Objective Loss 0.324255 LR 0.001000 Time 0.022524 -2023-02-13 17:40:13,488 - Epoch: [49][ 230/ 1207] Overall Loss 0.324436 Objective Loss 0.324436 LR 0.001000 Time 0.022373 -2023-02-13 17:40:13,678 - Epoch: [49][ 240/ 1207] Overall Loss 0.324267 Objective Loss 0.324267 LR 0.001000 Time 0.022232 -2023-02-13 17:40:13,868 - Epoch: [49][ 250/ 1207] Overall Loss 0.323951 Objective Loss 0.323951 LR 0.001000 Time 0.022100 -2023-02-13 17:40:14,057 - Epoch: [49][ 260/ 1207] Overall Loss 0.323767 Objective Loss 0.323767 LR 0.001000 Time 0.021978 -2023-02-13 17:40:14,247 - Epoch: [49][ 270/ 1207] Overall Loss 0.323487 Objective Loss 0.323487 LR 0.001000 Time 0.021865 -2023-02-13 17:40:14,437 - Epoch: [49][ 280/ 1207] Overall Loss 0.324523 Objective Loss 0.324523 LR 0.001000 Time 0.021762 -2023-02-13 17:40:14,627 - Epoch: [49][ 290/ 1207] Overall Loss 0.325082 Objective Loss 0.325082 LR 0.001000 Time 0.021664 -2023-02-13 17:40:14,817 - Epoch: [49][ 300/ 1207] Overall Loss 0.325033 Objective Loss 0.325033 LR 0.001000 Time 0.021574 -2023-02-13 17:40:15,006 - Epoch: [49][ 310/ 1207] Overall Loss 0.324693 Objective Loss 0.324693 LR 0.001000 Time 0.021488 -2023-02-13 17:40:15,196 - Epoch: [49][ 320/ 1207] Overall Loss 0.325821 Objective Loss 0.325821 LR 0.001000 Time 0.021407 -2023-02-13 17:40:15,385 - Epoch: [49][ 330/ 1207] Overall Loss 0.327040 Objective Loss 0.327040 LR 0.001000 Time 0.021331 -2023-02-13 17:40:15,575 - Epoch: [49][ 340/ 1207] Overall Loss 0.327927 Objective Loss 0.327927 LR 0.001000 Time 0.021262 -2023-02-13 17:40:15,765 - Epoch: [49][ 350/ 1207] Overall Loss 0.328659 Objective Loss 0.328659 LR 0.001000 Time 0.021195 -2023-02-13 17:40:15,955 - Epoch: [49][ 360/ 1207] Overall Loss 0.329376 Objective Loss 0.329376 LR 0.001000 Time 0.021135 -2023-02-13 17:40:16,144 - Epoch: [49][ 370/ 1207] Overall Loss 0.330301 Objective Loss 0.330301 LR 0.001000 Time 0.021074 -2023-02-13 17:40:16,334 - Epoch: [49][ 380/ 1207] Overall Loss 0.330342 Objective Loss 0.330342 LR 0.001000 Time 0.021017 -2023-02-13 17:40:16,525 - Epoch: [49][ 390/ 1207] Overall Loss 0.330891 Objective Loss 0.330891 LR 0.001000 Time 0.020966 -2023-02-13 17:40:16,715 - Epoch: [49][ 400/ 1207] Overall Loss 0.331044 Objective Loss 0.331044 LR 0.001000 Time 0.020916 -2023-02-13 17:40:16,905 - Epoch: [49][ 410/ 1207] Overall Loss 0.330819 Objective Loss 0.330819 LR 0.001000 Time 0.020869 -2023-02-13 17:40:17,095 - Epoch: [49][ 420/ 1207] Overall Loss 0.331165 Objective Loss 0.331165 LR 0.001000 Time 0.020823 -2023-02-13 17:40:17,284 - Epoch: [49][ 430/ 1207] Overall Loss 0.330890 Objective Loss 0.330890 LR 0.001000 Time 0.020779 -2023-02-13 17:40:17,475 - Epoch: [49][ 440/ 1207] Overall Loss 0.331634 Objective Loss 0.331634 LR 0.001000 Time 0.020738 -2023-02-13 17:40:17,665 - Epoch: [49][ 450/ 1207] Overall Loss 0.331165 Objective Loss 0.331165 LR 0.001000 Time 0.020700 -2023-02-13 17:40:17,855 - Epoch: [49][ 460/ 1207] Overall Loss 0.330677 Objective Loss 0.330677 LR 0.001000 Time 0.020662 -2023-02-13 17:40:18,044 - Epoch: [49][ 470/ 1207] Overall Loss 0.330817 Objective Loss 0.330817 LR 0.001000 Time 0.020625 -2023-02-13 17:40:18,235 - Epoch: [49][ 480/ 1207] Overall Loss 0.331330 Objective Loss 0.331330 LR 0.001000 Time 0.020590 -2023-02-13 17:40:18,425 - Epoch: [49][ 490/ 1207] Overall Loss 0.331954 Objective Loss 0.331954 LR 0.001000 Time 0.020558 -2023-02-13 17:40:18,615 - Epoch: [49][ 500/ 1207] Overall Loss 0.332582 Objective Loss 0.332582 LR 0.001000 Time 0.020527 -2023-02-13 17:40:18,805 - Epoch: [49][ 510/ 1207] Overall Loss 0.333336 Objective Loss 0.333336 LR 0.001000 Time 0.020495 -2023-02-13 17:40:18,995 - Epoch: [49][ 520/ 1207] Overall Loss 0.334244 Objective Loss 0.334244 LR 0.001000 Time 0.020466 -2023-02-13 17:40:19,185 - Epoch: [49][ 530/ 1207] Overall Loss 0.334755 Objective Loss 0.334755 LR 0.001000 Time 0.020437 -2023-02-13 17:40:19,375 - Epoch: [49][ 540/ 1207] Overall Loss 0.334750 Objective Loss 0.334750 LR 0.001000 Time 0.020410 -2023-02-13 17:40:19,565 - Epoch: [49][ 550/ 1207] Overall Loss 0.335852 Objective Loss 0.335852 LR 0.001000 Time 0.020384 -2023-02-13 17:40:19,755 - Epoch: [49][ 560/ 1207] Overall Loss 0.335559 Objective Loss 0.335559 LR 0.001000 Time 0.020358 -2023-02-13 17:40:19,945 - Epoch: [49][ 570/ 1207] Overall Loss 0.335297 Objective Loss 0.335297 LR 0.001000 Time 0.020333 -2023-02-13 17:40:20,134 - Epoch: [49][ 580/ 1207] Overall Loss 0.335742 Objective Loss 0.335742 LR 0.001000 Time 0.020309 -2023-02-13 17:40:20,325 - Epoch: [49][ 590/ 1207] Overall Loss 0.335493 Objective Loss 0.335493 LR 0.001000 Time 0.020287 -2023-02-13 17:40:20,515 - Epoch: [49][ 600/ 1207] Overall Loss 0.335640 Objective Loss 0.335640 LR 0.001000 Time 0.020266 -2023-02-13 17:40:20,705 - Epoch: [49][ 610/ 1207] Overall Loss 0.335739 Objective Loss 0.335739 LR 0.001000 Time 0.020244 -2023-02-13 17:40:20,897 - Epoch: [49][ 620/ 1207] Overall Loss 0.335514 Objective Loss 0.335514 LR 0.001000 Time 0.020226 -2023-02-13 17:40:21,086 - Epoch: [49][ 630/ 1207] Overall Loss 0.336073 Objective Loss 0.336073 LR 0.001000 Time 0.020205 -2023-02-13 17:40:21,275 - Epoch: [49][ 640/ 1207] Overall Loss 0.336180 Objective Loss 0.336180 LR 0.001000 Time 0.020184 -2023-02-13 17:40:21,465 - Epoch: [49][ 650/ 1207] Overall Loss 0.336538 Objective Loss 0.336538 LR 0.001000 Time 0.020166 -2023-02-13 17:40:21,656 - Epoch: [49][ 660/ 1207] Overall Loss 0.336613 Objective Loss 0.336613 LR 0.001000 Time 0.020148 -2023-02-13 17:40:21,846 - Epoch: [49][ 670/ 1207] Overall Loss 0.337086 Objective Loss 0.337086 LR 0.001000 Time 0.020131 -2023-02-13 17:40:22,036 - Epoch: [49][ 680/ 1207] Overall Loss 0.336866 Objective Loss 0.336866 LR 0.001000 Time 0.020113 -2023-02-13 17:40:22,225 - Epoch: [49][ 690/ 1207] Overall Loss 0.336582 Objective Loss 0.336582 LR 0.001000 Time 0.020096 -2023-02-13 17:40:22,416 - Epoch: [49][ 700/ 1207] Overall Loss 0.337034 Objective Loss 0.337034 LR 0.001000 Time 0.020081 -2023-02-13 17:40:22,606 - Epoch: [49][ 710/ 1207] Overall Loss 0.337587 Objective Loss 0.337587 LR 0.001000 Time 0.020065 -2023-02-13 17:40:22,795 - Epoch: [49][ 720/ 1207] Overall Loss 0.337754 Objective Loss 0.337754 LR 0.001000 Time 0.020049 -2023-02-13 17:40:22,985 - Epoch: [49][ 730/ 1207] Overall Loss 0.337991 Objective Loss 0.337991 LR 0.001000 Time 0.020034 -2023-02-13 17:40:23,175 - Epoch: [49][ 740/ 1207] Overall Loss 0.338246 Objective Loss 0.338246 LR 0.001000 Time 0.020019 -2023-02-13 17:40:23,365 - Epoch: [49][ 750/ 1207] Overall Loss 0.338596 Objective Loss 0.338596 LR 0.001000 Time 0.020005 -2023-02-13 17:40:23,556 - Epoch: [49][ 760/ 1207] Overall Loss 0.338618 Objective Loss 0.338618 LR 0.001000 Time 0.019992 -2023-02-13 17:40:23,746 - Epoch: [49][ 770/ 1207] Overall Loss 0.338582 Objective Loss 0.338582 LR 0.001000 Time 0.019979 -2023-02-13 17:40:23,936 - Epoch: [49][ 780/ 1207] Overall Loss 0.338420 Objective Loss 0.338420 LR 0.001000 Time 0.019966 -2023-02-13 17:40:24,125 - Epoch: [49][ 790/ 1207] Overall Loss 0.338186 Objective Loss 0.338186 LR 0.001000 Time 0.019952 -2023-02-13 17:40:24,315 - Epoch: [49][ 800/ 1207] Overall Loss 0.338251 Objective Loss 0.338251 LR 0.001000 Time 0.019940 -2023-02-13 17:40:24,506 - Epoch: [49][ 810/ 1207] Overall Loss 0.338561 Objective Loss 0.338561 LR 0.001000 Time 0.019929 -2023-02-13 17:40:24,696 - Epoch: [49][ 820/ 1207] Overall Loss 0.338928 Objective Loss 0.338928 LR 0.001000 Time 0.019917 -2023-02-13 17:40:24,886 - Epoch: [49][ 830/ 1207] Overall Loss 0.338485 Objective Loss 0.338485 LR 0.001000 Time 0.019905 -2023-02-13 17:40:25,075 - Epoch: [49][ 840/ 1207] Overall Loss 0.338446 Objective Loss 0.338446 LR 0.001000 Time 0.019894 -2023-02-13 17:40:25,265 - Epoch: [49][ 850/ 1207] Overall Loss 0.338221 Objective Loss 0.338221 LR 0.001000 Time 0.019882 -2023-02-13 17:40:25,456 - Epoch: [49][ 860/ 1207] Overall Loss 0.338082 Objective Loss 0.338082 LR 0.001000 Time 0.019872 -2023-02-13 17:40:25,646 - Epoch: [49][ 870/ 1207] Overall Loss 0.338263 Objective Loss 0.338263 LR 0.001000 Time 0.019862 -2023-02-13 17:40:25,839 - Epoch: [49][ 880/ 1207] Overall Loss 0.338126 Objective Loss 0.338126 LR 0.001000 Time 0.019855 -2023-02-13 17:40:26,029 - Epoch: [49][ 890/ 1207] Overall Loss 0.337919 Objective Loss 0.337919 LR 0.001000 Time 0.019846 -2023-02-13 17:40:26,220 - Epoch: [49][ 900/ 1207] Overall Loss 0.338159 Objective Loss 0.338159 LR 0.001000 Time 0.019836 -2023-02-13 17:40:26,410 - Epoch: [49][ 910/ 1207] Overall Loss 0.338066 Objective Loss 0.338066 LR 0.001000 Time 0.019827 -2023-02-13 17:40:26,602 - Epoch: [49][ 920/ 1207] Overall Loss 0.338152 Objective Loss 0.338152 LR 0.001000 Time 0.019819 -2023-02-13 17:40:26,793 - Epoch: [49][ 930/ 1207] Overall Loss 0.338171 Objective Loss 0.338171 LR 0.001000 Time 0.019811 -2023-02-13 17:40:26,984 - Epoch: [49][ 940/ 1207] Overall Loss 0.338424 Objective Loss 0.338424 LR 0.001000 Time 0.019803 -2023-02-13 17:40:27,174 - Epoch: [49][ 950/ 1207] Overall Loss 0.338284 Objective Loss 0.338284 LR 0.001000 Time 0.019795 -2023-02-13 17:40:27,364 - Epoch: [49][ 960/ 1207] Overall Loss 0.338077 Objective Loss 0.338077 LR 0.001000 Time 0.019786 -2023-02-13 17:40:27,556 - Epoch: [49][ 970/ 1207] Overall Loss 0.338045 Objective Loss 0.338045 LR 0.001000 Time 0.019779 -2023-02-13 17:40:27,746 - Epoch: [49][ 980/ 1207] Overall Loss 0.337798 Objective Loss 0.337798 LR 0.001000 Time 0.019772 -2023-02-13 17:40:27,936 - Epoch: [49][ 990/ 1207] Overall Loss 0.337792 Objective Loss 0.337792 LR 0.001000 Time 0.019764 -2023-02-13 17:40:28,127 - Epoch: [49][ 1000/ 1207] Overall Loss 0.337959 Objective Loss 0.337959 LR 0.001000 Time 0.019756 -2023-02-13 17:40:28,317 - Epoch: [49][ 1010/ 1207] Overall Loss 0.338019 Objective Loss 0.338019 LR 0.001000 Time 0.019749 -2023-02-13 17:40:28,508 - Epoch: [49][ 1020/ 1207] Overall Loss 0.338215 Objective Loss 0.338215 LR 0.001000 Time 0.019742 -2023-02-13 17:40:28,698 - Epoch: [49][ 1030/ 1207] Overall Loss 0.338141 Objective Loss 0.338141 LR 0.001000 Time 0.019734 -2023-02-13 17:40:28,888 - Epoch: [49][ 1040/ 1207] Overall Loss 0.337983 Objective Loss 0.337983 LR 0.001000 Time 0.019727 -2023-02-13 17:40:29,078 - Epoch: [49][ 1050/ 1207] Overall Loss 0.338124 Objective Loss 0.338124 LR 0.001000 Time 0.019719 -2023-02-13 17:40:29,269 - Epoch: [49][ 1060/ 1207] Overall Loss 0.338028 Objective Loss 0.338028 LR 0.001000 Time 0.019713 -2023-02-13 17:40:29,459 - Epoch: [49][ 1070/ 1207] Overall Loss 0.338228 Objective Loss 0.338228 LR 0.001000 Time 0.019706 -2023-02-13 17:40:29,649 - Epoch: [49][ 1080/ 1207] Overall Loss 0.338209 Objective Loss 0.338209 LR 0.001000 Time 0.019700 -2023-02-13 17:40:29,839 - Epoch: [49][ 1090/ 1207] Overall Loss 0.338056 Objective Loss 0.338056 LR 0.001000 Time 0.019693 -2023-02-13 17:40:30,029 - Epoch: [49][ 1100/ 1207] Overall Loss 0.338117 Objective Loss 0.338117 LR 0.001000 Time 0.019686 -2023-02-13 17:40:30,219 - Epoch: [49][ 1110/ 1207] Overall Loss 0.337905 Objective Loss 0.337905 LR 0.001000 Time 0.019679 -2023-02-13 17:40:30,409 - Epoch: [49][ 1120/ 1207] Overall Loss 0.338052 Objective Loss 0.338052 LR 0.001000 Time 0.019673 -2023-02-13 17:40:30,599 - Epoch: [49][ 1130/ 1207] Overall Loss 0.337860 Objective Loss 0.337860 LR 0.001000 Time 0.019667 -2023-02-13 17:40:30,790 - Epoch: [49][ 1140/ 1207] Overall Loss 0.337591 Objective Loss 0.337591 LR 0.001000 Time 0.019661 -2023-02-13 17:40:30,980 - Epoch: [49][ 1150/ 1207] Overall Loss 0.337706 Objective Loss 0.337706 LR 0.001000 Time 0.019655 -2023-02-13 17:40:31,170 - Epoch: [49][ 1160/ 1207] Overall Loss 0.338089 Objective Loss 0.338089 LR 0.001000 Time 0.019649 -2023-02-13 17:40:31,360 - Epoch: [49][ 1170/ 1207] Overall Loss 0.338005 Objective Loss 0.338005 LR 0.001000 Time 0.019643 -2023-02-13 17:40:31,551 - Epoch: [49][ 1180/ 1207] Overall Loss 0.338596 Objective Loss 0.338596 LR 0.001000 Time 0.019639 -2023-02-13 17:40:31,742 - Epoch: [49][ 1190/ 1207] Overall Loss 0.338642 Objective Loss 0.338642 LR 0.001000 Time 0.019634 -2023-02-13 17:40:31,983 - Epoch: [49][ 1200/ 1207] Overall Loss 0.338449 Objective Loss 0.338449 LR 0.001000 Time 0.019671 -2023-02-13 17:40:32,098 - Epoch: [49][ 1207/ 1207] Overall Loss 0.338503 Objective Loss 0.338503 Top1 83.841463 Top5 96.951220 LR 0.001000 Time 0.019652 -2023-02-13 17:40:32,169 - --- validate (epoch=49)----------- -2023-02-13 17:40:32,170 - 34311 samples (256 per mini-batch) -2023-02-13 17:40:32,572 - Epoch: [49][ 10/ 135] Loss 0.382777 Top1 80.273438 Top5 97.695312 -2023-02-13 17:40:32,697 - Epoch: [49][ 20/ 135] Loss 0.370395 Top1 80.234375 Top5 97.382812 -2023-02-13 17:40:32,830 - Epoch: [49][ 30/ 135] Loss 0.367390 Top1 80.911458 Top5 97.174479 -2023-02-13 17:40:32,961 - Epoch: [49][ 40/ 135] Loss 0.370446 Top1 80.625000 Top5 97.050781 -2023-02-13 17:40:33,091 - Epoch: [49][ 50/ 135] Loss 0.370527 Top1 80.742188 Top5 97.085938 -2023-02-13 17:40:33,224 - Epoch: [49][ 60/ 135] Loss 0.369200 Top1 80.579427 Top5 97.070312 -2023-02-13 17:40:33,357 - Epoch: [49][ 70/ 135] Loss 0.367288 Top1 80.859375 Top5 97.137277 -2023-02-13 17:40:33,488 - Epoch: [49][ 80/ 135] Loss 0.361750 Top1 81.015625 Top5 97.124023 -2023-02-13 17:40:33,619 - Epoch: [49][ 90/ 135] Loss 0.358911 Top1 81.128472 Top5 97.109375 -2023-02-13 17:40:33,749 - Epoch: [49][ 100/ 135] Loss 0.359803 Top1 81.019531 Top5 97.113281 -2023-02-13 17:40:33,880 - Epoch: [49][ 110/ 135] Loss 0.359702 Top1 81.093750 Top5 97.137784 -2023-02-13 17:40:34,010 - Epoch: [49][ 120/ 135] Loss 0.361780 Top1 81.113281 Top5 97.128906 -2023-02-13 17:40:34,144 - Epoch: [49][ 130/ 135] Loss 0.362472 Top1 81.117788 Top5 97.124399 -2023-02-13 17:40:34,189 - Epoch: [49][ 135/ 135] Loss 0.373684 Top1 81.148903 Top5 97.114628 -2023-02-13 17:40:34,262 - ==> Top1: 81.149 Top5: 97.115 Loss: 0.374 - -2023-02-13 17:40:34,263 - ==> Confusion: -[[ 848 6 4 2 12 4 0 1 7 56 0 4 1 2 3 4 6 2 1 2 2] - [ 2 929 2 4 10 27 3 10 4 2 3 2 1 0 2 3 7 2 6 2 12] - [ 11 5 948 10 3 3 11 20 0 0 2 3 0 2 4 12 3 3 8 4 6] - [ 5 1 28 884 2 5 0 3 2 2 15 0 5 2 20 3 4 7 20 0 8] - [ 21 8 1 0 978 10 1 4 1 4 0 5 0 3 7 7 9 0 0 5 2] - [ 2 25 1 4 3 965 2 16 0 2 2 16 3 7 2 2 5 2 0 5 6] - [ 1 3 25 4 1 8 1017 9 0 2 5 1 2 1 1 2 2 2 1 10 2] - [ 2 13 14 1 2 38 2 885 2 1 6 4 4 1 1 0 1 2 29 11 5] - [ 16 4 0 1 1 1 1 0 896 38 16 2 1 4 16 3 2 0 7 0 0] - [ 82 3 3 1 4 2 0 3 44 840 1 0 1 11 5 3 1 1 1 1 5] - [ 6 2 10 7 1 0 4 4 18 2 968 1 0 9 2 0 3 0 10 1 3] - [ 3 5 2 1 1 11 2 7 5 1 0 882 30 12 0 13 9 10 0 9 2] - [ 0 0 0 10 1 1 0 1 3 0 2 44 851 1 4 9 5 16 1 2 8] - [ 6 5 3 1 7 17 2 0 28 25 14 7 4 874 8 8 5 0 2 3 5] - [ 17 4 4 17 4 3 0 0 20 3 5 1 4 0 986 2 3 4 10 0 5] - [ 7 3 5 1 5 2 4 1 1 0 0 4 4 6 3 960 14 12 0 8 6] - [ 5 5 0 2 9 3 0 0 2 1 1 2 1 0 2 18 998 1 4 4 3] - [ 8 3 2 5 1 0 0 2 1 1 1 10 27 2 1 17 2 963 2 1 2] - [ 2 6 6 17 1 2 0 22 6 1 7 4 2 0 11 1 1 3 989 2 3] - [ 1 4 4 1 1 7 6 18 0 2 2 23 4 5 0 4 11 2 1 1045 7] - [ 225 306 269 171 152 260 81 178 132 119 234 129 320 296 196 155 480 128 171 295 9137]] - -2023-02-13 17:40:34,264 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:40:34,264 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:40:34,270 - - -2023-02-13 17:40:34,270 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:40:35,164 - Epoch: [50][ 10/ 1207] Overall Loss 0.358324 Objective Loss 0.358324 LR 0.001000 Time 0.089282 -2023-02-13 17:40:35,361 - Epoch: [50][ 20/ 1207] Overall Loss 0.361917 Objective Loss 0.361917 LR 0.001000 Time 0.054475 -2023-02-13 17:40:35,551 - Epoch: [50][ 30/ 1207] Overall Loss 0.354199 Objective Loss 0.354199 LR 0.001000 Time 0.042646 -2023-02-13 17:40:35,740 - Epoch: [50][ 40/ 1207] Overall Loss 0.337604 Objective Loss 0.337604 LR 0.001000 Time 0.036704 -2023-02-13 17:40:35,930 - Epoch: [50][ 50/ 1207] Overall Loss 0.342955 Objective Loss 0.342955 LR 0.001000 Time 0.033145 -2023-02-13 17:40:36,119 - Epoch: [50][ 60/ 1207] Overall Loss 0.339537 Objective Loss 0.339537 LR 0.001000 Time 0.030759 -2023-02-13 17:40:36,307 - Epoch: [50][ 70/ 1207] Overall Loss 0.342198 Objective Loss 0.342198 LR 0.001000 Time 0.029052 -2023-02-13 17:40:36,496 - Epoch: [50][ 80/ 1207] Overall Loss 0.337325 Objective Loss 0.337325 LR 0.001000 Time 0.027776 -2023-02-13 17:40:36,685 - Epoch: [50][ 90/ 1207] Overall Loss 0.336964 Objective Loss 0.336964 LR 0.001000 Time 0.026787 -2023-02-13 17:40:36,874 - Epoch: [50][ 100/ 1207] Overall Loss 0.338175 Objective Loss 0.338175 LR 0.001000 Time 0.025994 -2023-02-13 17:40:37,063 - Epoch: [50][ 110/ 1207] Overall Loss 0.338328 Objective Loss 0.338328 LR 0.001000 Time 0.025346 -2023-02-13 17:40:37,252 - Epoch: [50][ 120/ 1207] Overall Loss 0.335853 Objective Loss 0.335853 LR 0.001000 Time 0.024810 -2023-02-13 17:40:37,443 - Epoch: [50][ 130/ 1207] Overall Loss 0.332169 Objective Loss 0.332169 LR 0.001000 Time 0.024368 -2023-02-13 17:40:37,635 - Epoch: [50][ 140/ 1207] Overall Loss 0.333371 Objective Loss 0.333371 LR 0.001000 Time 0.023995 -2023-02-13 17:40:37,826 - Epoch: [50][ 150/ 1207] Overall Loss 0.333535 Objective Loss 0.333535 LR 0.001000 Time 0.023664 -2023-02-13 17:40:38,016 - Epoch: [50][ 160/ 1207] Overall Loss 0.333239 Objective Loss 0.333239 LR 0.001000 Time 0.023374 -2023-02-13 17:40:38,207 - Epoch: [50][ 170/ 1207] Overall Loss 0.333835 Objective Loss 0.333835 LR 0.001000 Time 0.023116 -2023-02-13 17:40:38,398 - Epoch: [50][ 180/ 1207] Overall Loss 0.333381 Objective Loss 0.333381 LR 0.001000 Time 0.022891 -2023-02-13 17:40:38,589 - Epoch: [50][ 190/ 1207] Overall Loss 0.332908 Objective Loss 0.332908 LR 0.001000 Time 0.022693 -2023-02-13 17:40:38,781 - Epoch: [50][ 200/ 1207] Overall Loss 0.333885 Objective Loss 0.333885 LR 0.001000 Time 0.022514 -2023-02-13 17:40:38,972 - Epoch: [50][ 210/ 1207] Overall Loss 0.333474 Objective Loss 0.333474 LR 0.001000 Time 0.022348 -2023-02-13 17:40:39,162 - Epoch: [50][ 220/ 1207] Overall Loss 0.332130 Objective Loss 0.332130 LR 0.001000 Time 0.022195 -2023-02-13 17:40:39,353 - Epoch: [50][ 230/ 1207] Overall Loss 0.331372 Objective Loss 0.331372 LR 0.001000 Time 0.022059 -2023-02-13 17:40:39,545 - Epoch: [50][ 240/ 1207] Overall Loss 0.330869 Objective Loss 0.330869 LR 0.001000 Time 0.021937 -2023-02-13 17:40:39,735 - Epoch: [50][ 250/ 1207] Overall Loss 0.331553 Objective Loss 0.331553 LR 0.001000 Time 0.021821 -2023-02-13 17:40:39,926 - Epoch: [50][ 260/ 1207] Overall Loss 0.332851 Objective Loss 0.332851 LR 0.001000 Time 0.021713 -2023-02-13 17:40:40,116 - Epoch: [50][ 270/ 1207] Overall Loss 0.333396 Objective Loss 0.333396 LR 0.001000 Time 0.021613 -2023-02-13 17:40:40,307 - Epoch: [50][ 280/ 1207] Overall Loss 0.333501 Objective Loss 0.333501 LR 0.001000 Time 0.021519 -2023-02-13 17:40:40,497 - Epoch: [50][ 290/ 1207] Overall Loss 0.334895 Objective Loss 0.334895 LR 0.001000 Time 0.021434 -2023-02-13 17:40:40,688 - Epoch: [50][ 300/ 1207] Overall Loss 0.335222 Objective Loss 0.335222 LR 0.001000 Time 0.021355 -2023-02-13 17:40:40,881 - Epoch: [50][ 310/ 1207] Overall Loss 0.336629 Objective Loss 0.336629 LR 0.001000 Time 0.021286 -2023-02-13 17:40:41,072 - Epoch: [50][ 320/ 1207] Overall Loss 0.336515 Objective Loss 0.336515 LR 0.001000 Time 0.021216 -2023-02-13 17:40:41,262 - Epoch: [50][ 330/ 1207] Overall Loss 0.337464 Objective Loss 0.337464 LR 0.001000 Time 0.021149 -2023-02-13 17:40:41,453 - Epoch: [50][ 340/ 1207] Overall Loss 0.337741 Objective Loss 0.337741 LR 0.001000 Time 0.021088 -2023-02-13 17:40:41,645 - Epoch: [50][ 350/ 1207] Overall Loss 0.338231 Objective Loss 0.338231 LR 0.001000 Time 0.021031 -2023-02-13 17:40:41,837 - Epoch: [50][ 360/ 1207] Overall Loss 0.338397 Objective Loss 0.338397 LR 0.001000 Time 0.020981 -2023-02-13 17:40:42,028 - Epoch: [50][ 370/ 1207] Overall Loss 0.339242 Objective Loss 0.339242 LR 0.001000 Time 0.020929 -2023-02-13 17:40:42,219 - Epoch: [50][ 380/ 1207] Overall Loss 0.339659 Objective Loss 0.339659 LR 0.001000 Time 0.020879 -2023-02-13 17:40:42,410 - Epoch: [50][ 390/ 1207] Overall Loss 0.340085 Objective Loss 0.340085 LR 0.001000 Time 0.020832 -2023-02-13 17:40:42,601 - Epoch: [50][ 400/ 1207] Overall Loss 0.339945 Objective Loss 0.339945 LR 0.001000 Time 0.020788 -2023-02-13 17:40:42,792 - Epoch: [50][ 410/ 1207] Overall Loss 0.339500 Objective Loss 0.339500 LR 0.001000 Time 0.020747 -2023-02-13 17:40:42,983 - Epoch: [50][ 420/ 1207] Overall Loss 0.338759 Objective Loss 0.338759 LR 0.001000 Time 0.020706 -2023-02-13 17:40:43,174 - Epoch: [50][ 430/ 1207] Overall Loss 0.338795 Objective Loss 0.338795 LR 0.001000 Time 0.020667 -2023-02-13 17:40:43,364 - Epoch: [50][ 440/ 1207] Overall Loss 0.338521 Objective Loss 0.338521 LR 0.001000 Time 0.020630 -2023-02-13 17:40:43,556 - Epoch: [50][ 450/ 1207] Overall Loss 0.338668 Objective Loss 0.338668 LR 0.001000 Time 0.020597 -2023-02-13 17:40:43,747 - Epoch: [50][ 460/ 1207] Overall Loss 0.338909 Objective Loss 0.338909 LR 0.001000 Time 0.020564 -2023-02-13 17:40:43,938 - Epoch: [50][ 470/ 1207] Overall Loss 0.338743 Objective Loss 0.338743 LR 0.001000 Time 0.020532 -2023-02-13 17:40:44,130 - Epoch: [50][ 480/ 1207] Overall Loss 0.339120 Objective Loss 0.339120 LR 0.001000 Time 0.020502 -2023-02-13 17:40:44,321 - Epoch: [50][ 490/ 1207] Overall Loss 0.339852 Objective Loss 0.339852 LR 0.001000 Time 0.020473 -2023-02-13 17:40:44,512 - Epoch: [50][ 500/ 1207] Overall Loss 0.340026 Objective Loss 0.340026 LR 0.001000 Time 0.020446 -2023-02-13 17:40:44,703 - Epoch: [50][ 510/ 1207] Overall Loss 0.339474 Objective Loss 0.339474 LR 0.001000 Time 0.020418 -2023-02-13 17:40:44,894 - Epoch: [50][ 520/ 1207] Overall Loss 0.339379 Objective Loss 0.339379 LR 0.001000 Time 0.020392 -2023-02-13 17:40:45,085 - Epoch: [50][ 530/ 1207] Overall Loss 0.339686 Objective Loss 0.339686 LR 0.001000 Time 0.020367 -2023-02-13 17:40:45,275 - Epoch: [50][ 540/ 1207] Overall Loss 0.340077 Objective Loss 0.340077 LR 0.001000 Time 0.020342 -2023-02-13 17:40:45,467 - Epoch: [50][ 550/ 1207] Overall Loss 0.339896 Objective Loss 0.339896 LR 0.001000 Time 0.020319 -2023-02-13 17:40:45,658 - Epoch: [50][ 560/ 1207] Overall Loss 0.339424 Objective Loss 0.339424 LR 0.001000 Time 0.020297 -2023-02-13 17:40:45,850 - Epoch: [50][ 570/ 1207] Overall Loss 0.339765 Objective Loss 0.339765 LR 0.001000 Time 0.020278 -2023-02-13 17:40:46,042 - Epoch: [50][ 580/ 1207] Overall Loss 0.340190 Objective Loss 0.340190 LR 0.001000 Time 0.020257 -2023-02-13 17:40:46,233 - Epoch: [50][ 590/ 1207] Overall Loss 0.340026 Objective Loss 0.340026 LR 0.001000 Time 0.020237 -2023-02-13 17:40:46,424 - Epoch: [50][ 600/ 1207] Overall Loss 0.340108 Objective Loss 0.340108 LR 0.001000 Time 0.020219 -2023-02-13 17:40:46,617 - Epoch: [50][ 610/ 1207] Overall Loss 0.339667 Objective Loss 0.339667 LR 0.001000 Time 0.020202 -2023-02-13 17:40:46,809 - Epoch: [50][ 620/ 1207] Overall Loss 0.339713 Objective Loss 0.339713 LR 0.001000 Time 0.020185 -2023-02-13 17:40:47,000 - Epoch: [50][ 630/ 1207] Overall Loss 0.340017 Objective Loss 0.340017 LR 0.001000 Time 0.020169 -2023-02-13 17:40:47,191 - Epoch: [50][ 640/ 1207] Overall Loss 0.340010 Objective Loss 0.340010 LR 0.001000 Time 0.020151 -2023-02-13 17:40:47,383 - Epoch: [50][ 650/ 1207] Overall Loss 0.340156 Objective Loss 0.340156 LR 0.001000 Time 0.020136 -2023-02-13 17:40:47,576 - Epoch: [50][ 660/ 1207] Overall Loss 0.340535 Objective Loss 0.340535 LR 0.001000 Time 0.020122 -2023-02-13 17:40:47,767 - Epoch: [50][ 670/ 1207] Overall Loss 0.340774 Objective Loss 0.340774 LR 0.001000 Time 0.020106 -2023-02-13 17:40:47,958 - Epoch: [50][ 680/ 1207] Overall Loss 0.340517 Objective Loss 0.340517 LR 0.001000 Time 0.020092 -2023-02-13 17:40:48,149 - Epoch: [50][ 690/ 1207] Overall Loss 0.340648 Objective Loss 0.340648 LR 0.001000 Time 0.020077 -2023-02-13 17:40:48,338 - Epoch: [50][ 700/ 1207] Overall Loss 0.340669 Objective Loss 0.340669 LR 0.001000 Time 0.020059 -2023-02-13 17:40:48,528 - Epoch: [50][ 710/ 1207] Overall Loss 0.340767 Objective Loss 0.340767 LR 0.001000 Time 0.020044 -2023-02-13 17:40:48,718 - Epoch: [50][ 720/ 1207] Overall Loss 0.340590 Objective Loss 0.340590 LR 0.001000 Time 0.020028 -2023-02-13 17:40:48,907 - Epoch: [50][ 730/ 1207] Overall Loss 0.340464 Objective Loss 0.340464 LR 0.001000 Time 0.020012 -2023-02-13 17:40:49,096 - Epoch: [50][ 740/ 1207] Overall Loss 0.340696 Objective Loss 0.340696 LR 0.001000 Time 0.019996 -2023-02-13 17:40:49,285 - Epoch: [50][ 750/ 1207] Overall Loss 0.340762 Objective Loss 0.340762 LR 0.001000 Time 0.019982 -2023-02-13 17:40:49,475 - Epoch: [50][ 760/ 1207] Overall Loss 0.341099 Objective Loss 0.341099 LR 0.001000 Time 0.019968 -2023-02-13 17:40:49,665 - Epoch: [50][ 770/ 1207] Overall Loss 0.340958 Objective Loss 0.340958 LR 0.001000 Time 0.019955 -2023-02-13 17:40:49,855 - Epoch: [50][ 780/ 1207] Overall Loss 0.340629 Objective Loss 0.340629 LR 0.001000 Time 0.019942 -2023-02-13 17:40:50,044 - Epoch: [50][ 790/ 1207] Overall Loss 0.340527 Objective Loss 0.340527 LR 0.001000 Time 0.019929 -2023-02-13 17:40:50,233 - Epoch: [50][ 800/ 1207] Overall Loss 0.340827 Objective Loss 0.340827 LR 0.001000 Time 0.019916 -2023-02-13 17:40:50,423 - Epoch: [50][ 810/ 1207] Overall Loss 0.341126 Objective Loss 0.341126 LR 0.001000 Time 0.019904 -2023-02-13 17:40:50,613 - Epoch: [50][ 820/ 1207] Overall Loss 0.341741 Objective Loss 0.341741 LR 0.001000 Time 0.019893 -2023-02-13 17:40:50,803 - Epoch: [50][ 830/ 1207] Overall Loss 0.342216 Objective Loss 0.342216 LR 0.001000 Time 0.019881 -2023-02-13 17:40:50,993 - Epoch: [50][ 840/ 1207] Overall Loss 0.342510 Objective Loss 0.342510 LR 0.001000 Time 0.019871 -2023-02-13 17:40:51,183 - Epoch: [50][ 850/ 1207] Overall Loss 0.342431 Objective Loss 0.342431 LR 0.001000 Time 0.019860 -2023-02-13 17:40:51,373 - Epoch: [50][ 860/ 1207] Overall Loss 0.342544 Objective Loss 0.342544 LR 0.001000 Time 0.019849 -2023-02-13 17:40:51,563 - Epoch: [50][ 870/ 1207] Overall Loss 0.342365 Objective Loss 0.342365 LR 0.001000 Time 0.019839 -2023-02-13 17:40:51,753 - Epoch: [50][ 880/ 1207] Overall Loss 0.342627 Objective Loss 0.342627 LR 0.001000 Time 0.019828 -2023-02-13 17:40:51,943 - Epoch: [50][ 890/ 1207] Overall Loss 0.342568 Objective Loss 0.342568 LR 0.001000 Time 0.019819 -2023-02-13 17:40:52,132 - Epoch: [50][ 900/ 1207] Overall Loss 0.342443 Objective Loss 0.342443 LR 0.001000 Time 0.019809 -2023-02-13 17:40:52,322 - Epoch: [50][ 910/ 1207] Overall Loss 0.342176 Objective Loss 0.342176 LR 0.001000 Time 0.019799 -2023-02-13 17:40:52,511 - Epoch: [50][ 920/ 1207] Overall Loss 0.342382 Objective Loss 0.342382 LR 0.001000 Time 0.019790 -2023-02-13 17:40:52,701 - Epoch: [50][ 930/ 1207] Overall Loss 0.342168 Objective Loss 0.342168 LR 0.001000 Time 0.019781 -2023-02-13 17:40:52,891 - Epoch: [50][ 940/ 1207] Overall Loss 0.341939 Objective Loss 0.341939 LR 0.001000 Time 0.019771 -2023-02-13 17:40:53,080 - Epoch: [50][ 950/ 1207] Overall Loss 0.342061 Objective Loss 0.342061 LR 0.001000 Time 0.019761 -2023-02-13 17:40:53,269 - Epoch: [50][ 960/ 1207] Overall Loss 0.342022 Objective Loss 0.342022 LR 0.001000 Time 0.019753 -2023-02-13 17:40:53,459 - Epoch: [50][ 970/ 1207] Overall Loss 0.342100 Objective Loss 0.342100 LR 0.001000 Time 0.019744 -2023-02-13 17:40:53,649 - Epoch: [50][ 980/ 1207] Overall Loss 0.342081 Objective Loss 0.342081 LR 0.001000 Time 0.019737 -2023-02-13 17:40:53,839 - Epoch: [50][ 990/ 1207] Overall Loss 0.341487 Objective Loss 0.341487 LR 0.001000 Time 0.019729 -2023-02-13 17:40:54,029 - Epoch: [50][ 1000/ 1207] Overall Loss 0.341444 Objective Loss 0.341444 LR 0.001000 Time 0.019721 -2023-02-13 17:40:54,218 - Epoch: [50][ 1010/ 1207] Overall Loss 0.341354 Objective Loss 0.341354 LR 0.001000 Time 0.019713 -2023-02-13 17:40:54,408 - Epoch: [50][ 1020/ 1207] Overall Loss 0.341482 Objective Loss 0.341482 LR 0.001000 Time 0.019705 -2023-02-13 17:40:54,598 - Epoch: [50][ 1030/ 1207] Overall Loss 0.341301 Objective Loss 0.341301 LR 0.001000 Time 0.019698 -2023-02-13 17:40:54,788 - Epoch: [50][ 1040/ 1207] Overall Loss 0.341226 Objective Loss 0.341226 LR 0.001000 Time 0.019691 -2023-02-13 17:40:54,978 - Epoch: [50][ 1050/ 1207] Overall Loss 0.341111 Objective Loss 0.341111 LR 0.001000 Time 0.019683 -2023-02-13 17:40:55,167 - Epoch: [50][ 1060/ 1207] Overall Loss 0.341087 Objective Loss 0.341087 LR 0.001000 Time 0.019676 -2023-02-13 17:40:55,356 - Epoch: [50][ 1070/ 1207] Overall Loss 0.340748 Objective Loss 0.340748 LR 0.001000 Time 0.019669 -2023-02-13 17:40:55,546 - Epoch: [50][ 1080/ 1207] Overall Loss 0.340164 Objective Loss 0.340164 LR 0.001000 Time 0.019662 -2023-02-13 17:40:55,737 - Epoch: [50][ 1090/ 1207] Overall Loss 0.340446 Objective Loss 0.340446 LR 0.001000 Time 0.019656 -2023-02-13 17:40:55,928 - Epoch: [50][ 1100/ 1207] Overall Loss 0.340482 Objective Loss 0.340482 LR 0.001000 Time 0.019651 -2023-02-13 17:40:56,117 - Epoch: [50][ 1110/ 1207] Overall Loss 0.340559 Objective Loss 0.340559 LR 0.001000 Time 0.019644 -2023-02-13 17:40:56,307 - Epoch: [50][ 1120/ 1207] Overall Loss 0.340491 Objective Loss 0.340491 LR 0.001000 Time 0.019638 -2023-02-13 17:40:56,498 - Epoch: [50][ 1130/ 1207] Overall Loss 0.340269 Objective Loss 0.340269 LR 0.001000 Time 0.019632 -2023-02-13 17:40:56,688 - Epoch: [50][ 1140/ 1207] Overall Loss 0.340805 Objective Loss 0.340805 LR 0.001000 Time 0.019627 -2023-02-13 17:40:56,879 - Epoch: [50][ 1150/ 1207] Overall Loss 0.340938 Objective Loss 0.340938 LR 0.001000 Time 0.019622 -2023-02-13 17:40:57,068 - Epoch: [50][ 1160/ 1207] Overall Loss 0.340990 Objective Loss 0.340990 LR 0.001000 Time 0.019616 -2023-02-13 17:40:57,258 - Epoch: [50][ 1170/ 1207] Overall Loss 0.341308 Objective Loss 0.341308 LR 0.001000 Time 0.019610 -2023-02-13 17:40:57,448 - Epoch: [50][ 1180/ 1207] Overall Loss 0.341326 Objective Loss 0.341326 LR 0.001000 Time 0.019604 -2023-02-13 17:40:57,639 - Epoch: [50][ 1190/ 1207] Overall Loss 0.341311 Objective Loss 0.341311 LR 0.001000 Time 0.019600 -2023-02-13 17:40:57,879 - Epoch: [50][ 1200/ 1207] Overall Loss 0.341294 Objective Loss 0.341294 LR 0.001000 Time 0.019636 -2023-02-13 17:40:57,995 - Epoch: [50][ 1207/ 1207] Overall Loss 0.341288 Objective Loss 0.341288 Top1 83.231707 Top5 98.475610 LR 0.001000 Time 0.019618 -2023-02-13 17:40:58,070 - --- validate (epoch=50)----------- -2023-02-13 17:40:58,070 - 34311 samples (256 per mini-batch) -2023-02-13 17:40:58,468 - Epoch: [50][ 10/ 135] Loss 0.397406 Top1 81.914062 Top5 96.601562 -2023-02-13 17:40:58,598 - Epoch: [50][ 20/ 135] Loss 0.392727 Top1 81.269531 Top5 96.855469 -2023-02-13 17:40:58,726 - Epoch: [50][ 30/ 135] Loss 0.383680 Top1 81.562500 Top5 96.992188 -2023-02-13 17:40:58,855 - Epoch: [50][ 40/ 135] Loss 0.380102 Top1 81.884766 Top5 97.089844 -2023-02-13 17:40:58,983 - Epoch: [50][ 50/ 135] Loss 0.373036 Top1 81.773438 Top5 97.179688 -2023-02-13 17:40:59,115 - Epoch: [50][ 60/ 135] Loss 0.372585 Top1 81.861979 Top5 97.220052 -2023-02-13 17:40:59,239 - Epoch: [50][ 70/ 135] Loss 0.368492 Top1 81.986607 Top5 97.209821 -2023-02-13 17:40:59,364 - Epoch: [50][ 80/ 135] Loss 0.364639 Top1 81.967773 Top5 97.226562 -2023-02-13 17:40:59,486 - Epoch: [50][ 90/ 135] Loss 0.365143 Top1 81.914062 Top5 97.183160 -2023-02-13 17:40:59,610 - Epoch: [50][ 100/ 135] Loss 0.363437 Top1 81.988281 Top5 97.222656 -2023-02-13 17:40:59,736 - Epoch: [50][ 110/ 135] Loss 0.364494 Top1 81.974432 Top5 97.183949 -2023-02-13 17:40:59,865 - Epoch: [50][ 120/ 135] Loss 0.366797 Top1 81.975911 Top5 97.145182 -2023-02-13 17:40:59,998 - Epoch: [50][ 130/ 135] Loss 0.367486 Top1 81.986178 Top5 97.178486 -2023-02-13 17:41:00,045 - Epoch: [50][ 135/ 135] Loss 0.366527 Top1 82.032001 Top5 97.199149 -2023-02-13 17:41:00,118 - ==> Top1: 82.032 Top5: 97.199 Loss: 0.367 - -2023-02-13 17:41:00,119 - ==> Confusion: -[[ 829 5 7 4 7 3 0 1 7 64 1 2 0 4 5 5 4 2 2 2 13] - [ 2 910 1 6 5 37 2 29 4 2 0 3 3 1 2 3 1 1 8 3 10] - [ 9 9 915 22 2 0 21 20 0 3 4 2 2 4 4 9 1 4 12 6 9] - [ 6 2 15 897 1 6 1 1 2 0 14 0 10 1 20 1 4 2 24 1 8] - [ 20 9 1 1 957 17 1 3 0 5 0 6 3 5 12 7 4 4 0 4 7] - [ 4 13 0 8 3 951 3 21 1 2 2 19 5 13 1 3 2 7 4 5 3] - [ 3 4 24 4 0 3 1010 9 0 1 7 5 2 2 0 5 0 4 2 9 5] - [ 1 8 7 2 2 28 0 926 3 0 1 7 2 2 0 0 1 0 18 9 7] - [ 12 4 1 4 1 1 0 1 866 41 10 1 2 16 32 4 1 2 4 1 5] - [ 74 1 0 1 3 6 0 2 35 848 1 1 1 23 7 2 0 1 2 0 4] - [ 3 4 3 7 0 3 1 7 16 0 952 1 4 15 4 2 1 1 17 2 8] - [ 1 5 0 1 3 13 0 4 1 0 0 891 39 11 1 11 3 9 3 8 1] - [ 2 0 0 9 0 3 0 2 0 0 1 38 863 1 7 7 2 14 3 0 7] - [ 5 4 2 2 5 13 1 2 8 12 8 10 3 925 6 6 3 3 0 3 3] - [ 8 3 1 28 6 0 0 3 16 7 2 2 6 1 982 1 0 6 10 0 10] - [ 6 2 4 0 4 2 9 1 1 2 0 4 7 6 0 962 8 12 1 7 8] - [ 1 8 2 2 8 3 0 0 5 0 1 2 2 6 2 18 981 3 0 4 13] - [ 8 3 3 7 0 2 2 2 0 1 1 9 37 4 1 19 0 948 2 0 2] - [ 3 4 4 14 1 1 0 35 0 1 7 2 7 1 13 1 1 0 984 1 6] - [ 0 4 0 1 1 3 8 16 1 0 1 26 5 3 0 8 6 3 2 1051 9] - [ 183 244 196 198 134 257 72 230 89 100 185 143 419 324 200 165 197 126 183 291 9498]] - -2023-02-13 17:41:00,120 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:41:00,120 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:41:00,126 - - -2023-02-13 17:41:00,126 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:41:01,111 - Epoch: [51][ 10/ 1207] Overall Loss 0.323393 Objective Loss 0.323393 LR 0.001000 Time 0.098448 -2023-02-13 17:41:01,304 - Epoch: [51][ 20/ 1207] Overall Loss 0.320182 Objective Loss 0.320182 LR 0.001000 Time 0.058820 -2023-02-13 17:41:01,494 - Epoch: [51][ 30/ 1207] Overall Loss 0.326954 Objective Loss 0.326954 LR 0.001000 Time 0.045556 -2023-02-13 17:41:01,685 - Epoch: [51][ 40/ 1207] Overall Loss 0.326273 Objective Loss 0.326273 LR 0.001000 Time 0.038924 -2023-02-13 17:41:01,874 - Epoch: [51][ 50/ 1207] Overall Loss 0.329994 Objective Loss 0.329994 LR 0.001000 Time 0.034926 -2023-02-13 17:41:02,064 - Epoch: [51][ 60/ 1207] Overall Loss 0.328324 Objective Loss 0.328324 LR 0.001000 Time 0.032257 -2023-02-13 17:41:02,253 - Epoch: [51][ 70/ 1207] Overall Loss 0.330506 Objective Loss 0.330506 LR 0.001000 Time 0.030343 -2023-02-13 17:41:02,442 - Epoch: [51][ 80/ 1207] Overall Loss 0.330571 Objective Loss 0.330571 LR 0.001000 Time 0.028912 -2023-02-13 17:41:02,633 - Epoch: [51][ 90/ 1207] Overall Loss 0.332943 Objective Loss 0.332943 LR 0.001000 Time 0.027813 -2023-02-13 17:41:02,823 - Epoch: [51][ 100/ 1207] Overall Loss 0.333691 Objective Loss 0.333691 LR 0.001000 Time 0.026926 -2023-02-13 17:41:03,011 - Epoch: [51][ 110/ 1207] Overall Loss 0.334582 Objective Loss 0.334582 LR 0.001000 Time 0.026190 -2023-02-13 17:41:03,201 - Epoch: [51][ 120/ 1207] Overall Loss 0.333330 Objective Loss 0.333330 LR 0.001000 Time 0.025581 -2023-02-13 17:41:03,390 - Epoch: [51][ 130/ 1207] Overall Loss 0.333790 Objective Loss 0.333790 LR 0.001000 Time 0.025069 -2023-02-13 17:41:03,580 - Epoch: [51][ 140/ 1207] Overall Loss 0.334460 Objective Loss 0.334460 LR 0.001000 Time 0.024632 -2023-02-13 17:41:03,770 - Epoch: [51][ 150/ 1207] Overall Loss 0.333864 Objective Loss 0.333864 LR 0.001000 Time 0.024255 -2023-02-13 17:41:03,960 - Epoch: [51][ 160/ 1207] Overall Loss 0.334934 Objective Loss 0.334934 LR 0.001000 Time 0.023925 -2023-02-13 17:41:04,150 - Epoch: [51][ 170/ 1207] Overall Loss 0.335499 Objective Loss 0.335499 LR 0.001000 Time 0.023630 -2023-02-13 17:41:04,339 - Epoch: [51][ 180/ 1207] Overall Loss 0.336560 Objective Loss 0.336560 LR 0.001000 Time 0.023367 -2023-02-13 17:41:04,529 - Epoch: [51][ 190/ 1207] Overall Loss 0.337482 Objective Loss 0.337482 LR 0.001000 Time 0.023135 -2023-02-13 17:41:04,719 - Epoch: [51][ 200/ 1207] Overall Loss 0.338477 Objective Loss 0.338477 LR 0.001000 Time 0.022928 -2023-02-13 17:41:04,909 - Epoch: [51][ 210/ 1207] Overall Loss 0.338378 Objective Loss 0.338378 LR 0.001000 Time 0.022738 -2023-02-13 17:41:05,098 - Epoch: [51][ 220/ 1207] Overall Loss 0.337636 Objective Loss 0.337636 LR 0.001000 Time 0.022563 -2023-02-13 17:41:05,288 - Epoch: [51][ 230/ 1207] Overall Loss 0.337798 Objective Loss 0.337798 LR 0.001000 Time 0.022404 -2023-02-13 17:41:05,488 - Epoch: [51][ 240/ 1207] Overall Loss 0.336573 Objective Loss 0.336573 LR 0.001000 Time 0.022301 -2023-02-13 17:41:05,692 - Epoch: [51][ 250/ 1207] Overall Loss 0.338630 Objective Loss 0.338630 LR 0.001000 Time 0.022227 -2023-02-13 17:41:05,894 - Epoch: [51][ 260/ 1207] Overall Loss 0.338247 Objective Loss 0.338247 LR 0.001000 Time 0.022145 -2023-02-13 17:41:06,099 - Epoch: [51][ 270/ 1207] Overall Loss 0.338181 Objective Loss 0.338181 LR 0.001000 Time 0.022084 -2023-02-13 17:41:06,299 - Epoch: [51][ 280/ 1207] Overall Loss 0.338417 Objective Loss 0.338417 LR 0.001000 Time 0.022008 -2023-02-13 17:41:06,504 - Epoch: [51][ 290/ 1207] Overall Loss 0.338409 Objective Loss 0.338409 LR 0.001000 Time 0.021954 -2023-02-13 17:41:06,704 - Epoch: [51][ 300/ 1207] Overall Loss 0.338120 Objective Loss 0.338120 LR 0.001000 Time 0.021888 -2023-02-13 17:41:06,910 - Epoch: [51][ 310/ 1207] Overall Loss 0.338261 Objective Loss 0.338261 LR 0.001000 Time 0.021843 -2023-02-13 17:41:07,110 - Epoch: [51][ 320/ 1207] Overall Loss 0.337504 Objective Loss 0.337504 LR 0.001000 Time 0.021786 -2023-02-13 17:41:07,315 - Epoch: [51][ 330/ 1207] Overall Loss 0.337584 Objective Loss 0.337584 LR 0.001000 Time 0.021746 -2023-02-13 17:41:07,508 - Epoch: [51][ 340/ 1207] Overall Loss 0.336918 Objective Loss 0.336918 LR 0.001000 Time 0.021673 -2023-02-13 17:41:07,699 - Epoch: [51][ 350/ 1207] Overall Loss 0.337217 Objective Loss 0.337217 LR 0.001000 Time 0.021598 -2023-02-13 17:41:07,889 - Epoch: [51][ 360/ 1207] Overall Loss 0.336554 Objective Loss 0.336554 LR 0.001000 Time 0.021524 -2023-02-13 17:41:08,079 - Epoch: [51][ 370/ 1207] Overall Loss 0.335907 Objective Loss 0.335907 LR 0.001000 Time 0.021456 -2023-02-13 17:41:08,269 - Epoch: [51][ 380/ 1207] Overall Loss 0.335460 Objective Loss 0.335460 LR 0.001000 Time 0.021391 -2023-02-13 17:41:08,459 - Epoch: [51][ 390/ 1207] Overall Loss 0.334931 Objective Loss 0.334931 LR 0.001000 Time 0.021328 -2023-02-13 17:41:08,649 - Epoch: [51][ 400/ 1207] Overall Loss 0.334902 Objective Loss 0.334902 LR 0.001000 Time 0.021269 -2023-02-13 17:41:08,839 - Epoch: [51][ 410/ 1207] Overall Loss 0.334786 Objective Loss 0.334786 LR 0.001000 Time 0.021212 -2023-02-13 17:41:09,029 - Epoch: [51][ 420/ 1207] Overall Loss 0.334875 Objective Loss 0.334875 LR 0.001000 Time 0.021159 -2023-02-13 17:41:09,219 - Epoch: [51][ 430/ 1207] Overall Loss 0.334565 Objective Loss 0.334565 LR 0.001000 Time 0.021107 -2023-02-13 17:41:09,407 - Epoch: [51][ 440/ 1207] Overall Loss 0.333324 Objective Loss 0.333324 LR 0.001000 Time 0.021055 -2023-02-13 17:41:09,598 - Epoch: [51][ 450/ 1207] Overall Loss 0.333171 Objective Loss 0.333171 LR 0.001000 Time 0.021010 -2023-02-13 17:41:09,788 - Epoch: [51][ 460/ 1207] Overall Loss 0.333943 Objective Loss 0.333943 LR 0.001000 Time 0.020965 -2023-02-13 17:41:09,977 - Epoch: [51][ 470/ 1207] Overall Loss 0.334011 Objective Loss 0.334011 LR 0.001000 Time 0.020921 -2023-02-13 17:41:10,167 - Epoch: [51][ 480/ 1207] Overall Loss 0.334223 Objective Loss 0.334223 LR 0.001000 Time 0.020880 -2023-02-13 17:41:10,356 - Epoch: [51][ 490/ 1207] Overall Loss 0.333920 Objective Loss 0.333920 LR 0.001000 Time 0.020840 -2023-02-13 17:41:10,546 - Epoch: [51][ 500/ 1207] Overall Loss 0.333378 Objective Loss 0.333378 LR 0.001000 Time 0.020800 -2023-02-13 17:41:10,736 - Epoch: [51][ 510/ 1207] Overall Loss 0.333239 Objective Loss 0.333239 LR 0.001000 Time 0.020765 -2023-02-13 17:41:10,926 - Epoch: [51][ 520/ 1207] Overall Loss 0.333254 Objective Loss 0.333254 LR 0.001000 Time 0.020730 -2023-02-13 17:41:11,115 - Epoch: [51][ 530/ 1207] Overall Loss 0.334020 Objective Loss 0.334020 LR 0.001000 Time 0.020696 -2023-02-13 17:41:11,305 - Epoch: [51][ 540/ 1207] Overall Loss 0.333839 Objective Loss 0.333839 LR 0.001000 Time 0.020663 -2023-02-13 17:41:11,495 - Epoch: [51][ 550/ 1207] Overall Loss 0.332918 Objective Loss 0.332918 LR 0.001000 Time 0.020632 -2023-02-13 17:41:11,685 - Epoch: [51][ 560/ 1207] Overall Loss 0.333856 Objective Loss 0.333856 LR 0.001000 Time 0.020603 -2023-02-13 17:41:11,876 - Epoch: [51][ 570/ 1207] Overall Loss 0.334633 Objective Loss 0.334633 LR 0.001000 Time 0.020575 -2023-02-13 17:41:12,065 - Epoch: [51][ 580/ 1207] Overall Loss 0.334166 Objective Loss 0.334166 LR 0.001000 Time 0.020546 -2023-02-13 17:41:12,255 - Epoch: [51][ 590/ 1207] Overall Loss 0.334111 Objective Loss 0.334111 LR 0.001000 Time 0.020519 -2023-02-13 17:41:12,444 - Epoch: [51][ 600/ 1207] Overall Loss 0.334233 Objective Loss 0.334233 LR 0.001000 Time 0.020491 -2023-02-13 17:41:12,634 - Epoch: [51][ 610/ 1207] Overall Loss 0.334656 Objective Loss 0.334656 LR 0.001000 Time 0.020467 -2023-02-13 17:41:12,823 - Epoch: [51][ 620/ 1207] Overall Loss 0.334880 Objective Loss 0.334880 LR 0.001000 Time 0.020441 -2023-02-13 17:41:13,012 - Epoch: [51][ 630/ 1207] Overall Loss 0.335067 Objective Loss 0.335067 LR 0.001000 Time 0.020416 -2023-02-13 17:41:13,202 - Epoch: [51][ 640/ 1207] Overall Loss 0.334983 Objective Loss 0.334983 LR 0.001000 Time 0.020393 -2023-02-13 17:41:13,392 - Epoch: [51][ 650/ 1207] Overall Loss 0.334402 Objective Loss 0.334402 LR 0.001000 Time 0.020370 -2023-02-13 17:41:13,581 - Epoch: [51][ 660/ 1207] Overall Loss 0.333959 Objective Loss 0.333959 LR 0.001000 Time 0.020349 -2023-02-13 17:41:13,772 - Epoch: [51][ 670/ 1207] Overall Loss 0.333916 Objective Loss 0.333916 LR 0.001000 Time 0.020328 -2023-02-13 17:41:13,961 - Epoch: [51][ 680/ 1207] Overall Loss 0.333632 Objective Loss 0.333632 LR 0.001000 Time 0.020307 -2023-02-13 17:41:14,150 - Epoch: [51][ 690/ 1207] Overall Loss 0.333773 Objective Loss 0.333773 LR 0.001000 Time 0.020286 -2023-02-13 17:41:14,339 - Epoch: [51][ 700/ 1207] Overall Loss 0.333743 Objective Loss 0.333743 LR 0.001000 Time 0.020266 -2023-02-13 17:41:14,529 - Epoch: [51][ 710/ 1207] Overall Loss 0.333624 Objective Loss 0.333624 LR 0.001000 Time 0.020247 -2023-02-13 17:41:14,719 - Epoch: [51][ 720/ 1207] Overall Loss 0.333920 Objective Loss 0.333920 LR 0.001000 Time 0.020229 -2023-02-13 17:41:14,909 - Epoch: [51][ 730/ 1207] Overall Loss 0.333887 Objective Loss 0.333887 LR 0.001000 Time 0.020212 -2023-02-13 17:41:15,098 - Epoch: [51][ 740/ 1207] Overall Loss 0.333809 Objective Loss 0.333809 LR 0.001000 Time 0.020194 -2023-02-13 17:41:15,288 - Epoch: [51][ 750/ 1207] Overall Loss 0.334230 Objective Loss 0.334230 LR 0.001000 Time 0.020177 -2023-02-13 17:41:15,477 - Epoch: [51][ 760/ 1207] Overall Loss 0.333915 Objective Loss 0.333915 LR 0.001000 Time 0.020161 -2023-02-13 17:41:15,668 - Epoch: [51][ 770/ 1207] Overall Loss 0.334288 Objective Loss 0.334288 LR 0.001000 Time 0.020146 -2023-02-13 17:41:15,859 - Epoch: [51][ 780/ 1207] Overall Loss 0.334297 Objective Loss 0.334297 LR 0.001000 Time 0.020133 -2023-02-13 17:41:16,049 - Epoch: [51][ 790/ 1207] Overall Loss 0.334471 Objective Loss 0.334471 LR 0.001000 Time 0.020117 -2023-02-13 17:41:16,239 - Epoch: [51][ 800/ 1207] Overall Loss 0.334499 Objective Loss 0.334499 LR 0.001000 Time 0.020103 -2023-02-13 17:41:16,429 - Epoch: [51][ 810/ 1207] Overall Loss 0.334259 Objective Loss 0.334259 LR 0.001000 Time 0.020089 -2023-02-13 17:41:16,619 - Epoch: [51][ 820/ 1207] Overall Loss 0.334403 Objective Loss 0.334403 LR 0.001000 Time 0.020075 -2023-02-13 17:41:16,809 - Epoch: [51][ 830/ 1207] Overall Loss 0.334847 Objective Loss 0.334847 LR 0.001000 Time 0.020062 -2023-02-13 17:41:16,999 - Epoch: [51][ 840/ 1207] Overall Loss 0.334830 Objective Loss 0.334830 LR 0.001000 Time 0.020049 -2023-02-13 17:41:17,189 - Epoch: [51][ 850/ 1207] Overall Loss 0.335044 Objective Loss 0.335044 LR 0.001000 Time 0.020036 -2023-02-13 17:41:17,379 - Epoch: [51][ 860/ 1207] Overall Loss 0.335003 Objective Loss 0.335003 LR 0.001000 Time 0.020024 -2023-02-13 17:41:17,570 - Epoch: [51][ 870/ 1207] Overall Loss 0.335040 Objective Loss 0.335040 LR 0.001000 Time 0.020012 -2023-02-13 17:41:17,760 - Epoch: [51][ 880/ 1207] Overall Loss 0.335384 Objective Loss 0.335384 LR 0.001000 Time 0.020000 -2023-02-13 17:41:17,949 - Epoch: [51][ 890/ 1207] Overall Loss 0.335365 Objective Loss 0.335365 LR 0.001000 Time 0.019988 -2023-02-13 17:41:18,138 - Epoch: [51][ 900/ 1207] Overall Loss 0.335081 Objective Loss 0.335081 LR 0.001000 Time 0.019976 -2023-02-13 17:41:18,328 - Epoch: [51][ 910/ 1207] Overall Loss 0.335238 Objective Loss 0.335238 LR 0.001000 Time 0.019964 -2023-02-13 17:41:18,518 - Epoch: [51][ 920/ 1207] Overall Loss 0.334940 Objective Loss 0.334940 LR 0.001000 Time 0.019953 -2023-02-13 17:41:18,708 - Epoch: [51][ 930/ 1207] Overall Loss 0.335334 Objective Loss 0.335334 LR 0.001000 Time 0.019942 -2023-02-13 17:41:18,898 - Epoch: [51][ 940/ 1207] Overall Loss 0.335146 Objective Loss 0.335146 LR 0.001000 Time 0.019932 -2023-02-13 17:41:19,087 - Epoch: [51][ 950/ 1207] Overall Loss 0.335057 Objective Loss 0.335057 LR 0.001000 Time 0.019921 -2023-02-13 17:41:19,277 - Epoch: [51][ 960/ 1207] Overall Loss 0.335118 Objective Loss 0.335118 LR 0.001000 Time 0.019911 -2023-02-13 17:41:19,467 - Epoch: [51][ 970/ 1207] Overall Loss 0.335143 Objective Loss 0.335143 LR 0.001000 Time 0.019902 -2023-02-13 17:41:19,658 - Epoch: [51][ 980/ 1207] Overall Loss 0.334977 Objective Loss 0.334977 LR 0.001000 Time 0.019893 -2023-02-13 17:41:19,847 - Epoch: [51][ 990/ 1207] Overall Loss 0.335060 Objective Loss 0.335060 LR 0.001000 Time 0.019883 -2023-02-13 17:41:20,037 - Epoch: [51][ 1000/ 1207] Overall Loss 0.335159 Objective Loss 0.335159 LR 0.001000 Time 0.019873 -2023-02-13 17:41:20,227 - Epoch: [51][ 1010/ 1207] Overall Loss 0.335260 Objective Loss 0.335260 LR 0.001000 Time 0.019864 -2023-02-13 17:41:20,417 - Epoch: [51][ 1020/ 1207] Overall Loss 0.335353 Objective Loss 0.335353 LR 0.001000 Time 0.019855 -2023-02-13 17:41:20,608 - Epoch: [51][ 1030/ 1207] Overall Loss 0.335201 Objective Loss 0.335201 LR 0.001000 Time 0.019847 -2023-02-13 17:41:20,799 - Epoch: [51][ 1040/ 1207] Overall Loss 0.335694 Objective Loss 0.335694 LR 0.001000 Time 0.019840 -2023-02-13 17:41:20,990 - Epoch: [51][ 1050/ 1207] Overall Loss 0.336166 Objective Loss 0.336166 LR 0.001000 Time 0.019833 -2023-02-13 17:41:21,180 - Epoch: [51][ 1060/ 1207] Overall Loss 0.336152 Objective Loss 0.336152 LR 0.001000 Time 0.019824 -2023-02-13 17:41:21,369 - Epoch: [51][ 1070/ 1207] Overall Loss 0.336392 Objective Loss 0.336392 LR 0.001000 Time 0.019816 -2023-02-13 17:41:21,560 - Epoch: [51][ 1080/ 1207] Overall Loss 0.336409 Objective Loss 0.336409 LR 0.001000 Time 0.019808 -2023-02-13 17:41:21,751 - Epoch: [51][ 1090/ 1207] Overall Loss 0.336745 Objective Loss 0.336745 LR 0.001000 Time 0.019801 -2023-02-13 17:41:21,941 - Epoch: [51][ 1100/ 1207] Overall Loss 0.336804 Objective Loss 0.336804 LR 0.001000 Time 0.019794 -2023-02-13 17:41:22,131 - Epoch: [51][ 1110/ 1207] Overall Loss 0.336989 Objective Loss 0.336989 LR 0.001000 Time 0.019786 -2023-02-13 17:41:22,321 - Epoch: [51][ 1120/ 1207] Overall Loss 0.337387 Objective Loss 0.337387 LR 0.001000 Time 0.019779 -2023-02-13 17:41:22,511 - Epoch: [51][ 1130/ 1207] Overall Loss 0.337662 Objective Loss 0.337662 LR 0.001000 Time 0.019771 -2023-02-13 17:41:22,702 - Epoch: [51][ 1140/ 1207] Overall Loss 0.337879 Objective Loss 0.337879 LR 0.001000 Time 0.019765 -2023-02-13 17:41:22,891 - Epoch: [51][ 1150/ 1207] Overall Loss 0.337831 Objective Loss 0.337831 LR 0.001000 Time 0.019758 -2023-02-13 17:41:23,081 - Epoch: [51][ 1160/ 1207] Overall Loss 0.337950 Objective Loss 0.337950 LR 0.001000 Time 0.019751 -2023-02-13 17:41:23,271 - Epoch: [51][ 1170/ 1207] Overall Loss 0.338167 Objective Loss 0.338167 LR 0.001000 Time 0.019744 -2023-02-13 17:41:23,461 - Epoch: [51][ 1180/ 1207] Overall Loss 0.338402 Objective Loss 0.338402 LR 0.001000 Time 0.019737 -2023-02-13 17:41:23,651 - Epoch: [51][ 1190/ 1207] Overall Loss 0.338513 Objective Loss 0.338513 LR 0.001000 Time 0.019731 -2023-02-13 17:41:23,893 - Epoch: [51][ 1200/ 1207] Overall Loss 0.338901 Objective Loss 0.338901 LR 0.001000 Time 0.019768 -2023-02-13 17:41:24,009 - Epoch: [51][ 1207/ 1207] Overall Loss 0.339154 Objective Loss 0.339154 Top1 81.097561 Top5 96.951220 LR 0.001000 Time 0.019749 -2023-02-13 17:41:24,081 - --- validate (epoch=51)----------- -2023-02-13 17:41:24,081 - 34311 samples (256 per mini-batch) -2023-02-13 17:41:24,488 - Epoch: [51][ 10/ 135] Loss 0.354848 Top1 81.562500 Top5 97.343750 -2023-02-13 17:41:24,620 - Epoch: [51][ 20/ 135] Loss 0.371279 Top1 81.015625 Top5 96.992188 -2023-02-13 17:41:24,749 - Epoch: [51][ 30/ 135] Loss 0.376060 Top1 80.924479 Top5 96.953125 -2023-02-13 17:41:24,877 - Epoch: [51][ 40/ 135] Loss 0.373949 Top1 80.917969 Top5 96.914062 -2023-02-13 17:41:25,007 - Epoch: [51][ 50/ 135] Loss 0.376207 Top1 80.757812 Top5 96.859375 -2023-02-13 17:41:25,137 - Epoch: [51][ 60/ 135] Loss 0.375412 Top1 80.898438 Top5 96.848958 -2023-02-13 17:41:25,262 - Epoch: [51][ 70/ 135] Loss 0.375530 Top1 81.004464 Top5 96.847098 -2023-02-13 17:41:25,386 - Epoch: [51][ 80/ 135] Loss 0.375013 Top1 81.005859 Top5 96.831055 -2023-02-13 17:41:25,519 - Epoch: [51][ 90/ 135] Loss 0.376968 Top1 81.028646 Top5 96.827257 -2023-02-13 17:41:25,654 - Epoch: [51][ 100/ 135] Loss 0.378231 Top1 81.000000 Top5 96.855469 -2023-02-13 17:41:25,781 - Epoch: [51][ 110/ 135] Loss 0.379439 Top1 80.916193 Top5 96.814631 -2023-02-13 17:41:25,910 - Epoch: [51][ 120/ 135] Loss 0.378173 Top1 80.898438 Top5 96.793620 -2023-02-13 17:41:26,040 - Epoch: [51][ 130/ 135] Loss 0.377939 Top1 80.904447 Top5 96.820913 -2023-02-13 17:41:26,085 - Epoch: [51][ 135/ 135] Loss 0.376950 Top1 80.883682 Top5 96.829005 -2023-02-13 17:41:26,153 - ==> Top1: 80.884 Top5: 96.829 Loss: 0.377 - -2023-02-13 17:41:26,154 - ==> Confusion: -[[ 848 5 5 0 10 5 0 3 5 57 0 4 0 4 2 2 7 4 0 0 6] - [ 5 935 2 3 11 33 2 7 5 2 2 4 0 1 1 4 3 0 5 2 6] - [ 10 7 949 8 7 5 13 20 0 0 5 1 1 2 2 8 3 2 8 1 6] - [ 7 2 28 865 3 10 0 2 2 1 25 0 5 2 29 2 5 3 19 0 6] - [ 16 12 2 0 981 10 1 1 0 5 1 5 0 9 8 4 7 0 0 2 2] - [ 5 27 2 4 6 971 2 14 1 2 3 8 1 12 1 3 2 2 0 2 2] - [ 2 6 20 4 1 7 1011 17 2 2 7 1 1 3 1 5 0 1 3 4 1] - [ 2 13 8 2 4 40 2 915 3 1 4 10 0 3 0 0 0 0 8 5 4] - [ 18 6 0 1 2 2 0 1 899 36 7 1 1 9 15 2 1 2 3 0 3] - [ 103 3 0 1 6 6 0 3 56 799 1 0 0 20 5 0 1 1 0 0 7] - [ 0 7 7 3 3 4 1 5 18 1 966 3 0 11 0 1 1 1 14 0 5] - [ 3 2 3 1 5 26 1 4 3 2 0 902 17 18 1 5 3 3 0 6 0] - [ 2 1 4 6 1 11 0 2 6 1 2 66 813 1 2 6 4 20 2 1 8] - [ 6 2 1 0 4 18 0 2 21 20 6 8 3 907 8 7 2 1 0 0 8] - [ 13 5 2 9 7 3 0 0 34 5 1 3 2 1 980 1 1 6 10 0 9] - [ 6 3 6 0 13 2 4 2 1 1 1 14 5 9 0 946 18 8 0 4 3] - [ 6 7 4 1 14 5 1 0 0 1 0 2 2 7 5 7 987 0 3 3 6] - [ 10 3 2 4 2 5 0 1 0 2 2 19 20 6 2 15 1 954 0 1 2] - [ 5 14 6 10 3 4 0 38 6 2 6 4 3 0 14 1 1 0 966 0 3] - [ 1 4 0 0 3 18 7 37 0 1 3 29 0 7 0 7 13 1 1 1008 8] - [ 222 405 292 137 190 276 75 222 130 102 261 176 307 320 214 133 305 103 176 238 9150]] - -2023-02-13 17:41:26,155 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:41:26,155 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:41:26,161 - - -2023-02-13 17:41:26,161 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:41:27,056 - Epoch: [52][ 10/ 1207] Overall Loss 0.357850 Objective Loss 0.357850 LR 0.001000 Time 0.089426 -2023-02-13 17:41:27,255 - Epoch: [52][ 20/ 1207] Overall Loss 0.345317 Objective Loss 0.345317 LR 0.001000 Time 0.054636 -2023-02-13 17:41:27,444 - Epoch: [52][ 30/ 1207] Overall Loss 0.341931 Objective Loss 0.341931 LR 0.001000 Time 0.042729 -2023-02-13 17:41:27,633 - Epoch: [52][ 40/ 1207] Overall Loss 0.336493 Objective Loss 0.336493 LR 0.001000 Time 0.036751 -2023-02-13 17:41:27,821 - Epoch: [52][ 50/ 1207] Overall Loss 0.336576 Objective Loss 0.336576 LR 0.001000 Time 0.033165 -2023-02-13 17:41:28,010 - Epoch: [52][ 60/ 1207] Overall Loss 0.339691 Objective Loss 0.339691 LR 0.001000 Time 0.030766 -2023-02-13 17:41:28,198 - Epoch: [52][ 70/ 1207] Overall Loss 0.341521 Objective Loss 0.341521 LR 0.001000 Time 0.029062 -2023-02-13 17:41:28,386 - Epoch: [52][ 80/ 1207] Overall Loss 0.339510 Objective Loss 0.339510 LR 0.001000 Time 0.027776 -2023-02-13 17:41:28,575 - Epoch: [52][ 90/ 1207] Overall Loss 0.339346 Objective Loss 0.339346 LR 0.001000 Time 0.026784 -2023-02-13 17:41:28,764 - Epoch: [52][ 100/ 1207] Overall Loss 0.335765 Objective Loss 0.335765 LR 0.001000 Time 0.025986 -2023-02-13 17:41:28,953 - Epoch: [52][ 110/ 1207] Overall Loss 0.337140 Objective Loss 0.337140 LR 0.001000 Time 0.025339 -2023-02-13 17:41:29,148 - Epoch: [52][ 120/ 1207] Overall Loss 0.336140 Objective Loss 0.336140 LR 0.001000 Time 0.024851 -2023-02-13 17:41:29,341 - Epoch: [52][ 130/ 1207] Overall Loss 0.334674 Objective Loss 0.334674 LR 0.001000 Time 0.024424 -2023-02-13 17:41:29,537 - Epoch: [52][ 140/ 1207] Overall Loss 0.331635 Objective Loss 0.331635 LR 0.001000 Time 0.024073 -2023-02-13 17:41:29,731 - Epoch: [52][ 150/ 1207] Overall Loss 0.332872 Objective Loss 0.332872 LR 0.001000 Time 0.023761 -2023-02-13 17:41:29,926 - Epoch: [52][ 160/ 1207] Overall Loss 0.330639 Objective Loss 0.330639 LR 0.001000 Time 0.023494 -2023-02-13 17:41:30,120 - Epoch: [52][ 170/ 1207] Overall Loss 0.331173 Objective Loss 0.331173 LR 0.001000 Time 0.023248 -2023-02-13 17:41:30,315 - Epoch: [52][ 180/ 1207] Overall Loss 0.332002 Objective Loss 0.332002 LR 0.001000 Time 0.023038 -2023-02-13 17:41:30,509 - Epoch: [52][ 190/ 1207] Overall Loss 0.332824 Objective Loss 0.332824 LR 0.001000 Time 0.022843 -2023-02-13 17:41:30,705 - Epoch: [52][ 200/ 1207] Overall Loss 0.331420 Objective Loss 0.331420 LR 0.001000 Time 0.022679 -2023-02-13 17:41:30,899 - Epoch: [52][ 210/ 1207] Overall Loss 0.331178 Objective Loss 0.331178 LR 0.001000 Time 0.022522 -2023-02-13 17:41:31,094 - Epoch: [52][ 220/ 1207] Overall Loss 0.332004 Objective Loss 0.332004 LR 0.001000 Time 0.022386 -2023-02-13 17:41:31,287 - Epoch: [52][ 230/ 1207] Overall Loss 0.332085 Objective Loss 0.332085 LR 0.001000 Time 0.022250 -2023-02-13 17:41:31,483 - Epoch: [52][ 240/ 1207] Overall Loss 0.332561 Objective Loss 0.332561 LR 0.001000 Time 0.022135 -2023-02-13 17:41:31,677 - Epoch: [52][ 250/ 1207] Overall Loss 0.332509 Objective Loss 0.332509 LR 0.001000 Time 0.022026 -2023-02-13 17:41:31,872 - Epoch: [52][ 260/ 1207] Overall Loss 0.333326 Objective Loss 0.333326 LR 0.001000 Time 0.021928 -2023-02-13 17:41:32,066 - Epoch: [52][ 270/ 1207] Overall Loss 0.334050 Objective Loss 0.334050 LR 0.001000 Time 0.021830 -2023-02-13 17:41:32,260 - Epoch: [52][ 280/ 1207] Overall Loss 0.336549 Objective Loss 0.336549 LR 0.001000 Time 0.021745 -2023-02-13 17:41:32,454 - Epoch: [52][ 290/ 1207] Overall Loss 0.337059 Objective Loss 0.337059 LR 0.001000 Time 0.021662 -2023-02-13 17:41:32,650 - Epoch: [52][ 300/ 1207] Overall Loss 0.336819 Objective Loss 0.336819 LR 0.001000 Time 0.021591 -2023-02-13 17:41:32,844 - Epoch: [52][ 310/ 1207] Overall Loss 0.335988 Objective Loss 0.335988 LR 0.001000 Time 0.021519 -2023-02-13 17:41:33,039 - Epoch: [52][ 320/ 1207] Overall Loss 0.335178 Objective Loss 0.335178 LR 0.001000 Time 0.021455 -2023-02-13 17:41:33,233 - Epoch: [52][ 330/ 1207] Overall Loss 0.334461 Objective Loss 0.334461 LR 0.001000 Time 0.021391 -2023-02-13 17:41:33,428 - Epoch: [52][ 340/ 1207] Overall Loss 0.333000 Objective Loss 0.333000 LR 0.001000 Time 0.021336 -2023-02-13 17:41:33,622 - Epoch: [52][ 350/ 1207] Overall Loss 0.333244 Objective Loss 0.333244 LR 0.001000 Time 0.021277 -2023-02-13 17:41:33,818 - Epoch: [52][ 360/ 1207] Overall Loss 0.333576 Objective Loss 0.333576 LR 0.001000 Time 0.021231 -2023-02-13 17:41:34,012 - Epoch: [52][ 370/ 1207] Overall Loss 0.334252 Objective Loss 0.334252 LR 0.001000 Time 0.021179 -2023-02-13 17:41:34,207 - Epoch: [52][ 380/ 1207] Overall Loss 0.333420 Objective Loss 0.333420 LR 0.001000 Time 0.021135 -2023-02-13 17:41:34,400 - Epoch: [52][ 390/ 1207] Overall Loss 0.333559 Objective Loss 0.333559 LR 0.001000 Time 0.021087 -2023-02-13 17:41:34,595 - Epoch: [52][ 400/ 1207] Overall Loss 0.333345 Objective Loss 0.333345 LR 0.001000 Time 0.021047 -2023-02-13 17:41:34,789 - Epoch: [52][ 410/ 1207] Overall Loss 0.334206 Objective Loss 0.334206 LR 0.001000 Time 0.021006 -2023-02-13 17:41:34,985 - Epoch: [52][ 420/ 1207] Overall Loss 0.335324 Objective Loss 0.335324 LR 0.001000 Time 0.020970 -2023-02-13 17:41:35,178 - Epoch: [52][ 430/ 1207] Overall Loss 0.335117 Objective Loss 0.335117 LR 0.001000 Time 0.020932 -2023-02-13 17:41:35,374 - Epoch: [52][ 440/ 1207] Overall Loss 0.334777 Objective Loss 0.334777 LR 0.001000 Time 0.020901 -2023-02-13 17:41:35,568 - Epoch: [52][ 450/ 1207] Overall Loss 0.335293 Objective Loss 0.335293 LR 0.001000 Time 0.020866 -2023-02-13 17:41:35,765 - Epoch: [52][ 460/ 1207] Overall Loss 0.336329 Objective Loss 0.336329 LR 0.001000 Time 0.020839 -2023-02-13 17:41:35,959 - Epoch: [52][ 470/ 1207] Overall Loss 0.336338 Objective Loss 0.336338 LR 0.001000 Time 0.020808 -2023-02-13 17:41:36,154 - Epoch: [52][ 480/ 1207] Overall Loss 0.336421 Objective Loss 0.336421 LR 0.001000 Time 0.020781 -2023-02-13 17:41:36,348 - Epoch: [52][ 490/ 1207] Overall Loss 0.337033 Objective Loss 0.337033 LR 0.001000 Time 0.020752 -2023-02-13 17:41:36,543 - Epoch: [52][ 500/ 1207] Overall Loss 0.337074 Objective Loss 0.337074 LR 0.001000 Time 0.020726 -2023-02-13 17:41:36,738 - Epoch: [52][ 510/ 1207] Overall Loss 0.337222 Objective Loss 0.337222 LR 0.001000 Time 0.020701 -2023-02-13 17:41:36,934 - Epoch: [52][ 520/ 1207] Overall Loss 0.336804 Objective Loss 0.336804 LR 0.001000 Time 0.020680 -2023-02-13 17:41:37,128 - Epoch: [52][ 530/ 1207] Overall Loss 0.337048 Objective Loss 0.337048 LR 0.001000 Time 0.020653 -2023-02-13 17:41:37,323 - Epoch: [52][ 540/ 1207] Overall Loss 0.337063 Objective Loss 0.337063 LR 0.001000 Time 0.020633 -2023-02-13 17:41:37,517 - Epoch: [52][ 550/ 1207] Overall Loss 0.337156 Objective Loss 0.337156 LR 0.001000 Time 0.020609 -2023-02-13 17:41:37,714 - Epoch: [52][ 560/ 1207] Overall Loss 0.336604 Objective Loss 0.336604 LR 0.001000 Time 0.020592 -2023-02-13 17:41:37,907 - Epoch: [52][ 570/ 1207] Overall Loss 0.336511 Objective Loss 0.336511 LR 0.001000 Time 0.020569 -2023-02-13 17:41:38,103 - Epoch: [52][ 580/ 1207] Overall Loss 0.336233 Objective Loss 0.336233 LR 0.001000 Time 0.020551 -2023-02-13 17:41:38,297 - Epoch: [52][ 590/ 1207] Overall Loss 0.336883 Objective Loss 0.336883 LR 0.001000 Time 0.020531 -2023-02-13 17:41:38,493 - Epoch: [52][ 600/ 1207] Overall Loss 0.337086 Objective Loss 0.337086 LR 0.001000 Time 0.020515 -2023-02-13 17:41:38,687 - Epoch: [52][ 610/ 1207] Overall Loss 0.336830 Objective Loss 0.336830 LR 0.001000 Time 0.020497 -2023-02-13 17:41:38,885 - Epoch: [52][ 620/ 1207] Overall Loss 0.336573 Objective Loss 0.336573 LR 0.001000 Time 0.020484 -2023-02-13 17:41:39,080 - Epoch: [52][ 630/ 1207] Overall Loss 0.336430 Objective Loss 0.336430 LR 0.001000 Time 0.020469 -2023-02-13 17:41:39,278 - Epoch: [52][ 640/ 1207] Overall Loss 0.335910 Objective Loss 0.335910 LR 0.001000 Time 0.020457 -2023-02-13 17:41:39,474 - Epoch: [52][ 650/ 1207] Overall Loss 0.335921 Objective Loss 0.335921 LR 0.001000 Time 0.020443 -2023-02-13 17:41:39,672 - Epoch: [52][ 660/ 1207] Overall Loss 0.335385 Objective Loss 0.335385 LR 0.001000 Time 0.020433 -2023-02-13 17:41:39,868 - Epoch: [52][ 670/ 1207] Overall Loss 0.335356 Objective Loss 0.335356 LR 0.001000 Time 0.020421 -2023-02-13 17:41:40,066 - Epoch: [52][ 680/ 1207] Overall Loss 0.335738 Objective Loss 0.335738 LR 0.001000 Time 0.020410 -2023-02-13 17:41:40,262 - Epoch: [52][ 690/ 1207] Overall Loss 0.335650 Objective Loss 0.335650 LR 0.001000 Time 0.020398 -2023-02-13 17:41:40,459 - Epoch: [52][ 700/ 1207] Overall Loss 0.335305 Objective Loss 0.335305 LR 0.001000 Time 0.020388 -2023-02-13 17:41:40,656 - Epoch: [52][ 710/ 1207] Overall Loss 0.335462 Objective Loss 0.335462 LR 0.001000 Time 0.020377 -2023-02-13 17:41:40,855 - Epoch: [52][ 720/ 1207] Overall Loss 0.334944 Objective Loss 0.334944 LR 0.001000 Time 0.020370 -2023-02-13 17:41:41,051 - Epoch: [52][ 730/ 1207] Overall Loss 0.335065 Objective Loss 0.335065 LR 0.001000 Time 0.020360 -2023-02-13 17:41:41,249 - Epoch: [52][ 740/ 1207] Overall Loss 0.335049 Objective Loss 0.335049 LR 0.001000 Time 0.020351 -2023-02-13 17:41:41,446 - Epoch: [52][ 750/ 1207] Overall Loss 0.335879 Objective Loss 0.335879 LR 0.001000 Time 0.020341 -2023-02-13 17:41:41,644 - Epoch: [52][ 760/ 1207] Overall Loss 0.336284 Objective Loss 0.336284 LR 0.001000 Time 0.020334 -2023-02-13 17:41:41,842 - Epoch: [52][ 770/ 1207] Overall Loss 0.335929 Objective Loss 0.335929 LR 0.001000 Time 0.020326 -2023-02-13 17:41:42,040 - Epoch: [52][ 780/ 1207] Overall Loss 0.336068 Objective Loss 0.336068 LR 0.001000 Time 0.020319 -2023-02-13 17:41:42,236 - Epoch: [52][ 790/ 1207] Overall Loss 0.336161 Objective Loss 0.336161 LR 0.001000 Time 0.020311 -2023-02-13 17:41:42,434 - Epoch: [52][ 800/ 1207] Overall Loss 0.335879 Objective Loss 0.335879 LR 0.001000 Time 0.020303 -2023-02-13 17:41:42,631 - Epoch: [52][ 810/ 1207] Overall Loss 0.336037 Objective Loss 0.336037 LR 0.001000 Time 0.020295 -2023-02-13 17:41:42,827 - Epoch: [52][ 820/ 1207] Overall Loss 0.336061 Objective Loss 0.336061 LR 0.001000 Time 0.020287 -2023-02-13 17:41:43,021 - Epoch: [52][ 830/ 1207] Overall Loss 0.336202 Objective Loss 0.336202 LR 0.001000 Time 0.020275 -2023-02-13 17:41:43,216 - Epoch: [52][ 840/ 1207] Overall Loss 0.336331 Objective Loss 0.336331 LR 0.001000 Time 0.020266 -2023-02-13 17:41:43,410 - Epoch: [52][ 850/ 1207] Overall Loss 0.336161 Objective Loss 0.336161 LR 0.001000 Time 0.020255 -2023-02-13 17:41:43,606 - Epoch: [52][ 860/ 1207] Overall Loss 0.336504 Objective Loss 0.336504 LR 0.001000 Time 0.020246 -2023-02-13 17:41:43,800 - Epoch: [52][ 870/ 1207] Overall Loss 0.336638 Objective Loss 0.336638 LR 0.001000 Time 0.020237 -2023-02-13 17:41:43,996 - Epoch: [52][ 880/ 1207] Overall Loss 0.336889 Objective Loss 0.336889 LR 0.001000 Time 0.020229 -2023-02-13 17:41:44,190 - Epoch: [52][ 890/ 1207] Overall Loss 0.336664 Objective Loss 0.336664 LR 0.001000 Time 0.020219 -2023-02-13 17:41:44,386 - Epoch: [52][ 900/ 1207] Overall Loss 0.336270 Objective Loss 0.336270 LR 0.001000 Time 0.020212 -2023-02-13 17:41:44,580 - Epoch: [52][ 910/ 1207] Overall Loss 0.336363 Objective Loss 0.336363 LR 0.001000 Time 0.020202 -2023-02-13 17:41:44,776 - Epoch: [52][ 920/ 1207] Overall Loss 0.336712 Objective Loss 0.336712 LR 0.001000 Time 0.020196 -2023-02-13 17:41:44,971 - Epoch: [52][ 930/ 1207] Overall Loss 0.336547 Objective Loss 0.336547 LR 0.001000 Time 0.020187 -2023-02-13 17:41:45,166 - Epoch: [52][ 940/ 1207] Overall Loss 0.336418 Objective Loss 0.336418 LR 0.001000 Time 0.020180 -2023-02-13 17:41:45,360 - Epoch: [52][ 950/ 1207] Overall Loss 0.336610 Objective Loss 0.336610 LR 0.001000 Time 0.020171 -2023-02-13 17:41:45,555 - Epoch: [52][ 960/ 1207] Overall Loss 0.336618 Objective Loss 0.336618 LR 0.001000 Time 0.020164 -2023-02-13 17:41:45,750 - Epoch: [52][ 970/ 1207] Overall Loss 0.336557 Objective Loss 0.336557 LR 0.001000 Time 0.020157 -2023-02-13 17:41:45,948 - Epoch: [52][ 980/ 1207] Overall Loss 0.336122 Objective Loss 0.336122 LR 0.001000 Time 0.020152 -2023-02-13 17:41:46,141 - Epoch: [52][ 990/ 1207] Overall Loss 0.336032 Objective Loss 0.336032 LR 0.001000 Time 0.020144 -2023-02-13 17:41:46,337 - Epoch: [52][ 1000/ 1207] Overall Loss 0.336225 Objective Loss 0.336225 LR 0.001000 Time 0.020138 -2023-02-13 17:41:46,531 - Epoch: [52][ 1010/ 1207] Overall Loss 0.336126 Objective Loss 0.336126 LR 0.001000 Time 0.020131 -2023-02-13 17:41:46,728 - Epoch: [52][ 1020/ 1207] Overall Loss 0.336363 Objective Loss 0.336363 LR 0.001000 Time 0.020126 -2023-02-13 17:41:46,923 - Epoch: [52][ 1030/ 1207] Overall Loss 0.336471 Objective Loss 0.336471 LR 0.001000 Time 0.020119 -2023-02-13 17:41:47,119 - Epoch: [52][ 1040/ 1207] Overall Loss 0.336718 Objective Loss 0.336718 LR 0.001000 Time 0.020114 -2023-02-13 17:41:47,313 - Epoch: [52][ 1050/ 1207] Overall Loss 0.336726 Objective Loss 0.336726 LR 0.001000 Time 0.020107 -2023-02-13 17:41:47,505 - Epoch: [52][ 1060/ 1207] Overall Loss 0.336891 Objective Loss 0.336891 LR 0.001000 Time 0.020098 -2023-02-13 17:41:47,696 - Epoch: [52][ 1070/ 1207] Overall Loss 0.336866 Objective Loss 0.336866 LR 0.001000 Time 0.020088 -2023-02-13 17:41:47,887 - Epoch: [52][ 1080/ 1207] Overall Loss 0.336950 Objective Loss 0.336950 LR 0.001000 Time 0.020078 -2023-02-13 17:41:48,077 - Epoch: [52][ 1090/ 1207] Overall Loss 0.336888 Objective Loss 0.336888 LR 0.001000 Time 0.020068 -2023-02-13 17:41:48,267 - Epoch: [52][ 1100/ 1207] Overall Loss 0.337182 Objective Loss 0.337182 LR 0.001000 Time 0.020058 -2023-02-13 17:41:48,456 - Epoch: [52][ 1110/ 1207] Overall Loss 0.337297 Objective Loss 0.337297 LR 0.001000 Time 0.020048 -2023-02-13 17:41:48,646 - Epoch: [52][ 1120/ 1207] Overall Loss 0.337319 Objective Loss 0.337319 LR 0.001000 Time 0.020038 -2023-02-13 17:41:48,836 - Epoch: [52][ 1130/ 1207] Overall Loss 0.337405 Objective Loss 0.337405 LR 0.001000 Time 0.020029 -2023-02-13 17:41:49,026 - Epoch: [52][ 1140/ 1207] Overall Loss 0.337505 Objective Loss 0.337505 LR 0.001000 Time 0.020019 -2023-02-13 17:41:49,216 - Epoch: [52][ 1150/ 1207] Overall Loss 0.337859 Objective Loss 0.337859 LR 0.001000 Time 0.020010 -2023-02-13 17:41:49,406 - Epoch: [52][ 1160/ 1207] Overall Loss 0.337931 Objective Loss 0.337931 LR 0.001000 Time 0.020001 -2023-02-13 17:41:49,595 - Epoch: [52][ 1170/ 1207] Overall Loss 0.338070 Objective Loss 0.338070 LR 0.001000 Time 0.019991 -2023-02-13 17:41:49,786 - Epoch: [52][ 1180/ 1207] Overall Loss 0.338183 Objective Loss 0.338183 LR 0.001000 Time 0.019983 -2023-02-13 17:41:49,975 - Epoch: [52][ 1190/ 1207] Overall Loss 0.338253 Objective Loss 0.338253 LR 0.001000 Time 0.019974 -2023-02-13 17:41:50,221 - Epoch: [52][ 1200/ 1207] Overall Loss 0.338471 Objective Loss 0.338471 LR 0.001000 Time 0.020012 -2023-02-13 17:41:50,336 - Epoch: [52][ 1207/ 1207] Overall Loss 0.338951 Objective Loss 0.338951 Top1 83.231707 Top5 98.780488 LR 0.001000 Time 0.019991 -2023-02-13 17:41:50,425 - --- validate (epoch=52)----------- -2023-02-13 17:41:50,425 - 34311 samples (256 per mini-batch) -2023-02-13 17:41:50,825 - Epoch: [52][ 10/ 135] Loss 0.341605 Top1 82.109375 Top5 97.773438 -2023-02-13 17:41:50,952 - Epoch: [52][ 20/ 135] Loss 0.391269 Top1 81.738281 Top5 97.343750 -2023-02-13 17:41:51,077 - Epoch: [52][ 30/ 135] Loss 0.388029 Top1 81.588542 Top5 97.096354 -2023-02-13 17:41:51,203 - Epoch: [52][ 40/ 135] Loss 0.390731 Top1 81.591797 Top5 97.099609 -2023-02-13 17:41:51,333 - Epoch: [52][ 50/ 135] Loss 0.384755 Top1 81.648438 Top5 97.125000 -2023-02-13 17:41:51,462 - Epoch: [52][ 60/ 135] Loss 0.377514 Top1 81.731771 Top5 97.135417 -2023-02-13 17:41:51,591 - Epoch: [52][ 70/ 135] Loss 0.381064 Top1 81.640625 Top5 97.092634 -2023-02-13 17:41:51,723 - Epoch: [52][ 80/ 135] Loss 0.374704 Top1 81.796875 Top5 97.138672 -2023-02-13 17:41:51,864 - Epoch: [52][ 90/ 135] Loss 0.378521 Top1 81.684028 Top5 97.065972 -2023-02-13 17:41:51,993 - Epoch: [52][ 100/ 135] Loss 0.378098 Top1 81.652344 Top5 97.136719 -2023-02-13 17:41:52,115 - Epoch: [52][ 110/ 135] Loss 0.382902 Top1 81.558949 Top5 97.130682 -2023-02-13 17:41:52,242 - Epoch: [52][ 120/ 135] Loss 0.384017 Top1 81.500651 Top5 97.106120 -2023-02-13 17:41:52,371 - Epoch: [52][ 130/ 135] Loss 0.382723 Top1 81.502404 Top5 97.103365 -2023-02-13 17:41:52,419 - Epoch: [52][ 135/ 135] Loss 0.379150 Top1 81.571508 Top5 97.111713 -2023-02-13 17:41:52,490 - ==> Top1: 81.572 Top5: 97.112 Loss: 0.379 - -2023-02-13 17:41:52,491 - ==> Confusion: -[[ 844 2 7 6 15 1 0 1 0 49 0 2 1 2 3 6 6 7 0 1 14] - [ 4 893 1 5 14 44 8 13 3 2 1 4 4 0 3 2 9 2 5 4 12] - [ 12 4 905 21 3 4 38 11 0 0 1 4 3 4 1 17 3 6 6 9 6] - [ 4 0 20 892 3 5 2 3 2 3 16 1 16 0 13 5 3 6 14 1 7] - [ 17 6 1 3 989 9 2 1 0 3 1 4 2 5 4 5 3 3 0 3 5] - [ 9 9 4 5 2 956 7 11 1 2 2 12 9 14 1 2 2 5 3 11 3] - [ 3 2 12 3 0 8 1041 3 0 1 0 2 1 1 1 5 1 2 1 9 3] - [ 2 13 12 8 3 42 9 867 2 2 1 8 7 2 0 0 0 1 18 21 6] - [ 24 0 1 3 0 0 1 1 843 43 12 2 3 21 42 3 2 1 4 0 3] - [ 94 1 7 2 5 1 0 0 27 834 0 1 0 23 3 2 2 1 0 2 7] - [ 2 4 10 10 0 0 7 1 15 1 965 1 4 6 3 1 1 1 12 3 4] - [ 2 2 1 0 3 12 1 4 0 1 0 867 65 5 2 14 2 7 3 12 2] - [ 2 0 3 5 4 2 0 0 2 0 0 22 878 1 2 7 0 21 1 4 5] - [ 5 4 0 1 8 6 0 1 7 18 7 9 8 925 4 12 1 0 0 3 5] - [ 12 2 3 24 7 2 0 2 11 3 8 3 3 2 984 1 1 6 7 1 10] - [ 4 2 1 0 6 1 4 1 0 1 0 8 12 2 1 969 6 11 0 9 8] - [ 5 3 0 1 12 3 1 0 2 1 0 5 6 1 2 17 980 3 1 4 14] - [ 9 1 2 6 0 1 0 1 0 0 0 12 45 0 1 22 1 946 0 1 3] - [ 5 1 9 23 1 2 2 29 3 0 6 1 9 0 17 4 2 1 963 3 5] - [ 0 1 1 2 1 4 13 12 0 0 0 17 7 4 0 8 2 6 0 1063 7] - [ 187 202 221 202 159 250 166 152 79 88 208 113 440 376 171 197 221 150 149 319 9384]] - -2023-02-13 17:41:52,493 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:41:52,493 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:41:52,498 - - -2023-02-13 17:41:52,499 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:41:53,383 - Epoch: [53][ 10/ 1207] Overall Loss 0.349253 Objective Loss 0.349253 LR 0.001000 Time 0.088420 -2023-02-13 17:41:53,589 - Epoch: [53][ 20/ 1207] Overall Loss 0.341387 Objective Loss 0.341387 LR 0.001000 Time 0.054461 -2023-02-13 17:41:53,791 - Epoch: [53][ 30/ 1207] Overall Loss 0.346389 Objective Loss 0.346389 LR 0.001000 Time 0.043032 -2023-02-13 17:41:53,988 - Epoch: [53][ 40/ 1207] Overall Loss 0.351378 Objective Loss 0.351378 LR 0.001000 Time 0.037187 -2023-02-13 17:41:54,189 - Epoch: [53][ 50/ 1207] Overall Loss 0.353101 Objective Loss 0.353101 LR 0.001000 Time 0.033766 -2023-02-13 17:41:54,386 - Epoch: [53][ 60/ 1207] Overall Loss 0.350411 Objective Loss 0.350411 LR 0.001000 Time 0.031419 -2023-02-13 17:41:54,589 - Epoch: [53][ 70/ 1207] Overall Loss 0.342335 Objective Loss 0.342335 LR 0.001000 Time 0.029818 -2023-02-13 17:41:54,787 - Epoch: [53][ 80/ 1207] Overall Loss 0.341396 Objective Loss 0.341396 LR 0.001000 Time 0.028559 -2023-02-13 17:41:54,989 - Epoch: [53][ 90/ 1207] Overall Loss 0.338729 Objective Loss 0.338729 LR 0.001000 Time 0.027627 -2023-02-13 17:41:55,186 - Epoch: [53][ 100/ 1207] Overall Loss 0.338335 Objective Loss 0.338335 LR 0.001000 Time 0.026835 -2023-02-13 17:41:55,388 - Epoch: [53][ 110/ 1207] Overall Loss 0.338556 Objective Loss 0.338556 LR 0.001000 Time 0.026226 -2023-02-13 17:41:55,585 - Epoch: [53][ 120/ 1207] Overall Loss 0.337857 Objective Loss 0.337857 LR 0.001000 Time 0.025681 -2023-02-13 17:41:55,789 - Epoch: [53][ 130/ 1207] Overall Loss 0.334714 Objective Loss 0.334714 LR 0.001000 Time 0.025267 -2023-02-13 17:41:55,986 - Epoch: [53][ 140/ 1207] Overall Loss 0.335968 Objective Loss 0.335968 LR 0.001000 Time 0.024871 -2023-02-13 17:41:56,189 - Epoch: [53][ 150/ 1207] Overall Loss 0.335985 Objective Loss 0.335985 LR 0.001000 Time 0.024559 -2023-02-13 17:41:56,386 - Epoch: [53][ 160/ 1207] Overall Loss 0.335688 Objective Loss 0.335688 LR 0.001000 Time 0.024255 -2023-02-13 17:41:56,588 - Epoch: [53][ 170/ 1207] Overall Loss 0.334787 Objective Loss 0.334787 LR 0.001000 Time 0.024012 -2023-02-13 17:41:56,786 - Epoch: [53][ 180/ 1207] Overall Loss 0.333763 Objective Loss 0.333763 LR 0.001000 Time 0.023776 -2023-02-13 17:41:56,988 - Epoch: [53][ 190/ 1207] Overall Loss 0.335002 Objective Loss 0.335002 LR 0.001000 Time 0.023589 -2023-02-13 17:41:57,185 - Epoch: [53][ 200/ 1207] Overall Loss 0.333096 Objective Loss 0.333096 LR 0.001000 Time 0.023391 -2023-02-13 17:41:57,388 - Epoch: [53][ 210/ 1207] Overall Loss 0.334053 Objective Loss 0.334053 LR 0.001000 Time 0.023242 -2023-02-13 17:41:57,585 - Epoch: [53][ 220/ 1207] Overall Loss 0.334681 Objective Loss 0.334681 LR 0.001000 Time 0.023081 -2023-02-13 17:41:57,788 - Epoch: [53][ 230/ 1207] Overall Loss 0.334430 Objective Loss 0.334430 LR 0.001000 Time 0.022958 -2023-02-13 17:41:57,986 - Epoch: [53][ 240/ 1207] Overall Loss 0.333611 Objective Loss 0.333611 LR 0.001000 Time 0.022824 -2023-02-13 17:41:58,188 - Epoch: [53][ 250/ 1207] Overall Loss 0.334263 Objective Loss 0.334263 LR 0.001000 Time 0.022717 -2023-02-13 17:41:58,386 - Epoch: [53][ 260/ 1207] Overall Loss 0.334903 Objective Loss 0.334903 LR 0.001000 Time 0.022601 -2023-02-13 17:41:58,588 - Epoch: [53][ 270/ 1207] Overall Loss 0.334688 Objective Loss 0.334688 LR 0.001000 Time 0.022512 -2023-02-13 17:41:58,785 - Epoch: [53][ 280/ 1207] Overall Loss 0.335061 Objective Loss 0.335061 LR 0.001000 Time 0.022410 -2023-02-13 17:41:58,987 - Epoch: [53][ 290/ 1207] Overall Loss 0.336049 Objective Loss 0.336049 LR 0.001000 Time 0.022333 -2023-02-13 17:41:59,184 - Epoch: [53][ 300/ 1207] Overall Loss 0.335381 Objective Loss 0.335381 LR 0.001000 Time 0.022245 -2023-02-13 17:41:59,386 - Epoch: [53][ 310/ 1207] Overall Loss 0.335313 Objective Loss 0.335313 LR 0.001000 Time 0.022178 -2023-02-13 17:41:59,583 - Epoch: [53][ 320/ 1207] Overall Loss 0.335590 Objective Loss 0.335590 LR 0.001000 Time 0.022099 -2023-02-13 17:41:59,786 - Epoch: [53][ 330/ 1207] Overall Loss 0.335483 Objective Loss 0.335483 LR 0.001000 Time 0.022044 -2023-02-13 17:41:59,985 - Epoch: [53][ 340/ 1207] Overall Loss 0.335100 Objective Loss 0.335100 LR 0.001000 Time 0.021978 -2023-02-13 17:42:00,187 - Epoch: [53][ 350/ 1207] Overall Loss 0.334645 Objective Loss 0.334645 LR 0.001000 Time 0.021927 -2023-02-13 17:42:00,386 - Epoch: [53][ 360/ 1207] Overall Loss 0.335822 Objective Loss 0.335822 LR 0.001000 Time 0.021868 -2023-02-13 17:42:00,589 - Epoch: [53][ 370/ 1207] Overall Loss 0.335907 Objective Loss 0.335907 LR 0.001000 Time 0.021825 -2023-02-13 17:42:00,789 - Epoch: [53][ 380/ 1207] Overall Loss 0.335857 Objective Loss 0.335857 LR 0.001000 Time 0.021776 -2023-02-13 17:42:00,991 - Epoch: [53][ 390/ 1207] Overall Loss 0.335408 Objective Loss 0.335408 LR 0.001000 Time 0.021735 -2023-02-13 17:42:01,188 - Epoch: [53][ 400/ 1207] Overall Loss 0.335232 Objective Loss 0.335232 LR 0.001000 Time 0.021683 -2023-02-13 17:42:01,390 - Epoch: [53][ 410/ 1207] Overall Loss 0.335858 Objective Loss 0.335858 LR 0.001000 Time 0.021645 -2023-02-13 17:42:01,588 - Epoch: [53][ 420/ 1207] Overall Loss 0.335082 Objective Loss 0.335082 LR 0.001000 Time 0.021601 -2023-02-13 17:42:01,792 - Epoch: [53][ 430/ 1207] Overall Loss 0.335655 Objective Loss 0.335655 LR 0.001000 Time 0.021573 -2023-02-13 17:42:01,991 - Epoch: [53][ 440/ 1207] Overall Loss 0.335162 Objective Loss 0.335162 LR 0.001000 Time 0.021532 -2023-02-13 17:42:02,192 - Epoch: [53][ 450/ 1207] Overall Loss 0.335148 Objective Loss 0.335148 LR 0.001000 Time 0.021500 -2023-02-13 17:42:02,390 - Epoch: [53][ 460/ 1207] Overall Loss 0.335136 Objective Loss 0.335136 LR 0.001000 Time 0.021464 -2023-02-13 17:42:02,594 - Epoch: [53][ 470/ 1207] Overall Loss 0.335056 Objective Loss 0.335056 LR 0.001000 Time 0.021439 -2023-02-13 17:42:02,792 - Epoch: [53][ 480/ 1207] Overall Loss 0.335052 Objective Loss 0.335052 LR 0.001000 Time 0.021405 -2023-02-13 17:42:02,994 - Epoch: [53][ 490/ 1207] Overall Loss 0.334953 Objective Loss 0.334953 LR 0.001000 Time 0.021379 -2023-02-13 17:42:03,194 - Epoch: [53][ 500/ 1207] Overall Loss 0.335413 Objective Loss 0.335413 LR 0.001000 Time 0.021350 -2023-02-13 17:42:03,396 - Epoch: [53][ 510/ 1207] Overall Loss 0.335665 Objective Loss 0.335665 LR 0.001000 Time 0.021328 -2023-02-13 17:42:03,594 - Epoch: [53][ 520/ 1207] Overall Loss 0.336038 Objective Loss 0.336038 LR 0.001000 Time 0.021298 -2023-02-13 17:42:03,797 - Epoch: [53][ 530/ 1207] Overall Loss 0.335369 Objective Loss 0.335369 LR 0.001000 Time 0.021278 -2023-02-13 17:42:03,994 - Epoch: [53][ 540/ 1207] Overall Loss 0.335090 Objective Loss 0.335090 LR 0.001000 Time 0.021247 -2023-02-13 17:42:04,195 - Epoch: [53][ 550/ 1207] Overall Loss 0.335003 Objective Loss 0.335003 LR 0.001000 Time 0.021226 -2023-02-13 17:42:04,393 - Epoch: [53][ 560/ 1207] Overall Loss 0.335454 Objective Loss 0.335454 LR 0.001000 Time 0.021200 -2023-02-13 17:42:04,594 - Epoch: [53][ 570/ 1207] Overall Loss 0.335006 Objective Loss 0.335006 LR 0.001000 Time 0.021181 -2023-02-13 17:42:04,793 - Epoch: [53][ 580/ 1207] Overall Loss 0.335173 Objective Loss 0.335173 LR 0.001000 Time 0.021157 -2023-02-13 17:42:04,995 - Epoch: [53][ 590/ 1207] Overall Loss 0.334807 Objective Loss 0.334807 LR 0.001000 Time 0.021140 -2023-02-13 17:42:05,194 - Epoch: [53][ 600/ 1207] Overall Loss 0.334324 Objective Loss 0.334324 LR 0.001000 Time 0.021118 -2023-02-13 17:42:05,396 - Epoch: [53][ 610/ 1207] Overall Loss 0.334431 Objective Loss 0.334431 LR 0.001000 Time 0.021103 -2023-02-13 17:42:05,595 - Epoch: [53][ 620/ 1207] Overall Loss 0.333867 Objective Loss 0.333867 LR 0.001000 Time 0.021084 -2023-02-13 17:42:05,798 - Epoch: [53][ 630/ 1207] Overall Loss 0.333919 Objective Loss 0.333919 LR 0.001000 Time 0.021070 -2023-02-13 17:42:05,996 - Epoch: [53][ 640/ 1207] Overall Loss 0.333750 Objective Loss 0.333750 LR 0.001000 Time 0.021049 -2023-02-13 17:42:06,198 - Epoch: [53][ 650/ 1207] Overall Loss 0.333321 Objective Loss 0.333321 LR 0.001000 Time 0.021035 -2023-02-13 17:42:06,395 - Epoch: [53][ 660/ 1207] Overall Loss 0.333165 Objective Loss 0.333165 LR 0.001000 Time 0.021015 -2023-02-13 17:42:06,598 - Epoch: [53][ 670/ 1207] Overall Loss 0.333223 Objective Loss 0.333223 LR 0.001000 Time 0.021004 -2023-02-13 17:42:06,797 - Epoch: [53][ 680/ 1207] Overall Loss 0.333793 Objective Loss 0.333793 LR 0.001000 Time 0.020986 -2023-02-13 17:42:06,999 - Epoch: [53][ 690/ 1207] Overall Loss 0.333420 Objective Loss 0.333420 LR 0.001000 Time 0.020975 -2023-02-13 17:42:07,198 - Epoch: [53][ 700/ 1207] Overall Loss 0.334168 Objective Loss 0.334168 LR 0.001000 Time 0.020959 -2023-02-13 17:42:07,398 - Epoch: [53][ 710/ 1207] Overall Loss 0.334484 Objective Loss 0.334484 LR 0.001000 Time 0.020946 -2023-02-13 17:42:07,597 - Epoch: [53][ 720/ 1207] Overall Loss 0.334604 Objective Loss 0.334604 LR 0.001000 Time 0.020930 -2023-02-13 17:42:07,800 - Epoch: [53][ 730/ 1207] Overall Loss 0.334708 Objective Loss 0.334708 LR 0.001000 Time 0.020921 -2023-02-13 17:42:07,998 - Epoch: [53][ 740/ 1207] Overall Loss 0.334593 Objective Loss 0.334593 LR 0.001000 Time 0.020906 -2023-02-13 17:42:08,201 - Epoch: [53][ 750/ 1207] Overall Loss 0.334314 Objective Loss 0.334314 LR 0.001000 Time 0.020896 -2023-02-13 17:42:08,399 - Epoch: [53][ 760/ 1207] Overall Loss 0.334181 Objective Loss 0.334181 LR 0.001000 Time 0.020881 -2023-02-13 17:42:08,601 - Epoch: [53][ 770/ 1207] Overall Loss 0.334524 Objective Loss 0.334524 LR 0.001000 Time 0.020872 -2023-02-13 17:42:08,799 - Epoch: [53][ 780/ 1207] Overall Loss 0.334539 Objective Loss 0.334539 LR 0.001000 Time 0.020858 -2023-02-13 17:42:09,003 - Epoch: [53][ 790/ 1207] Overall Loss 0.335022 Objective Loss 0.335022 LR 0.001000 Time 0.020852 -2023-02-13 17:42:09,206 - Epoch: [53][ 800/ 1207] Overall Loss 0.334928 Objective Loss 0.334928 LR 0.001000 Time 0.020844 -2023-02-13 17:42:09,414 - Epoch: [53][ 810/ 1207] Overall Loss 0.335353 Objective Loss 0.335353 LR 0.001000 Time 0.020843 -2023-02-13 17:42:09,613 - Epoch: [53][ 820/ 1207] Overall Loss 0.335467 Objective Loss 0.335467 LR 0.001000 Time 0.020832 -2023-02-13 17:42:09,810 - Epoch: [53][ 830/ 1207] Overall Loss 0.335797 Objective Loss 0.335797 LR 0.001000 Time 0.020817 -2023-02-13 17:42:10,004 - Epoch: [53][ 840/ 1207] Overall Loss 0.335394 Objective Loss 0.335394 LR 0.001000 Time 0.020799 -2023-02-13 17:42:10,199 - Epoch: [53][ 850/ 1207] Overall Loss 0.335514 Objective Loss 0.335514 LR 0.001000 Time 0.020785 -2023-02-13 17:42:10,393 - Epoch: [53][ 860/ 1207] Overall Loss 0.335279 Objective Loss 0.335279 LR 0.001000 Time 0.020767 -2023-02-13 17:42:10,589 - Epoch: [53][ 870/ 1207] Overall Loss 0.335333 Objective Loss 0.335333 LR 0.001000 Time 0.020753 -2023-02-13 17:42:10,784 - Epoch: [53][ 880/ 1207] Overall Loss 0.335416 Objective Loss 0.335416 LR 0.001000 Time 0.020739 -2023-02-13 17:42:10,981 - Epoch: [53][ 890/ 1207] Overall Loss 0.335660 Objective Loss 0.335660 LR 0.001000 Time 0.020727 -2023-02-13 17:42:11,175 - Epoch: [53][ 900/ 1207] Overall Loss 0.335836 Objective Loss 0.335836 LR 0.001000 Time 0.020711 -2023-02-13 17:42:11,370 - Epoch: [53][ 910/ 1207] Overall Loss 0.335527 Objective Loss 0.335527 LR 0.001000 Time 0.020698 -2023-02-13 17:42:11,565 - Epoch: [53][ 920/ 1207] Overall Loss 0.335867 Objective Loss 0.335867 LR 0.001000 Time 0.020684 -2023-02-13 17:42:11,761 - Epoch: [53][ 930/ 1207] Overall Loss 0.336187 Objective Loss 0.336187 LR 0.001000 Time 0.020673 -2023-02-13 17:42:11,957 - Epoch: [53][ 940/ 1207] Overall Loss 0.335819 Objective Loss 0.335819 LR 0.001000 Time 0.020660 -2023-02-13 17:42:12,152 - Epoch: [53][ 950/ 1207] Overall Loss 0.335799 Objective Loss 0.335799 LR 0.001000 Time 0.020648 -2023-02-13 17:42:12,347 - Epoch: [53][ 960/ 1207] Overall Loss 0.335898 Objective Loss 0.335898 LR 0.001000 Time 0.020636 -2023-02-13 17:42:12,542 - Epoch: [53][ 970/ 1207] Overall Loss 0.335939 Objective Loss 0.335939 LR 0.001000 Time 0.020624 -2023-02-13 17:42:12,737 - Epoch: [53][ 980/ 1207] Overall Loss 0.336107 Objective Loss 0.336107 LR 0.001000 Time 0.020612 -2023-02-13 17:42:12,934 - Epoch: [53][ 990/ 1207] Overall Loss 0.336420 Objective Loss 0.336420 LR 0.001000 Time 0.020602 -2023-02-13 17:42:13,129 - Epoch: [53][ 1000/ 1207] Overall Loss 0.336440 Objective Loss 0.336440 LR 0.001000 Time 0.020590 -2023-02-13 17:42:13,324 - Epoch: [53][ 1010/ 1207] Overall Loss 0.336496 Objective Loss 0.336496 LR 0.001000 Time 0.020580 -2023-02-13 17:42:13,519 - Epoch: [53][ 1020/ 1207] Overall Loss 0.336684 Objective Loss 0.336684 LR 0.001000 Time 0.020569 -2023-02-13 17:42:13,716 - Epoch: [53][ 1030/ 1207] Overall Loss 0.336795 Objective Loss 0.336795 LR 0.001000 Time 0.020560 -2023-02-13 17:42:13,911 - Epoch: [53][ 1040/ 1207] Overall Loss 0.336753 Objective Loss 0.336753 LR 0.001000 Time 0.020549 -2023-02-13 17:42:14,107 - Epoch: [53][ 1050/ 1207] Overall Loss 0.336700 Objective Loss 0.336700 LR 0.001000 Time 0.020540 -2023-02-13 17:42:14,300 - Epoch: [53][ 1060/ 1207] Overall Loss 0.336833 Objective Loss 0.336833 LR 0.001000 Time 0.020528 -2023-02-13 17:42:14,496 - Epoch: [53][ 1070/ 1207] Overall Loss 0.336812 Objective Loss 0.336812 LR 0.001000 Time 0.020519 -2023-02-13 17:42:14,691 - Epoch: [53][ 1080/ 1207] Overall Loss 0.336862 Objective Loss 0.336862 LR 0.001000 Time 0.020510 -2023-02-13 17:42:14,888 - Epoch: [53][ 1090/ 1207] Overall Loss 0.336880 Objective Loss 0.336880 LR 0.001000 Time 0.020502 -2023-02-13 17:42:15,084 - Epoch: [53][ 1100/ 1207] Overall Loss 0.336843 Objective Loss 0.336843 LR 0.001000 Time 0.020492 -2023-02-13 17:42:15,280 - Epoch: [53][ 1110/ 1207] Overall Loss 0.336969 Objective Loss 0.336969 LR 0.001000 Time 0.020485 -2023-02-13 17:42:15,474 - Epoch: [53][ 1120/ 1207] Overall Loss 0.336855 Objective Loss 0.336855 LR 0.001000 Time 0.020475 -2023-02-13 17:42:15,671 - Epoch: [53][ 1130/ 1207] Overall Loss 0.336941 Objective Loss 0.336941 LR 0.001000 Time 0.020467 -2023-02-13 17:42:15,867 - Epoch: [53][ 1140/ 1207] Overall Loss 0.337001 Objective Loss 0.337001 LR 0.001000 Time 0.020459 -2023-02-13 17:42:16,063 - Epoch: [53][ 1150/ 1207] Overall Loss 0.336842 Objective Loss 0.336842 LR 0.001000 Time 0.020452 -2023-02-13 17:42:16,257 - Epoch: [53][ 1160/ 1207] Overall Loss 0.336830 Objective Loss 0.336830 LR 0.001000 Time 0.020442 -2023-02-13 17:42:16,453 - Epoch: [53][ 1170/ 1207] Overall Loss 0.336979 Objective Loss 0.336979 LR 0.001000 Time 0.020435 -2023-02-13 17:42:16,648 - Epoch: [53][ 1180/ 1207] Overall Loss 0.336936 Objective Loss 0.336936 LR 0.001000 Time 0.020427 -2023-02-13 17:42:16,846 - Epoch: [53][ 1190/ 1207] Overall Loss 0.337142 Objective Loss 0.337142 LR 0.001000 Time 0.020421 -2023-02-13 17:42:17,096 - Epoch: [53][ 1200/ 1207] Overall Loss 0.337634 Objective Loss 0.337634 LR 0.001000 Time 0.020459 -2023-02-13 17:42:17,211 - Epoch: [53][ 1207/ 1207] Overall Loss 0.337557 Objective Loss 0.337557 Top1 81.402439 Top5 96.646341 LR 0.001000 Time 0.020435 -2023-02-13 17:42:17,282 - --- validate (epoch=53)----------- -2023-02-13 17:42:17,282 - 34311 samples (256 per mini-batch) -2023-02-13 17:42:17,783 - Epoch: [53][ 10/ 135] Loss 0.382001 Top1 81.914062 Top5 97.265625 -2023-02-13 17:42:17,913 - Epoch: [53][ 20/ 135] Loss 0.368967 Top1 81.464844 Top5 97.128906 -2023-02-13 17:42:18,043 - Epoch: [53][ 30/ 135] Loss 0.380230 Top1 81.119792 Top5 97.161458 -2023-02-13 17:42:18,171 - Epoch: [53][ 40/ 135] Loss 0.389863 Top1 80.986328 Top5 97.031250 -2023-02-13 17:42:18,298 - Epoch: [53][ 50/ 135] Loss 0.388927 Top1 80.898438 Top5 97.000000 -2023-02-13 17:42:18,424 - Epoch: [53][ 60/ 135] Loss 0.383661 Top1 81.028646 Top5 96.979167 -2023-02-13 17:42:18,554 - Epoch: [53][ 70/ 135] Loss 0.382568 Top1 80.993304 Top5 96.953125 -2023-02-13 17:42:18,699 - Epoch: [53][ 80/ 135] Loss 0.382100 Top1 80.903320 Top5 96.870117 -2023-02-13 17:42:18,838 - Epoch: [53][ 90/ 135] Loss 0.382402 Top1 80.733507 Top5 96.827257 -2023-02-13 17:42:18,984 - Epoch: [53][ 100/ 135] Loss 0.378965 Top1 80.769531 Top5 96.859375 -2023-02-13 17:42:19,123 - Epoch: [53][ 110/ 135] Loss 0.377856 Top1 80.724432 Top5 96.832386 -2023-02-13 17:42:19,268 - Epoch: [53][ 120/ 135] Loss 0.375647 Top1 80.764974 Top5 96.816406 -2023-02-13 17:42:19,408 - Epoch: [53][ 130/ 135] Loss 0.376268 Top1 80.769231 Top5 96.832933 -2023-02-13 17:42:19,453 - Epoch: [53][ 135/ 135] Loss 0.379333 Top1 80.740870 Top5 96.852321 -2023-02-13 17:42:19,532 - ==> Top1: 80.741 Top5: 96.852 Loss: 0.379 - -2023-02-13 17:42:19,532 - ==> Confusion: -[[ 832 3 6 3 11 2 0 0 8 64 0 5 0 2 6 4 6 3 2 2 8] - [ 3 941 3 3 6 19 1 20 2 2 1 2 0 3 3 3 3 0 12 2 4] - [ 9 2 946 12 0 3 13 17 1 1 5 0 2 4 3 8 1 4 18 5 4] - [ 5 3 23 865 1 5 1 3 3 4 19 0 2 4 36 0 2 9 27 0 4] - [ 22 9 1 1 982 13 0 3 2 5 1 3 0 0 6 6 3 0 0 6 3] - [ 2 41 2 5 6 921 2 30 5 7 4 12 2 9 3 1 2 0 8 6 2] - [ 3 5 21 4 1 4 1016 14 0 2 3 2 3 2 1 3 3 4 1 4 3] - [ 7 5 9 1 3 21 2 916 4 1 4 3 2 1 0 0 0 3 34 6 2] - [ 19 1 0 1 0 1 0 0 897 40 7 1 1 5 18 3 1 1 13 0 0] - [ 66 1 2 1 3 4 0 1 52 858 2 0 1 8 4 2 1 1 1 0 4] - [ 2 6 5 2 1 3 3 6 18 1 968 0 1 9 6 1 1 1 16 1 0] - [ 2 5 1 0 5 13 0 6 1 1 0 866 30 10 2 8 6 21 5 15 8] - [ 4 0 1 14 0 3 0 0 6 0 1 24 827 3 9 14 1 29 7 1 15] - [ 9 1 1 0 7 13 1 1 32 29 9 7 1 885 8 9 3 1 1 3 3] - [ 11 3 2 16 4 4 0 3 23 6 2 1 2 2 988 0 1 5 13 0 6] - [ 4 0 11 0 9 3 4 2 1 2 1 1 7 6 1 956 13 13 1 5 6] - [ 5 13 3 2 14 5 1 1 7 0 1 4 0 1 2 16 970 2 1 3 10] - [ 8 2 0 3 1 3 0 2 1 2 1 6 10 0 4 16 0 984 1 1 6] - [ 6 4 5 4 1 0 0 25 4 1 5 0 6 1 10 0 2 1 1009 1 1] - [ 1 3 0 1 3 6 7 19 0 0 1 18 1 5 0 7 4 4 5 1053 10] - [ 198 337 269 156 169 214 76 217 192 111 253 113 291 362 303 102 312 175 295 266 9023]] - -2023-02-13 17:42:19,534 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:42:19,534 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:42:19,540 - - -2023-02-13 17:42:19,540 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:42:20,418 - Epoch: [54][ 10/ 1207] Overall Loss 0.343898 Objective Loss 0.343898 LR 0.001000 Time 0.087739 -2023-02-13 17:42:20,620 - Epoch: [54][ 20/ 1207] Overall Loss 0.325408 Objective Loss 0.325408 LR 0.001000 Time 0.053901 -2023-02-13 17:42:20,815 - Epoch: [54][ 30/ 1207] Overall Loss 0.324460 Objective Loss 0.324460 LR 0.001000 Time 0.042445 -2023-02-13 17:42:21,006 - Epoch: [54][ 40/ 1207] Overall Loss 0.331642 Objective Loss 0.331642 LR 0.001000 Time 0.036594 -2023-02-13 17:42:21,193 - Epoch: [54][ 50/ 1207] Overall Loss 0.332835 Objective Loss 0.332835 LR 0.001000 Time 0.033012 -2023-02-13 17:42:21,381 - Epoch: [54][ 60/ 1207] Overall Loss 0.332599 Objective Loss 0.332599 LR 0.001000 Time 0.030638 -2023-02-13 17:42:21,569 - Epoch: [54][ 70/ 1207] Overall Loss 0.329053 Objective Loss 0.329053 LR 0.001000 Time 0.028940 -2023-02-13 17:42:21,757 - Epoch: [54][ 80/ 1207] Overall Loss 0.326092 Objective Loss 0.326092 LR 0.001000 Time 0.027672 -2023-02-13 17:42:21,946 - Epoch: [54][ 90/ 1207] Overall Loss 0.329380 Objective Loss 0.329380 LR 0.001000 Time 0.026686 -2023-02-13 17:42:22,133 - Epoch: [54][ 100/ 1207] Overall Loss 0.328284 Objective Loss 0.328284 LR 0.001000 Time 0.025888 -2023-02-13 17:42:22,322 - Epoch: [54][ 110/ 1207] Overall Loss 0.327997 Objective Loss 0.327997 LR 0.001000 Time 0.025243 -2023-02-13 17:42:22,509 - Epoch: [54][ 120/ 1207] Overall Loss 0.331458 Objective Loss 0.331458 LR 0.001000 Time 0.024702 -2023-02-13 17:42:22,698 - Epoch: [54][ 130/ 1207] Overall Loss 0.329295 Objective Loss 0.329295 LR 0.001000 Time 0.024248 -2023-02-13 17:42:22,886 - Epoch: [54][ 140/ 1207] Overall Loss 0.331862 Objective Loss 0.331862 LR 0.001000 Time 0.023860 -2023-02-13 17:42:23,074 - Epoch: [54][ 150/ 1207] Overall Loss 0.331528 Objective Loss 0.331528 LR 0.001000 Time 0.023519 -2023-02-13 17:42:23,262 - Epoch: [54][ 160/ 1207] Overall Loss 0.330193 Objective Loss 0.330193 LR 0.001000 Time 0.023222 -2023-02-13 17:42:23,451 - Epoch: [54][ 170/ 1207] Overall Loss 0.331865 Objective Loss 0.331865 LR 0.001000 Time 0.022966 -2023-02-13 17:42:23,640 - Epoch: [54][ 180/ 1207] Overall Loss 0.331545 Objective Loss 0.331545 LR 0.001000 Time 0.022735 -2023-02-13 17:42:23,829 - Epoch: [54][ 190/ 1207] Overall Loss 0.331264 Objective Loss 0.331264 LR 0.001000 Time 0.022534 -2023-02-13 17:42:24,018 - Epoch: [54][ 200/ 1207] Overall Loss 0.331700 Objective Loss 0.331700 LR 0.001000 Time 0.022350 -2023-02-13 17:42:24,207 - Epoch: [54][ 210/ 1207] Overall Loss 0.330957 Objective Loss 0.330957 LR 0.001000 Time 0.022181 -2023-02-13 17:42:24,395 - Epoch: [54][ 220/ 1207] Overall Loss 0.330207 Objective Loss 0.330207 LR 0.001000 Time 0.022027 -2023-02-13 17:42:24,584 - Epoch: [54][ 230/ 1207] Overall Loss 0.330311 Objective Loss 0.330311 LR 0.001000 Time 0.021888 -2023-02-13 17:42:24,772 - Epoch: [54][ 240/ 1207] Overall Loss 0.330026 Objective Loss 0.330026 LR 0.001000 Time 0.021758 -2023-02-13 17:42:24,961 - Epoch: [54][ 250/ 1207] Overall Loss 0.329712 Objective Loss 0.329712 LR 0.001000 Time 0.021644 -2023-02-13 17:42:25,149 - Epoch: [54][ 260/ 1207] Overall Loss 0.329878 Objective Loss 0.329878 LR 0.001000 Time 0.021534 -2023-02-13 17:42:25,338 - Epoch: [54][ 270/ 1207] Overall Loss 0.330188 Objective Loss 0.330188 LR 0.001000 Time 0.021434 -2023-02-13 17:42:25,526 - Epoch: [54][ 280/ 1207] Overall Loss 0.330386 Objective Loss 0.330386 LR 0.001000 Time 0.021340 -2023-02-13 17:42:25,716 - Epoch: [54][ 290/ 1207] Overall Loss 0.330531 Objective Loss 0.330531 LR 0.001000 Time 0.021255 -2023-02-13 17:42:25,906 - Epoch: [54][ 300/ 1207] Overall Loss 0.330688 Objective Loss 0.330688 LR 0.001000 Time 0.021179 -2023-02-13 17:42:26,094 - Epoch: [54][ 310/ 1207] Overall Loss 0.331098 Objective Loss 0.331098 LR 0.001000 Time 0.021101 -2023-02-13 17:42:26,281 - Epoch: [54][ 320/ 1207] Overall Loss 0.331297 Objective Loss 0.331297 LR 0.001000 Time 0.021027 -2023-02-13 17:42:26,470 - Epoch: [54][ 330/ 1207] Overall Loss 0.332167 Objective Loss 0.332167 LR 0.001000 Time 0.020960 -2023-02-13 17:42:26,659 - Epoch: [54][ 340/ 1207] Overall Loss 0.332441 Objective Loss 0.332441 LR 0.001000 Time 0.020898 -2023-02-13 17:42:26,848 - Epoch: [54][ 350/ 1207] Overall Loss 0.332491 Objective Loss 0.332491 LR 0.001000 Time 0.020840 -2023-02-13 17:42:27,037 - Epoch: [54][ 360/ 1207] Overall Loss 0.332193 Objective Loss 0.332193 LR 0.001000 Time 0.020784 -2023-02-13 17:42:27,225 - Epoch: [54][ 370/ 1207] Overall Loss 0.332058 Objective Loss 0.332058 LR 0.001000 Time 0.020730 -2023-02-13 17:42:27,414 - Epoch: [54][ 380/ 1207] Overall Loss 0.331525 Objective Loss 0.331525 LR 0.001000 Time 0.020681 -2023-02-13 17:42:27,603 - Epoch: [54][ 390/ 1207] Overall Loss 0.332316 Objective Loss 0.332316 LR 0.001000 Time 0.020635 -2023-02-13 17:42:27,791 - Epoch: [54][ 400/ 1207] Overall Loss 0.331877 Objective Loss 0.331877 LR 0.001000 Time 0.020589 -2023-02-13 17:42:27,981 - Epoch: [54][ 410/ 1207] Overall Loss 0.331971 Objective Loss 0.331971 LR 0.001000 Time 0.020549 -2023-02-13 17:42:28,169 - Epoch: [54][ 420/ 1207] Overall Loss 0.330889 Objective Loss 0.330889 LR 0.001000 Time 0.020507 -2023-02-13 17:42:28,358 - Epoch: [54][ 430/ 1207] Overall Loss 0.330536 Objective Loss 0.330536 LR 0.001000 Time 0.020469 -2023-02-13 17:42:28,548 - Epoch: [54][ 440/ 1207] Overall Loss 0.330370 Objective Loss 0.330370 LR 0.001000 Time 0.020434 -2023-02-13 17:42:28,737 - Epoch: [54][ 450/ 1207] Overall Loss 0.329905 Objective Loss 0.329905 LR 0.001000 Time 0.020400 -2023-02-13 17:42:28,927 - Epoch: [54][ 460/ 1207] Overall Loss 0.330119 Objective Loss 0.330119 LR 0.001000 Time 0.020367 -2023-02-13 17:42:29,116 - Epoch: [54][ 470/ 1207] Overall Loss 0.330335 Objective Loss 0.330335 LR 0.001000 Time 0.020335 -2023-02-13 17:42:29,304 - Epoch: [54][ 480/ 1207] Overall Loss 0.329967 Objective Loss 0.329967 LR 0.001000 Time 0.020304 -2023-02-13 17:42:29,494 - Epoch: [54][ 490/ 1207] Overall Loss 0.329472 Objective Loss 0.329472 LR 0.001000 Time 0.020275 -2023-02-13 17:42:29,683 - Epoch: [54][ 500/ 1207] Overall Loss 0.329324 Objective Loss 0.329324 LR 0.001000 Time 0.020248 -2023-02-13 17:42:29,873 - Epoch: [54][ 510/ 1207] Overall Loss 0.329936 Objective Loss 0.329936 LR 0.001000 Time 0.020221 -2023-02-13 17:42:30,062 - Epoch: [54][ 520/ 1207] Overall Loss 0.330203 Objective Loss 0.330203 LR 0.001000 Time 0.020195 -2023-02-13 17:42:30,251 - Epoch: [54][ 530/ 1207] Overall Loss 0.329716 Objective Loss 0.329716 LR 0.001000 Time 0.020170 -2023-02-13 17:42:30,440 - Epoch: [54][ 540/ 1207] Overall Loss 0.329840 Objective Loss 0.329840 LR 0.001000 Time 0.020146 -2023-02-13 17:42:30,630 - Epoch: [54][ 550/ 1207] Overall Loss 0.329905 Objective Loss 0.329905 LR 0.001000 Time 0.020124 -2023-02-13 17:42:30,819 - Epoch: [54][ 560/ 1207] Overall Loss 0.329629 Objective Loss 0.329629 LR 0.001000 Time 0.020103 -2023-02-13 17:42:31,009 - Epoch: [54][ 570/ 1207] Overall Loss 0.329476 Objective Loss 0.329476 LR 0.001000 Time 0.020082 -2023-02-13 17:42:31,197 - Epoch: [54][ 580/ 1207] Overall Loss 0.329660 Objective Loss 0.329660 LR 0.001000 Time 0.020060 -2023-02-13 17:42:31,386 - Epoch: [54][ 590/ 1207] Overall Loss 0.329686 Objective Loss 0.329686 LR 0.001000 Time 0.020040 -2023-02-13 17:42:31,575 - Epoch: [54][ 600/ 1207] Overall Loss 0.330508 Objective Loss 0.330508 LR 0.001000 Time 0.020020 -2023-02-13 17:42:31,765 - Epoch: [54][ 610/ 1207] Overall Loss 0.330371 Objective Loss 0.330371 LR 0.001000 Time 0.020003 -2023-02-13 17:42:31,955 - Epoch: [54][ 620/ 1207] Overall Loss 0.330513 Objective Loss 0.330513 LR 0.001000 Time 0.019986 -2023-02-13 17:42:32,144 - Epoch: [54][ 630/ 1207] Overall Loss 0.330405 Objective Loss 0.330405 LR 0.001000 Time 0.019968 -2023-02-13 17:42:32,333 - Epoch: [54][ 640/ 1207] Overall Loss 0.331082 Objective Loss 0.331082 LR 0.001000 Time 0.019950 -2023-02-13 17:42:32,523 - Epoch: [54][ 650/ 1207] Overall Loss 0.331570 Objective Loss 0.331570 LR 0.001000 Time 0.019935 -2023-02-13 17:42:32,711 - Epoch: [54][ 660/ 1207] Overall Loss 0.331930 Objective Loss 0.331930 LR 0.001000 Time 0.019918 -2023-02-13 17:42:32,902 - Epoch: [54][ 670/ 1207] Overall Loss 0.331635 Objective Loss 0.331635 LR 0.001000 Time 0.019904 -2023-02-13 17:42:33,091 - Epoch: [54][ 680/ 1207] Overall Loss 0.332272 Objective Loss 0.332272 LR 0.001000 Time 0.019890 -2023-02-13 17:42:33,281 - Epoch: [54][ 690/ 1207] Overall Loss 0.331994 Objective Loss 0.331994 LR 0.001000 Time 0.019876 -2023-02-13 17:42:33,470 - Epoch: [54][ 700/ 1207] Overall Loss 0.332522 Objective Loss 0.332522 LR 0.001000 Time 0.019862 -2023-02-13 17:42:33,659 - Epoch: [54][ 710/ 1207] Overall Loss 0.332366 Objective Loss 0.332366 LR 0.001000 Time 0.019848 -2023-02-13 17:42:33,849 - Epoch: [54][ 720/ 1207] Overall Loss 0.332703 Objective Loss 0.332703 LR 0.001000 Time 0.019835 -2023-02-13 17:42:34,045 - Epoch: [54][ 730/ 1207] Overall Loss 0.332648 Objective Loss 0.332648 LR 0.001000 Time 0.019831 -2023-02-13 17:42:34,237 - Epoch: [54][ 740/ 1207] Overall Loss 0.333238 Objective Loss 0.333238 LR 0.001000 Time 0.019823 -2023-02-13 17:42:34,433 - Epoch: [54][ 750/ 1207] Overall Loss 0.333697 Objective Loss 0.333697 LR 0.001000 Time 0.019819 -2023-02-13 17:42:34,626 - Epoch: [54][ 760/ 1207] Overall Loss 0.333573 Objective Loss 0.333573 LR 0.001000 Time 0.019812 -2023-02-13 17:42:34,822 - Epoch: [54][ 770/ 1207] Overall Loss 0.333875 Objective Loss 0.333875 LR 0.001000 Time 0.019809 -2023-02-13 17:42:35,016 - Epoch: [54][ 780/ 1207] Overall Loss 0.334271 Objective Loss 0.334271 LR 0.001000 Time 0.019803 -2023-02-13 17:42:35,212 - Epoch: [54][ 790/ 1207] Overall Loss 0.334135 Objective Loss 0.334135 LR 0.001000 Time 0.019800 -2023-02-13 17:42:35,406 - Epoch: [54][ 800/ 1207] Overall Loss 0.333903 Objective Loss 0.333903 LR 0.001000 Time 0.019794 -2023-02-13 17:42:35,602 - Epoch: [54][ 810/ 1207] Overall Loss 0.333558 Objective Loss 0.333558 LR 0.001000 Time 0.019791 -2023-02-13 17:42:35,796 - Epoch: [54][ 820/ 1207] Overall Loss 0.333590 Objective Loss 0.333590 LR 0.001000 Time 0.019786 -2023-02-13 17:42:35,993 - Epoch: [54][ 830/ 1207] Overall Loss 0.333282 Objective Loss 0.333282 LR 0.001000 Time 0.019784 -2023-02-13 17:42:36,186 - Epoch: [54][ 840/ 1207] Overall Loss 0.333248 Objective Loss 0.333248 LR 0.001000 Time 0.019779 -2023-02-13 17:42:36,382 - Epoch: [54][ 850/ 1207] Overall Loss 0.333489 Objective Loss 0.333489 LR 0.001000 Time 0.019776 -2023-02-13 17:42:36,576 - Epoch: [54][ 860/ 1207] Overall Loss 0.333379 Objective Loss 0.333379 LR 0.001000 Time 0.019771 -2023-02-13 17:42:36,772 - Epoch: [54][ 870/ 1207] Overall Loss 0.333536 Objective Loss 0.333536 LR 0.001000 Time 0.019769 -2023-02-13 17:42:36,966 - Epoch: [54][ 880/ 1207] Overall Loss 0.333849 Objective Loss 0.333849 LR 0.001000 Time 0.019764 -2023-02-13 17:42:37,162 - Epoch: [54][ 890/ 1207] Overall Loss 0.333773 Objective Loss 0.333773 LR 0.001000 Time 0.019762 -2023-02-13 17:42:37,356 - Epoch: [54][ 900/ 1207] Overall Loss 0.333487 Objective Loss 0.333487 LR 0.001000 Time 0.019757 -2023-02-13 17:42:37,553 - Epoch: [54][ 910/ 1207] Overall Loss 0.333543 Objective Loss 0.333543 LR 0.001000 Time 0.019756 -2023-02-13 17:42:37,747 - Epoch: [54][ 920/ 1207] Overall Loss 0.333828 Objective Loss 0.333828 LR 0.001000 Time 0.019752 -2023-02-13 17:42:37,943 - Epoch: [54][ 930/ 1207] Overall Loss 0.334100 Objective Loss 0.334100 LR 0.001000 Time 0.019750 -2023-02-13 17:42:38,136 - Epoch: [54][ 940/ 1207] Overall Loss 0.334323 Objective Loss 0.334323 LR 0.001000 Time 0.019745 -2023-02-13 17:42:38,332 - Epoch: [54][ 950/ 1207] Overall Loss 0.334670 Objective Loss 0.334670 LR 0.001000 Time 0.019743 -2023-02-13 17:42:38,526 - Epoch: [54][ 960/ 1207] Overall Loss 0.334630 Objective Loss 0.334630 LR 0.001000 Time 0.019739 -2023-02-13 17:42:38,722 - Epoch: [54][ 970/ 1207] Overall Loss 0.334913 Objective Loss 0.334913 LR 0.001000 Time 0.019737 -2023-02-13 17:42:38,916 - Epoch: [54][ 980/ 1207] Overall Loss 0.336167 Objective Loss 0.336167 LR 0.001000 Time 0.019733 -2023-02-13 17:42:39,113 - Epoch: [54][ 990/ 1207] Overall Loss 0.337790 Objective Loss 0.337790 LR 0.001000 Time 0.019732 -2023-02-13 17:42:39,306 - Epoch: [54][ 1000/ 1207] Overall Loss 0.338939 Objective Loss 0.338939 LR 0.001000 Time 0.019728 -2023-02-13 17:42:39,502 - Epoch: [54][ 1010/ 1207] Overall Loss 0.340274 Objective Loss 0.340274 LR 0.001000 Time 0.019727 -2023-02-13 17:42:39,696 - Epoch: [54][ 1020/ 1207] Overall Loss 0.341102 Objective Loss 0.341102 LR 0.001000 Time 0.019722 -2023-02-13 17:42:39,892 - Epoch: [54][ 1030/ 1207] Overall Loss 0.341936 Objective Loss 0.341936 LR 0.001000 Time 0.019721 -2023-02-13 17:42:40,085 - Epoch: [54][ 1040/ 1207] Overall Loss 0.342833 Objective Loss 0.342833 LR 0.001000 Time 0.019717 -2023-02-13 17:42:40,281 - Epoch: [54][ 1050/ 1207] Overall Loss 0.343680 Objective Loss 0.343680 LR 0.001000 Time 0.019715 -2023-02-13 17:42:40,475 - Epoch: [54][ 1060/ 1207] Overall Loss 0.344587 Objective Loss 0.344587 LR 0.001000 Time 0.019711 -2023-02-13 17:42:40,670 - Epoch: [54][ 1070/ 1207] Overall Loss 0.345176 Objective Loss 0.345176 LR 0.001000 Time 0.019710 -2023-02-13 17:42:40,864 - Epoch: [54][ 1080/ 1207] Overall Loss 0.345738 Objective Loss 0.345738 LR 0.001000 Time 0.019706 -2023-02-13 17:42:41,061 - Epoch: [54][ 1090/ 1207] Overall Loss 0.346158 Objective Loss 0.346158 LR 0.001000 Time 0.019706 -2023-02-13 17:42:41,255 - Epoch: [54][ 1100/ 1207] Overall Loss 0.346890 Objective Loss 0.346890 LR 0.001000 Time 0.019703 -2023-02-13 17:42:41,451 - Epoch: [54][ 1110/ 1207] Overall Loss 0.347567 Objective Loss 0.347567 LR 0.001000 Time 0.019702 -2023-02-13 17:42:41,646 - Epoch: [54][ 1120/ 1207] Overall Loss 0.348132 Objective Loss 0.348132 LR 0.001000 Time 0.019699 -2023-02-13 17:42:41,842 - Epoch: [54][ 1130/ 1207] Overall Loss 0.348684 Objective Loss 0.348684 LR 0.001000 Time 0.019698 -2023-02-13 17:42:42,036 - Epoch: [54][ 1140/ 1207] Overall Loss 0.349049 Objective Loss 0.349049 LR 0.001000 Time 0.019695 -2023-02-13 17:42:42,231 - Epoch: [54][ 1150/ 1207] Overall Loss 0.349545 Objective Loss 0.349545 LR 0.001000 Time 0.019693 -2023-02-13 17:42:42,425 - Epoch: [54][ 1160/ 1207] Overall Loss 0.349981 Objective Loss 0.349981 LR 0.001000 Time 0.019690 -2023-02-13 17:42:42,621 - Epoch: [54][ 1170/ 1207] Overall Loss 0.350483 Objective Loss 0.350483 LR 0.001000 Time 0.019689 -2023-02-13 17:42:42,815 - Epoch: [54][ 1180/ 1207] Overall Loss 0.350647 Objective Loss 0.350647 LR 0.001000 Time 0.019686 -2023-02-13 17:42:43,011 - Epoch: [54][ 1190/ 1207] Overall Loss 0.350885 Objective Loss 0.350885 LR 0.001000 Time 0.019686 -2023-02-13 17:42:43,255 - Epoch: [54][ 1200/ 1207] Overall Loss 0.351186 Objective Loss 0.351186 LR 0.001000 Time 0.019725 -2023-02-13 17:42:43,370 - Epoch: [54][ 1207/ 1207] Overall Loss 0.351537 Objective Loss 0.351537 Top1 81.097561 Top5 98.475610 LR 0.001000 Time 0.019705 -2023-02-13 17:42:43,448 - --- validate (epoch=54)----------- -2023-02-13 17:42:43,449 - 34311 samples (256 per mini-batch) -2023-02-13 17:42:43,839 - Epoch: [54][ 10/ 135] Loss 0.411194 Top1 80.234375 Top5 96.484375 -2023-02-13 17:42:43,965 - Epoch: [54][ 20/ 135] Loss 0.419323 Top1 79.746094 Top5 96.503906 -2023-02-13 17:42:44,092 - Epoch: [54][ 30/ 135] Loss 0.410328 Top1 80.000000 Top5 96.458333 -2023-02-13 17:42:44,221 - Epoch: [54][ 40/ 135] Loss 0.411056 Top1 79.765625 Top5 96.562500 -2023-02-13 17:42:44,351 - Epoch: [54][ 50/ 135] Loss 0.417894 Top1 79.539062 Top5 96.609375 -2023-02-13 17:42:44,478 - Epoch: [54][ 60/ 135] Loss 0.412954 Top1 79.492188 Top5 96.640625 -2023-02-13 17:42:44,605 - Epoch: [54][ 70/ 135] Loss 0.415733 Top1 79.441964 Top5 96.679688 -2023-02-13 17:42:44,730 - Epoch: [54][ 80/ 135] Loss 0.416606 Top1 79.404297 Top5 96.713867 -2023-02-13 17:42:44,856 - Epoch: [54][ 90/ 135] Loss 0.414567 Top1 79.470486 Top5 96.744792 -2023-02-13 17:42:44,982 - Epoch: [54][ 100/ 135] Loss 0.415741 Top1 79.476562 Top5 96.703125 -2023-02-13 17:42:45,110 - Epoch: [54][ 110/ 135] Loss 0.414853 Top1 79.630682 Top5 96.700994 -2023-02-13 17:42:45,239 - Epoch: [54][ 120/ 135] Loss 0.414812 Top1 79.554036 Top5 96.731771 -2023-02-13 17:42:45,369 - Epoch: [54][ 130/ 135] Loss 0.413674 Top1 79.648438 Top5 96.748798 -2023-02-13 17:42:45,416 - Epoch: [54][ 135/ 135] Loss 0.411402 Top1 79.735362 Top5 96.796946 -2023-02-13 17:42:45,489 - ==> Top1: 79.735 Top5: 96.797 Loss: 0.411 - -2023-02-13 17:42:45,490 - ==> Confusion: -[[ 857 3 3 1 15 2 0 5 2 45 0 3 1 5 4 6 5 1 1 0 8] - [ 2 942 3 2 5 33 3 15 4 1 0 2 1 0 7 2 2 2 6 0 1] - [ 13 6 946 13 1 4 5 20 2 1 5 2 0 3 3 9 1 4 11 4 5] - [ 6 2 26 889 3 4 0 3 2 2 12 2 8 0 22 2 4 7 15 0 7] - [ 18 10 0 1 986 12 1 1 0 5 0 7 3 4 4 3 4 3 0 1 3] - [ 3 32 2 7 5 959 1 9 2 2 2 14 1 15 4 2 5 0 2 1 2] - [ 4 4 21 2 1 8 1016 8 0 1 4 3 4 1 1 5 4 3 0 8 1] - [ 3 16 9 2 2 32 3 905 1 1 4 8 1 1 0 1 1 2 22 8 2] - [ 20 2 0 1 2 2 0 2 878 39 13 1 1 9 29 2 0 3 4 1 0] - [ 104 2 3 0 8 4 0 1 31 821 0 1 1 19 6 2 0 1 1 1 6] - [ 2 5 4 7 1 2 2 6 16 0 959 5 3 14 3 1 1 0 18 0 2] - [ 3 3 0 0 2 10 1 4 3 1 0 924 23 7 2 7 2 9 1 3 0] - [ 1 0 0 6 0 3 0 0 3 0 0 51 849 2 4 8 4 20 3 2 3] - [ 8 3 2 0 6 14 0 1 11 17 9 10 3 912 8 9 5 1 2 2 1] - [ 12 3 2 18 8 0 0 2 13 7 4 4 4 1 990 1 0 5 12 0 6] - [ 7 2 5 0 7 3 5 1 0 0 0 10 5 1 0 965 8 17 0 4 6] - [ 3 8 0 2 9 4 0 0 4 0 2 3 2 2 3 11 994 3 1 1 9] - [ 9 3 0 4 0 3 0 1 1 0 0 15 13 0 1 13 0 982 1 1 4] - [ 6 6 7 11 1 0 0 25 3 1 5 3 5 1 14 1 1 2 993 0 1] - [ 1 4 1 1 2 6 7 19 1 0 3 29 2 3 0 9 5 3 2 1045 5] - [ 243 424 237 169 168 294 76 221 128 87 261 214 446 398 258 158 359 135 280 331 8547]] - -2023-02-13 17:42:45,491 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:42:45,491 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:42:45,497 - - -2023-02-13 17:42:45,497 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:42:46,459 - Epoch: [55][ 10/ 1207] Overall Loss 0.375989 Objective Loss 0.375989 LR 0.001000 Time 0.096150 -2023-02-13 17:42:46,650 - Epoch: [55][ 20/ 1207] Overall Loss 0.377382 Objective Loss 0.377382 LR 0.001000 Time 0.057605 -2023-02-13 17:42:46,840 - Epoch: [55][ 30/ 1207] Overall Loss 0.370434 Objective Loss 0.370434 LR 0.001000 Time 0.044725 -2023-02-13 17:42:47,030 - Epoch: [55][ 40/ 1207] Overall Loss 0.377194 Objective Loss 0.377194 LR 0.001000 Time 0.038283 -2023-02-13 17:42:47,219 - Epoch: [55][ 50/ 1207] Overall Loss 0.378135 Objective Loss 0.378135 LR 0.001000 Time 0.034394 -2023-02-13 17:42:47,408 - Epoch: [55][ 60/ 1207] Overall Loss 0.379442 Objective Loss 0.379442 LR 0.001000 Time 0.031801 -2023-02-13 17:42:47,599 - Epoch: [55][ 70/ 1207] Overall Loss 0.377821 Objective Loss 0.377821 LR 0.001000 Time 0.029983 -2023-02-13 17:42:47,787 - Epoch: [55][ 80/ 1207] Overall Loss 0.379217 Objective Loss 0.379217 LR 0.001000 Time 0.028587 -2023-02-13 17:42:47,977 - Epoch: [55][ 90/ 1207] Overall Loss 0.377859 Objective Loss 0.377859 LR 0.001000 Time 0.027514 -2023-02-13 17:42:48,167 - Epoch: [55][ 100/ 1207] Overall Loss 0.377373 Objective Loss 0.377373 LR 0.001000 Time 0.026659 -2023-02-13 17:42:48,357 - Epoch: [55][ 110/ 1207] Overall Loss 0.378840 Objective Loss 0.378840 LR 0.001000 Time 0.025958 -2023-02-13 17:42:48,546 - Epoch: [55][ 120/ 1207] Overall Loss 0.377688 Objective Loss 0.377688 LR 0.001000 Time 0.025372 -2023-02-13 17:42:48,735 - Epoch: [55][ 130/ 1207] Overall Loss 0.378484 Objective Loss 0.378484 LR 0.001000 Time 0.024868 -2023-02-13 17:42:48,924 - Epoch: [55][ 140/ 1207] Overall Loss 0.376470 Objective Loss 0.376470 LR 0.001000 Time 0.024438 -2023-02-13 17:42:49,113 - Epoch: [55][ 150/ 1207] Overall Loss 0.376279 Objective Loss 0.376279 LR 0.001000 Time 0.024067 -2023-02-13 17:42:49,302 - Epoch: [55][ 160/ 1207] Overall Loss 0.377357 Objective Loss 0.377357 LR 0.001000 Time 0.023741 -2023-02-13 17:42:49,491 - Epoch: [55][ 170/ 1207] Overall Loss 0.377079 Objective Loss 0.377079 LR 0.001000 Time 0.023453 -2023-02-13 17:42:49,680 - Epoch: [55][ 180/ 1207] Overall Loss 0.377107 Objective Loss 0.377107 LR 0.001000 Time 0.023198 -2023-02-13 17:42:49,869 - Epoch: [55][ 190/ 1207] Overall Loss 0.378085 Objective Loss 0.378085 LR 0.001000 Time 0.022971 -2023-02-13 17:42:50,059 - Epoch: [55][ 200/ 1207] Overall Loss 0.376815 Objective Loss 0.376815 LR 0.001000 Time 0.022770 -2023-02-13 17:42:50,248 - Epoch: [55][ 210/ 1207] Overall Loss 0.376544 Objective Loss 0.376544 LR 0.001000 Time 0.022585 -2023-02-13 17:42:50,437 - Epoch: [55][ 220/ 1207] Overall Loss 0.375972 Objective Loss 0.375972 LR 0.001000 Time 0.022417 -2023-02-13 17:42:50,628 - Epoch: [55][ 230/ 1207] Overall Loss 0.376067 Objective Loss 0.376067 LR 0.001000 Time 0.022271 -2023-02-13 17:42:50,819 - Epoch: [55][ 240/ 1207] Overall Loss 0.376171 Objective Loss 0.376171 LR 0.001000 Time 0.022135 -2023-02-13 17:42:51,009 - Epoch: [55][ 250/ 1207] Overall Loss 0.375469 Objective Loss 0.375469 LR 0.001000 Time 0.022009 -2023-02-13 17:42:51,199 - Epoch: [55][ 260/ 1207] Overall Loss 0.374323 Objective Loss 0.374323 LR 0.001000 Time 0.021893 -2023-02-13 17:42:51,388 - Epoch: [55][ 270/ 1207] Overall Loss 0.373976 Objective Loss 0.373976 LR 0.001000 Time 0.021780 -2023-02-13 17:42:51,577 - Epoch: [55][ 280/ 1207] Overall Loss 0.373539 Objective Loss 0.373539 LR 0.001000 Time 0.021675 -2023-02-13 17:42:51,767 - Epoch: [55][ 290/ 1207] Overall Loss 0.373668 Objective Loss 0.373668 LR 0.001000 Time 0.021581 -2023-02-13 17:42:51,957 - Epoch: [55][ 300/ 1207] Overall Loss 0.372980 Objective Loss 0.372980 LR 0.001000 Time 0.021494 -2023-02-13 17:42:52,146 - Epoch: [55][ 310/ 1207] Overall Loss 0.373418 Objective Loss 0.373418 LR 0.001000 Time 0.021411 -2023-02-13 17:42:52,335 - Epoch: [55][ 320/ 1207] Overall Loss 0.373097 Objective Loss 0.373097 LR 0.001000 Time 0.021330 -2023-02-13 17:42:52,524 - Epoch: [55][ 330/ 1207] Overall Loss 0.372369 Objective Loss 0.372369 LR 0.001000 Time 0.021255 -2023-02-13 17:42:52,713 - Epoch: [55][ 340/ 1207] Overall Loss 0.372410 Objective Loss 0.372410 LR 0.001000 Time 0.021187 -2023-02-13 17:42:52,903 - Epoch: [55][ 350/ 1207] Overall Loss 0.371890 Objective Loss 0.371890 LR 0.001000 Time 0.021121 -2023-02-13 17:42:53,093 - Epoch: [55][ 360/ 1207] Overall Loss 0.371431 Objective Loss 0.371431 LR 0.001000 Time 0.021063 -2023-02-13 17:42:53,285 - Epoch: [55][ 370/ 1207] Overall Loss 0.370486 Objective Loss 0.370486 LR 0.001000 Time 0.021010 -2023-02-13 17:42:53,476 - Epoch: [55][ 380/ 1207] Overall Loss 0.369671 Objective Loss 0.369671 LR 0.001000 Time 0.020960 -2023-02-13 17:42:53,669 - Epoch: [55][ 390/ 1207] Overall Loss 0.369728 Objective Loss 0.369728 LR 0.001000 Time 0.020915 -2023-02-13 17:42:53,859 - Epoch: [55][ 400/ 1207] Overall Loss 0.370099 Objective Loss 0.370099 LR 0.001000 Time 0.020866 -2023-02-13 17:42:54,050 - Epoch: [55][ 410/ 1207] Overall Loss 0.370198 Objective Loss 0.370198 LR 0.001000 Time 0.020823 -2023-02-13 17:42:54,240 - Epoch: [55][ 420/ 1207] Overall Loss 0.370946 Objective Loss 0.370946 LR 0.001000 Time 0.020778 -2023-02-13 17:42:54,430 - Epoch: [55][ 430/ 1207] Overall Loss 0.370313 Objective Loss 0.370313 LR 0.001000 Time 0.020735 -2023-02-13 17:42:54,620 - Epoch: [55][ 440/ 1207] Overall Loss 0.370343 Objective Loss 0.370343 LR 0.001000 Time 0.020695 -2023-02-13 17:42:54,810 - Epoch: [55][ 450/ 1207] Overall Loss 0.369554 Objective Loss 0.369554 LR 0.001000 Time 0.020657 -2023-02-13 17:42:55,000 - Epoch: [55][ 460/ 1207] Overall Loss 0.369558 Objective Loss 0.369558 LR 0.001000 Time 0.020620 -2023-02-13 17:42:55,189 - Epoch: [55][ 470/ 1207] Overall Loss 0.369008 Objective Loss 0.369008 LR 0.001000 Time 0.020584 -2023-02-13 17:42:55,379 - Epoch: [55][ 480/ 1207] Overall Loss 0.368808 Objective Loss 0.368808 LR 0.001000 Time 0.020549 -2023-02-13 17:42:55,568 - Epoch: [55][ 490/ 1207] Overall Loss 0.368255 Objective Loss 0.368255 LR 0.001000 Time 0.020515 -2023-02-13 17:42:55,758 - Epoch: [55][ 500/ 1207] Overall Loss 0.367909 Objective Loss 0.367909 LR 0.001000 Time 0.020483 -2023-02-13 17:42:55,949 - Epoch: [55][ 510/ 1207] Overall Loss 0.368362 Objective Loss 0.368362 LR 0.001000 Time 0.020455 -2023-02-13 17:42:56,139 - Epoch: [55][ 520/ 1207] Overall Loss 0.367783 Objective Loss 0.367783 LR 0.001000 Time 0.020426 -2023-02-13 17:42:56,328 - Epoch: [55][ 530/ 1207] Overall Loss 0.368602 Objective Loss 0.368602 LR 0.001000 Time 0.020398 -2023-02-13 17:42:56,519 - Epoch: [55][ 540/ 1207] Overall Loss 0.368395 Objective Loss 0.368395 LR 0.001000 Time 0.020373 -2023-02-13 17:42:56,709 - Epoch: [55][ 550/ 1207] Overall Loss 0.367734 Objective Loss 0.367734 LR 0.001000 Time 0.020347 -2023-02-13 17:42:56,900 - Epoch: [55][ 560/ 1207] Overall Loss 0.367600 Objective Loss 0.367600 LR 0.001000 Time 0.020324 -2023-02-13 17:42:57,091 - Epoch: [55][ 570/ 1207] Overall Loss 0.367372 Objective Loss 0.367372 LR 0.001000 Time 0.020303 -2023-02-13 17:42:57,281 - Epoch: [55][ 580/ 1207] Overall Loss 0.367458 Objective Loss 0.367458 LR 0.001000 Time 0.020280 -2023-02-13 17:42:57,471 - Epoch: [55][ 590/ 1207] Overall Loss 0.367334 Objective Loss 0.367334 LR 0.001000 Time 0.020257 -2023-02-13 17:42:57,661 - Epoch: [55][ 600/ 1207] Overall Loss 0.367539 Objective Loss 0.367539 LR 0.001000 Time 0.020235 -2023-02-13 17:42:57,850 - Epoch: [55][ 610/ 1207] Overall Loss 0.367181 Objective Loss 0.367181 LR 0.001000 Time 0.020213 -2023-02-13 17:42:58,040 - Epoch: [55][ 620/ 1207] Overall Loss 0.367100 Objective Loss 0.367100 LR 0.001000 Time 0.020193 -2023-02-13 17:42:58,230 - Epoch: [55][ 630/ 1207] Overall Loss 0.367712 Objective Loss 0.367712 LR 0.001000 Time 0.020173 -2023-02-13 17:42:58,419 - Epoch: [55][ 640/ 1207] Overall Loss 0.368127 Objective Loss 0.368127 LR 0.001000 Time 0.020153 -2023-02-13 17:42:58,609 - Epoch: [55][ 650/ 1207] Overall Loss 0.367893 Objective Loss 0.367893 LR 0.001000 Time 0.020134 -2023-02-13 17:42:58,799 - Epoch: [55][ 660/ 1207] Overall Loss 0.367653 Objective Loss 0.367653 LR 0.001000 Time 0.020116 -2023-02-13 17:42:58,990 - Epoch: [55][ 670/ 1207] Overall Loss 0.367917 Objective Loss 0.367917 LR 0.001000 Time 0.020100 -2023-02-13 17:42:59,180 - Epoch: [55][ 680/ 1207] Overall Loss 0.367992 Objective Loss 0.367992 LR 0.001000 Time 0.020084 -2023-02-13 17:42:59,370 - Epoch: [55][ 690/ 1207] Overall Loss 0.368208 Objective Loss 0.368208 LR 0.001000 Time 0.020068 -2023-02-13 17:42:59,565 - Epoch: [55][ 700/ 1207] Overall Loss 0.368326 Objective Loss 0.368326 LR 0.001000 Time 0.020058 -2023-02-13 17:42:59,757 - Epoch: [55][ 710/ 1207] Overall Loss 0.368396 Objective Loss 0.368396 LR 0.001000 Time 0.020046 -2023-02-13 17:42:59,947 - Epoch: [55][ 720/ 1207] Overall Loss 0.368343 Objective Loss 0.368343 LR 0.001000 Time 0.020032 -2023-02-13 17:43:00,137 - Epoch: [55][ 730/ 1207] Overall Loss 0.368146 Objective Loss 0.368146 LR 0.001000 Time 0.020017 -2023-02-13 17:43:00,327 - Epoch: [55][ 740/ 1207] Overall Loss 0.367915 Objective Loss 0.367915 LR 0.001000 Time 0.020002 -2023-02-13 17:43:00,516 - Epoch: [55][ 750/ 1207] Overall Loss 0.367816 Objective Loss 0.367816 LR 0.001000 Time 0.019987 -2023-02-13 17:43:00,706 - Epoch: [55][ 760/ 1207] Overall Loss 0.367955 Objective Loss 0.367955 LR 0.001000 Time 0.019974 -2023-02-13 17:43:00,897 - Epoch: [55][ 770/ 1207] Overall Loss 0.368463 Objective Loss 0.368463 LR 0.001000 Time 0.019962 -2023-02-13 17:43:01,087 - Epoch: [55][ 780/ 1207] Overall Loss 0.368224 Objective Loss 0.368224 LR 0.001000 Time 0.019949 -2023-02-13 17:43:01,277 - Epoch: [55][ 790/ 1207] Overall Loss 0.368249 Objective Loss 0.368249 LR 0.001000 Time 0.019936 -2023-02-13 17:43:01,466 - Epoch: [55][ 800/ 1207] Overall Loss 0.368243 Objective Loss 0.368243 LR 0.001000 Time 0.019923 -2023-02-13 17:43:01,656 - Epoch: [55][ 810/ 1207] Overall Loss 0.368021 Objective Loss 0.368021 LR 0.001000 Time 0.019911 -2023-02-13 17:43:01,845 - Epoch: [55][ 820/ 1207] Overall Loss 0.368108 Objective Loss 0.368108 LR 0.001000 Time 0.019899 -2023-02-13 17:43:02,036 - Epoch: [55][ 830/ 1207] Overall Loss 0.368271 Objective Loss 0.368271 LR 0.001000 Time 0.019888 -2023-02-13 17:43:02,225 - Epoch: [55][ 840/ 1207] Overall Loss 0.368326 Objective Loss 0.368326 LR 0.001000 Time 0.019876 -2023-02-13 17:43:02,415 - Epoch: [55][ 850/ 1207] Overall Loss 0.368483 Objective Loss 0.368483 LR 0.001000 Time 0.019865 -2023-02-13 17:43:02,605 - Epoch: [55][ 860/ 1207] Overall Loss 0.368438 Objective Loss 0.368438 LR 0.001000 Time 0.019854 -2023-02-13 17:43:02,794 - Epoch: [55][ 870/ 1207] Overall Loss 0.368476 Objective Loss 0.368476 LR 0.001000 Time 0.019844 -2023-02-13 17:43:02,985 - Epoch: [55][ 880/ 1207] Overall Loss 0.368282 Objective Loss 0.368282 LR 0.001000 Time 0.019834 -2023-02-13 17:43:03,175 - Epoch: [55][ 890/ 1207] Overall Loss 0.368033 Objective Loss 0.368033 LR 0.001000 Time 0.019824 -2023-02-13 17:43:03,364 - Epoch: [55][ 900/ 1207] Overall Loss 0.368044 Objective Loss 0.368044 LR 0.001000 Time 0.019814 -2023-02-13 17:43:03,556 - Epoch: [55][ 910/ 1207] Overall Loss 0.368005 Objective Loss 0.368005 LR 0.001000 Time 0.019807 -2023-02-13 17:43:03,749 - Epoch: [55][ 920/ 1207] Overall Loss 0.367921 Objective Loss 0.367921 LR 0.001000 Time 0.019801 -2023-02-13 17:43:03,942 - Epoch: [55][ 930/ 1207] Overall Loss 0.367695 Objective Loss 0.367695 LR 0.001000 Time 0.019795 -2023-02-13 17:43:04,134 - Epoch: [55][ 940/ 1207] Overall Loss 0.367847 Objective Loss 0.367847 LR 0.001000 Time 0.019789 -2023-02-13 17:43:04,327 - Epoch: [55][ 950/ 1207] Overall Loss 0.367798 Objective Loss 0.367798 LR 0.001000 Time 0.019783 -2023-02-13 17:43:04,519 - Epoch: [55][ 960/ 1207] Overall Loss 0.367647 Objective Loss 0.367647 LR 0.001000 Time 0.019776 -2023-02-13 17:43:04,712 - Epoch: [55][ 970/ 1207] Overall Loss 0.367386 Objective Loss 0.367386 LR 0.001000 Time 0.019771 -2023-02-13 17:43:04,904 - Epoch: [55][ 980/ 1207] Overall Loss 0.366866 Objective Loss 0.366866 LR 0.001000 Time 0.019765 -2023-02-13 17:43:05,096 - Epoch: [55][ 990/ 1207] Overall Loss 0.366875 Objective Loss 0.366875 LR 0.001000 Time 0.019759 -2023-02-13 17:43:05,289 - Epoch: [55][ 1000/ 1207] Overall Loss 0.366829 Objective Loss 0.366829 LR 0.001000 Time 0.019753 -2023-02-13 17:43:05,481 - Epoch: [55][ 1010/ 1207] Overall Loss 0.367007 Objective Loss 0.367007 LR 0.001000 Time 0.019748 -2023-02-13 17:43:05,673 - Epoch: [55][ 1020/ 1207] Overall Loss 0.367264 Objective Loss 0.367264 LR 0.001000 Time 0.019742 -2023-02-13 17:43:05,867 - Epoch: [55][ 1030/ 1207] Overall Loss 0.367099 Objective Loss 0.367099 LR 0.001000 Time 0.019738 -2023-02-13 17:43:06,061 - Epoch: [55][ 1040/ 1207] Overall Loss 0.367108 Objective Loss 0.367108 LR 0.001000 Time 0.019734 -2023-02-13 17:43:06,253 - Epoch: [55][ 1050/ 1207] Overall Loss 0.366922 Objective Loss 0.366922 LR 0.001000 Time 0.019729 -2023-02-13 17:43:06,445 - Epoch: [55][ 1060/ 1207] Overall Loss 0.366916 Objective Loss 0.366916 LR 0.001000 Time 0.019724 -2023-02-13 17:43:06,637 - Epoch: [55][ 1070/ 1207] Overall Loss 0.366814 Objective Loss 0.366814 LR 0.001000 Time 0.019719 -2023-02-13 17:43:06,829 - Epoch: [55][ 1080/ 1207] Overall Loss 0.366907 Objective Loss 0.366907 LR 0.001000 Time 0.019714 -2023-02-13 17:43:07,022 - Epoch: [55][ 1090/ 1207] Overall Loss 0.366716 Objective Loss 0.366716 LR 0.001000 Time 0.019710 -2023-02-13 17:43:07,211 - Epoch: [55][ 1100/ 1207] Overall Loss 0.366546 Objective Loss 0.366546 LR 0.001000 Time 0.019702 -2023-02-13 17:43:07,401 - Epoch: [55][ 1110/ 1207] Overall Loss 0.366521 Objective Loss 0.366521 LR 0.001000 Time 0.019695 -2023-02-13 17:43:07,591 - Epoch: [55][ 1120/ 1207] Overall Loss 0.366600 Objective Loss 0.366600 LR 0.001000 Time 0.019689 -2023-02-13 17:43:07,784 - Epoch: [55][ 1130/ 1207] Overall Loss 0.366625 Objective Loss 0.366625 LR 0.001000 Time 0.019684 -2023-02-13 17:43:07,974 - Epoch: [55][ 1140/ 1207] Overall Loss 0.366591 Objective Loss 0.366591 LR 0.001000 Time 0.019678 -2023-02-13 17:43:08,165 - Epoch: [55][ 1150/ 1207] Overall Loss 0.366847 Objective Loss 0.366847 LR 0.001000 Time 0.019673 -2023-02-13 17:43:08,356 - Epoch: [55][ 1160/ 1207] Overall Loss 0.367111 Objective Loss 0.367111 LR 0.001000 Time 0.019668 -2023-02-13 17:43:08,547 - Epoch: [55][ 1170/ 1207] Overall Loss 0.367009 Objective Loss 0.367009 LR 0.001000 Time 0.019663 -2023-02-13 17:43:08,737 - Epoch: [55][ 1180/ 1207] Overall Loss 0.367014 Objective Loss 0.367014 LR 0.001000 Time 0.019657 -2023-02-13 17:43:08,927 - Epoch: [55][ 1190/ 1207] Overall Loss 0.366997 Objective Loss 0.366997 LR 0.001000 Time 0.019651 -2023-02-13 17:43:09,170 - Epoch: [55][ 1200/ 1207] Overall Loss 0.366733 Objective Loss 0.366733 LR 0.001000 Time 0.019689 -2023-02-13 17:43:09,286 - Epoch: [55][ 1207/ 1207] Overall Loss 0.366602 Objective Loss 0.366602 Top1 81.707317 Top5 98.475610 LR 0.001000 Time 0.019671 -2023-02-13 17:43:09,362 - --- validate (epoch=55)----------- -2023-02-13 17:43:09,362 - 34311 samples (256 per mini-batch) -2023-02-13 17:43:09,785 - Epoch: [55][ 10/ 135] Loss 0.383640 Top1 81.992188 Top5 97.187500 -2023-02-13 17:43:09,911 - Epoch: [55][ 20/ 135] Loss 0.360927 Top1 81.542969 Top5 96.914062 -2023-02-13 17:43:10,042 - Epoch: [55][ 30/ 135] Loss 0.379814 Top1 80.807292 Top5 96.718750 -2023-02-13 17:43:10,168 - Epoch: [55][ 40/ 135] Loss 0.381580 Top1 80.556641 Top5 96.826172 -2023-02-13 17:43:10,295 - Epoch: [55][ 50/ 135] Loss 0.380363 Top1 80.531250 Top5 96.898438 -2023-02-13 17:43:10,425 - Epoch: [55][ 60/ 135] Loss 0.380620 Top1 80.475260 Top5 96.816406 -2023-02-13 17:43:10,555 - Epoch: [55][ 70/ 135] Loss 0.381910 Top1 80.379464 Top5 96.768973 -2023-02-13 17:43:10,685 - Epoch: [55][ 80/ 135] Loss 0.383317 Top1 80.341797 Top5 96.791992 -2023-02-13 17:43:10,816 - Epoch: [55][ 90/ 135] Loss 0.380111 Top1 80.355903 Top5 96.861979 -2023-02-13 17:43:10,946 - Epoch: [55][ 100/ 135] Loss 0.378886 Top1 80.351562 Top5 96.835938 -2023-02-13 17:43:11,074 - Epoch: [55][ 110/ 135] Loss 0.382065 Top1 80.319602 Top5 96.800426 -2023-02-13 17:43:11,203 - Epoch: [55][ 120/ 135] Loss 0.383984 Top1 80.419922 Top5 96.813151 -2023-02-13 17:43:11,337 - Epoch: [55][ 130/ 135] Loss 0.384352 Top1 80.456731 Top5 96.829928 -2023-02-13 17:43:11,384 - Epoch: [55][ 135/ 135] Loss 0.382668 Top1 80.466906 Top5 96.834834 -2023-02-13 17:43:11,452 - ==> Top1: 80.467 Top5: 96.835 Loss: 0.383 - -2023-02-13 17:43:11,453 - ==> Confusion: -[[ 840 5 9 3 6 5 1 1 4 59 1 2 1 3 6 3 8 3 0 2 5] - [ 1 935 0 5 9 26 4 16 4 2 2 2 1 0 4 5 9 1 3 1 3] - [ 9 4 962 7 3 1 13 16 1 1 3 1 1 4 4 5 6 4 6 2 5] - [ 10 1 26 883 2 4 5 5 4 2 10 0 5 1 15 2 3 9 21 1 7] - [ 15 14 1 0 986 11 2 1 0 4 1 4 1 2 3 3 7 3 0 4 4] - [ 2 23 2 7 3 958 3 16 5 3 1 18 2 11 2 2 2 2 1 3 4] - [ 3 3 15 3 0 6 1033 10 0 1 1 1 3 2 0 5 2 2 0 8 1] - [ 6 18 15 2 2 28 4 906 1 1 2 8 3 0 0 0 1 2 15 9 1] - [ 18 1 1 1 2 0 0 1 893 47 6 3 1 5 23 2 0 0 4 0 1] - [ 95 1 3 0 4 2 1 0 39 847 1 0 0 10 4 1 0 0 0 0 4] - [ 4 2 7 6 1 2 3 7 28 1 960 2 3 5 1 1 2 0 8 0 8] - [ 4 2 1 1 2 11 0 4 3 3 0 918 12 6 2 5 3 11 1 16 0] - [ 2 2 2 6 2 2 0 2 5 0 0 39 854 1 2 8 3 21 1 3 4] - [ 10 4 4 1 9 11 1 3 34 36 13 9 2 852 7 6 6 3 1 5 7] - [ 17 2 5 11 7 6 0 3 27 10 1 1 6 0 963 0 3 7 15 0 8] - [ 3 2 7 0 3 1 5 2 0 0 0 13 5 0 1 962 12 17 0 9 4] - [ 3 8 3 1 7 6 0 0 4 2 0 3 3 0 0 13 997 0 1 2 8] - [ 6 2 0 4 2 1 2 2 0 2 1 11 23 1 0 12 0 977 0 3 2] - [ 7 11 10 10 1 0 1 42 6 3 1 3 6 0 8 0 2 1 972 0 2] - [ 0 5 0 1 1 5 7 18 1 2 2 29 4 2 0 4 5 1 1 1055 5] - [ 251 327 324 135 180 269 129 210 183 140 196 173 420 281 171 151 392 120 172 354 8856]] - -2023-02-13 17:43:11,455 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:43:11,455 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:43:11,461 - - -2023-02-13 17:43:11,461 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:43:12,454 - Epoch: [56][ 10/ 1207] Overall Loss 0.345352 Objective Loss 0.345352 LR 0.001000 Time 0.099257 -2023-02-13 17:43:12,653 - Epoch: [56][ 20/ 1207] Overall Loss 0.350661 Objective Loss 0.350661 LR 0.001000 Time 0.059549 -2023-02-13 17:43:12,843 - Epoch: [56][ 30/ 1207] Overall Loss 0.347872 Objective Loss 0.347872 LR 0.001000 Time 0.046025 -2023-02-13 17:43:13,034 - Epoch: [56][ 40/ 1207] Overall Loss 0.348175 Objective Loss 0.348175 LR 0.001000 Time 0.039265 -2023-02-13 17:43:13,223 - Epoch: [56][ 50/ 1207] Overall Loss 0.346260 Objective Loss 0.346260 LR 0.001000 Time 0.035197 -2023-02-13 17:43:13,414 - Epoch: [56][ 60/ 1207] Overall Loss 0.345669 Objective Loss 0.345669 LR 0.001000 Time 0.032506 -2023-02-13 17:43:13,604 - Epoch: [56][ 70/ 1207] Overall Loss 0.347164 Objective Loss 0.347164 LR 0.001000 Time 0.030573 -2023-02-13 17:43:13,794 - Epoch: [56][ 80/ 1207] Overall Loss 0.353024 Objective Loss 0.353024 LR 0.001000 Time 0.029125 -2023-02-13 17:43:13,984 - Epoch: [56][ 90/ 1207] Overall Loss 0.350507 Objective Loss 0.350507 LR 0.001000 Time 0.027992 -2023-02-13 17:43:14,175 - Epoch: [56][ 100/ 1207] Overall Loss 0.349463 Objective Loss 0.349463 LR 0.001000 Time 0.027097 -2023-02-13 17:43:14,365 - Epoch: [56][ 110/ 1207] Overall Loss 0.350403 Objective Loss 0.350403 LR 0.001000 Time 0.026356 -2023-02-13 17:43:14,555 - Epoch: [56][ 120/ 1207] Overall Loss 0.354329 Objective Loss 0.354329 LR 0.001000 Time 0.025740 -2023-02-13 17:43:14,745 - Epoch: [56][ 130/ 1207] Overall Loss 0.356254 Objective Loss 0.356254 LR 0.001000 Time 0.025225 -2023-02-13 17:43:14,937 - Epoch: [56][ 140/ 1207] Overall Loss 0.355603 Objective Loss 0.355603 LR 0.001000 Time 0.024786 -2023-02-13 17:43:15,127 - Epoch: [56][ 150/ 1207] Overall Loss 0.356542 Objective Loss 0.356542 LR 0.001000 Time 0.024402 -2023-02-13 17:43:15,318 - Epoch: [56][ 160/ 1207] Overall Loss 0.357695 Objective Loss 0.357695 LR 0.001000 Time 0.024065 -2023-02-13 17:43:15,508 - Epoch: [56][ 170/ 1207] Overall Loss 0.356424 Objective Loss 0.356424 LR 0.001000 Time 0.023765 -2023-02-13 17:43:15,698 - Epoch: [56][ 180/ 1207] Overall Loss 0.356454 Objective Loss 0.356454 LR 0.001000 Time 0.023502 -2023-02-13 17:43:15,890 - Epoch: [56][ 190/ 1207] Overall Loss 0.357505 Objective Loss 0.357505 LR 0.001000 Time 0.023269 -2023-02-13 17:43:16,081 - Epoch: [56][ 200/ 1207] Overall Loss 0.356763 Objective Loss 0.356763 LR 0.001000 Time 0.023060 -2023-02-13 17:43:16,270 - Epoch: [56][ 210/ 1207] Overall Loss 0.355715 Objective Loss 0.355715 LR 0.001000 Time 0.022861 -2023-02-13 17:43:16,461 - Epoch: [56][ 220/ 1207] Overall Loss 0.355242 Objective Loss 0.355242 LR 0.001000 Time 0.022688 -2023-02-13 17:43:16,651 - Epoch: [56][ 230/ 1207] Overall Loss 0.355839 Objective Loss 0.355839 LR 0.001000 Time 0.022528 -2023-02-13 17:43:16,842 - Epoch: [56][ 240/ 1207] Overall Loss 0.356398 Objective Loss 0.356398 LR 0.001000 Time 0.022384 -2023-02-13 17:43:17,034 - Epoch: [56][ 250/ 1207] Overall Loss 0.355964 Objective Loss 0.355964 LR 0.001000 Time 0.022253 -2023-02-13 17:43:17,224 - Epoch: [56][ 260/ 1207] Overall Loss 0.356493 Objective Loss 0.356493 LR 0.001000 Time 0.022129 -2023-02-13 17:43:17,414 - Epoch: [56][ 270/ 1207] Overall Loss 0.355967 Objective Loss 0.355967 LR 0.001000 Time 0.022011 -2023-02-13 17:43:17,605 - Epoch: [56][ 280/ 1207] Overall Loss 0.355482 Objective Loss 0.355482 LR 0.001000 Time 0.021905 -2023-02-13 17:43:17,796 - Epoch: [56][ 290/ 1207] Overall Loss 0.356508 Objective Loss 0.356508 LR 0.001000 Time 0.021805 -2023-02-13 17:43:17,986 - Epoch: [56][ 300/ 1207] Overall Loss 0.357236 Objective Loss 0.357236 LR 0.001000 Time 0.021712 -2023-02-13 17:43:18,177 - Epoch: [56][ 310/ 1207] Overall Loss 0.357491 Objective Loss 0.357491 LR 0.001000 Time 0.021627 -2023-02-13 17:43:18,368 - Epoch: [56][ 320/ 1207] Overall Loss 0.357435 Objective Loss 0.357435 LR 0.001000 Time 0.021546 -2023-02-13 17:43:18,558 - Epoch: [56][ 330/ 1207] Overall Loss 0.357130 Objective Loss 0.357130 LR 0.001000 Time 0.021468 -2023-02-13 17:43:18,749 - Epoch: [56][ 340/ 1207] Overall Loss 0.356685 Objective Loss 0.356685 LR 0.001000 Time 0.021397 -2023-02-13 17:43:18,939 - Epoch: [56][ 350/ 1207] Overall Loss 0.356110 Objective Loss 0.356110 LR 0.001000 Time 0.021327 -2023-02-13 17:43:19,130 - Epoch: [56][ 360/ 1207] Overall Loss 0.356487 Objective Loss 0.356487 LR 0.001000 Time 0.021266 -2023-02-13 17:43:19,321 - Epoch: [56][ 370/ 1207] Overall Loss 0.356562 Objective Loss 0.356562 LR 0.001000 Time 0.021205 -2023-02-13 17:43:19,511 - Epoch: [56][ 380/ 1207] Overall Loss 0.357370 Objective Loss 0.357370 LR 0.001000 Time 0.021147 -2023-02-13 17:43:19,702 - Epoch: [56][ 390/ 1207] Overall Loss 0.357331 Objective Loss 0.357331 LR 0.001000 Time 0.021092 -2023-02-13 17:43:19,893 - Epoch: [56][ 400/ 1207] Overall Loss 0.357686 Objective Loss 0.357686 LR 0.001000 Time 0.021042 -2023-02-13 17:43:20,084 - Epoch: [56][ 410/ 1207] Overall Loss 0.357310 Objective Loss 0.357310 LR 0.001000 Time 0.020993 -2023-02-13 17:43:20,275 - Epoch: [56][ 420/ 1207] Overall Loss 0.356853 Objective Loss 0.356853 LR 0.001000 Time 0.020949 -2023-02-13 17:43:20,466 - Epoch: [56][ 430/ 1207] Overall Loss 0.356523 Objective Loss 0.356523 LR 0.001000 Time 0.020903 -2023-02-13 17:43:20,658 - Epoch: [56][ 440/ 1207] Overall Loss 0.356833 Objective Loss 0.356833 LR 0.001000 Time 0.020863 -2023-02-13 17:43:20,849 - Epoch: [56][ 450/ 1207] Overall Loss 0.356537 Objective Loss 0.356537 LR 0.001000 Time 0.020825 -2023-02-13 17:43:21,041 - Epoch: [56][ 460/ 1207] Overall Loss 0.356461 Objective Loss 0.356461 LR 0.001000 Time 0.020788 -2023-02-13 17:43:21,232 - Epoch: [56][ 470/ 1207] Overall Loss 0.356491 Objective Loss 0.356491 LR 0.001000 Time 0.020751 -2023-02-13 17:43:21,422 - Epoch: [56][ 480/ 1207] Overall Loss 0.356537 Objective Loss 0.356537 LR 0.001000 Time 0.020714 -2023-02-13 17:43:21,613 - Epoch: [56][ 490/ 1207] Overall Loss 0.355945 Objective Loss 0.355945 LR 0.001000 Time 0.020680 -2023-02-13 17:43:21,804 - Epoch: [56][ 500/ 1207] Overall Loss 0.355756 Objective Loss 0.355756 LR 0.001000 Time 0.020648 -2023-02-13 17:43:21,995 - Epoch: [56][ 510/ 1207] Overall Loss 0.355580 Objective Loss 0.355580 LR 0.001000 Time 0.020617 -2023-02-13 17:43:22,187 - Epoch: [56][ 520/ 1207] Overall Loss 0.355656 Objective Loss 0.355656 LR 0.001000 Time 0.020588 -2023-02-13 17:43:22,377 - Epoch: [56][ 530/ 1207] Overall Loss 0.355838 Objective Loss 0.355838 LR 0.001000 Time 0.020559 -2023-02-13 17:43:22,568 - Epoch: [56][ 540/ 1207] Overall Loss 0.356143 Objective Loss 0.356143 LR 0.001000 Time 0.020530 -2023-02-13 17:43:22,758 - Epoch: [56][ 550/ 1207] Overall Loss 0.356333 Objective Loss 0.356333 LR 0.001000 Time 0.020503 -2023-02-13 17:43:22,948 - Epoch: [56][ 560/ 1207] Overall Loss 0.356496 Objective Loss 0.356496 LR 0.001000 Time 0.020476 -2023-02-13 17:43:23,139 - Epoch: [56][ 570/ 1207] Overall Loss 0.356124 Objective Loss 0.356124 LR 0.001000 Time 0.020451 -2023-02-13 17:43:23,330 - Epoch: [56][ 580/ 1207] Overall Loss 0.356089 Objective Loss 0.356089 LR 0.001000 Time 0.020426 -2023-02-13 17:43:23,520 - Epoch: [56][ 590/ 1207] Overall Loss 0.356333 Objective Loss 0.356333 LR 0.001000 Time 0.020402 -2023-02-13 17:43:23,712 - Epoch: [56][ 600/ 1207] Overall Loss 0.356808 Objective Loss 0.356808 LR 0.001000 Time 0.020380 -2023-02-13 17:43:23,901 - Epoch: [56][ 610/ 1207] Overall Loss 0.357379 Objective Loss 0.357379 LR 0.001000 Time 0.020356 -2023-02-13 17:43:24,093 - Epoch: [56][ 620/ 1207] Overall Loss 0.357605 Objective Loss 0.357605 LR 0.001000 Time 0.020336 -2023-02-13 17:43:24,283 - Epoch: [56][ 630/ 1207] Overall Loss 0.357499 Objective Loss 0.357499 LR 0.001000 Time 0.020316 -2023-02-13 17:43:24,474 - Epoch: [56][ 640/ 1207] Overall Loss 0.357481 Objective Loss 0.357481 LR 0.001000 Time 0.020296 -2023-02-13 17:43:24,665 - Epoch: [56][ 650/ 1207] Overall Loss 0.357638 Objective Loss 0.357638 LR 0.001000 Time 0.020276 -2023-02-13 17:43:24,855 - Epoch: [56][ 660/ 1207] Overall Loss 0.357500 Objective Loss 0.357500 LR 0.001000 Time 0.020257 -2023-02-13 17:43:25,046 - Epoch: [56][ 670/ 1207] Overall Loss 0.358114 Objective Loss 0.358114 LR 0.001000 Time 0.020239 -2023-02-13 17:43:25,237 - Epoch: [56][ 680/ 1207] Overall Loss 0.358391 Objective Loss 0.358391 LR 0.001000 Time 0.020222 -2023-02-13 17:43:25,428 - Epoch: [56][ 690/ 1207] Overall Loss 0.358698 Objective Loss 0.358698 LR 0.001000 Time 0.020205 -2023-02-13 17:43:25,619 - Epoch: [56][ 700/ 1207] Overall Loss 0.358856 Objective Loss 0.358856 LR 0.001000 Time 0.020189 -2023-02-13 17:43:25,811 - Epoch: [56][ 710/ 1207] Overall Loss 0.358654 Objective Loss 0.358654 LR 0.001000 Time 0.020175 -2023-02-13 17:43:26,002 - Epoch: [56][ 720/ 1207] Overall Loss 0.358830 Objective Loss 0.358830 LR 0.001000 Time 0.020159 -2023-02-13 17:43:26,193 - Epoch: [56][ 730/ 1207] Overall Loss 0.359238 Objective Loss 0.359238 LR 0.001000 Time 0.020144 -2023-02-13 17:43:26,384 - Epoch: [56][ 740/ 1207] Overall Loss 0.359219 Objective Loss 0.359219 LR 0.001000 Time 0.020129 -2023-02-13 17:43:26,574 - Epoch: [56][ 750/ 1207] Overall Loss 0.359416 Objective Loss 0.359416 LR 0.001000 Time 0.020114 -2023-02-13 17:43:26,766 - Epoch: [56][ 760/ 1207] Overall Loss 0.359678 Objective Loss 0.359678 LR 0.001000 Time 0.020101 -2023-02-13 17:43:26,958 - Epoch: [56][ 770/ 1207] Overall Loss 0.359710 Objective Loss 0.359710 LR 0.001000 Time 0.020089 -2023-02-13 17:43:27,150 - Epoch: [56][ 780/ 1207] Overall Loss 0.359658 Objective Loss 0.359658 LR 0.001000 Time 0.020077 -2023-02-13 17:43:27,341 - Epoch: [56][ 790/ 1207] Overall Loss 0.359249 Objective Loss 0.359249 LR 0.001000 Time 0.020063 -2023-02-13 17:43:27,532 - Epoch: [56][ 800/ 1207] Overall Loss 0.359279 Objective Loss 0.359279 LR 0.001000 Time 0.020051 -2023-02-13 17:43:27,723 - Epoch: [56][ 810/ 1207] Overall Loss 0.359128 Objective Loss 0.359128 LR 0.001000 Time 0.020039 -2023-02-13 17:43:27,913 - Epoch: [56][ 820/ 1207] Overall Loss 0.359049 Objective Loss 0.359049 LR 0.001000 Time 0.020026 -2023-02-13 17:43:28,104 - Epoch: [56][ 830/ 1207] Overall Loss 0.359267 Objective Loss 0.359267 LR 0.001000 Time 0.020015 -2023-02-13 17:43:28,295 - Epoch: [56][ 840/ 1207] Overall Loss 0.358774 Objective Loss 0.358774 LR 0.001000 Time 0.020004 -2023-02-13 17:43:28,486 - Epoch: [56][ 850/ 1207] Overall Loss 0.359197 Objective Loss 0.359197 LR 0.001000 Time 0.019992 -2023-02-13 17:43:28,677 - Epoch: [56][ 860/ 1207] Overall Loss 0.359155 Objective Loss 0.359155 LR 0.001000 Time 0.019982 -2023-02-13 17:43:28,868 - Epoch: [56][ 870/ 1207] Overall Loss 0.359098 Objective Loss 0.359098 LR 0.001000 Time 0.019971 -2023-02-13 17:43:29,060 - Epoch: [56][ 880/ 1207] Overall Loss 0.358825 Objective Loss 0.358825 LR 0.001000 Time 0.019962 -2023-02-13 17:43:29,251 - Epoch: [56][ 890/ 1207] Overall Loss 0.358504 Objective Loss 0.358504 LR 0.001000 Time 0.019952 -2023-02-13 17:43:29,442 - Epoch: [56][ 900/ 1207] Overall Loss 0.358833 Objective Loss 0.358833 LR 0.001000 Time 0.019942 -2023-02-13 17:43:29,634 - Epoch: [56][ 910/ 1207] Overall Loss 0.358655 Objective Loss 0.358655 LR 0.001000 Time 0.019933 -2023-02-13 17:43:29,825 - Epoch: [56][ 920/ 1207] Overall Loss 0.358516 Objective Loss 0.358516 LR 0.001000 Time 0.019923 -2023-02-13 17:43:30,015 - Epoch: [56][ 930/ 1207] Overall Loss 0.358329 Objective Loss 0.358329 LR 0.001000 Time 0.019913 -2023-02-13 17:43:30,207 - Epoch: [56][ 940/ 1207] Overall Loss 0.358226 Objective Loss 0.358226 LR 0.001000 Time 0.019905 -2023-02-13 17:43:30,397 - Epoch: [56][ 950/ 1207] Overall Loss 0.358341 Objective Loss 0.358341 LR 0.001000 Time 0.019895 -2023-02-13 17:43:30,588 - Epoch: [56][ 960/ 1207] Overall Loss 0.358448 Objective Loss 0.358448 LR 0.001000 Time 0.019887 -2023-02-13 17:43:30,779 - Epoch: [56][ 970/ 1207] Overall Loss 0.358184 Objective Loss 0.358184 LR 0.001000 Time 0.019878 -2023-02-13 17:43:30,971 - Epoch: [56][ 980/ 1207] Overall Loss 0.358158 Objective Loss 0.358158 LR 0.001000 Time 0.019871 -2023-02-13 17:43:31,163 - Epoch: [56][ 990/ 1207] Overall Loss 0.358138 Objective Loss 0.358138 LR 0.001000 Time 0.019864 -2023-02-13 17:43:31,354 - Epoch: [56][ 1000/ 1207] Overall Loss 0.358131 Objective Loss 0.358131 LR 0.001000 Time 0.019856 -2023-02-13 17:43:31,545 - Epoch: [56][ 1010/ 1207] Overall Loss 0.358023 Objective Loss 0.358023 LR 0.001000 Time 0.019848 -2023-02-13 17:43:31,737 - Epoch: [56][ 1020/ 1207] Overall Loss 0.358129 Objective Loss 0.358129 LR 0.001000 Time 0.019841 -2023-02-13 17:43:31,928 - Epoch: [56][ 1030/ 1207] Overall Loss 0.358112 Objective Loss 0.358112 LR 0.001000 Time 0.019833 -2023-02-13 17:43:32,120 - Epoch: [56][ 1040/ 1207] Overall Loss 0.358252 Objective Loss 0.358252 LR 0.001000 Time 0.019827 -2023-02-13 17:43:32,310 - Epoch: [56][ 1050/ 1207] Overall Loss 0.357955 Objective Loss 0.357955 LR 0.001000 Time 0.019819 -2023-02-13 17:43:32,501 - Epoch: [56][ 1060/ 1207] Overall Loss 0.357894 Objective Loss 0.357894 LR 0.001000 Time 0.019812 -2023-02-13 17:43:32,692 - Epoch: [56][ 1070/ 1207] Overall Loss 0.357670 Objective Loss 0.357670 LR 0.001000 Time 0.019805 -2023-02-13 17:43:32,883 - Epoch: [56][ 1080/ 1207] Overall Loss 0.357898 Objective Loss 0.357898 LR 0.001000 Time 0.019798 -2023-02-13 17:43:33,074 - Epoch: [56][ 1090/ 1207] Overall Loss 0.358114 Objective Loss 0.358114 LR 0.001000 Time 0.019791 -2023-02-13 17:43:33,266 - Epoch: [56][ 1100/ 1207] Overall Loss 0.358156 Objective Loss 0.358156 LR 0.001000 Time 0.019785 -2023-02-13 17:43:33,457 - Epoch: [56][ 1110/ 1207] Overall Loss 0.358160 Objective Loss 0.358160 LR 0.001000 Time 0.019779 -2023-02-13 17:43:33,648 - Epoch: [56][ 1120/ 1207] Overall Loss 0.358024 Objective Loss 0.358024 LR 0.001000 Time 0.019772 -2023-02-13 17:43:33,839 - Epoch: [56][ 1130/ 1207] Overall Loss 0.358148 Objective Loss 0.358148 LR 0.001000 Time 0.019766 -2023-02-13 17:43:34,030 - Epoch: [56][ 1140/ 1207] Overall Loss 0.358287 Objective Loss 0.358287 LR 0.001000 Time 0.019760 -2023-02-13 17:43:34,221 - Epoch: [56][ 1150/ 1207] Overall Loss 0.358568 Objective Loss 0.358568 LR 0.001000 Time 0.019754 -2023-02-13 17:43:34,412 - Epoch: [56][ 1160/ 1207] Overall Loss 0.358565 Objective Loss 0.358565 LR 0.001000 Time 0.019748 -2023-02-13 17:43:34,603 - Epoch: [56][ 1170/ 1207] Overall Loss 0.358679 Objective Loss 0.358679 LR 0.001000 Time 0.019742 -2023-02-13 17:43:34,793 - Epoch: [56][ 1180/ 1207] Overall Loss 0.358822 Objective Loss 0.358822 LR 0.001000 Time 0.019736 -2023-02-13 17:43:34,984 - Epoch: [56][ 1190/ 1207] Overall Loss 0.358648 Objective Loss 0.358648 LR 0.001000 Time 0.019730 -2023-02-13 17:43:35,227 - Epoch: [56][ 1200/ 1207] Overall Loss 0.358572 Objective Loss 0.358572 LR 0.001000 Time 0.019767 -2023-02-13 17:43:35,342 - Epoch: [56][ 1207/ 1207] Overall Loss 0.358450 Objective Loss 0.358450 Top1 84.756098 Top5 97.256098 LR 0.001000 Time 0.019748 -2023-02-13 17:43:35,414 - --- validate (epoch=56)----------- -2023-02-13 17:43:35,415 - 34311 samples (256 per mini-batch) -2023-02-13 17:43:35,822 - Epoch: [56][ 10/ 135] Loss 0.352783 Top1 82.148438 Top5 97.500000 -2023-02-13 17:43:35,953 - Epoch: [56][ 20/ 135] Loss 0.350552 Top1 82.519531 Top5 97.226562 -2023-02-13 17:43:36,084 - Epoch: [56][ 30/ 135] Loss 0.352787 Top1 82.330729 Top5 97.135417 -2023-02-13 17:43:36,212 - Epoch: [56][ 40/ 135] Loss 0.351899 Top1 82.216797 Top5 97.138672 -2023-02-13 17:43:36,343 - Epoch: [56][ 50/ 135] Loss 0.355884 Top1 81.953125 Top5 97.164062 -2023-02-13 17:43:36,471 - Epoch: [56][ 60/ 135] Loss 0.360394 Top1 81.848958 Top5 97.187500 -2023-02-13 17:43:36,597 - Epoch: [56][ 70/ 135] Loss 0.365920 Top1 81.713170 Top5 97.170759 -2023-02-13 17:43:36,721 - Epoch: [56][ 80/ 135] Loss 0.369571 Top1 81.494141 Top5 97.182617 -2023-02-13 17:43:36,847 - Epoch: [56][ 90/ 135] Loss 0.370767 Top1 81.623264 Top5 97.252604 -2023-02-13 17:43:36,972 - Epoch: [56][ 100/ 135] Loss 0.374561 Top1 81.722656 Top5 97.167969 -2023-02-13 17:43:37,096 - Epoch: [56][ 110/ 135] Loss 0.375358 Top1 81.583807 Top5 97.166193 -2023-02-13 17:43:37,220 - Epoch: [56][ 120/ 135] Loss 0.376434 Top1 81.494141 Top5 97.109375 -2023-02-13 17:43:37,345 - Epoch: [56][ 130/ 135] Loss 0.376472 Top1 81.502404 Top5 97.100361 -2023-02-13 17:43:37,389 - Epoch: [56][ 135/ 135] Loss 0.376946 Top1 81.554021 Top5 97.123372 -2023-02-13 17:43:37,472 - ==> Top1: 81.554 Top5: 97.123 Loss: 0.377 - -2023-02-13 17:43:37,473 - ==> Confusion: -[[ 827 3 8 0 19 7 0 1 0 70 0 5 1 5 3 2 3 2 2 1 8] - [ 8 919 1 4 9 45 1 20 3 2 4 2 1 0 1 0 4 0 6 1 2] - [ 11 7 955 13 0 3 21 16 1 1 4 1 3 5 0 5 1 1 7 0 3] - [ 4 2 30 884 3 7 1 1 1 3 17 1 5 1 18 2 2 6 15 2 11] - [ 12 9 2 1 985 14 2 1 2 2 1 7 1 5 7 3 1 0 0 4 7] - [ 7 31 2 6 6 949 3 16 1 5 2 10 3 18 0 1 1 2 1 3 3] - [ 3 5 19 2 0 4 1032 6 0 2 8 3 1 2 1 3 0 1 0 5 2] - [ 1 14 12 4 4 31 4 909 0 1 3 9 2 2 0 1 0 1 15 7 4] - [ 27 1 0 1 2 0 0 1 849 54 11 1 1 22 28 1 1 1 5 1 2] - [ 55 2 5 0 4 5 0 3 24 865 2 1 1 28 6 2 1 1 0 0 7] - [ 5 2 5 11 1 2 3 4 20 4 957 2 1 11 3 0 1 0 11 1 7] - [ 4 5 1 1 3 17 2 9 3 1 0 894 21 9 2 7 1 3 1 19 2] - [ 4 0 0 10 2 8 1 1 1 0 0 37 852 1 7 5 1 19 0 1 9] - [ 8 1 3 0 3 11 1 1 13 10 9 6 3 932 5 4 3 0 1 6 4] - [ 14 3 3 19 3 2 0 3 15 9 5 2 4 2 990 0 0 1 9 1 7] - [ 6 4 8 2 10 4 9 1 0 0 0 8 4 3 0 949 10 12 0 10 6] - [ 4 9 1 0 15 6 0 0 4 0 1 4 1 5 1 5 981 1 4 6 13] - [ 9 1 2 7 1 2 0 1 1 1 1 9 26 3 1 16 1 959 0 2 8] - [ 4 3 7 14 1 4 0 34 4 2 5 2 4 0 14 0 1 1 980 1 5] - [ 1 4 2 2 1 11 11 16 1 0 2 17 5 4 0 4 4 3 1 1049 10] - [ 187 292 301 152 182 293 98 195 76 104 240 132 391 390 167 106 260 102 185 316 9265]] - -2023-02-13 17:43:37,474 - ==> Best [Top1: 82.254 Top5: 97.260 Sparsity:0.00 Params: 148928 on epoch: 37] -2023-02-13 17:43:37,474 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:43:37,480 - - -2023-02-13 17:43:37,480 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:43:38,379 - Epoch: [57][ 10/ 1207] Overall Loss 0.332622 Objective Loss 0.332622 LR 0.001000 Time 0.089875 -2023-02-13 17:43:38,571 - Epoch: [57][ 20/ 1207] Overall Loss 0.336710 Objective Loss 0.336710 LR 0.001000 Time 0.054476 -2023-02-13 17:43:38,759 - Epoch: [57][ 30/ 1207] Overall Loss 0.338775 Objective Loss 0.338775 LR 0.001000 Time 0.042598 -2023-02-13 17:43:38,947 - Epoch: [57][ 40/ 1207] Overall Loss 0.342723 Objective Loss 0.342723 LR 0.001000 Time 0.036633 -2023-02-13 17:43:39,136 - Epoch: [57][ 50/ 1207] Overall Loss 0.347017 Objective Loss 0.347017 LR 0.001000 Time 0.033076 -2023-02-13 17:43:39,324 - Epoch: [57][ 60/ 1207] Overall Loss 0.341720 Objective Loss 0.341720 LR 0.001000 Time 0.030691 -2023-02-13 17:43:39,512 - Epoch: [57][ 70/ 1207] Overall Loss 0.346183 Objective Loss 0.346183 LR 0.001000 Time 0.028988 -2023-02-13 17:43:39,700 - Epoch: [57][ 80/ 1207] Overall Loss 0.346441 Objective Loss 0.346441 LR 0.001000 Time 0.027715 -2023-02-13 17:43:39,888 - Epoch: [57][ 90/ 1207] Overall Loss 0.344065 Objective Loss 0.344065 LR 0.001000 Time 0.026717 -2023-02-13 17:43:40,076 - Epoch: [57][ 100/ 1207] Overall Loss 0.344136 Objective Loss 0.344136 LR 0.001000 Time 0.025920 -2023-02-13 17:43:40,264 - Epoch: [57][ 110/ 1207] Overall Loss 0.345611 Objective Loss 0.345611 LR 0.001000 Time 0.025273 -2023-02-13 17:43:40,452 - Epoch: [57][ 120/ 1207] Overall Loss 0.345151 Objective Loss 0.345151 LR 0.001000 Time 0.024730 -2023-02-13 17:43:40,640 - Epoch: [57][ 130/ 1207] Overall Loss 0.347015 Objective Loss 0.347015 LR 0.001000 Time 0.024269 -2023-02-13 17:43:40,829 - Epoch: [57][ 140/ 1207] Overall Loss 0.345455 Objective Loss 0.345455 LR 0.001000 Time 0.023885 -2023-02-13 17:43:41,017 - Epoch: [57][ 150/ 1207] Overall Loss 0.345439 Objective Loss 0.345439 LR 0.001000 Time 0.023544 -2023-02-13 17:43:41,206 - Epoch: [57][ 160/ 1207] Overall Loss 0.347034 Objective Loss 0.347034 LR 0.001000 Time 0.023249 -2023-02-13 17:43:41,394 - Epoch: [57][ 170/ 1207] Overall Loss 0.347128 Objective Loss 0.347128 LR 0.001000 Time 0.022986 -2023-02-13 17:43:41,582 - Epoch: [57][ 180/ 1207] Overall Loss 0.344923 Objective Loss 0.344923 LR 0.001000 Time 0.022751 -2023-02-13 17:43:41,771 - Epoch: [57][ 190/ 1207] Overall Loss 0.345387 Objective Loss 0.345387 LR 0.001000 Time 0.022545 -2023-02-13 17:43:41,959 - Epoch: [57][ 200/ 1207] Overall Loss 0.344811 Objective Loss 0.344811 LR 0.001000 Time 0.022359 -2023-02-13 17:43:42,148 - Epoch: [57][ 210/ 1207] Overall Loss 0.345926 Objective Loss 0.345926 LR 0.001000 Time 0.022191 -2023-02-13 17:43:42,337 - Epoch: [57][ 220/ 1207] Overall Loss 0.345722 Objective Loss 0.345722 LR 0.001000 Time 0.022039 -2023-02-13 17:43:42,525 - Epoch: [57][ 230/ 1207] Overall Loss 0.346017 Objective Loss 0.346017 LR 0.001000 Time 0.021898 -2023-02-13 17:43:42,714 - Epoch: [57][ 240/ 1207] Overall Loss 0.347123 Objective Loss 0.347123 LR 0.001000 Time 0.021769 -2023-02-13 17:43:42,902 - Epoch: [57][ 250/ 1207] Overall Loss 0.348042 Objective Loss 0.348042 LR 0.001000 Time 0.021651 -2023-02-13 17:43:43,090 - Epoch: [57][ 260/ 1207] Overall Loss 0.349004 Objective Loss 0.349004 LR 0.001000 Time 0.021540 -2023-02-13 17:43:43,279 - Epoch: [57][ 270/ 1207] Overall Loss 0.348436 Objective Loss 0.348436 LR 0.001000 Time 0.021441 -2023-02-13 17:43:43,467 - Epoch: [57][ 280/ 1207] Overall Loss 0.346853 Objective Loss 0.346853 LR 0.001000 Time 0.021346 -2023-02-13 17:43:43,656 - Epoch: [57][ 290/ 1207] Overall Loss 0.347448 Objective Loss 0.347448 LR 0.001000 Time 0.021257 -2023-02-13 17:43:43,844 - Epoch: [57][ 300/ 1207] Overall Loss 0.348855 Objective Loss 0.348855 LR 0.001000 Time 0.021176 -2023-02-13 17:43:44,033 - Epoch: [57][ 310/ 1207] Overall Loss 0.348447 Objective Loss 0.348447 LR 0.001000 Time 0.021101 -2023-02-13 17:43:44,222 - Epoch: [57][ 320/ 1207] Overall Loss 0.349259 Objective Loss 0.349259 LR 0.001000 Time 0.021030 -2023-02-13 17:43:44,410 - Epoch: [57][ 330/ 1207] Overall Loss 0.349652 Objective Loss 0.349652 LR 0.001000 Time 0.020962 -2023-02-13 17:43:44,598 - Epoch: [57][ 340/ 1207] Overall Loss 0.349131 Objective Loss 0.349131 LR 0.001000 Time 0.020898 -2023-02-13 17:43:44,787 - Epoch: [57][ 350/ 1207] Overall Loss 0.349862 Objective Loss 0.349862 LR 0.001000 Time 0.020839 -2023-02-13 17:43:44,975 - Epoch: [57][ 360/ 1207] Overall Loss 0.349240 Objective Loss 0.349240 LR 0.001000 Time 0.020782 -2023-02-13 17:43:45,164 - Epoch: [57][ 370/ 1207] Overall Loss 0.349300 Objective Loss 0.349300 LR 0.001000 Time 0.020729 -2023-02-13 17:43:45,351 - Epoch: [57][ 380/ 1207] Overall Loss 0.348653 Objective Loss 0.348653 LR 0.001000 Time 0.020676 -2023-02-13 17:43:45,539 - Epoch: [57][ 390/ 1207] Overall Loss 0.348267 Objective Loss 0.348267 LR 0.001000 Time 0.020627 -2023-02-13 17:43:45,727 - Epoch: [57][ 400/ 1207] Overall Loss 0.347275 Objective Loss 0.347275 LR 0.001000 Time 0.020580 -2023-02-13 17:43:45,917 - Epoch: [57][ 410/ 1207] Overall Loss 0.347411 Objective Loss 0.347411 LR 0.001000 Time 0.020539 -2023-02-13 17:43:46,105 - Epoch: [57][ 420/ 1207] Overall Loss 0.348042 Objective Loss 0.348042 LR 0.001000 Time 0.020498 -2023-02-13 17:43:46,294 - Epoch: [57][ 430/ 1207] Overall Loss 0.348038 Objective Loss 0.348038 LR 0.001000 Time 0.020459 -2023-02-13 17:43:46,482 - Epoch: [57][ 440/ 1207] Overall Loss 0.348245 Objective Loss 0.348245 LR 0.001000 Time 0.020421 -2023-02-13 17:43:46,671 - Epoch: [57][ 450/ 1207] Overall Loss 0.347810 Objective Loss 0.347810 LR 0.001000 Time 0.020386 -2023-02-13 17:43:46,859 - Epoch: [57][ 460/ 1207] Overall Loss 0.347916 Objective Loss 0.347916 LR 0.001000 Time 0.020352 -2023-02-13 17:43:47,047 - Epoch: [57][ 470/ 1207] Overall Loss 0.348466 Objective Loss 0.348466 LR 0.001000 Time 0.020318 -2023-02-13 17:43:47,236 - Epoch: [57][ 480/ 1207] Overall Loss 0.348344 Objective Loss 0.348344 LR 0.001000 Time 0.020287 -2023-02-13 17:43:47,424 - Epoch: [57][ 490/ 1207] Overall Loss 0.348069 Objective Loss 0.348069 LR 0.001000 Time 0.020256 -2023-02-13 17:43:47,612 - Epoch: [57][ 500/ 1207] Overall Loss 0.347846 Objective Loss 0.347846 LR 0.001000 Time 0.020227 -2023-02-13 17:43:47,801 - Epoch: [57][ 510/ 1207] Overall Loss 0.347505 Objective Loss 0.347505 LR 0.001000 Time 0.020200 -2023-02-13 17:43:47,989 - Epoch: [57][ 520/ 1207] Overall Loss 0.347303 Objective Loss 0.347303 LR 0.001000 Time 0.020172 -2023-02-13 17:43:48,178 - Epoch: [57][ 530/ 1207] Overall Loss 0.347368 Objective Loss 0.347368 LR 0.001000 Time 0.020147 -2023-02-13 17:43:48,366 - Epoch: [57][ 540/ 1207] Overall Loss 0.347547 Objective Loss 0.347547 LR 0.001000 Time 0.020122 -2023-02-13 17:43:48,554 - Epoch: [57][ 550/ 1207] Overall Loss 0.347801 Objective Loss 0.347801 LR 0.001000 Time 0.020098 -2023-02-13 17:43:48,743 - Epoch: [57][ 560/ 1207] Overall Loss 0.347666 Objective Loss 0.347666 LR 0.001000 Time 0.020074 -2023-02-13 17:43:48,931 - Epoch: [57][ 570/ 1207] Overall Loss 0.347641 Objective Loss 0.347641 LR 0.001000 Time 0.020052 -2023-02-13 17:43:49,119 - Epoch: [57][ 580/ 1207] Overall Loss 0.348090 Objective Loss 0.348090 LR 0.001000 Time 0.020030 -2023-02-13 17:43:49,308 - Epoch: [57][ 590/ 1207] Overall Loss 0.348639 Objective Loss 0.348639 LR 0.001000 Time 0.020010 -2023-02-13 17:43:49,496 - Epoch: [57][ 600/ 1207] Overall Loss 0.348970 Objective Loss 0.348970 LR 0.001000 Time 0.019989 -2023-02-13 17:43:49,684 - Epoch: [57][ 610/ 1207] Overall Loss 0.348899 Objective Loss 0.348899 LR 0.001000 Time 0.019970 -2023-02-13 17:43:49,873 - Epoch: [57][ 620/ 1207] Overall Loss 0.348755 Objective Loss 0.348755 LR 0.001000 Time 0.019951 -2023-02-13 17:43:50,061 - Epoch: [57][ 630/ 1207] Overall Loss 0.348599 Objective Loss 0.348599 LR 0.001000 Time 0.019933 -2023-02-13 17:43:50,251 - Epoch: [57][ 640/ 1207] Overall Loss 0.348526 Objective Loss 0.348526 LR 0.001000 Time 0.019917 -2023-02-13 17:43:50,439 - Epoch: [57][ 650/ 1207] Overall Loss 0.348772 Objective Loss 0.348772 LR 0.001000 Time 0.019900 -2023-02-13 17:43:50,628 - Epoch: [57][ 660/ 1207] Overall Loss 0.348951 Objective Loss 0.348951 LR 0.001000 Time 0.019883 -2023-02-13 17:43:50,817 - Epoch: [57][ 670/ 1207] Overall Loss 0.349185 Objective Loss 0.349185 LR 0.001000 Time 0.019869 -2023-02-13 17:43:51,005 - Epoch: [57][ 680/ 1207] Overall Loss 0.349439 Objective Loss 0.349439 LR 0.001000 Time 0.019853 -2023-02-13 17:43:51,195 - Epoch: [57][ 690/ 1207] Overall Loss 0.349756 Objective Loss 0.349756 LR 0.001000 Time 0.019839 -2023-02-13 17:43:51,383 - Epoch: [57][ 700/ 1207] Overall Loss 0.349093 Objective Loss 0.349093 LR 0.001000 Time 0.019824 -2023-02-13 17:43:51,571 - Epoch: [57][ 710/ 1207] Overall Loss 0.349202 Objective Loss 0.349202 LR 0.001000 Time 0.019809 -2023-02-13 17:43:51,759 - Epoch: [57][ 720/ 1207] Overall Loss 0.349588 Objective Loss 0.349588 LR 0.001000 Time 0.019795 -2023-02-13 17:43:51,948 - Epoch: [57][ 730/ 1207] Overall Loss 0.349785 Objective Loss 0.349785 LR 0.001000 Time 0.019782 -2023-02-13 17:43:52,136 - Epoch: [57][ 740/ 1207] Overall Loss 0.349672 Objective Loss 0.349672 LR 0.001000 Time 0.019768 -2023-02-13 17:43:52,325 - Epoch: [57][ 750/ 1207] Overall Loss 0.349491 Objective Loss 0.349491 LR 0.001000 Time 0.019756 -2023-02-13 17:43:52,512 - Epoch: [57][ 760/ 1207] Overall Loss 0.349262 Objective Loss 0.349262 LR 0.001000 Time 0.019742 -2023-02-13 17:43:52,701 - Epoch: [57][ 770/ 1207] Overall Loss 0.349410 Objective Loss 0.349410 LR 0.001000 Time 0.019730 -2023-02-13 17:43:52,889 - Epoch: [57][ 780/ 1207] Overall Loss 0.349352 Objective Loss 0.349352 LR 0.001000 Time 0.019717 -2023-02-13 17:43:53,077 - Epoch: [57][ 790/ 1207] Overall Loss 0.349086 Objective Loss 0.349086 LR 0.001000 Time 0.019705 -2023-02-13 17:43:53,265 - Epoch: [57][ 800/ 1207] Overall Loss 0.349006 Objective Loss 0.349006 LR 0.001000 Time 0.019694 -2023-02-13 17:43:53,453 - Epoch: [57][ 810/ 1207] Overall Loss 0.348809 Objective Loss 0.348809 LR 0.001000 Time 0.019682 -2023-02-13 17:43:53,641 - Epoch: [57][ 820/ 1207] Overall Loss 0.349022 Objective Loss 0.349022 LR 0.001000 Time 0.019671 -2023-02-13 17:43:53,829 - Epoch: [57][ 830/ 1207] Overall Loss 0.349059 Objective Loss 0.349059 LR 0.001000 Time 0.019661 -2023-02-13 17:43:54,018 - Epoch: [57][ 840/ 1207] Overall Loss 0.349420 Objective Loss 0.349420 LR 0.001000 Time 0.019651 -2023-02-13 17:43:54,207 - Epoch: [57][ 850/ 1207] Overall Loss 0.349694 Objective Loss 0.349694 LR 0.001000 Time 0.019642 -2023-02-13 17:43:54,395 - Epoch: [57][ 860/ 1207] Overall Loss 0.349737 Objective Loss 0.349737 LR 0.001000 Time 0.019632 -2023-02-13 17:43:54,583 - Epoch: [57][ 870/ 1207] Overall Loss 0.349765 Objective Loss 0.349765 LR 0.001000 Time 0.019622 -2023-02-13 17:43:54,772 - Epoch: [57][ 880/ 1207] Overall Loss 0.349880 Objective Loss 0.349880 LR 0.001000 Time 0.019613 -2023-02-13 17:43:54,960 - Epoch: [57][ 890/ 1207] Overall Loss 0.350060 Objective Loss 0.350060 LR 0.001000 Time 0.019603 -2023-02-13 17:43:55,149 - Epoch: [57][ 900/ 1207] Overall Loss 0.350646 Objective Loss 0.350646 LR 0.001000 Time 0.019595 -2023-02-13 17:43:55,338 - Epoch: [57][ 910/ 1207] Overall Loss 0.350859 Objective Loss 0.350859 LR 0.001000 Time 0.019587 -2023-02-13 17:43:55,526 - Epoch: [57][ 920/ 1207] Overall Loss 0.350976 Objective Loss 0.350976 LR 0.001000 Time 0.019578 -2023-02-13 17:43:55,715 - Epoch: [57][ 930/ 1207] Overall Loss 0.351009 Objective Loss 0.351009 LR 0.001000 Time 0.019570 -2023-02-13 17:43:55,905 - Epoch: [57][ 940/ 1207] Overall Loss 0.351147 Objective Loss 0.351147 LR 0.001000 Time 0.019564 -2023-02-13 17:43:56,093 - Epoch: [57][ 950/ 1207] Overall Loss 0.351029 Objective Loss 0.351029 LR 0.001000 Time 0.019556 -2023-02-13 17:43:56,282 - Epoch: [57][ 960/ 1207] Overall Loss 0.350902 Objective Loss 0.350902 LR 0.001000 Time 0.019548 -2023-02-13 17:43:56,469 - Epoch: [57][ 970/ 1207] Overall Loss 0.350819 Objective Loss 0.350819 LR 0.001000 Time 0.019539 -2023-02-13 17:43:56,658 - Epoch: [57][ 980/ 1207] Overall Loss 0.351037 Objective Loss 0.351037 LR 0.001000 Time 0.019532 -2023-02-13 17:43:56,848 - Epoch: [57][ 990/ 1207] Overall Loss 0.350888 Objective Loss 0.350888 LR 0.001000 Time 0.019527 -2023-02-13 17:43:57,037 - Epoch: [57][ 1000/ 1207] Overall Loss 0.351208 Objective Loss 0.351208 LR 0.001000 Time 0.019520 -2023-02-13 17:43:57,226 - Epoch: [57][ 1010/ 1207] Overall Loss 0.351202 Objective Loss 0.351202 LR 0.001000 Time 0.019513 -2023-02-13 17:43:57,414 - Epoch: [57][ 1020/ 1207] Overall Loss 0.351445 Objective Loss 0.351445 LR 0.001000 Time 0.019506 -2023-02-13 17:43:57,602 - Epoch: [57][ 1030/ 1207] Overall Loss 0.351671 Objective Loss 0.351671 LR 0.001000 Time 0.019499 -2023-02-13 17:43:57,791 - Epoch: [57][ 1040/ 1207] Overall Loss 0.351673 Objective Loss 0.351673 LR 0.001000 Time 0.019493 -2023-02-13 17:43:57,980 - Epoch: [57][ 1050/ 1207] Overall Loss 0.351718 Objective Loss 0.351718 LR 0.001000 Time 0.019487 -2023-02-13 17:43:58,168 - Epoch: [57][ 1060/ 1207] Overall Loss 0.351903 Objective Loss 0.351903 LR 0.001000 Time 0.019480 -2023-02-13 17:43:58,357 - Epoch: [57][ 1070/ 1207] Overall Loss 0.352004 Objective Loss 0.352004 LR 0.001000 Time 0.019474 -2023-02-13 17:43:58,545 - Epoch: [57][ 1080/ 1207] Overall Loss 0.352357 Objective Loss 0.352357 LR 0.001000 Time 0.019468 -2023-02-13 17:43:58,734 - Epoch: [57][ 1090/ 1207] Overall Loss 0.352633 Objective Loss 0.352633 LR 0.001000 Time 0.019462 -2023-02-13 17:43:58,922 - Epoch: [57][ 1100/ 1207] Overall Loss 0.352822 Objective Loss 0.352822 LR 0.001000 Time 0.019456 -2023-02-13 17:43:59,111 - Epoch: [57][ 1110/ 1207] Overall Loss 0.352956 Objective Loss 0.352956 LR 0.001000 Time 0.019450 -2023-02-13 17:43:59,299 - Epoch: [57][ 1120/ 1207] Overall Loss 0.352906 Objective Loss 0.352906 LR 0.001000 Time 0.019444 -2023-02-13 17:43:59,487 - Epoch: [57][ 1130/ 1207] Overall Loss 0.353235 Objective Loss 0.353235 LR 0.001000 Time 0.019438 -2023-02-13 17:43:59,675 - Epoch: [57][ 1140/ 1207] Overall Loss 0.353216 Objective Loss 0.353216 LR 0.001000 Time 0.019432 -2023-02-13 17:43:59,864 - Epoch: [57][ 1150/ 1207] Overall Loss 0.353110 Objective Loss 0.353110 LR 0.001000 Time 0.019427 -2023-02-13 17:44:00,053 - Epoch: [57][ 1160/ 1207] Overall Loss 0.353184 Objective Loss 0.353184 LR 0.001000 Time 0.019422 -2023-02-13 17:44:00,242 - Epoch: [57][ 1170/ 1207] Overall Loss 0.353056 Objective Loss 0.353056 LR 0.001000 Time 0.019417 -2023-02-13 17:44:00,430 - Epoch: [57][ 1180/ 1207] Overall Loss 0.353157 Objective Loss 0.353157 LR 0.001000 Time 0.019411 -2023-02-13 17:44:00,618 - Epoch: [57][ 1190/ 1207] Overall Loss 0.353202 Objective Loss 0.353202 LR 0.001000 Time 0.019406 -2023-02-13 17:44:00,859 - Epoch: [57][ 1200/ 1207] Overall Loss 0.353220 Objective Loss 0.353220 LR 0.001000 Time 0.019445 -2023-02-13 17:44:00,974 - Epoch: [57][ 1207/ 1207] Overall Loss 0.353005 Objective Loss 0.353005 Top1 83.536585 Top5 97.865854 LR 0.001000 Time 0.019428 -2023-02-13 17:44:01,045 - --- validate (epoch=57)----------- -2023-02-13 17:44:01,045 - 34311 samples (256 per mini-batch) -2023-02-13 17:44:01,438 - Epoch: [57][ 10/ 135] Loss 0.389697 Top1 81.953125 Top5 97.031250 -2023-02-13 17:44:01,565 - Epoch: [57][ 20/ 135] Loss 0.393263 Top1 82.324219 Top5 97.031250 -2023-02-13 17:44:01,698 - Epoch: [57][ 30/ 135] Loss 0.385360 Top1 82.669271 Top5 97.070312 -2023-02-13 17:44:01,838 - Epoch: [57][ 40/ 135] Loss 0.384001 Top1 82.441406 Top5 97.109375 -2023-02-13 17:44:01,965 - Epoch: [57][ 50/ 135] Loss 0.382935 Top1 82.554688 Top5 97.070312 -2023-02-13 17:44:02,098 - Epoch: [57][ 60/ 135] Loss 0.381489 Top1 82.597656 Top5 97.128906 -2023-02-13 17:44:02,226 - Epoch: [57][ 70/ 135] Loss 0.379915 Top1 82.645089 Top5 97.114955 -2023-02-13 17:44:02,359 - Epoch: [57][ 80/ 135] Loss 0.380584 Top1 82.500000 Top5 97.119141 -2023-02-13 17:44:02,488 - Epoch: [57][ 90/ 135] Loss 0.377904 Top1 82.482639 Top5 97.126736 -2023-02-13 17:44:02,620 - Epoch: [57][ 100/ 135] Loss 0.377322 Top1 82.570312 Top5 97.128906 -2023-02-13 17:44:02,746 - Epoch: [57][ 110/ 135] Loss 0.376225 Top1 82.531960 Top5 97.084517 -2023-02-13 17:44:02,873 - Epoch: [57][ 120/ 135] Loss 0.376632 Top1 82.574870 Top5 97.073568 -2023-02-13 17:44:03,005 - Epoch: [57][ 130/ 135] Loss 0.375593 Top1 82.521034 Top5 97.097356 -2023-02-13 17:44:03,053 - Epoch: [57][ 135/ 135] Loss 0.375782 Top1 82.498324 Top5 97.120457 -2023-02-13 17:44:03,121 - ==> Top1: 82.498 Top5: 97.120 Loss: 0.376 - -2023-02-13 17:44:03,122 - ==> Confusion: -[[ 830 6 5 1 17 5 0 3 4 59 2 5 1 4 3 5 4 4 0 0 9] - [ 6 898 0 3 12 39 4 26 3 1 5 2 5 0 3 0 1 3 10 1 11] - [ 10 3 936 9 5 1 12 20 0 0 4 1 2 6 2 13 1 5 11 4 13] - [ 4 4 21 902 2 4 0 1 3 0 17 2 7 0 18 1 2 2 16 0 10] - [ 21 10 0 0 982 11 1 2 0 2 0 6 1 3 4 8 7 1 1 2 4] - [ 5 18 3 5 5 948 1 16 1 3 4 15 8 11 2 3 4 3 4 6 5] - [ 4 4 27 1 1 7 1013 3 0 0 7 3 3 2 0 9 0 3 1 4 7] - [ 1 10 8 3 4 33 1 900 1 2 0 7 3 0 1 0 2 3 31 8 6] - [ 19 4 0 1 0 1 0 0 880 46 13 1 0 9 24 3 1 1 5 0 1] - [ 78 2 3 1 9 3 0 3 29 841 1 1 0 24 2 2 3 1 2 1 6] - [ 3 3 4 4 1 3 3 5 18 2 960 2 5 10 3 0 1 0 18 1 5] - [ 2 3 2 0 4 9 1 6 1 0 0 880 52 4 2 12 3 11 1 9 3] - [ 4 2 1 9 2 1 1 0 2 0 1 29 850 1 4 10 2 19 0 4 17] - [ 9 3 3 0 10 13 0 1 22 14 11 12 4 894 5 7 5 1 1 3 6] - [ 10 1 2 15 9 3 1 2 17 8 4 1 1 0 993 1 1 4 10 0 9] - [ 6 1 5 0 8 3 3 1 0 0 0 3 8 4 0 976 6 6 1 6 9] - [ 3 4 0 0 10 1 0 0 2 0 0 2 3 0 3 14 997 2 3 2 15] - [ 10 2 0 8 1 1 1 1 0 1 0 9 30 0 2 31 0 949 0 2 3] - [ 2 4 2 19 3 2 0 29 2 0 1 1 8 0 20 2 0 3 983 2 3] - [ 0 6 0 0 1 9 9 17 1 0 1 13 7 3 0 11 7 2 2 1048 11] - [ 208 238 190 149 150 211 61 186 109 100 207 122 326 314 205 148 246 109 230 278 9647]] - -2023-02-13 17:44:03,123 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:44:03,124 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:44:03,130 - - -2023-02-13 17:44:03,130 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:44:04,019 - Epoch: [58][ 10/ 1207] Overall Loss 0.314860 Objective Loss 0.314860 LR 0.001000 Time 0.088860 -2023-02-13 17:44:04,218 - Epoch: [58][ 20/ 1207] Overall Loss 0.339632 Objective Loss 0.339632 LR 0.001000 Time 0.054354 -2023-02-13 17:44:04,413 - Epoch: [58][ 30/ 1207] Overall Loss 0.352479 Objective Loss 0.352479 LR 0.001000 Time 0.042716 -2023-02-13 17:44:04,608 - Epoch: [58][ 40/ 1207] Overall Loss 0.362117 Objective Loss 0.362117 LR 0.001000 Time 0.036898 -2023-02-13 17:44:04,799 - Epoch: [58][ 50/ 1207] Overall Loss 0.358886 Objective Loss 0.358886 LR 0.001000 Time 0.033337 -2023-02-13 17:44:04,988 - Epoch: [58][ 60/ 1207] Overall Loss 0.357846 Objective Loss 0.357846 LR 0.001000 Time 0.030913 -2023-02-13 17:44:05,176 - Epoch: [58][ 70/ 1207] Overall Loss 0.358134 Objective Loss 0.358134 LR 0.001000 Time 0.029185 -2023-02-13 17:44:05,365 - Epoch: [58][ 80/ 1207] Overall Loss 0.356030 Objective Loss 0.356030 LR 0.001000 Time 0.027889 -2023-02-13 17:44:05,552 - Epoch: [58][ 90/ 1207] Overall Loss 0.355383 Objective Loss 0.355383 LR 0.001000 Time 0.026873 -2023-02-13 17:44:05,741 - Epoch: [58][ 100/ 1207] Overall Loss 0.357018 Objective Loss 0.357018 LR 0.001000 Time 0.026069 -2023-02-13 17:44:05,931 - Epoch: [58][ 110/ 1207] Overall Loss 0.356138 Objective Loss 0.356138 LR 0.001000 Time 0.025421 -2023-02-13 17:44:06,120 - Epoch: [58][ 120/ 1207] Overall Loss 0.354852 Objective Loss 0.354852 LR 0.001000 Time 0.024872 -2023-02-13 17:44:06,309 - Epoch: [58][ 130/ 1207] Overall Loss 0.354461 Objective Loss 0.354461 LR 0.001000 Time 0.024415 -2023-02-13 17:44:06,498 - Epoch: [58][ 140/ 1207] Overall Loss 0.354706 Objective Loss 0.354706 LR 0.001000 Time 0.024013 -2023-02-13 17:44:06,687 - Epoch: [58][ 150/ 1207] Overall Loss 0.352961 Objective Loss 0.352961 LR 0.001000 Time 0.023671 -2023-02-13 17:44:06,876 - Epoch: [58][ 160/ 1207] Overall Loss 0.351210 Objective Loss 0.351210 LR 0.001000 Time 0.023373 -2023-02-13 17:44:07,065 - Epoch: [58][ 170/ 1207] Overall Loss 0.352533 Objective Loss 0.352533 LR 0.001000 Time 0.023106 -2023-02-13 17:44:07,254 - Epoch: [58][ 180/ 1207] Overall Loss 0.351501 Objective Loss 0.351501 LR 0.001000 Time 0.022871 -2023-02-13 17:44:07,443 - Epoch: [58][ 190/ 1207] Overall Loss 0.351822 Objective Loss 0.351822 LR 0.001000 Time 0.022660 -2023-02-13 17:44:07,632 - Epoch: [58][ 200/ 1207] Overall Loss 0.351966 Objective Loss 0.351966 LR 0.001000 Time 0.022470 -2023-02-13 17:44:07,821 - Epoch: [58][ 210/ 1207] Overall Loss 0.350509 Objective Loss 0.350509 LR 0.001000 Time 0.022299 -2023-02-13 17:44:08,010 - Epoch: [58][ 220/ 1207] Overall Loss 0.350874 Objective Loss 0.350874 LR 0.001000 Time 0.022143 -2023-02-13 17:44:08,200 - Epoch: [58][ 230/ 1207] Overall Loss 0.348878 Objective Loss 0.348878 LR 0.001000 Time 0.022004 -2023-02-13 17:44:08,389 - Epoch: [58][ 240/ 1207] Overall Loss 0.347720 Objective Loss 0.347720 LR 0.001000 Time 0.021875 -2023-02-13 17:44:08,578 - Epoch: [58][ 250/ 1207] Overall Loss 0.347985 Objective Loss 0.347985 LR 0.001000 Time 0.021754 -2023-02-13 17:44:08,768 - Epoch: [58][ 260/ 1207] Overall Loss 0.348677 Objective Loss 0.348677 LR 0.001000 Time 0.021646 -2023-02-13 17:44:08,957 - Epoch: [58][ 270/ 1207] Overall Loss 0.348507 Objective Loss 0.348507 LR 0.001000 Time 0.021543 -2023-02-13 17:44:09,146 - Epoch: [58][ 280/ 1207] Overall Loss 0.348194 Objective Loss 0.348194 LR 0.001000 Time 0.021447 -2023-02-13 17:44:09,336 - Epoch: [58][ 290/ 1207] Overall Loss 0.348245 Objective Loss 0.348245 LR 0.001000 Time 0.021361 -2023-02-13 17:44:09,525 - Epoch: [58][ 300/ 1207] Overall Loss 0.347461 Objective Loss 0.347461 LR 0.001000 Time 0.021278 -2023-02-13 17:44:09,714 - Epoch: [58][ 310/ 1207] Overall Loss 0.347968 Objective Loss 0.347968 LR 0.001000 Time 0.021200 -2023-02-13 17:44:09,903 - Epoch: [58][ 320/ 1207] Overall Loss 0.346858 Objective Loss 0.346858 LR 0.001000 Time 0.021127 -2023-02-13 17:44:10,093 - Epoch: [58][ 330/ 1207] Overall Loss 0.346331 Objective Loss 0.346331 LR 0.001000 Time 0.021060 -2023-02-13 17:44:10,282 - Epoch: [58][ 340/ 1207] Overall Loss 0.346662 Objective Loss 0.346662 LR 0.001000 Time 0.020996 -2023-02-13 17:44:10,471 - Epoch: [58][ 350/ 1207] Overall Loss 0.347333 Objective Loss 0.347333 LR 0.001000 Time 0.020935 -2023-02-13 17:44:10,660 - Epoch: [58][ 360/ 1207] Overall Loss 0.346722 Objective Loss 0.346722 LR 0.001000 Time 0.020878 -2023-02-13 17:44:10,850 - Epoch: [58][ 370/ 1207] Overall Loss 0.346378 Objective Loss 0.346378 LR 0.001000 Time 0.020826 -2023-02-13 17:44:11,039 - Epoch: [58][ 380/ 1207] Overall Loss 0.346221 Objective Loss 0.346221 LR 0.001000 Time 0.020774 -2023-02-13 17:44:11,228 - Epoch: [58][ 390/ 1207] Overall Loss 0.346184 Objective Loss 0.346184 LR 0.001000 Time 0.020726 -2023-02-13 17:44:11,418 - Epoch: [58][ 400/ 1207] Overall Loss 0.346878 Objective Loss 0.346878 LR 0.001000 Time 0.020680 -2023-02-13 17:44:11,607 - Epoch: [58][ 410/ 1207] Overall Loss 0.347233 Objective Loss 0.347233 LR 0.001000 Time 0.020637 -2023-02-13 17:44:11,798 - Epoch: [58][ 420/ 1207] Overall Loss 0.348087 Objective Loss 0.348087 LR 0.001000 Time 0.020600 -2023-02-13 17:44:11,987 - Epoch: [58][ 430/ 1207] Overall Loss 0.347738 Objective Loss 0.347738 LR 0.001000 Time 0.020559 -2023-02-13 17:44:12,176 - Epoch: [58][ 440/ 1207] Overall Loss 0.347704 Objective Loss 0.347704 LR 0.001000 Time 0.020521 -2023-02-13 17:44:12,367 - Epoch: [58][ 450/ 1207] Overall Loss 0.347804 Objective Loss 0.347804 LR 0.001000 Time 0.020487 -2023-02-13 17:44:12,556 - Epoch: [58][ 460/ 1207] Overall Loss 0.347554 Objective Loss 0.347554 LR 0.001000 Time 0.020452 -2023-02-13 17:44:12,746 - Epoch: [58][ 470/ 1207] Overall Loss 0.347532 Objective Loss 0.347532 LR 0.001000 Time 0.020420 -2023-02-13 17:44:12,935 - Epoch: [58][ 480/ 1207] Overall Loss 0.347654 Objective Loss 0.347654 LR 0.001000 Time 0.020388 -2023-02-13 17:44:13,124 - Epoch: [58][ 490/ 1207] Overall Loss 0.347901 Objective Loss 0.347901 LR 0.001000 Time 0.020358 -2023-02-13 17:44:13,314 - Epoch: [58][ 500/ 1207] Overall Loss 0.348408 Objective Loss 0.348408 LR 0.001000 Time 0.020330 -2023-02-13 17:44:13,503 - Epoch: [58][ 510/ 1207] Overall Loss 0.347903 Objective Loss 0.347903 LR 0.001000 Time 0.020301 -2023-02-13 17:44:13,693 - Epoch: [58][ 520/ 1207] Overall Loss 0.348006 Objective Loss 0.348006 LR 0.001000 Time 0.020276 -2023-02-13 17:44:13,883 - Epoch: [58][ 530/ 1207] Overall Loss 0.348053 Objective Loss 0.348053 LR 0.001000 Time 0.020250 -2023-02-13 17:44:14,072 - Epoch: [58][ 540/ 1207] Overall Loss 0.347707 Objective Loss 0.347707 LR 0.001000 Time 0.020225 -2023-02-13 17:44:14,262 - Epoch: [58][ 550/ 1207] Overall Loss 0.347349 Objective Loss 0.347349 LR 0.001000 Time 0.020202 -2023-02-13 17:44:14,451 - Epoch: [58][ 560/ 1207] Overall Loss 0.347604 Objective Loss 0.347604 LR 0.001000 Time 0.020179 -2023-02-13 17:44:14,641 - Epoch: [58][ 570/ 1207] Overall Loss 0.346932 Objective Loss 0.346932 LR 0.001000 Time 0.020156 -2023-02-13 17:44:14,830 - Epoch: [58][ 580/ 1207] Overall Loss 0.346626 Objective Loss 0.346626 LR 0.001000 Time 0.020135 -2023-02-13 17:44:15,019 - Epoch: [58][ 590/ 1207] Overall Loss 0.346297 Objective Loss 0.346297 LR 0.001000 Time 0.020113 -2023-02-13 17:44:15,208 - Epoch: [58][ 600/ 1207] Overall Loss 0.346160 Objective Loss 0.346160 LR 0.001000 Time 0.020092 -2023-02-13 17:44:15,398 - Epoch: [58][ 610/ 1207] Overall Loss 0.345834 Objective Loss 0.345834 LR 0.001000 Time 0.020073 -2023-02-13 17:44:15,587 - Epoch: [58][ 620/ 1207] Overall Loss 0.345864 Objective Loss 0.345864 LR 0.001000 Time 0.020054 -2023-02-13 17:44:15,776 - Epoch: [58][ 630/ 1207] Overall Loss 0.346500 Objective Loss 0.346500 LR 0.001000 Time 0.020035 -2023-02-13 17:44:15,966 - Epoch: [58][ 640/ 1207] Overall Loss 0.346397 Objective Loss 0.346397 LR 0.001000 Time 0.020019 -2023-02-13 17:44:16,155 - Epoch: [58][ 650/ 1207] Overall Loss 0.346849 Objective Loss 0.346849 LR 0.001000 Time 0.020000 -2023-02-13 17:44:16,345 - Epoch: [58][ 660/ 1207] Overall Loss 0.346657 Objective Loss 0.346657 LR 0.001000 Time 0.019985 -2023-02-13 17:44:16,535 - Epoch: [58][ 670/ 1207] Overall Loss 0.346685 Objective Loss 0.346685 LR 0.001000 Time 0.019969 -2023-02-13 17:44:16,725 - Epoch: [58][ 680/ 1207] Overall Loss 0.346405 Objective Loss 0.346405 LR 0.001000 Time 0.019954 -2023-02-13 17:44:16,915 - Epoch: [58][ 690/ 1207] Overall Loss 0.347026 Objective Loss 0.347026 LR 0.001000 Time 0.019940 -2023-02-13 17:44:17,104 - Epoch: [58][ 700/ 1207] Overall Loss 0.346917 Objective Loss 0.346917 LR 0.001000 Time 0.019925 -2023-02-13 17:44:17,294 - Epoch: [58][ 710/ 1207] Overall Loss 0.346384 Objective Loss 0.346384 LR 0.001000 Time 0.019911 -2023-02-13 17:44:17,483 - Epoch: [58][ 720/ 1207] Overall Loss 0.346577 Objective Loss 0.346577 LR 0.001000 Time 0.019897 -2023-02-13 17:44:17,672 - Epoch: [58][ 730/ 1207] Overall Loss 0.346595 Objective Loss 0.346595 LR 0.001000 Time 0.019883 -2023-02-13 17:44:17,863 - Epoch: [58][ 740/ 1207] Overall Loss 0.346944 Objective Loss 0.346944 LR 0.001000 Time 0.019871 -2023-02-13 17:44:18,052 - Epoch: [58][ 750/ 1207] Overall Loss 0.346470 Objective Loss 0.346470 LR 0.001000 Time 0.019858 -2023-02-13 17:44:18,243 - Epoch: [58][ 760/ 1207] Overall Loss 0.345989 Objective Loss 0.345989 LR 0.001000 Time 0.019848 -2023-02-13 17:44:18,433 - Epoch: [58][ 770/ 1207] Overall Loss 0.345832 Objective Loss 0.345832 LR 0.001000 Time 0.019836 -2023-02-13 17:44:18,622 - Epoch: [58][ 780/ 1207] Overall Loss 0.345712 Objective Loss 0.345712 LR 0.001000 Time 0.019824 -2023-02-13 17:44:18,812 - Epoch: [58][ 790/ 1207] Overall Loss 0.345110 Objective Loss 0.345110 LR 0.001000 Time 0.019813 -2023-02-13 17:44:19,002 - Epoch: [58][ 800/ 1207] Overall Loss 0.344802 Objective Loss 0.344802 LR 0.001000 Time 0.019802 -2023-02-13 17:44:19,191 - Epoch: [58][ 810/ 1207] Overall Loss 0.344726 Objective Loss 0.344726 LR 0.001000 Time 0.019791 -2023-02-13 17:44:19,381 - Epoch: [58][ 820/ 1207] Overall Loss 0.345193 Objective Loss 0.345193 LR 0.001000 Time 0.019780 -2023-02-13 17:44:19,571 - Epoch: [58][ 830/ 1207] Overall Loss 0.345206 Objective Loss 0.345206 LR 0.001000 Time 0.019771 -2023-02-13 17:44:19,761 - Epoch: [58][ 840/ 1207] Overall Loss 0.345641 Objective Loss 0.345641 LR 0.001000 Time 0.019760 -2023-02-13 17:44:19,950 - Epoch: [58][ 850/ 1207] Overall Loss 0.345741 Objective Loss 0.345741 LR 0.001000 Time 0.019751 -2023-02-13 17:44:20,140 - Epoch: [58][ 860/ 1207] Overall Loss 0.345997 Objective Loss 0.345997 LR 0.001000 Time 0.019741 -2023-02-13 17:44:20,330 - Epoch: [58][ 870/ 1207] Overall Loss 0.346026 Objective Loss 0.346026 LR 0.001000 Time 0.019732 -2023-02-13 17:44:20,519 - Epoch: [58][ 880/ 1207] Overall Loss 0.346260 Objective Loss 0.346260 LR 0.001000 Time 0.019722 -2023-02-13 17:44:20,709 - Epoch: [58][ 890/ 1207] Overall Loss 0.346625 Objective Loss 0.346625 LR 0.001000 Time 0.019714 -2023-02-13 17:44:20,900 - Epoch: [58][ 900/ 1207] Overall Loss 0.347054 Objective Loss 0.347054 LR 0.001000 Time 0.019707 -2023-02-13 17:44:21,090 - Epoch: [58][ 910/ 1207] Overall Loss 0.346893 Objective Loss 0.346893 LR 0.001000 Time 0.019698 -2023-02-13 17:44:21,281 - Epoch: [58][ 920/ 1207] Overall Loss 0.346875 Objective Loss 0.346875 LR 0.001000 Time 0.019691 -2023-02-13 17:44:21,470 - Epoch: [58][ 930/ 1207] Overall Loss 0.346981 Objective Loss 0.346981 LR 0.001000 Time 0.019683 -2023-02-13 17:44:21,660 - Epoch: [58][ 940/ 1207] Overall Loss 0.347555 Objective Loss 0.347555 LR 0.001000 Time 0.019675 -2023-02-13 17:44:21,850 - Epoch: [58][ 950/ 1207] Overall Loss 0.347749 Objective Loss 0.347749 LR 0.001000 Time 0.019668 -2023-02-13 17:44:22,040 - Epoch: [58][ 960/ 1207] Overall Loss 0.348056 Objective Loss 0.348056 LR 0.001000 Time 0.019660 -2023-02-13 17:44:22,229 - Epoch: [58][ 970/ 1207] Overall Loss 0.348328 Objective Loss 0.348328 LR 0.001000 Time 0.019652 -2023-02-13 17:44:22,419 - Epoch: [58][ 980/ 1207] Overall Loss 0.348317 Objective Loss 0.348317 LR 0.001000 Time 0.019645 -2023-02-13 17:44:22,608 - Epoch: [58][ 990/ 1207] Overall Loss 0.348526 Objective Loss 0.348526 LR 0.001000 Time 0.019637 -2023-02-13 17:44:22,798 - Epoch: [58][ 1000/ 1207] Overall Loss 0.349005 Objective Loss 0.349005 LR 0.001000 Time 0.019630 -2023-02-13 17:44:22,988 - Epoch: [58][ 1010/ 1207] Overall Loss 0.349473 Objective Loss 0.349473 LR 0.001000 Time 0.019624 -2023-02-13 17:44:23,178 - Epoch: [58][ 1020/ 1207] Overall Loss 0.349652 Objective Loss 0.349652 LR 0.001000 Time 0.019617 -2023-02-13 17:44:23,367 - Epoch: [58][ 1030/ 1207] Overall Loss 0.349919 Objective Loss 0.349919 LR 0.001000 Time 0.019610 -2023-02-13 17:44:23,557 - Epoch: [58][ 1040/ 1207] Overall Loss 0.350177 Objective Loss 0.350177 LR 0.001000 Time 0.019604 -2023-02-13 17:44:23,747 - Epoch: [58][ 1050/ 1207] Overall Loss 0.350191 Objective Loss 0.350191 LR 0.001000 Time 0.019598 -2023-02-13 17:44:23,937 - Epoch: [58][ 1060/ 1207] Overall Loss 0.350326 Objective Loss 0.350326 LR 0.001000 Time 0.019592 -2023-02-13 17:44:24,127 - Epoch: [58][ 1070/ 1207] Overall Loss 0.350716 Objective Loss 0.350716 LR 0.001000 Time 0.019586 -2023-02-13 17:44:24,317 - Epoch: [58][ 1080/ 1207] Overall Loss 0.351010 Objective Loss 0.351010 LR 0.001000 Time 0.019580 -2023-02-13 17:44:24,507 - Epoch: [58][ 1090/ 1207] Overall Loss 0.351360 Objective Loss 0.351360 LR 0.001000 Time 0.019574 -2023-02-13 17:44:24,698 - Epoch: [58][ 1100/ 1207] Overall Loss 0.351635 Objective Loss 0.351635 LR 0.001000 Time 0.019569 -2023-02-13 17:44:24,888 - Epoch: [58][ 1110/ 1207] Overall Loss 0.351979 Objective Loss 0.351979 LR 0.001000 Time 0.019564 -2023-02-13 17:44:25,078 - Epoch: [58][ 1120/ 1207] Overall Loss 0.352086 Objective Loss 0.352086 LR 0.001000 Time 0.019558 -2023-02-13 17:44:25,267 - Epoch: [58][ 1130/ 1207] Overall Loss 0.352133 Objective Loss 0.352133 LR 0.001000 Time 0.019553 -2023-02-13 17:44:25,458 - Epoch: [58][ 1140/ 1207] Overall Loss 0.352327 Objective Loss 0.352327 LR 0.001000 Time 0.019548 -2023-02-13 17:44:25,648 - Epoch: [58][ 1150/ 1207] Overall Loss 0.352328 Objective Loss 0.352328 LR 0.001000 Time 0.019543 -2023-02-13 17:44:25,839 - Epoch: [58][ 1160/ 1207] Overall Loss 0.352261 Objective Loss 0.352261 LR 0.001000 Time 0.019539 -2023-02-13 17:44:26,029 - Epoch: [58][ 1170/ 1207] Overall Loss 0.352385 Objective Loss 0.352385 LR 0.001000 Time 0.019534 -2023-02-13 17:44:26,218 - Epoch: [58][ 1180/ 1207] Overall Loss 0.352351 Objective Loss 0.352351 LR 0.001000 Time 0.019528 -2023-02-13 17:44:26,408 - Epoch: [58][ 1190/ 1207] Overall Loss 0.352587 Objective Loss 0.352587 LR 0.001000 Time 0.019524 -2023-02-13 17:44:26,653 - Epoch: [58][ 1200/ 1207] Overall Loss 0.352790 Objective Loss 0.352790 LR 0.001000 Time 0.019565 -2023-02-13 17:44:26,769 - Epoch: [58][ 1207/ 1207] Overall Loss 0.352661 Objective Loss 0.352661 Top1 83.231707 Top5 97.865854 LR 0.001000 Time 0.019547 -2023-02-13 17:44:26,842 - --- validate (epoch=58)----------- -2023-02-13 17:44:26,842 - 34311 samples (256 per mini-batch) -2023-02-13 17:44:27,242 - Epoch: [58][ 10/ 135] Loss 0.372862 Top1 81.718750 Top5 96.640625 -2023-02-13 17:44:27,374 - Epoch: [58][ 20/ 135] Loss 0.377244 Top1 81.308594 Top5 96.386719 -2023-02-13 17:44:27,504 - Epoch: [58][ 30/ 135] Loss 0.382825 Top1 81.302083 Top5 96.510417 -2023-02-13 17:44:27,633 - Epoch: [58][ 40/ 135] Loss 0.391717 Top1 80.517578 Top5 96.435547 -2023-02-13 17:44:27,763 - Epoch: [58][ 50/ 135] Loss 0.394015 Top1 80.640625 Top5 96.523438 -2023-02-13 17:44:27,887 - Epoch: [58][ 60/ 135] Loss 0.390248 Top1 80.878906 Top5 96.588542 -2023-02-13 17:44:28,013 - Epoch: [58][ 70/ 135] Loss 0.388623 Top1 80.820312 Top5 96.601562 -2023-02-13 17:44:28,135 - Epoch: [58][ 80/ 135] Loss 0.384466 Top1 80.917969 Top5 96.669922 -2023-02-13 17:44:28,279 - Epoch: [58][ 90/ 135] Loss 0.384021 Top1 80.928819 Top5 96.662326 -2023-02-13 17:44:28,417 - Epoch: [58][ 100/ 135] Loss 0.380219 Top1 80.910156 Top5 96.707031 -2023-02-13 17:44:28,541 - Epoch: [58][ 110/ 135] Loss 0.381797 Top1 80.763494 Top5 96.697443 -2023-02-13 17:44:28,668 - Epoch: [58][ 120/ 135] Loss 0.384275 Top1 80.628255 Top5 96.689453 -2023-02-13 17:44:28,797 - Epoch: [58][ 130/ 135] Loss 0.382367 Top1 80.682091 Top5 96.742788 -2023-02-13 17:44:28,842 - Epoch: [58][ 135/ 135] Loss 0.380758 Top1 80.665093 Top5 96.764886 -2023-02-13 17:44:28,926 - ==> Top1: 80.665 Top5: 96.765 Loss: 0.381 - -2023-02-13 17:44:28,927 - ==> Confusion: -[[ 816 4 9 2 15 8 0 3 3 81 0 2 1 2 7 2 3 3 0 0 6] - [ 3 912 1 5 9 41 3 21 4 3 3 2 0 0 6 1 2 0 10 3 4] - [ 7 6 932 15 3 4 31 15 1 1 5 1 1 3 5 5 2 5 7 4 5] - [ 8 0 17 885 2 8 1 5 1 3 17 1 5 2 24 4 3 6 16 1 7] - [ 23 11 2 0 969 11 1 2 1 6 0 7 0 9 10 3 5 1 1 1 3] - [ 1 15 2 4 5 968 4 13 1 6 2 21 4 13 4 1 1 1 1 1 2] - [ 2 5 12 2 2 8 1033 11 0 2 1 3 0 1 0 6 0 3 0 7 1] - [ 0 7 14 1 2 43 1 910 1 3 4 8 6 2 0 0 1 0 10 10 1] - [ 16 5 0 1 2 0 0 4 852 64 13 2 1 22 15 0 0 1 8 1 2] - [ 75 1 2 0 6 2 0 1 20 870 2 1 2 18 6 1 0 0 0 1 4] - [ 4 3 5 7 0 3 1 5 18 2 956 2 1 22 4 0 1 1 9 0 7] - [ 2 1 1 0 2 14 0 6 5 0 0 900 36 11 4 6 2 7 1 6 1] - [ 1 0 0 2 0 2 0 3 3 1 0 36 859 2 11 4 2 24 4 1 4] - [ 6 1 4 0 4 18 2 2 6 15 6 10 2 929 5 4 3 4 0 1 2] - [ 9 2 2 12 6 6 0 1 26 11 7 1 4 3 984 1 1 3 7 0 6] - [ 5 1 3 1 12 4 9 0 0 1 0 11 5 4 0 960 5 12 0 7 6] - [ 4 12 2 0 11 7 1 0 6 1 0 1 2 7 4 19 967 1 1 5 10] - [ 7 1 1 8 1 1 1 1 3 1 1 10 36 2 1 20 0 949 0 1 6] - [ 5 4 10 19 1 2 0 42 4 3 3 2 2 1 14 0 1 2 970 1 0] - [ 0 3 1 0 1 10 9 22 1 0 2 25 5 7 0 5 4 1 2 1038 12] - [ 168 239 248 132 127 321 136 243 131 155 226 170 422 461 206 165 270 120 184 292 9018]] - -2023-02-13 17:44:28,929 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:44:28,929 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:44:28,934 - - -2023-02-13 17:44:28,934 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:44:29,914 - Epoch: [59][ 10/ 1207] Overall Loss 0.403607 Objective Loss 0.403607 LR 0.001000 Time 0.097869 -2023-02-13 17:44:30,108 - Epoch: [59][ 20/ 1207] Overall Loss 0.388206 Objective Loss 0.388206 LR 0.001000 Time 0.058625 -2023-02-13 17:44:30,297 - Epoch: [59][ 30/ 1207] Overall Loss 0.373070 Objective Loss 0.373070 LR 0.001000 Time 0.045368 -2023-02-13 17:44:30,486 - Epoch: [59][ 40/ 1207] Overall Loss 0.361760 Objective Loss 0.361760 LR 0.001000 Time 0.038732 -2023-02-13 17:44:30,673 - Epoch: [59][ 50/ 1207] Overall Loss 0.358669 Objective Loss 0.358669 LR 0.001000 Time 0.034727 -2023-02-13 17:44:30,863 - Epoch: [59][ 60/ 1207] Overall Loss 0.364836 Objective Loss 0.364836 LR 0.001000 Time 0.032094 -2023-02-13 17:44:31,051 - Epoch: [59][ 70/ 1207] Overall Loss 0.363199 Objective Loss 0.363199 LR 0.001000 Time 0.030192 -2023-02-13 17:44:31,239 - Epoch: [59][ 80/ 1207] Overall Loss 0.362247 Objective Loss 0.362247 LR 0.001000 Time 0.028766 -2023-02-13 17:44:31,428 - Epoch: [59][ 90/ 1207] Overall Loss 0.363432 Objective Loss 0.363432 LR 0.001000 Time 0.027669 -2023-02-13 17:44:31,617 - Epoch: [59][ 100/ 1207] Overall Loss 0.361369 Objective Loss 0.361369 LR 0.001000 Time 0.026784 -2023-02-13 17:44:31,806 - Epoch: [59][ 110/ 1207] Overall Loss 0.360730 Objective Loss 0.360730 LR 0.001000 Time 0.026068 -2023-02-13 17:44:31,995 - Epoch: [59][ 120/ 1207] Overall Loss 0.362577 Objective Loss 0.362577 LR 0.001000 Time 0.025461 -2023-02-13 17:44:32,183 - Epoch: [59][ 130/ 1207] Overall Loss 0.359086 Objective Loss 0.359086 LR 0.001000 Time 0.024951 -2023-02-13 17:44:32,372 - Epoch: [59][ 140/ 1207] Overall Loss 0.358376 Objective Loss 0.358376 LR 0.001000 Time 0.024511 -2023-02-13 17:44:32,561 - Epoch: [59][ 150/ 1207] Overall Loss 0.357598 Objective Loss 0.357598 LR 0.001000 Time 0.024134 -2023-02-13 17:44:32,750 - Epoch: [59][ 160/ 1207] Overall Loss 0.356283 Objective Loss 0.356283 LR 0.001000 Time 0.023805 -2023-02-13 17:44:32,938 - Epoch: [59][ 170/ 1207] Overall Loss 0.353440 Objective Loss 0.353440 LR 0.001000 Time 0.023511 -2023-02-13 17:44:33,127 - Epoch: [59][ 180/ 1207] Overall Loss 0.352546 Objective Loss 0.352546 LR 0.001000 Time 0.023251 -2023-02-13 17:44:33,315 - Epoch: [59][ 190/ 1207] Overall Loss 0.353772 Objective Loss 0.353772 LR 0.001000 Time 0.023018 -2023-02-13 17:44:33,504 - Epoch: [59][ 200/ 1207] Overall Loss 0.353804 Objective Loss 0.353804 LR 0.001000 Time 0.022809 -2023-02-13 17:44:33,692 - Epoch: [59][ 210/ 1207] Overall Loss 0.353192 Objective Loss 0.353192 LR 0.001000 Time 0.022617 -2023-02-13 17:44:33,881 - Epoch: [59][ 220/ 1207] Overall Loss 0.354513 Objective Loss 0.354513 LR 0.001000 Time 0.022443 -2023-02-13 17:44:34,068 - Epoch: [59][ 230/ 1207] Overall Loss 0.354298 Objective Loss 0.354298 LR 0.001000 Time 0.022282 -2023-02-13 17:44:34,257 - Epoch: [59][ 240/ 1207] Overall Loss 0.354241 Objective Loss 0.354241 LR 0.001000 Time 0.022139 -2023-02-13 17:44:34,445 - Epoch: [59][ 250/ 1207] Overall Loss 0.354039 Objective Loss 0.354039 LR 0.001000 Time 0.022004 -2023-02-13 17:44:34,634 - Epoch: [59][ 260/ 1207] Overall Loss 0.354492 Objective Loss 0.354492 LR 0.001000 Time 0.021883 -2023-02-13 17:44:34,823 - Epoch: [59][ 270/ 1207] Overall Loss 0.353943 Objective Loss 0.353943 LR 0.001000 Time 0.021769 -2023-02-13 17:44:35,011 - Epoch: [59][ 280/ 1207] Overall Loss 0.351795 Objective Loss 0.351795 LR 0.001000 Time 0.021662 -2023-02-13 17:44:35,200 - Epoch: [59][ 290/ 1207] Overall Loss 0.351995 Objective Loss 0.351995 LR 0.001000 Time 0.021565 -2023-02-13 17:44:35,391 - Epoch: [59][ 300/ 1207] Overall Loss 0.351148 Objective Loss 0.351148 LR 0.001000 Time 0.021482 -2023-02-13 17:44:35,587 - Epoch: [59][ 310/ 1207] Overall Loss 0.350792 Objective Loss 0.350792 LR 0.001000 Time 0.021420 -2023-02-13 17:44:35,781 - Epoch: [59][ 320/ 1207] Overall Loss 0.350053 Objective Loss 0.350053 LR 0.001000 Time 0.021355 -2023-02-13 17:44:35,977 - Epoch: [59][ 330/ 1207] Overall Loss 0.349166 Objective Loss 0.349166 LR 0.001000 Time 0.021303 -2023-02-13 17:44:36,170 - Epoch: [59][ 340/ 1207] Overall Loss 0.348337 Objective Loss 0.348337 LR 0.001000 Time 0.021242 -2023-02-13 17:44:36,366 - Epoch: [59][ 350/ 1207] Overall Loss 0.348272 Objective Loss 0.348272 LR 0.001000 Time 0.021194 -2023-02-13 17:44:36,559 - Epoch: [59][ 360/ 1207] Overall Loss 0.347641 Objective Loss 0.347641 LR 0.001000 Time 0.021141 -2023-02-13 17:44:36,756 - Epoch: [59][ 370/ 1207] Overall Loss 0.348230 Objective Loss 0.348230 LR 0.001000 Time 0.021100 -2023-02-13 17:44:36,949 - Epoch: [59][ 380/ 1207] Overall Loss 0.348124 Objective Loss 0.348124 LR 0.001000 Time 0.021052 -2023-02-13 17:44:37,145 - Epoch: [59][ 390/ 1207] Overall Loss 0.348189 Objective Loss 0.348189 LR 0.001000 Time 0.021015 -2023-02-13 17:44:37,338 - Epoch: [59][ 400/ 1207] Overall Loss 0.347905 Objective Loss 0.347905 LR 0.001000 Time 0.020971 -2023-02-13 17:44:37,534 - Epoch: [59][ 410/ 1207] Overall Loss 0.347856 Objective Loss 0.347856 LR 0.001000 Time 0.020936 -2023-02-13 17:44:37,728 - Epoch: [59][ 420/ 1207] Overall Loss 0.347822 Objective Loss 0.347822 LR 0.001000 Time 0.020898 -2023-02-13 17:44:37,924 - Epoch: [59][ 430/ 1207] Overall Loss 0.347287 Objective Loss 0.347287 LR 0.001000 Time 0.020867 -2023-02-13 17:44:38,117 - Epoch: [59][ 440/ 1207] Overall Loss 0.347799 Objective Loss 0.347799 LR 0.001000 Time 0.020830 -2023-02-13 17:44:38,314 - Epoch: [59][ 450/ 1207] Overall Loss 0.347974 Objective Loss 0.347974 LR 0.001000 Time 0.020804 -2023-02-13 17:44:38,507 - Epoch: [59][ 460/ 1207] Overall Loss 0.347316 Objective Loss 0.347316 LR 0.001000 Time 0.020771 -2023-02-13 17:44:38,704 - Epoch: [59][ 470/ 1207] Overall Loss 0.347599 Objective Loss 0.347599 LR 0.001000 Time 0.020748 -2023-02-13 17:44:38,898 - Epoch: [59][ 480/ 1207] Overall Loss 0.348338 Objective Loss 0.348338 LR 0.001000 Time 0.020719 -2023-02-13 17:44:39,096 - Epoch: [59][ 490/ 1207] Overall Loss 0.348076 Objective Loss 0.348076 LR 0.001000 Time 0.020698 -2023-02-13 17:44:39,290 - Epoch: [59][ 500/ 1207] Overall Loss 0.347624 Objective Loss 0.347624 LR 0.001000 Time 0.020672 -2023-02-13 17:44:39,489 - Epoch: [59][ 510/ 1207] Overall Loss 0.347499 Objective Loss 0.347499 LR 0.001000 Time 0.020656 -2023-02-13 17:44:39,683 - Epoch: [59][ 520/ 1207] Overall Loss 0.347790 Objective Loss 0.347790 LR 0.001000 Time 0.020631 -2023-02-13 17:44:39,882 - Epoch: [59][ 530/ 1207] Overall Loss 0.347687 Objective Loss 0.347687 LR 0.001000 Time 0.020617 -2023-02-13 17:44:40,076 - Epoch: [59][ 540/ 1207] Overall Loss 0.347036 Objective Loss 0.347036 LR 0.001000 Time 0.020594 -2023-02-13 17:44:40,274 - Epoch: [59][ 550/ 1207] Overall Loss 0.346440 Objective Loss 0.346440 LR 0.001000 Time 0.020578 -2023-02-13 17:44:40,469 - Epoch: [59][ 560/ 1207] Overall Loss 0.346344 Objective Loss 0.346344 LR 0.001000 Time 0.020559 -2023-02-13 17:44:40,666 - Epoch: [59][ 570/ 1207] Overall Loss 0.346721 Objective Loss 0.346721 LR 0.001000 Time 0.020543 -2023-02-13 17:44:40,862 - Epoch: [59][ 580/ 1207] Overall Loss 0.346193 Objective Loss 0.346193 LR 0.001000 Time 0.020526 -2023-02-13 17:44:41,060 - Epoch: [59][ 590/ 1207] Overall Loss 0.346115 Objective Loss 0.346115 LR 0.001000 Time 0.020513 -2023-02-13 17:44:41,254 - Epoch: [59][ 600/ 1207] Overall Loss 0.346281 Objective Loss 0.346281 LR 0.001000 Time 0.020494 -2023-02-13 17:44:41,452 - Epoch: [59][ 610/ 1207] Overall Loss 0.346106 Objective Loss 0.346106 LR 0.001000 Time 0.020482 -2023-02-13 17:44:41,646 - Epoch: [59][ 620/ 1207] Overall Loss 0.346557 Objective Loss 0.346557 LR 0.001000 Time 0.020464 -2023-02-13 17:44:41,845 - Epoch: [59][ 630/ 1207] Overall Loss 0.346438 Objective Loss 0.346438 LR 0.001000 Time 0.020454 -2023-02-13 17:44:42,039 - Epoch: [59][ 640/ 1207] Overall Loss 0.346053 Objective Loss 0.346053 LR 0.001000 Time 0.020437 -2023-02-13 17:44:42,237 - Epoch: [59][ 650/ 1207] Overall Loss 0.345406 Objective Loss 0.345406 LR 0.001000 Time 0.020427 -2023-02-13 17:44:42,431 - Epoch: [59][ 660/ 1207] Overall Loss 0.345450 Objective Loss 0.345450 LR 0.001000 Time 0.020411 -2023-02-13 17:44:42,629 - Epoch: [59][ 670/ 1207] Overall Loss 0.345235 Objective Loss 0.345235 LR 0.001000 Time 0.020401 -2023-02-13 17:44:42,823 - Epoch: [59][ 680/ 1207] Overall Loss 0.345626 Objective Loss 0.345626 LR 0.001000 Time 0.020386 -2023-02-13 17:44:43,022 - Epoch: [59][ 690/ 1207] Overall Loss 0.345810 Objective Loss 0.345810 LR 0.001000 Time 0.020378 -2023-02-13 17:44:43,215 - Epoch: [59][ 700/ 1207] Overall Loss 0.346108 Objective Loss 0.346108 LR 0.001000 Time 0.020363 -2023-02-13 17:44:43,414 - Epoch: [59][ 710/ 1207] Overall Loss 0.346652 Objective Loss 0.346652 LR 0.001000 Time 0.020355 -2023-02-13 17:44:43,608 - Epoch: [59][ 720/ 1207] Overall Loss 0.346504 Objective Loss 0.346504 LR 0.001000 Time 0.020341 -2023-02-13 17:44:43,806 - Epoch: [59][ 730/ 1207] Overall Loss 0.346820 Objective Loss 0.346820 LR 0.001000 Time 0.020333 -2023-02-13 17:44:44,000 - Epoch: [59][ 740/ 1207] Overall Loss 0.347076 Objective Loss 0.347076 LR 0.001000 Time 0.020321 -2023-02-13 17:44:44,198 - Epoch: [59][ 750/ 1207] Overall Loss 0.347361 Objective Loss 0.347361 LR 0.001000 Time 0.020313 -2023-02-13 17:44:44,392 - Epoch: [59][ 760/ 1207] Overall Loss 0.347248 Objective Loss 0.347248 LR 0.001000 Time 0.020301 -2023-02-13 17:44:44,588 - Epoch: [59][ 770/ 1207] Overall Loss 0.347241 Objective Loss 0.347241 LR 0.001000 Time 0.020291 -2023-02-13 17:44:44,781 - Epoch: [59][ 780/ 1207] Overall Loss 0.347218 Objective Loss 0.347218 LR 0.001000 Time 0.020278 -2023-02-13 17:44:44,978 - Epoch: [59][ 790/ 1207] Overall Loss 0.347491 Objective Loss 0.347491 LR 0.001000 Time 0.020269 -2023-02-13 17:44:45,171 - Epoch: [59][ 800/ 1207] Overall Loss 0.347614 Objective Loss 0.347614 LR 0.001000 Time 0.020257 -2023-02-13 17:44:45,367 - Epoch: [59][ 810/ 1207] Overall Loss 0.347673 Objective Loss 0.347673 LR 0.001000 Time 0.020249 -2023-02-13 17:44:45,560 - Epoch: [59][ 820/ 1207] Overall Loss 0.347649 Objective Loss 0.347649 LR 0.001000 Time 0.020236 -2023-02-13 17:44:45,756 - Epoch: [59][ 830/ 1207] Overall Loss 0.347602 Objective Loss 0.347602 LR 0.001000 Time 0.020228 -2023-02-13 17:44:45,951 - Epoch: [59][ 840/ 1207] Overall Loss 0.347928 Objective Loss 0.347928 LR 0.001000 Time 0.020219 -2023-02-13 17:44:46,146 - Epoch: [59][ 850/ 1207] Overall Loss 0.347978 Objective Loss 0.347978 LR 0.001000 Time 0.020211 -2023-02-13 17:44:46,340 - Epoch: [59][ 860/ 1207] Overall Loss 0.348233 Objective Loss 0.348233 LR 0.001000 Time 0.020201 -2023-02-13 17:44:46,536 - Epoch: [59][ 870/ 1207] Overall Loss 0.348520 Objective Loss 0.348520 LR 0.001000 Time 0.020194 -2023-02-13 17:44:46,730 - Epoch: [59][ 880/ 1207] Overall Loss 0.348341 Objective Loss 0.348341 LR 0.001000 Time 0.020183 -2023-02-13 17:44:46,927 - Epoch: [59][ 890/ 1207] Overall Loss 0.348422 Objective Loss 0.348422 LR 0.001000 Time 0.020177 -2023-02-13 17:44:47,120 - Epoch: [59][ 900/ 1207] Overall Loss 0.348746 Objective Loss 0.348746 LR 0.001000 Time 0.020168 -2023-02-13 17:44:47,318 - Epoch: [59][ 910/ 1207] Overall Loss 0.349048 Objective Loss 0.349048 LR 0.001000 Time 0.020164 -2023-02-13 17:44:47,513 - Epoch: [59][ 920/ 1207] Overall Loss 0.349258 Objective Loss 0.349258 LR 0.001000 Time 0.020156 -2023-02-13 17:44:47,712 - Epoch: [59][ 930/ 1207] Overall Loss 0.349212 Objective Loss 0.349212 LR 0.001000 Time 0.020152 -2023-02-13 17:44:47,907 - Epoch: [59][ 940/ 1207] Overall Loss 0.349253 Objective Loss 0.349253 LR 0.001000 Time 0.020145 -2023-02-13 17:44:48,105 - Epoch: [59][ 950/ 1207] Overall Loss 0.349410 Objective Loss 0.349410 LR 0.001000 Time 0.020141 -2023-02-13 17:44:48,299 - Epoch: [59][ 960/ 1207] Overall Loss 0.349787 Objective Loss 0.349787 LR 0.001000 Time 0.020133 -2023-02-13 17:44:48,498 - Epoch: [59][ 970/ 1207] Overall Loss 0.350225 Objective Loss 0.350225 LR 0.001000 Time 0.020130 -2023-02-13 17:44:48,693 - Epoch: [59][ 980/ 1207] Overall Loss 0.350269 Objective Loss 0.350269 LR 0.001000 Time 0.020123 -2023-02-13 17:44:48,891 - Epoch: [59][ 990/ 1207] Overall Loss 0.350430 Objective Loss 0.350430 LR 0.001000 Time 0.020120 -2023-02-13 17:44:49,086 - Epoch: [59][ 1000/ 1207] Overall Loss 0.350553 Objective Loss 0.350553 LR 0.001000 Time 0.020113 -2023-02-13 17:44:49,284 - Epoch: [59][ 1010/ 1207] Overall Loss 0.350573 Objective Loss 0.350573 LR 0.001000 Time 0.020110 -2023-02-13 17:44:49,479 - Epoch: [59][ 1020/ 1207] Overall Loss 0.350388 Objective Loss 0.350388 LR 0.001000 Time 0.020103 -2023-02-13 17:44:49,677 - Epoch: [59][ 1030/ 1207] Overall Loss 0.350566 Objective Loss 0.350566 LR 0.001000 Time 0.020100 -2023-02-13 17:44:49,872 - Epoch: [59][ 1040/ 1207] Overall Loss 0.350310 Objective Loss 0.350310 LR 0.001000 Time 0.020094 -2023-02-13 17:44:50,070 - Epoch: [59][ 1050/ 1207] Overall Loss 0.350345 Objective Loss 0.350345 LR 0.001000 Time 0.020091 -2023-02-13 17:44:50,265 - Epoch: [59][ 1060/ 1207] Overall Loss 0.350361 Objective Loss 0.350361 LR 0.001000 Time 0.020085 -2023-02-13 17:44:50,464 - Epoch: [59][ 1070/ 1207] Overall Loss 0.350639 Objective Loss 0.350639 LR 0.001000 Time 0.020082 -2023-02-13 17:44:50,658 - Epoch: [59][ 1080/ 1207] Overall Loss 0.350941 Objective Loss 0.350941 LR 0.001000 Time 0.020076 -2023-02-13 17:44:50,859 - Epoch: [59][ 1090/ 1207] Overall Loss 0.350966 Objective Loss 0.350966 LR 0.001000 Time 0.020076 -2023-02-13 17:44:51,055 - Epoch: [59][ 1100/ 1207] Overall Loss 0.351450 Objective Loss 0.351450 LR 0.001000 Time 0.020071 -2023-02-13 17:44:51,254 - Epoch: [59][ 1110/ 1207] Overall Loss 0.351577 Objective Loss 0.351577 LR 0.001000 Time 0.020070 -2023-02-13 17:44:51,451 - Epoch: [59][ 1120/ 1207] Overall Loss 0.351717 Objective Loss 0.351717 LR 0.001000 Time 0.020065 -2023-02-13 17:44:51,649 - Epoch: [59][ 1130/ 1207] Overall Loss 0.351747 Objective Loss 0.351747 LR 0.001000 Time 0.020063 -2023-02-13 17:44:51,846 - Epoch: [59][ 1140/ 1207] Overall Loss 0.351744 Objective Loss 0.351744 LR 0.001000 Time 0.020059 -2023-02-13 17:44:52,045 - Epoch: [59][ 1150/ 1207] Overall Loss 0.351740 Objective Loss 0.351740 LR 0.001000 Time 0.020058 -2023-02-13 17:44:52,240 - Epoch: [59][ 1160/ 1207] Overall Loss 0.351939 Objective Loss 0.351939 LR 0.001000 Time 0.020053 -2023-02-13 17:44:52,440 - Epoch: [59][ 1170/ 1207] Overall Loss 0.351849 Objective Loss 0.351849 LR 0.001000 Time 0.020052 -2023-02-13 17:44:52,636 - Epoch: [59][ 1180/ 1207] Overall Loss 0.351704 Objective Loss 0.351704 LR 0.001000 Time 0.020047 -2023-02-13 17:44:52,835 - Epoch: [59][ 1190/ 1207] Overall Loss 0.351439 Objective Loss 0.351439 LR 0.001000 Time 0.020046 -2023-02-13 17:44:53,082 - Epoch: [59][ 1200/ 1207] Overall Loss 0.351476 Objective Loss 0.351476 LR 0.001000 Time 0.020084 -2023-02-13 17:44:53,197 - Epoch: [59][ 1207/ 1207] Overall Loss 0.351447 Objective Loss 0.351447 Top1 79.573171 Top5 96.646341 LR 0.001000 Time 0.020063 -2023-02-13 17:44:53,276 - --- validate (epoch=59)----------- -2023-02-13 17:44:53,277 - 34311 samples (256 per mini-batch) -2023-02-13 17:44:53,689 - Epoch: [59][ 10/ 135] Loss 0.379718 Top1 79.648438 Top5 96.171875 -2023-02-13 17:44:53,817 - Epoch: [59][ 20/ 135] Loss 0.377436 Top1 80.117188 Top5 96.621094 -2023-02-13 17:44:53,949 - Epoch: [59][ 30/ 135] Loss 0.388859 Top1 79.817708 Top5 96.497396 -2023-02-13 17:44:54,078 - Epoch: [59][ 40/ 135] Loss 0.387979 Top1 80.048828 Top5 96.513672 -2023-02-13 17:44:54,208 - Epoch: [59][ 50/ 135] Loss 0.383179 Top1 80.257812 Top5 96.570312 -2023-02-13 17:44:54,340 - Epoch: [59][ 60/ 135] Loss 0.388866 Top1 80.247396 Top5 96.627604 -2023-02-13 17:44:54,472 - Epoch: [59][ 70/ 135] Loss 0.388967 Top1 80.345982 Top5 96.640625 -2023-02-13 17:44:54,593 - Epoch: [59][ 80/ 135] Loss 0.391881 Top1 80.239258 Top5 96.586914 -2023-02-13 17:44:54,720 - Epoch: [59][ 90/ 135] Loss 0.390306 Top1 80.251736 Top5 96.605903 -2023-02-13 17:44:54,851 - Epoch: [59][ 100/ 135] Loss 0.387968 Top1 80.292969 Top5 96.574219 -2023-02-13 17:44:54,982 - Epoch: [59][ 110/ 135] Loss 0.387041 Top1 80.305398 Top5 96.580256 -2023-02-13 17:44:55,113 - Epoch: [59][ 120/ 135] Loss 0.384210 Top1 80.364583 Top5 96.608073 -2023-02-13 17:44:55,247 - Epoch: [59][ 130/ 135] Loss 0.384188 Top1 80.276442 Top5 96.592548 -2023-02-13 17:44:55,294 - Epoch: [59][ 135/ 135] Loss 0.385124 Top1 80.312436 Top5 96.610416 -2023-02-13 17:44:55,367 - ==> Top1: 80.312 Top5: 96.610 Loss: 0.385 - -2023-02-13 17:44:55,368 - ==> Confusion: -[[ 849 5 9 2 15 4 0 3 4 37 0 3 0 5 9 3 8 5 1 1 4] - [ 5 924 4 1 12 31 2 16 4 1 0 5 1 1 1 1 9 0 9 2 4] - [ 8 5 952 12 2 0 18 20 0 0 3 4 3 2 4 6 4 4 2 4 5] - [ 6 1 21 886 1 4 2 4 1 3 15 1 10 2 17 5 1 6 20 1 9] - [ 19 9 3 1 969 14 1 2 1 1 0 8 0 4 5 6 13 1 0 2 7] - [ 5 19 2 6 4 934 3 23 1 3 1 18 7 15 2 3 11 2 4 4 3] - [ 3 3 28 3 1 5 1015 4 0 0 2 2 1 3 0 11 1 5 1 9 2] - [ 3 17 11 1 2 32 4 899 1 1 0 12 5 1 0 0 1 1 20 10 3] - [ 20 2 1 1 2 1 0 1 880 40 12 3 2 8 25 1 3 1 5 0 1] - [ 97 1 5 1 4 0 2 4 35 817 0 2 2 24 6 2 3 3 0 1 3] - [ 1 6 7 3 1 4 5 6 17 1 958 3 2 12 2 2 2 2 14 2 1] - [ 3 2 1 1 3 11 1 5 2 0 0 909 24 4 1 7 7 14 1 8 1] - [ 4 0 0 8 1 1 0 2 3 0 0 43 865 1 4 5 2 14 2 1 3] - [ 8 7 4 0 4 13 0 2 15 21 8 11 3 901 3 8 9 2 0 2 3] - [ 13 6 2 19 8 1 0 2 20 6 2 1 4 0 980 1 3 5 11 0 8] - [ 3 1 7 0 6 1 1 1 0 0 0 8 6 2 1 973 15 12 0 5 4] - [ 6 7 0 1 7 2 1 0 3 1 0 2 2 1 3 6 1008 3 2 1 5] - [ 8 1 2 7 1 0 0 1 1 1 1 14 36 0 2 26 0 946 0 2 2] - [ 5 7 11 14 0 0 0 37 3 0 2 2 5 1 14 1 1 2 980 0 1] - [ 1 3 1 0 2 3 9 24 1 0 1 33 4 5 0 7 11 1 0 1036 6] - [ 216 275 315 123 176 251 92 238 145 89 214 217 402 325 237 171 458 110 209 296 8875]] - -2023-02-13 17:44:55,369 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:44:55,370 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:44:55,375 - - -2023-02-13 17:44:55,375 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:44:56,267 - Epoch: [60][ 10/ 1207] Overall Loss 0.341366 Objective Loss 0.341366 LR 0.001000 Time 0.089132 -2023-02-13 17:44:56,463 - Epoch: [60][ 20/ 1207] Overall Loss 0.329072 Objective Loss 0.329072 LR 0.001000 Time 0.054303 -2023-02-13 17:44:56,659 - Epoch: [60][ 30/ 1207] Overall Loss 0.328348 Objective Loss 0.328348 LR 0.001000 Time 0.042729 -2023-02-13 17:44:56,853 - Epoch: [60][ 40/ 1207] Overall Loss 0.327637 Objective Loss 0.327637 LR 0.001000 Time 0.036901 -2023-02-13 17:44:57,050 - Epoch: [60][ 50/ 1207] Overall Loss 0.332334 Objective Loss 0.332334 LR 0.001000 Time 0.033456 -2023-02-13 17:44:57,244 - Epoch: [60][ 60/ 1207] Overall Loss 0.336706 Objective Loss 0.336706 LR 0.001000 Time 0.031106 -2023-02-13 17:44:57,441 - Epoch: [60][ 70/ 1207] Overall Loss 0.332502 Objective Loss 0.332502 LR 0.001000 Time 0.029468 -2023-02-13 17:44:57,634 - Epoch: [60][ 80/ 1207] Overall Loss 0.334514 Objective Loss 0.334514 LR 0.001000 Time 0.028197 -2023-02-13 17:44:57,827 - Epoch: [60][ 90/ 1207] Overall Loss 0.334622 Objective Loss 0.334622 LR 0.001000 Time 0.027200 -2023-02-13 17:44:58,017 - Epoch: [60][ 100/ 1207] Overall Loss 0.337895 Objective Loss 0.337895 LR 0.001000 Time 0.026374 -2023-02-13 17:44:58,206 - Epoch: [60][ 110/ 1207] Overall Loss 0.335782 Objective Loss 0.335782 LR 0.001000 Time 0.025697 -2023-02-13 17:44:58,396 - Epoch: [60][ 120/ 1207] Overall Loss 0.335702 Objective Loss 0.335702 LR 0.001000 Time 0.025136 -2023-02-13 17:44:58,586 - Epoch: [60][ 130/ 1207] Overall Loss 0.335006 Objective Loss 0.335006 LR 0.001000 Time 0.024657 -2023-02-13 17:44:58,776 - Epoch: [60][ 140/ 1207] Overall Loss 0.335914 Objective Loss 0.335914 LR 0.001000 Time 0.024254 -2023-02-13 17:44:58,966 - Epoch: [60][ 150/ 1207] Overall Loss 0.337353 Objective Loss 0.337353 LR 0.001000 Time 0.023900 -2023-02-13 17:44:59,156 - Epoch: [60][ 160/ 1207] Overall Loss 0.339918 Objective Loss 0.339918 LR 0.001000 Time 0.023590 -2023-02-13 17:44:59,345 - Epoch: [60][ 170/ 1207] Overall Loss 0.338779 Objective Loss 0.338779 LR 0.001000 Time 0.023314 -2023-02-13 17:44:59,535 - Epoch: [60][ 180/ 1207] Overall Loss 0.337506 Objective Loss 0.337506 LR 0.001000 Time 0.023068 -2023-02-13 17:44:59,724 - Epoch: [60][ 190/ 1207] Overall Loss 0.337387 Objective Loss 0.337387 LR 0.001000 Time 0.022850 -2023-02-13 17:44:59,915 - Epoch: [60][ 200/ 1207] Overall Loss 0.337263 Objective Loss 0.337263 LR 0.001000 Time 0.022657 -2023-02-13 17:45:00,104 - Epoch: [60][ 210/ 1207] Overall Loss 0.336756 Objective Loss 0.336756 LR 0.001000 Time 0.022477 -2023-02-13 17:45:00,293 - Epoch: [60][ 220/ 1207] Overall Loss 0.336160 Objective Loss 0.336160 LR 0.001000 Time 0.022314 -2023-02-13 17:45:00,482 - Epoch: [60][ 230/ 1207] Overall Loss 0.336466 Objective Loss 0.336466 LR 0.001000 Time 0.022166 -2023-02-13 17:45:00,672 - Epoch: [60][ 240/ 1207] Overall Loss 0.338450 Objective Loss 0.338450 LR 0.001000 Time 0.022032 -2023-02-13 17:45:00,864 - Epoch: [60][ 250/ 1207] Overall Loss 0.340236 Objective Loss 0.340236 LR 0.001000 Time 0.021916 -2023-02-13 17:45:01,054 - Epoch: [60][ 260/ 1207] Overall Loss 0.339974 Objective Loss 0.339974 LR 0.001000 Time 0.021804 -2023-02-13 17:45:01,244 - Epoch: [60][ 270/ 1207] Overall Loss 0.339837 Objective Loss 0.339837 LR 0.001000 Time 0.021698 -2023-02-13 17:45:01,435 - Epoch: [60][ 280/ 1207] Overall Loss 0.339860 Objective Loss 0.339860 LR 0.001000 Time 0.021603 -2023-02-13 17:45:01,625 - Epoch: [60][ 290/ 1207] Overall Loss 0.339992 Objective Loss 0.339992 LR 0.001000 Time 0.021512 -2023-02-13 17:45:01,815 - Epoch: [60][ 300/ 1207] Overall Loss 0.341130 Objective Loss 0.341130 LR 0.001000 Time 0.021427 -2023-02-13 17:45:02,004 - Epoch: [60][ 310/ 1207] Overall Loss 0.342479 Objective Loss 0.342479 LR 0.001000 Time 0.021346 -2023-02-13 17:45:02,194 - Epoch: [60][ 320/ 1207] Overall Loss 0.342064 Objective Loss 0.342064 LR 0.001000 Time 0.021271 -2023-02-13 17:45:02,383 - Epoch: [60][ 330/ 1207] Overall Loss 0.341721 Objective Loss 0.341721 LR 0.001000 Time 0.021199 -2023-02-13 17:45:02,574 - Epoch: [60][ 340/ 1207] Overall Loss 0.342248 Objective Loss 0.342248 LR 0.001000 Time 0.021134 -2023-02-13 17:45:02,765 - Epoch: [60][ 350/ 1207] Overall Loss 0.342525 Objective Loss 0.342525 LR 0.001000 Time 0.021075 -2023-02-13 17:45:02,954 - Epoch: [60][ 360/ 1207] Overall Loss 0.341900 Objective Loss 0.341900 LR 0.001000 Time 0.021015 -2023-02-13 17:45:03,143 - Epoch: [60][ 370/ 1207] Overall Loss 0.342107 Objective Loss 0.342107 LR 0.001000 Time 0.020957 -2023-02-13 17:45:03,333 - Epoch: [60][ 380/ 1207] Overall Loss 0.342377 Objective Loss 0.342377 LR 0.001000 Time 0.020904 -2023-02-13 17:45:03,523 - Epoch: [60][ 390/ 1207] Overall Loss 0.342013 Objective Loss 0.342013 LR 0.001000 Time 0.020853 -2023-02-13 17:45:03,713 - Epoch: [60][ 400/ 1207] Overall Loss 0.342646 Objective Loss 0.342646 LR 0.001000 Time 0.020806 -2023-02-13 17:45:03,902 - Epoch: [60][ 410/ 1207] Overall Loss 0.342952 Objective Loss 0.342952 LR 0.001000 Time 0.020760 -2023-02-13 17:45:04,092 - Epoch: [60][ 420/ 1207] Overall Loss 0.343024 Objective Loss 0.343024 LR 0.001000 Time 0.020716 -2023-02-13 17:45:04,280 - Epoch: [60][ 430/ 1207] Overall Loss 0.342947 Objective Loss 0.342947 LR 0.001000 Time 0.020672 -2023-02-13 17:45:04,470 - Epoch: [60][ 440/ 1207] Overall Loss 0.342068 Objective Loss 0.342068 LR 0.001000 Time 0.020633 -2023-02-13 17:45:04,660 - Epoch: [60][ 450/ 1207] Overall Loss 0.342012 Objective Loss 0.342012 LR 0.001000 Time 0.020595 -2023-02-13 17:45:04,850 - Epoch: [60][ 460/ 1207] Overall Loss 0.341731 Objective Loss 0.341731 LR 0.001000 Time 0.020560 -2023-02-13 17:45:05,039 - Epoch: [60][ 470/ 1207] Overall Loss 0.342232 Objective Loss 0.342232 LR 0.001000 Time 0.020525 -2023-02-13 17:45:05,230 - Epoch: [60][ 480/ 1207] Overall Loss 0.342550 Objective Loss 0.342550 LR 0.001000 Time 0.020492 -2023-02-13 17:45:05,420 - Epoch: [60][ 490/ 1207] Overall Loss 0.343059 Objective Loss 0.343059 LR 0.001000 Time 0.020461 -2023-02-13 17:45:05,611 - Epoch: [60][ 500/ 1207] Overall Loss 0.342665 Objective Loss 0.342665 LR 0.001000 Time 0.020433 -2023-02-13 17:45:05,802 - Epoch: [60][ 510/ 1207] Overall Loss 0.343426 Objective Loss 0.343426 LR 0.001000 Time 0.020406 -2023-02-13 17:45:05,992 - Epoch: [60][ 520/ 1207] Overall Loss 0.343777 Objective Loss 0.343777 LR 0.001000 Time 0.020379 -2023-02-13 17:45:06,181 - Epoch: [60][ 530/ 1207] Overall Loss 0.343647 Objective Loss 0.343647 LR 0.001000 Time 0.020351 -2023-02-13 17:45:06,371 - Epoch: [60][ 540/ 1207] Overall Loss 0.343640 Objective Loss 0.343640 LR 0.001000 Time 0.020325 -2023-02-13 17:45:06,562 - Epoch: [60][ 550/ 1207] Overall Loss 0.344153 Objective Loss 0.344153 LR 0.001000 Time 0.020302 -2023-02-13 17:45:06,752 - Epoch: [60][ 560/ 1207] Overall Loss 0.344393 Objective Loss 0.344393 LR 0.001000 Time 0.020279 -2023-02-13 17:45:06,943 - Epoch: [60][ 570/ 1207] Overall Loss 0.344966 Objective Loss 0.344966 LR 0.001000 Time 0.020257 -2023-02-13 17:45:07,133 - Epoch: [60][ 580/ 1207] Overall Loss 0.345110 Objective Loss 0.345110 LR 0.001000 Time 0.020234 -2023-02-13 17:45:07,322 - Epoch: [60][ 590/ 1207] Overall Loss 0.345766 Objective Loss 0.345766 LR 0.001000 Time 0.020212 -2023-02-13 17:45:07,513 - Epoch: [60][ 600/ 1207] Overall Loss 0.345697 Objective Loss 0.345697 LR 0.001000 Time 0.020192 -2023-02-13 17:45:07,703 - Epoch: [60][ 610/ 1207] Overall Loss 0.345489 Objective Loss 0.345489 LR 0.001000 Time 0.020172 -2023-02-13 17:45:07,894 - Epoch: [60][ 620/ 1207] Overall Loss 0.346073 Objective Loss 0.346073 LR 0.001000 Time 0.020154 -2023-02-13 17:45:08,084 - Epoch: [60][ 630/ 1207] Overall Loss 0.346189 Objective Loss 0.346189 LR 0.001000 Time 0.020135 -2023-02-13 17:45:08,274 - Epoch: [60][ 640/ 1207] Overall Loss 0.346343 Objective Loss 0.346343 LR 0.001000 Time 0.020117 -2023-02-13 17:45:08,464 - Epoch: [60][ 650/ 1207] Overall Loss 0.346000 Objective Loss 0.346000 LR 0.001000 Time 0.020099 -2023-02-13 17:45:08,655 - Epoch: [60][ 660/ 1207] Overall Loss 0.346066 Objective Loss 0.346066 LR 0.001000 Time 0.020083 -2023-02-13 17:45:08,845 - Epoch: [60][ 670/ 1207] Overall Loss 0.346203 Objective Loss 0.346203 LR 0.001000 Time 0.020067 -2023-02-13 17:45:09,035 - Epoch: [60][ 680/ 1207] Overall Loss 0.346467 Objective Loss 0.346467 LR 0.001000 Time 0.020051 -2023-02-13 17:45:09,226 - Epoch: [60][ 690/ 1207] Overall Loss 0.345981 Objective Loss 0.345981 LR 0.001000 Time 0.020036 -2023-02-13 17:45:09,416 - Epoch: [60][ 700/ 1207] Overall Loss 0.346111 Objective Loss 0.346111 LR 0.001000 Time 0.020021 -2023-02-13 17:45:09,607 - Epoch: [60][ 710/ 1207] Overall Loss 0.346014 Objective Loss 0.346014 LR 0.001000 Time 0.020007 -2023-02-13 17:45:09,798 - Epoch: [60][ 720/ 1207] Overall Loss 0.346685 Objective Loss 0.346685 LR 0.001000 Time 0.019994 -2023-02-13 17:45:09,988 - Epoch: [60][ 730/ 1207] Overall Loss 0.346172 Objective Loss 0.346172 LR 0.001000 Time 0.019981 -2023-02-13 17:45:10,179 - Epoch: [60][ 740/ 1207] Overall Loss 0.346297 Objective Loss 0.346297 LR 0.001000 Time 0.019968 -2023-02-13 17:45:10,370 - Epoch: [60][ 750/ 1207] Overall Loss 0.346354 Objective Loss 0.346354 LR 0.001000 Time 0.019955 -2023-02-13 17:45:10,560 - Epoch: [60][ 760/ 1207] Overall Loss 0.346267 Objective Loss 0.346267 LR 0.001000 Time 0.019943 -2023-02-13 17:45:10,750 - Epoch: [60][ 770/ 1207] Overall Loss 0.346254 Objective Loss 0.346254 LR 0.001000 Time 0.019930 -2023-02-13 17:45:10,942 - Epoch: [60][ 780/ 1207] Overall Loss 0.346295 Objective Loss 0.346295 LR 0.001000 Time 0.019920 -2023-02-13 17:45:11,131 - Epoch: [60][ 790/ 1207] Overall Loss 0.345911 Objective Loss 0.345911 LR 0.001000 Time 0.019907 -2023-02-13 17:45:11,321 - Epoch: [60][ 800/ 1207] Overall Loss 0.345828 Objective Loss 0.345828 LR 0.001000 Time 0.019895 -2023-02-13 17:45:11,511 - Epoch: [60][ 810/ 1207] Overall Loss 0.345738 Objective Loss 0.345738 LR 0.001000 Time 0.019884 -2023-02-13 17:45:11,702 - Epoch: [60][ 820/ 1207] Overall Loss 0.345590 Objective Loss 0.345590 LR 0.001000 Time 0.019873 -2023-02-13 17:45:11,892 - Epoch: [60][ 830/ 1207] Overall Loss 0.345362 Objective Loss 0.345362 LR 0.001000 Time 0.019863 -2023-02-13 17:45:12,083 - Epoch: [60][ 840/ 1207] Overall Loss 0.345311 Objective Loss 0.345311 LR 0.001000 Time 0.019852 -2023-02-13 17:45:12,272 - Epoch: [60][ 850/ 1207] Overall Loss 0.345569 Objective Loss 0.345569 LR 0.001000 Time 0.019841 -2023-02-13 17:45:12,462 - Epoch: [60][ 860/ 1207] Overall Loss 0.345771 Objective Loss 0.345771 LR 0.001000 Time 0.019831 -2023-02-13 17:45:12,652 - Epoch: [60][ 870/ 1207] Overall Loss 0.345797 Objective Loss 0.345797 LR 0.001000 Time 0.019821 -2023-02-13 17:45:12,843 - Epoch: [60][ 880/ 1207] Overall Loss 0.345683 Objective Loss 0.345683 LR 0.001000 Time 0.019813 -2023-02-13 17:45:13,033 - Epoch: [60][ 890/ 1207] Overall Loss 0.345536 Objective Loss 0.345536 LR 0.001000 Time 0.019803 -2023-02-13 17:45:13,223 - Epoch: [60][ 900/ 1207] Overall Loss 0.345786 Objective Loss 0.345786 LR 0.001000 Time 0.019794 -2023-02-13 17:45:13,413 - Epoch: [60][ 910/ 1207] Overall Loss 0.345770 Objective Loss 0.345770 LR 0.001000 Time 0.019784 -2023-02-13 17:45:13,604 - Epoch: [60][ 920/ 1207] Overall Loss 0.345732 Objective Loss 0.345732 LR 0.001000 Time 0.019776 -2023-02-13 17:45:13,793 - Epoch: [60][ 930/ 1207] Overall Loss 0.345782 Objective Loss 0.345782 LR 0.001000 Time 0.019767 -2023-02-13 17:45:13,983 - Epoch: [60][ 940/ 1207] Overall Loss 0.345993 Objective Loss 0.345993 LR 0.001000 Time 0.019758 -2023-02-13 17:45:14,173 - Epoch: [60][ 950/ 1207] Overall Loss 0.346031 Objective Loss 0.346031 LR 0.001000 Time 0.019749 -2023-02-13 17:45:14,362 - Epoch: [60][ 960/ 1207] Overall Loss 0.345697 Objective Loss 0.345697 LR 0.001000 Time 0.019741 -2023-02-13 17:45:14,552 - Epoch: [60][ 970/ 1207] Overall Loss 0.345732 Objective Loss 0.345732 LR 0.001000 Time 0.019733 -2023-02-13 17:45:14,742 - Epoch: [60][ 980/ 1207] Overall Loss 0.345769 Objective Loss 0.345769 LR 0.001000 Time 0.019725 -2023-02-13 17:45:14,933 - Epoch: [60][ 990/ 1207] Overall Loss 0.345761 Objective Loss 0.345761 LR 0.001000 Time 0.019718 -2023-02-13 17:45:15,122 - Epoch: [60][ 1000/ 1207] Overall Loss 0.345729 Objective Loss 0.345729 LR 0.001000 Time 0.019710 -2023-02-13 17:45:15,311 - Epoch: [60][ 1010/ 1207] Overall Loss 0.346133 Objective Loss 0.346133 LR 0.001000 Time 0.019701 -2023-02-13 17:45:15,502 - Epoch: [60][ 1020/ 1207] Overall Loss 0.346167 Objective Loss 0.346167 LR 0.001000 Time 0.019695 -2023-02-13 17:45:15,692 - Epoch: [60][ 1030/ 1207] Overall Loss 0.346007 Objective Loss 0.346007 LR 0.001000 Time 0.019688 -2023-02-13 17:45:15,884 - Epoch: [60][ 1040/ 1207] Overall Loss 0.346334 Objective Loss 0.346334 LR 0.001000 Time 0.019683 -2023-02-13 17:45:16,073 - Epoch: [60][ 1050/ 1207] Overall Loss 0.346224 Objective Loss 0.346224 LR 0.001000 Time 0.019675 -2023-02-13 17:45:16,263 - Epoch: [60][ 1060/ 1207] Overall Loss 0.346442 Objective Loss 0.346442 LR 0.001000 Time 0.019668 -2023-02-13 17:45:16,453 - Epoch: [60][ 1070/ 1207] Overall Loss 0.346605 Objective Loss 0.346605 LR 0.001000 Time 0.019661 -2023-02-13 17:45:16,644 - Epoch: [60][ 1080/ 1207] Overall Loss 0.346713 Objective Loss 0.346713 LR 0.001000 Time 0.019656 -2023-02-13 17:45:16,834 - Epoch: [60][ 1090/ 1207] Overall Loss 0.346839 Objective Loss 0.346839 LR 0.001000 Time 0.019650 -2023-02-13 17:45:17,025 - Epoch: [60][ 1100/ 1207] Overall Loss 0.346823 Objective Loss 0.346823 LR 0.001000 Time 0.019644 -2023-02-13 17:45:17,214 - Epoch: [60][ 1110/ 1207] Overall Loss 0.346771 Objective Loss 0.346771 LR 0.001000 Time 0.019637 -2023-02-13 17:45:17,404 - Epoch: [60][ 1120/ 1207] Overall Loss 0.346631 Objective Loss 0.346631 LR 0.001000 Time 0.019631 -2023-02-13 17:45:17,594 - Epoch: [60][ 1130/ 1207] Overall Loss 0.346800 Objective Loss 0.346800 LR 0.001000 Time 0.019626 -2023-02-13 17:45:17,785 - Epoch: [60][ 1140/ 1207] Overall Loss 0.346877 Objective Loss 0.346877 LR 0.001000 Time 0.019621 -2023-02-13 17:45:17,975 - Epoch: [60][ 1150/ 1207] Overall Loss 0.347020 Objective Loss 0.347020 LR 0.001000 Time 0.019615 -2023-02-13 17:45:18,165 - Epoch: [60][ 1160/ 1207] Overall Loss 0.347040 Objective Loss 0.347040 LR 0.001000 Time 0.019609 -2023-02-13 17:45:18,354 - Epoch: [60][ 1170/ 1207] Overall Loss 0.347133 Objective Loss 0.347133 LR 0.001000 Time 0.019603 -2023-02-13 17:45:18,545 - Epoch: [60][ 1180/ 1207] Overall Loss 0.347288 Objective Loss 0.347288 LR 0.001000 Time 0.019598 -2023-02-13 17:45:18,735 - Epoch: [60][ 1190/ 1207] Overall Loss 0.347187 Objective Loss 0.347187 LR 0.001000 Time 0.019593 -2023-02-13 17:45:18,975 - Epoch: [60][ 1200/ 1207] Overall Loss 0.347218 Objective Loss 0.347218 LR 0.001000 Time 0.019629 -2023-02-13 17:45:19,091 - Epoch: [60][ 1207/ 1207] Overall Loss 0.347185 Objective Loss 0.347185 Top1 83.536585 Top5 96.951220 LR 0.001000 Time 0.019612 -2023-02-13 17:45:19,163 - --- validate (epoch=60)----------- -2023-02-13 17:45:19,163 - 34311 samples (256 per mini-batch) -2023-02-13 17:45:19,574 - Epoch: [60][ 10/ 135] Loss 0.392878 Top1 81.796875 Top5 97.578125 -2023-02-13 17:45:19,707 - Epoch: [60][ 20/ 135] Loss 0.368009 Top1 82.597656 Top5 97.421875 -2023-02-13 17:45:19,847 - Epoch: [60][ 30/ 135] Loss 0.363607 Top1 82.721354 Top5 97.408854 -2023-02-13 17:45:19,974 - Epoch: [60][ 40/ 135] Loss 0.369349 Top1 82.470703 Top5 97.314453 -2023-02-13 17:45:20,102 - Epoch: [60][ 50/ 135] Loss 0.369822 Top1 82.578125 Top5 97.351562 -2023-02-13 17:45:20,232 - Epoch: [60][ 60/ 135] Loss 0.373814 Top1 82.415365 Top5 97.291667 -2023-02-13 17:45:20,356 - Epoch: [60][ 70/ 135] Loss 0.376714 Top1 82.310268 Top5 97.271205 -2023-02-13 17:45:20,484 - Epoch: [60][ 80/ 135] Loss 0.377680 Top1 82.353516 Top5 97.197266 -2023-02-13 17:45:20,611 - Epoch: [60][ 90/ 135] Loss 0.375338 Top1 82.348090 Top5 97.196181 -2023-02-13 17:45:20,736 - Epoch: [60][ 100/ 135] Loss 0.375881 Top1 82.406250 Top5 97.183594 -2023-02-13 17:45:20,865 - Epoch: [60][ 110/ 135] Loss 0.374431 Top1 82.404119 Top5 97.212358 -2023-02-13 17:45:20,995 - Epoch: [60][ 120/ 135] Loss 0.375740 Top1 82.307943 Top5 97.213542 -2023-02-13 17:45:21,125 - Epoch: [60][ 130/ 135] Loss 0.376533 Top1 82.412861 Top5 97.181490 -2023-02-13 17:45:21,170 - Epoch: [60][ 135/ 135] Loss 0.378579 Top1 82.399231 Top5 97.164175 -2023-02-13 17:45:21,237 - ==> Top1: 82.399 Top5: 97.164 Loss: 0.379 - -2023-02-13 17:45:21,238 - ==> Confusion: -[[ 886 3 6 3 4 3 0 2 2 27 1 5 0 3 2 3 4 3 2 1 7] - [ 4 924 0 3 12 30 1 21 3 1 1 4 2 0 5 3 4 1 3 2 9] - [ 9 5 955 17 4 0 11 14 1 0 3 3 2 3 2 3 3 3 7 7 6] - [ 7 2 26 914 1 5 0 1 1 1 11 2 5 2 14 2 1 6 6 1 8] - [ 32 8 1 2 972 6 1 2 0 1 1 6 0 3 7 6 4 4 0 3 7] - [ 6 30 2 8 8 933 3 20 1 2 4 14 6 11 3 2 4 0 3 2 8] - [ 4 3 31 4 0 6 1001 11 0 1 3 2 3 1 1 6 3 3 1 8 7] - [ 5 6 12 3 1 37 1 905 3 1 1 2 4 0 0 1 1 2 19 11 9] - [ 36 2 1 2 0 0 0 1 858 40 19 4 2 9 23 2 4 2 1 0 3] - [ 169 1 4 1 4 4 0 2 30 762 1 1 1 18 5 2 1 1 1 0 4] - [ 2 3 7 12 1 1 2 3 11 2 977 1 3 7 2 1 2 1 8 0 5] - [ 3 5 1 3 3 19 0 9 1 1 0 889 30 7 0 4 2 15 0 13 0] - [ 1 0 0 11 1 2 0 2 1 0 0 34 858 0 2 8 2 27 1 1 8] - [ 7 3 7 0 3 15 0 2 17 19 12 5 4 892 7 5 9 1 1 8 7] - [ 24 1 4 28 7 2 0 3 26 4 5 2 3 1 951 1 4 3 8 0 15] - [ 5 0 11 4 7 1 2 0 0 0 0 8 10 4 1 949 11 17 1 7 8] - [ 4 4 0 1 8 1 1 0 2 1 0 2 3 3 2 6 1001 0 1 4 17] - [ 8 1 0 6 1 1 0 1 0 0 1 6 17 3 0 12 0 989 1 1 3] - [ 7 2 6 25 0 1 0 39 1 1 5 2 10 1 18 0 1 2 955 0 10] - [ 1 5 1 2 2 2 4 21 0 0 2 21 5 3 0 3 5 5 0 1057 9] - [ 287 191 295 193 163 199 47 182 93 94 175 143 373 268 170 110 244 155 130 278 9644]] - -2023-02-13 17:45:21,240 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:45:21,240 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:45:21,246 - - -2023-02-13 17:45:21,246 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:45:22,125 - Epoch: [61][ 10/ 1207] Overall Loss 0.339234 Objective Loss 0.339234 LR 0.001000 Time 0.087818 -2023-02-13 17:45:22,317 - Epoch: [61][ 20/ 1207] Overall Loss 0.329243 Objective Loss 0.329243 LR 0.001000 Time 0.053508 -2023-02-13 17:45:22,513 - Epoch: [61][ 30/ 1207] Overall Loss 0.341597 Objective Loss 0.341597 LR 0.001000 Time 0.042200 -2023-02-13 17:45:22,707 - Epoch: [61][ 40/ 1207] Overall Loss 0.345901 Objective Loss 0.345901 LR 0.001000 Time 0.036468 -2023-02-13 17:45:22,904 - Epoch: [61][ 50/ 1207] Overall Loss 0.346441 Objective Loss 0.346441 LR 0.001000 Time 0.033110 -2023-02-13 17:45:23,096 - Epoch: [61][ 60/ 1207] Overall Loss 0.345832 Objective Loss 0.345832 LR 0.001000 Time 0.030795 -2023-02-13 17:45:23,293 - Epoch: [61][ 70/ 1207] Overall Loss 0.348294 Objective Loss 0.348294 LR 0.001000 Time 0.029201 -2023-02-13 17:45:23,486 - Epoch: [61][ 80/ 1207] Overall Loss 0.346706 Objective Loss 0.346706 LR 0.001000 Time 0.027960 -2023-02-13 17:45:23,683 - Epoch: [61][ 90/ 1207] Overall Loss 0.342827 Objective Loss 0.342827 LR 0.001000 Time 0.027032 -2023-02-13 17:45:23,876 - Epoch: [61][ 100/ 1207] Overall Loss 0.342681 Objective Loss 0.342681 LR 0.001000 Time 0.026255 -2023-02-13 17:45:24,072 - Epoch: [61][ 110/ 1207] Overall Loss 0.342921 Objective Loss 0.342921 LR 0.001000 Time 0.025650 -2023-02-13 17:45:24,265 - Epoch: [61][ 120/ 1207] Overall Loss 0.341210 Objective Loss 0.341210 LR 0.001000 Time 0.025117 -2023-02-13 17:45:24,461 - Epoch: [61][ 130/ 1207] Overall Loss 0.340823 Objective Loss 0.340823 LR 0.001000 Time 0.024692 -2023-02-13 17:45:24,653 - Epoch: [61][ 140/ 1207] Overall Loss 0.340099 Objective Loss 0.340099 LR 0.001000 Time 0.024298 -2023-02-13 17:45:24,848 - Epoch: [61][ 150/ 1207] Overall Loss 0.339130 Objective Loss 0.339130 LR 0.001000 Time 0.023973 -2023-02-13 17:45:25,039 - Epoch: [61][ 160/ 1207] Overall Loss 0.338450 Objective Loss 0.338450 LR 0.001000 Time 0.023668 -2023-02-13 17:45:25,233 - Epoch: [61][ 170/ 1207] Overall Loss 0.336492 Objective Loss 0.336492 LR 0.001000 Time 0.023417 -2023-02-13 17:45:25,424 - Epoch: [61][ 180/ 1207] Overall Loss 0.335127 Objective Loss 0.335127 LR 0.001000 Time 0.023174 -2023-02-13 17:45:25,620 - Epoch: [61][ 190/ 1207] Overall Loss 0.334978 Objective Loss 0.334978 LR 0.001000 Time 0.022980 -2023-02-13 17:45:25,812 - Epoch: [61][ 200/ 1207] Overall Loss 0.334382 Objective Loss 0.334382 LR 0.001000 Time 0.022791 -2023-02-13 17:45:26,007 - Epoch: [61][ 210/ 1207] Overall Loss 0.335502 Objective Loss 0.335502 LR 0.001000 Time 0.022632 -2023-02-13 17:45:26,198 - Epoch: [61][ 220/ 1207] Overall Loss 0.336253 Objective Loss 0.336253 LR 0.001000 Time 0.022471 -2023-02-13 17:45:26,393 - Epoch: [61][ 230/ 1207] Overall Loss 0.336752 Objective Loss 0.336752 LR 0.001000 Time 0.022341 -2023-02-13 17:45:26,586 - Epoch: [61][ 240/ 1207] Overall Loss 0.336240 Objective Loss 0.336240 LR 0.001000 Time 0.022211 -2023-02-13 17:45:26,782 - Epoch: [61][ 250/ 1207] Overall Loss 0.335281 Objective Loss 0.335281 LR 0.001000 Time 0.022105 -2023-02-13 17:45:26,974 - Epoch: [61][ 260/ 1207] Overall Loss 0.335617 Objective Loss 0.335617 LR 0.001000 Time 0.021991 -2023-02-13 17:45:27,168 - Epoch: [61][ 270/ 1207] Overall Loss 0.334970 Objective Loss 0.334970 LR 0.001000 Time 0.021896 -2023-02-13 17:45:27,360 - Epoch: [61][ 280/ 1207] Overall Loss 0.333970 Objective Loss 0.333970 LR 0.001000 Time 0.021797 -2023-02-13 17:45:27,555 - Epoch: [61][ 290/ 1207] Overall Loss 0.334546 Objective Loss 0.334546 LR 0.001000 Time 0.021718 -2023-02-13 17:45:27,747 - Epoch: [61][ 300/ 1207] Overall Loss 0.334932 Objective Loss 0.334932 LR 0.001000 Time 0.021633 -2023-02-13 17:45:27,941 - Epoch: [61][ 310/ 1207] Overall Loss 0.335869 Objective Loss 0.335869 LR 0.001000 Time 0.021560 -2023-02-13 17:45:28,132 - Epoch: [61][ 320/ 1207] Overall Loss 0.336921 Objective Loss 0.336921 LR 0.001000 Time 0.021482 -2023-02-13 17:45:28,326 - Epoch: [61][ 330/ 1207] Overall Loss 0.337365 Objective Loss 0.337365 LR 0.001000 Time 0.021417 -2023-02-13 17:45:28,518 - Epoch: [61][ 340/ 1207] Overall Loss 0.337778 Objective Loss 0.337778 LR 0.001000 Time 0.021350 -2023-02-13 17:45:28,713 - Epoch: [61][ 350/ 1207] Overall Loss 0.337311 Objective Loss 0.337311 LR 0.001000 Time 0.021296 -2023-02-13 17:45:28,904 - Epoch: [61][ 360/ 1207] Overall Loss 0.337644 Objective Loss 0.337644 LR 0.001000 Time 0.021234 -2023-02-13 17:45:29,098 - Epoch: [61][ 370/ 1207] Overall Loss 0.337156 Objective Loss 0.337156 LR 0.001000 Time 0.021184 -2023-02-13 17:45:29,289 - Epoch: [61][ 380/ 1207] Overall Loss 0.338373 Objective Loss 0.338373 LR 0.001000 Time 0.021129 -2023-02-13 17:45:29,483 - Epoch: [61][ 390/ 1207] Overall Loss 0.338181 Objective Loss 0.338181 LR 0.001000 Time 0.021083 -2023-02-13 17:45:29,676 - Epoch: [61][ 400/ 1207] Overall Loss 0.338067 Objective Loss 0.338067 LR 0.001000 Time 0.021037 -2023-02-13 17:45:29,870 - Epoch: [61][ 410/ 1207] Overall Loss 0.338247 Objective Loss 0.338247 LR 0.001000 Time 0.020997 -2023-02-13 17:45:30,062 - Epoch: [61][ 420/ 1207] Overall Loss 0.338393 Objective Loss 0.338393 LR 0.001000 Time 0.020952 -2023-02-13 17:45:30,256 - Epoch: [61][ 430/ 1207] Overall Loss 0.338856 Objective Loss 0.338856 LR 0.001000 Time 0.020916 -2023-02-13 17:45:30,447 - Epoch: [61][ 440/ 1207] Overall Loss 0.338935 Objective Loss 0.338935 LR 0.001000 Time 0.020874 -2023-02-13 17:45:30,643 - Epoch: [61][ 450/ 1207] Overall Loss 0.339792 Objective Loss 0.339792 LR 0.001000 Time 0.020845 -2023-02-13 17:45:30,835 - Epoch: [61][ 460/ 1207] Overall Loss 0.339970 Objective Loss 0.339970 LR 0.001000 Time 0.020809 -2023-02-13 17:45:31,030 - Epoch: [61][ 470/ 1207] Overall Loss 0.340025 Objective Loss 0.340025 LR 0.001000 Time 0.020780 -2023-02-13 17:45:31,222 - Epoch: [61][ 480/ 1207] Overall Loss 0.340359 Objective Loss 0.340359 LR 0.001000 Time 0.020745 -2023-02-13 17:45:31,417 - Epoch: [61][ 490/ 1207] Overall Loss 0.340245 Objective Loss 0.340245 LR 0.001000 Time 0.020719 -2023-02-13 17:45:31,610 - Epoch: [61][ 500/ 1207] Overall Loss 0.340238 Objective Loss 0.340238 LR 0.001000 Time 0.020689 -2023-02-13 17:45:31,805 - Epoch: [61][ 510/ 1207] Overall Loss 0.340481 Objective Loss 0.340481 LR 0.001000 Time 0.020667 -2023-02-13 17:45:31,998 - Epoch: [61][ 520/ 1207] Overall Loss 0.340616 Objective Loss 0.340616 LR 0.001000 Time 0.020638 -2023-02-13 17:45:32,192 - Epoch: [61][ 530/ 1207] Overall Loss 0.340841 Objective Loss 0.340841 LR 0.001000 Time 0.020615 -2023-02-13 17:45:32,384 - Epoch: [61][ 540/ 1207] Overall Loss 0.341138 Objective Loss 0.341138 LR 0.001000 Time 0.020587 -2023-02-13 17:45:32,578 - Epoch: [61][ 550/ 1207] Overall Loss 0.341162 Objective Loss 0.341162 LR 0.001000 Time 0.020567 -2023-02-13 17:45:32,771 - Epoch: [61][ 560/ 1207] Overall Loss 0.341175 Objective Loss 0.341175 LR 0.001000 Time 0.020543 -2023-02-13 17:45:32,966 - Epoch: [61][ 570/ 1207] Overall Loss 0.340943 Objective Loss 0.340943 LR 0.001000 Time 0.020523 -2023-02-13 17:45:33,157 - Epoch: [61][ 580/ 1207] Overall Loss 0.340584 Objective Loss 0.340584 LR 0.001000 Time 0.020499 -2023-02-13 17:45:33,352 - Epoch: [61][ 590/ 1207] Overall Loss 0.340732 Objective Loss 0.340732 LR 0.001000 Time 0.020481 -2023-02-13 17:45:33,544 - Epoch: [61][ 600/ 1207] Overall Loss 0.340737 Objective Loss 0.340737 LR 0.001000 Time 0.020460 -2023-02-13 17:45:33,740 - Epoch: [61][ 610/ 1207] Overall Loss 0.340492 Objective Loss 0.340492 LR 0.001000 Time 0.020444 -2023-02-13 17:45:33,933 - Epoch: [61][ 620/ 1207] Overall Loss 0.340352 Objective Loss 0.340352 LR 0.001000 Time 0.020425 -2023-02-13 17:45:34,128 - Epoch: [61][ 630/ 1207] Overall Loss 0.340092 Objective Loss 0.340092 LR 0.001000 Time 0.020410 -2023-02-13 17:45:34,319 - Epoch: [61][ 640/ 1207] Overall Loss 0.339588 Objective Loss 0.339588 LR 0.001000 Time 0.020389 -2023-02-13 17:45:34,515 - Epoch: [61][ 650/ 1207] Overall Loss 0.339339 Objective Loss 0.339339 LR 0.001000 Time 0.020376 -2023-02-13 17:45:34,707 - Epoch: [61][ 660/ 1207] Overall Loss 0.339414 Objective Loss 0.339414 LR 0.001000 Time 0.020357 -2023-02-13 17:45:34,902 - Epoch: [61][ 670/ 1207] Overall Loss 0.339385 Objective Loss 0.339385 LR 0.001000 Time 0.020345 -2023-02-13 17:45:35,094 - Epoch: [61][ 680/ 1207] Overall Loss 0.339475 Objective Loss 0.339475 LR 0.001000 Time 0.020327 -2023-02-13 17:45:35,289 - Epoch: [61][ 690/ 1207] Overall Loss 0.339507 Objective Loss 0.339507 LR 0.001000 Time 0.020315 -2023-02-13 17:45:35,481 - Epoch: [61][ 700/ 1207] Overall Loss 0.339761 Objective Loss 0.339761 LR 0.001000 Time 0.020298 -2023-02-13 17:45:35,677 - Epoch: [61][ 710/ 1207] Overall Loss 0.339716 Objective Loss 0.339716 LR 0.001000 Time 0.020287 -2023-02-13 17:45:35,870 - Epoch: [61][ 720/ 1207] Overall Loss 0.339940 Objective Loss 0.339940 LR 0.001000 Time 0.020274 -2023-02-13 17:45:36,065 - Epoch: [61][ 730/ 1207] Overall Loss 0.339955 Objective Loss 0.339955 LR 0.001000 Time 0.020262 -2023-02-13 17:45:36,257 - Epoch: [61][ 740/ 1207] Overall Loss 0.339656 Objective Loss 0.339656 LR 0.001000 Time 0.020247 -2023-02-13 17:45:36,452 - Epoch: [61][ 750/ 1207] Overall Loss 0.339520 Objective Loss 0.339520 LR 0.001000 Time 0.020237 -2023-02-13 17:45:36,645 - Epoch: [61][ 760/ 1207] Overall Loss 0.339835 Objective Loss 0.339835 LR 0.001000 Time 0.020224 -2023-02-13 17:45:36,841 - Epoch: [61][ 770/ 1207] Overall Loss 0.339922 Objective Loss 0.339922 LR 0.001000 Time 0.020215 -2023-02-13 17:45:37,033 - Epoch: [61][ 780/ 1207] Overall Loss 0.339769 Objective Loss 0.339769 LR 0.001000 Time 0.020202 -2023-02-13 17:45:37,228 - Epoch: [61][ 790/ 1207] Overall Loss 0.339999 Objective Loss 0.339999 LR 0.001000 Time 0.020192 -2023-02-13 17:45:37,419 - Epoch: [61][ 800/ 1207] Overall Loss 0.340503 Objective Loss 0.340503 LR 0.001000 Time 0.020179 -2023-02-13 17:45:37,615 - Epoch: [61][ 810/ 1207] Overall Loss 0.340445 Objective Loss 0.340445 LR 0.001000 Time 0.020171 -2023-02-13 17:45:37,807 - Epoch: [61][ 820/ 1207] Overall Loss 0.341011 Objective Loss 0.341011 LR 0.001000 Time 0.020158 -2023-02-13 17:45:38,002 - Epoch: [61][ 830/ 1207] Overall Loss 0.340940 Objective Loss 0.340940 LR 0.001000 Time 0.020150 -2023-02-13 17:45:38,194 - Epoch: [61][ 840/ 1207] Overall Loss 0.341171 Objective Loss 0.341171 LR 0.001000 Time 0.020138 -2023-02-13 17:45:38,388 - Epoch: [61][ 850/ 1207] Overall Loss 0.341369 Objective Loss 0.341369 LR 0.001000 Time 0.020129 -2023-02-13 17:45:38,580 - Epoch: [61][ 860/ 1207] Overall Loss 0.341448 Objective Loss 0.341448 LR 0.001000 Time 0.020118 -2023-02-13 17:45:38,775 - Epoch: [61][ 870/ 1207] Overall Loss 0.341097 Objective Loss 0.341097 LR 0.001000 Time 0.020110 -2023-02-13 17:45:38,968 - Epoch: [61][ 880/ 1207] Overall Loss 0.340737 Objective Loss 0.340737 LR 0.001000 Time 0.020100 -2023-02-13 17:45:39,162 - Epoch: [61][ 890/ 1207] Overall Loss 0.340695 Objective Loss 0.340695 LR 0.001000 Time 0.020093 -2023-02-13 17:45:39,354 - Epoch: [61][ 900/ 1207] Overall Loss 0.340929 Objective Loss 0.340929 LR 0.001000 Time 0.020083 -2023-02-13 17:45:39,550 - Epoch: [61][ 910/ 1207] Overall Loss 0.341223 Objective Loss 0.341223 LR 0.001000 Time 0.020076 -2023-02-13 17:45:39,742 - Epoch: [61][ 920/ 1207] Overall Loss 0.341099 Objective Loss 0.341099 LR 0.001000 Time 0.020067 -2023-02-13 17:45:39,937 - Epoch: [61][ 930/ 1207] Overall Loss 0.341640 Objective Loss 0.341640 LR 0.001000 Time 0.020061 -2023-02-13 17:45:40,129 - Epoch: [61][ 940/ 1207] Overall Loss 0.341409 Objective Loss 0.341409 LR 0.001000 Time 0.020051 -2023-02-13 17:45:40,324 - Epoch: [61][ 950/ 1207] Overall Loss 0.341359 Objective Loss 0.341359 LR 0.001000 Time 0.020044 -2023-02-13 17:45:40,515 - Epoch: [61][ 960/ 1207] Overall Loss 0.341485 Objective Loss 0.341485 LR 0.001000 Time 0.020034 -2023-02-13 17:45:40,711 - Epoch: [61][ 970/ 1207] Overall Loss 0.341349 Objective Loss 0.341349 LR 0.001000 Time 0.020029 -2023-02-13 17:45:40,904 - Epoch: [61][ 980/ 1207] Overall Loss 0.341353 Objective Loss 0.341353 LR 0.001000 Time 0.020021 -2023-02-13 17:45:41,098 - Epoch: [61][ 990/ 1207] Overall Loss 0.341552 Objective Loss 0.341552 LR 0.001000 Time 0.020015 -2023-02-13 17:45:41,290 - Epoch: [61][ 1000/ 1207] Overall Loss 0.341709 Objective Loss 0.341709 LR 0.001000 Time 0.020006 -2023-02-13 17:45:41,484 - Epoch: [61][ 1010/ 1207] Overall Loss 0.341781 Objective Loss 0.341781 LR 0.001000 Time 0.020000 -2023-02-13 17:45:41,677 - Epoch: [61][ 1020/ 1207] Overall Loss 0.341795 Objective Loss 0.341795 LR 0.001000 Time 0.019992 -2023-02-13 17:45:41,873 - Epoch: [61][ 1030/ 1207] Overall Loss 0.342230 Objective Loss 0.342230 LR 0.001000 Time 0.019988 -2023-02-13 17:45:42,065 - Epoch: [61][ 1040/ 1207] Overall Loss 0.342590 Objective Loss 0.342590 LR 0.001000 Time 0.019980 -2023-02-13 17:45:42,259 - Epoch: [61][ 1050/ 1207] Overall Loss 0.342855 Objective Loss 0.342855 LR 0.001000 Time 0.019975 -2023-02-13 17:45:42,451 - Epoch: [61][ 1060/ 1207] Overall Loss 0.343139 Objective Loss 0.343139 LR 0.001000 Time 0.019967 -2023-02-13 17:45:42,647 - Epoch: [61][ 1070/ 1207] Overall Loss 0.343054 Objective Loss 0.343054 LR 0.001000 Time 0.019963 -2023-02-13 17:45:42,839 - Epoch: [61][ 1080/ 1207] Overall Loss 0.343118 Objective Loss 0.343118 LR 0.001000 Time 0.019956 -2023-02-13 17:45:43,034 - Epoch: [61][ 1090/ 1207] Overall Loss 0.343019 Objective Loss 0.343019 LR 0.001000 Time 0.019951 -2023-02-13 17:45:43,226 - Epoch: [61][ 1100/ 1207] Overall Loss 0.343235 Objective Loss 0.343235 LR 0.001000 Time 0.019944 -2023-02-13 17:45:43,420 - Epoch: [61][ 1110/ 1207] Overall Loss 0.343431 Objective Loss 0.343431 LR 0.001000 Time 0.019939 -2023-02-13 17:45:43,613 - Epoch: [61][ 1120/ 1207] Overall Loss 0.343333 Objective Loss 0.343333 LR 0.001000 Time 0.019933 -2023-02-13 17:45:43,808 - Epoch: [61][ 1130/ 1207] Overall Loss 0.343110 Objective Loss 0.343110 LR 0.001000 Time 0.019929 -2023-02-13 17:45:44,000 - Epoch: [61][ 1140/ 1207] Overall Loss 0.343404 Objective Loss 0.343404 LR 0.001000 Time 0.019922 -2023-02-13 17:45:44,195 - Epoch: [61][ 1150/ 1207] Overall Loss 0.343472 Objective Loss 0.343472 LR 0.001000 Time 0.019918 -2023-02-13 17:45:44,386 - Epoch: [61][ 1160/ 1207] Overall Loss 0.343528 Objective Loss 0.343528 LR 0.001000 Time 0.019911 -2023-02-13 17:45:44,581 - Epoch: [61][ 1170/ 1207] Overall Loss 0.343152 Objective Loss 0.343152 LR 0.001000 Time 0.019907 -2023-02-13 17:45:44,773 - Epoch: [61][ 1180/ 1207] Overall Loss 0.342990 Objective Loss 0.342990 LR 0.001000 Time 0.019901 -2023-02-13 17:45:44,969 - Epoch: [61][ 1190/ 1207] Overall Loss 0.343042 Objective Loss 0.343042 LR 0.001000 Time 0.019897 -2023-02-13 17:45:45,217 - Epoch: [61][ 1200/ 1207] Overall Loss 0.342984 Objective Loss 0.342984 LR 0.001000 Time 0.019938 -2023-02-13 17:45:45,331 - Epoch: [61][ 1207/ 1207] Overall Loss 0.342986 Objective Loss 0.342986 Top1 82.317073 Top5 96.951220 LR 0.001000 Time 0.019917 -2023-02-13 17:45:45,402 - --- validate (epoch=61)----------- -2023-02-13 17:45:45,402 - 34311 samples (256 per mini-batch) -2023-02-13 17:45:45,906 - Epoch: [61][ 10/ 135] Loss 0.390573 Top1 80.468750 Top5 96.796875 -2023-02-13 17:45:46,036 - Epoch: [61][ 20/ 135] Loss 0.398445 Top1 81.152344 Top5 96.621094 -2023-02-13 17:45:46,163 - Epoch: [61][ 30/ 135] Loss 0.393555 Top1 81.328125 Top5 96.549479 -2023-02-13 17:45:46,289 - Epoch: [61][ 40/ 135] Loss 0.383865 Top1 81.699219 Top5 96.630859 -2023-02-13 17:45:46,416 - Epoch: [61][ 50/ 135] Loss 0.382375 Top1 81.820312 Top5 96.664062 -2023-02-13 17:45:46,542 - Epoch: [61][ 60/ 135] Loss 0.381816 Top1 81.868490 Top5 96.686198 -2023-02-13 17:45:46,669 - Epoch: [61][ 70/ 135] Loss 0.379950 Top1 81.841518 Top5 96.785714 -2023-02-13 17:45:46,796 - Epoch: [61][ 80/ 135] Loss 0.376646 Top1 81.865234 Top5 96.831055 -2023-02-13 17:45:46,925 - Epoch: [61][ 90/ 135] Loss 0.374604 Top1 81.914062 Top5 96.853299 -2023-02-13 17:45:47,052 - Epoch: [61][ 100/ 135] Loss 0.371621 Top1 81.882812 Top5 96.796875 -2023-02-13 17:45:47,179 - Epoch: [61][ 110/ 135] Loss 0.371056 Top1 81.796875 Top5 96.832386 -2023-02-13 17:45:47,306 - Epoch: [61][ 120/ 135] Loss 0.370224 Top1 81.761068 Top5 96.839193 -2023-02-13 17:45:47,435 - Epoch: [61][ 130/ 135] Loss 0.373289 Top1 81.688702 Top5 96.817909 -2023-02-13 17:45:47,481 - Epoch: [61][ 135/ 135] Loss 0.369153 Top1 81.723063 Top5 96.846492 -2023-02-13 17:45:47,552 - ==> Top1: 81.723 Top5: 96.846 Loss: 0.369 - -2023-02-13 17:45:47,553 - ==> Confusion: -[[ 867 4 13 0 10 6 0 1 1 32 1 3 2 3 6 2 3 3 2 0 8] - [ 1 914 3 1 7 42 3 19 1 3 3 4 2 1 6 3 3 0 10 4 3] - [ 13 2 957 8 3 2 19 13 1 0 3 1 3 5 1 3 1 2 10 6 5] - [ 6 0 34 889 4 4 2 3 1 3 14 0 3 2 19 3 2 3 17 0 7] - [ 24 7 1 1 981 8 1 2 0 3 0 5 0 3 12 6 4 0 1 3 4] - [ 3 16 2 5 6 957 5 16 1 5 2 16 5 13 3 4 2 1 2 5 1] - [ 2 4 25 3 1 6 1027 10 0 0 2 1 3 1 0 4 0 2 2 5 1] - [ 1 4 6 4 2 25 4 926 0 1 3 6 2 1 0 0 1 1 26 10 1] - [ 31 4 1 2 1 2 0 1 824 55 14 2 1 13 42 2 2 2 9 0 1] - [ 128 1 6 0 6 2 0 2 18 810 0 1 1 16 8 2 1 2 1 1 6] - [ 4 2 8 6 1 2 3 3 17 2 968 2 0 8 3 0 2 1 14 0 5] - [ 1 0 0 0 3 12 1 8 3 1 1 909 20 8 2 3 7 8 2 14 2] - [ 4 0 3 12 2 3 1 1 2 0 1 43 836 1 6 13 3 18 3 2 5] - [ 4 3 6 0 9 11 1 2 6 30 16 8 2 895 7 3 5 3 2 7 4] - [ 13 2 5 18 5 4 0 2 6 3 2 2 7 2 991 1 2 3 16 0 8] - [ 6 2 6 1 6 3 3 1 1 0 0 9 5 4 1 975 7 5 0 6 5] - [ 5 11 2 0 14 3 0 0 0 0 0 2 0 4 1 11 983 1 4 7 13] - [ 10 2 0 7 1 2 1 2 0 1 0 14 20 1 3 27 1 949 2 2 6] - [ 5 3 7 15 1 1 0 28 0 0 5 3 6 0 9 1 2 1 995 1 3] - [ 1 3 4 1 0 3 8 16 1 0 1 25 2 6 0 4 3 3 3 1058 6] - [ 197 222 326 170 194 251 104 231 80 93 194 184 308 337 203 131 274 106 209 291 9329]] - -2023-02-13 17:45:47,554 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:45:47,554 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:45:47,560 - - -2023-02-13 17:45:47,560 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:45:48,433 - Epoch: [62][ 10/ 1207] Overall Loss 0.334543 Objective Loss 0.334543 LR 0.001000 Time 0.087245 -2023-02-13 17:45:48,623 - Epoch: [62][ 20/ 1207] Overall Loss 0.334609 Objective Loss 0.334609 LR 0.001000 Time 0.053095 -2023-02-13 17:45:48,812 - Epoch: [62][ 30/ 1207] Overall Loss 0.341566 Objective Loss 0.341566 LR 0.001000 Time 0.041679 -2023-02-13 17:45:49,001 - Epoch: [62][ 40/ 1207] Overall Loss 0.345635 Objective Loss 0.345635 LR 0.001000 Time 0.035967 -2023-02-13 17:45:49,189 - Epoch: [62][ 50/ 1207] Overall Loss 0.342879 Objective Loss 0.342879 LR 0.001000 Time 0.032531 -2023-02-13 17:45:49,377 - Epoch: [62][ 60/ 1207] Overall Loss 0.345641 Objective Loss 0.345641 LR 0.001000 Time 0.030238 -2023-02-13 17:45:49,565 - Epoch: [62][ 70/ 1207] Overall Loss 0.345780 Objective Loss 0.345780 LR 0.001000 Time 0.028598 -2023-02-13 17:45:49,754 - Epoch: [62][ 80/ 1207] Overall Loss 0.342693 Objective Loss 0.342693 LR 0.001000 Time 0.027376 -2023-02-13 17:45:49,941 - Epoch: [62][ 90/ 1207] Overall Loss 0.344854 Objective Loss 0.344854 LR 0.001000 Time 0.026415 -2023-02-13 17:45:50,129 - Epoch: [62][ 100/ 1207] Overall Loss 0.344306 Objective Loss 0.344306 LR 0.001000 Time 0.025648 -2023-02-13 17:45:50,316 - Epoch: [62][ 110/ 1207] Overall Loss 0.343036 Objective Loss 0.343036 LR 0.001000 Time 0.025015 -2023-02-13 17:45:50,504 - Epoch: [62][ 120/ 1207] Overall Loss 0.341985 Objective Loss 0.341985 LR 0.001000 Time 0.024492 -2023-02-13 17:45:50,693 - Epoch: [62][ 130/ 1207] Overall Loss 0.341224 Objective Loss 0.341224 LR 0.001000 Time 0.024055 -2023-02-13 17:45:50,882 - Epoch: [62][ 140/ 1207] Overall Loss 0.340699 Objective Loss 0.340699 LR 0.001000 Time 0.023688 -2023-02-13 17:45:51,069 - Epoch: [62][ 150/ 1207] Overall Loss 0.340049 Objective Loss 0.340049 LR 0.001000 Time 0.023355 -2023-02-13 17:45:51,257 - Epoch: [62][ 160/ 1207] Overall Loss 0.340573 Objective Loss 0.340573 LR 0.001000 Time 0.023067 -2023-02-13 17:45:51,445 - Epoch: [62][ 170/ 1207] Overall Loss 0.339763 Objective Loss 0.339763 LR 0.001000 Time 0.022813 -2023-02-13 17:45:51,634 - Epoch: [62][ 180/ 1207] Overall Loss 0.339770 Objective Loss 0.339770 LR 0.001000 Time 0.022591 -2023-02-13 17:45:51,822 - Epoch: [62][ 190/ 1207] Overall Loss 0.337525 Objective Loss 0.337525 LR 0.001000 Time 0.022394 -2023-02-13 17:45:52,011 - Epoch: [62][ 200/ 1207] Overall Loss 0.339710 Objective Loss 0.339710 LR 0.001000 Time 0.022213 -2023-02-13 17:45:52,198 - Epoch: [62][ 210/ 1207] Overall Loss 0.340098 Objective Loss 0.340098 LR 0.001000 Time 0.022046 -2023-02-13 17:45:52,386 - Epoch: [62][ 220/ 1207] Overall Loss 0.340549 Objective Loss 0.340549 LR 0.001000 Time 0.021897 -2023-02-13 17:45:52,574 - Epoch: [62][ 230/ 1207] Overall Loss 0.340002 Objective Loss 0.340002 LR 0.001000 Time 0.021760 -2023-02-13 17:45:52,763 - Epoch: [62][ 240/ 1207] Overall Loss 0.339737 Objective Loss 0.339737 LR 0.001000 Time 0.021638 -2023-02-13 17:45:52,951 - Epoch: [62][ 250/ 1207] Overall Loss 0.339603 Objective Loss 0.339603 LR 0.001000 Time 0.021524 -2023-02-13 17:45:53,139 - Epoch: [62][ 260/ 1207] Overall Loss 0.339094 Objective Loss 0.339094 LR 0.001000 Time 0.021418 -2023-02-13 17:45:53,326 - Epoch: [62][ 270/ 1207] Overall Loss 0.339047 Objective Loss 0.339047 LR 0.001000 Time 0.021317 -2023-02-13 17:45:53,514 - Epoch: [62][ 280/ 1207] Overall Loss 0.339085 Objective Loss 0.339085 LR 0.001000 Time 0.021226 -2023-02-13 17:45:53,703 - Epoch: [62][ 290/ 1207] Overall Loss 0.338210 Objective Loss 0.338210 LR 0.001000 Time 0.021142 -2023-02-13 17:45:53,891 - Epoch: [62][ 300/ 1207] Overall Loss 0.338131 Objective Loss 0.338131 LR 0.001000 Time 0.021064 -2023-02-13 17:45:54,079 - Epoch: [62][ 310/ 1207] Overall Loss 0.338558 Objective Loss 0.338558 LR 0.001000 Time 0.020989 -2023-02-13 17:45:54,267 - Epoch: [62][ 320/ 1207] Overall Loss 0.338499 Objective Loss 0.338499 LR 0.001000 Time 0.020920 -2023-02-13 17:45:54,455 - Epoch: [62][ 330/ 1207] Overall Loss 0.339252 Objective Loss 0.339252 LR 0.001000 Time 0.020854 -2023-02-13 17:45:54,643 - Epoch: [62][ 340/ 1207] Overall Loss 0.340532 Objective Loss 0.340532 LR 0.001000 Time 0.020794 -2023-02-13 17:45:54,832 - Epoch: [62][ 350/ 1207] Overall Loss 0.340351 Objective Loss 0.340351 LR 0.001000 Time 0.020738 -2023-02-13 17:45:55,020 - Epoch: [62][ 360/ 1207] Overall Loss 0.340127 Objective Loss 0.340127 LR 0.001000 Time 0.020684 -2023-02-13 17:45:55,208 - Epoch: [62][ 370/ 1207] Overall Loss 0.340508 Objective Loss 0.340508 LR 0.001000 Time 0.020632 -2023-02-13 17:45:55,397 - Epoch: [62][ 380/ 1207] Overall Loss 0.340721 Objective Loss 0.340721 LR 0.001000 Time 0.020584 -2023-02-13 17:45:55,585 - Epoch: [62][ 390/ 1207] Overall Loss 0.340850 Objective Loss 0.340850 LR 0.001000 Time 0.020538 -2023-02-13 17:45:55,774 - Epoch: [62][ 400/ 1207] Overall Loss 0.340675 Objective Loss 0.340675 LR 0.001000 Time 0.020495 -2023-02-13 17:45:55,963 - Epoch: [62][ 410/ 1207] Overall Loss 0.341028 Objective Loss 0.341028 LR 0.001000 Time 0.020456 -2023-02-13 17:45:56,150 - Epoch: [62][ 420/ 1207] Overall Loss 0.341526 Objective Loss 0.341526 LR 0.001000 Time 0.020414 -2023-02-13 17:45:56,339 - Epoch: [62][ 430/ 1207] Overall Loss 0.341005 Objective Loss 0.341005 LR 0.001000 Time 0.020377 -2023-02-13 17:45:56,528 - Epoch: [62][ 440/ 1207] Overall Loss 0.341153 Objective Loss 0.341153 LR 0.001000 Time 0.020342 -2023-02-13 17:45:56,717 - Epoch: [62][ 450/ 1207] Overall Loss 0.341577 Objective Loss 0.341577 LR 0.001000 Time 0.020310 -2023-02-13 17:45:56,908 - Epoch: [62][ 460/ 1207] Overall Loss 0.341334 Objective Loss 0.341334 LR 0.001000 Time 0.020283 -2023-02-13 17:45:57,099 - Epoch: [62][ 470/ 1207] Overall Loss 0.341274 Objective Loss 0.341274 LR 0.001000 Time 0.020257 -2023-02-13 17:45:57,289 - Epoch: [62][ 480/ 1207] Overall Loss 0.342580 Objective Loss 0.342580 LR 0.001000 Time 0.020229 -2023-02-13 17:45:57,480 - Epoch: [62][ 490/ 1207] Overall Loss 0.342554 Objective Loss 0.342554 LR 0.001000 Time 0.020207 -2023-02-13 17:45:57,670 - Epoch: [62][ 500/ 1207] Overall Loss 0.342494 Objective Loss 0.342494 LR 0.001000 Time 0.020182 -2023-02-13 17:45:57,861 - Epoch: [62][ 510/ 1207] Overall Loss 0.342511 Objective Loss 0.342511 LR 0.001000 Time 0.020159 -2023-02-13 17:45:58,050 - Epoch: [62][ 520/ 1207] Overall Loss 0.342595 Objective Loss 0.342595 LR 0.001000 Time 0.020135 -2023-02-13 17:45:58,240 - Epoch: [62][ 530/ 1207] Overall Loss 0.342439 Objective Loss 0.342439 LR 0.001000 Time 0.020113 -2023-02-13 17:45:58,430 - Epoch: [62][ 540/ 1207] Overall Loss 0.342226 Objective Loss 0.342226 LR 0.001000 Time 0.020092 -2023-02-13 17:45:58,622 - Epoch: [62][ 550/ 1207] Overall Loss 0.342277 Objective Loss 0.342277 LR 0.001000 Time 0.020074 -2023-02-13 17:45:58,812 - Epoch: [62][ 560/ 1207] Overall Loss 0.341948 Objective Loss 0.341948 LR 0.001000 Time 0.020055 -2023-02-13 17:45:59,003 - Epoch: [62][ 570/ 1207] Overall Loss 0.341562 Objective Loss 0.341562 LR 0.001000 Time 0.020037 -2023-02-13 17:45:59,193 - Epoch: [62][ 580/ 1207] Overall Loss 0.341899 Objective Loss 0.341899 LR 0.001000 Time 0.020018 -2023-02-13 17:45:59,383 - Epoch: [62][ 590/ 1207] Overall Loss 0.341546 Objective Loss 0.341546 LR 0.001000 Time 0.020001 -2023-02-13 17:45:59,573 - Epoch: [62][ 600/ 1207] Overall Loss 0.342073 Objective Loss 0.342073 LR 0.001000 Time 0.019984 -2023-02-13 17:45:59,765 - Epoch: [62][ 610/ 1207] Overall Loss 0.342689 Objective Loss 0.342689 LR 0.001000 Time 0.019969 -2023-02-13 17:45:59,955 - Epoch: [62][ 620/ 1207] Overall Loss 0.342607 Objective Loss 0.342607 LR 0.001000 Time 0.019953 -2023-02-13 17:46:00,145 - Epoch: [62][ 630/ 1207] Overall Loss 0.342175 Objective Loss 0.342175 LR 0.001000 Time 0.019938 -2023-02-13 17:46:00,335 - Epoch: [62][ 640/ 1207] Overall Loss 0.342296 Objective Loss 0.342296 LR 0.001000 Time 0.019923 -2023-02-13 17:46:00,526 - Epoch: [62][ 650/ 1207] Overall Loss 0.342240 Objective Loss 0.342240 LR 0.001000 Time 0.019910 -2023-02-13 17:46:00,718 - Epoch: [62][ 660/ 1207] Overall Loss 0.342121 Objective Loss 0.342121 LR 0.001000 Time 0.019898 -2023-02-13 17:46:00,910 - Epoch: [62][ 670/ 1207] Overall Loss 0.341415 Objective Loss 0.341415 LR 0.001000 Time 0.019887 -2023-02-13 17:46:01,100 - Epoch: [62][ 680/ 1207] Overall Loss 0.341631 Objective Loss 0.341631 LR 0.001000 Time 0.019874 -2023-02-13 17:46:01,291 - Epoch: [62][ 690/ 1207] Overall Loss 0.341647 Objective Loss 0.341647 LR 0.001000 Time 0.019861 -2023-02-13 17:46:01,480 - Epoch: [62][ 700/ 1207] Overall Loss 0.341788 Objective Loss 0.341788 LR 0.001000 Time 0.019848 -2023-02-13 17:46:01,672 - Epoch: [62][ 710/ 1207] Overall Loss 0.341807 Objective Loss 0.341807 LR 0.001000 Time 0.019838 -2023-02-13 17:46:01,863 - Epoch: [62][ 720/ 1207] Overall Loss 0.341978 Objective Loss 0.341978 LR 0.001000 Time 0.019827 -2023-02-13 17:46:02,054 - Epoch: [62][ 730/ 1207] Overall Loss 0.342358 Objective Loss 0.342358 LR 0.001000 Time 0.019817 -2023-02-13 17:46:02,245 - Epoch: [62][ 740/ 1207] Overall Loss 0.342366 Objective Loss 0.342366 LR 0.001000 Time 0.019806 -2023-02-13 17:46:02,435 - Epoch: [62][ 750/ 1207] Overall Loss 0.342397 Objective Loss 0.342397 LR 0.001000 Time 0.019795 -2023-02-13 17:46:02,625 - Epoch: [62][ 760/ 1207] Overall Loss 0.342219 Objective Loss 0.342219 LR 0.001000 Time 0.019784 -2023-02-13 17:46:02,817 - Epoch: [62][ 770/ 1207] Overall Loss 0.342400 Objective Loss 0.342400 LR 0.001000 Time 0.019776 -2023-02-13 17:46:03,007 - Epoch: [62][ 780/ 1207] Overall Loss 0.341875 Objective Loss 0.341875 LR 0.001000 Time 0.019766 -2023-02-13 17:46:03,198 - Epoch: [62][ 790/ 1207] Overall Loss 0.342003 Objective Loss 0.342003 LR 0.001000 Time 0.019756 -2023-02-13 17:46:03,388 - Epoch: [62][ 800/ 1207] Overall Loss 0.342000 Objective Loss 0.342000 LR 0.001000 Time 0.019747 -2023-02-13 17:46:03,579 - Epoch: [62][ 810/ 1207] Overall Loss 0.342315 Objective Loss 0.342315 LR 0.001000 Time 0.019739 -2023-02-13 17:46:03,771 - Epoch: [62][ 820/ 1207] Overall Loss 0.342662 Objective Loss 0.342662 LR 0.001000 Time 0.019731 -2023-02-13 17:46:03,962 - Epoch: [62][ 830/ 1207] Overall Loss 0.342703 Objective Loss 0.342703 LR 0.001000 Time 0.019723 -2023-02-13 17:46:04,153 - Epoch: [62][ 840/ 1207] Overall Loss 0.342585 Objective Loss 0.342585 LR 0.001000 Time 0.019715 -2023-02-13 17:46:04,342 - Epoch: [62][ 850/ 1207] Overall Loss 0.342052 Objective Loss 0.342052 LR 0.001000 Time 0.019706 -2023-02-13 17:46:04,531 - Epoch: [62][ 860/ 1207] Overall Loss 0.342127 Objective Loss 0.342127 LR 0.001000 Time 0.019696 -2023-02-13 17:46:04,721 - Epoch: [62][ 870/ 1207] Overall Loss 0.342321 Objective Loss 0.342321 LR 0.001000 Time 0.019688 -2023-02-13 17:46:04,911 - Epoch: [62][ 880/ 1207] Overall Loss 0.342352 Objective Loss 0.342352 LR 0.001000 Time 0.019679 -2023-02-13 17:46:05,100 - Epoch: [62][ 890/ 1207] Overall Loss 0.342179 Objective Loss 0.342179 LR 0.001000 Time 0.019670 -2023-02-13 17:46:05,289 - Epoch: [62][ 900/ 1207] Overall Loss 0.342326 Objective Loss 0.342326 LR 0.001000 Time 0.019661 -2023-02-13 17:46:05,478 - Epoch: [62][ 910/ 1207] Overall Loss 0.341876 Objective Loss 0.341876 LR 0.001000 Time 0.019652 -2023-02-13 17:46:05,668 - Epoch: [62][ 920/ 1207] Overall Loss 0.342229 Objective Loss 0.342229 LR 0.001000 Time 0.019644 -2023-02-13 17:46:05,860 - Epoch: [62][ 930/ 1207] Overall Loss 0.342531 Objective Loss 0.342531 LR 0.001000 Time 0.019639 -2023-02-13 17:46:06,050 - Epoch: [62][ 940/ 1207] Overall Loss 0.342799 Objective Loss 0.342799 LR 0.001000 Time 0.019633 -2023-02-13 17:46:06,241 - Epoch: [62][ 950/ 1207] Overall Loss 0.342451 Objective Loss 0.342451 LR 0.001000 Time 0.019626 -2023-02-13 17:46:06,431 - Epoch: [62][ 960/ 1207] Overall Loss 0.342622 Objective Loss 0.342622 LR 0.001000 Time 0.019619 -2023-02-13 17:46:06,623 - Epoch: [62][ 970/ 1207] Overall Loss 0.342507 Objective Loss 0.342507 LR 0.001000 Time 0.019614 -2023-02-13 17:46:06,814 - Epoch: [62][ 980/ 1207] Overall Loss 0.342685 Objective Loss 0.342685 LR 0.001000 Time 0.019609 -2023-02-13 17:46:07,005 - Epoch: [62][ 990/ 1207] Overall Loss 0.342592 Objective Loss 0.342592 LR 0.001000 Time 0.019604 -2023-02-13 17:46:07,195 - Epoch: [62][ 1000/ 1207] Overall Loss 0.342531 Objective Loss 0.342531 LR 0.001000 Time 0.019597 -2023-02-13 17:46:07,385 - Epoch: [62][ 1010/ 1207] Overall Loss 0.342457 Objective Loss 0.342457 LR 0.001000 Time 0.019591 -2023-02-13 17:46:07,575 - Epoch: [62][ 1020/ 1207] Overall Loss 0.342416 Objective Loss 0.342416 LR 0.001000 Time 0.019585 -2023-02-13 17:46:07,768 - Epoch: [62][ 1030/ 1207] Overall Loss 0.342545 Objective Loss 0.342545 LR 0.001000 Time 0.019581 -2023-02-13 17:46:07,958 - Epoch: [62][ 1040/ 1207] Overall Loss 0.342532 Objective Loss 0.342532 LR 0.001000 Time 0.019576 -2023-02-13 17:46:08,149 - Epoch: [62][ 1050/ 1207] Overall Loss 0.342624 Objective Loss 0.342624 LR 0.001000 Time 0.019570 -2023-02-13 17:46:08,338 - Epoch: [62][ 1060/ 1207] Overall Loss 0.342738 Objective Loss 0.342738 LR 0.001000 Time 0.019564 -2023-02-13 17:46:08,529 - Epoch: [62][ 1070/ 1207] Overall Loss 0.342776 Objective Loss 0.342776 LR 0.001000 Time 0.019559 -2023-02-13 17:46:08,720 - Epoch: [62][ 1080/ 1207] Overall Loss 0.342765 Objective Loss 0.342765 LR 0.001000 Time 0.019554 -2023-02-13 17:46:08,912 - Epoch: [62][ 1090/ 1207] Overall Loss 0.342784 Objective Loss 0.342784 LR 0.001000 Time 0.019551 -2023-02-13 17:46:09,101 - Epoch: [62][ 1100/ 1207] Overall Loss 0.342736 Objective Loss 0.342736 LR 0.001000 Time 0.019544 -2023-02-13 17:46:09,289 - Epoch: [62][ 1110/ 1207] Overall Loss 0.342478 Objective Loss 0.342478 LR 0.001000 Time 0.019538 -2023-02-13 17:46:09,478 - Epoch: [62][ 1120/ 1207] Overall Loss 0.342445 Objective Loss 0.342445 LR 0.001000 Time 0.019531 -2023-02-13 17:46:09,667 - Epoch: [62][ 1130/ 1207] Overall Loss 0.342548 Objective Loss 0.342548 LR 0.001000 Time 0.019526 -2023-02-13 17:46:09,859 - Epoch: [62][ 1140/ 1207] Overall Loss 0.342779 Objective Loss 0.342779 LR 0.001000 Time 0.019523 -2023-02-13 17:46:10,050 - Epoch: [62][ 1150/ 1207] Overall Loss 0.342637 Objective Loss 0.342637 LR 0.001000 Time 0.019519 -2023-02-13 17:46:10,242 - Epoch: [62][ 1160/ 1207] Overall Loss 0.342612 Objective Loss 0.342612 LR 0.001000 Time 0.019515 -2023-02-13 17:46:10,433 - Epoch: [62][ 1170/ 1207] Overall Loss 0.342451 Objective Loss 0.342451 LR 0.001000 Time 0.019511 -2023-02-13 17:46:10,625 - Epoch: [62][ 1180/ 1207] Overall Loss 0.342542 Objective Loss 0.342542 LR 0.001000 Time 0.019508 -2023-02-13 17:46:10,817 - Epoch: [62][ 1190/ 1207] Overall Loss 0.342469 Objective Loss 0.342469 LR 0.001000 Time 0.019505 -2023-02-13 17:46:11,065 - Epoch: [62][ 1200/ 1207] Overall Loss 0.342588 Objective Loss 0.342588 LR 0.001000 Time 0.019550 -2023-02-13 17:46:11,180 - Epoch: [62][ 1207/ 1207] Overall Loss 0.342576 Objective Loss 0.342576 Top1 83.536585 Top5 98.170732 LR 0.001000 Time 0.019531 -2023-02-13 17:46:11,257 - --- validate (epoch=62)----------- -2023-02-13 17:46:11,257 - 34311 samples (256 per mini-batch) -2023-02-13 17:46:11,660 - Epoch: [62][ 10/ 135] Loss 0.387186 Top1 80.742188 Top5 97.109375 -2023-02-13 17:46:11,786 - Epoch: [62][ 20/ 135] Loss 0.380652 Top1 80.839844 Top5 97.128906 -2023-02-13 17:46:11,913 - Epoch: [62][ 30/ 135] Loss 0.383667 Top1 80.950521 Top5 97.148438 -2023-02-13 17:46:12,037 - Epoch: [62][ 40/ 135] Loss 0.375319 Top1 81.484375 Top5 97.167969 -2023-02-13 17:46:12,163 - Epoch: [62][ 50/ 135] Loss 0.379722 Top1 81.406250 Top5 96.968750 -2023-02-13 17:46:12,287 - Epoch: [62][ 60/ 135] Loss 0.374226 Top1 81.464844 Top5 96.985677 -2023-02-13 17:46:12,418 - Epoch: [62][ 70/ 135] Loss 0.377848 Top1 81.439732 Top5 96.992188 -2023-02-13 17:46:12,542 - Epoch: [62][ 80/ 135] Loss 0.377996 Top1 81.279297 Top5 97.001953 -2023-02-13 17:46:12,668 - Epoch: [62][ 90/ 135] Loss 0.374067 Top1 81.427951 Top5 97.057292 -2023-02-13 17:46:12,796 - Epoch: [62][ 100/ 135] Loss 0.371987 Top1 81.562500 Top5 97.101562 -2023-02-13 17:46:12,927 - Epoch: [62][ 110/ 135] Loss 0.373355 Top1 81.534091 Top5 97.105824 -2023-02-13 17:46:13,058 - Epoch: [62][ 120/ 135] Loss 0.374242 Top1 81.510417 Top5 97.070312 -2023-02-13 17:46:13,191 - Epoch: [62][ 130/ 135] Loss 0.371613 Top1 81.520433 Top5 97.052284 -2023-02-13 17:46:13,239 - Epoch: [62][ 135/ 135] Loss 0.372686 Top1 81.486987 Top5 97.038851 -2023-02-13 17:46:13,316 - ==> Top1: 81.487 Top5: 97.039 Loss: 0.373 - -2023-02-13 17:46:13,317 - ==> Confusion: -[[ 791 6 16 0 12 6 0 2 3 94 1 3 1 4 3 3 5 5 2 1 9] - [ 1 940 1 5 7 31 2 23 3 2 2 1 2 1 0 0 3 0 3 3 3] - [ 7 8 938 9 2 3 30 14 0 0 3 2 2 5 3 4 5 4 6 7 6] - [ 5 4 19 911 1 4 2 3 2 1 10 0 8 0 6 5 3 11 17 0 4] - [ 12 19 1 0 966 14 1 3 0 3 1 5 1 3 11 6 9 2 1 2 6] - [ 5 28 0 3 5 950 7 21 1 2 4 8 4 16 1 1 0 1 4 5 4] - [ 1 8 13 2 0 6 1037 6 0 1 2 0 2 4 0 4 0 1 1 7 4] - [ 1 11 11 4 2 24 5 920 0 1 1 5 6 1 0 1 1 2 10 14 4] - [ 18 6 1 1 1 3 0 1 874 37 16 3 2 13 23 0 2 2 5 0 1] - [ 49 3 3 2 4 4 0 2 53 851 1 2 1 20 6 1 0 3 1 0 6] - [ 2 4 4 6 0 4 2 5 12 2 976 2 4 8 6 0 1 3 7 1 2] - [ 2 4 4 1 2 19 2 6 0 1 0 872 32 8 0 11 3 17 1 20 0] - [ 1 0 1 6 3 5 1 3 0 0 0 26 861 1 0 5 3 33 2 3 5] - [ 4 4 3 0 6 19 2 4 9 9 11 6 4 924 5 5 2 0 0 4 3] - [ 8 2 4 30 6 5 0 1 17 5 10 2 4 0 960 1 3 13 8 1 12] - [ 4 1 6 1 7 6 6 0 1 0 0 8 9 2 0 952 11 17 0 11 4] - [ 4 14 1 1 10 5 0 1 2 0 0 2 2 2 2 11 974 4 0 7 19] - [ 6 5 0 3 1 3 4 1 0 1 1 8 11 3 0 8 0 989 0 3 4] - [ 2 4 7 15 0 3 0 56 1 0 7 1 8 0 10 1 0 2 959 6 4] - [ 0 3 2 1 1 10 7 16 1 0 0 12 2 7 0 3 1 5 0 1074 3] - [ 132 366 254 182 105 277 95 270 90 87 252 106 386 388 126 107 221 165 162 424 9239]] - -2023-02-13 17:46:13,318 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:46:13,319 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:46:13,324 - - -2023-02-13 17:46:13,325 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:46:14,362 - Epoch: [63][ 10/ 1207] Overall Loss 0.311628 Objective Loss 0.311628 LR 0.001000 Time 0.103663 -2023-02-13 17:46:14,563 - Epoch: [63][ 20/ 1207] Overall Loss 0.308944 Objective Loss 0.308944 LR 0.001000 Time 0.061884 -2023-02-13 17:46:14,755 - Epoch: [63][ 30/ 1207] Overall Loss 0.310492 Objective Loss 0.310492 LR 0.001000 Time 0.047614 -2023-02-13 17:46:14,944 - Epoch: [63][ 40/ 1207] Overall Loss 0.328196 Objective Loss 0.328196 LR 0.001000 Time 0.040446 -2023-02-13 17:46:15,135 - Epoch: [63][ 50/ 1207] Overall Loss 0.324564 Objective Loss 0.324564 LR 0.001000 Time 0.036168 -2023-02-13 17:46:15,325 - Epoch: [63][ 60/ 1207] Overall Loss 0.324706 Objective Loss 0.324706 LR 0.001000 Time 0.033300 -2023-02-13 17:46:15,516 - Epoch: [63][ 70/ 1207] Overall Loss 0.326278 Objective Loss 0.326278 LR 0.001000 Time 0.031265 -2023-02-13 17:46:15,707 - Epoch: [63][ 80/ 1207] Overall Loss 0.327935 Objective Loss 0.327935 LR 0.001000 Time 0.029732 -2023-02-13 17:46:15,899 - Epoch: [63][ 90/ 1207] Overall Loss 0.329962 Objective Loss 0.329962 LR 0.001000 Time 0.028560 -2023-02-13 17:46:16,089 - Epoch: [63][ 100/ 1207] Overall Loss 0.330506 Objective Loss 0.330506 LR 0.001000 Time 0.027601 -2023-02-13 17:46:16,279 - Epoch: [63][ 110/ 1207] Overall Loss 0.329213 Objective Loss 0.329213 LR 0.001000 Time 0.026816 -2023-02-13 17:46:16,469 - Epoch: [63][ 120/ 1207] Overall Loss 0.329924 Objective Loss 0.329924 LR 0.001000 Time 0.026161 -2023-02-13 17:46:16,658 - Epoch: [63][ 130/ 1207] Overall Loss 0.329297 Objective Loss 0.329297 LR 0.001000 Time 0.025604 -2023-02-13 17:46:16,849 - Epoch: [63][ 140/ 1207] Overall Loss 0.332148 Objective Loss 0.332148 LR 0.001000 Time 0.025135 -2023-02-13 17:46:17,039 - Epoch: [63][ 150/ 1207] Overall Loss 0.333849 Objective Loss 0.333849 LR 0.001000 Time 0.024724 -2023-02-13 17:46:17,229 - Epoch: [63][ 160/ 1207] Overall Loss 0.334312 Objective Loss 0.334312 LR 0.001000 Time 0.024364 -2023-02-13 17:46:17,420 - Epoch: [63][ 170/ 1207] Overall Loss 0.334829 Objective Loss 0.334829 LR 0.001000 Time 0.024051 -2023-02-13 17:46:17,610 - Epoch: [63][ 180/ 1207] Overall Loss 0.334194 Objective Loss 0.334194 LR 0.001000 Time 0.023771 -2023-02-13 17:46:17,802 - Epoch: [63][ 190/ 1207] Overall Loss 0.334663 Objective Loss 0.334663 LR 0.001000 Time 0.023524 -2023-02-13 17:46:17,992 - Epoch: [63][ 200/ 1207] Overall Loss 0.335218 Objective Loss 0.335218 LR 0.001000 Time 0.023298 -2023-02-13 17:46:18,183 - Epoch: [63][ 210/ 1207] Overall Loss 0.335295 Objective Loss 0.335295 LR 0.001000 Time 0.023094 -2023-02-13 17:46:18,373 - Epoch: [63][ 220/ 1207] Overall Loss 0.335328 Objective Loss 0.335328 LR 0.001000 Time 0.022910 -2023-02-13 17:46:18,565 - Epoch: [63][ 230/ 1207] Overall Loss 0.334890 Objective Loss 0.334890 LR 0.001000 Time 0.022746 -2023-02-13 17:46:18,757 - Epoch: [63][ 240/ 1207] Overall Loss 0.334520 Objective Loss 0.334520 LR 0.001000 Time 0.022595 -2023-02-13 17:46:18,947 - Epoch: [63][ 250/ 1207] Overall Loss 0.333510 Objective Loss 0.333510 LR 0.001000 Time 0.022452 -2023-02-13 17:46:19,138 - Epoch: [63][ 260/ 1207] Overall Loss 0.333127 Objective Loss 0.333127 LR 0.001000 Time 0.022319 -2023-02-13 17:46:19,329 - Epoch: [63][ 270/ 1207] Overall Loss 0.334060 Objective Loss 0.334060 LR 0.001000 Time 0.022200 -2023-02-13 17:46:19,520 - Epoch: [63][ 280/ 1207] Overall Loss 0.334506 Objective Loss 0.334506 LR 0.001000 Time 0.022087 -2023-02-13 17:46:19,712 - Epoch: [63][ 290/ 1207] Overall Loss 0.335035 Objective Loss 0.335035 LR 0.001000 Time 0.021987 -2023-02-13 17:46:19,903 - Epoch: [63][ 300/ 1207] Overall Loss 0.335676 Objective Loss 0.335676 LR 0.001000 Time 0.021890 -2023-02-13 17:46:20,094 - Epoch: [63][ 310/ 1207] Overall Loss 0.334873 Objective Loss 0.334873 LR 0.001000 Time 0.021798 -2023-02-13 17:46:20,284 - Epoch: [63][ 320/ 1207] Overall Loss 0.334251 Objective Loss 0.334251 LR 0.001000 Time 0.021709 -2023-02-13 17:46:20,475 - Epoch: [63][ 330/ 1207] Overall Loss 0.333296 Objective Loss 0.333296 LR 0.001000 Time 0.021629 -2023-02-13 17:46:20,665 - Epoch: [63][ 340/ 1207] Overall Loss 0.333980 Objective Loss 0.333980 LR 0.001000 Time 0.021552 -2023-02-13 17:46:20,858 - Epoch: [63][ 350/ 1207] Overall Loss 0.334023 Objective Loss 0.334023 LR 0.001000 Time 0.021486 -2023-02-13 17:46:21,049 - Epoch: [63][ 360/ 1207] Overall Loss 0.334457 Objective Loss 0.334457 LR 0.001000 Time 0.021418 -2023-02-13 17:46:21,239 - Epoch: [63][ 370/ 1207] Overall Loss 0.334877 Objective Loss 0.334877 LR 0.001000 Time 0.021351 -2023-02-13 17:46:21,429 - Epoch: [63][ 380/ 1207] Overall Loss 0.334919 Objective Loss 0.334919 LR 0.001000 Time 0.021289 -2023-02-13 17:46:21,619 - Epoch: [63][ 390/ 1207] Overall Loss 0.335071 Objective Loss 0.335071 LR 0.001000 Time 0.021230 -2023-02-13 17:46:21,811 - Epoch: [63][ 400/ 1207] Overall Loss 0.334864 Objective Loss 0.334864 LR 0.001000 Time 0.021177 -2023-02-13 17:46:22,002 - Epoch: [63][ 410/ 1207] Overall Loss 0.335332 Objective Loss 0.335332 LR 0.001000 Time 0.021127 -2023-02-13 17:46:22,193 - Epoch: [63][ 420/ 1207] Overall Loss 0.334590 Objective Loss 0.334590 LR 0.001000 Time 0.021078 -2023-02-13 17:46:22,384 - Epoch: [63][ 430/ 1207] Overall Loss 0.334618 Objective Loss 0.334618 LR 0.001000 Time 0.021031 -2023-02-13 17:46:22,575 - Epoch: [63][ 440/ 1207] Overall Loss 0.333698 Objective Loss 0.333698 LR 0.001000 Time 0.020986 -2023-02-13 17:46:22,767 - Epoch: [63][ 450/ 1207] Overall Loss 0.333675 Objective Loss 0.333675 LR 0.001000 Time 0.020945 -2023-02-13 17:46:22,958 - Epoch: [63][ 460/ 1207] Overall Loss 0.333421 Objective Loss 0.333421 LR 0.001000 Time 0.020904 -2023-02-13 17:46:23,149 - Epoch: [63][ 470/ 1207] Overall Loss 0.333577 Objective Loss 0.333577 LR 0.001000 Time 0.020865 -2023-02-13 17:46:23,340 - Epoch: [63][ 480/ 1207] Overall Loss 0.333862 Objective Loss 0.333862 LR 0.001000 Time 0.020827 -2023-02-13 17:46:23,531 - Epoch: [63][ 490/ 1207] Overall Loss 0.334038 Objective Loss 0.334038 LR 0.001000 Time 0.020791 -2023-02-13 17:46:23,722 - Epoch: [63][ 500/ 1207] Overall Loss 0.334698 Objective Loss 0.334698 LR 0.001000 Time 0.020757 -2023-02-13 17:46:23,914 - Epoch: [63][ 510/ 1207] Overall Loss 0.334868 Objective Loss 0.334868 LR 0.001000 Time 0.020725 -2023-02-13 17:46:24,107 - Epoch: [63][ 520/ 1207] Overall Loss 0.335280 Objective Loss 0.335280 LR 0.001000 Time 0.020697 -2023-02-13 17:46:24,298 - Epoch: [63][ 530/ 1207] Overall Loss 0.335205 Objective Loss 0.335205 LR 0.001000 Time 0.020666 -2023-02-13 17:46:24,488 - Epoch: [63][ 540/ 1207] Overall Loss 0.336117 Objective Loss 0.336117 LR 0.001000 Time 0.020636 -2023-02-13 17:46:24,679 - Epoch: [63][ 550/ 1207] Overall Loss 0.336650 Objective Loss 0.336650 LR 0.001000 Time 0.020607 -2023-02-13 17:46:24,871 - Epoch: [63][ 560/ 1207] Overall Loss 0.336309 Objective Loss 0.336309 LR 0.001000 Time 0.020581 -2023-02-13 17:46:25,062 - Epoch: [63][ 570/ 1207] Overall Loss 0.336621 Objective Loss 0.336621 LR 0.001000 Time 0.020555 -2023-02-13 17:46:25,253 - Epoch: [63][ 580/ 1207] Overall Loss 0.336006 Objective Loss 0.336006 LR 0.001000 Time 0.020529 -2023-02-13 17:46:25,444 - Epoch: [63][ 590/ 1207] Overall Loss 0.336706 Objective Loss 0.336706 LR 0.001000 Time 0.020504 -2023-02-13 17:46:25,636 - Epoch: [63][ 600/ 1207] Overall Loss 0.336445 Objective Loss 0.336445 LR 0.001000 Time 0.020480 -2023-02-13 17:46:25,828 - Epoch: [63][ 610/ 1207] Overall Loss 0.336792 Objective Loss 0.336792 LR 0.001000 Time 0.020460 -2023-02-13 17:46:26,019 - Epoch: [63][ 620/ 1207] Overall Loss 0.336991 Objective Loss 0.336991 LR 0.001000 Time 0.020437 -2023-02-13 17:46:26,210 - Epoch: [63][ 630/ 1207] Overall Loss 0.337233 Objective Loss 0.337233 LR 0.001000 Time 0.020414 -2023-02-13 17:46:26,401 - Epoch: [63][ 640/ 1207] Overall Loss 0.337354 Objective Loss 0.337354 LR 0.001000 Time 0.020394 -2023-02-13 17:46:26,593 - Epoch: [63][ 650/ 1207] Overall Loss 0.337473 Objective Loss 0.337473 LR 0.001000 Time 0.020375 -2023-02-13 17:46:26,785 - Epoch: [63][ 660/ 1207] Overall Loss 0.337543 Objective Loss 0.337543 LR 0.001000 Time 0.020357 -2023-02-13 17:46:26,977 - Epoch: [63][ 670/ 1207] Overall Loss 0.337887 Objective Loss 0.337887 LR 0.001000 Time 0.020339 -2023-02-13 17:46:27,168 - Epoch: [63][ 680/ 1207] Overall Loss 0.338018 Objective Loss 0.338018 LR 0.001000 Time 0.020320 -2023-02-13 17:46:27,360 - Epoch: [63][ 690/ 1207] Overall Loss 0.338037 Objective Loss 0.338037 LR 0.001000 Time 0.020303 -2023-02-13 17:46:27,551 - Epoch: [63][ 700/ 1207] Overall Loss 0.338316 Objective Loss 0.338316 LR 0.001000 Time 0.020286 -2023-02-13 17:46:27,743 - Epoch: [63][ 710/ 1207] Overall Loss 0.338191 Objective Loss 0.338191 LR 0.001000 Time 0.020269 -2023-02-13 17:46:27,933 - Epoch: [63][ 720/ 1207] Overall Loss 0.338808 Objective Loss 0.338808 LR 0.001000 Time 0.020252 -2023-02-13 17:46:28,124 - Epoch: [63][ 730/ 1207] Overall Loss 0.339106 Objective Loss 0.339106 LR 0.001000 Time 0.020236 -2023-02-13 17:46:28,315 - Epoch: [63][ 740/ 1207] Overall Loss 0.339258 Objective Loss 0.339258 LR 0.001000 Time 0.020219 -2023-02-13 17:46:28,506 - Epoch: [63][ 750/ 1207] Overall Loss 0.338982 Objective Loss 0.338982 LR 0.001000 Time 0.020204 -2023-02-13 17:46:28,697 - Epoch: [63][ 760/ 1207] Overall Loss 0.339146 Objective Loss 0.339146 LR 0.001000 Time 0.020189 -2023-02-13 17:46:28,890 - Epoch: [63][ 770/ 1207] Overall Loss 0.338805 Objective Loss 0.338805 LR 0.001000 Time 0.020176 -2023-02-13 17:46:29,081 - Epoch: [63][ 780/ 1207] Overall Loss 0.338896 Objective Loss 0.338896 LR 0.001000 Time 0.020162 -2023-02-13 17:46:29,272 - Epoch: [63][ 790/ 1207] Overall Loss 0.339257 Objective Loss 0.339257 LR 0.001000 Time 0.020149 -2023-02-13 17:46:29,463 - Epoch: [63][ 800/ 1207] Overall Loss 0.339076 Objective Loss 0.339076 LR 0.001000 Time 0.020135 -2023-02-13 17:46:29,655 - Epoch: [63][ 810/ 1207] Overall Loss 0.339273 Objective Loss 0.339273 LR 0.001000 Time 0.020123 -2023-02-13 17:46:29,846 - Epoch: [63][ 820/ 1207] Overall Loss 0.340091 Objective Loss 0.340091 LR 0.001000 Time 0.020111 -2023-02-13 17:46:30,038 - Epoch: [63][ 830/ 1207] Overall Loss 0.340315 Objective Loss 0.340315 LR 0.001000 Time 0.020099 -2023-02-13 17:46:30,230 - Epoch: [63][ 840/ 1207] Overall Loss 0.339919 Objective Loss 0.339919 LR 0.001000 Time 0.020087 -2023-02-13 17:46:30,421 - Epoch: [63][ 850/ 1207] Overall Loss 0.339675 Objective Loss 0.339675 LR 0.001000 Time 0.020076 -2023-02-13 17:46:30,612 - Epoch: [63][ 860/ 1207] Overall Loss 0.339510 Objective Loss 0.339510 LR 0.001000 Time 0.020064 -2023-02-13 17:46:30,805 - Epoch: [63][ 870/ 1207] Overall Loss 0.339282 Objective Loss 0.339282 LR 0.001000 Time 0.020054 -2023-02-13 17:46:30,996 - Epoch: [63][ 880/ 1207] Overall Loss 0.339233 Objective Loss 0.339233 LR 0.001000 Time 0.020043 -2023-02-13 17:46:31,187 - Epoch: [63][ 890/ 1207] Overall Loss 0.339165 Objective Loss 0.339165 LR 0.001000 Time 0.020032 -2023-02-13 17:46:31,379 - Epoch: [63][ 900/ 1207] Overall Loss 0.339276 Objective Loss 0.339276 LR 0.001000 Time 0.020022 -2023-02-13 17:46:31,570 - Epoch: [63][ 910/ 1207] Overall Loss 0.339222 Objective Loss 0.339222 LR 0.001000 Time 0.020012 -2023-02-13 17:46:31,762 - Epoch: [63][ 920/ 1207] Overall Loss 0.339002 Objective Loss 0.339002 LR 0.001000 Time 0.020003 -2023-02-13 17:46:31,954 - Epoch: [63][ 930/ 1207] Overall Loss 0.339087 Objective Loss 0.339087 LR 0.001000 Time 0.019994 -2023-02-13 17:46:32,145 - Epoch: [63][ 940/ 1207] Overall Loss 0.339063 Objective Loss 0.339063 LR 0.001000 Time 0.019984 -2023-02-13 17:46:32,337 - Epoch: [63][ 950/ 1207] Overall Loss 0.338877 Objective Loss 0.338877 LR 0.001000 Time 0.019975 -2023-02-13 17:46:32,529 - Epoch: [63][ 960/ 1207] Overall Loss 0.339207 Objective Loss 0.339207 LR 0.001000 Time 0.019966 -2023-02-13 17:46:32,720 - Epoch: [63][ 970/ 1207] Overall Loss 0.339582 Objective Loss 0.339582 LR 0.001000 Time 0.019957 -2023-02-13 17:46:32,911 - Epoch: [63][ 980/ 1207] Overall Loss 0.339403 Objective Loss 0.339403 LR 0.001000 Time 0.019948 -2023-02-13 17:46:33,103 - Epoch: [63][ 990/ 1207] Overall Loss 0.339504 Objective Loss 0.339504 LR 0.001000 Time 0.019940 -2023-02-13 17:46:33,293 - Epoch: [63][ 1000/ 1207] Overall Loss 0.339797 Objective Loss 0.339797 LR 0.001000 Time 0.019931 -2023-02-13 17:46:33,485 - Epoch: [63][ 1010/ 1207] Overall Loss 0.339881 Objective Loss 0.339881 LR 0.001000 Time 0.019923 -2023-02-13 17:46:33,677 - Epoch: [63][ 1020/ 1207] Overall Loss 0.340260 Objective Loss 0.340260 LR 0.001000 Time 0.019916 -2023-02-13 17:46:33,870 - Epoch: [63][ 1030/ 1207] Overall Loss 0.340757 Objective Loss 0.340757 LR 0.001000 Time 0.019909 -2023-02-13 17:46:34,061 - Epoch: [63][ 1040/ 1207] Overall Loss 0.340592 Objective Loss 0.340592 LR 0.001000 Time 0.019901 -2023-02-13 17:46:34,253 - Epoch: [63][ 1050/ 1207] Overall Loss 0.340602 Objective Loss 0.340602 LR 0.001000 Time 0.019894 -2023-02-13 17:46:34,444 - Epoch: [63][ 1060/ 1207] Overall Loss 0.340629 Objective Loss 0.340629 LR 0.001000 Time 0.019886 -2023-02-13 17:46:34,636 - Epoch: [63][ 1070/ 1207] Overall Loss 0.340501 Objective Loss 0.340501 LR 0.001000 Time 0.019879 -2023-02-13 17:46:34,828 - Epoch: [63][ 1080/ 1207] Overall Loss 0.340956 Objective Loss 0.340956 LR 0.001000 Time 0.019872 -2023-02-13 17:46:35,019 - Epoch: [63][ 1090/ 1207] Overall Loss 0.340756 Objective Loss 0.340756 LR 0.001000 Time 0.019865 -2023-02-13 17:46:35,211 - Epoch: [63][ 1100/ 1207] Overall Loss 0.341022 Objective Loss 0.341022 LR 0.001000 Time 0.019859 -2023-02-13 17:46:35,403 - Epoch: [63][ 1110/ 1207] Overall Loss 0.340980 Objective Loss 0.340980 LR 0.001000 Time 0.019853 -2023-02-13 17:46:35,596 - Epoch: [63][ 1120/ 1207] Overall Loss 0.340721 Objective Loss 0.340721 LR 0.001000 Time 0.019847 -2023-02-13 17:46:35,787 - Epoch: [63][ 1130/ 1207] Overall Loss 0.341150 Objective Loss 0.341150 LR 0.001000 Time 0.019840 -2023-02-13 17:46:35,979 - Epoch: [63][ 1140/ 1207] Overall Loss 0.341099 Objective Loss 0.341099 LR 0.001000 Time 0.019835 -2023-02-13 17:46:36,171 - Epoch: [63][ 1150/ 1207] Overall Loss 0.341262 Objective Loss 0.341262 LR 0.001000 Time 0.019829 -2023-02-13 17:46:36,362 - Epoch: [63][ 1160/ 1207] Overall Loss 0.340927 Objective Loss 0.340927 LR 0.001000 Time 0.019822 -2023-02-13 17:46:36,552 - Epoch: [63][ 1170/ 1207] Overall Loss 0.340818 Objective Loss 0.340818 LR 0.001000 Time 0.019815 -2023-02-13 17:46:36,741 - Epoch: [63][ 1180/ 1207] Overall Loss 0.340603 Objective Loss 0.340603 LR 0.001000 Time 0.019807 -2023-02-13 17:46:36,930 - Epoch: [63][ 1190/ 1207] Overall Loss 0.340472 Objective Loss 0.340472 LR 0.001000 Time 0.019799 -2023-02-13 17:46:37,175 - Epoch: [63][ 1200/ 1207] Overall Loss 0.340380 Objective Loss 0.340380 LR 0.001000 Time 0.019838 -2023-02-13 17:46:37,291 - Epoch: [63][ 1207/ 1207] Overall Loss 0.340336 Objective Loss 0.340336 Top1 84.146341 Top5 98.170732 LR 0.001000 Time 0.019819 -2023-02-13 17:46:37,374 - --- validate (epoch=63)----------- -2023-02-13 17:46:37,374 - 34311 samples (256 per mini-batch) -2023-02-13 17:46:37,784 - Epoch: [63][ 10/ 135] Loss 0.379071 Top1 81.328125 Top5 97.187500 -2023-02-13 17:46:37,916 - Epoch: [63][ 20/ 135] Loss 0.386864 Top1 80.898438 Top5 97.109375 -2023-02-13 17:46:38,046 - Epoch: [63][ 30/ 135] Loss 0.384825 Top1 80.768229 Top5 97.031250 -2023-02-13 17:46:38,176 - Epoch: [63][ 40/ 135] Loss 0.375107 Top1 80.927734 Top5 97.050781 -2023-02-13 17:46:38,305 - Epoch: [63][ 50/ 135] Loss 0.379029 Top1 80.992188 Top5 96.960938 -2023-02-13 17:46:38,436 - Epoch: [63][ 60/ 135] Loss 0.380108 Top1 81.002604 Top5 96.959635 -2023-02-13 17:46:38,565 - Epoch: [63][ 70/ 135] Loss 0.377255 Top1 81.032366 Top5 96.997768 -2023-02-13 17:46:38,696 - Epoch: [63][ 80/ 135] Loss 0.378878 Top1 80.942383 Top5 96.967773 -2023-02-13 17:46:38,828 - Epoch: [63][ 90/ 135] Loss 0.374581 Top1 81.163194 Top5 96.992188 -2023-02-13 17:46:38,960 - Epoch: [63][ 100/ 135] Loss 0.370138 Top1 81.187500 Top5 96.925781 -2023-02-13 17:46:39,091 - Epoch: [63][ 110/ 135] Loss 0.369835 Top1 81.242898 Top5 96.938920 -2023-02-13 17:46:39,221 - Epoch: [63][ 120/ 135] Loss 0.370751 Top1 81.184896 Top5 96.910807 -2023-02-13 17:46:39,356 - Epoch: [63][ 130/ 135] Loss 0.372012 Top1 81.147837 Top5 96.914062 -2023-02-13 17:46:39,404 - Epoch: [63][ 135/ 135] Loss 0.374688 Top1 81.090612 Top5 96.898954 -2023-02-13 17:46:39,488 - ==> Top1: 81.091 Top5: 96.899 Loss: 0.375 - -2023-02-13 17:46:39,488 - ==> Confusion: -[[ 820 5 9 1 22 8 0 0 7 59 0 3 1 3 5 5 7 3 2 0 7] - [ 2 926 0 4 15 30 5 14 4 1 0 5 2 1 3 0 6 2 7 1 5] - [ 7 2 937 15 4 2 15 15 0 1 3 3 5 3 2 10 5 3 16 2 8] - [ 5 2 21 884 1 7 0 1 2 2 16 0 9 2 14 5 5 5 25 0 10] - [ 10 8 0 1 988 8 2 1 0 3 0 6 2 4 5 8 8 3 0 4 5] - [ 3 19 0 8 11 960 3 14 2 3 3 17 2 7 0 2 9 0 2 4 1] - [ 2 5 23 5 2 9 1012 7 0 1 4 2 3 2 0 7 4 2 2 5 2] - [ 1 14 8 5 2 39 3 891 0 2 5 8 2 2 0 0 3 0 27 6 6] - [ 16 1 0 1 2 1 0 0 880 36 11 3 2 22 18 3 2 1 7 1 2] - [ 82 2 1 0 15 3 1 1 36 811 2 4 0 32 7 2 2 2 2 2 5] - [ 2 2 3 4 1 3 3 6 15 0 965 2 0 15 2 1 5 1 17 1 3] - [ 1 2 1 0 3 14 2 6 0 1 0 909 27 4 1 9 5 12 1 6 1] - [ 2 0 2 8 2 3 0 4 3 0 1 45 840 0 3 11 1 24 4 0 6] - [ 5 4 1 0 9 12 1 5 7 12 5 10 3 926 4 4 5 4 3 1 3] - [ 8 3 3 19 11 3 0 2 27 4 7 1 6 3 947 4 11 11 13 0 9] - [ 3 2 2 0 4 3 4 0 0 0 0 11 4 2 0 981 7 13 1 5 4] - [ 4 3 0 1 10 3 0 0 3 0 1 2 0 1 3 16 1006 2 3 0 3] - [ 6 1 0 5 1 2 1 1 0 1 0 16 19 1 0 13 3 977 1 1 2] - [ 2 5 6 8 5 1 0 25 6 0 4 6 3 0 11 0 2 0 999 1 2] - [ 0 5 1 0 3 12 6 24 0 0 0 30 3 3 0 9 8 2 4 1029 9] - [ 138 290 225 142 234 285 83 193 106 72 216 185 338 365 166 182 449 155 226 249 9135]] - -2023-02-13 17:46:39,490 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:46:39,490 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:46:39,496 - - -2023-02-13 17:46:39,496 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:46:40,479 - Epoch: [64][ 10/ 1207] Overall Loss 0.352207 Objective Loss 0.352207 LR 0.001000 Time 0.098308 -2023-02-13 17:46:40,676 - Epoch: [64][ 20/ 1207] Overall Loss 0.333247 Objective Loss 0.333247 LR 0.001000 Time 0.058945 -2023-02-13 17:46:40,866 - Epoch: [64][ 30/ 1207] Overall Loss 0.335974 Objective Loss 0.335974 LR 0.001000 Time 0.045628 -2023-02-13 17:46:41,054 - Epoch: [64][ 40/ 1207] Overall Loss 0.331032 Objective Loss 0.331032 LR 0.001000 Time 0.038916 -2023-02-13 17:46:41,242 - Epoch: [64][ 50/ 1207] Overall Loss 0.339936 Objective Loss 0.339936 LR 0.001000 Time 0.034881 -2023-02-13 17:46:41,430 - Epoch: [64][ 60/ 1207] Overall Loss 0.335475 Objective Loss 0.335475 LR 0.001000 Time 0.032193 -2023-02-13 17:46:41,618 - Epoch: [64][ 70/ 1207] Overall Loss 0.341819 Objective Loss 0.341819 LR 0.001000 Time 0.030278 -2023-02-13 17:46:41,807 - Epoch: [64][ 80/ 1207] Overall Loss 0.341907 Objective Loss 0.341907 LR 0.001000 Time 0.028845 -2023-02-13 17:46:41,995 - Epoch: [64][ 90/ 1207] Overall Loss 0.342632 Objective Loss 0.342632 LR 0.001000 Time 0.027734 -2023-02-13 17:46:42,183 - Epoch: [64][ 100/ 1207] Overall Loss 0.338150 Objective Loss 0.338150 LR 0.001000 Time 0.026835 -2023-02-13 17:46:42,372 - Epoch: [64][ 110/ 1207] Overall Loss 0.336441 Objective Loss 0.336441 LR 0.001000 Time 0.026105 -2023-02-13 17:46:42,560 - Epoch: [64][ 120/ 1207] Overall Loss 0.336263 Objective Loss 0.336263 LR 0.001000 Time 0.025498 -2023-02-13 17:46:42,748 - Epoch: [64][ 130/ 1207] Overall Loss 0.337839 Objective Loss 0.337839 LR 0.001000 Time 0.024978 -2023-02-13 17:46:42,938 - Epoch: [64][ 140/ 1207] Overall Loss 0.336556 Objective Loss 0.336556 LR 0.001000 Time 0.024547 -2023-02-13 17:46:43,126 - Epoch: [64][ 150/ 1207] Overall Loss 0.336169 Objective Loss 0.336169 LR 0.001000 Time 0.024166 -2023-02-13 17:46:43,315 - Epoch: [64][ 160/ 1207] Overall Loss 0.335906 Objective Loss 0.335906 LR 0.001000 Time 0.023833 -2023-02-13 17:46:43,508 - Epoch: [64][ 170/ 1207] Overall Loss 0.335539 Objective Loss 0.335539 LR 0.001000 Time 0.023566 -2023-02-13 17:46:43,702 - Epoch: [64][ 180/ 1207] Overall Loss 0.335297 Objective Loss 0.335297 LR 0.001000 Time 0.023332 -2023-02-13 17:46:43,899 - Epoch: [64][ 190/ 1207] Overall Loss 0.335173 Objective Loss 0.335173 LR 0.001000 Time 0.023139 -2023-02-13 17:46:44,093 - Epoch: [64][ 200/ 1207] Overall Loss 0.335276 Objective Loss 0.335276 LR 0.001000 Time 0.022945 -2023-02-13 17:46:44,287 - Epoch: [64][ 210/ 1207] Overall Loss 0.334079 Objective Loss 0.334079 LR 0.001000 Time 0.022779 -2023-02-13 17:46:44,481 - Epoch: [64][ 220/ 1207] Overall Loss 0.334245 Objective Loss 0.334245 LR 0.001000 Time 0.022622 -2023-02-13 17:46:44,677 - Epoch: [64][ 230/ 1207] Overall Loss 0.334271 Objective Loss 0.334271 LR 0.001000 Time 0.022487 -2023-02-13 17:46:44,871 - Epoch: [64][ 240/ 1207] Overall Loss 0.335524 Objective Loss 0.335524 LR 0.001000 Time 0.022359 -2023-02-13 17:46:45,067 - Epoch: [64][ 250/ 1207] Overall Loss 0.335410 Objective Loss 0.335410 LR 0.001000 Time 0.022246 -2023-02-13 17:46:45,261 - Epoch: [64][ 260/ 1207] Overall Loss 0.335565 Objective Loss 0.335565 LR 0.001000 Time 0.022136 -2023-02-13 17:46:45,457 - Epoch: [64][ 270/ 1207] Overall Loss 0.335464 Objective Loss 0.335464 LR 0.001000 Time 0.022039 -2023-02-13 17:46:45,651 - Epoch: [64][ 280/ 1207] Overall Loss 0.335062 Objective Loss 0.335062 LR 0.001000 Time 0.021945 -2023-02-13 17:46:45,848 - Epoch: [64][ 290/ 1207] Overall Loss 0.335848 Objective Loss 0.335848 LR 0.001000 Time 0.021866 -2023-02-13 17:46:46,041 - Epoch: [64][ 300/ 1207] Overall Loss 0.335782 Objective Loss 0.335782 LR 0.001000 Time 0.021781 -2023-02-13 17:46:46,237 - Epoch: [64][ 310/ 1207] Overall Loss 0.335459 Objective Loss 0.335459 LR 0.001000 Time 0.021708 -2023-02-13 17:46:46,431 - Epoch: [64][ 320/ 1207] Overall Loss 0.335795 Objective Loss 0.335795 LR 0.001000 Time 0.021634 -2023-02-13 17:46:46,627 - Epoch: [64][ 330/ 1207] Overall Loss 0.335448 Objective Loss 0.335448 LR 0.001000 Time 0.021571 -2023-02-13 17:46:46,821 - Epoch: [64][ 340/ 1207] Overall Loss 0.335849 Objective Loss 0.335849 LR 0.001000 Time 0.021508 -2023-02-13 17:46:47,018 - Epoch: [64][ 350/ 1207] Overall Loss 0.336794 Objective Loss 0.336794 LR 0.001000 Time 0.021453 -2023-02-13 17:46:47,211 - Epoch: [64][ 360/ 1207] Overall Loss 0.336350 Objective Loss 0.336350 LR 0.001000 Time 0.021394 -2023-02-13 17:46:47,407 - Epoch: [64][ 370/ 1207] Overall Loss 0.336868 Objective Loss 0.336868 LR 0.001000 Time 0.021344 -2023-02-13 17:46:47,601 - Epoch: [64][ 380/ 1207] Overall Loss 0.336888 Objective Loss 0.336888 LR 0.001000 Time 0.021292 -2023-02-13 17:46:47,797 - Epoch: [64][ 390/ 1207] Overall Loss 0.337098 Objective Loss 0.337098 LR 0.001000 Time 0.021248 -2023-02-13 17:46:47,991 - Epoch: [64][ 400/ 1207] Overall Loss 0.335731 Objective Loss 0.335731 LR 0.001000 Time 0.021200 -2023-02-13 17:46:48,187 - Epoch: [64][ 410/ 1207] Overall Loss 0.335386 Objective Loss 0.335386 LR 0.001000 Time 0.021161 -2023-02-13 17:46:48,381 - Epoch: [64][ 420/ 1207] Overall Loss 0.335279 Objective Loss 0.335279 LR 0.001000 Time 0.021118 -2023-02-13 17:46:48,578 - Epoch: [64][ 430/ 1207] Overall Loss 0.334450 Objective Loss 0.334450 LR 0.001000 Time 0.021083 -2023-02-13 17:46:48,772 - Epoch: [64][ 440/ 1207] Overall Loss 0.334274 Objective Loss 0.334274 LR 0.001000 Time 0.021044 -2023-02-13 17:46:48,968 - Epoch: [64][ 450/ 1207] Overall Loss 0.334541 Objective Loss 0.334541 LR 0.001000 Time 0.021011 -2023-02-13 17:46:49,162 - Epoch: [64][ 460/ 1207] Overall Loss 0.333668 Objective Loss 0.333668 LR 0.001000 Time 0.020975 -2023-02-13 17:46:49,358 - Epoch: [64][ 470/ 1207] Overall Loss 0.333378 Objective Loss 0.333378 LR 0.001000 Time 0.020945 -2023-02-13 17:46:49,552 - Epoch: [64][ 480/ 1207] Overall Loss 0.333935 Objective Loss 0.333935 LR 0.001000 Time 0.020913 -2023-02-13 17:46:49,749 - Epoch: [64][ 490/ 1207] Overall Loss 0.333605 Objective Loss 0.333605 LR 0.001000 Time 0.020886 -2023-02-13 17:46:49,943 - Epoch: [64][ 500/ 1207] Overall Loss 0.333970 Objective Loss 0.333970 LR 0.001000 Time 0.020856 -2023-02-13 17:46:50,138 - Epoch: [64][ 510/ 1207] Overall Loss 0.334390 Objective Loss 0.334390 LR 0.001000 Time 0.020830 -2023-02-13 17:46:50,332 - Epoch: [64][ 520/ 1207] Overall Loss 0.334483 Objective Loss 0.334483 LR 0.001000 Time 0.020801 -2023-02-13 17:46:50,527 - Epoch: [64][ 530/ 1207] Overall Loss 0.334589 Objective Loss 0.334589 LR 0.001000 Time 0.020777 -2023-02-13 17:46:50,722 - Epoch: [64][ 540/ 1207] Overall Loss 0.335218 Objective Loss 0.335218 LR 0.001000 Time 0.020751 -2023-02-13 17:46:50,919 - Epoch: [64][ 550/ 1207] Overall Loss 0.335329 Objective Loss 0.335329 LR 0.001000 Time 0.020731 -2023-02-13 17:46:51,114 - Epoch: [64][ 560/ 1207] Overall Loss 0.334906 Objective Loss 0.334906 LR 0.001000 Time 0.020708 -2023-02-13 17:46:51,310 - Epoch: [64][ 570/ 1207] Overall Loss 0.334597 Objective Loss 0.334597 LR 0.001000 Time 0.020689 -2023-02-13 17:46:51,504 - Epoch: [64][ 580/ 1207] Overall Loss 0.335075 Objective Loss 0.335075 LR 0.001000 Time 0.020665 -2023-02-13 17:46:51,700 - Epoch: [64][ 590/ 1207] Overall Loss 0.335225 Objective Loss 0.335225 LR 0.001000 Time 0.020647 -2023-02-13 17:46:51,895 - Epoch: [64][ 600/ 1207] Overall Loss 0.335213 Objective Loss 0.335213 LR 0.001000 Time 0.020628 -2023-02-13 17:46:52,092 - Epoch: [64][ 610/ 1207] Overall Loss 0.334990 Objective Loss 0.334990 LR 0.001000 Time 0.020611 -2023-02-13 17:46:52,285 - Epoch: [64][ 620/ 1207] Overall Loss 0.334736 Objective Loss 0.334736 LR 0.001000 Time 0.020590 -2023-02-13 17:46:52,482 - Epoch: [64][ 630/ 1207] Overall Loss 0.335239 Objective Loss 0.335239 LR 0.001000 Time 0.020575 -2023-02-13 17:46:52,676 - Epoch: [64][ 640/ 1207] Overall Loss 0.335103 Objective Loss 0.335103 LR 0.001000 Time 0.020556 -2023-02-13 17:46:52,873 - Epoch: [64][ 650/ 1207] Overall Loss 0.334953 Objective Loss 0.334953 LR 0.001000 Time 0.020542 -2023-02-13 17:46:53,067 - Epoch: [64][ 660/ 1207] Overall Loss 0.335052 Objective Loss 0.335052 LR 0.001000 Time 0.020525 -2023-02-13 17:46:53,263 - Epoch: [64][ 670/ 1207] Overall Loss 0.334917 Objective Loss 0.334917 LR 0.001000 Time 0.020510 -2023-02-13 17:46:53,457 - Epoch: [64][ 680/ 1207] Overall Loss 0.335155 Objective Loss 0.335155 LR 0.001000 Time 0.020493 -2023-02-13 17:46:53,653 - Epoch: [64][ 690/ 1207] Overall Loss 0.335715 Objective Loss 0.335715 LR 0.001000 Time 0.020480 -2023-02-13 17:46:53,848 - Epoch: [64][ 700/ 1207] Overall Loss 0.336344 Objective Loss 0.336344 LR 0.001000 Time 0.020465 -2023-02-13 17:46:54,044 - Epoch: [64][ 710/ 1207] Overall Loss 0.336167 Objective Loss 0.336167 LR 0.001000 Time 0.020453 -2023-02-13 17:46:54,238 - Epoch: [64][ 720/ 1207] Overall Loss 0.336504 Objective Loss 0.336504 LR 0.001000 Time 0.020438 -2023-02-13 17:46:54,434 - Epoch: [64][ 730/ 1207] Overall Loss 0.336401 Objective Loss 0.336401 LR 0.001000 Time 0.020426 -2023-02-13 17:46:54,628 - Epoch: [64][ 740/ 1207] Overall Loss 0.336882 Objective Loss 0.336882 LR 0.001000 Time 0.020411 -2023-02-13 17:46:54,824 - Epoch: [64][ 750/ 1207] Overall Loss 0.337575 Objective Loss 0.337575 LR 0.001000 Time 0.020400 -2023-02-13 17:46:55,018 - Epoch: [64][ 760/ 1207] Overall Loss 0.337844 Objective Loss 0.337844 LR 0.001000 Time 0.020385 -2023-02-13 17:46:55,213 - Epoch: [64][ 770/ 1207] Overall Loss 0.337692 Objective Loss 0.337692 LR 0.001000 Time 0.020375 -2023-02-13 17:46:55,406 - Epoch: [64][ 780/ 1207] Overall Loss 0.337566 Objective Loss 0.337566 LR 0.001000 Time 0.020360 -2023-02-13 17:46:55,602 - Epoch: [64][ 790/ 1207] Overall Loss 0.337632 Objective Loss 0.337632 LR 0.001000 Time 0.020349 -2023-02-13 17:46:55,796 - Epoch: [64][ 800/ 1207] Overall Loss 0.337492 Objective Loss 0.337492 LR 0.001000 Time 0.020337 -2023-02-13 17:46:55,992 - Epoch: [64][ 810/ 1207] Overall Loss 0.337223 Objective Loss 0.337223 LR 0.001000 Time 0.020328 -2023-02-13 17:46:56,185 - Epoch: [64][ 820/ 1207] Overall Loss 0.337192 Objective Loss 0.337192 LR 0.001000 Time 0.020315 -2023-02-13 17:46:56,380 - Epoch: [64][ 830/ 1207] Overall Loss 0.337132 Objective Loss 0.337132 LR 0.001000 Time 0.020305 -2023-02-13 17:46:56,573 - Epoch: [64][ 840/ 1207] Overall Loss 0.336830 Objective Loss 0.336830 LR 0.001000 Time 0.020293 -2023-02-13 17:46:56,769 - Epoch: [64][ 850/ 1207] Overall Loss 0.336551 Objective Loss 0.336551 LR 0.001000 Time 0.020284 -2023-02-13 17:46:56,963 - Epoch: [64][ 860/ 1207] Overall Loss 0.336500 Objective Loss 0.336500 LR 0.001000 Time 0.020273 -2023-02-13 17:46:57,160 - Epoch: [64][ 870/ 1207] Overall Loss 0.336616 Objective Loss 0.336616 LR 0.001000 Time 0.020266 -2023-02-13 17:46:57,354 - Epoch: [64][ 880/ 1207] Overall Loss 0.336496 Objective Loss 0.336496 LR 0.001000 Time 0.020255 -2023-02-13 17:46:57,549 - Epoch: [64][ 890/ 1207] Overall Loss 0.336478 Objective Loss 0.336478 LR 0.001000 Time 0.020247 -2023-02-13 17:46:57,743 - Epoch: [64][ 900/ 1207] Overall Loss 0.336421 Objective Loss 0.336421 LR 0.001000 Time 0.020237 -2023-02-13 17:46:57,940 - Epoch: [64][ 910/ 1207] Overall Loss 0.336262 Objective Loss 0.336262 LR 0.001000 Time 0.020231 -2023-02-13 17:46:58,134 - Epoch: [64][ 920/ 1207] Overall Loss 0.336310 Objective Loss 0.336310 LR 0.001000 Time 0.020221 -2023-02-13 17:46:58,329 - Epoch: [64][ 930/ 1207] Overall Loss 0.336227 Objective Loss 0.336227 LR 0.001000 Time 0.020213 -2023-02-13 17:46:58,522 - Epoch: [64][ 940/ 1207] Overall Loss 0.336507 Objective Loss 0.336507 LR 0.001000 Time 0.020203 -2023-02-13 17:46:58,718 - Epoch: [64][ 950/ 1207] Overall Loss 0.336361 Objective Loss 0.336361 LR 0.001000 Time 0.020196 -2023-02-13 17:46:58,912 - Epoch: [64][ 960/ 1207] Overall Loss 0.336526 Objective Loss 0.336526 LR 0.001000 Time 0.020188 -2023-02-13 17:46:59,109 - Epoch: [64][ 970/ 1207] Overall Loss 0.336680 Objective Loss 0.336680 LR 0.001000 Time 0.020182 -2023-02-13 17:46:59,302 - Epoch: [64][ 980/ 1207] Overall Loss 0.336634 Objective Loss 0.336634 LR 0.001000 Time 0.020173 -2023-02-13 17:46:59,497 - Epoch: [64][ 990/ 1207] Overall Loss 0.336549 Objective Loss 0.336549 LR 0.001000 Time 0.020166 -2023-02-13 17:46:59,691 - Epoch: [64][ 1000/ 1207] Overall Loss 0.336615 Objective Loss 0.336615 LR 0.001000 Time 0.020158 -2023-02-13 17:46:59,888 - Epoch: [64][ 1010/ 1207] Overall Loss 0.336532 Objective Loss 0.336532 LR 0.001000 Time 0.020153 -2023-02-13 17:47:00,083 - Epoch: [64][ 1020/ 1207] Overall Loss 0.336467 Objective Loss 0.336467 LR 0.001000 Time 0.020146 -2023-02-13 17:47:00,279 - Epoch: [64][ 1030/ 1207] Overall Loss 0.336405 Objective Loss 0.336405 LR 0.001000 Time 0.020140 -2023-02-13 17:47:00,472 - Epoch: [64][ 1040/ 1207] Overall Loss 0.336062 Objective Loss 0.336062 LR 0.001000 Time 0.020132 -2023-02-13 17:47:00,669 - Epoch: [64][ 1050/ 1207] Overall Loss 0.336030 Objective Loss 0.336030 LR 0.001000 Time 0.020127 -2023-02-13 17:47:00,865 - Epoch: [64][ 1060/ 1207] Overall Loss 0.335954 Objective Loss 0.335954 LR 0.001000 Time 0.020122 -2023-02-13 17:47:01,061 - Epoch: [64][ 1070/ 1207] Overall Loss 0.335713 Objective Loss 0.335713 LR 0.001000 Time 0.020117 -2023-02-13 17:47:01,255 - Epoch: [64][ 1080/ 1207] Overall Loss 0.335844 Objective Loss 0.335844 LR 0.001000 Time 0.020110 -2023-02-13 17:47:01,451 - Epoch: [64][ 1090/ 1207] Overall Loss 0.335802 Objective Loss 0.335802 LR 0.001000 Time 0.020105 -2023-02-13 17:47:01,645 - Epoch: [64][ 1100/ 1207] Overall Loss 0.335791 Objective Loss 0.335791 LR 0.001000 Time 0.020098 -2023-02-13 17:47:01,842 - Epoch: [64][ 1110/ 1207] Overall Loss 0.335538 Objective Loss 0.335538 LR 0.001000 Time 0.020094 -2023-02-13 17:47:02,036 - Epoch: [64][ 1120/ 1207] Overall Loss 0.335708 Objective Loss 0.335708 LR 0.001000 Time 0.020088 -2023-02-13 17:47:02,232 - Epoch: [64][ 1130/ 1207] Overall Loss 0.335610 Objective Loss 0.335610 LR 0.001000 Time 0.020083 -2023-02-13 17:47:02,426 - Epoch: [64][ 1140/ 1207] Overall Loss 0.335473 Objective Loss 0.335473 LR 0.001000 Time 0.020077 -2023-02-13 17:47:02,622 - Epoch: [64][ 1150/ 1207] Overall Loss 0.335552 Objective Loss 0.335552 LR 0.001000 Time 0.020072 -2023-02-13 17:47:02,816 - Epoch: [64][ 1160/ 1207] Overall Loss 0.335805 Objective Loss 0.335805 LR 0.001000 Time 0.020066 -2023-02-13 17:47:03,013 - Epoch: [64][ 1170/ 1207] Overall Loss 0.336158 Objective Loss 0.336158 LR 0.001000 Time 0.020063 -2023-02-13 17:47:03,208 - Epoch: [64][ 1180/ 1207] Overall Loss 0.335944 Objective Loss 0.335944 LR 0.001000 Time 0.020057 -2023-02-13 17:47:03,404 - Epoch: [64][ 1190/ 1207] Overall Loss 0.335927 Objective Loss 0.335927 LR 0.001000 Time 0.020053 -2023-02-13 17:47:03,650 - Epoch: [64][ 1200/ 1207] Overall Loss 0.335793 Objective Loss 0.335793 LR 0.001000 Time 0.020091 -2023-02-13 17:47:03,765 - Epoch: [64][ 1207/ 1207] Overall Loss 0.336085 Objective Loss 0.336085 Top1 81.097561 Top5 96.036585 LR 0.001000 Time 0.020069 -2023-02-13 17:47:03,837 - --- validate (epoch=64)----------- -2023-02-13 17:47:03,837 - 34311 samples (256 per mini-batch) -2023-02-13 17:47:04,241 - Epoch: [64][ 10/ 135] Loss 0.346475 Top1 82.421875 Top5 97.460938 -2023-02-13 17:47:04,383 - Epoch: [64][ 20/ 135] Loss 0.337152 Top1 82.070312 Top5 97.343750 -2023-02-13 17:47:04,517 - Epoch: [64][ 30/ 135] Loss 0.336712 Top1 82.265625 Top5 97.291667 -2023-02-13 17:47:04,645 - Epoch: [64][ 40/ 135] Loss 0.341204 Top1 82.011719 Top5 97.324219 -2023-02-13 17:47:04,776 - Epoch: [64][ 50/ 135] Loss 0.342769 Top1 82.164062 Top5 97.312500 -2023-02-13 17:47:04,912 - Epoch: [64][ 60/ 135] Loss 0.343773 Top1 82.089844 Top5 97.259115 -2023-02-13 17:47:05,050 - Epoch: [64][ 70/ 135] Loss 0.347538 Top1 82.003348 Top5 97.087054 -2023-02-13 17:47:05,177 - Epoch: [64][ 80/ 135] Loss 0.350490 Top1 81.953125 Top5 97.006836 -2023-02-13 17:47:05,304 - Epoch: [64][ 90/ 135] Loss 0.356240 Top1 81.822917 Top5 96.987847 -2023-02-13 17:47:05,430 - Epoch: [64][ 100/ 135] Loss 0.356014 Top1 81.917969 Top5 96.968750 -2023-02-13 17:47:05,557 - Epoch: [64][ 110/ 135] Loss 0.356114 Top1 81.914062 Top5 96.963778 -2023-02-13 17:47:05,683 - Epoch: [64][ 120/ 135] Loss 0.358879 Top1 81.832682 Top5 96.923828 -2023-02-13 17:47:05,812 - Epoch: [64][ 130/ 135] Loss 0.359543 Top1 81.841947 Top5 96.914062 -2023-02-13 17:47:05,858 - Epoch: [64][ 135/ 135] Loss 0.355618 Top1 81.877532 Top5 96.916441 -2023-02-13 17:47:05,926 - ==> Top1: 81.878 Top5: 96.916 Loss: 0.356 - -2023-02-13 17:47:05,927 - ==> Confusion: -[[ 849 6 6 2 11 2 1 3 3 54 0 5 1 5 3 1 4 3 1 1 6] - [ 3 933 2 4 3 36 1 18 4 3 4 1 2 0 3 0 5 0 4 0 7] - [ 14 3 952 13 4 1 14 18 0 0 4 0 1 4 4 4 4 5 6 4 3] - [ 7 0 22 900 2 1 0 3 3 3 17 0 6 2 21 1 3 5 17 0 3] - [ 22 13 1 1 971 17 1 5 1 4 2 4 1 5 4 2 2 2 0 1 7] - [ 5 19 3 5 7 948 1 24 5 2 1 10 5 17 2 1 3 2 2 6 2] - [ 5 4 19 4 2 7 1015 11 1 2 7 1 1 0 0 2 4 3 2 4 5] - [ 2 8 3 4 1 34 0 932 1 1 2 4 2 3 0 0 1 2 15 3 6] - [ 18 4 0 1 2 0 0 4 893 45 5 2 1 4 23 0 2 0 4 0 1] - [ 89 1 4 0 3 4 0 0 45 838 1 0 1 15 5 0 0 1 2 1 2] - [ 1 3 6 5 1 0 0 4 22 3 973 2 1 10 4 0 1 1 11 0 3] - [ 2 1 0 0 2 15 0 7 2 2 0 899 23 8 2 7 7 11 2 14 1] - [ 2 0 2 5 5 1 0 1 5 0 1 40 851 2 2 3 4 27 1 3 4] - [ 4 2 2 0 11 19 1 4 19 20 14 6 2 905 3 0 4 1 0 1 6] - [ 17 1 1 11 5 3 0 2 24 7 3 1 1 2 986 1 0 8 7 0 12] - [ 11 2 3 0 11 1 2 1 0 1 1 6 6 5 2 955 12 12 1 7 7] - [ 3 7 0 1 10 8 0 1 1 1 0 1 0 0 4 5 1007 1 1 1 9] - [ 5 2 0 5 1 1 2 2 1 1 1 7 18 5 3 7 0 986 1 0 3] - [ 3 6 5 13 2 3 2 34 5 0 8 2 3 0 21 1 1 1 972 1 3] - [ 2 3 2 2 1 6 6 33 0 0 2 15 5 9 0 4 7 1 0 1042 8] - [ 238 245 221 141 162 262 72 242 138 112 249 126 407 312 220 94 288 122 198 299 9286]] - -2023-02-13 17:47:05,928 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:47:05,928 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:47:05,934 - - -2023-02-13 17:47:05,934 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:47:06,833 - Epoch: [65][ 10/ 1207] Overall Loss 0.299052 Objective Loss 0.299052 LR 0.001000 Time 0.089812 -2023-02-13 17:47:07,033 - Epoch: [65][ 20/ 1207] Overall Loss 0.316891 Objective Loss 0.316891 LR 0.001000 Time 0.054913 -2023-02-13 17:47:07,227 - Epoch: [65][ 30/ 1207] Overall Loss 0.319100 Objective Loss 0.319100 LR 0.001000 Time 0.043032 -2023-02-13 17:47:07,422 - Epoch: [65][ 40/ 1207] Overall Loss 0.329070 Objective Loss 0.329070 LR 0.001000 Time 0.037160 -2023-02-13 17:47:07,616 - Epoch: [65][ 50/ 1207] Overall Loss 0.325213 Objective Loss 0.325213 LR 0.001000 Time 0.033598 -2023-02-13 17:47:07,812 - Epoch: [65][ 60/ 1207] Overall Loss 0.324928 Objective Loss 0.324928 LR 0.001000 Time 0.031263 -2023-02-13 17:47:08,006 - Epoch: [65][ 70/ 1207] Overall Loss 0.323879 Objective Loss 0.323879 LR 0.001000 Time 0.029556 -2023-02-13 17:47:08,202 - Epoch: [65][ 80/ 1207] Overall Loss 0.326148 Objective Loss 0.326148 LR 0.001000 Time 0.028306 -2023-02-13 17:47:08,395 - Epoch: [65][ 90/ 1207] Overall Loss 0.326942 Objective Loss 0.326942 LR 0.001000 Time 0.027306 -2023-02-13 17:47:08,592 - Epoch: [65][ 100/ 1207] Overall Loss 0.324398 Objective Loss 0.324398 LR 0.001000 Time 0.026535 -2023-02-13 17:47:08,785 - Epoch: [65][ 110/ 1207] Overall Loss 0.324769 Objective Loss 0.324769 LR 0.001000 Time 0.025876 -2023-02-13 17:47:08,983 - Epoch: [65][ 120/ 1207] Overall Loss 0.326061 Objective Loss 0.326061 LR 0.001000 Time 0.025364 -2023-02-13 17:47:09,177 - Epoch: [65][ 130/ 1207] Overall Loss 0.328532 Objective Loss 0.328532 LR 0.001000 Time 0.024908 -2023-02-13 17:47:09,375 - Epoch: [65][ 140/ 1207] Overall Loss 0.330516 Objective Loss 0.330516 LR 0.001000 Time 0.024538 -2023-02-13 17:47:09,570 - Epoch: [65][ 150/ 1207] Overall Loss 0.331371 Objective Loss 0.331371 LR 0.001000 Time 0.024199 -2023-02-13 17:47:09,761 - Epoch: [65][ 160/ 1207] Overall Loss 0.331627 Objective Loss 0.331627 LR 0.001000 Time 0.023877 -2023-02-13 17:47:09,953 - Epoch: [65][ 170/ 1207] Overall Loss 0.333128 Objective Loss 0.333128 LR 0.001000 Time 0.023600 -2023-02-13 17:47:10,143 - Epoch: [65][ 180/ 1207] Overall Loss 0.332525 Objective Loss 0.332525 LR 0.001000 Time 0.023343 -2023-02-13 17:47:10,335 - Epoch: [65][ 190/ 1207] Overall Loss 0.331826 Objective Loss 0.331826 LR 0.001000 Time 0.023122 -2023-02-13 17:47:10,525 - Epoch: [65][ 200/ 1207] Overall Loss 0.330294 Objective Loss 0.330294 LR 0.001000 Time 0.022916 -2023-02-13 17:47:10,716 - Epoch: [65][ 210/ 1207] Overall Loss 0.329767 Objective Loss 0.329767 LR 0.001000 Time 0.022734 -2023-02-13 17:47:10,910 - Epoch: [65][ 220/ 1207] Overall Loss 0.330494 Objective Loss 0.330494 LR 0.001000 Time 0.022580 -2023-02-13 17:47:11,102 - Epoch: [65][ 230/ 1207] Overall Loss 0.331111 Objective Loss 0.331111 LR 0.001000 Time 0.022431 -2023-02-13 17:47:11,293 - Epoch: [65][ 240/ 1207] Overall Loss 0.330421 Objective Loss 0.330421 LR 0.001000 Time 0.022292 -2023-02-13 17:47:11,482 - Epoch: [65][ 250/ 1207] Overall Loss 0.330014 Objective Loss 0.330014 LR 0.001000 Time 0.022152 -2023-02-13 17:47:11,671 - Epoch: [65][ 260/ 1207] Overall Loss 0.329331 Objective Loss 0.329331 LR 0.001000 Time 0.022026 -2023-02-13 17:47:11,860 - Epoch: [65][ 270/ 1207] Overall Loss 0.330898 Objective Loss 0.330898 LR 0.001000 Time 0.021911 -2023-02-13 17:47:12,051 - Epoch: [65][ 280/ 1207] Overall Loss 0.331458 Objective Loss 0.331458 LR 0.001000 Time 0.021807 -2023-02-13 17:47:12,240 - Epoch: [65][ 290/ 1207] Overall Loss 0.330191 Objective Loss 0.330191 LR 0.001000 Time 0.021706 -2023-02-13 17:47:12,429 - Epoch: [65][ 300/ 1207] Overall Loss 0.330661 Objective Loss 0.330661 LR 0.001000 Time 0.021611 -2023-02-13 17:47:12,618 - Epoch: [65][ 310/ 1207] Overall Loss 0.330552 Objective Loss 0.330552 LR 0.001000 Time 0.021521 -2023-02-13 17:47:12,807 - Epoch: [65][ 320/ 1207] Overall Loss 0.330499 Objective Loss 0.330499 LR 0.001000 Time 0.021439 -2023-02-13 17:47:12,996 - Epoch: [65][ 330/ 1207] Overall Loss 0.329572 Objective Loss 0.329572 LR 0.001000 Time 0.021362 -2023-02-13 17:47:13,185 - Epoch: [65][ 340/ 1207] Overall Loss 0.329674 Objective Loss 0.329674 LR 0.001000 Time 0.021289 -2023-02-13 17:47:13,373 - Epoch: [65][ 350/ 1207] Overall Loss 0.330005 Objective Loss 0.330005 LR 0.001000 Time 0.021217 -2023-02-13 17:47:13,562 - Epoch: [65][ 360/ 1207] Overall Loss 0.331294 Objective Loss 0.331294 LR 0.001000 Time 0.021151 -2023-02-13 17:47:13,751 - Epoch: [65][ 370/ 1207] Overall Loss 0.332148 Objective Loss 0.332148 LR 0.001000 Time 0.021088 -2023-02-13 17:47:13,940 - Epoch: [65][ 380/ 1207] Overall Loss 0.332321 Objective Loss 0.332321 LR 0.001000 Time 0.021030 -2023-02-13 17:47:14,128 - Epoch: [65][ 390/ 1207] Overall Loss 0.332639 Objective Loss 0.332639 LR 0.001000 Time 0.020973 -2023-02-13 17:47:14,317 - Epoch: [65][ 400/ 1207] Overall Loss 0.332959 Objective Loss 0.332959 LR 0.001000 Time 0.020919 -2023-02-13 17:47:14,505 - Epoch: [65][ 410/ 1207] Overall Loss 0.333450 Objective Loss 0.333450 LR 0.001000 Time 0.020867 -2023-02-13 17:47:14,694 - Epoch: [65][ 420/ 1207] Overall Loss 0.334415 Objective Loss 0.334415 LR 0.001000 Time 0.020819 -2023-02-13 17:47:14,884 - Epoch: [65][ 430/ 1207] Overall Loss 0.335250 Objective Loss 0.335250 LR 0.001000 Time 0.020775 -2023-02-13 17:47:15,073 - Epoch: [65][ 440/ 1207] Overall Loss 0.335270 Objective Loss 0.335270 LR 0.001000 Time 0.020733 -2023-02-13 17:47:15,262 - Epoch: [65][ 450/ 1207] Overall Loss 0.335047 Objective Loss 0.335047 LR 0.001000 Time 0.020690 -2023-02-13 17:47:15,450 - Epoch: [65][ 460/ 1207] Overall Loss 0.335103 Objective Loss 0.335103 LR 0.001000 Time 0.020650 -2023-02-13 17:47:15,639 - Epoch: [65][ 470/ 1207] Overall Loss 0.334635 Objective Loss 0.334635 LR 0.001000 Time 0.020612 -2023-02-13 17:47:15,830 - Epoch: [65][ 480/ 1207] Overall Loss 0.334500 Objective Loss 0.334500 LR 0.001000 Time 0.020578 -2023-02-13 17:47:16,020 - Epoch: [65][ 490/ 1207] Overall Loss 0.334179 Objective Loss 0.334179 LR 0.001000 Time 0.020546 -2023-02-13 17:47:16,208 - Epoch: [65][ 500/ 1207] Overall Loss 0.333813 Objective Loss 0.333813 LR 0.001000 Time 0.020511 -2023-02-13 17:47:16,396 - Epoch: [65][ 510/ 1207] Overall Loss 0.333592 Objective Loss 0.333592 LR 0.001000 Time 0.020477 -2023-02-13 17:47:16,585 - Epoch: [65][ 520/ 1207] Overall Loss 0.333503 Objective Loss 0.333503 LR 0.001000 Time 0.020445 -2023-02-13 17:47:16,774 - Epoch: [65][ 530/ 1207] Overall Loss 0.333633 Objective Loss 0.333633 LR 0.001000 Time 0.020415 -2023-02-13 17:47:16,964 - Epoch: [65][ 540/ 1207] Overall Loss 0.333844 Objective Loss 0.333844 LR 0.001000 Time 0.020388 -2023-02-13 17:47:17,152 - Epoch: [65][ 550/ 1207] Overall Loss 0.334092 Objective Loss 0.334092 LR 0.001000 Time 0.020358 -2023-02-13 17:47:17,340 - Epoch: [65][ 560/ 1207] Overall Loss 0.334606 Objective Loss 0.334606 LR 0.001000 Time 0.020330 -2023-02-13 17:47:17,528 - Epoch: [65][ 570/ 1207] Overall Loss 0.334750 Objective Loss 0.334750 LR 0.001000 Time 0.020303 -2023-02-13 17:47:17,716 - Epoch: [65][ 580/ 1207] Overall Loss 0.334345 Objective Loss 0.334345 LR 0.001000 Time 0.020277 -2023-02-13 17:47:17,905 - Epoch: [65][ 590/ 1207] Overall Loss 0.333953 Objective Loss 0.333953 LR 0.001000 Time 0.020252 -2023-02-13 17:47:18,094 - Epoch: [65][ 600/ 1207] Overall Loss 0.334515 Objective Loss 0.334515 LR 0.001000 Time 0.020230 -2023-02-13 17:47:18,283 - Epoch: [65][ 610/ 1207] Overall Loss 0.334135 Objective Loss 0.334135 LR 0.001000 Time 0.020206 -2023-02-13 17:47:18,471 - Epoch: [65][ 620/ 1207] Overall Loss 0.334283 Objective Loss 0.334283 LR 0.001000 Time 0.020184 -2023-02-13 17:47:18,660 - Epoch: [65][ 630/ 1207] Overall Loss 0.334982 Objective Loss 0.334982 LR 0.001000 Time 0.020162 -2023-02-13 17:47:18,848 - Epoch: [65][ 640/ 1207] Overall Loss 0.334690 Objective Loss 0.334690 LR 0.001000 Time 0.020141 -2023-02-13 17:47:19,038 - Epoch: [65][ 650/ 1207] Overall Loss 0.334892 Objective Loss 0.334892 LR 0.001000 Time 0.020122 -2023-02-13 17:47:19,227 - Epoch: [65][ 660/ 1207] Overall Loss 0.335245 Objective Loss 0.335245 LR 0.001000 Time 0.020103 -2023-02-13 17:47:19,415 - Epoch: [65][ 670/ 1207] Overall Loss 0.335690 Objective Loss 0.335690 LR 0.001000 Time 0.020083 -2023-02-13 17:47:19,603 - Epoch: [65][ 680/ 1207] Overall Loss 0.335678 Objective Loss 0.335678 LR 0.001000 Time 0.020064 -2023-02-13 17:47:19,792 - Epoch: [65][ 690/ 1207] Overall Loss 0.335946 Objective Loss 0.335946 LR 0.001000 Time 0.020047 -2023-02-13 17:47:19,982 - Epoch: [65][ 700/ 1207] Overall Loss 0.335571 Objective Loss 0.335571 LR 0.001000 Time 0.020031 -2023-02-13 17:47:20,170 - Epoch: [65][ 710/ 1207] Overall Loss 0.335577 Objective Loss 0.335577 LR 0.001000 Time 0.020013 -2023-02-13 17:47:20,359 - Epoch: [65][ 720/ 1207] Overall Loss 0.335690 Objective Loss 0.335690 LR 0.001000 Time 0.019997 -2023-02-13 17:47:20,547 - Epoch: [65][ 730/ 1207] Overall Loss 0.335643 Objective Loss 0.335643 LR 0.001000 Time 0.019981 -2023-02-13 17:47:20,736 - Epoch: [65][ 740/ 1207] Overall Loss 0.335437 Objective Loss 0.335437 LR 0.001000 Time 0.019965 -2023-02-13 17:47:20,926 - Epoch: [65][ 750/ 1207] Overall Loss 0.335874 Objective Loss 0.335874 LR 0.001000 Time 0.019952 -2023-02-13 17:47:21,116 - Epoch: [65][ 760/ 1207] Overall Loss 0.335643 Objective Loss 0.335643 LR 0.001000 Time 0.019938 -2023-02-13 17:47:21,304 - Epoch: [65][ 770/ 1207] Overall Loss 0.335530 Objective Loss 0.335530 LR 0.001000 Time 0.019923 -2023-02-13 17:47:21,493 - Epoch: [65][ 780/ 1207] Overall Loss 0.335355 Objective Loss 0.335355 LR 0.001000 Time 0.019909 -2023-02-13 17:47:21,682 - Epoch: [65][ 790/ 1207] Overall Loss 0.335114 Objective Loss 0.335114 LR 0.001000 Time 0.019896 -2023-02-13 17:47:21,871 - Epoch: [65][ 800/ 1207] Overall Loss 0.335271 Objective Loss 0.335271 LR 0.001000 Time 0.019884 -2023-02-13 17:47:22,061 - Epoch: [65][ 810/ 1207] Overall Loss 0.335279 Objective Loss 0.335279 LR 0.001000 Time 0.019873 -2023-02-13 17:47:22,250 - Epoch: [65][ 820/ 1207] Overall Loss 0.335689 Objective Loss 0.335689 LR 0.001000 Time 0.019860 -2023-02-13 17:47:22,439 - Epoch: [65][ 830/ 1207] Overall Loss 0.335996 Objective Loss 0.335996 LR 0.001000 Time 0.019847 -2023-02-13 17:47:22,627 - Epoch: [65][ 840/ 1207] Overall Loss 0.335927 Objective Loss 0.335927 LR 0.001000 Time 0.019835 -2023-02-13 17:47:22,816 - Epoch: [65][ 850/ 1207] Overall Loss 0.336067 Objective Loss 0.336067 LR 0.001000 Time 0.019823 -2023-02-13 17:47:23,006 - Epoch: [65][ 860/ 1207] Overall Loss 0.335882 Objective Loss 0.335882 LR 0.001000 Time 0.019814 -2023-02-13 17:47:23,195 - Epoch: [65][ 870/ 1207] Overall Loss 0.335744 Objective Loss 0.335744 LR 0.001000 Time 0.019803 -2023-02-13 17:47:23,384 - Epoch: [65][ 880/ 1207] Overall Loss 0.335438 Objective Loss 0.335438 LR 0.001000 Time 0.019792 -2023-02-13 17:47:23,573 - Epoch: [65][ 890/ 1207] Overall Loss 0.335470 Objective Loss 0.335470 LR 0.001000 Time 0.019782 -2023-02-13 17:47:23,762 - Epoch: [65][ 900/ 1207] Overall Loss 0.335678 Objective Loss 0.335678 LR 0.001000 Time 0.019772 -2023-02-13 17:47:23,952 - Epoch: [65][ 910/ 1207] Overall Loss 0.335573 Objective Loss 0.335573 LR 0.001000 Time 0.019762 -2023-02-13 17:47:24,142 - Epoch: [65][ 920/ 1207] Overall Loss 0.335728 Objective Loss 0.335728 LR 0.001000 Time 0.019753 -2023-02-13 17:47:24,331 - Epoch: [65][ 930/ 1207] Overall Loss 0.335952 Objective Loss 0.335952 LR 0.001000 Time 0.019744 -2023-02-13 17:47:24,519 - Epoch: [65][ 940/ 1207] Overall Loss 0.335984 Objective Loss 0.335984 LR 0.001000 Time 0.019734 -2023-02-13 17:47:24,708 - Epoch: [65][ 950/ 1207] Overall Loss 0.335788 Objective Loss 0.335788 LR 0.001000 Time 0.019725 -2023-02-13 17:47:24,897 - Epoch: [65][ 960/ 1207] Overall Loss 0.335663 Objective Loss 0.335663 LR 0.001000 Time 0.019716 -2023-02-13 17:47:25,086 - Epoch: [65][ 970/ 1207] Overall Loss 0.335857 Objective Loss 0.335857 LR 0.001000 Time 0.019707 -2023-02-13 17:47:25,276 - Epoch: [65][ 980/ 1207] Overall Loss 0.335898 Objective Loss 0.335898 LR 0.001000 Time 0.019699 -2023-02-13 17:47:25,464 - Epoch: [65][ 990/ 1207] Overall Loss 0.336198 Objective Loss 0.336198 LR 0.001000 Time 0.019690 -2023-02-13 17:47:25,654 - Epoch: [65][ 1000/ 1207] Overall Loss 0.336186 Objective Loss 0.336186 LR 0.001000 Time 0.019682 -2023-02-13 17:47:25,843 - Epoch: [65][ 1010/ 1207] Overall Loss 0.336135 Objective Loss 0.336135 LR 0.001000 Time 0.019675 -2023-02-13 17:47:26,033 - Epoch: [65][ 1020/ 1207] Overall Loss 0.336344 Objective Loss 0.336344 LR 0.001000 Time 0.019667 -2023-02-13 17:47:26,222 - Epoch: [65][ 1030/ 1207] Overall Loss 0.336442 Objective Loss 0.336442 LR 0.001000 Time 0.019659 -2023-02-13 17:47:26,411 - Epoch: [65][ 1040/ 1207] Overall Loss 0.336696 Objective Loss 0.336696 LR 0.001000 Time 0.019652 -2023-02-13 17:47:26,600 - Epoch: [65][ 1050/ 1207] Overall Loss 0.336741 Objective Loss 0.336741 LR 0.001000 Time 0.019644 -2023-02-13 17:47:26,789 - Epoch: [65][ 1060/ 1207] Overall Loss 0.336849 Objective Loss 0.336849 LR 0.001000 Time 0.019637 -2023-02-13 17:47:26,979 - Epoch: [65][ 1070/ 1207] Overall Loss 0.337262 Objective Loss 0.337262 LR 0.001000 Time 0.019631 -2023-02-13 17:47:27,169 - Epoch: [65][ 1080/ 1207] Overall Loss 0.337421 Objective Loss 0.337421 LR 0.001000 Time 0.019624 -2023-02-13 17:47:27,358 - Epoch: [65][ 1090/ 1207] Overall Loss 0.337520 Objective Loss 0.337520 LR 0.001000 Time 0.019617 -2023-02-13 17:47:27,547 - Epoch: [65][ 1100/ 1207] Overall Loss 0.337446 Objective Loss 0.337446 LR 0.001000 Time 0.019611 -2023-02-13 17:47:27,737 - Epoch: [65][ 1110/ 1207] Overall Loss 0.337278 Objective Loss 0.337278 LR 0.001000 Time 0.019605 -2023-02-13 17:47:27,927 - Epoch: [65][ 1120/ 1207] Overall Loss 0.337051 Objective Loss 0.337051 LR 0.001000 Time 0.019599 -2023-02-13 17:47:28,116 - Epoch: [65][ 1130/ 1207] Overall Loss 0.337035 Objective Loss 0.337035 LR 0.001000 Time 0.019593 -2023-02-13 17:47:28,305 - Epoch: [65][ 1140/ 1207] Overall Loss 0.337247 Objective Loss 0.337247 LR 0.001000 Time 0.019586 -2023-02-13 17:47:28,494 - Epoch: [65][ 1150/ 1207] Overall Loss 0.337205 Objective Loss 0.337205 LR 0.001000 Time 0.019580 -2023-02-13 17:47:28,682 - Epoch: [65][ 1160/ 1207] Overall Loss 0.337394 Objective Loss 0.337394 LR 0.001000 Time 0.019573 -2023-02-13 17:47:28,871 - Epoch: [65][ 1170/ 1207] Overall Loss 0.337373 Objective Loss 0.337373 LR 0.001000 Time 0.019567 -2023-02-13 17:47:29,061 - Epoch: [65][ 1180/ 1207] Overall Loss 0.337644 Objective Loss 0.337644 LR 0.001000 Time 0.019562 -2023-02-13 17:47:29,249 - Epoch: [65][ 1190/ 1207] Overall Loss 0.337632 Objective Loss 0.337632 LR 0.001000 Time 0.019555 -2023-02-13 17:47:29,495 - Epoch: [65][ 1200/ 1207] Overall Loss 0.337651 Objective Loss 0.337651 LR 0.001000 Time 0.019596 -2023-02-13 17:47:29,610 - Epoch: [65][ 1207/ 1207] Overall Loss 0.337493 Objective Loss 0.337493 Top1 80.487805 Top5 96.646341 LR 0.001000 Time 0.019578 -2023-02-13 17:47:29,681 - --- validate (epoch=65)----------- -2023-02-13 17:47:29,681 - 34311 samples (256 per mini-batch) -2023-02-13 17:47:30,089 - Epoch: [65][ 10/ 135] Loss 0.356628 Top1 83.046875 Top5 97.343750 -2023-02-13 17:47:30,230 - Epoch: [65][ 20/ 135] Loss 0.357493 Top1 82.304688 Top5 97.187500 -2023-02-13 17:47:30,366 - Epoch: [65][ 30/ 135] Loss 0.363909 Top1 81.927083 Top5 97.187500 -2023-02-13 17:47:30,493 - Epoch: [65][ 40/ 135] Loss 0.360383 Top1 81.865234 Top5 97.187500 -2023-02-13 17:47:30,619 - Epoch: [65][ 50/ 135] Loss 0.355434 Top1 82.046875 Top5 97.265625 -2023-02-13 17:47:30,745 - Epoch: [65][ 60/ 135] Loss 0.359707 Top1 82.109375 Top5 97.220052 -2023-02-13 17:47:30,870 - Epoch: [65][ 70/ 135] Loss 0.367493 Top1 81.886161 Top5 97.176339 -2023-02-13 17:47:30,997 - Epoch: [65][ 80/ 135] Loss 0.368078 Top1 81.821289 Top5 97.133789 -2023-02-13 17:47:31,123 - Epoch: [65][ 90/ 135] Loss 0.366400 Top1 81.883681 Top5 97.152778 -2023-02-13 17:47:31,252 - Epoch: [65][ 100/ 135] Loss 0.364216 Top1 81.824219 Top5 97.132812 -2023-02-13 17:47:31,380 - Epoch: [65][ 110/ 135] Loss 0.362424 Top1 81.892756 Top5 97.169744 -2023-02-13 17:47:31,509 - Epoch: [65][ 120/ 135] Loss 0.365744 Top1 81.796875 Top5 97.128906 -2023-02-13 17:47:31,641 - Epoch: [65][ 130/ 135] Loss 0.365573 Top1 81.844952 Top5 97.157452 -2023-02-13 17:47:31,685 - Epoch: [65][ 135/ 135] Loss 0.366943 Top1 81.743464 Top5 97.135030 -2023-02-13 17:47:31,762 - ==> Top1: 81.743 Top5: 97.135 Loss: 0.367 - -2023-02-13 17:47:31,762 - ==> Confusion: -[[ 822 3 5 4 16 3 0 5 3 67 0 7 0 4 3 2 9 4 2 1 7] - [ 2 888 0 5 12 48 3 31 3 1 3 8 2 0 0 4 6 0 9 2 6] - [ 8 6 911 16 3 3 34 20 0 0 4 3 0 5 3 9 1 5 9 7 11] - [ 4 1 14 910 2 6 3 2 2 3 13 1 6 3 10 2 2 9 14 0 9] - [ 15 9 1 0 975 17 1 4 0 8 0 6 2 4 3 4 6 2 0 3 6] - [ 1 10 1 3 5 974 3 20 0 3 1 14 7 15 1 1 2 2 1 3 3] - [ 4 2 6 1 1 4 1041 7 1 0 0 2 1 2 0 8 3 3 1 5 7] - [ 1 3 6 2 4 37 6 906 0 1 4 8 2 1 1 0 1 3 25 11 2] - [ 20 4 0 1 2 3 0 3 859 46 12 3 0 16 20 3 2 1 13 0 1] - [ 71 1 2 1 5 3 1 0 27 854 1 1 0 28 3 1 1 3 1 1 7] - [ 3 4 2 14 1 3 5 11 14 3 946 1 1 14 4 0 1 0 17 0 7] - [ 0 2 0 1 4 8 1 7 2 2 0 912 30 12 0 5 0 6 2 6 5] - [ 1 0 0 8 1 5 0 1 1 0 0 45 851 1 2 10 3 20 2 1 7] - [ 5 1 1 0 10 14 0 3 6 11 6 6 1 943 4 5 4 2 0 1 1] - [ 11 3 4 30 10 9 0 2 11 5 2 1 4 4 958 1 5 9 15 0 8] - [ 3 1 1 0 6 2 6 2 0 2 0 8 7 3 0 972 5 13 0 8 7] - [ 3 5 1 0 11 4 0 1 2 1 0 2 0 2 1 12 998 1 1 5 11] - [ 7 0 1 8 1 1 1 0 1 0 0 12 19 1 0 13 0 978 0 1 7] - [ 4 4 4 16 1 5 0 39 5 2 5 2 9 0 11 0 1 1 973 1 3] - [ 0 2 0 0 2 16 6 16 1 0 0 24 5 5 0 6 4 3 1 1049 8] - [ 156 271 168 158 146 342 126 272 82 97 165 162 332 401 158 112 316 146 184 313 9327]] - -2023-02-13 17:47:31,764 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:47:31,764 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:47:31,770 - - -2023-02-13 17:47:31,770 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:47:32,664 - Epoch: [66][ 10/ 1207] Overall Loss 0.348010 Objective Loss 0.348010 LR 0.001000 Time 0.089279 -2023-02-13 17:47:32,865 - Epoch: [66][ 20/ 1207] Overall Loss 0.333128 Objective Loss 0.333128 LR 0.001000 Time 0.054669 -2023-02-13 17:47:33,062 - Epoch: [66][ 30/ 1207] Overall Loss 0.336255 Objective Loss 0.336255 LR 0.001000 Time 0.043003 -2023-02-13 17:47:33,256 - Epoch: [66][ 40/ 1207] Overall Loss 0.335148 Objective Loss 0.335148 LR 0.001000 Time 0.037093 -2023-02-13 17:47:33,451 - Epoch: [66][ 50/ 1207] Overall Loss 0.336123 Objective Loss 0.336123 LR 0.001000 Time 0.033583 -2023-02-13 17:47:33,645 - Epoch: [66][ 60/ 1207] Overall Loss 0.331429 Objective Loss 0.331429 LR 0.001000 Time 0.031212 -2023-02-13 17:47:33,842 - Epoch: [66][ 70/ 1207] Overall Loss 0.330716 Objective Loss 0.330716 LR 0.001000 Time 0.029553 -2023-02-13 17:47:34,037 - Epoch: [66][ 80/ 1207] Overall Loss 0.328457 Objective Loss 0.328457 LR 0.001000 Time 0.028288 -2023-02-13 17:47:34,232 - Epoch: [66][ 90/ 1207] Overall Loss 0.326722 Objective Loss 0.326722 LR 0.001000 Time 0.027313 -2023-02-13 17:47:34,426 - Epoch: [66][ 100/ 1207] Overall Loss 0.332024 Objective Loss 0.332024 LR 0.001000 Time 0.026513 -2023-02-13 17:47:34,622 - Epoch: [66][ 110/ 1207] Overall Loss 0.327588 Objective Loss 0.327588 LR 0.001000 Time 0.025885 -2023-02-13 17:47:34,816 - Epoch: [66][ 120/ 1207] Overall Loss 0.326120 Objective Loss 0.326120 LR 0.001000 Time 0.025341 -2023-02-13 17:47:35,012 - Epoch: [66][ 130/ 1207] Overall Loss 0.325209 Objective Loss 0.325209 LR 0.001000 Time 0.024901 -2023-02-13 17:47:35,207 - Epoch: [66][ 140/ 1207] Overall Loss 0.325827 Objective Loss 0.325827 LR 0.001000 Time 0.024507 -2023-02-13 17:47:35,402 - Epoch: [66][ 150/ 1207] Overall Loss 0.325524 Objective Loss 0.325524 LR 0.001000 Time 0.024174 -2023-02-13 17:47:35,596 - Epoch: [66][ 160/ 1207] Overall Loss 0.324855 Objective Loss 0.324855 LR 0.001000 Time 0.023872 -2023-02-13 17:47:35,793 - Epoch: [66][ 170/ 1207] Overall Loss 0.325709 Objective Loss 0.325709 LR 0.001000 Time 0.023624 -2023-02-13 17:47:35,988 - Epoch: [66][ 180/ 1207] Overall Loss 0.325874 Objective Loss 0.325874 LR 0.001000 Time 0.023394 -2023-02-13 17:47:36,185 - Epoch: [66][ 190/ 1207] Overall Loss 0.326845 Objective Loss 0.326845 LR 0.001000 Time 0.023196 -2023-02-13 17:47:36,379 - Epoch: [66][ 200/ 1207] Overall Loss 0.330027 Objective Loss 0.330027 LR 0.001000 Time 0.023003 -2023-02-13 17:47:36,574 - Epoch: [66][ 210/ 1207] Overall Loss 0.331304 Objective Loss 0.331304 LR 0.001000 Time 0.022838 -2023-02-13 17:47:36,769 - Epoch: [66][ 220/ 1207] Overall Loss 0.332061 Objective Loss 0.332061 LR 0.001000 Time 0.022681 -2023-02-13 17:47:36,965 - Epoch: [66][ 230/ 1207] Overall Loss 0.331937 Objective Loss 0.331937 LR 0.001000 Time 0.022549 -2023-02-13 17:47:37,160 - Epoch: [66][ 240/ 1207] Overall Loss 0.331705 Objective Loss 0.331705 LR 0.001000 Time 0.022421 -2023-02-13 17:47:37,357 - Epoch: [66][ 250/ 1207] Overall Loss 0.331108 Objective Loss 0.331108 LR 0.001000 Time 0.022307 -2023-02-13 17:47:37,550 - Epoch: [66][ 260/ 1207] Overall Loss 0.330739 Objective Loss 0.330739 LR 0.001000 Time 0.022192 -2023-02-13 17:47:37,746 - Epoch: [66][ 270/ 1207] Overall Loss 0.331165 Objective Loss 0.331165 LR 0.001000 Time 0.022095 -2023-02-13 17:47:37,941 - Epoch: [66][ 280/ 1207] Overall Loss 0.330725 Objective Loss 0.330725 LR 0.001000 Time 0.022000 -2023-02-13 17:47:38,139 - Epoch: [66][ 290/ 1207] Overall Loss 0.330012 Objective Loss 0.330012 LR 0.001000 Time 0.021924 -2023-02-13 17:47:38,335 - Epoch: [66][ 300/ 1207] Overall Loss 0.330405 Objective Loss 0.330405 LR 0.001000 Time 0.021844 -2023-02-13 17:47:38,533 - Epoch: [66][ 310/ 1207] Overall Loss 0.330251 Objective Loss 0.330251 LR 0.001000 Time 0.021778 -2023-02-13 17:47:38,729 - Epoch: [66][ 320/ 1207] Overall Loss 0.328999 Objective Loss 0.328999 LR 0.001000 Time 0.021707 -2023-02-13 17:47:38,928 - Epoch: [66][ 330/ 1207] Overall Loss 0.330358 Objective Loss 0.330358 LR 0.001000 Time 0.021652 -2023-02-13 17:47:39,124 - Epoch: [66][ 340/ 1207] Overall Loss 0.330379 Objective Loss 0.330379 LR 0.001000 Time 0.021591 -2023-02-13 17:47:39,322 - Epoch: [66][ 350/ 1207] Overall Loss 0.331274 Objective Loss 0.331274 LR 0.001000 Time 0.021538 -2023-02-13 17:47:39,517 - Epoch: [66][ 360/ 1207] Overall Loss 0.331091 Objective Loss 0.331091 LR 0.001000 Time 0.021482 -2023-02-13 17:47:39,717 - Epoch: [66][ 370/ 1207] Overall Loss 0.331156 Objective Loss 0.331156 LR 0.001000 Time 0.021439 -2023-02-13 17:47:39,913 - Epoch: [66][ 380/ 1207] Overall Loss 0.330716 Objective Loss 0.330716 LR 0.001000 Time 0.021389 -2023-02-13 17:47:40,112 - Epoch: [66][ 390/ 1207] Overall Loss 0.330370 Objective Loss 0.330370 LR 0.001000 Time 0.021350 -2023-02-13 17:47:40,307 - Epoch: [66][ 400/ 1207] Overall Loss 0.329377 Objective Loss 0.329377 LR 0.001000 Time 0.021304 -2023-02-13 17:47:40,505 - Epoch: [66][ 410/ 1207] Overall Loss 0.328373 Objective Loss 0.328373 LR 0.001000 Time 0.021267 -2023-02-13 17:47:40,701 - Epoch: [66][ 420/ 1207] Overall Loss 0.327626 Objective Loss 0.327626 LR 0.001000 Time 0.021225 -2023-02-13 17:47:40,900 - Epoch: [66][ 430/ 1207] Overall Loss 0.327190 Objective Loss 0.327190 LR 0.001000 Time 0.021194 -2023-02-13 17:47:41,096 - Epoch: [66][ 440/ 1207] Overall Loss 0.327287 Objective Loss 0.327287 LR 0.001000 Time 0.021158 -2023-02-13 17:47:41,296 - Epoch: [66][ 450/ 1207] Overall Loss 0.327100 Objective Loss 0.327100 LR 0.001000 Time 0.021131 -2023-02-13 17:47:41,491 - Epoch: [66][ 460/ 1207] Overall Loss 0.326661 Objective Loss 0.326661 LR 0.001000 Time 0.021094 -2023-02-13 17:47:41,690 - Epoch: [66][ 470/ 1207] Overall Loss 0.326905 Objective Loss 0.326905 LR 0.001000 Time 0.021068 -2023-02-13 17:47:41,886 - Epoch: [66][ 480/ 1207] Overall Loss 0.327194 Objective Loss 0.327194 LR 0.001000 Time 0.021037 -2023-02-13 17:47:42,086 - Epoch: [66][ 490/ 1207] Overall Loss 0.327539 Objective Loss 0.327539 LR 0.001000 Time 0.021014 -2023-02-13 17:47:42,281 - Epoch: [66][ 500/ 1207] Overall Loss 0.327316 Objective Loss 0.327316 LR 0.001000 Time 0.020984 -2023-02-13 17:47:42,481 - Epoch: [66][ 510/ 1207] Overall Loss 0.327916 Objective Loss 0.327916 LR 0.001000 Time 0.020963 -2023-02-13 17:47:42,677 - Epoch: [66][ 520/ 1207] Overall Loss 0.327671 Objective Loss 0.327671 LR 0.001000 Time 0.020936 -2023-02-13 17:47:42,875 - Epoch: [66][ 530/ 1207] Overall Loss 0.327288 Objective Loss 0.327288 LR 0.001000 Time 0.020915 -2023-02-13 17:47:43,072 - Epoch: [66][ 540/ 1207] Overall Loss 0.327728 Objective Loss 0.327728 LR 0.001000 Time 0.020891 -2023-02-13 17:47:43,271 - Epoch: [66][ 550/ 1207] Overall Loss 0.327461 Objective Loss 0.327461 LR 0.001000 Time 0.020872 -2023-02-13 17:47:43,466 - Epoch: [66][ 560/ 1207] Overall Loss 0.327451 Objective Loss 0.327451 LR 0.001000 Time 0.020847 -2023-02-13 17:47:43,664 - Epoch: [66][ 570/ 1207] Overall Loss 0.327387 Objective Loss 0.327387 LR 0.001000 Time 0.020828 -2023-02-13 17:47:43,859 - Epoch: [66][ 580/ 1207] Overall Loss 0.327063 Objective Loss 0.327063 LR 0.001000 Time 0.020805 -2023-02-13 17:47:44,059 - Epoch: [66][ 590/ 1207] Overall Loss 0.327555 Objective Loss 0.327555 LR 0.001000 Time 0.020790 -2023-02-13 17:47:44,254 - Epoch: [66][ 600/ 1207] Overall Loss 0.327812 Objective Loss 0.327812 LR 0.001000 Time 0.020768 -2023-02-13 17:47:44,452 - Epoch: [66][ 610/ 1207] Overall Loss 0.327799 Objective Loss 0.327799 LR 0.001000 Time 0.020753 -2023-02-13 17:47:44,647 - Epoch: [66][ 620/ 1207] Overall Loss 0.327986 Objective Loss 0.327986 LR 0.001000 Time 0.020731 -2023-02-13 17:47:44,846 - Epoch: [66][ 630/ 1207] Overall Loss 0.327726 Objective Loss 0.327726 LR 0.001000 Time 0.020717 -2023-02-13 17:47:45,041 - Epoch: [66][ 640/ 1207] Overall Loss 0.328065 Objective Loss 0.328065 LR 0.001000 Time 0.020698 -2023-02-13 17:47:45,241 - Epoch: [66][ 650/ 1207] Overall Loss 0.328064 Objective Loss 0.328064 LR 0.001000 Time 0.020686 -2023-02-13 17:47:45,436 - Epoch: [66][ 660/ 1207] Overall Loss 0.328308 Objective Loss 0.328308 LR 0.001000 Time 0.020668 -2023-02-13 17:47:45,634 - Epoch: [66][ 670/ 1207] Overall Loss 0.328353 Objective Loss 0.328353 LR 0.001000 Time 0.020654 -2023-02-13 17:47:45,830 - Epoch: [66][ 680/ 1207] Overall Loss 0.328584 Objective Loss 0.328584 LR 0.001000 Time 0.020638 -2023-02-13 17:47:46,030 - Epoch: [66][ 690/ 1207] Overall Loss 0.328399 Objective Loss 0.328399 LR 0.001000 Time 0.020628 -2023-02-13 17:47:46,226 - Epoch: [66][ 700/ 1207] Overall Loss 0.328539 Objective Loss 0.328539 LR 0.001000 Time 0.020613 -2023-02-13 17:47:46,424 - Epoch: [66][ 710/ 1207] Overall Loss 0.328759 Objective Loss 0.328759 LR 0.001000 Time 0.020602 -2023-02-13 17:47:46,620 - Epoch: [66][ 720/ 1207] Overall Loss 0.328650 Objective Loss 0.328650 LR 0.001000 Time 0.020587 -2023-02-13 17:47:46,819 - Epoch: [66][ 730/ 1207] Overall Loss 0.328228 Objective Loss 0.328228 LR 0.001000 Time 0.020576 -2023-02-13 17:47:47,015 - Epoch: [66][ 740/ 1207] Overall Loss 0.328441 Objective Loss 0.328441 LR 0.001000 Time 0.020564 -2023-02-13 17:47:47,214 - Epoch: [66][ 750/ 1207] Overall Loss 0.329227 Objective Loss 0.329227 LR 0.001000 Time 0.020554 -2023-02-13 17:47:47,411 - Epoch: [66][ 760/ 1207] Overall Loss 0.329195 Objective Loss 0.329195 LR 0.001000 Time 0.020541 -2023-02-13 17:47:47,609 - Epoch: [66][ 770/ 1207] Overall Loss 0.329255 Objective Loss 0.329255 LR 0.001000 Time 0.020533 -2023-02-13 17:47:47,805 - Epoch: [66][ 780/ 1207] Overall Loss 0.328905 Objective Loss 0.328905 LR 0.001000 Time 0.020520 -2023-02-13 17:47:48,004 - Epoch: [66][ 790/ 1207] Overall Loss 0.328700 Objective Loss 0.328700 LR 0.001000 Time 0.020511 -2023-02-13 17:47:48,201 - Epoch: [66][ 800/ 1207] Overall Loss 0.328856 Objective Loss 0.328856 LR 0.001000 Time 0.020501 -2023-02-13 17:47:48,401 - Epoch: [66][ 810/ 1207] Overall Loss 0.329045 Objective Loss 0.329045 LR 0.001000 Time 0.020493 -2023-02-13 17:47:48,596 - Epoch: [66][ 820/ 1207] Overall Loss 0.329218 Objective Loss 0.329218 LR 0.001000 Time 0.020481 -2023-02-13 17:47:48,794 - Epoch: [66][ 830/ 1207] Overall Loss 0.328945 Objective Loss 0.328945 LR 0.001000 Time 0.020473 -2023-02-13 17:47:48,990 - Epoch: [66][ 840/ 1207] Overall Loss 0.329517 Objective Loss 0.329517 LR 0.001000 Time 0.020462 -2023-02-13 17:47:49,189 - Epoch: [66][ 850/ 1207] Overall Loss 0.329166 Objective Loss 0.329166 LR 0.001000 Time 0.020455 -2023-02-13 17:47:49,385 - Epoch: [66][ 860/ 1207] Overall Loss 0.329335 Objective Loss 0.329335 LR 0.001000 Time 0.020445 -2023-02-13 17:47:49,585 - Epoch: [66][ 870/ 1207] Overall Loss 0.329665 Objective Loss 0.329665 LR 0.001000 Time 0.020439 -2023-02-13 17:47:49,781 - Epoch: [66][ 880/ 1207] Overall Loss 0.329772 Objective Loss 0.329772 LR 0.001000 Time 0.020429 -2023-02-13 17:47:49,980 - Epoch: [66][ 890/ 1207] Overall Loss 0.330108 Objective Loss 0.330108 LR 0.001000 Time 0.020423 -2023-02-13 17:47:50,177 - Epoch: [66][ 900/ 1207] Overall Loss 0.330275 Objective Loss 0.330275 LR 0.001000 Time 0.020414 -2023-02-13 17:47:50,375 - Epoch: [66][ 910/ 1207] Overall Loss 0.330498 Objective Loss 0.330498 LR 0.001000 Time 0.020407 -2023-02-13 17:47:50,571 - Epoch: [66][ 920/ 1207] Overall Loss 0.330401 Objective Loss 0.330401 LR 0.001000 Time 0.020398 -2023-02-13 17:47:50,770 - Epoch: [66][ 930/ 1207] Overall Loss 0.330150 Objective Loss 0.330150 LR 0.001000 Time 0.020392 -2023-02-13 17:47:50,967 - Epoch: [66][ 940/ 1207] Overall Loss 0.330243 Objective Loss 0.330243 LR 0.001000 Time 0.020384 -2023-02-13 17:47:51,167 - Epoch: [66][ 950/ 1207] Overall Loss 0.330660 Objective Loss 0.330660 LR 0.001000 Time 0.020380 -2023-02-13 17:47:51,363 - Epoch: [66][ 960/ 1207] Overall Loss 0.330604 Objective Loss 0.330604 LR 0.001000 Time 0.020371 -2023-02-13 17:47:51,562 - Epoch: [66][ 970/ 1207] Overall Loss 0.330732 Objective Loss 0.330732 LR 0.001000 Time 0.020366 -2023-02-13 17:47:51,758 - Epoch: [66][ 980/ 1207] Overall Loss 0.330669 Objective Loss 0.330669 LR 0.001000 Time 0.020358 -2023-02-13 17:47:51,958 - Epoch: [66][ 990/ 1207] Overall Loss 0.330862 Objective Loss 0.330862 LR 0.001000 Time 0.020354 -2023-02-13 17:47:52,154 - Epoch: [66][ 1000/ 1207] Overall Loss 0.331268 Objective Loss 0.331268 LR 0.001000 Time 0.020346 -2023-02-13 17:47:52,353 - Epoch: [66][ 1010/ 1207] Overall Loss 0.331305 Objective Loss 0.331305 LR 0.001000 Time 0.020341 -2023-02-13 17:47:52,549 - Epoch: [66][ 1020/ 1207] Overall Loss 0.331377 Objective Loss 0.331377 LR 0.001000 Time 0.020334 -2023-02-13 17:47:52,748 - Epoch: [66][ 1030/ 1207] Overall Loss 0.331460 Objective Loss 0.331460 LR 0.001000 Time 0.020329 -2023-02-13 17:47:52,944 - Epoch: [66][ 1040/ 1207] Overall Loss 0.331641 Objective Loss 0.331641 LR 0.001000 Time 0.020322 -2023-02-13 17:47:53,143 - Epoch: [66][ 1050/ 1207] Overall Loss 0.331521 Objective Loss 0.331521 LR 0.001000 Time 0.020318 -2023-02-13 17:47:53,339 - Epoch: [66][ 1060/ 1207] Overall Loss 0.331495 Objective Loss 0.331495 LR 0.001000 Time 0.020310 -2023-02-13 17:47:53,538 - Epoch: [66][ 1070/ 1207] Overall Loss 0.331737 Objective Loss 0.331737 LR 0.001000 Time 0.020306 -2023-02-13 17:47:53,734 - Epoch: [66][ 1080/ 1207] Overall Loss 0.331791 Objective Loss 0.331791 LR 0.001000 Time 0.020299 -2023-02-13 17:47:53,933 - Epoch: [66][ 1090/ 1207] Overall Loss 0.331913 Objective Loss 0.331913 LR 0.001000 Time 0.020295 -2023-02-13 17:47:54,130 - Epoch: [66][ 1100/ 1207] Overall Loss 0.332209 Objective Loss 0.332209 LR 0.001000 Time 0.020289 -2023-02-13 17:47:54,329 - Epoch: [66][ 1110/ 1207] Overall Loss 0.332458 Objective Loss 0.332458 LR 0.001000 Time 0.020285 -2023-02-13 17:47:54,524 - Epoch: [66][ 1120/ 1207] Overall Loss 0.332568 Objective Loss 0.332568 LR 0.001000 Time 0.020278 -2023-02-13 17:47:54,723 - Epoch: [66][ 1130/ 1207] Overall Loss 0.332693 Objective Loss 0.332693 LR 0.001000 Time 0.020275 -2023-02-13 17:47:54,919 - Epoch: [66][ 1140/ 1207] Overall Loss 0.332895 Objective Loss 0.332895 LR 0.001000 Time 0.020268 -2023-02-13 17:47:55,119 - Epoch: [66][ 1150/ 1207] Overall Loss 0.333290 Objective Loss 0.333290 LR 0.001000 Time 0.020265 -2023-02-13 17:47:55,314 - Epoch: [66][ 1160/ 1207] Overall Loss 0.333342 Objective Loss 0.333342 LR 0.001000 Time 0.020259 -2023-02-13 17:47:55,513 - Epoch: [66][ 1170/ 1207] Overall Loss 0.333263 Objective Loss 0.333263 LR 0.001000 Time 0.020255 -2023-02-13 17:47:55,710 - Epoch: [66][ 1180/ 1207] Overall Loss 0.333260 Objective Loss 0.333260 LR 0.001000 Time 0.020250 -2023-02-13 17:47:55,910 - Epoch: [66][ 1190/ 1207] Overall Loss 0.333327 Objective Loss 0.333327 LR 0.001000 Time 0.020248 -2023-02-13 17:47:56,160 - Epoch: [66][ 1200/ 1207] Overall Loss 0.333392 Objective Loss 0.333392 LR 0.001000 Time 0.020287 -2023-02-13 17:47:56,275 - Epoch: [66][ 1207/ 1207] Overall Loss 0.333358 Objective Loss 0.333358 Top1 81.707317 Top5 97.256098 LR 0.001000 Time 0.020264 -2023-02-13 17:47:56,350 - --- validate (epoch=66)----------- -2023-02-13 17:47:56,350 - 34311 samples (256 per mini-batch) -2023-02-13 17:47:56,751 - Epoch: [66][ 10/ 135] Loss 0.372293 Top1 81.640625 Top5 96.875000 -2023-02-13 17:47:56,878 - Epoch: [66][ 20/ 135] Loss 0.369421 Top1 81.621094 Top5 96.875000 -2023-02-13 17:47:57,001 - Epoch: [66][ 30/ 135] Loss 0.366998 Top1 81.445312 Top5 96.861979 -2023-02-13 17:47:57,141 - Epoch: [66][ 40/ 135] Loss 0.363958 Top1 81.669922 Top5 97.001953 -2023-02-13 17:47:57,276 - Epoch: [66][ 50/ 135] Loss 0.359394 Top1 81.687500 Top5 97.031250 -2023-02-13 17:47:57,420 - Epoch: [66][ 60/ 135] Loss 0.359315 Top1 81.809896 Top5 97.063802 -2023-02-13 17:47:57,557 - Epoch: [66][ 70/ 135] Loss 0.363753 Top1 81.897321 Top5 97.014509 -2023-02-13 17:47:57,702 - Epoch: [66][ 80/ 135] Loss 0.366155 Top1 81.860352 Top5 97.006836 -2023-02-13 17:47:57,839 - Epoch: [66][ 90/ 135] Loss 0.367012 Top1 81.848958 Top5 96.966146 -2023-02-13 17:47:57,983 - Epoch: [66][ 100/ 135] Loss 0.366048 Top1 81.828125 Top5 96.984375 -2023-02-13 17:47:58,122 - Epoch: [66][ 110/ 135] Loss 0.362427 Top1 81.938920 Top5 96.960227 -2023-02-13 17:47:58,268 - Epoch: [66][ 120/ 135] Loss 0.364101 Top1 81.881510 Top5 97.001953 -2023-02-13 17:47:58,402 - Epoch: [66][ 130/ 135] Loss 0.363211 Top1 81.893029 Top5 97.022236 -2023-02-13 17:47:58,447 - Epoch: [66][ 135/ 135] Loss 0.367956 Top1 81.950395 Top5 97.021363 -2023-02-13 17:47:58,515 - ==> Top1: 81.950 Top5: 97.021 Loss: 0.368 - -2023-02-13 17:47:58,516 - ==> Confusion: -[[ 828 5 9 2 12 5 0 2 2 67 1 3 1 2 6 5 3 1 3 2 8] - [ 2 937 1 2 9 25 2 12 5 1 0 2 2 0 0 3 3 2 11 4 10] - [ 11 3 948 19 4 0 19 12 0 1 3 2 4 3 1 3 2 2 13 3 5] - [ 2 3 18 902 3 2 4 3 2 2 11 0 4 3 22 2 3 7 14 1 8] - [ 17 10 3 0 979 13 1 1 1 5 0 4 4 1 6 5 3 3 2 2 6] - [ 5 27 0 7 7 937 2 20 0 7 3 11 3 15 2 3 6 2 3 7 3] - [ 4 4 18 2 1 5 1034 3 0 0 3 0 3 0 0 4 0 3 1 11 3] - [ 1 18 10 3 2 28 9 886 1 2 2 4 2 1 0 0 3 2 29 14 7] - [ 22 2 0 1 1 0 1 0 876 44 14 2 0 7 25 3 1 3 6 0 1] - [ 85 1 4 0 7 1 1 0 41 831 1 4 0 13 11 3 0 3 2 2 2] - [ 1 4 1 9 4 2 3 1 14 3 966 1 2 7 6 1 1 1 16 3 5] - [ 2 2 1 0 6 16 2 10 0 1 1 885 32 4 3 7 5 8 4 15 1] - [ 2 0 0 5 1 4 0 4 1 0 0 32 862 0 4 5 3 26 3 2 5] - [ 5 5 4 0 9 14 2 3 24 27 17 4 1 872 7 8 7 3 2 5 5] - [ 11 5 1 22 8 4 0 2 14 7 6 1 6 2 972 3 3 7 13 0 5] - [ 4 1 5 0 4 1 6 2 1 0 0 5 10 2 0 969 14 13 1 4 4] - [ 5 11 0 0 10 4 1 0 2 1 0 1 3 3 3 7 992 2 1 3 12] - [ 4 2 0 6 2 3 3 1 0 1 1 11 24 1 3 17 1 963 2 3 3] - [ 2 2 9 20 3 1 1 21 4 1 7 3 4 0 12 1 0 1 991 0 3] - [ 0 3 1 0 3 5 9 10 1 0 0 13 4 2 0 7 7 0 4 1070 9] - [ 179 303 246 182 161 214 98 179 109 99 223 124 323 268 221 118 271 141 219 338 9418]] - -2023-02-13 17:47:58,517 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:47:58,517 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:47:58,523 - - -2023-02-13 17:47:58,523 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:47:59,501 - Epoch: [67][ 10/ 1207] Overall Loss 0.302024 Objective Loss 0.302024 LR 0.001000 Time 0.097716 -2023-02-13 17:47:59,699 - Epoch: [67][ 20/ 1207] Overall Loss 0.308031 Objective Loss 0.308031 LR 0.001000 Time 0.058728 -2023-02-13 17:47:59,887 - Epoch: [67][ 30/ 1207] Overall Loss 0.307546 Objective Loss 0.307546 LR 0.001000 Time 0.045419 -2023-02-13 17:48:00,075 - Epoch: [67][ 40/ 1207] Overall Loss 0.310993 Objective Loss 0.310993 LR 0.001000 Time 0.038765 -2023-02-13 17:48:00,277 - Epoch: [67][ 50/ 1207] Overall Loss 0.313686 Objective Loss 0.313686 LR 0.001000 Time 0.035030 -2023-02-13 17:48:00,472 - Epoch: [67][ 60/ 1207] Overall Loss 0.317441 Objective Loss 0.317441 LR 0.001000 Time 0.032443 -2023-02-13 17:48:00,665 - Epoch: [67][ 70/ 1207] Overall Loss 0.321183 Objective Loss 0.321183 LR 0.001000 Time 0.030555 -2023-02-13 17:48:00,861 - Epoch: [67][ 80/ 1207] Overall Loss 0.323078 Objective Loss 0.323078 LR 0.001000 Time 0.029184 -2023-02-13 17:48:01,055 - Epoch: [67][ 90/ 1207] Overall Loss 0.324888 Objective Loss 0.324888 LR 0.001000 Time 0.028087 -2023-02-13 17:48:01,250 - Epoch: [67][ 100/ 1207] Overall Loss 0.328192 Objective Loss 0.328192 LR 0.001000 Time 0.027232 -2023-02-13 17:48:01,442 - Epoch: [67][ 110/ 1207] Overall Loss 0.327375 Objective Loss 0.327375 LR 0.001000 Time 0.026500 -2023-02-13 17:48:01,638 - Epoch: [67][ 120/ 1207] Overall Loss 0.328956 Objective Loss 0.328956 LR 0.001000 Time 0.025921 -2023-02-13 17:48:01,831 - Epoch: [67][ 130/ 1207] Overall Loss 0.328677 Objective Loss 0.328677 LR 0.001000 Time 0.025409 -2023-02-13 17:48:02,027 - Epoch: [67][ 140/ 1207] Overall Loss 0.330803 Objective Loss 0.330803 LR 0.001000 Time 0.024991 -2023-02-13 17:48:02,220 - Epoch: [67][ 150/ 1207] Overall Loss 0.329757 Objective Loss 0.329757 LR 0.001000 Time 0.024607 -2023-02-13 17:48:02,415 - Epoch: [67][ 160/ 1207] Overall Loss 0.329959 Objective Loss 0.329959 LR 0.001000 Time 0.024286 -2023-02-13 17:48:02,607 - Epoch: [67][ 170/ 1207] Overall Loss 0.330152 Objective Loss 0.330152 LR 0.001000 Time 0.023985 -2023-02-13 17:48:02,803 - Epoch: [67][ 180/ 1207] Overall Loss 0.329535 Objective Loss 0.329535 LR 0.001000 Time 0.023738 -2023-02-13 17:48:02,996 - Epoch: [67][ 190/ 1207] Overall Loss 0.329883 Objective Loss 0.329883 LR 0.001000 Time 0.023501 -2023-02-13 17:48:03,192 - Epoch: [67][ 200/ 1207] Overall Loss 0.329826 Objective Loss 0.329826 LR 0.001000 Time 0.023305 -2023-02-13 17:48:03,384 - Epoch: [67][ 210/ 1207] Overall Loss 0.330028 Objective Loss 0.330028 LR 0.001000 Time 0.023109 -2023-02-13 17:48:03,579 - Epoch: [67][ 220/ 1207] Overall Loss 0.329158 Objective Loss 0.329158 LR 0.001000 Time 0.022944 -2023-02-13 17:48:03,772 - Epoch: [67][ 230/ 1207] Overall Loss 0.328763 Objective Loss 0.328763 LR 0.001000 Time 0.022783 -2023-02-13 17:48:03,968 - Epoch: [67][ 240/ 1207] Overall Loss 0.327705 Objective Loss 0.327705 LR 0.001000 Time 0.022649 -2023-02-13 17:48:04,161 - Epoch: [67][ 250/ 1207] Overall Loss 0.327698 Objective Loss 0.327698 LR 0.001000 Time 0.022514 -2023-02-13 17:48:04,357 - Epoch: [67][ 260/ 1207] Overall Loss 0.327476 Objective Loss 0.327476 LR 0.001000 Time 0.022399 -2023-02-13 17:48:04,549 - Epoch: [67][ 270/ 1207] Overall Loss 0.327777 Objective Loss 0.327777 LR 0.001000 Time 0.022280 -2023-02-13 17:48:04,745 - Epoch: [67][ 280/ 1207] Overall Loss 0.326908 Objective Loss 0.326908 LR 0.001000 Time 0.022181 -2023-02-13 17:48:04,938 - Epoch: [67][ 290/ 1207] Overall Loss 0.326207 Objective Loss 0.326207 LR 0.001000 Time 0.022082 -2023-02-13 17:48:05,134 - Epoch: [67][ 300/ 1207] Overall Loss 0.328633 Objective Loss 0.328633 LR 0.001000 Time 0.021998 -2023-02-13 17:48:05,326 - Epoch: [67][ 310/ 1207] Overall Loss 0.328956 Objective Loss 0.328956 LR 0.001000 Time 0.021907 -2023-02-13 17:48:05,522 - Epoch: [67][ 320/ 1207] Overall Loss 0.328927 Objective Loss 0.328927 LR 0.001000 Time 0.021833 -2023-02-13 17:48:05,714 - Epoch: [67][ 330/ 1207] Overall Loss 0.329560 Objective Loss 0.329560 LR 0.001000 Time 0.021754 -2023-02-13 17:48:05,911 - Epoch: [67][ 340/ 1207] Overall Loss 0.329772 Objective Loss 0.329772 LR 0.001000 Time 0.021691 -2023-02-13 17:48:06,104 - Epoch: [67][ 350/ 1207] Overall Loss 0.329720 Objective Loss 0.329720 LR 0.001000 Time 0.021623 -2023-02-13 17:48:06,301 - Epoch: [67][ 360/ 1207] Overall Loss 0.329381 Objective Loss 0.329381 LR 0.001000 Time 0.021567 -2023-02-13 17:48:06,493 - Epoch: [67][ 370/ 1207] Overall Loss 0.329133 Objective Loss 0.329133 LR 0.001000 Time 0.021504 -2023-02-13 17:48:06,689 - Epoch: [67][ 380/ 1207] Overall Loss 0.329247 Objective Loss 0.329247 LR 0.001000 Time 0.021452 -2023-02-13 17:48:06,882 - Epoch: [67][ 390/ 1207] Overall Loss 0.329265 Objective Loss 0.329265 LR 0.001000 Time 0.021395 -2023-02-13 17:48:07,078 - Epoch: [67][ 400/ 1207] Overall Loss 0.329273 Objective Loss 0.329273 LR 0.001000 Time 0.021348 -2023-02-13 17:48:07,270 - Epoch: [67][ 410/ 1207] Overall Loss 0.329332 Objective Loss 0.329332 LR 0.001000 Time 0.021297 -2023-02-13 17:48:07,466 - Epoch: [67][ 420/ 1207] Overall Loss 0.329461 Objective Loss 0.329461 LR 0.001000 Time 0.021254 -2023-02-13 17:48:07,660 - Epoch: [67][ 430/ 1207] Overall Loss 0.330134 Objective Loss 0.330134 LR 0.001000 Time 0.021210 -2023-02-13 17:48:07,856 - Epoch: [67][ 440/ 1207] Overall Loss 0.329663 Objective Loss 0.329663 LR 0.001000 Time 0.021173 -2023-02-13 17:48:08,049 - Epoch: [67][ 450/ 1207] Overall Loss 0.328666 Objective Loss 0.328666 LR 0.001000 Time 0.021130 -2023-02-13 17:48:08,245 - Epoch: [67][ 460/ 1207] Overall Loss 0.329211 Objective Loss 0.329211 LR 0.001000 Time 0.021098 -2023-02-13 17:48:08,438 - Epoch: [67][ 470/ 1207] Overall Loss 0.328688 Objective Loss 0.328688 LR 0.001000 Time 0.021058 -2023-02-13 17:48:08,633 - Epoch: [67][ 480/ 1207] Overall Loss 0.328473 Objective Loss 0.328473 LR 0.001000 Time 0.021025 -2023-02-13 17:48:08,827 - Epoch: [67][ 490/ 1207] Overall Loss 0.327973 Objective Loss 0.327973 LR 0.001000 Time 0.020990 -2023-02-13 17:48:09,022 - Epoch: [67][ 500/ 1207] Overall Loss 0.327925 Objective Loss 0.327925 LR 0.001000 Time 0.020960 -2023-02-13 17:48:09,215 - Epoch: [67][ 510/ 1207] Overall Loss 0.327592 Objective Loss 0.327592 LR 0.001000 Time 0.020927 -2023-02-13 17:48:09,410 - Epoch: [67][ 520/ 1207] Overall Loss 0.328046 Objective Loss 0.328046 LR 0.001000 Time 0.020899 -2023-02-13 17:48:09,603 - Epoch: [67][ 530/ 1207] Overall Loss 0.328179 Objective Loss 0.328179 LR 0.001000 Time 0.020868 -2023-02-13 17:48:09,800 - Epoch: [67][ 540/ 1207] Overall Loss 0.328203 Objective Loss 0.328203 LR 0.001000 Time 0.020845 -2023-02-13 17:48:09,992 - Epoch: [67][ 550/ 1207] Overall Loss 0.328922 Objective Loss 0.328922 LR 0.001000 Time 0.020815 -2023-02-13 17:48:10,189 - Epoch: [67][ 560/ 1207] Overall Loss 0.328554 Objective Loss 0.328554 LR 0.001000 Time 0.020793 -2023-02-13 17:48:10,381 - Epoch: [67][ 570/ 1207] Overall Loss 0.328559 Objective Loss 0.328559 LR 0.001000 Time 0.020765 -2023-02-13 17:48:10,576 - Epoch: [67][ 580/ 1207] Overall Loss 0.328988 Objective Loss 0.328988 LR 0.001000 Time 0.020744 -2023-02-13 17:48:10,769 - Epoch: [67][ 590/ 1207] Overall Loss 0.329167 Objective Loss 0.329167 LR 0.001000 Time 0.020718 -2023-02-13 17:48:10,968 - Epoch: [67][ 600/ 1207] Overall Loss 0.329227 Objective Loss 0.329227 LR 0.001000 Time 0.020703 -2023-02-13 17:48:11,161 - Epoch: [67][ 610/ 1207] Overall Loss 0.329394 Objective Loss 0.329394 LR 0.001000 Time 0.020680 -2023-02-13 17:48:11,356 - Epoch: [67][ 620/ 1207] Overall Loss 0.328746 Objective Loss 0.328746 LR 0.001000 Time 0.020660 -2023-02-13 17:48:11,549 - Epoch: [67][ 630/ 1207] Overall Loss 0.328867 Objective Loss 0.328867 LR 0.001000 Time 0.020638 -2023-02-13 17:48:11,744 - Epoch: [67][ 640/ 1207] Overall Loss 0.328823 Objective Loss 0.328823 LR 0.001000 Time 0.020621 -2023-02-13 17:48:11,938 - Epoch: [67][ 650/ 1207] Overall Loss 0.329485 Objective Loss 0.329485 LR 0.001000 Time 0.020601 -2023-02-13 17:48:12,134 - Epoch: [67][ 660/ 1207] Overall Loss 0.329618 Objective Loss 0.329618 LR 0.001000 Time 0.020585 -2023-02-13 17:48:12,327 - Epoch: [67][ 670/ 1207] Overall Loss 0.329452 Objective Loss 0.329452 LR 0.001000 Time 0.020565 -2023-02-13 17:48:12,522 - Epoch: [67][ 680/ 1207] Overall Loss 0.330256 Objective Loss 0.330256 LR 0.001000 Time 0.020549 -2023-02-13 17:48:12,715 - Epoch: [67][ 690/ 1207] Overall Loss 0.330435 Objective Loss 0.330435 LR 0.001000 Time 0.020530 -2023-02-13 17:48:12,912 - Epoch: [67][ 700/ 1207] Overall Loss 0.330046 Objective Loss 0.330046 LR 0.001000 Time 0.020518 -2023-02-13 17:48:13,106 - Epoch: [67][ 710/ 1207] Overall Loss 0.329810 Objective Loss 0.329810 LR 0.001000 Time 0.020501 -2023-02-13 17:48:13,303 - Epoch: [67][ 720/ 1207] Overall Loss 0.330094 Objective Loss 0.330094 LR 0.001000 Time 0.020489 -2023-02-13 17:48:13,496 - Epoch: [67][ 730/ 1207] Overall Loss 0.330194 Objective Loss 0.330194 LR 0.001000 Time 0.020473 -2023-02-13 17:48:13,691 - Epoch: [67][ 740/ 1207] Overall Loss 0.330463 Objective Loss 0.330463 LR 0.001000 Time 0.020460 -2023-02-13 17:48:13,885 - Epoch: [67][ 750/ 1207] Overall Loss 0.330953 Objective Loss 0.330953 LR 0.001000 Time 0.020444 -2023-02-13 17:48:14,081 - Epoch: [67][ 760/ 1207] Overall Loss 0.331243 Objective Loss 0.331243 LR 0.001000 Time 0.020433 -2023-02-13 17:48:14,274 - Epoch: [67][ 770/ 1207] Overall Loss 0.331346 Objective Loss 0.331346 LR 0.001000 Time 0.020418 -2023-02-13 17:48:14,470 - Epoch: [67][ 780/ 1207] Overall Loss 0.331425 Objective Loss 0.331425 LR 0.001000 Time 0.020407 -2023-02-13 17:48:14,664 - Epoch: [67][ 790/ 1207] Overall Loss 0.331541 Objective Loss 0.331541 LR 0.001000 Time 0.020394 -2023-02-13 17:48:14,860 - Epoch: [67][ 800/ 1207] Overall Loss 0.331981 Objective Loss 0.331981 LR 0.001000 Time 0.020384 -2023-02-13 17:48:15,054 - Epoch: [67][ 810/ 1207] Overall Loss 0.332114 Objective Loss 0.332114 LR 0.001000 Time 0.020371 -2023-02-13 17:48:15,251 - Epoch: [67][ 820/ 1207] Overall Loss 0.332627 Objective Loss 0.332627 LR 0.001000 Time 0.020362 -2023-02-13 17:48:15,444 - Epoch: [67][ 830/ 1207] Overall Loss 0.332917 Objective Loss 0.332917 LR 0.001000 Time 0.020349 -2023-02-13 17:48:15,641 - Epoch: [67][ 840/ 1207] Overall Loss 0.332993 Objective Loss 0.332993 LR 0.001000 Time 0.020341 -2023-02-13 17:48:15,835 - Epoch: [67][ 850/ 1207] Overall Loss 0.332971 Objective Loss 0.332971 LR 0.001000 Time 0.020329 -2023-02-13 17:48:16,031 - Epoch: [67][ 860/ 1207] Overall Loss 0.332763 Objective Loss 0.332763 LR 0.001000 Time 0.020321 -2023-02-13 17:48:16,225 - Epoch: [67][ 870/ 1207] Overall Loss 0.332949 Objective Loss 0.332949 LR 0.001000 Time 0.020310 -2023-02-13 17:48:16,421 - Epoch: [67][ 880/ 1207] Overall Loss 0.332695 Objective Loss 0.332695 LR 0.001000 Time 0.020301 -2023-02-13 17:48:16,614 - Epoch: [67][ 890/ 1207] Overall Loss 0.332621 Objective Loss 0.332621 LR 0.001000 Time 0.020289 -2023-02-13 17:48:16,810 - Epoch: [67][ 900/ 1207] Overall Loss 0.332869 Objective Loss 0.332869 LR 0.001000 Time 0.020281 -2023-02-13 17:48:17,004 - Epoch: [67][ 910/ 1207] Overall Loss 0.333071 Objective Loss 0.333071 LR 0.001000 Time 0.020271 -2023-02-13 17:48:17,201 - Epoch: [67][ 920/ 1207] Overall Loss 0.333475 Objective Loss 0.333475 LR 0.001000 Time 0.020265 -2023-02-13 17:48:17,394 - Epoch: [67][ 930/ 1207] Overall Loss 0.333482 Objective Loss 0.333482 LR 0.001000 Time 0.020254 -2023-02-13 17:48:17,590 - Epoch: [67][ 940/ 1207] Overall Loss 0.333258 Objective Loss 0.333258 LR 0.001000 Time 0.020246 -2023-02-13 17:48:17,783 - Epoch: [67][ 950/ 1207] Overall Loss 0.333232 Objective Loss 0.333232 LR 0.001000 Time 0.020236 -2023-02-13 17:48:17,980 - Epoch: [67][ 960/ 1207] Overall Loss 0.333008 Objective Loss 0.333008 LR 0.001000 Time 0.020230 -2023-02-13 17:48:18,173 - Epoch: [67][ 970/ 1207] Overall Loss 0.333171 Objective Loss 0.333171 LR 0.001000 Time 0.020220 -2023-02-13 17:48:18,369 - Epoch: [67][ 980/ 1207] Overall Loss 0.333168 Objective Loss 0.333168 LR 0.001000 Time 0.020213 -2023-02-13 17:48:18,562 - Epoch: [67][ 990/ 1207] Overall Loss 0.332977 Objective Loss 0.332977 LR 0.001000 Time 0.020203 -2023-02-13 17:48:18,758 - Epoch: [67][ 1000/ 1207] Overall Loss 0.332664 Objective Loss 0.332664 LR 0.001000 Time 0.020197 -2023-02-13 17:48:18,951 - Epoch: [67][ 1010/ 1207] Overall Loss 0.333257 Objective Loss 0.333257 LR 0.001000 Time 0.020188 -2023-02-13 17:48:19,148 - Epoch: [67][ 1020/ 1207] Overall Loss 0.333384 Objective Loss 0.333384 LR 0.001000 Time 0.020182 -2023-02-13 17:48:19,341 - Epoch: [67][ 1030/ 1207] Overall Loss 0.333278 Objective Loss 0.333278 LR 0.001000 Time 0.020174 -2023-02-13 17:48:19,537 - Epoch: [67][ 1040/ 1207] Overall Loss 0.333348 Objective Loss 0.333348 LR 0.001000 Time 0.020168 -2023-02-13 17:48:19,729 - Epoch: [67][ 1050/ 1207] Overall Loss 0.333549 Objective Loss 0.333549 LR 0.001000 Time 0.020159 -2023-02-13 17:48:19,926 - Epoch: [67][ 1060/ 1207] Overall Loss 0.333537 Objective Loss 0.333537 LR 0.001000 Time 0.020154 -2023-02-13 17:48:20,119 - Epoch: [67][ 1070/ 1207] Overall Loss 0.333552 Objective Loss 0.333552 LR 0.001000 Time 0.020146 -2023-02-13 17:48:20,315 - Epoch: [67][ 1080/ 1207] Overall Loss 0.333585 Objective Loss 0.333585 LR 0.001000 Time 0.020140 -2023-02-13 17:48:20,508 - Epoch: [67][ 1090/ 1207] Overall Loss 0.333728 Objective Loss 0.333728 LR 0.001000 Time 0.020132 -2023-02-13 17:48:20,705 - Epoch: [67][ 1100/ 1207] Overall Loss 0.333861 Objective Loss 0.333861 LR 0.001000 Time 0.020127 -2023-02-13 17:48:20,899 - Epoch: [67][ 1110/ 1207] Overall Loss 0.333615 Objective Loss 0.333615 LR 0.001000 Time 0.020121 -2023-02-13 17:48:21,094 - Epoch: [67][ 1120/ 1207] Overall Loss 0.333646 Objective Loss 0.333646 LR 0.001000 Time 0.020115 -2023-02-13 17:48:21,288 - Epoch: [67][ 1130/ 1207] Overall Loss 0.333838 Objective Loss 0.333838 LR 0.001000 Time 0.020108 -2023-02-13 17:48:21,484 - Epoch: [67][ 1140/ 1207] Overall Loss 0.334009 Objective Loss 0.334009 LR 0.001000 Time 0.020103 -2023-02-13 17:48:21,677 - Epoch: [67][ 1150/ 1207] Overall Loss 0.334077 Objective Loss 0.334077 LR 0.001000 Time 0.020096 -2023-02-13 17:48:21,874 - Epoch: [67][ 1160/ 1207] Overall Loss 0.334291 Objective Loss 0.334291 LR 0.001000 Time 0.020093 -2023-02-13 17:48:22,067 - Epoch: [67][ 1170/ 1207] Overall Loss 0.334255 Objective Loss 0.334255 LR 0.001000 Time 0.020085 -2023-02-13 17:48:22,264 - Epoch: [67][ 1180/ 1207] Overall Loss 0.334220 Objective Loss 0.334220 LR 0.001000 Time 0.020082 -2023-02-13 17:48:22,457 - Epoch: [67][ 1190/ 1207] Overall Loss 0.334186 Objective Loss 0.334186 LR 0.001000 Time 0.020075 -2023-02-13 17:48:22,703 - Epoch: [67][ 1200/ 1207] Overall Loss 0.334158 Objective Loss 0.334158 LR 0.001000 Time 0.020112 -2023-02-13 17:48:22,818 - Epoch: [67][ 1207/ 1207] Overall Loss 0.334239 Objective Loss 0.334239 Top1 81.402439 Top5 96.951220 LR 0.001000 Time 0.020090 -2023-02-13 17:48:22,899 - --- validate (epoch=67)----------- -2023-02-13 17:48:22,899 - 34311 samples (256 per mini-batch) -2023-02-13 17:48:23,299 - Epoch: [67][ 10/ 135] Loss 0.362224 Top1 81.406250 Top5 97.421875 -2023-02-13 17:48:23,435 - Epoch: [67][ 20/ 135] Loss 0.374264 Top1 81.542969 Top5 96.855469 -2023-02-13 17:48:23,563 - Epoch: [67][ 30/ 135] Loss 0.373011 Top1 80.937500 Top5 96.757812 -2023-02-13 17:48:23,689 - Epoch: [67][ 40/ 135] Loss 0.374053 Top1 80.869141 Top5 96.845703 -2023-02-13 17:48:23,817 - Epoch: [67][ 50/ 135] Loss 0.372859 Top1 80.960938 Top5 96.804688 -2023-02-13 17:48:23,945 - Epoch: [67][ 60/ 135] Loss 0.374873 Top1 80.696615 Top5 96.751302 -2023-02-13 17:48:24,076 - Epoch: [67][ 70/ 135] Loss 0.369384 Top1 80.825893 Top5 96.696429 -2023-02-13 17:48:24,207 - Epoch: [67][ 80/ 135] Loss 0.370187 Top1 80.834961 Top5 96.718750 -2023-02-13 17:48:24,338 - Epoch: [67][ 90/ 135] Loss 0.367392 Top1 80.733507 Top5 96.740451 -2023-02-13 17:48:24,470 - Epoch: [67][ 100/ 135] Loss 0.362184 Top1 80.785156 Top5 96.820312 -2023-02-13 17:48:24,601 - Epoch: [67][ 110/ 135] Loss 0.363141 Top1 80.838068 Top5 96.793324 -2023-02-13 17:48:24,733 - Epoch: [67][ 120/ 135] Loss 0.364775 Top1 80.774740 Top5 96.780599 -2023-02-13 17:48:24,868 - Epoch: [67][ 130/ 135] Loss 0.364038 Top1 80.805288 Top5 96.808894 -2023-02-13 17:48:24,914 - Epoch: [67][ 135/ 135] Loss 0.363024 Top1 80.828306 Top5 96.826091 -2023-02-13 17:48:24,985 - ==> Top1: 80.828 Top5: 96.826 Loss: 0.363 - -2023-02-13 17:48:24,986 - ==> Confusion: -[[ 847 7 8 2 7 2 0 0 6 50 0 8 2 5 6 3 4 2 2 1 5] - [ 2 938 3 2 7 24 1 17 5 2 2 4 2 1 1 2 4 1 6 1 8] - [ 7 3 959 11 5 0 12 9 1 1 2 1 2 3 3 11 5 3 17 2 1] - [ 5 1 26 895 2 4 1 2 1 2 6 0 10 3 23 2 3 9 15 0 6] - [ 16 7 5 0 969 13 0 3 0 4 0 6 1 2 9 10 10 1 1 1 8] - [ 3 28 4 7 7 919 7 32 1 3 2 19 11 13 2 1 1 1 1 6 2] - [ 2 3 29 4 1 5 1016 4 0 1 4 4 1 0 0 6 1 6 3 5 4] - [ 2 7 16 3 2 20 5 924 0 1 0 10 1 0 0 1 0 4 19 5 4] - [ 20 3 1 2 2 1 0 0 872 38 12 4 1 11 29 2 0 3 7 1 0] - [ 85 4 4 0 7 1 0 2 34 836 2 1 0 16 6 3 0 5 1 2 3] - [ 2 4 11 11 3 2 0 2 17 3 942 2 1 21 4 0 1 1 19 0 5] - [ 4 3 0 0 2 8 0 8 0 1 0 920 24 4 1 7 2 15 0 4 2] - [ 3 0 0 10 2 0 0 2 1 0 0 54 837 0 5 7 3 24 1 1 9] - [ 6 5 2 0 6 12 1 0 6 22 8 9 2 918 8 8 6 1 2 0 2] - [ 10 3 1 22 5 2 0 2 14 5 3 4 2 2 990 2 1 8 7 0 9] - [ 6 1 7 0 6 1 5 2 0 0 0 7 5 2 0 970 12 10 0 7 5] - [ 2 5 2 1 8 4 0 0 3 0 1 4 1 2 2 13 999 5 0 2 7] - [ 3 3 0 3 2 3 0 0 0 1 1 13 13 1 3 16 1 985 0 1 2] - [ 2 6 7 12 3 1 0 25 1 0 3 2 7 0 16 1 1 1 993 1 4] - [ 0 5 6 0 1 6 7 21 0 0 1 33 4 2 0 6 6 4 1 1038 7] - [ 204 286 412 197 134 214 94 215 95 92 197 192 376 322 243 185 377 162 226 245 8966]] - -2023-02-13 17:48:24,987 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:48:24,987 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:48:24,993 - - -2023-02-13 17:48:24,993 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:48:25,900 - Epoch: [68][ 10/ 1207] Overall Loss 0.338865 Objective Loss 0.338865 LR 0.001000 Time 0.090625 -2023-02-13 17:48:26,101 - Epoch: [68][ 20/ 1207] Overall Loss 0.334560 Objective Loss 0.334560 LR 0.001000 Time 0.055355 -2023-02-13 17:48:26,304 - Epoch: [68][ 30/ 1207] Overall Loss 0.336854 Objective Loss 0.336854 LR 0.001000 Time 0.043645 -2023-02-13 17:48:26,498 - Epoch: [68][ 40/ 1207] Overall Loss 0.331028 Objective Loss 0.331028 LR 0.001000 Time 0.037586 -2023-02-13 17:48:26,695 - Epoch: [68][ 50/ 1207] Overall Loss 0.328013 Objective Loss 0.328013 LR 0.001000 Time 0.033999 -2023-02-13 17:48:26,889 - Epoch: [68][ 60/ 1207] Overall Loss 0.322275 Objective Loss 0.322275 LR 0.001000 Time 0.031554 -2023-02-13 17:48:27,085 - Epoch: [68][ 70/ 1207] Overall Loss 0.323350 Objective Loss 0.323350 LR 0.001000 Time 0.029838 -2023-02-13 17:48:27,278 - Epoch: [68][ 80/ 1207] Overall Loss 0.325805 Objective Loss 0.325805 LR 0.001000 Time 0.028523 -2023-02-13 17:48:27,475 - Epoch: [68][ 90/ 1207] Overall Loss 0.326137 Objective Loss 0.326137 LR 0.001000 Time 0.027531 -2023-02-13 17:48:27,668 - Epoch: [68][ 100/ 1207] Overall Loss 0.330932 Objective Loss 0.330932 LR 0.001000 Time 0.026705 -2023-02-13 17:48:27,864 - Epoch: [68][ 110/ 1207] Overall Loss 0.331965 Objective Loss 0.331965 LR 0.001000 Time 0.026062 -2023-02-13 17:48:28,058 - Epoch: [68][ 120/ 1207] Overall Loss 0.333368 Objective Loss 0.333368 LR 0.001000 Time 0.025499 -2023-02-13 17:48:28,255 - Epoch: [68][ 130/ 1207] Overall Loss 0.333242 Objective Loss 0.333242 LR 0.001000 Time 0.025048 -2023-02-13 17:48:28,449 - Epoch: [68][ 140/ 1207] Overall Loss 0.335250 Objective Loss 0.335250 LR 0.001000 Time 0.024643 -2023-02-13 17:48:28,645 - Epoch: [68][ 150/ 1207] Overall Loss 0.332176 Objective Loss 0.332176 LR 0.001000 Time 0.024307 -2023-02-13 17:48:28,839 - Epoch: [68][ 160/ 1207] Overall Loss 0.332104 Objective Loss 0.332104 LR 0.001000 Time 0.023996 -2023-02-13 17:48:29,035 - Epoch: [68][ 170/ 1207] Overall Loss 0.332830 Objective Loss 0.332830 LR 0.001000 Time 0.023737 -2023-02-13 17:48:29,229 - Epoch: [68][ 180/ 1207] Overall Loss 0.332440 Objective Loss 0.332440 LR 0.001000 Time 0.023492 -2023-02-13 17:48:29,425 - Epoch: [68][ 190/ 1207] Overall Loss 0.331108 Objective Loss 0.331108 LR 0.001000 Time 0.023287 -2023-02-13 17:48:29,618 - Epoch: [68][ 200/ 1207] Overall Loss 0.330542 Objective Loss 0.330542 LR 0.001000 Time 0.023084 -2023-02-13 17:48:29,811 - Epoch: [68][ 210/ 1207] Overall Loss 0.331195 Objective Loss 0.331195 LR 0.001000 Time 0.022905 -2023-02-13 17:48:30,000 - Epoch: [68][ 220/ 1207] Overall Loss 0.332093 Objective Loss 0.332093 LR 0.001000 Time 0.022719 -2023-02-13 17:48:30,189 - Epoch: [68][ 230/ 1207] Overall Loss 0.332134 Objective Loss 0.332134 LR 0.001000 Time 0.022551 -2023-02-13 17:48:30,378 - Epoch: [68][ 240/ 1207] Overall Loss 0.332472 Objective Loss 0.332472 LR 0.001000 Time 0.022400 -2023-02-13 17:48:30,567 - Epoch: [68][ 250/ 1207] Overall Loss 0.332489 Objective Loss 0.332489 LR 0.001000 Time 0.022258 -2023-02-13 17:48:30,756 - Epoch: [68][ 260/ 1207] Overall Loss 0.331407 Objective Loss 0.331407 LR 0.001000 Time 0.022127 -2023-02-13 17:48:30,947 - Epoch: [68][ 270/ 1207] Overall Loss 0.329812 Objective Loss 0.329812 LR 0.001000 Time 0.022012 -2023-02-13 17:48:31,135 - Epoch: [68][ 280/ 1207] Overall Loss 0.330365 Objective Loss 0.330365 LR 0.001000 Time 0.021898 -2023-02-13 17:48:31,325 - Epoch: [68][ 290/ 1207] Overall Loss 0.329462 Objective Loss 0.329462 LR 0.001000 Time 0.021795 -2023-02-13 17:48:31,514 - Epoch: [68][ 300/ 1207] Overall Loss 0.329614 Objective Loss 0.329614 LR 0.001000 Time 0.021698 -2023-02-13 17:48:31,703 - Epoch: [68][ 310/ 1207] Overall Loss 0.330035 Objective Loss 0.330035 LR 0.001000 Time 0.021605 -2023-02-13 17:48:31,891 - Epoch: [68][ 320/ 1207] Overall Loss 0.330100 Objective Loss 0.330100 LR 0.001000 Time 0.021518 -2023-02-13 17:48:32,080 - Epoch: [68][ 330/ 1207] Overall Loss 0.329418 Objective Loss 0.329418 LR 0.001000 Time 0.021438 -2023-02-13 17:48:32,269 - Epoch: [68][ 340/ 1207] Overall Loss 0.329140 Objective Loss 0.329140 LR 0.001000 Time 0.021362 -2023-02-13 17:48:32,458 - Epoch: [68][ 350/ 1207] Overall Loss 0.329069 Objective Loss 0.329069 LR 0.001000 Time 0.021291 -2023-02-13 17:48:32,646 - Epoch: [68][ 360/ 1207] Overall Loss 0.329988 Objective Loss 0.329988 LR 0.001000 Time 0.021221 -2023-02-13 17:48:32,834 - Epoch: [68][ 370/ 1207] Overall Loss 0.330114 Objective Loss 0.330114 LR 0.001000 Time 0.021155 -2023-02-13 17:48:33,023 - Epoch: [68][ 380/ 1207] Overall Loss 0.329844 Objective Loss 0.329844 LR 0.001000 Time 0.021093 -2023-02-13 17:48:33,212 - Epoch: [68][ 390/ 1207] Overall Loss 0.330205 Objective Loss 0.330205 LR 0.001000 Time 0.021036 -2023-02-13 17:48:33,401 - Epoch: [68][ 400/ 1207] Overall Loss 0.330047 Objective Loss 0.330047 LR 0.001000 Time 0.020981 -2023-02-13 17:48:33,589 - Epoch: [68][ 410/ 1207] Overall Loss 0.329750 Objective Loss 0.329750 LR 0.001000 Time 0.020927 -2023-02-13 17:48:33,778 - Epoch: [68][ 420/ 1207] Overall Loss 0.329686 Objective Loss 0.329686 LR 0.001000 Time 0.020879 -2023-02-13 17:48:33,969 - Epoch: [68][ 430/ 1207] Overall Loss 0.330218 Objective Loss 0.330218 LR 0.001000 Time 0.020837 -2023-02-13 17:48:34,160 - Epoch: [68][ 440/ 1207] Overall Loss 0.330560 Objective Loss 0.330560 LR 0.001000 Time 0.020796 -2023-02-13 17:48:34,352 - Epoch: [68][ 450/ 1207] Overall Loss 0.330349 Objective Loss 0.330349 LR 0.001000 Time 0.020761 -2023-02-13 17:48:34,543 - Epoch: [68][ 460/ 1207] Overall Loss 0.330383 Objective Loss 0.330383 LR 0.001000 Time 0.020723 -2023-02-13 17:48:34,732 - Epoch: [68][ 470/ 1207] Overall Loss 0.330277 Objective Loss 0.330277 LR 0.001000 Time 0.020684 -2023-02-13 17:48:34,921 - Epoch: [68][ 480/ 1207] Overall Loss 0.330229 Objective Loss 0.330229 LR 0.001000 Time 0.020646 -2023-02-13 17:48:35,110 - Epoch: [68][ 490/ 1207] Overall Loss 0.330707 Objective Loss 0.330707 LR 0.001000 Time 0.020610 -2023-02-13 17:48:35,301 - Epoch: [68][ 500/ 1207] Overall Loss 0.332164 Objective Loss 0.332164 LR 0.001000 Time 0.020577 -2023-02-13 17:48:35,490 - Epoch: [68][ 510/ 1207] Overall Loss 0.332407 Objective Loss 0.332407 LR 0.001000 Time 0.020544 -2023-02-13 17:48:35,679 - Epoch: [68][ 520/ 1207] Overall Loss 0.331404 Objective Loss 0.331404 LR 0.001000 Time 0.020512 -2023-02-13 17:48:35,869 - Epoch: [68][ 530/ 1207] Overall Loss 0.331242 Objective Loss 0.331242 LR 0.001000 Time 0.020483 -2023-02-13 17:48:36,059 - Epoch: [68][ 540/ 1207] Overall Loss 0.331535 Objective Loss 0.331535 LR 0.001000 Time 0.020454 -2023-02-13 17:48:36,250 - Epoch: [68][ 550/ 1207] Overall Loss 0.331401 Objective Loss 0.331401 LR 0.001000 Time 0.020429 -2023-02-13 17:48:36,439 - Epoch: [68][ 560/ 1207] Overall Loss 0.331307 Objective Loss 0.331307 LR 0.001000 Time 0.020402 -2023-02-13 17:48:36,629 - Epoch: [68][ 570/ 1207] Overall Loss 0.330958 Objective Loss 0.330958 LR 0.001000 Time 0.020375 -2023-02-13 17:48:36,818 - Epoch: [68][ 580/ 1207] Overall Loss 0.330951 Objective Loss 0.330951 LR 0.001000 Time 0.020350 -2023-02-13 17:48:37,008 - Epoch: [68][ 590/ 1207] Overall Loss 0.330495 Objective Loss 0.330495 LR 0.001000 Time 0.020326 -2023-02-13 17:48:37,197 - Epoch: [68][ 600/ 1207] Overall Loss 0.330999 Objective Loss 0.330999 LR 0.001000 Time 0.020302 -2023-02-13 17:48:37,387 - Epoch: [68][ 610/ 1207] Overall Loss 0.330796 Objective Loss 0.330796 LR 0.001000 Time 0.020280 -2023-02-13 17:48:37,577 - Epoch: [68][ 620/ 1207] Overall Loss 0.330898 Objective Loss 0.330898 LR 0.001000 Time 0.020258 -2023-02-13 17:48:37,766 - Epoch: [68][ 630/ 1207] Overall Loss 0.330997 Objective Loss 0.330997 LR 0.001000 Time 0.020237 -2023-02-13 17:48:37,955 - Epoch: [68][ 640/ 1207] Overall Loss 0.331512 Objective Loss 0.331512 LR 0.001000 Time 0.020216 -2023-02-13 17:48:38,145 - Epoch: [68][ 650/ 1207] Overall Loss 0.331604 Objective Loss 0.331604 LR 0.001000 Time 0.020196 -2023-02-13 17:48:38,335 - Epoch: [68][ 660/ 1207] Overall Loss 0.331746 Objective Loss 0.331746 LR 0.001000 Time 0.020177 -2023-02-13 17:48:38,524 - Epoch: [68][ 670/ 1207] Overall Loss 0.331846 Objective Loss 0.331846 LR 0.001000 Time 0.020158 -2023-02-13 17:48:38,714 - Epoch: [68][ 680/ 1207] Overall Loss 0.332298 Objective Loss 0.332298 LR 0.001000 Time 0.020141 -2023-02-13 17:48:38,904 - Epoch: [68][ 690/ 1207] Overall Loss 0.332442 Objective Loss 0.332442 LR 0.001000 Time 0.020122 -2023-02-13 17:48:39,093 - Epoch: [68][ 700/ 1207] Overall Loss 0.332113 Objective Loss 0.332113 LR 0.001000 Time 0.020105 -2023-02-13 17:48:39,284 - Epoch: [68][ 710/ 1207] Overall Loss 0.331771 Objective Loss 0.331771 LR 0.001000 Time 0.020090 -2023-02-13 17:48:39,474 - Epoch: [68][ 720/ 1207] Overall Loss 0.331965 Objective Loss 0.331965 LR 0.001000 Time 0.020075 -2023-02-13 17:48:39,664 - Epoch: [68][ 730/ 1207] Overall Loss 0.331954 Objective Loss 0.331954 LR 0.001000 Time 0.020059 -2023-02-13 17:48:39,854 - Epoch: [68][ 740/ 1207] Overall Loss 0.331821 Objective Loss 0.331821 LR 0.001000 Time 0.020045 -2023-02-13 17:48:40,044 - Epoch: [68][ 750/ 1207] Overall Loss 0.331625 Objective Loss 0.331625 LR 0.001000 Time 0.020030 -2023-02-13 17:48:40,234 - Epoch: [68][ 760/ 1207] Overall Loss 0.331546 Objective Loss 0.331546 LR 0.001000 Time 0.020016 -2023-02-13 17:48:40,424 - Epoch: [68][ 770/ 1207] Overall Loss 0.331879 Objective Loss 0.331879 LR 0.001000 Time 0.020002 -2023-02-13 17:48:40,613 - Epoch: [68][ 780/ 1207] Overall Loss 0.332291 Objective Loss 0.332291 LR 0.001000 Time 0.019988 -2023-02-13 17:48:40,803 - Epoch: [68][ 790/ 1207] Overall Loss 0.332196 Objective Loss 0.332196 LR 0.001000 Time 0.019974 -2023-02-13 17:48:40,992 - Epoch: [68][ 800/ 1207] Overall Loss 0.332222 Objective Loss 0.332222 LR 0.001000 Time 0.019961 -2023-02-13 17:48:41,182 - Epoch: [68][ 810/ 1207] Overall Loss 0.332666 Objective Loss 0.332666 LR 0.001000 Time 0.019948 -2023-02-13 17:48:41,371 - Epoch: [68][ 820/ 1207] Overall Loss 0.332366 Objective Loss 0.332366 LR 0.001000 Time 0.019935 -2023-02-13 17:48:41,561 - Epoch: [68][ 830/ 1207] Overall Loss 0.333036 Objective Loss 0.333036 LR 0.001000 Time 0.019923 -2023-02-13 17:48:41,751 - Epoch: [68][ 840/ 1207] Overall Loss 0.332852 Objective Loss 0.332852 LR 0.001000 Time 0.019912 -2023-02-13 17:48:41,941 - Epoch: [68][ 850/ 1207] Overall Loss 0.332635 Objective Loss 0.332635 LR 0.001000 Time 0.019900 -2023-02-13 17:48:42,131 - Epoch: [68][ 860/ 1207] Overall Loss 0.332352 Objective Loss 0.332352 LR 0.001000 Time 0.019890 -2023-02-13 17:48:42,321 - Epoch: [68][ 870/ 1207] Overall Loss 0.332469 Objective Loss 0.332469 LR 0.001000 Time 0.019879 -2023-02-13 17:48:42,509 - Epoch: [68][ 880/ 1207] Overall Loss 0.332594 Objective Loss 0.332594 LR 0.001000 Time 0.019867 -2023-02-13 17:48:42,699 - Epoch: [68][ 890/ 1207] Overall Loss 0.332968 Objective Loss 0.332968 LR 0.001000 Time 0.019857 -2023-02-13 17:48:42,890 - Epoch: [68][ 900/ 1207] Overall Loss 0.333059 Objective Loss 0.333059 LR 0.001000 Time 0.019847 -2023-02-13 17:48:43,079 - Epoch: [68][ 910/ 1207] Overall Loss 0.333119 Objective Loss 0.333119 LR 0.001000 Time 0.019837 -2023-02-13 17:48:43,269 - Epoch: [68][ 920/ 1207] Overall Loss 0.333057 Objective Loss 0.333057 LR 0.001000 Time 0.019828 -2023-02-13 17:48:43,459 - Epoch: [68][ 930/ 1207] Overall Loss 0.332943 Objective Loss 0.332943 LR 0.001000 Time 0.019818 -2023-02-13 17:48:43,649 - Epoch: [68][ 940/ 1207] Overall Loss 0.332940 Objective Loss 0.332940 LR 0.001000 Time 0.019809 -2023-02-13 17:48:43,839 - Epoch: [68][ 950/ 1207] Overall Loss 0.333085 Objective Loss 0.333085 LR 0.001000 Time 0.019800 -2023-02-13 17:48:44,029 - Epoch: [68][ 960/ 1207] Overall Loss 0.332918 Objective Loss 0.332918 LR 0.001000 Time 0.019791 -2023-02-13 17:48:44,218 - Epoch: [68][ 970/ 1207] Overall Loss 0.333096 Objective Loss 0.333096 LR 0.001000 Time 0.019782 -2023-02-13 17:48:44,408 - Epoch: [68][ 980/ 1207] Overall Loss 0.333244 Objective Loss 0.333244 LR 0.001000 Time 0.019773 -2023-02-13 17:48:44,599 - Epoch: [68][ 990/ 1207] Overall Loss 0.333291 Objective Loss 0.333291 LR 0.001000 Time 0.019766 -2023-02-13 17:48:44,788 - Epoch: [68][ 1000/ 1207] Overall Loss 0.333207 Objective Loss 0.333207 LR 0.001000 Time 0.019757 -2023-02-13 17:48:44,978 - Epoch: [68][ 1010/ 1207] Overall Loss 0.333329 Objective Loss 0.333329 LR 0.001000 Time 0.019749 -2023-02-13 17:48:45,167 - Epoch: [68][ 1020/ 1207] Overall Loss 0.333177 Objective Loss 0.333177 LR 0.001000 Time 0.019741 -2023-02-13 17:48:45,357 - Epoch: [68][ 1030/ 1207] Overall Loss 0.332942 Objective Loss 0.332942 LR 0.001000 Time 0.019733 -2023-02-13 17:48:45,547 - Epoch: [68][ 1040/ 1207] Overall Loss 0.332984 Objective Loss 0.332984 LR 0.001000 Time 0.019726 -2023-02-13 17:48:45,737 - Epoch: [68][ 1050/ 1207] Overall Loss 0.332970 Objective Loss 0.332970 LR 0.001000 Time 0.019719 -2023-02-13 17:48:45,929 - Epoch: [68][ 1060/ 1207] Overall Loss 0.333065 Objective Loss 0.333065 LR 0.001000 Time 0.019713 -2023-02-13 17:48:46,118 - Epoch: [68][ 1070/ 1207] Overall Loss 0.333220 Objective Loss 0.333220 LR 0.001000 Time 0.019706 -2023-02-13 17:48:46,309 - Epoch: [68][ 1080/ 1207] Overall Loss 0.333134 Objective Loss 0.333134 LR 0.001000 Time 0.019699 -2023-02-13 17:48:46,499 - Epoch: [68][ 1090/ 1207] Overall Loss 0.333131 Objective Loss 0.333131 LR 0.001000 Time 0.019692 -2023-02-13 17:48:46,688 - Epoch: [68][ 1100/ 1207] Overall Loss 0.333421 Objective Loss 0.333421 LR 0.001000 Time 0.019685 -2023-02-13 17:48:46,878 - Epoch: [68][ 1110/ 1207] Overall Loss 0.333944 Objective Loss 0.333944 LR 0.001000 Time 0.019678 -2023-02-13 17:48:47,068 - Epoch: [68][ 1120/ 1207] Overall Loss 0.334062 Objective Loss 0.334062 LR 0.001000 Time 0.019672 -2023-02-13 17:48:47,259 - Epoch: [68][ 1130/ 1207] Overall Loss 0.333904 Objective Loss 0.333904 LR 0.001000 Time 0.019666 -2023-02-13 17:48:47,448 - Epoch: [68][ 1140/ 1207] Overall Loss 0.334010 Objective Loss 0.334010 LR 0.001000 Time 0.019660 -2023-02-13 17:48:47,638 - Epoch: [68][ 1150/ 1207] Overall Loss 0.334112 Objective Loss 0.334112 LR 0.001000 Time 0.019654 -2023-02-13 17:48:47,828 - Epoch: [68][ 1160/ 1207] Overall Loss 0.334073 Objective Loss 0.334073 LR 0.001000 Time 0.019648 -2023-02-13 17:48:48,018 - Epoch: [68][ 1170/ 1207] Overall Loss 0.334163 Objective Loss 0.334163 LR 0.001000 Time 0.019642 -2023-02-13 17:48:48,208 - Epoch: [68][ 1180/ 1207] Overall Loss 0.334275 Objective Loss 0.334275 LR 0.001000 Time 0.019636 -2023-02-13 17:48:48,398 - Epoch: [68][ 1190/ 1207] Overall Loss 0.334190 Objective Loss 0.334190 LR 0.001000 Time 0.019630 -2023-02-13 17:48:48,645 - Epoch: [68][ 1200/ 1207] Overall Loss 0.334237 Objective Loss 0.334237 LR 0.001000 Time 0.019672 -2023-02-13 17:48:48,761 - Epoch: [68][ 1207/ 1207] Overall Loss 0.334417 Objective Loss 0.334417 Top1 83.231707 Top5 97.560976 LR 0.001000 Time 0.019654 -2023-02-13 17:48:48,833 - --- validate (epoch=68)----------- -2023-02-13 17:48:48,834 - 34311 samples (256 per mini-batch) -2023-02-13 17:48:49,239 - Epoch: [68][ 10/ 135] Loss 0.352378 Top1 82.382812 Top5 97.109375 -2023-02-13 17:48:49,381 - Epoch: [68][ 20/ 135] Loss 0.352164 Top1 82.011719 Top5 97.246094 -2023-02-13 17:48:49,527 - Epoch: [68][ 30/ 135] Loss 0.360598 Top1 81.992188 Top5 97.174479 -2023-02-13 17:48:49,667 - Epoch: [68][ 40/ 135] Loss 0.353230 Top1 82.050781 Top5 97.265625 -2023-02-13 17:48:49,793 - Epoch: [68][ 50/ 135] Loss 0.355761 Top1 82.023438 Top5 97.289062 -2023-02-13 17:48:49,921 - Epoch: [68][ 60/ 135] Loss 0.358221 Top1 81.881510 Top5 97.167969 -2023-02-13 17:48:50,048 - Epoch: [68][ 70/ 135] Loss 0.356274 Top1 81.886161 Top5 97.276786 -2023-02-13 17:48:50,175 - Epoch: [68][ 80/ 135] Loss 0.354197 Top1 81.958008 Top5 97.333984 -2023-02-13 17:48:50,304 - Epoch: [68][ 90/ 135] Loss 0.360258 Top1 81.792535 Top5 97.191840 -2023-02-13 17:48:50,431 - Epoch: [68][ 100/ 135] Loss 0.364790 Top1 81.718750 Top5 97.148438 -2023-02-13 17:48:50,557 - Epoch: [68][ 110/ 135] Loss 0.365039 Top1 81.768466 Top5 97.155540 -2023-02-13 17:48:50,691 - Epoch: [68][ 120/ 135] Loss 0.365520 Top1 81.728516 Top5 97.145182 -2023-02-13 17:48:50,829 - Epoch: [68][ 130/ 135] Loss 0.364830 Top1 81.727764 Top5 97.172476 -2023-02-13 17:48:50,873 - Epoch: [68][ 135/ 135] Loss 0.363450 Top1 81.693917 Top5 97.170004 -2023-02-13 17:48:50,944 - ==> Top1: 81.694 Top5: 97.170 Loss: 0.363 - -2023-02-13 17:48:50,944 - ==> Confusion: -[[ 778 2 3 2 19 4 0 3 10 100 2 7 3 9 7 2 2 0 1 2 11] - [ 1 926 0 1 6 56 2 13 1 1 1 2 3 0 4 2 2 1 4 3 4] - [ 5 3 933 13 2 1 16 18 1 2 3 2 2 11 3 6 6 3 7 6 15] - [ 6 0 17 891 2 6 4 4 4 4 13 0 7 0 26 3 4 3 13 3 6] - [ 9 11 2 0 978 13 1 1 1 2 0 3 0 11 10 2 8 2 0 5 7] - [ 2 17 1 5 5 972 3 13 3 2 3 7 4 22 2 0 2 2 0 2 3] - [ 1 6 15 2 0 7 1027 7 0 2 3 2 1 3 1 3 0 4 2 9 4] - [ 1 10 9 3 2 42 8 907 2 1 2 4 3 2 1 1 0 0 14 9 3] - [ 13 2 0 1 2 1 0 1 874 45 11 2 1 18 30 0 1 3 4 0 0] - [ 51 1 0 0 6 3 0 1 39 870 1 0 0 28 4 0 2 0 1 0 5] - [ 2 3 4 6 2 1 1 6 18 1 963 3 0 19 2 0 1 0 12 3 4] - [ 1 3 1 0 2 20 0 6 4 1 0 885 22 13 5 6 8 11 2 14 1] - [ 1 0 1 8 2 5 0 2 2 0 1 40 846 0 2 2 2 27 3 4 11] - [ 3 5 1 0 5 9 0 3 10 12 7 6 3 935 8 3 4 2 1 6 1] - [ 6 3 1 14 5 5 0 1 18 10 1 3 2 4 995 1 3 6 8 0 6] - [ 1 2 5 0 10 8 3 0 0 0 0 5 8 7 1 958 16 10 0 7 5] - [ 1 9 1 0 13 1 0 0 2 0 0 4 1 2 4 6 994 3 2 7 11] - [ 7 1 1 6 2 3 2 1 0 2 0 7 21 4 2 9 0 972 2 5 4] - [ 2 10 6 15 0 2 0 44 3 1 5 1 5 0 22 1 3 1 957 5 3] - [ 1 2 0 3 3 7 6 14 0 0 0 14 4 5 0 1 3 2 0 1079 4] - [ 119 308 205 156 150 322 90 189 122 98 185 116 331 430 202 83 396 130 149 362 9291]] - -2023-02-13 17:48:50,946 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:48:50,946 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:48:50,952 - - -2023-02-13 17:48:50,952 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:48:51,825 - Epoch: [69][ 10/ 1207] Overall Loss 0.309575 Objective Loss 0.309575 LR 0.001000 Time 0.087223 -2023-02-13 17:48:52,022 - Epoch: [69][ 20/ 1207] Overall Loss 0.307877 Objective Loss 0.307877 LR 0.001000 Time 0.053456 -2023-02-13 17:48:52,224 - Epoch: [69][ 30/ 1207] Overall Loss 0.312209 Objective Loss 0.312209 LR 0.001000 Time 0.042365 -2023-02-13 17:48:52,422 - Epoch: [69][ 40/ 1207] Overall Loss 0.326685 Objective Loss 0.326685 LR 0.001000 Time 0.036719 -2023-02-13 17:48:52,625 - Epoch: [69][ 50/ 1207] Overall Loss 0.327843 Objective Loss 0.327843 LR 0.001000 Time 0.033416 -2023-02-13 17:48:52,822 - Epoch: [69][ 60/ 1207] Overall Loss 0.327637 Objective Loss 0.327637 LR 0.001000 Time 0.031137 -2023-02-13 17:48:53,024 - Epoch: [69][ 70/ 1207] Overall Loss 0.323637 Objective Loss 0.323637 LR 0.001000 Time 0.029568 -2023-02-13 17:48:53,222 - Epoch: [69][ 80/ 1207] Overall Loss 0.320178 Objective Loss 0.320178 LR 0.001000 Time 0.028333 -2023-02-13 17:48:53,424 - Epoch: [69][ 90/ 1207] Overall Loss 0.322520 Objective Loss 0.322520 LR 0.001000 Time 0.027430 -2023-02-13 17:48:53,621 - Epoch: [69][ 100/ 1207] Overall Loss 0.322937 Objective Loss 0.322937 LR 0.001000 Time 0.026656 -2023-02-13 17:48:53,823 - Epoch: [69][ 110/ 1207] Overall Loss 0.321495 Objective Loss 0.321495 LR 0.001000 Time 0.026059 -2023-02-13 17:48:54,020 - Epoch: [69][ 120/ 1207] Overall Loss 0.319371 Objective Loss 0.319371 LR 0.001000 Time 0.025532 -2023-02-13 17:48:54,222 - Epoch: [69][ 130/ 1207] Overall Loss 0.317316 Objective Loss 0.317316 LR 0.001000 Time 0.025114 -2023-02-13 17:48:54,420 - Epoch: [69][ 140/ 1207] Overall Loss 0.317861 Objective Loss 0.317861 LR 0.001000 Time 0.024734 -2023-02-13 17:48:54,622 - Epoch: [69][ 150/ 1207] Overall Loss 0.322320 Objective Loss 0.322320 LR 0.001000 Time 0.024428 -2023-02-13 17:48:54,819 - Epoch: [69][ 160/ 1207] Overall Loss 0.325580 Objective Loss 0.325580 LR 0.001000 Time 0.024132 -2023-02-13 17:48:55,021 - Epoch: [69][ 170/ 1207] Overall Loss 0.325013 Objective Loss 0.325013 LR 0.001000 Time 0.023897 -2023-02-13 17:48:55,218 - Epoch: [69][ 180/ 1207] Overall Loss 0.326001 Objective Loss 0.326001 LR 0.001000 Time 0.023664 -2023-02-13 17:48:55,421 - Epoch: [69][ 190/ 1207] Overall Loss 0.325594 Objective Loss 0.325594 LR 0.001000 Time 0.023485 -2023-02-13 17:48:55,619 - Epoch: [69][ 200/ 1207] Overall Loss 0.325168 Objective Loss 0.325168 LR 0.001000 Time 0.023296 -2023-02-13 17:48:55,821 - Epoch: [69][ 210/ 1207] Overall Loss 0.325163 Objective Loss 0.325163 LR 0.001000 Time 0.023147 -2023-02-13 17:48:56,018 - Epoch: [69][ 220/ 1207] Overall Loss 0.325716 Objective Loss 0.325716 LR 0.001000 Time 0.022992 -2023-02-13 17:48:56,221 - Epoch: [69][ 230/ 1207] Overall Loss 0.325764 Objective Loss 0.325764 LR 0.001000 Time 0.022870 -2023-02-13 17:48:56,419 - Epoch: [69][ 240/ 1207] Overall Loss 0.327703 Objective Loss 0.327703 LR 0.001000 Time 0.022742 -2023-02-13 17:48:56,620 - Epoch: [69][ 250/ 1207] Overall Loss 0.328018 Objective Loss 0.328018 LR 0.001000 Time 0.022636 -2023-02-13 17:48:56,818 - Epoch: [69][ 260/ 1207] Overall Loss 0.328683 Objective Loss 0.328683 LR 0.001000 Time 0.022525 -2023-02-13 17:48:57,021 - Epoch: [69][ 270/ 1207] Overall Loss 0.328690 Objective Loss 0.328690 LR 0.001000 Time 0.022439 -2023-02-13 17:48:57,219 - Epoch: [69][ 280/ 1207] Overall Loss 0.329434 Objective Loss 0.329434 LR 0.001000 Time 0.022343 -2023-02-13 17:48:57,420 - Epoch: [69][ 290/ 1207] Overall Loss 0.329543 Objective Loss 0.329543 LR 0.001000 Time 0.022266 -2023-02-13 17:48:57,618 - Epoch: [69][ 300/ 1207] Overall Loss 0.329648 Objective Loss 0.329648 LR 0.001000 Time 0.022184 -2023-02-13 17:48:57,821 - Epoch: [69][ 310/ 1207] Overall Loss 0.329952 Objective Loss 0.329952 LR 0.001000 Time 0.022120 -2023-02-13 17:48:58,019 - Epoch: [69][ 320/ 1207] Overall Loss 0.329783 Objective Loss 0.329783 LR 0.001000 Time 0.022047 -2023-02-13 17:48:58,221 - Epoch: [69][ 330/ 1207] Overall Loss 0.328832 Objective Loss 0.328832 LR 0.001000 Time 0.021990 -2023-02-13 17:48:58,419 - Epoch: [69][ 340/ 1207] Overall Loss 0.329346 Objective Loss 0.329346 LR 0.001000 Time 0.021923 -2023-02-13 17:48:58,621 - Epoch: [69][ 350/ 1207] Overall Loss 0.329055 Objective Loss 0.329055 LR 0.001000 Time 0.021873 -2023-02-13 17:48:58,819 - Epoch: [69][ 360/ 1207] Overall Loss 0.329264 Objective Loss 0.329264 LR 0.001000 Time 0.021815 -2023-02-13 17:48:59,022 - Epoch: [69][ 370/ 1207] Overall Loss 0.329655 Objective Loss 0.329655 LR 0.001000 Time 0.021772 -2023-02-13 17:48:59,220 - Epoch: [69][ 380/ 1207] Overall Loss 0.330444 Objective Loss 0.330444 LR 0.001000 Time 0.021720 -2023-02-13 17:48:59,421 - Epoch: [69][ 390/ 1207] Overall Loss 0.330395 Objective Loss 0.330395 LR 0.001000 Time 0.021679 -2023-02-13 17:48:59,619 - Epoch: [69][ 400/ 1207] Overall Loss 0.330332 Objective Loss 0.330332 LR 0.001000 Time 0.021630 -2023-02-13 17:48:59,821 - Epoch: [69][ 410/ 1207] Overall Loss 0.330565 Objective Loss 0.330565 LR 0.001000 Time 0.021593 -2023-02-13 17:49:00,019 - Epoch: [69][ 420/ 1207] Overall Loss 0.330779 Objective Loss 0.330779 LR 0.001000 Time 0.021550 -2023-02-13 17:49:00,221 - Epoch: [69][ 430/ 1207] Overall Loss 0.331007 Objective Loss 0.331007 LR 0.001000 Time 0.021518 -2023-02-13 17:49:00,419 - Epoch: [69][ 440/ 1207] Overall Loss 0.331033 Objective Loss 0.331033 LR 0.001000 Time 0.021478 -2023-02-13 17:49:00,621 - Epoch: [69][ 450/ 1207] Overall Loss 0.330849 Objective Loss 0.330849 LR 0.001000 Time 0.021449 -2023-02-13 17:49:00,819 - Epoch: [69][ 460/ 1207] Overall Loss 0.331405 Objective Loss 0.331405 LR 0.001000 Time 0.021413 -2023-02-13 17:49:01,022 - Epoch: [69][ 470/ 1207] Overall Loss 0.330811 Objective Loss 0.330811 LR 0.001000 Time 0.021388 -2023-02-13 17:49:01,219 - Epoch: [69][ 480/ 1207] Overall Loss 0.330961 Objective Loss 0.330961 LR 0.001000 Time 0.021353 -2023-02-13 17:49:01,422 - Epoch: [69][ 490/ 1207] Overall Loss 0.331476 Objective Loss 0.331476 LR 0.001000 Time 0.021330 -2023-02-13 17:49:01,620 - Epoch: [69][ 500/ 1207] Overall Loss 0.331154 Objective Loss 0.331154 LR 0.001000 Time 0.021299 -2023-02-13 17:49:01,823 - Epoch: [69][ 510/ 1207] Overall Loss 0.331127 Objective Loss 0.331127 LR 0.001000 Time 0.021278 -2023-02-13 17:49:02,022 - Epoch: [69][ 520/ 1207] Overall Loss 0.330612 Objective Loss 0.330612 LR 0.001000 Time 0.021250 -2023-02-13 17:49:02,224 - Epoch: [69][ 530/ 1207] Overall Loss 0.330596 Objective Loss 0.330596 LR 0.001000 Time 0.021230 -2023-02-13 17:49:02,423 - Epoch: [69][ 540/ 1207] Overall Loss 0.330376 Objective Loss 0.330376 LR 0.001000 Time 0.021205 -2023-02-13 17:49:02,625 - Epoch: [69][ 550/ 1207] Overall Loss 0.330435 Objective Loss 0.330435 LR 0.001000 Time 0.021186 -2023-02-13 17:49:02,823 - Epoch: [69][ 560/ 1207] Overall Loss 0.331144 Objective Loss 0.331144 LR 0.001000 Time 0.021161 -2023-02-13 17:49:03,026 - Epoch: [69][ 570/ 1207] Overall Loss 0.331499 Objective Loss 0.331499 LR 0.001000 Time 0.021145 -2023-02-13 17:49:03,224 - Epoch: [69][ 580/ 1207] Overall Loss 0.331732 Objective Loss 0.331732 LR 0.001000 Time 0.021121 -2023-02-13 17:49:03,419 - Epoch: [69][ 590/ 1207] Overall Loss 0.332058 Objective Loss 0.332058 LR 0.001000 Time 0.021093 -2023-02-13 17:49:03,611 - Epoch: [69][ 600/ 1207] Overall Loss 0.331520 Objective Loss 0.331520 LR 0.001000 Time 0.021060 -2023-02-13 17:49:03,805 - Epoch: [69][ 610/ 1207] Overall Loss 0.330999 Objective Loss 0.330999 LR 0.001000 Time 0.021034 -2023-02-13 17:49:03,998 - Epoch: [69][ 620/ 1207] Overall Loss 0.331088 Objective Loss 0.331088 LR 0.001000 Time 0.021004 -2023-02-13 17:49:04,192 - Epoch: [69][ 630/ 1207] Overall Loss 0.330973 Objective Loss 0.330973 LR 0.001000 Time 0.020978 -2023-02-13 17:49:04,385 - Epoch: [69][ 640/ 1207] Overall Loss 0.330841 Objective Loss 0.330841 LR 0.001000 Time 0.020951 -2023-02-13 17:49:04,579 - Epoch: [69][ 650/ 1207] Overall Loss 0.330666 Objective Loss 0.330666 LR 0.001000 Time 0.020927 -2023-02-13 17:49:04,773 - Epoch: [69][ 660/ 1207] Overall Loss 0.331009 Objective Loss 0.331009 LR 0.001000 Time 0.020904 -2023-02-13 17:49:04,970 - Epoch: [69][ 670/ 1207] Overall Loss 0.330872 Objective Loss 0.330872 LR 0.001000 Time 0.020885 -2023-02-13 17:49:05,166 - Epoch: [69][ 680/ 1207] Overall Loss 0.330955 Objective Loss 0.330955 LR 0.001000 Time 0.020866 -2023-02-13 17:49:05,364 - Epoch: [69][ 690/ 1207] Overall Loss 0.330607 Objective Loss 0.330607 LR 0.001000 Time 0.020849 -2023-02-13 17:49:05,560 - Epoch: [69][ 700/ 1207] Overall Loss 0.330940 Objective Loss 0.330940 LR 0.001000 Time 0.020831 -2023-02-13 17:49:05,756 - Epoch: [69][ 710/ 1207] Overall Loss 0.330819 Objective Loss 0.330819 LR 0.001000 Time 0.020813 -2023-02-13 17:49:05,953 - Epoch: [69][ 720/ 1207] Overall Loss 0.330438 Objective Loss 0.330438 LR 0.001000 Time 0.020798 -2023-02-13 17:49:06,149 - Epoch: [69][ 730/ 1207] Overall Loss 0.329815 Objective Loss 0.329815 LR 0.001000 Time 0.020781 -2023-02-13 17:49:06,347 - Epoch: [69][ 740/ 1207] Overall Loss 0.329887 Objective Loss 0.329887 LR 0.001000 Time 0.020766 -2023-02-13 17:49:06,543 - Epoch: [69][ 750/ 1207] Overall Loss 0.330132 Objective Loss 0.330132 LR 0.001000 Time 0.020751 -2023-02-13 17:49:06,739 - Epoch: [69][ 760/ 1207] Overall Loss 0.329658 Objective Loss 0.329658 LR 0.001000 Time 0.020735 -2023-02-13 17:49:06,937 - Epoch: [69][ 770/ 1207] Overall Loss 0.329994 Objective Loss 0.329994 LR 0.001000 Time 0.020722 -2023-02-13 17:49:07,132 - Epoch: [69][ 780/ 1207] Overall Loss 0.330104 Objective Loss 0.330104 LR 0.001000 Time 0.020707 -2023-02-13 17:49:07,330 - Epoch: [69][ 790/ 1207] Overall Loss 0.330411 Objective Loss 0.330411 LR 0.001000 Time 0.020694 -2023-02-13 17:49:07,526 - Epoch: [69][ 800/ 1207] Overall Loss 0.330291 Objective Loss 0.330291 LR 0.001000 Time 0.020680 -2023-02-13 17:49:07,722 - Epoch: [69][ 810/ 1207] Overall Loss 0.330347 Objective Loss 0.330347 LR 0.001000 Time 0.020666 -2023-02-13 17:49:07,918 - Epoch: [69][ 820/ 1207] Overall Loss 0.329933 Objective Loss 0.329933 LR 0.001000 Time 0.020653 -2023-02-13 17:49:08,115 - Epoch: [69][ 830/ 1207] Overall Loss 0.330017 Objective Loss 0.330017 LR 0.001000 Time 0.020641 -2023-02-13 17:49:08,311 - Epoch: [69][ 840/ 1207] Overall Loss 0.330570 Objective Loss 0.330570 LR 0.001000 Time 0.020628 -2023-02-13 17:49:08,507 - Epoch: [69][ 850/ 1207] Overall Loss 0.330477 Objective Loss 0.330477 LR 0.001000 Time 0.020616 -2023-02-13 17:49:08,703 - Epoch: [69][ 860/ 1207] Overall Loss 0.330699 Objective Loss 0.330699 LR 0.001000 Time 0.020603 -2023-02-13 17:49:08,900 - Epoch: [69][ 870/ 1207] Overall Loss 0.330836 Objective Loss 0.330836 LR 0.001000 Time 0.020593 -2023-02-13 17:49:09,096 - Epoch: [69][ 880/ 1207] Overall Loss 0.330734 Objective Loss 0.330734 LR 0.001000 Time 0.020581 -2023-02-13 17:49:09,293 - Epoch: [69][ 890/ 1207] Overall Loss 0.330665 Objective Loss 0.330665 LR 0.001000 Time 0.020571 -2023-02-13 17:49:09,489 - Epoch: [69][ 900/ 1207] Overall Loss 0.330691 Objective Loss 0.330691 LR 0.001000 Time 0.020560 -2023-02-13 17:49:09,685 - Epoch: [69][ 910/ 1207] Overall Loss 0.331050 Objective Loss 0.331050 LR 0.001000 Time 0.020549 -2023-02-13 17:49:09,881 - Epoch: [69][ 920/ 1207] Overall Loss 0.331281 Objective Loss 0.331281 LR 0.001000 Time 0.020537 -2023-02-13 17:49:10,077 - Epoch: [69][ 930/ 1207] Overall Loss 0.331553 Objective Loss 0.331553 LR 0.001000 Time 0.020528 -2023-02-13 17:49:10,273 - Epoch: [69][ 940/ 1207] Overall Loss 0.331423 Objective Loss 0.331423 LR 0.001000 Time 0.020517 -2023-02-13 17:49:10,469 - Epoch: [69][ 950/ 1207] Overall Loss 0.331693 Objective Loss 0.331693 LR 0.001000 Time 0.020507 -2023-02-13 17:49:10,665 - Epoch: [69][ 960/ 1207] Overall Loss 0.331964 Objective Loss 0.331964 LR 0.001000 Time 0.020497 -2023-02-13 17:49:10,862 - Epoch: [69][ 970/ 1207] Overall Loss 0.331943 Objective Loss 0.331943 LR 0.001000 Time 0.020488 -2023-02-13 17:49:11,057 - Epoch: [69][ 980/ 1207] Overall Loss 0.331782 Objective Loss 0.331782 LR 0.001000 Time 0.020478 -2023-02-13 17:49:11,253 - Epoch: [69][ 990/ 1207] Overall Loss 0.331950 Objective Loss 0.331950 LR 0.001000 Time 0.020469 -2023-02-13 17:49:11,450 - Epoch: [69][ 1000/ 1207] Overall Loss 0.331801 Objective Loss 0.331801 LR 0.001000 Time 0.020461 -2023-02-13 17:49:11,647 - Epoch: [69][ 1010/ 1207] Overall Loss 0.331582 Objective Loss 0.331582 LR 0.001000 Time 0.020452 -2023-02-13 17:49:11,842 - Epoch: [69][ 1020/ 1207] Overall Loss 0.331515 Objective Loss 0.331515 LR 0.001000 Time 0.020443 -2023-02-13 17:49:12,039 - Epoch: [69][ 1030/ 1207] Overall Loss 0.332186 Objective Loss 0.332186 LR 0.001000 Time 0.020436 -2023-02-13 17:49:12,235 - Epoch: [69][ 1040/ 1207] Overall Loss 0.332219 Objective Loss 0.332219 LR 0.001000 Time 0.020427 -2023-02-13 17:49:12,432 - Epoch: [69][ 1050/ 1207] Overall Loss 0.332558 Objective Loss 0.332558 LR 0.001000 Time 0.020420 -2023-02-13 17:49:12,629 - Epoch: [69][ 1060/ 1207] Overall Loss 0.332383 Objective Loss 0.332383 LR 0.001000 Time 0.020412 -2023-02-13 17:49:12,825 - Epoch: [69][ 1070/ 1207] Overall Loss 0.332333 Objective Loss 0.332333 LR 0.001000 Time 0.020405 -2023-02-13 17:49:13,022 - Epoch: [69][ 1080/ 1207] Overall Loss 0.332540 Objective Loss 0.332540 LR 0.001000 Time 0.020398 -2023-02-13 17:49:13,218 - Epoch: [69][ 1090/ 1207] Overall Loss 0.332457 Objective Loss 0.332457 LR 0.001000 Time 0.020390 -2023-02-13 17:49:13,415 - Epoch: [69][ 1100/ 1207] Overall Loss 0.332321 Objective Loss 0.332321 LR 0.001000 Time 0.020384 -2023-02-13 17:49:13,611 - Epoch: [69][ 1110/ 1207] Overall Loss 0.332567 Objective Loss 0.332567 LR 0.001000 Time 0.020377 -2023-02-13 17:49:13,806 - Epoch: [69][ 1120/ 1207] Overall Loss 0.332199 Objective Loss 0.332199 LR 0.001000 Time 0.020368 -2023-02-13 17:49:14,003 - Epoch: [69][ 1130/ 1207] Overall Loss 0.332114 Objective Loss 0.332114 LR 0.001000 Time 0.020362 -2023-02-13 17:49:14,198 - Epoch: [69][ 1140/ 1207] Overall Loss 0.332239 Objective Loss 0.332239 LR 0.001000 Time 0.020354 -2023-02-13 17:49:14,395 - Epoch: [69][ 1150/ 1207] Overall Loss 0.332207 Objective Loss 0.332207 LR 0.001000 Time 0.020348 -2023-02-13 17:49:14,592 - Epoch: [69][ 1160/ 1207] Overall Loss 0.332213 Objective Loss 0.332213 LR 0.001000 Time 0.020342 -2023-02-13 17:49:14,788 - Epoch: [69][ 1170/ 1207] Overall Loss 0.332304 Objective Loss 0.332304 LR 0.001000 Time 0.020336 -2023-02-13 17:49:14,981 - Epoch: [69][ 1180/ 1207] Overall Loss 0.332550 Objective Loss 0.332550 LR 0.001000 Time 0.020326 -2023-02-13 17:49:15,175 - Epoch: [69][ 1190/ 1207] Overall Loss 0.332602 Objective Loss 0.332602 LR 0.001000 Time 0.020318 -2023-02-13 17:49:15,419 - Epoch: [69][ 1200/ 1207] Overall Loss 0.332920 Objective Loss 0.332920 LR 0.001000 Time 0.020352 -2023-02-13 17:49:15,534 - Epoch: [69][ 1207/ 1207] Overall Loss 0.333021 Objective Loss 0.333021 Top1 81.097561 Top5 97.256098 LR 0.001000 Time 0.020329 -2023-02-13 17:49:15,615 - --- validate (epoch=69)----------- -2023-02-13 17:49:15,615 - 34311 samples (256 per mini-batch) -2023-02-13 17:49:16,124 - Epoch: [69][ 10/ 135] Loss 0.380688 Top1 81.601562 Top5 97.109375 -2023-02-13 17:49:16,253 - Epoch: [69][ 20/ 135] Loss 0.381709 Top1 81.582031 Top5 96.855469 -2023-02-13 17:49:16,382 - Epoch: [69][ 30/ 135] Loss 0.375407 Top1 81.744792 Top5 96.848958 -2023-02-13 17:49:16,508 - Epoch: [69][ 40/ 135] Loss 0.368922 Top1 81.953125 Top5 96.982422 -2023-02-13 17:49:16,636 - Epoch: [69][ 50/ 135] Loss 0.371748 Top1 81.718750 Top5 96.929688 -2023-02-13 17:49:16,764 - Epoch: [69][ 60/ 135] Loss 0.366941 Top1 81.549479 Top5 96.998698 -2023-02-13 17:49:16,898 - Epoch: [69][ 70/ 135] Loss 0.364303 Top1 81.674107 Top5 97.042411 -2023-02-13 17:49:17,042 - Epoch: [69][ 80/ 135] Loss 0.362666 Top1 81.723633 Top5 97.041016 -2023-02-13 17:49:17,167 - Epoch: [69][ 90/ 135] Loss 0.361660 Top1 81.762153 Top5 97.157118 -2023-02-13 17:49:17,296 - Epoch: [69][ 100/ 135] Loss 0.365622 Top1 81.707031 Top5 97.152344 -2023-02-13 17:49:17,426 - Epoch: [69][ 110/ 135] Loss 0.369827 Top1 81.637074 Top5 97.052557 -2023-02-13 17:49:17,557 - Epoch: [69][ 120/ 135] Loss 0.371705 Top1 81.458333 Top5 97.034505 -2023-02-13 17:49:17,689 - Epoch: [69][ 130/ 135] Loss 0.371187 Top1 81.433293 Top5 97.043269 -2023-02-13 17:49:17,738 - Epoch: [69][ 135/ 135] Loss 0.369600 Top1 81.370406 Top5 97.035936 -2023-02-13 17:49:17,810 - ==> Top1: 81.370 Top5: 97.036 Loss: 0.370 - -2023-02-13 17:49:17,811 - ==> Confusion: -[[ 839 5 7 3 8 4 0 4 5 63 0 4 1 2 3 2 4 2 2 2 7] - [ 1 901 3 6 6 47 9 30 4 2 3 1 0 0 1 2 5 0 6 2 4] - [ 8 4 943 13 1 2 26 22 1 1 3 1 2 3 2 6 3 5 3 5 4] - [ 2 2 22 896 1 3 4 2 2 2 14 0 9 1 12 3 2 11 17 1 10] - [ 25 9 4 1 960 16 2 2 1 3 1 4 2 1 13 8 6 1 0 3 4] - [ 4 15 0 6 3 962 5 24 2 4 2 13 2 8 2 1 5 1 1 9 1] - [ 3 3 18 1 0 6 1040 7 0 2 1 0 2 0 0 4 0 1 1 8 2] - [ 3 7 13 3 1 31 4 922 0 1 3 9 1 1 0 0 0 1 9 13 2] - [ 22 3 0 1 0 0 0 1 867 50 18 3 3 11 18 1 1 3 6 0 1] - [ 85 2 4 0 5 2 1 2 29 852 0 0 0 14 3 1 1 6 1 2 2] - [ 1 3 6 8 2 5 3 5 10 0 980 2 3 4 1 0 2 1 10 1 4] - [ 1 5 2 1 3 31 0 8 2 0 0 856 25 11 1 2 2 15 6 30 4] - [ 0 0 2 7 1 4 0 5 0 1 1 41 845 0 2 7 1 29 0 3 10] - [ 3 4 8 2 2 11 1 4 10 20 18 5 5 903 7 5 5 3 0 5 3] - [ 12 3 3 19 2 7 0 2 28 4 9 3 2 2 964 2 2 7 8 1 12] - [ 1 2 10 3 7 6 11 3 0 0 0 8 8 3 0 943 10 16 0 10 5] - [ 1 6 1 1 13 4 1 0 3 0 1 2 4 2 1 5 993 2 2 8 11] - [ 6 2 3 4 1 3 2 0 0 0 1 12 13 0 1 11 0 983 1 5 3] - [ 5 2 11 14 0 2 0 49 5 1 7 3 2 0 15 0 2 1 958 5 4] - [ 1 4 2 1 2 6 9 20 1 0 0 12 4 3 0 4 1 4 1 1067 6] - [ 178 258 298 189 137 311 127 260 89 124 239 148 320 297 130 108 300 148 139 388 9246]] - -2023-02-13 17:49:17,812 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:49:17,812 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:49:17,818 - - -2023-02-13 17:49:17,818 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:49:18,701 - Epoch: [70][ 10/ 1207] Overall Loss 0.307871 Objective Loss 0.307871 LR 0.001000 Time 0.088189 -2023-02-13 17:49:18,891 - Epoch: [70][ 20/ 1207] Overall Loss 0.336419 Objective Loss 0.336419 LR 0.001000 Time 0.053610 -2023-02-13 17:49:19,083 - Epoch: [70][ 30/ 1207] Overall Loss 0.344912 Objective Loss 0.344912 LR 0.001000 Time 0.042098 -2023-02-13 17:49:19,273 - Epoch: [70][ 40/ 1207] Overall Loss 0.335999 Objective Loss 0.335999 LR 0.001000 Time 0.036323 -2023-02-13 17:49:19,464 - Epoch: [70][ 50/ 1207] Overall Loss 0.338120 Objective Loss 0.338120 LR 0.001000 Time 0.032876 -2023-02-13 17:49:19,654 - Epoch: [70][ 60/ 1207] Overall Loss 0.335075 Objective Loss 0.335075 LR 0.001000 Time 0.030563 -2023-02-13 17:49:19,844 - Epoch: [70][ 70/ 1207] Overall Loss 0.336458 Objective Loss 0.336458 LR 0.001000 Time 0.028905 -2023-02-13 17:49:20,035 - Epoch: [70][ 80/ 1207] Overall Loss 0.336499 Objective Loss 0.336499 LR 0.001000 Time 0.027664 -2023-02-13 17:49:20,224 - Epoch: [70][ 90/ 1207] Overall Loss 0.334559 Objective Loss 0.334559 LR 0.001000 Time 0.026695 -2023-02-13 17:49:20,415 - Epoch: [70][ 100/ 1207] Overall Loss 0.335193 Objective Loss 0.335193 LR 0.001000 Time 0.025928 -2023-02-13 17:49:20,605 - Epoch: [70][ 110/ 1207] Overall Loss 0.335812 Objective Loss 0.335812 LR 0.001000 Time 0.025292 -2023-02-13 17:49:20,795 - Epoch: [70][ 120/ 1207] Overall Loss 0.335399 Objective Loss 0.335399 LR 0.001000 Time 0.024767 -2023-02-13 17:49:20,985 - Epoch: [70][ 130/ 1207] Overall Loss 0.334488 Objective Loss 0.334488 LR 0.001000 Time 0.024325 -2023-02-13 17:49:21,176 - Epoch: [70][ 140/ 1207] Overall Loss 0.333387 Objective Loss 0.333387 LR 0.001000 Time 0.023944 -2023-02-13 17:49:21,366 - Epoch: [70][ 150/ 1207] Overall Loss 0.332384 Objective Loss 0.332384 LR 0.001000 Time 0.023614 -2023-02-13 17:49:21,557 - Epoch: [70][ 160/ 1207] Overall Loss 0.329936 Objective Loss 0.329936 LR 0.001000 Time 0.023328 -2023-02-13 17:49:21,747 - Epoch: [70][ 170/ 1207] Overall Loss 0.329627 Objective Loss 0.329627 LR 0.001000 Time 0.023071 -2023-02-13 17:49:21,938 - Epoch: [70][ 180/ 1207] Overall Loss 0.329611 Objective Loss 0.329611 LR 0.001000 Time 0.022847 -2023-02-13 17:49:22,127 - Epoch: [70][ 190/ 1207] Overall Loss 0.328327 Objective Loss 0.328327 LR 0.001000 Time 0.022642 -2023-02-13 17:49:22,318 - Epoch: [70][ 200/ 1207] Overall Loss 0.329459 Objective Loss 0.329459 LR 0.001000 Time 0.022459 -2023-02-13 17:49:22,508 - Epoch: [70][ 210/ 1207] Overall Loss 0.330145 Objective Loss 0.330145 LR 0.001000 Time 0.022295 -2023-02-13 17:49:22,698 - Epoch: [70][ 220/ 1207] Overall Loss 0.329511 Objective Loss 0.329511 LR 0.001000 Time 0.022144 -2023-02-13 17:49:22,889 - Epoch: [70][ 230/ 1207] Overall Loss 0.327514 Objective Loss 0.327514 LR 0.001000 Time 0.022008 -2023-02-13 17:49:23,079 - Epoch: [70][ 240/ 1207] Overall Loss 0.327125 Objective Loss 0.327125 LR 0.001000 Time 0.021883 -2023-02-13 17:49:23,269 - Epoch: [70][ 250/ 1207] Overall Loss 0.327295 Objective Loss 0.327295 LR 0.001000 Time 0.021765 -2023-02-13 17:49:23,461 - Epoch: [70][ 260/ 1207] Overall Loss 0.327438 Objective Loss 0.327438 LR 0.001000 Time 0.021665 -2023-02-13 17:49:23,651 - Epoch: [70][ 270/ 1207] Overall Loss 0.327148 Objective Loss 0.327148 LR 0.001000 Time 0.021565 -2023-02-13 17:49:23,841 - Epoch: [70][ 280/ 1207] Overall Loss 0.326862 Objective Loss 0.326862 LR 0.001000 Time 0.021473 -2023-02-13 17:49:24,031 - Epoch: [70][ 290/ 1207] Overall Loss 0.326725 Objective Loss 0.326725 LR 0.001000 Time 0.021387 -2023-02-13 17:49:24,221 - Epoch: [70][ 300/ 1207] Overall Loss 0.327373 Objective Loss 0.327373 LR 0.001000 Time 0.021305 -2023-02-13 17:49:24,412 - Epoch: [70][ 310/ 1207] Overall Loss 0.328211 Objective Loss 0.328211 LR 0.001000 Time 0.021234 -2023-02-13 17:49:24,602 - Epoch: [70][ 320/ 1207] Overall Loss 0.328972 Objective Loss 0.328972 LR 0.001000 Time 0.021160 -2023-02-13 17:49:24,792 - Epoch: [70][ 330/ 1207] Overall Loss 0.328484 Objective Loss 0.328484 LR 0.001000 Time 0.021096 -2023-02-13 17:49:24,982 - Epoch: [70][ 340/ 1207] Overall Loss 0.327912 Objective Loss 0.327912 LR 0.001000 Time 0.021032 -2023-02-13 17:49:25,172 - Epoch: [70][ 350/ 1207] Overall Loss 0.327562 Objective Loss 0.327562 LR 0.001000 Time 0.020974 -2023-02-13 17:49:25,362 - Epoch: [70][ 360/ 1207] Overall Loss 0.328387 Objective Loss 0.328387 LR 0.001000 Time 0.020918 -2023-02-13 17:49:25,553 - Epoch: [70][ 370/ 1207] Overall Loss 0.327922 Objective Loss 0.327922 LR 0.001000 Time 0.020868 -2023-02-13 17:49:25,744 - Epoch: [70][ 380/ 1207] Overall Loss 0.328566 Objective Loss 0.328566 LR 0.001000 Time 0.020819 -2023-02-13 17:49:25,937 - Epoch: [70][ 390/ 1207] Overall Loss 0.328231 Objective Loss 0.328231 LR 0.001000 Time 0.020779 -2023-02-13 17:49:26,127 - Epoch: [70][ 400/ 1207] Overall Loss 0.328503 Objective Loss 0.328503 LR 0.001000 Time 0.020734 -2023-02-13 17:49:26,319 - Epoch: [70][ 410/ 1207] Overall Loss 0.328408 Objective Loss 0.328408 LR 0.001000 Time 0.020696 -2023-02-13 17:49:26,511 - Epoch: [70][ 420/ 1207] Overall Loss 0.327677 Objective Loss 0.327677 LR 0.001000 Time 0.020659 -2023-02-13 17:49:26,702 - Epoch: [70][ 430/ 1207] Overall Loss 0.327923 Objective Loss 0.327923 LR 0.001000 Time 0.020623 -2023-02-13 17:49:26,893 - Epoch: [70][ 440/ 1207] Overall Loss 0.327660 Objective Loss 0.327660 LR 0.001000 Time 0.020588 -2023-02-13 17:49:27,085 - Epoch: [70][ 450/ 1207] Overall Loss 0.326864 Objective Loss 0.326864 LR 0.001000 Time 0.020555 -2023-02-13 17:49:27,276 - Epoch: [70][ 460/ 1207] Overall Loss 0.326524 Objective Loss 0.326524 LR 0.001000 Time 0.020522 -2023-02-13 17:49:27,468 - Epoch: [70][ 470/ 1207] Overall Loss 0.326398 Objective Loss 0.326398 LR 0.001000 Time 0.020495 -2023-02-13 17:49:27,659 - Epoch: [70][ 480/ 1207] Overall Loss 0.326282 Objective Loss 0.326282 LR 0.001000 Time 0.020464 -2023-02-13 17:49:27,850 - Epoch: [70][ 490/ 1207] Overall Loss 0.326900 Objective Loss 0.326900 LR 0.001000 Time 0.020436 -2023-02-13 17:49:28,042 - Epoch: [70][ 500/ 1207] Overall Loss 0.326482 Objective Loss 0.326482 LR 0.001000 Time 0.020410 -2023-02-13 17:49:28,233 - Epoch: [70][ 510/ 1207] Overall Loss 0.326413 Objective Loss 0.326413 LR 0.001000 Time 0.020384 -2023-02-13 17:49:28,424 - Epoch: [70][ 520/ 1207] Overall Loss 0.326738 Objective Loss 0.326738 LR 0.001000 Time 0.020359 -2023-02-13 17:49:28,615 - Epoch: [70][ 530/ 1207] Overall Loss 0.327049 Objective Loss 0.327049 LR 0.001000 Time 0.020335 -2023-02-13 17:49:28,806 - Epoch: [70][ 540/ 1207] Overall Loss 0.327757 Objective Loss 0.327757 LR 0.001000 Time 0.020310 -2023-02-13 17:49:28,998 - Epoch: [70][ 550/ 1207] Overall Loss 0.328608 Objective Loss 0.328608 LR 0.001000 Time 0.020289 -2023-02-13 17:49:29,189 - Epoch: [70][ 560/ 1207] Overall Loss 0.328825 Objective Loss 0.328825 LR 0.001000 Time 0.020268 -2023-02-13 17:49:29,378 - Epoch: [70][ 570/ 1207] Overall Loss 0.328682 Objective Loss 0.328682 LR 0.001000 Time 0.020244 -2023-02-13 17:49:29,568 - Epoch: [70][ 580/ 1207] Overall Loss 0.328789 Objective Loss 0.328789 LR 0.001000 Time 0.020220 -2023-02-13 17:49:29,764 - Epoch: [70][ 590/ 1207] Overall Loss 0.329171 Objective Loss 0.329171 LR 0.001000 Time 0.020210 -2023-02-13 17:49:29,965 - Epoch: [70][ 600/ 1207] Overall Loss 0.328588 Objective Loss 0.328588 LR 0.001000 Time 0.020207 -2023-02-13 17:49:30,173 - Epoch: [70][ 610/ 1207] Overall Loss 0.328553 Objective Loss 0.328553 LR 0.001000 Time 0.020216 -2023-02-13 17:49:30,384 - Epoch: [70][ 620/ 1207] Overall Loss 0.328166 Objective Loss 0.328166 LR 0.001000 Time 0.020231 -2023-02-13 17:49:30,592 - Epoch: [70][ 630/ 1207] Overall Loss 0.329031 Objective Loss 0.329031 LR 0.001000 Time 0.020239 -2023-02-13 17:49:30,803 - Epoch: [70][ 640/ 1207] Overall Loss 0.329249 Objective Loss 0.329249 LR 0.001000 Time 0.020251 -2023-02-13 17:49:31,010 - Epoch: [70][ 650/ 1207] Overall Loss 0.328949 Objective Loss 0.328949 LR 0.001000 Time 0.020258 -2023-02-13 17:49:31,220 - Epoch: [70][ 660/ 1207] Overall Loss 0.329278 Objective Loss 0.329278 LR 0.001000 Time 0.020269 -2023-02-13 17:49:31,428 - Epoch: [70][ 670/ 1207] Overall Loss 0.329281 Objective Loss 0.329281 LR 0.001000 Time 0.020276 -2023-02-13 17:49:31,638 - Epoch: [70][ 680/ 1207] Overall Loss 0.329392 Objective Loss 0.329392 LR 0.001000 Time 0.020286 -2023-02-13 17:49:31,846 - Epoch: [70][ 690/ 1207] Overall Loss 0.329262 Objective Loss 0.329262 LR 0.001000 Time 0.020292 -2023-02-13 17:49:32,056 - Epoch: [70][ 700/ 1207] Overall Loss 0.329520 Objective Loss 0.329520 LR 0.001000 Time 0.020303 -2023-02-13 17:49:32,263 - Epoch: [70][ 710/ 1207] Overall Loss 0.329603 Objective Loss 0.329603 LR 0.001000 Time 0.020307 -2023-02-13 17:49:32,473 - Epoch: [70][ 720/ 1207] Overall Loss 0.329727 Objective Loss 0.329727 LR 0.001000 Time 0.020316 -2023-02-13 17:49:32,680 - Epoch: [70][ 730/ 1207] Overall Loss 0.330374 Objective Loss 0.330374 LR 0.001000 Time 0.020321 -2023-02-13 17:49:32,890 - Epoch: [70][ 740/ 1207] Overall Loss 0.330439 Objective Loss 0.330439 LR 0.001000 Time 0.020330 -2023-02-13 17:49:33,098 - Epoch: [70][ 750/ 1207] Overall Loss 0.330651 Objective Loss 0.330651 LR 0.001000 Time 0.020335 -2023-02-13 17:49:33,307 - Epoch: [70][ 760/ 1207] Overall Loss 0.330993 Objective Loss 0.330993 LR 0.001000 Time 0.020343 -2023-02-13 17:49:33,515 - Epoch: [70][ 770/ 1207] Overall Loss 0.331492 Objective Loss 0.331492 LR 0.001000 Time 0.020348 -2023-02-13 17:49:33,711 - Epoch: [70][ 780/ 1207] Overall Loss 0.331555 Objective Loss 0.331555 LR 0.001000 Time 0.020338 -2023-02-13 17:49:33,909 - Epoch: [70][ 790/ 1207] Overall Loss 0.331488 Objective Loss 0.331488 LR 0.001000 Time 0.020331 -2023-02-13 17:49:34,105 - Epoch: [70][ 800/ 1207] Overall Loss 0.331384 Objective Loss 0.331384 LR 0.001000 Time 0.020322 -2023-02-13 17:49:34,304 - Epoch: [70][ 810/ 1207] Overall Loss 0.331376 Objective Loss 0.331376 LR 0.001000 Time 0.020315 -2023-02-13 17:49:34,500 - Epoch: [70][ 820/ 1207] Overall Loss 0.331269 Objective Loss 0.331269 LR 0.001000 Time 0.020306 -2023-02-13 17:49:34,698 - Epoch: [70][ 830/ 1207] Overall Loss 0.331437 Objective Loss 0.331437 LR 0.001000 Time 0.020300 -2023-02-13 17:49:34,894 - Epoch: [70][ 840/ 1207] Overall Loss 0.331462 Objective Loss 0.331462 LR 0.001000 Time 0.020291 -2023-02-13 17:49:35,092 - Epoch: [70][ 850/ 1207] Overall Loss 0.331451 Objective Loss 0.331451 LR 0.001000 Time 0.020285 -2023-02-13 17:49:35,291 - Epoch: [70][ 860/ 1207] Overall Loss 0.331142 Objective Loss 0.331142 LR 0.001000 Time 0.020280 -2023-02-13 17:49:35,489 - Epoch: [70][ 870/ 1207] Overall Loss 0.331058 Objective Loss 0.331058 LR 0.001000 Time 0.020273 -2023-02-13 17:49:35,688 - Epoch: [70][ 880/ 1207] Overall Loss 0.331212 Objective Loss 0.331212 LR 0.001000 Time 0.020269 -2023-02-13 17:49:35,886 - Epoch: [70][ 890/ 1207] Overall Loss 0.331185 Objective Loss 0.331185 LR 0.001000 Time 0.020264 -2023-02-13 17:49:36,086 - Epoch: [70][ 900/ 1207] Overall Loss 0.331140 Objective Loss 0.331140 LR 0.001000 Time 0.020260 -2023-02-13 17:49:36,283 - Epoch: [70][ 910/ 1207] Overall Loss 0.331400 Objective Loss 0.331400 LR 0.001000 Time 0.020253 -2023-02-13 17:49:36,483 - Epoch: [70][ 920/ 1207] Overall Loss 0.331330 Objective Loss 0.331330 LR 0.001000 Time 0.020250 -2023-02-13 17:49:36,679 - Epoch: [70][ 930/ 1207] Overall Loss 0.331481 Objective Loss 0.331481 LR 0.001000 Time 0.020243 -2023-02-13 17:49:36,879 - Epoch: [70][ 940/ 1207] Overall Loss 0.331280 Objective Loss 0.331280 LR 0.001000 Time 0.020240 -2023-02-13 17:49:37,076 - Epoch: [70][ 950/ 1207] Overall Loss 0.331257 Objective Loss 0.331257 LR 0.001000 Time 0.020234 -2023-02-13 17:49:37,277 - Epoch: [70][ 960/ 1207] Overall Loss 0.331460 Objective Loss 0.331460 LR 0.001000 Time 0.020232 -2023-02-13 17:49:37,474 - Epoch: [70][ 970/ 1207] Overall Loss 0.331631 Objective Loss 0.331631 LR 0.001000 Time 0.020226 -2023-02-13 17:49:37,674 - Epoch: [70][ 980/ 1207] Overall Loss 0.331819 Objective Loss 0.331819 LR 0.001000 Time 0.020223 -2023-02-13 17:49:37,871 - Epoch: [70][ 990/ 1207] Overall Loss 0.332009 Objective Loss 0.332009 LR 0.001000 Time 0.020218 -2023-02-13 17:49:38,071 - Epoch: [70][ 1000/ 1207] Overall Loss 0.331937 Objective Loss 0.331937 LR 0.001000 Time 0.020216 -2023-02-13 17:49:38,268 - Epoch: [70][ 1010/ 1207] Overall Loss 0.332186 Objective Loss 0.332186 LR 0.001000 Time 0.020210 -2023-02-13 17:49:38,469 - Epoch: [70][ 1020/ 1207] Overall Loss 0.332061 Objective Loss 0.332061 LR 0.001000 Time 0.020209 -2023-02-13 17:49:38,666 - Epoch: [70][ 1030/ 1207] Overall Loss 0.332423 Objective Loss 0.332423 LR 0.001000 Time 0.020203 -2023-02-13 17:49:38,866 - Epoch: [70][ 1040/ 1207] Overall Loss 0.332416 Objective Loss 0.332416 LR 0.001000 Time 0.020201 -2023-02-13 17:49:39,062 - Epoch: [70][ 1050/ 1207] Overall Loss 0.332558 Objective Loss 0.332558 LR 0.001000 Time 0.020195 -2023-02-13 17:49:39,263 - Epoch: [70][ 1060/ 1207] Overall Loss 0.333005 Objective Loss 0.333005 LR 0.001000 Time 0.020193 -2023-02-13 17:49:39,461 - Epoch: [70][ 1070/ 1207] Overall Loss 0.332923 Objective Loss 0.332923 LR 0.001000 Time 0.020189 -2023-02-13 17:49:39,661 - Epoch: [70][ 1080/ 1207] Overall Loss 0.333102 Objective Loss 0.333102 LR 0.001000 Time 0.020187 -2023-02-13 17:49:39,857 - Epoch: [70][ 1090/ 1207] Overall Loss 0.332963 Objective Loss 0.332963 LR 0.001000 Time 0.020182 -2023-02-13 17:49:40,058 - Epoch: [70][ 1100/ 1207] Overall Loss 0.333207 Objective Loss 0.333207 LR 0.001000 Time 0.020181 -2023-02-13 17:49:40,255 - Epoch: [70][ 1110/ 1207] Overall Loss 0.333208 Objective Loss 0.333208 LR 0.001000 Time 0.020176 -2023-02-13 17:49:40,455 - Epoch: [70][ 1120/ 1207] Overall Loss 0.333160 Objective Loss 0.333160 LR 0.001000 Time 0.020174 -2023-02-13 17:49:40,652 - Epoch: [70][ 1130/ 1207] Overall Loss 0.333186 Objective Loss 0.333186 LR 0.001000 Time 0.020169 -2023-02-13 17:49:40,853 - Epoch: [70][ 1140/ 1207] Overall Loss 0.333149 Objective Loss 0.333149 LR 0.001000 Time 0.020168 -2023-02-13 17:49:41,048 - Epoch: [70][ 1150/ 1207] Overall Loss 0.333113 Objective Loss 0.333113 LR 0.001000 Time 0.020163 -2023-02-13 17:49:41,247 - Epoch: [70][ 1160/ 1207] Overall Loss 0.333197 Objective Loss 0.333197 LR 0.001000 Time 0.020160 -2023-02-13 17:49:41,444 - Epoch: [70][ 1170/ 1207] Overall Loss 0.333219 Objective Loss 0.333219 LR 0.001000 Time 0.020156 -2023-02-13 17:49:41,644 - Epoch: [70][ 1180/ 1207] Overall Loss 0.333123 Objective Loss 0.333123 LR 0.001000 Time 0.020154 -2023-02-13 17:49:41,841 - Epoch: [70][ 1190/ 1207] Overall Loss 0.333244 Objective Loss 0.333244 LR 0.001000 Time 0.020150 -2023-02-13 17:49:42,095 - Epoch: [70][ 1200/ 1207] Overall Loss 0.333170 Objective Loss 0.333170 LR 0.001000 Time 0.020193 -2023-02-13 17:49:42,210 - Epoch: [70][ 1207/ 1207] Overall Loss 0.333224 Objective Loss 0.333224 Top1 82.012195 Top5 98.170732 LR 0.001000 Time 0.020171 -2023-02-13 17:49:42,282 - --- validate (epoch=70)----------- -2023-02-13 17:49:42,282 - 34311 samples (256 per mini-batch) -2023-02-13 17:49:42,683 - Epoch: [70][ 10/ 135] Loss 0.366938 Top1 80.195312 Top5 96.601562 -2023-02-13 17:49:42,813 - Epoch: [70][ 20/ 135] Loss 0.361675 Top1 80.781250 Top5 96.660156 -2023-02-13 17:49:42,942 - Epoch: [70][ 30/ 135] Loss 0.364748 Top1 81.041667 Top5 96.796875 -2023-02-13 17:49:43,068 - Epoch: [70][ 40/ 135] Loss 0.360189 Top1 81.064453 Top5 96.757812 -2023-02-13 17:49:43,195 - Epoch: [70][ 50/ 135] Loss 0.364390 Top1 81.226562 Top5 96.796875 -2023-02-13 17:49:43,319 - Epoch: [70][ 60/ 135] Loss 0.357129 Top1 81.360677 Top5 96.790365 -2023-02-13 17:49:43,447 - Epoch: [70][ 70/ 135] Loss 0.352924 Top1 81.562500 Top5 96.847098 -2023-02-13 17:49:43,578 - Epoch: [70][ 80/ 135] Loss 0.357119 Top1 81.411133 Top5 96.835938 -2023-02-13 17:49:43,709 - Epoch: [70][ 90/ 135] Loss 0.361774 Top1 81.241319 Top5 96.827257 -2023-02-13 17:49:43,840 - Epoch: [70][ 100/ 135] Loss 0.361125 Top1 81.277344 Top5 96.808594 -2023-02-13 17:49:43,970 - Epoch: [70][ 110/ 135] Loss 0.363073 Top1 81.214489 Top5 96.800426 -2023-02-13 17:49:44,103 - Epoch: [70][ 120/ 135] Loss 0.365368 Top1 81.155599 Top5 96.783854 -2023-02-13 17:49:44,232 - Epoch: [70][ 130/ 135] Loss 0.365335 Top1 81.183894 Top5 96.793870 -2023-02-13 17:49:44,276 - Epoch: [70][ 135/ 135] Loss 0.364584 Top1 81.189706 Top5 96.773629 -2023-02-13 17:49:44,359 - ==> Top1: 81.190 Top5: 96.774 Loss: 0.365 - -2023-02-13 17:49:44,360 - ==> Confusion: -[[ 831 7 4 4 10 6 0 2 5 61 1 5 2 3 2 3 4 2 5 3 7] - [ 0 933 1 1 3 35 2 25 2 1 2 2 3 2 1 4 5 0 6 2 3] - [ 7 6 919 21 4 2 20 22 0 1 6 3 2 6 5 8 3 2 8 7 6] - [ 5 4 18 892 2 6 1 3 2 0 13 1 8 2 18 1 4 6 24 0 6] - [ 19 11 0 2 979 11 1 3 1 1 0 7 1 4 2 6 4 2 3 3 6] - [ 3 23 2 3 5 952 4 23 3 5 3 17 3 11 1 3 2 0 4 2 1] - [ 2 7 17 4 0 7 1022 10 1 2 4 0 3 2 0 4 0 4 4 3 3] - [ 0 10 6 1 4 25 5 929 1 1 3 4 1 1 0 0 1 4 21 5 2] - [ 22 3 1 1 1 0 1 1 869 50 11 4 0 12 25 2 2 0 3 0 1] - [ 91 2 3 0 8 0 0 0 32 846 0 1 0 17 4 0 1 1 3 1 2] - [ 2 5 2 6 2 4 1 5 15 2 964 3 1 11 3 1 2 1 12 3 6] - [ 0 5 0 0 2 17 1 5 2 1 1 917 26 2 2 5 2 5 2 8 2] - [ 0 0 1 2 1 3 0 2 2 1 0 53 854 0 2 8 2 18 3 1 6] - [ 6 5 2 0 4 14 1 4 11 21 9 7 3 915 5 6 4 1 0 3 3] - [ 12 10 3 17 6 5 0 3 15 6 4 4 6 4 968 1 2 5 15 0 6] - [ 3 5 5 1 5 2 5 2 0 0 0 15 6 3 1 962 5 16 0 5 5] - [ 2 10 1 2 12 6 0 2 1 1 0 3 4 6 1 13 978 1 2 5 11] - [ 4 3 0 5 1 4 3 0 4 1 2 12 29 1 1 8 0 967 1 1 4] - [ 4 7 2 10 3 1 0 40 4 0 2 4 9 0 10 1 2 1 983 2 1] - [ 0 3 0 1 1 8 14 23 1 0 1 35 2 3 0 5 3 1 2 1036 9] - [ 160 356 208 201 143 260 106 320 101 113 215 176 367 330 166 126 282 122 223 318 9141]] - -2023-02-13 17:49:44,361 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:49:44,361 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:49:44,367 - - -2023-02-13 17:49:44,367 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:49:45,349 - Epoch: [71][ 10/ 1207] Overall Loss 0.305489 Objective Loss 0.305489 LR 0.001000 Time 0.098108 -2023-02-13 17:49:45,547 - Epoch: [71][ 20/ 1207] Overall Loss 0.326209 Objective Loss 0.326209 LR 0.001000 Time 0.058924 -2023-02-13 17:49:45,742 - Epoch: [71][ 30/ 1207] Overall Loss 0.330092 Objective Loss 0.330092 LR 0.001000 Time 0.045755 -2023-02-13 17:49:45,938 - Epoch: [71][ 40/ 1207] Overall Loss 0.333715 Objective Loss 0.333715 LR 0.001000 Time 0.039227 -2023-02-13 17:49:46,133 - Epoch: [71][ 50/ 1207] Overall Loss 0.336892 Objective Loss 0.336892 LR 0.001000 Time 0.035259 -2023-02-13 17:49:46,327 - Epoch: [71][ 60/ 1207] Overall Loss 0.332514 Objective Loss 0.332514 LR 0.001000 Time 0.032609 -2023-02-13 17:49:46,522 - Epoch: [71][ 70/ 1207] Overall Loss 0.331988 Objective Loss 0.331988 LR 0.001000 Time 0.030734 -2023-02-13 17:49:46,716 - Epoch: [71][ 80/ 1207] Overall Loss 0.334194 Objective Loss 0.334194 LR 0.001000 Time 0.029317 -2023-02-13 17:49:46,913 - Epoch: [71][ 90/ 1207] Overall Loss 0.328218 Objective Loss 0.328218 LR 0.001000 Time 0.028247 -2023-02-13 17:49:47,114 - Epoch: [71][ 100/ 1207] Overall Loss 0.329914 Objective Loss 0.329914 LR 0.001000 Time 0.027423 -2023-02-13 17:49:47,311 - Epoch: [71][ 110/ 1207] Overall Loss 0.329410 Objective Loss 0.329410 LR 0.001000 Time 0.026714 -2023-02-13 17:49:47,511 - Epoch: [71][ 120/ 1207] Overall Loss 0.329075 Objective Loss 0.329075 LR 0.001000 Time 0.026152 -2023-02-13 17:49:47,705 - Epoch: [71][ 130/ 1207] Overall Loss 0.328259 Objective Loss 0.328259 LR 0.001000 Time 0.025633 -2023-02-13 17:49:47,899 - Epoch: [71][ 140/ 1207] Overall Loss 0.330202 Objective Loss 0.330202 LR 0.001000 Time 0.025185 -2023-02-13 17:49:48,094 - Epoch: [71][ 150/ 1207] Overall Loss 0.327990 Objective Loss 0.327990 LR 0.001000 Time 0.024802 -2023-02-13 17:49:48,288 - Epoch: [71][ 160/ 1207] Overall Loss 0.324854 Objective Loss 0.324854 LR 0.001000 Time 0.024464 -2023-02-13 17:49:48,483 - Epoch: [71][ 170/ 1207] Overall Loss 0.324894 Objective Loss 0.324894 LR 0.001000 Time 0.024171 -2023-02-13 17:49:48,675 - Epoch: [71][ 180/ 1207] Overall Loss 0.324755 Objective Loss 0.324755 LR 0.001000 Time 0.023893 -2023-02-13 17:49:48,865 - Epoch: [71][ 190/ 1207] Overall Loss 0.323203 Objective Loss 0.323203 LR 0.001000 Time 0.023630 -2023-02-13 17:49:49,053 - Epoch: [71][ 200/ 1207] Overall Loss 0.323444 Objective Loss 0.323444 LR 0.001000 Time 0.023388 -2023-02-13 17:49:49,241 - Epoch: [71][ 210/ 1207] Overall Loss 0.323670 Objective Loss 0.323670 LR 0.001000 Time 0.023170 -2023-02-13 17:49:49,430 - Epoch: [71][ 220/ 1207] Overall Loss 0.323374 Objective Loss 0.323374 LR 0.001000 Time 0.022971 -2023-02-13 17:49:49,618 - Epoch: [71][ 230/ 1207] Overall Loss 0.322986 Objective Loss 0.322986 LR 0.001000 Time 0.022791 -2023-02-13 17:49:49,806 - Epoch: [71][ 240/ 1207] Overall Loss 0.322638 Objective Loss 0.322638 LR 0.001000 Time 0.022623 -2023-02-13 17:49:49,995 - Epoch: [71][ 250/ 1207] Overall Loss 0.323770 Objective Loss 0.323770 LR 0.001000 Time 0.022472 -2023-02-13 17:49:50,183 - Epoch: [71][ 260/ 1207] Overall Loss 0.323329 Objective Loss 0.323329 LR 0.001000 Time 0.022329 -2023-02-13 17:49:50,373 - Epoch: [71][ 270/ 1207] Overall Loss 0.323504 Objective Loss 0.323504 LR 0.001000 Time 0.022202 -2023-02-13 17:49:50,562 - Epoch: [71][ 280/ 1207] Overall Loss 0.324464 Objective Loss 0.324464 LR 0.001000 Time 0.022083 -2023-02-13 17:49:50,750 - Epoch: [71][ 290/ 1207] Overall Loss 0.324820 Objective Loss 0.324820 LR 0.001000 Time 0.021971 -2023-02-13 17:49:50,940 - Epoch: [71][ 300/ 1207] Overall Loss 0.324803 Objective Loss 0.324803 LR 0.001000 Time 0.021871 -2023-02-13 17:49:51,129 - Epoch: [71][ 310/ 1207] Overall Loss 0.325222 Objective Loss 0.325222 LR 0.001000 Time 0.021773 -2023-02-13 17:49:51,318 - Epoch: [71][ 320/ 1207] Overall Loss 0.325834 Objective Loss 0.325834 LR 0.001000 Time 0.021681 -2023-02-13 17:49:51,507 - Epoch: [71][ 330/ 1207] Overall Loss 0.325893 Objective Loss 0.325893 LR 0.001000 Time 0.021598 -2023-02-13 17:49:51,696 - Epoch: [71][ 340/ 1207] Overall Loss 0.326188 Objective Loss 0.326188 LR 0.001000 Time 0.021516 -2023-02-13 17:49:51,885 - Epoch: [71][ 350/ 1207] Overall Loss 0.325707 Objective Loss 0.325707 LR 0.001000 Time 0.021440 -2023-02-13 17:49:52,074 - Epoch: [71][ 360/ 1207] Overall Loss 0.326391 Objective Loss 0.326391 LR 0.001000 Time 0.021368 -2023-02-13 17:49:52,262 - Epoch: [71][ 370/ 1207] Overall Loss 0.326311 Objective Loss 0.326311 LR 0.001000 Time 0.021298 -2023-02-13 17:49:52,451 - Epoch: [71][ 380/ 1207] Overall Loss 0.326663 Objective Loss 0.326663 LR 0.001000 Time 0.021233 -2023-02-13 17:49:52,639 - Epoch: [71][ 390/ 1207] Overall Loss 0.326743 Objective Loss 0.326743 LR 0.001000 Time 0.021171 -2023-02-13 17:49:52,827 - Epoch: [71][ 400/ 1207] Overall Loss 0.326640 Objective Loss 0.326640 LR 0.001000 Time 0.021111 -2023-02-13 17:49:53,016 - Epoch: [71][ 410/ 1207] Overall Loss 0.326382 Objective Loss 0.326382 LR 0.001000 Time 0.021056 -2023-02-13 17:49:53,205 - Epoch: [71][ 420/ 1207] Overall Loss 0.326165 Objective Loss 0.326165 LR 0.001000 Time 0.021003 -2023-02-13 17:49:53,393 - Epoch: [71][ 430/ 1207] Overall Loss 0.326711 Objective Loss 0.326711 LR 0.001000 Time 0.020952 -2023-02-13 17:49:53,582 - Epoch: [71][ 440/ 1207] Overall Loss 0.326922 Objective Loss 0.326922 LR 0.001000 Time 0.020905 -2023-02-13 17:49:53,771 - Epoch: [71][ 450/ 1207] Overall Loss 0.327668 Objective Loss 0.327668 LR 0.001000 Time 0.020859 -2023-02-13 17:49:53,960 - Epoch: [71][ 460/ 1207] Overall Loss 0.328260 Objective Loss 0.328260 LR 0.001000 Time 0.020815 -2023-02-13 17:49:54,149 - Epoch: [71][ 470/ 1207] Overall Loss 0.328217 Objective Loss 0.328217 LR 0.001000 Time 0.020774 -2023-02-13 17:49:54,338 - Epoch: [71][ 480/ 1207] Overall Loss 0.328479 Objective Loss 0.328479 LR 0.001000 Time 0.020733 -2023-02-13 17:49:54,529 - Epoch: [71][ 490/ 1207] Overall Loss 0.328603 Objective Loss 0.328603 LR 0.001000 Time 0.020701 -2023-02-13 17:49:54,718 - Epoch: [71][ 500/ 1207] Overall Loss 0.328628 Objective Loss 0.328628 LR 0.001000 Time 0.020663 -2023-02-13 17:49:54,908 - Epoch: [71][ 510/ 1207] Overall Loss 0.328775 Objective Loss 0.328775 LR 0.001000 Time 0.020630 -2023-02-13 17:49:55,096 - Epoch: [71][ 520/ 1207] Overall Loss 0.329070 Objective Loss 0.329070 LR 0.001000 Time 0.020594 -2023-02-13 17:49:55,286 - Epoch: [71][ 530/ 1207] Overall Loss 0.328979 Objective Loss 0.328979 LR 0.001000 Time 0.020564 -2023-02-13 17:49:55,478 - Epoch: [71][ 540/ 1207] Overall Loss 0.329309 Objective Loss 0.329309 LR 0.001000 Time 0.020536 -2023-02-13 17:49:55,667 - Epoch: [71][ 550/ 1207] Overall Loss 0.329039 Objective Loss 0.329039 LR 0.001000 Time 0.020507 -2023-02-13 17:49:55,860 - Epoch: [71][ 560/ 1207] Overall Loss 0.329237 Objective Loss 0.329237 LR 0.001000 Time 0.020484 -2023-02-13 17:49:56,050 - Epoch: [71][ 570/ 1207] Overall Loss 0.329334 Objective Loss 0.329334 LR 0.001000 Time 0.020458 -2023-02-13 17:49:56,240 - Epoch: [71][ 580/ 1207] Overall Loss 0.329473 Objective Loss 0.329473 LR 0.001000 Time 0.020433 -2023-02-13 17:49:56,431 - Epoch: [71][ 590/ 1207] Overall Loss 0.329647 Objective Loss 0.329647 LR 0.001000 Time 0.020409 -2023-02-13 17:49:56,621 - Epoch: [71][ 600/ 1207] Overall Loss 0.330133 Objective Loss 0.330133 LR 0.001000 Time 0.020384 -2023-02-13 17:49:56,811 - Epoch: [71][ 610/ 1207] Overall Loss 0.329584 Objective Loss 0.329584 LR 0.001000 Time 0.020361 -2023-02-13 17:49:57,000 - Epoch: [71][ 620/ 1207] Overall Loss 0.329712 Objective Loss 0.329712 LR 0.001000 Time 0.020338 -2023-02-13 17:49:57,189 - Epoch: [71][ 630/ 1207] Overall Loss 0.330104 Objective Loss 0.330104 LR 0.001000 Time 0.020314 -2023-02-13 17:49:57,379 - Epoch: [71][ 640/ 1207] Overall Loss 0.330143 Objective Loss 0.330143 LR 0.001000 Time 0.020293 -2023-02-13 17:49:57,569 - Epoch: [71][ 650/ 1207] Overall Loss 0.330114 Objective Loss 0.330114 LR 0.001000 Time 0.020272 -2023-02-13 17:49:57,758 - Epoch: [71][ 660/ 1207] Overall Loss 0.330872 Objective Loss 0.330872 LR 0.001000 Time 0.020251 -2023-02-13 17:49:57,948 - Epoch: [71][ 670/ 1207] Overall Loss 0.330963 Objective Loss 0.330963 LR 0.001000 Time 0.020231 -2023-02-13 17:49:58,138 - Epoch: [71][ 680/ 1207] Overall Loss 0.331132 Objective Loss 0.331132 LR 0.001000 Time 0.020213 -2023-02-13 17:49:58,327 - Epoch: [71][ 690/ 1207] Overall Loss 0.331929 Objective Loss 0.331929 LR 0.001000 Time 0.020194 -2023-02-13 17:49:58,517 - Epoch: [71][ 700/ 1207] Overall Loss 0.331674 Objective Loss 0.331674 LR 0.001000 Time 0.020176 -2023-02-13 17:49:58,707 - Epoch: [71][ 710/ 1207] Overall Loss 0.331700 Objective Loss 0.331700 LR 0.001000 Time 0.020159 -2023-02-13 17:49:58,897 - Epoch: [71][ 720/ 1207] Overall Loss 0.331887 Objective Loss 0.331887 LR 0.001000 Time 0.020143 -2023-02-13 17:49:59,087 - Epoch: [71][ 730/ 1207] Overall Loss 0.331672 Objective Loss 0.331672 LR 0.001000 Time 0.020126 -2023-02-13 17:49:59,277 - Epoch: [71][ 740/ 1207] Overall Loss 0.331696 Objective Loss 0.331696 LR 0.001000 Time 0.020110 -2023-02-13 17:49:59,467 - Epoch: [71][ 750/ 1207] Overall Loss 0.331834 Objective Loss 0.331834 LR 0.001000 Time 0.020095 -2023-02-13 17:49:59,658 - Epoch: [71][ 760/ 1207] Overall Loss 0.331939 Objective Loss 0.331939 LR 0.001000 Time 0.020081 -2023-02-13 17:49:59,847 - Epoch: [71][ 770/ 1207] Overall Loss 0.332365 Objective Loss 0.332365 LR 0.001000 Time 0.020066 -2023-02-13 17:50:00,037 - Epoch: [71][ 780/ 1207] Overall Loss 0.331953 Objective Loss 0.331953 LR 0.001000 Time 0.020052 -2023-02-13 17:50:00,227 - Epoch: [71][ 790/ 1207] Overall Loss 0.331927 Objective Loss 0.331927 LR 0.001000 Time 0.020037 -2023-02-13 17:50:00,416 - Epoch: [71][ 800/ 1207] Overall Loss 0.332313 Objective Loss 0.332313 LR 0.001000 Time 0.020023 -2023-02-13 17:50:00,607 - Epoch: [71][ 810/ 1207] Overall Loss 0.332582 Objective Loss 0.332582 LR 0.001000 Time 0.020011 -2023-02-13 17:50:00,800 - Epoch: [71][ 820/ 1207] Overall Loss 0.332613 Objective Loss 0.332613 LR 0.001000 Time 0.020002 -2023-02-13 17:50:00,989 - Epoch: [71][ 830/ 1207] Overall Loss 0.332681 Objective Loss 0.332681 LR 0.001000 Time 0.019989 -2023-02-13 17:50:01,179 - Epoch: [71][ 840/ 1207] Overall Loss 0.332545 Objective Loss 0.332545 LR 0.001000 Time 0.019976 -2023-02-13 17:50:01,368 - Epoch: [71][ 850/ 1207] Overall Loss 0.332317 Objective Loss 0.332317 LR 0.001000 Time 0.019962 -2023-02-13 17:50:01,558 - Epoch: [71][ 860/ 1207] Overall Loss 0.332099 Objective Loss 0.332099 LR 0.001000 Time 0.019951 -2023-02-13 17:50:01,748 - Epoch: [71][ 870/ 1207] Overall Loss 0.331888 Objective Loss 0.331888 LR 0.001000 Time 0.019940 -2023-02-13 17:50:01,938 - Epoch: [71][ 880/ 1207] Overall Loss 0.331985 Objective Loss 0.331985 LR 0.001000 Time 0.019929 -2023-02-13 17:50:02,127 - Epoch: [71][ 890/ 1207] Overall Loss 0.332001 Objective Loss 0.332001 LR 0.001000 Time 0.019917 -2023-02-13 17:50:02,318 - Epoch: [71][ 900/ 1207] Overall Loss 0.332024 Objective Loss 0.332024 LR 0.001000 Time 0.019907 -2023-02-13 17:50:02,508 - Epoch: [71][ 910/ 1207] Overall Loss 0.331707 Objective Loss 0.331707 LR 0.001000 Time 0.019897 -2023-02-13 17:50:02,698 - Epoch: [71][ 920/ 1207] Overall Loss 0.331590 Objective Loss 0.331590 LR 0.001000 Time 0.019887 -2023-02-13 17:50:02,888 - Epoch: [71][ 930/ 1207] Overall Loss 0.332224 Objective Loss 0.332224 LR 0.001000 Time 0.019877 -2023-02-13 17:50:03,078 - Epoch: [71][ 940/ 1207] Overall Loss 0.332034 Objective Loss 0.332034 LR 0.001000 Time 0.019867 -2023-02-13 17:50:03,268 - Epoch: [71][ 950/ 1207] Overall Loss 0.332017 Objective Loss 0.332017 LR 0.001000 Time 0.019857 -2023-02-13 17:50:03,459 - Epoch: [71][ 960/ 1207] Overall Loss 0.331946 Objective Loss 0.331946 LR 0.001000 Time 0.019849 -2023-02-13 17:50:03,649 - Epoch: [71][ 970/ 1207] Overall Loss 0.331939 Objective Loss 0.331939 LR 0.001000 Time 0.019840 -2023-02-13 17:50:03,839 - Epoch: [71][ 980/ 1207] Overall Loss 0.331861 Objective Loss 0.331861 LR 0.001000 Time 0.019831 -2023-02-13 17:50:04,029 - Epoch: [71][ 990/ 1207] Overall Loss 0.332074 Objective Loss 0.332074 LR 0.001000 Time 0.019823 -2023-02-13 17:50:04,219 - Epoch: [71][ 1000/ 1207] Overall Loss 0.332118 Objective Loss 0.332118 LR 0.001000 Time 0.019814 -2023-02-13 17:50:04,409 - Epoch: [71][ 1010/ 1207] Overall Loss 0.332466 Objective Loss 0.332466 LR 0.001000 Time 0.019805 -2023-02-13 17:50:04,599 - Epoch: [71][ 1020/ 1207] Overall Loss 0.332733 Objective Loss 0.332733 LR 0.001000 Time 0.019798 -2023-02-13 17:50:04,789 - Epoch: [71][ 1030/ 1207] Overall Loss 0.332496 Objective Loss 0.332496 LR 0.001000 Time 0.019789 -2023-02-13 17:50:04,979 - Epoch: [71][ 1040/ 1207] Overall Loss 0.332338 Objective Loss 0.332338 LR 0.001000 Time 0.019781 -2023-02-13 17:50:05,169 - Epoch: [71][ 1050/ 1207] Overall Loss 0.332351 Objective Loss 0.332351 LR 0.001000 Time 0.019773 -2023-02-13 17:50:05,359 - Epoch: [71][ 1060/ 1207] Overall Loss 0.332604 Objective Loss 0.332604 LR 0.001000 Time 0.019766 -2023-02-13 17:50:05,550 - Epoch: [71][ 1070/ 1207] Overall Loss 0.332511 Objective Loss 0.332511 LR 0.001000 Time 0.019759 -2023-02-13 17:50:05,740 - Epoch: [71][ 1080/ 1207] Overall Loss 0.332498 Objective Loss 0.332498 LR 0.001000 Time 0.019752 -2023-02-13 17:50:05,932 - Epoch: [71][ 1090/ 1207] Overall Loss 0.332629 Objective Loss 0.332629 LR 0.001000 Time 0.019746 -2023-02-13 17:50:06,122 - Epoch: [71][ 1100/ 1207] Overall Loss 0.332656 Objective Loss 0.332656 LR 0.001000 Time 0.019739 -2023-02-13 17:50:06,311 - Epoch: [71][ 1110/ 1207] Overall Loss 0.332688 Objective Loss 0.332688 LR 0.001000 Time 0.019732 -2023-02-13 17:50:06,502 - Epoch: [71][ 1120/ 1207] Overall Loss 0.332603 Objective Loss 0.332603 LR 0.001000 Time 0.019725 -2023-02-13 17:50:06,693 - Epoch: [71][ 1130/ 1207] Overall Loss 0.332858 Objective Loss 0.332858 LR 0.001000 Time 0.019719 -2023-02-13 17:50:06,883 - Epoch: [71][ 1140/ 1207] Overall Loss 0.332708 Objective Loss 0.332708 LR 0.001000 Time 0.019713 -2023-02-13 17:50:07,074 - Epoch: [71][ 1150/ 1207] Overall Loss 0.332778 Objective Loss 0.332778 LR 0.001000 Time 0.019707 -2023-02-13 17:50:07,263 - Epoch: [71][ 1160/ 1207] Overall Loss 0.332598 Objective Loss 0.332598 LR 0.001000 Time 0.019700 -2023-02-13 17:50:07,453 - Epoch: [71][ 1170/ 1207] Overall Loss 0.332366 Objective Loss 0.332366 LR 0.001000 Time 0.019694 -2023-02-13 17:50:07,643 - Epoch: [71][ 1180/ 1207] Overall Loss 0.332279 Objective Loss 0.332279 LR 0.001000 Time 0.019687 -2023-02-13 17:50:07,833 - Epoch: [71][ 1190/ 1207] Overall Loss 0.332256 Objective Loss 0.332256 LR 0.001000 Time 0.019681 -2023-02-13 17:50:08,074 - Epoch: [71][ 1200/ 1207] Overall Loss 0.332307 Objective Loss 0.332307 LR 0.001000 Time 0.019718 -2023-02-13 17:50:08,189 - Epoch: [71][ 1207/ 1207] Overall Loss 0.332539 Objective Loss 0.332539 Top1 81.707317 Top5 97.560976 LR 0.001000 Time 0.019699 -2023-02-13 17:50:08,261 - --- validate (epoch=71)----------- -2023-02-13 17:50:08,261 - 34311 samples (256 per mini-batch) -2023-02-13 17:50:08,661 - Epoch: [71][ 10/ 135] Loss 0.360204 Top1 82.773438 Top5 96.953125 -2023-02-13 17:50:08,786 - Epoch: [71][ 20/ 135] Loss 0.366951 Top1 82.167969 Top5 96.972656 -2023-02-13 17:50:08,915 - Epoch: [71][ 30/ 135] Loss 0.355358 Top1 82.343750 Top5 97.265625 -2023-02-13 17:50:09,041 - Epoch: [71][ 40/ 135] Loss 0.352814 Top1 82.041016 Top5 97.265625 -2023-02-13 17:50:09,167 - Epoch: [71][ 50/ 135] Loss 0.350433 Top1 82.015625 Top5 97.210938 -2023-02-13 17:50:09,292 - Epoch: [71][ 60/ 135] Loss 0.343457 Top1 81.861979 Top5 97.174479 -2023-02-13 17:50:09,420 - Epoch: [71][ 70/ 135] Loss 0.348967 Top1 81.623884 Top5 97.059152 -2023-02-13 17:50:09,544 - Epoch: [71][ 80/ 135] Loss 0.354176 Top1 81.777344 Top5 97.119141 -2023-02-13 17:50:09,671 - Epoch: [71][ 90/ 135] Loss 0.356241 Top1 81.649306 Top5 97.087674 -2023-02-13 17:50:09,795 - Epoch: [71][ 100/ 135] Loss 0.358522 Top1 81.578125 Top5 97.078125 -2023-02-13 17:50:09,919 - Epoch: [71][ 110/ 135] Loss 0.360208 Top1 81.544744 Top5 97.084517 -2023-02-13 17:50:10,046 - Epoch: [71][ 120/ 135] Loss 0.360195 Top1 81.539714 Top5 97.135417 -2023-02-13 17:50:10,178 - Epoch: [71][ 130/ 135] Loss 0.357557 Top1 81.634615 Top5 97.172476 -2023-02-13 17:50:10,223 - Epoch: [71][ 135/ 135] Loss 0.361559 Top1 81.597738 Top5 97.149602 -2023-02-13 17:50:10,295 - ==> Top1: 81.598 Top5: 97.150 Loss: 0.362 - -2023-02-13 17:50:10,296 - ==> Confusion: -[[ 863 6 4 2 6 1 0 0 3 43 0 5 1 3 3 3 8 2 2 2 10] - [ 1 937 4 3 8 18 2 16 8 2 3 1 1 0 3 3 5 1 9 1 7] - [ 13 2 939 12 4 1 15 13 0 1 4 1 1 6 7 5 5 2 18 7 2] - [ 5 1 23 884 2 3 0 1 3 3 17 0 7 1 24 4 3 8 20 2 5] - [ 25 10 0 3 971 4 0 1 2 4 0 4 3 6 6 7 9 2 0 2 7] - [ 8 30 2 1 9 936 4 21 1 7 1 11 6 9 0 3 5 2 6 5 3] - [ 4 4 23 2 0 3 1016 7 0 0 5 2 1 3 0 9 4 4 5 6 1] - [ 1 12 9 1 4 25 4 891 1 1 6 7 2 1 0 0 1 1 42 10 5] - [ 20 4 0 3 0 0 0 1 868 45 11 1 3 7 26 3 0 2 14 0 1] - [ 105 1 4 1 3 0 0 1 31 832 0 2 0 18 4 1 2 2 1 0 4] - [ 2 1 3 2 1 2 3 3 27 2 963 3 2 11 3 1 1 2 15 3 1] - [ 3 1 0 0 3 17 2 7 0 0 0 871 46 10 1 6 2 15 6 14 1] - [ 0 0 0 4 1 1 0 1 2 0 0 18 872 1 3 12 2 31 1 4 6] - [ 3 5 0 0 8 15 1 3 13 33 7 6 5 896 5 7 5 2 0 2 8] - [ 6 1 3 13 2 4 0 3 22 3 3 1 4 3 994 3 1 7 13 1 5] - [ 3 1 8 1 6 0 0 1 0 0 0 5 10 3 0 971 10 12 1 8 6] - [ 5 7 2 0 6 3 0 0 4 0 0 1 4 2 1 7 1003 1 2 4 9] - [ 5 2 2 5 1 1 0 0 2 0 1 5 14 2 1 17 1 990 0 0 2] - [ 4 6 2 9 1 0 0 24 4 1 3 2 7 0 15 1 3 1 1002 0 1] - [ 0 3 1 1 2 6 11 17 1 0 3 13 4 3 0 7 8 2 1 1059 6] - [ 215 296 248 133 142 170 72 162 147 117 236 141 396 294 203 174 353 135 228 333 9239]] - -2023-02-13 17:50:10,297 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:50:10,297 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:50:10,303 - - -2023-02-13 17:50:10,303 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:50:11,291 - Epoch: [72][ 10/ 1207] Overall Loss 0.307806 Objective Loss 0.307806 LR 0.001000 Time 0.098744 -2023-02-13 17:50:11,487 - Epoch: [72][ 20/ 1207] Overall Loss 0.325778 Objective Loss 0.325778 LR 0.001000 Time 0.059138 -2023-02-13 17:50:11,678 - Epoch: [72][ 30/ 1207] Overall Loss 0.328301 Objective Loss 0.328301 LR 0.001000 Time 0.045764 -2023-02-13 17:50:11,868 - Epoch: [72][ 40/ 1207] Overall Loss 0.328401 Objective Loss 0.328401 LR 0.001000 Time 0.039063 -2023-02-13 17:50:12,057 - Epoch: [72][ 50/ 1207] Overall Loss 0.331483 Objective Loss 0.331483 LR 0.001000 Time 0.035029 -2023-02-13 17:50:12,247 - Epoch: [72][ 60/ 1207] Overall Loss 0.326802 Objective Loss 0.326802 LR 0.001000 Time 0.032351 -2023-02-13 17:50:12,437 - Epoch: [72][ 70/ 1207] Overall Loss 0.329413 Objective Loss 0.329413 LR 0.001000 Time 0.030433 -2023-02-13 17:50:12,627 - Epoch: [72][ 80/ 1207] Overall Loss 0.332552 Objective Loss 0.332552 LR 0.001000 Time 0.029005 -2023-02-13 17:50:12,816 - Epoch: [72][ 90/ 1207] Overall Loss 0.329945 Objective Loss 0.329945 LR 0.001000 Time 0.027881 -2023-02-13 17:50:13,006 - Epoch: [72][ 100/ 1207] Overall Loss 0.329513 Objective Loss 0.329513 LR 0.001000 Time 0.026990 -2023-02-13 17:50:13,196 - Epoch: [72][ 110/ 1207] Overall Loss 0.327831 Objective Loss 0.327831 LR 0.001000 Time 0.026258 -2023-02-13 17:50:13,386 - Epoch: [72][ 120/ 1207] Overall Loss 0.328158 Objective Loss 0.328158 LR 0.001000 Time 0.025648 -2023-02-13 17:50:13,575 - Epoch: [72][ 130/ 1207] Overall Loss 0.328121 Objective Loss 0.328121 LR 0.001000 Time 0.025129 -2023-02-13 17:50:13,766 - Epoch: [72][ 140/ 1207] Overall Loss 0.329034 Objective Loss 0.329034 LR 0.001000 Time 0.024692 -2023-02-13 17:50:13,955 - Epoch: [72][ 150/ 1207] Overall Loss 0.328032 Objective Loss 0.328032 LR 0.001000 Time 0.024305 -2023-02-13 17:50:14,145 - Epoch: [72][ 160/ 1207] Overall Loss 0.328706 Objective Loss 0.328706 LR 0.001000 Time 0.023972 -2023-02-13 17:50:14,335 - Epoch: [72][ 170/ 1207] Overall Loss 0.330841 Objective Loss 0.330841 LR 0.001000 Time 0.023677 -2023-02-13 17:50:14,525 - Epoch: [72][ 180/ 1207] Overall Loss 0.330990 Objective Loss 0.330990 LR 0.001000 Time 0.023416 -2023-02-13 17:50:14,715 - Epoch: [72][ 190/ 1207] Overall Loss 0.333126 Objective Loss 0.333126 LR 0.001000 Time 0.023180 -2023-02-13 17:50:14,905 - Epoch: [72][ 200/ 1207] Overall Loss 0.331703 Objective Loss 0.331703 LR 0.001000 Time 0.022971 -2023-02-13 17:50:15,095 - Epoch: [72][ 210/ 1207] Overall Loss 0.333231 Objective Loss 0.333231 LR 0.001000 Time 0.022777 -2023-02-13 17:50:15,283 - Epoch: [72][ 220/ 1207] Overall Loss 0.333414 Objective Loss 0.333414 LR 0.001000 Time 0.022597 -2023-02-13 17:50:15,472 - Epoch: [72][ 230/ 1207] Overall Loss 0.334273 Objective Loss 0.334273 LR 0.001000 Time 0.022434 -2023-02-13 17:50:15,663 - Epoch: [72][ 240/ 1207] Overall Loss 0.334829 Objective Loss 0.334829 LR 0.001000 Time 0.022292 -2023-02-13 17:50:15,853 - Epoch: [72][ 250/ 1207] Overall Loss 0.334364 Objective Loss 0.334364 LR 0.001000 Time 0.022159 -2023-02-13 17:50:16,042 - Epoch: [72][ 260/ 1207] Overall Loss 0.333386 Objective Loss 0.333386 LR 0.001000 Time 0.022034 -2023-02-13 17:50:16,230 - Epoch: [72][ 270/ 1207] Overall Loss 0.332494 Objective Loss 0.332494 LR 0.001000 Time 0.021914 -2023-02-13 17:50:16,419 - Epoch: [72][ 280/ 1207] Overall Loss 0.332307 Objective Loss 0.332307 LR 0.001000 Time 0.021805 -2023-02-13 17:50:16,609 - Epoch: [72][ 290/ 1207] Overall Loss 0.332063 Objective Loss 0.332063 LR 0.001000 Time 0.021707 -2023-02-13 17:50:16,798 - Epoch: [72][ 300/ 1207] Overall Loss 0.331382 Objective Loss 0.331382 LR 0.001000 Time 0.021612 -2023-02-13 17:50:16,988 - Epoch: [72][ 310/ 1207] Overall Loss 0.331073 Objective Loss 0.331073 LR 0.001000 Time 0.021525 -2023-02-13 17:50:17,178 - Epoch: [72][ 320/ 1207] Overall Loss 0.331247 Objective Loss 0.331247 LR 0.001000 Time 0.021444 -2023-02-13 17:50:17,367 - Epoch: [72][ 330/ 1207] Overall Loss 0.332463 Objective Loss 0.332463 LR 0.001000 Time 0.021367 -2023-02-13 17:50:17,557 - Epoch: [72][ 340/ 1207] Overall Loss 0.331178 Objective Loss 0.331178 LR 0.001000 Time 0.021297 -2023-02-13 17:50:17,747 - Epoch: [72][ 350/ 1207] Overall Loss 0.331317 Objective Loss 0.331317 LR 0.001000 Time 0.021229 -2023-02-13 17:50:17,936 - Epoch: [72][ 360/ 1207] Overall Loss 0.330449 Objective Loss 0.330449 LR 0.001000 Time 0.021164 -2023-02-13 17:50:18,124 - Epoch: [72][ 370/ 1207] Overall Loss 0.330860 Objective Loss 0.330860 LR 0.001000 Time 0.021100 -2023-02-13 17:50:18,314 - Epoch: [72][ 380/ 1207] Overall Loss 0.331086 Objective Loss 0.331086 LR 0.001000 Time 0.021042 -2023-02-13 17:50:18,504 - Epoch: [72][ 390/ 1207] Overall Loss 0.331314 Objective Loss 0.331314 LR 0.001000 Time 0.020988 -2023-02-13 17:50:18,694 - Epoch: [72][ 400/ 1207] Overall Loss 0.331293 Objective Loss 0.331293 LR 0.001000 Time 0.020938 -2023-02-13 17:50:18,883 - Epoch: [72][ 410/ 1207] Overall Loss 0.331456 Objective Loss 0.331456 LR 0.001000 Time 0.020889 -2023-02-13 17:50:19,074 - Epoch: [72][ 420/ 1207] Overall Loss 0.330639 Objective Loss 0.330639 LR 0.001000 Time 0.020843 -2023-02-13 17:50:19,262 - Epoch: [72][ 430/ 1207] Overall Loss 0.330594 Objective Loss 0.330594 LR 0.001000 Time 0.020797 -2023-02-13 17:50:19,452 - Epoch: [72][ 440/ 1207] Overall Loss 0.331468 Objective Loss 0.331468 LR 0.001000 Time 0.020755 -2023-02-13 17:50:19,643 - Epoch: [72][ 450/ 1207] Overall Loss 0.332170 Objective Loss 0.332170 LR 0.001000 Time 0.020716 -2023-02-13 17:50:19,832 - Epoch: [72][ 460/ 1207] Overall Loss 0.331609 Objective Loss 0.331609 LR 0.001000 Time 0.020677 -2023-02-13 17:50:20,022 - Epoch: [72][ 470/ 1207] Overall Loss 0.332094 Objective Loss 0.332094 LR 0.001000 Time 0.020639 -2023-02-13 17:50:20,211 - Epoch: [72][ 480/ 1207] Overall Loss 0.332351 Objective Loss 0.332351 LR 0.001000 Time 0.020603 -2023-02-13 17:50:20,399 - Epoch: [72][ 490/ 1207] Overall Loss 0.332290 Objective Loss 0.332290 LR 0.001000 Time 0.020566 -2023-02-13 17:50:20,590 - Epoch: [72][ 500/ 1207] Overall Loss 0.332172 Objective Loss 0.332172 LR 0.001000 Time 0.020535 -2023-02-13 17:50:20,779 - Epoch: [72][ 510/ 1207] Overall Loss 0.331623 Objective Loss 0.331623 LR 0.001000 Time 0.020502 -2023-02-13 17:50:20,970 - Epoch: [72][ 520/ 1207] Overall Loss 0.331331 Objective Loss 0.331331 LR 0.001000 Time 0.020474 -2023-02-13 17:50:21,159 - Epoch: [72][ 530/ 1207] Overall Loss 0.331325 Objective Loss 0.331325 LR 0.001000 Time 0.020445 -2023-02-13 17:50:21,349 - Epoch: [72][ 540/ 1207] Overall Loss 0.330791 Objective Loss 0.330791 LR 0.001000 Time 0.020416 -2023-02-13 17:50:21,538 - Epoch: [72][ 550/ 1207] Overall Loss 0.330307 Objective Loss 0.330307 LR 0.001000 Time 0.020389 -2023-02-13 17:50:21,728 - Epoch: [72][ 560/ 1207] Overall Loss 0.330327 Objective Loss 0.330327 LR 0.001000 Time 0.020364 -2023-02-13 17:50:21,918 - Epoch: [72][ 570/ 1207] Overall Loss 0.329913 Objective Loss 0.329913 LR 0.001000 Time 0.020339 -2023-02-13 17:50:22,109 - Epoch: [72][ 580/ 1207] Overall Loss 0.329155 Objective Loss 0.329155 LR 0.001000 Time 0.020316 -2023-02-13 17:50:22,297 - Epoch: [72][ 590/ 1207] Overall Loss 0.329681 Objective Loss 0.329681 LR 0.001000 Time 0.020290 -2023-02-13 17:50:22,485 - Epoch: [72][ 600/ 1207] Overall Loss 0.329701 Objective Loss 0.329701 LR 0.001000 Time 0.020265 -2023-02-13 17:50:22,675 - Epoch: [72][ 610/ 1207] Overall Loss 0.330101 Objective Loss 0.330101 LR 0.001000 Time 0.020244 -2023-02-13 17:50:22,865 - Epoch: [72][ 620/ 1207] Overall Loss 0.330153 Objective Loss 0.330153 LR 0.001000 Time 0.020223 -2023-02-13 17:50:23,053 - Epoch: [72][ 630/ 1207] Overall Loss 0.329654 Objective Loss 0.329654 LR 0.001000 Time 0.020200 -2023-02-13 17:50:23,242 - Epoch: [72][ 640/ 1207] Overall Loss 0.329272 Objective Loss 0.329272 LR 0.001000 Time 0.020179 -2023-02-13 17:50:23,430 - Epoch: [72][ 650/ 1207] Overall Loss 0.329281 Objective Loss 0.329281 LR 0.001000 Time 0.020157 -2023-02-13 17:50:23,620 - Epoch: [72][ 660/ 1207] Overall Loss 0.329146 Objective Loss 0.329146 LR 0.001000 Time 0.020139 -2023-02-13 17:50:23,809 - Epoch: [72][ 670/ 1207] Overall Loss 0.329287 Objective Loss 0.329287 LR 0.001000 Time 0.020120 -2023-02-13 17:50:23,999 - Epoch: [72][ 680/ 1207] Overall Loss 0.329545 Objective Loss 0.329545 LR 0.001000 Time 0.020102 -2023-02-13 17:50:24,188 - Epoch: [72][ 690/ 1207] Overall Loss 0.329455 Objective Loss 0.329455 LR 0.001000 Time 0.020085 -2023-02-13 17:50:24,377 - Epoch: [72][ 700/ 1207] Overall Loss 0.329125 Objective Loss 0.329125 LR 0.001000 Time 0.020068 -2023-02-13 17:50:24,568 - Epoch: [72][ 710/ 1207] Overall Loss 0.328793 Objective Loss 0.328793 LR 0.001000 Time 0.020053 -2023-02-13 17:50:24,758 - Epoch: [72][ 720/ 1207] Overall Loss 0.328664 Objective Loss 0.328664 LR 0.001000 Time 0.020038 -2023-02-13 17:50:24,947 - Epoch: [72][ 730/ 1207] Overall Loss 0.328901 Objective Loss 0.328901 LR 0.001000 Time 0.020022 -2023-02-13 17:50:25,137 - Epoch: [72][ 740/ 1207] Overall Loss 0.329043 Objective Loss 0.329043 LR 0.001000 Time 0.020008 -2023-02-13 17:50:25,326 - Epoch: [72][ 750/ 1207] Overall Loss 0.329058 Objective Loss 0.329058 LR 0.001000 Time 0.019992 -2023-02-13 17:50:25,515 - Epoch: [72][ 760/ 1207] Overall Loss 0.329505 Objective Loss 0.329505 LR 0.001000 Time 0.019977 -2023-02-13 17:50:25,705 - Epoch: [72][ 770/ 1207] Overall Loss 0.329709 Objective Loss 0.329709 LR 0.001000 Time 0.019964 -2023-02-13 17:50:25,896 - Epoch: [72][ 780/ 1207] Overall Loss 0.330058 Objective Loss 0.330058 LR 0.001000 Time 0.019952 -2023-02-13 17:50:26,085 - Epoch: [72][ 790/ 1207] Overall Loss 0.330164 Objective Loss 0.330164 LR 0.001000 Time 0.019939 -2023-02-13 17:50:26,274 - Epoch: [72][ 800/ 1207] Overall Loss 0.330261 Objective Loss 0.330261 LR 0.001000 Time 0.019925 -2023-02-13 17:50:26,463 - Epoch: [72][ 810/ 1207] Overall Loss 0.330782 Objective Loss 0.330782 LR 0.001000 Time 0.019912 -2023-02-13 17:50:26,654 - Epoch: [72][ 820/ 1207] Overall Loss 0.330780 Objective Loss 0.330780 LR 0.001000 Time 0.019901 -2023-02-13 17:50:26,843 - Epoch: [72][ 830/ 1207] Overall Loss 0.330840 Objective Loss 0.330840 LR 0.001000 Time 0.019890 -2023-02-13 17:50:27,033 - Epoch: [72][ 840/ 1207] Overall Loss 0.330704 Objective Loss 0.330704 LR 0.001000 Time 0.019879 -2023-02-13 17:50:27,222 - Epoch: [72][ 850/ 1207] Overall Loss 0.330824 Objective Loss 0.330824 LR 0.001000 Time 0.019866 -2023-02-13 17:50:27,411 - Epoch: [72][ 860/ 1207] Overall Loss 0.330958 Objective Loss 0.330958 LR 0.001000 Time 0.019854 -2023-02-13 17:50:27,601 - Epoch: [72][ 870/ 1207] Overall Loss 0.331092 Objective Loss 0.331092 LR 0.001000 Time 0.019845 -2023-02-13 17:50:27,792 - Epoch: [72][ 880/ 1207] Overall Loss 0.331049 Objective Loss 0.331049 LR 0.001000 Time 0.019835 -2023-02-13 17:50:27,982 - Epoch: [72][ 890/ 1207] Overall Loss 0.331442 Objective Loss 0.331442 LR 0.001000 Time 0.019825 -2023-02-13 17:50:28,172 - Epoch: [72][ 900/ 1207] Overall Loss 0.331283 Objective Loss 0.331283 LR 0.001000 Time 0.019816 -2023-02-13 17:50:28,361 - Epoch: [72][ 910/ 1207] Overall Loss 0.330853 Objective Loss 0.330853 LR 0.001000 Time 0.019806 -2023-02-13 17:50:28,551 - Epoch: [72][ 920/ 1207] Overall Loss 0.331005 Objective Loss 0.331005 LR 0.001000 Time 0.019797 -2023-02-13 17:50:28,741 - Epoch: [72][ 930/ 1207] Overall Loss 0.331214 Objective Loss 0.331214 LR 0.001000 Time 0.019788 -2023-02-13 17:50:28,931 - Epoch: [72][ 940/ 1207] Overall Loss 0.330963 Objective Loss 0.330963 LR 0.001000 Time 0.019778 -2023-02-13 17:50:29,120 - Epoch: [72][ 950/ 1207] Overall Loss 0.331415 Objective Loss 0.331415 LR 0.001000 Time 0.019769 -2023-02-13 17:50:29,309 - Epoch: [72][ 960/ 1207] Overall Loss 0.331668 Objective Loss 0.331668 LR 0.001000 Time 0.019760 -2023-02-13 17:50:29,499 - Epoch: [72][ 970/ 1207] Overall Loss 0.331973 Objective Loss 0.331973 LR 0.001000 Time 0.019751 -2023-02-13 17:50:29,689 - Epoch: [72][ 980/ 1207] Overall Loss 0.331822 Objective Loss 0.331822 LR 0.001000 Time 0.019743 -2023-02-13 17:50:29,878 - Epoch: [72][ 990/ 1207] Overall Loss 0.331699 Objective Loss 0.331699 LR 0.001000 Time 0.019734 -2023-02-13 17:50:30,068 - Epoch: [72][ 1000/ 1207] Overall Loss 0.331974 Objective Loss 0.331974 LR 0.001000 Time 0.019726 -2023-02-13 17:50:30,257 - Epoch: [72][ 1010/ 1207] Overall Loss 0.331901 Objective Loss 0.331901 LR 0.001000 Time 0.019718 -2023-02-13 17:50:30,446 - Epoch: [72][ 1020/ 1207] Overall Loss 0.332172 Objective Loss 0.332172 LR 0.001000 Time 0.019710 -2023-02-13 17:50:30,636 - Epoch: [72][ 1030/ 1207] Overall Loss 0.332076 Objective Loss 0.332076 LR 0.001000 Time 0.019702 -2023-02-13 17:50:30,828 - Epoch: [72][ 1040/ 1207] Overall Loss 0.332056 Objective Loss 0.332056 LR 0.001000 Time 0.019697 -2023-02-13 17:50:31,019 - Epoch: [72][ 1050/ 1207] Overall Loss 0.332204 Objective Loss 0.332204 LR 0.001000 Time 0.019691 -2023-02-13 17:50:31,209 - Epoch: [72][ 1060/ 1207] Overall Loss 0.332402 Objective Loss 0.332402 LR 0.001000 Time 0.019684 -2023-02-13 17:50:31,399 - Epoch: [72][ 1070/ 1207] Overall Loss 0.332470 Objective Loss 0.332470 LR 0.001000 Time 0.019677 -2023-02-13 17:50:31,589 - Epoch: [72][ 1080/ 1207] Overall Loss 0.332351 Objective Loss 0.332351 LR 0.001000 Time 0.019671 -2023-02-13 17:50:31,779 - Epoch: [72][ 1090/ 1207] Overall Loss 0.332560 Objective Loss 0.332560 LR 0.001000 Time 0.019664 -2023-02-13 17:50:31,970 - Epoch: [72][ 1100/ 1207] Overall Loss 0.332558 Objective Loss 0.332558 LR 0.001000 Time 0.019659 -2023-02-13 17:50:32,159 - Epoch: [72][ 1110/ 1207] Overall Loss 0.332680 Objective Loss 0.332680 LR 0.001000 Time 0.019652 -2023-02-13 17:50:32,349 - Epoch: [72][ 1120/ 1207] Overall Loss 0.332876 Objective Loss 0.332876 LR 0.001000 Time 0.019646 -2023-02-13 17:50:32,539 - Epoch: [72][ 1130/ 1207] Overall Loss 0.333102 Objective Loss 0.333102 LR 0.001000 Time 0.019640 -2023-02-13 17:50:32,730 - Epoch: [72][ 1140/ 1207] Overall Loss 0.333136 Objective Loss 0.333136 LR 0.001000 Time 0.019635 -2023-02-13 17:50:32,920 - Epoch: [72][ 1150/ 1207] Overall Loss 0.333058 Objective Loss 0.333058 LR 0.001000 Time 0.019629 -2023-02-13 17:50:33,110 - Epoch: [72][ 1160/ 1207] Overall Loss 0.333264 Objective Loss 0.333264 LR 0.001000 Time 0.019623 -2023-02-13 17:50:33,300 - Epoch: [72][ 1170/ 1207] Overall Loss 0.333130 Objective Loss 0.333130 LR 0.001000 Time 0.019617 -2023-02-13 17:50:33,490 - Epoch: [72][ 1180/ 1207] Overall Loss 0.333192 Objective Loss 0.333192 LR 0.001000 Time 0.019612 -2023-02-13 17:50:33,680 - Epoch: [72][ 1190/ 1207] Overall Loss 0.333338 Objective Loss 0.333338 LR 0.001000 Time 0.019606 -2023-02-13 17:50:33,920 - Epoch: [72][ 1200/ 1207] Overall Loss 0.333521 Objective Loss 0.333521 LR 0.001000 Time 0.019643 -2023-02-13 17:50:34,038 - Epoch: [72][ 1207/ 1207] Overall Loss 0.333446 Objective Loss 0.333446 Top1 81.707317 Top5 96.951220 LR 0.001000 Time 0.019626 -2023-02-13 17:50:34,115 - --- validate (epoch=72)----------- -2023-02-13 17:50:34,116 - 34311 samples (256 per mini-batch) -2023-02-13 17:50:34,521 - Epoch: [72][ 10/ 135] Loss 0.341848 Top1 82.460938 Top5 97.382812 -2023-02-13 17:50:34,651 - Epoch: [72][ 20/ 135] Loss 0.362632 Top1 81.875000 Top5 96.933594 -2023-02-13 17:50:34,780 - Epoch: [72][ 30/ 135] Loss 0.364012 Top1 81.861979 Top5 96.901042 -2023-02-13 17:50:34,908 - Epoch: [72][ 40/ 135] Loss 0.354308 Top1 81.884766 Top5 96.972656 -2023-02-13 17:50:35,039 - Epoch: [72][ 50/ 135] Loss 0.360543 Top1 81.859375 Top5 97.007812 -2023-02-13 17:50:35,170 - Epoch: [72][ 60/ 135] Loss 0.361977 Top1 81.790365 Top5 96.946615 -2023-02-13 17:50:35,301 - Epoch: [72][ 70/ 135] Loss 0.361908 Top1 81.891741 Top5 96.852679 -2023-02-13 17:50:35,431 - Epoch: [72][ 80/ 135] Loss 0.363024 Top1 81.938477 Top5 96.909180 -2023-02-13 17:50:35,559 - Epoch: [72][ 90/ 135] Loss 0.366829 Top1 81.792535 Top5 96.931424 -2023-02-13 17:50:35,686 - Epoch: [72][ 100/ 135] Loss 0.365909 Top1 81.863281 Top5 96.964844 -2023-02-13 17:50:35,818 - Epoch: [72][ 110/ 135] Loss 0.366722 Top1 81.839489 Top5 96.960227 -2023-02-13 17:50:35,945 - Epoch: [72][ 120/ 135] Loss 0.364552 Top1 81.897786 Top5 96.962891 -2023-02-13 17:50:36,078 - Epoch: [72][ 130/ 135] Loss 0.365630 Top1 81.947115 Top5 96.974159 -2023-02-13 17:50:36,126 - Epoch: [72][ 135/ 135] Loss 0.371086 Top1 81.991198 Top5 96.995133 -2023-02-13 17:50:36,202 - ==> Top1: 81.991 Top5: 96.995 Loss: 0.371 - -2023-02-13 17:50:36,203 - ==> Confusion: -[[ 842 8 10 6 6 2 0 3 2 51 1 2 0 6 8 5 1 2 2 3 7] - [ 1 937 1 2 7 29 3 31 2 0 1 1 1 0 2 2 1 0 2 3 7] - [ 4 9 953 11 1 0 16 10 0 1 5 1 4 3 3 4 4 5 7 9 8] - [ 3 4 32 902 3 5 2 4 1 1 10 0 8 0 9 1 5 4 18 0 4] - [ 15 6 3 1 982 11 1 2 0 5 0 8 2 5 8 6 5 2 0 1 3] - [ 5 24 1 5 8 945 3 25 1 2 2 14 1 10 2 2 5 0 2 10 3] - [ 1 6 19 3 1 6 1025 11 1 1 3 1 0 2 0 3 1 2 1 9 3] - [ 2 9 10 1 3 30 5 918 1 2 2 7 0 1 1 0 0 1 10 19 2] - [ 18 5 0 2 2 2 1 1 858 46 17 4 1 15 29 1 1 1 4 0 1] - [ 78 2 6 2 7 2 2 1 28 843 1 0 1 23 5 1 0 1 1 2 6] - [ 2 5 7 10 0 2 4 8 14 1 963 3 3 12 3 0 1 0 9 2 2] - [ 3 3 0 0 3 19 1 4 2 0 1 905 17 6 3 6 5 8 0 17 2] - [ 1 1 1 6 2 7 0 2 2 0 1 53 841 1 2 10 1 13 4 1 10] - [ 5 3 5 0 9 10 1 3 10 15 10 8 3 912 6 3 7 0 0 8 6] - [ 10 7 3 21 4 3 0 3 14 4 4 0 3 1 974 4 3 5 16 1 12] - [ 4 3 4 1 8 2 4 2 0 0 0 8 10 3 1 964 11 5 0 5 11] - [ 3 11 4 0 9 3 1 0 3 0 0 2 4 4 5 12 980 0 2 3 15] - [ 9 2 2 7 2 1 3 0 0 1 1 25 20 2 0 20 1 946 0 4 5] - [ 4 5 10 18 4 0 0 38 2 0 4 3 5 1 16 1 1 1 967 2 4] - [ 0 4 1 1 1 2 7 17 1 0 0 18 4 4 0 6 4 2 2 1068 6] - [ 145 326 290 159 165 201 117 228 87 78 224 162 343 318 153 113 258 93 179 388 9407]] - -2023-02-13 17:50:36,204 - ==> Best [Top1: 82.498 Top5: 97.120 Sparsity:0.00 Params: 148928 on epoch: 57] -2023-02-13 17:50:36,204 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:50:36,210 - - -2023-02-13 17:50:36,210 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:50:37,108 - Epoch: [73][ 10/ 1207] Overall Loss 0.344421 Objective Loss 0.344421 LR 0.001000 Time 0.089708 -2023-02-13 17:50:37,300 - Epoch: [73][ 20/ 1207] Overall Loss 0.345456 Objective Loss 0.345456 LR 0.001000 Time 0.054454 -2023-02-13 17:50:37,489 - Epoch: [73][ 30/ 1207] Overall Loss 0.337135 Objective Loss 0.337135 LR 0.001000 Time 0.042598 -2023-02-13 17:50:37,679 - Epoch: [73][ 40/ 1207] Overall Loss 0.332490 Objective Loss 0.332490 LR 0.001000 Time 0.036675 -2023-02-13 17:50:37,867 - Epoch: [73][ 50/ 1207] Overall Loss 0.331043 Objective Loss 0.331043 LR 0.001000 Time 0.033101 -2023-02-13 17:50:38,056 - Epoch: [73][ 60/ 1207] Overall Loss 0.330992 Objective Loss 0.330992 LR 0.001000 Time 0.030723 -2023-02-13 17:50:38,244 - Epoch: [73][ 70/ 1207] Overall Loss 0.329861 Objective Loss 0.329861 LR 0.001000 Time 0.029022 -2023-02-13 17:50:38,433 - Epoch: [73][ 80/ 1207] Overall Loss 0.329989 Objective Loss 0.329989 LR 0.001000 Time 0.027744 -2023-02-13 17:50:38,622 - Epoch: [73][ 90/ 1207] Overall Loss 0.329643 Objective Loss 0.329643 LR 0.001000 Time 0.026762 -2023-02-13 17:50:38,811 - Epoch: [73][ 100/ 1207] Overall Loss 0.329943 Objective Loss 0.329943 LR 0.001000 Time 0.025968 -2023-02-13 17:50:39,000 - Epoch: [73][ 110/ 1207] Overall Loss 0.330395 Objective Loss 0.330395 LR 0.001000 Time 0.025324 -2023-02-13 17:50:39,190 - Epoch: [73][ 120/ 1207] Overall Loss 0.328837 Objective Loss 0.328837 LR 0.001000 Time 0.024790 -2023-02-13 17:50:39,378 - Epoch: [73][ 130/ 1207] Overall Loss 0.329792 Objective Loss 0.329792 LR 0.001000 Time 0.024333 -2023-02-13 17:50:39,568 - Epoch: [73][ 140/ 1207] Overall Loss 0.329634 Objective Loss 0.329634 LR 0.001000 Time 0.023947 -2023-02-13 17:50:39,757 - Epoch: [73][ 150/ 1207] Overall Loss 0.328470 Objective Loss 0.328470 LR 0.001000 Time 0.023610 -2023-02-13 17:50:39,946 - Epoch: [73][ 160/ 1207] Overall Loss 0.327802 Objective Loss 0.327802 LR 0.001000 Time 0.023309 -2023-02-13 17:50:40,135 - Epoch: [73][ 170/ 1207] Overall Loss 0.328805 Objective Loss 0.328805 LR 0.001000 Time 0.023047 -2023-02-13 17:50:40,324 - Epoch: [73][ 180/ 1207] Overall Loss 0.332655 Objective Loss 0.332655 LR 0.001000 Time 0.022814 -2023-02-13 17:50:40,512 - Epoch: [73][ 190/ 1207] Overall Loss 0.332596 Objective Loss 0.332596 LR 0.001000 Time 0.022605 -2023-02-13 17:50:40,701 - Epoch: [73][ 200/ 1207] Overall Loss 0.332622 Objective Loss 0.332622 LR 0.001000 Time 0.022418 -2023-02-13 17:50:40,893 - Epoch: [73][ 210/ 1207] Overall Loss 0.332867 Objective Loss 0.332867 LR 0.001000 Time 0.022260 -2023-02-13 17:50:41,082 - Epoch: [73][ 220/ 1207] Overall Loss 0.332833 Objective Loss 0.332833 LR 0.001000 Time 0.022105 -2023-02-13 17:50:41,270 - Epoch: [73][ 230/ 1207] Overall Loss 0.332466 Objective Loss 0.332466 LR 0.001000 Time 0.021961 -2023-02-13 17:50:41,459 - Epoch: [73][ 240/ 1207] Overall Loss 0.331645 Objective Loss 0.331645 LR 0.001000 Time 0.021830 -2023-02-13 17:50:41,648 - Epoch: [73][ 250/ 1207] Overall Loss 0.332986 Objective Loss 0.332986 LR 0.001000 Time 0.021711 -2023-02-13 17:50:41,837 - Epoch: [73][ 260/ 1207] Overall Loss 0.332772 Objective Loss 0.332772 LR 0.001000 Time 0.021603 -2023-02-13 17:50:42,026 - Epoch: [73][ 270/ 1207] Overall Loss 0.332741 Objective Loss 0.332741 LR 0.001000 Time 0.021502 -2023-02-13 17:50:42,214 - Epoch: [73][ 280/ 1207] Overall Loss 0.332576 Objective Loss 0.332576 LR 0.001000 Time 0.021404 -2023-02-13 17:50:42,403 - Epoch: [73][ 290/ 1207] Overall Loss 0.332822 Objective Loss 0.332822 LR 0.001000 Time 0.021315 -2023-02-13 17:50:42,592 - Epoch: [73][ 300/ 1207] Overall Loss 0.332523 Objective Loss 0.332523 LR 0.001000 Time 0.021235 -2023-02-13 17:50:42,782 - Epoch: [73][ 310/ 1207] Overall Loss 0.331672 Objective Loss 0.331672 LR 0.001000 Time 0.021160 -2023-02-13 17:50:42,970 - Epoch: [73][ 320/ 1207] Overall Loss 0.330360 Objective Loss 0.330360 LR 0.001000 Time 0.021088 -2023-02-13 17:50:43,159 - Epoch: [73][ 330/ 1207] Overall Loss 0.330526 Objective Loss 0.330526 LR 0.001000 Time 0.021020 -2023-02-13 17:50:43,348 - Epoch: [73][ 340/ 1207] Overall Loss 0.330504 Objective Loss 0.330504 LR 0.001000 Time 0.020954 -2023-02-13 17:50:43,537 - Epoch: [73][ 350/ 1207] Overall Loss 0.330881 Objective Loss 0.330881 LR 0.001000 Time 0.020895 -2023-02-13 17:50:43,726 - Epoch: [73][ 360/ 1207] Overall Loss 0.330961 Objective Loss 0.330961 LR 0.001000 Time 0.020838 -2023-02-13 17:50:43,914 - Epoch: [73][ 370/ 1207] Overall Loss 0.330616 Objective Loss 0.330616 LR 0.001000 Time 0.020783 -2023-02-13 17:50:44,103 - Epoch: [73][ 380/ 1207] Overall Loss 0.331007 Objective Loss 0.331007 LR 0.001000 Time 0.020732 -2023-02-13 17:50:44,291 - Epoch: [73][ 390/ 1207] Overall Loss 0.331400 Objective Loss 0.331400 LR 0.001000 Time 0.020682 -2023-02-13 17:50:44,479 - Epoch: [73][ 400/ 1207] Overall Loss 0.332092 Objective Loss 0.332092 LR 0.001000 Time 0.020635 -2023-02-13 17:50:44,669 - Epoch: [73][ 410/ 1207] Overall Loss 0.332007 Objective Loss 0.332007 LR 0.001000 Time 0.020593 -2023-02-13 17:50:44,858 - Epoch: [73][ 420/ 1207] Overall Loss 0.332032 Objective Loss 0.332032 LR 0.001000 Time 0.020551 -2023-02-13 17:50:45,047 - Epoch: [73][ 430/ 1207] Overall Loss 0.333023 Objective Loss 0.333023 LR 0.001000 Time 0.020512 -2023-02-13 17:50:45,235 - Epoch: [73][ 440/ 1207] Overall Loss 0.332876 Objective Loss 0.332876 LR 0.001000 Time 0.020473 -2023-02-13 17:50:45,424 - Epoch: [73][ 450/ 1207] Overall Loss 0.332392 Objective Loss 0.332392 LR 0.001000 Time 0.020438 -2023-02-13 17:50:45,614 - Epoch: [73][ 460/ 1207] Overall Loss 0.332312 Objective Loss 0.332312 LR 0.001000 Time 0.020405 -2023-02-13 17:50:45,804 - Epoch: [73][ 470/ 1207] Overall Loss 0.331931 Objective Loss 0.331931 LR 0.001000 Time 0.020375 -2023-02-13 17:50:45,994 - Epoch: [73][ 480/ 1207] Overall Loss 0.331843 Objective Loss 0.331843 LR 0.001000 Time 0.020345 -2023-02-13 17:50:46,183 - Epoch: [73][ 490/ 1207] Overall Loss 0.331432 Objective Loss 0.331432 LR 0.001000 Time 0.020314 -2023-02-13 17:50:46,372 - Epoch: [73][ 500/ 1207] Overall Loss 0.331430 Objective Loss 0.331430 LR 0.001000 Time 0.020286 -2023-02-13 17:50:46,561 - Epoch: [73][ 510/ 1207] Overall Loss 0.331419 Objective Loss 0.331419 LR 0.001000 Time 0.020258 -2023-02-13 17:50:46,751 - Epoch: [73][ 520/ 1207] Overall Loss 0.331519 Objective Loss 0.331519 LR 0.001000 Time 0.020233 -2023-02-13 17:50:46,941 - Epoch: [73][ 530/ 1207] Overall Loss 0.331487 Objective Loss 0.331487 LR 0.001000 Time 0.020208 -2023-02-13 17:50:47,130 - Epoch: [73][ 540/ 1207] Overall Loss 0.331164 Objective Loss 0.331164 LR 0.001000 Time 0.020183 -2023-02-13 17:50:47,319 - Epoch: [73][ 550/ 1207] Overall Loss 0.330713 Objective Loss 0.330713 LR 0.001000 Time 0.020159 -2023-02-13 17:50:47,508 - Epoch: [73][ 560/ 1207] Overall Loss 0.331041 Objective Loss 0.331041 LR 0.001000 Time 0.020136 -2023-02-13 17:50:47,698 - Epoch: [73][ 570/ 1207] Overall Loss 0.331236 Objective Loss 0.331236 LR 0.001000 Time 0.020116 -2023-02-13 17:50:47,886 - Epoch: [73][ 580/ 1207] Overall Loss 0.331045 Objective Loss 0.331045 LR 0.001000 Time 0.020093 -2023-02-13 17:50:48,075 - Epoch: [73][ 590/ 1207] Overall Loss 0.331141 Objective Loss 0.331141 LR 0.001000 Time 0.020072 -2023-02-13 17:50:48,264 - Epoch: [73][ 600/ 1207] Overall Loss 0.331082 Objective Loss 0.331082 LR 0.001000 Time 0.020051 -2023-02-13 17:50:48,453 - Epoch: [73][ 610/ 1207] Overall Loss 0.331318 Objective Loss 0.331318 LR 0.001000 Time 0.020031 -2023-02-13 17:50:48,642 - Epoch: [73][ 620/ 1207] Overall Loss 0.331718 Objective Loss 0.331718 LR 0.001000 Time 0.020013 -2023-02-13 17:50:48,831 - Epoch: [73][ 630/ 1207] Overall Loss 0.331768 Objective Loss 0.331768 LR 0.001000 Time 0.019995 -2023-02-13 17:50:49,020 - Epoch: [73][ 640/ 1207] Overall Loss 0.331615 Objective Loss 0.331615 LR 0.001000 Time 0.019977 -2023-02-13 17:50:49,209 - Epoch: [73][ 650/ 1207] Overall Loss 0.331353 Objective Loss 0.331353 LR 0.001000 Time 0.019961 -2023-02-13 17:50:49,399 - Epoch: [73][ 660/ 1207] Overall Loss 0.331305 Objective Loss 0.331305 LR 0.001000 Time 0.019945 -2023-02-13 17:50:49,588 - Epoch: [73][ 670/ 1207] Overall Loss 0.331233 Objective Loss 0.331233 LR 0.001000 Time 0.019929 -2023-02-13 17:50:49,777 - Epoch: [73][ 680/ 1207] Overall Loss 0.331357 Objective Loss 0.331357 LR 0.001000 Time 0.019914 -2023-02-13 17:50:49,966 - Epoch: [73][ 690/ 1207] Overall Loss 0.331550 Objective Loss 0.331550 LR 0.001000 Time 0.019898 -2023-02-13 17:50:50,155 - Epoch: [73][ 700/ 1207] Overall Loss 0.331136 Objective Loss 0.331136 LR 0.001000 Time 0.019883 -2023-02-13 17:50:50,344 - Epoch: [73][ 710/ 1207] Overall Loss 0.331314 Objective Loss 0.331314 LR 0.001000 Time 0.019868 -2023-02-13 17:50:50,533 - Epoch: [73][ 720/ 1207] Overall Loss 0.331122 Objective Loss 0.331122 LR 0.001000 Time 0.019855 -2023-02-13 17:50:50,723 - Epoch: [73][ 730/ 1207] Overall Loss 0.331120 Objective Loss 0.331120 LR 0.001000 Time 0.019843 -2023-02-13 17:50:50,914 - Epoch: [73][ 740/ 1207] Overall Loss 0.331249 Objective Loss 0.331249 LR 0.001000 Time 0.019832 -2023-02-13 17:50:51,103 - Epoch: [73][ 750/ 1207] Overall Loss 0.331421 Objective Loss 0.331421 LR 0.001000 Time 0.019820 -2023-02-13 17:50:51,293 - Epoch: [73][ 760/ 1207] Overall Loss 0.331446 Objective Loss 0.331446 LR 0.001000 Time 0.019807 -2023-02-13 17:50:51,482 - Epoch: [73][ 770/ 1207] Overall Loss 0.331940 Objective Loss 0.331940 LR 0.001000 Time 0.019795 -2023-02-13 17:50:51,671 - Epoch: [73][ 780/ 1207] Overall Loss 0.332013 Objective Loss 0.332013 LR 0.001000 Time 0.019784 -2023-02-13 17:50:51,861 - Epoch: [73][ 790/ 1207] Overall Loss 0.331652 Objective Loss 0.331652 LR 0.001000 Time 0.019774 -2023-02-13 17:50:52,051 - Epoch: [73][ 800/ 1207] Overall Loss 0.331628 Objective Loss 0.331628 LR 0.001000 Time 0.019763 -2023-02-13 17:50:52,240 - Epoch: [73][ 810/ 1207] Overall Loss 0.331764 Objective Loss 0.331764 LR 0.001000 Time 0.019752 -2023-02-13 17:50:52,429 - Epoch: [73][ 820/ 1207] Overall Loss 0.331826 Objective Loss 0.331826 LR 0.001000 Time 0.019742 -2023-02-13 17:50:52,619 - Epoch: [73][ 830/ 1207] Overall Loss 0.331508 Objective Loss 0.331508 LR 0.001000 Time 0.019732 -2023-02-13 17:50:52,809 - Epoch: [73][ 840/ 1207] Overall Loss 0.331507 Objective Loss 0.331507 LR 0.001000 Time 0.019722 -2023-02-13 17:50:52,998 - Epoch: [73][ 850/ 1207] Overall Loss 0.331347 Objective Loss 0.331347 LR 0.001000 Time 0.019712 -2023-02-13 17:50:53,187 - Epoch: [73][ 860/ 1207] Overall Loss 0.331191 Objective Loss 0.331191 LR 0.001000 Time 0.019702 -2023-02-13 17:50:53,377 - Epoch: [73][ 870/ 1207] Overall Loss 0.331011 Objective Loss 0.331011 LR 0.001000 Time 0.019694 -2023-02-13 17:50:53,566 - Epoch: [73][ 880/ 1207] Overall Loss 0.331079 Objective Loss 0.331079 LR 0.001000 Time 0.019684 -2023-02-13 17:50:53,756 - Epoch: [73][ 890/ 1207] Overall Loss 0.331087 Objective Loss 0.331087 LR 0.001000 Time 0.019676 -2023-02-13 17:50:53,946 - Epoch: [73][ 900/ 1207] Overall Loss 0.331146 Objective Loss 0.331146 LR 0.001000 Time 0.019668 -2023-02-13 17:50:54,136 - Epoch: [73][ 910/ 1207] Overall Loss 0.331222 Objective Loss 0.331222 LR 0.001000 Time 0.019660 -2023-02-13 17:50:54,325 - Epoch: [73][ 920/ 1207] Overall Loss 0.331277 Objective Loss 0.331277 LR 0.001000 Time 0.019652 -2023-02-13 17:50:54,514 - Epoch: [73][ 930/ 1207] Overall Loss 0.331521 Objective Loss 0.331521 LR 0.001000 Time 0.019644 -2023-02-13 17:50:54,704 - Epoch: [73][ 940/ 1207] Overall Loss 0.331153 Objective Loss 0.331153 LR 0.001000 Time 0.019636 -2023-02-13 17:50:54,894 - Epoch: [73][ 950/ 1207] Overall Loss 0.331185 Objective Loss 0.331185 LR 0.001000 Time 0.019629 -2023-02-13 17:50:55,084 - Epoch: [73][ 960/ 1207] Overall Loss 0.331238 Objective Loss 0.331238 LR 0.001000 Time 0.019622 -2023-02-13 17:50:55,273 - Epoch: [73][ 970/ 1207] Overall Loss 0.330791 Objective Loss 0.330791 LR 0.001000 Time 0.019615 -2023-02-13 17:50:55,462 - Epoch: [73][ 980/ 1207] Overall Loss 0.330434 Objective Loss 0.330434 LR 0.001000 Time 0.019607 -2023-02-13 17:50:55,651 - Epoch: [73][ 990/ 1207] Overall Loss 0.330880 Objective Loss 0.330880 LR 0.001000 Time 0.019600 -2023-02-13 17:50:55,842 - Epoch: [73][ 1000/ 1207] Overall Loss 0.330818 Objective Loss 0.330818 LR 0.001000 Time 0.019594 -2023-02-13 17:50:56,032 - Epoch: [73][ 1010/ 1207] Overall Loss 0.330842 Objective Loss 0.330842 LR 0.001000 Time 0.019588 -2023-02-13 17:50:56,221 - Epoch: [73][ 1020/ 1207] Overall Loss 0.330835 Objective Loss 0.330835 LR 0.001000 Time 0.019581 -2023-02-13 17:50:56,412 - Epoch: [73][ 1030/ 1207] Overall Loss 0.330784 Objective Loss 0.330784 LR 0.001000 Time 0.019576 -2023-02-13 17:50:56,605 - Epoch: [73][ 1040/ 1207] Overall Loss 0.330887 Objective Loss 0.330887 LR 0.001000 Time 0.019572 -2023-02-13 17:50:56,798 - Epoch: [73][ 1050/ 1207] Overall Loss 0.330715 Objective Loss 0.330715 LR 0.001000 Time 0.019569 -2023-02-13 17:50:56,992 - Epoch: [73][ 1060/ 1207] Overall Loss 0.330715 Objective Loss 0.330715 LR 0.001000 Time 0.019568 -2023-02-13 17:50:57,182 - Epoch: [73][ 1070/ 1207] Overall Loss 0.330814 Objective Loss 0.330814 LR 0.001000 Time 0.019562 -2023-02-13 17:50:57,373 - Epoch: [73][ 1080/ 1207] Overall Loss 0.330668 Objective Loss 0.330668 LR 0.001000 Time 0.019557 -2023-02-13 17:50:57,563 - Epoch: [73][ 1090/ 1207] Overall Loss 0.330634 Objective Loss 0.330634 LR 0.001000 Time 0.019552 -2023-02-13 17:50:57,755 - Epoch: [73][ 1100/ 1207] Overall Loss 0.330786 Objective Loss 0.330786 LR 0.001000 Time 0.019548 -2023-02-13 17:50:57,945 - Epoch: [73][ 1110/ 1207] Overall Loss 0.330825 Objective Loss 0.330825 LR 0.001000 Time 0.019543 -2023-02-13 17:50:58,136 - Epoch: [73][ 1120/ 1207] Overall Loss 0.331155 Objective Loss 0.331155 LR 0.001000 Time 0.019539 -2023-02-13 17:50:58,326 - Epoch: [73][ 1130/ 1207] Overall Loss 0.331142 Objective Loss 0.331142 LR 0.001000 Time 0.019534 -2023-02-13 17:50:58,516 - Epoch: [73][ 1140/ 1207] Overall Loss 0.331185 Objective Loss 0.331185 LR 0.001000 Time 0.019529 -2023-02-13 17:50:58,708 - Epoch: [73][ 1150/ 1207] Overall Loss 0.331164 Objective Loss 0.331164 LR 0.001000 Time 0.019525 -2023-02-13 17:50:58,899 - Epoch: [73][ 1160/ 1207] Overall Loss 0.331057 Objective Loss 0.331057 LR 0.001000 Time 0.019521 -2023-02-13 17:50:59,090 - Epoch: [73][ 1170/ 1207] Overall Loss 0.331248 Objective Loss 0.331248 LR 0.001000 Time 0.019517 -2023-02-13 17:50:59,280 - Epoch: [73][ 1180/ 1207] Overall Loss 0.331065 Objective Loss 0.331065 LR 0.001000 Time 0.019513 -2023-02-13 17:50:59,470 - Epoch: [73][ 1190/ 1207] Overall Loss 0.331454 Objective Loss 0.331454 LR 0.001000 Time 0.019508 -2023-02-13 17:50:59,711 - Epoch: [73][ 1200/ 1207] Overall Loss 0.331414 Objective Loss 0.331414 LR 0.001000 Time 0.019546 -2023-02-13 17:50:59,825 - Epoch: [73][ 1207/ 1207] Overall Loss 0.331232 Objective Loss 0.331232 Top1 83.536585 Top5 97.865854 LR 0.001000 Time 0.019527 -2023-02-13 17:50:59,902 - --- validate (epoch=73)----------- -2023-02-13 17:50:59,902 - 34311 samples (256 per mini-batch) -2023-02-13 17:51:00,325 - Epoch: [73][ 10/ 135] Loss 0.352098 Top1 82.773438 Top5 97.031250 -2023-02-13 17:51:00,457 - Epoch: [73][ 20/ 135] Loss 0.355210 Top1 82.363281 Top5 97.207031 -2023-02-13 17:51:00,586 - Epoch: [73][ 30/ 135] Loss 0.358126 Top1 82.434896 Top5 97.356771 -2023-02-13 17:51:00,714 - Epoch: [73][ 40/ 135] Loss 0.358239 Top1 82.539062 Top5 97.392578 -2023-02-13 17:51:00,845 - Epoch: [73][ 50/ 135] Loss 0.361009 Top1 82.570312 Top5 97.390625 -2023-02-13 17:51:00,974 - Epoch: [73][ 60/ 135] Loss 0.352873 Top1 82.753906 Top5 97.486979 -2023-02-13 17:51:01,105 - Epoch: [73][ 70/ 135] Loss 0.349309 Top1 82.728795 Top5 97.511161 -2023-02-13 17:51:01,234 - Epoch: [73][ 80/ 135] Loss 0.352043 Top1 82.709961 Top5 97.543945 -2023-02-13 17:51:01,363 - Epoch: [73][ 90/ 135] Loss 0.357838 Top1 82.604167 Top5 97.456597 -2023-02-13 17:51:01,492 - Epoch: [73][ 100/ 135] Loss 0.360110 Top1 82.613281 Top5 97.492188 -2023-02-13 17:51:01,621 - Epoch: [73][ 110/ 135] Loss 0.365665 Top1 82.563920 Top5 97.460938 -2023-02-13 17:51:01,752 - Epoch: [73][ 120/ 135] Loss 0.365023 Top1 82.526042 Top5 97.418620 -2023-02-13 17:51:01,880 - Epoch: [73][ 130/ 135] Loss 0.364247 Top1 82.548077 Top5 97.430889 -2023-02-13 17:51:01,924 - Epoch: [73][ 135/ 135] Loss 0.367212 Top1 82.539127 Top5 97.403165 -2023-02-13 17:51:01,993 - ==> Top1: 82.539 Top5: 97.403 Loss: 0.367 - -2023-02-13 17:51:01,993 - ==> Confusion: -[[ 843 5 4 1 13 5 0 2 3 54 1 4 1 3 7 5 3 0 3 3 7] - [ 4 939 2 2 14 37 2 10 2 0 0 1 1 0 2 1 4 0 4 2 6] - [ 9 5 930 13 3 4 26 18 0 1 2 1 3 2 4 6 3 3 8 1 16] - [ 5 3 22 891 1 7 3 2 1 2 11 1 5 0 28 1 2 5 14 1 11] - [ 14 12 0 2 983 14 1 0 1 1 0 5 2 4 6 3 6 4 0 2 6] - [ 3 23 1 2 8 959 3 13 3 8 2 18 2 9 0 2 3 0 0 9 2] - [ 3 5 13 0 1 11 1038 5 0 1 2 0 0 1 0 4 0 3 1 5 6] - [ 2 17 9 2 2 40 3 893 0 1 6 3 1 2 1 0 4 2 18 11 7] - [ 19 5 1 1 3 0 0 2 866 44 10 5 0 9 25 3 0 3 6 1 6] - [ 88 2 6 1 4 1 0 1 44 823 0 4 0 14 10 0 2 3 2 1 6] - [ 2 3 4 7 3 4 5 5 17 0 960 3 1 9 4 0 1 1 13 3 6] - [ 4 2 0 0 4 15 1 6 2 0 2 914 17 6 0 8 2 3 4 11 4] - [ 2 0 1 6 1 4 0 1 2 0 0 56 834 1 4 6 4 19 4 4 10] - [ 3 4 3 0 9 27 1 2 13 22 7 9 4 891 8 3 6 2 0 3 7] - [ 7 3 4 18 7 3 0 3 21 4 2 0 3 4 985 2 1 5 8 0 12] - [ 3 2 4 0 10 2 7 0 2 0 0 7 8 3 1 962 10 7 0 7 11] - [ 0 7 2 1 9 2 0 1 4 1 0 4 0 3 3 11 995 0 2 3 13] - [ 4 3 1 3 2 2 4 0 1 1 2 20 12 0 0 24 1 959 0 2 10] - [ 3 5 9 18 2 1 0 36 3 0 8 4 5 2 15 0 2 0 971 1 1] - [ 0 5 2 1 2 11 12 13 1 0 1 17 4 2 0 7 0 4 0 1055 11] - [ 194 319 180 133 151 288 91 214 97 101 175 150 292 316 173 116 261 89 179 286 9629]] - -2023-02-13 17:51:01,995 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:51:01,995 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:51:02,001 - - -2023-02-13 17:51:02,002 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:51:02,880 - Epoch: [74][ 10/ 1207] Overall Loss 0.309595 Objective Loss 0.309595 LR 0.001000 Time 0.087829 -2023-02-13 17:51:03,080 - Epoch: [74][ 20/ 1207] Overall Loss 0.313506 Objective Loss 0.313506 LR 0.001000 Time 0.053884 -2023-02-13 17:51:03,268 - Epoch: [74][ 30/ 1207] Overall Loss 0.327217 Objective Loss 0.327217 LR 0.001000 Time 0.042182 -2023-02-13 17:51:03,457 - Epoch: [74][ 40/ 1207] Overall Loss 0.322241 Objective Loss 0.322241 LR 0.001000 Time 0.036333 -2023-02-13 17:51:03,645 - Epoch: [74][ 50/ 1207] Overall Loss 0.312815 Objective Loss 0.312815 LR 0.001000 Time 0.032821 -2023-02-13 17:51:03,833 - Epoch: [74][ 60/ 1207] Overall Loss 0.311777 Objective Loss 0.311777 LR 0.001000 Time 0.030485 -2023-02-13 17:51:04,020 - Epoch: [74][ 70/ 1207] Overall Loss 0.313458 Objective Loss 0.313458 LR 0.001000 Time 0.028797 -2023-02-13 17:51:04,208 - Epoch: [74][ 80/ 1207] Overall Loss 0.316166 Objective Loss 0.316166 LR 0.001000 Time 0.027544 -2023-02-13 17:51:04,396 - Epoch: [74][ 90/ 1207] Overall Loss 0.313974 Objective Loss 0.313974 LR 0.001000 Time 0.026561 -2023-02-13 17:51:04,583 - Epoch: [74][ 100/ 1207] Overall Loss 0.315627 Objective Loss 0.315627 LR 0.001000 Time 0.025779 -2023-02-13 17:51:04,771 - Epoch: [74][ 110/ 1207] Overall Loss 0.319931 Objective Loss 0.319931 LR 0.001000 Time 0.025141 -2023-02-13 17:51:04,960 - Epoch: [74][ 120/ 1207] Overall Loss 0.319605 Objective Loss 0.319605 LR 0.001000 Time 0.024613 -2023-02-13 17:51:05,147 - Epoch: [74][ 130/ 1207] Overall Loss 0.321243 Objective Loss 0.321243 LR 0.001000 Time 0.024160 -2023-02-13 17:51:05,335 - Epoch: [74][ 140/ 1207] Overall Loss 0.322660 Objective Loss 0.322660 LR 0.001000 Time 0.023772 -2023-02-13 17:51:05,523 - Epoch: [74][ 150/ 1207] Overall Loss 0.323665 Objective Loss 0.323665 LR 0.001000 Time 0.023435 -2023-02-13 17:51:05,711 - Epoch: [74][ 160/ 1207] Overall Loss 0.324987 Objective Loss 0.324987 LR 0.001000 Time 0.023144 -2023-02-13 17:51:05,900 - Epoch: [74][ 170/ 1207] Overall Loss 0.324513 Objective Loss 0.324513 LR 0.001000 Time 0.022892 -2023-02-13 17:51:06,088 - Epoch: [74][ 180/ 1207] Overall Loss 0.326461 Objective Loss 0.326461 LR 0.001000 Time 0.022665 -2023-02-13 17:51:06,275 - Epoch: [74][ 190/ 1207] Overall Loss 0.325987 Objective Loss 0.325987 LR 0.001000 Time 0.022455 -2023-02-13 17:51:06,463 - Epoch: [74][ 200/ 1207] Overall Loss 0.326271 Objective Loss 0.326271 LR 0.001000 Time 0.022267 -2023-02-13 17:51:06,650 - Epoch: [74][ 210/ 1207] Overall Loss 0.325729 Objective Loss 0.325729 LR 0.001000 Time 0.022099 -2023-02-13 17:51:06,839 - Epoch: [74][ 220/ 1207] Overall Loss 0.325950 Objective Loss 0.325950 LR 0.001000 Time 0.021950 -2023-02-13 17:51:07,027 - Epoch: [74][ 230/ 1207] Overall Loss 0.325765 Objective Loss 0.325765 LR 0.001000 Time 0.021811 -2023-02-13 17:51:07,216 - Epoch: [74][ 240/ 1207] Overall Loss 0.326344 Objective Loss 0.326344 LR 0.001000 Time 0.021687 -2023-02-13 17:51:07,403 - Epoch: [74][ 250/ 1207] Overall Loss 0.326526 Objective Loss 0.326526 LR 0.001000 Time 0.021567 -2023-02-13 17:51:07,591 - Epoch: [74][ 260/ 1207] Overall Loss 0.326606 Objective Loss 0.326606 LR 0.001000 Time 0.021458 -2023-02-13 17:51:07,779 - Epoch: [74][ 270/ 1207] Overall Loss 0.326889 Objective Loss 0.326889 LR 0.001000 Time 0.021360 -2023-02-13 17:51:07,966 - Epoch: [74][ 280/ 1207] Overall Loss 0.327905 Objective Loss 0.327905 LR 0.001000 Time 0.021265 -2023-02-13 17:51:08,154 - Epoch: [74][ 290/ 1207] Overall Loss 0.328456 Objective Loss 0.328456 LR 0.001000 Time 0.021177 -2023-02-13 17:51:08,342 - Epoch: [74][ 300/ 1207] Overall Loss 0.329331 Objective Loss 0.329331 LR 0.001000 Time 0.021097 -2023-02-13 17:51:08,529 - Epoch: [74][ 310/ 1207] Overall Loss 0.329943 Objective Loss 0.329943 LR 0.001000 Time 0.021020 -2023-02-13 17:51:08,718 - Epoch: [74][ 320/ 1207] Overall Loss 0.330695 Objective Loss 0.330695 LR 0.001000 Time 0.020951 -2023-02-13 17:51:08,906 - Epoch: [74][ 330/ 1207] Overall Loss 0.330906 Objective Loss 0.330906 LR 0.001000 Time 0.020885 -2023-02-13 17:51:09,094 - Epoch: [74][ 340/ 1207] Overall Loss 0.331112 Objective Loss 0.331112 LR 0.001000 Time 0.020823 -2023-02-13 17:51:09,282 - Epoch: [74][ 350/ 1207] Overall Loss 0.330879 Objective Loss 0.330879 LR 0.001000 Time 0.020762 -2023-02-13 17:51:09,469 - Epoch: [74][ 360/ 1207] Overall Loss 0.331625 Objective Loss 0.331625 LR 0.001000 Time 0.020704 -2023-02-13 17:51:09,656 - Epoch: [74][ 370/ 1207] Overall Loss 0.331295 Objective Loss 0.331295 LR 0.001000 Time 0.020649 -2023-02-13 17:51:09,844 - Epoch: [74][ 380/ 1207] Overall Loss 0.331860 Objective Loss 0.331860 LR 0.001000 Time 0.020601 -2023-02-13 17:51:10,032 - Epoch: [74][ 390/ 1207] Overall Loss 0.332273 Objective Loss 0.332273 LR 0.001000 Time 0.020552 -2023-02-13 17:51:10,219 - Epoch: [74][ 400/ 1207] Overall Loss 0.331741 Objective Loss 0.331741 LR 0.001000 Time 0.020506 -2023-02-13 17:51:10,406 - Epoch: [74][ 410/ 1207] Overall Loss 0.332315 Objective Loss 0.332315 LR 0.001000 Time 0.020462 -2023-02-13 17:51:10,595 - Epoch: [74][ 420/ 1207] Overall Loss 0.332492 Objective Loss 0.332492 LR 0.001000 Time 0.020422 -2023-02-13 17:51:10,783 - Epoch: [74][ 430/ 1207] Overall Loss 0.332772 Objective Loss 0.332772 LR 0.001000 Time 0.020384 -2023-02-13 17:51:10,972 - Epoch: [74][ 440/ 1207] Overall Loss 0.333367 Objective Loss 0.333367 LR 0.001000 Time 0.020350 -2023-02-13 17:51:11,160 - Epoch: [74][ 450/ 1207] Overall Loss 0.333299 Objective Loss 0.333299 LR 0.001000 Time 0.020314 -2023-02-13 17:51:11,347 - Epoch: [74][ 460/ 1207] Overall Loss 0.333319 Objective Loss 0.333319 LR 0.001000 Time 0.020280 -2023-02-13 17:51:11,535 - Epoch: [74][ 470/ 1207] Overall Loss 0.332877 Objective Loss 0.332877 LR 0.001000 Time 0.020246 -2023-02-13 17:51:11,723 - Epoch: [74][ 480/ 1207] Overall Loss 0.333500 Objective Loss 0.333500 LR 0.001000 Time 0.020216 -2023-02-13 17:51:11,912 - Epoch: [74][ 490/ 1207] Overall Loss 0.333397 Objective Loss 0.333397 LR 0.001000 Time 0.020187 -2023-02-13 17:51:12,100 - Epoch: [74][ 500/ 1207] Overall Loss 0.333789 Objective Loss 0.333789 LR 0.001000 Time 0.020159 -2023-02-13 17:51:12,288 - Epoch: [74][ 510/ 1207] Overall Loss 0.333587 Objective Loss 0.333587 LR 0.001000 Time 0.020131 -2023-02-13 17:51:12,475 - Epoch: [74][ 520/ 1207] Overall Loss 0.333551 Objective Loss 0.333551 LR 0.001000 Time 0.020104 -2023-02-13 17:51:12,663 - Epoch: [74][ 530/ 1207] Overall Loss 0.333360 Objective Loss 0.333360 LR 0.001000 Time 0.020079 -2023-02-13 17:51:12,852 - Epoch: [74][ 540/ 1207] Overall Loss 0.333734 Objective Loss 0.333734 LR 0.001000 Time 0.020056 -2023-02-13 17:51:13,041 - Epoch: [74][ 550/ 1207] Overall Loss 0.333631 Objective Loss 0.333631 LR 0.001000 Time 0.020033 -2023-02-13 17:51:13,229 - Epoch: [74][ 560/ 1207] Overall Loss 0.332984 Objective Loss 0.332984 LR 0.001000 Time 0.020011 -2023-02-13 17:51:13,417 - Epoch: [74][ 570/ 1207] Overall Loss 0.333043 Objective Loss 0.333043 LR 0.001000 Time 0.019989 -2023-02-13 17:51:13,604 - Epoch: [74][ 580/ 1207] Overall Loss 0.333194 Objective Loss 0.333194 LR 0.001000 Time 0.019967 -2023-02-13 17:51:13,792 - Epoch: [74][ 590/ 1207] Overall Loss 0.332997 Objective Loss 0.332997 LR 0.001000 Time 0.019947 -2023-02-13 17:51:13,981 - Epoch: [74][ 600/ 1207] Overall Loss 0.332609 Objective Loss 0.332609 LR 0.001000 Time 0.019928 -2023-02-13 17:51:14,169 - Epoch: [74][ 610/ 1207] Overall Loss 0.332639 Objective Loss 0.332639 LR 0.001000 Time 0.019909 -2023-02-13 17:51:14,357 - Epoch: [74][ 620/ 1207] Overall Loss 0.333307 Objective Loss 0.333307 LR 0.001000 Time 0.019891 -2023-02-13 17:51:14,544 - Epoch: [74][ 630/ 1207] Overall Loss 0.332842 Objective Loss 0.332842 LR 0.001000 Time 0.019871 -2023-02-13 17:51:14,732 - Epoch: [74][ 640/ 1207] Overall Loss 0.332556 Objective Loss 0.332556 LR 0.001000 Time 0.019854 -2023-02-13 17:51:14,921 - Epoch: [74][ 650/ 1207] Overall Loss 0.332702 Objective Loss 0.332702 LR 0.001000 Time 0.019839 -2023-02-13 17:51:15,109 - Epoch: [74][ 660/ 1207] Overall Loss 0.332243 Objective Loss 0.332243 LR 0.001000 Time 0.019822 -2023-02-13 17:51:15,297 - Epoch: [74][ 670/ 1207] Overall Loss 0.332299 Objective Loss 0.332299 LR 0.001000 Time 0.019806 -2023-02-13 17:51:15,484 - Epoch: [74][ 680/ 1207] Overall Loss 0.331834 Objective Loss 0.331834 LR 0.001000 Time 0.019790 -2023-02-13 17:51:15,673 - Epoch: [74][ 690/ 1207] Overall Loss 0.331838 Objective Loss 0.331838 LR 0.001000 Time 0.019776 -2023-02-13 17:51:15,862 - Epoch: [74][ 700/ 1207] Overall Loss 0.331687 Objective Loss 0.331687 LR 0.001000 Time 0.019763 -2023-02-13 17:51:16,050 - Epoch: [74][ 710/ 1207] Overall Loss 0.331844 Objective Loss 0.331844 LR 0.001000 Time 0.019750 -2023-02-13 17:51:16,238 - Epoch: [74][ 720/ 1207] Overall Loss 0.331394 Objective Loss 0.331394 LR 0.001000 Time 0.019736 -2023-02-13 17:51:16,426 - Epoch: [74][ 730/ 1207] Overall Loss 0.331920 Objective Loss 0.331920 LR 0.001000 Time 0.019722 -2023-02-13 17:51:16,615 - Epoch: [74][ 740/ 1207] Overall Loss 0.332072 Objective Loss 0.332072 LR 0.001000 Time 0.019710 -2023-02-13 17:51:16,803 - Epoch: [74][ 750/ 1207] Overall Loss 0.331813 Objective Loss 0.331813 LR 0.001000 Time 0.019698 -2023-02-13 17:51:16,992 - Epoch: [74][ 760/ 1207] Overall Loss 0.332119 Objective Loss 0.332119 LR 0.001000 Time 0.019686 -2023-02-13 17:51:17,179 - Epoch: [74][ 770/ 1207] Overall Loss 0.332036 Objective Loss 0.332036 LR 0.001000 Time 0.019674 -2023-02-13 17:51:17,368 - Epoch: [74][ 780/ 1207] Overall Loss 0.331987 Objective Loss 0.331987 LR 0.001000 Time 0.019663 -2023-02-13 17:51:17,556 - Epoch: [74][ 790/ 1207] Overall Loss 0.331808 Objective Loss 0.331808 LR 0.001000 Time 0.019652 -2023-02-13 17:51:17,744 - Epoch: [74][ 800/ 1207] Overall Loss 0.331556 Objective Loss 0.331556 LR 0.001000 Time 0.019640 -2023-02-13 17:51:17,932 - Epoch: [74][ 810/ 1207] Overall Loss 0.331926 Objective Loss 0.331926 LR 0.001000 Time 0.019630 -2023-02-13 17:51:18,121 - Epoch: [74][ 820/ 1207] Overall Loss 0.331872 Objective Loss 0.331872 LR 0.001000 Time 0.019620 -2023-02-13 17:51:18,309 - Epoch: [74][ 830/ 1207] Overall Loss 0.331948 Objective Loss 0.331948 LR 0.001000 Time 0.019610 -2023-02-13 17:51:18,497 - Epoch: [74][ 840/ 1207] Overall Loss 0.331866 Objective Loss 0.331866 LR 0.001000 Time 0.019600 -2023-02-13 17:51:18,685 - Epoch: [74][ 850/ 1207] Overall Loss 0.331588 Objective Loss 0.331588 LR 0.001000 Time 0.019590 -2023-02-13 17:51:18,874 - Epoch: [74][ 860/ 1207] Overall Loss 0.331727 Objective Loss 0.331727 LR 0.001000 Time 0.019582 -2023-02-13 17:51:19,062 - Epoch: [74][ 870/ 1207] Overall Loss 0.331543 Objective Loss 0.331543 LR 0.001000 Time 0.019572 -2023-02-13 17:51:19,250 - Epoch: [74][ 880/ 1207] Overall Loss 0.331527 Objective Loss 0.331527 LR 0.001000 Time 0.019563 -2023-02-13 17:51:19,438 - Epoch: [74][ 890/ 1207] Overall Loss 0.331765 Objective Loss 0.331765 LR 0.001000 Time 0.019554 -2023-02-13 17:51:19,626 - Epoch: [74][ 900/ 1207] Overall Loss 0.331259 Objective Loss 0.331259 LR 0.001000 Time 0.019545 -2023-02-13 17:51:19,814 - Epoch: [74][ 910/ 1207] Overall Loss 0.331302 Objective Loss 0.331302 LR 0.001000 Time 0.019537 -2023-02-13 17:51:20,003 - Epoch: [74][ 920/ 1207] Overall Loss 0.331245 Objective Loss 0.331245 LR 0.001000 Time 0.019529 -2023-02-13 17:51:20,191 - Epoch: [74][ 930/ 1207] Overall Loss 0.331306 Objective Loss 0.331306 LR 0.001000 Time 0.019521 -2023-02-13 17:51:20,379 - Epoch: [74][ 940/ 1207] Overall Loss 0.331575 Objective Loss 0.331575 LR 0.001000 Time 0.019513 -2023-02-13 17:51:20,567 - Epoch: [74][ 950/ 1207] Overall Loss 0.331707 Objective Loss 0.331707 LR 0.001000 Time 0.019505 -2023-02-13 17:51:20,754 - Epoch: [74][ 960/ 1207] Overall Loss 0.331511 Objective Loss 0.331511 LR 0.001000 Time 0.019497 -2023-02-13 17:51:20,944 - Epoch: [74][ 970/ 1207] Overall Loss 0.331420 Objective Loss 0.331420 LR 0.001000 Time 0.019490 -2023-02-13 17:51:21,132 - Epoch: [74][ 980/ 1207] Overall Loss 0.331590 Objective Loss 0.331590 LR 0.001000 Time 0.019483 -2023-02-13 17:51:21,321 - Epoch: [74][ 990/ 1207] Overall Loss 0.331484 Objective Loss 0.331484 LR 0.001000 Time 0.019477 -2023-02-13 17:51:21,509 - Epoch: [74][ 1000/ 1207] Overall Loss 0.331262 Objective Loss 0.331262 LR 0.001000 Time 0.019469 -2023-02-13 17:51:21,697 - Epoch: [74][ 1010/ 1207] Overall Loss 0.331148 Objective Loss 0.331148 LR 0.001000 Time 0.019463 -2023-02-13 17:51:21,886 - Epoch: [74][ 1020/ 1207] Overall Loss 0.331152 Objective Loss 0.331152 LR 0.001000 Time 0.019457 -2023-02-13 17:51:22,075 - Epoch: [74][ 1030/ 1207] Overall Loss 0.330933 Objective Loss 0.330933 LR 0.001000 Time 0.019451 -2023-02-13 17:51:22,263 - Epoch: [74][ 1040/ 1207] Overall Loss 0.330971 Objective Loss 0.330971 LR 0.001000 Time 0.019445 -2023-02-13 17:51:22,451 - Epoch: [74][ 1050/ 1207] Overall Loss 0.330887 Objective Loss 0.330887 LR 0.001000 Time 0.019439 -2023-02-13 17:51:22,640 - Epoch: [74][ 1060/ 1207] Overall Loss 0.331045 Objective Loss 0.331045 LR 0.001000 Time 0.019432 -2023-02-13 17:51:22,829 - Epoch: [74][ 1070/ 1207] Overall Loss 0.330878 Objective Loss 0.330878 LR 0.001000 Time 0.019427 -2023-02-13 17:51:23,017 - Epoch: [74][ 1080/ 1207] Overall Loss 0.330898 Objective Loss 0.330898 LR 0.001000 Time 0.019422 -2023-02-13 17:51:23,205 - Epoch: [74][ 1090/ 1207] Overall Loss 0.330923 Objective Loss 0.330923 LR 0.001000 Time 0.019415 -2023-02-13 17:51:23,394 - Epoch: [74][ 1100/ 1207] Overall Loss 0.331097 Objective Loss 0.331097 LR 0.001000 Time 0.019410 -2023-02-13 17:51:23,582 - Epoch: [74][ 1110/ 1207] Overall Loss 0.331235 Objective Loss 0.331235 LR 0.001000 Time 0.019404 -2023-02-13 17:51:23,771 - Epoch: [74][ 1120/ 1207] Overall Loss 0.331301 Objective Loss 0.331301 LR 0.001000 Time 0.019399 -2023-02-13 17:51:23,959 - Epoch: [74][ 1130/ 1207] Overall Loss 0.331127 Objective Loss 0.331127 LR 0.001000 Time 0.019394 -2023-02-13 17:51:24,147 - Epoch: [74][ 1140/ 1207] Overall Loss 0.331312 Objective Loss 0.331312 LR 0.001000 Time 0.019388 -2023-02-13 17:51:24,335 - Epoch: [74][ 1150/ 1207] Overall Loss 0.331148 Objective Loss 0.331148 LR 0.001000 Time 0.019383 -2023-02-13 17:51:24,523 - Epoch: [74][ 1160/ 1207] Overall Loss 0.331289 Objective Loss 0.331289 LR 0.001000 Time 0.019378 -2023-02-13 17:51:24,711 - Epoch: [74][ 1170/ 1207] Overall Loss 0.331187 Objective Loss 0.331187 LR 0.001000 Time 0.019372 -2023-02-13 17:51:24,900 - Epoch: [74][ 1180/ 1207] Overall Loss 0.331152 Objective Loss 0.331152 LR 0.001000 Time 0.019368 -2023-02-13 17:51:25,088 - Epoch: [74][ 1190/ 1207] Overall Loss 0.331196 Objective Loss 0.331196 LR 0.001000 Time 0.019363 -2023-02-13 17:51:25,332 - Epoch: [74][ 1200/ 1207] Overall Loss 0.331109 Objective Loss 0.331109 LR 0.001000 Time 0.019405 -2023-02-13 17:51:25,446 - Epoch: [74][ 1207/ 1207] Overall Loss 0.331231 Objective Loss 0.331231 Top1 79.573171 Top5 95.426829 LR 0.001000 Time 0.019387 -2023-02-13 17:51:25,518 - --- validate (epoch=74)----------- -2023-02-13 17:51:25,518 - 34311 samples (256 per mini-batch) -2023-02-13 17:51:25,925 - Epoch: [74][ 10/ 135] Loss 0.361032 Top1 81.796875 Top5 97.226562 -2023-02-13 17:51:26,059 - Epoch: [74][ 20/ 135] Loss 0.366014 Top1 81.347656 Top5 97.128906 -2023-02-13 17:51:26,191 - Epoch: [74][ 30/ 135] Loss 0.375068 Top1 81.354167 Top5 97.330729 -2023-02-13 17:51:26,322 - Epoch: [74][ 40/ 135] Loss 0.368696 Top1 81.591797 Top5 97.353516 -2023-02-13 17:51:26,455 - Epoch: [74][ 50/ 135] Loss 0.363956 Top1 81.796875 Top5 97.375000 -2023-02-13 17:51:26,587 - Epoch: [74][ 60/ 135] Loss 0.363141 Top1 81.757812 Top5 97.363281 -2023-02-13 17:51:26,719 - Epoch: [74][ 70/ 135] Loss 0.359108 Top1 81.768973 Top5 97.399554 -2023-02-13 17:51:26,851 - Epoch: [74][ 80/ 135] Loss 0.359064 Top1 81.777344 Top5 97.368164 -2023-02-13 17:51:26,983 - Epoch: [74][ 90/ 135] Loss 0.359428 Top1 81.770833 Top5 97.326389 -2023-02-13 17:51:27,115 - Epoch: [74][ 100/ 135] Loss 0.359388 Top1 81.738281 Top5 97.363281 -2023-02-13 17:51:27,243 - Epoch: [74][ 110/ 135] Loss 0.362114 Top1 81.598011 Top5 97.276278 -2023-02-13 17:51:27,378 - Epoch: [74][ 120/ 135] Loss 0.362333 Top1 81.695964 Top5 97.272135 -2023-02-13 17:51:27,508 - Epoch: [74][ 130/ 135] Loss 0.361438 Top1 81.799880 Top5 97.244591 -2023-02-13 17:51:27,552 - Epoch: [74][ 135/ 135] Loss 0.363853 Top1 81.857130 Top5 97.263268 -2023-02-13 17:51:27,633 - ==> Top1: 81.857 Top5: 97.263 Loss: 0.364 - -2023-02-13 17:51:27,633 - ==> Confusion: -[[ 790 5 8 3 9 3 1 3 2 102 2 5 0 4 8 3 3 4 2 3 7] - [ 1 933 2 3 3 33 4 10 4 2 5 0 1 1 1 3 8 0 9 3 7] - [ 9 2 942 21 3 2 24 8 1 2 5 2 1 5 4 5 3 1 10 4 4] - [ 3 2 16 907 3 5 2 3 4 2 10 1 6 4 13 0 4 4 19 1 7] - [ 11 10 3 1 977 11 1 1 1 12 0 5 1 2 8 5 4 4 2 2 5] - [ 3 27 3 5 8 954 1 16 2 6 2 5 4 15 1 2 2 2 1 6 5] - [ 3 6 20 4 1 7 1021 5 1 2 5 1 0 1 0 4 1 3 4 6 4] - [ 1 25 11 3 4 38 2 877 3 2 4 2 1 1 2 0 0 2 29 11 6] - [ 17 2 0 2 1 0 1 1 893 45 8 1 1 5 19 3 2 2 4 0 2] - [ 45 2 5 0 5 0 0 0 45 874 0 0 1 16 7 3 1 1 1 0 6] - [ 2 2 7 5 3 4 3 2 24 1 968 2 3 5 2 1 1 1 11 0 4] - [ 3 7 6 0 6 23 0 7 2 4 0 878 24 8 3 4 3 9 4 11 3] - [ 1 1 2 3 2 4 0 2 1 0 1 36 853 1 2 8 0 26 5 3 8] - [ 5 6 3 0 7 16 0 2 21 31 9 6 1 893 7 3 4 3 1 4 2] - [ 10 3 4 17 7 4 1 0 21 8 3 3 3 6 974 3 1 4 13 0 7] - [ 4 2 9 0 8 1 7 0 1 0 0 11 2 7 1 966 4 12 0 7 4] - [ 3 6 1 1 6 2 0 0 3 0 1 1 3 4 1 13 999 0 3 4 10] - [ 4 3 2 6 4 1 3 1 2 1 0 9 11 2 2 18 1 974 1 0 6] - [ 3 8 11 18 3 1 0 21 4 0 4 3 4 1 13 1 0 2 982 1 6] - [ 2 1 3 1 3 8 7 20 1 0 1 19 4 5 0 7 0 1 4 1053 8] - [ 159 277 286 176 160 245 88 155 143 121 227 128 337 314 187 141 338 109 197 268 9378]] - -2023-02-13 17:51:27,635 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:51:27,635 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:51:27,641 - - -2023-02-13 17:51:27,641 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:51:28,627 - Epoch: [75][ 10/ 1207] Overall Loss 0.323787 Objective Loss 0.323787 LR 0.001000 Time 0.098499 -2023-02-13 17:51:28,827 - Epoch: [75][ 20/ 1207] Overall Loss 0.324776 Objective Loss 0.324776 LR 0.001000 Time 0.059237 -2023-02-13 17:51:29,024 - Epoch: [75][ 30/ 1207] Overall Loss 0.325204 Objective Loss 0.325204 LR 0.001000 Time 0.046047 -2023-02-13 17:51:29,218 - Epoch: [75][ 40/ 1207] Overall Loss 0.332842 Objective Loss 0.332842 LR 0.001000 Time 0.039385 -2023-02-13 17:51:29,415 - Epoch: [75][ 50/ 1207] Overall Loss 0.332275 Objective Loss 0.332275 LR 0.001000 Time 0.035428 -2023-02-13 17:51:29,609 - Epoch: [75][ 60/ 1207] Overall Loss 0.328533 Objective Loss 0.328533 LR 0.001000 Time 0.032753 -2023-02-13 17:51:29,805 - Epoch: [75][ 70/ 1207] Overall Loss 0.328464 Objective Loss 0.328464 LR 0.001000 Time 0.030875 -2023-02-13 17:51:30,000 - Epoch: [75][ 80/ 1207] Overall Loss 0.327232 Objective Loss 0.327232 LR 0.001000 Time 0.029439 -2023-02-13 17:51:30,196 - Epoch: [75][ 90/ 1207] Overall Loss 0.325711 Objective Loss 0.325711 LR 0.001000 Time 0.028349 -2023-02-13 17:51:30,390 - Epoch: [75][ 100/ 1207] Overall Loss 0.322308 Objective Loss 0.322308 LR 0.001000 Time 0.027451 -2023-02-13 17:51:30,588 - Epoch: [75][ 110/ 1207] Overall Loss 0.321317 Objective Loss 0.321317 LR 0.001000 Time 0.026744 -2023-02-13 17:51:30,782 - Epoch: [75][ 120/ 1207] Overall Loss 0.319963 Objective Loss 0.319963 LR 0.001000 Time 0.026135 -2023-02-13 17:51:30,983 - Epoch: [75][ 130/ 1207] Overall Loss 0.322041 Objective Loss 0.322041 LR 0.001000 Time 0.025664 -2023-02-13 17:51:31,179 - Epoch: [75][ 140/ 1207] Overall Loss 0.320933 Objective Loss 0.320933 LR 0.001000 Time 0.025226 -2023-02-13 17:51:31,378 - Epoch: [75][ 150/ 1207] Overall Loss 0.320714 Objective Loss 0.320714 LR 0.001000 Time 0.024872 -2023-02-13 17:51:31,575 - Epoch: [75][ 160/ 1207] Overall Loss 0.321209 Objective Loss 0.321209 LR 0.001000 Time 0.024542 -2023-02-13 17:51:31,774 - Epoch: [75][ 170/ 1207] Overall Loss 0.321079 Objective Loss 0.321079 LR 0.001000 Time 0.024270 -2023-02-13 17:51:31,972 - Epoch: [75][ 180/ 1207] Overall Loss 0.320712 Objective Loss 0.320712 LR 0.001000 Time 0.024017 -2023-02-13 17:51:32,171 - Epoch: [75][ 190/ 1207] Overall Loss 0.320274 Objective Loss 0.320274 LR 0.001000 Time 0.023798 -2023-02-13 17:51:32,367 - Epoch: [75][ 200/ 1207] Overall Loss 0.319887 Objective Loss 0.319887 LR 0.001000 Time 0.023589 -2023-02-13 17:51:32,566 - Epoch: [75][ 210/ 1207] Overall Loss 0.320152 Objective Loss 0.320152 LR 0.001000 Time 0.023413 -2023-02-13 17:51:32,763 - Epoch: [75][ 220/ 1207] Overall Loss 0.321199 Objective Loss 0.321199 LR 0.001000 Time 0.023241 -2023-02-13 17:51:32,963 - Epoch: [75][ 230/ 1207] Overall Loss 0.320067 Objective Loss 0.320067 LR 0.001000 Time 0.023097 -2023-02-13 17:51:33,159 - Epoch: [75][ 240/ 1207] Overall Loss 0.320194 Objective Loss 0.320194 LR 0.001000 Time 0.022952 -2023-02-13 17:51:33,359 - Epoch: [75][ 250/ 1207] Overall Loss 0.321542 Objective Loss 0.321542 LR 0.001000 Time 0.022831 -2023-02-13 17:51:33,555 - Epoch: [75][ 260/ 1207] Overall Loss 0.322582 Objective Loss 0.322582 LR 0.001000 Time 0.022707 -2023-02-13 17:51:33,756 - Epoch: [75][ 270/ 1207] Overall Loss 0.323134 Objective Loss 0.323134 LR 0.001000 Time 0.022607 -2023-02-13 17:51:33,953 - Epoch: [75][ 280/ 1207] Overall Loss 0.325497 Objective Loss 0.325497 LR 0.001000 Time 0.022502 -2023-02-13 17:51:34,153 - Epoch: [75][ 290/ 1207] Overall Loss 0.326740 Objective Loss 0.326740 LR 0.001000 Time 0.022414 -2023-02-13 17:51:34,350 - Epoch: [75][ 300/ 1207] Overall Loss 0.326491 Objective Loss 0.326491 LR 0.001000 Time 0.022322 -2023-02-13 17:51:34,548 - Epoch: [75][ 310/ 1207] Overall Loss 0.327145 Objective Loss 0.327145 LR 0.001000 Time 0.022242 -2023-02-13 17:51:34,744 - Epoch: [75][ 320/ 1207] Overall Loss 0.327305 Objective Loss 0.327305 LR 0.001000 Time 0.022158 -2023-02-13 17:51:34,945 - Epoch: [75][ 330/ 1207] Overall Loss 0.327851 Objective Loss 0.327851 LR 0.001000 Time 0.022093 -2023-02-13 17:51:35,141 - Epoch: [75][ 340/ 1207] Overall Loss 0.328491 Objective Loss 0.328491 LR 0.001000 Time 0.022020 -2023-02-13 17:51:35,340 - Epoch: [75][ 350/ 1207] Overall Loss 0.328557 Objective Loss 0.328557 LR 0.001000 Time 0.021958 -2023-02-13 17:51:35,537 - Epoch: [75][ 360/ 1207] Overall Loss 0.328897 Objective Loss 0.328897 LR 0.001000 Time 0.021893 -2023-02-13 17:51:35,736 - Epoch: [75][ 370/ 1207] Overall Loss 0.329867 Objective Loss 0.329867 LR 0.001000 Time 0.021838 -2023-02-13 17:51:35,934 - Epoch: [75][ 380/ 1207] Overall Loss 0.331011 Objective Loss 0.331011 LR 0.001000 Time 0.021783 -2023-02-13 17:51:36,132 - Epoch: [75][ 390/ 1207] Overall Loss 0.330454 Objective Loss 0.330454 LR 0.001000 Time 0.021733 -2023-02-13 17:51:36,329 - Epoch: [75][ 400/ 1207] Overall Loss 0.329837 Objective Loss 0.329837 LR 0.001000 Time 0.021680 -2023-02-13 17:51:36,528 - Epoch: [75][ 410/ 1207] Overall Loss 0.329655 Objective Loss 0.329655 LR 0.001000 Time 0.021635 -2023-02-13 17:51:36,722 - Epoch: [75][ 420/ 1207] Overall Loss 0.330172 Objective Loss 0.330172 LR 0.001000 Time 0.021582 -2023-02-13 17:51:36,921 - Epoch: [75][ 430/ 1207] Overall Loss 0.331233 Objective Loss 0.331233 LR 0.001000 Time 0.021541 -2023-02-13 17:51:37,115 - Epoch: [75][ 440/ 1207] Overall Loss 0.331240 Objective Loss 0.331240 LR 0.001000 Time 0.021492 -2023-02-13 17:51:37,312 - Epoch: [75][ 450/ 1207] Overall Loss 0.330707 Objective Loss 0.330707 LR 0.001000 Time 0.021452 -2023-02-13 17:51:37,506 - Epoch: [75][ 460/ 1207] Overall Loss 0.330117 Objective Loss 0.330117 LR 0.001000 Time 0.021406 -2023-02-13 17:51:37,703 - Epoch: [75][ 470/ 1207] Overall Loss 0.330423 Objective Loss 0.330423 LR 0.001000 Time 0.021369 -2023-02-13 17:51:37,898 - Epoch: [75][ 480/ 1207] Overall Loss 0.330666 Objective Loss 0.330666 LR 0.001000 Time 0.021329 -2023-02-13 17:51:38,095 - Epoch: [75][ 490/ 1207] Overall Loss 0.330429 Objective Loss 0.330429 LR 0.001000 Time 0.021296 -2023-02-13 17:51:38,290 - Epoch: [75][ 500/ 1207] Overall Loss 0.330413 Objective Loss 0.330413 LR 0.001000 Time 0.021259 -2023-02-13 17:51:38,488 - Epoch: [75][ 510/ 1207] Overall Loss 0.330451 Objective Loss 0.330451 LR 0.001000 Time 0.021229 -2023-02-13 17:51:38,682 - Epoch: [75][ 520/ 1207] Overall Loss 0.329871 Objective Loss 0.329871 LR 0.001000 Time 0.021194 -2023-02-13 17:51:38,880 - Epoch: [75][ 530/ 1207] Overall Loss 0.329279 Objective Loss 0.329279 LR 0.001000 Time 0.021167 -2023-02-13 17:51:39,075 - Epoch: [75][ 540/ 1207] Overall Loss 0.329367 Objective Loss 0.329367 LR 0.001000 Time 0.021135 -2023-02-13 17:51:39,272 - Epoch: [75][ 550/ 1207] Overall Loss 0.329501 Objective Loss 0.329501 LR 0.001000 Time 0.021109 -2023-02-13 17:51:39,467 - Epoch: [75][ 560/ 1207] Overall Loss 0.329582 Objective Loss 0.329582 LR 0.001000 Time 0.021079 -2023-02-13 17:51:39,665 - Epoch: [75][ 570/ 1207] Overall Loss 0.330655 Objective Loss 0.330655 LR 0.001000 Time 0.021056 -2023-02-13 17:51:39,860 - Epoch: [75][ 580/ 1207] Overall Loss 0.331387 Objective Loss 0.331387 LR 0.001000 Time 0.021028 -2023-02-13 17:51:40,058 - Epoch: [75][ 590/ 1207] Overall Loss 0.331764 Objective Loss 0.331764 LR 0.001000 Time 0.021007 -2023-02-13 17:51:40,253 - Epoch: [75][ 600/ 1207] Overall Loss 0.332084 Objective Loss 0.332084 LR 0.001000 Time 0.020980 -2023-02-13 17:51:40,450 - Epoch: [75][ 610/ 1207] Overall Loss 0.332316 Objective Loss 0.332316 LR 0.001000 Time 0.020960 -2023-02-13 17:51:40,645 - Epoch: [75][ 620/ 1207] Overall Loss 0.332411 Objective Loss 0.332411 LR 0.001000 Time 0.020936 -2023-02-13 17:51:40,843 - Epoch: [75][ 630/ 1207] Overall Loss 0.332520 Objective Loss 0.332520 LR 0.001000 Time 0.020917 -2023-02-13 17:51:41,038 - Epoch: [75][ 640/ 1207] Overall Loss 0.332579 Objective Loss 0.332579 LR 0.001000 Time 0.020894 -2023-02-13 17:51:41,236 - Epoch: [75][ 650/ 1207] Overall Loss 0.332628 Objective Loss 0.332628 LR 0.001000 Time 0.020876 -2023-02-13 17:51:41,430 - Epoch: [75][ 660/ 1207] Overall Loss 0.332894 Objective Loss 0.332894 LR 0.001000 Time 0.020853 -2023-02-13 17:51:41,627 - Epoch: [75][ 670/ 1207] Overall Loss 0.333126 Objective Loss 0.333126 LR 0.001000 Time 0.020836 -2023-02-13 17:51:41,823 - Epoch: [75][ 680/ 1207] Overall Loss 0.333022 Objective Loss 0.333022 LR 0.001000 Time 0.020816 -2023-02-13 17:51:42,021 - Epoch: [75][ 690/ 1207] Overall Loss 0.332617 Objective Loss 0.332617 LR 0.001000 Time 0.020802 -2023-02-13 17:51:42,216 - Epoch: [75][ 700/ 1207] Overall Loss 0.332873 Objective Loss 0.332873 LR 0.001000 Time 0.020783 -2023-02-13 17:51:42,416 - Epoch: [75][ 710/ 1207] Overall Loss 0.333207 Objective Loss 0.333207 LR 0.001000 Time 0.020770 -2023-02-13 17:51:42,612 - Epoch: [75][ 720/ 1207] Overall Loss 0.333123 Objective Loss 0.333123 LR 0.001000 Time 0.020754 -2023-02-13 17:51:42,810 - Epoch: [75][ 730/ 1207] Overall Loss 0.332868 Objective Loss 0.332868 LR 0.001000 Time 0.020741 -2023-02-13 17:51:43,007 - Epoch: [75][ 740/ 1207] Overall Loss 0.332840 Objective Loss 0.332840 LR 0.001000 Time 0.020726 -2023-02-13 17:51:43,206 - Epoch: [75][ 750/ 1207] Overall Loss 0.332420 Objective Loss 0.332420 LR 0.001000 Time 0.020715 -2023-02-13 17:51:43,402 - Epoch: [75][ 760/ 1207] Overall Loss 0.332631 Objective Loss 0.332631 LR 0.001000 Time 0.020699 -2023-02-13 17:51:43,601 - Epoch: [75][ 770/ 1207] Overall Loss 0.332237 Objective Loss 0.332237 LR 0.001000 Time 0.020688 -2023-02-13 17:51:43,798 - Epoch: [75][ 780/ 1207] Overall Loss 0.332554 Objective Loss 0.332554 LR 0.001000 Time 0.020675 -2023-02-13 17:51:43,997 - Epoch: [75][ 790/ 1207] Overall Loss 0.332051 Objective Loss 0.332051 LR 0.001000 Time 0.020665 -2023-02-13 17:51:44,194 - Epoch: [75][ 800/ 1207] Overall Loss 0.331953 Objective Loss 0.331953 LR 0.001000 Time 0.020653 -2023-02-13 17:51:44,393 - Epoch: [75][ 810/ 1207] Overall Loss 0.331893 Objective Loss 0.331893 LR 0.001000 Time 0.020643 -2023-02-13 17:51:44,589 - Epoch: [75][ 820/ 1207] Overall Loss 0.331662 Objective Loss 0.331662 LR 0.001000 Time 0.020630 -2023-02-13 17:51:44,788 - Epoch: [75][ 830/ 1207] Overall Loss 0.331400 Objective Loss 0.331400 LR 0.001000 Time 0.020620 -2023-02-13 17:51:44,983 - Epoch: [75][ 840/ 1207] Overall Loss 0.331110 Objective Loss 0.331110 LR 0.001000 Time 0.020607 -2023-02-13 17:51:45,180 - Epoch: [75][ 850/ 1207] Overall Loss 0.330998 Objective Loss 0.330998 LR 0.001000 Time 0.020596 -2023-02-13 17:51:45,375 - Epoch: [75][ 860/ 1207] Overall Loss 0.330536 Objective Loss 0.330536 LR 0.001000 Time 0.020582 -2023-02-13 17:51:45,572 - Epoch: [75][ 870/ 1207] Overall Loss 0.330753 Objective Loss 0.330753 LR 0.001000 Time 0.020572 -2023-02-13 17:51:45,767 - Epoch: [75][ 880/ 1207] Overall Loss 0.330763 Objective Loss 0.330763 LR 0.001000 Time 0.020559 -2023-02-13 17:51:45,965 - Epoch: [75][ 890/ 1207] Overall Loss 0.330771 Objective Loss 0.330771 LR 0.001000 Time 0.020551 -2023-02-13 17:51:46,160 - Epoch: [75][ 900/ 1207] Overall Loss 0.330726 Objective Loss 0.330726 LR 0.001000 Time 0.020538 -2023-02-13 17:51:46,357 - Epoch: [75][ 910/ 1207] Overall Loss 0.330284 Objective Loss 0.330284 LR 0.001000 Time 0.020529 -2023-02-13 17:51:46,552 - Epoch: [75][ 920/ 1207] Overall Loss 0.330170 Objective Loss 0.330170 LR 0.001000 Time 0.020517 -2023-02-13 17:51:46,749 - Epoch: [75][ 930/ 1207] Overall Loss 0.330354 Objective Loss 0.330354 LR 0.001000 Time 0.020508 -2023-02-13 17:51:46,939 - Epoch: [75][ 940/ 1207] Overall Loss 0.330310 Objective Loss 0.330310 LR 0.001000 Time 0.020491 -2023-02-13 17:51:47,129 - Epoch: [75][ 950/ 1207] Overall Loss 0.329959 Objective Loss 0.329959 LR 0.001000 Time 0.020475 -2023-02-13 17:51:47,319 - Epoch: [75][ 960/ 1207] Overall Loss 0.329796 Objective Loss 0.329796 LR 0.001000 Time 0.020459 -2023-02-13 17:51:47,509 - Epoch: [75][ 970/ 1207] Overall Loss 0.330046 Objective Loss 0.330046 LR 0.001000 Time 0.020443 -2023-02-13 17:51:47,698 - Epoch: [75][ 980/ 1207] Overall Loss 0.329860 Objective Loss 0.329860 LR 0.001000 Time 0.020428 -2023-02-13 17:51:47,889 - Epoch: [75][ 990/ 1207] Overall Loss 0.330140 Objective Loss 0.330140 LR 0.001000 Time 0.020414 -2023-02-13 17:51:48,080 - Epoch: [75][ 1000/ 1207] Overall Loss 0.330121 Objective Loss 0.330121 LR 0.001000 Time 0.020400 -2023-02-13 17:51:48,270 - Epoch: [75][ 1010/ 1207] Overall Loss 0.330486 Objective Loss 0.330486 LR 0.001000 Time 0.020386 -2023-02-13 17:51:48,460 - Epoch: [75][ 1020/ 1207] Overall Loss 0.330637 Objective Loss 0.330637 LR 0.001000 Time 0.020373 -2023-02-13 17:51:48,651 - Epoch: [75][ 1030/ 1207] Overall Loss 0.330843 Objective Loss 0.330843 LR 0.001000 Time 0.020359 -2023-02-13 17:51:48,841 - Epoch: [75][ 1040/ 1207] Overall Loss 0.330908 Objective Loss 0.330908 LR 0.001000 Time 0.020346 -2023-02-13 17:51:49,032 - Epoch: [75][ 1050/ 1207] Overall Loss 0.331106 Objective Loss 0.331106 LR 0.001000 Time 0.020334 -2023-02-13 17:51:49,222 - Epoch: [75][ 1060/ 1207] Overall Loss 0.331026 Objective Loss 0.331026 LR 0.001000 Time 0.020321 -2023-02-13 17:51:49,412 - Epoch: [75][ 1070/ 1207] Overall Loss 0.331104 Objective Loss 0.331104 LR 0.001000 Time 0.020308 -2023-02-13 17:51:49,602 - Epoch: [75][ 1080/ 1207] Overall Loss 0.331133 Objective Loss 0.331133 LR 0.001000 Time 0.020296 -2023-02-13 17:51:49,792 - Epoch: [75][ 1090/ 1207] Overall Loss 0.331084 Objective Loss 0.331084 LR 0.001000 Time 0.020284 -2023-02-13 17:51:49,982 - Epoch: [75][ 1100/ 1207] Overall Loss 0.331150 Objective Loss 0.331150 LR 0.001000 Time 0.020272 -2023-02-13 17:51:50,173 - Epoch: [75][ 1110/ 1207] Overall Loss 0.331120 Objective Loss 0.331120 LR 0.001000 Time 0.020261 -2023-02-13 17:51:50,363 - Epoch: [75][ 1120/ 1207] Overall Loss 0.331002 Objective Loss 0.331002 LR 0.001000 Time 0.020249 -2023-02-13 17:51:50,553 - Epoch: [75][ 1130/ 1207] Overall Loss 0.331270 Objective Loss 0.331270 LR 0.001000 Time 0.020238 -2023-02-13 17:51:50,743 - Epoch: [75][ 1140/ 1207] Overall Loss 0.331168 Objective Loss 0.331168 LR 0.001000 Time 0.020227 -2023-02-13 17:51:50,936 - Epoch: [75][ 1150/ 1207] Overall Loss 0.331000 Objective Loss 0.331000 LR 0.001000 Time 0.020218 -2023-02-13 17:51:51,126 - Epoch: [75][ 1160/ 1207] Overall Loss 0.330833 Objective Loss 0.330833 LR 0.001000 Time 0.020207 -2023-02-13 17:51:51,317 - Epoch: [75][ 1170/ 1207] Overall Loss 0.331022 Objective Loss 0.331022 LR 0.001000 Time 0.020197 -2023-02-13 17:51:51,507 - Epoch: [75][ 1180/ 1207] Overall Loss 0.331092 Objective Loss 0.331092 LR 0.001000 Time 0.020187 -2023-02-13 17:51:51,698 - Epoch: [75][ 1190/ 1207] Overall Loss 0.330927 Objective Loss 0.330927 LR 0.001000 Time 0.020178 -2023-02-13 17:51:51,946 - Epoch: [75][ 1200/ 1207] Overall Loss 0.331035 Objective Loss 0.331035 LR 0.001000 Time 0.020216 -2023-02-13 17:51:52,062 - Epoch: [75][ 1207/ 1207] Overall Loss 0.330907 Objective Loss 0.330907 Top1 84.451220 Top5 97.865854 LR 0.001000 Time 0.020194 -2023-02-13 17:51:52,134 - --- validate (epoch=75)----------- -2023-02-13 17:51:52,134 - 34311 samples (256 per mini-batch) -2023-02-13 17:51:52,540 - Epoch: [75][ 10/ 135] Loss 0.365153 Top1 81.835938 Top5 97.109375 -2023-02-13 17:51:52,663 - Epoch: [75][ 20/ 135] Loss 0.367195 Top1 81.269531 Top5 97.226562 -2023-02-13 17:51:52,785 - Epoch: [75][ 30/ 135] Loss 0.356902 Top1 81.731771 Top5 97.213542 -2023-02-13 17:51:52,916 - Epoch: [75][ 40/ 135] Loss 0.360535 Top1 81.630859 Top5 97.285156 -2023-02-13 17:51:53,058 - Epoch: [75][ 50/ 135] Loss 0.366615 Top1 81.507812 Top5 97.242188 -2023-02-13 17:51:53,194 - Epoch: [75][ 60/ 135] Loss 0.370897 Top1 81.334635 Top5 97.174479 -2023-02-13 17:51:53,322 - Epoch: [75][ 70/ 135] Loss 0.371156 Top1 81.177455 Top5 97.059152 -2023-02-13 17:51:53,454 - Epoch: [75][ 80/ 135] Loss 0.369747 Top1 81.264648 Top5 97.021484 -2023-02-13 17:51:53,582 - Epoch: [75][ 90/ 135] Loss 0.367899 Top1 81.228299 Top5 97.005208 -2023-02-13 17:51:53,715 - Epoch: [75][ 100/ 135] Loss 0.364783 Top1 81.281250 Top5 96.988281 -2023-02-13 17:51:53,843 - Epoch: [75][ 110/ 135] Loss 0.365199 Top1 81.260653 Top5 97.002841 -2023-02-13 17:51:53,975 - Epoch: [75][ 120/ 135] Loss 0.366787 Top1 81.276042 Top5 96.998698 -2023-02-13 17:51:54,106 - Epoch: [75][ 130/ 135] Loss 0.364983 Top1 81.379207 Top5 97.034255 -2023-02-13 17:51:54,154 - Epoch: [75][ 135/ 135] Loss 0.362729 Top1 81.419953 Top5 97.047594 -2023-02-13 17:51:54,225 - ==> Top1: 81.420 Top5: 97.048 Loss: 0.363 - -2023-02-13 17:51:54,226 - ==> Confusion: -[[ 835 4 1 0 7 4 0 5 9 73 1 5 0 3 2 3 0 5 3 1 6] - [ 3 904 0 4 4 42 6 28 9 1 2 3 2 2 1 4 5 0 8 1 4] - [ 8 1 931 14 2 2 24 21 0 2 5 4 5 3 4 6 3 2 10 4 7] - [ 7 2 21 876 2 7 2 3 4 2 11 2 7 3 23 1 2 11 21 0 9] - [ 25 13 0 1 953 15 1 4 3 9 1 7 0 5 6 4 7 3 0 3 6] - [ 6 20 2 1 3 968 1 23 3 4 2 13 3 12 0 2 2 0 2 1 2] - [ 4 5 15 1 0 10 1027 9 0 1 3 1 2 1 1 3 3 3 1 5 4] - [ 3 6 10 0 1 37 3 912 2 1 1 5 4 2 0 0 0 0 21 14 2] - [ 12 3 0 3 1 1 0 4 896 40 13 1 0 8 10 4 1 3 9 0 0] - [ 75 2 3 0 1 3 0 2 47 852 0 1 1 12 3 1 0 3 1 1 4] - [ 3 1 3 5 1 4 4 9 19 2 969 1 1 9 2 0 1 1 11 0 5] - [ 3 3 0 1 4 15 0 7 2 0 0 884 43 8 0 6 2 14 2 10 1] - [ 2 1 1 6 0 6 0 2 1 0 0 36 855 1 3 6 2 22 3 1 11] - [ 4 3 3 0 8 23 0 4 15 21 16 6 6 891 3 2 6 2 1 5 5] - [ 12 3 1 16 4 5 0 4 42 9 3 2 4 3 956 2 1 10 10 0 5] - [ 6 1 3 2 7 1 1 3 0 1 0 8 6 4 0 963 10 14 1 9 6] - [ 4 5 1 0 10 5 0 2 5 0 1 0 2 2 4 13 981 1 4 6 15] - [ 6 3 0 1 0 2 1 3 0 0 0 8 12 5 0 8 0 992 0 4 6] - [ 2 2 6 13 0 2 0 36 5 0 7 0 11 0 15 2 2 3 979 0 1] - [ 0 2 1 0 0 10 4 16 0 0 0 16 3 5 0 7 1 4 0 1067 12] - [ 203 272 212 137 119 316 91 256 159 121 262 143 353 320 158 131 251 157 222 306 9245]] - -2023-02-13 17:51:54,227 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:51:54,227 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:51:54,233 - - -2023-02-13 17:51:54,233 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:51:55,128 - Epoch: [76][ 10/ 1207] Overall Loss 0.314471 Objective Loss 0.314471 LR 0.001000 Time 0.089410 -2023-02-13 17:51:55,328 - Epoch: [76][ 20/ 1207] Overall Loss 0.319480 Objective Loss 0.319480 LR 0.001000 Time 0.054706 -2023-02-13 17:51:55,522 - Epoch: [76][ 30/ 1207] Overall Loss 0.332899 Objective Loss 0.332899 LR 0.001000 Time 0.042918 -2023-02-13 17:51:55,718 - Epoch: [76][ 40/ 1207] Overall Loss 0.339158 Objective Loss 0.339158 LR 0.001000 Time 0.037083 -2023-02-13 17:51:55,914 - Epoch: [76][ 50/ 1207] Overall Loss 0.331033 Objective Loss 0.331033 LR 0.001000 Time 0.033573 -2023-02-13 17:51:56,111 - Epoch: [76][ 60/ 1207] Overall Loss 0.335909 Objective Loss 0.335909 LR 0.001000 Time 0.031250 -2023-02-13 17:51:56,305 - Epoch: [76][ 70/ 1207] Overall Loss 0.333769 Objective Loss 0.333769 LR 0.001000 Time 0.029555 -2023-02-13 17:51:56,504 - Epoch: [76][ 80/ 1207] Overall Loss 0.333062 Objective Loss 0.333062 LR 0.001000 Time 0.028336 -2023-02-13 17:51:56,699 - Epoch: [76][ 90/ 1207] Overall Loss 0.331615 Objective Loss 0.331615 LR 0.001000 Time 0.027355 -2023-02-13 17:51:56,898 - Epoch: [76][ 100/ 1207] Overall Loss 0.331881 Objective Loss 0.331881 LR 0.001000 Time 0.026603 -2023-02-13 17:51:57,095 - Epoch: [76][ 110/ 1207] Overall Loss 0.333253 Objective Loss 0.333253 LR 0.001000 Time 0.025973 -2023-02-13 17:51:57,293 - Epoch: [76][ 120/ 1207] Overall Loss 0.332984 Objective Loss 0.332984 LR 0.001000 Time 0.025454 -2023-02-13 17:51:57,489 - Epoch: [76][ 130/ 1207] Overall Loss 0.332974 Objective Loss 0.332974 LR 0.001000 Time 0.025003 -2023-02-13 17:51:57,686 - Epoch: [76][ 140/ 1207] Overall Loss 0.332071 Objective Loss 0.332071 LR 0.001000 Time 0.024623 -2023-02-13 17:51:57,880 - Epoch: [76][ 150/ 1207] Overall Loss 0.331201 Objective Loss 0.331201 LR 0.001000 Time 0.024270 -2023-02-13 17:51:58,077 - Epoch: [76][ 160/ 1207] Overall Loss 0.329560 Objective Loss 0.329560 LR 0.001000 Time 0.023982 -2023-02-13 17:51:58,271 - Epoch: [76][ 170/ 1207] Overall Loss 0.330627 Objective Loss 0.330627 LR 0.001000 Time 0.023709 -2023-02-13 17:51:58,466 - Epoch: [76][ 180/ 1207] Overall Loss 0.329026 Objective Loss 0.329026 LR 0.001000 Time 0.023477 -2023-02-13 17:51:58,660 - Epoch: [76][ 190/ 1207] Overall Loss 0.327138 Objective Loss 0.327138 LR 0.001000 Time 0.023257 -2023-02-13 17:51:58,856 - Epoch: [76][ 200/ 1207] Overall Loss 0.326175 Objective Loss 0.326175 LR 0.001000 Time 0.023074 -2023-02-13 17:51:59,051 - Epoch: [76][ 210/ 1207] Overall Loss 0.326541 Objective Loss 0.326541 LR 0.001000 Time 0.022901 -2023-02-13 17:51:59,246 - Epoch: [76][ 220/ 1207] Overall Loss 0.327081 Objective Loss 0.327081 LR 0.001000 Time 0.022747 -2023-02-13 17:51:59,440 - Epoch: [76][ 230/ 1207] Overall Loss 0.327290 Objective Loss 0.327290 LR 0.001000 Time 0.022598 -2023-02-13 17:51:59,636 - Epoch: [76][ 240/ 1207] Overall Loss 0.327149 Objective Loss 0.327149 LR 0.001000 Time 0.022472 -2023-02-13 17:51:59,830 - Epoch: [76][ 250/ 1207] Overall Loss 0.326296 Objective Loss 0.326296 LR 0.001000 Time 0.022349 -2023-02-13 17:52:00,027 - Epoch: [76][ 260/ 1207] Overall Loss 0.326104 Objective Loss 0.326104 LR 0.001000 Time 0.022246 -2023-02-13 17:52:00,222 - Epoch: [76][ 270/ 1207] Overall Loss 0.325883 Objective Loss 0.325883 LR 0.001000 Time 0.022140 -2023-02-13 17:52:00,418 - Epoch: [76][ 280/ 1207] Overall Loss 0.325812 Objective Loss 0.325812 LR 0.001000 Time 0.022049 -2023-02-13 17:52:00,612 - Epoch: [76][ 290/ 1207] Overall Loss 0.325459 Objective Loss 0.325459 LR 0.001000 Time 0.021957 -2023-02-13 17:52:00,809 - Epoch: [76][ 300/ 1207] Overall Loss 0.326201 Objective Loss 0.326201 LR 0.001000 Time 0.021880 -2023-02-13 17:52:01,004 - Epoch: [76][ 310/ 1207] Overall Loss 0.327153 Objective Loss 0.327153 LR 0.001000 Time 0.021802 -2023-02-13 17:52:01,200 - Epoch: [76][ 320/ 1207] Overall Loss 0.327707 Objective Loss 0.327707 LR 0.001000 Time 0.021731 -2023-02-13 17:52:01,393 - Epoch: [76][ 330/ 1207] Overall Loss 0.327724 Objective Loss 0.327724 LR 0.001000 Time 0.021659 -2023-02-13 17:52:01,589 - Epoch: [76][ 340/ 1207] Overall Loss 0.327550 Objective Loss 0.327550 LR 0.001000 Time 0.021596 -2023-02-13 17:52:01,783 - Epoch: [76][ 350/ 1207] Overall Loss 0.327964 Objective Loss 0.327964 LR 0.001000 Time 0.021532 -2023-02-13 17:52:01,981 - Epoch: [76][ 360/ 1207] Overall Loss 0.328180 Objective Loss 0.328180 LR 0.001000 Time 0.021483 -2023-02-13 17:52:02,175 - Epoch: [76][ 370/ 1207] Overall Loss 0.328299 Objective Loss 0.328299 LR 0.001000 Time 0.021426 -2023-02-13 17:52:02,371 - Epoch: [76][ 380/ 1207] Overall Loss 0.327448 Objective Loss 0.327448 LR 0.001000 Time 0.021377 -2023-02-13 17:52:02,565 - Epoch: [76][ 390/ 1207] Overall Loss 0.328060 Objective Loss 0.328060 LR 0.001000 Time 0.021325 -2023-02-13 17:52:02,761 - Epoch: [76][ 400/ 1207] Overall Loss 0.329386 Objective Loss 0.329386 LR 0.001000 Time 0.021280 -2023-02-13 17:52:02,955 - Epoch: [76][ 410/ 1207] Overall Loss 0.329531 Objective Loss 0.329531 LR 0.001000 Time 0.021235 -2023-02-13 17:52:03,151 - Epoch: [76][ 420/ 1207] Overall Loss 0.329422 Objective Loss 0.329422 LR 0.001000 Time 0.021195 -2023-02-13 17:52:03,345 - Epoch: [76][ 430/ 1207] Overall Loss 0.328616 Objective Loss 0.328616 LR 0.001000 Time 0.021151 -2023-02-13 17:52:03,541 - Epoch: [76][ 440/ 1207] Overall Loss 0.327762 Objective Loss 0.327762 LR 0.001000 Time 0.021115 -2023-02-13 17:52:03,736 - Epoch: [76][ 450/ 1207] Overall Loss 0.327816 Objective Loss 0.327816 LR 0.001000 Time 0.021078 -2023-02-13 17:52:03,932 - Epoch: [76][ 460/ 1207] Overall Loss 0.328129 Objective Loss 0.328129 LR 0.001000 Time 0.021046 -2023-02-13 17:52:04,127 - Epoch: [76][ 470/ 1207] Overall Loss 0.327977 Objective Loss 0.327977 LR 0.001000 Time 0.021013 -2023-02-13 17:52:04,324 - Epoch: [76][ 480/ 1207] Overall Loss 0.328289 Objective Loss 0.328289 LR 0.001000 Time 0.020984 -2023-02-13 17:52:04,519 - Epoch: [76][ 490/ 1207] Overall Loss 0.328427 Objective Loss 0.328427 LR 0.001000 Time 0.020952 -2023-02-13 17:52:04,715 - Epoch: [76][ 500/ 1207] Overall Loss 0.328507 Objective Loss 0.328507 LR 0.001000 Time 0.020925 -2023-02-13 17:52:04,909 - Epoch: [76][ 510/ 1207] Overall Loss 0.328986 Objective Loss 0.328986 LR 0.001000 Time 0.020895 -2023-02-13 17:52:05,106 - Epoch: [76][ 520/ 1207] Overall Loss 0.329476 Objective Loss 0.329476 LR 0.001000 Time 0.020871 -2023-02-13 17:52:05,300 - Epoch: [76][ 530/ 1207] Overall Loss 0.329809 Objective Loss 0.329809 LR 0.001000 Time 0.020841 -2023-02-13 17:52:05,496 - Epoch: [76][ 540/ 1207] Overall Loss 0.329861 Objective Loss 0.329861 LR 0.001000 Time 0.020818 -2023-02-13 17:52:05,689 - Epoch: [76][ 550/ 1207] Overall Loss 0.329417 Objective Loss 0.329417 LR 0.001000 Time 0.020790 -2023-02-13 17:52:05,886 - Epoch: [76][ 560/ 1207] Overall Loss 0.329211 Objective Loss 0.329211 LR 0.001000 Time 0.020769 -2023-02-13 17:52:06,083 - Epoch: [76][ 570/ 1207] Overall Loss 0.328756 Objective Loss 0.328756 LR 0.001000 Time 0.020750 -2023-02-13 17:52:06,281 - Epoch: [76][ 580/ 1207] Overall Loss 0.329038 Objective Loss 0.329038 LR 0.001000 Time 0.020734 -2023-02-13 17:52:06,478 - Epoch: [76][ 590/ 1207] Overall Loss 0.328935 Objective Loss 0.328935 LR 0.001000 Time 0.020715 -2023-02-13 17:52:06,676 - Epoch: [76][ 600/ 1207] Overall Loss 0.329685 Objective Loss 0.329685 LR 0.001000 Time 0.020700 -2023-02-13 17:52:06,874 - Epoch: [76][ 610/ 1207] Overall Loss 0.329856 Objective Loss 0.329856 LR 0.001000 Time 0.020684 -2023-02-13 17:52:07,073 - Epoch: [76][ 620/ 1207] Overall Loss 0.329753 Objective Loss 0.329753 LR 0.001000 Time 0.020671 -2023-02-13 17:52:07,271 - Epoch: [76][ 630/ 1207] Overall Loss 0.329942 Objective Loss 0.329942 LR 0.001000 Time 0.020656 -2023-02-13 17:52:07,469 - Epoch: [76][ 640/ 1207] Overall Loss 0.329580 Objective Loss 0.329580 LR 0.001000 Time 0.020642 -2023-02-13 17:52:07,666 - Epoch: [76][ 650/ 1207] Overall Loss 0.329532 Objective Loss 0.329532 LR 0.001000 Time 0.020628 -2023-02-13 17:52:07,864 - Epoch: [76][ 660/ 1207] Overall Loss 0.329483 Objective Loss 0.329483 LR 0.001000 Time 0.020615 -2023-02-13 17:52:08,063 - Epoch: [76][ 670/ 1207] Overall Loss 0.330052 Objective Loss 0.330052 LR 0.001000 Time 0.020604 -2023-02-13 17:52:08,262 - Epoch: [76][ 680/ 1207] Overall Loss 0.330308 Objective Loss 0.330308 LR 0.001000 Time 0.020592 -2023-02-13 17:52:08,459 - Epoch: [76][ 690/ 1207] Overall Loss 0.330650 Objective Loss 0.330650 LR 0.001000 Time 0.020579 -2023-02-13 17:52:08,658 - Epoch: [76][ 700/ 1207] Overall Loss 0.330237 Objective Loss 0.330237 LR 0.001000 Time 0.020568 -2023-02-13 17:52:08,855 - Epoch: [76][ 710/ 1207] Overall Loss 0.329941 Objective Loss 0.329941 LR 0.001000 Time 0.020556 -2023-02-13 17:52:09,054 - Epoch: [76][ 720/ 1207] Overall Loss 0.329853 Objective Loss 0.329853 LR 0.001000 Time 0.020546 -2023-02-13 17:52:09,252 - Epoch: [76][ 730/ 1207] Overall Loss 0.329516 Objective Loss 0.329516 LR 0.001000 Time 0.020535 -2023-02-13 17:52:09,449 - Epoch: [76][ 740/ 1207] Overall Loss 0.329321 Objective Loss 0.329321 LR 0.001000 Time 0.020524 -2023-02-13 17:52:09,647 - Epoch: [76][ 750/ 1207] Overall Loss 0.329561 Objective Loss 0.329561 LR 0.001000 Time 0.020513 -2023-02-13 17:52:09,845 - Epoch: [76][ 760/ 1207] Overall Loss 0.329728 Objective Loss 0.329728 LR 0.001000 Time 0.020503 -2023-02-13 17:52:10,043 - Epoch: [76][ 770/ 1207] Overall Loss 0.329803 Objective Loss 0.329803 LR 0.001000 Time 0.020494 -2023-02-13 17:52:10,243 - Epoch: [76][ 780/ 1207] Overall Loss 0.330139 Objective Loss 0.330139 LR 0.001000 Time 0.020486 -2023-02-13 17:52:10,439 - Epoch: [76][ 790/ 1207] Overall Loss 0.330004 Objective Loss 0.330004 LR 0.001000 Time 0.020476 -2023-02-13 17:52:10,637 - Epoch: [76][ 800/ 1207] Overall Loss 0.330201 Objective Loss 0.330201 LR 0.001000 Time 0.020467 -2023-02-13 17:52:10,835 - Epoch: [76][ 810/ 1207] Overall Loss 0.330660 Objective Loss 0.330660 LR 0.001000 Time 0.020458 -2023-02-13 17:52:11,034 - Epoch: [76][ 820/ 1207] Overall Loss 0.330886 Objective Loss 0.330886 LR 0.001000 Time 0.020451 -2023-02-13 17:52:11,233 - Epoch: [76][ 830/ 1207] Overall Loss 0.330750 Objective Loss 0.330750 LR 0.001000 Time 0.020443 -2023-02-13 17:52:11,431 - Epoch: [76][ 840/ 1207] Overall Loss 0.331487 Objective Loss 0.331487 LR 0.001000 Time 0.020435 -2023-02-13 17:52:11,629 - Epoch: [76][ 850/ 1207] Overall Loss 0.331778 Objective Loss 0.331778 LR 0.001000 Time 0.020427 -2023-02-13 17:52:11,827 - Epoch: [76][ 860/ 1207] Overall Loss 0.332390 Objective Loss 0.332390 LR 0.001000 Time 0.020419 -2023-02-13 17:52:12,025 - Epoch: [76][ 870/ 1207] Overall Loss 0.332406 Objective Loss 0.332406 LR 0.001000 Time 0.020412 -2023-02-13 17:52:12,224 - Epoch: [76][ 880/ 1207] Overall Loss 0.332577 Objective Loss 0.332577 LR 0.001000 Time 0.020406 -2023-02-13 17:52:12,421 - Epoch: [76][ 890/ 1207] Overall Loss 0.332115 Objective Loss 0.332115 LR 0.001000 Time 0.020398 -2023-02-13 17:52:12,619 - Epoch: [76][ 900/ 1207] Overall Loss 0.332210 Objective Loss 0.332210 LR 0.001000 Time 0.020391 -2023-02-13 17:52:12,817 - Epoch: [76][ 910/ 1207] Overall Loss 0.332565 Objective Loss 0.332565 LR 0.001000 Time 0.020383 -2023-02-13 17:52:13,015 - Epoch: [76][ 920/ 1207] Overall Loss 0.332402 Objective Loss 0.332402 LR 0.001000 Time 0.020377 -2023-02-13 17:52:13,213 - Epoch: [76][ 930/ 1207] Overall Loss 0.332276 Objective Loss 0.332276 LR 0.001000 Time 0.020370 -2023-02-13 17:52:13,412 - Epoch: [76][ 940/ 1207] Overall Loss 0.332337 Objective Loss 0.332337 LR 0.001000 Time 0.020364 -2023-02-13 17:52:13,609 - Epoch: [76][ 950/ 1207] Overall Loss 0.332359 Objective Loss 0.332359 LR 0.001000 Time 0.020357 -2023-02-13 17:52:13,807 - Epoch: [76][ 960/ 1207] Overall Loss 0.332285 Objective Loss 0.332285 LR 0.001000 Time 0.020351 -2023-02-13 17:52:14,006 - Epoch: [76][ 970/ 1207] Overall Loss 0.332088 Objective Loss 0.332088 LR 0.001000 Time 0.020346 -2023-02-13 17:52:14,205 - Epoch: [76][ 980/ 1207] Overall Loss 0.332448 Objective Loss 0.332448 LR 0.001000 Time 0.020341 -2023-02-13 17:52:14,401 - Epoch: [76][ 990/ 1207] Overall Loss 0.332672 Objective Loss 0.332672 LR 0.001000 Time 0.020333 -2023-02-13 17:52:14,599 - Epoch: [76][ 1000/ 1207] Overall Loss 0.332698 Objective Loss 0.332698 LR 0.001000 Time 0.020328 -2023-02-13 17:52:14,797 - Epoch: [76][ 1010/ 1207] Overall Loss 0.332796 Objective Loss 0.332796 LR 0.001000 Time 0.020322 -2023-02-13 17:52:14,995 - Epoch: [76][ 1020/ 1207] Overall Loss 0.332900 Objective Loss 0.332900 LR 0.001000 Time 0.020317 -2023-02-13 17:52:15,193 - Epoch: [76][ 1030/ 1207] Overall Loss 0.332616 Objective Loss 0.332616 LR 0.001000 Time 0.020311 -2023-02-13 17:52:15,391 - Epoch: [76][ 1040/ 1207] Overall Loss 0.332633 Objective Loss 0.332633 LR 0.001000 Time 0.020306 -2023-02-13 17:52:15,589 - Epoch: [76][ 1050/ 1207] Overall Loss 0.332734 Objective Loss 0.332734 LR 0.001000 Time 0.020301 -2023-02-13 17:52:15,788 - Epoch: [76][ 1060/ 1207] Overall Loss 0.332646 Objective Loss 0.332646 LR 0.001000 Time 0.020296 -2023-02-13 17:52:15,986 - Epoch: [76][ 1070/ 1207] Overall Loss 0.332794 Objective Loss 0.332794 LR 0.001000 Time 0.020291 -2023-02-13 17:52:16,184 - Epoch: [76][ 1080/ 1207] Overall Loss 0.332890 Objective Loss 0.332890 LR 0.001000 Time 0.020287 -2023-02-13 17:52:16,392 - Epoch: [76][ 1090/ 1207] Overall Loss 0.333222 Objective Loss 0.333222 LR 0.001000 Time 0.020291 -2023-02-13 17:52:16,597 - Epoch: [76][ 1100/ 1207] Overall Loss 0.333023 Objective Loss 0.333023 LR 0.001000 Time 0.020292 -2023-02-13 17:52:16,805 - Epoch: [76][ 1110/ 1207] Overall Loss 0.332677 Objective Loss 0.332677 LR 0.001000 Time 0.020296 -2023-02-13 17:52:17,008 - Epoch: [76][ 1120/ 1207] Overall Loss 0.332479 Objective Loss 0.332479 LR 0.001000 Time 0.020296 -2023-02-13 17:52:17,215 - Epoch: [76][ 1130/ 1207] Overall Loss 0.332680 Objective Loss 0.332680 LR 0.001000 Time 0.020299 -2023-02-13 17:52:17,417 - Epoch: [76][ 1140/ 1207] Overall Loss 0.332707 Objective Loss 0.332707 LR 0.001000 Time 0.020298 -2023-02-13 17:52:17,624 - Epoch: [76][ 1150/ 1207] Overall Loss 0.332964 Objective Loss 0.332964 LR 0.001000 Time 0.020301 -2023-02-13 17:52:17,826 - Epoch: [76][ 1160/ 1207] Overall Loss 0.333033 Objective Loss 0.333033 LR 0.001000 Time 0.020300 -2023-02-13 17:52:18,034 - Epoch: [76][ 1170/ 1207] Overall Loss 0.333155 Objective Loss 0.333155 LR 0.001000 Time 0.020304 -2023-02-13 17:52:18,236 - Epoch: [76][ 1180/ 1207] Overall Loss 0.333174 Objective Loss 0.333174 LR 0.001000 Time 0.020303 -2023-02-13 17:52:18,442 - Epoch: [76][ 1190/ 1207] Overall Loss 0.333430 Objective Loss 0.333430 LR 0.001000 Time 0.020305 -2023-02-13 17:52:18,697 - Epoch: [76][ 1200/ 1207] Overall Loss 0.333464 Objective Loss 0.333464 LR 0.001000 Time 0.020348 -2023-02-13 17:52:18,813 - Epoch: [76][ 1207/ 1207] Overall Loss 0.333602 Objective Loss 0.333602 Top1 81.707317 Top5 97.560976 LR 0.001000 Time 0.020326 -2023-02-13 17:52:18,885 - --- validate (epoch=76)----------- -2023-02-13 17:52:18,886 - 34311 samples (256 per mini-batch) -2023-02-13 17:52:19,302 - Epoch: [76][ 10/ 135] Loss 0.351594 Top1 83.359375 Top5 97.304688 -2023-02-13 17:52:19,436 - Epoch: [76][ 20/ 135] Loss 0.361170 Top1 81.914062 Top5 97.050781 -2023-02-13 17:52:19,561 - Epoch: [76][ 30/ 135] Loss 0.360420 Top1 81.927083 Top5 97.239583 -2023-02-13 17:52:19,683 - Epoch: [76][ 40/ 135] Loss 0.352678 Top1 82.451172 Top5 97.324219 -2023-02-13 17:52:19,809 - Epoch: [76][ 50/ 135] Loss 0.350311 Top1 82.406250 Top5 97.289062 -2023-02-13 17:52:19,937 - Epoch: [76][ 60/ 135] Loss 0.352413 Top1 82.402344 Top5 97.291667 -2023-02-13 17:52:20,067 - Epoch: [76][ 70/ 135] Loss 0.353659 Top1 82.388393 Top5 97.304688 -2023-02-13 17:52:20,197 - Epoch: [76][ 80/ 135] Loss 0.357392 Top1 82.211914 Top5 97.294922 -2023-02-13 17:52:20,328 - Epoch: [76][ 90/ 135] Loss 0.359607 Top1 82.174479 Top5 97.282986 -2023-02-13 17:52:20,461 - Epoch: [76][ 100/ 135] Loss 0.362280 Top1 82.144531 Top5 97.292969 -2023-02-13 17:52:20,590 - Epoch: [76][ 110/ 135] Loss 0.363930 Top1 82.176847 Top5 97.294034 -2023-02-13 17:52:20,723 - Epoch: [76][ 120/ 135] Loss 0.362651 Top1 82.226562 Top5 97.317708 -2023-02-13 17:52:20,855 - Epoch: [76][ 130/ 135] Loss 0.364074 Top1 82.190505 Top5 97.319712 -2023-02-13 17:52:20,903 - Epoch: [76][ 135/ 135] Loss 0.360475 Top1 82.212701 Top5 97.336131 -2023-02-13 17:52:20,974 - ==> Top1: 82.213 Top5: 97.336 Loss: 0.360 - -2023-02-13 17:52:20,975 - ==> Confusion: -[[ 843 3 6 0 7 5 0 0 9 62 1 0 0 5 3 4 3 4 3 2 7] - [ 2 915 1 4 14 33 7 22 4 3 1 1 2 1 1 4 6 0 6 2 4] - [ 8 0 964 7 2 2 17 12 0 2 3 1 2 4 4 4 4 3 4 6 9] - [ 5 0 27 898 4 4 1 2 1 3 19 1 7 0 11 2 2 6 15 1 7] - [ 29 8 4 2 964 8 2 1 0 7 0 3 0 7 5 5 9 4 1 3 4] - [ 7 21 5 5 2 933 1 27 2 6 2 11 7 20 0 2 4 2 2 8 3] - [ 3 6 22 2 0 4 1027 5 0 0 4 1 1 3 0 3 2 1 1 10 4] - [ 3 4 17 4 2 23 3 908 3 2 2 5 1 0 0 2 0 3 26 11 5] - [ 19 1 0 1 0 1 0 2 883 44 7 2 1 19 16 1 3 2 6 0 1] - [ 85 0 6 1 6 1 0 0 31 850 2 0 1 15 8 1 0 1 0 0 4] - [ 3 5 8 8 1 1 2 7 14 2 969 1 1 8 6 1 0 0 8 0 6] - [ 3 1 6 1 0 14 0 7 4 1 1 878 26 8 1 12 3 14 2 21 2] - [ 0 1 2 9 0 3 1 3 2 2 2 38 840 1 4 11 3 24 2 3 8] - [ 4 2 2 2 6 12 0 1 12 25 14 5 3 911 4 3 7 1 1 3 6] - [ 13 5 6 28 4 1 0 3 34 8 4 1 2 1 949 3 3 7 8 0 12] - [ 5 2 7 3 6 0 6 1 0 2 0 6 8 0 0 970 3 13 0 9 5] - [ 4 6 4 2 6 1 1 2 3 1 0 0 2 2 3 18 980 1 1 5 19] - [ 7 1 1 7 1 1 2 0 0 1 1 11 22 2 0 15 0 972 0 3 4] - [ 5 4 9 13 2 2 0 25 9 1 7 2 5 0 15 1 1 3 980 1 1] - [ 1 4 3 0 0 4 7 15 1 2 2 14 4 6 0 5 6 3 1 1065 5] - [ 218 240 339 179 103 209 113 179 113 132 239 101 295 351 156 129 210 130 172 317 9509]] - -2023-02-13 17:52:20,977 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:52:20,977 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:52:20,983 - - -2023-02-13 17:52:20,983 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:52:21,866 - Epoch: [77][ 10/ 1207] Overall Loss 0.343579 Objective Loss 0.343579 LR 0.001000 Time 0.088254 -2023-02-13 17:52:22,059 - Epoch: [77][ 20/ 1207] Overall Loss 0.353564 Objective Loss 0.353564 LR 0.001000 Time 0.053730 -2023-02-13 17:52:22,250 - Epoch: [77][ 30/ 1207] Overall Loss 0.348085 Objective Loss 0.348085 LR 0.001000 Time 0.042173 -2023-02-13 17:52:22,440 - Epoch: [77][ 40/ 1207] Overall Loss 0.342353 Objective Loss 0.342353 LR 0.001000 Time 0.036369 -2023-02-13 17:52:22,630 - Epoch: [77][ 50/ 1207] Overall Loss 0.344987 Objective Loss 0.344987 LR 0.001000 Time 0.032890 -2023-02-13 17:52:22,821 - Epoch: [77][ 60/ 1207] Overall Loss 0.342354 Objective Loss 0.342354 LR 0.001000 Time 0.030585 -2023-02-13 17:52:23,012 - Epoch: [77][ 70/ 1207] Overall Loss 0.346878 Objective Loss 0.346878 LR 0.001000 Time 0.028939 -2023-02-13 17:52:23,203 - Epoch: [77][ 80/ 1207] Overall Loss 0.345196 Objective Loss 0.345196 LR 0.001000 Time 0.027706 -2023-02-13 17:52:23,392 - Epoch: [77][ 90/ 1207] Overall Loss 0.346293 Objective Loss 0.346293 LR 0.001000 Time 0.026721 -2023-02-13 17:52:23,581 - Epoch: [77][ 100/ 1207] Overall Loss 0.346515 Objective Loss 0.346515 LR 0.001000 Time 0.025937 -2023-02-13 17:52:23,770 - Epoch: [77][ 110/ 1207] Overall Loss 0.344001 Objective Loss 0.344001 LR 0.001000 Time 0.025294 -2023-02-13 17:52:23,959 - Epoch: [77][ 120/ 1207] Overall Loss 0.342395 Objective Loss 0.342395 LR 0.001000 Time 0.024762 -2023-02-13 17:52:24,149 - Epoch: [77][ 130/ 1207] Overall Loss 0.339512 Objective Loss 0.339512 LR 0.001000 Time 0.024316 -2023-02-13 17:52:24,339 - Epoch: [77][ 140/ 1207] Overall Loss 0.338983 Objective Loss 0.338983 LR 0.001000 Time 0.023928 -2023-02-13 17:52:24,528 - Epoch: [77][ 150/ 1207] Overall Loss 0.338791 Objective Loss 0.338791 LR 0.001000 Time 0.023594 -2023-02-13 17:52:24,717 - Epoch: [77][ 160/ 1207] Overall Loss 0.337924 Objective Loss 0.337924 LR 0.001000 Time 0.023299 -2023-02-13 17:52:24,906 - Epoch: [77][ 170/ 1207] Overall Loss 0.337906 Objective Loss 0.337906 LR 0.001000 Time 0.023039 -2023-02-13 17:52:25,098 - Epoch: [77][ 180/ 1207] Overall Loss 0.336323 Objective Loss 0.336323 LR 0.001000 Time 0.022819 -2023-02-13 17:52:25,287 - Epoch: [77][ 190/ 1207] Overall Loss 0.334665 Objective Loss 0.334665 LR 0.001000 Time 0.022612 -2023-02-13 17:52:25,477 - Epoch: [77][ 200/ 1207] Overall Loss 0.334412 Objective Loss 0.334412 LR 0.001000 Time 0.022429 -2023-02-13 17:52:25,666 - Epoch: [77][ 210/ 1207] Overall Loss 0.332144 Objective Loss 0.332144 LR 0.001000 Time 0.022260 -2023-02-13 17:52:25,856 - Epoch: [77][ 220/ 1207] Overall Loss 0.332132 Objective Loss 0.332132 LR 0.001000 Time 0.022110 -2023-02-13 17:52:26,047 - Epoch: [77][ 230/ 1207] Overall Loss 0.330283 Objective Loss 0.330283 LR 0.001000 Time 0.021977 -2023-02-13 17:52:26,236 - Epoch: [77][ 240/ 1207] Overall Loss 0.329503 Objective Loss 0.329503 LR 0.001000 Time 0.021850 -2023-02-13 17:52:26,426 - Epoch: [77][ 250/ 1207] Overall Loss 0.329721 Objective Loss 0.329721 LR 0.001000 Time 0.021733 -2023-02-13 17:52:26,616 - Epoch: [77][ 260/ 1207] Overall Loss 0.329527 Objective Loss 0.329527 LR 0.001000 Time 0.021625 -2023-02-13 17:52:26,805 - Epoch: [77][ 270/ 1207] Overall Loss 0.328809 Objective Loss 0.328809 LR 0.001000 Time 0.021524 -2023-02-13 17:52:26,995 - Epoch: [77][ 280/ 1207] Overall Loss 0.329578 Objective Loss 0.329578 LR 0.001000 Time 0.021433 -2023-02-13 17:52:27,187 - Epoch: [77][ 290/ 1207] Overall Loss 0.327815 Objective Loss 0.327815 LR 0.001000 Time 0.021355 -2023-02-13 17:52:27,378 - Epoch: [77][ 300/ 1207] Overall Loss 0.327998 Objective Loss 0.327998 LR 0.001000 Time 0.021278 -2023-02-13 17:52:27,569 - Epoch: [77][ 310/ 1207] Overall Loss 0.327417 Objective Loss 0.327417 LR 0.001000 Time 0.021206 -2023-02-13 17:52:27,760 - Epoch: [77][ 320/ 1207] Overall Loss 0.327684 Objective Loss 0.327684 LR 0.001000 Time 0.021140 -2023-02-13 17:52:27,951 - Epoch: [77][ 330/ 1207] Overall Loss 0.327405 Objective Loss 0.327405 LR 0.001000 Time 0.021077 -2023-02-13 17:52:28,143 - Epoch: [77][ 340/ 1207] Overall Loss 0.328333 Objective Loss 0.328333 LR 0.001000 Time 0.021019 -2023-02-13 17:52:28,334 - Epoch: [77][ 350/ 1207] Overall Loss 0.328642 Objective Loss 0.328642 LR 0.001000 Time 0.020963 -2023-02-13 17:52:28,524 - Epoch: [77][ 360/ 1207] Overall Loss 0.328332 Objective Loss 0.328332 LR 0.001000 Time 0.020909 -2023-02-13 17:52:28,715 - Epoch: [77][ 370/ 1207] Overall Loss 0.328959 Objective Loss 0.328959 LR 0.001000 Time 0.020859 -2023-02-13 17:52:28,906 - Epoch: [77][ 380/ 1207] Overall Loss 0.329290 Objective Loss 0.329290 LR 0.001000 Time 0.020811 -2023-02-13 17:52:29,097 - Epoch: [77][ 390/ 1207] Overall Loss 0.329043 Objective Loss 0.329043 LR 0.001000 Time 0.020768 -2023-02-13 17:52:29,289 - Epoch: [77][ 400/ 1207] Overall Loss 0.329008 Objective Loss 0.329008 LR 0.001000 Time 0.020726 -2023-02-13 17:52:29,480 - Epoch: [77][ 410/ 1207] Overall Loss 0.330461 Objective Loss 0.330461 LR 0.001000 Time 0.020685 -2023-02-13 17:52:29,671 - Epoch: [77][ 420/ 1207] Overall Loss 0.330196 Objective Loss 0.330196 LR 0.001000 Time 0.020647 -2023-02-13 17:52:29,862 - Epoch: [77][ 430/ 1207] Overall Loss 0.330784 Objective Loss 0.330784 LR 0.001000 Time 0.020610 -2023-02-13 17:52:30,054 - Epoch: [77][ 440/ 1207] Overall Loss 0.331333 Objective Loss 0.331333 LR 0.001000 Time 0.020577 -2023-02-13 17:52:30,246 - Epoch: [77][ 450/ 1207] Overall Loss 0.331385 Objective Loss 0.331385 LR 0.001000 Time 0.020545 -2023-02-13 17:52:30,436 - Epoch: [77][ 460/ 1207] Overall Loss 0.331660 Objective Loss 0.331660 LR 0.001000 Time 0.020512 -2023-02-13 17:52:30,628 - Epoch: [77][ 470/ 1207] Overall Loss 0.331617 Objective Loss 0.331617 LR 0.001000 Time 0.020482 -2023-02-13 17:52:30,820 - Epoch: [77][ 480/ 1207] Overall Loss 0.331404 Objective Loss 0.331404 LR 0.001000 Time 0.020455 -2023-02-13 17:52:31,011 - Epoch: [77][ 490/ 1207] Overall Loss 0.330730 Objective Loss 0.330730 LR 0.001000 Time 0.020426 -2023-02-13 17:52:31,203 - Epoch: [77][ 500/ 1207] Overall Loss 0.330805 Objective Loss 0.330805 LR 0.001000 Time 0.020402 -2023-02-13 17:52:31,395 - Epoch: [77][ 510/ 1207] Overall Loss 0.330491 Objective Loss 0.330491 LR 0.001000 Time 0.020376 -2023-02-13 17:52:31,586 - Epoch: [77][ 520/ 1207] Overall Loss 0.330165 Objective Loss 0.330165 LR 0.001000 Time 0.020351 -2023-02-13 17:52:31,775 - Epoch: [77][ 530/ 1207] Overall Loss 0.330074 Objective Loss 0.330074 LR 0.001000 Time 0.020325 -2023-02-13 17:52:31,966 - Epoch: [77][ 540/ 1207] Overall Loss 0.329733 Objective Loss 0.329733 LR 0.001000 Time 0.020301 -2023-02-13 17:52:32,156 - Epoch: [77][ 550/ 1207] Overall Loss 0.329853 Objective Loss 0.329853 LR 0.001000 Time 0.020277 -2023-02-13 17:52:32,347 - Epoch: [77][ 560/ 1207] Overall Loss 0.329963 Objective Loss 0.329963 LR 0.001000 Time 0.020255 -2023-02-13 17:52:32,537 - Epoch: [77][ 570/ 1207] Overall Loss 0.330947 Objective Loss 0.330947 LR 0.001000 Time 0.020232 -2023-02-13 17:52:32,728 - Epoch: [77][ 580/ 1207] Overall Loss 0.330783 Objective Loss 0.330783 LR 0.001000 Time 0.020211 -2023-02-13 17:52:32,919 - Epoch: [77][ 590/ 1207] Overall Loss 0.331208 Objective Loss 0.331208 LR 0.001000 Time 0.020192 -2023-02-13 17:52:33,116 - Epoch: [77][ 600/ 1207] Overall Loss 0.331088 Objective Loss 0.331088 LR 0.001000 Time 0.020184 -2023-02-13 17:52:33,313 - Epoch: [77][ 610/ 1207] Overall Loss 0.330824 Objective Loss 0.330824 LR 0.001000 Time 0.020175 -2023-02-13 17:52:33,506 - Epoch: [77][ 620/ 1207] Overall Loss 0.331376 Objective Loss 0.331376 LR 0.001000 Time 0.020160 -2023-02-13 17:52:33,696 - Epoch: [77][ 630/ 1207] Overall Loss 0.331465 Objective Loss 0.331465 LR 0.001000 Time 0.020142 -2023-02-13 17:52:33,887 - Epoch: [77][ 640/ 1207] Overall Loss 0.331976 Objective Loss 0.331976 LR 0.001000 Time 0.020124 -2023-02-13 17:52:34,078 - Epoch: [77][ 650/ 1207] Overall Loss 0.331958 Objective Loss 0.331958 LR 0.001000 Time 0.020108 -2023-02-13 17:52:34,269 - Epoch: [77][ 660/ 1207] Overall Loss 0.331730 Objective Loss 0.331730 LR 0.001000 Time 0.020092 -2023-02-13 17:52:34,460 - Epoch: [77][ 670/ 1207] Overall Loss 0.331627 Objective Loss 0.331627 LR 0.001000 Time 0.020076 -2023-02-13 17:52:34,650 - Epoch: [77][ 680/ 1207] Overall Loss 0.331310 Objective Loss 0.331310 LR 0.001000 Time 0.020061 -2023-02-13 17:52:34,842 - Epoch: [77][ 690/ 1207] Overall Loss 0.331122 Objective Loss 0.331122 LR 0.001000 Time 0.020047 -2023-02-13 17:52:35,034 - Epoch: [77][ 700/ 1207] Overall Loss 0.331115 Objective Loss 0.331115 LR 0.001000 Time 0.020035 -2023-02-13 17:52:35,226 - Epoch: [77][ 710/ 1207] Overall Loss 0.331281 Objective Loss 0.331281 LR 0.001000 Time 0.020022 -2023-02-13 17:52:35,418 - Epoch: [77][ 720/ 1207] Overall Loss 0.331731 Objective Loss 0.331731 LR 0.001000 Time 0.020010 -2023-02-13 17:52:35,609 - Epoch: [77][ 730/ 1207] Overall Loss 0.331644 Objective Loss 0.331644 LR 0.001000 Time 0.019997 -2023-02-13 17:52:35,800 - Epoch: [77][ 740/ 1207] Overall Loss 0.331744 Objective Loss 0.331744 LR 0.001000 Time 0.019984 -2023-02-13 17:52:35,991 - Epoch: [77][ 750/ 1207] Overall Loss 0.331520 Objective Loss 0.331520 LR 0.001000 Time 0.019972 -2023-02-13 17:52:36,183 - Epoch: [77][ 760/ 1207] Overall Loss 0.331428 Objective Loss 0.331428 LR 0.001000 Time 0.019961 -2023-02-13 17:52:36,373 - Epoch: [77][ 770/ 1207] Overall Loss 0.331728 Objective Loss 0.331728 LR 0.001000 Time 0.019948 -2023-02-13 17:52:36,564 - Epoch: [77][ 780/ 1207] Overall Loss 0.331404 Objective Loss 0.331404 LR 0.001000 Time 0.019938 -2023-02-13 17:52:36,756 - Epoch: [77][ 790/ 1207] Overall Loss 0.331287 Objective Loss 0.331287 LR 0.001000 Time 0.019927 -2023-02-13 17:52:36,948 - Epoch: [77][ 800/ 1207] Overall Loss 0.331290 Objective Loss 0.331290 LR 0.001000 Time 0.019918 -2023-02-13 17:52:37,139 - Epoch: [77][ 810/ 1207] Overall Loss 0.331417 Objective Loss 0.331417 LR 0.001000 Time 0.019907 -2023-02-13 17:52:37,331 - Epoch: [77][ 820/ 1207] Overall Loss 0.331264 Objective Loss 0.331264 LR 0.001000 Time 0.019898 -2023-02-13 17:52:37,523 - Epoch: [77][ 830/ 1207] Overall Loss 0.331050 Objective Loss 0.331050 LR 0.001000 Time 0.019889 -2023-02-13 17:52:37,713 - Epoch: [77][ 840/ 1207] Overall Loss 0.330764 Objective Loss 0.330764 LR 0.001000 Time 0.019879 -2023-02-13 17:52:37,905 - Epoch: [77][ 850/ 1207] Overall Loss 0.330955 Objective Loss 0.330955 LR 0.001000 Time 0.019870 -2023-02-13 17:52:38,097 - Epoch: [77][ 860/ 1207] Overall Loss 0.330992 Objective Loss 0.330992 LR 0.001000 Time 0.019862 -2023-02-13 17:52:38,289 - Epoch: [77][ 870/ 1207] Overall Loss 0.331384 Objective Loss 0.331384 LR 0.001000 Time 0.019853 -2023-02-13 17:52:38,481 - Epoch: [77][ 880/ 1207] Overall Loss 0.331709 Objective Loss 0.331709 LR 0.001000 Time 0.019845 -2023-02-13 17:52:38,673 - Epoch: [77][ 890/ 1207] Overall Loss 0.331455 Objective Loss 0.331455 LR 0.001000 Time 0.019838 -2023-02-13 17:52:38,863 - Epoch: [77][ 900/ 1207] Overall Loss 0.331716 Objective Loss 0.331716 LR 0.001000 Time 0.019829 -2023-02-13 17:52:39,054 - Epoch: [77][ 910/ 1207] Overall Loss 0.331382 Objective Loss 0.331382 LR 0.001000 Time 0.019820 -2023-02-13 17:52:39,246 - Epoch: [77][ 920/ 1207] Overall Loss 0.331296 Objective Loss 0.331296 LR 0.001000 Time 0.019812 -2023-02-13 17:52:39,436 - Epoch: [77][ 930/ 1207] Overall Loss 0.330982 Objective Loss 0.330982 LR 0.001000 Time 0.019804 -2023-02-13 17:52:39,627 - Epoch: [77][ 940/ 1207] Overall Loss 0.330987 Objective Loss 0.330987 LR 0.001000 Time 0.019796 -2023-02-13 17:52:39,818 - Epoch: [77][ 950/ 1207] Overall Loss 0.330971 Objective Loss 0.330971 LR 0.001000 Time 0.019788 -2023-02-13 17:52:40,010 - Epoch: [77][ 960/ 1207] Overall Loss 0.330879 Objective Loss 0.330879 LR 0.001000 Time 0.019781 -2023-02-13 17:52:40,201 - Epoch: [77][ 970/ 1207] Overall Loss 0.331127 Objective Loss 0.331127 LR 0.001000 Time 0.019775 -2023-02-13 17:52:40,394 - Epoch: [77][ 980/ 1207] Overall Loss 0.331140 Objective Loss 0.331140 LR 0.001000 Time 0.019769 -2023-02-13 17:52:40,585 - Epoch: [77][ 990/ 1207] Overall Loss 0.331178 Objective Loss 0.331178 LR 0.001000 Time 0.019762 -2023-02-13 17:52:40,776 - Epoch: [77][ 1000/ 1207] Overall Loss 0.331292 Objective Loss 0.331292 LR 0.001000 Time 0.019755 -2023-02-13 17:52:40,969 - Epoch: [77][ 1010/ 1207] Overall Loss 0.330964 Objective Loss 0.330964 LR 0.001000 Time 0.019750 -2023-02-13 17:52:41,161 - Epoch: [77][ 1020/ 1207] Overall Loss 0.331501 Objective Loss 0.331501 LR 0.001000 Time 0.019745 -2023-02-13 17:52:41,353 - Epoch: [77][ 1030/ 1207] Overall Loss 0.331214 Objective Loss 0.331214 LR 0.001000 Time 0.019739 -2023-02-13 17:52:41,544 - Epoch: [77][ 1040/ 1207] Overall Loss 0.331313 Objective Loss 0.331313 LR 0.001000 Time 0.019732 -2023-02-13 17:52:41,736 - Epoch: [77][ 1050/ 1207] Overall Loss 0.331319 Objective Loss 0.331319 LR 0.001000 Time 0.019726 -2023-02-13 17:52:41,927 - Epoch: [77][ 1060/ 1207] Overall Loss 0.331437 Objective Loss 0.331437 LR 0.001000 Time 0.019721 -2023-02-13 17:52:42,119 - Epoch: [77][ 1070/ 1207] Overall Loss 0.331576 Objective Loss 0.331576 LR 0.001000 Time 0.019715 -2023-02-13 17:52:42,310 - Epoch: [77][ 1080/ 1207] Overall Loss 0.331493 Objective Loss 0.331493 LR 0.001000 Time 0.019709 -2023-02-13 17:52:42,502 - Epoch: [77][ 1090/ 1207] Overall Loss 0.331169 Objective Loss 0.331169 LR 0.001000 Time 0.019704 -2023-02-13 17:52:42,694 - Epoch: [77][ 1100/ 1207] Overall Loss 0.331133 Objective Loss 0.331133 LR 0.001000 Time 0.019699 -2023-02-13 17:52:42,885 - Epoch: [77][ 1110/ 1207] Overall Loss 0.331009 Objective Loss 0.331009 LR 0.001000 Time 0.019693 -2023-02-13 17:52:43,076 - Epoch: [77][ 1120/ 1207] Overall Loss 0.331265 Objective Loss 0.331265 LR 0.001000 Time 0.019688 -2023-02-13 17:52:43,268 - Epoch: [77][ 1130/ 1207] Overall Loss 0.331577 Objective Loss 0.331577 LR 0.001000 Time 0.019683 -2023-02-13 17:52:43,460 - Epoch: [77][ 1140/ 1207] Overall Loss 0.331638 Objective Loss 0.331638 LR 0.001000 Time 0.019678 -2023-02-13 17:52:43,650 - Epoch: [77][ 1150/ 1207] Overall Loss 0.331747 Objective Loss 0.331747 LR 0.001000 Time 0.019672 -2023-02-13 17:52:43,841 - Epoch: [77][ 1160/ 1207] Overall Loss 0.331900 Objective Loss 0.331900 LR 0.001000 Time 0.019667 -2023-02-13 17:52:44,032 - Epoch: [77][ 1170/ 1207] Overall Loss 0.332029 Objective Loss 0.332029 LR 0.001000 Time 0.019661 -2023-02-13 17:52:44,224 - Epoch: [77][ 1180/ 1207] Overall Loss 0.332035 Objective Loss 0.332035 LR 0.001000 Time 0.019657 -2023-02-13 17:52:44,415 - Epoch: [77][ 1190/ 1207] Overall Loss 0.332036 Objective Loss 0.332036 LR 0.001000 Time 0.019652 -2023-02-13 17:52:44,656 - Epoch: [77][ 1200/ 1207] Overall Loss 0.332071 Objective Loss 0.332071 LR 0.001000 Time 0.019689 -2023-02-13 17:52:44,773 - Epoch: [77][ 1207/ 1207] Overall Loss 0.331848 Objective Loss 0.331848 Top1 87.195122 Top5 98.170732 LR 0.001000 Time 0.019672 -2023-02-13 17:52:44,844 - --- validate (epoch=77)----------- -2023-02-13 17:52:44,844 - 34311 samples (256 per mini-batch) -2023-02-13 17:52:45,352 - Epoch: [77][ 10/ 135] Loss 0.380807 Top1 82.617188 Top5 97.421875 -2023-02-13 17:52:45,475 - Epoch: [77][ 20/ 135] Loss 0.361541 Top1 82.558594 Top5 97.148438 -2023-02-13 17:52:45,601 - Epoch: [77][ 30/ 135] Loss 0.363936 Top1 82.408854 Top5 97.148438 -2023-02-13 17:52:45,732 - Epoch: [77][ 40/ 135] Loss 0.378304 Top1 82.021484 Top5 97.001953 -2023-02-13 17:52:45,883 - Epoch: [77][ 50/ 135] Loss 0.375414 Top1 82.289062 Top5 96.976562 -2023-02-13 17:52:46,012 - Epoch: [77][ 60/ 135] Loss 0.381512 Top1 82.109375 Top5 96.992188 -2023-02-13 17:52:46,141 - Epoch: [77][ 70/ 135] Loss 0.379120 Top1 82.170759 Top5 97.047991 -2023-02-13 17:52:46,271 - Epoch: [77][ 80/ 135] Loss 0.376503 Top1 82.143555 Top5 97.089844 -2023-02-13 17:52:46,401 - Epoch: [77][ 90/ 135] Loss 0.378202 Top1 82.083333 Top5 97.074653 -2023-02-13 17:52:46,530 - Epoch: [77][ 100/ 135] Loss 0.376674 Top1 82.164062 Top5 97.109375 -2023-02-13 17:52:46,660 - Epoch: [77][ 110/ 135] Loss 0.374643 Top1 82.159091 Top5 97.088068 -2023-02-13 17:52:46,791 - Epoch: [77][ 120/ 135] Loss 0.374575 Top1 82.138672 Top5 97.086589 -2023-02-13 17:52:46,922 - Epoch: [77][ 130/ 135] Loss 0.372742 Top1 82.076322 Top5 97.103365 -2023-02-13 17:52:46,968 - Epoch: [77][ 135/ 135] Loss 0.374067 Top1 81.982455 Top5 97.117542 -2023-02-13 17:52:47,040 - ==> Top1: 81.982 Top5: 97.118 Loss: 0.374 - -2023-02-13 17:52:47,040 - ==> Confusion: -[[ 841 9 3 2 13 7 0 1 3 60 0 2 2 5 5 4 1 3 1 1 4] - [ 4 893 1 6 10 50 2 24 6 1 5 2 2 0 2 0 2 1 7 8 7] - [ 8 6 923 21 0 2 31 12 0 1 4 1 4 6 1 7 5 6 8 5 7] - [ 3 2 17 888 3 4 3 1 2 2 20 0 6 2 17 6 7 9 13 0 11] - [ 14 7 0 3 967 16 1 3 3 2 0 5 3 10 8 4 9 2 0 2 7] - [ 4 14 1 5 2 974 3 10 1 3 5 8 3 17 0 3 3 1 1 10 2] - [ 4 3 17 1 0 4 1035 10 0 0 3 3 3 2 1 3 2 2 1 4 1] - [ 3 6 5 2 2 35 9 916 1 1 4 7 4 2 1 0 2 2 8 10 4] - [ 22 3 0 1 1 3 0 0 862 45 21 0 1 15 17 2 2 2 5 1 6] - [ 81 0 4 0 4 2 0 1 44 827 2 0 1 28 2 1 2 5 0 0 8] - [ 1 2 5 3 1 3 3 5 16 1 984 1 0 7 3 1 1 2 7 2 3] - [ 3 1 3 1 3 24 1 6 1 2 2 888 27 10 1 5 5 9 2 8 3] - [ 3 1 2 5 1 4 1 3 2 0 1 30 834 4 3 7 1 43 3 2 9] - [ 7 3 6 0 3 9 1 3 12 13 15 3 1 921 8 6 4 2 0 3 4] - [ 15 3 2 23 6 4 2 1 22 2 3 1 3 3 963 2 5 8 13 0 11] - [ 5 3 8 1 5 3 5 1 0 1 1 7 4 3 1 957 5 20 0 9 7] - [ 3 7 0 1 4 3 0 1 6 0 1 2 2 5 1 8 992 4 1 6 14] - [ 8 3 0 4 3 4 1 2 1 0 0 11 6 1 2 13 0 985 0 4 3] - [ 3 3 4 16 3 3 0 33 5 0 8 4 9 0 16 1 0 6 969 0 3] - [ 0 4 1 2 2 9 4 13 1 0 2 18 4 4 0 2 4 5 2 1065 6] - [ 180 232 222 174 112 330 107 198 88 85 266 138 317 396 163 98 286 125 158 314 9445]] - -2023-02-13 17:52:47,042 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:52:47,042 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:52:47,047 - - -2023-02-13 17:52:47,048 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:52:47,932 - Epoch: [78][ 10/ 1207] Overall Loss 0.334797 Objective Loss 0.334797 LR 0.001000 Time 0.088408 -2023-02-13 17:52:48,126 - Epoch: [78][ 20/ 1207] Overall Loss 0.337109 Objective Loss 0.337109 LR 0.001000 Time 0.053882 -2023-02-13 17:52:48,321 - Epoch: [78][ 30/ 1207] Overall Loss 0.342058 Objective Loss 0.342058 LR 0.001000 Time 0.042410 -2023-02-13 17:52:48,513 - Epoch: [78][ 40/ 1207] Overall Loss 0.342136 Objective Loss 0.342136 LR 0.001000 Time 0.036583 -2023-02-13 17:52:48,708 - Epoch: [78][ 50/ 1207] Overall Loss 0.341164 Objective Loss 0.341164 LR 0.001000 Time 0.033166 -2023-02-13 17:52:48,900 - Epoch: [78][ 60/ 1207] Overall Loss 0.334443 Objective Loss 0.334443 LR 0.001000 Time 0.030830 -2023-02-13 17:52:49,096 - Epoch: [78][ 70/ 1207] Overall Loss 0.337293 Objective Loss 0.337293 LR 0.001000 Time 0.029223 -2023-02-13 17:52:49,289 - Epoch: [78][ 80/ 1207] Overall Loss 0.338808 Objective Loss 0.338808 LR 0.001000 Time 0.027971 -2023-02-13 17:52:49,484 - Epoch: [78][ 90/ 1207] Overall Loss 0.334951 Objective Loss 0.334951 LR 0.001000 Time 0.027029 -2023-02-13 17:52:49,676 - Epoch: [78][ 100/ 1207] Overall Loss 0.333792 Objective Loss 0.333792 LR 0.001000 Time 0.026239 -2023-02-13 17:52:49,871 - Epoch: [78][ 110/ 1207] Overall Loss 0.327767 Objective Loss 0.327767 LR 0.001000 Time 0.025629 -2023-02-13 17:52:50,063 - Epoch: [78][ 120/ 1207] Overall Loss 0.325772 Objective Loss 0.325772 LR 0.001000 Time 0.025092 -2023-02-13 17:52:50,259 - Epoch: [78][ 130/ 1207] Overall Loss 0.325941 Objective Loss 0.325941 LR 0.001000 Time 0.024664 -2023-02-13 17:52:50,451 - Epoch: [78][ 140/ 1207] Overall Loss 0.326016 Objective Loss 0.326016 LR 0.001000 Time 0.024272 -2023-02-13 17:52:50,647 - Epoch: [78][ 150/ 1207] Overall Loss 0.324342 Objective Loss 0.324342 LR 0.001000 Time 0.023955 -2023-02-13 17:52:50,839 - Epoch: [78][ 160/ 1207] Overall Loss 0.324780 Objective Loss 0.324780 LR 0.001000 Time 0.023660 -2023-02-13 17:52:51,035 - Epoch: [78][ 170/ 1207] Overall Loss 0.323432 Objective Loss 0.323432 LR 0.001000 Time 0.023418 -2023-02-13 17:52:51,228 - Epoch: [78][ 180/ 1207] Overall Loss 0.323733 Objective Loss 0.323733 LR 0.001000 Time 0.023183 -2023-02-13 17:52:51,423 - Epoch: [78][ 190/ 1207] Overall Loss 0.323041 Objective Loss 0.323041 LR 0.001000 Time 0.022991 -2023-02-13 17:52:51,616 - Epoch: [78][ 200/ 1207] Overall Loss 0.323152 Objective Loss 0.323152 LR 0.001000 Time 0.022804 -2023-02-13 17:52:51,812 - Epoch: [78][ 210/ 1207] Overall Loss 0.323281 Objective Loss 0.323281 LR 0.001000 Time 0.022647 -2023-02-13 17:52:52,004 - Epoch: [78][ 220/ 1207] Overall Loss 0.322894 Objective Loss 0.322894 LR 0.001000 Time 0.022490 -2023-02-13 17:52:52,200 - Epoch: [78][ 230/ 1207] Overall Loss 0.322065 Objective Loss 0.322065 LR 0.001000 Time 0.022365 -2023-02-13 17:52:52,392 - Epoch: [78][ 240/ 1207] Overall Loss 0.322750 Objective Loss 0.322750 LR 0.001000 Time 0.022231 -2023-02-13 17:52:52,588 - Epoch: [78][ 250/ 1207] Overall Loss 0.322082 Objective Loss 0.322082 LR 0.001000 Time 0.022124 -2023-02-13 17:52:52,780 - Epoch: [78][ 260/ 1207] Overall Loss 0.322340 Objective Loss 0.322340 LR 0.001000 Time 0.022009 -2023-02-13 17:52:52,975 - Epoch: [78][ 270/ 1207] Overall Loss 0.322694 Objective Loss 0.322694 LR 0.001000 Time 0.021916 -2023-02-13 17:52:53,168 - Epoch: [78][ 280/ 1207] Overall Loss 0.322189 Objective Loss 0.322189 LR 0.001000 Time 0.021819 -2023-02-13 17:52:53,366 - Epoch: [78][ 290/ 1207] Overall Loss 0.321456 Objective Loss 0.321456 LR 0.001000 Time 0.021747 -2023-02-13 17:52:53,561 - Epoch: [78][ 300/ 1207] Overall Loss 0.321810 Objective Loss 0.321810 LR 0.001000 Time 0.021671 -2023-02-13 17:52:53,758 - Epoch: [78][ 310/ 1207] Overall Loss 0.322184 Objective Loss 0.322184 LR 0.001000 Time 0.021609 -2023-02-13 17:52:53,953 - Epoch: [78][ 320/ 1207] Overall Loss 0.321563 Objective Loss 0.321563 LR 0.001000 Time 0.021542 -2023-02-13 17:52:54,149 - Epoch: [78][ 330/ 1207] Overall Loss 0.321327 Objective Loss 0.321327 LR 0.001000 Time 0.021482 -2023-02-13 17:52:54,341 - Epoch: [78][ 340/ 1207] Overall Loss 0.322668 Objective Loss 0.322668 LR 0.001000 Time 0.021412 -2023-02-13 17:52:54,535 - Epoch: [78][ 350/ 1207] Overall Loss 0.323376 Objective Loss 0.323376 LR 0.001000 Time 0.021355 -2023-02-13 17:52:54,728 - Epoch: [78][ 360/ 1207] Overall Loss 0.324001 Objective Loss 0.324001 LR 0.001000 Time 0.021297 -2023-02-13 17:52:54,924 - Epoch: [78][ 370/ 1207] Overall Loss 0.324491 Objective Loss 0.324491 LR 0.001000 Time 0.021248 -2023-02-13 17:52:55,116 - Epoch: [78][ 380/ 1207] Overall Loss 0.324450 Objective Loss 0.324450 LR 0.001000 Time 0.021194 -2023-02-13 17:52:55,312 - Epoch: [78][ 390/ 1207] Overall Loss 0.324194 Objective Loss 0.324194 LR 0.001000 Time 0.021152 -2023-02-13 17:52:55,504 - Epoch: [78][ 400/ 1207] Overall Loss 0.324409 Objective Loss 0.324409 LR 0.001000 Time 0.021103 -2023-02-13 17:52:55,699 - Epoch: [78][ 410/ 1207] Overall Loss 0.325760 Objective Loss 0.325760 LR 0.001000 Time 0.021064 -2023-02-13 17:52:55,893 - Epoch: [78][ 420/ 1207] Overall Loss 0.326228 Objective Loss 0.326228 LR 0.001000 Time 0.021023 -2023-02-13 17:52:56,090 - Epoch: [78][ 430/ 1207] Overall Loss 0.325966 Objective Loss 0.325966 LR 0.001000 Time 0.020990 -2023-02-13 17:52:56,283 - Epoch: [78][ 440/ 1207] Overall Loss 0.325823 Objective Loss 0.325823 LR 0.001000 Time 0.020951 -2023-02-13 17:52:56,478 - Epoch: [78][ 450/ 1207] Overall Loss 0.326261 Objective Loss 0.326261 LR 0.001000 Time 0.020919 -2023-02-13 17:52:56,671 - Epoch: [78][ 460/ 1207] Overall Loss 0.326519 Objective Loss 0.326519 LR 0.001000 Time 0.020882 -2023-02-13 17:52:56,867 - Epoch: [78][ 470/ 1207] Overall Loss 0.326315 Objective Loss 0.326315 LR 0.001000 Time 0.020854 -2023-02-13 17:52:57,060 - Epoch: [78][ 480/ 1207] Overall Loss 0.326079 Objective Loss 0.326079 LR 0.001000 Time 0.020821 -2023-02-13 17:52:57,257 - Epoch: [78][ 490/ 1207] Overall Loss 0.326144 Objective Loss 0.326144 LR 0.001000 Time 0.020797 -2023-02-13 17:52:57,449 - Epoch: [78][ 500/ 1207] Overall Loss 0.326279 Objective Loss 0.326279 LR 0.001000 Time 0.020765 -2023-02-13 17:52:57,645 - Epoch: [78][ 510/ 1207] Overall Loss 0.326266 Objective Loss 0.326266 LR 0.001000 Time 0.020741 -2023-02-13 17:52:57,837 - Epoch: [78][ 520/ 1207] Overall Loss 0.326227 Objective Loss 0.326227 LR 0.001000 Time 0.020712 -2023-02-13 17:52:58,033 - Epoch: [78][ 530/ 1207] Overall Loss 0.326226 Objective Loss 0.326226 LR 0.001000 Time 0.020690 -2023-02-13 17:52:58,227 - Epoch: [78][ 540/ 1207] Overall Loss 0.325624 Objective Loss 0.325624 LR 0.001000 Time 0.020665 -2023-02-13 17:52:58,425 - Epoch: [78][ 550/ 1207] Overall Loss 0.325600 Objective Loss 0.325600 LR 0.001000 Time 0.020649 -2023-02-13 17:52:58,620 - Epoch: [78][ 560/ 1207] Overall Loss 0.325483 Objective Loss 0.325483 LR 0.001000 Time 0.020627 -2023-02-13 17:52:58,819 - Epoch: [78][ 570/ 1207] Overall Loss 0.325111 Objective Loss 0.325111 LR 0.001000 Time 0.020614 -2023-02-13 17:52:59,014 - Epoch: [78][ 580/ 1207] Overall Loss 0.324851 Objective Loss 0.324851 LR 0.001000 Time 0.020594 -2023-02-13 17:52:59,213 - Epoch: [78][ 590/ 1207] Overall Loss 0.324226 Objective Loss 0.324226 LR 0.001000 Time 0.020581 -2023-02-13 17:52:59,408 - Epoch: [78][ 600/ 1207] Overall Loss 0.324532 Objective Loss 0.324532 LR 0.001000 Time 0.020563 -2023-02-13 17:52:59,606 - Epoch: [78][ 610/ 1207] Overall Loss 0.324789 Objective Loss 0.324789 LR 0.001000 Time 0.020550 -2023-02-13 17:52:59,800 - Epoch: [78][ 620/ 1207] Overall Loss 0.324690 Objective Loss 0.324690 LR 0.001000 Time 0.020530 -2023-02-13 17:52:59,996 - Epoch: [78][ 630/ 1207] Overall Loss 0.325098 Objective Loss 0.325098 LR 0.001000 Time 0.020515 -2023-02-13 17:53:00,189 - Epoch: [78][ 640/ 1207] Overall Loss 0.324822 Objective Loss 0.324822 LR 0.001000 Time 0.020495 -2023-02-13 17:53:00,385 - Epoch: [78][ 650/ 1207] Overall Loss 0.325066 Objective Loss 0.325066 LR 0.001000 Time 0.020481 -2023-02-13 17:53:00,577 - Epoch: [78][ 660/ 1207] Overall Loss 0.325312 Objective Loss 0.325312 LR 0.001000 Time 0.020462 -2023-02-13 17:53:00,780 - Epoch: [78][ 670/ 1207] Overall Loss 0.325404 Objective Loss 0.325404 LR 0.001000 Time 0.020458 -2023-02-13 17:53:00,974 - Epoch: [78][ 680/ 1207] Overall Loss 0.325623 Objective Loss 0.325623 LR 0.001000 Time 0.020443 -2023-02-13 17:53:01,172 - Epoch: [78][ 690/ 1207] Overall Loss 0.325525 Objective Loss 0.325525 LR 0.001000 Time 0.020432 -2023-02-13 17:53:01,364 - Epoch: [78][ 700/ 1207] Overall Loss 0.325464 Objective Loss 0.325464 LR 0.001000 Time 0.020415 -2023-02-13 17:53:01,561 - Epoch: [78][ 710/ 1207] Overall Loss 0.325613 Objective Loss 0.325613 LR 0.001000 Time 0.020403 -2023-02-13 17:53:01,753 - Epoch: [78][ 720/ 1207] Overall Loss 0.325305 Objective Loss 0.325305 LR 0.001000 Time 0.020387 -2023-02-13 17:53:01,949 - Epoch: [78][ 730/ 1207] Overall Loss 0.325583 Objective Loss 0.325583 LR 0.001000 Time 0.020376 -2023-02-13 17:53:02,142 - Epoch: [78][ 740/ 1207] Overall Loss 0.325448 Objective Loss 0.325448 LR 0.001000 Time 0.020360 -2023-02-13 17:53:02,338 - Epoch: [78][ 750/ 1207] Overall Loss 0.325600 Objective Loss 0.325600 LR 0.001000 Time 0.020350 -2023-02-13 17:53:02,531 - Epoch: [78][ 760/ 1207] Overall Loss 0.325868 Objective Loss 0.325868 LR 0.001000 Time 0.020335 -2023-02-13 17:53:02,727 - Epoch: [78][ 770/ 1207] Overall Loss 0.325925 Objective Loss 0.325925 LR 0.001000 Time 0.020325 -2023-02-13 17:53:02,920 - Epoch: [78][ 780/ 1207] Overall Loss 0.326010 Objective Loss 0.326010 LR 0.001000 Time 0.020311 -2023-02-13 17:53:03,116 - Epoch: [78][ 790/ 1207] Overall Loss 0.325944 Objective Loss 0.325944 LR 0.001000 Time 0.020302 -2023-02-13 17:53:03,309 - Epoch: [78][ 800/ 1207] Overall Loss 0.326131 Objective Loss 0.326131 LR 0.001000 Time 0.020289 -2023-02-13 17:53:03,505 - Epoch: [78][ 810/ 1207] Overall Loss 0.325911 Objective Loss 0.325911 LR 0.001000 Time 0.020281 -2023-02-13 17:53:03,698 - Epoch: [78][ 820/ 1207] Overall Loss 0.325736 Objective Loss 0.325736 LR 0.001000 Time 0.020268 -2023-02-13 17:53:03,894 - Epoch: [78][ 830/ 1207] Overall Loss 0.325868 Objective Loss 0.325868 LR 0.001000 Time 0.020259 -2023-02-13 17:53:04,087 - Epoch: [78][ 840/ 1207] Overall Loss 0.325928 Objective Loss 0.325928 LR 0.001000 Time 0.020247 -2023-02-13 17:53:04,283 - Epoch: [78][ 850/ 1207] Overall Loss 0.325920 Objective Loss 0.325920 LR 0.001000 Time 0.020240 -2023-02-13 17:53:04,476 - Epoch: [78][ 860/ 1207] Overall Loss 0.325566 Objective Loss 0.325566 LR 0.001000 Time 0.020228 -2023-02-13 17:53:04,673 - Epoch: [78][ 870/ 1207] Overall Loss 0.325627 Objective Loss 0.325627 LR 0.001000 Time 0.020221 -2023-02-13 17:53:04,865 - Epoch: [78][ 880/ 1207] Overall Loss 0.325506 Objective Loss 0.325506 LR 0.001000 Time 0.020209 -2023-02-13 17:53:05,061 - Epoch: [78][ 890/ 1207] Overall Loss 0.325280 Objective Loss 0.325280 LR 0.001000 Time 0.020203 -2023-02-13 17:53:05,255 - Epoch: [78][ 900/ 1207] Overall Loss 0.325469 Objective Loss 0.325469 LR 0.001000 Time 0.020192 -2023-02-13 17:53:05,451 - Epoch: [78][ 910/ 1207] Overall Loss 0.325237 Objective Loss 0.325237 LR 0.001000 Time 0.020186 -2023-02-13 17:53:05,644 - Epoch: [78][ 920/ 1207] Overall Loss 0.325092 Objective Loss 0.325092 LR 0.001000 Time 0.020175 -2023-02-13 17:53:05,841 - Epoch: [78][ 930/ 1207] Overall Loss 0.325412 Objective Loss 0.325412 LR 0.001000 Time 0.020170 -2023-02-13 17:53:06,034 - Epoch: [78][ 940/ 1207] Overall Loss 0.325499 Objective Loss 0.325499 LR 0.001000 Time 0.020160 -2023-02-13 17:53:06,230 - Epoch: [78][ 950/ 1207] Overall Loss 0.325456 Objective Loss 0.325456 LR 0.001000 Time 0.020155 -2023-02-13 17:53:06,423 - Epoch: [78][ 960/ 1207] Overall Loss 0.325757 Objective Loss 0.325757 LR 0.001000 Time 0.020145 -2023-02-13 17:53:06,618 - Epoch: [78][ 970/ 1207] Overall Loss 0.326265 Objective Loss 0.326265 LR 0.001000 Time 0.020139 -2023-02-13 17:53:06,812 - Epoch: [78][ 980/ 1207] Overall Loss 0.326135 Objective Loss 0.326135 LR 0.001000 Time 0.020131 -2023-02-13 17:53:07,008 - Epoch: [78][ 990/ 1207] Overall Loss 0.326416 Objective Loss 0.326416 LR 0.001000 Time 0.020125 -2023-02-13 17:53:07,201 - Epoch: [78][ 1000/ 1207] Overall Loss 0.326610 Objective Loss 0.326610 LR 0.001000 Time 0.020116 -2023-02-13 17:53:07,398 - Epoch: [78][ 1010/ 1207] Overall Loss 0.326556 Objective Loss 0.326556 LR 0.001000 Time 0.020111 -2023-02-13 17:53:07,590 - Epoch: [78][ 1020/ 1207] Overall Loss 0.327085 Objective Loss 0.327085 LR 0.001000 Time 0.020102 -2023-02-13 17:53:07,787 - Epoch: [78][ 1030/ 1207] Overall Loss 0.327212 Objective Loss 0.327212 LR 0.001000 Time 0.020098 -2023-02-13 17:53:07,979 - Epoch: [78][ 1040/ 1207] Overall Loss 0.327123 Objective Loss 0.327123 LR 0.001000 Time 0.020089 -2023-02-13 17:53:08,176 - Epoch: [78][ 1050/ 1207] Overall Loss 0.327106 Objective Loss 0.327106 LR 0.001000 Time 0.020085 -2023-02-13 17:53:08,369 - Epoch: [78][ 1060/ 1207] Overall Loss 0.327210 Objective Loss 0.327210 LR 0.001000 Time 0.020077 -2023-02-13 17:53:08,566 - Epoch: [78][ 1070/ 1207] Overall Loss 0.327255 Objective Loss 0.327255 LR 0.001000 Time 0.020073 -2023-02-13 17:53:08,758 - Epoch: [78][ 1080/ 1207] Overall Loss 0.327338 Objective Loss 0.327338 LR 0.001000 Time 0.020065 -2023-02-13 17:53:08,955 - Epoch: [78][ 1090/ 1207] Overall Loss 0.327176 Objective Loss 0.327176 LR 0.001000 Time 0.020061 -2023-02-13 17:53:09,148 - Epoch: [78][ 1100/ 1207] Overall Loss 0.327053 Objective Loss 0.327053 LR 0.001000 Time 0.020054 -2023-02-13 17:53:09,344 - Epoch: [78][ 1110/ 1207] Overall Loss 0.326891 Objective Loss 0.326891 LR 0.001000 Time 0.020049 -2023-02-13 17:53:09,537 - Epoch: [78][ 1120/ 1207] Overall Loss 0.326895 Objective Loss 0.326895 LR 0.001000 Time 0.020042 -2023-02-13 17:53:09,732 - Epoch: [78][ 1130/ 1207] Overall Loss 0.327086 Objective Loss 0.327086 LR 0.001000 Time 0.020037 -2023-02-13 17:53:09,924 - Epoch: [78][ 1140/ 1207] Overall Loss 0.327032 Objective Loss 0.327032 LR 0.001000 Time 0.020030 -2023-02-13 17:53:10,121 - Epoch: [78][ 1150/ 1207] Overall Loss 0.327031 Objective Loss 0.327031 LR 0.001000 Time 0.020026 -2023-02-13 17:53:10,314 - Epoch: [78][ 1160/ 1207] Overall Loss 0.327117 Objective Loss 0.327117 LR 0.001000 Time 0.020020 -2023-02-13 17:53:10,510 - Epoch: [78][ 1170/ 1207] Overall Loss 0.327176 Objective Loss 0.327176 LR 0.001000 Time 0.020016 -2023-02-13 17:53:10,703 - Epoch: [78][ 1180/ 1207] Overall Loss 0.327157 Objective Loss 0.327157 LR 0.001000 Time 0.020009 -2023-02-13 17:53:10,900 - Epoch: [78][ 1190/ 1207] Overall Loss 0.327460 Objective Loss 0.327460 LR 0.001000 Time 0.020007 -2023-02-13 17:53:11,150 - Epoch: [78][ 1200/ 1207] Overall Loss 0.327394 Objective Loss 0.327394 LR 0.001000 Time 0.020048 -2023-02-13 17:53:11,266 - Epoch: [78][ 1207/ 1207] Overall Loss 0.327481 Objective Loss 0.327481 Top1 85.365854 Top5 98.780488 LR 0.001000 Time 0.020027 -2023-02-13 17:53:11,339 - --- validate (epoch=78)----------- -2023-02-13 17:53:11,339 - 34311 samples (256 per mini-batch) -2023-02-13 17:53:11,740 - Epoch: [78][ 10/ 135] Loss 0.340686 Top1 82.304688 Top5 97.460938 -2023-02-13 17:53:11,871 - Epoch: [78][ 20/ 135] Loss 0.351897 Top1 82.109375 Top5 97.226562 -2023-02-13 17:53:11,999 - Epoch: [78][ 30/ 135] Loss 0.349042 Top1 81.875000 Top5 97.343750 -2023-02-13 17:53:12,125 - Epoch: [78][ 40/ 135] Loss 0.355593 Top1 81.923828 Top5 97.275391 -2023-02-13 17:53:12,254 - Epoch: [78][ 50/ 135] Loss 0.355497 Top1 81.890625 Top5 97.257812 -2023-02-13 17:53:12,383 - Epoch: [78][ 60/ 135] Loss 0.353794 Top1 82.018229 Top5 97.272135 -2023-02-13 17:53:12,507 - Epoch: [78][ 70/ 135] Loss 0.356751 Top1 81.858259 Top5 97.237723 -2023-02-13 17:53:12,634 - Epoch: [78][ 80/ 135] Loss 0.353408 Top1 82.001953 Top5 97.226562 -2023-02-13 17:53:12,763 - Epoch: [78][ 90/ 135] Loss 0.352648 Top1 82.118056 Top5 97.196181 -2023-02-13 17:53:12,885 - Epoch: [78][ 100/ 135] Loss 0.355386 Top1 82.085938 Top5 97.187500 -2023-02-13 17:53:13,011 - Epoch: [78][ 110/ 135] Loss 0.356814 Top1 82.034801 Top5 97.226562 -2023-02-13 17:53:13,137 - Epoch: [78][ 120/ 135] Loss 0.355714 Top1 82.089844 Top5 97.203776 -2023-02-13 17:53:13,265 - Epoch: [78][ 130/ 135] Loss 0.359409 Top1 81.995192 Top5 97.187500 -2023-02-13 17:53:13,309 - Epoch: [78][ 135/ 135] Loss 0.358251 Top1 81.967882 Top5 97.193320 -2023-02-13 17:53:13,376 - ==> Top1: 81.968 Top5: 97.193 Loss: 0.358 - -2023-02-13 17:53:13,377 - ==> Confusion: -[[ 848 5 8 3 9 3 0 0 5 46 1 4 1 5 3 4 4 2 3 2 11] - [ 2 918 5 2 7 33 11 21 3 1 2 1 3 1 0 2 5 0 6 4 6] - [ 10 4 927 10 3 2 29 15 1 0 2 3 2 5 3 14 3 5 7 3 10] - [ 8 4 26 901 1 3 0 2 3 0 14 0 5 1 11 2 5 5 19 2 4] - [ 19 12 1 2 967 18 1 1 2 0 0 7 0 0 7 7 8 1 2 3 8] - [ 3 24 0 3 5 953 5 14 0 3 3 15 9 14 0 5 3 1 1 3 6] - [ 2 4 16 0 0 6 1034 5 0 0 2 1 1 2 0 8 2 5 4 5 2] - [ 3 8 8 0 1 34 9 910 1 0 0 12 3 1 0 1 0 3 18 10 2] - [ 19 3 1 3 1 5 0 2 871 48 13 2 3 13 11 1 0 4 6 0 3] - [ 96 4 3 0 7 5 1 1 32 830 1 2 1 15 3 1 0 4 0 1 5] - [ 2 5 5 7 2 5 8 5 15 2 957 2 1 10 2 1 2 2 13 0 5] - [ 2 3 2 0 1 11 1 1 1 0 0 912 33 8 0 3 2 16 2 7 0] - [ 1 0 1 5 0 6 4 0 2 0 0 42 863 2 0 10 1 16 2 0 4] - [ 7 4 2 1 6 18 0 2 9 19 8 6 3 917 3 8 3 2 0 1 5] - [ 18 4 2 34 6 5 0 2 21 6 5 1 5 3 938 4 4 9 15 0 10] - [ 2 1 4 0 7 2 3 0 0 0 0 12 9 1 0 981 5 10 0 5 4] - [ 0 6 1 0 3 3 0 0 2 1 1 5 5 0 0 12 1001 5 1 3 12] - [ 3 3 0 3 0 1 2 0 3 0 1 11 26 1 1 17 0 973 0 1 5] - [ 6 4 6 14 1 2 0 36 6 0 3 2 7 0 13 2 1 1 980 1 1] - [ 1 4 1 0 1 11 9 12 0 0 2 25 3 2 0 5 5 3 2 1048 14] - [ 156 268 255 122 137 317 121 212 93 97 189 190 343 340 123 173 318 126 194 265 9395]] - -2023-02-13 17:53:13,378 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:53:13,379 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:53:13,384 - - -2023-02-13 17:53:13,384 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:53:14,364 - Epoch: [79][ 10/ 1207] Overall Loss 0.313614 Objective Loss 0.313614 LR 0.001000 Time 0.097914 -2023-02-13 17:53:14,565 - Epoch: [79][ 20/ 1207] Overall Loss 0.303156 Objective Loss 0.303156 LR 0.001000 Time 0.058955 -2023-02-13 17:53:14,755 - Epoch: [79][ 30/ 1207] Overall Loss 0.309456 Objective Loss 0.309456 LR 0.001000 Time 0.045650 -2023-02-13 17:53:14,949 - Epoch: [79][ 40/ 1207] Overall Loss 0.314595 Objective Loss 0.314595 LR 0.001000 Time 0.039083 -2023-02-13 17:53:15,141 - Epoch: [79][ 50/ 1207] Overall Loss 0.316745 Objective Loss 0.316745 LR 0.001000 Time 0.035083 -2023-02-13 17:53:15,336 - Epoch: [79][ 60/ 1207] Overall Loss 0.312986 Objective Loss 0.312986 LR 0.001000 Time 0.032479 -2023-02-13 17:53:15,526 - Epoch: [79][ 70/ 1207] Overall Loss 0.312845 Objective Loss 0.312845 LR 0.001000 Time 0.030553 -2023-02-13 17:53:15,720 - Epoch: [79][ 80/ 1207] Overall Loss 0.314445 Objective Loss 0.314445 LR 0.001000 Time 0.029157 -2023-02-13 17:53:15,912 - Epoch: [79][ 90/ 1207] Overall Loss 0.318546 Objective Loss 0.318546 LR 0.001000 Time 0.028047 -2023-02-13 17:53:16,107 - Epoch: [79][ 100/ 1207] Overall Loss 0.318249 Objective Loss 0.318249 LR 0.001000 Time 0.027183 -2023-02-13 17:53:16,298 - Epoch: [79][ 110/ 1207] Overall Loss 0.317488 Objective Loss 0.317488 LR 0.001000 Time 0.026451 -2023-02-13 17:53:16,495 - Epoch: [79][ 120/ 1207] Overall Loss 0.316125 Objective Loss 0.316125 LR 0.001000 Time 0.025879 -2023-02-13 17:53:16,689 - Epoch: [79][ 130/ 1207] Overall Loss 0.318775 Objective Loss 0.318775 LR 0.001000 Time 0.025378 -2023-02-13 17:53:16,886 - Epoch: [79][ 140/ 1207] Overall Loss 0.319304 Objective Loss 0.319304 LR 0.001000 Time 0.024971 -2023-02-13 17:53:17,080 - Epoch: [79][ 150/ 1207] Overall Loss 0.320893 Objective Loss 0.320893 LR 0.001000 Time 0.024595 -2023-02-13 17:53:17,277 - Epoch: [79][ 160/ 1207] Overall Loss 0.319856 Objective Loss 0.319856 LR 0.001000 Time 0.024292 -2023-02-13 17:53:17,471 - Epoch: [79][ 170/ 1207] Overall Loss 0.319186 Objective Loss 0.319186 LR 0.001000 Time 0.023998 -2023-02-13 17:53:17,667 - Epoch: [79][ 180/ 1207] Overall Loss 0.319926 Objective Loss 0.319926 LR 0.001000 Time 0.023756 -2023-02-13 17:53:17,861 - Epoch: [79][ 190/ 1207] Overall Loss 0.319444 Objective Loss 0.319444 LR 0.001000 Time 0.023523 -2023-02-13 17:53:18,058 - Epoch: [79][ 200/ 1207] Overall Loss 0.318167 Objective Loss 0.318167 LR 0.001000 Time 0.023328 -2023-02-13 17:53:18,252 - Epoch: [79][ 210/ 1207] Overall Loss 0.316930 Objective Loss 0.316930 LR 0.001000 Time 0.023138 -2023-02-13 17:53:18,447 - Epoch: [79][ 220/ 1207] Overall Loss 0.316794 Objective Loss 0.316794 LR 0.001000 Time 0.022973 -2023-02-13 17:53:18,639 - Epoch: [79][ 230/ 1207] Overall Loss 0.316887 Objective Loss 0.316887 LR 0.001000 Time 0.022808 -2023-02-13 17:53:18,835 - Epoch: [79][ 240/ 1207] Overall Loss 0.317717 Objective Loss 0.317717 LR 0.001000 Time 0.022674 -2023-02-13 17:53:19,028 - Epoch: [79][ 250/ 1207] Overall Loss 0.316884 Objective Loss 0.316884 LR 0.001000 Time 0.022534 -2023-02-13 17:53:19,224 - Epoch: [79][ 260/ 1207] Overall Loss 0.316726 Objective Loss 0.316726 LR 0.001000 Time 0.022420 -2023-02-13 17:53:19,417 - Epoch: [79][ 270/ 1207] Overall Loss 0.316910 Objective Loss 0.316910 LR 0.001000 Time 0.022303 -2023-02-13 17:53:19,612 - Epoch: [79][ 280/ 1207] Overall Loss 0.316708 Objective Loss 0.316708 LR 0.001000 Time 0.022204 -2023-02-13 17:53:19,804 - Epoch: [79][ 290/ 1207] Overall Loss 0.316380 Objective Loss 0.316380 LR 0.001000 Time 0.022099 -2023-02-13 17:53:20,000 - Epoch: [79][ 300/ 1207] Overall Loss 0.317147 Objective Loss 0.317147 LR 0.001000 Time 0.022013 -2023-02-13 17:53:20,193 - Epoch: [79][ 310/ 1207] Overall Loss 0.318227 Objective Loss 0.318227 LR 0.001000 Time 0.021924 -2023-02-13 17:53:20,389 - Epoch: [79][ 320/ 1207] Overall Loss 0.318201 Objective Loss 0.318201 LR 0.001000 Time 0.021851 -2023-02-13 17:53:20,580 - Epoch: [79][ 330/ 1207] Overall Loss 0.318248 Objective Loss 0.318248 LR 0.001000 Time 0.021768 -2023-02-13 17:53:20,775 - Epoch: [79][ 340/ 1207] Overall Loss 0.319231 Objective Loss 0.319231 LR 0.001000 Time 0.021699 -2023-02-13 17:53:20,969 - Epoch: [79][ 350/ 1207] Overall Loss 0.319619 Objective Loss 0.319619 LR 0.001000 Time 0.021630 -2023-02-13 17:53:21,163 - Epoch: [79][ 360/ 1207] Overall Loss 0.319893 Objective Loss 0.319893 LR 0.001000 Time 0.021570 -2023-02-13 17:53:21,356 - Epoch: [79][ 370/ 1207] Overall Loss 0.319514 Objective Loss 0.319514 LR 0.001000 Time 0.021506 -2023-02-13 17:53:21,550 - Epoch: [79][ 380/ 1207] Overall Loss 0.320217 Objective Loss 0.320217 LR 0.001000 Time 0.021450 -2023-02-13 17:53:21,742 - Epoch: [79][ 390/ 1207] Overall Loss 0.320904 Objective Loss 0.320904 LR 0.001000 Time 0.021392 -2023-02-13 17:53:21,938 - Epoch: [79][ 400/ 1207] Overall Loss 0.320434 Objective Loss 0.320434 LR 0.001000 Time 0.021344 -2023-02-13 17:53:22,130 - Epoch: [79][ 410/ 1207] Overall Loss 0.320132 Objective Loss 0.320132 LR 0.001000 Time 0.021292 -2023-02-13 17:53:22,326 - Epoch: [79][ 420/ 1207] Overall Loss 0.320479 Objective Loss 0.320479 LR 0.001000 Time 0.021251 -2023-02-13 17:53:22,518 - Epoch: [79][ 430/ 1207] Overall Loss 0.320458 Objective Loss 0.320458 LR 0.001000 Time 0.021203 -2023-02-13 17:53:22,714 - Epoch: [79][ 440/ 1207] Overall Loss 0.320879 Objective Loss 0.320879 LR 0.001000 Time 0.021166 -2023-02-13 17:53:22,906 - Epoch: [79][ 450/ 1207] Overall Loss 0.320077 Objective Loss 0.320077 LR 0.001000 Time 0.021121 -2023-02-13 17:53:23,103 - Epoch: [79][ 460/ 1207] Overall Loss 0.319539 Objective Loss 0.319539 LR 0.001000 Time 0.021089 -2023-02-13 17:53:23,296 - Epoch: [79][ 470/ 1207] Overall Loss 0.319172 Objective Loss 0.319172 LR 0.001000 Time 0.021051 -2023-02-13 17:53:23,493 - Epoch: [79][ 480/ 1207] Overall Loss 0.319259 Objective Loss 0.319259 LR 0.001000 Time 0.021021 -2023-02-13 17:53:23,685 - Epoch: [79][ 490/ 1207] Overall Loss 0.319202 Objective Loss 0.319202 LR 0.001000 Time 0.020984 -2023-02-13 17:53:23,881 - Epoch: [79][ 500/ 1207] Overall Loss 0.320571 Objective Loss 0.320571 LR 0.001000 Time 0.020955 -2023-02-13 17:53:24,074 - Epoch: [79][ 510/ 1207] Overall Loss 0.320706 Objective Loss 0.320706 LR 0.001000 Time 0.020921 -2023-02-13 17:53:24,271 - Epoch: [79][ 520/ 1207] Overall Loss 0.321174 Objective Loss 0.321174 LR 0.001000 Time 0.020897 -2023-02-13 17:53:24,463 - Epoch: [79][ 530/ 1207] Overall Loss 0.321408 Objective Loss 0.321408 LR 0.001000 Time 0.020865 -2023-02-13 17:53:24,660 - Epoch: [79][ 540/ 1207] Overall Loss 0.321868 Objective Loss 0.321868 LR 0.001000 Time 0.020843 -2023-02-13 17:53:24,853 - Epoch: [79][ 550/ 1207] Overall Loss 0.321878 Objective Loss 0.321878 LR 0.001000 Time 0.020813 -2023-02-13 17:53:25,049 - Epoch: [79][ 560/ 1207] Overall Loss 0.322644 Objective Loss 0.322644 LR 0.001000 Time 0.020791 -2023-02-13 17:53:25,242 - Epoch: [79][ 570/ 1207] Overall Loss 0.322896 Objective Loss 0.322896 LR 0.001000 Time 0.020765 -2023-02-13 17:53:25,439 - Epoch: [79][ 580/ 1207] Overall Loss 0.322438 Objective Loss 0.322438 LR 0.001000 Time 0.020747 -2023-02-13 17:53:25,632 - Epoch: [79][ 590/ 1207] Overall Loss 0.322539 Objective Loss 0.322539 LR 0.001000 Time 0.020721 -2023-02-13 17:53:25,829 - Epoch: [79][ 600/ 1207] Overall Loss 0.322856 Objective Loss 0.322856 LR 0.001000 Time 0.020704 -2023-02-13 17:53:26,022 - Epoch: [79][ 610/ 1207] Overall Loss 0.322598 Objective Loss 0.322598 LR 0.001000 Time 0.020680 -2023-02-13 17:53:26,217 - Epoch: [79][ 620/ 1207] Overall Loss 0.322723 Objective Loss 0.322723 LR 0.001000 Time 0.020659 -2023-02-13 17:53:26,406 - Epoch: [79][ 630/ 1207] Overall Loss 0.323014 Objective Loss 0.323014 LR 0.001000 Time 0.020631 -2023-02-13 17:53:26,596 - Epoch: [79][ 640/ 1207] Overall Loss 0.323203 Objective Loss 0.323203 LR 0.001000 Time 0.020605 -2023-02-13 17:53:26,786 - Epoch: [79][ 650/ 1207] Overall Loss 0.323816 Objective Loss 0.323816 LR 0.001000 Time 0.020580 -2023-02-13 17:53:26,977 - Epoch: [79][ 660/ 1207] Overall Loss 0.323575 Objective Loss 0.323575 LR 0.001000 Time 0.020556 -2023-02-13 17:53:27,166 - Epoch: [79][ 670/ 1207] Overall Loss 0.323752 Objective Loss 0.323752 LR 0.001000 Time 0.020531 -2023-02-13 17:53:27,357 - Epoch: [79][ 680/ 1207] Overall Loss 0.323780 Objective Loss 0.323780 LR 0.001000 Time 0.020510 -2023-02-13 17:53:27,546 - Epoch: [79][ 690/ 1207] Overall Loss 0.323617 Objective Loss 0.323617 LR 0.001000 Time 0.020487 -2023-02-13 17:53:27,737 - Epoch: [79][ 700/ 1207] Overall Loss 0.323783 Objective Loss 0.323783 LR 0.001000 Time 0.020466 -2023-02-13 17:53:27,927 - Epoch: [79][ 710/ 1207] Overall Loss 0.323858 Objective Loss 0.323858 LR 0.001000 Time 0.020445 -2023-02-13 17:53:28,118 - Epoch: [79][ 720/ 1207] Overall Loss 0.324019 Objective Loss 0.324019 LR 0.001000 Time 0.020426 -2023-02-13 17:53:28,308 - Epoch: [79][ 730/ 1207] Overall Loss 0.324368 Objective Loss 0.324368 LR 0.001000 Time 0.020406 -2023-02-13 17:53:28,500 - Epoch: [79][ 740/ 1207] Overall Loss 0.324599 Objective Loss 0.324599 LR 0.001000 Time 0.020389 -2023-02-13 17:53:28,690 - Epoch: [79][ 750/ 1207] Overall Loss 0.324559 Objective Loss 0.324559 LR 0.001000 Time 0.020369 -2023-02-13 17:53:28,880 - Epoch: [79][ 760/ 1207] Overall Loss 0.324639 Objective Loss 0.324639 LR 0.001000 Time 0.020352 -2023-02-13 17:53:29,070 - Epoch: [79][ 770/ 1207] Overall Loss 0.325114 Objective Loss 0.325114 LR 0.001000 Time 0.020334 -2023-02-13 17:53:29,261 - Epoch: [79][ 780/ 1207] Overall Loss 0.325183 Objective Loss 0.325183 LR 0.001000 Time 0.020317 -2023-02-13 17:53:29,451 - Epoch: [79][ 790/ 1207] Overall Loss 0.325253 Objective Loss 0.325253 LR 0.001000 Time 0.020300 -2023-02-13 17:53:29,641 - Epoch: [79][ 800/ 1207] Overall Loss 0.325116 Objective Loss 0.325116 LR 0.001000 Time 0.020284 -2023-02-13 17:53:29,831 - Epoch: [79][ 810/ 1207] Overall Loss 0.325338 Objective Loss 0.325338 LR 0.001000 Time 0.020267 -2023-02-13 17:53:30,022 - Epoch: [79][ 820/ 1207] Overall Loss 0.325566 Objective Loss 0.325566 LR 0.001000 Time 0.020252 -2023-02-13 17:53:30,211 - Epoch: [79][ 830/ 1207] Overall Loss 0.325487 Objective Loss 0.325487 LR 0.001000 Time 0.020236 -2023-02-13 17:53:30,403 - Epoch: [79][ 840/ 1207] Overall Loss 0.325055 Objective Loss 0.325055 LR 0.001000 Time 0.020223 -2023-02-13 17:53:30,592 - Epoch: [79][ 850/ 1207] Overall Loss 0.324834 Objective Loss 0.324834 LR 0.001000 Time 0.020207 -2023-02-13 17:53:30,783 - Epoch: [79][ 860/ 1207] Overall Loss 0.324777 Objective Loss 0.324777 LR 0.001000 Time 0.020193 -2023-02-13 17:53:30,978 - Epoch: [79][ 870/ 1207] Overall Loss 0.325083 Objective Loss 0.325083 LR 0.001000 Time 0.020185 -2023-02-13 17:53:31,168 - Epoch: [79][ 880/ 1207] Overall Loss 0.325482 Objective Loss 0.325482 LR 0.001000 Time 0.020172 -2023-02-13 17:53:31,359 - Epoch: [79][ 890/ 1207] Overall Loss 0.325450 Objective Loss 0.325450 LR 0.001000 Time 0.020159 -2023-02-13 17:53:31,550 - Epoch: [79][ 900/ 1207] Overall Loss 0.325252 Objective Loss 0.325252 LR 0.001000 Time 0.020146 -2023-02-13 17:53:31,740 - Epoch: [79][ 910/ 1207] Overall Loss 0.325063 Objective Loss 0.325063 LR 0.001000 Time 0.020133 -2023-02-13 17:53:31,931 - Epoch: [79][ 920/ 1207] Overall Loss 0.325277 Objective Loss 0.325277 LR 0.001000 Time 0.020121 -2023-02-13 17:53:32,120 - Epoch: [79][ 930/ 1207] Overall Loss 0.325838 Objective Loss 0.325838 LR 0.001000 Time 0.020109 -2023-02-13 17:53:32,311 - Epoch: [79][ 940/ 1207] Overall Loss 0.326155 Objective Loss 0.326155 LR 0.001000 Time 0.020097 -2023-02-13 17:53:32,501 - Epoch: [79][ 950/ 1207] Overall Loss 0.326403 Objective Loss 0.326403 LR 0.001000 Time 0.020085 -2023-02-13 17:53:32,691 - Epoch: [79][ 960/ 1207] Overall Loss 0.326633 Objective Loss 0.326633 LR 0.001000 Time 0.020073 -2023-02-13 17:53:32,880 - Epoch: [79][ 970/ 1207] Overall Loss 0.326417 Objective Loss 0.326417 LR 0.001000 Time 0.020061 -2023-02-13 17:53:33,070 - Epoch: [79][ 980/ 1207] Overall Loss 0.326576 Objective Loss 0.326576 LR 0.001000 Time 0.020050 -2023-02-13 17:53:33,260 - Epoch: [79][ 990/ 1207] Overall Loss 0.326595 Objective Loss 0.326595 LR 0.001000 Time 0.020039 -2023-02-13 17:53:33,450 - Epoch: [79][ 1000/ 1207] Overall Loss 0.326722 Objective Loss 0.326722 LR 0.001000 Time 0.020028 -2023-02-13 17:53:33,640 - Epoch: [79][ 1010/ 1207] Overall Loss 0.326881 Objective Loss 0.326881 LR 0.001000 Time 0.020017 -2023-02-13 17:53:33,831 - Epoch: [79][ 1020/ 1207] Overall Loss 0.327433 Objective Loss 0.327433 LR 0.001000 Time 0.020008 -2023-02-13 17:53:34,021 - Epoch: [79][ 1030/ 1207] Overall Loss 0.327384 Objective Loss 0.327384 LR 0.001000 Time 0.019998 -2023-02-13 17:53:34,211 - Epoch: [79][ 1040/ 1207] Overall Loss 0.327480 Objective Loss 0.327480 LR 0.001000 Time 0.019988 -2023-02-13 17:53:34,402 - Epoch: [79][ 1050/ 1207] Overall Loss 0.327473 Objective Loss 0.327473 LR 0.001000 Time 0.019979 -2023-02-13 17:53:34,593 - Epoch: [79][ 1060/ 1207] Overall Loss 0.327413 Objective Loss 0.327413 LR 0.001000 Time 0.019970 -2023-02-13 17:53:34,782 - Epoch: [79][ 1070/ 1207] Overall Loss 0.327799 Objective Loss 0.327799 LR 0.001000 Time 0.019961 -2023-02-13 17:53:34,973 - Epoch: [79][ 1080/ 1207] Overall Loss 0.327913 Objective Loss 0.327913 LR 0.001000 Time 0.019952 -2023-02-13 17:53:35,163 - Epoch: [79][ 1090/ 1207] Overall Loss 0.328154 Objective Loss 0.328154 LR 0.001000 Time 0.019943 -2023-02-13 17:53:35,354 - Epoch: [79][ 1100/ 1207] Overall Loss 0.328183 Objective Loss 0.328183 LR 0.001000 Time 0.019935 -2023-02-13 17:53:35,544 - Epoch: [79][ 1110/ 1207] Overall Loss 0.328155 Objective Loss 0.328155 LR 0.001000 Time 0.019926 -2023-02-13 17:53:35,734 - Epoch: [79][ 1120/ 1207] Overall Loss 0.328023 Objective Loss 0.328023 LR 0.001000 Time 0.019917 -2023-02-13 17:53:35,925 - Epoch: [79][ 1130/ 1207] Overall Loss 0.327889 Objective Loss 0.327889 LR 0.001000 Time 0.019910 -2023-02-13 17:53:36,116 - Epoch: [79][ 1140/ 1207] Overall Loss 0.327634 Objective Loss 0.327634 LR 0.001000 Time 0.019903 -2023-02-13 17:53:36,306 - Epoch: [79][ 1150/ 1207] Overall Loss 0.327700 Objective Loss 0.327700 LR 0.001000 Time 0.019894 -2023-02-13 17:53:36,497 - Epoch: [79][ 1160/ 1207] Overall Loss 0.327779 Objective Loss 0.327779 LR 0.001000 Time 0.019887 -2023-02-13 17:53:36,686 - Epoch: [79][ 1170/ 1207] Overall Loss 0.327558 Objective Loss 0.327558 LR 0.001000 Time 0.019879 -2023-02-13 17:53:36,879 - Epoch: [79][ 1180/ 1207] Overall Loss 0.327387 Objective Loss 0.327387 LR 0.001000 Time 0.019873 -2023-02-13 17:53:37,069 - Epoch: [79][ 1190/ 1207] Overall Loss 0.327378 Objective Loss 0.327378 LR 0.001000 Time 0.019866 -2023-02-13 17:53:37,315 - Epoch: [79][ 1200/ 1207] Overall Loss 0.327298 Objective Loss 0.327298 LR 0.001000 Time 0.019905 -2023-02-13 17:53:37,429 - Epoch: [79][ 1207/ 1207] Overall Loss 0.327428 Objective Loss 0.327428 Top1 83.841463 Top5 95.731707 LR 0.001000 Time 0.019884 -2023-02-13 17:53:37,501 - --- validate (epoch=79)----------- -2023-02-13 17:53:37,502 - 34311 samples (256 per mini-batch) -2023-02-13 17:53:37,895 - Epoch: [79][ 10/ 135] Loss 0.386655 Top1 81.093750 Top5 96.953125 -2023-02-13 17:53:38,017 - Epoch: [79][ 20/ 135] Loss 0.353754 Top1 81.464844 Top5 97.207031 -2023-02-13 17:53:38,138 - Epoch: [79][ 30/ 135] Loss 0.362292 Top1 81.471354 Top5 97.044271 -2023-02-13 17:53:38,261 - Epoch: [79][ 40/ 135] Loss 0.365987 Top1 81.542969 Top5 97.128906 -2023-02-13 17:53:38,384 - Epoch: [79][ 50/ 135] Loss 0.368435 Top1 81.625000 Top5 97.054688 -2023-02-13 17:53:38,508 - Epoch: [79][ 60/ 135] Loss 0.370552 Top1 81.575521 Top5 97.109375 -2023-02-13 17:53:38,630 - Epoch: [79][ 70/ 135] Loss 0.367942 Top1 81.791295 Top5 97.154018 -2023-02-13 17:53:38,754 - Epoch: [79][ 80/ 135] Loss 0.372089 Top1 81.582031 Top5 97.114258 -2023-02-13 17:53:38,876 - Epoch: [79][ 90/ 135] Loss 0.370280 Top1 81.627604 Top5 97.105035 -2023-02-13 17:53:38,999 - Epoch: [79][ 100/ 135] Loss 0.371237 Top1 81.640625 Top5 97.128906 -2023-02-13 17:53:39,121 - Epoch: [79][ 110/ 135] Loss 0.372472 Top1 81.622869 Top5 97.091619 -2023-02-13 17:53:39,245 - Epoch: [79][ 120/ 135] Loss 0.368934 Top1 81.682943 Top5 97.125651 -2023-02-13 17:53:39,370 - Epoch: [79][ 130/ 135] Loss 0.365912 Top1 81.721755 Top5 97.109375 -2023-02-13 17:53:39,414 - Epoch: [79][ 135/ 135] Loss 0.364156 Top1 81.688088 Top5 97.117542 -2023-02-13 17:53:39,481 - ==> Top1: 81.688 Top5: 97.118 Loss: 0.364 - -2023-02-13 17:53:39,482 - ==> Confusion: -[[ 857 2 8 0 8 4 1 2 2 44 0 5 0 4 3 6 1 2 1 3 14] - [ 1 919 3 1 12 29 10 22 7 3 1 3 0 1 0 2 5 0 5 3 6] - [ 10 5 923 20 1 1 25 22 0 2 3 2 4 2 3 5 2 4 9 4 11] - [ 7 1 24 885 2 3 2 3 2 1 15 1 10 2 21 2 4 3 19 2 7] - [ 21 9 3 1 965 9 1 2 1 4 3 8 2 5 7 5 11 1 0 3 5] - [ 2 23 3 4 10 937 5 15 3 6 3 17 6 12 1 2 3 1 2 9 6] - [ 4 5 17 3 1 5 1025 10 0 0 1 1 1 2 0 8 1 6 3 3 3] - [ 2 8 8 2 2 39 5 905 3 3 0 8 2 0 0 1 0 1 21 9 5] - [ 21 3 0 1 2 0 1 1 882 49 13 1 2 10 15 1 0 2 2 0 3] - [ 96 0 3 0 4 3 0 0 30 842 1 2 2 18 2 0 1 3 2 0 3] - [ 4 1 6 6 2 1 8 7 12 1 970 3 0 11 4 1 0 0 8 2 4] - [ 3 3 1 1 1 6 1 6 0 1 0 902 38 9 0 4 7 7 2 10 3] - [ 1 0 2 6 1 3 1 3 0 0 0 36 868 0 4 6 2 17 0 1 8] - [ 7 5 2 0 8 9 1 0 22 16 15 9 3 902 8 4 4 1 1 3 4] - [ 11 2 0 19 7 2 0 2 17 7 6 2 5 3 978 1 1 4 12 0 13] - [ 4 1 4 0 13 1 3 1 0 1 0 4 9 1 1 968 8 15 1 8 3] - [ 3 8 1 0 5 4 0 1 3 0 0 3 2 3 5 9 1000 0 2 2 10] - [ 6 2 1 5 1 1 1 1 0 1 0 12 35 1 1 17 4 958 0 0 4] - [ 4 3 6 11 2 0 0 34 4 0 3 1 10 0 13 1 1 1 986 3 3] - [ 2 2 2 0 1 5 7 20 0 0 1 17 3 3 0 7 8 4 0 1059 7] - [ 189 258 249 140 167 210 76 200 142 107 260 177 362 323 202 153 278 100 214 330 9297]] - -2023-02-13 17:53:39,484 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:53:39,484 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:53:39,489 - - -2023-02-13 17:53:39,489 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:53:40,484 - Epoch: [80][ 10/ 1207] Overall Loss 0.314447 Objective Loss 0.314447 LR 0.001000 Time 0.099395 -2023-02-13 17:53:40,678 - Epoch: [80][ 20/ 1207] Overall Loss 0.320906 Objective Loss 0.320906 LR 0.001000 Time 0.059374 -2023-02-13 17:53:40,873 - Epoch: [80][ 30/ 1207] Overall Loss 0.318654 Objective Loss 0.318654 LR 0.001000 Time 0.046061 -2023-02-13 17:53:41,065 - Epoch: [80][ 40/ 1207] Overall Loss 0.322130 Objective Loss 0.322130 LR 0.001000 Time 0.039346 -2023-02-13 17:53:41,260 - Epoch: [80][ 50/ 1207] Overall Loss 0.318138 Objective Loss 0.318138 LR 0.001000 Time 0.035365 -2023-02-13 17:53:41,452 - Epoch: [80][ 60/ 1207] Overall Loss 0.320490 Objective Loss 0.320490 LR 0.001000 Time 0.032672 -2023-02-13 17:53:41,647 - Epoch: [80][ 70/ 1207] Overall Loss 0.325818 Objective Loss 0.325818 LR 0.001000 Time 0.030785 -2023-02-13 17:53:41,840 - Epoch: [80][ 80/ 1207] Overall Loss 0.324941 Objective Loss 0.324941 LR 0.001000 Time 0.029340 -2023-02-13 17:53:42,035 - Epoch: [80][ 90/ 1207] Overall Loss 0.323488 Objective Loss 0.323488 LR 0.001000 Time 0.028245 -2023-02-13 17:53:42,227 - Epoch: [80][ 100/ 1207] Overall Loss 0.321791 Objective Loss 0.321791 LR 0.001000 Time 0.027338 -2023-02-13 17:53:42,423 - Epoch: [80][ 110/ 1207] Overall Loss 0.323425 Objective Loss 0.323425 LR 0.001000 Time 0.026633 -2023-02-13 17:53:42,616 - Epoch: [80][ 120/ 1207] Overall Loss 0.322312 Objective Loss 0.322312 LR 0.001000 Time 0.026011 -2023-02-13 17:53:42,810 - Epoch: [80][ 130/ 1207] Overall Loss 0.323910 Objective Loss 0.323910 LR 0.001000 Time 0.025504 -2023-02-13 17:53:43,002 - Epoch: [80][ 140/ 1207] Overall Loss 0.325486 Objective Loss 0.325486 LR 0.001000 Time 0.025052 -2023-02-13 17:53:43,197 - Epoch: [80][ 150/ 1207] Overall Loss 0.325424 Objective Loss 0.325424 LR 0.001000 Time 0.024681 -2023-02-13 17:53:43,390 - Epoch: [80][ 160/ 1207] Overall Loss 0.326283 Objective Loss 0.326283 LR 0.001000 Time 0.024340 -2023-02-13 17:53:43,586 - Epoch: [80][ 170/ 1207] Overall Loss 0.326857 Objective Loss 0.326857 LR 0.001000 Time 0.024056 -2023-02-13 17:53:43,778 - Epoch: [80][ 180/ 1207] Overall Loss 0.325948 Objective Loss 0.325948 LR 0.001000 Time 0.023786 -2023-02-13 17:53:43,973 - Epoch: [80][ 190/ 1207] Overall Loss 0.326520 Objective Loss 0.326520 LR 0.001000 Time 0.023560 -2023-02-13 17:53:44,166 - Epoch: [80][ 200/ 1207] Overall Loss 0.325813 Objective Loss 0.325813 LR 0.001000 Time 0.023342 -2023-02-13 17:53:44,361 - Epoch: [80][ 210/ 1207] Overall Loss 0.325840 Objective Loss 0.325840 LR 0.001000 Time 0.023160 -2023-02-13 17:53:44,553 - Epoch: [80][ 220/ 1207] Overall Loss 0.325612 Objective Loss 0.325612 LR 0.001000 Time 0.022977 -2023-02-13 17:53:44,749 - Epoch: [80][ 230/ 1207] Overall Loss 0.325417 Objective Loss 0.325417 LR 0.001000 Time 0.022828 -2023-02-13 17:53:44,941 - Epoch: [80][ 240/ 1207] Overall Loss 0.322731 Objective Loss 0.322731 LR 0.001000 Time 0.022675 -2023-02-13 17:53:45,136 - Epoch: [80][ 250/ 1207] Overall Loss 0.321659 Objective Loss 0.321659 LR 0.001000 Time 0.022547 -2023-02-13 17:53:45,329 - Epoch: [80][ 260/ 1207] Overall Loss 0.321134 Objective Loss 0.321134 LR 0.001000 Time 0.022420 -2023-02-13 17:53:45,524 - Epoch: [80][ 270/ 1207] Overall Loss 0.321579 Objective Loss 0.321579 LR 0.001000 Time 0.022310 -2023-02-13 17:53:45,716 - Epoch: [80][ 280/ 1207] Overall Loss 0.321201 Objective Loss 0.321201 LR 0.001000 Time 0.022199 -2023-02-13 17:53:45,913 - Epoch: [80][ 290/ 1207] Overall Loss 0.320783 Objective Loss 0.320783 LR 0.001000 Time 0.022112 -2023-02-13 17:53:46,105 - Epoch: [80][ 300/ 1207] Overall Loss 0.322023 Objective Loss 0.322023 LR 0.001000 Time 0.022015 -2023-02-13 17:53:46,301 - Epoch: [80][ 310/ 1207] Overall Loss 0.321813 Objective Loss 0.321813 LR 0.001000 Time 0.021936 -2023-02-13 17:53:46,494 - Epoch: [80][ 320/ 1207] Overall Loss 0.321523 Objective Loss 0.321523 LR 0.001000 Time 0.021851 -2023-02-13 17:53:46,684 - Epoch: [80][ 330/ 1207] Overall Loss 0.322485 Objective Loss 0.322485 LR 0.001000 Time 0.021764 -2023-02-13 17:53:46,873 - Epoch: [80][ 340/ 1207] Overall Loss 0.323451 Objective Loss 0.323451 LR 0.001000 Time 0.021678 -2023-02-13 17:53:47,061 - Epoch: [80][ 350/ 1207] Overall Loss 0.324761 Objective Loss 0.324761 LR 0.001000 Time 0.021596 -2023-02-13 17:53:47,249 - Epoch: [80][ 360/ 1207] Overall Loss 0.326051 Objective Loss 0.326051 LR 0.001000 Time 0.021517 -2023-02-13 17:53:47,438 - Epoch: [80][ 370/ 1207] Overall Loss 0.325382 Objective Loss 0.325382 LR 0.001000 Time 0.021444 -2023-02-13 17:53:47,629 - Epoch: [80][ 380/ 1207] Overall Loss 0.325236 Objective Loss 0.325236 LR 0.001000 Time 0.021382 -2023-02-13 17:53:47,821 - Epoch: [80][ 390/ 1207] Overall Loss 0.325780 Objective Loss 0.325780 LR 0.001000 Time 0.021325 -2023-02-13 17:53:48,016 - Epoch: [80][ 400/ 1207] Overall Loss 0.326379 Objective Loss 0.326379 LR 0.001000 Time 0.021279 -2023-02-13 17:53:48,209 - Epoch: [80][ 410/ 1207] Overall Loss 0.326244 Objective Loss 0.326244 LR 0.001000 Time 0.021230 -2023-02-13 17:53:48,406 - Epoch: [80][ 420/ 1207] Overall Loss 0.326575 Objective Loss 0.326575 LR 0.001000 Time 0.021191 -2023-02-13 17:53:48,598 - Epoch: [80][ 430/ 1207] Overall Loss 0.327187 Objective Loss 0.327187 LR 0.001000 Time 0.021144 -2023-02-13 17:53:48,793 - Epoch: [80][ 440/ 1207] Overall Loss 0.327074 Objective Loss 0.327074 LR 0.001000 Time 0.021107 -2023-02-13 17:53:48,987 - Epoch: [80][ 450/ 1207] Overall Loss 0.327573 Objective Loss 0.327573 LR 0.001000 Time 0.021068 -2023-02-13 17:53:49,183 - Epoch: [80][ 460/ 1207] Overall Loss 0.327654 Objective Loss 0.327654 LR 0.001000 Time 0.021036 -2023-02-13 17:53:49,376 - Epoch: [80][ 470/ 1207] Overall Loss 0.327177 Objective Loss 0.327177 LR 0.001000 Time 0.020998 -2023-02-13 17:53:49,572 - Epoch: [80][ 480/ 1207] Overall Loss 0.327306 Objective Loss 0.327306 LR 0.001000 Time 0.020968 -2023-02-13 17:53:49,765 - Epoch: [80][ 490/ 1207] Overall Loss 0.327380 Objective Loss 0.327380 LR 0.001000 Time 0.020933 -2023-02-13 17:53:49,961 - Epoch: [80][ 500/ 1207] Overall Loss 0.328229 Objective Loss 0.328229 LR 0.001000 Time 0.020905 -2023-02-13 17:53:50,154 - Epoch: [80][ 510/ 1207] Overall Loss 0.328188 Objective Loss 0.328188 LR 0.001000 Time 0.020873 -2023-02-13 17:53:50,350 - Epoch: [80][ 520/ 1207] Overall Loss 0.329016 Objective Loss 0.329016 LR 0.001000 Time 0.020849 -2023-02-13 17:53:50,543 - Epoch: [80][ 530/ 1207] Overall Loss 0.328781 Objective Loss 0.328781 LR 0.001000 Time 0.020819 -2023-02-13 17:53:50,739 - Epoch: [80][ 540/ 1207] Overall Loss 0.328677 Objective Loss 0.328677 LR 0.001000 Time 0.020795 -2023-02-13 17:53:50,932 - Epoch: [80][ 550/ 1207] Overall Loss 0.328509 Objective Loss 0.328509 LR 0.001000 Time 0.020768 -2023-02-13 17:53:51,128 - Epoch: [80][ 560/ 1207] Overall Loss 0.328259 Objective Loss 0.328259 LR 0.001000 Time 0.020746 -2023-02-13 17:53:51,321 - Epoch: [80][ 570/ 1207] Overall Loss 0.328146 Objective Loss 0.328146 LR 0.001000 Time 0.020720 -2023-02-13 17:53:51,517 - Epoch: [80][ 580/ 1207] Overall Loss 0.328211 Objective Loss 0.328211 LR 0.001000 Time 0.020700 -2023-02-13 17:53:51,710 - Epoch: [80][ 590/ 1207] Overall Loss 0.328273 Objective Loss 0.328273 LR 0.001000 Time 0.020675 -2023-02-13 17:53:51,907 - Epoch: [80][ 600/ 1207] Overall Loss 0.327804 Objective Loss 0.327804 LR 0.001000 Time 0.020657 -2023-02-13 17:53:52,099 - Epoch: [80][ 610/ 1207] Overall Loss 0.327797 Objective Loss 0.327797 LR 0.001000 Time 0.020634 -2023-02-13 17:53:52,295 - Epoch: [80][ 620/ 1207] Overall Loss 0.328178 Objective Loss 0.328178 LR 0.001000 Time 0.020617 -2023-02-13 17:53:52,489 - Epoch: [80][ 630/ 1207] Overall Loss 0.328633 Objective Loss 0.328633 LR 0.001000 Time 0.020596 -2023-02-13 17:53:52,684 - Epoch: [80][ 640/ 1207] Overall Loss 0.328611 Objective Loss 0.328611 LR 0.001000 Time 0.020579 -2023-02-13 17:53:52,877 - Epoch: [80][ 650/ 1207] Overall Loss 0.328949 Objective Loss 0.328949 LR 0.001000 Time 0.020558 -2023-02-13 17:53:53,072 - Epoch: [80][ 660/ 1207] Overall Loss 0.329407 Objective Loss 0.329407 LR 0.001000 Time 0.020543 -2023-02-13 17:53:53,265 - Epoch: [80][ 670/ 1207] Overall Loss 0.329535 Objective Loss 0.329535 LR 0.001000 Time 0.020523 -2023-02-13 17:53:53,462 - Epoch: [80][ 680/ 1207] Overall Loss 0.329361 Objective Loss 0.329361 LR 0.001000 Time 0.020510 -2023-02-13 17:53:53,654 - Epoch: [80][ 690/ 1207] Overall Loss 0.329086 Objective Loss 0.329086 LR 0.001000 Time 0.020491 -2023-02-13 17:53:53,851 - Epoch: [80][ 700/ 1207] Overall Loss 0.329247 Objective Loss 0.329247 LR 0.001000 Time 0.020478 -2023-02-13 17:53:54,043 - Epoch: [80][ 710/ 1207] Overall Loss 0.329150 Objective Loss 0.329150 LR 0.001000 Time 0.020461 -2023-02-13 17:53:54,239 - Epoch: [80][ 720/ 1207] Overall Loss 0.329258 Objective Loss 0.329258 LR 0.001000 Time 0.020448 -2023-02-13 17:53:54,432 - Epoch: [80][ 730/ 1207] Overall Loss 0.329104 Objective Loss 0.329104 LR 0.001000 Time 0.020432 -2023-02-13 17:53:54,628 - Epoch: [80][ 740/ 1207] Overall Loss 0.329167 Objective Loss 0.329167 LR 0.001000 Time 0.020420 -2023-02-13 17:53:54,820 - Epoch: [80][ 750/ 1207] Overall Loss 0.329221 Objective Loss 0.329221 LR 0.001000 Time 0.020403 -2023-02-13 17:53:55,016 - Epoch: [80][ 760/ 1207] Overall Loss 0.329140 Objective Loss 0.329140 LR 0.001000 Time 0.020391 -2023-02-13 17:53:55,209 - Epoch: [80][ 770/ 1207] Overall Loss 0.328799 Objective Loss 0.328799 LR 0.001000 Time 0.020377 -2023-02-13 17:53:55,405 - Epoch: [80][ 780/ 1207] Overall Loss 0.328530 Objective Loss 0.328530 LR 0.001000 Time 0.020367 -2023-02-13 17:53:55,598 - Epoch: [80][ 790/ 1207] Overall Loss 0.328378 Objective Loss 0.328378 LR 0.001000 Time 0.020353 -2023-02-13 17:53:55,794 - Epoch: [80][ 800/ 1207] Overall Loss 0.328219 Objective Loss 0.328219 LR 0.001000 Time 0.020343 -2023-02-13 17:53:55,987 - Epoch: [80][ 810/ 1207] Overall Loss 0.328187 Objective Loss 0.328187 LR 0.001000 Time 0.020330 -2023-02-13 17:53:56,183 - Epoch: [80][ 820/ 1207] Overall Loss 0.328346 Objective Loss 0.328346 LR 0.001000 Time 0.020320 -2023-02-13 17:53:56,376 - Epoch: [80][ 830/ 1207] Overall Loss 0.328415 Objective Loss 0.328415 LR 0.001000 Time 0.020308 -2023-02-13 17:53:56,572 - Epoch: [80][ 840/ 1207] Overall Loss 0.328885 Objective Loss 0.328885 LR 0.001000 Time 0.020299 -2023-02-13 17:53:56,765 - Epoch: [80][ 850/ 1207] Overall Loss 0.328782 Objective Loss 0.328782 LR 0.001000 Time 0.020286 -2023-02-13 17:53:56,961 - Epoch: [80][ 860/ 1207] Overall Loss 0.328201 Objective Loss 0.328201 LR 0.001000 Time 0.020278 -2023-02-13 17:53:57,154 - Epoch: [80][ 870/ 1207] Overall Loss 0.328745 Objective Loss 0.328745 LR 0.001000 Time 0.020267 -2023-02-13 17:53:57,351 - Epoch: [80][ 880/ 1207] Overall Loss 0.329018 Objective Loss 0.329018 LR 0.001000 Time 0.020259 -2023-02-13 17:53:57,544 - Epoch: [80][ 890/ 1207] Overall Loss 0.329261 Objective Loss 0.329261 LR 0.001000 Time 0.020248 -2023-02-13 17:53:57,740 - Epoch: [80][ 900/ 1207] Overall Loss 0.329552 Objective Loss 0.329552 LR 0.001000 Time 0.020240 -2023-02-13 17:53:57,932 - Epoch: [80][ 910/ 1207] Overall Loss 0.329493 Objective Loss 0.329493 LR 0.001000 Time 0.020229 -2023-02-13 17:53:58,128 - Epoch: [80][ 920/ 1207] Overall Loss 0.329973 Objective Loss 0.329973 LR 0.001000 Time 0.020222 -2023-02-13 17:53:58,321 - Epoch: [80][ 930/ 1207] Overall Loss 0.330118 Objective Loss 0.330118 LR 0.001000 Time 0.020212 -2023-02-13 17:53:58,519 - Epoch: [80][ 940/ 1207] Overall Loss 0.330596 Objective Loss 0.330596 LR 0.001000 Time 0.020207 -2023-02-13 17:53:58,712 - Epoch: [80][ 950/ 1207] Overall Loss 0.330863 Objective Loss 0.330863 LR 0.001000 Time 0.020196 -2023-02-13 17:53:58,907 - Epoch: [80][ 960/ 1207] Overall Loss 0.331173 Objective Loss 0.331173 LR 0.001000 Time 0.020189 -2023-02-13 17:53:59,100 - Epoch: [80][ 970/ 1207] Overall Loss 0.331402 Objective Loss 0.331402 LR 0.001000 Time 0.020179 -2023-02-13 17:53:59,295 - Epoch: [80][ 980/ 1207] Overall Loss 0.331178 Objective Loss 0.331178 LR 0.001000 Time 0.020172 -2023-02-13 17:53:59,488 - Epoch: [80][ 990/ 1207] Overall Loss 0.331036 Objective Loss 0.331036 LR 0.001000 Time 0.020163 -2023-02-13 17:53:59,684 - Epoch: [80][ 1000/ 1207] Overall Loss 0.330865 Objective Loss 0.330865 LR 0.001000 Time 0.020157 -2023-02-13 17:53:59,877 - Epoch: [80][ 1010/ 1207] Overall Loss 0.330897 Objective Loss 0.330897 LR 0.001000 Time 0.020148 -2023-02-13 17:54:00,073 - Epoch: [80][ 1020/ 1207] Overall Loss 0.331234 Objective Loss 0.331234 LR 0.001000 Time 0.020142 -2023-02-13 17:54:00,266 - Epoch: [80][ 1030/ 1207] Overall Loss 0.330977 Objective Loss 0.330977 LR 0.001000 Time 0.020134 -2023-02-13 17:54:00,462 - Epoch: [80][ 1040/ 1207] Overall Loss 0.330768 Objective Loss 0.330768 LR 0.001000 Time 0.020129 -2023-02-13 17:54:00,655 - Epoch: [80][ 1050/ 1207] Overall Loss 0.330375 Objective Loss 0.330375 LR 0.001000 Time 0.020120 -2023-02-13 17:54:00,851 - Epoch: [80][ 1060/ 1207] Overall Loss 0.330056 Objective Loss 0.330056 LR 0.001000 Time 0.020115 -2023-02-13 17:54:01,045 - Epoch: [80][ 1070/ 1207] Overall Loss 0.330300 Objective Loss 0.330300 LR 0.001000 Time 0.020108 -2023-02-13 17:54:01,241 - Epoch: [80][ 1080/ 1207] Overall Loss 0.330425 Objective Loss 0.330425 LR 0.001000 Time 0.020103 -2023-02-13 17:54:01,435 - Epoch: [80][ 1090/ 1207] Overall Loss 0.330904 Objective Loss 0.330904 LR 0.001000 Time 0.020096 -2023-02-13 17:54:01,631 - Epoch: [80][ 1100/ 1207] Overall Loss 0.330760 Objective Loss 0.330760 LR 0.001000 Time 0.020091 -2023-02-13 17:54:01,825 - Epoch: [80][ 1110/ 1207] Overall Loss 0.330976 Objective Loss 0.330976 LR 0.001000 Time 0.020084 -2023-02-13 17:54:02,020 - Epoch: [80][ 1120/ 1207] Overall Loss 0.330741 Objective Loss 0.330741 LR 0.001000 Time 0.020079 -2023-02-13 17:54:02,213 - Epoch: [80][ 1130/ 1207] Overall Loss 0.330666 Objective Loss 0.330666 LR 0.001000 Time 0.020072 -2023-02-13 17:54:02,410 - Epoch: [80][ 1140/ 1207] Overall Loss 0.330342 Objective Loss 0.330342 LR 0.001000 Time 0.020068 -2023-02-13 17:54:02,604 - Epoch: [80][ 1150/ 1207] Overall Loss 0.330461 Objective Loss 0.330461 LR 0.001000 Time 0.020062 -2023-02-13 17:54:02,800 - Epoch: [80][ 1160/ 1207] Overall Loss 0.330468 Objective Loss 0.330468 LR 0.001000 Time 0.020058 -2023-02-13 17:54:02,993 - Epoch: [80][ 1170/ 1207] Overall Loss 0.330259 Objective Loss 0.330259 LR 0.001000 Time 0.020051 -2023-02-13 17:54:03,190 - Epoch: [80][ 1180/ 1207] Overall Loss 0.330499 Objective Loss 0.330499 LR 0.001000 Time 0.020048 -2023-02-13 17:54:03,383 - Epoch: [80][ 1190/ 1207] Overall Loss 0.330508 Objective Loss 0.330508 LR 0.001000 Time 0.020041 -2023-02-13 17:54:03,632 - Epoch: [80][ 1200/ 1207] Overall Loss 0.330492 Objective Loss 0.330492 LR 0.001000 Time 0.020082 -2023-02-13 17:54:03,747 - Epoch: [80][ 1207/ 1207] Overall Loss 0.330359 Objective Loss 0.330359 Top1 88.719512 Top5 98.780488 LR 0.001000 Time 0.020060 -2023-02-13 17:54:03,820 - --- validate (epoch=80)----------- -2023-02-13 17:54:03,820 - 34311 samples (256 per mini-batch) -2023-02-13 17:54:04,237 - Epoch: [80][ 10/ 135] Loss 0.383007 Top1 82.148438 Top5 96.992188 -2023-02-13 17:54:04,372 - Epoch: [80][ 20/ 135] Loss 0.401894 Top1 81.074219 Top5 97.089844 -2023-02-13 17:54:04,500 - Epoch: [80][ 30/ 135] Loss 0.385458 Top1 81.250000 Top5 97.187500 -2023-02-13 17:54:04,640 - Epoch: [80][ 40/ 135] Loss 0.385786 Top1 81.162109 Top5 97.304688 -2023-02-13 17:54:04,776 - Epoch: [80][ 50/ 135] Loss 0.376395 Top1 81.414062 Top5 97.359375 -2023-02-13 17:54:04,919 - Epoch: [80][ 60/ 135] Loss 0.374436 Top1 81.738281 Top5 97.369792 -2023-02-13 17:54:05,056 - Epoch: [80][ 70/ 135] Loss 0.368712 Top1 81.858259 Top5 97.427455 -2023-02-13 17:54:05,188 - Epoch: [80][ 80/ 135] Loss 0.365132 Top1 81.870117 Top5 97.421875 -2023-02-13 17:54:05,314 - Epoch: [80][ 90/ 135] Loss 0.357486 Top1 81.809896 Top5 97.443576 -2023-02-13 17:54:05,442 - Epoch: [80][ 100/ 135] Loss 0.360117 Top1 81.761719 Top5 97.425781 -2023-02-13 17:54:05,569 - Epoch: [80][ 110/ 135] Loss 0.364716 Top1 81.697443 Top5 97.425426 -2023-02-13 17:54:05,697 - Epoch: [80][ 120/ 135] Loss 0.361437 Top1 81.735026 Top5 97.418620 -2023-02-13 17:54:05,829 - Epoch: [80][ 130/ 135] Loss 0.363188 Top1 81.622596 Top5 97.391827 -2023-02-13 17:54:05,878 - Epoch: [80][ 135/ 135] Loss 0.361125 Top1 81.673516 Top5 97.379849 -2023-02-13 17:54:05,949 - ==> Top1: 81.674 Top5: 97.380 Loss: 0.361 - -2023-02-13 17:54:05,949 - ==> Confusion: -[[ 831 2 7 2 15 3 0 2 2 65 0 3 1 3 7 4 3 3 1 2 11] - [ 2 940 1 2 10 25 6 15 4 2 1 1 2 1 3 2 4 0 6 1 5] - [ 9 4 930 17 3 1 29 16 0 3 2 1 5 5 4 7 0 4 4 8 6] - [ 6 1 20 886 4 5 2 2 2 2 11 0 14 2 24 4 7 4 13 1 6] - [ 12 7 0 0 985 9 1 2 1 3 0 9 0 5 9 8 5 2 0 2 6] - [ 4 23 2 3 4 957 2 14 0 4 4 8 5 11 3 4 5 2 0 10 5] - [ 4 6 15 1 0 5 1032 3 0 3 4 0 3 1 0 5 2 1 2 7 5] - [ 2 15 14 4 3 45 10 874 1 0 2 7 5 1 0 1 1 2 21 12 4] - [ 22 1 1 0 2 0 0 1 858 53 9 3 2 16 31 4 0 3 1 0 2] - [ 70 0 4 0 8 4 0 1 30 860 0 2 1 16 3 1 0 4 2 2 4] - [ 2 3 3 8 1 2 8 4 20 2 954 1 1 17 4 1 1 2 12 1 4] - [ 4 2 0 0 2 12 1 4 1 1 0 888 44 12 0 9 3 7 1 11 3] - [ 1 0 0 8 2 2 2 1 2 0 0 23 867 1 5 9 1 24 1 0 10] - [ 12 3 1 0 5 8 3 2 7 14 9 5 2 922 7 7 5 5 0 2 5] - [ 13 1 3 15 7 2 1 4 15 9 4 1 4 2 983 0 1 8 8 1 10] - [ 4 1 5 0 5 1 7 1 0 2 0 6 5 2 0 973 6 17 1 5 5] - [ 4 6 0 1 12 4 1 0 2 1 0 1 4 4 0 21 984 3 0 2 11] - [ 4 0 2 2 1 2 3 0 1 1 0 11 24 1 1 19 0 972 0 2 5] - [ 2 4 7 18 3 2 0 25 5 1 3 1 8 2 21 4 0 2 971 3 4] - [ 2 3 1 0 2 9 12 9 1 0 1 19 6 7 0 5 9 6 3 1044 9] - [ 179 288 228 148 181 250 137 165 94 123 212 125 406 329 219 144 305 130 170 289 9312]] - -2023-02-13 17:54:05,951 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:54:05,951 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:54:05,956 - - -2023-02-13 17:54:05,957 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:54:06,854 - Epoch: [81][ 10/ 1207] Overall Loss 0.319572 Objective Loss 0.319572 LR 0.001000 Time 0.089697 -2023-02-13 17:54:07,058 - Epoch: [81][ 20/ 1207] Overall Loss 0.317592 Objective Loss 0.317592 LR 0.001000 Time 0.055017 -2023-02-13 17:54:07,253 - Epoch: [81][ 30/ 1207] Overall Loss 0.332500 Objective Loss 0.332500 LR 0.001000 Time 0.043150 -2023-02-13 17:54:07,447 - Epoch: [81][ 40/ 1207] Overall Loss 0.327433 Objective Loss 0.327433 LR 0.001000 Time 0.037215 -2023-02-13 17:54:07,642 - Epoch: [81][ 50/ 1207] Overall Loss 0.327045 Objective Loss 0.327045 LR 0.001000 Time 0.033672 -2023-02-13 17:54:07,837 - Epoch: [81][ 60/ 1207] Overall Loss 0.326457 Objective Loss 0.326457 LR 0.001000 Time 0.031289 -2023-02-13 17:54:08,032 - Epoch: [81][ 70/ 1207] Overall Loss 0.328317 Objective Loss 0.328317 LR 0.001000 Time 0.029613 -2023-02-13 17:54:08,226 - Epoch: [81][ 80/ 1207] Overall Loss 0.326596 Objective Loss 0.326596 LR 0.001000 Time 0.028324 -2023-02-13 17:54:08,420 - Epoch: [81][ 90/ 1207] Overall Loss 0.328193 Objective Loss 0.328193 LR 0.001000 Time 0.027328 -2023-02-13 17:54:08,613 - Epoch: [81][ 100/ 1207] Overall Loss 0.329877 Objective Loss 0.329877 LR 0.001000 Time 0.026528 -2023-02-13 17:54:08,808 - Epoch: [81][ 110/ 1207] Overall Loss 0.327374 Objective Loss 0.327374 LR 0.001000 Time 0.025880 -2023-02-13 17:54:09,001 - Epoch: [81][ 120/ 1207] Overall Loss 0.329148 Objective Loss 0.329148 LR 0.001000 Time 0.025326 -2023-02-13 17:54:09,196 - Epoch: [81][ 130/ 1207] Overall Loss 0.326378 Objective Loss 0.326378 LR 0.001000 Time 0.024877 -2023-02-13 17:54:09,389 - Epoch: [81][ 140/ 1207] Overall Loss 0.329258 Objective Loss 0.329258 LR 0.001000 Time 0.024478 -2023-02-13 17:54:09,585 - Epoch: [81][ 150/ 1207] Overall Loss 0.330477 Objective Loss 0.330477 LR 0.001000 Time 0.024150 -2023-02-13 17:54:09,778 - Epoch: [81][ 160/ 1207] Overall Loss 0.329156 Objective Loss 0.329156 LR 0.001000 Time 0.023846 -2023-02-13 17:54:09,973 - Epoch: [81][ 170/ 1207] Overall Loss 0.328238 Objective Loss 0.328238 LR 0.001000 Time 0.023588 -2023-02-13 17:54:10,167 - Epoch: [81][ 180/ 1207] Overall Loss 0.328091 Objective Loss 0.328091 LR 0.001000 Time 0.023351 -2023-02-13 17:54:10,362 - Epoch: [81][ 190/ 1207] Overall Loss 0.327528 Objective Loss 0.327528 LR 0.001000 Time 0.023147 -2023-02-13 17:54:10,556 - Epoch: [81][ 200/ 1207] Overall Loss 0.326316 Objective Loss 0.326316 LR 0.001000 Time 0.022959 -2023-02-13 17:54:10,751 - Epoch: [81][ 210/ 1207] Overall Loss 0.327462 Objective Loss 0.327462 LR 0.001000 Time 0.022794 -2023-02-13 17:54:10,945 - Epoch: [81][ 220/ 1207] Overall Loss 0.326846 Objective Loss 0.326846 LR 0.001000 Time 0.022638 -2023-02-13 17:54:11,141 - Epoch: [81][ 230/ 1207] Overall Loss 0.327053 Objective Loss 0.327053 LR 0.001000 Time 0.022502 -2023-02-13 17:54:11,335 - Epoch: [81][ 240/ 1207] Overall Loss 0.325828 Objective Loss 0.325828 LR 0.001000 Time 0.022372 -2023-02-13 17:54:11,531 - Epoch: [81][ 250/ 1207] Overall Loss 0.325687 Objective Loss 0.325687 LR 0.001000 Time 0.022257 -2023-02-13 17:54:11,724 - Epoch: [81][ 260/ 1207] Overall Loss 0.326031 Objective Loss 0.326031 LR 0.001000 Time 0.022144 -2023-02-13 17:54:11,920 - Epoch: [81][ 270/ 1207] Overall Loss 0.325407 Objective Loss 0.325407 LR 0.001000 Time 0.022046 -2023-02-13 17:54:12,114 - Epoch: [81][ 280/ 1207] Overall Loss 0.324943 Objective Loss 0.324943 LR 0.001000 Time 0.021950 -2023-02-13 17:54:12,309 - Epoch: [81][ 290/ 1207] Overall Loss 0.325597 Objective Loss 0.325597 LR 0.001000 Time 0.021865 -2023-02-13 17:54:12,503 - Epoch: [81][ 300/ 1207] Overall Loss 0.326942 Objective Loss 0.326942 LR 0.001000 Time 0.021783 -2023-02-13 17:54:12,699 - Epoch: [81][ 310/ 1207] Overall Loss 0.327042 Objective Loss 0.327042 LR 0.001000 Time 0.021712 -2023-02-13 17:54:12,893 - Epoch: [81][ 320/ 1207] Overall Loss 0.326972 Objective Loss 0.326972 LR 0.001000 Time 0.021639 -2023-02-13 17:54:13,089 - Epoch: [81][ 330/ 1207] Overall Loss 0.326440 Objective Loss 0.326440 LR 0.001000 Time 0.021576 -2023-02-13 17:54:13,283 - Epoch: [81][ 340/ 1207] Overall Loss 0.326181 Objective Loss 0.326181 LR 0.001000 Time 0.021509 -2023-02-13 17:54:13,479 - Epoch: [81][ 350/ 1207] Overall Loss 0.326028 Objective Loss 0.326028 LR 0.001000 Time 0.021454 -2023-02-13 17:54:13,672 - Epoch: [81][ 360/ 1207] Overall Loss 0.325179 Objective Loss 0.325179 LR 0.001000 Time 0.021394 -2023-02-13 17:54:13,867 - Epoch: [81][ 370/ 1207] Overall Loss 0.325013 Objective Loss 0.325013 LR 0.001000 Time 0.021342 -2023-02-13 17:54:14,061 - Epoch: [81][ 380/ 1207] Overall Loss 0.325063 Objective Loss 0.325063 LR 0.001000 Time 0.021289 -2023-02-13 17:54:14,256 - Epoch: [81][ 390/ 1207] Overall Loss 0.324792 Objective Loss 0.324792 LR 0.001000 Time 0.021243 -2023-02-13 17:54:14,450 - Epoch: [81][ 400/ 1207] Overall Loss 0.324848 Objective Loss 0.324848 LR 0.001000 Time 0.021195 -2023-02-13 17:54:14,646 - Epoch: [81][ 410/ 1207] Overall Loss 0.324497 Objective Loss 0.324497 LR 0.001000 Time 0.021155 -2023-02-13 17:54:14,840 - Epoch: [81][ 420/ 1207] Overall Loss 0.324985 Objective Loss 0.324985 LR 0.001000 Time 0.021112 -2023-02-13 17:54:15,035 - Epoch: [81][ 430/ 1207] Overall Loss 0.324625 Objective Loss 0.324625 LR 0.001000 Time 0.021074 -2023-02-13 17:54:15,229 - Epoch: [81][ 440/ 1207] Overall Loss 0.325781 Objective Loss 0.325781 LR 0.001000 Time 0.021034 -2023-02-13 17:54:15,425 - Epoch: [81][ 450/ 1207] Overall Loss 0.326004 Objective Loss 0.326004 LR 0.001000 Time 0.021002 -2023-02-13 17:54:15,618 - Epoch: [81][ 460/ 1207] Overall Loss 0.326008 Objective Loss 0.326008 LR 0.001000 Time 0.020965 -2023-02-13 17:54:15,814 - Epoch: [81][ 470/ 1207] Overall Loss 0.325700 Objective Loss 0.325700 LR 0.001000 Time 0.020935 -2023-02-13 17:54:16,009 - Epoch: [81][ 480/ 1207] Overall Loss 0.325407 Objective Loss 0.325407 LR 0.001000 Time 0.020905 -2023-02-13 17:54:16,205 - Epoch: [81][ 490/ 1207] Overall Loss 0.324903 Objective Loss 0.324903 LR 0.001000 Time 0.020877 -2023-02-13 17:54:16,400 - Epoch: [81][ 500/ 1207] Overall Loss 0.325022 Objective Loss 0.325022 LR 0.001000 Time 0.020848 -2023-02-13 17:54:16,595 - Epoch: [81][ 510/ 1207] Overall Loss 0.325055 Objective Loss 0.325055 LR 0.001000 Time 0.020822 -2023-02-13 17:54:16,790 - Epoch: [81][ 520/ 1207] Overall Loss 0.324666 Objective Loss 0.324666 LR 0.001000 Time 0.020795 -2023-02-13 17:54:16,986 - Epoch: [81][ 530/ 1207] Overall Loss 0.324443 Objective Loss 0.324443 LR 0.001000 Time 0.020772 -2023-02-13 17:54:17,181 - Epoch: [81][ 540/ 1207] Overall Loss 0.324599 Objective Loss 0.324599 LR 0.001000 Time 0.020747 -2023-02-13 17:54:17,376 - Epoch: [81][ 550/ 1207] Overall Loss 0.324165 Objective Loss 0.324165 LR 0.001000 Time 0.020724 -2023-02-13 17:54:17,570 - Epoch: [81][ 560/ 1207] Overall Loss 0.324057 Objective Loss 0.324057 LR 0.001000 Time 0.020700 -2023-02-13 17:54:17,766 - Epoch: [81][ 570/ 1207] Overall Loss 0.323418 Objective Loss 0.323418 LR 0.001000 Time 0.020680 -2023-02-13 17:54:17,959 - Epoch: [81][ 580/ 1207] Overall Loss 0.323614 Objective Loss 0.323614 LR 0.001000 Time 0.020656 -2023-02-13 17:54:18,155 - Epoch: [81][ 590/ 1207] Overall Loss 0.323334 Objective Loss 0.323334 LR 0.001000 Time 0.020637 -2023-02-13 17:54:18,350 - Epoch: [81][ 600/ 1207] Overall Loss 0.323488 Objective Loss 0.323488 LR 0.001000 Time 0.020617 -2023-02-13 17:54:18,547 - Epoch: [81][ 610/ 1207] Overall Loss 0.323152 Objective Loss 0.323152 LR 0.001000 Time 0.020602 -2023-02-13 17:54:18,744 - Epoch: [81][ 620/ 1207] Overall Loss 0.323212 Objective Loss 0.323212 LR 0.001000 Time 0.020587 -2023-02-13 17:54:18,942 - Epoch: [81][ 630/ 1207] Overall Loss 0.323652 Objective Loss 0.323652 LR 0.001000 Time 0.020574 -2023-02-13 17:54:19,140 - Epoch: [81][ 640/ 1207] Overall Loss 0.324054 Objective Loss 0.324054 LR 0.001000 Time 0.020561 -2023-02-13 17:54:19,337 - Epoch: [81][ 650/ 1207] Overall Loss 0.324400 Objective Loss 0.324400 LR 0.001000 Time 0.020547 -2023-02-13 17:54:19,535 - Epoch: [81][ 660/ 1207] Overall Loss 0.324486 Objective Loss 0.324486 LR 0.001000 Time 0.020535 -2023-02-13 17:54:19,733 - Epoch: [81][ 670/ 1207] Overall Loss 0.324719 Objective Loss 0.324719 LR 0.001000 Time 0.020523 -2023-02-13 17:54:19,929 - Epoch: [81][ 680/ 1207] Overall Loss 0.324524 Objective Loss 0.324524 LR 0.001000 Time 0.020510 -2023-02-13 17:54:20,127 - Epoch: [81][ 690/ 1207] Overall Loss 0.324522 Objective Loss 0.324522 LR 0.001000 Time 0.020499 -2023-02-13 17:54:20,324 - Epoch: [81][ 700/ 1207] Overall Loss 0.324669 Objective Loss 0.324669 LR 0.001000 Time 0.020487 -2023-02-13 17:54:20,523 - Epoch: [81][ 710/ 1207] Overall Loss 0.325181 Objective Loss 0.325181 LR 0.001000 Time 0.020478 -2023-02-13 17:54:20,720 - Epoch: [81][ 720/ 1207] Overall Loss 0.325155 Objective Loss 0.325155 LR 0.001000 Time 0.020466 -2023-02-13 17:54:20,918 - Epoch: [81][ 730/ 1207] Overall Loss 0.325089 Objective Loss 0.325089 LR 0.001000 Time 0.020457 -2023-02-13 17:54:21,115 - Epoch: [81][ 740/ 1207] Overall Loss 0.324941 Objective Loss 0.324941 LR 0.001000 Time 0.020447 -2023-02-13 17:54:21,313 - Epoch: [81][ 750/ 1207] Overall Loss 0.324628 Objective Loss 0.324628 LR 0.001000 Time 0.020437 -2023-02-13 17:54:21,510 - Epoch: [81][ 760/ 1207] Overall Loss 0.324481 Objective Loss 0.324481 LR 0.001000 Time 0.020427 -2023-02-13 17:54:21,707 - Epoch: [81][ 770/ 1207] Overall Loss 0.325008 Objective Loss 0.325008 LR 0.001000 Time 0.020418 -2023-02-13 17:54:21,905 - Epoch: [81][ 780/ 1207] Overall Loss 0.324944 Objective Loss 0.324944 LR 0.001000 Time 0.020408 -2023-02-13 17:54:22,103 - Epoch: [81][ 790/ 1207] Overall Loss 0.325154 Objective Loss 0.325154 LR 0.001000 Time 0.020400 -2023-02-13 17:54:22,300 - Epoch: [81][ 800/ 1207] Overall Loss 0.324939 Objective Loss 0.324939 LR 0.001000 Time 0.020392 -2023-02-13 17:54:22,499 - Epoch: [81][ 810/ 1207] Overall Loss 0.325031 Objective Loss 0.325031 LR 0.001000 Time 0.020384 -2023-02-13 17:54:22,696 - Epoch: [81][ 820/ 1207] Overall Loss 0.325176 Objective Loss 0.325176 LR 0.001000 Time 0.020375 -2023-02-13 17:54:22,893 - Epoch: [81][ 830/ 1207] Overall Loss 0.325424 Objective Loss 0.325424 LR 0.001000 Time 0.020368 -2023-02-13 17:54:23,091 - Epoch: [81][ 840/ 1207] Overall Loss 0.325441 Objective Loss 0.325441 LR 0.001000 Time 0.020360 -2023-02-13 17:54:23,288 - Epoch: [81][ 850/ 1207] Overall Loss 0.325417 Objective Loss 0.325417 LR 0.001000 Time 0.020352 -2023-02-13 17:54:23,486 - Epoch: [81][ 860/ 1207] Overall Loss 0.325471 Objective Loss 0.325471 LR 0.001000 Time 0.020345 -2023-02-13 17:54:23,684 - Epoch: [81][ 870/ 1207] Overall Loss 0.325495 Objective Loss 0.325495 LR 0.001000 Time 0.020338 -2023-02-13 17:54:23,881 - Epoch: [81][ 880/ 1207] Overall Loss 0.325655 Objective Loss 0.325655 LR 0.001000 Time 0.020330 -2023-02-13 17:54:24,079 - Epoch: [81][ 890/ 1207] Overall Loss 0.325714 Objective Loss 0.325714 LR 0.001000 Time 0.020324 -2023-02-13 17:54:24,275 - Epoch: [81][ 900/ 1207] Overall Loss 0.325581 Objective Loss 0.325581 LR 0.001000 Time 0.020316 -2023-02-13 17:54:24,473 - Epoch: [81][ 910/ 1207] Overall Loss 0.325686 Objective Loss 0.325686 LR 0.001000 Time 0.020310 -2023-02-13 17:54:24,670 - Epoch: [81][ 920/ 1207] Overall Loss 0.325750 Objective Loss 0.325750 LR 0.001000 Time 0.020303 -2023-02-13 17:54:24,867 - Epoch: [81][ 930/ 1207] Overall Loss 0.325928 Objective Loss 0.325928 LR 0.001000 Time 0.020296 -2023-02-13 17:54:25,064 - Epoch: [81][ 940/ 1207] Overall Loss 0.325750 Objective Loss 0.325750 LR 0.001000 Time 0.020289 -2023-02-13 17:54:25,262 - Epoch: [81][ 950/ 1207] Overall Loss 0.325681 Objective Loss 0.325681 LR 0.001000 Time 0.020283 -2023-02-13 17:54:25,459 - Epoch: [81][ 960/ 1207] Overall Loss 0.326518 Objective Loss 0.326518 LR 0.001000 Time 0.020277 -2023-02-13 17:54:25,656 - Epoch: [81][ 970/ 1207] Overall Loss 0.326972 Objective Loss 0.326972 LR 0.001000 Time 0.020271 -2023-02-13 17:54:25,852 - Epoch: [81][ 980/ 1207] Overall Loss 0.327139 Objective Loss 0.327139 LR 0.001000 Time 0.020263 -2023-02-13 17:54:26,048 - Epoch: [81][ 990/ 1207] Overall Loss 0.327433 Objective Loss 0.327433 LR 0.001000 Time 0.020256 -2023-02-13 17:54:26,244 - Epoch: [81][ 1000/ 1207] Overall Loss 0.327789 Objective Loss 0.327789 LR 0.001000 Time 0.020249 -2023-02-13 17:54:26,440 - Epoch: [81][ 1010/ 1207] Overall Loss 0.328241 Objective Loss 0.328241 LR 0.001000 Time 0.020243 -2023-02-13 17:54:26,636 - Epoch: [81][ 1020/ 1207] Overall Loss 0.328135 Objective Loss 0.328135 LR 0.001000 Time 0.020236 -2023-02-13 17:54:26,833 - Epoch: [81][ 1030/ 1207] Overall Loss 0.328468 Objective Loss 0.328468 LR 0.001000 Time 0.020230 -2023-02-13 17:54:27,028 - Epoch: [81][ 1040/ 1207] Overall Loss 0.328355 Objective Loss 0.328355 LR 0.001000 Time 0.020223 -2023-02-13 17:54:27,224 - Epoch: [81][ 1050/ 1207] Overall Loss 0.328143 Objective Loss 0.328143 LR 0.001000 Time 0.020217 -2023-02-13 17:54:27,420 - Epoch: [81][ 1060/ 1207] Overall Loss 0.328081 Objective Loss 0.328081 LR 0.001000 Time 0.020210 -2023-02-13 17:54:27,617 - Epoch: [81][ 1070/ 1207] Overall Loss 0.328545 Objective Loss 0.328545 LR 0.001000 Time 0.020205 -2023-02-13 17:54:27,812 - Epoch: [81][ 1080/ 1207] Overall Loss 0.328812 Objective Loss 0.328812 LR 0.001000 Time 0.020198 -2023-02-13 17:54:28,008 - Epoch: [81][ 1090/ 1207] Overall Loss 0.329020 Objective Loss 0.329020 LR 0.001000 Time 0.020193 -2023-02-13 17:54:28,203 - Epoch: [81][ 1100/ 1207] Overall Loss 0.329119 Objective Loss 0.329119 LR 0.001000 Time 0.020186 -2023-02-13 17:54:28,399 - Epoch: [81][ 1110/ 1207] Overall Loss 0.329190 Objective Loss 0.329190 LR 0.001000 Time 0.020180 -2023-02-13 17:54:28,595 - Epoch: [81][ 1120/ 1207] Overall Loss 0.329229 Objective Loss 0.329229 LR 0.001000 Time 0.020174 -2023-02-13 17:54:28,790 - Epoch: [81][ 1130/ 1207] Overall Loss 0.329185 Objective Loss 0.329185 LR 0.001000 Time 0.020169 -2023-02-13 17:54:28,985 - Epoch: [81][ 1140/ 1207] Overall Loss 0.329377 Objective Loss 0.329377 LR 0.001000 Time 0.020163 -2023-02-13 17:54:29,182 - Epoch: [81][ 1150/ 1207] Overall Loss 0.329583 Objective Loss 0.329583 LR 0.001000 Time 0.020158 -2023-02-13 17:54:29,377 - Epoch: [81][ 1160/ 1207] Overall Loss 0.329833 Objective Loss 0.329833 LR 0.001000 Time 0.020152 -2023-02-13 17:54:29,574 - Epoch: [81][ 1170/ 1207] Overall Loss 0.330214 Objective Loss 0.330214 LR 0.001000 Time 0.020148 -2023-02-13 17:54:29,769 - Epoch: [81][ 1180/ 1207] Overall Loss 0.330374 Objective Loss 0.330374 LR 0.001000 Time 0.020142 -2023-02-13 17:54:29,964 - Epoch: [81][ 1190/ 1207] Overall Loss 0.330348 Objective Loss 0.330348 LR 0.001000 Time 0.020137 -2023-02-13 17:54:30,211 - Epoch: [81][ 1200/ 1207] Overall Loss 0.330655 Objective Loss 0.330655 LR 0.001000 Time 0.020174 -2023-02-13 17:54:30,325 - Epoch: [81][ 1207/ 1207] Overall Loss 0.330678 Objective Loss 0.330678 Top1 82.012195 Top5 97.865854 LR 0.001000 Time 0.020151 -2023-02-13 17:54:30,402 - --- validate (epoch=81)----------- -2023-02-13 17:54:30,403 - 34311 samples (256 per mini-batch) -2023-02-13 17:54:30,816 - Epoch: [81][ 10/ 135] Loss 0.393309 Top1 81.445312 Top5 97.070312 -2023-02-13 17:54:30,947 - Epoch: [81][ 20/ 135] Loss 0.381006 Top1 81.464844 Top5 96.953125 -2023-02-13 17:54:31,091 - Epoch: [81][ 30/ 135] Loss 0.380080 Top1 81.236979 Top5 97.018229 -2023-02-13 17:54:31,247 - Epoch: [81][ 40/ 135] Loss 0.375918 Top1 81.796875 Top5 97.099609 -2023-02-13 17:54:31,389 - Epoch: [81][ 50/ 135] Loss 0.375391 Top1 81.710938 Top5 97.078125 -2023-02-13 17:54:31,522 - Epoch: [81][ 60/ 135] Loss 0.375162 Top1 81.855469 Top5 96.972656 -2023-02-13 17:54:31,651 - Epoch: [81][ 70/ 135] Loss 0.379000 Top1 81.774554 Top5 96.958705 -2023-02-13 17:54:31,775 - Epoch: [81][ 80/ 135] Loss 0.374898 Top1 81.899414 Top5 97.016602 -2023-02-13 17:54:31,899 - Epoch: [81][ 90/ 135] Loss 0.372794 Top1 82.131076 Top5 97.100694 -2023-02-13 17:54:32,028 - Epoch: [81][ 100/ 135] Loss 0.369251 Top1 82.156250 Top5 97.160156 -2023-02-13 17:54:32,155 - Epoch: [81][ 110/ 135] Loss 0.370524 Top1 82.169744 Top5 97.198153 -2023-02-13 17:54:32,282 - Epoch: [81][ 120/ 135] Loss 0.370266 Top1 82.106120 Top5 97.161458 -2023-02-13 17:54:32,410 - Epoch: [81][ 130/ 135] Loss 0.368454 Top1 82.199519 Top5 97.196514 -2023-02-13 17:54:32,456 - Epoch: [81][ 135/ 135] Loss 0.373198 Top1 82.212701 Top5 97.184576 -2023-02-13 17:54:32,525 - ==> Top1: 82.213 Top5: 97.185 Loss: 0.373 - -2023-02-13 17:54:32,525 - ==> Confusion: -[[ 832 5 5 0 12 4 1 1 4 62 1 5 2 5 8 2 4 2 1 3 8] - [ 4 930 0 3 6 30 6 17 3 0 3 3 1 1 2 5 3 1 3 4 8] - [ 8 1 922 18 2 3 29 16 0 4 3 3 2 4 4 8 5 4 8 5 9] - [ 6 2 19 869 2 9 2 2 3 1 15 0 16 1 14 0 6 12 25 0 12] - [ 19 18 1 0 960 11 1 2 2 4 1 5 1 5 4 6 10 2 2 6 6] - [ 3 16 0 3 6 955 5 18 2 5 1 10 5 19 1 4 0 2 0 9 6] - [ 4 9 13 2 1 4 1038 5 1 1 1 2 1 1 0 3 3 3 1 5 1] - [ 0 12 14 1 2 31 4 922 0 1 1 6 1 1 1 2 0 2 9 11 3] - [ 18 2 1 1 1 2 0 1 861 64 12 1 0 15 18 2 0 3 5 1 1] - [ 74 2 3 0 1 3 1 1 28 856 1 2 0 24 5 3 0 1 2 1 4] - [ 1 1 6 11 1 1 3 6 14 4 957 1 2 19 7 1 0 0 10 1 5] - [ 3 4 2 1 4 14 0 5 1 2 0 894 22 10 1 6 4 12 1 17 2] - [ 1 1 1 4 1 6 1 2 2 0 1 53 837 1 3 10 0 21 0 3 11] - [ 5 2 0 0 6 8 0 1 10 20 7 8 2 929 7 6 5 2 0 3 3] - [ 14 7 4 16 7 5 0 2 26 14 4 0 5 7 942 1 2 6 15 2 13] - [ 7 0 5 0 4 2 2 3 0 2 0 7 7 6 1 973 6 7 1 5 8] - [ 0 5 0 0 8 3 0 0 3 0 0 3 1 6 4 20 985 3 3 6 11] - [ 3 1 3 2 0 0 2 0 1 0 1 13 16 2 0 20 2 972 1 4 8] - [ 3 9 2 17 2 1 0 53 3 1 5 1 5 1 7 4 1 2 963 1 5] - [ 0 2 1 0 0 4 11 13 1 0 0 16 3 2 0 6 10 4 0 1069 6] - [ 163 319 206 115 121 259 108 235 94 115 207 152 269 353 122 137 294 105 167 351 9542]] - -2023-02-13 17:54:32,527 - ==> Best [Top1: 82.539 Top5: 97.403 Sparsity:0.00 Params: 148928 on epoch: 73] -2023-02-13 17:54:32,527 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:54:32,533 - - -2023-02-13 17:54:32,533 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:54:33,430 - Epoch: [82][ 10/ 1207] Overall Loss 0.329585 Objective Loss 0.329585 LR 0.001000 Time 0.089650 -2023-02-13 17:54:33,623 - Epoch: [82][ 20/ 1207] Overall Loss 0.318172 Objective Loss 0.318172 LR 0.001000 Time 0.054457 -2023-02-13 17:54:33,813 - Epoch: [82][ 30/ 1207] Overall Loss 0.316328 Objective Loss 0.316328 LR 0.001000 Time 0.042615 -2023-02-13 17:54:34,003 - Epoch: [82][ 40/ 1207] Overall Loss 0.315090 Objective Loss 0.315090 LR 0.001000 Time 0.036689 -2023-02-13 17:54:34,192 - Epoch: [82][ 50/ 1207] Overall Loss 0.312644 Objective Loss 0.312644 LR 0.001000 Time 0.033139 -2023-02-13 17:54:34,382 - Epoch: [82][ 60/ 1207] Overall Loss 0.317027 Objective Loss 0.317027 LR 0.001000 Time 0.030765 -2023-02-13 17:54:34,571 - Epoch: [82][ 70/ 1207] Overall Loss 0.315216 Objective Loss 0.315216 LR 0.001000 Time 0.029075 -2023-02-13 17:54:34,761 - Epoch: [82][ 80/ 1207] Overall Loss 0.318367 Objective Loss 0.318367 LR 0.001000 Time 0.027805 -2023-02-13 17:54:34,951 - Epoch: [82][ 90/ 1207] Overall Loss 0.321932 Objective Loss 0.321932 LR 0.001000 Time 0.026821 -2023-02-13 17:54:35,141 - Epoch: [82][ 100/ 1207] Overall Loss 0.326713 Objective Loss 0.326713 LR 0.001000 Time 0.026037 -2023-02-13 17:54:35,330 - Epoch: [82][ 110/ 1207] Overall Loss 0.324143 Objective Loss 0.324143 LR 0.001000 Time 0.025387 -2023-02-13 17:54:35,524 - Epoch: [82][ 120/ 1207] Overall Loss 0.322771 Objective Loss 0.322771 LR 0.001000 Time 0.024887 -2023-02-13 17:54:35,716 - Epoch: [82][ 130/ 1207] Overall Loss 0.324669 Objective Loss 0.324669 LR 0.001000 Time 0.024446 -2023-02-13 17:54:35,911 - Epoch: [82][ 140/ 1207] Overall Loss 0.324666 Objective Loss 0.324666 LR 0.001000 Time 0.024088 -2023-02-13 17:54:36,103 - Epoch: [82][ 150/ 1207] Overall Loss 0.324942 Objective Loss 0.324942 LR 0.001000 Time 0.023762 -2023-02-13 17:54:36,297 - Epoch: [82][ 160/ 1207] Overall Loss 0.327354 Objective Loss 0.327354 LR 0.001000 Time 0.023485 -2023-02-13 17:54:36,490 - Epoch: [82][ 170/ 1207] Overall Loss 0.327388 Objective Loss 0.327388 LR 0.001000 Time 0.023235 -2023-02-13 17:54:36,684 - Epoch: [82][ 180/ 1207] Overall Loss 0.327969 Objective Loss 0.327969 LR 0.001000 Time 0.023020 -2023-02-13 17:54:36,876 - Epoch: [82][ 190/ 1207] Overall Loss 0.327285 Objective Loss 0.327285 LR 0.001000 Time 0.022817 -2023-02-13 17:54:37,070 - Epoch: [82][ 200/ 1207] Overall Loss 0.327976 Objective Loss 0.327976 LR 0.001000 Time 0.022644 -2023-02-13 17:54:37,262 - Epoch: [82][ 210/ 1207] Overall Loss 0.327948 Objective Loss 0.327948 LR 0.001000 Time 0.022479 -2023-02-13 17:54:37,456 - Epoch: [82][ 220/ 1207] Overall Loss 0.327380 Objective Loss 0.327380 LR 0.001000 Time 0.022337 -2023-02-13 17:54:37,649 - Epoch: [82][ 230/ 1207] Overall Loss 0.326910 Objective Loss 0.326910 LR 0.001000 Time 0.022204 -2023-02-13 17:54:37,843 - Epoch: [82][ 240/ 1207] Overall Loss 0.326740 Objective Loss 0.326740 LR 0.001000 Time 0.022085 -2023-02-13 17:54:38,036 - Epoch: [82][ 250/ 1207] Overall Loss 0.326425 Objective Loss 0.326425 LR 0.001000 Time 0.021971 -2023-02-13 17:54:38,230 - Epoch: [82][ 260/ 1207] Overall Loss 0.325828 Objective Loss 0.325828 LR 0.001000 Time 0.021871 -2023-02-13 17:54:38,422 - Epoch: [82][ 270/ 1207] Overall Loss 0.325001 Objective Loss 0.325001 LR 0.001000 Time 0.021773 -2023-02-13 17:54:38,618 - Epoch: [82][ 280/ 1207] Overall Loss 0.325861 Objective Loss 0.325861 LR 0.001000 Time 0.021693 -2023-02-13 17:54:38,810 - Epoch: [82][ 290/ 1207] Overall Loss 0.325946 Objective Loss 0.325946 LR 0.001000 Time 0.021607 -2023-02-13 17:54:39,004 - Epoch: [82][ 300/ 1207] Overall Loss 0.325121 Objective Loss 0.325121 LR 0.001000 Time 0.021532 -2023-02-13 17:54:39,197 - Epoch: [82][ 310/ 1207] Overall Loss 0.324430 Objective Loss 0.324430 LR 0.001000 Time 0.021459 -2023-02-13 17:54:39,391 - Epoch: [82][ 320/ 1207] Overall Loss 0.324403 Objective Loss 0.324403 LR 0.001000 Time 0.021393 -2023-02-13 17:54:39,584 - Epoch: [82][ 330/ 1207] Overall Loss 0.323447 Objective Loss 0.323447 LR 0.001000 Time 0.021329 -2023-02-13 17:54:39,778 - Epoch: [82][ 340/ 1207] Overall Loss 0.323333 Objective Loss 0.323333 LR 0.001000 Time 0.021271 -2023-02-13 17:54:39,970 - Epoch: [82][ 350/ 1207] Overall Loss 0.323463 Objective Loss 0.323463 LR 0.001000 Time 0.021211 -2023-02-13 17:54:40,165 - Epoch: [82][ 360/ 1207] Overall Loss 0.323759 Objective Loss 0.323759 LR 0.001000 Time 0.021160 -2023-02-13 17:54:40,356 - Epoch: [82][ 370/ 1207] Overall Loss 0.324714 Objective Loss 0.324714 LR 0.001000 Time 0.021105 -2023-02-13 17:54:40,551 - Epoch: [82][ 380/ 1207] Overall Loss 0.324378 Objective Loss 0.324378 LR 0.001000 Time 0.021062 -2023-02-13 17:54:40,744 - Epoch: [82][ 390/ 1207] Overall Loss 0.324431 Objective Loss 0.324431 LR 0.001000 Time 0.021015 -2023-02-13 17:54:40,938 - Epoch: [82][ 400/ 1207] Overall Loss 0.325044 Objective Loss 0.325044 LR 0.001000 Time 0.020975 -2023-02-13 17:54:41,131 - Epoch: [82][ 410/ 1207] Overall Loss 0.325888 Objective Loss 0.325888 LR 0.001000 Time 0.020931 -2023-02-13 17:54:41,325 - Epoch: [82][ 420/ 1207] Overall Loss 0.325679 Objective Loss 0.325679 LR 0.001000 Time 0.020894 -2023-02-13 17:54:41,517 - Epoch: [82][ 430/ 1207] Overall Loss 0.326003 Objective Loss 0.326003 LR 0.001000 Time 0.020855 -2023-02-13 17:54:41,711 - Epoch: [82][ 440/ 1207] Overall Loss 0.326321 Objective Loss 0.326321 LR 0.001000 Time 0.020821 -2023-02-13 17:54:41,904 - Epoch: [82][ 450/ 1207] Overall Loss 0.326392 Objective Loss 0.326392 LR 0.001000 Time 0.020786 -2023-02-13 17:54:42,098 - Epoch: [82][ 460/ 1207] Overall Loss 0.326274 Objective Loss 0.326274 LR 0.001000 Time 0.020754 -2023-02-13 17:54:42,290 - Epoch: [82][ 470/ 1207] Overall Loss 0.326338 Objective Loss 0.326338 LR 0.001000 Time 0.020720 -2023-02-13 17:54:42,483 - Epoch: [82][ 480/ 1207] Overall Loss 0.325990 Objective Loss 0.325990 LR 0.001000 Time 0.020690 -2023-02-13 17:54:42,676 - Epoch: [82][ 490/ 1207] Overall Loss 0.326444 Objective Loss 0.326444 LR 0.001000 Time 0.020661 -2023-02-13 17:54:42,870 - Epoch: [82][ 500/ 1207] Overall Loss 0.326115 Objective Loss 0.326115 LR 0.001000 Time 0.020635 -2023-02-13 17:54:43,062 - Epoch: [82][ 510/ 1207] Overall Loss 0.325906 Objective Loss 0.325906 LR 0.001000 Time 0.020606 -2023-02-13 17:54:43,255 - Epoch: [82][ 520/ 1207] Overall Loss 0.326237 Objective Loss 0.326237 LR 0.001000 Time 0.020581 -2023-02-13 17:54:43,447 - Epoch: [82][ 530/ 1207] Overall Loss 0.326989 Objective Loss 0.326989 LR 0.001000 Time 0.020554 -2023-02-13 17:54:43,642 - Epoch: [82][ 540/ 1207] Overall Loss 0.326942 Objective Loss 0.326942 LR 0.001000 Time 0.020533 -2023-02-13 17:54:43,834 - Epoch: [82][ 550/ 1207] Overall Loss 0.327313 Objective Loss 0.327313 LR 0.001000 Time 0.020508 -2023-02-13 17:54:44,027 - Epoch: [82][ 560/ 1207] Overall Loss 0.326870 Objective Loss 0.326870 LR 0.001000 Time 0.020487 -2023-02-13 17:54:44,220 - Epoch: [82][ 570/ 1207] Overall Loss 0.326499 Objective Loss 0.326499 LR 0.001000 Time 0.020465 -2023-02-13 17:54:44,414 - Epoch: [82][ 580/ 1207] Overall Loss 0.326612 Objective Loss 0.326612 LR 0.001000 Time 0.020445 -2023-02-13 17:54:44,606 - Epoch: [82][ 590/ 1207] Overall Loss 0.326688 Objective Loss 0.326688 LR 0.001000 Time 0.020425 -2023-02-13 17:54:44,800 - Epoch: [82][ 600/ 1207] Overall Loss 0.326832 Objective Loss 0.326832 LR 0.001000 Time 0.020406 -2023-02-13 17:54:44,992 - Epoch: [82][ 610/ 1207] Overall Loss 0.326463 Objective Loss 0.326463 LR 0.001000 Time 0.020387 -2023-02-13 17:54:45,186 - Epoch: [82][ 620/ 1207] Overall Loss 0.327305 Objective Loss 0.327305 LR 0.001000 Time 0.020370 -2023-02-13 17:54:45,379 - Epoch: [82][ 630/ 1207] Overall Loss 0.327303 Objective Loss 0.327303 LR 0.001000 Time 0.020352 -2023-02-13 17:54:45,573 - Epoch: [82][ 640/ 1207] Overall Loss 0.327279 Objective Loss 0.327279 LR 0.001000 Time 0.020337 -2023-02-13 17:54:45,765 - Epoch: [82][ 650/ 1207] Overall Loss 0.327987 Objective Loss 0.327987 LR 0.001000 Time 0.020320 -2023-02-13 17:54:45,960 - Epoch: [82][ 660/ 1207] Overall Loss 0.328010 Objective Loss 0.328010 LR 0.001000 Time 0.020306 -2023-02-13 17:54:46,153 - Epoch: [82][ 670/ 1207] Overall Loss 0.327894 Objective Loss 0.327894 LR 0.001000 Time 0.020291 -2023-02-13 17:54:46,347 - Epoch: [82][ 680/ 1207] Overall Loss 0.327916 Objective Loss 0.327916 LR 0.001000 Time 0.020277 -2023-02-13 17:54:46,539 - Epoch: [82][ 690/ 1207] Overall Loss 0.327802 Objective Loss 0.327802 LR 0.001000 Time 0.020261 -2023-02-13 17:54:46,734 - Epoch: [82][ 700/ 1207] Overall Loss 0.327558 Objective Loss 0.327558 LR 0.001000 Time 0.020249 -2023-02-13 17:54:46,926 - Epoch: [82][ 710/ 1207] Overall Loss 0.327663 Objective Loss 0.327663 LR 0.001000 Time 0.020234 -2023-02-13 17:54:47,120 - Epoch: [82][ 720/ 1207] Overall Loss 0.328032 Objective Loss 0.328032 LR 0.001000 Time 0.020222 -2023-02-13 17:54:47,313 - Epoch: [82][ 730/ 1207] Overall Loss 0.328293 Objective Loss 0.328293 LR 0.001000 Time 0.020208 -2023-02-13 17:54:47,507 - Epoch: [82][ 740/ 1207] Overall Loss 0.328488 Objective Loss 0.328488 LR 0.001000 Time 0.020197 -2023-02-13 17:54:47,699 - Epoch: [82][ 750/ 1207] Overall Loss 0.328831 Objective Loss 0.328831 LR 0.001000 Time 0.020183 -2023-02-13 17:54:47,893 - Epoch: [82][ 760/ 1207] Overall Loss 0.328922 Objective Loss 0.328922 LR 0.001000 Time 0.020172 -2023-02-13 17:54:48,085 - Epoch: [82][ 770/ 1207] Overall Loss 0.328826 Objective Loss 0.328826 LR 0.001000 Time 0.020160 -2023-02-13 17:54:48,279 - Epoch: [82][ 780/ 1207] Overall Loss 0.328697 Objective Loss 0.328697 LR 0.001000 Time 0.020149 -2023-02-13 17:54:48,471 - Epoch: [82][ 790/ 1207] Overall Loss 0.328641 Objective Loss 0.328641 LR 0.001000 Time 0.020137 -2023-02-13 17:54:48,666 - Epoch: [82][ 800/ 1207] Overall Loss 0.328820 Objective Loss 0.328820 LR 0.001000 Time 0.020128 -2023-02-13 17:54:48,859 - Epoch: [82][ 810/ 1207] Overall Loss 0.329063 Objective Loss 0.329063 LR 0.001000 Time 0.020117 -2023-02-13 17:54:49,054 - Epoch: [82][ 820/ 1207] Overall Loss 0.329179 Objective Loss 0.329179 LR 0.001000 Time 0.020109 -2023-02-13 17:54:49,246 - Epoch: [82][ 830/ 1207] Overall Loss 0.329067 Objective Loss 0.329067 LR 0.001000 Time 0.020099 -2023-02-13 17:54:49,441 - Epoch: [82][ 840/ 1207] Overall Loss 0.329033 Objective Loss 0.329033 LR 0.001000 Time 0.020090 -2023-02-13 17:54:49,635 - Epoch: [82][ 850/ 1207] Overall Loss 0.329107 Objective Loss 0.329107 LR 0.001000 Time 0.020082 -2023-02-13 17:54:49,829 - Epoch: [82][ 860/ 1207] Overall Loss 0.329167 Objective Loss 0.329167 LR 0.001000 Time 0.020073 -2023-02-13 17:54:50,022 - Epoch: [82][ 870/ 1207] Overall Loss 0.329235 Objective Loss 0.329235 LR 0.001000 Time 0.020064 -2023-02-13 17:54:50,217 - Epoch: [82][ 880/ 1207] Overall Loss 0.329693 Objective Loss 0.329693 LR 0.001000 Time 0.020057 -2023-02-13 17:54:50,409 - Epoch: [82][ 890/ 1207] Overall Loss 0.329591 Objective Loss 0.329591 LR 0.001000 Time 0.020048 -2023-02-13 17:54:50,605 - Epoch: [82][ 900/ 1207] Overall Loss 0.329802 Objective Loss 0.329802 LR 0.001000 Time 0.020042 -2023-02-13 17:54:50,797 - Epoch: [82][ 910/ 1207] Overall Loss 0.330451 Objective Loss 0.330451 LR 0.001000 Time 0.020032 -2023-02-13 17:54:50,992 - Epoch: [82][ 920/ 1207] Overall Loss 0.330823 Objective Loss 0.330823 LR 0.001000 Time 0.020026 -2023-02-13 17:54:51,185 - Epoch: [82][ 930/ 1207] Overall Loss 0.331223 Objective Loss 0.331223 LR 0.001000 Time 0.020018 -2023-02-13 17:54:51,378 - Epoch: [82][ 940/ 1207] Overall Loss 0.331582 Objective Loss 0.331582 LR 0.001000 Time 0.020010 -2023-02-13 17:54:51,571 - Epoch: [82][ 950/ 1207] Overall Loss 0.331735 Objective Loss 0.331735 LR 0.001000 Time 0.020002 -2023-02-13 17:54:51,765 - Epoch: [82][ 960/ 1207] Overall Loss 0.332583 Objective Loss 0.332583 LR 0.001000 Time 0.019996 -2023-02-13 17:54:51,959 - Epoch: [82][ 970/ 1207] Overall Loss 0.332681 Objective Loss 0.332681 LR 0.001000 Time 0.019989 -2023-02-13 17:54:52,153 - Epoch: [82][ 980/ 1207] Overall Loss 0.332818 Objective Loss 0.332818 LR 0.001000 Time 0.019982 -2023-02-13 17:54:52,345 - Epoch: [82][ 990/ 1207] Overall Loss 0.333360 Objective Loss 0.333360 LR 0.001000 Time 0.019974 -2023-02-13 17:54:52,540 - Epoch: [82][ 1000/ 1207] Overall Loss 0.333326 Objective Loss 0.333326 LR 0.001000 Time 0.019969 -2023-02-13 17:54:52,733 - Epoch: [82][ 1010/ 1207] Overall Loss 0.333396 Objective Loss 0.333396 LR 0.001000 Time 0.019962 -2023-02-13 17:54:52,927 - Epoch: [82][ 1020/ 1207] Overall Loss 0.333355 Objective Loss 0.333355 LR 0.001000 Time 0.019957 -2023-02-13 17:54:53,120 - Epoch: [82][ 1030/ 1207] Overall Loss 0.333228 Objective Loss 0.333228 LR 0.001000 Time 0.019950 -2023-02-13 17:54:53,315 - Epoch: [82][ 1040/ 1207] Overall Loss 0.333097 Objective Loss 0.333097 LR 0.001000 Time 0.019945 -2023-02-13 17:54:53,507 - Epoch: [82][ 1050/ 1207] Overall Loss 0.333025 Objective Loss 0.333025 LR 0.001000 Time 0.019938 -2023-02-13 17:54:53,702 - Epoch: [82][ 1060/ 1207] Overall Loss 0.333158 Objective Loss 0.333158 LR 0.001000 Time 0.019933 -2023-02-13 17:54:53,895 - Epoch: [82][ 1070/ 1207] Overall Loss 0.333218 Objective Loss 0.333218 LR 0.001000 Time 0.019926 -2023-02-13 17:54:54,089 - Epoch: [82][ 1080/ 1207] Overall Loss 0.333134 Objective Loss 0.333134 LR 0.001000 Time 0.019921 -2023-02-13 17:54:54,281 - Epoch: [82][ 1090/ 1207] Overall Loss 0.333228 Objective Loss 0.333228 LR 0.001000 Time 0.019915 -2023-02-13 17:54:54,475 - Epoch: [82][ 1100/ 1207] Overall Loss 0.332934 Objective Loss 0.332934 LR 0.001000 Time 0.019910 -2023-02-13 17:54:54,669 - Epoch: [82][ 1110/ 1207] Overall Loss 0.333165 Objective Loss 0.333165 LR 0.001000 Time 0.019904 -2023-02-13 17:54:54,863 - Epoch: [82][ 1120/ 1207] Overall Loss 0.333341 Objective Loss 0.333341 LR 0.001000 Time 0.019900 -2023-02-13 17:54:55,056 - Epoch: [82][ 1130/ 1207] Overall Loss 0.333397 Objective Loss 0.333397 LR 0.001000 Time 0.019894 -2023-02-13 17:54:55,250 - Epoch: [82][ 1140/ 1207] Overall Loss 0.333702 Objective Loss 0.333702 LR 0.001000 Time 0.019890 -2023-02-13 17:54:55,443 - Epoch: [82][ 1150/ 1207] Overall Loss 0.333849 Objective Loss 0.333849 LR 0.001000 Time 0.019884 -2023-02-13 17:54:55,638 - Epoch: [82][ 1160/ 1207] Overall Loss 0.334076 Objective Loss 0.334076 LR 0.001000 Time 0.019880 -2023-02-13 17:54:55,831 - Epoch: [82][ 1170/ 1207] Overall Loss 0.334014 Objective Loss 0.334014 LR 0.001000 Time 0.019875 -2023-02-13 17:54:56,026 - Epoch: [82][ 1180/ 1207] Overall Loss 0.334010 Objective Loss 0.334010 LR 0.001000 Time 0.019871 -2023-02-13 17:54:56,219 - Epoch: [82][ 1190/ 1207] Overall Loss 0.334009 Objective Loss 0.334009 LR 0.001000 Time 0.019866 -2023-02-13 17:54:56,467 - Epoch: [82][ 1200/ 1207] Overall Loss 0.333759 Objective Loss 0.333759 LR 0.001000 Time 0.019908 -2023-02-13 17:54:56,583 - Epoch: [82][ 1207/ 1207] Overall Loss 0.334035 Objective Loss 0.334035 Top1 83.231707 Top5 97.560976 LR 0.001000 Time 0.019888 -2023-02-13 17:54:56,665 - --- validate (epoch=82)----------- -2023-02-13 17:54:56,665 - 34311 samples (256 per mini-batch) -2023-02-13 17:54:57,071 - Epoch: [82][ 10/ 135] Loss 0.335683 Top1 83.164062 Top5 97.734375 -2023-02-13 17:54:57,199 - Epoch: [82][ 20/ 135] Loss 0.338082 Top1 83.281250 Top5 97.421875 -2023-02-13 17:54:57,323 - Epoch: [82][ 30/ 135] Loss 0.341253 Top1 83.203125 Top5 97.369792 -2023-02-13 17:54:57,450 - Epoch: [82][ 40/ 135] Loss 0.362546 Top1 82.792969 Top5 97.382812 -2023-02-13 17:54:57,573 - Epoch: [82][ 50/ 135] Loss 0.360783 Top1 82.812500 Top5 97.500000 -2023-02-13 17:54:57,697 - Epoch: [82][ 60/ 135] Loss 0.360870 Top1 82.578125 Top5 97.408854 -2023-02-13 17:54:57,824 - Epoch: [82][ 70/ 135] Loss 0.367287 Top1 82.611607 Top5 97.332589 -2023-02-13 17:54:57,952 - Epoch: [82][ 80/ 135] Loss 0.360134 Top1 82.797852 Top5 97.431641 -2023-02-13 17:54:58,083 - Epoch: [82][ 90/ 135] Loss 0.358097 Top1 82.782118 Top5 97.426215 -2023-02-13 17:54:58,212 - Epoch: [82][ 100/ 135] Loss 0.359953 Top1 82.738281 Top5 97.445312 -2023-02-13 17:54:58,343 - Epoch: [82][ 110/ 135] Loss 0.359287 Top1 82.798295 Top5 97.450284 -2023-02-13 17:54:58,472 - Epoch: [82][ 120/ 135] Loss 0.359169 Top1 82.802734 Top5 97.464193 -2023-02-13 17:54:58,605 - Epoch: [82][ 130/ 135] Loss 0.360056 Top1 82.743389 Top5 97.409856 -2023-02-13 17:54:58,651 - Epoch: [82][ 135/ 135] Loss 0.357898 Top1 82.760631 Top5 97.391507 -2023-02-13 17:54:58,734 - ==> Top1: 82.761 Top5: 97.392 Loss: 0.358 - -2023-02-13 17:54:58,735 - ==> Confusion: -[[ 839 7 6 2 14 3 1 2 4 55 0 4 0 3 6 2 7 4 1 1 6] - [ 1 947 1 3 13 25 3 5 2 2 2 7 1 1 1 1 4 0 6 1 7] - [ 7 4 935 11 6 1 27 12 0 1 4 2 2 2 4 12 2 4 8 3 11] - [ 8 2 29 854 2 7 0 1 1 2 27 0 7 2 33 4 5 7 19 1 5] - [ 15 7 0 0 990 5 1 1 0 2 2 3 0 1 13 9 6 4 0 1 6] - [ 3 35 0 5 12 927 6 14 1 4 4 14 3 17 2 4 6 0 0 7 6] - [ 4 6 18 2 1 5 1024 5 0 1 5 2 1 2 0 5 2 4 1 8 3] - [ 1 32 14 2 5 31 2 876 0 1 4 6 5 2 1 1 1 3 25 6 6] - [ 20 4 2 2 1 1 0 2 852 45 22 1 1 10 32 3 1 2 6 0 2] - [ 74 6 3 0 8 0 1 0 33 849 1 1 0 16 10 2 0 1 1 0 6] - [ 3 3 4 5 1 1 3 4 10 1 980 1 0 8 5 1 2 2 13 0 4] - [ 4 2 0 0 6 5 2 2 0 2 0 915 23 9 1 4 3 9 2 13 3] - [ 1 1 0 5 1 3 0 2 2 0 2 44 858 1 4 5 2 12 2 2 12] - [ 3 2 3 0 6 8 0 3 12 22 25 4 1 898 7 9 7 1 1 7 5] - [ 13 2 4 7 8 3 1 0 17 5 4 2 2 1 997 3 2 5 8 0 8] - [ 5 1 5 0 5 0 5 1 0 1 0 9 7 5 1 967 10 11 1 8 4] - [ 1 8 2 1 10 3 0 0 1 0 0 3 3 4 3 14 992 0 1 5 10] - [ 3 2 2 4 1 0 4 0 1 1 0 10 22 0 2 18 1 969 0 3 8] - [ 4 3 5 8 3 2 0 27 1 0 5 2 4 0 19 1 3 2 994 1 2] - [ 1 3 0 2 2 4 11 11 1 0 1 15 4 4 0 6 7 3 1 1063 9] - [ 171 285 240 96 185 153 112 132 73 90 261 131 333 295 221 130 273 112 207 264 9670]] - -2023-02-13 17:54:58,736 - ==> Best [Top1: 82.761 Top5: 97.392 Sparsity:0.00 Params: 148928 on epoch: 82] -2023-02-13 17:54:58,736 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:54:58,743 - - -2023-02-13 17:54:58,743 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:54:59,721 - Epoch: [83][ 10/ 1207] Overall Loss 0.336216 Objective Loss 0.336216 LR 0.001000 Time 0.097741 -2023-02-13 17:54:59,912 - Epoch: [83][ 20/ 1207] Overall Loss 0.334076 Objective Loss 0.334076 LR 0.001000 Time 0.058392 -2023-02-13 17:55:00,099 - Epoch: [83][ 30/ 1207] Overall Loss 0.332690 Objective Loss 0.332690 LR 0.001000 Time 0.045172 -2023-02-13 17:55:00,287 - Epoch: [83][ 40/ 1207] Overall Loss 0.322139 Objective Loss 0.322139 LR 0.001000 Time 0.038552 -2023-02-13 17:55:00,474 - Epoch: [83][ 50/ 1207] Overall Loss 0.323382 Objective Loss 0.323382 LR 0.001000 Time 0.034574 -2023-02-13 17:55:00,662 - Epoch: [83][ 60/ 1207] Overall Loss 0.321070 Objective Loss 0.321070 LR 0.001000 Time 0.031940 -2023-02-13 17:55:00,850 - Epoch: [83][ 70/ 1207] Overall Loss 0.319241 Objective Loss 0.319241 LR 0.001000 Time 0.030056 -2023-02-13 17:55:01,037 - Epoch: [83][ 80/ 1207] Overall Loss 0.320000 Objective Loss 0.320000 LR 0.001000 Time 0.028639 -2023-02-13 17:55:01,225 - Epoch: [83][ 90/ 1207] Overall Loss 0.320850 Objective Loss 0.320850 LR 0.001000 Time 0.027542 -2023-02-13 17:55:01,413 - Epoch: [83][ 100/ 1207] Overall Loss 0.319715 Objective Loss 0.319715 LR 0.001000 Time 0.026653 -2023-02-13 17:55:01,601 - Epoch: [83][ 110/ 1207] Overall Loss 0.321206 Objective Loss 0.321206 LR 0.001000 Time 0.025937 -2023-02-13 17:55:01,788 - Epoch: [83][ 120/ 1207] Overall Loss 0.320899 Objective Loss 0.320899 LR 0.001000 Time 0.025336 -2023-02-13 17:55:01,976 - Epoch: [83][ 130/ 1207] Overall Loss 0.321572 Objective Loss 0.321572 LR 0.001000 Time 0.024829 -2023-02-13 17:55:02,164 - Epoch: [83][ 140/ 1207] Overall Loss 0.321277 Objective Loss 0.321277 LR 0.001000 Time 0.024397 -2023-02-13 17:55:02,352 - Epoch: [83][ 150/ 1207] Overall Loss 0.318361 Objective Loss 0.318361 LR 0.001000 Time 0.024020 -2023-02-13 17:55:02,539 - Epoch: [83][ 160/ 1207] Overall Loss 0.317947 Objective Loss 0.317947 LR 0.001000 Time 0.023688 -2023-02-13 17:55:02,727 - Epoch: [83][ 170/ 1207] Overall Loss 0.316942 Objective Loss 0.316942 LR 0.001000 Time 0.023398 -2023-02-13 17:55:02,915 - Epoch: [83][ 180/ 1207] Overall Loss 0.318357 Objective Loss 0.318357 LR 0.001000 Time 0.023137 -2023-02-13 17:55:03,102 - Epoch: [83][ 190/ 1207] Overall Loss 0.320517 Objective Loss 0.320517 LR 0.001000 Time 0.022904 -2023-02-13 17:55:03,291 - Epoch: [83][ 200/ 1207] Overall Loss 0.320560 Objective Loss 0.320560 LR 0.001000 Time 0.022699 -2023-02-13 17:55:03,479 - Epoch: [83][ 210/ 1207] Overall Loss 0.321221 Objective Loss 0.321221 LR 0.001000 Time 0.022511 -2023-02-13 17:55:03,667 - Epoch: [83][ 220/ 1207] Overall Loss 0.322910 Objective Loss 0.322910 LR 0.001000 Time 0.022343 -2023-02-13 17:55:03,855 - Epoch: [83][ 230/ 1207] Overall Loss 0.321416 Objective Loss 0.321416 LR 0.001000 Time 0.022185 -2023-02-13 17:55:04,043 - Epoch: [83][ 240/ 1207] Overall Loss 0.321864 Objective Loss 0.321864 LR 0.001000 Time 0.022044 -2023-02-13 17:55:04,231 - Epoch: [83][ 250/ 1207] Overall Loss 0.321802 Objective Loss 0.321802 LR 0.001000 Time 0.021912 -2023-02-13 17:55:04,419 - Epoch: [83][ 260/ 1207] Overall Loss 0.322030 Objective Loss 0.322030 LR 0.001000 Time 0.021790 -2023-02-13 17:55:04,607 - Epoch: [83][ 270/ 1207] Overall Loss 0.321709 Objective Loss 0.321709 LR 0.001000 Time 0.021679 -2023-02-13 17:55:04,795 - Epoch: [83][ 280/ 1207] Overall Loss 0.320981 Objective Loss 0.320981 LR 0.001000 Time 0.021575 -2023-02-13 17:55:04,982 - Epoch: [83][ 290/ 1207] Overall Loss 0.320674 Objective Loss 0.320674 LR 0.001000 Time 0.021475 -2023-02-13 17:55:05,170 - Epoch: [83][ 300/ 1207] Overall Loss 0.321058 Objective Loss 0.321058 LR 0.001000 Time 0.021385 -2023-02-13 17:55:05,357 - Epoch: [83][ 310/ 1207] Overall Loss 0.321380 Objective Loss 0.321380 LR 0.001000 Time 0.021298 -2023-02-13 17:55:05,545 - Epoch: [83][ 320/ 1207] Overall Loss 0.321325 Objective Loss 0.321325 LR 0.001000 Time 0.021217 -2023-02-13 17:55:05,733 - Epoch: [83][ 330/ 1207] Overall Loss 0.322175 Objective Loss 0.322175 LR 0.001000 Time 0.021142 -2023-02-13 17:55:05,921 - Epoch: [83][ 340/ 1207] Overall Loss 0.322258 Objective Loss 0.322258 LR 0.001000 Time 0.021072 -2023-02-13 17:55:06,108 - Epoch: [83][ 350/ 1207] Overall Loss 0.322309 Objective Loss 0.322309 LR 0.001000 Time 0.021005 -2023-02-13 17:55:06,296 - Epoch: [83][ 360/ 1207] Overall Loss 0.321489 Objective Loss 0.321489 LR 0.001000 Time 0.020943 -2023-02-13 17:55:06,484 - Epoch: [83][ 370/ 1207] Overall Loss 0.322243 Objective Loss 0.322243 LR 0.001000 Time 0.020883 -2023-02-13 17:55:06,672 - Epoch: [83][ 380/ 1207] Overall Loss 0.323067 Objective Loss 0.323067 LR 0.001000 Time 0.020828 -2023-02-13 17:55:06,860 - Epoch: [83][ 390/ 1207] Overall Loss 0.323310 Objective Loss 0.323310 LR 0.001000 Time 0.020773 -2023-02-13 17:55:07,047 - Epoch: [83][ 400/ 1207] Overall Loss 0.323605 Objective Loss 0.323605 LR 0.001000 Time 0.020723 -2023-02-13 17:55:07,235 - Epoch: [83][ 410/ 1207] Overall Loss 0.324077 Objective Loss 0.324077 LR 0.001000 Time 0.020674 -2023-02-13 17:55:07,423 - Epoch: [83][ 420/ 1207] Overall Loss 0.324613 Objective Loss 0.324613 LR 0.001000 Time 0.020629 -2023-02-13 17:55:07,612 - Epoch: [83][ 430/ 1207] Overall Loss 0.325183 Objective Loss 0.325183 LR 0.001000 Time 0.020588 -2023-02-13 17:55:07,801 - Epoch: [83][ 440/ 1207] Overall Loss 0.325668 Objective Loss 0.325668 LR 0.001000 Time 0.020547 -2023-02-13 17:55:07,987 - Epoch: [83][ 450/ 1207] Overall Loss 0.325858 Objective Loss 0.325858 LR 0.001000 Time 0.020504 -2023-02-13 17:55:08,175 - Epoch: [83][ 460/ 1207] Overall Loss 0.326107 Objective Loss 0.326107 LR 0.001000 Time 0.020466 -2023-02-13 17:55:08,362 - Epoch: [83][ 470/ 1207] Overall Loss 0.325877 Objective Loss 0.325877 LR 0.001000 Time 0.020427 -2023-02-13 17:55:08,550 - Epoch: [83][ 480/ 1207] Overall Loss 0.326396 Objective Loss 0.326396 LR 0.001000 Time 0.020392 -2023-02-13 17:55:08,737 - Epoch: [83][ 490/ 1207] Overall Loss 0.326032 Objective Loss 0.326032 LR 0.001000 Time 0.020359 -2023-02-13 17:55:08,925 - Epoch: [83][ 500/ 1207] Overall Loss 0.325614 Objective Loss 0.325614 LR 0.001000 Time 0.020325 -2023-02-13 17:55:09,112 - Epoch: [83][ 510/ 1207] Overall Loss 0.324241 Objective Loss 0.324241 LR 0.001000 Time 0.020293 -2023-02-13 17:55:09,299 - Epoch: [83][ 520/ 1207] Overall Loss 0.324250 Objective Loss 0.324250 LR 0.001000 Time 0.020263 -2023-02-13 17:55:09,487 - Epoch: [83][ 530/ 1207] Overall Loss 0.324201 Objective Loss 0.324201 LR 0.001000 Time 0.020233 -2023-02-13 17:55:09,675 - Epoch: [83][ 540/ 1207] Overall Loss 0.324454 Objective Loss 0.324454 LR 0.001000 Time 0.020207 -2023-02-13 17:55:09,863 - Epoch: [83][ 550/ 1207] Overall Loss 0.324948 Objective Loss 0.324948 LR 0.001000 Time 0.020179 -2023-02-13 17:55:10,050 - Epoch: [83][ 560/ 1207] Overall Loss 0.325098 Objective Loss 0.325098 LR 0.001000 Time 0.020153 -2023-02-13 17:55:10,238 - Epoch: [83][ 570/ 1207] Overall Loss 0.325671 Objective Loss 0.325671 LR 0.001000 Time 0.020129 -2023-02-13 17:55:10,425 - Epoch: [83][ 580/ 1207] Overall Loss 0.325828 Objective Loss 0.325828 LR 0.001000 Time 0.020104 -2023-02-13 17:55:10,614 - Epoch: [83][ 590/ 1207] Overall Loss 0.325637 Objective Loss 0.325637 LR 0.001000 Time 0.020082 -2023-02-13 17:55:10,803 - Epoch: [83][ 600/ 1207] Overall Loss 0.325804 Objective Loss 0.325804 LR 0.001000 Time 0.020061 -2023-02-13 17:55:10,991 - Epoch: [83][ 610/ 1207] Overall Loss 0.325862 Objective Loss 0.325862 LR 0.001000 Time 0.020040 -2023-02-13 17:55:11,178 - Epoch: [83][ 620/ 1207] Overall Loss 0.325976 Objective Loss 0.325976 LR 0.001000 Time 0.020019 -2023-02-13 17:55:11,365 - Epoch: [83][ 630/ 1207] Overall Loss 0.326530 Objective Loss 0.326530 LR 0.001000 Time 0.019997 -2023-02-13 17:55:11,553 - Epoch: [83][ 640/ 1207] Overall Loss 0.326110 Objective Loss 0.326110 LR 0.001000 Time 0.019978 -2023-02-13 17:55:11,742 - Epoch: [83][ 650/ 1207] Overall Loss 0.326213 Objective Loss 0.326213 LR 0.001000 Time 0.019960 -2023-02-13 17:55:11,929 - Epoch: [83][ 660/ 1207] Overall Loss 0.326333 Objective Loss 0.326333 LR 0.001000 Time 0.019940 -2023-02-13 17:55:12,116 - Epoch: [83][ 670/ 1207] Overall Loss 0.325512 Objective Loss 0.325512 LR 0.001000 Time 0.019921 -2023-02-13 17:55:12,303 - Epoch: [83][ 680/ 1207] Overall Loss 0.325648 Objective Loss 0.325648 LR 0.001000 Time 0.019903 -2023-02-13 17:55:12,490 - Epoch: [83][ 690/ 1207] Overall Loss 0.325734 Objective Loss 0.325734 LR 0.001000 Time 0.019886 -2023-02-13 17:55:12,680 - Epoch: [83][ 700/ 1207] Overall Loss 0.325948 Objective Loss 0.325948 LR 0.001000 Time 0.019871 -2023-02-13 17:55:12,866 - Epoch: [83][ 710/ 1207] Overall Loss 0.325762 Objective Loss 0.325762 LR 0.001000 Time 0.019854 -2023-02-13 17:55:13,055 - Epoch: [83][ 720/ 1207] Overall Loss 0.325918 Objective Loss 0.325918 LR 0.001000 Time 0.019840 -2023-02-13 17:55:13,242 - Epoch: [83][ 730/ 1207] Overall Loss 0.326309 Objective Loss 0.326309 LR 0.001000 Time 0.019824 -2023-02-13 17:55:13,429 - Epoch: [83][ 740/ 1207] Overall Loss 0.325859 Objective Loss 0.325859 LR 0.001000 Time 0.019809 -2023-02-13 17:55:13,618 - Epoch: [83][ 750/ 1207] Overall Loss 0.326214 Objective Loss 0.326214 LR 0.001000 Time 0.019796 -2023-02-13 17:55:13,806 - Epoch: [83][ 760/ 1207] Overall Loss 0.326446 Objective Loss 0.326446 LR 0.001000 Time 0.019782 -2023-02-13 17:55:13,993 - Epoch: [83][ 770/ 1207] Overall Loss 0.326617 Objective Loss 0.326617 LR 0.001000 Time 0.019768 -2023-02-13 17:55:14,182 - Epoch: [83][ 780/ 1207] Overall Loss 0.326857 Objective Loss 0.326857 LR 0.001000 Time 0.019756 -2023-02-13 17:55:14,370 - Epoch: [83][ 790/ 1207] Overall Loss 0.327040 Objective Loss 0.327040 LR 0.001000 Time 0.019743 -2023-02-13 17:55:14,557 - Epoch: [83][ 800/ 1207] Overall Loss 0.327195 Objective Loss 0.327195 LR 0.001000 Time 0.019730 -2023-02-13 17:55:14,746 - Epoch: [83][ 810/ 1207] Overall Loss 0.327029 Objective Loss 0.327029 LR 0.001000 Time 0.019719 -2023-02-13 17:55:14,934 - Epoch: [83][ 820/ 1207] Overall Loss 0.327177 Objective Loss 0.327177 LR 0.001000 Time 0.019707 -2023-02-13 17:55:15,121 - Epoch: [83][ 830/ 1207] Overall Loss 0.327301 Objective Loss 0.327301 LR 0.001000 Time 0.019695 -2023-02-13 17:55:15,309 - Epoch: [83][ 840/ 1207] Overall Loss 0.327073 Objective Loss 0.327073 LR 0.001000 Time 0.019684 -2023-02-13 17:55:15,497 - Epoch: [83][ 850/ 1207] Overall Loss 0.327044 Objective Loss 0.327044 LR 0.001000 Time 0.019672 -2023-02-13 17:55:15,685 - Epoch: [83][ 860/ 1207] Overall Loss 0.327134 Objective Loss 0.327134 LR 0.001000 Time 0.019662 -2023-02-13 17:55:15,874 - Epoch: [83][ 870/ 1207] Overall Loss 0.327082 Objective Loss 0.327082 LR 0.001000 Time 0.019653 -2023-02-13 17:55:16,062 - Epoch: [83][ 880/ 1207] Overall Loss 0.326926 Objective Loss 0.326926 LR 0.001000 Time 0.019643 -2023-02-13 17:55:16,251 - Epoch: [83][ 890/ 1207] Overall Loss 0.327210 Objective Loss 0.327210 LR 0.001000 Time 0.019634 -2023-02-13 17:55:16,439 - Epoch: [83][ 900/ 1207] Overall Loss 0.326785 Objective Loss 0.326785 LR 0.001000 Time 0.019625 -2023-02-13 17:55:16,628 - Epoch: [83][ 910/ 1207] Overall Loss 0.327256 Objective Loss 0.327256 LR 0.001000 Time 0.019616 -2023-02-13 17:55:16,817 - Epoch: [83][ 920/ 1207] Overall Loss 0.327690 Objective Loss 0.327690 LR 0.001000 Time 0.019608 -2023-02-13 17:55:17,005 - Epoch: [83][ 930/ 1207] Overall Loss 0.327941 Objective Loss 0.327941 LR 0.001000 Time 0.019599 -2023-02-13 17:55:17,193 - Epoch: [83][ 940/ 1207] Overall Loss 0.327942 Objective Loss 0.327942 LR 0.001000 Time 0.019590 -2023-02-13 17:55:17,381 - Epoch: [83][ 950/ 1207] Overall Loss 0.327978 Objective Loss 0.327978 LR 0.001000 Time 0.019581 -2023-02-13 17:55:17,569 - Epoch: [83][ 960/ 1207] Overall Loss 0.327895 Objective Loss 0.327895 LR 0.001000 Time 0.019572 -2023-02-13 17:55:17,757 - Epoch: [83][ 970/ 1207] Overall Loss 0.328237 Objective Loss 0.328237 LR 0.001000 Time 0.019565 -2023-02-13 17:55:17,945 - Epoch: [83][ 980/ 1207] Overall Loss 0.328285 Objective Loss 0.328285 LR 0.001000 Time 0.019556 -2023-02-13 17:55:18,133 - Epoch: [83][ 990/ 1207] Overall Loss 0.328145 Objective Loss 0.328145 LR 0.001000 Time 0.019548 -2023-02-13 17:55:18,322 - Epoch: [83][ 1000/ 1207] Overall Loss 0.327959 Objective Loss 0.327959 LR 0.001000 Time 0.019541 -2023-02-13 17:55:18,510 - Epoch: [83][ 1010/ 1207] Overall Loss 0.328172 Objective Loss 0.328172 LR 0.001000 Time 0.019533 -2023-02-13 17:55:18,698 - Epoch: [83][ 1020/ 1207] Overall Loss 0.328056 Objective Loss 0.328056 LR 0.001000 Time 0.019526 -2023-02-13 17:55:18,886 - Epoch: [83][ 1030/ 1207] Overall Loss 0.328069 Objective Loss 0.328069 LR 0.001000 Time 0.019519 -2023-02-13 17:55:19,075 - Epoch: [83][ 1040/ 1207] Overall Loss 0.328156 Objective Loss 0.328156 LR 0.001000 Time 0.019512 -2023-02-13 17:55:19,263 - Epoch: [83][ 1050/ 1207] Overall Loss 0.328038 Objective Loss 0.328038 LR 0.001000 Time 0.019505 -2023-02-13 17:55:19,451 - Epoch: [83][ 1060/ 1207] Overall Loss 0.328072 Objective Loss 0.328072 LR 0.001000 Time 0.019499 -2023-02-13 17:55:19,640 - Epoch: [83][ 1070/ 1207] Overall Loss 0.328197 Objective Loss 0.328197 LR 0.001000 Time 0.019492 -2023-02-13 17:55:19,829 - Epoch: [83][ 1080/ 1207] Overall Loss 0.328180 Objective Loss 0.328180 LR 0.001000 Time 0.019486 -2023-02-13 17:55:20,017 - Epoch: [83][ 1090/ 1207] Overall Loss 0.328388 Objective Loss 0.328388 LR 0.001000 Time 0.019480 -2023-02-13 17:55:20,205 - Epoch: [83][ 1100/ 1207] Overall Loss 0.328362 Objective Loss 0.328362 LR 0.001000 Time 0.019473 -2023-02-13 17:55:20,393 - Epoch: [83][ 1110/ 1207] Overall Loss 0.328438 Objective Loss 0.328438 LR 0.001000 Time 0.019467 -2023-02-13 17:55:20,581 - Epoch: [83][ 1120/ 1207] Overall Loss 0.328335 Objective Loss 0.328335 LR 0.001000 Time 0.019461 -2023-02-13 17:55:20,770 - Epoch: [83][ 1130/ 1207] Overall Loss 0.328385 Objective Loss 0.328385 LR 0.001000 Time 0.019455 -2023-02-13 17:55:20,959 - Epoch: [83][ 1140/ 1207] Overall Loss 0.328415 Objective Loss 0.328415 LR 0.001000 Time 0.019450 -2023-02-13 17:55:21,147 - Epoch: [83][ 1150/ 1207] Overall Loss 0.328474 Objective Loss 0.328474 LR 0.001000 Time 0.019444 -2023-02-13 17:55:21,335 - Epoch: [83][ 1160/ 1207] Overall Loss 0.328917 Objective Loss 0.328917 LR 0.001000 Time 0.019439 -2023-02-13 17:55:21,523 - Epoch: [83][ 1170/ 1207] Overall Loss 0.329022 Objective Loss 0.329022 LR 0.001000 Time 0.019433 -2023-02-13 17:55:21,712 - Epoch: [83][ 1180/ 1207] Overall Loss 0.329063 Objective Loss 0.329063 LR 0.001000 Time 0.019428 -2023-02-13 17:55:21,901 - Epoch: [83][ 1190/ 1207] Overall Loss 0.329199 Objective Loss 0.329199 LR 0.001000 Time 0.019423 -2023-02-13 17:55:22,140 - Epoch: [83][ 1200/ 1207] Overall Loss 0.329136 Objective Loss 0.329136 LR 0.001000 Time 0.019460 -2023-02-13 17:55:22,254 - Epoch: [83][ 1207/ 1207] Overall Loss 0.328794 Objective Loss 0.328794 Top1 82.926829 Top5 98.475610 LR 0.001000 Time 0.019441 -2023-02-13 17:55:22,326 - --- validate (epoch=83)----------- -2023-02-13 17:55:22,326 - 34311 samples (256 per mini-batch) -2023-02-13 17:55:22,719 - Epoch: [83][ 10/ 135] Loss 0.355564 Top1 82.617188 Top5 97.421875 -2023-02-13 17:55:22,849 - Epoch: [83][ 20/ 135] Loss 0.339485 Top1 82.812500 Top5 97.597656 -2023-02-13 17:55:22,979 - Epoch: [83][ 30/ 135] Loss 0.352496 Top1 82.187500 Top5 97.395833 -2023-02-13 17:55:23,109 - Epoch: [83][ 40/ 135] Loss 0.358014 Top1 82.099609 Top5 97.421875 -2023-02-13 17:55:23,237 - Epoch: [83][ 50/ 135] Loss 0.368552 Top1 81.914062 Top5 97.320312 -2023-02-13 17:55:23,371 - Epoch: [83][ 60/ 135] Loss 0.363410 Top1 82.076823 Top5 97.330729 -2023-02-13 17:55:23,508 - Epoch: [83][ 70/ 135] Loss 0.370959 Top1 81.997768 Top5 97.304688 -2023-02-13 17:55:23,649 - Epoch: [83][ 80/ 135] Loss 0.365541 Top1 82.153320 Top5 97.280273 -2023-02-13 17:55:23,787 - Epoch: [83][ 90/ 135] Loss 0.362150 Top1 82.226562 Top5 97.274306 -2023-02-13 17:55:23,925 - Epoch: [83][ 100/ 135] Loss 0.364140 Top1 82.183594 Top5 97.273438 -2023-02-13 17:55:24,063 - Epoch: [83][ 110/ 135] Loss 0.365648 Top1 82.144886 Top5 97.301136 -2023-02-13 17:55:24,203 - Epoch: [83][ 120/ 135] Loss 0.363831 Top1 82.187500 Top5 97.337240 -2023-02-13 17:55:24,336 - Epoch: [83][ 130/ 135] Loss 0.366533 Top1 82.145433 Top5 97.370793 -2023-02-13 17:55:24,380 - Epoch: [83][ 135/ 135] Loss 0.367065 Top1 82.145668 Top5 97.353618 -2023-02-13 17:55:24,459 - ==> Top1: 82.146 Top5: 97.354 Loss: 0.367 - -2023-02-13 17:55:24,460 - ==> Confusion: -[[ 857 5 6 2 8 3 0 1 4 43 2 3 2 4 5 4 1 6 1 1 9] - [ 3 920 2 4 5 23 10 22 4 1 3 2 2 1 5 5 7 1 4 1 8] - [ 5 4 942 11 4 0 34 8 1 1 1 2 3 5 3 12 1 3 4 4 10] - [ 4 1 27 889 1 4 5 3 1 0 10 0 7 3 23 4 4 8 15 0 7] - [ 26 10 2 3 954 12 1 2 2 4 0 5 4 7 8 7 9 2 0 0 8] - [ 3 22 2 2 5 938 7 29 2 5 3 9 4 18 1 2 5 3 0 4 6] - [ 3 2 22 1 0 2 1031 5 0 3 1 0 1 2 0 7 1 3 4 7 4] - [ 2 9 17 3 2 30 8 894 1 3 2 7 6 3 2 0 2 3 16 11 3] - [ 22 1 2 1 1 0 0 1 839 50 13 2 2 17 41 2 2 4 7 0 2] - [ 92 1 4 0 5 1 1 1 30 836 0 1 1 20 9 1 1 0 0 1 7] - [ 2 4 9 15 1 1 7 5 14 3 949 2 2 13 5 1 1 0 15 0 2] - [ 1 2 1 0 4 9 2 3 0 0 0 881 46 14 1 15 5 8 3 7 3] - [ 1 0 1 8 2 4 0 0 1 0 0 32 858 3 3 13 4 21 0 1 7] - [ 6 3 4 0 3 3 1 3 11 12 9 5 4 937 6 7 3 3 0 1 3] - [ 10 4 7 15 5 1 0 2 18 3 2 1 3 4 986 3 0 8 10 1 9] - [ 4 1 6 1 6 0 2 1 0 0 0 6 8 5 2 984 4 6 2 3 5] - [ 6 6 4 0 6 3 1 0 1 0 1 4 2 5 1 23 980 0 0 6 12] - [ 7 1 2 4 0 2 0 1 0 0 0 10 22 1 1 30 1 964 0 0 5] - [ 3 7 13 17 3 1 0 39 6 0 3 4 8 0 14 2 1 1 960 3 1] - [ 0 2 1 1 2 4 14 14 1 0 1 16 7 6 0 10 6 4 0 1050 9] - [ 152 267 343 146 98 191 104 181 98 95 192 122 356 398 202 192 238 110 170 243 9536]] - -2023-02-13 17:55:24,461 - ==> Best [Top1: 82.761 Top5: 97.392 Sparsity:0.00 Params: 148928 on epoch: 82] -2023-02-13 17:55:24,462 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:55:24,467 - - -2023-02-13 17:55:24,468 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:55:25,361 - Epoch: [84][ 10/ 1207] Overall Loss 0.324140 Objective Loss 0.324140 LR 0.001000 Time 0.089244 -2023-02-13 17:55:25,554 - Epoch: [84][ 20/ 1207] Overall Loss 0.311342 Objective Loss 0.311342 LR 0.001000 Time 0.054288 -2023-02-13 17:55:25,743 - Epoch: [84][ 30/ 1207] Overall Loss 0.311995 Objective Loss 0.311995 LR 0.001000 Time 0.042475 -2023-02-13 17:55:25,931 - Epoch: [84][ 40/ 1207] Overall Loss 0.317978 Objective Loss 0.317978 LR 0.001000 Time 0.036549 -2023-02-13 17:55:26,118 - Epoch: [84][ 50/ 1207] Overall Loss 0.313729 Objective Loss 0.313729 LR 0.001000 Time 0.032975 -2023-02-13 17:55:26,305 - Epoch: [84][ 60/ 1207] Overall Loss 0.315493 Objective Loss 0.315493 LR 0.001000 Time 0.030582 -2023-02-13 17:55:26,491 - Epoch: [84][ 70/ 1207] Overall Loss 0.317662 Objective Loss 0.317662 LR 0.001000 Time 0.028871 -2023-02-13 17:55:26,678 - Epoch: [84][ 80/ 1207] Overall Loss 0.320557 Objective Loss 0.320557 LR 0.001000 Time 0.027592 -2023-02-13 17:55:26,866 - Epoch: [84][ 90/ 1207] Overall Loss 0.321114 Objective Loss 0.321114 LR 0.001000 Time 0.026607 -2023-02-13 17:55:27,053 - Epoch: [84][ 100/ 1207] Overall Loss 0.319221 Objective Loss 0.319221 LR 0.001000 Time 0.025809 -2023-02-13 17:55:27,239 - Epoch: [84][ 110/ 1207] Overall Loss 0.318562 Objective Loss 0.318562 LR 0.001000 Time 0.025156 -2023-02-13 17:55:27,426 - Epoch: [84][ 120/ 1207] Overall Loss 0.321132 Objective Loss 0.321132 LR 0.001000 Time 0.024610 -2023-02-13 17:55:27,612 - Epoch: [84][ 130/ 1207] Overall Loss 0.320616 Objective Loss 0.320616 LR 0.001000 Time 0.024152 -2023-02-13 17:55:27,799 - Epoch: [84][ 140/ 1207] Overall Loss 0.319201 Objective Loss 0.319201 LR 0.001000 Time 0.023758 -2023-02-13 17:55:27,986 - Epoch: [84][ 150/ 1207] Overall Loss 0.320081 Objective Loss 0.320081 LR 0.001000 Time 0.023416 -2023-02-13 17:55:28,173 - Epoch: [84][ 160/ 1207] Overall Loss 0.318364 Objective Loss 0.318364 LR 0.001000 Time 0.023119 -2023-02-13 17:55:28,360 - Epoch: [84][ 170/ 1207] Overall Loss 0.321131 Objective Loss 0.321131 LR 0.001000 Time 0.022857 -2023-02-13 17:55:28,546 - Epoch: [84][ 180/ 1207] Overall Loss 0.319732 Objective Loss 0.319732 LR 0.001000 Time 0.022618 -2023-02-13 17:55:28,734 - Epoch: [84][ 190/ 1207] Overall Loss 0.319800 Objective Loss 0.319800 LR 0.001000 Time 0.022416 -2023-02-13 17:55:28,920 - Epoch: [84][ 200/ 1207] Overall Loss 0.319907 Objective Loss 0.319907 LR 0.001000 Time 0.022224 -2023-02-13 17:55:29,107 - Epoch: [84][ 210/ 1207] Overall Loss 0.322074 Objective Loss 0.322074 LR 0.001000 Time 0.022054 -2023-02-13 17:55:29,294 - Epoch: [84][ 220/ 1207] Overall Loss 0.323004 Objective Loss 0.323004 LR 0.001000 Time 0.021900 -2023-02-13 17:55:29,481 - Epoch: [84][ 230/ 1207] Overall Loss 0.322934 Objective Loss 0.322934 LR 0.001000 Time 0.021757 -2023-02-13 17:55:29,667 - Epoch: [84][ 240/ 1207] Overall Loss 0.322985 Objective Loss 0.322985 LR 0.001000 Time 0.021626 -2023-02-13 17:55:29,855 - Epoch: [84][ 250/ 1207] Overall Loss 0.322082 Objective Loss 0.322082 LR 0.001000 Time 0.021510 -2023-02-13 17:55:30,041 - Epoch: [84][ 260/ 1207] Overall Loss 0.321625 Objective Loss 0.321625 LR 0.001000 Time 0.021397 -2023-02-13 17:55:30,228 - Epoch: [84][ 270/ 1207] Overall Loss 0.322363 Objective Loss 0.322363 LR 0.001000 Time 0.021298 -2023-02-13 17:55:30,415 - Epoch: [84][ 280/ 1207] Overall Loss 0.323013 Objective Loss 0.323013 LR 0.001000 Time 0.021203 -2023-02-13 17:55:30,603 - Epoch: [84][ 290/ 1207] Overall Loss 0.324074 Objective Loss 0.324074 LR 0.001000 Time 0.021116 -2023-02-13 17:55:30,790 - Epoch: [84][ 300/ 1207] Overall Loss 0.324414 Objective Loss 0.324414 LR 0.001000 Time 0.021037 -2023-02-13 17:55:30,978 - Epoch: [84][ 310/ 1207] Overall Loss 0.324387 Objective Loss 0.324387 LR 0.001000 Time 0.020964 -2023-02-13 17:55:31,166 - Epoch: [84][ 320/ 1207] Overall Loss 0.325558 Objective Loss 0.325558 LR 0.001000 Time 0.020893 -2023-02-13 17:55:31,353 - Epoch: [84][ 330/ 1207] Overall Loss 0.324848 Objective Loss 0.324848 LR 0.001000 Time 0.020827 -2023-02-13 17:55:31,540 - Epoch: [84][ 340/ 1207] Overall Loss 0.324075 Objective Loss 0.324075 LR 0.001000 Time 0.020764 -2023-02-13 17:55:31,729 - Epoch: [84][ 350/ 1207] Overall Loss 0.324119 Objective Loss 0.324119 LR 0.001000 Time 0.020707 -2023-02-13 17:55:31,916 - Epoch: [84][ 360/ 1207] Overall Loss 0.324576 Objective Loss 0.324576 LR 0.001000 Time 0.020652 -2023-02-13 17:55:32,103 - Epoch: [84][ 370/ 1207] Overall Loss 0.323830 Objective Loss 0.323830 LR 0.001000 Time 0.020598 -2023-02-13 17:55:32,290 - Epoch: [84][ 380/ 1207] Overall Loss 0.324457 Objective Loss 0.324457 LR 0.001000 Time 0.020547 -2023-02-13 17:55:32,477 - Epoch: [84][ 390/ 1207] Overall Loss 0.323955 Objective Loss 0.323955 LR 0.001000 Time 0.020499 -2023-02-13 17:55:32,664 - Epoch: [84][ 400/ 1207] Overall Loss 0.324622 Objective Loss 0.324622 LR 0.001000 Time 0.020453 -2023-02-13 17:55:32,852 - Epoch: [84][ 410/ 1207] Overall Loss 0.325098 Objective Loss 0.325098 LR 0.001000 Time 0.020412 -2023-02-13 17:55:33,040 - Epoch: [84][ 420/ 1207] Overall Loss 0.325542 Objective Loss 0.325542 LR 0.001000 Time 0.020371 -2023-02-13 17:55:33,227 - Epoch: [84][ 430/ 1207] Overall Loss 0.325781 Objective Loss 0.325781 LR 0.001000 Time 0.020331 -2023-02-13 17:55:33,413 - Epoch: [84][ 440/ 1207] Overall Loss 0.325954 Objective Loss 0.325954 LR 0.001000 Time 0.020293 -2023-02-13 17:55:33,601 - Epoch: [84][ 450/ 1207] Overall Loss 0.326007 Objective Loss 0.326007 LR 0.001000 Time 0.020258 -2023-02-13 17:55:33,789 - Epoch: [84][ 460/ 1207] Overall Loss 0.326218 Objective Loss 0.326218 LR 0.001000 Time 0.020225 -2023-02-13 17:55:33,976 - Epoch: [84][ 470/ 1207] Overall Loss 0.326386 Objective Loss 0.326386 LR 0.001000 Time 0.020192 -2023-02-13 17:55:34,163 - Epoch: [84][ 480/ 1207] Overall Loss 0.326114 Objective Loss 0.326114 LR 0.001000 Time 0.020161 -2023-02-13 17:55:34,351 - Epoch: [84][ 490/ 1207] Overall Loss 0.325724 Objective Loss 0.325724 LR 0.001000 Time 0.020131 -2023-02-13 17:55:34,538 - Epoch: [84][ 500/ 1207] Overall Loss 0.325593 Objective Loss 0.325593 LR 0.001000 Time 0.020102 -2023-02-13 17:55:34,726 - Epoch: [84][ 510/ 1207] Overall Loss 0.325617 Objective Loss 0.325617 LR 0.001000 Time 0.020076 -2023-02-13 17:55:34,913 - Epoch: [84][ 520/ 1207] Overall Loss 0.325863 Objective Loss 0.325863 LR 0.001000 Time 0.020049 -2023-02-13 17:55:35,100 - Epoch: [84][ 530/ 1207] Overall Loss 0.326433 Objective Loss 0.326433 LR 0.001000 Time 0.020023 -2023-02-13 17:55:35,287 - Epoch: [84][ 540/ 1207] Overall Loss 0.326591 Objective Loss 0.326591 LR 0.001000 Time 0.019998 -2023-02-13 17:55:35,474 - Epoch: [84][ 550/ 1207] Overall Loss 0.326445 Objective Loss 0.326445 LR 0.001000 Time 0.019974 -2023-02-13 17:55:35,662 - Epoch: [84][ 560/ 1207] Overall Loss 0.326044 Objective Loss 0.326044 LR 0.001000 Time 0.019951 -2023-02-13 17:55:35,851 - Epoch: [84][ 570/ 1207] Overall Loss 0.326291 Objective Loss 0.326291 LR 0.001000 Time 0.019933 -2023-02-13 17:55:36,040 - Epoch: [84][ 580/ 1207] Overall Loss 0.325934 Objective Loss 0.325934 LR 0.001000 Time 0.019914 -2023-02-13 17:55:36,229 - Epoch: [84][ 590/ 1207] Overall Loss 0.326370 Objective Loss 0.326370 LR 0.001000 Time 0.019896 -2023-02-13 17:55:36,417 - Epoch: [84][ 600/ 1207] Overall Loss 0.326837 Objective Loss 0.326837 LR 0.001000 Time 0.019877 -2023-02-13 17:55:36,605 - Epoch: [84][ 610/ 1207] Overall Loss 0.326320 Objective Loss 0.326320 LR 0.001000 Time 0.019860 -2023-02-13 17:55:36,794 - Epoch: [84][ 620/ 1207] Overall Loss 0.326425 Objective Loss 0.326425 LR 0.001000 Time 0.019844 -2023-02-13 17:55:36,983 - Epoch: [84][ 630/ 1207] Overall Loss 0.327017 Objective Loss 0.327017 LR 0.001000 Time 0.019828 -2023-02-13 17:55:37,172 - Epoch: [84][ 640/ 1207] Overall Loss 0.327160 Objective Loss 0.327160 LR 0.001000 Time 0.019812 -2023-02-13 17:55:37,360 - Epoch: [84][ 650/ 1207] Overall Loss 0.327810 Objective Loss 0.327810 LR 0.001000 Time 0.019796 -2023-02-13 17:55:37,548 - Epoch: [84][ 660/ 1207] Overall Loss 0.327155 Objective Loss 0.327155 LR 0.001000 Time 0.019781 -2023-02-13 17:55:37,737 - Epoch: [84][ 670/ 1207] Overall Loss 0.327010 Objective Loss 0.327010 LR 0.001000 Time 0.019767 -2023-02-13 17:55:37,926 - Epoch: [84][ 680/ 1207] Overall Loss 0.327045 Objective Loss 0.327045 LR 0.001000 Time 0.019753 -2023-02-13 17:55:38,114 - Epoch: [84][ 690/ 1207] Overall Loss 0.327503 Objective Loss 0.327503 LR 0.001000 Time 0.019739 -2023-02-13 17:55:38,302 - Epoch: [84][ 700/ 1207] Overall Loss 0.328189 Objective Loss 0.328189 LR 0.001000 Time 0.019726 -2023-02-13 17:55:38,491 - Epoch: [84][ 710/ 1207] Overall Loss 0.328178 Objective Loss 0.328178 LR 0.001000 Time 0.019713 -2023-02-13 17:55:38,679 - Epoch: [84][ 720/ 1207] Overall Loss 0.327975 Objective Loss 0.327975 LR 0.001000 Time 0.019700 -2023-02-13 17:55:38,868 - Epoch: [84][ 730/ 1207] Overall Loss 0.328243 Objective Loss 0.328243 LR 0.001000 Time 0.019689 -2023-02-13 17:55:39,057 - Epoch: [84][ 740/ 1207] Overall Loss 0.328483 Objective Loss 0.328483 LR 0.001000 Time 0.019677 -2023-02-13 17:55:39,246 - Epoch: [84][ 750/ 1207] Overall Loss 0.328779 Objective Loss 0.328779 LR 0.001000 Time 0.019666 -2023-02-13 17:55:39,434 - Epoch: [84][ 760/ 1207] Overall Loss 0.329096 Objective Loss 0.329096 LR 0.001000 Time 0.019654 -2023-02-13 17:55:39,622 - Epoch: [84][ 770/ 1207] Overall Loss 0.328996 Objective Loss 0.328996 LR 0.001000 Time 0.019643 -2023-02-13 17:55:39,811 - Epoch: [84][ 780/ 1207] Overall Loss 0.328986 Objective Loss 0.328986 LR 0.001000 Time 0.019633 -2023-02-13 17:55:40,000 - Epoch: [84][ 790/ 1207] Overall Loss 0.328724 Objective Loss 0.328724 LR 0.001000 Time 0.019623 -2023-02-13 17:55:40,188 - Epoch: [84][ 800/ 1207] Overall Loss 0.328911 Objective Loss 0.328911 LR 0.001000 Time 0.019613 -2023-02-13 17:55:40,377 - Epoch: [84][ 810/ 1207] Overall Loss 0.328567 Objective Loss 0.328567 LR 0.001000 Time 0.019603 -2023-02-13 17:55:40,565 - Epoch: [84][ 820/ 1207] Overall Loss 0.329112 Objective Loss 0.329112 LR 0.001000 Time 0.019592 -2023-02-13 17:55:40,753 - Epoch: [84][ 830/ 1207] Overall Loss 0.328953 Objective Loss 0.328953 LR 0.001000 Time 0.019583 -2023-02-13 17:55:40,943 - Epoch: [84][ 840/ 1207] Overall Loss 0.329258 Objective Loss 0.329258 LR 0.001000 Time 0.019575 -2023-02-13 17:55:41,131 - Epoch: [84][ 850/ 1207] Overall Loss 0.329280 Objective Loss 0.329280 LR 0.001000 Time 0.019566 -2023-02-13 17:55:41,319 - Epoch: [84][ 860/ 1207] Overall Loss 0.329513 Objective Loss 0.329513 LR 0.001000 Time 0.019557 -2023-02-13 17:55:41,507 - Epoch: [84][ 870/ 1207] Overall Loss 0.329507 Objective Loss 0.329507 LR 0.001000 Time 0.019547 -2023-02-13 17:55:41,695 - Epoch: [84][ 880/ 1207] Overall Loss 0.329204 Objective Loss 0.329204 LR 0.001000 Time 0.019538 -2023-02-13 17:55:41,883 - Epoch: [84][ 890/ 1207] Overall Loss 0.329149 Objective Loss 0.329149 LR 0.001000 Time 0.019530 -2023-02-13 17:55:42,071 - Epoch: [84][ 900/ 1207] Overall Loss 0.329045 Objective Loss 0.329045 LR 0.001000 Time 0.019522 -2023-02-13 17:55:42,261 - Epoch: [84][ 910/ 1207] Overall Loss 0.328914 Objective Loss 0.328914 LR 0.001000 Time 0.019515 -2023-02-13 17:55:42,448 - Epoch: [84][ 920/ 1207] Overall Loss 0.328901 Objective Loss 0.328901 LR 0.001000 Time 0.019506 -2023-02-13 17:55:42,636 - Epoch: [84][ 930/ 1207] Overall Loss 0.328905 Objective Loss 0.328905 LR 0.001000 Time 0.019499 -2023-02-13 17:55:42,825 - Epoch: [84][ 940/ 1207] Overall Loss 0.329343 Objective Loss 0.329343 LR 0.001000 Time 0.019492 -2023-02-13 17:55:43,014 - Epoch: [84][ 950/ 1207] Overall Loss 0.329699 Objective Loss 0.329699 LR 0.001000 Time 0.019484 -2023-02-13 17:55:43,202 - Epoch: [84][ 960/ 1207] Overall Loss 0.329943 Objective Loss 0.329943 LR 0.001000 Time 0.019478 -2023-02-13 17:55:43,390 - Epoch: [84][ 970/ 1207] Overall Loss 0.329868 Objective Loss 0.329868 LR 0.001000 Time 0.019470 -2023-02-13 17:55:43,578 - Epoch: [84][ 980/ 1207] Overall Loss 0.329633 Objective Loss 0.329633 LR 0.001000 Time 0.019463 -2023-02-13 17:55:43,767 - Epoch: [84][ 990/ 1207] Overall Loss 0.329631 Objective Loss 0.329631 LR 0.001000 Time 0.019457 -2023-02-13 17:55:43,955 - Epoch: [84][ 1000/ 1207] Overall Loss 0.329777 Objective Loss 0.329777 LR 0.001000 Time 0.019450 -2023-02-13 17:55:44,144 - Epoch: [84][ 1010/ 1207] Overall Loss 0.330423 Objective Loss 0.330423 LR 0.001000 Time 0.019444 -2023-02-13 17:55:44,331 - Epoch: [84][ 1020/ 1207] Overall Loss 0.330370 Objective Loss 0.330370 LR 0.001000 Time 0.019436 -2023-02-13 17:55:44,519 - Epoch: [84][ 1030/ 1207] Overall Loss 0.330388 Objective Loss 0.330388 LR 0.001000 Time 0.019430 -2023-02-13 17:55:44,707 - Epoch: [84][ 1040/ 1207] Overall Loss 0.330290 Objective Loss 0.330290 LR 0.001000 Time 0.019423 -2023-02-13 17:55:44,896 - Epoch: [84][ 1050/ 1207] Overall Loss 0.330489 Objective Loss 0.330489 LR 0.001000 Time 0.019418 -2023-02-13 17:55:45,085 - Epoch: [84][ 1060/ 1207] Overall Loss 0.330395 Objective Loss 0.330395 LR 0.001000 Time 0.019412 -2023-02-13 17:55:45,274 - Epoch: [84][ 1070/ 1207] Overall Loss 0.330256 Objective Loss 0.330256 LR 0.001000 Time 0.019407 -2023-02-13 17:55:45,461 - Epoch: [84][ 1080/ 1207] Overall Loss 0.330585 Objective Loss 0.330585 LR 0.001000 Time 0.019401 -2023-02-13 17:55:45,649 - Epoch: [84][ 1090/ 1207] Overall Loss 0.331151 Objective Loss 0.331151 LR 0.001000 Time 0.019395 -2023-02-13 17:55:45,839 - Epoch: [84][ 1100/ 1207] Overall Loss 0.331691 Objective Loss 0.331691 LR 0.001000 Time 0.019391 -2023-02-13 17:55:46,028 - Epoch: [84][ 1110/ 1207] Overall Loss 0.332081 Objective Loss 0.332081 LR 0.001000 Time 0.019386 -2023-02-13 17:55:46,217 - Epoch: [84][ 1120/ 1207] Overall Loss 0.332069 Objective Loss 0.332069 LR 0.001000 Time 0.019382 -2023-02-13 17:55:46,406 - Epoch: [84][ 1130/ 1207] Overall Loss 0.332182 Objective Loss 0.332182 LR 0.001000 Time 0.019376 -2023-02-13 17:55:46,594 - Epoch: [84][ 1140/ 1207] Overall Loss 0.332385 Objective Loss 0.332385 LR 0.001000 Time 0.019371 -2023-02-13 17:55:46,783 - Epoch: [84][ 1150/ 1207] Overall Loss 0.331966 Objective Loss 0.331966 LR 0.001000 Time 0.019367 -2023-02-13 17:55:46,973 - Epoch: [84][ 1160/ 1207] Overall Loss 0.332202 Objective Loss 0.332202 LR 0.001000 Time 0.019363 -2023-02-13 17:55:47,162 - Epoch: [84][ 1170/ 1207] Overall Loss 0.332114 Objective Loss 0.332114 LR 0.001000 Time 0.019359 -2023-02-13 17:55:47,350 - Epoch: [84][ 1180/ 1207] Overall Loss 0.332090 Objective Loss 0.332090 LR 0.001000 Time 0.019354 -2023-02-13 17:55:47,539 - Epoch: [84][ 1190/ 1207] Overall Loss 0.332129 Objective Loss 0.332129 LR 0.001000 Time 0.019350 -2023-02-13 17:55:47,784 - Epoch: [84][ 1200/ 1207] Overall Loss 0.332061 Objective Loss 0.332061 LR 0.001000 Time 0.019393 -2023-02-13 17:55:47,899 - Epoch: [84][ 1207/ 1207] Overall Loss 0.332076 Objective Loss 0.332076 Top1 83.536585 Top5 98.780488 LR 0.001000 Time 0.019375 -2023-02-13 17:55:47,971 - --- validate (epoch=84)----------- -2023-02-13 17:55:47,971 - 34311 samples (256 per mini-batch) -2023-02-13 17:55:48,391 - Epoch: [84][ 10/ 135] Loss 0.358067 Top1 82.382812 Top5 97.226562 -2023-02-13 17:55:48,537 - Epoch: [84][ 20/ 135] Loss 0.365901 Top1 82.617188 Top5 97.148438 -2023-02-13 17:55:48,673 - Epoch: [84][ 30/ 135] Loss 0.372991 Top1 82.356771 Top5 97.044271 -2023-02-13 17:55:48,814 - Epoch: [84][ 40/ 135] Loss 0.364487 Top1 82.558594 Top5 97.177734 -2023-02-13 17:55:48,937 - Epoch: [84][ 50/ 135] Loss 0.368972 Top1 82.289062 Top5 97.195312 -2023-02-13 17:55:49,061 - Epoch: [84][ 60/ 135] Loss 0.371560 Top1 82.350260 Top5 97.233073 -2023-02-13 17:55:49,187 - Epoch: [84][ 70/ 135] Loss 0.371608 Top1 82.416295 Top5 97.170759 -2023-02-13 17:55:49,324 - Epoch: [84][ 80/ 135] Loss 0.369638 Top1 82.495117 Top5 97.163086 -2023-02-13 17:55:49,469 - Epoch: [84][ 90/ 135] Loss 0.367400 Top1 82.547743 Top5 97.196181 -2023-02-13 17:55:49,606 - Epoch: [84][ 100/ 135] Loss 0.366370 Top1 82.464844 Top5 97.199219 -2023-02-13 17:55:49,743 - Epoch: [84][ 110/ 135] Loss 0.370002 Top1 82.333097 Top5 97.176847 -2023-02-13 17:55:49,875 - Epoch: [84][ 120/ 135] Loss 0.371847 Top1 82.268880 Top5 97.148438 -2023-02-13 17:55:50,007 - Epoch: [84][ 130/ 135] Loss 0.369522 Top1 82.268630 Top5 97.169471 -2023-02-13 17:55:50,051 - Epoch: [84][ 135/ 135] Loss 0.369005 Top1 82.203958 Top5 97.149602 -2023-02-13 17:55:50,123 - ==> Top1: 82.204 Top5: 97.150 Loss: 0.369 - -2023-02-13 17:55:50,124 - ==> Confusion: -[[ 843 3 4 1 8 4 0 2 4 53 4 7 2 3 4 4 5 4 0 7 5] - [ 1 917 2 2 6 44 2 18 3 3 3 3 3 2 0 0 7 1 6 2 8] - [ 7 2 915 14 3 4 42 11 0 2 2 2 6 2 3 6 2 5 12 3 15] - [ 5 3 19 884 0 4 4 2 3 2 20 1 8 3 13 2 4 11 20 1 7] - [ 18 11 2 0 967 9 1 0 4 5 2 7 2 3 6 6 13 3 0 4 3] - [ 3 19 0 3 5 954 7 19 5 4 4 11 5 10 1 1 3 2 3 3 8] - [ 1 5 9 3 0 4 1037 5 0 2 4 0 1 0 0 5 4 3 1 12 3] - [ 2 11 9 4 2 36 9 890 1 0 4 6 4 2 0 0 2 1 24 12 5] - [ 18 3 0 3 1 2 0 2 881 36 19 4 0 16 10 0 2 3 5 2 2] - [ 82 3 2 0 9 1 0 2 41 833 2 0 2 20 2 1 2 3 2 1 4] - [ 3 7 5 9 0 1 4 5 12 2 970 2 4 8 1 0 0 0 14 0 4] - [ 1 3 0 1 2 9 1 2 0 1 1 915 35 5 0 2 4 12 2 9 0] - [ 2 1 1 9 0 3 0 0 1 1 2 24 873 0 1 3 3 23 1 2 9] - [ 5 4 1 0 5 12 0 0 11 22 12 10 6 902 5 3 11 5 0 5 5] - [ 9 3 1 23 4 6 0 2 36 7 9 1 11 3 946 2 1 8 8 1 11] - [ 6 2 3 2 4 2 7 1 0 0 0 11 12 2 1 941 20 14 0 9 9] - [ 2 7 0 1 7 2 1 2 1 1 1 5 2 1 0 3 1006 3 1 4 11] - [ 4 3 2 1 1 0 0 1 2 0 0 13 19 0 3 10 1 985 0 4 2] - [ 3 6 4 15 2 3 1 23 9 0 9 1 5 1 15 1 0 1 982 2 3] - [ 1 5 1 0 1 10 11 10 0 0 2 19 2 6 0 3 9 5 3 1050 10] - [ 175 257 184 146 146 263 122 172 127 94 237 153 350 332 150 88 319 165 172 268 9514]] - -2023-02-13 17:55:50,126 - ==> Best [Top1: 82.761 Top5: 97.392 Sparsity:0.00 Params: 148928 on epoch: 82] -2023-02-13 17:55:50,126 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:55:50,131 - - -2023-02-13 17:55:50,131 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:55:51,004 - Epoch: [85][ 10/ 1207] Overall Loss 0.379390 Objective Loss 0.379390 LR 0.001000 Time 0.087202 -2023-02-13 17:55:51,210 - Epoch: [85][ 20/ 1207] Overall Loss 0.376182 Objective Loss 0.376182 LR 0.001000 Time 0.053892 -2023-02-13 17:55:51,406 - Epoch: [85][ 30/ 1207] Overall Loss 0.359463 Objective Loss 0.359463 LR 0.001000 Time 0.042449 -2023-02-13 17:55:51,604 - Epoch: [85][ 40/ 1207] Overall Loss 0.354927 Objective Loss 0.354927 LR 0.001000 Time 0.036754 -2023-02-13 17:55:51,799 - Epoch: [85][ 50/ 1207] Overall Loss 0.352807 Objective Loss 0.352807 LR 0.001000 Time 0.033301 -2023-02-13 17:55:51,997 - Epoch: [85][ 60/ 1207] Overall Loss 0.349896 Objective Loss 0.349896 LR 0.001000 Time 0.031043 -2023-02-13 17:55:52,192 - Epoch: [85][ 70/ 1207] Overall Loss 0.344430 Objective Loss 0.344430 LR 0.001000 Time 0.029389 -2023-02-13 17:55:52,389 - Epoch: [85][ 80/ 1207] Overall Loss 0.340090 Objective Loss 0.340090 LR 0.001000 Time 0.028182 -2023-02-13 17:55:52,585 - Epoch: [85][ 90/ 1207] Overall Loss 0.337890 Objective Loss 0.337890 LR 0.001000 Time 0.027216 -2023-02-13 17:55:52,782 - Epoch: [85][ 100/ 1207] Overall Loss 0.334962 Objective Loss 0.334962 LR 0.001000 Time 0.026469 -2023-02-13 17:55:52,978 - Epoch: [85][ 110/ 1207] Overall Loss 0.332692 Objective Loss 0.332692 LR 0.001000 Time 0.025835 -2023-02-13 17:55:53,175 - Epoch: [85][ 120/ 1207] Overall Loss 0.333181 Objective Loss 0.333181 LR 0.001000 Time 0.025325 -2023-02-13 17:55:53,370 - Epoch: [85][ 130/ 1207] Overall Loss 0.337257 Objective Loss 0.337257 LR 0.001000 Time 0.024872 -2023-02-13 17:55:53,567 - Epoch: [85][ 140/ 1207] Overall Loss 0.336945 Objective Loss 0.336945 LR 0.001000 Time 0.024503 -2023-02-13 17:55:53,763 - Epoch: [85][ 150/ 1207] Overall Loss 0.336522 Objective Loss 0.336522 LR 0.001000 Time 0.024169 -2023-02-13 17:55:53,959 - Epoch: [85][ 160/ 1207] Overall Loss 0.335611 Objective Loss 0.335611 LR 0.001000 Time 0.023884 -2023-02-13 17:55:54,151 - Epoch: [85][ 170/ 1207] Overall Loss 0.333063 Objective Loss 0.333063 LR 0.001000 Time 0.023607 -2023-02-13 17:55:54,347 - Epoch: [85][ 180/ 1207] Overall Loss 0.333379 Objective Loss 0.333379 LR 0.001000 Time 0.023379 -2023-02-13 17:55:54,539 - Epoch: [85][ 190/ 1207] Overall Loss 0.330935 Objective Loss 0.330935 LR 0.001000 Time 0.023156 -2023-02-13 17:55:54,734 - Epoch: [85][ 200/ 1207] Overall Loss 0.330056 Objective Loss 0.330056 LR 0.001000 Time 0.022972 -2023-02-13 17:55:54,926 - Epoch: [85][ 210/ 1207] Overall Loss 0.329577 Objective Loss 0.329577 LR 0.001000 Time 0.022792 -2023-02-13 17:55:55,121 - Epoch: [85][ 220/ 1207] Overall Loss 0.328414 Objective Loss 0.328414 LR 0.001000 Time 0.022639 -2023-02-13 17:55:55,313 - Epoch: [85][ 230/ 1207] Overall Loss 0.328532 Objective Loss 0.328532 LR 0.001000 Time 0.022491 -2023-02-13 17:55:55,509 - Epoch: [85][ 240/ 1207] Overall Loss 0.329214 Objective Loss 0.329214 LR 0.001000 Time 0.022367 -2023-02-13 17:55:55,701 - Epoch: [85][ 250/ 1207] Overall Loss 0.330408 Objective Loss 0.330408 LR 0.001000 Time 0.022240 -2023-02-13 17:55:55,898 - Epoch: [85][ 260/ 1207] Overall Loss 0.330122 Objective Loss 0.330122 LR 0.001000 Time 0.022140 -2023-02-13 17:55:56,091 - Epoch: [85][ 270/ 1207] Overall Loss 0.330035 Objective Loss 0.330035 LR 0.001000 Time 0.022032 -2023-02-13 17:55:56,286 - Epoch: [85][ 280/ 1207] Overall Loss 0.330064 Objective Loss 0.330064 LR 0.001000 Time 0.021941 -2023-02-13 17:55:56,478 - Epoch: [85][ 290/ 1207] Overall Loss 0.329707 Objective Loss 0.329707 LR 0.001000 Time 0.021847 -2023-02-13 17:55:56,674 - Epoch: [85][ 300/ 1207] Overall Loss 0.329353 Objective Loss 0.329353 LR 0.001000 Time 0.021769 -2023-02-13 17:55:56,867 - Epoch: [85][ 310/ 1207] Overall Loss 0.327995 Objective Loss 0.327995 LR 0.001000 Time 0.021690 -2023-02-13 17:55:57,062 - Epoch: [85][ 320/ 1207] Overall Loss 0.329069 Objective Loss 0.329069 LR 0.001000 Time 0.021620 -2023-02-13 17:55:57,255 - Epoch: [85][ 330/ 1207] Overall Loss 0.328995 Objective Loss 0.328995 LR 0.001000 Time 0.021547 -2023-02-13 17:55:57,450 - Epoch: [85][ 340/ 1207] Overall Loss 0.329395 Objective Loss 0.329395 LR 0.001000 Time 0.021486 -2023-02-13 17:55:57,642 - Epoch: [85][ 350/ 1207] Overall Loss 0.330819 Objective Loss 0.330819 LR 0.001000 Time 0.021419 -2023-02-13 17:55:57,837 - Epoch: [85][ 360/ 1207] Overall Loss 0.331104 Objective Loss 0.331104 LR 0.001000 Time 0.021366 -2023-02-13 17:55:58,029 - Epoch: [85][ 370/ 1207] Overall Loss 0.331683 Objective Loss 0.331683 LR 0.001000 Time 0.021306 -2023-02-13 17:55:58,225 - Epoch: [85][ 380/ 1207] Overall Loss 0.331508 Objective Loss 0.331508 LR 0.001000 Time 0.021260 -2023-02-13 17:55:58,416 - Epoch: [85][ 390/ 1207] Overall Loss 0.331082 Objective Loss 0.331082 LR 0.001000 Time 0.021205 -2023-02-13 17:55:58,612 - Epoch: [85][ 400/ 1207] Overall Loss 0.330269 Objective Loss 0.330269 LR 0.001000 Time 0.021163 -2023-02-13 17:55:58,806 - Epoch: [85][ 410/ 1207] Overall Loss 0.329599 Objective Loss 0.329599 LR 0.001000 Time 0.021120 -2023-02-13 17:55:59,003 - Epoch: [85][ 420/ 1207] Overall Loss 0.329637 Objective Loss 0.329637 LR 0.001000 Time 0.021085 -2023-02-13 17:55:59,197 - Epoch: [85][ 430/ 1207] Overall Loss 0.329722 Objective Loss 0.329722 LR 0.001000 Time 0.021043 -2023-02-13 17:55:59,393 - Epoch: [85][ 440/ 1207] Overall Loss 0.330685 Objective Loss 0.330685 LR 0.001000 Time 0.021011 -2023-02-13 17:55:59,587 - Epoch: [85][ 450/ 1207] Overall Loss 0.331433 Objective Loss 0.331433 LR 0.001000 Time 0.020973 -2023-02-13 17:55:59,783 - Epoch: [85][ 460/ 1207] Overall Loss 0.332411 Objective Loss 0.332411 LR 0.001000 Time 0.020944 -2023-02-13 17:55:59,977 - Epoch: [85][ 470/ 1207] Overall Loss 0.332921 Objective Loss 0.332921 LR 0.001000 Time 0.020909 -2023-02-13 17:56:00,174 - Epoch: [85][ 480/ 1207] Overall Loss 0.333203 Objective Loss 0.333203 LR 0.001000 Time 0.020883 -2023-02-13 17:56:00,367 - Epoch: [85][ 490/ 1207] Overall Loss 0.333294 Objective Loss 0.333294 LR 0.001000 Time 0.020851 -2023-02-13 17:56:00,564 - Epoch: [85][ 500/ 1207] Overall Loss 0.333225 Objective Loss 0.333225 LR 0.001000 Time 0.020826 -2023-02-13 17:56:00,756 - Epoch: [85][ 510/ 1207] Overall Loss 0.333466 Objective Loss 0.333466 LR 0.001000 Time 0.020795 -2023-02-13 17:56:00,954 - Epoch: [85][ 520/ 1207] Overall Loss 0.333900 Objective Loss 0.333900 LR 0.001000 Time 0.020774 -2023-02-13 17:56:01,147 - Epoch: [85][ 530/ 1207] Overall Loss 0.334101 Objective Loss 0.334101 LR 0.001000 Time 0.020746 -2023-02-13 17:56:01,344 - Epoch: [85][ 540/ 1207] Overall Loss 0.334392 Objective Loss 0.334392 LR 0.001000 Time 0.020725 -2023-02-13 17:56:01,537 - Epoch: [85][ 550/ 1207] Overall Loss 0.334152 Objective Loss 0.334152 LR 0.001000 Time 0.020698 -2023-02-13 17:56:01,733 - Epoch: [85][ 560/ 1207] Overall Loss 0.334414 Objective Loss 0.334414 LR 0.001000 Time 0.020679 -2023-02-13 17:56:01,927 - Epoch: [85][ 570/ 1207] Overall Loss 0.334711 Objective Loss 0.334711 LR 0.001000 Time 0.020656 -2023-02-13 17:56:02,124 - Epoch: [85][ 580/ 1207] Overall Loss 0.334334 Objective Loss 0.334334 LR 0.001000 Time 0.020639 -2023-02-13 17:56:02,317 - Epoch: [85][ 590/ 1207] Overall Loss 0.334892 Objective Loss 0.334892 LR 0.001000 Time 0.020615 -2023-02-13 17:56:02,513 - Epoch: [85][ 600/ 1207] Overall Loss 0.335267 Objective Loss 0.335267 LR 0.001000 Time 0.020598 -2023-02-13 17:56:02,707 - Epoch: [85][ 610/ 1207] Overall Loss 0.336268 Objective Loss 0.336268 LR 0.001000 Time 0.020577 -2023-02-13 17:56:02,903 - Epoch: [85][ 620/ 1207] Overall Loss 0.336333 Objective Loss 0.336333 LR 0.001000 Time 0.020561 -2023-02-13 17:56:03,097 - Epoch: [85][ 630/ 1207] Overall Loss 0.336370 Objective Loss 0.336370 LR 0.001000 Time 0.020542 -2023-02-13 17:56:03,293 - Epoch: [85][ 640/ 1207] Overall Loss 0.336565 Objective Loss 0.336565 LR 0.001000 Time 0.020527 -2023-02-13 17:56:03,486 - Epoch: [85][ 650/ 1207] Overall Loss 0.336219 Objective Loss 0.336219 LR 0.001000 Time 0.020508 -2023-02-13 17:56:03,683 - Epoch: [85][ 660/ 1207] Overall Loss 0.336388 Objective Loss 0.336388 LR 0.001000 Time 0.020494 -2023-02-13 17:56:03,877 - Epoch: [85][ 670/ 1207] Overall Loss 0.336399 Objective Loss 0.336399 LR 0.001000 Time 0.020478 -2023-02-13 17:56:04,074 - Epoch: [85][ 680/ 1207] Overall Loss 0.336535 Objective Loss 0.336535 LR 0.001000 Time 0.020465 -2023-02-13 17:56:04,268 - Epoch: [85][ 690/ 1207] Overall Loss 0.336713 Objective Loss 0.336713 LR 0.001000 Time 0.020449 -2023-02-13 17:56:04,459 - Epoch: [85][ 700/ 1207] Overall Loss 0.337230 Objective Loss 0.337230 LR 0.001000 Time 0.020430 -2023-02-13 17:56:04,647 - Epoch: [85][ 710/ 1207] Overall Loss 0.337318 Objective Loss 0.337318 LR 0.001000 Time 0.020407 -2023-02-13 17:56:04,842 - Epoch: [85][ 720/ 1207] Overall Loss 0.337112 Objective Loss 0.337112 LR 0.001000 Time 0.020393 -2023-02-13 17:56:05,036 - Epoch: [85][ 730/ 1207] Overall Loss 0.336994 Objective Loss 0.336994 LR 0.001000 Time 0.020379 -2023-02-13 17:56:05,233 - Epoch: [85][ 740/ 1207] Overall Loss 0.337313 Objective Loss 0.337313 LR 0.001000 Time 0.020369 -2023-02-13 17:56:05,426 - Epoch: [85][ 750/ 1207] Overall Loss 0.337085 Objective Loss 0.337085 LR 0.001000 Time 0.020355 -2023-02-13 17:56:05,622 - Epoch: [85][ 760/ 1207] Overall Loss 0.336676 Objective Loss 0.336676 LR 0.001000 Time 0.020344 -2023-02-13 17:56:05,817 - Epoch: [85][ 770/ 1207] Overall Loss 0.336533 Objective Loss 0.336533 LR 0.001000 Time 0.020332 -2023-02-13 17:56:06,013 - Epoch: [85][ 780/ 1207] Overall Loss 0.336263 Objective Loss 0.336263 LR 0.001000 Time 0.020323 -2023-02-13 17:56:06,207 - Epoch: [85][ 790/ 1207] Overall Loss 0.336163 Objective Loss 0.336163 LR 0.001000 Time 0.020311 -2023-02-13 17:56:06,403 - Epoch: [85][ 800/ 1207] Overall Loss 0.336067 Objective Loss 0.336067 LR 0.001000 Time 0.020302 -2023-02-13 17:56:06,597 - Epoch: [85][ 810/ 1207] Overall Loss 0.335952 Objective Loss 0.335952 LR 0.001000 Time 0.020289 -2023-02-13 17:56:06,792 - Epoch: [85][ 820/ 1207] Overall Loss 0.335791 Objective Loss 0.335791 LR 0.001000 Time 0.020280 -2023-02-13 17:56:06,987 - Epoch: [85][ 830/ 1207] Overall Loss 0.335552 Objective Loss 0.335552 LR 0.001000 Time 0.020270 -2023-02-13 17:56:07,183 - Epoch: [85][ 840/ 1207] Overall Loss 0.335400 Objective Loss 0.335400 LR 0.001000 Time 0.020261 -2023-02-13 17:56:07,378 - Epoch: [85][ 850/ 1207] Overall Loss 0.335068 Objective Loss 0.335068 LR 0.001000 Time 0.020252 -2023-02-13 17:56:07,573 - Epoch: [85][ 860/ 1207] Overall Loss 0.335136 Objective Loss 0.335136 LR 0.001000 Time 0.020243 -2023-02-13 17:56:07,768 - Epoch: [85][ 870/ 1207] Overall Loss 0.335159 Objective Loss 0.335159 LR 0.001000 Time 0.020233 -2023-02-13 17:56:07,963 - Epoch: [85][ 880/ 1207] Overall Loss 0.334900 Objective Loss 0.334900 LR 0.001000 Time 0.020225 -2023-02-13 17:56:08,156 - Epoch: [85][ 890/ 1207] Overall Loss 0.334743 Objective Loss 0.334743 LR 0.001000 Time 0.020214 -2023-02-13 17:56:08,351 - Epoch: [85][ 900/ 1207] Overall Loss 0.335100 Objective Loss 0.335100 LR 0.001000 Time 0.020206 -2023-02-13 17:56:08,545 - Epoch: [85][ 910/ 1207] Overall Loss 0.335324 Objective Loss 0.335324 LR 0.001000 Time 0.020197 -2023-02-13 17:56:08,740 - Epoch: [85][ 920/ 1207] Overall Loss 0.335364 Objective Loss 0.335364 LR 0.001000 Time 0.020189 -2023-02-13 17:56:08,934 - Epoch: [85][ 930/ 1207] Overall Loss 0.335683 Objective Loss 0.335683 LR 0.001000 Time 0.020180 -2023-02-13 17:56:09,129 - Epoch: [85][ 940/ 1207] Overall Loss 0.335402 Objective Loss 0.335402 LR 0.001000 Time 0.020172 -2023-02-13 17:56:09,321 - Epoch: [85][ 950/ 1207] Overall Loss 0.335541 Objective Loss 0.335541 LR 0.001000 Time 0.020162 -2023-02-13 17:56:09,516 - Epoch: [85][ 960/ 1207] Overall Loss 0.335793 Objective Loss 0.335793 LR 0.001000 Time 0.020155 -2023-02-13 17:56:09,710 - Epoch: [85][ 970/ 1207] Overall Loss 0.335842 Objective Loss 0.335842 LR 0.001000 Time 0.020146 -2023-02-13 17:56:09,905 - Epoch: [85][ 980/ 1207] Overall Loss 0.335681 Objective Loss 0.335681 LR 0.001000 Time 0.020139 -2023-02-13 17:56:10,097 - Epoch: [85][ 990/ 1207] Overall Loss 0.335872 Objective Loss 0.335872 LR 0.001000 Time 0.020129 -2023-02-13 17:56:10,293 - Epoch: [85][ 1000/ 1207] Overall Loss 0.336150 Objective Loss 0.336150 LR 0.001000 Time 0.020123 -2023-02-13 17:56:10,486 - Epoch: [85][ 1010/ 1207] Overall Loss 0.336050 Objective Loss 0.336050 LR 0.001000 Time 0.020115 -2023-02-13 17:56:10,680 - Epoch: [85][ 1020/ 1207] Overall Loss 0.335726 Objective Loss 0.335726 LR 0.001000 Time 0.020108 -2023-02-13 17:56:10,874 - Epoch: [85][ 1030/ 1207] Overall Loss 0.335758 Objective Loss 0.335758 LR 0.001000 Time 0.020101 -2023-02-13 17:56:11,069 - Epoch: [85][ 1040/ 1207] Overall Loss 0.335699 Objective Loss 0.335699 LR 0.001000 Time 0.020095 -2023-02-13 17:56:11,263 - Epoch: [85][ 1050/ 1207] Overall Loss 0.336013 Objective Loss 0.336013 LR 0.001000 Time 0.020087 -2023-02-13 17:56:11,458 - Epoch: [85][ 1060/ 1207] Overall Loss 0.336023 Objective Loss 0.336023 LR 0.001000 Time 0.020081 -2023-02-13 17:56:11,651 - Epoch: [85][ 1070/ 1207] Overall Loss 0.336209 Objective Loss 0.336209 LR 0.001000 Time 0.020073 -2023-02-13 17:56:11,845 - Epoch: [85][ 1080/ 1207] Overall Loss 0.336330 Objective Loss 0.336330 LR 0.001000 Time 0.020067 -2023-02-13 17:56:12,039 - Epoch: [85][ 1090/ 1207] Overall Loss 0.336462 Objective Loss 0.336462 LR 0.001000 Time 0.020061 -2023-02-13 17:56:12,234 - Epoch: [85][ 1100/ 1207] Overall Loss 0.336656 Objective Loss 0.336656 LR 0.001000 Time 0.020055 -2023-02-13 17:56:12,426 - Epoch: [85][ 1110/ 1207] Overall Loss 0.336787 Objective Loss 0.336787 LR 0.001000 Time 0.020047 -2023-02-13 17:56:12,622 - Epoch: [85][ 1120/ 1207] Overall Loss 0.336727 Objective Loss 0.336727 LR 0.001000 Time 0.020042 -2023-02-13 17:56:12,815 - Epoch: [85][ 1130/ 1207] Overall Loss 0.336594 Objective Loss 0.336594 LR 0.001000 Time 0.020036 -2023-02-13 17:56:13,011 - Epoch: [85][ 1140/ 1207] Overall Loss 0.336943 Objective Loss 0.336943 LR 0.001000 Time 0.020032 -2023-02-13 17:56:13,204 - Epoch: [85][ 1150/ 1207] Overall Loss 0.336875 Objective Loss 0.336875 LR 0.001000 Time 0.020025 -2023-02-13 17:56:13,399 - Epoch: [85][ 1160/ 1207] Overall Loss 0.336942 Objective Loss 0.336942 LR 0.001000 Time 0.020020 -2023-02-13 17:56:13,593 - Epoch: [85][ 1170/ 1207] Overall Loss 0.337130 Objective Loss 0.337130 LR 0.001000 Time 0.020014 -2023-02-13 17:56:13,789 - Epoch: [85][ 1180/ 1207] Overall Loss 0.337003 Objective Loss 0.337003 LR 0.001000 Time 0.020010 -2023-02-13 17:56:13,982 - Epoch: [85][ 1190/ 1207] Overall Loss 0.337025 Objective Loss 0.337025 LR 0.001000 Time 0.020004 -2023-02-13 17:56:14,231 - Epoch: [85][ 1200/ 1207] Overall Loss 0.336959 Objective Loss 0.336959 LR 0.001000 Time 0.020045 -2023-02-13 17:56:14,347 - Epoch: [85][ 1207/ 1207] Overall Loss 0.336856 Objective Loss 0.336856 Top1 85.060976 Top5 98.475610 LR 0.001000 Time 0.020024 -2023-02-13 17:56:14,418 - --- validate (epoch=85)----------- -2023-02-13 17:56:14,418 - 34311 samples (256 per mini-batch) -2023-02-13 17:56:14,917 - Epoch: [85][ 10/ 135] Loss 0.369951 Top1 82.070312 Top5 97.304688 -2023-02-13 17:56:15,046 - Epoch: [85][ 20/ 135] Loss 0.374823 Top1 82.070312 Top5 97.285156 -2023-02-13 17:56:15,175 - Epoch: [85][ 30/ 135] Loss 0.377813 Top1 82.291667 Top5 97.031250 -2023-02-13 17:56:15,304 - Epoch: [85][ 40/ 135] Loss 0.371227 Top1 82.294922 Top5 97.099609 -2023-02-13 17:56:15,432 - Epoch: [85][ 50/ 135] Loss 0.366221 Top1 82.351562 Top5 97.179688 -2023-02-13 17:56:15,560 - Epoch: [85][ 60/ 135] Loss 0.365546 Top1 82.343750 Top5 97.233073 -2023-02-13 17:56:15,688 - Epoch: [85][ 70/ 135] Loss 0.369393 Top1 82.159598 Top5 97.159598 -2023-02-13 17:56:15,817 - Epoch: [85][ 80/ 135] Loss 0.366539 Top1 82.387695 Top5 97.211914 -2023-02-13 17:56:15,947 - Epoch: [85][ 90/ 135] Loss 0.367018 Top1 82.378472 Top5 97.187500 -2023-02-13 17:56:16,073 - Epoch: [85][ 100/ 135] Loss 0.365303 Top1 82.445312 Top5 97.214844 -2023-02-13 17:56:16,198 - Epoch: [85][ 110/ 135] Loss 0.366636 Top1 82.290483 Top5 97.194602 -2023-02-13 17:56:16,327 - Epoch: [85][ 120/ 135] Loss 0.365536 Top1 82.272135 Top5 97.194010 -2023-02-13 17:56:16,459 - Epoch: [85][ 130/ 135] Loss 0.366059 Top1 82.316707 Top5 97.193510 -2023-02-13 17:56:16,507 - Epoch: [85][ 135/ 135] Loss 0.364046 Top1 82.358427 Top5 97.202063 -2023-02-13 17:56:16,575 - ==> Top1: 82.358 Top5: 97.202 Loss: 0.364 - -2023-02-13 17:56:16,576 - ==> Confusion: -[[ 820 4 13 4 11 7 0 5 6 59 1 2 3 5 9 2 5 3 1 2 5] - [ 2 864 3 5 12 50 4 33 6 2 5 9 2 0 1 3 7 0 5 2 18] - [ 4 0 949 17 2 1 17 18 2 0 8 2 4 5 2 8 3 5 4 1 6] - [ 2 0 18 909 2 4 1 1 4 2 8 0 6 0 24 3 5 5 11 2 9] - [ 14 8 0 0 991 10 1 1 2 2 1 8 2 7 5 5 3 2 0 2 2] - [ 3 16 0 6 8 965 6 21 1 1 2 11 2 12 1 3 3 1 1 5 2] - [ 2 2 25 2 3 8 1019 8 0 2 3 0 2 2 0 3 3 4 1 8 2] - [ 1 7 10 1 2 36 5 917 0 3 2 8 3 3 0 0 0 0 9 13 4] - [ 14 2 1 2 0 1 0 1 871 43 15 2 2 12 23 2 2 4 8 0 4] - [ 76 1 3 1 8 6 0 5 39 829 0 1 0 23 8 2 1 1 2 2 4] - [ 0 1 4 8 0 4 2 5 19 3 964 3 4 9 4 0 1 1 12 0 7] - [ 0 2 1 0 5 13 3 2 1 1 0 912 22 8 2 3 2 10 2 16 0] - [ 2 0 1 7 2 6 0 4 1 0 1 44 852 2 3 6 2 15 0 2 9] - [ 5 3 4 0 8 16 2 1 10 17 10 10 1 919 3 2 2 2 0 2 7] - [ 2 1 3 24 3 5 0 3 21 4 3 3 2 2 994 2 0 6 5 1 8] - [ 2 2 4 2 7 1 3 1 0 1 0 8 11 5 1 954 11 11 0 12 10] - [ 3 5 0 2 15 4 0 0 1 1 0 7 2 4 2 12 985 1 1 5 11] - [ 3 4 0 6 1 1 3 1 0 1 0 12 27 2 1 15 2 964 0 3 5] - [ 6 6 9 21 3 2 0 47 3 0 8 2 6 1 17 0 1 1 948 2 3] - [ 2 1 1 3 1 9 10 10 1 0 0 14 6 7 0 1 5 1 1 1066 9] - [ 160 179 257 178 159 273 106 244 92 106 198 147 306 370 172 126 252 93 149 301 9566]] - -2023-02-13 17:56:16,578 - ==> Best [Top1: 82.761 Top5: 97.392 Sparsity:0.00 Params: 148928 on epoch: 82] -2023-02-13 17:56:16,578 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:56:16,583 - - -2023-02-13 17:56:16,584 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:56:17,452 - Epoch: [86][ 10/ 1207] Overall Loss 0.310236 Objective Loss 0.310236 LR 0.001000 Time 0.086799 -2023-02-13 17:56:17,651 - Epoch: [86][ 20/ 1207] Overall Loss 0.317474 Objective Loss 0.317474 LR 0.001000 Time 0.053310 -2023-02-13 17:56:17,842 - Epoch: [86][ 30/ 1207] Overall Loss 0.319053 Objective Loss 0.319053 LR 0.001000 Time 0.041891 -2023-02-13 17:56:18,031 - Epoch: [86][ 40/ 1207] Overall Loss 0.322383 Objective Loss 0.322383 LR 0.001000 Time 0.036143 -2023-02-13 17:56:18,222 - Epoch: [86][ 50/ 1207] Overall Loss 0.322649 Objective Loss 0.322649 LR 0.001000 Time 0.032725 -2023-02-13 17:56:18,412 - Epoch: [86][ 60/ 1207] Overall Loss 0.322716 Objective Loss 0.322716 LR 0.001000 Time 0.030430 -2023-02-13 17:56:18,602 - Epoch: [86][ 70/ 1207] Overall Loss 0.326936 Objective Loss 0.326936 LR 0.001000 Time 0.028790 -2023-02-13 17:56:18,792 - Epoch: [86][ 80/ 1207] Overall Loss 0.327803 Objective Loss 0.327803 LR 0.001000 Time 0.027557 -2023-02-13 17:56:18,984 - Epoch: [86][ 90/ 1207] Overall Loss 0.329191 Objective Loss 0.329191 LR 0.001000 Time 0.026624 -2023-02-13 17:56:19,173 - Epoch: [86][ 100/ 1207] Overall Loss 0.332154 Objective Loss 0.332154 LR 0.001000 Time 0.025851 -2023-02-13 17:56:19,361 - Epoch: [86][ 110/ 1207] Overall Loss 0.332565 Objective Loss 0.332565 LR 0.001000 Time 0.025211 -2023-02-13 17:56:19,549 - Epoch: [86][ 120/ 1207] Overall Loss 0.332154 Objective Loss 0.332154 LR 0.001000 Time 0.024674 -2023-02-13 17:56:19,738 - Epoch: [86][ 130/ 1207] Overall Loss 0.329174 Objective Loss 0.329174 LR 0.001000 Time 0.024227 -2023-02-13 17:56:19,927 - Epoch: [86][ 140/ 1207] Overall Loss 0.327776 Objective Loss 0.327776 LR 0.001000 Time 0.023845 -2023-02-13 17:56:20,117 - Epoch: [86][ 150/ 1207] Overall Loss 0.327359 Objective Loss 0.327359 LR 0.001000 Time 0.023519 -2023-02-13 17:56:20,307 - Epoch: [86][ 160/ 1207] Overall Loss 0.330215 Objective Loss 0.330215 LR 0.001000 Time 0.023229 -2023-02-13 17:56:20,497 - Epoch: [86][ 170/ 1207] Overall Loss 0.331868 Objective Loss 0.331868 LR 0.001000 Time 0.022979 -2023-02-13 17:56:20,686 - Epoch: [86][ 180/ 1207] Overall Loss 0.331958 Objective Loss 0.331958 LR 0.001000 Time 0.022751 -2023-02-13 17:56:20,876 - Epoch: [86][ 190/ 1207] Overall Loss 0.332698 Objective Loss 0.332698 LR 0.001000 Time 0.022553 -2023-02-13 17:56:21,065 - Epoch: [86][ 200/ 1207] Overall Loss 0.333462 Objective Loss 0.333462 LR 0.001000 Time 0.022369 -2023-02-13 17:56:21,255 - Epoch: [86][ 210/ 1207] Overall Loss 0.332498 Objective Loss 0.332498 LR 0.001000 Time 0.022204 -2023-02-13 17:56:21,443 - Epoch: [86][ 220/ 1207] Overall Loss 0.333250 Objective Loss 0.333250 LR 0.001000 Time 0.022050 -2023-02-13 17:56:21,633 - Epoch: [86][ 230/ 1207] Overall Loss 0.333771 Objective Loss 0.333771 LR 0.001000 Time 0.021917 -2023-02-13 17:56:21,822 - Epoch: [86][ 240/ 1207] Overall Loss 0.333541 Objective Loss 0.333541 LR 0.001000 Time 0.021790 -2023-02-13 17:56:22,013 - Epoch: [86][ 250/ 1207] Overall Loss 0.335852 Objective Loss 0.335852 LR 0.001000 Time 0.021678 -2023-02-13 17:56:22,202 - Epoch: [86][ 260/ 1207] Overall Loss 0.336029 Objective Loss 0.336029 LR 0.001000 Time 0.021572 -2023-02-13 17:56:22,392 - Epoch: [86][ 270/ 1207] Overall Loss 0.336306 Objective Loss 0.336306 LR 0.001000 Time 0.021474 -2023-02-13 17:56:22,581 - Epoch: [86][ 280/ 1207] Overall Loss 0.336612 Objective Loss 0.336612 LR 0.001000 Time 0.021381 -2023-02-13 17:56:22,771 - Epoch: [86][ 290/ 1207] Overall Loss 0.336650 Objective Loss 0.336650 LR 0.001000 Time 0.021296 -2023-02-13 17:56:22,960 - Epoch: [86][ 300/ 1207] Overall Loss 0.336323 Objective Loss 0.336323 LR 0.001000 Time 0.021218 -2023-02-13 17:56:23,150 - Epoch: [86][ 310/ 1207] Overall Loss 0.335383 Objective Loss 0.335383 LR 0.001000 Time 0.021144 -2023-02-13 17:56:23,339 - Epoch: [86][ 320/ 1207] Overall Loss 0.335784 Objective Loss 0.335784 LR 0.001000 Time 0.021072 -2023-02-13 17:56:23,529 - Epoch: [86][ 330/ 1207] Overall Loss 0.336104 Objective Loss 0.336104 LR 0.001000 Time 0.021009 -2023-02-13 17:56:23,718 - Epoch: [86][ 340/ 1207] Overall Loss 0.336341 Objective Loss 0.336341 LR 0.001000 Time 0.020945 -2023-02-13 17:56:23,908 - Epoch: [86][ 350/ 1207] Overall Loss 0.336643 Objective Loss 0.336643 LR 0.001000 Time 0.020890 -2023-02-13 17:56:24,097 - Epoch: [86][ 360/ 1207] Overall Loss 0.336983 Objective Loss 0.336983 LR 0.001000 Time 0.020832 -2023-02-13 17:56:24,286 - Epoch: [86][ 370/ 1207] Overall Loss 0.336920 Objective Loss 0.336920 LR 0.001000 Time 0.020780 -2023-02-13 17:56:24,475 - Epoch: [86][ 380/ 1207] Overall Loss 0.337376 Objective Loss 0.337376 LR 0.001000 Time 0.020729 -2023-02-13 17:56:24,665 - Epoch: [86][ 390/ 1207] Overall Loss 0.336285 Objective Loss 0.336285 LR 0.001000 Time 0.020684 -2023-02-13 17:56:24,854 - Epoch: [86][ 400/ 1207] Overall Loss 0.336224 Objective Loss 0.336224 LR 0.001000 Time 0.020638 -2023-02-13 17:56:25,044 - Epoch: [86][ 410/ 1207] Overall Loss 0.336044 Objective Loss 0.336044 LR 0.001000 Time 0.020597 -2023-02-13 17:56:25,234 - Epoch: [86][ 420/ 1207] Overall Loss 0.335454 Objective Loss 0.335454 LR 0.001000 Time 0.020558 -2023-02-13 17:56:25,423 - Epoch: [86][ 430/ 1207] Overall Loss 0.335060 Objective Loss 0.335060 LR 0.001000 Time 0.020518 -2023-02-13 17:56:25,611 - Epoch: [86][ 440/ 1207] Overall Loss 0.335062 Objective Loss 0.335062 LR 0.001000 Time 0.020479 -2023-02-13 17:56:25,801 - Epoch: [86][ 450/ 1207] Overall Loss 0.335804 Objective Loss 0.335804 LR 0.001000 Time 0.020446 -2023-02-13 17:56:25,990 - Epoch: [86][ 460/ 1207] Overall Loss 0.335637 Objective Loss 0.335637 LR 0.001000 Time 0.020411 -2023-02-13 17:56:26,179 - Epoch: [86][ 470/ 1207] Overall Loss 0.335197 Objective Loss 0.335197 LR 0.001000 Time 0.020377 -2023-02-13 17:56:26,367 - Epoch: [86][ 480/ 1207] Overall Loss 0.334533 Objective Loss 0.334533 LR 0.001000 Time 0.020343 -2023-02-13 17:56:26,554 - Epoch: [86][ 490/ 1207] Overall Loss 0.334258 Objective Loss 0.334258 LR 0.001000 Time 0.020310 -2023-02-13 17:56:26,742 - Epoch: [86][ 500/ 1207] Overall Loss 0.335039 Objective Loss 0.335039 LR 0.001000 Time 0.020279 -2023-02-13 17:56:26,931 - Epoch: [86][ 510/ 1207] Overall Loss 0.334797 Objective Loss 0.334797 LR 0.001000 Time 0.020251 -2023-02-13 17:56:27,118 - Epoch: [86][ 520/ 1207] Overall Loss 0.334168 Objective Loss 0.334168 LR 0.001000 Time 0.020221 -2023-02-13 17:56:27,306 - Epoch: [86][ 530/ 1207] Overall Loss 0.334147 Objective Loss 0.334147 LR 0.001000 Time 0.020193 -2023-02-13 17:56:27,494 - Epoch: [86][ 540/ 1207] Overall Loss 0.333942 Objective Loss 0.333942 LR 0.001000 Time 0.020166 -2023-02-13 17:56:27,682 - Epoch: [86][ 550/ 1207] Overall Loss 0.333772 Objective Loss 0.333772 LR 0.001000 Time 0.020141 -2023-02-13 17:56:27,869 - Epoch: [86][ 560/ 1207] Overall Loss 0.333062 Objective Loss 0.333062 LR 0.001000 Time 0.020116 -2023-02-13 17:56:28,058 - Epoch: [86][ 570/ 1207] Overall Loss 0.332591 Objective Loss 0.332591 LR 0.001000 Time 0.020092 -2023-02-13 17:56:28,246 - Epoch: [86][ 580/ 1207] Overall Loss 0.333016 Objective Loss 0.333016 LR 0.001000 Time 0.020070 -2023-02-13 17:56:28,434 - Epoch: [86][ 590/ 1207] Overall Loss 0.333145 Objective Loss 0.333145 LR 0.001000 Time 0.020048 -2023-02-13 17:56:28,621 - Epoch: [86][ 600/ 1207] Overall Loss 0.333141 Objective Loss 0.333141 LR 0.001000 Time 0.020025 -2023-02-13 17:56:28,809 - Epoch: [86][ 610/ 1207] Overall Loss 0.333127 Objective Loss 0.333127 LR 0.001000 Time 0.020004 -2023-02-13 17:56:28,997 - Epoch: [86][ 620/ 1207] Overall Loss 0.332768 Objective Loss 0.332768 LR 0.001000 Time 0.019984 -2023-02-13 17:56:29,185 - Epoch: [86][ 630/ 1207] Overall Loss 0.332428 Objective Loss 0.332428 LR 0.001000 Time 0.019964 -2023-02-13 17:56:29,373 - Epoch: [86][ 640/ 1207] Overall Loss 0.332346 Objective Loss 0.332346 LR 0.001000 Time 0.019945 -2023-02-13 17:56:29,561 - Epoch: [86][ 650/ 1207] Overall Loss 0.332560 Objective Loss 0.332560 LR 0.001000 Time 0.019928 -2023-02-13 17:56:29,749 - Epoch: [86][ 660/ 1207] Overall Loss 0.332866 Objective Loss 0.332866 LR 0.001000 Time 0.019910 -2023-02-13 17:56:29,938 - Epoch: [86][ 670/ 1207] Overall Loss 0.332937 Objective Loss 0.332937 LR 0.001000 Time 0.019894 -2023-02-13 17:56:30,125 - Epoch: [86][ 680/ 1207] Overall Loss 0.333332 Objective Loss 0.333332 LR 0.001000 Time 0.019877 -2023-02-13 17:56:30,313 - Epoch: [86][ 690/ 1207] Overall Loss 0.333569 Objective Loss 0.333569 LR 0.001000 Time 0.019860 -2023-02-13 17:56:30,501 - Epoch: [86][ 700/ 1207] Overall Loss 0.333691 Objective Loss 0.333691 LR 0.001000 Time 0.019845 -2023-02-13 17:56:30,689 - Epoch: [86][ 710/ 1207] Overall Loss 0.333207 Objective Loss 0.333207 LR 0.001000 Time 0.019830 -2023-02-13 17:56:30,878 - Epoch: [86][ 720/ 1207] Overall Loss 0.333329 Objective Loss 0.333329 LR 0.001000 Time 0.019816 -2023-02-13 17:56:31,067 - Epoch: [86][ 730/ 1207] Overall Loss 0.333489 Objective Loss 0.333489 LR 0.001000 Time 0.019803 -2023-02-13 17:56:31,255 - Epoch: [86][ 740/ 1207] Overall Loss 0.333264 Objective Loss 0.333264 LR 0.001000 Time 0.019789 -2023-02-13 17:56:31,443 - Epoch: [86][ 750/ 1207] Overall Loss 0.333822 Objective Loss 0.333822 LR 0.001000 Time 0.019776 -2023-02-13 17:56:31,632 - Epoch: [86][ 760/ 1207] Overall Loss 0.333939 Objective Loss 0.333939 LR 0.001000 Time 0.019763 -2023-02-13 17:56:31,820 - Epoch: [86][ 770/ 1207] Overall Loss 0.333695 Objective Loss 0.333695 LR 0.001000 Time 0.019751 -2023-02-13 17:56:32,009 - Epoch: [86][ 780/ 1207] Overall Loss 0.333492 Objective Loss 0.333492 LR 0.001000 Time 0.019738 -2023-02-13 17:56:32,196 - Epoch: [86][ 790/ 1207] Overall Loss 0.333361 Objective Loss 0.333361 LR 0.001000 Time 0.019725 -2023-02-13 17:56:32,383 - Epoch: [86][ 800/ 1207] Overall Loss 0.333587 Objective Loss 0.333587 LR 0.001000 Time 0.019712 -2023-02-13 17:56:32,572 - Epoch: [86][ 810/ 1207] Overall Loss 0.333605 Objective Loss 0.333605 LR 0.001000 Time 0.019701 -2023-02-13 17:56:32,759 - Epoch: [86][ 820/ 1207] Overall Loss 0.333799 Objective Loss 0.333799 LR 0.001000 Time 0.019689 -2023-02-13 17:56:32,948 - Epoch: [86][ 830/ 1207] Overall Loss 0.333703 Objective Loss 0.333703 LR 0.001000 Time 0.019679 -2023-02-13 17:56:33,137 - Epoch: [86][ 840/ 1207] Overall Loss 0.333879 Objective Loss 0.333879 LR 0.001000 Time 0.019669 -2023-02-13 17:56:33,325 - Epoch: [86][ 850/ 1207] Overall Loss 0.334157 Objective Loss 0.334157 LR 0.001000 Time 0.019659 -2023-02-13 17:56:33,514 - Epoch: [86][ 860/ 1207] Overall Loss 0.334554 Objective Loss 0.334554 LR 0.001000 Time 0.019649 -2023-02-13 17:56:33,702 - Epoch: [86][ 870/ 1207] Overall Loss 0.334172 Objective Loss 0.334172 LR 0.001000 Time 0.019639 -2023-02-13 17:56:33,889 - Epoch: [86][ 880/ 1207] Overall Loss 0.334149 Objective Loss 0.334149 LR 0.001000 Time 0.019628 -2023-02-13 17:56:34,078 - Epoch: [86][ 890/ 1207] Overall Loss 0.334033 Objective Loss 0.334033 LR 0.001000 Time 0.019620 -2023-02-13 17:56:34,267 - Epoch: [86][ 900/ 1207] Overall Loss 0.333996 Objective Loss 0.333996 LR 0.001000 Time 0.019611 -2023-02-13 17:56:34,455 - Epoch: [86][ 910/ 1207] Overall Loss 0.333903 Objective Loss 0.333903 LR 0.001000 Time 0.019602 -2023-02-13 17:56:34,643 - Epoch: [86][ 920/ 1207] Overall Loss 0.333975 Objective Loss 0.333975 LR 0.001000 Time 0.019592 -2023-02-13 17:56:34,831 - Epoch: [86][ 930/ 1207] Overall Loss 0.334518 Objective Loss 0.334518 LR 0.001000 Time 0.019584 -2023-02-13 17:56:35,019 - Epoch: [86][ 940/ 1207] Overall Loss 0.334517 Objective Loss 0.334517 LR 0.001000 Time 0.019575 -2023-02-13 17:56:35,208 - Epoch: [86][ 950/ 1207] Overall Loss 0.334495 Objective Loss 0.334495 LR 0.001000 Time 0.019567 -2023-02-13 17:56:35,396 - Epoch: [86][ 960/ 1207] Overall Loss 0.334664 Objective Loss 0.334664 LR 0.001000 Time 0.019559 -2023-02-13 17:56:35,584 - Epoch: [86][ 970/ 1207] Overall Loss 0.334693 Objective Loss 0.334693 LR 0.001000 Time 0.019551 -2023-02-13 17:56:35,773 - Epoch: [86][ 980/ 1207] Overall Loss 0.334867 Objective Loss 0.334867 LR 0.001000 Time 0.019544 -2023-02-13 17:56:35,963 - Epoch: [86][ 990/ 1207] Overall Loss 0.335253 Objective Loss 0.335253 LR 0.001000 Time 0.019538 -2023-02-13 17:56:36,151 - Epoch: [86][ 1000/ 1207] Overall Loss 0.335725 Objective Loss 0.335725 LR 0.001000 Time 0.019530 -2023-02-13 17:56:36,339 - Epoch: [86][ 1010/ 1207] Overall Loss 0.335707 Objective Loss 0.335707 LR 0.001000 Time 0.019523 -2023-02-13 17:56:36,527 - Epoch: [86][ 1020/ 1207] Overall Loss 0.335853 Objective Loss 0.335853 LR 0.001000 Time 0.019515 -2023-02-13 17:56:36,716 - Epoch: [86][ 1030/ 1207] Overall Loss 0.336212 Objective Loss 0.336212 LR 0.001000 Time 0.019508 -2023-02-13 17:56:36,904 - Epoch: [86][ 1040/ 1207] Overall Loss 0.336062 Objective Loss 0.336062 LR 0.001000 Time 0.019502 -2023-02-13 17:56:37,091 - Epoch: [86][ 1050/ 1207] Overall Loss 0.336455 Objective Loss 0.336455 LR 0.001000 Time 0.019494 -2023-02-13 17:56:37,277 - Epoch: [86][ 1060/ 1207] Overall Loss 0.336413 Objective Loss 0.336413 LR 0.001000 Time 0.019485 -2023-02-13 17:56:37,463 - Epoch: [86][ 1070/ 1207] Overall Loss 0.336342 Objective Loss 0.336342 LR 0.001000 Time 0.019476 -2023-02-13 17:56:37,648 - Epoch: [86][ 1080/ 1207] Overall Loss 0.336030 Objective Loss 0.336030 LR 0.001000 Time 0.019467 -2023-02-13 17:56:37,836 - Epoch: [86][ 1090/ 1207] Overall Loss 0.335991 Objective Loss 0.335991 LR 0.001000 Time 0.019461 -2023-02-13 17:56:38,025 - Epoch: [86][ 1100/ 1207] Overall Loss 0.335606 Objective Loss 0.335606 LR 0.001000 Time 0.019455 -2023-02-13 17:56:38,214 - Epoch: [86][ 1110/ 1207] Overall Loss 0.335928 Objective Loss 0.335928 LR 0.001000 Time 0.019450 -2023-02-13 17:56:38,403 - Epoch: [86][ 1120/ 1207] Overall Loss 0.336128 Objective Loss 0.336128 LR 0.001000 Time 0.019444 -2023-02-13 17:56:38,592 - Epoch: [86][ 1130/ 1207] Overall Loss 0.336007 Objective Loss 0.336007 LR 0.001000 Time 0.019440 -2023-02-13 17:56:38,780 - Epoch: [86][ 1140/ 1207] Overall Loss 0.335876 Objective Loss 0.335876 LR 0.001000 Time 0.019434 -2023-02-13 17:56:38,970 - Epoch: [86][ 1150/ 1207] Overall Loss 0.335911 Objective Loss 0.335911 LR 0.001000 Time 0.019429 -2023-02-13 17:56:39,159 - Epoch: [86][ 1160/ 1207] Overall Loss 0.336031 Objective Loss 0.336031 LR 0.001000 Time 0.019424 -2023-02-13 17:56:39,349 - Epoch: [86][ 1170/ 1207] Overall Loss 0.336058 Objective Loss 0.336058 LR 0.001000 Time 0.019421 -2023-02-13 17:56:39,538 - Epoch: [86][ 1180/ 1207] Overall Loss 0.336078 Objective Loss 0.336078 LR 0.001000 Time 0.019415 -2023-02-13 17:56:39,726 - Epoch: [86][ 1190/ 1207] Overall Loss 0.336261 Objective Loss 0.336261 LR 0.001000 Time 0.019411 -2023-02-13 17:56:39,966 - Epoch: [86][ 1200/ 1207] Overall Loss 0.336264 Objective Loss 0.336264 LR 0.001000 Time 0.019449 -2023-02-13 17:56:40,080 - Epoch: [86][ 1207/ 1207] Overall Loss 0.336448 Objective Loss 0.336448 Top1 82.621951 Top5 96.951220 LR 0.001000 Time 0.019430 -2023-02-13 17:56:40,152 - --- validate (epoch=86)----------- -2023-02-13 17:56:40,153 - 34311 samples (256 per mini-batch) -2023-02-13 17:56:40,557 - Epoch: [86][ 10/ 135] Loss 0.381574 Top1 82.304688 Top5 97.421875 -2023-02-13 17:56:40,682 - Epoch: [86][ 20/ 135] Loss 0.393717 Top1 81.855469 Top5 97.363281 -2023-02-13 17:56:40,810 - Epoch: [86][ 30/ 135] Loss 0.386730 Top1 81.705729 Top5 97.239583 -2023-02-13 17:56:40,938 - Epoch: [86][ 40/ 135] Loss 0.378873 Top1 82.070312 Top5 97.265625 -2023-02-13 17:56:41,065 - Epoch: [86][ 50/ 135] Loss 0.377910 Top1 81.835938 Top5 97.304688 -2023-02-13 17:56:41,191 - Epoch: [86][ 60/ 135] Loss 0.376133 Top1 82.259115 Top5 97.434896 -2023-02-13 17:56:41,318 - Epoch: [86][ 70/ 135] Loss 0.374138 Top1 82.237723 Top5 97.455357 -2023-02-13 17:56:41,444 - Epoch: [86][ 80/ 135] Loss 0.372668 Top1 82.182617 Top5 97.446289 -2023-02-13 17:56:41,572 - Epoch: [86][ 90/ 135] Loss 0.372822 Top1 82.178819 Top5 97.408854 -2023-02-13 17:56:41,696 - Epoch: [86][ 100/ 135] Loss 0.373429 Top1 81.933594 Top5 97.378906 -2023-02-13 17:56:41,821 - Epoch: [86][ 110/ 135] Loss 0.372195 Top1 82.009943 Top5 97.407670 -2023-02-13 17:56:41,948 - Epoch: [86][ 120/ 135] Loss 0.373318 Top1 81.998698 Top5 97.369792 -2023-02-13 17:56:42,076 - Epoch: [86][ 130/ 135] Loss 0.374155 Top1 82.094351 Top5 97.364784 -2023-02-13 17:56:42,120 - Epoch: [86][ 135/ 135] Loss 0.378516 Top1 82.110693 Top5 97.371105 -2023-02-13 17:56:42,203 - ==> Top1: 82.111 Top5: 97.371 Loss: 0.379 - -2023-02-13 17:56:42,204 - ==> Confusion: -[[ 848 5 7 5 13 4 0 2 7 39 3 7 2 4 5 2 0 0 1 2 11] - [ 2 918 2 1 10 22 5 21 7 2 3 4 2 2 1 1 9 0 6 3 12] - [ 7 4 920 12 4 1 33 15 2 1 7 1 0 6 1 9 6 3 6 4 16] - [ 6 2 24 872 4 5 4 2 2 1 24 1 5 1 21 5 3 4 16 1 13] - [ 20 13 0 0 964 9 0 3 0 4 2 4 3 3 8 7 14 2 0 4 6] - [ 5 27 1 3 3 941 3 21 0 3 1 9 7 18 4 0 7 0 3 7 7] - [ 3 6 16 2 1 4 1033 5 0 2 2 4 2 1 1 2 1 1 1 9 3] - [ 0 17 11 1 5 27 9 892 1 1 5 7 5 1 0 0 1 2 22 14 3] - [ 16 2 0 1 2 0 0 2 859 44 29 3 0 11 29 2 1 0 4 1 3] - [ 106 2 3 0 10 1 2 4 37 798 1 0 0 26 7 0 3 3 0 2 7] - [ 2 4 2 4 0 2 4 5 6 2 991 2 1 9 1 0 2 1 9 0 4] - [ 1 3 2 0 3 10 1 6 1 0 0 917 25 6 1 6 2 5 2 10 4] - [ 1 2 1 5 2 1 1 0 3 1 0 66 841 0 2 7 4 11 0 3 8] - [ 7 2 1 0 6 3 1 4 9 14 21 7 1 925 3 5 5 1 0 2 7] - [ 10 7 5 16 6 3 1 1 15 6 8 3 6 5 970 1 1 4 13 0 11] - [ 2 1 3 0 10 1 3 1 1 0 0 8 5 3 0 956 17 12 0 14 9] - [ 1 8 0 1 10 2 0 1 1 0 0 3 1 3 0 9 1002 0 2 6 11] - [ 4 5 3 5 3 3 3 1 1 1 1 17 30 1 0 12 0 952 2 1 6] - [ 4 8 7 7 1 1 1 30 5 1 13 1 5 0 19 1 1 2 973 4 2] - [ 0 4 1 1 2 3 8 12 0 0 1 18 2 5 0 4 5 2 2 1073 5] - [ 144 259 189 90 164 182 119 161 108 106 301 173 305 351 173 108 378 69 174 352 9528]] - -2023-02-13 17:56:42,205 - ==> Best [Top1: 82.761 Top5: 97.392 Sparsity:0.00 Params: 148928 on epoch: 82] -2023-02-13 17:56:42,205 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:56:42,211 - - -2023-02-13 17:56:42,211 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:56:43,180 - Epoch: [87][ 10/ 1207] Overall Loss 0.352431 Objective Loss 0.352431 LR 0.001000 Time 0.096821 -2023-02-13 17:56:43,388 - Epoch: [87][ 20/ 1207] Overall Loss 0.341211 Objective Loss 0.341211 LR 0.001000 Time 0.058789 -2023-02-13 17:56:43,585 - Epoch: [87][ 30/ 1207] Overall Loss 0.340847 Objective Loss 0.340847 LR 0.001000 Time 0.045745 -2023-02-13 17:56:43,786 - Epoch: [87][ 40/ 1207] Overall Loss 0.340490 Objective Loss 0.340490 LR 0.001000 Time 0.039319 -2023-02-13 17:56:43,983 - Epoch: [87][ 50/ 1207] Overall Loss 0.341150 Objective Loss 0.341150 LR 0.001000 Time 0.035396 -2023-02-13 17:56:44,184 - Epoch: [87][ 60/ 1207] Overall Loss 0.341563 Objective Loss 0.341563 LR 0.001000 Time 0.032843 -2023-02-13 17:56:44,382 - Epoch: [87][ 70/ 1207] Overall Loss 0.339525 Objective Loss 0.339525 LR 0.001000 Time 0.030964 -2023-02-13 17:56:44,583 - Epoch: [87][ 80/ 1207] Overall Loss 0.335465 Objective Loss 0.335465 LR 0.001000 Time 0.029599 -2023-02-13 17:56:44,779 - Epoch: [87][ 90/ 1207] Overall Loss 0.333409 Objective Loss 0.333409 LR 0.001000 Time 0.028488 -2023-02-13 17:56:44,981 - Epoch: [87][ 100/ 1207] Overall Loss 0.331516 Objective Loss 0.331516 LR 0.001000 Time 0.027658 -2023-02-13 17:56:45,178 - Epoch: [87][ 110/ 1207] Overall Loss 0.333606 Objective Loss 0.333606 LR 0.001000 Time 0.026930 -2023-02-13 17:56:45,379 - Epoch: [87][ 120/ 1207] Overall Loss 0.333385 Objective Loss 0.333385 LR 0.001000 Time 0.026358 -2023-02-13 17:56:45,575 - Epoch: [87][ 130/ 1207] Overall Loss 0.333871 Objective Loss 0.333871 LR 0.001000 Time 0.025837 -2023-02-13 17:56:45,776 - Epoch: [87][ 140/ 1207] Overall Loss 0.333496 Objective Loss 0.333496 LR 0.001000 Time 0.025424 -2023-02-13 17:56:45,974 - Epoch: [87][ 150/ 1207] Overall Loss 0.333027 Objective Loss 0.333027 LR 0.001000 Time 0.025043 -2023-02-13 17:56:46,176 - Epoch: [87][ 160/ 1207] Overall Loss 0.333600 Objective Loss 0.333600 LR 0.001000 Time 0.024738 -2023-02-13 17:56:46,372 - Epoch: [87][ 170/ 1207] Overall Loss 0.334001 Objective Loss 0.334001 LR 0.001000 Time 0.024432 -2023-02-13 17:56:46,573 - Epoch: [87][ 180/ 1207] Overall Loss 0.332847 Objective Loss 0.332847 LR 0.001000 Time 0.024192 -2023-02-13 17:56:46,770 - Epoch: [87][ 190/ 1207] Overall Loss 0.333330 Objective Loss 0.333330 LR 0.001000 Time 0.023952 -2023-02-13 17:56:46,971 - Epoch: [87][ 200/ 1207] Overall Loss 0.331414 Objective Loss 0.331414 LR 0.001000 Time 0.023758 -2023-02-13 17:56:47,168 - Epoch: [87][ 210/ 1207] Overall Loss 0.330371 Objective Loss 0.330371 LR 0.001000 Time 0.023561 -2023-02-13 17:56:47,369 - Epoch: [87][ 220/ 1207] Overall Loss 0.330261 Objective Loss 0.330261 LR 0.001000 Time 0.023404 -2023-02-13 17:56:47,566 - Epoch: [87][ 230/ 1207] Overall Loss 0.330281 Objective Loss 0.330281 LR 0.001000 Time 0.023239 -2023-02-13 17:56:47,766 - Epoch: [87][ 240/ 1207] Overall Loss 0.330076 Objective Loss 0.330076 LR 0.001000 Time 0.023104 -2023-02-13 17:56:47,963 - Epoch: [87][ 250/ 1207] Overall Loss 0.330688 Objective Loss 0.330688 LR 0.001000 Time 0.022968 -2023-02-13 17:56:48,164 - Epoch: [87][ 260/ 1207] Overall Loss 0.329869 Objective Loss 0.329869 LR 0.001000 Time 0.022857 -2023-02-13 17:56:48,362 - Epoch: [87][ 270/ 1207] Overall Loss 0.328627 Objective Loss 0.328627 LR 0.001000 Time 0.022740 -2023-02-13 17:56:48,562 - Epoch: [87][ 280/ 1207] Overall Loss 0.328566 Objective Loss 0.328566 LR 0.001000 Time 0.022642 -2023-02-13 17:56:48,759 - Epoch: [87][ 290/ 1207] Overall Loss 0.329286 Objective Loss 0.329286 LR 0.001000 Time 0.022538 -2023-02-13 17:56:48,961 - Epoch: [87][ 300/ 1207] Overall Loss 0.329673 Objective Loss 0.329673 LR 0.001000 Time 0.022458 -2023-02-13 17:56:49,159 - Epoch: [87][ 310/ 1207] Overall Loss 0.330558 Objective Loss 0.330558 LR 0.001000 Time 0.022372 -2023-02-13 17:56:49,361 - Epoch: [87][ 320/ 1207] Overall Loss 0.331431 Objective Loss 0.331431 LR 0.001000 Time 0.022302 -2023-02-13 17:56:49,558 - Epoch: [87][ 330/ 1207] Overall Loss 0.332563 Objective Loss 0.332563 LR 0.001000 Time 0.022222 -2023-02-13 17:56:49,759 - Epoch: [87][ 340/ 1207] Overall Loss 0.332947 Objective Loss 0.332947 LR 0.001000 Time 0.022160 -2023-02-13 17:56:49,956 - Epoch: [87][ 350/ 1207] Overall Loss 0.332640 Objective Loss 0.332640 LR 0.001000 Time 0.022087 -2023-02-13 17:56:50,158 - Epoch: [87][ 360/ 1207] Overall Loss 0.332510 Objective Loss 0.332510 LR 0.001000 Time 0.022034 -2023-02-13 17:56:50,355 - Epoch: [87][ 370/ 1207] Overall Loss 0.332556 Objective Loss 0.332556 LR 0.001000 Time 0.021970 -2023-02-13 17:56:50,556 - Epoch: [87][ 380/ 1207] Overall Loss 0.332737 Objective Loss 0.332737 LR 0.001000 Time 0.021920 -2023-02-13 17:56:50,753 - Epoch: [87][ 390/ 1207] Overall Loss 0.331933 Objective Loss 0.331933 LR 0.001000 Time 0.021861 -2023-02-13 17:56:50,955 - Epoch: [87][ 400/ 1207] Overall Loss 0.331874 Objective Loss 0.331874 LR 0.001000 Time 0.021820 -2023-02-13 17:56:51,153 - Epoch: [87][ 410/ 1207] Overall Loss 0.331770 Objective Loss 0.331770 LR 0.001000 Time 0.021769 -2023-02-13 17:56:51,355 - Epoch: [87][ 420/ 1207] Overall Loss 0.330810 Objective Loss 0.330810 LR 0.001000 Time 0.021731 -2023-02-13 17:56:51,552 - Epoch: [87][ 430/ 1207] Overall Loss 0.330299 Objective Loss 0.330299 LR 0.001000 Time 0.021682 -2023-02-13 17:56:51,753 - Epoch: [87][ 440/ 1207] Overall Loss 0.329712 Objective Loss 0.329712 LR 0.001000 Time 0.021646 -2023-02-13 17:56:51,951 - Epoch: [87][ 450/ 1207] Overall Loss 0.329653 Objective Loss 0.329653 LR 0.001000 Time 0.021603 -2023-02-13 17:56:52,152 - Epoch: [87][ 460/ 1207] Overall Loss 0.329936 Objective Loss 0.329936 LR 0.001000 Time 0.021571 -2023-02-13 17:56:52,350 - Epoch: [87][ 470/ 1207] Overall Loss 0.330188 Objective Loss 0.330188 LR 0.001000 Time 0.021531 -2023-02-13 17:56:52,552 - Epoch: [87][ 480/ 1207] Overall Loss 0.330448 Objective Loss 0.330448 LR 0.001000 Time 0.021502 -2023-02-13 17:56:52,749 - Epoch: [87][ 490/ 1207] Overall Loss 0.330013 Objective Loss 0.330013 LR 0.001000 Time 0.021466 -2023-02-13 17:56:52,951 - Epoch: [87][ 500/ 1207] Overall Loss 0.329735 Objective Loss 0.329735 LR 0.001000 Time 0.021438 -2023-02-13 17:56:53,157 - Epoch: [87][ 510/ 1207] Overall Loss 0.329459 Objective Loss 0.329459 LR 0.001000 Time 0.021422 -2023-02-13 17:56:53,365 - Epoch: [87][ 520/ 1207] Overall Loss 0.329956 Objective Loss 0.329956 LR 0.001000 Time 0.021409 -2023-02-13 17:56:53,570 - Epoch: [87][ 530/ 1207] Overall Loss 0.330160 Objective Loss 0.330160 LR 0.001000 Time 0.021391 -2023-02-13 17:56:53,779 - Epoch: [87][ 540/ 1207] Overall Loss 0.330597 Objective Loss 0.330597 LR 0.001000 Time 0.021382 -2023-02-13 17:56:53,986 - Epoch: [87][ 550/ 1207] Overall Loss 0.330135 Objective Loss 0.330135 LR 0.001000 Time 0.021368 -2023-02-13 17:56:54,194 - Epoch: [87][ 560/ 1207] Overall Loss 0.330326 Objective Loss 0.330326 LR 0.001000 Time 0.021357 -2023-02-13 17:56:54,399 - Epoch: [87][ 570/ 1207] Overall Loss 0.330855 Objective Loss 0.330855 LR 0.001000 Time 0.021343 -2023-02-13 17:56:54,608 - Epoch: [87][ 580/ 1207] Overall Loss 0.331295 Objective Loss 0.331295 LR 0.001000 Time 0.021333 -2023-02-13 17:56:54,813 - Epoch: [87][ 590/ 1207] Overall Loss 0.331015 Objective Loss 0.331015 LR 0.001000 Time 0.021320 -2023-02-13 17:56:55,020 - Epoch: [87][ 600/ 1207] Overall Loss 0.330506 Objective Loss 0.330506 LR 0.001000 Time 0.021308 -2023-02-13 17:56:55,226 - Epoch: [87][ 610/ 1207] Overall Loss 0.331350 Objective Loss 0.331350 LR 0.001000 Time 0.021296 -2023-02-13 17:56:55,434 - Epoch: [87][ 620/ 1207] Overall Loss 0.331767 Objective Loss 0.331767 LR 0.001000 Time 0.021287 -2023-02-13 17:56:55,640 - Epoch: [87][ 630/ 1207] Overall Loss 0.332483 Objective Loss 0.332483 LR 0.001000 Time 0.021276 -2023-02-13 17:56:55,849 - Epoch: [87][ 640/ 1207] Overall Loss 0.333003 Objective Loss 0.333003 LR 0.001000 Time 0.021268 -2023-02-13 17:56:56,055 - Epoch: [87][ 650/ 1207] Overall Loss 0.332770 Objective Loss 0.332770 LR 0.001000 Time 0.021259 -2023-02-13 17:56:56,262 - Epoch: [87][ 660/ 1207] Overall Loss 0.332503 Objective Loss 0.332503 LR 0.001000 Time 0.021249 -2023-02-13 17:56:56,468 - Epoch: [87][ 670/ 1207] Overall Loss 0.333074 Objective Loss 0.333074 LR 0.001000 Time 0.021239 -2023-02-13 17:56:56,676 - Epoch: [87][ 680/ 1207] Overall Loss 0.332832 Objective Loss 0.332832 LR 0.001000 Time 0.021232 -2023-02-13 17:56:56,883 - Epoch: [87][ 690/ 1207] Overall Loss 0.332895 Objective Loss 0.332895 LR 0.001000 Time 0.021223 -2023-02-13 17:56:57,091 - Epoch: [87][ 700/ 1207] Overall Loss 0.332063 Objective Loss 0.332063 LR 0.001000 Time 0.021216 -2023-02-13 17:56:57,297 - Epoch: [87][ 710/ 1207] Overall Loss 0.332270 Objective Loss 0.332270 LR 0.001000 Time 0.021208 -2023-02-13 17:56:57,506 - Epoch: [87][ 720/ 1207] Overall Loss 0.332362 Objective Loss 0.332362 LR 0.001000 Time 0.021202 -2023-02-13 17:56:57,713 - Epoch: [87][ 730/ 1207] Overall Loss 0.332752 Objective Loss 0.332752 LR 0.001000 Time 0.021195 -2023-02-13 17:56:57,920 - Epoch: [87][ 740/ 1207] Overall Loss 0.332496 Objective Loss 0.332496 LR 0.001000 Time 0.021188 -2023-02-13 17:56:58,126 - Epoch: [87][ 750/ 1207] Overall Loss 0.332740 Objective Loss 0.332740 LR 0.001000 Time 0.021180 -2023-02-13 17:56:58,335 - Epoch: [87][ 760/ 1207] Overall Loss 0.332596 Objective Loss 0.332596 LR 0.001000 Time 0.021176 -2023-02-13 17:56:58,542 - Epoch: [87][ 770/ 1207] Overall Loss 0.332684 Objective Loss 0.332684 LR 0.001000 Time 0.021168 -2023-02-13 17:56:58,742 - Epoch: [87][ 780/ 1207] Overall Loss 0.332810 Objective Loss 0.332810 LR 0.001000 Time 0.021153 -2023-02-13 17:56:58,939 - Epoch: [87][ 790/ 1207] Overall Loss 0.332863 Objective Loss 0.332863 LR 0.001000 Time 0.021134 -2023-02-13 17:56:59,133 - Epoch: [87][ 800/ 1207] Overall Loss 0.332923 Objective Loss 0.332923 LR 0.001000 Time 0.021113 -2023-02-13 17:56:59,330 - Epoch: [87][ 810/ 1207] Overall Loss 0.333172 Objective Loss 0.333172 LR 0.001000 Time 0.021094 -2023-02-13 17:56:59,524 - Epoch: [87][ 820/ 1207] Overall Loss 0.332937 Objective Loss 0.332937 LR 0.001000 Time 0.021074 -2023-02-13 17:56:59,721 - Epoch: [87][ 830/ 1207] Overall Loss 0.333163 Objective Loss 0.333163 LR 0.001000 Time 0.021057 -2023-02-13 17:56:59,916 - Epoch: [87][ 840/ 1207] Overall Loss 0.333058 Objective Loss 0.333058 LR 0.001000 Time 0.021038 -2023-02-13 17:57:00,114 - Epoch: [87][ 850/ 1207] Overall Loss 0.333310 Objective Loss 0.333310 LR 0.001000 Time 0.021022 -2023-02-13 17:57:00,308 - Epoch: [87][ 860/ 1207] Overall Loss 0.333022 Objective Loss 0.333022 LR 0.001000 Time 0.021003 -2023-02-13 17:57:00,504 - Epoch: [87][ 870/ 1207] Overall Loss 0.332630 Objective Loss 0.332630 LR 0.001000 Time 0.020987 -2023-02-13 17:57:00,700 - Epoch: [87][ 880/ 1207] Overall Loss 0.332765 Objective Loss 0.332765 LR 0.001000 Time 0.020970 -2023-02-13 17:57:00,897 - Epoch: [87][ 890/ 1207] Overall Loss 0.332855 Objective Loss 0.332855 LR 0.001000 Time 0.020956 -2023-02-13 17:57:01,093 - Epoch: [87][ 900/ 1207] Overall Loss 0.333302 Objective Loss 0.333302 LR 0.001000 Time 0.020940 -2023-02-13 17:57:01,290 - Epoch: [87][ 910/ 1207] Overall Loss 0.333464 Objective Loss 0.333464 LR 0.001000 Time 0.020926 -2023-02-13 17:57:01,485 - Epoch: [87][ 920/ 1207] Overall Loss 0.333297 Objective Loss 0.333297 LR 0.001000 Time 0.020910 -2023-02-13 17:57:01,682 - Epoch: [87][ 930/ 1207] Overall Loss 0.333574 Objective Loss 0.333574 LR 0.001000 Time 0.020896 -2023-02-13 17:57:01,878 - Epoch: [87][ 940/ 1207] Overall Loss 0.333435 Objective Loss 0.333435 LR 0.001000 Time 0.020882 -2023-02-13 17:57:02,076 - Epoch: [87][ 950/ 1207] Overall Loss 0.333454 Objective Loss 0.333454 LR 0.001000 Time 0.020870 -2023-02-13 17:57:02,271 - Epoch: [87][ 960/ 1207] Overall Loss 0.333537 Objective Loss 0.333537 LR 0.001000 Time 0.020856 -2023-02-13 17:57:02,468 - Epoch: [87][ 970/ 1207] Overall Loss 0.333872 Objective Loss 0.333872 LR 0.001000 Time 0.020844 -2023-02-13 17:57:02,663 - Epoch: [87][ 980/ 1207] Overall Loss 0.333984 Objective Loss 0.333984 LR 0.001000 Time 0.020830 -2023-02-13 17:57:02,861 - Epoch: [87][ 990/ 1207] Overall Loss 0.333708 Objective Loss 0.333708 LR 0.001000 Time 0.020818 -2023-02-13 17:57:03,056 - Epoch: [87][ 1000/ 1207] Overall Loss 0.333991 Objective Loss 0.333991 LR 0.001000 Time 0.020806 -2023-02-13 17:57:03,254 - Epoch: [87][ 1010/ 1207] Overall Loss 0.333854 Objective Loss 0.333854 LR 0.001000 Time 0.020795 -2023-02-13 17:57:03,449 - Epoch: [87][ 1020/ 1207] Overall Loss 0.333960 Objective Loss 0.333960 LR 0.001000 Time 0.020782 -2023-02-13 17:57:03,646 - Epoch: [87][ 1030/ 1207] Overall Loss 0.334077 Objective Loss 0.334077 LR 0.001000 Time 0.020771 -2023-02-13 17:57:03,840 - Epoch: [87][ 1040/ 1207] Overall Loss 0.334116 Objective Loss 0.334116 LR 0.001000 Time 0.020758 -2023-02-13 17:57:04,037 - Epoch: [87][ 1050/ 1207] Overall Loss 0.334347 Objective Loss 0.334347 LR 0.001000 Time 0.020747 -2023-02-13 17:57:04,232 - Epoch: [87][ 1060/ 1207] Overall Loss 0.334413 Objective Loss 0.334413 LR 0.001000 Time 0.020736 -2023-02-13 17:57:04,430 - Epoch: [87][ 1070/ 1207] Overall Loss 0.334608 Objective Loss 0.334608 LR 0.001000 Time 0.020726 -2023-02-13 17:57:04,624 - Epoch: [87][ 1080/ 1207] Overall Loss 0.334708 Objective Loss 0.334708 LR 0.001000 Time 0.020714 -2023-02-13 17:57:04,822 - Epoch: [87][ 1090/ 1207] Overall Loss 0.335222 Objective Loss 0.335222 LR 0.001000 Time 0.020705 -2023-02-13 17:57:05,012 - Epoch: [87][ 1100/ 1207] Overall Loss 0.335116 Objective Loss 0.335116 LR 0.001000 Time 0.020689 -2023-02-13 17:57:05,202 - Epoch: [87][ 1110/ 1207] Overall Loss 0.335080 Objective Loss 0.335080 LR 0.001000 Time 0.020673 -2023-02-13 17:57:05,391 - Epoch: [87][ 1120/ 1207] Overall Loss 0.335148 Objective Loss 0.335148 LR 0.001000 Time 0.020657 -2023-02-13 17:57:05,579 - Epoch: [87][ 1130/ 1207] Overall Loss 0.335087 Objective Loss 0.335087 LR 0.001000 Time 0.020640 -2023-02-13 17:57:05,768 - Epoch: [87][ 1140/ 1207] Overall Loss 0.334889 Objective Loss 0.334889 LR 0.001000 Time 0.020625 -2023-02-13 17:57:05,958 - Epoch: [87][ 1150/ 1207] Overall Loss 0.334849 Objective Loss 0.334849 LR 0.001000 Time 0.020610 -2023-02-13 17:57:06,148 - Epoch: [87][ 1160/ 1207] Overall Loss 0.335011 Objective Loss 0.335011 LR 0.001000 Time 0.020596 -2023-02-13 17:57:06,337 - Epoch: [87][ 1170/ 1207] Overall Loss 0.335052 Objective Loss 0.335052 LR 0.001000 Time 0.020581 -2023-02-13 17:57:06,526 - Epoch: [87][ 1180/ 1207] Overall Loss 0.335046 Objective Loss 0.335046 LR 0.001000 Time 0.020567 -2023-02-13 17:57:06,715 - Epoch: [87][ 1190/ 1207] Overall Loss 0.334985 Objective Loss 0.334985 LR 0.001000 Time 0.020552 -2023-02-13 17:57:06,958 - Epoch: [87][ 1200/ 1207] Overall Loss 0.335052 Objective Loss 0.335052 LR 0.001000 Time 0.020583 -2023-02-13 17:57:07,073 - Epoch: [87][ 1207/ 1207] Overall Loss 0.335038 Objective Loss 0.335038 Top1 82.012195 Top5 96.646341 LR 0.001000 Time 0.020559 -2023-02-13 17:57:07,144 - --- validate (epoch=87)----------- -2023-02-13 17:57:07,145 - 34311 samples (256 per mini-batch) -2023-02-13 17:57:07,546 - Epoch: [87][ 10/ 135] Loss 0.336804 Top1 84.414062 Top5 97.617188 -2023-02-13 17:57:07,672 - Epoch: [87][ 20/ 135] Loss 0.351027 Top1 83.613281 Top5 97.675781 -2023-02-13 17:57:07,797 - Epoch: [87][ 30/ 135] Loss 0.348356 Top1 83.411458 Top5 97.578125 -2023-02-13 17:57:07,926 - Epoch: [87][ 40/ 135] Loss 0.356221 Top1 83.085938 Top5 97.431641 -2023-02-13 17:57:08,052 - Epoch: [87][ 50/ 135] Loss 0.349282 Top1 83.367188 Top5 97.500000 -2023-02-13 17:57:08,177 - Epoch: [87][ 60/ 135] Loss 0.351668 Top1 83.320312 Top5 97.460938 -2023-02-13 17:57:08,304 - Epoch: [87][ 70/ 135] Loss 0.355956 Top1 83.275670 Top5 97.410714 -2023-02-13 17:57:08,432 - Epoch: [87][ 80/ 135] Loss 0.361712 Top1 83.178711 Top5 97.353516 -2023-02-13 17:57:08,559 - Epoch: [87][ 90/ 135] Loss 0.360224 Top1 83.185764 Top5 97.387153 -2023-02-13 17:57:08,689 - Epoch: [87][ 100/ 135] Loss 0.364252 Top1 83.042969 Top5 97.324219 -2023-02-13 17:57:08,818 - Epoch: [87][ 110/ 135] Loss 0.363830 Top1 83.011364 Top5 97.286932 -2023-02-13 17:57:08,948 - Epoch: [87][ 120/ 135] Loss 0.364831 Top1 83.033854 Top5 97.294922 -2023-02-13 17:57:09,079 - Epoch: [87][ 130/ 135] Loss 0.366354 Top1 83.016827 Top5 97.283654 -2023-02-13 17:57:09,124 - Epoch: [87][ 135/ 135] Loss 0.366330 Top1 82.976305 Top5 97.277841 -2023-02-13 17:57:09,195 - ==> Top1: 82.976 Top5: 97.278 Loss: 0.366 - -2023-02-13 17:57:09,196 - ==> Confusion: -[[ 851 3 6 1 8 3 0 1 6 52 1 4 3 2 7 1 4 3 2 3 6] - [ 3 898 4 4 12 39 8 20 5 2 4 1 3 1 2 1 4 1 3 4 14] - [ 13 3 929 11 4 2 21 18 0 1 5 3 3 4 3 9 2 5 7 4 11] - [ 7 0 13 896 1 5 2 2 3 2 21 1 5 0 21 2 4 5 21 0 5] - [ 34 10 0 0 962 14 1 1 2 3 1 7 1 7 6 3 8 3 0 0 3] - [ 2 13 0 4 7 957 2 17 1 2 2 19 11 11 2 4 5 0 3 5 3] - [ 1 4 10 3 1 4 1031 10 0 3 4 1 2 1 0 6 1 4 1 7 5] - [ 2 9 8 1 1 36 5 912 2 2 5 6 4 1 2 0 1 1 12 9 5] - [ 24 1 0 1 2 0 0 2 883 41 11 4 1 8 15 1 1 4 5 0 5] - [ 93 1 3 1 5 1 1 2 33 845 1 2 0 13 5 0 1 1 0 0 4] - [ 3 4 3 4 0 2 4 1 20 1 974 0 1 11 2 1 0 1 11 0 8] - [ 5 2 1 0 1 8 0 9 3 1 0 864 61 5 0 7 3 19 2 4 10] - [ 2 0 0 9 5 4 0 1 1 0 1 30 847 0 3 4 1 30 1 1 19] - [ 6 3 3 2 6 13 2 3 14 19 9 9 3 907 4 5 6 0 3 1 6] - [ 12 1 6 14 7 6 1 2 27 6 7 1 2 2 968 2 1 7 7 0 13] - [ 5 2 2 2 11 0 6 1 0 0 0 6 7 2 0 960 12 17 0 6 7] - [ 7 3 0 1 10 2 0 0 1 1 1 3 4 1 1 16 991 2 1 1 15] - [ 4 1 1 5 1 1 0 0 1 1 0 8 18 2 1 18 1 980 0 1 7] - [ 6 3 3 11 2 2 1 38 9 0 7 3 4 0 15 0 2 2 975 0 3] - [ 0 2 2 2 2 13 9 24 1 0 2 23 8 0 0 8 10 4 1 1021 16] - [ 214 185 210 132 144 270 95 210 114 93 226 122 316 332 133 91 212 112 183 221 9819]] - -2023-02-13 17:57:09,197 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:57:09,197 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:57:09,204 - - -2023-02-13 17:57:09,204 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:57:10,187 - Epoch: [88][ 10/ 1207] Overall Loss 0.341304 Objective Loss 0.341304 LR 0.001000 Time 0.098292 -2023-02-13 17:57:10,379 - Epoch: [88][ 20/ 1207] Overall Loss 0.345943 Objective Loss 0.345943 LR 0.001000 Time 0.058695 -2023-02-13 17:57:10,568 - Epoch: [88][ 30/ 1207] Overall Loss 0.336200 Objective Loss 0.336200 LR 0.001000 Time 0.045413 -2023-02-13 17:57:10,755 - Epoch: [88][ 40/ 1207] Overall Loss 0.332190 Objective Loss 0.332190 LR 0.001000 Time 0.038739 -2023-02-13 17:57:10,944 - Epoch: [88][ 50/ 1207] Overall Loss 0.338472 Objective Loss 0.338472 LR 0.001000 Time 0.034760 -2023-02-13 17:57:11,132 - Epoch: [88][ 60/ 1207] Overall Loss 0.335935 Objective Loss 0.335935 LR 0.001000 Time 0.032092 -2023-02-13 17:57:11,319 - Epoch: [88][ 70/ 1207] Overall Loss 0.335399 Objective Loss 0.335399 LR 0.001000 Time 0.030179 -2023-02-13 17:57:11,507 - Epoch: [88][ 80/ 1207] Overall Loss 0.336729 Objective Loss 0.336729 LR 0.001000 Time 0.028750 -2023-02-13 17:57:11,695 - Epoch: [88][ 90/ 1207] Overall Loss 0.328945 Objective Loss 0.328945 LR 0.001000 Time 0.027643 -2023-02-13 17:57:11,883 - Epoch: [88][ 100/ 1207] Overall Loss 0.329897 Objective Loss 0.329897 LR 0.001000 Time 0.026753 -2023-02-13 17:57:12,071 - Epoch: [88][ 110/ 1207] Overall Loss 0.330844 Objective Loss 0.330844 LR 0.001000 Time 0.026028 -2023-02-13 17:57:12,259 - Epoch: [88][ 120/ 1207] Overall Loss 0.330582 Objective Loss 0.330582 LR 0.001000 Time 0.025424 -2023-02-13 17:57:12,448 - Epoch: [88][ 130/ 1207] Overall Loss 0.330090 Objective Loss 0.330090 LR 0.001000 Time 0.024914 -2023-02-13 17:57:12,635 - Epoch: [88][ 140/ 1207] Overall Loss 0.329969 Objective Loss 0.329969 LR 0.001000 Time 0.024471 -2023-02-13 17:57:12,823 - Epoch: [88][ 150/ 1207] Overall Loss 0.329668 Objective Loss 0.329668 LR 0.001000 Time 0.024089 -2023-02-13 17:57:13,011 - Epoch: [88][ 160/ 1207] Overall Loss 0.330520 Objective Loss 0.330520 LR 0.001000 Time 0.023756 -2023-02-13 17:57:13,199 - Epoch: [88][ 170/ 1207] Overall Loss 0.330767 Objective Loss 0.330767 LR 0.001000 Time 0.023464 -2023-02-13 17:57:13,388 - Epoch: [88][ 180/ 1207] Overall Loss 0.330553 Objective Loss 0.330553 LR 0.001000 Time 0.023204 -2023-02-13 17:57:13,575 - Epoch: [88][ 190/ 1207] Overall Loss 0.329555 Objective Loss 0.329555 LR 0.001000 Time 0.022969 -2023-02-13 17:57:13,763 - Epoch: [88][ 200/ 1207] Overall Loss 0.328303 Objective Loss 0.328303 LR 0.001000 Time 0.022759 -2023-02-13 17:57:13,951 - Epoch: [88][ 210/ 1207] Overall Loss 0.328180 Objective Loss 0.328180 LR 0.001000 Time 0.022566 -2023-02-13 17:57:14,139 - Epoch: [88][ 220/ 1207] Overall Loss 0.329628 Objective Loss 0.329628 LR 0.001000 Time 0.022395 -2023-02-13 17:57:14,328 - Epoch: [88][ 230/ 1207] Overall Loss 0.328833 Objective Loss 0.328833 LR 0.001000 Time 0.022239 -2023-02-13 17:57:14,516 - Epoch: [88][ 240/ 1207] Overall Loss 0.328705 Objective Loss 0.328705 LR 0.001000 Time 0.022093 -2023-02-13 17:57:14,703 - Epoch: [88][ 250/ 1207] Overall Loss 0.329262 Objective Loss 0.329262 LR 0.001000 Time 0.021959 -2023-02-13 17:57:14,891 - Epoch: [88][ 260/ 1207] Overall Loss 0.328230 Objective Loss 0.328230 LR 0.001000 Time 0.021836 -2023-02-13 17:57:15,079 - Epoch: [88][ 270/ 1207] Overall Loss 0.327637 Objective Loss 0.327637 LR 0.001000 Time 0.021723 -2023-02-13 17:57:15,268 - Epoch: [88][ 280/ 1207] Overall Loss 0.327614 Objective Loss 0.327614 LR 0.001000 Time 0.021618 -2023-02-13 17:57:15,456 - Epoch: [88][ 290/ 1207] Overall Loss 0.327892 Objective Loss 0.327892 LR 0.001000 Time 0.021520 -2023-02-13 17:57:15,643 - Epoch: [88][ 300/ 1207] Overall Loss 0.327995 Objective Loss 0.327995 LR 0.001000 Time 0.021427 -2023-02-13 17:57:15,832 - Epoch: [88][ 310/ 1207] Overall Loss 0.329354 Objective Loss 0.329354 LR 0.001000 Time 0.021343 -2023-02-13 17:57:16,020 - Epoch: [88][ 320/ 1207] Overall Loss 0.328447 Objective Loss 0.328447 LR 0.001000 Time 0.021262 -2023-02-13 17:57:16,209 - Epoch: [88][ 330/ 1207] Overall Loss 0.328084 Objective Loss 0.328084 LR 0.001000 Time 0.021190 -2023-02-13 17:57:16,398 - Epoch: [88][ 340/ 1207] Overall Loss 0.328439 Objective Loss 0.328439 LR 0.001000 Time 0.021122 -2023-02-13 17:57:16,586 - Epoch: [88][ 350/ 1207] Overall Loss 0.328746 Objective Loss 0.328746 LR 0.001000 Time 0.021054 -2023-02-13 17:57:16,775 - Epoch: [88][ 360/ 1207] Overall Loss 0.329274 Objective Loss 0.329274 LR 0.001000 Time 0.020992 -2023-02-13 17:57:16,963 - Epoch: [88][ 370/ 1207] Overall Loss 0.329244 Objective Loss 0.329244 LR 0.001000 Time 0.020933 -2023-02-13 17:57:17,152 - Epoch: [88][ 380/ 1207] Overall Loss 0.329202 Objective Loss 0.329202 LR 0.001000 Time 0.020877 -2023-02-13 17:57:17,340 - Epoch: [88][ 390/ 1207] Overall Loss 0.328532 Objective Loss 0.328532 LR 0.001000 Time 0.020822 -2023-02-13 17:57:17,528 - Epoch: [88][ 400/ 1207] Overall Loss 0.328407 Objective Loss 0.328407 LR 0.001000 Time 0.020772 -2023-02-13 17:57:17,716 - Epoch: [88][ 410/ 1207] Overall Loss 0.328882 Objective Loss 0.328882 LR 0.001000 Time 0.020723 -2023-02-13 17:57:17,904 - Epoch: [88][ 420/ 1207] Overall Loss 0.329484 Objective Loss 0.329484 LR 0.001000 Time 0.020676 -2023-02-13 17:57:18,093 - Epoch: [88][ 430/ 1207] Overall Loss 0.329521 Objective Loss 0.329521 LR 0.001000 Time 0.020634 -2023-02-13 17:57:18,281 - Epoch: [88][ 440/ 1207] Overall Loss 0.329168 Objective Loss 0.329168 LR 0.001000 Time 0.020593 -2023-02-13 17:57:18,470 - Epoch: [88][ 450/ 1207] Overall Loss 0.329855 Objective Loss 0.329855 LR 0.001000 Time 0.020552 -2023-02-13 17:57:18,658 - Epoch: [88][ 460/ 1207] Overall Loss 0.330126 Objective Loss 0.330126 LR 0.001000 Time 0.020514 -2023-02-13 17:57:18,845 - Epoch: [88][ 470/ 1207] Overall Loss 0.330384 Objective Loss 0.330384 LR 0.001000 Time 0.020475 -2023-02-13 17:57:19,033 - Epoch: [88][ 480/ 1207] Overall Loss 0.330095 Objective Loss 0.330095 LR 0.001000 Time 0.020440 -2023-02-13 17:57:19,221 - Epoch: [88][ 490/ 1207] Overall Loss 0.329905 Objective Loss 0.329905 LR 0.001000 Time 0.020406 -2023-02-13 17:57:19,409 - Epoch: [88][ 500/ 1207] Overall Loss 0.330302 Objective Loss 0.330302 LR 0.001000 Time 0.020373 -2023-02-13 17:57:19,597 - Epoch: [88][ 510/ 1207] Overall Loss 0.330863 Objective Loss 0.330863 LR 0.001000 Time 0.020341 -2023-02-13 17:57:19,785 - Epoch: [88][ 520/ 1207] Overall Loss 0.330960 Objective Loss 0.330960 LR 0.001000 Time 0.020310 -2023-02-13 17:57:19,972 - Epoch: [88][ 530/ 1207] Overall Loss 0.330607 Objective Loss 0.330607 LR 0.001000 Time 0.020280 -2023-02-13 17:57:20,161 - Epoch: [88][ 540/ 1207] Overall Loss 0.330430 Objective Loss 0.330430 LR 0.001000 Time 0.020252 -2023-02-13 17:57:20,349 - Epoch: [88][ 550/ 1207] Overall Loss 0.330004 Objective Loss 0.330004 LR 0.001000 Time 0.020226 -2023-02-13 17:57:20,536 - Epoch: [88][ 560/ 1207] Overall Loss 0.329933 Objective Loss 0.329933 LR 0.001000 Time 0.020199 -2023-02-13 17:57:20,724 - Epoch: [88][ 570/ 1207] Overall Loss 0.329544 Objective Loss 0.329544 LR 0.001000 Time 0.020173 -2023-02-13 17:57:20,912 - Epoch: [88][ 580/ 1207] Overall Loss 0.329460 Objective Loss 0.329460 LR 0.001000 Time 0.020148 -2023-02-13 17:57:21,100 - Epoch: [88][ 590/ 1207] Overall Loss 0.329479 Objective Loss 0.329479 LR 0.001000 Time 0.020124 -2023-02-13 17:57:21,288 - Epoch: [88][ 600/ 1207] Overall Loss 0.329496 Objective Loss 0.329496 LR 0.001000 Time 0.020102 -2023-02-13 17:57:21,477 - Epoch: [88][ 610/ 1207] Overall Loss 0.329518 Objective Loss 0.329518 LR 0.001000 Time 0.020081 -2023-02-13 17:57:21,665 - Epoch: [88][ 620/ 1207] Overall Loss 0.329295 Objective Loss 0.329295 LR 0.001000 Time 0.020061 -2023-02-13 17:57:21,854 - Epoch: [88][ 630/ 1207] Overall Loss 0.328788 Objective Loss 0.328788 LR 0.001000 Time 0.020042 -2023-02-13 17:57:22,043 - Epoch: [88][ 640/ 1207] Overall Loss 0.328402 Objective Loss 0.328402 LR 0.001000 Time 0.020023 -2023-02-13 17:57:22,232 - Epoch: [88][ 650/ 1207] Overall Loss 0.327763 Objective Loss 0.327763 LR 0.001000 Time 0.020006 -2023-02-13 17:57:22,421 - Epoch: [88][ 660/ 1207] Overall Loss 0.327853 Objective Loss 0.327853 LR 0.001000 Time 0.019988 -2023-02-13 17:57:22,609 - Epoch: [88][ 670/ 1207] Overall Loss 0.328008 Objective Loss 0.328008 LR 0.001000 Time 0.019970 -2023-02-13 17:57:22,797 - Epoch: [88][ 680/ 1207] Overall Loss 0.328436 Objective Loss 0.328436 LR 0.001000 Time 0.019952 -2023-02-13 17:57:22,985 - Epoch: [88][ 690/ 1207] Overall Loss 0.328693 Objective Loss 0.328693 LR 0.001000 Time 0.019935 -2023-02-13 17:57:23,174 - Epoch: [88][ 700/ 1207] Overall Loss 0.328799 Objective Loss 0.328799 LR 0.001000 Time 0.019919 -2023-02-13 17:57:23,362 - Epoch: [88][ 710/ 1207] Overall Loss 0.329075 Objective Loss 0.329075 LR 0.001000 Time 0.019903 -2023-02-13 17:57:23,550 - Epoch: [88][ 720/ 1207] Overall Loss 0.329078 Objective Loss 0.329078 LR 0.001000 Time 0.019887 -2023-02-13 17:57:23,738 - Epoch: [88][ 730/ 1207] Overall Loss 0.328672 Objective Loss 0.328672 LR 0.001000 Time 0.019872 -2023-02-13 17:57:23,925 - Epoch: [88][ 740/ 1207] Overall Loss 0.329096 Objective Loss 0.329096 LR 0.001000 Time 0.019856 -2023-02-13 17:57:24,114 - Epoch: [88][ 750/ 1207] Overall Loss 0.329201 Objective Loss 0.329201 LR 0.001000 Time 0.019842 -2023-02-13 17:57:24,302 - Epoch: [88][ 760/ 1207] Overall Loss 0.329378 Objective Loss 0.329378 LR 0.001000 Time 0.019829 -2023-02-13 17:57:24,490 - Epoch: [88][ 770/ 1207] Overall Loss 0.329039 Objective Loss 0.329039 LR 0.001000 Time 0.019815 -2023-02-13 17:57:24,678 - Epoch: [88][ 780/ 1207] Overall Loss 0.329102 Objective Loss 0.329102 LR 0.001000 Time 0.019801 -2023-02-13 17:57:24,867 - Epoch: [88][ 790/ 1207] Overall Loss 0.329202 Objective Loss 0.329202 LR 0.001000 Time 0.019789 -2023-02-13 17:57:25,054 - Epoch: [88][ 800/ 1207] Overall Loss 0.329858 Objective Loss 0.329858 LR 0.001000 Time 0.019776 -2023-02-13 17:57:25,243 - Epoch: [88][ 810/ 1207] Overall Loss 0.329866 Objective Loss 0.329866 LR 0.001000 Time 0.019764 -2023-02-13 17:57:25,432 - Epoch: [88][ 820/ 1207] Overall Loss 0.330048 Objective Loss 0.330048 LR 0.001000 Time 0.019753 -2023-02-13 17:57:25,620 - Epoch: [88][ 830/ 1207] Overall Loss 0.329951 Objective Loss 0.329951 LR 0.001000 Time 0.019741 -2023-02-13 17:57:25,809 - Epoch: [88][ 840/ 1207] Overall Loss 0.329883 Objective Loss 0.329883 LR 0.001000 Time 0.019730 -2023-02-13 17:57:25,998 - Epoch: [88][ 850/ 1207] Overall Loss 0.329991 Objective Loss 0.329991 LR 0.001000 Time 0.019720 -2023-02-13 17:57:26,186 - Epoch: [88][ 860/ 1207] Overall Loss 0.329893 Objective Loss 0.329893 LR 0.001000 Time 0.019710 -2023-02-13 17:57:26,375 - Epoch: [88][ 870/ 1207] Overall Loss 0.329841 Objective Loss 0.329841 LR 0.001000 Time 0.019699 -2023-02-13 17:57:26,563 - Epoch: [88][ 880/ 1207] Overall Loss 0.330175 Objective Loss 0.330175 LR 0.001000 Time 0.019688 -2023-02-13 17:57:26,751 - Epoch: [88][ 890/ 1207] Overall Loss 0.330330 Objective Loss 0.330330 LR 0.001000 Time 0.019678 -2023-02-13 17:57:26,939 - Epoch: [88][ 900/ 1207] Overall Loss 0.330143 Objective Loss 0.330143 LR 0.001000 Time 0.019669 -2023-02-13 17:57:27,128 - Epoch: [88][ 910/ 1207] Overall Loss 0.329956 Objective Loss 0.329956 LR 0.001000 Time 0.019659 -2023-02-13 17:57:27,316 - Epoch: [88][ 920/ 1207] Overall Loss 0.329819 Objective Loss 0.329819 LR 0.001000 Time 0.019650 -2023-02-13 17:57:27,505 - Epoch: [88][ 930/ 1207] Overall Loss 0.329938 Objective Loss 0.329938 LR 0.001000 Time 0.019641 -2023-02-13 17:57:27,693 - Epoch: [88][ 940/ 1207] Overall Loss 0.330003 Objective Loss 0.330003 LR 0.001000 Time 0.019632 -2023-02-13 17:57:27,881 - Epoch: [88][ 950/ 1207] Overall Loss 0.329920 Objective Loss 0.329920 LR 0.001000 Time 0.019622 -2023-02-13 17:57:28,069 - Epoch: [88][ 960/ 1207] Overall Loss 0.329759 Objective Loss 0.329759 LR 0.001000 Time 0.019614 -2023-02-13 17:57:28,258 - Epoch: [88][ 970/ 1207] Overall Loss 0.329772 Objective Loss 0.329772 LR 0.001000 Time 0.019606 -2023-02-13 17:57:28,447 - Epoch: [88][ 980/ 1207] Overall Loss 0.329928 Objective Loss 0.329928 LR 0.001000 Time 0.019598 -2023-02-13 17:57:28,635 - Epoch: [88][ 990/ 1207] Overall Loss 0.329909 Objective Loss 0.329909 LR 0.001000 Time 0.019590 -2023-02-13 17:57:28,823 - Epoch: [88][ 1000/ 1207] Overall Loss 0.330201 Objective Loss 0.330201 LR 0.001000 Time 0.019581 -2023-02-13 17:57:29,011 - Epoch: [88][ 1010/ 1207] Overall Loss 0.330313 Objective Loss 0.330313 LR 0.001000 Time 0.019574 -2023-02-13 17:57:29,200 - Epoch: [88][ 1020/ 1207] Overall Loss 0.330186 Objective Loss 0.330186 LR 0.001000 Time 0.019567 -2023-02-13 17:57:29,389 - Epoch: [88][ 1030/ 1207] Overall Loss 0.329898 Objective Loss 0.329898 LR 0.001000 Time 0.019560 -2023-02-13 17:57:29,576 - Epoch: [88][ 1040/ 1207] Overall Loss 0.330069 Objective Loss 0.330069 LR 0.001000 Time 0.019551 -2023-02-13 17:57:29,764 - Epoch: [88][ 1050/ 1207] Overall Loss 0.330408 Objective Loss 0.330408 LR 0.001000 Time 0.019543 -2023-02-13 17:57:29,951 - Epoch: [88][ 1060/ 1207] Overall Loss 0.330456 Objective Loss 0.330456 LR 0.001000 Time 0.019536 -2023-02-13 17:57:30,140 - Epoch: [88][ 1070/ 1207] Overall Loss 0.330585 Objective Loss 0.330585 LR 0.001000 Time 0.019529 -2023-02-13 17:57:30,328 - Epoch: [88][ 1080/ 1207] Overall Loss 0.330458 Objective Loss 0.330458 LR 0.001000 Time 0.019522 -2023-02-13 17:57:30,517 - Epoch: [88][ 1090/ 1207] Overall Loss 0.330492 Objective Loss 0.330492 LR 0.001000 Time 0.019516 -2023-02-13 17:57:30,704 - Epoch: [88][ 1100/ 1207] Overall Loss 0.330691 Objective Loss 0.330691 LR 0.001000 Time 0.019508 -2023-02-13 17:57:30,894 - Epoch: [88][ 1110/ 1207] Overall Loss 0.330787 Objective Loss 0.330787 LR 0.001000 Time 0.019503 -2023-02-13 17:57:31,082 - Epoch: [88][ 1120/ 1207] Overall Loss 0.330797 Objective Loss 0.330797 LR 0.001000 Time 0.019497 -2023-02-13 17:57:31,272 - Epoch: [88][ 1130/ 1207] Overall Loss 0.330710 Objective Loss 0.330710 LR 0.001000 Time 0.019492 -2023-02-13 17:57:31,461 - Epoch: [88][ 1140/ 1207] Overall Loss 0.330840 Objective Loss 0.330840 LR 0.001000 Time 0.019486 -2023-02-13 17:57:31,649 - Epoch: [88][ 1150/ 1207] Overall Loss 0.330806 Objective Loss 0.330806 LR 0.001000 Time 0.019480 -2023-02-13 17:57:31,837 - Epoch: [88][ 1160/ 1207] Overall Loss 0.331066 Objective Loss 0.331066 LR 0.001000 Time 0.019474 -2023-02-13 17:57:32,026 - Epoch: [88][ 1170/ 1207] Overall Loss 0.331331 Objective Loss 0.331331 LR 0.001000 Time 0.019469 -2023-02-13 17:57:32,215 - Epoch: [88][ 1180/ 1207] Overall Loss 0.331297 Objective Loss 0.331297 LR 0.001000 Time 0.019464 -2023-02-13 17:57:32,403 - Epoch: [88][ 1190/ 1207] Overall Loss 0.330983 Objective Loss 0.330983 LR 0.001000 Time 0.019458 -2023-02-13 17:57:32,641 - Epoch: [88][ 1200/ 1207] Overall Loss 0.330766 Objective Loss 0.330766 LR 0.001000 Time 0.019494 -2023-02-13 17:57:32,755 - Epoch: [88][ 1207/ 1207] Overall Loss 0.330586 Objective Loss 0.330586 Top1 86.585366 Top5 97.560976 LR 0.001000 Time 0.019475 -2023-02-13 17:57:32,828 - --- validate (epoch=88)----------- -2023-02-13 17:57:32,829 - 34311 samples (256 per mini-batch) -2023-02-13 17:57:33,227 - Epoch: [88][ 10/ 135] Loss 0.355695 Top1 82.031250 Top5 96.679688 -2023-02-13 17:57:33,357 - Epoch: [88][ 20/ 135] Loss 0.345468 Top1 82.070312 Top5 96.796875 -2023-02-13 17:57:33,487 - Epoch: [88][ 30/ 135] Loss 0.356825 Top1 81.679688 Top5 96.953125 -2023-02-13 17:57:33,616 - Epoch: [88][ 40/ 135] Loss 0.360553 Top1 81.669922 Top5 97.031250 -2023-02-13 17:57:33,745 - Epoch: [88][ 50/ 135] Loss 0.362377 Top1 81.609375 Top5 97.015625 -2023-02-13 17:57:33,874 - Epoch: [88][ 60/ 135] Loss 0.369412 Top1 81.464844 Top5 96.985677 -2023-02-13 17:57:34,001 - Epoch: [88][ 70/ 135] Loss 0.367309 Top1 81.579241 Top5 96.908482 -2023-02-13 17:57:34,131 - Epoch: [88][ 80/ 135] Loss 0.362399 Top1 81.772461 Top5 96.938477 -2023-02-13 17:57:34,260 - Epoch: [88][ 90/ 135] Loss 0.364292 Top1 81.775174 Top5 96.966146 -2023-02-13 17:57:34,389 - Epoch: [88][ 100/ 135] Loss 0.364505 Top1 81.742188 Top5 96.984375 -2023-02-13 17:57:34,517 - Epoch: [88][ 110/ 135] Loss 0.363379 Top1 81.711648 Top5 96.995739 -2023-02-13 17:57:34,646 - Epoch: [88][ 120/ 135] Loss 0.363084 Top1 81.669922 Top5 96.969401 -2023-02-13 17:57:34,778 - Epoch: [88][ 130/ 135] Loss 0.362750 Top1 81.670673 Top5 97.001202 -2023-02-13 17:57:34,825 - Epoch: [88][ 135/ 135] Loss 0.361989 Top1 81.693917 Top5 96.983475 -2023-02-13 17:57:34,893 - ==> Top1: 81.694 Top5: 96.983 Loss: 0.362 - -2023-02-13 17:57:34,894 - ==> Confusion: -[[ 825 2 7 1 12 4 0 5 11 64 0 2 0 5 7 3 3 5 2 3 6] - [ 0 942 0 1 9 25 2 14 3 1 0 2 3 1 3 2 8 1 12 0 4] - [ 6 2 950 10 4 0 13 16 2 3 4 3 4 7 3 4 4 3 9 3 8] - [ 5 0 22 885 3 2 1 2 3 4 16 0 4 3 24 3 2 7 25 0 5] - [ 16 11 0 0 971 4 2 3 2 8 1 7 1 4 12 3 13 2 0 1 5] - [ 2 37 0 6 13 921 5 21 2 6 5 13 7 13 1 1 4 2 1 3 7] - [ 4 3 29 0 0 5 1010 10 0 2 6 2 1 0 2 6 2 6 2 6 3] - [ 5 15 7 2 2 22 4 903 1 1 0 7 5 1 2 1 1 2 32 7 4] - [ 15 1 0 2 0 1 0 2 890 39 7 2 1 11 24 1 1 3 8 1 0] - [ 73 2 1 0 4 1 0 1 40 858 2 1 0 11 4 4 1 2 1 1 5] - [ 0 3 5 7 0 2 4 9 30 2 942 4 1 11 3 0 2 2 22 0 2] - [ 1 3 1 0 5 13 2 2 3 1 0 925 20 7 2 0 1 12 3 1 3] - [ 2 0 0 7 4 3 1 0 2 1 1 45 855 2 5 4 0 18 2 1 6] - [ 2 1 3 0 7 17 0 2 16 18 10 12 4 909 6 4 6 3 0 0 4] - [ 8 2 4 14 7 4 0 2 25 7 2 2 1 4 983 2 2 5 13 0 5] - [ 5 2 5 0 5 2 3 2 3 1 0 8 7 1 1 964 8 15 0 6 8] - [ 6 6 1 2 5 2 0 2 3 2 0 2 1 2 3 12 994 3 1 5 9] - [ 7 0 0 2 0 0 1 3 1 0 0 16 20 1 3 9 1 983 0 0 4] - [ 3 3 6 7 1 2 0 21 6 0 3 3 5 1 21 1 1 3 997 1 1] - [ 1 4 4 0 2 8 5 23 1 0 1 30 3 7 2 4 8 3 4 1032 6] - [ 184 294 299 139 153 209 64 210 144 109 195 196 336 378 203 110 272 143 273 232 9291]] - -2023-02-13 17:57:34,895 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:57:34,895 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:57:34,901 - - -2023-02-13 17:57:34,901 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:57:35,803 - Epoch: [89][ 10/ 1207] Overall Loss 0.288763 Objective Loss 0.288763 LR 0.001000 Time 0.090133 -2023-02-13 17:57:36,001 - Epoch: [89][ 20/ 1207] Overall Loss 0.310775 Objective Loss 0.310775 LR 0.001000 Time 0.054940 -2023-02-13 17:57:36,198 - Epoch: [89][ 30/ 1207] Overall Loss 0.322706 Objective Loss 0.322706 LR 0.001000 Time 0.043157 -2023-02-13 17:57:36,392 - Epoch: [89][ 40/ 1207] Overall Loss 0.315482 Objective Loss 0.315482 LR 0.001000 Time 0.037216 -2023-02-13 17:57:36,587 - Epoch: [89][ 50/ 1207] Overall Loss 0.320940 Objective Loss 0.320940 LR 0.001000 Time 0.033666 -2023-02-13 17:57:36,781 - Epoch: [89][ 60/ 1207] Overall Loss 0.318845 Objective Loss 0.318845 LR 0.001000 Time 0.031285 -2023-02-13 17:57:36,978 - Epoch: [89][ 70/ 1207] Overall Loss 0.318989 Objective Loss 0.318989 LR 0.001000 Time 0.029623 -2023-02-13 17:57:37,173 - Epoch: [89][ 80/ 1207] Overall Loss 0.323000 Objective Loss 0.323000 LR 0.001000 Time 0.028350 -2023-02-13 17:57:37,370 - Epoch: [89][ 90/ 1207] Overall Loss 0.325261 Objective Loss 0.325261 LR 0.001000 Time 0.027393 -2023-02-13 17:57:37,565 - Epoch: [89][ 100/ 1207] Overall Loss 0.324539 Objective Loss 0.324539 LR 0.001000 Time 0.026599 -2023-02-13 17:57:37,762 - Epoch: [89][ 110/ 1207] Overall Loss 0.324689 Objective Loss 0.324689 LR 0.001000 Time 0.025967 -2023-02-13 17:57:37,956 - Epoch: [89][ 120/ 1207] Overall Loss 0.321657 Objective Loss 0.321657 LR 0.001000 Time 0.025414 -2023-02-13 17:57:38,151 - Epoch: [89][ 130/ 1207] Overall Loss 0.323239 Objective Loss 0.323239 LR 0.001000 Time 0.024957 -2023-02-13 17:57:38,345 - Epoch: [89][ 140/ 1207] Overall Loss 0.321985 Objective Loss 0.321985 LR 0.001000 Time 0.024557 -2023-02-13 17:57:38,540 - Epoch: [89][ 150/ 1207] Overall Loss 0.321845 Objective Loss 0.321845 LR 0.001000 Time 0.024218 -2023-02-13 17:57:38,736 - Epoch: [89][ 160/ 1207] Overall Loss 0.321173 Objective Loss 0.321173 LR 0.001000 Time 0.023927 -2023-02-13 17:57:38,936 - Epoch: [89][ 170/ 1207] Overall Loss 0.320491 Objective Loss 0.320491 LR 0.001000 Time 0.023691 -2023-02-13 17:57:39,133 - Epoch: [89][ 180/ 1207] Overall Loss 0.320798 Objective Loss 0.320798 LR 0.001000 Time 0.023471 -2023-02-13 17:57:39,330 - Epoch: [89][ 190/ 1207] Overall Loss 0.320912 Objective Loss 0.320912 LR 0.001000 Time 0.023271 -2023-02-13 17:57:39,524 - Epoch: [89][ 200/ 1207] Overall Loss 0.320572 Objective Loss 0.320572 LR 0.001000 Time 0.023075 -2023-02-13 17:57:39,720 - Epoch: [89][ 210/ 1207] Overall Loss 0.321171 Objective Loss 0.321171 LR 0.001000 Time 0.022907 -2023-02-13 17:57:39,914 - Epoch: [89][ 220/ 1207] Overall Loss 0.320841 Objective Loss 0.320841 LR 0.001000 Time 0.022746 -2023-02-13 17:57:40,110 - Epoch: [89][ 230/ 1207] Overall Loss 0.321053 Objective Loss 0.321053 LR 0.001000 Time 0.022607 -2023-02-13 17:57:40,305 - Epoch: [89][ 240/ 1207] Overall Loss 0.319656 Objective Loss 0.319656 LR 0.001000 Time 0.022477 -2023-02-13 17:57:40,502 - Epoch: [89][ 250/ 1207] Overall Loss 0.320004 Objective Loss 0.320004 LR 0.001000 Time 0.022363 -2023-02-13 17:57:40,695 - Epoch: [89][ 260/ 1207] Overall Loss 0.320085 Objective Loss 0.320085 LR 0.001000 Time 0.022246 -2023-02-13 17:57:40,891 - Epoch: [89][ 270/ 1207] Overall Loss 0.321408 Objective Loss 0.321408 LR 0.001000 Time 0.022145 -2023-02-13 17:57:41,084 - Epoch: [89][ 280/ 1207] Overall Loss 0.321778 Objective Loss 0.321778 LR 0.001000 Time 0.022041 -2023-02-13 17:57:41,280 - Epoch: [89][ 290/ 1207] Overall Loss 0.321451 Objective Loss 0.321451 LR 0.001000 Time 0.021956 -2023-02-13 17:57:41,473 - Epoch: [89][ 300/ 1207] Overall Loss 0.320897 Objective Loss 0.320897 LR 0.001000 Time 0.021868 -2023-02-13 17:57:41,669 - Epoch: [89][ 310/ 1207] Overall Loss 0.320889 Objective Loss 0.320889 LR 0.001000 Time 0.021791 -2023-02-13 17:57:41,862 - Epoch: [89][ 320/ 1207] Overall Loss 0.321805 Objective Loss 0.321805 LR 0.001000 Time 0.021713 -2023-02-13 17:57:42,057 - Epoch: [89][ 330/ 1207] Overall Loss 0.320761 Objective Loss 0.320761 LR 0.001000 Time 0.021646 -2023-02-13 17:57:42,251 - Epoch: [89][ 340/ 1207] Overall Loss 0.321375 Objective Loss 0.321375 LR 0.001000 Time 0.021579 -2023-02-13 17:57:42,447 - Epoch: [89][ 350/ 1207] Overall Loss 0.322025 Objective Loss 0.322025 LR 0.001000 Time 0.021520 -2023-02-13 17:57:42,640 - Epoch: [89][ 360/ 1207] Overall Loss 0.322373 Objective Loss 0.322373 LR 0.001000 Time 0.021458 -2023-02-13 17:57:42,835 - Epoch: [89][ 370/ 1207] Overall Loss 0.321901 Objective Loss 0.321901 LR 0.001000 Time 0.021405 -2023-02-13 17:57:43,028 - Epoch: [89][ 380/ 1207] Overall Loss 0.321742 Objective Loss 0.321742 LR 0.001000 Time 0.021348 -2023-02-13 17:57:43,224 - Epoch: [89][ 390/ 1207] Overall Loss 0.321765 Objective Loss 0.321765 LR 0.001000 Time 0.021302 -2023-02-13 17:57:43,417 - Epoch: [89][ 400/ 1207] Overall Loss 0.321878 Objective Loss 0.321878 LR 0.001000 Time 0.021252 -2023-02-13 17:57:43,612 - Epoch: [89][ 410/ 1207] Overall Loss 0.321701 Objective Loss 0.321701 LR 0.001000 Time 0.021208 -2023-02-13 17:57:43,805 - Epoch: [89][ 420/ 1207] Overall Loss 0.322499 Objective Loss 0.322499 LR 0.001000 Time 0.021161 -2023-02-13 17:57:44,001 - Epoch: [89][ 430/ 1207] Overall Loss 0.322556 Objective Loss 0.322556 LR 0.001000 Time 0.021123 -2023-02-13 17:57:44,194 - Epoch: [89][ 440/ 1207] Overall Loss 0.323066 Objective Loss 0.323066 LR 0.001000 Time 0.021081 -2023-02-13 17:57:44,390 - Epoch: [89][ 450/ 1207] Overall Loss 0.323529 Objective Loss 0.323529 LR 0.001000 Time 0.021047 -2023-02-13 17:57:44,583 - Epoch: [89][ 460/ 1207] Overall Loss 0.323581 Objective Loss 0.323581 LR 0.001000 Time 0.021009 -2023-02-13 17:57:44,778 - Epoch: [89][ 470/ 1207] Overall Loss 0.323894 Objective Loss 0.323894 LR 0.001000 Time 0.020976 -2023-02-13 17:57:44,971 - Epoch: [89][ 480/ 1207] Overall Loss 0.324306 Objective Loss 0.324306 LR 0.001000 Time 0.020940 -2023-02-13 17:57:45,166 - Epoch: [89][ 490/ 1207] Overall Loss 0.323814 Objective Loss 0.323814 LR 0.001000 Time 0.020909 -2023-02-13 17:57:45,359 - Epoch: [89][ 500/ 1207] Overall Loss 0.324629 Objective Loss 0.324629 LR 0.001000 Time 0.020877 -2023-02-13 17:57:45,555 - Epoch: [89][ 510/ 1207] Overall Loss 0.324961 Objective Loss 0.324961 LR 0.001000 Time 0.020851 -2023-02-13 17:57:45,748 - Epoch: [89][ 520/ 1207] Overall Loss 0.325057 Objective Loss 0.325057 LR 0.001000 Time 0.020821 -2023-02-13 17:57:45,944 - Epoch: [89][ 530/ 1207] Overall Loss 0.325710 Objective Loss 0.325710 LR 0.001000 Time 0.020798 -2023-02-13 17:57:46,137 - Epoch: [89][ 540/ 1207] Overall Loss 0.325665 Objective Loss 0.325665 LR 0.001000 Time 0.020769 -2023-02-13 17:57:46,333 - Epoch: [89][ 550/ 1207] Overall Loss 0.325574 Objective Loss 0.325574 LR 0.001000 Time 0.020747 -2023-02-13 17:57:46,526 - Epoch: [89][ 560/ 1207] Overall Loss 0.325574 Objective Loss 0.325574 LR 0.001000 Time 0.020720 -2023-02-13 17:57:46,721 - Epoch: [89][ 570/ 1207] Overall Loss 0.325788 Objective Loss 0.325788 LR 0.001000 Time 0.020698 -2023-02-13 17:57:46,915 - Epoch: [89][ 580/ 1207] Overall Loss 0.325842 Objective Loss 0.325842 LR 0.001000 Time 0.020674 -2023-02-13 17:57:47,110 - Epoch: [89][ 590/ 1207] Overall Loss 0.326553 Objective Loss 0.326553 LR 0.001000 Time 0.020654 -2023-02-13 17:57:47,303 - Epoch: [89][ 600/ 1207] Overall Loss 0.326528 Objective Loss 0.326528 LR 0.001000 Time 0.020632 -2023-02-13 17:57:47,498 - Epoch: [89][ 610/ 1207] Overall Loss 0.326616 Objective Loss 0.326616 LR 0.001000 Time 0.020612 -2023-02-13 17:57:47,691 - Epoch: [89][ 620/ 1207] Overall Loss 0.326433 Objective Loss 0.326433 LR 0.001000 Time 0.020591 -2023-02-13 17:57:47,886 - Epoch: [89][ 630/ 1207] Overall Loss 0.326520 Objective Loss 0.326520 LR 0.001000 Time 0.020572 -2023-02-13 17:57:48,078 - Epoch: [89][ 640/ 1207] Overall Loss 0.326389 Objective Loss 0.326389 LR 0.001000 Time 0.020550 -2023-02-13 17:57:48,273 - Epoch: [89][ 650/ 1207] Overall Loss 0.326337 Objective Loss 0.326337 LR 0.001000 Time 0.020534 -2023-02-13 17:57:48,466 - Epoch: [89][ 660/ 1207] Overall Loss 0.326313 Objective Loss 0.326313 LR 0.001000 Time 0.020515 -2023-02-13 17:57:48,661 - Epoch: [89][ 670/ 1207] Overall Loss 0.326593 Objective Loss 0.326593 LR 0.001000 Time 0.020499 -2023-02-13 17:57:48,854 - Epoch: [89][ 680/ 1207] Overall Loss 0.326971 Objective Loss 0.326971 LR 0.001000 Time 0.020481 -2023-02-13 17:57:49,049 - Epoch: [89][ 690/ 1207] Overall Loss 0.327515 Objective Loss 0.327515 LR 0.001000 Time 0.020467 -2023-02-13 17:57:49,244 - Epoch: [89][ 700/ 1207] Overall Loss 0.327963 Objective Loss 0.327963 LR 0.001000 Time 0.020452 -2023-02-13 17:57:49,440 - Epoch: [89][ 710/ 1207] Overall Loss 0.328453 Objective Loss 0.328453 LR 0.001000 Time 0.020438 -2023-02-13 17:57:49,633 - Epoch: [89][ 720/ 1207] Overall Loss 0.328244 Objective Loss 0.328244 LR 0.001000 Time 0.020423 -2023-02-13 17:57:49,829 - Epoch: [89][ 730/ 1207] Overall Loss 0.328340 Objective Loss 0.328340 LR 0.001000 Time 0.020410 -2023-02-13 17:57:50,022 - Epoch: [89][ 740/ 1207] Overall Loss 0.328435 Objective Loss 0.328435 LR 0.001000 Time 0.020395 -2023-02-13 17:57:50,217 - Epoch: [89][ 750/ 1207] Overall Loss 0.328227 Objective Loss 0.328227 LR 0.001000 Time 0.020384 -2023-02-13 17:57:50,411 - Epoch: [89][ 760/ 1207] Overall Loss 0.328382 Objective Loss 0.328382 LR 0.001000 Time 0.020369 -2023-02-13 17:57:50,606 - Epoch: [89][ 770/ 1207] Overall Loss 0.328509 Objective Loss 0.328509 LR 0.001000 Time 0.020358 -2023-02-13 17:57:50,799 - Epoch: [89][ 780/ 1207] Overall Loss 0.328727 Objective Loss 0.328727 LR 0.001000 Time 0.020344 -2023-02-13 17:57:50,995 - Epoch: [89][ 790/ 1207] Overall Loss 0.328819 Objective Loss 0.328819 LR 0.001000 Time 0.020333 -2023-02-13 17:57:51,188 - Epoch: [89][ 800/ 1207] Overall Loss 0.328738 Objective Loss 0.328738 LR 0.001000 Time 0.020320 -2023-02-13 17:57:51,384 - Epoch: [89][ 810/ 1207] Overall Loss 0.328373 Objective Loss 0.328373 LR 0.001000 Time 0.020311 -2023-02-13 17:57:51,577 - Epoch: [89][ 820/ 1207] Overall Loss 0.328462 Objective Loss 0.328462 LR 0.001000 Time 0.020298 -2023-02-13 17:57:51,773 - Epoch: [89][ 830/ 1207] Overall Loss 0.328104 Objective Loss 0.328104 LR 0.001000 Time 0.020289 -2023-02-13 17:57:51,966 - Epoch: [89][ 840/ 1207] Overall Loss 0.328201 Objective Loss 0.328201 LR 0.001000 Time 0.020277 -2023-02-13 17:57:52,160 - Epoch: [89][ 850/ 1207] Overall Loss 0.328403 Objective Loss 0.328403 LR 0.001000 Time 0.020266 -2023-02-13 17:57:52,354 - Epoch: [89][ 860/ 1207] Overall Loss 0.328489 Objective Loss 0.328489 LR 0.001000 Time 0.020256 -2023-02-13 17:57:52,549 - Epoch: [89][ 870/ 1207] Overall Loss 0.327953 Objective Loss 0.327953 LR 0.001000 Time 0.020247 -2023-02-13 17:57:52,742 - Epoch: [89][ 880/ 1207] Overall Loss 0.328355 Objective Loss 0.328355 LR 0.001000 Time 0.020235 -2023-02-13 17:57:52,936 - Epoch: [89][ 890/ 1207] Overall Loss 0.328432 Objective Loss 0.328432 LR 0.001000 Time 0.020226 -2023-02-13 17:57:53,130 - Epoch: [89][ 900/ 1207] Overall Loss 0.328631 Objective Loss 0.328631 LR 0.001000 Time 0.020216 -2023-02-13 17:57:53,325 - Epoch: [89][ 910/ 1207] Overall Loss 0.328499 Objective Loss 0.328499 LR 0.001000 Time 0.020208 -2023-02-13 17:57:53,518 - Epoch: [89][ 920/ 1207] Overall Loss 0.328744 Objective Loss 0.328744 LR 0.001000 Time 0.020198 -2023-02-13 17:57:53,714 - Epoch: [89][ 930/ 1207] Overall Loss 0.329180 Objective Loss 0.329180 LR 0.001000 Time 0.020191 -2023-02-13 17:57:53,908 - Epoch: [89][ 940/ 1207] Overall Loss 0.329477 Objective Loss 0.329477 LR 0.001000 Time 0.020182 -2023-02-13 17:57:54,103 - Epoch: [89][ 950/ 1207] Overall Loss 0.329691 Objective Loss 0.329691 LR 0.001000 Time 0.020175 -2023-02-13 17:57:54,298 - Epoch: [89][ 960/ 1207] Overall Loss 0.330040 Objective Loss 0.330040 LR 0.001000 Time 0.020167 -2023-02-13 17:57:54,493 - Epoch: [89][ 970/ 1207] Overall Loss 0.329701 Objective Loss 0.329701 LR 0.001000 Time 0.020159 -2023-02-13 17:57:54,686 - Epoch: [89][ 980/ 1207] Overall Loss 0.329723 Objective Loss 0.329723 LR 0.001000 Time 0.020150 -2023-02-13 17:57:54,881 - Epoch: [89][ 990/ 1207] Overall Loss 0.329755 Objective Loss 0.329755 LR 0.001000 Time 0.020143 -2023-02-13 17:57:55,074 - Epoch: [89][ 1000/ 1207] Overall Loss 0.329972 Objective Loss 0.329972 LR 0.001000 Time 0.020135 -2023-02-13 17:57:55,269 - Epoch: [89][ 1010/ 1207] Overall Loss 0.330151 Objective Loss 0.330151 LR 0.001000 Time 0.020129 -2023-02-13 17:57:55,462 - Epoch: [89][ 1020/ 1207] Overall Loss 0.330330 Objective Loss 0.330330 LR 0.001000 Time 0.020120 -2023-02-13 17:57:55,657 - Epoch: [89][ 1030/ 1207] Overall Loss 0.330433 Objective Loss 0.330433 LR 0.001000 Time 0.020114 -2023-02-13 17:57:55,851 - Epoch: [89][ 1040/ 1207] Overall Loss 0.330625 Objective Loss 0.330625 LR 0.001000 Time 0.020106 -2023-02-13 17:57:56,047 - Epoch: [89][ 1050/ 1207] Overall Loss 0.330756 Objective Loss 0.330756 LR 0.001000 Time 0.020101 -2023-02-13 17:57:56,240 - Epoch: [89][ 1060/ 1207] Overall Loss 0.331103 Objective Loss 0.331103 LR 0.001000 Time 0.020093 -2023-02-13 17:57:56,435 - Epoch: [89][ 1070/ 1207] Overall Loss 0.331027 Objective Loss 0.331027 LR 0.001000 Time 0.020087 -2023-02-13 17:57:56,628 - Epoch: [89][ 1080/ 1207] Overall Loss 0.331048 Objective Loss 0.331048 LR 0.001000 Time 0.020079 -2023-02-13 17:57:56,822 - Epoch: [89][ 1090/ 1207] Overall Loss 0.330948 Objective Loss 0.330948 LR 0.001000 Time 0.020073 -2023-02-13 17:57:57,016 - Epoch: [89][ 1100/ 1207] Overall Loss 0.330877 Objective Loss 0.330877 LR 0.001000 Time 0.020067 -2023-02-13 17:57:57,212 - Epoch: [89][ 1110/ 1207] Overall Loss 0.331104 Objective Loss 0.331104 LR 0.001000 Time 0.020062 -2023-02-13 17:57:57,406 - Epoch: [89][ 1120/ 1207] Overall Loss 0.331047 Objective Loss 0.331047 LR 0.001000 Time 0.020055 -2023-02-13 17:57:57,600 - Epoch: [89][ 1130/ 1207] Overall Loss 0.331124 Objective Loss 0.331124 LR 0.001000 Time 0.020050 -2023-02-13 17:57:57,793 - Epoch: [89][ 1140/ 1207] Overall Loss 0.331114 Objective Loss 0.331114 LR 0.001000 Time 0.020043 -2023-02-13 17:57:57,988 - Epoch: [89][ 1150/ 1207] Overall Loss 0.330936 Objective Loss 0.330936 LR 0.001000 Time 0.020037 -2023-02-13 17:57:58,181 - Epoch: [89][ 1160/ 1207] Overall Loss 0.330960 Objective Loss 0.330960 LR 0.001000 Time 0.020031 -2023-02-13 17:57:58,377 - Epoch: [89][ 1170/ 1207] Overall Loss 0.330875 Objective Loss 0.330875 LR 0.001000 Time 0.020027 -2023-02-13 17:57:58,571 - Epoch: [89][ 1180/ 1207] Overall Loss 0.330827 Objective Loss 0.330827 LR 0.001000 Time 0.020021 -2023-02-13 17:57:58,765 - Epoch: [89][ 1190/ 1207] Overall Loss 0.330769 Objective Loss 0.330769 LR 0.001000 Time 0.020016 -2023-02-13 17:57:59,015 - Epoch: [89][ 1200/ 1207] Overall Loss 0.330951 Objective Loss 0.330951 LR 0.001000 Time 0.020057 -2023-02-13 17:57:59,131 - Epoch: [89][ 1207/ 1207] Overall Loss 0.331144 Objective Loss 0.331144 Top1 81.402439 Top5 96.341463 LR 0.001000 Time 0.020036 -2023-02-13 17:57:59,201 - --- validate (epoch=89)----------- -2023-02-13 17:57:59,202 - 34311 samples (256 per mini-batch) -2023-02-13 17:57:59,605 - Epoch: [89][ 10/ 135] Loss 0.364671 Top1 82.421875 Top5 97.148438 -2023-02-13 17:57:59,734 - Epoch: [89][ 20/ 135] Loss 0.368693 Top1 82.109375 Top5 97.246094 -2023-02-13 17:57:59,863 - Epoch: [89][ 30/ 135] Loss 0.363312 Top1 81.992188 Top5 97.213542 -2023-02-13 17:57:59,991 - Epoch: [89][ 40/ 135] Loss 0.360449 Top1 81.718750 Top5 97.099609 -2023-02-13 17:58:00,119 - Epoch: [89][ 50/ 135] Loss 0.365976 Top1 81.695312 Top5 97.000000 -2023-02-13 17:58:00,248 - Epoch: [89][ 60/ 135] Loss 0.366270 Top1 81.816406 Top5 97.024740 -2023-02-13 17:58:00,375 - Epoch: [89][ 70/ 135] Loss 0.366427 Top1 81.813616 Top5 97.020089 -2023-02-13 17:58:00,503 - Epoch: [89][ 80/ 135] Loss 0.370477 Top1 81.650391 Top5 97.031250 -2023-02-13 17:58:00,631 - Epoch: [89][ 90/ 135] Loss 0.371054 Top1 81.432292 Top5 96.918403 -2023-02-13 17:58:00,762 - Epoch: [89][ 100/ 135] Loss 0.373029 Top1 81.343750 Top5 96.878906 -2023-02-13 17:58:00,889 - Epoch: [89][ 110/ 135] Loss 0.374337 Top1 81.321023 Top5 96.882102 -2023-02-13 17:58:01,018 - Epoch: [89][ 120/ 135] Loss 0.373819 Top1 81.344401 Top5 96.884766 -2023-02-13 17:58:01,149 - Epoch: [89][ 130/ 135] Loss 0.372119 Top1 81.349159 Top5 96.884014 -2023-02-13 17:58:01,193 - Epoch: [89][ 135/ 135] Loss 0.368900 Top1 81.364577 Top5 96.884381 -2023-02-13 17:58:01,268 - ==> Top1: 81.365 Top5: 96.884 Loss: 0.369 - -2023-02-13 17:58:01,269 - ==> Confusion: -[[ 839 5 7 0 20 3 0 0 1 58 0 5 1 4 6 7 1 2 3 1 4] - [ 3 935 3 5 19 19 3 12 2 0 3 6 3 0 0 2 3 0 6 2 7] - [ 7 2 949 17 3 2 18 11 0 2 2 2 3 5 4 11 0 7 4 5 4] - [ 7 1 11 898 2 6 2 3 2 3 15 1 10 2 23 6 1 4 14 0 5] - [ 13 8 1 0 994 4 1 1 1 2 1 12 2 3 5 6 7 1 0 2 2] - [ 6 29 1 2 11 943 5 16 0 3 3 15 4 11 1 1 3 0 0 9 7] - [ 2 3 22 0 3 6 1018 9 0 4 4 1 2 2 0 7 2 1 1 11 1] - [ 2 14 11 1 4 33 6 911 1 1 3 9 1 2 1 0 0 0 9 12 3] - [ 27 4 0 2 3 0 0 1 852 58 4 5 2 18 23 2 1 1 5 0 1] - [ 81 2 1 0 9 1 0 2 26 859 0 2 1 14 5 1 2 1 0 2 3] - [ 1 3 6 6 1 3 2 4 18 4 958 3 1 15 2 1 2 0 14 0 7] - [ 1 5 0 0 7 17 2 3 3 1 0 911 18 5 1 6 4 5 1 11 4] - [ 1 1 1 3 2 2 0 3 3 0 0 62 830 3 3 10 4 17 2 3 9] - [ 5 2 4 0 14 10 1 4 10 20 12 6 3 911 4 5 4 2 0 4 3] - [ 8 2 3 8 18 4 0 1 25 12 2 3 7 2 975 2 1 4 9 0 6] - [ 3 2 3 0 6 1 5 1 0 0 0 5 6 6 0 979 12 8 0 4 5] - [ 3 7 3 1 10 1 0 0 1 3 1 2 2 3 0 14 990 3 0 4 13] - [ 6 2 1 6 2 5 2 1 1 1 1 16 19 0 0 28 1 948 1 5 5] - [ 5 10 5 18 1 3 1 39 3 0 4 3 7 0 16 4 0 1 963 2 1] - [ 0 3 2 0 4 12 4 13 0 0 1 19 3 4 0 5 3 4 0 1065 6] - [ 207 286 252 128 260 242 116 201 93 127 188 162 313 344 162 274 289 83 150 368 9189]] - -2023-02-13 17:58:01,270 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:58:01,270 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:58:01,276 - - -2023-02-13 17:58:01,276 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:58:02,160 - Epoch: [90][ 10/ 1207] Overall Loss 0.320997 Objective Loss 0.320997 LR 0.001000 Time 0.088317 -2023-02-13 17:58:02,355 - Epoch: [90][ 20/ 1207] Overall Loss 0.303072 Objective Loss 0.303072 LR 0.001000 Time 0.053891 -2023-02-13 17:58:02,549 - Epoch: [90][ 30/ 1207] Overall Loss 0.314520 Objective Loss 0.314520 LR 0.001000 Time 0.042376 -2023-02-13 17:58:02,740 - Epoch: [90][ 40/ 1207] Overall Loss 0.322940 Objective Loss 0.322940 LR 0.001000 Time 0.036549 -2023-02-13 17:58:02,934 - Epoch: [90][ 50/ 1207] Overall Loss 0.325608 Objective Loss 0.325608 LR 0.001000 Time 0.033106 -2023-02-13 17:58:03,124 - Epoch: [90][ 60/ 1207] Overall Loss 0.331813 Objective Loss 0.331813 LR 0.001000 Time 0.030757 -2023-02-13 17:58:03,318 - Epoch: [90][ 70/ 1207] Overall Loss 0.331021 Objective Loss 0.331021 LR 0.001000 Time 0.029132 -2023-02-13 17:58:03,509 - Epoch: [90][ 80/ 1207] Overall Loss 0.332163 Objective Loss 0.332163 LR 0.001000 Time 0.027867 -2023-02-13 17:58:03,703 - Epoch: [90][ 90/ 1207] Overall Loss 0.331505 Objective Loss 0.331505 LR 0.001000 Time 0.026921 -2023-02-13 17:58:03,893 - Epoch: [90][ 100/ 1207] Overall Loss 0.331013 Objective Loss 0.331013 LR 0.001000 Time 0.026130 -2023-02-13 17:58:04,087 - Epoch: [90][ 110/ 1207] Overall Loss 0.329214 Objective Loss 0.329214 LR 0.001000 Time 0.025514 -2023-02-13 17:58:04,279 - Epoch: [90][ 120/ 1207] Overall Loss 0.331234 Objective Loss 0.331234 LR 0.001000 Time 0.024983 -2023-02-13 17:58:04,473 - Epoch: [90][ 130/ 1207] Overall Loss 0.330331 Objective Loss 0.330331 LR 0.001000 Time 0.024551 -2023-02-13 17:58:04,663 - Epoch: [90][ 140/ 1207] Overall Loss 0.329153 Objective Loss 0.329153 LR 0.001000 Time 0.024154 -2023-02-13 17:58:04,857 - Epoch: [90][ 150/ 1207] Overall Loss 0.327444 Objective Loss 0.327444 LR 0.001000 Time 0.023833 -2023-02-13 17:58:05,047 - Epoch: [90][ 160/ 1207] Overall Loss 0.328174 Objective Loss 0.328174 LR 0.001000 Time 0.023529 -2023-02-13 17:58:05,242 - Epoch: [90][ 170/ 1207] Overall Loss 0.331738 Objective Loss 0.331738 LR 0.001000 Time 0.023290 -2023-02-13 17:58:05,433 - Epoch: [90][ 180/ 1207] Overall Loss 0.331364 Objective Loss 0.331364 LR 0.001000 Time 0.023054 -2023-02-13 17:58:05,627 - Epoch: [90][ 190/ 1207] Overall Loss 0.328421 Objective Loss 0.328421 LR 0.001000 Time 0.022857 -2023-02-13 17:58:05,818 - Epoch: [90][ 200/ 1207] Overall Loss 0.328075 Objective Loss 0.328075 LR 0.001000 Time 0.022671 -2023-02-13 17:58:06,013 - Epoch: [90][ 210/ 1207] Overall Loss 0.328944 Objective Loss 0.328944 LR 0.001000 Time 0.022519 -2023-02-13 17:58:06,204 - Epoch: [90][ 220/ 1207] Overall Loss 0.330160 Objective Loss 0.330160 LR 0.001000 Time 0.022362 -2023-02-13 17:58:06,400 - Epoch: [90][ 230/ 1207] Overall Loss 0.329239 Objective Loss 0.329239 LR 0.001000 Time 0.022236 -2023-02-13 17:58:06,591 - Epoch: [90][ 240/ 1207] Overall Loss 0.328721 Objective Loss 0.328721 LR 0.001000 Time 0.022108 -2023-02-13 17:58:06,786 - Epoch: [90][ 250/ 1207] Overall Loss 0.329797 Objective Loss 0.329797 LR 0.001000 Time 0.022000 -2023-02-13 17:58:06,978 - Epoch: [90][ 260/ 1207] Overall Loss 0.329328 Objective Loss 0.329328 LR 0.001000 Time 0.021890 -2023-02-13 17:58:07,172 - Epoch: [90][ 270/ 1207] Overall Loss 0.329508 Objective Loss 0.329508 LR 0.001000 Time 0.021799 -2023-02-13 17:58:07,364 - Epoch: [90][ 280/ 1207] Overall Loss 0.329729 Objective Loss 0.329729 LR 0.001000 Time 0.021703 -2023-02-13 17:58:07,558 - Epoch: [90][ 290/ 1207] Overall Loss 0.330528 Objective Loss 0.330528 LR 0.001000 Time 0.021623 -2023-02-13 17:58:07,749 - Epoch: [90][ 300/ 1207] Overall Loss 0.329762 Objective Loss 0.329762 LR 0.001000 Time 0.021536 -2023-02-13 17:58:07,943 - Epoch: [90][ 310/ 1207] Overall Loss 0.329573 Objective Loss 0.329573 LR 0.001000 Time 0.021466 -2023-02-13 17:58:08,133 - Epoch: [90][ 320/ 1207] Overall Loss 0.329957 Objective Loss 0.329957 LR 0.001000 Time 0.021391 -2023-02-13 17:58:08,328 - Epoch: [90][ 330/ 1207] Overall Loss 0.330285 Objective Loss 0.330285 LR 0.001000 Time 0.021332 -2023-02-13 17:58:08,519 - Epoch: [90][ 340/ 1207] Overall Loss 0.330021 Objective Loss 0.330021 LR 0.001000 Time 0.021265 -2023-02-13 17:58:08,713 - Epoch: [90][ 350/ 1207] Overall Loss 0.329702 Objective Loss 0.329702 LR 0.001000 Time 0.021211 -2023-02-13 17:58:08,904 - Epoch: [90][ 360/ 1207] Overall Loss 0.330314 Objective Loss 0.330314 LR 0.001000 Time 0.021149 -2023-02-13 17:58:09,098 - Epoch: [90][ 370/ 1207] Overall Loss 0.330360 Objective Loss 0.330360 LR 0.001000 Time 0.021102 -2023-02-13 17:58:09,290 - Epoch: [90][ 380/ 1207] Overall Loss 0.329348 Objective Loss 0.329348 LR 0.001000 Time 0.021050 -2023-02-13 17:58:09,484 - Epoch: [90][ 390/ 1207] Overall Loss 0.329111 Objective Loss 0.329111 LR 0.001000 Time 0.021007 -2023-02-13 17:58:09,675 - Epoch: [90][ 400/ 1207] Overall Loss 0.329015 Objective Loss 0.329015 LR 0.001000 Time 0.020959 -2023-02-13 17:58:09,870 - Epoch: [90][ 410/ 1207] Overall Loss 0.327741 Objective Loss 0.327741 LR 0.001000 Time 0.020921 -2023-02-13 17:58:10,061 - Epoch: [90][ 420/ 1207] Overall Loss 0.328008 Objective Loss 0.328008 LR 0.001000 Time 0.020878 -2023-02-13 17:58:10,256 - Epoch: [90][ 430/ 1207] Overall Loss 0.328390 Objective Loss 0.328390 LR 0.001000 Time 0.020845 -2023-02-13 17:58:10,447 - Epoch: [90][ 440/ 1207] Overall Loss 0.328955 Objective Loss 0.328955 LR 0.001000 Time 0.020806 -2023-02-13 17:58:10,642 - Epoch: [90][ 450/ 1207] Overall Loss 0.328875 Objective Loss 0.328875 LR 0.001000 Time 0.020774 -2023-02-13 17:58:10,833 - Epoch: [90][ 460/ 1207] Overall Loss 0.328605 Objective Loss 0.328605 LR 0.001000 Time 0.020739 -2023-02-13 17:58:11,023 - Epoch: [90][ 470/ 1207] Overall Loss 0.329752 Objective Loss 0.329752 LR 0.001000 Time 0.020701 -2023-02-13 17:58:11,218 - Epoch: [90][ 480/ 1207] Overall Loss 0.330017 Objective Loss 0.330017 LR 0.001000 Time 0.020675 -2023-02-13 17:58:11,411 - Epoch: [90][ 490/ 1207] Overall Loss 0.330042 Objective Loss 0.330042 LR 0.001000 Time 0.020646 -2023-02-13 17:58:11,606 - Epoch: [90][ 500/ 1207] Overall Loss 0.330079 Objective Loss 0.330079 LR 0.001000 Time 0.020621 -2023-02-13 17:58:11,798 - Epoch: [90][ 510/ 1207] Overall Loss 0.330091 Objective Loss 0.330091 LR 0.001000 Time 0.020594 -2023-02-13 17:58:11,994 - Epoch: [90][ 520/ 1207] Overall Loss 0.330326 Objective Loss 0.330326 LR 0.001000 Time 0.020573 -2023-02-13 17:58:12,186 - Epoch: [90][ 530/ 1207] Overall Loss 0.330675 Objective Loss 0.330675 LR 0.001000 Time 0.020548 -2023-02-13 17:58:12,382 - Epoch: [90][ 540/ 1207] Overall Loss 0.330984 Objective Loss 0.330984 LR 0.001000 Time 0.020528 -2023-02-13 17:58:12,574 - Epoch: [90][ 550/ 1207] Overall Loss 0.330892 Objective Loss 0.330892 LR 0.001000 Time 0.020504 -2023-02-13 17:58:12,769 - Epoch: [90][ 560/ 1207] Overall Loss 0.331339 Objective Loss 0.331339 LR 0.001000 Time 0.020485 -2023-02-13 17:58:12,961 - Epoch: [90][ 570/ 1207] Overall Loss 0.330979 Objective Loss 0.330979 LR 0.001000 Time 0.020462 -2023-02-13 17:58:13,155 - Epoch: [90][ 580/ 1207] Overall Loss 0.332190 Objective Loss 0.332190 LR 0.001000 Time 0.020443 -2023-02-13 17:58:13,348 - Epoch: [90][ 590/ 1207] Overall Loss 0.332073 Objective Loss 0.332073 LR 0.001000 Time 0.020423 -2023-02-13 17:58:13,543 - Epoch: [90][ 600/ 1207] Overall Loss 0.332308 Objective Loss 0.332308 LR 0.001000 Time 0.020407 -2023-02-13 17:58:13,735 - Epoch: [90][ 610/ 1207] Overall Loss 0.332813 Objective Loss 0.332813 LR 0.001000 Time 0.020386 -2023-02-13 17:58:13,930 - Epoch: [90][ 620/ 1207] Overall Loss 0.332715 Objective Loss 0.332715 LR 0.001000 Time 0.020371 -2023-02-13 17:58:14,122 - Epoch: [90][ 630/ 1207] Overall Loss 0.333487 Objective Loss 0.333487 LR 0.001000 Time 0.020353 -2023-02-13 17:58:14,318 - Epoch: [90][ 640/ 1207] Overall Loss 0.333638 Objective Loss 0.333638 LR 0.001000 Time 0.020340 -2023-02-13 17:58:14,510 - Epoch: [90][ 650/ 1207] Overall Loss 0.333785 Objective Loss 0.333785 LR 0.001000 Time 0.020322 -2023-02-13 17:58:14,705 - Epoch: [90][ 660/ 1207] Overall Loss 0.332968 Objective Loss 0.332968 LR 0.001000 Time 0.020308 -2023-02-13 17:58:14,897 - Epoch: [90][ 670/ 1207] Overall Loss 0.333295 Objective Loss 0.333295 LR 0.001000 Time 0.020292 -2023-02-13 17:58:15,092 - Epoch: [90][ 680/ 1207] Overall Loss 0.333504 Objective Loss 0.333504 LR 0.001000 Time 0.020279 -2023-02-13 17:58:15,285 - Epoch: [90][ 690/ 1207] Overall Loss 0.334090 Objective Loss 0.334090 LR 0.001000 Time 0.020265 -2023-02-13 17:58:15,481 - Epoch: [90][ 700/ 1207] Overall Loss 0.334142 Objective Loss 0.334142 LR 0.001000 Time 0.020254 -2023-02-13 17:58:15,674 - Epoch: [90][ 710/ 1207] Overall Loss 0.334206 Objective Loss 0.334206 LR 0.001000 Time 0.020240 -2023-02-13 17:58:15,869 - Epoch: [90][ 720/ 1207] Overall Loss 0.334345 Objective Loss 0.334345 LR 0.001000 Time 0.020231 -2023-02-13 17:58:16,062 - Epoch: [90][ 730/ 1207] Overall Loss 0.334864 Objective Loss 0.334864 LR 0.001000 Time 0.020217 -2023-02-13 17:58:16,257 - Epoch: [90][ 740/ 1207] Overall Loss 0.334926 Objective Loss 0.334926 LR 0.001000 Time 0.020206 -2023-02-13 17:58:16,446 - Epoch: [90][ 750/ 1207] Overall Loss 0.335329 Objective Loss 0.335329 LR 0.001000 Time 0.020189 -2023-02-13 17:58:16,635 - Epoch: [90][ 760/ 1207] Overall Loss 0.335431 Objective Loss 0.335431 LR 0.001000 Time 0.020171 -2023-02-13 17:58:16,823 - Epoch: [90][ 770/ 1207] Overall Loss 0.335429 Objective Loss 0.335429 LR 0.001000 Time 0.020153 -2023-02-13 17:58:17,011 - Epoch: [90][ 780/ 1207] Overall Loss 0.335368 Objective Loss 0.335368 LR 0.001000 Time 0.020136 -2023-02-13 17:58:17,200 - Epoch: [90][ 790/ 1207] Overall Loss 0.335309 Objective Loss 0.335309 LR 0.001000 Time 0.020119 -2023-02-13 17:58:17,389 - Epoch: [90][ 800/ 1207] Overall Loss 0.335527 Objective Loss 0.335527 LR 0.001000 Time 0.020104 -2023-02-13 17:58:17,578 - Epoch: [90][ 810/ 1207] Overall Loss 0.335710 Objective Loss 0.335710 LR 0.001000 Time 0.020088 -2023-02-13 17:58:17,766 - Epoch: [90][ 820/ 1207] Overall Loss 0.335799 Objective Loss 0.335799 LR 0.001000 Time 0.020072 -2023-02-13 17:58:17,954 - Epoch: [90][ 830/ 1207] Overall Loss 0.335648 Objective Loss 0.335648 LR 0.001000 Time 0.020056 -2023-02-13 17:58:18,142 - Epoch: [90][ 840/ 1207] Overall Loss 0.335266 Objective Loss 0.335266 LR 0.001000 Time 0.020041 -2023-02-13 17:58:18,331 - Epoch: [90][ 850/ 1207] Overall Loss 0.335195 Objective Loss 0.335195 LR 0.001000 Time 0.020027 -2023-02-13 17:58:18,520 - Epoch: [90][ 860/ 1207] Overall Loss 0.335497 Objective Loss 0.335497 LR 0.001000 Time 0.020014 -2023-02-13 17:58:18,709 - Epoch: [90][ 870/ 1207] Overall Loss 0.335798 Objective Loss 0.335798 LR 0.001000 Time 0.020000 -2023-02-13 17:58:18,897 - Epoch: [90][ 880/ 1207] Overall Loss 0.335696 Objective Loss 0.335696 LR 0.001000 Time 0.019987 -2023-02-13 17:58:19,086 - Epoch: [90][ 890/ 1207] Overall Loss 0.335868 Objective Loss 0.335868 LR 0.001000 Time 0.019973 -2023-02-13 17:58:19,274 - Epoch: [90][ 900/ 1207] Overall Loss 0.335674 Objective Loss 0.335674 LR 0.001000 Time 0.019961 -2023-02-13 17:58:19,464 - Epoch: [90][ 910/ 1207] Overall Loss 0.335853 Objective Loss 0.335853 LR 0.001000 Time 0.019949 -2023-02-13 17:58:19,653 - Epoch: [90][ 920/ 1207] Overall Loss 0.335647 Objective Loss 0.335647 LR 0.001000 Time 0.019937 -2023-02-13 17:58:19,841 - Epoch: [90][ 930/ 1207] Overall Loss 0.335529 Objective Loss 0.335529 LR 0.001000 Time 0.019925 -2023-02-13 17:58:20,030 - Epoch: [90][ 940/ 1207] Overall Loss 0.335434 Objective Loss 0.335434 LR 0.001000 Time 0.019913 -2023-02-13 17:58:20,218 - Epoch: [90][ 950/ 1207] Overall Loss 0.335488 Objective Loss 0.335488 LR 0.001000 Time 0.019902 -2023-02-13 17:58:20,408 - Epoch: [90][ 960/ 1207] Overall Loss 0.335339 Objective Loss 0.335339 LR 0.001000 Time 0.019891 -2023-02-13 17:58:20,596 - Epoch: [90][ 970/ 1207] Overall Loss 0.335410 Objective Loss 0.335410 LR 0.001000 Time 0.019880 -2023-02-13 17:58:20,784 - Epoch: [90][ 980/ 1207] Overall Loss 0.335622 Objective Loss 0.335622 LR 0.001000 Time 0.019869 -2023-02-13 17:58:20,974 - Epoch: [90][ 990/ 1207] Overall Loss 0.335880 Objective Loss 0.335880 LR 0.001000 Time 0.019859 -2023-02-13 17:58:21,163 - Epoch: [90][ 1000/ 1207] Overall Loss 0.335872 Objective Loss 0.335872 LR 0.001000 Time 0.019849 -2023-02-13 17:58:21,351 - Epoch: [90][ 1010/ 1207] Overall Loss 0.335581 Objective Loss 0.335581 LR 0.001000 Time 0.019839 -2023-02-13 17:58:21,539 - Epoch: [90][ 1020/ 1207] Overall Loss 0.335701 Objective Loss 0.335701 LR 0.001000 Time 0.019828 -2023-02-13 17:58:21,728 - Epoch: [90][ 1030/ 1207] Overall Loss 0.335607 Objective Loss 0.335607 LR 0.001000 Time 0.019819 -2023-02-13 17:58:21,917 - Epoch: [90][ 1040/ 1207] Overall Loss 0.335648 Objective Loss 0.335648 LR 0.001000 Time 0.019810 -2023-02-13 17:58:22,105 - Epoch: [90][ 1050/ 1207] Overall Loss 0.335566 Objective Loss 0.335566 LR 0.001000 Time 0.019800 -2023-02-13 17:58:22,294 - Epoch: [90][ 1060/ 1207] Overall Loss 0.335437 Objective Loss 0.335437 LR 0.001000 Time 0.019791 -2023-02-13 17:58:22,484 - Epoch: [90][ 1070/ 1207] Overall Loss 0.335799 Objective Loss 0.335799 LR 0.001000 Time 0.019783 -2023-02-13 17:58:22,672 - Epoch: [90][ 1080/ 1207] Overall Loss 0.336098 Objective Loss 0.336098 LR 0.001000 Time 0.019774 -2023-02-13 17:58:22,861 - Epoch: [90][ 1090/ 1207] Overall Loss 0.336310 Objective Loss 0.336310 LR 0.001000 Time 0.019765 -2023-02-13 17:58:23,050 - Epoch: [90][ 1100/ 1207] Overall Loss 0.336477 Objective Loss 0.336477 LR 0.001000 Time 0.019757 -2023-02-13 17:58:23,239 - Epoch: [90][ 1110/ 1207] Overall Loss 0.336403 Objective Loss 0.336403 LR 0.001000 Time 0.019749 -2023-02-13 17:58:23,428 - Epoch: [90][ 1120/ 1207] Overall Loss 0.336202 Objective Loss 0.336202 LR 0.001000 Time 0.019741 -2023-02-13 17:58:23,617 - Epoch: [90][ 1130/ 1207] Overall Loss 0.335964 Objective Loss 0.335964 LR 0.001000 Time 0.019733 -2023-02-13 17:58:23,806 - Epoch: [90][ 1140/ 1207] Overall Loss 0.335990 Objective Loss 0.335990 LR 0.001000 Time 0.019726 -2023-02-13 17:58:23,995 - Epoch: [90][ 1150/ 1207] Overall Loss 0.335962 Objective Loss 0.335962 LR 0.001000 Time 0.019718 -2023-02-13 17:58:24,184 - Epoch: [90][ 1160/ 1207] Overall Loss 0.336239 Objective Loss 0.336239 LR 0.001000 Time 0.019711 -2023-02-13 17:58:24,373 - Epoch: [90][ 1170/ 1207] Overall Loss 0.336211 Objective Loss 0.336211 LR 0.001000 Time 0.019703 -2023-02-13 17:58:24,562 - Epoch: [90][ 1180/ 1207] Overall Loss 0.336014 Objective Loss 0.336014 LR 0.001000 Time 0.019696 -2023-02-13 17:58:24,752 - Epoch: [90][ 1190/ 1207] Overall Loss 0.336110 Objective Loss 0.336110 LR 0.001000 Time 0.019690 -2023-02-13 17:58:24,996 - Epoch: [90][ 1200/ 1207] Overall Loss 0.335990 Objective Loss 0.335990 LR 0.001000 Time 0.019729 -2023-02-13 17:58:25,112 - Epoch: [90][ 1207/ 1207] Overall Loss 0.336148 Objective Loss 0.336148 Top1 82.926829 Top5 97.865854 LR 0.001000 Time 0.019710 -2023-02-13 17:58:25,193 - --- validate (epoch=90)----------- -2023-02-13 17:58:25,194 - 34311 samples (256 per mini-batch) -2023-02-13 17:58:25,597 - Epoch: [90][ 10/ 135] Loss 0.403501 Top1 81.250000 Top5 96.484375 -2023-02-13 17:58:25,727 - Epoch: [90][ 20/ 135] Loss 0.387316 Top1 81.621094 Top5 96.855469 -2023-02-13 17:58:25,856 - Epoch: [90][ 30/ 135] Loss 0.376814 Top1 81.770833 Top5 97.057292 -2023-02-13 17:58:25,982 - Epoch: [90][ 40/ 135] Loss 0.373284 Top1 81.582031 Top5 96.972656 -2023-02-13 17:58:26,107 - Epoch: [90][ 50/ 135] Loss 0.371839 Top1 81.554688 Top5 97.054688 -2023-02-13 17:58:26,231 - Epoch: [90][ 60/ 135] Loss 0.371166 Top1 81.725260 Top5 97.076823 -2023-02-13 17:58:26,357 - Epoch: [90][ 70/ 135] Loss 0.363809 Top1 81.863839 Top5 97.070312 -2023-02-13 17:58:26,480 - Epoch: [90][ 80/ 135] Loss 0.362424 Top1 82.001953 Top5 97.163086 -2023-02-13 17:58:26,605 - Epoch: [90][ 90/ 135] Loss 0.365336 Top1 82.057292 Top5 97.057292 -2023-02-13 17:58:26,731 - Epoch: [90][ 100/ 135] Loss 0.363617 Top1 81.960938 Top5 97.062500 -2023-02-13 17:58:26,862 - Epoch: [90][ 110/ 135] Loss 0.364256 Top1 81.931818 Top5 97.066761 -2023-02-13 17:58:26,993 - Epoch: [90][ 120/ 135] Loss 0.365023 Top1 81.914062 Top5 97.057292 -2023-02-13 17:58:27,123 - Epoch: [90][ 130/ 135] Loss 0.363037 Top1 81.935096 Top5 97.070312 -2023-02-13 17:58:27,167 - Epoch: [90][ 135/ 135] Loss 0.364108 Top1 81.900848 Top5 97.059252 -2023-02-13 17:58:27,238 - ==> Top1: 81.901 Top5: 97.059 Loss: 0.364 - -2023-02-13 17:58:27,239 - ==> Confusion: -[[ 835 5 6 0 16 5 0 3 8 46 0 7 1 5 8 2 2 4 1 4 9] - [ 0 918 0 1 10 30 1 27 3 2 7 4 5 1 3 2 2 1 6 4 6] - [ 6 4 937 15 4 2 23 18 0 1 4 1 1 5 5 11 0 6 7 3 5] - [ 5 0 14 897 5 6 6 0 3 0 11 1 10 1 16 1 2 7 22 1 8] - [ 12 7 0 1 982 11 2 2 1 4 2 9 1 4 9 6 4 1 0 3 5] - [ 1 20 0 6 4 942 6 21 2 2 2 10 8 18 2 1 4 4 2 8 7] - [ 0 7 11 0 3 3 1031 10 0 3 5 1 1 2 0 2 3 5 0 10 2] - [ 2 9 8 0 3 32 3 915 1 1 4 4 5 1 1 0 1 2 17 13 2] - [ 20 4 0 2 1 2 0 1 860 45 14 3 1 11 27 0 1 1 13 1 2] - [ 89 1 0 0 9 6 1 2 31 830 2 0 0 21 8 2 1 3 1 2 3] - [ 1 2 3 6 0 2 4 7 12 2 964 0 2 12 5 1 3 1 19 1 4] - [ 0 3 1 0 4 10 2 3 1 0 0 887 49 11 0 3 0 16 2 12 1] - [ 1 0 1 5 1 4 0 5 1 0 0 24 862 0 3 4 1 32 1 1 13] - [ 1 3 0 0 6 11 2 2 13 13 11 9 4 927 4 5 2 2 0 3 6] - [ 9 4 3 27 8 2 0 2 19 4 1 2 3 3 969 3 3 10 11 0 9] - [ 2 3 6 2 10 0 6 1 0 0 0 9 3 5 1 965 4 18 0 5 6] - [ 2 11 1 3 9 3 1 1 3 1 1 2 2 4 2 14 985 2 1 3 10] - [ 3 3 1 2 1 2 5 0 1 2 1 9 9 1 1 11 0 993 0 2 4] - [ 3 7 8 11 3 2 0 37 3 1 7 1 7 0 14 0 2 3 976 1 0] - [ 1 4 0 1 0 10 9 10 0 0 2 18 5 5 0 5 8 7 2 1056 5] - [ 142 255 223 170 196 279 134 202 110 85 228 141 340 400 142 156 241 142 178 300 9370]] - -2023-02-13 17:58:27,241 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:58:27,241 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:58:27,246 - - -2023-02-13 17:58:27,247 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:58:28,228 - Epoch: [91][ 10/ 1207] Overall Loss 0.321030 Objective Loss 0.321030 LR 0.001000 Time 0.098116 -2023-02-13 17:58:28,426 - Epoch: [91][ 20/ 1207] Overall Loss 0.334263 Objective Loss 0.334263 LR 0.001000 Time 0.058918 -2023-02-13 17:58:28,619 - Epoch: [91][ 30/ 1207] Overall Loss 0.328451 Objective Loss 0.328451 LR 0.001000 Time 0.045689 -2023-02-13 17:58:28,815 - Epoch: [91][ 40/ 1207] Overall Loss 0.328766 Objective Loss 0.328766 LR 0.001000 Time 0.039160 -2023-02-13 17:58:29,007 - Epoch: [91][ 50/ 1207] Overall Loss 0.322536 Objective Loss 0.322536 LR 0.001000 Time 0.035171 -2023-02-13 17:58:29,203 - Epoch: [91][ 60/ 1207] Overall Loss 0.326468 Objective Loss 0.326468 LR 0.001000 Time 0.032561 -2023-02-13 17:58:29,395 - Epoch: [91][ 70/ 1207] Overall Loss 0.321837 Objective Loss 0.321837 LR 0.001000 Time 0.030652 -2023-02-13 17:58:29,591 - Epoch: [91][ 80/ 1207] Overall Loss 0.320120 Objective Loss 0.320120 LR 0.001000 Time 0.029258 -2023-02-13 17:58:29,783 - Epoch: [91][ 90/ 1207] Overall Loss 0.320906 Objective Loss 0.320906 LR 0.001000 Time 0.028139 -2023-02-13 17:58:29,980 - Epoch: [91][ 100/ 1207] Overall Loss 0.323420 Objective Loss 0.323420 LR 0.001000 Time 0.027291 -2023-02-13 17:58:30,174 - Epoch: [91][ 110/ 1207] Overall Loss 0.322520 Objective Loss 0.322520 LR 0.001000 Time 0.026571 -2023-02-13 17:58:30,372 - Epoch: [91][ 120/ 1207] Overall Loss 0.325066 Objective Loss 0.325066 LR 0.001000 Time 0.026002 -2023-02-13 17:58:30,566 - Epoch: [91][ 130/ 1207] Overall Loss 0.329584 Objective Loss 0.329584 LR 0.001000 Time 0.025493 -2023-02-13 17:58:30,764 - Epoch: [91][ 140/ 1207] Overall Loss 0.328473 Objective Loss 0.328473 LR 0.001000 Time 0.025082 -2023-02-13 17:58:30,959 - Epoch: [91][ 150/ 1207] Overall Loss 0.327924 Objective Loss 0.327924 LR 0.001000 Time 0.024709 -2023-02-13 17:58:31,156 - Epoch: [91][ 160/ 1207] Overall Loss 0.326861 Objective Loss 0.326861 LR 0.001000 Time 0.024396 -2023-02-13 17:58:31,352 - Epoch: [91][ 170/ 1207] Overall Loss 0.326799 Objective Loss 0.326799 LR 0.001000 Time 0.024108 -2023-02-13 17:58:31,550 - Epoch: [91][ 180/ 1207] Overall Loss 0.328053 Objective Loss 0.328053 LR 0.001000 Time 0.023868 -2023-02-13 17:58:31,745 - Epoch: [91][ 190/ 1207] Overall Loss 0.326402 Objective Loss 0.326402 LR 0.001000 Time 0.023636 -2023-02-13 17:58:31,943 - Epoch: [91][ 200/ 1207] Overall Loss 0.327736 Objective Loss 0.327736 LR 0.001000 Time 0.023440 -2023-02-13 17:58:32,137 - Epoch: [91][ 210/ 1207] Overall Loss 0.327968 Objective Loss 0.327968 LR 0.001000 Time 0.023246 -2023-02-13 17:58:32,334 - Epoch: [91][ 220/ 1207] Overall Loss 0.328052 Objective Loss 0.328052 LR 0.001000 Time 0.023087 -2023-02-13 17:58:32,529 - Epoch: [91][ 230/ 1207] Overall Loss 0.328553 Objective Loss 0.328553 LR 0.001000 Time 0.022929 -2023-02-13 17:58:32,726 - Epoch: [91][ 240/ 1207] Overall Loss 0.328472 Objective Loss 0.328472 LR 0.001000 Time 0.022792 -2023-02-13 17:58:32,921 - Epoch: [91][ 250/ 1207] Overall Loss 0.329052 Objective Loss 0.329052 LR 0.001000 Time 0.022657 -2023-02-13 17:58:33,119 - Epoch: [91][ 260/ 1207] Overall Loss 0.330028 Objective Loss 0.330028 LR 0.001000 Time 0.022545 -2023-02-13 17:58:33,314 - Epoch: [91][ 270/ 1207] Overall Loss 0.329503 Objective Loss 0.329503 LR 0.001000 Time 0.022431 -2023-02-13 17:58:33,511 - Epoch: [91][ 280/ 1207] Overall Loss 0.330407 Objective Loss 0.330407 LR 0.001000 Time 0.022335 -2023-02-13 17:58:33,706 - Epoch: [91][ 290/ 1207] Overall Loss 0.331241 Objective Loss 0.331241 LR 0.001000 Time 0.022236 -2023-02-13 17:58:33,903 - Epoch: [91][ 300/ 1207] Overall Loss 0.331298 Objective Loss 0.331298 LR 0.001000 Time 0.022150 -2023-02-13 17:58:34,098 - Epoch: [91][ 310/ 1207] Overall Loss 0.330744 Objective Loss 0.330744 LR 0.001000 Time 0.022062 -2023-02-13 17:58:34,297 - Epoch: [91][ 320/ 1207] Overall Loss 0.330523 Objective Loss 0.330523 LR 0.001000 Time 0.021992 -2023-02-13 17:58:34,491 - Epoch: [91][ 330/ 1207] Overall Loss 0.330654 Objective Loss 0.330654 LR 0.001000 Time 0.021914 -2023-02-13 17:58:34,693 - Epoch: [91][ 340/ 1207] Overall Loss 0.331296 Objective Loss 0.331296 LR 0.001000 Time 0.021863 -2023-02-13 17:58:34,895 - Epoch: [91][ 350/ 1207] Overall Loss 0.331319 Objective Loss 0.331319 LR 0.001000 Time 0.021814 -2023-02-13 17:58:35,097 - Epoch: [91][ 360/ 1207] Overall Loss 0.331319 Objective Loss 0.331319 LR 0.001000 Time 0.021768 -2023-02-13 17:58:35,298 - Epoch: [91][ 370/ 1207] Overall Loss 0.330625 Objective Loss 0.330625 LR 0.001000 Time 0.021723 -2023-02-13 17:58:35,500 - Epoch: [91][ 380/ 1207] Overall Loss 0.330743 Objective Loss 0.330743 LR 0.001000 Time 0.021680 -2023-02-13 17:58:35,700 - Epoch: [91][ 390/ 1207] Overall Loss 0.331444 Objective Loss 0.331444 LR 0.001000 Time 0.021636 -2023-02-13 17:58:35,900 - Epoch: [91][ 400/ 1207] Overall Loss 0.332809 Objective Loss 0.332809 LR 0.001000 Time 0.021596 -2023-02-13 17:58:36,100 - Epoch: [91][ 410/ 1207] Overall Loss 0.333428 Objective Loss 0.333428 LR 0.001000 Time 0.021556 -2023-02-13 17:58:36,301 - Epoch: [91][ 420/ 1207] Overall Loss 0.333813 Objective Loss 0.333813 LR 0.001000 Time 0.021520 -2023-02-13 17:58:36,502 - Epoch: [91][ 430/ 1207] Overall Loss 0.333661 Objective Loss 0.333661 LR 0.001000 Time 0.021486 -2023-02-13 17:58:36,703 - Epoch: [91][ 440/ 1207] Overall Loss 0.333848 Objective Loss 0.333848 LR 0.001000 Time 0.021453 -2023-02-13 17:58:36,904 - Epoch: [91][ 450/ 1207] Overall Loss 0.334021 Objective Loss 0.334021 LR 0.001000 Time 0.021424 -2023-02-13 17:58:37,105 - Epoch: [91][ 460/ 1207] Overall Loss 0.333833 Objective Loss 0.333833 LR 0.001000 Time 0.021392 -2023-02-13 17:58:37,306 - Epoch: [91][ 470/ 1207] Overall Loss 0.333675 Objective Loss 0.333675 LR 0.001000 Time 0.021365 -2023-02-13 17:58:37,507 - Epoch: [91][ 480/ 1207] Overall Loss 0.333455 Objective Loss 0.333455 LR 0.001000 Time 0.021339 -2023-02-13 17:58:37,708 - Epoch: [91][ 490/ 1207] Overall Loss 0.333003 Objective Loss 0.333003 LR 0.001000 Time 0.021312 -2023-02-13 17:58:37,909 - Epoch: [91][ 500/ 1207] Overall Loss 0.333784 Objective Loss 0.333784 LR 0.001000 Time 0.021286 -2023-02-13 17:58:38,109 - Epoch: [91][ 510/ 1207] Overall Loss 0.333769 Objective Loss 0.333769 LR 0.001000 Time 0.021261 -2023-02-13 17:58:38,310 - Epoch: [91][ 520/ 1207] Overall Loss 0.333432 Objective Loss 0.333432 LR 0.001000 Time 0.021238 -2023-02-13 17:58:38,511 - Epoch: [91][ 530/ 1207] Overall Loss 0.333254 Objective Loss 0.333254 LR 0.001000 Time 0.021216 -2023-02-13 17:58:38,712 - Epoch: [91][ 540/ 1207] Overall Loss 0.333868 Objective Loss 0.333868 LR 0.001000 Time 0.021193 -2023-02-13 17:58:38,912 - Epoch: [91][ 550/ 1207] Overall Loss 0.333567 Objective Loss 0.333567 LR 0.001000 Time 0.021171 -2023-02-13 17:58:39,112 - Epoch: [91][ 560/ 1207] Overall Loss 0.333552 Objective Loss 0.333552 LR 0.001000 Time 0.021150 -2023-02-13 17:58:39,312 - Epoch: [91][ 570/ 1207] Overall Loss 0.332889 Objective Loss 0.332889 LR 0.001000 Time 0.021129 -2023-02-13 17:58:39,513 - Epoch: [91][ 580/ 1207] Overall Loss 0.333236 Objective Loss 0.333236 LR 0.001000 Time 0.021110 -2023-02-13 17:58:39,713 - Epoch: [91][ 590/ 1207] Overall Loss 0.332829 Objective Loss 0.332829 LR 0.001000 Time 0.021091 -2023-02-13 17:58:39,912 - Epoch: [91][ 600/ 1207] Overall Loss 0.332425 Objective Loss 0.332425 LR 0.001000 Time 0.021072 -2023-02-13 17:58:40,113 - Epoch: [91][ 610/ 1207] Overall Loss 0.332464 Objective Loss 0.332464 LR 0.001000 Time 0.021054 -2023-02-13 17:58:40,314 - Epoch: [91][ 620/ 1207] Overall Loss 0.332669 Objective Loss 0.332669 LR 0.001000 Time 0.021038 -2023-02-13 17:58:40,515 - Epoch: [91][ 630/ 1207] Overall Loss 0.332981 Objective Loss 0.332981 LR 0.001000 Time 0.021023 -2023-02-13 17:58:40,715 - Epoch: [91][ 640/ 1207] Overall Loss 0.332708 Objective Loss 0.332708 LR 0.001000 Time 0.021007 -2023-02-13 17:58:40,917 - Epoch: [91][ 650/ 1207] Overall Loss 0.332945 Objective Loss 0.332945 LR 0.001000 Time 0.020993 -2023-02-13 17:58:41,117 - Epoch: [91][ 660/ 1207] Overall Loss 0.332749 Objective Loss 0.332749 LR 0.001000 Time 0.020978 -2023-02-13 17:58:41,318 - Epoch: [91][ 670/ 1207] Overall Loss 0.332856 Objective Loss 0.332856 LR 0.001000 Time 0.020964 -2023-02-13 17:58:41,520 - Epoch: [91][ 680/ 1207] Overall Loss 0.332525 Objective Loss 0.332525 LR 0.001000 Time 0.020952 -2023-02-13 17:58:41,720 - Epoch: [91][ 690/ 1207] Overall Loss 0.332765 Objective Loss 0.332765 LR 0.001000 Time 0.020938 -2023-02-13 17:58:41,921 - Epoch: [91][ 700/ 1207] Overall Loss 0.332828 Objective Loss 0.332828 LR 0.001000 Time 0.020925 -2023-02-13 17:58:42,121 - Epoch: [91][ 710/ 1207] Overall Loss 0.333276 Objective Loss 0.333276 LR 0.001000 Time 0.020912 -2023-02-13 17:58:42,322 - Epoch: [91][ 720/ 1207] Overall Loss 0.333454 Objective Loss 0.333454 LR 0.001000 Time 0.020900 -2023-02-13 17:58:42,524 - Epoch: [91][ 730/ 1207] Overall Loss 0.333366 Objective Loss 0.333366 LR 0.001000 Time 0.020889 -2023-02-13 17:58:42,725 - Epoch: [91][ 740/ 1207] Overall Loss 0.333473 Objective Loss 0.333473 LR 0.001000 Time 0.020879 -2023-02-13 17:58:42,927 - Epoch: [91][ 750/ 1207] Overall Loss 0.333422 Objective Loss 0.333422 LR 0.001000 Time 0.020869 -2023-02-13 17:58:43,128 - Epoch: [91][ 760/ 1207] Overall Loss 0.333633 Objective Loss 0.333633 LR 0.001000 Time 0.020858 -2023-02-13 17:58:43,330 - Epoch: [91][ 770/ 1207] Overall Loss 0.333524 Objective Loss 0.333524 LR 0.001000 Time 0.020849 -2023-02-13 17:58:43,532 - Epoch: [91][ 780/ 1207] Overall Loss 0.333344 Objective Loss 0.333344 LR 0.001000 Time 0.020841 -2023-02-13 17:58:43,734 - Epoch: [91][ 790/ 1207] Overall Loss 0.333206 Objective Loss 0.333206 LR 0.001000 Time 0.020832 -2023-02-13 17:58:43,936 - Epoch: [91][ 800/ 1207] Overall Loss 0.332998 Objective Loss 0.332998 LR 0.001000 Time 0.020823 -2023-02-13 17:58:44,138 - Epoch: [91][ 810/ 1207] Overall Loss 0.332929 Objective Loss 0.332929 LR 0.001000 Time 0.020815 -2023-02-13 17:58:44,339 - Epoch: [91][ 820/ 1207] Overall Loss 0.332539 Objective Loss 0.332539 LR 0.001000 Time 0.020807 -2023-02-13 17:58:44,541 - Epoch: [91][ 830/ 1207] Overall Loss 0.332690 Objective Loss 0.332690 LR 0.001000 Time 0.020799 -2023-02-13 17:58:44,743 - Epoch: [91][ 840/ 1207] Overall Loss 0.333107 Objective Loss 0.333107 LR 0.001000 Time 0.020791 -2023-02-13 17:58:44,944 - Epoch: [91][ 850/ 1207] Overall Loss 0.333308 Objective Loss 0.333308 LR 0.001000 Time 0.020783 -2023-02-13 17:58:45,145 - Epoch: [91][ 860/ 1207] Overall Loss 0.333381 Objective Loss 0.333381 LR 0.001000 Time 0.020775 -2023-02-13 17:58:45,347 - Epoch: [91][ 870/ 1207] Overall Loss 0.333396 Objective Loss 0.333396 LR 0.001000 Time 0.020766 -2023-02-13 17:58:45,549 - Epoch: [91][ 880/ 1207] Overall Loss 0.333396 Objective Loss 0.333396 LR 0.001000 Time 0.020760 -2023-02-13 17:58:45,751 - Epoch: [91][ 890/ 1207] Overall Loss 0.333510 Objective Loss 0.333510 LR 0.001000 Time 0.020754 -2023-02-13 17:58:45,954 - Epoch: [91][ 900/ 1207] Overall Loss 0.333370 Objective Loss 0.333370 LR 0.001000 Time 0.020748 -2023-02-13 17:58:46,156 - Epoch: [91][ 910/ 1207] Overall Loss 0.333684 Objective Loss 0.333684 LR 0.001000 Time 0.020741 -2023-02-13 17:58:46,358 - Epoch: [91][ 920/ 1207] Overall Loss 0.334041 Objective Loss 0.334041 LR 0.001000 Time 0.020735 -2023-02-13 17:58:46,560 - Epoch: [91][ 930/ 1207] Overall Loss 0.333782 Objective Loss 0.333782 LR 0.001000 Time 0.020729 -2023-02-13 17:58:46,762 - Epoch: [91][ 940/ 1207] Overall Loss 0.334089 Objective Loss 0.334089 LR 0.001000 Time 0.020723 -2023-02-13 17:58:46,964 - Epoch: [91][ 950/ 1207] Overall Loss 0.334045 Objective Loss 0.334045 LR 0.001000 Time 0.020717 -2023-02-13 17:58:47,166 - Epoch: [91][ 960/ 1207] Overall Loss 0.334037 Objective Loss 0.334037 LR 0.001000 Time 0.020711 -2023-02-13 17:58:47,368 - Epoch: [91][ 970/ 1207] Overall Loss 0.334182 Objective Loss 0.334182 LR 0.001000 Time 0.020705 -2023-02-13 17:58:47,570 - Epoch: [91][ 980/ 1207] Overall Loss 0.334242 Objective Loss 0.334242 LR 0.001000 Time 0.020700 -2023-02-13 17:58:47,772 - Epoch: [91][ 990/ 1207] Overall Loss 0.334187 Objective Loss 0.334187 LR 0.001000 Time 0.020694 -2023-02-13 17:58:47,973 - Epoch: [91][ 1000/ 1207] Overall Loss 0.334255 Objective Loss 0.334255 LR 0.001000 Time 0.020688 -2023-02-13 17:58:48,174 - Epoch: [91][ 1010/ 1207] Overall Loss 0.334452 Objective Loss 0.334452 LR 0.001000 Time 0.020682 -2023-02-13 17:58:48,376 - Epoch: [91][ 1020/ 1207] Overall Loss 0.334544 Objective Loss 0.334544 LR 0.001000 Time 0.020677 -2023-02-13 17:58:48,578 - Epoch: [91][ 1030/ 1207] Overall Loss 0.334646 Objective Loss 0.334646 LR 0.001000 Time 0.020672 -2023-02-13 17:58:48,781 - Epoch: [91][ 1040/ 1207] Overall Loss 0.334834 Objective Loss 0.334834 LR 0.001000 Time 0.020668 -2023-02-13 17:58:48,982 - Epoch: [91][ 1050/ 1207] Overall Loss 0.335016 Objective Loss 0.335016 LR 0.001000 Time 0.020662 -2023-02-13 17:58:49,184 - Epoch: [91][ 1060/ 1207] Overall Loss 0.335358 Objective Loss 0.335358 LR 0.001000 Time 0.020657 -2023-02-13 17:58:49,386 - Epoch: [91][ 1070/ 1207] Overall Loss 0.335517 Objective Loss 0.335517 LR 0.001000 Time 0.020653 -2023-02-13 17:58:49,588 - Epoch: [91][ 1080/ 1207] Overall Loss 0.335168 Objective Loss 0.335168 LR 0.001000 Time 0.020648 -2023-02-13 17:58:49,790 - Epoch: [91][ 1090/ 1207] Overall Loss 0.335204 Objective Loss 0.335204 LR 0.001000 Time 0.020644 -2023-02-13 17:58:49,991 - Epoch: [91][ 1100/ 1207] Overall Loss 0.335077 Objective Loss 0.335077 LR 0.001000 Time 0.020639 -2023-02-13 17:58:50,194 - Epoch: [91][ 1110/ 1207] Overall Loss 0.334976 Objective Loss 0.334976 LR 0.001000 Time 0.020635 -2023-02-13 17:58:50,395 - Epoch: [91][ 1120/ 1207] Overall Loss 0.335139 Objective Loss 0.335139 LR 0.001000 Time 0.020630 -2023-02-13 17:58:50,598 - Epoch: [91][ 1130/ 1207] Overall Loss 0.335312 Objective Loss 0.335312 LR 0.001000 Time 0.020626 -2023-02-13 17:58:50,791 - Epoch: [91][ 1140/ 1207] Overall Loss 0.335022 Objective Loss 0.335022 LR 0.001000 Time 0.020615 -2023-02-13 17:58:50,988 - Epoch: [91][ 1150/ 1207] Overall Loss 0.335092 Objective Loss 0.335092 LR 0.001000 Time 0.020607 -2023-02-13 17:58:51,181 - Epoch: [91][ 1160/ 1207] Overall Loss 0.335055 Objective Loss 0.335055 LR 0.001000 Time 0.020595 -2023-02-13 17:58:51,377 - Epoch: [91][ 1170/ 1207] Overall Loss 0.335221 Objective Loss 0.335221 LR 0.001000 Time 0.020586 -2023-02-13 17:58:51,571 - Epoch: [91][ 1180/ 1207] Overall Loss 0.335504 Objective Loss 0.335504 LR 0.001000 Time 0.020575 -2023-02-13 17:58:51,766 - Epoch: [91][ 1190/ 1207] Overall Loss 0.335312 Objective Loss 0.335312 LR 0.001000 Time 0.020566 -2023-02-13 17:58:52,011 - Epoch: [91][ 1200/ 1207] Overall Loss 0.335307 Objective Loss 0.335307 LR 0.001000 Time 0.020599 -2023-02-13 17:58:52,126 - Epoch: [91][ 1207/ 1207] Overall Loss 0.335409 Objective Loss 0.335409 Top1 78.963415 Top5 96.951220 LR 0.001000 Time 0.020575 -2023-02-13 17:58:52,197 - --- validate (epoch=91)----------- -2023-02-13 17:58:52,197 - 34311 samples (256 per mini-batch) -2023-02-13 17:58:52,601 - Epoch: [91][ 10/ 135] Loss 0.338702 Top1 81.953125 Top5 97.187500 -2023-02-13 17:58:52,732 - Epoch: [91][ 20/ 135] Loss 0.357965 Top1 81.992188 Top5 96.816406 -2023-02-13 17:58:52,861 - Epoch: [91][ 30/ 135] Loss 0.354666 Top1 81.718750 Top5 96.809896 -2023-02-13 17:58:52,991 - Epoch: [91][ 40/ 135] Loss 0.351521 Top1 81.708984 Top5 96.835938 -2023-02-13 17:58:53,123 - Epoch: [91][ 50/ 135] Loss 0.359800 Top1 81.585938 Top5 96.804688 -2023-02-13 17:58:53,252 - Epoch: [91][ 60/ 135] Loss 0.354989 Top1 81.790365 Top5 96.907552 -2023-02-13 17:58:53,382 - Epoch: [91][ 70/ 135] Loss 0.355896 Top1 81.808036 Top5 96.925223 -2023-02-13 17:58:53,513 - Epoch: [91][ 80/ 135] Loss 0.359492 Top1 81.625977 Top5 96.909180 -2023-02-13 17:58:53,643 - Epoch: [91][ 90/ 135] Loss 0.361710 Top1 81.762153 Top5 96.979167 -2023-02-13 17:58:53,770 - Epoch: [91][ 100/ 135] Loss 0.362704 Top1 81.765625 Top5 97.027344 -2023-02-13 17:58:53,903 - Epoch: [91][ 110/ 135] Loss 0.362792 Top1 81.672585 Top5 97.009943 -2023-02-13 17:58:54,032 - Epoch: [91][ 120/ 135] Loss 0.364389 Top1 81.647135 Top5 96.975911 -2023-02-13 17:58:54,166 - Epoch: [91][ 130/ 135] Loss 0.364570 Top1 81.595553 Top5 96.986178 -2023-02-13 17:58:54,211 - Epoch: [91][ 135/ 135] Loss 0.362967 Top1 81.586080 Top5 97.018449 -2023-02-13 17:58:54,280 - ==> Top1: 81.586 Top5: 97.018 Loss: 0.363 - -2023-02-13 17:58:54,281 - ==> Confusion: -[[ 824 6 11 1 12 4 0 3 4 67 2 3 1 3 8 1 1 2 0 5 9] - [ 1 923 1 4 10 35 3 29 2 1 2 0 2 1 2 2 3 2 5 1 4] - [ 4 5 932 17 4 2 31 15 1 1 3 3 2 8 6 7 5 3 5 1 3] - [ 4 2 19 890 4 4 1 2 1 1 19 0 6 2 24 2 2 6 19 1 7] - [ 13 14 2 0 966 15 0 2 3 7 0 8 1 3 4 4 6 3 1 5 9] - [ 2 21 0 5 4 960 3 21 1 4 4 11 4 11 0 2 2 2 3 7 3] - [ 2 5 12 1 0 6 1042 7 1 2 2 1 1 1 0 5 0 3 1 5 2] - [ 1 6 8 0 2 30 7 915 2 4 2 9 1 2 2 0 0 3 18 12 0] - [ 16 7 0 3 1 0 0 3 857 51 16 4 0 17 21 3 0 1 7 0 2] - [ 73 2 2 1 5 3 0 1 32 855 1 2 0 19 3 2 0 6 0 1 4] - [ 1 5 5 10 2 6 6 6 13 2 960 1 2 7 5 0 0 1 14 1 4] - [ 2 4 0 0 4 16 1 4 1 3 1 892 38 9 0 3 3 9 3 9 3] - [ 1 2 1 4 2 4 1 2 2 0 2 27 865 0 4 5 3 18 1 1 14] - [ 1 2 3 0 6 11 1 1 11 23 9 7 3 917 7 7 3 4 1 2 5] - [ 5 4 5 17 2 3 0 3 19 10 4 1 4 5 984 1 1 7 13 1 3] - [ 5 4 3 0 8 1 8 1 0 0 0 6 7 2 1 962 9 15 0 5 9] - [ 2 16 3 1 13 4 0 1 2 0 2 6 5 2 3 10 966 1 2 7 15] - [ 5 3 4 4 1 1 1 2 0 0 1 10 28 3 1 14 1 967 1 1 3] - [ 4 4 6 9 2 1 1 25 5 0 6 0 5 1 18 2 1 3 989 2 2] - [ 0 4 1 2 3 5 9 22 1 0 0 19 1 2 0 6 6 6 2 1052 7] - [ 139 343 278 195 155 253 123 232 106 128 262 119 328 344 202 119 214 141 243 235 9275]] - -2023-02-13 17:58:54,282 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:58:54,282 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:58:54,288 - - -2023-02-13 17:58:54,288 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:58:55,189 - Epoch: [92][ 10/ 1207] Overall Loss 0.314484 Objective Loss 0.314484 LR 0.001000 Time 0.090061 -2023-02-13 17:58:55,401 - Epoch: [92][ 20/ 1207] Overall Loss 0.330149 Objective Loss 0.330149 LR 0.001000 Time 0.055594 -2023-02-13 17:58:55,600 - Epoch: [92][ 30/ 1207] Overall Loss 0.331166 Objective Loss 0.331166 LR 0.001000 Time 0.043680 -2023-02-13 17:58:55,803 - Epoch: [92][ 40/ 1207] Overall Loss 0.339482 Objective Loss 0.339482 LR 0.001000 Time 0.037829 -2023-02-13 17:58:56,001 - Epoch: [92][ 50/ 1207] Overall Loss 0.335941 Objective Loss 0.335941 LR 0.001000 Time 0.034221 -2023-02-13 17:58:56,204 - Epoch: [92][ 60/ 1207] Overall Loss 0.340778 Objective Loss 0.340778 LR 0.001000 Time 0.031888 -2023-02-13 17:58:56,402 - Epoch: [92][ 70/ 1207] Overall Loss 0.335905 Objective Loss 0.335905 LR 0.001000 Time 0.030149 -2023-02-13 17:58:56,604 - Epoch: [92][ 80/ 1207] Overall Loss 0.336710 Objective Loss 0.336710 LR 0.001000 Time 0.028909 -2023-02-13 17:58:56,802 - Epoch: [92][ 90/ 1207] Overall Loss 0.334526 Objective Loss 0.334526 LR 0.001000 Time 0.027891 -2023-02-13 17:58:57,004 - Epoch: [92][ 100/ 1207] Overall Loss 0.332386 Objective Loss 0.332386 LR 0.001000 Time 0.027116 -2023-02-13 17:58:57,202 - Epoch: [92][ 110/ 1207] Overall Loss 0.336196 Objective Loss 0.336196 LR 0.001000 Time 0.026446 -2023-02-13 17:58:57,404 - Epoch: [92][ 120/ 1207] Overall Loss 0.335741 Objective Loss 0.335741 LR 0.001000 Time 0.025926 -2023-02-13 17:58:57,602 - Epoch: [92][ 130/ 1207] Overall Loss 0.334315 Objective Loss 0.334315 LR 0.001000 Time 0.025453 -2023-02-13 17:58:57,805 - Epoch: [92][ 140/ 1207] Overall Loss 0.331339 Objective Loss 0.331339 LR 0.001000 Time 0.025080 -2023-02-13 17:58:58,002 - Epoch: [92][ 150/ 1207] Overall Loss 0.331154 Objective Loss 0.331154 LR 0.001000 Time 0.024721 -2023-02-13 17:58:58,206 - Epoch: [92][ 160/ 1207] Overall Loss 0.330806 Objective Loss 0.330806 LR 0.001000 Time 0.024446 -2023-02-13 17:58:58,404 - Epoch: [92][ 170/ 1207] Overall Loss 0.330934 Objective Loss 0.330934 LR 0.001000 Time 0.024173 -2023-02-13 17:58:58,609 - Epoch: [92][ 180/ 1207] Overall Loss 0.330288 Objective Loss 0.330288 LR 0.001000 Time 0.023963 -2023-02-13 17:58:58,807 - Epoch: [92][ 190/ 1207] Overall Loss 0.330549 Objective Loss 0.330549 LR 0.001000 Time 0.023743 -2023-02-13 17:58:59,010 - Epoch: [92][ 200/ 1207] Overall Loss 0.331065 Objective Loss 0.331065 LR 0.001000 Time 0.023567 -2023-02-13 17:58:59,208 - Epoch: [92][ 210/ 1207] Overall Loss 0.332045 Objective Loss 0.332045 LR 0.001000 Time 0.023385 -2023-02-13 17:58:59,411 - Epoch: [92][ 220/ 1207] Overall Loss 0.330212 Objective Loss 0.330212 LR 0.001000 Time 0.023243 -2023-02-13 17:58:59,609 - Epoch: [92][ 230/ 1207] Overall Loss 0.328960 Objective Loss 0.328960 LR 0.001000 Time 0.023094 -2023-02-13 17:58:59,812 - Epoch: [92][ 240/ 1207] Overall Loss 0.330580 Objective Loss 0.330580 LR 0.001000 Time 0.022976 -2023-02-13 17:59:00,011 - Epoch: [92][ 250/ 1207] Overall Loss 0.331910 Objective Loss 0.331910 LR 0.001000 Time 0.022850 -2023-02-13 17:59:00,214 - Epoch: [92][ 260/ 1207] Overall Loss 0.331951 Objective Loss 0.331951 LR 0.001000 Time 0.022752 -2023-02-13 17:59:00,403 - Epoch: [92][ 270/ 1207] Overall Loss 0.331764 Objective Loss 0.331764 LR 0.001000 Time 0.022606 -2023-02-13 17:59:00,592 - Epoch: [92][ 280/ 1207] Overall Loss 0.332651 Objective Loss 0.332651 LR 0.001000 Time 0.022473 -2023-02-13 17:59:00,790 - Epoch: [92][ 290/ 1207] Overall Loss 0.334815 Objective Loss 0.334815 LR 0.001000 Time 0.022379 -2023-02-13 17:59:00,988 - Epoch: [92][ 300/ 1207] Overall Loss 0.335460 Objective Loss 0.335460 LR 0.001000 Time 0.022294 -2023-02-13 17:59:01,191 - Epoch: [92][ 310/ 1207] Overall Loss 0.335572 Objective Loss 0.335572 LR 0.001000 Time 0.022227 -2023-02-13 17:59:01,389 - Epoch: [92][ 320/ 1207] Overall Loss 0.335911 Objective Loss 0.335911 LR 0.001000 Time 0.022151 -2023-02-13 17:59:01,592 - Epoch: [92][ 330/ 1207] Overall Loss 0.335864 Objective Loss 0.335864 LR 0.001000 Time 0.022093 -2023-02-13 17:59:01,790 - Epoch: [92][ 340/ 1207] Overall Loss 0.336262 Objective Loss 0.336262 LR 0.001000 Time 0.022026 -2023-02-13 17:59:01,994 - Epoch: [92][ 350/ 1207] Overall Loss 0.336335 Objective Loss 0.336335 LR 0.001000 Time 0.021976 -2023-02-13 17:59:02,191 - Epoch: [92][ 360/ 1207] Overall Loss 0.335825 Objective Loss 0.335825 LR 0.001000 Time 0.021914 -2023-02-13 17:59:02,394 - Epoch: [92][ 370/ 1207] Overall Loss 0.335116 Objective Loss 0.335116 LR 0.001000 Time 0.021869 -2023-02-13 17:59:02,591 - Epoch: [92][ 380/ 1207] Overall Loss 0.334267 Objective Loss 0.334267 LR 0.001000 Time 0.021811 -2023-02-13 17:59:02,793 - Epoch: [92][ 390/ 1207] Overall Loss 0.333757 Objective Loss 0.333757 LR 0.001000 Time 0.021768 -2023-02-13 17:59:02,991 - Epoch: [92][ 400/ 1207] Overall Loss 0.333217 Objective Loss 0.333217 LR 0.001000 Time 0.021718 -2023-02-13 17:59:03,193 - Epoch: [92][ 410/ 1207] Overall Loss 0.333263 Objective Loss 0.333263 LR 0.001000 Time 0.021681 -2023-02-13 17:59:03,391 - Epoch: [92][ 420/ 1207] Overall Loss 0.333626 Objective Loss 0.333626 LR 0.001000 Time 0.021635 -2023-02-13 17:59:03,595 - Epoch: [92][ 430/ 1207] Overall Loss 0.333906 Objective Loss 0.333906 LR 0.001000 Time 0.021605 -2023-02-13 17:59:03,793 - Epoch: [92][ 440/ 1207] Overall Loss 0.333672 Objective Loss 0.333672 LR 0.001000 Time 0.021563 -2023-02-13 17:59:03,996 - Epoch: [92][ 450/ 1207] Overall Loss 0.332923 Objective Loss 0.332923 LR 0.001000 Time 0.021534 -2023-02-13 17:59:04,195 - Epoch: [92][ 460/ 1207] Overall Loss 0.332600 Objective Loss 0.332600 LR 0.001000 Time 0.021497 -2023-02-13 17:59:04,397 - Epoch: [92][ 470/ 1207] Overall Loss 0.333340 Objective Loss 0.333340 LR 0.001000 Time 0.021469 -2023-02-13 17:59:04,596 - Epoch: [92][ 480/ 1207] Overall Loss 0.333260 Objective Loss 0.333260 LR 0.001000 Time 0.021437 -2023-02-13 17:59:04,799 - Epoch: [92][ 490/ 1207] Overall Loss 0.333751 Objective Loss 0.333751 LR 0.001000 Time 0.021413 -2023-02-13 17:59:04,997 - Epoch: [92][ 500/ 1207] Overall Loss 0.333929 Objective Loss 0.333929 LR 0.001000 Time 0.021380 -2023-02-13 17:59:05,201 - Epoch: [92][ 510/ 1207] Overall Loss 0.333219 Objective Loss 0.333219 LR 0.001000 Time 0.021359 -2023-02-13 17:59:05,399 - Epoch: [92][ 520/ 1207] Overall Loss 0.332828 Objective Loss 0.332828 LR 0.001000 Time 0.021328 -2023-02-13 17:59:05,597 - Epoch: [92][ 530/ 1207] Overall Loss 0.332636 Objective Loss 0.332636 LR 0.001000 Time 0.021298 -2023-02-13 17:59:05,785 - Epoch: [92][ 540/ 1207] Overall Loss 0.332282 Objective Loss 0.332282 LR 0.001000 Time 0.021251 -2023-02-13 17:59:05,973 - Epoch: [92][ 550/ 1207] Overall Loss 0.331861 Objective Loss 0.331861 LR 0.001000 Time 0.021206 -2023-02-13 17:59:06,160 - Epoch: [92][ 560/ 1207] Overall Loss 0.332251 Objective Loss 0.332251 LR 0.001000 Time 0.021162 -2023-02-13 17:59:06,348 - Epoch: [92][ 570/ 1207] Overall Loss 0.332252 Objective Loss 0.332252 LR 0.001000 Time 0.021119 -2023-02-13 17:59:06,538 - Epoch: [92][ 580/ 1207] Overall Loss 0.332383 Objective Loss 0.332383 LR 0.001000 Time 0.021082 -2023-02-13 17:59:06,727 - Epoch: [92][ 590/ 1207] Overall Loss 0.332394 Objective Loss 0.332394 LR 0.001000 Time 0.021045 -2023-02-13 17:59:06,919 - Epoch: [92][ 600/ 1207] Overall Loss 0.332087 Objective Loss 0.332087 LR 0.001000 Time 0.021013 -2023-02-13 17:59:07,109 - Epoch: [92][ 610/ 1207] Overall Loss 0.332162 Objective Loss 0.332162 LR 0.001000 Time 0.020979 -2023-02-13 17:59:07,300 - Epoch: [92][ 620/ 1207] Overall Loss 0.331736 Objective Loss 0.331736 LR 0.001000 Time 0.020948 -2023-02-13 17:59:07,488 - Epoch: [92][ 630/ 1207] Overall Loss 0.331912 Objective Loss 0.331912 LR 0.001000 Time 0.020913 -2023-02-13 17:59:07,677 - Epoch: [92][ 640/ 1207] Overall Loss 0.331745 Objective Loss 0.331745 LR 0.001000 Time 0.020881 -2023-02-13 17:59:07,865 - Epoch: [92][ 650/ 1207] Overall Loss 0.332204 Objective Loss 0.332204 LR 0.001000 Time 0.020848 -2023-02-13 17:59:08,053 - Epoch: [92][ 660/ 1207] Overall Loss 0.332719 Objective Loss 0.332719 LR 0.001000 Time 0.020817 -2023-02-13 17:59:08,241 - Epoch: [92][ 670/ 1207] Overall Loss 0.332767 Objective Loss 0.332767 LR 0.001000 Time 0.020786 -2023-02-13 17:59:08,429 - Epoch: [92][ 680/ 1207] Overall Loss 0.332853 Objective Loss 0.332853 LR 0.001000 Time 0.020756 -2023-02-13 17:59:08,618 - Epoch: [92][ 690/ 1207] Overall Loss 0.333371 Objective Loss 0.333371 LR 0.001000 Time 0.020729 -2023-02-13 17:59:08,806 - Epoch: [92][ 700/ 1207] Overall Loss 0.333588 Objective Loss 0.333588 LR 0.001000 Time 0.020702 -2023-02-13 17:59:08,994 - Epoch: [92][ 710/ 1207] Overall Loss 0.333785 Objective Loss 0.333785 LR 0.001000 Time 0.020674 -2023-02-13 17:59:09,182 - Epoch: [92][ 720/ 1207] Overall Loss 0.333928 Objective Loss 0.333928 LR 0.001000 Time 0.020647 -2023-02-13 17:59:09,370 - Epoch: [92][ 730/ 1207] Overall Loss 0.333488 Objective Loss 0.333488 LR 0.001000 Time 0.020621 -2023-02-13 17:59:09,559 - Epoch: [92][ 740/ 1207] Overall Loss 0.333911 Objective Loss 0.333911 LR 0.001000 Time 0.020597 -2023-02-13 17:59:09,746 - Epoch: [92][ 750/ 1207] Overall Loss 0.334025 Objective Loss 0.334025 LR 0.001000 Time 0.020573 -2023-02-13 17:59:09,934 - Epoch: [92][ 760/ 1207] Overall Loss 0.334055 Objective Loss 0.334055 LR 0.001000 Time 0.020549 -2023-02-13 17:59:10,122 - Epoch: [92][ 770/ 1207] Overall Loss 0.334053 Objective Loss 0.334053 LR 0.001000 Time 0.020525 -2023-02-13 17:59:10,310 - Epoch: [92][ 780/ 1207] Overall Loss 0.334144 Objective Loss 0.334144 LR 0.001000 Time 0.020503 -2023-02-13 17:59:10,498 - Epoch: [92][ 790/ 1207] Overall Loss 0.334083 Objective Loss 0.334083 LR 0.001000 Time 0.020481 -2023-02-13 17:59:10,687 - Epoch: [92][ 800/ 1207] Overall Loss 0.333986 Objective Loss 0.333986 LR 0.001000 Time 0.020460 -2023-02-13 17:59:10,875 - Epoch: [92][ 810/ 1207] Overall Loss 0.333977 Objective Loss 0.333977 LR 0.001000 Time 0.020440 -2023-02-13 17:59:11,063 - Epoch: [92][ 820/ 1207] Overall Loss 0.333960 Objective Loss 0.333960 LR 0.001000 Time 0.020419 -2023-02-13 17:59:11,254 - Epoch: [92][ 830/ 1207] Overall Loss 0.334434 Objective Loss 0.334434 LR 0.001000 Time 0.020403 -2023-02-13 17:59:11,448 - Epoch: [92][ 840/ 1207] Overall Loss 0.334544 Objective Loss 0.334544 LR 0.001000 Time 0.020390 -2023-02-13 17:59:11,644 - Epoch: [92][ 850/ 1207] Overall Loss 0.334907 Objective Loss 0.334907 LR 0.001000 Time 0.020380 -2023-02-13 17:59:11,837 - Epoch: [92][ 860/ 1207] Overall Loss 0.334651 Objective Loss 0.334651 LR 0.001000 Time 0.020367 -2023-02-13 17:59:12,033 - Epoch: [92][ 870/ 1207] Overall Loss 0.334765 Objective Loss 0.334765 LR 0.001000 Time 0.020358 -2023-02-13 17:59:12,226 - Epoch: [92][ 880/ 1207] Overall Loss 0.334717 Objective Loss 0.334717 LR 0.001000 Time 0.020345 -2023-02-13 17:59:12,422 - Epoch: [92][ 890/ 1207] Overall Loss 0.334743 Objective Loss 0.334743 LR 0.001000 Time 0.020337 -2023-02-13 17:59:12,616 - Epoch: [92][ 900/ 1207] Overall Loss 0.334782 Objective Loss 0.334782 LR 0.001000 Time 0.020326 -2023-02-13 17:59:12,811 - Epoch: [92][ 910/ 1207] Overall Loss 0.334963 Objective Loss 0.334963 LR 0.001000 Time 0.020317 -2023-02-13 17:59:13,005 - Epoch: [92][ 920/ 1207] Overall Loss 0.334712 Objective Loss 0.334712 LR 0.001000 Time 0.020306 -2023-02-13 17:59:13,200 - Epoch: [92][ 930/ 1207] Overall Loss 0.334774 Objective Loss 0.334774 LR 0.001000 Time 0.020297 -2023-02-13 17:59:13,393 - Epoch: [92][ 940/ 1207] Overall Loss 0.334901 Objective Loss 0.334901 LR 0.001000 Time 0.020286 -2023-02-13 17:59:13,590 - Epoch: [92][ 950/ 1207] Overall Loss 0.335016 Objective Loss 0.335016 LR 0.001000 Time 0.020279 -2023-02-13 17:59:13,786 - Epoch: [92][ 960/ 1207] Overall Loss 0.334972 Objective Loss 0.334972 LR 0.001000 Time 0.020272 -2023-02-13 17:59:13,983 - Epoch: [92][ 970/ 1207] Overall Loss 0.335059 Objective Loss 0.335059 LR 0.001000 Time 0.020266 -2023-02-13 17:59:14,179 - Epoch: [92][ 980/ 1207] Overall Loss 0.335151 Objective Loss 0.335151 LR 0.001000 Time 0.020258 -2023-02-13 17:59:14,377 - Epoch: [92][ 990/ 1207] Overall Loss 0.335020 Objective Loss 0.335020 LR 0.001000 Time 0.020253 -2023-02-13 17:59:14,572 - Epoch: [92][ 1000/ 1207] Overall Loss 0.335125 Objective Loss 0.335125 LR 0.001000 Time 0.020246 -2023-02-13 17:59:14,770 - Epoch: [92][ 1010/ 1207] Overall Loss 0.335113 Objective Loss 0.335113 LR 0.001000 Time 0.020240 -2023-02-13 17:59:14,965 - Epoch: [92][ 1020/ 1207] Overall Loss 0.335066 Objective Loss 0.335066 LR 0.001000 Time 0.020233 -2023-02-13 17:59:15,162 - Epoch: [92][ 1030/ 1207] Overall Loss 0.334722 Objective Loss 0.334722 LR 0.001000 Time 0.020228 -2023-02-13 17:59:15,358 - Epoch: [92][ 1040/ 1207] Overall Loss 0.334860 Objective Loss 0.334860 LR 0.001000 Time 0.020221 -2023-02-13 17:59:15,556 - Epoch: [92][ 1050/ 1207] Overall Loss 0.335009 Objective Loss 0.335009 LR 0.001000 Time 0.020217 -2023-02-13 17:59:15,752 - Epoch: [92][ 1060/ 1207] Overall Loss 0.335133 Objective Loss 0.335133 LR 0.001000 Time 0.020211 -2023-02-13 17:59:15,950 - Epoch: [92][ 1070/ 1207] Overall Loss 0.334932 Objective Loss 0.334932 LR 0.001000 Time 0.020207 -2023-02-13 17:59:16,146 - Epoch: [92][ 1080/ 1207] Overall Loss 0.335241 Objective Loss 0.335241 LR 0.001000 Time 0.020200 -2023-02-13 17:59:16,344 - Epoch: [92][ 1090/ 1207] Overall Loss 0.335318 Objective Loss 0.335318 LR 0.001000 Time 0.020196 -2023-02-13 17:59:16,540 - Epoch: [92][ 1100/ 1207] Overall Loss 0.335510 Objective Loss 0.335510 LR 0.001000 Time 0.020190 -2023-02-13 17:59:16,737 - Epoch: [92][ 1110/ 1207] Overall Loss 0.335175 Objective Loss 0.335175 LR 0.001000 Time 0.020186 -2023-02-13 17:59:16,933 - Epoch: [92][ 1120/ 1207] Overall Loss 0.335049 Objective Loss 0.335049 LR 0.001000 Time 0.020181 -2023-02-13 17:59:17,131 - Epoch: [92][ 1130/ 1207] Overall Loss 0.335088 Objective Loss 0.335088 LR 0.001000 Time 0.020177 -2023-02-13 17:59:17,325 - Epoch: [92][ 1140/ 1207] Overall Loss 0.334878 Objective Loss 0.334878 LR 0.001000 Time 0.020170 -2023-02-13 17:59:17,516 - Epoch: [92][ 1150/ 1207] Overall Loss 0.334832 Objective Loss 0.334832 LR 0.001000 Time 0.020160 -2023-02-13 17:59:17,708 - Epoch: [92][ 1160/ 1207] Overall Loss 0.334701 Objective Loss 0.334701 LR 0.001000 Time 0.020151 -2023-02-13 17:59:17,898 - Epoch: [92][ 1170/ 1207] Overall Loss 0.334791 Objective Loss 0.334791 LR 0.001000 Time 0.020142 -2023-02-13 17:59:18,090 - Epoch: [92][ 1180/ 1207] Overall Loss 0.334612 Objective Loss 0.334612 LR 0.001000 Time 0.020133 -2023-02-13 17:59:18,281 - Epoch: [92][ 1190/ 1207] Overall Loss 0.334497 Objective Loss 0.334497 LR 0.001000 Time 0.020124 -2023-02-13 17:59:18,529 - Epoch: [92][ 1200/ 1207] Overall Loss 0.334369 Objective Loss 0.334369 LR 0.001000 Time 0.020163 -2023-02-13 17:59:18,645 - Epoch: [92][ 1207/ 1207] Overall Loss 0.334378 Objective Loss 0.334378 Top1 82.621951 Top5 96.951220 LR 0.001000 Time 0.020141 -2023-02-13 17:59:18,715 - --- validate (epoch=92)----------- -2023-02-13 17:59:18,716 - 34311 samples (256 per mini-batch) -2023-02-13 17:59:19,106 - Epoch: [92][ 10/ 135] Loss 0.398829 Top1 80.585938 Top5 97.265625 -2023-02-13 17:59:19,234 - Epoch: [92][ 20/ 135] Loss 0.388668 Top1 80.742188 Top5 97.089844 -2023-02-13 17:59:19,358 - Epoch: [92][ 30/ 135] Loss 0.395740 Top1 80.729167 Top5 96.927083 -2023-02-13 17:59:19,489 - Epoch: [92][ 40/ 135] Loss 0.385486 Top1 80.937500 Top5 97.041016 -2023-02-13 17:59:19,641 - Epoch: [92][ 50/ 135] Loss 0.388659 Top1 80.968750 Top5 97.093750 -2023-02-13 17:59:19,766 - Epoch: [92][ 60/ 135] Loss 0.379790 Top1 81.243490 Top5 97.141927 -2023-02-13 17:59:19,889 - Epoch: [92][ 70/ 135] Loss 0.375627 Top1 81.328125 Top5 97.209821 -2023-02-13 17:59:20,012 - Epoch: [92][ 80/ 135] Loss 0.376116 Top1 81.347656 Top5 97.153320 -2023-02-13 17:59:20,136 - Epoch: [92][ 90/ 135] Loss 0.374318 Top1 81.419271 Top5 97.174479 -2023-02-13 17:59:20,258 - Epoch: [92][ 100/ 135] Loss 0.372305 Top1 81.445312 Top5 97.199219 -2023-02-13 17:59:20,381 - Epoch: [92][ 110/ 135] Loss 0.369928 Top1 81.470170 Top5 97.208807 -2023-02-13 17:59:20,504 - Epoch: [92][ 120/ 135] Loss 0.369801 Top1 81.367188 Top5 97.226562 -2023-02-13 17:59:20,630 - Epoch: [92][ 130/ 135] Loss 0.371254 Top1 81.340144 Top5 97.181490 -2023-02-13 17:59:20,674 - Epoch: [92][ 135/ 135] Loss 0.377814 Top1 81.323774 Top5 97.178747 -2023-02-13 17:59:20,755 - ==> Top1: 81.324 Top5: 97.179 Loss: 0.378 - -2023-02-13 17:59:20,756 - ==> Confusion: -[[ 853 5 10 2 12 1 0 0 5 47 1 6 1 3 5 4 1 4 0 1 6] - [ 0 935 4 2 16 21 3 20 7 4 1 3 1 0 0 2 4 0 5 1 4] - [ 6 4 947 13 7 1 17 16 0 0 2 2 2 5 2 4 4 4 10 4 8] - [ 2 3 31 851 5 5 2 2 3 2 19 0 11 5 22 4 5 6 26 1 11] - [ 18 8 2 0 982 5 1 2 1 4 3 7 2 4 3 7 9 4 0 1 3] - [ 5 35 3 5 8 936 6 18 0 3 4 14 2 14 0 5 4 1 0 5 2] - [ 1 4 23 0 1 5 1033 4 2 0 3 2 0 0 1 4 0 4 2 4 6] - [ 4 16 10 0 4 31 8 894 3 1 6 2 2 1 0 1 2 2 25 10 2] - [ 23 4 0 1 1 0 0 2 860 59 10 2 1 10 25 3 0 0 5 0 3] - [ 110 2 3 1 9 3 0 0 31 821 0 2 0 17 4 1 2 1 1 1 3] - [ 1 4 10 8 4 1 5 5 20 3 959 1 1 10 2 2 2 0 12 0 1] - [ 1 3 0 0 5 13 1 6 2 1 0 897 29 7 2 7 7 7 1 13 3] - [ 1 1 0 3 1 6 1 0 5 0 1 39 857 0 2 14 1 14 3 0 10] - [ 6 2 5 0 8 15 1 1 13 22 4 6 1 910 6 6 8 1 1 2 6] - [ 18 3 3 12 5 3 1 2 25 5 2 0 1 2 974 4 3 7 12 0 10] - [ 2 1 9 1 8 0 6 1 0 0 0 10 6 4 0 971 8 9 0 2 8] - [ 2 9 1 2 13 2 0 0 2 2 0 2 2 3 0 15 988 1 2 6 9] - [ 6 1 3 1 1 2 1 1 2 0 0 9 28 1 2 27 1 956 0 5 4] - [ 5 9 9 6 4 1 1 26 2 1 5 1 3 1 13 1 0 3 989 2 4] - [ 0 5 4 1 1 7 13 17 1 0 0 18 3 1 0 9 10 1 2 1047 8] - [ 217 291 296 80 236 205 103 187 113 114 207 144 340 347 185 153 354 103 225 291 9243]] - -2023-02-13 17:59:20,757 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:59:20,757 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:59:20,763 - - -2023-02-13 17:59:20,763 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:59:21,658 - Epoch: [93][ 10/ 1207] Overall Loss 0.346768 Objective Loss 0.346768 LR 0.001000 Time 0.089385 -2023-02-13 17:59:21,859 - Epoch: [93][ 20/ 1207] Overall Loss 0.330805 Objective Loss 0.330805 LR 0.001000 Time 0.054744 -2023-02-13 17:59:22,055 - Epoch: [93][ 30/ 1207] Overall Loss 0.325841 Objective Loss 0.325841 LR 0.001000 Time 0.043025 -2023-02-13 17:59:22,248 - Epoch: [93][ 40/ 1207] Overall Loss 0.316807 Objective Loss 0.316807 LR 0.001000 Time 0.037080 -2023-02-13 17:59:22,444 - Epoch: [93][ 50/ 1207] Overall Loss 0.318380 Objective Loss 0.318380 LR 0.001000 Time 0.033578 -2023-02-13 17:59:22,638 - Epoch: [93][ 60/ 1207] Overall Loss 0.321427 Objective Loss 0.321427 LR 0.001000 Time 0.031207 -2023-02-13 17:59:22,834 - Epoch: [93][ 70/ 1207] Overall Loss 0.320872 Objective Loss 0.320872 LR 0.001000 Time 0.029537 -2023-02-13 17:59:23,026 - Epoch: [93][ 80/ 1207] Overall Loss 0.322378 Objective Loss 0.322378 LR 0.001000 Time 0.028245 -2023-02-13 17:59:23,222 - Epoch: [93][ 90/ 1207] Overall Loss 0.327567 Objective Loss 0.327567 LR 0.001000 Time 0.027279 -2023-02-13 17:59:23,415 - Epoch: [93][ 100/ 1207] Overall Loss 0.325358 Objective Loss 0.325358 LR 0.001000 Time 0.026478 -2023-02-13 17:59:23,611 - Epoch: [93][ 110/ 1207] Overall Loss 0.327902 Objective Loss 0.327902 LR 0.001000 Time 0.025851 -2023-02-13 17:59:23,803 - Epoch: [93][ 120/ 1207] Overall Loss 0.328480 Objective Loss 0.328480 LR 0.001000 Time 0.025294 -2023-02-13 17:59:23,999 - Epoch: [93][ 130/ 1207] Overall Loss 0.326651 Objective Loss 0.326651 LR 0.001000 Time 0.024853 -2023-02-13 17:59:24,192 - Epoch: [93][ 140/ 1207] Overall Loss 0.326151 Objective Loss 0.326151 LR 0.001000 Time 0.024452 -2023-02-13 17:59:24,388 - Epoch: [93][ 150/ 1207] Overall Loss 0.327074 Objective Loss 0.327074 LR 0.001000 Time 0.024128 -2023-02-13 17:59:24,581 - Epoch: [93][ 160/ 1207] Overall Loss 0.327785 Objective Loss 0.327785 LR 0.001000 Time 0.023822 -2023-02-13 17:59:24,777 - Epoch: [93][ 170/ 1207] Overall Loss 0.329189 Objective Loss 0.329189 LR 0.001000 Time 0.023570 -2023-02-13 17:59:24,970 - Epoch: [93][ 180/ 1207] Overall Loss 0.329446 Objective Loss 0.329446 LR 0.001000 Time 0.023329 -2023-02-13 17:59:25,165 - Epoch: [93][ 190/ 1207] Overall Loss 0.330305 Objective Loss 0.330305 LR 0.001000 Time 0.023127 -2023-02-13 17:59:25,357 - Epoch: [93][ 200/ 1207] Overall Loss 0.330336 Objective Loss 0.330336 LR 0.001000 Time 0.022931 -2023-02-13 17:59:25,555 - Epoch: [93][ 210/ 1207] Overall Loss 0.331749 Objective Loss 0.331749 LR 0.001000 Time 0.022779 -2023-02-13 17:59:25,747 - Epoch: [93][ 220/ 1207] Overall Loss 0.332091 Objective Loss 0.332091 LR 0.001000 Time 0.022616 -2023-02-13 17:59:25,944 - Epoch: [93][ 230/ 1207] Overall Loss 0.332984 Objective Loss 0.332984 LR 0.001000 Time 0.022484 -2023-02-13 17:59:26,137 - Epoch: [93][ 240/ 1207] Overall Loss 0.331990 Objective Loss 0.331990 LR 0.001000 Time 0.022351 -2023-02-13 17:59:26,332 - Epoch: [93][ 250/ 1207] Overall Loss 0.331080 Objective Loss 0.331080 LR 0.001000 Time 0.022237 -2023-02-13 17:59:26,525 - Epoch: [93][ 260/ 1207] Overall Loss 0.329796 Objective Loss 0.329796 LR 0.001000 Time 0.022123 -2023-02-13 17:59:26,721 - Epoch: [93][ 270/ 1207] Overall Loss 0.329802 Objective Loss 0.329802 LR 0.001000 Time 0.022028 -2023-02-13 17:59:26,914 - Epoch: [93][ 280/ 1207] Overall Loss 0.329594 Objective Loss 0.329594 LR 0.001000 Time 0.021928 -2023-02-13 17:59:27,109 - Epoch: [93][ 290/ 1207] Overall Loss 0.329504 Objective Loss 0.329504 LR 0.001000 Time 0.021843 -2023-02-13 17:59:27,301 - Epoch: [93][ 300/ 1207] Overall Loss 0.329665 Objective Loss 0.329665 LR 0.001000 Time 0.021754 -2023-02-13 17:59:27,497 - Epoch: [93][ 310/ 1207] Overall Loss 0.329003 Objective Loss 0.329003 LR 0.001000 Time 0.021682 -2023-02-13 17:59:27,690 - Epoch: [93][ 320/ 1207] Overall Loss 0.328532 Objective Loss 0.328532 LR 0.001000 Time 0.021607 -2023-02-13 17:59:27,885 - Epoch: [93][ 330/ 1207] Overall Loss 0.327461 Objective Loss 0.327461 LR 0.001000 Time 0.021542 -2023-02-13 17:59:28,077 - Epoch: [93][ 340/ 1207] Overall Loss 0.326838 Objective Loss 0.326838 LR 0.001000 Time 0.021473 -2023-02-13 17:59:28,273 - Epoch: [93][ 350/ 1207] Overall Loss 0.327000 Objective Loss 0.327000 LR 0.001000 Time 0.021417 -2023-02-13 17:59:28,465 - Epoch: [93][ 360/ 1207] Overall Loss 0.326419 Objective Loss 0.326419 LR 0.001000 Time 0.021356 -2023-02-13 17:59:28,662 - Epoch: [93][ 370/ 1207] Overall Loss 0.326008 Objective Loss 0.326008 LR 0.001000 Time 0.021310 -2023-02-13 17:59:28,854 - Epoch: [93][ 380/ 1207] Overall Loss 0.324790 Objective Loss 0.324790 LR 0.001000 Time 0.021254 -2023-02-13 17:59:29,050 - Epoch: [93][ 390/ 1207] Overall Loss 0.325303 Objective Loss 0.325303 LR 0.001000 Time 0.021210 -2023-02-13 17:59:29,243 - Epoch: [93][ 400/ 1207] Overall Loss 0.325012 Objective Loss 0.325012 LR 0.001000 Time 0.021161 -2023-02-13 17:59:29,439 - Epoch: [93][ 410/ 1207] Overall Loss 0.325121 Objective Loss 0.325121 LR 0.001000 Time 0.021121 -2023-02-13 17:59:29,632 - Epoch: [93][ 420/ 1207] Overall Loss 0.324467 Objective Loss 0.324467 LR 0.001000 Time 0.021078 -2023-02-13 17:59:29,828 - Epoch: [93][ 430/ 1207] Overall Loss 0.324492 Objective Loss 0.324492 LR 0.001000 Time 0.021042 -2023-02-13 17:59:30,020 - Epoch: [93][ 440/ 1207] Overall Loss 0.325078 Objective Loss 0.325078 LR 0.001000 Time 0.021001 -2023-02-13 17:59:30,216 - Epoch: [93][ 450/ 1207] Overall Loss 0.326144 Objective Loss 0.326144 LR 0.001000 Time 0.020967 -2023-02-13 17:59:30,409 - Epoch: [93][ 460/ 1207] Overall Loss 0.326679 Objective Loss 0.326679 LR 0.001000 Time 0.020931 -2023-02-13 17:59:30,606 - Epoch: [93][ 470/ 1207] Overall Loss 0.327023 Objective Loss 0.327023 LR 0.001000 Time 0.020904 -2023-02-13 17:59:30,799 - Epoch: [93][ 480/ 1207] Overall Loss 0.326497 Objective Loss 0.326497 LR 0.001000 Time 0.020870 -2023-02-13 17:59:30,996 - Epoch: [93][ 490/ 1207] Overall Loss 0.326345 Objective Loss 0.326345 LR 0.001000 Time 0.020845 -2023-02-13 17:59:31,189 - Epoch: [93][ 500/ 1207] Overall Loss 0.326822 Objective Loss 0.326822 LR 0.001000 Time 0.020813 -2023-02-13 17:59:31,385 - Epoch: [93][ 510/ 1207] Overall Loss 0.326898 Objective Loss 0.326898 LR 0.001000 Time 0.020789 -2023-02-13 17:59:31,578 - Epoch: [93][ 520/ 1207] Overall Loss 0.326945 Objective Loss 0.326945 LR 0.001000 Time 0.020760 -2023-02-13 17:59:31,775 - Epoch: [93][ 530/ 1207] Overall Loss 0.326883 Objective Loss 0.326883 LR 0.001000 Time 0.020739 -2023-02-13 17:59:31,969 - Epoch: [93][ 540/ 1207] Overall Loss 0.326457 Objective Loss 0.326457 LR 0.001000 Time 0.020713 -2023-02-13 17:59:32,165 - Epoch: [93][ 550/ 1207] Overall Loss 0.327290 Objective Loss 0.327290 LR 0.001000 Time 0.020692 -2023-02-13 17:59:32,359 - Epoch: [93][ 560/ 1207] Overall Loss 0.327742 Objective Loss 0.327742 LR 0.001000 Time 0.020667 -2023-02-13 17:59:32,555 - Epoch: [93][ 570/ 1207] Overall Loss 0.327552 Objective Loss 0.327552 LR 0.001000 Time 0.020649 -2023-02-13 17:59:32,749 - Epoch: [93][ 580/ 1207] Overall Loss 0.327823 Objective Loss 0.327823 LR 0.001000 Time 0.020627 -2023-02-13 17:59:32,945 - Epoch: [93][ 590/ 1207] Overall Loss 0.328377 Objective Loss 0.328377 LR 0.001000 Time 0.020609 -2023-02-13 17:59:33,138 - Epoch: [93][ 600/ 1207] Overall Loss 0.328402 Objective Loss 0.328402 LR 0.001000 Time 0.020587 -2023-02-13 17:59:33,334 - Epoch: [93][ 610/ 1207] Overall Loss 0.328057 Objective Loss 0.328057 LR 0.001000 Time 0.020569 -2023-02-13 17:59:33,527 - Epoch: [93][ 620/ 1207] Overall Loss 0.327747 Objective Loss 0.327747 LR 0.001000 Time 0.020548 -2023-02-13 17:59:33,723 - Epoch: [93][ 630/ 1207] Overall Loss 0.327572 Objective Loss 0.327572 LR 0.001000 Time 0.020533 -2023-02-13 17:59:33,916 - Epoch: [93][ 640/ 1207] Overall Loss 0.327834 Objective Loss 0.327834 LR 0.001000 Time 0.020513 -2023-02-13 17:59:34,112 - Epoch: [93][ 650/ 1207] Overall Loss 0.328258 Objective Loss 0.328258 LR 0.001000 Time 0.020498 -2023-02-13 17:59:34,310 - Epoch: [93][ 660/ 1207] Overall Loss 0.328882 Objective Loss 0.328882 LR 0.001000 Time 0.020477 -2023-02-13 17:59:34,506 - Epoch: [93][ 670/ 1207] Overall Loss 0.329011 Objective Loss 0.329011 LR 0.001000 Time 0.020464 -2023-02-13 17:59:34,700 - Epoch: [93][ 680/ 1207] Overall Loss 0.328999 Objective Loss 0.328999 LR 0.001000 Time 0.020448 -2023-02-13 17:59:34,897 - Epoch: [93][ 690/ 1207] Overall Loss 0.328978 Objective Loss 0.328978 LR 0.001000 Time 0.020436 -2023-02-13 17:59:35,090 - Epoch: [93][ 700/ 1207] Overall Loss 0.329671 Objective Loss 0.329671 LR 0.001000 Time 0.020420 -2023-02-13 17:59:35,287 - Epoch: [93][ 710/ 1207] Overall Loss 0.329553 Objective Loss 0.329553 LR 0.001000 Time 0.020409 -2023-02-13 17:59:35,480 - Epoch: [93][ 720/ 1207] Overall Loss 0.329045 Objective Loss 0.329045 LR 0.001000 Time 0.020394 -2023-02-13 17:59:35,677 - Epoch: [93][ 730/ 1207] Overall Loss 0.328737 Objective Loss 0.328737 LR 0.001000 Time 0.020383 -2023-02-13 17:59:35,871 - Epoch: [93][ 740/ 1207] Overall Loss 0.329017 Objective Loss 0.329017 LR 0.001000 Time 0.020370 -2023-02-13 17:59:36,067 - Epoch: [93][ 750/ 1207] Overall Loss 0.329768 Objective Loss 0.329768 LR 0.001000 Time 0.020359 -2023-02-13 17:59:36,261 - Epoch: [93][ 760/ 1207] Overall Loss 0.330016 Objective Loss 0.330016 LR 0.001000 Time 0.020345 -2023-02-13 17:59:36,457 - Epoch: [93][ 770/ 1207] Overall Loss 0.330056 Objective Loss 0.330056 LR 0.001000 Time 0.020335 -2023-02-13 17:59:36,650 - Epoch: [93][ 780/ 1207] Overall Loss 0.329880 Objective Loss 0.329880 LR 0.001000 Time 0.020322 -2023-02-13 17:59:36,847 - Epoch: [93][ 790/ 1207] Overall Loss 0.330042 Objective Loss 0.330042 LR 0.001000 Time 0.020313 -2023-02-13 17:59:37,040 - Epoch: [93][ 800/ 1207] Overall Loss 0.330183 Objective Loss 0.330183 LR 0.001000 Time 0.020300 -2023-02-13 17:59:37,237 - Epoch: [93][ 810/ 1207] Overall Loss 0.330054 Objective Loss 0.330054 LR 0.001000 Time 0.020292 -2023-02-13 17:59:37,430 - Epoch: [93][ 820/ 1207] Overall Loss 0.330129 Objective Loss 0.330129 LR 0.001000 Time 0.020280 -2023-02-13 17:59:37,627 - Epoch: [93][ 830/ 1207] Overall Loss 0.329982 Objective Loss 0.329982 LR 0.001000 Time 0.020272 -2023-02-13 17:59:37,820 - Epoch: [93][ 840/ 1207] Overall Loss 0.329786 Objective Loss 0.329786 LR 0.001000 Time 0.020260 -2023-02-13 17:59:38,017 - Epoch: [93][ 850/ 1207] Overall Loss 0.329795 Objective Loss 0.329795 LR 0.001000 Time 0.020253 -2023-02-13 17:59:38,210 - Epoch: [93][ 860/ 1207] Overall Loss 0.329768 Objective Loss 0.329768 LR 0.001000 Time 0.020242 -2023-02-13 17:59:38,407 - Epoch: [93][ 870/ 1207] Overall Loss 0.329535 Objective Loss 0.329535 LR 0.001000 Time 0.020235 -2023-02-13 17:59:38,601 - Epoch: [93][ 880/ 1207] Overall Loss 0.329411 Objective Loss 0.329411 LR 0.001000 Time 0.020225 -2023-02-13 17:59:38,797 - Epoch: [93][ 890/ 1207] Overall Loss 0.329306 Objective Loss 0.329306 LR 0.001000 Time 0.020218 -2023-02-13 17:59:38,991 - Epoch: [93][ 900/ 1207] Overall Loss 0.329731 Objective Loss 0.329731 LR 0.001000 Time 0.020208 -2023-02-13 17:59:39,187 - Epoch: [93][ 910/ 1207] Overall Loss 0.329816 Objective Loss 0.329816 LR 0.001000 Time 0.020201 -2023-02-13 17:59:39,381 - Epoch: [93][ 920/ 1207] Overall Loss 0.330064 Objective Loss 0.330064 LR 0.001000 Time 0.020192 -2023-02-13 17:59:39,578 - Epoch: [93][ 930/ 1207] Overall Loss 0.329980 Objective Loss 0.329980 LR 0.001000 Time 0.020186 -2023-02-13 17:59:39,771 - Epoch: [93][ 940/ 1207] Overall Loss 0.330153 Objective Loss 0.330153 LR 0.001000 Time 0.020176 -2023-02-13 17:59:39,967 - Epoch: [93][ 950/ 1207] Overall Loss 0.330361 Objective Loss 0.330361 LR 0.001000 Time 0.020170 -2023-02-13 17:59:40,160 - Epoch: [93][ 960/ 1207] Overall Loss 0.330355 Objective Loss 0.330355 LR 0.001000 Time 0.020161 -2023-02-13 17:59:40,357 - Epoch: [93][ 970/ 1207] Overall Loss 0.330765 Objective Loss 0.330765 LR 0.001000 Time 0.020155 -2023-02-13 17:59:40,549 - Epoch: [93][ 980/ 1207] Overall Loss 0.330732 Objective Loss 0.330732 LR 0.001000 Time 0.020146 -2023-02-13 17:59:40,746 - Epoch: [93][ 990/ 1207] Overall Loss 0.330547 Objective Loss 0.330547 LR 0.001000 Time 0.020141 -2023-02-13 17:59:40,940 - Epoch: [93][ 1000/ 1207] Overall Loss 0.330388 Objective Loss 0.330388 LR 0.001000 Time 0.020133 -2023-02-13 17:59:41,137 - Epoch: [93][ 1010/ 1207] Overall Loss 0.330941 Objective Loss 0.330941 LR 0.001000 Time 0.020128 -2023-02-13 17:59:41,330 - Epoch: [93][ 1020/ 1207] Overall Loss 0.331108 Objective Loss 0.331108 LR 0.001000 Time 0.020119 -2023-02-13 17:59:41,526 - Epoch: [93][ 1030/ 1207] Overall Loss 0.331193 Objective Loss 0.331193 LR 0.001000 Time 0.020114 -2023-02-13 17:59:41,720 - Epoch: [93][ 1040/ 1207] Overall Loss 0.331223 Objective Loss 0.331223 LR 0.001000 Time 0.020107 -2023-02-13 17:59:41,916 - Epoch: [93][ 1050/ 1207] Overall Loss 0.331222 Objective Loss 0.331222 LR 0.001000 Time 0.020102 -2023-02-13 17:59:42,110 - Epoch: [93][ 1060/ 1207] Overall Loss 0.331389 Objective Loss 0.331389 LR 0.001000 Time 0.020094 -2023-02-13 17:59:42,306 - Epoch: [93][ 1070/ 1207] Overall Loss 0.331329 Objective Loss 0.331329 LR 0.001000 Time 0.020090 -2023-02-13 17:59:42,499 - Epoch: [93][ 1080/ 1207] Overall Loss 0.331320 Objective Loss 0.331320 LR 0.001000 Time 0.020082 -2023-02-13 17:59:42,696 - Epoch: [93][ 1090/ 1207] Overall Loss 0.331509 Objective Loss 0.331509 LR 0.001000 Time 0.020078 -2023-02-13 17:59:42,889 - Epoch: [93][ 1100/ 1207] Overall Loss 0.331324 Objective Loss 0.331324 LR 0.001000 Time 0.020071 -2023-02-13 17:59:43,085 - Epoch: [93][ 1110/ 1207] Overall Loss 0.331500 Objective Loss 0.331500 LR 0.001000 Time 0.020066 -2023-02-13 17:59:43,279 - Epoch: [93][ 1120/ 1207] Overall Loss 0.331471 Objective Loss 0.331471 LR 0.001000 Time 0.020060 -2023-02-13 17:59:43,475 - Epoch: [93][ 1130/ 1207] Overall Loss 0.331372 Objective Loss 0.331372 LR 0.001000 Time 0.020056 -2023-02-13 17:59:43,669 - Epoch: [93][ 1140/ 1207] Overall Loss 0.331561 Objective Loss 0.331561 LR 0.001000 Time 0.020050 -2023-02-13 17:59:43,866 - Epoch: [93][ 1150/ 1207] Overall Loss 0.331820 Objective Loss 0.331820 LR 0.001000 Time 0.020046 -2023-02-13 17:59:44,059 - Epoch: [93][ 1160/ 1207] Overall Loss 0.331984 Objective Loss 0.331984 LR 0.001000 Time 0.020039 -2023-02-13 17:59:44,255 - Epoch: [93][ 1170/ 1207] Overall Loss 0.331956 Objective Loss 0.331956 LR 0.001000 Time 0.020035 -2023-02-13 17:59:44,449 - Epoch: [93][ 1180/ 1207] Overall Loss 0.332127 Objective Loss 0.332127 LR 0.001000 Time 0.020029 -2023-02-13 17:59:44,645 - Epoch: [93][ 1190/ 1207] Overall Loss 0.332085 Objective Loss 0.332085 LR 0.001000 Time 0.020026 -2023-02-13 17:59:44,896 - Epoch: [93][ 1200/ 1207] Overall Loss 0.332092 Objective Loss 0.332092 LR 0.001000 Time 0.020067 -2023-02-13 17:59:45,010 - Epoch: [93][ 1207/ 1207] Overall Loss 0.332190 Objective Loss 0.332190 Top1 81.402439 Top5 96.036585 LR 0.001000 Time 0.020045 -2023-02-13 17:59:45,093 - --- validate (epoch=93)----------- -2023-02-13 17:59:45,094 - 34311 samples (256 per mini-batch) -2023-02-13 17:59:45,589 - Epoch: [93][ 10/ 135] Loss 0.349121 Top1 81.640625 Top5 97.226562 -2023-02-13 17:59:45,715 - Epoch: [93][ 20/ 135] Loss 0.367954 Top1 81.113281 Top5 96.894531 -2023-02-13 17:59:45,841 - Epoch: [93][ 30/ 135] Loss 0.370480 Top1 81.119792 Top5 96.914062 -2023-02-13 17:59:45,967 - Epoch: [93][ 40/ 135] Loss 0.371822 Top1 80.927734 Top5 96.953125 -2023-02-13 17:59:46,093 - Epoch: [93][ 50/ 135] Loss 0.376310 Top1 80.835938 Top5 96.906250 -2023-02-13 17:59:46,220 - Epoch: [93][ 60/ 135] Loss 0.382552 Top1 80.520833 Top5 96.888021 -2023-02-13 17:59:46,345 - Epoch: [93][ 70/ 135] Loss 0.378021 Top1 80.708705 Top5 96.863839 -2023-02-13 17:59:46,482 - Epoch: [93][ 80/ 135] Loss 0.377446 Top1 80.805664 Top5 96.870117 -2023-02-13 17:59:46,622 - Epoch: [93][ 90/ 135] Loss 0.377813 Top1 80.690104 Top5 96.870660 -2023-02-13 17:59:46,752 - Epoch: [93][ 100/ 135] Loss 0.376699 Top1 80.832031 Top5 96.906250 -2023-02-13 17:59:46,880 - Epoch: [93][ 110/ 135] Loss 0.373072 Top1 80.965909 Top5 96.949574 -2023-02-13 17:59:47,005 - Epoch: [93][ 120/ 135] Loss 0.369054 Top1 80.917969 Top5 96.962891 -2023-02-13 17:59:47,135 - Epoch: [93][ 130/ 135] Loss 0.372302 Top1 80.805288 Top5 96.908053 -2023-02-13 17:59:47,182 - Epoch: [93][ 135/ 135] Loss 0.378470 Top1 80.790417 Top5 96.901868 -2023-02-13 17:59:47,270 - ==> Top1: 80.790 Top5: 96.902 Loss: 0.378 - -2023-02-13 17:59:47,270 - ==> Confusion: -[[ 879 3 2 1 13 4 1 2 2 32 1 5 1 5 6 1 1 1 1 2 4] - [ 0 958 1 2 9 13 1 11 4 1 2 5 2 0 1 0 6 0 9 2 6] - [ 9 5 929 16 8 1 30 17 1 1 8 0 2 3 4 5 1 1 9 3 5] - [ 6 1 31 855 6 2 3 3 6 0 22 0 5 2 27 2 6 4 27 0 8] - [ 22 8 0 0 984 5 2 0 1 4 1 7 1 4 7 5 10 1 1 1 2] - [ 2 38 3 6 12 923 4 28 2 2 2 13 3 13 1 0 3 1 2 7 5] - [ 2 6 15 3 1 4 1036 5 1 1 4 1 0 0 2 6 3 1 2 3 3] - [ 3 17 13 2 3 24 6 902 2 4 1 6 5 2 1 0 0 0 25 6 2] - [ 27 2 0 1 1 0 1 2 896 39 13 3 0 8 10 0 0 0 6 0 0] - [ 118 1 2 0 5 0 0 2 39 818 0 1 0 13 4 1 1 2 2 1 2] - [ 1 3 3 5 4 0 4 5 19 2 970 1 2 6 3 0 0 0 19 0 4] - [ 3 3 2 0 5 13 1 9 2 1 3 896 23 5 2 6 5 8 2 15 1] - [ 1 1 3 6 6 4 0 3 3 0 2 48 828 1 5 7 2 23 4 0 12] - [ 2 5 6 1 10 7 2 6 25 24 15 6 0 886 5 4 7 4 0 4 5] - [ 12 5 2 7 6 2 1 2 34 7 6 3 1 3 975 1 1 8 11 0 5] - [ 5 2 10 0 10 0 11 0 0 0 0 6 4 1 1 966 10 8 1 6 5] - [ 3 13 2 1 10 2 0 2 3 1 1 4 2 1 3 9 989 1 3 2 9] - [ 6 4 2 1 0 3 3 2 3 0 3 11 20 0 2 16 4 961 0 3 7] - [ 4 5 4 5 4 0 1 18 7 1 9 0 4 0 10 0 1 2 1006 4 1] - [ 1 4 1 1 1 9 14 30 1 0 2 15 2 4 0 5 11 3 3 1036 5] - [ 269 376 252 109 193 162 144 209 175 131 323 140 281 325 231 127 311 109 267 273 9027]] - -2023-02-13 17:59:47,272 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 17:59:47,272 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 17:59:47,278 - - -2023-02-13 17:59:47,278 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 17:59:48,153 - Epoch: [94][ 10/ 1207] Overall Loss 0.325715 Objective Loss 0.325715 LR 0.001000 Time 0.087379 -2023-02-13 17:59:48,352 - Epoch: [94][ 20/ 1207] Overall Loss 0.327613 Objective Loss 0.327613 LR 0.001000 Time 0.053663 -2023-02-13 17:59:48,542 - Epoch: [94][ 30/ 1207] Overall Loss 0.327455 Objective Loss 0.327455 LR 0.001000 Time 0.042085 -2023-02-13 17:59:48,731 - Epoch: [94][ 40/ 1207] Overall Loss 0.326836 Objective Loss 0.326836 LR 0.001000 Time 0.036279 -2023-02-13 17:59:48,920 - Epoch: [94][ 50/ 1207] Overall Loss 0.327993 Objective Loss 0.327993 LR 0.001000 Time 0.032803 -2023-02-13 17:59:49,109 - Epoch: [94][ 60/ 1207] Overall Loss 0.326426 Objective Loss 0.326426 LR 0.001000 Time 0.030468 -2023-02-13 17:59:49,298 - Epoch: [94][ 70/ 1207] Overall Loss 0.331087 Objective Loss 0.331087 LR 0.001000 Time 0.028817 -2023-02-13 17:59:49,487 - Epoch: [94][ 80/ 1207] Overall Loss 0.332538 Objective Loss 0.332538 LR 0.001000 Time 0.027565 -2023-02-13 17:59:49,677 - Epoch: [94][ 90/ 1207] Overall Loss 0.331845 Objective Loss 0.331845 LR 0.001000 Time 0.026614 -2023-02-13 17:59:49,866 - Epoch: [94][ 100/ 1207] Overall Loss 0.332064 Objective Loss 0.332064 LR 0.001000 Time 0.025837 -2023-02-13 17:59:50,055 - Epoch: [94][ 110/ 1207] Overall Loss 0.330609 Objective Loss 0.330609 LR 0.001000 Time 0.025202 -2023-02-13 17:59:50,244 - Epoch: [94][ 120/ 1207] Overall Loss 0.329730 Objective Loss 0.329730 LR 0.001000 Time 0.024677 -2023-02-13 17:59:50,434 - Epoch: [94][ 130/ 1207] Overall Loss 0.324866 Objective Loss 0.324866 LR 0.001000 Time 0.024236 -2023-02-13 17:59:50,623 - Epoch: [94][ 140/ 1207] Overall Loss 0.324149 Objective Loss 0.324149 LR 0.001000 Time 0.023855 -2023-02-13 17:59:50,813 - Epoch: [94][ 150/ 1207] Overall Loss 0.324600 Objective Loss 0.324600 LR 0.001000 Time 0.023526 -2023-02-13 17:59:51,003 - Epoch: [94][ 160/ 1207] Overall Loss 0.325199 Objective Loss 0.325199 LR 0.001000 Time 0.023242 -2023-02-13 17:59:51,193 - Epoch: [94][ 170/ 1207] Overall Loss 0.324294 Objective Loss 0.324294 LR 0.001000 Time 0.022988 -2023-02-13 17:59:51,381 - Epoch: [94][ 180/ 1207] Overall Loss 0.323026 Objective Loss 0.323026 LR 0.001000 Time 0.022753 -2023-02-13 17:59:51,570 - Epoch: [94][ 190/ 1207] Overall Loss 0.323144 Objective Loss 0.323144 LR 0.001000 Time 0.022552 -2023-02-13 17:59:51,760 - Epoch: [94][ 200/ 1207] Overall Loss 0.321726 Objective Loss 0.321726 LR 0.001000 Time 0.022370 -2023-02-13 17:59:51,950 - Epoch: [94][ 210/ 1207] Overall Loss 0.321433 Objective Loss 0.321433 LR 0.001000 Time 0.022209 -2023-02-13 17:59:52,139 - Epoch: [94][ 220/ 1207] Overall Loss 0.320719 Objective Loss 0.320719 LR 0.001000 Time 0.022055 -2023-02-13 17:59:52,329 - Epoch: [94][ 230/ 1207] Overall Loss 0.320626 Objective Loss 0.320626 LR 0.001000 Time 0.021921 -2023-02-13 17:59:52,518 - Epoch: [94][ 240/ 1207] Overall Loss 0.320682 Objective Loss 0.320682 LR 0.001000 Time 0.021793 -2023-02-13 17:59:52,709 - Epoch: [94][ 250/ 1207] Overall Loss 0.321051 Objective Loss 0.321051 LR 0.001000 Time 0.021684 -2023-02-13 17:59:52,897 - Epoch: [94][ 260/ 1207] Overall Loss 0.321233 Objective Loss 0.321233 LR 0.001000 Time 0.021574 -2023-02-13 17:59:53,087 - Epoch: [94][ 270/ 1207] Overall Loss 0.321232 Objective Loss 0.321232 LR 0.001000 Time 0.021475 -2023-02-13 17:59:53,276 - Epoch: [94][ 280/ 1207] Overall Loss 0.321572 Objective Loss 0.321572 LR 0.001000 Time 0.021384 -2023-02-13 17:59:53,465 - Epoch: [94][ 290/ 1207] Overall Loss 0.322836 Objective Loss 0.322836 LR 0.001000 Time 0.021297 -2023-02-13 17:59:53,655 - Epoch: [94][ 300/ 1207] Overall Loss 0.323047 Objective Loss 0.323047 LR 0.001000 Time 0.021216 -2023-02-13 17:59:53,844 - Epoch: [94][ 310/ 1207] Overall Loss 0.322875 Objective Loss 0.322875 LR 0.001000 Time 0.021144 -2023-02-13 17:59:54,032 - Epoch: [94][ 320/ 1207] Overall Loss 0.323618 Objective Loss 0.323618 LR 0.001000 Time 0.021069 -2023-02-13 17:59:54,222 - Epoch: [94][ 330/ 1207] Overall Loss 0.324194 Objective Loss 0.324194 LR 0.001000 Time 0.021004 -2023-02-13 17:59:54,411 - Epoch: [94][ 340/ 1207] Overall Loss 0.324791 Objective Loss 0.324791 LR 0.001000 Time 0.020940 -2023-02-13 17:59:54,600 - Epoch: [94][ 350/ 1207] Overall Loss 0.323574 Objective Loss 0.323574 LR 0.001000 Time 0.020880 -2023-02-13 17:59:54,788 - Epoch: [94][ 360/ 1207] Overall Loss 0.323737 Objective Loss 0.323737 LR 0.001000 Time 0.020824 -2023-02-13 17:59:54,978 - Epoch: [94][ 370/ 1207] Overall Loss 0.323945 Objective Loss 0.323945 LR 0.001000 Time 0.020772 -2023-02-13 17:59:55,167 - Epoch: [94][ 380/ 1207] Overall Loss 0.323574 Objective Loss 0.323574 LR 0.001000 Time 0.020722 -2023-02-13 17:59:55,356 - Epoch: [94][ 390/ 1207] Overall Loss 0.324324 Objective Loss 0.324324 LR 0.001000 Time 0.020676 -2023-02-13 17:59:55,545 - Epoch: [94][ 400/ 1207] Overall Loss 0.324410 Objective Loss 0.324410 LR 0.001000 Time 0.020629 -2023-02-13 17:59:55,735 - Epoch: [94][ 410/ 1207] Overall Loss 0.324825 Objective Loss 0.324825 LR 0.001000 Time 0.020589 -2023-02-13 17:59:55,925 - Epoch: [94][ 420/ 1207] Overall Loss 0.325163 Objective Loss 0.325163 LR 0.001000 Time 0.020550 -2023-02-13 17:59:56,114 - Epoch: [94][ 430/ 1207] Overall Loss 0.325617 Objective Loss 0.325617 LR 0.001000 Time 0.020511 -2023-02-13 17:59:56,303 - Epoch: [94][ 440/ 1207] Overall Loss 0.326393 Objective Loss 0.326393 LR 0.001000 Time 0.020473 -2023-02-13 17:59:56,493 - Epoch: [94][ 450/ 1207] Overall Loss 0.326617 Objective Loss 0.326617 LR 0.001000 Time 0.020439 -2023-02-13 17:59:56,682 - Epoch: [94][ 460/ 1207] Overall Loss 0.326922 Objective Loss 0.326922 LR 0.001000 Time 0.020405 -2023-02-13 17:59:56,872 - Epoch: [94][ 470/ 1207] Overall Loss 0.326989 Objective Loss 0.326989 LR 0.001000 Time 0.020374 -2023-02-13 17:59:57,061 - Epoch: [94][ 480/ 1207] Overall Loss 0.326368 Objective Loss 0.326368 LR 0.001000 Time 0.020342 -2023-02-13 17:59:57,251 - Epoch: [94][ 490/ 1207] Overall Loss 0.326804 Objective Loss 0.326804 LR 0.001000 Time 0.020315 -2023-02-13 17:59:57,440 - Epoch: [94][ 500/ 1207] Overall Loss 0.326173 Objective Loss 0.326173 LR 0.001000 Time 0.020286 -2023-02-13 17:59:57,630 - Epoch: [94][ 510/ 1207] Overall Loss 0.326264 Objective Loss 0.326264 LR 0.001000 Time 0.020260 -2023-02-13 17:59:57,820 - Epoch: [94][ 520/ 1207] Overall Loss 0.326203 Objective Loss 0.326203 LR 0.001000 Time 0.020234 -2023-02-13 17:59:58,009 - Epoch: [94][ 530/ 1207] Overall Loss 0.325697 Objective Loss 0.325697 LR 0.001000 Time 0.020210 -2023-02-13 17:59:58,198 - Epoch: [94][ 540/ 1207] Overall Loss 0.326177 Objective Loss 0.326177 LR 0.001000 Time 0.020184 -2023-02-13 17:59:58,388 - Epoch: [94][ 550/ 1207] Overall Loss 0.326204 Objective Loss 0.326204 LR 0.001000 Time 0.020161 -2023-02-13 17:59:58,577 - Epoch: [94][ 560/ 1207] Overall Loss 0.326490 Objective Loss 0.326490 LR 0.001000 Time 0.020138 -2023-02-13 17:59:58,767 - Epoch: [94][ 570/ 1207] Overall Loss 0.326827 Objective Loss 0.326827 LR 0.001000 Time 0.020118 -2023-02-13 17:59:58,956 - Epoch: [94][ 580/ 1207] Overall Loss 0.326906 Objective Loss 0.326906 LR 0.001000 Time 0.020096 -2023-02-13 17:59:59,146 - Epoch: [94][ 590/ 1207] Overall Loss 0.327058 Objective Loss 0.327058 LR 0.001000 Time 0.020077 -2023-02-13 17:59:59,335 - Epoch: [94][ 600/ 1207] Overall Loss 0.327215 Objective Loss 0.327215 LR 0.001000 Time 0.020057 -2023-02-13 17:59:59,525 - Epoch: [94][ 610/ 1207] Overall Loss 0.327988 Objective Loss 0.327988 LR 0.001000 Time 0.020038 -2023-02-13 17:59:59,714 - Epoch: [94][ 620/ 1207] Overall Loss 0.328319 Objective Loss 0.328319 LR 0.001000 Time 0.020020 -2023-02-13 17:59:59,903 - Epoch: [94][ 630/ 1207] Overall Loss 0.328341 Objective Loss 0.328341 LR 0.001000 Time 0.020002 -2023-02-13 18:00:00,092 - Epoch: [94][ 640/ 1207] Overall Loss 0.327891 Objective Loss 0.327891 LR 0.001000 Time 0.019984 -2023-02-13 18:00:00,282 - Epoch: [94][ 650/ 1207] Overall Loss 0.327405 Objective Loss 0.327405 LR 0.001000 Time 0.019968 -2023-02-13 18:00:00,471 - Epoch: [94][ 660/ 1207] Overall Loss 0.327687 Objective Loss 0.327687 LR 0.001000 Time 0.019951 -2023-02-13 18:00:00,661 - Epoch: [94][ 670/ 1207] Overall Loss 0.326921 Objective Loss 0.326921 LR 0.001000 Time 0.019937 -2023-02-13 18:00:00,852 - Epoch: [94][ 680/ 1207] Overall Loss 0.327103 Objective Loss 0.327103 LR 0.001000 Time 0.019924 -2023-02-13 18:00:01,042 - Epoch: [94][ 690/ 1207] Overall Loss 0.326779 Objective Loss 0.326779 LR 0.001000 Time 0.019909 -2023-02-13 18:00:01,231 - Epoch: [94][ 700/ 1207] Overall Loss 0.327083 Objective Loss 0.327083 LR 0.001000 Time 0.019895 -2023-02-13 18:00:01,421 - Epoch: [94][ 710/ 1207] Overall Loss 0.327904 Objective Loss 0.327904 LR 0.001000 Time 0.019882 -2023-02-13 18:00:01,611 - Epoch: [94][ 720/ 1207] Overall Loss 0.327642 Objective Loss 0.327642 LR 0.001000 Time 0.019868 -2023-02-13 18:00:01,801 - Epoch: [94][ 730/ 1207] Overall Loss 0.327525 Objective Loss 0.327525 LR 0.001000 Time 0.019857 -2023-02-13 18:00:01,991 - Epoch: [94][ 740/ 1207] Overall Loss 0.327471 Objective Loss 0.327471 LR 0.001000 Time 0.019844 -2023-02-13 18:00:02,181 - Epoch: [94][ 750/ 1207] Overall Loss 0.327297 Objective Loss 0.327297 LR 0.001000 Time 0.019833 -2023-02-13 18:00:02,371 - Epoch: [94][ 760/ 1207] Overall Loss 0.326671 Objective Loss 0.326671 LR 0.001000 Time 0.019821 -2023-02-13 18:00:02,561 - Epoch: [94][ 770/ 1207] Overall Loss 0.326595 Objective Loss 0.326595 LR 0.001000 Time 0.019810 -2023-02-13 18:00:02,750 - Epoch: [94][ 780/ 1207] Overall Loss 0.326759 Objective Loss 0.326759 LR 0.001000 Time 0.019798 -2023-02-13 18:00:02,940 - Epoch: [94][ 790/ 1207] Overall Loss 0.327038 Objective Loss 0.327038 LR 0.001000 Time 0.019787 -2023-02-13 18:00:03,128 - Epoch: [94][ 800/ 1207] Overall Loss 0.327274 Objective Loss 0.327274 LR 0.001000 Time 0.019775 -2023-02-13 18:00:03,318 - Epoch: [94][ 810/ 1207] Overall Loss 0.327479 Objective Loss 0.327479 LR 0.001000 Time 0.019765 -2023-02-13 18:00:03,507 - Epoch: [94][ 820/ 1207] Overall Loss 0.327476 Objective Loss 0.327476 LR 0.001000 Time 0.019753 -2023-02-13 18:00:03,697 - Epoch: [94][ 830/ 1207] Overall Loss 0.327239 Objective Loss 0.327239 LR 0.001000 Time 0.019744 -2023-02-13 18:00:03,886 - Epoch: [94][ 840/ 1207] Overall Loss 0.327411 Objective Loss 0.327411 LR 0.001000 Time 0.019734 -2023-02-13 18:00:04,076 - Epoch: [94][ 850/ 1207] Overall Loss 0.327422 Objective Loss 0.327422 LR 0.001000 Time 0.019724 -2023-02-13 18:00:04,266 - Epoch: [94][ 860/ 1207] Overall Loss 0.327675 Objective Loss 0.327675 LR 0.001000 Time 0.019715 -2023-02-13 18:00:04,455 - Epoch: [94][ 870/ 1207] Overall Loss 0.327566 Objective Loss 0.327566 LR 0.001000 Time 0.019706 -2023-02-13 18:00:04,644 - Epoch: [94][ 880/ 1207] Overall Loss 0.327612 Objective Loss 0.327612 LR 0.001000 Time 0.019696 -2023-02-13 18:00:04,834 - Epoch: [94][ 890/ 1207] Overall Loss 0.327638 Objective Loss 0.327638 LR 0.001000 Time 0.019688 -2023-02-13 18:00:05,023 - Epoch: [94][ 900/ 1207] Overall Loss 0.328061 Objective Loss 0.328061 LR 0.001000 Time 0.019679 -2023-02-13 18:00:05,213 - Epoch: [94][ 910/ 1207] Overall Loss 0.328493 Objective Loss 0.328493 LR 0.001000 Time 0.019671 -2023-02-13 18:00:05,403 - Epoch: [94][ 920/ 1207] Overall Loss 0.328544 Objective Loss 0.328544 LR 0.001000 Time 0.019663 -2023-02-13 18:00:05,594 - Epoch: [94][ 930/ 1207] Overall Loss 0.328849 Objective Loss 0.328849 LR 0.001000 Time 0.019656 -2023-02-13 18:00:05,783 - Epoch: [94][ 940/ 1207] Overall Loss 0.328924 Objective Loss 0.328924 LR 0.001000 Time 0.019649 -2023-02-13 18:00:05,977 - Epoch: [94][ 950/ 1207] Overall Loss 0.329291 Objective Loss 0.329291 LR 0.001000 Time 0.019645 -2023-02-13 18:00:06,166 - Epoch: [94][ 960/ 1207] Overall Loss 0.329357 Objective Loss 0.329357 LR 0.001000 Time 0.019637 -2023-02-13 18:00:06,356 - Epoch: [94][ 970/ 1207] Overall Loss 0.329583 Objective Loss 0.329583 LR 0.001000 Time 0.019630 -2023-02-13 18:00:06,545 - Epoch: [94][ 980/ 1207] Overall Loss 0.329773 Objective Loss 0.329773 LR 0.001000 Time 0.019622 -2023-02-13 18:00:06,735 - Epoch: [94][ 990/ 1207] Overall Loss 0.329875 Objective Loss 0.329875 LR 0.001000 Time 0.019616 -2023-02-13 18:00:06,925 - Epoch: [94][ 1000/ 1207] Overall Loss 0.329783 Objective Loss 0.329783 LR 0.001000 Time 0.019609 -2023-02-13 18:00:07,115 - Epoch: [94][ 1010/ 1207] Overall Loss 0.330006 Objective Loss 0.330006 LR 0.001000 Time 0.019603 -2023-02-13 18:00:07,305 - Epoch: [94][ 1020/ 1207] Overall Loss 0.329935 Objective Loss 0.329935 LR 0.001000 Time 0.019596 -2023-02-13 18:00:07,495 - Epoch: [94][ 1030/ 1207] Overall Loss 0.330025 Objective Loss 0.330025 LR 0.001000 Time 0.019590 -2023-02-13 18:00:07,685 - Epoch: [94][ 1040/ 1207] Overall Loss 0.330063 Objective Loss 0.330063 LR 0.001000 Time 0.019584 -2023-02-13 18:00:07,875 - Epoch: [94][ 1050/ 1207] Overall Loss 0.330299 Objective Loss 0.330299 LR 0.001000 Time 0.019578 -2023-02-13 18:00:08,064 - Epoch: [94][ 1060/ 1207] Overall Loss 0.330533 Objective Loss 0.330533 LR 0.001000 Time 0.019571 -2023-02-13 18:00:08,255 - Epoch: [94][ 1070/ 1207] Overall Loss 0.330400 Objective Loss 0.330400 LR 0.001000 Time 0.019566 -2023-02-13 18:00:08,445 - Epoch: [94][ 1080/ 1207] Overall Loss 0.330426 Objective Loss 0.330426 LR 0.001000 Time 0.019561 -2023-02-13 18:00:08,635 - Epoch: [94][ 1090/ 1207] Overall Loss 0.330386 Objective Loss 0.330386 LR 0.001000 Time 0.019555 -2023-02-13 18:00:08,825 - Epoch: [94][ 1100/ 1207] Overall Loss 0.330348 Objective Loss 0.330348 LR 0.001000 Time 0.019550 -2023-02-13 18:00:09,014 - Epoch: [94][ 1110/ 1207] Overall Loss 0.330158 Objective Loss 0.330158 LR 0.001000 Time 0.019544 -2023-02-13 18:00:09,203 - Epoch: [94][ 1120/ 1207] Overall Loss 0.330320 Objective Loss 0.330320 LR 0.001000 Time 0.019538 -2023-02-13 18:00:09,393 - Epoch: [94][ 1130/ 1207] Overall Loss 0.330427 Objective Loss 0.330427 LR 0.001000 Time 0.019533 -2023-02-13 18:00:09,583 - Epoch: [94][ 1140/ 1207] Overall Loss 0.330385 Objective Loss 0.330385 LR 0.001000 Time 0.019528 -2023-02-13 18:00:09,774 - Epoch: [94][ 1150/ 1207] Overall Loss 0.330723 Objective Loss 0.330723 LR 0.001000 Time 0.019523 -2023-02-13 18:00:09,963 - Epoch: [94][ 1160/ 1207] Overall Loss 0.330974 Objective Loss 0.330974 LR 0.001000 Time 0.019518 -2023-02-13 18:00:10,153 - Epoch: [94][ 1170/ 1207] Overall Loss 0.331097 Objective Loss 0.331097 LR 0.001000 Time 0.019513 -2023-02-13 18:00:10,343 - Epoch: [94][ 1180/ 1207] Overall Loss 0.330903 Objective Loss 0.330903 LR 0.001000 Time 0.019508 -2023-02-13 18:00:10,533 - Epoch: [94][ 1190/ 1207] Overall Loss 0.330918 Objective Loss 0.330918 LR 0.001000 Time 0.019504 -2023-02-13 18:00:10,775 - Epoch: [94][ 1200/ 1207] Overall Loss 0.331119 Objective Loss 0.331119 LR 0.001000 Time 0.019543 -2023-02-13 18:00:10,892 - Epoch: [94][ 1207/ 1207] Overall Loss 0.330956 Objective Loss 0.330956 Top1 83.841463 Top5 97.256098 LR 0.001000 Time 0.019526 -2023-02-13 18:00:10,964 - --- validate (epoch=94)----------- -2023-02-13 18:00:10,965 - 34311 samples (256 per mini-batch) -2023-02-13 18:00:11,357 - Epoch: [94][ 10/ 135] Loss 0.384181 Top1 80.781250 Top5 96.796875 -2023-02-13 18:00:11,483 - Epoch: [94][ 20/ 135] Loss 0.381206 Top1 81.210938 Top5 97.031250 -2023-02-13 18:00:11,615 - Epoch: [94][ 30/ 135] Loss 0.369316 Top1 81.419271 Top5 97.135417 -2023-02-13 18:00:11,744 - Epoch: [94][ 40/ 135] Loss 0.365809 Top1 81.435547 Top5 97.285156 -2023-02-13 18:00:11,872 - Epoch: [94][ 50/ 135] Loss 0.365245 Top1 81.281250 Top5 97.210938 -2023-02-13 18:00:12,002 - Epoch: [94][ 60/ 135] Loss 0.369433 Top1 81.256510 Top5 97.180990 -2023-02-13 18:00:12,134 - Epoch: [94][ 70/ 135] Loss 0.369267 Top1 81.143973 Top5 97.148438 -2023-02-13 18:00:12,262 - Epoch: [94][ 80/ 135] Loss 0.368582 Top1 81.196289 Top5 97.148438 -2023-02-13 18:00:12,392 - Epoch: [94][ 90/ 135] Loss 0.371419 Top1 81.063368 Top5 97.139757 -2023-02-13 18:00:12,520 - Epoch: [94][ 100/ 135] Loss 0.375341 Top1 80.984375 Top5 97.144531 -2023-02-13 18:00:12,651 - Epoch: [94][ 110/ 135] Loss 0.376208 Top1 80.951705 Top5 97.180398 -2023-02-13 18:00:12,778 - Epoch: [94][ 120/ 135] Loss 0.377640 Top1 80.875651 Top5 97.138672 -2023-02-13 18:00:12,905 - Epoch: [94][ 130/ 135] Loss 0.374948 Top1 80.937500 Top5 97.163462 -2023-02-13 18:00:12,949 - Epoch: [94][ 135/ 135] Loss 0.373839 Top1 80.915741 Top5 97.161260 -2023-02-13 18:00:13,021 - ==> Top1: 80.916 Top5: 97.161 Loss: 0.374 - -2023-02-13 18:00:13,021 - ==> Confusion: -[[ 851 5 9 2 10 2 0 0 3 49 1 5 2 9 3 3 2 3 1 1 6] - [ 1 938 0 1 5 28 4 16 3 1 1 5 3 2 0 3 1 0 9 4 8] - [ 8 7 922 13 4 4 28 14 0 2 4 2 1 6 3 9 2 8 11 4 6] - [ 4 1 28 872 3 4 2 2 3 2 14 0 7 5 28 1 2 11 19 2 6] - [ 24 14 3 0 967 5 1 2 2 6 0 8 3 8 5 5 3 4 0 3 3] - [ 0 27 0 3 4 960 6 14 3 5 5 10 2 16 2 2 1 3 1 3 3] - [ 3 5 13 0 1 5 1035 5 3 1 4 4 1 3 0 3 0 4 0 5 4] - [ 2 10 10 2 0 47 8 894 1 1 4 8 4 1 1 0 0 2 20 8 1] - [ 20 3 0 0 0 1 0 2 888 52 5 3 1 6 14 1 1 3 9 0 0] - [ 90 1 3 0 1 1 0 0 37 846 1 0 0 14 4 2 1 5 1 2 3] - [ 3 3 1 6 1 1 4 3 25 2 968 2 1 10 1 0 1 1 16 0 2] - [ 0 2 0 0 5 12 2 3 1 2 1 903 32 7 0 2 2 15 2 11 3] - [ 1 1 0 4 2 3 1 1 3 0 2 36 860 3 3 2 1 27 2 1 6] - [ 3 1 0 0 9 10 0 3 20 20 10 8 2 919 1 6 2 3 0 2 5] - [ 16 2 4 9 6 2 0 2 27 8 3 2 6 4 964 1 0 10 18 1 7] - [ 4 2 6 0 6 2 4 0 1 0 0 14 10 3 0 941 9 32 0 6 6] - [ 3 14 1 2 15 5 1 1 2 4 0 3 4 3 2 7 965 1 2 12 14] - [ 5 4 2 2 0 1 2 1 1 1 1 10 18 0 1 8 0 990 0 1 3] - [ 4 2 6 8 2 0 1 24 3 0 8 2 7 0 12 1 1 2 998 4 1] - [ 0 4 0 1 1 5 7 16 1 0 0 21 4 3 0 4 3 7 0 1064 7] - [ 205 359 238 129 149 275 121 200 140 102 241 160 385 408 168 142 193 205 265 331 9018]] - -2023-02-13 18:00:13,023 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:00:13,023 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:00:13,029 - - -2023-02-13 18:00:13,029 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:00:13,990 - Epoch: [95][ 10/ 1207] Overall Loss 0.349087 Objective Loss 0.349087 LR 0.001000 Time 0.096034 -2023-02-13 18:00:14,181 - Epoch: [95][ 20/ 1207] Overall Loss 0.342188 Objective Loss 0.342188 LR 0.001000 Time 0.057576 -2023-02-13 18:00:14,370 - Epoch: [95][ 30/ 1207] Overall Loss 0.340972 Objective Loss 0.340972 LR 0.001000 Time 0.044647 -2023-02-13 18:00:14,557 - Epoch: [95][ 40/ 1207] Overall Loss 0.338211 Objective Loss 0.338211 LR 0.001000 Time 0.038159 -2023-02-13 18:00:14,745 - Epoch: [95][ 50/ 1207] Overall Loss 0.337867 Objective Loss 0.337867 LR 0.001000 Time 0.034280 -2023-02-13 18:00:14,933 - Epoch: [95][ 60/ 1207] Overall Loss 0.333401 Objective Loss 0.333401 LR 0.001000 Time 0.031698 -2023-02-13 18:00:15,120 - Epoch: [95][ 70/ 1207] Overall Loss 0.329968 Objective Loss 0.329968 LR 0.001000 Time 0.029834 -2023-02-13 18:00:15,309 - Epoch: [95][ 80/ 1207] Overall Loss 0.331655 Objective Loss 0.331655 LR 0.001000 Time 0.028463 -2023-02-13 18:00:15,497 - Epoch: [95][ 90/ 1207] Overall Loss 0.330638 Objective Loss 0.330638 LR 0.001000 Time 0.027382 -2023-02-13 18:00:15,685 - Epoch: [95][ 100/ 1207] Overall Loss 0.329447 Objective Loss 0.329447 LR 0.001000 Time 0.026516 -2023-02-13 18:00:15,873 - Epoch: [95][ 110/ 1207] Overall Loss 0.327578 Objective Loss 0.327578 LR 0.001000 Time 0.025819 -2023-02-13 18:00:16,061 - Epoch: [95][ 120/ 1207] Overall Loss 0.326218 Objective Loss 0.326218 LR 0.001000 Time 0.025230 -2023-02-13 18:00:16,248 - Epoch: [95][ 130/ 1207] Overall Loss 0.323798 Objective Loss 0.323798 LR 0.001000 Time 0.024726 -2023-02-13 18:00:16,437 - Epoch: [95][ 140/ 1207] Overall Loss 0.325149 Objective Loss 0.325149 LR 0.001000 Time 0.024303 -2023-02-13 18:00:16,624 - Epoch: [95][ 150/ 1207] Overall Loss 0.323426 Objective Loss 0.323426 LR 0.001000 Time 0.023930 -2023-02-13 18:00:16,814 - Epoch: [95][ 160/ 1207] Overall Loss 0.324172 Objective Loss 0.324172 LR 0.001000 Time 0.023616 -2023-02-13 18:00:17,002 - Epoch: [95][ 170/ 1207] Overall Loss 0.322713 Objective Loss 0.322713 LR 0.001000 Time 0.023330 -2023-02-13 18:00:17,189 - Epoch: [95][ 180/ 1207] Overall Loss 0.322931 Objective Loss 0.322931 LR 0.001000 Time 0.023074 -2023-02-13 18:00:17,377 - Epoch: [95][ 190/ 1207] Overall Loss 0.320878 Objective Loss 0.320878 LR 0.001000 Time 0.022847 -2023-02-13 18:00:17,565 - Epoch: [95][ 200/ 1207] Overall Loss 0.322233 Objective Loss 0.322233 LR 0.001000 Time 0.022642 -2023-02-13 18:00:17,753 - Epoch: [95][ 210/ 1207] Overall Loss 0.323661 Objective Loss 0.323661 LR 0.001000 Time 0.022457 -2023-02-13 18:00:17,942 - Epoch: [95][ 220/ 1207] Overall Loss 0.323177 Objective Loss 0.323177 LR 0.001000 Time 0.022293 -2023-02-13 18:00:18,131 - Epoch: [95][ 230/ 1207] Overall Loss 0.323416 Objective Loss 0.323416 LR 0.001000 Time 0.022142 -2023-02-13 18:00:18,319 - Epoch: [95][ 240/ 1207] Overall Loss 0.322507 Objective Loss 0.322507 LR 0.001000 Time 0.022002 -2023-02-13 18:00:18,508 - Epoch: [95][ 250/ 1207] Overall Loss 0.322756 Objective Loss 0.322756 LR 0.001000 Time 0.021878 -2023-02-13 18:00:18,697 - Epoch: [95][ 260/ 1207] Overall Loss 0.322136 Objective Loss 0.322136 LR 0.001000 Time 0.021763 -2023-02-13 18:00:18,886 - Epoch: [95][ 270/ 1207] Overall Loss 0.323052 Objective Loss 0.323052 LR 0.001000 Time 0.021655 -2023-02-13 18:00:19,074 - Epoch: [95][ 280/ 1207] Overall Loss 0.321831 Objective Loss 0.321831 LR 0.001000 Time 0.021551 -2023-02-13 18:00:19,262 - Epoch: [95][ 290/ 1207] Overall Loss 0.321993 Objective Loss 0.321993 LR 0.001000 Time 0.021455 -2023-02-13 18:00:19,450 - Epoch: [95][ 300/ 1207] Overall Loss 0.323096 Objective Loss 0.323096 LR 0.001000 Time 0.021367 -2023-02-13 18:00:19,638 - Epoch: [95][ 310/ 1207] Overall Loss 0.322167 Objective Loss 0.322167 LR 0.001000 Time 0.021282 -2023-02-13 18:00:19,827 - Epoch: [95][ 320/ 1207] Overall Loss 0.321892 Objective Loss 0.321892 LR 0.001000 Time 0.021207 -2023-02-13 18:00:20,015 - Epoch: [95][ 330/ 1207] Overall Loss 0.322466 Objective Loss 0.322466 LR 0.001000 Time 0.021133 -2023-02-13 18:00:20,204 - Epoch: [95][ 340/ 1207] Overall Loss 0.322480 Objective Loss 0.322480 LR 0.001000 Time 0.021066 -2023-02-13 18:00:20,393 - Epoch: [95][ 350/ 1207] Overall Loss 0.322854 Objective Loss 0.322854 LR 0.001000 Time 0.021001 -2023-02-13 18:00:20,581 - Epoch: [95][ 360/ 1207] Overall Loss 0.323345 Objective Loss 0.323345 LR 0.001000 Time 0.020941 -2023-02-13 18:00:20,769 - Epoch: [95][ 370/ 1207] Overall Loss 0.323009 Objective Loss 0.323009 LR 0.001000 Time 0.020882 -2023-02-13 18:00:20,959 - Epoch: [95][ 380/ 1207] Overall Loss 0.322931 Objective Loss 0.322931 LR 0.001000 Time 0.020830 -2023-02-13 18:00:21,147 - Epoch: [95][ 390/ 1207] Overall Loss 0.323459 Objective Loss 0.323459 LR 0.001000 Time 0.020777 -2023-02-13 18:00:21,335 - Epoch: [95][ 400/ 1207] Overall Loss 0.323547 Objective Loss 0.323547 LR 0.001000 Time 0.020728 -2023-02-13 18:00:21,524 - Epoch: [95][ 410/ 1207] Overall Loss 0.324104 Objective Loss 0.324104 LR 0.001000 Time 0.020681 -2023-02-13 18:00:21,713 - Epoch: [95][ 420/ 1207] Overall Loss 0.323645 Objective Loss 0.323645 LR 0.001000 Time 0.020638 -2023-02-13 18:00:21,903 - Epoch: [95][ 430/ 1207] Overall Loss 0.324654 Objective Loss 0.324654 LR 0.001000 Time 0.020600 -2023-02-13 18:00:22,093 - Epoch: [95][ 440/ 1207] Overall Loss 0.325358 Objective Loss 0.325358 LR 0.001000 Time 0.020561 -2023-02-13 18:00:22,281 - Epoch: [95][ 450/ 1207] Overall Loss 0.326552 Objective Loss 0.326552 LR 0.001000 Time 0.020522 -2023-02-13 18:00:22,470 - Epoch: [95][ 460/ 1207] Overall Loss 0.326481 Objective Loss 0.326481 LR 0.001000 Time 0.020486 -2023-02-13 18:00:22,659 - Epoch: [95][ 470/ 1207] Overall Loss 0.326447 Objective Loss 0.326447 LR 0.001000 Time 0.020452 -2023-02-13 18:00:22,849 - Epoch: [95][ 480/ 1207] Overall Loss 0.327091 Objective Loss 0.327091 LR 0.001000 Time 0.020421 -2023-02-13 18:00:23,038 - Epoch: [95][ 490/ 1207] Overall Loss 0.327171 Objective Loss 0.327171 LR 0.001000 Time 0.020389 -2023-02-13 18:00:23,227 - Epoch: [95][ 500/ 1207] Overall Loss 0.327086 Objective Loss 0.327086 LR 0.001000 Time 0.020358 -2023-02-13 18:00:23,415 - Epoch: [95][ 510/ 1207] Overall Loss 0.326862 Objective Loss 0.326862 LR 0.001000 Time 0.020327 -2023-02-13 18:00:23,604 - Epoch: [95][ 520/ 1207] Overall Loss 0.327615 Objective Loss 0.327615 LR 0.001000 Time 0.020299 -2023-02-13 18:00:23,792 - Epoch: [95][ 530/ 1207] Overall Loss 0.327246 Objective Loss 0.327246 LR 0.001000 Time 0.020270 -2023-02-13 18:00:23,981 - Epoch: [95][ 540/ 1207] Overall Loss 0.327102 Objective Loss 0.327102 LR 0.001000 Time 0.020245 -2023-02-13 18:00:24,170 - Epoch: [95][ 550/ 1207] Overall Loss 0.326846 Objective Loss 0.326846 LR 0.001000 Time 0.020219 -2023-02-13 18:00:24,358 - Epoch: [95][ 560/ 1207] Overall Loss 0.327010 Objective Loss 0.327010 LR 0.001000 Time 0.020193 -2023-02-13 18:00:24,546 - Epoch: [95][ 570/ 1207] Overall Loss 0.327263 Objective Loss 0.327263 LR 0.001000 Time 0.020168 -2023-02-13 18:00:24,735 - Epoch: [95][ 580/ 1207] Overall Loss 0.327642 Objective Loss 0.327642 LR 0.001000 Time 0.020145 -2023-02-13 18:00:24,923 - Epoch: [95][ 590/ 1207] Overall Loss 0.328588 Objective Loss 0.328588 LR 0.001000 Time 0.020122 -2023-02-13 18:00:25,113 - Epoch: [95][ 600/ 1207] Overall Loss 0.328611 Objective Loss 0.328611 LR 0.001000 Time 0.020102 -2023-02-13 18:00:25,301 - Epoch: [95][ 610/ 1207] Overall Loss 0.329024 Objective Loss 0.329024 LR 0.001000 Time 0.020080 -2023-02-13 18:00:25,490 - Epoch: [95][ 620/ 1207] Overall Loss 0.328706 Objective Loss 0.328706 LR 0.001000 Time 0.020060 -2023-02-13 18:00:25,679 - Epoch: [95][ 630/ 1207] Overall Loss 0.328525 Objective Loss 0.328525 LR 0.001000 Time 0.020041 -2023-02-13 18:00:25,870 - Epoch: [95][ 640/ 1207] Overall Loss 0.328437 Objective Loss 0.328437 LR 0.001000 Time 0.020026 -2023-02-13 18:00:26,059 - Epoch: [95][ 650/ 1207] Overall Loss 0.328669 Objective Loss 0.328669 LR 0.001000 Time 0.020008 -2023-02-13 18:00:26,248 - Epoch: [95][ 660/ 1207] Overall Loss 0.328663 Objective Loss 0.328663 LR 0.001000 Time 0.019991 -2023-02-13 18:00:26,436 - Epoch: [95][ 670/ 1207] Overall Loss 0.329327 Objective Loss 0.329327 LR 0.001000 Time 0.019973 -2023-02-13 18:00:26,626 - Epoch: [95][ 680/ 1207] Overall Loss 0.329433 Objective Loss 0.329433 LR 0.001000 Time 0.019958 -2023-02-13 18:00:26,815 - Epoch: [95][ 690/ 1207] Overall Loss 0.329553 Objective Loss 0.329553 LR 0.001000 Time 0.019943 -2023-02-13 18:00:27,005 - Epoch: [95][ 700/ 1207] Overall Loss 0.329080 Objective Loss 0.329080 LR 0.001000 Time 0.019928 -2023-02-13 18:00:27,193 - Epoch: [95][ 710/ 1207] Overall Loss 0.329469 Objective Loss 0.329469 LR 0.001000 Time 0.019911 -2023-02-13 18:00:27,382 - Epoch: [95][ 720/ 1207] Overall Loss 0.329979 Objective Loss 0.329979 LR 0.001000 Time 0.019897 -2023-02-13 18:00:27,571 - Epoch: [95][ 730/ 1207] Overall Loss 0.330158 Objective Loss 0.330158 LR 0.001000 Time 0.019882 -2023-02-13 18:00:27,760 - Epoch: [95][ 740/ 1207] Overall Loss 0.330083 Objective Loss 0.330083 LR 0.001000 Time 0.019870 -2023-02-13 18:00:27,949 - Epoch: [95][ 750/ 1207] Overall Loss 0.330304 Objective Loss 0.330304 LR 0.001000 Time 0.019855 -2023-02-13 18:00:28,138 - Epoch: [95][ 760/ 1207] Overall Loss 0.330877 Objective Loss 0.330877 LR 0.001000 Time 0.019843 -2023-02-13 18:00:28,327 - Epoch: [95][ 770/ 1207] Overall Loss 0.330952 Objective Loss 0.330952 LR 0.001000 Time 0.019830 -2023-02-13 18:00:28,516 - Epoch: [95][ 780/ 1207] Overall Loss 0.330773 Objective Loss 0.330773 LR 0.001000 Time 0.019817 -2023-02-13 18:00:28,704 - Epoch: [95][ 790/ 1207] Overall Loss 0.331330 Objective Loss 0.331330 LR 0.001000 Time 0.019804 -2023-02-13 18:00:28,894 - Epoch: [95][ 800/ 1207] Overall Loss 0.331457 Objective Loss 0.331457 LR 0.001000 Time 0.019794 -2023-02-13 18:00:29,082 - Epoch: [95][ 810/ 1207] Overall Loss 0.331363 Objective Loss 0.331363 LR 0.001000 Time 0.019781 -2023-02-13 18:00:29,271 - Epoch: [95][ 820/ 1207] Overall Loss 0.331433 Objective Loss 0.331433 LR 0.001000 Time 0.019769 -2023-02-13 18:00:29,459 - Epoch: [95][ 830/ 1207] Overall Loss 0.331393 Objective Loss 0.331393 LR 0.001000 Time 0.019758 -2023-02-13 18:00:29,648 - Epoch: [95][ 840/ 1207] Overall Loss 0.331203 Objective Loss 0.331203 LR 0.001000 Time 0.019747 -2023-02-13 18:00:29,837 - Epoch: [95][ 850/ 1207] Overall Loss 0.331082 Objective Loss 0.331082 LR 0.001000 Time 0.019737 -2023-02-13 18:00:30,026 - Epoch: [95][ 860/ 1207] Overall Loss 0.331058 Objective Loss 0.331058 LR 0.001000 Time 0.019726 -2023-02-13 18:00:30,214 - Epoch: [95][ 870/ 1207] Overall Loss 0.331112 Objective Loss 0.331112 LR 0.001000 Time 0.019715 -2023-02-13 18:00:30,403 - Epoch: [95][ 880/ 1207] Overall Loss 0.331183 Objective Loss 0.331183 LR 0.001000 Time 0.019706 -2023-02-13 18:00:30,592 - Epoch: [95][ 890/ 1207] Overall Loss 0.331289 Objective Loss 0.331289 LR 0.001000 Time 0.019696 -2023-02-13 18:00:30,782 - Epoch: [95][ 900/ 1207] Overall Loss 0.331397 Objective Loss 0.331397 LR 0.001000 Time 0.019688 -2023-02-13 18:00:30,973 - Epoch: [95][ 910/ 1207] Overall Loss 0.331600 Objective Loss 0.331600 LR 0.001000 Time 0.019681 -2023-02-13 18:00:31,163 - Epoch: [95][ 920/ 1207] Overall Loss 0.331829 Objective Loss 0.331829 LR 0.001000 Time 0.019673 -2023-02-13 18:00:31,352 - Epoch: [95][ 930/ 1207] Overall Loss 0.332038 Objective Loss 0.332038 LR 0.001000 Time 0.019665 -2023-02-13 18:00:31,541 - Epoch: [95][ 940/ 1207] Overall Loss 0.331671 Objective Loss 0.331671 LR 0.001000 Time 0.019656 -2023-02-13 18:00:31,731 - Epoch: [95][ 950/ 1207] Overall Loss 0.331476 Objective Loss 0.331476 LR 0.001000 Time 0.019649 -2023-02-13 18:00:31,922 - Epoch: [95][ 960/ 1207] Overall Loss 0.331660 Objective Loss 0.331660 LR 0.001000 Time 0.019642 -2023-02-13 18:00:32,111 - Epoch: [95][ 970/ 1207] Overall Loss 0.331552 Objective Loss 0.331552 LR 0.001000 Time 0.019635 -2023-02-13 18:00:32,301 - Epoch: [95][ 980/ 1207] Overall Loss 0.331524 Objective Loss 0.331524 LR 0.001000 Time 0.019627 -2023-02-13 18:00:32,490 - Epoch: [95][ 990/ 1207] Overall Loss 0.331659 Objective Loss 0.331659 LR 0.001000 Time 0.019620 -2023-02-13 18:00:32,679 - Epoch: [95][ 1000/ 1207] Overall Loss 0.331292 Objective Loss 0.331292 LR 0.001000 Time 0.019613 -2023-02-13 18:00:32,870 - Epoch: [95][ 1010/ 1207] Overall Loss 0.331367 Objective Loss 0.331367 LR 0.001000 Time 0.019607 -2023-02-13 18:00:33,059 - Epoch: [95][ 1020/ 1207] Overall Loss 0.331434 Objective Loss 0.331434 LR 0.001000 Time 0.019600 -2023-02-13 18:00:33,249 - Epoch: [95][ 1030/ 1207] Overall Loss 0.331614 Objective Loss 0.331614 LR 0.001000 Time 0.019593 -2023-02-13 18:00:33,438 - Epoch: [95][ 1040/ 1207] Overall Loss 0.331807 Objective Loss 0.331807 LR 0.001000 Time 0.019587 -2023-02-13 18:00:33,628 - Epoch: [95][ 1050/ 1207] Overall Loss 0.331855 Objective Loss 0.331855 LR 0.001000 Time 0.019581 -2023-02-13 18:00:33,818 - Epoch: [95][ 1060/ 1207] Overall Loss 0.332161 Objective Loss 0.332161 LR 0.001000 Time 0.019575 -2023-02-13 18:00:34,008 - Epoch: [95][ 1070/ 1207] Overall Loss 0.332357 Objective Loss 0.332357 LR 0.001000 Time 0.019569 -2023-02-13 18:00:34,197 - Epoch: [95][ 1080/ 1207] Overall Loss 0.332329 Objective Loss 0.332329 LR 0.001000 Time 0.019563 -2023-02-13 18:00:34,387 - Epoch: [95][ 1090/ 1207] Overall Loss 0.332494 Objective Loss 0.332494 LR 0.001000 Time 0.019557 -2023-02-13 18:00:34,577 - Epoch: [95][ 1100/ 1207] Overall Loss 0.332906 Objective Loss 0.332906 LR 0.001000 Time 0.019551 -2023-02-13 18:00:34,767 - Epoch: [95][ 1110/ 1207] Overall Loss 0.333082 Objective Loss 0.333082 LR 0.001000 Time 0.019546 -2023-02-13 18:00:34,959 - Epoch: [95][ 1120/ 1207] Overall Loss 0.333305 Objective Loss 0.333305 LR 0.001000 Time 0.019543 -2023-02-13 18:00:35,150 - Epoch: [95][ 1130/ 1207] Overall Loss 0.333367 Objective Loss 0.333367 LR 0.001000 Time 0.019538 -2023-02-13 18:00:35,340 - Epoch: [95][ 1140/ 1207] Overall Loss 0.333394 Objective Loss 0.333394 LR 0.001000 Time 0.019534 -2023-02-13 18:00:35,531 - Epoch: [95][ 1150/ 1207] Overall Loss 0.333649 Objective Loss 0.333649 LR 0.001000 Time 0.019529 -2023-02-13 18:00:35,721 - Epoch: [95][ 1160/ 1207] Overall Loss 0.333816 Objective Loss 0.333816 LR 0.001000 Time 0.019524 -2023-02-13 18:00:35,915 - Epoch: [95][ 1170/ 1207] Overall Loss 0.333819 Objective Loss 0.333819 LR 0.001000 Time 0.019523 -2023-02-13 18:00:36,105 - Epoch: [95][ 1180/ 1207] Overall Loss 0.333949 Objective Loss 0.333949 LR 0.001000 Time 0.019519 -2023-02-13 18:00:36,296 - Epoch: [95][ 1190/ 1207] Overall Loss 0.333927 Objective Loss 0.333927 LR 0.001000 Time 0.019515 -2023-02-13 18:00:36,543 - Epoch: [95][ 1200/ 1207] Overall Loss 0.333846 Objective Loss 0.333846 LR 0.001000 Time 0.019558 -2023-02-13 18:00:36,659 - Epoch: [95][ 1207/ 1207] Overall Loss 0.333797 Objective Loss 0.333797 Top1 88.719512 Top5 98.475610 LR 0.001000 Time 0.019540 -2023-02-13 18:00:36,734 - --- validate (epoch=95)----------- -2023-02-13 18:00:36,734 - 34311 samples (256 per mini-batch) -2023-02-13 18:00:37,137 - Epoch: [95][ 10/ 135] Loss 0.365964 Top1 83.007812 Top5 97.656250 -2023-02-13 18:00:37,265 - Epoch: [95][ 20/ 135] Loss 0.379071 Top1 82.617188 Top5 97.636719 -2023-02-13 18:00:37,390 - Epoch: [95][ 30/ 135] Loss 0.361146 Top1 82.851562 Top5 97.630208 -2023-02-13 18:00:37,531 - Epoch: [95][ 40/ 135] Loss 0.356033 Top1 83.066406 Top5 97.626953 -2023-02-13 18:00:37,655 - Epoch: [95][ 50/ 135] Loss 0.358519 Top1 82.937500 Top5 97.570312 -2023-02-13 18:00:37,784 - Epoch: [95][ 60/ 135] Loss 0.357328 Top1 82.975260 Top5 97.558594 -2023-02-13 18:00:37,906 - Epoch: [95][ 70/ 135] Loss 0.356808 Top1 82.974330 Top5 97.544643 -2023-02-13 18:00:38,032 - Epoch: [95][ 80/ 135] Loss 0.353093 Top1 82.954102 Top5 97.583008 -2023-02-13 18:00:38,158 - Epoch: [95][ 90/ 135] Loss 0.354245 Top1 82.894965 Top5 97.530382 -2023-02-13 18:00:38,284 - Epoch: [95][ 100/ 135] Loss 0.354682 Top1 83.007812 Top5 97.527344 -2023-02-13 18:00:38,408 - Epoch: [95][ 110/ 135] Loss 0.355402 Top1 82.904830 Top5 97.510653 -2023-02-13 18:00:38,531 - Epoch: [95][ 120/ 135] Loss 0.357844 Top1 82.890625 Top5 97.444661 -2023-02-13 18:00:38,657 - Epoch: [95][ 130/ 135] Loss 0.356798 Top1 82.893630 Top5 97.448918 -2023-02-13 18:00:38,701 - Epoch: [95][ 135/ 135] Loss 0.356316 Top1 82.839323 Top5 97.411909 -2023-02-13 18:00:38,769 - ==> Top1: 82.839 Top5: 97.412 Loss: 0.356 - -2023-02-13 18:00:38,770 - ==> Confusion: -[[ 830 6 10 4 8 4 0 0 5 64 0 4 0 6 2 3 8 4 0 2 7] - [ 3 904 3 1 10 41 4 21 3 2 3 4 1 1 0 2 12 0 4 4 10] - [ 6 2 928 25 3 2 17 14 0 2 3 3 1 8 2 10 4 2 4 8 14] - [ 5 2 19 892 3 4 3 2 1 2 11 0 19 2 17 5 4 4 12 0 9] - [ 17 10 0 1 967 13 2 2 3 4 1 3 3 4 6 8 10 0 0 2 10] - [ 1 13 1 5 6 966 5 13 1 6 4 11 4 16 0 2 4 1 2 5 4] - [ 4 4 19 1 0 9 1018 7 0 0 2 1 4 1 0 6 3 4 1 10 5] - [ 3 11 11 1 0 46 5 896 1 1 2 6 4 1 0 1 1 2 8 20 4] - [ 20 2 0 1 1 0 0 1 881 51 6 1 1 8 17 3 3 4 7 0 2] - [ 69 2 6 0 4 4 1 1 43 852 1 1 0 11 2 3 1 4 1 1 5] - [ 1 1 3 6 1 1 7 5 19 3 958 1 7 14 3 0 1 2 9 1 8] - [ 1 3 2 0 3 9 1 3 0 2 0 888 44 7 0 8 6 10 2 13 3] - [ 1 0 0 3 1 4 0 2 3 0 1 26 874 0 0 11 3 16 0 0 14] - [ 4 3 2 1 9 13 1 2 16 24 10 7 0 901 3 6 7 6 1 4 4] - [ 14 5 1 19 7 3 0 2 20 8 2 2 8 2 969 2 5 4 6 0 13] - [ 3 1 5 1 5 1 2 0 0 2 0 2 10 1 1 974 10 17 0 3 8] - [ 2 3 0 2 4 1 0 1 2 1 1 2 3 1 2 15 1002 1 1 5 12] - [ 7 1 2 3 0 1 1 1 0 2 0 5 27 1 0 17 0 973 0 5 5] - [ 6 3 11 21 0 3 0 35 5 2 10 0 7 0 14 0 3 2 959 2 3] - [ 1 2 1 1 0 4 4 10 1 0 0 16 6 2 1 9 5 3 1 1069 12] - [ 158 220 233 126 108 253 71 160 104 116 199 126 319 312 145 137 369 109 142 305 9722]] - -2023-02-13 18:00:38,771 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:00:38,771 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:00:38,777 - - -2023-02-13 18:00:38,777 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:00:39,763 - Epoch: [96][ 10/ 1207] Overall Loss 0.323353 Objective Loss 0.323353 LR 0.001000 Time 0.098574 -2023-02-13 18:00:39,957 - Epoch: [96][ 20/ 1207] Overall Loss 0.318600 Objective Loss 0.318600 LR 0.001000 Time 0.058951 -2023-02-13 18:00:40,143 - Epoch: [96][ 30/ 1207] Overall Loss 0.315535 Objective Loss 0.315535 LR 0.001000 Time 0.045503 -2023-02-13 18:00:40,330 - Epoch: [96][ 40/ 1207] Overall Loss 0.327591 Objective Loss 0.327591 LR 0.001000 Time 0.038771 -2023-02-13 18:00:40,524 - Epoch: [96][ 50/ 1207] Overall Loss 0.326973 Objective Loss 0.326973 LR 0.001000 Time 0.034904 -2023-02-13 18:00:40,721 - Epoch: [96][ 60/ 1207] Overall Loss 0.326348 Objective Loss 0.326348 LR 0.001000 Time 0.032366 -2023-02-13 18:00:40,916 - Epoch: [96][ 70/ 1207] Overall Loss 0.326101 Objective Loss 0.326101 LR 0.001000 Time 0.030509 -2023-02-13 18:00:41,101 - Epoch: [96][ 80/ 1207] Overall Loss 0.326100 Objective Loss 0.326100 LR 0.001000 Time 0.029007 -2023-02-13 18:00:41,286 - Epoch: [96][ 90/ 1207] Overall Loss 0.326499 Objective Loss 0.326499 LR 0.001000 Time 0.027842 -2023-02-13 18:00:41,488 - Epoch: [96][ 100/ 1207] Overall Loss 0.326135 Objective Loss 0.326135 LR 0.001000 Time 0.027071 -2023-02-13 18:00:41,684 - Epoch: [96][ 110/ 1207] Overall Loss 0.327978 Objective Loss 0.327978 LR 0.001000 Time 0.026387 -2023-02-13 18:00:41,886 - Epoch: [96][ 120/ 1207] Overall Loss 0.327694 Objective Loss 0.327694 LR 0.001000 Time 0.025867 -2023-02-13 18:00:42,082 - Epoch: [96][ 130/ 1207] Overall Loss 0.329683 Objective Loss 0.329683 LR 0.001000 Time 0.025384 -2023-02-13 18:00:42,283 - Epoch: [96][ 140/ 1207] Overall Loss 0.331030 Objective Loss 0.331030 LR 0.001000 Time 0.025004 -2023-02-13 18:00:42,479 - Epoch: [96][ 150/ 1207] Overall Loss 0.330768 Objective Loss 0.330768 LR 0.001000 Time 0.024638 -2023-02-13 18:00:42,680 - Epoch: [96][ 160/ 1207] Overall Loss 0.331994 Objective Loss 0.331994 LR 0.001000 Time 0.024355 -2023-02-13 18:00:42,877 - Epoch: [96][ 170/ 1207] Overall Loss 0.331444 Objective Loss 0.331444 LR 0.001000 Time 0.024076 -2023-02-13 18:00:43,078 - Epoch: [96][ 180/ 1207] Overall Loss 0.329707 Objective Loss 0.329707 LR 0.001000 Time 0.023855 -2023-02-13 18:00:43,274 - Epoch: [96][ 190/ 1207] Overall Loss 0.329540 Objective Loss 0.329540 LR 0.001000 Time 0.023630 -2023-02-13 18:00:43,476 - Epoch: [96][ 200/ 1207] Overall Loss 0.330409 Objective Loss 0.330409 LR 0.001000 Time 0.023455 -2023-02-13 18:00:43,672 - Epoch: [96][ 210/ 1207] Overall Loss 0.329395 Objective Loss 0.329395 LR 0.001000 Time 0.023270 -2023-02-13 18:00:43,875 - Epoch: [96][ 220/ 1207] Overall Loss 0.329362 Objective Loss 0.329362 LR 0.001000 Time 0.023132 -2023-02-13 18:00:44,072 - Epoch: [96][ 230/ 1207] Overall Loss 0.330374 Objective Loss 0.330374 LR 0.001000 Time 0.022982 -2023-02-13 18:00:44,274 - Epoch: [96][ 240/ 1207] Overall Loss 0.330825 Objective Loss 0.330825 LR 0.001000 Time 0.022863 -2023-02-13 18:00:44,470 - Epoch: [96][ 250/ 1207] Overall Loss 0.330743 Objective Loss 0.330743 LR 0.001000 Time 0.022734 -2023-02-13 18:00:44,672 - Epoch: [96][ 260/ 1207] Overall Loss 0.331178 Objective Loss 0.331178 LR 0.001000 Time 0.022634 -2023-02-13 18:00:44,869 - Epoch: [96][ 270/ 1207] Overall Loss 0.330900 Objective Loss 0.330900 LR 0.001000 Time 0.022523 -2023-02-13 18:00:45,070 - Epoch: [96][ 280/ 1207] Overall Loss 0.329850 Objective Loss 0.329850 LR 0.001000 Time 0.022438 -2023-02-13 18:00:45,268 - Epoch: [96][ 290/ 1207] Overall Loss 0.329615 Objective Loss 0.329615 LR 0.001000 Time 0.022344 -2023-02-13 18:00:45,471 - Epoch: [96][ 300/ 1207] Overall Loss 0.330117 Objective Loss 0.330117 LR 0.001000 Time 0.022275 -2023-02-13 18:00:45,668 - Epoch: [96][ 310/ 1207] Overall Loss 0.330139 Objective Loss 0.330139 LR 0.001000 Time 0.022190 -2023-02-13 18:00:45,871 - Epoch: [96][ 320/ 1207] Overall Loss 0.329764 Objective Loss 0.329764 LR 0.001000 Time 0.022130 -2023-02-13 18:00:46,068 - Epoch: [96][ 330/ 1207] Overall Loss 0.329308 Objective Loss 0.329308 LR 0.001000 Time 0.022055 -2023-02-13 18:00:46,269 - Epoch: [96][ 340/ 1207] Overall Loss 0.330225 Objective Loss 0.330225 LR 0.001000 Time 0.021997 -2023-02-13 18:00:46,465 - Epoch: [96][ 350/ 1207] Overall Loss 0.330564 Objective Loss 0.330564 LR 0.001000 Time 0.021928 -2023-02-13 18:00:46,667 - Epoch: [96][ 360/ 1207] Overall Loss 0.330178 Objective Loss 0.330178 LR 0.001000 Time 0.021879 -2023-02-13 18:00:46,855 - Epoch: [96][ 370/ 1207] Overall Loss 0.329689 Objective Loss 0.329689 LR 0.001000 Time 0.021794 -2023-02-13 18:00:47,042 - Epoch: [96][ 380/ 1207] Overall Loss 0.329460 Objective Loss 0.329460 LR 0.001000 Time 0.021711 -2023-02-13 18:00:47,229 - Epoch: [96][ 390/ 1207] Overall Loss 0.330341 Objective Loss 0.330341 LR 0.001000 Time 0.021634 -2023-02-13 18:00:47,417 - Epoch: [96][ 400/ 1207] Overall Loss 0.330582 Objective Loss 0.330582 LR 0.001000 Time 0.021561 -2023-02-13 18:00:47,604 - Epoch: [96][ 410/ 1207] Overall Loss 0.329741 Objective Loss 0.329741 LR 0.001000 Time 0.021490 -2023-02-13 18:00:47,790 - Epoch: [96][ 420/ 1207] Overall Loss 0.329913 Objective Loss 0.329913 LR 0.001000 Time 0.021421 -2023-02-13 18:00:47,978 - Epoch: [96][ 430/ 1207] Overall Loss 0.329780 Objective Loss 0.329780 LR 0.001000 Time 0.021360 -2023-02-13 18:00:48,166 - Epoch: [96][ 440/ 1207] Overall Loss 0.329457 Objective Loss 0.329457 LR 0.001000 Time 0.021300 -2023-02-13 18:00:48,353 - Epoch: [96][ 450/ 1207] Overall Loss 0.329517 Objective Loss 0.329517 LR 0.001000 Time 0.021242 -2023-02-13 18:00:48,541 - Epoch: [96][ 460/ 1207] Overall Loss 0.330691 Objective Loss 0.330691 LR 0.001000 Time 0.021188 -2023-02-13 18:00:48,729 - Epoch: [96][ 470/ 1207] Overall Loss 0.329940 Objective Loss 0.329940 LR 0.001000 Time 0.021136 -2023-02-13 18:00:48,917 - Epoch: [96][ 480/ 1207] Overall Loss 0.329452 Objective Loss 0.329452 LR 0.001000 Time 0.021087 -2023-02-13 18:00:49,105 - Epoch: [96][ 490/ 1207] Overall Loss 0.329265 Objective Loss 0.329265 LR 0.001000 Time 0.021040 -2023-02-13 18:00:49,293 - Epoch: [96][ 500/ 1207] Overall Loss 0.329458 Objective Loss 0.329458 LR 0.001000 Time 0.020993 -2023-02-13 18:00:49,480 - Epoch: [96][ 510/ 1207] Overall Loss 0.329883 Objective Loss 0.329883 LR 0.001000 Time 0.020948 -2023-02-13 18:00:49,667 - Epoch: [96][ 520/ 1207] Overall Loss 0.330247 Objective Loss 0.330247 LR 0.001000 Time 0.020904 -2023-02-13 18:00:49,856 - Epoch: [96][ 530/ 1207] Overall Loss 0.330413 Objective Loss 0.330413 LR 0.001000 Time 0.020865 -2023-02-13 18:00:50,043 - Epoch: [96][ 540/ 1207] Overall Loss 0.330498 Objective Loss 0.330498 LR 0.001000 Time 0.020826 -2023-02-13 18:00:50,231 - Epoch: [96][ 550/ 1207] Overall Loss 0.330644 Objective Loss 0.330644 LR 0.001000 Time 0.020787 -2023-02-13 18:00:50,419 - Epoch: [96][ 560/ 1207] Overall Loss 0.330718 Objective Loss 0.330718 LR 0.001000 Time 0.020751 -2023-02-13 18:00:50,606 - Epoch: [96][ 570/ 1207] Overall Loss 0.330703 Objective Loss 0.330703 LR 0.001000 Time 0.020714 -2023-02-13 18:00:50,793 - Epoch: [96][ 580/ 1207] Overall Loss 0.330383 Objective Loss 0.330383 LR 0.001000 Time 0.020679 -2023-02-13 18:00:50,982 - Epoch: [96][ 590/ 1207] Overall Loss 0.329521 Objective Loss 0.329521 LR 0.001000 Time 0.020648 -2023-02-13 18:00:51,169 - Epoch: [96][ 600/ 1207] Overall Loss 0.329819 Objective Loss 0.329819 LR 0.001000 Time 0.020616 -2023-02-13 18:00:51,357 - Epoch: [96][ 610/ 1207] Overall Loss 0.329602 Objective Loss 0.329602 LR 0.001000 Time 0.020585 -2023-02-13 18:00:51,544 - Epoch: [96][ 620/ 1207] Overall Loss 0.329157 Objective Loss 0.329157 LR 0.001000 Time 0.020554 -2023-02-13 18:00:51,732 - Epoch: [96][ 630/ 1207] Overall Loss 0.328752 Objective Loss 0.328752 LR 0.001000 Time 0.020525 -2023-02-13 18:00:51,920 - Epoch: [96][ 640/ 1207] Overall Loss 0.328839 Objective Loss 0.328839 LR 0.001000 Time 0.020498 -2023-02-13 18:00:52,108 - Epoch: [96][ 650/ 1207] Overall Loss 0.328827 Objective Loss 0.328827 LR 0.001000 Time 0.020471 -2023-02-13 18:00:52,296 - Epoch: [96][ 660/ 1207] Overall Loss 0.328661 Objective Loss 0.328661 LR 0.001000 Time 0.020445 -2023-02-13 18:00:52,484 - Epoch: [96][ 670/ 1207] Overall Loss 0.328707 Objective Loss 0.328707 LR 0.001000 Time 0.020420 -2023-02-13 18:00:52,671 - Epoch: [96][ 680/ 1207] Overall Loss 0.329243 Objective Loss 0.329243 LR 0.001000 Time 0.020395 -2023-02-13 18:00:52,860 - Epoch: [96][ 690/ 1207] Overall Loss 0.329298 Objective Loss 0.329298 LR 0.001000 Time 0.020372 -2023-02-13 18:00:53,047 - Epoch: [96][ 700/ 1207] Overall Loss 0.328977 Objective Loss 0.328977 LR 0.001000 Time 0.020348 -2023-02-13 18:00:53,235 - Epoch: [96][ 710/ 1207] Overall Loss 0.328924 Objective Loss 0.328924 LR 0.001000 Time 0.020326 -2023-02-13 18:00:53,423 - Epoch: [96][ 720/ 1207] Overall Loss 0.328982 Objective Loss 0.328982 LR 0.001000 Time 0.020304 -2023-02-13 18:00:53,610 - Epoch: [96][ 730/ 1207] Overall Loss 0.329065 Objective Loss 0.329065 LR 0.001000 Time 0.020281 -2023-02-13 18:00:53,797 - Epoch: [96][ 740/ 1207] Overall Loss 0.329047 Objective Loss 0.329047 LR 0.001000 Time 0.020260 -2023-02-13 18:00:53,986 - Epoch: [96][ 750/ 1207] Overall Loss 0.329301 Objective Loss 0.329301 LR 0.001000 Time 0.020241 -2023-02-13 18:00:54,174 - Epoch: [96][ 760/ 1207] Overall Loss 0.329688 Objective Loss 0.329688 LR 0.001000 Time 0.020221 -2023-02-13 18:00:54,361 - Epoch: [96][ 770/ 1207] Overall Loss 0.329458 Objective Loss 0.329458 LR 0.001000 Time 0.020201 -2023-02-13 18:00:54,550 - Epoch: [96][ 780/ 1207] Overall Loss 0.329187 Objective Loss 0.329187 LR 0.001000 Time 0.020184 -2023-02-13 18:00:54,738 - Epoch: [96][ 790/ 1207] Overall Loss 0.329234 Objective Loss 0.329234 LR 0.001000 Time 0.020166 -2023-02-13 18:00:54,925 - Epoch: [96][ 800/ 1207] Overall Loss 0.329405 Objective Loss 0.329405 LR 0.001000 Time 0.020148 -2023-02-13 18:00:55,113 - Epoch: [96][ 810/ 1207] Overall Loss 0.329046 Objective Loss 0.329046 LR 0.001000 Time 0.020130 -2023-02-13 18:00:55,300 - Epoch: [96][ 820/ 1207] Overall Loss 0.329037 Objective Loss 0.329037 LR 0.001000 Time 0.020112 -2023-02-13 18:00:55,489 - Epoch: [96][ 830/ 1207] Overall Loss 0.328686 Objective Loss 0.328686 LR 0.001000 Time 0.020097 -2023-02-13 18:00:55,677 - Epoch: [96][ 840/ 1207] Overall Loss 0.328987 Objective Loss 0.328987 LR 0.001000 Time 0.020081 -2023-02-13 18:00:55,871 - Epoch: [96][ 850/ 1207] Overall Loss 0.329049 Objective Loss 0.329049 LR 0.001000 Time 0.020073 -2023-02-13 18:00:56,069 - Epoch: [96][ 860/ 1207] Overall Loss 0.329152 Objective Loss 0.329152 LR 0.001000 Time 0.020069 -2023-02-13 18:00:56,273 - Epoch: [96][ 870/ 1207] Overall Loss 0.329130 Objective Loss 0.329130 LR 0.001000 Time 0.020072 -2023-02-13 18:00:56,471 - Epoch: [96][ 880/ 1207] Overall Loss 0.328779 Objective Loss 0.328779 LR 0.001000 Time 0.020069 -2023-02-13 18:00:56,675 - Epoch: [96][ 890/ 1207] Overall Loss 0.328973 Objective Loss 0.328973 LR 0.001000 Time 0.020072 -2023-02-13 18:00:56,874 - Epoch: [96][ 900/ 1207] Overall Loss 0.329398 Objective Loss 0.329398 LR 0.001000 Time 0.020070 -2023-02-13 18:00:57,078 - Epoch: [96][ 910/ 1207] Overall Loss 0.329274 Objective Loss 0.329274 LR 0.001000 Time 0.020073 -2023-02-13 18:00:57,276 - Epoch: [96][ 920/ 1207] Overall Loss 0.329280 Objective Loss 0.329280 LR 0.001000 Time 0.020069 -2023-02-13 18:00:57,479 - Epoch: [96][ 930/ 1207] Overall Loss 0.329402 Objective Loss 0.329402 LR 0.001000 Time 0.020071 -2023-02-13 18:00:57,677 - Epoch: [96][ 940/ 1207] Overall Loss 0.329923 Objective Loss 0.329923 LR 0.001000 Time 0.020069 -2023-02-13 18:00:57,881 - Epoch: [96][ 950/ 1207] Overall Loss 0.329658 Objective Loss 0.329658 LR 0.001000 Time 0.020071 -2023-02-13 18:00:58,080 - Epoch: [96][ 960/ 1207] Overall Loss 0.329954 Objective Loss 0.329954 LR 0.001000 Time 0.020069 -2023-02-13 18:00:58,280 - Epoch: [96][ 970/ 1207] Overall Loss 0.330429 Objective Loss 0.330429 LR 0.001000 Time 0.020069 -2023-02-13 18:00:58,474 - Epoch: [96][ 980/ 1207] Overall Loss 0.330601 Objective Loss 0.330601 LR 0.001000 Time 0.020061 -2023-02-13 18:00:58,670 - Epoch: [96][ 990/ 1207] Overall Loss 0.330810 Objective Loss 0.330810 LR 0.001000 Time 0.020056 -2023-02-13 18:00:58,863 - Epoch: [96][ 1000/ 1207] Overall Loss 0.330747 Objective Loss 0.330747 LR 0.001000 Time 0.020048 -2023-02-13 18:00:59,059 - Epoch: [96][ 1010/ 1207] Overall Loss 0.331227 Objective Loss 0.331227 LR 0.001000 Time 0.020044 -2023-02-13 18:00:59,252 - Epoch: [96][ 1020/ 1207] Overall Loss 0.331309 Objective Loss 0.331309 LR 0.001000 Time 0.020036 -2023-02-13 18:00:59,447 - Epoch: [96][ 1030/ 1207] Overall Loss 0.331682 Objective Loss 0.331682 LR 0.001000 Time 0.020030 -2023-02-13 18:00:59,642 - Epoch: [96][ 1040/ 1207] Overall Loss 0.331702 Objective Loss 0.331702 LR 0.001000 Time 0.020025 -2023-02-13 18:00:59,838 - Epoch: [96][ 1050/ 1207] Overall Loss 0.331474 Objective Loss 0.331474 LR 0.001000 Time 0.020020 -2023-02-13 18:01:00,032 - Epoch: [96][ 1060/ 1207] Overall Loss 0.331128 Objective Loss 0.331128 LR 0.001000 Time 0.020014 -2023-02-13 18:01:00,224 - Epoch: [96][ 1070/ 1207] Overall Loss 0.331042 Objective Loss 0.331042 LR 0.001000 Time 0.020006 -2023-02-13 18:01:00,418 - Epoch: [96][ 1080/ 1207] Overall Loss 0.331174 Objective Loss 0.331174 LR 0.001000 Time 0.020000 -2023-02-13 18:01:00,610 - Epoch: [96][ 1090/ 1207] Overall Loss 0.331051 Objective Loss 0.331051 LR 0.001000 Time 0.019992 -2023-02-13 18:01:00,804 - Epoch: [96][ 1100/ 1207] Overall Loss 0.331364 Objective Loss 0.331364 LR 0.001000 Time 0.019986 -2023-02-13 18:01:00,997 - Epoch: [96][ 1110/ 1207] Overall Loss 0.331594 Objective Loss 0.331594 LR 0.001000 Time 0.019980 -2023-02-13 18:01:01,190 - Epoch: [96][ 1120/ 1207] Overall Loss 0.331711 Objective Loss 0.331711 LR 0.001000 Time 0.019974 -2023-02-13 18:01:01,383 - Epoch: [96][ 1130/ 1207] Overall Loss 0.331968 Objective Loss 0.331968 LR 0.001000 Time 0.019967 -2023-02-13 18:01:01,577 - Epoch: [96][ 1140/ 1207] Overall Loss 0.331893 Objective Loss 0.331893 LR 0.001000 Time 0.019962 -2023-02-13 18:01:01,769 - Epoch: [96][ 1150/ 1207] Overall Loss 0.331842 Objective Loss 0.331842 LR 0.001000 Time 0.019956 -2023-02-13 18:01:01,964 - Epoch: [96][ 1160/ 1207] Overall Loss 0.331916 Objective Loss 0.331916 LR 0.001000 Time 0.019951 -2023-02-13 18:01:02,156 - Epoch: [96][ 1170/ 1207] Overall Loss 0.332270 Objective Loss 0.332270 LR 0.001000 Time 0.019944 -2023-02-13 18:01:02,349 - Epoch: [96][ 1180/ 1207] Overall Loss 0.332452 Objective Loss 0.332452 LR 0.001000 Time 0.019938 -2023-02-13 18:01:02,541 - Epoch: [96][ 1190/ 1207] Overall Loss 0.332543 Objective Loss 0.332543 LR 0.001000 Time 0.019932 -2023-02-13 18:01:02,789 - Epoch: [96][ 1200/ 1207] Overall Loss 0.332340 Objective Loss 0.332340 LR 0.001000 Time 0.019972 -2023-02-13 18:01:02,904 - Epoch: [96][ 1207/ 1207] Overall Loss 0.332340 Objective Loss 0.332340 Top1 82.317073 Top5 96.951220 LR 0.001000 Time 0.019951 -2023-02-13 18:01:02,976 - --- validate (epoch=96)----------- -2023-02-13 18:01:02,977 - 34311 samples (256 per mini-batch) -2023-02-13 18:01:03,378 - Epoch: [96][ 10/ 135] Loss 0.395329 Top1 81.796875 Top5 97.226562 -2023-02-13 18:01:03,505 - Epoch: [96][ 20/ 135] Loss 0.379694 Top1 82.363281 Top5 97.109375 -2023-02-13 18:01:03,635 - Epoch: [96][ 30/ 135] Loss 0.384119 Top1 81.888021 Top5 96.914062 -2023-02-13 18:01:03,767 - Epoch: [96][ 40/ 135] Loss 0.376655 Top1 81.884766 Top5 96.992188 -2023-02-13 18:01:03,898 - Epoch: [96][ 50/ 135] Loss 0.374662 Top1 81.875000 Top5 96.968750 -2023-02-13 18:01:04,030 - Epoch: [96][ 60/ 135] Loss 0.369489 Top1 81.933594 Top5 96.894531 -2023-02-13 18:01:04,161 - Epoch: [96][ 70/ 135] Loss 0.366882 Top1 81.964286 Top5 96.886161 -2023-02-13 18:01:04,291 - Epoch: [96][ 80/ 135] Loss 0.366256 Top1 81.923828 Top5 96.933594 -2023-02-13 18:01:04,423 - Epoch: [96][ 90/ 135] Loss 0.369383 Top1 81.905382 Top5 96.931424 -2023-02-13 18:01:04,553 - Epoch: [96][ 100/ 135] Loss 0.364171 Top1 81.859375 Top5 96.968750 -2023-02-13 18:01:04,684 - Epoch: [96][ 110/ 135] Loss 0.365655 Top1 81.647727 Top5 96.946023 -2023-02-13 18:01:04,815 - Epoch: [96][ 120/ 135] Loss 0.365233 Top1 81.539714 Top5 96.933594 -2023-02-13 18:01:04,952 - Epoch: [96][ 130/ 135] Loss 0.365329 Top1 81.538462 Top5 97.004207 -2023-02-13 18:01:04,999 - Epoch: [96][ 135/ 135] Loss 0.374375 Top1 81.454927 Top5 96.980560 -2023-02-13 18:01:05,079 - ==> Top1: 81.455 Top5: 96.981 Loss: 0.374 - -2023-02-13 18:01:05,080 - ==> Confusion: -[[ 844 2 15 3 11 4 0 1 10 38 1 5 2 4 4 5 5 4 0 0 9] - [ 2 932 6 2 8 31 2 14 4 1 3 3 3 1 0 1 5 0 8 2 5] - [ 6 2 970 9 5 2 7 15 2 1 3 2 5 4 2 5 3 3 4 3 5] - [ 6 1 42 860 4 8 0 1 1 2 15 0 11 2 18 2 2 13 22 1 5] - [ 23 12 5 1 957 9 1 1 0 7 2 6 2 9 5 8 9 3 0 3 3] - [ 2 27 1 3 8 930 7 23 3 2 3 19 3 20 0 2 4 2 2 2 7] - [ 5 2 32 1 0 4 1013 6 1 1 4 0 2 2 0 6 1 7 1 10 1] - [ 4 13 12 0 2 40 5 877 3 2 4 6 3 1 0 1 1 3 28 15 4] - [ 22 6 1 3 0 0 0 2 886 32 11 4 0 14 16 3 1 1 7 0 0] - [ 102 2 5 1 6 4 0 1 56 801 0 3 0 19 3 2 0 3 0 1 3] - [ 3 3 12 7 0 3 2 1 19 1 962 2 2 12 2 0 0 1 16 2 1] - [ 2 1 0 1 7 8 0 4 0 1 0 914 33 8 0 1 4 10 3 7 1] - [ 0 0 1 5 1 3 0 0 2 1 1 37 862 1 0 5 3 28 1 0 8] - [ 4 3 6 0 10 15 0 2 14 19 7 5 5 919 1 2 1 2 0 3 6] - [ 16 4 3 23 4 5 0 1 15 7 7 1 6 6 961 2 4 7 9 3 8] - [ 3 1 8 1 8 0 3 1 0 0 0 9 7 5 1 968 4 18 0 5 4] - [ 2 6 3 1 5 2 0 0 5 0 1 6 0 4 1 12 983 4 4 3 19] - [ 5 2 1 1 1 1 1 1 0 2 0 8 11 1 0 15 0 996 0 1 4] - [ 3 4 8 8 2 2 0 21 3 0 6 1 7 0 10 0 1 2 1005 1 2] - [ 0 3 4 0 1 10 7 13 1 1 2 29 6 6 0 9 4 4 1 1042 5] - [ 174 302 406 122 106 241 103 160 156 86 225 169 359 362 158 134 231 177 231 266 9266]] - -2023-02-13 18:01:05,082 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:01:05,082 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:01:05,088 - - -2023-02-13 18:01:05,088 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:01:05,977 - Epoch: [97][ 10/ 1207] Overall Loss 0.344843 Objective Loss 0.344843 LR 0.001000 Time 0.088829 -2023-02-13 18:01:06,175 - Epoch: [97][ 20/ 1207] Overall Loss 0.344912 Objective Loss 0.344912 LR 0.001000 Time 0.054306 -2023-02-13 18:01:06,363 - Epoch: [97][ 30/ 1207] Overall Loss 0.342418 Objective Loss 0.342418 LR 0.001000 Time 0.042466 -2023-02-13 18:01:06,551 - Epoch: [97][ 40/ 1207] Overall Loss 0.335055 Objective Loss 0.335055 LR 0.001000 Time 0.036536 -2023-02-13 18:01:06,738 - Epoch: [97][ 50/ 1207] Overall Loss 0.332725 Objective Loss 0.332725 LR 0.001000 Time 0.032963 -2023-02-13 18:01:06,926 - Epoch: [97][ 60/ 1207] Overall Loss 0.325365 Objective Loss 0.325365 LR 0.001000 Time 0.030595 -2023-02-13 18:01:07,114 - Epoch: [97][ 70/ 1207] Overall Loss 0.319432 Objective Loss 0.319432 LR 0.001000 Time 0.028898 -2023-02-13 18:01:07,301 - Epoch: [97][ 80/ 1207] Overall Loss 0.321271 Objective Loss 0.321271 LR 0.001000 Time 0.027620 -2023-02-13 18:01:07,488 - Epoch: [97][ 90/ 1207] Overall Loss 0.317420 Objective Loss 0.317420 LR 0.001000 Time 0.026632 -2023-02-13 18:01:07,676 - Epoch: [97][ 100/ 1207] Overall Loss 0.317840 Objective Loss 0.317840 LR 0.001000 Time 0.025838 -2023-02-13 18:01:07,864 - Epoch: [97][ 110/ 1207] Overall Loss 0.319015 Objective Loss 0.319015 LR 0.001000 Time 0.025195 -2023-02-13 18:01:08,052 - Epoch: [97][ 120/ 1207] Overall Loss 0.319207 Objective Loss 0.319207 LR 0.001000 Time 0.024662 -2023-02-13 18:01:08,239 - Epoch: [97][ 130/ 1207] Overall Loss 0.318783 Objective Loss 0.318783 LR 0.001000 Time 0.024203 -2023-02-13 18:01:08,427 - Epoch: [97][ 140/ 1207] Overall Loss 0.320181 Objective Loss 0.320181 LR 0.001000 Time 0.023814 -2023-02-13 18:01:08,615 - Epoch: [97][ 150/ 1207] Overall Loss 0.321429 Objective Loss 0.321429 LR 0.001000 Time 0.023477 -2023-02-13 18:01:08,803 - Epoch: [97][ 160/ 1207] Overall Loss 0.323723 Objective Loss 0.323723 LR 0.001000 Time 0.023180 -2023-02-13 18:01:08,991 - Epoch: [97][ 170/ 1207] Overall Loss 0.322445 Objective Loss 0.322445 LR 0.001000 Time 0.022921 -2023-02-13 18:01:09,179 - Epoch: [97][ 180/ 1207] Overall Loss 0.323089 Objective Loss 0.323089 LR 0.001000 Time 0.022688 -2023-02-13 18:01:09,367 - Epoch: [97][ 190/ 1207] Overall Loss 0.325392 Objective Loss 0.325392 LR 0.001000 Time 0.022483 -2023-02-13 18:01:09,555 - Epoch: [97][ 200/ 1207] Overall Loss 0.324190 Objective Loss 0.324190 LR 0.001000 Time 0.022296 -2023-02-13 18:01:09,742 - Epoch: [97][ 210/ 1207] Overall Loss 0.324283 Objective Loss 0.324283 LR 0.001000 Time 0.022125 -2023-02-13 18:01:09,930 - Epoch: [97][ 220/ 1207] Overall Loss 0.325156 Objective Loss 0.325156 LR 0.001000 Time 0.021973 -2023-02-13 18:01:10,118 - Epoch: [97][ 230/ 1207] Overall Loss 0.326537 Objective Loss 0.326537 LR 0.001000 Time 0.021833 -2023-02-13 18:01:10,306 - Epoch: [97][ 240/ 1207] Overall Loss 0.327430 Objective Loss 0.327430 LR 0.001000 Time 0.021705 -2023-02-13 18:01:10,495 - Epoch: [97][ 250/ 1207] Overall Loss 0.327400 Objective Loss 0.327400 LR 0.001000 Time 0.021589 -2023-02-13 18:01:10,682 - Epoch: [97][ 260/ 1207] Overall Loss 0.327972 Objective Loss 0.327972 LR 0.001000 Time 0.021478 -2023-02-13 18:01:10,871 - Epoch: [97][ 270/ 1207] Overall Loss 0.327616 Objective Loss 0.327616 LR 0.001000 Time 0.021379 -2023-02-13 18:01:11,059 - Epoch: [97][ 280/ 1207] Overall Loss 0.328708 Objective Loss 0.328708 LR 0.001000 Time 0.021288 -2023-02-13 18:01:11,247 - Epoch: [97][ 290/ 1207] Overall Loss 0.329020 Objective Loss 0.329020 LR 0.001000 Time 0.021200 -2023-02-13 18:01:11,435 - Epoch: [97][ 300/ 1207] Overall Loss 0.329530 Objective Loss 0.329530 LR 0.001000 Time 0.021119 -2023-02-13 18:01:11,623 - Epoch: [97][ 310/ 1207] Overall Loss 0.330500 Objective Loss 0.330500 LR 0.001000 Time 0.021043 -2023-02-13 18:01:11,811 - Epoch: [97][ 320/ 1207] Overall Loss 0.330372 Objective Loss 0.330372 LR 0.001000 Time 0.020971 -2023-02-13 18:01:11,999 - Epoch: [97][ 330/ 1207] Overall Loss 0.330530 Objective Loss 0.330530 LR 0.001000 Time 0.020906 -2023-02-13 18:01:12,188 - Epoch: [97][ 340/ 1207] Overall Loss 0.331616 Objective Loss 0.331616 LR 0.001000 Time 0.020843 -2023-02-13 18:01:12,376 - Epoch: [97][ 350/ 1207] Overall Loss 0.331341 Objective Loss 0.331341 LR 0.001000 Time 0.020784 -2023-02-13 18:01:12,563 - Epoch: [97][ 360/ 1207] Overall Loss 0.331086 Objective Loss 0.331086 LR 0.001000 Time 0.020727 -2023-02-13 18:01:12,752 - Epoch: [97][ 370/ 1207] Overall Loss 0.329928 Objective Loss 0.329928 LR 0.001000 Time 0.020674 -2023-02-13 18:01:12,940 - Epoch: [97][ 380/ 1207] Overall Loss 0.330172 Objective Loss 0.330172 LR 0.001000 Time 0.020624 -2023-02-13 18:01:13,128 - Epoch: [97][ 390/ 1207] Overall Loss 0.330445 Objective Loss 0.330445 LR 0.001000 Time 0.020577 -2023-02-13 18:01:13,315 - Epoch: [97][ 400/ 1207] Overall Loss 0.329868 Objective Loss 0.329868 LR 0.001000 Time 0.020531 -2023-02-13 18:01:13,503 - Epoch: [97][ 410/ 1207] Overall Loss 0.329939 Objective Loss 0.329939 LR 0.001000 Time 0.020487 -2023-02-13 18:01:13,691 - Epoch: [97][ 420/ 1207] Overall Loss 0.330817 Objective Loss 0.330817 LR 0.001000 Time 0.020447 -2023-02-13 18:01:13,879 - Epoch: [97][ 430/ 1207] Overall Loss 0.330357 Objective Loss 0.330357 LR 0.001000 Time 0.020407 -2023-02-13 18:01:14,068 - Epoch: [97][ 440/ 1207] Overall Loss 0.330640 Objective Loss 0.330640 LR 0.001000 Time 0.020371 -2023-02-13 18:01:14,257 - Epoch: [97][ 450/ 1207] Overall Loss 0.330821 Objective Loss 0.330821 LR 0.001000 Time 0.020337 -2023-02-13 18:01:14,445 - Epoch: [97][ 460/ 1207] Overall Loss 0.331021 Objective Loss 0.331021 LR 0.001000 Time 0.020303 -2023-02-13 18:01:14,632 - Epoch: [97][ 470/ 1207] Overall Loss 0.331205 Objective Loss 0.331205 LR 0.001000 Time 0.020269 -2023-02-13 18:01:14,820 - Epoch: [97][ 480/ 1207] Overall Loss 0.331459 Objective Loss 0.331459 LR 0.001000 Time 0.020237 -2023-02-13 18:01:15,009 - Epoch: [97][ 490/ 1207] Overall Loss 0.332306 Objective Loss 0.332306 LR 0.001000 Time 0.020209 -2023-02-13 18:01:15,197 - Epoch: [97][ 500/ 1207] Overall Loss 0.331969 Objective Loss 0.331969 LR 0.001000 Time 0.020181 -2023-02-13 18:01:15,386 - Epoch: [97][ 510/ 1207] Overall Loss 0.332166 Objective Loss 0.332166 LR 0.001000 Time 0.020153 -2023-02-13 18:01:15,574 - Epoch: [97][ 520/ 1207] Overall Loss 0.332249 Objective Loss 0.332249 LR 0.001000 Time 0.020127 -2023-02-13 18:01:15,762 - Epoch: [97][ 530/ 1207] Overall Loss 0.333052 Objective Loss 0.333052 LR 0.001000 Time 0.020102 -2023-02-13 18:01:15,953 - Epoch: [97][ 540/ 1207] Overall Loss 0.333067 Objective Loss 0.333067 LR 0.001000 Time 0.020082 -2023-02-13 18:01:16,141 - Epoch: [97][ 550/ 1207] Overall Loss 0.333351 Objective Loss 0.333351 LR 0.001000 Time 0.020059 -2023-02-13 18:01:16,329 - Epoch: [97][ 560/ 1207] Overall Loss 0.333159 Objective Loss 0.333159 LR 0.001000 Time 0.020036 -2023-02-13 18:01:16,517 - Epoch: [97][ 570/ 1207] Overall Loss 0.332691 Objective Loss 0.332691 LR 0.001000 Time 0.020014 -2023-02-13 18:01:16,706 - Epoch: [97][ 580/ 1207] Overall Loss 0.332627 Objective Loss 0.332627 LR 0.001000 Time 0.019993 -2023-02-13 18:01:16,895 - Epoch: [97][ 590/ 1207] Overall Loss 0.332180 Objective Loss 0.332180 LR 0.001000 Time 0.019974 -2023-02-13 18:01:17,084 - Epoch: [97][ 600/ 1207] Overall Loss 0.332819 Objective Loss 0.332819 LR 0.001000 Time 0.019955 -2023-02-13 18:01:17,272 - Epoch: [97][ 610/ 1207] Overall Loss 0.332776 Objective Loss 0.332776 LR 0.001000 Time 0.019937 -2023-02-13 18:01:17,461 - Epoch: [97][ 620/ 1207] Overall Loss 0.332412 Objective Loss 0.332412 LR 0.001000 Time 0.019919 -2023-02-13 18:01:17,650 - Epoch: [97][ 630/ 1207] Overall Loss 0.332390 Objective Loss 0.332390 LR 0.001000 Time 0.019901 -2023-02-13 18:01:17,838 - Epoch: [97][ 640/ 1207] Overall Loss 0.332815 Objective Loss 0.332815 LR 0.001000 Time 0.019884 -2023-02-13 18:01:18,027 - Epoch: [97][ 650/ 1207] Overall Loss 0.332966 Objective Loss 0.332966 LR 0.001000 Time 0.019869 -2023-02-13 18:01:18,215 - Epoch: [97][ 660/ 1207] Overall Loss 0.332745 Objective Loss 0.332745 LR 0.001000 Time 0.019852 -2023-02-13 18:01:18,404 - Epoch: [97][ 670/ 1207] Overall Loss 0.332834 Objective Loss 0.332834 LR 0.001000 Time 0.019837 -2023-02-13 18:01:18,593 - Epoch: [97][ 680/ 1207] Overall Loss 0.333197 Objective Loss 0.333197 LR 0.001000 Time 0.019822 -2023-02-13 18:01:18,781 - Epoch: [97][ 690/ 1207] Overall Loss 0.333194 Objective Loss 0.333194 LR 0.001000 Time 0.019808 -2023-02-13 18:01:18,970 - Epoch: [97][ 700/ 1207] Overall Loss 0.333012 Objective Loss 0.333012 LR 0.001000 Time 0.019794 -2023-02-13 18:01:19,161 - Epoch: [97][ 710/ 1207] Overall Loss 0.333516 Objective Loss 0.333516 LR 0.001000 Time 0.019783 -2023-02-13 18:01:19,349 - Epoch: [97][ 720/ 1207] Overall Loss 0.334006 Objective Loss 0.334006 LR 0.001000 Time 0.019770 -2023-02-13 18:01:19,538 - Epoch: [97][ 730/ 1207] Overall Loss 0.334397 Objective Loss 0.334397 LR 0.001000 Time 0.019757 -2023-02-13 18:01:19,727 - Epoch: [97][ 740/ 1207] Overall Loss 0.334206 Objective Loss 0.334206 LR 0.001000 Time 0.019744 -2023-02-13 18:01:19,916 - Epoch: [97][ 750/ 1207] Overall Loss 0.333843 Objective Loss 0.333843 LR 0.001000 Time 0.019733 -2023-02-13 18:01:20,105 - Epoch: [97][ 760/ 1207] Overall Loss 0.333848 Objective Loss 0.333848 LR 0.001000 Time 0.019722 -2023-02-13 18:01:20,294 - Epoch: [97][ 770/ 1207] Overall Loss 0.333863 Objective Loss 0.333863 LR 0.001000 Time 0.019710 -2023-02-13 18:01:20,483 - Epoch: [97][ 780/ 1207] Overall Loss 0.333817 Objective Loss 0.333817 LR 0.001000 Time 0.019699 -2023-02-13 18:01:20,675 - Epoch: [97][ 790/ 1207] Overall Loss 0.333994 Objective Loss 0.333994 LR 0.001000 Time 0.019693 -2023-02-13 18:01:20,871 - Epoch: [97][ 800/ 1207] Overall Loss 0.334140 Objective Loss 0.334140 LR 0.001000 Time 0.019691 -2023-02-13 18:01:21,064 - Epoch: [97][ 810/ 1207] Overall Loss 0.334336 Objective Loss 0.334336 LR 0.001000 Time 0.019685 -2023-02-13 18:01:21,258 - Epoch: [97][ 820/ 1207] Overall Loss 0.334011 Objective Loss 0.334011 LR 0.001000 Time 0.019682 -2023-02-13 18:01:21,450 - Epoch: [97][ 830/ 1207] Overall Loss 0.333622 Objective Loss 0.333622 LR 0.001000 Time 0.019676 -2023-02-13 18:01:21,645 - Epoch: [97][ 840/ 1207] Overall Loss 0.334135 Objective Loss 0.334135 LR 0.001000 Time 0.019674 -2023-02-13 18:01:21,837 - Epoch: [97][ 850/ 1207] Overall Loss 0.333924 Objective Loss 0.333924 LR 0.001000 Time 0.019667 -2023-02-13 18:01:22,033 - Epoch: [97][ 860/ 1207] Overall Loss 0.334196 Objective Loss 0.334196 LR 0.001000 Time 0.019666 -2023-02-13 18:01:22,225 - Epoch: [97][ 870/ 1207] Overall Loss 0.334140 Objective Loss 0.334140 LR 0.001000 Time 0.019660 -2023-02-13 18:01:22,420 - Epoch: [97][ 880/ 1207] Overall Loss 0.334279 Objective Loss 0.334279 LR 0.001000 Time 0.019658 -2023-02-13 18:01:22,612 - Epoch: [97][ 890/ 1207] Overall Loss 0.334225 Objective Loss 0.334225 LR 0.001000 Time 0.019652 -2023-02-13 18:01:22,806 - Epoch: [97][ 900/ 1207] Overall Loss 0.333790 Objective Loss 0.333790 LR 0.001000 Time 0.019650 -2023-02-13 18:01:22,998 - Epoch: [97][ 910/ 1207] Overall Loss 0.333670 Objective Loss 0.333670 LR 0.001000 Time 0.019644 -2023-02-13 18:01:23,193 - Epoch: [97][ 920/ 1207] Overall Loss 0.333589 Objective Loss 0.333589 LR 0.001000 Time 0.019642 -2023-02-13 18:01:23,385 - Epoch: [97][ 930/ 1207] Overall Loss 0.333739 Objective Loss 0.333739 LR 0.001000 Time 0.019637 -2023-02-13 18:01:23,580 - Epoch: [97][ 940/ 1207] Overall Loss 0.333790 Objective Loss 0.333790 LR 0.001000 Time 0.019634 -2023-02-13 18:01:23,772 - Epoch: [97][ 950/ 1207] Overall Loss 0.333547 Objective Loss 0.333547 LR 0.001000 Time 0.019629 -2023-02-13 18:01:23,967 - Epoch: [97][ 960/ 1207] Overall Loss 0.333516 Objective Loss 0.333516 LR 0.001000 Time 0.019628 -2023-02-13 18:01:24,158 - Epoch: [97][ 970/ 1207] Overall Loss 0.334028 Objective Loss 0.334028 LR 0.001000 Time 0.019623 -2023-02-13 18:01:24,353 - Epoch: [97][ 980/ 1207] Overall Loss 0.334302 Objective Loss 0.334302 LR 0.001000 Time 0.019621 -2023-02-13 18:01:24,545 - Epoch: [97][ 990/ 1207] Overall Loss 0.334500 Objective Loss 0.334500 LR 0.001000 Time 0.019616 -2023-02-13 18:01:24,740 - Epoch: [97][ 1000/ 1207] Overall Loss 0.334670 Objective Loss 0.334670 LR 0.001000 Time 0.019614 -2023-02-13 18:01:24,931 - Epoch: [97][ 1010/ 1207] Overall Loss 0.334950 Objective Loss 0.334950 LR 0.001000 Time 0.019609 -2023-02-13 18:01:25,126 - Epoch: [97][ 1020/ 1207] Overall Loss 0.335231 Objective Loss 0.335231 LR 0.001000 Time 0.019608 -2023-02-13 18:01:25,318 - Epoch: [97][ 1030/ 1207] Overall Loss 0.335374 Objective Loss 0.335374 LR 0.001000 Time 0.019603 -2023-02-13 18:01:25,513 - Epoch: [97][ 1040/ 1207] Overall Loss 0.335615 Objective Loss 0.335615 LR 0.001000 Time 0.019602 -2023-02-13 18:01:25,704 - Epoch: [97][ 1050/ 1207] Overall Loss 0.335942 Objective Loss 0.335942 LR 0.001000 Time 0.019597 -2023-02-13 18:01:25,900 - Epoch: [97][ 1060/ 1207] Overall Loss 0.335896 Objective Loss 0.335896 LR 0.001000 Time 0.019596 -2023-02-13 18:01:26,092 - Epoch: [97][ 1070/ 1207] Overall Loss 0.335683 Objective Loss 0.335683 LR 0.001000 Time 0.019593 -2023-02-13 18:01:26,287 - Epoch: [97][ 1080/ 1207] Overall Loss 0.335960 Objective Loss 0.335960 LR 0.001000 Time 0.019591 -2023-02-13 18:01:26,479 - Epoch: [97][ 1090/ 1207] Overall Loss 0.336073 Objective Loss 0.336073 LR 0.001000 Time 0.019587 -2023-02-13 18:01:26,674 - Epoch: [97][ 1100/ 1207] Overall Loss 0.336092 Objective Loss 0.336092 LR 0.001000 Time 0.019586 -2023-02-13 18:01:26,866 - Epoch: [97][ 1110/ 1207] Overall Loss 0.336023 Objective Loss 0.336023 LR 0.001000 Time 0.019583 -2023-02-13 18:01:27,062 - Epoch: [97][ 1120/ 1207] Overall Loss 0.336359 Objective Loss 0.336359 LR 0.001000 Time 0.019582 -2023-02-13 18:01:27,253 - Epoch: [97][ 1130/ 1207] Overall Loss 0.336418 Objective Loss 0.336418 LR 0.001000 Time 0.019578 -2023-02-13 18:01:27,449 - Epoch: [97][ 1140/ 1207] Overall Loss 0.336443 Objective Loss 0.336443 LR 0.001000 Time 0.019577 -2023-02-13 18:01:27,641 - Epoch: [97][ 1150/ 1207] Overall Loss 0.336380 Objective Loss 0.336380 LR 0.001000 Time 0.019573 -2023-02-13 18:01:27,835 - Epoch: [97][ 1160/ 1207] Overall Loss 0.336343 Objective Loss 0.336343 LR 0.001000 Time 0.019572 -2023-02-13 18:01:28,027 - Epoch: [97][ 1170/ 1207] Overall Loss 0.336583 Objective Loss 0.336583 LR 0.001000 Time 0.019568 -2023-02-13 18:01:28,221 - Epoch: [97][ 1180/ 1207] Overall Loss 0.336511 Objective Loss 0.336511 LR 0.001000 Time 0.019567 -2023-02-13 18:01:28,413 - Epoch: [97][ 1190/ 1207] Overall Loss 0.336384 Objective Loss 0.336384 LR 0.001000 Time 0.019563 -2023-02-13 18:01:28,661 - Epoch: [97][ 1200/ 1207] Overall Loss 0.336616 Objective Loss 0.336616 LR 0.001000 Time 0.019606 -2023-02-13 18:01:28,775 - Epoch: [97][ 1207/ 1207] Overall Loss 0.336302 Objective Loss 0.336302 Top1 80.487805 Top5 96.341463 LR 0.001000 Time 0.019587 -2023-02-13 18:01:28,847 - --- validate (epoch=97)----------- -2023-02-13 18:01:28,847 - 34311 samples (256 per mini-batch) -2023-02-13 18:01:29,250 - Epoch: [97][ 10/ 135] Loss 0.408005 Top1 80.898438 Top5 96.328125 -2023-02-13 18:01:29,381 - Epoch: [97][ 20/ 135] Loss 0.393346 Top1 81.640625 Top5 96.679688 -2023-02-13 18:01:29,511 - Epoch: [97][ 30/ 135] Loss 0.385615 Top1 81.875000 Top5 96.888021 -2023-02-13 18:01:29,641 - Epoch: [97][ 40/ 135] Loss 0.377287 Top1 81.972656 Top5 97.031250 -2023-02-13 18:01:29,771 - Epoch: [97][ 50/ 135] Loss 0.363594 Top1 82.234375 Top5 97.070312 -2023-02-13 18:01:29,901 - Epoch: [97][ 60/ 135] Loss 0.361888 Top1 82.063802 Top5 97.005208 -2023-02-13 18:01:30,031 - Epoch: [97][ 70/ 135] Loss 0.364211 Top1 82.008929 Top5 97.081473 -2023-02-13 18:01:30,153 - Epoch: [97][ 80/ 135] Loss 0.362763 Top1 81.855469 Top5 97.055664 -2023-02-13 18:01:30,277 - Epoch: [97][ 90/ 135] Loss 0.361957 Top1 81.809896 Top5 97.087674 -2023-02-13 18:01:30,398 - Epoch: [97][ 100/ 135] Loss 0.367220 Top1 81.570312 Top5 97.070312 -2023-02-13 18:01:30,525 - Epoch: [97][ 110/ 135] Loss 0.369538 Top1 81.477273 Top5 97.049006 -2023-02-13 18:01:30,654 - Epoch: [97][ 120/ 135] Loss 0.367802 Top1 81.445312 Top5 97.060547 -2023-02-13 18:01:30,785 - Epoch: [97][ 130/ 135] Loss 0.365936 Top1 81.385216 Top5 97.061298 -2023-02-13 18:01:30,833 - Epoch: [97][ 135/ 135] Loss 0.376123 Top1 81.405380 Top5 97.067996 -2023-02-13 18:01:30,902 - ==> Top1: 81.405 Top5: 97.068 Loss: 0.376 - -2023-02-13 18:01:30,903 - ==> Confusion: -[[ 828 5 9 2 14 1 0 2 5 71 2 5 0 4 6 3 3 0 1 2 4] - [ 3 931 2 2 13 23 4 23 3 3 1 1 0 1 2 2 5 0 8 2 4] - [ 9 4 952 5 4 1 23 20 0 0 2 2 1 8 3 4 2 4 4 4 6] - [ 8 1 40 853 4 5 3 3 3 2 15 0 5 4 19 1 4 7 36 0 3] - [ 22 11 2 1 974 4 1 2 0 6 0 9 1 4 6 2 13 0 0 3 5] - [ 1 26 3 5 9 921 7 39 1 6 3 9 3 16 1 4 5 1 1 8 1] - [ 1 3 24 2 2 3 1023 8 2 3 5 4 1 1 0 1 2 3 1 9 1] - [ 3 5 10 2 2 22 3 938 0 1 2 7 2 2 0 0 1 1 11 10 2] - [ 17 3 0 1 3 0 0 2 871 54 7 3 1 11 17 2 1 2 14 0 0] - [ 72 1 7 0 10 1 0 2 31 866 0 1 0 13 3 1 1 2 0 0 1] - [ 3 2 7 4 1 1 8 6 23 1 953 2 1 10 3 0 1 2 18 1 4] - [ 3 1 0 0 6 9 1 11 0 3 1 891 22 10 1 8 3 10 1 23 1] - [ 0 1 2 9 1 2 0 4 2 2 0 33 832 2 1 11 6 35 2 6 8] - [ 3 1 2 1 9 11 1 3 12 30 5 6 2 918 3 5 7 1 2 2 0] - [ 10 3 4 13 5 5 0 3 29 6 2 0 3 4 979 0 1 3 16 1 5] - [ 4 2 7 0 4 2 11 1 0 0 0 7 6 3 0 958 12 13 0 8 8] - [ 2 7 4 0 6 1 0 1 0 3 0 1 1 2 0 10 999 1 3 6 14] - [ 8 3 0 1 1 3 1 2 0 0 1 8 11 1 1 15 0 988 0 3 4] - [ 7 5 6 9 1 1 0 48 1 0 3 5 3 0 13 1 0 2 977 4 0] - [ 0 3 3 0 1 4 6 21 1 0 1 17 1 4 0 5 6 4 1 1063 7] - [ 191 277 347 107 169 198 127 308 119 124 179 133 267 390 147 130 288 130 242 345 9216]] - -2023-02-13 18:01:30,904 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:01:30,904 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:01:30,910 - - -2023-02-13 18:01:30,910 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:01:31,796 - Epoch: [98][ 10/ 1207] Overall Loss 0.331282 Objective Loss 0.331282 LR 0.001000 Time 0.088565 -2023-02-13 18:01:31,987 - Epoch: [98][ 20/ 1207] Overall Loss 0.326322 Objective Loss 0.326322 LR 0.001000 Time 0.053787 -2023-02-13 18:01:32,178 - Epoch: [98][ 30/ 1207] Overall Loss 0.320996 Objective Loss 0.320996 LR 0.001000 Time 0.042213 -2023-02-13 18:01:32,367 - Epoch: [98][ 40/ 1207] Overall Loss 0.327997 Objective Loss 0.327997 LR 0.001000 Time 0.036376 -2023-02-13 18:01:32,556 - Epoch: [98][ 50/ 1207] Overall Loss 0.328539 Objective Loss 0.328539 LR 0.001000 Time 0.032878 -2023-02-13 18:01:32,745 - Epoch: [98][ 60/ 1207] Overall Loss 0.328723 Objective Loss 0.328723 LR 0.001000 Time 0.030535 -2023-02-13 18:01:32,935 - Epoch: [98][ 70/ 1207] Overall Loss 0.327551 Objective Loss 0.327551 LR 0.001000 Time 0.028881 -2023-02-13 18:01:33,124 - Epoch: [98][ 80/ 1207] Overall Loss 0.326632 Objective Loss 0.326632 LR 0.001000 Time 0.027637 -2023-02-13 18:01:33,314 - Epoch: [98][ 90/ 1207] Overall Loss 0.327753 Objective Loss 0.327753 LR 0.001000 Time 0.026672 -2023-02-13 18:01:33,503 - Epoch: [98][ 100/ 1207] Overall Loss 0.325461 Objective Loss 0.325461 LR 0.001000 Time 0.025890 -2023-02-13 18:01:33,694 - Epoch: [98][ 110/ 1207] Overall Loss 0.326133 Objective Loss 0.326133 LR 0.001000 Time 0.025264 -2023-02-13 18:01:33,882 - Epoch: [98][ 120/ 1207] Overall Loss 0.323642 Objective Loss 0.323642 LR 0.001000 Time 0.024727 -2023-02-13 18:01:34,073 - Epoch: [98][ 130/ 1207] Overall Loss 0.320943 Objective Loss 0.320943 LR 0.001000 Time 0.024289 -2023-02-13 18:01:34,261 - Epoch: [98][ 140/ 1207] Overall Loss 0.320127 Objective Loss 0.320127 LR 0.001000 Time 0.023899 -2023-02-13 18:01:34,452 - Epoch: [98][ 150/ 1207] Overall Loss 0.319099 Objective Loss 0.319099 LR 0.001000 Time 0.023572 -2023-02-13 18:01:34,640 - Epoch: [98][ 160/ 1207] Overall Loss 0.319045 Objective Loss 0.319045 LR 0.001000 Time 0.023275 -2023-02-13 18:01:34,831 - Epoch: [98][ 170/ 1207] Overall Loss 0.319125 Objective Loss 0.319125 LR 0.001000 Time 0.023023 -2023-02-13 18:01:35,020 - Epoch: [98][ 180/ 1207] Overall Loss 0.319740 Objective Loss 0.319740 LR 0.001000 Time 0.022797 -2023-02-13 18:01:35,211 - Epoch: [98][ 190/ 1207] Overall Loss 0.318807 Objective Loss 0.318807 LR 0.001000 Time 0.022598 -2023-02-13 18:01:35,401 - Epoch: [98][ 200/ 1207] Overall Loss 0.319627 Objective Loss 0.319627 LR 0.001000 Time 0.022414 -2023-02-13 18:01:35,591 - Epoch: [98][ 210/ 1207] Overall Loss 0.320018 Objective Loss 0.320018 LR 0.001000 Time 0.022250 -2023-02-13 18:01:35,780 - Epoch: [98][ 220/ 1207] Overall Loss 0.320405 Objective Loss 0.320405 LR 0.001000 Time 0.022096 -2023-02-13 18:01:35,971 - Epoch: [98][ 230/ 1207] Overall Loss 0.319958 Objective Loss 0.319958 LR 0.001000 Time 0.021966 -2023-02-13 18:01:36,161 - Epoch: [98][ 240/ 1207] Overall Loss 0.319553 Objective Loss 0.319553 LR 0.001000 Time 0.021840 -2023-02-13 18:01:36,351 - Epoch: [98][ 250/ 1207] Overall Loss 0.319194 Objective Loss 0.319194 LR 0.001000 Time 0.021727 -2023-02-13 18:01:36,541 - Epoch: [98][ 260/ 1207] Overall Loss 0.318921 Objective Loss 0.318921 LR 0.001000 Time 0.021618 -2023-02-13 18:01:36,731 - Epoch: [98][ 270/ 1207] Overall Loss 0.320407 Objective Loss 0.320407 LR 0.001000 Time 0.021521 -2023-02-13 18:01:36,921 - Epoch: [98][ 280/ 1207] Overall Loss 0.321554 Objective Loss 0.321554 LR 0.001000 Time 0.021428 -2023-02-13 18:01:37,112 - Epoch: [98][ 290/ 1207] Overall Loss 0.321418 Objective Loss 0.321418 LR 0.001000 Time 0.021347 -2023-02-13 18:01:37,300 - Epoch: [98][ 300/ 1207] Overall Loss 0.321837 Objective Loss 0.321837 LR 0.001000 Time 0.021262 -2023-02-13 18:01:37,492 - Epoch: [98][ 310/ 1207] Overall Loss 0.322439 Objective Loss 0.322439 LR 0.001000 Time 0.021194 -2023-02-13 18:01:37,681 - Epoch: [98][ 320/ 1207] Overall Loss 0.323005 Objective Loss 0.323005 LR 0.001000 Time 0.021120 -2023-02-13 18:01:37,871 - Epoch: [98][ 330/ 1207] Overall Loss 0.323111 Objective Loss 0.323111 LR 0.001000 Time 0.021057 -2023-02-13 18:01:38,062 - Epoch: [98][ 340/ 1207] Overall Loss 0.323899 Objective Loss 0.323899 LR 0.001000 Time 0.020996 -2023-02-13 18:01:38,252 - Epoch: [98][ 350/ 1207] Overall Loss 0.323479 Objective Loss 0.323479 LR 0.001000 Time 0.020939 -2023-02-13 18:01:38,442 - Epoch: [98][ 360/ 1207] Overall Loss 0.323537 Objective Loss 0.323537 LR 0.001000 Time 0.020883 -2023-02-13 18:01:38,632 - Epoch: [98][ 370/ 1207] Overall Loss 0.322509 Objective Loss 0.322509 LR 0.001000 Time 0.020832 -2023-02-13 18:01:38,822 - Epoch: [98][ 380/ 1207] Overall Loss 0.322309 Objective Loss 0.322309 LR 0.001000 Time 0.020784 -2023-02-13 18:01:39,012 - Epoch: [98][ 390/ 1207] Overall Loss 0.321848 Objective Loss 0.321848 LR 0.001000 Time 0.020737 -2023-02-13 18:01:39,202 - Epoch: [98][ 400/ 1207] Overall Loss 0.321607 Objective Loss 0.321607 LR 0.001000 Time 0.020692 -2023-02-13 18:01:39,392 - Epoch: [98][ 410/ 1207] Overall Loss 0.321466 Objective Loss 0.321466 LR 0.001000 Time 0.020649 -2023-02-13 18:01:39,581 - Epoch: [98][ 420/ 1207] Overall Loss 0.321575 Objective Loss 0.321575 LR 0.001000 Time 0.020608 -2023-02-13 18:01:39,771 - Epoch: [98][ 430/ 1207] Overall Loss 0.321865 Objective Loss 0.321865 LR 0.001000 Time 0.020569 -2023-02-13 18:01:39,960 - Epoch: [98][ 440/ 1207] Overall Loss 0.322163 Objective Loss 0.322163 LR 0.001000 Time 0.020529 -2023-02-13 18:01:40,151 - Epoch: [98][ 450/ 1207] Overall Loss 0.322420 Objective Loss 0.322420 LR 0.001000 Time 0.020497 -2023-02-13 18:01:40,340 - Epoch: [98][ 460/ 1207] Overall Loss 0.322220 Objective Loss 0.322220 LR 0.001000 Time 0.020461 -2023-02-13 18:01:40,530 - Epoch: [98][ 470/ 1207] Overall Loss 0.322252 Objective Loss 0.322252 LR 0.001000 Time 0.020431 -2023-02-13 18:01:40,719 - Epoch: [98][ 480/ 1207] Overall Loss 0.322097 Objective Loss 0.322097 LR 0.001000 Time 0.020398 -2023-02-13 18:01:40,911 - Epoch: [98][ 490/ 1207] Overall Loss 0.323041 Objective Loss 0.323041 LR 0.001000 Time 0.020372 -2023-02-13 18:01:41,100 - Epoch: [98][ 500/ 1207] Overall Loss 0.322946 Objective Loss 0.322946 LR 0.001000 Time 0.020343 -2023-02-13 18:01:41,291 - Epoch: [98][ 510/ 1207] Overall Loss 0.323029 Objective Loss 0.323029 LR 0.001000 Time 0.020316 -2023-02-13 18:01:41,480 - Epoch: [98][ 520/ 1207] Overall Loss 0.323590 Objective Loss 0.323590 LR 0.001000 Time 0.020290 -2023-02-13 18:01:41,671 - Epoch: [98][ 530/ 1207] Overall Loss 0.323925 Objective Loss 0.323925 LR 0.001000 Time 0.020265 -2023-02-13 18:01:41,860 - Epoch: [98][ 540/ 1207] Overall Loss 0.324305 Objective Loss 0.324305 LR 0.001000 Time 0.020240 -2023-02-13 18:01:42,051 - Epoch: [98][ 550/ 1207] Overall Loss 0.324195 Objective Loss 0.324195 LR 0.001000 Time 0.020218 -2023-02-13 18:01:42,240 - Epoch: [98][ 560/ 1207] Overall Loss 0.324270 Objective Loss 0.324270 LR 0.001000 Time 0.020195 -2023-02-13 18:01:42,431 - Epoch: [98][ 570/ 1207] Overall Loss 0.324285 Objective Loss 0.324285 LR 0.001000 Time 0.020174 -2023-02-13 18:01:42,620 - Epoch: [98][ 580/ 1207] Overall Loss 0.324844 Objective Loss 0.324844 LR 0.001000 Time 0.020151 -2023-02-13 18:01:42,809 - Epoch: [98][ 590/ 1207] Overall Loss 0.324732 Objective Loss 0.324732 LR 0.001000 Time 0.020131 -2023-02-13 18:01:42,998 - Epoch: [98][ 600/ 1207] Overall Loss 0.325179 Objective Loss 0.325179 LR 0.001000 Time 0.020108 -2023-02-13 18:01:43,188 - Epoch: [98][ 610/ 1207] Overall Loss 0.325277 Objective Loss 0.325277 LR 0.001000 Time 0.020090 -2023-02-13 18:01:43,377 - Epoch: [98][ 620/ 1207] Overall Loss 0.325076 Objective Loss 0.325076 LR 0.001000 Time 0.020070 -2023-02-13 18:01:43,567 - Epoch: [98][ 630/ 1207] Overall Loss 0.325734 Objective Loss 0.325734 LR 0.001000 Time 0.020053 -2023-02-13 18:01:43,757 - Epoch: [98][ 640/ 1207] Overall Loss 0.326380 Objective Loss 0.326380 LR 0.001000 Time 0.020035 -2023-02-13 18:01:43,947 - Epoch: [98][ 650/ 1207] Overall Loss 0.326390 Objective Loss 0.326390 LR 0.001000 Time 0.020019 -2023-02-13 18:01:44,136 - Epoch: [98][ 660/ 1207] Overall Loss 0.326481 Objective Loss 0.326481 LR 0.001000 Time 0.020002 -2023-02-13 18:01:44,324 - Epoch: [98][ 670/ 1207] Overall Loss 0.326568 Objective Loss 0.326568 LR 0.001000 Time 0.019983 -2023-02-13 18:01:44,513 - Epoch: [98][ 680/ 1207] Overall Loss 0.326242 Objective Loss 0.326242 LR 0.001000 Time 0.019966 -2023-02-13 18:01:44,701 - Epoch: [98][ 690/ 1207] Overall Loss 0.326053 Objective Loss 0.326053 LR 0.001000 Time 0.019949 -2023-02-13 18:01:44,889 - Epoch: [98][ 700/ 1207] Overall Loss 0.326186 Objective Loss 0.326186 LR 0.001000 Time 0.019932 -2023-02-13 18:01:45,078 - Epoch: [98][ 710/ 1207] Overall Loss 0.326368 Objective Loss 0.326368 LR 0.001000 Time 0.019917 -2023-02-13 18:01:45,266 - Epoch: [98][ 720/ 1207] Overall Loss 0.326160 Objective Loss 0.326160 LR 0.001000 Time 0.019902 -2023-02-13 18:01:45,455 - Epoch: [98][ 730/ 1207] Overall Loss 0.326075 Objective Loss 0.326075 LR 0.001000 Time 0.019887 -2023-02-13 18:01:45,643 - Epoch: [98][ 740/ 1207] Overall Loss 0.326452 Objective Loss 0.326452 LR 0.001000 Time 0.019872 -2023-02-13 18:01:45,832 - Epoch: [98][ 750/ 1207] Overall Loss 0.326737 Objective Loss 0.326737 LR 0.001000 Time 0.019858 -2023-02-13 18:01:46,022 - Epoch: [98][ 760/ 1207] Overall Loss 0.326726 Objective Loss 0.326726 LR 0.001000 Time 0.019846 -2023-02-13 18:01:46,211 - Epoch: [98][ 770/ 1207] Overall Loss 0.326724 Objective Loss 0.326724 LR 0.001000 Time 0.019833 -2023-02-13 18:01:46,398 - Epoch: [98][ 780/ 1207] Overall Loss 0.326976 Objective Loss 0.326976 LR 0.001000 Time 0.019819 -2023-02-13 18:01:46,587 - Epoch: [98][ 790/ 1207] Overall Loss 0.327101 Objective Loss 0.327101 LR 0.001000 Time 0.019806 -2023-02-13 18:01:46,774 - Epoch: [98][ 800/ 1207] Overall Loss 0.327072 Objective Loss 0.327072 LR 0.001000 Time 0.019793 -2023-02-13 18:01:46,963 - Epoch: [98][ 810/ 1207] Overall Loss 0.327762 Objective Loss 0.327762 LR 0.001000 Time 0.019781 -2023-02-13 18:01:47,153 - Epoch: [98][ 820/ 1207] Overall Loss 0.327661 Objective Loss 0.327661 LR 0.001000 Time 0.019771 -2023-02-13 18:01:47,342 - Epoch: [98][ 830/ 1207] Overall Loss 0.327808 Objective Loss 0.327808 LR 0.001000 Time 0.019760 -2023-02-13 18:01:47,530 - Epoch: [98][ 840/ 1207] Overall Loss 0.328165 Objective Loss 0.328165 LR 0.001000 Time 0.019748 -2023-02-13 18:01:47,718 - Epoch: [98][ 850/ 1207] Overall Loss 0.328257 Objective Loss 0.328257 LR 0.001000 Time 0.019737 -2023-02-13 18:01:47,907 - Epoch: [98][ 860/ 1207] Overall Loss 0.328311 Objective Loss 0.328311 LR 0.001000 Time 0.019727 -2023-02-13 18:01:48,098 - Epoch: [98][ 870/ 1207] Overall Loss 0.328352 Objective Loss 0.328352 LR 0.001000 Time 0.019719 -2023-02-13 18:01:48,286 - Epoch: [98][ 880/ 1207] Overall Loss 0.328327 Objective Loss 0.328327 LR 0.001000 Time 0.019708 -2023-02-13 18:01:48,475 - Epoch: [98][ 890/ 1207] Overall Loss 0.328381 Objective Loss 0.328381 LR 0.001000 Time 0.019698 -2023-02-13 18:01:48,663 - Epoch: [98][ 900/ 1207] Overall Loss 0.328415 Objective Loss 0.328415 LR 0.001000 Time 0.019687 -2023-02-13 18:01:48,850 - Epoch: [98][ 910/ 1207] Overall Loss 0.328495 Objective Loss 0.328495 LR 0.001000 Time 0.019677 -2023-02-13 18:01:49,039 - Epoch: [98][ 920/ 1207] Overall Loss 0.328548 Objective Loss 0.328548 LR 0.001000 Time 0.019668 -2023-02-13 18:01:49,227 - Epoch: [98][ 930/ 1207] Overall Loss 0.328727 Objective Loss 0.328727 LR 0.001000 Time 0.019658 -2023-02-13 18:01:49,415 - Epoch: [98][ 940/ 1207] Overall Loss 0.328883 Objective Loss 0.328883 LR 0.001000 Time 0.019648 -2023-02-13 18:01:49,604 - Epoch: [98][ 950/ 1207] Overall Loss 0.328823 Objective Loss 0.328823 LR 0.001000 Time 0.019640 -2023-02-13 18:01:49,791 - Epoch: [98][ 960/ 1207] Overall Loss 0.329086 Objective Loss 0.329086 LR 0.001000 Time 0.019630 -2023-02-13 18:01:49,982 - Epoch: [98][ 970/ 1207] Overall Loss 0.329424 Objective Loss 0.329424 LR 0.001000 Time 0.019624 -2023-02-13 18:01:50,171 - Epoch: [98][ 980/ 1207] Overall Loss 0.329962 Objective Loss 0.329962 LR 0.001000 Time 0.019616 -2023-02-13 18:01:50,359 - Epoch: [98][ 990/ 1207] Overall Loss 0.330097 Objective Loss 0.330097 LR 0.001000 Time 0.019608 -2023-02-13 18:01:50,547 - Epoch: [98][ 1000/ 1207] Overall Loss 0.330432 Objective Loss 0.330432 LR 0.001000 Time 0.019600 -2023-02-13 18:01:50,735 - Epoch: [98][ 1010/ 1207] Overall Loss 0.330546 Objective Loss 0.330546 LR 0.001000 Time 0.019591 -2023-02-13 18:01:50,925 - Epoch: [98][ 1020/ 1207] Overall Loss 0.330622 Objective Loss 0.330622 LR 0.001000 Time 0.019585 -2023-02-13 18:01:51,114 - Epoch: [98][ 1030/ 1207] Overall Loss 0.330643 Objective Loss 0.330643 LR 0.001000 Time 0.019578 -2023-02-13 18:01:51,302 - Epoch: [98][ 1040/ 1207] Overall Loss 0.330942 Objective Loss 0.330942 LR 0.001000 Time 0.019570 -2023-02-13 18:01:51,490 - Epoch: [98][ 1050/ 1207] Overall Loss 0.330888 Objective Loss 0.330888 LR 0.001000 Time 0.019563 -2023-02-13 18:01:51,678 - Epoch: [98][ 1060/ 1207] Overall Loss 0.331337 Objective Loss 0.331337 LR 0.001000 Time 0.019555 -2023-02-13 18:01:51,866 - Epoch: [98][ 1070/ 1207] Overall Loss 0.331805 Objective Loss 0.331805 LR 0.001000 Time 0.019548 -2023-02-13 18:01:52,055 - Epoch: [98][ 1080/ 1207] Overall Loss 0.331955 Objective Loss 0.331955 LR 0.001000 Time 0.019541 -2023-02-13 18:01:52,243 - Epoch: [98][ 1090/ 1207] Overall Loss 0.332388 Objective Loss 0.332388 LR 0.001000 Time 0.019534 -2023-02-13 18:01:52,431 - Epoch: [98][ 1100/ 1207] Overall Loss 0.332320 Objective Loss 0.332320 LR 0.001000 Time 0.019527 -2023-02-13 18:01:52,619 - Epoch: [98][ 1110/ 1207] Overall Loss 0.332726 Objective Loss 0.332726 LR 0.001000 Time 0.019520 -2023-02-13 18:01:52,807 - Epoch: [98][ 1120/ 1207] Overall Loss 0.332878 Objective Loss 0.332878 LR 0.001000 Time 0.019513 -2023-02-13 18:01:52,996 - Epoch: [98][ 1130/ 1207] Overall Loss 0.332779 Objective Loss 0.332779 LR 0.001000 Time 0.019507 -2023-02-13 18:01:53,184 - Epoch: [98][ 1140/ 1207] Overall Loss 0.332906 Objective Loss 0.332906 LR 0.001000 Time 0.019501 -2023-02-13 18:01:53,372 - Epoch: [98][ 1150/ 1207] Overall Loss 0.333072 Objective Loss 0.333072 LR 0.001000 Time 0.019495 -2023-02-13 18:01:53,560 - Epoch: [98][ 1160/ 1207] Overall Loss 0.333455 Objective Loss 0.333455 LR 0.001000 Time 0.019488 -2023-02-13 18:01:53,748 - Epoch: [98][ 1170/ 1207] Overall Loss 0.333457 Objective Loss 0.333457 LR 0.001000 Time 0.019482 -2023-02-13 18:01:53,936 - Epoch: [98][ 1180/ 1207] Overall Loss 0.333292 Objective Loss 0.333292 LR 0.001000 Time 0.019476 -2023-02-13 18:01:54,125 - Epoch: [98][ 1190/ 1207] Overall Loss 0.333445 Objective Loss 0.333445 LR 0.001000 Time 0.019470 -2023-02-13 18:01:54,363 - Epoch: [98][ 1200/ 1207] Overall Loss 0.333465 Objective Loss 0.333465 LR 0.001000 Time 0.019507 -2023-02-13 18:01:54,478 - Epoch: [98][ 1207/ 1207] Overall Loss 0.333356 Objective Loss 0.333356 Top1 82.317073 Top5 97.865854 LR 0.001000 Time 0.019489 -2023-02-13 18:01:54,550 - --- validate (epoch=98)----------- -2023-02-13 18:01:54,551 - 34311 samples (256 per mini-batch) -2023-02-13 18:01:54,955 - Epoch: [98][ 10/ 135] Loss 0.402025 Top1 82.226562 Top5 97.421875 -2023-02-13 18:01:55,088 - Epoch: [98][ 20/ 135] Loss 0.399383 Top1 82.363281 Top5 97.539062 -2023-02-13 18:01:55,214 - Epoch: [98][ 30/ 135] Loss 0.389359 Top1 82.356771 Top5 97.330729 -2023-02-13 18:01:55,340 - Epoch: [98][ 40/ 135] Loss 0.384533 Top1 82.080078 Top5 97.177734 -2023-02-13 18:01:55,462 - Epoch: [98][ 50/ 135] Loss 0.372859 Top1 82.367188 Top5 97.250000 -2023-02-13 18:01:55,587 - Epoch: [98][ 60/ 135] Loss 0.372504 Top1 82.389323 Top5 97.259115 -2023-02-13 18:01:55,713 - Epoch: [98][ 70/ 135] Loss 0.368786 Top1 82.483259 Top5 97.204241 -2023-02-13 18:01:55,846 - Epoch: [98][ 80/ 135] Loss 0.368337 Top1 82.407227 Top5 97.246094 -2023-02-13 18:01:55,985 - Epoch: [98][ 90/ 135] Loss 0.368256 Top1 82.335069 Top5 97.256944 -2023-02-13 18:01:56,129 - Epoch: [98][ 100/ 135] Loss 0.370526 Top1 82.320312 Top5 97.246094 -2023-02-13 18:01:56,272 - Epoch: [98][ 110/ 135] Loss 0.372005 Top1 82.382812 Top5 97.226562 -2023-02-13 18:01:56,416 - Epoch: [98][ 120/ 135] Loss 0.372695 Top1 82.412109 Top5 97.233073 -2023-02-13 18:01:56,554 - Epoch: [98][ 130/ 135] Loss 0.372063 Top1 82.313702 Top5 97.190505 -2023-02-13 18:01:56,599 - Epoch: [98][ 135/ 135] Loss 0.369845 Top1 82.340940 Top5 97.184576 -2023-02-13 18:01:56,672 - ==> Top1: 82.341 Top5: 97.185 Loss: 0.370 - -2023-02-13 18:01:56,673 - ==> Confusion: -[[ 772 4 11 3 21 5 0 4 4 101 1 4 0 4 8 6 5 5 0 1 8] - [ 1 913 0 4 9 31 10 21 5 1 3 8 1 0 4 3 5 0 6 2 6] - [ 8 4 919 13 7 0 37 18 0 2 2 1 2 5 1 10 5 5 7 2 10] - [ 5 2 17 883 4 4 3 3 3 3 14 1 10 2 16 3 3 8 23 1 8] - [ 11 9 0 0 981 8 2 2 1 8 1 5 2 8 6 7 4 0 0 2 9] - [ 4 18 3 7 5 935 8 22 2 5 1 23 6 16 0 0 3 0 2 6 4] - [ 3 3 16 2 1 3 1023 6 1 3 1 2 3 1 1 10 3 7 1 6 3] - [ 3 11 11 3 4 35 7 879 0 1 1 5 6 1 1 0 1 1 30 18 6] - [ 13 2 2 2 2 1 1 1 874 51 13 3 1 8 20 2 1 2 7 0 3] - [ 62 0 2 0 4 2 0 0 33 867 2 1 0 17 6 4 1 4 2 0 5] - [ 2 2 1 6 3 2 4 3 20 2 969 3 2 9 4 0 0 1 12 0 6] - [ 2 2 0 0 4 8 1 6 1 1 0 913 24 8 1 5 2 12 2 10 3] - [ 1 1 2 3 1 3 0 1 2 0 0 43 842 1 3 12 3 24 0 0 17] - [ 4 2 1 1 8 9 1 0 15 30 12 14 2 890 7 8 4 3 1 2 10] - [ 7 1 1 22 6 2 1 2 22 4 3 5 1 5 968 4 4 5 12 1 16] - [ 1 2 6 0 8 0 4 0 1 0 0 8 6 3 3 979 1 11 0 5 8] - [ 0 3 0 1 11 2 0 0 2 0 1 2 1 3 2 17 993 4 1 5 13] - [ 1 1 0 4 0 0 4 2 0 4 0 15 17 0 0 19 0 972 0 4 8] - [ 2 3 8 14 4 2 1 24 2 1 7 2 9 0 17 1 1 2 981 2 3] - [ 1 3 3 3 0 4 13 15 0 0 1 26 3 2 0 12 11 4 1 1037 9] - [ 111 231 240 134 174 200 107 187 99 125 209 180 279 327 156 136 315 111 187 264 9662]] - -2023-02-13 18:01:56,674 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:01:56,674 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:01:56,680 - - -2023-02-13 18:01:56,680 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:01:57,668 - Epoch: [99][ 10/ 1207] Overall Loss 0.338493 Objective Loss 0.338493 LR 0.001000 Time 0.098770 -2023-02-13 18:01:57,859 - Epoch: [99][ 20/ 1207] Overall Loss 0.328806 Objective Loss 0.328806 LR 0.001000 Time 0.058874 -2023-02-13 18:01:58,046 - Epoch: [99][ 30/ 1207] Overall Loss 0.321940 Objective Loss 0.321940 LR 0.001000 Time 0.045499 -2023-02-13 18:01:58,234 - Epoch: [99][ 40/ 1207] Overall Loss 0.323335 Objective Loss 0.323335 LR 0.001000 Time 0.038810 -2023-02-13 18:01:58,422 - Epoch: [99][ 50/ 1207] Overall Loss 0.328014 Objective Loss 0.328014 LR 0.001000 Time 0.034797 -2023-02-13 18:01:58,609 - Epoch: [99][ 60/ 1207] Overall Loss 0.330152 Objective Loss 0.330152 LR 0.001000 Time 0.032107 -2023-02-13 18:01:58,796 - Epoch: [99][ 70/ 1207] Overall Loss 0.327480 Objective Loss 0.327480 LR 0.001000 Time 0.030188 -2023-02-13 18:01:58,983 - Epoch: [99][ 80/ 1207] Overall Loss 0.329347 Objective Loss 0.329347 LR 0.001000 Time 0.028749 -2023-02-13 18:01:59,171 - Epoch: [99][ 90/ 1207] Overall Loss 0.328531 Objective Loss 0.328531 LR 0.001000 Time 0.027639 -2023-02-13 18:01:59,358 - Epoch: [99][ 100/ 1207] Overall Loss 0.326548 Objective Loss 0.326548 LR 0.001000 Time 0.026742 -2023-02-13 18:01:59,546 - Epoch: [99][ 110/ 1207] Overall Loss 0.325958 Objective Loss 0.325958 LR 0.001000 Time 0.026015 -2023-02-13 18:01:59,733 - Epoch: [99][ 120/ 1207] Overall Loss 0.327222 Objective Loss 0.327222 LR 0.001000 Time 0.025404 -2023-02-13 18:01:59,920 - Epoch: [99][ 130/ 1207] Overall Loss 0.325922 Objective Loss 0.325922 LR 0.001000 Time 0.024887 -2023-02-13 18:02:00,108 - Epoch: [99][ 140/ 1207] Overall Loss 0.326922 Objective Loss 0.326922 LR 0.001000 Time 0.024449 -2023-02-13 18:02:00,296 - Epoch: [99][ 150/ 1207] Overall Loss 0.327904 Objective Loss 0.327904 LR 0.001000 Time 0.024066 -2023-02-13 18:02:00,483 - Epoch: [99][ 160/ 1207] Overall Loss 0.328929 Objective Loss 0.328929 LR 0.001000 Time 0.023732 -2023-02-13 18:02:00,671 - Epoch: [99][ 170/ 1207] Overall Loss 0.327200 Objective Loss 0.327200 LR 0.001000 Time 0.023438 -2023-02-13 18:02:00,859 - Epoch: [99][ 180/ 1207] Overall Loss 0.327656 Objective Loss 0.327656 LR 0.001000 Time 0.023178 -2023-02-13 18:02:01,046 - Epoch: [99][ 190/ 1207] Overall Loss 0.328061 Objective Loss 0.328061 LR 0.001000 Time 0.022942 -2023-02-13 18:02:01,234 - Epoch: [99][ 200/ 1207] Overall Loss 0.327574 Objective Loss 0.327574 LR 0.001000 Time 0.022731 -2023-02-13 18:02:01,422 - Epoch: [99][ 210/ 1207] Overall Loss 0.328561 Objective Loss 0.328561 LR 0.001000 Time 0.022541 -2023-02-13 18:02:01,610 - Epoch: [99][ 220/ 1207] Overall Loss 0.328365 Objective Loss 0.328365 LR 0.001000 Time 0.022369 -2023-02-13 18:02:01,798 - Epoch: [99][ 230/ 1207] Overall Loss 0.329547 Objective Loss 0.329547 LR 0.001000 Time 0.022212 -2023-02-13 18:02:01,986 - Epoch: [99][ 240/ 1207] Overall Loss 0.328175 Objective Loss 0.328175 LR 0.001000 Time 0.022072 -2023-02-13 18:02:02,175 - Epoch: [99][ 250/ 1207] Overall Loss 0.329391 Objective Loss 0.329391 LR 0.001000 Time 0.021940 -2023-02-13 18:02:02,362 - Epoch: [99][ 260/ 1207] Overall Loss 0.329049 Objective Loss 0.329049 LR 0.001000 Time 0.021817 -2023-02-13 18:02:02,551 - Epoch: [99][ 270/ 1207] Overall Loss 0.329222 Objective Loss 0.329222 LR 0.001000 Time 0.021706 -2023-02-13 18:02:02,738 - Epoch: [99][ 280/ 1207] Overall Loss 0.330534 Objective Loss 0.330534 LR 0.001000 Time 0.021599 -2023-02-13 18:02:02,926 - Epoch: [99][ 290/ 1207] Overall Loss 0.330650 Objective Loss 0.330650 LR 0.001000 Time 0.021500 -2023-02-13 18:02:03,113 - Epoch: [99][ 300/ 1207] Overall Loss 0.329663 Objective Loss 0.329663 LR 0.001000 Time 0.021406 -2023-02-13 18:02:03,301 - Epoch: [99][ 310/ 1207] Overall Loss 0.330293 Objective Loss 0.330293 LR 0.001000 Time 0.021320 -2023-02-13 18:02:03,488 - Epoch: [99][ 320/ 1207] Overall Loss 0.330216 Objective Loss 0.330216 LR 0.001000 Time 0.021238 -2023-02-13 18:02:03,676 - Epoch: [99][ 330/ 1207] Overall Loss 0.329930 Objective Loss 0.329930 LR 0.001000 Time 0.021162 -2023-02-13 18:02:03,863 - Epoch: [99][ 340/ 1207] Overall Loss 0.329908 Objective Loss 0.329908 LR 0.001000 Time 0.021089 -2023-02-13 18:02:04,050 - Epoch: [99][ 350/ 1207] Overall Loss 0.330003 Objective Loss 0.330003 LR 0.001000 Time 0.021021 -2023-02-13 18:02:04,239 - Epoch: [99][ 360/ 1207] Overall Loss 0.330812 Objective Loss 0.330812 LR 0.001000 Time 0.020959 -2023-02-13 18:02:04,427 - Epoch: [99][ 370/ 1207] Overall Loss 0.330717 Objective Loss 0.330717 LR 0.001000 Time 0.020899 -2023-02-13 18:02:04,614 - Epoch: [99][ 380/ 1207] Overall Loss 0.330582 Objective Loss 0.330582 LR 0.001000 Time 0.020840 -2023-02-13 18:02:04,801 - Epoch: [99][ 390/ 1207] Overall Loss 0.330107 Objective Loss 0.330107 LR 0.001000 Time 0.020785 -2023-02-13 18:02:04,989 - Epoch: [99][ 400/ 1207] Overall Loss 0.329967 Objective Loss 0.329967 LR 0.001000 Time 0.020735 -2023-02-13 18:02:05,177 - Epoch: [99][ 410/ 1207] Overall Loss 0.330665 Objective Loss 0.330665 LR 0.001000 Time 0.020688 -2023-02-13 18:02:05,365 - Epoch: [99][ 420/ 1207] Overall Loss 0.330532 Objective Loss 0.330532 LR 0.001000 Time 0.020640 -2023-02-13 18:02:05,553 - Epoch: [99][ 430/ 1207] Overall Loss 0.329720 Objective Loss 0.329720 LR 0.001000 Time 0.020597 -2023-02-13 18:02:05,741 - Epoch: [99][ 440/ 1207] Overall Loss 0.329774 Objective Loss 0.329774 LR 0.001000 Time 0.020554 -2023-02-13 18:02:05,929 - Epoch: [99][ 450/ 1207] Overall Loss 0.329433 Objective Loss 0.329433 LR 0.001000 Time 0.020516 -2023-02-13 18:02:06,118 - Epoch: [99][ 460/ 1207] Overall Loss 0.328925 Objective Loss 0.328925 LR 0.001000 Time 0.020479 -2023-02-13 18:02:06,306 - Epoch: [99][ 470/ 1207] Overall Loss 0.329564 Objective Loss 0.329564 LR 0.001000 Time 0.020442 -2023-02-13 18:02:06,493 - Epoch: [99][ 480/ 1207] Overall Loss 0.329676 Objective Loss 0.329676 LR 0.001000 Time 0.020407 -2023-02-13 18:02:06,681 - Epoch: [99][ 490/ 1207] Overall Loss 0.329902 Objective Loss 0.329902 LR 0.001000 Time 0.020374 -2023-02-13 18:02:06,870 - Epoch: [99][ 500/ 1207] Overall Loss 0.329626 Objective Loss 0.329626 LR 0.001000 Time 0.020342 -2023-02-13 18:02:07,057 - Epoch: [99][ 510/ 1207] Overall Loss 0.329497 Objective Loss 0.329497 LR 0.001000 Time 0.020310 -2023-02-13 18:02:07,245 - Epoch: [99][ 520/ 1207] Overall Loss 0.328611 Objective Loss 0.328611 LR 0.001000 Time 0.020281 -2023-02-13 18:02:07,433 - Epoch: [99][ 530/ 1207] Overall Loss 0.328347 Objective Loss 0.328347 LR 0.001000 Time 0.020252 -2023-02-13 18:02:07,621 - Epoch: [99][ 540/ 1207] Overall Loss 0.328360 Objective Loss 0.328360 LR 0.001000 Time 0.020223 -2023-02-13 18:02:07,809 - Epoch: [99][ 550/ 1207] Overall Loss 0.328276 Objective Loss 0.328276 LR 0.001000 Time 0.020196 -2023-02-13 18:02:07,997 - Epoch: [99][ 560/ 1207] Overall Loss 0.328336 Objective Loss 0.328336 LR 0.001000 Time 0.020172 -2023-02-13 18:02:08,185 - Epoch: [99][ 570/ 1207] Overall Loss 0.328613 Objective Loss 0.328613 LR 0.001000 Time 0.020147 -2023-02-13 18:02:08,372 - Epoch: [99][ 580/ 1207] Overall Loss 0.329302 Objective Loss 0.329302 LR 0.001000 Time 0.020122 -2023-02-13 18:02:08,560 - Epoch: [99][ 590/ 1207] Overall Loss 0.329087 Objective Loss 0.329087 LR 0.001000 Time 0.020099 -2023-02-13 18:02:08,748 - Epoch: [99][ 600/ 1207] Overall Loss 0.328979 Objective Loss 0.328979 LR 0.001000 Time 0.020075 -2023-02-13 18:02:08,936 - Epoch: [99][ 610/ 1207] Overall Loss 0.329272 Objective Loss 0.329272 LR 0.001000 Time 0.020053 -2023-02-13 18:02:09,123 - Epoch: [99][ 620/ 1207] Overall Loss 0.329207 Objective Loss 0.329207 LR 0.001000 Time 0.020032 -2023-02-13 18:02:09,311 - Epoch: [99][ 630/ 1207] Overall Loss 0.329008 Objective Loss 0.329008 LR 0.001000 Time 0.020012 -2023-02-13 18:02:09,499 - Epoch: [99][ 640/ 1207] Overall Loss 0.328938 Objective Loss 0.328938 LR 0.001000 Time 0.019992 -2023-02-13 18:02:09,686 - Epoch: [99][ 650/ 1207] Overall Loss 0.329126 Objective Loss 0.329126 LR 0.001000 Time 0.019972 -2023-02-13 18:02:09,874 - Epoch: [99][ 660/ 1207] Overall Loss 0.328894 Objective Loss 0.328894 LR 0.001000 Time 0.019953 -2023-02-13 18:02:10,062 - Epoch: [99][ 670/ 1207] Overall Loss 0.328865 Objective Loss 0.328865 LR 0.001000 Time 0.019935 -2023-02-13 18:02:10,249 - Epoch: [99][ 680/ 1207] Overall Loss 0.328738 Objective Loss 0.328738 LR 0.001000 Time 0.019918 -2023-02-13 18:02:10,437 - Epoch: [99][ 690/ 1207] Overall Loss 0.328880 Objective Loss 0.328880 LR 0.001000 Time 0.019900 -2023-02-13 18:02:10,624 - Epoch: [99][ 700/ 1207] Overall Loss 0.328355 Objective Loss 0.328355 LR 0.001000 Time 0.019883 -2023-02-13 18:02:10,812 - Epoch: [99][ 710/ 1207] Overall Loss 0.328285 Objective Loss 0.328285 LR 0.001000 Time 0.019867 -2023-02-13 18:02:11,003 - Epoch: [99][ 720/ 1207] Overall Loss 0.328581 Objective Loss 0.328581 LR 0.001000 Time 0.019855 -2023-02-13 18:02:11,190 - Epoch: [99][ 730/ 1207] Overall Loss 0.328400 Objective Loss 0.328400 LR 0.001000 Time 0.019839 -2023-02-13 18:02:11,377 - Epoch: [99][ 740/ 1207] Overall Loss 0.328236 Objective Loss 0.328236 LR 0.001000 Time 0.019824 -2023-02-13 18:02:11,565 - Epoch: [99][ 750/ 1207] Overall Loss 0.328513 Objective Loss 0.328513 LR 0.001000 Time 0.019810 -2023-02-13 18:02:11,753 - Epoch: [99][ 760/ 1207] Overall Loss 0.328368 Objective Loss 0.328368 LR 0.001000 Time 0.019795 -2023-02-13 18:02:11,941 - Epoch: [99][ 770/ 1207] Overall Loss 0.328094 Objective Loss 0.328094 LR 0.001000 Time 0.019782 -2023-02-13 18:02:12,129 - Epoch: [99][ 780/ 1207] Overall Loss 0.328323 Objective Loss 0.328323 LR 0.001000 Time 0.019769 -2023-02-13 18:02:12,317 - Epoch: [99][ 790/ 1207] Overall Loss 0.328691 Objective Loss 0.328691 LR 0.001000 Time 0.019756 -2023-02-13 18:02:12,504 - Epoch: [99][ 800/ 1207] Overall Loss 0.328653 Objective Loss 0.328653 LR 0.001000 Time 0.019742 -2023-02-13 18:02:12,692 - Epoch: [99][ 810/ 1207] Overall Loss 0.328480 Objective Loss 0.328480 LR 0.001000 Time 0.019730 -2023-02-13 18:02:12,880 - Epoch: [99][ 820/ 1207] Overall Loss 0.328897 Objective Loss 0.328897 LR 0.001000 Time 0.019718 -2023-02-13 18:02:13,068 - Epoch: [99][ 830/ 1207] Overall Loss 0.329218 Objective Loss 0.329218 LR 0.001000 Time 0.019708 -2023-02-13 18:02:13,257 - Epoch: [99][ 840/ 1207] Overall Loss 0.329180 Objective Loss 0.329180 LR 0.001000 Time 0.019697 -2023-02-13 18:02:13,445 - Epoch: [99][ 850/ 1207] Overall Loss 0.329581 Objective Loss 0.329581 LR 0.001000 Time 0.019686 -2023-02-13 18:02:13,633 - Epoch: [99][ 860/ 1207] Overall Loss 0.329782 Objective Loss 0.329782 LR 0.001000 Time 0.019676 -2023-02-13 18:02:13,821 - Epoch: [99][ 870/ 1207] Overall Loss 0.329896 Objective Loss 0.329896 LR 0.001000 Time 0.019665 -2023-02-13 18:02:14,011 - Epoch: [99][ 880/ 1207] Overall Loss 0.330188 Objective Loss 0.330188 LR 0.001000 Time 0.019657 -2023-02-13 18:02:14,200 - Epoch: [99][ 890/ 1207] Overall Loss 0.330232 Objective Loss 0.330232 LR 0.001000 Time 0.019648 -2023-02-13 18:02:14,388 - Epoch: [99][ 900/ 1207] Overall Loss 0.330735 Objective Loss 0.330735 LR 0.001000 Time 0.019638 -2023-02-13 18:02:14,576 - Epoch: [99][ 910/ 1207] Overall Loss 0.330714 Objective Loss 0.330714 LR 0.001000 Time 0.019629 -2023-02-13 18:02:14,764 - Epoch: [99][ 920/ 1207] Overall Loss 0.330858 Objective Loss 0.330858 LR 0.001000 Time 0.019619 -2023-02-13 18:02:14,953 - Epoch: [99][ 930/ 1207] Overall Loss 0.331287 Objective Loss 0.331287 LR 0.001000 Time 0.019611 -2023-02-13 18:02:15,141 - Epoch: [99][ 940/ 1207] Overall Loss 0.331710 Objective Loss 0.331710 LR 0.001000 Time 0.019602 -2023-02-13 18:02:15,330 - Epoch: [99][ 950/ 1207] Overall Loss 0.331970 Objective Loss 0.331970 LR 0.001000 Time 0.019594 -2023-02-13 18:02:15,518 - Epoch: [99][ 960/ 1207] Overall Loss 0.331897 Objective Loss 0.331897 LR 0.001000 Time 0.019586 -2023-02-13 18:02:15,706 - Epoch: [99][ 970/ 1207] Overall Loss 0.332047 Objective Loss 0.332047 LR 0.001000 Time 0.019577 -2023-02-13 18:02:15,895 - Epoch: [99][ 980/ 1207] Overall Loss 0.331623 Objective Loss 0.331623 LR 0.001000 Time 0.019570 -2023-02-13 18:02:16,083 - Epoch: [99][ 990/ 1207] Overall Loss 0.331359 Objective Loss 0.331359 LR 0.001000 Time 0.019562 -2023-02-13 18:02:16,271 - Epoch: [99][ 1000/ 1207] Overall Loss 0.331214 Objective Loss 0.331214 LR 0.001000 Time 0.019554 -2023-02-13 18:02:16,460 - Epoch: [99][ 1010/ 1207] Overall Loss 0.331222 Objective Loss 0.331222 LR 0.001000 Time 0.019546 -2023-02-13 18:02:16,648 - Epoch: [99][ 1020/ 1207] Overall Loss 0.331334 Objective Loss 0.331334 LR 0.001000 Time 0.019539 -2023-02-13 18:02:16,836 - Epoch: [99][ 1030/ 1207] Overall Loss 0.331330 Objective Loss 0.331330 LR 0.001000 Time 0.019532 -2023-02-13 18:02:17,026 - Epoch: [99][ 1040/ 1207] Overall Loss 0.331378 Objective Loss 0.331378 LR 0.001000 Time 0.019526 -2023-02-13 18:02:17,214 - Epoch: [99][ 1050/ 1207] Overall Loss 0.331351 Objective Loss 0.331351 LR 0.001000 Time 0.019519 -2023-02-13 18:02:17,403 - Epoch: [99][ 1060/ 1207] Overall Loss 0.331160 Objective Loss 0.331160 LR 0.001000 Time 0.019512 -2023-02-13 18:02:17,591 - Epoch: [99][ 1070/ 1207] Overall Loss 0.331214 Objective Loss 0.331214 LR 0.001000 Time 0.019506 -2023-02-13 18:02:17,779 - Epoch: [99][ 1080/ 1207] Overall Loss 0.331299 Objective Loss 0.331299 LR 0.001000 Time 0.019499 -2023-02-13 18:02:17,968 - Epoch: [99][ 1090/ 1207] Overall Loss 0.331254 Objective Loss 0.331254 LR 0.001000 Time 0.019492 -2023-02-13 18:02:18,156 - Epoch: [99][ 1100/ 1207] Overall Loss 0.330920 Objective Loss 0.330920 LR 0.001000 Time 0.019486 -2023-02-13 18:02:18,345 - Epoch: [99][ 1110/ 1207] Overall Loss 0.331016 Objective Loss 0.331016 LR 0.001000 Time 0.019480 -2023-02-13 18:02:18,533 - Epoch: [99][ 1120/ 1207] Overall Loss 0.330950 Objective Loss 0.330950 LR 0.001000 Time 0.019474 -2023-02-13 18:02:18,721 - Epoch: [99][ 1130/ 1207] Overall Loss 0.330861 Objective Loss 0.330861 LR 0.001000 Time 0.019468 -2023-02-13 18:02:18,909 - Epoch: [99][ 1140/ 1207] Overall Loss 0.330938 Objective Loss 0.330938 LR 0.001000 Time 0.019462 -2023-02-13 18:02:19,102 - Epoch: [99][ 1150/ 1207] Overall Loss 0.331435 Objective Loss 0.331435 LR 0.001000 Time 0.019460 -2023-02-13 18:02:19,295 - Epoch: [99][ 1160/ 1207] Overall Loss 0.331724 Objective Loss 0.331724 LR 0.001000 Time 0.019458 -2023-02-13 18:02:19,490 - Epoch: [99][ 1170/ 1207] Overall Loss 0.331943 Objective Loss 0.331943 LR 0.001000 Time 0.019458 -2023-02-13 18:02:19,684 - Epoch: [99][ 1180/ 1207] Overall Loss 0.332019 Objective Loss 0.332019 LR 0.001000 Time 0.019457 -2023-02-13 18:02:19,880 - Epoch: [99][ 1190/ 1207] Overall Loss 0.332237 Objective Loss 0.332237 LR 0.001000 Time 0.019458 -2023-02-13 18:02:20,132 - Epoch: [99][ 1200/ 1207] Overall Loss 0.332303 Objective Loss 0.332303 LR 0.001000 Time 0.019506 -2023-02-13 18:02:20,250 - Epoch: [99][ 1207/ 1207] Overall Loss 0.332643 Objective Loss 0.332643 Top1 81.097561 Top5 96.951220 LR 0.001000 Time 0.019490 -2023-02-13 18:02:20,330 - --- validate (epoch=99)----------- -2023-02-13 18:02:20,331 - 34311 samples (256 per mini-batch) -2023-02-13 18:02:20,737 - Epoch: [99][ 10/ 135] Loss 0.353220 Top1 81.601562 Top5 97.578125 -2023-02-13 18:02:20,870 - Epoch: [99][ 20/ 135] Loss 0.360206 Top1 81.425781 Top5 97.578125 -2023-02-13 18:02:20,993 - Epoch: [99][ 30/ 135] Loss 0.364905 Top1 81.145833 Top5 97.382812 -2023-02-13 18:02:21,133 - Epoch: [99][ 40/ 135] Loss 0.375108 Top1 80.996094 Top5 97.343750 -2023-02-13 18:02:21,261 - Epoch: [99][ 50/ 135] Loss 0.373304 Top1 80.968750 Top5 97.296875 -2023-02-13 18:02:21,391 - Epoch: [99][ 60/ 135] Loss 0.368391 Top1 81.210938 Top5 97.298177 -2023-02-13 18:02:21,520 - Epoch: [99][ 70/ 135] Loss 0.371562 Top1 81.210938 Top5 97.243304 -2023-02-13 18:02:21,650 - Epoch: [99][ 80/ 135] Loss 0.371157 Top1 81.420898 Top5 97.333984 -2023-02-13 18:02:21,779 - Epoch: [99][ 90/ 135] Loss 0.372390 Top1 81.427951 Top5 97.269965 -2023-02-13 18:02:21,905 - Epoch: [99][ 100/ 135] Loss 0.375058 Top1 81.324219 Top5 97.238281 -2023-02-13 18:02:22,032 - Epoch: [99][ 110/ 135] Loss 0.375060 Top1 81.335227 Top5 97.208807 -2023-02-13 18:02:22,160 - Epoch: [99][ 120/ 135] Loss 0.373952 Top1 81.435547 Top5 97.213542 -2023-02-13 18:02:22,291 - Epoch: [99][ 130/ 135] Loss 0.373402 Top1 81.370192 Top5 97.232572 -2023-02-13 18:02:22,338 - Epoch: [99][ 135/ 135] Loss 0.370657 Top1 81.350004 Top5 97.237038 -2023-02-13 18:02:22,407 - ==> Top1: 81.350 Top5: 97.237 Loss: 0.371 - -2023-02-13 18:02:22,408 - ==> Confusion: -[[ 843 6 6 3 11 5 0 1 7 48 2 6 1 6 4 5 1 1 3 3 5] - [ 4 918 0 3 7 41 4 25 5 0 0 3 1 2 2 0 3 0 7 1 7] - [ 9 6 933 14 3 2 27 21 0 1 3 3 4 3 2 7 1 2 9 4 4] - [ 8 2 16 857 1 7 6 2 2 2 14 3 5 0 30 3 4 7 39 0 8] - [ 17 11 4 3 970 12 1 3 2 5 0 5 2 1 6 8 6 1 1 3 5] - [ 2 18 1 4 3 960 4 21 4 2 2 14 2 14 0 1 4 1 4 4 5] - [ 2 6 18 4 1 7 1027 11 0 1 1 2 1 0 0 3 0 3 1 8 3] - [ 2 5 9 1 2 29 2 928 1 1 1 8 1 1 0 0 0 3 22 4 4] - [ 15 4 2 3 1 1 0 5 883 40 12 3 0 5 24 0 0 2 8 1 0] - [ 102 3 7 1 8 1 2 4 47 800 0 1 0 18 3 2 1 2 3 1 6] - [ 3 4 6 5 0 1 3 8 25 0 960 2 1 6 7 0 1 0 18 0 1] - [ 2 3 1 0 3 16 4 5 2 1 2 914 29 5 0 0 1 6 6 3 2] - [ 0 1 0 6 0 2 0 2 2 0 0 32 869 0 3 5 1 23 7 0 6] - [ 2 3 2 1 4 14 3 3 17 20 10 7 3 905 4 5 7 4 1 5 4] - [ 12 4 2 20 4 6 1 2 25 3 5 0 5 4 965 3 1 5 15 1 9] - [ 3 4 12 0 8 1 9 0 0 2 0 16 8 3 1 929 10 20 1 10 9] - [ 2 6 2 2 10 5 0 2 3 1 0 1 4 1 2 14 983 1 2 5 15] - [ 2 4 0 6 1 0 2 2 1 0 1 7 26 2 1 9 1 977 2 2 5] - [ 2 3 7 10 1 1 0 32 6 0 4 1 5 1 16 0 0 2 991 2 2] - [ 1 4 2 0 1 16 8 21 1 0 3 26 3 1 0 2 6 4 1 1037 11] - [ 174 250 216 155 131 295 101 262 149 84 235 150 371 322 203 91 270 135 293 284 9263]] - -2023-02-13 18:02:22,409 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:02:22,409 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:02:22,415 - - -2023-02-13 18:02:22,415 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:02:23,305 - Epoch: [100][ 10/ 1207] Overall Loss 0.296506 Objective Loss 0.296506 LR 0.000500 Time 0.088878 -2023-02-13 18:02:23,500 - Epoch: [100][ 20/ 1207] Overall Loss 0.301999 Objective Loss 0.301999 LR 0.000500 Time 0.054166 -2023-02-13 18:02:23,689 - Epoch: [100][ 30/ 1207] Overall Loss 0.323583 Objective Loss 0.323583 LR 0.000500 Time 0.042416 -2023-02-13 18:02:23,878 - Epoch: [100][ 40/ 1207] Overall Loss 0.326474 Objective Loss 0.326474 LR 0.000500 Time 0.036528 -2023-02-13 18:02:24,067 - Epoch: [100][ 50/ 1207] Overall Loss 0.326751 Objective Loss 0.326751 LR 0.000500 Time 0.033002 -2023-02-13 18:02:24,258 - Epoch: [100][ 60/ 1207] Overall Loss 0.329797 Objective Loss 0.329797 LR 0.000500 Time 0.030668 -2023-02-13 18:02:24,447 - Epoch: [100][ 70/ 1207] Overall Loss 0.326832 Objective Loss 0.326832 LR 0.000500 Time 0.028989 -2023-02-13 18:02:24,637 - Epoch: [100][ 80/ 1207] Overall Loss 0.327069 Objective Loss 0.327069 LR 0.000500 Time 0.027733 -2023-02-13 18:02:24,826 - Epoch: [100][ 90/ 1207] Overall Loss 0.325300 Objective Loss 0.325300 LR 0.000500 Time 0.026748 -2023-02-13 18:02:25,016 - Epoch: [100][ 100/ 1207] Overall Loss 0.322209 Objective Loss 0.322209 LR 0.000500 Time 0.025966 -2023-02-13 18:02:25,206 - Epoch: [100][ 110/ 1207] Overall Loss 0.322126 Objective Loss 0.322126 LR 0.000500 Time 0.025330 -2023-02-13 18:02:25,395 - Epoch: [100][ 120/ 1207] Overall Loss 0.318062 Objective Loss 0.318062 LR 0.000500 Time 0.024796 -2023-02-13 18:02:25,585 - Epoch: [100][ 130/ 1207] Overall Loss 0.318223 Objective Loss 0.318223 LR 0.000500 Time 0.024344 -2023-02-13 18:02:25,775 - Epoch: [100][ 140/ 1207] Overall Loss 0.316445 Objective Loss 0.316445 LR 0.000500 Time 0.023956 -2023-02-13 18:02:25,966 - Epoch: [100][ 150/ 1207] Overall Loss 0.316220 Objective Loss 0.316220 LR 0.000500 Time 0.023630 -2023-02-13 18:02:26,155 - Epoch: [100][ 160/ 1207] Overall Loss 0.313793 Objective Loss 0.313793 LR 0.000500 Time 0.023335 -2023-02-13 18:02:26,345 - Epoch: [100][ 170/ 1207] Overall Loss 0.310922 Objective Loss 0.310922 LR 0.000500 Time 0.023077 -2023-02-13 18:02:26,535 - Epoch: [100][ 180/ 1207] Overall Loss 0.312114 Objective Loss 0.312114 LR 0.000500 Time 0.022846 -2023-02-13 18:02:26,724 - Epoch: [100][ 190/ 1207] Overall Loss 0.312175 Objective Loss 0.312175 LR 0.000500 Time 0.022640 -2023-02-13 18:02:26,914 - Epoch: [100][ 200/ 1207] Overall Loss 0.312093 Objective Loss 0.312093 LR 0.000500 Time 0.022456 -2023-02-13 18:02:27,104 - Epoch: [100][ 210/ 1207] Overall Loss 0.310989 Objective Loss 0.310989 LR 0.000500 Time 0.022288 -2023-02-13 18:02:27,294 - Epoch: [100][ 220/ 1207] Overall Loss 0.309898 Objective Loss 0.309898 LR 0.000500 Time 0.022138 -2023-02-13 18:02:27,495 - Epoch: [100][ 230/ 1207] Overall Loss 0.309519 Objective Loss 0.309519 LR 0.000500 Time 0.022048 -2023-02-13 18:02:27,702 - Epoch: [100][ 240/ 1207] Overall Loss 0.310072 Objective Loss 0.310072 LR 0.000500 Time 0.021989 -2023-02-13 18:02:27,902 - Epoch: [100][ 250/ 1207] Overall Loss 0.310782 Objective Loss 0.310782 LR 0.000500 Time 0.021910 -2023-02-13 18:02:28,108 - Epoch: [100][ 260/ 1207] Overall Loss 0.311910 Objective Loss 0.311910 LR 0.000500 Time 0.021855 -2023-02-13 18:02:28,309 - Epoch: [100][ 270/ 1207] Overall Loss 0.310797 Objective Loss 0.310797 LR 0.000500 Time 0.021790 -2023-02-13 18:02:28,515 - Epoch: [100][ 280/ 1207] Overall Loss 0.310452 Objective Loss 0.310452 LR 0.000500 Time 0.021746 -2023-02-13 18:02:28,716 - Epoch: [100][ 290/ 1207] Overall Loss 0.310023 Objective Loss 0.310023 LR 0.000500 Time 0.021688 -2023-02-13 18:02:28,922 - Epoch: [100][ 300/ 1207] Overall Loss 0.309970 Objective Loss 0.309970 LR 0.000500 Time 0.021650 -2023-02-13 18:02:29,123 - Epoch: [100][ 310/ 1207] Overall Loss 0.309458 Objective Loss 0.309458 LR 0.000500 Time 0.021600 -2023-02-13 18:02:29,330 - Epoch: [100][ 320/ 1207] Overall Loss 0.308963 Objective Loss 0.308963 LR 0.000500 Time 0.021569 -2023-02-13 18:02:29,531 - Epoch: [100][ 330/ 1207] Overall Loss 0.307922 Objective Loss 0.307922 LR 0.000500 Time 0.021523 -2023-02-13 18:02:29,737 - Epoch: [100][ 340/ 1207] Overall Loss 0.307227 Objective Loss 0.307227 LR 0.000500 Time 0.021496 -2023-02-13 18:02:29,938 - Epoch: [100][ 350/ 1207] Overall Loss 0.307390 Objective Loss 0.307390 LR 0.000500 Time 0.021455 -2023-02-13 18:02:30,143 - Epoch: [100][ 360/ 1207] Overall Loss 0.306752 Objective Loss 0.306752 LR 0.000500 Time 0.021428 -2023-02-13 18:02:30,344 - Epoch: [100][ 370/ 1207] Overall Loss 0.305509 Objective Loss 0.305509 LR 0.000500 Time 0.021390 -2023-02-13 18:02:30,550 - Epoch: [100][ 380/ 1207] Overall Loss 0.304944 Objective Loss 0.304944 LR 0.000500 Time 0.021370 -2023-02-13 18:02:30,751 - Epoch: [100][ 390/ 1207] Overall Loss 0.304995 Objective Loss 0.304995 LR 0.000500 Time 0.021335 -2023-02-13 18:02:30,957 - Epoch: [100][ 400/ 1207] Overall Loss 0.305486 Objective Loss 0.305486 LR 0.000500 Time 0.021317 -2023-02-13 18:02:31,159 - Epoch: [100][ 410/ 1207] Overall Loss 0.304961 Objective Loss 0.304961 LR 0.000500 Time 0.021288 -2023-02-13 18:02:31,366 - Epoch: [100][ 420/ 1207] Overall Loss 0.304633 Objective Loss 0.304633 LR 0.000500 Time 0.021272 -2023-02-13 18:02:31,567 - Epoch: [100][ 430/ 1207] Overall Loss 0.304083 Objective Loss 0.304083 LR 0.000500 Time 0.021245 -2023-02-13 18:02:31,773 - Epoch: [100][ 440/ 1207] Overall Loss 0.304083 Objective Loss 0.304083 LR 0.000500 Time 0.021230 -2023-02-13 18:02:31,975 - Epoch: [100][ 450/ 1207] Overall Loss 0.303713 Objective Loss 0.303713 LR 0.000500 Time 0.021205 -2023-02-13 18:02:32,180 - Epoch: [100][ 460/ 1207] Overall Loss 0.303517 Objective Loss 0.303517 LR 0.000500 Time 0.021189 -2023-02-13 18:02:32,381 - Epoch: [100][ 470/ 1207] Overall Loss 0.303170 Objective Loss 0.303170 LR 0.000500 Time 0.021165 -2023-02-13 18:02:32,587 - Epoch: [100][ 480/ 1207] Overall Loss 0.303179 Objective Loss 0.303179 LR 0.000500 Time 0.021154 -2023-02-13 18:02:32,788 - Epoch: [100][ 490/ 1207] Overall Loss 0.302398 Objective Loss 0.302398 LR 0.000500 Time 0.021130 -2023-02-13 18:02:32,994 - Epoch: [100][ 500/ 1207] Overall Loss 0.301308 Objective Loss 0.301308 LR 0.000500 Time 0.021119 -2023-02-13 18:02:33,195 - Epoch: [100][ 510/ 1207] Overall Loss 0.301142 Objective Loss 0.301142 LR 0.000500 Time 0.021099 -2023-02-13 18:02:33,402 - Epoch: [100][ 520/ 1207] Overall Loss 0.301062 Objective Loss 0.301062 LR 0.000500 Time 0.021090 -2023-02-13 18:02:33,604 - Epoch: [100][ 530/ 1207] Overall Loss 0.301226 Objective Loss 0.301226 LR 0.000500 Time 0.021071 -2023-02-13 18:02:33,809 - Epoch: [100][ 540/ 1207] Overall Loss 0.301032 Objective Loss 0.301032 LR 0.000500 Time 0.021062 -2023-02-13 18:02:34,010 - Epoch: [100][ 550/ 1207] Overall Loss 0.300357 Objective Loss 0.300357 LR 0.000500 Time 0.021043 -2023-02-13 18:02:34,217 - Epoch: [100][ 560/ 1207] Overall Loss 0.299772 Objective Loss 0.299772 LR 0.000500 Time 0.021036 -2023-02-13 18:02:34,419 - Epoch: [100][ 570/ 1207] Overall Loss 0.299437 Objective Loss 0.299437 LR 0.000500 Time 0.021021 -2023-02-13 18:02:34,625 - Epoch: [100][ 580/ 1207] Overall Loss 0.299300 Objective Loss 0.299300 LR 0.000500 Time 0.021013 -2023-02-13 18:02:34,826 - Epoch: [100][ 590/ 1207] Overall Loss 0.299423 Objective Loss 0.299423 LR 0.000500 Time 0.020997 -2023-02-13 18:02:35,033 - Epoch: [100][ 600/ 1207] Overall Loss 0.299013 Objective Loss 0.299013 LR 0.000500 Time 0.020990 -2023-02-13 18:02:35,233 - Epoch: [100][ 610/ 1207] Overall Loss 0.299114 Objective Loss 0.299114 LR 0.000500 Time 0.020974 -2023-02-13 18:02:35,439 - Epoch: [100][ 620/ 1207] Overall Loss 0.299139 Objective Loss 0.299139 LR 0.000500 Time 0.020968 -2023-02-13 18:02:35,640 - Epoch: [100][ 630/ 1207] Overall Loss 0.299464 Objective Loss 0.299464 LR 0.000500 Time 0.020953 -2023-02-13 18:02:35,847 - Epoch: [100][ 640/ 1207] Overall Loss 0.299252 Objective Loss 0.299252 LR 0.000500 Time 0.020948 -2023-02-13 18:02:36,048 - Epoch: [100][ 650/ 1207] Overall Loss 0.299142 Objective Loss 0.299142 LR 0.000500 Time 0.020935 -2023-02-13 18:02:36,254 - Epoch: [100][ 660/ 1207] Overall Loss 0.299007 Objective Loss 0.299007 LR 0.000500 Time 0.020929 -2023-02-13 18:02:36,456 - Epoch: [100][ 670/ 1207] Overall Loss 0.298730 Objective Loss 0.298730 LR 0.000500 Time 0.020917 -2023-02-13 18:02:36,662 - Epoch: [100][ 680/ 1207] Overall Loss 0.298475 Objective Loss 0.298475 LR 0.000500 Time 0.020912 -2023-02-13 18:02:36,863 - Epoch: [100][ 690/ 1207] Overall Loss 0.298195 Objective Loss 0.298195 LR 0.000500 Time 0.020900 -2023-02-13 18:02:37,069 - Epoch: [100][ 700/ 1207] Overall Loss 0.297931 Objective Loss 0.297931 LR 0.000500 Time 0.020895 -2023-02-13 18:02:37,270 - Epoch: [100][ 710/ 1207] Overall Loss 0.297474 Objective Loss 0.297474 LR 0.000500 Time 0.020884 -2023-02-13 18:02:37,477 - Epoch: [100][ 720/ 1207] Overall Loss 0.296855 Objective Loss 0.296855 LR 0.000500 Time 0.020881 -2023-02-13 18:02:37,678 - Epoch: [100][ 730/ 1207] Overall Loss 0.296599 Objective Loss 0.296599 LR 0.000500 Time 0.020869 -2023-02-13 18:02:37,884 - Epoch: [100][ 740/ 1207] Overall Loss 0.296521 Objective Loss 0.296521 LR 0.000500 Time 0.020865 -2023-02-13 18:02:38,086 - Epoch: [100][ 750/ 1207] Overall Loss 0.296125 Objective Loss 0.296125 LR 0.000500 Time 0.020855 -2023-02-13 18:02:38,292 - Epoch: [100][ 760/ 1207] Overall Loss 0.295558 Objective Loss 0.295558 LR 0.000500 Time 0.020852 -2023-02-13 18:02:38,493 - Epoch: [100][ 770/ 1207] Overall Loss 0.295199 Objective Loss 0.295199 LR 0.000500 Time 0.020842 -2023-02-13 18:02:38,699 - Epoch: [100][ 780/ 1207] Overall Loss 0.294983 Objective Loss 0.294983 LR 0.000500 Time 0.020838 -2023-02-13 18:02:38,900 - Epoch: [100][ 790/ 1207] Overall Loss 0.294254 Objective Loss 0.294254 LR 0.000500 Time 0.020828 -2023-02-13 18:02:39,106 - Epoch: [100][ 800/ 1207] Overall Loss 0.294145 Objective Loss 0.294145 LR 0.000500 Time 0.020824 -2023-02-13 18:02:39,305 - Epoch: [100][ 810/ 1207] Overall Loss 0.294054 Objective Loss 0.294054 LR 0.000500 Time 0.020813 -2023-02-13 18:02:39,511 - Epoch: [100][ 820/ 1207] Overall Loss 0.294167 Objective Loss 0.294167 LR 0.000500 Time 0.020810 -2023-02-13 18:02:39,712 - Epoch: [100][ 830/ 1207] Overall Loss 0.294620 Objective Loss 0.294620 LR 0.000500 Time 0.020800 -2023-02-13 18:02:39,917 - Epoch: [100][ 840/ 1207] Overall Loss 0.294509 Objective Loss 0.294509 LR 0.000500 Time 0.020796 -2023-02-13 18:02:40,117 - Epoch: [100][ 850/ 1207] Overall Loss 0.294240 Objective Loss 0.294240 LR 0.000500 Time 0.020787 -2023-02-13 18:02:40,323 - Epoch: [100][ 860/ 1207] Overall Loss 0.294740 Objective Loss 0.294740 LR 0.000500 Time 0.020784 -2023-02-13 18:02:40,520 - Epoch: [100][ 870/ 1207] Overall Loss 0.294271 Objective Loss 0.294271 LR 0.000500 Time 0.020772 -2023-02-13 18:02:40,717 - Epoch: [100][ 880/ 1207] Overall Loss 0.293935 Objective Loss 0.293935 LR 0.000500 Time 0.020759 -2023-02-13 18:02:40,913 - Epoch: [100][ 890/ 1207] Overall Loss 0.294067 Objective Loss 0.294067 LR 0.000500 Time 0.020746 -2023-02-13 18:02:41,111 - Epoch: [100][ 900/ 1207] Overall Loss 0.293832 Objective Loss 0.293832 LR 0.000500 Time 0.020734 -2023-02-13 18:02:41,305 - Epoch: [100][ 910/ 1207] Overall Loss 0.293817 Objective Loss 0.293817 LR 0.000500 Time 0.020720 -2023-02-13 18:02:41,503 - Epoch: [100][ 920/ 1207] Overall Loss 0.293456 Objective Loss 0.293456 LR 0.000500 Time 0.020709 -2023-02-13 18:02:41,698 - Epoch: [100][ 930/ 1207] Overall Loss 0.293559 Objective Loss 0.293559 LR 0.000500 Time 0.020695 -2023-02-13 18:02:41,895 - Epoch: [100][ 940/ 1207] Overall Loss 0.293434 Objective Loss 0.293434 LR 0.000500 Time 0.020684 -2023-02-13 18:02:42,090 - Epoch: [100][ 950/ 1207] Overall Loss 0.293339 Objective Loss 0.293339 LR 0.000500 Time 0.020672 -2023-02-13 18:02:42,288 - Epoch: [100][ 960/ 1207] Overall Loss 0.293116 Objective Loss 0.293116 LR 0.000500 Time 0.020662 -2023-02-13 18:02:42,483 - Epoch: [100][ 970/ 1207] Overall Loss 0.293123 Objective Loss 0.293123 LR 0.000500 Time 0.020650 -2023-02-13 18:02:42,680 - Epoch: [100][ 980/ 1207] Overall Loss 0.293375 Objective Loss 0.293375 LR 0.000500 Time 0.020640 -2023-02-13 18:02:42,876 - Epoch: [100][ 990/ 1207] Overall Loss 0.293764 Objective Loss 0.293764 LR 0.000500 Time 0.020629 -2023-02-13 18:02:43,074 - Epoch: [100][ 1000/ 1207] Overall Loss 0.293631 Objective Loss 0.293631 LR 0.000500 Time 0.020620 -2023-02-13 18:02:43,269 - Epoch: [100][ 1010/ 1207] Overall Loss 0.293307 Objective Loss 0.293307 LR 0.000500 Time 0.020609 -2023-02-13 18:02:43,466 - Epoch: [100][ 1020/ 1207] Overall Loss 0.293196 Objective Loss 0.293196 LR 0.000500 Time 0.020600 -2023-02-13 18:02:43,661 - Epoch: [100][ 1030/ 1207] Overall Loss 0.293073 Objective Loss 0.293073 LR 0.000500 Time 0.020588 -2023-02-13 18:02:43,858 - Epoch: [100][ 1040/ 1207] Overall Loss 0.293088 Objective Loss 0.293088 LR 0.000500 Time 0.020579 -2023-02-13 18:02:44,053 - Epoch: [100][ 1050/ 1207] Overall Loss 0.293050 Objective Loss 0.293050 LR 0.000500 Time 0.020569 -2023-02-13 18:02:44,250 - Epoch: [100][ 1060/ 1207] Overall Loss 0.292891 Objective Loss 0.292891 LR 0.000500 Time 0.020560 -2023-02-13 18:02:44,445 - Epoch: [100][ 1070/ 1207] Overall Loss 0.293039 Objective Loss 0.293039 LR 0.000500 Time 0.020550 -2023-02-13 18:02:44,642 - Epoch: [100][ 1080/ 1207] Overall Loss 0.293079 Objective Loss 0.293079 LR 0.000500 Time 0.020542 -2023-02-13 18:02:44,837 - Epoch: [100][ 1090/ 1207] Overall Loss 0.292867 Objective Loss 0.292867 LR 0.000500 Time 0.020532 -2023-02-13 18:02:45,034 - Epoch: [100][ 1100/ 1207] Overall Loss 0.292859 Objective Loss 0.292859 LR 0.000500 Time 0.020524 -2023-02-13 18:02:45,229 - Epoch: [100][ 1110/ 1207] Overall Loss 0.292752 Objective Loss 0.292752 LR 0.000500 Time 0.020514 -2023-02-13 18:02:45,426 - Epoch: [100][ 1120/ 1207] Overall Loss 0.292730 Objective Loss 0.292730 LR 0.000500 Time 0.020507 -2023-02-13 18:02:45,621 - Epoch: [100][ 1130/ 1207] Overall Loss 0.292613 Objective Loss 0.292613 LR 0.000500 Time 0.020498 -2023-02-13 18:02:45,817 - Epoch: [100][ 1140/ 1207] Overall Loss 0.292734 Objective Loss 0.292734 LR 0.000500 Time 0.020489 -2023-02-13 18:02:46,013 - Epoch: [100][ 1150/ 1207] Overall Loss 0.292683 Objective Loss 0.292683 LR 0.000500 Time 0.020481 -2023-02-13 18:02:46,209 - Epoch: [100][ 1160/ 1207] Overall Loss 0.292751 Objective Loss 0.292751 LR 0.000500 Time 0.020474 -2023-02-13 18:02:46,405 - Epoch: [100][ 1170/ 1207] Overall Loss 0.292196 Objective Loss 0.292196 LR 0.000500 Time 0.020465 -2023-02-13 18:02:46,602 - Epoch: [100][ 1180/ 1207] Overall Loss 0.291919 Objective Loss 0.291919 LR 0.000500 Time 0.020458 -2023-02-13 18:02:46,795 - Epoch: [100][ 1190/ 1207] Overall Loss 0.291756 Objective Loss 0.291756 LR 0.000500 Time 0.020448 -2023-02-13 18:02:47,039 - Epoch: [100][ 1200/ 1207] Overall Loss 0.291864 Objective Loss 0.291864 LR 0.000500 Time 0.020481 -2023-02-13 18:02:47,153 - Epoch: [100][ 1207/ 1207] Overall Loss 0.292038 Objective Loss 0.292038 Top1 80.792683 Top5 98.170732 LR 0.000500 Time 0.020457 -2023-02-13 18:02:47,225 - --- validate (epoch=100)----------- -2023-02-13 18:02:47,225 - 34311 samples (256 per mini-batch) -2023-02-13 18:02:47,637 - Epoch: [100][ 10/ 135] Loss 0.349402 Top1 81.210938 Top5 97.812500 -2023-02-13 18:02:47,780 - Epoch: [100][ 20/ 135] Loss 0.330420 Top1 82.167969 Top5 97.578125 -2023-02-13 18:02:47,924 - Epoch: [100][ 30/ 135] Loss 0.328346 Top1 82.591146 Top5 97.578125 -2023-02-13 18:02:48,068 - Epoch: [100][ 40/ 135] Loss 0.337897 Top1 82.402344 Top5 97.314453 -2023-02-13 18:02:48,209 - Epoch: [100][ 50/ 135] Loss 0.339191 Top1 82.437500 Top5 97.234375 -2023-02-13 18:02:48,350 - Epoch: [100][ 60/ 135] Loss 0.338099 Top1 82.526042 Top5 97.285156 -2023-02-13 18:02:48,489 - Epoch: [100][ 70/ 135] Loss 0.335455 Top1 82.410714 Top5 97.310268 -2023-02-13 18:02:48,634 - Epoch: [100][ 80/ 135] Loss 0.335720 Top1 82.456055 Top5 97.304688 -2023-02-13 18:02:48,772 - Epoch: [100][ 90/ 135] Loss 0.335963 Top1 82.317708 Top5 97.313368 -2023-02-13 18:02:48,917 - Epoch: [100][ 100/ 135] Loss 0.336555 Top1 82.308594 Top5 97.347656 -2023-02-13 18:02:49,057 - Epoch: [100][ 110/ 135] Loss 0.335948 Top1 82.411222 Top5 97.336648 -2023-02-13 18:02:49,201 - Epoch: [100][ 120/ 135] Loss 0.335664 Top1 82.434896 Top5 97.347005 -2023-02-13 18:02:49,328 - Epoch: [100][ 130/ 135] Loss 0.335209 Top1 82.460938 Top5 97.301683 -2023-02-13 18:02:49,372 - Epoch: [100][ 135/ 135] Loss 0.334566 Top1 82.428376 Top5 97.289499 -2023-02-13 18:02:49,456 - ==> Top1: 82.428 Top5: 97.289 Loss: 0.335 - -2023-02-13 18:02:49,456 - ==> Confusion: -[[ 837 6 5 3 13 4 1 0 3 54 0 8 0 5 10 1 4 0 3 3 7] - [ 3 950 0 5 10 24 3 7 2 0 3 5 1 0 1 0 5 1 7 1 5] - [ 5 6 939 15 4 1 21 16 0 1 4 2 2 5 5 10 2 3 10 5 2] - [ 4 1 20 898 4 6 2 2 0 1 14 0 9 0 26 4 2 3 16 1 3] - [ 12 10 1 0 985 7 2 1 0 4 1 10 2 1 9 4 9 0 0 4 4] - [ 2 31 0 8 7 947 2 14 1 2 2 18 3 12 1 2 6 0 1 6 5] - [ 3 6 13 1 2 5 1034 4 0 0 3 4 1 0 0 5 0 2 0 12 4] - [ 0 20 8 1 4 29 8 892 1 2 2 8 4 1 1 0 0 1 22 14 6] - [ 18 3 1 3 1 0 1 4 871 40 10 2 0 12 31 2 0 2 7 0 1] - [ 73 1 2 0 9 1 0 4 27 861 1 1 1 17 4 1 2 1 1 1 4] - [ 1 0 5 10 1 4 2 3 14 0 972 3 1 13 5 0 0 1 14 1 1] - [ 1 2 1 0 4 13 0 8 1 1 0 921 20 5 3 4 2 9 2 7 1] - [ 1 0 0 6 4 2 1 2 2 0 0 48 857 1 3 5 3 11 2 1 10] - [ 5 2 2 0 8 10 0 1 12 20 7 10 4 920 7 3 5 1 1 1 5] - [ 6 2 2 15 4 4 0 2 16 4 3 2 3 0 1007 2 0 6 11 0 3] - [ 3 3 4 0 11 1 4 1 1 0 0 9 7 4 0 969 6 10 0 6 7] - [ 0 10 0 2 9 2 0 0 1 0 1 3 0 4 3 13 993 2 4 8 6] - [ 6 3 2 2 1 1 2 2 0 1 0 11 24 3 0 18 0 965 0 4 6] - [ 5 4 6 12 2 2 0 17 1 0 6 3 5 1 15 2 0 0 1001 3 1] - [ 0 4 1 1 0 6 8 12 0 0 0 20 3 6 0 5 5 2 1 1065 9] - [ 149 318 226 163 182 217 106 169 105 105 244 144 304 359 242 140 260 91 196 316 9398]] - -2023-02-13 18:02:49,458 - ==> Best [Top1: 82.976 Top5: 97.278 Sparsity:0.00 Params: 148928 on epoch: 87] -2023-02-13 18:02:49,458 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:02:49,463 - - -2023-02-13 18:02:49,464 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:02:50,330 - Epoch: [101][ 10/ 1207] Overall Loss 0.291124 Objective Loss 0.291124 LR 0.000500 Time 0.086597 -2023-02-13 18:02:50,530 - Epoch: [101][ 20/ 1207] Overall Loss 0.285362 Objective Loss 0.285362 LR 0.000500 Time 0.053264 -2023-02-13 18:02:50,719 - Epoch: [101][ 30/ 1207] Overall Loss 0.276283 Objective Loss 0.276283 LR 0.000500 Time 0.041800 -2023-02-13 18:02:50,910 - Epoch: [101][ 40/ 1207] Overall Loss 0.280576 Objective Loss 0.280576 LR 0.000500 Time 0.036107 -2023-02-13 18:02:51,099 - Epoch: [101][ 50/ 1207] Overall Loss 0.277986 Objective Loss 0.277986 LR 0.000500 Time 0.032669 -2023-02-13 18:02:51,288 - Epoch: [101][ 60/ 1207] Overall Loss 0.276071 Objective Loss 0.276071 LR 0.000500 Time 0.030369 -2023-02-13 18:02:51,478 - Epoch: [101][ 70/ 1207] Overall Loss 0.275629 Objective Loss 0.275629 LR 0.000500 Time 0.028735 -2023-02-13 18:02:51,667 - Epoch: [101][ 80/ 1207] Overall Loss 0.275267 Objective Loss 0.275267 LR 0.000500 Time 0.027497 -2023-02-13 18:02:51,856 - Epoch: [101][ 90/ 1207] Overall Loss 0.274675 Objective Loss 0.274675 LR 0.000500 Time 0.026541 -2023-02-13 18:02:52,045 - Epoch: [101][ 100/ 1207] Overall Loss 0.273349 Objective Loss 0.273349 LR 0.000500 Time 0.025775 -2023-02-13 18:02:52,235 - Epoch: [101][ 110/ 1207] Overall Loss 0.274849 Objective Loss 0.274849 LR 0.000500 Time 0.025153 -2023-02-13 18:02:52,425 - Epoch: [101][ 120/ 1207] Overall Loss 0.276420 Objective Loss 0.276420 LR 0.000500 Time 0.024634 -2023-02-13 18:02:52,613 - Epoch: [101][ 130/ 1207] Overall Loss 0.278397 Objective Loss 0.278397 LR 0.000500 Time 0.024187 -2023-02-13 18:02:52,802 - Epoch: [101][ 140/ 1207] Overall Loss 0.278612 Objective Loss 0.278612 LR 0.000500 Time 0.023805 -2023-02-13 18:02:52,991 - Epoch: [101][ 150/ 1207] Overall Loss 0.277711 Objective Loss 0.277711 LR 0.000500 Time 0.023478 -2023-02-13 18:02:53,181 - Epoch: [101][ 160/ 1207] Overall Loss 0.278998 Objective Loss 0.278998 LR 0.000500 Time 0.023195 -2023-02-13 18:02:53,372 - Epoch: [101][ 170/ 1207] Overall Loss 0.276744 Objective Loss 0.276744 LR 0.000500 Time 0.022949 -2023-02-13 18:02:53,562 - Epoch: [101][ 180/ 1207] Overall Loss 0.276763 Objective Loss 0.276763 LR 0.000500 Time 0.022727 -2023-02-13 18:02:53,751 - Epoch: [101][ 190/ 1207] Overall Loss 0.277562 Objective Loss 0.277562 LR 0.000500 Time 0.022527 -2023-02-13 18:02:53,941 - Epoch: [101][ 200/ 1207] Overall Loss 0.277195 Objective Loss 0.277195 LR 0.000500 Time 0.022350 -2023-02-13 18:02:54,131 - Epoch: [101][ 210/ 1207] Overall Loss 0.277534 Objective Loss 0.277534 LR 0.000500 Time 0.022186 -2023-02-13 18:02:54,321 - Epoch: [101][ 220/ 1207] Overall Loss 0.277069 Objective Loss 0.277069 LR 0.000500 Time 0.022040 -2023-02-13 18:02:54,511 - Epoch: [101][ 230/ 1207] Overall Loss 0.277593 Objective Loss 0.277593 LR 0.000500 Time 0.021905 -2023-02-13 18:02:54,701 - Epoch: [101][ 240/ 1207] Overall Loss 0.277454 Objective Loss 0.277454 LR 0.000500 Time 0.021782 -2023-02-13 18:02:54,891 - Epoch: [101][ 250/ 1207] Overall Loss 0.276971 Objective Loss 0.276971 LR 0.000500 Time 0.021669 -2023-02-13 18:02:55,081 - Epoch: [101][ 260/ 1207] Overall Loss 0.276697 Objective Loss 0.276697 LR 0.000500 Time 0.021567 -2023-02-13 18:02:55,271 - Epoch: [101][ 270/ 1207] Overall Loss 0.277881 Objective Loss 0.277881 LR 0.000500 Time 0.021470 -2023-02-13 18:02:55,461 - Epoch: [101][ 280/ 1207] Overall Loss 0.277988 Objective Loss 0.277988 LR 0.000500 Time 0.021382 -2023-02-13 18:02:55,651 - Epoch: [101][ 290/ 1207] Overall Loss 0.278060 Objective Loss 0.278060 LR 0.000500 Time 0.021298 -2023-02-13 18:02:55,841 - Epoch: [101][ 300/ 1207] Overall Loss 0.277060 Objective Loss 0.277060 LR 0.000500 Time 0.021218 -2023-02-13 18:02:56,031 - Epoch: [101][ 310/ 1207] Overall Loss 0.277093 Objective Loss 0.277093 LR 0.000500 Time 0.021147 -2023-02-13 18:02:56,221 - Epoch: [101][ 320/ 1207] Overall Loss 0.278674 Objective Loss 0.278674 LR 0.000500 Time 0.021077 -2023-02-13 18:02:56,411 - Epoch: [101][ 330/ 1207] Overall Loss 0.278123 Objective Loss 0.278123 LR 0.000500 Time 0.021015 -2023-02-13 18:02:56,602 - Epoch: [101][ 340/ 1207] Overall Loss 0.278097 Objective Loss 0.278097 LR 0.000500 Time 0.020956 -2023-02-13 18:02:56,791 - Epoch: [101][ 350/ 1207] Overall Loss 0.278550 Objective Loss 0.278550 LR 0.000500 Time 0.020897 -2023-02-13 18:02:56,981 - Epoch: [101][ 360/ 1207] Overall Loss 0.278587 Objective Loss 0.278587 LR 0.000500 Time 0.020844 -2023-02-13 18:02:57,171 - Epoch: [101][ 370/ 1207] Overall Loss 0.278737 Objective Loss 0.278737 LR 0.000500 Time 0.020793 -2023-02-13 18:02:57,362 - Epoch: [101][ 380/ 1207] Overall Loss 0.278606 Objective Loss 0.278606 LR 0.000500 Time 0.020746 -2023-02-13 18:02:57,552 - Epoch: [101][ 390/ 1207] Overall Loss 0.279168 Objective Loss 0.279168 LR 0.000500 Time 0.020700 -2023-02-13 18:02:57,741 - Epoch: [101][ 400/ 1207] Overall Loss 0.279361 Objective Loss 0.279361 LR 0.000500 Time 0.020656 -2023-02-13 18:02:57,932 - Epoch: [101][ 410/ 1207] Overall Loss 0.279756 Objective Loss 0.279756 LR 0.000500 Time 0.020615 -2023-02-13 18:02:58,121 - Epoch: [101][ 420/ 1207] Overall Loss 0.279589 Objective Loss 0.279589 LR 0.000500 Time 0.020574 -2023-02-13 18:02:58,312 - Epoch: [101][ 430/ 1207] Overall Loss 0.279881 Objective Loss 0.279881 LR 0.000500 Time 0.020538 -2023-02-13 18:02:58,502 - Epoch: [101][ 440/ 1207] Overall Loss 0.279532 Objective Loss 0.279532 LR 0.000500 Time 0.020504 -2023-02-13 18:02:58,692 - Epoch: [101][ 450/ 1207] Overall Loss 0.279174 Objective Loss 0.279174 LR 0.000500 Time 0.020469 -2023-02-13 18:02:58,884 - Epoch: [101][ 460/ 1207] Overall Loss 0.279514 Objective Loss 0.279514 LR 0.000500 Time 0.020441 -2023-02-13 18:02:59,080 - Epoch: [101][ 470/ 1207] Overall Loss 0.279793 Objective Loss 0.279793 LR 0.000500 Time 0.020422 -2023-02-13 18:02:59,278 - Epoch: [101][ 480/ 1207] Overall Loss 0.279798 Objective Loss 0.279798 LR 0.000500 Time 0.020409 -2023-02-13 18:02:59,475 - Epoch: [101][ 490/ 1207] Overall Loss 0.280165 Objective Loss 0.280165 LR 0.000500 Time 0.020394 -2023-02-13 18:02:59,674 - Epoch: [101][ 500/ 1207] Overall Loss 0.280005 Objective Loss 0.280005 LR 0.000500 Time 0.020382 -2023-02-13 18:02:59,870 - Epoch: [101][ 510/ 1207] Overall Loss 0.279529 Objective Loss 0.279529 LR 0.000500 Time 0.020367 -2023-02-13 18:03:00,068 - Epoch: [101][ 520/ 1207] Overall Loss 0.280248 Objective Loss 0.280248 LR 0.000500 Time 0.020355 -2023-02-13 18:03:00,264 - Epoch: [101][ 530/ 1207] Overall Loss 0.280241 Objective Loss 0.280241 LR 0.000500 Time 0.020340 -2023-02-13 18:03:00,463 - Epoch: [101][ 540/ 1207] Overall Loss 0.280259 Objective Loss 0.280259 LR 0.000500 Time 0.020331 -2023-02-13 18:03:00,659 - Epoch: [101][ 550/ 1207] Overall Loss 0.279834 Objective Loss 0.279834 LR 0.000500 Time 0.020317 -2023-02-13 18:03:00,857 - Epoch: [101][ 560/ 1207] Overall Loss 0.279892 Objective Loss 0.279892 LR 0.000500 Time 0.020308 -2023-02-13 18:03:01,053 - Epoch: [101][ 570/ 1207] Overall Loss 0.279961 Objective Loss 0.279961 LR 0.000500 Time 0.020295 -2023-02-13 18:03:01,252 - Epoch: [101][ 580/ 1207] Overall Loss 0.280211 Objective Loss 0.280211 LR 0.000500 Time 0.020287 -2023-02-13 18:03:01,449 - Epoch: [101][ 590/ 1207] Overall Loss 0.279998 Objective Loss 0.279998 LR 0.000500 Time 0.020275 -2023-02-13 18:03:01,647 - Epoch: [101][ 600/ 1207] Overall Loss 0.280451 Objective Loss 0.280451 LR 0.000500 Time 0.020268 -2023-02-13 18:03:01,844 - Epoch: [101][ 610/ 1207] Overall Loss 0.280834 Objective Loss 0.280834 LR 0.000500 Time 0.020258 -2023-02-13 18:03:02,035 - Epoch: [101][ 620/ 1207] Overall Loss 0.280464 Objective Loss 0.280464 LR 0.000500 Time 0.020239 -2023-02-13 18:03:02,225 - Epoch: [101][ 630/ 1207] Overall Loss 0.280669 Objective Loss 0.280669 LR 0.000500 Time 0.020217 -2023-02-13 18:03:02,415 - Epoch: [101][ 640/ 1207] Overall Loss 0.280296 Objective Loss 0.280296 LR 0.000500 Time 0.020199 -2023-02-13 18:03:02,605 - Epoch: [101][ 650/ 1207] Overall Loss 0.280436 Objective Loss 0.280436 LR 0.000500 Time 0.020180 -2023-02-13 18:03:02,799 - Epoch: [101][ 660/ 1207] Overall Loss 0.280080 Objective Loss 0.280080 LR 0.000500 Time 0.020167 -2023-02-13 18:03:02,994 - Epoch: [101][ 670/ 1207] Overall Loss 0.280710 Objective Loss 0.280710 LR 0.000500 Time 0.020156 -2023-02-13 18:03:03,188 - Epoch: [101][ 680/ 1207] Overall Loss 0.280587 Objective Loss 0.280587 LR 0.000500 Time 0.020145 -2023-02-13 18:03:03,384 - Epoch: [101][ 690/ 1207] Overall Loss 0.280759 Objective Loss 0.280759 LR 0.000500 Time 0.020136 -2023-02-13 18:03:03,579 - Epoch: [101][ 700/ 1207] Overall Loss 0.280958 Objective Loss 0.280958 LR 0.000500 Time 0.020127 -2023-02-13 18:03:03,774 - Epoch: [101][ 710/ 1207] Overall Loss 0.280639 Objective Loss 0.280639 LR 0.000500 Time 0.020117 -2023-02-13 18:03:03,969 - Epoch: [101][ 720/ 1207] Overall Loss 0.281116 Objective Loss 0.281116 LR 0.000500 Time 0.020108 -2023-02-13 18:03:04,164 - Epoch: [101][ 730/ 1207] Overall Loss 0.280847 Objective Loss 0.280847 LR 0.000500 Time 0.020099 -2023-02-13 18:03:04,359 - Epoch: [101][ 740/ 1207] Overall Loss 0.280817 Objective Loss 0.280817 LR 0.000500 Time 0.020091 -2023-02-13 18:03:04,550 - Epoch: [101][ 750/ 1207] Overall Loss 0.280948 Objective Loss 0.280948 LR 0.000500 Time 0.020077 -2023-02-13 18:03:04,739 - Epoch: [101][ 760/ 1207] Overall Loss 0.280927 Objective Loss 0.280927 LR 0.000500 Time 0.020062 -2023-02-13 18:03:04,929 - Epoch: [101][ 770/ 1207] Overall Loss 0.281009 Objective Loss 0.281009 LR 0.000500 Time 0.020046 -2023-02-13 18:03:05,118 - Epoch: [101][ 780/ 1207] Overall Loss 0.281137 Objective Loss 0.281137 LR 0.000500 Time 0.020032 -2023-02-13 18:03:05,307 - Epoch: [101][ 790/ 1207] Overall Loss 0.281297 Objective Loss 0.281297 LR 0.000500 Time 0.020017 -2023-02-13 18:03:05,497 - Epoch: [101][ 800/ 1207] Overall Loss 0.281672 Objective Loss 0.281672 LR 0.000500 Time 0.020004 -2023-02-13 18:03:05,687 - Epoch: [101][ 810/ 1207] Overall Loss 0.281297 Objective Loss 0.281297 LR 0.000500 Time 0.019991 -2023-02-13 18:03:05,876 - Epoch: [101][ 820/ 1207] Overall Loss 0.280948 Objective Loss 0.280948 LR 0.000500 Time 0.019977 -2023-02-13 18:03:06,065 - Epoch: [101][ 830/ 1207] Overall Loss 0.281246 Objective Loss 0.281246 LR 0.000500 Time 0.019964 -2023-02-13 18:03:06,254 - Epoch: [101][ 840/ 1207] Overall Loss 0.281443 Objective Loss 0.281443 LR 0.000500 Time 0.019951 -2023-02-13 18:03:06,444 - Epoch: [101][ 850/ 1207] Overall Loss 0.281306 Objective Loss 0.281306 LR 0.000500 Time 0.019940 -2023-02-13 18:03:06,634 - Epoch: [101][ 860/ 1207] Overall Loss 0.281184 Objective Loss 0.281184 LR 0.000500 Time 0.019927 -2023-02-13 18:03:06,823 - Epoch: [101][ 870/ 1207] Overall Loss 0.280999 Objective Loss 0.280999 LR 0.000500 Time 0.019915 -2023-02-13 18:03:07,012 - Epoch: [101][ 880/ 1207] Overall Loss 0.281060 Objective Loss 0.281060 LR 0.000500 Time 0.019904 -2023-02-13 18:03:07,201 - Epoch: [101][ 890/ 1207] Overall Loss 0.281050 Objective Loss 0.281050 LR 0.000500 Time 0.019892 -2023-02-13 18:03:07,391 - Epoch: [101][ 900/ 1207] Overall Loss 0.281283 Objective Loss 0.281283 LR 0.000500 Time 0.019882 -2023-02-13 18:03:07,582 - Epoch: [101][ 910/ 1207] Overall Loss 0.281178 Objective Loss 0.281178 LR 0.000500 Time 0.019872 -2023-02-13 18:03:07,771 - Epoch: [101][ 920/ 1207] Overall Loss 0.281372 Objective Loss 0.281372 LR 0.000500 Time 0.019861 -2023-02-13 18:03:07,960 - Epoch: [101][ 930/ 1207] Overall Loss 0.281146 Objective Loss 0.281146 LR 0.000500 Time 0.019851 -2023-02-13 18:03:08,149 - Epoch: [101][ 940/ 1207] Overall Loss 0.281233 Objective Loss 0.281233 LR 0.000500 Time 0.019840 -2023-02-13 18:03:08,338 - Epoch: [101][ 950/ 1207] Overall Loss 0.281150 Objective Loss 0.281150 LR 0.000500 Time 0.019830 -2023-02-13 18:03:08,529 - Epoch: [101][ 960/ 1207] Overall Loss 0.281283 Objective Loss 0.281283 LR 0.000500 Time 0.019822 -2023-02-13 18:03:08,718 - Epoch: [101][ 970/ 1207] Overall Loss 0.281546 Objective Loss 0.281546 LR 0.000500 Time 0.019812 -2023-02-13 18:03:08,907 - Epoch: [101][ 980/ 1207] Overall Loss 0.281512 Objective Loss 0.281512 LR 0.000500 Time 0.019802 -2023-02-13 18:03:09,096 - Epoch: [101][ 990/ 1207] Overall Loss 0.281657 Objective Loss 0.281657 LR 0.000500 Time 0.019793 -2023-02-13 18:03:09,284 - Epoch: [101][ 1000/ 1207] Overall Loss 0.281937 Objective Loss 0.281937 LR 0.000500 Time 0.019783 -2023-02-13 18:03:09,474 - Epoch: [101][ 1010/ 1207] Overall Loss 0.281774 Objective Loss 0.281774 LR 0.000500 Time 0.019775 -2023-02-13 18:03:09,663 - Epoch: [101][ 1020/ 1207] Overall Loss 0.281893 Objective Loss 0.281893 LR 0.000500 Time 0.019766 -2023-02-13 18:03:09,855 - Epoch: [101][ 1030/ 1207] Overall Loss 0.282049 Objective Loss 0.282049 LR 0.000500 Time 0.019760 -2023-02-13 18:03:10,051 - Epoch: [101][ 1040/ 1207] Overall Loss 0.282246 Objective Loss 0.282246 LR 0.000500 Time 0.019758 -2023-02-13 18:03:10,247 - Epoch: [101][ 1050/ 1207] Overall Loss 0.282320 Objective Loss 0.282320 LR 0.000500 Time 0.019756 -2023-02-13 18:03:10,444 - Epoch: [101][ 1060/ 1207] Overall Loss 0.282299 Objective Loss 0.282299 LR 0.000500 Time 0.019755 -2023-02-13 18:03:10,637 - Epoch: [101][ 1070/ 1207] Overall Loss 0.282395 Objective Loss 0.282395 LR 0.000500 Time 0.019751 -2023-02-13 18:03:10,827 - Epoch: [101][ 1080/ 1207] Overall Loss 0.282130 Objective Loss 0.282130 LR 0.000500 Time 0.019743 -2023-02-13 18:03:11,018 - Epoch: [101][ 1090/ 1207] Overall Loss 0.282098 Objective Loss 0.282098 LR 0.000500 Time 0.019737 -2023-02-13 18:03:11,208 - Epoch: [101][ 1100/ 1207] Overall Loss 0.281931 Objective Loss 0.281931 LR 0.000500 Time 0.019730 -2023-02-13 18:03:11,399 - Epoch: [101][ 1110/ 1207] Overall Loss 0.282000 Objective Loss 0.282000 LR 0.000500 Time 0.019724 -2023-02-13 18:03:11,588 - Epoch: [101][ 1120/ 1207] Overall Loss 0.282196 Objective Loss 0.282196 LR 0.000500 Time 0.019717 -2023-02-13 18:03:11,779 - Epoch: [101][ 1130/ 1207] Overall Loss 0.282385 Objective Loss 0.282385 LR 0.000500 Time 0.019710 -2023-02-13 18:03:11,969 - Epoch: [101][ 1140/ 1207] Overall Loss 0.282451 Objective Loss 0.282451 LR 0.000500 Time 0.019704 -2023-02-13 18:03:12,159 - Epoch: [101][ 1150/ 1207] Overall Loss 0.282418 Objective Loss 0.282418 LR 0.000500 Time 0.019697 -2023-02-13 18:03:12,348 - Epoch: [101][ 1160/ 1207] Overall Loss 0.282226 Objective Loss 0.282226 LR 0.000500 Time 0.019690 -2023-02-13 18:03:12,539 - Epoch: [101][ 1170/ 1207] Overall Loss 0.282242 Objective Loss 0.282242 LR 0.000500 Time 0.019685 -2023-02-13 18:03:12,729 - Epoch: [101][ 1180/ 1207] Overall Loss 0.282378 Objective Loss 0.282378 LR 0.000500 Time 0.019679 -2023-02-13 18:03:12,919 - Epoch: [101][ 1190/ 1207] Overall Loss 0.282389 Objective Loss 0.282389 LR 0.000500 Time 0.019673 -2023-02-13 18:03:13,165 - Epoch: [101][ 1200/ 1207] Overall Loss 0.282395 Objective Loss 0.282395 LR 0.000500 Time 0.019714 -2023-02-13 18:03:13,281 - Epoch: [101][ 1207/ 1207] Overall Loss 0.282314 Objective Loss 0.282314 Top1 83.536585 Top5 96.036585 LR 0.000500 Time 0.019695 -2023-02-13 18:03:13,359 - --- validate (epoch=101)----------- -2023-02-13 18:03:13,360 - 34311 samples (256 per mini-batch) -2023-02-13 18:03:13,852 - Epoch: [101][ 10/ 135] Loss 0.346936 Top1 83.125000 Top5 97.382812 -2023-02-13 18:03:13,982 - Epoch: [101][ 20/ 135] Loss 0.356622 Top1 82.851562 Top5 97.363281 -2023-02-13 18:03:14,116 - Epoch: [101][ 30/ 135] Loss 0.359404 Top1 82.903646 Top5 97.291667 -2023-02-13 18:03:14,245 - Epoch: [101][ 40/ 135] Loss 0.347169 Top1 83.134766 Top5 97.402344 -2023-02-13 18:03:14,378 - Epoch: [101][ 50/ 135] Loss 0.343068 Top1 83.171875 Top5 97.414062 -2023-02-13 18:03:14,506 - Epoch: [101][ 60/ 135] Loss 0.341192 Top1 83.072917 Top5 97.447917 -2023-02-13 18:03:14,639 - Epoch: [101][ 70/ 135] Loss 0.342804 Top1 82.991071 Top5 97.483259 -2023-02-13 18:03:14,766 - Epoch: [101][ 80/ 135] Loss 0.339487 Top1 83.027344 Top5 97.514648 -2023-02-13 18:03:14,900 - Epoch: [101][ 90/ 135] Loss 0.336985 Top1 83.133681 Top5 97.543403 -2023-02-13 18:03:15,028 - Epoch: [101][ 100/ 135] Loss 0.334689 Top1 83.140625 Top5 97.535156 -2023-02-13 18:03:15,161 - Epoch: [101][ 110/ 135] Loss 0.338082 Top1 83.064631 Top5 97.485795 -2023-02-13 18:03:15,290 - Epoch: [101][ 120/ 135] Loss 0.336380 Top1 83.079427 Top5 97.506510 -2023-02-13 18:03:15,420 - Epoch: [101][ 130/ 135] Loss 0.333288 Top1 83.125000 Top5 97.542067 -2023-02-13 18:03:15,465 - Epoch: [101][ 135/ 135] Loss 0.335800 Top1 83.148261 Top5 97.557635 -2023-02-13 18:03:15,537 - ==> Top1: 83.148 Top5: 97.558 Loss: 0.336 - -2023-02-13 18:03:15,538 - ==> Confusion: -[[ 826 3 8 0 8 2 0 2 8 70 2 8 3 3 7 5 5 2 0 0 5] - [ 1 933 2 3 9 32 4 16 7 3 3 1 0 0 4 1 3 0 3 0 8] - [ 7 5 945 17 3 1 20 12 0 2 8 2 2 4 3 9 1 2 3 3 9] - [ 4 0 15 901 4 3 1 1 1 1 21 0 7 2 27 4 4 3 11 1 5] - [ 11 8 1 1 992 10 1 1 2 4 0 6 1 2 7 4 6 0 1 2 6] - [ 0 15 1 4 4 962 3 12 2 4 3 19 5 14 1 2 6 1 4 3 5] - [ 4 4 13 3 0 7 1039 5 1 0 6 2 1 0 0 3 1 3 0 3 4] - [ 3 8 8 1 3 36 6 897 1 1 4 6 5 1 0 0 0 0 32 6 6] - [ 13 1 0 1 3 1 0 1 908 42 6 3 0 10 13 1 0 1 4 0 1] - [ 67 0 2 0 6 0 0 0 33 867 1 2 1 17 3 0 1 3 2 1 6] - [ 2 0 0 6 1 1 7 3 19 2 981 2 1 10 4 0 0 1 6 1 4] - [ 1 2 0 0 4 10 0 5 4 1 0 918 25 5 2 4 2 10 2 6 4] - [ 0 0 0 2 1 6 0 1 3 0 2 38 864 0 3 6 3 15 3 2 10] - [ 8 2 2 0 5 7 0 0 22 17 10 10 3 921 3 4 0 3 1 2 4] - [ 9 3 3 8 4 5 0 1 25 8 4 2 3 4 990 2 0 8 6 0 7] - [ 2 1 6 0 8 2 5 0 0 1 0 12 11 3 0 962 4 13 1 4 11] - [ 3 4 0 1 10 4 0 1 2 0 1 4 2 3 2 13 986 0 3 5 17] - [ 3 2 1 1 1 0 1 1 0 1 2 16 23 0 0 12 1 976 0 2 8] - [ 3 4 7 13 2 1 1 17 7 2 7 1 7 1 17 1 0 1 982 4 8] - [ 0 3 2 0 2 6 5 14 1 0 1 27 3 4 1 6 4 3 1 1052 13] - [ 124 247 186 126 170 257 103 139 140 115 275 157 324 361 199 102 220 116 191 255 9627]] - -2023-02-13 18:03:15,539 - ==> Best [Top1: 83.148 Top5: 97.558 Sparsity:0.00 Params: 148928 on epoch: 101] -2023-02-13 18:03:15,539 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:03:15,545 - - -2023-02-13 18:03:15,546 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:03:16,420 - Epoch: [102][ 10/ 1207] Overall Loss 0.257526 Objective Loss 0.257526 LR 0.000500 Time 0.087429 -2023-02-13 18:03:16,618 - Epoch: [102][ 20/ 1207] Overall Loss 0.275587 Objective Loss 0.275587 LR 0.000500 Time 0.053583 -2023-02-13 18:03:16,807 - Epoch: [102][ 30/ 1207] Overall Loss 0.274472 Objective Loss 0.274472 LR 0.000500 Time 0.042004 -2023-02-13 18:03:16,996 - Epoch: [102][ 40/ 1207] Overall Loss 0.274471 Objective Loss 0.274471 LR 0.000500 Time 0.036206 -2023-02-13 18:03:17,185 - Epoch: [102][ 50/ 1207] Overall Loss 0.268622 Objective Loss 0.268622 LR 0.000500 Time 0.032753 -2023-02-13 18:03:17,374 - Epoch: [102][ 60/ 1207] Overall Loss 0.268760 Objective Loss 0.268760 LR 0.000500 Time 0.030425 -2023-02-13 18:03:17,562 - Epoch: [102][ 70/ 1207] Overall Loss 0.271380 Objective Loss 0.271380 LR 0.000500 Time 0.028767 -2023-02-13 18:03:17,749 - Epoch: [102][ 80/ 1207] Overall Loss 0.270400 Objective Loss 0.270400 LR 0.000500 Time 0.027507 -2023-02-13 18:03:17,937 - Epoch: [102][ 90/ 1207] Overall Loss 0.269397 Objective Loss 0.269397 LR 0.000500 Time 0.026531 -2023-02-13 18:03:18,124 - Epoch: [102][ 100/ 1207] Overall Loss 0.267205 Objective Loss 0.267205 LR 0.000500 Time 0.025749 -2023-02-13 18:03:18,313 - Epoch: [102][ 110/ 1207] Overall Loss 0.265184 Objective Loss 0.265184 LR 0.000500 Time 0.025115 -2023-02-13 18:03:18,501 - Epoch: [102][ 120/ 1207] Overall Loss 0.265238 Objective Loss 0.265238 LR 0.000500 Time 0.024590 -2023-02-13 18:03:18,689 - Epoch: [102][ 130/ 1207] Overall Loss 0.266960 Objective Loss 0.266960 LR 0.000500 Time 0.024143 -2023-02-13 18:03:18,877 - Epoch: [102][ 140/ 1207] Overall Loss 0.270116 Objective Loss 0.270116 LR 0.000500 Time 0.023756 -2023-02-13 18:03:19,065 - Epoch: [102][ 150/ 1207] Overall Loss 0.272928 Objective Loss 0.272928 LR 0.000500 Time 0.023424 -2023-02-13 18:03:19,252 - Epoch: [102][ 160/ 1207] Overall Loss 0.274569 Objective Loss 0.274569 LR 0.000500 Time 0.023126 -2023-02-13 18:03:19,441 - Epoch: [102][ 170/ 1207] Overall Loss 0.273063 Objective Loss 0.273063 LR 0.000500 Time 0.022876 -2023-02-13 18:03:19,629 - Epoch: [102][ 180/ 1207] Overall Loss 0.271375 Objective Loss 0.271375 LR 0.000500 Time 0.022647 -2023-02-13 18:03:19,816 - Epoch: [102][ 190/ 1207] Overall Loss 0.272475 Objective Loss 0.272475 LR 0.000500 Time 0.022440 -2023-02-13 18:03:20,003 - Epoch: [102][ 200/ 1207] Overall Loss 0.271985 Objective Loss 0.271985 LR 0.000500 Time 0.022250 -2023-02-13 18:03:20,193 - Epoch: [102][ 210/ 1207] Overall Loss 0.270211 Objective Loss 0.270211 LR 0.000500 Time 0.022091 -2023-02-13 18:03:20,380 - Epoch: [102][ 220/ 1207] Overall Loss 0.270980 Objective Loss 0.270980 LR 0.000500 Time 0.021937 -2023-02-13 18:03:20,569 - Epoch: [102][ 230/ 1207] Overall Loss 0.270433 Objective Loss 0.270433 LR 0.000500 Time 0.021804 -2023-02-13 18:03:20,756 - Epoch: [102][ 240/ 1207] Overall Loss 0.270322 Objective Loss 0.270322 LR 0.000500 Time 0.021674 -2023-02-13 18:03:20,946 - Epoch: [102][ 250/ 1207] Overall Loss 0.271109 Objective Loss 0.271109 LR 0.000500 Time 0.021563 -2023-02-13 18:03:21,133 - Epoch: [102][ 260/ 1207] Overall Loss 0.272034 Objective Loss 0.272034 LR 0.000500 Time 0.021452 -2023-02-13 18:03:21,321 - Epoch: [102][ 270/ 1207] Overall Loss 0.272117 Objective Loss 0.272117 LR 0.000500 Time 0.021353 -2023-02-13 18:03:21,509 - Epoch: [102][ 280/ 1207] Overall Loss 0.272106 Objective Loss 0.272106 LR 0.000500 Time 0.021261 -2023-02-13 18:03:21,698 - Epoch: [102][ 290/ 1207] Overall Loss 0.273141 Objective Loss 0.273141 LR 0.000500 Time 0.021176 -2023-02-13 18:03:21,886 - Epoch: [102][ 300/ 1207] Overall Loss 0.272636 Objective Loss 0.272636 LR 0.000500 Time 0.021096 -2023-02-13 18:03:22,074 - Epoch: [102][ 310/ 1207] Overall Loss 0.273208 Objective Loss 0.273208 LR 0.000500 Time 0.021020 -2023-02-13 18:03:22,262 - Epoch: [102][ 320/ 1207] Overall Loss 0.272941 Objective Loss 0.272941 LR 0.000500 Time 0.020952 -2023-02-13 18:03:22,454 - Epoch: [102][ 330/ 1207] Overall Loss 0.273270 Objective Loss 0.273270 LR 0.000500 Time 0.020896 -2023-02-13 18:03:22,644 - Epoch: [102][ 340/ 1207] Overall Loss 0.273078 Objective Loss 0.273078 LR 0.000500 Time 0.020840 -2023-02-13 18:03:22,834 - Epoch: [102][ 350/ 1207] Overall Loss 0.274304 Objective Loss 0.274304 LR 0.000500 Time 0.020786 -2023-02-13 18:03:23,024 - Epoch: [102][ 360/ 1207] Overall Loss 0.274624 Objective Loss 0.274624 LR 0.000500 Time 0.020734 -2023-02-13 18:03:23,215 - Epoch: [102][ 370/ 1207] Overall Loss 0.274791 Objective Loss 0.274791 LR 0.000500 Time 0.020690 -2023-02-13 18:03:23,406 - Epoch: [102][ 380/ 1207] Overall Loss 0.275290 Objective Loss 0.275290 LR 0.000500 Time 0.020648 -2023-02-13 18:03:23,597 - Epoch: [102][ 390/ 1207] Overall Loss 0.275768 Objective Loss 0.275768 LR 0.000500 Time 0.020607 -2023-02-13 18:03:23,788 - Epoch: [102][ 400/ 1207] Overall Loss 0.276741 Objective Loss 0.276741 LR 0.000500 Time 0.020567 -2023-02-13 18:03:23,978 - Epoch: [102][ 410/ 1207] Overall Loss 0.277106 Objective Loss 0.277106 LR 0.000500 Time 0.020529 -2023-02-13 18:03:24,168 - Epoch: [102][ 420/ 1207] Overall Loss 0.277535 Objective Loss 0.277535 LR 0.000500 Time 0.020492 -2023-02-13 18:03:24,358 - Epoch: [102][ 430/ 1207] Overall Loss 0.276572 Objective Loss 0.276572 LR 0.000500 Time 0.020456 -2023-02-13 18:03:24,549 - Epoch: [102][ 440/ 1207] Overall Loss 0.276041 Objective Loss 0.276041 LR 0.000500 Time 0.020424 -2023-02-13 18:03:24,739 - Epoch: [102][ 450/ 1207] Overall Loss 0.275484 Objective Loss 0.275484 LR 0.000500 Time 0.020393 -2023-02-13 18:03:24,930 - Epoch: [102][ 460/ 1207] Overall Loss 0.274795 Objective Loss 0.274795 LR 0.000500 Time 0.020363 -2023-02-13 18:03:25,121 - Epoch: [102][ 470/ 1207] Overall Loss 0.274937 Objective Loss 0.274937 LR 0.000500 Time 0.020335 -2023-02-13 18:03:25,311 - Epoch: [102][ 480/ 1207] Overall Loss 0.275124 Objective Loss 0.275124 LR 0.000500 Time 0.020307 -2023-02-13 18:03:25,503 - Epoch: [102][ 490/ 1207] Overall Loss 0.275381 Objective Loss 0.275381 LR 0.000500 Time 0.020283 -2023-02-13 18:03:25,693 - Epoch: [102][ 500/ 1207] Overall Loss 0.275965 Objective Loss 0.275965 LR 0.000500 Time 0.020258 -2023-02-13 18:03:25,885 - Epoch: [102][ 510/ 1207] Overall Loss 0.275354 Objective Loss 0.275354 LR 0.000500 Time 0.020235 -2023-02-13 18:03:26,076 - Epoch: [102][ 520/ 1207] Overall Loss 0.274928 Objective Loss 0.274928 LR 0.000500 Time 0.020213 -2023-02-13 18:03:26,266 - Epoch: [102][ 530/ 1207] Overall Loss 0.275218 Objective Loss 0.275218 LR 0.000500 Time 0.020189 -2023-02-13 18:03:26,457 - Epoch: [102][ 540/ 1207] Overall Loss 0.274940 Objective Loss 0.274940 LR 0.000500 Time 0.020168 -2023-02-13 18:03:26,648 - Epoch: [102][ 550/ 1207] Overall Loss 0.275061 Objective Loss 0.275061 LR 0.000500 Time 0.020149 -2023-02-13 18:03:26,840 - Epoch: [102][ 560/ 1207] Overall Loss 0.275425 Objective Loss 0.275425 LR 0.000500 Time 0.020131 -2023-02-13 18:03:27,031 - Epoch: [102][ 570/ 1207] Overall Loss 0.275236 Objective Loss 0.275236 LR 0.000500 Time 0.020111 -2023-02-13 18:03:27,221 - Epoch: [102][ 580/ 1207] Overall Loss 0.275456 Objective Loss 0.275456 LR 0.000500 Time 0.020092 -2023-02-13 18:03:27,412 - Epoch: [102][ 590/ 1207] Overall Loss 0.275014 Objective Loss 0.275014 LR 0.000500 Time 0.020075 -2023-02-13 18:03:27,603 - Epoch: [102][ 600/ 1207] Overall Loss 0.275508 Objective Loss 0.275508 LR 0.000500 Time 0.020058 -2023-02-13 18:03:27,793 - Epoch: [102][ 610/ 1207] Overall Loss 0.275410 Objective Loss 0.275410 LR 0.000500 Time 0.020040 -2023-02-13 18:03:27,984 - Epoch: [102][ 620/ 1207] Overall Loss 0.275580 Objective Loss 0.275580 LR 0.000500 Time 0.020024 -2023-02-13 18:03:28,174 - Epoch: [102][ 630/ 1207] Overall Loss 0.275435 Objective Loss 0.275435 LR 0.000500 Time 0.020007 -2023-02-13 18:03:28,365 - Epoch: [102][ 640/ 1207] Overall Loss 0.275772 Objective Loss 0.275772 LR 0.000500 Time 0.019992 -2023-02-13 18:03:28,557 - Epoch: [102][ 650/ 1207] Overall Loss 0.275929 Objective Loss 0.275929 LR 0.000500 Time 0.019979 -2023-02-13 18:03:28,747 - Epoch: [102][ 660/ 1207] Overall Loss 0.276008 Objective Loss 0.276008 LR 0.000500 Time 0.019965 -2023-02-13 18:03:28,938 - Epoch: [102][ 670/ 1207] Overall Loss 0.276393 Objective Loss 0.276393 LR 0.000500 Time 0.019950 -2023-02-13 18:03:29,130 - Epoch: [102][ 680/ 1207] Overall Loss 0.276924 Objective Loss 0.276924 LR 0.000500 Time 0.019940 -2023-02-13 18:03:29,321 - Epoch: [102][ 690/ 1207] Overall Loss 0.277136 Objective Loss 0.277136 LR 0.000500 Time 0.019927 -2023-02-13 18:03:29,513 - Epoch: [102][ 700/ 1207] Overall Loss 0.276801 Objective Loss 0.276801 LR 0.000500 Time 0.019916 -2023-02-13 18:03:29,704 - Epoch: [102][ 710/ 1207] Overall Loss 0.276553 Objective Loss 0.276553 LR 0.000500 Time 0.019904 -2023-02-13 18:03:29,894 - Epoch: [102][ 720/ 1207] Overall Loss 0.276752 Objective Loss 0.276752 LR 0.000500 Time 0.019891 -2023-02-13 18:03:30,085 - Epoch: [102][ 730/ 1207] Overall Loss 0.276395 Objective Loss 0.276395 LR 0.000500 Time 0.019879 -2023-02-13 18:03:30,275 - Epoch: [102][ 740/ 1207] Overall Loss 0.276832 Objective Loss 0.276832 LR 0.000500 Time 0.019867 -2023-02-13 18:03:30,466 - Epoch: [102][ 750/ 1207] Overall Loss 0.277114 Objective Loss 0.277114 LR 0.000500 Time 0.019856 -2023-02-13 18:03:30,657 - Epoch: [102][ 760/ 1207] Overall Loss 0.277052 Objective Loss 0.277052 LR 0.000500 Time 0.019845 -2023-02-13 18:03:30,848 - Epoch: [102][ 770/ 1207] Overall Loss 0.276612 Objective Loss 0.276612 LR 0.000500 Time 0.019836 -2023-02-13 18:03:31,039 - Epoch: [102][ 780/ 1207] Overall Loss 0.276696 Objective Loss 0.276696 LR 0.000500 Time 0.019826 -2023-02-13 18:03:31,230 - Epoch: [102][ 790/ 1207] Overall Loss 0.276475 Objective Loss 0.276475 LR 0.000500 Time 0.019816 -2023-02-13 18:03:31,420 - Epoch: [102][ 800/ 1207] Overall Loss 0.276748 Objective Loss 0.276748 LR 0.000500 Time 0.019806 -2023-02-13 18:03:31,612 - Epoch: [102][ 810/ 1207] Overall Loss 0.276854 Objective Loss 0.276854 LR 0.000500 Time 0.019797 -2023-02-13 18:03:31,802 - Epoch: [102][ 820/ 1207] Overall Loss 0.277202 Objective Loss 0.277202 LR 0.000500 Time 0.019787 -2023-02-13 18:03:31,993 - Epoch: [102][ 830/ 1207] Overall Loss 0.277243 Objective Loss 0.277243 LR 0.000500 Time 0.019779 -2023-02-13 18:03:32,185 - Epoch: [102][ 840/ 1207] Overall Loss 0.277308 Objective Loss 0.277308 LR 0.000500 Time 0.019771 -2023-02-13 18:03:32,376 - Epoch: [102][ 850/ 1207] Overall Loss 0.277367 Objective Loss 0.277367 LR 0.000500 Time 0.019763 -2023-02-13 18:03:32,567 - Epoch: [102][ 860/ 1207] Overall Loss 0.277486 Objective Loss 0.277486 LR 0.000500 Time 0.019754 -2023-02-13 18:03:32,758 - Epoch: [102][ 870/ 1207] Overall Loss 0.277575 Objective Loss 0.277575 LR 0.000500 Time 0.019746 -2023-02-13 18:03:32,948 - Epoch: [102][ 880/ 1207] Overall Loss 0.277412 Objective Loss 0.277412 LR 0.000500 Time 0.019738 -2023-02-13 18:03:33,138 - Epoch: [102][ 890/ 1207] Overall Loss 0.277340 Objective Loss 0.277340 LR 0.000500 Time 0.019729 -2023-02-13 18:03:33,328 - Epoch: [102][ 900/ 1207] Overall Loss 0.277148 Objective Loss 0.277148 LR 0.000500 Time 0.019721 -2023-02-13 18:03:33,519 - Epoch: [102][ 910/ 1207] Overall Loss 0.277242 Objective Loss 0.277242 LR 0.000500 Time 0.019714 -2023-02-13 18:03:33,711 - Epoch: [102][ 920/ 1207] Overall Loss 0.277215 Objective Loss 0.277215 LR 0.000500 Time 0.019707 -2023-02-13 18:03:33,901 - Epoch: [102][ 930/ 1207] Overall Loss 0.277267 Objective Loss 0.277267 LR 0.000500 Time 0.019699 -2023-02-13 18:03:34,091 - Epoch: [102][ 940/ 1207] Overall Loss 0.277008 Objective Loss 0.277008 LR 0.000500 Time 0.019692 -2023-02-13 18:03:34,282 - Epoch: [102][ 950/ 1207] Overall Loss 0.276876 Objective Loss 0.276876 LR 0.000500 Time 0.019685 -2023-02-13 18:03:34,473 - Epoch: [102][ 960/ 1207] Overall Loss 0.277116 Objective Loss 0.277116 LR 0.000500 Time 0.019678 -2023-02-13 18:03:34,664 - Epoch: [102][ 970/ 1207] Overall Loss 0.276920 Objective Loss 0.276920 LR 0.000500 Time 0.019672 -2023-02-13 18:03:34,855 - Epoch: [102][ 980/ 1207] Overall Loss 0.277197 Objective Loss 0.277197 LR 0.000500 Time 0.019666 -2023-02-13 18:03:35,046 - Epoch: [102][ 990/ 1207] Overall Loss 0.277448 Objective Loss 0.277448 LR 0.000500 Time 0.019659 -2023-02-13 18:03:35,237 - Epoch: [102][ 1000/ 1207] Overall Loss 0.277352 Objective Loss 0.277352 LR 0.000500 Time 0.019654 -2023-02-13 18:03:35,428 - Epoch: [102][ 1010/ 1207] Overall Loss 0.277509 Objective Loss 0.277509 LR 0.000500 Time 0.019648 -2023-02-13 18:03:35,619 - Epoch: [102][ 1020/ 1207] Overall Loss 0.277660 Objective Loss 0.277660 LR 0.000500 Time 0.019642 -2023-02-13 18:03:35,810 - Epoch: [102][ 1030/ 1207] Overall Loss 0.277779 Objective Loss 0.277779 LR 0.000500 Time 0.019636 -2023-02-13 18:03:36,000 - Epoch: [102][ 1040/ 1207] Overall Loss 0.277531 Objective Loss 0.277531 LR 0.000500 Time 0.019630 -2023-02-13 18:03:36,192 - Epoch: [102][ 1050/ 1207] Overall Loss 0.277762 Objective Loss 0.277762 LR 0.000500 Time 0.019626 -2023-02-13 18:03:36,384 - Epoch: [102][ 1060/ 1207] Overall Loss 0.277616 Objective Loss 0.277616 LR 0.000500 Time 0.019621 -2023-02-13 18:03:36,577 - Epoch: [102][ 1070/ 1207] Overall Loss 0.277716 Objective Loss 0.277716 LR 0.000500 Time 0.019618 -2023-02-13 18:03:36,769 - Epoch: [102][ 1080/ 1207] Overall Loss 0.277765 Objective Loss 0.277765 LR 0.000500 Time 0.019614 -2023-02-13 18:03:36,962 - Epoch: [102][ 1090/ 1207] Overall Loss 0.277846 Objective Loss 0.277846 LR 0.000500 Time 0.019611 -2023-02-13 18:03:37,153 - Epoch: [102][ 1100/ 1207] Overall Loss 0.277618 Objective Loss 0.277618 LR 0.000500 Time 0.019606 -2023-02-13 18:03:37,345 - Epoch: [102][ 1110/ 1207] Overall Loss 0.277394 Objective Loss 0.277394 LR 0.000500 Time 0.019602 -2023-02-13 18:03:37,538 - Epoch: [102][ 1120/ 1207] Overall Loss 0.277468 Objective Loss 0.277468 LR 0.000500 Time 0.019599 -2023-02-13 18:03:37,730 - Epoch: [102][ 1130/ 1207] Overall Loss 0.277275 Objective Loss 0.277275 LR 0.000500 Time 0.019595 -2023-02-13 18:03:37,922 - Epoch: [102][ 1140/ 1207] Overall Loss 0.277306 Objective Loss 0.277306 LR 0.000500 Time 0.019591 -2023-02-13 18:03:38,115 - Epoch: [102][ 1150/ 1207] Overall Loss 0.277407 Objective Loss 0.277407 LR 0.000500 Time 0.019588 -2023-02-13 18:03:38,308 - Epoch: [102][ 1160/ 1207] Overall Loss 0.277243 Objective Loss 0.277243 LR 0.000500 Time 0.019585 -2023-02-13 18:03:38,501 - Epoch: [102][ 1170/ 1207] Overall Loss 0.277307 Objective Loss 0.277307 LR 0.000500 Time 0.019582 -2023-02-13 18:03:38,693 - Epoch: [102][ 1180/ 1207] Overall Loss 0.277042 Objective Loss 0.277042 LR 0.000500 Time 0.019579 -2023-02-13 18:03:38,885 - Epoch: [102][ 1190/ 1207] Overall Loss 0.277079 Objective Loss 0.277079 LR 0.000500 Time 0.019576 -2023-02-13 18:03:39,130 - Epoch: [102][ 1200/ 1207] Overall Loss 0.277275 Objective Loss 0.277275 LR 0.000500 Time 0.019616 -2023-02-13 18:03:39,246 - Epoch: [102][ 1207/ 1207] Overall Loss 0.277407 Objective Loss 0.277407 Top1 81.402439 Top5 96.951220 LR 0.000500 Time 0.019598 -2023-02-13 18:03:39,318 - --- validate (epoch=102)----------- -2023-02-13 18:03:39,318 - 34311 samples (256 per mini-batch) -2023-02-13 18:03:39,731 - Epoch: [102][ 10/ 135] Loss 0.338653 Top1 82.070312 Top5 97.226562 -2023-02-13 18:03:39,860 - Epoch: [102][ 20/ 135] Loss 0.315790 Top1 83.281250 Top5 97.460938 -2023-02-13 18:03:39,989 - Epoch: [102][ 30/ 135] Loss 0.305354 Top1 83.450521 Top5 97.552083 -2023-02-13 18:03:40,133 - Epoch: [102][ 40/ 135] Loss 0.309387 Top1 83.437500 Top5 97.412109 -2023-02-13 18:03:40,256 - Epoch: [102][ 50/ 135] Loss 0.307654 Top1 83.554688 Top5 97.421875 -2023-02-13 18:03:40,380 - Epoch: [102][ 60/ 135] Loss 0.309703 Top1 83.704427 Top5 97.493490 -2023-02-13 18:03:40,513 - Epoch: [102][ 70/ 135] Loss 0.310633 Top1 83.777902 Top5 97.516741 -2023-02-13 18:03:40,655 - Epoch: [102][ 80/ 135] Loss 0.314174 Top1 83.657227 Top5 97.441406 -2023-02-13 18:03:40,790 - Epoch: [102][ 90/ 135] Loss 0.312176 Top1 83.784722 Top5 97.482639 -2023-02-13 18:03:40,917 - Epoch: [102][ 100/ 135] Loss 0.311435 Top1 83.769531 Top5 97.535156 -2023-02-13 18:03:41,043 - Epoch: [102][ 110/ 135] Loss 0.312462 Top1 83.668324 Top5 97.553267 -2023-02-13 18:03:41,175 - Epoch: [102][ 120/ 135] Loss 0.313926 Top1 83.688151 Top5 97.548828 -2023-02-13 18:03:41,304 - Epoch: [102][ 130/ 135] Loss 0.317252 Top1 83.608774 Top5 97.533053 -2023-02-13 18:03:41,347 - Epoch: [102][ 135/ 135] Loss 0.322954 Top1 83.594183 Top5 97.511002 -2023-02-13 18:03:41,420 - ==> Top1: 83.594 Top5: 97.511 Loss: 0.323 - -2023-02-13 18:03:41,421 - ==> Confusion: -[[ 863 4 5 0 12 2 0 1 5 41 1 5 0 4 4 5 5 2 1 2 5] - [ 1 949 1 1 9 29 5 15 5 1 1 2 1 0 0 1 5 1 2 0 4] - [ 10 6 939 14 6 2 17 16 0 3 3 1 1 3 4 11 1 2 5 5 9] - [ 4 2 11 904 4 4 3 3 2 1 17 0 8 0 20 1 5 6 17 0 4] - [ 17 8 0 1 994 5 1 3 1 4 0 4 1 3 5 5 6 1 0 3 4] - [ 3 17 0 5 6 964 4 13 1 3 4 14 2 17 1 2 8 0 1 2 3] - [ 5 2 14 3 0 5 1034 5 0 1 3 2 2 0 0 6 1 3 0 8 5] - [ 2 7 5 2 2 24 6 931 2 1 1 6 2 1 0 0 1 1 20 7 3] - [ 20 3 1 4 1 0 0 3 905 32 6 2 1 7 13 1 1 2 4 1 2] - [ 84 2 1 0 6 2 0 2 36 844 0 0 0 18 2 3 2 3 1 0 6] - [ 0 1 2 10 0 4 4 5 22 0 977 2 1 5 3 1 0 1 9 1 3] - [ 3 3 0 0 7 12 1 6 2 0 0 909 21 6 2 5 6 10 3 7 2] - [ 1 0 1 6 0 4 0 2 4 0 2 24 868 0 2 12 2 15 2 1 13] - [ 2 1 1 1 10 6 0 5 17 21 7 7 0 923 4 4 3 5 1 3 3] - [ 13 2 1 22 5 3 0 1 21 6 4 1 3 2 982 1 1 7 9 1 7] - [ 2 1 4 1 10 0 5 0 1 0 0 6 4 2 0 978 7 15 1 3 6] - [ 2 2 0 0 9 1 0 2 2 0 0 0 3 3 1 13 1007 1 2 4 9] - [ 4 2 1 4 0 0 3 0 0 0 2 6 12 3 0 17 0 991 0 4 2] - [ 6 7 2 14 3 1 0 26 6 0 5 0 5 0 8 1 1 2 994 2 3] - [ 0 4 3 1 2 7 6 11 1 0 2 15 1 4 1 6 5 4 0 1068 7] - [ 165 276 193 132 189 250 83 166 126 85 217 99 290 314 163 135 319 121 168 285 9658]] - -2023-02-13 18:03:41,422 - ==> Best [Top1: 83.594 Top5: 97.511 Sparsity:0.00 Params: 148928 on epoch: 102] -2023-02-13 18:03:41,422 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:03:41,524 - - -2023-02-13 18:03:41,525 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:03:42,556 - Epoch: [103][ 10/ 1207] Overall Loss 0.282885 Objective Loss 0.282885 LR 0.000500 Time 0.102873 -2023-02-13 18:03:42,750 - Epoch: [103][ 20/ 1207] Overall Loss 0.280945 Objective Loss 0.280945 LR 0.000500 Time 0.061132 -2023-02-13 18:03:42,939 - Epoch: [103][ 30/ 1207] Overall Loss 0.275930 Objective Loss 0.275930 LR 0.000500 Time 0.047043 -2023-02-13 18:03:43,129 - Epoch: [103][ 40/ 1207] Overall Loss 0.269399 Objective Loss 0.269399 LR 0.000500 Time 0.040012 -2023-02-13 18:03:43,317 - Epoch: [103][ 50/ 1207] Overall Loss 0.271036 Objective Loss 0.271036 LR 0.000500 Time 0.035781 -2023-02-13 18:03:43,507 - Epoch: [103][ 60/ 1207] Overall Loss 0.270789 Objective Loss 0.270789 LR 0.000500 Time 0.032972 -2023-02-13 18:03:43,696 - Epoch: [103][ 70/ 1207] Overall Loss 0.271789 Objective Loss 0.271789 LR 0.000500 Time 0.030956 -2023-02-13 18:03:43,886 - Epoch: [103][ 80/ 1207] Overall Loss 0.271216 Objective Loss 0.271216 LR 0.000500 Time 0.029452 -2023-02-13 18:03:44,073 - Epoch: [103][ 90/ 1207] Overall Loss 0.272251 Objective Loss 0.272251 LR 0.000500 Time 0.028257 -2023-02-13 18:03:44,261 - Epoch: [103][ 100/ 1207] Overall Loss 0.274673 Objective Loss 0.274673 LR 0.000500 Time 0.027303 -2023-02-13 18:03:44,448 - Epoch: [103][ 110/ 1207] Overall Loss 0.276771 Objective Loss 0.276771 LR 0.000500 Time 0.026526 -2023-02-13 18:03:44,636 - Epoch: [103][ 120/ 1207] Overall Loss 0.275888 Objective Loss 0.275888 LR 0.000500 Time 0.025880 -2023-02-13 18:03:44,823 - Epoch: [103][ 130/ 1207] Overall Loss 0.274978 Objective Loss 0.274978 LR 0.000500 Time 0.025324 -2023-02-13 18:03:45,011 - Epoch: [103][ 140/ 1207] Overall Loss 0.273014 Objective Loss 0.273014 LR 0.000500 Time 0.024849 -2023-02-13 18:03:45,198 - Epoch: [103][ 150/ 1207] Overall Loss 0.271030 Objective Loss 0.271030 LR 0.000500 Time 0.024438 -2023-02-13 18:03:45,385 - Epoch: [103][ 160/ 1207] Overall Loss 0.269649 Objective Loss 0.269649 LR 0.000500 Time 0.024080 -2023-02-13 18:03:45,573 - Epoch: [103][ 170/ 1207] Overall Loss 0.272556 Objective Loss 0.272556 LR 0.000500 Time 0.023765 -2023-02-13 18:03:45,760 - Epoch: [103][ 180/ 1207] Overall Loss 0.272245 Objective Loss 0.272245 LR 0.000500 Time 0.023481 -2023-02-13 18:03:45,949 - Epoch: [103][ 190/ 1207] Overall Loss 0.272581 Objective Loss 0.272581 LR 0.000500 Time 0.023238 -2023-02-13 18:03:46,136 - Epoch: [103][ 200/ 1207] Overall Loss 0.272182 Objective Loss 0.272182 LR 0.000500 Time 0.023009 -2023-02-13 18:03:46,323 - Epoch: [103][ 210/ 1207] Overall Loss 0.273182 Objective Loss 0.273182 LR 0.000500 Time 0.022805 -2023-02-13 18:03:46,511 - Epoch: [103][ 220/ 1207] Overall Loss 0.273184 Objective Loss 0.273184 LR 0.000500 Time 0.022621 -2023-02-13 18:03:46,699 - Epoch: [103][ 230/ 1207] Overall Loss 0.271800 Objective Loss 0.271800 LR 0.000500 Time 0.022453 -2023-02-13 18:03:46,887 - Epoch: [103][ 240/ 1207] Overall Loss 0.270655 Objective Loss 0.270655 LR 0.000500 Time 0.022297 -2023-02-13 18:03:47,074 - Epoch: [103][ 250/ 1207] Overall Loss 0.271278 Objective Loss 0.271278 LR 0.000500 Time 0.022153 -2023-02-13 18:03:47,261 - Epoch: [103][ 260/ 1207] Overall Loss 0.270718 Objective Loss 0.270718 LR 0.000500 Time 0.022020 -2023-02-13 18:03:47,449 - Epoch: [103][ 270/ 1207] Overall Loss 0.270762 Objective Loss 0.270762 LR 0.000500 Time 0.021900 -2023-02-13 18:03:47,637 - Epoch: [103][ 280/ 1207] Overall Loss 0.270718 Objective Loss 0.270718 LR 0.000500 Time 0.021788 -2023-02-13 18:03:47,825 - Epoch: [103][ 290/ 1207] Overall Loss 0.271423 Objective Loss 0.271423 LR 0.000500 Time 0.021682 -2023-02-13 18:03:48,012 - Epoch: [103][ 300/ 1207] Overall Loss 0.272142 Objective Loss 0.272142 LR 0.000500 Time 0.021582 -2023-02-13 18:03:48,199 - Epoch: [103][ 310/ 1207] Overall Loss 0.272255 Objective Loss 0.272255 LR 0.000500 Time 0.021489 -2023-02-13 18:03:48,386 - Epoch: [103][ 320/ 1207] Overall Loss 0.272717 Objective Loss 0.272717 LR 0.000500 Time 0.021401 -2023-02-13 18:03:48,574 - Epoch: [103][ 330/ 1207] Overall Loss 0.273589 Objective Loss 0.273589 LR 0.000500 Time 0.021321 -2023-02-13 18:03:48,761 - Epoch: [103][ 340/ 1207] Overall Loss 0.273870 Objective Loss 0.273870 LR 0.000500 Time 0.021242 -2023-02-13 18:03:48,947 - Epoch: [103][ 350/ 1207] Overall Loss 0.273959 Objective Loss 0.273959 LR 0.000500 Time 0.021166 -2023-02-13 18:03:49,135 - Epoch: [103][ 360/ 1207] Overall Loss 0.274511 Objective Loss 0.274511 LR 0.000500 Time 0.021097 -2023-02-13 18:03:49,321 - Epoch: [103][ 370/ 1207] Overall Loss 0.274619 Objective Loss 0.274619 LR 0.000500 Time 0.021031 -2023-02-13 18:03:49,509 - Epoch: [103][ 380/ 1207] Overall Loss 0.274820 Objective Loss 0.274820 LR 0.000500 Time 0.020972 -2023-02-13 18:03:49,697 - Epoch: [103][ 390/ 1207] Overall Loss 0.274186 Objective Loss 0.274186 LR 0.000500 Time 0.020913 -2023-02-13 18:03:49,884 - Epoch: [103][ 400/ 1207] Overall Loss 0.274228 Objective Loss 0.274228 LR 0.000500 Time 0.020856 -2023-02-13 18:03:50,070 - Epoch: [103][ 410/ 1207] Overall Loss 0.274093 Objective Loss 0.274093 LR 0.000500 Time 0.020801 -2023-02-13 18:03:50,258 - Epoch: [103][ 420/ 1207] Overall Loss 0.274292 Objective Loss 0.274292 LR 0.000500 Time 0.020752 -2023-02-13 18:03:50,444 - Epoch: [103][ 430/ 1207] Overall Loss 0.274205 Objective Loss 0.274205 LR 0.000500 Time 0.020703 -2023-02-13 18:03:50,633 - Epoch: [103][ 440/ 1207] Overall Loss 0.274515 Objective Loss 0.274515 LR 0.000500 Time 0.020660 -2023-02-13 18:03:50,820 - Epoch: [103][ 450/ 1207] Overall Loss 0.274281 Objective Loss 0.274281 LR 0.000500 Time 0.020615 -2023-02-13 18:03:51,008 - Epoch: [103][ 460/ 1207] Overall Loss 0.274483 Objective Loss 0.274483 LR 0.000500 Time 0.020576 -2023-02-13 18:03:51,196 - Epoch: [103][ 470/ 1207] Overall Loss 0.274311 Objective Loss 0.274311 LR 0.000500 Time 0.020537 -2023-02-13 18:03:51,383 - Epoch: [103][ 480/ 1207] Overall Loss 0.274703 Objective Loss 0.274703 LR 0.000500 Time 0.020498 -2023-02-13 18:03:51,571 - Epoch: [103][ 490/ 1207] Overall Loss 0.274634 Objective Loss 0.274634 LR 0.000500 Time 0.020463 -2023-02-13 18:03:51,759 - Epoch: [103][ 500/ 1207] Overall Loss 0.275260 Objective Loss 0.275260 LR 0.000500 Time 0.020428 -2023-02-13 18:03:51,947 - Epoch: [103][ 510/ 1207] Overall Loss 0.275248 Objective Loss 0.275248 LR 0.000500 Time 0.020396 -2023-02-13 18:03:52,134 - Epoch: [103][ 520/ 1207] Overall Loss 0.275129 Objective Loss 0.275129 LR 0.000500 Time 0.020363 -2023-02-13 18:03:52,322 - Epoch: [103][ 530/ 1207] Overall Loss 0.275090 Objective Loss 0.275090 LR 0.000500 Time 0.020332 -2023-02-13 18:03:52,510 - Epoch: [103][ 540/ 1207] Overall Loss 0.275087 Objective Loss 0.275087 LR 0.000500 Time 0.020303 -2023-02-13 18:03:52,698 - Epoch: [103][ 550/ 1207] Overall Loss 0.275449 Objective Loss 0.275449 LR 0.000500 Time 0.020275 -2023-02-13 18:03:52,885 - Epoch: [103][ 560/ 1207] Overall Loss 0.275418 Objective Loss 0.275418 LR 0.000500 Time 0.020246 -2023-02-13 18:03:53,072 - Epoch: [103][ 570/ 1207] Overall Loss 0.275731 Objective Loss 0.275731 LR 0.000500 Time 0.020219 -2023-02-13 18:03:53,259 - Epoch: [103][ 580/ 1207] Overall Loss 0.275563 Objective Loss 0.275563 LR 0.000500 Time 0.020193 -2023-02-13 18:03:53,447 - Epoch: [103][ 590/ 1207] Overall Loss 0.275674 Objective Loss 0.275674 LR 0.000500 Time 0.020168 -2023-02-13 18:03:53,636 - Epoch: [103][ 600/ 1207] Overall Loss 0.275665 Objective Loss 0.275665 LR 0.000500 Time 0.020146 -2023-02-13 18:03:53,823 - Epoch: [103][ 610/ 1207] Overall Loss 0.275334 Objective Loss 0.275334 LR 0.000500 Time 0.020122 -2023-02-13 18:03:54,010 - Epoch: [103][ 620/ 1207] Overall Loss 0.274885 Objective Loss 0.274885 LR 0.000500 Time 0.020098 -2023-02-13 18:03:54,197 - Epoch: [103][ 630/ 1207] Overall Loss 0.274797 Objective Loss 0.274797 LR 0.000500 Time 0.020076 -2023-02-13 18:03:54,384 - Epoch: [103][ 640/ 1207] Overall Loss 0.275270 Objective Loss 0.275270 LR 0.000500 Time 0.020053 -2023-02-13 18:03:54,573 - Epoch: [103][ 650/ 1207] Overall Loss 0.275372 Objective Loss 0.275372 LR 0.000500 Time 0.020035 -2023-02-13 18:03:54,760 - Epoch: [103][ 660/ 1207] Overall Loss 0.275495 Objective Loss 0.275495 LR 0.000500 Time 0.020015 -2023-02-13 18:03:54,947 - Epoch: [103][ 670/ 1207] Overall Loss 0.275294 Objective Loss 0.275294 LR 0.000500 Time 0.019995 -2023-02-13 18:03:55,135 - Epoch: [103][ 680/ 1207] Overall Loss 0.275451 Objective Loss 0.275451 LR 0.000500 Time 0.019976 -2023-02-13 18:03:55,322 - Epoch: [103][ 690/ 1207] Overall Loss 0.275201 Objective Loss 0.275201 LR 0.000500 Time 0.019958 -2023-02-13 18:03:55,511 - Epoch: [103][ 700/ 1207] Overall Loss 0.274936 Objective Loss 0.274936 LR 0.000500 Time 0.019941 -2023-02-13 18:03:55,699 - Epoch: [103][ 710/ 1207] Overall Loss 0.275018 Objective Loss 0.275018 LR 0.000500 Time 0.019924 -2023-02-13 18:03:55,889 - Epoch: [103][ 720/ 1207] Overall Loss 0.274938 Objective Loss 0.274938 LR 0.000500 Time 0.019911 -2023-02-13 18:03:56,077 - Epoch: [103][ 730/ 1207] Overall Loss 0.274774 Objective Loss 0.274774 LR 0.000500 Time 0.019895 -2023-02-13 18:03:56,264 - Epoch: [103][ 740/ 1207] Overall Loss 0.274755 Objective Loss 0.274755 LR 0.000500 Time 0.019879 -2023-02-13 18:03:56,451 - Epoch: [103][ 750/ 1207] Overall Loss 0.274565 Objective Loss 0.274565 LR 0.000500 Time 0.019864 -2023-02-13 18:03:56,640 - Epoch: [103][ 760/ 1207] Overall Loss 0.274141 Objective Loss 0.274141 LR 0.000500 Time 0.019850 -2023-02-13 18:03:56,828 - Epoch: [103][ 770/ 1207] Overall Loss 0.274188 Objective Loss 0.274188 LR 0.000500 Time 0.019836 -2023-02-13 18:03:57,015 - Epoch: [103][ 780/ 1207] Overall Loss 0.274544 Objective Loss 0.274544 LR 0.000500 Time 0.019821 -2023-02-13 18:03:57,202 - Epoch: [103][ 790/ 1207] Overall Loss 0.274989 Objective Loss 0.274989 LR 0.000500 Time 0.019806 -2023-02-13 18:03:57,390 - Epoch: [103][ 800/ 1207] Overall Loss 0.274934 Objective Loss 0.274934 LR 0.000500 Time 0.019793 -2023-02-13 18:03:57,578 - Epoch: [103][ 810/ 1207] Overall Loss 0.274855 Objective Loss 0.274855 LR 0.000500 Time 0.019780 -2023-02-13 18:03:57,765 - Epoch: [103][ 820/ 1207] Overall Loss 0.275093 Objective Loss 0.275093 LR 0.000500 Time 0.019766 -2023-02-13 18:03:57,952 - Epoch: [103][ 830/ 1207] Overall Loss 0.274980 Objective Loss 0.274980 LR 0.000500 Time 0.019754 -2023-02-13 18:03:58,140 - Epoch: [103][ 840/ 1207] Overall Loss 0.275210 Objective Loss 0.275210 LR 0.000500 Time 0.019741 -2023-02-13 18:03:58,328 - Epoch: [103][ 850/ 1207] Overall Loss 0.275028 Objective Loss 0.275028 LR 0.000500 Time 0.019730 -2023-02-13 18:03:58,516 - Epoch: [103][ 860/ 1207] Overall Loss 0.274676 Objective Loss 0.274676 LR 0.000500 Time 0.019718 -2023-02-13 18:03:58,703 - Epoch: [103][ 870/ 1207] Overall Loss 0.274912 Objective Loss 0.274912 LR 0.000500 Time 0.019707 -2023-02-13 18:03:58,890 - Epoch: [103][ 880/ 1207] Overall Loss 0.274902 Objective Loss 0.274902 LR 0.000500 Time 0.019695 -2023-02-13 18:03:59,078 - Epoch: [103][ 890/ 1207] Overall Loss 0.275190 Objective Loss 0.275190 LR 0.000500 Time 0.019684 -2023-02-13 18:03:59,265 - Epoch: [103][ 900/ 1207] Overall Loss 0.275436 Objective Loss 0.275436 LR 0.000500 Time 0.019673 -2023-02-13 18:03:59,453 - Epoch: [103][ 910/ 1207] Overall Loss 0.275464 Objective Loss 0.275464 LR 0.000500 Time 0.019663 -2023-02-13 18:03:59,641 - Epoch: [103][ 920/ 1207] Overall Loss 0.275871 Objective Loss 0.275871 LR 0.000500 Time 0.019653 -2023-02-13 18:03:59,829 - Epoch: [103][ 930/ 1207] Overall Loss 0.275619 Objective Loss 0.275619 LR 0.000500 Time 0.019643 -2023-02-13 18:04:00,015 - Epoch: [103][ 940/ 1207] Overall Loss 0.275999 Objective Loss 0.275999 LR 0.000500 Time 0.019633 -2023-02-13 18:04:00,203 - Epoch: [103][ 950/ 1207] Overall Loss 0.276125 Objective Loss 0.276125 LR 0.000500 Time 0.019623 -2023-02-13 18:04:00,390 - Epoch: [103][ 960/ 1207] Overall Loss 0.275970 Objective Loss 0.275970 LR 0.000500 Time 0.019613 -2023-02-13 18:04:00,578 - Epoch: [103][ 970/ 1207] Overall Loss 0.276105 Objective Loss 0.276105 LR 0.000500 Time 0.019604 -2023-02-13 18:04:00,765 - Epoch: [103][ 980/ 1207] Overall Loss 0.276080 Objective Loss 0.276080 LR 0.000500 Time 0.019595 -2023-02-13 18:04:00,954 - Epoch: [103][ 990/ 1207] Overall Loss 0.275898 Objective Loss 0.275898 LR 0.000500 Time 0.019588 -2023-02-13 18:04:01,141 - Epoch: [103][ 1000/ 1207] Overall Loss 0.275575 Objective Loss 0.275575 LR 0.000500 Time 0.019578 -2023-02-13 18:04:01,328 - Epoch: [103][ 1010/ 1207] Overall Loss 0.275474 Objective Loss 0.275474 LR 0.000500 Time 0.019569 -2023-02-13 18:04:01,515 - Epoch: [103][ 1020/ 1207] Overall Loss 0.275222 Objective Loss 0.275222 LR 0.000500 Time 0.019560 -2023-02-13 18:04:01,703 - Epoch: [103][ 1030/ 1207] Overall Loss 0.275334 Objective Loss 0.275334 LR 0.000500 Time 0.019553 -2023-02-13 18:04:01,891 - Epoch: [103][ 1040/ 1207] Overall Loss 0.275455 Objective Loss 0.275455 LR 0.000500 Time 0.019545 -2023-02-13 18:04:02,079 - Epoch: [103][ 1050/ 1207] Overall Loss 0.275506 Objective Loss 0.275506 LR 0.000500 Time 0.019537 -2023-02-13 18:04:02,267 - Epoch: [103][ 1060/ 1207] Overall Loss 0.275545 Objective Loss 0.275545 LR 0.000500 Time 0.019530 -2023-02-13 18:04:02,455 - Epoch: [103][ 1070/ 1207] Overall Loss 0.275560 Objective Loss 0.275560 LR 0.000500 Time 0.019523 -2023-02-13 18:04:02,643 - Epoch: [103][ 1080/ 1207] Overall Loss 0.275645 Objective Loss 0.275645 LR 0.000500 Time 0.019516 -2023-02-13 18:04:02,831 - Epoch: [103][ 1090/ 1207] Overall Loss 0.275804 Objective Loss 0.275804 LR 0.000500 Time 0.019509 -2023-02-13 18:04:03,018 - Epoch: [103][ 1100/ 1207] Overall Loss 0.275860 Objective Loss 0.275860 LR 0.000500 Time 0.019502 -2023-02-13 18:04:03,206 - Epoch: [103][ 1110/ 1207] Overall Loss 0.276079 Objective Loss 0.276079 LR 0.000500 Time 0.019494 -2023-02-13 18:04:03,393 - Epoch: [103][ 1120/ 1207] Overall Loss 0.275937 Objective Loss 0.275937 LR 0.000500 Time 0.019487 -2023-02-13 18:04:03,581 - Epoch: [103][ 1130/ 1207] Overall Loss 0.275653 Objective Loss 0.275653 LR 0.000500 Time 0.019481 -2023-02-13 18:04:03,769 - Epoch: [103][ 1140/ 1207] Overall Loss 0.275742 Objective Loss 0.275742 LR 0.000500 Time 0.019474 -2023-02-13 18:04:03,957 - Epoch: [103][ 1150/ 1207] Overall Loss 0.275534 Objective Loss 0.275534 LR 0.000500 Time 0.019468 -2023-02-13 18:04:04,144 - Epoch: [103][ 1160/ 1207] Overall Loss 0.275600 Objective Loss 0.275600 LR 0.000500 Time 0.019461 -2023-02-13 18:04:04,331 - Epoch: [103][ 1170/ 1207] Overall Loss 0.275435 Objective Loss 0.275435 LR 0.000500 Time 0.019455 -2023-02-13 18:04:04,519 - Epoch: [103][ 1180/ 1207] Overall Loss 0.275365 Objective Loss 0.275365 LR 0.000500 Time 0.019448 -2023-02-13 18:04:04,708 - Epoch: [103][ 1190/ 1207] Overall Loss 0.275229 Objective Loss 0.275229 LR 0.000500 Time 0.019443 -2023-02-13 18:04:04,952 - Epoch: [103][ 1200/ 1207] Overall Loss 0.275006 Objective Loss 0.275006 LR 0.000500 Time 0.019484 -2023-02-13 18:04:05,066 - Epoch: [103][ 1207/ 1207] Overall Loss 0.274887 Objective Loss 0.274887 Top1 87.500000 Top5 99.085366 LR 0.000500 Time 0.019466 -2023-02-13 18:04:05,138 - --- validate (epoch=103)----------- -2023-02-13 18:04:05,138 - 34311 samples (256 per mini-batch) -2023-02-13 18:04:05,545 - Epoch: [103][ 10/ 135] Loss 0.310507 Top1 84.101562 Top5 97.070312 -2023-02-13 18:04:05,676 - Epoch: [103][ 20/ 135] Loss 0.314078 Top1 84.140625 Top5 97.500000 -2023-02-13 18:04:05,806 - Epoch: [103][ 30/ 135] Loss 0.319847 Top1 83.763021 Top5 97.526042 -2023-02-13 18:04:05,934 - Epoch: [103][ 40/ 135] Loss 0.320329 Top1 83.857422 Top5 97.607422 -2023-02-13 18:04:06,061 - Epoch: [103][ 50/ 135] Loss 0.317753 Top1 83.851562 Top5 97.554688 -2023-02-13 18:04:06,186 - Epoch: [103][ 60/ 135] Loss 0.321445 Top1 83.769531 Top5 97.519531 -2023-02-13 18:04:06,312 - Epoch: [103][ 70/ 135] Loss 0.321951 Top1 83.805804 Top5 97.472098 -2023-02-13 18:04:06,435 - Epoch: [103][ 80/ 135] Loss 0.322678 Top1 83.862305 Top5 97.490234 -2023-02-13 18:04:06,564 - Epoch: [103][ 90/ 135] Loss 0.320406 Top1 83.897569 Top5 97.491319 -2023-02-13 18:04:06,689 - Epoch: [103][ 100/ 135] Loss 0.320736 Top1 83.886719 Top5 97.535156 -2023-02-13 18:04:06,816 - Epoch: [103][ 110/ 135] Loss 0.323758 Top1 83.881392 Top5 97.535511 -2023-02-13 18:04:06,942 - Epoch: [103][ 120/ 135] Loss 0.323716 Top1 83.870443 Top5 97.513021 -2023-02-13 18:04:07,070 - Epoch: [103][ 130/ 135] Loss 0.324349 Top1 83.882212 Top5 97.496995 -2023-02-13 18:04:07,115 - Epoch: [103][ 135/ 135] Loss 0.325158 Top1 83.911865 Top5 97.508088 -2023-02-13 18:04:07,184 - ==> Top1: 83.912 Top5: 97.508 Loss: 0.325 - -2023-02-13 18:04:07,185 - ==> Confusion: -[[ 819 7 7 1 14 1 0 1 5 74 2 5 1 5 2 5 3 2 1 2 10] - [ 3 923 1 2 13 38 2 21 4 0 2 4 2 0 2 1 2 0 5 1 7] - [ 6 4 932 18 3 0 24 17 2 1 3 1 3 5 4 12 2 4 4 3 10] - [ 4 2 14 914 2 5 2 1 2 2 17 0 8 0 16 1 4 6 12 0 4] - [ 10 7 0 0 998 6 1 1 1 4 1 7 2 2 6 5 7 1 0 3 4] - [ 0 13 1 9 5 960 5 17 4 3 1 13 5 11 2 3 6 0 1 6 5] - [ 4 3 19 3 0 6 1022 2 0 4 2 2 2 2 0 4 1 7 1 11 4] - [ 3 10 8 0 3 28 3 915 1 1 2 5 5 1 0 0 1 2 21 11 4] - [ 12 4 0 2 1 0 0 5 889 46 8 1 1 6 23 1 1 2 5 0 2] - [ 63 0 2 0 6 2 0 2 42 860 3 0 0 16 3 1 1 3 1 1 6] - [ 0 1 1 9 0 2 2 4 16 1 991 3 4 4 3 0 0 2 5 1 2] - [ 0 3 0 0 4 14 0 4 2 2 0 918 27 4 3 3 3 7 3 5 3] - [ 1 0 1 5 1 4 0 1 1 0 0 25 888 0 2 9 1 11 3 0 6] - [ 6 1 0 1 6 11 0 2 16 12 11 6 4 924 4 5 1 3 1 5 5] - [ 8 2 3 23 3 4 0 2 18 6 6 1 3 1 992 1 2 4 7 1 5] - [ 3 0 7 0 6 1 2 1 1 0 1 8 6 1 2 967 8 19 1 8 4] - [ 3 1 0 2 10 4 0 1 1 1 0 5 2 1 2 16 992 1 3 4 12] - [ 6 4 0 4 0 0 1 0 0 2 2 10 21 1 0 13 0 979 1 2 5] - [ 4 2 4 11 1 0 1 17 5 0 9 1 6 1 15 1 0 2 1002 4 0] - [ 0 3 2 2 1 4 5 13 1 0 1 27 2 3 0 4 4 3 0 1063 10] - [ 127 229 195 182 140 210 81 156 98 85 224 119 378 263 174 122 225 131 187 265 9843]] - -2023-02-13 18:04:07,186 - ==> Best [Top1: 83.912 Top5: 97.508 Sparsity:0.00 Params: 148928 on epoch: 103] -2023-02-13 18:04:07,186 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:04:07,193 - - -2023-02-13 18:04:07,193 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:04:08,192 - Epoch: [104][ 10/ 1207] Overall Loss 0.275762 Objective Loss 0.275762 LR 0.000500 Time 0.099864 -2023-02-13 18:04:08,402 - Epoch: [104][ 20/ 1207] Overall Loss 0.275347 Objective Loss 0.275347 LR 0.000500 Time 0.060386 -2023-02-13 18:04:08,599 - Epoch: [104][ 30/ 1207] Overall Loss 0.271912 Objective Loss 0.271912 LR 0.000500 Time 0.046825 -2023-02-13 18:04:08,801 - Epoch: [104][ 40/ 1207] Overall Loss 0.272256 Objective Loss 0.272256 LR 0.000500 Time 0.040154 -2023-02-13 18:04:08,998 - Epoch: [104][ 50/ 1207] Overall Loss 0.269032 Objective Loss 0.269032 LR 0.000500 Time 0.036051 -2023-02-13 18:04:09,199 - Epoch: [104][ 60/ 1207] Overall Loss 0.273155 Objective Loss 0.273155 LR 0.000500 Time 0.033393 -2023-02-13 18:04:09,396 - Epoch: [104][ 70/ 1207] Overall Loss 0.272648 Objective Loss 0.272648 LR 0.000500 Time 0.031422 -2023-02-13 18:04:09,598 - Epoch: [104][ 80/ 1207] Overall Loss 0.271838 Objective Loss 0.271838 LR 0.000500 Time 0.030013 -2023-02-13 18:04:09,795 - Epoch: [104][ 90/ 1207] Overall Loss 0.269573 Objective Loss 0.269573 LR 0.000500 Time 0.028867 -2023-02-13 18:04:09,996 - Epoch: [104][ 100/ 1207] Overall Loss 0.268255 Objective Loss 0.268255 LR 0.000500 Time 0.027988 -2023-02-13 18:04:10,193 - Epoch: [104][ 110/ 1207] Overall Loss 0.268576 Objective Loss 0.268576 LR 0.000500 Time 0.027227 -2023-02-13 18:04:10,394 - Epoch: [104][ 120/ 1207] Overall Loss 0.268905 Objective Loss 0.268905 LR 0.000500 Time 0.026635 -2023-02-13 18:04:10,591 - Epoch: [104][ 130/ 1207] Overall Loss 0.269197 Objective Loss 0.269197 LR 0.000500 Time 0.026099 -2023-02-13 18:04:10,793 - Epoch: [104][ 140/ 1207] Overall Loss 0.270664 Objective Loss 0.270664 LR 0.000500 Time 0.025670 -2023-02-13 18:04:10,991 - Epoch: [104][ 150/ 1207] Overall Loss 0.269819 Objective Loss 0.269819 LR 0.000500 Time 0.025277 -2023-02-13 18:04:11,193 - Epoch: [104][ 160/ 1207] Overall Loss 0.269926 Objective Loss 0.269926 LR 0.000500 Time 0.024957 -2023-02-13 18:04:11,389 - Epoch: [104][ 170/ 1207] Overall Loss 0.268962 Objective Loss 0.268962 LR 0.000500 Time 0.024642 -2023-02-13 18:04:11,590 - Epoch: [104][ 180/ 1207] Overall Loss 0.268473 Objective Loss 0.268473 LR 0.000500 Time 0.024388 -2023-02-13 18:04:11,788 - Epoch: [104][ 190/ 1207] Overall Loss 0.267239 Objective Loss 0.267239 LR 0.000500 Time 0.024141 -2023-02-13 18:04:11,990 - Epoch: [104][ 200/ 1207] Overall Loss 0.268126 Objective Loss 0.268126 LR 0.000500 Time 0.023943 -2023-02-13 18:04:12,187 - Epoch: [104][ 210/ 1207] Overall Loss 0.268394 Objective Loss 0.268394 LR 0.000500 Time 0.023738 -2023-02-13 18:04:12,388 - Epoch: [104][ 220/ 1207] Overall Loss 0.270303 Objective Loss 0.270303 LR 0.000500 Time 0.023574 -2023-02-13 18:04:12,586 - Epoch: [104][ 230/ 1207] Overall Loss 0.270097 Objective Loss 0.270097 LR 0.000500 Time 0.023404 -2023-02-13 18:04:12,788 - Epoch: [104][ 240/ 1207] Overall Loss 0.270143 Objective Loss 0.270143 LR 0.000500 Time 0.023270 -2023-02-13 18:04:12,984 - Epoch: [104][ 250/ 1207] Overall Loss 0.269881 Objective Loss 0.269881 LR 0.000500 Time 0.023125 -2023-02-13 18:04:13,186 - Epoch: [104][ 260/ 1207] Overall Loss 0.269822 Objective Loss 0.269822 LR 0.000500 Time 0.023009 -2023-02-13 18:04:13,383 - Epoch: [104][ 270/ 1207] Overall Loss 0.268817 Objective Loss 0.268817 LR 0.000500 Time 0.022883 -2023-02-13 18:04:13,585 - Epoch: [104][ 280/ 1207] Overall Loss 0.269469 Objective Loss 0.269469 LR 0.000500 Time 0.022786 -2023-02-13 18:04:13,782 - Epoch: [104][ 290/ 1207] Overall Loss 0.270218 Objective Loss 0.270218 LR 0.000500 Time 0.022679 -2023-02-13 18:04:13,984 - Epoch: [104][ 300/ 1207] Overall Loss 0.269956 Objective Loss 0.269956 LR 0.000500 Time 0.022595 -2023-02-13 18:04:14,180 - Epoch: [104][ 310/ 1207] Overall Loss 0.268625 Objective Loss 0.268625 LR 0.000500 Time 0.022498 -2023-02-13 18:04:14,381 - Epoch: [104][ 320/ 1207] Overall Loss 0.268431 Objective Loss 0.268431 LR 0.000500 Time 0.022422 -2023-02-13 18:04:14,579 - Epoch: [104][ 330/ 1207] Overall Loss 0.267445 Objective Loss 0.267445 LR 0.000500 Time 0.022340 -2023-02-13 18:04:14,781 - Epoch: [104][ 340/ 1207] Overall Loss 0.268760 Objective Loss 0.268760 LR 0.000500 Time 0.022276 -2023-02-13 18:04:14,978 - Epoch: [104][ 350/ 1207] Overall Loss 0.269457 Objective Loss 0.269457 LR 0.000500 Time 0.022201 -2023-02-13 18:04:15,179 - Epoch: [104][ 360/ 1207] Overall Loss 0.269476 Objective Loss 0.269476 LR 0.000500 Time 0.022141 -2023-02-13 18:04:15,375 - Epoch: [104][ 370/ 1207] Overall Loss 0.268637 Objective Loss 0.268637 LR 0.000500 Time 0.022073 -2023-02-13 18:04:15,577 - Epoch: [104][ 380/ 1207] Overall Loss 0.268168 Objective Loss 0.268168 LR 0.000500 Time 0.022023 -2023-02-13 18:04:15,775 - Epoch: [104][ 390/ 1207] Overall Loss 0.268531 Objective Loss 0.268531 LR 0.000500 Time 0.021964 -2023-02-13 18:04:15,978 - Epoch: [104][ 400/ 1207] Overall Loss 0.268300 Objective Loss 0.268300 LR 0.000500 Time 0.021922 -2023-02-13 18:04:16,175 - Epoch: [104][ 410/ 1207] Overall Loss 0.267388 Objective Loss 0.267388 LR 0.000500 Time 0.021866 -2023-02-13 18:04:16,376 - Epoch: [104][ 420/ 1207] Overall Loss 0.267060 Objective Loss 0.267060 LR 0.000500 Time 0.021825 -2023-02-13 18:04:16,574 - Epoch: [104][ 430/ 1207] Overall Loss 0.266239 Objective Loss 0.266239 LR 0.000500 Time 0.021776 -2023-02-13 18:04:16,776 - Epoch: [104][ 440/ 1207] Overall Loss 0.266294 Objective Loss 0.266294 LR 0.000500 Time 0.021740 -2023-02-13 18:04:16,974 - Epoch: [104][ 450/ 1207] Overall Loss 0.266879 Objective Loss 0.266879 LR 0.000500 Time 0.021695 -2023-02-13 18:04:17,176 - Epoch: [104][ 460/ 1207] Overall Loss 0.266584 Objective Loss 0.266584 LR 0.000500 Time 0.021661 -2023-02-13 18:04:17,372 - Epoch: [104][ 470/ 1207] Overall Loss 0.266351 Objective Loss 0.266351 LR 0.000500 Time 0.021617 -2023-02-13 18:04:17,574 - Epoch: [104][ 480/ 1207] Overall Loss 0.266520 Objective Loss 0.266520 LR 0.000500 Time 0.021586 -2023-02-13 18:04:17,771 - Epoch: [104][ 490/ 1207] Overall Loss 0.266387 Objective Loss 0.266387 LR 0.000500 Time 0.021548 -2023-02-13 18:04:17,973 - Epoch: [104][ 500/ 1207] Overall Loss 0.266948 Objective Loss 0.266948 LR 0.000500 Time 0.021520 -2023-02-13 18:04:18,171 - Epoch: [104][ 510/ 1207] Overall Loss 0.267015 Objective Loss 0.267015 LR 0.000500 Time 0.021484 -2023-02-13 18:04:18,373 - Epoch: [104][ 520/ 1207] Overall Loss 0.267165 Objective Loss 0.267165 LR 0.000500 Time 0.021459 -2023-02-13 18:04:18,570 - Epoch: [104][ 530/ 1207] Overall Loss 0.267400 Objective Loss 0.267400 LR 0.000500 Time 0.021426 -2023-02-13 18:04:18,773 - Epoch: [104][ 540/ 1207] Overall Loss 0.267385 Objective Loss 0.267385 LR 0.000500 Time 0.021404 -2023-02-13 18:04:18,970 - Epoch: [104][ 550/ 1207] Overall Loss 0.268001 Objective Loss 0.268001 LR 0.000500 Time 0.021373 -2023-02-13 18:04:19,173 - Epoch: [104][ 560/ 1207] Overall Loss 0.268356 Objective Loss 0.268356 LR 0.000500 Time 0.021353 -2023-02-13 18:04:19,370 - Epoch: [104][ 570/ 1207] Overall Loss 0.268626 Objective Loss 0.268626 LR 0.000500 Time 0.021323 -2023-02-13 18:04:19,572 - Epoch: [104][ 580/ 1207] Overall Loss 0.268406 Objective Loss 0.268406 LR 0.000500 Time 0.021303 -2023-02-13 18:04:19,770 - Epoch: [104][ 590/ 1207] Overall Loss 0.268700 Objective Loss 0.268700 LR 0.000500 Time 0.021277 -2023-02-13 18:04:19,972 - Epoch: [104][ 600/ 1207] Overall Loss 0.269044 Objective Loss 0.269044 LR 0.000500 Time 0.021258 -2023-02-13 18:04:20,169 - Epoch: [104][ 610/ 1207] Overall Loss 0.269354 Objective Loss 0.269354 LR 0.000500 Time 0.021232 -2023-02-13 18:04:20,371 - Epoch: [104][ 620/ 1207] Overall Loss 0.269467 Objective Loss 0.269467 LR 0.000500 Time 0.021215 -2023-02-13 18:04:20,568 - Epoch: [104][ 630/ 1207] Overall Loss 0.269432 Objective Loss 0.269432 LR 0.000500 Time 0.021190 -2023-02-13 18:04:20,770 - Epoch: [104][ 640/ 1207] Overall Loss 0.269663 Objective Loss 0.269663 LR 0.000500 Time 0.021174 -2023-02-13 18:04:20,967 - Epoch: [104][ 650/ 1207] Overall Loss 0.269670 Objective Loss 0.269670 LR 0.000500 Time 0.021151 -2023-02-13 18:04:21,170 - Epoch: [104][ 660/ 1207] Overall Loss 0.269635 Objective Loss 0.269635 LR 0.000500 Time 0.021136 -2023-02-13 18:04:21,367 - Epoch: [104][ 670/ 1207] Overall Loss 0.270375 Objective Loss 0.270375 LR 0.000500 Time 0.021115 -2023-02-13 18:04:21,569 - Epoch: [104][ 680/ 1207] Overall Loss 0.269977 Objective Loss 0.269977 LR 0.000500 Time 0.021101 -2023-02-13 18:04:21,768 - Epoch: [104][ 690/ 1207] Overall Loss 0.269673 Objective Loss 0.269673 LR 0.000500 Time 0.021083 -2023-02-13 18:04:21,970 - Epoch: [104][ 700/ 1207] Overall Loss 0.269899 Objective Loss 0.269899 LR 0.000500 Time 0.021070 -2023-02-13 18:04:22,167 - Epoch: [104][ 710/ 1207] Overall Loss 0.269357 Objective Loss 0.269357 LR 0.000500 Time 0.021050 -2023-02-13 18:04:22,369 - Epoch: [104][ 720/ 1207] Overall Loss 0.268949 Objective Loss 0.268949 LR 0.000500 Time 0.021038 -2023-02-13 18:04:22,567 - Epoch: [104][ 730/ 1207] Overall Loss 0.268680 Objective Loss 0.268680 LR 0.000500 Time 0.021020 -2023-02-13 18:04:22,770 - Epoch: [104][ 740/ 1207] Overall Loss 0.268562 Objective Loss 0.268562 LR 0.000500 Time 0.021010 -2023-02-13 18:04:22,967 - Epoch: [104][ 750/ 1207] Overall Loss 0.268650 Objective Loss 0.268650 LR 0.000500 Time 0.020992 -2023-02-13 18:04:23,169 - Epoch: [104][ 760/ 1207] Overall Loss 0.268434 Objective Loss 0.268434 LR 0.000500 Time 0.020981 -2023-02-13 18:04:23,367 - Epoch: [104][ 770/ 1207] Overall Loss 0.268543 Objective Loss 0.268543 LR 0.000500 Time 0.020965 -2023-02-13 18:04:23,569 - Epoch: [104][ 780/ 1207] Overall Loss 0.268506 Objective Loss 0.268506 LR 0.000500 Time 0.020955 -2023-02-13 18:04:23,767 - Epoch: [104][ 790/ 1207] Overall Loss 0.268734 Objective Loss 0.268734 LR 0.000500 Time 0.020940 -2023-02-13 18:04:23,969 - Epoch: [104][ 800/ 1207] Overall Loss 0.268551 Objective Loss 0.268551 LR 0.000500 Time 0.020930 -2023-02-13 18:04:24,167 - Epoch: [104][ 810/ 1207] Overall Loss 0.268573 Objective Loss 0.268573 LR 0.000500 Time 0.020916 -2023-02-13 18:04:24,369 - Epoch: [104][ 820/ 1207] Overall Loss 0.268461 Objective Loss 0.268461 LR 0.000500 Time 0.020907 -2023-02-13 18:04:24,567 - Epoch: [104][ 830/ 1207] Overall Loss 0.268763 Objective Loss 0.268763 LR 0.000500 Time 0.020892 -2023-02-13 18:04:24,770 - Epoch: [104][ 840/ 1207] Overall Loss 0.268594 Objective Loss 0.268594 LR 0.000500 Time 0.020885 -2023-02-13 18:04:24,967 - Epoch: [104][ 850/ 1207] Overall Loss 0.269086 Objective Loss 0.269086 LR 0.000500 Time 0.020870 -2023-02-13 18:04:25,169 - Epoch: [104][ 860/ 1207] Overall Loss 0.268903 Objective Loss 0.268903 LR 0.000500 Time 0.020862 -2023-02-13 18:04:25,366 - Epoch: [104][ 870/ 1207] Overall Loss 0.269018 Objective Loss 0.269018 LR 0.000500 Time 0.020849 -2023-02-13 18:04:25,568 - Epoch: [104][ 880/ 1207] Overall Loss 0.268695 Objective Loss 0.268695 LR 0.000500 Time 0.020841 -2023-02-13 18:04:25,766 - Epoch: [104][ 890/ 1207] Overall Loss 0.268761 Objective Loss 0.268761 LR 0.000500 Time 0.020829 -2023-02-13 18:04:25,970 - Epoch: [104][ 900/ 1207] Overall Loss 0.268548 Objective Loss 0.268548 LR 0.000500 Time 0.020823 -2023-02-13 18:04:26,167 - Epoch: [104][ 910/ 1207] Overall Loss 0.268454 Objective Loss 0.268454 LR 0.000500 Time 0.020811 -2023-02-13 18:04:26,370 - Epoch: [104][ 920/ 1207] Overall Loss 0.268500 Objective Loss 0.268500 LR 0.000500 Time 0.020804 -2023-02-13 18:04:26,568 - Epoch: [104][ 930/ 1207] Overall Loss 0.268748 Objective Loss 0.268748 LR 0.000500 Time 0.020793 -2023-02-13 18:04:26,770 - Epoch: [104][ 940/ 1207] Overall Loss 0.268648 Objective Loss 0.268648 LR 0.000500 Time 0.020787 -2023-02-13 18:04:26,968 - Epoch: [104][ 950/ 1207] Overall Loss 0.268742 Objective Loss 0.268742 LR 0.000500 Time 0.020776 -2023-02-13 18:04:27,170 - Epoch: [104][ 960/ 1207] Overall Loss 0.268684 Objective Loss 0.268684 LR 0.000500 Time 0.020769 -2023-02-13 18:04:27,367 - Epoch: [104][ 970/ 1207] Overall Loss 0.268412 Objective Loss 0.268412 LR 0.000500 Time 0.020758 -2023-02-13 18:04:27,570 - Epoch: [104][ 980/ 1207] Overall Loss 0.268394 Objective Loss 0.268394 LR 0.000500 Time 0.020753 -2023-02-13 18:04:27,768 - Epoch: [104][ 990/ 1207] Overall Loss 0.268351 Objective Loss 0.268351 LR 0.000500 Time 0.020743 -2023-02-13 18:04:27,970 - Epoch: [104][ 1000/ 1207] Overall Loss 0.267952 Objective Loss 0.267952 LR 0.000500 Time 0.020737 -2023-02-13 18:04:28,167 - Epoch: [104][ 1010/ 1207] Overall Loss 0.267864 Objective Loss 0.267864 LR 0.000500 Time 0.020727 -2023-02-13 18:04:28,370 - Epoch: [104][ 1020/ 1207] Overall Loss 0.267900 Objective Loss 0.267900 LR 0.000500 Time 0.020721 -2023-02-13 18:04:28,567 - Epoch: [104][ 1030/ 1207] Overall Loss 0.267831 Objective Loss 0.267831 LR 0.000500 Time 0.020712 -2023-02-13 18:04:28,770 - Epoch: [104][ 1040/ 1207] Overall Loss 0.267819 Objective Loss 0.267819 LR 0.000500 Time 0.020708 -2023-02-13 18:04:28,968 - Epoch: [104][ 1050/ 1207] Overall Loss 0.267858 Objective Loss 0.267858 LR 0.000500 Time 0.020698 -2023-02-13 18:04:29,170 - Epoch: [104][ 1060/ 1207] Overall Loss 0.267775 Objective Loss 0.267775 LR 0.000500 Time 0.020693 -2023-02-13 18:04:29,368 - Epoch: [104][ 1070/ 1207] Overall Loss 0.267773 Objective Loss 0.267773 LR 0.000500 Time 0.020684 -2023-02-13 18:04:29,570 - Epoch: [104][ 1080/ 1207] Overall Loss 0.267827 Objective Loss 0.267827 LR 0.000500 Time 0.020679 -2023-02-13 18:04:29,768 - Epoch: [104][ 1090/ 1207] Overall Loss 0.267854 Objective Loss 0.267854 LR 0.000500 Time 0.020671 -2023-02-13 18:04:29,970 - Epoch: [104][ 1100/ 1207] Overall Loss 0.267855 Objective Loss 0.267855 LR 0.000500 Time 0.020667 -2023-02-13 18:04:30,168 - Epoch: [104][ 1110/ 1207] Overall Loss 0.267703 Objective Loss 0.267703 LR 0.000500 Time 0.020658 -2023-02-13 18:04:30,370 - Epoch: [104][ 1120/ 1207] Overall Loss 0.267641 Objective Loss 0.267641 LR 0.000500 Time 0.020654 -2023-02-13 18:04:30,568 - Epoch: [104][ 1130/ 1207] Overall Loss 0.267670 Objective Loss 0.267670 LR 0.000500 Time 0.020646 -2023-02-13 18:04:30,771 - Epoch: [104][ 1140/ 1207] Overall Loss 0.267701 Objective Loss 0.267701 LR 0.000500 Time 0.020642 -2023-02-13 18:04:30,968 - Epoch: [104][ 1150/ 1207] Overall Loss 0.267765 Objective Loss 0.267765 LR 0.000500 Time 0.020634 -2023-02-13 18:04:31,170 - Epoch: [104][ 1160/ 1207] Overall Loss 0.267903 Objective Loss 0.267903 LR 0.000500 Time 0.020630 -2023-02-13 18:04:31,368 - Epoch: [104][ 1170/ 1207] Overall Loss 0.267894 Objective Loss 0.267894 LR 0.000500 Time 0.020622 -2023-02-13 18:04:31,570 - Epoch: [104][ 1180/ 1207] Overall Loss 0.268110 Objective Loss 0.268110 LR 0.000500 Time 0.020619 -2023-02-13 18:04:31,768 - Epoch: [104][ 1190/ 1207] Overall Loss 0.268260 Objective Loss 0.268260 LR 0.000500 Time 0.020612 -2023-02-13 18:04:32,022 - Epoch: [104][ 1200/ 1207] Overall Loss 0.268442 Objective Loss 0.268442 LR 0.000500 Time 0.020651 -2023-02-13 18:04:32,136 - Epoch: [104][ 1207/ 1207] Overall Loss 0.268578 Objective Loss 0.268578 Top1 79.268293 Top5 97.256098 LR 0.000500 Time 0.020626 -2023-02-13 18:04:32,223 - --- validate (epoch=104)----------- -2023-02-13 18:04:32,223 - 34311 samples (256 per mini-batch) -2023-02-13 18:04:32,620 - Epoch: [104][ 10/ 135] Loss 0.317903 Top1 84.804688 Top5 97.421875 -2023-02-13 18:04:32,747 - Epoch: [104][ 20/ 135] Loss 0.330738 Top1 84.101562 Top5 97.558594 -2023-02-13 18:04:32,871 - Epoch: [104][ 30/ 135] Loss 0.338059 Top1 83.710938 Top5 97.382812 -2023-02-13 18:04:32,996 - Epoch: [104][ 40/ 135] Loss 0.333817 Top1 83.779297 Top5 97.500000 -2023-02-13 18:04:33,121 - Epoch: [104][ 50/ 135] Loss 0.333719 Top1 83.578125 Top5 97.390625 -2023-02-13 18:04:33,245 - Epoch: [104][ 60/ 135] Loss 0.338062 Top1 83.593750 Top5 97.324219 -2023-02-13 18:04:33,370 - Epoch: [104][ 70/ 135] Loss 0.332126 Top1 83.588170 Top5 97.321429 -2023-02-13 18:04:33,494 - Epoch: [104][ 80/ 135] Loss 0.327773 Top1 83.676758 Top5 97.431641 -2023-02-13 18:04:33,621 - Epoch: [104][ 90/ 135] Loss 0.328862 Top1 83.563368 Top5 97.434896 -2023-02-13 18:04:33,746 - Epoch: [104][ 100/ 135] Loss 0.330567 Top1 83.519531 Top5 97.437500 -2023-02-13 18:04:33,870 - Epoch: [104][ 110/ 135] Loss 0.327193 Top1 83.540483 Top5 97.478693 -2023-02-13 18:04:33,993 - Epoch: [104][ 120/ 135] Loss 0.325101 Top1 83.554688 Top5 97.513021 -2023-02-13 18:04:34,121 - Epoch: [104][ 130/ 135] Loss 0.325729 Top1 83.542668 Top5 97.500000 -2023-02-13 18:04:34,168 - Epoch: [104][ 135/ 135] Loss 0.323560 Top1 83.518405 Top5 97.508088 -2023-02-13 18:04:34,235 - ==> Top1: 83.518 Top5: 97.508 Loss: 0.324 - -2023-02-13 18:04:34,236 - ==> Confusion: -[[ 821 6 7 2 16 3 0 2 6 69 0 6 0 5 4 5 0 5 0 2 8] - [ 2 943 0 4 8 24 4 20 4 3 2 2 1 0 1 1 5 0 5 1 3] - [ 9 5 944 7 7 0 20 15 2 1 3 1 5 6 4 8 2 4 5 2 8] - [ 3 2 21 889 3 4 1 2 3 4 12 0 12 1 22 4 3 7 16 0 7] - [ 10 12 1 0 993 10 1 2 1 3 1 8 1 3 4 8 2 0 1 3 2] - [ 1 14 1 3 4 962 5 21 2 2 6 15 5 13 0 2 1 3 0 6 4] - [ 3 6 22 2 1 3 1024 9 0 0 3 3 2 0 0 4 1 5 0 8 3] - [ 4 12 10 0 4 27 2 925 0 1 2 6 5 1 0 1 0 1 13 8 2] - [ 16 3 0 1 2 0 0 3 885 48 9 3 0 9 18 2 0 1 6 1 2] - [ 62 3 3 1 10 2 0 1 31 867 0 2 0 18 2 1 1 2 1 0 5] - [ 0 4 5 11 0 2 3 4 18 2 967 4 0 11 4 0 1 1 11 0 3] - [ 1 3 0 0 5 10 0 4 1 1 1 924 25 2 0 7 2 7 2 5 5] - [ 0 0 1 4 3 3 0 2 3 0 0 27 881 1 1 5 3 13 3 1 8] - [ 3 2 5 0 7 8 0 2 17 19 9 10 2 915 3 6 1 5 1 1 8] - [ 16 3 2 7 8 6 0 2 21 10 2 2 2 1 982 1 3 8 7 0 9] - [ 3 2 5 0 8 0 2 1 1 0 0 9 7 1 1 977 5 16 1 4 3] - [ 0 6 1 1 10 2 0 3 2 0 1 3 2 3 3 14 996 1 1 5 7] - [ 3 4 0 2 0 1 1 0 0 0 0 10 17 1 1 19 0 983 1 2 6] - [ 3 6 8 7 4 2 0 36 3 1 4 2 5 0 15 0 0 2 984 3 1] - [ 0 3 1 0 3 7 6 15 0 0 1 25 4 3 0 7 5 4 0 1056 8] - [ 126 304 237 130 187 233 86 209 109 91 187 133 341 292 153 140 209 117 161 251 9738]] - -2023-02-13 18:04:34,237 - ==> Best [Top1: 83.912 Top5: 97.508 Sparsity:0.00 Params: 148928 on epoch: 103] -2023-02-13 18:04:34,238 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:04:34,243 - - -2023-02-13 18:04:34,243 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:04:35,148 - Epoch: [105][ 10/ 1207] Overall Loss 0.300480 Objective Loss 0.300480 LR 0.000500 Time 0.090433 -2023-02-13 18:04:35,347 - Epoch: [105][ 20/ 1207] Overall Loss 0.281701 Objective Loss 0.281701 LR 0.000500 Time 0.055111 -2023-02-13 18:04:35,541 - Epoch: [105][ 30/ 1207] Overall Loss 0.276304 Objective Loss 0.276304 LR 0.000500 Time 0.043204 -2023-02-13 18:04:35,737 - Epoch: [105][ 40/ 1207] Overall Loss 0.277294 Objective Loss 0.277294 LR 0.000500 Time 0.037307 -2023-02-13 18:04:35,933 - Epoch: [105][ 50/ 1207] Overall Loss 0.276437 Objective Loss 0.276437 LR 0.000500 Time 0.033741 -2023-02-13 18:04:36,129 - Epoch: [105][ 60/ 1207] Overall Loss 0.278719 Objective Loss 0.278719 LR 0.000500 Time 0.031382 -2023-02-13 18:04:36,323 - Epoch: [105][ 70/ 1207] Overall Loss 0.281091 Objective Loss 0.281091 LR 0.000500 Time 0.029670 -2023-02-13 18:04:36,519 - Epoch: [105][ 80/ 1207] Overall Loss 0.281626 Objective Loss 0.281626 LR 0.000500 Time 0.028406 -2023-02-13 18:04:36,713 - Epoch: [105][ 90/ 1207] Overall Loss 0.276834 Objective Loss 0.276834 LR 0.000500 Time 0.027400 -2023-02-13 18:04:36,909 - Epoch: [105][ 100/ 1207] Overall Loss 0.274548 Objective Loss 0.274548 LR 0.000500 Time 0.026618 -2023-02-13 18:04:37,104 - Epoch: [105][ 110/ 1207] Overall Loss 0.276318 Objective Loss 0.276318 LR 0.000500 Time 0.025963 -2023-02-13 18:04:37,299 - Epoch: [105][ 120/ 1207] Overall Loss 0.274349 Objective Loss 0.274349 LR 0.000500 Time 0.025423 -2023-02-13 18:04:37,493 - Epoch: [105][ 130/ 1207] Overall Loss 0.273813 Objective Loss 0.273813 LR 0.000500 Time 0.024955 -2023-02-13 18:04:37,689 - Epoch: [105][ 140/ 1207] Overall Loss 0.273695 Objective Loss 0.273695 LR 0.000500 Time 0.024575 -2023-02-13 18:04:37,884 - Epoch: [105][ 150/ 1207] Overall Loss 0.273392 Objective Loss 0.273392 LR 0.000500 Time 0.024231 -2023-02-13 18:04:38,080 - Epoch: [105][ 160/ 1207] Overall Loss 0.271538 Objective Loss 0.271538 LR 0.000500 Time 0.023938 -2023-02-13 18:04:38,274 - Epoch: [105][ 170/ 1207] Overall Loss 0.270093 Objective Loss 0.270093 LR 0.000500 Time 0.023670 -2023-02-13 18:04:38,470 - Epoch: [105][ 180/ 1207] Overall Loss 0.269180 Objective Loss 0.269180 LR 0.000500 Time 0.023443 -2023-02-13 18:04:38,665 - Epoch: [105][ 190/ 1207] Overall Loss 0.268826 Objective Loss 0.268826 LR 0.000500 Time 0.023231 -2023-02-13 18:04:38,861 - Epoch: [105][ 200/ 1207] Overall Loss 0.269697 Objective Loss 0.269697 LR 0.000500 Time 0.023051 -2023-02-13 18:04:39,056 - Epoch: [105][ 210/ 1207] Overall Loss 0.268024 Objective Loss 0.268024 LR 0.000500 Time 0.022876 -2023-02-13 18:04:39,251 - Epoch: [105][ 220/ 1207] Overall Loss 0.267701 Objective Loss 0.267701 LR 0.000500 Time 0.022724 -2023-02-13 18:04:39,445 - Epoch: [105][ 230/ 1207] Overall Loss 0.267373 Objective Loss 0.267373 LR 0.000500 Time 0.022577 -2023-02-13 18:04:39,641 - Epoch: [105][ 240/ 1207] Overall Loss 0.267451 Objective Loss 0.267451 LR 0.000500 Time 0.022451 -2023-02-13 18:04:39,837 - Epoch: [105][ 250/ 1207] Overall Loss 0.267171 Objective Loss 0.267171 LR 0.000500 Time 0.022337 -2023-02-13 18:04:40,035 - Epoch: [105][ 260/ 1207] Overall Loss 0.267373 Objective Loss 0.267373 LR 0.000500 Time 0.022238 -2023-02-13 18:04:40,232 - Epoch: [105][ 270/ 1207] Overall Loss 0.267092 Objective Loss 0.267092 LR 0.000500 Time 0.022140 -2023-02-13 18:04:40,429 - Epoch: [105][ 280/ 1207] Overall Loss 0.266920 Objective Loss 0.266920 LR 0.000500 Time 0.022053 -2023-02-13 18:04:40,625 - Epoch: [105][ 290/ 1207] Overall Loss 0.266679 Objective Loss 0.266679 LR 0.000500 Time 0.021968 -2023-02-13 18:04:40,824 - Epoch: [105][ 300/ 1207] Overall Loss 0.267197 Objective Loss 0.267197 LR 0.000500 Time 0.021895 -2023-02-13 18:04:41,021 - Epoch: [105][ 310/ 1207] Overall Loss 0.267591 Objective Loss 0.267591 LR 0.000500 Time 0.021825 -2023-02-13 18:04:41,219 - Epoch: [105][ 320/ 1207] Overall Loss 0.268026 Objective Loss 0.268026 LR 0.000500 Time 0.021760 -2023-02-13 18:04:41,416 - Epoch: [105][ 330/ 1207] Overall Loss 0.268067 Objective Loss 0.268067 LR 0.000500 Time 0.021697 -2023-02-13 18:04:41,615 - Epoch: [105][ 340/ 1207] Overall Loss 0.268020 Objective Loss 0.268020 LR 0.000500 Time 0.021641 -2023-02-13 18:04:41,812 - Epoch: [105][ 350/ 1207] Overall Loss 0.267650 Objective Loss 0.267650 LR 0.000500 Time 0.021586 -2023-02-13 18:04:42,011 - Epoch: [105][ 360/ 1207] Overall Loss 0.267157 Objective Loss 0.267157 LR 0.000500 Time 0.021539 -2023-02-13 18:04:42,208 - Epoch: [105][ 370/ 1207] Overall Loss 0.267895 Objective Loss 0.267895 LR 0.000500 Time 0.021486 -2023-02-13 18:04:42,406 - Epoch: [105][ 380/ 1207] Overall Loss 0.267569 Objective Loss 0.267569 LR 0.000500 Time 0.021443 -2023-02-13 18:04:42,603 - Epoch: [105][ 390/ 1207] Overall Loss 0.267381 Objective Loss 0.267381 LR 0.000500 Time 0.021396 -2023-02-13 18:04:42,802 - Epoch: [105][ 400/ 1207] Overall Loss 0.267665 Objective Loss 0.267665 LR 0.000500 Time 0.021358 -2023-02-13 18:04:42,998 - Epoch: [105][ 410/ 1207] Overall Loss 0.267494 Objective Loss 0.267494 LR 0.000500 Time 0.021315 -2023-02-13 18:04:43,196 - Epoch: [105][ 420/ 1207] Overall Loss 0.267286 Objective Loss 0.267286 LR 0.000500 Time 0.021278 -2023-02-13 18:04:43,393 - Epoch: [105][ 430/ 1207] Overall Loss 0.266619 Objective Loss 0.266619 LR 0.000500 Time 0.021240 -2023-02-13 18:04:43,592 - Epoch: [105][ 440/ 1207] Overall Loss 0.266924 Objective Loss 0.266924 LR 0.000500 Time 0.021209 -2023-02-13 18:04:43,789 - Epoch: [105][ 450/ 1207] Overall Loss 0.267026 Objective Loss 0.267026 LR 0.000500 Time 0.021174 -2023-02-13 18:04:43,987 - Epoch: [105][ 460/ 1207] Overall Loss 0.267664 Objective Loss 0.267664 LR 0.000500 Time 0.021143 -2023-02-13 18:04:44,184 - Epoch: [105][ 470/ 1207] Overall Loss 0.267942 Objective Loss 0.267942 LR 0.000500 Time 0.021111 -2023-02-13 18:04:44,383 - Epoch: [105][ 480/ 1207] Overall Loss 0.267863 Objective Loss 0.267863 LR 0.000500 Time 0.021086 -2023-02-13 18:04:44,580 - Epoch: [105][ 490/ 1207] Overall Loss 0.267652 Objective Loss 0.267652 LR 0.000500 Time 0.021057 -2023-02-13 18:04:44,779 - Epoch: [105][ 500/ 1207] Overall Loss 0.267683 Objective Loss 0.267683 LR 0.000500 Time 0.021033 -2023-02-13 18:04:44,976 - Epoch: [105][ 510/ 1207] Overall Loss 0.267600 Objective Loss 0.267600 LR 0.000500 Time 0.021006 -2023-02-13 18:04:45,174 - Epoch: [105][ 520/ 1207] Overall Loss 0.267986 Objective Loss 0.267986 LR 0.000500 Time 0.020982 -2023-02-13 18:04:45,370 - Epoch: [105][ 530/ 1207] Overall Loss 0.267501 Objective Loss 0.267501 LR 0.000500 Time 0.020955 -2023-02-13 18:04:45,569 - Epoch: [105][ 540/ 1207] Overall Loss 0.267571 Objective Loss 0.267571 LR 0.000500 Time 0.020934 -2023-02-13 18:04:45,766 - Epoch: [105][ 550/ 1207] Overall Loss 0.268067 Objective Loss 0.268067 LR 0.000500 Time 0.020912 -2023-02-13 18:04:45,966 - Epoch: [105][ 560/ 1207] Overall Loss 0.267838 Objective Loss 0.267838 LR 0.000500 Time 0.020894 -2023-02-13 18:04:46,162 - Epoch: [105][ 570/ 1207] Overall Loss 0.268006 Objective Loss 0.268006 LR 0.000500 Time 0.020872 -2023-02-13 18:04:46,360 - Epoch: [105][ 580/ 1207] Overall Loss 0.268207 Objective Loss 0.268207 LR 0.000500 Time 0.020852 -2023-02-13 18:04:46,556 - Epoch: [105][ 590/ 1207] Overall Loss 0.268225 Objective Loss 0.268225 LR 0.000500 Time 0.020831 -2023-02-13 18:04:46,756 - Epoch: [105][ 600/ 1207] Overall Loss 0.268258 Objective Loss 0.268258 LR 0.000500 Time 0.020816 -2023-02-13 18:04:46,953 - Epoch: [105][ 610/ 1207] Overall Loss 0.268001 Objective Loss 0.268001 LR 0.000500 Time 0.020797 -2023-02-13 18:04:47,152 - Epoch: [105][ 620/ 1207] Overall Loss 0.267870 Objective Loss 0.267870 LR 0.000500 Time 0.020781 -2023-02-13 18:04:47,349 - Epoch: [105][ 630/ 1207] Overall Loss 0.268052 Objective Loss 0.268052 LR 0.000500 Time 0.020763 -2023-02-13 18:04:47,547 - Epoch: [105][ 640/ 1207] Overall Loss 0.268126 Objective Loss 0.268126 LR 0.000500 Time 0.020748 -2023-02-13 18:04:47,745 - Epoch: [105][ 650/ 1207] Overall Loss 0.267821 Objective Loss 0.267821 LR 0.000500 Time 0.020733 -2023-02-13 18:04:47,943 - Epoch: [105][ 660/ 1207] Overall Loss 0.267850 Objective Loss 0.267850 LR 0.000500 Time 0.020718 -2023-02-13 18:04:48,139 - Epoch: [105][ 670/ 1207] Overall Loss 0.267873 Objective Loss 0.267873 LR 0.000500 Time 0.020702 -2023-02-13 18:04:48,338 - Epoch: [105][ 680/ 1207] Overall Loss 0.267917 Objective Loss 0.267917 LR 0.000500 Time 0.020689 -2023-02-13 18:04:48,535 - Epoch: [105][ 690/ 1207] Overall Loss 0.267938 Objective Loss 0.267938 LR 0.000500 Time 0.020675 -2023-02-13 18:04:48,734 - Epoch: [105][ 700/ 1207] Overall Loss 0.267693 Objective Loss 0.267693 LR 0.000500 Time 0.020663 -2023-02-13 18:04:48,930 - Epoch: [105][ 710/ 1207] Overall Loss 0.267419 Objective Loss 0.267419 LR 0.000500 Time 0.020648 -2023-02-13 18:04:49,129 - Epoch: [105][ 720/ 1207] Overall Loss 0.267459 Objective Loss 0.267459 LR 0.000500 Time 0.020636 -2023-02-13 18:04:49,325 - Epoch: [105][ 730/ 1207] Overall Loss 0.267550 Objective Loss 0.267550 LR 0.000500 Time 0.020622 -2023-02-13 18:04:49,524 - Epoch: [105][ 740/ 1207] Overall Loss 0.267625 Objective Loss 0.267625 LR 0.000500 Time 0.020611 -2023-02-13 18:04:49,720 - Epoch: [105][ 750/ 1207] Overall Loss 0.267160 Objective Loss 0.267160 LR 0.000500 Time 0.020598 -2023-02-13 18:04:49,919 - Epoch: [105][ 760/ 1207] Overall Loss 0.267028 Objective Loss 0.267028 LR 0.000500 Time 0.020587 -2023-02-13 18:04:50,116 - Epoch: [105][ 770/ 1207] Overall Loss 0.266998 Objective Loss 0.266998 LR 0.000500 Time 0.020575 -2023-02-13 18:04:50,315 - Epoch: [105][ 780/ 1207] Overall Loss 0.266844 Objective Loss 0.266844 LR 0.000500 Time 0.020567 -2023-02-13 18:04:50,512 - Epoch: [105][ 790/ 1207] Overall Loss 0.267207 Objective Loss 0.267207 LR 0.000500 Time 0.020555 -2023-02-13 18:04:50,711 - Epoch: [105][ 800/ 1207] Overall Loss 0.267278 Objective Loss 0.267278 LR 0.000500 Time 0.020547 -2023-02-13 18:04:50,910 - Epoch: [105][ 810/ 1207] Overall Loss 0.267901 Objective Loss 0.267901 LR 0.000500 Time 0.020538 -2023-02-13 18:04:51,108 - Epoch: [105][ 820/ 1207] Overall Loss 0.267974 Objective Loss 0.267974 LR 0.000500 Time 0.020529 -2023-02-13 18:04:51,305 - Epoch: [105][ 830/ 1207] Overall Loss 0.267871 Objective Loss 0.267871 LR 0.000500 Time 0.020518 -2023-02-13 18:04:51,504 - Epoch: [105][ 840/ 1207] Overall Loss 0.268213 Objective Loss 0.268213 LR 0.000500 Time 0.020510 -2023-02-13 18:04:51,701 - Epoch: [105][ 850/ 1207] Overall Loss 0.268343 Objective Loss 0.268343 LR 0.000500 Time 0.020500 -2023-02-13 18:04:51,901 - Epoch: [105][ 860/ 1207] Overall Loss 0.268642 Objective Loss 0.268642 LR 0.000500 Time 0.020493 -2023-02-13 18:04:52,097 - Epoch: [105][ 870/ 1207] Overall Loss 0.268438 Objective Loss 0.268438 LR 0.000500 Time 0.020483 -2023-02-13 18:04:52,296 - Epoch: [105][ 880/ 1207] Overall Loss 0.268212 Objective Loss 0.268212 LR 0.000500 Time 0.020476 -2023-02-13 18:04:52,493 - Epoch: [105][ 890/ 1207] Overall Loss 0.268170 Objective Loss 0.268170 LR 0.000500 Time 0.020467 -2023-02-13 18:04:52,692 - Epoch: [105][ 900/ 1207] Overall Loss 0.268156 Objective Loss 0.268156 LR 0.000500 Time 0.020461 -2023-02-13 18:04:52,890 - Epoch: [105][ 910/ 1207] Overall Loss 0.268184 Objective Loss 0.268184 LR 0.000500 Time 0.020452 -2023-02-13 18:04:53,088 - Epoch: [105][ 920/ 1207] Overall Loss 0.268006 Objective Loss 0.268006 LR 0.000500 Time 0.020445 -2023-02-13 18:04:53,286 - Epoch: [105][ 930/ 1207] Overall Loss 0.268311 Objective Loss 0.268311 LR 0.000500 Time 0.020437 -2023-02-13 18:04:53,485 - Epoch: [105][ 940/ 1207] Overall Loss 0.268402 Objective Loss 0.268402 LR 0.000500 Time 0.020431 -2023-02-13 18:04:53,681 - Epoch: [105][ 950/ 1207] Overall Loss 0.267951 Objective Loss 0.267951 LR 0.000500 Time 0.020422 -2023-02-13 18:04:53,880 - Epoch: [105][ 960/ 1207] Overall Loss 0.267968 Objective Loss 0.267968 LR 0.000500 Time 0.020417 -2023-02-13 18:04:54,078 - Epoch: [105][ 970/ 1207] Overall Loss 0.268072 Objective Loss 0.268072 LR 0.000500 Time 0.020410 -2023-02-13 18:04:54,276 - Epoch: [105][ 980/ 1207] Overall Loss 0.267831 Objective Loss 0.267831 LR 0.000500 Time 0.020403 -2023-02-13 18:04:54,474 - Epoch: [105][ 990/ 1207] Overall Loss 0.267827 Objective Loss 0.267827 LR 0.000500 Time 0.020397 -2023-02-13 18:04:54,673 - Epoch: [105][ 1000/ 1207] Overall Loss 0.268029 Objective Loss 0.268029 LR 0.000500 Time 0.020391 -2023-02-13 18:04:54,870 - Epoch: [105][ 1010/ 1207] Overall Loss 0.267944 Objective Loss 0.267944 LR 0.000500 Time 0.020384 -2023-02-13 18:04:55,068 - Epoch: [105][ 1020/ 1207] Overall Loss 0.268177 Objective Loss 0.268177 LR 0.000500 Time 0.020378 -2023-02-13 18:04:55,265 - Epoch: [105][ 1030/ 1207] Overall Loss 0.268175 Objective Loss 0.268175 LR 0.000500 Time 0.020371 -2023-02-13 18:04:55,464 - Epoch: [105][ 1040/ 1207] Overall Loss 0.268239 Objective Loss 0.268239 LR 0.000500 Time 0.020366 -2023-02-13 18:04:55,661 - Epoch: [105][ 1050/ 1207] Overall Loss 0.267992 Objective Loss 0.267992 LR 0.000500 Time 0.020360 -2023-02-13 18:04:55,861 - Epoch: [105][ 1060/ 1207] Overall Loss 0.267966 Objective Loss 0.267966 LR 0.000500 Time 0.020356 -2023-02-13 18:04:56,059 - Epoch: [105][ 1070/ 1207] Overall Loss 0.267841 Objective Loss 0.267841 LR 0.000500 Time 0.020350 -2023-02-13 18:04:56,257 - Epoch: [105][ 1080/ 1207] Overall Loss 0.267745 Objective Loss 0.267745 LR 0.000500 Time 0.020345 -2023-02-13 18:04:56,454 - Epoch: [105][ 1090/ 1207] Overall Loss 0.267596 Objective Loss 0.267596 LR 0.000500 Time 0.020339 -2023-02-13 18:04:56,653 - Epoch: [105][ 1100/ 1207] Overall Loss 0.267931 Objective Loss 0.267931 LR 0.000500 Time 0.020334 -2023-02-13 18:04:56,852 - Epoch: [105][ 1110/ 1207] Overall Loss 0.267712 Objective Loss 0.267712 LR 0.000500 Time 0.020330 -2023-02-13 18:04:57,050 - Epoch: [105][ 1120/ 1207] Overall Loss 0.267696 Objective Loss 0.267696 LR 0.000500 Time 0.020325 -2023-02-13 18:04:57,253 - Epoch: [105][ 1130/ 1207] Overall Loss 0.267839 Objective Loss 0.267839 LR 0.000500 Time 0.020324 -2023-02-13 18:04:57,466 - Epoch: [105][ 1140/ 1207] Overall Loss 0.267668 Objective Loss 0.267668 LR 0.000500 Time 0.020333 -2023-02-13 18:04:57,674 - Epoch: [105][ 1150/ 1207] Overall Loss 0.267648 Objective Loss 0.267648 LR 0.000500 Time 0.020336 -2023-02-13 18:04:57,889 - Epoch: [105][ 1160/ 1207] Overall Loss 0.267841 Objective Loss 0.267841 LR 0.000500 Time 0.020346 -2023-02-13 18:04:58,095 - Epoch: [105][ 1170/ 1207] Overall Loss 0.267834 Objective Loss 0.267834 LR 0.000500 Time 0.020348 -2023-02-13 18:04:58,309 - Epoch: [105][ 1180/ 1207] Overall Loss 0.267965 Objective Loss 0.267965 LR 0.000500 Time 0.020356 -2023-02-13 18:04:58,515 - Epoch: [105][ 1190/ 1207] Overall Loss 0.268079 Objective Loss 0.268079 LR 0.000500 Time 0.020358 -2023-02-13 18:04:58,779 - Epoch: [105][ 1200/ 1207] Overall Loss 0.268357 Objective Loss 0.268357 LR 0.000500 Time 0.020408 -2023-02-13 18:04:58,896 - Epoch: [105][ 1207/ 1207] Overall Loss 0.268318 Objective Loss 0.268318 Top1 84.451220 Top5 97.560976 LR 0.000500 Time 0.020386 -2023-02-13 18:04:58,968 - --- validate (epoch=105)----------- -2023-02-13 18:04:58,968 - 34311 samples (256 per mini-batch) -2023-02-13 18:04:59,370 - Epoch: [105][ 10/ 135] Loss 0.353395 Top1 82.460938 Top5 98.242188 -2023-02-13 18:04:59,500 - Epoch: [105][ 20/ 135] Loss 0.334851 Top1 82.910156 Top5 97.871094 -2023-02-13 18:04:59,627 - Epoch: [105][ 30/ 135] Loss 0.321922 Top1 83.125000 Top5 97.682292 -2023-02-13 18:04:59,757 - Epoch: [105][ 40/ 135] Loss 0.322147 Top1 83.457031 Top5 97.695312 -2023-02-13 18:04:59,882 - Epoch: [105][ 50/ 135] Loss 0.321902 Top1 83.664062 Top5 97.625000 -2023-02-13 18:05:00,011 - Epoch: [105][ 60/ 135] Loss 0.319862 Top1 83.697917 Top5 97.636719 -2023-02-13 18:05:00,139 - Epoch: [105][ 70/ 135] Loss 0.323606 Top1 83.738839 Top5 97.617188 -2023-02-13 18:05:00,268 - Epoch: [105][ 80/ 135] Loss 0.323669 Top1 83.798828 Top5 97.641602 -2023-02-13 18:05:00,397 - Epoch: [105][ 90/ 135] Loss 0.320354 Top1 83.923611 Top5 97.647569 -2023-02-13 18:05:00,525 - Epoch: [105][ 100/ 135] Loss 0.318444 Top1 83.937500 Top5 97.691406 -2023-02-13 18:05:00,654 - Epoch: [105][ 110/ 135] Loss 0.318760 Top1 83.987926 Top5 97.634943 -2023-02-13 18:05:00,782 - Epoch: [105][ 120/ 135] Loss 0.317412 Top1 84.075521 Top5 97.646484 -2023-02-13 18:05:00,912 - Epoch: [105][ 130/ 135] Loss 0.317472 Top1 84.104567 Top5 97.659255 -2023-02-13 18:05:00,958 - Epoch: [105][ 135/ 135] Loss 0.322678 Top1 84.104223 Top5 97.665472 -2023-02-13 18:05:01,036 - ==> Top1: 84.104 Top5: 97.665 Loss: 0.323 - -2023-02-13 18:05:01,037 - ==> Confusion: -[[ 844 6 4 0 12 5 0 2 4 45 1 6 0 10 7 5 2 4 0 1 9] - [ 2 927 0 3 11 38 4 18 4 0 2 4 2 0 2 0 2 0 6 2 6] - [ 8 2 920 24 6 1 27 16 0 1 2 1 2 4 4 11 1 3 12 5 8] - [ 5 1 13 896 4 4 0 2 2 1 14 1 12 0 21 3 4 6 20 1 6] - [ 13 9 0 0 976 11 1 1 1 3 1 8 1 3 12 6 9 3 0 2 6] - [ 2 13 0 4 5 973 4 20 1 1 2 13 5 12 0 1 6 1 2 1 4] - [ 4 5 15 0 1 7 1036 6 0 0 3 0 1 1 0 5 0 2 1 8 4] - [ 2 5 8 3 3 29 1 929 0 1 0 6 3 1 1 0 0 2 21 8 1] - [ 13 4 0 2 2 0 0 3 898 35 8 3 2 11 17 1 0 1 4 1 4] - [ 77 1 5 0 7 3 0 1 33 840 0 1 0 22 9 3 2 3 0 3 2] - [ 0 1 4 10 0 3 2 4 17 1 972 3 2 7 2 1 2 1 14 1 4] - [ 2 5 0 0 5 5 1 6 0 2 0 915 28 7 1 4 2 9 3 7 3] - [ 3 0 0 4 1 5 0 1 1 0 0 29 879 1 1 4 1 15 1 1 12] - [ 2 5 0 0 7 8 1 1 17 9 9 9 0 931 2 6 6 2 1 1 7] - [ 9 3 1 17 3 1 0 1 18 3 5 6 2 1 997 1 0 7 11 0 6] - [ 2 3 4 1 7 1 3 0 0 0 0 5 10 1 1 969 9 15 0 6 9] - [ 3 5 0 1 5 3 0 1 0 1 0 4 2 5 2 16 996 1 2 4 10] - [ 4 4 0 3 0 1 0 1 0 1 3 10 15 0 2 18 1 981 0 3 4] - [ 1 3 3 10 2 1 0 29 0 0 2 6 6 1 14 0 0 2 1002 3 1] - [ 0 3 0 1 2 6 5 12 0 0 1 15 3 2 0 4 3 5 0 1078 8] - [ 142 220 175 134 121 252 88 183 91 71 186 136 309 299 199 111 248 106 182 283 9898]] - -2023-02-13 18:05:01,038 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:05:01,038 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:05:01,045 - - -2023-02-13 18:05:01,045 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:05:01,992 - Epoch: [106][ 10/ 1207] Overall Loss 0.324827 Objective Loss 0.324827 LR 0.000500 Time 0.094602 -2023-02-13 18:05:02,186 - Epoch: [106][ 20/ 1207] Overall Loss 0.296655 Objective Loss 0.296655 LR 0.000500 Time 0.056979 -2023-02-13 18:05:02,375 - Epoch: [106][ 30/ 1207] Overall Loss 0.283305 Objective Loss 0.283305 LR 0.000500 Time 0.044278 -2023-02-13 18:05:02,562 - Epoch: [106][ 40/ 1207] Overall Loss 0.281411 Objective Loss 0.281411 LR 0.000500 Time 0.037879 -2023-02-13 18:05:02,750 - Epoch: [106][ 50/ 1207] Overall Loss 0.273605 Objective Loss 0.273605 LR 0.000500 Time 0.034050 -2023-02-13 18:05:02,937 - Epoch: [106][ 60/ 1207] Overall Loss 0.271228 Objective Loss 0.271228 LR 0.000500 Time 0.031498 -2023-02-13 18:05:03,125 - Epoch: [106][ 70/ 1207] Overall Loss 0.266954 Objective Loss 0.266954 LR 0.000500 Time 0.029672 -2023-02-13 18:05:03,312 - Epoch: [106][ 80/ 1207] Overall Loss 0.264467 Objective Loss 0.264467 LR 0.000500 Time 0.028301 -2023-02-13 18:05:03,499 - Epoch: [106][ 90/ 1207] Overall Loss 0.264846 Objective Loss 0.264846 LR 0.000500 Time 0.027231 -2023-02-13 18:05:03,687 - Epoch: [106][ 100/ 1207] Overall Loss 0.263395 Objective Loss 0.263395 LR 0.000500 Time 0.026380 -2023-02-13 18:05:03,875 - Epoch: [106][ 110/ 1207] Overall Loss 0.263995 Objective Loss 0.263995 LR 0.000500 Time 0.025691 -2023-02-13 18:05:04,063 - Epoch: [106][ 120/ 1207] Overall Loss 0.264321 Objective Loss 0.264321 LR 0.000500 Time 0.025109 -2023-02-13 18:05:04,250 - Epoch: [106][ 130/ 1207] Overall Loss 0.263910 Objective Loss 0.263910 LR 0.000500 Time 0.024617 -2023-02-13 18:05:04,438 - Epoch: [106][ 140/ 1207] Overall Loss 0.264162 Objective Loss 0.264162 LR 0.000500 Time 0.024193 -2023-02-13 18:05:04,625 - Epoch: [106][ 150/ 1207] Overall Loss 0.263852 Objective Loss 0.263852 LR 0.000500 Time 0.023827 -2023-02-13 18:05:04,812 - Epoch: [106][ 160/ 1207] Overall Loss 0.264891 Objective Loss 0.264891 LR 0.000500 Time 0.023508 -2023-02-13 18:05:05,000 - Epoch: [106][ 170/ 1207] Overall Loss 0.266231 Objective Loss 0.266231 LR 0.000500 Time 0.023223 -2023-02-13 18:05:05,187 - Epoch: [106][ 180/ 1207] Overall Loss 0.267427 Objective Loss 0.267427 LR 0.000500 Time 0.022970 -2023-02-13 18:05:05,375 - Epoch: [106][ 190/ 1207] Overall Loss 0.268617 Objective Loss 0.268617 LR 0.000500 Time 0.022752 -2023-02-13 18:05:05,563 - Epoch: [106][ 200/ 1207] Overall Loss 0.269761 Objective Loss 0.269761 LR 0.000500 Time 0.022549 -2023-02-13 18:05:05,750 - Epoch: [106][ 210/ 1207] Overall Loss 0.268668 Objective Loss 0.268668 LR 0.000500 Time 0.022366 -2023-02-13 18:05:05,939 - Epoch: [106][ 220/ 1207] Overall Loss 0.267600 Objective Loss 0.267600 LR 0.000500 Time 0.022206 -2023-02-13 18:05:06,127 - Epoch: [106][ 230/ 1207] Overall Loss 0.267778 Objective Loss 0.267778 LR 0.000500 Time 0.022057 -2023-02-13 18:05:06,314 - Epoch: [106][ 240/ 1207] Overall Loss 0.266465 Objective Loss 0.266465 LR 0.000500 Time 0.021917 -2023-02-13 18:05:06,502 - Epoch: [106][ 250/ 1207] Overall Loss 0.266025 Objective Loss 0.266025 LR 0.000500 Time 0.021790 -2023-02-13 18:05:06,690 - Epoch: [106][ 260/ 1207] Overall Loss 0.265868 Objective Loss 0.265868 LR 0.000500 Time 0.021672 -2023-02-13 18:05:06,879 - Epoch: [106][ 270/ 1207] Overall Loss 0.266233 Objective Loss 0.266233 LR 0.000500 Time 0.021569 -2023-02-13 18:05:07,067 - Epoch: [106][ 280/ 1207] Overall Loss 0.265573 Objective Loss 0.265573 LR 0.000500 Time 0.021467 -2023-02-13 18:05:07,254 - Epoch: [106][ 290/ 1207] Overall Loss 0.264594 Objective Loss 0.264594 LR 0.000500 Time 0.021374 -2023-02-13 18:05:07,441 - Epoch: [106][ 300/ 1207] Overall Loss 0.264786 Objective Loss 0.264786 LR 0.000500 Time 0.021282 -2023-02-13 18:05:07,629 - Epoch: [106][ 310/ 1207] Overall Loss 0.263427 Objective Loss 0.263427 LR 0.000500 Time 0.021200 -2023-02-13 18:05:07,817 - Epoch: [106][ 320/ 1207] Overall Loss 0.263456 Objective Loss 0.263456 LR 0.000500 Time 0.021123 -2023-02-13 18:05:08,004 - Epoch: [106][ 330/ 1207] Overall Loss 0.264040 Objective Loss 0.264040 LR 0.000500 Time 0.021050 -2023-02-13 18:05:08,191 - Epoch: [106][ 340/ 1207] Overall Loss 0.263833 Objective Loss 0.263833 LR 0.000500 Time 0.020981 -2023-02-13 18:05:08,379 - Epoch: [106][ 350/ 1207] Overall Loss 0.264293 Objective Loss 0.264293 LR 0.000500 Time 0.020916 -2023-02-13 18:05:08,566 - Epoch: [106][ 360/ 1207] Overall Loss 0.263902 Objective Loss 0.263902 LR 0.000500 Time 0.020853 -2023-02-13 18:05:08,754 - Epoch: [106][ 370/ 1207] Overall Loss 0.263696 Objective Loss 0.263696 LR 0.000500 Time 0.020797 -2023-02-13 18:05:08,942 - Epoch: [106][ 380/ 1207] Overall Loss 0.263672 Objective Loss 0.263672 LR 0.000500 Time 0.020744 -2023-02-13 18:05:09,130 - Epoch: [106][ 390/ 1207] Overall Loss 0.263544 Objective Loss 0.263544 LR 0.000500 Time 0.020693 -2023-02-13 18:05:09,317 - Epoch: [106][ 400/ 1207] Overall Loss 0.263133 Objective Loss 0.263133 LR 0.000500 Time 0.020643 -2023-02-13 18:05:09,505 - Epoch: [106][ 410/ 1207] Overall Loss 0.262535 Objective Loss 0.262535 LR 0.000500 Time 0.020596 -2023-02-13 18:05:09,693 - Epoch: [106][ 420/ 1207] Overall Loss 0.262830 Objective Loss 0.262830 LR 0.000500 Time 0.020552 -2023-02-13 18:05:09,881 - Epoch: [106][ 430/ 1207] Overall Loss 0.262186 Objective Loss 0.262186 LR 0.000500 Time 0.020511 -2023-02-13 18:05:10,068 - Epoch: [106][ 440/ 1207] Overall Loss 0.261765 Objective Loss 0.261765 LR 0.000500 Time 0.020470 -2023-02-13 18:05:10,257 - Epoch: [106][ 450/ 1207] Overall Loss 0.261961 Objective Loss 0.261961 LR 0.000500 Time 0.020433 -2023-02-13 18:05:10,444 - Epoch: [106][ 460/ 1207] Overall Loss 0.263107 Objective Loss 0.263107 LR 0.000500 Time 0.020394 -2023-02-13 18:05:10,632 - Epoch: [106][ 470/ 1207] Overall Loss 0.263655 Objective Loss 0.263655 LR 0.000500 Time 0.020361 -2023-02-13 18:05:10,820 - Epoch: [106][ 480/ 1207] Overall Loss 0.263319 Objective Loss 0.263319 LR 0.000500 Time 0.020328 -2023-02-13 18:05:11,009 - Epoch: [106][ 490/ 1207] Overall Loss 0.262940 Objective Loss 0.262940 LR 0.000500 Time 0.020298 -2023-02-13 18:05:11,197 - Epoch: [106][ 500/ 1207] Overall Loss 0.262639 Objective Loss 0.262639 LR 0.000500 Time 0.020265 -2023-02-13 18:05:11,385 - Epoch: [106][ 510/ 1207] Overall Loss 0.262649 Objective Loss 0.262649 LR 0.000500 Time 0.020237 -2023-02-13 18:05:11,573 - Epoch: [106][ 520/ 1207] Overall Loss 0.263289 Objective Loss 0.263289 LR 0.000500 Time 0.020208 -2023-02-13 18:05:11,761 - Epoch: [106][ 530/ 1207] Overall Loss 0.262849 Objective Loss 0.262849 LR 0.000500 Time 0.020181 -2023-02-13 18:05:11,950 - Epoch: [106][ 540/ 1207] Overall Loss 0.263061 Objective Loss 0.263061 LR 0.000500 Time 0.020156 -2023-02-13 18:05:12,137 - Epoch: [106][ 550/ 1207] Overall Loss 0.262830 Objective Loss 0.262830 LR 0.000500 Time 0.020130 -2023-02-13 18:05:12,324 - Epoch: [106][ 560/ 1207] Overall Loss 0.262772 Objective Loss 0.262772 LR 0.000500 Time 0.020104 -2023-02-13 18:05:12,512 - Epoch: [106][ 570/ 1207] Overall Loss 0.262800 Objective Loss 0.262800 LR 0.000500 Time 0.020079 -2023-02-13 18:05:12,699 - Epoch: [106][ 580/ 1207] Overall Loss 0.263324 Objective Loss 0.263324 LR 0.000500 Time 0.020056 -2023-02-13 18:05:12,887 - Epoch: [106][ 590/ 1207] Overall Loss 0.263290 Objective Loss 0.263290 LR 0.000500 Time 0.020034 -2023-02-13 18:05:13,075 - Epoch: [106][ 600/ 1207] Overall Loss 0.263659 Objective Loss 0.263659 LR 0.000500 Time 0.020012 -2023-02-13 18:05:13,263 - Epoch: [106][ 610/ 1207] Overall Loss 0.263546 Objective Loss 0.263546 LR 0.000500 Time 0.019992 -2023-02-13 18:05:13,450 - Epoch: [106][ 620/ 1207] Overall Loss 0.263358 Objective Loss 0.263358 LR 0.000500 Time 0.019971 -2023-02-13 18:05:13,638 - Epoch: [106][ 630/ 1207] Overall Loss 0.263146 Objective Loss 0.263146 LR 0.000500 Time 0.019952 -2023-02-13 18:05:13,826 - Epoch: [106][ 640/ 1207] Overall Loss 0.263476 Objective Loss 0.263476 LR 0.000500 Time 0.019933 -2023-02-13 18:05:14,014 - Epoch: [106][ 650/ 1207] Overall Loss 0.263694 Objective Loss 0.263694 LR 0.000500 Time 0.019914 -2023-02-13 18:05:14,202 - Epoch: [106][ 660/ 1207] Overall Loss 0.264123 Objective Loss 0.264123 LR 0.000500 Time 0.019897 -2023-02-13 18:05:14,390 - Epoch: [106][ 670/ 1207] Overall Loss 0.264079 Objective Loss 0.264079 LR 0.000500 Time 0.019881 -2023-02-13 18:05:14,578 - Epoch: [106][ 680/ 1207] Overall Loss 0.264008 Objective Loss 0.264008 LR 0.000500 Time 0.019864 -2023-02-13 18:05:14,766 - Epoch: [106][ 690/ 1207] Overall Loss 0.264109 Objective Loss 0.264109 LR 0.000500 Time 0.019848 -2023-02-13 18:05:14,954 - Epoch: [106][ 700/ 1207] Overall Loss 0.264447 Objective Loss 0.264447 LR 0.000500 Time 0.019832 -2023-02-13 18:05:15,142 - Epoch: [106][ 710/ 1207] Overall Loss 0.264977 Objective Loss 0.264977 LR 0.000500 Time 0.019818 -2023-02-13 18:05:15,330 - Epoch: [106][ 720/ 1207] Overall Loss 0.265027 Objective Loss 0.265027 LR 0.000500 Time 0.019803 -2023-02-13 18:05:15,518 - Epoch: [106][ 730/ 1207] Overall Loss 0.265256 Objective Loss 0.265256 LR 0.000500 Time 0.019789 -2023-02-13 18:05:15,706 - Epoch: [106][ 740/ 1207] Overall Loss 0.265369 Objective Loss 0.265369 LR 0.000500 Time 0.019775 -2023-02-13 18:05:15,897 - Epoch: [106][ 750/ 1207] Overall Loss 0.265144 Objective Loss 0.265144 LR 0.000500 Time 0.019765 -2023-02-13 18:05:16,084 - Epoch: [106][ 760/ 1207] Overall Loss 0.265005 Objective Loss 0.265005 LR 0.000500 Time 0.019751 -2023-02-13 18:05:16,272 - Epoch: [106][ 770/ 1207] Overall Loss 0.264835 Objective Loss 0.264835 LR 0.000500 Time 0.019738 -2023-02-13 18:05:16,460 - Epoch: [106][ 780/ 1207] Overall Loss 0.265031 Objective Loss 0.265031 LR 0.000500 Time 0.019726 -2023-02-13 18:05:16,649 - Epoch: [106][ 790/ 1207] Overall Loss 0.265158 Objective Loss 0.265158 LR 0.000500 Time 0.019714 -2023-02-13 18:05:16,837 - Epoch: [106][ 800/ 1207] Overall Loss 0.264950 Objective Loss 0.264950 LR 0.000500 Time 0.019703 -2023-02-13 18:05:17,025 - Epoch: [106][ 810/ 1207] Overall Loss 0.265175 Objective Loss 0.265175 LR 0.000500 Time 0.019691 -2023-02-13 18:05:17,213 - Epoch: [106][ 820/ 1207] Overall Loss 0.264733 Objective Loss 0.264733 LR 0.000500 Time 0.019679 -2023-02-13 18:05:17,401 - Epoch: [106][ 830/ 1207] Overall Loss 0.264883 Objective Loss 0.264883 LR 0.000500 Time 0.019668 -2023-02-13 18:05:17,588 - Epoch: [106][ 840/ 1207] Overall Loss 0.264724 Objective Loss 0.264724 LR 0.000500 Time 0.019657 -2023-02-13 18:05:17,777 - Epoch: [106][ 850/ 1207] Overall Loss 0.265019 Objective Loss 0.265019 LR 0.000500 Time 0.019647 -2023-02-13 18:05:17,964 - Epoch: [106][ 860/ 1207] Overall Loss 0.265268 Objective Loss 0.265268 LR 0.000500 Time 0.019636 -2023-02-13 18:05:18,153 - Epoch: [106][ 870/ 1207] Overall Loss 0.265292 Objective Loss 0.265292 LR 0.000500 Time 0.019627 -2023-02-13 18:05:18,341 - Epoch: [106][ 880/ 1207] Overall Loss 0.265360 Objective Loss 0.265360 LR 0.000500 Time 0.019617 -2023-02-13 18:05:18,529 - Epoch: [106][ 890/ 1207] Overall Loss 0.265493 Objective Loss 0.265493 LR 0.000500 Time 0.019608 -2023-02-13 18:05:18,717 - Epoch: [106][ 900/ 1207] Overall Loss 0.265406 Objective Loss 0.265406 LR 0.000500 Time 0.019599 -2023-02-13 18:05:18,906 - Epoch: [106][ 910/ 1207] Overall Loss 0.265389 Objective Loss 0.265389 LR 0.000500 Time 0.019591 -2023-02-13 18:05:19,094 - Epoch: [106][ 920/ 1207] Overall Loss 0.265618 Objective Loss 0.265618 LR 0.000500 Time 0.019581 -2023-02-13 18:05:19,282 - Epoch: [106][ 930/ 1207] Overall Loss 0.265683 Objective Loss 0.265683 LR 0.000500 Time 0.019572 -2023-02-13 18:05:19,471 - Epoch: [106][ 940/ 1207] Overall Loss 0.265654 Objective Loss 0.265654 LR 0.000500 Time 0.019564 -2023-02-13 18:05:19,660 - Epoch: [106][ 950/ 1207] Overall Loss 0.266063 Objective Loss 0.266063 LR 0.000500 Time 0.019557 -2023-02-13 18:05:19,849 - Epoch: [106][ 960/ 1207] Overall Loss 0.266135 Objective Loss 0.266135 LR 0.000500 Time 0.019550 -2023-02-13 18:05:20,038 - Epoch: [106][ 970/ 1207] Overall Loss 0.266032 Objective Loss 0.266032 LR 0.000500 Time 0.019543 -2023-02-13 18:05:20,227 - Epoch: [106][ 980/ 1207] Overall Loss 0.265977 Objective Loss 0.265977 LR 0.000500 Time 0.019536 -2023-02-13 18:05:20,416 - Epoch: [106][ 990/ 1207] Overall Loss 0.265807 Objective Loss 0.265807 LR 0.000500 Time 0.019529 -2023-02-13 18:05:20,605 - Epoch: [106][ 1000/ 1207] Overall Loss 0.266057 Objective Loss 0.266057 LR 0.000500 Time 0.019522 -2023-02-13 18:05:20,794 - Epoch: [106][ 1010/ 1207] Overall Loss 0.265969 Objective Loss 0.265969 LR 0.000500 Time 0.019516 -2023-02-13 18:05:20,985 - Epoch: [106][ 1020/ 1207] Overall Loss 0.265965 Objective Loss 0.265965 LR 0.000500 Time 0.019512 -2023-02-13 18:05:21,174 - Epoch: [106][ 1030/ 1207] Overall Loss 0.265937 Objective Loss 0.265937 LR 0.000500 Time 0.019506 -2023-02-13 18:05:21,363 - Epoch: [106][ 1040/ 1207] Overall Loss 0.266095 Objective Loss 0.266095 LR 0.000500 Time 0.019499 -2023-02-13 18:05:21,552 - Epoch: [106][ 1050/ 1207] Overall Loss 0.266187 Objective Loss 0.266187 LR 0.000500 Time 0.019493 -2023-02-13 18:05:21,741 - Epoch: [106][ 1060/ 1207] Overall Loss 0.266249 Objective Loss 0.266249 LR 0.000500 Time 0.019487 -2023-02-13 18:05:21,933 - Epoch: [106][ 1070/ 1207] Overall Loss 0.266609 Objective Loss 0.266609 LR 0.000500 Time 0.019484 -2023-02-13 18:05:22,123 - Epoch: [106][ 1080/ 1207] Overall Loss 0.266833 Objective Loss 0.266833 LR 0.000500 Time 0.019479 -2023-02-13 18:05:22,313 - Epoch: [106][ 1090/ 1207] Overall Loss 0.266791 Objective Loss 0.266791 LR 0.000500 Time 0.019475 -2023-02-13 18:05:22,502 - Epoch: [106][ 1100/ 1207] Overall Loss 0.266802 Objective Loss 0.266802 LR 0.000500 Time 0.019469 -2023-02-13 18:05:22,691 - Epoch: [106][ 1110/ 1207] Overall Loss 0.267016 Objective Loss 0.267016 LR 0.000500 Time 0.019464 -2023-02-13 18:05:22,881 - Epoch: [106][ 1120/ 1207] Overall Loss 0.267265 Objective Loss 0.267265 LR 0.000500 Time 0.019459 -2023-02-13 18:05:23,071 - Epoch: [106][ 1130/ 1207] Overall Loss 0.267227 Objective Loss 0.267227 LR 0.000500 Time 0.019454 -2023-02-13 18:05:23,259 - Epoch: [106][ 1140/ 1207] Overall Loss 0.267288 Objective Loss 0.267288 LR 0.000500 Time 0.019449 -2023-02-13 18:05:23,449 - Epoch: [106][ 1150/ 1207] Overall Loss 0.267149 Objective Loss 0.267149 LR 0.000500 Time 0.019444 -2023-02-13 18:05:23,638 - Epoch: [106][ 1160/ 1207] Overall Loss 0.267061 Objective Loss 0.267061 LR 0.000500 Time 0.019439 -2023-02-13 18:05:23,827 - Epoch: [106][ 1170/ 1207] Overall Loss 0.267002 Objective Loss 0.267002 LR 0.000500 Time 0.019435 -2023-02-13 18:05:24,016 - Epoch: [106][ 1180/ 1207] Overall Loss 0.267211 Objective Loss 0.267211 LR 0.000500 Time 0.019430 -2023-02-13 18:05:24,206 - Epoch: [106][ 1190/ 1207] Overall Loss 0.267148 Objective Loss 0.267148 LR 0.000500 Time 0.019426 -2023-02-13 18:05:24,452 - Epoch: [106][ 1200/ 1207] Overall Loss 0.267284 Objective Loss 0.267284 LR 0.000500 Time 0.019468 -2023-02-13 18:05:24,567 - Epoch: [106][ 1207/ 1207] Overall Loss 0.267334 Objective Loss 0.267334 Top1 89.939024 Top5 97.256098 LR 0.000500 Time 0.019451 -2023-02-13 18:05:24,649 - --- validate (epoch=106)----------- -2023-02-13 18:05:24,649 - 34311 samples (256 per mini-batch) -2023-02-13 18:05:25,061 - Epoch: [106][ 10/ 135] Loss 0.329244 Top1 83.828125 Top5 97.890625 -2023-02-13 18:05:25,204 - Epoch: [106][ 20/ 135] Loss 0.339541 Top1 84.023438 Top5 97.558594 -2023-02-13 18:05:25,343 - Epoch: [106][ 30/ 135] Loss 0.343308 Top1 83.541667 Top5 97.356771 -2023-02-13 18:05:25,483 - Epoch: [106][ 40/ 135] Loss 0.345094 Top1 83.535156 Top5 97.353516 -2023-02-13 18:05:25,606 - Epoch: [106][ 50/ 135] Loss 0.340950 Top1 83.515625 Top5 97.421875 -2023-02-13 18:05:25,731 - Epoch: [106][ 60/ 135] Loss 0.334866 Top1 83.802083 Top5 97.460938 -2023-02-13 18:05:25,856 - Epoch: [106][ 70/ 135] Loss 0.328917 Top1 83.844866 Top5 97.500000 -2023-02-13 18:05:25,994 - Epoch: [106][ 80/ 135] Loss 0.327885 Top1 83.867188 Top5 97.500000 -2023-02-13 18:05:26,130 - Epoch: [106][ 90/ 135] Loss 0.328917 Top1 83.819444 Top5 97.517361 -2023-02-13 18:05:26,261 - Epoch: [106][ 100/ 135] Loss 0.326187 Top1 83.871094 Top5 97.519531 -2023-02-13 18:05:26,391 - Epoch: [106][ 110/ 135] Loss 0.325616 Top1 83.924006 Top5 97.517756 -2023-02-13 18:05:26,522 - Epoch: [106][ 120/ 135] Loss 0.323301 Top1 84.013672 Top5 97.532552 -2023-02-13 18:05:26,655 - Epoch: [106][ 130/ 135] Loss 0.323087 Top1 83.972356 Top5 97.512019 -2023-02-13 18:05:26,703 - Epoch: [106][ 135/ 135] Loss 0.319781 Top1 83.978899 Top5 97.525575 -2023-02-13 18:05:26,775 - ==> Top1: 83.979 Top5: 97.526 Loss: 0.320 - -2023-02-13 18:05:26,776 - ==> Confusion: -[[ 843 6 8 1 14 2 1 3 4 50 1 3 1 4 5 7 2 1 1 1 9] - [ 1 950 0 2 11 16 4 18 3 1 2 3 2 0 1 1 4 0 6 2 6] - [ 6 5 936 11 6 2 28 21 0 1 3 1 2 6 3 7 3 2 6 3 6] - [ 3 2 13 903 5 5 2 2 2 1 11 1 3 1 15 0 4 8 28 0 7] - [ 11 11 1 0 991 8 1 0 3 3 1 7 1 3 6 6 7 1 0 1 4] - [ 3 30 0 3 7 964 7 15 1 2 0 8 4 11 0 1 3 1 2 5 3] - [ 5 5 13 2 0 3 1040 5 0 1 3 2 0 0 0 3 0 5 0 5 7] - [ 2 7 8 2 4 22 6 921 0 1 1 8 3 1 0 0 0 2 19 13 4] - [ 19 5 0 2 2 0 0 2 892 37 10 1 0 6 13 1 2 5 8 1 3] - [ 86 2 2 0 8 1 0 2 42 836 1 1 1 16 3 1 1 2 0 1 6] - [ 1 1 2 10 1 1 3 1 15 2 984 1 1 8 4 0 1 1 12 0 2] - [ 2 4 1 0 3 9 1 5 1 1 0 898 27 4 0 13 5 16 3 8 4] - [ 1 1 0 6 4 6 0 1 1 0 0 20 866 0 2 7 1 28 0 1 14] - [ 5 3 3 0 10 7 0 3 13 14 10 6 2 924 3 4 6 3 0 1 7] - [ 9 5 3 17 7 3 0 2 20 9 2 2 3 1 980 1 1 7 11 0 9] - [ 2 3 7 0 11 0 5 0 0 0 0 5 4 1 1 968 7 19 1 5 7] - [ 1 5 1 1 9 4 0 2 1 1 1 1 2 1 0 13 1004 0 2 2 10] - [ 4 2 1 4 2 1 2 1 0 1 2 3 8 1 0 5 0 1006 1 1 6] - [ 3 5 5 10 1 1 0 31 3 0 3 1 5 1 11 2 1 1 994 5 3] - [ 0 3 1 1 1 7 7 14 0 0 1 14 2 5 0 8 5 6 1 1064 8] - [ 138 310 208 153 143 209 100 163 88 87 225 96 302 259 146 92 282 148 182 253 9850]] - -2023-02-13 18:05:26,778 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:05:26,778 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:05:26,784 - - -2023-02-13 18:05:26,784 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:05:27,761 - Epoch: [107][ 10/ 1207] Overall Loss 0.253009 Objective Loss 0.253009 LR 0.000500 Time 0.097696 -2023-02-13 18:05:27,952 - Epoch: [107][ 20/ 1207] Overall Loss 0.261628 Objective Loss 0.261628 LR 0.000500 Time 0.058383 -2023-02-13 18:05:28,141 - Epoch: [107][ 30/ 1207] Overall Loss 0.255114 Objective Loss 0.255114 LR 0.000500 Time 0.045181 -2023-02-13 18:05:28,328 - Epoch: [107][ 40/ 1207] Overall Loss 0.252080 Objective Loss 0.252080 LR 0.000500 Time 0.038556 -2023-02-13 18:05:28,515 - Epoch: [107][ 50/ 1207] Overall Loss 0.250769 Objective Loss 0.250769 LR 0.000500 Time 0.034589 -2023-02-13 18:05:28,703 - Epoch: [107][ 60/ 1207] Overall Loss 0.254455 Objective Loss 0.254455 LR 0.000500 Time 0.031946 -2023-02-13 18:05:28,891 - Epoch: [107][ 70/ 1207] Overall Loss 0.254656 Objective Loss 0.254656 LR 0.000500 Time 0.030060 -2023-02-13 18:05:29,078 - Epoch: [107][ 80/ 1207] Overall Loss 0.252914 Objective Loss 0.252914 LR 0.000500 Time 0.028643 -2023-02-13 18:05:29,267 - Epoch: [107][ 90/ 1207] Overall Loss 0.251503 Objective Loss 0.251503 LR 0.000500 Time 0.027549 -2023-02-13 18:05:29,455 - Epoch: [107][ 100/ 1207] Overall Loss 0.250299 Objective Loss 0.250299 LR 0.000500 Time 0.026676 -2023-02-13 18:05:29,644 - Epoch: [107][ 110/ 1207] Overall Loss 0.250467 Objective Loss 0.250467 LR 0.000500 Time 0.025965 -2023-02-13 18:05:29,833 - Epoch: [107][ 120/ 1207] Overall Loss 0.250584 Objective Loss 0.250584 LR 0.000500 Time 0.025367 -2023-02-13 18:05:30,022 - Epoch: [107][ 130/ 1207] Overall Loss 0.253954 Objective Loss 0.253954 LR 0.000500 Time 0.024874 -2023-02-13 18:05:30,212 - Epoch: [107][ 140/ 1207] Overall Loss 0.254513 Objective Loss 0.254513 LR 0.000500 Time 0.024446 -2023-02-13 18:05:30,401 - Epoch: [107][ 150/ 1207] Overall Loss 0.255508 Objective Loss 0.255508 LR 0.000500 Time 0.024073 -2023-02-13 18:05:30,589 - Epoch: [107][ 160/ 1207] Overall Loss 0.255439 Objective Loss 0.255439 LR 0.000500 Time 0.023745 -2023-02-13 18:05:30,778 - Epoch: [107][ 170/ 1207] Overall Loss 0.255984 Objective Loss 0.255984 LR 0.000500 Time 0.023460 -2023-02-13 18:05:30,969 - Epoch: [107][ 180/ 1207] Overall Loss 0.255461 Objective Loss 0.255461 LR 0.000500 Time 0.023213 -2023-02-13 18:05:31,158 - Epoch: [107][ 190/ 1207] Overall Loss 0.254433 Objective Loss 0.254433 LR 0.000500 Time 0.022984 -2023-02-13 18:05:31,347 - Epoch: [107][ 200/ 1207] Overall Loss 0.253442 Objective Loss 0.253442 LR 0.000500 Time 0.022777 -2023-02-13 18:05:31,535 - Epoch: [107][ 210/ 1207] Overall Loss 0.253427 Objective Loss 0.253427 LR 0.000500 Time 0.022588 -2023-02-13 18:05:31,737 - Epoch: [107][ 220/ 1207] Overall Loss 0.254730 Objective Loss 0.254730 LR 0.000500 Time 0.022476 -2023-02-13 18:05:31,935 - Epoch: [107][ 230/ 1207] Overall Loss 0.255118 Objective Loss 0.255118 LR 0.000500 Time 0.022359 -2023-02-13 18:05:32,136 - Epoch: [107][ 240/ 1207] Overall Loss 0.255800 Objective Loss 0.255800 LR 0.000500 Time 0.022263 -2023-02-13 18:05:32,335 - Epoch: [107][ 250/ 1207] Overall Loss 0.257152 Objective Loss 0.257152 LR 0.000500 Time 0.022168 -2023-02-13 18:05:32,537 - Epoch: [107][ 260/ 1207] Overall Loss 0.256914 Objective Loss 0.256914 LR 0.000500 Time 0.022090 -2023-02-13 18:05:32,735 - Epoch: [107][ 270/ 1207] Overall Loss 0.257005 Objective Loss 0.257005 LR 0.000500 Time 0.022002 -2023-02-13 18:05:32,937 - Epoch: [107][ 280/ 1207] Overall Loss 0.256396 Objective Loss 0.256396 LR 0.000500 Time 0.021938 -2023-02-13 18:05:33,135 - Epoch: [107][ 290/ 1207] Overall Loss 0.257329 Objective Loss 0.257329 LR 0.000500 Time 0.021864 -2023-02-13 18:05:33,337 - Epoch: [107][ 300/ 1207] Overall Loss 0.257221 Objective Loss 0.257221 LR 0.000500 Time 0.021805 -2023-02-13 18:05:33,535 - Epoch: [107][ 310/ 1207] Overall Loss 0.256673 Objective Loss 0.256673 LR 0.000500 Time 0.021741 -2023-02-13 18:05:33,737 - Epoch: [107][ 320/ 1207] Overall Loss 0.257008 Objective Loss 0.257008 LR 0.000500 Time 0.021691 -2023-02-13 18:05:33,936 - Epoch: [107][ 330/ 1207] Overall Loss 0.257875 Objective Loss 0.257875 LR 0.000500 Time 0.021637 -2023-02-13 18:05:34,138 - Epoch: [107][ 340/ 1207] Overall Loss 0.259299 Objective Loss 0.259299 LR 0.000500 Time 0.021591 -2023-02-13 18:05:34,336 - Epoch: [107][ 350/ 1207] Overall Loss 0.259112 Objective Loss 0.259112 LR 0.000500 Time 0.021539 -2023-02-13 18:05:34,537 - Epoch: [107][ 360/ 1207] Overall Loss 0.259006 Objective Loss 0.259006 LR 0.000500 Time 0.021499 -2023-02-13 18:05:34,735 - Epoch: [107][ 370/ 1207] Overall Loss 0.258562 Objective Loss 0.258562 LR 0.000500 Time 0.021451 -2023-02-13 18:05:34,937 - Epoch: [107][ 380/ 1207] Overall Loss 0.259436 Objective Loss 0.259436 LR 0.000500 Time 0.021418 -2023-02-13 18:05:35,135 - Epoch: [107][ 390/ 1207] Overall Loss 0.259915 Objective Loss 0.259915 LR 0.000500 Time 0.021375 -2023-02-13 18:05:35,337 - Epoch: [107][ 400/ 1207] Overall Loss 0.259388 Objective Loss 0.259388 LR 0.000500 Time 0.021344 -2023-02-13 18:05:35,535 - Epoch: [107][ 410/ 1207] Overall Loss 0.259285 Objective Loss 0.259285 LR 0.000500 Time 0.021306 -2023-02-13 18:05:35,736 - Epoch: [107][ 420/ 1207] Overall Loss 0.259507 Objective Loss 0.259507 LR 0.000500 Time 0.021277 -2023-02-13 18:05:35,937 - Epoch: [107][ 430/ 1207] Overall Loss 0.259319 Objective Loss 0.259319 LR 0.000500 Time 0.021248 -2023-02-13 18:05:36,138 - Epoch: [107][ 440/ 1207] Overall Loss 0.259656 Objective Loss 0.259656 LR 0.000500 Time 0.021221 -2023-02-13 18:05:36,336 - Epoch: [107][ 450/ 1207] Overall Loss 0.259298 Objective Loss 0.259298 LR 0.000500 Time 0.021188 -2023-02-13 18:05:36,537 - Epoch: [107][ 460/ 1207] Overall Loss 0.259302 Objective Loss 0.259302 LR 0.000500 Time 0.021165 -2023-02-13 18:05:36,735 - Epoch: [107][ 470/ 1207] Overall Loss 0.259235 Objective Loss 0.259235 LR 0.000500 Time 0.021135 -2023-02-13 18:05:36,938 - Epoch: [107][ 480/ 1207] Overall Loss 0.259466 Objective Loss 0.259466 LR 0.000500 Time 0.021116 -2023-02-13 18:05:37,136 - Epoch: [107][ 490/ 1207] Overall Loss 0.259438 Objective Loss 0.259438 LR 0.000500 Time 0.021088 -2023-02-13 18:05:37,338 - Epoch: [107][ 500/ 1207] Overall Loss 0.259747 Objective Loss 0.259747 LR 0.000500 Time 0.021071 -2023-02-13 18:05:37,537 - Epoch: [107][ 510/ 1207] Overall Loss 0.259934 Objective Loss 0.259934 LR 0.000500 Time 0.021047 -2023-02-13 18:05:37,739 - Epoch: [107][ 520/ 1207] Overall Loss 0.260210 Objective Loss 0.260210 LR 0.000500 Time 0.021029 -2023-02-13 18:05:37,938 - Epoch: [107][ 530/ 1207] Overall Loss 0.260221 Objective Loss 0.260221 LR 0.000500 Time 0.021008 -2023-02-13 18:05:38,140 - Epoch: [107][ 540/ 1207] Overall Loss 0.260250 Objective Loss 0.260250 LR 0.000500 Time 0.020993 -2023-02-13 18:05:38,339 - Epoch: [107][ 550/ 1207] Overall Loss 0.260592 Objective Loss 0.260592 LR 0.000500 Time 0.020971 -2023-02-13 18:05:38,541 - Epoch: [107][ 560/ 1207] Overall Loss 0.261216 Objective Loss 0.261216 LR 0.000500 Time 0.020956 -2023-02-13 18:05:38,739 - Epoch: [107][ 570/ 1207] Overall Loss 0.261967 Objective Loss 0.261967 LR 0.000500 Time 0.020936 -2023-02-13 18:05:38,941 - Epoch: [107][ 580/ 1207] Overall Loss 0.261782 Objective Loss 0.261782 LR 0.000500 Time 0.020923 -2023-02-13 18:05:39,139 - Epoch: [107][ 590/ 1207] Overall Loss 0.261801 Objective Loss 0.261801 LR 0.000500 Time 0.020903 -2023-02-13 18:05:39,341 - Epoch: [107][ 600/ 1207] Overall Loss 0.261632 Objective Loss 0.261632 LR 0.000500 Time 0.020890 -2023-02-13 18:05:39,539 - Epoch: [107][ 610/ 1207] Overall Loss 0.261134 Objective Loss 0.261134 LR 0.000500 Time 0.020872 -2023-02-13 18:05:39,741 - Epoch: [107][ 620/ 1207] Overall Loss 0.261148 Objective Loss 0.261148 LR 0.000500 Time 0.020860 -2023-02-13 18:05:39,939 - Epoch: [107][ 630/ 1207] Overall Loss 0.261064 Objective Loss 0.261064 LR 0.000500 Time 0.020843 -2023-02-13 18:05:40,140 - Epoch: [107][ 640/ 1207] Overall Loss 0.260677 Objective Loss 0.260677 LR 0.000500 Time 0.020830 -2023-02-13 18:05:40,339 - Epoch: [107][ 650/ 1207] Overall Loss 0.260721 Objective Loss 0.260721 LR 0.000500 Time 0.020815 -2023-02-13 18:05:40,540 - Epoch: [107][ 660/ 1207] Overall Loss 0.260357 Objective Loss 0.260357 LR 0.000500 Time 0.020805 -2023-02-13 18:05:40,739 - Epoch: [107][ 670/ 1207] Overall Loss 0.260313 Objective Loss 0.260313 LR 0.000500 Time 0.020790 -2023-02-13 18:05:40,943 - Epoch: [107][ 680/ 1207] Overall Loss 0.260600 Objective Loss 0.260600 LR 0.000500 Time 0.020784 -2023-02-13 18:05:41,141 - Epoch: [107][ 690/ 1207] Overall Loss 0.260301 Objective Loss 0.260301 LR 0.000500 Time 0.020770 -2023-02-13 18:05:41,344 - Epoch: [107][ 700/ 1207] Overall Loss 0.260405 Objective Loss 0.260405 LR 0.000500 Time 0.020762 -2023-02-13 18:05:41,543 - Epoch: [107][ 710/ 1207] Overall Loss 0.260620 Objective Loss 0.260620 LR 0.000500 Time 0.020749 -2023-02-13 18:05:41,745 - Epoch: [107][ 720/ 1207] Overall Loss 0.260568 Objective Loss 0.260568 LR 0.000500 Time 0.020742 -2023-02-13 18:05:41,944 - Epoch: [107][ 730/ 1207] Overall Loss 0.260677 Objective Loss 0.260677 LR 0.000500 Time 0.020730 -2023-02-13 18:05:42,146 - Epoch: [107][ 740/ 1207] Overall Loss 0.260669 Objective Loss 0.260669 LR 0.000500 Time 0.020721 -2023-02-13 18:05:42,344 - Epoch: [107][ 750/ 1207] Overall Loss 0.260462 Objective Loss 0.260462 LR 0.000500 Time 0.020709 -2023-02-13 18:05:42,546 - Epoch: [107][ 760/ 1207] Overall Loss 0.260617 Objective Loss 0.260617 LR 0.000500 Time 0.020701 -2023-02-13 18:05:42,746 - Epoch: [107][ 770/ 1207] Overall Loss 0.260638 Objective Loss 0.260638 LR 0.000500 Time 0.020692 -2023-02-13 18:05:42,949 - Epoch: [107][ 780/ 1207] Overall Loss 0.260495 Objective Loss 0.260495 LR 0.000500 Time 0.020686 -2023-02-13 18:05:43,147 - Epoch: [107][ 790/ 1207] Overall Loss 0.260135 Objective Loss 0.260135 LR 0.000500 Time 0.020674 -2023-02-13 18:05:43,348 - Epoch: [107][ 800/ 1207] Overall Loss 0.260176 Objective Loss 0.260176 LR 0.000500 Time 0.020667 -2023-02-13 18:05:43,553 - Epoch: [107][ 810/ 1207] Overall Loss 0.259692 Objective Loss 0.259692 LR 0.000500 Time 0.020664 -2023-02-13 18:05:43,758 - Epoch: [107][ 820/ 1207] Overall Loss 0.259855 Objective Loss 0.259855 LR 0.000500 Time 0.020662 -2023-02-13 18:05:43,963 - Epoch: [107][ 830/ 1207] Overall Loss 0.259907 Objective Loss 0.259907 LR 0.000500 Time 0.020659 -2023-02-13 18:05:44,167 - Epoch: [107][ 840/ 1207] Overall Loss 0.259696 Objective Loss 0.259696 LR 0.000500 Time 0.020656 -2023-02-13 18:05:44,371 - Epoch: [107][ 850/ 1207] Overall Loss 0.259808 Objective Loss 0.259808 LR 0.000500 Time 0.020653 -2023-02-13 18:05:44,576 - Epoch: [107][ 860/ 1207] Overall Loss 0.259659 Objective Loss 0.259659 LR 0.000500 Time 0.020651 -2023-02-13 18:05:44,780 - Epoch: [107][ 870/ 1207] Overall Loss 0.259677 Objective Loss 0.259677 LR 0.000500 Time 0.020647 -2023-02-13 18:05:44,986 - Epoch: [107][ 880/ 1207] Overall Loss 0.259775 Objective Loss 0.259775 LR 0.000500 Time 0.020646 -2023-02-13 18:05:45,189 - Epoch: [107][ 890/ 1207] Overall Loss 0.259703 Objective Loss 0.259703 LR 0.000500 Time 0.020642 -2023-02-13 18:05:45,395 - Epoch: [107][ 900/ 1207] Overall Loss 0.259727 Objective Loss 0.259727 LR 0.000500 Time 0.020641 -2023-02-13 18:05:45,599 - Epoch: [107][ 910/ 1207] Overall Loss 0.259764 Objective Loss 0.259764 LR 0.000500 Time 0.020638 -2023-02-13 18:05:45,804 - Epoch: [107][ 920/ 1207] Overall Loss 0.260124 Objective Loss 0.260124 LR 0.000500 Time 0.020636 -2023-02-13 18:05:46,010 - Epoch: [107][ 930/ 1207] Overall Loss 0.260392 Objective Loss 0.260392 LR 0.000500 Time 0.020636 -2023-02-13 18:05:46,215 - Epoch: [107][ 940/ 1207] Overall Loss 0.260532 Objective Loss 0.260532 LR 0.000500 Time 0.020634 -2023-02-13 18:05:46,421 - Epoch: [107][ 950/ 1207] Overall Loss 0.260774 Objective Loss 0.260774 LR 0.000500 Time 0.020632 -2023-02-13 18:05:46,626 - Epoch: [107][ 960/ 1207] Overall Loss 0.260807 Objective Loss 0.260807 LR 0.000500 Time 0.020631 -2023-02-13 18:05:46,830 - Epoch: [107][ 970/ 1207] Overall Loss 0.260982 Objective Loss 0.260982 LR 0.000500 Time 0.020628 -2023-02-13 18:05:47,036 - Epoch: [107][ 980/ 1207] Overall Loss 0.261126 Objective Loss 0.261126 LR 0.000500 Time 0.020628 -2023-02-13 18:05:47,240 - Epoch: [107][ 990/ 1207] Overall Loss 0.261064 Objective Loss 0.261064 LR 0.000500 Time 0.020625 -2023-02-13 18:05:47,445 - Epoch: [107][ 1000/ 1207] Overall Loss 0.261157 Objective Loss 0.261157 LR 0.000500 Time 0.020623 -2023-02-13 18:05:47,650 - Epoch: [107][ 1010/ 1207] Overall Loss 0.261215 Objective Loss 0.261215 LR 0.000500 Time 0.020621 -2023-02-13 18:05:47,854 - Epoch: [107][ 1020/ 1207] Overall Loss 0.261063 Objective Loss 0.261063 LR 0.000500 Time 0.020619 -2023-02-13 18:05:48,059 - Epoch: [107][ 1030/ 1207] Overall Loss 0.261069 Objective Loss 0.261069 LR 0.000500 Time 0.020618 -2023-02-13 18:05:48,264 - Epoch: [107][ 1040/ 1207] Overall Loss 0.260992 Objective Loss 0.260992 LR 0.000500 Time 0.020616 -2023-02-13 18:05:48,469 - Epoch: [107][ 1050/ 1207] Overall Loss 0.260916 Objective Loss 0.260916 LR 0.000500 Time 0.020614 -2023-02-13 18:05:48,674 - Epoch: [107][ 1060/ 1207] Overall Loss 0.260916 Objective Loss 0.260916 LR 0.000500 Time 0.020613 -2023-02-13 18:05:48,878 - Epoch: [107][ 1070/ 1207] Overall Loss 0.260968 Objective Loss 0.260968 LR 0.000500 Time 0.020611 -2023-02-13 18:05:49,083 - Epoch: [107][ 1080/ 1207] Overall Loss 0.261084 Objective Loss 0.261084 LR 0.000500 Time 0.020609 -2023-02-13 18:05:49,287 - Epoch: [107][ 1090/ 1207] Overall Loss 0.261235 Objective Loss 0.261235 LR 0.000500 Time 0.020607 -2023-02-13 18:05:49,493 - Epoch: [107][ 1100/ 1207] Overall Loss 0.261433 Objective Loss 0.261433 LR 0.000500 Time 0.020606 -2023-02-13 18:05:49,697 - Epoch: [107][ 1110/ 1207] Overall Loss 0.261599 Objective Loss 0.261599 LR 0.000500 Time 0.020604 -2023-02-13 18:05:49,903 - Epoch: [107][ 1120/ 1207] Overall Loss 0.261642 Objective Loss 0.261642 LR 0.000500 Time 0.020604 -2023-02-13 18:05:50,108 - Epoch: [107][ 1130/ 1207] Overall Loss 0.261855 Objective Loss 0.261855 LR 0.000500 Time 0.020602 -2023-02-13 18:05:50,313 - Epoch: [107][ 1140/ 1207] Overall Loss 0.262299 Objective Loss 0.262299 LR 0.000500 Time 0.020601 -2023-02-13 18:05:50,517 - Epoch: [107][ 1150/ 1207] Overall Loss 0.262157 Objective Loss 0.262157 LR 0.000500 Time 0.020600 -2023-02-13 18:05:50,722 - Epoch: [107][ 1160/ 1207] Overall Loss 0.262164 Objective Loss 0.262164 LR 0.000500 Time 0.020598 -2023-02-13 18:05:50,928 - Epoch: [107][ 1170/ 1207] Overall Loss 0.262130 Objective Loss 0.262130 LR 0.000500 Time 0.020598 -2023-02-13 18:05:51,133 - Epoch: [107][ 1180/ 1207] Overall Loss 0.262113 Objective Loss 0.262113 LR 0.000500 Time 0.020596 -2023-02-13 18:05:51,338 - Epoch: [107][ 1190/ 1207] Overall Loss 0.261994 Objective Loss 0.261994 LR 0.000500 Time 0.020596 -2023-02-13 18:05:51,596 - Epoch: [107][ 1200/ 1207] Overall Loss 0.262013 Objective Loss 0.262013 LR 0.000500 Time 0.020638 -2023-02-13 18:05:51,710 - Epoch: [107][ 1207/ 1207] Overall Loss 0.262223 Objective Loss 0.262223 Top1 83.536585 Top5 97.560976 LR 0.000500 Time 0.020613 -2023-02-13 18:05:51,789 - --- validate (epoch=107)----------- -2023-02-13 18:05:51,789 - 34311 samples (256 per mini-batch) -2023-02-13 18:05:52,181 - Epoch: [107][ 10/ 135] Loss 0.302610 Top1 83.398438 Top5 97.695312 -2023-02-13 18:05:52,315 - Epoch: [107][ 20/ 135] Loss 0.318107 Top1 83.398438 Top5 97.519531 -2023-02-13 18:05:52,448 - Epoch: [107][ 30/ 135] Loss 0.325555 Top1 83.489583 Top5 97.382812 -2023-02-13 18:05:52,577 - Epoch: [107][ 40/ 135] Loss 0.313891 Top1 83.984375 Top5 97.568359 -2023-02-13 18:05:52,701 - Epoch: [107][ 50/ 135] Loss 0.314348 Top1 84.078125 Top5 97.656250 -2023-02-13 18:05:52,829 - Epoch: [107][ 60/ 135] Loss 0.319689 Top1 83.932292 Top5 97.610677 -2023-02-13 18:05:52,956 - Epoch: [107][ 70/ 135] Loss 0.327187 Top1 83.537946 Top5 97.566964 -2023-02-13 18:05:53,084 - Epoch: [107][ 80/ 135] Loss 0.332238 Top1 83.535156 Top5 97.568359 -2023-02-13 18:05:53,211 - Epoch: [107][ 90/ 135] Loss 0.331702 Top1 83.498264 Top5 97.573785 -2023-02-13 18:05:53,339 - Epoch: [107][ 100/ 135] Loss 0.327553 Top1 83.613281 Top5 97.625000 -2023-02-13 18:05:53,468 - Epoch: [107][ 110/ 135] Loss 0.329152 Top1 83.583097 Top5 97.620739 -2023-02-13 18:05:53,595 - Epoch: [107][ 120/ 135] Loss 0.325502 Top1 83.658854 Top5 97.630208 -2023-02-13 18:05:53,725 - Epoch: [107][ 130/ 135] Loss 0.326704 Top1 83.605769 Top5 97.617188 -2023-02-13 18:05:53,770 - Epoch: [107][ 135/ 135] Loss 0.327786 Top1 83.579610 Top5 97.592609 -2023-02-13 18:05:53,854 - ==> Top1: 83.580 Top5: 97.593 Loss: 0.328 - -2023-02-13 18:05:53,855 - ==> Confusion: -[[ 868 5 9 3 9 2 0 2 1 30 2 6 1 5 7 4 5 0 0 2 6] - [ 1 946 0 1 7 28 5 18 3 0 2 1 1 0 1 1 8 0 3 2 5] - [ 7 4 946 15 5 1 23 13 0 0 2 1 1 6 5 7 3 2 9 2 6] - [ 2 4 16 906 3 3 1 2 1 3 14 1 4 1 21 2 2 7 20 1 2] - [ 19 10 1 2 970 13 1 0 1 2 2 7 2 4 7 5 7 2 0 3 8] - [ 2 20 1 6 6 974 3 15 2 1 3 9 3 10 0 1 5 0 1 4 4] - [ 4 3 8 1 0 3 1047 8 1 0 2 1 4 1 0 3 2 2 1 6 2] - [ 3 10 9 0 2 27 6 925 1 0 3 5 3 1 0 0 1 1 17 7 3] - [ 26 1 0 2 1 0 0 4 880 25 16 2 1 11 22 2 2 1 8 0 5] - [ 115 1 4 0 7 2 0 4 39 799 0 1 1 20 7 2 1 2 0 2 5] - [ 0 1 3 4 0 2 6 2 10 1 986 3 2 11 3 0 3 1 9 0 4] - [ 2 5 1 0 3 18 1 9 0 0 0 898 29 8 0 7 2 8 1 10 3] - [ 2 0 0 7 2 3 0 2 1 0 0 25 880 1 6 5 2 13 2 0 8] - [ 3 1 3 0 9 9 1 1 16 11 14 5 3 924 3 2 6 4 1 2 6] - [ 9 3 3 19 3 2 0 3 15 3 6 3 1 1 994 2 1 5 11 0 8] - [ 4 1 6 0 6 0 2 0 0 1 0 6 9 2 1 968 8 15 1 7 9] - [ 0 4 0 1 5 3 0 2 1 0 0 1 2 3 2 8 1006 2 4 7 10] - [ 3 2 1 4 2 0 4 3 0 1 0 10 14 0 1 19 0 977 0 5 5] - [ 2 3 3 11 0 2 0 28 4 0 6 2 4 1 11 1 1 2 1000 3 2] - [ 1 2 1 2 3 10 12 9 0 0 2 18 1 4 0 3 8 2 2 1063 5] - [ 180 261 248 148 102 240 120 173 81 55 200 113 322 283 176 138 284 111 201 278 9720]] - -2023-02-13 18:05:53,856 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:05:53,856 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:05:53,862 - - -2023-02-13 18:05:53,862 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:05:54,754 - Epoch: [108][ 10/ 1207] Overall Loss 0.277408 Objective Loss 0.277408 LR 0.000500 Time 0.089086 -2023-02-13 18:05:54,948 - Epoch: [108][ 20/ 1207] Overall Loss 0.265991 Objective Loss 0.265991 LR 0.000500 Time 0.054267 -2023-02-13 18:05:55,138 - Epoch: [108][ 30/ 1207] Overall Loss 0.264088 Objective Loss 0.264088 LR 0.000500 Time 0.042469 -2023-02-13 18:05:55,326 - Epoch: [108][ 40/ 1207] Overall Loss 0.258920 Objective Loss 0.258920 LR 0.000500 Time 0.036550 -2023-02-13 18:05:55,514 - Epoch: [108][ 50/ 1207] Overall Loss 0.263318 Objective Loss 0.263318 LR 0.000500 Time 0.032988 -2023-02-13 18:05:55,702 - Epoch: [108][ 60/ 1207] Overall Loss 0.262284 Objective Loss 0.262284 LR 0.000500 Time 0.030616 -2023-02-13 18:05:55,890 - Epoch: [108][ 70/ 1207] Overall Loss 0.265100 Objective Loss 0.265100 LR 0.000500 Time 0.028924 -2023-02-13 18:05:56,079 - Epoch: [108][ 80/ 1207] Overall Loss 0.266460 Objective Loss 0.266460 LR 0.000500 Time 0.027669 -2023-02-13 18:05:56,267 - Epoch: [108][ 90/ 1207] Overall Loss 0.265389 Objective Loss 0.265389 LR 0.000500 Time 0.026676 -2023-02-13 18:05:56,455 - Epoch: [108][ 100/ 1207] Overall Loss 0.264992 Objective Loss 0.264992 LR 0.000500 Time 0.025889 -2023-02-13 18:05:56,643 - Epoch: [108][ 110/ 1207] Overall Loss 0.265476 Objective Loss 0.265476 LR 0.000500 Time 0.025239 -2023-02-13 18:05:56,831 - Epoch: [108][ 120/ 1207] Overall Loss 0.263592 Objective Loss 0.263592 LR 0.000500 Time 0.024702 -2023-02-13 18:05:57,030 - Epoch: [108][ 130/ 1207] Overall Loss 0.262579 Objective Loss 0.262579 LR 0.000500 Time 0.024331 -2023-02-13 18:05:57,228 - Epoch: [108][ 140/ 1207] Overall Loss 0.260593 Objective Loss 0.260593 LR 0.000500 Time 0.024004 -2023-02-13 18:05:57,430 - Epoch: [108][ 150/ 1207] Overall Loss 0.258865 Objective Loss 0.258865 LR 0.000500 Time 0.023748 -2023-02-13 18:05:57,629 - Epoch: [108][ 160/ 1207] Overall Loss 0.258947 Objective Loss 0.258947 LR 0.000500 Time 0.023501 -2023-02-13 18:05:57,831 - Epoch: [108][ 170/ 1207] Overall Loss 0.257659 Objective Loss 0.257659 LR 0.000500 Time 0.023306 -2023-02-13 18:05:58,029 - Epoch: [108][ 180/ 1207] Overall Loss 0.256803 Objective Loss 0.256803 LR 0.000500 Time 0.023110 -2023-02-13 18:05:58,231 - Epoch: [108][ 190/ 1207] Overall Loss 0.257803 Objective Loss 0.257803 LR 0.000500 Time 0.022954 -2023-02-13 18:05:58,428 - Epoch: [108][ 200/ 1207] Overall Loss 0.259972 Objective Loss 0.259972 LR 0.000500 Time 0.022792 -2023-02-13 18:05:58,631 - Epoch: [108][ 210/ 1207] Overall Loss 0.261294 Objective Loss 0.261294 LR 0.000500 Time 0.022669 -2023-02-13 18:05:58,829 - Epoch: [108][ 220/ 1207] Overall Loss 0.261096 Objective Loss 0.261096 LR 0.000500 Time 0.022536 -2023-02-13 18:05:59,032 - Epoch: [108][ 230/ 1207] Overall Loss 0.261330 Objective Loss 0.261330 LR 0.000500 Time 0.022437 -2023-02-13 18:05:59,230 - Epoch: [108][ 240/ 1207] Overall Loss 0.262386 Objective Loss 0.262386 LR 0.000500 Time 0.022327 -2023-02-13 18:05:59,433 - Epoch: [108][ 250/ 1207] Overall Loss 0.262105 Objective Loss 0.262105 LR 0.000500 Time 0.022246 -2023-02-13 18:05:59,632 - Epoch: [108][ 260/ 1207] Overall Loss 0.262940 Objective Loss 0.262940 LR 0.000500 Time 0.022152 -2023-02-13 18:05:59,835 - Epoch: [108][ 270/ 1207] Overall Loss 0.261375 Objective Loss 0.261375 LR 0.000500 Time 0.022081 -2023-02-13 18:06:00,034 - Epoch: [108][ 280/ 1207] Overall Loss 0.260697 Objective Loss 0.260697 LR 0.000500 Time 0.022002 -2023-02-13 18:06:00,236 - Epoch: [108][ 290/ 1207] Overall Loss 0.261216 Objective Loss 0.261216 LR 0.000500 Time 0.021940 -2023-02-13 18:06:00,434 - Epoch: [108][ 300/ 1207] Overall Loss 0.261712 Objective Loss 0.261712 LR 0.000500 Time 0.021868 -2023-02-13 18:06:00,638 - Epoch: [108][ 310/ 1207] Overall Loss 0.260823 Objective Loss 0.260823 LR 0.000500 Time 0.021817 -2023-02-13 18:06:00,837 - Epoch: [108][ 320/ 1207] Overall Loss 0.261702 Objective Loss 0.261702 LR 0.000500 Time 0.021757 -2023-02-13 18:06:01,040 - Epoch: [108][ 330/ 1207] Overall Loss 0.261787 Objective Loss 0.261787 LR 0.000500 Time 0.021712 -2023-02-13 18:06:01,238 - Epoch: [108][ 340/ 1207] Overall Loss 0.261385 Objective Loss 0.261385 LR 0.000500 Time 0.021655 -2023-02-13 18:06:01,441 - Epoch: [108][ 350/ 1207] Overall Loss 0.261338 Objective Loss 0.261338 LR 0.000500 Time 0.021614 -2023-02-13 18:06:01,639 - Epoch: [108][ 360/ 1207] Overall Loss 0.260844 Objective Loss 0.260844 LR 0.000500 Time 0.021564 -2023-02-13 18:06:01,843 - Epoch: [108][ 370/ 1207] Overall Loss 0.260613 Objective Loss 0.260613 LR 0.000500 Time 0.021529 -2023-02-13 18:06:02,041 - Epoch: [108][ 380/ 1207] Overall Loss 0.260541 Objective Loss 0.260541 LR 0.000500 Time 0.021485 -2023-02-13 18:06:02,244 - Epoch: [108][ 390/ 1207] Overall Loss 0.260061 Objective Loss 0.260061 LR 0.000500 Time 0.021452 -2023-02-13 18:06:02,442 - Epoch: [108][ 400/ 1207] Overall Loss 0.259429 Objective Loss 0.259429 LR 0.000500 Time 0.021410 -2023-02-13 18:06:02,645 - Epoch: [108][ 410/ 1207] Overall Loss 0.259646 Objective Loss 0.259646 LR 0.000500 Time 0.021383 -2023-02-13 18:06:02,844 - Epoch: [108][ 420/ 1207] Overall Loss 0.259420 Objective Loss 0.259420 LR 0.000500 Time 0.021345 -2023-02-13 18:06:03,048 - Epoch: [108][ 430/ 1207] Overall Loss 0.259580 Objective Loss 0.259580 LR 0.000500 Time 0.021324 -2023-02-13 18:06:03,247 - Epoch: [108][ 440/ 1207] Overall Loss 0.259457 Objective Loss 0.259457 LR 0.000500 Time 0.021289 -2023-02-13 18:06:03,449 - Epoch: [108][ 450/ 1207] Overall Loss 0.259631 Objective Loss 0.259631 LR 0.000500 Time 0.021265 -2023-02-13 18:06:03,648 - Epoch: [108][ 460/ 1207] Overall Loss 0.259782 Objective Loss 0.259782 LR 0.000500 Time 0.021233 -2023-02-13 18:06:03,838 - Epoch: [108][ 470/ 1207] Overall Loss 0.260160 Objective Loss 0.260160 LR 0.000500 Time 0.021186 -2023-02-13 18:06:04,028 - Epoch: [108][ 480/ 1207] Overall Loss 0.260514 Objective Loss 0.260514 LR 0.000500 Time 0.021140 -2023-02-13 18:06:04,218 - Epoch: [108][ 490/ 1207] Overall Loss 0.260784 Objective Loss 0.260784 LR 0.000500 Time 0.021096 -2023-02-13 18:06:04,408 - Epoch: [108][ 500/ 1207] Overall Loss 0.261344 Objective Loss 0.261344 LR 0.000500 Time 0.021052 -2023-02-13 18:06:04,598 - Epoch: [108][ 510/ 1207] Overall Loss 0.261321 Objective Loss 0.261321 LR 0.000500 Time 0.021012 -2023-02-13 18:06:04,788 - Epoch: [108][ 520/ 1207] Overall Loss 0.261713 Objective Loss 0.261713 LR 0.000500 Time 0.020972 -2023-02-13 18:06:04,979 - Epoch: [108][ 530/ 1207] Overall Loss 0.261687 Objective Loss 0.261687 LR 0.000500 Time 0.020935 -2023-02-13 18:06:05,169 - Epoch: [108][ 540/ 1207] Overall Loss 0.261621 Objective Loss 0.261621 LR 0.000500 Time 0.020899 -2023-02-13 18:06:05,359 - Epoch: [108][ 550/ 1207] Overall Loss 0.261710 Objective Loss 0.261710 LR 0.000500 Time 0.020864 -2023-02-13 18:06:05,549 - Epoch: [108][ 560/ 1207] Overall Loss 0.261554 Objective Loss 0.261554 LR 0.000500 Time 0.020830 -2023-02-13 18:06:05,740 - Epoch: [108][ 570/ 1207] Overall Loss 0.261815 Objective Loss 0.261815 LR 0.000500 Time 0.020799 -2023-02-13 18:06:05,930 - Epoch: [108][ 580/ 1207] Overall Loss 0.262203 Objective Loss 0.262203 LR 0.000500 Time 0.020768 -2023-02-13 18:06:06,121 - Epoch: [108][ 590/ 1207] Overall Loss 0.261981 Objective Loss 0.261981 LR 0.000500 Time 0.020739 -2023-02-13 18:06:06,311 - Epoch: [108][ 600/ 1207] Overall Loss 0.261991 Objective Loss 0.261991 LR 0.000500 Time 0.020709 -2023-02-13 18:06:06,501 - Epoch: [108][ 610/ 1207] Overall Loss 0.262559 Objective Loss 0.262559 LR 0.000500 Time 0.020681 -2023-02-13 18:06:06,691 - Epoch: [108][ 620/ 1207] Overall Loss 0.262655 Objective Loss 0.262655 LR 0.000500 Time 0.020653 -2023-02-13 18:06:06,881 - Epoch: [108][ 630/ 1207] Overall Loss 0.262908 Objective Loss 0.262908 LR 0.000500 Time 0.020626 -2023-02-13 18:06:07,070 - Epoch: [108][ 640/ 1207] Overall Loss 0.262907 Objective Loss 0.262907 LR 0.000500 Time 0.020598 -2023-02-13 18:06:07,257 - Epoch: [108][ 650/ 1207] Overall Loss 0.262523 Objective Loss 0.262523 LR 0.000500 Time 0.020569 -2023-02-13 18:06:07,445 - Epoch: [108][ 660/ 1207] Overall Loss 0.262466 Objective Loss 0.262466 LR 0.000500 Time 0.020541 -2023-02-13 18:06:07,633 - Epoch: [108][ 670/ 1207] Overall Loss 0.262767 Objective Loss 0.262767 LR 0.000500 Time 0.020515 -2023-02-13 18:06:07,821 - Epoch: [108][ 680/ 1207] Overall Loss 0.262044 Objective Loss 0.262044 LR 0.000500 Time 0.020489 -2023-02-13 18:06:08,010 - Epoch: [108][ 690/ 1207] Overall Loss 0.262223 Objective Loss 0.262223 LR 0.000500 Time 0.020465 -2023-02-13 18:06:08,198 - Epoch: [108][ 700/ 1207] Overall Loss 0.262201 Objective Loss 0.262201 LR 0.000500 Time 0.020441 -2023-02-13 18:06:08,386 - Epoch: [108][ 710/ 1207] Overall Loss 0.262136 Objective Loss 0.262136 LR 0.000500 Time 0.020417 -2023-02-13 18:06:08,574 - Epoch: [108][ 720/ 1207] Overall Loss 0.261867 Objective Loss 0.261867 LR 0.000500 Time 0.020394 -2023-02-13 18:06:08,762 - Epoch: [108][ 730/ 1207] Overall Loss 0.261587 Objective Loss 0.261587 LR 0.000500 Time 0.020372 -2023-02-13 18:06:08,951 - Epoch: [108][ 740/ 1207] Overall Loss 0.261466 Objective Loss 0.261466 LR 0.000500 Time 0.020351 -2023-02-13 18:06:09,140 - Epoch: [108][ 750/ 1207] Overall Loss 0.261737 Objective Loss 0.261737 LR 0.000500 Time 0.020331 -2023-02-13 18:06:09,327 - Epoch: [108][ 760/ 1207] Overall Loss 0.261957 Objective Loss 0.261957 LR 0.000500 Time 0.020310 -2023-02-13 18:06:09,515 - Epoch: [108][ 770/ 1207] Overall Loss 0.261944 Objective Loss 0.261944 LR 0.000500 Time 0.020290 -2023-02-13 18:06:09,703 - Epoch: [108][ 780/ 1207] Overall Loss 0.262444 Objective Loss 0.262444 LR 0.000500 Time 0.020271 -2023-02-13 18:06:09,891 - Epoch: [108][ 790/ 1207] Overall Loss 0.262907 Objective Loss 0.262907 LR 0.000500 Time 0.020251 -2023-02-13 18:06:10,079 - Epoch: [108][ 800/ 1207] Overall Loss 0.262636 Objective Loss 0.262636 LR 0.000500 Time 0.020233 -2023-02-13 18:06:10,267 - Epoch: [108][ 810/ 1207] Overall Loss 0.262521 Objective Loss 0.262521 LR 0.000500 Time 0.020214 -2023-02-13 18:06:10,454 - Epoch: [108][ 820/ 1207] Overall Loss 0.262568 Objective Loss 0.262568 LR 0.000500 Time 0.020196 -2023-02-13 18:06:10,642 - Epoch: [108][ 830/ 1207] Overall Loss 0.262636 Objective Loss 0.262636 LR 0.000500 Time 0.020178 -2023-02-13 18:06:10,831 - Epoch: [108][ 840/ 1207] Overall Loss 0.262745 Objective Loss 0.262745 LR 0.000500 Time 0.020162 -2023-02-13 18:06:11,020 - Epoch: [108][ 850/ 1207] Overall Loss 0.262704 Objective Loss 0.262704 LR 0.000500 Time 0.020147 -2023-02-13 18:06:11,208 - Epoch: [108][ 860/ 1207] Overall Loss 0.262673 Objective Loss 0.262673 LR 0.000500 Time 0.020131 -2023-02-13 18:06:11,396 - Epoch: [108][ 870/ 1207] Overall Loss 0.262624 Objective Loss 0.262624 LR 0.000500 Time 0.020115 -2023-02-13 18:06:11,585 - Epoch: [108][ 880/ 1207] Overall Loss 0.262596 Objective Loss 0.262596 LR 0.000500 Time 0.020101 -2023-02-13 18:06:11,773 - Epoch: [108][ 890/ 1207] Overall Loss 0.262758 Objective Loss 0.262758 LR 0.000500 Time 0.020086 -2023-02-13 18:06:11,961 - Epoch: [108][ 900/ 1207] Overall Loss 0.262588 Objective Loss 0.262588 LR 0.000500 Time 0.020072 -2023-02-13 18:06:12,150 - Epoch: [108][ 910/ 1207] Overall Loss 0.262262 Objective Loss 0.262262 LR 0.000500 Time 0.020058 -2023-02-13 18:06:12,338 - Epoch: [108][ 920/ 1207] Overall Loss 0.262335 Objective Loss 0.262335 LR 0.000500 Time 0.020044 -2023-02-13 18:06:12,526 - Epoch: [108][ 930/ 1207] Overall Loss 0.262553 Objective Loss 0.262553 LR 0.000500 Time 0.020030 -2023-02-13 18:06:12,714 - Epoch: [108][ 940/ 1207] Overall Loss 0.262691 Objective Loss 0.262691 LR 0.000500 Time 0.020017 -2023-02-13 18:06:12,902 - Epoch: [108][ 950/ 1207] Overall Loss 0.262766 Objective Loss 0.262766 LR 0.000500 Time 0.020004 -2023-02-13 18:06:13,091 - Epoch: [108][ 960/ 1207] Overall Loss 0.263006 Objective Loss 0.263006 LR 0.000500 Time 0.019992 -2023-02-13 18:06:13,279 - Epoch: [108][ 970/ 1207] Overall Loss 0.263050 Objective Loss 0.263050 LR 0.000500 Time 0.019979 -2023-02-13 18:06:13,467 - Epoch: [108][ 980/ 1207] Overall Loss 0.263296 Objective Loss 0.263296 LR 0.000500 Time 0.019967 -2023-02-13 18:06:13,655 - Epoch: [108][ 990/ 1207] Overall Loss 0.263428 Objective Loss 0.263428 LR 0.000500 Time 0.019955 -2023-02-13 18:06:13,843 - Epoch: [108][ 1000/ 1207] Overall Loss 0.263457 Objective Loss 0.263457 LR 0.000500 Time 0.019943 -2023-02-13 18:06:14,032 - Epoch: [108][ 1010/ 1207] Overall Loss 0.263637 Objective Loss 0.263637 LR 0.000500 Time 0.019932 -2023-02-13 18:06:14,220 - Epoch: [108][ 1020/ 1207] Overall Loss 0.263666 Objective Loss 0.263666 LR 0.000500 Time 0.019920 -2023-02-13 18:06:14,408 - Epoch: [108][ 1030/ 1207] Overall Loss 0.263609 Objective Loss 0.263609 LR 0.000500 Time 0.019909 -2023-02-13 18:06:14,596 - Epoch: [108][ 1040/ 1207] Overall Loss 0.263545 Objective Loss 0.263545 LR 0.000500 Time 0.019898 -2023-02-13 18:06:14,784 - Epoch: [108][ 1050/ 1207] Overall Loss 0.263333 Objective Loss 0.263333 LR 0.000500 Time 0.019888 -2023-02-13 18:06:14,973 - Epoch: [108][ 1060/ 1207] Overall Loss 0.263177 Objective Loss 0.263177 LR 0.000500 Time 0.019877 -2023-02-13 18:06:15,162 - Epoch: [108][ 1070/ 1207] Overall Loss 0.263203 Objective Loss 0.263203 LR 0.000500 Time 0.019868 -2023-02-13 18:06:15,350 - Epoch: [108][ 1080/ 1207] Overall Loss 0.263247 Objective Loss 0.263247 LR 0.000500 Time 0.019858 -2023-02-13 18:06:15,538 - Epoch: [108][ 1090/ 1207] Overall Loss 0.263322 Objective Loss 0.263322 LR 0.000500 Time 0.019848 -2023-02-13 18:06:15,726 - Epoch: [108][ 1100/ 1207] Overall Loss 0.263345 Objective Loss 0.263345 LR 0.000500 Time 0.019838 -2023-02-13 18:06:15,915 - Epoch: [108][ 1110/ 1207] Overall Loss 0.263411 Objective Loss 0.263411 LR 0.000500 Time 0.019830 -2023-02-13 18:06:16,104 - Epoch: [108][ 1120/ 1207] Overall Loss 0.263432 Objective Loss 0.263432 LR 0.000500 Time 0.019821 -2023-02-13 18:06:16,292 - Epoch: [108][ 1130/ 1207] Overall Loss 0.263478 Objective Loss 0.263478 LR 0.000500 Time 0.019812 -2023-02-13 18:06:16,481 - Epoch: [108][ 1140/ 1207] Overall Loss 0.263607 Objective Loss 0.263607 LR 0.000500 Time 0.019803 -2023-02-13 18:06:16,670 - Epoch: [108][ 1150/ 1207] Overall Loss 0.263539 Objective Loss 0.263539 LR 0.000500 Time 0.019795 -2023-02-13 18:06:16,858 - Epoch: [108][ 1160/ 1207] Overall Loss 0.263670 Objective Loss 0.263670 LR 0.000500 Time 0.019786 -2023-02-13 18:06:17,047 - Epoch: [108][ 1170/ 1207] Overall Loss 0.263721 Objective Loss 0.263721 LR 0.000500 Time 0.019778 -2023-02-13 18:06:17,236 - Epoch: [108][ 1180/ 1207] Overall Loss 0.263509 Objective Loss 0.263509 LR 0.000500 Time 0.019770 -2023-02-13 18:06:17,424 - Epoch: [108][ 1190/ 1207] Overall Loss 0.263134 Objective Loss 0.263134 LR 0.000500 Time 0.019762 -2023-02-13 18:06:17,663 - Epoch: [108][ 1200/ 1207] Overall Loss 0.263151 Objective Loss 0.263151 LR 0.000500 Time 0.019796 -2023-02-13 18:06:17,779 - Epoch: [108][ 1207/ 1207] Overall Loss 0.263123 Objective Loss 0.263123 Top1 86.280488 Top5 97.865854 LR 0.000500 Time 0.019777 -2023-02-13 18:06:17,861 - --- validate (epoch=108)----------- -2023-02-13 18:06:17,862 - 34311 samples (256 per mini-batch) -2023-02-13 18:06:18,270 - Epoch: [108][ 10/ 135] Loss 0.328987 Top1 83.906250 Top5 97.656250 -2023-02-13 18:06:18,398 - Epoch: [108][ 20/ 135] Loss 0.326491 Top1 83.300781 Top5 97.480469 -2023-02-13 18:06:18,528 - Epoch: [108][ 30/ 135] Loss 0.330698 Top1 83.411458 Top5 97.473958 -2023-02-13 18:06:18,654 - Epoch: [108][ 40/ 135] Loss 0.325846 Top1 83.349609 Top5 97.519531 -2023-02-13 18:06:18,782 - Epoch: [108][ 50/ 135] Loss 0.328546 Top1 83.531250 Top5 97.476562 -2023-02-13 18:06:18,910 - Epoch: [108][ 60/ 135] Loss 0.329599 Top1 83.444010 Top5 97.428385 -2023-02-13 18:06:19,036 - Epoch: [108][ 70/ 135] Loss 0.330235 Top1 83.364955 Top5 97.332589 -2023-02-13 18:06:19,167 - Epoch: [108][ 80/ 135] Loss 0.327427 Top1 83.408203 Top5 97.377930 -2023-02-13 18:06:19,298 - Epoch: [108][ 90/ 135] Loss 0.330206 Top1 83.398438 Top5 97.404514 -2023-02-13 18:06:19,427 - Epoch: [108][ 100/ 135] Loss 0.329215 Top1 83.355469 Top5 97.394531 -2023-02-13 18:06:19,557 - Epoch: [108][ 110/ 135] Loss 0.331344 Top1 83.277699 Top5 97.425426 -2023-02-13 18:06:19,688 - Epoch: [108][ 120/ 135] Loss 0.330497 Top1 83.411458 Top5 97.457682 -2023-02-13 18:06:19,821 - Epoch: [108][ 130/ 135] Loss 0.331477 Top1 83.353365 Top5 97.463942 -2023-02-13 18:06:19,868 - Epoch: [108][ 135/ 135] Loss 0.329508 Top1 83.404739 Top5 97.476028 -2023-02-13 18:06:19,936 - ==> Top1: 83.405 Top5: 97.476 Loss: 0.330 - -2023-02-13 18:06:19,937 - ==> Confusion: -[[ 859 6 10 2 7 2 1 1 3 44 1 10 2 3 5 2 2 0 0 0 7] - [ 3 931 1 3 16 25 1 27 5 1 1 2 2 0 0 3 2 0 3 2 5] - [ 8 5 948 12 5 1 19 20 0 1 2 2 3 3 4 7 1 4 9 1 3] - [ 5 1 30 878 3 4 0 4 1 3 11 0 8 2 23 2 3 9 22 2 5] - [ 22 3 0 1 988 5 2 1 0 8 0 7 1 4 5 7 4 0 0 5 3] - [ 3 15 1 6 8 955 2 25 0 3 2 11 4 13 0 2 6 1 0 8 5] - [ 4 5 15 2 1 3 1044 3 0 0 3 0 2 2 1 2 1 2 0 7 2] - [ 1 2 7 1 4 16 0 952 0 4 1 7 3 1 0 1 0 0 7 12 5] - [ 24 4 1 2 1 1 0 2 880 49 8 3 0 10 14 2 0 1 4 0 3] - [ 93 0 4 1 5 0 0 1 26 852 0 1 0 13 4 3 0 4 1 0 4] - [ 3 2 4 9 2 0 3 5 20 2 964 2 2 7 5 2 1 2 9 0 7] - [ 2 3 2 0 4 9 2 10 1 2 0 908 20 10 1 5 2 9 2 10 3] - [ 0 0 1 10 0 4 0 3 3 0 0 38 842 0 2 8 3 27 4 2 12] - [ 8 3 4 1 8 5 0 3 13 23 11 3 2 916 4 7 0 2 1 4 6] - [ 9 4 5 18 8 4 0 1 15 7 2 3 2 1 986 1 1 7 11 0 7] - [ 4 3 8 0 7 2 5 1 0 0 0 12 5 1 1 963 3 17 0 9 5] - [ 5 4 2 1 9 1 0 1 1 0 1 3 1 3 2 14 988 1 1 9 14] - [ 6 3 0 3 2 0 3 2 0 1 1 13 12 0 1 14 0 984 0 3 3] - [ 6 4 7 10 1 0 0 38 3 0 3 1 2 0 17 1 0 2 986 3 2] - [ 0 2 2 0 2 4 11 11 0 0 1 18 4 3 0 3 2 1 1 1077 6] - [ 213 251 306 114 159 176 106 234 88 99 152 148 253 315 163 120 215 129 188 289 9716]] - -2023-02-13 18:06:19,938 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:06:19,938 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:06:19,944 - - -2023-02-13 18:06:19,944 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:06:20,827 - Epoch: [109][ 10/ 1207] Overall Loss 0.297394 Objective Loss 0.297394 LR 0.000500 Time 0.088282 -2023-02-13 18:06:21,038 - Epoch: [109][ 20/ 1207] Overall Loss 0.274149 Objective Loss 0.274149 LR 0.000500 Time 0.054627 -2023-02-13 18:06:21,240 - Epoch: [109][ 30/ 1207] Overall Loss 0.266911 Objective Loss 0.266911 LR 0.000500 Time 0.043144 -2023-02-13 18:06:21,443 - Epoch: [109][ 40/ 1207] Overall Loss 0.263366 Objective Loss 0.263366 LR 0.000500 Time 0.037437 -2023-02-13 18:06:21,646 - Epoch: [109][ 50/ 1207] Overall Loss 0.258050 Objective Loss 0.258050 LR 0.000500 Time 0.033999 -2023-02-13 18:06:21,850 - Epoch: [109][ 60/ 1207] Overall Loss 0.254196 Objective Loss 0.254196 LR 0.000500 Time 0.031719 -2023-02-13 18:06:22,053 - Epoch: [109][ 70/ 1207] Overall Loss 0.255428 Objective Loss 0.255428 LR 0.000500 Time 0.030087 -2023-02-13 18:06:22,257 - Epoch: [109][ 80/ 1207] Overall Loss 0.253133 Objective Loss 0.253133 LR 0.000500 Time 0.028875 -2023-02-13 18:06:22,459 - Epoch: [109][ 90/ 1207] Overall Loss 0.253186 Objective Loss 0.253186 LR 0.000500 Time 0.027908 -2023-02-13 18:06:22,663 - Epoch: [109][ 100/ 1207] Overall Loss 0.253432 Objective Loss 0.253432 LR 0.000500 Time 0.027152 -2023-02-13 18:06:22,865 - Epoch: [109][ 110/ 1207] Overall Loss 0.252174 Objective Loss 0.252174 LR 0.000500 Time 0.026519 -2023-02-13 18:06:23,069 - Epoch: [109][ 120/ 1207] Overall Loss 0.252266 Objective Loss 0.252266 LR 0.000500 Time 0.026000 -2023-02-13 18:06:23,263 - Epoch: [109][ 130/ 1207] Overall Loss 0.251979 Objective Loss 0.251979 LR 0.000500 Time 0.025488 -2023-02-13 18:06:23,453 - Epoch: [109][ 140/ 1207] Overall Loss 0.251792 Objective Loss 0.251792 LR 0.000500 Time 0.025026 -2023-02-13 18:06:23,647 - Epoch: [109][ 150/ 1207] Overall Loss 0.252558 Objective Loss 0.252558 LR 0.000500 Time 0.024650 -2023-02-13 18:06:23,838 - Epoch: [109][ 160/ 1207] Overall Loss 0.253139 Objective Loss 0.253139 LR 0.000500 Time 0.024297 -2023-02-13 18:06:24,032 - Epoch: [109][ 170/ 1207] Overall Loss 0.252173 Objective Loss 0.252173 LR 0.000500 Time 0.024008 -2023-02-13 18:06:24,222 - Epoch: [109][ 180/ 1207] Overall Loss 0.251681 Objective Loss 0.251681 LR 0.000500 Time 0.023729 -2023-02-13 18:06:24,415 - Epoch: [109][ 190/ 1207] Overall Loss 0.250734 Objective Loss 0.250734 LR 0.000500 Time 0.023495 -2023-02-13 18:06:24,607 - Epoch: [109][ 200/ 1207] Overall Loss 0.250284 Objective Loss 0.250284 LR 0.000500 Time 0.023275 -2023-02-13 18:06:24,801 - Epoch: [109][ 210/ 1207] Overall Loss 0.251103 Objective Loss 0.251103 LR 0.000500 Time 0.023088 -2023-02-13 18:06:24,991 - Epoch: [109][ 220/ 1207] Overall Loss 0.251100 Objective Loss 0.251100 LR 0.000500 Time 0.022902 -2023-02-13 18:06:25,185 - Epoch: [109][ 230/ 1207] Overall Loss 0.251012 Objective Loss 0.251012 LR 0.000500 Time 0.022751 -2023-02-13 18:06:25,376 - Epoch: [109][ 240/ 1207] Overall Loss 0.251746 Objective Loss 0.251746 LR 0.000500 Time 0.022595 -2023-02-13 18:06:25,570 - Epoch: [109][ 250/ 1207] Overall Loss 0.251793 Objective Loss 0.251793 LR 0.000500 Time 0.022464 -2023-02-13 18:06:25,760 - Epoch: [109][ 260/ 1207] Overall Loss 0.251965 Objective Loss 0.251965 LR 0.000500 Time 0.022331 -2023-02-13 18:06:25,954 - Epoch: [109][ 270/ 1207] Overall Loss 0.250715 Objective Loss 0.250715 LR 0.000500 Time 0.022221 -2023-02-13 18:06:26,145 - Epoch: [109][ 280/ 1207] Overall Loss 0.250184 Objective Loss 0.250184 LR 0.000500 Time 0.022110 -2023-02-13 18:06:26,338 - Epoch: [109][ 290/ 1207] Overall Loss 0.251662 Objective Loss 0.251662 LR 0.000500 Time 0.022011 -2023-02-13 18:06:26,529 - Epoch: [109][ 300/ 1207] Overall Loss 0.253625 Objective Loss 0.253625 LR 0.000500 Time 0.021912 -2023-02-13 18:06:26,723 - Epoch: [109][ 310/ 1207] Overall Loss 0.254590 Objective Loss 0.254590 LR 0.000500 Time 0.021829 -2023-02-13 18:06:26,914 - Epoch: [109][ 320/ 1207] Overall Loss 0.255080 Objective Loss 0.255080 LR 0.000500 Time 0.021744 -2023-02-13 18:06:27,108 - Epoch: [109][ 330/ 1207] Overall Loss 0.254603 Objective Loss 0.254603 LR 0.000500 Time 0.021673 -2023-02-13 18:06:27,298 - Epoch: [109][ 340/ 1207] Overall Loss 0.254653 Objective Loss 0.254653 LR 0.000500 Time 0.021593 -2023-02-13 18:06:27,492 - Epoch: [109][ 350/ 1207] Overall Loss 0.255668 Objective Loss 0.255668 LR 0.000500 Time 0.021528 -2023-02-13 18:06:27,683 - Epoch: [109][ 360/ 1207] Overall Loss 0.255450 Objective Loss 0.255450 LR 0.000500 Time 0.021459 -2023-02-13 18:06:27,877 - Epoch: [109][ 370/ 1207] Overall Loss 0.255222 Objective Loss 0.255222 LR 0.000500 Time 0.021403 -2023-02-13 18:06:28,068 - Epoch: [109][ 380/ 1207] Overall Loss 0.255167 Objective Loss 0.255167 LR 0.000500 Time 0.021341 -2023-02-13 18:06:28,262 - Epoch: [109][ 390/ 1207] Overall Loss 0.255700 Objective Loss 0.255700 LR 0.000500 Time 0.021291 -2023-02-13 18:06:28,453 - Epoch: [109][ 400/ 1207] Overall Loss 0.256606 Objective Loss 0.256606 LR 0.000500 Time 0.021234 -2023-02-13 18:06:28,646 - Epoch: [109][ 410/ 1207] Overall Loss 0.257272 Objective Loss 0.257272 LR 0.000500 Time 0.021188 -2023-02-13 18:06:28,838 - Epoch: [109][ 420/ 1207] Overall Loss 0.257632 Objective Loss 0.257632 LR 0.000500 Time 0.021139 -2023-02-13 18:06:29,032 - Epoch: [109][ 430/ 1207] Overall Loss 0.257624 Objective Loss 0.257624 LR 0.000500 Time 0.021097 -2023-02-13 18:06:29,223 - Epoch: [109][ 440/ 1207] Overall Loss 0.257165 Objective Loss 0.257165 LR 0.000500 Time 0.021051 -2023-02-13 18:06:29,416 - Epoch: [109][ 450/ 1207] Overall Loss 0.256721 Objective Loss 0.256721 LR 0.000500 Time 0.021012 -2023-02-13 18:06:29,607 - Epoch: [109][ 460/ 1207] Overall Loss 0.256805 Objective Loss 0.256805 LR 0.000500 Time 0.020970 -2023-02-13 18:06:29,801 - Epoch: [109][ 470/ 1207] Overall Loss 0.256511 Objective Loss 0.256511 LR 0.000500 Time 0.020935 -2023-02-13 18:06:29,992 - Epoch: [109][ 480/ 1207] Overall Loss 0.257188 Objective Loss 0.257188 LR 0.000500 Time 0.020896 -2023-02-13 18:06:30,189 - Epoch: [109][ 490/ 1207] Overall Loss 0.257041 Objective Loss 0.257041 LR 0.000500 Time 0.020871 -2023-02-13 18:06:30,383 - Epoch: [109][ 500/ 1207] Overall Loss 0.256925 Objective Loss 0.256925 LR 0.000500 Time 0.020842 -2023-02-13 18:06:30,581 - Epoch: [109][ 510/ 1207] Overall Loss 0.257187 Objective Loss 0.257187 LR 0.000500 Time 0.020819 -2023-02-13 18:06:30,775 - Epoch: [109][ 520/ 1207] Overall Loss 0.257252 Objective Loss 0.257252 LR 0.000500 Time 0.020791 -2023-02-13 18:06:30,973 - Epoch: [109][ 530/ 1207] Overall Loss 0.257236 Objective Loss 0.257236 LR 0.000500 Time 0.020772 -2023-02-13 18:06:31,165 - Epoch: [109][ 540/ 1207] Overall Loss 0.257388 Objective Loss 0.257388 LR 0.000500 Time 0.020743 -2023-02-13 18:06:31,360 - Epoch: [109][ 550/ 1207] Overall Loss 0.257603 Objective Loss 0.257603 LR 0.000500 Time 0.020718 -2023-02-13 18:06:31,551 - Epoch: [109][ 560/ 1207] Overall Loss 0.257688 Objective Loss 0.257688 LR 0.000500 Time 0.020690 -2023-02-13 18:06:31,746 - Epoch: [109][ 570/ 1207] Overall Loss 0.257718 Objective Loss 0.257718 LR 0.000500 Time 0.020668 -2023-02-13 18:06:31,938 - Epoch: [109][ 580/ 1207] Overall Loss 0.258290 Objective Loss 0.258290 LR 0.000500 Time 0.020641 -2023-02-13 18:06:32,133 - Epoch: [109][ 590/ 1207] Overall Loss 0.258631 Objective Loss 0.258631 LR 0.000500 Time 0.020622 -2023-02-13 18:06:32,325 - Epoch: [109][ 600/ 1207] Overall Loss 0.259324 Objective Loss 0.259324 LR 0.000500 Time 0.020598 -2023-02-13 18:06:32,519 - Epoch: [109][ 610/ 1207] Overall Loss 0.259768 Objective Loss 0.259768 LR 0.000500 Time 0.020578 -2023-02-13 18:06:32,712 - Epoch: [109][ 620/ 1207] Overall Loss 0.260105 Objective Loss 0.260105 LR 0.000500 Time 0.020556 -2023-02-13 18:06:32,906 - Epoch: [109][ 630/ 1207] Overall Loss 0.260009 Objective Loss 0.260009 LR 0.000500 Time 0.020537 -2023-02-13 18:06:33,098 - Epoch: [109][ 640/ 1207] Overall Loss 0.260474 Objective Loss 0.260474 LR 0.000500 Time 0.020517 -2023-02-13 18:06:33,293 - Epoch: [109][ 650/ 1207] Overall Loss 0.260677 Objective Loss 0.260677 LR 0.000500 Time 0.020500 -2023-02-13 18:06:33,485 - Epoch: [109][ 660/ 1207] Overall Loss 0.260552 Objective Loss 0.260552 LR 0.000500 Time 0.020479 -2023-02-13 18:06:33,678 - Epoch: [109][ 670/ 1207] Overall Loss 0.260387 Objective Loss 0.260387 LR 0.000500 Time 0.020462 -2023-02-13 18:06:33,873 - Epoch: [109][ 680/ 1207] Overall Loss 0.260467 Objective Loss 0.260467 LR 0.000500 Time 0.020447 -2023-02-13 18:06:34,071 - Epoch: [109][ 690/ 1207] Overall Loss 0.260698 Objective Loss 0.260698 LR 0.000500 Time 0.020437 -2023-02-13 18:06:34,266 - Epoch: [109][ 700/ 1207] Overall Loss 0.260600 Objective Loss 0.260600 LR 0.000500 Time 0.020423 -2023-02-13 18:06:34,463 - Epoch: [109][ 710/ 1207] Overall Loss 0.260873 Objective Loss 0.260873 LR 0.000500 Time 0.020412 -2023-02-13 18:06:34,657 - Epoch: [109][ 720/ 1207] Overall Loss 0.260935 Objective Loss 0.260935 LR 0.000500 Time 0.020398 -2023-02-13 18:06:34,855 - Epoch: [109][ 730/ 1207] Overall Loss 0.260465 Objective Loss 0.260465 LR 0.000500 Time 0.020388 -2023-02-13 18:06:35,046 - Epoch: [109][ 740/ 1207] Overall Loss 0.260689 Objective Loss 0.260689 LR 0.000500 Time 0.020371 -2023-02-13 18:06:35,242 - Epoch: [109][ 750/ 1207] Overall Loss 0.260775 Objective Loss 0.260775 LR 0.000500 Time 0.020360 -2023-02-13 18:06:35,434 - Epoch: [109][ 760/ 1207] Overall Loss 0.260526 Objective Loss 0.260526 LR 0.000500 Time 0.020344 -2023-02-13 18:06:35,629 - Epoch: [109][ 770/ 1207] Overall Loss 0.260663 Objective Loss 0.260663 LR 0.000500 Time 0.020333 -2023-02-13 18:06:35,821 - Epoch: [109][ 780/ 1207] Overall Loss 0.261343 Objective Loss 0.261343 LR 0.000500 Time 0.020318 -2023-02-13 18:06:36,017 - Epoch: [109][ 790/ 1207] Overall Loss 0.260987 Objective Loss 0.260987 LR 0.000500 Time 0.020308 -2023-02-13 18:06:36,209 - Epoch: [109][ 800/ 1207] Overall Loss 0.260999 Objective Loss 0.260999 LR 0.000500 Time 0.020294 -2023-02-13 18:06:36,404 - Epoch: [109][ 810/ 1207] Overall Loss 0.260902 Objective Loss 0.260902 LR 0.000500 Time 0.020284 -2023-02-13 18:06:36,595 - Epoch: [109][ 820/ 1207] Overall Loss 0.261130 Objective Loss 0.261130 LR 0.000500 Time 0.020269 -2023-02-13 18:06:36,790 - Epoch: [109][ 830/ 1207] Overall Loss 0.261800 Objective Loss 0.261800 LR 0.000500 Time 0.020259 -2023-02-13 18:06:36,982 - Epoch: [109][ 840/ 1207] Overall Loss 0.262002 Objective Loss 0.262002 LR 0.000500 Time 0.020246 -2023-02-13 18:06:37,177 - Epoch: [109][ 850/ 1207] Overall Loss 0.261849 Objective Loss 0.261849 LR 0.000500 Time 0.020237 -2023-02-13 18:06:37,368 - Epoch: [109][ 860/ 1207] Overall Loss 0.262149 Objective Loss 0.262149 LR 0.000500 Time 0.020224 -2023-02-13 18:06:37,563 - Epoch: [109][ 870/ 1207] Overall Loss 0.262207 Objective Loss 0.262207 LR 0.000500 Time 0.020215 -2023-02-13 18:06:37,756 - Epoch: [109][ 880/ 1207] Overall Loss 0.262260 Objective Loss 0.262260 LR 0.000500 Time 0.020204 -2023-02-13 18:06:37,950 - Epoch: [109][ 890/ 1207] Overall Loss 0.262145 Objective Loss 0.262145 LR 0.000500 Time 0.020195 -2023-02-13 18:06:38,143 - Epoch: [109][ 900/ 1207] Overall Loss 0.261859 Objective Loss 0.261859 LR 0.000500 Time 0.020183 -2023-02-13 18:06:38,337 - Epoch: [109][ 910/ 1207] Overall Loss 0.261484 Objective Loss 0.261484 LR 0.000500 Time 0.020175 -2023-02-13 18:06:38,529 - Epoch: [109][ 920/ 1207] Overall Loss 0.261690 Objective Loss 0.261690 LR 0.000500 Time 0.020164 -2023-02-13 18:06:38,723 - Epoch: [109][ 930/ 1207] Overall Loss 0.261689 Objective Loss 0.261689 LR 0.000500 Time 0.020156 -2023-02-13 18:06:38,916 - Epoch: [109][ 940/ 1207] Overall Loss 0.261726 Objective Loss 0.261726 LR 0.000500 Time 0.020145 -2023-02-13 18:06:39,110 - Epoch: [109][ 950/ 1207] Overall Loss 0.261379 Objective Loss 0.261379 LR 0.000500 Time 0.020138 -2023-02-13 18:06:39,302 - Epoch: [109][ 960/ 1207] Overall Loss 0.261408 Objective Loss 0.261408 LR 0.000500 Time 0.020128 -2023-02-13 18:06:39,497 - Epoch: [109][ 970/ 1207] Overall Loss 0.261536 Objective Loss 0.261536 LR 0.000500 Time 0.020121 -2023-02-13 18:06:39,689 - Epoch: [109][ 980/ 1207] Overall Loss 0.261338 Objective Loss 0.261338 LR 0.000500 Time 0.020111 -2023-02-13 18:06:39,884 - Epoch: [109][ 990/ 1207] Overall Loss 0.261354 Objective Loss 0.261354 LR 0.000500 Time 0.020104 -2023-02-13 18:06:40,076 - Epoch: [109][ 1000/ 1207] Overall Loss 0.261425 Objective Loss 0.261425 LR 0.000500 Time 0.020094 -2023-02-13 18:06:40,270 - Epoch: [109][ 1010/ 1207] Overall Loss 0.261270 Objective Loss 0.261270 LR 0.000500 Time 0.020088 -2023-02-13 18:06:40,462 - Epoch: [109][ 1020/ 1207] Overall Loss 0.260893 Objective Loss 0.260893 LR 0.000500 Time 0.020079 -2023-02-13 18:06:40,657 - Epoch: [109][ 1030/ 1207] Overall Loss 0.260834 Objective Loss 0.260834 LR 0.000500 Time 0.020073 -2023-02-13 18:06:40,850 - Epoch: [109][ 1040/ 1207] Overall Loss 0.260947 Objective Loss 0.260947 LR 0.000500 Time 0.020064 -2023-02-13 18:06:41,045 - Epoch: [109][ 1050/ 1207] Overall Loss 0.260795 Objective Loss 0.260795 LR 0.000500 Time 0.020059 -2023-02-13 18:06:41,236 - Epoch: [109][ 1060/ 1207] Overall Loss 0.260878 Objective Loss 0.260878 LR 0.000500 Time 0.020050 -2023-02-13 18:06:41,431 - Epoch: [109][ 1070/ 1207] Overall Loss 0.260725 Objective Loss 0.260725 LR 0.000500 Time 0.020044 -2023-02-13 18:06:41,624 - Epoch: [109][ 1080/ 1207] Overall Loss 0.260495 Objective Loss 0.260495 LR 0.000500 Time 0.020037 -2023-02-13 18:06:41,818 - Epoch: [109][ 1090/ 1207] Overall Loss 0.260479 Objective Loss 0.260479 LR 0.000500 Time 0.020031 -2023-02-13 18:06:42,011 - Epoch: [109][ 1100/ 1207] Overall Loss 0.260703 Objective Loss 0.260703 LR 0.000500 Time 0.020024 -2023-02-13 18:06:42,205 - Epoch: [109][ 1110/ 1207] Overall Loss 0.260595 Objective Loss 0.260595 LR 0.000500 Time 0.020018 -2023-02-13 18:06:42,397 - Epoch: [109][ 1120/ 1207] Overall Loss 0.260452 Objective Loss 0.260452 LR 0.000500 Time 0.020010 -2023-02-13 18:06:42,592 - Epoch: [109][ 1130/ 1207] Overall Loss 0.260461 Objective Loss 0.260461 LR 0.000500 Time 0.020005 -2023-02-13 18:06:42,784 - Epoch: [109][ 1140/ 1207] Overall Loss 0.260694 Objective Loss 0.260694 LR 0.000500 Time 0.019998 -2023-02-13 18:06:42,978 - Epoch: [109][ 1150/ 1207] Overall Loss 0.260625 Objective Loss 0.260625 LR 0.000500 Time 0.019992 -2023-02-13 18:06:43,170 - Epoch: [109][ 1160/ 1207] Overall Loss 0.260760 Objective Loss 0.260760 LR 0.000500 Time 0.019986 -2023-02-13 18:06:43,365 - Epoch: [109][ 1170/ 1207] Overall Loss 0.260671 Objective Loss 0.260671 LR 0.000500 Time 0.019981 -2023-02-13 18:06:43,556 - Epoch: [109][ 1180/ 1207] Overall Loss 0.260854 Objective Loss 0.260854 LR 0.000500 Time 0.019973 -2023-02-13 18:06:43,750 - Epoch: [109][ 1190/ 1207] Overall Loss 0.260585 Objective Loss 0.260585 LR 0.000500 Time 0.019968 -2023-02-13 18:06:43,999 - Epoch: [109][ 1200/ 1207] Overall Loss 0.260481 Objective Loss 0.260481 LR 0.000500 Time 0.020009 -2023-02-13 18:06:44,114 - Epoch: [109][ 1207/ 1207] Overall Loss 0.260378 Objective Loss 0.260378 Top1 86.280488 Top5 98.780488 LR 0.000500 Time 0.019988 -2023-02-13 18:06:44,185 - --- validate (epoch=109)----------- -2023-02-13 18:06:44,185 - 34311 samples (256 per mini-batch) -2023-02-13 18:06:44,688 - Epoch: [109][ 10/ 135] Loss 0.283700 Top1 84.179688 Top5 97.851562 -2023-02-13 18:06:44,820 - Epoch: [109][ 20/ 135] Loss 0.309518 Top1 83.867188 Top5 97.890625 -2023-02-13 18:06:44,951 - Epoch: [109][ 30/ 135] Loss 0.310870 Top1 83.789062 Top5 97.838542 -2023-02-13 18:06:45,083 - Epoch: [109][ 40/ 135] Loss 0.299532 Top1 84.169922 Top5 97.871094 -2023-02-13 18:06:45,213 - Epoch: [109][ 50/ 135] Loss 0.308310 Top1 84.171875 Top5 97.835938 -2023-02-13 18:06:45,344 - Epoch: [109][ 60/ 135] Loss 0.316934 Top1 84.062500 Top5 97.871094 -2023-02-13 18:06:45,473 - Epoch: [109][ 70/ 135] Loss 0.314857 Top1 84.017857 Top5 97.795759 -2023-02-13 18:06:45,608 - Epoch: [109][ 80/ 135] Loss 0.319412 Top1 83.891602 Top5 97.714844 -2023-02-13 18:06:45,737 - Epoch: [109][ 90/ 135] Loss 0.321864 Top1 83.975694 Top5 97.708333 -2023-02-13 18:06:45,864 - Epoch: [109][ 100/ 135] Loss 0.321769 Top1 83.945312 Top5 97.734375 -2023-02-13 18:06:45,992 - Epoch: [109][ 110/ 135] Loss 0.325983 Top1 83.902699 Top5 97.681108 -2023-02-13 18:06:46,120 - Epoch: [109][ 120/ 135] Loss 0.328577 Top1 83.818359 Top5 97.643229 -2023-02-13 18:06:46,251 - Epoch: [109][ 130/ 135] Loss 0.327131 Top1 83.768029 Top5 97.605168 -2023-02-13 18:06:46,297 - Epoch: [109][ 135/ 135] Loss 0.324253 Top1 83.795284 Top5 97.610096 -2023-02-13 18:06:46,369 - ==> Top1: 83.795 Top5: 97.610 Loss: 0.324 - -2023-02-13 18:06:46,370 - ==> Confusion: -[[ 844 6 8 4 6 2 1 1 6 53 0 6 0 4 5 3 4 1 2 3 8] - [ 4 945 1 3 12 21 1 17 4 1 3 0 1 0 2 1 5 0 3 3 6] - [ 5 6 962 12 1 1 13 13 1 1 5 2 2 6 5 3 2 4 5 4 5] - [ 4 3 24 885 1 5 1 2 2 2 17 0 3 4 22 3 3 11 18 1 5] - [ 23 12 0 1 976 7 1 3 3 3 0 7 1 5 9 5 6 0 0 1 3] - [ 1 20 2 7 4 949 1 26 4 8 3 12 1 18 1 2 4 0 2 3 2] - [ 2 5 17 2 1 5 1030 10 1 1 4 1 1 0 0 3 2 2 3 4 5] - [ 2 7 10 2 1 21 2 931 1 2 3 5 3 1 1 1 2 1 19 5 4] - [ 19 3 1 1 1 0 0 1 908 33 11 1 0 9 16 0 0 1 2 0 2] - [ 81 2 4 0 5 1 0 2 43 850 0 0 0 14 2 1 1 1 1 0 4] - [ 1 2 1 9 0 1 4 5 22 2 981 1 3 6 4 0 1 1 6 0 1] - [ 2 3 1 0 3 12 1 7 1 2 0 914 18 12 4 3 3 7 2 7 3] - [ 0 0 1 7 0 4 1 2 2 1 1 40 848 2 4 6 4 16 2 1 17] - [ 3 3 2 3 3 3 1 1 23 19 13 3 4 925 3 2 4 1 1 0 7] - [ 8 3 3 15 2 2 0 2 26 7 5 1 1 2 990 1 1 7 8 0 8] - [ 5 3 7 1 6 2 6 1 0 0 0 11 7 6 1 949 10 15 1 7 8] - [ 4 6 2 1 7 3 1 0 3 1 2 0 2 4 3 10 992 3 4 2 11] - [ 6 5 1 3 1 0 2 0 0 1 2 12 15 2 2 17 1 971 0 1 9] - [ 4 4 8 13 1 2 0 26 3 1 3 0 3 2 15 1 0 2 992 3 3] - [ 0 4 1 1 1 4 5 16 1 0 2 17 2 4 0 6 6 4 0 1061 13] - [ 163 294 237 114 130 183 84 171 152 110 240 115 263 320 167 74 248 111 178 232 9848]] - -2023-02-13 18:06:46,371 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:06:46,372 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:06:46,377 - - -2023-02-13 18:06:46,377 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:06:47,254 - Epoch: [110][ 10/ 1207] Overall Loss 0.242771 Objective Loss 0.242771 LR 0.000500 Time 0.087618 -2023-02-13 18:06:47,455 - Epoch: [110][ 20/ 1207] Overall Loss 0.264904 Objective Loss 0.264904 LR 0.000500 Time 0.053820 -2023-02-13 18:06:47,651 - Epoch: [110][ 30/ 1207] Overall Loss 0.258298 Objective Loss 0.258298 LR 0.000500 Time 0.042392 -2023-02-13 18:06:47,843 - Epoch: [110][ 40/ 1207] Overall Loss 0.252671 Objective Loss 0.252671 LR 0.000500 Time 0.036592 -2023-02-13 18:06:48,038 - Epoch: [110][ 50/ 1207] Overall Loss 0.254207 Objective Loss 0.254207 LR 0.000500 Time 0.033167 -2023-02-13 18:06:48,231 - Epoch: [110][ 60/ 1207] Overall Loss 0.251413 Objective Loss 0.251413 LR 0.000500 Time 0.030859 -2023-02-13 18:06:48,426 - Epoch: [110][ 70/ 1207] Overall Loss 0.251650 Objective Loss 0.251650 LR 0.000500 Time 0.029227 -2023-02-13 18:06:48,618 - Epoch: [110][ 80/ 1207] Overall Loss 0.250913 Objective Loss 0.250913 LR 0.000500 Time 0.027972 -2023-02-13 18:06:48,813 - Epoch: [110][ 90/ 1207] Overall Loss 0.254509 Objective Loss 0.254509 LR 0.000500 Time 0.027026 -2023-02-13 18:06:49,005 - Epoch: [110][ 100/ 1207] Overall Loss 0.253128 Objective Loss 0.253128 LR 0.000500 Time 0.026238 -2023-02-13 18:06:49,201 - Epoch: [110][ 110/ 1207] Overall Loss 0.256947 Objective Loss 0.256947 LR 0.000500 Time 0.025629 -2023-02-13 18:06:49,393 - Epoch: [110][ 120/ 1207] Overall Loss 0.256134 Objective Loss 0.256134 LR 0.000500 Time 0.025089 -2023-02-13 18:06:49,587 - Epoch: [110][ 130/ 1207] Overall Loss 0.255517 Objective Loss 0.255517 LR 0.000500 Time 0.024653 -2023-02-13 18:06:49,780 - Epoch: [110][ 140/ 1207] Overall Loss 0.254646 Objective Loss 0.254646 LR 0.000500 Time 0.024264 -2023-02-13 18:06:49,975 - Epoch: [110][ 150/ 1207] Overall Loss 0.255591 Objective Loss 0.255591 LR 0.000500 Time 0.023944 -2023-02-13 18:06:50,168 - Epoch: [110][ 160/ 1207] Overall Loss 0.255337 Objective Loss 0.255337 LR 0.000500 Time 0.023650 -2023-02-13 18:06:50,363 - Epoch: [110][ 170/ 1207] Overall Loss 0.254929 Objective Loss 0.254929 LR 0.000500 Time 0.023403 -2023-02-13 18:06:50,556 - Epoch: [110][ 180/ 1207] Overall Loss 0.255337 Objective Loss 0.255337 LR 0.000500 Time 0.023173 -2023-02-13 18:06:50,753 - Epoch: [110][ 190/ 1207] Overall Loss 0.255097 Objective Loss 0.255097 LR 0.000500 Time 0.022990 -2023-02-13 18:06:50,947 - Epoch: [110][ 200/ 1207] Overall Loss 0.256543 Objective Loss 0.256543 LR 0.000500 Time 0.022810 -2023-02-13 18:06:51,145 - Epoch: [110][ 210/ 1207] Overall Loss 0.256091 Objective Loss 0.256091 LR 0.000500 Time 0.022665 -2023-02-13 18:06:51,338 - Epoch: [110][ 220/ 1207] Overall Loss 0.255164 Objective Loss 0.255164 LR 0.000500 Time 0.022510 -2023-02-13 18:06:51,535 - Epoch: [110][ 230/ 1207] Overall Loss 0.255233 Objective Loss 0.255233 LR 0.000500 Time 0.022387 -2023-02-13 18:06:51,730 - Epoch: [110][ 240/ 1207] Overall Loss 0.256221 Objective Loss 0.256221 LR 0.000500 Time 0.022262 -2023-02-13 18:06:51,927 - Epoch: [110][ 250/ 1207] Overall Loss 0.256690 Objective Loss 0.256690 LR 0.000500 Time 0.022158 -2023-02-13 18:06:52,119 - Epoch: [110][ 260/ 1207] Overall Loss 0.257452 Objective Loss 0.257452 LR 0.000500 Time 0.022045 -2023-02-13 18:06:52,317 - Epoch: [110][ 270/ 1207] Overall Loss 0.257880 Objective Loss 0.257880 LR 0.000500 Time 0.021960 -2023-02-13 18:06:52,511 - Epoch: [110][ 280/ 1207] Overall Loss 0.257721 Objective Loss 0.257721 LR 0.000500 Time 0.021866 -2023-02-13 18:06:52,708 - Epoch: [110][ 290/ 1207] Overall Loss 0.257927 Objective Loss 0.257927 LR 0.000500 Time 0.021790 -2023-02-13 18:06:52,901 - Epoch: [110][ 300/ 1207] Overall Loss 0.258528 Objective Loss 0.258528 LR 0.000500 Time 0.021707 -2023-02-13 18:06:53,099 - Epoch: [110][ 310/ 1207] Overall Loss 0.258131 Objective Loss 0.258131 LR 0.000500 Time 0.021642 -2023-02-13 18:06:53,292 - Epoch: [110][ 320/ 1207] Overall Loss 0.258966 Objective Loss 0.258966 LR 0.000500 Time 0.021569 -2023-02-13 18:06:53,489 - Epoch: [110][ 330/ 1207] Overall Loss 0.259075 Objective Loss 0.259075 LR 0.000500 Time 0.021512 -2023-02-13 18:06:53,684 - Epoch: [110][ 340/ 1207] Overall Loss 0.258968 Objective Loss 0.258968 LR 0.000500 Time 0.021451 -2023-02-13 18:06:53,882 - Epoch: [110][ 350/ 1207] Overall Loss 0.258973 Objective Loss 0.258973 LR 0.000500 Time 0.021403 -2023-02-13 18:06:54,077 - Epoch: [110][ 360/ 1207] Overall Loss 0.257841 Objective Loss 0.257841 LR 0.000500 Time 0.021350 -2023-02-13 18:06:54,276 - Epoch: [110][ 370/ 1207] Overall Loss 0.257429 Objective Loss 0.257429 LR 0.000500 Time 0.021309 -2023-02-13 18:06:54,471 - Epoch: [110][ 380/ 1207] Overall Loss 0.257808 Objective Loss 0.257808 LR 0.000500 Time 0.021260 -2023-02-13 18:06:54,669 - Epoch: [110][ 390/ 1207] Overall Loss 0.257175 Objective Loss 0.257175 LR 0.000500 Time 0.021221 -2023-02-13 18:06:54,864 - Epoch: [110][ 400/ 1207] Overall Loss 0.257598 Objective Loss 0.257598 LR 0.000500 Time 0.021178 -2023-02-13 18:06:55,062 - Epoch: [110][ 410/ 1207] Overall Loss 0.256757 Objective Loss 0.256757 LR 0.000500 Time 0.021143 -2023-02-13 18:06:55,258 - Epoch: [110][ 420/ 1207] Overall Loss 0.256550 Objective Loss 0.256550 LR 0.000500 Time 0.021105 -2023-02-13 18:06:55,456 - Epoch: [110][ 430/ 1207] Overall Loss 0.256246 Objective Loss 0.256246 LR 0.000500 Time 0.021074 -2023-02-13 18:06:55,651 - Epoch: [110][ 440/ 1207] Overall Loss 0.256337 Objective Loss 0.256337 LR 0.000500 Time 0.021037 -2023-02-13 18:06:55,849 - Epoch: [110][ 450/ 1207] Overall Loss 0.256186 Objective Loss 0.256186 LR 0.000500 Time 0.021010 -2023-02-13 18:06:56,045 - Epoch: [110][ 460/ 1207] Overall Loss 0.255706 Objective Loss 0.255706 LR 0.000500 Time 0.020978 -2023-02-13 18:06:56,244 - Epoch: [110][ 470/ 1207] Overall Loss 0.255791 Objective Loss 0.255791 LR 0.000500 Time 0.020954 -2023-02-13 18:06:56,439 - Epoch: [110][ 480/ 1207] Overall Loss 0.255390 Objective Loss 0.255390 LR 0.000500 Time 0.020923 -2023-02-13 18:06:56,638 - Epoch: [110][ 490/ 1207] Overall Loss 0.255469 Objective Loss 0.255469 LR 0.000500 Time 0.020902 -2023-02-13 18:06:56,834 - Epoch: [110][ 500/ 1207] Overall Loss 0.255485 Objective Loss 0.255485 LR 0.000500 Time 0.020874 -2023-02-13 18:06:57,033 - Epoch: [110][ 510/ 1207] Overall Loss 0.255524 Objective Loss 0.255524 LR 0.000500 Time 0.020854 -2023-02-13 18:06:57,229 - Epoch: [110][ 520/ 1207] Overall Loss 0.255417 Objective Loss 0.255417 LR 0.000500 Time 0.020829 -2023-02-13 18:06:57,425 - Epoch: [110][ 530/ 1207] Overall Loss 0.254783 Objective Loss 0.254783 LR 0.000500 Time 0.020807 -2023-02-13 18:06:57,618 - Epoch: [110][ 540/ 1207] Overall Loss 0.254877 Objective Loss 0.254877 LR 0.000500 Time 0.020778 -2023-02-13 18:06:57,814 - Epoch: [110][ 550/ 1207] Overall Loss 0.255220 Objective Loss 0.255220 LR 0.000500 Time 0.020755 -2023-02-13 18:06:58,007 - Epoch: [110][ 560/ 1207] Overall Loss 0.255101 Objective Loss 0.255101 LR 0.000500 Time 0.020728 -2023-02-13 18:06:58,200 - Epoch: [110][ 570/ 1207] Overall Loss 0.255335 Objective Loss 0.255335 LR 0.000500 Time 0.020703 -2023-02-13 18:06:58,389 - Epoch: [110][ 580/ 1207] Overall Loss 0.255681 Objective Loss 0.255681 LR 0.000500 Time 0.020671 -2023-02-13 18:06:58,577 - Epoch: [110][ 590/ 1207] Overall Loss 0.255596 Objective Loss 0.255596 LR 0.000500 Time 0.020639 -2023-02-13 18:06:58,766 - Epoch: [110][ 600/ 1207] Overall Loss 0.255425 Objective Loss 0.255425 LR 0.000500 Time 0.020609 -2023-02-13 18:06:58,956 - Epoch: [110][ 610/ 1207] Overall Loss 0.255914 Objective Loss 0.255914 LR 0.000500 Time 0.020581 -2023-02-13 18:06:59,144 - Epoch: [110][ 620/ 1207] Overall Loss 0.256117 Objective Loss 0.256117 LR 0.000500 Time 0.020553 -2023-02-13 18:06:59,334 - Epoch: [110][ 630/ 1207] Overall Loss 0.256170 Objective Loss 0.256170 LR 0.000500 Time 0.020528 -2023-02-13 18:06:59,523 - Epoch: [110][ 640/ 1207] Overall Loss 0.256075 Objective Loss 0.256075 LR 0.000500 Time 0.020502 -2023-02-13 18:06:59,713 - Epoch: [110][ 650/ 1207] Overall Loss 0.256166 Objective Loss 0.256166 LR 0.000500 Time 0.020477 -2023-02-13 18:06:59,901 - Epoch: [110][ 660/ 1207] Overall Loss 0.256145 Objective Loss 0.256145 LR 0.000500 Time 0.020453 -2023-02-13 18:07:00,091 - Epoch: [110][ 670/ 1207] Overall Loss 0.256430 Objective Loss 0.256430 LR 0.000500 Time 0.020429 -2023-02-13 18:07:00,280 - Epoch: [110][ 680/ 1207] Overall Loss 0.256448 Objective Loss 0.256448 LR 0.000500 Time 0.020407 -2023-02-13 18:07:00,470 - Epoch: [110][ 690/ 1207] Overall Loss 0.256709 Objective Loss 0.256709 LR 0.000500 Time 0.020385 -2023-02-13 18:07:00,659 - Epoch: [110][ 700/ 1207] Overall Loss 0.257221 Objective Loss 0.257221 LR 0.000500 Time 0.020364 -2023-02-13 18:07:00,849 - Epoch: [110][ 710/ 1207] Overall Loss 0.257231 Objective Loss 0.257231 LR 0.000500 Time 0.020345 -2023-02-13 18:07:01,038 - Epoch: [110][ 720/ 1207] Overall Loss 0.257604 Objective Loss 0.257604 LR 0.000500 Time 0.020324 -2023-02-13 18:07:01,228 - Epoch: [110][ 730/ 1207] Overall Loss 0.257653 Objective Loss 0.257653 LR 0.000500 Time 0.020305 -2023-02-13 18:07:01,418 - Epoch: [110][ 740/ 1207] Overall Loss 0.257903 Objective Loss 0.257903 LR 0.000500 Time 0.020286 -2023-02-13 18:07:01,607 - Epoch: [110][ 750/ 1207] Overall Loss 0.257865 Objective Loss 0.257865 LR 0.000500 Time 0.020267 -2023-02-13 18:07:01,796 - Epoch: [110][ 760/ 1207] Overall Loss 0.257920 Objective Loss 0.257920 LR 0.000500 Time 0.020250 -2023-02-13 18:07:01,986 - Epoch: [110][ 770/ 1207] Overall Loss 0.258214 Objective Loss 0.258214 LR 0.000500 Time 0.020232 -2023-02-13 18:07:02,175 - Epoch: [110][ 780/ 1207] Overall Loss 0.258264 Objective Loss 0.258264 LR 0.000500 Time 0.020215 -2023-02-13 18:07:02,364 - Epoch: [110][ 790/ 1207] Overall Loss 0.257965 Objective Loss 0.257965 LR 0.000500 Time 0.020198 -2023-02-13 18:07:02,555 - Epoch: [110][ 800/ 1207] Overall Loss 0.258119 Objective Loss 0.258119 LR 0.000500 Time 0.020184 -2023-02-13 18:07:02,745 - Epoch: [110][ 810/ 1207] Overall Loss 0.258304 Objective Loss 0.258304 LR 0.000500 Time 0.020168 -2023-02-13 18:07:02,934 - Epoch: [110][ 820/ 1207] Overall Loss 0.258207 Objective Loss 0.258207 LR 0.000500 Time 0.020152 -2023-02-13 18:07:03,123 - Epoch: [110][ 830/ 1207] Overall Loss 0.258521 Objective Loss 0.258521 LR 0.000500 Time 0.020138 -2023-02-13 18:07:03,313 - Epoch: [110][ 840/ 1207] Overall Loss 0.258476 Objective Loss 0.258476 LR 0.000500 Time 0.020123 -2023-02-13 18:07:03,502 - Epoch: [110][ 850/ 1207] Overall Loss 0.258235 Objective Loss 0.258235 LR 0.000500 Time 0.020108 -2023-02-13 18:07:03,691 - Epoch: [110][ 860/ 1207] Overall Loss 0.258163 Objective Loss 0.258163 LR 0.000500 Time 0.020094 -2023-02-13 18:07:03,881 - Epoch: [110][ 870/ 1207] Overall Loss 0.258317 Objective Loss 0.258317 LR 0.000500 Time 0.020080 -2023-02-13 18:07:04,070 - Epoch: [110][ 880/ 1207] Overall Loss 0.258168 Objective Loss 0.258168 LR 0.000500 Time 0.020067 -2023-02-13 18:07:04,259 - Epoch: [110][ 890/ 1207] Overall Loss 0.258391 Objective Loss 0.258391 LR 0.000500 Time 0.020054 -2023-02-13 18:07:04,448 - Epoch: [110][ 900/ 1207] Overall Loss 0.258146 Objective Loss 0.258146 LR 0.000500 Time 0.020041 -2023-02-13 18:07:04,638 - Epoch: [110][ 910/ 1207] Overall Loss 0.258177 Objective Loss 0.258177 LR 0.000500 Time 0.020028 -2023-02-13 18:07:04,826 - Epoch: [110][ 920/ 1207] Overall Loss 0.258189 Objective Loss 0.258189 LR 0.000500 Time 0.020015 -2023-02-13 18:07:05,016 - Epoch: [110][ 930/ 1207] Overall Loss 0.258262 Objective Loss 0.258262 LR 0.000500 Time 0.020003 -2023-02-13 18:07:05,205 - Epoch: [110][ 940/ 1207] Overall Loss 0.258554 Objective Loss 0.258554 LR 0.000500 Time 0.019991 -2023-02-13 18:07:05,394 - Epoch: [110][ 950/ 1207] Overall Loss 0.258651 Objective Loss 0.258651 LR 0.000500 Time 0.019979 -2023-02-13 18:07:05,584 - Epoch: [110][ 960/ 1207] Overall Loss 0.258514 Objective Loss 0.258514 LR 0.000500 Time 0.019969 -2023-02-13 18:07:05,774 - Epoch: [110][ 970/ 1207] Overall Loss 0.258490 Objective Loss 0.258490 LR 0.000500 Time 0.019958 -2023-02-13 18:07:05,964 - Epoch: [110][ 980/ 1207] Overall Loss 0.258676 Objective Loss 0.258676 LR 0.000500 Time 0.019948 -2023-02-13 18:07:06,153 - Epoch: [110][ 990/ 1207] Overall Loss 0.258732 Objective Loss 0.258732 LR 0.000500 Time 0.019937 -2023-02-13 18:07:06,343 - Epoch: [110][ 1000/ 1207] Overall Loss 0.258721 Objective Loss 0.258721 LR 0.000500 Time 0.019927 -2023-02-13 18:07:06,532 - Epoch: [110][ 1010/ 1207] Overall Loss 0.258721 Objective Loss 0.258721 LR 0.000500 Time 0.019917 -2023-02-13 18:07:06,722 - Epoch: [110][ 1020/ 1207] Overall Loss 0.258856 Objective Loss 0.258856 LR 0.000500 Time 0.019907 -2023-02-13 18:07:06,911 - Epoch: [110][ 1030/ 1207] Overall Loss 0.258693 Objective Loss 0.258693 LR 0.000500 Time 0.019898 -2023-02-13 18:07:07,101 - Epoch: [110][ 1040/ 1207] Overall Loss 0.258494 Objective Loss 0.258494 LR 0.000500 Time 0.019888 -2023-02-13 18:07:07,291 - Epoch: [110][ 1050/ 1207] Overall Loss 0.258429 Objective Loss 0.258429 LR 0.000500 Time 0.019879 -2023-02-13 18:07:07,479 - Epoch: [110][ 1060/ 1207] Overall Loss 0.258474 Objective Loss 0.258474 LR 0.000500 Time 0.019869 -2023-02-13 18:07:07,669 - Epoch: [110][ 1070/ 1207] Overall Loss 0.258604 Objective Loss 0.258604 LR 0.000500 Time 0.019860 -2023-02-13 18:07:07,859 - Epoch: [110][ 1080/ 1207] Overall Loss 0.258873 Objective Loss 0.258873 LR 0.000500 Time 0.019852 -2023-02-13 18:07:08,048 - Epoch: [110][ 1090/ 1207] Overall Loss 0.259095 Objective Loss 0.259095 LR 0.000500 Time 0.019843 -2023-02-13 18:07:08,237 - Epoch: [110][ 1100/ 1207] Overall Loss 0.259080 Objective Loss 0.259080 LR 0.000500 Time 0.019834 -2023-02-13 18:07:08,427 - Epoch: [110][ 1110/ 1207] Overall Loss 0.259015 Objective Loss 0.259015 LR 0.000500 Time 0.019826 -2023-02-13 18:07:08,616 - Epoch: [110][ 1120/ 1207] Overall Loss 0.259177 Objective Loss 0.259177 LR 0.000500 Time 0.019818 -2023-02-13 18:07:08,806 - Epoch: [110][ 1130/ 1207] Overall Loss 0.259253 Objective Loss 0.259253 LR 0.000500 Time 0.019810 -2023-02-13 18:07:08,995 - Epoch: [110][ 1140/ 1207] Overall Loss 0.259037 Objective Loss 0.259037 LR 0.000500 Time 0.019802 -2023-02-13 18:07:09,184 - Epoch: [110][ 1150/ 1207] Overall Loss 0.259076 Objective Loss 0.259076 LR 0.000500 Time 0.019794 -2023-02-13 18:07:09,374 - Epoch: [110][ 1160/ 1207] Overall Loss 0.258919 Objective Loss 0.258919 LR 0.000500 Time 0.019786 -2023-02-13 18:07:09,563 - Epoch: [110][ 1170/ 1207] Overall Loss 0.258906 Objective Loss 0.258906 LR 0.000500 Time 0.019778 -2023-02-13 18:07:09,752 - Epoch: [110][ 1180/ 1207] Overall Loss 0.259126 Objective Loss 0.259126 LR 0.000500 Time 0.019771 -2023-02-13 18:07:09,941 - Epoch: [110][ 1190/ 1207] Overall Loss 0.259099 Objective Loss 0.259099 LR 0.000500 Time 0.019763 -2023-02-13 18:07:10,182 - Epoch: [110][ 1200/ 1207] Overall Loss 0.259200 Objective Loss 0.259200 LR 0.000500 Time 0.019799 -2023-02-13 18:07:10,298 - Epoch: [110][ 1207/ 1207] Overall Loss 0.259401 Objective Loss 0.259401 Top1 85.670732 Top5 97.865854 LR 0.000500 Time 0.019780 -2023-02-13 18:07:10,370 - --- validate (epoch=110)----------- -2023-02-13 18:07:10,370 - 34311 samples (256 per mini-batch) -2023-02-13 18:07:10,780 - Epoch: [110][ 10/ 135] Loss 0.364854 Top1 82.226562 Top5 96.914062 -2023-02-13 18:07:10,910 - Epoch: [110][ 20/ 135] Loss 0.341998 Top1 82.773438 Top5 97.285156 -2023-02-13 18:07:11,041 - Epoch: [110][ 30/ 135] Loss 0.325305 Top1 83.372396 Top5 97.330729 -2023-02-13 18:07:11,170 - Epoch: [110][ 40/ 135] Loss 0.326217 Top1 83.525391 Top5 97.421875 -2023-02-13 18:07:11,301 - Epoch: [110][ 50/ 135] Loss 0.318536 Top1 83.726562 Top5 97.570312 -2023-02-13 18:07:11,426 - Epoch: [110][ 60/ 135] Loss 0.322238 Top1 83.697917 Top5 97.597656 -2023-02-13 18:07:11,555 - Epoch: [110][ 70/ 135] Loss 0.321939 Top1 83.744420 Top5 97.544643 -2023-02-13 18:07:11,684 - Epoch: [110][ 80/ 135] Loss 0.322185 Top1 83.632812 Top5 97.597656 -2023-02-13 18:07:11,826 - Epoch: [110][ 90/ 135] Loss 0.322109 Top1 83.650174 Top5 97.595486 -2023-02-13 18:07:11,961 - Epoch: [110][ 100/ 135] Loss 0.324517 Top1 83.621094 Top5 97.605469 -2023-02-13 18:07:12,105 - Epoch: [110][ 110/ 135] Loss 0.322760 Top1 83.668324 Top5 97.602983 -2023-02-13 18:07:12,236 - Epoch: [110][ 120/ 135] Loss 0.323500 Top1 83.697917 Top5 97.548828 -2023-02-13 18:07:12,367 - Epoch: [110][ 130/ 135] Loss 0.324409 Top1 83.617788 Top5 97.551082 -2023-02-13 18:07:12,415 - Epoch: [110][ 135/ 135] Loss 0.325180 Top1 83.565037 Top5 97.531404 -2023-02-13 18:07:12,484 - ==> Top1: 83.565 Top5: 97.531 Loss: 0.325 - -2023-02-13 18:07:12,484 - ==> Confusion: -[[ 848 5 5 2 8 1 1 1 4 62 0 3 2 4 7 5 1 0 2 4 2] - [ 2 933 0 2 12 27 2 23 3 1 1 0 2 0 3 2 5 0 6 3 6] - [ 8 4 943 17 5 1 19 18 0 3 4 1 4 5 4 2 3 1 8 3 5] - [ 4 1 19 893 3 3 1 2 2 4 9 0 7 3 29 1 4 9 15 1 6] - [ 22 10 0 1 983 7 1 1 0 7 1 6 0 3 5 5 5 2 0 5 2] - [ 1 15 2 6 8 958 3 20 1 8 2 10 4 15 2 3 2 1 1 6 2] - [ 4 4 14 4 1 5 1033 12 1 1 3 0 0 0 0 2 0 4 3 4 4] - [ 0 5 9 1 1 21 4 945 1 2 1 5 3 1 0 0 1 1 11 8 4] - [ 13 2 0 2 1 1 0 3 880 62 7 2 1 10 15 3 0 1 4 0 2] - [ 61 2 5 0 3 0 0 4 25 887 0 0 0 14 3 2 1 2 1 1 1] - [ 2 3 7 7 1 2 1 5 15 1 969 3 0 10 7 1 1 1 12 0 3] - [ 1 3 1 1 8 13 0 7 0 3 0 910 17 9 1 7 3 6 3 8 4] - [ 1 0 1 7 0 5 0 2 3 0 1 31 859 0 6 8 3 21 2 1 8] - [ 3 7 3 0 7 7 0 3 7 26 6 4 4 931 3 7 2 2 0 0 2] - [ 11 2 1 8 4 4 0 2 21 11 2 2 6 1 997 1 0 6 6 1 6] - [ 3 3 4 0 9 1 3 0 0 0 0 4 5 6 1 974 7 15 0 6 5] - [ 3 4 0 1 10 3 0 1 2 1 1 0 2 3 2 13 998 1 1 4 11] - [ 5 3 0 4 0 1 1 2 1 1 0 6 11 2 3 17 0 989 0 3 2] - [ 3 4 4 13 2 1 0 27 5 1 1 2 3 0 17 2 0 3 995 2 1] - [ 1 4 2 1 2 8 9 22 0 0 1 12 4 6 1 4 4 4 0 1058 5] - [ 179 252 244 137 158 196 94 216 98 123 163 106 316 342 192 100 249 124 198 258 9689]] - -2023-02-13 18:07:12,486 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:07:12,486 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:07:12,492 - - -2023-02-13 18:07:12,492 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:07:13,460 - Epoch: [111][ 10/ 1207] Overall Loss 0.228782 Objective Loss 0.228782 LR 0.000500 Time 0.096786 -2023-02-13 18:07:13,649 - Epoch: [111][ 20/ 1207] Overall Loss 0.234569 Objective Loss 0.234569 LR 0.000500 Time 0.057818 -2023-02-13 18:07:13,837 - Epoch: [111][ 30/ 1207] Overall Loss 0.244622 Objective Loss 0.244622 LR 0.000500 Time 0.044787 -2023-02-13 18:07:14,025 - Epoch: [111][ 40/ 1207] Overall Loss 0.250493 Objective Loss 0.250493 LR 0.000500 Time 0.038274 -2023-02-13 18:07:14,212 - Epoch: [111][ 50/ 1207] Overall Loss 0.247589 Objective Loss 0.247589 LR 0.000500 Time 0.034364 -2023-02-13 18:07:14,400 - Epoch: [111][ 60/ 1207] Overall Loss 0.243589 Objective Loss 0.243589 LR 0.000500 Time 0.031767 -2023-02-13 18:07:14,589 - Epoch: [111][ 70/ 1207] Overall Loss 0.247235 Objective Loss 0.247235 LR 0.000500 Time 0.029918 -2023-02-13 18:07:14,777 - Epoch: [111][ 80/ 1207] Overall Loss 0.247771 Objective Loss 0.247771 LR 0.000500 Time 0.028526 -2023-02-13 18:07:14,965 - Epoch: [111][ 90/ 1207] Overall Loss 0.248177 Objective Loss 0.248177 LR 0.000500 Time 0.027434 -2023-02-13 18:07:15,153 - Epoch: [111][ 100/ 1207] Overall Loss 0.252572 Objective Loss 0.252572 LR 0.000500 Time 0.026572 -2023-02-13 18:07:15,341 - Epoch: [111][ 110/ 1207] Overall Loss 0.254881 Objective Loss 0.254881 LR 0.000500 Time 0.025863 -2023-02-13 18:07:15,529 - Epoch: [111][ 120/ 1207] Overall Loss 0.253880 Objective Loss 0.253880 LR 0.000500 Time 0.025268 -2023-02-13 18:07:15,716 - Epoch: [111][ 130/ 1207] Overall Loss 0.252846 Objective Loss 0.252846 LR 0.000500 Time 0.024763 -2023-02-13 18:07:15,905 - Epoch: [111][ 140/ 1207] Overall Loss 0.252324 Objective Loss 0.252324 LR 0.000500 Time 0.024341 -2023-02-13 18:07:16,093 - Epoch: [111][ 150/ 1207] Overall Loss 0.253039 Objective Loss 0.253039 LR 0.000500 Time 0.023967 -2023-02-13 18:07:16,282 - Epoch: [111][ 160/ 1207] Overall Loss 0.252390 Objective Loss 0.252390 LR 0.000500 Time 0.023647 -2023-02-13 18:07:16,468 - Epoch: [111][ 170/ 1207] Overall Loss 0.253063 Objective Loss 0.253063 LR 0.000500 Time 0.023351 -2023-02-13 18:07:16,656 - Epoch: [111][ 180/ 1207] Overall Loss 0.253012 Objective Loss 0.253012 LR 0.000500 Time 0.023096 -2023-02-13 18:07:16,844 - Epoch: [111][ 190/ 1207] Overall Loss 0.253812 Objective Loss 0.253812 LR 0.000500 Time 0.022865 -2023-02-13 18:07:17,032 - Epoch: [111][ 200/ 1207] Overall Loss 0.255009 Objective Loss 0.255009 LR 0.000500 Time 0.022662 -2023-02-13 18:07:17,219 - Epoch: [111][ 210/ 1207] Overall Loss 0.255488 Objective Loss 0.255488 LR 0.000500 Time 0.022471 -2023-02-13 18:07:17,408 - Epoch: [111][ 220/ 1207] Overall Loss 0.255389 Objective Loss 0.255389 LR 0.000500 Time 0.022305 -2023-02-13 18:07:17,595 - Epoch: [111][ 230/ 1207] Overall Loss 0.255123 Objective Loss 0.255123 LR 0.000500 Time 0.022150 -2023-02-13 18:07:17,784 - Epoch: [111][ 240/ 1207] Overall Loss 0.256576 Objective Loss 0.256576 LR 0.000500 Time 0.022010 -2023-02-13 18:07:17,971 - Epoch: [111][ 250/ 1207] Overall Loss 0.256863 Objective Loss 0.256863 LR 0.000500 Time 0.021879 -2023-02-13 18:07:18,160 - Epoch: [111][ 260/ 1207] Overall Loss 0.256459 Objective Loss 0.256459 LR 0.000500 Time 0.021760 -2023-02-13 18:07:18,348 - Epoch: [111][ 270/ 1207] Overall Loss 0.256819 Objective Loss 0.256819 LR 0.000500 Time 0.021651 -2023-02-13 18:07:18,537 - Epoch: [111][ 280/ 1207] Overall Loss 0.257427 Objective Loss 0.257427 LR 0.000500 Time 0.021552 -2023-02-13 18:07:18,727 - Epoch: [111][ 290/ 1207] Overall Loss 0.257670 Objective Loss 0.257670 LR 0.000500 Time 0.021461 -2023-02-13 18:07:18,916 - Epoch: [111][ 300/ 1207] Overall Loss 0.257682 Objective Loss 0.257682 LR 0.000500 Time 0.021376 -2023-02-13 18:07:19,106 - Epoch: [111][ 310/ 1207] Overall Loss 0.256887 Objective Loss 0.256887 LR 0.000500 Time 0.021297 -2023-02-13 18:07:19,297 - Epoch: [111][ 320/ 1207] Overall Loss 0.255456 Objective Loss 0.255456 LR 0.000500 Time 0.021227 -2023-02-13 18:07:19,487 - Epoch: [111][ 330/ 1207] Overall Loss 0.255788 Objective Loss 0.255788 LR 0.000500 Time 0.021158 -2023-02-13 18:07:19,676 - Epoch: [111][ 340/ 1207] Overall Loss 0.256016 Objective Loss 0.256016 LR 0.000500 Time 0.021092 -2023-02-13 18:07:19,866 - Epoch: [111][ 350/ 1207] Overall Loss 0.256109 Objective Loss 0.256109 LR 0.000500 Time 0.021030 -2023-02-13 18:07:20,055 - Epoch: [111][ 360/ 1207] Overall Loss 0.255843 Objective Loss 0.255843 LR 0.000500 Time 0.020970 -2023-02-13 18:07:20,245 - Epoch: [111][ 370/ 1207] Overall Loss 0.256566 Objective Loss 0.256566 LR 0.000500 Time 0.020916 -2023-02-13 18:07:20,435 - Epoch: [111][ 380/ 1207] Overall Loss 0.256849 Objective Loss 0.256849 LR 0.000500 Time 0.020864 -2023-02-13 18:07:20,624 - Epoch: [111][ 390/ 1207] Overall Loss 0.257135 Objective Loss 0.257135 LR 0.000500 Time 0.020812 -2023-02-13 18:07:20,812 - Epoch: [111][ 400/ 1207] Overall Loss 0.256422 Objective Loss 0.256422 LR 0.000500 Time 0.020762 -2023-02-13 18:07:21,001 - Epoch: [111][ 410/ 1207] Overall Loss 0.255126 Objective Loss 0.255126 LR 0.000500 Time 0.020715 -2023-02-13 18:07:21,189 - Epoch: [111][ 420/ 1207] Overall Loss 0.254883 Objective Loss 0.254883 LR 0.000500 Time 0.020670 -2023-02-13 18:07:21,377 - Epoch: [111][ 430/ 1207] Overall Loss 0.255046 Objective Loss 0.255046 LR 0.000500 Time 0.020625 -2023-02-13 18:07:21,565 - Epoch: [111][ 440/ 1207] Overall Loss 0.254198 Objective Loss 0.254198 LR 0.000500 Time 0.020583 -2023-02-13 18:07:21,754 - Epoch: [111][ 450/ 1207] Overall Loss 0.254245 Objective Loss 0.254245 LR 0.000500 Time 0.020544 -2023-02-13 18:07:21,942 - Epoch: [111][ 460/ 1207] Overall Loss 0.253644 Objective Loss 0.253644 LR 0.000500 Time 0.020506 -2023-02-13 18:07:22,130 - Epoch: [111][ 470/ 1207] Overall Loss 0.253718 Objective Loss 0.253718 LR 0.000500 Time 0.020468 -2023-02-13 18:07:22,319 - Epoch: [111][ 480/ 1207] Overall Loss 0.253526 Objective Loss 0.253526 LR 0.000500 Time 0.020435 -2023-02-13 18:07:22,507 - Epoch: [111][ 490/ 1207] Overall Loss 0.253370 Objective Loss 0.253370 LR 0.000500 Time 0.020400 -2023-02-13 18:07:22,695 - Epoch: [111][ 500/ 1207] Overall Loss 0.252955 Objective Loss 0.252955 LR 0.000500 Time 0.020368 -2023-02-13 18:07:22,883 - Epoch: [111][ 510/ 1207] Overall Loss 0.252969 Objective Loss 0.252969 LR 0.000500 Time 0.020336 -2023-02-13 18:07:23,071 - Epoch: [111][ 520/ 1207] Overall Loss 0.252937 Objective Loss 0.252937 LR 0.000500 Time 0.020306 -2023-02-13 18:07:23,259 - Epoch: [111][ 530/ 1207] Overall Loss 0.252681 Objective Loss 0.252681 LR 0.000500 Time 0.020277 -2023-02-13 18:07:23,448 - Epoch: [111][ 540/ 1207] Overall Loss 0.252480 Objective Loss 0.252480 LR 0.000500 Time 0.020250 -2023-02-13 18:07:23,636 - Epoch: [111][ 550/ 1207] Overall Loss 0.252767 Objective Loss 0.252767 LR 0.000500 Time 0.020223 -2023-02-13 18:07:23,824 - Epoch: [111][ 560/ 1207] Overall Loss 0.253279 Objective Loss 0.253279 LR 0.000500 Time 0.020197 -2023-02-13 18:07:24,011 - Epoch: [111][ 570/ 1207] Overall Loss 0.253246 Objective Loss 0.253246 LR 0.000500 Time 0.020171 -2023-02-13 18:07:24,199 - Epoch: [111][ 580/ 1207] Overall Loss 0.253109 Objective Loss 0.253109 LR 0.000500 Time 0.020147 -2023-02-13 18:07:24,387 - Epoch: [111][ 590/ 1207] Overall Loss 0.253226 Objective Loss 0.253226 LR 0.000500 Time 0.020123 -2023-02-13 18:07:24,574 - Epoch: [111][ 600/ 1207] Overall Loss 0.253138 Objective Loss 0.253138 LR 0.000500 Time 0.020099 -2023-02-13 18:07:24,762 - Epoch: [111][ 610/ 1207] Overall Loss 0.253501 Objective Loss 0.253501 LR 0.000500 Time 0.020077 -2023-02-13 18:07:24,949 - Epoch: [111][ 620/ 1207] Overall Loss 0.253490 Objective Loss 0.253490 LR 0.000500 Time 0.020054 -2023-02-13 18:07:25,136 - Epoch: [111][ 630/ 1207] Overall Loss 0.253713 Objective Loss 0.253713 LR 0.000500 Time 0.020033 -2023-02-13 18:07:25,324 - Epoch: [111][ 640/ 1207] Overall Loss 0.253578 Objective Loss 0.253578 LR 0.000500 Time 0.020012 -2023-02-13 18:07:25,512 - Epoch: [111][ 650/ 1207] Overall Loss 0.253763 Objective Loss 0.253763 LR 0.000500 Time 0.019993 -2023-02-13 18:07:25,700 - Epoch: [111][ 660/ 1207] Overall Loss 0.253664 Objective Loss 0.253664 LR 0.000500 Time 0.019974 -2023-02-13 18:07:25,888 - Epoch: [111][ 670/ 1207] Overall Loss 0.253946 Objective Loss 0.253946 LR 0.000500 Time 0.019956 -2023-02-13 18:07:26,076 - Epoch: [111][ 680/ 1207] Overall Loss 0.254129 Objective Loss 0.254129 LR 0.000500 Time 0.019939 -2023-02-13 18:07:26,264 - Epoch: [111][ 690/ 1207] Overall Loss 0.254334 Objective Loss 0.254334 LR 0.000500 Time 0.019921 -2023-02-13 18:07:26,452 - Epoch: [111][ 700/ 1207] Overall Loss 0.254119 Objective Loss 0.254119 LR 0.000500 Time 0.019905 -2023-02-13 18:07:26,640 - Epoch: [111][ 710/ 1207] Overall Loss 0.254044 Objective Loss 0.254044 LR 0.000500 Time 0.019889 -2023-02-13 18:07:26,828 - Epoch: [111][ 720/ 1207] Overall Loss 0.253599 Objective Loss 0.253599 LR 0.000500 Time 0.019874 -2023-02-13 18:07:27,016 - Epoch: [111][ 730/ 1207] Overall Loss 0.253552 Objective Loss 0.253552 LR 0.000500 Time 0.019858 -2023-02-13 18:07:27,204 - Epoch: [111][ 740/ 1207] Overall Loss 0.253634 Objective Loss 0.253634 LR 0.000500 Time 0.019843 -2023-02-13 18:07:27,391 - Epoch: [111][ 750/ 1207] Overall Loss 0.253605 Objective Loss 0.253605 LR 0.000500 Time 0.019828 -2023-02-13 18:07:27,579 - Epoch: [111][ 760/ 1207] Overall Loss 0.253369 Objective Loss 0.253369 LR 0.000500 Time 0.019814 -2023-02-13 18:07:27,766 - Epoch: [111][ 770/ 1207] Overall Loss 0.253225 Objective Loss 0.253225 LR 0.000500 Time 0.019799 -2023-02-13 18:07:27,954 - Epoch: [111][ 780/ 1207] Overall Loss 0.253287 Objective Loss 0.253287 LR 0.000500 Time 0.019785 -2023-02-13 18:07:28,141 - Epoch: [111][ 790/ 1207] Overall Loss 0.253310 Objective Loss 0.253310 LR 0.000500 Time 0.019771 -2023-02-13 18:07:28,329 - Epoch: [111][ 800/ 1207] Overall Loss 0.253087 Objective Loss 0.253087 LR 0.000500 Time 0.019758 -2023-02-13 18:07:28,517 - Epoch: [111][ 810/ 1207] Overall Loss 0.253009 Objective Loss 0.253009 LR 0.000500 Time 0.019745 -2023-02-13 18:07:28,705 - Epoch: [111][ 820/ 1207] Overall Loss 0.253138 Objective Loss 0.253138 LR 0.000500 Time 0.019734 -2023-02-13 18:07:28,892 - Epoch: [111][ 830/ 1207] Overall Loss 0.253539 Objective Loss 0.253539 LR 0.000500 Time 0.019721 -2023-02-13 18:07:29,080 - Epoch: [111][ 840/ 1207] Overall Loss 0.253638 Objective Loss 0.253638 LR 0.000500 Time 0.019710 -2023-02-13 18:07:29,268 - Epoch: [111][ 850/ 1207] Overall Loss 0.253719 Objective Loss 0.253719 LR 0.000500 Time 0.019698 -2023-02-13 18:07:29,456 - Epoch: [111][ 860/ 1207] Overall Loss 0.253802 Objective Loss 0.253802 LR 0.000500 Time 0.019688 -2023-02-13 18:07:29,644 - Epoch: [111][ 870/ 1207] Overall Loss 0.254050 Objective Loss 0.254050 LR 0.000500 Time 0.019677 -2023-02-13 18:07:29,833 - Epoch: [111][ 880/ 1207] Overall Loss 0.253913 Objective Loss 0.253913 LR 0.000500 Time 0.019668 -2023-02-13 18:07:30,020 - Epoch: [111][ 890/ 1207] Overall Loss 0.254019 Objective Loss 0.254019 LR 0.000500 Time 0.019657 -2023-02-13 18:07:30,208 - Epoch: [111][ 900/ 1207] Overall Loss 0.254051 Objective Loss 0.254051 LR 0.000500 Time 0.019647 -2023-02-13 18:07:30,396 - Epoch: [111][ 910/ 1207] Overall Loss 0.253861 Objective Loss 0.253861 LR 0.000500 Time 0.019636 -2023-02-13 18:07:30,583 - Epoch: [111][ 920/ 1207] Overall Loss 0.253969 Objective Loss 0.253969 LR 0.000500 Time 0.019626 -2023-02-13 18:07:30,770 - Epoch: [111][ 930/ 1207] Overall Loss 0.253750 Objective Loss 0.253750 LR 0.000500 Time 0.019616 -2023-02-13 18:07:30,959 - Epoch: [111][ 940/ 1207] Overall Loss 0.253963 Objective Loss 0.253963 LR 0.000500 Time 0.019608 -2023-02-13 18:07:31,146 - Epoch: [111][ 950/ 1207] Overall Loss 0.254148 Objective Loss 0.254148 LR 0.000500 Time 0.019598 -2023-02-13 18:07:31,334 - Epoch: [111][ 960/ 1207] Overall Loss 0.254105 Objective Loss 0.254105 LR 0.000500 Time 0.019590 -2023-02-13 18:07:31,522 - Epoch: [111][ 970/ 1207] Overall Loss 0.254353 Objective Loss 0.254353 LR 0.000500 Time 0.019581 -2023-02-13 18:07:31,710 - Epoch: [111][ 980/ 1207] Overall Loss 0.254525 Objective Loss 0.254525 LR 0.000500 Time 0.019572 -2023-02-13 18:07:31,898 - Epoch: [111][ 990/ 1207] Overall Loss 0.254797 Objective Loss 0.254797 LR 0.000500 Time 0.019564 -2023-02-13 18:07:32,086 - Epoch: [111][ 1000/ 1207] Overall Loss 0.254998 Objective Loss 0.254998 LR 0.000500 Time 0.019556 -2023-02-13 18:07:32,272 - Epoch: [111][ 1010/ 1207] Overall Loss 0.255243 Objective Loss 0.255243 LR 0.000500 Time 0.019547 -2023-02-13 18:07:32,460 - Epoch: [111][ 1020/ 1207] Overall Loss 0.255392 Objective Loss 0.255392 LR 0.000500 Time 0.019539 -2023-02-13 18:07:32,647 - Epoch: [111][ 1030/ 1207] Overall Loss 0.255510 Objective Loss 0.255510 LR 0.000500 Time 0.019530 -2023-02-13 18:07:32,835 - Epoch: [111][ 1040/ 1207] Overall Loss 0.255510 Objective Loss 0.255510 LR 0.000500 Time 0.019522 -2023-02-13 18:07:33,021 - Epoch: [111][ 1050/ 1207] Overall Loss 0.255584 Objective Loss 0.255584 LR 0.000500 Time 0.019514 -2023-02-13 18:07:33,209 - Epoch: [111][ 1060/ 1207] Overall Loss 0.255768 Objective Loss 0.255768 LR 0.000500 Time 0.019506 -2023-02-13 18:07:33,397 - Epoch: [111][ 1070/ 1207] Overall Loss 0.255471 Objective Loss 0.255471 LR 0.000500 Time 0.019499 -2023-02-13 18:07:33,585 - Epoch: [111][ 1080/ 1207] Overall Loss 0.255610 Objective Loss 0.255610 LR 0.000500 Time 0.019492 -2023-02-13 18:07:33,774 - Epoch: [111][ 1090/ 1207] Overall Loss 0.255964 Objective Loss 0.255964 LR 0.000500 Time 0.019487 -2023-02-13 18:07:33,965 - Epoch: [111][ 1100/ 1207] Overall Loss 0.255875 Objective Loss 0.255875 LR 0.000500 Time 0.019483 -2023-02-13 18:07:34,156 - Epoch: [111][ 1110/ 1207] Overall Loss 0.255835 Objective Loss 0.255835 LR 0.000500 Time 0.019479 -2023-02-13 18:07:34,349 - Epoch: [111][ 1120/ 1207] Overall Loss 0.255739 Objective Loss 0.255739 LR 0.000500 Time 0.019477 -2023-02-13 18:07:34,540 - Epoch: [111][ 1130/ 1207] Overall Loss 0.255833 Objective Loss 0.255833 LR 0.000500 Time 0.019473 -2023-02-13 18:07:34,731 - Epoch: [111][ 1140/ 1207] Overall Loss 0.255966 Objective Loss 0.255966 LR 0.000500 Time 0.019470 -2023-02-13 18:07:34,922 - Epoch: [111][ 1150/ 1207] Overall Loss 0.256221 Objective Loss 0.256221 LR 0.000500 Time 0.019466 -2023-02-13 18:07:35,114 - Epoch: [111][ 1160/ 1207] Overall Loss 0.256071 Objective Loss 0.256071 LR 0.000500 Time 0.019463 -2023-02-13 18:07:35,304 - Epoch: [111][ 1170/ 1207] Overall Loss 0.256120 Objective Loss 0.256120 LR 0.000500 Time 0.019459 -2023-02-13 18:07:35,495 - Epoch: [111][ 1180/ 1207] Overall Loss 0.256042 Objective Loss 0.256042 LR 0.000500 Time 0.019456 -2023-02-13 18:07:35,685 - Epoch: [111][ 1190/ 1207] Overall Loss 0.256381 Objective Loss 0.256381 LR 0.000500 Time 0.019452 -2023-02-13 18:07:35,932 - Epoch: [111][ 1200/ 1207] Overall Loss 0.256448 Objective Loss 0.256448 LR 0.000500 Time 0.019495 -2023-02-13 18:07:36,048 - Epoch: [111][ 1207/ 1207] Overall Loss 0.256269 Objective Loss 0.256269 Top1 85.060976 Top5 97.865854 LR 0.000500 Time 0.019478 -2023-02-13 18:07:36,129 - --- validate (epoch=111)----------- -2023-02-13 18:07:36,129 - 34311 samples (256 per mini-batch) -2023-02-13 18:07:36,538 - Epoch: [111][ 10/ 135] Loss 0.320537 Top1 84.179688 Top5 97.656250 -2023-02-13 18:07:36,668 - Epoch: [111][ 20/ 135] Loss 0.329086 Top1 84.316406 Top5 97.597656 -2023-02-13 18:07:36,798 - Epoch: [111][ 30/ 135] Loss 0.317291 Top1 84.505208 Top5 97.500000 -2023-02-13 18:07:36,930 - Epoch: [111][ 40/ 135] Loss 0.314166 Top1 84.257812 Top5 97.460938 -2023-02-13 18:07:37,057 - Epoch: [111][ 50/ 135] Loss 0.317660 Top1 83.859375 Top5 97.445312 -2023-02-13 18:07:37,188 - Epoch: [111][ 60/ 135] Loss 0.315714 Top1 83.938802 Top5 97.493490 -2023-02-13 18:07:37,315 - Epoch: [111][ 70/ 135] Loss 0.321529 Top1 83.794643 Top5 97.500000 -2023-02-13 18:07:37,447 - Epoch: [111][ 80/ 135] Loss 0.323433 Top1 83.867188 Top5 97.509766 -2023-02-13 18:07:37,573 - Epoch: [111][ 90/ 135] Loss 0.322182 Top1 83.884549 Top5 97.547743 -2023-02-13 18:07:37,705 - Epoch: [111][ 100/ 135] Loss 0.323869 Top1 83.753906 Top5 97.488281 -2023-02-13 18:07:37,835 - Epoch: [111][ 110/ 135] Loss 0.322376 Top1 83.760653 Top5 97.542614 -2023-02-13 18:07:37,965 - Epoch: [111][ 120/ 135] Loss 0.323542 Top1 83.707682 Top5 97.532552 -2023-02-13 18:07:38,098 - Epoch: [111][ 130/ 135] Loss 0.322020 Top1 83.731971 Top5 97.545072 -2023-02-13 18:07:38,146 - Epoch: [111][ 135/ 135] Loss 0.319673 Top1 83.739908 Top5 97.543062 -2023-02-13 18:07:38,216 - ==> Top1: 83.740 Top5: 97.543 Loss: 0.320 - -2023-02-13 18:07:38,217 - ==> Confusion: -[[ 867 5 8 0 13 2 0 1 4 34 0 5 2 5 7 3 2 1 0 3 5] - [ 2 947 1 1 14 26 1 15 3 1 2 0 3 0 0 2 6 1 2 1 5] - [ 10 4 938 15 9 1 24 15 0 1 4 1 4 4 2 10 2 3 4 3 4] - [ 5 4 21 901 6 4 0 1 2 1 14 0 8 2 16 1 4 9 12 2 3] - [ 14 8 0 1 995 9 1 0 2 1 1 7 4 3 5 7 5 0 0 1 2] - [ 2 18 0 4 8 963 4 10 0 2 5 14 4 13 0 2 5 3 1 7 5] - [ 5 4 13 1 1 5 1035 5 1 0 3 2 2 2 1 5 1 5 1 4 3] - [ 2 9 6 0 1 29 5 930 1 2 3 8 4 2 0 0 1 2 8 8 3] - [ 21 2 0 2 0 0 0 4 910 34 5 4 1 8 8 2 2 1 4 0 1] - [ 88 0 5 0 6 1 0 2 39 840 0 0 0 16 4 1 0 4 0 2 4] - [ 1 4 3 6 1 3 4 1 21 1 971 3 3 10 4 0 2 1 7 1 4] - [ 1 3 0 0 5 8 1 4 1 2 0 923 23 5 3 4 2 11 4 3 2] - [ 1 1 1 5 3 3 0 2 1 1 1 27 871 1 0 5 4 22 2 0 8] - [ 4 2 4 0 7 11 0 2 15 13 6 6 2 931 1 8 5 1 0 3 3] - [ 11 4 3 20 5 3 0 0 27 7 1 4 2 0 979 2 3 6 6 1 8] - [ 3 1 4 2 4 1 6 0 0 0 0 8 8 2 0 974 5 18 0 5 5] - [ 1 6 1 1 11 4 0 2 1 0 0 0 2 3 4 12 997 2 1 8 5] - [ 3 4 0 3 0 1 3 1 1 1 0 9 12 1 0 9 0 998 0 1 4] - [ 5 5 3 14 4 2 0 39 2 0 5 1 4 0 12 1 2 2 977 4 4] - [ 0 4 2 0 1 7 4 8 0 0 1 21 3 4 0 9 4 4 1 1064 11] - [ 173 294 202 132 168 238 96 193 94 86 183 134 328 296 126 132 278 133 131 297 9720]] - -2023-02-13 18:07:38,219 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:07:38,219 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:07:38,225 - - -2023-02-13 18:07:38,225 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:07:39,100 - Epoch: [112][ 10/ 1207] Overall Loss 0.256608 Objective Loss 0.256608 LR 0.000500 Time 0.087434 -2023-02-13 18:07:39,295 - Epoch: [112][ 20/ 1207] Overall Loss 0.253471 Objective Loss 0.253471 LR 0.000500 Time 0.053486 -2023-02-13 18:07:39,487 - Epoch: [112][ 30/ 1207] Overall Loss 0.258314 Objective Loss 0.258314 LR 0.000500 Time 0.042013 -2023-02-13 18:07:39,678 - Epoch: [112][ 40/ 1207] Overall Loss 0.259678 Objective Loss 0.259678 LR 0.000500 Time 0.036286 -2023-02-13 18:07:39,868 - Epoch: [112][ 50/ 1207] Overall Loss 0.257444 Objective Loss 0.257444 LR 0.000500 Time 0.032821 -2023-02-13 18:07:40,059 - Epoch: [112][ 60/ 1207] Overall Loss 0.253728 Objective Loss 0.253728 LR 0.000500 Time 0.030531 -2023-02-13 18:07:40,249 - Epoch: [112][ 70/ 1207] Overall Loss 0.252248 Objective Loss 0.252248 LR 0.000500 Time 0.028871 -2023-02-13 18:07:40,440 - Epoch: [112][ 80/ 1207] Overall Loss 0.249844 Objective Loss 0.249844 LR 0.000500 Time 0.027645 -2023-02-13 18:07:40,629 - Epoch: [112][ 90/ 1207] Overall Loss 0.248883 Objective Loss 0.248883 LR 0.000500 Time 0.026671 -2023-02-13 18:07:40,819 - Epoch: [112][ 100/ 1207] Overall Loss 0.246196 Objective Loss 0.246196 LR 0.000500 Time 0.025904 -2023-02-13 18:07:41,013 - Epoch: [112][ 110/ 1207] Overall Loss 0.247611 Objective Loss 0.247611 LR 0.000500 Time 0.025304 -2023-02-13 18:07:41,203 - Epoch: [112][ 120/ 1207] Overall Loss 0.249349 Objective Loss 0.249349 LR 0.000500 Time 0.024780 -2023-02-13 18:07:41,393 - Epoch: [112][ 130/ 1207] Overall Loss 0.251744 Objective Loss 0.251744 LR 0.000500 Time 0.024335 -2023-02-13 18:07:41,583 - Epoch: [112][ 140/ 1207] Overall Loss 0.249996 Objective Loss 0.249996 LR 0.000500 Time 0.023949 -2023-02-13 18:07:41,775 - Epoch: [112][ 150/ 1207] Overall Loss 0.250796 Objective Loss 0.250796 LR 0.000500 Time 0.023628 -2023-02-13 18:07:41,966 - Epoch: [112][ 160/ 1207] Overall Loss 0.249829 Objective Loss 0.249829 LR 0.000500 Time 0.023342 -2023-02-13 18:07:42,158 - Epoch: [112][ 170/ 1207] Overall Loss 0.247826 Objective Loss 0.247826 LR 0.000500 Time 0.023096 -2023-02-13 18:07:42,348 - Epoch: [112][ 180/ 1207] Overall Loss 0.248930 Objective Loss 0.248930 LR 0.000500 Time 0.022869 -2023-02-13 18:07:42,539 - Epoch: [112][ 190/ 1207] Overall Loss 0.249251 Objective Loss 0.249251 LR 0.000500 Time 0.022667 -2023-02-13 18:07:42,729 - Epoch: [112][ 200/ 1207] Overall Loss 0.250062 Objective Loss 0.250062 LR 0.000500 Time 0.022482 -2023-02-13 18:07:42,920 - Epoch: [112][ 210/ 1207] Overall Loss 0.250600 Objective Loss 0.250600 LR 0.000500 Time 0.022320 -2023-02-13 18:07:43,111 - Epoch: [112][ 220/ 1207] Overall Loss 0.250157 Objective Loss 0.250157 LR 0.000500 Time 0.022171 -2023-02-13 18:07:43,303 - Epoch: [112][ 230/ 1207] Overall Loss 0.250291 Objective Loss 0.250291 LR 0.000500 Time 0.022040 -2023-02-13 18:07:43,494 - Epoch: [112][ 240/ 1207] Overall Loss 0.250289 Objective Loss 0.250289 LR 0.000500 Time 0.021917 -2023-02-13 18:07:43,686 - Epoch: [112][ 250/ 1207] Overall Loss 0.250291 Objective Loss 0.250291 LR 0.000500 Time 0.021806 -2023-02-13 18:07:43,876 - Epoch: [112][ 260/ 1207] Overall Loss 0.251172 Objective Loss 0.251172 LR 0.000500 Time 0.021698 -2023-02-13 18:07:44,068 - Epoch: [112][ 270/ 1207] Overall Loss 0.251958 Objective Loss 0.251958 LR 0.000500 Time 0.021602 -2023-02-13 18:07:44,259 - Epoch: [112][ 280/ 1207] Overall Loss 0.251962 Objective Loss 0.251962 LR 0.000500 Time 0.021511 -2023-02-13 18:07:44,447 - Epoch: [112][ 290/ 1207] Overall Loss 0.252550 Objective Loss 0.252550 LR 0.000500 Time 0.021418 -2023-02-13 18:07:44,635 - Epoch: [112][ 300/ 1207] Overall Loss 0.252964 Objective Loss 0.252964 LR 0.000500 Time 0.021330 -2023-02-13 18:07:44,827 - Epoch: [112][ 310/ 1207] Overall Loss 0.253027 Objective Loss 0.253027 LR 0.000500 Time 0.021259 -2023-02-13 18:07:45,017 - Epoch: [112][ 320/ 1207] Overall Loss 0.253174 Objective Loss 0.253174 LR 0.000500 Time 0.021188 -2023-02-13 18:07:45,209 - Epoch: [112][ 330/ 1207] Overall Loss 0.253174 Objective Loss 0.253174 LR 0.000500 Time 0.021125 -2023-02-13 18:07:45,402 - Epoch: [112][ 340/ 1207] Overall Loss 0.253188 Objective Loss 0.253188 LR 0.000500 Time 0.021070 -2023-02-13 18:07:45,598 - Epoch: [112][ 350/ 1207] Overall Loss 0.252486 Objective Loss 0.252486 LR 0.000500 Time 0.021027 -2023-02-13 18:07:45,794 - Epoch: [112][ 360/ 1207] Overall Loss 0.251986 Objective Loss 0.251986 LR 0.000500 Time 0.020986 -2023-02-13 18:07:45,991 - Epoch: [112][ 370/ 1207] Overall Loss 0.252897 Objective Loss 0.252897 LR 0.000500 Time 0.020950 -2023-02-13 18:07:46,186 - Epoch: [112][ 380/ 1207] Overall Loss 0.252480 Objective Loss 0.252480 LR 0.000500 Time 0.020911 -2023-02-13 18:07:46,382 - Epoch: [112][ 390/ 1207] Overall Loss 0.251963 Objective Loss 0.251963 LR 0.000500 Time 0.020878 -2023-02-13 18:07:46,578 - Epoch: [112][ 400/ 1207] Overall Loss 0.251636 Objective Loss 0.251636 LR 0.000500 Time 0.020845 -2023-02-13 18:07:46,775 - Epoch: [112][ 410/ 1207] Overall Loss 0.251479 Objective Loss 0.251479 LR 0.000500 Time 0.020815 -2023-02-13 18:07:46,968 - Epoch: [112][ 420/ 1207] Overall Loss 0.251658 Objective Loss 0.251658 LR 0.000500 Time 0.020780 -2023-02-13 18:07:47,157 - Epoch: [112][ 430/ 1207] Overall Loss 0.251782 Objective Loss 0.251782 LR 0.000500 Time 0.020735 -2023-02-13 18:07:47,347 - Epoch: [112][ 440/ 1207] Overall Loss 0.252556 Objective Loss 0.252556 LR 0.000500 Time 0.020695 -2023-02-13 18:07:47,538 - Epoch: [112][ 450/ 1207] Overall Loss 0.252220 Objective Loss 0.252220 LR 0.000500 Time 0.020657 -2023-02-13 18:07:47,727 - Epoch: [112][ 460/ 1207] Overall Loss 0.252639 Objective Loss 0.252639 LR 0.000500 Time 0.020619 -2023-02-13 18:07:47,917 - Epoch: [112][ 470/ 1207] Overall Loss 0.252546 Objective Loss 0.252546 LR 0.000500 Time 0.020584 -2023-02-13 18:07:48,107 - Epoch: [112][ 480/ 1207] Overall Loss 0.253098 Objective Loss 0.253098 LR 0.000500 Time 0.020550 -2023-02-13 18:07:48,297 - Epoch: [112][ 490/ 1207] Overall Loss 0.254178 Objective Loss 0.254178 LR 0.000500 Time 0.020517 -2023-02-13 18:07:48,486 - Epoch: [112][ 500/ 1207] Overall Loss 0.254370 Objective Loss 0.254370 LR 0.000500 Time 0.020484 -2023-02-13 18:07:48,677 - Epoch: [112][ 510/ 1207] Overall Loss 0.253838 Objective Loss 0.253838 LR 0.000500 Time 0.020456 -2023-02-13 18:07:48,866 - Epoch: [112][ 520/ 1207] Overall Loss 0.254025 Objective Loss 0.254025 LR 0.000500 Time 0.020424 -2023-02-13 18:07:49,054 - Epoch: [112][ 530/ 1207] Overall Loss 0.254080 Objective Loss 0.254080 LR 0.000500 Time 0.020394 -2023-02-13 18:07:49,243 - Epoch: [112][ 540/ 1207] Overall Loss 0.253670 Objective Loss 0.253670 LR 0.000500 Time 0.020365 -2023-02-13 18:07:49,432 - Epoch: [112][ 550/ 1207] Overall Loss 0.253815 Objective Loss 0.253815 LR 0.000500 Time 0.020339 -2023-02-13 18:07:49,622 - Epoch: [112][ 560/ 1207] Overall Loss 0.253745 Objective Loss 0.253745 LR 0.000500 Time 0.020314 -2023-02-13 18:07:49,811 - Epoch: [112][ 570/ 1207] Overall Loss 0.253466 Objective Loss 0.253466 LR 0.000500 Time 0.020288 -2023-02-13 18:07:50,000 - Epoch: [112][ 580/ 1207] Overall Loss 0.253905 Objective Loss 0.253905 LR 0.000500 Time 0.020264 -2023-02-13 18:07:50,189 - Epoch: [112][ 590/ 1207] Overall Loss 0.254310 Objective Loss 0.254310 LR 0.000500 Time 0.020240 -2023-02-13 18:07:50,380 - Epoch: [112][ 600/ 1207] Overall Loss 0.254600 Objective Loss 0.254600 LR 0.000500 Time 0.020219 -2023-02-13 18:07:50,570 - Epoch: [112][ 610/ 1207] Overall Loss 0.255044 Objective Loss 0.255044 LR 0.000500 Time 0.020200 -2023-02-13 18:07:50,760 - Epoch: [112][ 620/ 1207] Overall Loss 0.255669 Objective Loss 0.255669 LR 0.000500 Time 0.020180 -2023-02-13 18:07:50,950 - Epoch: [112][ 630/ 1207] Overall Loss 0.255496 Objective Loss 0.255496 LR 0.000500 Time 0.020161 -2023-02-13 18:07:51,139 - Epoch: [112][ 640/ 1207] Overall Loss 0.255369 Objective Loss 0.255369 LR 0.000500 Time 0.020140 -2023-02-13 18:07:51,329 - Epoch: [112][ 650/ 1207] Overall Loss 0.255697 Objective Loss 0.255697 LR 0.000500 Time 0.020121 -2023-02-13 18:07:51,521 - Epoch: [112][ 660/ 1207] Overall Loss 0.255465 Objective Loss 0.255465 LR 0.000500 Time 0.020107 -2023-02-13 18:07:51,710 - Epoch: [112][ 670/ 1207] Overall Loss 0.255411 Objective Loss 0.255411 LR 0.000500 Time 0.020088 -2023-02-13 18:07:51,899 - Epoch: [112][ 680/ 1207] Overall Loss 0.255570 Objective Loss 0.255570 LR 0.000500 Time 0.020071 -2023-02-13 18:07:52,089 - Epoch: [112][ 690/ 1207] Overall Loss 0.255407 Objective Loss 0.255407 LR 0.000500 Time 0.020055 -2023-02-13 18:07:52,278 - Epoch: [112][ 700/ 1207] Overall Loss 0.255356 Objective Loss 0.255356 LR 0.000500 Time 0.020038 -2023-02-13 18:07:52,470 - Epoch: [112][ 710/ 1207] Overall Loss 0.255387 Objective Loss 0.255387 LR 0.000500 Time 0.020025 -2023-02-13 18:07:52,658 - Epoch: [112][ 720/ 1207] Overall Loss 0.255196 Objective Loss 0.255196 LR 0.000500 Time 0.020008 -2023-02-13 18:07:52,849 - Epoch: [112][ 730/ 1207] Overall Loss 0.255356 Objective Loss 0.255356 LR 0.000500 Time 0.019995 -2023-02-13 18:07:53,044 - Epoch: [112][ 740/ 1207] Overall Loss 0.255101 Objective Loss 0.255101 LR 0.000500 Time 0.019987 -2023-02-13 18:07:53,240 - Epoch: [112][ 750/ 1207] Overall Loss 0.255376 Objective Loss 0.255376 LR 0.000500 Time 0.019982 -2023-02-13 18:07:53,436 - Epoch: [112][ 760/ 1207] Overall Loss 0.255276 Objective Loss 0.255276 LR 0.000500 Time 0.019977 -2023-02-13 18:07:53,626 - Epoch: [112][ 770/ 1207] Overall Loss 0.255140 Objective Loss 0.255140 LR 0.000500 Time 0.019963 -2023-02-13 18:07:53,816 - Epoch: [112][ 780/ 1207] Overall Loss 0.255175 Objective Loss 0.255175 LR 0.000500 Time 0.019950 -2023-02-13 18:07:54,005 - Epoch: [112][ 790/ 1207] Overall Loss 0.254810 Objective Loss 0.254810 LR 0.000500 Time 0.019936 -2023-02-13 18:07:54,194 - Epoch: [112][ 800/ 1207] Overall Loss 0.254891 Objective Loss 0.254891 LR 0.000500 Time 0.019923 -2023-02-13 18:07:54,383 - Epoch: [112][ 810/ 1207] Overall Loss 0.255423 Objective Loss 0.255423 LR 0.000500 Time 0.019910 -2023-02-13 18:07:54,573 - Epoch: [112][ 820/ 1207] Overall Loss 0.255330 Objective Loss 0.255330 LR 0.000500 Time 0.019898 -2023-02-13 18:07:54,762 - Epoch: [112][ 830/ 1207] Overall Loss 0.255685 Objective Loss 0.255685 LR 0.000500 Time 0.019886 -2023-02-13 18:07:54,951 - Epoch: [112][ 840/ 1207] Overall Loss 0.255526 Objective Loss 0.255526 LR 0.000500 Time 0.019874 -2023-02-13 18:07:55,140 - Epoch: [112][ 850/ 1207] Overall Loss 0.255683 Objective Loss 0.255683 LR 0.000500 Time 0.019862 -2023-02-13 18:07:55,329 - Epoch: [112][ 860/ 1207] Overall Loss 0.255783 Objective Loss 0.255783 LR 0.000500 Time 0.019850 -2023-02-13 18:07:55,519 - Epoch: [112][ 870/ 1207] Overall Loss 0.255673 Objective Loss 0.255673 LR 0.000500 Time 0.019840 -2023-02-13 18:07:55,708 - Epoch: [112][ 880/ 1207] Overall Loss 0.255498 Objective Loss 0.255498 LR 0.000500 Time 0.019829 -2023-02-13 18:07:55,899 - Epoch: [112][ 890/ 1207] Overall Loss 0.255648 Objective Loss 0.255648 LR 0.000500 Time 0.019820 -2023-02-13 18:07:56,088 - Epoch: [112][ 900/ 1207] Overall Loss 0.255816 Objective Loss 0.255816 LR 0.000500 Time 0.019809 -2023-02-13 18:07:56,277 - Epoch: [112][ 910/ 1207] Overall Loss 0.255857 Objective Loss 0.255857 LR 0.000500 Time 0.019799 -2023-02-13 18:07:56,467 - Epoch: [112][ 920/ 1207] Overall Loss 0.255885 Objective Loss 0.255885 LR 0.000500 Time 0.019790 -2023-02-13 18:07:56,656 - Epoch: [112][ 930/ 1207] Overall Loss 0.256143 Objective Loss 0.256143 LR 0.000500 Time 0.019780 -2023-02-13 18:07:56,846 - Epoch: [112][ 940/ 1207] Overall Loss 0.256102 Objective Loss 0.256102 LR 0.000500 Time 0.019771 -2023-02-13 18:07:57,035 - Epoch: [112][ 950/ 1207] Overall Loss 0.256117 Objective Loss 0.256117 LR 0.000500 Time 0.019762 -2023-02-13 18:07:57,224 - Epoch: [112][ 960/ 1207] Overall Loss 0.256042 Objective Loss 0.256042 LR 0.000500 Time 0.019753 -2023-02-13 18:07:57,413 - Epoch: [112][ 970/ 1207] Overall Loss 0.256143 Objective Loss 0.256143 LR 0.000500 Time 0.019744 -2023-02-13 18:07:57,602 - Epoch: [112][ 980/ 1207] Overall Loss 0.255945 Objective Loss 0.255945 LR 0.000500 Time 0.019735 -2023-02-13 18:07:57,792 - Epoch: [112][ 990/ 1207] Overall Loss 0.255949 Objective Loss 0.255949 LR 0.000500 Time 0.019727 -2023-02-13 18:07:57,981 - Epoch: [112][ 1000/ 1207] Overall Loss 0.255880 Objective Loss 0.255880 LR 0.000500 Time 0.019718 -2023-02-13 18:07:58,170 - Epoch: [112][ 1010/ 1207] Overall Loss 0.256190 Objective Loss 0.256190 LR 0.000500 Time 0.019709 -2023-02-13 18:07:58,358 - Epoch: [112][ 1020/ 1207] Overall Loss 0.256016 Objective Loss 0.256016 LR 0.000500 Time 0.019700 -2023-02-13 18:07:58,549 - Epoch: [112][ 1030/ 1207] Overall Loss 0.256094 Objective Loss 0.256094 LR 0.000500 Time 0.019694 -2023-02-13 18:07:58,739 - Epoch: [112][ 1040/ 1207] Overall Loss 0.255859 Objective Loss 0.255859 LR 0.000500 Time 0.019687 -2023-02-13 18:07:58,929 - Epoch: [112][ 1050/ 1207] Overall Loss 0.255713 Objective Loss 0.255713 LR 0.000500 Time 0.019680 -2023-02-13 18:07:59,119 - Epoch: [112][ 1060/ 1207] Overall Loss 0.255497 Objective Loss 0.255497 LR 0.000500 Time 0.019673 -2023-02-13 18:07:59,308 - Epoch: [112][ 1070/ 1207] Overall Loss 0.255465 Objective Loss 0.255465 LR 0.000500 Time 0.019666 -2023-02-13 18:07:59,498 - Epoch: [112][ 1080/ 1207] Overall Loss 0.255285 Objective Loss 0.255285 LR 0.000500 Time 0.019659 -2023-02-13 18:07:59,688 - Epoch: [112][ 1090/ 1207] Overall Loss 0.255309 Objective Loss 0.255309 LR 0.000500 Time 0.019652 -2023-02-13 18:07:59,877 - Epoch: [112][ 1100/ 1207] Overall Loss 0.255326 Objective Loss 0.255326 LR 0.000500 Time 0.019646 -2023-02-13 18:08:00,066 - Epoch: [112][ 1110/ 1207] Overall Loss 0.255300 Objective Loss 0.255300 LR 0.000500 Time 0.019639 -2023-02-13 18:08:00,256 - Epoch: [112][ 1120/ 1207] Overall Loss 0.255514 Objective Loss 0.255514 LR 0.000500 Time 0.019632 -2023-02-13 18:08:00,446 - Epoch: [112][ 1130/ 1207] Overall Loss 0.255482 Objective Loss 0.255482 LR 0.000500 Time 0.019626 -2023-02-13 18:08:00,636 - Epoch: [112][ 1140/ 1207] Overall Loss 0.255535 Objective Loss 0.255535 LR 0.000500 Time 0.019621 -2023-02-13 18:08:00,826 - Epoch: [112][ 1150/ 1207] Overall Loss 0.255278 Objective Loss 0.255278 LR 0.000500 Time 0.019615 -2023-02-13 18:08:01,017 - Epoch: [112][ 1160/ 1207] Overall Loss 0.255383 Objective Loss 0.255383 LR 0.000500 Time 0.019610 -2023-02-13 18:08:01,206 - Epoch: [112][ 1170/ 1207] Overall Loss 0.255439 Objective Loss 0.255439 LR 0.000500 Time 0.019604 -2023-02-13 18:08:01,396 - Epoch: [112][ 1180/ 1207] Overall Loss 0.255470 Objective Loss 0.255470 LR 0.000500 Time 0.019598 -2023-02-13 18:08:01,586 - Epoch: [112][ 1190/ 1207] Overall Loss 0.255592 Objective Loss 0.255592 LR 0.000500 Time 0.019593 -2023-02-13 18:08:01,827 - Epoch: [112][ 1200/ 1207] Overall Loss 0.255499 Objective Loss 0.255499 LR 0.000500 Time 0.019630 -2023-02-13 18:08:01,944 - Epoch: [112][ 1207/ 1207] Overall Loss 0.255414 Objective Loss 0.255414 Top1 86.280488 Top5 99.085366 LR 0.000500 Time 0.019613 -2023-02-13 18:08:02,023 - --- validate (epoch=112)----------- -2023-02-13 18:08:02,023 - 34311 samples (256 per mini-batch) -2023-02-13 18:08:02,544 - Epoch: [112][ 10/ 135] Loss 0.317793 Top1 83.046875 Top5 97.578125 -2023-02-13 18:08:02,682 - Epoch: [112][ 20/ 135] Loss 0.318657 Top1 83.710938 Top5 97.343750 -2023-02-13 18:08:02,807 - Epoch: [112][ 30/ 135] Loss 0.328996 Top1 83.307292 Top5 97.317708 -2023-02-13 18:08:02,929 - Epoch: [112][ 40/ 135] Loss 0.316764 Top1 83.623047 Top5 97.548828 -2023-02-13 18:08:03,050 - Epoch: [112][ 50/ 135] Loss 0.313809 Top1 83.796875 Top5 97.640625 -2023-02-13 18:08:03,175 - Epoch: [112][ 60/ 135] Loss 0.316459 Top1 83.717448 Top5 97.610677 -2023-02-13 18:08:03,302 - Epoch: [112][ 70/ 135] Loss 0.317022 Top1 83.727679 Top5 97.555804 -2023-02-13 18:08:03,429 - Epoch: [112][ 80/ 135] Loss 0.317241 Top1 83.710938 Top5 97.495117 -2023-02-13 18:08:03,557 - Epoch: [112][ 90/ 135] Loss 0.321493 Top1 83.611111 Top5 97.491319 -2023-02-13 18:08:03,684 - Epoch: [112][ 100/ 135] Loss 0.321093 Top1 83.578125 Top5 97.503906 -2023-02-13 18:08:03,812 - Epoch: [112][ 110/ 135] Loss 0.320640 Top1 83.636364 Top5 97.514205 -2023-02-13 18:08:03,939 - Epoch: [112][ 120/ 135] Loss 0.320776 Top1 83.736979 Top5 97.516276 -2023-02-13 18:08:04,071 - Epoch: [112][ 130/ 135] Loss 0.323738 Top1 83.653846 Top5 97.536058 -2023-02-13 18:08:04,117 - Epoch: [112][ 135/ 135] Loss 0.322633 Top1 83.664131 Top5 97.525575 -2023-02-13 18:08:04,186 - ==> Top1: 83.664 Top5: 97.526 Loss: 0.323 - -2023-02-13 18:08:04,187 - ==> Confusion: -[[ 847 6 3 0 23 3 0 0 3 48 2 5 1 4 6 3 2 1 1 1 8] - [ 1 959 2 1 12 19 3 13 4 1 1 0 1 0 0 2 6 1 2 0 5] - [ 5 4 942 16 4 1 20 16 0 3 2 3 2 4 3 5 2 3 8 6 9] - [ 3 3 11 908 3 4 0 2 2 2 16 1 5 1 16 1 4 7 21 0 6] - [ 7 10 0 0 1005 7 1 3 2 3 0 6 1 3 4 5 4 0 0 2 3] - [ 1 24 1 3 6 962 4 14 3 3 2 14 2 12 3 2 4 1 1 5 3] - [ 3 4 12 1 0 7 1032 4 0 2 4 3 2 2 1 5 0 5 2 5 5] - [ 0 19 7 0 1 24 5 912 1 1 3 3 6 2 0 1 0 1 22 13 3] - [ 14 3 0 1 1 0 0 2 891 39 11 3 0 12 19 2 2 1 6 0 2] - [ 71 3 2 0 9 2 0 2 41 845 0 1 0 21 7 1 1 3 0 2 1] - [ 3 3 1 7 1 2 3 2 11 1 983 4 1 8 5 1 1 0 10 0 4] - [ 2 5 2 0 2 10 1 4 1 0 1 917 25 5 0 6 1 13 4 6 0] - [ 1 0 0 6 3 6 0 1 1 0 0 28 867 0 2 6 3 23 2 3 7] - [ 2 3 0 2 9 12 2 2 9 13 7 8 1 931 3 7 2 2 0 2 7] - [ 3 3 1 15 8 4 0 2 20 6 6 3 2 2 994 0 0 9 9 0 5] - [ 4 3 5 1 10 0 3 0 0 0 0 10 10 5 0 955 10 16 1 6 7] - [ 0 8 2 0 11 2 0 0 1 0 0 2 0 2 4 6 1008 1 1 4 9] - [ 4 4 0 6 0 1 4 0 0 0 1 4 13 0 1 14 0 992 0 2 5] - [ 5 10 4 8 2 0 2 23 4 1 5 1 4 0 13 0 1 1 998 3 1] - [ 0 6 1 0 1 6 8 8 1 0 0 23 4 4 0 5 6 4 1 1066 4] - [ 137 321 207 128 172 201 83 165 99 84 236 130 321 270 174 92 322 134 183 283 9692]] - -2023-02-13 18:08:04,188 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:08:04,188 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:08:04,194 - - -2023-02-13 18:08:04,194 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:08:05,073 - Epoch: [113][ 10/ 1207] Overall Loss 0.271043 Objective Loss 0.271043 LR 0.000500 Time 0.087863 -2023-02-13 18:08:05,271 - Epoch: [113][ 20/ 1207] Overall Loss 0.277029 Objective Loss 0.277029 LR 0.000500 Time 0.053789 -2023-02-13 18:08:05,466 - Epoch: [113][ 30/ 1207] Overall Loss 0.263361 Objective Loss 0.263361 LR 0.000500 Time 0.042354 -2023-02-13 18:08:05,658 - Epoch: [113][ 40/ 1207] Overall Loss 0.263827 Objective Loss 0.263827 LR 0.000500 Time 0.036556 -2023-02-13 18:08:05,852 - Epoch: [113][ 50/ 1207] Overall Loss 0.266044 Objective Loss 0.266044 LR 0.000500 Time 0.033121 -2023-02-13 18:08:06,045 - Epoch: [113][ 60/ 1207] Overall Loss 0.263332 Objective Loss 0.263332 LR 0.000500 Time 0.030800 -2023-02-13 18:08:06,239 - Epoch: [113][ 70/ 1207] Overall Loss 0.260785 Objective Loss 0.260785 LR 0.000500 Time 0.029173 -2023-02-13 18:08:06,431 - Epoch: [113][ 80/ 1207] Overall Loss 0.261164 Objective Loss 0.261164 LR 0.000500 Time 0.027918 -2023-02-13 18:08:06,625 - Epoch: [113][ 90/ 1207] Overall Loss 0.259958 Objective Loss 0.259958 LR 0.000500 Time 0.026973 -2023-02-13 18:08:06,817 - Epoch: [113][ 100/ 1207] Overall Loss 0.257875 Objective Loss 0.257875 LR 0.000500 Time 0.026190 -2023-02-13 18:08:07,011 - Epoch: [113][ 110/ 1207] Overall Loss 0.258737 Objective Loss 0.258737 LR 0.000500 Time 0.025567 -2023-02-13 18:08:07,203 - Epoch: [113][ 120/ 1207] Overall Loss 0.259524 Objective Loss 0.259524 LR 0.000500 Time 0.025033 -2023-02-13 18:08:07,396 - Epoch: [113][ 130/ 1207] Overall Loss 0.259485 Objective Loss 0.259485 LR 0.000500 Time 0.024592 -2023-02-13 18:08:07,588 - Epoch: [113][ 140/ 1207] Overall Loss 0.260412 Objective Loss 0.260412 LR 0.000500 Time 0.024207 -2023-02-13 18:08:07,782 - Epoch: [113][ 150/ 1207] Overall Loss 0.259655 Objective Loss 0.259655 LR 0.000500 Time 0.023881 -2023-02-13 18:08:07,973 - Epoch: [113][ 160/ 1207] Overall Loss 0.260482 Objective Loss 0.260482 LR 0.000500 Time 0.023580 -2023-02-13 18:08:08,166 - Epoch: [113][ 170/ 1207] Overall Loss 0.258980 Objective Loss 0.258980 LR 0.000500 Time 0.023326 -2023-02-13 18:08:08,357 - Epoch: [113][ 180/ 1207] Overall Loss 0.257589 Objective Loss 0.257589 LR 0.000500 Time 0.023090 -2023-02-13 18:08:08,551 - Epoch: [113][ 190/ 1207] Overall Loss 0.256959 Objective Loss 0.256959 LR 0.000500 Time 0.022895 -2023-02-13 18:08:08,743 - Epoch: [113][ 200/ 1207] Overall Loss 0.256622 Objective Loss 0.256622 LR 0.000500 Time 0.022707 -2023-02-13 18:08:08,936 - Epoch: [113][ 210/ 1207] Overall Loss 0.255478 Objective Loss 0.255478 LR 0.000500 Time 0.022544 -2023-02-13 18:08:09,128 - Epoch: [113][ 220/ 1207] Overall Loss 0.256228 Objective Loss 0.256228 LR 0.000500 Time 0.022388 -2023-02-13 18:08:09,322 - Epoch: [113][ 230/ 1207] Overall Loss 0.255341 Objective Loss 0.255341 LR 0.000500 Time 0.022260 -2023-02-13 18:08:09,515 - Epoch: [113][ 240/ 1207] Overall Loss 0.255943 Objective Loss 0.255943 LR 0.000500 Time 0.022133 -2023-02-13 18:08:09,709 - Epoch: [113][ 250/ 1207] Overall Loss 0.255776 Objective Loss 0.255776 LR 0.000500 Time 0.022023 -2023-02-13 18:08:09,901 - Epoch: [113][ 260/ 1207] Overall Loss 0.254506 Objective Loss 0.254506 LR 0.000500 Time 0.021913 -2023-02-13 18:08:10,095 - Epoch: [113][ 270/ 1207] Overall Loss 0.254418 Objective Loss 0.254418 LR 0.000500 Time 0.021817 -2023-02-13 18:08:10,287 - Epoch: [113][ 280/ 1207] Overall Loss 0.253458 Objective Loss 0.253458 LR 0.000500 Time 0.021721 -2023-02-13 18:08:10,482 - Epoch: [113][ 290/ 1207] Overall Loss 0.253492 Objective Loss 0.253492 LR 0.000500 Time 0.021644 -2023-02-13 18:08:10,674 - Epoch: [113][ 300/ 1207] Overall Loss 0.253128 Objective Loss 0.253128 LR 0.000500 Time 0.021563 -2023-02-13 18:08:10,869 - Epoch: [113][ 310/ 1207] Overall Loss 0.253432 Objective Loss 0.253432 LR 0.000500 Time 0.021494 -2023-02-13 18:08:11,061 - Epoch: [113][ 320/ 1207] Overall Loss 0.253304 Objective Loss 0.253304 LR 0.000500 Time 0.021422 -2023-02-13 18:08:11,254 - Epoch: [113][ 330/ 1207] Overall Loss 0.253130 Objective Loss 0.253130 LR 0.000500 Time 0.021358 -2023-02-13 18:08:11,447 - Epoch: [113][ 340/ 1207] Overall Loss 0.252029 Objective Loss 0.252029 LR 0.000500 Time 0.021294 -2023-02-13 18:08:11,641 - Epoch: [113][ 350/ 1207] Overall Loss 0.251638 Objective Loss 0.251638 LR 0.000500 Time 0.021240 -2023-02-13 18:08:11,834 - Epoch: [113][ 360/ 1207] Overall Loss 0.251060 Objective Loss 0.251060 LR 0.000500 Time 0.021183 -2023-02-13 18:08:12,027 - Epoch: [113][ 370/ 1207] Overall Loss 0.251689 Objective Loss 0.251689 LR 0.000500 Time 0.021134 -2023-02-13 18:08:12,220 - Epoch: [113][ 380/ 1207] Overall Loss 0.251927 Objective Loss 0.251927 LR 0.000500 Time 0.021084 -2023-02-13 18:08:12,414 - Epoch: [113][ 390/ 1207] Overall Loss 0.250836 Objective Loss 0.250836 LR 0.000500 Time 0.021039 -2023-02-13 18:08:12,607 - Epoch: [113][ 400/ 1207] Overall Loss 0.250938 Objective Loss 0.250938 LR 0.000500 Time 0.020995 -2023-02-13 18:08:12,801 - Epoch: [113][ 410/ 1207] Overall Loss 0.250835 Objective Loss 0.250835 LR 0.000500 Time 0.020955 -2023-02-13 18:08:12,993 - Epoch: [113][ 420/ 1207] Overall Loss 0.250882 Objective Loss 0.250882 LR 0.000500 Time 0.020912 -2023-02-13 18:08:13,186 - Epoch: [113][ 430/ 1207] Overall Loss 0.250736 Objective Loss 0.250736 LR 0.000500 Time 0.020875 -2023-02-13 18:08:13,378 - Epoch: [113][ 440/ 1207] Overall Loss 0.250884 Objective Loss 0.250884 LR 0.000500 Time 0.020836 -2023-02-13 18:08:13,573 - Epoch: [113][ 450/ 1207] Overall Loss 0.250627 Objective Loss 0.250627 LR 0.000500 Time 0.020804 -2023-02-13 18:08:13,765 - Epoch: [113][ 460/ 1207] Overall Loss 0.250595 Objective Loss 0.250595 LR 0.000500 Time 0.020769 -2023-02-13 18:08:13,959 - Epoch: [113][ 470/ 1207] Overall Loss 0.250692 Objective Loss 0.250692 LR 0.000500 Time 0.020739 -2023-02-13 18:08:14,151 - Epoch: [113][ 480/ 1207] Overall Loss 0.250909 Objective Loss 0.250909 LR 0.000500 Time 0.020706 -2023-02-13 18:08:14,345 - Epoch: [113][ 490/ 1207] Overall Loss 0.250472 Objective Loss 0.250472 LR 0.000500 Time 0.020679 -2023-02-13 18:08:14,538 - Epoch: [113][ 500/ 1207] Overall Loss 0.250807 Objective Loss 0.250807 LR 0.000500 Time 0.020651 -2023-02-13 18:08:14,732 - Epoch: [113][ 510/ 1207] Overall Loss 0.250566 Objective Loss 0.250566 LR 0.000500 Time 0.020626 -2023-02-13 18:08:14,924 - Epoch: [113][ 520/ 1207] Overall Loss 0.250860 Objective Loss 0.250860 LR 0.000500 Time 0.020598 -2023-02-13 18:08:15,118 - Epoch: [113][ 530/ 1207] Overall Loss 0.251057 Objective Loss 0.251057 LR 0.000500 Time 0.020575 -2023-02-13 18:08:15,311 - Epoch: [113][ 540/ 1207] Overall Loss 0.250597 Objective Loss 0.250597 LR 0.000500 Time 0.020549 -2023-02-13 18:08:15,504 - Epoch: [113][ 550/ 1207] Overall Loss 0.250700 Objective Loss 0.250700 LR 0.000500 Time 0.020527 -2023-02-13 18:08:15,697 - Epoch: [113][ 560/ 1207] Overall Loss 0.250444 Objective Loss 0.250444 LR 0.000500 Time 0.020504 -2023-02-13 18:08:15,892 - Epoch: [113][ 570/ 1207] Overall Loss 0.250477 Objective Loss 0.250477 LR 0.000500 Time 0.020486 -2023-02-13 18:08:16,084 - Epoch: [113][ 580/ 1207] Overall Loss 0.250665 Objective Loss 0.250665 LR 0.000500 Time 0.020462 -2023-02-13 18:08:16,278 - Epoch: [113][ 590/ 1207] Overall Loss 0.250610 Objective Loss 0.250610 LR 0.000500 Time 0.020444 -2023-02-13 18:08:16,470 - Epoch: [113][ 600/ 1207] Overall Loss 0.250792 Objective Loss 0.250792 LR 0.000500 Time 0.020422 -2023-02-13 18:08:16,664 - Epoch: [113][ 610/ 1207] Overall Loss 0.250775 Objective Loss 0.250775 LR 0.000500 Time 0.020406 -2023-02-13 18:08:16,857 - Epoch: [113][ 620/ 1207] Overall Loss 0.251256 Objective Loss 0.251256 LR 0.000500 Time 0.020387 -2023-02-13 18:08:17,051 - Epoch: [113][ 630/ 1207] Overall Loss 0.251540 Objective Loss 0.251540 LR 0.000500 Time 0.020371 -2023-02-13 18:08:17,243 - Epoch: [113][ 640/ 1207] Overall Loss 0.251965 Objective Loss 0.251965 LR 0.000500 Time 0.020352 -2023-02-13 18:08:17,438 - Epoch: [113][ 650/ 1207] Overall Loss 0.251858 Objective Loss 0.251858 LR 0.000500 Time 0.020338 -2023-02-13 18:08:17,631 - Epoch: [113][ 660/ 1207] Overall Loss 0.251454 Objective Loss 0.251454 LR 0.000500 Time 0.020321 -2023-02-13 18:08:17,826 - Epoch: [113][ 670/ 1207] Overall Loss 0.251148 Objective Loss 0.251148 LR 0.000500 Time 0.020310 -2023-02-13 18:08:18,018 - Epoch: [113][ 680/ 1207] Overall Loss 0.251256 Objective Loss 0.251256 LR 0.000500 Time 0.020293 -2023-02-13 18:08:18,213 - Epoch: [113][ 690/ 1207] Overall Loss 0.251492 Objective Loss 0.251492 LR 0.000500 Time 0.020280 -2023-02-13 18:08:18,405 - Epoch: [113][ 700/ 1207] Overall Loss 0.251436 Objective Loss 0.251436 LR 0.000500 Time 0.020264 -2023-02-13 18:08:18,599 - Epoch: [113][ 710/ 1207] Overall Loss 0.251796 Objective Loss 0.251796 LR 0.000500 Time 0.020252 -2023-02-13 18:08:18,792 - Epoch: [113][ 720/ 1207] Overall Loss 0.251655 Objective Loss 0.251655 LR 0.000500 Time 0.020238 -2023-02-13 18:08:18,986 - Epoch: [113][ 730/ 1207] Overall Loss 0.251135 Objective Loss 0.251135 LR 0.000500 Time 0.020226 -2023-02-13 18:08:19,178 - Epoch: [113][ 740/ 1207] Overall Loss 0.251247 Objective Loss 0.251247 LR 0.000500 Time 0.020212 -2023-02-13 18:08:19,372 - Epoch: [113][ 750/ 1207] Overall Loss 0.251181 Objective Loss 0.251181 LR 0.000500 Time 0.020200 -2023-02-13 18:08:19,565 - Epoch: [113][ 760/ 1207] Overall Loss 0.251132 Objective Loss 0.251132 LR 0.000500 Time 0.020188 -2023-02-13 18:08:19,759 - Epoch: [113][ 770/ 1207] Overall Loss 0.251101 Objective Loss 0.251101 LR 0.000500 Time 0.020178 -2023-02-13 18:08:19,952 - Epoch: [113][ 780/ 1207] Overall Loss 0.251260 Objective Loss 0.251260 LR 0.000500 Time 0.020165 -2023-02-13 18:08:20,146 - Epoch: [113][ 790/ 1207] Overall Loss 0.251726 Objective Loss 0.251726 LR 0.000500 Time 0.020155 -2023-02-13 18:08:20,338 - Epoch: [113][ 800/ 1207] Overall Loss 0.251735 Objective Loss 0.251735 LR 0.000500 Time 0.020143 -2023-02-13 18:08:20,532 - Epoch: [113][ 810/ 1207] Overall Loss 0.251587 Objective Loss 0.251587 LR 0.000500 Time 0.020134 -2023-02-13 18:08:20,725 - Epoch: [113][ 820/ 1207] Overall Loss 0.251564 Objective Loss 0.251564 LR 0.000500 Time 0.020123 -2023-02-13 18:08:20,921 - Epoch: [113][ 830/ 1207] Overall Loss 0.251799 Objective Loss 0.251799 LR 0.000500 Time 0.020116 -2023-02-13 18:08:21,113 - Epoch: [113][ 840/ 1207] Overall Loss 0.251704 Objective Loss 0.251704 LR 0.000500 Time 0.020105 -2023-02-13 18:08:21,306 - Epoch: [113][ 850/ 1207] Overall Loss 0.251571 Objective Loss 0.251571 LR 0.000500 Time 0.020095 -2023-02-13 18:08:21,499 - Epoch: [113][ 860/ 1207] Overall Loss 0.251433 Objective Loss 0.251433 LR 0.000500 Time 0.020085 -2023-02-13 18:08:21,693 - Epoch: [113][ 870/ 1207] Overall Loss 0.251688 Objective Loss 0.251688 LR 0.000500 Time 0.020077 -2023-02-13 18:08:21,885 - Epoch: [113][ 880/ 1207] Overall Loss 0.251661 Objective Loss 0.251661 LR 0.000500 Time 0.020067 -2023-02-13 18:08:22,079 - Epoch: [113][ 890/ 1207] Overall Loss 0.251673 Objective Loss 0.251673 LR 0.000500 Time 0.020059 -2023-02-13 18:08:22,271 - Epoch: [113][ 900/ 1207] Overall Loss 0.251599 Objective Loss 0.251599 LR 0.000500 Time 0.020049 -2023-02-13 18:08:22,465 - Epoch: [113][ 910/ 1207] Overall Loss 0.251789 Objective Loss 0.251789 LR 0.000500 Time 0.020041 -2023-02-13 18:08:22,658 - Epoch: [113][ 920/ 1207] Overall Loss 0.251931 Objective Loss 0.251931 LR 0.000500 Time 0.020033 -2023-02-13 18:08:22,853 - Epoch: [113][ 930/ 1207] Overall Loss 0.252281 Objective Loss 0.252281 LR 0.000500 Time 0.020026 -2023-02-13 18:08:23,044 - Epoch: [113][ 940/ 1207] Overall Loss 0.252426 Objective Loss 0.252426 LR 0.000500 Time 0.020017 -2023-02-13 18:08:23,238 - Epoch: [113][ 950/ 1207] Overall Loss 0.252576 Objective Loss 0.252576 LR 0.000500 Time 0.020010 -2023-02-13 18:08:23,430 - Epoch: [113][ 960/ 1207] Overall Loss 0.252922 Objective Loss 0.252922 LR 0.000500 Time 0.020001 -2023-02-13 18:08:23,625 - Epoch: [113][ 970/ 1207] Overall Loss 0.252896 Objective Loss 0.252896 LR 0.000500 Time 0.019995 -2023-02-13 18:08:23,818 - Epoch: [113][ 980/ 1207] Overall Loss 0.253065 Objective Loss 0.253065 LR 0.000500 Time 0.019987 -2023-02-13 18:08:24,012 - Epoch: [113][ 990/ 1207] Overall Loss 0.253183 Objective Loss 0.253183 LR 0.000500 Time 0.019981 -2023-02-13 18:08:24,204 - Epoch: [113][ 1000/ 1207] Overall Loss 0.253099 Objective Loss 0.253099 LR 0.000500 Time 0.019973 -2023-02-13 18:08:24,399 - Epoch: [113][ 1010/ 1207] Overall Loss 0.253172 Objective Loss 0.253172 LR 0.000500 Time 0.019968 -2023-02-13 18:08:24,593 - Epoch: [113][ 1020/ 1207] Overall Loss 0.253251 Objective Loss 0.253251 LR 0.000500 Time 0.019962 -2023-02-13 18:08:24,789 - Epoch: [113][ 1030/ 1207] Overall Loss 0.253376 Objective Loss 0.253376 LR 0.000500 Time 0.019958 -2023-02-13 18:08:24,983 - Epoch: [113][ 1040/ 1207] Overall Loss 0.253536 Objective Loss 0.253536 LR 0.000500 Time 0.019952 -2023-02-13 18:08:25,178 - Epoch: [113][ 1050/ 1207] Overall Loss 0.253221 Objective Loss 0.253221 LR 0.000500 Time 0.019948 -2023-02-13 18:08:25,373 - Epoch: [113][ 1060/ 1207] Overall Loss 0.253248 Objective Loss 0.253248 LR 0.000500 Time 0.019943 -2023-02-13 18:08:25,569 - Epoch: [113][ 1070/ 1207] Overall Loss 0.253620 Objective Loss 0.253620 LR 0.000500 Time 0.019940 -2023-02-13 18:08:25,764 - Epoch: [113][ 1080/ 1207] Overall Loss 0.253523 Objective Loss 0.253523 LR 0.000500 Time 0.019935 -2023-02-13 18:08:25,961 - Epoch: [113][ 1090/ 1207] Overall Loss 0.253240 Objective Loss 0.253240 LR 0.000500 Time 0.019933 -2023-02-13 18:08:26,156 - Epoch: [113][ 1100/ 1207] Overall Loss 0.253365 Objective Loss 0.253365 LR 0.000500 Time 0.019928 -2023-02-13 18:08:26,351 - Epoch: [113][ 1110/ 1207] Overall Loss 0.253573 Objective Loss 0.253573 LR 0.000500 Time 0.019924 -2023-02-13 18:08:26,546 - Epoch: [113][ 1120/ 1207] Overall Loss 0.253600 Objective Loss 0.253600 LR 0.000500 Time 0.019920 -2023-02-13 18:08:26,743 - Epoch: [113][ 1130/ 1207] Overall Loss 0.253694 Objective Loss 0.253694 LR 0.000500 Time 0.019918 -2023-02-13 18:08:26,939 - Epoch: [113][ 1140/ 1207] Overall Loss 0.253609 Objective Loss 0.253609 LR 0.000500 Time 0.019915 -2023-02-13 18:08:27,134 - Epoch: [113][ 1150/ 1207] Overall Loss 0.253607 Objective Loss 0.253607 LR 0.000500 Time 0.019911 -2023-02-13 18:08:27,329 - Epoch: [113][ 1160/ 1207] Overall Loss 0.253743 Objective Loss 0.253743 LR 0.000500 Time 0.019907 -2023-02-13 18:08:27,525 - Epoch: [113][ 1170/ 1207] Overall Loss 0.253754 Objective Loss 0.253754 LR 0.000500 Time 0.019904 -2023-02-13 18:08:27,720 - Epoch: [113][ 1180/ 1207] Overall Loss 0.253707 Objective Loss 0.253707 LR 0.000500 Time 0.019901 -2023-02-13 18:08:27,916 - Epoch: [113][ 1190/ 1207] Overall Loss 0.253794 Objective Loss 0.253794 LR 0.000500 Time 0.019898 -2023-02-13 18:08:28,161 - Epoch: [113][ 1200/ 1207] Overall Loss 0.253804 Objective Loss 0.253804 LR 0.000500 Time 0.019936 -2023-02-13 18:08:28,275 - Epoch: [113][ 1207/ 1207] Overall Loss 0.253881 Objective Loss 0.253881 Top1 86.585366 Top5 97.865854 LR 0.000500 Time 0.019914 -2023-02-13 18:08:28,347 - --- validate (epoch=113)----------- -2023-02-13 18:08:28,347 - 34311 samples (256 per mini-batch) -2023-02-13 18:08:28,745 - Epoch: [113][ 10/ 135] Loss 0.298267 Top1 83.710938 Top5 96.601562 -2023-02-13 18:08:28,876 - Epoch: [113][ 20/ 135] Loss 0.304034 Top1 83.183594 Top5 97.285156 -2023-02-13 18:08:29,002 - Epoch: [113][ 30/ 135] Loss 0.304808 Top1 83.085938 Top5 97.447917 -2023-02-13 18:08:29,128 - Epoch: [113][ 40/ 135] Loss 0.305768 Top1 82.910156 Top5 97.441406 -2023-02-13 18:08:29,254 - Epoch: [113][ 50/ 135] Loss 0.302527 Top1 83.117188 Top5 97.625000 -2023-02-13 18:08:29,384 - Epoch: [113][ 60/ 135] Loss 0.305418 Top1 83.125000 Top5 97.636719 -2023-02-13 18:08:29,510 - Epoch: [113][ 70/ 135] Loss 0.311910 Top1 83.052455 Top5 97.672991 -2023-02-13 18:08:29,639 - Epoch: [113][ 80/ 135] Loss 0.312455 Top1 83.144531 Top5 97.646484 -2023-02-13 18:08:29,767 - Epoch: [113][ 90/ 135] Loss 0.309032 Top1 83.272569 Top5 97.651910 -2023-02-13 18:08:29,896 - Epoch: [113][ 100/ 135] Loss 0.310589 Top1 83.273438 Top5 97.640625 -2023-02-13 18:08:30,023 - Epoch: [113][ 110/ 135] Loss 0.313519 Top1 83.235085 Top5 97.627841 -2023-02-13 18:08:30,154 - Epoch: [113][ 120/ 135] Loss 0.314608 Top1 83.212891 Top5 97.607422 -2023-02-13 18:08:30,282 - Epoch: [113][ 130/ 135] Loss 0.317506 Top1 83.128005 Top5 97.575120 -2023-02-13 18:08:30,328 - Epoch: [113][ 135/ 135] Loss 0.317606 Top1 83.122031 Top5 97.566378 -2023-02-13 18:08:30,399 - ==> Top1: 83.122 Top5: 97.566 Loss: 0.318 - -2023-02-13 18:08:30,400 - ==> Confusion: -[[ 851 4 8 3 11 5 0 0 3 44 1 8 3 5 0 5 6 2 1 0 7] - [ 1 962 1 2 10 15 3 20 3 2 1 1 2 0 1 0 3 0 4 0 2] - [ 8 6 951 8 3 2 20 17 0 1 3 2 3 2 4 10 1 2 7 5 3] - [ 6 3 19 895 3 4 1 2 1 1 10 1 10 0 19 1 5 7 22 1 5] - [ 11 13 0 1 993 6 1 2 2 3 1 5 1 3 6 5 6 0 1 3 3] - [ 2 41 1 3 5 936 4 19 0 2 1 16 5 17 3 1 4 0 2 3 5] - [ 4 5 16 0 1 3 1041 6 0 0 3 2 2 0 0 5 0 3 1 4 3] - [ 0 10 5 0 1 17 5 939 0 1 1 7 5 1 0 0 0 5 19 7 1] - [ 14 4 2 1 1 0 0 2 881 47 8 3 3 10 17 3 1 1 9 1 1] - [ 88 1 1 0 7 4 0 2 28 847 3 1 1 18 4 1 1 2 2 0 1] - [ 3 6 5 10 1 3 2 6 17 3 967 3 3 6 2 0 1 0 10 0 3] - [ 0 4 0 0 2 7 1 4 2 2 0 921 29 6 4 3 2 7 3 5 3] - [ 1 0 0 6 0 2 0 1 2 0 2 20 891 1 2 6 3 15 0 1 6] - [ 4 4 2 0 6 9 0 2 8 16 8 9 3 929 9 7 1 2 0 3 2] - [ 10 4 1 16 6 3 0 2 12 7 0 2 4 2 993 2 1 4 12 1 10] - [ 2 3 7 0 10 0 2 1 0 0 0 9 13 1 1 958 6 17 1 9 6] - [ 1 8 0 1 8 0 0 3 2 1 0 0 1 2 2 12 999 3 4 7 7] - [ 3 3 0 4 0 1 1 1 0 2 0 10 24 0 1 14 0 981 2 2 2] - [ 2 6 5 10 4 1 0 23 1 0 3 2 5 0 12 1 0 2 1005 3 1] - [ 1 5 2 0 1 4 8 9 0 0 0 21 4 1 0 5 6 3 1 1071 6] - [ 140 329 251 141 165 184 112 190 93 86 156 153 388 321 180 108 269 120 231 308 9509]] - -2023-02-13 18:08:30,402 - ==> Best [Top1: 84.104 Top5: 97.665 Sparsity:0.00 Params: 148928 on epoch: 105] -2023-02-13 18:08:30,402 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:08:30,408 - - -2023-02-13 18:08:30,408 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:08:31,384 - Epoch: [114][ 10/ 1207] Overall Loss 0.228033 Objective Loss 0.228033 LR 0.000500 Time 0.097519 -2023-02-13 18:08:31,585 - Epoch: [114][ 20/ 1207] Overall Loss 0.238629 Objective Loss 0.238629 LR 0.000500 Time 0.058807 -2023-02-13 18:08:31,779 - Epoch: [114][ 30/ 1207] Overall Loss 0.246352 Objective Loss 0.246352 LR 0.000500 Time 0.045676 -2023-02-13 18:08:31,974 - Epoch: [114][ 40/ 1207] Overall Loss 0.241657 Objective Loss 0.241657 LR 0.000500 Time 0.039113 -2023-02-13 18:08:32,168 - Epoch: [114][ 50/ 1207] Overall Loss 0.243156 Objective Loss 0.243156 LR 0.000500 Time 0.035159 -2023-02-13 18:08:32,362 - Epoch: [114][ 60/ 1207] Overall Loss 0.248055 Objective Loss 0.248055 LR 0.000500 Time 0.032523 -2023-02-13 18:08:32,556 - Epoch: [114][ 70/ 1207] Overall Loss 0.247453 Objective Loss 0.247453 LR 0.000500 Time 0.030652 -2023-02-13 18:08:32,748 - Epoch: [114][ 80/ 1207] Overall Loss 0.248545 Objective Loss 0.248545 LR 0.000500 Time 0.029215 -2023-02-13 18:08:32,940 - Epoch: [114][ 90/ 1207] Overall Loss 0.248697 Objective Loss 0.248697 LR 0.000500 Time 0.028090 -2023-02-13 18:08:33,130 - Epoch: [114][ 100/ 1207] Overall Loss 0.250402 Objective Loss 0.250402 LR 0.000500 Time 0.027185 -2023-02-13 18:08:33,321 - Epoch: [114][ 110/ 1207] Overall Loss 0.247420 Objective Loss 0.247420 LR 0.000500 Time 0.026439 -2023-02-13 18:08:33,511 - Epoch: [114][ 120/ 1207] Overall Loss 0.247544 Objective Loss 0.247544 LR 0.000500 Time 0.025822 -2023-02-13 18:08:33,702 - Epoch: [114][ 130/ 1207] Overall Loss 0.249917 Objective Loss 0.249917 LR 0.000500 Time 0.025300 -2023-02-13 18:08:33,893 - Epoch: [114][ 140/ 1207] Overall Loss 0.250863 Objective Loss 0.250863 LR 0.000500 Time 0.024851 -2023-02-13 18:08:34,082 - Epoch: [114][ 150/ 1207] Overall Loss 0.250102 Objective Loss 0.250102 LR 0.000500 Time 0.024456 -2023-02-13 18:08:34,273 - Epoch: [114][ 160/ 1207] Overall Loss 0.249565 Objective Loss 0.249565 LR 0.000500 Time 0.024118 -2023-02-13 18:08:34,464 - Epoch: [114][ 170/ 1207] Overall Loss 0.250012 Objective Loss 0.250012 LR 0.000500 Time 0.023819 -2023-02-13 18:08:34,656 - Epoch: [114][ 180/ 1207] Overall Loss 0.249264 Objective Loss 0.249264 LR 0.000500 Time 0.023562 -2023-02-13 18:08:34,846 - Epoch: [114][ 190/ 1207] Overall Loss 0.248862 Objective Loss 0.248862 LR 0.000500 Time 0.023319 -2023-02-13 18:08:35,037 - Epoch: [114][ 200/ 1207] Overall Loss 0.249960 Objective Loss 0.249960 LR 0.000500 Time 0.023106 -2023-02-13 18:08:35,227 - Epoch: [114][ 210/ 1207] Overall Loss 0.249618 Objective Loss 0.249618 LR 0.000500 Time 0.022909 -2023-02-13 18:08:35,418 - Epoch: [114][ 220/ 1207] Overall Loss 0.248610 Objective Loss 0.248610 LR 0.000500 Time 0.022735 -2023-02-13 18:08:35,609 - Epoch: [114][ 230/ 1207] Overall Loss 0.248895 Objective Loss 0.248895 LR 0.000500 Time 0.022577 -2023-02-13 18:08:35,803 - Epoch: [114][ 240/ 1207] Overall Loss 0.249105 Objective Loss 0.249105 LR 0.000500 Time 0.022440 -2023-02-13 18:08:35,995 - Epoch: [114][ 250/ 1207] Overall Loss 0.249058 Objective Loss 0.249058 LR 0.000500 Time 0.022308 -2023-02-13 18:08:36,186 - Epoch: [114][ 260/ 1207] Overall Loss 0.248836 Objective Loss 0.248836 LR 0.000500 Time 0.022185 -2023-02-13 18:08:36,376 - Epoch: [114][ 270/ 1207] Overall Loss 0.249426 Objective Loss 0.249426 LR 0.000500 Time 0.022065 -2023-02-13 18:08:36,567 - Epoch: [114][ 280/ 1207] Overall Loss 0.248788 Objective Loss 0.248788 LR 0.000500 Time 0.021959 -2023-02-13 18:08:36,756 - Epoch: [114][ 290/ 1207] Overall Loss 0.248863 Objective Loss 0.248863 LR 0.000500 Time 0.021853 -2023-02-13 18:08:36,945 - Epoch: [114][ 300/ 1207] Overall Loss 0.248411 Objective Loss 0.248411 LR 0.000500 Time 0.021752 -2023-02-13 18:08:37,133 - Epoch: [114][ 310/ 1207] Overall Loss 0.247564 Objective Loss 0.247564 LR 0.000500 Time 0.021657 -2023-02-13 18:08:37,322 - Epoch: [114][ 320/ 1207] Overall Loss 0.246931 Objective Loss 0.246931 LR 0.000500 Time 0.021568 -2023-02-13 18:08:37,511 - Epoch: [114][ 330/ 1207] Overall Loss 0.247190 Objective Loss 0.247190 LR 0.000500 Time 0.021485 -2023-02-13 18:08:37,700 - Epoch: [114][ 340/ 1207] Overall Loss 0.247860 Objective Loss 0.247860 LR 0.000500 Time 0.021409 -2023-02-13 18:08:37,889 - Epoch: [114][ 350/ 1207] Overall Loss 0.249118 Objective Loss 0.249118 LR 0.000500 Time 0.021337 -2023-02-13 18:08:38,078 - Epoch: [114][ 360/ 1207] Overall Loss 0.249535 Objective Loss 0.249535 LR 0.000500 Time 0.021267 -2023-02-13 18:08:38,267 - Epoch: [114][ 370/ 1207] Overall Loss 0.249566 Objective Loss 0.249566 LR 0.000500 Time 0.021203 -2023-02-13 18:08:38,455 - Epoch: [114][ 380/ 1207] Overall Loss 0.250214 Objective Loss 0.250214 LR 0.000500 Time 0.021139 -2023-02-13 18:08:38,645 - Epoch: [114][ 390/ 1207] Overall Loss 0.249712 Objective Loss 0.249712 LR 0.000500 Time 0.021082 -2023-02-13 18:08:38,834 - Epoch: [114][ 400/ 1207] Overall Loss 0.248450 Objective Loss 0.248450 LR 0.000500 Time 0.021026 -2023-02-13 18:08:39,023 - Epoch: [114][ 410/ 1207] Overall Loss 0.248675 Objective Loss 0.248675 LR 0.000500 Time 0.020973 -2023-02-13 18:08:39,211 - Epoch: [114][ 420/ 1207] Overall Loss 0.248442 Objective Loss 0.248442 LR 0.000500 Time 0.020922 -2023-02-13 18:08:39,400 - Epoch: [114][ 430/ 1207] Overall Loss 0.248591 Objective Loss 0.248591 LR 0.000500 Time 0.020874 -2023-02-13 18:08:39,589 - Epoch: [114][ 440/ 1207] Overall Loss 0.249083 Objective Loss 0.249083 LR 0.000500 Time 0.020829 -2023-02-13 18:08:39,779 - Epoch: [114][ 450/ 1207] Overall Loss 0.249172 Objective Loss 0.249172 LR 0.000500 Time 0.020786 -2023-02-13 18:08:39,968 - Epoch: [114][ 460/ 1207] Overall Loss 0.248776 Objective Loss 0.248776 LR 0.000500 Time 0.020744 -2023-02-13 18:08:40,157 - Epoch: [114][ 470/ 1207] Overall Loss 0.248878 Objective Loss 0.248878 LR 0.000500 Time 0.020705 -2023-02-13 18:08:40,346 - Epoch: [114][ 480/ 1207] Overall Loss 0.249080 Objective Loss 0.249080 LR 0.000500 Time 0.020667 -2023-02-13 18:08:40,535 - Epoch: [114][ 490/ 1207] Overall Loss 0.249086 Objective Loss 0.249086 LR 0.000500 Time 0.020630 -2023-02-13 18:08:40,726 - Epoch: [114][ 500/ 1207] Overall Loss 0.248759 Objective Loss 0.248759 LR 0.000500 Time 0.020597 -2023-02-13 18:08:40,916 - Epoch: [114][ 510/ 1207] Overall Loss 0.248662 Objective Loss 0.248662 LR 0.000500 Time 0.020566 -2023-02-13 18:08:41,105 - Epoch: [114][ 520/ 1207] Overall Loss 0.248659 Objective Loss 0.248659 LR 0.000500 Time 0.020534 -2023-02-13 18:08:41,295 - Epoch: [114][ 530/ 1207] Overall Loss 0.248853 Objective Loss 0.248853 LR 0.000500 Time 0.020503 -2023-02-13 18:08:41,483 - Epoch: [114][ 540/ 1207] Overall Loss 0.248651 Objective Loss 0.248651 LR 0.000500 Time 0.020472 -2023-02-13 18:08:41,674 - Epoch: [114][ 550/ 1207] Overall Loss 0.248876 Objective Loss 0.248876 LR 0.000500 Time 0.020446 -2023-02-13 18:08:41,865 - Epoch: [114][ 560/ 1207] Overall Loss 0.248991 Objective Loss 0.248991 LR 0.000500 Time 0.020421 -2023-02-13 18:08:42,055 - Epoch: [114][ 570/ 1207] Overall Loss 0.249218 Objective Loss 0.249218 LR 0.000500 Time 0.020395 -2023-02-13 18:08:42,244 - Epoch: [114][ 580/ 1207] Overall Loss 0.249165 Objective Loss 0.249165 LR 0.000500 Time 0.020368 -2023-02-13 18:08:42,434 - Epoch: [114][ 590/ 1207] Overall Loss 0.249079 Objective Loss 0.249079 LR 0.000500 Time 0.020345 -2023-02-13 18:08:42,624 - Epoch: [114][ 600/ 1207] Overall Loss 0.248814 Objective Loss 0.248814 LR 0.000500 Time 0.020321 -2023-02-13 18:08:42,813 - Epoch: [114][ 610/ 1207] Overall Loss 0.249061 Objective Loss 0.249061 LR 0.000500 Time 0.020299 -2023-02-13 18:08:43,003 - Epoch: [114][ 620/ 1207] Overall Loss 0.248731 Objective Loss 0.248731 LR 0.000500 Time 0.020276 -2023-02-13 18:08:43,192 - Epoch: [114][ 630/ 1207] Overall Loss 0.249100 Objective Loss 0.249100 LR 0.000500 Time 0.020254 -2023-02-13 18:08:43,382 - Epoch: [114][ 640/ 1207] Overall Loss 0.249018 Objective Loss 0.249018 LR 0.000500 Time 0.020234 -2023-02-13 18:08:43,572 - Epoch: [114][ 650/ 1207] Overall Loss 0.249483 Objective Loss 0.249483 LR 0.000500 Time 0.020214 -2023-02-13 18:08:43,761 - Epoch: [114][ 660/ 1207] Overall Loss 0.249738 Objective Loss 0.249738 LR 0.000500 Time 0.020195 -2023-02-13 18:08:43,951 - Epoch: [114][ 670/ 1207] Overall Loss 0.249247 Objective Loss 0.249247 LR 0.000500 Time 0.020176 -2023-02-13 18:08:44,141 - Epoch: [114][ 680/ 1207] Overall Loss 0.249430 Objective Loss 0.249430 LR 0.000500 Time 0.020157 -2023-02-13 18:08:44,330 - Epoch: [114][ 690/ 1207] Overall Loss 0.249687 Objective Loss 0.249687 LR 0.000500 Time 0.020139 -2023-02-13 18:08:44,519 - Epoch: [114][ 700/ 1207] Overall Loss 0.249988 Objective Loss 0.249988 LR 0.000500 Time 0.020121 -2023-02-13 18:08:44,709 - Epoch: [114][ 710/ 1207] Overall Loss 0.249869 Objective Loss 0.249869 LR 0.000500 Time 0.020105 -2023-02-13 18:08:44,899 - Epoch: [114][ 720/ 1207] Overall Loss 0.249578 Objective Loss 0.249578 LR 0.000500 Time 0.020089 -2023-02-13 18:08:45,089 - Epoch: [114][ 730/ 1207] Overall Loss 0.249821 Objective Loss 0.249821 LR 0.000500 Time 0.020073 -2023-02-13 18:08:45,278 - Epoch: [114][ 740/ 1207] Overall Loss 0.249674 Objective Loss 0.249674 LR 0.000500 Time 0.020057 -2023-02-13 18:08:45,467 - Epoch: [114][ 750/ 1207] Overall Loss 0.249466 Objective Loss 0.249466 LR 0.000500 Time 0.020041 -2023-02-13 18:08:45,657 - Epoch: [114][ 760/ 1207] Overall Loss 0.249015 Objective Loss 0.249015 LR 0.000500 Time 0.020027 -2023-02-13 18:08:45,847 - Epoch: [114][ 770/ 1207] Overall Loss 0.249377 Objective Loss 0.249377 LR 0.000500 Time 0.020013 -2023-02-13 18:08:46,037 - Epoch: [114][ 780/ 1207] Overall Loss 0.249537 Objective Loss 0.249537 LR 0.000500 Time 0.019999 -2023-02-13 18:08:46,226 - Epoch: [114][ 790/ 1207] Overall Loss 0.249781 Objective Loss 0.249781 LR 0.000500 Time 0.019985 -2023-02-13 18:08:46,415 - Epoch: [114][ 800/ 1207] Overall Loss 0.250162 Objective Loss 0.250162 LR 0.000500 Time 0.019970 -2023-02-13 18:08:46,605 - Epoch: [114][ 810/ 1207] Overall Loss 0.250184 Objective Loss 0.250184 LR 0.000500 Time 0.019958 -2023-02-13 18:08:46,794 - Epoch: [114][ 820/ 1207] Overall Loss 0.250144 Objective Loss 0.250144 LR 0.000500 Time 0.019945 -2023-02-13 18:08:46,984 - Epoch: [114][ 830/ 1207] Overall Loss 0.250146 Objective Loss 0.250146 LR 0.000500 Time 0.019934 -2023-02-13 18:08:47,174 - Epoch: [114][ 840/ 1207] Overall Loss 0.250297 Objective Loss 0.250297 LR 0.000500 Time 0.019921 -2023-02-13 18:08:47,363 - Epoch: [114][ 850/ 1207] Overall Loss 0.250644 Objective Loss 0.250644 LR 0.000500 Time 0.019910 -2023-02-13 18:08:47,552 - Epoch: [114][ 860/ 1207] Overall Loss 0.250689 Objective Loss 0.250689 LR 0.000500 Time 0.019898 -2023-02-13 18:08:47,743 - Epoch: [114][ 870/ 1207] Overall Loss 0.250825 Objective Loss 0.250825 LR 0.000500 Time 0.019888 -2023-02-13 18:08:47,933 - Epoch: [114][ 880/ 1207] Overall Loss 0.250766 Objective Loss 0.250766 LR 0.000500 Time 0.019877 -2023-02-13 18:08:48,122 - Epoch: [114][ 890/ 1207] Overall Loss 0.251007 Objective Loss 0.251007 LR 0.000500 Time 0.019866 -2023-02-13 18:08:48,312 - Epoch: [114][ 900/ 1207] Overall Loss 0.251025 Objective Loss 0.251025 LR 0.000500 Time 0.019855 -2023-02-13 18:08:48,501 - Epoch: [114][ 910/ 1207] Overall Loss 0.251291 Objective Loss 0.251291 LR 0.000500 Time 0.019845 -2023-02-13 18:08:48,691 - Epoch: [114][ 920/ 1207] Overall Loss 0.251028 Objective Loss 0.251028 LR 0.000500 Time 0.019836 -2023-02-13 18:08:48,881 - Epoch: [114][ 930/ 1207] Overall Loss 0.251277 Objective Loss 0.251277 LR 0.000500 Time 0.019826 -2023-02-13 18:08:49,070 - Epoch: [114][ 940/ 1207] Overall Loss 0.251199 Objective Loss 0.251199 LR 0.000500 Time 0.019815 -2023-02-13 18:08:49,259 - Epoch: [114][ 950/ 1207] Overall Loss 0.251404 Objective Loss 0.251404 LR 0.000500 Time 0.019806 -2023-02-13 18:08:49,448 - Epoch: [114][ 960/ 1207] Overall Loss 0.251486 Objective Loss 0.251486 LR 0.000500 Time 0.019796 -2023-02-13 18:08:49,638 - Epoch: [114][ 970/ 1207] Overall Loss 0.251526 Objective Loss 0.251526 LR 0.000500 Time 0.019787 -2023-02-13 18:08:49,828 - Epoch: [114][ 980/ 1207] Overall Loss 0.251655 Objective Loss 0.251655 LR 0.000500 Time 0.019778 -2023-02-13 18:08:50,017 - Epoch: [114][ 990/ 1207] Overall Loss 0.251646 Objective Loss 0.251646 LR 0.000500 Time 0.019770 -2023-02-13 18:08:50,206 - Epoch: [114][ 1000/ 1207] Overall Loss 0.251754 Objective Loss 0.251754 LR 0.000500 Time 0.019761 -2023-02-13 18:08:50,396 - Epoch: [114][ 1010/ 1207] Overall Loss 0.251592 Objective Loss 0.251592 LR 0.000500 Time 0.019752 -2023-02-13 18:08:50,586 - Epoch: [114][ 1020/ 1207] Overall Loss 0.251700 Objective Loss 0.251700 LR 0.000500 Time 0.019745 -2023-02-13 18:08:50,777 - Epoch: [114][ 1030/ 1207] Overall Loss 0.251807 Objective Loss 0.251807 LR 0.000500 Time 0.019738 -2023-02-13 18:08:50,968 - Epoch: [114][ 1040/ 1207] Overall Loss 0.251643 Objective Loss 0.251643 LR 0.000500 Time 0.019731 -2023-02-13 18:08:51,157 - Epoch: [114][ 1050/ 1207] Overall Loss 0.251575 Objective Loss 0.251575 LR 0.000500 Time 0.019724 -2023-02-13 18:08:51,347 - Epoch: [114][ 1060/ 1207] Overall Loss 0.251808 Objective Loss 0.251808 LR 0.000500 Time 0.019716 -2023-02-13 18:08:51,537 - Epoch: [114][ 1070/ 1207] Overall Loss 0.251580 Objective Loss 0.251580 LR 0.000500 Time 0.019709 -2023-02-13 18:08:51,727 - Epoch: [114][ 1080/ 1207] Overall Loss 0.251624 Objective Loss 0.251624 LR 0.000500 Time 0.019702 -2023-02-13 18:08:51,917 - Epoch: [114][ 1090/ 1207] Overall Loss 0.251687 Objective Loss 0.251687 LR 0.000500 Time 0.019695 -2023-02-13 18:08:52,106 - Epoch: [114][ 1100/ 1207] Overall Loss 0.251610 Objective Loss 0.251610 LR 0.000500 Time 0.019688 -2023-02-13 18:08:52,295 - Epoch: [114][ 1110/ 1207] Overall Loss 0.251707 Objective Loss 0.251707 LR 0.000500 Time 0.019681 -2023-02-13 18:08:52,484 - Epoch: [114][ 1120/ 1207] Overall Loss 0.251588 Objective Loss 0.251588 LR 0.000500 Time 0.019673 -2023-02-13 18:08:52,674 - Epoch: [114][ 1130/ 1207] Overall Loss 0.251609 Objective Loss 0.251609 LR 0.000500 Time 0.019667 -2023-02-13 18:08:52,863 - Epoch: [114][ 1140/ 1207] Overall Loss 0.251385 Objective Loss 0.251385 LR 0.000500 Time 0.019659 -2023-02-13 18:08:53,052 - Epoch: [114][ 1150/ 1207] Overall Loss 0.251159 Objective Loss 0.251159 LR 0.000500 Time 0.019653 -2023-02-13 18:08:53,240 - Epoch: [114][ 1160/ 1207] Overall Loss 0.251255 Objective Loss 0.251255 LR 0.000500 Time 0.019646 -2023-02-13 18:08:53,430 - Epoch: [114][ 1170/ 1207] Overall Loss 0.251353 Objective Loss 0.251353 LR 0.000500 Time 0.019639 -2023-02-13 18:08:53,619 - Epoch: [114][ 1180/ 1207] Overall Loss 0.251338 Objective Loss 0.251338 LR 0.000500 Time 0.019633 -2023-02-13 18:08:53,809 - Epoch: [114][ 1190/ 1207] Overall Loss 0.251293 Objective Loss 0.251293 LR 0.000500 Time 0.019627 -2023-02-13 18:08:54,050 - Epoch: [114][ 1200/ 1207] Overall Loss 0.251315 Objective Loss 0.251315 LR 0.000500 Time 0.019664 -2023-02-13 18:08:54,166 - Epoch: [114][ 1207/ 1207] Overall Loss 0.251391 Objective Loss 0.251391 Top1 88.414634 Top5 99.390244 LR 0.000500 Time 0.019646 -2023-02-13 18:08:54,237 - --- validate (epoch=114)----------- -2023-02-13 18:08:54,237 - 34311 samples (256 per mini-batch) -2023-02-13 18:08:54,635 - Epoch: [114][ 10/ 135] Loss 0.321318 Top1 83.789062 Top5 97.929688 -2023-02-13 18:08:54,767 - Epoch: [114][ 20/ 135] Loss 0.335493 Top1 83.457031 Top5 97.773438 -2023-02-13 18:08:54,897 - Epoch: [114][ 30/ 135] Loss 0.327271 Top1 83.789062 Top5 97.773438 -2023-02-13 18:08:55,027 - Epoch: [114][ 40/ 135] Loss 0.335656 Top1 83.974609 Top5 97.519531 -2023-02-13 18:08:55,158 - Epoch: [114][ 50/ 135] Loss 0.329484 Top1 84.132812 Top5 97.539062 -2023-02-13 18:08:55,288 - Epoch: [114][ 60/ 135] Loss 0.321670 Top1 84.277344 Top5 97.656250 -2023-02-13 18:08:55,417 - Epoch: [114][ 70/ 135] Loss 0.318959 Top1 84.341518 Top5 97.678571 -2023-02-13 18:08:55,546 - Epoch: [114][ 80/ 135] Loss 0.319953 Top1 84.311523 Top5 97.705078 -2023-02-13 18:08:55,673 - Epoch: [114][ 90/ 135] Loss 0.317400 Top1 84.231771 Top5 97.738715 -2023-02-13 18:08:55,801 - Epoch: [114][ 100/ 135] Loss 0.317635 Top1 84.265625 Top5 97.769531 -2023-02-13 18:08:55,926 - Epoch: [114][ 110/ 135] Loss 0.316773 Top1 84.268466 Top5 97.727273 -2023-02-13 18:08:56,052 - Epoch: [114][ 120/ 135] Loss 0.317360 Top1 84.238281 Top5 97.737630 -2023-02-13 18:08:56,180 - Epoch: [114][ 130/ 135] Loss 0.315313 Top1 84.299880 Top5 97.791466 -2023-02-13 18:08:56,224 - Epoch: [114][ 135/ 135] Loss 0.312489 Top1 84.354872 Top5 97.811198 -2023-02-13 18:08:56,296 - ==> Top1: 84.355 Top5: 97.811 Loss: 0.312 - -2023-02-13 18:08:56,297 - ==> Confusion: -[[ 865 4 5 1 6 3 0 1 2 42 1 6 1 8 4 4 1 1 2 1 9] - [ 0 945 3 1 6 28 2 14 3 1 3 2 0 0 1 3 4 1 6 0 10] - [ 10 3 953 11 2 1 21 12 0 1 3 2 3 5 7 6 2 4 5 2 5] - [ 3 1 21 917 3 4 2 1 2 0 12 0 4 2 12 2 6 7 11 0 6] - [ 21 10 0 2 972 8 2 2 0 5 2 7 1 4 7 8 4 0 2 1 8] - [ 3 19 1 7 5 968 2 21 0 1 2 9 1 12 1 2 7 2 2 4 1] - [ 1 4 16 2 0 8 1039 7 1 1 2 0 3 1 0 3 0 3 2 5 1] - [ 0 8 11 3 4 26 1 934 0 1 1 5 5 1 0 0 2 1 12 6 3] - [ 16 4 1 1 0 0 1 4 901 36 11 2 1 9 15 3 0 0 4 0 0] - [ 70 2 3 1 2 5 0 3 40 855 1 1 0 15 6 3 0 2 1 0 2] - [ 2 5 3 10 1 3 4 4 10 0 983 3 1 7 2 0 1 0 9 0 3] - [ 3 5 0 0 3 11 0 6 1 0 1 926 15 3 2 7 1 10 2 5 4] - [ 0 0 0 7 0 4 0 2 2 1 2 26 868 1 3 6 3 21 1 0 12] - [ 3 5 0 0 5 12 0 1 12 14 15 4 2 922 5 6 6 2 1 1 8] - [ 9 2 4 25 1 4 0 1 22 5 7 0 4 2 978 2 1 5 8 1 11] - [ 2 1 7 0 7 2 3 0 0 0 1 8 6 1 2 966 8 18 0 6 8] - [ 1 6 2 1 8 1 0 2 1 1 0 0 2 3 2 10 1001 2 2 3 13] - [ 5 0 0 3 0 1 1 2 0 1 1 10 18 0 1 14 0 990 0 0 4] - [ 4 2 3 14 1 2 1 23 4 0 3 1 4 0 16 1 0 1 1000 2 4] - [ 0 3 2 1 3 12 9 12 1 0 1 14 2 1 0 3 3 5 1 1065 10] - [ 181 205 239 167 98 227 96 162 93 97 207 124 296 272 160 101 260 132 186 236 9895]] - -2023-02-13 18:08:56,298 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:08:56,298 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:08:56,305 - - -2023-02-13 18:08:56,305 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:08:57,294 - Epoch: [115][ 10/ 1207] Overall Loss 0.266930 Objective Loss 0.266930 LR 0.000500 Time 0.098821 -2023-02-13 18:08:57,491 - Epoch: [115][ 20/ 1207] Overall Loss 0.272089 Objective Loss 0.272089 LR 0.000500 Time 0.059241 -2023-02-13 18:08:57,681 - Epoch: [115][ 30/ 1207] Overall Loss 0.264602 Objective Loss 0.264602 LR 0.000500 Time 0.045804 -2023-02-13 18:08:57,869 - Epoch: [115][ 40/ 1207] Overall Loss 0.259132 Objective Loss 0.259132 LR 0.000500 Time 0.039046 -2023-02-13 18:08:58,057 - Epoch: [115][ 50/ 1207] Overall Loss 0.256205 Objective Loss 0.256205 LR 0.000500 Time 0.035004 -2023-02-13 18:08:58,245 - Epoch: [115][ 60/ 1207] Overall Loss 0.256848 Objective Loss 0.256848 LR 0.000500 Time 0.032286 -2023-02-13 18:08:58,433 - Epoch: [115][ 70/ 1207] Overall Loss 0.255117 Objective Loss 0.255117 LR 0.000500 Time 0.030359 -2023-02-13 18:08:58,621 - Epoch: [115][ 80/ 1207] Overall Loss 0.256671 Objective Loss 0.256671 LR 0.000500 Time 0.028907 -2023-02-13 18:08:58,809 - Epoch: [115][ 90/ 1207] Overall Loss 0.255504 Objective Loss 0.255504 LR 0.000500 Time 0.027784 -2023-02-13 18:08:58,997 - Epoch: [115][ 100/ 1207] Overall Loss 0.251880 Objective Loss 0.251880 LR 0.000500 Time 0.026875 -2023-02-13 18:08:59,185 - Epoch: [115][ 110/ 1207] Overall Loss 0.254470 Objective Loss 0.254470 LR 0.000500 Time 0.026140 -2023-02-13 18:08:59,372 - Epoch: [115][ 120/ 1207] Overall Loss 0.253739 Objective Loss 0.253739 LR 0.000500 Time 0.025516 -2023-02-13 18:08:59,561 - Epoch: [115][ 130/ 1207] Overall Loss 0.252933 Objective Loss 0.252933 LR 0.000500 Time 0.025006 -2023-02-13 18:08:59,750 - Epoch: [115][ 140/ 1207] Overall Loss 0.251811 Objective Loss 0.251811 LR 0.000500 Time 0.024570 -2023-02-13 18:08:59,940 - Epoch: [115][ 150/ 1207] Overall Loss 0.251646 Objective Loss 0.251646 LR 0.000500 Time 0.024193 -2023-02-13 18:09:00,128 - Epoch: [115][ 160/ 1207] Overall Loss 0.250985 Objective Loss 0.250985 LR 0.000500 Time 0.023857 -2023-02-13 18:09:00,318 - Epoch: [115][ 170/ 1207] Overall Loss 0.250700 Objective Loss 0.250700 LR 0.000500 Time 0.023565 -2023-02-13 18:09:00,506 - Epoch: [115][ 180/ 1207] Overall Loss 0.249871 Objective Loss 0.249871 LR 0.000500 Time 0.023300 -2023-02-13 18:09:00,696 - Epoch: [115][ 190/ 1207] Overall Loss 0.248829 Objective Loss 0.248829 LR 0.000500 Time 0.023069 -2023-02-13 18:09:00,885 - Epoch: [115][ 200/ 1207] Overall Loss 0.248456 Objective Loss 0.248456 LR 0.000500 Time 0.022861 -2023-02-13 18:09:01,074 - Epoch: [115][ 210/ 1207] Overall Loss 0.247951 Objective Loss 0.247951 LR 0.000500 Time 0.022672 -2023-02-13 18:09:01,263 - Epoch: [115][ 220/ 1207] Overall Loss 0.247645 Objective Loss 0.247645 LR 0.000500 Time 0.022495 -2023-02-13 18:09:01,452 - Epoch: [115][ 230/ 1207] Overall Loss 0.246611 Objective Loss 0.246611 LR 0.000500 Time 0.022339 -2023-02-13 18:09:01,641 - Epoch: [115][ 240/ 1207] Overall Loss 0.247070 Objective Loss 0.247070 LR 0.000500 Time 0.022196 -2023-02-13 18:09:01,831 - Epoch: [115][ 250/ 1207] Overall Loss 0.247865 Objective Loss 0.247865 LR 0.000500 Time 0.022064 -2023-02-13 18:09:02,020 - Epoch: [115][ 260/ 1207] Overall Loss 0.247913 Objective Loss 0.247913 LR 0.000500 Time 0.021943 -2023-02-13 18:09:02,209 - Epoch: [115][ 270/ 1207] Overall Loss 0.248567 Objective Loss 0.248567 LR 0.000500 Time 0.021829 -2023-02-13 18:09:02,397 - Epoch: [115][ 280/ 1207] Overall Loss 0.248693 Objective Loss 0.248693 LR 0.000500 Time 0.021720 -2023-02-13 18:09:02,587 - Epoch: [115][ 290/ 1207] Overall Loss 0.248986 Objective Loss 0.248986 LR 0.000500 Time 0.021623 -2023-02-13 18:09:02,776 - Epoch: [115][ 300/ 1207] Overall Loss 0.248456 Objective Loss 0.248456 LR 0.000500 Time 0.021532 -2023-02-13 18:09:02,965 - Epoch: [115][ 310/ 1207] Overall Loss 0.248932 Objective Loss 0.248932 LR 0.000500 Time 0.021446 -2023-02-13 18:09:03,154 - Epoch: [115][ 320/ 1207] Overall Loss 0.248640 Objective Loss 0.248640 LR 0.000500 Time 0.021363 -2023-02-13 18:09:03,343 - Epoch: [115][ 330/ 1207] Overall Loss 0.249252 Objective Loss 0.249252 LR 0.000500 Time 0.021287 -2023-02-13 18:09:03,531 - Epoch: [115][ 340/ 1207] Overall Loss 0.249615 Objective Loss 0.249615 LR 0.000500 Time 0.021213 -2023-02-13 18:09:03,720 - Epoch: [115][ 350/ 1207] Overall Loss 0.249697 Objective Loss 0.249697 LR 0.000500 Time 0.021148 -2023-02-13 18:09:03,909 - Epoch: [115][ 360/ 1207] Overall Loss 0.249993 Objective Loss 0.249993 LR 0.000500 Time 0.021082 -2023-02-13 18:09:04,097 - Epoch: [115][ 370/ 1207] Overall Loss 0.250152 Objective Loss 0.250152 LR 0.000500 Time 0.021022 -2023-02-13 18:09:04,285 - Epoch: [115][ 380/ 1207] Overall Loss 0.250121 Objective Loss 0.250121 LR 0.000500 Time 0.020963 -2023-02-13 18:09:04,475 - Epoch: [115][ 390/ 1207] Overall Loss 0.250035 Objective Loss 0.250035 LR 0.000500 Time 0.020909 -2023-02-13 18:09:04,663 - Epoch: [115][ 400/ 1207] Overall Loss 0.249911 Objective Loss 0.249911 LR 0.000500 Time 0.020857 -2023-02-13 18:09:04,852 - Epoch: [115][ 410/ 1207] Overall Loss 0.250363 Objective Loss 0.250363 LR 0.000500 Time 0.020809 -2023-02-13 18:09:05,041 - Epoch: [115][ 420/ 1207] Overall Loss 0.249951 Objective Loss 0.249951 LR 0.000500 Time 0.020761 -2023-02-13 18:09:05,230 - Epoch: [115][ 430/ 1207] Overall Loss 0.249896 Objective Loss 0.249896 LR 0.000500 Time 0.020717 -2023-02-13 18:09:05,418 - Epoch: [115][ 440/ 1207] Overall Loss 0.250295 Objective Loss 0.250295 LR 0.000500 Time 0.020673 -2023-02-13 18:09:05,607 - Epoch: [115][ 450/ 1207] Overall Loss 0.250620 Objective Loss 0.250620 LR 0.000500 Time 0.020633 -2023-02-13 18:09:05,797 - Epoch: [115][ 460/ 1207] Overall Loss 0.251134 Objective Loss 0.251134 LR 0.000500 Time 0.020595 -2023-02-13 18:09:05,987 - Epoch: [115][ 470/ 1207] Overall Loss 0.250835 Objective Loss 0.250835 LR 0.000500 Time 0.020562 -2023-02-13 18:09:06,175 - Epoch: [115][ 480/ 1207] Overall Loss 0.251628 Objective Loss 0.251628 LR 0.000500 Time 0.020525 -2023-02-13 18:09:06,366 - Epoch: [115][ 490/ 1207] Overall Loss 0.251567 Objective Loss 0.251567 LR 0.000500 Time 0.020493 -2023-02-13 18:09:06,554 - Epoch: [115][ 500/ 1207] Overall Loss 0.251722 Objective Loss 0.251722 LR 0.000500 Time 0.020460 -2023-02-13 18:09:06,745 - Epoch: [115][ 510/ 1207] Overall Loss 0.251946 Objective Loss 0.251946 LR 0.000500 Time 0.020433 -2023-02-13 18:09:06,935 - Epoch: [115][ 520/ 1207] Overall Loss 0.251381 Objective Loss 0.251381 LR 0.000500 Time 0.020404 -2023-02-13 18:09:07,124 - Epoch: [115][ 530/ 1207] Overall Loss 0.251735 Objective Loss 0.251735 LR 0.000500 Time 0.020376 -2023-02-13 18:09:07,313 - Epoch: [115][ 540/ 1207] Overall Loss 0.251916 Objective Loss 0.251916 LR 0.000500 Time 0.020346 -2023-02-13 18:09:07,502 - Epoch: [115][ 550/ 1207] Overall Loss 0.251770 Objective Loss 0.251770 LR 0.000500 Time 0.020320 -2023-02-13 18:09:07,691 - Epoch: [115][ 560/ 1207] Overall Loss 0.251763 Objective Loss 0.251763 LR 0.000500 Time 0.020294 -2023-02-13 18:09:07,881 - Epoch: [115][ 570/ 1207] Overall Loss 0.251502 Objective Loss 0.251502 LR 0.000500 Time 0.020271 -2023-02-13 18:09:08,070 - Epoch: [115][ 580/ 1207] Overall Loss 0.251733 Objective Loss 0.251733 LR 0.000500 Time 0.020246 -2023-02-13 18:09:08,259 - Epoch: [115][ 590/ 1207] Overall Loss 0.251655 Objective Loss 0.251655 LR 0.000500 Time 0.020222 -2023-02-13 18:09:08,447 - Epoch: [115][ 600/ 1207] Overall Loss 0.251837 Objective Loss 0.251837 LR 0.000500 Time 0.020198 -2023-02-13 18:09:08,636 - Epoch: [115][ 610/ 1207] Overall Loss 0.252158 Objective Loss 0.252158 LR 0.000500 Time 0.020176 -2023-02-13 18:09:08,826 - Epoch: [115][ 620/ 1207] Overall Loss 0.251672 Objective Loss 0.251672 LR 0.000500 Time 0.020156 -2023-02-13 18:09:09,015 - Epoch: [115][ 630/ 1207] Overall Loss 0.251539 Objective Loss 0.251539 LR 0.000500 Time 0.020137 -2023-02-13 18:09:09,204 - Epoch: [115][ 640/ 1207] Overall Loss 0.251618 Objective Loss 0.251618 LR 0.000500 Time 0.020116 -2023-02-13 18:09:09,393 - Epoch: [115][ 650/ 1207] Overall Loss 0.251452 Objective Loss 0.251452 LR 0.000500 Time 0.020097 -2023-02-13 18:09:09,582 - Epoch: [115][ 660/ 1207] Overall Loss 0.251198 Objective Loss 0.251198 LR 0.000500 Time 0.020078 -2023-02-13 18:09:09,771 - Epoch: [115][ 670/ 1207] Overall Loss 0.251075 Objective Loss 0.251075 LR 0.000500 Time 0.020060 -2023-02-13 18:09:09,960 - Epoch: [115][ 680/ 1207] Overall Loss 0.251047 Objective Loss 0.251047 LR 0.000500 Time 0.020042 -2023-02-13 18:09:10,149 - Epoch: [115][ 690/ 1207] Overall Loss 0.251038 Objective Loss 0.251038 LR 0.000500 Time 0.020025 -2023-02-13 18:09:10,337 - Epoch: [115][ 700/ 1207] Overall Loss 0.251046 Objective Loss 0.251046 LR 0.000500 Time 0.020008 -2023-02-13 18:09:10,526 - Epoch: [115][ 710/ 1207] Overall Loss 0.250664 Objective Loss 0.250664 LR 0.000500 Time 0.019992 -2023-02-13 18:09:10,715 - Epoch: [115][ 720/ 1207] Overall Loss 0.250492 Objective Loss 0.250492 LR 0.000500 Time 0.019976 -2023-02-13 18:09:10,908 - Epoch: [115][ 730/ 1207] Overall Loss 0.250307 Objective Loss 0.250307 LR 0.000500 Time 0.019966 -2023-02-13 18:09:11,105 - Epoch: [115][ 740/ 1207] Overall Loss 0.250773 Objective Loss 0.250773 LR 0.000500 Time 0.019961 -2023-02-13 18:09:11,299 - Epoch: [115][ 750/ 1207] Overall Loss 0.250960 Objective Loss 0.250960 LR 0.000500 Time 0.019953 -2023-02-13 18:09:11,495 - Epoch: [115][ 760/ 1207] Overall Loss 0.251033 Objective Loss 0.251033 LR 0.000500 Time 0.019949 -2023-02-13 18:09:11,689 - Epoch: [115][ 770/ 1207] Overall Loss 0.251471 Objective Loss 0.251471 LR 0.000500 Time 0.019942 -2023-02-13 18:09:11,885 - Epoch: [115][ 780/ 1207] Overall Loss 0.251591 Objective Loss 0.251591 LR 0.000500 Time 0.019936 -2023-02-13 18:09:12,080 - Epoch: [115][ 790/ 1207] Overall Loss 0.251589 Objective Loss 0.251589 LR 0.000500 Time 0.019931 -2023-02-13 18:09:12,272 - Epoch: [115][ 800/ 1207] Overall Loss 0.251577 Objective Loss 0.251577 LR 0.000500 Time 0.019921 -2023-02-13 18:09:12,469 - Epoch: [115][ 810/ 1207] Overall Loss 0.252248 Objective Loss 0.252248 LR 0.000500 Time 0.019917 -2023-02-13 18:09:12,661 - Epoch: [115][ 820/ 1207] Overall Loss 0.252316 Objective Loss 0.252316 LR 0.000500 Time 0.019908 -2023-02-13 18:09:12,857 - Epoch: [115][ 830/ 1207] Overall Loss 0.252312 Objective Loss 0.252312 LR 0.000500 Time 0.019904 -2023-02-13 18:09:13,050 - Epoch: [115][ 840/ 1207] Overall Loss 0.252583 Objective Loss 0.252583 LR 0.000500 Time 0.019896 -2023-02-13 18:09:13,245 - Epoch: [115][ 850/ 1207] Overall Loss 0.252453 Objective Loss 0.252453 LR 0.000500 Time 0.019892 -2023-02-13 18:09:13,438 - Epoch: [115][ 860/ 1207] Overall Loss 0.252209 Objective Loss 0.252209 LR 0.000500 Time 0.019884 -2023-02-13 18:09:13,634 - Epoch: [115][ 870/ 1207] Overall Loss 0.252268 Objective Loss 0.252268 LR 0.000500 Time 0.019880 -2023-02-13 18:09:13,827 - Epoch: [115][ 880/ 1207] Overall Loss 0.252150 Objective Loss 0.252150 LR 0.000500 Time 0.019874 -2023-02-13 18:09:14,023 - Epoch: [115][ 890/ 1207] Overall Loss 0.252172 Objective Loss 0.252172 LR 0.000500 Time 0.019870 -2023-02-13 18:09:14,216 - Epoch: [115][ 900/ 1207] Overall Loss 0.252207 Objective Loss 0.252207 LR 0.000500 Time 0.019863 -2023-02-13 18:09:14,411 - Epoch: [115][ 910/ 1207] Overall Loss 0.252247 Objective Loss 0.252247 LR 0.000500 Time 0.019859 -2023-02-13 18:09:14,604 - Epoch: [115][ 920/ 1207] Overall Loss 0.252211 Objective Loss 0.252211 LR 0.000500 Time 0.019852 -2023-02-13 18:09:14,801 - Epoch: [115][ 930/ 1207] Overall Loss 0.252458 Objective Loss 0.252458 LR 0.000500 Time 0.019850 -2023-02-13 18:09:14,994 - Epoch: [115][ 940/ 1207] Overall Loss 0.252422 Objective Loss 0.252422 LR 0.000500 Time 0.019843 -2023-02-13 18:09:15,189 - Epoch: [115][ 950/ 1207] Overall Loss 0.252332 Objective Loss 0.252332 LR 0.000500 Time 0.019840 -2023-02-13 18:09:15,382 - Epoch: [115][ 960/ 1207] Overall Loss 0.252252 Objective Loss 0.252252 LR 0.000500 Time 0.019834 -2023-02-13 18:09:15,578 - Epoch: [115][ 970/ 1207] Overall Loss 0.252220 Objective Loss 0.252220 LR 0.000500 Time 0.019831 -2023-02-13 18:09:15,772 - Epoch: [115][ 980/ 1207] Overall Loss 0.252412 Objective Loss 0.252412 LR 0.000500 Time 0.019826 -2023-02-13 18:09:15,968 - Epoch: [115][ 990/ 1207] Overall Loss 0.252307 Objective Loss 0.252307 LR 0.000500 Time 0.019824 -2023-02-13 18:09:16,161 - Epoch: [115][ 1000/ 1207] Overall Loss 0.252179 Objective Loss 0.252179 LR 0.000500 Time 0.019818 -2023-02-13 18:09:16,356 - Epoch: [115][ 1010/ 1207] Overall Loss 0.252122 Objective Loss 0.252122 LR 0.000500 Time 0.019815 -2023-02-13 18:09:16,549 - Epoch: [115][ 1020/ 1207] Overall Loss 0.251980 Objective Loss 0.251980 LR 0.000500 Time 0.019809 -2023-02-13 18:09:16,745 - Epoch: [115][ 1030/ 1207] Overall Loss 0.251901 Objective Loss 0.251901 LR 0.000500 Time 0.019807 -2023-02-13 18:09:16,938 - Epoch: [115][ 1040/ 1207] Overall Loss 0.251864 Objective Loss 0.251864 LR 0.000500 Time 0.019802 -2023-02-13 18:09:17,134 - Epoch: [115][ 1050/ 1207] Overall Loss 0.251646 Objective Loss 0.251646 LR 0.000500 Time 0.019799 -2023-02-13 18:09:17,326 - Epoch: [115][ 1060/ 1207] Overall Loss 0.251752 Objective Loss 0.251752 LR 0.000500 Time 0.019793 -2023-02-13 18:09:17,522 - Epoch: [115][ 1070/ 1207] Overall Loss 0.251748 Objective Loss 0.251748 LR 0.000500 Time 0.019791 -2023-02-13 18:09:17,715 - Epoch: [115][ 1080/ 1207] Overall Loss 0.251674 Objective Loss 0.251674 LR 0.000500 Time 0.019786 -2023-02-13 18:09:17,911 - Epoch: [115][ 1090/ 1207] Overall Loss 0.251805 Objective Loss 0.251805 LR 0.000500 Time 0.019784 -2023-02-13 18:09:18,104 - Epoch: [115][ 1100/ 1207] Overall Loss 0.251763 Objective Loss 0.251763 LR 0.000500 Time 0.019779 -2023-02-13 18:09:18,300 - Epoch: [115][ 1110/ 1207] Overall Loss 0.251790 Objective Loss 0.251790 LR 0.000500 Time 0.019777 -2023-02-13 18:09:18,492 - Epoch: [115][ 1120/ 1207] Overall Loss 0.251815 Objective Loss 0.251815 LR 0.000500 Time 0.019772 -2023-02-13 18:09:18,688 - Epoch: [115][ 1130/ 1207] Overall Loss 0.252173 Objective Loss 0.252173 LR 0.000500 Time 0.019770 -2023-02-13 18:09:18,881 - Epoch: [115][ 1140/ 1207] Overall Loss 0.252371 Objective Loss 0.252371 LR 0.000500 Time 0.019766 -2023-02-13 18:09:19,077 - Epoch: [115][ 1150/ 1207] Overall Loss 0.252842 Objective Loss 0.252842 LR 0.000500 Time 0.019763 -2023-02-13 18:09:19,270 - Epoch: [115][ 1160/ 1207] Overall Loss 0.252856 Objective Loss 0.252856 LR 0.000500 Time 0.019759 -2023-02-13 18:09:19,465 - Epoch: [115][ 1170/ 1207] Overall Loss 0.252803 Objective Loss 0.252803 LR 0.000500 Time 0.019757 -2023-02-13 18:09:19,657 - Epoch: [115][ 1180/ 1207] Overall Loss 0.252837 Objective Loss 0.252837 LR 0.000500 Time 0.019752 -2023-02-13 18:09:19,854 - Epoch: [115][ 1190/ 1207] Overall Loss 0.252916 Objective Loss 0.252916 LR 0.000500 Time 0.019751 -2023-02-13 18:09:20,098 - Epoch: [115][ 1200/ 1207] Overall Loss 0.252940 Objective Loss 0.252940 LR 0.000500 Time 0.019789 -2023-02-13 18:09:20,213 - Epoch: [115][ 1207/ 1207] Overall Loss 0.253020 Objective Loss 0.253020 Top1 87.804878 Top5 98.170732 LR 0.000500 Time 0.019770 -2023-02-13 18:09:20,285 - --- validate (epoch=115)----------- -2023-02-13 18:09:20,285 - 34311 samples (256 per mini-batch) -2023-02-13 18:09:20,681 - Epoch: [115][ 10/ 135] Loss 0.343442 Top1 82.734375 Top5 97.578125 -2023-02-13 18:09:20,812 - Epoch: [115][ 20/ 135] Loss 0.376804 Top1 82.363281 Top5 97.207031 -2023-02-13 18:09:20,938 - Epoch: [115][ 30/ 135] Loss 0.348944 Top1 82.916667 Top5 97.486979 -2023-02-13 18:09:21,063 - Epoch: [115][ 40/ 135] Loss 0.334248 Top1 83.505859 Top5 97.548828 -2023-02-13 18:09:21,186 - Epoch: [115][ 50/ 135] Loss 0.328112 Top1 83.507812 Top5 97.554688 -2023-02-13 18:09:21,311 - Epoch: [115][ 60/ 135] Loss 0.326362 Top1 83.593750 Top5 97.526042 -2023-02-13 18:09:21,438 - Epoch: [115][ 70/ 135] Loss 0.326614 Top1 83.487723 Top5 97.505580 -2023-02-13 18:09:21,571 - Epoch: [115][ 80/ 135] Loss 0.319271 Top1 83.662109 Top5 97.534180 -2023-02-13 18:09:21,701 - Epoch: [115][ 90/ 135] Loss 0.318097 Top1 83.658854 Top5 97.517361 -2023-02-13 18:09:21,831 - Epoch: [115][ 100/ 135] Loss 0.316843 Top1 83.699219 Top5 97.562500 -2023-02-13 18:09:21,958 - Epoch: [115][ 110/ 135] Loss 0.317313 Top1 83.700284 Top5 97.546165 -2023-02-13 18:09:22,088 - Epoch: [115][ 120/ 135] Loss 0.319387 Top1 83.701172 Top5 97.558594 -2023-02-13 18:09:22,221 - Epoch: [115][ 130/ 135] Loss 0.321754 Top1 83.608774 Top5 97.518029 -2023-02-13 18:09:22,267 - Epoch: [115][ 135/ 135] Loss 0.317750 Top1 83.629157 Top5 97.519746 -2023-02-13 18:09:22,345 - ==> Top1: 83.629 Top5: 97.520 Loss: 0.318 - -2023-02-13 18:09:22,346 - ==> Confusion: -[[ 842 5 5 1 8 1 0 0 4 62 0 5 0 8 8 6 3 3 1 1 4] - [ 2 933 0 3 11 33 1 16 3 2 2 2 1 0 1 3 3 0 7 4 6] - [ 8 3 941 14 5 1 15 15 0 2 3 1 4 7 6 14 2 3 5 2 7] - [ 2 2 22 895 4 4 0 4 1 0 12 0 7 2 29 3 6 4 14 0 5] - [ 9 8 0 1 993 7 0 0 1 4 0 3 2 5 8 6 8 1 0 6 4] - [ 2 19 1 3 5 970 2 12 0 4 1 11 5 19 1 2 6 1 1 3 2] - [ 2 4 16 3 0 10 1022 7 0 1 6 0 4 1 0 4 1 6 2 9 1] - [ 0 8 4 1 4 30 2 926 1 2 2 7 4 0 0 1 1 1 15 12 3] - [ 16 2 1 1 2 0 0 3 897 32 8 2 0 11 26 2 1 1 2 0 2] - [ 59 1 4 0 5 0 0 1 37 871 0 3 0 18 6 1 1 2 1 1 1] - [ 1 2 4 5 1 5 1 5 20 1 973 1 0 14 5 0 0 1 8 1 3] - [ 1 1 0 0 6 14 0 7 0 3 0 917 22 5 0 7 2 8 2 8 2] - [ 1 0 0 3 1 3 0 3 2 0 2 35 872 1 4 7 4 13 3 0 5] - [ 4 1 1 1 11 10 1 1 14 13 4 6 2 944 2 5 2 0 0 0 2] - [ 7 3 0 8 4 7 0 1 16 6 1 1 2 1 1006 3 1 6 8 1 10] - [ 5 1 6 0 7 0 2 0 0 0 0 6 8 5 2 973 14 7 0 5 5] - [ 1 7 1 1 7 2 0 1 3 0 1 1 0 3 1 15 1000 2 2 3 10] - [ 3 2 1 2 0 2 2 2 1 1 1 7 25 2 3 18 0 972 0 1 6] - [ 2 1 4 6 1 2 1 26 6 0 6 4 4 0 14 1 1 2 1002 1 2] - [ 1 6 1 0 1 12 5 12 0 0 1 15 4 7 0 5 8 4 0 1060 6] - [ 157 234 199 121 152 232 78 199 123 111 186 127 300 352 202 125 268 101 223 260 9684]] - -2023-02-13 18:09:22,347 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:09:22,347 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:09:22,353 - - -2023-02-13 18:09:22,353 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:09:23,252 - Epoch: [116][ 10/ 1207] Overall Loss 0.234460 Objective Loss 0.234460 LR 0.000500 Time 0.089868 -2023-02-13 18:09:23,445 - Epoch: [116][ 20/ 1207] Overall Loss 0.234281 Objective Loss 0.234281 LR 0.000500 Time 0.054532 -2023-02-13 18:09:23,634 - Epoch: [116][ 30/ 1207] Overall Loss 0.238332 Objective Loss 0.238332 LR 0.000500 Time 0.042662 -2023-02-13 18:09:23,824 - Epoch: [116][ 40/ 1207] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.000500 Time 0.036734 -2023-02-13 18:09:24,012 - Epoch: [116][ 50/ 1207] Overall Loss 0.245133 Objective Loss 0.245133 LR 0.000500 Time 0.033140 -2023-02-13 18:09:24,200 - Epoch: [116][ 60/ 1207] Overall Loss 0.247201 Objective Loss 0.247201 LR 0.000500 Time 0.030743 -2023-02-13 18:09:24,389 - Epoch: [116][ 70/ 1207] Overall Loss 0.249179 Objective Loss 0.249179 LR 0.000500 Time 0.029035 -2023-02-13 18:09:24,576 - Epoch: [116][ 80/ 1207] Overall Loss 0.247830 Objective Loss 0.247830 LR 0.000500 Time 0.027748 -2023-02-13 18:09:24,765 - Epoch: [116][ 90/ 1207] Overall Loss 0.247473 Objective Loss 0.247473 LR 0.000500 Time 0.026762 -2023-02-13 18:09:24,954 - Epoch: [116][ 100/ 1207] Overall Loss 0.245878 Objective Loss 0.245878 LR 0.000500 Time 0.025967 -2023-02-13 18:09:25,141 - Epoch: [116][ 110/ 1207] Overall Loss 0.247221 Objective Loss 0.247221 LR 0.000500 Time 0.025306 -2023-02-13 18:09:25,329 - Epoch: [116][ 120/ 1207] Overall Loss 0.247590 Objective Loss 0.247590 LR 0.000500 Time 0.024762 -2023-02-13 18:09:25,517 - Epoch: [116][ 130/ 1207] Overall Loss 0.247387 Objective Loss 0.247387 LR 0.000500 Time 0.024298 -2023-02-13 18:09:25,706 - Epoch: [116][ 140/ 1207] Overall Loss 0.247763 Objective Loss 0.247763 LR 0.000500 Time 0.023906 -2023-02-13 18:09:25,896 - Epoch: [116][ 150/ 1207] Overall Loss 0.245491 Objective Loss 0.245491 LR 0.000500 Time 0.023582 -2023-02-13 18:09:26,085 - Epoch: [116][ 160/ 1207] Overall Loss 0.244570 Objective Loss 0.244570 LR 0.000500 Time 0.023287 -2023-02-13 18:09:26,274 - Epoch: [116][ 170/ 1207] Overall Loss 0.245625 Objective Loss 0.245625 LR 0.000500 Time 0.023024 -2023-02-13 18:09:26,463 - Epoch: [116][ 180/ 1207] Overall Loss 0.245489 Objective Loss 0.245489 LR 0.000500 Time 0.022791 -2023-02-13 18:09:26,651 - Epoch: [116][ 190/ 1207] Overall Loss 0.243985 Objective Loss 0.243985 LR 0.000500 Time 0.022582 -2023-02-13 18:09:26,841 - Epoch: [116][ 200/ 1207] Overall Loss 0.245651 Objective Loss 0.245651 LR 0.000500 Time 0.022402 -2023-02-13 18:09:27,030 - Epoch: [116][ 210/ 1207] Overall Loss 0.245936 Objective Loss 0.245936 LR 0.000500 Time 0.022233 -2023-02-13 18:09:27,218 - Epoch: [116][ 220/ 1207] Overall Loss 0.246329 Objective Loss 0.246329 LR 0.000500 Time 0.022077 -2023-02-13 18:09:27,408 - Epoch: [116][ 230/ 1207] Overall Loss 0.245999 Objective Loss 0.245999 LR 0.000500 Time 0.021939 -2023-02-13 18:09:27,597 - Epoch: [116][ 240/ 1207] Overall Loss 0.246705 Objective Loss 0.246705 LR 0.000500 Time 0.021809 -2023-02-13 18:09:27,785 - Epoch: [116][ 250/ 1207] Overall Loss 0.246359 Objective Loss 0.246359 LR 0.000500 Time 0.021691 -2023-02-13 18:09:27,975 - Epoch: [116][ 260/ 1207] Overall Loss 0.246172 Objective Loss 0.246172 LR 0.000500 Time 0.021585 -2023-02-13 18:09:28,164 - Epoch: [116][ 270/ 1207] Overall Loss 0.245031 Objective Loss 0.245031 LR 0.000500 Time 0.021484 -2023-02-13 18:09:28,352 - Epoch: [116][ 280/ 1207] Overall Loss 0.245118 Objective Loss 0.245118 LR 0.000500 Time 0.021388 -2023-02-13 18:09:28,540 - Epoch: [116][ 290/ 1207] Overall Loss 0.244199 Objective Loss 0.244199 LR 0.000500 Time 0.021298 -2023-02-13 18:09:28,729 - Epoch: [116][ 300/ 1207] Overall Loss 0.244740 Objective Loss 0.244740 LR 0.000500 Time 0.021216 -2023-02-13 18:09:28,919 - Epoch: [116][ 310/ 1207] Overall Loss 0.244060 Objective Loss 0.244060 LR 0.000500 Time 0.021141 -2023-02-13 18:09:29,108 - Epoch: [116][ 320/ 1207] Overall Loss 0.244805 Objective Loss 0.244805 LR 0.000500 Time 0.021071 -2023-02-13 18:09:29,297 - Epoch: [116][ 330/ 1207] Overall Loss 0.244835 Objective Loss 0.244835 LR 0.000500 Time 0.021003 -2023-02-13 18:09:29,485 - Epoch: [116][ 340/ 1207] Overall Loss 0.244476 Objective Loss 0.244476 LR 0.000500 Time 0.020939 -2023-02-13 18:09:29,673 - Epoch: [116][ 350/ 1207] Overall Loss 0.245138 Objective Loss 0.245138 LR 0.000500 Time 0.020878 -2023-02-13 18:09:29,863 - Epoch: [116][ 360/ 1207] Overall Loss 0.244847 Objective Loss 0.244847 LR 0.000500 Time 0.020824 -2023-02-13 18:09:30,052 - Epoch: [116][ 370/ 1207] Overall Loss 0.244600 Objective Loss 0.244600 LR 0.000500 Time 0.020771 -2023-02-13 18:09:30,242 - Epoch: [116][ 380/ 1207] Overall Loss 0.244777 Objective Loss 0.244777 LR 0.000500 Time 0.020722 -2023-02-13 18:09:30,431 - Epoch: [116][ 390/ 1207] Overall Loss 0.244751 Objective Loss 0.244751 LR 0.000500 Time 0.020675 -2023-02-13 18:09:30,621 - Epoch: [116][ 400/ 1207] Overall Loss 0.244705 Objective Loss 0.244705 LR 0.000500 Time 0.020632 -2023-02-13 18:09:30,811 - Epoch: [116][ 410/ 1207] Overall Loss 0.244633 Objective Loss 0.244633 LR 0.000500 Time 0.020592 -2023-02-13 18:09:31,004 - Epoch: [116][ 420/ 1207] Overall Loss 0.245058 Objective Loss 0.245058 LR 0.000500 Time 0.020560 -2023-02-13 18:09:31,195 - Epoch: [116][ 430/ 1207] Overall Loss 0.244883 Objective Loss 0.244883 LR 0.000500 Time 0.020525 -2023-02-13 18:09:31,388 - Epoch: [116][ 440/ 1207] Overall Loss 0.244501 Objective Loss 0.244501 LR 0.000500 Time 0.020496 -2023-02-13 18:09:31,578 - Epoch: [116][ 450/ 1207] Overall Loss 0.244255 Objective Loss 0.244255 LR 0.000500 Time 0.020463 -2023-02-13 18:09:31,771 - Epoch: [116][ 460/ 1207] Overall Loss 0.244140 Objective Loss 0.244140 LR 0.000500 Time 0.020436 -2023-02-13 18:09:31,963 - Epoch: [116][ 470/ 1207] Overall Loss 0.244567 Objective Loss 0.244567 LR 0.000500 Time 0.020408 -2023-02-13 18:09:32,154 - Epoch: [116][ 480/ 1207] Overall Loss 0.244474 Objective Loss 0.244474 LR 0.000500 Time 0.020382 -2023-02-13 18:09:32,346 - Epoch: [116][ 490/ 1207] Overall Loss 0.244797 Objective Loss 0.244797 LR 0.000500 Time 0.020355 -2023-02-13 18:09:32,537 - Epoch: [116][ 500/ 1207] Overall Loss 0.244900 Objective Loss 0.244900 LR 0.000500 Time 0.020331 -2023-02-13 18:09:32,729 - Epoch: [116][ 510/ 1207] Overall Loss 0.244705 Objective Loss 0.244705 LR 0.000500 Time 0.020307 -2023-02-13 18:09:32,921 - Epoch: [116][ 520/ 1207] Overall Loss 0.245132 Objective Loss 0.245132 LR 0.000500 Time 0.020286 -2023-02-13 18:09:33,112 - Epoch: [116][ 530/ 1207] Overall Loss 0.245325 Objective Loss 0.245325 LR 0.000500 Time 0.020263 -2023-02-13 18:09:33,304 - Epoch: [116][ 540/ 1207] Overall Loss 0.245231 Objective Loss 0.245231 LR 0.000500 Time 0.020242 -2023-02-13 18:09:33,496 - Epoch: [116][ 550/ 1207] Overall Loss 0.245204 Objective Loss 0.245204 LR 0.000500 Time 0.020222 -2023-02-13 18:09:33,688 - Epoch: [116][ 560/ 1207] Overall Loss 0.245188 Objective Loss 0.245188 LR 0.000500 Time 0.020203 -2023-02-13 18:09:33,880 - Epoch: [116][ 570/ 1207] Overall Loss 0.245153 Objective Loss 0.245153 LR 0.000500 Time 0.020186 -2023-02-13 18:09:34,072 - Epoch: [116][ 580/ 1207] Overall Loss 0.244852 Objective Loss 0.244852 LR 0.000500 Time 0.020168 -2023-02-13 18:09:34,264 - Epoch: [116][ 590/ 1207] Overall Loss 0.245040 Objective Loss 0.245040 LR 0.000500 Time 0.020151 -2023-02-13 18:09:34,456 - Epoch: [116][ 600/ 1207] Overall Loss 0.244825 Objective Loss 0.244825 LR 0.000500 Time 0.020134 -2023-02-13 18:09:34,648 - Epoch: [116][ 610/ 1207] Overall Loss 0.245110 Objective Loss 0.245110 LR 0.000500 Time 0.020118 -2023-02-13 18:09:34,841 - Epoch: [116][ 620/ 1207] Overall Loss 0.245689 Objective Loss 0.245689 LR 0.000500 Time 0.020104 -2023-02-13 18:09:35,032 - Epoch: [116][ 630/ 1207] Overall Loss 0.245886 Objective Loss 0.245886 LR 0.000500 Time 0.020088 -2023-02-13 18:09:35,224 - Epoch: [116][ 640/ 1207] Overall Loss 0.245978 Objective Loss 0.245978 LR 0.000500 Time 0.020073 -2023-02-13 18:09:35,415 - Epoch: [116][ 650/ 1207] Overall Loss 0.246204 Objective Loss 0.246204 LR 0.000500 Time 0.020058 -2023-02-13 18:09:35,608 - Epoch: [116][ 660/ 1207] Overall Loss 0.246199 Objective Loss 0.246199 LR 0.000500 Time 0.020046 -2023-02-13 18:09:35,801 - Epoch: [116][ 670/ 1207] Overall Loss 0.246349 Objective Loss 0.246349 LR 0.000500 Time 0.020034 -2023-02-13 18:09:35,996 - Epoch: [116][ 680/ 1207] Overall Loss 0.246589 Objective Loss 0.246589 LR 0.000500 Time 0.020026 -2023-02-13 18:09:36,188 - Epoch: [116][ 690/ 1207] Overall Loss 0.246832 Objective Loss 0.246832 LR 0.000500 Time 0.020013 -2023-02-13 18:09:36,383 - Epoch: [116][ 700/ 1207] Overall Loss 0.247238 Objective Loss 0.247238 LR 0.000500 Time 0.020005 -2023-02-13 18:09:36,579 - Epoch: [116][ 710/ 1207] Overall Loss 0.247342 Objective Loss 0.247342 LR 0.000500 Time 0.020000 -2023-02-13 18:09:36,780 - Epoch: [116][ 720/ 1207] Overall Loss 0.247236 Objective Loss 0.247236 LR 0.000500 Time 0.020000 -2023-02-13 18:09:36,977 - Epoch: [116][ 730/ 1207] Overall Loss 0.247766 Objective Loss 0.247766 LR 0.000500 Time 0.019995 -2023-02-13 18:09:37,176 - Epoch: [116][ 740/ 1207] Overall Loss 0.247833 Objective Loss 0.247833 LR 0.000500 Time 0.019994 -2023-02-13 18:09:37,372 - Epoch: [116][ 750/ 1207] Overall Loss 0.248379 Objective Loss 0.248379 LR 0.000500 Time 0.019988 -2023-02-13 18:09:37,572 - Epoch: [116][ 760/ 1207] Overall Loss 0.247995 Objective Loss 0.247995 LR 0.000500 Time 0.019987 -2023-02-13 18:09:37,769 - Epoch: [116][ 770/ 1207] Overall Loss 0.247957 Objective Loss 0.247957 LR 0.000500 Time 0.019983 -2023-02-13 18:09:37,969 - Epoch: [116][ 780/ 1207] Overall Loss 0.247979 Objective Loss 0.247979 LR 0.000500 Time 0.019983 -2023-02-13 18:09:38,165 - Epoch: [116][ 790/ 1207] Overall Loss 0.247848 Objective Loss 0.247848 LR 0.000500 Time 0.019978 -2023-02-13 18:09:38,365 - Epoch: [116][ 800/ 1207] Overall Loss 0.247963 Objective Loss 0.247963 LR 0.000500 Time 0.019977 -2023-02-13 18:09:38,561 - Epoch: [116][ 810/ 1207] Overall Loss 0.247992 Objective Loss 0.247992 LR 0.000500 Time 0.019973 -2023-02-13 18:09:38,761 - Epoch: [116][ 820/ 1207] Overall Loss 0.248017 Objective Loss 0.248017 LR 0.000500 Time 0.019973 -2023-02-13 18:09:38,958 - Epoch: [116][ 830/ 1207] Overall Loss 0.247981 Objective Loss 0.247981 LR 0.000500 Time 0.019968 -2023-02-13 18:09:39,157 - Epoch: [116][ 840/ 1207] Overall Loss 0.247731 Objective Loss 0.247731 LR 0.000500 Time 0.019968 -2023-02-13 18:09:39,353 - Epoch: [116][ 850/ 1207] Overall Loss 0.247663 Objective Loss 0.247663 LR 0.000500 Time 0.019963 -2023-02-13 18:09:39,553 - Epoch: [116][ 860/ 1207] Overall Loss 0.247744 Objective Loss 0.247744 LR 0.000500 Time 0.019963 -2023-02-13 18:09:39,750 - Epoch: [116][ 870/ 1207] Overall Loss 0.247916 Objective Loss 0.247916 LR 0.000500 Time 0.019959 -2023-02-13 18:09:39,950 - Epoch: [116][ 880/ 1207] Overall Loss 0.248195 Objective Loss 0.248195 LR 0.000500 Time 0.019959 -2023-02-13 18:09:40,147 - Epoch: [116][ 890/ 1207] Overall Loss 0.248454 Objective Loss 0.248454 LR 0.000500 Time 0.019955 -2023-02-13 18:09:40,346 - Epoch: [116][ 900/ 1207] Overall Loss 0.248815 Objective Loss 0.248815 LR 0.000500 Time 0.019955 -2023-02-13 18:09:40,544 - Epoch: [116][ 910/ 1207] Overall Loss 0.248909 Objective Loss 0.248909 LR 0.000500 Time 0.019952 -2023-02-13 18:09:40,743 - Epoch: [116][ 920/ 1207] Overall Loss 0.248551 Objective Loss 0.248551 LR 0.000500 Time 0.019952 -2023-02-13 18:09:40,941 - Epoch: [116][ 930/ 1207] Overall Loss 0.248897 Objective Loss 0.248897 LR 0.000500 Time 0.019950 -2023-02-13 18:09:41,139 - Epoch: [116][ 940/ 1207] Overall Loss 0.248911 Objective Loss 0.248911 LR 0.000500 Time 0.019947 -2023-02-13 18:09:41,333 - Epoch: [116][ 950/ 1207] Overall Loss 0.248976 Objective Loss 0.248976 LR 0.000500 Time 0.019942 -2023-02-13 18:09:41,531 - Epoch: [116][ 960/ 1207] Overall Loss 0.249375 Objective Loss 0.249375 LR 0.000500 Time 0.019939 -2023-02-13 18:09:41,726 - Epoch: [116][ 970/ 1207] Overall Loss 0.249425 Objective Loss 0.249425 LR 0.000500 Time 0.019934 -2023-02-13 18:09:41,926 - Epoch: [116][ 980/ 1207] Overall Loss 0.249516 Objective Loss 0.249516 LR 0.000500 Time 0.019935 -2023-02-13 18:09:42,123 - Epoch: [116][ 990/ 1207] Overall Loss 0.249650 Objective Loss 0.249650 LR 0.000500 Time 0.019932 -2023-02-13 18:09:42,323 - Epoch: [116][ 1000/ 1207] Overall Loss 0.249886 Objective Loss 0.249886 LR 0.000500 Time 0.019932 -2023-02-13 18:09:42,521 - Epoch: [116][ 1010/ 1207] Overall Loss 0.249922 Objective Loss 0.249922 LR 0.000500 Time 0.019931 -2023-02-13 18:09:42,721 - Epoch: [116][ 1020/ 1207] Overall Loss 0.249964 Objective Loss 0.249964 LR 0.000500 Time 0.019931 -2023-02-13 18:09:42,917 - Epoch: [116][ 1030/ 1207] Overall Loss 0.250055 Objective Loss 0.250055 LR 0.000500 Time 0.019928 -2023-02-13 18:09:43,110 - Epoch: [116][ 1040/ 1207] Overall Loss 0.250347 Objective Loss 0.250347 LR 0.000500 Time 0.019921 -2023-02-13 18:09:43,303 - Epoch: [116][ 1050/ 1207] Overall Loss 0.250613 Objective Loss 0.250613 LR 0.000500 Time 0.019915 -2023-02-13 18:09:43,495 - Epoch: [116][ 1060/ 1207] Overall Loss 0.250728 Objective Loss 0.250728 LR 0.000500 Time 0.019907 -2023-02-13 18:09:43,688 - Epoch: [116][ 1070/ 1207] Overall Loss 0.250861 Objective Loss 0.250861 LR 0.000500 Time 0.019902 -2023-02-13 18:09:43,881 - Epoch: [116][ 1080/ 1207] Overall Loss 0.251000 Objective Loss 0.251000 LR 0.000500 Time 0.019896 -2023-02-13 18:09:44,074 - Epoch: [116][ 1090/ 1207] Overall Loss 0.250822 Objective Loss 0.250822 LR 0.000500 Time 0.019890 -2023-02-13 18:09:44,267 - Epoch: [116][ 1100/ 1207] Overall Loss 0.250558 Objective Loss 0.250558 LR 0.000500 Time 0.019884 -2023-02-13 18:09:44,459 - Epoch: [116][ 1110/ 1207] Overall Loss 0.250432 Objective Loss 0.250432 LR 0.000500 Time 0.019878 -2023-02-13 18:09:44,651 - Epoch: [116][ 1120/ 1207] Overall Loss 0.250361 Objective Loss 0.250361 LR 0.000500 Time 0.019872 -2023-02-13 18:09:44,845 - Epoch: [116][ 1130/ 1207] Overall Loss 0.250434 Objective Loss 0.250434 LR 0.000500 Time 0.019867 -2023-02-13 18:09:45,039 - Epoch: [116][ 1140/ 1207] Overall Loss 0.250360 Objective Loss 0.250360 LR 0.000500 Time 0.019862 -2023-02-13 18:09:45,231 - Epoch: [116][ 1150/ 1207] Overall Loss 0.250224 Objective Loss 0.250224 LR 0.000500 Time 0.019856 -2023-02-13 18:09:45,423 - Epoch: [116][ 1160/ 1207] Overall Loss 0.250011 Objective Loss 0.250011 LR 0.000500 Time 0.019851 -2023-02-13 18:09:45,615 - Epoch: [116][ 1170/ 1207] Overall Loss 0.250291 Objective Loss 0.250291 LR 0.000500 Time 0.019844 -2023-02-13 18:09:45,807 - Epoch: [116][ 1180/ 1207] Overall Loss 0.250320 Objective Loss 0.250320 LR 0.000500 Time 0.019839 -2023-02-13 18:09:46,000 - Epoch: [116][ 1190/ 1207] Overall Loss 0.250259 Objective Loss 0.250259 LR 0.000500 Time 0.019834 -2023-02-13 18:09:46,241 - Epoch: [116][ 1200/ 1207] Overall Loss 0.250297 Objective Loss 0.250297 LR 0.000500 Time 0.019870 -2023-02-13 18:09:46,355 - Epoch: [116][ 1207/ 1207] Overall Loss 0.250133 Objective Loss 0.250133 Top1 87.804878 Top5 97.865854 LR 0.000500 Time 0.019849 -2023-02-13 18:09:46,427 - --- validate (epoch=116)----------- -2023-02-13 18:09:46,427 - 34311 samples (256 per mini-batch) -2023-02-13 18:09:46,844 - Epoch: [116][ 10/ 135] Loss 0.306150 Top1 84.335938 Top5 97.617188 -2023-02-13 18:09:46,984 - Epoch: [116][ 20/ 135] Loss 0.299208 Top1 84.140625 Top5 97.578125 -2023-02-13 18:09:47,122 - Epoch: [116][ 30/ 135] Loss 0.299722 Top1 84.361979 Top5 97.434896 -2023-02-13 18:09:47,267 - Epoch: [116][ 40/ 135] Loss 0.307728 Top1 84.101562 Top5 97.460938 -2023-02-13 18:09:47,396 - Epoch: [116][ 50/ 135] Loss 0.303926 Top1 84.453125 Top5 97.617188 -2023-02-13 18:09:47,525 - Epoch: [116][ 60/ 135] Loss 0.307401 Top1 84.401042 Top5 97.558594 -2023-02-13 18:09:47,656 - Epoch: [116][ 70/ 135] Loss 0.304605 Top1 84.369420 Top5 97.600446 -2023-02-13 18:09:47,785 - Epoch: [116][ 80/ 135] Loss 0.303312 Top1 84.394531 Top5 97.587891 -2023-02-13 18:09:47,913 - Epoch: [116][ 90/ 135] Loss 0.300422 Top1 84.388021 Top5 97.677951 -2023-02-13 18:09:48,042 - Epoch: [116][ 100/ 135] Loss 0.303945 Top1 84.371094 Top5 97.695312 -2023-02-13 18:09:48,172 - Epoch: [116][ 110/ 135] Loss 0.307061 Top1 84.225852 Top5 97.713068 -2023-02-13 18:09:48,303 - Epoch: [116][ 120/ 135] Loss 0.310612 Top1 84.215495 Top5 97.705078 -2023-02-13 18:09:48,436 - Epoch: [116][ 130/ 135] Loss 0.310868 Top1 84.272837 Top5 97.707332 -2023-02-13 18:09:48,483 - Epoch: [116][ 135/ 135] Loss 0.309979 Top1 84.273265 Top5 97.715019 -2023-02-13 18:09:48,568 - ==> Top1: 84.273 Top5: 97.715 Loss: 0.310 - -2023-02-13 18:09:48,569 - ==> Confusion: -[[ 856 2 5 0 8 3 0 0 6 55 2 4 1 6 3 7 0 1 1 2 5] - [ 3 940 0 4 10 35 2 15 5 2 1 3 2 0 0 2 1 0 2 0 6] - [ 9 2 950 12 5 1 22 12 0 1 2 2 5 6 3 8 1 3 2 4 8] - [ 3 2 16 907 6 5 3 2 1 2 12 0 6 2 20 2 4 5 15 0 3] - [ 15 8 0 1 996 9 1 0 4 4 1 6 0 3 6 4 2 0 0 2 4] - [ 3 14 0 3 7 991 3 12 1 4 3 7 2 9 1 2 4 0 1 2 1] - [ 2 5 17 2 0 9 1039 2 0 1 4 0 4 0 0 4 0 1 1 6 2] - [ 0 5 5 1 3 32 3 930 1 1 5 3 5 2 0 1 0 2 11 6 8] - [ 21 3 1 1 1 1 0 4 883 37 11 1 0 11 20 3 0 2 6 0 3] - [ 72 0 2 0 5 1 0 2 42 846 2 1 0 26 3 2 1 0 1 1 5] - [ 2 1 2 6 3 5 1 3 15 0 985 1 1 10 3 0 1 0 9 0 3] - [ 1 2 0 0 7 17 1 6 2 2 1 907 26 8 1 3 1 7 3 10 0] - [ 0 0 1 6 0 7 0 3 2 1 0 23 886 1 2 7 2 12 1 0 5] - [ 3 1 2 0 9 10 0 2 7 13 10 4 3 944 4 5 2 0 0 0 5] - [ 12 2 1 17 5 4 1 1 12 8 2 0 4 4 993 2 0 5 9 0 10] - [ 2 2 4 2 5 0 5 0 2 0 0 8 8 3 1 975 9 8 0 7 5] - [ 4 7 1 2 10 3 0 0 0 0 1 0 4 4 1 13 989 0 3 3 16] - [ 3 3 0 5 1 3 2 1 1 0 1 8 23 1 1 16 0 973 1 3 5] - [ 2 3 5 12 2 4 1 36 3 1 6 1 3 0 17 1 0 1 985 2 1] - [ 1 2 0 0 2 9 7 12 1 0 1 15 4 4 0 5 3 1 0 1072 9] - [ 169 241 203 117 158 287 101 190 99 78 207 125 309 302 149 116 198 92 152 273 9868]] - -2023-02-13 18:09:48,570 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:09:48,570 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:09:48,576 - - -2023-02-13 18:09:48,576 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:09:49,473 - Epoch: [117][ 10/ 1207] Overall Loss 0.230182 Objective Loss 0.230182 LR 0.000500 Time 0.089581 -2023-02-13 18:09:49,674 - Epoch: [117][ 20/ 1207] Overall Loss 0.245718 Objective Loss 0.245718 LR 0.000500 Time 0.054851 -2023-02-13 18:09:49,874 - Epoch: [117][ 30/ 1207] Overall Loss 0.240294 Objective Loss 0.240294 LR 0.000500 Time 0.043197 -2023-02-13 18:09:50,069 - Epoch: [117][ 40/ 1207] Overall Loss 0.245481 Objective Loss 0.245481 LR 0.000500 Time 0.037270 -2023-02-13 18:09:50,267 - Epoch: [117][ 50/ 1207] Overall Loss 0.245318 Objective Loss 0.245318 LR 0.000500 Time 0.033768 -2023-02-13 18:09:50,462 - Epoch: [117][ 60/ 1207] Overall Loss 0.244638 Objective Loss 0.244638 LR 0.000500 Time 0.031377 -2023-02-13 18:09:50,660 - Epoch: [117][ 70/ 1207] Overall Loss 0.242161 Objective Loss 0.242161 LR 0.000500 Time 0.029726 -2023-02-13 18:09:50,856 - Epoch: [117][ 80/ 1207] Overall Loss 0.247910 Objective Loss 0.247910 LR 0.000500 Time 0.028453 -2023-02-13 18:09:51,054 - Epoch: [117][ 90/ 1207] Overall Loss 0.247753 Objective Loss 0.247753 LR 0.000500 Time 0.027492 -2023-02-13 18:09:51,249 - Epoch: [117][ 100/ 1207] Overall Loss 0.246116 Objective Loss 0.246116 LR 0.000500 Time 0.026689 -2023-02-13 18:09:51,448 - Epoch: [117][ 110/ 1207] Overall Loss 0.245515 Objective Loss 0.245515 LR 0.000500 Time 0.026065 -2023-02-13 18:09:51,643 - Epoch: [117][ 120/ 1207] Overall Loss 0.245645 Objective Loss 0.245645 LR 0.000500 Time 0.025513 -2023-02-13 18:09:51,842 - Epoch: [117][ 130/ 1207] Overall Loss 0.247756 Objective Loss 0.247756 LR 0.000500 Time 0.025077 -2023-02-13 18:09:52,038 - Epoch: [117][ 140/ 1207] Overall Loss 0.246349 Objective Loss 0.246349 LR 0.000500 Time 0.024683 -2023-02-13 18:09:52,236 - Epoch: [117][ 150/ 1207] Overall Loss 0.245685 Objective Loss 0.245685 LR 0.000500 Time 0.024357 -2023-02-13 18:09:52,431 - Epoch: [117][ 160/ 1207] Overall Loss 0.246746 Objective Loss 0.246746 LR 0.000500 Time 0.024050 -2023-02-13 18:09:52,629 - Epoch: [117][ 170/ 1207] Overall Loss 0.248459 Objective Loss 0.248459 LR 0.000500 Time 0.023796 -2023-02-13 18:09:52,824 - Epoch: [117][ 180/ 1207] Overall Loss 0.248639 Objective Loss 0.248639 LR 0.000500 Time 0.023560 -2023-02-13 18:09:53,023 - Epoch: [117][ 190/ 1207] Overall Loss 0.247359 Objective Loss 0.247359 LR 0.000500 Time 0.023362 -2023-02-13 18:09:53,218 - Epoch: [117][ 200/ 1207] Overall Loss 0.247617 Objective Loss 0.247617 LR 0.000500 Time 0.023170 -2023-02-13 18:09:53,416 - Epoch: [117][ 210/ 1207] Overall Loss 0.247199 Objective Loss 0.247199 LR 0.000500 Time 0.023007 -2023-02-13 18:09:53,611 - Epoch: [117][ 220/ 1207] Overall Loss 0.246283 Objective Loss 0.246283 LR 0.000500 Time 0.022845 -2023-02-13 18:09:53,810 - Epoch: [117][ 230/ 1207] Overall Loss 0.246158 Objective Loss 0.246158 LR 0.000500 Time 0.022713 -2023-02-13 18:09:54,008 - Epoch: [117][ 240/ 1207] Overall Loss 0.246740 Objective Loss 0.246740 LR 0.000500 Time 0.022592 -2023-02-13 18:09:54,208 - Epoch: [117][ 250/ 1207] Overall Loss 0.246709 Objective Loss 0.246709 LR 0.000500 Time 0.022487 -2023-02-13 18:09:54,405 - Epoch: [117][ 260/ 1207] Overall Loss 0.247587 Objective Loss 0.247587 LR 0.000500 Time 0.022379 -2023-02-13 18:09:54,605 - Epoch: [117][ 270/ 1207] Overall Loss 0.247404 Objective Loss 0.247404 LR 0.000500 Time 0.022290 -2023-02-13 18:09:54,803 - Epoch: [117][ 280/ 1207] Overall Loss 0.247430 Objective Loss 0.247430 LR 0.000500 Time 0.022199 -2023-02-13 18:09:55,004 - Epoch: [117][ 290/ 1207] Overall Loss 0.247949 Objective Loss 0.247949 LR 0.000500 Time 0.022125 -2023-02-13 18:09:55,202 - Epoch: [117][ 300/ 1207] Overall Loss 0.248280 Objective Loss 0.248280 LR 0.000500 Time 0.022045 -2023-02-13 18:09:55,402 - Epoch: [117][ 310/ 1207] Overall Loss 0.248202 Objective Loss 0.248202 LR 0.000500 Time 0.021978 -2023-02-13 18:09:55,599 - Epoch: [117][ 320/ 1207] Overall Loss 0.247598 Objective Loss 0.247598 LR 0.000500 Time 0.021907 -2023-02-13 18:09:55,799 - Epoch: [117][ 330/ 1207] Overall Loss 0.248061 Objective Loss 0.248061 LR 0.000500 Time 0.021849 -2023-02-13 18:09:55,998 - Epoch: [117][ 340/ 1207] Overall Loss 0.248208 Objective Loss 0.248208 LR 0.000500 Time 0.021790 -2023-02-13 18:09:56,196 - Epoch: [117][ 350/ 1207] Overall Loss 0.247750 Objective Loss 0.247750 LR 0.000500 Time 0.021732 -2023-02-13 18:09:56,391 - Epoch: [117][ 360/ 1207] Overall Loss 0.248435 Objective Loss 0.248435 LR 0.000500 Time 0.021668 -2023-02-13 18:09:56,590 - Epoch: [117][ 370/ 1207] Overall Loss 0.248379 Objective Loss 0.248379 LR 0.000500 Time 0.021618 -2023-02-13 18:09:56,785 - Epoch: [117][ 380/ 1207] Overall Loss 0.248313 Objective Loss 0.248313 LR 0.000500 Time 0.021563 -2023-02-13 18:09:56,985 - Epoch: [117][ 390/ 1207] Overall Loss 0.246960 Objective Loss 0.246960 LR 0.000500 Time 0.021521 -2023-02-13 18:09:57,180 - Epoch: [117][ 400/ 1207] Overall Loss 0.245983 Objective Loss 0.245983 LR 0.000500 Time 0.021470 -2023-02-13 18:09:57,379 - Epoch: [117][ 410/ 1207] Overall Loss 0.245929 Objective Loss 0.245929 LR 0.000500 Time 0.021431 -2023-02-13 18:09:57,574 - Epoch: [117][ 420/ 1207] Overall Loss 0.245907 Objective Loss 0.245907 LR 0.000500 Time 0.021385 -2023-02-13 18:09:57,773 - Epoch: [117][ 430/ 1207] Overall Loss 0.245969 Objective Loss 0.245969 LR 0.000500 Time 0.021348 -2023-02-13 18:09:57,969 - Epoch: [117][ 440/ 1207] Overall Loss 0.246261 Objective Loss 0.246261 LR 0.000500 Time 0.021307 -2023-02-13 18:09:58,167 - Epoch: [117][ 450/ 1207] Overall Loss 0.246419 Objective Loss 0.246419 LR 0.000500 Time 0.021273 -2023-02-13 18:09:58,363 - Epoch: [117][ 460/ 1207] Overall Loss 0.246209 Objective Loss 0.246209 LR 0.000500 Time 0.021235 -2023-02-13 18:09:58,561 - Epoch: [117][ 470/ 1207] Overall Loss 0.246205 Objective Loss 0.246205 LR 0.000500 Time 0.021205 -2023-02-13 18:09:58,757 - Epoch: [117][ 480/ 1207] Overall Loss 0.246078 Objective Loss 0.246078 LR 0.000500 Time 0.021171 -2023-02-13 18:09:58,957 - Epoch: [117][ 490/ 1207] Overall Loss 0.246191 Objective Loss 0.246191 LR 0.000500 Time 0.021146 -2023-02-13 18:09:59,152 - Epoch: [117][ 500/ 1207] Overall Loss 0.245926 Objective Loss 0.245926 LR 0.000500 Time 0.021113 -2023-02-13 18:09:59,351 - Epoch: [117][ 510/ 1207] Overall Loss 0.246114 Objective Loss 0.246114 LR 0.000500 Time 0.021087 -2023-02-13 18:09:59,546 - Epoch: [117][ 520/ 1207] Overall Loss 0.245872 Objective Loss 0.245872 LR 0.000500 Time 0.021056 -2023-02-13 18:09:59,745 - Epoch: [117][ 530/ 1207] Overall Loss 0.245749 Objective Loss 0.245749 LR 0.000500 Time 0.021033 -2023-02-13 18:09:59,941 - Epoch: [117][ 540/ 1207] Overall Loss 0.245851 Objective Loss 0.245851 LR 0.000500 Time 0.021007 -2023-02-13 18:10:00,140 - Epoch: [117][ 550/ 1207] Overall Loss 0.246419 Objective Loss 0.246419 LR 0.000500 Time 0.020985 -2023-02-13 18:10:00,335 - Epoch: [117][ 560/ 1207] Overall Loss 0.246364 Objective Loss 0.246364 LR 0.000500 Time 0.020958 -2023-02-13 18:10:00,533 - Epoch: [117][ 570/ 1207] Overall Loss 0.246439 Objective Loss 0.246439 LR 0.000500 Time 0.020938 -2023-02-13 18:10:00,729 - Epoch: [117][ 580/ 1207] Overall Loss 0.247036 Objective Loss 0.247036 LR 0.000500 Time 0.020913 -2023-02-13 18:10:00,929 - Epoch: [117][ 590/ 1207] Overall Loss 0.246885 Objective Loss 0.246885 LR 0.000500 Time 0.020898 -2023-02-13 18:10:01,122 - Epoch: [117][ 600/ 1207] Overall Loss 0.247107 Objective Loss 0.247107 LR 0.000500 Time 0.020871 -2023-02-13 18:10:01,314 - Epoch: [117][ 610/ 1207] Overall Loss 0.247218 Objective Loss 0.247218 LR 0.000500 Time 0.020842 -2023-02-13 18:10:01,506 - Epoch: [117][ 620/ 1207] Overall Loss 0.246666 Objective Loss 0.246666 LR 0.000500 Time 0.020815 -2023-02-13 18:10:01,697 - Epoch: [117][ 630/ 1207] Overall Loss 0.246840 Objective Loss 0.246840 LR 0.000500 Time 0.020788 -2023-02-13 18:10:01,890 - Epoch: [117][ 640/ 1207] Overall Loss 0.246975 Objective Loss 0.246975 LR 0.000500 Time 0.020763 -2023-02-13 18:10:02,081 - Epoch: [117][ 650/ 1207] Overall Loss 0.247887 Objective Loss 0.247887 LR 0.000500 Time 0.020738 -2023-02-13 18:10:02,273 - Epoch: [117][ 660/ 1207] Overall Loss 0.247791 Objective Loss 0.247791 LR 0.000500 Time 0.020713 -2023-02-13 18:10:02,465 - Epoch: [117][ 670/ 1207] Overall Loss 0.247966 Objective Loss 0.247966 LR 0.000500 Time 0.020690 -2023-02-13 18:10:02,656 - Epoch: [117][ 680/ 1207] Overall Loss 0.248302 Objective Loss 0.248302 LR 0.000500 Time 0.020666 -2023-02-13 18:10:02,848 - Epoch: [117][ 690/ 1207] Overall Loss 0.248338 Objective Loss 0.248338 LR 0.000500 Time 0.020645 -2023-02-13 18:10:03,041 - Epoch: [117][ 700/ 1207] Overall Loss 0.247953 Objective Loss 0.247953 LR 0.000500 Time 0.020624 -2023-02-13 18:10:03,233 - Epoch: [117][ 710/ 1207] Overall Loss 0.248381 Objective Loss 0.248381 LR 0.000500 Time 0.020604 -2023-02-13 18:10:03,424 - Epoch: [117][ 720/ 1207] Overall Loss 0.248438 Objective Loss 0.248438 LR 0.000500 Time 0.020583 -2023-02-13 18:10:03,616 - Epoch: [117][ 730/ 1207] Overall Loss 0.248604 Objective Loss 0.248604 LR 0.000500 Time 0.020564 -2023-02-13 18:10:03,808 - Epoch: [117][ 740/ 1207] Overall Loss 0.248863 Objective Loss 0.248863 LR 0.000500 Time 0.020545 -2023-02-13 18:10:04,001 - Epoch: [117][ 750/ 1207] Overall Loss 0.248678 Objective Loss 0.248678 LR 0.000500 Time 0.020527 -2023-02-13 18:10:04,192 - Epoch: [117][ 760/ 1207] Overall Loss 0.248750 Objective Loss 0.248750 LR 0.000500 Time 0.020508 -2023-02-13 18:10:04,384 - Epoch: [117][ 770/ 1207] Overall Loss 0.248597 Objective Loss 0.248597 LR 0.000500 Time 0.020491 -2023-02-13 18:10:04,575 - Epoch: [117][ 780/ 1207] Overall Loss 0.248336 Objective Loss 0.248336 LR 0.000500 Time 0.020473 -2023-02-13 18:10:04,767 - Epoch: [117][ 790/ 1207] Overall Loss 0.248431 Objective Loss 0.248431 LR 0.000500 Time 0.020456 -2023-02-13 18:10:04,958 - Epoch: [117][ 800/ 1207] Overall Loss 0.248509 Objective Loss 0.248509 LR 0.000500 Time 0.020439 -2023-02-13 18:10:05,150 - Epoch: [117][ 810/ 1207] Overall Loss 0.248498 Objective Loss 0.248498 LR 0.000500 Time 0.020422 -2023-02-13 18:10:05,341 - Epoch: [117][ 820/ 1207] Overall Loss 0.248661 Objective Loss 0.248661 LR 0.000500 Time 0.020406 -2023-02-13 18:10:05,533 - Epoch: [117][ 830/ 1207] Overall Loss 0.248848 Objective Loss 0.248848 LR 0.000500 Time 0.020391 -2023-02-13 18:10:05,725 - Epoch: [117][ 840/ 1207] Overall Loss 0.248813 Objective Loss 0.248813 LR 0.000500 Time 0.020376 -2023-02-13 18:10:05,918 - Epoch: [117][ 850/ 1207] Overall Loss 0.248646 Objective Loss 0.248646 LR 0.000500 Time 0.020363 -2023-02-13 18:10:06,110 - Epoch: [117][ 860/ 1207] Overall Loss 0.248966 Objective Loss 0.248966 LR 0.000500 Time 0.020349 -2023-02-13 18:10:06,301 - Epoch: [117][ 870/ 1207] Overall Loss 0.248690 Objective Loss 0.248690 LR 0.000500 Time 0.020335 -2023-02-13 18:10:06,493 - Epoch: [117][ 880/ 1207] Overall Loss 0.248589 Objective Loss 0.248589 LR 0.000500 Time 0.020321 -2023-02-13 18:10:06,685 - Epoch: [117][ 890/ 1207] Overall Loss 0.248671 Objective Loss 0.248671 LR 0.000500 Time 0.020308 -2023-02-13 18:10:06,878 - Epoch: [117][ 900/ 1207] Overall Loss 0.248759 Objective Loss 0.248759 LR 0.000500 Time 0.020296 -2023-02-13 18:10:07,069 - Epoch: [117][ 910/ 1207] Overall Loss 0.248743 Objective Loss 0.248743 LR 0.000500 Time 0.020283 -2023-02-13 18:10:07,261 - Epoch: [117][ 920/ 1207] Overall Loss 0.248991 Objective Loss 0.248991 LR 0.000500 Time 0.020270 -2023-02-13 18:10:07,452 - Epoch: [117][ 930/ 1207] Overall Loss 0.249313 Objective Loss 0.249313 LR 0.000500 Time 0.020258 -2023-02-13 18:10:07,644 - Epoch: [117][ 940/ 1207] Overall Loss 0.249328 Objective Loss 0.249328 LR 0.000500 Time 0.020246 -2023-02-13 18:10:07,836 - Epoch: [117][ 950/ 1207] Overall Loss 0.249502 Objective Loss 0.249502 LR 0.000500 Time 0.020235 -2023-02-13 18:10:08,028 - Epoch: [117][ 960/ 1207] Overall Loss 0.249566 Objective Loss 0.249566 LR 0.000500 Time 0.020223 -2023-02-13 18:10:08,220 - Epoch: [117][ 970/ 1207] Overall Loss 0.249522 Objective Loss 0.249522 LR 0.000500 Time 0.020212 -2023-02-13 18:10:08,411 - Epoch: [117][ 980/ 1207] Overall Loss 0.249459 Objective Loss 0.249459 LR 0.000500 Time 0.020201 -2023-02-13 18:10:08,602 - Epoch: [117][ 990/ 1207] Overall Loss 0.249717 Objective Loss 0.249717 LR 0.000500 Time 0.020190 -2023-02-13 18:10:08,794 - Epoch: [117][ 1000/ 1207] Overall Loss 0.249907 Objective Loss 0.249907 LR 0.000500 Time 0.020179 -2023-02-13 18:10:08,986 - Epoch: [117][ 1010/ 1207] Overall Loss 0.249830 Objective Loss 0.249830 LR 0.000500 Time 0.020169 -2023-02-13 18:10:09,178 - Epoch: [117][ 1020/ 1207] Overall Loss 0.250048 Objective Loss 0.250048 LR 0.000500 Time 0.020159 -2023-02-13 18:10:09,369 - Epoch: [117][ 1030/ 1207] Overall Loss 0.250256 Objective Loss 0.250256 LR 0.000500 Time 0.020149 -2023-02-13 18:10:09,561 - Epoch: [117][ 1040/ 1207] Overall Loss 0.250279 Objective Loss 0.250279 LR 0.000500 Time 0.020139 -2023-02-13 18:10:09,753 - Epoch: [117][ 1050/ 1207] Overall Loss 0.250374 Objective Loss 0.250374 LR 0.000500 Time 0.020129 -2023-02-13 18:10:09,945 - Epoch: [117][ 1060/ 1207] Overall Loss 0.250523 Objective Loss 0.250523 LR 0.000500 Time 0.020120 -2023-02-13 18:10:10,137 - Epoch: [117][ 1070/ 1207] Overall Loss 0.250476 Objective Loss 0.250476 LR 0.000500 Time 0.020111 -2023-02-13 18:10:10,328 - Epoch: [117][ 1080/ 1207] Overall Loss 0.250527 Objective Loss 0.250527 LR 0.000500 Time 0.020102 -2023-02-13 18:10:10,520 - Epoch: [117][ 1090/ 1207] Overall Loss 0.250845 Objective Loss 0.250845 LR 0.000500 Time 0.020094 -2023-02-13 18:10:10,712 - Epoch: [117][ 1100/ 1207] Overall Loss 0.250787 Objective Loss 0.250787 LR 0.000500 Time 0.020084 -2023-02-13 18:10:10,904 - Epoch: [117][ 1110/ 1207] Overall Loss 0.251056 Objective Loss 0.251056 LR 0.000500 Time 0.020076 -2023-02-13 18:10:11,096 - Epoch: [117][ 1120/ 1207] Overall Loss 0.251055 Objective Loss 0.251055 LR 0.000500 Time 0.020068 -2023-02-13 18:10:11,288 - Epoch: [117][ 1130/ 1207] Overall Loss 0.251221 Objective Loss 0.251221 LR 0.000500 Time 0.020060 -2023-02-13 18:10:11,479 - Epoch: [117][ 1140/ 1207] Overall Loss 0.251182 Objective Loss 0.251182 LR 0.000500 Time 0.020052 -2023-02-13 18:10:11,672 - Epoch: [117][ 1150/ 1207] Overall Loss 0.251306 Objective Loss 0.251306 LR 0.000500 Time 0.020044 -2023-02-13 18:10:11,864 - Epoch: [117][ 1160/ 1207] Overall Loss 0.251057 Objective Loss 0.251057 LR 0.000500 Time 0.020037 -2023-02-13 18:10:12,057 - Epoch: [117][ 1170/ 1207] Overall Loss 0.251058 Objective Loss 0.251058 LR 0.000500 Time 0.020030 -2023-02-13 18:10:12,249 - Epoch: [117][ 1180/ 1207] Overall Loss 0.250985 Objective Loss 0.250985 LR 0.000500 Time 0.020023 -2023-02-13 18:10:12,440 - Epoch: [117][ 1190/ 1207] Overall Loss 0.251059 Objective Loss 0.251059 LR 0.000500 Time 0.020015 -2023-02-13 18:10:12,683 - Epoch: [117][ 1200/ 1207] Overall Loss 0.250983 Objective Loss 0.250983 LR 0.000500 Time 0.020051 -2023-02-13 18:10:12,800 - Epoch: [117][ 1207/ 1207] Overall Loss 0.250806 Objective Loss 0.250806 Top1 86.890244 Top5 98.170732 LR 0.000500 Time 0.020031 -2023-02-13 18:10:12,872 - --- validate (epoch=117)----------- -2023-02-13 18:10:12,873 - 34311 samples (256 per mini-batch) -2023-02-13 18:10:13,282 - Epoch: [117][ 10/ 135] Loss 0.267575 Top1 85.507812 Top5 98.281250 -2023-02-13 18:10:13,417 - Epoch: [117][ 20/ 135] Loss 0.295736 Top1 84.804688 Top5 97.832031 -2023-02-13 18:10:13,546 - Epoch: [117][ 30/ 135] Loss 0.298188 Top1 84.388021 Top5 97.851562 -2023-02-13 18:10:13,677 - Epoch: [117][ 40/ 135] Loss 0.304635 Top1 84.433594 Top5 97.773438 -2023-02-13 18:10:13,808 - Epoch: [117][ 50/ 135] Loss 0.294074 Top1 84.507812 Top5 97.789062 -2023-02-13 18:10:13,937 - Epoch: [117][ 60/ 135] Loss 0.296210 Top1 84.524740 Top5 97.760417 -2023-02-13 18:10:14,068 - Epoch: [117][ 70/ 135] Loss 0.300665 Top1 84.386161 Top5 97.689732 -2023-02-13 18:10:14,195 - Epoch: [117][ 80/ 135] Loss 0.305159 Top1 84.218750 Top5 97.612305 -2023-02-13 18:10:14,324 - Epoch: [117][ 90/ 135] Loss 0.305520 Top1 84.201389 Top5 97.573785 -2023-02-13 18:10:14,451 - Epoch: [117][ 100/ 135] Loss 0.310581 Top1 84.140625 Top5 97.566406 -2023-02-13 18:10:14,580 - Epoch: [117][ 110/ 135] Loss 0.314509 Top1 84.066051 Top5 97.524858 -2023-02-13 18:10:14,710 - Epoch: [117][ 120/ 135] Loss 0.317566 Top1 84.026693 Top5 97.500000 -2023-02-13 18:10:14,841 - Epoch: [117][ 130/ 135] Loss 0.315133 Top1 84.044471 Top5 97.527043 -2023-02-13 18:10:14,888 - Epoch: [117][ 135/ 135] Loss 0.314018 Top1 84.051762 Top5 97.540147 -2023-02-13 18:10:14,959 - ==> Top1: 84.052 Top5: 97.540 Loss: 0.314 - -2023-02-13 18:10:14,960 - ==> Confusion: -[[ 838 6 10 2 9 3 0 3 4 59 1 4 1 5 2 5 1 6 2 2 4] - [ 2 926 0 2 12 45 2 13 2 2 3 3 1 0 3 2 6 0 2 1 6] - [ 7 3 961 13 3 3 12 13 1 1 3 0 4 5 5 6 3 2 5 1 7] - [ 1 3 20 910 3 3 1 2 1 2 17 0 7 0 14 3 6 7 12 0 4] - [ 11 6 0 1 995 13 0 1 2 4 0 7 1 1 6 5 5 2 1 0 5] - [ 1 8 0 4 7 983 5 15 0 2 4 10 4 11 0 3 3 2 0 6 2] - [ 3 4 18 5 0 6 1036 5 0 1 2 0 5 1 0 2 1 3 2 3 2] - [ 2 6 10 2 2 31 4 931 2 1 1 4 4 1 0 0 0 2 10 7 4] - [ 10 6 1 2 1 0 1 2 907 38 7 1 0 13 9 1 1 3 4 0 2] - [ 61 3 4 0 9 1 0 1 40 854 0 1 0 24 4 1 0 2 0 0 7] - [ 1 3 2 5 0 4 3 4 15 1 993 1 1 6 3 0 0 1 5 1 2] - [ 1 1 1 0 2 7 1 6 1 0 0 930 20 4 1 6 1 9 2 11 1] - [ 0 0 2 5 2 3 0 2 2 0 3 25 875 1 1 6 2 19 2 0 9] - [ 2 3 2 0 6 11 1 3 8 13 12 7 2 931 1 5 3 2 1 3 8] - [ 5 3 4 23 5 3 0 1 24 9 3 0 4 2 981 1 2 6 8 0 8] - [ 2 1 9 0 5 2 4 0 1 0 0 9 7 3 0 971 6 15 0 7 4] - [ 1 4 1 1 12 3 0 1 1 1 2 3 2 2 1 19 991 2 1 4 9] - [ 5 4 1 3 0 1 0 1 0 1 2 6 14 1 0 11 0 994 0 2 5] - [ 2 5 2 10 2 1 0 27 5 0 8 3 5 2 19 1 1 2 988 3 0] - [ 1 2 1 1 1 9 9 12 0 0 1 21 5 2 0 8 6 3 0 1061 5] - [ 125 226 280 144 134 283 109 161 100 77 214 148 311 302 141 119 235 143 157 242 9783]] - -2023-02-13 18:10:14,961 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:10:14,961 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:10:14,967 - - -2023-02-13 18:10:14,967 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:10:15,950 - Epoch: [118][ 10/ 1207] Overall Loss 0.229307 Objective Loss 0.229307 LR 0.000500 Time 0.098291 -2023-02-13 18:10:16,147 - Epoch: [118][ 20/ 1207] Overall Loss 0.241690 Objective Loss 0.241690 LR 0.000500 Time 0.058948 -2023-02-13 18:10:16,334 - Epoch: [118][ 30/ 1207] Overall Loss 0.242927 Objective Loss 0.242927 LR 0.000500 Time 0.045525 -2023-02-13 18:10:16,521 - Epoch: [118][ 40/ 1207] Overall Loss 0.241319 Objective Loss 0.241319 LR 0.000500 Time 0.038814 -2023-02-13 18:10:16,719 - Epoch: [118][ 50/ 1207] Overall Loss 0.240317 Objective Loss 0.240317 LR 0.000500 Time 0.034998 -2023-02-13 18:10:16,923 - Epoch: [118][ 60/ 1207] Overall Loss 0.237907 Objective Loss 0.237907 LR 0.000500 Time 0.032565 -2023-02-13 18:10:17,122 - Epoch: [118][ 70/ 1207] Overall Loss 0.235649 Objective Loss 0.235649 LR 0.000500 Time 0.030746 -2023-02-13 18:10:17,326 - Epoch: [118][ 80/ 1207] Overall Loss 0.237851 Objective Loss 0.237851 LR 0.000500 Time 0.029442 -2023-02-13 18:10:17,525 - Epoch: [118][ 90/ 1207] Overall Loss 0.237330 Objective Loss 0.237330 LR 0.000500 Time 0.028382 -2023-02-13 18:10:17,728 - Epoch: [118][ 100/ 1207] Overall Loss 0.240029 Objective Loss 0.240029 LR 0.000500 Time 0.027571 -2023-02-13 18:10:17,929 - Epoch: [118][ 110/ 1207] Overall Loss 0.238927 Objective Loss 0.238927 LR 0.000500 Time 0.026884 -2023-02-13 18:10:18,131 - Epoch: [118][ 120/ 1207] Overall Loss 0.237099 Objective Loss 0.237099 LR 0.000500 Time 0.026326 -2023-02-13 18:10:18,331 - Epoch: [118][ 130/ 1207] Overall Loss 0.237895 Objective Loss 0.237895 LR 0.000500 Time 0.025833 -2023-02-13 18:10:18,533 - Epoch: [118][ 140/ 1207] Overall Loss 0.238798 Objective Loss 0.238798 LR 0.000500 Time 0.025432 -2023-02-13 18:10:18,731 - Epoch: [118][ 150/ 1207] Overall Loss 0.239176 Objective Loss 0.239176 LR 0.000500 Time 0.025054 -2023-02-13 18:10:18,935 - Epoch: [118][ 160/ 1207] Overall Loss 0.239395 Objective Loss 0.239395 LR 0.000500 Time 0.024757 -2023-02-13 18:10:19,133 - Epoch: [118][ 170/ 1207] Overall Loss 0.241145 Objective Loss 0.241145 LR 0.000500 Time 0.024464 -2023-02-13 18:10:19,335 - Epoch: [118][ 180/ 1207] Overall Loss 0.241893 Objective Loss 0.241893 LR 0.000500 Time 0.024226 -2023-02-13 18:10:19,535 - Epoch: [118][ 190/ 1207] Overall Loss 0.242840 Objective Loss 0.242840 LR 0.000500 Time 0.024001 -2023-02-13 18:10:19,737 - Epoch: [118][ 200/ 1207] Overall Loss 0.242262 Objective Loss 0.242262 LR 0.000500 Time 0.023808 -2023-02-13 18:10:19,937 - Epoch: [118][ 210/ 1207] Overall Loss 0.241201 Objective Loss 0.241201 LR 0.000500 Time 0.023625 -2023-02-13 18:10:20,139 - Epoch: [118][ 220/ 1207] Overall Loss 0.241736 Objective Loss 0.241736 LR 0.000500 Time 0.023467 -2023-02-13 18:10:20,336 - Epoch: [118][ 230/ 1207] Overall Loss 0.243105 Objective Loss 0.243105 LR 0.000500 Time 0.023305 -2023-02-13 18:10:20,538 - Epoch: [118][ 240/ 1207] Overall Loss 0.242520 Objective Loss 0.242520 LR 0.000500 Time 0.023174 -2023-02-13 18:10:20,736 - Epoch: [118][ 250/ 1207] Overall Loss 0.242798 Objective Loss 0.242798 LR 0.000500 Time 0.023036 -2023-02-13 18:10:20,940 - Epoch: [118][ 260/ 1207] Overall Loss 0.242416 Objective Loss 0.242416 LR 0.000500 Time 0.022933 -2023-02-13 18:10:21,139 - Epoch: [118][ 270/ 1207] Overall Loss 0.242300 Objective Loss 0.242300 LR 0.000500 Time 0.022819 -2023-02-13 18:10:21,341 - Epoch: [118][ 280/ 1207] Overall Loss 0.243458 Objective Loss 0.243458 LR 0.000500 Time 0.022725 -2023-02-13 18:10:21,539 - Epoch: [118][ 290/ 1207] Overall Loss 0.244244 Objective Loss 0.244244 LR 0.000500 Time 0.022622 -2023-02-13 18:10:21,742 - Epoch: [118][ 300/ 1207] Overall Loss 0.245374 Objective Loss 0.245374 LR 0.000500 Time 0.022542 -2023-02-13 18:10:21,941 - Epoch: [118][ 310/ 1207] Overall Loss 0.245356 Objective Loss 0.245356 LR 0.000500 Time 0.022457 -2023-02-13 18:10:22,144 - Epoch: [118][ 320/ 1207] Overall Loss 0.245518 Objective Loss 0.245518 LR 0.000500 Time 0.022388 -2023-02-13 18:10:22,343 - Epoch: [118][ 330/ 1207] Overall Loss 0.245877 Objective Loss 0.245877 LR 0.000500 Time 0.022310 -2023-02-13 18:10:22,545 - Epoch: [118][ 340/ 1207] Overall Loss 0.245847 Objective Loss 0.245847 LR 0.000500 Time 0.022249 -2023-02-13 18:10:22,744 - Epoch: [118][ 350/ 1207] Overall Loss 0.246312 Objective Loss 0.246312 LR 0.000500 Time 0.022179 -2023-02-13 18:10:22,948 - Epoch: [118][ 360/ 1207] Overall Loss 0.247138 Objective Loss 0.247138 LR 0.000500 Time 0.022128 -2023-02-13 18:10:23,146 - Epoch: [118][ 370/ 1207] Overall Loss 0.246551 Objective Loss 0.246551 LR 0.000500 Time 0.022064 -2023-02-13 18:10:23,348 - Epoch: [118][ 380/ 1207] Overall Loss 0.246585 Objective Loss 0.246585 LR 0.000500 Time 0.022016 -2023-02-13 18:10:23,547 - Epoch: [118][ 390/ 1207] Overall Loss 0.247070 Objective Loss 0.247070 LR 0.000500 Time 0.021960 -2023-02-13 18:10:23,750 - Epoch: [118][ 400/ 1207] Overall Loss 0.246770 Objective Loss 0.246770 LR 0.000500 Time 0.021916 -2023-02-13 18:10:23,949 - Epoch: [118][ 410/ 1207] Overall Loss 0.246831 Objective Loss 0.246831 LR 0.000500 Time 0.021866 -2023-02-13 18:10:24,151 - Epoch: [118][ 420/ 1207] Overall Loss 0.246827 Objective Loss 0.246827 LR 0.000500 Time 0.021827 -2023-02-13 18:10:24,349 - Epoch: [118][ 430/ 1207] Overall Loss 0.246707 Objective Loss 0.246707 LR 0.000500 Time 0.021779 -2023-02-13 18:10:24,552 - Epoch: [118][ 440/ 1207] Overall Loss 0.246254 Objective Loss 0.246254 LR 0.000500 Time 0.021743 -2023-02-13 18:10:24,750 - Epoch: [118][ 450/ 1207] Overall Loss 0.246181 Objective Loss 0.246181 LR 0.000500 Time 0.021701 -2023-02-13 18:10:24,954 - Epoch: [118][ 460/ 1207] Overall Loss 0.246124 Objective Loss 0.246124 LR 0.000500 Time 0.021671 -2023-02-13 18:10:25,154 - Epoch: [118][ 470/ 1207] Overall Loss 0.246137 Objective Loss 0.246137 LR 0.000500 Time 0.021634 -2023-02-13 18:10:25,357 - Epoch: [118][ 480/ 1207] Overall Loss 0.246197 Objective Loss 0.246197 LR 0.000500 Time 0.021606 -2023-02-13 18:10:25,556 - Epoch: [118][ 490/ 1207] Overall Loss 0.245903 Objective Loss 0.245903 LR 0.000500 Time 0.021571 -2023-02-13 18:10:25,759 - Epoch: [118][ 500/ 1207] Overall Loss 0.246494 Objective Loss 0.246494 LR 0.000500 Time 0.021544 -2023-02-13 18:10:25,959 - Epoch: [118][ 510/ 1207] Overall Loss 0.246301 Objective Loss 0.246301 LR 0.000500 Time 0.021513 -2023-02-13 18:10:26,163 - Epoch: [118][ 520/ 1207] Overall Loss 0.246701 Objective Loss 0.246701 LR 0.000500 Time 0.021491 -2023-02-13 18:10:26,363 - Epoch: [118][ 530/ 1207] Overall Loss 0.246434 Objective Loss 0.246434 LR 0.000500 Time 0.021462 -2023-02-13 18:10:26,567 - Epoch: [118][ 540/ 1207] Overall Loss 0.246762 Objective Loss 0.246762 LR 0.000500 Time 0.021441 -2023-02-13 18:10:26,766 - Epoch: [118][ 550/ 1207] Overall Loss 0.246742 Objective Loss 0.246742 LR 0.000500 Time 0.021413 -2023-02-13 18:10:26,971 - Epoch: [118][ 560/ 1207] Overall Loss 0.246930 Objective Loss 0.246930 LR 0.000500 Time 0.021396 -2023-02-13 18:10:27,170 - Epoch: [118][ 570/ 1207] Overall Loss 0.247257 Objective Loss 0.247257 LR 0.000500 Time 0.021369 -2023-02-13 18:10:27,374 - Epoch: [118][ 580/ 1207] Overall Loss 0.247553 Objective Loss 0.247553 LR 0.000500 Time 0.021352 -2023-02-13 18:10:27,573 - Epoch: [118][ 590/ 1207] Overall Loss 0.247858 Objective Loss 0.247858 LR 0.000500 Time 0.021326 -2023-02-13 18:10:27,777 - Epoch: [118][ 600/ 1207] Overall Loss 0.247976 Objective Loss 0.247976 LR 0.000500 Time 0.021309 -2023-02-13 18:10:27,976 - Epoch: [118][ 610/ 1207] Overall Loss 0.247846 Objective Loss 0.247846 LR 0.000500 Time 0.021286 -2023-02-13 18:10:28,179 - Epoch: [118][ 620/ 1207] Overall Loss 0.247536 Objective Loss 0.247536 LR 0.000500 Time 0.021269 -2023-02-13 18:10:28,377 - Epoch: [118][ 630/ 1207] Overall Loss 0.247667 Objective Loss 0.247667 LR 0.000500 Time 0.021246 -2023-02-13 18:10:28,581 - Epoch: [118][ 640/ 1207] Overall Loss 0.246826 Objective Loss 0.246826 LR 0.000500 Time 0.021232 -2023-02-13 18:10:28,780 - Epoch: [118][ 650/ 1207] Overall Loss 0.246685 Objective Loss 0.246685 LR 0.000500 Time 0.021211 -2023-02-13 18:10:28,985 - Epoch: [118][ 660/ 1207] Overall Loss 0.246980 Objective Loss 0.246980 LR 0.000500 Time 0.021199 -2023-02-13 18:10:29,184 - Epoch: [118][ 670/ 1207] Overall Loss 0.247103 Objective Loss 0.247103 LR 0.000500 Time 0.021180 -2023-02-13 18:10:29,389 - Epoch: [118][ 680/ 1207] Overall Loss 0.246889 Objective Loss 0.246889 LR 0.000500 Time 0.021168 -2023-02-13 18:10:29,588 - Epoch: [118][ 690/ 1207] Overall Loss 0.247018 Objective Loss 0.247018 LR 0.000500 Time 0.021150 -2023-02-13 18:10:29,792 - Epoch: [118][ 700/ 1207] Overall Loss 0.246886 Objective Loss 0.246886 LR 0.000500 Time 0.021139 -2023-02-13 18:10:29,992 - Epoch: [118][ 710/ 1207] Overall Loss 0.247120 Objective Loss 0.247120 LR 0.000500 Time 0.021122 -2023-02-13 18:10:30,196 - Epoch: [118][ 720/ 1207] Overall Loss 0.247070 Objective Loss 0.247070 LR 0.000500 Time 0.021112 -2023-02-13 18:10:30,395 - Epoch: [118][ 730/ 1207] Overall Loss 0.247484 Objective Loss 0.247484 LR 0.000500 Time 0.021094 -2023-02-13 18:10:30,600 - Epoch: [118][ 740/ 1207] Overall Loss 0.247540 Objective Loss 0.247540 LR 0.000500 Time 0.021085 -2023-02-13 18:10:30,799 - Epoch: [118][ 750/ 1207] Overall Loss 0.247704 Objective Loss 0.247704 LR 0.000500 Time 0.021070 -2023-02-13 18:10:31,004 - Epoch: [118][ 760/ 1207] Overall Loss 0.247943 Objective Loss 0.247943 LR 0.000500 Time 0.021061 -2023-02-13 18:10:31,203 - Epoch: [118][ 770/ 1207] Overall Loss 0.247529 Objective Loss 0.247529 LR 0.000500 Time 0.021046 -2023-02-13 18:10:31,408 - Epoch: [118][ 780/ 1207] Overall Loss 0.247407 Objective Loss 0.247407 LR 0.000500 Time 0.021037 -2023-02-13 18:10:31,607 - Epoch: [118][ 790/ 1207] Overall Loss 0.247505 Objective Loss 0.247505 LR 0.000500 Time 0.021023 -2023-02-13 18:10:31,811 - Epoch: [118][ 800/ 1207] Overall Loss 0.247818 Objective Loss 0.247818 LR 0.000500 Time 0.021015 -2023-02-13 18:10:32,011 - Epoch: [118][ 810/ 1207] Overall Loss 0.247626 Objective Loss 0.247626 LR 0.000500 Time 0.021002 -2023-02-13 18:10:32,199 - Epoch: [118][ 820/ 1207] Overall Loss 0.247460 Objective Loss 0.247460 LR 0.000500 Time 0.020975 -2023-02-13 18:10:32,388 - Epoch: [118][ 830/ 1207] Overall Loss 0.247643 Objective Loss 0.247643 LR 0.000500 Time 0.020949 -2023-02-13 18:10:32,575 - Epoch: [118][ 840/ 1207] Overall Loss 0.247475 Objective Loss 0.247475 LR 0.000500 Time 0.020922 -2023-02-13 18:10:32,764 - Epoch: [118][ 850/ 1207] Overall Loss 0.247445 Objective Loss 0.247445 LR 0.000500 Time 0.020898 -2023-02-13 18:10:32,952 - Epoch: [118][ 860/ 1207] Overall Loss 0.247585 Objective Loss 0.247585 LR 0.000500 Time 0.020873 -2023-02-13 18:10:33,141 - Epoch: [118][ 870/ 1207] Overall Loss 0.247514 Objective Loss 0.247514 LR 0.000500 Time 0.020850 -2023-02-13 18:10:33,330 - Epoch: [118][ 880/ 1207] Overall Loss 0.247960 Objective Loss 0.247960 LR 0.000500 Time 0.020827 -2023-02-13 18:10:33,518 - Epoch: [118][ 890/ 1207] Overall Loss 0.248275 Objective Loss 0.248275 LR 0.000500 Time 0.020804 -2023-02-13 18:10:33,707 - Epoch: [118][ 900/ 1207] Overall Loss 0.248471 Objective Loss 0.248471 LR 0.000500 Time 0.020782 -2023-02-13 18:10:33,896 - Epoch: [118][ 910/ 1207] Overall Loss 0.248498 Objective Loss 0.248498 LR 0.000500 Time 0.020761 -2023-02-13 18:10:34,085 - Epoch: [118][ 920/ 1207] Overall Loss 0.248556 Objective Loss 0.248556 LR 0.000500 Time 0.020740 -2023-02-13 18:10:34,273 - Epoch: [118][ 930/ 1207] Overall Loss 0.248514 Objective Loss 0.248514 LR 0.000500 Time 0.020720 -2023-02-13 18:10:34,461 - Epoch: [118][ 940/ 1207] Overall Loss 0.248521 Objective Loss 0.248521 LR 0.000500 Time 0.020699 -2023-02-13 18:10:34,649 - Epoch: [118][ 950/ 1207] Overall Loss 0.248414 Objective Loss 0.248414 LR 0.000500 Time 0.020679 -2023-02-13 18:10:34,838 - Epoch: [118][ 960/ 1207] Overall Loss 0.248515 Objective Loss 0.248515 LR 0.000500 Time 0.020659 -2023-02-13 18:10:35,026 - Epoch: [118][ 970/ 1207] Overall Loss 0.248537 Objective Loss 0.248537 LR 0.000500 Time 0.020640 -2023-02-13 18:10:35,214 - Epoch: [118][ 980/ 1207] Overall Loss 0.248640 Objective Loss 0.248640 LR 0.000500 Time 0.020620 -2023-02-13 18:10:35,402 - Epoch: [118][ 990/ 1207] Overall Loss 0.248775 Objective Loss 0.248775 LR 0.000500 Time 0.020602 -2023-02-13 18:10:35,591 - Epoch: [118][ 1000/ 1207] Overall Loss 0.248879 Objective Loss 0.248879 LR 0.000500 Time 0.020584 -2023-02-13 18:10:35,779 - Epoch: [118][ 1010/ 1207] Overall Loss 0.248739 Objective Loss 0.248739 LR 0.000500 Time 0.020567 -2023-02-13 18:10:35,970 - Epoch: [118][ 1020/ 1207] Overall Loss 0.248515 Objective Loss 0.248515 LR 0.000500 Time 0.020551 -2023-02-13 18:10:36,159 - Epoch: [118][ 1030/ 1207] Overall Loss 0.248479 Objective Loss 0.248479 LR 0.000500 Time 0.020535 -2023-02-13 18:10:36,347 - Epoch: [118][ 1040/ 1207] Overall Loss 0.248422 Objective Loss 0.248422 LR 0.000500 Time 0.020518 -2023-02-13 18:10:36,535 - Epoch: [118][ 1050/ 1207] Overall Loss 0.248535 Objective Loss 0.248535 LR 0.000500 Time 0.020502 -2023-02-13 18:10:36,724 - Epoch: [118][ 1060/ 1207] Overall Loss 0.248844 Objective Loss 0.248844 LR 0.000500 Time 0.020486 -2023-02-13 18:10:36,913 - Epoch: [118][ 1070/ 1207] Overall Loss 0.248721 Objective Loss 0.248721 LR 0.000500 Time 0.020471 -2023-02-13 18:10:37,101 - Epoch: [118][ 1080/ 1207] Overall Loss 0.248604 Objective Loss 0.248604 LR 0.000500 Time 0.020455 -2023-02-13 18:10:37,290 - Epoch: [118][ 1090/ 1207] Overall Loss 0.248783 Objective Loss 0.248783 LR 0.000500 Time 0.020440 -2023-02-13 18:10:37,478 - Epoch: [118][ 1100/ 1207] Overall Loss 0.248734 Objective Loss 0.248734 LR 0.000500 Time 0.020425 -2023-02-13 18:10:37,666 - Epoch: [118][ 1110/ 1207] Overall Loss 0.248913 Objective Loss 0.248913 LR 0.000500 Time 0.020410 -2023-02-13 18:10:37,854 - Epoch: [118][ 1120/ 1207] Overall Loss 0.248940 Objective Loss 0.248940 LR 0.000500 Time 0.020395 -2023-02-13 18:10:38,043 - Epoch: [118][ 1130/ 1207] Overall Loss 0.249059 Objective Loss 0.249059 LR 0.000500 Time 0.020382 -2023-02-13 18:10:38,231 - Epoch: [118][ 1140/ 1207] Overall Loss 0.249133 Objective Loss 0.249133 LR 0.000500 Time 0.020368 -2023-02-13 18:10:38,420 - Epoch: [118][ 1150/ 1207] Overall Loss 0.249260 Objective Loss 0.249260 LR 0.000500 Time 0.020354 -2023-02-13 18:10:38,608 - Epoch: [118][ 1160/ 1207] Overall Loss 0.249223 Objective Loss 0.249223 LR 0.000500 Time 0.020340 -2023-02-13 18:10:38,797 - Epoch: [118][ 1170/ 1207] Overall Loss 0.249243 Objective Loss 0.249243 LR 0.000500 Time 0.020328 -2023-02-13 18:10:38,986 - Epoch: [118][ 1180/ 1207] Overall Loss 0.249260 Objective Loss 0.249260 LR 0.000500 Time 0.020316 -2023-02-13 18:10:39,175 - Epoch: [118][ 1190/ 1207] Overall Loss 0.249365 Objective Loss 0.249365 LR 0.000500 Time 0.020303 -2023-02-13 18:10:39,420 - Epoch: [118][ 1200/ 1207] Overall Loss 0.249432 Objective Loss 0.249432 LR 0.000500 Time 0.020338 -2023-02-13 18:10:39,536 - Epoch: [118][ 1207/ 1207] Overall Loss 0.249256 Objective Loss 0.249256 Top1 88.109756 Top5 98.170732 LR 0.000500 Time 0.020316 -2023-02-13 18:10:39,608 - --- validate (epoch=118)----------- -2023-02-13 18:10:39,608 - 34311 samples (256 per mini-batch) -2023-02-13 18:10:40,010 - Epoch: [118][ 10/ 135] Loss 0.371245 Top1 83.750000 Top5 97.304688 -2023-02-13 18:10:40,151 - Epoch: [118][ 20/ 135] Loss 0.311074 Top1 85.097656 Top5 97.753906 -2023-02-13 18:10:40,292 - Epoch: [118][ 30/ 135] Loss 0.311077 Top1 84.869792 Top5 97.695312 -2023-02-13 18:10:40,434 - Epoch: [118][ 40/ 135] Loss 0.318210 Top1 84.580078 Top5 97.763672 -2023-02-13 18:10:40,573 - Epoch: [118][ 50/ 135] Loss 0.319373 Top1 84.375000 Top5 97.578125 -2023-02-13 18:10:40,711 - Epoch: [118][ 60/ 135] Loss 0.317110 Top1 84.335938 Top5 97.526042 -2023-02-13 18:10:40,839 - Epoch: [118][ 70/ 135] Loss 0.318519 Top1 84.252232 Top5 97.511161 -2023-02-13 18:10:40,967 - Epoch: [118][ 80/ 135] Loss 0.318373 Top1 84.228516 Top5 97.495117 -2023-02-13 18:10:41,095 - Epoch: [118][ 90/ 135] Loss 0.311978 Top1 84.249132 Top5 97.543403 -2023-02-13 18:10:41,220 - Epoch: [118][ 100/ 135] Loss 0.314274 Top1 84.253906 Top5 97.507812 -2023-02-13 18:10:41,349 - Epoch: [118][ 110/ 135] Loss 0.316509 Top1 84.176136 Top5 97.556818 -2023-02-13 18:10:41,471 - Epoch: [118][ 120/ 135] Loss 0.315530 Top1 84.130859 Top5 97.604167 -2023-02-13 18:10:41,595 - Epoch: [118][ 130/ 135] Loss 0.315691 Top1 84.104567 Top5 97.626202 -2023-02-13 18:10:41,639 - Epoch: [118][ 135/ 135] Loss 0.314588 Top1 84.153770 Top5 97.653814 -2023-02-13 18:10:41,714 - ==> Top1: 84.154 Top5: 97.654 Loss: 0.315 - -2023-02-13 18:10:41,715 - ==> Confusion: -[[ 883 5 5 0 7 3 0 1 5 28 2 3 1 2 4 5 1 3 1 1 7] - [ 3 962 3 3 7 20 2 10 3 0 1 1 1 0 1 2 4 1 2 0 7] - [ 9 4 947 19 3 1 14 15 0 2 2 1 2 4 5 10 2 6 6 2 4] - [ 1 2 18 923 3 3 0 2 2 2 11 0 6 2 10 2 6 5 13 1 4] - [ 15 9 0 1 983 13 1 3 1 2 0 5 0 2 6 6 8 2 0 3 6] - [ 2 24 0 6 8 953 5 19 1 2 4 14 7 12 0 2 4 1 1 1 4] - [ 4 3 18 2 0 2 1030 8 1 1 2 2 3 3 0 6 2 1 1 6 4] - [ 1 8 10 1 3 19 1 942 1 1 0 7 3 1 0 0 1 2 15 5 3] - [ 17 3 0 2 1 0 1 1 901 31 8 4 1 8 22 2 0 1 4 0 2] - [ 99 2 3 0 8 1 0 2 46 810 1 1 0 19 7 2 1 3 1 0 6] - [ 1 1 3 12 2 1 3 4 11 2 978 4 1 6 3 0 1 0 16 0 2] - [ 3 4 2 0 4 12 0 6 0 0 1 905 23 4 2 8 6 10 3 9 3] - [ 1 0 0 7 0 5 0 3 2 0 1 28 863 1 4 7 3 24 2 0 8] - [ 3 3 0 1 10 11 1 4 10 11 10 6 4 928 2 5 4 1 0 0 10] - [ 18 2 1 27 5 2 0 2 12 5 3 1 5 2 977 1 4 8 9 0 8] - [ 3 1 5 0 10 1 4 1 0 0 0 8 8 1 0 975 8 11 0 4 6] - [ 3 3 1 1 8 2 0 1 0 0 0 0 1 5 2 14 1006 2 1 4 7] - [ 3 2 1 3 0 1 1 1 0 0 0 7 18 0 2 15 0 989 0 1 7] - [ 3 4 3 10 1 2 1 23 2 0 2 3 4 0 16 1 0 1 1007 2 1] - [ 0 3 1 0 3 6 7 12 0 0 1 13 5 5 0 7 7 2 0 1068 8] - [ 172 243 233 160 132 196 89 180 87 75 171 126 299 269 157 138 305 130 193 235 9844]] - -2023-02-13 18:10:41,716 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:10:41,717 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:10:41,722 - - -2023-02-13 18:10:41,722 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:10:42,609 - Epoch: [119][ 10/ 1207] Overall Loss 0.243783 Objective Loss 0.243783 LR 0.000500 Time 0.088568 -2023-02-13 18:10:42,810 - Epoch: [119][ 20/ 1207] Overall Loss 0.236878 Objective Loss 0.236878 LR 0.000500 Time 0.054342 -2023-02-13 18:10:43,000 - Epoch: [119][ 30/ 1207] Overall Loss 0.232392 Objective Loss 0.232392 LR 0.000500 Time 0.042550 -2023-02-13 18:10:43,191 - Epoch: [119][ 40/ 1207] Overall Loss 0.238549 Objective Loss 0.238549 LR 0.000500 Time 0.036663 -2023-02-13 18:10:43,381 - Epoch: [119][ 50/ 1207] Overall Loss 0.245154 Objective Loss 0.245154 LR 0.000500 Time 0.033122 -2023-02-13 18:10:43,570 - Epoch: [119][ 60/ 1207] Overall Loss 0.243066 Objective Loss 0.243066 LR 0.000500 Time 0.030757 -2023-02-13 18:10:43,760 - Epoch: [119][ 70/ 1207] Overall Loss 0.246239 Objective Loss 0.246239 LR 0.000500 Time 0.029064 -2023-02-13 18:10:43,949 - Epoch: [119][ 80/ 1207] Overall Loss 0.248033 Objective Loss 0.248033 LR 0.000500 Time 0.027791 -2023-02-13 18:10:44,139 - Epoch: [119][ 90/ 1207] Overall Loss 0.244943 Objective Loss 0.244943 LR 0.000500 Time 0.026812 -2023-02-13 18:10:44,329 - Epoch: [119][ 100/ 1207] Overall Loss 0.242932 Objective Loss 0.242932 LR 0.000500 Time 0.026021 -2023-02-13 18:10:44,519 - Epoch: [119][ 110/ 1207] Overall Loss 0.244196 Objective Loss 0.244196 LR 0.000500 Time 0.025379 -2023-02-13 18:10:44,708 - Epoch: [119][ 120/ 1207] Overall Loss 0.243126 Objective Loss 0.243126 LR 0.000500 Time 0.024837 -2023-02-13 18:10:44,897 - Epoch: [119][ 130/ 1207] Overall Loss 0.245650 Objective Loss 0.245650 LR 0.000500 Time 0.024383 -2023-02-13 18:10:45,087 - Epoch: [119][ 140/ 1207] Overall Loss 0.245020 Objective Loss 0.245020 LR 0.000500 Time 0.023995 -2023-02-13 18:10:45,277 - Epoch: [119][ 150/ 1207] Overall Loss 0.244184 Objective Loss 0.244184 LR 0.000500 Time 0.023655 -2023-02-13 18:10:45,466 - Epoch: [119][ 160/ 1207] Overall Loss 0.244049 Objective Loss 0.244049 LR 0.000500 Time 0.023360 -2023-02-13 18:10:45,657 - Epoch: [119][ 170/ 1207] Overall Loss 0.245072 Objective Loss 0.245072 LR 0.000500 Time 0.023102 -2023-02-13 18:10:45,847 - Epoch: [119][ 180/ 1207] Overall Loss 0.243989 Objective Loss 0.243989 LR 0.000500 Time 0.022877 -2023-02-13 18:10:46,038 - Epoch: [119][ 190/ 1207] Overall Loss 0.244406 Objective Loss 0.244406 LR 0.000500 Time 0.022673 -2023-02-13 18:10:46,227 - Epoch: [119][ 200/ 1207] Overall Loss 0.244996 Objective Loss 0.244996 LR 0.000500 Time 0.022484 -2023-02-13 18:10:46,417 - Epoch: [119][ 210/ 1207] Overall Loss 0.244617 Objective Loss 0.244617 LR 0.000500 Time 0.022314 -2023-02-13 18:10:46,606 - Epoch: [119][ 220/ 1207] Overall Loss 0.243494 Objective Loss 0.243494 LR 0.000500 Time 0.022158 -2023-02-13 18:10:46,796 - Epoch: [119][ 230/ 1207] Overall Loss 0.245155 Objective Loss 0.245155 LR 0.000500 Time 0.022020 -2023-02-13 18:10:46,986 - Epoch: [119][ 240/ 1207] Overall Loss 0.244119 Objective Loss 0.244119 LR 0.000500 Time 0.021894 -2023-02-13 18:10:47,177 - Epoch: [119][ 250/ 1207] Overall Loss 0.244919 Objective Loss 0.244919 LR 0.000500 Time 0.021778 -2023-02-13 18:10:47,366 - Epoch: [119][ 260/ 1207] Overall Loss 0.244764 Objective Loss 0.244764 LR 0.000500 Time 0.021668 -2023-02-13 18:10:47,556 - Epoch: [119][ 270/ 1207] Overall Loss 0.243769 Objective Loss 0.243769 LR 0.000500 Time 0.021566 -2023-02-13 18:10:47,746 - Epoch: [119][ 280/ 1207] Overall Loss 0.244409 Objective Loss 0.244409 LR 0.000500 Time 0.021473 -2023-02-13 18:10:47,935 - Epoch: [119][ 290/ 1207] Overall Loss 0.244696 Objective Loss 0.244696 LR 0.000500 Time 0.021384 -2023-02-13 18:10:48,126 - Epoch: [119][ 300/ 1207] Overall Loss 0.245756 Objective Loss 0.245756 LR 0.000500 Time 0.021305 -2023-02-13 18:10:48,316 - Epoch: [119][ 310/ 1207] Overall Loss 0.244655 Objective Loss 0.244655 LR 0.000500 Time 0.021231 -2023-02-13 18:10:48,506 - Epoch: [119][ 320/ 1207] Overall Loss 0.245960 Objective Loss 0.245960 LR 0.000500 Time 0.021159 -2023-02-13 18:10:48,696 - Epoch: [119][ 330/ 1207] Overall Loss 0.245988 Objective Loss 0.245988 LR 0.000500 Time 0.021092 -2023-02-13 18:10:48,885 - Epoch: [119][ 340/ 1207] Overall Loss 0.245757 Objective Loss 0.245757 LR 0.000500 Time 0.021029 -2023-02-13 18:10:49,076 - Epoch: [119][ 350/ 1207] Overall Loss 0.246040 Objective Loss 0.246040 LR 0.000500 Time 0.020970 -2023-02-13 18:10:49,265 - Epoch: [119][ 360/ 1207] Overall Loss 0.246907 Objective Loss 0.246907 LR 0.000500 Time 0.020914 -2023-02-13 18:10:49,455 - Epoch: [119][ 370/ 1207] Overall Loss 0.246938 Objective Loss 0.246938 LR 0.000500 Time 0.020859 -2023-02-13 18:10:49,645 - Epoch: [119][ 380/ 1207] Overall Loss 0.246566 Objective Loss 0.246566 LR 0.000500 Time 0.020809 -2023-02-13 18:10:49,835 - Epoch: [119][ 390/ 1207] Overall Loss 0.246757 Objective Loss 0.246757 LR 0.000500 Time 0.020763 -2023-02-13 18:10:50,024 - Epoch: [119][ 400/ 1207] Overall Loss 0.246473 Objective Loss 0.246473 LR 0.000500 Time 0.020716 -2023-02-13 18:10:50,214 - Epoch: [119][ 410/ 1207] Overall Loss 0.246588 Objective Loss 0.246588 LR 0.000500 Time 0.020674 -2023-02-13 18:10:50,404 - Epoch: [119][ 420/ 1207] Overall Loss 0.246836 Objective Loss 0.246836 LR 0.000500 Time 0.020631 -2023-02-13 18:10:50,593 - Epoch: [119][ 430/ 1207] Overall Loss 0.246651 Objective Loss 0.246651 LR 0.000500 Time 0.020591 -2023-02-13 18:10:50,784 - Epoch: [119][ 440/ 1207] Overall Loss 0.246519 Objective Loss 0.246519 LR 0.000500 Time 0.020555 -2023-02-13 18:10:50,976 - Epoch: [119][ 450/ 1207] Overall Loss 0.246007 Objective Loss 0.246007 LR 0.000500 Time 0.020526 -2023-02-13 18:10:51,166 - Epoch: [119][ 460/ 1207] Overall Loss 0.246815 Objective Loss 0.246815 LR 0.000500 Time 0.020491 -2023-02-13 18:10:51,356 - Epoch: [119][ 470/ 1207] Overall Loss 0.246895 Objective Loss 0.246895 LR 0.000500 Time 0.020458 -2023-02-13 18:10:51,546 - Epoch: [119][ 480/ 1207] Overall Loss 0.246602 Objective Loss 0.246602 LR 0.000500 Time 0.020427 -2023-02-13 18:10:51,736 - Epoch: [119][ 490/ 1207] Overall Loss 0.246742 Objective Loss 0.246742 LR 0.000500 Time 0.020397 -2023-02-13 18:10:51,927 - Epoch: [119][ 500/ 1207] Overall Loss 0.246736 Objective Loss 0.246736 LR 0.000500 Time 0.020370 -2023-02-13 18:10:52,119 - Epoch: [119][ 510/ 1207] Overall Loss 0.247153 Objective Loss 0.247153 LR 0.000500 Time 0.020346 -2023-02-13 18:10:52,310 - Epoch: [119][ 520/ 1207] Overall Loss 0.247271 Objective Loss 0.247271 LR 0.000500 Time 0.020323 -2023-02-13 18:10:52,503 - Epoch: [119][ 530/ 1207] Overall Loss 0.247282 Objective Loss 0.247282 LR 0.000500 Time 0.020302 -2023-02-13 18:10:52,695 - Epoch: [119][ 540/ 1207] Overall Loss 0.247191 Objective Loss 0.247191 LR 0.000500 Time 0.020281 -2023-02-13 18:10:52,888 - Epoch: [119][ 550/ 1207] Overall Loss 0.247235 Objective Loss 0.247235 LR 0.000500 Time 0.020263 -2023-02-13 18:10:53,080 - Epoch: [119][ 560/ 1207] Overall Loss 0.247181 Objective Loss 0.247181 LR 0.000500 Time 0.020244 -2023-02-13 18:10:53,271 - Epoch: [119][ 570/ 1207] Overall Loss 0.246814 Objective Loss 0.246814 LR 0.000500 Time 0.020222 -2023-02-13 18:10:53,461 - Epoch: [119][ 580/ 1207] Overall Loss 0.246794 Objective Loss 0.246794 LR 0.000500 Time 0.020200 -2023-02-13 18:10:53,650 - Epoch: [119][ 590/ 1207] Overall Loss 0.246893 Objective Loss 0.246893 LR 0.000500 Time 0.020179 -2023-02-13 18:10:53,840 - Epoch: [119][ 600/ 1207] Overall Loss 0.247152 Objective Loss 0.247152 LR 0.000500 Time 0.020158 -2023-02-13 18:10:54,031 - Epoch: [119][ 610/ 1207] Overall Loss 0.247356 Objective Loss 0.247356 LR 0.000500 Time 0.020140 -2023-02-13 18:10:54,222 - Epoch: [119][ 620/ 1207] Overall Loss 0.247533 Objective Loss 0.247533 LR 0.000500 Time 0.020122 -2023-02-13 18:10:54,412 - Epoch: [119][ 630/ 1207] Overall Loss 0.247721 Objective Loss 0.247721 LR 0.000500 Time 0.020104 -2023-02-13 18:10:54,603 - Epoch: [119][ 640/ 1207] Overall Loss 0.248392 Objective Loss 0.248392 LR 0.000500 Time 0.020088 -2023-02-13 18:10:54,795 - Epoch: [119][ 650/ 1207] Overall Loss 0.248327 Objective Loss 0.248327 LR 0.000500 Time 0.020074 -2023-02-13 18:10:54,988 - Epoch: [119][ 660/ 1207] Overall Loss 0.248310 Objective Loss 0.248310 LR 0.000500 Time 0.020061 -2023-02-13 18:10:55,181 - Epoch: [119][ 670/ 1207] Overall Loss 0.248228 Objective Loss 0.248228 LR 0.000500 Time 0.020049 -2023-02-13 18:10:55,372 - Epoch: [119][ 680/ 1207] Overall Loss 0.248518 Objective Loss 0.248518 LR 0.000500 Time 0.020035 -2023-02-13 18:10:55,566 - Epoch: [119][ 690/ 1207] Overall Loss 0.248605 Objective Loss 0.248605 LR 0.000500 Time 0.020024 -2023-02-13 18:10:55,757 - Epoch: [119][ 700/ 1207] Overall Loss 0.248747 Objective Loss 0.248747 LR 0.000500 Time 0.020012 -2023-02-13 18:10:55,952 - Epoch: [119][ 710/ 1207] Overall Loss 0.248686 Objective Loss 0.248686 LR 0.000500 Time 0.020003 -2023-02-13 18:10:56,145 - Epoch: [119][ 720/ 1207] Overall Loss 0.248693 Objective Loss 0.248693 LR 0.000500 Time 0.019993 -2023-02-13 18:10:56,338 - Epoch: [119][ 730/ 1207] Overall Loss 0.248542 Objective Loss 0.248542 LR 0.000500 Time 0.019982 -2023-02-13 18:10:56,529 - Epoch: [119][ 740/ 1207] Overall Loss 0.248344 Objective Loss 0.248344 LR 0.000500 Time 0.019971 -2023-02-13 18:10:56,722 - Epoch: [119][ 750/ 1207] Overall Loss 0.248496 Objective Loss 0.248496 LR 0.000500 Time 0.019962 -2023-02-13 18:10:56,914 - Epoch: [119][ 760/ 1207] Overall Loss 0.248334 Objective Loss 0.248334 LR 0.000500 Time 0.019951 -2023-02-13 18:10:57,108 - Epoch: [119][ 770/ 1207] Overall Loss 0.248862 Objective Loss 0.248862 LR 0.000500 Time 0.019943 -2023-02-13 18:10:57,299 - Epoch: [119][ 780/ 1207] Overall Loss 0.248903 Objective Loss 0.248903 LR 0.000500 Time 0.019931 -2023-02-13 18:10:57,489 - Epoch: [119][ 790/ 1207] Overall Loss 0.248796 Objective Loss 0.248796 LR 0.000500 Time 0.019920 -2023-02-13 18:10:57,680 - Epoch: [119][ 800/ 1207] Overall Loss 0.248752 Objective Loss 0.248752 LR 0.000500 Time 0.019908 -2023-02-13 18:10:57,871 - Epoch: [119][ 810/ 1207] Overall Loss 0.248839 Objective Loss 0.248839 LR 0.000500 Time 0.019898 -2023-02-13 18:10:58,061 - Epoch: [119][ 820/ 1207] Overall Loss 0.249221 Objective Loss 0.249221 LR 0.000500 Time 0.019886 -2023-02-13 18:10:58,252 - Epoch: [119][ 830/ 1207] Overall Loss 0.249266 Objective Loss 0.249266 LR 0.000500 Time 0.019877 -2023-02-13 18:10:58,442 - Epoch: [119][ 840/ 1207] Overall Loss 0.249223 Objective Loss 0.249223 LR 0.000500 Time 0.019866 -2023-02-13 18:10:58,632 - Epoch: [119][ 850/ 1207] Overall Loss 0.249193 Objective Loss 0.249193 LR 0.000500 Time 0.019856 -2023-02-13 18:10:58,823 - Epoch: [119][ 860/ 1207] Overall Loss 0.249565 Objective Loss 0.249565 LR 0.000500 Time 0.019846 -2023-02-13 18:10:59,014 - Epoch: [119][ 870/ 1207] Overall Loss 0.249650 Objective Loss 0.249650 LR 0.000500 Time 0.019837 -2023-02-13 18:10:59,205 - Epoch: [119][ 880/ 1207] Overall Loss 0.249441 Objective Loss 0.249441 LR 0.000500 Time 0.019828 -2023-02-13 18:10:59,396 - Epoch: [119][ 890/ 1207] Overall Loss 0.249686 Objective Loss 0.249686 LR 0.000500 Time 0.019819 -2023-02-13 18:10:59,587 - Epoch: [119][ 900/ 1207] Overall Loss 0.249541 Objective Loss 0.249541 LR 0.000500 Time 0.019811 -2023-02-13 18:10:59,777 - Epoch: [119][ 910/ 1207] Overall Loss 0.249173 Objective Loss 0.249173 LR 0.000500 Time 0.019802 -2023-02-13 18:10:59,967 - Epoch: [119][ 920/ 1207] Overall Loss 0.249033 Objective Loss 0.249033 LR 0.000500 Time 0.019793 -2023-02-13 18:11:00,158 - Epoch: [119][ 930/ 1207] Overall Loss 0.248809 Objective Loss 0.248809 LR 0.000500 Time 0.019785 -2023-02-13 18:11:00,348 - Epoch: [119][ 940/ 1207] Overall Loss 0.248750 Objective Loss 0.248750 LR 0.000500 Time 0.019776 -2023-02-13 18:11:00,538 - Epoch: [119][ 950/ 1207] Overall Loss 0.248973 Objective Loss 0.248973 LR 0.000500 Time 0.019768 -2023-02-13 18:11:00,729 - Epoch: [119][ 960/ 1207] Overall Loss 0.248932 Objective Loss 0.248932 LR 0.000500 Time 0.019760 -2023-02-13 18:11:00,922 - Epoch: [119][ 970/ 1207] Overall Loss 0.249024 Objective Loss 0.249024 LR 0.000500 Time 0.019755 -2023-02-13 18:11:01,113 - Epoch: [119][ 980/ 1207] Overall Loss 0.249175 Objective Loss 0.249175 LR 0.000500 Time 0.019748 -2023-02-13 18:11:01,303 - Epoch: [119][ 990/ 1207] Overall Loss 0.249374 Objective Loss 0.249374 LR 0.000500 Time 0.019740 -2023-02-13 18:11:01,493 - Epoch: [119][ 1000/ 1207] Overall Loss 0.249410 Objective Loss 0.249410 LR 0.000500 Time 0.019732 -2023-02-13 18:11:01,684 - Epoch: [119][ 1010/ 1207] Overall Loss 0.249261 Objective Loss 0.249261 LR 0.000500 Time 0.019725 -2023-02-13 18:11:01,875 - Epoch: [119][ 1020/ 1207] Overall Loss 0.248972 Objective Loss 0.248972 LR 0.000500 Time 0.019719 -2023-02-13 18:11:02,066 - Epoch: [119][ 1030/ 1207] Overall Loss 0.249137 Objective Loss 0.249137 LR 0.000500 Time 0.019712 -2023-02-13 18:11:02,256 - Epoch: [119][ 1040/ 1207] Overall Loss 0.249370 Objective Loss 0.249370 LR 0.000500 Time 0.019706 -2023-02-13 18:11:02,447 - Epoch: [119][ 1050/ 1207] Overall Loss 0.249349 Objective Loss 0.249349 LR 0.000500 Time 0.019700 -2023-02-13 18:11:02,637 - Epoch: [119][ 1060/ 1207] Overall Loss 0.249146 Objective Loss 0.249146 LR 0.000500 Time 0.019693 -2023-02-13 18:11:02,828 - Epoch: [119][ 1070/ 1207] Overall Loss 0.249078 Objective Loss 0.249078 LR 0.000500 Time 0.019687 -2023-02-13 18:11:03,018 - Epoch: [119][ 1080/ 1207] Overall Loss 0.248842 Objective Loss 0.248842 LR 0.000500 Time 0.019680 -2023-02-13 18:11:03,209 - Epoch: [119][ 1090/ 1207] Overall Loss 0.248702 Objective Loss 0.248702 LR 0.000500 Time 0.019675 -2023-02-13 18:11:03,400 - Epoch: [119][ 1100/ 1207] Overall Loss 0.248575 Objective Loss 0.248575 LR 0.000500 Time 0.019669 -2023-02-13 18:11:03,591 - Epoch: [119][ 1110/ 1207] Overall Loss 0.248572 Objective Loss 0.248572 LR 0.000500 Time 0.019663 -2023-02-13 18:11:03,781 - Epoch: [119][ 1120/ 1207] Overall Loss 0.248387 Objective Loss 0.248387 LR 0.000500 Time 0.019657 -2023-02-13 18:11:03,972 - Epoch: [119][ 1130/ 1207] Overall Loss 0.248346 Objective Loss 0.248346 LR 0.000500 Time 0.019651 -2023-02-13 18:11:04,163 - Epoch: [119][ 1140/ 1207] Overall Loss 0.248382 Objective Loss 0.248382 LR 0.000500 Time 0.019646 -2023-02-13 18:11:04,354 - Epoch: [119][ 1150/ 1207] Overall Loss 0.248186 Objective Loss 0.248186 LR 0.000500 Time 0.019641 -2023-02-13 18:11:04,544 - Epoch: [119][ 1160/ 1207] Overall Loss 0.248048 Objective Loss 0.248048 LR 0.000500 Time 0.019636 -2023-02-13 18:11:04,734 - Epoch: [119][ 1170/ 1207] Overall Loss 0.247961 Objective Loss 0.247961 LR 0.000500 Time 0.019630 -2023-02-13 18:11:04,925 - Epoch: [119][ 1180/ 1207] Overall Loss 0.247906 Objective Loss 0.247906 LR 0.000500 Time 0.019625 -2023-02-13 18:11:05,115 - Epoch: [119][ 1190/ 1207] Overall Loss 0.248023 Objective Loss 0.248023 LR 0.000500 Time 0.019620 -2023-02-13 18:11:05,362 - Epoch: [119][ 1200/ 1207] Overall Loss 0.247852 Objective Loss 0.247852 LR 0.000500 Time 0.019662 -2023-02-13 18:11:05,478 - Epoch: [119][ 1207/ 1207] Overall Loss 0.247659 Objective Loss 0.247659 Top1 86.280488 Top5 99.085366 LR 0.000500 Time 0.019643 -2023-02-13 18:11:05,548 - --- validate (epoch=119)----------- -2023-02-13 18:11:05,548 - 34311 samples (256 per mini-batch) -2023-02-13 18:11:05,946 - Epoch: [119][ 10/ 135] Loss 0.361850 Top1 81.875000 Top5 97.421875 -2023-02-13 18:11:06,083 - Epoch: [119][ 20/ 135] Loss 0.340170 Top1 82.597656 Top5 97.285156 -2023-02-13 18:11:06,227 - Epoch: [119][ 30/ 135] Loss 0.343448 Top1 82.812500 Top5 97.343750 -2023-02-13 18:11:06,365 - Epoch: [119][ 40/ 135] Loss 0.340742 Top1 83.125000 Top5 97.314453 -2023-02-13 18:11:06,510 - Epoch: [119][ 50/ 135] Loss 0.333438 Top1 83.226562 Top5 97.359375 -2023-02-13 18:11:06,640 - Epoch: [119][ 60/ 135] Loss 0.329524 Top1 83.294271 Top5 97.473958 -2023-02-13 18:11:06,770 - Epoch: [119][ 70/ 135] Loss 0.328263 Top1 83.381696 Top5 97.555804 -2023-02-13 18:11:06,901 - Epoch: [119][ 80/ 135] Loss 0.324921 Top1 83.505859 Top5 97.539062 -2023-02-13 18:11:07,031 - Epoch: [119][ 90/ 135] Loss 0.323312 Top1 83.554688 Top5 97.552083 -2023-02-13 18:11:07,158 - Epoch: [119][ 100/ 135] Loss 0.325771 Top1 83.468750 Top5 97.484375 -2023-02-13 18:11:07,286 - Epoch: [119][ 110/ 135] Loss 0.324218 Top1 83.519176 Top5 97.503551 -2023-02-13 18:11:07,414 - Epoch: [119][ 120/ 135] Loss 0.324511 Top1 83.450521 Top5 97.477214 -2023-02-13 18:11:07,547 - Epoch: [119][ 130/ 135] Loss 0.322537 Top1 83.539663 Top5 97.521034 -2023-02-13 18:11:07,593 - Epoch: [119][ 135/ 135] Loss 0.320351 Top1 83.565037 Top5 97.545977 -2023-02-13 18:11:07,677 - ==> Top1: 83.565 Top5: 97.546 Loss: 0.320 - -2023-02-13 18:11:07,678 - ==> Confusion: -[[ 877 5 7 1 10 2 0 0 3 33 0 4 3 4 3 4 0 1 1 2 7] - [ 2 939 0 3 18 17 3 18 6 1 0 1 2 0 1 1 6 1 4 3 7] - [ 7 2 961 8 9 0 13 16 0 0 2 2 3 6 5 4 3 3 6 5 3] - [ 7 1 22 897 4 1 2 2 1 3 13 0 5 2 28 1 6 4 14 1 2] - [ 15 8 0 1 998 6 1 0 0 4 0 6 2 4 6 7 2 1 0 3 2] - [ 1 27 0 5 13 940 3 25 2 3 3 15 2 11 1 3 3 1 1 4 7] - [ 4 8 21 2 1 4 1031 5 2 0 2 2 2 1 0 1 0 1 2 7 3] - [ 1 15 7 1 5 22 1 930 1 1 1 2 3 1 0 0 0 2 18 9 4] - [ 23 4 1 2 3 1 0 2 879 52 6 1 1 7 20 1 0 0 4 1 1] - [ 95 2 3 0 9 1 0 2 30 838 0 3 0 18 5 1 1 1 0 2 1] - [ 4 3 7 9 3 4 3 3 16 1 971 2 1 8 2 1 2 0 9 0 2] - [ 1 5 1 0 8 10 1 4 3 0 0 913 22 8 0 5 2 8 2 10 2] - [ 0 0 0 8 6 5 0 2 3 0 2 32 851 0 5 9 4 20 1 2 9] - [ 6 2 2 0 10 6 2 1 10 16 8 5 1 940 2 4 3 1 0 0 5] - [ 8 1 2 15 9 3 0 1 18 8 0 1 2 2 1002 1 0 6 5 1 7] - [ 4 3 8 0 9 0 4 2 0 0 0 8 6 5 0 964 14 9 0 5 5] - [ 3 5 2 1 13 2 1 1 2 1 0 0 2 3 3 11 997 1 2 4 7] - [ 4 2 1 8 2 3 5 0 2 0 2 9 12 0 2 20 0 973 1 1 4] - [ 5 10 4 11 1 1 0 20 5 0 4 3 3 0 15 1 1 2 995 4 1] - [ 0 4 1 0 1 4 7 11 1 0 1 15 3 5 0 4 7 1 1 1078 4] - [ 211 247 281 121 203 198 91 154 94 94 140 102 290 337 197 119 305 102 169 281 9698]] - -2023-02-13 18:11:07,679 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:11:07,679 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:11:07,685 - - -2023-02-13 18:11:07,685 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:11:08,550 - Epoch: [120][ 10/ 1207] Overall Loss 0.251660 Objective Loss 0.251660 LR 0.000500 Time 0.086435 -2023-02-13 18:11:08,746 - Epoch: [120][ 20/ 1207] Overall Loss 0.242468 Objective Loss 0.242468 LR 0.000500 Time 0.052970 -2023-02-13 18:11:08,934 - Epoch: [120][ 30/ 1207] Overall Loss 0.242456 Objective Loss 0.242456 LR 0.000500 Time 0.041589 -2023-02-13 18:11:09,123 - Epoch: [120][ 40/ 1207] Overall Loss 0.242597 Objective Loss 0.242597 LR 0.000500 Time 0.035908 -2023-02-13 18:11:09,312 - Epoch: [120][ 50/ 1207] Overall Loss 0.245480 Objective Loss 0.245480 LR 0.000500 Time 0.032492 -2023-02-13 18:11:09,500 - Epoch: [120][ 60/ 1207] Overall Loss 0.241549 Objective Loss 0.241549 LR 0.000500 Time 0.030207 -2023-02-13 18:11:09,689 - Epoch: [120][ 70/ 1207] Overall Loss 0.241360 Objective Loss 0.241360 LR 0.000500 Time 0.028584 -2023-02-13 18:11:09,877 - Epoch: [120][ 80/ 1207] Overall Loss 0.241906 Objective Loss 0.241906 LR 0.000500 Time 0.027363 -2023-02-13 18:11:10,066 - Epoch: [120][ 90/ 1207] Overall Loss 0.241042 Objective Loss 0.241042 LR 0.000500 Time 0.026411 -2023-02-13 18:11:10,254 - Epoch: [120][ 100/ 1207] Overall Loss 0.239884 Objective Loss 0.239884 LR 0.000500 Time 0.025652 -2023-02-13 18:11:10,443 - Epoch: [120][ 110/ 1207] Overall Loss 0.238896 Objective Loss 0.238896 LR 0.000500 Time 0.025030 -2023-02-13 18:11:10,631 - Epoch: [120][ 120/ 1207] Overall Loss 0.236566 Objective Loss 0.236566 LR 0.000500 Time 0.024513 -2023-02-13 18:11:10,820 - Epoch: [120][ 130/ 1207] Overall Loss 0.238759 Objective Loss 0.238759 LR 0.000500 Time 0.024072 -2023-02-13 18:11:11,009 - Epoch: [120][ 140/ 1207] Overall Loss 0.239728 Objective Loss 0.239728 LR 0.000500 Time 0.023701 -2023-02-13 18:11:11,197 - Epoch: [120][ 150/ 1207] Overall Loss 0.240847 Objective Loss 0.240847 LR 0.000500 Time 0.023376 -2023-02-13 18:11:11,385 - Epoch: [120][ 160/ 1207] Overall Loss 0.240592 Objective Loss 0.240592 LR 0.000500 Time 0.023087 -2023-02-13 18:11:11,573 - Epoch: [120][ 170/ 1207] Overall Loss 0.240124 Objective Loss 0.240124 LR 0.000500 Time 0.022833 -2023-02-13 18:11:11,762 - Epoch: [120][ 180/ 1207] Overall Loss 0.240888 Objective Loss 0.240888 LR 0.000500 Time 0.022609 -2023-02-13 18:11:11,951 - Epoch: [120][ 190/ 1207] Overall Loss 0.240782 Objective Loss 0.240782 LR 0.000500 Time 0.022412 -2023-02-13 18:11:12,140 - Epoch: [120][ 200/ 1207] Overall Loss 0.240482 Objective Loss 0.240482 LR 0.000500 Time 0.022234 -2023-02-13 18:11:12,327 - Epoch: [120][ 210/ 1207] Overall Loss 0.240929 Objective Loss 0.240929 LR 0.000500 Time 0.022067 -2023-02-13 18:11:12,515 - Epoch: [120][ 220/ 1207] Overall Loss 0.240206 Objective Loss 0.240206 LR 0.000500 Time 0.021917 -2023-02-13 18:11:12,704 - Epoch: [120][ 230/ 1207] Overall Loss 0.240526 Objective Loss 0.240526 LR 0.000500 Time 0.021783 -2023-02-13 18:11:12,892 - Epoch: [120][ 240/ 1207] Overall Loss 0.241496 Objective Loss 0.241496 LR 0.000500 Time 0.021658 -2023-02-13 18:11:13,080 - Epoch: [120][ 250/ 1207] Overall Loss 0.240761 Objective Loss 0.240761 LR 0.000500 Time 0.021541 -2023-02-13 18:11:13,269 - Epoch: [120][ 260/ 1207] Overall Loss 0.240524 Objective Loss 0.240524 LR 0.000500 Time 0.021437 -2023-02-13 18:11:13,456 - Epoch: [120][ 270/ 1207] Overall Loss 0.240440 Objective Loss 0.240440 LR 0.000500 Time 0.021336 -2023-02-13 18:11:13,644 - Epoch: [120][ 280/ 1207] Overall Loss 0.239707 Objective Loss 0.239707 LR 0.000500 Time 0.021243 -2023-02-13 18:11:13,832 - Epoch: [120][ 290/ 1207] Overall Loss 0.239654 Objective Loss 0.239654 LR 0.000500 Time 0.021158 -2023-02-13 18:11:14,020 - Epoch: [120][ 300/ 1207] Overall Loss 0.239962 Objective Loss 0.239962 LR 0.000500 Time 0.021078 -2023-02-13 18:11:14,209 - Epoch: [120][ 310/ 1207] Overall Loss 0.240648 Objective Loss 0.240648 LR 0.000500 Time 0.021005 -2023-02-13 18:11:14,397 - Epoch: [120][ 320/ 1207] Overall Loss 0.240254 Objective Loss 0.240254 LR 0.000500 Time 0.020935 -2023-02-13 18:11:14,584 - Epoch: [120][ 330/ 1207] Overall Loss 0.240411 Objective Loss 0.240411 LR 0.000500 Time 0.020868 -2023-02-13 18:11:14,772 - Epoch: [120][ 340/ 1207] Overall Loss 0.241033 Objective Loss 0.241033 LR 0.000500 Time 0.020807 -2023-02-13 18:11:14,961 - Epoch: [120][ 350/ 1207] Overall Loss 0.240763 Objective Loss 0.240763 LR 0.000500 Time 0.020749 -2023-02-13 18:11:15,150 - Epoch: [120][ 360/ 1207] Overall Loss 0.241229 Objective Loss 0.241229 LR 0.000500 Time 0.020696 -2023-02-13 18:11:15,337 - Epoch: [120][ 370/ 1207] Overall Loss 0.241349 Objective Loss 0.241349 LR 0.000500 Time 0.020643 -2023-02-13 18:11:15,525 - Epoch: [120][ 380/ 1207] Overall Loss 0.242186 Objective Loss 0.242186 LR 0.000500 Time 0.020594 -2023-02-13 18:11:15,713 - Epoch: [120][ 390/ 1207] Overall Loss 0.241764 Objective Loss 0.241764 LR 0.000500 Time 0.020547 -2023-02-13 18:11:15,904 - Epoch: [120][ 400/ 1207] Overall Loss 0.242461 Objective Loss 0.242461 LR 0.000500 Time 0.020510 -2023-02-13 18:11:16,093 - Epoch: [120][ 410/ 1207] Overall Loss 0.242239 Objective Loss 0.242239 LR 0.000500 Time 0.020468 -2023-02-13 18:11:16,281 - Epoch: [120][ 420/ 1207] Overall Loss 0.242803 Objective Loss 0.242803 LR 0.000500 Time 0.020429 -2023-02-13 18:11:16,469 - Epoch: [120][ 430/ 1207] Overall Loss 0.242677 Objective Loss 0.242677 LR 0.000500 Time 0.020389 -2023-02-13 18:11:16,657 - Epoch: [120][ 440/ 1207] Overall Loss 0.242706 Objective Loss 0.242706 LR 0.000500 Time 0.020352 -2023-02-13 18:11:16,845 - Epoch: [120][ 450/ 1207] Overall Loss 0.242657 Objective Loss 0.242657 LR 0.000500 Time 0.020317 -2023-02-13 18:11:17,033 - Epoch: [120][ 460/ 1207] Overall Loss 0.242839 Objective Loss 0.242839 LR 0.000500 Time 0.020285 -2023-02-13 18:11:17,222 - Epoch: [120][ 470/ 1207] Overall Loss 0.243634 Objective Loss 0.243634 LR 0.000500 Time 0.020254 -2023-02-13 18:11:17,410 - Epoch: [120][ 480/ 1207] Overall Loss 0.243980 Objective Loss 0.243980 LR 0.000500 Time 0.020223 -2023-02-13 18:11:17,598 - Epoch: [120][ 490/ 1207] Overall Loss 0.244003 Objective Loss 0.244003 LR 0.000500 Time 0.020193 -2023-02-13 18:11:17,786 - Epoch: [120][ 500/ 1207] Overall Loss 0.243861 Objective Loss 0.243861 LR 0.000500 Time 0.020164 -2023-02-13 18:11:17,974 - Epoch: [120][ 510/ 1207] Overall Loss 0.244036 Objective Loss 0.244036 LR 0.000500 Time 0.020137 -2023-02-13 18:11:18,163 - Epoch: [120][ 520/ 1207] Overall Loss 0.244330 Objective Loss 0.244330 LR 0.000500 Time 0.020112 -2023-02-13 18:11:18,351 - Epoch: [120][ 530/ 1207] Overall Loss 0.243761 Objective Loss 0.243761 LR 0.000500 Time 0.020086 -2023-02-13 18:11:18,539 - Epoch: [120][ 540/ 1207] Overall Loss 0.244003 Objective Loss 0.244003 LR 0.000500 Time 0.020063 -2023-02-13 18:11:18,727 - Epoch: [120][ 550/ 1207] Overall Loss 0.243879 Objective Loss 0.243879 LR 0.000500 Time 0.020039 -2023-02-13 18:11:18,916 - Epoch: [120][ 560/ 1207] Overall Loss 0.243822 Objective Loss 0.243822 LR 0.000500 Time 0.020017 -2023-02-13 18:11:19,103 - Epoch: [120][ 570/ 1207] Overall Loss 0.243816 Objective Loss 0.243816 LR 0.000500 Time 0.019994 -2023-02-13 18:11:19,292 - Epoch: [120][ 580/ 1207] Overall Loss 0.243372 Objective Loss 0.243372 LR 0.000500 Time 0.019974 -2023-02-13 18:11:19,479 - Epoch: [120][ 590/ 1207] Overall Loss 0.243411 Objective Loss 0.243411 LR 0.000500 Time 0.019953 -2023-02-13 18:11:19,667 - Epoch: [120][ 600/ 1207] Overall Loss 0.243803 Objective Loss 0.243803 LR 0.000500 Time 0.019933 -2023-02-13 18:11:19,856 - Epoch: [120][ 610/ 1207] Overall Loss 0.243466 Objective Loss 0.243466 LR 0.000500 Time 0.019914 -2023-02-13 18:11:20,044 - Epoch: [120][ 620/ 1207] Overall Loss 0.243544 Objective Loss 0.243544 LR 0.000500 Time 0.019896 -2023-02-13 18:11:20,233 - Epoch: [120][ 630/ 1207] Overall Loss 0.243783 Objective Loss 0.243783 LR 0.000500 Time 0.019880 -2023-02-13 18:11:20,421 - Epoch: [120][ 640/ 1207] Overall Loss 0.243883 Objective Loss 0.243883 LR 0.000500 Time 0.019863 -2023-02-13 18:11:20,609 - Epoch: [120][ 650/ 1207] Overall Loss 0.244059 Objective Loss 0.244059 LR 0.000500 Time 0.019845 -2023-02-13 18:11:20,797 - Epoch: [120][ 660/ 1207] Overall Loss 0.244037 Objective Loss 0.244037 LR 0.000500 Time 0.019829 -2023-02-13 18:11:20,986 - Epoch: [120][ 670/ 1207] Overall Loss 0.244163 Objective Loss 0.244163 LR 0.000500 Time 0.019815 -2023-02-13 18:11:21,176 - Epoch: [120][ 680/ 1207] Overall Loss 0.244554 Objective Loss 0.244554 LR 0.000500 Time 0.019802 -2023-02-13 18:11:21,364 - Epoch: [120][ 690/ 1207] Overall Loss 0.244622 Objective Loss 0.244622 LR 0.000500 Time 0.019786 -2023-02-13 18:11:21,552 - Epoch: [120][ 700/ 1207] Overall Loss 0.244629 Objective Loss 0.244629 LR 0.000500 Time 0.019772 -2023-02-13 18:11:21,740 - Epoch: [120][ 710/ 1207] Overall Loss 0.244521 Objective Loss 0.244521 LR 0.000500 Time 0.019758 -2023-02-13 18:11:21,929 - Epoch: [120][ 720/ 1207] Overall Loss 0.244578 Objective Loss 0.244578 LR 0.000500 Time 0.019745 -2023-02-13 18:11:22,117 - Epoch: [120][ 730/ 1207] Overall Loss 0.244786 Objective Loss 0.244786 LR 0.000500 Time 0.019732 -2023-02-13 18:11:22,306 - Epoch: [120][ 740/ 1207] Overall Loss 0.244350 Objective Loss 0.244350 LR 0.000500 Time 0.019720 -2023-02-13 18:11:22,493 - Epoch: [120][ 750/ 1207] Overall Loss 0.244552 Objective Loss 0.244552 LR 0.000500 Time 0.019706 -2023-02-13 18:11:22,681 - Epoch: [120][ 760/ 1207] Overall Loss 0.244896 Objective Loss 0.244896 LR 0.000500 Time 0.019694 -2023-02-13 18:11:22,869 - Epoch: [120][ 770/ 1207] Overall Loss 0.245019 Objective Loss 0.245019 LR 0.000500 Time 0.019682 -2023-02-13 18:11:23,057 - Epoch: [120][ 780/ 1207] Overall Loss 0.244977 Objective Loss 0.244977 LR 0.000500 Time 0.019670 -2023-02-13 18:11:23,245 - Epoch: [120][ 790/ 1207] Overall Loss 0.245148 Objective Loss 0.245148 LR 0.000500 Time 0.019659 -2023-02-13 18:11:23,433 - Epoch: [120][ 800/ 1207] Overall Loss 0.245255 Objective Loss 0.245255 LR 0.000500 Time 0.019647 -2023-02-13 18:11:23,621 - Epoch: [120][ 810/ 1207] Overall Loss 0.245541 Objective Loss 0.245541 LR 0.000500 Time 0.019636 -2023-02-13 18:11:23,809 - Epoch: [120][ 820/ 1207] Overall Loss 0.245423 Objective Loss 0.245423 LR 0.000500 Time 0.019626 -2023-02-13 18:11:23,997 - Epoch: [120][ 830/ 1207] Overall Loss 0.245616 Objective Loss 0.245616 LR 0.000500 Time 0.019615 -2023-02-13 18:11:24,186 - Epoch: [120][ 840/ 1207] Overall Loss 0.245483 Objective Loss 0.245483 LR 0.000500 Time 0.019606 -2023-02-13 18:11:24,518 - Epoch: [120][ 850/ 1207] Overall Loss 0.245579 Objective Loss 0.245579 LR 0.000500 Time 0.019644 -2023-02-13 18:11:24,707 - Epoch: [120][ 860/ 1207] Overall Loss 0.245570 Objective Loss 0.245570 LR 0.000500 Time 0.019635 -2023-02-13 18:11:24,897 - Epoch: [120][ 870/ 1207] Overall Loss 0.245450 Objective Loss 0.245450 LR 0.000500 Time 0.019628 -2023-02-13 18:11:25,087 - Epoch: [120][ 880/ 1207] Overall Loss 0.245910 Objective Loss 0.245910 LR 0.000500 Time 0.019619 -2023-02-13 18:11:25,278 - Epoch: [120][ 890/ 1207] Overall Loss 0.246000 Objective Loss 0.246000 LR 0.000500 Time 0.019613 -2023-02-13 18:11:25,467 - Epoch: [120][ 900/ 1207] Overall Loss 0.246155 Objective Loss 0.246155 LR 0.000500 Time 0.019605 -2023-02-13 18:11:25,657 - Epoch: [120][ 910/ 1207] Overall Loss 0.246447 Objective Loss 0.246447 LR 0.000500 Time 0.019598 -2023-02-13 18:11:25,849 - Epoch: [120][ 920/ 1207] Overall Loss 0.246817 Objective Loss 0.246817 LR 0.000500 Time 0.019593 -2023-02-13 18:11:26,040 - Epoch: [120][ 930/ 1207] Overall Loss 0.247210 Objective Loss 0.247210 LR 0.000500 Time 0.019587 -2023-02-13 18:11:26,230 - Epoch: [120][ 940/ 1207] Overall Loss 0.247152 Objective Loss 0.247152 LR 0.000500 Time 0.019581 -2023-02-13 18:11:26,420 - Epoch: [120][ 950/ 1207] Overall Loss 0.247292 Objective Loss 0.247292 LR 0.000500 Time 0.019574 -2023-02-13 18:11:26,609 - Epoch: [120][ 960/ 1207] Overall Loss 0.247465 Objective Loss 0.247465 LR 0.000500 Time 0.019567 -2023-02-13 18:11:26,799 - Epoch: [120][ 970/ 1207] Overall Loss 0.247507 Objective Loss 0.247507 LR 0.000500 Time 0.019561 -2023-02-13 18:11:26,989 - Epoch: [120][ 980/ 1207] Overall Loss 0.247454 Objective Loss 0.247454 LR 0.000500 Time 0.019555 -2023-02-13 18:11:27,181 - Epoch: [120][ 990/ 1207] Overall Loss 0.247637 Objective Loss 0.247637 LR 0.000500 Time 0.019550 -2023-02-13 18:11:27,370 - Epoch: [120][ 1000/ 1207] Overall Loss 0.247489 Objective Loss 0.247489 LR 0.000500 Time 0.019544 -2023-02-13 18:11:27,560 - Epoch: [120][ 1010/ 1207] Overall Loss 0.247393 Objective Loss 0.247393 LR 0.000500 Time 0.019538 -2023-02-13 18:11:27,750 - Epoch: [120][ 1020/ 1207] Overall Loss 0.247486 Objective Loss 0.247486 LR 0.000500 Time 0.019532 -2023-02-13 18:11:27,940 - Epoch: [120][ 1030/ 1207] Overall Loss 0.247512 Objective Loss 0.247512 LR 0.000500 Time 0.019527 -2023-02-13 18:11:28,130 - Epoch: [120][ 1040/ 1207] Overall Loss 0.247453 Objective Loss 0.247453 LR 0.000500 Time 0.019522 -2023-02-13 18:11:28,321 - Epoch: [120][ 1050/ 1207] Overall Loss 0.247769 Objective Loss 0.247769 LR 0.000500 Time 0.019517 -2023-02-13 18:11:28,511 - Epoch: [120][ 1060/ 1207] Overall Loss 0.248001 Objective Loss 0.248001 LR 0.000500 Time 0.019512 -2023-02-13 18:11:28,701 - Epoch: [120][ 1070/ 1207] Overall Loss 0.247887 Objective Loss 0.247887 LR 0.000500 Time 0.019506 -2023-02-13 18:11:28,891 - Epoch: [120][ 1080/ 1207] Overall Loss 0.248037 Objective Loss 0.248037 LR 0.000500 Time 0.019502 -2023-02-13 18:11:29,081 - Epoch: [120][ 1090/ 1207] Overall Loss 0.247914 Objective Loss 0.247914 LR 0.000500 Time 0.019497 -2023-02-13 18:11:29,271 - Epoch: [120][ 1100/ 1207] Overall Loss 0.247853 Objective Loss 0.247853 LR 0.000500 Time 0.019492 -2023-02-13 18:11:29,461 - Epoch: [120][ 1110/ 1207] Overall Loss 0.247802 Objective Loss 0.247802 LR 0.000500 Time 0.019487 -2023-02-13 18:11:29,651 - Epoch: [120][ 1120/ 1207] Overall Loss 0.247939 Objective Loss 0.247939 LR 0.000500 Time 0.019482 -2023-02-13 18:11:29,841 - Epoch: [120][ 1130/ 1207] Overall Loss 0.247654 Objective Loss 0.247654 LR 0.000500 Time 0.019478 -2023-02-13 18:11:30,031 - Epoch: [120][ 1140/ 1207] Overall Loss 0.247613 Objective Loss 0.247613 LR 0.000500 Time 0.019473 -2023-02-13 18:11:30,221 - Epoch: [120][ 1150/ 1207] Overall Loss 0.247625 Objective Loss 0.247625 LR 0.000500 Time 0.019469 -2023-02-13 18:11:30,411 - Epoch: [120][ 1160/ 1207] Overall Loss 0.247543 Objective Loss 0.247543 LR 0.000500 Time 0.019465 -2023-02-13 18:11:30,601 - Epoch: [120][ 1170/ 1207] Overall Loss 0.247737 Objective Loss 0.247737 LR 0.000500 Time 0.019461 -2023-02-13 18:11:30,791 - Epoch: [120][ 1180/ 1207] Overall Loss 0.247727 Objective Loss 0.247727 LR 0.000500 Time 0.019456 -2023-02-13 18:11:30,984 - Epoch: [120][ 1190/ 1207] Overall Loss 0.247637 Objective Loss 0.247637 LR 0.000500 Time 0.019454 -2023-02-13 18:11:31,224 - Epoch: [120][ 1200/ 1207] Overall Loss 0.247563 Objective Loss 0.247563 LR 0.000500 Time 0.019492 -2023-02-13 18:11:31,339 - Epoch: [120][ 1207/ 1207] Overall Loss 0.247319 Objective Loss 0.247319 Top1 87.500000 Top5 98.475610 LR 0.000500 Time 0.019474 -2023-02-13 18:11:31,409 - --- validate (epoch=120)----------- -2023-02-13 18:11:31,410 - 34311 samples (256 per mini-batch) -2023-02-13 18:11:32,041 - Epoch: [120][ 10/ 135] Loss 0.336049 Top1 84.062500 Top5 97.773438 -2023-02-13 18:11:32,172 - Epoch: [120][ 20/ 135] Loss 0.315302 Top1 84.531250 Top5 97.773438 -2023-02-13 18:11:32,302 - Epoch: [120][ 30/ 135] Loss 0.307516 Top1 84.375000 Top5 97.773438 -2023-02-13 18:11:32,432 - Epoch: [120][ 40/ 135] Loss 0.301552 Top1 84.433594 Top5 97.822266 -2023-02-13 18:11:32,560 - Epoch: [120][ 50/ 135] Loss 0.306163 Top1 84.601562 Top5 97.835938 -2023-02-13 18:11:32,688 - Epoch: [120][ 60/ 135] Loss 0.305156 Top1 84.570312 Top5 97.825521 -2023-02-13 18:11:32,816 - Epoch: [120][ 70/ 135] Loss 0.302324 Top1 84.587054 Top5 97.868304 -2023-02-13 18:11:32,940 - Epoch: [120][ 80/ 135] Loss 0.303421 Top1 84.497070 Top5 97.866211 -2023-02-13 18:11:33,069 - Epoch: [120][ 90/ 135] Loss 0.305599 Top1 84.431424 Top5 97.816840 -2023-02-13 18:11:33,200 - Epoch: [120][ 100/ 135] Loss 0.309126 Top1 84.281250 Top5 97.769531 -2023-02-13 18:11:33,339 - Epoch: [120][ 110/ 135] Loss 0.314361 Top1 84.307528 Top5 97.741477 -2023-02-13 18:11:33,467 - Epoch: [120][ 120/ 135] Loss 0.313124 Top1 84.319661 Top5 97.721354 -2023-02-13 18:11:33,596 - Epoch: [120][ 130/ 135] Loss 0.314870 Top1 84.305889 Top5 97.698317 -2023-02-13 18:11:33,644 - Epoch: [120][ 135/ 135] Loss 0.324084 Top1 84.322812 Top5 97.697531 -2023-02-13 18:11:33,721 - ==> Top1: 84.323 Top5: 97.698 Loss: 0.324 - -2023-02-13 18:11:33,721 - ==> Confusion: -[[ 876 5 3 1 5 2 0 0 1 41 1 6 0 5 5 5 1 2 1 3 4] - [ 2 947 3 1 11 27 2 16 4 1 1 2 2 0 1 1 3 0 5 0 4] - [ 8 2 953 18 4 1 13 9 1 1 4 1 2 4 4 5 3 4 9 5 7] - [ 5 1 22 907 3 3 3 2 4 3 10 0 5 1 18 0 2 6 15 0 6] - [ 15 8 0 1 985 11 1 1 1 5 0 8 1 4 7 6 5 2 0 2 3] - [ 1 18 1 5 9 954 1 21 3 3 3 10 2 22 2 2 2 0 1 3 7] - [ 3 5 21 2 1 3 1027 6 0 0 5 1 1 2 0 4 2 4 1 5 6] - [ 1 7 5 1 5 27 3 932 2 2 2 5 2 1 1 0 1 1 18 5 3] - [ 19 2 4 1 1 0 0 2 897 37 10 1 0 10 16 1 0 2 4 0 2] - [ 90 0 2 0 6 2 0 1 37 838 1 1 1 16 5 2 1 2 2 1 4] - [ 1 1 5 8 3 3 2 6 15 1 975 1 0 13 3 0 0 0 10 1 3] - [ 1 3 1 0 5 10 0 7 0 1 1 894 29 15 0 6 2 13 1 12 4] - [ 0 0 1 9 2 3 0 2 2 0 2 21 867 0 5 5 1 24 4 0 11] - [ 4 4 3 0 9 5 1 1 12 15 5 4 1 944 3 4 2 0 1 2 4] - [ 13 2 0 19 0 3 0 1 20 5 4 4 3 0 994 1 1 8 7 0 7] - [ 3 2 6 0 7 3 1 0 1 0 0 8 13 3 0 961 9 17 0 6 6] - [ 3 8 4 2 12 2 0 1 1 0 0 2 2 2 3 10 986 2 0 7 14] - [ 5 5 1 5 1 1 1 0 1 0 0 5 12 1 1 9 0 997 1 0 5] - [ 5 3 3 10 1 1 0 26 5 0 5 2 3 0 11 1 2 2 999 3 4] - [ 0 2 2 1 4 3 6 15 0 0 2 9 2 3 0 8 6 3 1 1070 11] - [ 164 238 226 173 155 196 72 171 105 89 173 83 291 346 194 103 195 115 188 228 9929]] - -2023-02-13 18:11:33,723 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:11:33,723 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:11:33,729 - - -2023-02-13 18:11:33,729 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:11:34,601 - Epoch: [121][ 10/ 1207] Overall Loss 0.244551 Objective Loss 0.244551 LR 0.000500 Time 0.087154 -2023-02-13 18:11:34,799 - Epoch: [121][ 20/ 1207] Overall Loss 0.248777 Objective Loss 0.248777 LR 0.000500 Time 0.053451 -2023-02-13 18:11:34,988 - Epoch: [121][ 30/ 1207] Overall Loss 0.238625 Objective Loss 0.238625 LR 0.000500 Time 0.041920 -2023-02-13 18:11:35,177 - Epoch: [121][ 40/ 1207] Overall Loss 0.241452 Objective Loss 0.241452 LR 0.000500 Time 0.036142 -2023-02-13 18:11:35,365 - Epoch: [121][ 50/ 1207] Overall Loss 0.243267 Objective Loss 0.243267 LR 0.000500 Time 0.032669 -2023-02-13 18:11:35,553 - Epoch: [121][ 60/ 1207] Overall Loss 0.243267 Objective Loss 0.243267 LR 0.000500 Time 0.030350 -2023-02-13 18:11:35,740 - Epoch: [121][ 70/ 1207] Overall Loss 0.242285 Objective Loss 0.242285 LR 0.000500 Time 0.028691 -2023-02-13 18:11:35,930 - Epoch: [121][ 80/ 1207] Overall Loss 0.245053 Objective Loss 0.245053 LR 0.000500 Time 0.027468 -2023-02-13 18:11:36,118 - Epoch: [121][ 90/ 1207] Overall Loss 0.244088 Objective Loss 0.244088 LR 0.000500 Time 0.026502 -2023-02-13 18:11:36,306 - Epoch: [121][ 100/ 1207] Overall Loss 0.246024 Objective Loss 0.246024 LR 0.000500 Time 0.025732 -2023-02-13 18:11:36,494 - Epoch: [121][ 110/ 1207] Overall Loss 0.245808 Objective Loss 0.245808 LR 0.000500 Time 0.025098 -2023-02-13 18:11:36,682 - Epoch: [121][ 120/ 1207] Overall Loss 0.244626 Objective Loss 0.244626 LR 0.000500 Time 0.024570 -2023-02-13 18:11:36,871 - Epoch: [121][ 130/ 1207] Overall Loss 0.244240 Objective Loss 0.244240 LR 0.000500 Time 0.024127 -2023-02-13 18:11:37,059 - Epoch: [121][ 140/ 1207] Overall Loss 0.244822 Objective Loss 0.244822 LR 0.000500 Time 0.023744 -2023-02-13 18:11:37,248 - Epoch: [121][ 150/ 1207] Overall Loss 0.245347 Objective Loss 0.245347 LR 0.000500 Time 0.023418 -2023-02-13 18:11:37,435 - Epoch: [121][ 160/ 1207] Overall Loss 0.245742 Objective Loss 0.245742 LR 0.000500 Time 0.023126 -2023-02-13 18:11:37,624 - Epoch: [121][ 170/ 1207] Overall Loss 0.245671 Objective Loss 0.245671 LR 0.000500 Time 0.022870 -2023-02-13 18:11:37,812 - Epoch: [121][ 180/ 1207] Overall Loss 0.245326 Objective Loss 0.245326 LR 0.000500 Time 0.022646 -2023-02-13 18:11:38,001 - Epoch: [121][ 190/ 1207] Overall Loss 0.245936 Objective Loss 0.245936 LR 0.000500 Time 0.022444 -2023-02-13 18:11:38,189 - Epoch: [121][ 200/ 1207] Overall Loss 0.245257 Objective Loss 0.245257 LR 0.000500 Time 0.022262 -2023-02-13 18:11:38,377 - Epoch: [121][ 210/ 1207] Overall Loss 0.245691 Objective Loss 0.245691 LR 0.000500 Time 0.022097 -2023-02-13 18:11:38,566 - Epoch: [121][ 220/ 1207] Overall Loss 0.245209 Objective Loss 0.245209 LR 0.000500 Time 0.021947 -2023-02-13 18:11:38,754 - Epoch: [121][ 230/ 1207] Overall Loss 0.245019 Objective Loss 0.245019 LR 0.000500 Time 0.021808 -2023-02-13 18:11:38,942 - Epoch: [121][ 240/ 1207] Overall Loss 0.244407 Objective Loss 0.244407 LR 0.000500 Time 0.021683 -2023-02-13 18:11:39,136 - Epoch: [121][ 250/ 1207] Overall Loss 0.243829 Objective Loss 0.243829 LR 0.000500 Time 0.021590 -2023-02-13 18:11:39,336 - Epoch: [121][ 260/ 1207] Overall Loss 0.244308 Objective Loss 0.244308 LR 0.000500 Time 0.021526 -2023-02-13 18:11:39,541 - Epoch: [121][ 270/ 1207] Overall Loss 0.243948 Objective Loss 0.243948 LR 0.000500 Time 0.021486 -2023-02-13 18:11:39,741 - Epoch: [121][ 280/ 1207] Overall Loss 0.244167 Objective Loss 0.244167 LR 0.000500 Time 0.021431 -2023-02-13 18:11:39,931 - Epoch: [121][ 290/ 1207] Overall Loss 0.244255 Objective Loss 0.244255 LR 0.000500 Time 0.021347 -2023-02-13 18:11:40,120 - Epoch: [121][ 300/ 1207] Overall Loss 0.243775 Objective Loss 0.243775 LR 0.000500 Time 0.021266 -2023-02-13 18:11:40,311 - Epoch: [121][ 310/ 1207] Overall Loss 0.243525 Objective Loss 0.243525 LR 0.000500 Time 0.021193 -2023-02-13 18:11:40,500 - Epoch: [121][ 320/ 1207] Overall Loss 0.243256 Objective Loss 0.243256 LR 0.000500 Time 0.021122 -2023-02-13 18:11:40,690 - Epoch: [121][ 330/ 1207] Overall Loss 0.243619 Objective Loss 0.243619 LR 0.000500 Time 0.021055 -2023-02-13 18:11:40,880 - Epoch: [121][ 340/ 1207] Overall Loss 0.243888 Objective Loss 0.243888 LR 0.000500 Time 0.020995 -2023-02-13 18:11:41,071 - Epoch: [121][ 350/ 1207] Overall Loss 0.244106 Objective Loss 0.244106 LR 0.000500 Time 0.020937 -2023-02-13 18:11:41,261 - Epoch: [121][ 360/ 1207] Overall Loss 0.244664 Objective Loss 0.244664 LR 0.000500 Time 0.020884 -2023-02-13 18:11:41,451 - Epoch: [121][ 370/ 1207] Overall Loss 0.244925 Objective Loss 0.244925 LR 0.000500 Time 0.020831 -2023-02-13 18:11:41,641 - Epoch: [121][ 380/ 1207] Overall Loss 0.244879 Objective Loss 0.244879 LR 0.000500 Time 0.020781 -2023-02-13 18:11:41,831 - Epoch: [121][ 390/ 1207] Overall Loss 0.244731 Objective Loss 0.244731 LR 0.000500 Time 0.020735 -2023-02-13 18:11:42,024 - Epoch: [121][ 400/ 1207] Overall Loss 0.244935 Objective Loss 0.244935 LR 0.000500 Time 0.020700 -2023-02-13 18:11:42,216 - Epoch: [121][ 410/ 1207] Overall Loss 0.244866 Objective Loss 0.244866 LR 0.000500 Time 0.020661 -2023-02-13 18:11:42,408 - Epoch: [121][ 420/ 1207] Overall Loss 0.245243 Objective Loss 0.245243 LR 0.000500 Time 0.020627 -2023-02-13 18:11:42,599 - Epoch: [121][ 430/ 1207] Overall Loss 0.246264 Objective Loss 0.246264 LR 0.000500 Time 0.020590 -2023-02-13 18:11:42,792 - Epoch: [121][ 440/ 1207] Overall Loss 0.246456 Objective Loss 0.246456 LR 0.000500 Time 0.020560 -2023-02-13 18:11:42,984 - Epoch: [121][ 450/ 1207] Overall Loss 0.245861 Objective Loss 0.245861 LR 0.000500 Time 0.020528 -2023-02-13 18:11:43,177 - Epoch: [121][ 460/ 1207] Overall Loss 0.246051 Objective Loss 0.246051 LR 0.000500 Time 0.020501 -2023-02-13 18:11:43,369 - Epoch: [121][ 470/ 1207] Overall Loss 0.246124 Objective Loss 0.246124 LR 0.000500 Time 0.020472 -2023-02-13 18:11:43,562 - Epoch: [121][ 480/ 1207] Overall Loss 0.246137 Objective Loss 0.246137 LR 0.000500 Time 0.020447 -2023-02-13 18:11:43,753 - Epoch: [121][ 490/ 1207] Overall Loss 0.246433 Objective Loss 0.246433 LR 0.000500 Time 0.020419 -2023-02-13 18:11:43,947 - Epoch: [121][ 500/ 1207] Overall Loss 0.246514 Objective Loss 0.246514 LR 0.000500 Time 0.020397 -2023-02-13 18:11:44,138 - Epoch: [121][ 510/ 1207] Overall Loss 0.246442 Objective Loss 0.246442 LR 0.000500 Time 0.020372 -2023-02-13 18:11:44,332 - Epoch: [121][ 520/ 1207] Overall Loss 0.246671 Objective Loss 0.246671 LR 0.000500 Time 0.020353 -2023-02-13 18:11:44,524 - Epoch: [121][ 530/ 1207] Overall Loss 0.246270 Objective Loss 0.246270 LR 0.000500 Time 0.020330 -2023-02-13 18:11:44,717 - Epoch: [121][ 540/ 1207] Overall Loss 0.246565 Objective Loss 0.246565 LR 0.000500 Time 0.020311 -2023-02-13 18:11:44,909 - Epoch: [121][ 550/ 1207] Overall Loss 0.246609 Objective Loss 0.246609 LR 0.000500 Time 0.020288 -2023-02-13 18:11:45,100 - Epoch: [121][ 560/ 1207] Overall Loss 0.246267 Objective Loss 0.246267 LR 0.000500 Time 0.020267 -2023-02-13 18:11:45,291 - Epoch: [121][ 570/ 1207] Overall Loss 0.246306 Objective Loss 0.246306 LR 0.000500 Time 0.020245 -2023-02-13 18:11:45,481 - Epoch: [121][ 580/ 1207] Overall Loss 0.246171 Objective Loss 0.246171 LR 0.000500 Time 0.020224 -2023-02-13 18:11:45,671 - Epoch: [121][ 590/ 1207] Overall Loss 0.245958 Objective Loss 0.245958 LR 0.000500 Time 0.020203 -2023-02-13 18:11:45,862 - Epoch: [121][ 600/ 1207] Overall Loss 0.246158 Objective Loss 0.246158 LR 0.000500 Time 0.020184 -2023-02-13 18:11:46,053 - Epoch: [121][ 610/ 1207] Overall Loss 0.246363 Objective Loss 0.246363 LR 0.000500 Time 0.020165 -2023-02-13 18:11:46,243 - Epoch: [121][ 620/ 1207] Overall Loss 0.246180 Objective Loss 0.246180 LR 0.000500 Time 0.020146 -2023-02-13 18:11:46,433 - Epoch: [121][ 630/ 1207] Overall Loss 0.246281 Objective Loss 0.246281 LR 0.000500 Time 0.020127 -2023-02-13 18:11:46,623 - Epoch: [121][ 640/ 1207] Overall Loss 0.246234 Objective Loss 0.246234 LR 0.000500 Time 0.020109 -2023-02-13 18:11:46,814 - Epoch: [121][ 650/ 1207] Overall Loss 0.246315 Objective Loss 0.246315 LR 0.000500 Time 0.020092 -2023-02-13 18:11:47,004 - Epoch: [121][ 660/ 1207] Overall Loss 0.245836 Objective Loss 0.245836 LR 0.000500 Time 0.020076 -2023-02-13 18:11:47,194 - Epoch: [121][ 670/ 1207] Overall Loss 0.246197 Objective Loss 0.246197 LR 0.000500 Time 0.020059 -2023-02-13 18:11:47,392 - Epoch: [121][ 680/ 1207] Overall Loss 0.246308 Objective Loss 0.246308 LR 0.000500 Time 0.020055 -2023-02-13 18:11:47,586 - Epoch: [121][ 690/ 1207] Overall Loss 0.245901 Objective Loss 0.245901 LR 0.000500 Time 0.020045 -2023-02-13 18:11:47,784 - Epoch: [121][ 700/ 1207] Overall Loss 0.245974 Objective Loss 0.245974 LR 0.000500 Time 0.020040 -2023-02-13 18:11:47,978 - Epoch: [121][ 710/ 1207] Overall Loss 0.245724 Objective Loss 0.245724 LR 0.000500 Time 0.020032 -2023-02-13 18:11:48,176 - Epoch: [121][ 720/ 1207] Overall Loss 0.245836 Objective Loss 0.245836 LR 0.000500 Time 0.020027 -2023-02-13 18:11:48,370 - Epoch: [121][ 730/ 1207] Overall Loss 0.245647 Objective Loss 0.245647 LR 0.000500 Time 0.020019 -2023-02-13 18:11:48,568 - Epoch: [121][ 740/ 1207] Overall Loss 0.245545 Objective Loss 0.245545 LR 0.000500 Time 0.020014 -2023-02-13 18:11:48,763 - Epoch: [121][ 750/ 1207] Overall Loss 0.245536 Objective Loss 0.245536 LR 0.000500 Time 0.020007 -2023-02-13 18:11:48,960 - Epoch: [121][ 760/ 1207] Overall Loss 0.245184 Objective Loss 0.245184 LR 0.000500 Time 0.020003 -2023-02-13 18:11:49,155 - Epoch: [121][ 770/ 1207] Overall Loss 0.245056 Objective Loss 0.245056 LR 0.000500 Time 0.019996 -2023-02-13 18:11:49,354 - Epoch: [121][ 780/ 1207] Overall Loss 0.245480 Objective Loss 0.245480 LR 0.000500 Time 0.019994 -2023-02-13 18:11:49,548 - Epoch: [121][ 790/ 1207] Overall Loss 0.245666 Objective Loss 0.245666 LR 0.000500 Time 0.019986 -2023-02-13 18:11:49,745 - Epoch: [121][ 800/ 1207] Overall Loss 0.245359 Objective Loss 0.245359 LR 0.000500 Time 0.019982 -2023-02-13 18:11:49,940 - Epoch: [121][ 810/ 1207] Overall Loss 0.245060 Objective Loss 0.245060 LR 0.000500 Time 0.019975 -2023-02-13 18:11:50,137 - Epoch: [121][ 820/ 1207] Overall Loss 0.245011 Objective Loss 0.245011 LR 0.000500 Time 0.019971 -2023-02-13 18:11:50,332 - Epoch: [121][ 830/ 1207] Overall Loss 0.245256 Objective Loss 0.245256 LR 0.000500 Time 0.019965 -2023-02-13 18:11:50,529 - Epoch: [121][ 840/ 1207] Overall Loss 0.244966 Objective Loss 0.244966 LR 0.000500 Time 0.019962 -2023-02-13 18:11:50,723 - Epoch: [121][ 850/ 1207] Overall Loss 0.245152 Objective Loss 0.245152 LR 0.000500 Time 0.019955 -2023-02-13 18:11:50,922 - Epoch: [121][ 860/ 1207] Overall Loss 0.244999 Objective Loss 0.244999 LR 0.000500 Time 0.019953 -2023-02-13 18:11:51,116 - Epoch: [121][ 870/ 1207] Overall Loss 0.245605 Objective Loss 0.245605 LR 0.000500 Time 0.019947 -2023-02-13 18:11:51,314 - Epoch: [121][ 880/ 1207] Overall Loss 0.245675 Objective Loss 0.245675 LR 0.000500 Time 0.019946 -2023-02-13 18:11:51,509 - Epoch: [121][ 890/ 1207] Overall Loss 0.245815 Objective Loss 0.245815 LR 0.000500 Time 0.019939 -2023-02-13 18:11:51,707 - Epoch: [121][ 900/ 1207] Overall Loss 0.245454 Objective Loss 0.245454 LR 0.000500 Time 0.019938 -2023-02-13 18:11:51,902 - Epoch: [121][ 910/ 1207] Overall Loss 0.245460 Objective Loss 0.245460 LR 0.000500 Time 0.019933 -2023-02-13 18:11:52,100 - Epoch: [121][ 920/ 1207] Overall Loss 0.245465 Objective Loss 0.245465 LR 0.000500 Time 0.019931 -2023-02-13 18:11:52,295 - Epoch: [121][ 930/ 1207] Overall Loss 0.245886 Objective Loss 0.245886 LR 0.000500 Time 0.019925 -2023-02-13 18:11:52,492 - Epoch: [121][ 940/ 1207] Overall Loss 0.246054 Objective Loss 0.246054 LR 0.000500 Time 0.019923 -2023-02-13 18:11:52,687 - Epoch: [121][ 950/ 1207] Overall Loss 0.246133 Objective Loss 0.246133 LR 0.000500 Time 0.019918 -2023-02-13 18:11:52,885 - Epoch: [121][ 960/ 1207] Overall Loss 0.245961 Objective Loss 0.245961 LR 0.000500 Time 0.019916 -2023-02-13 18:11:53,080 - Epoch: [121][ 970/ 1207] Overall Loss 0.245852 Objective Loss 0.245852 LR 0.000500 Time 0.019911 -2023-02-13 18:11:53,277 - Epoch: [121][ 980/ 1207] Overall Loss 0.245766 Objective Loss 0.245766 LR 0.000500 Time 0.019909 -2023-02-13 18:11:53,472 - Epoch: [121][ 990/ 1207] Overall Loss 0.245478 Objective Loss 0.245478 LR 0.000500 Time 0.019905 -2023-02-13 18:11:53,662 - Epoch: [121][ 1000/ 1207] Overall Loss 0.245185 Objective Loss 0.245185 LR 0.000500 Time 0.019896 -2023-02-13 18:11:53,854 - Epoch: [121][ 1010/ 1207] Overall Loss 0.245143 Objective Loss 0.245143 LR 0.000500 Time 0.019888 -2023-02-13 18:11:54,044 - Epoch: [121][ 1020/ 1207] Overall Loss 0.245244 Objective Loss 0.245244 LR 0.000500 Time 0.019879 -2023-02-13 18:11:54,235 - Epoch: [121][ 1030/ 1207] Overall Loss 0.245173 Objective Loss 0.245173 LR 0.000500 Time 0.019871 -2023-02-13 18:11:54,426 - Epoch: [121][ 1040/ 1207] Overall Loss 0.245195 Objective Loss 0.245195 LR 0.000500 Time 0.019864 -2023-02-13 18:11:54,618 - Epoch: [121][ 1050/ 1207] Overall Loss 0.245526 Objective Loss 0.245526 LR 0.000500 Time 0.019856 -2023-02-13 18:11:54,815 - Epoch: [121][ 1060/ 1207] Overall Loss 0.245456 Objective Loss 0.245456 LR 0.000500 Time 0.019855 -2023-02-13 18:11:55,009 - Epoch: [121][ 1070/ 1207] Overall Loss 0.245568 Objective Loss 0.245568 LR 0.000500 Time 0.019850 -2023-02-13 18:11:55,205 - Epoch: [121][ 1080/ 1207] Overall Loss 0.245683 Objective Loss 0.245683 LR 0.000500 Time 0.019847 -2023-02-13 18:11:55,399 - Epoch: [121][ 1090/ 1207] Overall Loss 0.245783 Objective Loss 0.245783 LR 0.000500 Time 0.019842 -2023-02-13 18:11:55,595 - Epoch: [121][ 1100/ 1207] Overall Loss 0.245958 Objective Loss 0.245958 LR 0.000500 Time 0.019840 -2023-02-13 18:11:55,789 - Epoch: [121][ 1110/ 1207] Overall Loss 0.245889 Objective Loss 0.245889 LR 0.000500 Time 0.019836 -2023-02-13 18:11:55,987 - Epoch: [121][ 1120/ 1207] Overall Loss 0.245802 Objective Loss 0.245802 LR 0.000500 Time 0.019835 -2023-02-13 18:11:56,180 - Epoch: [121][ 1130/ 1207] Overall Loss 0.245877 Objective Loss 0.245877 LR 0.000500 Time 0.019831 -2023-02-13 18:11:56,378 - Epoch: [121][ 1140/ 1207] Overall Loss 0.246043 Objective Loss 0.246043 LR 0.000500 Time 0.019830 -2023-02-13 18:11:56,572 - Epoch: [121][ 1150/ 1207] Overall Loss 0.245934 Objective Loss 0.245934 LR 0.000500 Time 0.019826 -2023-02-13 18:11:56,770 - Epoch: [121][ 1160/ 1207] Overall Loss 0.245941 Objective Loss 0.245941 LR 0.000500 Time 0.019825 -2023-02-13 18:11:56,965 - Epoch: [121][ 1170/ 1207] Overall Loss 0.245929 Objective Loss 0.245929 LR 0.000500 Time 0.019822 -2023-02-13 18:11:57,162 - Epoch: [121][ 1180/ 1207] Overall Loss 0.245736 Objective Loss 0.245736 LR 0.000500 Time 0.019820 -2023-02-13 18:11:57,355 - Epoch: [121][ 1190/ 1207] Overall Loss 0.245718 Objective Loss 0.245718 LR 0.000500 Time 0.019816 -2023-02-13 18:11:57,606 - Epoch: [121][ 1200/ 1207] Overall Loss 0.245412 Objective Loss 0.245412 LR 0.000500 Time 0.019859 -2023-02-13 18:11:57,722 - Epoch: [121][ 1207/ 1207] Overall Loss 0.245455 Objective Loss 0.245455 Top1 84.451220 Top5 97.560976 LR 0.000500 Time 0.019840 -2023-02-13 18:11:57,795 - --- validate (epoch=121)----------- -2023-02-13 18:11:57,795 - 34311 samples (256 per mini-batch) -2023-02-13 18:11:58,310 - Epoch: [121][ 10/ 135] Loss 0.292640 Top1 85.859375 Top5 97.890625 -2023-02-13 18:11:58,440 - Epoch: [121][ 20/ 135] Loss 0.312631 Top1 84.218750 Top5 97.500000 -2023-02-13 18:11:58,567 - Epoch: [121][ 30/ 135] Loss 0.314483 Top1 84.218750 Top5 97.682292 -2023-02-13 18:11:58,690 - Epoch: [121][ 40/ 135] Loss 0.314474 Top1 84.199219 Top5 97.734375 -2023-02-13 18:11:58,829 - Epoch: [121][ 50/ 135] Loss 0.309777 Top1 84.187500 Top5 97.726562 -2023-02-13 18:11:58,964 - Epoch: [121][ 60/ 135] Loss 0.309360 Top1 83.906250 Top5 97.766927 -2023-02-13 18:11:59,099 - Epoch: [121][ 70/ 135] Loss 0.312596 Top1 83.856027 Top5 97.739955 -2023-02-13 18:11:59,230 - Epoch: [121][ 80/ 135] Loss 0.307810 Top1 83.920898 Top5 97.739258 -2023-02-13 18:11:59,362 - Epoch: [121][ 90/ 135] Loss 0.306321 Top1 84.105903 Top5 97.812500 -2023-02-13 18:11:59,492 - Epoch: [121][ 100/ 135] Loss 0.306998 Top1 84.019531 Top5 97.808594 -2023-02-13 18:11:59,623 - Epoch: [121][ 110/ 135] Loss 0.310041 Top1 84.012784 Top5 97.784091 -2023-02-13 18:11:59,753 - Epoch: [121][ 120/ 135] Loss 0.314130 Top1 83.974609 Top5 97.750651 -2023-02-13 18:11:59,884 - Epoch: [121][ 130/ 135] Loss 0.312651 Top1 84.053486 Top5 97.737380 -2023-02-13 18:11:59,929 - Epoch: [121][ 135/ 135] Loss 0.313422 Top1 84.008044 Top5 97.712104 -2023-02-13 18:11:59,998 - ==> Top1: 84.008 Top5: 97.712 Loss: 0.313 - -2023-02-13 18:11:59,999 - ==> Confusion: -[[ 868 5 8 0 8 2 0 2 3 35 1 4 1 8 7 4 1 3 2 2 3] - [ 2 950 0 1 10 19 3 15 3 1 1 2 2 0 4 4 5 0 1 0 10] - [ 5 3 960 11 4 1 13 16 0 1 4 1 2 8 6 8 2 1 4 4 4] - [ 3 0 23 894 5 6 1 2 1 3 12 0 8 2 32 1 1 4 8 1 9] - [ 14 9 0 0 987 12 0 2 1 5 1 8 1 3 6 5 6 0 0 3 3] - [ 3 24 1 3 5 965 3 18 1 1 1 7 4 18 2 1 5 1 0 4 3] - [ 5 4 23 2 0 7 1028 4 0 1 3 1 0 3 0 5 1 2 2 5 3] - [ 2 5 8 1 2 29 3 933 1 1 3 6 3 1 0 0 1 2 12 5 6] - [ 20 1 3 1 1 0 0 2 883 50 6 2 0 9 18 2 1 1 8 0 1] - [ 82 1 4 1 4 5 0 3 33 841 0 0 0 21 7 3 2 1 2 0 2] - [ 2 3 2 10 1 4 4 4 21 1 964 3 1 8 2 0 2 0 13 1 5] - [ 1 4 1 0 4 11 1 6 3 0 1 915 24 9 0 4 3 9 2 5 2] - [ 0 0 2 4 0 3 0 2 3 1 1 24 872 1 3 9 3 20 4 0 7] - [ 5 2 2 0 8 9 0 2 10 14 10 5 2 941 2 5 2 1 1 0 3] - [ 7 2 5 11 4 3 0 1 18 7 3 1 4 1 1005 2 0 4 8 1 5] - [ 2 1 5 0 13 1 4 0 1 1 1 7 7 2 1 967 10 12 1 4 6] - [ 2 4 1 1 8 4 0 2 0 1 0 0 1 3 3 7 1006 1 2 6 9] - [ 5 2 2 4 1 1 1 0 0 2 2 5 19 3 0 16 1 984 0 0 3] - [ 4 3 2 13 0 3 0 28 2 1 3 1 6 0 13 1 2 2 999 2 1] - [ 0 3 1 1 3 7 5 14 1 0 2 20 3 6 0 6 9 3 0 1060 4] - [ 172 234 232 131 158 236 85 172 94 81 162 111 317 352 208 111 254 125 167 230 9802]] - -2023-02-13 18:12:00,000 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:12:00,000 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:12:00,006 - - -2023-02-13 18:12:00,006 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:12:00,892 - Epoch: [122][ 10/ 1207] Overall Loss 0.236832 Objective Loss 0.236832 LR 0.000500 Time 0.088524 -2023-02-13 18:12:01,090 - Epoch: [122][ 20/ 1207] Overall Loss 0.234608 Objective Loss 0.234608 LR 0.000500 Time 0.054120 -2023-02-13 18:12:01,278 - Epoch: [122][ 30/ 1207] Overall Loss 0.233202 Objective Loss 0.233202 LR 0.000500 Time 0.042355 -2023-02-13 18:12:01,467 - Epoch: [122][ 40/ 1207] Overall Loss 0.233894 Objective Loss 0.233894 LR 0.000500 Time 0.036462 -2023-02-13 18:12:01,654 - Epoch: [122][ 50/ 1207] Overall Loss 0.234366 Objective Loss 0.234366 LR 0.000500 Time 0.032913 -2023-02-13 18:12:01,842 - Epoch: [122][ 60/ 1207] Overall Loss 0.234384 Objective Loss 0.234384 LR 0.000500 Time 0.030550 -2023-02-13 18:12:02,031 - Epoch: [122][ 70/ 1207] Overall Loss 0.235053 Objective Loss 0.235053 LR 0.000500 Time 0.028878 -2023-02-13 18:12:02,218 - Epoch: [122][ 80/ 1207] Overall Loss 0.234482 Objective Loss 0.234482 LR 0.000500 Time 0.027610 -2023-02-13 18:12:02,406 - Epoch: [122][ 90/ 1207] Overall Loss 0.234622 Objective Loss 0.234622 LR 0.000500 Time 0.026626 -2023-02-13 18:12:02,593 - Epoch: [122][ 100/ 1207] Overall Loss 0.236261 Objective Loss 0.236261 LR 0.000500 Time 0.025831 -2023-02-13 18:12:02,781 - Epoch: [122][ 110/ 1207] Overall Loss 0.236444 Objective Loss 0.236444 LR 0.000500 Time 0.025185 -2023-02-13 18:12:02,968 - Epoch: [122][ 120/ 1207] Overall Loss 0.236807 Objective Loss 0.236807 LR 0.000500 Time 0.024645 -2023-02-13 18:12:03,157 - Epoch: [122][ 130/ 1207] Overall Loss 0.236509 Objective Loss 0.236509 LR 0.000500 Time 0.024194 -2023-02-13 18:12:03,345 - Epoch: [122][ 140/ 1207] Overall Loss 0.238195 Objective Loss 0.238195 LR 0.000500 Time 0.023811 -2023-02-13 18:12:03,533 - Epoch: [122][ 150/ 1207] Overall Loss 0.240102 Objective Loss 0.240102 LR 0.000500 Time 0.023472 -2023-02-13 18:12:03,721 - Epoch: [122][ 160/ 1207] Overall Loss 0.239550 Objective Loss 0.239550 LR 0.000500 Time 0.023178 -2023-02-13 18:12:03,909 - Epoch: [122][ 170/ 1207] Overall Loss 0.240560 Objective Loss 0.240560 LR 0.000500 Time 0.022917 -2023-02-13 18:12:04,104 - Epoch: [122][ 180/ 1207] Overall Loss 0.241812 Objective Loss 0.241812 LR 0.000500 Time 0.022726 -2023-02-13 18:12:04,297 - Epoch: [122][ 190/ 1207] Overall Loss 0.241587 Objective Loss 0.241587 LR 0.000500 Time 0.022543 -2023-02-13 18:12:04,493 - Epoch: [122][ 200/ 1207] Overall Loss 0.242253 Objective Loss 0.242253 LR 0.000500 Time 0.022394 -2023-02-13 18:12:04,686 - Epoch: [122][ 210/ 1207] Overall Loss 0.242228 Objective Loss 0.242228 LR 0.000500 Time 0.022246 -2023-02-13 18:12:04,883 - Epoch: [122][ 220/ 1207] Overall Loss 0.241912 Objective Loss 0.241912 LR 0.000500 Time 0.022126 -2023-02-13 18:12:05,076 - Epoch: [122][ 230/ 1207] Overall Loss 0.241352 Objective Loss 0.241352 LR 0.000500 Time 0.022003 -2023-02-13 18:12:05,273 - Epoch: [122][ 240/ 1207] Overall Loss 0.241305 Objective Loss 0.241305 LR 0.000500 Time 0.021905 -2023-02-13 18:12:05,466 - Epoch: [122][ 250/ 1207] Overall Loss 0.240429 Objective Loss 0.240429 LR 0.000500 Time 0.021799 -2023-02-13 18:12:05,662 - Epoch: [122][ 260/ 1207] Overall Loss 0.239724 Objective Loss 0.239724 LR 0.000500 Time 0.021714 -2023-02-13 18:12:05,856 - Epoch: [122][ 270/ 1207] Overall Loss 0.240217 Objective Loss 0.240217 LR 0.000500 Time 0.021625 -2023-02-13 18:12:06,052 - Epoch: [122][ 280/ 1207] Overall Loss 0.241135 Objective Loss 0.241135 LR 0.000500 Time 0.021553 -2023-02-13 18:12:06,246 - Epoch: [122][ 290/ 1207] Overall Loss 0.240620 Objective Loss 0.240620 LR 0.000500 Time 0.021478 -2023-02-13 18:12:06,442 - Epoch: [122][ 300/ 1207] Overall Loss 0.241494 Objective Loss 0.241494 LR 0.000500 Time 0.021415 -2023-02-13 18:12:06,635 - Epoch: [122][ 310/ 1207] Overall Loss 0.240846 Objective Loss 0.240846 LR 0.000500 Time 0.021344 -2023-02-13 18:12:06,824 - Epoch: [122][ 320/ 1207] Overall Loss 0.241887 Objective Loss 0.241887 LR 0.000500 Time 0.021267 -2023-02-13 18:12:07,013 - Epoch: [122][ 330/ 1207] Overall Loss 0.242455 Objective Loss 0.242455 LR 0.000500 Time 0.021194 -2023-02-13 18:12:07,202 - Epoch: [122][ 340/ 1207] Overall Loss 0.242737 Objective Loss 0.242737 LR 0.000500 Time 0.021124 -2023-02-13 18:12:07,390 - Epoch: [122][ 350/ 1207] Overall Loss 0.243524 Objective Loss 0.243524 LR 0.000500 Time 0.021059 -2023-02-13 18:12:07,580 - Epoch: [122][ 360/ 1207] Overall Loss 0.243643 Objective Loss 0.243643 LR 0.000500 Time 0.020999 -2023-02-13 18:12:07,770 - Epoch: [122][ 370/ 1207] Overall Loss 0.244041 Objective Loss 0.244041 LR 0.000500 Time 0.020944 -2023-02-13 18:12:07,960 - Epoch: [122][ 380/ 1207] Overall Loss 0.243744 Objective Loss 0.243744 LR 0.000500 Time 0.020891 -2023-02-13 18:12:08,149 - Epoch: [122][ 390/ 1207] Overall Loss 0.244342 Objective Loss 0.244342 LR 0.000500 Time 0.020840 -2023-02-13 18:12:08,338 - Epoch: [122][ 400/ 1207] Overall Loss 0.244645 Objective Loss 0.244645 LR 0.000500 Time 0.020791 -2023-02-13 18:12:08,527 - Epoch: [122][ 410/ 1207] Overall Loss 0.243954 Objective Loss 0.243954 LR 0.000500 Time 0.020744 -2023-02-13 18:12:08,716 - Epoch: [122][ 420/ 1207] Overall Loss 0.244560 Objective Loss 0.244560 LR 0.000500 Time 0.020699 -2023-02-13 18:12:08,906 - Epoch: [122][ 430/ 1207] Overall Loss 0.244542 Objective Loss 0.244542 LR 0.000500 Time 0.020658 -2023-02-13 18:12:09,095 - Epoch: [122][ 440/ 1207] Overall Loss 0.244395 Objective Loss 0.244395 LR 0.000500 Time 0.020617 -2023-02-13 18:12:09,285 - Epoch: [122][ 450/ 1207] Overall Loss 0.244772 Objective Loss 0.244772 LR 0.000500 Time 0.020580 -2023-02-13 18:12:09,474 - Epoch: [122][ 460/ 1207] Overall Loss 0.245301 Objective Loss 0.245301 LR 0.000500 Time 0.020544 -2023-02-13 18:12:09,663 - Epoch: [122][ 470/ 1207] Overall Loss 0.244944 Objective Loss 0.244944 LR 0.000500 Time 0.020507 -2023-02-13 18:12:09,853 - Epoch: [122][ 480/ 1207] Overall Loss 0.245341 Objective Loss 0.245341 LR 0.000500 Time 0.020475 -2023-02-13 18:12:10,042 - Epoch: [122][ 490/ 1207] Overall Loss 0.245522 Objective Loss 0.245522 LR 0.000500 Time 0.020442 -2023-02-13 18:12:10,231 - Epoch: [122][ 500/ 1207] Overall Loss 0.246320 Objective Loss 0.246320 LR 0.000500 Time 0.020412 -2023-02-13 18:12:10,421 - Epoch: [122][ 510/ 1207] Overall Loss 0.246212 Objective Loss 0.246212 LR 0.000500 Time 0.020383 -2023-02-13 18:12:10,610 - Epoch: [122][ 520/ 1207] Overall Loss 0.246407 Objective Loss 0.246407 LR 0.000500 Time 0.020354 -2023-02-13 18:12:10,799 - Epoch: [122][ 530/ 1207] Overall Loss 0.246187 Objective Loss 0.246187 LR 0.000500 Time 0.020325 -2023-02-13 18:12:10,990 - Epoch: [122][ 540/ 1207] Overall Loss 0.246502 Objective Loss 0.246502 LR 0.000500 Time 0.020302 -2023-02-13 18:12:11,179 - Epoch: [122][ 550/ 1207] Overall Loss 0.246129 Objective Loss 0.246129 LR 0.000500 Time 0.020275 -2023-02-13 18:12:11,369 - Epoch: [122][ 560/ 1207] Overall Loss 0.246294 Objective Loss 0.246294 LR 0.000500 Time 0.020252 -2023-02-13 18:12:11,558 - Epoch: [122][ 570/ 1207] Overall Loss 0.246296 Objective Loss 0.246296 LR 0.000500 Time 0.020229 -2023-02-13 18:12:11,747 - Epoch: [122][ 580/ 1207] Overall Loss 0.246716 Objective Loss 0.246716 LR 0.000500 Time 0.020205 -2023-02-13 18:12:11,937 - Epoch: [122][ 590/ 1207] Overall Loss 0.246964 Objective Loss 0.246964 LR 0.000500 Time 0.020184 -2023-02-13 18:12:12,127 - Epoch: [122][ 600/ 1207] Overall Loss 0.246583 Objective Loss 0.246583 LR 0.000500 Time 0.020162 -2023-02-13 18:12:12,317 - Epoch: [122][ 610/ 1207] Overall Loss 0.246708 Objective Loss 0.246708 LR 0.000500 Time 0.020143 -2023-02-13 18:12:12,507 - Epoch: [122][ 620/ 1207] Overall Loss 0.246616 Objective Loss 0.246616 LR 0.000500 Time 0.020125 -2023-02-13 18:12:12,697 - Epoch: [122][ 630/ 1207] Overall Loss 0.246265 Objective Loss 0.246265 LR 0.000500 Time 0.020105 -2023-02-13 18:12:12,887 - Epoch: [122][ 640/ 1207] Overall Loss 0.246544 Objective Loss 0.246544 LR 0.000500 Time 0.020087 -2023-02-13 18:12:13,076 - Epoch: [122][ 650/ 1207] Overall Loss 0.246338 Objective Loss 0.246338 LR 0.000500 Time 0.020069 -2023-02-13 18:12:13,265 - Epoch: [122][ 660/ 1207] Overall Loss 0.246044 Objective Loss 0.246044 LR 0.000500 Time 0.020051 -2023-02-13 18:12:13,455 - Epoch: [122][ 670/ 1207] Overall Loss 0.246552 Objective Loss 0.246552 LR 0.000500 Time 0.020035 -2023-02-13 18:12:13,645 - Epoch: [122][ 680/ 1207] Overall Loss 0.246285 Objective Loss 0.246285 LR 0.000500 Time 0.020018 -2023-02-13 18:12:13,834 - Epoch: [122][ 690/ 1207] Overall Loss 0.246214 Objective Loss 0.246214 LR 0.000500 Time 0.020002 -2023-02-13 18:12:14,023 - Epoch: [122][ 700/ 1207] Overall Loss 0.246217 Objective Loss 0.246217 LR 0.000500 Time 0.019986 -2023-02-13 18:12:14,213 - Epoch: [122][ 710/ 1207] Overall Loss 0.246121 Objective Loss 0.246121 LR 0.000500 Time 0.019971 -2023-02-13 18:12:14,403 - Epoch: [122][ 720/ 1207] Overall Loss 0.246339 Objective Loss 0.246339 LR 0.000500 Time 0.019957 -2023-02-13 18:12:14,592 - Epoch: [122][ 730/ 1207] Overall Loss 0.246377 Objective Loss 0.246377 LR 0.000500 Time 0.019942 -2023-02-13 18:12:14,781 - Epoch: [122][ 740/ 1207] Overall Loss 0.246812 Objective Loss 0.246812 LR 0.000500 Time 0.019928 -2023-02-13 18:12:14,970 - Epoch: [122][ 750/ 1207] Overall Loss 0.247234 Objective Loss 0.247234 LR 0.000500 Time 0.019914 -2023-02-13 18:12:15,160 - Epoch: [122][ 760/ 1207] Overall Loss 0.247142 Objective Loss 0.247142 LR 0.000500 Time 0.019901 -2023-02-13 18:12:15,350 - Epoch: [122][ 770/ 1207] Overall Loss 0.247293 Objective Loss 0.247293 LR 0.000500 Time 0.019889 -2023-02-13 18:12:15,540 - Epoch: [122][ 780/ 1207] Overall Loss 0.247226 Objective Loss 0.247226 LR 0.000500 Time 0.019877 -2023-02-13 18:12:15,730 - Epoch: [122][ 790/ 1207] Overall Loss 0.247148 Objective Loss 0.247148 LR 0.000500 Time 0.019865 -2023-02-13 18:12:15,921 - Epoch: [122][ 800/ 1207] Overall Loss 0.246725 Objective Loss 0.246725 LR 0.000500 Time 0.019855 -2023-02-13 18:12:16,110 - Epoch: [122][ 810/ 1207] Overall Loss 0.246824 Objective Loss 0.246824 LR 0.000500 Time 0.019843 -2023-02-13 18:12:16,300 - Epoch: [122][ 820/ 1207] Overall Loss 0.246860 Objective Loss 0.246860 LR 0.000500 Time 0.019832 -2023-02-13 18:12:16,489 - Epoch: [122][ 830/ 1207] Overall Loss 0.246513 Objective Loss 0.246513 LR 0.000500 Time 0.019821 -2023-02-13 18:12:16,679 - Epoch: [122][ 840/ 1207] Overall Loss 0.246535 Objective Loss 0.246535 LR 0.000500 Time 0.019811 -2023-02-13 18:12:16,870 - Epoch: [122][ 850/ 1207] Overall Loss 0.246365 Objective Loss 0.246365 LR 0.000500 Time 0.019801 -2023-02-13 18:12:17,060 - Epoch: [122][ 860/ 1207] Overall Loss 0.246315 Objective Loss 0.246315 LR 0.000500 Time 0.019791 -2023-02-13 18:12:17,249 - Epoch: [122][ 870/ 1207] Overall Loss 0.245856 Objective Loss 0.245856 LR 0.000500 Time 0.019782 -2023-02-13 18:12:17,440 - Epoch: [122][ 880/ 1207] Overall Loss 0.245749 Objective Loss 0.245749 LR 0.000500 Time 0.019773 -2023-02-13 18:12:17,629 - Epoch: [122][ 890/ 1207] Overall Loss 0.245705 Objective Loss 0.245705 LR 0.000500 Time 0.019763 -2023-02-13 18:12:17,819 - Epoch: [122][ 900/ 1207] Overall Loss 0.245639 Objective Loss 0.245639 LR 0.000500 Time 0.019754 -2023-02-13 18:12:18,009 - Epoch: [122][ 910/ 1207] Overall Loss 0.245610 Objective Loss 0.245610 LR 0.000500 Time 0.019745 -2023-02-13 18:12:18,199 - Epoch: [122][ 920/ 1207] Overall Loss 0.245987 Objective Loss 0.245987 LR 0.000500 Time 0.019736 -2023-02-13 18:12:18,389 - Epoch: [122][ 930/ 1207] Overall Loss 0.245881 Objective Loss 0.245881 LR 0.000500 Time 0.019728 -2023-02-13 18:12:18,578 - Epoch: [122][ 940/ 1207] Overall Loss 0.245744 Objective Loss 0.245744 LR 0.000500 Time 0.019719 -2023-02-13 18:12:18,768 - Epoch: [122][ 950/ 1207] Overall Loss 0.245503 Objective Loss 0.245503 LR 0.000500 Time 0.019711 -2023-02-13 18:12:18,957 - Epoch: [122][ 960/ 1207] Overall Loss 0.245274 Objective Loss 0.245274 LR 0.000500 Time 0.019702 -2023-02-13 18:12:19,146 - Epoch: [122][ 970/ 1207] Overall Loss 0.245257 Objective Loss 0.245257 LR 0.000500 Time 0.019694 -2023-02-13 18:12:19,336 - Epoch: [122][ 980/ 1207] Overall Loss 0.245372 Objective Loss 0.245372 LR 0.000500 Time 0.019686 -2023-02-13 18:12:19,525 - Epoch: [122][ 990/ 1207] Overall Loss 0.245283 Objective Loss 0.245283 LR 0.000500 Time 0.019678 -2023-02-13 18:12:19,714 - Epoch: [122][ 1000/ 1207] Overall Loss 0.245420 Objective Loss 0.245420 LR 0.000500 Time 0.019670 -2023-02-13 18:12:19,904 - Epoch: [122][ 1010/ 1207] Overall Loss 0.245100 Objective Loss 0.245100 LR 0.000500 Time 0.019663 -2023-02-13 18:12:20,094 - Epoch: [122][ 1020/ 1207] Overall Loss 0.245324 Objective Loss 0.245324 LR 0.000500 Time 0.019656 -2023-02-13 18:12:20,283 - Epoch: [122][ 1030/ 1207] Overall Loss 0.245139 Objective Loss 0.245139 LR 0.000500 Time 0.019649 -2023-02-13 18:12:20,473 - Epoch: [122][ 1040/ 1207] Overall Loss 0.245135 Objective Loss 0.245135 LR 0.000500 Time 0.019642 -2023-02-13 18:12:20,663 - Epoch: [122][ 1050/ 1207] Overall Loss 0.244988 Objective Loss 0.244988 LR 0.000500 Time 0.019635 -2023-02-13 18:12:20,854 - Epoch: [122][ 1060/ 1207] Overall Loss 0.245384 Objective Loss 0.245384 LR 0.000500 Time 0.019630 -2023-02-13 18:12:21,044 - Epoch: [122][ 1070/ 1207] Overall Loss 0.245372 Objective Loss 0.245372 LR 0.000500 Time 0.019623 -2023-02-13 18:12:21,233 - Epoch: [122][ 1080/ 1207] Overall Loss 0.245655 Objective Loss 0.245655 LR 0.000500 Time 0.019617 -2023-02-13 18:12:21,423 - Epoch: [122][ 1090/ 1207] Overall Loss 0.245751 Objective Loss 0.245751 LR 0.000500 Time 0.019611 -2023-02-13 18:12:21,613 - Epoch: [122][ 1100/ 1207] Overall Loss 0.246106 Objective Loss 0.246106 LR 0.000500 Time 0.019604 -2023-02-13 18:12:21,803 - Epoch: [122][ 1110/ 1207] Overall Loss 0.246310 Objective Loss 0.246310 LR 0.000500 Time 0.019599 -2023-02-13 18:12:21,993 - Epoch: [122][ 1120/ 1207] Overall Loss 0.246322 Objective Loss 0.246322 LR 0.000500 Time 0.019593 -2023-02-13 18:12:22,183 - Epoch: [122][ 1130/ 1207] Overall Loss 0.246441 Objective Loss 0.246441 LR 0.000500 Time 0.019588 -2023-02-13 18:12:22,373 - Epoch: [122][ 1140/ 1207] Overall Loss 0.246449 Objective Loss 0.246449 LR 0.000500 Time 0.019582 -2023-02-13 18:12:22,562 - Epoch: [122][ 1150/ 1207] Overall Loss 0.246560 Objective Loss 0.246560 LR 0.000500 Time 0.019576 -2023-02-13 18:12:22,751 - Epoch: [122][ 1160/ 1207] Overall Loss 0.246621 Objective Loss 0.246621 LR 0.000500 Time 0.019570 -2023-02-13 18:12:22,941 - Epoch: [122][ 1170/ 1207] Overall Loss 0.246424 Objective Loss 0.246424 LR 0.000500 Time 0.019564 -2023-02-13 18:12:23,130 - Epoch: [122][ 1180/ 1207] Overall Loss 0.246282 Objective Loss 0.246282 LR 0.000500 Time 0.019559 -2023-02-13 18:12:23,320 - Epoch: [122][ 1190/ 1207] Overall Loss 0.246272 Objective Loss 0.246272 LR 0.000500 Time 0.019553 -2023-02-13 18:12:23,560 - Epoch: [122][ 1200/ 1207] Overall Loss 0.246331 Objective Loss 0.246331 LR 0.000500 Time 0.019590 -2023-02-13 18:12:23,676 - Epoch: [122][ 1207/ 1207] Overall Loss 0.246557 Objective Loss 0.246557 Top1 85.365854 Top5 98.475610 LR 0.000500 Time 0.019573 -2023-02-13 18:12:23,748 - --- validate (epoch=122)----------- -2023-02-13 18:12:23,748 - 34311 samples (256 per mini-batch) -2023-02-13 18:12:24,149 - Epoch: [122][ 10/ 135] Loss 0.347578 Top1 84.140625 Top5 97.070312 -2023-02-13 18:12:24,272 - Epoch: [122][ 20/ 135] Loss 0.340300 Top1 84.296875 Top5 97.558594 -2023-02-13 18:12:24,396 - Epoch: [122][ 30/ 135] Loss 0.331449 Top1 83.932292 Top5 97.486979 -2023-02-13 18:12:24,522 - Epoch: [122][ 40/ 135] Loss 0.324101 Top1 83.837891 Top5 97.558594 -2023-02-13 18:12:24,648 - Epoch: [122][ 50/ 135] Loss 0.323501 Top1 83.882812 Top5 97.585938 -2023-02-13 18:12:24,774 - Epoch: [122][ 60/ 135] Loss 0.318808 Top1 84.010417 Top5 97.597656 -2023-02-13 18:12:24,897 - Epoch: [122][ 70/ 135] Loss 0.317074 Top1 84.140625 Top5 97.633929 -2023-02-13 18:12:25,024 - Epoch: [122][ 80/ 135] Loss 0.313112 Top1 84.145508 Top5 97.651367 -2023-02-13 18:12:25,148 - Epoch: [122][ 90/ 135] Loss 0.312044 Top1 84.197049 Top5 97.647569 -2023-02-13 18:12:25,274 - Epoch: [122][ 100/ 135] Loss 0.312624 Top1 84.269531 Top5 97.640625 -2023-02-13 18:12:25,400 - Epoch: [122][ 110/ 135] Loss 0.311988 Top1 84.318182 Top5 97.652699 -2023-02-13 18:12:25,526 - Epoch: [122][ 120/ 135] Loss 0.311710 Top1 84.342448 Top5 97.659505 -2023-02-13 18:12:25,654 - Epoch: [122][ 130/ 135] Loss 0.312837 Top1 84.338942 Top5 97.638221 -2023-02-13 18:12:25,699 - Epoch: [122][ 135/ 135] Loss 0.311114 Top1 84.328641 Top5 97.645070 -2023-02-13 18:12:25,770 - ==> Top1: 84.329 Top5: 97.645 Loss: 0.311 - -2023-02-13 18:12:25,771 - ==> Confusion: -[[ 872 3 5 0 5 6 0 1 2 44 1 2 1 7 4 3 1 2 0 2 6] - [ 2 952 2 1 11 28 3 10 3 0 2 1 1 0 1 2 2 1 5 1 5] - [ 9 3 943 11 5 1 23 14 1 1 2 0 5 5 2 8 4 3 5 6 7] - [ 5 1 18 907 3 5 1 1 2 2 14 1 9 2 24 2 2 3 9 0 5] - [ 10 8 0 1 993 10 1 2 2 1 1 6 1 2 7 4 6 1 0 3 7] - [ 2 19 0 4 4 975 4 16 1 3 4 10 2 9 2 3 2 1 1 6 2] - [ 4 4 4 0 0 7 1053 3 3 1 2 1 0 2 1 3 0 2 1 4 4] - [ 2 8 7 2 2 25 6 923 1 2 2 5 2 2 0 0 1 1 19 7 7] - [ 17 2 1 2 1 0 0 3 885 52 7 1 3 6 16 2 1 1 4 2 3] - [ 79 1 2 0 5 0 0 3 31 857 0 1 0 18 5 1 4 1 1 1 2] - [ 2 2 3 11 0 1 4 3 21 0 980 2 1 5 3 0 1 0 7 0 5] - [ 2 2 1 0 2 13 1 4 1 2 0 918 26 7 0 4 3 9 2 6 2] - [ 2 0 2 4 2 3 0 2 2 0 0 24 878 0 2 9 4 15 1 0 9] - [ 2 2 0 0 9 7 2 0 13 21 12 3 1 928 3 7 2 2 1 1 8] - [ 6 2 3 15 5 3 0 2 21 6 2 2 2 0 999 1 2 6 7 1 7] - [ 4 1 7 0 10 0 3 2 0 0 0 6 12 3 1 961 12 13 0 5 6] - [ 4 6 0 2 11 3 1 1 2 0 0 2 3 4 2 5 998 2 3 4 8] - [ 5 4 1 3 1 3 1 0 1 1 1 6 18 1 1 19 0 981 0 0 4] - [ 2 1 5 12 0 2 1 22 6 2 7 3 3 0 14 1 1 1 995 3 5] - [ 0 4 1 0 1 8 7 12 1 0 1 15 4 5 0 4 9 1 0 1070 5] - [ 177 205 209 118 164 244 109 157 109 103 196 105 302 302 173 105 272 96 177 245 9866]] - -2023-02-13 18:12:25,773 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:12:25,773 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:12:25,779 - - -2023-02-13 18:12:25,779 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:12:26,771 - Epoch: [123][ 10/ 1207] Overall Loss 0.251537 Objective Loss 0.251537 LR 0.000500 Time 0.099201 -2023-02-13 18:12:26,970 - Epoch: [123][ 20/ 1207] Overall Loss 0.251964 Objective Loss 0.251964 LR 0.000500 Time 0.059515 -2023-02-13 18:12:27,159 - Epoch: [123][ 30/ 1207] Overall Loss 0.250438 Objective Loss 0.250438 LR 0.000500 Time 0.045968 -2023-02-13 18:12:27,348 - Epoch: [123][ 40/ 1207] Overall Loss 0.245386 Objective Loss 0.245386 LR 0.000500 Time 0.039191 -2023-02-13 18:12:27,537 - Epoch: [123][ 50/ 1207] Overall Loss 0.246914 Objective Loss 0.246914 LR 0.000500 Time 0.035129 -2023-02-13 18:12:27,726 - Epoch: [123][ 60/ 1207] Overall Loss 0.250179 Objective Loss 0.250179 LR 0.000500 Time 0.032411 -2023-02-13 18:12:27,915 - Epoch: [123][ 70/ 1207] Overall Loss 0.250363 Objective Loss 0.250363 LR 0.000500 Time 0.030473 -2023-02-13 18:12:28,103 - Epoch: [123][ 80/ 1207] Overall Loss 0.248699 Objective Loss 0.248699 LR 0.000500 Time 0.029017 -2023-02-13 18:12:28,292 - Epoch: [123][ 90/ 1207] Overall Loss 0.247864 Objective Loss 0.247864 LR 0.000500 Time 0.027883 -2023-02-13 18:12:28,482 - Epoch: [123][ 100/ 1207] Overall Loss 0.248386 Objective Loss 0.248386 LR 0.000500 Time 0.026992 -2023-02-13 18:12:28,670 - Epoch: [123][ 110/ 1207] Overall Loss 0.249173 Objective Loss 0.249173 LR 0.000500 Time 0.026246 -2023-02-13 18:12:28,858 - Epoch: [123][ 120/ 1207] Overall Loss 0.247355 Objective Loss 0.247355 LR 0.000500 Time 0.025623 -2023-02-13 18:12:29,047 - Epoch: [123][ 130/ 1207] Overall Loss 0.246799 Objective Loss 0.246799 LR 0.000500 Time 0.025098 -2023-02-13 18:12:29,236 - Epoch: [123][ 140/ 1207] Overall Loss 0.247909 Objective Loss 0.247909 LR 0.000500 Time 0.024653 -2023-02-13 18:12:29,425 - Epoch: [123][ 150/ 1207] Overall Loss 0.246585 Objective Loss 0.246585 LR 0.000500 Time 0.024269 -2023-02-13 18:12:29,613 - Epoch: [123][ 160/ 1207] Overall Loss 0.245983 Objective Loss 0.245983 LR 0.000500 Time 0.023925 -2023-02-13 18:12:29,802 - Epoch: [123][ 170/ 1207] Overall Loss 0.246456 Objective Loss 0.246456 LR 0.000500 Time 0.023628 -2023-02-13 18:12:29,991 - Epoch: [123][ 180/ 1207] Overall Loss 0.246643 Objective Loss 0.246643 LR 0.000500 Time 0.023361 -2023-02-13 18:12:30,180 - Epoch: [123][ 190/ 1207] Overall Loss 0.246800 Objective Loss 0.246800 LR 0.000500 Time 0.023126 -2023-02-13 18:12:30,369 - Epoch: [123][ 200/ 1207] Overall Loss 0.246494 Objective Loss 0.246494 LR 0.000500 Time 0.022914 -2023-02-13 18:12:30,565 - Epoch: [123][ 210/ 1207] Overall Loss 0.245743 Objective Loss 0.245743 LR 0.000500 Time 0.022754 -2023-02-13 18:12:30,761 - Epoch: [123][ 220/ 1207] Overall Loss 0.247153 Objective Loss 0.247153 LR 0.000500 Time 0.022606 -2023-02-13 18:12:30,959 - Epoch: [123][ 230/ 1207] Overall Loss 0.246498 Objective Loss 0.246498 LR 0.000500 Time 0.022483 -2023-02-13 18:12:31,152 - Epoch: [123][ 240/ 1207] Overall Loss 0.245891 Objective Loss 0.245891 LR 0.000500 Time 0.022351 -2023-02-13 18:12:31,348 - Epoch: [123][ 250/ 1207] Overall Loss 0.245939 Objective Loss 0.245939 LR 0.000500 Time 0.022238 -2023-02-13 18:12:31,544 - Epoch: [123][ 260/ 1207] Overall Loss 0.245278 Objective Loss 0.245278 LR 0.000500 Time 0.022135 -2023-02-13 18:12:31,740 - Epoch: [123][ 270/ 1207] Overall Loss 0.244808 Objective Loss 0.244808 LR 0.000500 Time 0.022041 -2023-02-13 18:12:31,935 - Epoch: [123][ 280/ 1207] Overall Loss 0.244578 Objective Loss 0.244578 LR 0.000500 Time 0.021947 -2023-02-13 18:12:32,131 - Epoch: [123][ 290/ 1207] Overall Loss 0.244535 Objective Loss 0.244535 LR 0.000500 Time 0.021866 -2023-02-13 18:12:32,326 - Epoch: [123][ 300/ 1207] Overall Loss 0.244167 Objective Loss 0.244167 LR 0.000500 Time 0.021786 -2023-02-13 18:12:32,523 - Epoch: [123][ 310/ 1207] Overall Loss 0.244038 Objective Loss 0.244038 LR 0.000500 Time 0.021719 -2023-02-13 18:12:32,718 - Epoch: [123][ 320/ 1207] Overall Loss 0.245027 Objective Loss 0.245027 LR 0.000500 Time 0.021645 -2023-02-13 18:12:32,915 - Epoch: [123][ 330/ 1207] Overall Loss 0.244693 Objective Loss 0.244693 LR 0.000500 Time 0.021588 -2023-02-13 18:12:33,109 - Epoch: [123][ 340/ 1207] Overall Loss 0.245453 Objective Loss 0.245453 LR 0.000500 Time 0.021520 -2023-02-13 18:12:33,306 - Epoch: [123][ 350/ 1207] Overall Loss 0.245211 Objective Loss 0.245211 LR 0.000500 Time 0.021468 -2023-02-13 18:12:33,501 - Epoch: [123][ 360/ 1207] Overall Loss 0.244743 Objective Loss 0.244743 LR 0.000500 Time 0.021412 -2023-02-13 18:12:33,697 - Epoch: [123][ 370/ 1207] Overall Loss 0.244542 Objective Loss 0.244542 LR 0.000500 Time 0.021361 -2023-02-13 18:12:33,891 - Epoch: [123][ 380/ 1207] Overall Loss 0.244149 Objective Loss 0.244149 LR 0.000500 Time 0.021310 -2023-02-13 18:12:34,087 - Epoch: [123][ 390/ 1207] Overall Loss 0.243665 Objective Loss 0.243665 LR 0.000500 Time 0.021265 -2023-02-13 18:12:34,282 - Epoch: [123][ 400/ 1207] Overall Loss 0.242840 Objective Loss 0.242840 LR 0.000500 Time 0.021221 -2023-02-13 18:12:34,479 - Epoch: [123][ 410/ 1207] Overall Loss 0.242505 Objective Loss 0.242505 LR 0.000500 Time 0.021183 -2023-02-13 18:12:34,674 - Epoch: [123][ 420/ 1207] Overall Loss 0.242137 Objective Loss 0.242137 LR 0.000500 Time 0.021141 -2023-02-13 18:12:34,870 - Epoch: [123][ 430/ 1207] Overall Loss 0.242492 Objective Loss 0.242492 LR 0.000500 Time 0.021106 -2023-02-13 18:12:35,066 - Epoch: [123][ 440/ 1207] Overall Loss 0.241678 Objective Loss 0.241678 LR 0.000500 Time 0.021071 -2023-02-13 18:12:35,264 - Epoch: [123][ 450/ 1207] Overall Loss 0.241450 Objective Loss 0.241450 LR 0.000500 Time 0.021040 -2023-02-13 18:12:35,461 - Epoch: [123][ 460/ 1207] Overall Loss 0.241520 Objective Loss 0.241520 LR 0.000500 Time 0.021010 -2023-02-13 18:12:35,658 - Epoch: [123][ 470/ 1207] Overall Loss 0.241615 Objective Loss 0.241615 LR 0.000500 Time 0.020983 -2023-02-13 18:12:35,854 - Epoch: [123][ 480/ 1207] Overall Loss 0.241448 Objective Loss 0.241448 LR 0.000500 Time 0.020953 -2023-02-13 18:12:36,053 - Epoch: [123][ 490/ 1207] Overall Loss 0.241635 Objective Loss 0.241635 LR 0.000500 Time 0.020929 -2023-02-13 18:12:36,248 - Epoch: [123][ 500/ 1207] Overall Loss 0.241214 Objective Loss 0.241214 LR 0.000500 Time 0.020901 -2023-02-13 18:12:36,446 - Epoch: [123][ 510/ 1207] Overall Loss 0.240967 Objective Loss 0.240967 LR 0.000500 Time 0.020878 -2023-02-13 18:12:36,642 - Epoch: [123][ 520/ 1207] Overall Loss 0.241065 Objective Loss 0.241065 LR 0.000500 Time 0.020852 -2023-02-13 18:12:36,839 - Epoch: [123][ 530/ 1207] Overall Loss 0.241129 Objective Loss 0.241129 LR 0.000500 Time 0.020831 -2023-02-13 18:12:37,034 - Epoch: [123][ 540/ 1207] Overall Loss 0.241168 Objective Loss 0.241168 LR 0.000500 Time 0.020804 -2023-02-13 18:12:37,228 - Epoch: [123][ 550/ 1207] Overall Loss 0.241177 Objective Loss 0.241177 LR 0.000500 Time 0.020779 -2023-02-13 18:12:37,421 - Epoch: [123][ 560/ 1207] Overall Loss 0.241745 Objective Loss 0.241745 LR 0.000500 Time 0.020752 -2023-02-13 18:12:37,616 - Epoch: [123][ 570/ 1207] Overall Loss 0.241871 Objective Loss 0.241871 LR 0.000500 Time 0.020730 -2023-02-13 18:12:37,810 - Epoch: [123][ 580/ 1207] Overall Loss 0.241818 Objective Loss 0.241818 LR 0.000500 Time 0.020706 -2023-02-13 18:12:38,004 - Epoch: [123][ 590/ 1207] Overall Loss 0.242078 Objective Loss 0.242078 LR 0.000500 Time 0.020683 -2023-02-13 18:12:38,198 - Epoch: [123][ 600/ 1207] Overall Loss 0.242208 Objective Loss 0.242208 LR 0.000500 Time 0.020661 -2023-02-13 18:12:38,393 - Epoch: [123][ 610/ 1207] Overall Loss 0.242169 Objective Loss 0.242169 LR 0.000500 Time 0.020641 -2023-02-13 18:12:38,588 - Epoch: [123][ 620/ 1207] Overall Loss 0.241841 Objective Loss 0.241841 LR 0.000500 Time 0.020622 -2023-02-13 18:12:38,784 - Epoch: [123][ 630/ 1207] Overall Loss 0.242013 Objective Loss 0.242013 LR 0.000500 Time 0.020605 -2023-02-13 18:12:38,978 - Epoch: [123][ 640/ 1207] Overall Loss 0.242217 Objective Loss 0.242217 LR 0.000500 Time 0.020586 -2023-02-13 18:12:39,173 - Epoch: [123][ 650/ 1207] Overall Loss 0.242418 Objective Loss 0.242418 LR 0.000500 Time 0.020568 -2023-02-13 18:12:39,366 - Epoch: [123][ 660/ 1207] Overall Loss 0.242198 Objective Loss 0.242198 LR 0.000500 Time 0.020550 -2023-02-13 18:12:39,563 - Epoch: [123][ 670/ 1207] Overall Loss 0.241949 Objective Loss 0.241949 LR 0.000500 Time 0.020535 -2023-02-13 18:12:39,757 - Epoch: [123][ 680/ 1207] Overall Loss 0.242473 Objective Loss 0.242473 LR 0.000500 Time 0.020518 -2023-02-13 18:12:39,952 - Epoch: [123][ 690/ 1207] Overall Loss 0.242459 Objective Loss 0.242459 LR 0.000500 Time 0.020504 -2023-02-13 18:12:40,146 - Epoch: [123][ 700/ 1207] Overall Loss 0.242156 Objective Loss 0.242156 LR 0.000500 Time 0.020487 -2023-02-13 18:12:40,341 - Epoch: [123][ 710/ 1207] Overall Loss 0.241912 Objective Loss 0.241912 LR 0.000500 Time 0.020472 -2023-02-13 18:12:40,535 - Epoch: [123][ 720/ 1207] Overall Loss 0.241505 Objective Loss 0.241505 LR 0.000500 Time 0.020457 -2023-02-13 18:12:40,729 - Epoch: [123][ 730/ 1207] Overall Loss 0.242311 Objective Loss 0.242311 LR 0.000500 Time 0.020442 -2023-02-13 18:12:40,924 - Epoch: [123][ 740/ 1207] Overall Loss 0.241988 Objective Loss 0.241988 LR 0.000500 Time 0.020428 -2023-02-13 18:12:41,119 - Epoch: [123][ 750/ 1207] Overall Loss 0.241742 Objective Loss 0.241742 LR 0.000500 Time 0.020416 -2023-02-13 18:12:41,313 - Epoch: [123][ 760/ 1207] Overall Loss 0.241385 Objective Loss 0.241385 LR 0.000500 Time 0.020402 -2023-02-13 18:12:41,508 - Epoch: [123][ 770/ 1207] Overall Loss 0.241349 Objective Loss 0.241349 LR 0.000500 Time 0.020390 -2023-02-13 18:12:41,702 - Epoch: [123][ 780/ 1207] Overall Loss 0.241119 Objective Loss 0.241119 LR 0.000500 Time 0.020377 -2023-02-13 18:12:41,898 - Epoch: [123][ 790/ 1207] Overall Loss 0.241493 Objective Loss 0.241493 LR 0.000500 Time 0.020367 -2023-02-13 18:12:42,092 - Epoch: [123][ 800/ 1207] Overall Loss 0.241445 Objective Loss 0.241445 LR 0.000500 Time 0.020354 -2023-02-13 18:12:42,285 - Epoch: [123][ 810/ 1207] Overall Loss 0.241468 Objective Loss 0.241468 LR 0.000500 Time 0.020341 -2023-02-13 18:12:42,475 - Epoch: [123][ 820/ 1207] Overall Loss 0.241199 Objective Loss 0.241199 LR 0.000500 Time 0.020324 -2023-02-13 18:12:42,665 - Epoch: [123][ 830/ 1207] Overall Loss 0.240884 Objective Loss 0.240884 LR 0.000500 Time 0.020307 -2023-02-13 18:12:42,855 - Epoch: [123][ 840/ 1207] Overall Loss 0.240856 Objective Loss 0.240856 LR 0.000500 Time 0.020292 -2023-02-13 18:12:43,045 - Epoch: [123][ 850/ 1207] Overall Loss 0.240653 Objective Loss 0.240653 LR 0.000500 Time 0.020276 -2023-02-13 18:12:43,234 - Epoch: [123][ 860/ 1207] Overall Loss 0.240430 Objective Loss 0.240430 LR 0.000500 Time 0.020260 -2023-02-13 18:12:43,424 - Epoch: [123][ 870/ 1207] Overall Loss 0.240491 Objective Loss 0.240491 LR 0.000500 Time 0.020244 -2023-02-13 18:12:43,614 - Epoch: [123][ 880/ 1207] Overall Loss 0.240523 Objective Loss 0.240523 LR 0.000500 Time 0.020229 -2023-02-13 18:12:43,803 - Epoch: [123][ 890/ 1207] Overall Loss 0.240421 Objective Loss 0.240421 LR 0.000500 Time 0.020215 -2023-02-13 18:12:43,993 - Epoch: [123][ 900/ 1207] Overall Loss 0.240226 Objective Loss 0.240226 LR 0.000500 Time 0.020201 -2023-02-13 18:12:44,183 - Epoch: [123][ 910/ 1207] Overall Loss 0.240504 Objective Loss 0.240504 LR 0.000500 Time 0.020187 -2023-02-13 18:12:44,372 - Epoch: [123][ 920/ 1207] Overall Loss 0.240673 Objective Loss 0.240673 LR 0.000500 Time 0.020173 -2023-02-13 18:12:44,562 - Epoch: [123][ 930/ 1207] Overall Loss 0.240634 Objective Loss 0.240634 LR 0.000500 Time 0.020160 -2023-02-13 18:12:44,751 - Epoch: [123][ 940/ 1207] Overall Loss 0.240561 Objective Loss 0.240561 LR 0.000500 Time 0.020146 -2023-02-13 18:12:44,941 - Epoch: [123][ 950/ 1207] Overall Loss 0.240474 Objective Loss 0.240474 LR 0.000500 Time 0.020133 -2023-02-13 18:12:45,130 - Epoch: [123][ 960/ 1207] Overall Loss 0.240580 Objective Loss 0.240580 LR 0.000500 Time 0.020120 -2023-02-13 18:12:45,320 - Epoch: [123][ 970/ 1207] Overall Loss 0.241004 Objective Loss 0.241004 LR 0.000500 Time 0.020108 -2023-02-13 18:12:45,510 - Epoch: [123][ 980/ 1207] Overall Loss 0.240935 Objective Loss 0.240935 LR 0.000500 Time 0.020096 -2023-02-13 18:12:45,699 - Epoch: [123][ 990/ 1207] Overall Loss 0.241000 Objective Loss 0.241000 LR 0.000500 Time 0.020084 -2023-02-13 18:12:45,889 - Epoch: [123][ 1000/ 1207] Overall Loss 0.240931 Objective Loss 0.240931 LR 0.000500 Time 0.020073 -2023-02-13 18:12:46,079 - Epoch: [123][ 1010/ 1207] Overall Loss 0.241021 Objective Loss 0.241021 LR 0.000500 Time 0.020062 -2023-02-13 18:12:46,269 - Epoch: [123][ 1020/ 1207] Overall Loss 0.240989 Objective Loss 0.240989 LR 0.000500 Time 0.020051 -2023-02-13 18:12:46,459 - Epoch: [123][ 1030/ 1207] Overall Loss 0.240947 Objective Loss 0.240947 LR 0.000500 Time 0.020040 -2023-02-13 18:12:46,649 - Epoch: [123][ 1040/ 1207] Overall Loss 0.241139 Objective Loss 0.241139 LR 0.000500 Time 0.020030 -2023-02-13 18:12:46,838 - Epoch: [123][ 1050/ 1207] Overall Loss 0.241228 Objective Loss 0.241228 LR 0.000500 Time 0.020019 -2023-02-13 18:12:47,029 - Epoch: [123][ 1060/ 1207] Overall Loss 0.241115 Objective Loss 0.241115 LR 0.000500 Time 0.020010 -2023-02-13 18:12:47,218 - Epoch: [123][ 1070/ 1207] Overall Loss 0.241245 Objective Loss 0.241245 LR 0.000500 Time 0.019999 -2023-02-13 18:12:47,408 - Epoch: [123][ 1080/ 1207] Overall Loss 0.241184 Objective Loss 0.241184 LR 0.000500 Time 0.019990 -2023-02-13 18:12:47,598 - Epoch: [123][ 1090/ 1207] Overall Loss 0.241109 Objective Loss 0.241109 LR 0.000500 Time 0.019980 -2023-02-13 18:12:47,788 - Epoch: [123][ 1100/ 1207] Overall Loss 0.241044 Objective Loss 0.241044 LR 0.000500 Time 0.019970 -2023-02-13 18:12:47,977 - Epoch: [123][ 1110/ 1207] Overall Loss 0.241263 Objective Loss 0.241263 LR 0.000500 Time 0.019961 -2023-02-13 18:12:48,167 - Epoch: [123][ 1120/ 1207] Overall Loss 0.241350 Objective Loss 0.241350 LR 0.000500 Time 0.019952 -2023-02-13 18:12:48,357 - Epoch: [123][ 1130/ 1207] Overall Loss 0.241647 Objective Loss 0.241647 LR 0.000500 Time 0.019944 -2023-02-13 18:12:48,547 - Epoch: [123][ 1140/ 1207] Overall Loss 0.241789 Objective Loss 0.241789 LR 0.000500 Time 0.019935 -2023-02-13 18:12:48,736 - Epoch: [123][ 1150/ 1207] Overall Loss 0.241796 Objective Loss 0.241796 LR 0.000500 Time 0.019926 -2023-02-13 18:12:48,927 - Epoch: [123][ 1160/ 1207] Overall Loss 0.242116 Objective Loss 0.242116 LR 0.000500 Time 0.019917 -2023-02-13 18:12:49,116 - Epoch: [123][ 1170/ 1207] Overall Loss 0.242368 Objective Loss 0.242368 LR 0.000500 Time 0.019909 -2023-02-13 18:12:49,306 - Epoch: [123][ 1180/ 1207] Overall Loss 0.242470 Objective Loss 0.242470 LR 0.000500 Time 0.019901 -2023-02-13 18:12:49,495 - Epoch: [123][ 1190/ 1207] Overall Loss 0.242321 Objective Loss 0.242321 LR 0.000500 Time 0.019892 -2023-02-13 18:12:49,741 - Epoch: [123][ 1200/ 1207] Overall Loss 0.242542 Objective Loss 0.242542 LR 0.000500 Time 0.019931 -2023-02-13 18:12:49,856 - Epoch: [123][ 1207/ 1207] Overall Loss 0.242685 Objective Loss 0.242685 Top1 85.670732 Top5 99.390244 LR 0.000500 Time 0.019910 -2023-02-13 18:12:49,927 - --- validate (epoch=123)----------- -2023-02-13 18:12:49,928 - 34311 samples (256 per mini-batch) -2023-02-13 18:12:50,323 - Epoch: [123][ 10/ 135] Loss 0.323478 Top1 84.414062 Top5 97.578125 -2023-02-13 18:12:50,454 - Epoch: [123][ 20/ 135] Loss 0.324164 Top1 84.277344 Top5 97.480469 -2023-02-13 18:12:50,584 - Epoch: [123][ 30/ 135] Loss 0.308402 Top1 83.984375 Top5 97.382812 -2023-02-13 18:12:50,714 - Epoch: [123][ 40/ 135] Loss 0.302185 Top1 84.160156 Top5 97.597656 -2023-02-13 18:12:50,844 - Epoch: [123][ 50/ 135] Loss 0.305425 Top1 84.085938 Top5 97.656250 -2023-02-13 18:12:50,971 - Epoch: [123][ 60/ 135] Loss 0.304785 Top1 84.114583 Top5 97.649740 -2023-02-13 18:12:51,100 - Epoch: [123][ 70/ 135] Loss 0.310814 Top1 84.045759 Top5 97.600446 -2023-02-13 18:12:51,229 - Epoch: [123][ 80/ 135] Loss 0.316693 Top1 83.979492 Top5 97.646484 -2023-02-13 18:12:51,360 - Epoch: [123][ 90/ 135] Loss 0.318343 Top1 83.888889 Top5 97.647569 -2023-02-13 18:12:51,487 - Epoch: [123][ 100/ 135] Loss 0.322874 Top1 83.843750 Top5 97.636719 -2023-02-13 18:12:51,612 - Epoch: [123][ 110/ 135] Loss 0.323632 Top1 83.892045 Top5 97.613636 -2023-02-13 18:12:51,735 - Epoch: [123][ 120/ 135] Loss 0.323603 Top1 83.893229 Top5 97.626953 -2023-02-13 18:12:51,862 - Epoch: [123][ 130/ 135] Loss 0.322180 Top1 83.969351 Top5 97.590144 -2023-02-13 18:12:51,906 - Epoch: [123][ 135/ 135] Loss 0.323156 Top1 83.967241 Top5 97.610096 -2023-02-13 18:12:51,975 - ==> Top1: 83.967 Top5: 97.610 Loss: 0.323 - -2023-02-13 18:12:51,976 - ==> Confusion: -[[ 896 4 1 1 3 3 0 1 4 15 2 8 0 5 6 6 5 2 2 1 2] - [ 3 937 3 1 8 30 2 16 4 0 1 0 2 0 1 1 9 2 3 0 10] - [ 10 6 940 13 4 1 20 10 0 2 2 3 2 5 3 10 2 2 11 7 5] - [ 3 1 20 901 3 4 2 2 1 2 13 0 7 0 24 1 6 6 17 1 2] - [ 18 8 0 1 978 12 0 4 2 2 0 6 1 1 6 12 8 3 0 1 3] - [ 0 12 1 6 6 976 3 20 2 3 2 6 4 7 3 2 6 2 0 5 4] - [ 5 1 15 1 1 10 1029 4 0 0 4 1 2 1 0 9 0 3 1 7 5] - [ 2 12 7 2 3 26 0 930 0 1 0 4 3 1 0 0 2 2 18 7 4] - [ 19 2 1 1 0 0 0 2 904 33 6 1 1 11 15 3 3 2 3 1 1] - [ 130 1 3 0 4 1 0 2 32 800 0 2 1 15 6 1 1 3 2 1 7] - [ 2 3 5 8 0 4 3 5 21 1 969 0 2 13 2 0 0 1 7 1 4] - [ 0 4 1 1 1 14 0 7 2 0 0 911 18 8 1 8 5 4 3 15 2] - [ 2 0 1 3 0 5 0 0 2 0 1 34 865 0 2 15 5 13 1 1 9] - [ 4 4 2 0 6 9 0 3 11 15 8 6 5 925 3 8 5 4 0 2 4] - [ 6 2 3 12 3 2 0 1 23 4 1 3 5 1 1001 3 3 5 9 0 5] - [ 1 2 6 1 5 1 5 1 0 1 0 7 6 3 0 975 10 9 1 6 6] - [ 3 3 1 1 7 1 0 0 1 0 0 2 4 3 1 6 1014 3 0 3 8] - [ 5 4 2 5 0 2 1 2 1 0 1 9 15 1 1 24 0 973 0 2 3] - [ 4 6 4 10 0 2 1 33 5 0 3 3 4 0 12 1 2 3 989 2 2] - [ 0 5 2 0 2 8 6 10 1 0 0 15 2 2 1 7 8 2 1 1068 8] - [ 177 218 230 127 126 230 86 171 103 71 148 139 304 290 171 128 323 100 184 280 9828]] - -2023-02-13 18:12:51,977 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:12:51,977 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:12:51,983 - - -2023-02-13 18:12:51,983 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:12:52,874 - Epoch: [124][ 10/ 1207] Overall Loss 0.230869 Objective Loss 0.230869 LR 0.000500 Time 0.089016 -2023-02-13 18:12:53,073 - Epoch: [124][ 20/ 1207] Overall Loss 0.224108 Objective Loss 0.224108 LR 0.000500 Time 0.054474 -2023-02-13 18:12:53,269 - Epoch: [124][ 30/ 1207] Overall Loss 0.227910 Objective Loss 0.227910 LR 0.000500 Time 0.042835 -2023-02-13 18:12:53,462 - Epoch: [124][ 40/ 1207] Overall Loss 0.232103 Objective Loss 0.232103 LR 0.000500 Time 0.036945 -2023-02-13 18:12:53,657 - Epoch: [124][ 50/ 1207] Overall Loss 0.231127 Objective Loss 0.231127 LR 0.000500 Time 0.033442 -2023-02-13 18:12:53,850 - Epoch: [124][ 60/ 1207] Overall Loss 0.231221 Objective Loss 0.231221 LR 0.000500 Time 0.031071 -2023-02-13 18:12:54,044 - Epoch: [124][ 70/ 1207] Overall Loss 0.232603 Objective Loss 0.232603 LR 0.000500 Time 0.029410 -2023-02-13 18:12:54,237 - Epoch: [124][ 80/ 1207] Overall Loss 0.232632 Objective Loss 0.232632 LR 0.000500 Time 0.028134 -2023-02-13 18:12:54,432 - Epoch: [124][ 90/ 1207] Overall Loss 0.229953 Objective Loss 0.229953 LR 0.000500 Time 0.027168 -2023-02-13 18:12:54,625 - Epoch: [124][ 100/ 1207] Overall Loss 0.233649 Objective Loss 0.233649 LR 0.000500 Time 0.026383 -2023-02-13 18:12:54,820 - Epoch: [124][ 110/ 1207] Overall Loss 0.234705 Objective Loss 0.234705 LR 0.000500 Time 0.025749 -2023-02-13 18:12:55,012 - Epoch: [124][ 120/ 1207] Overall Loss 0.235900 Objective Loss 0.235900 LR 0.000500 Time 0.025207 -2023-02-13 18:12:55,207 - Epoch: [124][ 130/ 1207] Overall Loss 0.236027 Objective Loss 0.236027 LR 0.000500 Time 0.024764 -2023-02-13 18:12:55,400 - Epoch: [124][ 140/ 1207] Overall Loss 0.236611 Objective Loss 0.236611 LR 0.000500 Time 0.024371 -2023-02-13 18:12:55,596 - Epoch: [124][ 150/ 1207] Overall Loss 0.235366 Objective Loss 0.235366 LR 0.000500 Time 0.024047 -2023-02-13 18:12:55,789 - Epoch: [124][ 160/ 1207] Overall Loss 0.236397 Objective Loss 0.236397 LR 0.000500 Time 0.023748 -2023-02-13 18:12:55,985 - Epoch: [124][ 170/ 1207] Overall Loss 0.239339 Objective Loss 0.239339 LR 0.000500 Time 0.023505 -2023-02-13 18:12:56,178 - Epoch: [124][ 180/ 1207] Overall Loss 0.239928 Objective Loss 0.239928 LR 0.000500 Time 0.023269 -2023-02-13 18:12:56,373 - Epoch: [124][ 190/ 1207] Overall Loss 0.240591 Objective Loss 0.240591 LR 0.000500 Time 0.023070 -2023-02-13 18:12:56,570 - Epoch: [124][ 200/ 1207] Overall Loss 0.241253 Objective Loss 0.241253 LR 0.000500 Time 0.022899 -2023-02-13 18:12:56,764 - Epoch: [124][ 210/ 1207] Overall Loss 0.239886 Objective Loss 0.239886 LR 0.000500 Time 0.022730 -2023-02-13 18:12:56,961 - Epoch: [124][ 220/ 1207] Overall Loss 0.239504 Objective Loss 0.239504 LR 0.000500 Time 0.022588 -2023-02-13 18:12:57,154 - Epoch: [124][ 230/ 1207] Overall Loss 0.239500 Objective Loss 0.239500 LR 0.000500 Time 0.022445 -2023-02-13 18:12:57,351 - Epoch: [124][ 240/ 1207] Overall Loss 0.239087 Objective Loss 0.239087 LR 0.000500 Time 0.022329 -2023-02-13 18:12:57,545 - Epoch: [124][ 250/ 1207] Overall Loss 0.238287 Objective Loss 0.238287 LR 0.000500 Time 0.022209 -2023-02-13 18:12:57,741 - Epoch: [124][ 260/ 1207] Overall Loss 0.238403 Objective Loss 0.238403 LR 0.000500 Time 0.022108 -2023-02-13 18:12:57,941 - Epoch: [124][ 270/ 1207] Overall Loss 0.239180 Objective Loss 0.239180 LR 0.000500 Time 0.022030 -2023-02-13 18:12:58,152 - Epoch: [124][ 280/ 1207] Overall Loss 0.239997 Objective Loss 0.239997 LR 0.000500 Time 0.021996 -2023-02-13 18:12:58,356 - Epoch: [124][ 290/ 1207] Overall Loss 0.240525 Objective Loss 0.240525 LR 0.000500 Time 0.021940 -2023-02-13 18:12:58,567 - Epoch: [124][ 300/ 1207] Overall Loss 0.240918 Objective Loss 0.240918 LR 0.000500 Time 0.021911 -2023-02-13 18:12:58,767 - Epoch: [124][ 310/ 1207] Overall Loss 0.240569 Objective Loss 0.240569 LR 0.000500 Time 0.021846 -2023-02-13 18:12:58,970 - Epoch: [124][ 320/ 1207] Overall Loss 0.241380 Objective Loss 0.241380 LR 0.000500 Time 0.021796 -2023-02-13 18:12:59,168 - Epoch: [124][ 330/ 1207] Overall Loss 0.241174 Objective Loss 0.241174 LR 0.000500 Time 0.021736 -2023-02-13 18:12:59,370 - Epoch: [124][ 340/ 1207] Overall Loss 0.241066 Objective Loss 0.241066 LR 0.000500 Time 0.021690 -2023-02-13 18:12:59,570 - Epoch: [124][ 350/ 1207] Overall Loss 0.240886 Objective Loss 0.240886 LR 0.000500 Time 0.021641 -2023-02-13 18:12:59,773 - Epoch: [124][ 360/ 1207] Overall Loss 0.240445 Objective Loss 0.240445 LR 0.000500 Time 0.021602 -2023-02-13 18:12:59,971 - Epoch: [124][ 370/ 1207] Overall Loss 0.240764 Objective Loss 0.240764 LR 0.000500 Time 0.021552 -2023-02-13 18:13:00,173 - Epoch: [124][ 380/ 1207] Overall Loss 0.240513 Objective Loss 0.240513 LR 0.000500 Time 0.021516 -2023-02-13 18:13:00,373 - Epoch: [124][ 390/ 1207] Overall Loss 0.240805 Objective Loss 0.240805 LR 0.000500 Time 0.021476 -2023-02-13 18:13:00,576 - Epoch: [124][ 400/ 1207] Overall Loss 0.241095 Objective Loss 0.241095 LR 0.000500 Time 0.021445 -2023-02-13 18:13:00,774 - Epoch: [124][ 410/ 1207] Overall Loss 0.241029 Objective Loss 0.241029 LR 0.000500 Time 0.021405 -2023-02-13 18:13:00,978 - Epoch: [124][ 420/ 1207] Overall Loss 0.240979 Objective Loss 0.240979 LR 0.000500 Time 0.021380 -2023-02-13 18:13:01,177 - Epoch: [124][ 430/ 1207] Overall Loss 0.241495 Objective Loss 0.241495 LR 0.000500 Time 0.021343 -2023-02-13 18:13:01,380 - Epoch: [124][ 440/ 1207] Overall Loss 0.241412 Objective Loss 0.241412 LR 0.000500 Time 0.021319 -2023-02-13 18:13:01,579 - Epoch: [124][ 450/ 1207] Overall Loss 0.241625 Objective Loss 0.241625 LR 0.000500 Time 0.021288 -2023-02-13 18:13:01,782 - Epoch: [124][ 460/ 1207] Overall Loss 0.241529 Objective Loss 0.241529 LR 0.000500 Time 0.021264 -2023-02-13 18:13:01,980 - Epoch: [124][ 470/ 1207] Overall Loss 0.241405 Objective Loss 0.241405 LR 0.000500 Time 0.021234 -2023-02-13 18:13:02,183 - Epoch: [124][ 480/ 1207] Overall Loss 0.241614 Objective Loss 0.241614 LR 0.000500 Time 0.021213 -2023-02-13 18:13:02,383 - Epoch: [124][ 490/ 1207] Overall Loss 0.241419 Objective Loss 0.241419 LR 0.000500 Time 0.021187 -2023-02-13 18:13:02,590 - Epoch: [124][ 500/ 1207] Overall Loss 0.241325 Objective Loss 0.241325 LR 0.000500 Time 0.021176 -2023-02-13 18:13:02,790 - Epoch: [124][ 510/ 1207] Overall Loss 0.241341 Objective Loss 0.241341 LR 0.000500 Time 0.021153 -2023-02-13 18:13:02,994 - Epoch: [124][ 520/ 1207] Overall Loss 0.241192 Objective Loss 0.241192 LR 0.000500 Time 0.021137 -2023-02-13 18:13:03,192 - Epoch: [124][ 530/ 1207] Overall Loss 0.241119 Objective Loss 0.241119 LR 0.000500 Time 0.021112 -2023-02-13 18:13:03,396 - Epoch: [124][ 540/ 1207] Overall Loss 0.241077 Objective Loss 0.241077 LR 0.000500 Time 0.021097 -2023-02-13 18:13:03,594 - Epoch: [124][ 550/ 1207] Overall Loss 0.240767 Objective Loss 0.240767 LR 0.000500 Time 0.021074 -2023-02-13 18:13:03,797 - Epoch: [124][ 560/ 1207] Overall Loss 0.240650 Objective Loss 0.240650 LR 0.000500 Time 0.021059 -2023-02-13 18:13:03,997 - Epoch: [124][ 570/ 1207] Overall Loss 0.240901 Objective Loss 0.240901 LR 0.000500 Time 0.021039 -2023-02-13 18:13:04,200 - Epoch: [124][ 580/ 1207] Overall Loss 0.241514 Objective Loss 0.241514 LR 0.000500 Time 0.021025 -2023-02-13 18:13:04,398 - Epoch: [124][ 590/ 1207] Overall Loss 0.241658 Objective Loss 0.241658 LR 0.000500 Time 0.021005 -2023-02-13 18:13:04,601 - Epoch: [124][ 600/ 1207] Overall Loss 0.242258 Objective Loss 0.242258 LR 0.000500 Time 0.020992 -2023-02-13 18:13:04,800 - Epoch: [124][ 610/ 1207] Overall Loss 0.242229 Objective Loss 0.242229 LR 0.000500 Time 0.020973 -2023-02-13 18:13:05,004 - Epoch: [124][ 620/ 1207] Overall Loss 0.242523 Objective Loss 0.242523 LR 0.000500 Time 0.020963 -2023-02-13 18:13:05,202 - Epoch: [124][ 630/ 1207] Overall Loss 0.242518 Objective Loss 0.242518 LR 0.000500 Time 0.020944 -2023-02-13 18:13:05,405 - Epoch: [124][ 640/ 1207] Overall Loss 0.242854 Objective Loss 0.242854 LR 0.000500 Time 0.020934 -2023-02-13 18:13:05,605 - Epoch: [124][ 650/ 1207] Overall Loss 0.243353 Objective Loss 0.243353 LR 0.000500 Time 0.020919 -2023-02-13 18:13:05,808 - Epoch: [124][ 660/ 1207] Overall Loss 0.243480 Objective Loss 0.243480 LR 0.000500 Time 0.020909 -2023-02-13 18:13:06,008 - Epoch: [124][ 670/ 1207] Overall Loss 0.243378 Objective Loss 0.243378 LR 0.000500 Time 0.020895 -2023-02-13 18:13:06,211 - Epoch: [124][ 680/ 1207] Overall Loss 0.243797 Objective Loss 0.243797 LR 0.000500 Time 0.020886 -2023-02-13 18:13:06,410 - Epoch: [124][ 690/ 1207] Overall Loss 0.243729 Objective Loss 0.243729 LR 0.000500 Time 0.020870 -2023-02-13 18:13:06,614 - Epoch: [124][ 700/ 1207] Overall Loss 0.243565 Objective Loss 0.243565 LR 0.000500 Time 0.020863 -2023-02-13 18:13:06,815 - Epoch: [124][ 710/ 1207] Overall Loss 0.243322 Objective Loss 0.243322 LR 0.000500 Time 0.020852 -2023-02-13 18:13:07,021 - Epoch: [124][ 720/ 1207] Overall Loss 0.243874 Objective Loss 0.243874 LR 0.000500 Time 0.020848 -2023-02-13 18:13:07,225 - Epoch: [124][ 730/ 1207] Overall Loss 0.243965 Objective Loss 0.243965 LR 0.000500 Time 0.020842 -2023-02-13 18:13:07,431 - Epoch: [124][ 740/ 1207] Overall Loss 0.243983 Objective Loss 0.243983 LR 0.000500 Time 0.020838 -2023-02-13 18:13:07,636 - Epoch: [124][ 750/ 1207] Overall Loss 0.244094 Objective Loss 0.244094 LR 0.000500 Time 0.020833 -2023-02-13 18:13:07,842 - Epoch: [124][ 760/ 1207] Overall Loss 0.243971 Objective Loss 0.243971 LR 0.000500 Time 0.020830 -2023-02-13 18:13:08,047 - Epoch: [124][ 770/ 1207] Overall Loss 0.244057 Objective Loss 0.244057 LR 0.000500 Time 0.020825 -2023-02-13 18:13:08,252 - Epoch: [124][ 780/ 1207] Overall Loss 0.244240 Objective Loss 0.244240 LR 0.000500 Time 0.020820 -2023-02-13 18:13:08,456 - Epoch: [124][ 790/ 1207] Overall Loss 0.244102 Objective Loss 0.244102 LR 0.000500 Time 0.020814 -2023-02-13 18:13:08,662 - Epoch: [124][ 800/ 1207] Overall Loss 0.243987 Objective Loss 0.243987 LR 0.000500 Time 0.020811 -2023-02-13 18:13:08,866 - Epoch: [124][ 810/ 1207] Overall Loss 0.243608 Objective Loss 0.243608 LR 0.000500 Time 0.020805 -2023-02-13 18:13:09,072 - Epoch: [124][ 820/ 1207] Overall Loss 0.243540 Objective Loss 0.243540 LR 0.000500 Time 0.020802 -2023-02-13 18:13:09,276 - Epoch: [124][ 830/ 1207] Overall Loss 0.243365 Objective Loss 0.243365 LR 0.000500 Time 0.020797 -2023-02-13 18:13:09,482 - Epoch: [124][ 840/ 1207] Overall Loss 0.243177 Objective Loss 0.243177 LR 0.000500 Time 0.020794 -2023-02-13 18:13:09,682 - Epoch: [124][ 850/ 1207] Overall Loss 0.242995 Objective Loss 0.242995 LR 0.000500 Time 0.020784 -2023-02-13 18:13:09,874 - Epoch: [124][ 860/ 1207] Overall Loss 0.242890 Objective Loss 0.242890 LR 0.000500 Time 0.020766 -2023-02-13 18:13:10,070 - Epoch: [124][ 870/ 1207] Overall Loss 0.242528 Objective Loss 0.242528 LR 0.000500 Time 0.020751 -2023-02-13 18:13:10,262 - Epoch: [124][ 880/ 1207] Overall Loss 0.242844 Objective Loss 0.242844 LR 0.000500 Time 0.020733 -2023-02-13 18:13:10,458 - Epoch: [124][ 890/ 1207] Overall Loss 0.242737 Objective Loss 0.242737 LR 0.000500 Time 0.020720 -2023-02-13 18:13:10,650 - Epoch: [124][ 900/ 1207] Overall Loss 0.242841 Objective Loss 0.242841 LR 0.000500 Time 0.020704 -2023-02-13 18:13:10,846 - Epoch: [124][ 910/ 1207] Overall Loss 0.242701 Objective Loss 0.242701 LR 0.000500 Time 0.020690 -2023-02-13 18:13:11,039 - Epoch: [124][ 920/ 1207] Overall Loss 0.242746 Objective Loss 0.242746 LR 0.000500 Time 0.020675 -2023-02-13 18:13:11,234 - Epoch: [124][ 930/ 1207] Overall Loss 0.242491 Objective Loss 0.242491 LR 0.000500 Time 0.020662 -2023-02-13 18:13:11,425 - Epoch: [124][ 940/ 1207] Overall Loss 0.242525 Objective Loss 0.242525 LR 0.000500 Time 0.020645 -2023-02-13 18:13:11,621 - Epoch: [124][ 950/ 1207] Overall Loss 0.242490 Objective Loss 0.242490 LR 0.000500 Time 0.020634 -2023-02-13 18:13:11,813 - Epoch: [124][ 960/ 1207] Overall Loss 0.242459 Objective Loss 0.242459 LR 0.000500 Time 0.020618 -2023-02-13 18:13:12,009 - Epoch: [124][ 970/ 1207] Overall Loss 0.242645 Objective Loss 0.242645 LR 0.000500 Time 0.020608 -2023-02-13 18:13:12,201 - Epoch: [124][ 980/ 1207] Overall Loss 0.242543 Objective Loss 0.242543 LR 0.000500 Time 0.020592 -2023-02-13 18:13:12,396 - Epoch: [124][ 990/ 1207] Overall Loss 0.242596 Objective Loss 0.242596 LR 0.000500 Time 0.020581 -2023-02-13 18:13:12,586 - Epoch: [124][ 1000/ 1207] Overall Loss 0.242464 Objective Loss 0.242464 LR 0.000500 Time 0.020566 -2023-02-13 18:13:12,776 - Epoch: [124][ 1010/ 1207] Overall Loss 0.242560 Objective Loss 0.242560 LR 0.000500 Time 0.020549 -2023-02-13 18:13:12,965 - Epoch: [124][ 1020/ 1207] Overall Loss 0.242972 Objective Loss 0.242972 LR 0.000500 Time 0.020533 -2023-02-13 18:13:13,154 - Epoch: [124][ 1030/ 1207] Overall Loss 0.242969 Objective Loss 0.242969 LR 0.000500 Time 0.020517 -2023-02-13 18:13:13,343 - Epoch: [124][ 1040/ 1207] Overall Loss 0.242992 Objective Loss 0.242992 LR 0.000500 Time 0.020501 -2023-02-13 18:13:13,533 - Epoch: [124][ 1050/ 1207] Overall Loss 0.242853 Objective Loss 0.242853 LR 0.000500 Time 0.020486 -2023-02-13 18:13:13,723 - Epoch: [124][ 1060/ 1207] Overall Loss 0.242972 Objective Loss 0.242972 LR 0.000500 Time 0.020471 -2023-02-13 18:13:13,913 - Epoch: [124][ 1070/ 1207] Overall Loss 0.243124 Objective Loss 0.243124 LR 0.000500 Time 0.020458 -2023-02-13 18:13:14,104 - Epoch: [124][ 1080/ 1207] Overall Loss 0.242885 Objective Loss 0.242885 LR 0.000500 Time 0.020444 -2023-02-13 18:13:14,294 - Epoch: [124][ 1090/ 1207] Overall Loss 0.243070 Objective Loss 0.243070 LR 0.000500 Time 0.020431 -2023-02-13 18:13:14,482 - Epoch: [124][ 1100/ 1207] Overall Loss 0.243406 Objective Loss 0.243406 LR 0.000500 Time 0.020416 -2023-02-13 18:13:14,672 - Epoch: [124][ 1110/ 1207] Overall Loss 0.243452 Objective Loss 0.243452 LR 0.000500 Time 0.020403 -2023-02-13 18:13:14,861 - Epoch: [124][ 1120/ 1207] Overall Loss 0.243425 Objective Loss 0.243425 LR 0.000500 Time 0.020389 -2023-02-13 18:13:15,051 - Epoch: [124][ 1130/ 1207] Overall Loss 0.243504 Objective Loss 0.243504 LR 0.000500 Time 0.020376 -2023-02-13 18:13:15,239 - Epoch: [124][ 1140/ 1207] Overall Loss 0.243754 Objective Loss 0.243754 LR 0.000500 Time 0.020362 -2023-02-13 18:13:15,428 - Epoch: [124][ 1150/ 1207] Overall Loss 0.244216 Objective Loss 0.244216 LR 0.000500 Time 0.020349 -2023-02-13 18:13:15,617 - Epoch: [124][ 1160/ 1207] Overall Loss 0.244355 Objective Loss 0.244355 LR 0.000500 Time 0.020337 -2023-02-13 18:13:15,807 - Epoch: [124][ 1170/ 1207] Overall Loss 0.244505 Objective Loss 0.244505 LR 0.000500 Time 0.020324 -2023-02-13 18:13:15,997 - Epoch: [124][ 1180/ 1207] Overall Loss 0.244486 Objective Loss 0.244486 LR 0.000500 Time 0.020313 -2023-02-13 18:13:16,186 - Epoch: [124][ 1190/ 1207] Overall Loss 0.244648 Objective Loss 0.244648 LR 0.000500 Time 0.020300 -2023-02-13 18:13:16,431 - Epoch: [124][ 1200/ 1207] Overall Loss 0.244702 Objective Loss 0.244702 LR 0.000500 Time 0.020336 -2023-02-13 18:13:16,547 - Epoch: [124][ 1207/ 1207] Overall Loss 0.244398 Objective Loss 0.244398 Top1 86.890244 Top5 97.865854 LR 0.000500 Time 0.020314 -2023-02-13 18:13:16,618 - --- validate (epoch=124)----------- -2023-02-13 18:13:16,618 - 34311 samples (256 per mini-batch) -2023-02-13 18:13:17,026 - Epoch: [124][ 10/ 135] Loss 0.331241 Top1 85.078125 Top5 97.617188 -2023-02-13 18:13:17,158 - Epoch: [124][ 20/ 135] Loss 0.296614 Top1 84.921875 Top5 97.851562 -2023-02-13 18:13:17,290 - Epoch: [124][ 30/ 135] Loss 0.305019 Top1 84.674479 Top5 97.864583 -2023-02-13 18:13:17,422 - Epoch: [124][ 40/ 135] Loss 0.316738 Top1 84.189453 Top5 97.753906 -2023-02-13 18:13:17,554 - Epoch: [124][ 50/ 135] Loss 0.316033 Top1 84.257812 Top5 97.671875 -2023-02-13 18:13:17,683 - Epoch: [124][ 60/ 135] Loss 0.315219 Top1 84.303385 Top5 97.682292 -2023-02-13 18:13:17,814 - Epoch: [124][ 70/ 135] Loss 0.318901 Top1 84.146205 Top5 97.712054 -2023-02-13 18:13:17,938 - Epoch: [124][ 80/ 135] Loss 0.317381 Top1 84.145508 Top5 97.690430 -2023-02-13 18:13:18,069 - Epoch: [124][ 90/ 135] Loss 0.314774 Top1 84.188368 Top5 97.690972 -2023-02-13 18:13:18,212 - Epoch: [124][ 100/ 135] Loss 0.317512 Top1 84.101562 Top5 97.585938 -2023-02-13 18:13:18,352 - Epoch: [124][ 110/ 135] Loss 0.317868 Top1 84.030540 Top5 97.581676 -2023-02-13 18:13:18,499 - Epoch: [124][ 120/ 135] Loss 0.317176 Top1 84.042969 Top5 97.587891 -2023-02-13 18:13:18,627 - Epoch: [124][ 130/ 135] Loss 0.318640 Top1 83.987380 Top5 97.578125 -2023-02-13 18:13:18,672 - Epoch: [124][ 135/ 135] Loss 0.317877 Top1 83.981813 Top5 97.566378 -2023-02-13 18:13:18,741 - ==> Top1: 83.982 Top5: 97.566 Loss: 0.318 - -2023-02-13 18:13:18,742 - ==> Confusion: -[[ 836 6 7 2 9 3 0 1 4 65 0 2 2 4 7 7 1 2 1 2 6] - [ 2 934 2 4 18 32 4 14 3 2 1 0 1 0 0 2 4 1 3 0 6] - [ 5 3 957 8 6 1 19 12 0 1 2 0 5 5 4 10 1 3 5 3 8] - [ 5 3 20 913 3 3 0 2 2 2 10 0 10 1 17 3 4 4 10 0 4] - [ 15 8 0 2 988 14 1 3 1 2 0 6 1 1 7 6 4 4 0 3 0] - [ 1 17 1 6 6 978 2 12 1 5 2 13 2 11 0 3 3 1 1 3 2] - [ 2 6 19 1 1 3 1034 5 1 0 7 2 2 1 0 2 1 1 1 5 5] - [ 1 8 9 1 2 31 7 930 0 2 2 4 1 1 0 0 2 2 13 7 1] - [ 22 4 1 1 2 0 1 2 859 57 13 2 0 17 18 1 1 2 3 0 3] - [ 57 2 5 0 9 2 0 4 28 874 0 0 1 14 6 0 3 1 2 0 4] - [ 2 2 8 9 0 2 4 4 10 1 980 2 3 9 3 0 0 1 5 1 5] - [ 2 2 2 1 5 12 1 4 0 0 1 912 21 7 0 7 3 7 1 15 2] - [ 2 0 1 8 3 4 0 1 3 0 1 35 852 0 4 12 4 17 3 1 8] - [ 4 3 2 1 10 10 0 1 6 15 12 8 1 931 4 6 1 3 1 1 4] - [ 6 4 1 30 8 3 1 2 13 8 4 2 4 2 979 2 0 7 7 0 9] - [ 3 4 4 0 4 0 4 1 0 1 0 9 10 1 1 976 9 8 0 5 6] - [ 6 7 1 1 9 2 0 1 0 0 0 0 4 4 1 8 1000 1 2 6 8] - [ 7 4 0 4 0 0 7 2 1 0 2 6 10 1 1 17 1 981 0 2 5] - [ 2 2 10 14 0 2 1 25 4 0 6 3 2 0 11 1 1 2 995 2 3] - [ 1 2 1 0 2 6 6 15 1 0 1 10 3 2 0 7 4 3 1 1078 5] - [ 162 214 248 157 158 251 80 183 74 92 188 113 286 312 146 132 230 107 168 305 9828]] - -2023-02-13 18:13:18,743 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:13:18,743 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:13:18,749 - - -2023-02-13 18:13:18,749 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:13:19,626 - Epoch: [125][ 10/ 1207] Overall Loss 0.240962 Objective Loss 0.240962 LR 0.000500 Time 0.087614 -2023-02-13 18:13:19,830 - Epoch: [125][ 20/ 1207] Overall Loss 0.244754 Objective Loss 0.244754 LR 0.000500 Time 0.053981 -2023-02-13 18:13:20,027 - Epoch: [125][ 30/ 1207] Overall Loss 0.233095 Objective Loss 0.233095 LR 0.000500 Time 0.042532 -2023-02-13 18:13:20,228 - Epoch: [125][ 40/ 1207] Overall Loss 0.227876 Objective Loss 0.227876 LR 0.000500 Time 0.036932 -2023-02-13 18:13:20,426 - Epoch: [125][ 50/ 1207] Overall Loss 0.232882 Objective Loss 0.232882 LR 0.000500 Time 0.033484 -2023-02-13 18:13:20,628 - Epoch: [125][ 60/ 1207] Overall Loss 0.236322 Objective Loss 0.236322 LR 0.000500 Time 0.031271 -2023-02-13 18:13:20,826 - Epoch: [125][ 70/ 1207] Overall Loss 0.239678 Objective Loss 0.239678 LR 0.000500 Time 0.029624 -2023-02-13 18:13:21,028 - Epoch: [125][ 80/ 1207] Overall Loss 0.239454 Objective Loss 0.239454 LR 0.000500 Time 0.028446 -2023-02-13 18:13:21,226 - Epoch: [125][ 90/ 1207] Overall Loss 0.237455 Objective Loss 0.237455 LR 0.000500 Time 0.027478 -2023-02-13 18:13:21,429 - Epoch: [125][ 100/ 1207] Overall Loss 0.237151 Objective Loss 0.237151 LR 0.000500 Time 0.026751 -2023-02-13 18:13:21,627 - Epoch: [125][ 110/ 1207] Overall Loss 0.235621 Objective Loss 0.235621 LR 0.000500 Time 0.026115 -2023-02-13 18:13:21,829 - Epoch: [125][ 120/ 1207] Overall Loss 0.236841 Objective Loss 0.236841 LR 0.000500 Time 0.025623 -2023-02-13 18:13:22,027 - Epoch: [125][ 130/ 1207] Overall Loss 0.238232 Objective Loss 0.238232 LR 0.000500 Time 0.025168 -2023-02-13 18:13:22,229 - Epoch: [125][ 140/ 1207] Overall Loss 0.237764 Objective Loss 0.237764 LR 0.000500 Time 0.024812 -2023-02-13 18:13:22,426 - Epoch: [125][ 150/ 1207] Overall Loss 0.237208 Objective Loss 0.237208 LR 0.000500 Time 0.024469 -2023-02-13 18:13:22,628 - Epoch: [125][ 160/ 1207] Overall Loss 0.238698 Objective Loss 0.238698 LR 0.000500 Time 0.024199 -2023-02-13 18:13:22,825 - Epoch: [125][ 170/ 1207] Overall Loss 0.238674 Objective Loss 0.238674 LR 0.000500 Time 0.023936 -2023-02-13 18:13:23,026 - Epoch: [125][ 180/ 1207] Overall Loss 0.237502 Objective Loss 0.237502 LR 0.000500 Time 0.023721 -2023-02-13 18:13:23,224 - Epoch: [125][ 190/ 1207] Overall Loss 0.236554 Objective Loss 0.236554 LR 0.000500 Time 0.023511 -2023-02-13 18:13:23,426 - Epoch: [125][ 200/ 1207] Overall Loss 0.238335 Objective Loss 0.238335 LR 0.000500 Time 0.023341 -2023-02-13 18:13:23,623 - Epoch: [125][ 210/ 1207] Overall Loss 0.237830 Objective Loss 0.237830 LR 0.000500 Time 0.023169 -2023-02-13 18:13:23,825 - Epoch: [125][ 220/ 1207] Overall Loss 0.237069 Objective Loss 0.237069 LR 0.000500 Time 0.023033 -2023-02-13 18:13:24,023 - Epoch: [125][ 230/ 1207] Overall Loss 0.237407 Objective Loss 0.237407 LR 0.000500 Time 0.022890 -2023-02-13 18:13:24,225 - Epoch: [125][ 240/ 1207] Overall Loss 0.238162 Objective Loss 0.238162 LR 0.000500 Time 0.022777 -2023-02-13 18:13:24,423 - Epoch: [125][ 250/ 1207] Overall Loss 0.237882 Objective Loss 0.237882 LR 0.000500 Time 0.022654 -2023-02-13 18:13:24,626 - Epoch: [125][ 260/ 1207] Overall Loss 0.238162 Objective Loss 0.238162 LR 0.000500 Time 0.022561 -2023-02-13 18:13:24,824 - Epoch: [125][ 270/ 1207] Overall Loss 0.239250 Objective Loss 0.239250 LR 0.000500 Time 0.022458 -2023-02-13 18:13:25,026 - Epoch: [125][ 280/ 1207] Overall Loss 0.239677 Objective Loss 0.239677 LR 0.000500 Time 0.022377 -2023-02-13 18:13:25,223 - Epoch: [125][ 290/ 1207] Overall Loss 0.239867 Objective Loss 0.239867 LR 0.000500 Time 0.022283 -2023-02-13 18:13:25,425 - Epoch: [125][ 300/ 1207] Overall Loss 0.240106 Objective Loss 0.240106 LR 0.000500 Time 0.022213 -2023-02-13 18:13:25,619 - Epoch: [125][ 310/ 1207] Overall Loss 0.239243 Objective Loss 0.239243 LR 0.000500 Time 0.022121 -2023-02-13 18:13:25,808 - Epoch: [125][ 320/ 1207] Overall Loss 0.238571 Objective Loss 0.238571 LR 0.000500 Time 0.022017 -2023-02-13 18:13:25,997 - Epoch: [125][ 330/ 1207] Overall Loss 0.238403 Objective Loss 0.238403 LR 0.000500 Time 0.021923 -2023-02-13 18:13:26,185 - Epoch: [125][ 340/ 1207] Overall Loss 0.238087 Objective Loss 0.238087 LR 0.000500 Time 0.021830 -2023-02-13 18:13:26,374 - Epoch: [125][ 350/ 1207] Overall Loss 0.238088 Objective Loss 0.238088 LR 0.000500 Time 0.021744 -2023-02-13 18:13:26,563 - Epoch: [125][ 360/ 1207] Overall Loss 0.238702 Objective Loss 0.238702 LR 0.000500 Time 0.021664 -2023-02-13 18:13:26,751 - Epoch: [125][ 370/ 1207] Overall Loss 0.238049 Objective Loss 0.238049 LR 0.000500 Time 0.021588 -2023-02-13 18:13:26,941 - Epoch: [125][ 380/ 1207] Overall Loss 0.238007 Objective Loss 0.238007 LR 0.000500 Time 0.021516 -2023-02-13 18:13:27,128 - Epoch: [125][ 390/ 1207] Overall Loss 0.237293 Objective Loss 0.237293 LR 0.000500 Time 0.021445 -2023-02-13 18:13:27,317 - Epoch: [125][ 400/ 1207] Overall Loss 0.237939 Objective Loss 0.237939 LR 0.000500 Time 0.021379 -2023-02-13 18:13:27,505 - Epoch: [125][ 410/ 1207] Overall Loss 0.238004 Objective Loss 0.238004 LR 0.000500 Time 0.021316 -2023-02-13 18:13:27,694 - Epoch: [125][ 420/ 1207] Overall Loss 0.237657 Objective Loss 0.237657 LR 0.000500 Time 0.021257 -2023-02-13 18:13:27,883 - Epoch: [125][ 430/ 1207] Overall Loss 0.237425 Objective Loss 0.237425 LR 0.000500 Time 0.021201 -2023-02-13 18:13:28,074 - Epoch: [125][ 440/ 1207] Overall Loss 0.238271 Objective Loss 0.238271 LR 0.000500 Time 0.021152 -2023-02-13 18:13:28,264 - Epoch: [125][ 450/ 1207] Overall Loss 0.238977 Objective Loss 0.238977 LR 0.000500 Time 0.021105 -2023-02-13 18:13:28,455 - Epoch: [125][ 460/ 1207] Overall Loss 0.238967 Objective Loss 0.238967 LR 0.000500 Time 0.021061 -2023-02-13 18:13:28,647 - Epoch: [125][ 470/ 1207] Overall Loss 0.238884 Objective Loss 0.238884 LR 0.000500 Time 0.021019 -2023-02-13 18:13:28,838 - Epoch: [125][ 480/ 1207] Overall Loss 0.239257 Objective Loss 0.239257 LR 0.000500 Time 0.020979 -2023-02-13 18:13:29,029 - Epoch: [125][ 490/ 1207] Overall Loss 0.239137 Objective Loss 0.239137 LR 0.000500 Time 0.020940 -2023-02-13 18:13:29,221 - Epoch: [125][ 500/ 1207] Overall Loss 0.238622 Objective Loss 0.238622 LR 0.000500 Time 0.020904 -2023-02-13 18:13:29,412 - Epoch: [125][ 510/ 1207] Overall Loss 0.238891 Objective Loss 0.238891 LR 0.000500 Time 0.020868 -2023-02-13 18:13:29,603 - Epoch: [125][ 520/ 1207] Overall Loss 0.238454 Objective Loss 0.238454 LR 0.000500 Time 0.020834 -2023-02-13 18:13:29,794 - Epoch: [125][ 530/ 1207] Overall Loss 0.238106 Objective Loss 0.238106 LR 0.000500 Time 0.020800 -2023-02-13 18:13:29,985 - Epoch: [125][ 540/ 1207] Overall Loss 0.238246 Objective Loss 0.238246 LR 0.000500 Time 0.020768 -2023-02-13 18:13:30,176 - Epoch: [125][ 550/ 1207] Overall Loss 0.237975 Objective Loss 0.237975 LR 0.000500 Time 0.020736 -2023-02-13 18:13:30,368 - Epoch: [125][ 560/ 1207] Overall Loss 0.238016 Objective Loss 0.238016 LR 0.000500 Time 0.020709 -2023-02-13 18:13:30,559 - Epoch: [125][ 570/ 1207] Overall Loss 0.238142 Objective Loss 0.238142 LR 0.000500 Time 0.020680 -2023-02-13 18:13:30,751 - Epoch: [125][ 580/ 1207] Overall Loss 0.237973 Objective Loss 0.237973 LR 0.000500 Time 0.020654 -2023-02-13 18:13:30,943 - Epoch: [125][ 590/ 1207] Overall Loss 0.238090 Objective Loss 0.238090 LR 0.000500 Time 0.020629 -2023-02-13 18:13:31,134 - Epoch: [125][ 600/ 1207] Overall Loss 0.238086 Objective Loss 0.238086 LR 0.000500 Time 0.020603 -2023-02-13 18:13:31,325 - Epoch: [125][ 610/ 1207] Overall Loss 0.238322 Objective Loss 0.238322 LR 0.000500 Time 0.020577 -2023-02-13 18:13:31,516 - Epoch: [125][ 620/ 1207] Overall Loss 0.238583 Objective Loss 0.238583 LR 0.000500 Time 0.020552 -2023-02-13 18:13:31,706 - Epoch: [125][ 630/ 1207] Overall Loss 0.238708 Objective Loss 0.238708 LR 0.000500 Time 0.020528 -2023-02-13 18:13:31,898 - Epoch: [125][ 640/ 1207] Overall Loss 0.239272 Objective Loss 0.239272 LR 0.000500 Time 0.020506 -2023-02-13 18:13:32,089 - Epoch: [125][ 650/ 1207] Overall Loss 0.239302 Objective Loss 0.239302 LR 0.000500 Time 0.020484 -2023-02-13 18:13:32,280 - Epoch: [125][ 660/ 1207] Overall Loss 0.239295 Objective Loss 0.239295 LR 0.000500 Time 0.020462 -2023-02-13 18:13:32,471 - Epoch: [125][ 670/ 1207] Overall Loss 0.239047 Objective Loss 0.239047 LR 0.000500 Time 0.020441 -2023-02-13 18:13:32,663 - Epoch: [125][ 680/ 1207] Overall Loss 0.239180 Objective Loss 0.239180 LR 0.000500 Time 0.020423 -2023-02-13 18:13:32,854 - Epoch: [125][ 690/ 1207] Overall Loss 0.238988 Objective Loss 0.238988 LR 0.000500 Time 0.020403 -2023-02-13 18:13:33,045 - Epoch: [125][ 700/ 1207] Overall Loss 0.239240 Objective Loss 0.239240 LR 0.000500 Time 0.020385 -2023-02-13 18:13:33,236 - Epoch: [125][ 710/ 1207] Overall Loss 0.239103 Objective Loss 0.239103 LR 0.000500 Time 0.020366 -2023-02-13 18:13:33,428 - Epoch: [125][ 720/ 1207] Overall Loss 0.239574 Objective Loss 0.239574 LR 0.000500 Time 0.020349 -2023-02-13 18:13:33,619 - Epoch: [125][ 730/ 1207] Overall Loss 0.239617 Objective Loss 0.239617 LR 0.000500 Time 0.020331 -2023-02-13 18:13:33,808 - Epoch: [125][ 740/ 1207] Overall Loss 0.239670 Objective Loss 0.239670 LR 0.000500 Time 0.020311 -2023-02-13 18:13:33,996 - Epoch: [125][ 750/ 1207] Overall Loss 0.239867 Objective Loss 0.239867 LR 0.000500 Time 0.020291 -2023-02-13 18:13:34,186 - Epoch: [125][ 760/ 1207] Overall Loss 0.239783 Objective Loss 0.239783 LR 0.000500 Time 0.020273 -2023-02-13 18:13:34,376 - Epoch: [125][ 770/ 1207] Overall Loss 0.239753 Objective Loss 0.239753 LR 0.000500 Time 0.020256 -2023-02-13 18:13:34,567 - Epoch: [125][ 780/ 1207] Overall Loss 0.239774 Objective Loss 0.239774 LR 0.000500 Time 0.020241 -2023-02-13 18:13:34,759 - Epoch: [125][ 790/ 1207] Overall Loss 0.239572 Objective Loss 0.239572 LR 0.000500 Time 0.020227 -2023-02-13 18:13:34,950 - Epoch: [125][ 800/ 1207] Overall Loss 0.239654 Objective Loss 0.239654 LR 0.000500 Time 0.020213 -2023-02-13 18:13:35,140 - Epoch: [125][ 810/ 1207] Overall Loss 0.239782 Objective Loss 0.239782 LR 0.000500 Time 0.020197 -2023-02-13 18:13:35,331 - Epoch: [125][ 820/ 1207] Overall Loss 0.240166 Objective Loss 0.240166 LR 0.000500 Time 0.020183 -2023-02-13 18:13:35,522 - Epoch: [125][ 830/ 1207] Overall Loss 0.240377 Objective Loss 0.240377 LR 0.000500 Time 0.020170 -2023-02-13 18:13:35,714 - Epoch: [125][ 840/ 1207] Overall Loss 0.240263 Objective Loss 0.240263 LR 0.000500 Time 0.020157 -2023-02-13 18:13:35,905 - Epoch: [125][ 850/ 1207] Overall Loss 0.240391 Objective Loss 0.240391 LR 0.000500 Time 0.020145 -2023-02-13 18:13:36,096 - Epoch: [125][ 860/ 1207] Overall Loss 0.240225 Objective Loss 0.240225 LR 0.000500 Time 0.020133 -2023-02-13 18:13:36,287 - Epoch: [125][ 870/ 1207] Overall Loss 0.240279 Objective Loss 0.240279 LR 0.000500 Time 0.020120 -2023-02-13 18:13:36,479 - Epoch: [125][ 880/ 1207] Overall Loss 0.240470 Objective Loss 0.240470 LR 0.000500 Time 0.020109 -2023-02-13 18:13:36,670 - Epoch: [125][ 890/ 1207] Overall Loss 0.240576 Objective Loss 0.240576 LR 0.000500 Time 0.020098 -2023-02-13 18:13:36,862 - Epoch: [125][ 900/ 1207] Overall Loss 0.240832 Objective Loss 0.240832 LR 0.000500 Time 0.020087 -2023-02-13 18:13:37,053 - Epoch: [125][ 910/ 1207] Overall Loss 0.240976 Objective Loss 0.240976 LR 0.000500 Time 0.020076 -2023-02-13 18:13:37,245 - Epoch: [125][ 920/ 1207] Overall Loss 0.241077 Objective Loss 0.241077 LR 0.000500 Time 0.020065 -2023-02-13 18:13:37,435 - Epoch: [125][ 930/ 1207] Overall Loss 0.241112 Objective Loss 0.241112 LR 0.000500 Time 0.020054 -2023-02-13 18:13:37,626 - Epoch: [125][ 940/ 1207] Overall Loss 0.240995 Objective Loss 0.240995 LR 0.000500 Time 0.020044 -2023-02-13 18:13:37,817 - Epoch: [125][ 950/ 1207] Overall Loss 0.241430 Objective Loss 0.241430 LR 0.000500 Time 0.020033 -2023-02-13 18:13:38,008 - Epoch: [125][ 960/ 1207] Overall Loss 0.241853 Objective Loss 0.241853 LR 0.000500 Time 0.020023 -2023-02-13 18:13:38,198 - Epoch: [125][ 970/ 1207] Overall Loss 0.241815 Objective Loss 0.241815 LR 0.000500 Time 0.020012 -2023-02-13 18:13:38,389 - Epoch: [125][ 980/ 1207] Overall Loss 0.241860 Objective Loss 0.241860 LR 0.000500 Time 0.020002 -2023-02-13 18:13:38,579 - Epoch: [125][ 990/ 1207] Overall Loss 0.241851 Objective Loss 0.241851 LR 0.000500 Time 0.019992 -2023-02-13 18:13:38,771 - Epoch: [125][ 1000/ 1207] Overall Loss 0.241827 Objective Loss 0.241827 LR 0.000500 Time 0.019983 -2023-02-13 18:13:38,961 - Epoch: [125][ 1010/ 1207] Overall Loss 0.242144 Objective Loss 0.242144 LR 0.000500 Time 0.019974 -2023-02-13 18:13:39,152 - Epoch: [125][ 1020/ 1207] Overall Loss 0.242392 Objective Loss 0.242392 LR 0.000500 Time 0.019965 -2023-02-13 18:13:39,343 - Epoch: [125][ 1030/ 1207] Overall Loss 0.242342 Objective Loss 0.242342 LR 0.000500 Time 0.019956 -2023-02-13 18:13:39,534 - Epoch: [125][ 1040/ 1207] Overall Loss 0.242377 Objective Loss 0.242377 LR 0.000500 Time 0.019947 -2023-02-13 18:13:39,725 - Epoch: [125][ 1050/ 1207] Overall Loss 0.242435 Objective Loss 0.242435 LR 0.000500 Time 0.019939 -2023-02-13 18:13:39,917 - Epoch: [125][ 1060/ 1207] Overall Loss 0.242422 Objective Loss 0.242422 LR 0.000500 Time 0.019931 -2023-02-13 18:13:40,107 - Epoch: [125][ 1070/ 1207] Overall Loss 0.242431 Objective Loss 0.242431 LR 0.000500 Time 0.019923 -2023-02-13 18:13:40,298 - Epoch: [125][ 1080/ 1207] Overall Loss 0.242491 Objective Loss 0.242491 LR 0.000500 Time 0.019914 -2023-02-13 18:13:40,488 - Epoch: [125][ 1090/ 1207] Overall Loss 0.242481 Objective Loss 0.242481 LR 0.000500 Time 0.019906 -2023-02-13 18:13:40,679 - Epoch: [125][ 1100/ 1207] Overall Loss 0.242484 Objective Loss 0.242484 LR 0.000500 Time 0.019898 -2023-02-13 18:13:40,872 - Epoch: [125][ 1110/ 1207] Overall Loss 0.242352 Objective Loss 0.242352 LR 0.000500 Time 0.019893 -2023-02-13 18:13:41,064 - Epoch: [125][ 1120/ 1207] Overall Loss 0.242549 Objective Loss 0.242549 LR 0.000500 Time 0.019886 -2023-02-13 18:13:41,271 - Epoch: [125][ 1130/ 1207] Overall Loss 0.242651 Objective Loss 0.242651 LR 0.000500 Time 0.019893 -2023-02-13 18:13:41,473 - Epoch: [125][ 1140/ 1207] Overall Loss 0.242499 Objective Loss 0.242499 LR 0.000500 Time 0.019895 -2023-02-13 18:13:41,680 - Epoch: [125][ 1150/ 1207] Overall Loss 0.242449 Objective Loss 0.242449 LR 0.000500 Time 0.019902 -2023-02-13 18:13:41,884 - Epoch: [125][ 1160/ 1207] Overall Loss 0.242397 Objective Loss 0.242397 LR 0.000500 Time 0.019905 -2023-02-13 18:13:42,091 - Epoch: [125][ 1170/ 1207] Overall Loss 0.242234 Objective Loss 0.242234 LR 0.000500 Time 0.019912 -2023-02-13 18:13:42,294 - Epoch: [125][ 1180/ 1207] Overall Loss 0.242083 Objective Loss 0.242083 LR 0.000500 Time 0.019916 -2023-02-13 18:13:42,500 - Epoch: [125][ 1190/ 1207] Overall Loss 0.242036 Objective Loss 0.242036 LR 0.000500 Time 0.019921 -2023-02-13 18:13:42,753 - Epoch: [125][ 1200/ 1207] Overall Loss 0.242153 Objective Loss 0.242153 LR 0.000500 Time 0.019965 -2023-02-13 18:13:42,870 - Epoch: [125][ 1207/ 1207] Overall Loss 0.242237 Objective Loss 0.242237 Top1 84.146341 Top5 98.780488 LR 0.000500 Time 0.019946 -2023-02-13 18:13:42,942 - --- validate (epoch=125)----------- -2023-02-13 18:13:42,942 - 34311 samples (256 per mini-batch) -2023-02-13 18:13:43,453 - Epoch: [125][ 10/ 135] Loss 0.327937 Top1 84.570312 Top5 97.773438 -2023-02-13 18:13:43,580 - Epoch: [125][ 20/ 135] Loss 0.316607 Top1 84.746094 Top5 97.597656 -2023-02-13 18:13:43,709 - Epoch: [125][ 30/ 135] Loss 0.310741 Top1 84.544271 Top5 97.643229 -2023-02-13 18:13:43,840 - Epoch: [125][ 40/ 135] Loss 0.312440 Top1 84.589844 Top5 97.666016 -2023-02-13 18:13:43,967 - Epoch: [125][ 50/ 135] Loss 0.312386 Top1 84.492188 Top5 97.726562 -2023-02-13 18:13:44,095 - Epoch: [125][ 60/ 135] Loss 0.316346 Top1 84.257812 Top5 97.695312 -2023-02-13 18:13:44,225 - Epoch: [125][ 70/ 135] Loss 0.315007 Top1 84.213170 Top5 97.689732 -2023-02-13 18:13:44,352 - Epoch: [125][ 80/ 135] Loss 0.312987 Top1 84.291992 Top5 97.680664 -2023-02-13 18:13:44,477 - Epoch: [125][ 90/ 135] Loss 0.315095 Top1 84.184028 Top5 97.695312 -2023-02-13 18:13:44,603 - Epoch: [125][ 100/ 135] Loss 0.314044 Top1 84.156250 Top5 97.675781 -2023-02-13 18:13:44,731 - Epoch: [125][ 110/ 135] Loss 0.313684 Top1 84.126420 Top5 97.645597 -2023-02-13 18:13:44,860 - Epoch: [125][ 120/ 135] Loss 0.316238 Top1 84.130859 Top5 97.662760 -2023-02-13 18:13:44,992 - Epoch: [125][ 130/ 135] Loss 0.315888 Top1 84.107572 Top5 97.650240 -2023-02-13 18:13:45,039 - Epoch: [125][ 135/ 135] Loss 0.332437 Top1 84.118796 Top5 97.639241 -2023-02-13 18:13:45,109 - ==> Top1: 84.119 Top5: 97.639 Loss: 0.332 - -2023-02-13 18:13:45,110 - ==> Confusion: -[[ 861 5 3 0 7 3 0 1 5 43 0 3 0 4 10 3 4 4 1 3 7] - [ 2 940 0 1 9 26 1 21 3 1 2 1 1 0 2 1 6 1 11 1 3] - [ 7 4 957 8 4 2 10 14 0 1 4 1 3 4 3 6 3 6 7 6 8] - [ 3 3 19 909 2 3 2 3 0 3 8 0 5 0 27 1 3 5 17 0 3] - [ 12 7 0 1 988 8 1 1 1 2 0 6 2 3 11 7 6 2 1 3 4] - [ 3 18 0 6 5 958 3 18 2 4 0 14 2 14 2 2 3 2 3 6 5] - [ 3 5 17 1 0 3 1028 8 2 1 5 1 0 1 1 3 1 6 2 8 3] - [ 1 4 7 3 0 25 1 932 3 1 1 5 2 0 1 0 1 1 23 7 6] - [ 17 1 1 2 0 0 1 2 908 35 6 2 1 7 18 2 0 1 3 1 1] - [ 93 1 3 1 6 2 0 2 37 832 0 1 0 17 6 1 3 3 2 0 2] - [ 2 1 4 9 1 1 2 8 16 1 968 3 1 9 2 0 1 2 16 0 4] - [ 0 2 0 0 4 11 2 6 0 2 0 924 22 4 0 5 5 9 1 7 1] - [ 2 0 0 6 3 3 0 1 2 0 1 23 881 1 1 9 4 9 4 0 9] - [ 4 3 0 0 10 9 0 2 13 17 12 6 4 922 2 5 4 2 0 1 8] - [ 3 3 2 11 3 0 1 2 16 4 2 1 3 1 1019 0 0 6 5 3 7] - [ 6 2 5 3 9 2 1 1 0 0 0 12 6 1 1 952 10 16 2 7 10] - [ 2 7 0 2 7 1 0 2 2 1 1 0 1 1 2 7 1009 1 2 5 8] - [ 3 2 0 3 1 1 1 1 0 0 1 14 14 0 0 11 0 992 0 1 6] - [ 2 4 3 6 0 1 0 22 5 0 0 4 4 0 14 2 1 2 1014 1 1] - [ 2 5 1 1 1 9 7 19 1 0 1 22 4 2 0 6 8 5 1 1045 8] - [ 164 207 216 128 157 215 69 204 133 83 195 129 310 274 235 82 276 102 219 214 9822]] - -2023-02-13 18:13:45,112 - ==> Best [Top1: 84.355 Top5: 97.811 Sparsity:0.00 Params: 148928 on epoch: 114] -2023-02-13 18:13:45,112 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:13:45,117 - - -2023-02-13 18:13:45,117 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:13:46,005 - Epoch: [126][ 10/ 1207] Overall Loss 0.229597 Objective Loss 0.229597 LR 0.000500 Time 0.088645 -2023-02-13 18:13:46,202 - Epoch: [126][ 20/ 1207] Overall Loss 0.227964 Objective Loss 0.227964 LR 0.000500 Time 0.054193 -2023-02-13 18:13:46,390 - Epoch: [126][ 30/ 1207] Overall Loss 0.238751 Objective Loss 0.238751 LR 0.000500 Time 0.042386 -2023-02-13 18:13:46,578 - Epoch: [126][ 40/ 1207] Overall Loss 0.230554 Objective Loss 0.230554 LR 0.000500 Time 0.036465 -2023-02-13 18:13:46,766 - Epoch: [126][ 50/ 1207] Overall Loss 0.229140 Objective Loss 0.229140 LR 0.000500 Time 0.032924 -2023-02-13 18:13:46,954 - Epoch: [126][ 60/ 1207] Overall Loss 0.228900 Objective Loss 0.228900 LR 0.000500 Time 0.030565 -2023-02-13 18:13:47,142 - Epoch: [126][ 70/ 1207] Overall Loss 0.235697 Objective Loss 0.235697 LR 0.000500 Time 0.028874 -2023-02-13 18:13:47,329 - Epoch: [126][ 80/ 1207] Overall Loss 0.232845 Objective Loss 0.232845 LR 0.000500 Time 0.027609 -2023-02-13 18:13:47,517 - Epoch: [126][ 90/ 1207] Overall Loss 0.230340 Objective Loss 0.230340 LR 0.000500 Time 0.026616 -2023-02-13 18:13:47,705 - Epoch: [126][ 100/ 1207] Overall Loss 0.229567 Objective Loss 0.229567 LR 0.000500 Time 0.025831 -2023-02-13 18:13:47,892 - Epoch: [126][ 110/ 1207] Overall Loss 0.232863 Objective Loss 0.232863 LR 0.000500 Time 0.025179 -2023-02-13 18:13:48,085 - Epoch: [126][ 120/ 1207] Overall Loss 0.236132 Objective Loss 0.236132 LR 0.000500 Time 0.024688 -2023-02-13 18:13:48,280 - Epoch: [126][ 130/ 1207] Overall Loss 0.237042 Objective Loss 0.237042 LR 0.000500 Time 0.024287 -2023-02-13 18:13:48,474 - Epoch: [126][ 140/ 1207] Overall Loss 0.236370 Objective Loss 0.236370 LR 0.000500 Time 0.023936 -2023-02-13 18:13:48,670 - Epoch: [126][ 150/ 1207] Overall Loss 0.235756 Objective Loss 0.235756 LR 0.000500 Time 0.023641 -2023-02-13 18:13:48,863 - Epoch: [126][ 160/ 1207] Overall Loss 0.236605 Objective Loss 0.236605 LR 0.000500 Time 0.023367 -2023-02-13 18:13:49,058 - Epoch: [126][ 170/ 1207] Overall Loss 0.236634 Objective Loss 0.236634 LR 0.000500 Time 0.023142 -2023-02-13 18:13:49,251 - Epoch: [126][ 180/ 1207] Overall Loss 0.236303 Objective Loss 0.236303 LR 0.000500 Time 0.022923 -2023-02-13 18:13:49,447 - Epoch: [126][ 190/ 1207] Overall Loss 0.236438 Objective Loss 0.236438 LR 0.000500 Time 0.022745 -2023-02-13 18:13:49,640 - Epoch: [126][ 200/ 1207] Overall Loss 0.235610 Objective Loss 0.235610 LR 0.000500 Time 0.022572 -2023-02-13 18:13:49,836 - Epoch: [126][ 210/ 1207] Overall Loss 0.236319 Objective Loss 0.236319 LR 0.000500 Time 0.022431 -2023-02-13 18:13:50,029 - Epoch: [126][ 220/ 1207] Overall Loss 0.235724 Objective Loss 0.235724 LR 0.000500 Time 0.022285 -2023-02-13 18:13:50,224 - Epoch: [126][ 230/ 1207] Overall Loss 0.236605 Objective Loss 0.236605 LR 0.000500 Time 0.022164 -2023-02-13 18:13:50,417 - Epoch: [126][ 240/ 1207] Overall Loss 0.236887 Objective Loss 0.236887 LR 0.000500 Time 0.022042 -2023-02-13 18:13:50,613 - Epoch: [126][ 250/ 1207] Overall Loss 0.236734 Objective Loss 0.236734 LR 0.000500 Time 0.021943 -2023-02-13 18:13:50,807 - Epoch: [126][ 260/ 1207] Overall Loss 0.236844 Objective Loss 0.236844 LR 0.000500 Time 0.021844 -2023-02-13 18:13:51,004 - Epoch: [126][ 270/ 1207] Overall Loss 0.237170 Objective Loss 0.237170 LR 0.000500 Time 0.021763 -2023-02-13 18:13:51,197 - Epoch: [126][ 280/ 1207] Overall Loss 0.237447 Objective Loss 0.237447 LR 0.000500 Time 0.021672 -2023-02-13 18:13:51,393 - Epoch: [126][ 290/ 1207] Overall Loss 0.237300 Objective Loss 0.237300 LR 0.000500 Time 0.021601 -2023-02-13 18:13:51,586 - Epoch: [126][ 300/ 1207] Overall Loss 0.237214 Objective Loss 0.237214 LR 0.000500 Time 0.021521 -2023-02-13 18:13:51,782 - Epoch: [126][ 310/ 1207] Overall Loss 0.237546 Objective Loss 0.237546 LR 0.000500 Time 0.021460 -2023-02-13 18:13:51,976 - Epoch: [126][ 320/ 1207] Overall Loss 0.237188 Objective Loss 0.237188 LR 0.000500 Time 0.021393 -2023-02-13 18:13:52,171 - Epoch: [126][ 330/ 1207] Overall Loss 0.237526 Objective Loss 0.237526 LR 0.000500 Time 0.021335 -2023-02-13 18:13:52,365 - Epoch: [126][ 340/ 1207] Overall Loss 0.237037 Objective Loss 0.237037 LR 0.000500 Time 0.021276 -2023-02-13 18:13:52,561 - Epoch: [126][ 350/ 1207] Overall Loss 0.236926 Objective Loss 0.236926 LR 0.000500 Time 0.021229 -2023-02-13 18:13:52,755 - Epoch: [126][ 360/ 1207] Overall Loss 0.236220 Objective Loss 0.236220 LR 0.000500 Time 0.021178 -2023-02-13 18:13:52,951 - Epoch: [126][ 370/ 1207] Overall Loss 0.236814 Objective Loss 0.236814 LR 0.000500 Time 0.021133 -2023-02-13 18:13:53,145 - Epoch: [126][ 380/ 1207] Overall Loss 0.236191 Objective Loss 0.236191 LR 0.000500 Time 0.021087 -2023-02-13 18:13:53,342 - Epoch: [126][ 390/ 1207] Overall Loss 0.236654 Objective Loss 0.236654 LR 0.000500 Time 0.021049 -2023-02-13 18:13:53,535 - Epoch: [126][ 400/ 1207] Overall Loss 0.237815 Objective Loss 0.237815 LR 0.000500 Time 0.021006 -2023-02-13 18:13:53,732 - Epoch: [126][ 410/ 1207] Overall Loss 0.237865 Objective Loss 0.237865 LR 0.000500 Time 0.020973 -2023-02-13 18:13:53,925 - Epoch: [126][ 420/ 1207] Overall Loss 0.238388 Objective Loss 0.238388 LR 0.000500 Time 0.020932 -2023-02-13 18:13:54,122 - Epoch: [126][ 430/ 1207] Overall Loss 0.238235 Objective Loss 0.238235 LR 0.000500 Time 0.020902 -2023-02-13 18:13:54,316 - Epoch: [126][ 440/ 1207] Overall Loss 0.237962 Objective Loss 0.237962 LR 0.000500 Time 0.020866 -2023-02-13 18:13:54,505 - Epoch: [126][ 450/ 1207] Overall Loss 0.237845 Objective Loss 0.237845 LR 0.000500 Time 0.020823 -2023-02-13 18:13:54,695 - Epoch: [126][ 460/ 1207] Overall Loss 0.237530 Objective Loss 0.237530 LR 0.000500 Time 0.020782 -2023-02-13 18:13:54,885 - Epoch: [126][ 470/ 1207] Overall Loss 0.237701 Objective Loss 0.237701 LR 0.000500 Time 0.020743 -2023-02-13 18:13:55,074 - Epoch: [126][ 480/ 1207] Overall Loss 0.237802 Objective Loss 0.237802 LR 0.000500 Time 0.020704 -2023-02-13 18:13:55,263 - Epoch: [126][ 490/ 1207] Overall Loss 0.238516 Objective Loss 0.238516 LR 0.000500 Time 0.020666 -2023-02-13 18:13:55,452 - Epoch: [126][ 500/ 1207] Overall Loss 0.238539 Objective Loss 0.238539 LR 0.000500 Time 0.020630 -2023-02-13 18:13:55,640 - Epoch: [126][ 510/ 1207] Overall Loss 0.238378 Objective Loss 0.238378 LR 0.000500 Time 0.020594 -2023-02-13 18:13:55,830 - Epoch: [126][ 520/ 1207] Overall Loss 0.238827 Objective Loss 0.238827 LR 0.000500 Time 0.020563 -2023-02-13 18:13:56,020 - Epoch: [126][ 530/ 1207] Overall Loss 0.239066 Objective Loss 0.239066 LR 0.000500 Time 0.020533 -2023-02-13 18:13:56,209 - Epoch: [126][ 540/ 1207] Overall Loss 0.239357 Objective Loss 0.239357 LR 0.000500 Time 0.020501 -2023-02-13 18:13:56,397 - Epoch: [126][ 550/ 1207] Overall Loss 0.239177 Objective Loss 0.239177 LR 0.000500 Time 0.020470 -2023-02-13 18:13:56,586 - Epoch: [126][ 560/ 1207] Overall Loss 0.239331 Objective Loss 0.239331 LR 0.000500 Time 0.020442 -2023-02-13 18:13:56,776 - Epoch: [126][ 570/ 1207] Overall Loss 0.239674 Objective Loss 0.239674 LR 0.000500 Time 0.020415 -2023-02-13 18:13:56,966 - Epoch: [126][ 580/ 1207] Overall Loss 0.239794 Objective Loss 0.239794 LR 0.000500 Time 0.020391 -2023-02-13 18:13:57,154 - Epoch: [126][ 590/ 1207] Overall Loss 0.239798 Objective Loss 0.239798 LR 0.000500 Time 0.020363 -2023-02-13 18:13:57,343 - Epoch: [126][ 600/ 1207] Overall Loss 0.239819 Objective Loss 0.239819 LR 0.000500 Time 0.020338 -2023-02-13 18:13:57,532 - Epoch: [126][ 610/ 1207] Overall Loss 0.239615 Objective Loss 0.239615 LR 0.000500 Time 0.020313 -2023-02-13 18:13:57,721 - Epoch: [126][ 620/ 1207] Overall Loss 0.239702 Objective Loss 0.239702 LR 0.000500 Time 0.020289 -2023-02-13 18:13:57,910 - Epoch: [126][ 630/ 1207] Overall Loss 0.239460 Objective Loss 0.239460 LR 0.000500 Time 0.020268 -2023-02-13 18:13:58,099 - Epoch: [126][ 640/ 1207] Overall Loss 0.239643 Objective Loss 0.239643 LR 0.000500 Time 0.020246 -2023-02-13 18:13:58,289 - Epoch: [126][ 650/ 1207] Overall Loss 0.239719 Objective Loss 0.239719 LR 0.000500 Time 0.020225 -2023-02-13 18:13:58,478 - Epoch: [126][ 660/ 1207] Overall Loss 0.239746 Objective Loss 0.239746 LR 0.000500 Time 0.020205 -2023-02-13 18:13:58,668 - Epoch: [126][ 670/ 1207] Overall Loss 0.239956 Objective Loss 0.239956 LR 0.000500 Time 0.020186 -2023-02-13 18:13:58,858 - Epoch: [126][ 680/ 1207] Overall Loss 0.240032 Objective Loss 0.240032 LR 0.000500 Time 0.020169 -2023-02-13 18:13:59,047 - Epoch: [126][ 690/ 1207] Overall Loss 0.239666 Objective Loss 0.239666 LR 0.000500 Time 0.020149 -2023-02-13 18:13:59,236 - Epoch: [126][ 700/ 1207] Overall Loss 0.239712 Objective Loss 0.239712 LR 0.000500 Time 0.020131 -2023-02-13 18:13:59,426 - Epoch: [126][ 710/ 1207] Overall Loss 0.239767 Objective Loss 0.239767 LR 0.000500 Time 0.020114 -2023-02-13 18:13:59,616 - Epoch: [126][ 720/ 1207] Overall Loss 0.239725 Objective Loss 0.239725 LR 0.000500 Time 0.020099 -2023-02-13 18:13:59,806 - Epoch: [126][ 730/ 1207] Overall Loss 0.239594 Objective Loss 0.239594 LR 0.000500 Time 0.020082 -2023-02-13 18:13:59,995 - Epoch: [126][ 740/ 1207] Overall Loss 0.239940 Objective Loss 0.239940 LR 0.000500 Time 0.020066 -2023-02-13 18:14:00,183 - Epoch: [126][ 750/ 1207] Overall Loss 0.239661 Objective Loss 0.239661 LR 0.000500 Time 0.020049 -2023-02-13 18:14:00,372 - Epoch: [126][ 760/ 1207] Overall Loss 0.239459 Objective Loss 0.239459 LR 0.000500 Time 0.020033 -2023-02-13 18:14:00,561 - Epoch: [126][ 770/ 1207] Overall Loss 0.239646 Objective Loss 0.239646 LR 0.000500 Time 0.020018 -2023-02-13 18:14:00,751 - Epoch: [126][ 780/ 1207] Overall Loss 0.239666 Objective Loss 0.239666 LR 0.000500 Time 0.020005 -2023-02-13 18:14:00,941 - Epoch: [126][ 790/ 1207] Overall Loss 0.239662 Objective Loss 0.239662 LR 0.000500 Time 0.019991 -2023-02-13 18:14:01,130 - Epoch: [126][ 800/ 1207] Overall Loss 0.239871 Objective Loss 0.239871 LR 0.000500 Time 0.019978 -2023-02-13 18:14:01,319 - Epoch: [126][ 810/ 1207] Overall Loss 0.239548 Objective Loss 0.239548 LR 0.000500 Time 0.019964 -2023-02-13 18:14:01,509 - Epoch: [126][ 820/ 1207] Overall Loss 0.239596 Objective Loss 0.239596 LR 0.000500 Time 0.019952 -2023-02-13 18:14:01,699 - Epoch: [126][ 830/ 1207] Overall Loss 0.239621 Objective Loss 0.239621 LR 0.000500 Time 0.019939 -2023-02-13 18:14:01,889 - Epoch: [126][ 840/ 1207] Overall Loss 0.239832 Objective Loss 0.239832 LR 0.000500 Time 0.019928 -2023-02-13 18:14:02,078 - Epoch: [126][ 850/ 1207] Overall Loss 0.239847 Objective Loss 0.239847 LR 0.000500 Time 0.019916 -2023-02-13 18:14:02,268 - Epoch: [126][ 860/ 1207] Overall Loss 0.239874 Objective Loss 0.239874 LR 0.000500 Time 0.019904 -2023-02-13 18:14:02,457 - Epoch: [126][ 870/ 1207] Overall Loss 0.240051 Objective Loss 0.240051 LR 0.000500 Time 0.019892 -2023-02-13 18:14:02,646 - Epoch: [126][ 880/ 1207] Overall Loss 0.239948 Objective Loss 0.239948 LR 0.000500 Time 0.019881 -2023-02-13 18:14:02,837 - Epoch: [126][ 890/ 1207] Overall Loss 0.240133 Objective Loss 0.240133 LR 0.000500 Time 0.019871 -2023-02-13 18:14:03,026 - Epoch: [126][ 900/ 1207] Overall Loss 0.240193 Objective Loss 0.240193 LR 0.000500 Time 0.019860 -2023-02-13 18:14:03,215 - Epoch: [126][ 910/ 1207] Overall Loss 0.240223 Objective Loss 0.240223 LR 0.000500 Time 0.019849 -2023-02-13 18:14:03,405 - Epoch: [126][ 920/ 1207] Overall Loss 0.240073 Objective Loss 0.240073 LR 0.000500 Time 0.019839 -2023-02-13 18:14:03,594 - Epoch: [126][ 930/ 1207] Overall Loss 0.239897 Objective Loss 0.239897 LR 0.000500 Time 0.019830 -2023-02-13 18:14:03,785 - Epoch: [126][ 940/ 1207] Overall Loss 0.239906 Objective Loss 0.239906 LR 0.000500 Time 0.019821 -2023-02-13 18:14:03,975 - Epoch: [126][ 950/ 1207] Overall Loss 0.240005 Objective Loss 0.240005 LR 0.000500 Time 0.019812 -2023-02-13 18:14:04,164 - Epoch: [126][ 960/ 1207] Overall Loss 0.239944 Objective Loss 0.239944 LR 0.000500 Time 0.019802 -2023-02-13 18:14:04,354 - Epoch: [126][ 970/ 1207] Overall Loss 0.240286 Objective Loss 0.240286 LR 0.000500 Time 0.019793 -2023-02-13 18:14:04,543 - Epoch: [126][ 980/ 1207] Overall Loss 0.240486 Objective Loss 0.240486 LR 0.000500 Time 0.019784 -2023-02-13 18:14:04,732 - Epoch: [126][ 990/ 1207] Overall Loss 0.240121 Objective Loss 0.240121 LR 0.000500 Time 0.019774 -2023-02-13 18:14:04,922 - Epoch: [126][ 1000/ 1207] Overall Loss 0.240111 Objective Loss 0.240111 LR 0.000500 Time 0.019766 -2023-02-13 18:14:05,111 - Epoch: [126][ 1010/ 1207] Overall Loss 0.240369 Objective Loss 0.240369 LR 0.000500 Time 0.019757 -2023-02-13 18:14:05,300 - Epoch: [126][ 1020/ 1207] Overall Loss 0.240317 Objective Loss 0.240317 LR 0.000500 Time 0.019749 -2023-02-13 18:14:05,489 - Epoch: [126][ 1030/ 1207] Overall Loss 0.240480 Objective Loss 0.240480 LR 0.000500 Time 0.019740 -2023-02-13 18:14:05,679 - Epoch: [126][ 1040/ 1207] Overall Loss 0.240875 Objective Loss 0.240875 LR 0.000500 Time 0.019732 -2023-02-13 18:14:05,869 - Epoch: [126][ 1050/ 1207] Overall Loss 0.241045 Objective Loss 0.241045 LR 0.000500 Time 0.019725 -2023-02-13 18:14:06,059 - Epoch: [126][ 1060/ 1207] Overall Loss 0.241156 Objective Loss 0.241156 LR 0.000500 Time 0.019718 -2023-02-13 18:14:06,248 - Epoch: [126][ 1070/ 1207] Overall Loss 0.241195 Objective Loss 0.241195 LR 0.000500 Time 0.019710 -2023-02-13 18:14:06,438 - Epoch: [126][ 1080/ 1207] Overall Loss 0.241508 Objective Loss 0.241508 LR 0.000500 Time 0.019703 -2023-02-13 18:14:06,627 - Epoch: [126][ 1090/ 1207] Overall Loss 0.241618 Objective Loss 0.241618 LR 0.000500 Time 0.019695 -2023-02-13 18:14:06,818 - Epoch: [126][ 1100/ 1207] Overall Loss 0.242106 Objective Loss 0.242106 LR 0.000500 Time 0.019689 -2023-02-13 18:14:07,007 - Epoch: [126][ 1110/ 1207] Overall Loss 0.242434 Objective Loss 0.242434 LR 0.000500 Time 0.019683 -2023-02-13 18:14:07,197 - Epoch: [126][ 1120/ 1207] Overall Loss 0.242545 Objective Loss 0.242545 LR 0.000500 Time 0.019676 -2023-02-13 18:14:07,387 - Epoch: [126][ 1130/ 1207] Overall Loss 0.242471 Objective Loss 0.242471 LR 0.000500 Time 0.019669 -2023-02-13 18:14:07,576 - Epoch: [126][ 1140/ 1207] Overall Loss 0.242627 Objective Loss 0.242627 LR 0.000500 Time 0.019663 -2023-02-13 18:14:07,767 - Epoch: [126][ 1150/ 1207] Overall Loss 0.242812 Objective Loss 0.242812 LR 0.000500 Time 0.019657 -2023-02-13 18:14:07,956 - Epoch: [126][ 1160/ 1207] Overall Loss 0.242981 Objective Loss 0.242981 LR 0.000500 Time 0.019651 -2023-02-13 18:14:08,146 - Epoch: [126][ 1170/ 1207] Overall Loss 0.242663 Objective Loss 0.242663 LR 0.000500 Time 0.019644 -2023-02-13 18:14:08,335 - Epoch: [126][ 1180/ 1207] Overall Loss 0.242748 Objective Loss 0.242748 LR 0.000500 Time 0.019638 -2023-02-13 18:14:08,525 - Epoch: [126][ 1190/ 1207] Overall Loss 0.243023 Objective Loss 0.243023 LR 0.000500 Time 0.019632 -2023-02-13 18:14:08,771 - Epoch: [126][ 1200/ 1207] Overall Loss 0.243090 Objective Loss 0.243090 LR 0.000500 Time 0.019673 -2023-02-13 18:14:08,888 - Epoch: [126][ 1207/ 1207] Overall Loss 0.242975 Objective Loss 0.242975 Top1 90.243902 Top5 98.170732 LR 0.000500 Time 0.019656 -2023-02-13 18:14:08,959 - --- validate (epoch=126)----------- -2023-02-13 18:14:08,959 - 34311 samples (256 per mini-batch) -2023-02-13 18:14:09,362 - Epoch: [126][ 10/ 135] Loss 0.310669 Top1 84.843750 Top5 97.500000 -2023-02-13 18:14:09,490 - Epoch: [126][ 20/ 135] Loss 0.308741 Top1 84.609375 Top5 97.656250 -2023-02-13 18:14:09,626 - Epoch: [126][ 30/ 135] Loss 0.305335 Top1 84.882812 Top5 97.656250 -2023-02-13 18:14:09,761 - Epoch: [126][ 40/ 135] Loss 0.299051 Top1 84.824219 Top5 97.695312 -2023-02-13 18:14:09,902 - Epoch: [126][ 50/ 135] Loss 0.291419 Top1 84.968750 Top5 97.789062 -2023-02-13 18:14:10,039 - Epoch: [126][ 60/ 135] Loss 0.298345 Top1 84.811198 Top5 97.721354 -2023-02-13 18:14:10,179 - Epoch: [126][ 70/ 135] Loss 0.296982 Top1 84.893973 Top5 97.812500 -2023-02-13 18:14:10,318 - Epoch: [126][ 80/ 135] Loss 0.302553 Top1 84.677734 Top5 97.807617 -2023-02-13 18:14:10,456 - Epoch: [126][ 90/ 135] Loss 0.305907 Top1 84.609375 Top5 97.760417 -2023-02-13 18:14:10,595 - Epoch: [126][ 100/ 135] Loss 0.309658 Top1 84.550781 Top5 97.765625 -2023-02-13 18:14:10,722 - Epoch: [126][ 110/ 135] Loss 0.311968 Top1 84.506392 Top5 97.737926 -2023-02-13 18:14:10,850 - Epoch: [126][ 120/ 135] Loss 0.312895 Top1 84.459635 Top5 97.734375 -2023-02-13 18:14:10,981 - Epoch: [126][ 130/ 135] Loss 0.311440 Top1 84.480168 Top5 97.731370 -2023-02-13 18:14:11,027 - Epoch: [126][ 135/ 135] Loss 0.310151 Top1 84.480196 Top5 97.709189 -2023-02-13 18:14:11,095 - ==> Top1: 84.480 Top5: 97.709 Loss: 0.310 - -2023-02-13 18:14:11,096 - ==> Confusion: -[[ 866 6 4 0 6 2 0 2 2 47 2 4 1 3 3 3 3 2 2 1 8] - [ 5 958 2 1 10 17 1 14 2 0 1 2 0 0 2 2 4 0 5 2 5] - [ 7 9 944 10 6 1 14 14 0 1 3 1 4 4 6 6 2 4 8 4 10] - [ 7 3 23 893 3 3 0 3 0 2 12 0 8 0 23 4 4 6 19 1 2] - [ 14 7 0 1 995 7 1 1 2 4 0 6 2 4 4 5 5 1 1 2 4] - [ 1 18 0 2 6 963 5 20 2 4 2 6 4 15 1 2 5 3 1 5 5] - [ 2 6 12 1 0 4 1035 5 2 1 2 2 3 2 0 2 1 4 3 9 3] - [ 0 11 13 1 3 29 2 924 1 1 1 3 2 0 0 1 0 1 22 6 3] - [ 15 4 1 1 1 1 1 3 882 51 4 1 1 11 22 3 0 1 2 0 4] - [ 81 0 2 0 7 2 0 2 30 858 1 1 1 15 5 1 1 1 1 0 3] - [ 1 2 7 7 0 3 3 4 17 0 968 4 0 12 3 0 1 1 13 0 5] - [ 0 3 1 1 2 9 0 3 0 1 0 921 22 7 0 4 8 10 1 10 2] - [ 0 0 0 5 1 3 0 1 2 0 1 31 870 1 3 8 3 18 1 1 10] - [ 4 4 1 0 9 8 0 2 15 20 12 4 1 917 3 4 5 2 2 2 9] - [ 6 3 2 15 5 6 0 1 11 6 2 0 4 2 1000 1 4 8 6 3 7] - [ 7 2 6 0 9 3 4 0 0 0 0 10 6 1 0 960 10 13 0 6 9] - [ 2 10 1 2 7 1 0 2 2 1 0 2 3 3 2 6 1001 2 1 5 8] - [ 7 3 2 4 1 2 2 0 1 0 1 9 11 0 2 10 1 990 0 1 4] - [ 6 5 2 4 1 1 1 18 4 0 2 2 6 0 11 1 0 3 1016 1 2] - [ 2 5 3 0 0 8 3 11 0 0 0 14 2 4 1 6 5 4 0 1072 8] - [ 155 297 201 105 147 172 75 175 97 91 147 126 288 243 175 106 258 116 226 281 9953]] - -2023-02-13 18:14:11,097 - ==> Best [Top1: 84.480 Top5: 97.709 Sparsity:0.00 Params: 148928 on epoch: 126] -2023-02-13 18:14:11,097 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:14:11,104 - - -2023-02-13 18:14:11,104 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:14:12,093 - Epoch: [127][ 10/ 1207] Overall Loss 0.226117 Objective Loss 0.226117 LR 0.000500 Time 0.098874 -2023-02-13 18:14:12,292 - Epoch: [127][ 20/ 1207] Overall Loss 0.236547 Objective Loss 0.236547 LR 0.000500 Time 0.059329 -2023-02-13 18:14:12,481 - Epoch: [127][ 30/ 1207] Overall Loss 0.227812 Objective Loss 0.227812 LR 0.000500 Time 0.045867 -2023-02-13 18:14:12,671 - Epoch: [127][ 40/ 1207] Overall Loss 0.231017 Objective Loss 0.231017 LR 0.000500 Time 0.039135 -2023-02-13 18:14:12,861 - Epoch: [127][ 50/ 1207] Overall Loss 0.230395 Objective Loss 0.230395 LR 0.000500 Time 0.035095 -2023-02-13 18:14:13,051 - Epoch: [127][ 60/ 1207] Overall Loss 0.228374 Objective Loss 0.228374 LR 0.000500 Time 0.032401 -2023-02-13 18:14:13,240 - Epoch: [127][ 70/ 1207] Overall Loss 0.231691 Objective Loss 0.231691 LR 0.000500 Time 0.030467 -2023-02-13 18:14:13,429 - Epoch: [127][ 80/ 1207] Overall Loss 0.235195 Objective Loss 0.235195 LR 0.000500 Time 0.029024 -2023-02-13 18:14:13,618 - Epoch: [127][ 90/ 1207] Overall Loss 0.238467 Objective Loss 0.238467 LR 0.000500 Time 0.027897 -2023-02-13 18:14:13,809 - Epoch: [127][ 100/ 1207] Overall Loss 0.238150 Objective Loss 0.238150 LR 0.000500 Time 0.027004 -2023-02-13 18:14:13,998 - Epoch: [127][ 110/ 1207] Overall Loss 0.236817 Objective Loss 0.236817 LR 0.000500 Time 0.026269 -2023-02-13 18:14:14,188 - Epoch: [127][ 120/ 1207] Overall Loss 0.236231 Objective Loss 0.236231 LR 0.000500 Time 0.025662 -2023-02-13 18:14:14,377 - Epoch: [127][ 130/ 1207] Overall Loss 0.239351 Objective Loss 0.239351 LR 0.000500 Time 0.025140 -2023-02-13 18:14:14,568 - Epoch: [127][ 140/ 1207] Overall Loss 0.241034 Objective Loss 0.241034 LR 0.000500 Time 0.024701 -2023-02-13 18:14:14,758 - Epoch: [127][ 150/ 1207] Overall Loss 0.241856 Objective Loss 0.241856 LR 0.000500 Time 0.024317 -2023-02-13 18:14:14,948 - Epoch: [127][ 160/ 1207] Overall Loss 0.240807 Objective Loss 0.240807 LR 0.000500 Time 0.023985 -2023-02-13 18:14:15,137 - Epoch: [127][ 170/ 1207] Overall Loss 0.241996 Objective Loss 0.241996 LR 0.000500 Time 0.023685 -2023-02-13 18:14:15,327 - Epoch: [127][ 180/ 1207] Overall Loss 0.241448 Objective Loss 0.241448 LR 0.000500 Time 0.023420 -2023-02-13 18:14:15,516 - Epoch: [127][ 190/ 1207] Overall Loss 0.241546 Objective Loss 0.241546 LR 0.000500 Time 0.023181 -2023-02-13 18:14:15,706 - Epoch: [127][ 200/ 1207] Overall Loss 0.241523 Objective Loss 0.241523 LR 0.000500 Time 0.022969 -2023-02-13 18:14:15,898 - Epoch: [127][ 210/ 1207] Overall Loss 0.241548 Objective Loss 0.241548 LR 0.000500 Time 0.022787 -2023-02-13 18:14:16,088 - Epoch: [127][ 220/ 1207] Overall Loss 0.241658 Objective Loss 0.241658 LR 0.000500 Time 0.022615 -2023-02-13 18:14:16,278 - Epoch: [127][ 230/ 1207] Overall Loss 0.239532 Objective Loss 0.239532 LR 0.000500 Time 0.022456 -2023-02-13 18:14:16,468 - Epoch: [127][ 240/ 1207] Overall Loss 0.239271 Objective Loss 0.239271 LR 0.000500 Time 0.022311 -2023-02-13 18:14:16,658 - Epoch: [127][ 250/ 1207] Overall Loss 0.238874 Objective Loss 0.238874 LR 0.000500 Time 0.022178 -2023-02-13 18:14:16,848 - Epoch: [127][ 260/ 1207] Overall Loss 0.239757 Objective Loss 0.239757 LR 0.000500 Time 0.022055 -2023-02-13 18:14:17,039 - Epoch: [127][ 270/ 1207] Overall Loss 0.239851 Objective Loss 0.239851 LR 0.000500 Time 0.021942 -2023-02-13 18:14:17,229 - Epoch: [127][ 280/ 1207] Overall Loss 0.239530 Objective Loss 0.239530 LR 0.000500 Time 0.021837 -2023-02-13 18:14:17,419 - Epoch: [127][ 290/ 1207] Overall Loss 0.239802 Objective Loss 0.239802 LR 0.000500 Time 0.021737 -2023-02-13 18:14:17,609 - Epoch: [127][ 300/ 1207] Overall Loss 0.240012 Objective Loss 0.240012 LR 0.000500 Time 0.021644 -2023-02-13 18:14:17,799 - Epoch: [127][ 310/ 1207] Overall Loss 0.241295 Objective Loss 0.241295 LR 0.000500 Time 0.021558 -2023-02-13 18:14:17,990 - Epoch: [127][ 320/ 1207] Overall Loss 0.241060 Objective Loss 0.241060 LR 0.000500 Time 0.021479 -2023-02-13 18:14:18,179 - Epoch: [127][ 330/ 1207] Overall Loss 0.240245 Objective Loss 0.240245 LR 0.000500 Time 0.021402 -2023-02-13 18:14:18,370 - Epoch: [127][ 340/ 1207] Overall Loss 0.240020 Objective Loss 0.240020 LR 0.000500 Time 0.021332 -2023-02-13 18:14:18,560 - Epoch: [127][ 350/ 1207] Overall Loss 0.240040 Objective Loss 0.240040 LR 0.000500 Time 0.021263 -2023-02-13 18:14:18,750 - Epoch: [127][ 360/ 1207] Overall Loss 0.239952 Objective Loss 0.239952 LR 0.000500 Time 0.021200 -2023-02-13 18:14:18,941 - Epoch: [127][ 370/ 1207] Overall Loss 0.239065 Objective Loss 0.239065 LR 0.000500 Time 0.021143 -2023-02-13 18:14:19,132 - Epoch: [127][ 380/ 1207] Overall Loss 0.238235 Objective Loss 0.238235 LR 0.000500 Time 0.021087 -2023-02-13 18:14:19,322 - Epoch: [127][ 390/ 1207] Overall Loss 0.238203 Objective Loss 0.238203 LR 0.000500 Time 0.021033 -2023-02-13 18:14:19,512 - Epoch: [127][ 400/ 1207] Overall Loss 0.238278 Objective Loss 0.238278 LR 0.000500 Time 0.020982 -2023-02-13 18:14:19,702 - Epoch: [127][ 410/ 1207] Overall Loss 0.238784 Objective Loss 0.238784 LR 0.000500 Time 0.020931 -2023-02-13 18:14:19,893 - Epoch: [127][ 420/ 1207] Overall Loss 0.238682 Objective Loss 0.238682 LR 0.000500 Time 0.020887 -2023-02-13 18:14:20,082 - Epoch: [127][ 430/ 1207] Overall Loss 0.238378 Objective Loss 0.238378 LR 0.000500 Time 0.020841 -2023-02-13 18:14:20,272 - Epoch: [127][ 440/ 1207] Overall Loss 0.238486 Objective Loss 0.238486 LR 0.000500 Time 0.020798 -2023-02-13 18:14:20,462 - Epoch: [127][ 450/ 1207] Overall Loss 0.238550 Objective Loss 0.238550 LR 0.000500 Time 0.020756 -2023-02-13 18:14:20,652 - Epoch: [127][ 460/ 1207] Overall Loss 0.238314 Objective Loss 0.238314 LR 0.000500 Time 0.020718 -2023-02-13 18:14:20,843 - Epoch: [127][ 470/ 1207] Overall Loss 0.238651 Objective Loss 0.238651 LR 0.000500 Time 0.020682 -2023-02-13 18:14:21,033 - Epoch: [127][ 480/ 1207] Overall Loss 0.239054 Objective Loss 0.239054 LR 0.000500 Time 0.020647 -2023-02-13 18:14:21,222 - Epoch: [127][ 490/ 1207] Overall Loss 0.239130 Objective Loss 0.239130 LR 0.000500 Time 0.020611 -2023-02-13 18:14:21,413 - Epoch: [127][ 500/ 1207] Overall Loss 0.239287 Objective Loss 0.239287 LR 0.000500 Time 0.020579 -2023-02-13 18:14:21,603 - Epoch: [127][ 510/ 1207] Overall Loss 0.239962 Objective Loss 0.239962 LR 0.000500 Time 0.020548 -2023-02-13 18:14:21,793 - Epoch: [127][ 520/ 1207] Overall Loss 0.240132 Objective Loss 0.240132 LR 0.000500 Time 0.020518 -2023-02-13 18:14:21,985 - Epoch: [127][ 530/ 1207] Overall Loss 0.239953 Objective Loss 0.239953 LR 0.000500 Time 0.020491 -2023-02-13 18:14:22,175 - Epoch: [127][ 540/ 1207] Overall Loss 0.239427 Objective Loss 0.239427 LR 0.000500 Time 0.020463 -2023-02-13 18:14:22,365 - Epoch: [127][ 550/ 1207] Overall Loss 0.239204 Objective Loss 0.239204 LR 0.000500 Time 0.020436 -2023-02-13 18:14:22,556 - Epoch: [127][ 560/ 1207] Overall Loss 0.239143 Objective Loss 0.239143 LR 0.000500 Time 0.020411 -2023-02-13 18:14:22,746 - Epoch: [127][ 570/ 1207] Overall Loss 0.238948 Objective Loss 0.238948 LR 0.000500 Time 0.020386 -2023-02-13 18:14:22,937 - Epoch: [127][ 580/ 1207] Overall Loss 0.238400 Objective Loss 0.238400 LR 0.000500 Time 0.020363 -2023-02-13 18:14:23,127 - Epoch: [127][ 590/ 1207] Overall Loss 0.238698 Objective Loss 0.238698 LR 0.000500 Time 0.020339 -2023-02-13 18:14:23,317 - Epoch: [127][ 600/ 1207] Overall Loss 0.238806 Objective Loss 0.238806 LR 0.000500 Time 0.020316 -2023-02-13 18:14:23,507 - Epoch: [127][ 610/ 1207] Overall Loss 0.239104 Objective Loss 0.239104 LR 0.000500 Time 0.020294 -2023-02-13 18:14:23,697 - Epoch: [127][ 620/ 1207] Overall Loss 0.238985 Objective Loss 0.238985 LR 0.000500 Time 0.020273 -2023-02-13 18:14:23,887 - Epoch: [127][ 630/ 1207] Overall Loss 0.239249 Objective Loss 0.239249 LR 0.000500 Time 0.020253 -2023-02-13 18:14:24,078 - Epoch: [127][ 640/ 1207] Overall Loss 0.239212 Objective Loss 0.239212 LR 0.000500 Time 0.020233 -2023-02-13 18:14:24,267 - Epoch: [127][ 650/ 1207] Overall Loss 0.239284 Objective Loss 0.239284 LR 0.000500 Time 0.020213 -2023-02-13 18:14:24,458 - Epoch: [127][ 660/ 1207] Overall Loss 0.239602 Objective Loss 0.239602 LR 0.000500 Time 0.020194 -2023-02-13 18:14:24,647 - Epoch: [127][ 670/ 1207] Overall Loss 0.239623 Objective Loss 0.239623 LR 0.000500 Time 0.020175 -2023-02-13 18:14:24,838 - Epoch: [127][ 680/ 1207] Overall Loss 0.239859 Objective Loss 0.239859 LR 0.000500 Time 0.020158 -2023-02-13 18:14:25,029 - Epoch: [127][ 690/ 1207] Overall Loss 0.239631 Objective Loss 0.239631 LR 0.000500 Time 0.020143 -2023-02-13 18:14:25,219 - Epoch: [127][ 700/ 1207] Overall Loss 0.239960 Objective Loss 0.239960 LR 0.000500 Time 0.020126 -2023-02-13 18:14:25,409 - Epoch: [127][ 710/ 1207] Overall Loss 0.239982 Objective Loss 0.239982 LR 0.000500 Time 0.020110 -2023-02-13 18:14:25,600 - Epoch: [127][ 720/ 1207] Overall Loss 0.239956 Objective Loss 0.239956 LR 0.000500 Time 0.020096 -2023-02-13 18:14:25,790 - Epoch: [127][ 730/ 1207] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.000500 Time 0.020080 -2023-02-13 18:14:25,982 - Epoch: [127][ 740/ 1207] Overall Loss 0.239719 Objective Loss 0.239719 LR 0.000500 Time 0.020067 -2023-02-13 18:14:26,172 - Epoch: [127][ 750/ 1207] Overall Loss 0.239728 Objective Loss 0.239728 LR 0.000500 Time 0.020053 -2023-02-13 18:14:26,363 - Epoch: [127][ 760/ 1207] Overall Loss 0.239428 Objective Loss 0.239428 LR 0.000500 Time 0.020039 -2023-02-13 18:14:26,553 - Epoch: [127][ 770/ 1207] Overall Loss 0.239698 Objective Loss 0.239698 LR 0.000500 Time 0.020025 -2023-02-13 18:14:26,744 - Epoch: [127][ 780/ 1207] Overall Loss 0.239695 Objective Loss 0.239695 LR 0.000500 Time 0.020013 -2023-02-13 18:14:26,935 - Epoch: [127][ 790/ 1207] Overall Loss 0.239932 Objective Loss 0.239932 LR 0.000500 Time 0.020001 -2023-02-13 18:14:27,126 - Epoch: [127][ 800/ 1207] Overall Loss 0.239759 Objective Loss 0.239759 LR 0.000500 Time 0.019989 -2023-02-13 18:14:27,316 - Epoch: [127][ 810/ 1207] Overall Loss 0.239295 Objective Loss 0.239295 LR 0.000500 Time 0.019976 -2023-02-13 18:14:27,506 - Epoch: [127][ 820/ 1207] Overall Loss 0.239291 Objective Loss 0.239291 LR 0.000500 Time 0.019965 -2023-02-13 18:14:27,697 - Epoch: [127][ 830/ 1207] Overall Loss 0.239465 Objective Loss 0.239465 LR 0.000500 Time 0.019953 -2023-02-13 18:14:27,888 - Epoch: [127][ 840/ 1207] Overall Loss 0.239343 Objective Loss 0.239343 LR 0.000500 Time 0.019943 -2023-02-13 18:14:28,078 - Epoch: [127][ 850/ 1207] Overall Loss 0.239445 Objective Loss 0.239445 LR 0.000500 Time 0.019931 -2023-02-13 18:14:28,269 - Epoch: [127][ 860/ 1207] Overall Loss 0.239871 Objective Loss 0.239871 LR 0.000500 Time 0.019921 -2023-02-13 18:14:28,458 - Epoch: [127][ 870/ 1207] Overall Loss 0.239869 Objective Loss 0.239869 LR 0.000500 Time 0.019909 -2023-02-13 18:14:28,649 - Epoch: [127][ 880/ 1207] Overall Loss 0.239957 Objective Loss 0.239957 LR 0.000500 Time 0.019899 -2023-02-13 18:14:28,839 - Epoch: [127][ 890/ 1207] Overall Loss 0.240615 Objective Loss 0.240615 LR 0.000500 Time 0.019889 -2023-02-13 18:14:29,030 - Epoch: [127][ 900/ 1207] Overall Loss 0.240927 Objective Loss 0.240927 LR 0.000500 Time 0.019880 -2023-02-13 18:14:29,220 - Epoch: [127][ 910/ 1207] Overall Loss 0.240968 Objective Loss 0.240968 LR 0.000500 Time 0.019870 -2023-02-13 18:14:29,412 - Epoch: [127][ 920/ 1207] Overall Loss 0.241058 Objective Loss 0.241058 LR 0.000500 Time 0.019861 -2023-02-13 18:14:29,602 - Epoch: [127][ 930/ 1207] Overall Loss 0.241195 Objective Loss 0.241195 LR 0.000500 Time 0.019852 -2023-02-13 18:14:29,793 - Epoch: [127][ 940/ 1207] Overall Loss 0.241108 Objective Loss 0.241108 LR 0.000500 Time 0.019843 -2023-02-13 18:14:29,984 - Epoch: [127][ 950/ 1207] Overall Loss 0.241311 Objective Loss 0.241311 LR 0.000500 Time 0.019836 -2023-02-13 18:14:30,175 - Epoch: [127][ 960/ 1207] Overall Loss 0.241060 Objective Loss 0.241060 LR 0.000500 Time 0.019827 -2023-02-13 18:14:30,365 - Epoch: [127][ 970/ 1207] Overall Loss 0.240926 Objective Loss 0.240926 LR 0.000500 Time 0.019819 -2023-02-13 18:14:30,556 - Epoch: [127][ 980/ 1207] Overall Loss 0.240874 Objective Loss 0.240874 LR 0.000500 Time 0.019811 -2023-02-13 18:14:30,747 - Epoch: [127][ 990/ 1207] Overall Loss 0.240673 Objective Loss 0.240673 LR 0.000500 Time 0.019803 -2023-02-13 18:14:30,940 - Epoch: [127][ 1000/ 1207] Overall Loss 0.240819 Objective Loss 0.240819 LR 0.000500 Time 0.019797 -2023-02-13 18:14:31,130 - Epoch: [127][ 1010/ 1207] Overall Loss 0.241097 Objective Loss 0.241097 LR 0.000500 Time 0.019789 -2023-02-13 18:14:31,320 - Epoch: [127][ 1020/ 1207] Overall Loss 0.241248 Objective Loss 0.241248 LR 0.000500 Time 0.019782 -2023-02-13 18:14:31,511 - Epoch: [127][ 1030/ 1207] Overall Loss 0.241324 Objective Loss 0.241324 LR 0.000500 Time 0.019774 -2023-02-13 18:14:31,701 - Epoch: [127][ 1040/ 1207] Overall Loss 0.241162 Objective Loss 0.241162 LR 0.000500 Time 0.019767 -2023-02-13 18:14:31,893 - Epoch: [127][ 1050/ 1207] Overall Loss 0.241099 Objective Loss 0.241099 LR 0.000500 Time 0.019761 -2023-02-13 18:14:32,084 - Epoch: [127][ 1060/ 1207] Overall Loss 0.240868 Objective Loss 0.240868 LR 0.000500 Time 0.019754 -2023-02-13 18:14:32,274 - Epoch: [127][ 1070/ 1207] Overall Loss 0.240509 Objective Loss 0.240509 LR 0.000500 Time 0.019747 -2023-02-13 18:14:32,464 - Epoch: [127][ 1080/ 1207] Overall Loss 0.240727 Objective Loss 0.240727 LR 0.000500 Time 0.019740 -2023-02-13 18:14:32,654 - Epoch: [127][ 1090/ 1207] Overall Loss 0.240973 Objective Loss 0.240973 LR 0.000500 Time 0.019733 -2023-02-13 18:14:32,845 - Epoch: [127][ 1100/ 1207] Overall Loss 0.240896 Objective Loss 0.240896 LR 0.000500 Time 0.019726 -2023-02-13 18:14:33,034 - Epoch: [127][ 1110/ 1207] Overall Loss 0.241006 Objective Loss 0.241006 LR 0.000500 Time 0.019719 -2023-02-13 18:14:33,224 - Epoch: [127][ 1120/ 1207] Overall Loss 0.241222 Objective Loss 0.241222 LR 0.000500 Time 0.019712 -2023-02-13 18:14:33,414 - Epoch: [127][ 1130/ 1207] Overall Loss 0.241402 Objective Loss 0.241402 LR 0.000500 Time 0.019705 -2023-02-13 18:14:33,605 - Epoch: [127][ 1140/ 1207] Overall Loss 0.241405 Objective Loss 0.241405 LR 0.000500 Time 0.019700 -2023-02-13 18:14:33,795 - Epoch: [127][ 1150/ 1207] Overall Loss 0.241295 Objective Loss 0.241295 LR 0.000500 Time 0.019693 -2023-02-13 18:14:33,987 - Epoch: [127][ 1160/ 1207] Overall Loss 0.241284 Objective Loss 0.241284 LR 0.000500 Time 0.019689 -2023-02-13 18:14:34,177 - Epoch: [127][ 1170/ 1207] Overall Loss 0.241291 Objective Loss 0.241291 LR 0.000500 Time 0.019683 -2023-02-13 18:14:34,368 - Epoch: [127][ 1180/ 1207] Overall Loss 0.241203 Objective Loss 0.241203 LR 0.000500 Time 0.019677 -2023-02-13 18:14:34,558 - Epoch: [127][ 1190/ 1207] Overall Loss 0.241194 Objective Loss 0.241194 LR 0.000500 Time 0.019671 -2023-02-13 18:14:34,800 - Epoch: [127][ 1200/ 1207] Overall Loss 0.241072 Objective Loss 0.241072 LR 0.000500 Time 0.019709 -2023-02-13 18:14:34,917 - Epoch: [127][ 1207/ 1207] Overall Loss 0.241220 Objective Loss 0.241220 Top1 86.890244 Top5 98.780488 LR 0.000500 Time 0.019691 -2023-02-13 18:14:34,988 - --- validate (epoch=127)----------- -2023-02-13 18:14:34,989 - 34311 samples (256 per mini-batch) -2023-02-13 18:14:35,383 - Epoch: [127][ 10/ 135] Loss 0.291022 Top1 85.742188 Top5 97.968750 -2023-02-13 18:14:35,510 - Epoch: [127][ 20/ 135] Loss 0.290403 Top1 85.253906 Top5 98.027344 -2023-02-13 18:14:35,634 - Epoch: [127][ 30/ 135] Loss 0.305410 Top1 84.895833 Top5 97.968750 -2023-02-13 18:14:35,760 - Epoch: [127][ 40/ 135] Loss 0.300770 Top1 85.126953 Top5 97.978516 -2023-02-13 18:14:35,884 - Epoch: [127][ 50/ 135] Loss 0.300904 Top1 85.078125 Top5 97.906250 -2023-02-13 18:14:36,010 - Epoch: [127][ 60/ 135] Loss 0.301501 Top1 85.071615 Top5 97.910156 -2023-02-13 18:14:36,137 - Epoch: [127][ 70/ 135] Loss 0.307410 Top1 84.916295 Top5 97.901786 -2023-02-13 18:14:36,262 - Epoch: [127][ 80/ 135] Loss 0.306460 Top1 84.892578 Top5 97.924805 -2023-02-13 18:14:36,389 - Epoch: [127][ 90/ 135] Loss 0.306300 Top1 84.830729 Top5 97.934028 -2023-02-13 18:14:36,517 - Epoch: [127][ 100/ 135] Loss 0.305802 Top1 84.800781 Top5 97.949219 -2023-02-13 18:14:36,644 - Epoch: [127][ 110/ 135] Loss 0.304140 Top1 84.712358 Top5 97.965199 -2023-02-13 18:14:36,774 - Epoch: [127][ 120/ 135] Loss 0.306365 Top1 84.622396 Top5 97.965495 -2023-02-13 18:14:36,905 - Epoch: [127][ 130/ 135] Loss 0.305541 Top1 84.609375 Top5 97.962740 -2023-02-13 18:14:36,952 - Epoch: [127][ 135/ 135] Loss 0.308726 Top1 84.617178 Top5 97.951094 -2023-02-13 18:14:37,023 - ==> Top1: 84.617 Top5: 97.951 Loss: 0.309 - -2023-02-13 18:14:37,024 - ==> Confusion: -[[ 859 5 4 1 10 2 0 1 1 53 0 3 1 5 6 2 3 2 0 1 8] - [ 1 931 2 3 12 41 1 16 2 1 2 1 2 0 1 2 1 0 0 1 13] - [ 8 2 962 14 4 1 14 12 1 2 1 0 2 4 3 5 2 4 5 3 9] - [ 5 1 27 907 4 6 0 3 2 1 12 0 8 1 15 2 2 5 10 0 5] - [ 13 8 0 0 988 11 0 2 1 3 0 7 2 5 8 5 6 2 0 1 4] - [ 3 14 1 4 6 980 2 15 0 4 1 5 1 17 0 1 2 2 1 4 7] - [ 3 6 18 1 0 6 1036 4 0 2 2 1 2 1 0 2 1 3 1 8 2] - [ 0 8 13 2 2 26 1 933 1 1 0 7 0 1 0 0 0 2 11 10 6] - [ 21 2 0 2 1 1 1 2 879 49 8 1 0 8 20 3 2 1 4 1 3] - [ 69 1 3 0 5 3 0 2 20 885 0 1 0 14 3 0 1 1 0 1 3] - [ 1 0 5 7 1 4 3 6 12 2 967 3 1 18 4 0 1 0 5 1 10] - [ 0 3 0 0 4 7 1 6 1 1 0 928 21 9 0 5 2 7 1 7 2] - [ 1 1 1 9 1 3 0 1 1 1 0 41 860 1 4 8 3 12 3 1 7] - [ 6 2 1 0 8 8 0 2 7 21 6 6 3 938 4 2 2 0 0 1 7] - [ 7 2 2 17 4 3 0 1 15 10 3 0 3 3 994 0 1 9 8 1 9] - [ 2 5 5 0 6 1 4 0 0 0 0 6 6 4 3 979 6 11 0 3 5] - [ 4 8 2 1 7 2 0 0 1 2 0 1 2 1 1 13 1003 1 0 5 7] - [ 8 3 0 4 0 1 1 0 0 0 1 13 12 1 0 13 0 986 0 1 7] - [ 3 2 6 11 1 2 1 36 2 1 5 1 4 0 12 1 1 1 987 4 5] - [ 0 4 1 0 3 7 3 9 0 0 1 13 1 8 0 6 3 4 0 1079 6] - [ 167 221 273 112 131 248 91 192 68 117 133 131 290 279 163 107 284 102 127 246 9952]] - -2023-02-13 18:14:37,026 - ==> Best [Top1: 84.617 Top5: 97.951 Sparsity:0.00 Params: 148928 on epoch: 127] -2023-02-13 18:14:37,026 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:14:37,032 - - -2023-02-13 18:14:37,032 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:14:37,925 - Epoch: [128][ 10/ 1207] Overall Loss 0.221872 Objective Loss 0.221872 LR 0.000500 Time 0.089239 -2023-02-13 18:14:38,127 - Epoch: [128][ 20/ 1207] Overall Loss 0.219925 Objective Loss 0.219925 LR 0.000500 Time 0.054661 -2023-02-13 18:14:38,320 - Epoch: [128][ 30/ 1207] Overall Loss 0.227707 Objective Loss 0.227707 LR 0.000500 Time 0.042889 -2023-02-13 18:14:38,516 - Epoch: [128][ 40/ 1207] Overall Loss 0.233025 Objective Loss 0.233025 LR 0.000500 Time 0.037038 -2023-02-13 18:14:38,709 - Epoch: [128][ 50/ 1207] Overall Loss 0.232684 Objective Loss 0.232684 LR 0.000500 Time 0.033483 -2023-02-13 18:14:38,905 - Epoch: [128][ 60/ 1207] Overall Loss 0.231304 Objective Loss 0.231304 LR 0.000500 Time 0.031161 -2023-02-13 18:14:39,098 - Epoch: [128][ 70/ 1207] Overall Loss 0.229446 Objective Loss 0.229446 LR 0.000500 Time 0.029467 -2023-02-13 18:14:39,293 - Epoch: [128][ 80/ 1207] Overall Loss 0.230551 Objective Loss 0.230551 LR 0.000500 Time 0.028220 -2023-02-13 18:14:39,486 - Epoch: [128][ 90/ 1207] Overall Loss 0.231596 Objective Loss 0.231596 LR 0.000500 Time 0.027226 -2023-02-13 18:14:39,682 - Epoch: [128][ 100/ 1207] Overall Loss 0.231672 Objective Loss 0.231672 LR 0.000500 Time 0.026458 -2023-02-13 18:14:39,876 - Epoch: [128][ 110/ 1207] Overall Loss 0.232871 Objective Loss 0.232871 LR 0.000500 Time 0.025808 -2023-02-13 18:14:40,072 - Epoch: [128][ 120/ 1207] Overall Loss 0.235446 Objective Loss 0.235446 LR 0.000500 Time 0.025292 -2023-02-13 18:14:40,265 - Epoch: [128][ 130/ 1207] Overall Loss 0.235806 Objective Loss 0.235806 LR 0.000500 Time 0.024828 -2023-02-13 18:14:40,461 - Epoch: [128][ 140/ 1207] Overall Loss 0.234339 Objective Loss 0.234339 LR 0.000500 Time 0.024451 -2023-02-13 18:14:40,654 - Epoch: [128][ 150/ 1207] Overall Loss 0.233010 Objective Loss 0.233010 LR 0.000500 Time 0.024105 -2023-02-13 18:14:40,851 - Epoch: [128][ 160/ 1207] Overall Loss 0.231108 Objective Loss 0.231108 LR 0.000500 Time 0.023824 -2023-02-13 18:14:41,044 - Epoch: [128][ 170/ 1207] Overall Loss 0.230864 Objective Loss 0.230864 LR 0.000500 Time 0.023560 -2023-02-13 18:14:41,240 - Epoch: [128][ 180/ 1207] Overall Loss 0.229919 Objective Loss 0.229919 LR 0.000500 Time 0.023339 -2023-02-13 18:14:41,433 - Epoch: [128][ 190/ 1207] Overall Loss 0.231322 Objective Loss 0.231322 LR 0.000500 Time 0.023124 -2023-02-13 18:14:41,630 - Epoch: [128][ 200/ 1207] Overall Loss 0.231339 Objective Loss 0.231339 LR 0.000500 Time 0.022948 -2023-02-13 18:14:41,823 - Epoch: [128][ 210/ 1207] Overall Loss 0.231771 Objective Loss 0.231771 LR 0.000500 Time 0.022773 -2023-02-13 18:14:42,020 - Epoch: [128][ 220/ 1207] Overall Loss 0.231841 Objective Loss 0.231841 LR 0.000500 Time 0.022632 -2023-02-13 18:14:42,213 - Epoch: [128][ 230/ 1207] Overall Loss 0.233139 Objective Loss 0.233139 LR 0.000500 Time 0.022485 -2023-02-13 18:14:42,409 - Epoch: [128][ 240/ 1207] Overall Loss 0.234008 Objective Loss 0.234008 LR 0.000500 Time 0.022362 -2023-02-13 18:14:42,603 - Epoch: [128][ 250/ 1207] Overall Loss 0.233736 Objective Loss 0.233736 LR 0.000500 Time 0.022242 -2023-02-13 18:14:42,798 - Epoch: [128][ 260/ 1207] Overall Loss 0.234468 Objective Loss 0.234468 LR 0.000500 Time 0.022135 -2023-02-13 18:14:42,987 - Epoch: [128][ 270/ 1207] Overall Loss 0.233323 Objective Loss 0.233323 LR 0.000500 Time 0.022016 -2023-02-13 18:14:43,176 - Epoch: [128][ 280/ 1207] Overall Loss 0.233811 Objective Loss 0.233811 LR 0.000500 Time 0.021902 -2023-02-13 18:14:43,364 - Epoch: [128][ 290/ 1207] Overall Loss 0.234161 Objective Loss 0.234161 LR 0.000500 Time 0.021795 -2023-02-13 18:14:43,553 - Epoch: [128][ 300/ 1207] Overall Loss 0.233665 Objective Loss 0.233665 LR 0.000500 Time 0.021696 -2023-02-13 18:14:43,741 - Epoch: [128][ 310/ 1207] Overall Loss 0.233387 Objective Loss 0.233387 LR 0.000500 Time 0.021602 -2023-02-13 18:14:43,931 - Epoch: [128][ 320/ 1207] Overall Loss 0.233516 Objective Loss 0.233516 LR 0.000500 Time 0.021519 -2023-02-13 18:14:44,119 - Epoch: [128][ 330/ 1207] Overall Loss 0.233492 Objective Loss 0.233492 LR 0.000500 Time 0.021435 -2023-02-13 18:14:44,308 - Epoch: [128][ 340/ 1207] Overall Loss 0.233935 Objective Loss 0.233935 LR 0.000500 Time 0.021360 -2023-02-13 18:14:44,497 - Epoch: [128][ 350/ 1207] Overall Loss 0.233935 Objective Loss 0.233935 LR 0.000500 Time 0.021288 -2023-02-13 18:14:44,685 - Epoch: [128][ 360/ 1207] Overall Loss 0.234389 Objective Loss 0.234389 LR 0.000500 Time 0.021219 -2023-02-13 18:14:44,874 - Epoch: [128][ 370/ 1207] Overall Loss 0.234742 Objective Loss 0.234742 LR 0.000500 Time 0.021156 -2023-02-13 18:14:45,063 - Epoch: [128][ 380/ 1207] Overall Loss 0.235546 Objective Loss 0.235546 LR 0.000500 Time 0.021096 -2023-02-13 18:14:45,252 - Epoch: [128][ 390/ 1207] Overall Loss 0.236085 Objective Loss 0.236085 LR 0.000500 Time 0.021038 -2023-02-13 18:14:45,441 - Epoch: [128][ 400/ 1207] Overall Loss 0.235662 Objective Loss 0.235662 LR 0.000500 Time 0.020982 -2023-02-13 18:14:45,630 - Epoch: [128][ 410/ 1207] Overall Loss 0.235687 Objective Loss 0.235687 LR 0.000500 Time 0.020930 -2023-02-13 18:14:45,819 - Epoch: [128][ 420/ 1207] Overall Loss 0.235825 Objective Loss 0.235825 LR 0.000500 Time 0.020881 -2023-02-13 18:14:46,010 - Epoch: [128][ 430/ 1207] Overall Loss 0.236408 Objective Loss 0.236408 LR 0.000500 Time 0.020839 -2023-02-13 18:14:46,199 - Epoch: [128][ 440/ 1207] Overall Loss 0.236848 Objective Loss 0.236848 LR 0.000500 Time 0.020794 -2023-02-13 18:14:46,387 - Epoch: [128][ 450/ 1207] Overall Loss 0.236960 Objective Loss 0.236960 LR 0.000500 Time 0.020750 -2023-02-13 18:14:46,577 - Epoch: [128][ 460/ 1207] Overall Loss 0.237171 Objective Loss 0.237171 LR 0.000500 Time 0.020710 -2023-02-13 18:14:46,766 - Epoch: [128][ 470/ 1207] Overall Loss 0.237177 Objective Loss 0.237177 LR 0.000500 Time 0.020670 -2023-02-13 18:14:46,955 - Epoch: [128][ 480/ 1207] Overall Loss 0.237101 Objective Loss 0.237101 LR 0.000500 Time 0.020634 -2023-02-13 18:14:47,144 - Epoch: [128][ 490/ 1207] Overall Loss 0.237292 Objective Loss 0.237292 LR 0.000500 Time 0.020597 -2023-02-13 18:14:47,332 - Epoch: [128][ 500/ 1207] Overall Loss 0.237677 Objective Loss 0.237677 LR 0.000500 Time 0.020560 -2023-02-13 18:14:47,520 - Epoch: [128][ 510/ 1207] Overall Loss 0.237410 Objective Loss 0.237410 LR 0.000500 Time 0.020526 -2023-02-13 18:14:47,709 - Epoch: [128][ 520/ 1207] Overall Loss 0.236966 Objective Loss 0.236966 LR 0.000500 Time 0.020493 -2023-02-13 18:14:47,898 - Epoch: [128][ 530/ 1207] Overall Loss 0.236759 Objective Loss 0.236759 LR 0.000500 Time 0.020462 -2023-02-13 18:14:48,087 - Epoch: [128][ 540/ 1207] Overall Loss 0.236434 Objective Loss 0.236434 LR 0.000500 Time 0.020433 -2023-02-13 18:14:48,276 - Epoch: [128][ 550/ 1207] Overall Loss 0.236725 Objective Loss 0.236725 LR 0.000500 Time 0.020404 -2023-02-13 18:14:48,465 - Epoch: [128][ 560/ 1207] Overall Loss 0.236851 Objective Loss 0.236851 LR 0.000500 Time 0.020376 -2023-02-13 18:14:48,653 - Epoch: [128][ 570/ 1207] Overall Loss 0.237041 Objective Loss 0.237041 LR 0.000500 Time 0.020349 -2023-02-13 18:14:48,842 - Epoch: [128][ 580/ 1207] Overall Loss 0.237603 Objective Loss 0.237603 LR 0.000500 Time 0.020324 -2023-02-13 18:14:49,033 - Epoch: [128][ 590/ 1207] Overall Loss 0.237969 Objective Loss 0.237969 LR 0.000500 Time 0.020301 -2023-02-13 18:14:49,221 - Epoch: [128][ 600/ 1207] Overall Loss 0.237715 Objective Loss 0.237715 LR 0.000500 Time 0.020276 -2023-02-13 18:14:49,409 - Epoch: [128][ 610/ 1207] Overall Loss 0.238232 Objective Loss 0.238232 LR 0.000500 Time 0.020251 -2023-02-13 18:14:49,598 - Epoch: [128][ 620/ 1207] Overall Loss 0.238349 Objective Loss 0.238349 LR 0.000500 Time 0.020228 -2023-02-13 18:14:49,787 - Epoch: [128][ 630/ 1207] Overall Loss 0.238678 Objective Loss 0.238678 LR 0.000500 Time 0.020207 -2023-02-13 18:14:49,977 - Epoch: [128][ 640/ 1207] Overall Loss 0.238490 Objective Loss 0.238490 LR 0.000500 Time 0.020187 -2023-02-13 18:14:50,166 - Epoch: [128][ 650/ 1207] Overall Loss 0.238689 Objective Loss 0.238689 LR 0.000500 Time 0.020167 -2023-02-13 18:14:50,355 - Epoch: [128][ 660/ 1207] Overall Loss 0.238842 Objective Loss 0.238842 LR 0.000500 Time 0.020148 -2023-02-13 18:14:50,544 - Epoch: [128][ 670/ 1207] Overall Loss 0.238809 Objective Loss 0.238809 LR 0.000500 Time 0.020129 -2023-02-13 18:14:50,733 - Epoch: [128][ 680/ 1207] Overall Loss 0.239022 Objective Loss 0.239022 LR 0.000500 Time 0.020110 -2023-02-13 18:14:50,924 - Epoch: [128][ 690/ 1207] Overall Loss 0.239314 Objective Loss 0.239314 LR 0.000500 Time 0.020094 -2023-02-13 18:14:51,113 - Epoch: [128][ 700/ 1207] Overall Loss 0.239045 Objective Loss 0.239045 LR 0.000500 Time 0.020077 -2023-02-13 18:14:51,302 - Epoch: [128][ 710/ 1207] Overall Loss 0.239278 Objective Loss 0.239278 LR 0.000500 Time 0.020060 -2023-02-13 18:14:51,492 - Epoch: [128][ 720/ 1207] Overall Loss 0.239532 Objective Loss 0.239532 LR 0.000500 Time 0.020044 -2023-02-13 18:14:51,681 - Epoch: [128][ 730/ 1207] Overall Loss 0.239717 Objective Loss 0.239717 LR 0.000500 Time 0.020028 -2023-02-13 18:14:51,869 - Epoch: [128][ 740/ 1207] Overall Loss 0.239806 Objective Loss 0.239806 LR 0.000500 Time 0.020011 -2023-02-13 18:14:52,059 - Epoch: [128][ 750/ 1207] Overall Loss 0.239616 Objective Loss 0.239616 LR 0.000500 Time 0.019997 -2023-02-13 18:14:52,248 - Epoch: [128][ 760/ 1207] Overall Loss 0.239465 Objective Loss 0.239465 LR 0.000500 Time 0.019982 -2023-02-13 18:14:52,437 - Epoch: [128][ 770/ 1207] Overall Loss 0.239515 Objective Loss 0.239515 LR 0.000500 Time 0.019968 -2023-02-13 18:14:52,626 - Epoch: [128][ 780/ 1207] Overall Loss 0.239396 Objective Loss 0.239396 LR 0.000500 Time 0.019954 -2023-02-13 18:14:52,816 - Epoch: [128][ 790/ 1207] Overall Loss 0.239395 Objective Loss 0.239395 LR 0.000500 Time 0.019941 -2023-02-13 18:14:53,006 - Epoch: [128][ 800/ 1207] Overall Loss 0.239644 Objective Loss 0.239644 LR 0.000500 Time 0.019929 -2023-02-13 18:14:53,195 - Epoch: [128][ 810/ 1207] Overall Loss 0.239828 Objective Loss 0.239828 LR 0.000500 Time 0.019916 -2023-02-13 18:14:53,384 - Epoch: [128][ 820/ 1207] Overall Loss 0.239919 Objective Loss 0.239919 LR 0.000500 Time 0.019903 -2023-02-13 18:14:53,573 - Epoch: [128][ 830/ 1207] Overall Loss 0.240011 Objective Loss 0.240011 LR 0.000500 Time 0.019891 -2023-02-13 18:14:53,763 - Epoch: [128][ 840/ 1207] Overall Loss 0.240068 Objective Loss 0.240068 LR 0.000500 Time 0.019879 -2023-02-13 18:14:53,953 - Epoch: [128][ 850/ 1207] Overall Loss 0.240544 Objective Loss 0.240544 LR 0.000500 Time 0.019869 -2023-02-13 18:14:54,142 - Epoch: [128][ 860/ 1207] Overall Loss 0.240454 Objective Loss 0.240454 LR 0.000500 Time 0.019857 -2023-02-13 18:14:54,332 - Epoch: [128][ 870/ 1207] Overall Loss 0.240687 Objective Loss 0.240687 LR 0.000500 Time 0.019846 -2023-02-13 18:14:54,522 - Epoch: [128][ 880/ 1207] Overall Loss 0.240791 Objective Loss 0.240791 LR 0.000500 Time 0.019836 -2023-02-13 18:14:54,711 - Epoch: [128][ 890/ 1207] Overall Loss 0.240768 Objective Loss 0.240768 LR 0.000500 Time 0.019825 -2023-02-13 18:14:54,901 - Epoch: [128][ 900/ 1207] Overall Loss 0.241027 Objective Loss 0.241027 LR 0.000500 Time 0.019815 -2023-02-13 18:14:55,090 - Epoch: [128][ 910/ 1207] Overall Loss 0.241240 Objective Loss 0.241240 LR 0.000500 Time 0.019806 -2023-02-13 18:14:55,279 - Epoch: [128][ 920/ 1207] Overall Loss 0.241289 Objective Loss 0.241289 LR 0.000500 Time 0.019795 -2023-02-13 18:14:55,468 - Epoch: [128][ 930/ 1207] Overall Loss 0.241230 Objective Loss 0.241230 LR 0.000500 Time 0.019785 -2023-02-13 18:14:55,658 - Epoch: [128][ 940/ 1207] Overall Loss 0.241578 Objective Loss 0.241578 LR 0.000500 Time 0.019776 -2023-02-13 18:14:55,847 - Epoch: [128][ 950/ 1207] Overall Loss 0.241463 Objective Loss 0.241463 LR 0.000500 Time 0.019767 -2023-02-13 18:14:56,038 - Epoch: [128][ 960/ 1207] Overall Loss 0.241612 Objective Loss 0.241612 LR 0.000500 Time 0.019759 -2023-02-13 18:14:56,227 - Epoch: [128][ 970/ 1207] Overall Loss 0.241667 Objective Loss 0.241667 LR 0.000500 Time 0.019750 -2023-02-13 18:14:56,417 - Epoch: [128][ 980/ 1207] Overall Loss 0.241913 Objective Loss 0.241913 LR 0.000500 Time 0.019742 -2023-02-13 18:14:56,606 - Epoch: [128][ 990/ 1207] Overall Loss 0.242008 Objective Loss 0.242008 LR 0.000500 Time 0.019733 -2023-02-13 18:14:56,796 - Epoch: [128][ 1000/ 1207] Overall Loss 0.242029 Objective Loss 0.242029 LR 0.000500 Time 0.019725 -2023-02-13 18:14:56,986 - Epoch: [128][ 1010/ 1207] Overall Loss 0.242313 Objective Loss 0.242313 LR 0.000500 Time 0.019718 -2023-02-13 18:14:57,176 - Epoch: [128][ 1020/ 1207] Overall Loss 0.242471 Objective Loss 0.242471 LR 0.000500 Time 0.019710 -2023-02-13 18:14:57,365 - Epoch: [128][ 1030/ 1207] Overall Loss 0.242607 Objective Loss 0.242607 LR 0.000500 Time 0.019702 -2023-02-13 18:14:57,555 - Epoch: [128][ 1040/ 1207] Overall Loss 0.243051 Objective Loss 0.243051 LR 0.000500 Time 0.019695 -2023-02-13 18:14:57,744 - Epoch: [128][ 1050/ 1207] Overall Loss 0.243104 Objective Loss 0.243104 LR 0.000500 Time 0.019687 -2023-02-13 18:14:57,935 - Epoch: [128][ 1060/ 1207] Overall Loss 0.243156 Objective Loss 0.243156 LR 0.000500 Time 0.019681 -2023-02-13 18:14:58,125 - Epoch: [128][ 1070/ 1207] Overall Loss 0.243205 Objective Loss 0.243205 LR 0.000500 Time 0.019674 -2023-02-13 18:14:58,314 - Epoch: [128][ 1080/ 1207] Overall Loss 0.243257 Objective Loss 0.243257 LR 0.000500 Time 0.019667 -2023-02-13 18:14:58,503 - Epoch: [128][ 1090/ 1207] Overall Loss 0.243081 Objective Loss 0.243081 LR 0.000500 Time 0.019660 -2023-02-13 18:14:58,693 - Epoch: [128][ 1100/ 1207] Overall Loss 0.243251 Objective Loss 0.243251 LR 0.000500 Time 0.019653 -2023-02-13 18:14:58,882 - Epoch: [128][ 1110/ 1207] Overall Loss 0.242850 Objective Loss 0.242850 LR 0.000500 Time 0.019646 -2023-02-13 18:14:59,073 - Epoch: [128][ 1120/ 1207] Overall Loss 0.242748 Objective Loss 0.242748 LR 0.000500 Time 0.019641 -2023-02-13 18:14:59,262 - Epoch: [128][ 1130/ 1207] Overall Loss 0.242794 Objective Loss 0.242794 LR 0.000500 Time 0.019634 -2023-02-13 18:14:59,451 - Epoch: [128][ 1140/ 1207] Overall Loss 0.242555 Objective Loss 0.242555 LR 0.000500 Time 0.019628 -2023-02-13 18:14:59,641 - Epoch: [128][ 1150/ 1207] Overall Loss 0.242450 Objective Loss 0.242450 LR 0.000500 Time 0.019622 -2023-02-13 18:14:59,831 - Epoch: [128][ 1160/ 1207] Overall Loss 0.242485 Objective Loss 0.242485 LR 0.000500 Time 0.019615 -2023-02-13 18:15:00,021 - Epoch: [128][ 1170/ 1207] Overall Loss 0.242549 Objective Loss 0.242549 LR 0.000500 Time 0.019610 -2023-02-13 18:15:00,210 - Epoch: [128][ 1180/ 1207] Overall Loss 0.242533 Objective Loss 0.242533 LR 0.000500 Time 0.019604 -2023-02-13 18:15:00,399 - Epoch: [128][ 1190/ 1207] Overall Loss 0.242334 Objective Loss 0.242334 LR 0.000500 Time 0.019598 -2023-02-13 18:15:00,639 - Epoch: [128][ 1200/ 1207] Overall Loss 0.242297 Objective Loss 0.242297 LR 0.000500 Time 0.019634 -2023-02-13 18:15:00,753 - Epoch: [128][ 1207/ 1207] Overall Loss 0.242169 Objective Loss 0.242169 Top1 89.939024 Top5 98.475610 LR 0.000500 Time 0.019615 -2023-02-13 18:15:00,825 - --- validate (epoch=128)----------- -2023-02-13 18:15:00,826 - 34311 samples (256 per mini-batch) -2023-02-13 18:15:01,232 - Epoch: [128][ 10/ 135] Loss 0.318162 Top1 84.335938 Top5 97.773438 -2023-02-13 18:15:01,363 - Epoch: [128][ 20/ 135] Loss 0.312557 Top1 84.707031 Top5 97.812500 -2023-02-13 18:15:01,492 - Epoch: [128][ 30/ 135] Loss 0.318515 Top1 84.947917 Top5 97.786458 -2023-02-13 18:15:01,621 - Epoch: [128][ 40/ 135] Loss 0.321227 Top1 84.912109 Top5 97.763672 -2023-02-13 18:15:01,750 - Epoch: [128][ 50/ 135] Loss 0.316667 Top1 85.031250 Top5 97.867188 -2023-02-13 18:15:01,878 - Epoch: [128][ 60/ 135] Loss 0.311055 Top1 85.299479 Top5 97.903646 -2023-02-13 18:15:02,006 - Epoch: [128][ 70/ 135] Loss 0.308975 Top1 85.167411 Top5 97.896205 -2023-02-13 18:15:02,132 - Epoch: [128][ 80/ 135] Loss 0.310565 Top1 85.258789 Top5 97.915039 -2023-02-13 18:15:02,253 - Epoch: [128][ 90/ 135] Loss 0.308757 Top1 85.329861 Top5 97.899306 -2023-02-13 18:15:02,376 - Epoch: [128][ 100/ 135] Loss 0.307186 Top1 85.277344 Top5 97.894531 -2023-02-13 18:15:02,502 - Epoch: [128][ 110/ 135] Loss 0.308017 Top1 85.152699 Top5 97.872869 -2023-02-13 18:15:02,628 - Epoch: [128][ 120/ 135] Loss 0.307745 Top1 85.169271 Top5 97.848307 -2023-02-13 18:15:02,760 - Epoch: [128][ 130/ 135] Loss 0.306003 Top1 85.222356 Top5 97.887620 -2023-02-13 18:15:02,807 - Epoch: [128][ 135/ 135] Loss 0.306041 Top1 85.159278 Top5 97.863659 -2023-02-13 18:15:02,875 - ==> Top1: 85.159 Top5: 97.864 Loss: 0.306 - -2023-02-13 18:15:02,876 - ==> Confusion: -[[ 857 6 2 0 10 1 0 1 3 49 1 2 1 5 7 2 1 3 1 3 12] - [ 2 971 2 2 5 11 5 9 3 2 1 0 2 0 0 1 4 1 2 2 8] - [ 5 4 931 14 4 1 25 14 1 1 4 1 2 6 5 12 1 4 9 2 12] - [ 3 1 8 913 3 3 4 2 4 2 11 0 5 1 17 1 4 4 20 0 10] - [ 11 14 0 0 996 7 2 1 4 2 1 4 1 2 3 5 7 0 0 2 4] - [ 2 28 0 5 5 956 4 21 2 5 2 8 2 12 1 3 5 1 0 3 5] - [ 4 2 16 1 0 4 1037 3 0 1 5 1 1 1 0 5 1 4 1 7 5] - [ 1 15 7 0 5 28 6 904 1 1 2 6 4 0 0 0 0 1 30 3 10] - [ 16 3 3 2 2 0 0 2 909 36 9 2 0 6 11 2 0 1 4 0 1] - [ 76 1 3 0 8 2 0 2 28 864 0 1 0 12 3 2 1 2 0 1 6] - [ 3 2 1 7 0 3 1 3 17 0 993 2 1 7 2 0 0 0 5 1 3] - [ 1 5 1 0 3 8 1 4 3 0 0 916 23 7 1 8 2 11 3 5 3] - [ 0 1 0 4 3 3 0 0 1 1 0 25 875 0 0 7 3 21 3 0 12] - [ 3 3 1 0 7 8 0 2 14 16 11 4 3 923 3 5 8 1 1 2 9] - [ 7 3 1 21 6 5 0 1 21 7 5 2 2 3 980 2 0 6 9 0 11] - [ 4 3 3 0 7 0 3 0 1 0 0 6 8 3 0 974 8 13 0 5 8] - [ 5 4 0 1 6 1 0 0 1 0 0 3 2 2 2 13 1000 2 2 5 12] - [ 2 2 0 4 1 2 1 0 1 2 1 7 11 0 1 14 0 997 0 2 3] - [ 4 7 5 10 4 1 0 15 2 0 6 2 5 0 9 1 1 1 1009 1 3] - [ 1 7 3 0 1 3 7 10 1 0 1 18 3 5 0 5 6 4 1 1063 9] - [ 137 291 156 122 142 151 73 132 117 88 226 131 273 243 135 114 213 107 212 220 10151]] - -2023-02-13 18:15:02,878 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:15:02,878 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:15:02,884 - - -2023-02-13 18:15:02,884 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:15:03,771 - Epoch: [129][ 10/ 1207] Overall Loss 0.247479 Objective Loss 0.247479 LR 0.000500 Time 0.088613 -2023-02-13 18:15:03,964 - Epoch: [129][ 20/ 1207] Overall Loss 0.260197 Objective Loss 0.260197 LR 0.000500 Time 0.053921 -2023-02-13 18:15:04,158 - Epoch: [129][ 30/ 1207] Overall Loss 0.243523 Objective Loss 0.243523 LR 0.000500 Time 0.042406 -2023-02-13 18:15:04,349 - Epoch: [129][ 40/ 1207] Overall Loss 0.242943 Objective Loss 0.242943 LR 0.000500 Time 0.036573 -2023-02-13 18:15:04,543 - Epoch: [129][ 50/ 1207] Overall Loss 0.244234 Objective Loss 0.244234 LR 0.000500 Time 0.033125 -2023-02-13 18:15:04,734 - Epoch: [129][ 60/ 1207] Overall Loss 0.244524 Objective Loss 0.244524 LR 0.000500 Time 0.030787 -2023-02-13 18:15:04,928 - Epoch: [129][ 70/ 1207] Overall Loss 0.244048 Objective Loss 0.244048 LR 0.000500 Time 0.029156 -2023-02-13 18:15:05,120 - Epoch: [129][ 80/ 1207] Overall Loss 0.241548 Objective Loss 0.241548 LR 0.000500 Time 0.027901 -2023-02-13 18:15:05,315 - Epoch: [129][ 90/ 1207] Overall Loss 0.239203 Objective Loss 0.239203 LR 0.000500 Time 0.026962 -2023-02-13 18:15:05,505 - Epoch: [129][ 100/ 1207] Overall Loss 0.236912 Objective Loss 0.236912 LR 0.000500 Time 0.026169 -2023-02-13 18:15:05,700 - Epoch: [129][ 110/ 1207] Overall Loss 0.236861 Objective Loss 0.236861 LR 0.000500 Time 0.025555 -2023-02-13 18:15:05,891 - Epoch: [129][ 120/ 1207] Overall Loss 0.235663 Objective Loss 0.235663 LR 0.000500 Time 0.025019 -2023-02-13 18:15:06,086 - Epoch: [129][ 130/ 1207] Overall Loss 0.236524 Objective Loss 0.236524 LR 0.000500 Time 0.024592 -2023-02-13 18:15:06,277 - Epoch: [129][ 140/ 1207] Overall Loss 0.235960 Objective Loss 0.235960 LR 0.000500 Time 0.024193 -2023-02-13 18:15:06,471 - Epoch: [129][ 150/ 1207] Overall Loss 0.234846 Objective Loss 0.234846 LR 0.000500 Time 0.023872 -2023-02-13 18:15:06,662 - Epoch: [129][ 160/ 1207] Overall Loss 0.233233 Objective Loss 0.233233 LR 0.000500 Time 0.023571 -2023-02-13 18:15:06,856 - Epoch: [129][ 170/ 1207] Overall Loss 0.233758 Objective Loss 0.233758 LR 0.000500 Time 0.023327 -2023-02-13 18:15:07,049 - Epoch: [129][ 180/ 1207] Overall Loss 0.235859 Objective Loss 0.235859 LR 0.000500 Time 0.023099 -2023-02-13 18:15:07,243 - Epoch: [129][ 190/ 1207] Overall Loss 0.236312 Objective Loss 0.236312 LR 0.000500 Time 0.022902 -2023-02-13 18:15:07,434 - Epoch: [129][ 200/ 1207] Overall Loss 0.237599 Objective Loss 0.237599 LR 0.000500 Time 0.022709 -2023-02-13 18:15:07,629 - Epoch: [129][ 210/ 1207] Overall Loss 0.238528 Objective Loss 0.238528 LR 0.000500 Time 0.022554 -2023-02-13 18:15:07,819 - Epoch: [129][ 220/ 1207] Overall Loss 0.238678 Objective Loss 0.238678 LR 0.000500 Time 0.022394 -2023-02-13 18:15:08,015 - Epoch: [129][ 230/ 1207] Overall Loss 0.239160 Objective Loss 0.239160 LR 0.000500 Time 0.022268 -2023-02-13 18:15:08,206 - Epoch: [129][ 240/ 1207] Overall Loss 0.239673 Objective Loss 0.239673 LR 0.000500 Time 0.022136 -2023-02-13 18:15:08,401 - Epoch: [129][ 250/ 1207] Overall Loss 0.239599 Objective Loss 0.239599 LR 0.000500 Time 0.022028 -2023-02-13 18:15:08,593 - Epoch: [129][ 260/ 1207] Overall Loss 0.239032 Objective Loss 0.239032 LR 0.000500 Time 0.021917 -2023-02-13 18:15:08,787 - Epoch: [129][ 270/ 1207] Overall Loss 0.239915 Objective Loss 0.239915 LR 0.000500 Time 0.021826 -2023-02-13 18:15:08,979 - Epoch: [129][ 280/ 1207] Overall Loss 0.240779 Objective Loss 0.240779 LR 0.000500 Time 0.021730 -2023-02-13 18:15:09,174 - Epoch: [129][ 290/ 1207] Overall Loss 0.241800 Objective Loss 0.241800 LR 0.000500 Time 0.021651 -2023-02-13 18:15:09,365 - Epoch: [129][ 300/ 1207] Overall Loss 0.241732 Objective Loss 0.241732 LR 0.000500 Time 0.021565 -2023-02-13 18:15:09,560 - Epoch: [129][ 310/ 1207] Overall Loss 0.241041 Objective Loss 0.241041 LR 0.000500 Time 0.021496 -2023-02-13 18:15:09,751 - Epoch: [129][ 320/ 1207] Overall Loss 0.240757 Objective Loss 0.240757 LR 0.000500 Time 0.021421 -2023-02-13 18:15:09,947 - Epoch: [129][ 330/ 1207] Overall Loss 0.241116 Objective Loss 0.241116 LR 0.000500 Time 0.021364 -2023-02-13 18:15:10,138 - Epoch: [129][ 340/ 1207] Overall Loss 0.240539 Objective Loss 0.240539 LR 0.000500 Time 0.021298 -2023-02-13 18:15:10,333 - Epoch: [129][ 350/ 1207] Overall Loss 0.240162 Objective Loss 0.240162 LR 0.000500 Time 0.021244 -2023-02-13 18:15:10,524 - Epoch: [129][ 360/ 1207] Overall Loss 0.240341 Objective Loss 0.240341 LR 0.000500 Time 0.021184 -2023-02-13 18:15:10,719 - Epoch: [129][ 370/ 1207] Overall Loss 0.239874 Objective Loss 0.239874 LR 0.000500 Time 0.021136 -2023-02-13 18:15:10,911 - Epoch: [129][ 380/ 1207] Overall Loss 0.239920 Objective Loss 0.239920 LR 0.000500 Time 0.021084 -2023-02-13 18:15:11,106 - Epoch: [129][ 390/ 1207] Overall Loss 0.239215 Objective Loss 0.239215 LR 0.000500 Time 0.021044 -2023-02-13 18:15:11,297 - Epoch: [129][ 400/ 1207] Overall Loss 0.239706 Objective Loss 0.239706 LR 0.000500 Time 0.020995 -2023-02-13 18:15:11,492 - Epoch: [129][ 410/ 1207] Overall Loss 0.239993 Objective Loss 0.239993 LR 0.000500 Time 0.020956 -2023-02-13 18:15:11,683 - Epoch: [129][ 420/ 1207] Overall Loss 0.239942 Objective Loss 0.239942 LR 0.000500 Time 0.020911 -2023-02-13 18:15:11,878 - Epoch: [129][ 430/ 1207] Overall Loss 0.240184 Objective Loss 0.240184 LR 0.000500 Time 0.020879 -2023-02-13 18:15:12,071 - Epoch: [129][ 440/ 1207] Overall Loss 0.240641 Objective Loss 0.240641 LR 0.000500 Time 0.020842 -2023-02-13 18:15:12,266 - Epoch: [129][ 450/ 1207] Overall Loss 0.240080 Objective Loss 0.240080 LR 0.000500 Time 0.020811 -2023-02-13 18:15:12,457 - Epoch: [129][ 460/ 1207] Overall Loss 0.240051 Objective Loss 0.240051 LR 0.000500 Time 0.020773 -2023-02-13 18:15:12,652 - Epoch: [129][ 470/ 1207] Overall Loss 0.239295 Objective Loss 0.239295 LR 0.000500 Time 0.020746 -2023-02-13 18:15:12,844 - Epoch: [129][ 480/ 1207] Overall Loss 0.239352 Objective Loss 0.239352 LR 0.000500 Time 0.020711 -2023-02-13 18:15:13,039 - Epoch: [129][ 490/ 1207] Overall Loss 0.239940 Objective Loss 0.239940 LR 0.000500 Time 0.020687 -2023-02-13 18:15:13,231 - Epoch: [129][ 500/ 1207] Overall Loss 0.240451 Objective Loss 0.240451 LR 0.000500 Time 0.020655 -2023-02-13 18:15:13,426 - Epoch: [129][ 510/ 1207] Overall Loss 0.240197 Objective Loss 0.240197 LR 0.000500 Time 0.020632 -2023-02-13 18:15:13,617 - Epoch: [129][ 520/ 1207] Overall Loss 0.240443 Objective Loss 0.240443 LR 0.000500 Time 0.020602 -2023-02-13 18:15:13,812 - Epoch: [129][ 530/ 1207] Overall Loss 0.240158 Objective Loss 0.240158 LR 0.000500 Time 0.020581 -2023-02-13 18:15:14,005 - Epoch: [129][ 540/ 1207] Overall Loss 0.239959 Objective Loss 0.239959 LR 0.000500 Time 0.020557 -2023-02-13 18:15:14,202 - Epoch: [129][ 550/ 1207] Overall Loss 0.239764 Objective Loss 0.239764 LR 0.000500 Time 0.020541 -2023-02-13 18:15:14,396 - Epoch: [129][ 560/ 1207] Overall Loss 0.239710 Objective Loss 0.239710 LR 0.000500 Time 0.020520 -2023-02-13 18:15:14,592 - Epoch: [129][ 570/ 1207] Overall Loss 0.239610 Objective Loss 0.239610 LR 0.000500 Time 0.020503 -2023-02-13 18:15:14,786 - Epoch: [129][ 580/ 1207] Overall Loss 0.240040 Objective Loss 0.240040 LR 0.000500 Time 0.020483 -2023-02-13 18:15:14,983 - Epoch: [129][ 590/ 1207] Overall Loss 0.240112 Objective Loss 0.240112 LR 0.000500 Time 0.020469 -2023-02-13 18:15:15,176 - Epoch: [129][ 600/ 1207] Overall Loss 0.240642 Objective Loss 0.240642 LR 0.000500 Time 0.020449 -2023-02-13 18:15:15,372 - Epoch: [129][ 610/ 1207] Overall Loss 0.240444 Objective Loss 0.240444 LR 0.000500 Time 0.020434 -2023-02-13 18:15:15,566 - Epoch: [129][ 620/ 1207] Overall Loss 0.240763 Objective Loss 0.240763 LR 0.000500 Time 0.020417 -2023-02-13 18:15:15,762 - Epoch: [129][ 630/ 1207] Overall Loss 0.240522 Objective Loss 0.240522 LR 0.000500 Time 0.020404 -2023-02-13 18:15:15,957 - Epoch: [129][ 640/ 1207] Overall Loss 0.240455 Objective Loss 0.240455 LR 0.000500 Time 0.020389 -2023-02-13 18:15:16,149 - Epoch: [129][ 650/ 1207] Overall Loss 0.240256 Objective Loss 0.240256 LR 0.000500 Time 0.020370 -2023-02-13 18:15:16,337 - Epoch: [129][ 660/ 1207] Overall Loss 0.240487 Objective Loss 0.240487 LR 0.000500 Time 0.020346 -2023-02-13 18:15:16,525 - Epoch: [129][ 670/ 1207] Overall Loss 0.240746 Objective Loss 0.240746 LR 0.000500 Time 0.020322 -2023-02-13 18:15:16,713 - Epoch: [129][ 680/ 1207] Overall Loss 0.241056 Objective Loss 0.241056 LR 0.000500 Time 0.020299 -2023-02-13 18:15:16,901 - Epoch: [129][ 690/ 1207] Overall Loss 0.240920 Objective Loss 0.240920 LR 0.000500 Time 0.020277 -2023-02-13 18:15:17,089 - Epoch: [129][ 700/ 1207] Overall Loss 0.241034 Objective Loss 0.241034 LR 0.000500 Time 0.020256 -2023-02-13 18:15:17,278 - Epoch: [129][ 710/ 1207] Overall Loss 0.241025 Objective Loss 0.241025 LR 0.000500 Time 0.020235 -2023-02-13 18:15:17,466 - Epoch: [129][ 720/ 1207] Overall Loss 0.240836 Objective Loss 0.240836 LR 0.000500 Time 0.020215 -2023-02-13 18:15:17,654 - Epoch: [129][ 730/ 1207] Overall Loss 0.241003 Objective Loss 0.241003 LR 0.000500 Time 0.020195 -2023-02-13 18:15:17,842 - Epoch: [129][ 740/ 1207] Overall Loss 0.240801 Objective Loss 0.240801 LR 0.000500 Time 0.020176 -2023-02-13 18:15:18,030 - Epoch: [129][ 750/ 1207] Overall Loss 0.240645 Objective Loss 0.240645 LR 0.000500 Time 0.020158 -2023-02-13 18:15:18,218 - Epoch: [129][ 760/ 1207] Overall Loss 0.240721 Objective Loss 0.240721 LR 0.000500 Time 0.020139 -2023-02-13 18:15:18,407 - Epoch: [129][ 770/ 1207] Overall Loss 0.240349 Objective Loss 0.240349 LR 0.000500 Time 0.020122 -2023-02-13 18:15:18,595 - Epoch: [129][ 780/ 1207] Overall Loss 0.240283 Objective Loss 0.240283 LR 0.000500 Time 0.020104 -2023-02-13 18:15:18,782 - Epoch: [129][ 790/ 1207] Overall Loss 0.240461 Objective Loss 0.240461 LR 0.000500 Time 0.020087 -2023-02-13 18:15:18,971 - Epoch: [129][ 800/ 1207] Overall Loss 0.240376 Objective Loss 0.240376 LR 0.000500 Time 0.020071 -2023-02-13 18:15:19,160 - Epoch: [129][ 810/ 1207] Overall Loss 0.240485 Objective Loss 0.240485 LR 0.000500 Time 0.020056 -2023-02-13 18:15:19,348 - Epoch: [129][ 820/ 1207] Overall Loss 0.240675 Objective Loss 0.240675 LR 0.000500 Time 0.020040 -2023-02-13 18:15:19,536 - Epoch: [129][ 830/ 1207] Overall Loss 0.240511 Objective Loss 0.240511 LR 0.000500 Time 0.020025 -2023-02-13 18:15:19,724 - Epoch: [129][ 840/ 1207] Overall Loss 0.240939 Objective Loss 0.240939 LR 0.000500 Time 0.020010 -2023-02-13 18:15:19,912 - Epoch: [129][ 850/ 1207] Overall Loss 0.240862 Objective Loss 0.240862 LR 0.000500 Time 0.019996 -2023-02-13 18:15:20,101 - Epoch: [129][ 860/ 1207] Overall Loss 0.241029 Objective Loss 0.241029 LR 0.000500 Time 0.019983 -2023-02-13 18:15:20,290 - Epoch: [129][ 870/ 1207] Overall Loss 0.240964 Objective Loss 0.240964 LR 0.000500 Time 0.019969 -2023-02-13 18:15:20,478 - Epoch: [129][ 880/ 1207] Overall Loss 0.241085 Objective Loss 0.241085 LR 0.000500 Time 0.019956 -2023-02-13 18:15:20,666 - Epoch: [129][ 890/ 1207] Overall Loss 0.240894 Objective Loss 0.240894 LR 0.000500 Time 0.019943 -2023-02-13 18:15:20,855 - Epoch: [129][ 900/ 1207] Overall Loss 0.240698 Objective Loss 0.240698 LR 0.000500 Time 0.019931 -2023-02-13 18:15:21,045 - Epoch: [129][ 910/ 1207] Overall Loss 0.240632 Objective Loss 0.240632 LR 0.000500 Time 0.019920 -2023-02-13 18:15:21,233 - Epoch: [129][ 920/ 1207] Overall Loss 0.240530 Objective Loss 0.240530 LR 0.000500 Time 0.019907 -2023-02-13 18:15:21,420 - Epoch: [129][ 930/ 1207] Overall Loss 0.240415 Objective Loss 0.240415 LR 0.000500 Time 0.019894 -2023-02-13 18:15:21,608 - Epoch: [129][ 940/ 1207] Overall Loss 0.240573 Objective Loss 0.240573 LR 0.000500 Time 0.019882 -2023-02-13 18:15:21,797 - Epoch: [129][ 950/ 1207] Overall Loss 0.240234 Objective Loss 0.240234 LR 0.000500 Time 0.019871 -2023-02-13 18:15:21,985 - Epoch: [129][ 960/ 1207] Overall Loss 0.240496 Objective Loss 0.240496 LR 0.000500 Time 0.019860 -2023-02-13 18:15:22,173 - Epoch: [129][ 970/ 1207] Overall Loss 0.240325 Objective Loss 0.240325 LR 0.000500 Time 0.019848 -2023-02-13 18:15:22,361 - Epoch: [129][ 980/ 1207] Overall Loss 0.240379 Objective Loss 0.240379 LR 0.000500 Time 0.019837 -2023-02-13 18:15:22,549 - Epoch: [129][ 990/ 1207] Overall Loss 0.240392 Objective Loss 0.240392 LR 0.000500 Time 0.019826 -2023-02-13 18:15:22,737 - Epoch: [129][ 1000/ 1207] Overall Loss 0.240359 Objective Loss 0.240359 LR 0.000500 Time 0.019816 -2023-02-13 18:15:22,925 - Epoch: [129][ 1010/ 1207] Overall Loss 0.240135 Objective Loss 0.240135 LR 0.000500 Time 0.019806 -2023-02-13 18:15:23,114 - Epoch: [129][ 1020/ 1207] Overall Loss 0.240183 Objective Loss 0.240183 LR 0.000500 Time 0.019796 -2023-02-13 18:15:23,302 - Epoch: [129][ 1030/ 1207] Overall Loss 0.240476 Objective Loss 0.240476 LR 0.000500 Time 0.019786 -2023-02-13 18:15:23,490 - Epoch: [129][ 1040/ 1207] Overall Loss 0.240565 Objective Loss 0.240565 LR 0.000500 Time 0.019776 -2023-02-13 18:15:23,678 - Epoch: [129][ 1050/ 1207] Overall Loss 0.240642 Objective Loss 0.240642 LR 0.000500 Time 0.019767 -2023-02-13 18:15:23,866 - Epoch: [129][ 1060/ 1207] Overall Loss 0.240899 Objective Loss 0.240899 LR 0.000500 Time 0.019757 -2023-02-13 18:15:24,056 - Epoch: [129][ 1070/ 1207] Overall Loss 0.240999 Objective Loss 0.240999 LR 0.000500 Time 0.019749 -2023-02-13 18:15:24,244 - Epoch: [129][ 1080/ 1207] Overall Loss 0.240866 Objective Loss 0.240866 LR 0.000500 Time 0.019741 -2023-02-13 18:15:24,433 - Epoch: [129][ 1090/ 1207] Overall Loss 0.241048 Objective Loss 0.241048 LR 0.000500 Time 0.019732 -2023-02-13 18:15:24,621 - Epoch: [129][ 1100/ 1207] Overall Loss 0.240928 Objective Loss 0.240928 LR 0.000500 Time 0.019724 -2023-02-13 18:15:24,809 - Epoch: [129][ 1110/ 1207] Overall Loss 0.240794 Objective Loss 0.240794 LR 0.000500 Time 0.019715 -2023-02-13 18:15:24,998 - Epoch: [129][ 1120/ 1207] Overall Loss 0.240662 Objective Loss 0.240662 LR 0.000500 Time 0.019707 -2023-02-13 18:15:25,186 - Epoch: [129][ 1130/ 1207] Overall Loss 0.240718 Objective Loss 0.240718 LR 0.000500 Time 0.019699 -2023-02-13 18:15:25,374 - Epoch: [129][ 1140/ 1207] Overall Loss 0.240837 Objective Loss 0.240837 LR 0.000500 Time 0.019691 -2023-02-13 18:15:25,562 - Epoch: [129][ 1150/ 1207] Overall Loss 0.241101 Objective Loss 0.241101 LR 0.000500 Time 0.019683 -2023-02-13 18:15:25,751 - Epoch: [129][ 1160/ 1207] Overall Loss 0.241111 Objective Loss 0.241111 LR 0.000500 Time 0.019676 -2023-02-13 18:15:25,941 - Epoch: [129][ 1170/ 1207] Overall Loss 0.241481 Objective Loss 0.241481 LR 0.000500 Time 0.019669 -2023-02-13 18:15:26,130 - Epoch: [129][ 1180/ 1207] Overall Loss 0.241581 Objective Loss 0.241581 LR 0.000500 Time 0.019663 -2023-02-13 18:15:26,318 - Epoch: [129][ 1190/ 1207] Overall Loss 0.241711 Objective Loss 0.241711 LR 0.000500 Time 0.019656 -2023-02-13 18:15:26,558 - Epoch: [129][ 1200/ 1207] Overall Loss 0.241328 Objective Loss 0.241328 LR 0.000500 Time 0.019691 -2023-02-13 18:15:26,673 - Epoch: [129][ 1207/ 1207] Overall Loss 0.241237 Objective Loss 0.241237 Top1 85.975610 Top5 97.560976 LR 0.000500 Time 0.019672 -2023-02-13 18:15:26,745 - --- validate (epoch=129)----------- -2023-02-13 18:15:26,746 - 34311 samples (256 per mini-batch) -2023-02-13 18:15:27,150 - Epoch: [129][ 10/ 135] Loss 0.335638 Top1 84.570312 Top5 97.773438 -2023-02-13 18:15:27,274 - Epoch: [129][ 20/ 135] Loss 0.314947 Top1 84.667969 Top5 97.910156 -2023-02-13 18:15:27,401 - Epoch: [129][ 30/ 135] Loss 0.313393 Top1 84.583333 Top5 97.812500 -2023-02-13 18:15:27,528 - Epoch: [129][ 40/ 135] Loss 0.309305 Top1 84.785156 Top5 97.851562 -2023-02-13 18:15:27,656 - Epoch: [129][ 50/ 135] Loss 0.312817 Top1 84.617188 Top5 97.882812 -2023-02-13 18:15:27,787 - Epoch: [129][ 60/ 135] Loss 0.311331 Top1 84.798177 Top5 97.910156 -2023-02-13 18:15:27,917 - Epoch: [129][ 70/ 135] Loss 0.307105 Top1 84.899554 Top5 97.924107 -2023-02-13 18:15:28,041 - Epoch: [129][ 80/ 135] Loss 0.310159 Top1 84.755859 Top5 97.875977 -2023-02-13 18:15:28,167 - Epoch: [129][ 90/ 135] Loss 0.311577 Top1 84.704861 Top5 97.855903 -2023-02-13 18:15:28,297 - Epoch: [129][ 100/ 135] Loss 0.309994 Top1 84.714844 Top5 97.835938 -2023-02-13 18:15:28,426 - Epoch: [129][ 110/ 135] Loss 0.310423 Top1 84.723011 Top5 97.858665 -2023-02-13 18:15:28,555 - Epoch: [129][ 120/ 135] Loss 0.309160 Top1 84.726562 Top5 97.825521 -2023-02-13 18:15:28,688 - Epoch: [129][ 130/ 135] Loss 0.312021 Top1 84.660457 Top5 97.800481 -2023-02-13 18:15:28,736 - Epoch: [129][ 135/ 135] Loss 0.311457 Top1 84.704614 Top5 97.793710 -2023-02-13 18:15:28,804 - ==> Top1: 84.705 Top5: 97.794 Loss: 0.311 - -2023-02-13 18:15:28,805 - ==> Confusion: -[[ 835 5 5 0 13 3 0 1 3 67 0 3 1 3 5 3 2 4 2 3 9] - [ 1 930 1 2 17 30 2 16 4 2 0 2 2 1 0 2 4 0 7 0 10] - [ 8 4 950 6 2 1 19 12 0 1 4 0 4 8 5 5 3 4 8 4 10] - [ 3 2 17 918 2 5 2 1 2 3 11 0 5 1 18 1 4 6 7 1 7] - [ 14 9 0 1 1000 6 1 1 1 3 0 4 1 3 5 6 5 0 0 3 3] - [ 2 15 2 4 9 973 3 11 2 4 1 11 4 13 0 2 2 1 3 2 6] - [ 2 5 16 2 0 9 1032 4 0 1 4 2 5 1 0 3 1 2 2 4 4] - [ 0 4 5 2 7 30 5 929 0 1 1 5 3 1 0 1 0 1 17 7 5] - [ 15 2 1 2 1 1 0 2 891 53 6 2 0 11 11 2 0 3 4 1 1] - [ 55 0 1 3 5 2 0 1 30 884 0 0 0 17 1 3 2 2 1 0 5] - [ 1 1 1 5 2 2 2 2 26 2 966 5 1 15 2 0 1 2 8 0 7] - [ 1 2 2 0 4 7 0 5 0 0 1 929 17 5 0 9 4 9 1 8 1] - [ 1 0 0 8 1 5 0 1 1 1 0 32 871 0 1 8 1 14 2 0 12] - [ 3 2 1 0 10 10 0 1 11 16 3 4 1 942 5 6 2 1 1 0 5] - [ 5 1 2 21 4 4 0 3 16 10 2 1 1 1 996 2 1 7 7 0 8] - [ 5 3 6 0 11 3 3 0 1 0 0 7 7 1 1 962 9 12 0 9 6] - [ 2 6 0 1 9 1 0 1 2 1 0 0 3 0 1 8 1012 2 2 3 7] - [ 7 6 2 3 0 2 0 0 0 1 1 11 11 0 1 14 0 984 0 4 4] - [ 5 9 3 13 2 0 2 20 4 0 5 1 5 0 13 1 0 1 994 2 6] - [ 0 3 2 1 2 8 6 12 1 0 0 21 1 5 0 9 3 3 0 1062 9] - [ 144 219 207 123 174 234 92 167 90 115 157 123 286 331 155 99 199 86 185 245 10003]] - -2023-02-13 18:15:28,807 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:15:28,807 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:15:28,813 - - -2023-02-13 18:15:28,813 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:15:29,815 - Epoch: [130][ 10/ 1207] Overall Loss 0.222718 Objective Loss 0.222718 LR 0.000500 Time 0.100215 -2023-02-13 18:15:30,013 - Epoch: [130][ 20/ 1207] Overall Loss 0.228125 Objective Loss 0.228125 LR 0.000500 Time 0.059955 -2023-02-13 18:15:30,209 - Epoch: [130][ 30/ 1207] Overall Loss 0.232881 Objective Loss 0.232881 LR 0.000500 Time 0.046493 -2023-02-13 18:15:30,404 - Epoch: [130][ 40/ 1207] Overall Loss 0.241998 Objective Loss 0.241998 LR 0.000500 Time 0.039734 -2023-02-13 18:15:30,599 - Epoch: [130][ 50/ 1207] Overall Loss 0.242661 Objective Loss 0.242661 LR 0.000500 Time 0.035691 -2023-02-13 18:15:30,794 - Epoch: [130][ 60/ 1207] Overall Loss 0.240666 Objective Loss 0.240666 LR 0.000500 Time 0.032987 -2023-02-13 18:15:30,989 - Epoch: [130][ 70/ 1207] Overall Loss 0.236822 Objective Loss 0.236822 LR 0.000500 Time 0.031054 -2023-02-13 18:15:31,181 - Epoch: [130][ 80/ 1207] Overall Loss 0.236699 Objective Loss 0.236699 LR 0.000500 Time 0.029561 -2023-02-13 18:15:31,373 - Epoch: [130][ 90/ 1207] Overall Loss 0.235603 Objective Loss 0.235603 LR 0.000500 Time 0.028403 -2023-02-13 18:15:31,563 - Epoch: [130][ 100/ 1207] Overall Loss 0.235049 Objective Loss 0.235049 LR 0.000500 Time 0.027468 -2023-02-13 18:15:31,755 - Epoch: [130][ 110/ 1207] Overall Loss 0.233689 Objective Loss 0.233689 LR 0.000500 Time 0.026711 -2023-02-13 18:15:31,946 - Epoch: [130][ 120/ 1207] Overall Loss 0.236314 Objective Loss 0.236314 LR 0.000500 Time 0.026075 -2023-02-13 18:15:32,139 - Epoch: [130][ 130/ 1207] Overall Loss 0.236314 Objective Loss 0.236314 LR 0.000500 Time 0.025547 -2023-02-13 18:15:32,329 - Epoch: [130][ 140/ 1207] Overall Loss 0.234518 Objective Loss 0.234518 LR 0.000500 Time 0.025075 -2023-02-13 18:15:32,520 - Epoch: [130][ 150/ 1207] Overall Loss 0.234227 Objective Loss 0.234227 LR 0.000500 Time 0.024675 -2023-02-13 18:15:32,710 - Epoch: [130][ 160/ 1207] Overall Loss 0.233926 Objective Loss 0.233926 LR 0.000500 Time 0.024320 -2023-02-13 18:15:32,902 - Epoch: [130][ 170/ 1207] Overall Loss 0.234665 Objective Loss 0.234665 LR 0.000500 Time 0.024014 -2023-02-13 18:15:33,092 - Epoch: [130][ 180/ 1207] Overall Loss 0.234008 Objective Loss 0.234008 LR 0.000500 Time 0.023737 -2023-02-13 18:15:33,284 - Epoch: [130][ 190/ 1207] Overall Loss 0.233892 Objective Loss 0.233892 LR 0.000500 Time 0.023494 -2023-02-13 18:15:33,474 - Epoch: [130][ 200/ 1207] Overall Loss 0.234000 Objective Loss 0.234000 LR 0.000500 Time 0.023269 -2023-02-13 18:15:33,666 - Epoch: [130][ 210/ 1207] Overall Loss 0.235819 Objective Loss 0.235819 LR 0.000500 Time 0.023073 -2023-02-13 18:15:33,856 - Epoch: [130][ 220/ 1207] Overall Loss 0.235595 Objective Loss 0.235595 LR 0.000500 Time 0.022886 -2023-02-13 18:15:34,049 - Epoch: [130][ 230/ 1207] Overall Loss 0.234932 Objective Loss 0.234932 LR 0.000500 Time 0.022726 -2023-02-13 18:15:34,239 - Epoch: [130][ 240/ 1207] Overall Loss 0.234753 Objective Loss 0.234753 LR 0.000500 Time 0.022572 -2023-02-13 18:15:34,431 - Epoch: [130][ 250/ 1207] Overall Loss 0.233573 Objective Loss 0.233573 LR 0.000500 Time 0.022435 -2023-02-13 18:15:34,621 - Epoch: [130][ 260/ 1207] Overall Loss 0.234129 Objective Loss 0.234129 LR 0.000500 Time 0.022302 -2023-02-13 18:15:34,814 - Epoch: [130][ 270/ 1207] Overall Loss 0.233552 Objective Loss 0.233552 LR 0.000500 Time 0.022187 -2023-02-13 18:15:35,004 - Epoch: [130][ 280/ 1207] Overall Loss 0.233837 Objective Loss 0.233837 LR 0.000500 Time 0.022073 -2023-02-13 18:15:35,197 - Epoch: [130][ 290/ 1207] Overall Loss 0.233190 Objective Loss 0.233190 LR 0.000500 Time 0.021976 -2023-02-13 18:15:35,387 - Epoch: [130][ 300/ 1207] Overall Loss 0.232822 Objective Loss 0.232822 LR 0.000500 Time 0.021877 -2023-02-13 18:15:35,579 - Epoch: [130][ 310/ 1207] Overall Loss 0.232423 Objective Loss 0.232423 LR 0.000500 Time 0.021787 -2023-02-13 18:15:35,769 - Epoch: [130][ 320/ 1207] Overall Loss 0.231859 Objective Loss 0.231859 LR 0.000500 Time 0.021700 -2023-02-13 18:15:35,962 - Epoch: [130][ 330/ 1207] Overall Loss 0.232387 Objective Loss 0.232387 LR 0.000500 Time 0.021626 -2023-02-13 18:15:36,153 - Epoch: [130][ 340/ 1207] Overall Loss 0.232916 Objective Loss 0.232916 LR 0.000500 Time 0.021551 -2023-02-13 18:15:36,345 - Epoch: [130][ 350/ 1207] Overall Loss 0.232945 Objective Loss 0.232945 LR 0.000500 Time 0.021482 -2023-02-13 18:15:36,535 - Epoch: [130][ 360/ 1207] Overall Loss 0.232498 Objective Loss 0.232498 LR 0.000500 Time 0.021414 -2023-02-13 18:15:36,727 - Epoch: [130][ 370/ 1207] Overall Loss 0.233458 Objective Loss 0.233458 LR 0.000500 Time 0.021353 -2023-02-13 18:15:36,918 - Epoch: [130][ 380/ 1207] Overall Loss 0.233266 Objective Loss 0.233266 LR 0.000500 Time 0.021293 -2023-02-13 18:15:37,112 - Epoch: [130][ 390/ 1207] Overall Loss 0.233651 Objective Loss 0.233651 LR 0.000500 Time 0.021243 -2023-02-13 18:15:37,303 - Epoch: [130][ 400/ 1207] Overall Loss 0.234014 Objective Loss 0.234014 LR 0.000500 Time 0.021187 -2023-02-13 18:15:37,494 - Epoch: [130][ 410/ 1207] Overall Loss 0.234173 Objective Loss 0.234173 LR 0.000500 Time 0.021137 -2023-02-13 18:15:37,684 - Epoch: [130][ 420/ 1207] Overall Loss 0.234567 Objective Loss 0.234567 LR 0.000500 Time 0.021085 -2023-02-13 18:15:37,876 - Epoch: [130][ 430/ 1207] Overall Loss 0.234473 Objective Loss 0.234473 LR 0.000500 Time 0.021041 -2023-02-13 18:15:38,067 - Epoch: [130][ 440/ 1207] Overall Loss 0.234944 Objective Loss 0.234944 LR 0.000500 Time 0.020996 -2023-02-13 18:15:38,260 - Epoch: [130][ 450/ 1207] Overall Loss 0.235294 Objective Loss 0.235294 LR 0.000500 Time 0.020956 -2023-02-13 18:15:38,450 - Epoch: [130][ 460/ 1207] Overall Loss 0.235860 Objective Loss 0.235860 LR 0.000500 Time 0.020913 -2023-02-13 18:15:38,642 - Epoch: [130][ 470/ 1207] Overall Loss 0.236256 Objective Loss 0.236256 LR 0.000500 Time 0.020876 -2023-02-13 18:15:38,836 - Epoch: [130][ 480/ 1207] Overall Loss 0.236105 Objective Loss 0.236105 LR 0.000500 Time 0.020844 -2023-02-13 18:15:39,032 - Epoch: [130][ 490/ 1207] Overall Loss 0.236108 Objective Loss 0.236108 LR 0.000500 Time 0.020818 -2023-02-13 18:15:39,231 - Epoch: [130][ 500/ 1207] Overall Loss 0.236463 Objective Loss 0.236463 LR 0.000500 Time 0.020800 -2023-02-13 18:15:39,427 - Epoch: [130][ 510/ 1207] Overall Loss 0.236843 Objective Loss 0.236843 LR 0.000500 Time 0.020774 -2023-02-13 18:15:39,625 - Epoch: [130][ 520/ 1207] Overall Loss 0.237124 Objective Loss 0.237124 LR 0.000500 Time 0.020755 -2023-02-13 18:15:39,820 - Epoch: [130][ 530/ 1207] Overall Loss 0.236879 Objective Loss 0.236879 LR 0.000500 Time 0.020731 -2023-02-13 18:15:40,018 - Epoch: [130][ 540/ 1207] Overall Loss 0.237235 Objective Loss 0.237235 LR 0.000500 Time 0.020714 -2023-02-13 18:15:40,215 - Epoch: [130][ 550/ 1207] Overall Loss 0.236859 Objective Loss 0.236859 LR 0.000500 Time 0.020693 -2023-02-13 18:15:40,413 - Epoch: [130][ 560/ 1207] Overall Loss 0.237151 Objective Loss 0.237151 LR 0.000500 Time 0.020677 -2023-02-13 18:15:40,609 - Epoch: [130][ 570/ 1207] Overall Loss 0.237332 Objective Loss 0.237332 LR 0.000500 Time 0.020657 -2023-02-13 18:15:40,806 - Epoch: [130][ 580/ 1207] Overall Loss 0.237378 Objective Loss 0.237378 LR 0.000500 Time 0.020640 -2023-02-13 18:15:41,004 - Epoch: [130][ 590/ 1207] Overall Loss 0.237526 Objective Loss 0.237526 LR 0.000500 Time 0.020625 -2023-02-13 18:15:41,202 - Epoch: [130][ 600/ 1207] Overall Loss 0.237369 Objective Loss 0.237369 LR 0.000500 Time 0.020611 -2023-02-13 18:15:41,398 - Epoch: [130][ 610/ 1207] Overall Loss 0.237506 Objective Loss 0.237506 LR 0.000500 Time 0.020594 -2023-02-13 18:15:41,596 - Epoch: [130][ 620/ 1207] Overall Loss 0.237392 Objective Loss 0.237392 LR 0.000500 Time 0.020580 -2023-02-13 18:15:41,791 - Epoch: [130][ 630/ 1207] Overall Loss 0.237791 Objective Loss 0.237791 LR 0.000500 Time 0.020563 -2023-02-13 18:15:41,990 - Epoch: [130][ 640/ 1207] Overall Loss 0.237676 Objective Loss 0.237676 LR 0.000500 Time 0.020552 -2023-02-13 18:15:42,186 - Epoch: [130][ 650/ 1207] Overall Loss 0.237658 Objective Loss 0.237658 LR 0.000500 Time 0.020537 -2023-02-13 18:15:42,385 - Epoch: [130][ 660/ 1207] Overall Loss 0.237934 Objective Loss 0.237934 LR 0.000500 Time 0.020526 -2023-02-13 18:15:42,581 - Epoch: [130][ 670/ 1207] Overall Loss 0.237654 Objective Loss 0.237654 LR 0.000500 Time 0.020512 -2023-02-13 18:15:42,779 - Epoch: [130][ 680/ 1207] Overall Loss 0.237751 Objective Loss 0.237751 LR 0.000500 Time 0.020501 -2023-02-13 18:15:42,975 - Epoch: [130][ 690/ 1207] Overall Loss 0.237837 Objective Loss 0.237837 LR 0.000500 Time 0.020488 -2023-02-13 18:15:43,175 - Epoch: [130][ 700/ 1207] Overall Loss 0.237846 Objective Loss 0.237846 LR 0.000500 Time 0.020479 -2023-02-13 18:15:43,371 - Epoch: [130][ 710/ 1207] Overall Loss 0.237934 Objective Loss 0.237934 LR 0.000500 Time 0.020467 -2023-02-13 18:15:43,569 - Epoch: [130][ 720/ 1207] Overall Loss 0.237897 Objective Loss 0.237897 LR 0.000500 Time 0.020458 -2023-02-13 18:15:43,765 - Epoch: [130][ 730/ 1207] Overall Loss 0.238005 Objective Loss 0.238005 LR 0.000500 Time 0.020445 -2023-02-13 18:15:43,963 - Epoch: [130][ 740/ 1207] Overall Loss 0.237884 Objective Loss 0.237884 LR 0.000500 Time 0.020436 -2023-02-13 18:15:44,160 - Epoch: [130][ 750/ 1207] Overall Loss 0.238484 Objective Loss 0.238484 LR 0.000500 Time 0.020426 -2023-02-13 18:15:44,358 - Epoch: [130][ 760/ 1207] Overall Loss 0.238715 Objective Loss 0.238715 LR 0.000500 Time 0.020417 -2023-02-13 18:15:44,554 - Epoch: [130][ 770/ 1207] Overall Loss 0.238592 Objective Loss 0.238592 LR 0.000500 Time 0.020406 -2023-02-13 18:15:44,752 - Epoch: [130][ 780/ 1207] Overall Loss 0.238324 Objective Loss 0.238324 LR 0.000500 Time 0.020398 -2023-02-13 18:15:44,948 - Epoch: [130][ 790/ 1207] Overall Loss 0.237952 Objective Loss 0.237952 LR 0.000500 Time 0.020387 -2023-02-13 18:15:45,148 - Epoch: [130][ 800/ 1207] Overall Loss 0.237899 Objective Loss 0.237899 LR 0.000500 Time 0.020382 -2023-02-13 18:15:45,344 - Epoch: [130][ 810/ 1207] Overall Loss 0.237734 Objective Loss 0.237734 LR 0.000500 Time 0.020372 -2023-02-13 18:15:45,542 - Epoch: [130][ 820/ 1207] Overall Loss 0.237641 Objective Loss 0.237641 LR 0.000500 Time 0.020365 -2023-02-13 18:15:45,739 - Epoch: [130][ 830/ 1207] Overall Loss 0.237688 Objective Loss 0.237688 LR 0.000500 Time 0.020356 -2023-02-13 18:15:45,938 - Epoch: [130][ 840/ 1207] Overall Loss 0.237916 Objective Loss 0.237916 LR 0.000500 Time 0.020350 -2023-02-13 18:15:46,135 - Epoch: [130][ 850/ 1207] Overall Loss 0.237566 Objective Loss 0.237566 LR 0.000500 Time 0.020341 -2023-02-13 18:15:46,333 - Epoch: [130][ 860/ 1207] Overall Loss 0.237510 Objective Loss 0.237510 LR 0.000500 Time 0.020335 -2023-02-13 18:15:46,529 - Epoch: [130][ 870/ 1207] Overall Loss 0.237622 Objective Loss 0.237622 LR 0.000500 Time 0.020326 -2023-02-13 18:15:46,728 - Epoch: [130][ 880/ 1207] Overall Loss 0.237571 Objective Loss 0.237571 LR 0.000500 Time 0.020320 -2023-02-13 18:15:46,925 - Epoch: [130][ 890/ 1207] Overall Loss 0.237617 Objective Loss 0.237617 LR 0.000500 Time 0.020313 -2023-02-13 18:15:47,123 - Epoch: [130][ 900/ 1207] Overall Loss 0.237621 Objective Loss 0.237621 LR 0.000500 Time 0.020308 -2023-02-13 18:15:47,320 - Epoch: [130][ 910/ 1207] Overall Loss 0.237905 Objective Loss 0.237905 LR 0.000500 Time 0.020300 -2023-02-13 18:15:47,519 - Epoch: [130][ 920/ 1207] Overall Loss 0.237945 Objective Loss 0.237945 LR 0.000500 Time 0.020295 -2023-02-13 18:15:47,715 - Epoch: [130][ 930/ 1207] Overall Loss 0.237939 Objective Loss 0.237939 LR 0.000500 Time 0.020288 -2023-02-13 18:15:47,914 - Epoch: [130][ 940/ 1207] Overall Loss 0.237995 Objective Loss 0.237995 LR 0.000500 Time 0.020283 -2023-02-13 18:15:48,110 - Epoch: [130][ 950/ 1207] Overall Loss 0.237956 Objective Loss 0.237956 LR 0.000500 Time 0.020276 -2023-02-13 18:15:48,309 - Epoch: [130][ 960/ 1207] Overall Loss 0.238419 Objective Loss 0.238419 LR 0.000500 Time 0.020271 -2023-02-13 18:15:48,505 - Epoch: [130][ 970/ 1207] Overall Loss 0.238486 Objective Loss 0.238486 LR 0.000500 Time 0.020264 -2023-02-13 18:15:48,704 - Epoch: [130][ 980/ 1207] Overall Loss 0.238445 Objective Loss 0.238445 LR 0.000500 Time 0.020260 -2023-02-13 18:15:48,899 - Epoch: [130][ 990/ 1207] Overall Loss 0.238378 Objective Loss 0.238378 LR 0.000500 Time 0.020252 -2023-02-13 18:15:49,098 - Epoch: [130][ 1000/ 1207] Overall Loss 0.238519 Objective Loss 0.238519 LR 0.000500 Time 0.020248 -2023-02-13 18:15:49,293 - Epoch: [130][ 1010/ 1207] Overall Loss 0.238593 Objective Loss 0.238593 LR 0.000500 Time 0.020241 -2023-02-13 18:15:49,491 - Epoch: [130][ 1020/ 1207] Overall Loss 0.238626 Objective Loss 0.238626 LR 0.000500 Time 0.020236 -2023-02-13 18:15:49,687 - Epoch: [130][ 1030/ 1207] Overall Loss 0.238877 Objective Loss 0.238877 LR 0.000500 Time 0.020229 -2023-02-13 18:15:49,886 - Epoch: [130][ 1040/ 1207] Overall Loss 0.238963 Objective Loss 0.238963 LR 0.000500 Time 0.020225 -2023-02-13 18:15:50,081 - Epoch: [130][ 1050/ 1207] Overall Loss 0.238876 Objective Loss 0.238876 LR 0.000500 Time 0.020218 -2023-02-13 18:15:50,279 - Epoch: [130][ 1060/ 1207] Overall Loss 0.239014 Objective Loss 0.239014 LR 0.000500 Time 0.020214 -2023-02-13 18:15:50,474 - Epoch: [130][ 1070/ 1207] Overall Loss 0.239121 Objective Loss 0.239121 LR 0.000500 Time 0.020207 -2023-02-13 18:15:50,671 - Epoch: [130][ 1080/ 1207] Overall Loss 0.239288 Objective Loss 0.239288 LR 0.000500 Time 0.020202 -2023-02-13 18:15:50,866 - Epoch: [130][ 1090/ 1207] Overall Loss 0.239368 Objective Loss 0.239368 LR 0.000500 Time 0.020195 -2023-02-13 18:15:51,064 - Epoch: [130][ 1100/ 1207] Overall Loss 0.239523 Objective Loss 0.239523 LR 0.000500 Time 0.020191 -2023-02-13 18:15:51,259 - Epoch: [130][ 1110/ 1207] Overall Loss 0.239478 Objective Loss 0.239478 LR 0.000500 Time 0.020184 -2023-02-13 18:15:51,456 - Epoch: [130][ 1120/ 1207] Overall Loss 0.239680 Objective Loss 0.239680 LR 0.000500 Time 0.020180 -2023-02-13 18:15:51,651 - Epoch: [130][ 1130/ 1207] Overall Loss 0.239750 Objective Loss 0.239750 LR 0.000500 Time 0.020173 -2023-02-13 18:15:51,848 - Epoch: [130][ 1140/ 1207] Overall Loss 0.239850 Objective Loss 0.239850 LR 0.000500 Time 0.020169 -2023-02-13 18:15:52,043 - Epoch: [130][ 1150/ 1207] Overall Loss 0.239899 Objective Loss 0.239899 LR 0.000500 Time 0.020163 -2023-02-13 18:15:52,241 - Epoch: [130][ 1160/ 1207] Overall Loss 0.239867 Objective Loss 0.239867 LR 0.000500 Time 0.020159 -2023-02-13 18:15:52,435 - Epoch: [130][ 1170/ 1207] Overall Loss 0.239978 Objective Loss 0.239978 LR 0.000500 Time 0.020153 -2023-02-13 18:15:52,632 - Epoch: [130][ 1180/ 1207] Overall Loss 0.240057 Objective Loss 0.240057 LR 0.000500 Time 0.020149 -2023-02-13 18:15:52,826 - Epoch: [130][ 1190/ 1207] Overall Loss 0.239913 Objective Loss 0.239913 LR 0.000500 Time 0.020142 -2023-02-13 18:15:53,078 - Epoch: [130][ 1200/ 1207] Overall Loss 0.239929 Objective Loss 0.239929 LR 0.000500 Time 0.020184 -2023-02-13 18:15:53,195 - Epoch: [130][ 1207/ 1207] Overall Loss 0.239825 Objective Loss 0.239825 Top1 86.890244 Top5 98.475610 LR 0.000500 Time 0.020163 -2023-02-13 18:15:53,283 - --- validate (epoch=130)----------- -2023-02-13 18:15:53,283 - 34311 samples (256 per mini-batch) -2023-02-13 18:15:53,687 - Epoch: [130][ 10/ 135] Loss 0.310939 Top1 84.179688 Top5 97.773438 -2023-02-13 18:15:53,818 - Epoch: [130][ 20/ 135] Loss 0.302530 Top1 84.492188 Top5 97.792969 -2023-02-13 18:15:53,948 - Epoch: [130][ 30/ 135] Loss 0.301593 Top1 85.065104 Top5 97.747396 -2023-02-13 18:15:54,077 - Epoch: [130][ 40/ 135] Loss 0.316584 Top1 84.912109 Top5 97.773438 -2023-02-13 18:15:54,209 - Epoch: [130][ 50/ 135] Loss 0.318048 Top1 84.898438 Top5 97.640625 -2023-02-13 18:15:54,333 - Epoch: [130][ 60/ 135] Loss 0.316961 Top1 84.778646 Top5 97.649740 -2023-02-13 18:15:54,476 - Epoch: [130][ 70/ 135] Loss 0.318284 Top1 84.587054 Top5 97.695312 -2023-02-13 18:15:54,616 - Epoch: [130][ 80/ 135] Loss 0.319716 Top1 84.580078 Top5 97.705078 -2023-02-13 18:15:54,759 - Epoch: [130][ 90/ 135] Loss 0.320250 Top1 84.548611 Top5 97.673611 -2023-02-13 18:15:54,888 - Epoch: [130][ 100/ 135] Loss 0.315978 Top1 84.687500 Top5 97.691406 -2023-02-13 18:15:55,021 - Epoch: [130][ 110/ 135] Loss 0.316535 Top1 84.740767 Top5 97.702415 -2023-02-13 18:15:55,151 - Epoch: [130][ 120/ 135] Loss 0.316024 Top1 84.664714 Top5 97.721354 -2023-02-13 18:15:55,285 - Epoch: [130][ 130/ 135] Loss 0.312530 Top1 84.669471 Top5 97.734375 -2023-02-13 18:15:55,333 - Epoch: [130][ 135/ 135] Loss 0.313717 Top1 84.660896 Top5 97.749993 -2023-02-13 18:15:55,406 - ==> Top1: 84.661 Top5: 97.750 Loss: 0.314 - -2023-02-13 18:15:55,406 - ==> Confusion: -[[ 857 4 9 0 10 2 0 2 2 52 0 5 0 2 5 3 1 2 2 4 5] - [ 0 943 3 1 11 30 1 16 2 0 1 0 1 0 1 1 7 0 3 1 11] - [ 6 4 970 6 4 1 12 12 0 2 4 0 4 3 1 5 2 3 4 8 7] - [ 6 1 24 888 2 6 2 2 1 2 18 0 7 0 23 2 2 5 18 1 6] - [ 10 10 1 1 992 8 1 4 1 2 0 5 2 2 6 4 8 1 0 3 5] - [ 1 18 2 2 8 960 2 17 2 5 2 11 2 14 1 1 5 0 2 10 5] - [ 2 6 25 3 0 9 1023 5 0 1 1 1 3 1 1 2 1 2 1 7 5] - [ 1 10 10 1 3 21 6 928 0 1 0 6 3 1 0 0 1 1 18 8 5] - [ 17 4 1 1 1 2 0 2 905 35 9 3 0 6 8 2 3 2 6 0 2] - [ 79 5 4 0 13 0 0 1 26 858 0 1 0 11 4 0 1 3 1 1 4] - [ 1 2 7 5 2 3 5 7 17 0 969 1 1 7 4 0 3 1 12 1 3] - [ 2 6 4 0 3 8 1 6 0 0 0 916 26 5 0 4 2 11 2 9 0] - [ 2 0 0 6 2 2 0 1 1 0 0 27 883 0 3 4 2 14 4 1 7] - [ 6 3 4 0 9 11 0 3 14 19 10 7 1 912 7 2 2 1 1 5 7] - [ 6 3 3 18 3 3 0 1 14 4 7 1 5 0 992 0 1 7 12 3 9] - [ 5 2 6 0 8 1 5 3 0 0 0 8 10 2 1 957 8 12 0 9 9] - [ 7 2 0 0 5 3 0 2 0 1 0 0 3 3 1 9 1010 1 2 4 8] - [ 6 3 4 4 0 3 1 1 1 1 1 7 15 1 0 12 0 984 0 4 3] - [ 5 4 7 8 0 2 0 24 3 0 2 2 2 0 7 1 1 2 1011 2 3] - [ 1 5 2 0 2 4 4 15 0 0 0 15 3 2 0 4 9 2 2 1070 8] - [ 149 264 302 92 142 181 87 182 90 86 166 98 285 238 148 68 287 87 209 253 10020]] - -2023-02-13 18:15:55,408 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:15:55,408 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:15:55,414 - - -2023-02-13 18:15:55,414 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:15:56,309 - Epoch: [131][ 10/ 1207] Overall Loss 0.213745 Objective Loss 0.213745 LR 0.000500 Time 0.089511 -2023-02-13 18:15:56,506 - Epoch: [131][ 20/ 1207] Overall Loss 0.226442 Objective Loss 0.226442 LR 0.000500 Time 0.054581 -2023-02-13 18:15:56,699 - Epoch: [131][ 30/ 1207] Overall Loss 0.233916 Objective Loss 0.233916 LR 0.000500 Time 0.042810 -2023-02-13 18:15:56,895 - Epoch: [131][ 40/ 1207] Overall Loss 0.237172 Objective Loss 0.237172 LR 0.000500 Time 0.036995 -2023-02-13 18:15:57,088 - Epoch: [131][ 50/ 1207] Overall Loss 0.235771 Objective Loss 0.235771 LR 0.000500 Time 0.033436 -2023-02-13 18:15:57,284 - Epoch: [131][ 60/ 1207] Overall Loss 0.237253 Objective Loss 0.237253 LR 0.000500 Time 0.031126 -2023-02-13 18:15:57,476 - Epoch: [131][ 70/ 1207] Overall Loss 0.241092 Objective Loss 0.241092 LR 0.000500 Time 0.029424 -2023-02-13 18:15:57,672 - Epoch: [131][ 80/ 1207] Overall Loss 0.241078 Objective Loss 0.241078 LR 0.000500 Time 0.028185 -2023-02-13 18:15:57,865 - Epoch: [131][ 90/ 1207] Overall Loss 0.240851 Objective Loss 0.240851 LR 0.000500 Time 0.027191 -2023-02-13 18:15:58,060 - Epoch: [131][ 100/ 1207] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.000500 Time 0.026419 -2023-02-13 18:15:58,253 - Epoch: [131][ 110/ 1207] Overall Loss 0.241129 Objective Loss 0.241129 LR 0.000500 Time 0.025773 -2023-02-13 18:15:58,448 - Epoch: [131][ 120/ 1207] Overall Loss 0.241077 Objective Loss 0.241077 LR 0.000500 Time 0.025247 -2023-02-13 18:15:58,641 - Epoch: [131][ 130/ 1207] Overall Loss 0.240281 Objective Loss 0.240281 LR 0.000500 Time 0.024786 -2023-02-13 18:15:58,836 - Epoch: [131][ 140/ 1207] Overall Loss 0.239227 Objective Loss 0.239227 LR 0.000500 Time 0.024405 -2023-02-13 18:15:59,029 - Epoch: [131][ 150/ 1207] Overall Loss 0.239936 Objective Loss 0.239936 LR 0.000500 Time 0.024060 -2023-02-13 18:15:59,225 - Epoch: [131][ 160/ 1207] Overall Loss 0.240916 Objective Loss 0.240916 LR 0.000500 Time 0.023780 -2023-02-13 18:15:59,417 - Epoch: [131][ 170/ 1207] Overall Loss 0.240661 Objective Loss 0.240661 LR 0.000500 Time 0.023507 -2023-02-13 18:15:59,611 - Epoch: [131][ 180/ 1207] Overall Loss 0.241289 Objective Loss 0.241289 LR 0.000500 Time 0.023282 -2023-02-13 18:15:59,804 - Epoch: [131][ 190/ 1207] Overall Loss 0.240527 Objective Loss 0.240527 LR 0.000500 Time 0.023066 -2023-02-13 18:15:59,999 - Epoch: [131][ 200/ 1207] Overall Loss 0.240272 Objective Loss 0.240272 LR 0.000500 Time 0.022886 -2023-02-13 18:16:00,191 - Epoch: [131][ 210/ 1207] Overall Loss 0.240887 Objective Loss 0.240887 LR 0.000500 Time 0.022711 -2023-02-13 18:16:00,386 - Epoch: [131][ 220/ 1207] Overall Loss 0.239779 Objective Loss 0.239779 LR 0.000500 Time 0.022565 -2023-02-13 18:16:00,579 - Epoch: [131][ 230/ 1207] Overall Loss 0.240259 Objective Loss 0.240259 LR 0.000500 Time 0.022421 -2023-02-13 18:16:00,775 - Epoch: [131][ 240/ 1207] Overall Loss 0.239897 Objective Loss 0.239897 LR 0.000500 Time 0.022301 -2023-02-13 18:16:00,970 - Epoch: [131][ 250/ 1207] Overall Loss 0.239658 Objective Loss 0.239658 LR 0.000500 Time 0.022185 -2023-02-13 18:16:01,166 - Epoch: [131][ 260/ 1207] Overall Loss 0.239597 Objective Loss 0.239597 LR 0.000500 Time 0.022086 -2023-02-13 18:16:01,359 - Epoch: [131][ 270/ 1207] Overall Loss 0.239157 Objective Loss 0.239157 LR 0.000500 Time 0.021981 -2023-02-13 18:16:01,554 - Epoch: [131][ 280/ 1207] Overall Loss 0.239413 Objective Loss 0.239413 LR 0.000500 Time 0.021892 -2023-02-13 18:16:01,747 - Epoch: [131][ 290/ 1207] Overall Loss 0.239762 Objective Loss 0.239762 LR 0.000500 Time 0.021802 -2023-02-13 18:16:01,943 - Epoch: [131][ 300/ 1207] Overall Loss 0.241016 Objective Loss 0.241016 LR 0.000500 Time 0.021727 -2023-02-13 18:16:02,136 - Epoch: [131][ 310/ 1207] Overall Loss 0.240567 Objective Loss 0.240567 LR 0.000500 Time 0.021647 -2023-02-13 18:16:02,332 - Epoch: [131][ 320/ 1207] Overall Loss 0.239615 Objective Loss 0.239615 LR 0.000500 Time 0.021581 -2023-02-13 18:16:02,525 - Epoch: [131][ 330/ 1207] Overall Loss 0.238538 Objective Loss 0.238538 LR 0.000500 Time 0.021511 -2023-02-13 18:16:02,720 - Epoch: [131][ 340/ 1207] Overall Loss 0.238559 Objective Loss 0.238559 LR 0.000500 Time 0.021450 -2023-02-13 18:16:02,912 - Epoch: [131][ 350/ 1207] Overall Loss 0.238664 Objective Loss 0.238664 LR 0.000500 Time 0.021386 -2023-02-13 18:16:03,107 - Epoch: [131][ 360/ 1207] Overall Loss 0.238836 Objective Loss 0.238836 LR 0.000500 Time 0.021333 -2023-02-13 18:16:03,301 - Epoch: [131][ 370/ 1207] Overall Loss 0.238482 Objective Loss 0.238482 LR 0.000500 Time 0.021278 -2023-02-13 18:16:03,496 - Epoch: [131][ 380/ 1207] Overall Loss 0.239266 Objective Loss 0.239266 LR 0.000500 Time 0.021232 -2023-02-13 18:16:03,690 - Epoch: [131][ 390/ 1207] Overall Loss 0.239632 Objective Loss 0.239632 LR 0.000500 Time 0.021182 -2023-02-13 18:16:03,885 - Epoch: [131][ 400/ 1207] Overall Loss 0.239651 Objective Loss 0.239651 LR 0.000500 Time 0.021140 -2023-02-13 18:16:04,077 - Epoch: [131][ 410/ 1207] Overall Loss 0.239110 Objective Loss 0.239110 LR 0.000500 Time 0.021093 -2023-02-13 18:16:04,274 - Epoch: [131][ 420/ 1207] Overall Loss 0.238366 Objective Loss 0.238366 LR 0.000500 Time 0.021057 -2023-02-13 18:16:04,467 - Epoch: [131][ 430/ 1207] Overall Loss 0.238783 Objective Loss 0.238783 LR 0.000500 Time 0.021015 -2023-02-13 18:16:04,662 - Epoch: [131][ 440/ 1207] Overall Loss 0.238903 Objective Loss 0.238903 LR 0.000500 Time 0.020981 -2023-02-13 18:16:04,855 - Epoch: [131][ 450/ 1207] Overall Loss 0.238934 Objective Loss 0.238934 LR 0.000500 Time 0.020943 -2023-02-13 18:16:05,051 - Epoch: [131][ 460/ 1207] Overall Loss 0.238926 Objective Loss 0.238926 LR 0.000500 Time 0.020912 -2023-02-13 18:16:05,244 - Epoch: [131][ 470/ 1207] Overall Loss 0.238967 Objective Loss 0.238967 LR 0.000500 Time 0.020877 -2023-02-13 18:16:05,440 - Epoch: [131][ 480/ 1207] Overall Loss 0.238717 Objective Loss 0.238717 LR 0.000500 Time 0.020849 -2023-02-13 18:16:05,632 - Epoch: [131][ 490/ 1207] Overall Loss 0.238923 Objective Loss 0.238923 LR 0.000500 Time 0.020816 -2023-02-13 18:16:05,828 - Epoch: [131][ 500/ 1207] Overall Loss 0.238699 Objective Loss 0.238699 LR 0.000500 Time 0.020789 -2023-02-13 18:16:06,022 - Epoch: [131][ 510/ 1207] Overall Loss 0.238517 Objective Loss 0.238517 LR 0.000500 Time 0.020763 -2023-02-13 18:16:06,218 - Epoch: [131][ 520/ 1207] Overall Loss 0.237904 Objective Loss 0.237904 LR 0.000500 Time 0.020739 -2023-02-13 18:16:06,411 - Epoch: [131][ 530/ 1207] Overall Loss 0.237637 Objective Loss 0.237637 LR 0.000500 Time 0.020712 -2023-02-13 18:16:06,606 - Epoch: [131][ 540/ 1207] Overall Loss 0.237095 Objective Loss 0.237095 LR 0.000500 Time 0.020689 -2023-02-13 18:16:06,800 - Epoch: [131][ 550/ 1207] Overall Loss 0.237499 Objective Loss 0.237499 LR 0.000500 Time 0.020664 -2023-02-13 18:16:06,996 - Epoch: [131][ 560/ 1207] Overall Loss 0.237354 Objective Loss 0.237354 LR 0.000500 Time 0.020644 -2023-02-13 18:16:07,189 - Epoch: [131][ 570/ 1207] Overall Loss 0.237452 Objective Loss 0.237452 LR 0.000500 Time 0.020620 -2023-02-13 18:16:07,386 - Epoch: [131][ 580/ 1207] Overall Loss 0.237431 Objective Loss 0.237431 LR 0.000500 Time 0.020604 -2023-02-13 18:16:07,582 - Epoch: [131][ 590/ 1207] Overall Loss 0.237500 Objective Loss 0.237500 LR 0.000500 Time 0.020586 -2023-02-13 18:16:07,781 - Epoch: [131][ 600/ 1207] Overall Loss 0.237513 Objective Loss 0.237513 LR 0.000500 Time 0.020573 -2023-02-13 18:16:07,977 - Epoch: [131][ 610/ 1207] Overall Loss 0.237842 Objective Loss 0.237842 LR 0.000500 Time 0.020557 -2023-02-13 18:16:08,175 - Epoch: [131][ 620/ 1207] Overall Loss 0.237597 Objective Loss 0.237597 LR 0.000500 Time 0.020545 -2023-02-13 18:16:08,372 - Epoch: [131][ 630/ 1207] Overall Loss 0.237482 Objective Loss 0.237482 LR 0.000500 Time 0.020530 -2023-02-13 18:16:08,570 - Epoch: [131][ 640/ 1207] Overall Loss 0.238009 Objective Loss 0.238009 LR 0.000500 Time 0.020518 -2023-02-13 18:16:08,766 - Epoch: [131][ 650/ 1207] Overall Loss 0.238137 Objective Loss 0.238137 LR 0.000500 Time 0.020504 -2023-02-13 18:16:08,965 - Epoch: [131][ 660/ 1207] Overall Loss 0.238152 Objective Loss 0.238152 LR 0.000500 Time 0.020494 -2023-02-13 18:16:09,161 - Epoch: [131][ 670/ 1207] Overall Loss 0.238213 Objective Loss 0.238213 LR 0.000500 Time 0.020480 -2023-02-13 18:16:09,359 - Epoch: [131][ 680/ 1207] Overall Loss 0.238253 Objective Loss 0.238253 LR 0.000500 Time 0.020470 -2023-02-13 18:16:09,556 - Epoch: [131][ 690/ 1207] Overall Loss 0.238514 Objective Loss 0.238514 LR 0.000500 Time 0.020458 -2023-02-13 18:16:09,755 - Epoch: [131][ 700/ 1207] Overall Loss 0.238437 Objective Loss 0.238437 LR 0.000500 Time 0.020449 -2023-02-13 18:16:09,951 - Epoch: [131][ 710/ 1207] Overall Loss 0.238313 Objective Loss 0.238313 LR 0.000500 Time 0.020437 -2023-02-13 18:16:10,150 - Epoch: [131][ 720/ 1207] Overall Loss 0.238082 Objective Loss 0.238082 LR 0.000500 Time 0.020429 -2023-02-13 18:16:10,345 - Epoch: [131][ 730/ 1207] Overall Loss 0.238312 Objective Loss 0.238312 LR 0.000500 Time 0.020416 -2023-02-13 18:16:10,543 - Epoch: [131][ 740/ 1207] Overall Loss 0.238770 Objective Loss 0.238770 LR 0.000500 Time 0.020408 -2023-02-13 18:16:10,740 - Epoch: [131][ 750/ 1207] Overall Loss 0.238814 Objective Loss 0.238814 LR 0.000500 Time 0.020397 -2023-02-13 18:16:10,939 - Epoch: [131][ 760/ 1207] Overall Loss 0.238955 Objective Loss 0.238955 LR 0.000500 Time 0.020390 -2023-02-13 18:16:11,136 - Epoch: [131][ 770/ 1207] Overall Loss 0.238941 Objective Loss 0.238941 LR 0.000500 Time 0.020380 -2023-02-13 18:16:11,335 - Epoch: [131][ 780/ 1207] Overall Loss 0.239213 Objective Loss 0.239213 LR 0.000500 Time 0.020374 -2023-02-13 18:16:11,530 - Epoch: [131][ 790/ 1207] Overall Loss 0.239461 Objective Loss 0.239461 LR 0.000500 Time 0.020362 -2023-02-13 18:16:11,728 - Epoch: [131][ 800/ 1207] Overall Loss 0.239778 Objective Loss 0.239778 LR 0.000500 Time 0.020355 -2023-02-13 18:16:11,923 - Epoch: [131][ 810/ 1207] Overall Loss 0.239818 Objective Loss 0.239818 LR 0.000500 Time 0.020344 -2023-02-13 18:16:12,122 - Epoch: [131][ 820/ 1207] Overall Loss 0.240241 Objective Loss 0.240241 LR 0.000500 Time 0.020338 -2023-02-13 18:16:12,317 - Epoch: [131][ 830/ 1207] Overall Loss 0.240438 Objective Loss 0.240438 LR 0.000500 Time 0.020327 -2023-02-13 18:16:12,515 - Epoch: [131][ 840/ 1207] Overall Loss 0.240433 Objective Loss 0.240433 LR 0.000500 Time 0.020320 -2023-02-13 18:16:12,710 - Epoch: [131][ 850/ 1207] Overall Loss 0.240564 Objective Loss 0.240564 LR 0.000500 Time 0.020311 -2023-02-13 18:16:12,908 - Epoch: [131][ 860/ 1207] Overall Loss 0.240817 Objective Loss 0.240817 LR 0.000500 Time 0.020304 -2023-02-13 18:16:13,103 - Epoch: [131][ 870/ 1207] Overall Loss 0.240786 Objective Loss 0.240786 LR 0.000500 Time 0.020295 -2023-02-13 18:16:13,303 - Epoch: [131][ 880/ 1207] Overall Loss 0.240741 Objective Loss 0.240741 LR 0.000500 Time 0.020290 -2023-02-13 18:16:13,498 - Epoch: [131][ 890/ 1207] Overall Loss 0.240558 Objective Loss 0.240558 LR 0.000500 Time 0.020281 -2023-02-13 18:16:13,696 - Epoch: [131][ 900/ 1207] Overall Loss 0.240933 Objective Loss 0.240933 LR 0.000500 Time 0.020276 -2023-02-13 18:16:13,891 - Epoch: [131][ 910/ 1207] Overall Loss 0.241181 Objective Loss 0.241181 LR 0.000500 Time 0.020266 -2023-02-13 18:16:14,089 - Epoch: [131][ 920/ 1207] Overall Loss 0.241136 Objective Loss 0.241136 LR 0.000500 Time 0.020261 -2023-02-13 18:16:14,284 - Epoch: [131][ 930/ 1207] Overall Loss 0.241142 Objective Loss 0.241142 LR 0.000500 Time 0.020253 -2023-02-13 18:16:14,482 - Epoch: [131][ 940/ 1207] Overall Loss 0.240916 Objective Loss 0.240916 LR 0.000500 Time 0.020247 -2023-02-13 18:16:14,676 - Epoch: [131][ 950/ 1207] Overall Loss 0.240939 Objective Loss 0.240939 LR 0.000500 Time 0.020238 -2023-02-13 18:16:14,875 - Epoch: [131][ 960/ 1207] Overall Loss 0.240987 Objective Loss 0.240987 LR 0.000500 Time 0.020234 -2023-02-13 18:16:15,069 - Epoch: [131][ 970/ 1207] Overall Loss 0.240991 Objective Loss 0.240991 LR 0.000500 Time 0.020225 -2023-02-13 18:16:15,268 - Epoch: [131][ 980/ 1207] Overall Loss 0.240780 Objective Loss 0.240780 LR 0.000500 Time 0.020222 -2023-02-13 18:16:15,464 - Epoch: [131][ 990/ 1207] Overall Loss 0.240761 Objective Loss 0.240761 LR 0.000500 Time 0.020214 -2023-02-13 18:16:15,662 - Epoch: [131][ 1000/ 1207] Overall Loss 0.240577 Objective Loss 0.240577 LR 0.000500 Time 0.020210 -2023-02-13 18:16:15,857 - Epoch: [131][ 1010/ 1207] Overall Loss 0.240905 Objective Loss 0.240905 LR 0.000500 Time 0.020202 -2023-02-13 18:16:16,055 - Epoch: [131][ 1020/ 1207] Overall Loss 0.240563 Objective Loss 0.240563 LR 0.000500 Time 0.020198 -2023-02-13 18:16:16,250 - Epoch: [131][ 1030/ 1207] Overall Loss 0.240558 Objective Loss 0.240558 LR 0.000500 Time 0.020192 -2023-02-13 18:16:16,448 - Epoch: [131][ 1040/ 1207] Overall Loss 0.240674 Objective Loss 0.240674 LR 0.000500 Time 0.020188 -2023-02-13 18:16:16,643 - Epoch: [131][ 1050/ 1207] Overall Loss 0.240400 Objective Loss 0.240400 LR 0.000500 Time 0.020181 -2023-02-13 18:16:16,841 - Epoch: [131][ 1060/ 1207] Overall Loss 0.240383 Objective Loss 0.240383 LR 0.000500 Time 0.020177 -2023-02-13 18:16:17,037 - Epoch: [131][ 1070/ 1207] Overall Loss 0.240105 Objective Loss 0.240105 LR 0.000500 Time 0.020170 -2023-02-13 18:16:17,236 - Epoch: [131][ 1080/ 1207] Overall Loss 0.240111 Objective Loss 0.240111 LR 0.000500 Time 0.020168 -2023-02-13 18:16:17,431 - Epoch: [131][ 1090/ 1207] Overall Loss 0.240052 Objective Loss 0.240052 LR 0.000500 Time 0.020161 -2023-02-13 18:16:17,629 - Epoch: [131][ 1100/ 1207] Overall Loss 0.240025 Objective Loss 0.240025 LR 0.000500 Time 0.020158 -2023-02-13 18:16:17,824 - Epoch: [131][ 1110/ 1207] Overall Loss 0.240173 Objective Loss 0.240173 LR 0.000500 Time 0.020152 -2023-02-13 18:16:18,022 - Epoch: [131][ 1120/ 1207] Overall Loss 0.239911 Objective Loss 0.239911 LR 0.000500 Time 0.020148 -2023-02-13 18:16:18,217 - Epoch: [131][ 1130/ 1207] Overall Loss 0.239914 Objective Loss 0.239914 LR 0.000500 Time 0.020142 -2023-02-13 18:16:18,416 - Epoch: [131][ 1140/ 1207] Overall Loss 0.239923 Objective Loss 0.239923 LR 0.000500 Time 0.020139 -2023-02-13 18:16:18,611 - Epoch: [131][ 1150/ 1207] Overall Loss 0.239698 Objective Loss 0.239698 LR 0.000500 Time 0.020133 -2023-02-13 18:16:18,809 - Epoch: [131][ 1160/ 1207] Overall Loss 0.239382 Objective Loss 0.239382 LR 0.000500 Time 0.020131 -2023-02-13 18:16:19,003 - Epoch: [131][ 1170/ 1207] Overall Loss 0.239519 Objective Loss 0.239519 LR 0.000500 Time 0.020124 -2023-02-13 18:16:19,202 - Epoch: [131][ 1180/ 1207] Overall Loss 0.239524 Objective Loss 0.239524 LR 0.000500 Time 0.020122 -2023-02-13 18:16:19,397 - Epoch: [131][ 1190/ 1207] Overall Loss 0.239549 Objective Loss 0.239549 LR 0.000500 Time 0.020116 -2023-02-13 18:16:19,642 - Epoch: [131][ 1200/ 1207] Overall Loss 0.239517 Objective Loss 0.239517 LR 0.000500 Time 0.020152 -2023-02-13 18:16:19,757 - Epoch: [131][ 1207/ 1207] Overall Loss 0.239680 Objective Loss 0.239680 Top1 82.621951 Top5 97.560976 LR 0.000500 Time 0.020131 -2023-02-13 18:16:19,838 - --- validate (epoch=131)----------- -2023-02-13 18:16:19,838 - 34311 samples (256 per mini-batch) -2023-02-13 18:16:20,234 - Epoch: [131][ 10/ 135] Loss 0.336350 Top1 82.851562 Top5 97.851562 -2023-02-13 18:16:20,365 - Epoch: [131][ 20/ 135] Loss 0.328264 Top1 83.691406 Top5 97.519531 -2023-02-13 18:16:20,493 - Epoch: [131][ 30/ 135] Loss 0.311239 Top1 84.283854 Top5 97.812500 -2023-02-13 18:16:20,638 - Epoch: [131][ 40/ 135] Loss 0.309917 Top1 84.785156 Top5 97.998047 -2023-02-13 18:16:20,776 - Epoch: [131][ 50/ 135] Loss 0.312759 Top1 84.679688 Top5 98.023438 -2023-02-13 18:16:20,910 - Epoch: [131][ 60/ 135] Loss 0.311181 Top1 84.765625 Top5 97.994792 -2023-02-13 18:16:21,037 - Epoch: [131][ 70/ 135] Loss 0.316715 Top1 84.592634 Top5 97.935268 -2023-02-13 18:16:21,166 - Epoch: [131][ 80/ 135] Loss 0.316397 Top1 84.560547 Top5 97.866211 -2023-02-13 18:16:21,295 - Epoch: [131][ 90/ 135] Loss 0.308978 Top1 84.678819 Top5 97.855903 -2023-02-13 18:16:21,427 - Epoch: [131][ 100/ 135] Loss 0.309770 Top1 84.625000 Top5 97.843750 -2023-02-13 18:16:21,555 - Epoch: [131][ 110/ 135] Loss 0.309953 Top1 84.609375 Top5 97.837358 -2023-02-13 18:16:21,686 - Epoch: [131][ 120/ 135] Loss 0.310321 Top1 84.661458 Top5 97.867839 -2023-02-13 18:16:21,820 - Epoch: [131][ 130/ 135] Loss 0.307600 Top1 84.837740 Top5 97.878606 -2023-02-13 18:16:21,867 - Epoch: [131][ 135/ 135] Loss 0.307889 Top1 84.827023 Top5 97.843257 -2023-02-13 18:16:21,940 - ==> Top1: 84.827 Top5: 97.843 Loss: 0.308 - -2023-02-13 18:16:21,941 - ==> Confusion: -[[ 836 4 3 0 16 3 0 2 6 59 0 2 3 5 7 3 4 3 2 1 8] - [ 2 955 3 2 6 20 3 12 4 0 1 1 2 0 0 3 5 0 5 2 7] - [ 8 5 945 10 5 1 19 15 0 2 5 2 2 6 7 9 3 1 2 4 7] - [ 4 3 16 895 2 5 1 2 0 3 13 0 9 1 21 6 4 4 19 0 8] - [ 9 15 0 1 985 9 1 2 1 3 0 5 3 2 8 7 8 1 0 2 4] - [ 0 17 0 5 2 975 5 15 2 4 2 7 2 14 0 2 4 3 2 2 7] - [ 4 6 9 5 0 3 1042 7 0 1 1 2 0 2 0 6 1 2 1 4 3] - [ 2 5 14 2 1 24 7 933 0 1 0 6 2 1 0 0 0 1 11 9 5] - [ 8 2 1 2 1 0 1 2 920 34 7 1 0 7 11 3 0 1 5 0 3] - [ 56 2 3 0 7 1 0 3 34 872 2 3 0 15 3 2 0 1 2 0 6] - [ 1 2 3 14 0 1 5 7 16 0 967 3 0 10 2 0 3 0 9 1 7] - [ 3 3 2 0 2 23 0 9 0 1 0 905 25 5 0 4 3 9 3 7 1] - [ 0 0 3 5 0 4 0 2 1 1 0 31 869 1 3 12 3 12 2 2 8] - [ 1 3 2 0 8 11 1 4 12 14 9 3 2 932 2 4 4 2 0 2 8] - [ 3 3 2 18 4 3 0 2 18 7 2 0 4 1 1001 2 4 5 8 0 5] - [ 0 2 6 0 8 2 4 1 1 0 0 6 6 0 2 979 12 7 0 6 4] - [ 3 5 0 2 6 1 0 1 2 0 0 0 2 1 2 14 1005 2 1 4 10] - [ 3 3 1 6 1 2 1 0 1 3 1 11 7 1 1 17 0 980 1 3 8] - [ 2 8 7 8 2 2 1 27 3 0 3 3 2 0 7 1 2 0 1002 1 5] - [ 0 3 0 1 2 4 8 10 0 1 0 15 2 1 0 9 5 2 0 1078 7] - [ 116 279 199 119 128 203 92 152 94 100 156 113 273 242 139 119 302 97 198 285 10028]] - -2023-02-13 18:16:21,943 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:16:21,943 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:16:21,949 - - -2023-02-13 18:16:21,949 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:16:22,835 - Epoch: [132][ 10/ 1207] Overall Loss 0.238381 Objective Loss 0.238381 LR 0.000500 Time 0.088521 -2023-02-13 18:16:23,030 - Epoch: [132][ 20/ 1207] Overall Loss 0.240574 Objective Loss 0.240574 LR 0.000500 Time 0.053972 -2023-02-13 18:16:23,220 - Epoch: [132][ 30/ 1207] Overall Loss 0.238068 Objective Loss 0.238068 LR 0.000500 Time 0.042317 -2023-02-13 18:16:23,411 - Epoch: [132][ 40/ 1207] Overall Loss 0.231740 Objective Loss 0.231740 LR 0.000500 Time 0.036497 -2023-02-13 18:16:23,601 - Epoch: [132][ 50/ 1207] Overall Loss 0.231936 Objective Loss 0.231936 LR 0.000500 Time 0.032993 -2023-02-13 18:16:23,791 - Epoch: [132][ 60/ 1207] Overall Loss 0.233128 Objective Loss 0.233128 LR 0.000500 Time 0.030657 -2023-02-13 18:16:23,981 - Epoch: [132][ 70/ 1207] Overall Loss 0.230220 Objective Loss 0.230220 LR 0.000500 Time 0.028991 -2023-02-13 18:16:24,171 - Epoch: [132][ 80/ 1207] Overall Loss 0.232450 Objective Loss 0.232450 LR 0.000500 Time 0.027737 -2023-02-13 18:16:24,362 - Epoch: [132][ 90/ 1207] Overall Loss 0.233725 Objective Loss 0.233725 LR 0.000500 Time 0.026767 -2023-02-13 18:16:24,551 - Epoch: [132][ 100/ 1207] Overall Loss 0.232599 Objective Loss 0.232599 LR 0.000500 Time 0.025983 -2023-02-13 18:16:24,741 - Epoch: [132][ 110/ 1207] Overall Loss 0.237761 Objective Loss 0.237761 LR 0.000500 Time 0.025344 -2023-02-13 18:16:24,932 - Epoch: [132][ 120/ 1207] Overall Loss 0.237810 Objective Loss 0.237810 LR 0.000500 Time 0.024817 -2023-02-13 18:16:25,123 - Epoch: [132][ 130/ 1207] Overall Loss 0.237908 Objective Loss 0.237908 LR 0.000500 Time 0.024372 -2023-02-13 18:16:25,314 - Epoch: [132][ 140/ 1207] Overall Loss 0.238375 Objective Loss 0.238375 LR 0.000500 Time 0.023992 -2023-02-13 18:16:25,504 - Epoch: [132][ 150/ 1207] Overall Loss 0.239144 Objective Loss 0.239144 LR 0.000500 Time 0.023660 -2023-02-13 18:16:25,694 - Epoch: [132][ 160/ 1207] Overall Loss 0.240510 Objective Loss 0.240510 LR 0.000500 Time 0.023367 -2023-02-13 18:16:25,885 - Epoch: [132][ 170/ 1207] Overall Loss 0.239029 Objective Loss 0.239029 LR 0.000500 Time 0.023112 -2023-02-13 18:16:26,075 - Epoch: [132][ 180/ 1207] Overall Loss 0.238449 Objective Loss 0.238449 LR 0.000500 Time 0.022885 -2023-02-13 18:16:26,266 - Epoch: [132][ 190/ 1207] Overall Loss 0.239600 Objective Loss 0.239600 LR 0.000500 Time 0.022680 -2023-02-13 18:16:26,456 - Epoch: [132][ 200/ 1207] Overall Loss 0.239288 Objective Loss 0.239288 LR 0.000500 Time 0.022497 -2023-02-13 18:16:26,647 - Epoch: [132][ 210/ 1207] Overall Loss 0.239370 Objective Loss 0.239370 LR 0.000500 Time 0.022331 -2023-02-13 18:16:26,837 - Epoch: [132][ 220/ 1207] Overall Loss 0.239590 Objective Loss 0.239590 LR 0.000500 Time 0.022178 -2023-02-13 18:16:27,027 - Epoch: [132][ 230/ 1207] Overall Loss 0.238291 Objective Loss 0.238291 LR 0.000500 Time 0.022040 -2023-02-13 18:16:27,218 - Epoch: [132][ 240/ 1207] Overall Loss 0.237522 Objective Loss 0.237522 LR 0.000500 Time 0.021916 -2023-02-13 18:16:27,410 - Epoch: [132][ 250/ 1207] Overall Loss 0.237527 Objective Loss 0.237527 LR 0.000500 Time 0.021804 -2023-02-13 18:16:27,600 - Epoch: [132][ 260/ 1207] Overall Loss 0.237142 Objective Loss 0.237142 LR 0.000500 Time 0.021695 -2023-02-13 18:16:27,791 - Epoch: [132][ 270/ 1207] Overall Loss 0.236623 Objective Loss 0.236623 LR 0.000500 Time 0.021596 -2023-02-13 18:16:27,982 - Epoch: [132][ 280/ 1207] Overall Loss 0.236694 Objective Loss 0.236694 LR 0.000500 Time 0.021506 -2023-02-13 18:16:28,173 - Epoch: [132][ 290/ 1207] Overall Loss 0.235827 Objective Loss 0.235827 LR 0.000500 Time 0.021422 -2023-02-13 18:16:28,364 - Epoch: [132][ 300/ 1207] Overall Loss 0.235978 Objective Loss 0.235978 LR 0.000500 Time 0.021344 -2023-02-13 18:16:28,556 - Epoch: [132][ 310/ 1207] Overall Loss 0.236222 Objective Loss 0.236222 LR 0.000500 Time 0.021275 -2023-02-13 18:16:28,753 - Epoch: [132][ 320/ 1207] Overall Loss 0.236279 Objective Loss 0.236279 LR 0.000500 Time 0.021224 -2023-02-13 18:16:28,949 - Epoch: [132][ 330/ 1207] Overall Loss 0.235977 Objective Loss 0.235977 LR 0.000500 Time 0.021172 -2023-02-13 18:16:29,146 - Epoch: [132][ 340/ 1207] Overall Loss 0.236698 Objective Loss 0.236698 LR 0.000500 Time 0.021129 -2023-02-13 18:16:29,342 - Epoch: [132][ 350/ 1207] Overall Loss 0.237364 Objective Loss 0.237364 LR 0.000500 Time 0.021084 -2023-02-13 18:16:29,540 - Epoch: [132][ 360/ 1207] Overall Loss 0.237000 Objective Loss 0.237000 LR 0.000500 Time 0.021046 -2023-02-13 18:16:29,735 - Epoch: [132][ 370/ 1207] Overall Loss 0.237069 Objective Loss 0.237069 LR 0.000500 Time 0.021004 -2023-02-13 18:16:29,932 - Epoch: [132][ 380/ 1207] Overall Loss 0.237055 Objective Loss 0.237055 LR 0.000500 Time 0.020968 -2023-02-13 18:16:30,128 - Epoch: [132][ 390/ 1207] Overall Loss 0.236881 Objective Loss 0.236881 LR 0.000500 Time 0.020932 -2023-02-13 18:16:30,326 - Epoch: [132][ 400/ 1207] Overall Loss 0.236981 Objective Loss 0.236981 LR 0.000500 Time 0.020902 -2023-02-13 18:16:30,521 - Epoch: [132][ 410/ 1207] Overall Loss 0.236984 Objective Loss 0.236984 LR 0.000500 Time 0.020868 -2023-02-13 18:16:30,719 - Epoch: [132][ 420/ 1207] Overall Loss 0.237456 Objective Loss 0.237456 LR 0.000500 Time 0.020841 -2023-02-13 18:16:30,915 - Epoch: [132][ 430/ 1207] Overall Loss 0.237408 Objective Loss 0.237408 LR 0.000500 Time 0.020811 -2023-02-13 18:16:31,112 - Epoch: [132][ 440/ 1207] Overall Loss 0.237334 Objective Loss 0.237334 LR 0.000500 Time 0.020787 -2023-02-13 18:16:31,309 - Epoch: [132][ 450/ 1207] Overall Loss 0.237514 Objective Loss 0.237514 LR 0.000500 Time 0.020761 -2023-02-13 18:16:31,507 - Epoch: [132][ 460/ 1207] Overall Loss 0.238132 Objective Loss 0.238132 LR 0.000500 Time 0.020738 -2023-02-13 18:16:31,702 - Epoch: [132][ 470/ 1207] Overall Loss 0.238189 Objective Loss 0.238189 LR 0.000500 Time 0.020713 -2023-02-13 18:16:31,901 - Epoch: [132][ 480/ 1207] Overall Loss 0.237907 Objective Loss 0.237907 LR 0.000500 Time 0.020694 -2023-02-13 18:16:32,097 - Epoch: [132][ 490/ 1207] Overall Loss 0.237655 Objective Loss 0.237655 LR 0.000500 Time 0.020671 -2023-02-13 18:16:32,295 - Epoch: [132][ 500/ 1207] Overall Loss 0.237337 Objective Loss 0.237337 LR 0.000500 Time 0.020652 -2023-02-13 18:16:32,490 - Epoch: [132][ 510/ 1207] Overall Loss 0.237978 Objective Loss 0.237978 LR 0.000500 Time 0.020630 -2023-02-13 18:16:32,687 - Epoch: [132][ 520/ 1207] Overall Loss 0.238153 Objective Loss 0.238153 LR 0.000500 Time 0.020612 -2023-02-13 18:16:32,883 - Epoch: [132][ 530/ 1207] Overall Loss 0.237900 Objective Loss 0.237900 LR 0.000500 Time 0.020591 -2023-02-13 18:16:33,081 - Epoch: [132][ 540/ 1207] Overall Loss 0.237891 Objective Loss 0.237891 LR 0.000500 Time 0.020575 -2023-02-13 18:16:33,276 - Epoch: [132][ 550/ 1207] Overall Loss 0.237926 Objective Loss 0.237926 LR 0.000500 Time 0.020556 -2023-02-13 18:16:33,474 - Epoch: [132][ 560/ 1207] Overall Loss 0.237837 Objective Loss 0.237837 LR 0.000500 Time 0.020541 -2023-02-13 18:16:33,669 - Epoch: [132][ 570/ 1207] Overall Loss 0.237749 Objective Loss 0.237749 LR 0.000500 Time 0.020522 -2023-02-13 18:16:33,866 - Epoch: [132][ 580/ 1207] Overall Loss 0.237576 Objective Loss 0.237576 LR 0.000500 Time 0.020508 -2023-02-13 18:16:34,062 - Epoch: [132][ 590/ 1207] Overall Loss 0.237427 Objective Loss 0.237427 LR 0.000500 Time 0.020492 -2023-02-13 18:16:34,260 - Epoch: [132][ 600/ 1207] Overall Loss 0.237608 Objective Loss 0.237608 LR 0.000500 Time 0.020479 -2023-02-13 18:16:34,455 - Epoch: [132][ 610/ 1207] Overall Loss 0.237486 Objective Loss 0.237486 LR 0.000500 Time 0.020463 -2023-02-13 18:16:34,652 - Epoch: [132][ 620/ 1207] Overall Loss 0.237255 Objective Loss 0.237255 LR 0.000500 Time 0.020451 -2023-02-13 18:16:34,848 - Epoch: [132][ 630/ 1207] Overall Loss 0.237562 Objective Loss 0.237562 LR 0.000500 Time 0.020436 -2023-02-13 18:16:35,045 - Epoch: [132][ 640/ 1207] Overall Loss 0.237722 Objective Loss 0.237722 LR 0.000500 Time 0.020424 -2023-02-13 18:16:35,241 - Epoch: [132][ 650/ 1207] Overall Loss 0.237143 Objective Loss 0.237143 LR 0.000500 Time 0.020410 -2023-02-13 18:16:35,438 - Epoch: [132][ 660/ 1207] Overall Loss 0.237403 Objective Loss 0.237403 LR 0.000500 Time 0.020399 -2023-02-13 18:16:35,634 - Epoch: [132][ 670/ 1207] Overall Loss 0.237731 Objective Loss 0.237731 LR 0.000500 Time 0.020386 -2023-02-13 18:16:35,831 - Epoch: [132][ 680/ 1207] Overall Loss 0.237779 Objective Loss 0.237779 LR 0.000500 Time 0.020376 -2023-02-13 18:16:36,028 - Epoch: [132][ 690/ 1207] Overall Loss 0.237364 Objective Loss 0.237364 LR 0.000500 Time 0.020366 -2023-02-13 18:16:36,225 - Epoch: [132][ 700/ 1207] Overall Loss 0.237168 Objective Loss 0.237168 LR 0.000500 Time 0.020356 -2023-02-13 18:16:36,421 - Epoch: [132][ 710/ 1207] Overall Loss 0.237521 Objective Loss 0.237521 LR 0.000500 Time 0.020345 -2023-02-13 18:16:36,619 - Epoch: [132][ 720/ 1207] Overall Loss 0.237311 Objective Loss 0.237311 LR 0.000500 Time 0.020336 -2023-02-13 18:16:36,814 - Epoch: [132][ 730/ 1207] Overall Loss 0.237202 Objective Loss 0.237202 LR 0.000500 Time 0.020324 -2023-02-13 18:16:37,012 - Epoch: [132][ 740/ 1207] Overall Loss 0.237159 Objective Loss 0.237159 LR 0.000500 Time 0.020317 -2023-02-13 18:16:37,210 - Epoch: [132][ 750/ 1207] Overall Loss 0.237233 Objective Loss 0.237233 LR 0.000500 Time 0.020310 -2023-02-13 18:16:37,410 - Epoch: [132][ 760/ 1207] Overall Loss 0.237072 Objective Loss 0.237072 LR 0.000500 Time 0.020305 -2023-02-13 18:16:37,609 - Epoch: [132][ 770/ 1207] Overall Loss 0.237220 Objective Loss 0.237220 LR 0.000500 Time 0.020300 -2023-02-13 18:16:37,809 - Epoch: [132][ 780/ 1207] Overall Loss 0.237168 Objective Loss 0.237168 LR 0.000500 Time 0.020295 -2023-02-13 18:16:38,008 - Epoch: [132][ 790/ 1207] Overall Loss 0.237727 Objective Loss 0.237727 LR 0.000500 Time 0.020290 -2023-02-13 18:16:38,209 - Epoch: [132][ 800/ 1207] Overall Loss 0.238111 Objective Loss 0.238111 LR 0.000500 Time 0.020286 -2023-02-13 18:16:38,407 - Epoch: [132][ 810/ 1207] Overall Loss 0.238470 Objective Loss 0.238470 LR 0.000500 Time 0.020280 -2023-02-13 18:16:38,606 - Epoch: [132][ 820/ 1207] Overall Loss 0.238561 Objective Loss 0.238561 LR 0.000500 Time 0.020274 -2023-02-13 18:16:38,805 - Epoch: [132][ 830/ 1207] Overall Loss 0.238494 Objective Loss 0.238494 LR 0.000500 Time 0.020269 -2023-02-13 18:16:39,005 - Epoch: [132][ 840/ 1207] Overall Loss 0.238606 Objective Loss 0.238606 LR 0.000500 Time 0.020266 -2023-02-13 18:16:39,203 - Epoch: [132][ 850/ 1207] Overall Loss 0.238851 Objective Loss 0.238851 LR 0.000500 Time 0.020260 -2023-02-13 18:16:39,403 - Epoch: [132][ 860/ 1207] Overall Loss 0.238734 Objective Loss 0.238734 LR 0.000500 Time 0.020257 -2023-02-13 18:16:39,600 - Epoch: [132][ 870/ 1207] Overall Loss 0.238752 Objective Loss 0.238752 LR 0.000500 Time 0.020250 -2023-02-13 18:16:39,799 - Epoch: [132][ 880/ 1207] Overall Loss 0.238509 Objective Loss 0.238509 LR 0.000500 Time 0.020245 -2023-02-13 18:16:39,996 - Epoch: [132][ 890/ 1207] Overall Loss 0.238726 Objective Loss 0.238726 LR 0.000500 Time 0.020239 -2023-02-13 18:16:40,196 - Epoch: [132][ 900/ 1207] Overall Loss 0.238800 Objective Loss 0.238800 LR 0.000500 Time 0.020236 -2023-02-13 18:16:40,391 - Epoch: [132][ 910/ 1207] Overall Loss 0.238596 Objective Loss 0.238596 LR 0.000500 Time 0.020227 -2023-02-13 18:16:40,583 - Epoch: [132][ 920/ 1207] Overall Loss 0.238473 Objective Loss 0.238473 LR 0.000500 Time 0.020216 -2023-02-13 18:16:40,775 - Epoch: [132][ 930/ 1207] Overall Loss 0.238424 Objective Loss 0.238424 LR 0.000500 Time 0.020204 -2023-02-13 18:16:40,969 - Epoch: [132][ 940/ 1207] Overall Loss 0.238577 Objective Loss 0.238577 LR 0.000500 Time 0.020195 -2023-02-13 18:16:41,161 - Epoch: [132][ 950/ 1207] Overall Loss 0.238806 Objective Loss 0.238806 LR 0.000500 Time 0.020185 -2023-02-13 18:16:41,354 - Epoch: [132][ 960/ 1207] Overall Loss 0.239074 Objective Loss 0.239074 LR 0.000500 Time 0.020175 -2023-02-13 18:16:41,546 - Epoch: [132][ 970/ 1207] Overall Loss 0.238952 Objective Loss 0.238952 LR 0.000500 Time 0.020165 -2023-02-13 18:16:41,738 - Epoch: [132][ 980/ 1207] Overall Loss 0.238635 Objective Loss 0.238635 LR 0.000500 Time 0.020155 -2023-02-13 18:16:41,931 - Epoch: [132][ 990/ 1207] Overall Loss 0.238611 Objective Loss 0.238611 LR 0.000500 Time 0.020145 -2023-02-13 18:16:42,123 - Epoch: [132][ 1000/ 1207] Overall Loss 0.238521 Objective Loss 0.238521 LR 0.000500 Time 0.020136 -2023-02-13 18:16:42,315 - Epoch: [132][ 1010/ 1207] Overall Loss 0.238615 Objective Loss 0.238615 LR 0.000500 Time 0.020126 -2023-02-13 18:16:42,507 - Epoch: [132][ 1020/ 1207] Overall Loss 0.238797 Objective Loss 0.238797 LR 0.000500 Time 0.020117 -2023-02-13 18:16:42,700 - Epoch: [132][ 1030/ 1207] Overall Loss 0.238529 Objective Loss 0.238529 LR 0.000500 Time 0.020108 -2023-02-13 18:16:42,893 - Epoch: [132][ 1040/ 1207] Overall Loss 0.238441 Objective Loss 0.238441 LR 0.000500 Time 0.020100 -2023-02-13 18:16:43,083 - Epoch: [132][ 1050/ 1207] Overall Loss 0.238603 Objective Loss 0.238603 LR 0.000500 Time 0.020089 -2023-02-13 18:16:43,274 - Epoch: [132][ 1060/ 1207] Overall Loss 0.238583 Objective Loss 0.238583 LR 0.000500 Time 0.020079 -2023-02-13 18:16:43,464 - Epoch: [132][ 1070/ 1207] Overall Loss 0.238364 Objective Loss 0.238364 LR 0.000500 Time 0.020069 -2023-02-13 18:16:43,655 - Epoch: [132][ 1080/ 1207] Overall Loss 0.238227 Objective Loss 0.238227 LR 0.000500 Time 0.020060 -2023-02-13 18:16:43,845 - Epoch: [132][ 1090/ 1207] Overall Loss 0.238079 Objective Loss 0.238079 LR 0.000500 Time 0.020050 -2023-02-13 18:16:44,036 - Epoch: [132][ 1100/ 1207] Overall Loss 0.238326 Objective Loss 0.238326 LR 0.000500 Time 0.020040 -2023-02-13 18:16:44,226 - Epoch: [132][ 1110/ 1207] Overall Loss 0.238748 Objective Loss 0.238748 LR 0.000500 Time 0.020031 -2023-02-13 18:16:44,417 - Epoch: [132][ 1120/ 1207] Overall Loss 0.238878 Objective Loss 0.238878 LR 0.000500 Time 0.020022 -2023-02-13 18:16:44,608 - Epoch: [132][ 1130/ 1207] Overall Loss 0.239056 Objective Loss 0.239056 LR 0.000500 Time 0.020013 -2023-02-13 18:16:44,798 - Epoch: [132][ 1140/ 1207] Overall Loss 0.239162 Objective Loss 0.239162 LR 0.000500 Time 0.020005 -2023-02-13 18:16:44,989 - Epoch: [132][ 1150/ 1207] Overall Loss 0.239349 Objective Loss 0.239349 LR 0.000500 Time 0.019996 -2023-02-13 18:16:45,180 - Epoch: [132][ 1160/ 1207] Overall Loss 0.239147 Objective Loss 0.239147 LR 0.000500 Time 0.019988 -2023-02-13 18:16:45,371 - Epoch: [132][ 1170/ 1207] Overall Loss 0.239253 Objective Loss 0.239253 LR 0.000500 Time 0.019980 -2023-02-13 18:16:45,562 - Epoch: [132][ 1180/ 1207] Overall Loss 0.239035 Objective Loss 0.239035 LR 0.000500 Time 0.019972 -2023-02-13 18:16:45,752 - Epoch: [132][ 1190/ 1207] Overall Loss 0.239167 Objective Loss 0.239167 LR 0.000500 Time 0.019964 -2023-02-13 18:16:45,995 - Epoch: [132][ 1200/ 1207] Overall Loss 0.239122 Objective Loss 0.239122 LR 0.000500 Time 0.020000 -2023-02-13 18:16:46,112 - Epoch: [132][ 1207/ 1207] Overall Loss 0.239048 Objective Loss 0.239048 Top1 84.756098 Top5 97.560976 LR 0.000500 Time 0.019980 -2023-02-13 18:16:46,183 - --- validate (epoch=132)----------- -2023-02-13 18:16:46,184 - 34311 samples (256 per mini-batch) -2023-02-13 18:16:46,697 - Epoch: [132][ 10/ 135] Loss 0.284330 Top1 85.195312 Top5 98.085938 -2023-02-13 18:16:46,826 - Epoch: [132][ 20/ 135] Loss 0.300034 Top1 85.683594 Top5 97.910156 -2023-02-13 18:16:46,957 - Epoch: [132][ 30/ 135] Loss 0.295427 Top1 85.625000 Top5 97.903646 -2023-02-13 18:16:47,086 - Epoch: [132][ 40/ 135] Loss 0.304292 Top1 85.419922 Top5 97.763672 -2023-02-13 18:16:47,216 - Epoch: [132][ 50/ 135] Loss 0.302375 Top1 85.507812 Top5 97.625000 -2023-02-13 18:16:47,346 - Epoch: [132][ 60/ 135] Loss 0.307558 Top1 85.436198 Top5 97.617188 -2023-02-13 18:16:47,476 - Epoch: [132][ 70/ 135] Loss 0.308419 Top1 85.312500 Top5 97.650670 -2023-02-13 18:16:47,605 - Epoch: [132][ 80/ 135] Loss 0.306260 Top1 85.390625 Top5 97.714844 -2023-02-13 18:16:47,736 - Epoch: [132][ 90/ 135] Loss 0.307002 Top1 85.399306 Top5 97.730035 -2023-02-13 18:16:47,864 - Epoch: [132][ 100/ 135] Loss 0.308991 Top1 85.304688 Top5 97.718750 -2023-02-13 18:16:47,990 - Epoch: [132][ 110/ 135] Loss 0.306681 Top1 85.248580 Top5 97.780540 -2023-02-13 18:16:48,118 - Epoch: [132][ 120/ 135] Loss 0.308286 Top1 85.205078 Top5 97.753906 -2023-02-13 18:16:48,246 - Epoch: [132][ 130/ 135] Loss 0.309532 Top1 85.159255 Top5 97.746394 -2023-02-13 18:16:48,293 - Epoch: [132][ 135/ 135] Loss 0.307885 Top1 85.112646 Top5 97.749993 -2023-02-13 18:16:48,362 - ==> Top1: 85.113 Top5: 97.750 Loss: 0.308 - -2023-02-13 18:16:48,362 - ==> Confusion: -[[ 859 4 8 0 11 2 0 2 5 43 1 4 0 4 7 2 0 1 2 3 9] - [ 1 926 0 1 13 36 2 20 3 1 2 1 2 1 1 2 7 2 2 3 7] - [ 8 2 963 5 5 2 12 16 0 2 4 1 4 8 4 6 2 2 2 5 5] - [ 7 0 20 909 4 4 1 2 2 2 15 1 7 0 17 1 4 3 11 0 6] - [ 13 6 0 1 1000 9 1 1 0 0 0 5 2 9 5 5 4 1 0 1 3] - [ 1 15 3 2 7 968 2 16 1 2 1 16 4 18 0 2 4 0 1 3 4] - [ 2 3 14 1 0 8 1031 8 0 3 0 2 1 2 0 3 1 5 2 8 5] - [ 1 7 10 2 1 31 3 931 1 1 2 6 3 3 0 1 1 2 10 5 3] - [ 17 2 1 1 1 1 0 3 884 43 10 2 1 12 20 1 2 0 6 0 2] - [ 83 0 2 1 8 2 0 2 25 847 1 2 2 20 4 0 2 3 1 1 6] - [ 2 2 5 9 0 1 1 6 15 1 974 3 2 14 3 0 1 0 7 0 5] - [ 0 3 1 0 3 4 0 6 1 1 1 938 17 6 1 3 2 7 3 5 3] - [ 0 0 1 7 0 2 0 0 1 3 1 34 865 2 5 9 2 13 4 2 8] - [ 4 3 0 0 9 5 1 4 8 12 8 3 2 952 3 2 1 3 0 1 3] - [ 7 4 1 23 6 3 0 1 14 5 4 3 1 2 991 1 4 6 7 0 9] - [ 3 3 6 0 8 0 1 1 0 1 0 11 10 3 3 971 2 11 0 6 6] - [ 2 5 0 1 10 2 0 0 2 0 1 0 4 4 1 8 1007 1 1 4 8] - [ 2 2 0 4 0 0 1 0 0 1 1 13 9 1 0 18 0 993 1 0 5] - [ 1 3 6 8 2 1 0 29 5 0 7 1 4 1 9 0 2 4 994 3 6] - [ 0 4 2 1 3 6 7 11 1 0 1 21 7 5 0 6 7 2 0 1052 12] - [ 132 182 228 118 166 202 86 195 73 65 180 128 270 320 150 89 253 111 143 195 10148]] - -2023-02-13 18:16:48,364 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:16:48,364 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:16:48,370 - - -2023-02-13 18:16:48,370 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:16:49,270 - Epoch: [133][ 10/ 1207] Overall Loss 0.207495 Objective Loss 0.207495 LR 0.000500 Time 0.089899 -2023-02-13 18:16:49,459 - Epoch: [133][ 20/ 1207] Overall Loss 0.225635 Objective Loss 0.225635 LR 0.000500 Time 0.054401 -2023-02-13 18:16:49,647 - Epoch: [133][ 30/ 1207] Overall Loss 0.226988 Objective Loss 0.226988 LR 0.000500 Time 0.042511 -2023-02-13 18:16:49,834 - Epoch: [133][ 40/ 1207] Overall Loss 0.227813 Objective Loss 0.227813 LR 0.000500 Time 0.036548 -2023-02-13 18:16:50,021 - Epoch: [133][ 50/ 1207] Overall Loss 0.225713 Objective Loss 0.225713 LR 0.000500 Time 0.032977 -2023-02-13 18:16:50,208 - Epoch: [133][ 60/ 1207] Overall Loss 0.231754 Objective Loss 0.231754 LR 0.000500 Time 0.030595 -2023-02-13 18:16:50,396 - Epoch: [133][ 70/ 1207] Overall Loss 0.234631 Objective Loss 0.234631 LR 0.000500 Time 0.028902 -2023-02-13 18:16:50,583 - Epoch: [133][ 80/ 1207] Overall Loss 0.236305 Objective Loss 0.236305 LR 0.000500 Time 0.027621 -2023-02-13 18:16:50,770 - Epoch: [133][ 90/ 1207] Overall Loss 0.233597 Objective Loss 0.233597 LR 0.000500 Time 0.026624 -2023-02-13 18:16:50,958 - Epoch: [133][ 100/ 1207] Overall Loss 0.232670 Objective Loss 0.232670 LR 0.000500 Time 0.025839 -2023-02-13 18:16:51,144 - Epoch: [133][ 110/ 1207] Overall Loss 0.232449 Objective Loss 0.232449 LR 0.000500 Time 0.025184 -2023-02-13 18:16:51,332 - Epoch: [133][ 120/ 1207] Overall Loss 0.231858 Objective Loss 0.231858 LR 0.000500 Time 0.024641 -2023-02-13 18:16:51,519 - Epoch: [133][ 130/ 1207] Overall Loss 0.231791 Objective Loss 0.231791 LR 0.000500 Time 0.024183 -2023-02-13 18:16:51,706 - Epoch: [133][ 140/ 1207] Overall Loss 0.234113 Objective Loss 0.234113 LR 0.000500 Time 0.023789 -2023-02-13 18:16:51,893 - Epoch: [133][ 150/ 1207] Overall Loss 0.233508 Objective Loss 0.233508 LR 0.000500 Time 0.023446 -2023-02-13 18:16:52,080 - Epoch: [133][ 160/ 1207] Overall Loss 0.233737 Objective Loss 0.233737 LR 0.000500 Time 0.023151 -2023-02-13 18:16:52,268 - Epoch: [133][ 170/ 1207] Overall Loss 0.236203 Objective Loss 0.236203 LR 0.000500 Time 0.022889 -2023-02-13 18:16:52,455 - Epoch: [133][ 180/ 1207] Overall Loss 0.237001 Objective Loss 0.237001 LR 0.000500 Time 0.022658 -2023-02-13 18:16:52,643 - Epoch: [133][ 190/ 1207] Overall Loss 0.236638 Objective Loss 0.236638 LR 0.000500 Time 0.022449 -2023-02-13 18:16:52,830 - Epoch: [133][ 200/ 1207] Overall Loss 0.237023 Objective Loss 0.237023 LR 0.000500 Time 0.022260 -2023-02-13 18:16:53,017 - Epoch: [133][ 210/ 1207] Overall Loss 0.237860 Objective Loss 0.237860 LR 0.000500 Time 0.022089 -2023-02-13 18:16:53,204 - Epoch: [133][ 220/ 1207] Overall Loss 0.238572 Objective Loss 0.238572 LR 0.000500 Time 0.021933 -2023-02-13 18:16:53,391 - Epoch: [133][ 230/ 1207] Overall Loss 0.239662 Objective Loss 0.239662 LR 0.000500 Time 0.021792 -2023-02-13 18:16:53,578 - Epoch: [133][ 240/ 1207] Overall Loss 0.239205 Objective Loss 0.239205 LR 0.000500 Time 0.021662 -2023-02-13 18:16:53,765 - Epoch: [133][ 250/ 1207] Overall Loss 0.239885 Objective Loss 0.239885 LR 0.000500 Time 0.021541 -2023-02-13 18:16:53,952 - Epoch: [133][ 260/ 1207] Overall Loss 0.238901 Objective Loss 0.238901 LR 0.000500 Time 0.021432 -2023-02-13 18:16:54,140 - Epoch: [133][ 270/ 1207] Overall Loss 0.239652 Objective Loss 0.239652 LR 0.000500 Time 0.021332 -2023-02-13 18:16:54,327 - Epoch: [133][ 280/ 1207] Overall Loss 0.240258 Objective Loss 0.240258 LR 0.000500 Time 0.021239 -2023-02-13 18:16:54,515 - Epoch: [133][ 290/ 1207] Overall Loss 0.239624 Objective Loss 0.239624 LR 0.000500 Time 0.021152 -2023-02-13 18:16:54,702 - Epoch: [133][ 300/ 1207] Overall Loss 0.239407 Objective Loss 0.239407 LR 0.000500 Time 0.021070 -2023-02-13 18:16:54,890 - Epoch: [133][ 310/ 1207] Overall Loss 0.238736 Objective Loss 0.238736 LR 0.000500 Time 0.020994 -2023-02-13 18:16:55,078 - Epoch: [133][ 320/ 1207] Overall Loss 0.238623 Objective Loss 0.238623 LR 0.000500 Time 0.020924 -2023-02-13 18:16:55,265 - Epoch: [133][ 330/ 1207] Overall Loss 0.238961 Objective Loss 0.238961 LR 0.000500 Time 0.020855 -2023-02-13 18:16:55,453 - Epoch: [133][ 340/ 1207] Overall Loss 0.238398 Objective Loss 0.238398 LR 0.000500 Time 0.020794 -2023-02-13 18:16:55,640 - Epoch: [133][ 350/ 1207] Overall Loss 0.238460 Objective Loss 0.238460 LR 0.000500 Time 0.020733 -2023-02-13 18:16:55,827 - Epoch: [133][ 360/ 1207] Overall Loss 0.238162 Objective Loss 0.238162 LR 0.000500 Time 0.020676 -2023-02-13 18:16:56,015 - Epoch: [133][ 370/ 1207] Overall Loss 0.238188 Objective Loss 0.238188 LR 0.000500 Time 0.020625 -2023-02-13 18:16:56,202 - Epoch: [133][ 380/ 1207] Overall Loss 0.238245 Objective Loss 0.238245 LR 0.000500 Time 0.020574 -2023-02-13 18:16:56,390 - Epoch: [133][ 390/ 1207] Overall Loss 0.238488 Objective Loss 0.238488 LR 0.000500 Time 0.020527 -2023-02-13 18:16:56,578 - Epoch: [133][ 400/ 1207] Overall Loss 0.238920 Objective Loss 0.238920 LR 0.000500 Time 0.020482 -2023-02-13 18:16:56,765 - Epoch: [133][ 410/ 1207] Overall Loss 0.239154 Objective Loss 0.239154 LR 0.000500 Time 0.020438 -2023-02-13 18:16:56,953 - Epoch: [133][ 420/ 1207] Overall Loss 0.239333 Objective Loss 0.239333 LR 0.000500 Time 0.020398 -2023-02-13 18:16:57,140 - Epoch: [133][ 430/ 1207] Overall Loss 0.239465 Objective Loss 0.239465 LR 0.000500 Time 0.020358 -2023-02-13 18:16:57,328 - Epoch: [133][ 440/ 1207] Overall Loss 0.239081 Objective Loss 0.239081 LR 0.000500 Time 0.020321 -2023-02-13 18:16:57,516 - Epoch: [133][ 450/ 1207] Overall Loss 0.239200 Objective Loss 0.239200 LR 0.000500 Time 0.020287 -2023-02-13 18:16:57,704 - Epoch: [133][ 460/ 1207] Overall Loss 0.239053 Objective Loss 0.239053 LR 0.000500 Time 0.020253 -2023-02-13 18:16:57,891 - Epoch: [133][ 470/ 1207] Overall Loss 0.239551 Objective Loss 0.239551 LR 0.000500 Time 0.020221 -2023-02-13 18:16:58,079 - Epoch: [133][ 480/ 1207] Overall Loss 0.239635 Objective Loss 0.239635 LR 0.000500 Time 0.020190 -2023-02-13 18:16:58,267 - Epoch: [133][ 490/ 1207] Overall Loss 0.239019 Objective Loss 0.239019 LR 0.000500 Time 0.020160 -2023-02-13 18:16:58,455 - Epoch: [133][ 500/ 1207] Overall Loss 0.239315 Objective Loss 0.239315 LR 0.000500 Time 0.020133 -2023-02-13 18:16:58,643 - Epoch: [133][ 510/ 1207] Overall Loss 0.239082 Objective Loss 0.239082 LR 0.000500 Time 0.020106 -2023-02-13 18:16:58,831 - Epoch: [133][ 520/ 1207] Overall Loss 0.239596 Objective Loss 0.239596 LR 0.000500 Time 0.020079 -2023-02-13 18:16:59,019 - Epoch: [133][ 530/ 1207] Overall Loss 0.239369 Objective Loss 0.239369 LR 0.000500 Time 0.020054 -2023-02-13 18:16:59,206 - Epoch: [133][ 540/ 1207] Overall Loss 0.239402 Objective Loss 0.239402 LR 0.000500 Time 0.020030 -2023-02-13 18:16:59,394 - Epoch: [133][ 550/ 1207] Overall Loss 0.239595 Objective Loss 0.239595 LR 0.000500 Time 0.020007 -2023-02-13 18:16:59,582 - Epoch: [133][ 560/ 1207] Overall Loss 0.240029 Objective Loss 0.240029 LR 0.000500 Time 0.019984 -2023-02-13 18:16:59,769 - Epoch: [133][ 570/ 1207] Overall Loss 0.239670 Objective Loss 0.239670 LR 0.000500 Time 0.019961 -2023-02-13 18:16:59,957 - Epoch: [133][ 580/ 1207] Overall Loss 0.239428 Objective Loss 0.239428 LR 0.000500 Time 0.019940 -2023-02-13 18:17:00,144 - Epoch: [133][ 590/ 1207] Overall Loss 0.239321 Objective Loss 0.239321 LR 0.000500 Time 0.019919 -2023-02-13 18:17:00,332 - Epoch: [133][ 600/ 1207] Overall Loss 0.238616 Objective Loss 0.238616 LR 0.000500 Time 0.019899 -2023-02-13 18:17:00,520 - Epoch: [133][ 610/ 1207] Overall Loss 0.238931 Objective Loss 0.238931 LR 0.000500 Time 0.019881 -2023-02-13 18:17:00,707 - Epoch: [133][ 620/ 1207] Overall Loss 0.238992 Objective Loss 0.238992 LR 0.000500 Time 0.019862 -2023-02-13 18:17:00,896 - Epoch: [133][ 630/ 1207] Overall Loss 0.239081 Objective Loss 0.239081 LR 0.000500 Time 0.019846 -2023-02-13 18:17:01,085 - Epoch: [133][ 640/ 1207] Overall Loss 0.238625 Objective Loss 0.238625 LR 0.000500 Time 0.019829 -2023-02-13 18:17:01,272 - Epoch: [133][ 650/ 1207] Overall Loss 0.238456 Objective Loss 0.238456 LR 0.000500 Time 0.019812 -2023-02-13 18:17:01,460 - Epoch: [133][ 660/ 1207] Overall Loss 0.238430 Objective Loss 0.238430 LR 0.000500 Time 0.019796 -2023-02-13 18:17:01,647 - Epoch: [133][ 670/ 1207] Overall Loss 0.238345 Objective Loss 0.238345 LR 0.000500 Time 0.019780 -2023-02-13 18:17:01,835 - Epoch: [133][ 680/ 1207] Overall Loss 0.238102 Objective Loss 0.238102 LR 0.000500 Time 0.019765 -2023-02-13 18:17:02,024 - Epoch: [133][ 690/ 1207] Overall Loss 0.238443 Objective Loss 0.238443 LR 0.000500 Time 0.019751 -2023-02-13 18:17:02,211 - Epoch: [133][ 700/ 1207] Overall Loss 0.238294 Objective Loss 0.238294 LR 0.000500 Time 0.019736 -2023-02-13 18:17:02,400 - Epoch: [133][ 710/ 1207] Overall Loss 0.238771 Objective Loss 0.238771 LR 0.000500 Time 0.019723 -2023-02-13 18:17:02,587 - Epoch: [133][ 720/ 1207] Overall Loss 0.238792 Objective Loss 0.238792 LR 0.000500 Time 0.019709 -2023-02-13 18:17:02,775 - Epoch: [133][ 730/ 1207] Overall Loss 0.238722 Objective Loss 0.238722 LR 0.000500 Time 0.019695 -2023-02-13 18:17:02,962 - Epoch: [133][ 740/ 1207] Overall Loss 0.238760 Objective Loss 0.238760 LR 0.000500 Time 0.019682 -2023-02-13 18:17:03,150 - Epoch: [133][ 750/ 1207] Overall Loss 0.238833 Objective Loss 0.238833 LR 0.000500 Time 0.019669 -2023-02-13 18:17:03,338 - Epoch: [133][ 760/ 1207] Overall Loss 0.238431 Objective Loss 0.238431 LR 0.000500 Time 0.019657 -2023-02-13 18:17:03,526 - Epoch: [133][ 770/ 1207] Overall Loss 0.238800 Objective Loss 0.238800 LR 0.000500 Time 0.019646 -2023-02-13 18:17:03,714 - Epoch: [133][ 780/ 1207] Overall Loss 0.239191 Objective Loss 0.239191 LR 0.000500 Time 0.019635 -2023-02-13 18:17:03,902 - Epoch: [133][ 790/ 1207] Overall Loss 0.239230 Objective Loss 0.239230 LR 0.000500 Time 0.019624 -2023-02-13 18:17:04,090 - Epoch: [133][ 800/ 1207] Overall Loss 0.239048 Objective Loss 0.239048 LR 0.000500 Time 0.019613 -2023-02-13 18:17:04,277 - Epoch: [133][ 810/ 1207] Overall Loss 0.239098 Objective Loss 0.239098 LR 0.000500 Time 0.019601 -2023-02-13 18:17:04,466 - Epoch: [133][ 820/ 1207] Overall Loss 0.239447 Objective Loss 0.239447 LR 0.000500 Time 0.019592 -2023-02-13 18:17:04,653 - Epoch: [133][ 830/ 1207] Overall Loss 0.239508 Objective Loss 0.239508 LR 0.000500 Time 0.019581 -2023-02-13 18:17:04,841 - Epoch: [133][ 840/ 1207] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.000500 Time 0.019571 -2023-02-13 18:17:05,029 - Epoch: [133][ 850/ 1207] Overall Loss 0.239446 Objective Loss 0.239446 LR 0.000500 Time 0.019562 -2023-02-13 18:17:05,217 - Epoch: [133][ 860/ 1207] Overall Loss 0.239268 Objective Loss 0.239268 LR 0.000500 Time 0.019553 -2023-02-13 18:17:05,405 - Epoch: [133][ 870/ 1207] Overall Loss 0.239392 Objective Loss 0.239392 LR 0.000500 Time 0.019544 -2023-02-13 18:17:05,593 - Epoch: [133][ 880/ 1207] Overall Loss 0.239130 Objective Loss 0.239130 LR 0.000500 Time 0.019534 -2023-02-13 18:17:05,780 - Epoch: [133][ 890/ 1207] Overall Loss 0.239061 Objective Loss 0.239061 LR 0.000500 Time 0.019525 -2023-02-13 18:17:05,969 - Epoch: [133][ 900/ 1207] Overall Loss 0.238886 Objective Loss 0.238886 LR 0.000500 Time 0.019517 -2023-02-13 18:17:06,157 - Epoch: [133][ 910/ 1207] Overall Loss 0.238891 Objective Loss 0.238891 LR 0.000500 Time 0.019509 -2023-02-13 18:17:06,344 - Epoch: [133][ 920/ 1207] Overall Loss 0.238786 Objective Loss 0.238786 LR 0.000500 Time 0.019500 -2023-02-13 18:17:06,532 - Epoch: [133][ 930/ 1207] Overall Loss 0.238730 Objective Loss 0.238730 LR 0.000500 Time 0.019492 -2023-02-13 18:17:06,721 - Epoch: [133][ 940/ 1207] Overall Loss 0.238934 Objective Loss 0.238934 LR 0.000500 Time 0.019486 -2023-02-13 18:17:06,911 - Epoch: [133][ 950/ 1207] Overall Loss 0.238980 Objective Loss 0.238980 LR 0.000500 Time 0.019479 -2023-02-13 18:17:07,100 - Epoch: [133][ 960/ 1207] Overall Loss 0.238708 Objective Loss 0.238708 LR 0.000500 Time 0.019473 -2023-02-13 18:17:07,288 - Epoch: [133][ 970/ 1207] Overall Loss 0.238948 Objective Loss 0.238948 LR 0.000500 Time 0.019466 -2023-02-13 18:17:07,477 - Epoch: [133][ 980/ 1207] Overall Loss 0.238908 Objective Loss 0.238908 LR 0.000500 Time 0.019459 -2023-02-13 18:17:07,664 - Epoch: [133][ 990/ 1207] Overall Loss 0.238815 Objective Loss 0.238815 LR 0.000500 Time 0.019452 -2023-02-13 18:17:07,851 - Epoch: [133][ 1000/ 1207] Overall Loss 0.238870 Objective Loss 0.238870 LR 0.000500 Time 0.019444 -2023-02-13 18:17:08,040 - Epoch: [133][ 1010/ 1207] Overall Loss 0.238801 Objective Loss 0.238801 LR 0.000500 Time 0.019438 -2023-02-13 18:17:08,227 - Epoch: [133][ 1020/ 1207] Overall Loss 0.238831 Objective Loss 0.238831 LR 0.000500 Time 0.019431 -2023-02-13 18:17:08,415 - Epoch: [133][ 1030/ 1207] Overall Loss 0.238778 Objective Loss 0.238778 LR 0.000500 Time 0.019424 -2023-02-13 18:17:08,603 - Epoch: [133][ 1040/ 1207] Overall Loss 0.239003 Objective Loss 0.239003 LR 0.000500 Time 0.019418 -2023-02-13 18:17:08,791 - Epoch: [133][ 1050/ 1207] Overall Loss 0.238915 Objective Loss 0.238915 LR 0.000500 Time 0.019411 -2023-02-13 18:17:08,979 - Epoch: [133][ 1060/ 1207] Overall Loss 0.238907 Objective Loss 0.238907 LR 0.000500 Time 0.019405 -2023-02-13 18:17:09,167 - Epoch: [133][ 1070/ 1207] Overall Loss 0.238765 Objective Loss 0.238765 LR 0.000500 Time 0.019399 -2023-02-13 18:17:09,355 - Epoch: [133][ 1080/ 1207] Overall Loss 0.238602 Objective Loss 0.238602 LR 0.000500 Time 0.019393 -2023-02-13 18:17:09,543 - Epoch: [133][ 1090/ 1207] Overall Loss 0.238855 Objective Loss 0.238855 LR 0.000500 Time 0.019388 -2023-02-13 18:17:09,731 - Epoch: [133][ 1100/ 1207] Overall Loss 0.238792 Objective Loss 0.238792 LR 0.000500 Time 0.019382 -2023-02-13 18:17:09,919 - Epoch: [133][ 1110/ 1207] Overall Loss 0.238835 Objective Loss 0.238835 LR 0.000500 Time 0.019377 -2023-02-13 18:17:10,108 - Epoch: [133][ 1120/ 1207] Overall Loss 0.238513 Objective Loss 0.238513 LR 0.000500 Time 0.019372 -2023-02-13 18:17:10,296 - Epoch: [133][ 1130/ 1207] Overall Loss 0.238475 Objective Loss 0.238475 LR 0.000500 Time 0.019367 -2023-02-13 18:17:10,485 - Epoch: [133][ 1140/ 1207] Overall Loss 0.238469 Objective Loss 0.238469 LR 0.000500 Time 0.019362 -2023-02-13 18:17:10,673 - Epoch: [133][ 1150/ 1207] Overall Loss 0.238525 Objective Loss 0.238525 LR 0.000500 Time 0.019356 -2023-02-13 18:17:10,862 - Epoch: [133][ 1160/ 1207] Overall Loss 0.238523 Objective Loss 0.238523 LR 0.000500 Time 0.019353 -2023-02-13 18:17:11,052 - Epoch: [133][ 1170/ 1207] Overall Loss 0.238476 Objective Loss 0.238476 LR 0.000500 Time 0.019349 -2023-02-13 18:17:11,241 - Epoch: [133][ 1180/ 1207] Overall Loss 0.238387 Objective Loss 0.238387 LR 0.000500 Time 0.019345 -2023-02-13 18:17:11,431 - Epoch: [133][ 1190/ 1207] Overall Loss 0.238426 Objective Loss 0.238426 LR 0.000500 Time 0.019342 -2023-02-13 18:17:11,670 - Epoch: [133][ 1200/ 1207] Overall Loss 0.238594 Objective Loss 0.238594 LR 0.000500 Time 0.019380 -2023-02-13 18:17:11,784 - Epoch: [133][ 1207/ 1207] Overall Loss 0.238685 Objective Loss 0.238685 Top1 81.707317 Top5 97.865854 LR 0.000500 Time 0.019362 -2023-02-13 18:17:11,855 - --- validate (epoch=133)----------- -2023-02-13 18:17:11,856 - 34311 samples (256 per mini-batch) -2023-02-13 18:17:12,269 - Epoch: [133][ 10/ 135] Loss 0.336919 Top1 83.750000 Top5 97.148438 -2023-02-13 18:17:12,411 - Epoch: [133][ 20/ 135] Loss 0.339140 Top1 83.769531 Top5 97.519531 -2023-02-13 18:17:12,540 - Epoch: [133][ 30/ 135] Loss 0.321448 Top1 84.049479 Top5 97.565104 -2023-02-13 18:17:12,663 - Epoch: [133][ 40/ 135] Loss 0.317716 Top1 84.238281 Top5 97.626953 -2023-02-13 18:17:12,787 - Epoch: [133][ 50/ 135] Loss 0.315104 Top1 84.164062 Top5 97.609375 -2023-02-13 18:17:12,909 - Epoch: [133][ 60/ 135] Loss 0.311877 Top1 84.251302 Top5 97.701823 -2023-02-13 18:17:13,033 - Epoch: [133][ 70/ 135] Loss 0.315130 Top1 84.207589 Top5 97.650670 -2023-02-13 18:17:13,156 - Epoch: [133][ 80/ 135] Loss 0.312394 Top1 84.179688 Top5 97.636719 -2023-02-13 18:17:13,280 - Epoch: [133][ 90/ 135] Loss 0.311432 Top1 84.149306 Top5 97.660590 -2023-02-13 18:17:13,416 - Epoch: [133][ 100/ 135] Loss 0.310053 Top1 84.183594 Top5 97.703125 -2023-02-13 18:17:13,542 - Epoch: [133][ 110/ 135] Loss 0.310310 Top1 84.261364 Top5 97.716619 -2023-02-13 18:17:13,662 - Epoch: [133][ 120/ 135] Loss 0.309727 Top1 84.326172 Top5 97.763672 -2023-02-13 18:17:13,789 - Epoch: [133][ 130/ 135] Loss 0.313576 Top1 84.182692 Top5 97.755409 -2023-02-13 18:17:13,833 - Epoch: [133][ 135/ 135] Loss 0.311292 Top1 84.180001 Top5 97.749993 -2023-02-13 18:17:13,903 - ==> Top1: 84.180 Top5: 97.750 Loss: 0.311 - -2023-02-13 18:17:13,904 - ==> Confusion: -[[ 864 5 6 0 8 2 1 4 5 44 0 3 1 4 4 3 3 1 1 2 6] - [ 3 932 1 1 15 24 2 27 3 1 3 2 1 0 1 1 4 0 4 1 7] - [ 8 3 958 7 6 1 15 13 0 1 5 3 2 5 6 9 1 1 5 2 7] - [ 3 0 20 898 3 5 0 1 2 3 19 0 9 0 19 2 6 3 19 0 4] - [ 13 7 0 1 995 10 1 2 2 2 0 6 0 2 7 6 7 1 0 2 2] - [ 0 20 2 2 5 964 5 21 2 5 2 9 2 15 1 2 3 1 0 4 5] - [ 3 5 10 1 1 4 1049 4 1 0 2 2 1 1 1 1 1 2 1 6 3] - [ 0 5 8 1 1 16 6 952 0 1 2 5 3 2 0 0 0 0 14 6 2] - [ 20 2 0 2 1 0 0 4 896 42 7 3 0 8 13 3 1 2 3 0 2] - [ 82 0 3 0 6 3 0 3 24 856 1 1 0 19 3 3 1 0 2 2 3] - [ 2 2 2 3 0 1 2 5 19 0 990 2 0 10 3 0 0 0 8 1 1] - [ 0 5 2 0 6 7 0 5 2 1 0 930 16 5 0 5 1 6 2 10 2] - [ 0 0 1 7 2 4 0 4 1 1 1 34 864 0 4 9 3 10 3 1 10] - [ 3 3 2 0 12 6 1 2 15 20 7 3 2 933 4 3 2 1 1 3 1] - [ 9 1 1 17 3 2 0 2 21 6 5 2 1 1 996 1 3 5 8 1 7] - [ 0 1 7 0 6 1 4 3 1 0 0 8 4 1 1 978 7 12 0 7 5] - [ 5 5 0 0 11 3 0 2 1 0 1 0 0 2 2 7 1009 3 0 5 5] - [ 4 3 1 3 0 1 4 1 1 0 1 12 13 1 0 20 0 978 1 0 7] - [ 4 3 3 7 1 2 1 24 2 1 7 1 1 0 14 1 0 1 1007 4 2] - [ 0 3 1 0 2 6 8 12 0 0 2 20 2 6 0 4 3 2 0 1072 5] - [ 165 205 237 95 184 193 105 235 109 95 223 133 242 291 190 108 329 86 177 270 9762]] - -2023-02-13 18:17:13,906 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:17:13,906 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:17:13,911 - - -2023-02-13 18:17:13,912 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:17:14,899 - Epoch: [134][ 10/ 1207] Overall Loss 0.227508 Objective Loss 0.227508 LR 0.000500 Time 0.098695 -2023-02-13 18:17:15,099 - Epoch: [134][ 20/ 1207] Overall Loss 0.236863 Objective Loss 0.236863 LR 0.000500 Time 0.059305 -2023-02-13 18:17:15,291 - Epoch: [134][ 30/ 1207] Overall Loss 0.232813 Objective Loss 0.232813 LR 0.000500 Time 0.045927 -2023-02-13 18:17:15,486 - Epoch: [134][ 40/ 1207] Overall Loss 0.231028 Objective Loss 0.231028 LR 0.000500 Time 0.039312 -2023-02-13 18:17:15,677 - Epoch: [134][ 50/ 1207] Overall Loss 0.237406 Objective Loss 0.237406 LR 0.000500 Time 0.035257 -2023-02-13 18:17:15,871 - Epoch: [134][ 60/ 1207] Overall Loss 0.237718 Objective Loss 0.237718 LR 0.000500 Time 0.032619 -2023-02-13 18:17:16,065 - Epoch: [134][ 70/ 1207] Overall Loss 0.236832 Objective Loss 0.236832 LR 0.000500 Time 0.030715 -2023-02-13 18:17:16,259 - Epoch: [134][ 80/ 1207] Overall Loss 0.237937 Objective Loss 0.237937 LR 0.000500 Time 0.029306 -2023-02-13 18:17:16,452 - Epoch: [134][ 90/ 1207] Overall Loss 0.233348 Objective Loss 0.233348 LR 0.000500 Time 0.028187 -2023-02-13 18:17:16,647 - Epoch: [134][ 100/ 1207] Overall Loss 0.232115 Objective Loss 0.232115 LR 0.000500 Time 0.027313 -2023-02-13 18:17:16,839 - Epoch: [134][ 110/ 1207] Overall Loss 0.231099 Objective Loss 0.231099 LR 0.000500 Time 0.026569 -2023-02-13 18:17:17,033 - Epoch: [134][ 120/ 1207] Overall Loss 0.232528 Objective Loss 0.232528 LR 0.000500 Time 0.025973 -2023-02-13 18:17:17,224 - Epoch: [134][ 130/ 1207] Overall Loss 0.233148 Objective Loss 0.233148 LR 0.000500 Time 0.025442 -2023-02-13 18:17:17,419 - Epoch: [134][ 140/ 1207] Overall Loss 0.233157 Objective Loss 0.233157 LR 0.000500 Time 0.025011 -2023-02-13 18:17:17,612 - Epoch: [134][ 150/ 1207] Overall Loss 0.233037 Objective Loss 0.233037 LR 0.000500 Time 0.024628 -2023-02-13 18:17:17,806 - Epoch: [134][ 160/ 1207] Overall Loss 0.234066 Objective Loss 0.234066 LR 0.000500 Time 0.024300 -2023-02-13 18:17:17,997 - Epoch: [134][ 170/ 1207] Overall Loss 0.234609 Objective Loss 0.234609 LR 0.000500 Time 0.023995 -2023-02-13 18:17:18,192 - Epoch: [134][ 180/ 1207] Overall Loss 0.235959 Objective Loss 0.235959 LR 0.000500 Time 0.023740 -2023-02-13 18:17:18,384 - Epoch: [134][ 190/ 1207] Overall Loss 0.236406 Objective Loss 0.236406 LR 0.000500 Time 0.023498 -2023-02-13 18:17:18,578 - Epoch: [134][ 200/ 1207] Overall Loss 0.236931 Objective Loss 0.236931 LR 0.000500 Time 0.023294 -2023-02-13 18:17:18,770 - Epoch: [134][ 210/ 1207] Overall Loss 0.236603 Objective Loss 0.236603 LR 0.000500 Time 0.023098 -2023-02-13 18:17:18,964 - Epoch: [134][ 220/ 1207] Overall Loss 0.235730 Objective Loss 0.235730 LR 0.000500 Time 0.022928 -2023-02-13 18:17:19,157 - Epoch: [134][ 230/ 1207] Overall Loss 0.236493 Objective Loss 0.236493 LR 0.000500 Time 0.022768 -2023-02-13 18:17:19,353 - Epoch: [134][ 240/ 1207] Overall Loss 0.236722 Objective Loss 0.236722 LR 0.000500 Time 0.022636 -2023-02-13 18:17:19,548 - Epoch: [134][ 250/ 1207] Overall Loss 0.236600 Objective Loss 0.236600 LR 0.000500 Time 0.022509 -2023-02-13 18:17:19,745 - Epoch: [134][ 260/ 1207] Overall Loss 0.235956 Objective Loss 0.235956 LR 0.000500 Time 0.022397 -2023-02-13 18:17:19,939 - Epoch: [134][ 270/ 1207] Overall Loss 0.235856 Objective Loss 0.235856 LR 0.000500 Time 0.022286 -2023-02-13 18:17:20,136 - Epoch: [134][ 280/ 1207] Overall Loss 0.235527 Objective Loss 0.235527 LR 0.000500 Time 0.022192 -2023-02-13 18:17:20,330 - Epoch: [134][ 290/ 1207] Overall Loss 0.236062 Objective Loss 0.236062 LR 0.000500 Time 0.022094 -2023-02-13 18:17:20,527 - Epoch: [134][ 300/ 1207] Overall Loss 0.236872 Objective Loss 0.236872 LR 0.000500 Time 0.022015 -2023-02-13 18:17:20,722 - Epoch: [134][ 310/ 1207] Overall Loss 0.236387 Objective Loss 0.236387 LR 0.000500 Time 0.021932 -2023-02-13 18:17:20,920 - Epoch: [134][ 320/ 1207] Overall Loss 0.236590 Objective Loss 0.236590 LR 0.000500 Time 0.021864 -2023-02-13 18:17:21,115 - Epoch: [134][ 330/ 1207] Overall Loss 0.236442 Objective Loss 0.236442 LR 0.000500 Time 0.021790 -2023-02-13 18:17:21,312 - Epoch: [134][ 340/ 1207] Overall Loss 0.236471 Objective Loss 0.236471 LR 0.000500 Time 0.021727 -2023-02-13 18:17:21,507 - Epoch: [134][ 350/ 1207] Overall Loss 0.236393 Objective Loss 0.236393 LR 0.000500 Time 0.021663 -2023-02-13 18:17:21,704 - Epoch: [134][ 360/ 1207] Overall Loss 0.235729 Objective Loss 0.235729 LR 0.000500 Time 0.021607 -2023-02-13 18:17:21,898 - Epoch: [134][ 370/ 1207] Overall Loss 0.235831 Objective Loss 0.235831 LR 0.000500 Time 0.021548 -2023-02-13 18:17:22,096 - Epoch: [134][ 380/ 1207] Overall Loss 0.236313 Objective Loss 0.236313 LR 0.000500 Time 0.021499 -2023-02-13 18:17:22,291 - Epoch: [134][ 390/ 1207] Overall Loss 0.236682 Objective Loss 0.236682 LR 0.000500 Time 0.021447 -2023-02-13 18:17:22,489 - Epoch: [134][ 400/ 1207] Overall Loss 0.236982 Objective Loss 0.236982 LR 0.000500 Time 0.021406 -2023-02-13 18:17:22,683 - Epoch: [134][ 410/ 1207] Overall Loss 0.236862 Objective Loss 0.236862 LR 0.000500 Time 0.021356 -2023-02-13 18:17:22,881 - Epoch: [134][ 420/ 1207] Overall Loss 0.236750 Objective Loss 0.236750 LR 0.000500 Time 0.021318 -2023-02-13 18:17:23,075 - Epoch: [134][ 430/ 1207] Overall Loss 0.236794 Objective Loss 0.236794 LR 0.000500 Time 0.021272 -2023-02-13 18:17:23,272 - Epoch: [134][ 440/ 1207] Overall Loss 0.236875 Objective Loss 0.236875 LR 0.000500 Time 0.021236 -2023-02-13 18:17:23,467 - Epoch: [134][ 450/ 1207] Overall Loss 0.236908 Objective Loss 0.236908 LR 0.000500 Time 0.021196 -2023-02-13 18:17:23,664 - Epoch: [134][ 460/ 1207] Overall Loss 0.236377 Objective Loss 0.236377 LR 0.000500 Time 0.021164 -2023-02-13 18:17:23,859 - Epoch: [134][ 470/ 1207] Overall Loss 0.236606 Objective Loss 0.236606 LR 0.000500 Time 0.021126 -2023-02-13 18:17:24,056 - Epoch: [134][ 480/ 1207] Overall Loss 0.236499 Objective Loss 0.236499 LR 0.000500 Time 0.021097 -2023-02-13 18:17:24,251 - Epoch: [134][ 490/ 1207] Overall Loss 0.236490 Objective Loss 0.236490 LR 0.000500 Time 0.021063 -2023-02-13 18:17:24,448 - Epoch: [134][ 500/ 1207] Overall Loss 0.236478 Objective Loss 0.236478 LR 0.000500 Time 0.021036 -2023-02-13 18:17:24,643 - Epoch: [134][ 510/ 1207] Overall Loss 0.236720 Objective Loss 0.236720 LR 0.000500 Time 0.021004 -2023-02-13 18:17:24,841 - Epoch: [134][ 520/ 1207] Overall Loss 0.237260 Objective Loss 0.237260 LR 0.000500 Time 0.020981 -2023-02-13 18:17:25,036 - Epoch: [134][ 530/ 1207] Overall Loss 0.237192 Objective Loss 0.237192 LR 0.000500 Time 0.020951 -2023-02-13 18:17:25,233 - Epoch: [134][ 540/ 1207] Overall Loss 0.236898 Objective Loss 0.236898 LR 0.000500 Time 0.020928 -2023-02-13 18:17:25,428 - Epoch: [134][ 550/ 1207] Overall Loss 0.237015 Objective Loss 0.237015 LR 0.000500 Time 0.020902 -2023-02-13 18:17:25,627 - Epoch: [134][ 560/ 1207] Overall Loss 0.236995 Objective Loss 0.236995 LR 0.000500 Time 0.020882 -2023-02-13 18:17:25,824 - Epoch: [134][ 570/ 1207] Overall Loss 0.236741 Objective Loss 0.236741 LR 0.000500 Time 0.020862 -2023-02-13 18:17:26,021 - Epoch: [134][ 580/ 1207] Overall Loss 0.237287 Objective Loss 0.237287 LR 0.000500 Time 0.020842 -2023-02-13 18:17:26,214 - Epoch: [134][ 590/ 1207] Overall Loss 0.237429 Objective Loss 0.237429 LR 0.000500 Time 0.020814 -2023-02-13 18:17:26,407 - Epoch: [134][ 600/ 1207] Overall Loss 0.237120 Objective Loss 0.237120 LR 0.000500 Time 0.020789 -2023-02-13 18:17:26,602 - Epoch: [134][ 610/ 1207] Overall Loss 0.236869 Objective Loss 0.236869 LR 0.000500 Time 0.020766 -2023-02-13 18:17:26,794 - Epoch: [134][ 620/ 1207] Overall Loss 0.237819 Objective Loss 0.237819 LR 0.000500 Time 0.020741 -2023-02-13 18:17:26,989 - Epoch: [134][ 630/ 1207] Overall Loss 0.237852 Objective Loss 0.237852 LR 0.000500 Time 0.020720 -2023-02-13 18:17:27,182 - Epoch: [134][ 640/ 1207] Overall Loss 0.237881 Objective Loss 0.237881 LR 0.000500 Time 0.020698 -2023-02-13 18:17:27,375 - Epoch: [134][ 650/ 1207] Overall Loss 0.238417 Objective Loss 0.238417 LR 0.000500 Time 0.020676 -2023-02-13 18:17:27,569 - Epoch: [134][ 660/ 1207] Overall Loss 0.238543 Objective Loss 0.238543 LR 0.000500 Time 0.020656 -2023-02-13 18:17:27,762 - Epoch: [134][ 670/ 1207] Overall Loss 0.238619 Objective Loss 0.238619 LR 0.000500 Time 0.020635 -2023-02-13 18:17:27,957 - Epoch: [134][ 680/ 1207] Overall Loss 0.238598 Objective Loss 0.238598 LR 0.000500 Time 0.020617 -2023-02-13 18:17:28,150 - Epoch: [134][ 690/ 1207] Overall Loss 0.238389 Objective Loss 0.238389 LR 0.000500 Time 0.020598 -2023-02-13 18:17:28,345 - Epoch: [134][ 700/ 1207] Overall Loss 0.238111 Objective Loss 0.238111 LR 0.000500 Time 0.020581 -2023-02-13 18:17:28,539 - Epoch: [134][ 710/ 1207] Overall Loss 0.238084 Objective Loss 0.238084 LR 0.000500 Time 0.020564 -2023-02-13 18:17:28,732 - Epoch: [134][ 720/ 1207] Overall Loss 0.238227 Objective Loss 0.238227 LR 0.000500 Time 0.020547 -2023-02-13 18:17:28,925 - Epoch: [134][ 730/ 1207] Overall Loss 0.238220 Objective Loss 0.238220 LR 0.000500 Time 0.020529 -2023-02-13 18:17:29,119 - Epoch: [134][ 740/ 1207] Overall Loss 0.238437 Objective Loss 0.238437 LR 0.000500 Time 0.020513 -2023-02-13 18:17:29,312 - Epoch: [134][ 750/ 1207] Overall Loss 0.238338 Objective Loss 0.238338 LR 0.000500 Time 0.020497 -2023-02-13 18:17:29,507 - Epoch: [134][ 760/ 1207] Overall Loss 0.238348 Objective Loss 0.238348 LR 0.000500 Time 0.020482 -2023-02-13 18:17:29,700 - Epoch: [134][ 770/ 1207] Overall Loss 0.238405 Objective Loss 0.238405 LR 0.000500 Time 0.020468 -2023-02-13 18:17:29,894 - Epoch: [134][ 780/ 1207] Overall Loss 0.238475 Objective Loss 0.238475 LR 0.000500 Time 0.020452 -2023-02-13 18:17:30,087 - Epoch: [134][ 790/ 1207] Overall Loss 0.238313 Objective Loss 0.238313 LR 0.000500 Time 0.020437 -2023-02-13 18:17:30,280 - Epoch: [134][ 800/ 1207] Overall Loss 0.238239 Objective Loss 0.238239 LR 0.000500 Time 0.020423 -2023-02-13 18:17:30,474 - Epoch: [134][ 810/ 1207] Overall Loss 0.238195 Objective Loss 0.238195 LR 0.000500 Time 0.020410 -2023-02-13 18:17:30,668 - Epoch: [134][ 820/ 1207] Overall Loss 0.238107 Objective Loss 0.238107 LR 0.000500 Time 0.020397 -2023-02-13 18:17:30,865 - Epoch: [134][ 830/ 1207] Overall Loss 0.238010 Objective Loss 0.238010 LR 0.000500 Time 0.020389 -2023-02-13 18:17:31,058 - Epoch: [134][ 840/ 1207] Overall Loss 0.237844 Objective Loss 0.237844 LR 0.000500 Time 0.020375 -2023-02-13 18:17:31,252 - Epoch: [134][ 850/ 1207] Overall Loss 0.237868 Objective Loss 0.237868 LR 0.000500 Time 0.020363 -2023-02-13 18:17:31,446 - Epoch: [134][ 860/ 1207] Overall Loss 0.237946 Objective Loss 0.237946 LR 0.000500 Time 0.020351 -2023-02-13 18:17:31,639 - Epoch: [134][ 870/ 1207] Overall Loss 0.237970 Objective Loss 0.237970 LR 0.000500 Time 0.020339 -2023-02-13 18:17:31,833 - Epoch: [134][ 880/ 1207] Overall Loss 0.238165 Objective Loss 0.238165 LR 0.000500 Time 0.020328 -2023-02-13 18:17:32,026 - Epoch: [134][ 890/ 1207] Overall Loss 0.238278 Objective Loss 0.238278 LR 0.000500 Time 0.020316 -2023-02-13 18:17:32,217 - Epoch: [134][ 900/ 1207] Overall Loss 0.238363 Objective Loss 0.238363 LR 0.000500 Time 0.020302 -2023-02-13 18:17:32,409 - Epoch: [134][ 910/ 1207] Overall Loss 0.238573 Objective Loss 0.238573 LR 0.000500 Time 0.020290 -2023-02-13 18:17:32,601 - Epoch: [134][ 920/ 1207] Overall Loss 0.238514 Objective Loss 0.238514 LR 0.000500 Time 0.020277 -2023-02-13 18:17:32,793 - Epoch: [134][ 930/ 1207] Overall Loss 0.238354 Objective Loss 0.238354 LR 0.000500 Time 0.020265 -2023-02-13 18:17:32,984 - Epoch: [134][ 940/ 1207] Overall Loss 0.238287 Objective Loss 0.238287 LR 0.000500 Time 0.020252 -2023-02-13 18:17:33,177 - Epoch: [134][ 950/ 1207] Overall Loss 0.238276 Objective Loss 0.238276 LR 0.000500 Time 0.020242 -2023-02-13 18:17:33,368 - Epoch: [134][ 960/ 1207] Overall Loss 0.237988 Objective Loss 0.237988 LR 0.000500 Time 0.020230 -2023-02-13 18:17:33,561 - Epoch: [134][ 970/ 1207] Overall Loss 0.237941 Objective Loss 0.237941 LR 0.000500 Time 0.020220 -2023-02-13 18:17:33,752 - Epoch: [134][ 980/ 1207] Overall Loss 0.237987 Objective Loss 0.237987 LR 0.000500 Time 0.020208 -2023-02-13 18:17:33,944 - Epoch: [134][ 990/ 1207] Overall Loss 0.238006 Objective Loss 0.238006 LR 0.000500 Time 0.020198 -2023-02-13 18:17:34,136 - Epoch: [134][ 1000/ 1207] Overall Loss 0.237954 Objective Loss 0.237954 LR 0.000500 Time 0.020187 -2023-02-13 18:17:34,328 - Epoch: [134][ 1010/ 1207] Overall Loss 0.238045 Objective Loss 0.238045 LR 0.000500 Time 0.020177 -2023-02-13 18:17:34,521 - Epoch: [134][ 1020/ 1207] Overall Loss 0.238183 Objective Loss 0.238183 LR 0.000500 Time 0.020167 -2023-02-13 18:17:34,713 - Epoch: [134][ 1030/ 1207] Overall Loss 0.238260 Objective Loss 0.238260 LR 0.000500 Time 0.020158 -2023-02-13 18:17:34,905 - Epoch: [134][ 1040/ 1207] Overall Loss 0.238283 Objective Loss 0.238283 LR 0.000500 Time 0.020148 -2023-02-13 18:17:35,097 - Epoch: [134][ 1050/ 1207] Overall Loss 0.238131 Objective Loss 0.238131 LR 0.000500 Time 0.020139 -2023-02-13 18:17:35,288 - Epoch: [134][ 1060/ 1207] Overall Loss 0.238277 Objective Loss 0.238277 LR 0.000500 Time 0.020129 -2023-02-13 18:17:35,480 - Epoch: [134][ 1070/ 1207] Overall Loss 0.238260 Objective Loss 0.238260 LR 0.000500 Time 0.020120 -2023-02-13 18:17:35,670 - Epoch: [134][ 1080/ 1207] Overall Loss 0.238313 Objective Loss 0.238313 LR 0.000500 Time 0.020110 -2023-02-13 18:17:35,860 - Epoch: [134][ 1090/ 1207] Overall Loss 0.238509 Objective Loss 0.238509 LR 0.000500 Time 0.020099 -2023-02-13 18:17:36,051 - Epoch: [134][ 1100/ 1207] Overall Loss 0.238185 Objective Loss 0.238185 LR 0.000500 Time 0.020090 -2023-02-13 18:17:36,242 - Epoch: [134][ 1110/ 1207] Overall Loss 0.238111 Objective Loss 0.238111 LR 0.000500 Time 0.020080 -2023-02-13 18:17:36,432 - Epoch: [134][ 1120/ 1207] Overall Loss 0.238225 Objective Loss 0.238225 LR 0.000500 Time 0.020070 -2023-02-13 18:17:36,622 - Epoch: [134][ 1130/ 1207] Overall Loss 0.238333 Objective Loss 0.238333 LR 0.000500 Time 0.020060 -2023-02-13 18:17:36,812 - Epoch: [134][ 1140/ 1207] Overall Loss 0.238333 Objective Loss 0.238333 LR 0.000500 Time 0.020051 -2023-02-13 18:17:37,002 - Epoch: [134][ 1150/ 1207] Overall Loss 0.238306 Objective Loss 0.238306 LR 0.000500 Time 0.020042 -2023-02-13 18:17:37,192 - Epoch: [134][ 1160/ 1207] Overall Loss 0.238610 Objective Loss 0.238610 LR 0.000500 Time 0.020032 -2023-02-13 18:17:37,382 - Epoch: [134][ 1170/ 1207] Overall Loss 0.238610 Objective Loss 0.238610 LR 0.000500 Time 0.020023 -2023-02-13 18:17:37,572 - Epoch: [134][ 1180/ 1207] Overall Loss 0.238771 Objective Loss 0.238771 LR 0.000500 Time 0.020014 -2023-02-13 18:17:37,762 - Epoch: [134][ 1190/ 1207] Overall Loss 0.238671 Objective Loss 0.238671 LR 0.000500 Time 0.020005 -2023-02-13 18:17:38,002 - Epoch: [134][ 1200/ 1207] Overall Loss 0.238460 Objective Loss 0.238460 LR 0.000500 Time 0.020038 -2023-02-13 18:17:38,117 - Epoch: [134][ 1207/ 1207] Overall Loss 0.238448 Objective Loss 0.238448 Top1 88.719512 Top5 99.695122 LR 0.000500 Time 0.020017 -2023-02-13 18:17:38,189 - --- validate (epoch=134)----------- -2023-02-13 18:17:38,189 - 34311 samples (256 per mini-batch) -2023-02-13 18:17:38,591 - Epoch: [134][ 10/ 135] Loss 0.319454 Top1 85.585938 Top5 97.890625 -2023-02-13 18:17:38,716 - Epoch: [134][ 20/ 135] Loss 0.326231 Top1 85.078125 Top5 97.949219 -2023-02-13 18:17:38,840 - Epoch: [134][ 30/ 135] Loss 0.326358 Top1 85.286458 Top5 97.838542 -2023-02-13 18:17:38,972 - Epoch: [134][ 40/ 135] Loss 0.321787 Top1 85.400391 Top5 97.792969 -2023-02-13 18:17:39,104 - Epoch: [134][ 50/ 135] Loss 0.321987 Top1 85.265625 Top5 97.718750 -2023-02-13 18:17:39,230 - Epoch: [134][ 60/ 135] Loss 0.322360 Top1 85.123698 Top5 97.747396 -2023-02-13 18:17:39,351 - Epoch: [134][ 70/ 135] Loss 0.318790 Top1 85.066964 Top5 97.806920 -2023-02-13 18:17:39,489 - Epoch: [134][ 80/ 135] Loss 0.315230 Top1 85.151367 Top5 97.871094 -2023-02-13 18:17:39,620 - Epoch: [134][ 90/ 135] Loss 0.312601 Top1 85.190972 Top5 97.929688 -2023-02-13 18:17:39,751 - Epoch: [134][ 100/ 135] Loss 0.310296 Top1 85.226562 Top5 97.960938 -2023-02-13 18:17:39,878 - Epoch: [134][ 110/ 135] Loss 0.311225 Top1 85.127841 Top5 97.908381 -2023-02-13 18:17:40,002 - Epoch: [134][ 120/ 135] Loss 0.314508 Top1 85.029297 Top5 97.893880 -2023-02-13 18:17:40,134 - Epoch: [134][ 130/ 135] Loss 0.314825 Top1 84.957933 Top5 97.896635 -2023-02-13 18:17:40,180 - Epoch: [134][ 135/ 135] Loss 0.314345 Top1 84.955262 Top5 97.921949 -2023-02-13 18:17:40,249 - ==> Top1: 84.955 Top5: 97.922 Loss: 0.314 - -2023-02-13 18:17:40,249 - ==> Confusion: -[[ 863 4 4 1 15 4 0 2 5 26 1 4 0 5 11 5 3 3 0 2 9] - [ 1 941 2 1 13 31 3 9 2 0 2 2 2 0 2 1 7 0 6 1 7] - [ 6 3 944 10 5 2 20 14 0 1 3 1 6 6 3 11 3 5 4 4 7] - [ 4 2 18 908 1 5 1 2 1 2 9 0 15 0 16 4 5 6 7 0 10] - [ 13 8 1 0 986 11 1 1 0 2 0 5 0 2 11 6 9 1 2 2 5] - [ 3 21 0 4 3 974 2 18 2 1 1 8 3 12 1 1 5 0 1 4 6] - [ 4 5 14 1 0 7 1035 7 3 0 3 1 2 0 0 4 1 2 1 6 3] - [ 2 18 9 3 1 36 4 900 1 1 2 6 3 1 0 1 0 3 20 9 4] - [ 26 3 1 1 2 0 0 4 887 26 8 2 0 9 27 1 1 2 2 1 6] - [ 106 2 2 1 12 3 0 2 31 814 2 0 0 18 8 1 1 3 3 1 2] - [ 0 1 6 10 0 4 2 3 9 1 988 0 3 11 2 1 0 0 9 0 1] - [ 2 2 2 0 3 10 0 3 1 1 1 915 29 5 0 2 3 10 2 10 4] - [ 1 0 0 4 2 1 0 2 1 3 1 28 887 0 2 6 4 10 0 0 7] - [ 5 3 1 2 5 17 0 0 10 11 9 2 5 926 6 2 5 0 1 3 11] - [ 3 3 3 18 2 3 0 2 11 4 4 2 2 0 1008 1 1 6 5 1 13] - [ 3 1 2 0 4 2 2 1 2 0 0 5 8 3 1 977 11 11 0 6 7] - [ 1 6 0 1 7 2 0 0 1 1 0 2 5 3 2 10 1003 1 0 4 12] - [ 4 3 0 3 1 0 2 0 0 0 0 9 14 0 1 19 1 983 1 2 8] - [ 1 3 4 11 0 2 0 20 3 0 5 1 9 0 17 1 1 1 1002 1 4] - [ 1 5 2 0 2 6 5 13 0 0 1 11 3 4 0 6 4 4 1 1073 7] - [ 137 221 185 131 140 214 95 143 66 67 171 94 323 261 173 128 226 95 163 266 10135]] - -2023-02-13 18:17:40,251 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:17:40,251 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:17:40,257 - - -2023-02-13 18:17:40,257 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:17:41,161 - Epoch: [135][ 10/ 1207] Overall Loss 0.232716 Objective Loss 0.232716 LR 0.000500 Time 0.090361 -2023-02-13 18:17:41,361 - Epoch: [135][ 20/ 1207] Overall Loss 0.226664 Objective Loss 0.226664 LR 0.000500 Time 0.055133 -2023-02-13 18:17:41,557 - Epoch: [135][ 30/ 1207] Overall Loss 0.230292 Objective Loss 0.230292 LR 0.000500 Time 0.043303 -2023-02-13 18:17:41,750 - Epoch: [135][ 40/ 1207] Overall Loss 0.228183 Objective Loss 0.228183 LR 0.000500 Time 0.037280 -2023-02-13 18:17:41,945 - Epoch: [135][ 50/ 1207] Overall Loss 0.232617 Objective Loss 0.232617 LR 0.000500 Time 0.033723 -2023-02-13 18:17:42,137 - Epoch: [135][ 60/ 1207] Overall Loss 0.231543 Objective Loss 0.231543 LR 0.000500 Time 0.031301 -2023-02-13 18:17:42,332 - Epoch: [135][ 70/ 1207] Overall Loss 0.231000 Objective Loss 0.231000 LR 0.000500 Time 0.029608 -2023-02-13 18:17:42,526 - Epoch: [135][ 80/ 1207] Overall Loss 0.226158 Objective Loss 0.226158 LR 0.000500 Time 0.028327 -2023-02-13 18:17:42,722 - Epoch: [135][ 90/ 1207] Overall Loss 0.227910 Objective Loss 0.227910 LR 0.000500 Time 0.027349 -2023-02-13 18:17:42,914 - Epoch: [135][ 100/ 1207] Overall Loss 0.228230 Objective Loss 0.228230 LR 0.000500 Time 0.026535 -2023-02-13 18:17:43,110 - Epoch: [135][ 110/ 1207] Overall Loss 0.228224 Objective Loss 0.228224 LR 0.000500 Time 0.025897 -2023-02-13 18:17:43,303 - Epoch: [135][ 120/ 1207] Overall Loss 0.228384 Objective Loss 0.228384 LR 0.000500 Time 0.025343 -2023-02-13 18:17:43,498 - Epoch: [135][ 130/ 1207] Overall Loss 0.228909 Objective Loss 0.228909 LR 0.000500 Time 0.024892 -2023-02-13 18:17:43,691 - Epoch: [135][ 140/ 1207] Overall Loss 0.229224 Objective Loss 0.229224 LR 0.000500 Time 0.024492 -2023-02-13 18:17:43,887 - Epoch: [135][ 150/ 1207] Overall Loss 0.228434 Objective Loss 0.228434 LR 0.000500 Time 0.024158 -2023-02-13 18:17:44,079 - Epoch: [135][ 160/ 1207] Overall Loss 0.227565 Objective Loss 0.227565 LR 0.000500 Time 0.023850 -2023-02-13 18:17:44,274 - Epoch: [135][ 170/ 1207] Overall Loss 0.229359 Objective Loss 0.229359 LR 0.000500 Time 0.023592 -2023-02-13 18:17:44,467 - Epoch: [135][ 180/ 1207] Overall Loss 0.230458 Objective Loss 0.230458 LR 0.000500 Time 0.023352 -2023-02-13 18:17:44,663 - Epoch: [135][ 190/ 1207] Overall Loss 0.229673 Objective Loss 0.229673 LR 0.000500 Time 0.023149 -2023-02-13 18:17:44,856 - Epoch: [135][ 200/ 1207] Overall Loss 0.229748 Objective Loss 0.229748 LR 0.000500 Time 0.022959 -2023-02-13 18:17:45,051 - Epoch: [135][ 210/ 1207] Overall Loss 0.230812 Objective Loss 0.230812 LR 0.000500 Time 0.022791 -2023-02-13 18:17:45,243 - Epoch: [135][ 220/ 1207] Overall Loss 0.230773 Objective Loss 0.230773 LR 0.000500 Time 0.022627 -2023-02-13 18:17:45,439 - Epoch: [135][ 230/ 1207] Overall Loss 0.230005 Objective Loss 0.230005 LR 0.000500 Time 0.022494 -2023-02-13 18:17:45,634 - Epoch: [135][ 240/ 1207] Overall Loss 0.230690 Objective Loss 0.230690 LR 0.000500 Time 0.022364 -2023-02-13 18:17:45,830 - Epoch: [135][ 250/ 1207] Overall Loss 0.230653 Objective Loss 0.230653 LR 0.000500 Time 0.022253 -2023-02-13 18:17:46,024 - Epoch: [135][ 260/ 1207] Overall Loss 0.230437 Objective Loss 0.230437 LR 0.000500 Time 0.022143 -2023-02-13 18:17:46,220 - Epoch: [135][ 270/ 1207] Overall Loss 0.231299 Objective Loss 0.231299 LR 0.000500 Time 0.022045 -2023-02-13 18:17:46,413 - Epoch: [135][ 280/ 1207] Overall Loss 0.231242 Objective Loss 0.231242 LR 0.000500 Time 0.021946 -2023-02-13 18:17:46,608 - Epoch: [135][ 290/ 1207] Overall Loss 0.231226 Objective Loss 0.231226 LR 0.000500 Time 0.021861 -2023-02-13 18:17:46,801 - Epoch: [135][ 300/ 1207] Overall Loss 0.231539 Objective Loss 0.231539 LR 0.000500 Time 0.021774 -2023-02-13 18:17:46,997 - Epoch: [135][ 310/ 1207] Overall Loss 0.231319 Objective Loss 0.231319 LR 0.000500 Time 0.021703 -2023-02-13 18:17:47,190 - Epoch: [135][ 320/ 1207] Overall Loss 0.231872 Objective Loss 0.231872 LR 0.000500 Time 0.021627 -2023-02-13 18:17:47,385 - Epoch: [135][ 330/ 1207] Overall Loss 0.232499 Objective Loss 0.232499 LR 0.000500 Time 0.021561 -2023-02-13 18:17:47,578 - Epoch: [135][ 340/ 1207] Overall Loss 0.232380 Objective Loss 0.232380 LR 0.000500 Time 0.021494 -2023-02-13 18:17:47,773 - Epoch: [135][ 350/ 1207] Overall Loss 0.232224 Objective Loss 0.232224 LR 0.000500 Time 0.021436 -2023-02-13 18:17:47,966 - Epoch: [135][ 360/ 1207] Overall Loss 0.232071 Objective Loss 0.232071 LR 0.000500 Time 0.021377 -2023-02-13 18:17:48,162 - Epoch: [135][ 370/ 1207] Overall Loss 0.231777 Objective Loss 0.231777 LR 0.000500 Time 0.021326 -2023-02-13 18:17:48,355 - Epoch: [135][ 380/ 1207] Overall Loss 0.231298 Objective Loss 0.231298 LR 0.000500 Time 0.021273 -2023-02-13 18:17:48,551 - Epoch: [135][ 390/ 1207] Overall Loss 0.231112 Objective Loss 0.231112 LR 0.000500 Time 0.021230 -2023-02-13 18:17:48,744 - Epoch: [135][ 400/ 1207] Overall Loss 0.231304 Objective Loss 0.231304 LR 0.000500 Time 0.021181 -2023-02-13 18:17:48,941 - Epoch: [135][ 410/ 1207] Overall Loss 0.231081 Objective Loss 0.231081 LR 0.000500 Time 0.021143 -2023-02-13 18:17:49,135 - Epoch: [135][ 420/ 1207] Overall Loss 0.230826 Objective Loss 0.230826 LR 0.000500 Time 0.021100 -2023-02-13 18:17:49,331 - Epoch: [135][ 430/ 1207] Overall Loss 0.231063 Objective Loss 0.231063 LR 0.000500 Time 0.021064 -2023-02-13 18:17:49,524 - Epoch: [135][ 440/ 1207] Overall Loss 0.231186 Objective Loss 0.231186 LR 0.000500 Time 0.021023 -2023-02-13 18:17:49,719 - Epoch: [135][ 450/ 1207] Overall Loss 0.231229 Objective Loss 0.231229 LR 0.000500 Time 0.020989 -2023-02-13 18:17:49,913 - Epoch: [135][ 460/ 1207] Overall Loss 0.231472 Objective Loss 0.231472 LR 0.000500 Time 0.020953 -2023-02-13 18:17:50,108 - Epoch: [135][ 470/ 1207] Overall Loss 0.231522 Objective Loss 0.231522 LR 0.000500 Time 0.020922 -2023-02-13 18:17:50,301 - Epoch: [135][ 480/ 1207] Overall Loss 0.231463 Objective Loss 0.231463 LR 0.000500 Time 0.020887 -2023-02-13 18:17:50,498 - Epoch: [135][ 490/ 1207] Overall Loss 0.231900 Objective Loss 0.231900 LR 0.000500 Time 0.020861 -2023-02-13 18:17:50,691 - Epoch: [135][ 500/ 1207] Overall Loss 0.232336 Objective Loss 0.232336 LR 0.000500 Time 0.020831 -2023-02-13 18:17:50,888 - Epoch: [135][ 510/ 1207] Overall Loss 0.232198 Objective Loss 0.232198 LR 0.000500 Time 0.020806 -2023-02-13 18:17:51,081 - Epoch: [135][ 520/ 1207] Overall Loss 0.232614 Objective Loss 0.232614 LR 0.000500 Time 0.020778 -2023-02-13 18:17:51,277 - Epoch: [135][ 530/ 1207] Overall Loss 0.232757 Objective Loss 0.232757 LR 0.000500 Time 0.020755 -2023-02-13 18:17:51,470 - Epoch: [135][ 540/ 1207] Overall Loss 0.232939 Objective Loss 0.232939 LR 0.000500 Time 0.020727 -2023-02-13 18:17:51,667 - Epoch: [135][ 550/ 1207] Overall Loss 0.233123 Objective Loss 0.233123 LR 0.000500 Time 0.020707 -2023-02-13 18:17:51,861 - Epoch: [135][ 560/ 1207] Overall Loss 0.233082 Objective Loss 0.233082 LR 0.000500 Time 0.020683 -2023-02-13 18:17:52,057 - Epoch: [135][ 570/ 1207] Overall Loss 0.232907 Objective Loss 0.232907 LR 0.000500 Time 0.020664 -2023-02-13 18:17:52,249 - Epoch: [135][ 580/ 1207] Overall Loss 0.232568 Objective Loss 0.232568 LR 0.000500 Time 0.020638 -2023-02-13 18:17:52,445 - Epoch: [135][ 590/ 1207] Overall Loss 0.233090 Objective Loss 0.233090 LR 0.000500 Time 0.020620 -2023-02-13 18:17:52,639 - Epoch: [135][ 600/ 1207] Overall Loss 0.233550 Objective Loss 0.233550 LR 0.000500 Time 0.020599 -2023-02-13 18:17:52,835 - Epoch: [135][ 610/ 1207] Overall Loss 0.233839 Objective Loss 0.233839 LR 0.000500 Time 0.020582 -2023-02-13 18:17:53,028 - Epoch: [135][ 620/ 1207] Overall Loss 0.233584 Objective Loss 0.233584 LR 0.000500 Time 0.020560 -2023-02-13 18:17:53,223 - Epoch: [135][ 630/ 1207] Overall Loss 0.233607 Objective Loss 0.233607 LR 0.000500 Time 0.020544 -2023-02-13 18:17:53,417 - Epoch: [135][ 640/ 1207] Overall Loss 0.233322 Objective Loss 0.233322 LR 0.000500 Time 0.020525 -2023-02-13 18:17:53,614 - Epoch: [135][ 650/ 1207] Overall Loss 0.233237 Objective Loss 0.233237 LR 0.000500 Time 0.020511 -2023-02-13 18:17:53,807 - Epoch: [135][ 660/ 1207] Overall Loss 0.233679 Objective Loss 0.233679 LR 0.000500 Time 0.020492 -2023-02-13 18:17:54,003 - Epoch: [135][ 670/ 1207] Overall Loss 0.234058 Objective Loss 0.234058 LR 0.000500 Time 0.020478 -2023-02-13 18:17:54,197 - Epoch: [135][ 680/ 1207] Overall Loss 0.234382 Objective Loss 0.234382 LR 0.000500 Time 0.020461 -2023-02-13 18:17:54,392 - Epoch: [135][ 690/ 1207] Overall Loss 0.234268 Objective Loss 0.234268 LR 0.000500 Time 0.020448 -2023-02-13 18:17:54,587 - Epoch: [135][ 700/ 1207] Overall Loss 0.234098 Objective Loss 0.234098 LR 0.000500 Time 0.020433 -2023-02-13 18:17:54,783 - Epoch: [135][ 710/ 1207] Overall Loss 0.234192 Objective Loss 0.234192 LR 0.000500 Time 0.020421 -2023-02-13 18:17:54,977 - Epoch: [135][ 720/ 1207] Overall Loss 0.234370 Objective Loss 0.234370 LR 0.000500 Time 0.020406 -2023-02-13 18:17:55,174 - Epoch: [135][ 730/ 1207] Overall Loss 0.234535 Objective Loss 0.234535 LR 0.000500 Time 0.020396 -2023-02-13 18:17:55,367 - Epoch: [135][ 740/ 1207] Overall Loss 0.234674 Objective Loss 0.234674 LR 0.000500 Time 0.020381 -2023-02-13 18:17:55,563 - Epoch: [135][ 750/ 1207] Overall Loss 0.235130 Objective Loss 0.235130 LR 0.000500 Time 0.020370 -2023-02-13 18:17:55,757 - Epoch: [135][ 760/ 1207] Overall Loss 0.235062 Objective Loss 0.235062 LR 0.000500 Time 0.020357 -2023-02-13 18:17:55,954 - Epoch: [135][ 770/ 1207] Overall Loss 0.234983 Objective Loss 0.234983 LR 0.000500 Time 0.020348 -2023-02-13 18:17:56,149 - Epoch: [135][ 780/ 1207] Overall Loss 0.234785 Objective Loss 0.234785 LR 0.000500 Time 0.020336 -2023-02-13 18:17:56,345 - Epoch: [135][ 790/ 1207] Overall Loss 0.234878 Objective Loss 0.234878 LR 0.000500 Time 0.020327 -2023-02-13 18:17:56,539 - Epoch: [135][ 800/ 1207] Overall Loss 0.235138 Objective Loss 0.235138 LR 0.000500 Time 0.020315 -2023-02-13 18:17:56,736 - Epoch: [135][ 810/ 1207] Overall Loss 0.235618 Objective Loss 0.235618 LR 0.000500 Time 0.020306 -2023-02-13 18:17:56,930 - Epoch: [135][ 820/ 1207] Overall Loss 0.235638 Objective Loss 0.235638 LR 0.000500 Time 0.020295 -2023-02-13 18:17:57,127 - Epoch: [135][ 830/ 1207] Overall Loss 0.235354 Objective Loss 0.235354 LR 0.000500 Time 0.020287 -2023-02-13 18:17:57,322 - Epoch: [135][ 840/ 1207] Overall Loss 0.235411 Objective Loss 0.235411 LR 0.000500 Time 0.020277 -2023-02-13 18:17:57,518 - Epoch: [135][ 850/ 1207] Overall Loss 0.235379 Objective Loss 0.235379 LR 0.000500 Time 0.020270 -2023-02-13 18:17:57,713 - Epoch: [135][ 860/ 1207] Overall Loss 0.235725 Objective Loss 0.235725 LR 0.000500 Time 0.020260 -2023-02-13 18:17:57,910 - Epoch: [135][ 870/ 1207] Overall Loss 0.235574 Objective Loss 0.235574 LR 0.000500 Time 0.020253 -2023-02-13 18:17:58,104 - Epoch: [135][ 880/ 1207] Overall Loss 0.235835 Objective Loss 0.235835 LR 0.000500 Time 0.020243 -2023-02-13 18:17:58,295 - Epoch: [135][ 890/ 1207] Overall Loss 0.236165 Objective Loss 0.236165 LR 0.000500 Time 0.020229 -2023-02-13 18:17:58,485 - Epoch: [135][ 900/ 1207] Overall Loss 0.236151 Objective Loss 0.236151 LR 0.000500 Time 0.020215 -2023-02-13 18:17:58,675 - Epoch: [135][ 910/ 1207] Overall Loss 0.235979 Objective Loss 0.235979 LR 0.000500 Time 0.020202 -2023-02-13 18:17:58,865 - Epoch: [135][ 920/ 1207] Overall Loss 0.235855 Objective Loss 0.235855 LR 0.000500 Time 0.020188 -2023-02-13 18:17:59,055 - Epoch: [135][ 930/ 1207] Overall Loss 0.235531 Objective Loss 0.235531 LR 0.000500 Time 0.020175 -2023-02-13 18:17:59,245 - Epoch: [135][ 940/ 1207] Overall Loss 0.235447 Objective Loss 0.235447 LR 0.000500 Time 0.020162 -2023-02-13 18:17:59,435 - Epoch: [135][ 950/ 1207] Overall Loss 0.235576 Objective Loss 0.235576 LR 0.000500 Time 0.020149 -2023-02-13 18:17:59,625 - Epoch: [135][ 960/ 1207] Overall Loss 0.235583 Objective Loss 0.235583 LR 0.000500 Time 0.020137 -2023-02-13 18:17:59,815 - Epoch: [135][ 970/ 1207] Overall Loss 0.235672 Objective Loss 0.235672 LR 0.000500 Time 0.020125 -2023-02-13 18:18:00,004 - Epoch: [135][ 980/ 1207] Overall Loss 0.235889 Objective Loss 0.235889 LR 0.000500 Time 0.020113 -2023-02-13 18:18:00,194 - Epoch: [135][ 990/ 1207] Overall Loss 0.235881 Objective Loss 0.235881 LR 0.000500 Time 0.020101 -2023-02-13 18:18:00,384 - Epoch: [135][ 1000/ 1207] Overall Loss 0.235936 Objective Loss 0.235936 LR 0.000500 Time 0.020089 -2023-02-13 18:18:00,573 - Epoch: [135][ 1010/ 1207] Overall Loss 0.236456 Objective Loss 0.236456 LR 0.000500 Time 0.020077 -2023-02-13 18:18:00,763 - Epoch: [135][ 1020/ 1207] Overall Loss 0.236388 Objective Loss 0.236388 LR 0.000500 Time 0.020066 -2023-02-13 18:18:00,954 - Epoch: [135][ 1030/ 1207] Overall Loss 0.236136 Objective Loss 0.236136 LR 0.000500 Time 0.020056 -2023-02-13 18:18:01,144 - Epoch: [135][ 1040/ 1207] Overall Loss 0.236404 Objective Loss 0.236404 LR 0.000500 Time 0.020046 -2023-02-13 18:18:01,335 - Epoch: [135][ 1050/ 1207] Overall Loss 0.236390 Objective Loss 0.236390 LR 0.000500 Time 0.020037 -2023-02-13 18:18:01,524 - Epoch: [135][ 1060/ 1207] Overall Loss 0.236366 Objective Loss 0.236366 LR 0.000500 Time 0.020026 -2023-02-13 18:18:01,715 - Epoch: [135][ 1070/ 1207] Overall Loss 0.236548 Objective Loss 0.236548 LR 0.000500 Time 0.020016 -2023-02-13 18:18:01,905 - Epoch: [135][ 1080/ 1207] Overall Loss 0.236405 Objective Loss 0.236405 LR 0.000500 Time 0.020007 -2023-02-13 18:18:02,095 - Epoch: [135][ 1090/ 1207] Overall Loss 0.236268 Objective Loss 0.236268 LR 0.000500 Time 0.019997 -2023-02-13 18:18:02,285 - Epoch: [135][ 1100/ 1207] Overall Loss 0.236123 Objective Loss 0.236123 LR 0.000500 Time 0.019988 -2023-02-13 18:18:02,475 - Epoch: [135][ 1110/ 1207] Overall Loss 0.235970 Objective Loss 0.235970 LR 0.000500 Time 0.019978 -2023-02-13 18:18:02,664 - Epoch: [135][ 1120/ 1207] Overall Loss 0.236429 Objective Loss 0.236429 LR 0.000500 Time 0.019969 -2023-02-13 18:18:02,854 - Epoch: [135][ 1130/ 1207] Overall Loss 0.236418 Objective Loss 0.236418 LR 0.000500 Time 0.019959 -2023-02-13 18:18:03,051 - Epoch: [135][ 1140/ 1207] Overall Loss 0.236453 Objective Loss 0.236453 LR 0.000500 Time 0.019957 -2023-02-13 18:18:03,247 - Epoch: [135][ 1150/ 1207] Overall Loss 0.236557 Objective Loss 0.236557 LR 0.000500 Time 0.019954 -2023-02-13 18:18:03,440 - Epoch: [135][ 1160/ 1207] Overall Loss 0.236672 Objective Loss 0.236672 LR 0.000500 Time 0.019948 -2023-02-13 18:18:03,637 - Epoch: [135][ 1170/ 1207] Overall Loss 0.236691 Objective Loss 0.236691 LR 0.000500 Time 0.019945 -2023-02-13 18:18:03,830 - Epoch: [135][ 1180/ 1207] Overall Loss 0.236344 Objective Loss 0.236344 LR 0.000500 Time 0.019939 -2023-02-13 18:18:04,027 - Epoch: [135][ 1190/ 1207] Overall Loss 0.236345 Objective Loss 0.236345 LR 0.000500 Time 0.019937 -2023-02-13 18:18:04,271 - Epoch: [135][ 1200/ 1207] Overall Loss 0.236274 Objective Loss 0.236274 LR 0.000500 Time 0.019974 -2023-02-13 18:18:04,386 - Epoch: [135][ 1207/ 1207] Overall Loss 0.236164 Objective Loss 0.236164 Top1 89.634146 Top5 99.085366 LR 0.000500 Time 0.019953 -2023-02-13 18:18:04,459 - --- validate (epoch=135)----------- -2023-02-13 18:18:04,459 - 34311 samples (256 per mini-batch) -2023-02-13 18:18:04,870 - Epoch: [135][ 10/ 135] Loss 0.338933 Top1 84.492188 Top5 97.695312 -2023-02-13 18:18:05,001 - Epoch: [135][ 20/ 135] Loss 0.332060 Top1 85.332031 Top5 98.007812 -2023-02-13 18:18:05,141 - Epoch: [135][ 30/ 135] Loss 0.314063 Top1 85.429688 Top5 97.851562 -2023-02-13 18:18:05,278 - Epoch: [135][ 40/ 135] Loss 0.315745 Top1 85.556641 Top5 97.812500 -2023-02-13 18:18:05,420 - Epoch: [135][ 50/ 135] Loss 0.306903 Top1 85.570312 Top5 97.828125 -2023-02-13 18:18:05,555 - Epoch: [135][ 60/ 135] Loss 0.309160 Top1 85.325521 Top5 97.786458 -2023-02-13 18:18:05,695 - Epoch: [135][ 70/ 135] Loss 0.309418 Top1 85.256696 Top5 97.739955 -2023-02-13 18:18:05,827 - Epoch: [135][ 80/ 135] Loss 0.313615 Top1 85.068359 Top5 97.685547 -2023-02-13 18:18:05,959 - Epoch: [135][ 90/ 135] Loss 0.309774 Top1 85.086806 Top5 97.773438 -2023-02-13 18:18:06,084 - Epoch: [135][ 100/ 135] Loss 0.308770 Top1 85.132812 Top5 97.781250 -2023-02-13 18:18:06,212 - Epoch: [135][ 110/ 135] Loss 0.306554 Top1 85.031960 Top5 97.784091 -2023-02-13 18:18:06,338 - Epoch: [135][ 120/ 135] Loss 0.306416 Top1 85.029297 Top5 97.799479 -2023-02-13 18:18:06,466 - Epoch: [135][ 130/ 135] Loss 0.306709 Top1 84.975962 Top5 97.806490 -2023-02-13 18:18:06,511 - Epoch: [135][ 135/ 135] Loss 0.308815 Top1 85.004809 Top5 97.811198 -2023-02-13 18:18:06,590 - ==> Top1: 85.005 Top5: 97.811 Loss: 0.309 - -2023-02-13 18:18:06,591 - ==> Confusion: -[[ 852 4 5 1 11 1 0 2 5 51 0 4 2 6 5 3 3 2 2 3 5] - [ 2 949 0 1 12 18 3 14 3 1 2 2 2 0 1 3 6 0 5 1 8] - [ 5 5 967 7 5 1 11 16 1 1 3 1 2 5 1 7 1 3 5 5 6] - [ 7 2 21 887 5 5 1 1 2 2 18 0 12 0 18 1 4 3 20 0 7] - [ 8 7 0 0 998 6 1 3 2 3 0 5 0 4 8 8 7 0 1 2 3] - [ 3 17 0 3 7 971 2 17 1 2 4 12 1 13 2 1 5 1 0 3 5] - [ 2 4 23 1 0 7 1029 6 0 2 4 3 2 2 0 2 0 2 1 6 3] - [ 0 6 6 2 3 25 1 937 0 2 3 7 3 1 0 0 0 2 17 5 4] - [ 13 3 1 1 0 1 0 2 899 45 9 1 1 11 11 3 2 0 2 0 4] - [ 58 1 2 1 9 1 0 2 29 872 0 5 1 15 5 2 0 3 1 0 5] - [ 1 1 3 3 2 4 1 6 16 0 980 3 0 12 3 1 0 0 9 1 5] - [ 2 3 2 0 4 9 0 2 2 1 0 936 15 9 0 3 2 6 2 5 2] - [ 1 1 0 7 1 3 0 2 2 1 0 39 865 1 4 7 1 13 1 2 8] - [ 4 1 1 0 10 11 1 3 6 14 6 5 2 947 1 4 0 1 0 0 7] - [ 2 2 0 19 4 4 0 1 23 6 4 1 2 1 1002 1 1 5 6 0 8] - [ 4 3 5 0 7 2 4 1 0 0 0 6 4 2 0 976 8 11 1 7 5] - [ 2 4 0 0 7 2 0 2 3 0 0 0 5 5 1 8 999 0 2 7 14] - [ 3 2 0 2 2 2 1 1 0 0 2 9 15 2 0 16 0 983 0 2 9] - [ 4 6 3 6 0 1 1 27 5 1 3 4 5 0 10 1 2 2 1002 0 3] - [ 0 3 1 1 1 11 5 8 0 1 0 25 4 6 0 4 4 5 0 1060 9] - [ 130 229 222 104 144 210 73 181 94 98 187 166 288 283 135 98 218 108 198 213 10055]] - -2023-02-13 18:18:06,592 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:18:06,592 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:18:06,598 - - -2023-02-13 18:18:06,598 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:18:07,488 - Epoch: [136][ 10/ 1207] Overall Loss 0.219319 Objective Loss 0.219319 LR 0.000500 Time 0.088934 -2023-02-13 18:18:07,688 - Epoch: [136][ 20/ 1207] Overall Loss 0.223804 Objective Loss 0.223804 LR 0.000500 Time 0.054424 -2023-02-13 18:18:07,880 - Epoch: [136][ 30/ 1207] Overall Loss 0.220175 Objective Loss 0.220175 LR 0.000500 Time 0.042662 -2023-02-13 18:18:08,074 - Epoch: [136][ 40/ 1207] Overall Loss 0.229286 Objective Loss 0.229286 LR 0.000500 Time 0.036847 -2023-02-13 18:18:08,265 - Epoch: [136][ 50/ 1207] Overall Loss 0.229339 Objective Loss 0.229339 LR 0.000500 Time 0.033297 -2023-02-13 18:18:08,460 - Epoch: [136][ 60/ 1207] Overall Loss 0.227303 Objective Loss 0.227303 LR 0.000500 Time 0.030985 -2023-02-13 18:18:08,651 - Epoch: [136][ 70/ 1207] Overall Loss 0.228216 Objective Loss 0.228216 LR 0.000500 Time 0.029291 -2023-02-13 18:18:08,846 - Epoch: [136][ 80/ 1207] Overall Loss 0.229274 Objective Loss 0.229274 LR 0.000500 Time 0.028052 -2023-02-13 18:18:09,040 - Epoch: [136][ 90/ 1207] Overall Loss 0.229348 Objective Loss 0.229348 LR 0.000500 Time 0.027085 -2023-02-13 18:18:09,232 - Epoch: [136][ 100/ 1207] Overall Loss 0.230721 Objective Loss 0.230721 LR 0.000500 Time 0.026301 -2023-02-13 18:18:09,428 - Epoch: [136][ 110/ 1207] Overall Loss 0.230817 Objective Loss 0.230817 LR 0.000500 Time 0.025684 -2023-02-13 18:18:09,621 - Epoch: [136][ 120/ 1207] Overall Loss 0.231989 Objective Loss 0.231989 LR 0.000500 Time 0.025149 -2023-02-13 18:18:09,817 - Epoch: [136][ 130/ 1207] Overall Loss 0.231009 Objective Loss 0.231009 LR 0.000500 Time 0.024719 -2023-02-13 18:18:10,010 - Epoch: [136][ 140/ 1207] Overall Loss 0.229856 Objective Loss 0.229856 LR 0.000500 Time 0.024327 -2023-02-13 18:18:10,205 - Epoch: [136][ 150/ 1207] Overall Loss 0.230003 Objective Loss 0.230003 LR 0.000500 Time 0.024006 -2023-02-13 18:18:10,397 - Epoch: [136][ 160/ 1207] Overall Loss 0.230435 Objective Loss 0.230435 LR 0.000500 Time 0.023703 -2023-02-13 18:18:10,592 - Epoch: [136][ 170/ 1207] Overall Loss 0.231667 Objective Loss 0.231667 LR 0.000500 Time 0.023454 -2023-02-13 18:18:10,784 - Epoch: [136][ 180/ 1207] Overall Loss 0.229638 Objective Loss 0.229638 LR 0.000500 Time 0.023218 -2023-02-13 18:18:10,981 - Epoch: [136][ 190/ 1207] Overall Loss 0.229781 Objective Loss 0.229781 LR 0.000500 Time 0.023030 -2023-02-13 18:18:11,174 - Epoch: [136][ 200/ 1207] Overall Loss 0.230575 Objective Loss 0.230575 LR 0.000500 Time 0.022841 -2023-02-13 18:18:11,370 - Epoch: [136][ 210/ 1207] Overall Loss 0.230832 Objective Loss 0.230832 LR 0.000500 Time 0.022685 -2023-02-13 18:18:11,563 - Epoch: [136][ 220/ 1207] Overall Loss 0.231439 Objective Loss 0.231439 LR 0.000500 Time 0.022529 -2023-02-13 18:18:11,759 - Epoch: [136][ 230/ 1207] Overall Loss 0.233144 Objective Loss 0.233144 LR 0.000500 Time 0.022399 -2023-02-13 18:18:11,952 - Epoch: [136][ 240/ 1207] Overall Loss 0.232563 Objective Loss 0.232563 LR 0.000500 Time 0.022269 -2023-02-13 18:18:12,148 - Epoch: [136][ 250/ 1207] Overall Loss 0.231952 Objective Loss 0.231952 LR 0.000500 Time 0.022161 -2023-02-13 18:18:12,341 - Epoch: [136][ 260/ 1207] Overall Loss 0.231499 Objective Loss 0.231499 LR 0.000500 Time 0.022048 -2023-02-13 18:18:12,537 - Epoch: [136][ 270/ 1207] Overall Loss 0.231116 Objective Loss 0.231116 LR 0.000500 Time 0.021957 -2023-02-13 18:18:12,730 - Epoch: [136][ 280/ 1207] Overall Loss 0.232017 Objective Loss 0.232017 LR 0.000500 Time 0.021862 -2023-02-13 18:18:12,926 - Epoch: [136][ 290/ 1207] Overall Loss 0.231859 Objective Loss 0.231859 LR 0.000500 Time 0.021781 -2023-02-13 18:18:13,119 - Epoch: [136][ 300/ 1207] Overall Loss 0.231575 Objective Loss 0.231575 LR 0.000500 Time 0.021697 -2023-02-13 18:18:13,315 - Epoch: [136][ 310/ 1207] Overall Loss 0.232067 Objective Loss 0.232067 LR 0.000500 Time 0.021629 -2023-02-13 18:18:13,508 - Epoch: [136][ 320/ 1207] Overall Loss 0.231844 Objective Loss 0.231844 LR 0.000500 Time 0.021554 -2023-02-13 18:18:13,704 - Epoch: [136][ 330/ 1207] Overall Loss 0.232725 Objective Loss 0.232725 LR 0.000500 Time 0.021494 -2023-02-13 18:18:13,896 - Epoch: [136][ 340/ 1207] Overall Loss 0.233042 Objective Loss 0.233042 LR 0.000500 Time 0.021426 -2023-02-13 18:18:14,092 - Epoch: [136][ 350/ 1207] Overall Loss 0.232244 Objective Loss 0.232244 LR 0.000500 Time 0.021374 -2023-02-13 18:18:14,285 - Epoch: [136][ 360/ 1207] Overall Loss 0.232315 Objective Loss 0.232315 LR 0.000500 Time 0.021313 -2023-02-13 18:18:14,480 - Epoch: [136][ 370/ 1207] Overall Loss 0.233740 Objective Loss 0.233740 LR 0.000500 Time 0.021264 -2023-02-13 18:18:14,673 - Epoch: [136][ 380/ 1207] Overall Loss 0.233919 Objective Loss 0.233919 LR 0.000500 Time 0.021212 -2023-02-13 18:18:14,869 - Epoch: [136][ 390/ 1207] Overall Loss 0.233786 Objective Loss 0.233786 LR 0.000500 Time 0.021168 -2023-02-13 18:18:15,062 - Epoch: [136][ 400/ 1207] Overall Loss 0.233558 Objective Loss 0.233558 LR 0.000500 Time 0.021120 -2023-02-13 18:18:15,258 - Epoch: [136][ 410/ 1207] Overall Loss 0.233332 Objective Loss 0.233332 LR 0.000500 Time 0.021082 -2023-02-13 18:18:15,451 - Epoch: [136][ 420/ 1207] Overall Loss 0.233132 Objective Loss 0.233132 LR 0.000500 Time 0.021039 -2023-02-13 18:18:15,647 - Epoch: [136][ 430/ 1207] Overall Loss 0.234109 Objective Loss 0.234109 LR 0.000500 Time 0.021007 -2023-02-13 18:18:15,840 - Epoch: [136][ 440/ 1207] Overall Loss 0.233832 Objective Loss 0.233832 LR 0.000500 Time 0.020967 -2023-02-13 18:18:16,039 - Epoch: [136][ 450/ 1207] Overall Loss 0.233622 Objective Loss 0.233622 LR 0.000500 Time 0.020941 -2023-02-13 18:18:16,232 - Epoch: [136][ 460/ 1207] Overall Loss 0.233218 Objective Loss 0.233218 LR 0.000500 Time 0.020904 -2023-02-13 18:18:16,428 - Epoch: [136][ 470/ 1207] Overall Loss 0.233415 Objective Loss 0.233415 LR 0.000500 Time 0.020876 -2023-02-13 18:18:16,622 - Epoch: [136][ 480/ 1207] Overall Loss 0.233017 Objective Loss 0.233017 LR 0.000500 Time 0.020845 -2023-02-13 18:18:16,818 - Epoch: [136][ 490/ 1207] Overall Loss 0.233222 Objective Loss 0.233222 LR 0.000500 Time 0.020819 -2023-02-13 18:18:17,012 - Epoch: [136][ 500/ 1207] Overall Loss 0.233351 Objective Loss 0.233351 LR 0.000500 Time 0.020789 -2023-02-13 18:18:17,208 - Epoch: [136][ 510/ 1207] Overall Loss 0.233605 Objective Loss 0.233605 LR 0.000500 Time 0.020765 -2023-02-13 18:18:17,401 - Epoch: [136][ 520/ 1207] Overall Loss 0.234115 Objective Loss 0.234115 LR 0.000500 Time 0.020737 -2023-02-13 18:18:17,597 - Epoch: [136][ 530/ 1207] Overall Loss 0.233573 Objective Loss 0.233573 LR 0.000500 Time 0.020715 -2023-02-13 18:18:17,790 - Epoch: [136][ 540/ 1207] Overall Loss 0.233751 Objective Loss 0.233751 LR 0.000500 Time 0.020688 -2023-02-13 18:18:17,986 - Epoch: [136][ 550/ 1207] Overall Loss 0.233605 Objective Loss 0.233605 LR 0.000500 Time 0.020668 -2023-02-13 18:18:18,180 - Epoch: [136][ 560/ 1207] Overall Loss 0.233472 Objective Loss 0.233472 LR 0.000500 Time 0.020643 -2023-02-13 18:18:18,376 - Epoch: [136][ 570/ 1207] Overall Loss 0.233125 Objective Loss 0.233125 LR 0.000500 Time 0.020624 -2023-02-13 18:18:18,568 - Epoch: [136][ 580/ 1207] Overall Loss 0.232916 Objective Loss 0.232916 LR 0.000500 Time 0.020600 -2023-02-13 18:18:18,764 - Epoch: [136][ 590/ 1207] Overall Loss 0.233119 Objective Loss 0.233119 LR 0.000500 Time 0.020582 -2023-02-13 18:18:18,957 - Epoch: [136][ 600/ 1207] Overall Loss 0.233059 Objective Loss 0.233059 LR 0.000500 Time 0.020560 -2023-02-13 18:18:19,153 - Epoch: [136][ 610/ 1207] Overall Loss 0.233268 Objective Loss 0.233268 LR 0.000500 Time 0.020544 -2023-02-13 18:18:19,346 - Epoch: [136][ 620/ 1207] Overall Loss 0.233311 Objective Loss 0.233311 LR 0.000500 Time 0.020522 -2023-02-13 18:18:19,541 - Epoch: [136][ 630/ 1207] Overall Loss 0.233391 Objective Loss 0.233391 LR 0.000500 Time 0.020507 -2023-02-13 18:18:19,735 - Epoch: [136][ 640/ 1207] Overall Loss 0.233633 Objective Loss 0.233633 LR 0.000500 Time 0.020488 -2023-02-13 18:18:19,931 - Epoch: [136][ 650/ 1207] Overall Loss 0.233939 Objective Loss 0.233939 LR 0.000500 Time 0.020474 -2023-02-13 18:18:20,124 - Epoch: [136][ 660/ 1207] Overall Loss 0.234061 Objective Loss 0.234061 LR 0.000500 Time 0.020455 -2023-02-13 18:18:20,320 - Epoch: [136][ 670/ 1207] Overall Loss 0.234262 Objective Loss 0.234262 LR 0.000500 Time 0.020442 -2023-02-13 18:18:20,512 - Epoch: [136][ 680/ 1207] Overall Loss 0.234107 Objective Loss 0.234107 LR 0.000500 Time 0.020424 -2023-02-13 18:18:20,709 - Epoch: [136][ 690/ 1207] Overall Loss 0.234002 Objective Loss 0.234002 LR 0.000500 Time 0.020412 -2023-02-13 18:18:20,902 - Epoch: [136][ 700/ 1207] Overall Loss 0.233888 Objective Loss 0.233888 LR 0.000500 Time 0.020396 -2023-02-13 18:18:21,098 - Epoch: [136][ 710/ 1207] Overall Loss 0.233891 Objective Loss 0.233891 LR 0.000500 Time 0.020385 -2023-02-13 18:18:21,291 - Epoch: [136][ 720/ 1207] Overall Loss 0.233519 Objective Loss 0.233519 LR 0.000500 Time 0.020369 -2023-02-13 18:18:21,487 - Epoch: [136][ 730/ 1207] Overall Loss 0.233358 Objective Loss 0.233358 LR 0.000500 Time 0.020358 -2023-02-13 18:18:21,680 - Epoch: [136][ 740/ 1207] Overall Loss 0.233364 Objective Loss 0.233364 LR 0.000500 Time 0.020343 -2023-02-13 18:18:21,876 - Epoch: [136][ 750/ 1207] Overall Loss 0.233638 Objective Loss 0.233638 LR 0.000500 Time 0.020333 -2023-02-13 18:18:22,067 - Epoch: [136][ 760/ 1207] Overall Loss 0.233698 Objective Loss 0.233698 LR 0.000500 Time 0.020316 -2023-02-13 18:18:22,256 - Epoch: [136][ 770/ 1207] Overall Loss 0.234031 Objective Loss 0.234031 LR 0.000500 Time 0.020297 -2023-02-13 18:18:22,445 - Epoch: [136][ 780/ 1207] Overall Loss 0.234085 Objective Loss 0.234085 LR 0.000500 Time 0.020279 -2023-02-13 18:18:22,635 - Epoch: [136][ 790/ 1207] Overall Loss 0.234298 Objective Loss 0.234298 LR 0.000500 Time 0.020261 -2023-02-13 18:18:22,824 - Epoch: [136][ 800/ 1207] Overall Loss 0.234553 Objective Loss 0.234553 LR 0.000500 Time 0.020244 -2023-02-13 18:18:23,014 - Epoch: [136][ 810/ 1207] Overall Loss 0.234423 Objective Loss 0.234423 LR 0.000500 Time 0.020228 -2023-02-13 18:18:23,203 - Epoch: [136][ 820/ 1207] Overall Loss 0.234340 Objective Loss 0.234340 LR 0.000500 Time 0.020212 -2023-02-13 18:18:23,393 - Epoch: [136][ 830/ 1207] Overall Loss 0.234532 Objective Loss 0.234532 LR 0.000500 Time 0.020197 -2023-02-13 18:18:23,582 - Epoch: [136][ 840/ 1207] Overall Loss 0.234451 Objective Loss 0.234451 LR 0.000500 Time 0.020181 -2023-02-13 18:18:23,772 - Epoch: [136][ 850/ 1207] Overall Loss 0.234551 Objective Loss 0.234551 LR 0.000500 Time 0.020167 -2023-02-13 18:18:23,961 - Epoch: [136][ 860/ 1207] Overall Loss 0.234957 Objective Loss 0.234957 LR 0.000500 Time 0.020152 -2023-02-13 18:18:24,151 - Epoch: [136][ 870/ 1207] Overall Loss 0.235103 Objective Loss 0.235103 LR 0.000500 Time 0.020138 -2023-02-13 18:18:24,340 - Epoch: [136][ 880/ 1207] Overall Loss 0.235250 Objective Loss 0.235250 LR 0.000500 Time 0.020123 -2023-02-13 18:18:24,530 - Epoch: [136][ 890/ 1207] Overall Loss 0.235375 Objective Loss 0.235375 LR 0.000500 Time 0.020110 -2023-02-13 18:18:24,720 - Epoch: [136][ 900/ 1207] Overall Loss 0.235378 Objective Loss 0.235378 LR 0.000500 Time 0.020097 -2023-02-13 18:18:24,909 - Epoch: [136][ 910/ 1207] Overall Loss 0.235380 Objective Loss 0.235380 LR 0.000500 Time 0.020084 -2023-02-13 18:18:25,099 - Epoch: [136][ 920/ 1207] Overall Loss 0.235449 Objective Loss 0.235449 LR 0.000500 Time 0.020072 -2023-02-13 18:18:25,289 - Epoch: [136][ 930/ 1207] Overall Loss 0.235945 Objective Loss 0.235945 LR 0.000500 Time 0.020059 -2023-02-13 18:18:25,478 - Epoch: [136][ 940/ 1207] Overall Loss 0.236016 Objective Loss 0.236016 LR 0.000500 Time 0.020047 -2023-02-13 18:18:25,668 - Epoch: [136][ 950/ 1207] Overall Loss 0.236187 Objective Loss 0.236187 LR 0.000500 Time 0.020036 -2023-02-13 18:18:25,858 - Epoch: [136][ 960/ 1207] Overall Loss 0.236121 Objective Loss 0.236121 LR 0.000500 Time 0.020024 -2023-02-13 18:18:26,048 - Epoch: [136][ 970/ 1207] Overall Loss 0.236025 Objective Loss 0.236025 LR 0.000500 Time 0.020013 -2023-02-13 18:18:26,237 - Epoch: [136][ 980/ 1207] Overall Loss 0.236230 Objective Loss 0.236230 LR 0.000500 Time 0.020002 -2023-02-13 18:18:26,426 - Epoch: [136][ 990/ 1207] Overall Loss 0.236382 Objective Loss 0.236382 LR 0.000500 Time 0.019991 -2023-02-13 18:18:26,616 - Epoch: [136][ 1000/ 1207] Overall Loss 0.236319 Objective Loss 0.236319 LR 0.000500 Time 0.019980 -2023-02-13 18:18:26,806 - Epoch: [136][ 1010/ 1207] Overall Loss 0.236072 Objective Loss 0.236072 LR 0.000500 Time 0.019970 -2023-02-13 18:18:26,996 - Epoch: [136][ 1020/ 1207] Overall Loss 0.236256 Objective Loss 0.236256 LR 0.000500 Time 0.019960 -2023-02-13 18:18:27,186 - Epoch: [136][ 1030/ 1207] Overall Loss 0.236400 Objective Loss 0.236400 LR 0.000500 Time 0.019951 -2023-02-13 18:18:27,376 - Epoch: [136][ 1040/ 1207] Overall Loss 0.236485 Objective Loss 0.236485 LR 0.000500 Time 0.019941 -2023-02-13 18:18:27,566 - Epoch: [136][ 1050/ 1207] Overall Loss 0.236539 Objective Loss 0.236539 LR 0.000500 Time 0.019932 -2023-02-13 18:18:27,756 - Epoch: [136][ 1060/ 1207] Overall Loss 0.236536 Objective Loss 0.236536 LR 0.000500 Time 0.019922 -2023-02-13 18:18:27,946 - Epoch: [136][ 1070/ 1207] Overall Loss 0.236469 Objective Loss 0.236469 LR 0.000500 Time 0.019913 -2023-02-13 18:18:28,135 - Epoch: [136][ 1080/ 1207] Overall Loss 0.236756 Objective Loss 0.236756 LR 0.000500 Time 0.019904 -2023-02-13 18:18:28,325 - Epoch: [136][ 1090/ 1207] Overall Loss 0.236860 Objective Loss 0.236860 LR 0.000500 Time 0.019895 -2023-02-13 18:18:28,515 - Epoch: [136][ 1100/ 1207] Overall Loss 0.236946 Objective Loss 0.236946 LR 0.000500 Time 0.019887 -2023-02-13 18:18:28,705 - Epoch: [136][ 1110/ 1207] Overall Loss 0.236851 Objective Loss 0.236851 LR 0.000500 Time 0.019878 -2023-02-13 18:18:28,894 - Epoch: [136][ 1120/ 1207] Overall Loss 0.236824 Objective Loss 0.236824 LR 0.000500 Time 0.019869 -2023-02-13 18:18:29,084 - Epoch: [136][ 1130/ 1207] Overall Loss 0.236826 Objective Loss 0.236826 LR 0.000500 Time 0.019861 -2023-02-13 18:18:29,273 - Epoch: [136][ 1140/ 1207] Overall Loss 0.236993 Objective Loss 0.236993 LR 0.000500 Time 0.019852 -2023-02-13 18:18:29,462 - Epoch: [136][ 1150/ 1207] Overall Loss 0.237116 Objective Loss 0.237116 LR 0.000500 Time 0.019844 -2023-02-13 18:18:29,651 - Epoch: [136][ 1160/ 1207] Overall Loss 0.237065 Objective Loss 0.237065 LR 0.000500 Time 0.019836 -2023-02-13 18:18:29,841 - Epoch: [136][ 1170/ 1207] Overall Loss 0.236954 Objective Loss 0.236954 LR 0.000500 Time 0.019828 -2023-02-13 18:18:30,030 - Epoch: [136][ 1180/ 1207] Overall Loss 0.236713 Objective Loss 0.236713 LR 0.000500 Time 0.019820 -2023-02-13 18:18:30,220 - Epoch: [136][ 1190/ 1207] Overall Loss 0.236600 Objective Loss 0.236600 LR 0.000500 Time 0.019812 -2023-02-13 18:18:30,464 - Epoch: [136][ 1200/ 1207] Overall Loss 0.236584 Objective Loss 0.236584 LR 0.000500 Time 0.019851 -2023-02-13 18:18:30,579 - Epoch: [136][ 1207/ 1207] Overall Loss 0.236665 Objective Loss 0.236665 Top1 86.890244 Top5 99.085366 LR 0.000500 Time 0.019830 -2023-02-13 18:18:30,650 - --- validate (epoch=136)----------- -2023-02-13 18:18:30,650 - 34311 samples (256 per mini-batch) -2023-02-13 18:18:31,059 - Epoch: [136][ 10/ 135] Loss 0.315801 Top1 84.414062 Top5 97.734375 -2023-02-13 18:18:31,188 - Epoch: [136][ 20/ 135] Loss 0.306119 Top1 85.117188 Top5 97.753906 -2023-02-13 18:18:31,319 - Epoch: [136][ 30/ 135] Loss 0.313687 Top1 85.104167 Top5 97.682292 -2023-02-13 18:18:31,452 - Epoch: [136][ 40/ 135] Loss 0.310544 Top1 84.990234 Top5 97.666016 -2023-02-13 18:18:31,580 - Epoch: [136][ 50/ 135] Loss 0.308599 Top1 84.984375 Top5 97.640625 -2023-02-13 18:18:31,710 - Epoch: [136][ 60/ 135] Loss 0.308919 Top1 85.065104 Top5 97.656250 -2023-02-13 18:18:31,838 - Epoch: [136][ 70/ 135] Loss 0.308604 Top1 85.156250 Top5 97.622768 -2023-02-13 18:18:31,968 - Epoch: [136][ 80/ 135] Loss 0.307666 Top1 85.151367 Top5 97.680664 -2023-02-13 18:18:32,097 - Epoch: [136][ 90/ 135] Loss 0.309898 Top1 85.099826 Top5 97.656250 -2023-02-13 18:18:32,227 - Epoch: [136][ 100/ 135] Loss 0.310106 Top1 85.074219 Top5 97.687500 -2023-02-13 18:18:32,357 - Epoch: [136][ 110/ 135] Loss 0.308940 Top1 85.085227 Top5 97.723722 -2023-02-13 18:18:32,487 - Epoch: [136][ 120/ 135] Loss 0.309814 Top1 85.016276 Top5 97.698568 -2023-02-13 18:18:32,621 - Epoch: [136][ 130/ 135] Loss 0.310595 Top1 85.051082 Top5 97.710337 -2023-02-13 18:18:32,668 - Epoch: [136][ 135/ 135] Loss 0.309095 Top1 85.100988 Top5 97.726677 -2023-02-13 18:18:32,738 - ==> Top1: 85.101 Top5: 97.727 Loss: 0.309 - -2023-02-13 18:18:32,739 - ==> Confusion: -[[ 866 5 10 2 13 1 0 1 3 35 0 6 0 2 3 3 4 2 1 4 6] - [ 2 947 2 2 13 15 1 17 4 1 1 1 2 0 2 1 4 1 6 1 10] - [ 8 2 969 11 6 0 16 13 0 1 1 1 1 5 1 7 1 4 4 5 2] - [ 4 2 21 909 2 4 0 2 0 3 8 0 6 1 19 1 4 6 17 1 6] - [ 10 7 0 0 993 6 1 4 0 3 0 6 1 2 11 5 8 2 0 4 3] - [ 3 21 1 6 7 944 2 23 0 5 2 13 2 18 2 1 3 0 2 8 7] - [ 3 4 21 2 1 4 1030 7 0 1 6 2 0 2 0 1 0 3 1 5 6] - [ 2 5 8 1 2 23 1 933 1 2 1 5 5 2 0 0 0 2 17 11 3] - [ 21 3 1 2 1 0 0 2 883 48 8 1 0 8 16 3 2 1 5 1 3] - [ 85 2 5 1 14 1 0 3 25 853 0 2 0 11 1 1 2 3 1 1 1] - [ 1 2 9 11 1 3 1 7 9 1 979 0 2 7 3 0 2 0 10 0 3] - [ 2 2 0 0 1 7 0 4 0 3 0 935 16 8 1 3 2 7 1 10 3] - [ 1 0 1 7 2 1 0 1 2 2 1 38 862 0 3 7 4 13 2 2 10] - [ 6 3 2 0 8 7 0 3 11 15 14 3 4 931 2 2 2 2 0 3 6] - [ 6 3 3 19 3 4 0 2 18 6 2 1 4 0 997 0 3 5 8 0 8] - [ 6 4 7 0 6 0 4 1 2 0 0 11 3 2 1 962 11 11 1 8 6] - [ 4 3 1 1 10 0 0 1 2 0 0 2 2 1 0 8 1001 1 1 7 16] - [ 5 3 0 7 2 0 3 0 0 1 1 12 12 0 0 17 1 975 2 5 5] - [ 3 3 6 9 0 2 0 27 2 0 2 3 2 0 12 1 0 3 1004 4 3] - [ 0 3 4 0 0 7 4 9 0 0 1 16 1 3 0 5 6 2 1 1079 7] - [ 158 225 270 118 155 146 78 172 81 77 157 127 275 261 133 79 228 95 190 262 10147]] - -2023-02-13 18:18:32,740 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:18:32,740 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:18:32,746 - - -2023-02-13 18:18:32,746 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:18:33,744 - Epoch: [137][ 10/ 1207] Overall Loss 0.228040 Objective Loss 0.228040 LR 0.000500 Time 0.099722 -2023-02-13 18:18:33,945 - Epoch: [137][ 20/ 1207] Overall Loss 0.239683 Objective Loss 0.239683 LR 0.000500 Time 0.059908 -2023-02-13 18:18:34,140 - Epoch: [137][ 30/ 1207] Overall Loss 0.235743 Objective Loss 0.235743 LR 0.000500 Time 0.046429 -2023-02-13 18:18:34,336 - Epoch: [137][ 40/ 1207] Overall Loss 0.227856 Objective Loss 0.227856 LR 0.000500 Time 0.039699 -2023-02-13 18:18:34,530 - Epoch: [137][ 50/ 1207] Overall Loss 0.222696 Objective Loss 0.222696 LR 0.000500 Time 0.035643 -2023-02-13 18:18:34,726 - Epoch: [137][ 60/ 1207] Overall Loss 0.222744 Objective Loss 0.222744 LR 0.000500 Time 0.032958 -2023-02-13 18:18:34,920 - Epoch: [137][ 70/ 1207] Overall Loss 0.226225 Objective Loss 0.226225 LR 0.000500 Time 0.031021 -2023-02-13 18:18:35,116 - Epoch: [137][ 80/ 1207] Overall Loss 0.227748 Objective Loss 0.227748 LR 0.000500 Time 0.029581 -2023-02-13 18:18:35,310 - Epoch: [137][ 90/ 1207] Overall Loss 0.231621 Objective Loss 0.231621 LR 0.000500 Time 0.028447 -2023-02-13 18:18:35,505 - Epoch: [137][ 100/ 1207] Overall Loss 0.232314 Objective Loss 0.232314 LR 0.000500 Time 0.027549 -2023-02-13 18:18:35,700 - Epoch: [137][ 110/ 1207] Overall Loss 0.232581 Objective Loss 0.232581 LR 0.000500 Time 0.026813 -2023-02-13 18:18:35,896 - Epoch: [137][ 120/ 1207] Overall Loss 0.231399 Objective Loss 0.231399 LR 0.000500 Time 0.026210 -2023-02-13 18:18:36,090 - Epoch: [137][ 130/ 1207] Overall Loss 0.231262 Objective Loss 0.231262 LR 0.000500 Time 0.025681 -2023-02-13 18:18:36,286 - Epoch: [137][ 140/ 1207] Overall Loss 0.232827 Objective Loss 0.232827 LR 0.000500 Time 0.025242 -2023-02-13 18:18:36,480 - Epoch: [137][ 150/ 1207] Overall Loss 0.233438 Objective Loss 0.233438 LR 0.000500 Time 0.024852 -2023-02-13 18:18:36,676 - Epoch: [137][ 160/ 1207] Overall Loss 0.233764 Objective Loss 0.233764 LR 0.000500 Time 0.024520 -2023-02-13 18:18:36,870 - Epoch: [137][ 170/ 1207] Overall Loss 0.234498 Objective Loss 0.234498 LR 0.000500 Time 0.024220 -2023-02-13 18:18:37,067 - Epoch: [137][ 180/ 1207] Overall Loss 0.232905 Objective Loss 0.232905 LR 0.000500 Time 0.023963 -2023-02-13 18:18:37,261 - Epoch: [137][ 190/ 1207] Overall Loss 0.233837 Objective Loss 0.233837 LR 0.000500 Time 0.023723 -2023-02-13 18:18:37,457 - Epoch: [137][ 200/ 1207] Overall Loss 0.234157 Objective Loss 0.234157 LR 0.000500 Time 0.023515 -2023-02-13 18:18:37,651 - Epoch: [137][ 210/ 1207] Overall Loss 0.233618 Objective Loss 0.233618 LR 0.000500 Time 0.023319 -2023-02-13 18:18:37,847 - Epoch: [137][ 220/ 1207] Overall Loss 0.233075 Objective Loss 0.233075 LR 0.000500 Time 0.023149 -2023-02-13 18:18:38,043 - Epoch: [137][ 230/ 1207] Overall Loss 0.234431 Objective Loss 0.234431 LR 0.000500 Time 0.022990 -2023-02-13 18:18:38,239 - Epoch: [137][ 240/ 1207] Overall Loss 0.234584 Objective Loss 0.234584 LR 0.000500 Time 0.022846 -2023-02-13 18:18:38,433 - Epoch: [137][ 250/ 1207] Overall Loss 0.234267 Objective Loss 0.234267 LR 0.000500 Time 0.022710 -2023-02-13 18:18:38,629 - Epoch: [137][ 260/ 1207] Overall Loss 0.234281 Objective Loss 0.234281 LR 0.000500 Time 0.022588 -2023-02-13 18:18:38,824 - Epoch: [137][ 270/ 1207] Overall Loss 0.235048 Objective Loss 0.235048 LR 0.000500 Time 0.022472 -2023-02-13 18:18:39,020 - Epoch: [137][ 280/ 1207] Overall Loss 0.235182 Objective Loss 0.235182 LR 0.000500 Time 0.022370 -2023-02-13 18:18:39,215 - Epoch: [137][ 290/ 1207] Overall Loss 0.234424 Objective Loss 0.234424 LR 0.000500 Time 0.022269 -2023-02-13 18:18:39,411 - Epoch: [137][ 300/ 1207] Overall Loss 0.234980 Objective Loss 0.234980 LR 0.000500 Time 0.022177 -2023-02-13 18:18:39,605 - Epoch: [137][ 310/ 1207] Overall Loss 0.233465 Objective Loss 0.233465 LR 0.000500 Time 0.022086 -2023-02-13 18:18:39,802 - Epoch: [137][ 320/ 1207] Overall Loss 0.232764 Objective Loss 0.232764 LR 0.000500 Time 0.022009 -2023-02-13 18:18:39,997 - Epoch: [137][ 330/ 1207] Overall Loss 0.232368 Objective Loss 0.232368 LR 0.000500 Time 0.021933 -2023-02-13 18:18:40,194 - Epoch: [137][ 340/ 1207] Overall Loss 0.231998 Objective Loss 0.231998 LR 0.000500 Time 0.021868 -2023-02-13 18:18:40,390 - Epoch: [137][ 350/ 1207] Overall Loss 0.232619 Objective Loss 0.232619 LR 0.000500 Time 0.021801 -2023-02-13 18:18:40,587 - Epoch: [137][ 360/ 1207] Overall Loss 0.232331 Objective Loss 0.232331 LR 0.000500 Time 0.021743 -2023-02-13 18:18:40,783 - Epoch: [137][ 370/ 1207] Overall Loss 0.232290 Objective Loss 0.232290 LR 0.000500 Time 0.021684 -2023-02-13 18:18:40,982 - Epoch: [137][ 380/ 1207] Overall Loss 0.231586 Objective Loss 0.231586 LR 0.000500 Time 0.021635 -2023-02-13 18:18:41,178 - Epoch: [137][ 390/ 1207] Overall Loss 0.231622 Objective Loss 0.231622 LR 0.000500 Time 0.021582 -2023-02-13 18:18:41,377 - Epoch: [137][ 400/ 1207] Overall Loss 0.232556 Objective Loss 0.232556 LR 0.000500 Time 0.021537 -2023-02-13 18:18:41,572 - Epoch: [137][ 410/ 1207] Overall Loss 0.232424 Objective Loss 0.232424 LR 0.000500 Time 0.021488 -2023-02-13 18:18:41,771 - Epoch: [137][ 420/ 1207] Overall Loss 0.232340 Objective Loss 0.232340 LR 0.000500 Time 0.021449 -2023-02-13 18:18:41,967 - Epoch: [137][ 430/ 1207] Overall Loss 0.231924 Objective Loss 0.231924 LR 0.000500 Time 0.021405 -2023-02-13 18:18:42,165 - Epoch: [137][ 440/ 1207] Overall Loss 0.232607 Objective Loss 0.232607 LR 0.000500 Time 0.021367 -2023-02-13 18:18:42,360 - Epoch: [137][ 450/ 1207] Overall Loss 0.232741 Objective Loss 0.232741 LR 0.000500 Time 0.021325 -2023-02-13 18:18:42,558 - Epoch: [137][ 460/ 1207] Overall Loss 0.232502 Objective Loss 0.232502 LR 0.000500 Time 0.021292 -2023-02-13 18:18:42,754 - Epoch: [137][ 470/ 1207] Overall Loss 0.232082 Objective Loss 0.232082 LR 0.000500 Time 0.021255 -2023-02-13 18:18:42,953 - Epoch: [137][ 480/ 1207] Overall Loss 0.231954 Objective Loss 0.231954 LR 0.000500 Time 0.021225 -2023-02-13 18:18:43,149 - Epoch: [137][ 490/ 1207] Overall Loss 0.231508 Objective Loss 0.231508 LR 0.000500 Time 0.021192 -2023-02-13 18:18:43,347 - Epoch: [137][ 500/ 1207] Overall Loss 0.231407 Objective Loss 0.231407 LR 0.000500 Time 0.021164 -2023-02-13 18:18:43,544 - Epoch: [137][ 510/ 1207] Overall Loss 0.231614 Objective Loss 0.231614 LR 0.000500 Time 0.021133 -2023-02-13 18:18:43,742 - Epoch: [137][ 520/ 1207] Overall Loss 0.231850 Objective Loss 0.231850 LR 0.000500 Time 0.021107 -2023-02-13 18:18:43,938 - Epoch: [137][ 530/ 1207] Overall Loss 0.231655 Objective Loss 0.231655 LR 0.000500 Time 0.021079 -2023-02-13 18:18:44,137 - Epoch: [137][ 540/ 1207] Overall Loss 0.231761 Objective Loss 0.231761 LR 0.000500 Time 0.021056 -2023-02-13 18:18:44,334 - Epoch: [137][ 550/ 1207] Overall Loss 0.231890 Objective Loss 0.231890 LR 0.000500 Time 0.021029 -2023-02-13 18:18:44,532 - Epoch: [137][ 560/ 1207] Overall Loss 0.232866 Objective Loss 0.232866 LR 0.000500 Time 0.021008 -2023-02-13 18:18:44,727 - Epoch: [137][ 570/ 1207] Overall Loss 0.232832 Objective Loss 0.232832 LR 0.000500 Time 0.020981 -2023-02-13 18:18:44,926 - Epoch: [137][ 580/ 1207] Overall Loss 0.233736 Objective Loss 0.233736 LR 0.000500 Time 0.020961 -2023-02-13 18:18:45,122 - Epoch: [137][ 590/ 1207] Overall Loss 0.233787 Objective Loss 0.233787 LR 0.000500 Time 0.020937 -2023-02-13 18:18:45,320 - Epoch: [137][ 600/ 1207] Overall Loss 0.233914 Objective Loss 0.233914 LR 0.000500 Time 0.020917 -2023-02-13 18:18:45,515 - Epoch: [137][ 610/ 1207] Overall Loss 0.234294 Objective Loss 0.234294 LR 0.000500 Time 0.020895 -2023-02-13 18:18:45,714 - Epoch: [137][ 620/ 1207] Overall Loss 0.234385 Objective Loss 0.234385 LR 0.000500 Time 0.020877 -2023-02-13 18:18:45,911 - Epoch: [137][ 630/ 1207] Overall Loss 0.234600 Objective Loss 0.234600 LR 0.000500 Time 0.020858 -2023-02-13 18:18:46,110 - Epoch: [137][ 640/ 1207] Overall Loss 0.234445 Objective Loss 0.234445 LR 0.000500 Time 0.020842 -2023-02-13 18:18:46,306 - Epoch: [137][ 650/ 1207] Overall Loss 0.234216 Objective Loss 0.234216 LR 0.000500 Time 0.020822 -2023-02-13 18:18:46,504 - Epoch: [137][ 660/ 1207] Overall Loss 0.234061 Objective Loss 0.234061 LR 0.000500 Time 0.020807 -2023-02-13 18:18:46,701 - Epoch: [137][ 670/ 1207] Overall Loss 0.234225 Objective Loss 0.234225 LR 0.000500 Time 0.020789 -2023-02-13 18:18:46,901 - Epoch: [137][ 680/ 1207] Overall Loss 0.234676 Objective Loss 0.234676 LR 0.000500 Time 0.020777 -2023-02-13 18:18:47,097 - Epoch: [137][ 690/ 1207] Overall Loss 0.234778 Objective Loss 0.234778 LR 0.000500 Time 0.020760 -2023-02-13 18:18:47,295 - Epoch: [137][ 700/ 1207] Overall Loss 0.234731 Objective Loss 0.234731 LR 0.000500 Time 0.020745 -2023-02-13 18:18:47,490 - Epoch: [137][ 710/ 1207] Overall Loss 0.234484 Objective Loss 0.234484 LR 0.000500 Time 0.020728 -2023-02-13 18:18:47,689 - Epoch: [137][ 720/ 1207] Overall Loss 0.234754 Objective Loss 0.234754 LR 0.000500 Time 0.020715 -2023-02-13 18:18:47,885 - Epoch: [137][ 730/ 1207] Overall Loss 0.234751 Objective Loss 0.234751 LR 0.000500 Time 0.020700 -2023-02-13 18:18:48,083 - Epoch: [137][ 740/ 1207] Overall Loss 0.234672 Objective Loss 0.234672 LR 0.000500 Time 0.020687 -2023-02-13 18:18:48,280 - Epoch: [137][ 750/ 1207] Overall Loss 0.234626 Objective Loss 0.234626 LR 0.000500 Time 0.020673 -2023-02-13 18:18:48,478 - Epoch: [137][ 760/ 1207] Overall Loss 0.234836 Objective Loss 0.234836 LR 0.000500 Time 0.020661 -2023-02-13 18:18:48,674 - Epoch: [137][ 770/ 1207] Overall Loss 0.234901 Objective Loss 0.234901 LR 0.000500 Time 0.020646 -2023-02-13 18:18:48,873 - Epoch: [137][ 780/ 1207] Overall Loss 0.234796 Objective Loss 0.234796 LR 0.000500 Time 0.020636 -2023-02-13 18:18:49,069 - Epoch: [137][ 790/ 1207] Overall Loss 0.234674 Objective Loss 0.234674 LR 0.000500 Time 0.020623 -2023-02-13 18:18:49,267 - Epoch: [137][ 800/ 1207] Overall Loss 0.235068 Objective Loss 0.235068 LR 0.000500 Time 0.020613 -2023-02-13 18:18:49,463 - Epoch: [137][ 810/ 1207] Overall Loss 0.235173 Objective Loss 0.235173 LR 0.000500 Time 0.020600 -2023-02-13 18:18:49,661 - Epoch: [137][ 820/ 1207] Overall Loss 0.235143 Objective Loss 0.235143 LR 0.000500 Time 0.020590 -2023-02-13 18:18:49,858 - Epoch: [137][ 830/ 1207] Overall Loss 0.235240 Objective Loss 0.235240 LR 0.000500 Time 0.020578 -2023-02-13 18:18:50,057 - Epoch: [137][ 840/ 1207] Overall Loss 0.234994 Objective Loss 0.234994 LR 0.000500 Time 0.020569 -2023-02-13 18:18:50,253 - Epoch: [137][ 850/ 1207] Overall Loss 0.234839 Objective Loss 0.234839 LR 0.000500 Time 0.020558 -2023-02-13 18:18:50,452 - Epoch: [137][ 860/ 1207] Overall Loss 0.234595 Objective Loss 0.234595 LR 0.000500 Time 0.020549 -2023-02-13 18:18:50,647 - Epoch: [137][ 870/ 1207] Overall Loss 0.234603 Objective Loss 0.234603 LR 0.000500 Time 0.020537 -2023-02-13 18:18:50,845 - Epoch: [137][ 880/ 1207] Overall Loss 0.234707 Objective Loss 0.234707 LR 0.000500 Time 0.020529 -2023-02-13 18:18:51,043 - Epoch: [137][ 890/ 1207] Overall Loss 0.234563 Objective Loss 0.234563 LR 0.000500 Time 0.020520 -2023-02-13 18:18:51,242 - Epoch: [137][ 900/ 1207] Overall Loss 0.234687 Objective Loss 0.234687 LR 0.000500 Time 0.020512 -2023-02-13 18:18:51,437 - Epoch: [137][ 910/ 1207] Overall Loss 0.234466 Objective Loss 0.234466 LR 0.000500 Time 0.020501 -2023-02-13 18:18:51,635 - Epoch: [137][ 920/ 1207] Overall Loss 0.234467 Objective Loss 0.234467 LR 0.000500 Time 0.020493 -2023-02-13 18:18:51,832 - Epoch: [137][ 930/ 1207] Overall Loss 0.234533 Objective Loss 0.234533 LR 0.000500 Time 0.020484 -2023-02-13 18:18:52,032 - Epoch: [137][ 940/ 1207] Overall Loss 0.234842 Objective Loss 0.234842 LR 0.000500 Time 0.020478 -2023-02-13 18:18:52,228 - Epoch: [137][ 950/ 1207] Overall Loss 0.235099 Objective Loss 0.235099 LR 0.000500 Time 0.020469 -2023-02-13 18:18:52,427 - Epoch: [137][ 960/ 1207] Overall Loss 0.235321 Objective Loss 0.235321 LR 0.000500 Time 0.020462 -2023-02-13 18:18:52,622 - Epoch: [137][ 970/ 1207] Overall Loss 0.235545 Objective Loss 0.235545 LR 0.000500 Time 0.020453 -2023-02-13 18:18:52,821 - Epoch: [137][ 980/ 1207] Overall Loss 0.235667 Objective Loss 0.235667 LR 0.000500 Time 0.020447 -2023-02-13 18:18:53,017 - Epoch: [137][ 990/ 1207] Overall Loss 0.235678 Objective Loss 0.235678 LR 0.000500 Time 0.020437 -2023-02-13 18:18:53,215 - Epoch: [137][ 1000/ 1207] Overall Loss 0.235597 Objective Loss 0.235597 LR 0.000500 Time 0.020431 -2023-02-13 18:18:53,411 - Epoch: [137][ 1010/ 1207] Overall Loss 0.235396 Objective Loss 0.235396 LR 0.000500 Time 0.020422 -2023-02-13 18:18:53,608 - Epoch: [137][ 1020/ 1207] Overall Loss 0.235145 Objective Loss 0.235145 LR 0.000500 Time 0.020414 -2023-02-13 18:18:53,805 - Epoch: [137][ 1030/ 1207] Overall Loss 0.235037 Objective Loss 0.235037 LR 0.000500 Time 0.020407 -2023-02-13 18:18:54,004 - Epoch: [137][ 1040/ 1207] Overall Loss 0.235101 Objective Loss 0.235101 LR 0.000500 Time 0.020401 -2023-02-13 18:18:54,200 - Epoch: [137][ 1050/ 1207] Overall Loss 0.235169 Objective Loss 0.235169 LR 0.000500 Time 0.020394 -2023-02-13 18:18:54,398 - Epoch: [137][ 1060/ 1207] Overall Loss 0.235156 Objective Loss 0.235156 LR 0.000500 Time 0.020388 -2023-02-13 18:18:54,594 - Epoch: [137][ 1070/ 1207] Overall Loss 0.235132 Objective Loss 0.235132 LR 0.000500 Time 0.020380 -2023-02-13 18:18:54,793 - Epoch: [137][ 1080/ 1207] Overall Loss 0.235050 Objective Loss 0.235050 LR 0.000500 Time 0.020375 -2023-02-13 18:18:54,990 - Epoch: [137][ 1090/ 1207] Overall Loss 0.235142 Objective Loss 0.235142 LR 0.000500 Time 0.020369 -2023-02-13 18:18:55,189 - Epoch: [137][ 1100/ 1207] Overall Loss 0.235073 Objective Loss 0.235073 LR 0.000500 Time 0.020364 -2023-02-13 18:18:55,384 - Epoch: [137][ 1110/ 1207] Overall Loss 0.235236 Objective Loss 0.235236 LR 0.000500 Time 0.020356 -2023-02-13 18:18:55,582 - Epoch: [137][ 1120/ 1207] Overall Loss 0.235161 Objective Loss 0.235161 LR 0.000500 Time 0.020350 -2023-02-13 18:18:55,779 - Epoch: [137][ 1130/ 1207] Overall Loss 0.235020 Objective Loss 0.235020 LR 0.000500 Time 0.020344 -2023-02-13 18:18:55,978 - Epoch: [137][ 1140/ 1207] Overall Loss 0.235231 Objective Loss 0.235231 LR 0.000500 Time 0.020340 -2023-02-13 18:18:56,175 - Epoch: [137][ 1150/ 1207] Overall Loss 0.235444 Objective Loss 0.235444 LR 0.000500 Time 0.020334 -2023-02-13 18:18:56,373 - Epoch: [137][ 1160/ 1207] Overall Loss 0.235456 Objective Loss 0.235456 LR 0.000500 Time 0.020330 -2023-02-13 18:18:56,570 - Epoch: [137][ 1170/ 1207] Overall Loss 0.235543 Objective Loss 0.235543 LR 0.000500 Time 0.020324 -2023-02-13 18:18:56,769 - Epoch: [137][ 1180/ 1207] Overall Loss 0.235783 Objective Loss 0.235783 LR 0.000500 Time 0.020320 -2023-02-13 18:18:56,967 - Epoch: [137][ 1190/ 1207] Overall Loss 0.235657 Objective Loss 0.235657 LR 0.000500 Time 0.020315 -2023-02-13 18:18:57,215 - Epoch: [137][ 1200/ 1207] Overall Loss 0.235829 Objective Loss 0.235829 LR 0.000500 Time 0.020352 -2023-02-13 18:18:57,331 - Epoch: [137][ 1207/ 1207] Overall Loss 0.235710 Objective Loss 0.235710 Top1 87.195122 Top5 99.085366 LR 0.000500 Time 0.020330 -2023-02-13 18:18:57,403 - --- validate (epoch=137)----------- -2023-02-13 18:18:57,403 - 34311 samples (256 per mini-batch) -2023-02-13 18:18:57,818 - Epoch: [137][ 10/ 135] Loss 0.315621 Top1 84.140625 Top5 97.500000 -2023-02-13 18:18:57,949 - Epoch: [137][ 20/ 135] Loss 0.308423 Top1 84.511719 Top5 97.539062 -2023-02-13 18:18:58,078 - Epoch: [137][ 30/ 135] Loss 0.302910 Top1 84.570312 Top5 97.708333 -2023-02-13 18:18:58,203 - Epoch: [137][ 40/ 135] Loss 0.302938 Top1 84.628906 Top5 97.792969 -2023-02-13 18:18:58,329 - Epoch: [137][ 50/ 135] Loss 0.304419 Top1 84.609375 Top5 97.828125 -2023-02-13 18:18:58,454 - Epoch: [137][ 60/ 135] Loss 0.307791 Top1 84.654948 Top5 97.812500 -2023-02-13 18:18:58,578 - Epoch: [137][ 70/ 135] Loss 0.305837 Top1 84.687500 Top5 97.834821 -2023-02-13 18:18:58,702 - Epoch: [137][ 80/ 135] Loss 0.305635 Top1 84.609375 Top5 97.836914 -2023-02-13 18:18:58,827 - Epoch: [137][ 90/ 135] Loss 0.306534 Top1 84.700521 Top5 97.842882 -2023-02-13 18:18:58,950 - Epoch: [137][ 100/ 135] Loss 0.309284 Top1 84.566406 Top5 97.785156 -2023-02-13 18:18:59,073 - Epoch: [137][ 110/ 135] Loss 0.313980 Top1 84.460227 Top5 97.727273 -2023-02-13 18:18:59,201 - Epoch: [137][ 120/ 135] Loss 0.312475 Top1 84.508464 Top5 97.753906 -2023-02-13 18:18:59,330 - Epoch: [137][ 130/ 135] Loss 0.313941 Top1 84.498197 Top5 97.728365 -2023-02-13 18:18:59,374 - Epoch: [137][ 135/ 135] Loss 0.312774 Top1 84.526828 Top5 97.723762 -2023-02-13 18:18:59,446 - ==> Top1: 84.527 Top5: 97.724 Loss: 0.313 - -2023-02-13 18:18:59,447 - ==> Confusion: -[[ 860 5 6 1 12 3 0 2 3 39 2 5 2 5 7 1 2 2 1 3 6] - [ 2 960 0 2 12 10 3 15 4 2 1 0 1 0 2 3 4 0 4 4 4] - [ 8 6 954 10 1 1 18 13 0 1 5 1 1 5 5 3 3 2 5 10 6] - [ 2 0 22 895 5 5 3 2 3 2 13 1 7 1 24 0 4 7 15 1 4] - [ 14 6 0 0 991 6 1 3 1 0 0 6 1 5 10 6 9 0 0 3 4] - [ 1 31 3 5 9 929 6 19 1 6 3 12 3 18 2 2 5 1 1 9 4] - [ 3 3 13 1 0 2 1040 6 0 1 5 2 1 2 0 1 1 3 3 7 5] - [ 2 12 7 3 2 29 5 893 0 2 2 6 3 2 0 1 2 0 30 14 9] - [ 18 3 0 2 0 0 0 1 894 47 5 4 0 10 12 2 2 1 6 0 2] - [ 81 0 5 1 10 4 0 1 32 852 0 1 0 13 2 1 1 3 0 2 3] - [ 1 1 2 8 3 1 4 4 16 1 981 1 3 10 3 0 0 1 6 1 4] - [ 0 3 0 0 2 3 1 7 1 1 0 917 31 3 1 5 6 7 2 13 2] - [ 0 0 2 9 2 2 0 2 2 2 1 24 870 0 2 12 2 17 1 1 8] - [ 2 3 1 0 9 10 0 0 12 16 12 8 3 929 3 7 3 2 1 0 3] - [ 8 3 4 17 4 2 0 2 20 4 2 0 2 2 1003 2 0 5 4 1 7] - [ 2 2 3 0 10 0 1 1 0 0 0 2 9 2 0 985 7 6 2 9 5] - [ 3 8 1 1 15 2 0 0 1 1 0 1 6 2 1 10 993 1 1 5 9] - [ 4 4 2 6 0 0 2 1 1 0 1 9 20 0 1 21 2 968 0 4 5] - [ 4 3 3 9 0 1 0 19 1 0 3 1 6 0 14 2 1 2 1012 2 3] - [ 0 5 0 2 1 3 9 8 0 0 0 13 2 4 0 6 4 2 1 1081 7] - [ 143 253 217 131 141 165 96 111 115 103 204 106 294 293 159 114 232 113 179 270 9995]] - -2023-02-13 18:18:59,449 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:18:59,449 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:18:59,455 - - -2023-02-13 18:18:59,455 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:19:00,348 - Epoch: [138][ 10/ 1207] Overall Loss 0.220509 Objective Loss 0.220509 LR 0.000500 Time 0.089261 -2023-02-13 18:19:00,551 - Epoch: [138][ 20/ 1207] Overall Loss 0.224693 Objective Loss 0.224693 LR 0.000500 Time 0.054739 -2023-02-13 18:19:00,741 - Epoch: [138][ 30/ 1207] Overall Loss 0.220617 Objective Loss 0.220617 LR 0.000500 Time 0.042829 -2023-02-13 18:19:00,934 - Epoch: [138][ 40/ 1207] Overall Loss 0.228279 Objective Loss 0.228279 LR 0.000500 Time 0.036940 -2023-02-13 18:19:01,126 - Epoch: [138][ 50/ 1207] Overall Loss 0.226983 Objective Loss 0.226983 LR 0.000500 Time 0.033371 -2023-02-13 18:19:01,316 - Epoch: [138][ 60/ 1207] Overall Loss 0.226372 Objective Loss 0.226372 LR 0.000500 Time 0.030978 -2023-02-13 18:19:01,507 - Epoch: [138][ 70/ 1207] Overall Loss 0.227624 Objective Loss 0.227624 LR 0.000500 Time 0.029270 -2023-02-13 18:19:01,697 - Epoch: [138][ 80/ 1207] Overall Loss 0.231543 Objective Loss 0.231543 LR 0.000500 Time 0.027988 -2023-02-13 18:19:01,888 - Epoch: [138][ 90/ 1207] Overall Loss 0.232531 Objective Loss 0.232531 LR 0.000500 Time 0.026993 -2023-02-13 18:19:02,078 - Epoch: [138][ 100/ 1207] Overall Loss 0.231284 Objective Loss 0.231284 LR 0.000500 Time 0.026194 -2023-02-13 18:19:02,268 - Epoch: [138][ 110/ 1207] Overall Loss 0.234811 Objective Loss 0.234811 LR 0.000500 Time 0.025533 -2023-02-13 18:19:02,458 - Epoch: [138][ 120/ 1207] Overall Loss 0.237811 Objective Loss 0.237811 LR 0.000500 Time 0.024987 -2023-02-13 18:19:02,648 - Epoch: [138][ 130/ 1207] Overall Loss 0.236949 Objective Loss 0.236949 LR 0.000500 Time 0.024522 -2023-02-13 18:19:02,840 - Epoch: [138][ 140/ 1207] Overall Loss 0.236815 Objective Loss 0.236815 LR 0.000500 Time 0.024140 -2023-02-13 18:19:03,031 - Epoch: [138][ 150/ 1207] Overall Loss 0.236316 Objective Loss 0.236316 LR 0.000500 Time 0.023799 -2023-02-13 18:19:03,222 - Epoch: [138][ 160/ 1207] Overall Loss 0.238104 Objective Loss 0.238104 LR 0.000500 Time 0.023508 -2023-02-13 18:19:03,413 - Epoch: [138][ 170/ 1207] Overall Loss 0.237906 Objective Loss 0.237906 LR 0.000500 Time 0.023245 -2023-02-13 18:19:03,605 - Epoch: [138][ 180/ 1207] Overall Loss 0.236754 Objective Loss 0.236754 LR 0.000500 Time 0.023015 -2023-02-13 18:19:03,795 - Epoch: [138][ 190/ 1207] Overall Loss 0.235354 Objective Loss 0.235354 LR 0.000500 Time 0.022804 -2023-02-13 18:19:03,987 - Epoch: [138][ 200/ 1207] Overall Loss 0.235050 Objective Loss 0.235050 LR 0.000500 Time 0.022623 -2023-02-13 18:19:04,179 - Epoch: [138][ 210/ 1207] Overall Loss 0.236163 Objective Loss 0.236163 LR 0.000500 Time 0.022457 -2023-02-13 18:19:04,371 - Epoch: [138][ 220/ 1207] Overall Loss 0.235630 Objective Loss 0.235630 LR 0.000500 Time 0.022305 -2023-02-13 18:19:04,563 - Epoch: [138][ 230/ 1207] Overall Loss 0.235629 Objective Loss 0.235629 LR 0.000500 Time 0.022170 -2023-02-13 18:19:04,755 - Epoch: [138][ 240/ 1207] Overall Loss 0.235218 Objective Loss 0.235218 LR 0.000500 Time 0.022044 -2023-02-13 18:19:04,946 - Epoch: [138][ 250/ 1207] Overall Loss 0.235960 Objective Loss 0.235960 LR 0.000500 Time 0.021927 -2023-02-13 18:19:05,138 - Epoch: [138][ 260/ 1207] Overall Loss 0.236431 Objective Loss 0.236431 LR 0.000500 Time 0.021820 -2023-02-13 18:19:05,329 - Epoch: [138][ 270/ 1207] Overall Loss 0.236870 Objective Loss 0.236870 LR 0.000500 Time 0.021716 -2023-02-13 18:19:05,519 - Epoch: [138][ 280/ 1207] Overall Loss 0.236548 Objective Loss 0.236548 LR 0.000500 Time 0.021620 -2023-02-13 18:19:05,710 - Epoch: [138][ 290/ 1207] Overall Loss 0.235503 Objective Loss 0.235503 LR 0.000500 Time 0.021530 -2023-02-13 18:19:05,901 - Epoch: [138][ 300/ 1207] Overall Loss 0.235911 Objective Loss 0.235911 LR 0.000500 Time 0.021449 -2023-02-13 18:19:06,092 - Epoch: [138][ 310/ 1207] Overall Loss 0.235830 Objective Loss 0.235830 LR 0.000500 Time 0.021371 -2023-02-13 18:19:06,282 - Epoch: [138][ 320/ 1207] Overall Loss 0.236874 Objective Loss 0.236874 LR 0.000500 Time 0.021297 -2023-02-13 18:19:06,474 - Epoch: [138][ 330/ 1207] Overall Loss 0.237098 Objective Loss 0.237098 LR 0.000500 Time 0.021232 -2023-02-13 18:19:06,666 - Epoch: [138][ 340/ 1207] Overall Loss 0.237099 Objective Loss 0.237099 LR 0.000500 Time 0.021171 -2023-02-13 18:19:06,859 - Epoch: [138][ 350/ 1207] Overall Loss 0.236983 Objective Loss 0.236983 LR 0.000500 Time 0.021115 -2023-02-13 18:19:07,052 - Epoch: [138][ 360/ 1207] Overall Loss 0.236220 Objective Loss 0.236220 LR 0.000500 Time 0.021064 -2023-02-13 18:19:07,244 - Epoch: [138][ 370/ 1207] Overall Loss 0.236432 Objective Loss 0.236432 LR 0.000500 Time 0.021013 -2023-02-13 18:19:07,436 - Epoch: [138][ 380/ 1207] Overall Loss 0.236987 Objective Loss 0.236987 LR 0.000500 Time 0.020964 -2023-02-13 18:19:07,629 - Epoch: [138][ 390/ 1207] Overall Loss 0.236077 Objective Loss 0.236077 LR 0.000500 Time 0.020920 -2023-02-13 18:19:07,821 - Epoch: [138][ 400/ 1207] Overall Loss 0.235731 Objective Loss 0.235731 LR 0.000500 Time 0.020876 -2023-02-13 18:19:08,014 - Epoch: [138][ 410/ 1207] Overall Loss 0.235411 Objective Loss 0.235411 LR 0.000500 Time 0.020837 -2023-02-13 18:19:08,206 - Epoch: [138][ 420/ 1207] Overall Loss 0.234858 Objective Loss 0.234858 LR 0.000500 Time 0.020797 -2023-02-13 18:19:08,398 - Epoch: [138][ 430/ 1207] Overall Loss 0.234543 Objective Loss 0.234543 LR 0.000500 Time 0.020760 -2023-02-13 18:19:08,590 - Epoch: [138][ 440/ 1207] Overall Loss 0.234063 Objective Loss 0.234063 LR 0.000500 Time 0.020724 -2023-02-13 18:19:08,782 - Epoch: [138][ 450/ 1207] Overall Loss 0.234070 Objective Loss 0.234070 LR 0.000500 Time 0.020689 -2023-02-13 18:19:08,975 - Epoch: [138][ 460/ 1207] Overall Loss 0.234346 Objective Loss 0.234346 LR 0.000500 Time 0.020657 -2023-02-13 18:19:09,167 - Epoch: [138][ 470/ 1207] Overall Loss 0.234324 Objective Loss 0.234324 LR 0.000500 Time 0.020627 -2023-02-13 18:19:09,359 - Epoch: [138][ 480/ 1207] Overall Loss 0.234010 Objective Loss 0.234010 LR 0.000500 Time 0.020595 -2023-02-13 18:19:09,552 - Epoch: [138][ 490/ 1207] Overall Loss 0.234347 Objective Loss 0.234347 LR 0.000500 Time 0.020567 -2023-02-13 18:19:09,744 - Epoch: [138][ 500/ 1207] Overall Loss 0.234454 Objective Loss 0.234454 LR 0.000500 Time 0.020540 -2023-02-13 18:19:09,937 - Epoch: [138][ 510/ 1207] Overall Loss 0.234363 Objective Loss 0.234363 LR 0.000500 Time 0.020515 -2023-02-13 18:19:10,129 - Epoch: [138][ 520/ 1207] Overall Loss 0.234159 Objective Loss 0.234159 LR 0.000500 Time 0.020490 -2023-02-13 18:19:10,322 - Epoch: [138][ 530/ 1207] Overall Loss 0.233938 Objective Loss 0.233938 LR 0.000500 Time 0.020465 -2023-02-13 18:19:10,514 - Epoch: [138][ 540/ 1207] Overall Loss 0.233778 Objective Loss 0.233778 LR 0.000500 Time 0.020441 -2023-02-13 18:19:10,706 - Epoch: [138][ 550/ 1207] Overall Loss 0.233892 Objective Loss 0.233892 LR 0.000500 Time 0.020417 -2023-02-13 18:19:10,899 - Epoch: [138][ 560/ 1207] Overall Loss 0.233647 Objective Loss 0.233647 LR 0.000500 Time 0.020398 -2023-02-13 18:19:11,091 - Epoch: [138][ 570/ 1207] Overall Loss 0.233942 Objective Loss 0.233942 LR 0.000500 Time 0.020376 -2023-02-13 18:19:11,282 - Epoch: [138][ 580/ 1207] Overall Loss 0.234508 Objective Loss 0.234508 LR 0.000500 Time 0.020353 -2023-02-13 18:19:11,474 - Epoch: [138][ 590/ 1207] Overall Loss 0.234250 Objective Loss 0.234250 LR 0.000500 Time 0.020333 -2023-02-13 18:19:11,666 - Epoch: [138][ 600/ 1207] Overall Loss 0.234244 Objective Loss 0.234244 LR 0.000500 Time 0.020313 -2023-02-13 18:19:11,859 - Epoch: [138][ 610/ 1207] Overall Loss 0.234421 Objective Loss 0.234421 LR 0.000500 Time 0.020296 -2023-02-13 18:19:12,052 - Epoch: [138][ 620/ 1207] Overall Loss 0.234479 Objective Loss 0.234479 LR 0.000500 Time 0.020279 -2023-02-13 18:19:12,244 - Epoch: [138][ 630/ 1207] Overall Loss 0.234620 Objective Loss 0.234620 LR 0.000500 Time 0.020261 -2023-02-13 18:19:12,435 - Epoch: [138][ 640/ 1207] Overall Loss 0.234781 Objective Loss 0.234781 LR 0.000500 Time 0.020244 -2023-02-13 18:19:12,627 - Epoch: [138][ 650/ 1207] Overall Loss 0.235063 Objective Loss 0.235063 LR 0.000500 Time 0.020227 -2023-02-13 18:19:12,820 - Epoch: [138][ 660/ 1207] Overall Loss 0.235093 Objective Loss 0.235093 LR 0.000500 Time 0.020211 -2023-02-13 18:19:13,013 - Epoch: [138][ 670/ 1207] Overall Loss 0.235379 Objective Loss 0.235379 LR 0.000500 Time 0.020197 -2023-02-13 18:19:13,204 - Epoch: [138][ 680/ 1207] Overall Loss 0.235142 Objective Loss 0.235142 LR 0.000500 Time 0.020181 -2023-02-13 18:19:13,397 - Epoch: [138][ 690/ 1207] Overall Loss 0.235546 Objective Loss 0.235546 LR 0.000500 Time 0.020168 -2023-02-13 18:19:13,589 - Epoch: [138][ 700/ 1207] Overall Loss 0.235772 Objective Loss 0.235772 LR 0.000500 Time 0.020154 -2023-02-13 18:19:13,782 - Epoch: [138][ 710/ 1207] Overall Loss 0.235486 Objective Loss 0.235486 LR 0.000500 Time 0.020140 -2023-02-13 18:19:13,974 - Epoch: [138][ 720/ 1207] Overall Loss 0.235484 Objective Loss 0.235484 LR 0.000500 Time 0.020127 -2023-02-13 18:19:14,167 - Epoch: [138][ 730/ 1207] Overall Loss 0.235369 Objective Loss 0.235369 LR 0.000500 Time 0.020115 -2023-02-13 18:19:14,358 - Epoch: [138][ 740/ 1207] Overall Loss 0.235112 Objective Loss 0.235112 LR 0.000500 Time 0.020101 -2023-02-13 18:19:14,551 - Epoch: [138][ 750/ 1207] Overall Loss 0.235431 Objective Loss 0.235431 LR 0.000500 Time 0.020090 -2023-02-13 18:19:14,743 - Epoch: [138][ 760/ 1207] Overall Loss 0.235298 Objective Loss 0.235298 LR 0.000500 Time 0.020078 -2023-02-13 18:19:14,937 - Epoch: [138][ 770/ 1207] Overall Loss 0.235371 Objective Loss 0.235371 LR 0.000500 Time 0.020068 -2023-02-13 18:19:15,129 - Epoch: [138][ 780/ 1207] Overall Loss 0.235235 Objective Loss 0.235235 LR 0.000500 Time 0.020056 -2023-02-13 18:19:15,321 - Epoch: [138][ 790/ 1207] Overall Loss 0.235345 Objective Loss 0.235345 LR 0.000500 Time 0.020045 -2023-02-13 18:19:15,512 - Epoch: [138][ 800/ 1207] Overall Loss 0.235253 Objective Loss 0.235253 LR 0.000500 Time 0.020033 -2023-02-13 18:19:15,705 - Epoch: [138][ 810/ 1207] Overall Loss 0.235242 Objective Loss 0.235242 LR 0.000500 Time 0.020023 -2023-02-13 18:19:15,897 - Epoch: [138][ 820/ 1207] Overall Loss 0.235083 Objective Loss 0.235083 LR 0.000500 Time 0.020013 -2023-02-13 18:19:16,090 - Epoch: [138][ 830/ 1207] Overall Loss 0.235272 Objective Loss 0.235272 LR 0.000500 Time 0.020004 -2023-02-13 18:19:16,282 - Epoch: [138][ 840/ 1207] Overall Loss 0.235257 Objective Loss 0.235257 LR 0.000500 Time 0.019994 -2023-02-13 18:19:16,475 - Epoch: [138][ 850/ 1207] Overall Loss 0.235370 Objective Loss 0.235370 LR 0.000500 Time 0.019985 -2023-02-13 18:19:16,668 - Epoch: [138][ 860/ 1207] Overall Loss 0.235768 Objective Loss 0.235768 LR 0.000500 Time 0.019977 -2023-02-13 18:19:16,861 - Epoch: [138][ 870/ 1207] Overall Loss 0.235875 Objective Loss 0.235875 LR 0.000500 Time 0.019969 -2023-02-13 18:19:17,053 - Epoch: [138][ 880/ 1207] Overall Loss 0.236194 Objective Loss 0.236194 LR 0.000500 Time 0.019960 -2023-02-13 18:19:17,245 - Epoch: [138][ 890/ 1207] Overall Loss 0.236246 Objective Loss 0.236246 LR 0.000500 Time 0.019951 -2023-02-13 18:19:17,437 - Epoch: [138][ 900/ 1207] Overall Loss 0.236279 Objective Loss 0.236279 LR 0.000500 Time 0.019942 -2023-02-13 18:19:17,629 - Epoch: [138][ 910/ 1207] Overall Loss 0.235979 Objective Loss 0.235979 LR 0.000500 Time 0.019933 -2023-02-13 18:19:17,822 - Epoch: [138][ 920/ 1207] Overall Loss 0.235787 Objective Loss 0.235787 LR 0.000500 Time 0.019926 -2023-02-13 18:19:18,015 - Epoch: [138][ 930/ 1207] Overall Loss 0.235916 Objective Loss 0.235916 LR 0.000500 Time 0.019919 -2023-02-13 18:19:18,209 - Epoch: [138][ 940/ 1207] Overall Loss 0.235756 Objective Loss 0.235756 LR 0.000500 Time 0.019913 -2023-02-13 18:19:18,402 - Epoch: [138][ 950/ 1207] Overall Loss 0.235618 Objective Loss 0.235618 LR 0.000500 Time 0.019906 -2023-02-13 18:19:18,595 - Epoch: [138][ 960/ 1207] Overall Loss 0.235348 Objective Loss 0.235348 LR 0.000500 Time 0.019899 -2023-02-13 18:19:18,787 - Epoch: [138][ 970/ 1207] Overall Loss 0.235336 Objective Loss 0.235336 LR 0.000500 Time 0.019892 -2023-02-13 18:19:18,981 - Epoch: [138][ 980/ 1207] Overall Loss 0.235214 Objective Loss 0.235214 LR 0.000500 Time 0.019887 -2023-02-13 18:19:19,175 - Epoch: [138][ 990/ 1207] Overall Loss 0.235133 Objective Loss 0.235133 LR 0.000500 Time 0.019881 -2023-02-13 18:19:19,368 - Epoch: [138][ 1000/ 1207] Overall Loss 0.235153 Objective Loss 0.235153 LR 0.000500 Time 0.019875 -2023-02-13 18:19:19,561 - Epoch: [138][ 1010/ 1207] Overall Loss 0.235238 Objective Loss 0.235238 LR 0.000500 Time 0.019869 -2023-02-13 18:19:19,755 - Epoch: [138][ 1020/ 1207] Overall Loss 0.235242 Objective Loss 0.235242 LR 0.000500 Time 0.019864 -2023-02-13 18:19:19,948 - Epoch: [138][ 1030/ 1207] Overall Loss 0.235551 Objective Loss 0.235551 LR 0.000500 Time 0.019858 -2023-02-13 18:19:20,142 - Epoch: [138][ 1040/ 1207] Overall Loss 0.235467 Objective Loss 0.235467 LR 0.000500 Time 0.019853 -2023-02-13 18:19:20,335 - Epoch: [138][ 1050/ 1207] Overall Loss 0.235574 Objective Loss 0.235574 LR 0.000500 Time 0.019848 -2023-02-13 18:19:20,528 - Epoch: [138][ 1060/ 1207] Overall Loss 0.235640 Objective Loss 0.235640 LR 0.000500 Time 0.019843 -2023-02-13 18:19:20,721 - Epoch: [138][ 1070/ 1207] Overall Loss 0.235456 Objective Loss 0.235456 LR 0.000500 Time 0.019837 -2023-02-13 18:19:20,916 - Epoch: [138][ 1080/ 1207] Overall Loss 0.235481 Objective Loss 0.235481 LR 0.000500 Time 0.019833 -2023-02-13 18:19:21,110 - Epoch: [138][ 1090/ 1207] Overall Loss 0.235518 Objective Loss 0.235518 LR 0.000500 Time 0.019829 -2023-02-13 18:19:21,304 - Epoch: [138][ 1100/ 1207] Overall Loss 0.235316 Objective Loss 0.235316 LR 0.000500 Time 0.019824 -2023-02-13 18:19:21,496 - Epoch: [138][ 1110/ 1207] Overall Loss 0.235368 Objective Loss 0.235368 LR 0.000500 Time 0.019819 -2023-02-13 18:19:21,689 - Epoch: [138][ 1120/ 1207] Overall Loss 0.235092 Objective Loss 0.235092 LR 0.000500 Time 0.019814 -2023-02-13 18:19:21,882 - Epoch: [138][ 1130/ 1207] Overall Loss 0.235178 Objective Loss 0.235178 LR 0.000500 Time 0.019809 -2023-02-13 18:19:22,077 - Epoch: [138][ 1140/ 1207] Overall Loss 0.235115 Objective Loss 0.235115 LR 0.000500 Time 0.019805 -2023-02-13 18:19:22,270 - Epoch: [138][ 1150/ 1207] Overall Loss 0.235250 Objective Loss 0.235250 LR 0.000500 Time 0.019801 -2023-02-13 18:19:22,463 - Epoch: [138][ 1160/ 1207] Overall Loss 0.235440 Objective Loss 0.235440 LR 0.000500 Time 0.019796 -2023-02-13 18:19:22,656 - Epoch: [138][ 1170/ 1207] Overall Loss 0.235553 Objective Loss 0.235553 LR 0.000500 Time 0.019792 -2023-02-13 18:19:22,849 - Epoch: [138][ 1180/ 1207] Overall Loss 0.235573 Objective Loss 0.235573 LR 0.000500 Time 0.019788 -2023-02-13 18:19:23,044 - Epoch: [138][ 1190/ 1207] Overall Loss 0.235582 Objective Loss 0.235582 LR 0.000500 Time 0.019785 -2023-02-13 18:19:23,288 - Epoch: [138][ 1200/ 1207] Overall Loss 0.235444 Objective Loss 0.235444 LR 0.000500 Time 0.019823 -2023-02-13 18:19:23,404 - Epoch: [138][ 1207/ 1207] Overall Loss 0.235452 Objective Loss 0.235452 Top1 86.280488 Top5 98.780488 LR 0.000500 Time 0.019804 -2023-02-13 18:19:23,475 - --- validate (epoch=138)----------- -2023-02-13 18:19:23,476 - 34311 samples (256 per mini-batch) -2023-02-13 18:19:23,882 - Epoch: [138][ 10/ 135] Loss 0.314851 Top1 83.671875 Top5 97.226562 -2023-02-13 18:19:24,010 - Epoch: [138][ 20/ 135] Loss 0.321230 Top1 83.632812 Top5 97.480469 -2023-02-13 18:19:24,139 - Epoch: [138][ 30/ 135] Loss 0.321608 Top1 83.971354 Top5 97.695312 -2023-02-13 18:19:24,263 - Epoch: [138][ 40/ 135] Loss 0.322848 Top1 83.769531 Top5 97.666016 -2023-02-13 18:19:24,407 - Epoch: [138][ 50/ 135] Loss 0.321749 Top1 83.890625 Top5 97.703125 -2023-02-13 18:19:24,544 - Epoch: [138][ 60/ 135] Loss 0.322987 Top1 83.776042 Top5 97.584635 -2023-02-13 18:19:24,684 - Epoch: [138][ 70/ 135] Loss 0.321278 Top1 83.816964 Top5 97.622768 -2023-02-13 18:19:24,812 - Epoch: [138][ 80/ 135] Loss 0.321312 Top1 83.891602 Top5 97.636719 -2023-02-13 18:19:24,943 - Epoch: [138][ 90/ 135] Loss 0.319753 Top1 83.823785 Top5 97.586806 -2023-02-13 18:19:25,069 - Epoch: [138][ 100/ 135] Loss 0.319527 Top1 83.882812 Top5 97.585938 -2023-02-13 18:19:25,197 - Epoch: [138][ 110/ 135] Loss 0.317783 Top1 83.998580 Top5 97.606534 -2023-02-13 18:19:25,328 - Epoch: [138][ 120/ 135] Loss 0.315209 Top1 84.020182 Top5 97.620443 -2023-02-13 18:19:25,463 - Epoch: [138][ 130/ 135] Loss 0.312822 Top1 84.050481 Top5 97.626202 -2023-02-13 18:19:25,510 - Epoch: [138][ 135/ 135] Loss 0.311396 Top1 84.086736 Top5 97.633412 -2023-02-13 18:19:25,582 - ==> Top1: 84.087 Top5: 97.633 Loss: 0.311 - -2023-02-13 18:19:25,582 - ==> Confusion: -[[ 845 1 3 1 10 2 0 1 7 58 1 4 1 5 8 4 3 3 3 1 6] - [ 1 957 2 1 14 16 1 9 4 2 2 0 1 0 4 1 5 0 7 1 5] - [ 5 6 962 8 2 1 19 12 1 1 4 2 1 4 3 6 2 3 3 6 7] - [ 5 1 30 887 2 3 0 1 3 2 12 1 7 0 28 0 3 6 21 0 4] - [ 10 6 1 0 996 5 1 3 2 6 0 5 0 2 9 7 6 2 0 3 2] - [ 3 31 1 1 7 950 4 17 1 5 1 8 5 16 1 3 4 0 1 5 6] - [ 3 4 7 2 2 5 1041 3 1 1 4 0 0 3 0 4 1 2 2 8 6] - [ 2 11 10 4 5 24 2 915 2 1 0 7 3 2 0 0 1 1 17 8 9] - [ 11 1 0 2 0 1 0 1 927 31 5 3 1 9 13 1 1 0 2 0 0] - [ 54 3 4 1 7 1 0 1 42 872 0 2 0 15 3 1 0 2 1 0 3] - [ 0 0 6 4 2 5 2 2 21 0 980 1 1 7 4 0 1 2 11 0 2] - [ 0 2 0 0 2 13 0 6 4 0 1 917 19 8 2 2 2 4 3 19 1] - [ 0 0 2 9 1 2 0 3 4 0 0 45 844 1 2 8 3 18 2 3 12] - [ 3 4 1 0 10 7 0 2 20 16 8 5 4 929 3 5 3 1 0 1 2] - [ 2 1 2 13 3 3 0 1 23 6 1 0 3 2 1007 3 3 6 7 1 5] - [ 1 2 7 1 7 1 3 1 2 0 0 11 7 2 0 959 14 14 1 10 3] - [ 1 8 1 2 8 0 0 0 3 2 0 2 5 3 3 6 1005 0 2 2 8] - [ 4 1 1 3 0 3 3 1 2 1 1 13 14 1 0 18 2 974 0 4 5] - [ 5 6 5 6 0 2 1 14 3 0 3 1 4 0 12 1 1 1 1016 3 2] - [ 0 4 2 0 1 6 3 7 0 0 2 17 2 3 0 3 9 1 1 1082 5] - [ 159 267 230 113 152 203 99 130 144 117 158 140 263 290 208 90 301 102 198 284 9786]] - -2023-02-13 18:19:25,584 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:19:25,584 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:19:25,590 - - -2023-02-13 18:19:25,590 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:19:26,474 - Epoch: [139][ 10/ 1207] Overall Loss 0.235437 Objective Loss 0.235437 LR 0.000500 Time 0.088365 -2023-02-13 18:19:26,674 - Epoch: [139][ 20/ 1207] Overall Loss 0.236525 Objective Loss 0.236525 LR 0.000500 Time 0.054148 -2023-02-13 18:19:26,871 - Epoch: [139][ 30/ 1207] Overall Loss 0.239500 Objective Loss 0.239500 LR 0.000500 Time 0.042647 -2023-02-13 18:19:27,065 - Epoch: [139][ 40/ 1207] Overall Loss 0.238984 Objective Loss 0.238984 LR 0.000500 Time 0.036840 -2023-02-13 18:19:27,262 - Epoch: [139][ 50/ 1207] Overall Loss 0.238571 Objective Loss 0.238571 LR 0.000500 Time 0.033409 -2023-02-13 18:19:27,456 - Epoch: [139][ 60/ 1207] Overall Loss 0.238460 Objective Loss 0.238460 LR 0.000500 Time 0.031058 -2023-02-13 18:19:27,652 - Epoch: [139][ 70/ 1207] Overall Loss 0.235553 Objective Loss 0.235553 LR 0.000500 Time 0.029418 -2023-02-13 18:19:27,845 - Epoch: [139][ 80/ 1207] Overall Loss 0.236798 Objective Loss 0.236798 LR 0.000500 Time 0.028150 -2023-02-13 18:19:28,042 - Epoch: [139][ 90/ 1207] Overall Loss 0.234625 Objective Loss 0.234625 LR 0.000500 Time 0.027203 -2023-02-13 18:19:28,236 - Epoch: [139][ 100/ 1207] Overall Loss 0.236309 Objective Loss 0.236309 LR 0.000500 Time 0.026418 -2023-02-13 18:19:28,432 - Epoch: [139][ 110/ 1207] Overall Loss 0.234749 Objective Loss 0.234749 LR 0.000500 Time 0.025796 -2023-02-13 18:19:28,625 - Epoch: [139][ 120/ 1207] Overall Loss 0.236522 Objective Loss 0.236522 LR 0.000500 Time 0.025254 -2023-02-13 18:19:28,820 - Epoch: [139][ 130/ 1207] Overall Loss 0.238535 Objective Loss 0.238535 LR 0.000500 Time 0.024807 -2023-02-13 18:19:29,012 - Epoch: [139][ 140/ 1207] Overall Loss 0.239718 Objective Loss 0.239718 LR 0.000500 Time 0.024405 -2023-02-13 18:19:29,207 - Epoch: [139][ 150/ 1207] Overall Loss 0.239005 Objective Loss 0.239005 LR 0.000500 Time 0.024073 -2023-02-13 18:19:29,398 - Epoch: [139][ 160/ 1207] Overall Loss 0.238026 Objective Loss 0.238026 LR 0.000500 Time 0.023761 -2023-02-13 18:19:29,593 - Epoch: [139][ 170/ 1207] Overall Loss 0.237948 Objective Loss 0.237948 LR 0.000500 Time 0.023507 -2023-02-13 18:19:29,784 - Epoch: [139][ 180/ 1207] Overall Loss 0.238176 Objective Loss 0.238176 LR 0.000500 Time 0.023261 -2023-02-13 18:19:29,979 - Epoch: [139][ 190/ 1207] Overall Loss 0.238406 Objective Loss 0.238406 LR 0.000500 Time 0.023063 -2023-02-13 18:19:30,171 - Epoch: [139][ 200/ 1207] Overall Loss 0.237399 Objective Loss 0.237399 LR 0.000500 Time 0.022867 -2023-02-13 18:19:30,365 - Epoch: [139][ 210/ 1207] Overall Loss 0.236728 Objective Loss 0.236728 LR 0.000500 Time 0.022703 -2023-02-13 18:19:30,557 - Epoch: [139][ 220/ 1207] Overall Loss 0.236700 Objective Loss 0.236700 LR 0.000500 Time 0.022541 -2023-02-13 18:19:30,752 - Epoch: [139][ 230/ 1207] Overall Loss 0.236999 Objective Loss 0.236999 LR 0.000500 Time 0.022404 -2023-02-13 18:19:30,944 - Epoch: [139][ 240/ 1207] Overall Loss 0.236366 Objective Loss 0.236366 LR 0.000500 Time 0.022272 -2023-02-13 18:19:31,133 - Epoch: [139][ 250/ 1207] Overall Loss 0.236275 Objective Loss 0.236275 LR 0.000500 Time 0.022135 -2023-02-13 18:19:31,322 - Epoch: [139][ 260/ 1207] Overall Loss 0.236249 Objective Loss 0.236249 LR 0.000500 Time 0.022007 -2023-02-13 18:19:31,510 - Epoch: [139][ 270/ 1207] Overall Loss 0.236611 Objective Loss 0.236611 LR 0.000500 Time 0.021890 -2023-02-13 18:19:31,698 - Epoch: [139][ 280/ 1207] Overall Loss 0.237748 Objective Loss 0.237748 LR 0.000500 Time 0.021778 -2023-02-13 18:19:31,887 - Epoch: [139][ 290/ 1207] Overall Loss 0.238126 Objective Loss 0.238126 LR 0.000500 Time 0.021676 -2023-02-13 18:19:32,076 - Epoch: [139][ 300/ 1207] Overall Loss 0.237814 Objective Loss 0.237814 LR 0.000500 Time 0.021582 -2023-02-13 18:19:32,265 - Epoch: [139][ 310/ 1207] Overall Loss 0.238060 Objective Loss 0.238060 LR 0.000500 Time 0.021494 -2023-02-13 18:19:32,453 - Epoch: [139][ 320/ 1207] Overall Loss 0.238686 Objective Loss 0.238686 LR 0.000500 Time 0.021409 -2023-02-13 18:19:32,642 - Epoch: [139][ 330/ 1207] Overall Loss 0.239230 Objective Loss 0.239230 LR 0.000500 Time 0.021332 -2023-02-13 18:19:32,830 - Epoch: [139][ 340/ 1207] Overall Loss 0.239371 Objective Loss 0.239371 LR 0.000500 Time 0.021256 -2023-02-13 18:19:33,019 - Epoch: [139][ 350/ 1207] Overall Loss 0.238440 Objective Loss 0.238440 LR 0.000500 Time 0.021188 -2023-02-13 18:19:33,207 - Epoch: [139][ 360/ 1207] Overall Loss 0.238603 Objective Loss 0.238603 LR 0.000500 Time 0.021122 -2023-02-13 18:19:33,396 - Epoch: [139][ 370/ 1207] Overall Loss 0.238453 Objective Loss 0.238453 LR 0.000500 Time 0.021059 -2023-02-13 18:19:33,584 - Epoch: [139][ 380/ 1207] Overall Loss 0.238153 Objective Loss 0.238153 LR 0.000500 Time 0.020999 -2023-02-13 18:19:33,772 - Epoch: [139][ 390/ 1207] Overall Loss 0.238221 Objective Loss 0.238221 LR 0.000500 Time 0.020943 -2023-02-13 18:19:33,961 - Epoch: [139][ 400/ 1207] Overall Loss 0.237834 Objective Loss 0.237834 LR 0.000500 Time 0.020890 -2023-02-13 18:19:34,150 - Epoch: [139][ 410/ 1207] Overall Loss 0.237549 Objective Loss 0.237549 LR 0.000500 Time 0.020841 -2023-02-13 18:19:34,338 - Epoch: [139][ 420/ 1207] Overall Loss 0.237156 Objective Loss 0.237156 LR 0.000500 Time 0.020792 -2023-02-13 18:19:34,527 - Epoch: [139][ 430/ 1207] Overall Loss 0.236496 Objective Loss 0.236496 LR 0.000500 Time 0.020747 -2023-02-13 18:19:34,716 - Epoch: [139][ 440/ 1207] Overall Loss 0.236339 Objective Loss 0.236339 LR 0.000500 Time 0.020704 -2023-02-13 18:19:34,905 - Epoch: [139][ 450/ 1207] Overall Loss 0.236446 Objective Loss 0.236446 LR 0.000500 Time 0.020663 -2023-02-13 18:19:35,094 - Epoch: [139][ 460/ 1207] Overall Loss 0.235773 Objective Loss 0.235773 LR 0.000500 Time 0.020624 -2023-02-13 18:19:35,283 - Epoch: [139][ 470/ 1207] Overall Loss 0.235613 Objective Loss 0.235613 LR 0.000500 Time 0.020587 -2023-02-13 18:19:35,472 - Epoch: [139][ 480/ 1207] Overall Loss 0.235341 Objective Loss 0.235341 LR 0.000500 Time 0.020551 -2023-02-13 18:19:35,661 - Epoch: [139][ 490/ 1207] Overall Loss 0.234772 Objective Loss 0.234772 LR 0.000500 Time 0.020516 -2023-02-13 18:19:35,849 - Epoch: [139][ 500/ 1207] Overall Loss 0.234688 Objective Loss 0.234688 LR 0.000500 Time 0.020482 -2023-02-13 18:19:36,041 - Epoch: [139][ 510/ 1207] Overall Loss 0.235032 Objective Loss 0.235032 LR 0.000500 Time 0.020454 -2023-02-13 18:19:36,229 - Epoch: [139][ 520/ 1207] Overall Loss 0.234934 Objective Loss 0.234934 LR 0.000500 Time 0.020423 -2023-02-13 18:19:36,418 - Epoch: [139][ 530/ 1207] Overall Loss 0.234647 Objective Loss 0.234647 LR 0.000500 Time 0.020393 -2023-02-13 18:19:36,606 - Epoch: [139][ 540/ 1207] Overall Loss 0.234718 Objective Loss 0.234718 LR 0.000500 Time 0.020363 -2023-02-13 18:19:36,795 - Epoch: [139][ 550/ 1207] Overall Loss 0.234652 Objective Loss 0.234652 LR 0.000500 Time 0.020336 -2023-02-13 18:19:36,985 - Epoch: [139][ 560/ 1207] Overall Loss 0.235224 Objective Loss 0.235224 LR 0.000500 Time 0.020311 -2023-02-13 18:19:37,173 - Epoch: [139][ 570/ 1207] Overall Loss 0.235198 Objective Loss 0.235198 LR 0.000500 Time 0.020284 -2023-02-13 18:19:37,362 - Epoch: [139][ 580/ 1207] Overall Loss 0.235057 Objective Loss 0.235057 LR 0.000500 Time 0.020259 -2023-02-13 18:19:37,550 - Epoch: [139][ 590/ 1207] Overall Loss 0.235059 Objective Loss 0.235059 LR 0.000500 Time 0.020235 -2023-02-13 18:19:37,740 - Epoch: [139][ 600/ 1207] Overall Loss 0.235276 Objective Loss 0.235276 LR 0.000500 Time 0.020213 -2023-02-13 18:19:37,931 - Epoch: [139][ 610/ 1207] Overall Loss 0.235417 Objective Loss 0.235417 LR 0.000500 Time 0.020194 -2023-02-13 18:19:38,122 - Epoch: [139][ 620/ 1207] Overall Loss 0.235623 Objective Loss 0.235623 LR 0.000500 Time 0.020175 -2023-02-13 18:19:38,312 - Epoch: [139][ 630/ 1207] Overall Loss 0.235861 Objective Loss 0.235861 LR 0.000500 Time 0.020157 -2023-02-13 18:19:38,503 - Epoch: [139][ 640/ 1207] Overall Loss 0.236089 Objective Loss 0.236089 LR 0.000500 Time 0.020139 -2023-02-13 18:19:38,693 - Epoch: [139][ 650/ 1207] Overall Loss 0.235909 Objective Loss 0.235909 LR 0.000500 Time 0.020121 -2023-02-13 18:19:38,884 - Epoch: [139][ 660/ 1207] Overall Loss 0.235958 Objective Loss 0.235958 LR 0.000500 Time 0.020106 -2023-02-13 18:19:39,075 - Epoch: [139][ 670/ 1207] Overall Loss 0.236437 Objective Loss 0.236437 LR 0.000500 Time 0.020090 -2023-02-13 18:19:39,266 - Epoch: [139][ 680/ 1207] Overall Loss 0.235907 Objective Loss 0.235907 LR 0.000500 Time 0.020075 -2023-02-13 18:19:39,455 - Epoch: [139][ 690/ 1207] Overall Loss 0.235992 Objective Loss 0.235992 LR 0.000500 Time 0.020058 -2023-02-13 18:19:39,645 - Epoch: [139][ 700/ 1207] Overall Loss 0.235563 Objective Loss 0.235563 LR 0.000500 Time 0.020042 -2023-02-13 18:19:39,835 - Epoch: [139][ 710/ 1207] Overall Loss 0.235596 Objective Loss 0.235596 LR 0.000500 Time 0.020027 -2023-02-13 18:19:40,026 - Epoch: [139][ 720/ 1207] Overall Loss 0.235298 Objective Loss 0.235298 LR 0.000500 Time 0.020012 -2023-02-13 18:19:40,218 - Epoch: [139][ 730/ 1207] Overall Loss 0.235403 Objective Loss 0.235403 LR 0.000500 Time 0.020001 -2023-02-13 18:19:40,408 - Epoch: [139][ 740/ 1207] Overall Loss 0.235573 Objective Loss 0.235573 LR 0.000500 Time 0.019987 -2023-02-13 18:19:40,598 - Epoch: [139][ 750/ 1207] Overall Loss 0.235814 Objective Loss 0.235814 LR 0.000500 Time 0.019973 -2023-02-13 18:19:40,787 - Epoch: [139][ 760/ 1207] Overall Loss 0.235698 Objective Loss 0.235698 LR 0.000500 Time 0.019960 -2023-02-13 18:19:40,979 - Epoch: [139][ 770/ 1207] Overall Loss 0.235725 Objective Loss 0.235725 LR 0.000500 Time 0.019948 -2023-02-13 18:19:41,169 - Epoch: [139][ 780/ 1207] Overall Loss 0.235688 Objective Loss 0.235688 LR 0.000500 Time 0.019936 -2023-02-13 18:19:41,359 - Epoch: [139][ 790/ 1207] Overall Loss 0.236084 Objective Loss 0.236084 LR 0.000500 Time 0.019924 -2023-02-13 18:19:41,549 - Epoch: [139][ 800/ 1207] Overall Loss 0.236231 Objective Loss 0.236231 LR 0.000500 Time 0.019912 -2023-02-13 18:19:41,740 - Epoch: [139][ 810/ 1207] Overall Loss 0.236253 Objective Loss 0.236253 LR 0.000500 Time 0.019901 -2023-02-13 18:19:41,930 - Epoch: [139][ 820/ 1207] Overall Loss 0.236422 Objective Loss 0.236422 LR 0.000500 Time 0.019889 -2023-02-13 18:19:42,121 - Epoch: [139][ 830/ 1207] Overall Loss 0.236257 Objective Loss 0.236257 LR 0.000500 Time 0.019879 -2023-02-13 18:19:42,311 - Epoch: [139][ 840/ 1207] Overall Loss 0.236095 Objective Loss 0.236095 LR 0.000500 Time 0.019869 -2023-02-13 18:19:42,501 - Epoch: [139][ 850/ 1207] Overall Loss 0.235888 Objective Loss 0.235888 LR 0.000500 Time 0.019858 -2023-02-13 18:19:42,691 - Epoch: [139][ 860/ 1207] Overall Loss 0.236139 Objective Loss 0.236139 LR 0.000500 Time 0.019847 -2023-02-13 18:19:42,881 - Epoch: [139][ 870/ 1207] Overall Loss 0.236106 Objective Loss 0.236106 LR 0.000500 Time 0.019837 -2023-02-13 18:19:43,072 - Epoch: [139][ 880/ 1207] Overall Loss 0.236474 Objective Loss 0.236474 LR 0.000500 Time 0.019829 -2023-02-13 18:19:43,263 - Epoch: [139][ 890/ 1207] Overall Loss 0.236156 Objective Loss 0.236156 LR 0.000500 Time 0.019820 -2023-02-13 18:19:43,452 - Epoch: [139][ 900/ 1207] Overall Loss 0.236094 Objective Loss 0.236094 LR 0.000500 Time 0.019810 -2023-02-13 18:19:43,643 - Epoch: [139][ 910/ 1207] Overall Loss 0.236236 Objective Loss 0.236236 LR 0.000500 Time 0.019801 -2023-02-13 18:19:43,832 - Epoch: [139][ 920/ 1207] Overall Loss 0.236024 Objective Loss 0.236024 LR 0.000500 Time 0.019791 -2023-02-13 18:19:44,023 - Epoch: [139][ 930/ 1207] Overall Loss 0.236299 Objective Loss 0.236299 LR 0.000500 Time 0.019783 -2023-02-13 18:19:44,214 - Epoch: [139][ 940/ 1207] Overall Loss 0.236094 Objective Loss 0.236094 LR 0.000500 Time 0.019775 -2023-02-13 18:19:44,404 - Epoch: [139][ 950/ 1207] Overall Loss 0.235808 Objective Loss 0.235808 LR 0.000500 Time 0.019766 -2023-02-13 18:19:44,594 - Epoch: [139][ 960/ 1207] Overall Loss 0.235426 Objective Loss 0.235426 LR 0.000500 Time 0.019758 -2023-02-13 18:19:44,784 - Epoch: [139][ 970/ 1207] Overall Loss 0.235538 Objective Loss 0.235538 LR 0.000500 Time 0.019750 -2023-02-13 18:19:44,974 - Epoch: [139][ 980/ 1207] Overall Loss 0.235346 Objective Loss 0.235346 LR 0.000500 Time 0.019743 -2023-02-13 18:19:45,165 - Epoch: [139][ 990/ 1207] Overall Loss 0.235226 Objective Loss 0.235226 LR 0.000500 Time 0.019735 -2023-02-13 18:19:45,355 - Epoch: [139][ 1000/ 1207] Overall Loss 0.235176 Objective Loss 0.235176 LR 0.000500 Time 0.019727 -2023-02-13 18:19:45,545 - Epoch: [139][ 1010/ 1207] Overall Loss 0.235330 Objective Loss 0.235330 LR 0.000500 Time 0.019720 -2023-02-13 18:19:45,735 - Epoch: [139][ 1020/ 1207] Overall Loss 0.235164 Objective Loss 0.235164 LR 0.000500 Time 0.019712 -2023-02-13 18:19:45,925 - Epoch: [139][ 1030/ 1207] Overall Loss 0.235002 Objective Loss 0.235002 LR 0.000500 Time 0.019706 -2023-02-13 18:19:46,116 - Epoch: [139][ 1040/ 1207] Overall Loss 0.234741 Objective Loss 0.234741 LR 0.000500 Time 0.019699 -2023-02-13 18:19:46,306 - Epoch: [139][ 1050/ 1207] Overall Loss 0.234733 Objective Loss 0.234733 LR 0.000500 Time 0.019692 -2023-02-13 18:19:46,496 - Epoch: [139][ 1060/ 1207] Overall Loss 0.234633 Objective Loss 0.234633 LR 0.000500 Time 0.019686 -2023-02-13 18:19:46,687 - Epoch: [139][ 1070/ 1207] Overall Loss 0.234826 Objective Loss 0.234826 LR 0.000500 Time 0.019679 -2023-02-13 18:19:46,877 - Epoch: [139][ 1080/ 1207] Overall Loss 0.234755 Objective Loss 0.234755 LR 0.000500 Time 0.019673 -2023-02-13 18:19:47,067 - Epoch: [139][ 1090/ 1207] Overall Loss 0.234952 Objective Loss 0.234952 LR 0.000500 Time 0.019667 -2023-02-13 18:19:47,258 - Epoch: [139][ 1100/ 1207] Overall Loss 0.234822 Objective Loss 0.234822 LR 0.000500 Time 0.019661 -2023-02-13 18:19:47,448 - Epoch: [139][ 1110/ 1207] Overall Loss 0.235052 Objective Loss 0.235052 LR 0.000500 Time 0.019655 -2023-02-13 18:19:47,638 - Epoch: [139][ 1120/ 1207] Overall Loss 0.234949 Objective Loss 0.234949 LR 0.000500 Time 0.019649 -2023-02-13 18:19:47,828 - Epoch: [139][ 1130/ 1207] Overall Loss 0.235017 Objective Loss 0.235017 LR 0.000500 Time 0.019643 -2023-02-13 18:19:48,019 - Epoch: [139][ 1140/ 1207] Overall Loss 0.234896 Objective Loss 0.234896 LR 0.000500 Time 0.019637 -2023-02-13 18:19:48,209 - Epoch: [139][ 1150/ 1207] Overall Loss 0.234671 Objective Loss 0.234671 LR 0.000500 Time 0.019631 -2023-02-13 18:19:48,399 - Epoch: [139][ 1160/ 1207] Overall Loss 0.234483 Objective Loss 0.234483 LR 0.000500 Time 0.019625 -2023-02-13 18:19:48,588 - Epoch: [139][ 1170/ 1207] Overall Loss 0.234540 Objective Loss 0.234540 LR 0.000500 Time 0.019619 -2023-02-13 18:19:48,778 - Epoch: [139][ 1180/ 1207] Overall Loss 0.234772 Objective Loss 0.234772 LR 0.000500 Time 0.019614 -2023-02-13 18:19:48,969 - Epoch: [139][ 1190/ 1207] Overall Loss 0.234990 Objective Loss 0.234990 LR 0.000500 Time 0.019609 -2023-02-13 18:19:49,211 - Epoch: [139][ 1200/ 1207] Overall Loss 0.235065 Objective Loss 0.235065 LR 0.000500 Time 0.019647 -2023-02-13 18:19:49,327 - Epoch: [139][ 1207/ 1207] Overall Loss 0.235169 Objective Loss 0.235169 Top1 83.536585 Top5 98.780488 LR 0.000500 Time 0.019629 -2023-02-13 18:19:49,399 - --- validate (epoch=139)----------- -2023-02-13 18:19:49,399 - 34311 samples (256 per mini-batch) -2023-02-13 18:19:49,914 - Epoch: [139][ 10/ 135] Loss 0.310442 Top1 85.937500 Top5 97.578125 -2023-02-13 18:19:50,056 - Epoch: [139][ 20/ 135] Loss 0.307563 Top1 84.531250 Top5 97.734375 -2023-02-13 18:19:50,203 - Epoch: [139][ 30/ 135] Loss 0.309538 Top1 84.895833 Top5 97.643229 -2023-02-13 18:19:50,342 - Epoch: [139][ 40/ 135] Loss 0.307362 Top1 84.755859 Top5 97.568359 -2023-02-13 18:19:50,482 - Epoch: [139][ 50/ 135] Loss 0.300094 Top1 84.765625 Top5 97.656250 -2023-02-13 18:19:50,607 - Epoch: [139][ 60/ 135] Loss 0.298979 Top1 84.628906 Top5 97.669271 -2023-02-13 18:19:50,733 - Epoch: [139][ 70/ 135] Loss 0.299730 Top1 84.497768 Top5 97.555804 -2023-02-13 18:19:50,863 - Epoch: [139][ 80/ 135] Loss 0.302417 Top1 84.599609 Top5 97.597656 -2023-02-13 18:19:51,007 - Epoch: [139][ 90/ 135] Loss 0.304341 Top1 84.448785 Top5 97.595486 -2023-02-13 18:19:51,139 - Epoch: [139][ 100/ 135] Loss 0.305599 Top1 84.234375 Top5 97.558594 -2023-02-13 18:19:51,272 - Epoch: [139][ 110/ 135] Loss 0.306678 Top1 84.186790 Top5 97.585227 -2023-02-13 18:19:51,404 - Epoch: [139][ 120/ 135] Loss 0.308462 Top1 84.212240 Top5 97.613932 -2023-02-13 18:19:51,538 - Epoch: [139][ 130/ 135] Loss 0.310984 Top1 84.182692 Top5 97.599159 -2023-02-13 18:19:51,585 - Epoch: [139][ 135/ 135] Loss 0.309453 Top1 84.168343 Top5 97.598438 -2023-02-13 18:19:51,658 - ==> Top1: 84.168 Top5: 97.598 Loss: 0.309 - -2023-02-13 18:19:51,658 - ==> Confusion: -[[ 864 6 4 2 9 1 0 1 1 48 0 3 2 3 5 5 2 1 2 2 6] - [ 4 965 0 4 9 16 0 12 4 2 1 2 1 0 0 1 4 0 3 2 3] - [ 8 5 959 13 2 1 13 15 1 1 3 2 1 6 4 8 2 3 1 5 5] - [ 4 0 25 901 1 7 4 3 5 4 7 0 3 0 20 3 5 6 12 0 6] - [ 14 8 3 0 989 7 3 1 2 3 0 4 3 5 5 7 7 1 0 1 3] - [ 4 23 1 6 7 962 6 19 3 6 0 7 3 12 2 1 2 0 0 4 2] - [ 2 5 16 2 0 7 1035 5 1 1 4 2 2 2 0 1 2 2 1 5 4] - [ 4 12 7 1 4 30 2 924 1 2 1 4 2 1 1 1 1 1 16 7 2] - [ 18 5 1 2 1 0 0 1 892 50 8 3 0 10 11 1 3 0 2 0 1] - [ 70 3 2 0 8 0 0 1 30 875 0 2 0 13 3 2 0 0 1 0 2] - [ 2 4 2 3 0 4 3 4 14 2 981 3 1 10 3 0 1 1 10 0 3] - [ 2 1 0 0 4 12 0 6 0 2 0 918 16 10 1 3 5 7 3 14 1] - [ 0 0 0 6 2 5 0 2 3 1 0 41 844 3 2 9 2 24 3 2 10] - [ 5 2 1 1 8 9 0 3 10 13 4 6 3 943 5 4 2 2 0 0 3] - [ 9 4 2 18 3 7 1 1 27 7 2 3 1 0 984 0 1 6 6 0 10] - [ 3 2 6 2 7 2 2 2 0 0 0 5 5 2 3 977 9 5 0 8 6] - [ 4 5 0 1 7 2 0 1 1 2 0 0 2 1 1 11 1004 2 1 9 7] - [ 8 5 1 4 1 3 2 0 0 1 3 12 9 1 2 21 1 967 0 4 6] - [ 4 5 4 6 1 4 0 31 3 0 5 0 3 0 13 0 1 2 999 2 3] - [ 0 2 1 0 0 8 1 9 1 0 1 12 1 4 0 9 3 3 0 1086 7] - [ 177 319 200 116 132 223 99 193 119 121 180 130 253 309 137 107 275 103 183 248 9810]] - -2023-02-13 18:19:51,660 - ==> Best [Top1: 85.159 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 128] -2023-02-13 18:19:51,660 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:19:51,665 - - -2023-02-13 18:19:51,666 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:19:52,555 - Epoch: [140][ 10/ 1207] Overall Loss 0.211858 Objective Loss 0.211858 LR 0.000250 Time 0.088841 -2023-02-13 18:19:52,767 - Epoch: [140][ 20/ 1207] Overall Loss 0.215566 Objective Loss 0.215566 LR 0.000250 Time 0.055015 -2023-02-13 18:19:52,967 - Epoch: [140][ 30/ 1207] Overall Loss 0.220233 Objective Loss 0.220233 LR 0.000250 Time 0.043334 -2023-02-13 18:19:53,173 - Epoch: [140][ 40/ 1207] Overall Loss 0.224134 Objective Loss 0.224134 LR 0.000250 Time 0.037642 -2023-02-13 18:19:53,373 - Epoch: [140][ 50/ 1207] Overall Loss 0.220655 Objective Loss 0.220655 LR 0.000250 Time 0.034098 -2023-02-13 18:19:53,578 - Epoch: [140][ 60/ 1207] Overall Loss 0.220124 Objective Loss 0.220124 LR 0.000250 Time 0.031825 -2023-02-13 18:19:53,777 - Epoch: [140][ 70/ 1207] Overall Loss 0.215094 Objective Loss 0.215094 LR 0.000250 Time 0.030126 -2023-02-13 18:19:53,982 - Epoch: [140][ 80/ 1207] Overall Loss 0.214772 Objective Loss 0.214772 LR 0.000250 Time 0.028918 -2023-02-13 18:19:54,184 - Epoch: [140][ 90/ 1207] Overall Loss 0.214911 Objective Loss 0.214911 LR 0.000250 Time 0.027941 -2023-02-13 18:19:54,392 - Epoch: [140][ 100/ 1207] Overall Loss 0.216959 Objective Loss 0.216959 LR 0.000250 Time 0.027227 -2023-02-13 18:19:54,598 - Epoch: [140][ 110/ 1207] Overall Loss 0.213931 Objective Loss 0.213931 LR 0.000250 Time 0.026618 -2023-02-13 18:19:54,807 - Epoch: [140][ 120/ 1207] Overall Loss 0.214174 Objective Loss 0.214174 LR 0.000250 Time 0.026136 -2023-02-13 18:19:55,014 - Epoch: [140][ 130/ 1207] Overall Loss 0.214558 Objective Loss 0.214558 LR 0.000250 Time 0.025716 -2023-02-13 18:19:55,211 - Epoch: [140][ 140/ 1207] Overall Loss 0.214787 Objective Loss 0.214787 LR 0.000250 Time 0.025280 -2023-02-13 18:19:55,400 - Epoch: [140][ 150/ 1207] Overall Loss 0.216719 Objective Loss 0.216719 LR 0.000250 Time 0.024852 -2023-02-13 18:19:55,589 - Epoch: [140][ 160/ 1207] Overall Loss 0.217953 Objective Loss 0.217953 LR 0.000250 Time 0.024478 -2023-02-13 18:19:55,777 - Epoch: [140][ 170/ 1207] Overall Loss 0.217719 Objective Loss 0.217719 LR 0.000250 Time 0.024147 -2023-02-13 18:19:55,967 - Epoch: [140][ 180/ 1207] Overall Loss 0.217146 Objective Loss 0.217146 LR 0.000250 Time 0.023854 -2023-02-13 18:19:56,156 - Epoch: [140][ 190/ 1207] Overall Loss 0.215952 Objective Loss 0.215952 LR 0.000250 Time 0.023594 -2023-02-13 18:19:56,345 - Epoch: [140][ 200/ 1207] Overall Loss 0.216697 Objective Loss 0.216697 LR 0.000250 Time 0.023357 -2023-02-13 18:19:56,534 - Epoch: [140][ 210/ 1207] Overall Loss 0.217437 Objective Loss 0.217437 LR 0.000250 Time 0.023144 -2023-02-13 18:19:56,723 - Epoch: [140][ 220/ 1207] Overall Loss 0.217789 Objective Loss 0.217789 LR 0.000250 Time 0.022950 -2023-02-13 18:19:56,912 - Epoch: [140][ 230/ 1207] Overall Loss 0.216831 Objective Loss 0.216831 LR 0.000250 Time 0.022773 -2023-02-13 18:19:57,102 - Epoch: [140][ 240/ 1207] Overall Loss 0.216652 Objective Loss 0.216652 LR 0.000250 Time 0.022611 -2023-02-13 18:19:57,291 - Epoch: [140][ 250/ 1207] Overall Loss 0.216412 Objective Loss 0.216412 LR 0.000250 Time 0.022464 -2023-02-13 18:19:57,480 - Epoch: [140][ 260/ 1207] Overall Loss 0.216481 Objective Loss 0.216481 LR 0.000250 Time 0.022325 -2023-02-13 18:19:57,670 - Epoch: [140][ 270/ 1207] Overall Loss 0.215423 Objective Loss 0.215423 LR 0.000250 Time 0.022198 -2023-02-13 18:19:57,859 - Epoch: [140][ 280/ 1207] Overall Loss 0.216257 Objective Loss 0.216257 LR 0.000250 Time 0.022082 -2023-02-13 18:19:58,049 - Epoch: [140][ 290/ 1207] Overall Loss 0.217414 Objective Loss 0.217414 LR 0.000250 Time 0.021973 -2023-02-13 18:19:58,238 - Epoch: [140][ 300/ 1207] Overall Loss 0.217157 Objective Loss 0.217157 LR 0.000250 Time 0.021870 -2023-02-13 18:19:58,427 - Epoch: [140][ 310/ 1207] Overall Loss 0.217861 Objective Loss 0.217861 LR 0.000250 Time 0.021773 -2023-02-13 18:19:58,617 - Epoch: [140][ 320/ 1207] Overall Loss 0.217440 Objective Loss 0.217440 LR 0.000250 Time 0.021683 -2023-02-13 18:19:58,806 - Epoch: [140][ 330/ 1207] Overall Loss 0.217268 Objective Loss 0.217268 LR 0.000250 Time 0.021597 -2023-02-13 18:19:58,994 - Epoch: [140][ 340/ 1207] Overall Loss 0.217520 Objective Loss 0.217520 LR 0.000250 Time 0.021515 -2023-02-13 18:19:59,183 - Epoch: [140][ 350/ 1207] Overall Loss 0.217725 Objective Loss 0.217725 LR 0.000250 Time 0.021440 -2023-02-13 18:19:59,372 - Epoch: [140][ 360/ 1207] Overall Loss 0.217462 Objective Loss 0.217462 LR 0.000250 Time 0.021368 -2023-02-13 18:19:59,561 - Epoch: [140][ 370/ 1207] Overall Loss 0.217347 Objective Loss 0.217347 LR 0.000250 Time 0.021300 -2023-02-13 18:19:59,750 - Epoch: [140][ 380/ 1207] Overall Loss 0.216893 Objective Loss 0.216893 LR 0.000250 Time 0.021235 -2023-02-13 18:19:59,939 - Epoch: [140][ 390/ 1207] Overall Loss 0.217006 Objective Loss 0.217006 LR 0.000250 Time 0.021175 -2023-02-13 18:20:00,128 - Epoch: [140][ 400/ 1207] Overall Loss 0.217126 Objective Loss 0.217126 LR 0.000250 Time 0.021118 -2023-02-13 18:20:00,317 - Epoch: [140][ 410/ 1207] Overall Loss 0.217349 Objective Loss 0.217349 LR 0.000250 Time 0.021063 -2023-02-13 18:20:00,506 - Epoch: [140][ 420/ 1207] Overall Loss 0.217930 Objective Loss 0.217930 LR 0.000250 Time 0.021010 -2023-02-13 18:20:00,696 - Epoch: [140][ 430/ 1207] Overall Loss 0.218229 Objective Loss 0.218229 LR 0.000250 Time 0.020962 -2023-02-13 18:20:00,886 - Epoch: [140][ 440/ 1207] Overall Loss 0.218348 Objective Loss 0.218348 LR 0.000250 Time 0.020916 -2023-02-13 18:20:01,075 - Epoch: [140][ 450/ 1207] Overall Loss 0.218256 Objective Loss 0.218256 LR 0.000250 Time 0.020872 -2023-02-13 18:20:01,265 - Epoch: [140][ 460/ 1207] Overall Loss 0.218182 Objective Loss 0.218182 LR 0.000250 Time 0.020830 -2023-02-13 18:20:01,454 - Epoch: [140][ 470/ 1207] Overall Loss 0.218145 Objective Loss 0.218145 LR 0.000250 Time 0.020788 -2023-02-13 18:20:01,643 - Epoch: [140][ 480/ 1207] Overall Loss 0.217652 Objective Loss 0.217652 LR 0.000250 Time 0.020749 -2023-02-13 18:20:01,833 - Epoch: [140][ 490/ 1207] Overall Loss 0.217408 Objective Loss 0.217408 LR 0.000250 Time 0.020711 -2023-02-13 18:20:02,022 - Epoch: [140][ 500/ 1207] Overall Loss 0.217341 Objective Loss 0.217341 LR 0.000250 Time 0.020675 -2023-02-13 18:20:02,212 - Epoch: [140][ 510/ 1207] Overall Loss 0.216772 Objective Loss 0.216772 LR 0.000250 Time 0.020640 -2023-02-13 18:20:02,401 - Epoch: [140][ 520/ 1207] Overall Loss 0.216434 Objective Loss 0.216434 LR 0.000250 Time 0.020607 -2023-02-13 18:20:02,590 - Epoch: [140][ 530/ 1207] Overall Loss 0.216456 Objective Loss 0.216456 LR 0.000250 Time 0.020573 -2023-02-13 18:20:02,778 - Epoch: [140][ 540/ 1207] Overall Loss 0.215917 Objective Loss 0.215917 LR 0.000250 Time 0.020541 -2023-02-13 18:20:02,967 - Epoch: [140][ 550/ 1207] Overall Loss 0.215791 Objective Loss 0.215791 LR 0.000250 Time 0.020511 -2023-02-13 18:20:03,157 - Epoch: [140][ 560/ 1207] Overall Loss 0.215571 Objective Loss 0.215571 LR 0.000250 Time 0.020483 -2023-02-13 18:20:03,346 - Epoch: [140][ 570/ 1207] Overall Loss 0.215740 Objective Loss 0.215740 LR 0.000250 Time 0.020455 -2023-02-13 18:20:03,535 - Epoch: [140][ 580/ 1207] Overall Loss 0.215551 Objective Loss 0.215551 LR 0.000250 Time 0.020427 -2023-02-13 18:20:03,725 - Epoch: [140][ 590/ 1207] Overall Loss 0.215868 Objective Loss 0.215868 LR 0.000250 Time 0.020401 -2023-02-13 18:20:03,913 - Epoch: [140][ 600/ 1207] Overall Loss 0.216095 Objective Loss 0.216095 LR 0.000250 Time 0.020374 -2023-02-13 18:20:04,102 - Epoch: [140][ 610/ 1207] Overall Loss 0.216418 Objective Loss 0.216418 LR 0.000250 Time 0.020350 -2023-02-13 18:20:04,292 - Epoch: [140][ 620/ 1207] Overall Loss 0.216033 Objective Loss 0.216033 LR 0.000250 Time 0.020327 -2023-02-13 18:20:04,481 - Epoch: [140][ 630/ 1207] Overall Loss 0.215831 Objective Loss 0.215831 LR 0.000250 Time 0.020304 -2023-02-13 18:20:04,670 - Epoch: [140][ 640/ 1207] Overall Loss 0.216274 Objective Loss 0.216274 LR 0.000250 Time 0.020281 -2023-02-13 18:20:04,859 - Epoch: [140][ 650/ 1207] Overall Loss 0.215921 Objective Loss 0.215921 LR 0.000250 Time 0.020260 -2023-02-13 18:20:05,049 - Epoch: [140][ 660/ 1207] Overall Loss 0.215795 Objective Loss 0.215795 LR 0.000250 Time 0.020240 -2023-02-13 18:20:05,239 - Epoch: [140][ 670/ 1207] Overall Loss 0.215859 Objective Loss 0.215859 LR 0.000250 Time 0.020221 -2023-02-13 18:20:05,428 - Epoch: [140][ 680/ 1207] Overall Loss 0.215990 Objective Loss 0.215990 LR 0.000250 Time 0.020201 -2023-02-13 18:20:05,617 - Epoch: [140][ 690/ 1207] Overall Loss 0.216002 Objective Loss 0.216002 LR 0.000250 Time 0.020182 -2023-02-13 18:20:05,806 - Epoch: [140][ 700/ 1207] Overall Loss 0.215660 Objective Loss 0.215660 LR 0.000250 Time 0.020162 -2023-02-13 18:20:05,996 - Epoch: [140][ 710/ 1207] Overall Loss 0.215674 Objective Loss 0.215674 LR 0.000250 Time 0.020146 -2023-02-13 18:20:06,186 - Epoch: [140][ 720/ 1207] Overall Loss 0.215499 Objective Loss 0.215499 LR 0.000250 Time 0.020129 -2023-02-13 18:20:06,375 - Epoch: [140][ 730/ 1207] Overall Loss 0.215886 Objective Loss 0.215886 LR 0.000250 Time 0.020112 -2023-02-13 18:20:06,564 - Epoch: [140][ 740/ 1207] Overall Loss 0.216103 Objective Loss 0.216103 LR 0.000250 Time 0.020095 -2023-02-13 18:20:06,754 - Epoch: [140][ 750/ 1207] Overall Loss 0.215997 Objective Loss 0.215997 LR 0.000250 Time 0.020080 -2023-02-13 18:20:06,943 - Epoch: [140][ 760/ 1207] Overall Loss 0.216584 Objective Loss 0.216584 LR 0.000250 Time 0.020065 -2023-02-13 18:20:07,134 - Epoch: [140][ 770/ 1207] Overall Loss 0.216727 Objective Loss 0.216727 LR 0.000250 Time 0.020051 -2023-02-13 18:20:07,324 - Epoch: [140][ 780/ 1207] Overall Loss 0.216570 Objective Loss 0.216570 LR 0.000250 Time 0.020037 -2023-02-13 18:20:07,513 - Epoch: [140][ 790/ 1207] Overall Loss 0.216795 Objective Loss 0.216795 LR 0.000250 Time 0.020022 -2023-02-13 18:20:07,702 - Epoch: [140][ 800/ 1207] Overall Loss 0.216696 Objective Loss 0.216696 LR 0.000250 Time 0.020007 -2023-02-13 18:20:07,892 - Epoch: [140][ 810/ 1207] Overall Loss 0.216870 Objective Loss 0.216870 LR 0.000250 Time 0.019994 -2023-02-13 18:20:08,081 - Epoch: [140][ 820/ 1207] Overall Loss 0.216642 Objective Loss 0.216642 LR 0.000250 Time 0.019981 -2023-02-13 18:20:08,272 - Epoch: [140][ 830/ 1207] Overall Loss 0.216603 Objective Loss 0.216603 LR 0.000250 Time 0.019970 -2023-02-13 18:20:08,461 - Epoch: [140][ 840/ 1207] Overall Loss 0.216861 Objective Loss 0.216861 LR 0.000250 Time 0.019957 -2023-02-13 18:20:08,650 - Epoch: [140][ 850/ 1207] Overall Loss 0.216942 Objective Loss 0.216942 LR 0.000250 Time 0.019944 -2023-02-13 18:20:08,839 - Epoch: [140][ 860/ 1207] Overall Loss 0.217358 Objective Loss 0.217358 LR 0.000250 Time 0.019931 -2023-02-13 18:20:09,029 - Epoch: [140][ 870/ 1207] Overall Loss 0.217177 Objective Loss 0.217177 LR 0.000250 Time 0.019919 -2023-02-13 18:20:09,219 - Epoch: [140][ 880/ 1207] Overall Loss 0.217088 Objective Loss 0.217088 LR 0.000250 Time 0.019909 -2023-02-13 18:20:09,408 - Epoch: [140][ 890/ 1207] Overall Loss 0.217022 Objective Loss 0.217022 LR 0.000250 Time 0.019897 -2023-02-13 18:20:09,597 - Epoch: [140][ 900/ 1207] Overall Loss 0.217111 Objective Loss 0.217111 LR 0.000250 Time 0.019886 -2023-02-13 18:20:09,787 - Epoch: [140][ 910/ 1207] Overall Loss 0.217118 Objective Loss 0.217118 LR 0.000250 Time 0.019876 -2023-02-13 18:20:09,976 - Epoch: [140][ 920/ 1207] Overall Loss 0.217106 Objective Loss 0.217106 LR 0.000250 Time 0.019865 -2023-02-13 18:20:10,165 - Epoch: [140][ 930/ 1207] Overall Loss 0.217189 Objective Loss 0.217189 LR 0.000250 Time 0.019854 -2023-02-13 18:20:10,354 - Epoch: [140][ 940/ 1207] Overall Loss 0.217213 Objective Loss 0.217213 LR 0.000250 Time 0.019844 -2023-02-13 18:20:10,543 - Epoch: [140][ 950/ 1207] Overall Loss 0.217229 Objective Loss 0.217229 LR 0.000250 Time 0.019833 -2023-02-13 18:20:10,733 - Epoch: [140][ 960/ 1207] Overall Loss 0.217171 Objective Loss 0.217171 LR 0.000250 Time 0.019824 -2023-02-13 18:20:10,922 - Epoch: [140][ 970/ 1207] Overall Loss 0.217410 Objective Loss 0.217410 LR 0.000250 Time 0.019815 -2023-02-13 18:20:11,113 - Epoch: [140][ 980/ 1207] Overall Loss 0.217279 Objective Loss 0.217279 LR 0.000250 Time 0.019806 -2023-02-13 18:20:11,302 - Epoch: [140][ 990/ 1207] Overall Loss 0.217247 Objective Loss 0.217247 LR 0.000250 Time 0.019797 -2023-02-13 18:20:11,490 - Epoch: [140][ 1000/ 1207] Overall Loss 0.217063 Objective Loss 0.217063 LR 0.000250 Time 0.019787 -2023-02-13 18:20:11,680 - Epoch: [140][ 1010/ 1207] Overall Loss 0.217235 Objective Loss 0.217235 LR 0.000250 Time 0.019778 -2023-02-13 18:20:11,868 - Epoch: [140][ 1020/ 1207] Overall Loss 0.217078 Objective Loss 0.217078 LR 0.000250 Time 0.019769 -2023-02-13 18:20:12,059 - Epoch: [140][ 1030/ 1207] Overall Loss 0.217016 Objective Loss 0.217016 LR 0.000250 Time 0.019761 -2023-02-13 18:20:12,248 - Epoch: [140][ 1040/ 1207] Overall Loss 0.216845 Objective Loss 0.216845 LR 0.000250 Time 0.019753 -2023-02-13 18:20:12,438 - Epoch: [140][ 1050/ 1207] Overall Loss 0.216825 Objective Loss 0.216825 LR 0.000250 Time 0.019745 -2023-02-13 18:20:12,626 - Epoch: [140][ 1060/ 1207] Overall Loss 0.217095 Objective Loss 0.217095 LR 0.000250 Time 0.019736 -2023-02-13 18:20:12,816 - Epoch: [140][ 1070/ 1207] Overall Loss 0.216886 Objective Loss 0.216886 LR 0.000250 Time 0.019729 -2023-02-13 18:20:13,005 - Epoch: [140][ 1080/ 1207] Overall Loss 0.216983 Objective Loss 0.216983 LR 0.000250 Time 0.019721 -2023-02-13 18:20:13,195 - Epoch: [140][ 1090/ 1207] Overall Loss 0.216926 Objective Loss 0.216926 LR 0.000250 Time 0.019714 -2023-02-13 18:20:13,385 - Epoch: [140][ 1100/ 1207] Overall Loss 0.216737 Objective Loss 0.216737 LR 0.000250 Time 0.019707 -2023-02-13 18:20:13,574 - Epoch: [140][ 1110/ 1207] Overall Loss 0.216710 Objective Loss 0.216710 LR 0.000250 Time 0.019700 -2023-02-13 18:20:13,764 - Epoch: [140][ 1120/ 1207] Overall Loss 0.216620 Objective Loss 0.216620 LR 0.000250 Time 0.019693 -2023-02-13 18:20:13,952 - Epoch: [140][ 1130/ 1207] Overall Loss 0.216634 Objective Loss 0.216634 LR 0.000250 Time 0.019685 -2023-02-13 18:20:14,142 - Epoch: [140][ 1140/ 1207] Overall Loss 0.216821 Objective Loss 0.216821 LR 0.000250 Time 0.019678 -2023-02-13 18:20:14,332 - Epoch: [140][ 1150/ 1207] Overall Loss 0.216664 Objective Loss 0.216664 LR 0.000250 Time 0.019672 -2023-02-13 18:20:14,522 - Epoch: [140][ 1160/ 1207] Overall Loss 0.216589 Objective Loss 0.216589 LR 0.000250 Time 0.019666 -2023-02-13 18:20:14,712 - Epoch: [140][ 1170/ 1207] Overall Loss 0.216538 Objective Loss 0.216538 LR 0.000250 Time 0.019660 -2023-02-13 18:20:14,901 - Epoch: [140][ 1180/ 1207] Overall Loss 0.216264 Objective Loss 0.216264 LR 0.000250 Time 0.019653 -2023-02-13 18:20:15,091 - Epoch: [140][ 1190/ 1207] Overall Loss 0.216156 Objective Loss 0.216156 LR 0.000250 Time 0.019647 -2023-02-13 18:20:15,331 - Epoch: [140][ 1200/ 1207] Overall Loss 0.216147 Objective Loss 0.216147 LR 0.000250 Time 0.019684 -2023-02-13 18:20:15,447 - Epoch: [140][ 1207/ 1207] Overall Loss 0.216228 Objective Loss 0.216228 Top1 86.890244 Top5 98.780488 LR 0.000250 Time 0.019665 -2023-02-13 18:20:15,520 - --- validate (epoch=140)----------- -2023-02-13 18:20:15,520 - 34311 samples (256 per mini-batch) -2023-02-13 18:20:15,921 - Epoch: [140][ 10/ 135] Loss 0.280064 Top1 85.703125 Top5 97.929688 -2023-02-13 18:20:16,066 - Epoch: [140][ 20/ 135] Loss 0.279933 Top1 85.820312 Top5 98.085938 -2023-02-13 18:20:16,205 - Epoch: [140][ 30/ 135] Loss 0.291697 Top1 85.716146 Top5 98.059896 -2023-02-13 18:20:16,339 - Epoch: [140][ 40/ 135] Loss 0.293645 Top1 85.517578 Top5 97.910156 -2023-02-13 18:20:16,464 - Epoch: [140][ 50/ 135] Loss 0.298781 Top1 85.375000 Top5 97.921875 -2023-02-13 18:20:16,592 - Epoch: [140][ 60/ 135] Loss 0.294962 Top1 85.358073 Top5 97.949219 -2023-02-13 18:20:16,721 - Epoch: [140][ 70/ 135] Loss 0.297041 Top1 85.290179 Top5 97.963170 -2023-02-13 18:20:16,850 - Epoch: [140][ 80/ 135] Loss 0.298929 Top1 85.180664 Top5 97.924805 -2023-02-13 18:20:16,980 - Epoch: [140][ 90/ 135] Loss 0.297960 Top1 85.221354 Top5 97.890625 -2023-02-13 18:20:17,111 - Epoch: [140][ 100/ 135] Loss 0.294937 Top1 85.304688 Top5 97.863281 -2023-02-13 18:20:17,241 - Epoch: [140][ 110/ 135] Loss 0.293399 Top1 85.316051 Top5 97.911932 -2023-02-13 18:20:17,370 - Epoch: [140][ 120/ 135] Loss 0.293474 Top1 85.273438 Top5 97.923177 -2023-02-13 18:20:17,501 - Epoch: [140][ 130/ 135] Loss 0.293566 Top1 85.333534 Top5 97.899639 -2023-02-13 18:20:17,548 - Epoch: [140][ 135/ 135] Loss 0.294494 Top1 85.328320 Top5 97.901548 -2023-02-13 18:20:17,617 - ==> Top1: 85.328 Top5: 97.902 Loss: 0.294 - -2023-02-13 18:20:17,617 - ==> Confusion: -[[ 863 7 5 0 10 1 0 2 2 44 1 3 1 5 6 1 2 3 1 1 9] - [ 5 954 1 2 11 20 1 12 2 2 1 0 1 0 0 1 5 0 3 2 10] - [ 7 2 960 13 4 2 17 13 0 1 2 0 3 5 3 4 3 3 5 3 8] - [ 3 0 24 909 4 4 0 2 4 1 10 0 8 0 21 2 3 6 12 0 3] - [ 17 9 1 2 989 5 1 0 1 4 0 1 1 5 9 6 5 2 1 6 1] - [ 3 21 3 3 8 965 5 11 2 4 2 8 4 14 0 1 4 0 0 4 8] - [ 3 4 12 0 0 8 1039 3 0 1 5 2 2 1 0 4 0 2 1 8 4] - [ 1 6 8 4 0 24 2 933 1 1 2 5 4 2 0 0 1 0 15 10 5] - [ 14 2 0 2 1 1 1 1 897 49 8 1 1 10 12 2 1 0 3 0 3] - [ 66 2 3 1 8 1 0 4 26 875 0 1 0 13 5 1 0 1 1 0 4] - [ 2 3 4 11 1 5 2 6 18 0 974 1 1 12 3 0 0 0 5 1 2] - [ 1 3 4 0 3 17 0 6 2 2 0 898 29 10 1 1 3 12 2 10 1] - [ 2 1 0 7 1 2 0 1 1 3 0 16 889 2 3 6 1 14 2 1 7] - [ 3 3 0 0 9 5 1 4 10 17 5 3 4 943 3 5 2 2 0 0 5] - [ 5 2 3 15 4 5 0 2 17 6 2 0 3 0 1003 0 2 8 7 0 8] - [ 1 3 9 1 5 2 4 1 0 0 0 8 8 2 1 975 5 10 1 6 4] - [ 4 6 0 1 12 3 0 1 1 1 0 1 2 2 1 11 1000 1 0 4 10] - [ 3 4 2 3 0 1 2 0 0 2 2 1 16 2 0 14 2 989 1 2 5] - [ 4 8 2 9 0 2 0 22 3 0 5 1 2 0 15 1 1 2 1006 2 1] - [ 1 2 1 0 3 6 4 8 1 1 0 9 3 4 0 5 6 3 0 1084 7] - [ 134 252 224 127 129 185 103 154 96 86 179 88 318 260 149 91 218 105 171 233 10132]] - -2023-02-13 18:20:17,619 - ==> Best [Top1: 85.328 Top5: 97.902 Sparsity:0.00 Params: 148928 on epoch: 140] -2023-02-13 18:20:17,619 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:20:17,625 - - -2023-02-13 18:20:17,626 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:20:18,628 - Epoch: [141][ 10/ 1207] Overall Loss 0.188210 Objective Loss 0.188210 LR 0.000250 Time 0.100163 -2023-02-13 18:20:18,829 - Epoch: [141][ 20/ 1207] Overall Loss 0.193847 Objective Loss 0.193847 LR 0.000250 Time 0.060136 -2023-02-13 18:20:19,024 - Epoch: [141][ 30/ 1207] Overall Loss 0.199867 Objective Loss 0.199867 LR 0.000250 Time 0.046562 -2023-02-13 18:20:19,217 - Epoch: [141][ 40/ 1207] Overall Loss 0.199274 Objective Loss 0.199274 LR 0.000250 Time 0.039732 -2023-02-13 18:20:19,406 - Epoch: [141][ 50/ 1207] Overall Loss 0.202043 Objective Loss 0.202043 LR 0.000250 Time 0.035564 -2023-02-13 18:20:19,595 - Epoch: [141][ 60/ 1207] Overall Loss 0.203838 Objective Loss 0.203838 LR 0.000250 Time 0.032788 -2023-02-13 18:20:19,786 - Epoch: [141][ 70/ 1207] Overall Loss 0.206946 Objective Loss 0.206946 LR 0.000250 Time 0.030815 -2023-02-13 18:20:19,975 - Epoch: [141][ 80/ 1207] Overall Loss 0.208461 Objective Loss 0.208461 LR 0.000250 Time 0.029329 -2023-02-13 18:20:20,166 - Epoch: [141][ 90/ 1207] Overall Loss 0.207461 Objective Loss 0.207461 LR 0.000250 Time 0.028182 -2023-02-13 18:20:20,356 - Epoch: [141][ 100/ 1207] Overall Loss 0.207399 Objective Loss 0.207399 LR 0.000250 Time 0.027260 -2023-02-13 18:20:20,545 - Epoch: [141][ 110/ 1207] Overall Loss 0.207345 Objective Loss 0.207345 LR 0.000250 Time 0.026500 -2023-02-13 18:20:20,735 - Epoch: [141][ 120/ 1207] Overall Loss 0.205958 Objective Loss 0.205958 LR 0.000250 Time 0.025869 -2023-02-13 18:20:20,925 - Epoch: [141][ 130/ 1207] Overall Loss 0.205089 Objective Loss 0.205089 LR 0.000250 Time 0.025339 -2023-02-13 18:20:21,114 - Epoch: [141][ 140/ 1207] Overall Loss 0.205600 Objective Loss 0.205600 LR 0.000250 Time 0.024878 -2023-02-13 18:20:21,304 - Epoch: [141][ 150/ 1207] Overall Loss 0.205394 Objective Loss 0.205394 LR 0.000250 Time 0.024481 -2023-02-13 18:20:21,493 - Epoch: [141][ 160/ 1207] Overall Loss 0.205144 Objective Loss 0.205144 LR 0.000250 Time 0.024132 -2023-02-13 18:20:21,683 - Epoch: [141][ 170/ 1207] Overall Loss 0.205985 Objective Loss 0.205985 LR 0.000250 Time 0.023825 -2023-02-13 18:20:21,872 - Epoch: [141][ 180/ 1207] Overall Loss 0.204759 Objective Loss 0.204759 LR 0.000250 Time 0.023550 -2023-02-13 18:20:22,062 - Epoch: [141][ 190/ 1207] Overall Loss 0.204734 Objective Loss 0.204734 LR 0.000250 Time 0.023310 -2023-02-13 18:20:22,253 - Epoch: [141][ 200/ 1207] Overall Loss 0.204146 Objective Loss 0.204146 LR 0.000250 Time 0.023096 -2023-02-13 18:20:22,442 - Epoch: [141][ 210/ 1207] Overall Loss 0.204456 Objective Loss 0.204456 LR 0.000250 Time 0.022895 -2023-02-13 18:20:22,632 - Epoch: [141][ 220/ 1207] Overall Loss 0.205475 Objective Loss 0.205475 LR 0.000250 Time 0.022716 -2023-02-13 18:20:22,822 - Epoch: [141][ 230/ 1207] Overall Loss 0.206901 Objective Loss 0.206901 LR 0.000250 Time 0.022552 -2023-02-13 18:20:23,011 - Epoch: [141][ 240/ 1207] Overall Loss 0.206909 Objective Loss 0.206909 LR 0.000250 Time 0.022400 -2023-02-13 18:20:23,201 - Epoch: [141][ 250/ 1207] Overall Loss 0.205520 Objective Loss 0.205520 LR 0.000250 Time 0.022263 -2023-02-13 18:20:23,390 - Epoch: [141][ 260/ 1207] Overall Loss 0.205499 Objective Loss 0.205499 LR 0.000250 Time 0.022133 -2023-02-13 18:20:23,579 - Epoch: [141][ 270/ 1207] Overall Loss 0.206543 Objective Loss 0.206543 LR 0.000250 Time 0.022013 -2023-02-13 18:20:23,770 - Epoch: [141][ 280/ 1207] Overall Loss 0.207084 Objective Loss 0.207084 LR 0.000250 Time 0.021904 -2023-02-13 18:20:23,959 - Epoch: [141][ 290/ 1207] Overall Loss 0.208341 Objective Loss 0.208341 LR 0.000250 Time 0.021801 -2023-02-13 18:20:24,149 - Epoch: [141][ 300/ 1207] Overall Loss 0.208080 Objective Loss 0.208080 LR 0.000250 Time 0.021705 -2023-02-13 18:20:24,338 - Epoch: [141][ 310/ 1207] Overall Loss 0.208062 Objective Loss 0.208062 LR 0.000250 Time 0.021616 -2023-02-13 18:20:24,528 - Epoch: [141][ 320/ 1207] Overall Loss 0.208349 Objective Loss 0.208349 LR 0.000250 Time 0.021530 -2023-02-13 18:20:24,717 - Epoch: [141][ 330/ 1207] Overall Loss 0.207785 Objective Loss 0.207785 LR 0.000250 Time 0.021451 -2023-02-13 18:20:24,907 - Epoch: [141][ 340/ 1207] Overall Loss 0.207943 Objective Loss 0.207943 LR 0.000250 Time 0.021377 -2023-02-13 18:20:25,097 - Epoch: [141][ 350/ 1207] Overall Loss 0.207796 Objective Loss 0.207796 LR 0.000250 Time 0.021308 -2023-02-13 18:20:25,287 - Epoch: [141][ 360/ 1207] Overall Loss 0.207056 Objective Loss 0.207056 LR 0.000250 Time 0.021243 -2023-02-13 18:20:25,476 - Epoch: [141][ 370/ 1207] Overall Loss 0.207048 Objective Loss 0.207048 LR 0.000250 Time 0.021180 -2023-02-13 18:20:25,666 - Epoch: [141][ 380/ 1207] Overall Loss 0.207594 Objective Loss 0.207594 LR 0.000250 Time 0.021120 -2023-02-13 18:20:25,856 - Epoch: [141][ 390/ 1207] Overall Loss 0.207415 Objective Loss 0.207415 LR 0.000250 Time 0.021065 -2023-02-13 18:20:26,046 - Epoch: [141][ 400/ 1207] Overall Loss 0.207909 Objective Loss 0.207909 LR 0.000250 Time 0.021014 -2023-02-13 18:20:26,237 - Epoch: [141][ 410/ 1207] Overall Loss 0.208095 Objective Loss 0.208095 LR 0.000250 Time 0.020964 -2023-02-13 18:20:26,426 - Epoch: [141][ 420/ 1207] Overall Loss 0.208401 Objective Loss 0.208401 LR 0.000250 Time 0.020916 -2023-02-13 18:20:26,616 - Epoch: [141][ 430/ 1207] Overall Loss 0.208277 Objective Loss 0.208277 LR 0.000250 Time 0.020870 -2023-02-13 18:20:26,806 - Epoch: [141][ 440/ 1207] Overall Loss 0.208524 Objective Loss 0.208524 LR 0.000250 Time 0.020825 -2023-02-13 18:20:26,996 - Epoch: [141][ 450/ 1207] Overall Loss 0.208572 Objective Loss 0.208572 LR 0.000250 Time 0.020785 -2023-02-13 18:20:27,186 - Epoch: [141][ 460/ 1207] Overall Loss 0.208636 Objective Loss 0.208636 LR 0.000250 Time 0.020745 -2023-02-13 18:20:27,376 - Epoch: [141][ 470/ 1207] Overall Loss 0.208986 Objective Loss 0.208986 LR 0.000250 Time 0.020707 -2023-02-13 18:20:27,565 - Epoch: [141][ 480/ 1207] Overall Loss 0.209033 Objective Loss 0.209033 LR 0.000250 Time 0.020670 -2023-02-13 18:20:27,755 - Epoch: [141][ 490/ 1207] Overall Loss 0.208930 Objective Loss 0.208930 LR 0.000250 Time 0.020634 -2023-02-13 18:20:27,945 - Epoch: [141][ 500/ 1207] Overall Loss 0.208614 Objective Loss 0.208614 LR 0.000250 Time 0.020600 -2023-02-13 18:20:28,135 - Epoch: [141][ 510/ 1207] Overall Loss 0.208653 Objective Loss 0.208653 LR 0.000250 Time 0.020568 -2023-02-13 18:20:28,324 - Epoch: [141][ 520/ 1207] Overall Loss 0.208434 Objective Loss 0.208434 LR 0.000250 Time 0.020536 -2023-02-13 18:20:28,513 - Epoch: [141][ 530/ 1207] Overall Loss 0.208684 Objective Loss 0.208684 LR 0.000250 Time 0.020505 -2023-02-13 18:20:28,703 - Epoch: [141][ 540/ 1207] Overall Loss 0.208752 Objective Loss 0.208752 LR 0.000250 Time 0.020475 -2023-02-13 18:20:28,892 - Epoch: [141][ 550/ 1207] Overall Loss 0.208831 Objective Loss 0.208831 LR 0.000250 Time 0.020446 -2023-02-13 18:20:29,082 - Epoch: [141][ 560/ 1207] Overall Loss 0.208617 Objective Loss 0.208617 LR 0.000250 Time 0.020419 -2023-02-13 18:20:29,272 - Epoch: [141][ 570/ 1207] Overall Loss 0.208727 Objective Loss 0.208727 LR 0.000250 Time 0.020394 -2023-02-13 18:20:29,462 - Epoch: [141][ 580/ 1207] Overall Loss 0.208446 Objective Loss 0.208446 LR 0.000250 Time 0.020369 -2023-02-13 18:20:29,651 - Epoch: [141][ 590/ 1207] Overall Loss 0.208330 Objective Loss 0.208330 LR 0.000250 Time 0.020344 -2023-02-13 18:20:29,840 - Epoch: [141][ 600/ 1207] Overall Loss 0.208013 Objective Loss 0.208013 LR 0.000250 Time 0.020320 -2023-02-13 18:20:30,030 - Epoch: [141][ 610/ 1207] Overall Loss 0.208201 Objective Loss 0.208201 LR 0.000250 Time 0.020297 -2023-02-13 18:20:30,221 - Epoch: [141][ 620/ 1207] Overall Loss 0.207898 Objective Loss 0.207898 LR 0.000250 Time 0.020277 -2023-02-13 18:20:30,411 - Epoch: [141][ 630/ 1207] Overall Loss 0.207981 Objective Loss 0.207981 LR 0.000250 Time 0.020256 -2023-02-13 18:20:30,600 - Epoch: [141][ 640/ 1207] Overall Loss 0.208075 Objective Loss 0.208075 LR 0.000250 Time 0.020235 -2023-02-13 18:20:30,789 - Epoch: [141][ 650/ 1207] Overall Loss 0.208346 Objective Loss 0.208346 LR 0.000250 Time 0.020214 -2023-02-13 18:20:30,980 - Epoch: [141][ 660/ 1207] Overall Loss 0.208585 Objective Loss 0.208585 LR 0.000250 Time 0.020196 -2023-02-13 18:20:31,171 - Epoch: [141][ 670/ 1207] Overall Loss 0.208434 Objective Loss 0.208434 LR 0.000250 Time 0.020179 -2023-02-13 18:20:31,361 - Epoch: [141][ 680/ 1207] Overall Loss 0.208423 Objective Loss 0.208423 LR 0.000250 Time 0.020161 -2023-02-13 18:20:31,550 - Epoch: [141][ 690/ 1207] Overall Loss 0.208618 Objective Loss 0.208618 LR 0.000250 Time 0.020143 -2023-02-13 18:20:31,741 - Epoch: [141][ 700/ 1207] Overall Loss 0.208713 Objective Loss 0.208713 LR 0.000250 Time 0.020126 -2023-02-13 18:20:31,930 - Epoch: [141][ 710/ 1207] Overall Loss 0.208899 Objective Loss 0.208899 LR 0.000250 Time 0.020109 -2023-02-13 18:20:32,121 - Epoch: [141][ 720/ 1207] Overall Loss 0.208586 Objective Loss 0.208586 LR 0.000250 Time 0.020094 -2023-02-13 18:20:32,311 - Epoch: [141][ 730/ 1207] Overall Loss 0.208557 Objective Loss 0.208557 LR 0.000250 Time 0.020079 -2023-02-13 18:20:32,500 - Epoch: [141][ 740/ 1207] Overall Loss 0.208465 Objective Loss 0.208465 LR 0.000250 Time 0.020063 -2023-02-13 18:20:32,689 - Epoch: [141][ 750/ 1207] Overall Loss 0.208645 Objective Loss 0.208645 LR 0.000250 Time 0.020047 -2023-02-13 18:20:32,878 - Epoch: [141][ 760/ 1207] Overall Loss 0.208314 Objective Loss 0.208314 LR 0.000250 Time 0.020031 -2023-02-13 18:20:33,068 - Epoch: [141][ 770/ 1207] Overall Loss 0.208310 Objective Loss 0.208310 LR 0.000250 Time 0.020017 -2023-02-13 18:20:33,260 - Epoch: [141][ 780/ 1207] Overall Loss 0.208582 Objective Loss 0.208582 LR 0.000250 Time 0.020006 -2023-02-13 18:20:33,449 - Epoch: [141][ 790/ 1207] Overall Loss 0.208551 Objective Loss 0.208551 LR 0.000250 Time 0.019991 -2023-02-13 18:20:33,638 - Epoch: [141][ 800/ 1207] Overall Loss 0.208402 Objective Loss 0.208402 LR 0.000250 Time 0.019978 -2023-02-13 18:20:33,827 - Epoch: [141][ 810/ 1207] Overall Loss 0.208483 Objective Loss 0.208483 LR 0.000250 Time 0.019964 -2023-02-13 18:20:34,017 - Epoch: [141][ 820/ 1207] Overall Loss 0.208506 Objective Loss 0.208506 LR 0.000250 Time 0.019952 -2023-02-13 18:20:34,208 - Epoch: [141][ 830/ 1207] Overall Loss 0.208641 Objective Loss 0.208641 LR 0.000250 Time 0.019941 -2023-02-13 18:20:34,398 - Epoch: [141][ 840/ 1207] Overall Loss 0.208490 Objective Loss 0.208490 LR 0.000250 Time 0.019930 -2023-02-13 18:20:34,588 - Epoch: [141][ 850/ 1207] Overall Loss 0.208586 Objective Loss 0.208586 LR 0.000250 Time 0.019918 -2023-02-13 18:20:34,778 - Epoch: [141][ 860/ 1207] Overall Loss 0.208230 Objective Loss 0.208230 LR 0.000250 Time 0.019907 -2023-02-13 18:20:34,968 - Epoch: [141][ 870/ 1207] Overall Loss 0.208142 Objective Loss 0.208142 LR 0.000250 Time 0.019896 -2023-02-13 18:20:35,158 - Epoch: [141][ 880/ 1207] Overall Loss 0.208527 Objective Loss 0.208527 LR 0.000250 Time 0.019886 -2023-02-13 18:20:35,348 - Epoch: [141][ 890/ 1207] Overall Loss 0.208607 Objective Loss 0.208607 LR 0.000250 Time 0.019875 -2023-02-13 18:20:35,538 - Epoch: [141][ 900/ 1207] Overall Loss 0.208514 Objective Loss 0.208514 LR 0.000250 Time 0.019864 -2023-02-13 18:20:35,727 - Epoch: [141][ 910/ 1207] Overall Loss 0.208573 Objective Loss 0.208573 LR 0.000250 Time 0.019854 -2023-02-13 18:20:35,918 - Epoch: [141][ 920/ 1207] Overall Loss 0.208546 Objective Loss 0.208546 LR 0.000250 Time 0.019845 -2023-02-13 18:20:36,108 - Epoch: [141][ 930/ 1207] Overall Loss 0.208853 Objective Loss 0.208853 LR 0.000250 Time 0.019835 -2023-02-13 18:20:36,299 - Epoch: [141][ 940/ 1207] Overall Loss 0.208743 Objective Loss 0.208743 LR 0.000250 Time 0.019828 -2023-02-13 18:20:36,488 - Epoch: [141][ 950/ 1207] Overall Loss 0.208924 Objective Loss 0.208924 LR 0.000250 Time 0.019817 -2023-02-13 18:20:36,678 - Epoch: [141][ 960/ 1207] Overall Loss 0.208901 Objective Loss 0.208901 LR 0.000250 Time 0.019809 -2023-02-13 18:20:36,868 - Epoch: [141][ 970/ 1207] Overall Loss 0.208653 Objective Loss 0.208653 LR 0.000250 Time 0.019800 -2023-02-13 18:20:37,059 - Epoch: [141][ 980/ 1207] Overall Loss 0.208842 Objective Loss 0.208842 LR 0.000250 Time 0.019792 -2023-02-13 18:20:37,250 - Epoch: [141][ 990/ 1207] Overall Loss 0.209058 Objective Loss 0.209058 LR 0.000250 Time 0.019785 -2023-02-13 18:20:37,441 - Epoch: [141][ 1000/ 1207] Overall Loss 0.209197 Objective Loss 0.209197 LR 0.000250 Time 0.019778 -2023-02-13 18:20:37,630 - Epoch: [141][ 1010/ 1207] Overall Loss 0.209000 Objective Loss 0.209000 LR 0.000250 Time 0.019769 -2023-02-13 18:20:37,821 - Epoch: [141][ 1020/ 1207] Overall Loss 0.209299 Objective Loss 0.209299 LR 0.000250 Time 0.019761 -2023-02-13 18:20:38,010 - Epoch: [141][ 1030/ 1207] Overall Loss 0.209459 Objective Loss 0.209459 LR 0.000250 Time 0.019753 -2023-02-13 18:20:38,201 - Epoch: [141][ 1040/ 1207] Overall Loss 0.209223 Objective Loss 0.209223 LR 0.000250 Time 0.019746 -2023-02-13 18:20:38,391 - Epoch: [141][ 1050/ 1207] Overall Loss 0.209063 Objective Loss 0.209063 LR 0.000250 Time 0.019739 -2023-02-13 18:20:38,582 - Epoch: [141][ 1060/ 1207] Overall Loss 0.209080 Objective Loss 0.209080 LR 0.000250 Time 0.019732 -2023-02-13 18:20:38,771 - Epoch: [141][ 1070/ 1207] Overall Loss 0.209169 Objective Loss 0.209169 LR 0.000250 Time 0.019724 -2023-02-13 18:20:38,961 - Epoch: [141][ 1080/ 1207] Overall Loss 0.209348 Objective Loss 0.209348 LR 0.000250 Time 0.019717 -2023-02-13 18:20:39,151 - Epoch: [141][ 1090/ 1207] Overall Loss 0.209198 Objective Loss 0.209198 LR 0.000250 Time 0.019710 -2023-02-13 18:20:39,342 - Epoch: [141][ 1100/ 1207] Overall Loss 0.209167 Objective Loss 0.209167 LR 0.000250 Time 0.019705 -2023-02-13 18:20:39,532 - Epoch: [141][ 1110/ 1207] Overall Loss 0.209022 Objective Loss 0.209022 LR 0.000250 Time 0.019697 -2023-02-13 18:20:39,723 - Epoch: [141][ 1120/ 1207] Overall Loss 0.208987 Objective Loss 0.208987 LR 0.000250 Time 0.019692 -2023-02-13 18:20:39,912 - Epoch: [141][ 1130/ 1207] Overall Loss 0.209121 Objective Loss 0.209121 LR 0.000250 Time 0.019685 -2023-02-13 18:20:40,102 - Epoch: [141][ 1140/ 1207] Overall Loss 0.209086 Objective Loss 0.209086 LR 0.000250 Time 0.019679 -2023-02-13 18:20:40,293 - Epoch: [141][ 1150/ 1207] Overall Loss 0.209058 Objective Loss 0.209058 LR 0.000250 Time 0.019673 -2023-02-13 18:20:40,484 - Epoch: [141][ 1160/ 1207] Overall Loss 0.209102 Objective Loss 0.209102 LR 0.000250 Time 0.019668 -2023-02-13 18:20:40,675 - Epoch: [141][ 1170/ 1207] Overall Loss 0.209107 Objective Loss 0.209107 LR 0.000250 Time 0.019662 -2023-02-13 18:20:40,866 - Epoch: [141][ 1180/ 1207] Overall Loss 0.208929 Objective Loss 0.208929 LR 0.000250 Time 0.019657 -2023-02-13 18:20:41,057 - Epoch: [141][ 1190/ 1207] Overall Loss 0.208886 Objective Loss 0.208886 LR 0.000250 Time 0.019652 -2023-02-13 18:20:41,303 - Epoch: [141][ 1200/ 1207] Overall Loss 0.209083 Objective Loss 0.209083 LR 0.000250 Time 0.019693 -2023-02-13 18:20:41,418 - Epoch: [141][ 1207/ 1207] Overall Loss 0.209189 Objective Loss 0.209189 Top1 85.975610 Top5 98.475610 LR 0.000250 Time 0.019674 -2023-02-13 18:20:41,502 - --- validate (epoch=141)----------- -2023-02-13 18:20:41,503 - 34311 samples (256 per mini-batch) -2023-02-13 18:20:41,906 - Epoch: [141][ 10/ 135] Loss 0.301836 Top1 85.781250 Top5 98.320312 -2023-02-13 18:20:42,030 - Epoch: [141][ 20/ 135] Loss 0.285103 Top1 85.976562 Top5 98.320312 -2023-02-13 18:20:42,161 - Epoch: [141][ 30/ 135] Loss 0.298246 Top1 85.481771 Top5 98.190104 -2023-02-13 18:20:42,289 - Epoch: [141][ 40/ 135] Loss 0.298014 Top1 85.087891 Top5 98.144531 -2023-02-13 18:20:42,417 - Epoch: [141][ 50/ 135] Loss 0.304159 Top1 85.062500 Top5 98.007812 -2023-02-13 18:20:42,544 - Epoch: [141][ 60/ 135] Loss 0.301401 Top1 85.065104 Top5 97.942708 -2023-02-13 18:20:42,673 - Epoch: [141][ 70/ 135] Loss 0.302999 Top1 85.016741 Top5 97.924107 -2023-02-13 18:20:42,800 - Epoch: [141][ 80/ 135] Loss 0.303829 Top1 84.873047 Top5 97.885742 -2023-02-13 18:20:42,927 - Epoch: [141][ 90/ 135] Loss 0.302083 Top1 84.856771 Top5 97.855903 -2023-02-13 18:20:43,056 - Epoch: [141][ 100/ 135] Loss 0.299249 Top1 84.960938 Top5 97.898438 -2023-02-13 18:20:43,184 - Epoch: [141][ 110/ 135] Loss 0.298722 Top1 84.982244 Top5 97.915483 -2023-02-13 18:20:43,312 - Epoch: [141][ 120/ 135] Loss 0.295362 Top1 85.117188 Top5 97.900391 -2023-02-13 18:20:43,442 - Epoch: [141][ 130/ 135] Loss 0.294140 Top1 85.222356 Top5 97.923678 -2023-02-13 18:20:43,486 - Epoch: [141][ 135/ 135] Loss 0.292999 Top1 85.202996 Top5 97.919035 -2023-02-13 18:20:43,557 - ==> Top1: 85.203 Top5: 97.919 Loss: 0.293 - -2023-02-13 18:20:43,558 - ==> Confusion: -[[ 862 4 5 1 10 2 0 1 2 48 0 5 0 4 6 2 2 1 1 1 10] - [ 2 945 1 1 14 27 3 19 4 0 1 0 2 0 1 0 6 0 3 0 4] - [ 7 3 967 7 3 1 15 16 2 1 1 1 2 7 5 4 2 2 2 4 6] - [ 5 0 24 888 2 3 3 1 2 2 13 1 9 0 26 1 4 8 18 0 6] - [ 11 7 0 2 1000 5 1 1 1 3 0 4 1 4 9 5 5 1 0 3 3] - [ 2 15 1 5 4 971 6 18 2 5 1 9 4 14 0 1 1 1 1 4 5] - [ 3 5 12 2 2 6 1043 5 1 1 2 1 2 1 0 1 0 2 1 5 4] - [ 0 6 11 1 5 25 3 943 0 2 0 6 4 0 0 0 1 0 6 6 5] - [ 12 3 0 2 1 1 1 2 904 44 5 3 1 7 17 1 1 0 4 0 0] - [ 72 1 4 1 11 1 0 4 30 863 0 1 0 10 4 1 0 4 2 0 3] - [ 3 2 5 7 0 2 6 7 9 0 981 1 1 7 5 0 2 1 9 0 3] - [ 1 4 4 0 3 9 0 6 0 1 0 913 29 9 0 2 2 11 3 7 1] - [ 0 0 1 8 1 2 0 1 1 3 0 20 883 2 4 6 4 14 3 0 6] - [ 4 3 0 0 8 5 0 2 7 22 5 6 3 936 6 4 2 6 0 1 4] - [ 5 1 1 13 4 2 0 1 17 7 3 1 2 1 1010 0 3 4 8 0 9] - [ 2 0 6 1 7 1 6 2 0 2 0 6 6 1 1 967 10 12 1 9 6] - [ 3 4 0 1 8 2 0 1 1 0 0 2 4 2 1 11 1004 0 1 5 11] - [ 2 3 1 3 1 1 1 0 1 1 2 8 11 0 1 14 0 993 1 2 5] - [ 5 6 2 4 0 1 0 26 2 1 3 4 2 0 14 1 0 0 1009 3 3] - [ 0 2 0 0 3 5 8 12 1 0 0 12 2 3 0 4 6 3 0 1078 9] - [ 128 212 198 123 151 188 102 200 97 98 167 129 290 281 175 93 208 100 191 229 10074]] - -2023-02-13 18:20:43,559 - ==> Best [Top1: 85.328 Top5: 97.902 Sparsity:0.00 Params: 148928 on epoch: 140] -2023-02-13 18:20:43,559 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:20:43,565 - - -2023-02-13 18:20:43,565 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:20:44,465 - Epoch: [142][ 10/ 1207] Overall Loss 0.212349 Objective Loss 0.212349 LR 0.000250 Time 0.089941 -2023-02-13 18:20:44,658 - Epoch: [142][ 20/ 1207] Overall Loss 0.202010 Objective Loss 0.202010 LR 0.000250 Time 0.054576 -2023-02-13 18:20:44,847 - Epoch: [142][ 30/ 1207] Overall Loss 0.198027 Objective Loss 0.198027 LR 0.000250 Time 0.042683 -2023-02-13 18:20:45,036 - Epoch: [142][ 40/ 1207] Overall Loss 0.201151 Objective Loss 0.201151 LR 0.000250 Time 0.036729 -2023-02-13 18:20:45,225 - Epoch: [142][ 50/ 1207] Overall Loss 0.206759 Objective Loss 0.206759 LR 0.000250 Time 0.033157 -2023-02-13 18:20:45,414 - Epoch: [142][ 60/ 1207] Overall Loss 0.210889 Objective Loss 0.210889 LR 0.000250 Time 0.030773 -2023-02-13 18:20:45,603 - Epoch: [142][ 70/ 1207] Overall Loss 0.209972 Objective Loss 0.209972 LR 0.000250 Time 0.029067 -2023-02-13 18:20:45,792 - Epoch: [142][ 80/ 1207] Overall Loss 0.211732 Objective Loss 0.211732 LR 0.000250 Time 0.027789 -2023-02-13 18:20:45,982 - Epoch: [142][ 90/ 1207] Overall Loss 0.210021 Objective Loss 0.210021 LR 0.000250 Time 0.026808 -2023-02-13 18:20:46,171 - Epoch: [142][ 100/ 1207] Overall Loss 0.211183 Objective Loss 0.211183 LR 0.000250 Time 0.026015 -2023-02-13 18:20:46,361 - Epoch: [142][ 110/ 1207] Overall Loss 0.211107 Objective Loss 0.211107 LR 0.000250 Time 0.025375 -2023-02-13 18:20:46,550 - Epoch: [142][ 120/ 1207] Overall Loss 0.209417 Objective Loss 0.209417 LR 0.000250 Time 0.024834 -2023-02-13 18:20:46,740 - Epoch: [142][ 130/ 1207] Overall Loss 0.209267 Objective Loss 0.209267 LR 0.000250 Time 0.024378 -2023-02-13 18:20:46,930 - Epoch: [142][ 140/ 1207] Overall Loss 0.208541 Objective Loss 0.208541 LR 0.000250 Time 0.023994 -2023-02-13 18:20:47,119 - Epoch: [142][ 150/ 1207] Overall Loss 0.207111 Objective Loss 0.207111 LR 0.000250 Time 0.023653 -2023-02-13 18:20:47,310 - Epoch: [142][ 160/ 1207] Overall Loss 0.206021 Objective Loss 0.206021 LR 0.000250 Time 0.023363 -2023-02-13 18:20:47,499 - Epoch: [142][ 170/ 1207] Overall Loss 0.206217 Objective Loss 0.206217 LR 0.000250 Time 0.023098 -2023-02-13 18:20:47,688 - Epoch: [142][ 180/ 1207] Overall Loss 0.206437 Objective Loss 0.206437 LR 0.000250 Time 0.022864 -2023-02-13 18:20:47,877 - Epoch: [142][ 190/ 1207] Overall Loss 0.205649 Objective Loss 0.205649 LR 0.000250 Time 0.022654 -2023-02-13 18:20:48,066 - Epoch: [142][ 200/ 1207] Overall Loss 0.204840 Objective Loss 0.204840 LR 0.000250 Time 0.022466 -2023-02-13 18:20:48,255 - Epoch: [142][ 210/ 1207] Overall Loss 0.204170 Objective Loss 0.204170 LR 0.000250 Time 0.022295 -2023-02-13 18:20:48,445 - Epoch: [142][ 220/ 1207] Overall Loss 0.203904 Objective Loss 0.203904 LR 0.000250 Time 0.022141 -2023-02-13 18:20:48,634 - Epoch: [142][ 230/ 1207] Overall Loss 0.203353 Objective Loss 0.203353 LR 0.000250 Time 0.022000 -2023-02-13 18:20:48,823 - Epoch: [142][ 240/ 1207] Overall Loss 0.203211 Objective Loss 0.203211 LR 0.000250 Time 0.021870 -2023-02-13 18:20:49,012 - Epoch: [142][ 250/ 1207] Overall Loss 0.203668 Objective Loss 0.203668 LR 0.000250 Time 0.021748 -2023-02-13 18:20:49,202 - Epoch: [142][ 260/ 1207] Overall Loss 0.205762 Objective Loss 0.205762 LR 0.000250 Time 0.021639 -2023-02-13 18:20:49,391 - Epoch: [142][ 270/ 1207] Overall Loss 0.205621 Objective Loss 0.205621 LR 0.000250 Time 0.021538 -2023-02-13 18:20:49,580 - Epoch: [142][ 280/ 1207] Overall Loss 0.205315 Objective Loss 0.205315 LR 0.000250 Time 0.021441 -2023-02-13 18:20:49,768 - Epoch: [142][ 290/ 1207] Overall Loss 0.205578 Objective Loss 0.205578 LR 0.000250 Time 0.021351 -2023-02-13 18:20:49,958 - Epoch: [142][ 300/ 1207] Overall Loss 0.205295 Objective Loss 0.205295 LR 0.000250 Time 0.021269 -2023-02-13 18:20:50,147 - Epoch: [142][ 310/ 1207] Overall Loss 0.205086 Objective Loss 0.205086 LR 0.000250 Time 0.021191 -2023-02-13 18:20:50,336 - Epoch: [142][ 320/ 1207] Overall Loss 0.204846 Objective Loss 0.204846 LR 0.000250 Time 0.021120 -2023-02-13 18:20:50,525 - Epoch: [142][ 330/ 1207] Overall Loss 0.205319 Objective Loss 0.205319 LR 0.000250 Time 0.021051 -2023-02-13 18:20:50,713 - Epoch: [142][ 340/ 1207] Overall Loss 0.205023 Objective Loss 0.205023 LR 0.000250 Time 0.020985 -2023-02-13 18:20:50,903 - Epoch: [142][ 350/ 1207] Overall Loss 0.204869 Objective Loss 0.204869 LR 0.000250 Time 0.020926 -2023-02-13 18:20:51,093 - Epoch: [142][ 360/ 1207] Overall Loss 0.204925 Objective Loss 0.204925 LR 0.000250 Time 0.020871 -2023-02-13 18:20:51,282 - Epoch: [142][ 370/ 1207] Overall Loss 0.205480 Objective Loss 0.205480 LR 0.000250 Time 0.020818 -2023-02-13 18:20:51,471 - Epoch: [142][ 380/ 1207] Overall Loss 0.205617 Objective Loss 0.205617 LR 0.000250 Time 0.020766 -2023-02-13 18:20:51,660 - Epoch: [142][ 390/ 1207] Overall Loss 0.205693 Objective Loss 0.205693 LR 0.000250 Time 0.020716 -2023-02-13 18:20:51,848 - Epoch: [142][ 400/ 1207] Overall Loss 0.205865 Objective Loss 0.205865 LR 0.000250 Time 0.020669 -2023-02-13 18:20:52,039 - Epoch: [142][ 410/ 1207] Overall Loss 0.205780 Objective Loss 0.205780 LR 0.000250 Time 0.020628 -2023-02-13 18:20:52,231 - Epoch: [142][ 420/ 1207] Overall Loss 0.205836 Objective Loss 0.205836 LR 0.000250 Time 0.020593 -2023-02-13 18:20:52,422 - Epoch: [142][ 430/ 1207] Overall Loss 0.205485 Objective Loss 0.205485 LR 0.000250 Time 0.020559 -2023-02-13 18:20:52,613 - Epoch: [142][ 440/ 1207] Overall Loss 0.205422 Objective Loss 0.205422 LR 0.000250 Time 0.020525 -2023-02-13 18:20:52,805 - Epoch: [142][ 450/ 1207] Overall Loss 0.205884 Objective Loss 0.205884 LR 0.000250 Time 0.020495 -2023-02-13 18:20:52,996 - Epoch: [142][ 460/ 1207] Overall Loss 0.206348 Objective Loss 0.206348 LR 0.000250 Time 0.020463 -2023-02-13 18:20:53,187 - Epoch: [142][ 470/ 1207] Overall Loss 0.206016 Objective Loss 0.206016 LR 0.000250 Time 0.020433 -2023-02-13 18:20:53,376 - Epoch: [142][ 480/ 1207] Overall Loss 0.205911 Objective Loss 0.205911 LR 0.000250 Time 0.020401 -2023-02-13 18:20:53,566 - Epoch: [142][ 490/ 1207] Overall Loss 0.205505 Objective Loss 0.205505 LR 0.000250 Time 0.020371 -2023-02-13 18:20:53,755 - Epoch: [142][ 500/ 1207] Overall Loss 0.205215 Objective Loss 0.205215 LR 0.000250 Time 0.020341 -2023-02-13 18:20:53,944 - Epoch: [142][ 510/ 1207] Overall Loss 0.205801 Objective Loss 0.205801 LR 0.000250 Time 0.020311 -2023-02-13 18:20:54,134 - Epoch: [142][ 520/ 1207] Overall Loss 0.206149 Objective Loss 0.206149 LR 0.000250 Time 0.020286 -2023-02-13 18:20:54,324 - Epoch: [142][ 530/ 1207] Overall Loss 0.205939 Objective Loss 0.205939 LR 0.000250 Time 0.020261 -2023-02-13 18:20:54,514 - Epoch: [142][ 540/ 1207] Overall Loss 0.206017 Objective Loss 0.206017 LR 0.000250 Time 0.020236 -2023-02-13 18:20:54,702 - Epoch: [142][ 550/ 1207] Overall Loss 0.206081 Objective Loss 0.206081 LR 0.000250 Time 0.020210 -2023-02-13 18:20:54,891 - Epoch: [142][ 560/ 1207] Overall Loss 0.206394 Objective Loss 0.206394 LR 0.000250 Time 0.020186 -2023-02-13 18:20:55,081 - Epoch: [142][ 570/ 1207] Overall Loss 0.206499 Objective Loss 0.206499 LR 0.000250 Time 0.020165 -2023-02-13 18:20:55,272 - Epoch: [142][ 580/ 1207] Overall Loss 0.206117 Objective Loss 0.206117 LR 0.000250 Time 0.020145 -2023-02-13 18:20:55,462 - Epoch: [142][ 590/ 1207] Overall Loss 0.205668 Objective Loss 0.205668 LR 0.000250 Time 0.020125 -2023-02-13 18:20:55,652 - Epoch: [142][ 600/ 1207] Overall Loss 0.205669 Objective Loss 0.205669 LR 0.000250 Time 0.020106 -2023-02-13 18:20:55,842 - Epoch: [142][ 610/ 1207] Overall Loss 0.205731 Objective Loss 0.205731 LR 0.000250 Time 0.020087 -2023-02-13 18:20:56,034 - Epoch: [142][ 620/ 1207] Overall Loss 0.205591 Objective Loss 0.205591 LR 0.000250 Time 0.020072 -2023-02-13 18:20:56,225 - Epoch: [142][ 630/ 1207] Overall Loss 0.205392 Objective Loss 0.205392 LR 0.000250 Time 0.020057 -2023-02-13 18:20:56,415 - Epoch: [142][ 640/ 1207] Overall Loss 0.205670 Objective Loss 0.205670 LR 0.000250 Time 0.020039 -2023-02-13 18:20:56,605 - Epoch: [142][ 650/ 1207] Overall Loss 0.206263 Objective Loss 0.206263 LR 0.000250 Time 0.020023 -2023-02-13 18:20:56,796 - Epoch: [142][ 660/ 1207] Overall Loss 0.206407 Objective Loss 0.206407 LR 0.000250 Time 0.020008 -2023-02-13 18:20:56,986 - Epoch: [142][ 670/ 1207] Overall Loss 0.206413 Objective Loss 0.206413 LR 0.000250 Time 0.019993 -2023-02-13 18:20:57,176 - Epoch: [142][ 680/ 1207] Overall Loss 0.206279 Objective Loss 0.206279 LR 0.000250 Time 0.019978 -2023-02-13 18:20:57,368 - Epoch: [142][ 690/ 1207] Overall Loss 0.206460 Objective Loss 0.206460 LR 0.000250 Time 0.019965 -2023-02-13 18:20:57,558 - Epoch: [142][ 700/ 1207] Overall Loss 0.206653 Objective Loss 0.206653 LR 0.000250 Time 0.019951 -2023-02-13 18:20:57,748 - Epoch: [142][ 710/ 1207] Overall Loss 0.207228 Objective Loss 0.207228 LR 0.000250 Time 0.019938 -2023-02-13 18:20:57,939 - Epoch: [142][ 720/ 1207] Overall Loss 0.207206 Objective Loss 0.207206 LR 0.000250 Time 0.019924 -2023-02-13 18:20:58,129 - Epoch: [142][ 730/ 1207] Overall Loss 0.207495 Objective Loss 0.207495 LR 0.000250 Time 0.019912 -2023-02-13 18:20:58,320 - Epoch: [142][ 740/ 1207] Overall Loss 0.207544 Objective Loss 0.207544 LR 0.000250 Time 0.019901 -2023-02-13 18:20:58,510 - Epoch: [142][ 750/ 1207] Overall Loss 0.207767 Objective Loss 0.207767 LR 0.000250 Time 0.019888 -2023-02-13 18:20:58,700 - Epoch: [142][ 760/ 1207] Overall Loss 0.207863 Objective Loss 0.207863 LR 0.000250 Time 0.019876 -2023-02-13 18:20:58,890 - Epoch: [142][ 770/ 1207] Overall Loss 0.207811 Objective Loss 0.207811 LR 0.000250 Time 0.019864 -2023-02-13 18:20:59,081 - Epoch: [142][ 780/ 1207] Overall Loss 0.208028 Objective Loss 0.208028 LR 0.000250 Time 0.019853 -2023-02-13 18:20:59,271 - Epoch: [142][ 790/ 1207] Overall Loss 0.207941 Objective Loss 0.207941 LR 0.000250 Time 0.019842 -2023-02-13 18:20:59,462 - Epoch: [142][ 800/ 1207] Overall Loss 0.208175 Objective Loss 0.208175 LR 0.000250 Time 0.019832 -2023-02-13 18:20:59,652 - Epoch: [142][ 810/ 1207] Overall Loss 0.208261 Objective Loss 0.208261 LR 0.000250 Time 0.019821 -2023-02-13 18:20:59,842 - Epoch: [142][ 820/ 1207] Overall Loss 0.208345 Objective Loss 0.208345 LR 0.000250 Time 0.019811 -2023-02-13 18:21:00,032 - Epoch: [142][ 830/ 1207] Overall Loss 0.208499 Objective Loss 0.208499 LR 0.000250 Time 0.019801 -2023-02-13 18:21:00,223 - Epoch: [142][ 840/ 1207] Overall Loss 0.208610 Objective Loss 0.208610 LR 0.000250 Time 0.019793 -2023-02-13 18:21:00,415 - Epoch: [142][ 850/ 1207] Overall Loss 0.208435 Objective Loss 0.208435 LR 0.000250 Time 0.019785 -2023-02-13 18:21:00,605 - Epoch: [142][ 860/ 1207] Overall Loss 0.208164 Objective Loss 0.208164 LR 0.000250 Time 0.019775 -2023-02-13 18:21:00,795 - Epoch: [142][ 870/ 1207] Overall Loss 0.207837 Objective Loss 0.207837 LR 0.000250 Time 0.019765 -2023-02-13 18:21:00,986 - Epoch: [142][ 880/ 1207] Overall Loss 0.207928 Objective Loss 0.207928 LR 0.000250 Time 0.019758 -2023-02-13 18:21:01,177 - Epoch: [142][ 890/ 1207] Overall Loss 0.207745 Objective Loss 0.207745 LR 0.000250 Time 0.019750 -2023-02-13 18:21:01,369 - Epoch: [142][ 900/ 1207] Overall Loss 0.207762 Objective Loss 0.207762 LR 0.000250 Time 0.019743 -2023-02-13 18:21:01,559 - Epoch: [142][ 910/ 1207] Overall Loss 0.207981 Objective Loss 0.207981 LR 0.000250 Time 0.019735 -2023-02-13 18:21:01,750 - Epoch: [142][ 920/ 1207] Overall Loss 0.207946 Objective Loss 0.207946 LR 0.000250 Time 0.019727 -2023-02-13 18:21:01,941 - Epoch: [142][ 930/ 1207] Overall Loss 0.207949 Objective Loss 0.207949 LR 0.000250 Time 0.019720 -2023-02-13 18:21:02,132 - Epoch: [142][ 940/ 1207] Overall Loss 0.207928 Objective Loss 0.207928 LR 0.000250 Time 0.019713 -2023-02-13 18:21:02,323 - Epoch: [142][ 950/ 1207] Overall Loss 0.207865 Objective Loss 0.207865 LR 0.000250 Time 0.019706 -2023-02-13 18:21:02,513 - Epoch: [142][ 960/ 1207] Overall Loss 0.207892 Objective Loss 0.207892 LR 0.000250 Time 0.019699 -2023-02-13 18:21:02,704 - Epoch: [142][ 970/ 1207] Overall Loss 0.207711 Objective Loss 0.207711 LR 0.000250 Time 0.019692 -2023-02-13 18:21:02,894 - Epoch: [142][ 980/ 1207] Overall Loss 0.207658 Objective Loss 0.207658 LR 0.000250 Time 0.019685 -2023-02-13 18:21:03,085 - Epoch: [142][ 990/ 1207] Overall Loss 0.207502 Objective Loss 0.207502 LR 0.000250 Time 0.019678 -2023-02-13 18:21:03,276 - Epoch: [142][ 1000/ 1207] Overall Loss 0.207605 Objective Loss 0.207605 LR 0.000250 Time 0.019672 -2023-02-13 18:21:03,466 - Epoch: [142][ 1010/ 1207] Overall Loss 0.207405 Objective Loss 0.207405 LR 0.000250 Time 0.019665 -2023-02-13 18:21:03,656 - Epoch: [142][ 1020/ 1207] Overall Loss 0.207460 Objective Loss 0.207460 LR 0.000250 Time 0.019658 -2023-02-13 18:21:03,846 - Epoch: [142][ 1030/ 1207] Overall Loss 0.207518 Objective Loss 0.207518 LR 0.000250 Time 0.019652 -2023-02-13 18:21:04,037 - Epoch: [142][ 1040/ 1207] Overall Loss 0.207587 Objective Loss 0.207587 LR 0.000250 Time 0.019646 -2023-02-13 18:21:04,229 - Epoch: [142][ 1050/ 1207] Overall Loss 0.207529 Objective Loss 0.207529 LR 0.000250 Time 0.019641 -2023-02-13 18:21:04,419 - Epoch: [142][ 1060/ 1207] Overall Loss 0.207604 Objective Loss 0.207604 LR 0.000250 Time 0.019635 -2023-02-13 18:21:04,610 - Epoch: [142][ 1070/ 1207] Overall Loss 0.207591 Objective Loss 0.207591 LR 0.000250 Time 0.019630 -2023-02-13 18:21:04,800 - Epoch: [142][ 1080/ 1207] Overall Loss 0.207702 Objective Loss 0.207702 LR 0.000250 Time 0.019624 -2023-02-13 18:21:04,991 - Epoch: [142][ 1090/ 1207] Overall Loss 0.208052 Objective Loss 0.208052 LR 0.000250 Time 0.019618 -2023-02-13 18:21:05,182 - Epoch: [142][ 1100/ 1207] Overall Loss 0.208191 Objective Loss 0.208191 LR 0.000250 Time 0.019613 -2023-02-13 18:21:05,373 - Epoch: [142][ 1110/ 1207] Overall Loss 0.208390 Objective Loss 0.208390 LR 0.000250 Time 0.019608 -2023-02-13 18:21:05,563 - Epoch: [142][ 1120/ 1207] Overall Loss 0.208519 Objective Loss 0.208519 LR 0.000250 Time 0.019602 -2023-02-13 18:21:05,753 - Epoch: [142][ 1130/ 1207] Overall Loss 0.208590 Objective Loss 0.208590 LR 0.000250 Time 0.019597 -2023-02-13 18:21:05,944 - Epoch: [142][ 1140/ 1207] Overall Loss 0.208855 Objective Loss 0.208855 LR 0.000250 Time 0.019592 -2023-02-13 18:21:06,135 - Epoch: [142][ 1150/ 1207] Overall Loss 0.208847 Objective Loss 0.208847 LR 0.000250 Time 0.019588 -2023-02-13 18:21:06,327 - Epoch: [142][ 1160/ 1207] Overall Loss 0.208802 Objective Loss 0.208802 LR 0.000250 Time 0.019583 -2023-02-13 18:21:06,517 - Epoch: [142][ 1170/ 1207] Overall Loss 0.208645 Objective Loss 0.208645 LR 0.000250 Time 0.019579 -2023-02-13 18:21:06,707 - Epoch: [142][ 1180/ 1207] Overall Loss 0.208512 Objective Loss 0.208512 LR 0.000250 Time 0.019573 -2023-02-13 18:21:06,897 - Epoch: [142][ 1190/ 1207] Overall Loss 0.208464 Objective Loss 0.208464 LR 0.000250 Time 0.019568 -2023-02-13 18:21:07,139 - Epoch: [142][ 1200/ 1207] Overall Loss 0.208737 Objective Loss 0.208737 LR 0.000250 Time 0.019606 -2023-02-13 18:21:07,256 - Epoch: [142][ 1207/ 1207] Overall Loss 0.208809 Objective Loss 0.208809 Top1 85.975610 Top5 96.951220 LR 0.000250 Time 0.019589 -2023-02-13 18:21:07,339 - --- validate (epoch=142)----------- -2023-02-13 18:21:07,339 - 34311 samples (256 per mini-batch) -2023-02-13 18:21:07,744 - Epoch: [142][ 10/ 135] Loss 0.331696 Top1 84.218750 Top5 97.773438 -2023-02-13 18:21:07,876 - Epoch: [142][ 20/ 135] Loss 0.309350 Top1 84.531250 Top5 97.773438 -2023-02-13 18:21:08,005 - Epoch: [142][ 30/ 135] Loss 0.308794 Top1 84.648438 Top5 97.630208 -2023-02-13 18:21:08,134 - Epoch: [142][ 40/ 135] Loss 0.304734 Top1 84.541016 Top5 97.744141 -2023-02-13 18:21:08,260 - Epoch: [142][ 50/ 135] Loss 0.303025 Top1 84.445312 Top5 97.781250 -2023-02-13 18:21:08,388 - Epoch: [142][ 60/ 135] Loss 0.298336 Top1 84.641927 Top5 97.766927 -2023-02-13 18:21:08,517 - Epoch: [142][ 70/ 135] Loss 0.297279 Top1 84.531250 Top5 97.739955 -2023-02-13 18:21:08,644 - Epoch: [142][ 80/ 135] Loss 0.295095 Top1 84.536133 Top5 97.758789 -2023-02-13 18:21:08,772 - Epoch: [142][ 90/ 135] Loss 0.302806 Top1 84.392361 Top5 97.756076 -2023-02-13 18:21:08,900 - Epoch: [142][ 100/ 135] Loss 0.299247 Top1 84.542969 Top5 97.738281 -2023-02-13 18:21:09,028 - Epoch: [142][ 110/ 135] Loss 0.300138 Top1 84.573864 Top5 97.766335 -2023-02-13 18:21:09,156 - Epoch: [142][ 120/ 135] Loss 0.296259 Top1 84.609375 Top5 97.770182 -2023-02-13 18:21:09,286 - Epoch: [142][ 130/ 135] Loss 0.295306 Top1 84.675481 Top5 97.806490 -2023-02-13 18:21:09,331 - Epoch: [142][ 135/ 135] Loss 0.295068 Top1 84.692956 Top5 97.793710 -2023-02-13 18:21:09,399 - ==> Top1: 84.693 Top5: 97.794 Loss: 0.295 - -2023-02-13 18:21:09,400 - ==> Confusion: -[[ 837 6 5 1 11 3 0 2 6 59 0 5 0 4 11 1 3 1 3 1 8] - [ 1 963 1 2 6 15 2 13 2 1 3 1 2 0 2 1 6 0 4 2 6] - [ 8 4 961 9 4 2 13 11 1 1 1 1 4 7 6 10 1 3 2 3 6] - [ 4 0 14 907 3 3 0 2 2 3 8 0 7 0 22 1 6 6 21 2 5] - [ 7 9 0 1 995 10 0 1 1 3 0 5 3 4 11 5 8 1 0 1 1] - [ 3 24 1 5 4 953 6 17 2 3 1 11 7 15 1 2 2 0 1 4 8] - [ 2 4 9 0 0 8 1049 3 1 2 3 0 3 2 0 1 0 1 1 7 3] - [ 0 9 9 2 3 21 2 942 0 1 0 7 1 2 0 0 0 1 13 6 5] - [ 17 4 0 2 1 0 1 1 909 32 7 2 1 13 14 1 1 0 2 0 1] - [ 56 4 1 1 7 2 0 3 37 866 1 2 0 18 6 2 0 2 1 0 3] - [ 1 2 4 5 1 4 2 3 18 1 982 2 4 8 2 0 1 1 7 1 2] - [ 2 2 2 0 3 14 0 4 1 3 1 913 22 8 0 6 2 10 3 8 1] - [ 0 0 1 6 2 5 0 1 1 2 0 22 887 2 4 7 4 8 2 1 4] - [ 4 2 1 0 9 7 1 3 9 11 8 3 3 939 8 8 3 2 0 1 2] - [ 2 3 0 11 3 3 1 1 20 5 2 2 2 0 1011 0 4 4 9 2 7] - [ 1 2 5 1 5 2 5 1 1 0 0 6 8 3 2 976 9 7 1 6 5] - [ 1 6 1 3 6 1 0 0 1 1 0 2 4 2 2 12 1005 2 1 3 8] - [ 3 2 1 4 1 1 3 0 1 2 1 8 13 1 0 16 0 983 1 4 6] - [ 0 1 5 6 1 3 0 21 3 0 3 2 5 0 15 2 0 1 1015 3 0] - [ 0 2 1 0 3 5 5 9 1 0 1 10 1 5 0 5 2 2 2 1089 5] - [ 111 248 197 127 124 214 107 179 88 90 176 97 316 276 192 120 273 99 221 302 9877]] - -2023-02-13 18:21:09,401 - ==> Best [Top1: 85.328 Top5: 97.902 Sparsity:0.00 Params: 148928 on epoch: 140] -2023-02-13 18:21:09,401 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:21:09,407 - - -2023-02-13 18:21:09,407 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:21:10,290 - Epoch: [143][ 10/ 1207] Overall Loss 0.206408 Objective Loss 0.206408 LR 0.000250 Time 0.088182 -2023-02-13 18:21:10,482 - Epoch: [143][ 20/ 1207] Overall Loss 0.203507 Objective Loss 0.203507 LR 0.000250 Time 0.053673 -2023-02-13 18:21:10,669 - Epoch: [143][ 30/ 1207] Overall Loss 0.206069 Objective Loss 0.206069 LR 0.000250 Time 0.042015 -2023-02-13 18:21:10,856 - Epoch: [143][ 40/ 1207] Overall Loss 0.207504 Objective Loss 0.207504 LR 0.000250 Time 0.036177 -2023-02-13 18:21:11,045 - Epoch: [143][ 50/ 1207] Overall Loss 0.203582 Objective Loss 0.203582 LR 0.000250 Time 0.032711 -2023-02-13 18:21:11,231 - Epoch: [143][ 60/ 1207] Overall Loss 0.203281 Objective Loss 0.203281 LR 0.000250 Time 0.030360 -2023-02-13 18:21:11,419 - Epoch: [143][ 70/ 1207] Overall Loss 0.201076 Objective Loss 0.201076 LR 0.000250 Time 0.028701 -2023-02-13 18:21:11,606 - Epoch: [143][ 80/ 1207] Overall Loss 0.199903 Objective Loss 0.199903 LR 0.000250 Time 0.027447 -2023-02-13 18:21:11,793 - Epoch: [143][ 90/ 1207] Overall Loss 0.199176 Objective Loss 0.199176 LR 0.000250 Time 0.026473 -2023-02-13 18:21:11,981 - Epoch: [143][ 100/ 1207] Overall Loss 0.199316 Objective Loss 0.199316 LR 0.000250 Time 0.025698 -2023-02-13 18:21:12,169 - Epoch: [143][ 110/ 1207] Overall Loss 0.199233 Objective Loss 0.199233 LR 0.000250 Time 0.025069 -2023-02-13 18:21:12,357 - Epoch: [143][ 120/ 1207] Overall Loss 0.197895 Objective Loss 0.197895 LR 0.000250 Time 0.024543 -2023-02-13 18:21:12,544 - Epoch: [143][ 130/ 1207] Overall Loss 0.197409 Objective Loss 0.197409 LR 0.000250 Time 0.024094 -2023-02-13 18:21:12,732 - Epoch: [143][ 140/ 1207] Overall Loss 0.197176 Objective Loss 0.197176 LR 0.000250 Time 0.023709 -2023-02-13 18:21:12,919 - Epoch: [143][ 150/ 1207] Overall Loss 0.198681 Objective Loss 0.198681 LR 0.000250 Time 0.023375 -2023-02-13 18:21:13,107 - Epoch: [143][ 160/ 1207] Overall Loss 0.198651 Objective Loss 0.198651 LR 0.000250 Time 0.023085 -2023-02-13 18:21:13,295 - Epoch: [143][ 170/ 1207] Overall Loss 0.198458 Objective Loss 0.198458 LR 0.000250 Time 0.022830 -2023-02-13 18:21:13,482 - Epoch: [143][ 180/ 1207] Overall Loss 0.199851 Objective Loss 0.199851 LR 0.000250 Time 0.022599 -2023-02-13 18:21:13,669 - Epoch: [143][ 190/ 1207] Overall Loss 0.199942 Objective Loss 0.199942 LR 0.000250 Time 0.022393 -2023-02-13 18:21:13,857 - Epoch: [143][ 200/ 1207] Overall Loss 0.200810 Objective Loss 0.200810 LR 0.000250 Time 0.022210 -2023-02-13 18:21:14,043 - Epoch: [143][ 210/ 1207] Overall Loss 0.201477 Objective Loss 0.201477 LR 0.000250 Time 0.022039 -2023-02-13 18:21:14,231 - Epoch: [143][ 220/ 1207] Overall Loss 0.200611 Objective Loss 0.200611 LR 0.000250 Time 0.021887 -2023-02-13 18:21:14,419 - Epoch: [143][ 230/ 1207] Overall Loss 0.201077 Objective Loss 0.201077 LR 0.000250 Time 0.021754 -2023-02-13 18:21:14,606 - Epoch: [143][ 240/ 1207] Overall Loss 0.200911 Objective Loss 0.200911 LR 0.000250 Time 0.021625 -2023-02-13 18:21:14,794 - Epoch: [143][ 250/ 1207] Overall Loss 0.201087 Objective Loss 0.201087 LR 0.000250 Time 0.021508 -2023-02-13 18:21:14,981 - Epoch: [143][ 260/ 1207] Overall Loss 0.201148 Objective Loss 0.201148 LR 0.000250 Time 0.021399 -2023-02-13 18:21:15,168 - Epoch: [143][ 270/ 1207] Overall Loss 0.200252 Objective Loss 0.200252 LR 0.000250 Time 0.021299 -2023-02-13 18:21:15,356 - Epoch: [143][ 280/ 1207] Overall Loss 0.200825 Objective Loss 0.200825 LR 0.000250 Time 0.021207 -2023-02-13 18:21:15,543 - Epoch: [143][ 290/ 1207] Overall Loss 0.200711 Objective Loss 0.200711 LR 0.000250 Time 0.021120 -2023-02-13 18:21:15,730 - Epoch: [143][ 300/ 1207] Overall Loss 0.200760 Objective Loss 0.200760 LR 0.000250 Time 0.021039 -2023-02-13 18:21:15,919 - Epoch: [143][ 310/ 1207] Overall Loss 0.201028 Objective Loss 0.201028 LR 0.000250 Time 0.020967 -2023-02-13 18:21:16,106 - Epoch: [143][ 320/ 1207] Overall Loss 0.201260 Objective Loss 0.201260 LR 0.000250 Time 0.020897 -2023-02-13 18:21:16,295 - Epoch: [143][ 330/ 1207] Overall Loss 0.202541 Objective Loss 0.202541 LR 0.000250 Time 0.020833 -2023-02-13 18:21:16,482 - Epoch: [143][ 340/ 1207] Overall Loss 0.202426 Objective Loss 0.202426 LR 0.000250 Time 0.020771 -2023-02-13 18:21:16,670 - Epoch: [143][ 350/ 1207] Overall Loss 0.201914 Objective Loss 0.201914 LR 0.000250 Time 0.020712 -2023-02-13 18:21:16,857 - Epoch: [143][ 360/ 1207] Overall Loss 0.202204 Objective Loss 0.202204 LR 0.000250 Time 0.020656 -2023-02-13 18:21:17,044 - Epoch: [143][ 370/ 1207] Overall Loss 0.202932 Objective Loss 0.202932 LR 0.000250 Time 0.020604 -2023-02-13 18:21:17,232 - Epoch: [143][ 380/ 1207] Overall Loss 0.202662 Objective Loss 0.202662 LR 0.000250 Time 0.020554 -2023-02-13 18:21:17,420 - Epoch: [143][ 390/ 1207] Overall Loss 0.202838 Objective Loss 0.202838 LR 0.000250 Time 0.020508 -2023-02-13 18:21:17,608 - Epoch: [143][ 400/ 1207] Overall Loss 0.202702 Objective Loss 0.202702 LR 0.000250 Time 0.020463 -2023-02-13 18:21:17,795 - Epoch: [143][ 410/ 1207] Overall Loss 0.203330 Objective Loss 0.203330 LR 0.000250 Time 0.020421 -2023-02-13 18:21:17,982 - Epoch: [143][ 420/ 1207] Overall Loss 0.203767 Objective Loss 0.203767 LR 0.000250 Time 0.020379 -2023-02-13 18:21:18,170 - Epoch: [143][ 430/ 1207] Overall Loss 0.203174 Objective Loss 0.203174 LR 0.000250 Time 0.020342 -2023-02-13 18:21:18,358 - Epoch: [143][ 440/ 1207] Overall Loss 0.203632 Objective Loss 0.203632 LR 0.000250 Time 0.020306 -2023-02-13 18:21:18,546 - Epoch: [143][ 450/ 1207] Overall Loss 0.204160 Objective Loss 0.204160 LR 0.000250 Time 0.020271 -2023-02-13 18:21:18,734 - Epoch: [143][ 460/ 1207] Overall Loss 0.203456 Objective Loss 0.203456 LR 0.000250 Time 0.020237 -2023-02-13 18:21:18,921 - Epoch: [143][ 470/ 1207] Overall Loss 0.203576 Objective Loss 0.203576 LR 0.000250 Time 0.020205 -2023-02-13 18:21:19,109 - Epoch: [143][ 480/ 1207] Overall Loss 0.203612 Objective Loss 0.203612 LR 0.000250 Time 0.020174 -2023-02-13 18:21:19,297 - Epoch: [143][ 490/ 1207] Overall Loss 0.203871 Objective Loss 0.203871 LR 0.000250 Time 0.020146 -2023-02-13 18:21:19,485 - Epoch: [143][ 500/ 1207] Overall Loss 0.203963 Objective Loss 0.203963 LR 0.000250 Time 0.020119 -2023-02-13 18:21:19,672 - Epoch: [143][ 510/ 1207] Overall Loss 0.203968 Objective Loss 0.203968 LR 0.000250 Time 0.020090 -2023-02-13 18:21:19,860 - Epoch: [143][ 520/ 1207] Overall Loss 0.203820 Objective Loss 0.203820 LR 0.000250 Time 0.020064 -2023-02-13 18:21:20,047 - Epoch: [143][ 530/ 1207] Overall Loss 0.203934 Objective Loss 0.203934 LR 0.000250 Time 0.020038 -2023-02-13 18:21:20,235 - Epoch: [143][ 540/ 1207] Overall Loss 0.203712 Objective Loss 0.203712 LR 0.000250 Time 0.020013 -2023-02-13 18:21:20,423 - Epoch: [143][ 550/ 1207] Overall Loss 0.203658 Objective Loss 0.203658 LR 0.000250 Time 0.019992 -2023-02-13 18:21:20,611 - Epoch: [143][ 560/ 1207] Overall Loss 0.203960 Objective Loss 0.203960 LR 0.000250 Time 0.019969 -2023-02-13 18:21:20,798 - Epoch: [143][ 570/ 1207] Overall Loss 0.203916 Objective Loss 0.203916 LR 0.000250 Time 0.019947 -2023-02-13 18:21:20,986 - Epoch: [143][ 580/ 1207] Overall Loss 0.204060 Objective Loss 0.204060 LR 0.000250 Time 0.019926 -2023-02-13 18:21:21,173 - Epoch: [143][ 590/ 1207] Overall Loss 0.203977 Objective Loss 0.203977 LR 0.000250 Time 0.019905 -2023-02-13 18:21:21,361 - Epoch: [143][ 600/ 1207] Overall Loss 0.204116 Objective Loss 0.204116 LR 0.000250 Time 0.019885 -2023-02-13 18:21:21,549 - Epoch: [143][ 610/ 1207] Overall Loss 0.204184 Objective Loss 0.204184 LR 0.000250 Time 0.019867 -2023-02-13 18:21:21,739 - Epoch: [143][ 620/ 1207] Overall Loss 0.204159 Objective Loss 0.204159 LR 0.000250 Time 0.019853 -2023-02-13 18:21:21,933 - Epoch: [143][ 630/ 1207] Overall Loss 0.204031 Objective Loss 0.204031 LR 0.000250 Time 0.019845 -2023-02-13 18:21:22,127 - Epoch: [143][ 640/ 1207] Overall Loss 0.204315 Objective Loss 0.204315 LR 0.000250 Time 0.019837 -2023-02-13 18:21:22,319 - Epoch: [143][ 650/ 1207] Overall Loss 0.204536 Objective Loss 0.204536 LR 0.000250 Time 0.019828 -2023-02-13 18:21:22,514 - Epoch: [143][ 660/ 1207] Overall Loss 0.204411 Objective Loss 0.204411 LR 0.000250 Time 0.019821 -2023-02-13 18:21:22,706 - Epoch: [143][ 670/ 1207] Overall Loss 0.204261 Objective Loss 0.204261 LR 0.000250 Time 0.019811 -2023-02-13 18:21:22,900 - Epoch: [143][ 680/ 1207] Overall Loss 0.204665 Objective Loss 0.204665 LR 0.000250 Time 0.019805 -2023-02-13 18:21:23,092 - Epoch: [143][ 690/ 1207] Overall Loss 0.204898 Objective Loss 0.204898 LR 0.000250 Time 0.019795 -2023-02-13 18:21:23,285 - Epoch: [143][ 700/ 1207] Overall Loss 0.205009 Objective Loss 0.205009 LR 0.000250 Time 0.019788 -2023-02-13 18:21:23,478 - Epoch: [143][ 710/ 1207] Overall Loss 0.204929 Objective Loss 0.204929 LR 0.000250 Time 0.019781 -2023-02-13 18:21:23,672 - Epoch: [143][ 720/ 1207] Overall Loss 0.204913 Objective Loss 0.204913 LR 0.000250 Time 0.019775 -2023-02-13 18:21:23,864 - Epoch: [143][ 730/ 1207] Overall Loss 0.205163 Objective Loss 0.205163 LR 0.000250 Time 0.019766 -2023-02-13 18:21:24,057 - Epoch: [143][ 740/ 1207] Overall Loss 0.205066 Objective Loss 0.205066 LR 0.000250 Time 0.019760 -2023-02-13 18:21:24,249 - Epoch: [143][ 750/ 1207] Overall Loss 0.205123 Objective Loss 0.205123 LR 0.000250 Time 0.019752 -2023-02-13 18:21:24,443 - Epoch: [143][ 760/ 1207] Overall Loss 0.205293 Objective Loss 0.205293 LR 0.000250 Time 0.019747 -2023-02-13 18:21:24,636 - Epoch: [143][ 770/ 1207] Overall Loss 0.205205 Objective Loss 0.205205 LR 0.000250 Time 0.019740 -2023-02-13 18:21:24,829 - Epoch: [143][ 780/ 1207] Overall Loss 0.204978 Objective Loss 0.204978 LR 0.000250 Time 0.019734 -2023-02-13 18:21:25,021 - Epoch: [143][ 790/ 1207] Overall Loss 0.205173 Objective Loss 0.205173 LR 0.000250 Time 0.019726 -2023-02-13 18:21:25,214 - Epoch: [143][ 800/ 1207] Overall Loss 0.204976 Objective Loss 0.204976 LR 0.000250 Time 0.019721 -2023-02-13 18:21:25,407 - Epoch: [143][ 810/ 1207] Overall Loss 0.205101 Objective Loss 0.205101 LR 0.000250 Time 0.019716 -2023-02-13 18:21:25,601 - Epoch: [143][ 820/ 1207] Overall Loss 0.205171 Objective Loss 0.205171 LR 0.000250 Time 0.019711 -2023-02-13 18:21:25,793 - Epoch: [143][ 830/ 1207] Overall Loss 0.205384 Objective Loss 0.205384 LR 0.000250 Time 0.019704 -2023-02-13 18:21:25,987 - Epoch: [143][ 840/ 1207] Overall Loss 0.205406 Objective Loss 0.205406 LR 0.000250 Time 0.019701 -2023-02-13 18:21:26,180 - Epoch: [143][ 850/ 1207] Overall Loss 0.205426 Objective Loss 0.205426 LR 0.000250 Time 0.019695 -2023-02-13 18:21:26,374 - Epoch: [143][ 860/ 1207] Overall Loss 0.205596 Objective Loss 0.205596 LR 0.000250 Time 0.019691 -2023-02-13 18:21:26,566 - Epoch: [143][ 870/ 1207] Overall Loss 0.205714 Objective Loss 0.205714 LR 0.000250 Time 0.019685 -2023-02-13 18:21:26,760 - Epoch: [143][ 880/ 1207] Overall Loss 0.205650 Objective Loss 0.205650 LR 0.000250 Time 0.019681 -2023-02-13 18:21:26,952 - Epoch: [143][ 890/ 1207] Overall Loss 0.205669 Objective Loss 0.205669 LR 0.000250 Time 0.019676 -2023-02-13 18:21:27,146 - Epoch: [143][ 900/ 1207] Overall Loss 0.205866 Objective Loss 0.205866 LR 0.000250 Time 0.019673 -2023-02-13 18:21:27,340 - Epoch: [143][ 910/ 1207] Overall Loss 0.206019 Objective Loss 0.206019 LR 0.000250 Time 0.019668 -2023-02-13 18:21:27,534 - Epoch: [143][ 920/ 1207] Overall Loss 0.206139 Objective Loss 0.206139 LR 0.000250 Time 0.019666 -2023-02-13 18:21:27,727 - Epoch: [143][ 930/ 1207] Overall Loss 0.206207 Objective Loss 0.206207 LR 0.000250 Time 0.019661 -2023-02-13 18:21:27,920 - Epoch: [143][ 940/ 1207] Overall Loss 0.206262 Objective Loss 0.206262 LR 0.000250 Time 0.019658 -2023-02-13 18:21:28,112 - Epoch: [143][ 950/ 1207] Overall Loss 0.205925 Objective Loss 0.205925 LR 0.000250 Time 0.019652 -2023-02-13 18:21:28,307 - Epoch: [143][ 960/ 1207] Overall Loss 0.205852 Objective Loss 0.205852 LR 0.000250 Time 0.019650 -2023-02-13 18:21:28,501 - Epoch: [143][ 970/ 1207] Overall Loss 0.205903 Objective Loss 0.205903 LR 0.000250 Time 0.019647 -2023-02-13 18:21:28,694 - Epoch: [143][ 980/ 1207] Overall Loss 0.205753 Objective Loss 0.205753 LR 0.000250 Time 0.019643 -2023-02-13 18:21:28,887 - Epoch: [143][ 990/ 1207] Overall Loss 0.205866 Objective Loss 0.205866 LR 0.000250 Time 0.019639 -2023-02-13 18:21:29,082 - Epoch: [143][ 1000/ 1207] Overall Loss 0.205884 Objective Loss 0.205884 LR 0.000250 Time 0.019637 -2023-02-13 18:21:29,274 - Epoch: [143][ 1010/ 1207] Overall Loss 0.206145 Objective Loss 0.206145 LR 0.000250 Time 0.019633 -2023-02-13 18:21:29,468 - Epoch: [143][ 1020/ 1207] Overall Loss 0.206178 Objective Loss 0.206178 LR 0.000250 Time 0.019631 -2023-02-13 18:21:29,660 - Epoch: [143][ 1030/ 1207] Overall Loss 0.206047 Objective Loss 0.206047 LR 0.000250 Time 0.019626 -2023-02-13 18:21:29,854 - Epoch: [143][ 1040/ 1207] Overall Loss 0.205822 Objective Loss 0.205822 LR 0.000250 Time 0.019623 -2023-02-13 18:21:30,047 - Epoch: [143][ 1050/ 1207] Overall Loss 0.205966 Objective Loss 0.205966 LR 0.000250 Time 0.019620 -2023-02-13 18:21:30,241 - Epoch: [143][ 1060/ 1207] Overall Loss 0.206293 Objective Loss 0.206293 LR 0.000250 Time 0.019617 -2023-02-13 18:21:30,434 - Epoch: [143][ 1070/ 1207] Overall Loss 0.206218 Objective Loss 0.206218 LR 0.000250 Time 0.019614 -2023-02-13 18:21:30,627 - Epoch: [143][ 1080/ 1207] Overall Loss 0.206436 Objective Loss 0.206436 LR 0.000250 Time 0.019611 -2023-02-13 18:21:30,820 - Epoch: [143][ 1090/ 1207] Overall Loss 0.206593 Objective Loss 0.206593 LR 0.000250 Time 0.019607 -2023-02-13 18:21:31,015 - Epoch: [143][ 1100/ 1207] Overall Loss 0.206381 Objective Loss 0.206381 LR 0.000250 Time 0.019606 -2023-02-13 18:21:31,207 - Epoch: [143][ 1110/ 1207] Overall Loss 0.206156 Objective Loss 0.206156 LR 0.000250 Time 0.019602 -2023-02-13 18:21:31,402 - Epoch: [143][ 1120/ 1207] Overall Loss 0.206178 Objective Loss 0.206178 LR 0.000250 Time 0.019601 -2023-02-13 18:21:31,594 - Epoch: [143][ 1130/ 1207] Overall Loss 0.206311 Objective Loss 0.206311 LR 0.000250 Time 0.019597 -2023-02-13 18:21:31,788 - Epoch: [143][ 1140/ 1207] Overall Loss 0.206198 Objective Loss 0.206198 LR 0.000250 Time 0.019595 -2023-02-13 18:21:31,983 - Epoch: [143][ 1150/ 1207] Overall Loss 0.206366 Objective Loss 0.206366 LR 0.000250 Time 0.019594 -2023-02-13 18:21:32,179 - Epoch: [143][ 1160/ 1207] Overall Loss 0.206390 Objective Loss 0.206390 LR 0.000250 Time 0.019594 -2023-02-13 18:21:32,376 - Epoch: [143][ 1170/ 1207] Overall Loss 0.206355 Objective Loss 0.206355 LR 0.000250 Time 0.019594 -2023-02-13 18:21:32,572 - Epoch: [143][ 1180/ 1207] Overall Loss 0.206359 Objective Loss 0.206359 LR 0.000250 Time 0.019594 -2023-02-13 18:21:32,767 - Epoch: [143][ 1190/ 1207] Overall Loss 0.206319 Objective Loss 0.206319 LR 0.000250 Time 0.019593 -2023-02-13 18:21:33,018 - Epoch: [143][ 1200/ 1207] Overall Loss 0.206330 Objective Loss 0.206330 LR 0.000250 Time 0.019639 -2023-02-13 18:21:33,133 - Epoch: [143][ 1207/ 1207] Overall Loss 0.206071 Objective Loss 0.206071 Top1 89.634146 Top5 98.170732 LR 0.000250 Time 0.019619 -2023-02-13 18:21:33,205 - --- validate (epoch=143)----------- -2023-02-13 18:21:33,205 - 34311 samples (256 per mini-batch) -2023-02-13 18:21:33,610 - Epoch: [143][ 10/ 135] Loss 0.327707 Top1 85.390625 Top5 97.851562 -2023-02-13 18:21:33,746 - Epoch: [143][ 20/ 135] Loss 0.298453 Top1 85.273438 Top5 97.949219 -2023-02-13 18:21:33,879 - Epoch: [143][ 30/ 135] Loss 0.294275 Top1 85.390625 Top5 97.955729 -2023-02-13 18:21:34,004 - Epoch: [143][ 40/ 135] Loss 0.294411 Top1 85.341797 Top5 97.861328 -2023-02-13 18:21:34,133 - Epoch: [143][ 50/ 135] Loss 0.300204 Top1 85.203125 Top5 97.882812 -2023-02-13 18:21:34,261 - Epoch: [143][ 60/ 135] Loss 0.301197 Top1 85.149740 Top5 97.884115 -2023-02-13 18:21:34,389 - Epoch: [143][ 70/ 135] Loss 0.298885 Top1 85.184152 Top5 97.890625 -2023-02-13 18:21:34,517 - Epoch: [143][ 80/ 135] Loss 0.296274 Top1 85.180664 Top5 97.915039 -2023-02-13 18:21:34,645 - Epoch: [143][ 90/ 135] Loss 0.296835 Top1 85.138889 Top5 97.868924 -2023-02-13 18:21:34,773 - Epoch: [143][ 100/ 135] Loss 0.294917 Top1 85.062500 Top5 97.902344 -2023-02-13 18:21:34,901 - Epoch: [143][ 110/ 135] Loss 0.295247 Top1 84.968040 Top5 97.926136 -2023-02-13 18:21:35,029 - Epoch: [143][ 120/ 135] Loss 0.296167 Top1 84.967448 Top5 97.916667 -2023-02-13 18:21:35,161 - Epoch: [143][ 130/ 135] Loss 0.295367 Top1 85.090144 Top5 97.896635 -2023-02-13 18:21:35,207 - Epoch: [143][ 135/ 135] Loss 0.297992 Top1 85.086415 Top5 97.904462 -2023-02-13 18:21:35,275 - ==> Top1: 85.086 Top5: 97.904 Loss: 0.298 - -2023-02-13 18:21:35,276 - ==> Confusion: -[[ 855 3 7 0 12 4 0 0 3 47 0 3 1 4 9 4 2 2 1 2 8] - [ 1 967 1 2 8 21 0 11 2 3 2 0 0 0 0 1 2 2 3 1 6] - [ 6 8 971 8 4 1 11 13 0 1 1 0 5 5 4 3 4 1 4 4 4] - [ 5 2 21 900 3 5 2 2 2 3 10 0 8 0 18 0 5 6 15 2 7] - [ 8 8 0 0 997 8 1 3 2 2 0 7 1 2 10 3 5 0 1 4 4] - [ 1 16 1 4 6 979 4 12 1 3 0 7 1 14 1 3 5 2 1 3 6] - [ 4 6 19 1 1 5 1043 1 0 0 2 1 4 2 0 0 1 2 1 3 3] - [ 1 14 13 3 3 31 0 924 0 2 2 5 2 2 0 0 0 1 10 4 7] - [ 13 2 0 1 2 0 0 1 911 36 8 3 0 13 11 3 0 0 2 1 2] - [ 68 2 4 0 8 3 0 3 32 862 0 0 0 16 3 1 1 3 3 0 3] - [ 2 1 6 7 1 1 1 6 12 1 989 2 1 8 3 0 1 0 8 0 1] - [ 0 2 1 1 4 12 1 7 2 2 1 912 18 11 0 4 7 7 2 9 2] - [ 1 0 2 4 2 3 0 1 3 2 1 28 867 1 2 4 4 23 2 0 9] - [ 3 3 1 0 12 4 0 1 10 11 5 4 3 951 3 6 1 1 0 1 4] - [ 4 4 0 15 4 3 0 1 24 5 3 1 3 3 997 1 3 4 8 1 8] - [ 1 3 6 0 9 1 4 0 1 0 0 7 3 5 2 958 17 13 0 8 8] - [ 2 6 1 3 7 0 0 1 0 0 0 0 2 4 1 11 1010 1 1 2 9] - [ 5 5 1 2 3 3 1 0 0 1 2 6 16 1 2 11 0 987 1 0 4] - [ 4 5 6 9 0 2 0 21 4 0 5 2 3 0 15 1 1 3 1002 2 1] - [ 0 4 0 0 3 9 5 9 1 0 0 9 3 5 0 6 9 5 0 1073 7] - [ 113 277 252 100 146 231 94 157 101 81 167 115 265 336 158 84 242 99 171 206 10039]] - -2023-02-13 18:21:35,278 - ==> Best [Top1: 85.328 Top5: 97.902 Sparsity:0.00 Params: 148928 on epoch: 140] -2023-02-13 18:21:35,278 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:21:35,284 - - -2023-02-13 18:21:35,284 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:21:36,289 - Epoch: [144][ 10/ 1207] Overall Loss 0.199034 Objective Loss 0.199034 LR 0.000250 Time 0.100497 -2023-02-13 18:21:36,494 - Epoch: [144][ 20/ 1207] Overall Loss 0.197917 Objective Loss 0.197917 LR 0.000250 Time 0.060455 -2023-02-13 18:21:36,691 - Epoch: [144][ 30/ 1207] Overall Loss 0.205669 Objective Loss 0.205669 LR 0.000250 Time 0.046870 -2023-02-13 18:21:36,886 - Epoch: [144][ 40/ 1207] Overall Loss 0.201428 Objective Loss 0.201428 LR 0.000250 Time 0.040004 -2023-02-13 18:21:37,083 - Epoch: [144][ 50/ 1207] Overall Loss 0.202228 Objective Loss 0.202228 LR 0.000250 Time 0.035947 -2023-02-13 18:21:37,278 - Epoch: [144][ 60/ 1207] Overall Loss 0.205615 Objective Loss 0.205615 LR 0.000250 Time 0.033194 -2023-02-13 18:21:37,476 - Epoch: [144][ 70/ 1207] Overall Loss 0.205509 Objective Loss 0.205509 LR 0.000250 Time 0.031275 -2023-02-13 18:21:37,671 - Epoch: [144][ 80/ 1207] Overall Loss 0.200860 Objective Loss 0.200860 LR 0.000250 Time 0.029795 -2023-02-13 18:21:37,868 - Epoch: [144][ 90/ 1207] Overall Loss 0.200188 Objective Loss 0.200188 LR 0.000250 Time 0.028669 -2023-02-13 18:21:38,062 - Epoch: [144][ 100/ 1207] Overall Loss 0.201259 Objective Loss 0.201259 LR 0.000250 Time 0.027744 -2023-02-13 18:21:38,259 - Epoch: [144][ 110/ 1207] Overall Loss 0.202702 Objective Loss 0.202702 LR 0.000250 Time 0.027009 -2023-02-13 18:21:38,455 - Epoch: [144][ 120/ 1207] Overall Loss 0.200899 Objective Loss 0.200899 LR 0.000250 Time 0.026385 -2023-02-13 18:21:38,651 - Epoch: [144][ 130/ 1207] Overall Loss 0.202463 Objective Loss 0.202463 LR 0.000250 Time 0.025863 -2023-02-13 18:21:38,845 - Epoch: [144][ 140/ 1207] Overall Loss 0.202890 Objective Loss 0.202890 LR 0.000250 Time 0.025401 -2023-02-13 18:21:39,043 - Epoch: [144][ 150/ 1207] Overall Loss 0.203989 Objective Loss 0.203989 LR 0.000250 Time 0.025019 -2023-02-13 18:21:39,237 - Epoch: [144][ 160/ 1207] Overall Loss 0.202121 Objective Loss 0.202121 LR 0.000250 Time 0.024670 -2023-02-13 18:21:39,435 - Epoch: [144][ 170/ 1207] Overall Loss 0.202746 Objective Loss 0.202746 LR 0.000250 Time 0.024380 -2023-02-13 18:21:39,630 - Epoch: [144][ 180/ 1207] Overall Loss 0.202755 Objective Loss 0.202755 LR 0.000250 Time 0.024109 -2023-02-13 18:21:39,828 - Epoch: [144][ 190/ 1207] Overall Loss 0.202458 Objective Loss 0.202458 LR 0.000250 Time 0.023876 -2023-02-13 18:21:40,022 - Epoch: [144][ 200/ 1207] Overall Loss 0.201972 Objective Loss 0.201972 LR 0.000250 Time 0.023654 -2023-02-13 18:21:40,220 - Epoch: [144][ 210/ 1207] Overall Loss 0.202913 Objective Loss 0.202913 LR 0.000250 Time 0.023465 -2023-02-13 18:21:40,415 - Epoch: [144][ 220/ 1207] Overall Loss 0.202671 Objective Loss 0.202671 LR 0.000250 Time 0.023285 -2023-02-13 18:21:40,612 - Epoch: [144][ 230/ 1207] Overall Loss 0.202752 Objective Loss 0.202752 LR 0.000250 Time 0.023128 -2023-02-13 18:21:40,809 - Epoch: [144][ 240/ 1207] Overall Loss 0.202435 Objective Loss 0.202435 LR 0.000250 Time 0.022981 -2023-02-13 18:21:41,009 - Epoch: [144][ 250/ 1207] Overall Loss 0.203215 Objective Loss 0.203215 LR 0.000250 Time 0.022862 -2023-02-13 18:21:41,205 - Epoch: [144][ 260/ 1207] Overall Loss 0.203333 Objective Loss 0.203333 LR 0.000250 Time 0.022737 -2023-02-13 18:21:41,402 - Epoch: [144][ 270/ 1207] Overall Loss 0.202526 Objective Loss 0.202526 LR 0.000250 Time 0.022622 -2023-02-13 18:21:41,593 - Epoch: [144][ 280/ 1207] Overall Loss 0.202610 Objective Loss 0.202610 LR 0.000250 Time 0.022493 -2023-02-13 18:21:41,783 - Epoch: [144][ 290/ 1207] Overall Loss 0.203096 Objective Loss 0.203096 LR 0.000250 Time 0.022372 -2023-02-13 18:21:41,973 - Epoch: [144][ 300/ 1207] Overall Loss 0.202736 Objective Loss 0.202736 LR 0.000250 Time 0.022260 -2023-02-13 18:21:42,164 - Epoch: [144][ 310/ 1207] Overall Loss 0.203197 Objective Loss 0.203197 LR 0.000250 Time 0.022156 -2023-02-13 18:21:42,355 - Epoch: [144][ 320/ 1207] Overall Loss 0.203484 Objective Loss 0.203484 LR 0.000250 Time 0.022060 -2023-02-13 18:21:42,547 - Epoch: [144][ 330/ 1207] Overall Loss 0.203333 Objective Loss 0.203333 LR 0.000250 Time 0.021971 -2023-02-13 18:21:42,738 - Epoch: [144][ 340/ 1207] Overall Loss 0.203029 Objective Loss 0.203029 LR 0.000250 Time 0.021885 -2023-02-13 18:21:42,929 - Epoch: [144][ 350/ 1207] Overall Loss 0.202812 Objective Loss 0.202812 LR 0.000250 Time 0.021805 -2023-02-13 18:21:43,120 - Epoch: [144][ 360/ 1207] Overall Loss 0.203421 Objective Loss 0.203421 LR 0.000250 Time 0.021727 -2023-02-13 18:21:43,311 - Epoch: [144][ 370/ 1207] Overall Loss 0.203445 Objective Loss 0.203445 LR 0.000250 Time 0.021656 -2023-02-13 18:21:43,502 - Epoch: [144][ 380/ 1207] Overall Loss 0.203044 Objective Loss 0.203044 LR 0.000250 Time 0.021588 -2023-02-13 18:21:43,693 - Epoch: [144][ 390/ 1207] Overall Loss 0.203328 Objective Loss 0.203328 LR 0.000250 Time 0.021523 -2023-02-13 18:21:43,883 - Epoch: [144][ 400/ 1207] Overall Loss 0.203548 Objective Loss 0.203548 LR 0.000250 Time 0.021461 -2023-02-13 18:21:44,075 - Epoch: [144][ 410/ 1207] Overall Loss 0.203881 Objective Loss 0.203881 LR 0.000250 Time 0.021402 -2023-02-13 18:21:44,265 - Epoch: [144][ 420/ 1207] Overall Loss 0.203431 Objective Loss 0.203431 LR 0.000250 Time 0.021346 -2023-02-13 18:21:44,457 - Epoch: [144][ 430/ 1207] Overall Loss 0.203453 Objective Loss 0.203453 LR 0.000250 Time 0.021295 -2023-02-13 18:21:44,648 - Epoch: [144][ 440/ 1207] Overall Loss 0.203788 Objective Loss 0.203788 LR 0.000250 Time 0.021244 -2023-02-13 18:21:44,839 - Epoch: [144][ 450/ 1207] Overall Loss 0.204263 Objective Loss 0.204263 LR 0.000250 Time 0.021196 -2023-02-13 18:21:45,030 - Epoch: [144][ 460/ 1207] Overall Loss 0.204511 Objective Loss 0.204511 LR 0.000250 Time 0.021149 -2023-02-13 18:21:45,222 - Epoch: [144][ 470/ 1207] Overall Loss 0.204514 Objective Loss 0.204514 LR 0.000250 Time 0.021106 -2023-02-13 18:21:45,413 - Epoch: [144][ 480/ 1207] Overall Loss 0.204982 Objective Loss 0.204982 LR 0.000250 Time 0.021063 -2023-02-13 18:21:45,604 - Epoch: [144][ 490/ 1207] Overall Loss 0.204918 Objective Loss 0.204918 LR 0.000250 Time 0.021022 -2023-02-13 18:21:45,794 - Epoch: [144][ 500/ 1207] Overall Loss 0.205012 Objective Loss 0.205012 LR 0.000250 Time 0.020983 -2023-02-13 18:21:45,986 - Epoch: [144][ 510/ 1207] Overall Loss 0.205096 Objective Loss 0.205096 LR 0.000250 Time 0.020947 -2023-02-13 18:21:46,177 - Epoch: [144][ 520/ 1207] Overall Loss 0.205077 Objective Loss 0.205077 LR 0.000250 Time 0.020910 -2023-02-13 18:21:46,368 - Epoch: [144][ 530/ 1207] Overall Loss 0.205542 Objective Loss 0.205542 LR 0.000250 Time 0.020876 -2023-02-13 18:21:46,560 - Epoch: [144][ 540/ 1207] Overall Loss 0.205620 Objective Loss 0.205620 LR 0.000250 Time 0.020844 -2023-02-13 18:21:46,751 - Epoch: [144][ 550/ 1207] Overall Loss 0.205303 Objective Loss 0.205303 LR 0.000250 Time 0.020811 -2023-02-13 18:21:46,942 - Epoch: [144][ 560/ 1207] Overall Loss 0.205030 Objective Loss 0.205030 LR 0.000250 Time 0.020780 -2023-02-13 18:21:47,133 - Epoch: [144][ 570/ 1207] Overall Loss 0.205274 Objective Loss 0.205274 LR 0.000250 Time 0.020751 -2023-02-13 18:21:47,324 - Epoch: [144][ 580/ 1207] Overall Loss 0.205163 Objective Loss 0.205163 LR 0.000250 Time 0.020721 -2023-02-13 18:21:47,516 - Epoch: [144][ 590/ 1207] Overall Loss 0.205239 Objective Loss 0.205239 LR 0.000250 Time 0.020693 -2023-02-13 18:21:47,707 - Epoch: [144][ 600/ 1207] Overall Loss 0.205602 Objective Loss 0.205602 LR 0.000250 Time 0.020667 -2023-02-13 18:21:47,898 - Epoch: [144][ 610/ 1207] Overall Loss 0.205842 Objective Loss 0.205842 LR 0.000250 Time 0.020641 -2023-02-13 18:21:48,089 - Epoch: [144][ 620/ 1207] Overall Loss 0.205946 Objective Loss 0.205946 LR 0.000250 Time 0.020615 -2023-02-13 18:21:48,279 - Epoch: [144][ 630/ 1207] Overall Loss 0.206309 Objective Loss 0.206309 LR 0.000250 Time 0.020590 -2023-02-13 18:21:48,471 - Epoch: [144][ 640/ 1207] Overall Loss 0.206420 Objective Loss 0.206420 LR 0.000250 Time 0.020566 -2023-02-13 18:21:48,662 - Epoch: [144][ 650/ 1207] Overall Loss 0.206443 Objective Loss 0.206443 LR 0.000250 Time 0.020543 -2023-02-13 18:21:48,853 - Epoch: [144][ 660/ 1207] Overall Loss 0.206149 Objective Loss 0.206149 LR 0.000250 Time 0.020521 -2023-02-13 18:21:49,044 - Epoch: [144][ 670/ 1207] Overall Loss 0.206053 Objective Loss 0.206053 LR 0.000250 Time 0.020499 -2023-02-13 18:21:49,235 - Epoch: [144][ 680/ 1207] Overall Loss 0.206324 Objective Loss 0.206324 LR 0.000250 Time 0.020478 -2023-02-13 18:21:49,426 - Epoch: [144][ 690/ 1207] Overall Loss 0.206252 Objective Loss 0.206252 LR 0.000250 Time 0.020458 -2023-02-13 18:21:49,617 - Epoch: [144][ 700/ 1207] Overall Loss 0.206310 Objective Loss 0.206310 LR 0.000250 Time 0.020438 -2023-02-13 18:21:49,808 - Epoch: [144][ 710/ 1207] Overall Loss 0.206295 Objective Loss 0.206295 LR 0.000250 Time 0.020418 -2023-02-13 18:21:49,999 - Epoch: [144][ 720/ 1207] Overall Loss 0.206149 Objective Loss 0.206149 LR 0.000250 Time 0.020399 -2023-02-13 18:21:50,190 - Epoch: [144][ 730/ 1207] Overall Loss 0.206366 Objective Loss 0.206366 LR 0.000250 Time 0.020381 -2023-02-13 18:21:50,381 - Epoch: [144][ 740/ 1207] Overall Loss 0.206487 Objective Loss 0.206487 LR 0.000250 Time 0.020363 -2023-02-13 18:21:50,572 - Epoch: [144][ 750/ 1207] Overall Loss 0.206263 Objective Loss 0.206263 LR 0.000250 Time 0.020346 -2023-02-13 18:21:50,764 - Epoch: [144][ 760/ 1207] Overall Loss 0.206280 Objective Loss 0.206280 LR 0.000250 Time 0.020330 -2023-02-13 18:21:50,956 - Epoch: [144][ 770/ 1207] Overall Loss 0.206428 Objective Loss 0.206428 LR 0.000250 Time 0.020315 -2023-02-13 18:21:51,147 - Epoch: [144][ 780/ 1207] Overall Loss 0.206473 Objective Loss 0.206473 LR 0.000250 Time 0.020299 -2023-02-13 18:21:51,338 - Epoch: [144][ 790/ 1207] Overall Loss 0.206327 Objective Loss 0.206327 LR 0.000250 Time 0.020284 -2023-02-13 18:21:51,529 - Epoch: [144][ 800/ 1207] Overall Loss 0.206443 Objective Loss 0.206443 LR 0.000250 Time 0.020269 -2023-02-13 18:21:51,720 - Epoch: [144][ 810/ 1207] Overall Loss 0.206551 Objective Loss 0.206551 LR 0.000250 Time 0.020253 -2023-02-13 18:21:51,911 - Epoch: [144][ 820/ 1207] Overall Loss 0.206494 Objective Loss 0.206494 LR 0.000250 Time 0.020239 -2023-02-13 18:21:52,103 - Epoch: [144][ 830/ 1207] Overall Loss 0.206519 Objective Loss 0.206519 LR 0.000250 Time 0.020226 -2023-02-13 18:21:52,294 - Epoch: [144][ 840/ 1207] Overall Loss 0.206288 Objective Loss 0.206288 LR 0.000250 Time 0.020211 -2023-02-13 18:21:52,486 - Epoch: [144][ 850/ 1207] Overall Loss 0.206288 Objective Loss 0.206288 LR 0.000250 Time 0.020199 -2023-02-13 18:21:52,676 - Epoch: [144][ 860/ 1207] Overall Loss 0.206242 Objective Loss 0.206242 LR 0.000250 Time 0.020185 -2023-02-13 18:21:52,867 - Epoch: [144][ 870/ 1207] Overall Loss 0.206099 Objective Loss 0.206099 LR 0.000250 Time 0.020172 -2023-02-13 18:21:53,057 - Epoch: [144][ 880/ 1207] Overall Loss 0.206068 Objective Loss 0.206068 LR 0.000250 Time 0.020158 -2023-02-13 18:21:53,248 - Epoch: [144][ 890/ 1207] Overall Loss 0.206354 Objective Loss 0.206354 LR 0.000250 Time 0.020146 -2023-02-13 18:21:53,439 - Epoch: [144][ 900/ 1207] Overall Loss 0.206620 Objective Loss 0.206620 LR 0.000250 Time 0.020134 -2023-02-13 18:21:53,629 - Epoch: [144][ 910/ 1207] Overall Loss 0.206714 Objective Loss 0.206714 LR 0.000250 Time 0.020122 -2023-02-13 18:21:53,821 - Epoch: [144][ 920/ 1207] Overall Loss 0.206723 Objective Loss 0.206723 LR 0.000250 Time 0.020110 -2023-02-13 18:21:54,012 - Epoch: [144][ 930/ 1207] Overall Loss 0.206687 Objective Loss 0.206687 LR 0.000250 Time 0.020099 -2023-02-13 18:21:54,202 - Epoch: [144][ 940/ 1207] Overall Loss 0.206872 Objective Loss 0.206872 LR 0.000250 Time 0.020088 -2023-02-13 18:21:54,393 - Epoch: [144][ 950/ 1207] Overall Loss 0.206852 Objective Loss 0.206852 LR 0.000250 Time 0.020077 -2023-02-13 18:21:54,585 - Epoch: [144][ 960/ 1207] Overall Loss 0.206907 Objective Loss 0.206907 LR 0.000250 Time 0.020067 -2023-02-13 18:21:54,776 - Epoch: [144][ 970/ 1207] Overall Loss 0.206826 Objective Loss 0.206826 LR 0.000250 Time 0.020056 -2023-02-13 18:21:54,967 - Epoch: [144][ 980/ 1207] Overall Loss 0.206821 Objective Loss 0.206821 LR 0.000250 Time 0.020046 -2023-02-13 18:21:55,159 - Epoch: [144][ 990/ 1207] Overall Loss 0.206938 Objective Loss 0.206938 LR 0.000250 Time 0.020037 -2023-02-13 18:21:55,350 - Epoch: [144][ 1000/ 1207] Overall Loss 0.206899 Objective Loss 0.206899 LR 0.000250 Time 0.020028 -2023-02-13 18:21:55,542 - Epoch: [144][ 1010/ 1207] Overall Loss 0.206783 Objective Loss 0.206783 LR 0.000250 Time 0.020019 -2023-02-13 18:21:55,733 - Epoch: [144][ 1020/ 1207] Overall Loss 0.206828 Objective Loss 0.206828 LR 0.000250 Time 0.020010 -2023-02-13 18:21:55,924 - Epoch: [144][ 1030/ 1207] Overall Loss 0.206798 Objective Loss 0.206798 LR 0.000250 Time 0.020001 -2023-02-13 18:21:56,115 - Epoch: [144][ 1040/ 1207] Overall Loss 0.206866 Objective Loss 0.206866 LR 0.000250 Time 0.019992 -2023-02-13 18:21:56,307 - Epoch: [144][ 1050/ 1207] Overall Loss 0.206815 Objective Loss 0.206815 LR 0.000250 Time 0.019984 -2023-02-13 18:21:56,498 - Epoch: [144][ 1060/ 1207] Overall Loss 0.206909 Objective Loss 0.206909 LR 0.000250 Time 0.019975 -2023-02-13 18:21:56,689 - Epoch: [144][ 1070/ 1207] Overall Loss 0.206834 Objective Loss 0.206834 LR 0.000250 Time 0.019967 -2023-02-13 18:21:56,880 - Epoch: [144][ 1080/ 1207] Overall Loss 0.206913 Objective Loss 0.206913 LR 0.000250 Time 0.019958 -2023-02-13 18:21:57,072 - Epoch: [144][ 1090/ 1207] Overall Loss 0.206904 Objective Loss 0.206904 LR 0.000250 Time 0.019951 -2023-02-13 18:21:57,264 - Epoch: [144][ 1100/ 1207] Overall Loss 0.206992 Objective Loss 0.206992 LR 0.000250 Time 0.019943 -2023-02-13 18:21:57,455 - Epoch: [144][ 1110/ 1207] Overall Loss 0.206906 Objective Loss 0.206906 LR 0.000250 Time 0.019936 -2023-02-13 18:21:57,647 - Epoch: [144][ 1120/ 1207] Overall Loss 0.206818 Objective Loss 0.206818 LR 0.000250 Time 0.019929 -2023-02-13 18:21:57,838 - Epoch: [144][ 1130/ 1207] Overall Loss 0.206833 Objective Loss 0.206833 LR 0.000250 Time 0.019921 -2023-02-13 18:21:58,028 - Epoch: [144][ 1140/ 1207] Overall Loss 0.206976 Objective Loss 0.206976 LR 0.000250 Time 0.019913 -2023-02-13 18:21:58,220 - Epoch: [144][ 1150/ 1207] Overall Loss 0.206982 Objective Loss 0.206982 LR 0.000250 Time 0.019906 -2023-02-13 18:21:58,411 - Epoch: [144][ 1160/ 1207] Overall Loss 0.206927 Objective Loss 0.206927 LR 0.000250 Time 0.019899 -2023-02-13 18:21:58,603 - Epoch: [144][ 1170/ 1207] Overall Loss 0.206811 Objective Loss 0.206811 LR 0.000250 Time 0.019893 -2023-02-13 18:21:58,794 - Epoch: [144][ 1180/ 1207] Overall Loss 0.206839 Objective Loss 0.206839 LR 0.000250 Time 0.019886 -2023-02-13 18:21:58,986 - Epoch: [144][ 1190/ 1207] Overall Loss 0.206776 Objective Loss 0.206776 LR 0.000250 Time 0.019880 -2023-02-13 18:21:59,229 - Epoch: [144][ 1200/ 1207] Overall Loss 0.206582 Objective Loss 0.206582 LR 0.000250 Time 0.019916 -2023-02-13 18:21:59,345 - Epoch: [144][ 1207/ 1207] Overall Loss 0.206656 Objective Loss 0.206656 Top1 85.670732 Top5 98.780488 LR 0.000250 Time 0.019897 -2023-02-13 18:21:59,416 - --- validate (epoch=144)----------- -2023-02-13 18:21:59,416 - 34311 samples (256 per mini-batch) -2023-02-13 18:21:59,808 - Epoch: [144][ 10/ 135] Loss 0.314325 Top1 83.945312 Top5 98.007812 -2023-02-13 18:21:59,933 - Epoch: [144][ 20/ 135] Loss 0.328973 Top1 84.238281 Top5 97.753906 -2023-02-13 18:22:00,065 - Epoch: [144][ 30/ 135] Loss 0.323643 Top1 84.596354 Top5 97.799479 -2023-02-13 18:22:00,196 - Epoch: [144][ 40/ 135] Loss 0.320386 Top1 84.882812 Top5 97.714844 -2023-02-13 18:22:00,326 - Epoch: [144][ 50/ 135] Loss 0.310216 Top1 85.210938 Top5 97.789062 -2023-02-13 18:22:00,464 - Epoch: [144][ 60/ 135] Loss 0.303229 Top1 85.332031 Top5 97.734375 -2023-02-13 18:22:00,610 - Epoch: [144][ 70/ 135] Loss 0.297748 Top1 85.463170 Top5 97.806920 -2023-02-13 18:22:00,744 - Epoch: [144][ 80/ 135] Loss 0.297572 Top1 85.439453 Top5 97.807617 -2023-02-13 18:22:00,873 - Epoch: [144][ 90/ 135] Loss 0.295051 Top1 85.490451 Top5 97.808160 -2023-02-13 18:22:00,999 - Epoch: [144][ 100/ 135] Loss 0.295707 Top1 85.320312 Top5 97.820312 -2023-02-13 18:22:01,124 - Epoch: [144][ 110/ 135] Loss 0.295916 Top1 85.387074 Top5 97.819602 -2023-02-13 18:22:01,250 - Epoch: [144][ 120/ 135] Loss 0.292934 Top1 85.478516 Top5 97.841797 -2023-02-13 18:22:01,384 - Epoch: [144][ 130/ 135] Loss 0.293274 Top1 85.480769 Top5 97.845553 -2023-02-13 18:22:01,430 - Epoch: [144][ 135/ 135] Loss 0.292571 Top1 85.506106 Top5 97.863659 -2023-02-13 18:22:01,503 - ==> Top1: 85.506 Top5: 97.864 Loss: 0.293 - -2023-02-13 18:22:01,504 - ==> Confusion: -[[ 864 4 4 1 10 3 0 0 4 41 1 3 1 4 8 5 2 1 1 1 9] - [ 3 959 2 2 12 16 2 16 3 1 1 0 0 0 1 1 4 0 3 0 7] - [ 6 5 973 12 5 1 8 13 0 2 2 0 3 3 4 5 3 3 4 2 4] - [ 6 1 18 913 4 3 0 1 3 1 9 1 7 0 15 2 2 4 21 0 5] - [ 9 7 0 0 1005 6 0 1 2 1 0 5 2 1 7 9 6 1 0 2 2] - [ 2 21 1 4 10 968 5 14 1 3 2 6 5 11 0 2 4 0 0 3 8] - [ 2 7 18 0 0 5 1032 4 0 2 2 1 3 2 0 4 2 1 2 7 5] - [ 2 8 8 1 3 23 1 934 0 1 0 3 2 1 0 1 0 1 22 6 7] - [ 20 2 0 1 1 0 1 1 894 39 9 1 1 11 14 4 2 1 4 0 3] - [ 73 1 3 0 10 1 0 2 27 866 0 2 1 15 4 2 0 1 1 0 3] - [ 2 1 1 7 1 3 3 2 14 0 994 1 1 6 1 0 0 2 8 0 4] - [ 2 2 2 0 3 9 1 5 2 1 0 911 28 10 0 4 1 12 1 9 2] - [ 1 0 0 7 1 4 0 2 1 2 1 19 880 1 3 5 3 17 1 1 10] - [ 5 3 1 0 8 7 1 2 8 14 8 5 1 942 3 5 1 2 0 0 8] - [ 8 2 2 22 5 4 0 1 17 6 5 1 1 1 991 0 1 7 10 1 7] - [ 3 1 10 0 5 0 5 1 1 0 0 6 7 3 1 964 12 16 0 6 5] - [ 4 4 0 1 9 3 0 0 1 1 0 2 2 1 1 13 1004 1 1 5 8] - [ 5 3 1 2 1 1 1 0 0 1 2 8 17 0 1 15 0 985 3 1 4] - [ 2 3 4 7 2 2 0 24 4 1 3 0 3 0 11 1 1 2 1009 4 3] - [ 1 3 1 0 2 4 2 8 0 0 0 11 3 5 0 4 4 4 0 1089 7] - [ 136 241 216 116 156 179 72 170 76 78 193 94 303 266 157 96 211 93 182 238 10161]] - -2023-02-13 18:22:01,506 - ==> Best [Top1: 85.506 Top5: 97.864 Sparsity:0.00 Params: 148928 on epoch: 144] -2023-02-13 18:22:01,506 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:22:01,513 - - -2023-02-13 18:22:01,513 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:22:02,410 - Epoch: [145][ 10/ 1207] Overall Loss 0.213365 Objective Loss 0.213365 LR 0.000250 Time 0.089696 -2023-02-13 18:22:02,607 - Epoch: [145][ 20/ 1207] Overall Loss 0.213709 Objective Loss 0.213709 LR 0.000250 Time 0.054659 -2023-02-13 18:22:02,796 - Epoch: [145][ 30/ 1207] Overall Loss 0.207695 Objective Loss 0.207695 LR 0.000250 Time 0.042709 -2023-02-13 18:22:02,985 - Epoch: [145][ 40/ 1207] Overall Loss 0.203609 Objective Loss 0.203609 LR 0.000250 Time 0.036749 -2023-02-13 18:22:03,173 - Epoch: [145][ 50/ 1207] Overall Loss 0.201079 Objective Loss 0.201079 LR 0.000250 Time 0.033156 -2023-02-13 18:22:03,361 - Epoch: [145][ 60/ 1207] Overall Loss 0.199837 Objective Loss 0.199837 LR 0.000250 Time 0.030769 -2023-02-13 18:22:03,550 - Epoch: [145][ 70/ 1207] Overall Loss 0.202403 Objective Loss 0.202403 LR 0.000250 Time 0.029066 -2023-02-13 18:22:03,739 - Epoch: [145][ 80/ 1207] Overall Loss 0.202851 Objective Loss 0.202851 LR 0.000250 Time 0.027781 -2023-02-13 18:22:03,927 - Epoch: [145][ 90/ 1207] Overall Loss 0.204160 Objective Loss 0.204160 LR 0.000250 Time 0.026784 -2023-02-13 18:22:04,116 - Epoch: [145][ 100/ 1207] Overall Loss 0.206156 Objective Loss 0.206156 LR 0.000250 Time 0.025988 -2023-02-13 18:22:04,305 - Epoch: [145][ 110/ 1207] Overall Loss 0.203792 Objective Loss 0.203792 LR 0.000250 Time 0.025341 -2023-02-13 18:22:04,494 - Epoch: [145][ 120/ 1207] Overall Loss 0.205380 Objective Loss 0.205380 LR 0.000250 Time 0.024804 -2023-02-13 18:22:04,683 - Epoch: [145][ 130/ 1207] Overall Loss 0.206615 Objective Loss 0.206615 LR 0.000250 Time 0.024343 -2023-02-13 18:22:04,871 - Epoch: [145][ 140/ 1207] Overall Loss 0.205361 Objective Loss 0.205361 LR 0.000250 Time 0.023949 -2023-02-13 18:22:05,060 - Epoch: [145][ 150/ 1207] Overall Loss 0.206436 Objective Loss 0.206436 LR 0.000250 Time 0.023605 -2023-02-13 18:22:05,248 - Epoch: [145][ 160/ 1207] Overall Loss 0.205428 Objective Loss 0.205428 LR 0.000250 Time 0.023308 -2023-02-13 18:22:05,437 - Epoch: [145][ 170/ 1207] Overall Loss 0.205095 Objective Loss 0.205095 LR 0.000250 Time 0.023042 -2023-02-13 18:22:05,626 - Epoch: [145][ 180/ 1207] Overall Loss 0.205463 Objective Loss 0.205463 LR 0.000250 Time 0.022809 -2023-02-13 18:22:05,814 - Epoch: [145][ 190/ 1207] Overall Loss 0.204995 Objective Loss 0.204995 LR 0.000250 Time 0.022597 -2023-02-13 18:22:06,003 - Epoch: [145][ 200/ 1207] Overall Loss 0.204729 Objective Loss 0.204729 LR 0.000250 Time 0.022412 -2023-02-13 18:22:06,192 - Epoch: [145][ 210/ 1207] Overall Loss 0.205213 Objective Loss 0.205213 LR 0.000250 Time 0.022242 -2023-02-13 18:22:06,381 - Epoch: [145][ 220/ 1207] Overall Loss 0.206001 Objective Loss 0.206001 LR 0.000250 Time 0.022089 -2023-02-13 18:22:06,571 - Epoch: [145][ 230/ 1207] Overall Loss 0.206878 Objective Loss 0.206878 LR 0.000250 Time 0.021954 -2023-02-13 18:22:06,759 - Epoch: [145][ 240/ 1207] Overall Loss 0.207000 Objective Loss 0.207000 LR 0.000250 Time 0.021822 -2023-02-13 18:22:06,949 - Epoch: [145][ 250/ 1207] Overall Loss 0.207594 Objective Loss 0.207594 LR 0.000250 Time 0.021705 -2023-02-13 18:22:07,137 - Epoch: [145][ 260/ 1207] Overall Loss 0.207347 Objective Loss 0.207347 LR 0.000250 Time 0.021593 -2023-02-13 18:22:07,326 - Epoch: [145][ 270/ 1207] Overall Loss 0.207095 Objective Loss 0.207095 LR 0.000250 Time 0.021491 -2023-02-13 18:22:07,515 - Epoch: [145][ 280/ 1207] Overall Loss 0.207172 Objective Loss 0.207172 LR 0.000250 Time 0.021398 -2023-02-13 18:22:07,703 - Epoch: [145][ 290/ 1207] Overall Loss 0.207589 Objective Loss 0.207589 LR 0.000250 Time 0.021308 -2023-02-13 18:22:07,892 - Epoch: [145][ 300/ 1207] Overall Loss 0.208165 Objective Loss 0.208165 LR 0.000250 Time 0.021227 -2023-02-13 18:22:08,081 - Epoch: [145][ 310/ 1207] Overall Loss 0.208953 Objective Loss 0.208953 LR 0.000250 Time 0.021148 -2023-02-13 18:22:08,269 - Epoch: [145][ 320/ 1207] Overall Loss 0.209984 Objective Loss 0.209984 LR 0.000250 Time 0.021075 -2023-02-13 18:22:08,458 - Epoch: [145][ 330/ 1207] Overall Loss 0.210285 Objective Loss 0.210285 LR 0.000250 Time 0.021008 -2023-02-13 18:22:08,647 - Epoch: [145][ 340/ 1207] Overall Loss 0.210012 Objective Loss 0.210012 LR 0.000250 Time 0.020946 -2023-02-13 18:22:08,836 - Epoch: [145][ 350/ 1207] Overall Loss 0.210081 Objective Loss 0.210081 LR 0.000250 Time 0.020885 -2023-02-13 18:22:09,025 - Epoch: [145][ 360/ 1207] Overall Loss 0.210316 Objective Loss 0.210316 LR 0.000250 Time 0.020828 -2023-02-13 18:22:09,213 - Epoch: [145][ 370/ 1207] Overall Loss 0.209714 Objective Loss 0.209714 LR 0.000250 Time 0.020774 -2023-02-13 18:22:09,403 - Epoch: [145][ 380/ 1207] Overall Loss 0.208982 Objective Loss 0.208982 LR 0.000250 Time 0.020725 -2023-02-13 18:22:09,592 - Epoch: [145][ 390/ 1207] Overall Loss 0.208978 Objective Loss 0.208978 LR 0.000250 Time 0.020677 -2023-02-13 18:22:09,781 - Epoch: [145][ 400/ 1207] Overall Loss 0.209423 Objective Loss 0.209423 LR 0.000250 Time 0.020631 -2023-02-13 18:22:09,969 - Epoch: [145][ 410/ 1207] Overall Loss 0.208834 Objective Loss 0.208834 LR 0.000250 Time 0.020586 -2023-02-13 18:22:10,158 - Epoch: [145][ 420/ 1207] Overall Loss 0.209340 Objective Loss 0.209340 LR 0.000250 Time 0.020545 -2023-02-13 18:22:10,347 - Epoch: [145][ 430/ 1207] Overall Loss 0.209013 Objective Loss 0.209013 LR 0.000250 Time 0.020507 -2023-02-13 18:22:10,537 - Epoch: [145][ 440/ 1207] Overall Loss 0.209148 Objective Loss 0.209148 LR 0.000250 Time 0.020471 -2023-02-13 18:22:10,725 - Epoch: [145][ 450/ 1207] Overall Loss 0.208944 Objective Loss 0.208944 LR 0.000250 Time 0.020434 -2023-02-13 18:22:10,916 - Epoch: [145][ 460/ 1207] Overall Loss 0.208486 Objective Loss 0.208486 LR 0.000250 Time 0.020403 -2023-02-13 18:22:11,105 - Epoch: [145][ 470/ 1207] Overall Loss 0.208071 Objective Loss 0.208071 LR 0.000250 Time 0.020371 -2023-02-13 18:22:11,295 - Epoch: [145][ 480/ 1207] Overall Loss 0.208242 Objective Loss 0.208242 LR 0.000250 Time 0.020341 -2023-02-13 18:22:11,484 - Epoch: [145][ 490/ 1207] Overall Loss 0.208031 Objective Loss 0.208031 LR 0.000250 Time 0.020311 -2023-02-13 18:22:11,673 - Epoch: [145][ 500/ 1207] Overall Loss 0.207807 Objective Loss 0.207807 LR 0.000250 Time 0.020283 -2023-02-13 18:22:11,862 - Epoch: [145][ 510/ 1207] Overall Loss 0.207712 Objective Loss 0.207712 LR 0.000250 Time 0.020255 -2023-02-13 18:22:12,052 - Epoch: [145][ 520/ 1207] Overall Loss 0.207712 Objective Loss 0.207712 LR 0.000250 Time 0.020229 -2023-02-13 18:22:12,241 - Epoch: [145][ 530/ 1207] Overall Loss 0.207913 Objective Loss 0.207913 LR 0.000250 Time 0.020203 -2023-02-13 18:22:12,430 - Epoch: [145][ 540/ 1207] Overall Loss 0.208080 Objective Loss 0.208080 LR 0.000250 Time 0.020179 -2023-02-13 18:22:12,619 - Epoch: [145][ 550/ 1207] Overall Loss 0.207634 Objective Loss 0.207634 LR 0.000250 Time 0.020155 -2023-02-13 18:22:12,807 - Epoch: [145][ 560/ 1207] Overall Loss 0.207875 Objective Loss 0.207875 LR 0.000250 Time 0.020131 -2023-02-13 18:22:12,996 - Epoch: [145][ 570/ 1207] Overall Loss 0.208031 Objective Loss 0.208031 LR 0.000250 Time 0.020108 -2023-02-13 18:22:13,184 - Epoch: [145][ 580/ 1207] Overall Loss 0.208185 Objective Loss 0.208185 LR 0.000250 Time 0.020085 -2023-02-13 18:22:13,374 - Epoch: [145][ 590/ 1207] Overall Loss 0.208183 Objective Loss 0.208183 LR 0.000250 Time 0.020065 -2023-02-13 18:22:13,563 - Epoch: [145][ 600/ 1207] Overall Loss 0.208085 Objective Loss 0.208085 LR 0.000250 Time 0.020046 -2023-02-13 18:22:13,752 - Epoch: [145][ 610/ 1207] Overall Loss 0.207968 Objective Loss 0.207968 LR 0.000250 Time 0.020026 -2023-02-13 18:22:13,941 - Epoch: [145][ 620/ 1207] Overall Loss 0.207709 Objective Loss 0.207709 LR 0.000250 Time 0.020007 -2023-02-13 18:22:14,130 - Epoch: [145][ 630/ 1207] Overall Loss 0.207398 Objective Loss 0.207398 LR 0.000250 Time 0.019989 -2023-02-13 18:22:14,319 - Epoch: [145][ 640/ 1207] Overall Loss 0.207030 Objective Loss 0.207030 LR 0.000250 Time 0.019972 -2023-02-13 18:22:14,508 - Epoch: [145][ 650/ 1207] Overall Loss 0.207223 Objective Loss 0.207223 LR 0.000250 Time 0.019955 -2023-02-13 18:22:14,697 - Epoch: [145][ 660/ 1207] Overall Loss 0.206846 Objective Loss 0.206846 LR 0.000250 Time 0.019939 -2023-02-13 18:22:14,886 - Epoch: [145][ 670/ 1207] Overall Loss 0.206936 Objective Loss 0.206936 LR 0.000250 Time 0.019922 -2023-02-13 18:22:15,075 - Epoch: [145][ 680/ 1207] Overall Loss 0.206886 Objective Loss 0.206886 LR 0.000250 Time 0.019906 -2023-02-13 18:22:15,264 - Epoch: [145][ 690/ 1207] Overall Loss 0.206857 Objective Loss 0.206857 LR 0.000250 Time 0.019892 -2023-02-13 18:22:15,453 - Epoch: [145][ 700/ 1207] Overall Loss 0.206729 Objective Loss 0.206729 LR 0.000250 Time 0.019877 -2023-02-13 18:22:15,642 - Epoch: [145][ 710/ 1207] Overall Loss 0.206778 Objective Loss 0.206778 LR 0.000250 Time 0.019863 -2023-02-13 18:22:15,832 - Epoch: [145][ 720/ 1207] Overall Loss 0.206731 Objective Loss 0.206731 LR 0.000250 Time 0.019850 -2023-02-13 18:22:16,022 - Epoch: [145][ 730/ 1207] Overall Loss 0.206861 Objective Loss 0.206861 LR 0.000250 Time 0.019838 -2023-02-13 18:22:16,211 - Epoch: [145][ 740/ 1207] Overall Loss 0.206780 Objective Loss 0.206780 LR 0.000250 Time 0.019825 -2023-02-13 18:22:16,401 - Epoch: [145][ 750/ 1207] Overall Loss 0.206804 Objective Loss 0.206804 LR 0.000250 Time 0.019813 -2023-02-13 18:22:16,591 - Epoch: [145][ 760/ 1207] Overall Loss 0.206499 Objective Loss 0.206499 LR 0.000250 Time 0.019802 -2023-02-13 18:22:16,780 - Epoch: [145][ 770/ 1207] Overall Loss 0.206383 Objective Loss 0.206383 LR 0.000250 Time 0.019790 -2023-02-13 18:22:16,970 - Epoch: [145][ 780/ 1207] Overall Loss 0.206743 Objective Loss 0.206743 LR 0.000250 Time 0.019779 -2023-02-13 18:22:17,159 - Epoch: [145][ 790/ 1207] Overall Loss 0.206590 Objective Loss 0.206590 LR 0.000250 Time 0.019767 -2023-02-13 18:22:17,348 - Epoch: [145][ 800/ 1207] Overall Loss 0.206732 Objective Loss 0.206732 LR 0.000250 Time 0.019756 -2023-02-13 18:22:17,537 - Epoch: [145][ 810/ 1207] Overall Loss 0.206909 Objective Loss 0.206909 LR 0.000250 Time 0.019746 -2023-02-13 18:22:17,726 - Epoch: [145][ 820/ 1207] Overall Loss 0.206858 Objective Loss 0.206858 LR 0.000250 Time 0.019735 -2023-02-13 18:22:17,915 - Epoch: [145][ 830/ 1207] Overall Loss 0.206757 Objective Loss 0.206757 LR 0.000250 Time 0.019724 -2023-02-13 18:22:18,104 - Epoch: [145][ 840/ 1207] Overall Loss 0.206650 Objective Loss 0.206650 LR 0.000250 Time 0.019714 -2023-02-13 18:22:18,294 - Epoch: [145][ 850/ 1207] Overall Loss 0.206510 Objective Loss 0.206510 LR 0.000250 Time 0.019705 -2023-02-13 18:22:18,483 - Epoch: [145][ 860/ 1207] Overall Loss 0.206542 Objective Loss 0.206542 LR 0.000250 Time 0.019695 -2023-02-13 18:22:18,673 - Epoch: [145][ 870/ 1207] Overall Loss 0.206594 Objective Loss 0.206594 LR 0.000250 Time 0.019686 -2023-02-13 18:22:18,861 - Epoch: [145][ 880/ 1207] Overall Loss 0.206402 Objective Loss 0.206402 LR 0.000250 Time 0.019677 -2023-02-13 18:22:19,051 - Epoch: [145][ 890/ 1207] Overall Loss 0.206369 Objective Loss 0.206369 LR 0.000250 Time 0.019668 -2023-02-13 18:22:19,240 - Epoch: [145][ 900/ 1207] Overall Loss 0.206356 Objective Loss 0.206356 LR 0.000250 Time 0.019659 -2023-02-13 18:22:19,429 - Epoch: [145][ 910/ 1207] Overall Loss 0.206221 Objective Loss 0.206221 LR 0.000250 Time 0.019651 -2023-02-13 18:22:19,620 - Epoch: [145][ 920/ 1207] Overall Loss 0.206156 Objective Loss 0.206156 LR 0.000250 Time 0.019644 -2023-02-13 18:22:19,808 - Epoch: [145][ 930/ 1207] Overall Loss 0.206203 Objective Loss 0.206203 LR 0.000250 Time 0.019635 -2023-02-13 18:22:19,998 - Epoch: [145][ 940/ 1207] Overall Loss 0.206133 Objective Loss 0.206133 LR 0.000250 Time 0.019627 -2023-02-13 18:22:20,187 - Epoch: [145][ 950/ 1207] Overall Loss 0.206087 Objective Loss 0.206087 LR 0.000250 Time 0.019620 -2023-02-13 18:22:20,377 - Epoch: [145][ 960/ 1207] Overall Loss 0.205997 Objective Loss 0.205997 LR 0.000250 Time 0.019612 -2023-02-13 18:22:20,567 - Epoch: [145][ 970/ 1207] Overall Loss 0.206067 Objective Loss 0.206067 LR 0.000250 Time 0.019606 -2023-02-13 18:22:20,757 - Epoch: [145][ 980/ 1207] Overall Loss 0.206211 Objective Loss 0.206211 LR 0.000250 Time 0.019600 -2023-02-13 18:22:20,949 - Epoch: [145][ 990/ 1207] Overall Loss 0.206219 Objective Loss 0.206219 LR 0.000250 Time 0.019595 -2023-02-13 18:22:21,139 - Epoch: [145][ 1000/ 1207] Overall Loss 0.206227 Objective Loss 0.206227 LR 0.000250 Time 0.019589 -2023-02-13 18:22:21,329 - Epoch: [145][ 1010/ 1207] Overall Loss 0.206267 Objective Loss 0.206267 LR 0.000250 Time 0.019583 -2023-02-13 18:22:21,519 - Epoch: [145][ 1020/ 1207] Overall Loss 0.206110 Objective Loss 0.206110 LR 0.000250 Time 0.019577 -2023-02-13 18:22:21,710 - Epoch: [145][ 1030/ 1207] Overall Loss 0.206138 Objective Loss 0.206138 LR 0.000250 Time 0.019571 -2023-02-13 18:22:21,899 - Epoch: [145][ 1040/ 1207] Overall Loss 0.206262 Objective Loss 0.206262 LR 0.000250 Time 0.019565 -2023-02-13 18:22:22,090 - Epoch: [145][ 1050/ 1207] Overall Loss 0.206195 Objective Loss 0.206195 LR 0.000250 Time 0.019560 -2023-02-13 18:22:22,281 - Epoch: [145][ 1060/ 1207] Overall Loss 0.206003 Objective Loss 0.206003 LR 0.000250 Time 0.019555 -2023-02-13 18:22:22,472 - Epoch: [145][ 1070/ 1207] Overall Loss 0.205994 Objective Loss 0.205994 LR 0.000250 Time 0.019550 -2023-02-13 18:22:22,662 - Epoch: [145][ 1080/ 1207] Overall Loss 0.206207 Objective Loss 0.206207 LR 0.000250 Time 0.019545 -2023-02-13 18:22:22,853 - Epoch: [145][ 1090/ 1207] Overall Loss 0.206091 Objective Loss 0.206091 LR 0.000250 Time 0.019541 -2023-02-13 18:22:23,043 - Epoch: [145][ 1100/ 1207] Overall Loss 0.205935 Objective Loss 0.205935 LR 0.000250 Time 0.019536 -2023-02-13 18:22:23,235 - Epoch: [145][ 1110/ 1207] Overall Loss 0.206131 Objective Loss 0.206131 LR 0.000250 Time 0.019532 -2023-02-13 18:22:23,425 - Epoch: [145][ 1120/ 1207] Overall Loss 0.206152 Objective Loss 0.206152 LR 0.000250 Time 0.019527 -2023-02-13 18:22:23,616 - Epoch: [145][ 1130/ 1207] Overall Loss 0.206296 Objective Loss 0.206296 LR 0.000250 Time 0.019523 -2023-02-13 18:22:23,807 - Epoch: [145][ 1140/ 1207] Overall Loss 0.206418 Objective Loss 0.206418 LR 0.000250 Time 0.019518 -2023-02-13 18:22:23,997 - Epoch: [145][ 1150/ 1207] Overall Loss 0.206447 Objective Loss 0.206447 LR 0.000250 Time 0.019514 -2023-02-13 18:22:24,188 - Epoch: [145][ 1160/ 1207] Overall Loss 0.206538 Objective Loss 0.206538 LR 0.000250 Time 0.019510 -2023-02-13 18:22:24,378 - Epoch: [145][ 1170/ 1207] Overall Loss 0.206520 Objective Loss 0.206520 LR 0.000250 Time 0.019506 -2023-02-13 18:22:24,573 - Epoch: [145][ 1180/ 1207] Overall Loss 0.206492 Objective Loss 0.206492 LR 0.000250 Time 0.019505 -2023-02-13 18:22:24,772 - Epoch: [145][ 1190/ 1207] Overall Loss 0.206385 Objective Loss 0.206385 LR 0.000250 Time 0.019508 -2023-02-13 18:22:25,027 - Epoch: [145][ 1200/ 1207] Overall Loss 0.206390 Objective Loss 0.206390 LR 0.000250 Time 0.019558 -2023-02-13 18:22:25,142 - Epoch: [145][ 1207/ 1207] Overall Loss 0.206483 Objective Loss 0.206483 Top1 88.414634 Top5 98.475610 LR 0.000250 Time 0.019539 -2023-02-13 18:22:25,220 - --- validate (epoch=145)----------- -2023-02-13 18:22:25,220 - 34311 samples (256 per mini-batch) -2023-02-13 18:22:25,638 - Epoch: [145][ 10/ 135] Loss 0.270888 Top1 87.421875 Top5 97.890625 -2023-02-13 18:22:25,781 - Epoch: [145][ 20/ 135] Loss 0.291725 Top1 86.425781 Top5 97.812500 -2023-02-13 18:22:25,925 - Epoch: [145][ 30/ 135] Loss 0.307148 Top1 86.263021 Top5 97.799479 -2023-02-13 18:22:26,067 - Epoch: [145][ 40/ 135] Loss 0.300521 Top1 86.210938 Top5 97.832031 -2023-02-13 18:22:26,209 - Epoch: [145][ 50/ 135] Loss 0.298027 Top1 86.125000 Top5 97.914062 -2023-02-13 18:22:26,351 - Epoch: [145][ 60/ 135] Loss 0.294698 Top1 86.067708 Top5 97.988281 -2023-02-13 18:22:26,492 - Epoch: [145][ 70/ 135] Loss 0.292416 Top1 86.127232 Top5 98.035714 -2023-02-13 18:22:26,621 - Epoch: [145][ 80/ 135] Loss 0.295062 Top1 85.961914 Top5 98.017578 -2023-02-13 18:22:26,751 - Epoch: [145][ 90/ 135] Loss 0.296196 Top1 85.898438 Top5 98.016493 -2023-02-13 18:22:26,881 - Epoch: [145][ 100/ 135] Loss 0.296123 Top1 85.906250 Top5 98.019531 -2023-02-13 18:22:27,009 - Epoch: [145][ 110/ 135] Loss 0.297134 Top1 85.809659 Top5 98.011364 -2023-02-13 18:22:27,138 - Epoch: [145][ 120/ 135] Loss 0.297088 Top1 85.716146 Top5 97.968750 -2023-02-13 18:22:27,272 - Epoch: [145][ 130/ 135] Loss 0.295476 Top1 85.715144 Top5 97.998798 -2023-02-13 18:22:27,320 - Epoch: [145][ 135/ 135] Loss 0.294764 Top1 85.701379 Top5 97.983154 -2023-02-13 18:22:27,389 - ==> Top1: 85.701 Top5: 97.983 Loss: 0.295 - -2023-02-13 18:22:27,389 - ==> Confusion: -[[ 870 4 3 0 12 3 0 3 4 38 0 4 1 4 6 3 2 1 1 1 7] - [ 3 958 2 2 9 20 0 22 0 1 1 0 0 0 0 2 4 0 2 0 7] - [ 10 6 968 7 3 1 13 11 0 1 4 1 3 3 3 5 4 4 2 3 6] - [ 7 1 24 897 3 3 0 3 1 2 11 1 7 0 22 2 3 5 13 0 11] - [ 8 10 0 1 1002 10 1 3 1 1 0 3 1 2 6 5 5 1 0 2 4] - [ 2 16 0 3 10 977 5 11 0 4 1 11 3 10 1 0 4 1 1 4 6] - [ 2 4 6 0 1 5 1047 4 0 2 2 1 2 2 0 6 4 2 2 4 3] - [ 0 7 7 1 4 39 3 926 1 1 2 5 2 0 0 1 0 2 13 7 3] - [ 18 2 1 1 1 0 0 3 898 35 9 3 0 10 12 2 3 0 3 1 7] - [ 84 1 3 1 10 2 0 4 31 844 0 1 1 15 3 2 1 4 1 0 4] - [ 2 2 6 6 2 0 3 3 9 0 993 1 1 6 1 1 1 1 9 1 3] - [ 1 3 2 0 4 11 1 9 1 1 0 922 19 4 1 5 2 5 1 12 1] - [ 0 0 1 5 2 3 0 1 3 2 2 32 861 2 3 6 4 17 2 2 11] - [ 4 4 2 0 9 8 0 4 7 9 6 7 2 939 3 7 1 2 0 2 8] - [ 12 2 1 13 6 6 0 1 15 4 2 0 2 1 996 0 4 7 6 1 13] - [ 4 1 8 1 5 1 7 1 1 0 0 5 7 2 0 971 11 8 0 8 5] - [ 2 4 1 1 10 0 0 0 1 0 0 1 1 1 1 8 1014 2 1 3 10] - [ 5 2 1 2 1 1 3 2 0 0 1 11 13 0 0 22 0 978 1 1 7] - [ 4 6 4 7 0 3 0 25 4 1 3 0 4 0 13 0 1 3 999 3 6] - [ 0 2 1 0 2 6 8 9 1 0 0 14 3 5 0 5 7 3 1 1073 8] - [ 136 232 208 85 135 211 100 156 86 55 166 112 261 245 135 100 267 88 145 239 10272]] - -2023-02-13 18:22:27,391 - ==> Best [Top1: 85.701 Top5: 97.983 Sparsity:0.00 Params: 148928 on epoch: 145] -2023-02-13 18:22:27,391 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:22:27,398 - - -2023-02-13 18:22:27,398 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:22:28,285 - Epoch: [146][ 10/ 1207] Overall Loss 0.192690 Objective Loss 0.192690 LR 0.000250 Time 0.088614 -2023-02-13 18:22:28,481 - Epoch: [146][ 20/ 1207] Overall Loss 0.196940 Objective Loss 0.196940 LR 0.000250 Time 0.054104 -2023-02-13 18:22:28,671 - Epoch: [146][ 30/ 1207] Overall Loss 0.203116 Objective Loss 0.203116 LR 0.000250 Time 0.042384 -2023-02-13 18:22:28,859 - Epoch: [146][ 40/ 1207] Overall Loss 0.199955 Objective Loss 0.199955 LR 0.000250 Time 0.036490 -2023-02-13 18:22:29,048 - Epoch: [146][ 50/ 1207] Overall Loss 0.195616 Objective Loss 0.195616 LR 0.000250 Time 0.032956 -2023-02-13 18:22:29,236 - Epoch: [146][ 60/ 1207] Overall Loss 0.194981 Objective Loss 0.194981 LR 0.000250 Time 0.030597 -2023-02-13 18:22:29,425 - Epoch: [146][ 70/ 1207] Overall Loss 0.196566 Objective Loss 0.196566 LR 0.000250 Time 0.028916 -2023-02-13 18:22:29,613 - Epoch: [146][ 80/ 1207] Overall Loss 0.195321 Objective Loss 0.195321 LR 0.000250 Time 0.027654 -2023-02-13 18:22:29,802 - Epoch: [146][ 90/ 1207] Overall Loss 0.196999 Objective Loss 0.196999 LR 0.000250 Time 0.026668 -2023-02-13 18:22:29,989 - Epoch: [146][ 100/ 1207] Overall Loss 0.196228 Objective Loss 0.196228 LR 0.000250 Time 0.025876 -2023-02-13 18:22:30,178 - Epoch: [146][ 110/ 1207] Overall Loss 0.196944 Objective Loss 0.196944 LR 0.000250 Time 0.025233 -2023-02-13 18:22:30,366 - Epoch: [146][ 120/ 1207] Overall Loss 0.195638 Objective Loss 0.195638 LR 0.000250 Time 0.024697 -2023-02-13 18:22:30,555 - Epoch: [146][ 130/ 1207] Overall Loss 0.198132 Objective Loss 0.198132 LR 0.000250 Time 0.024249 -2023-02-13 18:22:30,744 - Epoch: [146][ 140/ 1207] Overall Loss 0.197966 Objective Loss 0.197966 LR 0.000250 Time 0.023859 -2023-02-13 18:22:30,933 - Epoch: [146][ 150/ 1207] Overall Loss 0.197996 Objective Loss 0.197996 LR 0.000250 Time 0.023529 -2023-02-13 18:22:31,121 - Epoch: [146][ 160/ 1207] Overall Loss 0.196553 Objective Loss 0.196553 LR 0.000250 Time 0.023234 -2023-02-13 18:22:31,310 - Epoch: [146][ 170/ 1207] Overall Loss 0.196516 Objective Loss 0.196516 LR 0.000250 Time 0.022974 -2023-02-13 18:22:31,499 - Epoch: [146][ 180/ 1207] Overall Loss 0.196332 Objective Loss 0.196332 LR 0.000250 Time 0.022745 -2023-02-13 18:22:31,688 - Epoch: [146][ 190/ 1207] Overall Loss 0.196362 Objective Loss 0.196362 LR 0.000250 Time 0.022539 -2023-02-13 18:22:31,875 - Epoch: [146][ 200/ 1207] Overall Loss 0.196623 Objective Loss 0.196623 LR 0.000250 Time 0.022349 -2023-02-13 18:22:32,064 - Epoch: [146][ 210/ 1207] Overall Loss 0.197171 Objective Loss 0.197171 LR 0.000250 Time 0.022181 -2023-02-13 18:22:32,253 - Epoch: [146][ 220/ 1207] Overall Loss 0.196468 Objective Loss 0.196468 LR 0.000250 Time 0.022030 -2023-02-13 18:22:32,442 - Epoch: [146][ 230/ 1207] Overall Loss 0.196543 Objective Loss 0.196543 LR 0.000250 Time 0.021893 -2023-02-13 18:22:32,631 - Epoch: [146][ 240/ 1207] Overall Loss 0.196918 Objective Loss 0.196918 LR 0.000250 Time 0.021767 -2023-02-13 18:22:32,819 - Epoch: [146][ 250/ 1207] Overall Loss 0.196117 Objective Loss 0.196117 LR 0.000250 Time 0.021648 -2023-02-13 18:22:33,008 - Epoch: [146][ 260/ 1207] Overall Loss 0.196617 Objective Loss 0.196617 LR 0.000250 Time 0.021538 -2023-02-13 18:22:33,196 - Epoch: [146][ 270/ 1207] Overall Loss 0.197670 Objective Loss 0.197670 LR 0.000250 Time 0.021438 -2023-02-13 18:22:33,386 - Epoch: [146][ 280/ 1207] Overall Loss 0.198225 Objective Loss 0.198225 LR 0.000250 Time 0.021348 -2023-02-13 18:22:33,575 - Epoch: [146][ 290/ 1207] Overall Loss 0.198311 Objective Loss 0.198311 LR 0.000250 Time 0.021264 -2023-02-13 18:22:33,764 - Epoch: [146][ 300/ 1207] Overall Loss 0.199248 Objective Loss 0.199248 LR 0.000250 Time 0.021183 -2023-02-13 18:22:33,952 - Epoch: [146][ 310/ 1207] Overall Loss 0.199117 Objective Loss 0.199117 LR 0.000250 Time 0.021105 -2023-02-13 18:22:34,141 - Epoch: [146][ 320/ 1207] Overall Loss 0.199572 Objective Loss 0.199572 LR 0.000250 Time 0.021035 -2023-02-13 18:22:34,329 - Epoch: [146][ 330/ 1207] Overall Loss 0.199996 Objective Loss 0.199996 LR 0.000250 Time 0.020967 -2023-02-13 18:22:34,518 - Epoch: [146][ 340/ 1207] Overall Loss 0.199870 Objective Loss 0.199870 LR 0.000250 Time 0.020905 -2023-02-13 18:22:34,707 - Epoch: [146][ 350/ 1207] Overall Loss 0.199557 Objective Loss 0.199557 LR 0.000250 Time 0.020846 -2023-02-13 18:22:34,895 - Epoch: [146][ 360/ 1207] Overall Loss 0.199941 Objective Loss 0.199941 LR 0.000250 Time 0.020789 -2023-02-13 18:22:35,083 - Epoch: [146][ 370/ 1207] Overall Loss 0.200239 Objective Loss 0.200239 LR 0.000250 Time 0.020734 -2023-02-13 18:22:35,271 - Epoch: [146][ 380/ 1207] Overall Loss 0.200555 Objective Loss 0.200555 LR 0.000250 Time 0.020682 -2023-02-13 18:22:35,460 - Epoch: [146][ 390/ 1207] Overall Loss 0.200005 Objective Loss 0.200005 LR 0.000250 Time 0.020635 -2023-02-13 18:22:35,649 - Epoch: [146][ 400/ 1207] Overall Loss 0.199603 Objective Loss 0.199603 LR 0.000250 Time 0.020590 -2023-02-13 18:22:35,837 - Epoch: [146][ 410/ 1207] Overall Loss 0.200001 Objective Loss 0.200001 LR 0.000250 Time 0.020546 -2023-02-13 18:22:36,026 - Epoch: [146][ 420/ 1207] Overall Loss 0.200233 Objective Loss 0.200233 LR 0.000250 Time 0.020506 -2023-02-13 18:22:36,214 - Epoch: [146][ 430/ 1207] Overall Loss 0.199972 Objective Loss 0.199972 LR 0.000250 Time 0.020466 -2023-02-13 18:22:36,403 - Epoch: [146][ 440/ 1207] Overall Loss 0.199607 Objective Loss 0.199607 LR 0.000250 Time 0.020428 -2023-02-13 18:22:36,592 - Epoch: [146][ 450/ 1207] Overall Loss 0.199215 Objective Loss 0.199215 LR 0.000250 Time 0.020393 -2023-02-13 18:22:36,781 - Epoch: [146][ 460/ 1207] Overall Loss 0.199013 Objective Loss 0.199013 LR 0.000250 Time 0.020360 -2023-02-13 18:22:36,970 - Epoch: [146][ 470/ 1207] Overall Loss 0.199592 Objective Loss 0.199592 LR 0.000250 Time 0.020328 -2023-02-13 18:22:37,158 - Epoch: [146][ 480/ 1207] Overall Loss 0.199762 Objective Loss 0.199762 LR 0.000250 Time 0.020297 -2023-02-13 18:22:37,347 - Epoch: [146][ 490/ 1207] Overall Loss 0.199909 Objective Loss 0.199909 LR 0.000250 Time 0.020267 -2023-02-13 18:22:37,536 - Epoch: [146][ 500/ 1207] Overall Loss 0.199968 Objective Loss 0.199968 LR 0.000250 Time 0.020239 -2023-02-13 18:22:37,725 - Epoch: [146][ 510/ 1207] Overall Loss 0.199914 Objective Loss 0.199914 LR 0.000250 Time 0.020213 -2023-02-13 18:22:37,914 - Epoch: [146][ 520/ 1207] Overall Loss 0.199530 Objective Loss 0.199530 LR 0.000250 Time 0.020186 -2023-02-13 18:22:38,102 - Epoch: [146][ 530/ 1207] Overall Loss 0.199237 Objective Loss 0.199237 LR 0.000250 Time 0.020160 -2023-02-13 18:22:38,291 - Epoch: [146][ 540/ 1207] Overall Loss 0.199050 Objective Loss 0.199050 LR 0.000250 Time 0.020136 -2023-02-13 18:22:38,480 - Epoch: [146][ 550/ 1207] Overall Loss 0.199846 Objective Loss 0.199846 LR 0.000250 Time 0.020112 -2023-02-13 18:22:38,669 - Epoch: [146][ 560/ 1207] Overall Loss 0.200207 Objective Loss 0.200207 LR 0.000250 Time 0.020091 -2023-02-13 18:22:38,859 - Epoch: [146][ 570/ 1207] Overall Loss 0.200364 Objective Loss 0.200364 LR 0.000250 Time 0.020069 -2023-02-13 18:22:39,048 - Epoch: [146][ 580/ 1207] Overall Loss 0.200742 Objective Loss 0.200742 LR 0.000250 Time 0.020049 -2023-02-13 18:22:39,237 - Epoch: [146][ 590/ 1207] Overall Loss 0.200666 Objective Loss 0.200666 LR 0.000250 Time 0.020029 -2023-02-13 18:22:39,425 - Epoch: [146][ 600/ 1207] Overall Loss 0.201005 Objective Loss 0.201005 LR 0.000250 Time 0.020008 -2023-02-13 18:22:39,615 - Epoch: [146][ 610/ 1207] Overall Loss 0.200997 Objective Loss 0.200997 LR 0.000250 Time 0.019990 -2023-02-13 18:22:39,803 - Epoch: [146][ 620/ 1207] Overall Loss 0.200787 Objective Loss 0.200787 LR 0.000250 Time 0.019972 -2023-02-13 18:22:39,993 - Epoch: [146][ 630/ 1207] Overall Loss 0.200895 Objective Loss 0.200895 LR 0.000250 Time 0.019955 -2023-02-13 18:22:40,181 - Epoch: [146][ 640/ 1207] Overall Loss 0.200959 Objective Loss 0.200959 LR 0.000250 Time 0.019937 -2023-02-13 18:22:40,371 - Epoch: [146][ 650/ 1207] Overall Loss 0.201304 Objective Loss 0.201304 LR 0.000250 Time 0.019922 -2023-02-13 18:22:40,561 - Epoch: [146][ 660/ 1207] Overall Loss 0.201410 Objective Loss 0.201410 LR 0.000250 Time 0.019907 -2023-02-13 18:22:40,751 - Epoch: [146][ 670/ 1207] Overall Loss 0.201767 Objective Loss 0.201767 LR 0.000250 Time 0.019892 -2023-02-13 18:22:40,942 - Epoch: [146][ 680/ 1207] Overall Loss 0.201695 Objective Loss 0.201695 LR 0.000250 Time 0.019881 -2023-02-13 18:22:41,131 - Epoch: [146][ 690/ 1207] Overall Loss 0.201991 Objective Loss 0.201991 LR 0.000250 Time 0.019866 -2023-02-13 18:22:41,321 - Epoch: [146][ 700/ 1207] Overall Loss 0.202189 Objective Loss 0.202189 LR 0.000250 Time 0.019852 -2023-02-13 18:22:41,509 - Epoch: [146][ 710/ 1207] Overall Loss 0.202521 Objective Loss 0.202521 LR 0.000250 Time 0.019838 -2023-02-13 18:22:41,699 - Epoch: [146][ 720/ 1207] Overall Loss 0.202524 Objective Loss 0.202524 LR 0.000250 Time 0.019825 -2023-02-13 18:22:41,888 - Epoch: [146][ 730/ 1207] Overall Loss 0.202739 Objective Loss 0.202739 LR 0.000250 Time 0.019812 -2023-02-13 18:22:42,077 - Epoch: [146][ 740/ 1207] Overall Loss 0.202737 Objective Loss 0.202737 LR 0.000250 Time 0.019799 -2023-02-13 18:22:42,266 - Epoch: [146][ 750/ 1207] Overall Loss 0.202848 Objective Loss 0.202848 LR 0.000250 Time 0.019787 -2023-02-13 18:22:42,454 - Epoch: [146][ 760/ 1207] Overall Loss 0.202508 Objective Loss 0.202508 LR 0.000250 Time 0.019773 -2023-02-13 18:22:42,643 - Epoch: [146][ 770/ 1207] Overall Loss 0.202657 Objective Loss 0.202657 LR 0.000250 Time 0.019762 -2023-02-13 18:22:42,832 - Epoch: [146][ 780/ 1207] Overall Loss 0.202837 Objective Loss 0.202837 LR 0.000250 Time 0.019750 -2023-02-13 18:22:43,021 - Epoch: [146][ 790/ 1207] Overall Loss 0.202906 Objective Loss 0.202906 LR 0.000250 Time 0.019739 -2023-02-13 18:22:43,210 - Epoch: [146][ 800/ 1207] Overall Loss 0.202831 Objective Loss 0.202831 LR 0.000250 Time 0.019728 -2023-02-13 18:22:43,400 - Epoch: [146][ 810/ 1207] Overall Loss 0.202832 Objective Loss 0.202832 LR 0.000250 Time 0.019718 -2023-02-13 18:22:43,589 - Epoch: [146][ 820/ 1207] Overall Loss 0.202883 Objective Loss 0.202883 LR 0.000250 Time 0.019708 -2023-02-13 18:22:43,778 - Epoch: [146][ 830/ 1207] Overall Loss 0.203019 Objective Loss 0.203019 LR 0.000250 Time 0.019698 -2023-02-13 18:22:43,968 - Epoch: [146][ 840/ 1207] Overall Loss 0.202909 Objective Loss 0.202909 LR 0.000250 Time 0.019689 -2023-02-13 18:22:44,157 - Epoch: [146][ 850/ 1207] Overall Loss 0.202806 Objective Loss 0.202806 LR 0.000250 Time 0.019679 -2023-02-13 18:22:44,346 - Epoch: [146][ 860/ 1207] Overall Loss 0.203307 Objective Loss 0.203307 LR 0.000250 Time 0.019670 -2023-02-13 18:22:44,536 - Epoch: [146][ 870/ 1207] Overall Loss 0.203122 Objective Loss 0.203122 LR 0.000250 Time 0.019661 -2023-02-13 18:22:44,725 - Epoch: [146][ 880/ 1207] Overall Loss 0.203132 Objective Loss 0.203132 LR 0.000250 Time 0.019652 -2023-02-13 18:22:44,914 - Epoch: [146][ 890/ 1207] Overall Loss 0.202975 Objective Loss 0.202975 LR 0.000250 Time 0.019644 -2023-02-13 18:22:45,103 - Epoch: [146][ 900/ 1207] Overall Loss 0.202868 Objective Loss 0.202868 LR 0.000250 Time 0.019635 -2023-02-13 18:22:45,292 - Epoch: [146][ 910/ 1207] Overall Loss 0.202817 Objective Loss 0.202817 LR 0.000250 Time 0.019627 -2023-02-13 18:22:45,481 - Epoch: [146][ 920/ 1207] Overall Loss 0.202928 Objective Loss 0.202928 LR 0.000250 Time 0.019618 -2023-02-13 18:22:45,670 - Epoch: [146][ 930/ 1207] Overall Loss 0.202828 Objective Loss 0.202828 LR 0.000250 Time 0.019610 -2023-02-13 18:22:45,859 - Epoch: [146][ 940/ 1207] Overall Loss 0.202943 Objective Loss 0.202943 LR 0.000250 Time 0.019602 -2023-02-13 18:22:46,050 - Epoch: [146][ 950/ 1207] Overall Loss 0.202797 Objective Loss 0.202797 LR 0.000250 Time 0.019596 -2023-02-13 18:22:46,239 - Epoch: [146][ 960/ 1207] Overall Loss 0.203100 Objective Loss 0.203100 LR 0.000250 Time 0.019589 -2023-02-13 18:22:46,428 - Epoch: [146][ 970/ 1207] Overall Loss 0.203059 Objective Loss 0.203059 LR 0.000250 Time 0.019581 -2023-02-13 18:22:46,618 - Epoch: [146][ 980/ 1207] Overall Loss 0.203084 Objective Loss 0.203084 LR 0.000250 Time 0.019575 -2023-02-13 18:22:46,808 - Epoch: [146][ 990/ 1207] Overall Loss 0.203247 Objective Loss 0.203247 LR 0.000250 Time 0.019568 -2023-02-13 18:22:46,997 - Epoch: [146][ 1000/ 1207] Overall Loss 0.203227 Objective Loss 0.203227 LR 0.000250 Time 0.019562 -2023-02-13 18:22:47,186 - Epoch: [146][ 1010/ 1207] Overall Loss 0.203037 Objective Loss 0.203037 LR 0.000250 Time 0.019555 -2023-02-13 18:22:47,376 - Epoch: [146][ 1020/ 1207] Overall Loss 0.203203 Objective Loss 0.203203 LR 0.000250 Time 0.019549 -2023-02-13 18:22:47,566 - Epoch: [146][ 1030/ 1207] Overall Loss 0.203260 Objective Loss 0.203260 LR 0.000250 Time 0.019543 -2023-02-13 18:22:47,755 - Epoch: [146][ 1040/ 1207] Overall Loss 0.203600 Objective Loss 0.203600 LR 0.000250 Time 0.019537 -2023-02-13 18:22:47,944 - Epoch: [146][ 1050/ 1207] Overall Loss 0.203718 Objective Loss 0.203718 LR 0.000250 Time 0.019530 -2023-02-13 18:22:48,133 - Epoch: [146][ 1060/ 1207] Overall Loss 0.203736 Objective Loss 0.203736 LR 0.000250 Time 0.019524 -2023-02-13 18:22:48,322 - Epoch: [146][ 1070/ 1207] Overall Loss 0.203557 Objective Loss 0.203557 LR 0.000250 Time 0.019518 -2023-02-13 18:22:48,510 - Epoch: [146][ 1080/ 1207] Overall Loss 0.203608 Objective Loss 0.203608 LR 0.000250 Time 0.019511 -2023-02-13 18:22:48,700 - Epoch: [146][ 1090/ 1207] Overall Loss 0.203639 Objective Loss 0.203639 LR 0.000250 Time 0.019506 -2023-02-13 18:22:48,888 - Epoch: [146][ 1100/ 1207] Overall Loss 0.203678 Objective Loss 0.203678 LR 0.000250 Time 0.019499 -2023-02-13 18:22:49,078 - Epoch: [146][ 1110/ 1207] Overall Loss 0.203527 Objective Loss 0.203527 LR 0.000250 Time 0.019494 -2023-02-13 18:22:49,267 - Epoch: [146][ 1120/ 1207] Overall Loss 0.203623 Objective Loss 0.203623 LR 0.000250 Time 0.019488 -2023-02-13 18:22:49,456 - Epoch: [146][ 1130/ 1207] Overall Loss 0.203519 Objective Loss 0.203519 LR 0.000250 Time 0.019483 -2023-02-13 18:22:49,645 - Epoch: [146][ 1140/ 1207] Overall Loss 0.203205 Objective Loss 0.203205 LR 0.000250 Time 0.019478 -2023-02-13 18:22:49,834 - Epoch: [146][ 1150/ 1207] Overall Loss 0.202948 Objective Loss 0.202948 LR 0.000250 Time 0.019472 -2023-02-13 18:22:50,023 - Epoch: [146][ 1160/ 1207] Overall Loss 0.202973 Objective Loss 0.202973 LR 0.000250 Time 0.019467 -2023-02-13 18:22:50,213 - Epoch: [146][ 1170/ 1207] Overall Loss 0.203068 Objective Loss 0.203068 LR 0.000250 Time 0.019463 -2023-02-13 18:22:50,402 - Epoch: [146][ 1180/ 1207] Overall Loss 0.203223 Objective Loss 0.203223 LR 0.000250 Time 0.019458 -2023-02-13 18:22:50,591 - Epoch: [146][ 1190/ 1207] Overall Loss 0.203247 Objective Loss 0.203247 LR 0.000250 Time 0.019453 -2023-02-13 18:22:50,839 - Epoch: [146][ 1200/ 1207] Overall Loss 0.203196 Objective Loss 0.203196 LR 0.000250 Time 0.019496 -2023-02-13 18:22:50,955 - Epoch: [146][ 1207/ 1207] Overall Loss 0.203167 Objective Loss 0.203167 Top1 88.109756 Top5 98.475610 LR 0.000250 Time 0.019480 -2023-02-13 18:22:51,028 - --- validate (epoch=146)----------- -2023-02-13 18:22:51,028 - 34311 samples (256 per mini-batch) -2023-02-13 18:22:51,537 - Epoch: [146][ 10/ 135] Loss 0.297841 Top1 84.804688 Top5 98.437500 -2023-02-13 18:22:51,662 - Epoch: [146][ 20/ 135] Loss 0.305946 Top1 84.726562 Top5 98.183594 -2023-02-13 18:22:51,785 - Epoch: [146][ 30/ 135] Loss 0.301527 Top1 84.778646 Top5 98.098958 -2023-02-13 18:22:51,911 - Epoch: [146][ 40/ 135] Loss 0.299799 Top1 84.843750 Top5 98.076172 -2023-02-13 18:22:52,039 - Epoch: [146][ 50/ 135] Loss 0.303893 Top1 84.851562 Top5 98.046875 -2023-02-13 18:22:52,166 - Epoch: [146][ 60/ 135] Loss 0.301233 Top1 85.078125 Top5 98.118490 -2023-02-13 18:22:52,293 - Epoch: [146][ 70/ 135] Loss 0.302922 Top1 85.083705 Top5 98.097098 -2023-02-13 18:22:52,420 - Epoch: [146][ 80/ 135] Loss 0.304399 Top1 84.960938 Top5 98.051758 -2023-02-13 18:22:52,560 - Epoch: [146][ 90/ 135] Loss 0.303089 Top1 85.034722 Top5 97.999132 -2023-02-13 18:22:52,686 - Epoch: [146][ 100/ 135] Loss 0.302061 Top1 84.914062 Top5 97.980469 -2023-02-13 18:22:52,810 - Epoch: [146][ 110/ 135] Loss 0.299990 Top1 84.889915 Top5 97.943892 -2023-02-13 18:22:52,937 - Epoch: [146][ 120/ 135] Loss 0.301762 Top1 84.882812 Top5 97.913411 -2023-02-13 18:22:53,063 - Epoch: [146][ 130/ 135] Loss 0.300266 Top1 84.939904 Top5 97.914663 -2023-02-13 18:22:53,109 - Epoch: [146][ 135/ 135] Loss 0.296493 Top1 84.993151 Top5 97.916120 -2023-02-13 18:22:53,180 - ==> Top1: 84.993 Top5: 97.916 Loss: 0.296 - -2023-02-13 18:22:53,181 - ==> Confusion: -[[ 860 5 4 0 9 2 0 2 3 45 1 3 4 3 7 5 2 3 2 1 6] - [ 1 966 0 1 5 20 1 19 3 0 1 0 1 0 1 2 4 0 2 1 5] - [ 7 5 960 13 4 1 16 12 0 1 4 2 3 6 1 6 1 3 7 3 3] - [ 5 2 17 924 3 4 0 1 1 2 7 0 5 0 13 4 3 4 16 0 5] - [ 12 12 0 1 994 8 1 0 1 1 0 6 2 4 6 9 6 0 0 1 2] - [ 0 17 0 4 9 973 3 18 1 6 1 13 1 9 0 2 4 0 1 3 5] - [ 2 6 16 2 0 4 1040 3 1 1 2 1 2 1 0 4 0 3 2 6 3] - [ 1 10 8 2 2 25 3 942 0 2 4 3 3 1 0 1 0 1 11 4 1] - [ 15 4 1 1 1 0 1 1 907 39 7 3 0 5 12 3 1 0 4 1 3] - [ 79 2 3 0 8 3 0 4 28 864 0 1 0 10 3 0 1 3 0 1 2] - [ 2 5 3 11 1 2 3 5 15 2 976 3 1 5 2 0 2 0 11 0 2] - [ 1 4 3 0 2 15 0 6 2 1 0 915 25 5 1 4 2 8 4 6 1] - [ 0 0 1 10 1 5 0 0 2 1 0 26 871 1 1 8 3 16 5 0 8] - [ 4 3 2 1 10 12 0 3 17 19 5 8 2 921 1 7 2 2 0 1 4] - [ 11 1 0 22 7 4 0 1 16 7 2 0 5 1 988 0 3 7 7 0 10] - [ 4 3 3 0 7 2 6 1 1 0 0 4 8 1 0 973 9 10 0 9 5] - [ 5 5 0 1 6 0 0 0 2 1 0 1 3 1 0 9 1013 1 1 2 10] - [ 2 4 0 3 1 2 1 0 0 0 3 7 10 0 0 14 0 996 1 3 4] - [ 4 5 2 7 1 1 0 29 3 1 3 0 2 0 11 1 1 4 1005 3 3] - [ 0 3 2 1 2 9 6 18 1 0 0 12 2 4 0 5 3 2 1 1071 6] - [ 149 270 217 138 148 238 99 206 102 76 165 107 304 240 135 107 234 113 170 213 10003]] - -2023-02-13 18:22:53,182 - ==> Best [Top1: 85.701 Top5: 97.983 Sparsity:0.00 Params: 148928 on epoch: 145] -2023-02-13 18:22:53,182 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:22:53,188 - - -2023-02-13 18:22:53,188 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:22:54,078 - Epoch: [147][ 10/ 1207] Overall Loss 0.200029 Objective Loss 0.200029 LR 0.000250 Time 0.088874 -2023-02-13 18:22:54,269 - Epoch: [147][ 20/ 1207] Overall Loss 0.202297 Objective Loss 0.202297 LR 0.000250 Time 0.053970 -2023-02-13 18:22:54,456 - Epoch: [147][ 30/ 1207] Overall Loss 0.205603 Objective Loss 0.205603 LR 0.000250 Time 0.042213 -2023-02-13 18:22:54,643 - Epoch: [147][ 40/ 1207] Overall Loss 0.202974 Objective Loss 0.202974 LR 0.000250 Time 0.036323 -2023-02-13 18:22:54,831 - Epoch: [147][ 50/ 1207] Overall Loss 0.198033 Objective Loss 0.198033 LR 0.000250 Time 0.032802 -2023-02-13 18:22:55,018 - Epoch: [147][ 60/ 1207] Overall Loss 0.200133 Objective Loss 0.200133 LR 0.000250 Time 0.030442 -2023-02-13 18:22:55,205 - Epoch: [147][ 70/ 1207] Overall Loss 0.198798 Objective Loss 0.198798 LR 0.000250 Time 0.028764 -2023-02-13 18:22:55,392 - Epoch: [147][ 80/ 1207] Overall Loss 0.200484 Objective Loss 0.200484 LR 0.000250 Time 0.027499 -2023-02-13 18:22:55,579 - Epoch: [147][ 90/ 1207] Overall Loss 0.199639 Objective Loss 0.199639 LR 0.000250 Time 0.026524 -2023-02-13 18:22:55,767 - Epoch: [147][ 100/ 1207] Overall Loss 0.202158 Objective Loss 0.202158 LR 0.000250 Time 0.025741 -2023-02-13 18:22:55,955 - Epoch: [147][ 110/ 1207] Overall Loss 0.206551 Objective Loss 0.206551 LR 0.000250 Time 0.025106 -2023-02-13 18:22:56,142 - Epoch: [147][ 120/ 1207] Overall Loss 0.206781 Objective Loss 0.206781 LR 0.000250 Time 0.024571 -2023-02-13 18:22:56,329 - Epoch: [147][ 130/ 1207] Overall Loss 0.206865 Objective Loss 0.206865 LR 0.000250 Time 0.024120 -2023-02-13 18:22:56,517 - Epoch: [147][ 140/ 1207] Overall Loss 0.206696 Objective Loss 0.206696 LR 0.000250 Time 0.023734 -2023-02-13 18:22:56,705 - Epoch: [147][ 150/ 1207] Overall Loss 0.205727 Objective Loss 0.205727 LR 0.000250 Time 0.023402 -2023-02-13 18:22:56,891 - Epoch: [147][ 160/ 1207] Overall Loss 0.205280 Objective Loss 0.205280 LR 0.000250 Time 0.023103 -2023-02-13 18:22:57,078 - Epoch: [147][ 170/ 1207] Overall Loss 0.203386 Objective Loss 0.203386 LR 0.000250 Time 0.022844 -2023-02-13 18:22:57,266 - Epoch: [147][ 180/ 1207] Overall Loss 0.203806 Objective Loss 0.203806 LR 0.000250 Time 0.022615 -2023-02-13 18:22:57,453 - Epoch: [147][ 190/ 1207] Overall Loss 0.203997 Objective Loss 0.203997 LR 0.000250 Time 0.022407 -2023-02-13 18:22:57,641 - Epoch: [147][ 200/ 1207] Overall Loss 0.204510 Objective Loss 0.204510 LR 0.000250 Time 0.022224 -2023-02-13 18:22:57,829 - Epoch: [147][ 210/ 1207] Overall Loss 0.204002 Objective Loss 0.204002 LR 0.000250 Time 0.022058 -2023-02-13 18:22:58,015 - Epoch: [147][ 220/ 1207] Overall Loss 0.204134 Objective Loss 0.204134 LR 0.000250 Time 0.021902 -2023-02-13 18:22:58,202 - Epoch: [147][ 230/ 1207] Overall Loss 0.204385 Objective Loss 0.204385 LR 0.000250 Time 0.021761 -2023-02-13 18:22:58,390 - Epoch: [147][ 240/ 1207] Overall Loss 0.204173 Objective Loss 0.204173 LR 0.000250 Time 0.021633 -2023-02-13 18:22:58,577 - Epoch: [147][ 250/ 1207] Overall Loss 0.203123 Objective Loss 0.203123 LR 0.000250 Time 0.021515 -2023-02-13 18:22:58,765 - Epoch: [147][ 260/ 1207] Overall Loss 0.203249 Objective Loss 0.203249 LR 0.000250 Time 0.021409 -2023-02-13 18:22:58,953 - Epoch: [147][ 270/ 1207] Overall Loss 0.203073 Objective Loss 0.203073 LR 0.000250 Time 0.021310 -2023-02-13 18:22:59,139 - Epoch: [147][ 280/ 1207] Overall Loss 0.203058 Objective Loss 0.203058 LR 0.000250 Time 0.021215 -2023-02-13 18:22:59,327 - Epoch: [147][ 290/ 1207] Overall Loss 0.201950 Objective Loss 0.201950 LR 0.000250 Time 0.021130 -2023-02-13 18:22:59,514 - Epoch: [147][ 300/ 1207] Overall Loss 0.201347 Objective Loss 0.201347 LR 0.000250 Time 0.021048 -2023-02-13 18:22:59,702 - Epoch: [147][ 310/ 1207] Overall Loss 0.200989 Objective Loss 0.200989 LR 0.000250 Time 0.020974 -2023-02-13 18:22:59,889 - Epoch: [147][ 320/ 1207] Overall Loss 0.200978 Objective Loss 0.200978 LR 0.000250 Time 0.020902 -2023-02-13 18:23:00,076 - Epoch: [147][ 330/ 1207] Overall Loss 0.200505 Objective Loss 0.200505 LR 0.000250 Time 0.020834 -2023-02-13 18:23:00,263 - Epoch: [147][ 340/ 1207] Overall Loss 0.200511 Objective Loss 0.200511 LR 0.000250 Time 0.020770 -2023-02-13 18:23:00,451 - Epoch: [147][ 350/ 1207] Overall Loss 0.200388 Objective Loss 0.200388 LR 0.000250 Time 0.020711 -2023-02-13 18:23:00,637 - Epoch: [147][ 360/ 1207] Overall Loss 0.200101 Objective Loss 0.200101 LR 0.000250 Time 0.020653 -2023-02-13 18:23:00,825 - Epoch: [147][ 370/ 1207] Overall Loss 0.200887 Objective Loss 0.200887 LR 0.000250 Time 0.020601 -2023-02-13 18:23:01,013 - Epoch: [147][ 380/ 1207] Overall Loss 0.200980 Objective Loss 0.200980 LR 0.000250 Time 0.020552 -2023-02-13 18:23:01,199 - Epoch: [147][ 390/ 1207] Overall Loss 0.201435 Objective Loss 0.201435 LR 0.000250 Time 0.020503 -2023-02-13 18:23:01,386 - Epoch: [147][ 400/ 1207] Overall Loss 0.201699 Objective Loss 0.201699 LR 0.000250 Time 0.020456 -2023-02-13 18:23:01,573 - Epoch: [147][ 410/ 1207] Overall Loss 0.201691 Objective Loss 0.201691 LR 0.000250 Time 0.020412 -2023-02-13 18:23:01,760 - Epoch: [147][ 420/ 1207] Overall Loss 0.202045 Objective Loss 0.202045 LR 0.000250 Time 0.020372 -2023-02-13 18:23:01,948 - Epoch: [147][ 430/ 1207] Overall Loss 0.201677 Objective Loss 0.201677 LR 0.000250 Time 0.020334 -2023-02-13 18:23:02,136 - Epoch: [147][ 440/ 1207] Overall Loss 0.202154 Objective Loss 0.202154 LR 0.000250 Time 0.020297 -2023-02-13 18:23:02,323 - Epoch: [147][ 450/ 1207] Overall Loss 0.201749 Objective Loss 0.201749 LR 0.000250 Time 0.020261 -2023-02-13 18:23:02,510 - Epoch: [147][ 460/ 1207] Overall Loss 0.201709 Objective Loss 0.201709 LR 0.000250 Time 0.020227 -2023-02-13 18:23:02,697 - Epoch: [147][ 470/ 1207] Overall Loss 0.201776 Objective Loss 0.201776 LR 0.000250 Time 0.020193 -2023-02-13 18:23:02,884 - Epoch: [147][ 480/ 1207] Overall Loss 0.201290 Objective Loss 0.201290 LR 0.000250 Time 0.020162 -2023-02-13 18:23:03,071 - Epoch: [147][ 490/ 1207] Overall Loss 0.201131 Objective Loss 0.201131 LR 0.000250 Time 0.020131 -2023-02-13 18:23:03,258 - Epoch: [147][ 500/ 1207] Overall Loss 0.201324 Objective Loss 0.201324 LR 0.000250 Time 0.020102 -2023-02-13 18:23:03,450 - Epoch: [147][ 510/ 1207] Overall Loss 0.201400 Objective Loss 0.201400 LR 0.000250 Time 0.020084 -2023-02-13 18:23:03,647 - Epoch: [147][ 520/ 1207] Overall Loss 0.201487 Objective Loss 0.201487 LR 0.000250 Time 0.020075 -2023-02-13 18:23:03,840 - Epoch: [147][ 530/ 1207] Overall Loss 0.201511 Objective Loss 0.201511 LR 0.000250 Time 0.020059 -2023-02-13 18:23:04,035 - Epoch: [147][ 540/ 1207] Overall Loss 0.201095 Objective Loss 0.201095 LR 0.000250 Time 0.020049 -2023-02-13 18:23:04,227 - Epoch: [147][ 550/ 1207] Overall Loss 0.200837 Objective Loss 0.200837 LR 0.000250 Time 0.020033 -2023-02-13 18:23:04,422 - Epoch: [147][ 560/ 1207] Overall Loss 0.201186 Objective Loss 0.201186 LR 0.000250 Time 0.020022 -2023-02-13 18:23:04,614 - Epoch: [147][ 570/ 1207] Overall Loss 0.200807 Objective Loss 0.200807 LR 0.000250 Time 0.020008 -2023-02-13 18:23:04,810 - Epoch: [147][ 580/ 1207] Overall Loss 0.201474 Objective Loss 0.201474 LR 0.000250 Time 0.019999 -2023-02-13 18:23:05,001 - Epoch: [147][ 590/ 1207] Overall Loss 0.201809 Objective Loss 0.201809 LR 0.000250 Time 0.019984 -2023-02-13 18:23:05,196 - Epoch: [147][ 600/ 1207] Overall Loss 0.201607 Objective Loss 0.201607 LR 0.000250 Time 0.019976 -2023-02-13 18:23:05,388 - Epoch: [147][ 610/ 1207] Overall Loss 0.201852 Objective Loss 0.201852 LR 0.000250 Time 0.019961 -2023-02-13 18:23:05,582 - Epoch: [147][ 620/ 1207] Overall Loss 0.201145 Objective Loss 0.201145 LR 0.000250 Time 0.019953 -2023-02-13 18:23:05,775 - Epoch: [147][ 630/ 1207] Overall Loss 0.201107 Objective Loss 0.201107 LR 0.000250 Time 0.019941 -2023-02-13 18:23:05,971 - Epoch: [147][ 640/ 1207] Overall Loss 0.201014 Objective Loss 0.201014 LR 0.000250 Time 0.019935 -2023-02-13 18:23:06,163 - Epoch: [147][ 650/ 1207] Overall Loss 0.201094 Objective Loss 0.201094 LR 0.000250 Time 0.019924 -2023-02-13 18:23:06,359 - Epoch: [147][ 660/ 1207] Overall Loss 0.201130 Objective Loss 0.201130 LR 0.000250 Time 0.019918 -2023-02-13 18:23:06,551 - Epoch: [147][ 670/ 1207] Overall Loss 0.201336 Objective Loss 0.201336 LR 0.000250 Time 0.019907 -2023-02-13 18:23:06,746 - Epoch: [147][ 680/ 1207] Overall Loss 0.201408 Objective Loss 0.201408 LR 0.000250 Time 0.019901 -2023-02-13 18:23:06,940 - Epoch: [147][ 690/ 1207] Overall Loss 0.201684 Objective Loss 0.201684 LR 0.000250 Time 0.019892 -2023-02-13 18:23:07,135 - Epoch: [147][ 700/ 1207] Overall Loss 0.201647 Objective Loss 0.201647 LR 0.000250 Time 0.019886 -2023-02-13 18:23:07,327 - Epoch: [147][ 710/ 1207] Overall Loss 0.201704 Objective Loss 0.201704 LR 0.000250 Time 0.019876 -2023-02-13 18:23:07,522 - Epoch: [147][ 720/ 1207] Overall Loss 0.201827 Objective Loss 0.201827 LR 0.000250 Time 0.019871 -2023-02-13 18:23:07,715 - Epoch: [147][ 730/ 1207] Overall Loss 0.202342 Objective Loss 0.202342 LR 0.000250 Time 0.019861 -2023-02-13 18:23:07,910 - Epoch: [147][ 740/ 1207] Overall Loss 0.202165 Objective Loss 0.202165 LR 0.000250 Time 0.019856 -2023-02-13 18:23:08,102 - Epoch: [147][ 750/ 1207] Overall Loss 0.201983 Objective Loss 0.201983 LR 0.000250 Time 0.019847 -2023-02-13 18:23:08,298 - Epoch: [147][ 760/ 1207] Overall Loss 0.202235 Objective Loss 0.202235 LR 0.000250 Time 0.019843 -2023-02-13 18:23:08,490 - Epoch: [147][ 770/ 1207] Overall Loss 0.202016 Objective Loss 0.202016 LR 0.000250 Time 0.019835 -2023-02-13 18:23:08,685 - Epoch: [147][ 780/ 1207] Overall Loss 0.202160 Objective Loss 0.202160 LR 0.000250 Time 0.019830 -2023-02-13 18:23:08,877 - Epoch: [147][ 790/ 1207] Overall Loss 0.201999 Objective Loss 0.201999 LR 0.000250 Time 0.019822 -2023-02-13 18:23:09,072 - Epoch: [147][ 800/ 1207] Overall Loss 0.202178 Objective Loss 0.202178 LR 0.000250 Time 0.019817 -2023-02-13 18:23:09,264 - Epoch: [147][ 810/ 1207] Overall Loss 0.202277 Objective Loss 0.202277 LR 0.000250 Time 0.019808 -2023-02-13 18:23:09,458 - Epoch: [147][ 820/ 1207] Overall Loss 0.202112 Objective Loss 0.202112 LR 0.000250 Time 0.019803 -2023-02-13 18:23:09,650 - Epoch: [147][ 830/ 1207] Overall Loss 0.202610 Objective Loss 0.202610 LR 0.000250 Time 0.019796 -2023-02-13 18:23:09,846 - Epoch: [147][ 840/ 1207] Overall Loss 0.202864 Objective Loss 0.202864 LR 0.000250 Time 0.019792 -2023-02-13 18:23:10,038 - Epoch: [147][ 850/ 1207] Overall Loss 0.202953 Objective Loss 0.202953 LR 0.000250 Time 0.019785 -2023-02-13 18:23:10,233 - Epoch: [147][ 860/ 1207] Overall Loss 0.203110 Objective Loss 0.203110 LR 0.000250 Time 0.019781 -2023-02-13 18:23:10,425 - Epoch: [147][ 870/ 1207] Overall Loss 0.203073 Objective Loss 0.203073 LR 0.000250 Time 0.019775 -2023-02-13 18:23:10,620 - Epoch: [147][ 880/ 1207] Overall Loss 0.203144 Objective Loss 0.203144 LR 0.000250 Time 0.019771 -2023-02-13 18:23:10,813 - Epoch: [147][ 890/ 1207] Overall Loss 0.203236 Objective Loss 0.203236 LR 0.000250 Time 0.019765 -2023-02-13 18:23:11,006 - Epoch: [147][ 900/ 1207] Overall Loss 0.203182 Objective Loss 0.203182 LR 0.000250 Time 0.019759 -2023-02-13 18:23:11,196 - Epoch: [147][ 910/ 1207] Overall Loss 0.203095 Objective Loss 0.203095 LR 0.000250 Time 0.019751 -2023-02-13 18:23:11,387 - Epoch: [147][ 920/ 1207] Overall Loss 0.202901 Objective Loss 0.202901 LR 0.000250 Time 0.019744 -2023-02-13 18:23:11,577 - Epoch: [147][ 930/ 1207] Overall Loss 0.203076 Objective Loss 0.203076 LR 0.000250 Time 0.019735 -2023-02-13 18:23:11,768 - Epoch: [147][ 940/ 1207] Overall Loss 0.202994 Objective Loss 0.202994 LR 0.000250 Time 0.019728 -2023-02-13 18:23:11,959 - Epoch: [147][ 950/ 1207] Overall Loss 0.202942 Objective Loss 0.202942 LR 0.000250 Time 0.019721 -2023-02-13 18:23:12,149 - Epoch: [147][ 960/ 1207] Overall Loss 0.202940 Objective Loss 0.202940 LR 0.000250 Time 0.019713 -2023-02-13 18:23:12,340 - Epoch: [147][ 970/ 1207] Overall Loss 0.203009 Objective Loss 0.203009 LR 0.000250 Time 0.019706 -2023-02-13 18:23:12,529 - Epoch: [147][ 980/ 1207] Overall Loss 0.202790 Objective Loss 0.202790 LR 0.000250 Time 0.019698 -2023-02-13 18:23:12,720 - Epoch: [147][ 990/ 1207] Overall Loss 0.202948 Objective Loss 0.202948 LR 0.000250 Time 0.019691 -2023-02-13 18:23:12,910 - Epoch: [147][ 1000/ 1207] Overall Loss 0.203046 Objective Loss 0.203046 LR 0.000250 Time 0.019684 -2023-02-13 18:23:13,100 - Epoch: [147][ 1010/ 1207] Overall Loss 0.203084 Objective Loss 0.203084 LR 0.000250 Time 0.019677 -2023-02-13 18:23:13,290 - Epoch: [147][ 1020/ 1207] Overall Loss 0.203430 Objective Loss 0.203430 LR 0.000250 Time 0.019670 -2023-02-13 18:23:13,481 - Epoch: [147][ 1030/ 1207] Overall Loss 0.203415 Objective Loss 0.203415 LR 0.000250 Time 0.019664 -2023-02-13 18:23:13,672 - Epoch: [147][ 1040/ 1207] Overall Loss 0.203485 Objective Loss 0.203485 LR 0.000250 Time 0.019658 -2023-02-13 18:23:13,861 - Epoch: [147][ 1050/ 1207] Overall Loss 0.203400 Objective Loss 0.203400 LR 0.000250 Time 0.019651 -2023-02-13 18:23:14,051 - Epoch: [147][ 1060/ 1207] Overall Loss 0.203225 Objective Loss 0.203225 LR 0.000250 Time 0.019644 -2023-02-13 18:23:14,241 - Epoch: [147][ 1070/ 1207] Overall Loss 0.203312 Objective Loss 0.203312 LR 0.000250 Time 0.019638 -2023-02-13 18:23:14,431 - Epoch: [147][ 1080/ 1207] Overall Loss 0.203091 Objective Loss 0.203091 LR 0.000250 Time 0.019631 -2023-02-13 18:23:14,620 - Epoch: [147][ 1090/ 1207] Overall Loss 0.203171 Objective Loss 0.203171 LR 0.000250 Time 0.019625 -2023-02-13 18:23:14,810 - Epoch: [147][ 1100/ 1207] Overall Loss 0.203376 Objective Loss 0.203376 LR 0.000250 Time 0.019619 -2023-02-13 18:23:15,000 - Epoch: [147][ 1110/ 1207] Overall Loss 0.203479 Objective Loss 0.203479 LR 0.000250 Time 0.019612 -2023-02-13 18:23:15,189 - Epoch: [147][ 1120/ 1207] Overall Loss 0.203424 Objective Loss 0.203424 LR 0.000250 Time 0.019606 -2023-02-13 18:23:15,379 - Epoch: [147][ 1130/ 1207] Overall Loss 0.203429 Objective Loss 0.203429 LR 0.000250 Time 0.019600 -2023-02-13 18:23:15,568 - Epoch: [147][ 1140/ 1207] Overall Loss 0.203560 Objective Loss 0.203560 LR 0.000250 Time 0.019594 -2023-02-13 18:23:15,758 - Epoch: [147][ 1150/ 1207] Overall Loss 0.203571 Objective Loss 0.203571 LR 0.000250 Time 0.019588 -2023-02-13 18:23:15,949 - Epoch: [147][ 1160/ 1207] Overall Loss 0.203546 Objective Loss 0.203546 LR 0.000250 Time 0.019584 -2023-02-13 18:23:16,138 - Epoch: [147][ 1170/ 1207] Overall Loss 0.203577 Objective Loss 0.203577 LR 0.000250 Time 0.019578 -2023-02-13 18:23:16,328 - Epoch: [147][ 1180/ 1207] Overall Loss 0.203463 Objective Loss 0.203463 LR 0.000250 Time 0.019572 -2023-02-13 18:23:16,518 - Epoch: [147][ 1190/ 1207] Overall Loss 0.203371 Objective Loss 0.203371 LR 0.000250 Time 0.019567 -2023-02-13 18:23:16,758 - Epoch: [147][ 1200/ 1207] Overall Loss 0.203325 Objective Loss 0.203325 LR 0.000250 Time 0.019604 -2023-02-13 18:23:16,873 - Epoch: [147][ 1207/ 1207] Overall Loss 0.203296 Objective Loss 0.203296 Top1 90.548780 Top5 99.390244 LR 0.000250 Time 0.019586 -2023-02-13 18:23:16,946 - --- validate (epoch=147)----------- -2023-02-13 18:23:16,946 - 34311 samples (256 per mini-batch) -2023-02-13 18:23:17,348 - Epoch: [147][ 10/ 135] Loss 0.243187 Top1 88.007812 Top5 98.085938 -2023-02-13 18:23:17,481 - Epoch: [147][ 20/ 135] Loss 0.264561 Top1 87.285156 Top5 98.046875 -2023-02-13 18:23:17,611 - Epoch: [147][ 30/ 135] Loss 0.276207 Top1 86.848958 Top5 98.020833 -2023-02-13 18:23:17,741 - Epoch: [147][ 40/ 135] Loss 0.282177 Top1 86.445312 Top5 98.017578 -2023-02-13 18:23:17,870 - Epoch: [147][ 50/ 135] Loss 0.282682 Top1 86.117188 Top5 97.992188 -2023-02-13 18:23:18,003 - Epoch: [147][ 60/ 135] Loss 0.287058 Top1 85.989583 Top5 97.988281 -2023-02-13 18:23:18,134 - Epoch: [147][ 70/ 135] Loss 0.292396 Top1 86.010045 Top5 97.924107 -2023-02-13 18:23:18,264 - Epoch: [147][ 80/ 135] Loss 0.291722 Top1 86.044922 Top5 97.924805 -2023-02-13 18:23:18,394 - Epoch: [147][ 90/ 135] Loss 0.294147 Top1 85.902778 Top5 97.916667 -2023-02-13 18:23:18,525 - Epoch: [147][ 100/ 135] Loss 0.293077 Top1 85.960938 Top5 97.929688 -2023-02-13 18:23:18,655 - Epoch: [147][ 110/ 135] Loss 0.294201 Top1 85.955256 Top5 97.933239 -2023-02-13 18:23:18,788 - Epoch: [147][ 120/ 135] Loss 0.293991 Top1 85.888672 Top5 97.968750 -2023-02-13 18:23:18,921 - Epoch: [147][ 130/ 135] Loss 0.293687 Top1 85.898438 Top5 97.965745 -2023-02-13 18:23:18,970 - Epoch: [147][ 135/ 135] Loss 0.295770 Top1 85.887908 Top5 97.965667 -2023-02-13 18:23:19,047 - ==> Top1: 85.888 Top5: 97.966 Loss: 0.296 - -2023-02-13 18:23:19,048 - ==> Confusion: -[[ 848 3 7 1 12 2 0 2 3 51 0 3 2 5 9 3 4 2 1 3 6] - [ 3 966 1 2 9 15 1 15 3 0 2 0 1 0 0 1 3 2 2 3 4] - [ 6 3 963 10 3 1 16 10 1 1 2 0 5 4 3 7 5 4 5 3 6] - [ 6 2 21 905 3 2 1 1 2 2 9 1 8 0 21 3 4 3 11 0 11] - [ 10 9 0 0 994 9 1 0 3 1 0 5 3 2 9 6 6 1 0 4 3] - [ 1 14 0 5 4 973 3 16 1 4 0 10 5 13 1 2 4 0 1 7 6] - [ 1 4 10 2 0 5 1048 2 0 3 1 2 2 0 1 3 2 2 1 5 5] - [ 2 13 12 2 2 29 6 918 0 1 1 4 3 2 0 0 0 2 13 8 6] - [ 15 3 1 2 2 0 0 0 910 31 7 2 1 9 14 3 1 0 3 1 4] - [ 64 2 4 0 10 2 0 4 32 866 0 1 1 11 5 2 1 2 1 1 3] - [ 1 3 2 7 1 4 5 3 9 1 988 1 2 7 1 0 3 1 7 0 5] - [ 2 3 3 0 4 13 0 5 0 1 1 912 29 5 1 5 6 4 1 6 4] - [ 0 0 1 6 2 2 0 1 3 0 1 27 873 2 3 10 4 14 1 1 8] - [ 3 2 1 1 9 7 1 3 6 14 7 4 2 942 2 4 4 3 0 2 7] - [ 4 0 2 15 5 3 0 2 19 7 3 1 2 2 1003 1 2 2 5 0 14] - [ 3 1 4 0 5 0 6 2 0 0 0 5 2 4 3 984 7 7 1 6 6] - [ 1 4 0 1 7 1 0 0 1 1 0 3 2 1 1 9 1015 1 2 2 9] - [ 4 4 1 4 0 2 2 0 0 1 0 8 15 1 0 27 0 973 0 2 7] - [ 3 5 6 8 0 2 1 20 6 0 3 1 2 0 10 2 0 2 1010 2 3] - [ 0 2 2 0 3 5 5 6 0 0 0 17 5 2 0 6 5 5 0 1076 9] - [ 109 235 193 112 125 192 98 139 99 72 169 99 281 267 142 98 272 77 146 207 10302]] - -2023-02-13 18:23:19,049 - ==> Best [Top1: 85.888 Top5: 97.966 Sparsity:0.00 Params: 148928 on epoch: 147] -2023-02-13 18:23:19,050 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:23:19,056 - - -2023-02-13 18:23:19,056 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:23:20,059 - Epoch: [148][ 10/ 1207] Overall Loss 0.201909 Objective Loss 0.201909 LR 0.000250 Time 0.100180 -2023-02-13 18:23:20,262 - Epoch: [148][ 20/ 1207] Overall Loss 0.194448 Objective Loss 0.194448 LR 0.000250 Time 0.060244 -2023-02-13 18:23:20,458 - Epoch: [148][ 30/ 1207] Overall Loss 0.202630 Objective Loss 0.202630 LR 0.000250 Time 0.046674 -2023-02-13 18:23:20,654 - Epoch: [148][ 40/ 1207] Overall Loss 0.199957 Objective Loss 0.199957 LR 0.000250 Time 0.039902 -2023-02-13 18:23:20,849 - Epoch: [148][ 50/ 1207] Overall Loss 0.194806 Objective Loss 0.194806 LR 0.000250 Time 0.035819 -2023-02-13 18:23:21,047 - Epoch: [148][ 60/ 1207] Overall Loss 0.195351 Objective Loss 0.195351 LR 0.000250 Time 0.033133 -2023-02-13 18:23:21,242 - Epoch: [148][ 70/ 1207] Overall Loss 0.195496 Objective Loss 0.195496 LR 0.000250 Time 0.031182 -2023-02-13 18:23:21,439 - Epoch: [148][ 80/ 1207] Overall Loss 0.195550 Objective Loss 0.195550 LR 0.000250 Time 0.029739 -2023-02-13 18:23:21,634 - Epoch: [148][ 90/ 1207] Overall Loss 0.196749 Objective Loss 0.196749 LR 0.000250 Time 0.028597 -2023-02-13 18:23:21,831 - Epoch: [148][ 100/ 1207] Overall Loss 0.192635 Objective Loss 0.192635 LR 0.000250 Time 0.027711 -2023-02-13 18:23:22,028 - Epoch: [148][ 110/ 1207] Overall Loss 0.191875 Objective Loss 0.191875 LR 0.000250 Time 0.026971 -2023-02-13 18:23:22,224 - Epoch: [148][ 120/ 1207] Overall Loss 0.191819 Objective Loss 0.191819 LR 0.000250 Time 0.026355 -2023-02-13 18:23:22,419 - Epoch: [148][ 130/ 1207] Overall Loss 0.194193 Objective Loss 0.194193 LR 0.000250 Time 0.025828 -2023-02-13 18:23:22,616 - Epoch: [148][ 140/ 1207] Overall Loss 0.193544 Objective Loss 0.193544 LR 0.000250 Time 0.025389 -2023-02-13 18:23:22,812 - Epoch: [148][ 150/ 1207] Overall Loss 0.194242 Objective Loss 0.194242 LR 0.000250 Time 0.025000 -2023-02-13 18:23:23,009 - Epoch: [148][ 160/ 1207] Overall Loss 0.194612 Objective Loss 0.194612 LR 0.000250 Time 0.024666 -2023-02-13 18:23:23,204 - Epoch: [148][ 170/ 1207] Overall Loss 0.195743 Objective Loss 0.195743 LR 0.000250 Time 0.024359 -2023-02-13 18:23:23,401 - Epoch: [148][ 180/ 1207] Overall Loss 0.195562 Objective Loss 0.195562 LR 0.000250 Time 0.024096 -2023-02-13 18:23:23,596 - Epoch: [148][ 190/ 1207] Overall Loss 0.195284 Objective Loss 0.195284 LR 0.000250 Time 0.023853 -2023-02-13 18:23:23,793 - Epoch: [148][ 200/ 1207] Overall Loss 0.194808 Objective Loss 0.194808 LR 0.000250 Time 0.023643 -2023-02-13 18:23:23,987 - Epoch: [148][ 210/ 1207] Overall Loss 0.195350 Objective Loss 0.195350 LR 0.000250 Time 0.023443 -2023-02-13 18:23:24,185 - Epoch: [148][ 220/ 1207] Overall Loss 0.195881 Objective Loss 0.195881 LR 0.000250 Time 0.023272 -2023-02-13 18:23:24,380 - Epoch: [148][ 230/ 1207] Overall Loss 0.196525 Objective Loss 0.196525 LR 0.000250 Time 0.023111 -2023-02-13 18:23:24,577 - Epoch: [148][ 240/ 1207] Overall Loss 0.196775 Objective Loss 0.196775 LR 0.000250 Time 0.022966 -2023-02-13 18:23:24,772 - Epoch: [148][ 250/ 1207] Overall Loss 0.196621 Objective Loss 0.196621 LR 0.000250 Time 0.022827 -2023-02-13 18:23:24,969 - Epoch: [148][ 260/ 1207] Overall Loss 0.196514 Objective Loss 0.196514 LR 0.000250 Time 0.022703 -2023-02-13 18:23:25,165 - Epoch: [148][ 270/ 1207] Overall Loss 0.196634 Objective Loss 0.196634 LR 0.000250 Time 0.022587 -2023-02-13 18:23:25,362 - Epoch: [148][ 280/ 1207] Overall Loss 0.197244 Objective Loss 0.197244 LR 0.000250 Time 0.022482 -2023-02-13 18:23:25,558 - Epoch: [148][ 290/ 1207] Overall Loss 0.197786 Objective Loss 0.197786 LR 0.000250 Time 0.022382 -2023-02-13 18:23:25,755 - Epoch: [148][ 300/ 1207] Overall Loss 0.197242 Objective Loss 0.197242 LR 0.000250 Time 0.022292 -2023-02-13 18:23:25,952 - Epoch: [148][ 310/ 1207] Overall Loss 0.196513 Objective Loss 0.196513 LR 0.000250 Time 0.022206 -2023-02-13 18:23:26,150 - Epoch: [148][ 320/ 1207] Overall Loss 0.196572 Objective Loss 0.196572 LR 0.000250 Time 0.022129 -2023-02-13 18:23:26,345 - Epoch: [148][ 330/ 1207] Overall Loss 0.196851 Objective Loss 0.196851 LR 0.000250 Time 0.022048 -2023-02-13 18:23:26,541 - Epoch: [148][ 340/ 1207] Overall Loss 0.196859 Objective Loss 0.196859 LR 0.000250 Time 0.021977 -2023-02-13 18:23:26,737 - Epoch: [148][ 350/ 1207] Overall Loss 0.197371 Objective Loss 0.197371 LR 0.000250 Time 0.021906 -2023-02-13 18:23:26,934 - Epoch: [148][ 360/ 1207] Overall Loss 0.198667 Objective Loss 0.198667 LR 0.000250 Time 0.021846 -2023-02-13 18:23:27,130 - Epoch: [148][ 370/ 1207] Overall Loss 0.199027 Objective Loss 0.199027 LR 0.000250 Time 0.021782 -2023-02-13 18:23:27,327 - Epoch: [148][ 380/ 1207] Overall Loss 0.199707 Objective Loss 0.199707 LR 0.000250 Time 0.021727 -2023-02-13 18:23:27,522 - Epoch: [148][ 390/ 1207] Overall Loss 0.199646 Objective Loss 0.199646 LR 0.000250 Time 0.021669 -2023-02-13 18:23:27,719 - Epoch: [148][ 400/ 1207] Overall Loss 0.199444 Objective Loss 0.199444 LR 0.000250 Time 0.021619 -2023-02-13 18:23:27,917 - Epoch: [148][ 410/ 1207] Overall Loss 0.199476 Objective Loss 0.199476 LR 0.000250 Time 0.021574 -2023-02-13 18:23:28,114 - Epoch: [148][ 420/ 1207] Overall Loss 0.198924 Objective Loss 0.198924 LR 0.000250 Time 0.021530 -2023-02-13 18:23:28,305 - Epoch: [148][ 430/ 1207] Overall Loss 0.199291 Objective Loss 0.199291 LR 0.000250 Time 0.021472 -2023-02-13 18:23:28,495 - Epoch: [148][ 440/ 1207] Overall Loss 0.199058 Objective Loss 0.199058 LR 0.000250 Time 0.021413 -2023-02-13 18:23:28,685 - Epoch: [148][ 450/ 1207] Overall Loss 0.199491 Objective Loss 0.199491 LR 0.000250 Time 0.021360 -2023-02-13 18:23:28,875 - Epoch: [148][ 460/ 1207] Overall Loss 0.199639 Objective Loss 0.199639 LR 0.000250 Time 0.021308 -2023-02-13 18:23:29,065 - Epoch: [148][ 470/ 1207] Overall Loss 0.199860 Objective Loss 0.199860 LR 0.000250 Time 0.021257 -2023-02-13 18:23:29,254 - Epoch: [148][ 480/ 1207] Overall Loss 0.199587 Objective Loss 0.199587 LR 0.000250 Time 0.021208 -2023-02-13 18:23:29,443 - Epoch: [148][ 490/ 1207] Overall Loss 0.199305 Objective Loss 0.199305 LR 0.000250 Time 0.021160 -2023-02-13 18:23:29,633 - Epoch: [148][ 500/ 1207] Overall Loss 0.199421 Objective Loss 0.199421 LR 0.000250 Time 0.021115 -2023-02-13 18:23:29,822 - Epoch: [148][ 510/ 1207] Overall Loss 0.199037 Objective Loss 0.199037 LR 0.000250 Time 0.021072 -2023-02-13 18:23:30,012 - Epoch: [148][ 520/ 1207] Overall Loss 0.199384 Objective Loss 0.199384 LR 0.000250 Time 0.021031 -2023-02-13 18:23:30,201 - Epoch: [148][ 530/ 1207] Overall Loss 0.199346 Objective Loss 0.199346 LR 0.000250 Time 0.020990 -2023-02-13 18:23:30,391 - Epoch: [148][ 540/ 1207] Overall Loss 0.199375 Objective Loss 0.199375 LR 0.000250 Time 0.020952 -2023-02-13 18:23:30,580 - Epoch: [148][ 550/ 1207] Overall Loss 0.199401 Objective Loss 0.199401 LR 0.000250 Time 0.020915 -2023-02-13 18:23:30,770 - Epoch: [148][ 560/ 1207] Overall Loss 0.199210 Objective Loss 0.199210 LR 0.000250 Time 0.020879 -2023-02-13 18:23:30,960 - Epoch: [148][ 570/ 1207] Overall Loss 0.199384 Objective Loss 0.199384 LR 0.000250 Time 0.020847 -2023-02-13 18:23:31,150 - Epoch: [148][ 580/ 1207] Overall Loss 0.199146 Objective Loss 0.199146 LR 0.000250 Time 0.020813 -2023-02-13 18:23:31,339 - Epoch: [148][ 590/ 1207] Overall Loss 0.199831 Objective Loss 0.199831 LR 0.000250 Time 0.020780 -2023-02-13 18:23:31,528 - Epoch: [148][ 600/ 1207] Overall Loss 0.200053 Objective Loss 0.200053 LR 0.000250 Time 0.020749 -2023-02-13 18:23:31,718 - Epoch: [148][ 610/ 1207] Overall Loss 0.200273 Objective Loss 0.200273 LR 0.000250 Time 0.020720 -2023-02-13 18:23:31,909 - Epoch: [148][ 620/ 1207] Overall Loss 0.200552 Objective Loss 0.200552 LR 0.000250 Time 0.020693 -2023-02-13 18:23:32,098 - Epoch: [148][ 630/ 1207] Overall Loss 0.200587 Objective Loss 0.200587 LR 0.000250 Time 0.020664 -2023-02-13 18:23:32,288 - Epoch: [148][ 640/ 1207] Overall Loss 0.200562 Objective Loss 0.200562 LR 0.000250 Time 0.020638 -2023-02-13 18:23:32,478 - Epoch: [148][ 650/ 1207] Overall Loss 0.200716 Objective Loss 0.200716 LR 0.000250 Time 0.020611 -2023-02-13 18:23:32,668 - Epoch: [148][ 660/ 1207] Overall Loss 0.200629 Objective Loss 0.200629 LR 0.000250 Time 0.020586 -2023-02-13 18:23:32,857 - Epoch: [148][ 670/ 1207] Overall Loss 0.200720 Objective Loss 0.200720 LR 0.000250 Time 0.020561 -2023-02-13 18:23:33,047 - Epoch: [148][ 680/ 1207] Overall Loss 0.200385 Objective Loss 0.200385 LR 0.000250 Time 0.020537 -2023-02-13 18:23:33,236 - Epoch: [148][ 690/ 1207] Overall Loss 0.200456 Objective Loss 0.200456 LR 0.000250 Time 0.020513 -2023-02-13 18:23:33,426 - Epoch: [148][ 700/ 1207] Overall Loss 0.200578 Objective Loss 0.200578 LR 0.000250 Time 0.020491 -2023-02-13 18:23:33,614 - Epoch: [148][ 710/ 1207] Overall Loss 0.200686 Objective Loss 0.200686 LR 0.000250 Time 0.020467 -2023-02-13 18:23:33,804 - Epoch: [148][ 720/ 1207] Overall Loss 0.200741 Objective Loss 0.200741 LR 0.000250 Time 0.020446 -2023-02-13 18:23:33,994 - Epoch: [148][ 730/ 1207] Overall Loss 0.200841 Objective Loss 0.200841 LR 0.000250 Time 0.020425 -2023-02-13 18:23:34,184 - Epoch: [148][ 740/ 1207] Overall Loss 0.200680 Objective Loss 0.200680 LR 0.000250 Time 0.020405 -2023-02-13 18:23:34,373 - Epoch: [148][ 750/ 1207] Overall Loss 0.200206 Objective Loss 0.200206 LR 0.000250 Time 0.020385 -2023-02-13 18:23:34,563 - Epoch: [148][ 760/ 1207] Overall Loss 0.200095 Objective Loss 0.200095 LR 0.000250 Time 0.020366 -2023-02-13 18:23:34,752 - Epoch: [148][ 770/ 1207] Overall Loss 0.200192 Objective Loss 0.200192 LR 0.000250 Time 0.020347 -2023-02-13 18:23:34,942 - Epoch: [148][ 780/ 1207] Overall Loss 0.200511 Objective Loss 0.200511 LR 0.000250 Time 0.020329 -2023-02-13 18:23:35,132 - Epoch: [148][ 790/ 1207] Overall Loss 0.200426 Objective Loss 0.200426 LR 0.000250 Time 0.020311 -2023-02-13 18:23:35,322 - Epoch: [148][ 800/ 1207] Overall Loss 0.200328 Objective Loss 0.200328 LR 0.000250 Time 0.020294 -2023-02-13 18:23:35,511 - Epoch: [148][ 810/ 1207] Overall Loss 0.200198 Objective Loss 0.200198 LR 0.000250 Time 0.020278 -2023-02-13 18:23:35,701 - Epoch: [148][ 820/ 1207] Overall Loss 0.200111 Objective Loss 0.200111 LR 0.000250 Time 0.020261 -2023-02-13 18:23:35,892 - Epoch: [148][ 830/ 1207] Overall Loss 0.200326 Objective Loss 0.200326 LR 0.000250 Time 0.020246 -2023-02-13 18:23:36,082 - Epoch: [148][ 840/ 1207] Overall Loss 0.200138 Objective Loss 0.200138 LR 0.000250 Time 0.020231 -2023-02-13 18:23:36,272 - Epoch: [148][ 850/ 1207] Overall Loss 0.199995 Objective Loss 0.199995 LR 0.000250 Time 0.020216 -2023-02-13 18:23:36,461 - Epoch: [148][ 860/ 1207] Overall Loss 0.199898 Objective Loss 0.199898 LR 0.000250 Time 0.020201 -2023-02-13 18:23:36,651 - Epoch: [148][ 870/ 1207] Overall Loss 0.199976 Objective Loss 0.199976 LR 0.000250 Time 0.020186 -2023-02-13 18:23:36,841 - Epoch: [148][ 880/ 1207] Overall Loss 0.200052 Objective Loss 0.200052 LR 0.000250 Time 0.020173 -2023-02-13 18:23:37,031 - Epoch: [148][ 890/ 1207] Overall Loss 0.200028 Objective Loss 0.200028 LR 0.000250 Time 0.020159 -2023-02-13 18:23:37,221 - Epoch: [148][ 900/ 1207] Overall Loss 0.200348 Objective Loss 0.200348 LR 0.000250 Time 0.020146 -2023-02-13 18:23:37,411 - Epoch: [148][ 910/ 1207] Overall Loss 0.200501 Objective Loss 0.200501 LR 0.000250 Time 0.020132 -2023-02-13 18:23:37,601 - Epoch: [148][ 920/ 1207] Overall Loss 0.200609 Objective Loss 0.200609 LR 0.000250 Time 0.020120 -2023-02-13 18:23:37,790 - Epoch: [148][ 930/ 1207] Overall Loss 0.200866 Objective Loss 0.200866 LR 0.000250 Time 0.020107 -2023-02-13 18:23:37,981 - Epoch: [148][ 940/ 1207] Overall Loss 0.201038 Objective Loss 0.201038 LR 0.000250 Time 0.020095 -2023-02-13 18:23:38,170 - Epoch: [148][ 950/ 1207] Overall Loss 0.200826 Objective Loss 0.200826 LR 0.000250 Time 0.020082 -2023-02-13 18:23:38,360 - Epoch: [148][ 960/ 1207] Overall Loss 0.200974 Objective Loss 0.200974 LR 0.000250 Time 0.020071 -2023-02-13 18:23:38,549 - Epoch: [148][ 970/ 1207] Overall Loss 0.201138 Objective Loss 0.201138 LR 0.000250 Time 0.020059 -2023-02-13 18:23:38,739 - Epoch: [148][ 980/ 1207] Overall Loss 0.201407 Objective Loss 0.201407 LR 0.000250 Time 0.020047 -2023-02-13 18:23:38,930 - Epoch: [148][ 990/ 1207] Overall Loss 0.201541 Objective Loss 0.201541 LR 0.000250 Time 0.020037 -2023-02-13 18:23:39,120 - Epoch: [148][ 1000/ 1207] Overall Loss 0.201634 Objective Loss 0.201634 LR 0.000250 Time 0.020026 -2023-02-13 18:23:39,310 - Epoch: [148][ 1010/ 1207] Overall Loss 0.201347 Objective Loss 0.201347 LR 0.000250 Time 0.020015 -2023-02-13 18:23:39,499 - Epoch: [148][ 1020/ 1207] Overall Loss 0.201606 Objective Loss 0.201606 LR 0.000250 Time 0.020005 -2023-02-13 18:23:39,689 - Epoch: [148][ 1030/ 1207] Overall Loss 0.201751 Objective Loss 0.201751 LR 0.000250 Time 0.019994 -2023-02-13 18:23:39,879 - Epoch: [148][ 1040/ 1207] Overall Loss 0.201814 Objective Loss 0.201814 LR 0.000250 Time 0.019985 -2023-02-13 18:23:40,069 - Epoch: [148][ 1050/ 1207] Overall Loss 0.201694 Objective Loss 0.201694 LR 0.000250 Time 0.019975 -2023-02-13 18:23:40,258 - Epoch: [148][ 1060/ 1207] Overall Loss 0.201607 Objective Loss 0.201607 LR 0.000250 Time 0.019964 -2023-02-13 18:23:40,448 - Epoch: [148][ 1070/ 1207] Overall Loss 0.201624 Objective Loss 0.201624 LR 0.000250 Time 0.019955 -2023-02-13 18:23:40,639 - Epoch: [148][ 1080/ 1207] Overall Loss 0.201627 Objective Loss 0.201627 LR 0.000250 Time 0.019947 -2023-02-13 18:23:40,830 - Epoch: [148][ 1090/ 1207] Overall Loss 0.201471 Objective Loss 0.201471 LR 0.000250 Time 0.019939 -2023-02-13 18:23:41,022 - Epoch: [148][ 1100/ 1207] Overall Loss 0.201346 Objective Loss 0.201346 LR 0.000250 Time 0.019931 -2023-02-13 18:23:41,211 - Epoch: [148][ 1110/ 1207] Overall Loss 0.201492 Objective Loss 0.201492 LR 0.000250 Time 0.019922 -2023-02-13 18:23:41,401 - Epoch: [148][ 1120/ 1207] Overall Loss 0.201736 Objective Loss 0.201736 LR 0.000250 Time 0.019914 -2023-02-13 18:23:41,591 - Epoch: [148][ 1130/ 1207] Overall Loss 0.201624 Objective Loss 0.201624 LR 0.000250 Time 0.019905 -2023-02-13 18:23:41,781 - Epoch: [148][ 1140/ 1207] Overall Loss 0.201660 Objective Loss 0.201660 LR 0.000250 Time 0.019896 -2023-02-13 18:23:41,971 - Epoch: [148][ 1150/ 1207] Overall Loss 0.201456 Objective Loss 0.201456 LR 0.000250 Time 0.019889 -2023-02-13 18:23:42,161 - Epoch: [148][ 1160/ 1207] Overall Loss 0.201214 Objective Loss 0.201214 LR 0.000250 Time 0.019881 -2023-02-13 18:23:42,351 - Epoch: [148][ 1170/ 1207] Overall Loss 0.201351 Objective Loss 0.201351 LR 0.000250 Time 0.019873 -2023-02-13 18:23:42,540 - Epoch: [148][ 1180/ 1207] Overall Loss 0.201418 Objective Loss 0.201418 LR 0.000250 Time 0.019864 -2023-02-13 18:23:42,730 - Epoch: [148][ 1190/ 1207] Overall Loss 0.201380 Objective Loss 0.201380 LR 0.000250 Time 0.019857 -2023-02-13 18:23:42,969 - Epoch: [148][ 1200/ 1207] Overall Loss 0.201413 Objective Loss 0.201413 LR 0.000250 Time 0.019890 -2023-02-13 18:23:43,086 - Epoch: [148][ 1207/ 1207] Overall Loss 0.201537 Objective Loss 0.201537 Top1 84.756098 Top5 97.865854 LR 0.000250 Time 0.019871 -2023-02-13 18:23:43,158 - --- validate (epoch=148)----------- -2023-02-13 18:23:43,158 - 34311 samples (256 per mini-batch) -2023-02-13 18:23:43,558 - Epoch: [148][ 10/ 135] Loss 0.304168 Top1 85.585938 Top5 97.851562 -2023-02-13 18:23:43,687 - Epoch: [148][ 20/ 135] Loss 0.300511 Top1 85.703125 Top5 97.968750 -2023-02-13 18:23:43,815 - Epoch: [148][ 30/ 135] Loss 0.301916 Top1 85.781250 Top5 97.942708 -2023-02-13 18:23:43,942 - Epoch: [148][ 40/ 135] Loss 0.294964 Top1 85.625000 Top5 97.841797 -2023-02-13 18:23:44,069 - Epoch: [148][ 50/ 135] Loss 0.294757 Top1 85.257812 Top5 97.835938 -2023-02-13 18:23:44,199 - Epoch: [148][ 60/ 135] Loss 0.293350 Top1 85.319010 Top5 97.845052 -2023-02-13 18:23:44,329 - Epoch: [148][ 70/ 135] Loss 0.296069 Top1 85.234375 Top5 97.851562 -2023-02-13 18:23:44,458 - Epoch: [148][ 80/ 135] Loss 0.297432 Top1 85.297852 Top5 97.822266 -2023-02-13 18:23:44,588 - Epoch: [148][ 90/ 135] Loss 0.298382 Top1 85.347222 Top5 97.834201 -2023-02-13 18:23:44,718 - Epoch: [148][ 100/ 135] Loss 0.296772 Top1 85.312500 Top5 97.859375 -2023-02-13 18:23:44,848 - Epoch: [148][ 110/ 135] Loss 0.294449 Top1 85.248580 Top5 97.865767 -2023-02-13 18:23:44,978 - Epoch: [148][ 120/ 135] Loss 0.295840 Top1 85.286458 Top5 97.838542 -2023-02-13 18:23:45,108 - Epoch: [148][ 130/ 135] Loss 0.295058 Top1 85.339543 Top5 97.878606 -2023-02-13 18:23:45,154 - Epoch: [148][ 135/ 135] Loss 0.294191 Top1 85.299175 Top5 97.869488 -2023-02-13 18:23:45,227 - ==> Top1: 85.299 Top5: 97.869 Loss: 0.294 - -2023-02-13 18:23:45,228 - ==> Confusion: -[[ 859 4 2 2 8 4 0 0 5 42 0 5 4 5 8 4 2 2 0 0 11] - [ 1 956 3 2 11 21 0 18 3 0 2 1 2 1 0 1 1 1 3 2 4] - [ 7 4 951 16 5 1 12 13 1 1 3 1 3 5 2 13 3 1 8 3 5] - [ 5 1 18 903 4 3 0 1 3 0 11 1 8 0 19 1 3 7 20 1 7] - [ 11 10 1 0 1001 5 1 1 1 1 0 7 1 2 10 4 4 1 0 1 4] - [ 0 16 2 1 7 981 3 14 1 2 2 10 6 11 0 3 3 0 2 2 4] - [ 3 3 12 1 0 5 1033 6 0 2 3 3 3 2 0 7 1 5 3 3 4] - [ 3 10 7 2 2 24 2 935 0 1 2 4 4 2 0 1 1 2 12 7 3] - [ 12 3 0 1 2 0 1 1 911 34 4 3 1 8 15 3 0 1 6 1 2] - [ 70 1 3 0 7 2 0 4 35 866 0 1 0 9 5 1 1 3 1 0 3] - [ 2 2 0 7 2 1 3 4 15 1 988 2 1 6 4 0 0 1 8 0 4] - [ 1 1 3 1 4 10 0 3 1 2 1 932 24 4 2 1 2 7 1 3 2] - [ 0 0 1 4 2 4 0 1 1 2 1 29 883 1 3 4 2 12 1 0 8] - [ 4 3 1 1 9 11 0 1 15 10 7 8 1 934 3 6 3 3 1 0 3] - [ 5 0 1 13 6 2 0 2 15 7 1 1 4 1 1011 2 1 6 6 0 8] - [ 1 3 5 0 8 2 3 0 0 0 0 7 5 1 2 975 8 15 0 7 4] - [ 1 7 0 1 11 2 0 0 2 0 0 1 5 2 1 11 997 1 3 2 14] - [ 3 3 1 4 0 2 1 0 1 1 0 9 14 0 2 16 0 988 1 1 4] - [ 5 2 2 7 1 1 0 22 3 0 2 1 6 0 11 0 2 1 1016 2 2] - [ 0 3 1 0 3 6 2 10 0 0 1 19 3 5 0 5 5 6 2 1070 7] - [ 125 242 178 115 145 191 76 181 112 72 180 148 324 277 184 93 207 105 200 202 10077]] - -2023-02-13 18:23:45,230 - ==> Best [Top1: 85.888 Top5: 97.966 Sparsity:0.00 Params: 148928 on epoch: 147] -2023-02-13 18:23:45,230 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:23:45,235 - - -2023-02-13 18:23:45,235 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:23:46,127 - Epoch: [149][ 10/ 1207] Overall Loss 0.183774 Objective Loss 0.183774 LR 0.000250 Time 0.089048 -2023-02-13 18:23:46,327 - Epoch: [149][ 20/ 1207] Overall Loss 0.181603 Objective Loss 0.181603 LR 0.000250 Time 0.054530 -2023-02-13 18:23:46,520 - Epoch: [149][ 30/ 1207] Overall Loss 0.190629 Objective Loss 0.190629 LR 0.000250 Time 0.042776 -2023-02-13 18:23:46,715 - Epoch: [149][ 40/ 1207] Overall Loss 0.190823 Objective Loss 0.190823 LR 0.000250 Time 0.036935 -2023-02-13 18:23:46,908 - Epoch: [149][ 50/ 1207] Overall Loss 0.191849 Objective Loss 0.191849 LR 0.000250 Time 0.033404 -2023-02-13 18:23:47,103 - Epoch: [149][ 60/ 1207] Overall Loss 0.191418 Objective Loss 0.191418 LR 0.000250 Time 0.031081 -2023-02-13 18:23:47,295 - Epoch: [149][ 70/ 1207] Overall Loss 0.190545 Objective Loss 0.190545 LR 0.000250 Time 0.029384 -2023-02-13 18:23:47,489 - Epoch: [149][ 80/ 1207] Overall Loss 0.192799 Objective Loss 0.192799 LR 0.000250 Time 0.028130 -2023-02-13 18:23:47,681 - Epoch: [149][ 90/ 1207] Overall Loss 0.194069 Objective Loss 0.194069 LR 0.000250 Time 0.027129 -2023-02-13 18:23:47,875 - Epoch: [149][ 100/ 1207] Overall Loss 0.194901 Objective Loss 0.194901 LR 0.000250 Time 0.026357 -2023-02-13 18:23:48,068 - Epoch: [149][ 110/ 1207] Overall Loss 0.195043 Objective Loss 0.195043 LR 0.000250 Time 0.025707 -2023-02-13 18:23:48,262 - Epoch: [149][ 120/ 1207] Overall Loss 0.194533 Objective Loss 0.194533 LR 0.000250 Time 0.025179 -2023-02-13 18:23:48,454 - Epoch: [149][ 130/ 1207] Overall Loss 0.195016 Objective Loss 0.195016 LR 0.000250 Time 0.024716 -2023-02-13 18:23:48,648 - Epoch: [149][ 140/ 1207] Overall Loss 0.195185 Objective Loss 0.195185 LR 0.000250 Time 0.024335 -2023-02-13 18:23:48,841 - Epoch: [149][ 150/ 1207] Overall Loss 0.195732 Objective Loss 0.195732 LR 0.000250 Time 0.023995 -2023-02-13 18:23:49,035 - Epoch: [149][ 160/ 1207] Overall Loss 0.195271 Objective Loss 0.195271 LR 0.000250 Time 0.023705 -2023-02-13 18:23:49,227 - Epoch: [149][ 170/ 1207] Overall Loss 0.194652 Objective Loss 0.194652 LR 0.000250 Time 0.023438 -2023-02-13 18:23:49,421 - Epoch: [149][ 180/ 1207] Overall Loss 0.196119 Objective Loss 0.196119 LR 0.000250 Time 0.023211 -2023-02-13 18:23:49,612 - Epoch: [149][ 190/ 1207] Overall Loss 0.195949 Objective Loss 0.195949 LR 0.000250 Time 0.022998 -2023-02-13 18:23:49,806 - Epoch: [149][ 200/ 1207] Overall Loss 0.196137 Objective Loss 0.196137 LR 0.000250 Time 0.022815 -2023-02-13 18:23:49,999 - Epoch: [149][ 210/ 1207] Overall Loss 0.196680 Objective Loss 0.196680 LR 0.000250 Time 0.022642 -2023-02-13 18:23:50,192 - Epoch: [149][ 220/ 1207] Overall Loss 0.195896 Objective Loss 0.195896 LR 0.000250 Time 0.022493 -2023-02-13 18:23:50,385 - Epoch: [149][ 230/ 1207] Overall Loss 0.197131 Objective Loss 0.197131 LR 0.000250 Time 0.022351 -2023-02-13 18:23:50,579 - Epoch: [149][ 240/ 1207] Overall Loss 0.196748 Objective Loss 0.196748 LR 0.000250 Time 0.022225 -2023-02-13 18:23:50,776 - Epoch: [149][ 250/ 1207] Overall Loss 0.196152 Objective Loss 0.196152 LR 0.000250 Time 0.022123 -2023-02-13 18:23:50,986 - Epoch: [149][ 260/ 1207] Overall Loss 0.196283 Objective Loss 0.196283 LR 0.000250 Time 0.022078 -2023-02-13 18:23:51,180 - Epoch: [149][ 270/ 1207] Overall Loss 0.197223 Objective Loss 0.197223 LR 0.000250 Time 0.021976 -2023-02-13 18:23:51,375 - Epoch: [149][ 280/ 1207] Overall Loss 0.197310 Objective Loss 0.197310 LR 0.000250 Time 0.021888 -2023-02-13 18:23:51,569 - Epoch: [149][ 290/ 1207] Overall Loss 0.197515 Objective Loss 0.197515 LR 0.000250 Time 0.021800 -2023-02-13 18:23:51,765 - Epoch: [149][ 300/ 1207] Overall Loss 0.197255 Objective Loss 0.197255 LR 0.000250 Time 0.021725 -2023-02-13 18:23:51,959 - Epoch: [149][ 310/ 1207] Overall Loss 0.197134 Objective Loss 0.197134 LR 0.000250 Time 0.021651 -2023-02-13 18:23:52,155 - Epoch: [149][ 320/ 1207] Overall Loss 0.198069 Objective Loss 0.198069 LR 0.000250 Time 0.021584 -2023-02-13 18:23:52,348 - Epoch: [149][ 330/ 1207] Overall Loss 0.197688 Objective Loss 0.197688 LR 0.000250 Time 0.021516 -2023-02-13 18:23:52,544 - Epoch: [149][ 340/ 1207] Overall Loss 0.197868 Objective Loss 0.197868 LR 0.000250 Time 0.021455 -2023-02-13 18:23:52,738 - Epoch: [149][ 350/ 1207] Overall Loss 0.197754 Objective Loss 0.197754 LR 0.000250 Time 0.021396 -2023-02-13 18:23:52,934 - Epoch: [149][ 360/ 1207] Overall Loss 0.197874 Objective Loss 0.197874 LR 0.000250 Time 0.021345 -2023-02-13 18:23:53,127 - Epoch: [149][ 370/ 1207] Overall Loss 0.198091 Objective Loss 0.198091 LR 0.000250 Time 0.021290 -2023-02-13 18:23:53,323 - Epoch: [149][ 380/ 1207] Overall Loss 0.197643 Objective Loss 0.197643 LR 0.000250 Time 0.021243 -2023-02-13 18:23:53,516 - Epoch: [149][ 390/ 1207] Overall Loss 0.198100 Objective Loss 0.198100 LR 0.000250 Time 0.021194 -2023-02-13 18:23:53,711 - Epoch: [149][ 400/ 1207] Overall Loss 0.198175 Objective Loss 0.198175 LR 0.000250 Time 0.021151 -2023-02-13 18:23:53,906 - Epoch: [149][ 410/ 1207] Overall Loss 0.198486 Objective Loss 0.198486 LR 0.000250 Time 0.021108 -2023-02-13 18:23:54,101 - Epoch: [149][ 420/ 1207] Overall Loss 0.198524 Objective Loss 0.198524 LR 0.000250 Time 0.021071 -2023-02-13 18:23:54,295 - Epoch: [149][ 430/ 1207] Overall Loss 0.198569 Objective Loss 0.198569 LR 0.000250 Time 0.021031 -2023-02-13 18:23:54,491 - Epoch: [149][ 440/ 1207] Overall Loss 0.198701 Objective Loss 0.198701 LR 0.000250 Time 0.020997 -2023-02-13 18:23:54,685 - Epoch: [149][ 450/ 1207] Overall Loss 0.198484 Objective Loss 0.198484 LR 0.000250 Time 0.020960 -2023-02-13 18:23:54,882 - Epoch: [149][ 460/ 1207] Overall Loss 0.198398 Objective Loss 0.198398 LR 0.000250 Time 0.020933 -2023-02-13 18:23:55,076 - Epoch: [149][ 470/ 1207] Overall Loss 0.198082 Objective Loss 0.198082 LR 0.000250 Time 0.020899 -2023-02-13 18:23:55,273 - Epoch: [149][ 480/ 1207] Overall Loss 0.198622 Objective Loss 0.198622 LR 0.000250 Time 0.020872 -2023-02-13 18:23:55,467 - Epoch: [149][ 490/ 1207] Overall Loss 0.198518 Objective Loss 0.198518 LR 0.000250 Time 0.020843 -2023-02-13 18:23:55,664 - Epoch: [149][ 500/ 1207] Overall Loss 0.198610 Objective Loss 0.198610 LR 0.000250 Time 0.020818 -2023-02-13 18:23:55,859 - Epoch: [149][ 510/ 1207] Overall Loss 0.198267 Objective Loss 0.198267 LR 0.000250 Time 0.020791 -2023-02-13 18:23:56,055 - Epoch: [149][ 520/ 1207] Overall Loss 0.198392 Objective Loss 0.198392 LR 0.000250 Time 0.020768 -2023-02-13 18:23:56,249 - Epoch: [149][ 530/ 1207] Overall Loss 0.197978 Objective Loss 0.197978 LR 0.000250 Time 0.020742 -2023-02-13 18:23:56,446 - Epoch: [149][ 540/ 1207] Overall Loss 0.197887 Objective Loss 0.197887 LR 0.000250 Time 0.020721 -2023-02-13 18:23:56,639 - Epoch: [149][ 550/ 1207] Overall Loss 0.197918 Objective Loss 0.197918 LR 0.000250 Time 0.020696 -2023-02-13 18:23:56,835 - Epoch: [149][ 560/ 1207] Overall Loss 0.198168 Objective Loss 0.198168 LR 0.000250 Time 0.020674 -2023-02-13 18:23:57,030 - Epoch: [149][ 570/ 1207] Overall Loss 0.198594 Objective Loss 0.198594 LR 0.000250 Time 0.020653 -2023-02-13 18:23:57,226 - Epoch: [149][ 580/ 1207] Overall Loss 0.198371 Objective Loss 0.198371 LR 0.000250 Time 0.020634 -2023-02-13 18:23:57,420 - Epoch: [149][ 590/ 1207] Overall Loss 0.198353 Objective Loss 0.198353 LR 0.000250 Time 0.020613 -2023-02-13 18:23:57,616 - Epoch: [149][ 600/ 1207] Overall Loss 0.198192 Objective Loss 0.198192 LR 0.000250 Time 0.020596 -2023-02-13 18:23:57,810 - Epoch: [149][ 610/ 1207] Overall Loss 0.198485 Objective Loss 0.198485 LR 0.000250 Time 0.020575 -2023-02-13 18:23:58,006 - Epoch: [149][ 620/ 1207] Overall Loss 0.198503 Objective Loss 0.198503 LR 0.000250 Time 0.020559 -2023-02-13 18:23:58,200 - Epoch: [149][ 630/ 1207] Overall Loss 0.198329 Objective Loss 0.198329 LR 0.000250 Time 0.020540 -2023-02-13 18:23:58,396 - Epoch: [149][ 640/ 1207] Overall Loss 0.198316 Objective Loss 0.198316 LR 0.000250 Time 0.020524 -2023-02-13 18:23:58,590 - Epoch: [149][ 650/ 1207] Overall Loss 0.198174 Objective Loss 0.198174 LR 0.000250 Time 0.020507 -2023-02-13 18:23:58,787 - Epoch: [149][ 660/ 1207] Overall Loss 0.198249 Objective Loss 0.198249 LR 0.000250 Time 0.020494 -2023-02-13 18:23:58,983 - Epoch: [149][ 670/ 1207] Overall Loss 0.198009 Objective Loss 0.198009 LR 0.000250 Time 0.020480 -2023-02-13 18:23:59,180 - Epoch: [149][ 680/ 1207] Overall Loss 0.198233 Objective Loss 0.198233 LR 0.000250 Time 0.020468 -2023-02-13 18:23:59,376 - Epoch: [149][ 690/ 1207] Overall Loss 0.198291 Objective Loss 0.198291 LR 0.000250 Time 0.020454 -2023-02-13 18:23:59,573 - Epoch: [149][ 700/ 1207] Overall Loss 0.198541 Objective Loss 0.198541 LR 0.000250 Time 0.020443 -2023-02-13 18:23:59,768 - Epoch: [149][ 710/ 1207] Overall Loss 0.198519 Objective Loss 0.198519 LR 0.000250 Time 0.020429 -2023-02-13 18:23:59,965 - Epoch: [149][ 720/ 1207] Overall Loss 0.198778 Objective Loss 0.198778 LR 0.000250 Time 0.020419 -2023-02-13 18:24:00,160 - Epoch: [149][ 730/ 1207] Overall Loss 0.198688 Objective Loss 0.198688 LR 0.000250 Time 0.020406 -2023-02-13 18:24:00,356 - Epoch: [149][ 740/ 1207] Overall Loss 0.198842 Objective Loss 0.198842 LR 0.000250 Time 0.020394 -2023-02-13 18:24:00,551 - Epoch: [149][ 750/ 1207] Overall Loss 0.198735 Objective Loss 0.198735 LR 0.000250 Time 0.020382 -2023-02-13 18:24:00,748 - Epoch: [149][ 760/ 1207] Overall Loss 0.198522 Objective Loss 0.198522 LR 0.000250 Time 0.020373 -2023-02-13 18:24:00,944 - Epoch: [149][ 770/ 1207] Overall Loss 0.198739 Objective Loss 0.198739 LR 0.000250 Time 0.020362 -2023-02-13 18:24:01,141 - Epoch: [149][ 780/ 1207] Overall Loss 0.198986 Objective Loss 0.198986 LR 0.000250 Time 0.020353 -2023-02-13 18:24:01,336 - Epoch: [149][ 790/ 1207] Overall Loss 0.199008 Objective Loss 0.199008 LR 0.000250 Time 0.020341 -2023-02-13 18:24:01,532 - Epoch: [149][ 800/ 1207] Overall Loss 0.198771 Objective Loss 0.198771 LR 0.000250 Time 0.020332 -2023-02-13 18:24:01,727 - Epoch: [149][ 810/ 1207] Overall Loss 0.198518 Objective Loss 0.198518 LR 0.000250 Time 0.020322 -2023-02-13 18:24:01,925 - Epoch: [149][ 820/ 1207] Overall Loss 0.198865 Objective Loss 0.198865 LR 0.000250 Time 0.020315 -2023-02-13 18:24:02,121 - Epoch: [149][ 830/ 1207] Overall Loss 0.198804 Objective Loss 0.198804 LR 0.000250 Time 0.020305 -2023-02-13 18:24:02,318 - Epoch: [149][ 840/ 1207] Overall Loss 0.198889 Objective Loss 0.198889 LR 0.000250 Time 0.020297 -2023-02-13 18:24:02,513 - Epoch: [149][ 850/ 1207] Overall Loss 0.198524 Objective Loss 0.198524 LR 0.000250 Time 0.020288 -2023-02-13 18:24:02,710 - Epoch: [149][ 860/ 1207] Overall Loss 0.198448 Objective Loss 0.198448 LR 0.000250 Time 0.020281 -2023-02-13 18:24:02,906 - Epoch: [149][ 870/ 1207] Overall Loss 0.198382 Objective Loss 0.198382 LR 0.000250 Time 0.020272 -2023-02-13 18:24:03,103 - Epoch: [149][ 880/ 1207] Overall Loss 0.198402 Objective Loss 0.198402 LR 0.000250 Time 0.020265 -2023-02-13 18:24:03,297 - Epoch: [149][ 890/ 1207] Overall Loss 0.198697 Objective Loss 0.198697 LR 0.000250 Time 0.020255 -2023-02-13 18:24:03,493 - Epoch: [149][ 900/ 1207] Overall Loss 0.198730 Objective Loss 0.198730 LR 0.000250 Time 0.020248 -2023-02-13 18:24:03,688 - Epoch: [149][ 910/ 1207] Overall Loss 0.198652 Objective Loss 0.198652 LR 0.000250 Time 0.020239 -2023-02-13 18:24:03,885 - Epoch: [149][ 920/ 1207] Overall Loss 0.198755 Objective Loss 0.198755 LR 0.000250 Time 0.020232 -2023-02-13 18:24:04,080 - Epoch: [149][ 930/ 1207] Overall Loss 0.198774 Objective Loss 0.198774 LR 0.000250 Time 0.020224 -2023-02-13 18:24:04,277 - Epoch: [149][ 940/ 1207] Overall Loss 0.198975 Objective Loss 0.198975 LR 0.000250 Time 0.020218 -2023-02-13 18:24:04,472 - Epoch: [149][ 950/ 1207] Overall Loss 0.199132 Objective Loss 0.199132 LR 0.000250 Time 0.020210 -2023-02-13 18:24:04,668 - Epoch: [149][ 960/ 1207] Overall Loss 0.199151 Objective Loss 0.199151 LR 0.000250 Time 0.020204 -2023-02-13 18:24:04,863 - Epoch: [149][ 970/ 1207] Overall Loss 0.199022 Objective Loss 0.199022 LR 0.000250 Time 0.020196 -2023-02-13 18:24:05,060 - Epoch: [149][ 980/ 1207] Overall Loss 0.199196 Objective Loss 0.199196 LR 0.000250 Time 0.020191 -2023-02-13 18:24:05,256 - Epoch: [149][ 990/ 1207] Overall Loss 0.199178 Objective Loss 0.199178 LR 0.000250 Time 0.020184 -2023-02-13 18:24:05,453 - Epoch: [149][ 1000/ 1207] Overall Loss 0.199400 Objective Loss 0.199400 LR 0.000250 Time 0.020179 -2023-02-13 18:24:05,648 - Epoch: [149][ 1010/ 1207] Overall Loss 0.199656 Objective Loss 0.199656 LR 0.000250 Time 0.020172 -2023-02-13 18:24:05,844 - Epoch: [149][ 1020/ 1207] Overall Loss 0.199715 Objective Loss 0.199715 LR 0.000250 Time 0.020166 -2023-02-13 18:24:06,042 - Epoch: [149][ 1030/ 1207] Overall Loss 0.199836 Objective Loss 0.199836 LR 0.000250 Time 0.020162 -2023-02-13 18:24:06,238 - Epoch: [149][ 1040/ 1207] Overall Loss 0.199797 Objective Loss 0.199797 LR 0.000250 Time 0.020157 -2023-02-13 18:24:06,434 - Epoch: [149][ 1050/ 1207] Overall Loss 0.199726 Objective Loss 0.199726 LR 0.000250 Time 0.020151 -2023-02-13 18:24:06,631 - Epoch: [149][ 1060/ 1207] Overall Loss 0.199624 Objective Loss 0.199624 LR 0.000250 Time 0.020145 -2023-02-13 18:24:06,826 - Epoch: [149][ 1070/ 1207] Overall Loss 0.199827 Objective Loss 0.199827 LR 0.000250 Time 0.020140 -2023-02-13 18:24:07,024 - Epoch: [149][ 1080/ 1207] Overall Loss 0.199765 Objective Loss 0.199765 LR 0.000250 Time 0.020135 -2023-02-13 18:24:07,218 - Epoch: [149][ 1090/ 1207] Overall Loss 0.199812 Objective Loss 0.199812 LR 0.000250 Time 0.020129 -2023-02-13 18:24:07,415 - Epoch: [149][ 1100/ 1207] Overall Loss 0.200025 Objective Loss 0.200025 LR 0.000250 Time 0.020125 -2023-02-13 18:24:07,610 - Epoch: [149][ 1110/ 1207] Overall Loss 0.199996 Objective Loss 0.199996 LR 0.000250 Time 0.020118 -2023-02-13 18:24:07,807 - Epoch: [149][ 1120/ 1207] Overall Loss 0.199860 Objective Loss 0.199860 LR 0.000250 Time 0.020114 -2023-02-13 18:24:08,003 - Epoch: [149][ 1130/ 1207] Overall Loss 0.200015 Objective Loss 0.200015 LR 0.000250 Time 0.020109 -2023-02-13 18:24:08,199 - Epoch: [149][ 1140/ 1207] Overall Loss 0.200158 Objective Loss 0.200158 LR 0.000250 Time 0.020105 -2023-02-13 18:24:08,394 - Epoch: [149][ 1150/ 1207] Overall Loss 0.200109 Objective Loss 0.200109 LR 0.000250 Time 0.020099 -2023-02-13 18:24:08,589 - Epoch: [149][ 1160/ 1207] Overall Loss 0.200317 Objective Loss 0.200317 LR 0.000250 Time 0.020094 -2023-02-13 18:24:08,785 - Epoch: [149][ 1170/ 1207] Overall Loss 0.200375 Objective Loss 0.200375 LR 0.000250 Time 0.020089 -2023-02-13 18:24:08,982 - Epoch: [149][ 1180/ 1207] Overall Loss 0.200544 Objective Loss 0.200544 LR 0.000250 Time 0.020085 -2023-02-13 18:24:09,176 - Epoch: [149][ 1190/ 1207] Overall Loss 0.200402 Objective Loss 0.200402 LR 0.000250 Time 0.020080 -2023-02-13 18:24:09,424 - Epoch: [149][ 1200/ 1207] Overall Loss 0.200498 Objective Loss 0.200498 LR 0.000250 Time 0.020118 -2023-02-13 18:24:09,539 - Epoch: [149][ 1207/ 1207] Overall Loss 0.200369 Objective Loss 0.200369 Top1 88.109756 Top5 98.780488 LR 0.000250 Time 0.020097 -2023-02-13 18:24:09,623 - --- validate (epoch=149)----------- -2023-02-13 18:24:09,623 - 34311 samples (256 per mini-batch) -2023-02-13 18:24:10,025 - Epoch: [149][ 10/ 135] Loss 0.284692 Top1 86.132812 Top5 98.085938 -2023-02-13 18:24:10,156 - Epoch: [149][ 20/ 135] Loss 0.284017 Top1 86.132812 Top5 98.066406 -2023-02-13 18:24:10,300 - Epoch: [149][ 30/ 135] Loss 0.289254 Top1 86.093750 Top5 97.968750 -2023-02-13 18:24:10,423 - Epoch: [149][ 40/ 135] Loss 0.281066 Top1 86.113281 Top5 97.978516 -2023-02-13 18:24:10,549 - Epoch: [149][ 50/ 135] Loss 0.284583 Top1 86.101562 Top5 98.007812 -2023-02-13 18:24:10,677 - Epoch: [149][ 60/ 135] Loss 0.290117 Top1 86.165365 Top5 97.988281 -2023-02-13 18:24:10,804 - Epoch: [149][ 70/ 135] Loss 0.287679 Top1 86.344866 Top5 98.041295 -2023-02-13 18:24:10,933 - Epoch: [149][ 80/ 135] Loss 0.289371 Top1 86.284180 Top5 98.041992 -2023-02-13 18:24:11,056 - Epoch: [149][ 90/ 135] Loss 0.290242 Top1 86.271701 Top5 97.960069 -2023-02-13 18:24:11,181 - Epoch: [149][ 100/ 135] Loss 0.294621 Top1 86.187500 Top5 97.945312 -2023-02-13 18:24:11,303 - Epoch: [149][ 110/ 135] Loss 0.293953 Top1 86.154119 Top5 97.954545 -2023-02-13 18:24:11,427 - Epoch: [149][ 120/ 135] Loss 0.292643 Top1 86.113281 Top5 98.004557 -2023-02-13 18:24:11,551 - Epoch: [149][ 130/ 135] Loss 0.292140 Top1 86.123798 Top5 98.004808 -2023-02-13 18:24:11,595 - Epoch: [149][ 135/ 135] Loss 0.291082 Top1 86.094838 Top5 97.994812 -2023-02-13 18:24:11,671 - ==> Top1: 86.095 Top5: 97.995 Loss: 0.291 - -2023-02-13 18:24:11,672 - ==> Confusion: -[[ 862 6 6 0 11 3 0 1 4 45 0 5 1 3 6 4 2 2 1 0 5] - [ 4 967 0 1 6 17 2 14 4 1 1 1 0 0 1 2 3 0 1 0 8] - [ 5 6 963 11 4 1 12 14 0 1 3 0 3 5 5 7 0 4 6 2 6] - [ 4 1 19 898 2 4 0 1 2 0 13 0 10 0 28 2 5 4 16 0 7] - [ 11 9 0 1 991 8 1 2 1 0 0 7 2 3 9 5 7 2 0 3 4] - [ 1 20 0 3 3 973 4 19 2 4 2 9 3 10 0 3 4 0 1 4 5] - [ 2 5 14 0 0 7 1031 7 0 2 5 2 3 2 1 3 1 3 2 4 5] - [ 2 14 8 1 1 25 3 927 2 2 1 3 2 1 0 0 0 3 14 9 6] - [ 16 3 1 1 1 0 0 0 916 29 9 3 0 8 11 2 1 0 5 0 3] - [ 76 1 3 0 9 2 1 5 45 837 0 2 0 14 6 4 0 4 0 0 3] - [ 2 2 0 4 1 1 2 3 15 1 990 2 2 8 3 0 1 1 8 0 5] - [ 1 2 2 0 3 7 1 5 1 2 0 932 25 5 0 3 2 5 3 2 4] - [ 0 0 1 5 1 5 0 0 2 2 0 28 881 1 4 4 2 11 3 0 9] - [ 5 2 1 1 7 9 0 1 11 11 11 4 2 938 3 5 4 1 1 1 6] - [ 7 1 0 11 4 4 0 1 19 7 2 0 5 1 1008 1 2 4 4 0 11] - [ 4 2 7 0 9 1 5 0 0 0 0 6 8 1 0 977 10 6 0 6 4] - [ 2 4 0 1 9 1 0 1 2 0 0 3 3 1 3 9 1005 2 2 3 10] - [ 3 3 1 4 1 1 3 0 0 0 2 6 21 1 1 13 0 980 2 2 7] - [ 3 6 2 5 1 2 0 23 3 1 4 1 2 0 15 1 1 3 1010 2 1] - [ 1 4 1 1 1 6 3 7 0 0 0 22 3 4 0 5 4 4 0 1071 11] - [ 125 251 187 92 120 191 66 145 93 68 208 114 277 261 144 78 213 88 144 186 10383]] - -2023-02-13 18:24:11,674 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:24:11,674 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:24:11,680 - - -2023-02-13 18:24:11,680 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:24:12,573 - Epoch: [150][ 10/ 1207] Overall Loss 0.214720 Objective Loss 0.214720 LR 0.000250 Time 0.089182 -2023-02-13 18:24:12,777 - Epoch: [150][ 20/ 1207] Overall Loss 0.213031 Objective Loss 0.213031 LR 0.000250 Time 0.054799 -2023-02-13 18:24:12,971 - Epoch: [150][ 30/ 1207] Overall Loss 0.210364 Objective Loss 0.210364 LR 0.000250 Time 0.042959 -2023-02-13 18:24:13,167 - Epoch: [150][ 40/ 1207] Overall Loss 0.207448 Objective Loss 0.207448 LR 0.000250 Time 0.037129 -2023-02-13 18:24:13,361 - Epoch: [150][ 50/ 1207] Overall Loss 0.205435 Objective Loss 0.205435 LR 0.000250 Time 0.033562 -2023-02-13 18:24:13,557 - Epoch: [150][ 60/ 1207] Overall Loss 0.203176 Objective Loss 0.203176 LR 0.000250 Time 0.031225 -2023-02-13 18:24:13,749 - Epoch: [150][ 70/ 1207] Overall Loss 0.199772 Objective Loss 0.199772 LR 0.000250 Time 0.029508 -2023-02-13 18:24:13,945 - Epoch: [150][ 80/ 1207] Overall Loss 0.201643 Objective Loss 0.201643 LR 0.000250 Time 0.028264 -2023-02-13 18:24:14,138 - Epoch: [150][ 90/ 1207] Overall Loss 0.201885 Objective Loss 0.201885 LR 0.000250 Time 0.027266 -2023-02-13 18:24:14,334 - Epoch: [150][ 100/ 1207] Overall Loss 0.201769 Objective Loss 0.201769 LR 0.000250 Time 0.026492 -2023-02-13 18:24:14,527 - Epoch: [150][ 110/ 1207] Overall Loss 0.202878 Objective Loss 0.202878 LR 0.000250 Time 0.025833 -2023-02-13 18:24:14,723 - Epoch: [150][ 120/ 1207] Overall Loss 0.203603 Objective Loss 0.203603 LR 0.000250 Time 0.025311 -2023-02-13 18:24:14,915 - Epoch: [150][ 130/ 1207] Overall Loss 0.203067 Objective Loss 0.203067 LR 0.000250 Time 0.024842 -2023-02-13 18:24:15,111 - Epoch: [150][ 140/ 1207] Overall Loss 0.204608 Objective Loss 0.204608 LR 0.000250 Time 0.024464 -2023-02-13 18:24:15,305 - Epoch: [150][ 150/ 1207] Overall Loss 0.205443 Objective Loss 0.205443 LR 0.000250 Time 0.024121 -2023-02-13 18:24:15,501 - Epoch: [150][ 160/ 1207] Overall Loss 0.205005 Objective Loss 0.205005 LR 0.000250 Time 0.023837 -2023-02-13 18:24:15,694 - Epoch: [150][ 170/ 1207] Overall Loss 0.204519 Objective Loss 0.204519 LR 0.000250 Time 0.023568 -2023-02-13 18:24:15,890 - Epoch: [150][ 180/ 1207] Overall Loss 0.202957 Objective Loss 0.202957 LR 0.000250 Time 0.023348 -2023-02-13 18:24:16,084 - Epoch: [150][ 190/ 1207] Overall Loss 0.203722 Objective Loss 0.203722 LR 0.000250 Time 0.023135 -2023-02-13 18:24:16,279 - Epoch: [150][ 200/ 1207] Overall Loss 0.204163 Objective Loss 0.204163 LR 0.000250 Time 0.022953 -2023-02-13 18:24:16,472 - Epoch: [150][ 210/ 1207] Overall Loss 0.203894 Objective Loss 0.203894 LR 0.000250 Time 0.022777 -2023-02-13 18:24:16,667 - Epoch: [150][ 220/ 1207] Overall Loss 0.203915 Objective Loss 0.203915 LR 0.000250 Time 0.022629 -2023-02-13 18:24:16,860 - Epoch: [150][ 230/ 1207] Overall Loss 0.203379 Objective Loss 0.203379 LR 0.000250 Time 0.022481 -2023-02-13 18:24:17,057 - Epoch: [150][ 240/ 1207] Overall Loss 0.203631 Objective Loss 0.203631 LR 0.000250 Time 0.022364 -2023-02-13 18:24:17,249 - Epoch: [150][ 250/ 1207] Overall Loss 0.203970 Objective Loss 0.203970 LR 0.000250 Time 0.022237 -2023-02-13 18:24:17,444 - Epoch: [150][ 260/ 1207] Overall Loss 0.203157 Objective Loss 0.203157 LR 0.000250 Time 0.022129 -2023-02-13 18:24:17,637 - Epoch: [150][ 270/ 1207] Overall Loss 0.203350 Objective Loss 0.203350 LR 0.000250 Time 0.022021 -2023-02-13 18:24:17,832 - Epoch: [150][ 280/ 1207] Overall Loss 0.202680 Objective Loss 0.202680 LR 0.000250 Time 0.021930 -2023-02-13 18:24:18,025 - Epoch: [150][ 290/ 1207] Overall Loss 0.203281 Objective Loss 0.203281 LR 0.000250 Time 0.021840 -2023-02-13 18:24:18,220 - Epoch: [150][ 300/ 1207] Overall Loss 0.202758 Objective Loss 0.202758 LR 0.000250 Time 0.021761 -2023-02-13 18:24:18,413 - Epoch: [150][ 310/ 1207] Overall Loss 0.202848 Objective Loss 0.202848 LR 0.000250 Time 0.021679 -2023-02-13 18:24:18,609 - Epoch: [150][ 320/ 1207] Overall Loss 0.202814 Objective Loss 0.202814 LR 0.000250 Time 0.021612 -2023-02-13 18:24:18,801 - Epoch: [150][ 330/ 1207] Overall Loss 0.202572 Objective Loss 0.202572 LR 0.000250 Time 0.021538 -2023-02-13 18:24:18,997 - Epoch: [150][ 340/ 1207] Overall Loss 0.201936 Objective Loss 0.201936 LR 0.000250 Time 0.021480 -2023-02-13 18:24:19,190 - Epoch: [150][ 350/ 1207] Overall Loss 0.201621 Objective Loss 0.201621 LR 0.000250 Time 0.021416 -2023-02-13 18:24:19,385 - Epoch: [150][ 360/ 1207] Overall Loss 0.200963 Objective Loss 0.200963 LR 0.000250 Time 0.021364 -2023-02-13 18:24:19,578 - Epoch: [150][ 370/ 1207] Overall Loss 0.200496 Objective Loss 0.200496 LR 0.000250 Time 0.021307 -2023-02-13 18:24:19,774 - Epoch: [150][ 380/ 1207] Overall Loss 0.200392 Objective Loss 0.200392 LR 0.000250 Time 0.021260 -2023-02-13 18:24:19,966 - Epoch: [150][ 390/ 1207] Overall Loss 0.200333 Objective Loss 0.200333 LR 0.000250 Time 0.021207 -2023-02-13 18:24:20,161 - Epoch: [150][ 400/ 1207] Overall Loss 0.200305 Objective Loss 0.200305 LR 0.000250 Time 0.021164 -2023-02-13 18:24:20,353 - Epoch: [150][ 410/ 1207] Overall Loss 0.200411 Objective Loss 0.200411 LR 0.000250 Time 0.021115 -2023-02-13 18:24:20,549 - Epoch: [150][ 420/ 1207] Overall Loss 0.200218 Objective Loss 0.200218 LR 0.000250 Time 0.021078 -2023-02-13 18:24:20,742 - Epoch: [150][ 430/ 1207] Overall Loss 0.199772 Objective Loss 0.199772 LR 0.000250 Time 0.021035 -2023-02-13 18:24:20,938 - Epoch: [150][ 440/ 1207] Overall Loss 0.199967 Objective Loss 0.199967 LR 0.000250 Time 0.021001 -2023-02-13 18:24:21,130 - Epoch: [150][ 450/ 1207] Overall Loss 0.200084 Objective Loss 0.200084 LR 0.000250 Time 0.020961 -2023-02-13 18:24:21,326 - Epoch: [150][ 460/ 1207] Overall Loss 0.200334 Objective Loss 0.200334 LR 0.000250 Time 0.020930 -2023-02-13 18:24:21,518 - Epoch: [150][ 470/ 1207] Overall Loss 0.200146 Objective Loss 0.200146 LR 0.000250 Time 0.020894 -2023-02-13 18:24:21,713 - Epoch: [150][ 480/ 1207] Overall Loss 0.199981 Objective Loss 0.199981 LR 0.000250 Time 0.020863 -2023-02-13 18:24:21,905 - Epoch: [150][ 490/ 1207] Overall Loss 0.199554 Objective Loss 0.199554 LR 0.000250 Time 0.020829 -2023-02-13 18:24:22,101 - Epoch: [150][ 500/ 1207] Overall Loss 0.199673 Objective Loss 0.199673 LR 0.000250 Time 0.020804 -2023-02-13 18:24:22,294 - Epoch: [150][ 510/ 1207] Overall Loss 0.199325 Objective Loss 0.199325 LR 0.000250 Time 0.020773 -2023-02-13 18:24:22,489 - Epoch: [150][ 520/ 1207] Overall Loss 0.199417 Objective Loss 0.199417 LR 0.000250 Time 0.020748 -2023-02-13 18:24:22,682 - Epoch: [150][ 530/ 1207] Overall Loss 0.200116 Objective Loss 0.200116 LR 0.000250 Time 0.020719 -2023-02-13 18:24:22,878 - Epoch: [150][ 540/ 1207] Overall Loss 0.200613 Objective Loss 0.200613 LR 0.000250 Time 0.020697 -2023-02-13 18:24:23,071 - Epoch: [150][ 550/ 1207] Overall Loss 0.200476 Objective Loss 0.200476 LR 0.000250 Time 0.020671 -2023-02-13 18:24:23,266 - Epoch: [150][ 560/ 1207] Overall Loss 0.200290 Objective Loss 0.200290 LR 0.000250 Time 0.020651 -2023-02-13 18:24:23,459 - Epoch: [150][ 570/ 1207] Overall Loss 0.200561 Objective Loss 0.200561 LR 0.000250 Time 0.020626 -2023-02-13 18:24:23,654 - Epoch: [150][ 580/ 1207] Overall Loss 0.200824 Objective Loss 0.200824 LR 0.000250 Time 0.020607 -2023-02-13 18:24:23,847 - Epoch: [150][ 590/ 1207] Overall Loss 0.200571 Objective Loss 0.200571 LR 0.000250 Time 0.020583 -2023-02-13 18:24:24,043 - Epoch: [150][ 600/ 1207] Overall Loss 0.200467 Objective Loss 0.200467 LR 0.000250 Time 0.020566 -2023-02-13 18:24:24,235 - Epoch: [150][ 610/ 1207] Overall Loss 0.200578 Objective Loss 0.200578 LR 0.000250 Time 0.020543 -2023-02-13 18:24:24,430 - Epoch: [150][ 620/ 1207] Overall Loss 0.200896 Objective Loss 0.200896 LR 0.000250 Time 0.020525 -2023-02-13 18:24:24,622 - Epoch: [150][ 630/ 1207] Overall Loss 0.200799 Objective Loss 0.200799 LR 0.000250 Time 0.020504 -2023-02-13 18:24:24,818 - Epoch: [150][ 640/ 1207] Overall Loss 0.201103 Objective Loss 0.201103 LR 0.000250 Time 0.020489 -2023-02-13 18:24:25,011 - Epoch: [150][ 650/ 1207] Overall Loss 0.201233 Objective Loss 0.201233 LR 0.000250 Time 0.020470 -2023-02-13 18:24:25,206 - Epoch: [150][ 660/ 1207] Overall Loss 0.201444 Objective Loss 0.201444 LR 0.000250 Time 0.020456 -2023-02-13 18:24:25,399 - Epoch: [150][ 670/ 1207] Overall Loss 0.200973 Objective Loss 0.200973 LR 0.000250 Time 0.020438 -2023-02-13 18:24:25,595 - Epoch: [150][ 680/ 1207] Overall Loss 0.201135 Objective Loss 0.201135 LR 0.000250 Time 0.020425 -2023-02-13 18:24:25,789 - Epoch: [150][ 690/ 1207] Overall Loss 0.201360 Objective Loss 0.201360 LR 0.000250 Time 0.020409 -2023-02-13 18:24:25,986 - Epoch: [150][ 700/ 1207] Overall Loss 0.201513 Objective Loss 0.201513 LR 0.000250 Time 0.020398 -2023-02-13 18:24:26,179 - Epoch: [150][ 710/ 1207] Overall Loss 0.201762 Objective Loss 0.201762 LR 0.000250 Time 0.020382 -2023-02-13 18:24:26,374 - Epoch: [150][ 720/ 1207] Overall Loss 0.202075 Objective Loss 0.202075 LR 0.000250 Time 0.020370 -2023-02-13 18:24:26,568 - Epoch: [150][ 730/ 1207] Overall Loss 0.202055 Objective Loss 0.202055 LR 0.000250 Time 0.020355 -2023-02-13 18:24:26,764 - Epoch: [150][ 740/ 1207] Overall Loss 0.202313 Objective Loss 0.202313 LR 0.000250 Time 0.020345 -2023-02-13 18:24:26,957 - Epoch: [150][ 750/ 1207] Overall Loss 0.202435 Objective Loss 0.202435 LR 0.000250 Time 0.020331 -2023-02-13 18:24:27,154 - Epoch: [150][ 760/ 1207] Overall Loss 0.202457 Objective Loss 0.202457 LR 0.000250 Time 0.020321 -2023-02-13 18:24:27,346 - Epoch: [150][ 770/ 1207] Overall Loss 0.202479 Objective Loss 0.202479 LR 0.000250 Time 0.020307 -2023-02-13 18:24:27,543 - Epoch: [150][ 780/ 1207] Overall Loss 0.202326 Objective Loss 0.202326 LR 0.000250 Time 0.020298 -2023-02-13 18:24:27,736 - Epoch: [150][ 790/ 1207] Overall Loss 0.202677 Objective Loss 0.202677 LR 0.000250 Time 0.020286 -2023-02-13 18:24:27,932 - Epoch: [150][ 800/ 1207] Overall Loss 0.202661 Objective Loss 0.202661 LR 0.000250 Time 0.020277 -2023-02-13 18:24:28,125 - Epoch: [150][ 810/ 1207] Overall Loss 0.202662 Objective Loss 0.202662 LR 0.000250 Time 0.020264 -2023-02-13 18:24:28,320 - Epoch: [150][ 820/ 1207] Overall Loss 0.202828 Objective Loss 0.202828 LR 0.000250 Time 0.020254 -2023-02-13 18:24:28,512 - Epoch: [150][ 830/ 1207] Overall Loss 0.202777 Objective Loss 0.202777 LR 0.000250 Time 0.020241 -2023-02-13 18:24:28,707 - Epoch: [150][ 840/ 1207] Overall Loss 0.202801 Objective Loss 0.202801 LR 0.000250 Time 0.020232 -2023-02-13 18:24:28,901 - Epoch: [150][ 850/ 1207] Overall Loss 0.202619 Objective Loss 0.202619 LR 0.000250 Time 0.020221 -2023-02-13 18:24:29,096 - Epoch: [150][ 860/ 1207] Overall Loss 0.202702 Objective Loss 0.202702 LR 0.000250 Time 0.020213 -2023-02-13 18:24:29,289 - Epoch: [150][ 870/ 1207] Overall Loss 0.202933 Objective Loss 0.202933 LR 0.000250 Time 0.020202 -2023-02-13 18:24:29,484 - Epoch: [150][ 880/ 1207] Overall Loss 0.203163 Objective Loss 0.203163 LR 0.000250 Time 0.020194 -2023-02-13 18:24:29,682 - Epoch: [150][ 890/ 1207] Overall Loss 0.203369 Objective Loss 0.203369 LR 0.000250 Time 0.020188 -2023-02-13 18:24:29,891 - Epoch: [150][ 900/ 1207] Overall Loss 0.203861 Objective Loss 0.203861 LR 0.000250 Time 0.020196 -2023-02-13 18:24:30,094 - Epoch: [150][ 910/ 1207] Overall Loss 0.203556 Objective Loss 0.203556 LR 0.000250 Time 0.020197 -2023-02-13 18:24:30,304 - Epoch: [150][ 920/ 1207] Overall Loss 0.203310 Objective Loss 0.203310 LR 0.000250 Time 0.020205 -2023-02-13 18:24:30,507 - Epoch: [150][ 930/ 1207] Overall Loss 0.203333 Objective Loss 0.203333 LR 0.000250 Time 0.020205 -2023-02-13 18:24:30,716 - Epoch: [150][ 940/ 1207] Overall Loss 0.203320 Objective Loss 0.203320 LR 0.000250 Time 0.020213 -2023-02-13 18:24:30,920 - Epoch: [150][ 950/ 1207] Overall Loss 0.203462 Objective Loss 0.203462 LR 0.000250 Time 0.020215 -2023-02-13 18:24:31,130 - Epoch: [150][ 960/ 1207] Overall Loss 0.203456 Objective Loss 0.203456 LR 0.000250 Time 0.020223 -2023-02-13 18:24:31,333 - Epoch: [150][ 970/ 1207] Overall Loss 0.203381 Objective Loss 0.203381 LR 0.000250 Time 0.020223 -2023-02-13 18:24:31,543 - Epoch: [150][ 980/ 1207] Overall Loss 0.203280 Objective Loss 0.203280 LR 0.000250 Time 0.020230 -2023-02-13 18:24:31,746 - Epoch: [150][ 990/ 1207] Overall Loss 0.203139 Objective Loss 0.203139 LR 0.000250 Time 0.020230 -2023-02-13 18:24:31,956 - Epoch: [150][ 1000/ 1207] Overall Loss 0.202925 Objective Loss 0.202925 LR 0.000250 Time 0.020237 -2023-02-13 18:24:32,158 - Epoch: [150][ 1010/ 1207] Overall Loss 0.203225 Objective Loss 0.203225 LR 0.000250 Time 0.020237 -2023-02-13 18:24:32,355 - Epoch: [150][ 1020/ 1207] Overall Loss 0.203292 Objective Loss 0.203292 LR 0.000250 Time 0.020231 -2023-02-13 18:24:32,548 - Epoch: [150][ 1030/ 1207] Overall Loss 0.203252 Objective Loss 0.203252 LR 0.000250 Time 0.020221 -2023-02-13 18:24:32,744 - Epoch: [150][ 1040/ 1207] Overall Loss 0.203137 Objective Loss 0.203137 LR 0.000250 Time 0.020215 -2023-02-13 18:24:32,936 - Epoch: [150][ 1050/ 1207] Overall Loss 0.203083 Objective Loss 0.203083 LR 0.000250 Time 0.020206 -2023-02-13 18:24:33,133 - Epoch: [150][ 1060/ 1207] Overall Loss 0.203281 Objective Loss 0.203281 LR 0.000250 Time 0.020200 -2023-02-13 18:24:33,325 - Epoch: [150][ 1070/ 1207] Overall Loss 0.203116 Objective Loss 0.203116 LR 0.000250 Time 0.020191 -2023-02-13 18:24:33,521 - Epoch: [150][ 1080/ 1207] Overall Loss 0.202965 Objective Loss 0.202965 LR 0.000250 Time 0.020185 -2023-02-13 18:24:33,714 - Epoch: [150][ 1090/ 1207] Overall Loss 0.202964 Objective Loss 0.202964 LR 0.000250 Time 0.020176 -2023-02-13 18:24:33,910 - Epoch: [150][ 1100/ 1207] Overall Loss 0.202853 Objective Loss 0.202853 LR 0.000250 Time 0.020170 -2023-02-13 18:24:34,103 - Epoch: [150][ 1110/ 1207] Overall Loss 0.202863 Objective Loss 0.202863 LR 0.000250 Time 0.020162 -2023-02-13 18:24:34,298 - Epoch: [150][ 1120/ 1207] Overall Loss 0.202907 Objective Loss 0.202907 LR 0.000250 Time 0.020157 -2023-02-13 18:24:34,491 - Epoch: [150][ 1130/ 1207] Overall Loss 0.202866 Objective Loss 0.202866 LR 0.000250 Time 0.020148 -2023-02-13 18:24:34,687 - Epoch: [150][ 1140/ 1207] Overall Loss 0.203020 Objective Loss 0.203020 LR 0.000250 Time 0.020144 -2023-02-13 18:24:34,879 - Epoch: [150][ 1150/ 1207] Overall Loss 0.202949 Objective Loss 0.202949 LR 0.000250 Time 0.020135 -2023-02-13 18:24:35,075 - Epoch: [150][ 1160/ 1207] Overall Loss 0.202784 Objective Loss 0.202784 LR 0.000250 Time 0.020130 -2023-02-13 18:24:35,268 - Epoch: [150][ 1170/ 1207] Overall Loss 0.202865 Objective Loss 0.202865 LR 0.000250 Time 0.020123 -2023-02-13 18:24:35,465 - Epoch: [150][ 1180/ 1207] Overall Loss 0.202692 Objective Loss 0.202692 LR 0.000250 Time 0.020118 -2023-02-13 18:24:35,657 - Epoch: [150][ 1190/ 1207] Overall Loss 0.202802 Objective Loss 0.202802 LR 0.000250 Time 0.020111 -2023-02-13 18:24:35,908 - Epoch: [150][ 1200/ 1207] Overall Loss 0.202923 Objective Loss 0.202923 LR 0.000250 Time 0.020152 -2023-02-13 18:24:36,024 - Epoch: [150][ 1207/ 1207] Overall Loss 0.202916 Objective Loss 0.202916 Top1 84.451220 Top5 98.170732 LR 0.000250 Time 0.020131 -2023-02-13 18:24:36,114 - --- validate (epoch=150)----------- -2023-02-13 18:24:36,114 - 34311 samples (256 per mini-batch) -2023-02-13 18:24:36,516 - Epoch: [150][ 10/ 135] Loss 0.293745 Top1 85.000000 Top5 97.695312 -2023-02-13 18:24:36,648 - Epoch: [150][ 20/ 135] Loss 0.289417 Top1 85.351562 Top5 97.832031 -2023-02-13 18:24:36,781 - Epoch: [150][ 30/ 135] Loss 0.292606 Top1 85.625000 Top5 97.877604 -2023-02-13 18:24:36,912 - Epoch: [150][ 40/ 135] Loss 0.289503 Top1 85.488281 Top5 97.978516 -2023-02-13 18:24:37,041 - Epoch: [150][ 50/ 135] Loss 0.294064 Top1 85.320312 Top5 97.929688 -2023-02-13 18:24:37,161 - Epoch: [150][ 60/ 135] Loss 0.297968 Top1 85.201823 Top5 97.858073 -2023-02-13 18:24:37,285 - Epoch: [150][ 70/ 135] Loss 0.295819 Top1 85.217634 Top5 97.862723 -2023-02-13 18:24:37,408 - Epoch: [150][ 80/ 135] Loss 0.298811 Top1 85.278320 Top5 97.880859 -2023-02-13 18:24:37,531 - Epoch: [150][ 90/ 135] Loss 0.296475 Top1 85.312500 Top5 97.907986 -2023-02-13 18:24:37,653 - Epoch: [150][ 100/ 135] Loss 0.297946 Top1 85.210938 Top5 97.898438 -2023-02-13 18:24:37,776 - Epoch: [150][ 110/ 135] Loss 0.296890 Top1 85.301847 Top5 97.876420 -2023-02-13 18:24:37,914 - Epoch: [150][ 120/ 135] Loss 0.297272 Top1 85.253906 Top5 97.890625 -2023-02-13 18:24:38,045 - Epoch: [150][ 130/ 135] Loss 0.297929 Top1 85.267428 Top5 97.872596 -2023-02-13 18:24:38,089 - Epoch: [150][ 135/ 135] Loss 0.297181 Top1 85.264201 Top5 97.878231 -2023-02-13 18:24:38,162 - ==> Top1: 85.264 Top5: 97.878 Loss: 0.297 - -2023-02-13 18:24:38,163 - ==> Confusion: -[[ 849 2 8 1 10 5 0 0 4 53 1 4 1 4 8 4 3 3 0 1 6] - [ 2 975 1 3 7 13 3 9 3 1 1 1 1 0 0 1 4 0 1 2 5] - [ 5 6 955 16 2 1 20 11 0 1 4 2 1 6 4 5 4 1 3 5 6] - [ 2 0 19 907 6 2 2 1 3 0 11 0 4 0 21 2 4 4 19 0 9] - [ 11 8 0 2 1004 8 1 1 0 2 0 4 2 2 8 4 3 0 0 3 3] - [ 0 22 1 4 5 964 3 16 0 3 1 15 4 13 0 3 4 0 2 3 7] - [ 2 5 10 2 1 6 1042 7 1 2 2 2 2 1 0 3 1 1 2 4 3] - [ 1 18 11 3 4 25 5 917 0 2 1 1 1 1 0 1 1 2 17 6 7] - [ 11 6 1 1 0 0 0 0 897 42 7 3 0 10 19 3 2 1 5 0 1] - [ 56 2 6 1 7 2 0 2 35 868 0 1 0 14 8 3 0 2 1 0 4] - [ 2 2 2 5 3 0 4 4 12 1 987 1 3 7 3 0 1 2 7 0 5] - [ 2 4 2 0 3 10 1 5 4 2 0 918 18 9 2 3 4 7 2 6 3] - [ 0 1 2 10 2 3 0 1 3 1 1 30 855 1 5 6 2 18 7 1 10] - [ 7 2 2 1 10 5 0 1 11 14 7 6 2 940 6 3 2 2 0 0 3] - [ 4 0 0 13 2 4 0 2 14 4 3 1 3 0 1020 1 4 2 8 0 7] - [ 1 4 7 0 5 0 4 1 0 0 0 6 8 1 2 970 12 11 1 7 6] - [ 1 8 1 1 10 2 0 1 1 0 0 1 1 0 2 7 1003 2 1 4 15] - [ 2 5 0 5 1 0 4 0 0 1 2 7 10 1 0 16 0 990 1 1 5] - [ 2 6 6 7 1 1 0 22 3 0 5 1 2 0 10 1 2 4 1009 1 3] - [ 2 7 2 0 1 7 7 8 1 0 0 15 4 2 0 5 8 4 0 1068 7] - [ 121 280 203 112 153 178 90 147 90 81 179 119 273 290 186 86 245 100 170 214 10117]] - -2023-02-13 18:24:38,165 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:24:38,165 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:24:38,171 - - -2023-02-13 18:24:38,171 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:24:39,160 - Epoch: [151][ 10/ 1207] Overall Loss 0.189332 Objective Loss 0.189332 LR 0.000250 Time 0.098822 -2023-02-13 18:24:39,351 - Epoch: [151][ 20/ 1207] Overall Loss 0.195183 Objective Loss 0.195183 LR 0.000250 Time 0.058949 -2023-02-13 18:24:39,541 - Epoch: [151][ 30/ 1207] Overall Loss 0.195388 Objective Loss 0.195388 LR 0.000250 Time 0.045605 -2023-02-13 18:24:39,729 - Epoch: [151][ 40/ 1207] Overall Loss 0.191351 Objective Loss 0.191351 LR 0.000250 Time 0.038901 -2023-02-13 18:24:39,917 - Epoch: [151][ 50/ 1207] Overall Loss 0.195075 Objective Loss 0.195075 LR 0.000250 Time 0.034886 -2023-02-13 18:24:40,106 - Epoch: [151][ 60/ 1207] Overall Loss 0.192867 Objective Loss 0.192867 LR 0.000250 Time 0.032212 -2023-02-13 18:24:40,295 - Epoch: [151][ 70/ 1207] Overall Loss 0.193729 Objective Loss 0.193729 LR 0.000250 Time 0.030297 -2023-02-13 18:24:40,483 - Epoch: [151][ 80/ 1207] Overall Loss 0.194023 Objective Loss 0.194023 LR 0.000250 Time 0.028862 -2023-02-13 18:24:40,672 - Epoch: [151][ 90/ 1207] Overall Loss 0.195590 Objective Loss 0.195590 LR 0.000250 Time 0.027743 -2023-02-13 18:24:40,860 - Epoch: [151][ 100/ 1207] Overall Loss 0.195969 Objective Loss 0.195969 LR 0.000250 Time 0.026851 -2023-02-13 18:24:41,050 - Epoch: [151][ 110/ 1207] Overall Loss 0.196044 Objective Loss 0.196044 LR 0.000250 Time 0.026131 -2023-02-13 18:24:41,239 - Epoch: [151][ 120/ 1207] Overall Loss 0.196376 Objective Loss 0.196376 LR 0.000250 Time 0.025526 -2023-02-13 18:24:41,427 - Epoch: [151][ 130/ 1207] Overall Loss 0.197930 Objective Loss 0.197930 LR 0.000250 Time 0.025010 -2023-02-13 18:24:41,616 - Epoch: [151][ 140/ 1207] Overall Loss 0.197785 Objective Loss 0.197785 LR 0.000250 Time 0.024568 -2023-02-13 18:24:41,804 - Epoch: [151][ 150/ 1207] Overall Loss 0.198261 Objective Loss 0.198261 LR 0.000250 Time 0.024183 -2023-02-13 18:24:41,998 - Epoch: [151][ 160/ 1207] Overall Loss 0.198404 Objective Loss 0.198404 LR 0.000250 Time 0.023881 -2023-02-13 18:24:42,203 - Epoch: [151][ 170/ 1207] Overall Loss 0.197746 Objective Loss 0.197746 LR 0.000250 Time 0.023676 -2023-02-13 18:24:42,402 - Epoch: [151][ 180/ 1207] Overall Loss 0.197063 Objective Loss 0.197063 LR 0.000250 Time 0.023466 -2023-02-13 18:24:42,604 - Epoch: [151][ 190/ 1207] Overall Loss 0.196614 Objective Loss 0.196614 LR 0.000250 Time 0.023292 -2023-02-13 18:24:42,802 - Epoch: [151][ 200/ 1207] Overall Loss 0.195834 Objective Loss 0.195834 LR 0.000250 Time 0.023117 -2023-02-13 18:24:43,005 - Epoch: [151][ 210/ 1207] Overall Loss 0.195445 Objective Loss 0.195445 LR 0.000250 Time 0.022981 -2023-02-13 18:24:43,205 - Epoch: [151][ 220/ 1207] Overall Loss 0.196468 Objective Loss 0.196468 LR 0.000250 Time 0.022844 -2023-02-13 18:24:43,410 - Epoch: [151][ 230/ 1207] Overall Loss 0.196805 Objective Loss 0.196805 LR 0.000250 Time 0.022740 -2023-02-13 18:24:43,609 - Epoch: [151][ 240/ 1207] Overall Loss 0.196968 Objective Loss 0.196968 LR 0.000250 Time 0.022619 -2023-02-13 18:24:43,804 - Epoch: [151][ 250/ 1207] Overall Loss 0.198453 Objective Loss 0.198453 LR 0.000250 Time 0.022495 -2023-02-13 18:24:43,993 - Epoch: [151][ 260/ 1207] Overall Loss 0.198829 Objective Loss 0.198829 LR 0.000250 Time 0.022355 -2023-02-13 18:24:44,183 - Epoch: [151][ 270/ 1207] Overall Loss 0.199545 Objective Loss 0.199545 LR 0.000250 Time 0.022226 -2023-02-13 18:24:44,370 - Epoch: [151][ 280/ 1207] Overall Loss 0.200218 Objective Loss 0.200218 LR 0.000250 Time 0.022103 -2023-02-13 18:24:44,558 - Epoch: [151][ 290/ 1207] Overall Loss 0.200260 Objective Loss 0.200260 LR 0.000250 Time 0.021985 -2023-02-13 18:24:44,746 - Epoch: [151][ 300/ 1207] Overall Loss 0.200652 Objective Loss 0.200652 LR 0.000250 Time 0.021877 -2023-02-13 18:24:44,933 - Epoch: [151][ 310/ 1207] Overall Loss 0.200333 Objective Loss 0.200333 LR 0.000250 Time 0.021776 -2023-02-13 18:24:45,121 - Epoch: [151][ 320/ 1207] Overall Loss 0.199747 Objective Loss 0.199747 LR 0.000250 Time 0.021682 -2023-02-13 18:24:45,309 - Epoch: [151][ 330/ 1207] Overall Loss 0.199722 Objective Loss 0.199722 LR 0.000250 Time 0.021592 -2023-02-13 18:24:45,497 - Epoch: [151][ 340/ 1207] Overall Loss 0.199698 Objective Loss 0.199698 LR 0.000250 Time 0.021508 -2023-02-13 18:24:45,685 - Epoch: [151][ 350/ 1207] Overall Loss 0.199983 Objective Loss 0.199983 LR 0.000250 Time 0.021430 -2023-02-13 18:24:45,873 - Epoch: [151][ 360/ 1207] Overall Loss 0.200304 Objective Loss 0.200304 LR 0.000250 Time 0.021355 -2023-02-13 18:24:46,062 - Epoch: [151][ 370/ 1207] Overall Loss 0.200064 Objective Loss 0.200064 LR 0.000250 Time 0.021290 -2023-02-13 18:24:46,251 - Epoch: [151][ 380/ 1207] Overall Loss 0.199486 Objective Loss 0.199486 LR 0.000250 Time 0.021224 -2023-02-13 18:24:46,439 - Epoch: [151][ 390/ 1207] Overall Loss 0.199373 Objective Loss 0.199373 LR 0.000250 Time 0.021163 -2023-02-13 18:24:46,628 - Epoch: [151][ 400/ 1207] Overall Loss 0.199585 Objective Loss 0.199585 LR 0.000250 Time 0.021104 -2023-02-13 18:24:46,817 - Epoch: [151][ 410/ 1207] Overall Loss 0.200022 Objective Loss 0.200022 LR 0.000250 Time 0.021049 -2023-02-13 18:24:47,007 - Epoch: [151][ 420/ 1207] Overall Loss 0.200184 Objective Loss 0.200184 LR 0.000250 Time 0.021001 -2023-02-13 18:24:47,198 - Epoch: [151][ 430/ 1207] Overall Loss 0.199723 Objective Loss 0.199723 LR 0.000250 Time 0.020955 -2023-02-13 18:24:47,388 - Epoch: [151][ 440/ 1207] Overall Loss 0.199701 Objective Loss 0.199701 LR 0.000250 Time 0.020911 -2023-02-13 18:24:47,578 - Epoch: [151][ 450/ 1207] Overall Loss 0.199973 Objective Loss 0.199973 LR 0.000250 Time 0.020866 -2023-02-13 18:24:47,768 - Epoch: [151][ 460/ 1207] Overall Loss 0.200133 Objective Loss 0.200133 LR 0.000250 Time 0.020825 -2023-02-13 18:24:47,958 - Epoch: [151][ 470/ 1207] Overall Loss 0.199450 Objective Loss 0.199450 LR 0.000250 Time 0.020785 -2023-02-13 18:24:48,148 - Epoch: [151][ 480/ 1207] Overall Loss 0.198872 Objective Loss 0.198872 LR 0.000250 Time 0.020749 -2023-02-13 18:24:48,338 - Epoch: [151][ 490/ 1207] Overall Loss 0.199046 Objective Loss 0.199046 LR 0.000250 Time 0.020712 -2023-02-13 18:24:48,529 - Epoch: [151][ 500/ 1207] Overall Loss 0.198882 Objective Loss 0.198882 LR 0.000250 Time 0.020678 -2023-02-13 18:24:48,719 - Epoch: [151][ 510/ 1207] Overall Loss 0.198914 Objective Loss 0.198914 LR 0.000250 Time 0.020644 -2023-02-13 18:24:48,909 - Epoch: [151][ 520/ 1207] Overall Loss 0.199038 Objective Loss 0.199038 LR 0.000250 Time 0.020612 -2023-02-13 18:24:49,100 - Epoch: [151][ 530/ 1207] Overall Loss 0.199197 Objective Loss 0.199197 LR 0.000250 Time 0.020583 -2023-02-13 18:24:49,290 - Epoch: [151][ 540/ 1207] Overall Loss 0.199243 Objective Loss 0.199243 LR 0.000250 Time 0.020552 -2023-02-13 18:24:49,479 - Epoch: [151][ 550/ 1207] Overall Loss 0.199410 Objective Loss 0.199410 LR 0.000250 Time 0.020523 -2023-02-13 18:24:49,669 - Epoch: [151][ 560/ 1207] Overall Loss 0.199435 Objective Loss 0.199435 LR 0.000250 Time 0.020494 -2023-02-13 18:24:49,859 - Epoch: [151][ 570/ 1207] Overall Loss 0.199495 Objective Loss 0.199495 LR 0.000250 Time 0.020468 -2023-02-13 18:24:50,049 - Epoch: [151][ 580/ 1207] Overall Loss 0.199845 Objective Loss 0.199845 LR 0.000250 Time 0.020442 -2023-02-13 18:24:50,239 - Epoch: [151][ 590/ 1207] Overall Loss 0.200194 Objective Loss 0.200194 LR 0.000250 Time 0.020417 -2023-02-13 18:24:50,430 - Epoch: [151][ 600/ 1207] Overall Loss 0.200200 Objective Loss 0.200200 LR 0.000250 Time 0.020394 -2023-02-13 18:24:50,619 - Epoch: [151][ 610/ 1207] Overall Loss 0.200242 Objective Loss 0.200242 LR 0.000250 Time 0.020370 -2023-02-13 18:24:50,809 - Epoch: [151][ 620/ 1207] Overall Loss 0.200578 Objective Loss 0.200578 LR 0.000250 Time 0.020347 -2023-02-13 18:24:51,001 - Epoch: [151][ 630/ 1207] Overall Loss 0.200644 Objective Loss 0.200644 LR 0.000250 Time 0.020328 -2023-02-13 18:24:51,192 - Epoch: [151][ 640/ 1207] Overall Loss 0.200943 Objective Loss 0.200943 LR 0.000250 Time 0.020308 -2023-02-13 18:24:51,382 - Epoch: [151][ 650/ 1207] Overall Loss 0.201012 Objective Loss 0.201012 LR 0.000250 Time 0.020287 -2023-02-13 18:24:51,572 - Epoch: [151][ 660/ 1207] Overall Loss 0.201074 Objective Loss 0.201074 LR 0.000250 Time 0.020267 -2023-02-13 18:24:51,762 - Epoch: [151][ 670/ 1207] Overall Loss 0.201392 Objective Loss 0.201392 LR 0.000250 Time 0.020247 -2023-02-13 18:24:51,952 - Epoch: [151][ 680/ 1207] Overall Loss 0.201249 Objective Loss 0.201249 LR 0.000250 Time 0.020229 -2023-02-13 18:24:52,143 - Epoch: [151][ 690/ 1207] Overall Loss 0.201085 Objective Loss 0.201085 LR 0.000250 Time 0.020212 -2023-02-13 18:24:52,333 - Epoch: [151][ 700/ 1207] Overall Loss 0.201245 Objective Loss 0.201245 LR 0.000250 Time 0.020195 -2023-02-13 18:24:52,523 - Epoch: [151][ 710/ 1207] Overall Loss 0.201421 Objective Loss 0.201421 LR 0.000250 Time 0.020177 -2023-02-13 18:24:52,714 - Epoch: [151][ 720/ 1207] Overall Loss 0.201436 Objective Loss 0.201436 LR 0.000250 Time 0.020161 -2023-02-13 18:24:52,904 - Epoch: [151][ 730/ 1207] Overall Loss 0.201490 Objective Loss 0.201490 LR 0.000250 Time 0.020145 -2023-02-13 18:24:53,095 - Epoch: [151][ 740/ 1207] Overall Loss 0.201625 Objective Loss 0.201625 LR 0.000250 Time 0.020130 -2023-02-13 18:24:53,285 - Epoch: [151][ 750/ 1207] Overall Loss 0.201788 Objective Loss 0.201788 LR 0.000250 Time 0.020114 -2023-02-13 18:24:53,474 - Epoch: [151][ 760/ 1207] Overall Loss 0.202289 Objective Loss 0.202289 LR 0.000250 Time 0.020099 -2023-02-13 18:24:53,663 - Epoch: [151][ 770/ 1207] Overall Loss 0.202057 Objective Loss 0.202057 LR 0.000250 Time 0.020082 -2023-02-13 18:24:53,851 - Epoch: [151][ 780/ 1207] Overall Loss 0.202257 Objective Loss 0.202257 LR 0.000250 Time 0.020065 -2023-02-13 18:24:54,040 - Epoch: [151][ 790/ 1207] Overall Loss 0.202242 Objective Loss 0.202242 LR 0.000250 Time 0.020050 -2023-02-13 18:24:54,228 - Epoch: [151][ 800/ 1207] Overall Loss 0.202397 Objective Loss 0.202397 LR 0.000250 Time 0.020035 -2023-02-13 18:24:54,417 - Epoch: [151][ 810/ 1207] Overall Loss 0.202134 Objective Loss 0.202134 LR 0.000250 Time 0.020020 -2023-02-13 18:24:54,606 - Epoch: [151][ 820/ 1207] Overall Loss 0.202053 Objective Loss 0.202053 LR 0.000250 Time 0.020005 -2023-02-13 18:24:54,794 - Epoch: [151][ 830/ 1207] Overall Loss 0.202016 Objective Loss 0.202016 LR 0.000250 Time 0.019991 -2023-02-13 18:24:54,982 - Epoch: [151][ 840/ 1207] Overall Loss 0.202064 Objective Loss 0.202064 LR 0.000250 Time 0.019976 -2023-02-13 18:24:55,172 - Epoch: [151][ 850/ 1207] Overall Loss 0.201930 Objective Loss 0.201930 LR 0.000250 Time 0.019964 -2023-02-13 18:24:55,361 - Epoch: [151][ 860/ 1207] Overall Loss 0.201911 Objective Loss 0.201911 LR 0.000250 Time 0.019951 -2023-02-13 18:24:55,549 - Epoch: [151][ 870/ 1207] Overall Loss 0.202027 Objective Loss 0.202027 LR 0.000250 Time 0.019938 -2023-02-13 18:24:55,738 - Epoch: [151][ 880/ 1207] Overall Loss 0.201674 Objective Loss 0.201674 LR 0.000250 Time 0.019925 -2023-02-13 18:24:55,927 - Epoch: [151][ 890/ 1207] Overall Loss 0.201613 Objective Loss 0.201613 LR 0.000250 Time 0.019914 -2023-02-13 18:24:56,116 - Epoch: [151][ 900/ 1207] Overall Loss 0.201582 Objective Loss 0.201582 LR 0.000250 Time 0.019902 -2023-02-13 18:24:56,305 - Epoch: [151][ 910/ 1207] Overall Loss 0.201655 Objective Loss 0.201655 LR 0.000250 Time 0.019890 -2023-02-13 18:24:56,493 - Epoch: [151][ 920/ 1207] Overall Loss 0.201663 Objective Loss 0.201663 LR 0.000250 Time 0.019879 -2023-02-13 18:24:56,682 - Epoch: [151][ 930/ 1207] Overall Loss 0.201441 Objective Loss 0.201441 LR 0.000250 Time 0.019867 -2023-02-13 18:24:56,870 - Epoch: [151][ 940/ 1207] Overall Loss 0.201128 Objective Loss 0.201128 LR 0.000250 Time 0.019856 -2023-02-13 18:24:57,059 - Epoch: [151][ 950/ 1207] Overall Loss 0.201262 Objective Loss 0.201262 LR 0.000250 Time 0.019845 -2023-02-13 18:24:57,248 - Epoch: [151][ 960/ 1207] Overall Loss 0.201368 Objective Loss 0.201368 LR 0.000250 Time 0.019835 -2023-02-13 18:24:57,437 - Epoch: [151][ 970/ 1207] Overall Loss 0.201433 Objective Loss 0.201433 LR 0.000250 Time 0.019825 -2023-02-13 18:24:57,626 - Epoch: [151][ 980/ 1207] Overall Loss 0.201407 Objective Loss 0.201407 LR 0.000250 Time 0.019814 -2023-02-13 18:24:57,814 - Epoch: [151][ 990/ 1207] Overall Loss 0.201636 Objective Loss 0.201636 LR 0.000250 Time 0.019804 -2023-02-13 18:24:58,002 - Epoch: [151][ 1000/ 1207] Overall Loss 0.201650 Objective Loss 0.201650 LR 0.000250 Time 0.019794 -2023-02-13 18:24:58,192 - Epoch: [151][ 1010/ 1207] Overall Loss 0.201724 Objective Loss 0.201724 LR 0.000250 Time 0.019786 -2023-02-13 18:24:58,381 - Epoch: [151][ 1020/ 1207] Overall Loss 0.201731 Objective Loss 0.201731 LR 0.000250 Time 0.019776 -2023-02-13 18:24:58,569 - Epoch: [151][ 1030/ 1207] Overall Loss 0.201778 Objective Loss 0.201778 LR 0.000250 Time 0.019767 -2023-02-13 18:24:58,758 - Epoch: [151][ 1040/ 1207] Overall Loss 0.201843 Objective Loss 0.201843 LR 0.000250 Time 0.019758 -2023-02-13 18:24:58,947 - Epoch: [151][ 1050/ 1207] Overall Loss 0.201821 Objective Loss 0.201821 LR 0.000250 Time 0.019750 -2023-02-13 18:24:59,135 - Epoch: [151][ 1060/ 1207] Overall Loss 0.201643 Objective Loss 0.201643 LR 0.000250 Time 0.019741 -2023-02-13 18:24:59,324 - Epoch: [151][ 1070/ 1207] Overall Loss 0.201765 Objective Loss 0.201765 LR 0.000250 Time 0.019732 -2023-02-13 18:24:59,513 - Epoch: [151][ 1080/ 1207] Overall Loss 0.201976 Objective Loss 0.201976 LR 0.000250 Time 0.019724 -2023-02-13 18:24:59,702 - Epoch: [151][ 1090/ 1207] Overall Loss 0.201804 Objective Loss 0.201804 LR 0.000250 Time 0.019716 -2023-02-13 18:24:59,890 - Epoch: [151][ 1100/ 1207] Overall Loss 0.202022 Objective Loss 0.202022 LR 0.000250 Time 0.019707 -2023-02-13 18:25:00,078 - Epoch: [151][ 1110/ 1207] Overall Loss 0.202069 Objective Loss 0.202069 LR 0.000250 Time 0.019699 -2023-02-13 18:25:00,268 - Epoch: [151][ 1120/ 1207] Overall Loss 0.202312 Objective Loss 0.202312 LR 0.000250 Time 0.019692 -2023-02-13 18:25:00,456 - Epoch: [151][ 1130/ 1207] Overall Loss 0.202151 Objective Loss 0.202151 LR 0.000250 Time 0.019684 -2023-02-13 18:25:00,645 - Epoch: [151][ 1140/ 1207] Overall Loss 0.202251 Objective Loss 0.202251 LR 0.000250 Time 0.019677 -2023-02-13 18:25:00,833 - Epoch: [151][ 1150/ 1207] Overall Loss 0.202377 Objective Loss 0.202377 LR 0.000250 Time 0.019669 -2023-02-13 18:25:01,023 - Epoch: [151][ 1160/ 1207] Overall Loss 0.202332 Objective Loss 0.202332 LR 0.000250 Time 0.019663 -2023-02-13 18:25:01,212 - Epoch: [151][ 1170/ 1207] Overall Loss 0.202139 Objective Loss 0.202139 LR 0.000250 Time 0.019656 -2023-02-13 18:25:01,401 - Epoch: [151][ 1180/ 1207] Overall Loss 0.202252 Objective Loss 0.202252 LR 0.000250 Time 0.019649 -2023-02-13 18:25:01,589 - Epoch: [151][ 1190/ 1207] Overall Loss 0.202224 Objective Loss 0.202224 LR 0.000250 Time 0.019642 -2023-02-13 18:25:01,834 - Epoch: [151][ 1200/ 1207] Overall Loss 0.202181 Objective Loss 0.202181 LR 0.000250 Time 0.019683 -2023-02-13 18:25:01,951 - Epoch: [151][ 1207/ 1207] Overall Loss 0.202163 Objective Loss 0.202163 Top1 86.280488 Top5 98.170732 LR 0.000250 Time 0.019664 -2023-02-13 18:25:02,022 - --- validate (epoch=151)----------- -2023-02-13 18:25:02,023 - 34311 samples (256 per mini-batch) -2023-02-13 18:25:02,434 - Epoch: [151][ 10/ 135] Loss 0.281716 Top1 85.781250 Top5 98.593750 -2023-02-13 18:25:02,578 - Epoch: [151][ 20/ 135] Loss 0.280270 Top1 86.132812 Top5 98.222656 -2023-02-13 18:25:02,709 - Epoch: [151][ 30/ 135] Loss 0.289963 Top1 86.145833 Top5 98.072917 -2023-02-13 18:25:02,836 - Epoch: [151][ 40/ 135] Loss 0.286908 Top1 86.005859 Top5 98.125000 -2023-02-13 18:25:02,961 - Epoch: [151][ 50/ 135] Loss 0.283134 Top1 86.007812 Top5 98.132812 -2023-02-13 18:25:03,085 - Epoch: [151][ 60/ 135] Loss 0.289477 Top1 85.950521 Top5 98.138021 -2023-02-13 18:25:03,222 - Epoch: [151][ 70/ 135] Loss 0.286808 Top1 85.920759 Top5 98.125000 -2023-02-13 18:25:03,363 - Epoch: [151][ 80/ 135] Loss 0.288122 Top1 85.751953 Top5 98.090820 -2023-02-13 18:25:03,495 - Epoch: [151][ 90/ 135] Loss 0.291413 Top1 85.776910 Top5 98.046875 -2023-02-13 18:25:03,627 - Epoch: [151][ 100/ 135] Loss 0.291119 Top1 85.746094 Top5 98.070312 -2023-02-13 18:25:03,759 - Epoch: [151][ 110/ 135] Loss 0.294568 Top1 85.632102 Top5 98.057528 -2023-02-13 18:25:03,892 - Epoch: [151][ 120/ 135] Loss 0.294592 Top1 85.598958 Top5 98.050130 -2023-02-13 18:25:04,023 - Epoch: [151][ 130/ 135] Loss 0.292119 Top1 85.652043 Top5 98.037861 -2023-02-13 18:25:04,067 - Epoch: [151][ 135/ 135] Loss 0.314068 Top1 85.634345 Top5 98.015214 -2023-02-13 18:25:04,135 - ==> Top1: 85.634 Top5: 98.015 Loss: 0.314 - -2023-02-13 18:25:04,136 - ==> Confusion: -[[ 870 5 6 0 10 2 0 1 2 35 2 4 2 4 6 4 2 3 2 1 6] - [ 3 964 2 1 5 16 2 15 2 2 1 0 1 0 0 3 6 0 2 1 7] - [ 9 4 965 10 2 1 17 12 0 1 1 2 1 6 2 4 3 2 4 4 8] - [ 3 1 16 901 4 6 0 0 1 2 17 0 6 2 18 4 4 3 19 0 9] - [ 14 8 0 0 989 13 1 1 0 1 0 6 2 2 4 8 6 1 0 5 5] - [ 1 20 0 4 5 973 3 14 2 5 1 9 1 15 0 1 6 0 0 5 5] - [ 3 3 14 2 1 5 1042 2 0 2 3 1 1 1 0 3 2 1 3 5 5] - [ 0 14 8 0 2 25 6 935 0 2 1 4 2 0 0 0 1 0 9 9 6] - [ 17 5 1 1 0 0 0 1 905 40 9 3 0 8 10 2 2 1 4 0 0] - [ 88 1 4 0 11 2 0 2 24 852 0 2 1 13 4 1 1 2 1 0 3] - [ 2 2 2 5 1 2 2 5 16 1 986 2 1 7 2 0 1 1 7 0 6] - [ 0 2 2 0 5 11 1 8 1 0 1 911 21 8 0 4 6 7 3 11 3] - [ 0 1 2 6 3 5 0 1 1 2 0 24 878 1 1 6 2 11 6 1 8] - [ 3 3 3 0 8 7 1 2 5 17 6 6 3 941 1 3 5 2 0 3 5] - [ 13 2 2 14 3 4 0 2 19 9 3 0 3 1 986 1 4 6 6 2 12] - [ 1 2 6 0 7 2 4 2 0 0 0 3 5 3 2 979 7 10 0 7 6] - [ 2 4 0 1 7 1 0 0 1 0 0 1 2 1 1 11 1012 2 1 2 12] - [ 4 3 0 3 1 2 2 0 0 0 1 9 14 1 0 22 0 982 0 3 4] - [ 2 5 3 5 0 2 1 20 3 2 7 1 3 0 10 0 1 1 1014 2 4] - [ 1 4 1 0 1 6 7 7 1 0 1 10 4 3 0 4 3 4 0 1083 8] - [ 130 250 212 99 113 220 86 136 87 89 186 104 259 267 123 95 260 77 147 280 10214]] - -2023-02-13 18:25:04,137 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:25:04,138 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:25:04,143 - - -2023-02-13 18:25:04,143 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:25:05,035 - Epoch: [152][ 10/ 1207] Overall Loss 0.192062 Objective Loss 0.192062 LR 0.000250 Time 0.089131 -2023-02-13 18:25:05,233 - Epoch: [152][ 20/ 1207] Overall Loss 0.189596 Objective Loss 0.189596 LR 0.000250 Time 0.054417 -2023-02-13 18:25:05,427 - Epoch: [152][ 30/ 1207] Overall Loss 0.196688 Objective Loss 0.196688 LR 0.000250 Time 0.042735 -2023-02-13 18:25:05,621 - Epoch: [152][ 40/ 1207] Overall Loss 0.200386 Objective Loss 0.200386 LR 0.000250 Time 0.036903 -2023-02-13 18:25:05,814 - Epoch: [152][ 50/ 1207] Overall Loss 0.205763 Objective Loss 0.205763 LR 0.000250 Time 0.033379 -2023-02-13 18:25:06,009 - Epoch: [152][ 60/ 1207] Overall Loss 0.203648 Objective Loss 0.203648 LR 0.000250 Time 0.031059 -2023-02-13 18:25:06,202 - Epoch: [152][ 70/ 1207] Overall Loss 0.204776 Objective Loss 0.204776 LR 0.000250 Time 0.029375 -2023-02-13 18:25:06,397 - Epoch: [152][ 80/ 1207] Overall Loss 0.202982 Objective Loss 0.202982 LR 0.000250 Time 0.028126 -2023-02-13 18:25:06,589 - Epoch: [152][ 90/ 1207] Overall Loss 0.202305 Objective Loss 0.202305 LR 0.000250 Time 0.027137 -2023-02-13 18:25:06,783 - Epoch: [152][ 100/ 1207] Overall Loss 0.200260 Objective Loss 0.200260 LR 0.000250 Time 0.026358 -2023-02-13 18:25:06,977 - Epoch: [152][ 110/ 1207] Overall Loss 0.198186 Objective Loss 0.198186 LR 0.000250 Time 0.025723 -2023-02-13 18:25:07,171 - Epoch: [152][ 120/ 1207] Overall Loss 0.198254 Objective Loss 0.198254 LR 0.000250 Time 0.025194 -2023-02-13 18:25:07,364 - Epoch: [152][ 130/ 1207] Overall Loss 0.197219 Objective Loss 0.197219 LR 0.000250 Time 0.024738 -2023-02-13 18:25:07,558 - Epoch: [152][ 140/ 1207] Overall Loss 0.197975 Objective Loss 0.197975 LR 0.000250 Time 0.024355 -2023-02-13 18:25:07,751 - Epoch: [152][ 150/ 1207] Overall Loss 0.198565 Objective Loss 0.198565 LR 0.000250 Time 0.024015 -2023-02-13 18:25:07,946 - Epoch: [152][ 160/ 1207] Overall Loss 0.199563 Objective Loss 0.199563 LR 0.000250 Time 0.023728 -2023-02-13 18:25:08,139 - Epoch: [152][ 170/ 1207] Overall Loss 0.200448 Objective Loss 0.200448 LR 0.000250 Time 0.023465 -2023-02-13 18:25:08,334 - Epoch: [152][ 180/ 1207] Overall Loss 0.199214 Objective Loss 0.199214 LR 0.000250 Time 0.023243 -2023-02-13 18:25:08,527 - Epoch: [152][ 190/ 1207] Overall Loss 0.199100 Objective Loss 0.199100 LR 0.000250 Time 0.023031 -2023-02-13 18:25:08,721 - Epoch: [152][ 200/ 1207] Overall Loss 0.198168 Objective Loss 0.198168 LR 0.000250 Time 0.022848 -2023-02-13 18:25:08,914 - Epoch: [152][ 210/ 1207] Overall Loss 0.198749 Objective Loss 0.198749 LR 0.000250 Time 0.022679 -2023-02-13 18:25:09,108 - Epoch: [152][ 220/ 1207] Overall Loss 0.199282 Objective Loss 0.199282 LR 0.000250 Time 0.022526 -2023-02-13 18:25:09,301 - Epoch: [152][ 230/ 1207] Overall Loss 0.198235 Objective Loss 0.198235 LR 0.000250 Time 0.022387 -2023-02-13 18:25:09,495 - Epoch: [152][ 240/ 1207] Overall Loss 0.197562 Objective Loss 0.197562 LR 0.000250 Time 0.022262 -2023-02-13 18:25:09,688 - Epoch: [152][ 250/ 1207] Overall Loss 0.198087 Objective Loss 0.198087 LR 0.000250 Time 0.022142 -2023-02-13 18:25:09,882 - Epoch: [152][ 260/ 1207] Overall Loss 0.197585 Objective Loss 0.197585 LR 0.000250 Time 0.022034 -2023-02-13 18:25:10,074 - Epoch: [152][ 270/ 1207] Overall Loss 0.197969 Objective Loss 0.197969 LR 0.000250 Time 0.021929 -2023-02-13 18:25:10,269 - Epoch: [152][ 280/ 1207] Overall Loss 0.197547 Objective Loss 0.197547 LR 0.000250 Time 0.021840 -2023-02-13 18:25:10,462 - Epoch: [152][ 290/ 1207] Overall Loss 0.197904 Objective Loss 0.197904 LR 0.000250 Time 0.021750 -2023-02-13 18:25:10,655 - Epoch: [152][ 300/ 1207] Overall Loss 0.197782 Objective Loss 0.197782 LR 0.000250 Time 0.021668 -2023-02-13 18:25:10,848 - Epoch: [152][ 310/ 1207] Overall Loss 0.198480 Objective Loss 0.198480 LR 0.000250 Time 0.021589 -2023-02-13 18:25:11,044 - Epoch: [152][ 320/ 1207] Overall Loss 0.198849 Objective Loss 0.198849 LR 0.000250 Time 0.021525 -2023-02-13 18:25:11,237 - Epoch: [152][ 330/ 1207] Overall Loss 0.198588 Objective Loss 0.198588 LR 0.000250 Time 0.021459 -2023-02-13 18:25:11,431 - Epoch: [152][ 340/ 1207] Overall Loss 0.198863 Objective Loss 0.198863 LR 0.000250 Time 0.021398 -2023-02-13 18:25:11,624 - Epoch: [152][ 350/ 1207] Overall Loss 0.199314 Objective Loss 0.199314 LR 0.000250 Time 0.021336 -2023-02-13 18:25:11,818 - Epoch: [152][ 360/ 1207] Overall Loss 0.199507 Objective Loss 0.199507 LR 0.000250 Time 0.021281 -2023-02-13 18:25:12,011 - Epoch: [152][ 370/ 1207] Overall Loss 0.199316 Objective Loss 0.199316 LR 0.000250 Time 0.021226 -2023-02-13 18:25:12,206 - Epoch: [152][ 380/ 1207] Overall Loss 0.199344 Objective Loss 0.199344 LR 0.000250 Time 0.021180 -2023-02-13 18:25:12,398 - Epoch: [152][ 390/ 1207] Overall Loss 0.198924 Objective Loss 0.198924 LR 0.000250 Time 0.021129 -2023-02-13 18:25:12,593 - Epoch: [152][ 400/ 1207] Overall Loss 0.198320 Objective Loss 0.198320 LR 0.000250 Time 0.021086 -2023-02-13 18:25:12,785 - Epoch: [152][ 410/ 1207] Overall Loss 0.198299 Objective Loss 0.198299 LR 0.000250 Time 0.021039 -2023-02-13 18:25:12,979 - Epoch: [152][ 420/ 1207] Overall Loss 0.197982 Objective Loss 0.197982 LR 0.000250 Time 0.020999 -2023-02-13 18:25:13,172 - Epoch: [152][ 430/ 1207] Overall Loss 0.198124 Objective Loss 0.198124 LR 0.000250 Time 0.020959 -2023-02-13 18:25:13,366 - Epoch: [152][ 440/ 1207] Overall Loss 0.198596 Objective Loss 0.198596 LR 0.000250 Time 0.020924 -2023-02-13 18:25:13,560 - Epoch: [152][ 450/ 1207] Overall Loss 0.198450 Objective Loss 0.198450 LR 0.000250 Time 0.020888 -2023-02-13 18:25:13,754 - Epoch: [152][ 460/ 1207] Overall Loss 0.198616 Objective Loss 0.198616 LR 0.000250 Time 0.020856 -2023-02-13 18:25:13,947 - Epoch: [152][ 470/ 1207] Overall Loss 0.198653 Objective Loss 0.198653 LR 0.000250 Time 0.020822 -2023-02-13 18:25:14,142 - Epoch: [152][ 480/ 1207] Overall Loss 0.198680 Objective Loss 0.198680 LR 0.000250 Time 0.020793 -2023-02-13 18:25:14,336 - Epoch: [152][ 490/ 1207] Overall Loss 0.198201 Objective Loss 0.198201 LR 0.000250 Time 0.020763 -2023-02-13 18:25:14,530 - Epoch: [152][ 500/ 1207] Overall Loss 0.199342 Objective Loss 0.199342 LR 0.000250 Time 0.020736 -2023-02-13 18:25:14,723 - Epoch: [152][ 510/ 1207] Overall Loss 0.199614 Objective Loss 0.199614 LR 0.000250 Time 0.020707 -2023-02-13 18:25:14,918 - Epoch: [152][ 520/ 1207] Overall Loss 0.199543 Objective Loss 0.199543 LR 0.000250 Time 0.020682 -2023-02-13 18:25:15,111 - Epoch: [152][ 530/ 1207] Overall Loss 0.199734 Objective Loss 0.199734 LR 0.000250 Time 0.020656 -2023-02-13 18:25:15,306 - Epoch: [152][ 540/ 1207] Overall Loss 0.200317 Objective Loss 0.200317 LR 0.000250 Time 0.020634 -2023-02-13 18:25:15,499 - Epoch: [152][ 550/ 1207] Overall Loss 0.200698 Objective Loss 0.200698 LR 0.000250 Time 0.020609 -2023-02-13 18:25:15,693 - Epoch: [152][ 560/ 1207] Overall Loss 0.200327 Objective Loss 0.200327 LR 0.000250 Time 0.020587 -2023-02-13 18:25:15,886 - Epoch: [152][ 570/ 1207] Overall Loss 0.199907 Objective Loss 0.199907 LR 0.000250 Time 0.020564 -2023-02-13 18:25:16,080 - Epoch: [152][ 580/ 1207] Overall Loss 0.199769 Objective Loss 0.199769 LR 0.000250 Time 0.020544 -2023-02-13 18:25:16,274 - Epoch: [152][ 590/ 1207] Overall Loss 0.199901 Objective Loss 0.199901 LR 0.000250 Time 0.020523 -2023-02-13 18:25:16,467 - Epoch: [152][ 600/ 1207] Overall Loss 0.200148 Objective Loss 0.200148 LR 0.000250 Time 0.020503 -2023-02-13 18:25:16,660 - Epoch: [152][ 610/ 1207] Overall Loss 0.199952 Objective Loss 0.199952 LR 0.000250 Time 0.020482 -2023-02-13 18:25:16,855 - Epoch: [152][ 620/ 1207] Overall Loss 0.200367 Objective Loss 0.200367 LR 0.000250 Time 0.020465 -2023-02-13 18:25:17,048 - Epoch: [152][ 630/ 1207] Overall Loss 0.200233 Objective Loss 0.200233 LR 0.000250 Time 0.020446 -2023-02-13 18:25:17,243 - Epoch: [152][ 640/ 1207] Overall Loss 0.200316 Objective Loss 0.200316 LR 0.000250 Time 0.020431 -2023-02-13 18:25:17,436 - Epoch: [152][ 650/ 1207] Overall Loss 0.200052 Objective Loss 0.200052 LR 0.000250 Time 0.020412 -2023-02-13 18:25:17,630 - Epoch: [152][ 660/ 1207] Overall Loss 0.200481 Objective Loss 0.200481 LR 0.000250 Time 0.020396 -2023-02-13 18:25:17,822 - Epoch: [152][ 670/ 1207] Overall Loss 0.200546 Objective Loss 0.200546 LR 0.000250 Time 0.020379 -2023-02-13 18:25:18,016 - Epoch: [152][ 680/ 1207] Overall Loss 0.200843 Objective Loss 0.200843 LR 0.000250 Time 0.020364 -2023-02-13 18:25:18,209 - Epoch: [152][ 690/ 1207] Overall Loss 0.201000 Objective Loss 0.201000 LR 0.000250 Time 0.020348 -2023-02-13 18:25:18,404 - Epoch: [152][ 700/ 1207] Overall Loss 0.200950 Objective Loss 0.200950 LR 0.000250 Time 0.020335 -2023-02-13 18:25:18,600 - Epoch: [152][ 710/ 1207] Overall Loss 0.200777 Objective Loss 0.200777 LR 0.000250 Time 0.020324 -2023-02-13 18:25:18,796 - Epoch: [152][ 720/ 1207] Overall Loss 0.200684 Objective Loss 0.200684 LR 0.000250 Time 0.020313 -2023-02-13 18:25:18,991 - Epoch: [152][ 730/ 1207] Overall Loss 0.201019 Objective Loss 0.201019 LR 0.000250 Time 0.020302 -2023-02-13 18:25:19,189 - Epoch: [152][ 740/ 1207] Overall Loss 0.201223 Objective Loss 0.201223 LR 0.000250 Time 0.020294 -2023-02-13 18:25:19,384 - Epoch: [152][ 750/ 1207] Overall Loss 0.201089 Objective Loss 0.201089 LR 0.000250 Time 0.020284 -2023-02-13 18:25:19,581 - Epoch: [152][ 760/ 1207] Overall Loss 0.200832 Objective Loss 0.200832 LR 0.000250 Time 0.020275 -2023-02-13 18:25:19,777 - Epoch: [152][ 770/ 1207] Overall Loss 0.200902 Objective Loss 0.200902 LR 0.000250 Time 0.020266 -2023-02-13 18:25:19,973 - Epoch: [152][ 780/ 1207] Overall Loss 0.200907 Objective Loss 0.200907 LR 0.000250 Time 0.020257 -2023-02-13 18:25:20,169 - Epoch: [152][ 790/ 1207] Overall Loss 0.201114 Objective Loss 0.201114 LR 0.000250 Time 0.020249 -2023-02-13 18:25:20,366 - Epoch: [152][ 800/ 1207] Overall Loss 0.201225 Objective Loss 0.201225 LR 0.000250 Time 0.020241 -2023-02-13 18:25:20,562 - Epoch: [152][ 810/ 1207] Overall Loss 0.201758 Objective Loss 0.201758 LR 0.000250 Time 0.020232 -2023-02-13 18:25:20,758 - Epoch: [152][ 820/ 1207] Overall Loss 0.201715 Objective Loss 0.201715 LR 0.000250 Time 0.020225 -2023-02-13 18:25:20,956 - Epoch: [152][ 830/ 1207] Overall Loss 0.201888 Objective Loss 0.201888 LR 0.000250 Time 0.020219 -2023-02-13 18:25:21,152 - Epoch: [152][ 840/ 1207] Overall Loss 0.201725 Objective Loss 0.201725 LR 0.000250 Time 0.020211 -2023-02-13 18:25:21,349 - Epoch: [152][ 850/ 1207] Overall Loss 0.201609 Objective Loss 0.201609 LR 0.000250 Time 0.020205 -2023-02-13 18:25:21,546 - Epoch: [152][ 860/ 1207] Overall Loss 0.201393 Objective Loss 0.201393 LR 0.000250 Time 0.020198 -2023-02-13 18:25:21,742 - Epoch: [152][ 870/ 1207] Overall Loss 0.201389 Objective Loss 0.201389 LR 0.000250 Time 0.020191 -2023-02-13 18:25:21,939 - Epoch: [152][ 880/ 1207] Overall Loss 0.201403 Objective Loss 0.201403 LR 0.000250 Time 0.020185 -2023-02-13 18:25:22,135 - Epoch: [152][ 890/ 1207] Overall Loss 0.201436 Objective Loss 0.201436 LR 0.000250 Time 0.020178 -2023-02-13 18:25:22,333 - Epoch: [152][ 900/ 1207] Overall Loss 0.201579 Objective Loss 0.201579 LR 0.000250 Time 0.020173 -2023-02-13 18:25:22,528 - Epoch: [152][ 910/ 1207] Overall Loss 0.201389 Objective Loss 0.201389 LR 0.000250 Time 0.020166 -2023-02-13 18:25:22,725 - Epoch: [152][ 920/ 1207] Overall Loss 0.201237 Objective Loss 0.201237 LR 0.000250 Time 0.020160 -2023-02-13 18:25:22,919 - Epoch: [152][ 930/ 1207] Overall Loss 0.201105 Objective Loss 0.201105 LR 0.000250 Time 0.020152 -2023-02-13 18:25:23,110 - Epoch: [152][ 940/ 1207] Overall Loss 0.201200 Objective Loss 0.201200 LR 0.000250 Time 0.020140 -2023-02-13 18:25:23,302 - Epoch: [152][ 950/ 1207] Overall Loss 0.201031 Objective Loss 0.201031 LR 0.000250 Time 0.020130 -2023-02-13 18:25:23,492 - Epoch: [152][ 960/ 1207] Overall Loss 0.201103 Objective Loss 0.201103 LR 0.000250 Time 0.020118 -2023-02-13 18:25:23,684 - Epoch: [152][ 970/ 1207] Overall Loss 0.200859 Objective Loss 0.200859 LR 0.000250 Time 0.020108 -2023-02-13 18:25:23,874 - Epoch: [152][ 980/ 1207] Overall Loss 0.200717 Objective Loss 0.200717 LR 0.000250 Time 0.020096 -2023-02-13 18:25:24,065 - Epoch: [152][ 990/ 1207] Overall Loss 0.200697 Objective Loss 0.200697 LR 0.000250 Time 0.020085 -2023-02-13 18:25:24,254 - Epoch: [152][ 1000/ 1207] Overall Loss 0.200744 Objective Loss 0.200744 LR 0.000250 Time 0.020073 -2023-02-13 18:25:24,445 - Epoch: [152][ 1010/ 1207] Overall Loss 0.200942 Objective Loss 0.200942 LR 0.000250 Time 0.020063 -2023-02-13 18:25:24,636 - Epoch: [152][ 1020/ 1207] Overall Loss 0.201051 Objective Loss 0.201051 LR 0.000250 Time 0.020053 -2023-02-13 18:25:24,826 - Epoch: [152][ 1030/ 1207] Overall Loss 0.200788 Objective Loss 0.200788 LR 0.000250 Time 0.020043 -2023-02-13 18:25:25,016 - Epoch: [152][ 1040/ 1207] Overall Loss 0.200519 Objective Loss 0.200519 LR 0.000250 Time 0.020032 -2023-02-13 18:25:25,208 - Epoch: [152][ 1050/ 1207] Overall Loss 0.200444 Objective Loss 0.200444 LR 0.000250 Time 0.020024 -2023-02-13 18:25:25,398 - Epoch: [152][ 1060/ 1207] Overall Loss 0.200617 Objective Loss 0.200617 LR 0.000250 Time 0.020014 -2023-02-13 18:25:25,588 - Epoch: [152][ 1070/ 1207] Overall Loss 0.200613 Objective Loss 0.200613 LR 0.000250 Time 0.020005 -2023-02-13 18:25:25,777 - Epoch: [152][ 1080/ 1207] Overall Loss 0.200604 Objective Loss 0.200604 LR 0.000250 Time 0.019994 -2023-02-13 18:25:25,969 - Epoch: [152][ 1090/ 1207] Overall Loss 0.200624 Objective Loss 0.200624 LR 0.000250 Time 0.019986 -2023-02-13 18:25:26,158 - Epoch: [152][ 1100/ 1207] Overall Loss 0.200509 Objective Loss 0.200509 LR 0.000250 Time 0.019976 -2023-02-13 18:25:26,350 - Epoch: [152][ 1110/ 1207] Overall Loss 0.200252 Objective Loss 0.200252 LR 0.000250 Time 0.019969 -2023-02-13 18:25:26,541 - Epoch: [152][ 1120/ 1207] Overall Loss 0.200228 Objective Loss 0.200228 LR 0.000250 Time 0.019960 -2023-02-13 18:25:26,732 - Epoch: [152][ 1130/ 1207] Overall Loss 0.200178 Objective Loss 0.200178 LR 0.000250 Time 0.019953 -2023-02-13 18:25:26,922 - Epoch: [152][ 1140/ 1207] Overall Loss 0.200115 Objective Loss 0.200115 LR 0.000250 Time 0.019944 -2023-02-13 18:25:27,113 - Epoch: [152][ 1150/ 1207] Overall Loss 0.200146 Objective Loss 0.200146 LR 0.000250 Time 0.019936 -2023-02-13 18:25:27,304 - Epoch: [152][ 1160/ 1207] Overall Loss 0.200050 Objective Loss 0.200050 LR 0.000250 Time 0.019929 -2023-02-13 18:25:27,495 - Epoch: [152][ 1170/ 1207] Overall Loss 0.200165 Objective Loss 0.200165 LR 0.000250 Time 0.019921 -2023-02-13 18:25:27,684 - Epoch: [152][ 1180/ 1207] Overall Loss 0.200321 Objective Loss 0.200321 LR 0.000250 Time 0.019912 -2023-02-13 18:25:27,873 - Epoch: [152][ 1190/ 1207] Overall Loss 0.200374 Objective Loss 0.200374 LR 0.000250 Time 0.019904 -2023-02-13 18:25:28,112 - Epoch: [152][ 1200/ 1207] Overall Loss 0.200341 Objective Loss 0.200341 LR 0.000250 Time 0.019937 -2023-02-13 18:25:28,227 - Epoch: [152][ 1207/ 1207] Overall Loss 0.200368 Objective Loss 0.200368 Top1 86.585366 Top5 99.390244 LR 0.000250 Time 0.019916 -2023-02-13 18:25:28,298 - --- validate (epoch=152)----------- -2023-02-13 18:25:28,299 - 34311 samples (256 per mini-batch) -2023-02-13 18:25:28,698 - Epoch: [152][ 10/ 135] Loss 0.275220 Top1 84.804688 Top5 97.929688 -2023-02-13 18:25:28,826 - Epoch: [152][ 20/ 135] Loss 0.278345 Top1 84.863281 Top5 97.929688 -2023-02-13 18:25:28,967 - Epoch: [152][ 30/ 135] Loss 0.287385 Top1 85.104167 Top5 98.059896 -2023-02-13 18:25:29,095 - Epoch: [152][ 40/ 135] Loss 0.289339 Top1 85.175781 Top5 98.056641 -2023-02-13 18:25:29,224 - Epoch: [152][ 50/ 135] Loss 0.292166 Top1 85.156250 Top5 98.101562 -2023-02-13 18:25:29,354 - Epoch: [152][ 60/ 135] Loss 0.292855 Top1 85.143229 Top5 98.066406 -2023-02-13 18:25:29,483 - Epoch: [152][ 70/ 135] Loss 0.291477 Top1 85.156250 Top5 98.102679 -2023-02-13 18:25:29,616 - Epoch: [152][ 80/ 135] Loss 0.293516 Top1 85.190430 Top5 98.100586 -2023-02-13 18:25:29,745 - Epoch: [152][ 90/ 135] Loss 0.292301 Top1 85.151910 Top5 98.077257 -2023-02-13 18:25:29,879 - Epoch: [152][ 100/ 135] Loss 0.294406 Top1 85.226562 Top5 98.031250 -2023-02-13 18:25:30,009 - Epoch: [152][ 110/ 135] Loss 0.296958 Top1 85.177557 Top5 98.011364 -2023-02-13 18:25:30,141 - Epoch: [152][ 120/ 135] Loss 0.294998 Top1 85.332031 Top5 98.020833 -2023-02-13 18:25:30,276 - Epoch: [152][ 130/ 135] Loss 0.294482 Top1 85.360577 Top5 98.013822 -2023-02-13 18:25:30,323 - Epoch: [152][ 135/ 135] Loss 0.293544 Top1 85.392440 Top5 97.997727 -2023-02-13 18:25:30,396 - ==> Top1: 85.392 Top5: 97.998 Loss: 0.294 - -2023-02-13 18:25:30,397 - ==> Confusion: -[[ 870 3 3 1 8 3 0 0 5 40 1 4 3 4 8 2 1 2 3 0 6] - [ 2 956 1 2 6 16 5 11 3 2 2 2 1 0 0 1 4 1 5 4 9] - [ 5 1 965 9 2 2 19 15 0 2 3 0 1 6 4 8 3 0 4 1 8] - [ 5 1 25 897 0 4 0 0 2 0 12 0 10 1 22 3 5 7 15 0 7] - [ 13 9 0 1 992 10 1 1 3 1 0 4 1 2 9 5 5 1 1 2 5] - [ 1 21 0 4 10 950 4 22 1 3 1 11 4 18 0 2 4 1 0 5 8] - [ 3 2 7 0 0 5 1044 4 1 2 5 0 0 3 1 4 2 3 2 6 5] - [ 1 9 8 2 3 23 5 928 0 1 1 5 4 1 0 0 0 1 21 5 6] - [ 15 1 1 1 2 0 0 3 915 33 9 2 0 7 13 1 0 0 4 0 2] - [ 75 1 5 0 8 1 0 3 35 855 0 0 1 13 5 2 1 3 1 0 3] - [ 2 1 0 8 1 2 3 3 17 1 985 1 2 12 4 0 0 1 5 1 2] - [ 3 1 3 0 3 7 0 4 3 0 0 918 21 10 1 5 2 12 2 10 0] - [ 0 0 0 7 1 2 0 2 1 3 0 31 872 1 2 11 3 10 2 1 10] - [ 3 2 2 1 6 6 0 2 9 14 5 4 1 951 3 4 2 3 0 2 4] - [ 7 1 1 11 4 4 0 1 17 5 3 2 4 3 1007 2 1 3 7 1 8] - [ 5 3 3 0 4 2 3 1 0 0 0 6 8 1 0 979 7 12 1 7 4] - [ 1 5 0 1 7 1 0 1 2 0 1 4 2 3 0 13 998 3 0 4 15] - [ 4 2 0 3 1 1 1 0 0 1 1 9 15 1 1 16 1 984 2 1 7] - [ 2 4 4 7 1 1 0 22 2 1 5 1 2 0 13 0 0 3 1013 1 4] - [ 0 3 1 0 1 7 5 8 1 0 1 17 3 8 1 6 3 3 0 1071 9] - [ 155 216 241 108 116 174 90 142 88 91 199 125 265 299 150 116 212 102 182 214 10149]] - -2023-02-13 18:25:30,398 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:25:30,398 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:25:30,404 - - -2023-02-13 18:25:30,404 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:25:31,288 - Epoch: [153][ 10/ 1207] Overall Loss 0.209286 Objective Loss 0.209286 LR 0.000250 Time 0.088345 -2023-02-13 18:25:31,490 - Epoch: [153][ 20/ 1207] Overall Loss 0.215385 Objective Loss 0.215385 LR 0.000250 Time 0.054225 -2023-02-13 18:25:31,685 - Epoch: [153][ 30/ 1207] Overall Loss 0.204320 Objective Loss 0.204320 LR 0.000250 Time 0.042641 -2023-02-13 18:25:31,883 - Epoch: [153][ 40/ 1207] Overall Loss 0.204502 Objective Loss 0.204502 LR 0.000250 Time 0.036916 -2023-02-13 18:25:32,078 - Epoch: [153][ 50/ 1207] Overall Loss 0.201783 Objective Loss 0.201783 LR 0.000250 Time 0.033429 -2023-02-13 18:25:32,276 - Epoch: [153][ 60/ 1207] Overall Loss 0.201619 Objective Loss 0.201619 LR 0.000250 Time 0.031158 -2023-02-13 18:25:32,471 - Epoch: [153][ 70/ 1207] Overall Loss 0.200813 Objective Loss 0.200813 LR 0.000250 Time 0.029479 -2023-02-13 18:25:32,668 - Epoch: [153][ 80/ 1207] Overall Loss 0.199874 Objective Loss 0.199874 LR 0.000250 Time 0.028252 -2023-02-13 18:25:32,862 - Epoch: [153][ 90/ 1207] Overall Loss 0.200910 Objective Loss 0.200910 LR 0.000250 Time 0.027267 -2023-02-13 18:25:33,059 - Epoch: [153][ 100/ 1207] Overall Loss 0.197524 Objective Loss 0.197524 LR 0.000250 Time 0.026507 -2023-02-13 18:25:33,254 - Epoch: [153][ 110/ 1207] Overall Loss 0.196398 Objective Loss 0.196398 LR 0.000250 Time 0.025870 -2023-02-13 18:25:33,452 - Epoch: [153][ 120/ 1207] Overall Loss 0.194386 Objective Loss 0.194386 LR 0.000250 Time 0.025355 -2023-02-13 18:25:33,646 - Epoch: [153][ 130/ 1207] Overall Loss 0.194311 Objective Loss 0.194311 LR 0.000250 Time 0.024897 -2023-02-13 18:25:33,843 - Epoch: [153][ 140/ 1207] Overall Loss 0.194804 Objective Loss 0.194804 LR 0.000250 Time 0.024527 -2023-02-13 18:25:34,038 - Epoch: [153][ 150/ 1207] Overall Loss 0.196489 Objective Loss 0.196489 LR 0.000250 Time 0.024185 -2023-02-13 18:25:34,235 - Epoch: [153][ 160/ 1207] Overall Loss 0.197787 Objective Loss 0.197787 LR 0.000250 Time 0.023903 -2023-02-13 18:25:34,430 - Epoch: [153][ 170/ 1207] Overall Loss 0.197323 Objective Loss 0.197323 LR 0.000250 Time 0.023642 -2023-02-13 18:25:34,627 - Epoch: [153][ 180/ 1207] Overall Loss 0.196300 Objective Loss 0.196300 LR 0.000250 Time 0.023421 -2023-02-13 18:25:34,822 - Epoch: [153][ 190/ 1207] Overall Loss 0.197005 Objective Loss 0.197005 LR 0.000250 Time 0.023212 -2023-02-13 18:25:35,019 - Epoch: [153][ 200/ 1207] Overall Loss 0.198675 Objective Loss 0.198675 LR 0.000250 Time 0.023034 -2023-02-13 18:25:35,214 - Epoch: [153][ 210/ 1207] Overall Loss 0.198820 Objective Loss 0.198820 LR 0.000250 Time 0.022864 -2023-02-13 18:25:35,411 - Epoch: [153][ 220/ 1207] Overall Loss 0.198983 Objective Loss 0.198983 LR 0.000250 Time 0.022720 -2023-02-13 18:25:35,606 - Epoch: [153][ 230/ 1207] Overall Loss 0.199269 Objective Loss 0.199269 LR 0.000250 Time 0.022577 -2023-02-13 18:25:35,803 - Epoch: [153][ 240/ 1207] Overall Loss 0.199202 Objective Loss 0.199202 LR 0.000250 Time 0.022456 -2023-02-13 18:25:35,998 - Epoch: [153][ 250/ 1207] Overall Loss 0.198472 Objective Loss 0.198472 LR 0.000250 Time 0.022337 -2023-02-13 18:25:36,195 - Epoch: [153][ 260/ 1207] Overall Loss 0.198053 Objective Loss 0.198053 LR 0.000250 Time 0.022235 -2023-02-13 18:25:36,390 - Epoch: [153][ 270/ 1207] Overall Loss 0.198751 Objective Loss 0.198751 LR 0.000250 Time 0.022133 -2023-02-13 18:25:36,588 - Epoch: [153][ 280/ 1207] Overall Loss 0.198159 Objective Loss 0.198159 LR 0.000250 Time 0.022045 -2023-02-13 18:25:36,782 - Epoch: [153][ 290/ 1207] Overall Loss 0.197508 Objective Loss 0.197508 LR 0.000250 Time 0.021953 -2023-02-13 18:25:36,979 - Epoch: [153][ 300/ 1207] Overall Loss 0.197980 Objective Loss 0.197980 LR 0.000250 Time 0.021879 -2023-02-13 18:25:37,174 - Epoch: [153][ 310/ 1207] Overall Loss 0.198144 Objective Loss 0.198144 LR 0.000250 Time 0.021798 -2023-02-13 18:25:37,372 - Epoch: [153][ 320/ 1207] Overall Loss 0.198547 Objective Loss 0.198547 LR 0.000250 Time 0.021735 -2023-02-13 18:25:37,566 - Epoch: [153][ 330/ 1207] Overall Loss 0.198623 Objective Loss 0.198623 LR 0.000250 Time 0.021664 -2023-02-13 18:25:37,763 - Epoch: [153][ 340/ 1207] Overall Loss 0.198255 Objective Loss 0.198255 LR 0.000250 Time 0.021604 -2023-02-13 18:25:37,957 - Epoch: [153][ 350/ 1207] Overall Loss 0.197758 Objective Loss 0.197758 LR 0.000250 Time 0.021542 -2023-02-13 18:25:38,154 - Epoch: [153][ 360/ 1207] Overall Loss 0.198181 Objective Loss 0.198181 LR 0.000250 Time 0.021489 -2023-02-13 18:25:38,349 - Epoch: [153][ 370/ 1207] Overall Loss 0.197834 Objective Loss 0.197834 LR 0.000250 Time 0.021433 -2023-02-13 18:25:38,545 - Epoch: [153][ 380/ 1207] Overall Loss 0.198685 Objective Loss 0.198685 LR 0.000250 Time 0.021385 -2023-02-13 18:25:38,738 - Epoch: [153][ 390/ 1207] Overall Loss 0.198562 Objective Loss 0.198562 LR 0.000250 Time 0.021331 -2023-02-13 18:25:38,935 - Epoch: [153][ 400/ 1207] Overall Loss 0.198961 Objective Loss 0.198961 LR 0.000250 Time 0.021288 -2023-02-13 18:25:39,128 - Epoch: [153][ 410/ 1207] Overall Loss 0.199424 Objective Loss 0.199424 LR 0.000250 Time 0.021239 -2023-02-13 18:25:39,325 - Epoch: [153][ 420/ 1207] Overall Loss 0.199244 Objective Loss 0.199244 LR 0.000250 Time 0.021201 -2023-02-13 18:25:39,519 - Epoch: [153][ 430/ 1207] Overall Loss 0.199083 Objective Loss 0.199083 LR 0.000250 Time 0.021159 -2023-02-13 18:25:39,715 - Epoch: [153][ 440/ 1207] Overall Loss 0.198719 Objective Loss 0.198719 LR 0.000250 Time 0.021122 -2023-02-13 18:25:39,908 - Epoch: [153][ 450/ 1207] Overall Loss 0.198882 Objective Loss 0.198882 LR 0.000250 Time 0.021082 -2023-02-13 18:25:40,105 - Epoch: [153][ 460/ 1207] Overall Loss 0.198623 Objective Loss 0.198623 LR 0.000250 Time 0.021051 -2023-02-13 18:25:40,300 - Epoch: [153][ 470/ 1207] Overall Loss 0.198327 Objective Loss 0.198327 LR 0.000250 Time 0.021017 -2023-02-13 18:25:40,497 - Epoch: [153][ 480/ 1207] Overall Loss 0.198399 Objective Loss 0.198399 LR 0.000250 Time 0.020988 -2023-02-13 18:25:40,690 - Epoch: [153][ 490/ 1207] Overall Loss 0.198178 Objective Loss 0.198178 LR 0.000250 Time 0.020954 -2023-02-13 18:25:40,888 - Epoch: [153][ 500/ 1207] Overall Loss 0.198703 Objective Loss 0.198703 LR 0.000250 Time 0.020930 -2023-02-13 18:25:41,082 - Epoch: [153][ 510/ 1207] Overall Loss 0.199206 Objective Loss 0.199206 LR 0.000250 Time 0.020899 -2023-02-13 18:25:41,280 - Epoch: [153][ 520/ 1207] Overall Loss 0.198658 Objective Loss 0.198658 LR 0.000250 Time 0.020877 -2023-02-13 18:25:41,474 - Epoch: [153][ 530/ 1207] Overall Loss 0.198879 Objective Loss 0.198879 LR 0.000250 Time 0.020849 -2023-02-13 18:25:41,672 - Epoch: [153][ 540/ 1207] Overall Loss 0.198649 Objective Loss 0.198649 LR 0.000250 Time 0.020828 -2023-02-13 18:25:41,866 - Epoch: [153][ 550/ 1207] Overall Loss 0.198245 Objective Loss 0.198245 LR 0.000250 Time 0.020801 -2023-02-13 18:25:42,063 - Epoch: [153][ 560/ 1207] Overall Loss 0.198383 Objective Loss 0.198383 LR 0.000250 Time 0.020780 -2023-02-13 18:25:42,257 - Epoch: [153][ 570/ 1207] Overall Loss 0.198328 Objective Loss 0.198328 LR 0.000250 Time 0.020756 -2023-02-13 18:25:42,454 - Epoch: [153][ 580/ 1207] Overall Loss 0.198091 Objective Loss 0.198091 LR 0.000250 Time 0.020737 -2023-02-13 18:25:42,648 - Epoch: [153][ 590/ 1207] Overall Loss 0.197963 Objective Loss 0.197963 LR 0.000250 Time 0.020714 -2023-02-13 18:25:42,845 - Epoch: [153][ 600/ 1207] Overall Loss 0.198425 Objective Loss 0.198425 LR 0.000250 Time 0.020696 -2023-02-13 18:25:43,039 - Epoch: [153][ 610/ 1207] Overall Loss 0.198617 Objective Loss 0.198617 LR 0.000250 Time 0.020674 -2023-02-13 18:25:43,236 - Epoch: [153][ 620/ 1207] Overall Loss 0.198375 Objective Loss 0.198375 LR 0.000250 Time 0.020658 -2023-02-13 18:25:43,431 - Epoch: [153][ 630/ 1207] Overall Loss 0.198166 Objective Loss 0.198166 LR 0.000250 Time 0.020640 -2023-02-13 18:25:43,629 - Epoch: [153][ 640/ 1207] Overall Loss 0.198295 Objective Loss 0.198295 LR 0.000250 Time 0.020625 -2023-02-13 18:25:43,825 - Epoch: [153][ 650/ 1207] Overall Loss 0.198382 Objective Loss 0.198382 LR 0.000250 Time 0.020609 -2023-02-13 18:25:44,022 - Epoch: [153][ 660/ 1207] Overall Loss 0.198601 Objective Loss 0.198601 LR 0.000250 Time 0.020595 -2023-02-13 18:25:44,217 - Epoch: [153][ 670/ 1207] Overall Loss 0.198876 Objective Loss 0.198876 LR 0.000250 Time 0.020578 -2023-02-13 18:25:44,414 - Epoch: [153][ 680/ 1207] Overall Loss 0.198953 Objective Loss 0.198953 LR 0.000250 Time 0.020565 -2023-02-13 18:25:44,609 - Epoch: [153][ 690/ 1207] Overall Loss 0.199224 Objective Loss 0.199224 LR 0.000250 Time 0.020549 -2023-02-13 18:25:44,806 - Epoch: [153][ 700/ 1207] Overall Loss 0.199045 Objective Loss 0.199045 LR 0.000250 Time 0.020536 -2023-02-13 18:25:45,000 - Epoch: [153][ 710/ 1207] Overall Loss 0.199116 Objective Loss 0.199116 LR 0.000250 Time 0.020519 -2023-02-13 18:25:45,197 - Epoch: [153][ 720/ 1207] Overall Loss 0.199059 Objective Loss 0.199059 LR 0.000250 Time 0.020507 -2023-02-13 18:25:45,392 - Epoch: [153][ 730/ 1207] Overall Loss 0.199042 Objective Loss 0.199042 LR 0.000250 Time 0.020493 -2023-02-13 18:25:45,589 - Epoch: [153][ 740/ 1207] Overall Loss 0.198797 Objective Loss 0.198797 LR 0.000250 Time 0.020482 -2023-02-13 18:25:45,783 - Epoch: [153][ 750/ 1207] Overall Loss 0.198622 Objective Loss 0.198622 LR 0.000250 Time 0.020467 -2023-02-13 18:25:45,981 - Epoch: [153][ 760/ 1207] Overall Loss 0.198401 Objective Loss 0.198401 LR 0.000250 Time 0.020458 -2023-02-13 18:25:46,175 - Epoch: [153][ 770/ 1207] Overall Loss 0.198506 Objective Loss 0.198506 LR 0.000250 Time 0.020443 -2023-02-13 18:25:46,373 - Epoch: [153][ 780/ 1207] Overall Loss 0.198846 Objective Loss 0.198846 LR 0.000250 Time 0.020434 -2023-02-13 18:25:46,567 - Epoch: [153][ 790/ 1207] Overall Loss 0.198679 Objective Loss 0.198679 LR 0.000250 Time 0.020422 -2023-02-13 18:25:46,765 - Epoch: [153][ 800/ 1207] Overall Loss 0.198360 Objective Loss 0.198360 LR 0.000250 Time 0.020413 -2023-02-13 18:25:46,960 - Epoch: [153][ 810/ 1207] Overall Loss 0.198453 Objective Loss 0.198453 LR 0.000250 Time 0.020401 -2023-02-13 18:25:47,157 - Epoch: [153][ 820/ 1207] Overall Loss 0.198522 Objective Loss 0.198522 LR 0.000250 Time 0.020392 -2023-02-13 18:25:47,352 - Epoch: [153][ 830/ 1207] Overall Loss 0.198756 Objective Loss 0.198756 LR 0.000250 Time 0.020381 -2023-02-13 18:25:47,549 - Epoch: [153][ 840/ 1207] Overall Loss 0.199041 Objective Loss 0.199041 LR 0.000250 Time 0.020373 -2023-02-13 18:25:47,744 - Epoch: [153][ 850/ 1207] Overall Loss 0.199200 Objective Loss 0.199200 LR 0.000250 Time 0.020361 -2023-02-13 18:25:47,941 - Epoch: [153][ 860/ 1207] Overall Loss 0.198927 Objective Loss 0.198927 LR 0.000250 Time 0.020354 -2023-02-13 18:25:48,135 - Epoch: [153][ 870/ 1207] Overall Loss 0.198757 Objective Loss 0.198757 LR 0.000250 Time 0.020342 -2023-02-13 18:25:48,333 - Epoch: [153][ 880/ 1207] Overall Loss 0.198758 Objective Loss 0.198758 LR 0.000250 Time 0.020335 -2023-02-13 18:25:48,527 - Epoch: [153][ 890/ 1207] Overall Loss 0.198584 Objective Loss 0.198584 LR 0.000250 Time 0.020325 -2023-02-13 18:25:48,724 - Epoch: [153][ 900/ 1207] Overall Loss 0.198793 Objective Loss 0.198793 LR 0.000250 Time 0.020318 -2023-02-13 18:25:48,919 - Epoch: [153][ 910/ 1207] Overall Loss 0.198746 Objective Loss 0.198746 LR 0.000250 Time 0.020308 -2023-02-13 18:25:49,116 - Epoch: [153][ 920/ 1207] Overall Loss 0.198856 Objective Loss 0.198856 LR 0.000250 Time 0.020301 -2023-02-13 18:25:49,311 - Epoch: [153][ 930/ 1207] Overall Loss 0.198751 Objective Loss 0.198751 LR 0.000250 Time 0.020292 -2023-02-13 18:25:49,508 - Epoch: [153][ 940/ 1207] Overall Loss 0.198744 Objective Loss 0.198744 LR 0.000250 Time 0.020285 -2023-02-13 18:25:49,702 - Epoch: [153][ 950/ 1207] Overall Loss 0.198742 Objective Loss 0.198742 LR 0.000250 Time 0.020276 -2023-02-13 18:25:49,900 - Epoch: [153][ 960/ 1207] Overall Loss 0.198863 Objective Loss 0.198863 LR 0.000250 Time 0.020271 -2023-02-13 18:25:50,095 - Epoch: [153][ 970/ 1207] Overall Loss 0.198770 Objective Loss 0.198770 LR 0.000250 Time 0.020262 -2023-02-13 18:25:50,293 - Epoch: [153][ 980/ 1207] Overall Loss 0.199038 Objective Loss 0.199038 LR 0.000250 Time 0.020256 -2023-02-13 18:25:50,488 - Epoch: [153][ 990/ 1207] Overall Loss 0.199161 Objective Loss 0.199161 LR 0.000250 Time 0.020248 -2023-02-13 18:25:50,684 - Epoch: [153][ 1000/ 1207] Overall Loss 0.199236 Objective Loss 0.199236 LR 0.000250 Time 0.020242 -2023-02-13 18:25:50,879 - Epoch: [153][ 1010/ 1207] Overall Loss 0.199111 Objective Loss 0.199111 LR 0.000250 Time 0.020234 -2023-02-13 18:25:51,077 - Epoch: [153][ 1020/ 1207] Overall Loss 0.199058 Objective Loss 0.199058 LR 0.000250 Time 0.020230 -2023-02-13 18:25:51,272 - Epoch: [153][ 1030/ 1207] Overall Loss 0.199236 Objective Loss 0.199236 LR 0.000250 Time 0.020222 -2023-02-13 18:25:51,470 - Epoch: [153][ 1040/ 1207] Overall Loss 0.199310 Objective Loss 0.199310 LR 0.000250 Time 0.020217 -2023-02-13 18:25:51,664 - Epoch: [153][ 1050/ 1207] Overall Loss 0.199420 Objective Loss 0.199420 LR 0.000250 Time 0.020209 -2023-02-13 18:25:51,861 - Epoch: [153][ 1060/ 1207] Overall Loss 0.199219 Objective Loss 0.199219 LR 0.000250 Time 0.020205 -2023-02-13 18:25:52,056 - Epoch: [153][ 1070/ 1207] Overall Loss 0.199217 Objective Loss 0.199217 LR 0.000250 Time 0.020197 -2023-02-13 18:25:52,253 - Epoch: [153][ 1080/ 1207] Overall Loss 0.199278 Objective Loss 0.199278 LR 0.000250 Time 0.020192 -2023-02-13 18:25:52,447 - Epoch: [153][ 1090/ 1207] Overall Loss 0.199252 Objective Loss 0.199252 LR 0.000250 Time 0.020185 -2023-02-13 18:25:52,644 - Epoch: [153][ 1100/ 1207] Overall Loss 0.199689 Objective Loss 0.199689 LR 0.000250 Time 0.020180 -2023-02-13 18:25:52,838 - Epoch: [153][ 1110/ 1207] Overall Loss 0.199781 Objective Loss 0.199781 LR 0.000250 Time 0.020173 -2023-02-13 18:25:53,036 - Epoch: [153][ 1120/ 1207] Overall Loss 0.199792 Objective Loss 0.199792 LR 0.000250 Time 0.020169 -2023-02-13 18:25:53,230 - Epoch: [153][ 1130/ 1207] Overall Loss 0.199851 Objective Loss 0.199851 LR 0.000250 Time 0.020162 -2023-02-13 18:25:53,428 - Epoch: [153][ 1140/ 1207] Overall Loss 0.199872 Objective Loss 0.199872 LR 0.000250 Time 0.020159 -2023-02-13 18:25:53,622 - Epoch: [153][ 1150/ 1207] Overall Loss 0.199952 Objective Loss 0.199952 LR 0.000250 Time 0.020151 -2023-02-13 18:25:53,819 - Epoch: [153][ 1160/ 1207] Overall Loss 0.199884 Objective Loss 0.199884 LR 0.000250 Time 0.020147 -2023-02-13 18:25:54,013 - Epoch: [153][ 1170/ 1207] Overall Loss 0.199720 Objective Loss 0.199720 LR 0.000250 Time 0.020141 -2023-02-13 18:25:54,210 - Epoch: [153][ 1180/ 1207] Overall Loss 0.199609 Objective Loss 0.199609 LR 0.000250 Time 0.020137 -2023-02-13 18:25:54,405 - Epoch: [153][ 1190/ 1207] Overall Loss 0.199448 Objective Loss 0.199448 LR 0.000250 Time 0.020131 -2023-02-13 18:25:54,650 - Epoch: [153][ 1200/ 1207] Overall Loss 0.199541 Objective Loss 0.199541 LR 0.000250 Time 0.020167 -2023-02-13 18:25:54,765 - Epoch: [153][ 1207/ 1207] Overall Loss 0.199554 Objective Loss 0.199554 Top1 88.109756 Top5 99.085366 LR 0.000250 Time 0.020146 -2023-02-13 18:25:54,838 - --- validate (epoch=153)----------- -2023-02-13 18:25:54,838 - 34311 samples (256 per mini-batch) -2023-02-13 18:25:55,345 - Epoch: [153][ 10/ 135] Loss 0.314737 Top1 84.335938 Top5 97.421875 -2023-02-13 18:25:55,469 - Epoch: [153][ 20/ 135] Loss 0.288726 Top1 85.039062 Top5 97.949219 -2023-02-13 18:25:55,594 - Epoch: [153][ 30/ 135] Loss 0.293363 Top1 85.195312 Top5 97.916667 -2023-02-13 18:25:55,721 - Epoch: [153][ 40/ 135] Loss 0.296318 Top1 85.195312 Top5 97.968750 -2023-02-13 18:25:55,849 - Epoch: [153][ 50/ 135] Loss 0.291035 Top1 85.406250 Top5 97.960938 -2023-02-13 18:25:55,976 - Epoch: [153][ 60/ 135] Loss 0.288988 Top1 85.423177 Top5 97.968750 -2023-02-13 18:25:56,103 - Epoch: [153][ 70/ 135] Loss 0.286943 Top1 85.541295 Top5 97.991071 -2023-02-13 18:25:56,228 - Epoch: [153][ 80/ 135] Loss 0.287417 Top1 85.512695 Top5 97.998047 -2023-02-13 18:25:56,357 - Epoch: [153][ 90/ 135] Loss 0.288187 Top1 85.516493 Top5 97.968750 -2023-02-13 18:25:56,485 - Epoch: [153][ 100/ 135] Loss 0.289469 Top1 85.445312 Top5 97.941406 -2023-02-13 18:25:56,614 - Epoch: [153][ 110/ 135] Loss 0.289278 Top1 85.433239 Top5 97.965199 -2023-02-13 18:25:56,743 - Epoch: [153][ 120/ 135] Loss 0.291342 Top1 85.410156 Top5 97.945964 -2023-02-13 18:25:56,876 - Epoch: [153][ 130/ 135] Loss 0.293897 Top1 85.252404 Top5 97.938702 -2023-02-13 18:25:56,923 - Epoch: [153][ 135/ 135] Loss 0.293876 Top1 85.261286 Top5 97.936522 -2023-02-13 18:25:57,003 - ==> Top1: 85.261 Top5: 97.937 Loss: 0.294 - -2023-02-13 18:25:57,004 - ==> Confusion: -[[ 856 2 5 0 6 4 0 2 5 56 1 2 0 6 7 3 4 2 2 0 4] - [ 3 953 1 3 7 29 0 17 3 0 1 1 1 0 0 1 3 1 2 2 5] - [ 5 5 965 17 2 2 12 12 0 1 3 1 2 7 6 8 0 2 3 1 4] - [ 3 0 14 923 5 4 0 1 3 2 12 0 6 1 12 0 3 8 13 0 6] - [ 12 8 0 1 987 14 1 1 1 2 0 4 1 3 8 5 5 1 1 5 6] - [ 2 11 0 4 5 979 3 18 2 6 1 8 1 13 0 3 3 0 1 6 4] - [ 3 5 17 2 0 7 1030 4 1 1 5 0 3 1 0 1 2 2 1 9 5] - [ 1 5 8 1 1 21 2 949 0 2 2 5 2 1 0 0 1 1 12 6 4] - [ 15 1 1 1 1 0 1 2 915 34 6 3 0 11 9 3 2 0 3 0 1] - [ 54 1 3 1 6 2 0 4 32 883 0 0 0 13 4 3 0 2 0 1 3] - [ 1 1 4 6 1 4 3 4 17 0 984 2 1 7 4 0 0 1 5 1 5] - [ 0 2 0 1 5 15 0 8 2 2 1 925 16 6 1 2 2 6 2 8 1] - [ 0 0 1 9 1 4 0 0 3 3 1 38 870 2 1 4 3 8 3 2 6] - [ 3 1 1 0 7 10 0 2 13 11 8 6 3 942 2 6 2 2 1 0 4] - [ 5 2 2 22 4 5 0 1 23 7 5 2 2 1 988 0 2 4 7 0 10] - [ 3 2 9 0 7 3 3 1 0 0 0 6 4 3 0 964 11 15 0 7 8] - [ 5 3 0 1 10 3 0 0 3 0 1 2 2 1 0 9 1004 3 0 5 9] - [ 4 2 1 3 2 2 4 0 0 1 2 14 16 1 0 10 0 979 2 2 6] - [ 3 4 4 10 0 3 0 31 4 1 6 1 2 0 11 1 0 1 999 3 2] - [ 0 2 2 0 3 6 5 10 0 0 3 11 1 3 0 5 5 2 1 1082 7] - [ 136 221 219 138 97 254 91 191 96 85 189 105 286 300 119 79 278 93 137 242 10078]] - -2023-02-13 18:25:57,006 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:25:57,006 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:25:57,011 - - -2023-02-13 18:25:57,012 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:25:57,900 - Epoch: [154][ 10/ 1207] Overall Loss 0.195153 Objective Loss 0.195153 LR 0.000250 Time 0.088749 -2023-02-13 18:25:58,100 - Epoch: [154][ 20/ 1207] Overall Loss 0.202880 Objective Loss 0.202880 LR 0.000250 Time 0.054350 -2023-02-13 18:25:58,289 - Epoch: [154][ 30/ 1207] Overall Loss 0.198025 Objective Loss 0.198025 LR 0.000250 Time 0.042546 -2023-02-13 18:25:58,479 - Epoch: [154][ 40/ 1207] Overall Loss 0.200165 Objective Loss 0.200165 LR 0.000250 Time 0.036648 -2023-02-13 18:25:58,669 - Epoch: [154][ 50/ 1207] Overall Loss 0.207065 Objective Loss 0.207065 LR 0.000250 Time 0.033104 -2023-02-13 18:25:58,858 - Epoch: [154][ 60/ 1207] Overall Loss 0.205151 Objective Loss 0.205151 LR 0.000250 Time 0.030735 -2023-02-13 18:25:59,048 - Epoch: [154][ 70/ 1207] Overall Loss 0.204103 Objective Loss 0.204103 LR 0.000250 Time 0.029054 -2023-02-13 18:25:59,237 - Epoch: [154][ 80/ 1207] Overall Loss 0.202619 Objective Loss 0.202619 LR 0.000250 Time 0.027781 -2023-02-13 18:25:59,428 - Epoch: [154][ 90/ 1207] Overall Loss 0.200546 Objective Loss 0.200546 LR 0.000250 Time 0.026805 -2023-02-13 18:25:59,617 - Epoch: [154][ 100/ 1207] Overall Loss 0.202948 Objective Loss 0.202948 LR 0.000250 Time 0.026014 -2023-02-13 18:25:59,806 - Epoch: [154][ 110/ 1207] Overall Loss 0.200897 Objective Loss 0.200897 LR 0.000250 Time 0.025367 -2023-02-13 18:25:59,996 - Epoch: [154][ 120/ 1207] Overall Loss 0.197914 Objective Loss 0.197914 LR 0.000250 Time 0.024827 -2023-02-13 18:26:00,186 - Epoch: [154][ 130/ 1207] Overall Loss 0.197705 Objective Loss 0.197705 LR 0.000250 Time 0.024375 -2023-02-13 18:26:00,376 - Epoch: [154][ 140/ 1207] Overall Loss 0.198578 Objective Loss 0.198578 LR 0.000250 Time 0.023989 -2023-02-13 18:26:00,565 - Epoch: [154][ 150/ 1207] Overall Loss 0.198509 Objective Loss 0.198509 LR 0.000250 Time 0.023649 -2023-02-13 18:26:00,754 - Epoch: [154][ 160/ 1207] Overall Loss 0.198258 Objective Loss 0.198258 LR 0.000250 Time 0.023353 -2023-02-13 18:26:00,945 - Epoch: [154][ 170/ 1207] Overall Loss 0.197540 Objective Loss 0.197540 LR 0.000250 Time 0.023096 -2023-02-13 18:26:01,134 - Epoch: [154][ 180/ 1207] Overall Loss 0.197816 Objective Loss 0.197816 LR 0.000250 Time 0.022860 -2023-02-13 18:26:01,324 - Epoch: [154][ 190/ 1207] Overall Loss 0.198115 Objective Loss 0.198115 LR 0.000250 Time 0.022655 -2023-02-13 18:26:01,514 - Epoch: [154][ 200/ 1207] Overall Loss 0.198215 Objective Loss 0.198215 LR 0.000250 Time 0.022470 -2023-02-13 18:26:01,703 - Epoch: [154][ 210/ 1207] Overall Loss 0.197797 Objective Loss 0.197797 LR 0.000250 Time 0.022302 -2023-02-13 18:26:01,893 - Epoch: [154][ 220/ 1207] Overall Loss 0.197501 Objective Loss 0.197501 LR 0.000250 Time 0.022151 -2023-02-13 18:26:02,083 - Epoch: [154][ 230/ 1207] Overall Loss 0.197368 Objective Loss 0.197368 LR 0.000250 Time 0.022011 -2023-02-13 18:26:02,273 - Epoch: [154][ 240/ 1207] Overall Loss 0.196779 Objective Loss 0.196779 LR 0.000250 Time 0.021882 -2023-02-13 18:26:02,463 - Epoch: [154][ 250/ 1207] Overall Loss 0.197503 Objective Loss 0.197503 LR 0.000250 Time 0.021767 -2023-02-13 18:26:02,653 - Epoch: [154][ 260/ 1207] Overall Loss 0.197454 Objective Loss 0.197454 LR 0.000250 Time 0.021658 -2023-02-13 18:26:02,843 - Epoch: [154][ 270/ 1207] Overall Loss 0.197880 Objective Loss 0.197880 LR 0.000250 Time 0.021557 -2023-02-13 18:26:03,033 - Epoch: [154][ 280/ 1207] Overall Loss 0.198446 Objective Loss 0.198446 LR 0.000250 Time 0.021465 -2023-02-13 18:26:03,221 - Epoch: [154][ 290/ 1207] Overall Loss 0.198575 Objective Loss 0.198575 LR 0.000250 Time 0.021374 -2023-02-13 18:26:03,411 - Epoch: [154][ 300/ 1207] Overall Loss 0.198743 Objective Loss 0.198743 LR 0.000250 Time 0.021293 -2023-02-13 18:26:03,601 - Epoch: [154][ 310/ 1207] Overall Loss 0.198656 Objective Loss 0.198656 LR 0.000250 Time 0.021218 -2023-02-13 18:26:03,791 - Epoch: [154][ 320/ 1207] Overall Loss 0.198265 Objective Loss 0.198265 LR 0.000250 Time 0.021147 -2023-02-13 18:26:03,981 - Epoch: [154][ 330/ 1207] Overall Loss 0.197616 Objective Loss 0.197616 LR 0.000250 Time 0.021080 -2023-02-13 18:26:04,170 - Epoch: [154][ 340/ 1207] Overall Loss 0.197448 Objective Loss 0.197448 LR 0.000250 Time 0.021015 -2023-02-13 18:26:04,360 - Epoch: [154][ 350/ 1207] Overall Loss 0.197321 Objective Loss 0.197321 LR 0.000250 Time 0.020956 -2023-02-13 18:26:04,549 - Epoch: [154][ 360/ 1207] Overall Loss 0.197376 Objective Loss 0.197376 LR 0.000250 Time 0.020899 -2023-02-13 18:26:04,739 - Epoch: [154][ 370/ 1207] Overall Loss 0.197181 Objective Loss 0.197181 LR 0.000250 Time 0.020845 -2023-02-13 18:26:04,928 - Epoch: [154][ 380/ 1207] Overall Loss 0.197220 Objective Loss 0.197220 LR 0.000250 Time 0.020793 -2023-02-13 18:26:05,117 - Epoch: [154][ 390/ 1207] Overall Loss 0.196993 Objective Loss 0.196993 LR 0.000250 Time 0.020745 -2023-02-13 18:26:05,307 - Epoch: [154][ 400/ 1207] Overall Loss 0.197314 Objective Loss 0.197314 LR 0.000250 Time 0.020700 -2023-02-13 18:26:05,497 - Epoch: [154][ 410/ 1207] Overall Loss 0.197571 Objective Loss 0.197571 LR 0.000250 Time 0.020658 -2023-02-13 18:26:05,687 - Epoch: [154][ 420/ 1207] Overall Loss 0.197826 Objective Loss 0.197826 LR 0.000250 Time 0.020618 -2023-02-13 18:26:05,877 - Epoch: [154][ 430/ 1207] Overall Loss 0.197604 Objective Loss 0.197604 LR 0.000250 Time 0.020579 -2023-02-13 18:26:06,067 - Epoch: [154][ 440/ 1207] Overall Loss 0.198050 Objective Loss 0.198050 LR 0.000250 Time 0.020543 -2023-02-13 18:26:06,257 - Epoch: [154][ 450/ 1207] Overall Loss 0.198065 Objective Loss 0.198065 LR 0.000250 Time 0.020507 -2023-02-13 18:26:06,447 - Epoch: [154][ 460/ 1207] Overall Loss 0.197799 Objective Loss 0.197799 LR 0.000250 Time 0.020474 -2023-02-13 18:26:06,637 - Epoch: [154][ 470/ 1207] Overall Loss 0.197926 Objective Loss 0.197926 LR 0.000250 Time 0.020442 -2023-02-13 18:26:06,827 - Epoch: [154][ 480/ 1207] Overall Loss 0.197589 Objective Loss 0.197589 LR 0.000250 Time 0.020410 -2023-02-13 18:26:07,017 - Epoch: [154][ 490/ 1207] Overall Loss 0.197860 Objective Loss 0.197860 LR 0.000250 Time 0.020381 -2023-02-13 18:26:07,207 - Epoch: [154][ 500/ 1207] Overall Loss 0.197695 Objective Loss 0.197695 LR 0.000250 Time 0.020352 -2023-02-13 18:26:07,397 - Epoch: [154][ 510/ 1207] Overall Loss 0.197922 Objective Loss 0.197922 LR 0.000250 Time 0.020325 -2023-02-13 18:26:07,586 - Epoch: [154][ 520/ 1207] Overall Loss 0.198142 Objective Loss 0.198142 LR 0.000250 Time 0.020297 -2023-02-13 18:26:07,776 - Epoch: [154][ 530/ 1207] Overall Loss 0.197902 Objective Loss 0.197902 LR 0.000250 Time 0.020272 -2023-02-13 18:26:07,966 - Epoch: [154][ 540/ 1207] Overall Loss 0.198144 Objective Loss 0.198144 LR 0.000250 Time 0.020247 -2023-02-13 18:26:08,155 - Epoch: [154][ 550/ 1207] Overall Loss 0.198156 Objective Loss 0.198156 LR 0.000250 Time 0.020223 -2023-02-13 18:26:08,346 - Epoch: [154][ 560/ 1207] Overall Loss 0.197845 Objective Loss 0.197845 LR 0.000250 Time 0.020201 -2023-02-13 18:26:08,535 - Epoch: [154][ 570/ 1207] Overall Loss 0.197801 Objective Loss 0.197801 LR 0.000250 Time 0.020179 -2023-02-13 18:26:08,725 - Epoch: [154][ 580/ 1207] Overall Loss 0.197816 Objective Loss 0.197816 LR 0.000250 Time 0.020158 -2023-02-13 18:26:08,916 - Epoch: [154][ 590/ 1207] Overall Loss 0.197672 Objective Loss 0.197672 LR 0.000250 Time 0.020138 -2023-02-13 18:26:09,105 - Epoch: [154][ 600/ 1207] Overall Loss 0.197477 Objective Loss 0.197477 LR 0.000250 Time 0.020118 -2023-02-13 18:26:09,295 - Epoch: [154][ 610/ 1207] Overall Loss 0.197931 Objective Loss 0.197931 LR 0.000250 Time 0.020099 -2023-02-13 18:26:09,485 - Epoch: [154][ 620/ 1207] Overall Loss 0.198057 Objective Loss 0.198057 LR 0.000250 Time 0.020081 -2023-02-13 18:26:09,675 - Epoch: [154][ 630/ 1207] Overall Loss 0.198069 Objective Loss 0.198069 LR 0.000250 Time 0.020063 -2023-02-13 18:26:09,866 - Epoch: [154][ 640/ 1207] Overall Loss 0.197868 Objective Loss 0.197868 LR 0.000250 Time 0.020046 -2023-02-13 18:26:10,056 - Epoch: [154][ 650/ 1207] Overall Loss 0.198088 Objective Loss 0.198088 LR 0.000250 Time 0.020030 -2023-02-13 18:26:10,246 - Epoch: [154][ 660/ 1207] Overall Loss 0.197883 Objective Loss 0.197883 LR 0.000250 Time 0.020013 -2023-02-13 18:26:10,437 - Epoch: [154][ 670/ 1207] Overall Loss 0.198148 Objective Loss 0.198148 LR 0.000250 Time 0.019999 -2023-02-13 18:26:10,627 - Epoch: [154][ 680/ 1207] Overall Loss 0.197952 Objective Loss 0.197952 LR 0.000250 Time 0.019984 -2023-02-13 18:26:10,816 - Epoch: [154][ 690/ 1207] Overall Loss 0.197782 Objective Loss 0.197782 LR 0.000250 Time 0.019969 -2023-02-13 18:26:11,007 - Epoch: [154][ 700/ 1207] Overall Loss 0.197593 Objective Loss 0.197593 LR 0.000250 Time 0.019955 -2023-02-13 18:26:11,197 - Epoch: [154][ 710/ 1207] Overall Loss 0.197922 Objective Loss 0.197922 LR 0.000250 Time 0.019942 -2023-02-13 18:26:11,388 - Epoch: [154][ 720/ 1207] Overall Loss 0.197872 Objective Loss 0.197872 LR 0.000250 Time 0.019929 -2023-02-13 18:26:11,578 - Epoch: [154][ 730/ 1207] Overall Loss 0.197804 Objective Loss 0.197804 LR 0.000250 Time 0.019916 -2023-02-13 18:26:11,768 - Epoch: [154][ 740/ 1207] Overall Loss 0.197678 Objective Loss 0.197678 LR 0.000250 Time 0.019903 -2023-02-13 18:26:11,959 - Epoch: [154][ 750/ 1207] Overall Loss 0.197767 Objective Loss 0.197767 LR 0.000250 Time 0.019891 -2023-02-13 18:26:12,148 - Epoch: [154][ 760/ 1207] Overall Loss 0.197793 Objective Loss 0.197793 LR 0.000250 Time 0.019879 -2023-02-13 18:26:12,338 - Epoch: [154][ 770/ 1207] Overall Loss 0.197890 Objective Loss 0.197890 LR 0.000250 Time 0.019867 -2023-02-13 18:26:12,529 - Epoch: [154][ 780/ 1207] Overall Loss 0.197852 Objective Loss 0.197852 LR 0.000250 Time 0.019857 -2023-02-13 18:26:12,720 - Epoch: [154][ 790/ 1207] Overall Loss 0.197678 Objective Loss 0.197678 LR 0.000250 Time 0.019846 -2023-02-13 18:26:12,909 - Epoch: [154][ 800/ 1207] Overall Loss 0.197563 Objective Loss 0.197563 LR 0.000250 Time 0.019834 -2023-02-13 18:26:13,100 - Epoch: [154][ 810/ 1207] Overall Loss 0.197208 Objective Loss 0.197208 LR 0.000250 Time 0.019824 -2023-02-13 18:26:13,290 - Epoch: [154][ 820/ 1207] Overall Loss 0.197480 Objective Loss 0.197480 LR 0.000250 Time 0.019814 -2023-02-13 18:26:13,481 - Epoch: [154][ 830/ 1207] Overall Loss 0.197354 Objective Loss 0.197354 LR 0.000250 Time 0.019804 -2023-02-13 18:26:13,671 - Epoch: [154][ 840/ 1207] Overall Loss 0.197090 Objective Loss 0.197090 LR 0.000250 Time 0.019794 -2023-02-13 18:26:13,861 - Epoch: [154][ 850/ 1207] Overall Loss 0.197093 Objective Loss 0.197093 LR 0.000250 Time 0.019785 -2023-02-13 18:26:14,052 - Epoch: [154][ 860/ 1207] Overall Loss 0.197009 Objective Loss 0.197009 LR 0.000250 Time 0.019776 -2023-02-13 18:26:14,242 - Epoch: [154][ 870/ 1207] Overall Loss 0.196686 Objective Loss 0.196686 LR 0.000250 Time 0.019767 -2023-02-13 18:26:14,432 - Epoch: [154][ 880/ 1207] Overall Loss 0.196659 Objective Loss 0.196659 LR 0.000250 Time 0.019758 -2023-02-13 18:26:14,622 - Epoch: [154][ 890/ 1207] Overall Loss 0.196588 Objective Loss 0.196588 LR 0.000250 Time 0.019749 -2023-02-13 18:26:14,812 - Epoch: [154][ 900/ 1207] Overall Loss 0.196534 Objective Loss 0.196534 LR 0.000250 Time 0.019740 -2023-02-13 18:26:15,003 - Epoch: [154][ 910/ 1207] Overall Loss 0.196412 Objective Loss 0.196412 LR 0.000250 Time 0.019733 -2023-02-13 18:26:15,193 - Epoch: [154][ 920/ 1207] Overall Loss 0.196318 Objective Loss 0.196318 LR 0.000250 Time 0.019724 -2023-02-13 18:26:15,384 - Epoch: [154][ 930/ 1207] Overall Loss 0.196070 Objective Loss 0.196070 LR 0.000250 Time 0.019717 -2023-02-13 18:26:15,574 - Epoch: [154][ 940/ 1207] Overall Loss 0.196377 Objective Loss 0.196377 LR 0.000250 Time 0.019709 -2023-02-13 18:26:15,764 - Epoch: [154][ 950/ 1207] Overall Loss 0.196417 Objective Loss 0.196417 LR 0.000250 Time 0.019701 -2023-02-13 18:26:15,955 - Epoch: [154][ 960/ 1207] Overall Loss 0.196503 Objective Loss 0.196503 LR 0.000250 Time 0.019695 -2023-02-13 18:26:16,145 - Epoch: [154][ 970/ 1207] Overall Loss 0.196835 Objective Loss 0.196835 LR 0.000250 Time 0.019687 -2023-02-13 18:26:16,335 - Epoch: [154][ 980/ 1207] Overall Loss 0.196677 Objective Loss 0.196677 LR 0.000250 Time 0.019680 -2023-02-13 18:26:16,526 - Epoch: [154][ 990/ 1207] Overall Loss 0.196773 Objective Loss 0.196773 LR 0.000250 Time 0.019674 -2023-02-13 18:26:16,716 - Epoch: [154][ 1000/ 1207] Overall Loss 0.196547 Objective Loss 0.196547 LR 0.000250 Time 0.019666 -2023-02-13 18:26:16,907 - Epoch: [154][ 1010/ 1207] Overall Loss 0.196572 Objective Loss 0.196572 LR 0.000250 Time 0.019660 -2023-02-13 18:26:17,098 - Epoch: [154][ 1020/ 1207] Overall Loss 0.196569 Objective Loss 0.196569 LR 0.000250 Time 0.019654 -2023-02-13 18:26:17,288 - Epoch: [154][ 1030/ 1207] Overall Loss 0.196584 Objective Loss 0.196584 LR 0.000250 Time 0.019648 -2023-02-13 18:26:17,479 - Epoch: [154][ 1040/ 1207] Overall Loss 0.196600 Objective Loss 0.196600 LR 0.000250 Time 0.019642 -2023-02-13 18:26:17,669 - Epoch: [154][ 1050/ 1207] Overall Loss 0.196761 Objective Loss 0.196761 LR 0.000250 Time 0.019635 -2023-02-13 18:26:17,859 - Epoch: [154][ 1060/ 1207] Overall Loss 0.196824 Objective Loss 0.196824 LR 0.000250 Time 0.019629 -2023-02-13 18:26:18,049 - Epoch: [154][ 1070/ 1207] Overall Loss 0.196683 Objective Loss 0.196683 LR 0.000250 Time 0.019623 -2023-02-13 18:26:18,239 - Epoch: [154][ 1080/ 1207] Overall Loss 0.196985 Objective Loss 0.196985 LR 0.000250 Time 0.019617 -2023-02-13 18:26:18,429 - Epoch: [154][ 1090/ 1207] Overall Loss 0.197101 Objective Loss 0.197101 LR 0.000250 Time 0.019611 -2023-02-13 18:26:18,619 - Epoch: [154][ 1100/ 1207] Overall Loss 0.197199 Objective Loss 0.197199 LR 0.000250 Time 0.019605 -2023-02-13 18:26:18,809 - Epoch: [154][ 1110/ 1207] Overall Loss 0.197216 Objective Loss 0.197216 LR 0.000250 Time 0.019599 -2023-02-13 18:26:18,998 - Epoch: [154][ 1120/ 1207] Overall Loss 0.197206 Objective Loss 0.197206 LR 0.000250 Time 0.019593 -2023-02-13 18:26:19,188 - Epoch: [154][ 1130/ 1207] Overall Loss 0.197194 Objective Loss 0.197194 LR 0.000250 Time 0.019587 -2023-02-13 18:26:19,378 - Epoch: [154][ 1140/ 1207] Overall Loss 0.197030 Objective Loss 0.197030 LR 0.000250 Time 0.019582 -2023-02-13 18:26:19,568 - Epoch: [154][ 1150/ 1207] Overall Loss 0.197121 Objective Loss 0.197121 LR 0.000250 Time 0.019577 -2023-02-13 18:26:19,757 - Epoch: [154][ 1160/ 1207] Overall Loss 0.196927 Objective Loss 0.196927 LR 0.000250 Time 0.019570 -2023-02-13 18:26:19,948 - Epoch: [154][ 1170/ 1207] Overall Loss 0.196966 Objective Loss 0.196966 LR 0.000250 Time 0.019566 -2023-02-13 18:26:20,137 - Epoch: [154][ 1180/ 1207] Overall Loss 0.196910 Objective Loss 0.196910 LR 0.000250 Time 0.019560 -2023-02-13 18:26:20,327 - Epoch: [154][ 1190/ 1207] Overall Loss 0.196974 Objective Loss 0.196974 LR 0.000250 Time 0.019555 -2023-02-13 18:26:20,569 - Epoch: [154][ 1200/ 1207] Overall Loss 0.197040 Objective Loss 0.197040 LR 0.000250 Time 0.019593 -2023-02-13 18:26:20,684 - Epoch: [154][ 1207/ 1207] Overall Loss 0.197020 Objective Loss 0.197020 Top1 88.414634 Top5 99.085366 LR 0.000250 Time 0.019575 -2023-02-13 18:26:20,757 - --- validate (epoch=154)----------- -2023-02-13 18:26:20,757 - 34311 samples (256 per mini-batch) -2023-02-13 18:26:21,169 - Epoch: [154][ 10/ 135] Loss 0.284460 Top1 85.781250 Top5 97.929688 -2023-02-13 18:26:21,308 - Epoch: [154][ 20/ 135] Loss 0.300647 Top1 85.097656 Top5 97.656250 -2023-02-13 18:26:21,451 - Epoch: [154][ 30/ 135] Loss 0.303187 Top1 85.247396 Top5 97.708333 -2023-02-13 18:26:21,588 - Epoch: [154][ 40/ 135] Loss 0.296114 Top1 85.488281 Top5 97.822266 -2023-02-13 18:26:21,723 - Epoch: [154][ 50/ 135] Loss 0.298093 Top1 85.359375 Top5 97.898438 -2023-02-13 18:26:21,848 - Epoch: [154][ 60/ 135] Loss 0.299287 Top1 85.377604 Top5 97.910156 -2023-02-13 18:26:21,992 - Epoch: [154][ 70/ 135] Loss 0.299189 Top1 85.334821 Top5 97.868304 -2023-02-13 18:26:22,132 - Epoch: [154][ 80/ 135] Loss 0.298867 Top1 85.341797 Top5 97.910156 -2023-02-13 18:26:22,274 - Epoch: [154][ 90/ 135] Loss 0.298733 Top1 85.277778 Top5 97.921007 -2023-02-13 18:26:22,414 - Epoch: [154][ 100/ 135] Loss 0.299116 Top1 85.312500 Top5 97.945312 -2023-02-13 18:26:22,558 - Epoch: [154][ 110/ 135] Loss 0.297978 Top1 85.227273 Top5 97.904830 -2023-02-13 18:26:22,698 - Epoch: [154][ 120/ 135] Loss 0.297631 Top1 85.208333 Top5 97.871094 -2023-02-13 18:26:22,838 - Epoch: [154][ 130/ 135] Loss 0.296423 Top1 85.279447 Top5 97.884615 -2023-02-13 18:26:22,882 - Epoch: [154][ 135/ 135] Loss 0.302103 Top1 85.284603 Top5 97.892804 -2023-02-13 18:26:22,955 - ==> Top1: 85.285 Top5: 97.893 Loss: 0.302 - -2023-02-13 18:26:22,955 - ==> Confusion: -[[ 866 5 8 1 7 3 0 2 5 36 1 2 1 4 6 4 3 2 1 2 8] - [ 2 980 1 2 6 7 1 13 3 0 0 0 0 0 1 1 7 1 1 2 5] - [ 5 9 961 12 3 2 14 14 0 1 2 2 2 5 6 6 2 1 6 1 4] - [ 4 3 19 904 3 3 1 2 3 1 11 1 3 1 17 1 2 4 24 0 9] - [ 7 11 0 0 997 10 1 2 3 0 0 4 1 3 8 5 7 1 1 2 3] - [ 0 30 1 2 5 947 4 23 2 4 1 13 4 11 0 4 4 0 1 7 7] - [ 4 5 19 1 1 6 1034 6 0 1 1 1 3 2 0 2 2 1 1 5 4] - [ 1 15 6 1 1 22 3 941 0 2 1 5 1 0 0 0 2 1 12 5 5] - [ 14 1 1 1 1 0 1 3 915 31 6 2 0 9 14 2 3 1 3 0 1] - [ 87 3 3 0 9 1 0 4 33 842 0 0 1 16 4 2 0 2 0 0 5] - [ 0 4 2 6 1 0 2 7 17 1 984 1 1 6 3 0 1 0 8 0 7] - [ 1 3 0 1 2 9 0 4 2 2 0 936 14 4 2 7 4 4 1 7 2] - [ 0 0 0 8 2 4 0 1 1 1 0 39 864 0 4 11 4 10 3 0 7] - [ 3 2 1 0 8 11 0 1 9 15 5 7 1 938 3 5 3 2 1 3 6] - [ 6 5 0 10 4 3 0 1 22 6 3 1 2 2 999 2 2 5 10 1 8] - [ 3 3 6 0 5 1 3 0 0 0 0 3 4 3 1 981 13 9 1 6 4] - [ 4 6 0 1 12 1 0 1 1 0 0 0 0 2 1 6 1010 3 1 3 9] - [ 3 3 0 4 1 1 2 1 0 0 2 13 13 0 0 15 0 984 1 3 5] - [ 2 3 2 4 1 2 1 23 4 1 5 1 0 0 10 1 1 1 1020 1 3] - [ 0 3 1 0 2 6 5 9 0 0 0 20 3 3 1 4 8 2 0 1074 7] - [ 137 329 198 101 133 214 69 179 91 73 157 137 272 256 148 105 300 92 178 180 10085]] - -2023-02-13 18:26:22,957 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:26:22,957 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:26:22,963 - - -2023-02-13 18:26:22,963 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:26:23,962 - Epoch: [155][ 10/ 1207] Overall Loss 0.226612 Objective Loss 0.226612 LR 0.000250 Time 0.099862 -2023-02-13 18:26:24,164 - Epoch: [155][ 20/ 1207] Overall Loss 0.207364 Objective Loss 0.207364 LR 0.000250 Time 0.060016 -2023-02-13 18:26:24,359 - Epoch: [155][ 30/ 1207] Overall Loss 0.198063 Objective Loss 0.198063 LR 0.000250 Time 0.046499 -2023-02-13 18:26:24,553 - Epoch: [155][ 40/ 1207] Overall Loss 0.199795 Objective Loss 0.199795 LR 0.000250 Time 0.039701 -2023-02-13 18:26:24,748 - Epoch: [155][ 50/ 1207] Overall Loss 0.199613 Objective Loss 0.199613 LR 0.000250 Time 0.035660 -2023-02-13 18:26:24,941 - Epoch: [155][ 60/ 1207] Overall Loss 0.199830 Objective Loss 0.199830 LR 0.000250 Time 0.032927 -2023-02-13 18:26:25,136 - Epoch: [155][ 70/ 1207] Overall Loss 0.197128 Objective Loss 0.197128 LR 0.000250 Time 0.031009 -2023-02-13 18:26:25,329 - Epoch: [155][ 80/ 1207] Overall Loss 0.196476 Objective Loss 0.196476 LR 0.000250 Time 0.029538 -2023-02-13 18:26:25,525 - Epoch: [155][ 90/ 1207] Overall Loss 0.196719 Objective Loss 0.196719 LR 0.000250 Time 0.028430 -2023-02-13 18:26:25,718 - Epoch: [155][ 100/ 1207] Overall Loss 0.197878 Objective Loss 0.197878 LR 0.000250 Time 0.027515 -2023-02-13 18:26:25,915 - Epoch: [155][ 110/ 1207] Overall Loss 0.192660 Objective Loss 0.192660 LR 0.000250 Time 0.026796 -2023-02-13 18:26:26,108 - Epoch: [155][ 120/ 1207] Overall Loss 0.192765 Objective Loss 0.192765 LR 0.000250 Time 0.026171 -2023-02-13 18:26:26,304 - Epoch: [155][ 130/ 1207] Overall Loss 0.193415 Objective Loss 0.193415 LR 0.000250 Time 0.025660 -2023-02-13 18:26:26,498 - Epoch: [155][ 140/ 1207] Overall Loss 0.193014 Objective Loss 0.193014 LR 0.000250 Time 0.025211 -2023-02-13 18:26:26,693 - Epoch: [155][ 150/ 1207] Overall Loss 0.193438 Objective Loss 0.193438 LR 0.000250 Time 0.024829 -2023-02-13 18:26:26,887 - Epoch: [155][ 160/ 1207] Overall Loss 0.194539 Objective Loss 0.194539 LR 0.000250 Time 0.024484 -2023-02-13 18:26:27,083 - Epoch: [155][ 170/ 1207] Overall Loss 0.193821 Objective Loss 0.193821 LR 0.000250 Time 0.024195 -2023-02-13 18:26:27,276 - Epoch: [155][ 180/ 1207] Overall Loss 0.192800 Objective Loss 0.192800 LR 0.000250 Time 0.023925 -2023-02-13 18:26:27,473 - Epoch: [155][ 190/ 1207] Overall Loss 0.193642 Objective Loss 0.193642 LR 0.000250 Time 0.023696 -2023-02-13 18:26:27,666 - Epoch: [155][ 200/ 1207] Overall Loss 0.193183 Objective Loss 0.193183 LR 0.000250 Time 0.023476 -2023-02-13 18:26:27,862 - Epoch: [155][ 210/ 1207] Overall Loss 0.192872 Objective Loss 0.192872 LR 0.000250 Time 0.023289 -2023-02-13 18:26:28,055 - Epoch: [155][ 220/ 1207] Overall Loss 0.192979 Objective Loss 0.192979 LR 0.000250 Time 0.023106 -2023-02-13 18:26:28,251 - Epoch: [155][ 230/ 1207] Overall Loss 0.192941 Objective Loss 0.192941 LR 0.000250 Time 0.022951 -2023-02-13 18:26:28,444 - Epoch: [155][ 240/ 1207] Overall Loss 0.193452 Objective Loss 0.193452 LR 0.000250 Time 0.022799 -2023-02-13 18:26:28,640 - Epoch: [155][ 250/ 1207] Overall Loss 0.193756 Objective Loss 0.193756 LR 0.000250 Time 0.022668 -2023-02-13 18:26:28,833 - Epoch: [155][ 260/ 1207] Overall Loss 0.193591 Objective Loss 0.193591 LR 0.000250 Time 0.022538 -2023-02-13 18:26:29,028 - Epoch: [155][ 270/ 1207] Overall Loss 0.193448 Objective Loss 0.193448 LR 0.000250 Time 0.022426 -2023-02-13 18:26:29,220 - Epoch: [155][ 280/ 1207] Overall Loss 0.193062 Objective Loss 0.193062 LR 0.000250 Time 0.022310 -2023-02-13 18:26:29,416 - Epoch: [155][ 290/ 1207] Overall Loss 0.192645 Objective Loss 0.192645 LR 0.000250 Time 0.022215 -2023-02-13 18:26:29,610 - Epoch: [155][ 300/ 1207] Overall Loss 0.192834 Objective Loss 0.192834 LR 0.000250 Time 0.022119 -2023-02-13 18:26:29,805 - Epoch: [155][ 310/ 1207] Overall Loss 0.193073 Objective Loss 0.193073 LR 0.000250 Time 0.022034 -2023-02-13 18:26:29,995 - Epoch: [155][ 320/ 1207] Overall Loss 0.192745 Objective Loss 0.192745 LR 0.000250 Time 0.021936 -2023-02-13 18:26:30,184 - Epoch: [155][ 330/ 1207] Overall Loss 0.192559 Objective Loss 0.192559 LR 0.000250 Time 0.021843 -2023-02-13 18:26:30,373 - Epoch: [155][ 340/ 1207] Overall Loss 0.192834 Objective Loss 0.192834 LR 0.000250 Time 0.021755 -2023-02-13 18:26:30,563 - Epoch: [155][ 350/ 1207] Overall Loss 0.192702 Objective Loss 0.192702 LR 0.000250 Time 0.021676 -2023-02-13 18:26:30,752 - Epoch: [155][ 360/ 1207] Overall Loss 0.192793 Objective Loss 0.192793 LR 0.000250 Time 0.021598 -2023-02-13 18:26:30,942 - Epoch: [155][ 370/ 1207] Overall Loss 0.192958 Objective Loss 0.192958 LR 0.000250 Time 0.021527 -2023-02-13 18:26:31,131 - Epoch: [155][ 380/ 1207] Overall Loss 0.193350 Objective Loss 0.193350 LR 0.000250 Time 0.021458 -2023-02-13 18:26:31,321 - Epoch: [155][ 390/ 1207] Overall Loss 0.193502 Objective Loss 0.193502 LR 0.000250 Time 0.021393 -2023-02-13 18:26:31,511 - Epoch: [155][ 400/ 1207] Overall Loss 0.193426 Objective Loss 0.193426 LR 0.000250 Time 0.021332 -2023-02-13 18:26:31,701 - Epoch: [155][ 410/ 1207] Overall Loss 0.193223 Objective Loss 0.193223 LR 0.000250 Time 0.021273 -2023-02-13 18:26:31,890 - Epoch: [155][ 420/ 1207] Overall Loss 0.193411 Objective Loss 0.193411 LR 0.000250 Time 0.021217 -2023-02-13 18:26:32,080 - Epoch: [155][ 430/ 1207] Overall Loss 0.193885 Objective Loss 0.193885 LR 0.000250 Time 0.021165 -2023-02-13 18:26:32,269 - Epoch: [155][ 440/ 1207] Overall Loss 0.194219 Objective Loss 0.194219 LR 0.000250 Time 0.021113 -2023-02-13 18:26:32,459 - Epoch: [155][ 450/ 1207] Overall Loss 0.194483 Objective Loss 0.194483 LR 0.000250 Time 0.021065 -2023-02-13 18:26:32,649 - Epoch: [155][ 460/ 1207] Overall Loss 0.194873 Objective Loss 0.194873 LR 0.000250 Time 0.021018 -2023-02-13 18:26:32,838 - Epoch: [155][ 470/ 1207] Overall Loss 0.194979 Objective Loss 0.194979 LR 0.000250 Time 0.020973 -2023-02-13 18:26:33,028 - Epoch: [155][ 480/ 1207] Overall Loss 0.195394 Objective Loss 0.195394 LR 0.000250 Time 0.020930 -2023-02-13 18:26:33,217 - Epoch: [155][ 490/ 1207] Overall Loss 0.195741 Objective Loss 0.195741 LR 0.000250 Time 0.020889 -2023-02-13 18:26:33,406 - Epoch: [155][ 500/ 1207] Overall Loss 0.195777 Objective Loss 0.195777 LR 0.000250 Time 0.020848 -2023-02-13 18:26:33,596 - Epoch: [155][ 510/ 1207] Overall Loss 0.195870 Objective Loss 0.195870 LR 0.000250 Time 0.020811 -2023-02-13 18:26:33,785 - Epoch: [155][ 520/ 1207] Overall Loss 0.196082 Objective Loss 0.196082 LR 0.000250 Time 0.020774 -2023-02-13 18:26:33,974 - Epoch: [155][ 530/ 1207] Overall Loss 0.195863 Objective Loss 0.195863 LR 0.000250 Time 0.020738 -2023-02-13 18:26:34,163 - Epoch: [155][ 540/ 1207] Overall Loss 0.195857 Objective Loss 0.195857 LR 0.000250 Time 0.020704 -2023-02-13 18:26:34,352 - Epoch: [155][ 550/ 1207] Overall Loss 0.196310 Objective Loss 0.196310 LR 0.000250 Time 0.020670 -2023-02-13 18:26:34,542 - Epoch: [155][ 560/ 1207] Overall Loss 0.196411 Objective Loss 0.196411 LR 0.000250 Time 0.020638 -2023-02-13 18:26:34,731 - Epoch: [155][ 570/ 1207] Overall Loss 0.196357 Objective Loss 0.196357 LR 0.000250 Time 0.020608 -2023-02-13 18:26:34,921 - Epoch: [155][ 580/ 1207] Overall Loss 0.196357 Objective Loss 0.196357 LR 0.000250 Time 0.020579 -2023-02-13 18:26:35,111 - Epoch: [155][ 590/ 1207] Overall Loss 0.196303 Objective Loss 0.196303 LR 0.000250 Time 0.020552 -2023-02-13 18:26:35,300 - Epoch: [155][ 600/ 1207] Overall Loss 0.196530 Objective Loss 0.196530 LR 0.000250 Time 0.020524 -2023-02-13 18:26:35,491 - Epoch: [155][ 610/ 1207] Overall Loss 0.196337 Objective Loss 0.196337 LR 0.000250 Time 0.020500 -2023-02-13 18:26:35,680 - Epoch: [155][ 620/ 1207] Overall Loss 0.196381 Objective Loss 0.196381 LR 0.000250 Time 0.020474 -2023-02-13 18:26:35,869 - Epoch: [155][ 630/ 1207] Overall Loss 0.196583 Objective Loss 0.196583 LR 0.000250 Time 0.020449 -2023-02-13 18:26:36,060 - Epoch: [155][ 640/ 1207] Overall Loss 0.196768 Objective Loss 0.196768 LR 0.000250 Time 0.020426 -2023-02-13 18:26:36,249 - Epoch: [155][ 650/ 1207] Overall Loss 0.196751 Objective Loss 0.196751 LR 0.000250 Time 0.020402 -2023-02-13 18:26:36,438 - Epoch: [155][ 660/ 1207] Overall Loss 0.197027 Objective Loss 0.197027 LR 0.000250 Time 0.020379 -2023-02-13 18:26:36,628 - Epoch: [155][ 670/ 1207] Overall Loss 0.197203 Objective Loss 0.197203 LR 0.000250 Time 0.020358 -2023-02-13 18:26:36,817 - Epoch: [155][ 680/ 1207] Overall Loss 0.197426 Objective Loss 0.197426 LR 0.000250 Time 0.020336 -2023-02-13 18:26:37,007 - Epoch: [155][ 690/ 1207] Overall Loss 0.197417 Objective Loss 0.197417 LR 0.000250 Time 0.020315 -2023-02-13 18:26:37,196 - Epoch: [155][ 700/ 1207] Overall Loss 0.197842 Objective Loss 0.197842 LR 0.000250 Time 0.020294 -2023-02-13 18:26:37,385 - Epoch: [155][ 710/ 1207] Overall Loss 0.197889 Objective Loss 0.197889 LR 0.000250 Time 0.020275 -2023-02-13 18:26:37,575 - Epoch: [155][ 720/ 1207] Overall Loss 0.197799 Objective Loss 0.197799 LR 0.000250 Time 0.020256 -2023-02-13 18:26:37,763 - Epoch: [155][ 730/ 1207] Overall Loss 0.197979 Objective Loss 0.197979 LR 0.000250 Time 0.020236 -2023-02-13 18:26:37,952 - Epoch: [155][ 740/ 1207] Overall Loss 0.197784 Objective Loss 0.197784 LR 0.000250 Time 0.020217 -2023-02-13 18:26:38,141 - Epoch: [155][ 750/ 1207] Overall Loss 0.198123 Objective Loss 0.198123 LR 0.000250 Time 0.020200 -2023-02-13 18:26:38,330 - Epoch: [155][ 760/ 1207] Overall Loss 0.198097 Objective Loss 0.198097 LR 0.000250 Time 0.020181 -2023-02-13 18:26:38,519 - Epoch: [155][ 770/ 1207] Overall Loss 0.197898 Objective Loss 0.197898 LR 0.000250 Time 0.020165 -2023-02-13 18:26:38,707 - Epoch: [155][ 780/ 1207] Overall Loss 0.197876 Objective Loss 0.197876 LR 0.000250 Time 0.020147 -2023-02-13 18:26:38,897 - Epoch: [155][ 790/ 1207] Overall Loss 0.197738 Objective Loss 0.197738 LR 0.000250 Time 0.020131 -2023-02-13 18:26:39,085 - Epoch: [155][ 800/ 1207] Overall Loss 0.197803 Objective Loss 0.197803 LR 0.000250 Time 0.020115 -2023-02-13 18:26:39,274 - Epoch: [155][ 810/ 1207] Overall Loss 0.197520 Objective Loss 0.197520 LR 0.000250 Time 0.020099 -2023-02-13 18:26:39,463 - Epoch: [155][ 820/ 1207] Overall Loss 0.197681 Objective Loss 0.197681 LR 0.000250 Time 0.020084 -2023-02-13 18:26:39,652 - Epoch: [155][ 830/ 1207] Overall Loss 0.197857 Objective Loss 0.197857 LR 0.000250 Time 0.020070 -2023-02-13 18:26:39,841 - Epoch: [155][ 840/ 1207] Overall Loss 0.197993 Objective Loss 0.197993 LR 0.000250 Time 0.020055 -2023-02-13 18:26:40,029 - Epoch: [155][ 850/ 1207] Overall Loss 0.197793 Objective Loss 0.197793 LR 0.000250 Time 0.020040 -2023-02-13 18:26:40,218 - Epoch: [155][ 860/ 1207] Overall Loss 0.197667 Objective Loss 0.197667 LR 0.000250 Time 0.020026 -2023-02-13 18:26:40,407 - Epoch: [155][ 870/ 1207] Overall Loss 0.197850 Objective Loss 0.197850 LR 0.000250 Time 0.020013 -2023-02-13 18:26:40,596 - Epoch: [155][ 880/ 1207] Overall Loss 0.197688 Objective Loss 0.197688 LR 0.000250 Time 0.020000 -2023-02-13 18:26:40,785 - Epoch: [155][ 890/ 1207] Overall Loss 0.197793 Objective Loss 0.197793 LR 0.000250 Time 0.019987 -2023-02-13 18:26:40,978 - Epoch: [155][ 900/ 1207] Overall Loss 0.197870 Objective Loss 0.197870 LR 0.000250 Time 0.019979 -2023-02-13 18:26:41,168 - Epoch: [155][ 910/ 1207] Overall Loss 0.197778 Objective Loss 0.197778 LR 0.000250 Time 0.019968 -2023-02-13 18:26:41,357 - Epoch: [155][ 920/ 1207] Overall Loss 0.197845 Objective Loss 0.197845 LR 0.000250 Time 0.019956 -2023-02-13 18:26:41,549 - Epoch: [155][ 930/ 1207] Overall Loss 0.197959 Objective Loss 0.197959 LR 0.000250 Time 0.019947 -2023-02-13 18:26:41,740 - Epoch: [155][ 940/ 1207] Overall Loss 0.198160 Objective Loss 0.198160 LR 0.000250 Time 0.019938 -2023-02-13 18:26:41,932 - Epoch: [155][ 950/ 1207] Overall Loss 0.198174 Objective Loss 0.198174 LR 0.000250 Time 0.019930 -2023-02-13 18:26:42,124 - Epoch: [155][ 960/ 1207] Overall Loss 0.198217 Objective Loss 0.198217 LR 0.000250 Time 0.019921 -2023-02-13 18:26:42,316 - Epoch: [155][ 970/ 1207] Overall Loss 0.198321 Objective Loss 0.198321 LR 0.000250 Time 0.019913 -2023-02-13 18:26:42,508 - Epoch: [155][ 980/ 1207] Overall Loss 0.198334 Objective Loss 0.198334 LR 0.000250 Time 0.019906 -2023-02-13 18:26:42,699 - Epoch: [155][ 990/ 1207] Overall Loss 0.198167 Objective Loss 0.198167 LR 0.000250 Time 0.019898 -2023-02-13 18:26:42,891 - Epoch: [155][ 1000/ 1207] Overall Loss 0.198310 Objective Loss 0.198310 LR 0.000250 Time 0.019890 -2023-02-13 18:26:43,081 - Epoch: [155][ 1010/ 1207] Overall Loss 0.198373 Objective Loss 0.198373 LR 0.000250 Time 0.019881 -2023-02-13 18:26:43,272 - Epoch: [155][ 1020/ 1207] Overall Loss 0.198064 Objective Loss 0.198064 LR 0.000250 Time 0.019873 -2023-02-13 18:26:43,465 - Epoch: [155][ 1030/ 1207] Overall Loss 0.198252 Objective Loss 0.198252 LR 0.000250 Time 0.019867 -2023-02-13 18:26:43,657 - Epoch: [155][ 1040/ 1207] Overall Loss 0.198364 Objective Loss 0.198364 LR 0.000250 Time 0.019861 -2023-02-13 18:26:43,849 - Epoch: [155][ 1050/ 1207] Overall Loss 0.198545 Objective Loss 0.198545 LR 0.000250 Time 0.019854 -2023-02-13 18:26:44,041 - Epoch: [155][ 1060/ 1207] Overall Loss 0.198236 Objective Loss 0.198236 LR 0.000250 Time 0.019847 -2023-02-13 18:26:44,232 - Epoch: [155][ 1070/ 1207] Overall Loss 0.197956 Objective Loss 0.197956 LR 0.000250 Time 0.019840 -2023-02-13 18:26:44,424 - Epoch: [155][ 1080/ 1207] Overall Loss 0.197922 Objective Loss 0.197922 LR 0.000250 Time 0.019834 -2023-02-13 18:26:44,617 - Epoch: [155][ 1090/ 1207] Overall Loss 0.198003 Objective Loss 0.198003 LR 0.000250 Time 0.019828 -2023-02-13 18:26:44,808 - Epoch: [155][ 1100/ 1207] Overall Loss 0.198009 Objective Loss 0.198009 LR 0.000250 Time 0.019821 -2023-02-13 18:26:44,999 - Epoch: [155][ 1110/ 1207] Overall Loss 0.198155 Objective Loss 0.198155 LR 0.000250 Time 0.019814 -2023-02-13 18:26:45,190 - Epoch: [155][ 1120/ 1207] Overall Loss 0.198165 Objective Loss 0.198165 LR 0.000250 Time 0.019808 -2023-02-13 18:26:45,381 - Epoch: [155][ 1130/ 1207] Overall Loss 0.198008 Objective Loss 0.198008 LR 0.000250 Time 0.019801 -2023-02-13 18:26:45,574 - Epoch: [155][ 1140/ 1207] Overall Loss 0.197830 Objective Loss 0.197830 LR 0.000250 Time 0.019797 -2023-02-13 18:26:45,766 - Epoch: [155][ 1150/ 1207] Overall Loss 0.197798 Objective Loss 0.197798 LR 0.000250 Time 0.019791 -2023-02-13 18:26:45,962 - Epoch: [155][ 1160/ 1207] Overall Loss 0.197975 Objective Loss 0.197975 LR 0.000250 Time 0.019789 -2023-02-13 18:26:46,154 - Epoch: [155][ 1170/ 1207] Overall Loss 0.197843 Objective Loss 0.197843 LR 0.000250 Time 0.019783 -2023-02-13 18:26:46,346 - Epoch: [155][ 1180/ 1207] Overall Loss 0.197701 Objective Loss 0.197701 LR 0.000250 Time 0.019778 -2023-02-13 18:26:46,539 - Epoch: [155][ 1190/ 1207] Overall Loss 0.197655 Objective Loss 0.197655 LR 0.000250 Time 0.019774 -2023-02-13 18:26:46,787 - Epoch: [155][ 1200/ 1207] Overall Loss 0.197581 Objective Loss 0.197581 LR 0.000250 Time 0.019816 -2023-02-13 18:26:46,903 - Epoch: [155][ 1207/ 1207] Overall Loss 0.197648 Objective Loss 0.197648 Top1 89.024390 Top5 98.475610 LR 0.000250 Time 0.019797 -2023-02-13 18:26:46,986 - --- validate (epoch=155)----------- -2023-02-13 18:26:46,986 - 34311 samples (256 per mini-batch) -2023-02-13 18:26:47,391 - Epoch: [155][ 10/ 135] Loss 0.321855 Top1 84.179688 Top5 98.320312 -2023-02-13 18:26:47,524 - Epoch: [155][ 20/ 135] Loss 0.293451 Top1 84.726562 Top5 98.339844 -2023-02-13 18:26:47,670 - Epoch: [155][ 30/ 135] Loss 0.282358 Top1 85.377604 Top5 98.281250 -2023-02-13 18:26:47,807 - Epoch: [155][ 40/ 135] Loss 0.276786 Top1 85.546875 Top5 98.222656 -2023-02-13 18:26:47,952 - Epoch: [155][ 50/ 135] Loss 0.288152 Top1 85.304688 Top5 98.210938 -2023-02-13 18:26:48,089 - Epoch: [155][ 60/ 135] Loss 0.286317 Top1 85.325521 Top5 98.118490 -2023-02-13 18:26:48,233 - Epoch: [155][ 70/ 135] Loss 0.286346 Top1 85.424107 Top5 98.113839 -2023-02-13 18:26:48,370 - Epoch: [155][ 80/ 135] Loss 0.290342 Top1 85.283203 Top5 98.071289 -2023-02-13 18:26:48,516 - Epoch: [155][ 90/ 135] Loss 0.289858 Top1 85.325521 Top5 98.033854 -2023-02-13 18:26:48,654 - Epoch: [155][ 100/ 135] Loss 0.288985 Top1 85.277344 Top5 98.007812 -2023-02-13 18:26:48,797 - Epoch: [155][ 110/ 135] Loss 0.293837 Top1 85.213068 Top5 97.968750 -2023-02-13 18:26:48,935 - Epoch: [155][ 120/ 135] Loss 0.294098 Top1 85.227865 Top5 97.965495 -2023-02-13 18:26:49,076 - Epoch: [155][ 130/ 135] Loss 0.292256 Top1 85.228365 Top5 97.968750 -2023-02-13 18:26:49,120 - Epoch: [155][ 135/ 135] Loss 0.289065 Top1 85.240885 Top5 97.977325 -2023-02-13 18:26:49,192 - ==> Top1: 85.241 Top5: 97.977 Loss: 0.289 - -2023-02-13 18:26:49,193 - ==> Confusion: -[[ 837 4 4 0 8 3 0 0 7 66 1 4 0 5 7 2 2 3 3 3 8] - [ 3 952 2 2 5 27 1 17 3 1 1 2 0 0 1 1 5 1 2 1 6] - [ 6 8 957 10 4 1 18 13 0 1 2 2 4 5 5 5 2 1 4 2 8] - [ 3 1 16 918 2 5 0 3 4 0 8 1 6 0 19 1 3 6 13 0 7] - [ 10 10 0 0 990 12 1 2 3 1 0 7 2 4 8 5 5 0 0 2 4] - [ 1 13 0 2 6 980 2 22 0 3 1 11 2 14 0 3 1 1 1 2 5] - [ 1 6 16 1 1 7 1033 8 0 3 1 2 1 1 1 3 1 5 1 4 3] - [ 0 5 9 1 4 26 3 945 1 2 0 5 3 0 0 0 0 1 8 8 3] - [ 12 3 1 1 2 0 1 1 922 29 6 3 0 10 11 2 1 0 4 0 0] - [ 47 1 3 0 6 1 0 4 45 881 0 0 0 13 4 2 0 0 1 0 4] - [ 1 2 4 8 1 2 3 4 16 0 985 3 1 7 3 1 0 0 7 0 3] - [ 0 2 0 0 4 12 0 5 1 1 1 935 18 4 1 3 1 6 2 6 3] - [ 0 0 0 5 2 2 0 0 3 1 0 33 880 1 1 8 1 11 2 0 9] - [ 6 3 1 0 8 8 0 1 9 13 6 5 4 946 2 4 3 0 1 1 3] - [ 9 1 1 15 3 5 0 1 21 8 1 0 0 1 1003 0 3 6 6 0 8] - [ 6 4 7 1 6 2 2 0 0 0 0 9 6 3 0 973 9 6 0 6 6] - [ 3 9 1 1 8 4 0 0 3 0 0 2 3 2 0 7 1000 2 0 3 13] - [ 3 2 0 4 1 0 2 1 1 1 1 14 10 2 0 13 1 990 1 0 4] - [ 2 3 4 12 1 3 0 29 6 0 5 2 3 0 13 1 1 3 995 1 2] - [ 1 2 1 0 1 7 3 6 1 0 2 20 1 3 0 7 3 3 1 1077 9] - [ 119 248 218 109 119 216 80 202 97 86 141 144 288 313 154 94 250 93 178 237 10048]] - -2023-02-13 18:26:49,194 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:26:49,194 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:26:49,200 - - -2023-02-13 18:26:49,200 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:26:50,100 - Epoch: [156][ 10/ 1207] Overall Loss 0.205955 Objective Loss 0.205955 LR 0.000250 Time 0.089916 -2023-02-13 18:26:50,297 - Epoch: [156][ 20/ 1207] Overall Loss 0.203022 Objective Loss 0.203022 LR 0.000250 Time 0.054785 -2023-02-13 18:26:50,488 - Epoch: [156][ 30/ 1207] Overall Loss 0.208161 Objective Loss 0.208161 LR 0.000250 Time 0.042871 -2023-02-13 18:26:50,678 - Epoch: [156][ 40/ 1207] Overall Loss 0.200978 Objective Loss 0.200978 LR 0.000250 Time 0.036889 -2023-02-13 18:26:50,867 - Epoch: [156][ 50/ 1207] Overall Loss 0.204588 Objective Loss 0.204588 LR 0.000250 Time 0.033288 -2023-02-13 18:26:51,058 - Epoch: [156][ 60/ 1207] Overall Loss 0.203781 Objective Loss 0.203781 LR 0.000250 Time 0.030924 -2023-02-13 18:26:51,248 - Epoch: [156][ 70/ 1207] Overall Loss 0.201570 Objective Loss 0.201570 LR 0.000250 Time 0.029215 -2023-02-13 18:26:51,439 - Epoch: [156][ 80/ 1207] Overall Loss 0.200108 Objective Loss 0.200108 LR 0.000250 Time 0.027937 -2023-02-13 18:26:51,630 - Epoch: [156][ 90/ 1207] Overall Loss 0.199070 Objective Loss 0.199070 LR 0.000250 Time 0.026949 -2023-02-13 18:26:51,819 - Epoch: [156][ 100/ 1207] Overall Loss 0.199508 Objective Loss 0.199508 LR 0.000250 Time 0.026148 -2023-02-13 18:26:52,009 - Epoch: [156][ 110/ 1207] Overall Loss 0.197565 Objective Loss 0.197565 LR 0.000250 Time 0.025493 -2023-02-13 18:26:52,199 - Epoch: [156][ 120/ 1207] Overall Loss 0.196971 Objective Loss 0.196971 LR 0.000250 Time 0.024947 -2023-02-13 18:26:52,388 - Epoch: [156][ 130/ 1207] Overall Loss 0.198540 Objective Loss 0.198540 LR 0.000250 Time 0.024482 -2023-02-13 18:26:52,579 - Epoch: [156][ 140/ 1207] Overall Loss 0.198307 Objective Loss 0.198307 LR 0.000250 Time 0.024094 -2023-02-13 18:26:52,769 - Epoch: [156][ 150/ 1207] Overall Loss 0.197182 Objective Loss 0.197182 LR 0.000250 Time 0.023752 -2023-02-13 18:26:52,959 - Epoch: [156][ 160/ 1207] Overall Loss 0.197113 Objective Loss 0.197113 LR 0.000250 Time 0.023452 -2023-02-13 18:26:53,149 - Epoch: [156][ 170/ 1207] Overall Loss 0.196073 Objective Loss 0.196073 LR 0.000250 Time 0.023188 -2023-02-13 18:26:53,339 - Epoch: [156][ 180/ 1207] Overall Loss 0.195510 Objective Loss 0.195510 LR 0.000250 Time 0.022953 -2023-02-13 18:26:53,530 - Epoch: [156][ 190/ 1207] Overall Loss 0.196481 Objective Loss 0.196481 LR 0.000250 Time 0.022748 -2023-02-13 18:26:53,721 - Epoch: [156][ 200/ 1207] Overall Loss 0.196462 Objective Loss 0.196462 LR 0.000250 Time 0.022562 -2023-02-13 18:26:53,910 - Epoch: [156][ 210/ 1207] Overall Loss 0.196699 Objective Loss 0.196699 LR 0.000250 Time 0.022388 -2023-02-13 18:26:54,100 - Epoch: [156][ 220/ 1207] Overall Loss 0.196301 Objective Loss 0.196301 LR 0.000250 Time 0.022232 -2023-02-13 18:26:54,290 - Epoch: [156][ 230/ 1207] Overall Loss 0.196662 Objective Loss 0.196662 LR 0.000250 Time 0.022091 -2023-02-13 18:26:54,481 - Epoch: [156][ 240/ 1207] Overall Loss 0.196886 Objective Loss 0.196886 LR 0.000250 Time 0.021963 -2023-02-13 18:26:54,671 - Epoch: [156][ 250/ 1207] Overall Loss 0.196502 Objective Loss 0.196502 LR 0.000250 Time 0.021844 -2023-02-13 18:26:54,862 - Epoch: [156][ 260/ 1207] Overall Loss 0.196482 Objective Loss 0.196482 LR 0.000250 Time 0.021736 -2023-02-13 18:26:55,052 - Epoch: [156][ 270/ 1207] Overall Loss 0.196232 Objective Loss 0.196232 LR 0.000250 Time 0.021633 -2023-02-13 18:26:55,242 - Epoch: [156][ 280/ 1207] Overall Loss 0.196056 Objective Loss 0.196056 LR 0.000250 Time 0.021539 -2023-02-13 18:26:55,432 - Epoch: [156][ 290/ 1207] Overall Loss 0.195189 Objective Loss 0.195189 LR 0.000250 Time 0.021451 -2023-02-13 18:26:55,623 - Epoch: [156][ 300/ 1207] Overall Loss 0.195101 Objective Loss 0.195101 LR 0.000250 Time 0.021371 -2023-02-13 18:26:55,813 - Epoch: [156][ 310/ 1207] Overall Loss 0.195037 Objective Loss 0.195037 LR 0.000250 Time 0.021294 -2023-02-13 18:26:56,005 - Epoch: [156][ 320/ 1207] Overall Loss 0.196035 Objective Loss 0.196035 LR 0.000250 Time 0.021225 -2023-02-13 18:26:56,195 - Epoch: [156][ 330/ 1207] Overall Loss 0.195703 Objective Loss 0.195703 LR 0.000250 Time 0.021157 -2023-02-13 18:26:56,385 - Epoch: [156][ 340/ 1207] Overall Loss 0.195668 Objective Loss 0.195668 LR 0.000250 Time 0.021093 -2023-02-13 18:26:56,576 - Epoch: [156][ 350/ 1207] Overall Loss 0.196129 Objective Loss 0.196129 LR 0.000250 Time 0.021034 -2023-02-13 18:26:56,766 - Epoch: [156][ 360/ 1207] Overall Loss 0.196816 Objective Loss 0.196816 LR 0.000250 Time 0.020977 -2023-02-13 18:26:56,957 - Epoch: [156][ 370/ 1207] Overall Loss 0.196523 Objective Loss 0.196523 LR 0.000250 Time 0.020924 -2023-02-13 18:26:57,147 - Epoch: [156][ 380/ 1207] Overall Loss 0.196536 Objective Loss 0.196536 LR 0.000250 Time 0.020873 -2023-02-13 18:26:57,338 - Epoch: [156][ 390/ 1207] Overall Loss 0.196844 Objective Loss 0.196844 LR 0.000250 Time 0.020826 -2023-02-13 18:26:57,529 - Epoch: [156][ 400/ 1207] Overall Loss 0.196698 Objective Loss 0.196698 LR 0.000250 Time 0.020782 -2023-02-13 18:26:57,719 - Epoch: [156][ 410/ 1207] Overall Loss 0.196099 Objective Loss 0.196099 LR 0.000250 Time 0.020739 -2023-02-13 18:26:57,909 - Epoch: [156][ 420/ 1207] Overall Loss 0.196043 Objective Loss 0.196043 LR 0.000250 Time 0.020697 -2023-02-13 18:26:58,099 - Epoch: [156][ 430/ 1207] Overall Loss 0.196144 Objective Loss 0.196144 LR 0.000250 Time 0.020657 -2023-02-13 18:26:58,290 - Epoch: [156][ 440/ 1207] Overall Loss 0.195780 Objective Loss 0.195780 LR 0.000250 Time 0.020620 -2023-02-13 18:26:58,480 - Epoch: [156][ 450/ 1207] Overall Loss 0.195659 Objective Loss 0.195659 LR 0.000250 Time 0.020582 -2023-02-13 18:26:58,671 - Epoch: [156][ 460/ 1207] Overall Loss 0.195644 Objective Loss 0.195644 LR 0.000250 Time 0.020550 -2023-02-13 18:26:58,861 - Epoch: [156][ 470/ 1207] Overall Loss 0.195650 Objective Loss 0.195650 LR 0.000250 Time 0.020516 -2023-02-13 18:26:59,052 - Epoch: [156][ 480/ 1207] Overall Loss 0.195828 Objective Loss 0.195828 LR 0.000250 Time 0.020486 -2023-02-13 18:26:59,242 - Epoch: [156][ 490/ 1207] Overall Loss 0.196366 Objective Loss 0.196366 LR 0.000250 Time 0.020454 -2023-02-13 18:26:59,433 - Epoch: [156][ 500/ 1207] Overall Loss 0.196221 Objective Loss 0.196221 LR 0.000250 Time 0.020426 -2023-02-13 18:26:59,624 - Epoch: [156][ 510/ 1207] Overall Loss 0.196380 Objective Loss 0.196380 LR 0.000250 Time 0.020400 -2023-02-13 18:26:59,814 - Epoch: [156][ 520/ 1207] Overall Loss 0.196617 Objective Loss 0.196617 LR 0.000250 Time 0.020373 -2023-02-13 18:27:00,005 - Epoch: [156][ 530/ 1207] Overall Loss 0.196756 Objective Loss 0.196756 LR 0.000250 Time 0.020347 -2023-02-13 18:27:00,195 - Epoch: [156][ 540/ 1207] Overall Loss 0.197013 Objective Loss 0.197013 LR 0.000250 Time 0.020322 -2023-02-13 18:27:00,385 - Epoch: [156][ 550/ 1207] Overall Loss 0.196730 Objective Loss 0.196730 LR 0.000250 Time 0.020298 -2023-02-13 18:27:00,576 - Epoch: [156][ 560/ 1207] Overall Loss 0.196489 Objective Loss 0.196489 LR 0.000250 Time 0.020275 -2023-02-13 18:27:00,766 - Epoch: [156][ 570/ 1207] Overall Loss 0.196339 Objective Loss 0.196339 LR 0.000250 Time 0.020252 -2023-02-13 18:27:00,957 - Epoch: [156][ 580/ 1207] Overall Loss 0.196565 Objective Loss 0.196565 LR 0.000250 Time 0.020232 -2023-02-13 18:27:01,148 - Epoch: [156][ 590/ 1207] Overall Loss 0.196360 Objective Loss 0.196360 LR 0.000250 Time 0.020212 -2023-02-13 18:27:01,338 - Epoch: [156][ 600/ 1207] Overall Loss 0.196572 Objective Loss 0.196572 LR 0.000250 Time 0.020192 -2023-02-13 18:27:01,529 - Epoch: [156][ 610/ 1207] Overall Loss 0.196356 Objective Loss 0.196356 LR 0.000250 Time 0.020173 -2023-02-13 18:27:01,720 - Epoch: [156][ 620/ 1207] Overall Loss 0.196496 Objective Loss 0.196496 LR 0.000250 Time 0.020155 -2023-02-13 18:27:01,911 - Epoch: [156][ 630/ 1207] Overall Loss 0.196155 Objective Loss 0.196155 LR 0.000250 Time 0.020137 -2023-02-13 18:27:02,102 - Epoch: [156][ 640/ 1207] Overall Loss 0.196416 Objective Loss 0.196416 LR 0.000250 Time 0.020120 -2023-02-13 18:27:02,293 - Epoch: [156][ 650/ 1207] Overall Loss 0.197009 Objective Loss 0.197009 LR 0.000250 Time 0.020104 -2023-02-13 18:27:02,484 - Epoch: [156][ 660/ 1207] Overall Loss 0.197199 Objective Loss 0.197199 LR 0.000250 Time 0.020088 -2023-02-13 18:27:02,676 - Epoch: [156][ 670/ 1207] Overall Loss 0.197168 Objective Loss 0.197168 LR 0.000250 Time 0.020075 -2023-02-13 18:27:02,868 - Epoch: [156][ 680/ 1207] Overall Loss 0.197094 Objective Loss 0.197094 LR 0.000250 Time 0.020061 -2023-02-13 18:27:03,059 - Epoch: [156][ 690/ 1207] Overall Loss 0.196963 Objective Loss 0.196963 LR 0.000250 Time 0.020047 -2023-02-13 18:27:03,251 - Epoch: [156][ 700/ 1207] Overall Loss 0.197098 Objective Loss 0.197098 LR 0.000250 Time 0.020034 -2023-02-13 18:27:03,443 - Epoch: [156][ 710/ 1207] Overall Loss 0.197391 Objective Loss 0.197391 LR 0.000250 Time 0.020022 -2023-02-13 18:27:03,635 - Epoch: [156][ 720/ 1207] Overall Loss 0.197401 Objective Loss 0.197401 LR 0.000250 Time 0.020011 -2023-02-13 18:27:03,827 - Epoch: [156][ 730/ 1207] Overall Loss 0.197707 Objective Loss 0.197707 LR 0.000250 Time 0.019999 -2023-02-13 18:27:04,018 - Epoch: [156][ 740/ 1207] Overall Loss 0.197547 Objective Loss 0.197547 LR 0.000250 Time 0.019986 -2023-02-13 18:27:04,210 - Epoch: [156][ 750/ 1207] Overall Loss 0.197548 Objective Loss 0.197548 LR 0.000250 Time 0.019975 -2023-02-13 18:27:04,402 - Epoch: [156][ 760/ 1207] Overall Loss 0.197753 Objective Loss 0.197753 LR 0.000250 Time 0.019964 -2023-02-13 18:27:04,594 - Epoch: [156][ 770/ 1207] Overall Loss 0.197948 Objective Loss 0.197948 LR 0.000250 Time 0.019954 -2023-02-13 18:27:04,786 - Epoch: [156][ 780/ 1207] Overall Loss 0.198160 Objective Loss 0.198160 LR 0.000250 Time 0.019943 -2023-02-13 18:27:04,978 - Epoch: [156][ 790/ 1207] Overall Loss 0.197609 Objective Loss 0.197609 LR 0.000250 Time 0.019934 -2023-02-13 18:27:05,169 - Epoch: [156][ 800/ 1207] Overall Loss 0.197541 Objective Loss 0.197541 LR 0.000250 Time 0.019924 -2023-02-13 18:27:05,361 - Epoch: [156][ 810/ 1207] Overall Loss 0.197500 Objective Loss 0.197500 LR 0.000250 Time 0.019914 -2023-02-13 18:27:05,553 - Epoch: [156][ 820/ 1207] Overall Loss 0.197449 Objective Loss 0.197449 LR 0.000250 Time 0.019905 -2023-02-13 18:27:05,745 - Epoch: [156][ 830/ 1207] Overall Loss 0.197877 Objective Loss 0.197877 LR 0.000250 Time 0.019896 -2023-02-13 18:27:05,938 - Epoch: [156][ 840/ 1207] Overall Loss 0.197787 Objective Loss 0.197787 LR 0.000250 Time 0.019887 -2023-02-13 18:27:06,130 - Epoch: [156][ 850/ 1207] Overall Loss 0.198066 Objective Loss 0.198066 LR 0.000250 Time 0.019879 -2023-02-13 18:27:06,321 - Epoch: [156][ 860/ 1207] Overall Loss 0.198389 Objective Loss 0.198389 LR 0.000250 Time 0.019870 -2023-02-13 18:27:06,513 - Epoch: [156][ 870/ 1207] Overall Loss 0.198205 Objective Loss 0.198205 LR 0.000250 Time 0.019862 -2023-02-13 18:27:06,705 - Epoch: [156][ 880/ 1207] Overall Loss 0.198134 Objective Loss 0.198134 LR 0.000250 Time 0.019854 -2023-02-13 18:27:06,897 - Epoch: [156][ 890/ 1207] Overall Loss 0.198054 Objective Loss 0.198054 LR 0.000250 Time 0.019846 -2023-02-13 18:27:07,089 - Epoch: [156][ 900/ 1207] Overall Loss 0.197795 Objective Loss 0.197795 LR 0.000250 Time 0.019838 -2023-02-13 18:27:07,280 - Epoch: [156][ 910/ 1207] Overall Loss 0.197498 Objective Loss 0.197498 LR 0.000250 Time 0.019830 -2023-02-13 18:27:07,472 - Epoch: [156][ 920/ 1207] Overall Loss 0.197293 Objective Loss 0.197293 LR 0.000250 Time 0.019822 -2023-02-13 18:27:07,664 - Epoch: [156][ 930/ 1207] Overall Loss 0.197214 Objective Loss 0.197214 LR 0.000250 Time 0.019816 -2023-02-13 18:27:07,856 - Epoch: [156][ 940/ 1207] Overall Loss 0.197284 Objective Loss 0.197284 LR 0.000250 Time 0.019809 -2023-02-13 18:27:08,048 - Epoch: [156][ 950/ 1207] Overall Loss 0.197421 Objective Loss 0.197421 LR 0.000250 Time 0.019802 -2023-02-13 18:27:08,239 - Epoch: [156][ 960/ 1207] Overall Loss 0.197190 Objective Loss 0.197190 LR 0.000250 Time 0.019795 -2023-02-13 18:27:08,432 - Epoch: [156][ 970/ 1207] Overall Loss 0.197113 Objective Loss 0.197113 LR 0.000250 Time 0.019788 -2023-02-13 18:27:08,624 - Epoch: [156][ 980/ 1207] Overall Loss 0.196994 Objective Loss 0.196994 LR 0.000250 Time 0.019782 -2023-02-13 18:27:08,815 - Epoch: [156][ 990/ 1207] Overall Loss 0.196915 Objective Loss 0.196915 LR 0.000250 Time 0.019775 -2023-02-13 18:27:09,006 - Epoch: [156][ 1000/ 1207] Overall Loss 0.196863 Objective Loss 0.196863 LR 0.000250 Time 0.019768 -2023-02-13 18:27:09,197 - Epoch: [156][ 1010/ 1207] Overall Loss 0.196958 Objective Loss 0.196958 LR 0.000250 Time 0.019761 -2023-02-13 18:27:09,389 - Epoch: [156][ 1020/ 1207] Overall Loss 0.196836 Objective Loss 0.196836 LR 0.000250 Time 0.019755 -2023-02-13 18:27:09,582 - Epoch: [156][ 1030/ 1207] Overall Loss 0.196811 Objective Loss 0.196811 LR 0.000250 Time 0.019751 -2023-02-13 18:27:09,774 - Epoch: [156][ 1040/ 1207] Overall Loss 0.197024 Objective Loss 0.197024 LR 0.000250 Time 0.019745 -2023-02-13 18:27:09,966 - Epoch: [156][ 1050/ 1207] Overall Loss 0.197119 Objective Loss 0.197119 LR 0.000250 Time 0.019739 -2023-02-13 18:27:10,158 - Epoch: [156][ 1060/ 1207] Overall Loss 0.197151 Objective Loss 0.197151 LR 0.000250 Time 0.019734 -2023-02-13 18:27:10,349 - Epoch: [156][ 1070/ 1207] Overall Loss 0.197407 Objective Loss 0.197407 LR 0.000250 Time 0.019728 -2023-02-13 18:27:10,542 - Epoch: [156][ 1080/ 1207] Overall Loss 0.197432 Objective Loss 0.197432 LR 0.000250 Time 0.019723 -2023-02-13 18:27:10,734 - Epoch: [156][ 1090/ 1207] Overall Loss 0.197415 Objective Loss 0.197415 LR 0.000250 Time 0.019718 -2023-02-13 18:27:10,926 - Epoch: [156][ 1100/ 1207] Overall Loss 0.197280 Objective Loss 0.197280 LR 0.000250 Time 0.019713 -2023-02-13 18:27:11,119 - Epoch: [156][ 1110/ 1207] Overall Loss 0.197529 Objective Loss 0.197529 LR 0.000250 Time 0.019709 -2023-02-13 18:27:11,310 - Epoch: [156][ 1120/ 1207] Overall Loss 0.197433 Objective Loss 0.197433 LR 0.000250 Time 0.019703 -2023-02-13 18:27:11,503 - Epoch: [156][ 1130/ 1207] Overall Loss 0.197369 Objective Loss 0.197369 LR 0.000250 Time 0.019699 -2023-02-13 18:27:11,695 - Epoch: [156][ 1140/ 1207] Overall Loss 0.197350 Objective Loss 0.197350 LR 0.000250 Time 0.019695 -2023-02-13 18:27:11,887 - Epoch: [156][ 1150/ 1207] Overall Loss 0.197441 Objective Loss 0.197441 LR 0.000250 Time 0.019690 -2023-02-13 18:27:12,079 - Epoch: [156][ 1160/ 1207] Overall Loss 0.197550 Objective Loss 0.197550 LR 0.000250 Time 0.019685 -2023-02-13 18:27:12,270 - Epoch: [156][ 1170/ 1207] Overall Loss 0.197409 Objective Loss 0.197409 LR 0.000250 Time 0.019680 -2023-02-13 18:27:12,462 - Epoch: [156][ 1180/ 1207] Overall Loss 0.197365 Objective Loss 0.197365 LR 0.000250 Time 0.019676 -2023-02-13 18:27:12,655 - Epoch: [156][ 1190/ 1207] Overall Loss 0.197442 Objective Loss 0.197442 LR 0.000250 Time 0.019672 -2023-02-13 18:27:12,902 - Epoch: [156][ 1200/ 1207] Overall Loss 0.197620 Objective Loss 0.197620 LR 0.000250 Time 0.019714 -2023-02-13 18:27:13,017 - Epoch: [156][ 1207/ 1207] Overall Loss 0.197776 Objective Loss 0.197776 Top1 87.195122 Top5 97.865854 LR 0.000250 Time 0.019694 -2023-02-13 18:27:13,092 - --- validate (epoch=156)----------- -2023-02-13 18:27:13,092 - 34311 samples (256 per mini-batch) -2023-02-13 18:27:13,486 - Epoch: [156][ 10/ 135] Loss 0.278394 Top1 86.328125 Top5 98.164062 -2023-02-13 18:27:13,614 - Epoch: [156][ 20/ 135] Loss 0.279538 Top1 85.839844 Top5 98.007812 -2023-02-13 18:27:13,751 - Epoch: [156][ 30/ 135] Loss 0.282948 Top1 86.145833 Top5 98.033854 -2023-02-13 18:27:13,895 - Epoch: [156][ 40/ 135] Loss 0.288105 Top1 86.171875 Top5 97.871094 -2023-02-13 18:27:14,032 - Epoch: [156][ 50/ 135] Loss 0.293656 Top1 85.960938 Top5 97.867188 -2023-02-13 18:27:14,176 - Epoch: [156][ 60/ 135] Loss 0.291896 Top1 85.872396 Top5 97.877604 -2023-02-13 18:27:14,316 - Epoch: [156][ 70/ 135] Loss 0.290607 Top1 85.954241 Top5 97.918527 -2023-02-13 18:27:14,453 - Epoch: [156][ 80/ 135] Loss 0.289917 Top1 85.913086 Top5 97.929688 -2023-02-13 18:27:14,591 - Epoch: [156][ 90/ 135] Loss 0.293238 Top1 85.729167 Top5 97.903646 -2023-02-13 18:27:14,718 - Epoch: [156][ 100/ 135] Loss 0.291467 Top1 85.808594 Top5 97.894531 -2023-02-13 18:27:14,844 - Epoch: [156][ 110/ 135] Loss 0.291065 Top1 85.809659 Top5 97.926136 -2023-02-13 18:27:14,966 - Epoch: [156][ 120/ 135] Loss 0.291330 Top1 85.764974 Top5 97.958984 -2023-02-13 18:27:15,092 - Epoch: [156][ 130/ 135] Loss 0.295102 Top1 85.667067 Top5 97.944712 -2023-02-13 18:27:15,137 - Epoch: [156][ 135/ 135] Loss 0.292617 Top1 85.646003 Top5 97.927778 -2023-02-13 18:27:15,205 - ==> Top1: 85.646 Top5: 97.928 Loss: 0.293 - -2023-02-13 18:27:15,206 - ==> Confusion: -[[ 869 6 6 1 6 3 0 0 6 39 1 3 0 4 8 2 2 3 3 1 4] - [ 2 965 1 1 6 16 1 14 2 2 4 1 1 0 0 1 6 1 3 1 5] - [ 3 7 961 12 5 2 14 13 2 1 3 1 3 5 5 3 3 2 4 1 8] - [ 7 0 25 907 3 2 1 1 3 3 10 0 6 0 17 1 5 5 12 0 8] - [ 10 11 0 0 995 9 1 0 0 0 0 6 0 4 13 4 5 1 0 1 6] - [ 1 20 0 5 7 956 3 18 1 6 1 10 3 17 1 2 5 0 2 7 5] - [ 2 2 10 1 1 8 1044 3 0 2 5 1 2 1 1 2 1 2 1 6 4] - [ 1 9 10 2 6 29 3 925 0 3 3 4 1 3 0 1 1 1 12 5 5] - [ 14 3 1 1 2 0 0 1 913 29 7 2 1 13 14 1 1 0 5 0 1] - [ 78 1 4 0 10 1 0 4 26 857 0 1 4 14 5 1 0 2 1 0 3] - [ 3 2 3 6 1 2 1 5 16 0 985 2 1 11 3 0 1 1 3 0 5] - [ 2 2 2 0 1 13 0 5 4 2 0 930 17 6 0 4 2 9 2 2 2] - [ 0 0 2 7 4 3 0 0 2 3 0 29 878 2 0 6 2 13 1 0 7] - [ 2 3 1 0 9 9 0 1 7 16 9 6 1 945 2 4 1 4 0 0 4] - [ 7 2 0 18 2 3 0 2 28 5 4 1 1 4 990 0 1 6 6 1 11] - [ 4 1 10 3 8 1 4 1 1 0 0 7 6 3 0 965 8 11 0 6 7] - [ 1 5 0 1 11 2 0 1 2 1 1 1 3 2 1 9 1000 1 2 4 13] - [ 2 2 0 5 1 2 2 1 0 0 1 11 8 1 1 12 0 995 1 1 5] - [ 6 6 5 8 0 1 0 21 5 0 6 0 5 0 9 1 2 3 1002 2 4] - [ 1 4 1 0 0 5 1 7 0 0 0 21 3 8 0 9 3 3 0 1072 10] - [ 133 258 221 98 144 192 72 171 99 76 177 121 280 310 126 73 203 95 153 200 10232]] - -2023-02-13 18:27:15,207 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:27:15,207 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:27:15,213 - - -2023-02-13 18:27:15,213 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:27:16,100 - Epoch: [157][ 10/ 1207] Overall Loss 0.207624 Objective Loss 0.207624 LR 0.000250 Time 0.088561 -2023-02-13 18:27:16,292 - Epoch: [157][ 20/ 1207] Overall Loss 0.210572 Objective Loss 0.210572 LR 0.000250 Time 0.053893 -2023-02-13 18:27:16,482 - Epoch: [157][ 30/ 1207] Overall Loss 0.193797 Objective Loss 0.193797 LR 0.000250 Time 0.042250 -2023-02-13 18:27:16,673 - Epoch: [157][ 40/ 1207] Overall Loss 0.191069 Objective Loss 0.191069 LR 0.000250 Time 0.036437 -2023-02-13 18:27:16,863 - Epoch: [157][ 50/ 1207] Overall Loss 0.186446 Objective Loss 0.186446 LR 0.000250 Time 0.032942 -2023-02-13 18:27:17,053 - Epoch: [157][ 60/ 1207] Overall Loss 0.184621 Objective Loss 0.184621 LR 0.000250 Time 0.030619 -2023-02-13 18:27:17,243 - Epoch: [157][ 70/ 1207] Overall Loss 0.185380 Objective Loss 0.185380 LR 0.000250 Time 0.028959 -2023-02-13 18:27:17,433 - Epoch: [157][ 80/ 1207] Overall Loss 0.186073 Objective Loss 0.186073 LR 0.000250 Time 0.027709 -2023-02-13 18:27:17,624 - Epoch: [157][ 90/ 1207] Overall Loss 0.187841 Objective Loss 0.187841 LR 0.000250 Time 0.026745 -2023-02-13 18:27:17,813 - Epoch: [157][ 100/ 1207] Overall Loss 0.188857 Objective Loss 0.188857 LR 0.000250 Time 0.025959 -2023-02-13 18:27:18,004 - Epoch: [157][ 110/ 1207] Overall Loss 0.190367 Objective Loss 0.190367 LR 0.000250 Time 0.025326 -2023-02-13 18:27:18,195 - Epoch: [157][ 120/ 1207] Overall Loss 0.194849 Objective Loss 0.194849 LR 0.000250 Time 0.024810 -2023-02-13 18:27:18,386 - Epoch: [157][ 130/ 1207] Overall Loss 0.195351 Objective Loss 0.195351 LR 0.000250 Time 0.024362 -2023-02-13 18:27:18,578 - Epoch: [157][ 140/ 1207] Overall Loss 0.197224 Objective Loss 0.197224 LR 0.000250 Time 0.023990 -2023-02-13 18:27:18,769 - Epoch: [157][ 150/ 1207] Overall Loss 0.198298 Objective Loss 0.198298 LR 0.000250 Time 0.023667 -2023-02-13 18:27:18,962 - Epoch: [157][ 160/ 1207] Overall Loss 0.198580 Objective Loss 0.198580 LR 0.000250 Time 0.023388 -2023-02-13 18:27:19,153 - Epoch: [157][ 170/ 1207] Overall Loss 0.198292 Objective Loss 0.198292 LR 0.000250 Time 0.023134 -2023-02-13 18:27:19,345 - Epoch: [157][ 180/ 1207] Overall Loss 0.197005 Objective Loss 0.197005 LR 0.000250 Time 0.022914 -2023-02-13 18:27:19,536 - Epoch: [157][ 190/ 1207] Overall Loss 0.197669 Objective Loss 0.197669 LR 0.000250 Time 0.022710 -2023-02-13 18:27:19,728 - Epoch: [157][ 200/ 1207] Overall Loss 0.197113 Objective Loss 0.197113 LR 0.000250 Time 0.022536 -2023-02-13 18:27:19,919 - Epoch: [157][ 210/ 1207] Overall Loss 0.195874 Objective Loss 0.195874 LR 0.000250 Time 0.022368 -2023-02-13 18:27:20,111 - Epoch: [157][ 220/ 1207] Overall Loss 0.195698 Objective Loss 0.195698 LR 0.000250 Time 0.022223 -2023-02-13 18:27:20,302 - Epoch: [157][ 230/ 1207] Overall Loss 0.195107 Objective Loss 0.195107 LR 0.000250 Time 0.022084 -2023-02-13 18:27:20,494 - Epoch: [157][ 240/ 1207] Overall Loss 0.195166 Objective Loss 0.195166 LR 0.000250 Time 0.021962 -2023-02-13 18:27:20,685 - Epoch: [157][ 250/ 1207] Overall Loss 0.195415 Objective Loss 0.195415 LR 0.000250 Time 0.021846 -2023-02-13 18:27:20,876 - Epoch: [157][ 260/ 1207] Overall Loss 0.195193 Objective Loss 0.195193 LR 0.000250 Time 0.021742 -2023-02-13 18:27:21,068 - Epoch: [157][ 270/ 1207] Overall Loss 0.195438 Objective Loss 0.195438 LR 0.000250 Time 0.021645 -2023-02-13 18:27:21,259 - Epoch: [157][ 280/ 1207] Overall Loss 0.196042 Objective Loss 0.196042 LR 0.000250 Time 0.021554 -2023-02-13 18:27:21,450 - Epoch: [157][ 290/ 1207] Overall Loss 0.196259 Objective Loss 0.196259 LR 0.000250 Time 0.021468 -2023-02-13 18:27:21,643 - Epoch: [157][ 300/ 1207] Overall Loss 0.196956 Objective Loss 0.196956 LR 0.000250 Time 0.021394 -2023-02-13 18:27:21,834 - Epoch: [157][ 310/ 1207] Overall Loss 0.196877 Objective Loss 0.196877 LR 0.000250 Time 0.021320 -2023-02-13 18:27:22,028 - Epoch: [157][ 320/ 1207] Overall Loss 0.197005 Objective Loss 0.197005 LR 0.000250 Time 0.021256 -2023-02-13 18:27:22,220 - Epoch: [157][ 330/ 1207] Overall Loss 0.197530 Objective Loss 0.197530 LR 0.000250 Time 0.021193 -2023-02-13 18:27:22,412 - Epoch: [157][ 340/ 1207] Overall Loss 0.197666 Objective Loss 0.197666 LR 0.000250 Time 0.021133 -2023-02-13 18:27:22,603 - Epoch: [157][ 350/ 1207] Overall Loss 0.198194 Objective Loss 0.198194 LR 0.000250 Time 0.021075 -2023-02-13 18:27:22,795 - Epoch: [157][ 360/ 1207] Overall Loss 0.198172 Objective Loss 0.198172 LR 0.000250 Time 0.021022 -2023-02-13 18:27:22,987 - Epoch: [157][ 370/ 1207] Overall Loss 0.198371 Objective Loss 0.198371 LR 0.000250 Time 0.020970 -2023-02-13 18:27:23,180 - Epoch: [157][ 380/ 1207] Overall Loss 0.198286 Objective Loss 0.198286 LR 0.000250 Time 0.020926 -2023-02-13 18:27:23,371 - Epoch: [157][ 390/ 1207] Overall Loss 0.198359 Objective Loss 0.198359 LR 0.000250 Time 0.020879 -2023-02-13 18:27:23,564 - Epoch: [157][ 400/ 1207] Overall Loss 0.198343 Objective Loss 0.198343 LR 0.000250 Time 0.020837 -2023-02-13 18:27:23,755 - Epoch: [157][ 410/ 1207] Overall Loss 0.197807 Objective Loss 0.197807 LR 0.000250 Time 0.020796 -2023-02-13 18:27:23,948 - Epoch: [157][ 420/ 1207] Overall Loss 0.197752 Objective Loss 0.197752 LR 0.000250 Time 0.020758 -2023-02-13 18:27:24,139 - Epoch: [157][ 430/ 1207] Overall Loss 0.198058 Objective Loss 0.198058 LR 0.000250 Time 0.020719 -2023-02-13 18:27:24,331 - Epoch: [157][ 440/ 1207] Overall Loss 0.198091 Objective Loss 0.198091 LR 0.000250 Time 0.020684 -2023-02-13 18:27:24,523 - Epoch: [157][ 450/ 1207] Overall Loss 0.198117 Objective Loss 0.198117 LR 0.000250 Time 0.020649 -2023-02-13 18:27:24,716 - Epoch: [157][ 460/ 1207] Overall Loss 0.197093 Objective Loss 0.197093 LR 0.000250 Time 0.020619 -2023-02-13 18:27:24,907 - Epoch: [157][ 470/ 1207] Overall Loss 0.196491 Objective Loss 0.196491 LR 0.000250 Time 0.020586 -2023-02-13 18:27:25,099 - Epoch: [157][ 480/ 1207] Overall Loss 0.196875 Objective Loss 0.196875 LR 0.000250 Time 0.020557 -2023-02-13 18:27:25,290 - Epoch: [157][ 490/ 1207] Overall Loss 0.196801 Objective Loss 0.196801 LR 0.000250 Time 0.020527 -2023-02-13 18:27:25,483 - Epoch: [157][ 500/ 1207] Overall Loss 0.197351 Objective Loss 0.197351 LR 0.000250 Time 0.020502 -2023-02-13 18:27:25,675 - Epoch: [157][ 510/ 1207] Overall Loss 0.197250 Objective Loss 0.197250 LR 0.000250 Time 0.020476 -2023-02-13 18:27:25,867 - Epoch: [157][ 520/ 1207] Overall Loss 0.197556 Objective Loss 0.197556 LR 0.000250 Time 0.020450 -2023-02-13 18:27:26,061 - Epoch: [157][ 530/ 1207] Overall Loss 0.197175 Objective Loss 0.197175 LR 0.000250 Time 0.020428 -2023-02-13 18:27:26,253 - Epoch: [157][ 540/ 1207] Overall Loss 0.197017 Objective Loss 0.197017 LR 0.000250 Time 0.020405 -2023-02-13 18:27:26,444 - Epoch: [157][ 550/ 1207] Overall Loss 0.197169 Objective Loss 0.197169 LR 0.000250 Time 0.020380 -2023-02-13 18:27:26,638 - Epoch: [157][ 560/ 1207] Overall Loss 0.197324 Objective Loss 0.197324 LR 0.000250 Time 0.020362 -2023-02-13 18:27:26,829 - Epoch: [157][ 570/ 1207] Overall Loss 0.197197 Objective Loss 0.197197 LR 0.000250 Time 0.020340 -2023-02-13 18:27:27,022 - Epoch: [157][ 580/ 1207] Overall Loss 0.197155 Objective Loss 0.197155 LR 0.000250 Time 0.020322 -2023-02-13 18:27:27,213 - Epoch: [157][ 590/ 1207] Overall Loss 0.196852 Objective Loss 0.196852 LR 0.000250 Time 0.020301 -2023-02-13 18:27:27,405 - Epoch: [157][ 600/ 1207] Overall Loss 0.197008 Objective Loss 0.197008 LR 0.000250 Time 0.020282 -2023-02-13 18:27:27,598 - Epoch: [157][ 610/ 1207] Overall Loss 0.197248 Objective Loss 0.197248 LR 0.000250 Time 0.020264 -2023-02-13 18:27:27,791 - Epoch: [157][ 620/ 1207] Overall Loss 0.197142 Objective Loss 0.197142 LR 0.000250 Time 0.020248 -2023-02-13 18:27:27,982 - Epoch: [157][ 630/ 1207] Overall Loss 0.196941 Objective Loss 0.196941 LR 0.000250 Time 0.020229 -2023-02-13 18:27:28,173 - Epoch: [157][ 640/ 1207] Overall Loss 0.196912 Objective Loss 0.196912 LR 0.000250 Time 0.020212 -2023-02-13 18:27:28,365 - Epoch: [157][ 650/ 1207] Overall Loss 0.197098 Objective Loss 0.197098 LR 0.000250 Time 0.020195 -2023-02-13 18:27:28,557 - Epoch: [157][ 660/ 1207] Overall Loss 0.197261 Objective Loss 0.197261 LR 0.000250 Time 0.020180 -2023-02-13 18:27:28,754 - Epoch: [157][ 670/ 1207] Overall Loss 0.197322 Objective Loss 0.197322 LR 0.000250 Time 0.020171 -2023-02-13 18:27:28,960 - Epoch: [157][ 680/ 1207] Overall Loss 0.197109 Objective Loss 0.197109 LR 0.000250 Time 0.020178 -2023-02-13 18:27:29,162 - Epoch: [157][ 690/ 1207] Overall Loss 0.197041 Objective Loss 0.197041 LR 0.000250 Time 0.020177 -2023-02-13 18:27:29,368 - Epoch: [157][ 700/ 1207] Overall Loss 0.197205 Objective Loss 0.197205 LR 0.000250 Time 0.020183 -2023-02-13 18:27:29,566 - Epoch: [157][ 710/ 1207] Overall Loss 0.197270 Objective Loss 0.197270 LR 0.000250 Time 0.020177 -2023-02-13 18:27:29,759 - Epoch: [157][ 720/ 1207] Overall Loss 0.197299 Objective Loss 0.197299 LR 0.000250 Time 0.020164 -2023-02-13 18:27:29,950 - Epoch: [157][ 730/ 1207] Overall Loss 0.197366 Objective Loss 0.197366 LR 0.000250 Time 0.020149 -2023-02-13 18:27:30,142 - Epoch: [157][ 740/ 1207] Overall Loss 0.197463 Objective Loss 0.197463 LR 0.000250 Time 0.020136 -2023-02-13 18:27:30,333 - Epoch: [157][ 750/ 1207] Overall Loss 0.197154 Objective Loss 0.197154 LR 0.000250 Time 0.020122 -2023-02-13 18:27:30,525 - Epoch: [157][ 760/ 1207] Overall Loss 0.196908 Objective Loss 0.196908 LR 0.000250 Time 0.020109 -2023-02-13 18:27:30,717 - Epoch: [157][ 770/ 1207] Overall Loss 0.197124 Objective Loss 0.197124 LR 0.000250 Time 0.020097 -2023-02-13 18:27:30,910 - Epoch: [157][ 780/ 1207] Overall Loss 0.197010 Objective Loss 0.197010 LR 0.000250 Time 0.020085 -2023-02-13 18:27:31,101 - Epoch: [157][ 790/ 1207] Overall Loss 0.197169 Objective Loss 0.197169 LR 0.000250 Time 0.020073 -2023-02-13 18:27:31,294 - Epoch: [157][ 800/ 1207] Overall Loss 0.197365 Objective Loss 0.197365 LR 0.000250 Time 0.020063 -2023-02-13 18:27:31,485 - Epoch: [157][ 810/ 1207] Overall Loss 0.197377 Objective Loss 0.197377 LR 0.000250 Time 0.020050 -2023-02-13 18:27:31,678 - Epoch: [157][ 820/ 1207] Overall Loss 0.197402 Objective Loss 0.197402 LR 0.000250 Time 0.020041 -2023-02-13 18:27:31,870 - Epoch: [157][ 830/ 1207] Overall Loss 0.197346 Objective Loss 0.197346 LR 0.000250 Time 0.020030 -2023-02-13 18:27:32,062 - Epoch: [157][ 840/ 1207] Overall Loss 0.197201 Objective Loss 0.197201 LR 0.000250 Time 0.020020 -2023-02-13 18:27:32,253 - Epoch: [157][ 850/ 1207] Overall Loss 0.197037 Objective Loss 0.197037 LR 0.000250 Time 0.020009 -2023-02-13 18:27:32,446 - Epoch: [157][ 860/ 1207] Overall Loss 0.197017 Objective Loss 0.197017 LR 0.000250 Time 0.020000 -2023-02-13 18:27:32,638 - Epoch: [157][ 870/ 1207] Overall Loss 0.197141 Objective Loss 0.197141 LR 0.000250 Time 0.019991 -2023-02-13 18:27:32,831 - Epoch: [157][ 880/ 1207] Overall Loss 0.196931 Objective Loss 0.196931 LR 0.000250 Time 0.019983 -2023-02-13 18:27:33,022 - Epoch: [157][ 890/ 1207] Overall Loss 0.197066 Objective Loss 0.197066 LR 0.000250 Time 0.019972 -2023-02-13 18:27:33,214 - Epoch: [157][ 900/ 1207] Overall Loss 0.197095 Objective Loss 0.197095 LR 0.000250 Time 0.019963 -2023-02-13 18:27:33,405 - Epoch: [157][ 910/ 1207] Overall Loss 0.197051 Objective Loss 0.197051 LR 0.000250 Time 0.019953 -2023-02-13 18:27:33,598 - Epoch: [157][ 920/ 1207] Overall Loss 0.197186 Objective Loss 0.197186 LR 0.000250 Time 0.019946 -2023-02-13 18:27:33,790 - Epoch: [157][ 930/ 1207] Overall Loss 0.197059 Objective Loss 0.197059 LR 0.000250 Time 0.019937 -2023-02-13 18:27:33,983 - Epoch: [157][ 940/ 1207] Overall Loss 0.197186 Objective Loss 0.197186 LR 0.000250 Time 0.019930 -2023-02-13 18:27:34,174 - Epoch: [157][ 950/ 1207] Overall Loss 0.197031 Objective Loss 0.197031 LR 0.000250 Time 0.019921 -2023-02-13 18:27:34,366 - Epoch: [157][ 960/ 1207] Overall Loss 0.197109 Objective Loss 0.197109 LR 0.000250 Time 0.019913 -2023-02-13 18:27:34,558 - Epoch: [157][ 970/ 1207] Overall Loss 0.197114 Objective Loss 0.197114 LR 0.000250 Time 0.019905 -2023-02-13 18:27:34,751 - Epoch: [157][ 980/ 1207] Overall Loss 0.197194 Objective Loss 0.197194 LR 0.000250 Time 0.019898 -2023-02-13 18:27:34,940 - Epoch: [157][ 990/ 1207] Overall Loss 0.196910 Objective Loss 0.196910 LR 0.000250 Time 0.019888 -2023-02-13 18:27:35,130 - Epoch: [157][ 1000/ 1207] Overall Loss 0.196954 Objective Loss 0.196954 LR 0.000250 Time 0.019879 -2023-02-13 18:27:35,320 - Epoch: [157][ 1010/ 1207] Overall Loss 0.196797 Objective Loss 0.196797 LR 0.000250 Time 0.019870 -2023-02-13 18:27:35,510 - Epoch: [157][ 1020/ 1207] Overall Loss 0.196943 Objective Loss 0.196943 LR 0.000250 Time 0.019861 -2023-02-13 18:27:35,701 - Epoch: [157][ 1030/ 1207] Overall Loss 0.196819 Objective Loss 0.196819 LR 0.000250 Time 0.019853 -2023-02-13 18:27:35,892 - Epoch: [157][ 1040/ 1207] Overall Loss 0.196876 Objective Loss 0.196876 LR 0.000250 Time 0.019846 -2023-02-13 18:27:36,082 - Epoch: [157][ 1050/ 1207] Overall Loss 0.196978 Objective Loss 0.196978 LR 0.000250 Time 0.019837 -2023-02-13 18:27:36,272 - Epoch: [157][ 1060/ 1207] Overall Loss 0.197037 Objective Loss 0.197037 LR 0.000250 Time 0.019829 -2023-02-13 18:27:36,462 - Epoch: [157][ 1070/ 1207] Overall Loss 0.196898 Objective Loss 0.196898 LR 0.000250 Time 0.019821 -2023-02-13 18:27:36,653 - Epoch: [157][ 1080/ 1207] Overall Loss 0.196989 Objective Loss 0.196989 LR 0.000250 Time 0.019813 -2023-02-13 18:27:36,843 - Epoch: [157][ 1090/ 1207] Overall Loss 0.196919 Objective Loss 0.196919 LR 0.000250 Time 0.019806 -2023-02-13 18:27:37,034 - Epoch: [157][ 1100/ 1207] Overall Loss 0.197004 Objective Loss 0.197004 LR 0.000250 Time 0.019799 -2023-02-13 18:27:37,223 - Epoch: [157][ 1110/ 1207] Overall Loss 0.197137 Objective Loss 0.197137 LR 0.000250 Time 0.019791 -2023-02-13 18:27:37,412 - Epoch: [157][ 1120/ 1207] Overall Loss 0.197331 Objective Loss 0.197331 LR 0.000250 Time 0.019783 -2023-02-13 18:27:37,605 - Epoch: [157][ 1130/ 1207] Overall Loss 0.197374 Objective Loss 0.197374 LR 0.000250 Time 0.019778 -2023-02-13 18:27:37,799 - Epoch: [157][ 1140/ 1207] Overall Loss 0.197280 Objective Loss 0.197280 LR 0.000250 Time 0.019774 -2023-02-13 18:27:37,998 - Epoch: [157][ 1150/ 1207] Overall Loss 0.197321 Objective Loss 0.197321 LR 0.000250 Time 0.019775 -2023-02-13 18:27:38,194 - Epoch: [157][ 1160/ 1207] Overall Loss 0.197342 Objective Loss 0.197342 LR 0.000250 Time 0.019773 -2023-02-13 18:27:38,391 - Epoch: [157][ 1170/ 1207] Overall Loss 0.197357 Objective Loss 0.197357 LR 0.000250 Time 0.019772 -2023-02-13 18:27:38,583 - Epoch: [157][ 1180/ 1207] Overall Loss 0.197294 Objective Loss 0.197294 LR 0.000250 Time 0.019767 -2023-02-13 18:27:38,776 - Epoch: [157][ 1190/ 1207] Overall Loss 0.197510 Objective Loss 0.197510 LR 0.000250 Time 0.019763 -2023-02-13 18:27:39,025 - Epoch: [157][ 1200/ 1207] Overall Loss 0.197458 Objective Loss 0.197458 LR 0.000250 Time 0.019805 -2023-02-13 18:27:39,140 - Epoch: [157][ 1207/ 1207] Overall Loss 0.197497 Objective Loss 0.197497 Top1 89.939024 Top5 98.475610 LR 0.000250 Time 0.019786 -2023-02-13 18:27:39,213 - --- validate (epoch=157)----------- -2023-02-13 18:27:39,213 - 34311 samples (256 per mini-batch) -2023-02-13 18:27:39,609 - Epoch: [157][ 10/ 135] Loss 0.330110 Top1 86.132812 Top5 97.500000 -2023-02-13 18:27:39,738 - Epoch: [157][ 20/ 135] Loss 0.316437 Top1 85.273438 Top5 97.734375 -2023-02-13 18:27:39,864 - Epoch: [157][ 30/ 135] Loss 0.301623 Top1 85.572917 Top5 97.916667 -2023-02-13 18:27:39,994 - Epoch: [157][ 40/ 135] Loss 0.302439 Top1 85.556641 Top5 97.841797 -2023-02-13 18:27:40,124 - Epoch: [157][ 50/ 135] Loss 0.298797 Top1 85.726562 Top5 97.945312 -2023-02-13 18:27:40,253 - Epoch: [157][ 60/ 135] Loss 0.300081 Top1 85.579427 Top5 98.020833 -2023-02-13 18:27:40,380 - Epoch: [157][ 70/ 135] Loss 0.295445 Top1 85.758929 Top5 98.041295 -2023-02-13 18:27:40,509 - Epoch: [157][ 80/ 135] Loss 0.292952 Top1 85.742188 Top5 97.978516 -2023-02-13 18:27:40,636 - Epoch: [157][ 90/ 135] Loss 0.288014 Top1 85.889757 Top5 98.020833 -2023-02-13 18:27:40,760 - Epoch: [157][ 100/ 135] Loss 0.290019 Top1 85.871094 Top5 98.042969 -2023-02-13 18:27:40,887 - Epoch: [157][ 110/ 135] Loss 0.292006 Top1 85.848722 Top5 97.986506 -2023-02-13 18:27:41,018 - Epoch: [157][ 120/ 135] Loss 0.290603 Top1 85.895182 Top5 98.001302 -2023-02-13 18:27:41,147 - Epoch: [157][ 130/ 135] Loss 0.291689 Top1 85.751202 Top5 97.971755 -2023-02-13 18:27:41,193 - Epoch: [157][ 135/ 135] Loss 0.292039 Top1 85.680977 Top5 97.974411 -2023-02-13 18:27:41,265 - ==> Top1: 85.681 Top5: 97.974 Loss: 0.292 - -2023-02-13 18:27:41,266 - ==> Confusion: -[[ 870 2 8 1 8 3 0 0 5 32 1 4 1 4 10 2 2 2 1 4 7] - [ 3 955 1 1 12 24 3 11 2 0 3 1 0 0 0 1 2 1 5 1 7] - [ 9 3 965 10 7 1 15 11 0 1 1 1 2 3 4 3 3 2 7 3 7] - [ 3 0 21 914 4 4 0 1 2 1 9 0 8 1 16 2 3 6 17 0 4] - [ 8 11 3 1 990 8 1 3 0 1 0 5 3 3 8 5 7 1 2 1 5] - [ 3 20 2 7 6 956 2 21 0 4 2 8 2 16 1 2 2 1 1 7 7] - [ 2 4 16 2 0 5 1040 4 0 2 1 0 2 1 1 4 1 4 1 5 4] - [ 2 10 8 2 1 22 3 935 0 2 0 4 2 2 0 0 0 0 14 11 6] - [ 18 1 2 1 1 1 1 2 911 35 5 1 0 7 15 3 1 0 3 0 1] - [ 98 1 4 0 11 1 0 2 33 833 0 1 1 13 6 0 1 1 2 0 4] - [ 1 1 6 12 1 1 2 7 18 0 978 1 2 7 4 0 1 1 4 0 4] - [ 1 2 1 0 1 13 1 8 1 3 0 906 33 9 1 2 2 10 1 6 4] - [ 0 0 0 9 2 2 0 2 2 0 1 21 882 0 2 9 2 14 1 0 10] - [ 5 3 2 1 8 5 0 2 11 13 8 4 2 935 5 6 2 2 1 1 8] - [ 2 3 1 16 3 2 0 1 20 6 1 1 2 2 1011 1 1 5 5 2 7] - [ 3 3 8 1 5 1 4 1 0 1 1 6 7 2 0 970 11 7 0 5 10] - [ 4 6 0 1 7 0 0 0 2 0 0 0 0 1 1 8 1010 3 1 5 12] - [ 4 2 1 4 2 3 2 0 0 1 1 8 10 1 0 13 0 991 1 1 6] - [ 1 4 4 11 0 2 0 21 4 0 3 0 6 0 10 1 0 2 1009 3 5] - [ 0 3 2 0 2 6 8 11 1 0 2 11 5 4 0 5 6 3 0 1070 9] - [ 126 219 241 138 126 163 76 163 96 66 146 94 285 250 162 97 234 102 190 193 10267]] - -2023-02-13 18:27:41,267 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:27:41,267 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:27:41,273 - - -2023-02-13 18:27:41,273 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:27:42,266 - Epoch: [158][ 10/ 1207] Overall Loss 0.183168 Objective Loss 0.183168 LR 0.000250 Time 0.099219 -2023-02-13 18:27:42,463 - Epoch: [158][ 20/ 1207] Overall Loss 0.194535 Objective Loss 0.194535 LR 0.000250 Time 0.059476 -2023-02-13 18:27:42,653 - Epoch: [158][ 30/ 1207] Overall Loss 0.192547 Objective Loss 0.192547 LR 0.000250 Time 0.045971 -2023-02-13 18:27:42,843 - Epoch: [158][ 40/ 1207] Overall Loss 0.191364 Objective Loss 0.191364 LR 0.000250 Time 0.039220 -2023-02-13 18:27:43,033 - Epoch: [158][ 50/ 1207] Overall Loss 0.193182 Objective Loss 0.193182 LR 0.000250 Time 0.035168 -2023-02-13 18:27:43,223 - Epoch: [158][ 60/ 1207] Overall Loss 0.194668 Objective Loss 0.194668 LR 0.000250 Time 0.032459 -2023-02-13 18:27:43,412 - Epoch: [158][ 70/ 1207] Overall Loss 0.194623 Objective Loss 0.194623 LR 0.000250 Time 0.030520 -2023-02-13 18:27:43,602 - Epoch: [158][ 80/ 1207] Overall Loss 0.193904 Objective Loss 0.193904 LR 0.000250 Time 0.029069 -2023-02-13 18:27:43,791 - Epoch: [158][ 90/ 1207] Overall Loss 0.191974 Objective Loss 0.191974 LR 0.000250 Time 0.027937 -2023-02-13 18:27:43,980 - Epoch: [158][ 100/ 1207] Overall Loss 0.194208 Objective Loss 0.194208 LR 0.000250 Time 0.027035 -2023-02-13 18:27:44,170 - Epoch: [158][ 110/ 1207] Overall Loss 0.194004 Objective Loss 0.194004 LR 0.000250 Time 0.026302 -2023-02-13 18:27:44,359 - Epoch: [158][ 120/ 1207] Overall Loss 0.195274 Objective Loss 0.195274 LR 0.000250 Time 0.025682 -2023-02-13 18:27:44,549 - Epoch: [158][ 130/ 1207] Overall Loss 0.194333 Objective Loss 0.194333 LR 0.000250 Time 0.025158 -2023-02-13 18:27:44,738 - Epoch: [158][ 140/ 1207] Overall Loss 0.193446 Objective Loss 0.193446 LR 0.000250 Time 0.024713 -2023-02-13 18:27:44,927 - Epoch: [158][ 150/ 1207] Overall Loss 0.194159 Objective Loss 0.194159 LR 0.000250 Time 0.024320 -2023-02-13 18:27:45,117 - Epoch: [158][ 160/ 1207] Overall Loss 0.194717 Objective Loss 0.194717 LR 0.000250 Time 0.023984 -2023-02-13 18:27:45,306 - Epoch: [158][ 170/ 1207] Overall Loss 0.194579 Objective Loss 0.194579 LR 0.000250 Time 0.023683 -2023-02-13 18:27:45,496 - Epoch: [158][ 180/ 1207] Overall Loss 0.194501 Objective Loss 0.194501 LR 0.000250 Time 0.023420 -2023-02-13 18:27:45,686 - Epoch: [158][ 190/ 1207] Overall Loss 0.194014 Objective Loss 0.194014 LR 0.000250 Time 0.023186 -2023-02-13 18:27:45,876 - Epoch: [158][ 200/ 1207] Overall Loss 0.194618 Objective Loss 0.194618 LR 0.000250 Time 0.022976 -2023-02-13 18:27:46,067 - Epoch: [158][ 210/ 1207] Overall Loss 0.193712 Objective Loss 0.193712 LR 0.000250 Time 0.022788 -2023-02-13 18:27:46,256 - Epoch: [158][ 220/ 1207] Overall Loss 0.194569 Objective Loss 0.194569 LR 0.000250 Time 0.022611 -2023-02-13 18:27:46,445 - Epoch: [158][ 230/ 1207] Overall Loss 0.195124 Objective Loss 0.195124 LR 0.000250 Time 0.022450 -2023-02-13 18:27:46,635 - Epoch: [158][ 240/ 1207] Overall Loss 0.195974 Objective Loss 0.195974 LR 0.000250 Time 0.022303 -2023-02-13 18:27:46,825 - Epoch: [158][ 250/ 1207] Overall Loss 0.196968 Objective Loss 0.196968 LR 0.000250 Time 0.022169 -2023-02-13 18:27:47,015 - Epoch: [158][ 260/ 1207] Overall Loss 0.196409 Objective Loss 0.196409 LR 0.000250 Time 0.022046 -2023-02-13 18:27:47,205 - Epoch: [158][ 270/ 1207] Overall Loss 0.196316 Objective Loss 0.196316 LR 0.000250 Time 0.021931 -2023-02-13 18:27:47,395 - Epoch: [158][ 280/ 1207] Overall Loss 0.196433 Objective Loss 0.196433 LR 0.000250 Time 0.021824 -2023-02-13 18:27:47,585 - Epoch: [158][ 290/ 1207] Overall Loss 0.196397 Objective Loss 0.196397 LR 0.000250 Time 0.021726 -2023-02-13 18:27:47,775 - Epoch: [158][ 300/ 1207] Overall Loss 0.196912 Objective Loss 0.196912 LR 0.000250 Time 0.021634 -2023-02-13 18:27:47,964 - Epoch: [158][ 310/ 1207] Overall Loss 0.196833 Objective Loss 0.196833 LR 0.000250 Time 0.021546 -2023-02-13 18:27:48,153 - Epoch: [158][ 320/ 1207] Overall Loss 0.196071 Objective Loss 0.196071 LR 0.000250 Time 0.021462 -2023-02-13 18:27:48,342 - Epoch: [158][ 330/ 1207] Overall Loss 0.196030 Objective Loss 0.196030 LR 0.000250 Time 0.021384 -2023-02-13 18:27:48,532 - Epoch: [158][ 340/ 1207] Overall Loss 0.196570 Objective Loss 0.196570 LR 0.000250 Time 0.021311 -2023-02-13 18:27:48,721 - Epoch: [158][ 350/ 1207] Overall Loss 0.197159 Objective Loss 0.197159 LR 0.000250 Time 0.021243 -2023-02-13 18:27:48,910 - Epoch: [158][ 360/ 1207] Overall Loss 0.197373 Objective Loss 0.197373 LR 0.000250 Time 0.021176 -2023-02-13 18:27:49,099 - Epoch: [158][ 370/ 1207] Overall Loss 0.197178 Objective Loss 0.197178 LR 0.000250 Time 0.021113 -2023-02-13 18:27:49,289 - Epoch: [158][ 380/ 1207] Overall Loss 0.196740 Objective Loss 0.196740 LR 0.000250 Time 0.021055 -2023-02-13 18:27:49,478 - Epoch: [158][ 390/ 1207] Overall Loss 0.196788 Objective Loss 0.196788 LR 0.000250 Time 0.021000 -2023-02-13 18:27:49,667 - Epoch: [158][ 400/ 1207] Overall Loss 0.196276 Objective Loss 0.196276 LR 0.000250 Time 0.020947 -2023-02-13 18:27:49,857 - Epoch: [158][ 410/ 1207] Overall Loss 0.196381 Objective Loss 0.196381 LR 0.000250 Time 0.020899 -2023-02-13 18:27:50,047 - Epoch: [158][ 420/ 1207] Overall Loss 0.195969 Objective Loss 0.195969 LR 0.000250 Time 0.020851 -2023-02-13 18:27:50,236 - Epoch: [158][ 430/ 1207] Overall Loss 0.195872 Objective Loss 0.195872 LR 0.000250 Time 0.020807 -2023-02-13 18:27:50,427 - Epoch: [158][ 440/ 1207] Overall Loss 0.195948 Objective Loss 0.195948 LR 0.000250 Time 0.020766 -2023-02-13 18:27:50,617 - Epoch: [158][ 450/ 1207] Overall Loss 0.195914 Objective Loss 0.195914 LR 0.000250 Time 0.020725 -2023-02-13 18:27:50,807 - Epoch: [158][ 460/ 1207] Overall Loss 0.195739 Objective Loss 0.195739 LR 0.000250 Time 0.020688 -2023-02-13 18:27:50,998 - Epoch: [158][ 470/ 1207] Overall Loss 0.195727 Objective Loss 0.195727 LR 0.000250 Time 0.020653 -2023-02-13 18:27:51,187 - Epoch: [158][ 480/ 1207] Overall Loss 0.195667 Objective Loss 0.195667 LR 0.000250 Time 0.020616 -2023-02-13 18:27:51,377 - Epoch: [158][ 490/ 1207] Overall Loss 0.195889 Objective Loss 0.195889 LR 0.000250 Time 0.020582 -2023-02-13 18:27:51,567 - Epoch: [158][ 500/ 1207] Overall Loss 0.195981 Objective Loss 0.195981 LR 0.000250 Time 0.020549 -2023-02-13 18:27:51,757 - Epoch: [158][ 510/ 1207] Overall Loss 0.196024 Objective Loss 0.196024 LR 0.000250 Time 0.020518 -2023-02-13 18:27:51,947 - Epoch: [158][ 520/ 1207] Overall Loss 0.196371 Objective Loss 0.196371 LR 0.000250 Time 0.020489 -2023-02-13 18:27:52,136 - Epoch: [158][ 530/ 1207] Overall Loss 0.196521 Objective Loss 0.196521 LR 0.000250 Time 0.020458 -2023-02-13 18:27:52,326 - Epoch: [158][ 540/ 1207] Overall Loss 0.196373 Objective Loss 0.196373 LR 0.000250 Time 0.020429 -2023-02-13 18:27:52,515 - Epoch: [158][ 550/ 1207] Overall Loss 0.196457 Objective Loss 0.196457 LR 0.000250 Time 0.020401 -2023-02-13 18:27:52,705 - Epoch: [158][ 560/ 1207] Overall Loss 0.196350 Objective Loss 0.196350 LR 0.000250 Time 0.020375 -2023-02-13 18:27:52,895 - Epoch: [158][ 570/ 1207] Overall Loss 0.196146 Objective Loss 0.196146 LR 0.000250 Time 0.020350 -2023-02-13 18:27:53,084 - Epoch: [158][ 580/ 1207] Overall Loss 0.196127 Objective Loss 0.196127 LR 0.000250 Time 0.020325 -2023-02-13 18:27:53,273 - Epoch: [158][ 590/ 1207] Overall Loss 0.196333 Objective Loss 0.196333 LR 0.000250 Time 0.020300 -2023-02-13 18:27:53,463 - Epoch: [158][ 600/ 1207] Overall Loss 0.196400 Objective Loss 0.196400 LR 0.000250 Time 0.020278 -2023-02-13 18:27:53,652 - Epoch: [158][ 610/ 1207] Overall Loss 0.196189 Objective Loss 0.196189 LR 0.000250 Time 0.020256 -2023-02-13 18:27:53,842 - Epoch: [158][ 620/ 1207] Overall Loss 0.196256 Objective Loss 0.196256 LR 0.000250 Time 0.020235 -2023-02-13 18:27:54,031 - Epoch: [158][ 630/ 1207] Overall Loss 0.195858 Objective Loss 0.195858 LR 0.000250 Time 0.020213 -2023-02-13 18:27:54,221 - Epoch: [158][ 640/ 1207] Overall Loss 0.196020 Objective Loss 0.196020 LR 0.000250 Time 0.020193 -2023-02-13 18:27:54,411 - Epoch: [158][ 650/ 1207] Overall Loss 0.196128 Objective Loss 0.196128 LR 0.000250 Time 0.020173 -2023-02-13 18:27:54,600 - Epoch: [158][ 660/ 1207] Overall Loss 0.195966 Objective Loss 0.195966 LR 0.000250 Time 0.020155 -2023-02-13 18:27:54,791 - Epoch: [158][ 670/ 1207] Overall Loss 0.195998 Objective Loss 0.195998 LR 0.000250 Time 0.020137 -2023-02-13 18:27:54,980 - Epoch: [158][ 680/ 1207] Overall Loss 0.196187 Objective Loss 0.196187 LR 0.000250 Time 0.020119 -2023-02-13 18:27:55,169 - Epoch: [158][ 690/ 1207] Overall Loss 0.195756 Objective Loss 0.195756 LR 0.000250 Time 0.020100 -2023-02-13 18:27:55,359 - Epoch: [158][ 700/ 1207] Overall Loss 0.195548 Objective Loss 0.195548 LR 0.000250 Time 0.020084 -2023-02-13 18:27:55,548 - Epoch: [158][ 710/ 1207] Overall Loss 0.195735 Objective Loss 0.195735 LR 0.000250 Time 0.020067 -2023-02-13 18:27:55,739 - Epoch: [158][ 720/ 1207] Overall Loss 0.195775 Objective Loss 0.195775 LR 0.000250 Time 0.020054 -2023-02-13 18:27:55,929 - Epoch: [158][ 730/ 1207] Overall Loss 0.195984 Objective Loss 0.195984 LR 0.000250 Time 0.020039 -2023-02-13 18:27:56,119 - Epoch: [158][ 740/ 1207] Overall Loss 0.195858 Objective Loss 0.195858 LR 0.000250 Time 0.020024 -2023-02-13 18:27:56,309 - Epoch: [158][ 750/ 1207] Overall Loss 0.196055 Objective Loss 0.196055 LR 0.000250 Time 0.020009 -2023-02-13 18:27:56,498 - Epoch: [158][ 760/ 1207] Overall Loss 0.196123 Objective Loss 0.196123 LR 0.000250 Time 0.019995 -2023-02-13 18:27:56,689 - Epoch: [158][ 770/ 1207] Overall Loss 0.195736 Objective Loss 0.195736 LR 0.000250 Time 0.019982 -2023-02-13 18:27:56,879 - Epoch: [158][ 780/ 1207] Overall Loss 0.195558 Objective Loss 0.195558 LR 0.000250 Time 0.019969 -2023-02-13 18:27:57,068 - Epoch: [158][ 790/ 1207] Overall Loss 0.195839 Objective Loss 0.195839 LR 0.000250 Time 0.019955 -2023-02-13 18:27:57,257 - Epoch: [158][ 800/ 1207] Overall Loss 0.195765 Objective Loss 0.195765 LR 0.000250 Time 0.019942 -2023-02-13 18:27:57,447 - Epoch: [158][ 810/ 1207] Overall Loss 0.195440 Objective Loss 0.195440 LR 0.000250 Time 0.019929 -2023-02-13 18:27:57,637 - Epoch: [158][ 820/ 1207] Overall Loss 0.195386 Objective Loss 0.195386 LR 0.000250 Time 0.019917 -2023-02-13 18:27:57,827 - Epoch: [158][ 830/ 1207] Overall Loss 0.195523 Objective Loss 0.195523 LR 0.000250 Time 0.019906 -2023-02-13 18:27:58,016 - Epoch: [158][ 840/ 1207] Overall Loss 0.195465 Objective Loss 0.195465 LR 0.000250 Time 0.019893 -2023-02-13 18:27:58,205 - Epoch: [158][ 850/ 1207] Overall Loss 0.195450 Objective Loss 0.195450 LR 0.000250 Time 0.019881 -2023-02-13 18:27:58,395 - Epoch: [158][ 860/ 1207] Overall Loss 0.195582 Objective Loss 0.195582 LR 0.000250 Time 0.019871 -2023-02-13 18:27:58,583 - Epoch: [158][ 870/ 1207] Overall Loss 0.195516 Objective Loss 0.195516 LR 0.000250 Time 0.019859 -2023-02-13 18:27:58,774 - Epoch: [158][ 880/ 1207] Overall Loss 0.195642 Objective Loss 0.195642 LR 0.000250 Time 0.019849 -2023-02-13 18:27:58,963 - Epoch: [158][ 890/ 1207] Overall Loss 0.195520 Objective Loss 0.195520 LR 0.000250 Time 0.019838 -2023-02-13 18:27:59,153 - Epoch: [158][ 900/ 1207] Overall Loss 0.195510 Objective Loss 0.195510 LR 0.000250 Time 0.019828 -2023-02-13 18:27:59,342 - Epoch: [158][ 910/ 1207] Overall Loss 0.195659 Objective Loss 0.195659 LR 0.000250 Time 0.019818 -2023-02-13 18:27:59,532 - Epoch: [158][ 920/ 1207] Overall Loss 0.195692 Objective Loss 0.195692 LR 0.000250 Time 0.019809 -2023-02-13 18:27:59,722 - Epoch: [158][ 930/ 1207] Overall Loss 0.195648 Objective Loss 0.195648 LR 0.000250 Time 0.019799 -2023-02-13 18:27:59,912 - Epoch: [158][ 940/ 1207] Overall Loss 0.195600 Objective Loss 0.195600 LR 0.000250 Time 0.019791 -2023-02-13 18:28:00,101 - Epoch: [158][ 950/ 1207] Overall Loss 0.195414 Objective Loss 0.195414 LR 0.000250 Time 0.019781 -2023-02-13 18:28:00,291 - Epoch: [158][ 960/ 1207] Overall Loss 0.195265 Objective Loss 0.195265 LR 0.000250 Time 0.019772 -2023-02-13 18:28:00,480 - Epoch: [158][ 970/ 1207] Overall Loss 0.195266 Objective Loss 0.195266 LR 0.000250 Time 0.019763 -2023-02-13 18:28:00,670 - Epoch: [158][ 980/ 1207] Overall Loss 0.195328 Objective Loss 0.195328 LR 0.000250 Time 0.019754 -2023-02-13 18:28:00,860 - Epoch: [158][ 990/ 1207] Overall Loss 0.195450 Objective Loss 0.195450 LR 0.000250 Time 0.019746 -2023-02-13 18:28:01,050 - Epoch: [158][ 1000/ 1207] Overall Loss 0.195672 Objective Loss 0.195672 LR 0.000250 Time 0.019739 -2023-02-13 18:28:01,240 - Epoch: [158][ 1010/ 1207] Overall Loss 0.195754 Objective Loss 0.195754 LR 0.000250 Time 0.019731 -2023-02-13 18:28:01,430 - Epoch: [158][ 1020/ 1207] Overall Loss 0.195754 Objective Loss 0.195754 LR 0.000250 Time 0.019723 -2023-02-13 18:28:01,620 - Epoch: [158][ 1030/ 1207] Overall Loss 0.195728 Objective Loss 0.195728 LR 0.000250 Time 0.019716 -2023-02-13 18:28:01,811 - Epoch: [158][ 1040/ 1207] Overall Loss 0.196062 Objective Loss 0.196062 LR 0.000250 Time 0.019710 -2023-02-13 18:28:02,002 - Epoch: [158][ 1050/ 1207] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000250 Time 0.019703 -2023-02-13 18:28:02,191 - Epoch: [158][ 1060/ 1207] Overall Loss 0.195988 Objective Loss 0.195988 LR 0.000250 Time 0.019696 -2023-02-13 18:28:02,381 - Epoch: [158][ 1070/ 1207] Overall Loss 0.196256 Objective Loss 0.196256 LR 0.000250 Time 0.019688 -2023-02-13 18:28:02,571 - Epoch: [158][ 1080/ 1207] Overall Loss 0.196277 Objective Loss 0.196277 LR 0.000250 Time 0.019682 -2023-02-13 18:28:02,760 - Epoch: [158][ 1090/ 1207] Overall Loss 0.196520 Objective Loss 0.196520 LR 0.000250 Time 0.019674 -2023-02-13 18:28:02,950 - Epoch: [158][ 1100/ 1207] Overall Loss 0.196538 Objective Loss 0.196538 LR 0.000250 Time 0.019668 -2023-02-13 18:28:03,139 - Epoch: [158][ 1110/ 1207] Overall Loss 0.196678 Objective Loss 0.196678 LR 0.000250 Time 0.019661 -2023-02-13 18:28:03,328 - Epoch: [158][ 1120/ 1207] Overall Loss 0.196663 Objective Loss 0.196663 LR 0.000250 Time 0.019654 -2023-02-13 18:28:03,518 - Epoch: [158][ 1130/ 1207] Overall Loss 0.196901 Objective Loss 0.196901 LR 0.000250 Time 0.019647 -2023-02-13 18:28:03,707 - Epoch: [158][ 1140/ 1207] Overall Loss 0.196862 Objective Loss 0.196862 LR 0.000250 Time 0.019640 -2023-02-13 18:28:03,896 - Epoch: [158][ 1150/ 1207] Overall Loss 0.196791 Objective Loss 0.196791 LR 0.000250 Time 0.019634 -2023-02-13 18:28:04,093 - Epoch: [158][ 1160/ 1207] Overall Loss 0.196572 Objective Loss 0.196572 LR 0.000250 Time 0.019634 -2023-02-13 18:28:04,283 - Epoch: [158][ 1170/ 1207] Overall Loss 0.196459 Objective Loss 0.196459 LR 0.000250 Time 0.019628 -2023-02-13 18:28:04,472 - Epoch: [158][ 1180/ 1207] Overall Loss 0.196227 Objective Loss 0.196227 LR 0.000250 Time 0.019622 -2023-02-13 18:28:04,662 - Epoch: [158][ 1190/ 1207] Overall Loss 0.196112 Objective Loss 0.196112 LR 0.000250 Time 0.019616 -2023-02-13 18:28:04,909 - Epoch: [158][ 1200/ 1207] Overall Loss 0.196085 Objective Loss 0.196085 LR 0.000250 Time 0.019658 -2023-02-13 18:28:05,024 - Epoch: [158][ 1207/ 1207] Overall Loss 0.196112 Objective Loss 0.196112 Top1 89.024390 Top5 98.170732 LR 0.000250 Time 0.019640 -2023-02-13 18:28:05,096 - --- validate (epoch=158)----------- -2023-02-13 18:28:05,097 - 34311 samples (256 per mini-batch) -2023-02-13 18:28:05,499 - Epoch: [158][ 10/ 135] Loss 0.299058 Top1 86.640625 Top5 97.929688 -2023-02-13 18:28:05,626 - Epoch: [158][ 20/ 135] Loss 0.302167 Top1 86.328125 Top5 98.203125 -2023-02-13 18:28:05,752 - Epoch: [158][ 30/ 135] Loss 0.299135 Top1 85.729167 Top5 97.955729 -2023-02-13 18:28:05,876 - Epoch: [158][ 40/ 135] Loss 0.306388 Top1 85.507812 Top5 97.900391 -2023-02-13 18:28:06,001 - Epoch: [158][ 50/ 135] Loss 0.306624 Top1 85.515625 Top5 97.968750 -2023-02-13 18:28:06,134 - Epoch: [158][ 60/ 135] Loss 0.303614 Top1 85.501302 Top5 97.955729 -2023-02-13 18:28:06,281 - Epoch: [158][ 70/ 135] Loss 0.302239 Top1 85.625000 Top5 98.041295 -2023-02-13 18:28:06,423 - Epoch: [158][ 80/ 135] Loss 0.300345 Top1 85.561523 Top5 98.051758 -2023-02-13 18:28:06,566 - Epoch: [158][ 90/ 135] Loss 0.298535 Top1 85.651042 Top5 98.072917 -2023-02-13 18:28:06,712 - Epoch: [158][ 100/ 135] Loss 0.298191 Top1 85.695312 Top5 98.039062 -2023-02-13 18:28:06,845 - Epoch: [158][ 110/ 135] Loss 0.296355 Top1 85.724432 Top5 98.032670 -2023-02-13 18:28:06,973 - Epoch: [158][ 120/ 135] Loss 0.296296 Top1 85.800781 Top5 98.027344 -2023-02-13 18:28:07,107 - Epoch: [158][ 130/ 135] Loss 0.295139 Top1 85.856370 Top5 97.998798 -2023-02-13 18:28:07,154 - Epoch: [158][ 135/ 135] Loss 0.303201 Top1 85.873335 Top5 98.006470 -2023-02-13 18:28:07,222 - ==> Top1: 85.873 Top5: 98.006 Loss: 0.303 - -2023-02-13 18:28:07,223 - ==> Confusion: -[[ 869 2 4 0 10 4 0 2 5 35 0 6 0 3 9 1 2 3 0 2 10] - [ 3 944 1 1 6 28 4 15 3 1 1 1 1 1 1 3 8 0 4 1 6] - [ 5 3 966 10 7 1 14 12 0 1 3 1 5 4 3 4 5 2 5 1 6] - [ 3 0 20 912 3 4 2 2 4 2 6 0 8 0 19 2 4 8 12 0 5] - [ 14 9 0 0 984 10 2 5 0 1 0 5 3 1 9 8 4 1 0 1 9] - [ 3 10 2 3 6 972 6 15 0 4 1 10 6 14 1 1 3 0 1 6 6] - [ 1 3 14 1 0 8 1041 3 0 1 1 2 1 0 1 3 3 4 1 6 5] - [ 2 5 11 2 1 29 3 937 0 2 0 5 1 1 0 0 1 1 8 10 5] - [ 17 2 1 1 1 0 1 1 893 46 8 3 0 9 17 2 1 0 4 0 2] - [ 91 0 3 0 11 3 0 4 29 838 0 0 1 14 6 2 1 4 2 0 3] - [ 3 1 2 6 1 1 3 7 13 1 985 0 1 9 4 0 1 1 6 1 5] - [ 1 2 1 0 4 12 1 6 1 3 0 906 35 6 0 4 3 9 2 7 2] - [ 0 0 1 4 2 2 0 1 2 1 1 22 886 0 5 9 1 11 1 1 9] - [ 1 3 2 1 8 9 0 3 11 15 8 7 4 934 4 4 3 2 0 0 5] - [ 8 1 1 20 2 3 0 2 10 4 3 2 4 1 1007 2 2 4 9 0 7] - [ 4 2 6 0 8 2 3 1 0 0 0 5 9 1 0 979 6 8 0 7 5] - [ 3 4 1 1 6 3 0 1 0 0 0 2 3 2 1 13 1003 2 1 3 12] - [ 4 2 0 3 1 1 1 1 0 0 0 7 18 0 0 13 1 990 1 1 7] - [ 3 3 3 8 1 1 0 32 2 0 3 2 5 0 10 2 1 3 1000 2 5] - [ 1 3 1 0 1 7 10 8 1 0 0 17 4 3 0 7 3 1 0 1073 8] - [ 128 188 239 103 118 227 101 186 66 57 158 92 301 244 144 88 219 99 140 191 10345]] - -2023-02-13 18:28:07,224 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:28:07,224 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:28:07,230 - - -2023-02-13 18:28:07,230 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:28:08,127 - Epoch: [159][ 10/ 1207] Overall Loss 0.171602 Objective Loss 0.171602 LR 0.000250 Time 0.089661 -2023-02-13 18:28:08,330 - Epoch: [159][ 20/ 1207] Overall Loss 0.173006 Objective Loss 0.173006 LR 0.000250 Time 0.054925 -2023-02-13 18:28:08,522 - Epoch: [159][ 30/ 1207] Overall Loss 0.178296 Objective Loss 0.178296 LR 0.000250 Time 0.043001 -2023-02-13 18:28:08,711 - Epoch: [159][ 40/ 1207] Overall Loss 0.177848 Objective Loss 0.177848 LR 0.000250 Time 0.036983 -2023-02-13 18:28:08,902 - Epoch: [159][ 50/ 1207] Overall Loss 0.183265 Objective Loss 0.183265 LR 0.000250 Time 0.033392 -2023-02-13 18:28:09,091 - Epoch: [159][ 60/ 1207] Overall Loss 0.183026 Objective Loss 0.183026 LR 0.000250 Time 0.030976 -2023-02-13 18:28:09,281 - Epoch: [159][ 70/ 1207] Overall Loss 0.187699 Objective Loss 0.187699 LR 0.000250 Time 0.029253 -2023-02-13 18:28:09,471 - Epoch: [159][ 80/ 1207] Overall Loss 0.187611 Objective Loss 0.187611 LR 0.000250 Time 0.027967 -2023-02-13 18:28:09,661 - Epoch: [159][ 90/ 1207] Overall Loss 0.188607 Objective Loss 0.188607 LR 0.000250 Time 0.026965 -2023-02-13 18:28:09,851 - Epoch: [159][ 100/ 1207] Overall Loss 0.189623 Objective Loss 0.189623 LR 0.000250 Time 0.026168 -2023-02-13 18:28:10,041 - Epoch: [159][ 110/ 1207] Overall Loss 0.190763 Objective Loss 0.190763 LR 0.000250 Time 0.025512 -2023-02-13 18:28:10,230 - Epoch: [159][ 120/ 1207] Overall Loss 0.190346 Objective Loss 0.190346 LR 0.000250 Time 0.024962 -2023-02-13 18:28:10,420 - Epoch: [159][ 130/ 1207] Overall Loss 0.189725 Objective Loss 0.189725 LR 0.000250 Time 0.024502 -2023-02-13 18:28:10,610 - Epoch: [159][ 140/ 1207] Overall Loss 0.189869 Objective Loss 0.189869 LR 0.000250 Time 0.024104 -2023-02-13 18:28:10,800 - Epoch: [159][ 150/ 1207] Overall Loss 0.190165 Objective Loss 0.190165 LR 0.000250 Time 0.023757 -2023-02-13 18:28:10,991 - Epoch: [159][ 160/ 1207] Overall Loss 0.188933 Objective Loss 0.188933 LR 0.000250 Time 0.023467 -2023-02-13 18:28:11,181 - Epoch: [159][ 170/ 1207] Overall Loss 0.190059 Objective Loss 0.190059 LR 0.000250 Time 0.023202 -2023-02-13 18:28:11,371 - Epoch: [159][ 180/ 1207] Overall Loss 0.191239 Objective Loss 0.191239 LR 0.000250 Time 0.022965 -2023-02-13 18:28:11,562 - Epoch: [159][ 190/ 1207] Overall Loss 0.190255 Objective Loss 0.190255 LR 0.000250 Time 0.022759 -2023-02-13 18:28:11,751 - Epoch: [159][ 200/ 1207] Overall Loss 0.189933 Objective Loss 0.189933 LR 0.000250 Time 0.022568 -2023-02-13 18:28:11,942 - Epoch: [159][ 210/ 1207] Overall Loss 0.190246 Objective Loss 0.190246 LR 0.000250 Time 0.022399 -2023-02-13 18:28:12,132 - Epoch: [159][ 220/ 1207] Overall Loss 0.190251 Objective Loss 0.190251 LR 0.000250 Time 0.022241 -2023-02-13 18:28:12,323 - Epoch: [159][ 230/ 1207] Overall Loss 0.190123 Objective Loss 0.190123 LR 0.000250 Time 0.022105 -2023-02-13 18:28:12,515 - Epoch: [159][ 240/ 1207] Overall Loss 0.191500 Objective Loss 0.191500 LR 0.000250 Time 0.021981 -2023-02-13 18:28:12,706 - Epoch: [159][ 250/ 1207] Overall Loss 0.191236 Objective Loss 0.191236 LR 0.000250 Time 0.021866 -2023-02-13 18:28:12,900 - Epoch: [159][ 260/ 1207] Overall Loss 0.190735 Objective Loss 0.190735 LR 0.000250 Time 0.021768 -2023-02-13 18:28:13,091 - Epoch: [159][ 270/ 1207] Overall Loss 0.191550 Objective Loss 0.191550 LR 0.000250 Time 0.021668 -2023-02-13 18:28:13,284 - Epoch: [159][ 280/ 1207] Overall Loss 0.191450 Objective Loss 0.191450 LR 0.000250 Time 0.021581 -2023-02-13 18:28:13,475 - Epoch: [159][ 290/ 1207] Overall Loss 0.191857 Objective Loss 0.191857 LR 0.000250 Time 0.021496 -2023-02-13 18:28:13,668 - Epoch: [159][ 300/ 1207] Overall Loss 0.192028 Objective Loss 0.192028 LR 0.000250 Time 0.021422 -2023-02-13 18:28:13,860 - Epoch: [159][ 310/ 1207] Overall Loss 0.191817 Objective Loss 0.191817 LR 0.000250 Time 0.021348 -2023-02-13 18:28:14,052 - Epoch: [159][ 320/ 1207] Overall Loss 0.192159 Objective Loss 0.192159 LR 0.000250 Time 0.021280 -2023-02-13 18:28:14,244 - Epoch: [159][ 330/ 1207] Overall Loss 0.192027 Objective Loss 0.192027 LR 0.000250 Time 0.021215 -2023-02-13 18:28:14,435 - Epoch: [159][ 340/ 1207] Overall Loss 0.192068 Objective Loss 0.192068 LR 0.000250 Time 0.021154 -2023-02-13 18:28:14,628 - Epoch: [159][ 350/ 1207] Overall Loss 0.192094 Objective Loss 0.192094 LR 0.000250 Time 0.021098 -2023-02-13 18:28:14,820 - Epoch: [159][ 360/ 1207] Overall Loss 0.192019 Objective Loss 0.192019 LR 0.000250 Time 0.021045 -2023-02-13 18:28:15,011 - Epoch: [159][ 370/ 1207] Overall Loss 0.192051 Objective Loss 0.192051 LR 0.000250 Time 0.020991 -2023-02-13 18:28:15,202 - Epoch: [159][ 380/ 1207] Overall Loss 0.191879 Objective Loss 0.191879 LR 0.000250 Time 0.020942 -2023-02-13 18:28:15,394 - Epoch: [159][ 390/ 1207] Overall Loss 0.191826 Objective Loss 0.191826 LR 0.000250 Time 0.020895 -2023-02-13 18:28:15,585 - Epoch: [159][ 400/ 1207] Overall Loss 0.192628 Objective Loss 0.192628 LR 0.000250 Time 0.020849 -2023-02-13 18:28:15,776 - Epoch: [159][ 410/ 1207] Overall Loss 0.192855 Objective Loss 0.192855 LR 0.000250 Time 0.020805 -2023-02-13 18:28:15,969 - Epoch: [159][ 420/ 1207] Overall Loss 0.194380 Objective Loss 0.194380 LR 0.000250 Time 0.020768 -2023-02-13 18:28:16,159 - Epoch: [159][ 430/ 1207] Overall Loss 0.197508 Objective Loss 0.197508 LR 0.000250 Time 0.020728 -2023-02-13 18:28:16,350 - Epoch: [159][ 440/ 1207] Overall Loss 0.200724 Objective Loss 0.200724 LR 0.000250 Time 0.020689 -2023-02-13 18:28:16,541 - Epoch: [159][ 450/ 1207] Overall Loss 0.203843 Objective Loss 0.203843 LR 0.000250 Time 0.020654 -2023-02-13 18:28:16,733 - Epoch: [159][ 460/ 1207] Overall Loss 0.206295 Objective Loss 0.206295 LR 0.000250 Time 0.020621 -2023-02-13 18:28:16,923 - Epoch: [159][ 470/ 1207] Overall Loss 0.208926 Objective Loss 0.208926 LR 0.000250 Time 0.020586 -2023-02-13 18:28:17,113 - Epoch: [159][ 480/ 1207] Overall Loss 0.211269 Objective Loss 0.211269 LR 0.000250 Time 0.020551 -2023-02-13 18:28:17,303 - Epoch: [159][ 490/ 1207] Overall Loss 0.213467 Objective Loss 0.213467 LR 0.000250 Time 0.020518 -2023-02-13 18:28:17,492 - Epoch: [159][ 500/ 1207] Overall Loss 0.215761 Objective Loss 0.215761 LR 0.000250 Time 0.020486 -2023-02-13 18:28:17,683 - Epoch: [159][ 510/ 1207] Overall Loss 0.217167 Objective Loss 0.217167 LR 0.000250 Time 0.020457 -2023-02-13 18:28:17,873 - Epoch: [159][ 520/ 1207] Overall Loss 0.219329 Objective Loss 0.219329 LR 0.000250 Time 0.020430 -2023-02-13 18:28:18,063 - Epoch: [159][ 530/ 1207] Overall Loss 0.221100 Objective Loss 0.221100 LR 0.000250 Time 0.020401 -2023-02-13 18:28:18,252 - Epoch: [159][ 540/ 1207] Overall Loss 0.222565 Objective Loss 0.222565 LR 0.000250 Time 0.020374 -2023-02-13 18:28:18,442 - Epoch: [159][ 550/ 1207] Overall Loss 0.224178 Objective Loss 0.224178 LR 0.000250 Time 0.020348 -2023-02-13 18:28:18,632 - Epoch: [159][ 560/ 1207] Overall Loss 0.225461 Objective Loss 0.225461 LR 0.000250 Time 0.020322 -2023-02-13 18:28:18,822 - Epoch: [159][ 570/ 1207] Overall Loss 0.227605 Objective Loss 0.227605 LR 0.000250 Time 0.020299 -2023-02-13 18:28:19,012 - Epoch: [159][ 580/ 1207] Overall Loss 0.229253 Objective Loss 0.229253 LR 0.000250 Time 0.020276 -2023-02-13 18:28:19,202 - Epoch: [159][ 590/ 1207] Overall Loss 0.230448 Objective Loss 0.230448 LR 0.000250 Time 0.020253 -2023-02-13 18:28:19,391 - Epoch: [159][ 600/ 1207] Overall Loss 0.231583 Objective Loss 0.231583 LR 0.000250 Time 0.020231 -2023-02-13 18:28:19,581 - Epoch: [159][ 610/ 1207] Overall Loss 0.232659 Objective Loss 0.232659 LR 0.000250 Time 0.020211 -2023-02-13 18:28:19,771 - Epoch: [159][ 620/ 1207] Overall Loss 0.233821 Objective Loss 0.233821 LR 0.000250 Time 0.020189 -2023-02-13 18:28:19,961 - Epoch: [159][ 630/ 1207] Overall Loss 0.234567 Objective Loss 0.234567 LR 0.000250 Time 0.020171 -2023-02-13 18:28:20,151 - Epoch: [159][ 640/ 1207] Overall Loss 0.235696 Objective Loss 0.235696 LR 0.000250 Time 0.020151 -2023-02-13 18:28:20,341 - Epoch: [159][ 650/ 1207] Overall Loss 0.236887 Objective Loss 0.236887 LR 0.000250 Time 0.020133 -2023-02-13 18:28:20,530 - Epoch: [159][ 660/ 1207] Overall Loss 0.237675 Objective Loss 0.237675 LR 0.000250 Time 0.020115 -2023-02-13 18:28:20,722 - Epoch: [159][ 670/ 1207] Overall Loss 0.238456 Objective Loss 0.238456 LR 0.000250 Time 0.020099 -2023-02-13 18:28:20,913 - Epoch: [159][ 680/ 1207] Overall Loss 0.239330 Objective Loss 0.239330 LR 0.000250 Time 0.020084 -2023-02-13 18:28:21,103 - Epoch: [159][ 690/ 1207] Overall Loss 0.240210 Objective Loss 0.240210 LR 0.000250 Time 0.020068 -2023-02-13 18:28:21,293 - Epoch: [159][ 700/ 1207] Overall Loss 0.241181 Objective Loss 0.241181 LR 0.000250 Time 0.020052 -2023-02-13 18:28:21,483 - Epoch: [159][ 710/ 1207] Overall Loss 0.241862 Objective Loss 0.241862 LR 0.000250 Time 0.020037 -2023-02-13 18:28:21,673 - Epoch: [159][ 720/ 1207] Overall Loss 0.242571 Objective Loss 0.242571 LR 0.000250 Time 0.020022 -2023-02-13 18:28:21,864 - Epoch: [159][ 730/ 1207] Overall Loss 0.243125 Objective Loss 0.243125 LR 0.000250 Time 0.020008 -2023-02-13 18:28:22,054 - Epoch: [159][ 740/ 1207] Overall Loss 0.243872 Objective Loss 0.243872 LR 0.000250 Time 0.019995 -2023-02-13 18:28:22,244 - Epoch: [159][ 750/ 1207] Overall Loss 0.244409 Objective Loss 0.244409 LR 0.000250 Time 0.019981 -2023-02-13 18:28:22,433 - Epoch: [159][ 760/ 1207] Overall Loss 0.245011 Objective Loss 0.245011 LR 0.000250 Time 0.019967 -2023-02-13 18:28:22,623 - Epoch: [159][ 770/ 1207] Overall Loss 0.245625 Objective Loss 0.245625 LR 0.000250 Time 0.019954 -2023-02-13 18:28:22,813 - Epoch: [159][ 780/ 1207] Overall Loss 0.246072 Objective Loss 0.246072 LR 0.000250 Time 0.019941 -2023-02-13 18:28:23,003 - Epoch: [159][ 790/ 1207] Overall Loss 0.246797 Objective Loss 0.246797 LR 0.000250 Time 0.019929 -2023-02-13 18:28:23,194 - Epoch: [159][ 800/ 1207] Overall Loss 0.247232 Objective Loss 0.247232 LR 0.000250 Time 0.019917 -2023-02-13 18:28:23,383 - Epoch: [159][ 810/ 1207] Overall Loss 0.248008 Objective Loss 0.248008 LR 0.000250 Time 0.019905 -2023-02-13 18:28:23,572 - Epoch: [159][ 820/ 1207] Overall Loss 0.248392 Objective Loss 0.248392 LR 0.000250 Time 0.019892 -2023-02-13 18:28:23,762 - Epoch: [159][ 830/ 1207] Overall Loss 0.248934 Objective Loss 0.248934 LR 0.000250 Time 0.019880 -2023-02-13 18:28:23,952 - Epoch: [159][ 840/ 1207] Overall Loss 0.249404 Objective Loss 0.249404 LR 0.000250 Time 0.019869 -2023-02-13 18:28:24,141 - Epoch: [159][ 850/ 1207] Overall Loss 0.250005 Objective Loss 0.250005 LR 0.000250 Time 0.019858 -2023-02-13 18:28:24,331 - Epoch: [159][ 860/ 1207] Overall Loss 0.250558 Objective Loss 0.250558 LR 0.000250 Time 0.019848 -2023-02-13 18:28:24,521 - Epoch: [159][ 870/ 1207] Overall Loss 0.251072 Objective Loss 0.251072 LR 0.000250 Time 0.019837 -2023-02-13 18:28:24,711 - Epoch: [159][ 880/ 1207] Overall Loss 0.251805 Objective Loss 0.251805 LR 0.000250 Time 0.019827 -2023-02-13 18:28:24,901 - Epoch: [159][ 890/ 1207] Overall Loss 0.252143 Objective Loss 0.252143 LR 0.000250 Time 0.019818 -2023-02-13 18:28:25,091 - Epoch: [159][ 900/ 1207] Overall Loss 0.252525 Objective Loss 0.252525 LR 0.000250 Time 0.019808 -2023-02-13 18:28:25,280 - Epoch: [159][ 910/ 1207] Overall Loss 0.253283 Objective Loss 0.253283 LR 0.000250 Time 0.019798 -2023-02-13 18:28:25,470 - Epoch: [159][ 920/ 1207] Overall Loss 0.253647 Objective Loss 0.253647 LR 0.000250 Time 0.019789 -2023-02-13 18:28:25,660 - Epoch: [159][ 930/ 1207] Overall Loss 0.254101 Objective Loss 0.254101 LR 0.000250 Time 0.019779 -2023-02-13 18:28:25,849 - Epoch: [159][ 940/ 1207] Overall Loss 0.254464 Objective Loss 0.254464 LR 0.000250 Time 0.019771 -2023-02-13 18:28:26,040 - Epoch: [159][ 950/ 1207] Overall Loss 0.254797 Objective Loss 0.254797 LR 0.000250 Time 0.019763 -2023-02-13 18:28:26,230 - Epoch: [159][ 960/ 1207] Overall Loss 0.255131 Objective Loss 0.255131 LR 0.000250 Time 0.019754 -2023-02-13 18:28:26,419 - Epoch: [159][ 970/ 1207] Overall Loss 0.255463 Objective Loss 0.255463 LR 0.000250 Time 0.019745 -2023-02-13 18:28:26,610 - Epoch: [159][ 980/ 1207] Overall Loss 0.255674 Objective Loss 0.255674 LR 0.000250 Time 0.019738 -2023-02-13 18:28:26,799 - Epoch: [159][ 990/ 1207] Overall Loss 0.255913 Objective Loss 0.255913 LR 0.000250 Time 0.019729 -2023-02-13 18:28:26,990 - Epoch: [159][ 1000/ 1207] Overall Loss 0.256204 Objective Loss 0.256204 LR 0.000250 Time 0.019723 -2023-02-13 18:28:27,179 - Epoch: [159][ 1010/ 1207] Overall Loss 0.256551 Objective Loss 0.256551 LR 0.000250 Time 0.019714 -2023-02-13 18:28:27,369 - Epoch: [159][ 1020/ 1207] Overall Loss 0.257077 Objective Loss 0.257077 LR 0.000250 Time 0.019706 -2023-02-13 18:28:27,558 - Epoch: [159][ 1030/ 1207] Overall Loss 0.257252 Objective Loss 0.257252 LR 0.000250 Time 0.019699 -2023-02-13 18:28:27,747 - Epoch: [159][ 1040/ 1207] Overall Loss 0.257424 Objective Loss 0.257424 LR 0.000250 Time 0.019690 -2023-02-13 18:28:27,938 - Epoch: [159][ 1050/ 1207] Overall Loss 0.257723 Objective Loss 0.257723 LR 0.000250 Time 0.019684 -2023-02-13 18:28:28,127 - Epoch: [159][ 1060/ 1207] Overall Loss 0.258071 Objective Loss 0.258071 LR 0.000250 Time 0.019677 -2023-02-13 18:28:28,316 - Epoch: [159][ 1070/ 1207] Overall Loss 0.258205 Objective Loss 0.258205 LR 0.000250 Time 0.019669 -2023-02-13 18:28:28,505 - Epoch: [159][ 1080/ 1207] Overall Loss 0.258268 Objective Loss 0.258268 LR 0.000250 Time 0.019662 -2023-02-13 18:28:28,695 - Epoch: [159][ 1090/ 1207] Overall Loss 0.258636 Objective Loss 0.258636 LR 0.000250 Time 0.019655 -2023-02-13 18:28:28,884 - Epoch: [159][ 1100/ 1207] Overall Loss 0.258993 Objective Loss 0.258993 LR 0.000250 Time 0.019648 -2023-02-13 18:28:29,074 - Epoch: [159][ 1110/ 1207] Overall Loss 0.259026 Objective Loss 0.259026 LR 0.000250 Time 0.019642 -2023-02-13 18:28:29,264 - Epoch: [159][ 1120/ 1207] Overall Loss 0.259362 Objective Loss 0.259362 LR 0.000250 Time 0.019636 -2023-02-13 18:28:29,453 - Epoch: [159][ 1130/ 1207] Overall Loss 0.259643 Objective Loss 0.259643 LR 0.000250 Time 0.019629 -2023-02-13 18:28:29,643 - Epoch: [159][ 1140/ 1207] Overall Loss 0.259844 Objective Loss 0.259844 LR 0.000250 Time 0.019623 -2023-02-13 18:28:29,833 - Epoch: [159][ 1150/ 1207] Overall Loss 0.259893 Objective Loss 0.259893 LR 0.000250 Time 0.019617 -2023-02-13 18:28:30,023 - Epoch: [159][ 1160/ 1207] Overall Loss 0.260138 Objective Loss 0.260138 LR 0.000250 Time 0.019612 -2023-02-13 18:28:30,213 - Epoch: [159][ 1170/ 1207] Overall Loss 0.260187 Objective Loss 0.260187 LR 0.000250 Time 0.019606 -2023-02-13 18:28:30,403 - Epoch: [159][ 1180/ 1207] Overall Loss 0.260280 Objective Loss 0.260280 LR 0.000250 Time 0.019601 -2023-02-13 18:28:30,593 - Epoch: [159][ 1190/ 1207] Overall Loss 0.260401 Objective Loss 0.260401 LR 0.000250 Time 0.019595 -2023-02-13 18:28:30,832 - Epoch: [159][ 1200/ 1207] Overall Loss 0.260619 Objective Loss 0.260619 LR 0.000250 Time 0.019631 -2023-02-13 18:28:30,949 - Epoch: [159][ 1207/ 1207] Overall Loss 0.260817 Objective Loss 0.260817 Top1 85.365854 Top5 97.865854 LR 0.000250 Time 0.019614 -2023-02-13 18:28:31,022 - --- validate (epoch=159)----------- -2023-02-13 18:28:31,022 - 34311 samples (256 per mini-batch) -2023-02-13 18:28:31,429 - Epoch: [159][ 10/ 135] Loss 0.327283 Top1 84.726562 Top5 98.203125 -2023-02-13 18:28:31,552 - Epoch: [159][ 20/ 135] Loss 0.333743 Top1 83.984375 Top5 97.890625 -2023-02-13 18:28:31,674 - Epoch: [159][ 30/ 135] Loss 0.333596 Top1 84.205729 Top5 97.773438 -2023-02-13 18:28:31,816 - Epoch: [159][ 40/ 135] Loss 0.335187 Top1 84.062500 Top5 97.802734 -2023-02-13 18:28:31,954 - Epoch: [159][ 50/ 135] Loss 0.339283 Top1 84.179688 Top5 97.820312 -2023-02-13 18:28:32,093 - Epoch: [159][ 60/ 135] Loss 0.339889 Top1 84.075521 Top5 97.766927 -2023-02-13 18:28:32,228 - Epoch: [159][ 70/ 135] Loss 0.341701 Top1 83.967634 Top5 97.739955 -2023-02-13 18:28:32,368 - Epoch: [159][ 80/ 135] Loss 0.342429 Top1 83.798828 Top5 97.763672 -2023-02-13 18:28:32,502 - Epoch: [159][ 90/ 135] Loss 0.342119 Top1 83.806424 Top5 97.734375 -2023-02-13 18:28:32,628 - Epoch: [159][ 100/ 135] Loss 0.340279 Top1 83.871094 Top5 97.734375 -2023-02-13 18:28:32,754 - Epoch: [159][ 110/ 135] Loss 0.339962 Top1 83.842330 Top5 97.720170 -2023-02-13 18:28:32,882 - Epoch: [159][ 120/ 135] Loss 0.340436 Top1 83.805339 Top5 97.721354 -2023-02-13 18:28:33,007 - Epoch: [159][ 130/ 135] Loss 0.338104 Top1 83.858173 Top5 97.770433 -2023-02-13 18:28:33,051 - Epoch: [159][ 135/ 135] Loss 0.336607 Top1 83.891463 Top5 97.767480 -2023-02-13 18:28:33,119 - ==> Top1: 83.891 Top5: 97.767 Loss: 0.337 - -2023-02-13 18:28:33,120 - ==> Confusion: -[[ 862 2 7 0 9 5 0 3 4 45 0 5 1 5 7 4 4 1 0 1 2] - [ 4 948 1 1 7 21 3 19 5 0 1 2 0 0 1 2 8 0 3 3 4] - [ 7 3 965 10 6 2 18 12 0 2 2 1 4 4 4 2 4 5 3 1 3] - [ 5 1 18 897 3 3 1 2 3 2 15 0 7 0 26 0 4 7 16 0 6] - [ 14 8 1 0 989 10 1 0 3 3 0 8 1 2 9 6 2 1 2 2 4] - [ 1 20 0 4 5 971 3 12 1 4 2 9 3 17 2 1 4 0 2 4 5] - [ 2 2 16 2 1 5 1037 6 0 1 1 2 2 1 0 3 4 4 1 5 4] - [ 3 7 9 1 2 27 3 939 0 2 1 8 2 1 0 0 1 0 10 7 1] - [ 15 3 1 1 1 0 1 1 917 31 5 3 0 13 11 1 1 0 3 0 1] - [ 75 0 3 0 6 2 0 2 29 862 1 2 2 18 3 1 0 2 1 0 3] - [ 3 1 4 6 2 1 3 6 16 2 983 2 2 11 2 0 1 1 4 0 1] - [ 1 2 2 0 0 9 0 4 2 1 1 929 18 8 4 5 3 10 1 4 1] - [ 0 0 2 8 2 5 0 2 3 0 0 33 867 2 2 7 2 18 3 0 3] - [ 4 1 3 0 5 9 0 3 8 13 6 6 2 951 2 4 2 3 0 1 1] - [ 6 1 2 10 3 3 0 0 20 7 4 0 4 5 1007 1 3 5 5 0 6] - [ 4 2 8 0 4 1 4 1 0 0 0 8 3 4 0 977 8 11 0 6 5] - [ 1 5 1 0 9 3 0 0 3 0 0 3 2 3 1 15 1000 3 2 2 8] - [ 4 1 0 3 0 2 2 0 1 0 0 9 15 1 0 18 0 986 0 1 8] - [ 3 4 3 7 1 2 0 29 5 0 3 0 5 0 16 0 1 4 999 2 2] - [ 1 4 1 0 0 7 6 8 1 0 1 17 3 4 1 4 5 5 1 1074 5] - [ 187 246 249 99 156 216 102 207 108 87 201 120 353 362 176 109 278 113 170 271 9624]] - -2023-02-13 18:28:33,121 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:28:33,121 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:28:33,127 - - -2023-02-13 18:28:33,127 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:28:34,003 - Epoch: [160][ 10/ 1207] Overall Loss 0.273608 Objective Loss 0.273608 LR 0.000250 Time 0.087541 -2023-02-13 18:28:34,202 - Epoch: [160][ 20/ 1207] Overall Loss 0.278701 Objective Loss 0.278701 LR 0.000250 Time 0.053694 -2023-02-13 18:28:34,393 - Epoch: [160][ 30/ 1207] Overall Loss 0.282300 Objective Loss 0.282300 LR 0.000250 Time 0.042137 -2023-02-13 18:28:34,582 - Epoch: [160][ 40/ 1207] Overall Loss 0.282033 Objective Loss 0.282033 LR 0.000250 Time 0.036321 -2023-02-13 18:28:34,770 - Epoch: [160][ 50/ 1207] Overall Loss 0.280096 Objective Loss 0.280096 LR 0.000250 Time 0.032818 -2023-02-13 18:28:34,960 - Epoch: [160][ 60/ 1207] Overall Loss 0.278023 Objective Loss 0.278023 LR 0.000250 Time 0.030496 -2023-02-13 18:28:35,148 - Epoch: [160][ 70/ 1207] Overall Loss 0.274842 Objective Loss 0.274842 LR 0.000250 Time 0.028822 -2023-02-13 18:28:35,337 - Epoch: [160][ 80/ 1207] Overall Loss 0.274627 Objective Loss 0.274627 LR 0.000250 Time 0.027580 -2023-02-13 18:28:35,525 - Epoch: [160][ 90/ 1207] Overall Loss 0.277916 Objective Loss 0.277916 LR 0.000250 Time 0.026598 -2023-02-13 18:28:35,714 - Epoch: [160][ 100/ 1207] Overall Loss 0.277380 Objective Loss 0.277380 LR 0.000250 Time 0.025830 -2023-02-13 18:28:35,903 - Epoch: [160][ 110/ 1207] Overall Loss 0.276319 Objective Loss 0.276319 LR 0.000250 Time 0.025195 -2023-02-13 18:28:36,092 - Epoch: [160][ 120/ 1207] Overall Loss 0.275601 Objective Loss 0.275601 LR 0.000250 Time 0.024662 -2023-02-13 18:28:36,280 - Epoch: [160][ 130/ 1207] Overall Loss 0.275791 Objective Loss 0.275791 LR 0.000250 Time 0.024211 -2023-02-13 18:28:36,468 - Epoch: [160][ 140/ 1207] Overall Loss 0.275031 Objective Loss 0.275031 LR 0.000250 Time 0.023825 -2023-02-13 18:28:36,657 - Epoch: [160][ 150/ 1207] Overall Loss 0.275769 Objective Loss 0.275769 LR 0.000250 Time 0.023490 -2023-02-13 18:28:36,844 - Epoch: [160][ 160/ 1207] Overall Loss 0.274755 Objective Loss 0.274755 LR 0.000250 Time 0.023191 -2023-02-13 18:28:37,033 - Epoch: [160][ 170/ 1207] Overall Loss 0.273864 Objective Loss 0.273864 LR 0.000250 Time 0.022937 -2023-02-13 18:28:37,221 - Epoch: [160][ 180/ 1207] Overall Loss 0.273935 Objective Loss 0.273935 LR 0.000250 Time 0.022706 -2023-02-13 18:28:37,409 - Epoch: [160][ 190/ 1207] Overall Loss 0.273731 Objective Loss 0.273731 LR 0.000250 Time 0.022499 -2023-02-13 18:28:37,598 - Epoch: [160][ 200/ 1207] Overall Loss 0.273766 Objective Loss 0.273766 LR 0.000250 Time 0.022315 -2023-02-13 18:28:37,787 - Epoch: [160][ 210/ 1207] Overall Loss 0.273306 Objective Loss 0.273306 LR 0.000250 Time 0.022148 -2023-02-13 18:28:37,977 - Epoch: [160][ 220/ 1207] Overall Loss 0.273892 Objective Loss 0.273892 LR 0.000250 Time 0.022004 -2023-02-13 18:28:38,165 - Epoch: [160][ 230/ 1207] Overall Loss 0.273758 Objective Loss 0.273758 LR 0.000250 Time 0.021863 -2023-02-13 18:28:38,354 - Epoch: [160][ 240/ 1207] Overall Loss 0.273743 Objective Loss 0.273743 LR 0.000250 Time 0.021738 -2023-02-13 18:28:38,542 - Epoch: [160][ 250/ 1207] Overall Loss 0.273777 Objective Loss 0.273777 LR 0.000250 Time 0.021620 -2023-02-13 18:28:38,731 - Epoch: [160][ 260/ 1207] Overall Loss 0.272754 Objective Loss 0.272754 LR 0.000250 Time 0.021515 -2023-02-13 18:28:38,920 - Epoch: [160][ 270/ 1207] Overall Loss 0.273005 Objective Loss 0.273005 LR 0.000250 Time 0.021416 -2023-02-13 18:28:39,109 - Epoch: [160][ 280/ 1207] Overall Loss 0.273067 Objective Loss 0.273067 LR 0.000250 Time 0.021324 -2023-02-13 18:28:39,297 - Epoch: [160][ 290/ 1207] Overall Loss 0.273625 Objective Loss 0.273625 LR 0.000250 Time 0.021236 -2023-02-13 18:28:39,486 - Epoch: [160][ 300/ 1207] Overall Loss 0.273539 Objective Loss 0.273539 LR 0.000250 Time 0.021158 -2023-02-13 18:28:39,674 - Epoch: [160][ 310/ 1207] Overall Loss 0.273609 Objective Loss 0.273609 LR 0.000250 Time 0.021080 -2023-02-13 18:28:39,863 - Epoch: [160][ 320/ 1207] Overall Loss 0.273201 Objective Loss 0.273201 LR 0.000250 Time 0.021010 -2023-02-13 18:28:40,051 - Epoch: [160][ 330/ 1207] Overall Loss 0.273157 Objective Loss 0.273157 LR 0.000250 Time 0.020944 -2023-02-13 18:28:40,239 - Epoch: [160][ 340/ 1207] Overall Loss 0.273779 Objective Loss 0.273779 LR 0.000250 Time 0.020879 -2023-02-13 18:28:40,428 - Epoch: [160][ 350/ 1207] Overall Loss 0.273263 Objective Loss 0.273263 LR 0.000250 Time 0.020820 -2023-02-13 18:28:40,616 - Epoch: [160][ 360/ 1207] Overall Loss 0.273088 Objective Loss 0.273088 LR 0.000250 Time 0.020765 -2023-02-13 18:28:40,804 - Epoch: [160][ 370/ 1207] Overall Loss 0.273242 Objective Loss 0.273242 LR 0.000250 Time 0.020710 -2023-02-13 18:28:40,994 - Epoch: [160][ 380/ 1207] Overall Loss 0.273604 Objective Loss 0.273604 LR 0.000250 Time 0.020664 -2023-02-13 18:28:41,184 - Epoch: [160][ 390/ 1207] Overall Loss 0.273727 Objective Loss 0.273727 LR 0.000250 Time 0.020619 -2023-02-13 18:28:41,372 - Epoch: [160][ 400/ 1207] Overall Loss 0.273827 Objective Loss 0.273827 LR 0.000250 Time 0.020573 -2023-02-13 18:28:41,560 - Epoch: [160][ 410/ 1207] Overall Loss 0.273786 Objective Loss 0.273786 LR 0.000250 Time 0.020530 -2023-02-13 18:28:41,750 - Epoch: [160][ 420/ 1207] Overall Loss 0.273634 Objective Loss 0.273634 LR 0.000250 Time 0.020493 -2023-02-13 18:28:41,939 - Epoch: [160][ 430/ 1207] Overall Loss 0.274401 Objective Loss 0.274401 LR 0.000250 Time 0.020454 -2023-02-13 18:28:42,128 - Epoch: [160][ 440/ 1207] Overall Loss 0.274443 Objective Loss 0.274443 LR 0.000250 Time 0.020418 -2023-02-13 18:28:42,316 - Epoch: [160][ 450/ 1207] Overall Loss 0.274024 Objective Loss 0.274024 LR 0.000250 Time 0.020382 -2023-02-13 18:28:42,505 - Epoch: [160][ 460/ 1207] Overall Loss 0.274488 Objective Loss 0.274488 LR 0.000250 Time 0.020348 -2023-02-13 18:28:42,695 - Epoch: [160][ 470/ 1207] Overall Loss 0.274575 Objective Loss 0.274575 LR 0.000250 Time 0.020319 -2023-02-13 18:28:42,884 - Epoch: [160][ 480/ 1207] Overall Loss 0.274619 Objective Loss 0.274619 LR 0.000250 Time 0.020288 -2023-02-13 18:28:43,073 - Epoch: [160][ 490/ 1207] Overall Loss 0.274957 Objective Loss 0.274957 LR 0.000250 Time 0.020260 -2023-02-13 18:28:43,262 - Epoch: [160][ 500/ 1207] Overall Loss 0.274796 Objective Loss 0.274796 LR 0.000250 Time 0.020231 -2023-02-13 18:28:43,450 - Epoch: [160][ 510/ 1207] Overall Loss 0.274578 Objective Loss 0.274578 LR 0.000250 Time 0.020203 -2023-02-13 18:28:43,639 - Epoch: [160][ 520/ 1207] Overall Loss 0.274318 Objective Loss 0.274318 LR 0.000250 Time 0.020176 -2023-02-13 18:28:43,827 - Epoch: [160][ 530/ 1207] Overall Loss 0.274103 Objective Loss 0.274103 LR 0.000250 Time 0.020150 -2023-02-13 18:28:44,017 - Epoch: [160][ 540/ 1207] Overall Loss 0.273875 Objective Loss 0.273875 LR 0.000250 Time 0.020128 -2023-02-13 18:28:44,206 - Epoch: [160][ 550/ 1207] Overall Loss 0.273793 Objective Loss 0.273793 LR 0.000250 Time 0.020105 -2023-02-13 18:28:44,394 - Epoch: [160][ 560/ 1207] Overall Loss 0.273917 Objective Loss 0.273917 LR 0.000250 Time 0.020082 -2023-02-13 18:28:44,582 - Epoch: [160][ 570/ 1207] Overall Loss 0.273596 Objective Loss 0.273596 LR 0.000250 Time 0.020058 -2023-02-13 18:28:44,771 - Epoch: [160][ 580/ 1207] Overall Loss 0.274009 Objective Loss 0.274009 LR 0.000250 Time 0.020037 -2023-02-13 18:28:44,960 - Epoch: [160][ 590/ 1207] Overall Loss 0.274406 Objective Loss 0.274406 LR 0.000250 Time 0.020018 -2023-02-13 18:28:45,148 - Epoch: [160][ 600/ 1207] Overall Loss 0.274224 Objective Loss 0.274224 LR 0.000250 Time 0.019997 -2023-02-13 18:28:45,337 - Epoch: [160][ 610/ 1207] Overall Loss 0.274336 Objective Loss 0.274336 LR 0.000250 Time 0.019978 -2023-02-13 18:28:45,525 - Epoch: [160][ 620/ 1207] Overall Loss 0.274425 Objective Loss 0.274425 LR 0.000250 Time 0.019959 -2023-02-13 18:28:45,713 - Epoch: [160][ 630/ 1207] Overall Loss 0.274647 Objective Loss 0.274647 LR 0.000250 Time 0.019940 -2023-02-13 18:28:45,904 - Epoch: [160][ 640/ 1207] Overall Loss 0.274246 Objective Loss 0.274246 LR 0.000250 Time 0.019926 -2023-02-13 18:28:46,093 - Epoch: [160][ 650/ 1207] Overall Loss 0.274242 Objective Loss 0.274242 LR 0.000250 Time 0.019909 -2023-02-13 18:28:46,281 - Epoch: [160][ 660/ 1207] Overall Loss 0.273751 Objective Loss 0.273751 LR 0.000250 Time 0.019893 -2023-02-13 18:28:46,469 - Epoch: [160][ 670/ 1207] Overall Loss 0.273879 Objective Loss 0.273879 LR 0.000250 Time 0.019876 -2023-02-13 18:28:46,658 - Epoch: [160][ 680/ 1207] Overall Loss 0.274125 Objective Loss 0.274125 LR 0.000250 Time 0.019860 -2023-02-13 18:28:46,847 - Epoch: [160][ 690/ 1207] Overall Loss 0.274002 Objective Loss 0.274002 LR 0.000250 Time 0.019846 -2023-02-13 18:28:47,036 - Epoch: [160][ 700/ 1207] Overall Loss 0.273788 Objective Loss 0.273788 LR 0.000250 Time 0.019832 -2023-02-13 18:28:47,224 - Epoch: [160][ 710/ 1207] Overall Loss 0.273741 Objective Loss 0.273741 LR 0.000250 Time 0.019818 -2023-02-13 18:28:47,413 - Epoch: [160][ 720/ 1207] Overall Loss 0.274073 Objective Loss 0.274073 LR 0.000250 Time 0.019803 -2023-02-13 18:28:47,601 - Epoch: [160][ 730/ 1207] Overall Loss 0.273950 Objective Loss 0.273950 LR 0.000250 Time 0.019789 -2023-02-13 18:28:47,790 - Epoch: [160][ 740/ 1207] Overall Loss 0.274166 Objective Loss 0.274166 LR 0.000250 Time 0.019777 -2023-02-13 18:28:47,979 - Epoch: [160][ 750/ 1207] Overall Loss 0.274088 Objective Loss 0.274088 LR 0.000250 Time 0.019764 -2023-02-13 18:28:48,167 - Epoch: [160][ 760/ 1207] Overall Loss 0.274131 Objective Loss 0.274131 LR 0.000250 Time 0.019752 -2023-02-13 18:28:48,355 - Epoch: [160][ 770/ 1207] Overall Loss 0.274016 Objective Loss 0.274016 LR 0.000250 Time 0.019739 -2023-02-13 18:28:48,544 - Epoch: [160][ 780/ 1207] Overall Loss 0.274001 Objective Loss 0.274001 LR 0.000250 Time 0.019728 -2023-02-13 18:28:48,732 - Epoch: [160][ 790/ 1207] Overall Loss 0.274336 Objective Loss 0.274336 LR 0.000250 Time 0.019715 -2023-02-13 18:28:48,921 - Epoch: [160][ 800/ 1207] Overall Loss 0.274483 Objective Loss 0.274483 LR 0.000250 Time 0.019705 -2023-02-13 18:28:49,110 - Epoch: [160][ 810/ 1207] Overall Loss 0.274414 Objective Loss 0.274414 LR 0.000250 Time 0.019694 -2023-02-13 18:28:49,298 - Epoch: [160][ 820/ 1207] Overall Loss 0.274278 Objective Loss 0.274278 LR 0.000250 Time 0.019683 -2023-02-13 18:28:49,486 - Epoch: [160][ 830/ 1207] Overall Loss 0.274266 Objective Loss 0.274266 LR 0.000250 Time 0.019672 -2023-02-13 18:28:49,674 - Epoch: [160][ 840/ 1207] Overall Loss 0.274309 Objective Loss 0.274309 LR 0.000250 Time 0.019662 -2023-02-13 18:28:49,863 - Epoch: [160][ 850/ 1207] Overall Loss 0.274621 Objective Loss 0.274621 LR 0.000250 Time 0.019651 -2023-02-13 18:28:50,052 - Epoch: [160][ 860/ 1207] Overall Loss 0.274347 Objective Loss 0.274347 LR 0.000250 Time 0.019643 -2023-02-13 18:28:50,240 - Epoch: [160][ 870/ 1207] Overall Loss 0.274493 Objective Loss 0.274493 LR 0.000250 Time 0.019632 -2023-02-13 18:28:50,429 - Epoch: [160][ 880/ 1207] Overall Loss 0.274385 Objective Loss 0.274385 LR 0.000250 Time 0.019624 -2023-02-13 18:28:50,618 - Epoch: [160][ 890/ 1207] Overall Loss 0.274368 Objective Loss 0.274368 LR 0.000250 Time 0.019615 -2023-02-13 18:28:50,807 - Epoch: [160][ 900/ 1207] Overall Loss 0.274425 Objective Loss 0.274425 LR 0.000250 Time 0.019607 -2023-02-13 18:28:50,997 - Epoch: [160][ 910/ 1207] Overall Loss 0.274272 Objective Loss 0.274272 LR 0.000250 Time 0.019600 -2023-02-13 18:28:51,186 - Epoch: [160][ 920/ 1207] Overall Loss 0.274102 Objective Loss 0.274102 LR 0.000250 Time 0.019591 -2023-02-13 18:28:51,374 - Epoch: [160][ 930/ 1207] Overall Loss 0.274221 Objective Loss 0.274221 LR 0.000250 Time 0.019583 -2023-02-13 18:28:51,562 - Epoch: [160][ 940/ 1207] Overall Loss 0.274163 Objective Loss 0.274163 LR 0.000250 Time 0.019575 -2023-02-13 18:28:51,752 - Epoch: [160][ 950/ 1207] Overall Loss 0.274096 Objective Loss 0.274096 LR 0.000250 Time 0.019568 -2023-02-13 18:28:51,942 - Epoch: [160][ 960/ 1207] Overall Loss 0.273978 Objective Loss 0.273978 LR 0.000250 Time 0.019561 -2023-02-13 18:28:52,131 - Epoch: [160][ 970/ 1207] Overall Loss 0.273969 Objective Loss 0.273969 LR 0.000250 Time 0.019554 -2023-02-13 18:28:52,319 - Epoch: [160][ 980/ 1207] Overall Loss 0.273694 Objective Loss 0.273694 LR 0.000250 Time 0.019546 -2023-02-13 18:28:52,509 - Epoch: [160][ 990/ 1207] Overall Loss 0.273616 Objective Loss 0.273616 LR 0.000250 Time 0.019540 -2023-02-13 18:28:52,698 - Epoch: [160][ 1000/ 1207] Overall Loss 0.273592 Objective Loss 0.273592 LR 0.000250 Time 0.019533 -2023-02-13 18:28:52,886 - Epoch: [160][ 1010/ 1207] Overall Loss 0.273465 Objective Loss 0.273465 LR 0.000250 Time 0.019526 -2023-02-13 18:28:53,075 - Epoch: [160][ 1020/ 1207] Overall Loss 0.273555 Objective Loss 0.273555 LR 0.000250 Time 0.019519 -2023-02-13 18:28:53,264 - Epoch: [160][ 1030/ 1207] Overall Loss 0.273579 Objective Loss 0.273579 LR 0.000250 Time 0.019513 -2023-02-13 18:28:53,454 - Epoch: [160][ 1040/ 1207] Overall Loss 0.273813 Objective Loss 0.273813 LR 0.000250 Time 0.019507 -2023-02-13 18:28:53,643 - Epoch: [160][ 1050/ 1207] Overall Loss 0.273700 Objective Loss 0.273700 LR 0.000250 Time 0.019502 -2023-02-13 18:28:53,832 - Epoch: [160][ 1060/ 1207] Overall Loss 0.273928 Objective Loss 0.273928 LR 0.000250 Time 0.019496 -2023-02-13 18:28:54,021 - Epoch: [160][ 1070/ 1207] Overall Loss 0.273903 Objective Loss 0.273903 LR 0.000250 Time 0.019490 -2023-02-13 18:28:54,211 - Epoch: [160][ 1080/ 1207] Overall Loss 0.273801 Objective Loss 0.273801 LR 0.000250 Time 0.019484 -2023-02-13 18:28:54,400 - Epoch: [160][ 1090/ 1207] Overall Loss 0.273794 Objective Loss 0.273794 LR 0.000250 Time 0.019479 -2023-02-13 18:28:54,589 - Epoch: [160][ 1100/ 1207] Overall Loss 0.273765 Objective Loss 0.273765 LR 0.000250 Time 0.019473 -2023-02-13 18:28:54,778 - Epoch: [160][ 1110/ 1207] Overall Loss 0.273702 Objective Loss 0.273702 LR 0.000250 Time 0.019468 -2023-02-13 18:28:54,967 - Epoch: [160][ 1120/ 1207] Overall Loss 0.273741 Objective Loss 0.273741 LR 0.000250 Time 0.019463 -2023-02-13 18:28:55,157 - Epoch: [160][ 1130/ 1207] Overall Loss 0.273747 Objective Loss 0.273747 LR 0.000250 Time 0.019458 -2023-02-13 18:28:55,347 - Epoch: [160][ 1140/ 1207] Overall Loss 0.273844 Objective Loss 0.273844 LR 0.000250 Time 0.019453 -2023-02-13 18:28:55,536 - Epoch: [160][ 1150/ 1207] Overall Loss 0.273982 Objective Loss 0.273982 LR 0.000250 Time 0.019449 -2023-02-13 18:28:55,726 - Epoch: [160][ 1160/ 1207] Overall Loss 0.274102 Objective Loss 0.274102 LR 0.000250 Time 0.019444 -2023-02-13 18:28:55,917 - Epoch: [160][ 1170/ 1207] Overall Loss 0.274075 Objective Loss 0.274075 LR 0.000250 Time 0.019441 -2023-02-13 18:28:56,107 - Epoch: [160][ 1180/ 1207] Overall Loss 0.274125 Objective Loss 0.274125 LR 0.000250 Time 0.019437 -2023-02-13 18:28:56,297 - Epoch: [160][ 1190/ 1207] Overall Loss 0.274134 Objective Loss 0.274134 LR 0.000250 Time 0.019433 -2023-02-13 18:28:56,544 - Epoch: [160][ 1200/ 1207] Overall Loss 0.274085 Objective Loss 0.274085 LR 0.000250 Time 0.019476 -2023-02-13 18:28:56,660 - Epoch: [160][ 1207/ 1207] Overall Loss 0.273965 Objective Loss 0.273965 Top1 87.500000 Top5 98.475610 LR 0.000250 Time 0.019460 -2023-02-13 18:28:56,738 - --- validate (epoch=160)----------- -2023-02-13 18:28:56,738 - 34311 samples (256 per mini-batch) -2023-02-13 18:28:57,250 - Epoch: [160][ 10/ 135] Loss 0.324854 Top1 84.140625 Top5 97.578125 -2023-02-13 18:28:57,377 - Epoch: [160][ 20/ 135] Loss 0.332302 Top1 83.613281 Top5 97.656250 -2023-02-13 18:28:57,501 - Epoch: [160][ 30/ 135] Loss 0.327497 Top1 83.580729 Top5 97.773438 -2023-02-13 18:28:57,626 - Epoch: [160][ 40/ 135] Loss 0.328727 Top1 83.476562 Top5 97.705078 -2023-02-13 18:28:57,751 - Epoch: [160][ 50/ 135] Loss 0.330671 Top1 83.523438 Top5 97.742188 -2023-02-13 18:28:57,875 - Epoch: [160][ 60/ 135] Loss 0.327660 Top1 83.873698 Top5 97.708333 -2023-02-13 18:28:58,002 - Epoch: [160][ 70/ 135] Loss 0.328167 Top1 83.861607 Top5 97.667411 -2023-02-13 18:28:58,128 - Epoch: [160][ 80/ 135] Loss 0.327855 Top1 83.886719 Top5 97.670898 -2023-02-13 18:28:58,256 - Epoch: [160][ 90/ 135] Loss 0.331765 Top1 83.884549 Top5 97.612847 -2023-02-13 18:28:58,384 - Epoch: [160][ 100/ 135] Loss 0.331466 Top1 83.882812 Top5 97.589844 -2023-02-13 18:28:58,511 - Epoch: [160][ 110/ 135] Loss 0.331821 Top1 83.831676 Top5 97.595881 -2023-02-13 18:28:58,640 - Epoch: [160][ 120/ 135] Loss 0.331192 Top1 83.779297 Top5 97.591146 -2023-02-13 18:28:58,770 - Epoch: [160][ 130/ 135] Loss 0.328227 Top1 83.921274 Top5 97.608173 -2023-02-13 18:28:58,818 - Epoch: [160][ 135/ 135] Loss 0.328633 Top1 83.865233 Top5 97.592609 -2023-02-13 18:28:58,897 - ==> Top1: 83.865 Top5: 97.593 Loss: 0.329 - -2023-02-13 18:28:58,897 - ==> Confusion: -[[ 863 8 7 0 9 1 0 0 1 48 0 3 0 4 7 4 3 2 1 2 4] - [ 3 957 2 3 10 20 2 14 2 1 1 0 0 0 0 2 6 1 4 1 4] - [ 6 4 974 12 4 1 14 13 0 2 2 1 2 2 4 3 3 3 3 1 4] - [ 4 0 22 910 1 4 1 1 2 3 15 2 7 0 15 2 3 5 15 0 4] - [ 9 8 1 1 996 9 2 1 0 2 0 5 2 2 6 7 5 3 0 4 3] - [ 3 22 2 5 5 964 4 14 1 6 1 9 3 15 0 2 3 2 2 3 4] - [ 2 2 15 2 1 6 1048 4 0 2 1 3 1 1 0 2 2 0 1 2 4] - [ 4 14 13 1 2 27 3 916 0 1 1 6 4 0 0 0 1 2 16 9 4] - [ 20 2 0 1 2 0 0 1 888 43 7 3 0 10 22 1 1 1 5 0 2] - [ 75 0 2 0 12 1 0 3 26 859 0 1 0 20 4 2 1 1 2 0 3] - [ 2 2 3 8 1 1 3 4 15 2 981 2 1 8 2 1 2 1 10 0 2] - [ 4 2 1 1 0 13 0 4 1 3 1 923 17 7 1 7 3 8 1 6 2] - [ 1 1 1 9 1 4 0 0 2 0 1 29 865 0 3 7 5 22 2 0 6] - [ 5 1 5 0 7 10 0 2 9 13 10 6 2 935 4 6 4 3 0 0 2] - [ 8 4 1 17 6 4 0 1 13 6 3 1 1 1 1005 2 3 4 4 0 8] - [ 1 2 9 1 8 4 3 0 1 1 0 4 5 2 0 972 10 10 0 7 6] - [ 2 4 1 3 6 1 0 0 1 0 0 1 3 3 1 12 1012 2 0 2 7] - [ 2 2 0 5 1 2 2 0 0 0 1 8 9 1 0 17 0 995 0 2 4] - [ 3 7 6 8 1 2 0 24 3 0 4 1 1 1 14 0 0 3 1005 2 1] - [ 1 2 0 0 0 9 7 7 1 0 0 18 1 2 0 8 8 5 1 1073 5] - [ 167 276 292 127 194 202 112 157 93 92 205 115 329 296 154 103 328 111 182 265 9634]] - -2023-02-13 18:28:58,899 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:28:58,899 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:28:58,905 - - -2023-02-13 18:28:58,905 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:28:59,796 - Epoch: [161][ 10/ 1207] Overall Loss 0.264335 Objective Loss 0.264335 LR 0.000250 Time 0.089095 -2023-02-13 18:28:59,987 - Epoch: [161][ 20/ 1207] Overall Loss 0.275001 Objective Loss 0.275001 LR 0.000250 Time 0.054049 -2023-02-13 18:29:00,176 - Epoch: [161][ 30/ 1207] Overall Loss 0.272126 Objective Loss 0.272126 LR 0.000250 Time 0.042333 -2023-02-13 18:29:00,365 - Epoch: [161][ 40/ 1207] Overall Loss 0.275952 Objective Loss 0.275952 LR 0.000250 Time 0.036470 -2023-02-13 18:29:00,555 - Epoch: [161][ 50/ 1207] Overall Loss 0.278244 Objective Loss 0.278244 LR 0.000250 Time 0.032956 -2023-02-13 18:29:00,744 - Epoch: [161][ 60/ 1207] Overall Loss 0.278054 Objective Loss 0.278054 LR 0.000250 Time 0.030606 -2023-02-13 18:29:00,933 - Epoch: [161][ 70/ 1207] Overall Loss 0.277896 Objective Loss 0.277896 LR 0.000250 Time 0.028941 -2023-02-13 18:29:01,123 - Epoch: [161][ 80/ 1207] Overall Loss 0.276484 Objective Loss 0.276484 LR 0.000250 Time 0.027684 -2023-02-13 18:29:01,312 - Epoch: [161][ 90/ 1207] Overall Loss 0.273770 Objective Loss 0.273770 LR 0.000250 Time 0.026703 -2023-02-13 18:29:01,501 - Epoch: [161][ 100/ 1207] Overall Loss 0.272745 Objective Loss 0.272745 LR 0.000250 Time 0.025927 -2023-02-13 18:29:01,691 - Epoch: [161][ 110/ 1207] Overall Loss 0.271943 Objective Loss 0.271943 LR 0.000250 Time 0.025290 -2023-02-13 18:29:01,880 - Epoch: [161][ 120/ 1207] Overall Loss 0.269961 Objective Loss 0.269961 LR 0.000250 Time 0.024757 -2023-02-13 18:29:02,070 - Epoch: [161][ 130/ 1207] Overall Loss 0.268906 Objective Loss 0.268906 LR 0.000250 Time 0.024307 -2023-02-13 18:29:02,259 - Epoch: [161][ 140/ 1207] Overall Loss 0.269126 Objective Loss 0.269126 LR 0.000250 Time 0.023923 -2023-02-13 18:29:02,448 - Epoch: [161][ 150/ 1207] Overall Loss 0.270120 Objective Loss 0.270120 LR 0.000250 Time 0.023585 -2023-02-13 18:29:02,638 - Epoch: [161][ 160/ 1207] Overall Loss 0.270955 Objective Loss 0.270955 LR 0.000250 Time 0.023295 -2023-02-13 18:29:02,828 - Epoch: [161][ 170/ 1207] Overall Loss 0.270780 Objective Loss 0.270780 LR 0.000250 Time 0.023038 -2023-02-13 18:29:03,017 - Epoch: [161][ 180/ 1207] Overall Loss 0.271402 Objective Loss 0.271402 LR 0.000250 Time 0.022810 -2023-02-13 18:29:03,207 - Epoch: [161][ 190/ 1207] Overall Loss 0.270598 Objective Loss 0.270598 LR 0.000250 Time 0.022604 -2023-02-13 18:29:03,396 - Epoch: [161][ 200/ 1207] Overall Loss 0.270933 Objective Loss 0.270933 LR 0.000250 Time 0.022418 -2023-02-13 18:29:03,585 - Epoch: [161][ 210/ 1207] Overall Loss 0.270996 Objective Loss 0.270996 LR 0.000250 Time 0.022250 -2023-02-13 18:29:03,774 - Epoch: [161][ 220/ 1207] Overall Loss 0.270147 Objective Loss 0.270147 LR 0.000250 Time 0.022097 -2023-02-13 18:29:03,963 - Epoch: [161][ 230/ 1207] Overall Loss 0.269636 Objective Loss 0.269636 LR 0.000250 Time 0.021956 -2023-02-13 18:29:04,153 - Epoch: [161][ 240/ 1207] Overall Loss 0.269447 Objective Loss 0.269447 LR 0.000250 Time 0.021831 -2023-02-13 18:29:04,342 - Epoch: [161][ 250/ 1207] Overall Loss 0.269137 Objective Loss 0.269137 LR 0.000250 Time 0.021713 -2023-02-13 18:29:04,532 - Epoch: [161][ 260/ 1207] Overall Loss 0.268704 Objective Loss 0.268704 LR 0.000250 Time 0.021606 -2023-02-13 18:29:04,724 - Epoch: [161][ 270/ 1207] Overall Loss 0.268543 Objective Loss 0.268543 LR 0.000250 Time 0.021516 -2023-02-13 18:29:04,915 - Epoch: [161][ 280/ 1207] Overall Loss 0.268549 Objective Loss 0.268549 LR 0.000250 Time 0.021428 -2023-02-13 18:29:05,107 - Epoch: [161][ 290/ 1207] Overall Loss 0.269226 Objective Loss 0.269226 LR 0.000250 Time 0.021350 -2023-02-13 18:29:05,298 - Epoch: [161][ 300/ 1207] Overall Loss 0.269257 Objective Loss 0.269257 LR 0.000250 Time 0.021273 -2023-02-13 18:29:05,489 - Epoch: [161][ 310/ 1207] Overall Loss 0.268670 Objective Loss 0.268670 LR 0.000250 Time 0.021202 -2023-02-13 18:29:05,680 - Epoch: [161][ 320/ 1207] Overall Loss 0.268321 Objective Loss 0.268321 LR 0.000250 Time 0.021135 -2023-02-13 18:29:05,872 - Epoch: [161][ 330/ 1207] Overall Loss 0.267618 Objective Loss 0.267618 LR 0.000250 Time 0.021075 -2023-02-13 18:29:06,064 - Epoch: [161][ 340/ 1207] Overall Loss 0.267723 Objective Loss 0.267723 LR 0.000250 Time 0.021019 -2023-02-13 18:29:06,255 - Epoch: [161][ 350/ 1207] Overall Loss 0.267158 Objective Loss 0.267158 LR 0.000250 Time 0.020964 -2023-02-13 18:29:06,446 - Epoch: [161][ 360/ 1207] Overall Loss 0.266697 Objective Loss 0.266697 LR 0.000250 Time 0.020911 -2023-02-13 18:29:06,638 - Epoch: [161][ 370/ 1207] Overall Loss 0.267411 Objective Loss 0.267411 LR 0.000250 Time 0.020862 -2023-02-13 18:29:06,829 - Epoch: [161][ 380/ 1207] Overall Loss 0.267558 Objective Loss 0.267558 LR 0.000250 Time 0.020814 -2023-02-13 18:29:07,021 - Epoch: [161][ 390/ 1207] Overall Loss 0.267412 Objective Loss 0.267412 LR 0.000250 Time 0.020772 -2023-02-13 18:29:07,211 - Epoch: [161][ 400/ 1207] Overall Loss 0.267146 Objective Loss 0.267146 LR 0.000250 Time 0.020728 -2023-02-13 18:29:07,403 - Epoch: [161][ 410/ 1207] Overall Loss 0.266734 Objective Loss 0.266734 LR 0.000250 Time 0.020690 -2023-02-13 18:29:07,594 - Epoch: [161][ 420/ 1207] Overall Loss 0.267338 Objective Loss 0.267338 LR 0.000250 Time 0.020651 -2023-02-13 18:29:07,786 - Epoch: [161][ 430/ 1207] Overall Loss 0.267324 Objective Loss 0.267324 LR 0.000250 Time 0.020616 -2023-02-13 18:29:07,977 - Epoch: [161][ 440/ 1207] Overall Loss 0.267254 Objective Loss 0.267254 LR 0.000250 Time 0.020580 -2023-02-13 18:29:08,169 - Epoch: [161][ 450/ 1207] Overall Loss 0.266728 Objective Loss 0.266728 LR 0.000250 Time 0.020549 -2023-02-13 18:29:08,361 - Epoch: [161][ 460/ 1207] Overall Loss 0.266983 Objective Loss 0.266983 LR 0.000250 Time 0.020518 -2023-02-13 18:29:08,552 - Epoch: [161][ 470/ 1207] Overall Loss 0.266774 Objective Loss 0.266774 LR 0.000250 Time 0.020489 -2023-02-13 18:29:08,743 - Epoch: [161][ 480/ 1207] Overall Loss 0.266674 Objective Loss 0.266674 LR 0.000250 Time 0.020459 -2023-02-13 18:29:08,935 - Epoch: [161][ 490/ 1207] Overall Loss 0.267077 Objective Loss 0.267077 LR 0.000250 Time 0.020432 -2023-02-13 18:29:09,126 - Epoch: [161][ 500/ 1207] Overall Loss 0.267159 Objective Loss 0.267159 LR 0.000250 Time 0.020406 -2023-02-13 18:29:09,318 - Epoch: [161][ 510/ 1207] Overall Loss 0.267655 Objective Loss 0.267655 LR 0.000250 Time 0.020381 -2023-02-13 18:29:09,509 - Epoch: [161][ 520/ 1207] Overall Loss 0.267694 Objective Loss 0.267694 LR 0.000250 Time 0.020355 -2023-02-13 18:29:09,701 - Epoch: [161][ 530/ 1207] Overall Loss 0.267973 Objective Loss 0.267973 LR 0.000250 Time 0.020332 -2023-02-13 18:29:09,891 - Epoch: [161][ 540/ 1207] Overall Loss 0.268114 Objective Loss 0.268114 LR 0.000250 Time 0.020308 -2023-02-13 18:29:10,083 - Epoch: [161][ 550/ 1207] Overall Loss 0.268542 Objective Loss 0.268542 LR 0.000250 Time 0.020287 -2023-02-13 18:29:10,273 - Epoch: [161][ 560/ 1207] Overall Loss 0.268497 Objective Loss 0.268497 LR 0.000250 Time 0.020264 -2023-02-13 18:29:10,465 - Epoch: [161][ 570/ 1207] Overall Loss 0.268249 Objective Loss 0.268249 LR 0.000250 Time 0.020243 -2023-02-13 18:29:10,656 - Epoch: [161][ 580/ 1207] Overall Loss 0.268429 Objective Loss 0.268429 LR 0.000250 Time 0.020223 -2023-02-13 18:29:10,847 - Epoch: [161][ 590/ 1207] Overall Loss 0.268190 Objective Loss 0.268190 LR 0.000250 Time 0.020204 -2023-02-13 18:29:11,039 - Epoch: [161][ 600/ 1207] Overall Loss 0.268375 Objective Loss 0.268375 LR 0.000250 Time 0.020186 -2023-02-13 18:29:11,231 - Epoch: [161][ 610/ 1207] Overall Loss 0.268231 Objective Loss 0.268231 LR 0.000250 Time 0.020169 -2023-02-13 18:29:11,424 - Epoch: [161][ 620/ 1207] Overall Loss 0.268076 Objective Loss 0.268076 LR 0.000250 Time 0.020154 -2023-02-13 18:29:11,616 - Epoch: [161][ 630/ 1207] Overall Loss 0.268289 Objective Loss 0.268289 LR 0.000250 Time 0.020139 -2023-02-13 18:29:11,807 - Epoch: [161][ 640/ 1207] Overall Loss 0.267951 Objective Loss 0.267951 LR 0.000250 Time 0.020122 -2023-02-13 18:29:11,999 - Epoch: [161][ 650/ 1207] Overall Loss 0.267834 Objective Loss 0.267834 LR 0.000250 Time 0.020108 -2023-02-13 18:29:12,191 - Epoch: [161][ 660/ 1207] Overall Loss 0.267612 Objective Loss 0.267612 LR 0.000250 Time 0.020093 -2023-02-13 18:29:12,383 - Epoch: [161][ 670/ 1207] Overall Loss 0.267453 Objective Loss 0.267453 LR 0.000250 Time 0.020080 -2023-02-13 18:29:12,574 - Epoch: [161][ 680/ 1207] Overall Loss 0.267275 Objective Loss 0.267275 LR 0.000250 Time 0.020065 -2023-02-13 18:29:12,765 - Epoch: [161][ 690/ 1207] Overall Loss 0.267490 Objective Loss 0.267490 LR 0.000250 Time 0.020050 -2023-02-13 18:29:12,956 - Epoch: [161][ 700/ 1207] Overall Loss 0.267352 Objective Loss 0.267352 LR 0.000250 Time 0.020036 -2023-02-13 18:29:13,149 - Epoch: [161][ 710/ 1207] Overall Loss 0.267159 Objective Loss 0.267159 LR 0.000250 Time 0.020025 -2023-02-13 18:29:13,340 - Epoch: [161][ 720/ 1207] Overall Loss 0.266895 Objective Loss 0.266895 LR 0.000250 Time 0.020012 -2023-02-13 18:29:13,532 - Epoch: [161][ 730/ 1207] Overall Loss 0.266883 Objective Loss 0.266883 LR 0.000250 Time 0.020001 -2023-02-13 18:29:13,724 - Epoch: [161][ 740/ 1207] Overall Loss 0.266545 Objective Loss 0.266545 LR 0.000250 Time 0.019988 -2023-02-13 18:29:13,916 - Epoch: [161][ 750/ 1207] Overall Loss 0.266234 Objective Loss 0.266234 LR 0.000250 Time 0.019977 -2023-02-13 18:29:14,107 - Epoch: [161][ 760/ 1207] Overall Loss 0.266109 Objective Loss 0.266109 LR 0.000250 Time 0.019966 -2023-02-13 18:29:14,300 - Epoch: [161][ 770/ 1207] Overall Loss 0.265925 Objective Loss 0.265925 LR 0.000250 Time 0.019957 -2023-02-13 18:29:14,491 - Epoch: [161][ 780/ 1207] Overall Loss 0.265799 Objective Loss 0.265799 LR 0.000250 Time 0.019945 -2023-02-13 18:29:14,683 - Epoch: [161][ 790/ 1207] Overall Loss 0.265661 Objective Loss 0.265661 LR 0.000250 Time 0.019935 -2023-02-13 18:29:14,874 - Epoch: [161][ 800/ 1207] Overall Loss 0.265432 Objective Loss 0.265432 LR 0.000250 Time 0.019924 -2023-02-13 18:29:15,066 - Epoch: [161][ 810/ 1207] Overall Loss 0.265302 Objective Loss 0.265302 LR 0.000250 Time 0.019915 -2023-02-13 18:29:15,257 - Epoch: [161][ 820/ 1207] Overall Loss 0.265108 Objective Loss 0.265108 LR 0.000250 Time 0.019905 -2023-02-13 18:29:15,449 - Epoch: [161][ 830/ 1207] Overall Loss 0.265161 Objective Loss 0.265161 LR 0.000250 Time 0.019896 -2023-02-13 18:29:15,641 - Epoch: [161][ 840/ 1207] Overall Loss 0.264668 Objective Loss 0.264668 LR 0.000250 Time 0.019887 -2023-02-13 18:29:15,833 - Epoch: [161][ 850/ 1207] Overall Loss 0.264794 Objective Loss 0.264794 LR 0.000250 Time 0.019878 -2023-02-13 18:29:16,025 - Epoch: [161][ 860/ 1207] Overall Loss 0.264950 Objective Loss 0.264950 LR 0.000250 Time 0.019870 -2023-02-13 18:29:16,217 - Epoch: [161][ 870/ 1207] Overall Loss 0.264680 Objective Loss 0.264680 LR 0.000250 Time 0.019862 -2023-02-13 18:29:16,409 - Epoch: [161][ 880/ 1207] Overall Loss 0.264550 Objective Loss 0.264550 LR 0.000250 Time 0.019853 -2023-02-13 18:29:16,601 - Epoch: [161][ 890/ 1207] Overall Loss 0.264390 Objective Loss 0.264390 LR 0.000250 Time 0.019846 -2023-02-13 18:29:16,792 - Epoch: [161][ 900/ 1207] Overall Loss 0.264560 Objective Loss 0.264560 LR 0.000250 Time 0.019837 -2023-02-13 18:29:16,984 - Epoch: [161][ 910/ 1207] Overall Loss 0.264365 Objective Loss 0.264365 LR 0.000250 Time 0.019831 -2023-02-13 18:29:17,176 - Epoch: [161][ 920/ 1207] Overall Loss 0.264484 Objective Loss 0.264484 LR 0.000250 Time 0.019823 -2023-02-13 18:29:17,368 - Epoch: [161][ 930/ 1207] Overall Loss 0.264341 Objective Loss 0.264341 LR 0.000250 Time 0.019816 -2023-02-13 18:29:17,560 - Epoch: [161][ 940/ 1207] Overall Loss 0.264519 Objective Loss 0.264519 LR 0.000250 Time 0.019808 -2023-02-13 18:29:17,752 - Epoch: [161][ 950/ 1207] Overall Loss 0.264361 Objective Loss 0.264361 LR 0.000250 Time 0.019802 -2023-02-13 18:29:17,943 - Epoch: [161][ 960/ 1207] Overall Loss 0.264300 Objective Loss 0.264300 LR 0.000250 Time 0.019794 -2023-02-13 18:29:18,134 - Epoch: [161][ 970/ 1207] Overall Loss 0.264437 Objective Loss 0.264437 LR 0.000250 Time 0.019787 -2023-02-13 18:29:18,324 - Epoch: [161][ 980/ 1207] Overall Loss 0.264307 Objective Loss 0.264307 LR 0.000250 Time 0.019778 -2023-02-13 18:29:18,513 - Epoch: [161][ 990/ 1207] Overall Loss 0.264254 Objective Loss 0.264254 LR 0.000250 Time 0.019769 -2023-02-13 18:29:18,703 - Epoch: [161][ 1000/ 1207] Overall Loss 0.264128 Objective Loss 0.264128 LR 0.000250 Time 0.019761 -2023-02-13 18:29:18,892 - Epoch: [161][ 1010/ 1207] Overall Loss 0.264199 Objective Loss 0.264199 LR 0.000250 Time 0.019752 -2023-02-13 18:29:19,082 - Epoch: [161][ 1020/ 1207] Overall Loss 0.263968 Objective Loss 0.263968 LR 0.000250 Time 0.019745 -2023-02-13 18:29:19,272 - Epoch: [161][ 1030/ 1207] Overall Loss 0.264018 Objective Loss 0.264018 LR 0.000250 Time 0.019736 -2023-02-13 18:29:19,461 - Epoch: [161][ 1040/ 1207] Overall Loss 0.264024 Objective Loss 0.264024 LR 0.000250 Time 0.019729 -2023-02-13 18:29:19,651 - Epoch: [161][ 1050/ 1207] Overall Loss 0.263998 Objective Loss 0.263998 LR 0.000250 Time 0.019721 -2023-02-13 18:29:19,841 - Epoch: [161][ 1060/ 1207] Overall Loss 0.264044 Objective Loss 0.264044 LR 0.000250 Time 0.019714 -2023-02-13 18:29:20,031 - Epoch: [161][ 1070/ 1207] Overall Loss 0.263898 Objective Loss 0.263898 LR 0.000250 Time 0.019707 -2023-02-13 18:29:20,221 - Epoch: [161][ 1080/ 1207] Overall Loss 0.263799 Objective Loss 0.263799 LR 0.000250 Time 0.019700 -2023-02-13 18:29:20,411 - Epoch: [161][ 1090/ 1207] Overall Loss 0.263528 Objective Loss 0.263528 LR 0.000250 Time 0.019694 -2023-02-13 18:29:20,602 - Epoch: [161][ 1100/ 1207] Overall Loss 0.263730 Objective Loss 0.263730 LR 0.000250 Time 0.019687 -2023-02-13 18:29:20,791 - Epoch: [161][ 1110/ 1207] Overall Loss 0.263894 Objective Loss 0.263894 LR 0.000250 Time 0.019680 -2023-02-13 18:29:20,982 - Epoch: [161][ 1120/ 1207] Overall Loss 0.263797 Objective Loss 0.263797 LR 0.000250 Time 0.019674 -2023-02-13 18:29:21,172 - Epoch: [161][ 1130/ 1207] Overall Loss 0.263577 Objective Loss 0.263577 LR 0.000250 Time 0.019668 -2023-02-13 18:29:21,362 - Epoch: [161][ 1140/ 1207] Overall Loss 0.263525 Objective Loss 0.263525 LR 0.000250 Time 0.019662 -2023-02-13 18:29:21,551 - Epoch: [161][ 1150/ 1207] Overall Loss 0.263534 Objective Loss 0.263534 LR 0.000250 Time 0.019656 -2023-02-13 18:29:21,741 - Epoch: [161][ 1160/ 1207] Overall Loss 0.263811 Objective Loss 0.263811 LR 0.000250 Time 0.019649 -2023-02-13 18:29:21,932 - Epoch: [161][ 1170/ 1207] Overall Loss 0.263678 Objective Loss 0.263678 LR 0.000250 Time 0.019644 -2023-02-13 18:29:22,122 - Epoch: [161][ 1180/ 1207] Overall Loss 0.263662 Objective Loss 0.263662 LR 0.000250 Time 0.019638 -2023-02-13 18:29:22,311 - Epoch: [161][ 1190/ 1207] Overall Loss 0.263656 Objective Loss 0.263656 LR 0.000250 Time 0.019632 -2023-02-13 18:29:22,558 - Epoch: [161][ 1200/ 1207] Overall Loss 0.263690 Objective Loss 0.263690 LR 0.000250 Time 0.019674 -2023-02-13 18:29:22,673 - Epoch: [161][ 1207/ 1207] Overall Loss 0.263763 Objective Loss 0.263763 Top1 85.670732 Top5 97.865854 LR 0.000250 Time 0.019655 -2023-02-13 18:29:22,745 - --- validate (epoch=161)----------- -2023-02-13 18:29:22,746 - 34311 samples (256 per mini-batch) -2023-02-13 18:29:23,156 - Epoch: [161][ 10/ 135] Loss 0.335932 Top1 85.234375 Top5 97.851562 -2023-02-13 18:29:23,301 - Epoch: [161][ 20/ 135] Loss 0.335588 Top1 84.316406 Top5 97.617188 -2023-02-13 18:29:23,436 - Epoch: [161][ 30/ 135] Loss 0.332068 Top1 84.166667 Top5 97.734375 -2023-02-13 18:29:23,571 - Epoch: [161][ 40/ 135] Loss 0.333120 Top1 84.404297 Top5 97.783203 -2023-02-13 18:29:23,708 - Epoch: [161][ 50/ 135] Loss 0.329793 Top1 84.390625 Top5 97.835938 -2023-02-13 18:29:23,846 - Epoch: [161][ 60/ 135] Loss 0.328268 Top1 84.524740 Top5 97.858073 -2023-02-13 18:29:23,993 - Epoch: [161][ 70/ 135] Loss 0.331679 Top1 84.525670 Top5 97.873884 -2023-02-13 18:29:24,133 - Epoch: [161][ 80/ 135] Loss 0.328297 Top1 84.389648 Top5 97.797852 -2023-02-13 18:29:24,277 - Epoch: [161][ 90/ 135] Loss 0.330596 Top1 84.210069 Top5 97.786458 -2023-02-13 18:29:24,407 - Epoch: [161][ 100/ 135] Loss 0.327867 Top1 84.277344 Top5 97.753906 -2023-02-13 18:29:24,543 - Epoch: [161][ 110/ 135] Loss 0.326982 Top1 84.371449 Top5 97.720170 -2023-02-13 18:29:24,673 - Epoch: [161][ 120/ 135] Loss 0.325811 Top1 84.361979 Top5 97.727865 -2023-02-13 18:29:24,799 - Epoch: [161][ 130/ 135] Loss 0.323360 Top1 84.447115 Top5 97.746394 -2023-02-13 18:29:24,844 - Epoch: [161][ 135/ 135] Loss 0.322207 Top1 84.468538 Top5 97.764565 -2023-02-13 18:29:24,915 - ==> Top1: 84.469 Top5: 97.765 Loss: 0.322 - -2023-02-13 18:29:24,916 - ==> Confusion: -[[ 871 4 5 0 8 3 0 3 2 42 0 4 0 4 4 2 3 1 1 1 9] - [ 3 954 1 2 9 21 0 15 2 2 2 1 0 1 1 1 6 1 2 4 5] - [ 5 2 967 8 5 0 12 13 1 2 3 3 3 7 4 5 4 3 2 1 8] - [ 8 0 22 901 1 5 0 2 3 2 15 2 9 0 19 2 4 4 11 0 6] - [ 14 7 0 0 992 9 1 1 4 2 0 7 3 3 8 5 4 0 0 2 4] - [ 2 19 0 3 7 959 4 13 4 6 0 16 2 18 1 2 6 1 0 4 3] - [ 3 4 18 1 0 6 1033 5 1 0 2 2 2 1 0 5 2 3 2 3 6] - [ 2 10 12 2 3 32 3 927 0 2 2 5 1 1 0 0 0 2 8 7 5] - [ 15 1 1 2 1 0 1 1 917 30 4 2 0 10 14 2 1 0 4 1 2] - [ 84 1 2 0 7 0 0 1 32 855 0 2 0 16 5 2 0 2 0 0 3] - [ 2 2 3 5 0 2 2 5 18 1 986 2 1 9 3 0 2 0 6 0 2] - [ 1 2 2 0 1 7 1 8 1 2 1 923 23 5 5 6 1 5 3 6 2] - [ 0 0 0 6 1 3 0 1 2 1 0 35 873 1 2 9 3 13 2 1 6] - [ 5 2 3 1 4 10 1 1 14 13 8 4 2 938 4 4 4 2 0 2 2] - [ 5 4 0 10 5 3 0 0 20 7 3 0 2 3 1010 1 2 3 6 0 8] - [ 3 2 7 0 5 1 3 2 1 0 0 2 8 2 0 981 9 9 0 6 5] - [ 3 3 1 0 5 3 0 1 0 1 0 0 2 3 2 11 1009 2 1 2 12] - [ 4 1 0 2 1 0 1 0 1 0 2 8 16 0 1 20 0 988 0 1 5] - [ 4 7 6 11 0 1 0 27 5 1 5 1 5 0 15 2 0 3 991 2 0] - [ 2 2 1 0 1 3 7 9 1 0 0 13 3 2 1 5 6 3 0 1081 8] - [ 161 259 236 106 131 216 73 188 106 98 217 123 350 276 175 119 257 115 138 264 9826]] - -2023-02-13 18:29:24,918 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:29:24,918 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:29:24,923 - - -2023-02-13 18:29:24,923 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:29:25,924 - Epoch: [162][ 10/ 1207] Overall Loss 0.280955 Objective Loss 0.280955 LR 0.000250 Time 0.100016 -2023-02-13 18:29:26,136 - Epoch: [162][ 20/ 1207] Overall Loss 0.268062 Objective Loss 0.268062 LR 0.000250 Time 0.060587 -2023-02-13 18:29:26,335 - Epoch: [162][ 30/ 1207] Overall Loss 0.268880 Objective Loss 0.268880 LR 0.000250 Time 0.046989 -2023-02-13 18:29:26,537 - Epoch: [162][ 40/ 1207] Overall Loss 0.265021 Objective Loss 0.265021 LR 0.000250 Time 0.040284 -2023-02-13 18:29:26,735 - Epoch: [162][ 50/ 1207] Overall Loss 0.264822 Objective Loss 0.264822 LR 0.000250 Time 0.036179 -2023-02-13 18:29:26,937 - Epoch: [162][ 60/ 1207] Overall Loss 0.263697 Objective Loss 0.263697 LR 0.000250 Time 0.033519 -2023-02-13 18:29:27,136 - Epoch: [162][ 70/ 1207] Overall Loss 0.264444 Objective Loss 0.264444 LR 0.000250 Time 0.031561 -2023-02-13 18:29:27,338 - Epoch: [162][ 80/ 1207] Overall Loss 0.263545 Objective Loss 0.263545 LR 0.000250 Time 0.030135 -2023-02-13 18:29:27,535 - Epoch: [162][ 90/ 1207] Overall Loss 0.262823 Objective Loss 0.262823 LR 0.000250 Time 0.028977 -2023-02-13 18:29:27,737 - Epoch: [162][ 100/ 1207] Overall Loss 0.264129 Objective Loss 0.264129 LR 0.000250 Time 0.028097 -2023-02-13 18:29:27,935 - Epoch: [162][ 110/ 1207] Overall Loss 0.264636 Objective Loss 0.264636 LR 0.000250 Time 0.027334 -2023-02-13 18:29:28,138 - Epoch: [162][ 120/ 1207] Overall Loss 0.263830 Objective Loss 0.263830 LR 0.000250 Time 0.026749 -2023-02-13 18:29:28,336 - Epoch: [162][ 130/ 1207] Overall Loss 0.262897 Objective Loss 0.262897 LR 0.000250 Time 0.026212 -2023-02-13 18:29:28,539 - Epoch: [162][ 140/ 1207] Overall Loss 0.263013 Objective Loss 0.263013 LR 0.000250 Time 0.025783 -2023-02-13 18:29:28,736 - Epoch: [162][ 150/ 1207] Overall Loss 0.264097 Objective Loss 0.264097 LR 0.000250 Time 0.025377 -2023-02-13 18:29:28,940 - Epoch: [162][ 160/ 1207] Overall Loss 0.263468 Objective Loss 0.263468 LR 0.000250 Time 0.025063 -2023-02-13 18:29:29,138 - Epoch: [162][ 170/ 1207] Overall Loss 0.261667 Objective Loss 0.261667 LR 0.000250 Time 0.024753 -2023-02-13 18:29:29,341 - Epoch: [162][ 180/ 1207] Overall Loss 0.260699 Objective Loss 0.260699 LR 0.000250 Time 0.024499 -2023-02-13 18:29:29,538 - Epoch: [162][ 190/ 1207] Overall Loss 0.259758 Objective Loss 0.259758 LR 0.000250 Time 0.024249 -2023-02-13 18:29:29,741 - Epoch: [162][ 200/ 1207] Overall Loss 0.260756 Objective Loss 0.260756 LR 0.000250 Time 0.024046 -2023-02-13 18:29:29,938 - Epoch: [162][ 210/ 1207] Overall Loss 0.260524 Objective Loss 0.260524 LR 0.000250 Time 0.023840 -2023-02-13 18:29:30,142 - Epoch: [162][ 220/ 1207] Overall Loss 0.260502 Objective Loss 0.260502 LR 0.000250 Time 0.023680 -2023-02-13 18:29:30,339 - Epoch: [162][ 230/ 1207] Overall Loss 0.260452 Objective Loss 0.260452 LR 0.000250 Time 0.023505 -2023-02-13 18:29:30,541 - Epoch: [162][ 240/ 1207] Overall Loss 0.260688 Objective Loss 0.260688 LR 0.000250 Time 0.023368 -2023-02-13 18:29:30,739 - Epoch: [162][ 250/ 1207] Overall Loss 0.260076 Objective Loss 0.260076 LR 0.000250 Time 0.023222 -2023-02-13 18:29:30,942 - Epoch: [162][ 260/ 1207] Overall Loss 0.259801 Objective Loss 0.259801 LR 0.000250 Time 0.023109 -2023-02-13 18:29:31,140 - Epoch: [162][ 270/ 1207] Overall Loss 0.260570 Objective Loss 0.260570 LR 0.000250 Time 0.022985 -2023-02-13 18:29:31,342 - Epoch: [162][ 280/ 1207] Overall Loss 0.260716 Objective Loss 0.260716 LR 0.000250 Time 0.022883 -2023-02-13 18:29:31,539 - Epoch: [162][ 290/ 1207] Overall Loss 0.260977 Objective Loss 0.260977 LR 0.000250 Time 0.022774 -2023-02-13 18:29:31,742 - Epoch: [162][ 300/ 1207] Overall Loss 0.260900 Objective Loss 0.260900 LR 0.000250 Time 0.022688 -2023-02-13 18:29:31,939 - Epoch: [162][ 310/ 1207] Overall Loss 0.261186 Objective Loss 0.261186 LR 0.000250 Time 0.022593 -2023-02-13 18:29:32,143 - Epoch: [162][ 320/ 1207] Overall Loss 0.260736 Objective Loss 0.260736 LR 0.000250 Time 0.022521 -2023-02-13 18:29:32,341 - Epoch: [162][ 330/ 1207] Overall Loss 0.260563 Objective Loss 0.260563 LR 0.000250 Time 0.022437 -2023-02-13 18:29:32,543 - Epoch: [162][ 340/ 1207] Overall Loss 0.261181 Objective Loss 0.261181 LR 0.000250 Time 0.022373 -2023-02-13 18:29:32,742 - Epoch: [162][ 350/ 1207] Overall Loss 0.261135 Objective Loss 0.261135 LR 0.000250 Time 0.022299 -2023-02-13 18:29:32,944 - Epoch: [162][ 360/ 1207] Overall Loss 0.261127 Objective Loss 0.261127 LR 0.000250 Time 0.022241 -2023-02-13 18:29:33,143 - Epoch: [162][ 370/ 1207] Overall Loss 0.260962 Objective Loss 0.260962 LR 0.000250 Time 0.022177 -2023-02-13 18:29:33,346 - Epoch: [162][ 380/ 1207] Overall Loss 0.260442 Objective Loss 0.260442 LR 0.000250 Time 0.022125 -2023-02-13 18:29:33,543 - Epoch: [162][ 390/ 1207] Overall Loss 0.260435 Objective Loss 0.260435 LR 0.000250 Time 0.022063 -2023-02-13 18:29:33,748 - Epoch: [162][ 400/ 1207] Overall Loss 0.260264 Objective Loss 0.260264 LR 0.000250 Time 0.022023 -2023-02-13 18:29:33,950 - Epoch: [162][ 410/ 1207] Overall Loss 0.260618 Objective Loss 0.260618 LR 0.000250 Time 0.021976 -2023-02-13 18:29:34,155 - Epoch: [162][ 420/ 1207] Overall Loss 0.261055 Objective Loss 0.261055 LR 0.000250 Time 0.021942 -2023-02-13 18:29:34,356 - Epoch: [162][ 430/ 1207] Overall Loss 0.260335 Objective Loss 0.260335 LR 0.000250 Time 0.021899 -2023-02-13 18:29:34,561 - Epoch: [162][ 440/ 1207] Overall Loss 0.260310 Objective Loss 0.260310 LR 0.000250 Time 0.021864 -2023-02-13 18:29:34,761 - Epoch: [162][ 450/ 1207] Overall Loss 0.259668 Objective Loss 0.259668 LR 0.000250 Time 0.021822 -2023-02-13 18:29:34,964 - Epoch: [162][ 460/ 1207] Overall Loss 0.259536 Objective Loss 0.259536 LR 0.000250 Time 0.021789 -2023-02-13 18:29:35,163 - Epoch: [162][ 470/ 1207] Overall Loss 0.259636 Objective Loss 0.259636 LR 0.000250 Time 0.021747 -2023-02-13 18:29:35,366 - Epoch: [162][ 480/ 1207] Overall Loss 0.260086 Objective Loss 0.260086 LR 0.000250 Time 0.021716 -2023-02-13 18:29:35,563 - Epoch: [162][ 490/ 1207] Overall Loss 0.259779 Objective Loss 0.259779 LR 0.000250 Time 0.021676 -2023-02-13 18:29:35,766 - Epoch: [162][ 500/ 1207] Overall Loss 0.259790 Objective Loss 0.259790 LR 0.000250 Time 0.021647 -2023-02-13 18:29:35,965 - Epoch: [162][ 510/ 1207] Overall Loss 0.259845 Objective Loss 0.259845 LR 0.000250 Time 0.021611 -2023-02-13 18:29:36,168 - Epoch: [162][ 520/ 1207] Overall Loss 0.259801 Objective Loss 0.259801 LR 0.000250 Time 0.021585 -2023-02-13 18:29:36,366 - Epoch: [162][ 530/ 1207] Overall Loss 0.259802 Objective Loss 0.259802 LR 0.000250 Time 0.021550 -2023-02-13 18:29:36,568 - Epoch: [162][ 540/ 1207] Overall Loss 0.259937 Objective Loss 0.259937 LR 0.000250 Time 0.021526 -2023-02-13 18:29:36,766 - Epoch: [162][ 550/ 1207] Overall Loss 0.259919 Objective Loss 0.259919 LR 0.000250 Time 0.021494 -2023-02-13 18:29:36,969 - Epoch: [162][ 560/ 1207] Overall Loss 0.259766 Objective Loss 0.259766 LR 0.000250 Time 0.021471 -2023-02-13 18:29:37,167 - Epoch: [162][ 570/ 1207] Overall Loss 0.259765 Objective Loss 0.259765 LR 0.000250 Time 0.021441 -2023-02-13 18:29:37,369 - Epoch: [162][ 580/ 1207] Overall Loss 0.259918 Objective Loss 0.259918 LR 0.000250 Time 0.021419 -2023-02-13 18:29:37,567 - Epoch: [162][ 590/ 1207] Overall Loss 0.259963 Objective Loss 0.259963 LR 0.000250 Time 0.021392 -2023-02-13 18:29:37,771 - Epoch: [162][ 600/ 1207] Overall Loss 0.259917 Objective Loss 0.259917 LR 0.000250 Time 0.021374 -2023-02-13 18:29:37,970 - Epoch: [162][ 610/ 1207] Overall Loss 0.259935 Objective Loss 0.259935 LR 0.000250 Time 0.021350 -2023-02-13 18:29:38,175 - Epoch: [162][ 620/ 1207] Overall Loss 0.259860 Objective Loss 0.259860 LR 0.000250 Time 0.021334 -2023-02-13 18:29:38,373 - Epoch: [162][ 630/ 1207] Overall Loss 0.259566 Objective Loss 0.259566 LR 0.000250 Time 0.021311 -2023-02-13 18:29:38,577 - Epoch: [162][ 640/ 1207] Overall Loss 0.259454 Objective Loss 0.259454 LR 0.000250 Time 0.021294 -2023-02-13 18:29:38,776 - Epoch: [162][ 650/ 1207] Overall Loss 0.258769 Objective Loss 0.258769 LR 0.000250 Time 0.021273 -2023-02-13 18:29:38,979 - Epoch: [162][ 660/ 1207] Overall Loss 0.258625 Objective Loss 0.258625 LR 0.000250 Time 0.021258 -2023-02-13 18:29:39,179 - Epoch: [162][ 670/ 1207] Overall Loss 0.258510 Objective Loss 0.258510 LR 0.000250 Time 0.021238 -2023-02-13 18:29:39,383 - Epoch: [162][ 680/ 1207] Overall Loss 0.258499 Objective Loss 0.258499 LR 0.000250 Time 0.021225 -2023-02-13 18:29:39,582 - Epoch: [162][ 690/ 1207] Overall Loss 0.258354 Objective Loss 0.258354 LR 0.000250 Time 0.021206 -2023-02-13 18:29:39,786 - Epoch: [162][ 700/ 1207] Overall Loss 0.258371 Objective Loss 0.258371 LR 0.000250 Time 0.021193 -2023-02-13 18:29:39,984 - Epoch: [162][ 710/ 1207] Overall Loss 0.258194 Objective Loss 0.258194 LR 0.000250 Time 0.021173 -2023-02-13 18:29:40,188 - Epoch: [162][ 720/ 1207] Overall Loss 0.258409 Objective Loss 0.258409 LR 0.000250 Time 0.021162 -2023-02-13 18:29:40,387 - Epoch: [162][ 730/ 1207] Overall Loss 0.258607 Objective Loss 0.258607 LR 0.000250 Time 0.021144 -2023-02-13 18:29:40,590 - Epoch: [162][ 740/ 1207] Overall Loss 0.258613 Objective Loss 0.258613 LR 0.000250 Time 0.021132 -2023-02-13 18:29:40,788 - Epoch: [162][ 750/ 1207] Overall Loss 0.258440 Objective Loss 0.258440 LR 0.000250 Time 0.021113 -2023-02-13 18:29:40,992 - Epoch: [162][ 760/ 1207] Overall Loss 0.258516 Objective Loss 0.258516 LR 0.000250 Time 0.021104 -2023-02-13 18:29:41,191 - Epoch: [162][ 770/ 1207] Overall Loss 0.258501 Objective Loss 0.258501 LR 0.000250 Time 0.021088 -2023-02-13 18:29:41,394 - Epoch: [162][ 780/ 1207] Overall Loss 0.258574 Objective Loss 0.258574 LR 0.000250 Time 0.021078 -2023-02-13 18:29:41,593 - Epoch: [162][ 790/ 1207] Overall Loss 0.259003 Objective Loss 0.259003 LR 0.000250 Time 0.021062 -2023-02-13 18:29:41,796 - Epoch: [162][ 800/ 1207] Overall Loss 0.258862 Objective Loss 0.258862 LR 0.000250 Time 0.021052 -2023-02-13 18:29:41,995 - Epoch: [162][ 810/ 1207] Overall Loss 0.258829 Objective Loss 0.258829 LR 0.000250 Time 0.021037 -2023-02-13 18:29:42,198 - Epoch: [162][ 820/ 1207] Overall Loss 0.259038 Objective Loss 0.259038 LR 0.000250 Time 0.021028 -2023-02-13 18:29:42,396 - Epoch: [162][ 830/ 1207] Overall Loss 0.258923 Objective Loss 0.258923 LR 0.000250 Time 0.021012 -2023-02-13 18:29:42,598 - Epoch: [162][ 840/ 1207] Overall Loss 0.258860 Objective Loss 0.258860 LR 0.000250 Time 0.021002 -2023-02-13 18:29:42,797 - Epoch: [162][ 850/ 1207] Overall Loss 0.258898 Objective Loss 0.258898 LR 0.000250 Time 0.020989 -2023-02-13 18:29:43,001 - Epoch: [162][ 860/ 1207] Overall Loss 0.258694 Objective Loss 0.258694 LR 0.000250 Time 0.020982 -2023-02-13 18:29:43,201 - Epoch: [162][ 870/ 1207] Overall Loss 0.258892 Objective Loss 0.258892 LR 0.000250 Time 0.020970 -2023-02-13 18:29:43,405 - Epoch: [162][ 880/ 1207] Overall Loss 0.258914 Objective Loss 0.258914 LR 0.000250 Time 0.020964 -2023-02-13 18:29:43,606 - Epoch: [162][ 890/ 1207] Overall Loss 0.258680 Objective Loss 0.258680 LR 0.000250 Time 0.020952 -2023-02-13 18:29:43,810 - Epoch: [162][ 900/ 1207] Overall Loss 0.258571 Objective Loss 0.258571 LR 0.000250 Time 0.020946 -2023-02-13 18:29:44,010 - Epoch: [162][ 910/ 1207] Overall Loss 0.258695 Objective Loss 0.258695 LR 0.000250 Time 0.020936 -2023-02-13 18:29:44,215 - Epoch: [162][ 920/ 1207] Overall Loss 0.258666 Objective Loss 0.258666 LR 0.000250 Time 0.020931 -2023-02-13 18:29:44,416 - Epoch: [162][ 930/ 1207] Overall Loss 0.259013 Objective Loss 0.259013 LR 0.000250 Time 0.020921 -2023-02-13 18:29:44,620 - Epoch: [162][ 940/ 1207] Overall Loss 0.258718 Objective Loss 0.258718 LR 0.000250 Time 0.020915 -2023-02-13 18:29:44,821 - Epoch: [162][ 950/ 1207] Overall Loss 0.258652 Objective Loss 0.258652 LR 0.000250 Time 0.020906 -2023-02-13 18:29:45,024 - Epoch: [162][ 960/ 1207] Overall Loss 0.258825 Objective Loss 0.258825 LR 0.000250 Time 0.020900 -2023-02-13 18:29:45,225 - Epoch: [162][ 970/ 1207] Overall Loss 0.258681 Objective Loss 0.258681 LR 0.000250 Time 0.020891 -2023-02-13 18:29:45,429 - Epoch: [162][ 980/ 1207] Overall Loss 0.258528 Objective Loss 0.258528 LR 0.000250 Time 0.020886 -2023-02-13 18:29:45,630 - Epoch: [162][ 990/ 1207] Overall Loss 0.258292 Objective Loss 0.258292 LR 0.000250 Time 0.020877 -2023-02-13 18:29:45,834 - Epoch: [162][ 1000/ 1207] Overall Loss 0.258400 Objective Loss 0.258400 LR 0.000250 Time 0.020872 -2023-02-13 18:29:46,035 - Epoch: [162][ 1010/ 1207] Overall Loss 0.258553 Objective Loss 0.258553 LR 0.000250 Time 0.020864 -2023-02-13 18:29:46,239 - Epoch: [162][ 1020/ 1207] Overall Loss 0.258507 Objective Loss 0.258507 LR 0.000250 Time 0.020859 -2023-02-13 18:29:46,440 - Epoch: [162][ 1030/ 1207] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.000250 Time 0.020851 -2023-02-13 18:29:46,644 - Epoch: [162][ 1040/ 1207] Overall Loss 0.258672 Objective Loss 0.258672 LR 0.000250 Time 0.020846 -2023-02-13 18:29:46,843 - Epoch: [162][ 1050/ 1207] Overall Loss 0.258771 Objective Loss 0.258771 LR 0.000250 Time 0.020837 -2023-02-13 18:29:47,047 - Epoch: [162][ 1060/ 1207] Overall Loss 0.258860 Objective Loss 0.258860 LR 0.000250 Time 0.020832 -2023-02-13 18:29:47,248 - Epoch: [162][ 1070/ 1207] Overall Loss 0.258796 Objective Loss 0.258796 LR 0.000250 Time 0.020825 -2023-02-13 18:29:47,453 - Epoch: [162][ 1080/ 1207] Overall Loss 0.258650 Objective Loss 0.258650 LR 0.000250 Time 0.020822 -2023-02-13 18:29:47,653 - Epoch: [162][ 1090/ 1207] Overall Loss 0.258786 Objective Loss 0.258786 LR 0.000250 Time 0.020814 -2023-02-13 18:29:47,857 - Epoch: [162][ 1100/ 1207] Overall Loss 0.258830 Objective Loss 0.258830 LR 0.000250 Time 0.020810 -2023-02-13 18:29:48,049 - Epoch: [162][ 1110/ 1207] Overall Loss 0.258961 Objective Loss 0.258961 LR 0.000250 Time 0.020795 -2023-02-13 18:29:48,241 - Epoch: [162][ 1120/ 1207] Overall Loss 0.259206 Objective Loss 0.259206 LR 0.000250 Time 0.020781 -2023-02-13 18:29:48,432 - Epoch: [162][ 1130/ 1207] Overall Loss 0.259335 Objective Loss 0.259335 LR 0.000250 Time 0.020766 -2023-02-13 18:29:48,623 - Epoch: [162][ 1140/ 1207] Overall Loss 0.259232 Objective Loss 0.259232 LR 0.000250 Time 0.020751 -2023-02-13 18:29:48,814 - Epoch: [162][ 1150/ 1207] Overall Loss 0.259337 Objective Loss 0.259337 LR 0.000250 Time 0.020736 -2023-02-13 18:29:49,005 - Epoch: [162][ 1160/ 1207] Overall Loss 0.259408 Objective Loss 0.259408 LR 0.000250 Time 0.020721 -2023-02-13 18:29:49,197 - Epoch: [162][ 1170/ 1207] Overall Loss 0.259206 Objective Loss 0.259206 LR 0.000250 Time 0.020708 -2023-02-13 18:29:49,385 - Epoch: [162][ 1180/ 1207] Overall Loss 0.259228 Objective Loss 0.259228 LR 0.000250 Time 0.020692 -2023-02-13 18:29:49,574 - Epoch: [162][ 1190/ 1207] Overall Loss 0.259176 Objective Loss 0.259176 LR 0.000250 Time 0.020676 -2023-02-13 18:29:49,813 - Epoch: [162][ 1200/ 1207] Overall Loss 0.259298 Objective Loss 0.259298 LR 0.000250 Time 0.020703 -2023-02-13 18:29:49,928 - Epoch: [162][ 1207/ 1207] Overall Loss 0.259199 Objective Loss 0.259199 Top1 88.414634 Top5 99.695122 LR 0.000250 Time 0.020678 -2023-02-13 18:29:50,008 - --- validate (epoch=162)----------- -2023-02-13 18:29:50,008 - 34311 samples (256 per mini-batch) -2023-02-13 18:29:50,413 - Epoch: [162][ 10/ 135] Loss 0.306559 Top1 84.531250 Top5 97.773438 -2023-02-13 18:29:50,542 - Epoch: [162][ 20/ 135] Loss 0.320095 Top1 84.414062 Top5 97.500000 -2023-02-13 18:29:50,669 - Epoch: [162][ 30/ 135] Loss 0.308360 Top1 84.596354 Top5 97.643229 -2023-02-13 18:29:50,796 - Epoch: [162][ 40/ 135] Loss 0.315703 Top1 84.326172 Top5 97.685547 -2023-02-13 18:29:50,922 - Epoch: [162][ 50/ 135] Loss 0.324150 Top1 84.023438 Top5 97.601562 -2023-02-13 18:29:51,049 - Epoch: [162][ 60/ 135] Loss 0.321993 Top1 84.029948 Top5 97.669271 -2023-02-13 18:29:51,176 - Epoch: [162][ 70/ 135] Loss 0.319033 Top1 84.095982 Top5 97.672991 -2023-02-13 18:29:51,300 - Epoch: [162][ 80/ 135] Loss 0.321538 Top1 84.150391 Top5 97.666016 -2023-02-13 18:29:51,429 - Epoch: [162][ 90/ 135] Loss 0.320808 Top1 84.175347 Top5 97.612847 -2023-02-13 18:29:51,558 - Epoch: [162][ 100/ 135] Loss 0.323215 Top1 84.183594 Top5 97.628906 -2023-02-13 18:29:51,686 - Epoch: [162][ 110/ 135] Loss 0.324589 Top1 84.186790 Top5 97.606534 -2023-02-13 18:29:51,815 - Epoch: [162][ 120/ 135] Loss 0.321827 Top1 84.270833 Top5 97.652995 -2023-02-13 18:29:51,944 - Epoch: [162][ 130/ 135] Loss 0.320935 Top1 84.275841 Top5 97.662260 -2023-02-13 18:29:51,988 - Epoch: [162][ 135/ 135] Loss 0.320459 Top1 84.267436 Top5 97.656728 -2023-02-13 18:29:52,056 - ==> Top1: 84.267 Top5: 97.657 Loss: 0.320 - -2023-02-13 18:29:52,057 - ==> Confusion: -[[ 865 6 5 0 7 3 0 2 6 48 0 4 1 4 5 3 3 2 1 0 2] - [ 3 958 1 1 12 24 3 11 3 2 2 0 1 0 0 1 3 0 3 1 4] - [ 9 6 957 11 5 2 14 13 0 1 2 1 3 7 2 2 4 3 8 1 7] - [ 6 2 19 908 1 7 0 1 1 3 13 0 6 0 17 1 2 4 19 0 6] - [ 8 8 0 1 1000 9 1 1 1 1 0 8 3 2 8 7 1 2 0 2 3] - [ 3 21 0 5 6 980 4 9 0 4 1 9 1 14 0 2 3 2 2 1 3] - [ 3 6 11 1 0 9 1046 6 0 0 1 1 2 1 0 2 1 1 1 4 3] - [ 3 15 9 0 2 39 3 916 1 1 2 4 1 1 0 0 1 1 13 8 4] - [ 16 1 1 2 1 1 0 2 918 26 6 3 0 9 16 2 1 0 4 0 0] - [ 84 1 2 0 10 0 0 2 37 845 0 0 1 14 7 2 0 1 1 0 5] - [ 3 2 1 6 2 1 2 4 15 1 992 1 1 7 2 0 1 2 5 0 3] - [ 2 1 1 0 1 15 0 6 2 4 0 923 21 9 0 2 1 9 1 5 2] - [ 0 1 0 8 3 7 0 0 1 1 0 32 871 1 4 4 2 14 2 0 8] - [ 5 2 1 0 6 7 2 4 8 17 8 3 2 940 4 5 1 3 0 1 5] - [ 6 5 1 12 5 6 0 2 22 5 1 2 2 2 1000 0 3 4 7 0 7] - [ 5 3 8 1 7 4 3 1 0 0 0 6 10 4 0 961 12 8 0 5 8] - [ 3 9 1 1 8 3 0 0 2 0 0 1 3 3 2 9 998 3 0 4 11] - [ 7 3 0 6 3 3 2 0 0 1 1 9 10 1 0 15 0 985 1 0 4] - [ 2 6 5 6 0 2 0 26 4 0 5 2 2 0 14 1 1 4 1005 1 0] - [ 2 2 2 1 0 7 6 12 1 0 0 13 2 1 0 5 2 3 0 1085 4] - [ 169 328 213 132 179 252 102 186 94 84 227 130 320 304 160 68 205 123 172 226 9760]] - -2023-02-13 18:29:52,058 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:29:52,059 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:29:52,064 - - -2023-02-13 18:29:52,065 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:29:52,957 - Epoch: [163][ 10/ 1207] Overall Loss 0.246547 Objective Loss 0.246547 LR 0.000250 Time 0.089144 -2023-02-13 18:29:53,148 - Epoch: [163][ 20/ 1207] Overall Loss 0.244560 Objective Loss 0.244560 LR 0.000250 Time 0.054097 -2023-02-13 18:29:53,338 - Epoch: [163][ 30/ 1207] Overall Loss 0.246537 Objective Loss 0.246537 LR 0.000250 Time 0.042399 -2023-02-13 18:29:53,529 - Epoch: [163][ 40/ 1207] Overall Loss 0.252436 Objective Loss 0.252436 LR 0.000250 Time 0.036557 -2023-02-13 18:29:53,719 - Epoch: [163][ 50/ 1207] Overall Loss 0.256969 Objective Loss 0.256969 LR 0.000250 Time 0.033040 -2023-02-13 18:29:53,908 - Epoch: [163][ 60/ 1207] Overall Loss 0.259831 Objective Loss 0.259831 LR 0.000250 Time 0.030686 -2023-02-13 18:29:54,097 - Epoch: [163][ 70/ 1207] Overall Loss 0.257036 Objective Loss 0.257036 LR 0.000250 Time 0.028988 -2023-02-13 18:29:54,286 - Epoch: [163][ 80/ 1207] Overall Loss 0.254981 Objective Loss 0.254981 LR 0.000250 Time 0.027730 -2023-02-13 18:29:54,475 - Epoch: [163][ 90/ 1207] Overall Loss 0.255739 Objective Loss 0.255739 LR 0.000250 Time 0.026737 -2023-02-13 18:29:54,663 - Epoch: [163][ 100/ 1207] Overall Loss 0.254506 Objective Loss 0.254506 LR 0.000250 Time 0.025945 -2023-02-13 18:29:54,851 - Epoch: [163][ 110/ 1207] Overall Loss 0.252988 Objective Loss 0.252988 LR 0.000250 Time 0.025291 -2023-02-13 18:29:55,039 - Epoch: [163][ 120/ 1207] Overall Loss 0.251584 Objective Loss 0.251584 LR 0.000250 Time 0.024750 -2023-02-13 18:29:55,229 - Epoch: [163][ 130/ 1207] Overall Loss 0.253535 Objective Loss 0.253535 LR 0.000250 Time 0.024299 -2023-02-13 18:29:55,418 - Epoch: [163][ 140/ 1207] Overall Loss 0.253352 Objective Loss 0.253352 LR 0.000250 Time 0.023912 -2023-02-13 18:29:55,606 - Epoch: [163][ 150/ 1207] Overall Loss 0.253011 Objective Loss 0.253011 LR 0.000250 Time 0.023573 -2023-02-13 18:29:55,796 - Epoch: [163][ 160/ 1207] Overall Loss 0.253157 Objective Loss 0.253157 LR 0.000250 Time 0.023283 -2023-02-13 18:29:55,986 - Epoch: [163][ 170/ 1207] Overall Loss 0.253687 Objective Loss 0.253687 LR 0.000250 Time 0.023026 -2023-02-13 18:29:56,175 - Epoch: [163][ 180/ 1207] Overall Loss 0.254732 Objective Loss 0.254732 LR 0.000250 Time 0.022796 -2023-02-13 18:29:56,364 - Epoch: [163][ 190/ 1207] Overall Loss 0.254584 Objective Loss 0.254584 LR 0.000250 Time 0.022590 -2023-02-13 18:29:56,553 - Epoch: [163][ 200/ 1207] Overall Loss 0.254845 Objective Loss 0.254845 LR 0.000250 Time 0.022405 -2023-02-13 18:29:56,742 - Epoch: [163][ 210/ 1207] Overall Loss 0.253961 Objective Loss 0.253961 LR 0.000250 Time 0.022237 -2023-02-13 18:29:56,932 - Epoch: [163][ 220/ 1207] Overall Loss 0.254135 Objective Loss 0.254135 LR 0.000250 Time 0.022084 -2023-02-13 18:29:57,120 - Epoch: [163][ 230/ 1207] Overall Loss 0.253467 Objective Loss 0.253467 LR 0.000250 Time 0.021942 -2023-02-13 18:29:57,309 - Epoch: [163][ 240/ 1207] Overall Loss 0.253479 Objective Loss 0.253479 LR 0.000250 Time 0.021814 -2023-02-13 18:29:57,498 - Epoch: [163][ 250/ 1207] Overall Loss 0.252609 Objective Loss 0.252609 LR 0.000250 Time 0.021695 -2023-02-13 18:29:57,687 - Epoch: [163][ 260/ 1207] Overall Loss 0.252289 Objective Loss 0.252289 LR 0.000250 Time 0.021587 -2023-02-13 18:29:57,876 - Epoch: [163][ 270/ 1207] Overall Loss 0.252152 Objective Loss 0.252152 LR 0.000250 Time 0.021483 -2023-02-13 18:29:58,065 - Epoch: [163][ 280/ 1207] Overall Loss 0.252759 Objective Loss 0.252759 LR 0.000250 Time 0.021390 -2023-02-13 18:29:58,253 - Epoch: [163][ 290/ 1207] Overall Loss 0.252348 Objective Loss 0.252348 LR 0.000250 Time 0.021301 -2023-02-13 18:29:58,442 - Epoch: [163][ 300/ 1207] Overall Loss 0.252289 Objective Loss 0.252289 LR 0.000250 Time 0.021219 -2023-02-13 18:29:58,630 - Epoch: [163][ 310/ 1207] Overall Loss 0.251391 Objective Loss 0.251391 LR 0.000250 Time 0.021141 -2023-02-13 18:29:58,819 - Epoch: [163][ 320/ 1207] Overall Loss 0.251387 Objective Loss 0.251387 LR 0.000250 Time 0.021068 -2023-02-13 18:29:59,007 - Epoch: [163][ 330/ 1207] Overall Loss 0.251453 Objective Loss 0.251453 LR 0.000250 Time 0.020999 -2023-02-13 18:29:59,197 - Epoch: [163][ 340/ 1207] Overall Loss 0.251472 Objective Loss 0.251472 LR 0.000250 Time 0.020938 -2023-02-13 18:29:59,386 - Epoch: [163][ 350/ 1207] Overall Loss 0.251030 Objective Loss 0.251030 LR 0.000250 Time 0.020879 -2023-02-13 18:29:59,575 - Epoch: [163][ 360/ 1207] Overall Loss 0.251003 Objective Loss 0.251003 LR 0.000250 Time 0.020824 -2023-02-13 18:29:59,764 - Epoch: [163][ 370/ 1207] Overall Loss 0.250925 Objective Loss 0.250925 LR 0.000250 Time 0.020770 -2023-02-13 18:29:59,953 - Epoch: [163][ 380/ 1207] Overall Loss 0.251483 Objective Loss 0.251483 LR 0.000250 Time 0.020721 -2023-02-13 18:30:00,142 - Epoch: [163][ 390/ 1207] Overall Loss 0.251948 Objective Loss 0.251948 LR 0.000250 Time 0.020672 -2023-02-13 18:30:00,331 - Epoch: [163][ 400/ 1207] Overall Loss 0.251981 Objective Loss 0.251981 LR 0.000250 Time 0.020627 -2023-02-13 18:30:00,519 - Epoch: [163][ 410/ 1207] Overall Loss 0.251672 Objective Loss 0.251672 LR 0.000250 Time 0.020583 -2023-02-13 18:30:00,709 - Epoch: [163][ 420/ 1207] Overall Loss 0.251849 Objective Loss 0.251849 LR 0.000250 Time 0.020543 -2023-02-13 18:30:00,898 - Epoch: [163][ 430/ 1207] Overall Loss 0.251581 Objective Loss 0.251581 LR 0.000250 Time 0.020505 -2023-02-13 18:30:01,087 - Epoch: [163][ 440/ 1207] Overall Loss 0.251574 Objective Loss 0.251574 LR 0.000250 Time 0.020468 -2023-02-13 18:30:01,277 - Epoch: [163][ 450/ 1207] Overall Loss 0.251689 Objective Loss 0.251689 LR 0.000250 Time 0.020435 -2023-02-13 18:30:01,466 - Epoch: [163][ 460/ 1207] Overall Loss 0.252044 Objective Loss 0.252044 LR 0.000250 Time 0.020400 -2023-02-13 18:30:01,655 - Epoch: [163][ 470/ 1207] Overall Loss 0.251647 Objective Loss 0.251647 LR 0.000250 Time 0.020367 -2023-02-13 18:30:01,844 - Epoch: [163][ 480/ 1207] Overall Loss 0.251854 Objective Loss 0.251854 LR 0.000250 Time 0.020336 -2023-02-13 18:30:02,033 - Epoch: [163][ 490/ 1207] Overall Loss 0.251635 Objective Loss 0.251635 LR 0.000250 Time 0.020306 -2023-02-13 18:30:02,222 - Epoch: [163][ 500/ 1207] Overall Loss 0.251609 Objective Loss 0.251609 LR 0.000250 Time 0.020277 -2023-02-13 18:30:02,411 - Epoch: [163][ 510/ 1207] Overall Loss 0.251674 Objective Loss 0.251674 LR 0.000250 Time 0.020250 -2023-02-13 18:30:02,601 - Epoch: [163][ 520/ 1207] Overall Loss 0.251531 Objective Loss 0.251531 LR 0.000250 Time 0.020224 -2023-02-13 18:30:02,790 - Epoch: [163][ 530/ 1207] Overall Loss 0.251182 Objective Loss 0.251182 LR 0.000250 Time 0.020198 -2023-02-13 18:30:02,979 - Epoch: [163][ 540/ 1207] Overall Loss 0.250628 Objective Loss 0.250628 LR 0.000250 Time 0.020174 -2023-02-13 18:30:03,167 - Epoch: [163][ 550/ 1207] Overall Loss 0.250552 Objective Loss 0.250552 LR 0.000250 Time 0.020149 -2023-02-13 18:30:03,358 - Epoch: [163][ 560/ 1207] Overall Loss 0.250665 Objective Loss 0.250665 LR 0.000250 Time 0.020128 -2023-02-13 18:30:03,547 - Epoch: [163][ 570/ 1207] Overall Loss 0.250991 Objective Loss 0.250991 LR 0.000250 Time 0.020106 -2023-02-13 18:30:03,736 - Epoch: [163][ 580/ 1207] Overall Loss 0.251433 Objective Loss 0.251433 LR 0.000250 Time 0.020085 -2023-02-13 18:30:03,925 - Epoch: [163][ 590/ 1207] Overall Loss 0.251786 Objective Loss 0.251786 LR 0.000250 Time 0.020064 -2023-02-13 18:30:04,114 - Epoch: [163][ 600/ 1207] Overall Loss 0.252218 Objective Loss 0.252218 LR 0.000250 Time 0.020045 -2023-02-13 18:30:04,304 - Epoch: [163][ 610/ 1207] Overall Loss 0.252070 Objective Loss 0.252070 LR 0.000250 Time 0.020026 -2023-02-13 18:30:04,493 - Epoch: [163][ 620/ 1207] Overall Loss 0.251807 Objective Loss 0.251807 LR 0.000250 Time 0.020008 -2023-02-13 18:30:04,682 - Epoch: [163][ 630/ 1207] Overall Loss 0.251935 Objective Loss 0.251935 LR 0.000250 Time 0.019990 -2023-02-13 18:30:04,871 - Epoch: [163][ 640/ 1207] Overall Loss 0.252222 Objective Loss 0.252222 LR 0.000250 Time 0.019972 -2023-02-13 18:30:05,060 - Epoch: [163][ 650/ 1207] Overall Loss 0.252175 Objective Loss 0.252175 LR 0.000250 Time 0.019956 -2023-02-13 18:30:05,250 - Epoch: [163][ 660/ 1207] Overall Loss 0.252359 Objective Loss 0.252359 LR 0.000250 Time 0.019940 -2023-02-13 18:30:05,440 - Epoch: [163][ 670/ 1207] Overall Loss 0.252402 Objective Loss 0.252402 LR 0.000250 Time 0.019925 -2023-02-13 18:30:05,629 - Epoch: [163][ 680/ 1207] Overall Loss 0.252344 Objective Loss 0.252344 LR 0.000250 Time 0.019909 -2023-02-13 18:30:05,818 - Epoch: [163][ 690/ 1207] Overall Loss 0.252684 Objective Loss 0.252684 LR 0.000250 Time 0.019895 -2023-02-13 18:30:06,008 - Epoch: [163][ 700/ 1207] Overall Loss 0.252457 Objective Loss 0.252457 LR 0.000250 Time 0.019882 -2023-02-13 18:30:06,198 - Epoch: [163][ 710/ 1207] Overall Loss 0.252497 Objective Loss 0.252497 LR 0.000250 Time 0.019868 -2023-02-13 18:30:06,387 - Epoch: [163][ 720/ 1207] Overall Loss 0.252463 Objective Loss 0.252463 LR 0.000250 Time 0.019855 -2023-02-13 18:30:06,577 - Epoch: [163][ 730/ 1207] Overall Loss 0.252833 Objective Loss 0.252833 LR 0.000250 Time 0.019842 -2023-02-13 18:30:06,767 - Epoch: [163][ 740/ 1207] Overall Loss 0.252887 Objective Loss 0.252887 LR 0.000250 Time 0.019830 -2023-02-13 18:30:06,957 - Epoch: [163][ 750/ 1207] Overall Loss 0.252957 Objective Loss 0.252957 LR 0.000250 Time 0.019818 -2023-02-13 18:30:07,146 - Epoch: [163][ 760/ 1207] Overall Loss 0.252901 Objective Loss 0.252901 LR 0.000250 Time 0.019806 -2023-02-13 18:30:07,336 - Epoch: [163][ 770/ 1207] Overall Loss 0.252956 Objective Loss 0.252956 LR 0.000250 Time 0.019795 -2023-02-13 18:30:07,525 - Epoch: [163][ 780/ 1207] Overall Loss 0.252912 Objective Loss 0.252912 LR 0.000250 Time 0.019784 -2023-02-13 18:30:07,715 - Epoch: [163][ 790/ 1207] Overall Loss 0.253242 Objective Loss 0.253242 LR 0.000250 Time 0.019773 -2023-02-13 18:30:07,904 - Epoch: [163][ 800/ 1207] Overall Loss 0.253456 Objective Loss 0.253456 LR 0.000250 Time 0.019761 -2023-02-13 18:30:08,093 - Epoch: [163][ 810/ 1207] Overall Loss 0.253553 Objective Loss 0.253553 LR 0.000250 Time 0.019750 -2023-02-13 18:30:08,283 - Epoch: [163][ 820/ 1207] Overall Loss 0.253801 Objective Loss 0.253801 LR 0.000250 Time 0.019741 -2023-02-13 18:30:08,473 - Epoch: [163][ 830/ 1207] Overall Loss 0.253857 Objective Loss 0.253857 LR 0.000250 Time 0.019731 -2023-02-13 18:30:08,663 - Epoch: [163][ 840/ 1207] Overall Loss 0.253888 Objective Loss 0.253888 LR 0.000250 Time 0.019722 -2023-02-13 18:30:08,852 - Epoch: [163][ 850/ 1207] Overall Loss 0.253721 Objective Loss 0.253721 LR 0.000250 Time 0.019712 -2023-02-13 18:30:09,041 - Epoch: [163][ 860/ 1207] Overall Loss 0.253707 Objective Loss 0.253707 LR 0.000250 Time 0.019702 -2023-02-13 18:30:09,231 - Epoch: [163][ 870/ 1207] Overall Loss 0.253632 Objective Loss 0.253632 LR 0.000250 Time 0.019693 -2023-02-13 18:30:09,420 - Epoch: [163][ 880/ 1207] Overall Loss 0.253439 Objective Loss 0.253439 LR 0.000250 Time 0.019684 -2023-02-13 18:30:09,609 - Epoch: [163][ 890/ 1207] Overall Loss 0.253378 Objective Loss 0.253378 LR 0.000250 Time 0.019675 -2023-02-13 18:30:09,798 - Epoch: [163][ 900/ 1207] Overall Loss 0.253471 Objective Loss 0.253471 LR 0.000250 Time 0.019667 -2023-02-13 18:30:09,988 - Epoch: [163][ 910/ 1207] Overall Loss 0.253295 Objective Loss 0.253295 LR 0.000250 Time 0.019658 -2023-02-13 18:30:10,176 - Epoch: [163][ 920/ 1207] Overall Loss 0.253360 Objective Loss 0.253360 LR 0.000250 Time 0.019649 -2023-02-13 18:30:10,367 - Epoch: [163][ 930/ 1207] Overall Loss 0.253158 Objective Loss 0.253158 LR 0.000250 Time 0.019642 -2023-02-13 18:30:10,556 - Epoch: [163][ 940/ 1207] Overall Loss 0.253099 Objective Loss 0.253099 LR 0.000250 Time 0.019634 -2023-02-13 18:30:10,746 - Epoch: [163][ 950/ 1207] Overall Loss 0.253256 Objective Loss 0.253256 LR 0.000250 Time 0.019627 -2023-02-13 18:30:10,936 - Epoch: [163][ 960/ 1207] Overall Loss 0.253332 Objective Loss 0.253332 LR 0.000250 Time 0.019620 -2023-02-13 18:30:11,125 - Epoch: [163][ 970/ 1207] Overall Loss 0.253250 Objective Loss 0.253250 LR 0.000250 Time 0.019612 -2023-02-13 18:30:11,315 - Epoch: [163][ 980/ 1207] Overall Loss 0.253232 Objective Loss 0.253232 LR 0.000250 Time 0.019606 -2023-02-13 18:30:11,506 - Epoch: [163][ 990/ 1207] Overall Loss 0.253111 Objective Loss 0.253111 LR 0.000250 Time 0.019600 -2023-02-13 18:30:11,696 - Epoch: [163][ 1000/ 1207] Overall Loss 0.252982 Objective Loss 0.252982 LR 0.000250 Time 0.019593 -2023-02-13 18:30:11,885 - Epoch: [163][ 1010/ 1207] Overall Loss 0.253274 Objective Loss 0.253274 LR 0.000250 Time 0.019586 -2023-02-13 18:30:12,075 - Epoch: [163][ 1020/ 1207] Overall Loss 0.253215 Objective Loss 0.253215 LR 0.000250 Time 0.019580 -2023-02-13 18:30:12,265 - Epoch: [163][ 1030/ 1207] Overall Loss 0.253440 Objective Loss 0.253440 LR 0.000250 Time 0.019574 -2023-02-13 18:30:12,454 - Epoch: [163][ 1040/ 1207] Overall Loss 0.253499 Objective Loss 0.253499 LR 0.000250 Time 0.019568 -2023-02-13 18:30:12,643 - Epoch: [163][ 1050/ 1207] Overall Loss 0.253297 Objective Loss 0.253297 LR 0.000250 Time 0.019561 -2023-02-13 18:30:12,832 - Epoch: [163][ 1060/ 1207] Overall Loss 0.253185 Objective Loss 0.253185 LR 0.000250 Time 0.019555 -2023-02-13 18:30:13,021 - Epoch: [163][ 1070/ 1207] Overall Loss 0.253130 Objective Loss 0.253130 LR 0.000250 Time 0.019548 -2023-02-13 18:30:13,210 - Epoch: [163][ 1080/ 1207] Overall Loss 0.253053 Objective Loss 0.253053 LR 0.000250 Time 0.019541 -2023-02-13 18:30:13,399 - Epoch: [163][ 1090/ 1207] Overall Loss 0.253138 Objective Loss 0.253138 LR 0.000250 Time 0.019535 -2023-02-13 18:30:13,588 - Epoch: [163][ 1100/ 1207] Overall Loss 0.253186 Objective Loss 0.253186 LR 0.000250 Time 0.019529 -2023-02-13 18:30:13,777 - Epoch: [163][ 1110/ 1207] Overall Loss 0.253254 Objective Loss 0.253254 LR 0.000250 Time 0.019523 -2023-02-13 18:30:13,966 - Epoch: [163][ 1120/ 1207] Overall Loss 0.253426 Objective Loss 0.253426 LR 0.000250 Time 0.019517 -2023-02-13 18:30:14,155 - Epoch: [163][ 1130/ 1207] Overall Loss 0.253450 Objective Loss 0.253450 LR 0.000250 Time 0.019511 -2023-02-13 18:30:14,345 - Epoch: [163][ 1140/ 1207] Overall Loss 0.253343 Objective Loss 0.253343 LR 0.000250 Time 0.019507 -2023-02-13 18:30:14,535 - Epoch: [163][ 1150/ 1207] Overall Loss 0.253153 Objective Loss 0.253153 LR 0.000250 Time 0.019502 -2023-02-13 18:30:14,725 - Epoch: [163][ 1160/ 1207] Overall Loss 0.253148 Objective Loss 0.253148 LR 0.000250 Time 0.019497 -2023-02-13 18:30:14,914 - Epoch: [163][ 1170/ 1207] Overall Loss 0.253074 Objective Loss 0.253074 LR 0.000250 Time 0.019492 -2023-02-13 18:30:15,103 - Epoch: [163][ 1180/ 1207] Overall Loss 0.253142 Objective Loss 0.253142 LR 0.000250 Time 0.019487 -2023-02-13 18:30:15,293 - Epoch: [163][ 1190/ 1207] Overall Loss 0.253202 Objective Loss 0.253202 LR 0.000250 Time 0.019482 -2023-02-13 18:30:15,540 - Epoch: [163][ 1200/ 1207] Overall Loss 0.253259 Objective Loss 0.253259 LR 0.000250 Time 0.019525 -2023-02-13 18:30:15,653 - Epoch: [163][ 1207/ 1207] Overall Loss 0.253354 Objective Loss 0.253354 Top1 88.414634 Top5 97.560976 LR 0.000250 Time 0.019506 -2023-02-13 18:30:15,740 - --- validate (epoch=163)----------- -2023-02-13 18:30:15,740 - 34311 samples (256 per mini-batch) -2023-02-13 18:30:16,140 - Epoch: [163][ 10/ 135] Loss 0.304505 Top1 85.351562 Top5 98.046875 -2023-02-13 18:30:16,264 - Epoch: [163][ 20/ 135] Loss 0.308913 Top1 85.175781 Top5 97.949219 -2023-02-13 18:30:16,408 - Epoch: [163][ 30/ 135] Loss 0.315037 Top1 84.804688 Top5 97.812500 -2023-02-13 18:30:16,549 - Epoch: [163][ 40/ 135] Loss 0.308140 Top1 84.931641 Top5 97.880859 -2023-02-13 18:30:16,679 - Epoch: [163][ 50/ 135] Loss 0.317314 Top1 84.515625 Top5 97.796875 -2023-02-13 18:30:16,808 - Epoch: [163][ 60/ 135] Loss 0.318152 Top1 84.609375 Top5 97.740885 -2023-02-13 18:30:16,937 - Epoch: [163][ 70/ 135] Loss 0.318461 Top1 84.542411 Top5 97.779018 -2023-02-13 18:30:17,066 - Epoch: [163][ 80/ 135] Loss 0.318407 Top1 84.541016 Top5 97.705078 -2023-02-13 18:30:17,195 - Epoch: [163][ 90/ 135] Loss 0.319457 Top1 84.522569 Top5 97.695312 -2023-02-13 18:30:17,324 - Epoch: [163][ 100/ 135] Loss 0.316249 Top1 84.695312 Top5 97.738281 -2023-02-13 18:30:17,453 - Epoch: [163][ 110/ 135] Loss 0.317524 Top1 84.651989 Top5 97.748580 -2023-02-13 18:30:17,583 - Epoch: [163][ 120/ 135] Loss 0.317857 Top1 84.606120 Top5 97.757161 -2023-02-13 18:30:17,714 - Epoch: [163][ 130/ 135] Loss 0.318374 Top1 84.588341 Top5 97.752404 -2023-02-13 18:30:17,760 - Epoch: [163][ 135/ 135] Loss 0.318219 Top1 84.564717 Top5 97.726677 -2023-02-13 18:30:17,833 - ==> Top1: 84.565 Top5: 97.727 Loss: 0.318 - -2023-02-13 18:30:17,834 - ==> Confusion: -[[ 860 4 6 0 7 3 0 3 6 46 1 5 0 5 4 1 4 3 0 2 7] - [ 3 947 1 2 5 22 3 22 5 2 0 1 1 0 0 2 4 0 3 4 6] - [ 8 5 952 8 6 1 16 13 0 1 4 2 1 5 2 5 4 2 5 6 12] - [ 6 1 14 908 1 6 0 2 4 3 12 1 7 1 15 3 3 4 17 0 8] - [ 17 9 0 0 982 12 1 5 2 2 1 6 1 3 7 5 5 1 0 3 4] - [ 3 16 0 4 1 982 4 16 1 4 2 10 3 10 0 4 5 1 2 2 0] - [ 1 1 8 1 1 6 1041 8 1 1 3 1 2 2 0 5 3 2 1 5 6] - [ 2 5 9 0 0 25 2 942 0 1 0 8 2 2 0 0 2 1 11 6 6] - [ 16 3 1 2 1 0 0 1 909 32 7 2 0 14 9 3 2 1 4 1 1] - [ 81 0 2 0 4 1 0 1 40 850 0 0 0 17 6 2 1 3 1 0 3] - [ 0 2 1 6 0 2 5 5 11 2 991 3 1 6 2 0 1 0 7 0 6] - [ 3 2 1 0 1 8 1 5 3 1 1 935 16 7 2 2 3 8 2 3 1] - [ 0 0 0 7 2 3 0 2 2 0 0 42 864 0 2 8 4 15 0 0 8] - [ 4 2 1 0 7 8 1 3 8 11 7 5 1 942 1 5 6 2 0 3 7] - [ 4 6 0 13 5 5 0 2 29 10 6 1 4 1 976 1 3 4 11 1 10] - [ 2 3 4 1 7 1 3 0 0 0 0 6 3 3 0 976 10 12 0 7 8] - [ 1 7 1 0 8 3 0 0 1 0 0 2 3 3 2 9 1005 3 0 2 11] - [ 4 2 0 2 1 1 2 0 0 0 3 11 11 2 0 15 0 988 0 3 6] - [ 2 4 3 6 0 1 0 23 1 0 4 2 9 0 10 2 1 3 1012 0 3] - [ 2 2 0 0 1 6 6 8 1 0 0 19 1 4 1 5 1 5 0 1083 3] - [ 156 260 188 99 121 212 86 217 103 97 228 148 307 314 126 98 260 106 177 261 9870]] - -2023-02-13 18:30:17,835 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:30:17,835 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:30:17,841 - - -2023-02-13 18:30:17,841 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:30:18,727 - Epoch: [164][ 10/ 1207] Overall Loss 0.254445 Objective Loss 0.254445 LR 0.000250 Time 0.088524 -2023-02-13 18:30:18,930 - Epoch: [164][ 20/ 1207] Overall Loss 0.263548 Objective Loss 0.263548 LR 0.000250 Time 0.054384 -2023-02-13 18:30:19,121 - Epoch: [164][ 30/ 1207] Overall Loss 0.262177 Objective Loss 0.262177 LR 0.000250 Time 0.042606 -2023-02-13 18:30:19,315 - Epoch: [164][ 40/ 1207] Overall Loss 0.254966 Objective Loss 0.254966 LR 0.000250 Time 0.036812 -2023-02-13 18:30:19,507 - Epoch: [164][ 50/ 1207] Overall Loss 0.252284 Objective Loss 0.252284 LR 0.000250 Time 0.033272 -2023-02-13 18:30:19,702 - Epoch: [164][ 60/ 1207] Overall Loss 0.252475 Objective Loss 0.252475 LR 0.000250 Time 0.030968 -2023-02-13 18:30:19,894 - Epoch: [164][ 70/ 1207] Overall Loss 0.250637 Objective Loss 0.250637 LR 0.000250 Time 0.029277 -2023-02-13 18:30:20,088 - Epoch: [164][ 80/ 1207] Overall Loss 0.248633 Objective Loss 0.248633 LR 0.000250 Time 0.028045 -2023-02-13 18:30:20,280 - Epoch: [164][ 90/ 1207] Overall Loss 0.247965 Objective Loss 0.247965 LR 0.000250 Time 0.027053 -2023-02-13 18:30:20,480 - Epoch: [164][ 100/ 1207] Overall Loss 0.248852 Objective Loss 0.248852 LR 0.000250 Time 0.026343 -2023-02-13 18:30:20,684 - Epoch: [164][ 110/ 1207] Overall Loss 0.250827 Objective Loss 0.250827 LR 0.000250 Time 0.025806 -2023-02-13 18:30:20,889 - Epoch: [164][ 120/ 1207] Overall Loss 0.249161 Objective Loss 0.249161 LR 0.000250 Time 0.025354 -2023-02-13 18:30:21,085 - Epoch: [164][ 130/ 1207] Overall Loss 0.249253 Objective Loss 0.249253 LR 0.000250 Time 0.024909 -2023-02-13 18:30:21,280 - Epoch: [164][ 140/ 1207] Overall Loss 0.248922 Objective Loss 0.248922 LR 0.000250 Time 0.024522 -2023-02-13 18:30:21,476 - Epoch: [164][ 150/ 1207] Overall Loss 0.249255 Objective Loss 0.249255 LR 0.000250 Time 0.024192 -2023-02-13 18:30:21,676 - Epoch: [164][ 160/ 1207] Overall Loss 0.249346 Objective Loss 0.249346 LR 0.000250 Time 0.023930 -2023-02-13 18:30:21,873 - Epoch: [164][ 170/ 1207] Overall Loss 0.247931 Objective Loss 0.247931 LR 0.000250 Time 0.023678 -2023-02-13 18:30:22,073 - Epoch: [164][ 180/ 1207] Overall Loss 0.248312 Objective Loss 0.248312 LR 0.000250 Time 0.023470 -2023-02-13 18:30:22,269 - Epoch: [164][ 190/ 1207] Overall Loss 0.249126 Objective Loss 0.249126 LR 0.000250 Time 0.023267 -2023-02-13 18:30:22,470 - Epoch: [164][ 200/ 1207] Overall Loss 0.248987 Objective Loss 0.248987 LR 0.000250 Time 0.023104 -2023-02-13 18:30:22,667 - Epoch: [164][ 210/ 1207] Overall Loss 0.248824 Objective Loss 0.248824 LR 0.000250 Time 0.022939 -2023-02-13 18:30:22,858 - Epoch: [164][ 220/ 1207] Overall Loss 0.248147 Objective Loss 0.248147 LR 0.000250 Time 0.022763 -2023-02-13 18:30:23,047 - Epoch: [164][ 230/ 1207] Overall Loss 0.248662 Objective Loss 0.248662 LR 0.000250 Time 0.022595 -2023-02-13 18:30:23,235 - Epoch: [164][ 240/ 1207] Overall Loss 0.247937 Objective Loss 0.247937 LR 0.000250 Time 0.022437 -2023-02-13 18:30:23,425 - Epoch: [164][ 250/ 1207] Overall Loss 0.249308 Objective Loss 0.249308 LR 0.000250 Time 0.022296 -2023-02-13 18:30:23,614 - Epoch: [164][ 260/ 1207] Overall Loss 0.249391 Objective Loss 0.249391 LR 0.000250 Time 0.022165 -2023-02-13 18:30:23,803 - Epoch: [164][ 270/ 1207] Overall Loss 0.249259 Objective Loss 0.249259 LR 0.000250 Time 0.022044 -2023-02-13 18:30:23,992 - Epoch: [164][ 280/ 1207] Overall Loss 0.249885 Objective Loss 0.249885 LR 0.000250 Time 0.021928 -2023-02-13 18:30:24,181 - Epoch: [164][ 290/ 1207] Overall Loss 0.249943 Objective Loss 0.249943 LR 0.000250 Time 0.021823 -2023-02-13 18:30:24,370 - Epoch: [164][ 300/ 1207] Overall Loss 0.249855 Objective Loss 0.249855 LR 0.000250 Time 0.021723 -2023-02-13 18:30:24,559 - Epoch: [164][ 310/ 1207] Overall Loss 0.249731 Objective Loss 0.249731 LR 0.000250 Time 0.021633 -2023-02-13 18:30:24,748 - Epoch: [164][ 320/ 1207] Overall Loss 0.249669 Objective Loss 0.249669 LR 0.000250 Time 0.021545 -2023-02-13 18:30:24,936 - Epoch: [164][ 330/ 1207] Overall Loss 0.249474 Objective Loss 0.249474 LR 0.000250 Time 0.021462 -2023-02-13 18:30:25,125 - Epoch: [164][ 340/ 1207] Overall Loss 0.249385 Objective Loss 0.249385 LR 0.000250 Time 0.021385 -2023-02-13 18:30:25,314 - Epoch: [164][ 350/ 1207] Overall Loss 0.250002 Objective Loss 0.250002 LR 0.000250 Time 0.021312 -2023-02-13 18:30:25,503 - Epoch: [164][ 360/ 1207] Overall Loss 0.250267 Objective Loss 0.250267 LR 0.000250 Time 0.021244 -2023-02-13 18:30:25,692 - Epoch: [164][ 370/ 1207] Overall Loss 0.250455 Objective Loss 0.250455 LR 0.000250 Time 0.021180 -2023-02-13 18:30:25,881 - Epoch: [164][ 380/ 1207] Overall Loss 0.250412 Objective Loss 0.250412 LR 0.000250 Time 0.021118 -2023-02-13 18:30:26,071 - Epoch: [164][ 390/ 1207] Overall Loss 0.250220 Objective Loss 0.250220 LR 0.000250 Time 0.021063 -2023-02-13 18:30:26,259 - Epoch: [164][ 400/ 1207] Overall Loss 0.250226 Objective Loss 0.250226 LR 0.000250 Time 0.021007 -2023-02-13 18:30:26,449 - Epoch: [164][ 410/ 1207] Overall Loss 0.250605 Objective Loss 0.250605 LR 0.000250 Time 0.020958 -2023-02-13 18:30:26,639 - Epoch: [164][ 420/ 1207] Overall Loss 0.250520 Objective Loss 0.250520 LR 0.000250 Time 0.020908 -2023-02-13 18:30:26,828 - Epoch: [164][ 430/ 1207] Overall Loss 0.249963 Objective Loss 0.249963 LR 0.000250 Time 0.020861 -2023-02-13 18:30:27,017 - Epoch: [164][ 440/ 1207] Overall Loss 0.250222 Objective Loss 0.250222 LR 0.000250 Time 0.020816 -2023-02-13 18:30:27,205 - Epoch: [164][ 450/ 1207] Overall Loss 0.250168 Objective Loss 0.250168 LR 0.000250 Time 0.020771 -2023-02-13 18:30:27,395 - Epoch: [164][ 460/ 1207] Overall Loss 0.250478 Objective Loss 0.250478 LR 0.000250 Time 0.020731 -2023-02-13 18:30:27,584 - Epoch: [164][ 470/ 1207] Overall Loss 0.250086 Objective Loss 0.250086 LR 0.000250 Time 0.020691 -2023-02-13 18:30:27,773 - Epoch: [164][ 480/ 1207] Overall Loss 0.250543 Objective Loss 0.250543 LR 0.000250 Time 0.020653 -2023-02-13 18:30:27,962 - Epoch: [164][ 490/ 1207] Overall Loss 0.250163 Objective Loss 0.250163 LR 0.000250 Time 0.020616 -2023-02-13 18:30:28,150 - Epoch: [164][ 500/ 1207] Overall Loss 0.250306 Objective Loss 0.250306 LR 0.000250 Time 0.020580 -2023-02-13 18:30:28,339 - Epoch: [164][ 510/ 1207] Overall Loss 0.250615 Objective Loss 0.250615 LR 0.000250 Time 0.020547 -2023-02-13 18:30:28,528 - Epoch: [164][ 520/ 1207] Overall Loss 0.249805 Objective Loss 0.249805 LR 0.000250 Time 0.020514 -2023-02-13 18:30:28,717 - Epoch: [164][ 530/ 1207] Overall Loss 0.249640 Objective Loss 0.249640 LR 0.000250 Time 0.020482 -2023-02-13 18:30:28,906 - Epoch: [164][ 540/ 1207] Overall Loss 0.250161 Objective Loss 0.250161 LR 0.000250 Time 0.020452 -2023-02-13 18:30:29,094 - Epoch: [164][ 550/ 1207] Overall Loss 0.249662 Objective Loss 0.249662 LR 0.000250 Time 0.020422 -2023-02-13 18:30:29,283 - Epoch: [164][ 560/ 1207] Overall Loss 0.249981 Objective Loss 0.249981 LR 0.000250 Time 0.020394 -2023-02-13 18:30:29,473 - Epoch: [164][ 570/ 1207] Overall Loss 0.250150 Objective Loss 0.250150 LR 0.000250 Time 0.020368 -2023-02-13 18:30:29,662 - Epoch: [164][ 580/ 1207] Overall Loss 0.250024 Objective Loss 0.250024 LR 0.000250 Time 0.020342 -2023-02-13 18:30:29,850 - Epoch: [164][ 590/ 1207] Overall Loss 0.249961 Objective Loss 0.249961 LR 0.000250 Time 0.020317 -2023-02-13 18:30:30,039 - Epoch: [164][ 600/ 1207] Overall Loss 0.249871 Objective Loss 0.249871 LR 0.000250 Time 0.020292 -2023-02-13 18:30:30,228 - Epoch: [164][ 610/ 1207] Overall Loss 0.249729 Objective Loss 0.249729 LR 0.000250 Time 0.020268 -2023-02-13 18:30:30,417 - Epoch: [164][ 620/ 1207] Overall Loss 0.249462 Objective Loss 0.249462 LR 0.000250 Time 0.020245 -2023-02-13 18:30:30,605 - Epoch: [164][ 630/ 1207] Overall Loss 0.249346 Objective Loss 0.249346 LR 0.000250 Time 0.020222 -2023-02-13 18:30:30,794 - Epoch: [164][ 640/ 1207] Overall Loss 0.249565 Objective Loss 0.249565 LR 0.000250 Time 0.020200 -2023-02-13 18:30:30,983 - Epoch: [164][ 650/ 1207] Overall Loss 0.249325 Objective Loss 0.249325 LR 0.000250 Time 0.020180 -2023-02-13 18:30:31,171 - Epoch: [164][ 660/ 1207] Overall Loss 0.249070 Objective Loss 0.249070 LR 0.000250 Time 0.020159 -2023-02-13 18:30:31,362 - Epoch: [164][ 670/ 1207] Overall Loss 0.249028 Objective Loss 0.249028 LR 0.000250 Time 0.020143 -2023-02-13 18:30:31,555 - Epoch: [164][ 680/ 1207] Overall Loss 0.249069 Objective Loss 0.249069 LR 0.000250 Time 0.020129 -2023-02-13 18:30:31,749 - Epoch: [164][ 690/ 1207] Overall Loss 0.249211 Objective Loss 0.249211 LR 0.000250 Time 0.020119 -2023-02-13 18:30:31,942 - Epoch: [164][ 700/ 1207] Overall Loss 0.249208 Objective Loss 0.249208 LR 0.000250 Time 0.020107 -2023-02-13 18:30:32,136 - Epoch: [164][ 710/ 1207] Overall Loss 0.249438 Objective Loss 0.249438 LR 0.000250 Time 0.020096 -2023-02-13 18:30:32,329 - Epoch: [164][ 720/ 1207] Overall Loss 0.249484 Objective Loss 0.249484 LR 0.000250 Time 0.020084 -2023-02-13 18:30:32,524 - Epoch: [164][ 730/ 1207] Overall Loss 0.249430 Objective Loss 0.249430 LR 0.000250 Time 0.020076 -2023-02-13 18:30:32,716 - Epoch: [164][ 740/ 1207] Overall Loss 0.249437 Objective Loss 0.249437 LR 0.000250 Time 0.020063 -2023-02-13 18:30:32,910 - Epoch: [164][ 750/ 1207] Overall Loss 0.249584 Objective Loss 0.249584 LR 0.000250 Time 0.020054 -2023-02-13 18:30:33,103 - Epoch: [164][ 760/ 1207] Overall Loss 0.249443 Objective Loss 0.249443 LR 0.000250 Time 0.020043 -2023-02-13 18:30:33,299 - Epoch: [164][ 770/ 1207] Overall Loss 0.249379 Objective Loss 0.249379 LR 0.000250 Time 0.020037 -2023-02-13 18:30:33,493 - Epoch: [164][ 780/ 1207] Overall Loss 0.249365 Objective Loss 0.249365 LR 0.000250 Time 0.020028 -2023-02-13 18:30:33,689 - Epoch: [164][ 790/ 1207] Overall Loss 0.249274 Objective Loss 0.249274 LR 0.000250 Time 0.020023 -2023-02-13 18:30:33,881 - Epoch: [164][ 800/ 1207] Overall Loss 0.249489 Objective Loss 0.249489 LR 0.000250 Time 0.020013 -2023-02-13 18:30:34,077 - Epoch: [164][ 810/ 1207] Overall Loss 0.249392 Objective Loss 0.249392 LR 0.000250 Time 0.020007 -2023-02-13 18:30:34,270 - Epoch: [164][ 820/ 1207] Overall Loss 0.249273 Objective Loss 0.249273 LR 0.000250 Time 0.019998 -2023-02-13 18:30:34,467 - Epoch: [164][ 830/ 1207] Overall Loss 0.249342 Objective Loss 0.249342 LR 0.000250 Time 0.019993 -2023-02-13 18:30:34,660 - Epoch: [164][ 840/ 1207] Overall Loss 0.249610 Objective Loss 0.249610 LR 0.000250 Time 0.019985 -2023-02-13 18:30:34,856 - Epoch: [164][ 850/ 1207] Overall Loss 0.249885 Objective Loss 0.249885 LR 0.000250 Time 0.019980 -2023-02-13 18:30:35,049 - Epoch: [164][ 860/ 1207] Overall Loss 0.249684 Objective Loss 0.249684 LR 0.000250 Time 0.019971 -2023-02-13 18:30:35,245 - Epoch: [164][ 870/ 1207] Overall Loss 0.249358 Objective Loss 0.249358 LR 0.000250 Time 0.019966 -2023-02-13 18:30:35,439 - Epoch: [164][ 880/ 1207] Overall Loss 0.249168 Objective Loss 0.249168 LR 0.000250 Time 0.019960 -2023-02-13 18:30:35,634 - Epoch: [164][ 890/ 1207] Overall Loss 0.249362 Objective Loss 0.249362 LR 0.000250 Time 0.019955 -2023-02-13 18:30:35,828 - Epoch: [164][ 900/ 1207] Overall Loss 0.249386 Objective Loss 0.249386 LR 0.000250 Time 0.019948 -2023-02-13 18:30:36,025 - Epoch: [164][ 910/ 1207] Overall Loss 0.249221 Objective Loss 0.249221 LR 0.000250 Time 0.019945 -2023-02-13 18:30:36,218 - Epoch: [164][ 920/ 1207] Overall Loss 0.249142 Objective Loss 0.249142 LR 0.000250 Time 0.019937 -2023-02-13 18:30:36,414 - Epoch: [164][ 930/ 1207] Overall Loss 0.249526 Objective Loss 0.249526 LR 0.000250 Time 0.019933 -2023-02-13 18:30:36,607 - Epoch: [164][ 940/ 1207] Overall Loss 0.249630 Objective Loss 0.249630 LR 0.000250 Time 0.019926 -2023-02-13 18:30:36,803 - Epoch: [164][ 950/ 1207] Overall Loss 0.249720 Objective Loss 0.249720 LR 0.000250 Time 0.019922 -2023-02-13 18:30:36,996 - Epoch: [164][ 960/ 1207] Overall Loss 0.249457 Objective Loss 0.249457 LR 0.000250 Time 0.019915 -2023-02-13 18:30:37,192 - Epoch: [164][ 970/ 1207] Overall Loss 0.249518 Objective Loss 0.249518 LR 0.000250 Time 0.019912 -2023-02-13 18:30:37,385 - Epoch: [164][ 980/ 1207] Overall Loss 0.249581 Objective Loss 0.249581 LR 0.000250 Time 0.019905 -2023-02-13 18:30:37,582 - Epoch: [164][ 990/ 1207] Overall Loss 0.249600 Objective Loss 0.249600 LR 0.000250 Time 0.019902 -2023-02-13 18:30:37,775 - Epoch: [164][ 1000/ 1207] Overall Loss 0.249611 Objective Loss 0.249611 LR 0.000250 Time 0.019896 -2023-02-13 18:30:37,971 - Epoch: [164][ 1010/ 1207] Overall Loss 0.249652 Objective Loss 0.249652 LR 0.000250 Time 0.019893 -2023-02-13 18:30:38,164 - Epoch: [164][ 1020/ 1207] Overall Loss 0.249559 Objective Loss 0.249559 LR 0.000250 Time 0.019887 -2023-02-13 18:30:38,360 - Epoch: [164][ 1030/ 1207] Overall Loss 0.249789 Objective Loss 0.249789 LR 0.000250 Time 0.019884 -2023-02-13 18:30:38,554 - Epoch: [164][ 1040/ 1207] Overall Loss 0.249938 Objective Loss 0.249938 LR 0.000250 Time 0.019878 -2023-02-13 18:30:38,750 - Epoch: [164][ 1050/ 1207] Overall Loss 0.249904 Objective Loss 0.249904 LR 0.000250 Time 0.019875 -2023-02-13 18:30:38,943 - Epoch: [164][ 1060/ 1207] Overall Loss 0.249827 Objective Loss 0.249827 LR 0.000250 Time 0.019870 -2023-02-13 18:30:39,139 - Epoch: [164][ 1070/ 1207] Overall Loss 0.250028 Objective Loss 0.250028 LR 0.000250 Time 0.019867 -2023-02-13 18:30:39,332 - Epoch: [164][ 1080/ 1207] Overall Loss 0.249994 Objective Loss 0.249994 LR 0.000250 Time 0.019861 -2023-02-13 18:30:39,528 - Epoch: [164][ 1090/ 1207] Overall Loss 0.249845 Objective Loss 0.249845 LR 0.000250 Time 0.019859 -2023-02-13 18:30:39,721 - Epoch: [164][ 1100/ 1207] Overall Loss 0.249930 Objective Loss 0.249930 LR 0.000250 Time 0.019853 -2023-02-13 18:30:39,917 - Epoch: [164][ 1110/ 1207] Overall Loss 0.249620 Objective Loss 0.249620 LR 0.000250 Time 0.019851 -2023-02-13 18:30:40,110 - Epoch: [164][ 1120/ 1207] Overall Loss 0.249549 Objective Loss 0.249549 LR 0.000250 Time 0.019845 -2023-02-13 18:30:40,306 - Epoch: [164][ 1130/ 1207] Overall Loss 0.249646 Objective Loss 0.249646 LR 0.000250 Time 0.019842 -2023-02-13 18:30:40,499 - Epoch: [164][ 1140/ 1207] Overall Loss 0.249611 Objective Loss 0.249611 LR 0.000250 Time 0.019837 -2023-02-13 18:30:40,695 - Epoch: [164][ 1150/ 1207] Overall Loss 0.249616 Objective Loss 0.249616 LR 0.000250 Time 0.019835 -2023-02-13 18:30:40,888 - Epoch: [164][ 1160/ 1207] Overall Loss 0.249701 Objective Loss 0.249701 LR 0.000250 Time 0.019830 -2023-02-13 18:30:41,085 - Epoch: [164][ 1170/ 1207] Overall Loss 0.249455 Objective Loss 0.249455 LR 0.000250 Time 0.019828 -2023-02-13 18:30:41,277 - Epoch: [164][ 1180/ 1207] Overall Loss 0.249540 Objective Loss 0.249540 LR 0.000250 Time 0.019823 -2023-02-13 18:30:41,474 - Epoch: [164][ 1190/ 1207] Overall Loss 0.249572 Objective Loss 0.249572 LR 0.000250 Time 0.019822 -2023-02-13 18:30:41,721 - Epoch: [164][ 1200/ 1207] Overall Loss 0.249664 Objective Loss 0.249664 LR 0.000250 Time 0.019863 -2023-02-13 18:30:41,837 - Epoch: [164][ 1207/ 1207] Overall Loss 0.249671 Objective Loss 0.249671 Top1 83.231707 Top5 96.951220 LR 0.000250 Time 0.019843 -2023-02-13 18:30:41,909 - --- validate (epoch=164)----------- -2023-02-13 18:30:41,910 - 34311 samples (256 per mini-batch) -2023-02-13 18:30:42,318 - Epoch: [164][ 10/ 135] Loss 0.314481 Top1 83.828125 Top5 97.539062 -2023-02-13 18:30:42,448 - Epoch: [164][ 20/ 135] Loss 0.308019 Top1 84.277344 Top5 97.519531 -2023-02-13 18:30:42,593 - Epoch: [164][ 30/ 135] Loss 0.315864 Top1 83.919271 Top5 97.369792 -2023-02-13 18:30:42,737 - Epoch: [164][ 40/ 135] Loss 0.320365 Top1 83.740234 Top5 97.314453 -2023-02-13 18:30:42,877 - Epoch: [164][ 50/ 135] Loss 0.320222 Top1 83.843750 Top5 97.367188 -2023-02-13 18:30:43,018 - Epoch: [164][ 60/ 135] Loss 0.318039 Top1 83.782552 Top5 97.337240 -2023-02-13 18:30:43,157 - Epoch: [164][ 70/ 135] Loss 0.318006 Top1 83.861607 Top5 97.354911 -2023-02-13 18:30:43,298 - Epoch: [164][ 80/ 135] Loss 0.315441 Top1 83.964844 Top5 97.407227 -2023-02-13 18:30:43,426 - Epoch: [164][ 90/ 135] Loss 0.314802 Top1 83.932292 Top5 97.387153 -2023-02-13 18:30:43,554 - Epoch: [164][ 100/ 135] Loss 0.313145 Top1 83.988281 Top5 97.414062 -2023-02-13 18:30:43,683 - Epoch: [164][ 110/ 135] Loss 0.314528 Top1 83.973722 Top5 97.407670 -2023-02-13 18:30:43,813 - Epoch: [164][ 120/ 135] Loss 0.314889 Top1 83.981120 Top5 97.454427 -2023-02-13 18:30:43,940 - Epoch: [164][ 130/ 135] Loss 0.315422 Top1 83.933293 Top5 97.493990 -2023-02-13 18:30:43,985 - Epoch: [164][ 135/ 135] Loss 0.313961 Top1 83.932267 Top5 97.478943 -2023-02-13 18:30:44,056 - ==> Top1: 83.932 Top5: 97.479 Loss: 0.314 - -2023-02-13 18:30:44,057 - ==> Confusion: -[[ 868 3 5 0 8 4 0 1 4 45 0 4 0 7 6 4 1 2 2 0 3] - [ 3 941 1 2 12 30 2 17 2 1 2 2 2 0 0 2 4 1 3 1 5] - [ 8 3 951 10 5 2 17 15 3 1 2 4 3 5 4 7 3 2 8 1 4] - [ 5 1 14 892 1 6 0 2 2 2 16 1 10 0 23 4 3 4 23 0 7] - [ 10 6 0 0 999 11 1 2 1 2 0 6 0 4 8 6 2 1 0 4 3] - [ 2 16 1 3 4 973 3 14 2 5 0 10 4 16 1 3 6 1 1 5 0] - [ 2 2 9 2 0 8 1044 7 0 2 0 1 2 2 0 5 1 1 1 7 3] - [ 4 7 8 2 1 24 2 937 0 1 0 5 4 2 0 0 2 1 12 9 3] - [ 13 1 1 1 1 2 0 3 915 34 5 4 0 8 14 2 1 0 4 0 0] - [ 84 1 2 0 9 0 0 2 36 848 0 0 1 15 6 2 0 2 0 1 3] - [ 1 2 2 6 0 1 4 4 18 2 982 1 2 12 3 0 1 2 6 0 2] - [ 2 2 1 0 2 6 0 3 4 3 0 938 10 9 1 5 1 12 1 5 0] - [ 1 0 0 5 1 5 0 2 2 0 0 31 873 1 3 8 4 17 1 1 4] - [ 3 2 3 0 5 7 0 0 15 13 6 7 2 945 3 5 3 1 1 0 3] - [ 4 2 0 12 3 4 0 3 18 9 3 1 2 3 1009 0 0 6 7 1 5] - [ 3 2 3 0 7 3 4 1 1 2 0 4 7 2 0 981 7 7 0 5 7] - [ 2 6 0 2 9 2 0 1 5 0 0 1 3 3 0 12 1001 1 1 1 11] - [ 4 1 0 4 0 1 2 0 2 1 1 12 9 2 0 17 0 985 1 3 6] - [ 5 4 2 4 0 2 0 23 5 2 5 1 1 0 12 1 2 3 1012 2 0] - [ 1 3 0 0 0 5 7 12 0 1 0 20 4 6 1 7 5 3 0 1067 6] - [ 170 237 208 111 159 245 89 230 123 88 190 153 332 338 179 127 262 90 197 269 9637]] - -2023-02-13 18:30:44,058 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:30:44,059 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:30:44,064 - - -2023-02-13 18:30:44,065 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:30:45,066 - Epoch: [165][ 10/ 1207] Overall Loss 0.232156 Objective Loss 0.232156 LR 0.000250 Time 0.100046 -2023-02-13 18:30:45,260 - Epoch: [165][ 20/ 1207] Overall Loss 0.233732 Objective Loss 0.233732 LR 0.000250 Time 0.059722 -2023-02-13 18:30:45,450 - Epoch: [165][ 30/ 1207] Overall Loss 0.227901 Objective Loss 0.227901 LR 0.000250 Time 0.046137 -2023-02-13 18:30:45,639 - Epoch: [165][ 40/ 1207] Overall Loss 0.227931 Objective Loss 0.227931 LR 0.000250 Time 0.039317 -2023-02-13 18:30:45,827 - Epoch: [165][ 50/ 1207] Overall Loss 0.231639 Objective Loss 0.231639 LR 0.000250 Time 0.035212 -2023-02-13 18:30:46,018 - Epoch: [165][ 60/ 1207] Overall Loss 0.235939 Objective Loss 0.235939 LR 0.000250 Time 0.032507 -2023-02-13 18:30:46,207 - Epoch: [165][ 70/ 1207] Overall Loss 0.236977 Objective Loss 0.236977 LR 0.000250 Time 0.030564 -2023-02-13 18:30:46,397 - Epoch: [165][ 80/ 1207] Overall Loss 0.236720 Objective Loss 0.236720 LR 0.000250 Time 0.029112 -2023-02-13 18:30:46,586 - Epoch: [165][ 90/ 1207] Overall Loss 0.239492 Objective Loss 0.239492 LR 0.000250 Time 0.027974 -2023-02-13 18:30:46,775 - Epoch: [165][ 100/ 1207] Overall Loss 0.242010 Objective Loss 0.242010 LR 0.000250 Time 0.027068 -2023-02-13 18:30:46,965 - Epoch: [165][ 110/ 1207] Overall Loss 0.239719 Objective Loss 0.239719 LR 0.000250 Time 0.026329 -2023-02-13 18:30:47,154 - Epoch: [165][ 120/ 1207] Overall Loss 0.241961 Objective Loss 0.241961 LR 0.000250 Time 0.025708 -2023-02-13 18:30:47,344 - Epoch: [165][ 130/ 1207] Overall Loss 0.241854 Objective Loss 0.241854 LR 0.000250 Time 0.025186 -2023-02-13 18:30:47,534 - Epoch: [165][ 140/ 1207] Overall Loss 0.241840 Objective Loss 0.241840 LR 0.000250 Time 0.024740 -2023-02-13 18:30:47,723 - Epoch: [165][ 150/ 1207] Overall Loss 0.241245 Objective Loss 0.241245 LR 0.000250 Time 0.024353 -2023-02-13 18:30:47,913 - Epoch: [165][ 160/ 1207] Overall Loss 0.241309 Objective Loss 0.241309 LR 0.000250 Time 0.024016 -2023-02-13 18:30:48,103 - Epoch: [165][ 170/ 1207] Overall Loss 0.241116 Objective Loss 0.241116 LR 0.000250 Time 0.023713 -2023-02-13 18:30:48,292 - Epoch: [165][ 180/ 1207] Overall Loss 0.241878 Objective Loss 0.241878 LR 0.000250 Time 0.023445 -2023-02-13 18:30:48,481 - Epoch: [165][ 190/ 1207] Overall Loss 0.241508 Objective Loss 0.241508 LR 0.000250 Time 0.023207 -2023-02-13 18:30:48,671 - Epoch: [165][ 200/ 1207] Overall Loss 0.241908 Objective Loss 0.241908 LR 0.000250 Time 0.022993 -2023-02-13 18:30:48,860 - Epoch: [165][ 210/ 1207] Overall Loss 0.242247 Objective Loss 0.242247 LR 0.000250 Time 0.022798 -2023-02-13 18:30:49,049 - Epoch: [165][ 220/ 1207] Overall Loss 0.241256 Objective Loss 0.241256 LR 0.000250 Time 0.022620 -2023-02-13 18:30:49,239 - Epoch: [165][ 230/ 1207] Overall Loss 0.240615 Objective Loss 0.240615 LR 0.000250 Time 0.022457 -2023-02-13 18:30:49,429 - Epoch: [165][ 240/ 1207] Overall Loss 0.240546 Objective Loss 0.240546 LR 0.000250 Time 0.022312 -2023-02-13 18:30:49,618 - Epoch: [165][ 250/ 1207] Overall Loss 0.240482 Objective Loss 0.240482 LR 0.000250 Time 0.022176 -2023-02-13 18:30:49,807 - Epoch: [165][ 260/ 1207] Overall Loss 0.240632 Objective Loss 0.240632 LR 0.000250 Time 0.022048 -2023-02-13 18:30:49,996 - Epoch: [165][ 270/ 1207] Overall Loss 0.239789 Objective Loss 0.239789 LR 0.000250 Time 0.021931 -2023-02-13 18:30:50,185 - Epoch: [165][ 280/ 1207] Overall Loss 0.239762 Objective Loss 0.239762 LR 0.000250 Time 0.021821 -2023-02-13 18:30:50,374 - Epoch: [165][ 290/ 1207] Overall Loss 0.239469 Objective Loss 0.239469 LR 0.000250 Time 0.021718 -2023-02-13 18:30:50,563 - Epoch: [165][ 300/ 1207] Overall Loss 0.239553 Objective Loss 0.239553 LR 0.000250 Time 0.021625 -2023-02-13 18:30:50,752 - Epoch: [165][ 310/ 1207] Overall Loss 0.239927 Objective Loss 0.239927 LR 0.000250 Time 0.021535 -2023-02-13 18:30:50,943 - Epoch: [165][ 320/ 1207] Overall Loss 0.239916 Objective Loss 0.239916 LR 0.000250 Time 0.021458 -2023-02-13 18:30:51,132 - Epoch: [165][ 330/ 1207] Overall Loss 0.240492 Objective Loss 0.240492 LR 0.000250 Time 0.021379 -2023-02-13 18:30:51,321 - Epoch: [165][ 340/ 1207] Overall Loss 0.240378 Objective Loss 0.240378 LR 0.000250 Time 0.021305 -2023-02-13 18:30:51,510 - Epoch: [165][ 350/ 1207] Overall Loss 0.240905 Objective Loss 0.240905 LR 0.000250 Time 0.021236 -2023-02-13 18:30:51,699 - Epoch: [165][ 360/ 1207] Overall Loss 0.240917 Objective Loss 0.240917 LR 0.000250 Time 0.021170 -2023-02-13 18:30:51,888 - Epoch: [165][ 370/ 1207] Overall Loss 0.240952 Objective Loss 0.240952 LR 0.000250 Time 0.021105 -2023-02-13 18:30:52,077 - Epoch: [165][ 380/ 1207] Overall Loss 0.240923 Objective Loss 0.240923 LR 0.000250 Time 0.021047 -2023-02-13 18:30:52,265 - Epoch: [165][ 390/ 1207] Overall Loss 0.240279 Objective Loss 0.240279 LR 0.000250 Time 0.020989 -2023-02-13 18:30:52,454 - Epoch: [165][ 400/ 1207] Overall Loss 0.240566 Objective Loss 0.240566 LR 0.000250 Time 0.020936 -2023-02-13 18:30:52,643 - Epoch: [165][ 410/ 1207] Overall Loss 0.241091 Objective Loss 0.241091 LR 0.000250 Time 0.020884 -2023-02-13 18:30:52,832 - Epoch: [165][ 420/ 1207] Overall Loss 0.240732 Objective Loss 0.240732 LR 0.000250 Time 0.020836 -2023-02-13 18:30:53,020 - Epoch: [165][ 430/ 1207] Overall Loss 0.241005 Objective Loss 0.241005 LR 0.000250 Time 0.020788 -2023-02-13 18:30:53,209 - Epoch: [165][ 440/ 1207] Overall Loss 0.241656 Objective Loss 0.241656 LR 0.000250 Time 0.020745 -2023-02-13 18:30:53,398 - Epoch: [165][ 450/ 1207] Overall Loss 0.241427 Objective Loss 0.241427 LR 0.000250 Time 0.020702 -2023-02-13 18:30:53,587 - Epoch: [165][ 460/ 1207] Overall Loss 0.241563 Objective Loss 0.241563 LR 0.000250 Time 0.020663 -2023-02-13 18:30:53,776 - Epoch: [165][ 470/ 1207] Overall Loss 0.241482 Objective Loss 0.241482 LR 0.000250 Time 0.020624 -2023-02-13 18:30:53,967 - Epoch: [165][ 480/ 1207] Overall Loss 0.241651 Objective Loss 0.241651 LR 0.000250 Time 0.020591 -2023-02-13 18:30:54,155 - Epoch: [165][ 490/ 1207] Overall Loss 0.242050 Objective Loss 0.242050 LR 0.000250 Time 0.020555 -2023-02-13 18:30:54,344 - Epoch: [165][ 500/ 1207] Overall Loss 0.241955 Objective Loss 0.241955 LR 0.000250 Time 0.020522 -2023-02-13 18:30:54,534 - Epoch: [165][ 510/ 1207] Overall Loss 0.241648 Objective Loss 0.241648 LR 0.000250 Time 0.020490 -2023-02-13 18:30:54,723 - Epoch: [165][ 520/ 1207] Overall Loss 0.241658 Objective Loss 0.241658 LR 0.000250 Time 0.020460 -2023-02-13 18:30:54,912 - Epoch: [165][ 530/ 1207] Overall Loss 0.241893 Objective Loss 0.241893 LR 0.000250 Time 0.020429 -2023-02-13 18:30:55,101 - Epoch: [165][ 540/ 1207] Overall Loss 0.241733 Objective Loss 0.241733 LR 0.000250 Time 0.020401 -2023-02-13 18:30:55,290 - Epoch: [165][ 550/ 1207] Overall Loss 0.241642 Objective Loss 0.241642 LR 0.000250 Time 0.020372 -2023-02-13 18:30:55,479 - Epoch: [165][ 560/ 1207] Overall Loss 0.241856 Objective Loss 0.241856 LR 0.000250 Time 0.020346 -2023-02-13 18:30:55,668 - Epoch: [165][ 570/ 1207] Overall Loss 0.242130 Objective Loss 0.242130 LR 0.000250 Time 0.020319 -2023-02-13 18:30:55,857 - Epoch: [165][ 580/ 1207] Overall Loss 0.242596 Objective Loss 0.242596 LR 0.000250 Time 0.020293 -2023-02-13 18:30:56,046 - Epoch: [165][ 590/ 1207] Overall Loss 0.242738 Objective Loss 0.242738 LR 0.000250 Time 0.020270 -2023-02-13 18:30:56,235 - Epoch: [165][ 600/ 1207] Overall Loss 0.243099 Objective Loss 0.243099 LR 0.000250 Time 0.020247 -2023-02-13 18:30:56,424 - Epoch: [165][ 610/ 1207] Overall Loss 0.243024 Objective Loss 0.243024 LR 0.000250 Time 0.020224 -2023-02-13 18:30:56,614 - Epoch: [165][ 620/ 1207] Overall Loss 0.243244 Objective Loss 0.243244 LR 0.000250 Time 0.020204 -2023-02-13 18:30:56,804 - Epoch: [165][ 630/ 1207] Overall Loss 0.243217 Objective Loss 0.243217 LR 0.000250 Time 0.020183 -2023-02-13 18:30:56,995 - Epoch: [165][ 640/ 1207] Overall Loss 0.243360 Objective Loss 0.243360 LR 0.000250 Time 0.020166 -2023-02-13 18:30:57,185 - Epoch: [165][ 650/ 1207] Overall Loss 0.243458 Objective Loss 0.243458 LR 0.000250 Time 0.020147 -2023-02-13 18:30:57,374 - Epoch: [165][ 660/ 1207] Overall Loss 0.243289 Objective Loss 0.243289 LR 0.000250 Time 0.020127 -2023-02-13 18:30:57,563 - Epoch: [165][ 670/ 1207] Overall Loss 0.243402 Objective Loss 0.243402 LR 0.000250 Time 0.020110 -2023-02-13 18:30:57,753 - Epoch: [165][ 680/ 1207] Overall Loss 0.243531 Objective Loss 0.243531 LR 0.000250 Time 0.020092 -2023-02-13 18:30:57,942 - Epoch: [165][ 690/ 1207] Overall Loss 0.243489 Objective Loss 0.243489 LR 0.000250 Time 0.020074 -2023-02-13 18:30:58,131 - Epoch: [165][ 700/ 1207] Overall Loss 0.243612 Objective Loss 0.243612 LR 0.000250 Time 0.020057 -2023-02-13 18:30:58,320 - Epoch: [165][ 710/ 1207] Overall Loss 0.243734 Objective Loss 0.243734 LR 0.000250 Time 0.020040 -2023-02-13 18:30:58,510 - Epoch: [165][ 720/ 1207] Overall Loss 0.243650 Objective Loss 0.243650 LR 0.000250 Time 0.020025 -2023-02-13 18:30:58,699 - Epoch: [165][ 730/ 1207] Overall Loss 0.243320 Objective Loss 0.243320 LR 0.000250 Time 0.020009 -2023-02-13 18:30:58,888 - Epoch: [165][ 740/ 1207] Overall Loss 0.243507 Objective Loss 0.243507 LR 0.000250 Time 0.019994 -2023-02-13 18:30:59,078 - Epoch: [165][ 750/ 1207] Overall Loss 0.243794 Objective Loss 0.243794 LR 0.000250 Time 0.019980 -2023-02-13 18:30:59,266 - Epoch: [165][ 760/ 1207] Overall Loss 0.243740 Objective Loss 0.243740 LR 0.000250 Time 0.019965 -2023-02-13 18:30:59,456 - Epoch: [165][ 770/ 1207] Overall Loss 0.243637 Objective Loss 0.243637 LR 0.000250 Time 0.019951 -2023-02-13 18:30:59,645 - Epoch: [165][ 780/ 1207] Overall Loss 0.243713 Objective Loss 0.243713 LR 0.000250 Time 0.019938 -2023-02-13 18:30:59,835 - Epoch: [165][ 790/ 1207] Overall Loss 0.243697 Objective Loss 0.243697 LR 0.000250 Time 0.019925 -2023-02-13 18:31:00,024 - Epoch: [165][ 800/ 1207] Overall Loss 0.243690 Objective Loss 0.243690 LR 0.000250 Time 0.019912 -2023-02-13 18:31:00,214 - Epoch: [165][ 810/ 1207] Overall Loss 0.243405 Objective Loss 0.243405 LR 0.000250 Time 0.019900 -2023-02-13 18:31:00,402 - Epoch: [165][ 820/ 1207] Overall Loss 0.243289 Objective Loss 0.243289 LR 0.000250 Time 0.019886 -2023-02-13 18:31:00,592 - Epoch: [165][ 830/ 1207] Overall Loss 0.243201 Objective Loss 0.243201 LR 0.000250 Time 0.019874 -2023-02-13 18:31:00,781 - Epoch: [165][ 840/ 1207] Overall Loss 0.243344 Objective Loss 0.243344 LR 0.000250 Time 0.019863 -2023-02-13 18:31:00,971 - Epoch: [165][ 850/ 1207] Overall Loss 0.243335 Objective Loss 0.243335 LR 0.000250 Time 0.019853 -2023-02-13 18:31:01,161 - Epoch: [165][ 860/ 1207] Overall Loss 0.243088 Objective Loss 0.243088 LR 0.000250 Time 0.019842 -2023-02-13 18:31:01,350 - Epoch: [165][ 870/ 1207] Overall Loss 0.243098 Objective Loss 0.243098 LR 0.000250 Time 0.019830 -2023-02-13 18:31:01,539 - Epoch: [165][ 880/ 1207] Overall Loss 0.243118 Objective Loss 0.243118 LR 0.000250 Time 0.019820 -2023-02-13 18:31:01,729 - Epoch: [165][ 890/ 1207] Overall Loss 0.243143 Objective Loss 0.243143 LR 0.000250 Time 0.019810 -2023-02-13 18:31:01,918 - Epoch: [165][ 900/ 1207] Overall Loss 0.243280 Objective Loss 0.243280 LR 0.000250 Time 0.019799 -2023-02-13 18:31:02,108 - Epoch: [165][ 910/ 1207] Overall Loss 0.243245 Objective Loss 0.243245 LR 0.000250 Time 0.019790 -2023-02-13 18:31:02,297 - Epoch: [165][ 920/ 1207] Overall Loss 0.243202 Objective Loss 0.243202 LR 0.000250 Time 0.019781 -2023-02-13 18:31:02,488 - Epoch: [165][ 930/ 1207] Overall Loss 0.243239 Objective Loss 0.243239 LR 0.000250 Time 0.019772 -2023-02-13 18:31:02,677 - Epoch: [165][ 940/ 1207] Overall Loss 0.243194 Objective Loss 0.243194 LR 0.000250 Time 0.019763 -2023-02-13 18:31:02,866 - Epoch: [165][ 950/ 1207] Overall Loss 0.243263 Objective Loss 0.243263 LR 0.000250 Time 0.019753 -2023-02-13 18:31:03,056 - Epoch: [165][ 960/ 1207] Overall Loss 0.243313 Objective Loss 0.243313 LR 0.000250 Time 0.019745 -2023-02-13 18:31:03,245 - Epoch: [165][ 970/ 1207] Overall Loss 0.243368 Objective Loss 0.243368 LR 0.000250 Time 0.019736 -2023-02-13 18:31:03,434 - Epoch: [165][ 980/ 1207] Overall Loss 0.243319 Objective Loss 0.243319 LR 0.000250 Time 0.019727 -2023-02-13 18:31:03,624 - Epoch: [165][ 990/ 1207] Overall Loss 0.243300 Objective Loss 0.243300 LR 0.000250 Time 0.019719 -2023-02-13 18:31:03,813 - Epoch: [165][ 1000/ 1207] Overall Loss 0.243225 Objective Loss 0.243225 LR 0.000250 Time 0.019711 -2023-02-13 18:31:04,002 - Epoch: [165][ 1010/ 1207] Overall Loss 0.243233 Objective Loss 0.243233 LR 0.000250 Time 0.019702 -2023-02-13 18:31:04,191 - Epoch: [165][ 1020/ 1207] Overall Loss 0.243364 Objective Loss 0.243364 LR 0.000250 Time 0.019694 -2023-02-13 18:31:04,380 - Epoch: [165][ 1030/ 1207] Overall Loss 0.243329 Objective Loss 0.243329 LR 0.000250 Time 0.019686 -2023-02-13 18:31:04,582 - Epoch: [165][ 1040/ 1207] Overall Loss 0.243577 Objective Loss 0.243577 LR 0.000250 Time 0.019690 -2023-02-13 18:31:04,781 - Epoch: [165][ 1050/ 1207] Overall Loss 0.243743 Objective Loss 0.243743 LR 0.000250 Time 0.019692 -2023-02-13 18:31:04,982 - Epoch: [165][ 1060/ 1207] Overall Loss 0.243958 Objective Loss 0.243958 LR 0.000250 Time 0.019696 -2023-02-13 18:31:05,180 - Epoch: [165][ 1070/ 1207] Overall Loss 0.243901 Objective Loss 0.243901 LR 0.000250 Time 0.019696 -2023-02-13 18:31:05,383 - Epoch: [165][ 1080/ 1207] Overall Loss 0.243893 Objective Loss 0.243893 LR 0.000250 Time 0.019701 -2023-02-13 18:31:05,581 - Epoch: [165][ 1090/ 1207] Overall Loss 0.243980 Objective Loss 0.243980 LR 0.000250 Time 0.019702 -2023-02-13 18:31:05,783 - Epoch: [165][ 1100/ 1207] Overall Loss 0.244063 Objective Loss 0.244063 LR 0.000250 Time 0.019706 -2023-02-13 18:31:05,981 - Epoch: [165][ 1110/ 1207] Overall Loss 0.244217 Objective Loss 0.244217 LR 0.000250 Time 0.019707 -2023-02-13 18:31:06,184 - Epoch: [165][ 1120/ 1207] Overall Loss 0.244242 Objective Loss 0.244242 LR 0.000250 Time 0.019712 -2023-02-13 18:31:06,382 - Epoch: [165][ 1130/ 1207] Overall Loss 0.244250 Objective Loss 0.244250 LR 0.000250 Time 0.019712 -2023-02-13 18:31:06,585 - Epoch: [165][ 1140/ 1207] Overall Loss 0.244180 Objective Loss 0.244180 LR 0.000250 Time 0.019717 -2023-02-13 18:31:06,784 - Epoch: [165][ 1150/ 1207] Overall Loss 0.244245 Objective Loss 0.244245 LR 0.000250 Time 0.019718 -2023-02-13 18:31:06,987 - Epoch: [165][ 1160/ 1207] Overall Loss 0.244201 Objective Loss 0.244201 LR 0.000250 Time 0.019723 -2023-02-13 18:31:07,185 - Epoch: [165][ 1170/ 1207] Overall Loss 0.244346 Objective Loss 0.244346 LR 0.000250 Time 0.019723 -2023-02-13 18:31:07,388 - Epoch: [165][ 1180/ 1207] Overall Loss 0.244418 Objective Loss 0.244418 LR 0.000250 Time 0.019728 -2023-02-13 18:31:07,587 - Epoch: [165][ 1190/ 1207] Overall Loss 0.244348 Objective Loss 0.244348 LR 0.000250 Time 0.019729 -2023-02-13 18:31:07,842 - Epoch: [165][ 1200/ 1207] Overall Loss 0.244524 Objective Loss 0.244524 LR 0.000250 Time 0.019777 -2023-02-13 18:31:07,957 - Epoch: [165][ 1207/ 1207] Overall Loss 0.244618 Objective Loss 0.244618 Top1 86.280488 Top5 98.170732 LR 0.000250 Time 0.019757 -2023-02-13 18:31:08,030 - --- validate (epoch=165)----------- -2023-02-13 18:31:08,030 - 34311 samples (256 per mini-batch) -2023-02-13 18:31:08,446 - Epoch: [165][ 10/ 135] Loss 0.313452 Top1 84.414062 Top5 97.265625 -2023-02-13 18:31:08,588 - Epoch: [165][ 20/ 135] Loss 0.314978 Top1 84.531250 Top5 97.441406 -2023-02-13 18:31:08,719 - Epoch: [165][ 30/ 135] Loss 0.305922 Top1 84.778646 Top5 97.734375 -2023-02-13 18:31:08,843 - Epoch: [165][ 40/ 135] Loss 0.310879 Top1 84.472656 Top5 97.695312 -2023-02-13 18:31:08,970 - Epoch: [165][ 50/ 135] Loss 0.309487 Top1 84.500000 Top5 97.726562 -2023-02-13 18:31:09,097 - Epoch: [165][ 60/ 135] Loss 0.311186 Top1 84.570312 Top5 97.701823 -2023-02-13 18:31:09,218 - Epoch: [165][ 70/ 135] Loss 0.315324 Top1 84.375000 Top5 97.672991 -2023-02-13 18:31:09,339 - Epoch: [165][ 80/ 135] Loss 0.314334 Top1 84.365234 Top5 97.666016 -2023-02-13 18:31:09,470 - Epoch: [165][ 90/ 135] Loss 0.313231 Top1 84.552951 Top5 97.699653 -2023-02-13 18:31:09,608 - Epoch: [165][ 100/ 135] Loss 0.312658 Top1 84.460938 Top5 97.695312 -2023-02-13 18:31:09,735 - Epoch: [165][ 110/ 135] Loss 0.312866 Top1 84.371449 Top5 97.663352 -2023-02-13 18:31:09,863 - Epoch: [165][ 120/ 135] Loss 0.313568 Top1 84.352214 Top5 97.692057 -2023-02-13 18:31:09,990 - Epoch: [165][ 130/ 135] Loss 0.312849 Top1 84.362981 Top5 97.689303 -2023-02-13 18:31:10,034 - Epoch: [165][ 135/ 135] Loss 0.311069 Top1 84.340299 Top5 97.691702 -2023-02-13 18:31:10,102 - ==> Top1: 84.340 Top5: 97.692 Loss: 0.311 - -2023-02-13 18:31:10,103 - ==> Confusion: -[[ 867 5 8 1 8 2 0 1 4 44 0 2 2 3 2 2 2 3 1 2 8] - [ 2 945 1 2 7 25 4 20 5 1 2 0 1 0 0 1 3 2 2 5 5] - [ 7 2 972 10 1 0 11 15 1 1 3 1 2 6 2 4 3 3 7 4 3] - [ 4 0 22 917 1 2 1 2 3 1 9 0 9 1 13 2 3 4 17 0 5] - [ 12 8 0 2 990 12 2 2 2 1 0 4 0 2 8 5 3 5 1 3 4] - [ 4 14 1 5 3 960 4 22 3 4 2 7 3 17 2 3 4 2 1 7 2] - [ 3 3 25 2 0 6 1031 5 0 1 0 0 3 1 0 2 3 4 1 3 6] - [ 3 5 9 1 0 22 2 954 1 1 1 4 3 1 0 1 2 0 4 8 2] - [ 17 5 1 1 1 0 0 1 912 34 6 2 0 8 10 2 1 0 6 1 1] - [ 85 2 4 0 7 1 0 1 29 859 0 0 0 8 6 2 1 1 1 1 4] - [ 1 2 5 8 0 2 2 7 13 2 988 2 1 7 2 0 1 1 4 0 3] - [ 3 2 2 0 1 8 0 7 2 3 0 916 21 9 2 5 2 14 1 5 2] - [ 1 0 0 7 1 5 0 1 2 1 0 24 881 1 1 7 4 17 1 1 4] - [ 5 3 7 0 5 7 1 2 14 17 7 3 1 935 3 5 3 2 0 1 3] - [ 10 4 1 23 3 3 0 2 21 6 1 1 2 2 989 0 1 9 7 0 7] - [ 4 1 7 0 3 0 2 1 0 1 0 4 7 3 1 971 8 17 1 6 9] - [ 3 6 0 1 6 2 0 0 2 2 0 1 4 1 1 11 1008 2 0 4 7] - [ 6 2 0 4 0 0 1 0 0 1 1 8 9 1 1 14 0 997 0 1 5] - [ 5 4 4 8 0 1 0 30 2 2 4 1 3 0 10 0 0 4 1003 3 2] - [ 2 3 2 0 0 2 4 8 1 0 0 15 2 2 1 5 4 6 0 1086 5] - [ 164 238 275 137 118 211 96 213 109 91 187 104 320 311 143 97 264 134 189 276 9757]] - -2023-02-13 18:31:10,105 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:31:10,105 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:31:10,111 - - -2023-02-13 18:31:10,111 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:31:10,999 - Epoch: [166][ 10/ 1207] Overall Loss 0.247399 Objective Loss 0.247399 LR 0.000250 Time 0.088787 -2023-02-13 18:31:11,189 - Epoch: [166][ 20/ 1207] Overall Loss 0.236224 Objective Loss 0.236224 LR 0.000250 Time 0.053842 -2023-02-13 18:31:11,377 - Epoch: [166][ 30/ 1207] Overall Loss 0.236454 Objective Loss 0.236454 LR 0.000250 Time 0.042157 -2023-02-13 18:31:11,566 - Epoch: [166][ 40/ 1207] Overall Loss 0.235195 Objective Loss 0.235195 LR 0.000250 Time 0.036336 -2023-02-13 18:31:11,754 - Epoch: [166][ 50/ 1207] Overall Loss 0.235400 Objective Loss 0.235400 LR 0.000250 Time 0.032823 -2023-02-13 18:31:11,942 - Epoch: [166][ 60/ 1207] Overall Loss 0.237046 Objective Loss 0.237046 LR 0.000250 Time 0.030481 -2023-02-13 18:31:12,130 - Epoch: [166][ 70/ 1207] Overall Loss 0.237820 Objective Loss 0.237820 LR 0.000250 Time 0.028806 -2023-02-13 18:31:12,318 - Epoch: [166][ 80/ 1207] Overall Loss 0.241114 Objective Loss 0.241114 LR 0.000250 Time 0.027546 -2023-02-13 18:31:12,506 - Epoch: [166][ 90/ 1207] Overall Loss 0.240673 Objective Loss 0.240673 LR 0.000250 Time 0.026572 -2023-02-13 18:31:12,694 - Epoch: [166][ 100/ 1207] Overall Loss 0.241740 Objective Loss 0.241740 LR 0.000250 Time 0.025791 -2023-02-13 18:31:12,884 - Epoch: [166][ 110/ 1207] Overall Loss 0.239982 Objective Loss 0.239982 LR 0.000250 Time 0.025167 -2023-02-13 18:31:13,073 - Epoch: [166][ 120/ 1207] Overall Loss 0.239800 Objective Loss 0.239800 LR 0.000250 Time 0.024647 -2023-02-13 18:31:13,262 - Epoch: [166][ 130/ 1207] Overall Loss 0.239900 Objective Loss 0.239900 LR 0.000250 Time 0.024204 -2023-02-13 18:31:13,453 - Epoch: [166][ 140/ 1207] Overall Loss 0.240193 Objective Loss 0.240193 LR 0.000250 Time 0.023835 -2023-02-13 18:31:13,643 - Epoch: [166][ 150/ 1207] Overall Loss 0.240448 Objective Loss 0.240448 LR 0.000250 Time 0.023506 -2023-02-13 18:31:13,832 - Epoch: [166][ 160/ 1207] Overall Loss 0.240373 Objective Loss 0.240373 LR 0.000250 Time 0.023220 -2023-02-13 18:31:14,021 - Epoch: [166][ 170/ 1207] Overall Loss 0.240533 Objective Loss 0.240533 LR 0.000250 Time 0.022963 -2023-02-13 18:31:14,211 - Epoch: [166][ 180/ 1207] Overall Loss 0.241764 Objective Loss 0.241764 LR 0.000250 Time 0.022742 -2023-02-13 18:31:14,401 - Epoch: [166][ 190/ 1207] Overall Loss 0.240768 Objective Loss 0.240768 LR 0.000250 Time 0.022541 -2023-02-13 18:31:14,591 - Epoch: [166][ 200/ 1207] Overall Loss 0.240866 Objective Loss 0.240866 LR 0.000250 Time 0.022365 -2023-02-13 18:31:14,780 - Epoch: [166][ 210/ 1207] Overall Loss 0.240870 Objective Loss 0.240870 LR 0.000250 Time 0.022198 -2023-02-13 18:31:14,971 - Epoch: [166][ 220/ 1207] Overall Loss 0.240721 Objective Loss 0.240721 LR 0.000250 Time 0.022053 -2023-02-13 18:31:15,161 - Epoch: [166][ 230/ 1207] Overall Loss 0.240962 Objective Loss 0.240962 LR 0.000250 Time 0.021917 -2023-02-13 18:31:15,351 - Epoch: [166][ 240/ 1207] Overall Loss 0.241821 Objective Loss 0.241821 LR 0.000250 Time 0.021796 -2023-02-13 18:31:15,542 - Epoch: [166][ 250/ 1207] Overall Loss 0.242529 Objective Loss 0.242529 LR 0.000250 Time 0.021685 -2023-02-13 18:31:15,732 - Epoch: [166][ 260/ 1207] Overall Loss 0.242907 Objective Loss 0.242907 LR 0.000250 Time 0.021582 -2023-02-13 18:31:15,922 - Epoch: [166][ 270/ 1207] Overall Loss 0.243139 Objective Loss 0.243139 LR 0.000250 Time 0.021486 -2023-02-13 18:31:16,112 - Epoch: [166][ 280/ 1207] Overall Loss 0.242657 Objective Loss 0.242657 LR 0.000250 Time 0.021394 -2023-02-13 18:31:16,301 - Epoch: [166][ 290/ 1207] Overall Loss 0.242890 Objective Loss 0.242890 LR 0.000250 Time 0.021308 -2023-02-13 18:31:16,491 - Epoch: [166][ 300/ 1207] Overall Loss 0.242669 Objective Loss 0.242669 LR 0.000250 Time 0.021231 -2023-02-13 18:31:16,682 - Epoch: [166][ 310/ 1207] Overall Loss 0.243002 Objective Loss 0.243002 LR 0.000250 Time 0.021159 -2023-02-13 18:31:16,872 - Epoch: [166][ 320/ 1207] Overall Loss 0.243288 Objective Loss 0.243288 LR 0.000250 Time 0.021091 -2023-02-13 18:31:17,061 - Epoch: [166][ 330/ 1207] Overall Loss 0.243012 Objective Loss 0.243012 LR 0.000250 Time 0.021024 -2023-02-13 18:31:17,251 - Epoch: [166][ 340/ 1207] Overall Loss 0.243237 Objective Loss 0.243237 LR 0.000250 Time 0.020964 -2023-02-13 18:31:17,441 - Epoch: [166][ 350/ 1207] Overall Loss 0.243156 Objective Loss 0.243156 LR 0.000250 Time 0.020906 -2023-02-13 18:31:17,632 - Epoch: [166][ 360/ 1207] Overall Loss 0.243572 Objective Loss 0.243572 LR 0.000250 Time 0.020853 -2023-02-13 18:31:17,821 - Epoch: [166][ 370/ 1207] Overall Loss 0.243591 Objective Loss 0.243591 LR 0.000250 Time 0.020801 -2023-02-13 18:31:18,012 - Epoch: [166][ 380/ 1207] Overall Loss 0.243262 Objective Loss 0.243262 LR 0.000250 Time 0.020754 -2023-02-13 18:31:18,201 - Epoch: [166][ 390/ 1207] Overall Loss 0.242892 Objective Loss 0.242892 LR 0.000250 Time 0.020707 -2023-02-13 18:31:18,392 - Epoch: [166][ 400/ 1207] Overall Loss 0.242536 Objective Loss 0.242536 LR 0.000250 Time 0.020664 -2023-02-13 18:31:18,581 - Epoch: [166][ 410/ 1207] Overall Loss 0.242794 Objective Loss 0.242794 LR 0.000250 Time 0.020622 -2023-02-13 18:31:18,772 - Epoch: [166][ 420/ 1207] Overall Loss 0.242755 Objective Loss 0.242755 LR 0.000250 Time 0.020584 -2023-02-13 18:31:18,961 - Epoch: [166][ 430/ 1207] Overall Loss 0.243480 Objective Loss 0.243480 LR 0.000250 Time 0.020545 -2023-02-13 18:31:19,152 - Epoch: [166][ 440/ 1207] Overall Loss 0.243364 Objective Loss 0.243364 LR 0.000250 Time 0.020510 -2023-02-13 18:31:19,341 - Epoch: [166][ 450/ 1207] Overall Loss 0.243393 Objective Loss 0.243393 LR 0.000250 Time 0.020474 -2023-02-13 18:31:19,535 - Epoch: [166][ 460/ 1207] Overall Loss 0.243024 Objective Loss 0.243024 LR 0.000250 Time 0.020449 -2023-02-13 18:31:19,731 - Epoch: [166][ 470/ 1207] Overall Loss 0.242748 Objective Loss 0.242748 LR 0.000250 Time 0.020432 -2023-02-13 18:31:19,926 - Epoch: [166][ 480/ 1207] Overall Loss 0.242652 Objective Loss 0.242652 LR 0.000250 Time 0.020410 -2023-02-13 18:31:20,122 - Epoch: [166][ 490/ 1207] Overall Loss 0.242343 Objective Loss 0.242343 LR 0.000250 Time 0.020395 -2023-02-13 18:31:20,317 - Epoch: [166][ 500/ 1207] Overall Loss 0.242467 Objective Loss 0.242467 LR 0.000250 Time 0.020375 -2023-02-13 18:31:20,514 - Epoch: [166][ 510/ 1207] Overall Loss 0.242398 Objective Loss 0.242398 LR 0.000250 Time 0.020361 -2023-02-13 18:31:20,709 - Epoch: [166][ 520/ 1207] Overall Loss 0.241933 Objective Loss 0.241933 LR 0.000250 Time 0.020344 -2023-02-13 18:31:20,906 - Epoch: [166][ 530/ 1207] Overall Loss 0.241593 Objective Loss 0.241593 LR 0.000250 Time 0.020332 -2023-02-13 18:31:21,101 - Epoch: [166][ 540/ 1207] Overall Loss 0.241445 Objective Loss 0.241445 LR 0.000250 Time 0.020315 -2023-02-13 18:31:21,297 - Epoch: [166][ 550/ 1207] Overall Loss 0.241744 Objective Loss 0.241744 LR 0.000250 Time 0.020302 -2023-02-13 18:31:21,493 - Epoch: [166][ 560/ 1207] Overall Loss 0.242159 Objective Loss 0.242159 LR 0.000250 Time 0.020287 -2023-02-13 18:31:21,690 - Epoch: [166][ 570/ 1207] Overall Loss 0.242481 Objective Loss 0.242481 LR 0.000250 Time 0.020277 -2023-02-13 18:31:21,884 - Epoch: [166][ 580/ 1207] Overall Loss 0.243046 Objective Loss 0.243046 LR 0.000250 Time 0.020262 -2023-02-13 18:31:22,081 - Epoch: [166][ 590/ 1207] Overall Loss 0.242959 Objective Loss 0.242959 LR 0.000250 Time 0.020252 -2023-02-13 18:31:22,276 - Epoch: [166][ 600/ 1207] Overall Loss 0.242677 Objective Loss 0.242677 LR 0.000250 Time 0.020238 -2023-02-13 18:31:22,473 - Epoch: [166][ 610/ 1207] Overall Loss 0.243098 Objective Loss 0.243098 LR 0.000250 Time 0.020229 -2023-02-13 18:31:22,668 - Epoch: [166][ 620/ 1207] Overall Loss 0.243029 Objective Loss 0.243029 LR 0.000250 Time 0.020216 -2023-02-13 18:31:22,865 - Epoch: [166][ 630/ 1207] Overall Loss 0.243433 Objective Loss 0.243433 LR 0.000250 Time 0.020208 -2023-02-13 18:31:23,059 - Epoch: [166][ 640/ 1207] Overall Loss 0.243172 Objective Loss 0.243172 LR 0.000250 Time 0.020194 -2023-02-13 18:31:23,256 - Epoch: [166][ 650/ 1207] Overall Loss 0.242846 Objective Loss 0.242846 LR 0.000250 Time 0.020185 -2023-02-13 18:31:23,450 - Epoch: [166][ 660/ 1207] Overall Loss 0.242528 Objective Loss 0.242528 LR 0.000250 Time 0.020173 -2023-02-13 18:31:23,647 - Epoch: [166][ 670/ 1207] Overall Loss 0.242651 Objective Loss 0.242651 LR 0.000250 Time 0.020167 -2023-02-13 18:31:23,842 - Epoch: [166][ 680/ 1207] Overall Loss 0.242610 Objective Loss 0.242610 LR 0.000250 Time 0.020156 -2023-02-13 18:31:24,037 - Epoch: [166][ 690/ 1207] Overall Loss 0.242551 Objective Loss 0.242551 LR 0.000250 Time 0.020146 -2023-02-13 18:31:24,227 - Epoch: [166][ 700/ 1207] Overall Loss 0.242600 Objective Loss 0.242600 LR 0.000250 Time 0.020129 -2023-02-13 18:31:24,417 - Epoch: [166][ 710/ 1207] Overall Loss 0.242307 Objective Loss 0.242307 LR 0.000250 Time 0.020113 -2023-02-13 18:31:24,609 - Epoch: [166][ 720/ 1207] Overall Loss 0.242090 Objective Loss 0.242090 LR 0.000250 Time 0.020099 -2023-02-13 18:31:24,799 - Epoch: [166][ 730/ 1207] Overall Loss 0.241947 Objective Loss 0.241947 LR 0.000250 Time 0.020084 -2023-02-13 18:31:24,990 - Epoch: [166][ 740/ 1207] Overall Loss 0.241993 Objective Loss 0.241993 LR 0.000250 Time 0.020070 -2023-02-13 18:31:25,180 - Epoch: [166][ 750/ 1207] Overall Loss 0.242029 Objective Loss 0.242029 LR 0.000250 Time 0.020055 -2023-02-13 18:31:25,371 - Epoch: [166][ 760/ 1207] Overall Loss 0.242418 Objective Loss 0.242418 LR 0.000250 Time 0.020042 -2023-02-13 18:31:25,562 - Epoch: [166][ 770/ 1207] Overall Loss 0.242414 Objective Loss 0.242414 LR 0.000250 Time 0.020029 -2023-02-13 18:31:25,753 - Epoch: [166][ 780/ 1207] Overall Loss 0.242427 Objective Loss 0.242427 LR 0.000250 Time 0.020017 -2023-02-13 18:31:25,944 - Epoch: [166][ 790/ 1207] Overall Loss 0.242425 Objective Loss 0.242425 LR 0.000250 Time 0.020005 -2023-02-13 18:31:26,135 - Epoch: [166][ 800/ 1207] Overall Loss 0.242526 Objective Loss 0.242526 LR 0.000250 Time 0.019993 -2023-02-13 18:31:26,325 - Epoch: [166][ 810/ 1207] Overall Loss 0.242608 Objective Loss 0.242608 LR 0.000250 Time 0.019981 -2023-02-13 18:31:26,516 - Epoch: [166][ 820/ 1207] Overall Loss 0.242700 Objective Loss 0.242700 LR 0.000250 Time 0.019970 -2023-02-13 18:31:26,707 - Epoch: [166][ 830/ 1207] Overall Loss 0.242839 Objective Loss 0.242839 LR 0.000250 Time 0.019959 -2023-02-13 18:31:26,897 - Epoch: [166][ 840/ 1207] Overall Loss 0.242880 Objective Loss 0.242880 LR 0.000250 Time 0.019947 -2023-02-13 18:31:27,088 - Epoch: [166][ 850/ 1207] Overall Loss 0.242972 Objective Loss 0.242972 LR 0.000250 Time 0.019936 -2023-02-13 18:31:27,279 - Epoch: [166][ 860/ 1207] Overall Loss 0.242752 Objective Loss 0.242752 LR 0.000250 Time 0.019926 -2023-02-13 18:31:27,469 - Epoch: [166][ 870/ 1207] Overall Loss 0.242830 Objective Loss 0.242830 LR 0.000250 Time 0.019915 -2023-02-13 18:31:27,660 - Epoch: [166][ 880/ 1207] Overall Loss 0.242761 Objective Loss 0.242761 LR 0.000250 Time 0.019905 -2023-02-13 18:31:27,849 - Epoch: [166][ 890/ 1207] Overall Loss 0.242733 Objective Loss 0.242733 LR 0.000250 Time 0.019894 -2023-02-13 18:31:28,040 - Epoch: [166][ 900/ 1207] Overall Loss 0.242723 Objective Loss 0.242723 LR 0.000250 Time 0.019884 -2023-02-13 18:31:28,229 - Epoch: [166][ 910/ 1207] Overall Loss 0.242884 Objective Loss 0.242884 LR 0.000250 Time 0.019873 -2023-02-13 18:31:28,420 - Epoch: [166][ 920/ 1207] Overall Loss 0.243038 Objective Loss 0.243038 LR 0.000250 Time 0.019864 -2023-02-13 18:31:28,610 - Epoch: [166][ 930/ 1207] Overall Loss 0.242902 Objective Loss 0.242902 LR 0.000250 Time 0.019854 -2023-02-13 18:31:28,800 - Epoch: [166][ 940/ 1207] Overall Loss 0.242902 Objective Loss 0.242902 LR 0.000250 Time 0.019845 -2023-02-13 18:31:28,989 - Epoch: [166][ 950/ 1207] Overall Loss 0.243006 Objective Loss 0.243006 LR 0.000250 Time 0.019835 -2023-02-13 18:31:29,180 - Epoch: [166][ 960/ 1207] Overall Loss 0.242983 Objective Loss 0.242983 LR 0.000250 Time 0.019827 -2023-02-13 18:31:29,370 - Epoch: [166][ 970/ 1207] Overall Loss 0.242971 Objective Loss 0.242971 LR 0.000250 Time 0.019818 -2023-02-13 18:31:29,561 - Epoch: [166][ 980/ 1207] Overall Loss 0.243042 Objective Loss 0.243042 LR 0.000250 Time 0.019811 -2023-02-13 18:31:29,752 - Epoch: [166][ 990/ 1207] Overall Loss 0.243036 Objective Loss 0.243036 LR 0.000250 Time 0.019803 -2023-02-13 18:31:29,943 - Epoch: [166][ 1000/ 1207] Overall Loss 0.243026 Objective Loss 0.243026 LR 0.000250 Time 0.019795 -2023-02-13 18:31:30,134 - Epoch: [166][ 1010/ 1207] Overall Loss 0.242847 Objective Loss 0.242847 LR 0.000250 Time 0.019788 -2023-02-13 18:31:30,324 - Epoch: [166][ 1020/ 1207] Overall Loss 0.243024 Objective Loss 0.243024 LR 0.000250 Time 0.019780 -2023-02-13 18:31:30,514 - Epoch: [166][ 1030/ 1207] Overall Loss 0.242771 Objective Loss 0.242771 LR 0.000250 Time 0.019772 -2023-02-13 18:31:30,705 - Epoch: [166][ 1040/ 1207] Overall Loss 0.242746 Objective Loss 0.242746 LR 0.000250 Time 0.019766 -2023-02-13 18:31:30,896 - Epoch: [166][ 1050/ 1207] Overall Loss 0.242823 Objective Loss 0.242823 LR 0.000250 Time 0.019759 -2023-02-13 18:31:31,087 - Epoch: [166][ 1060/ 1207] Overall Loss 0.242709 Objective Loss 0.242709 LR 0.000250 Time 0.019752 -2023-02-13 18:31:31,277 - Epoch: [166][ 1070/ 1207] Overall Loss 0.242595 Objective Loss 0.242595 LR 0.000250 Time 0.019745 -2023-02-13 18:31:31,468 - Epoch: [166][ 1080/ 1207] Overall Loss 0.242642 Objective Loss 0.242642 LR 0.000250 Time 0.019738 -2023-02-13 18:31:31,659 - Epoch: [166][ 1090/ 1207] Overall Loss 0.242510 Objective Loss 0.242510 LR 0.000250 Time 0.019732 -2023-02-13 18:31:31,850 - Epoch: [166][ 1100/ 1207] Overall Loss 0.242525 Objective Loss 0.242525 LR 0.000250 Time 0.019726 -2023-02-13 18:31:32,041 - Epoch: [166][ 1110/ 1207] Overall Loss 0.242413 Objective Loss 0.242413 LR 0.000250 Time 0.019720 -2023-02-13 18:31:32,231 - Epoch: [166][ 1120/ 1207] Overall Loss 0.242342 Objective Loss 0.242342 LR 0.000250 Time 0.019714 -2023-02-13 18:31:32,421 - Epoch: [166][ 1130/ 1207] Overall Loss 0.242309 Objective Loss 0.242309 LR 0.000250 Time 0.019706 -2023-02-13 18:31:32,613 - Epoch: [166][ 1140/ 1207] Overall Loss 0.242327 Objective Loss 0.242327 LR 0.000250 Time 0.019702 -2023-02-13 18:31:32,804 - Epoch: [166][ 1150/ 1207] Overall Loss 0.242360 Objective Loss 0.242360 LR 0.000250 Time 0.019696 -2023-02-13 18:31:32,995 - Epoch: [166][ 1160/ 1207] Overall Loss 0.242376 Objective Loss 0.242376 LR 0.000250 Time 0.019691 -2023-02-13 18:31:33,186 - Epoch: [166][ 1170/ 1207] Overall Loss 0.242357 Objective Loss 0.242357 LR 0.000250 Time 0.019685 -2023-02-13 18:31:33,376 - Epoch: [166][ 1180/ 1207] Overall Loss 0.242178 Objective Loss 0.242178 LR 0.000250 Time 0.019680 -2023-02-13 18:31:33,566 - Epoch: [166][ 1190/ 1207] Overall Loss 0.242063 Objective Loss 0.242063 LR 0.000250 Time 0.019673 -2023-02-13 18:31:33,813 - Epoch: [166][ 1200/ 1207] Overall Loss 0.242226 Objective Loss 0.242226 LR 0.000250 Time 0.019715 -2023-02-13 18:31:33,928 - Epoch: [166][ 1207/ 1207] Overall Loss 0.242161 Objective Loss 0.242161 Top1 86.890244 Top5 98.170732 LR 0.000250 Time 0.019696 -2023-02-13 18:31:34,000 - --- validate (epoch=166)----------- -2023-02-13 18:31:34,000 - 34311 samples (256 per mini-batch) -2023-02-13 18:31:34,403 - Epoch: [166][ 10/ 135] Loss 0.280581 Top1 84.648438 Top5 97.968750 -2023-02-13 18:31:34,529 - Epoch: [166][ 20/ 135] Loss 0.321886 Top1 84.082031 Top5 97.597656 -2023-02-13 18:31:34,653 - Epoch: [166][ 30/ 135] Loss 0.317372 Top1 84.075521 Top5 97.656250 -2023-02-13 18:31:34,778 - Epoch: [166][ 40/ 135] Loss 0.313653 Top1 83.935547 Top5 97.714844 -2023-02-13 18:31:34,901 - Epoch: [166][ 50/ 135] Loss 0.311113 Top1 84.250000 Top5 97.835938 -2023-02-13 18:31:35,029 - Epoch: [166][ 60/ 135] Loss 0.304355 Top1 84.348958 Top5 97.825521 -2023-02-13 18:31:35,158 - Epoch: [166][ 70/ 135] Loss 0.304699 Top1 84.358259 Top5 97.823661 -2023-02-13 18:31:35,283 - Epoch: [166][ 80/ 135] Loss 0.305188 Top1 84.340820 Top5 97.822266 -2023-02-13 18:31:35,411 - Epoch: [166][ 90/ 135] Loss 0.308417 Top1 84.179688 Top5 97.825521 -2023-02-13 18:31:35,535 - Epoch: [166][ 100/ 135] Loss 0.308779 Top1 84.125000 Top5 97.777344 -2023-02-13 18:31:35,655 - Epoch: [166][ 110/ 135] Loss 0.312134 Top1 84.044744 Top5 97.773438 -2023-02-13 18:31:35,782 - Epoch: [166][ 120/ 135] Loss 0.313000 Top1 84.020182 Top5 97.692057 -2023-02-13 18:31:35,914 - Epoch: [166][ 130/ 135] Loss 0.313160 Top1 83.957332 Top5 97.671274 -2023-02-13 18:31:35,961 - Epoch: [166][ 135/ 135] Loss 0.314153 Top1 83.981813 Top5 97.682959 -2023-02-13 18:31:36,034 - ==> Top1: 83.982 Top5: 97.683 Loss: 0.314 - -2023-02-13 18:31:36,035 - ==> Confusion: -[[ 868 4 6 1 10 3 0 2 4 43 0 2 0 3 5 1 2 2 3 1 7] - [ 2 942 1 1 13 31 2 18 2 2 1 0 1 0 0 2 8 0 2 1 4] - [ 8 5 967 9 3 2 15 13 1 2 4 3 0 4 3 4 4 2 4 2 3] - [ 4 0 18 912 1 4 0 2 2 2 11 1 5 0 17 5 7 1 18 0 6] - [ 13 7 1 1 997 11 1 2 1 1 0 4 3 1 5 7 3 2 0 2 4] - [ 4 14 0 6 5 978 3 16 0 2 1 11 5 11 0 2 4 3 1 2 2] - [ 3 3 14 3 1 9 1040 3 0 2 0 1 2 2 0 3 2 2 1 2 6] - [ 3 6 8 1 1 24 1 950 0 2 0 4 3 2 0 0 2 1 9 5 2] - [ 18 2 1 1 1 0 1 3 893 41 4 3 0 14 18 3 2 0 3 0 1] - [ 92 1 4 0 8 0 0 2 25 851 0 2 0 14 6 1 1 1 0 1 3] - [ 3 2 2 10 0 4 2 6 16 2 977 1 1 12 2 0 2 0 6 0 3] - [ 4 2 1 0 3 13 0 8 3 1 0 924 17 6 0 4 2 8 1 7 1] - [ 2 0 0 11 2 6 0 0 2 1 1 38 871 1 2 6 3 8 0 0 5] - [ 5 1 1 1 6 12 0 4 11 17 6 5 4 934 5 4 3 1 0 1 3] - [ 9 3 3 20 4 5 0 3 21 8 2 2 2 3 990 0 2 3 5 0 7] - [ 4 1 7 0 7 3 3 3 2 0 0 7 4 3 1 973 10 6 0 7 5] - [ 2 4 2 1 10 2 0 0 3 1 0 2 1 2 1 12 1003 3 0 3 9] - [ 5 1 0 4 0 1 2 0 0 1 2 14 16 1 0 20 1 975 0 1 7] - [ 4 6 5 7 2 3 0 34 4 1 5 1 2 0 10 1 0 2 993 5 1] - [ 2 2 2 1 0 6 6 13 1 0 0 15 3 0 1 7 7 2 0 1075 5] - [ 188 259 238 142 162 249 92 227 83 99 164 123 324 304 145 103 335 90 157 248 9702]] - -2023-02-13 18:31:36,036 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:31:36,036 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:31:36,042 - - -2023-02-13 18:31:36,042 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:31:36,927 - Epoch: [167][ 10/ 1207] Overall Loss 0.232191 Objective Loss 0.232191 LR 0.000250 Time 0.088461 -2023-02-13 18:31:37,127 - Epoch: [167][ 20/ 1207] Overall Loss 0.237046 Objective Loss 0.237046 LR 0.000250 Time 0.054185 -2023-02-13 18:31:37,317 - Epoch: [167][ 30/ 1207] Overall Loss 0.237545 Objective Loss 0.237545 LR 0.000250 Time 0.042444 -2023-02-13 18:31:37,506 - Epoch: [167][ 40/ 1207] Overall Loss 0.235698 Objective Loss 0.235698 LR 0.000250 Time 0.036560 -2023-02-13 18:31:37,696 - Epoch: [167][ 50/ 1207] Overall Loss 0.237523 Objective Loss 0.237523 LR 0.000250 Time 0.033036 -2023-02-13 18:31:37,886 - Epoch: [167][ 60/ 1207] Overall Loss 0.237458 Objective Loss 0.237458 LR 0.000250 Time 0.030696 -2023-02-13 18:31:38,076 - Epoch: [167][ 70/ 1207] Overall Loss 0.237795 Objective Loss 0.237795 LR 0.000250 Time 0.029012 -2023-02-13 18:31:38,266 - Epoch: [167][ 80/ 1207] Overall Loss 0.236656 Objective Loss 0.236656 LR 0.000250 Time 0.027756 -2023-02-13 18:31:38,455 - Epoch: [167][ 90/ 1207] Overall Loss 0.235362 Objective Loss 0.235362 LR 0.000250 Time 0.026772 -2023-02-13 18:31:38,645 - Epoch: [167][ 100/ 1207] Overall Loss 0.235984 Objective Loss 0.235984 LR 0.000250 Time 0.025992 -2023-02-13 18:31:38,835 - Epoch: [167][ 110/ 1207] Overall Loss 0.235613 Objective Loss 0.235613 LR 0.000250 Time 0.025350 -2023-02-13 18:31:39,025 - Epoch: [167][ 120/ 1207] Overall Loss 0.236090 Objective Loss 0.236090 LR 0.000250 Time 0.024817 -2023-02-13 18:31:39,215 - Epoch: [167][ 130/ 1207] Overall Loss 0.237547 Objective Loss 0.237547 LR 0.000250 Time 0.024364 -2023-02-13 18:31:39,404 - Epoch: [167][ 140/ 1207] Overall Loss 0.236497 Objective Loss 0.236497 LR 0.000250 Time 0.023974 -2023-02-13 18:31:39,594 - Epoch: [167][ 150/ 1207] Overall Loss 0.235885 Objective Loss 0.235885 LR 0.000250 Time 0.023642 -2023-02-13 18:31:39,785 - Epoch: [167][ 160/ 1207] Overall Loss 0.236839 Objective Loss 0.236839 LR 0.000250 Time 0.023350 -2023-02-13 18:31:39,975 - Epoch: [167][ 170/ 1207] Overall Loss 0.236135 Objective Loss 0.236135 LR 0.000250 Time 0.023096 -2023-02-13 18:31:40,165 - Epoch: [167][ 180/ 1207] Overall Loss 0.235565 Objective Loss 0.235565 LR 0.000250 Time 0.022864 -2023-02-13 18:31:40,354 - Epoch: [167][ 190/ 1207] Overall Loss 0.235964 Objective Loss 0.235964 LR 0.000250 Time 0.022655 -2023-02-13 18:31:40,544 - Epoch: [167][ 200/ 1207] Overall Loss 0.234871 Objective Loss 0.234871 LR 0.000250 Time 0.022472 -2023-02-13 18:31:40,734 - Epoch: [167][ 210/ 1207] Overall Loss 0.235709 Objective Loss 0.235709 LR 0.000250 Time 0.022304 -2023-02-13 18:31:40,925 - Epoch: [167][ 220/ 1207] Overall Loss 0.236082 Objective Loss 0.236082 LR 0.000250 Time 0.022154 -2023-02-13 18:31:41,115 - Epoch: [167][ 230/ 1207] Overall Loss 0.236833 Objective Loss 0.236833 LR 0.000250 Time 0.022016 -2023-02-13 18:31:41,305 - Epoch: [167][ 240/ 1207] Overall Loss 0.236926 Objective Loss 0.236926 LR 0.000250 Time 0.021889 -2023-02-13 18:31:41,494 - Epoch: [167][ 250/ 1207] Overall Loss 0.237724 Objective Loss 0.237724 LR 0.000250 Time 0.021770 -2023-02-13 18:31:41,685 - Epoch: [167][ 260/ 1207] Overall Loss 0.238701 Objective Loss 0.238701 LR 0.000250 Time 0.021666 -2023-02-13 18:31:41,875 - Epoch: [167][ 270/ 1207] Overall Loss 0.239111 Objective Loss 0.239111 LR 0.000250 Time 0.021563 -2023-02-13 18:31:42,065 - Epoch: [167][ 280/ 1207] Overall Loss 0.239974 Objective Loss 0.239974 LR 0.000250 Time 0.021471 -2023-02-13 18:31:42,254 - Epoch: [167][ 290/ 1207] Overall Loss 0.239837 Objective Loss 0.239837 LR 0.000250 Time 0.021383 -2023-02-13 18:31:42,444 - Epoch: [167][ 300/ 1207] Overall Loss 0.239573 Objective Loss 0.239573 LR 0.000250 Time 0.021302 -2023-02-13 18:31:42,635 - Epoch: [167][ 310/ 1207] Overall Loss 0.240029 Objective Loss 0.240029 LR 0.000250 Time 0.021227 -2023-02-13 18:31:42,825 - Epoch: [167][ 320/ 1207] Overall Loss 0.239603 Objective Loss 0.239603 LR 0.000250 Time 0.021158 -2023-02-13 18:31:43,013 - Epoch: [167][ 330/ 1207] Overall Loss 0.238969 Objective Loss 0.238969 LR 0.000250 Time 0.021086 -2023-02-13 18:31:43,202 - Epoch: [167][ 340/ 1207] Overall Loss 0.238147 Objective Loss 0.238147 LR 0.000250 Time 0.021019 -2023-02-13 18:31:43,390 - Epoch: [167][ 350/ 1207] Overall Loss 0.238711 Objective Loss 0.238711 LR 0.000250 Time 0.020954 -2023-02-13 18:31:43,578 - Epoch: [167][ 360/ 1207] Overall Loss 0.238440 Objective Loss 0.238440 LR 0.000250 Time 0.020895 -2023-02-13 18:31:43,767 - Epoch: [167][ 370/ 1207] Overall Loss 0.238941 Objective Loss 0.238941 LR 0.000250 Time 0.020840 -2023-02-13 18:31:43,955 - Epoch: [167][ 380/ 1207] Overall Loss 0.238796 Objective Loss 0.238796 LR 0.000250 Time 0.020786 -2023-02-13 18:31:44,144 - Epoch: [167][ 390/ 1207] Overall Loss 0.238830 Objective Loss 0.238830 LR 0.000250 Time 0.020736 -2023-02-13 18:31:44,332 - Epoch: [167][ 400/ 1207] Overall Loss 0.238910 Objective Loss 0.238910 LR 0.000250 Time 0.020687 -2023-02-13 18:31:44,521 - Epoch: [167][ 410/ 1207] Overall Loss 0.238275 Objective Loss 0.238275 LR 0.000250 Time 0.020641 -2023-02-13 18:31:44,710 - Epoch: [167][ 420/ 1207] Overall Loss 0.238100 Objective Loss 0.238100 LR 0.000250 Time 0.020599 -2023-02-13 18:31:44,898 - Epoch: [167][ 430/ 1207] Overall Loss 0.238533 Objective Loss 0.238533 LR 0.000250 Time 0.020556 -2023-02-13 18:31:45,086 - Epoch: [167][ 440/ 1207] Overall Loss 0.239280 Objective Loss 0.239280 LR 0.000250 Time 0.020516 -2023-02-13 18:31:45,274 - Epoch: [167][ 450/ 1207] Overall Loss 0.239451 Objective Loss 0.239451 LR 0.000250 Time 0.020477 -2023-02-13 18:31:45,463 - Epoch: [167][ 460/ 1207] Overall Loss 0.239312 Objective Loss 0.239312 LR 0.000250 Time 0.020442 -2023-02-13 18:31:45,654 - Epoch: [167][ 470/ 1207] Overall Loss 0.239143 Objective Loss 0.239143 LR 0.000250 Time 0.020412 -2023-02-13 18:31:45,844 - Epoch: [167][ 480/ 1207] Overall Loss 0.239081 Objective Loss 0.239081 LR 0.000250 Time 0.020382 -2023-02-13 18:31:46,034 - Epoch: [167][ 490/ 1207] Overall Loss 0.238926 Objective Loss 0.238926 LR 0.000250 Time 0.020353 -2023-02-13 18:31:46,224 - Epoch: [167][ 500/ 1207] Overall Loss 0.239381 Objective Loss 0.239381 LR 0.000250 Time 0.020326 -2023-02-13 18:31:46,413 - Epoch: [167][ 510/ 1207] Overall Loss 0.239137 Objective Loss 0.239137 LR 0.000250 Time 0.020297 -2023-02-13 18:31:46,605 - Epoch: [167][ 520/ 1207] Overall Loss 0.239208 Objective Loss 0.239208 LR 0.000250 Time 0.020274 -2023-02-13 18:31:46,796 - Epoch: [167][ 530/ 1207] Overall Loss 0.239060 Objective Loss 0.239060 LR 0.000250 Time 0.020252 -2023-02-13 18:31:46,988 - Epoch: [167][ 540/ 1207] Overall Loss 0.239208 Objective Loss 0.239208 LR 0.000250 Time 0.020232 -2023-02-13 18:31:47,178 - Epoch: [167][ 550/ 1207] Overall Loss 0.239239 Objective Loss 0.239239 LR 0.000250 Time 0.020210 -2023-02-13 18:31:47,370 - Epoch: [167][ 560/ 1207] Overall Loss 0.239345 Objective Loss 0.239345 LR 0.000250 Time 0.020190 -2023-02-13 18:31:47,561 - Epoch: [167][ 570/ 1207] Overall Loss 0.238964 Objective Loss 0.238964 LR 0.000250 Time 0.020170 -2023-02-13 18:31:47,753 - Epoch: [167][ 580/ 1207] Overall Loss 0.238738 Objective Loss 0.238738 LR 0.000250 Time 0.020152 -2023-02-13 18:31:47,944 - Epoch: [167][ 590/ 1207] Overall Loss 0.238688 Objective Loss 0.238688 LR 0.000250 Time 0.020135 -2023-02-13 18:31:48,135 - Epoch: [167][ 600/ 1207] Overall Loss 0.238762 Objective Loss 0.238762 LR 0.000250 Time 0.020117 -2023-02-13 18:31:48,326 - Epoch: [167][ 610/ 1207] Overall Loss 0.238714 Objective Loss 0.238714 LR 0.000250 Time 0.020099 -2023-02-13 18:31:48,518 - Epoch: [167][ 620/ 1207] Overall Loss 0.238729 Objective Loss 0.238729 LR 0.000250 Time 0.020084 -2023-02-13 18:31:48,710 - Epoch: [167][ 630/ 1207] Overall Loss 0.238428 Objective Loss 0.238428 LR 0.000250 Time 0.020069 -2023-02-13 18:31:48,901 - Epoch: [167][ 640/ 1207] Overall Loss 0.237986 Objective Loss 0.237986 LR 0.000250 Time 0.020054 -2023-02-13 18:31:49,091 - Epoch: [167][ 650/ 1207] Overall Loss 0.238116 Objective Loss 0.238116 LR 0.000250 Time 0.020038 -2023-02-13 18:31:49,283 - Epoch: [167][ 660/ 1207] Overall Loss 0.238468 Objective Loss 0.238468 LR 0.000250 Time 0.020024 -2023-02-13 18:31:49,474 - Epoch: [167][ 670/ 1207] Overall Loss 0.238550 Objective Loss 0.238550 LR 0.000250 Time 0.020010 -2023-02-13 18:31:49,667 - Epoch: [167][ 680/ 1207] Overall Loss 0.238614 Objective Loss 0.238614 LR 0.000250 Time 0.019999 -2023-02-13 18:31:49,858 - Epoch: [167][ 690/ 1207] Overall Loss 0.238853 Objective Loss 0.238853 LR 0.000250 Time 0.019985 -2023-02-13 18:31:50,050 - Epoch: [167][ 700/ 1207] Overall Loss 0.238581 Objective Loss 0.238581 LR 0.000250 Time 0.019973 -2023-02-13 18:31:50,241 - Epoch: [167][ 710/ 1207] Overall Loss 0.238916 Objective Loss 0.238916 LR 0.000250 Time 0.019960 -2023-02-13 18:31:50,433 - Epoch: [167][ 720/ 1207] Overall Loss 0.239135 Objective Loss 0.239135 LR 0.000250 Time 0.019949 -2023-02-13 18:31:50,624 - Epoch: [167][ 730/ 1207] Overall Loss 0.239140 Objective Loss 0.239140 LR 0.000250 Time 0.019938 -2023-02-13 18:31:50,817 - Epoch: [167][ 740/ 1207] Overall Loss 0.239503 Objective Loss 0.239503 LR 0.000250 Time 0.019928 -2023-02-13 18:31:51,009 - Epoch: [167][ 750/ 1207] Overall Loss 0.239555 Objective Loss 0.239555 LR 0.000250 Time 0.019918 -2023-02-13 18:31:51,201 - Epoch: [167][ 760/ 1207] Overall Loss 0.239789 Objective Loss 0.239789 LR 0.000250 Time 0.019907 -2023-02-13 18:31:51,391 - Epoch: [167][ 770/ 1207] Overall Loss 0.239649 Objective Loss 0.239649 LR 0.000250 Time 0.019896 -2023-02-13 18:31:51,584 - Epoch: [167][ 780/ 1207] Overall Loss 0.239704 Objective Loss 0.239704 LR 0.000250 Time 0.019887 -2023-02-13 18:31:51,776 - Epoch: [167][ 790/ 1207] Overall Loss 0.239738 Objective Loss 0.239738 LR 0.000250 Time 0.019878 -2023-02-13 18:31:51,969 - Epoch: [167][ 800/ 1207] Overall Loss 0.239855 Objective Loss 0.239855 LR 0.000250 Time 0.019870 -2023-02-13 18:31:52,160 - Epoch: [167][ 810/ 1207] Overall Loss 0.239990 Objective Loss 0.239990 LR 0.000250 Time 0.019860 -2023-02-13 18:31:52,352 - Epoch: [167][ 820/ 1207] Overall Loss 0.239759 Objective Loss 0.239759 LR 0.000250 Time 0.019852 -2023-02-13 18:31:52,543 - Epoch: [167][ 830/ 1207] Overall Loss 0.239625 Objective Loss 0.239625 LR 0.000250 Time 0.019843 -2023-02-13 18:31:52,736 - Epoch: [167][ 840/ 1207] Overall Loss 0.239597 Objective Loss 0.239597 LR 0.000250 Time 0.019836 -2023-02-13 18:31:52,928 - Epoch: [167][ 850/ 1207] Overall Loss 0.239394 Objective Loss 0.239394 LR 0.000250 Time 0.019828 -2023-02-13 18:31:53,120 - Epoch: [167][ 860/ 1207] Overall Loss 0.239189 Objective Loss 0.239189 LR 0.000250 Time 0.019820 -2023-02-13 18:31:53,311 - Epoch: [167][ 870/ 1207] Overall Loss 0.239361 Objective Loss 0.239361 LR 0.000250 Time 0.019812 -2023-02-13 18:31:53,504 - Epoch: [167][ 880/ 1207] Overall Loss 0.239360 Objective Loss 0.239360 LR 0.000250 Time 0.019805 -2023-02-13 18:31:53,696 - Epoch: [167][ 890/ 1207] Overall Loss 0.239540 Objective Loss 0.239540 LR 0.000250 Time 0.019798 -2023-02-13 18:31:53,888 - Epoch: [167][ 900/ 1207] Overall Loss 0.239383 Objective Loss 0.239383 LR 0.000250 Time 0.019791 -2023-02-13 18:31:54,079 - Epoch: [167][ 910/ 1207] Overall Loss 0.239484 Objective Loss 0.239484 LR 0.000250 Time 0.019783 -2023-02-13 18:31:54,272 - Epoch: [167][ 920/ 1207] Overall Loss 0.239549 Objective Loss 0.239549 LR 0.000250 Time 0.019777 -2023-02-13 18:31:54,462 - Epoch: [167][ 930/ 1207] Overall Loss 0.239482 Objective Loss 0.239482 LR 0.000250 Time 0.019768 -2023-02-13 18:31:54,654 - Epoch: [167][ 940/ 1207] Overall Loss 0.239278 Objective Loss 0.239278 LR 0.000250 Time 0.019762 -2023-02-13 18:31:54,846 - Epoch: [167][ 950/ 1207] Overall Loss 0.239140 Objective Loss 0.239140 LR 0.000250 Time 0.019755 -2023-02-13 18:31:55,037 - Epoch: [167][ 960/ 1207] Overall Loss 0.239163 Objective Loss 0.239163 LR 0.000250 Time 0.019749 -2023-02-13 18:31:55,229 - Epoch: [167][ 970/ 1207] Overall Loss 0.239097 Objective Loss 0.239097 LR 0.000250 Time 0.019742 -2023-02-13 18:31:55,420 - Epoch: [167][ 980/ 1207] Overall Loss 0.238950 Objective Loss 0.238950 LR 0.000250 Time 0.019736 -2023-02-13 18:31:55,612 - Epoch: [167][ 990/ 1207] Overall Loss 0.239142 Objective Loss 0.239142 LR 0.000250 Time 0.019730 -2023-02-13 18:31:55,805 - Epoch: [167][ 1000/ 1207] Overall Loss 0.239014 Objective Loss 0.239014 LR 0.000250 Time 0.019725 -2023-02-13 18:31:55,997 - Epoch: [167][ 1010/ 1207] Overall Loss 0.239202 Objective Loss 0.239202 LR 0.000250 Time 0.019719 -2023-02-13 18:31:56,189 - Epoch: [167][ 1020/ 1207] Overall Loss 0.239218 Objective Loss 0.239218 LR 0.000250 Time 0.019714 -2023-02-13 18:31:56,381 - Epoch: [167][ 1030/ 1207] Overall Loss 0.239447 Objective Loss 0.239447 LR 0.000250 Time 0.019708 -2023-02-13 18:31:56,573 - Epoch: [167][ 1040/ 1207] Overall Loss 0.239308 Objective Loss 0.239308 LR 0.000250 Time 0.019703 -2023-02-13 18:31:56,766 - Epoch: [167][ 1050/ 1207] Overall Loss 0.239513 Objective Loss 0.239513 LR 0.000250 Time 0.019699 -2023-02-13 18:31:56,959 - Epoch: [167][ 1060/ 1207] Overall Loss 0.239587 Objective Loss 0.239587 LR 0.000250 Time 0.019695 -2023-02-13 18:31:57,150 - Epoch: [167][ 1070/ 1207] Overall Loss 0.239596 Objective Loss 0.239596 LR 0.000250 Time 0.019689 -2023-02-13 18:31:57,343 - Epoch: [167][ 1080/ 1207] Overall Loss 0.239799 Objective Loss 0.239799 LR 0.000250 Time 0.019685 -2023-02-13 18:31:57,535 - Epoch: [167][ 1090/ 1207] Overall Loss 0.239853 Objective Loss 0.239853 LR 0.000250 Time 0.019680 -2023-02-13 18:31:57,728 - Epoch: [167][ 1100/ 1207] Overall Loss 0.239967 Objective Loss 0.239967 LR 0.000250 Time 0.019677 -2023-02-13 18:31:57,920 - Epoch: [167][ 1110/ 1207] Overall Loss 0.239823 Objective Loss 0.239823 LR 0.000250 Time 0.019672 -2023-02-13 18:31:58,112 - Epoch: [167][ 1120/ 1207] Overall Loss 0.239810 Objective Loss 0.239810 LR 0.000250 Time 0.019667 -2023-02-13 18:31:58,304 - Epoch: [167][ 1130/ 1207] Overall Loss 0.239664 Objective Loss 0.239664 LR 0.000250 Time 0.019663 -2023-02-13 18:31:58,496 - Epoch: [167][ 1140/ 1207] Overall Loss 0.239647 Objective Loss 0.239647 LR 0.000250 Time 0.019659 -2023-02-13 18:31:58,687 - Epoch: [167][ 1150/ 1207] Overall Loss 0.239698 Objective Loss 0.239698 LR 0.000250 Time 0.019653 -2023-02-13 18:31:58,878 - Epoch: [167][ 1160/ 1207] Overall Loss 0.239684 Objective Loss 0.239684 LR 0.000250 Time 0.019648 -2023-02-13 18:31:59,069 - Epoch: [167][ 1170/ 1207] Overall Loss 0.239610 Objective Loss 0.239610 LR 0.000250 Time 0.019643 -2023-02-13 18:31:59,260 - Epoch: [167][ 1180/ 1207] Overall Loss 0.239606 Objective Loss 0.239606 LR 0.000250 Time 0.019638 -2023-02-13 18:31:59,451 - Epoch: [167][ 1190/ 1207] Overall Loss 0.239499 Objective Loss 0.239499 LR 0.000250 Time 0.019633 -2023-02-13 18:31:59,693 - Epoch: [167][ 1200/ 1207] Overall Loss 0.239419 Objective Loss 0.239419 LR 0.000250 Time 0.019671 -2023-02-13 18:31:59,808 - Epoch: [167][ 1207/ 1207] Overall Loss 0.239397 Objective Loss 0.239397 Top1 83.231707 Top5 98.170732 LR 0.000250 Time 0.019652 -2023-02-13 18:31:59,878 - --- validate (epoch=167)----------- -2023-02-13 18:31:59,878 - 34311 samples (256 per mini-batch) -2023-02-13 18:32:00,374 - Epoch: [167][ 10/ 135] Loss 0.333149 Top1 83.671875 Top5 97.851562 -2023-02-13 18:32:00,498 - Epoch: [167][ 20/ 135] Loss 0.313422 Top1 84.316406 Top5 97.968750 -2023-02-13 18:32:00,627 - Epoch: [167][ 30/ 135] Loss 0.308050 Top1 84.114583 Top5 98.007812 -2023-02-13 18:32:00,756 - Epoch: [167][ 40/ 135] Loss 0.307213 Top1 84.345703 Top5 97.783203 -2023-02-13 18:32:00,886 - Epoch: [167][ 50/ 135] Loss 0.311346 Top1 83.960938 Top5 97.796875 -2023-02-13 18:32:01,015 - Epoch: [167][ 60/ 135] Loss 0.310597 Top1 84.023438 Top5 97.766927 -2023-02-13 18:32:01,144 - Epoch: [167][ 70/ 135] Loss 0.310327 Top1 84.095982 Top5 97.745536 -2023-02-13 18:32:01,273 - Epoch: [167][ 80/ 135] Loss 0.313282 Top1 83.984375 Top5 97.729492 -2023-02-13 18:32:01,401 - Epoch: [167][ 90/ 135] Loss 0.313112 Top1 83.862847 Top5 97.656250 -2023-02-13 18:32:01,529 - Epoch: [167][ 100/ 135] Loss 0.312841 Top1 83.796875 Top5 97.625000 -2023-02-13 18:32:01,657 - Epoch: [167][ 110/ 135] Loss 0.311479 Top1 83.860085 Top5 97.638494 -2023-02-13 18:32:01,786 - Epoch: [167][ 120/ 135] Loss 0.309265 Top1 83.876953 Top5 97.649740 -2023-02-13 18:32:01,917 - Epoch: [167][ 130/ 135] Loss 0.307201 Top1 83.963341 Top5 97.671274 -2023-02-13 18:32:01,964 - Epoch: [167][ 135/ 135] Loss 0.308609 Top1 83.990557 Top5 97.680044 -2023-02-13 18:32:02,046 - ==> Top1: 83.991 Top5: 97.680 Loss: 0.309 - -2023-02-13 18:32:02,046 - ==> Confusion: -[[ 859 3 6 2 6 4 0 3 4 48 0 3 0 5 7 4 3 3 1 3 3] - [ 3 951 1 1 8 31 0 16 2 1 1 0 2 0 0 1 3 1 2 3 6] - [ 6 4 954 11 2 2 14 15 0 2 4 2 2 4 3 11 6 4 3 5 4] - [ 7 1 15 904 1 6 1 1 3 2 13 0 9 0 20 2 4 6 15 0 6] - [ 14 5 0 1 996 12 2 1 0 1 0 4 3 2 9 6 3 3 0 1 3] - [ 2 17 1 4 4 977 4 16 1 3 1 8 1 15 1 2 4 3 1 3 2] - [ 4 5 11 2 0 3 1042 6 0 2 0 1 2 2 0 6 2 1 2 3 5] - [ 2 5 9 1 4 30 1 938 0 1 1 4 5 1 0 0 2 1 9 9 1] - [ 15 1 1 1 2 0 1 0 905 39 9 2 0 13 11 2 2 0 4 0 1] - [ 74 0 4 0 10 1 0 1 34 853 0 0 0 19 4 2 1 3 3 0 3] - [ 0 1 5 4 2 4 3 4 13 2 988 1 1 8 2 0 1 1 6 0 5] - [ 2 3 1 0 2 10 0 4 3 2 0 922 24 9 1 6 1 7 2 3 3] - [ 1 0 1 7 3 4 0 1 4 1 0 22 888 1 1 7 2 12 0 0 4] - [ 6 3 3 0 8 10 1 1 10 9 5 8 2 939 5 2 4 2 0 1 5] - [ 4 3 1 17 6 3 0 1 21 6 2 0 2 1 1004 0 0 7 7 0 7] - [ 3 3 5 2 4 1 2 2 1 0 0 4 7 2 0 972 15 10 1 6 6] - [ 2 5 2 1 7 3 0 1 2 1 0 2 3 1 1 10 1007 3 0 4 6] - [ 6 2 0 4 2 1 2 0 0 0 0 9 18 1 0 20 0 976 0 2 8] - [ 3 6 5 7 1 1 0 31 3 0 5 1 4 0 11 0 0 4 996 2 6] - [ 1 3 0 0 0 5 7 8 0 1 0 15 3 1 0 7 2 4 0 1086 5] - [ 159 251 227 129 146 231 88 173 91 86 195 128 350 327 159 105 334 106 181 307 9661]] - -2023-02-13 18:32:02,048 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:32:02,048 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:32:02,054 - - -2023-02-13 18:32:02,054 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:32:02,973 - Epoch: [168][ 10/ 1207] Overall Loss 0.235741 Objective Loss 0.235741 LR 0.000250 Time 0.091853 -2023-02-13 18:32:03,164 - Epoch: [168][ 20/ 1207] Overall Loss 0.238170 Objective Loss 0.238170 LR 0.000250 Time 0.055457 -2023-02-13 18:32:03,354 - Epoch: [168][ 30/ 1207] Overall Loss 0.242662 Objective Loss 0.242662 LR 0.000250 Time 0.043290 -2023-02-13 18:32:03,544 - Epoch: [168][ 40/ 1207] Overall Loss 0.242532 Objective Loss 0.242532 LR 0.000250 Time 0.037211 -2023-02-13 18:32:03,734 - Epoch: [168][ 50/ 1207] Overall Loss 0.244285 Objective Loss 0.244285 LR 0.000250 Time 0.033573 -2023-02-13 18:32:03,924 - Epoch: [168][ 60/ 1207] Overall Loss 0.242879 Objective Loss 0.242879 LR 0.000250 Time 0.031135 -2023-02-13 18:32:04,114 - Epoch: [168][ 70/ 1207] Overall Loss 0.241156 Objective Loss 0.241156 LR 0.000250 Time 0.029392 -2023-02-13 18:32:04,303 - Epoch: [168][ 80/ 1207] Overall Loss 0.240112 Objective Loss 0.240112 LR 0.000250 Time 0.028081 -2023-02-13 18:32:04,493 - Epoch: [168][ 90/ 1207] Overall Loss 0.240657 Objective Loss 0.240657 LR 0.000250 Time 0.027065 -2023-02-13 18:32:04,684 - Epoch: [168][ 100/ 1207] Overall Loss 0.240512 Objective Loss 0.240512 LR 0.000250 Time 0.026267 -2023-02-13 18:32:04,875 - Epoch: [168][ 110/ 1207] Overall Loss 0.240358 Objective Loss 0.240358 LR 0.000250 Time 0.025607 -2023-02-13 18:32:05,065 - Epoch: [168][ 120/ 1207] Overall Loss 0.239738 Objective Loss 0.239738 LR 0.000250 Time 0.025058 -2023-02-13 18:32:05,255 - Epoch: [168][ 130/ 1207] Overall Loss 0.240853 Objective Loss 0.240853 LR 0.000250 Time 0.024589 -2023-02-13 18:32:05,445 - Epoch: [168][ 140/ 1207] Overall Loss 0.240618 Objective Loss 0.240618 LR 0.000250 Time 0.024188 -2023-02-13 18:32:05,636 - Epoch: [168][ 150/ 1207] Overall Loss 0.240366 Objective Loss 0.240366 LR 0.000250 Time 0.023845 -2023-02-13 18:32:05,827 - Epoch: [168][ 160/ 1207] Overall Loss 0.238902 Objective Loss 0.238902 LR 0.000250 Time 0.023544 -2023-02-13 18:32:06,018 - Epoch: [168][ 170/ 1207] Overall Loss 0.238848 Objective Loss 0.238848 LR 0.000250 Time 0.023281 -2023-02-13 18:32:06,209 - Epoch: [168][ 180/ 1207] Overall Loss 0.237905 Objective Loss 0.237905 LR 0.000250 Time 0.023045 -2023-02-13 18:32:06,399 - Epoch: [168][ 190/ 1207] Overall Loss 0.236721 Objective Loss 0.236721 LR 0.000250 Time 0.022830 -2023-02-13 18:32:06,589 - Epoch: [168][ 200/ 1207] Overall Loss 0.237338 Objective Loss 0.237338 LR 0.000250 Time 0.022637 -2023-02-13 18:32:06,780 - Epoch: [168][ 210/ 1207] Overall Loss 0.237923 Objective Loss 0.237923 LR 0.000250 Time 0.022467 -2023-02-13 18:32:06,971 - Epoch: [168][ 220/ 1207] Overall Loss 0.238607 Objective Loss 0.238607 LR 0.000250 Time 0.022311 -2023-02-13 18:32:07,161 - Epoch: [168][ 230/ 1207] Overall Loss 0.238024 Objective Loss 0.238024 LR 0.000250 Time 0.022168 -2023-02-13 18:32:07,351 - Epoch: [168][ 240/ 1207] Overall Loss 0.237143 Objective Loss 0.237143 LR 0.000250 Time 0.022035 -2023-02-13 18:32:07,542 - Epoch: [168][ 250/ 1207] Overall Loss 0.236781 Objective Loss 0.236781 LR 0.000250 Time 0.021914 -2023-02-13 18:32:07,733 - Epoch: [168][ 260/ 1207] Overall Loss 0.236855 Objective Loss 0.236855 LR 0.000250 Time 0.021806 -2023-02-13 18:32:07,923 - Epoch: [168][ 270/ 1207] Overall Loss 0.236864 Objective Loss 0.236864 LR 0.000250 Time 0.021700 -2023-02-13 18:32:08,113 - Epoch: [168][ 280/ 1207] Overall Loss 0.236413 Objective Loss 0.236413 LR 0.000250 Time 0.021603 -2023-02-13 18:32:08,303 - Epoch: [168][ 290/ 1207] Overall Loss 0.236607 Objective Loss 0.236607 LR 0.000250 Time 0.021512 -2023-02-13 18:32:08,494 - Epoch: [168][ 300/ 1207] Overall Loss 0.236355 Objective Loss 0.236355 LR 0.000250 Time 0.021428 -2023-02-13 18:32:08,685 - Epoch: [168][ 310/ 1207] Overall Loss 0.236130 Objective Loss 0.236130 LR 0.000250 Time 0.021352 -2023-02-13 18:32:08,875 - Epoch: [168][ 320/ 1207] Overall Loss 0.235934 Objective Loss 0.235934 LR 0.000250 Time 0.021279 -2023-02-13 18:32:09,065 - Epoch: [168][ 330/ 1207] Overall Loss 0.236289 Objective Loss 0.236289 LR 0.000250 Time 0.021210 -2023-02-13 18:32:09,258 - Epoch: [168][ 340/ 1207] Overall Loss 0.236469 Objective Loss 0.236469 LR 0.000250 Time 0.021150 -2023-02-13 18:32:09,450 - Epoch: [168][ 350/ 1207] Overall Loss 0.236786 Objective Loss 0.236786 LR 0.000250 Time 0.021094 -2023-02-13 18:32:09,642 - Epoch: [168][ 360/ 1207] Overall Loss 0.236922 Objective Loss 0.236922 LR 0.000250 Time 0.021042 -2023-02-13 18:32:09,836 - Epoch: [168][ 370/ 1207] Overall Loss 0.237337 Objective Loss 0.237337 LR 0.000250 Time 0.020995 -2023-02-13 18:32:10,028 - Epoch: [168][ 380/ 1207] Overall Loss 0.237813 Objective Loss 0.237813 LR 0.000250 Time 0.020946 -2023-02-13 18:32:10,220 - Epoch: [168][ 390/ 1207] Overall Loss 0.238168 Objective Loss 0.238168 LR 0.000250 Time 0.020902 -2023-02-13 18:32:10,412 - Epoch: [168][ 400/ 1207] Overall Loss 0.238548 Objective Loss 0.238548 LR 0.000250 Time 0.020858 -2023-02-13 18:32:10,605 - Epoch: [168][ 410/ 1207] Overall Loss 0.238261 Objective Loss 0.238261 LR 0.000250 Time 0.020819 -2023-02-13 18:32:10,798 - Epoch: [168][ 420/ 1207] Overall Loss 0.238216 Objective Loss 0.238216 LR 0.000250 Time 0.020782 -2023-02-13 18:32:10,992 - Epoch: [168][ 430/ 1207] Overall Loss 0.238459 Objective Loss 0.238459 LR 0.000250 Time 0.020750 -2023-02-13 18:32:11,185 - Epoch: [168][ 440/ 1207] Overall Loss 0.238117 Objective Loss 0.238117 LR 0.000250 Time 0.020715 -2023-02-13 18:32:11,378 - Epoch: [168][ 450/ 1207] Overall Loss 0.238369 Objective Loss 0.238369 LR 0.000250 Time 0.020682 -2023-02-13 18:32:11,570 - Epoch: [168][ 460/ 1207] Overall Loss 0.238371 Objective Loss 0.238371 LR 0.000250 Time 0.020650 -2023-02-13 18:32:11,763 - Epoch: [168][ 470/ 1207] Overall Loss 0.238294 Objective Loss 0.238294 LR 0.000250 Time 0.020620 -2023-02-13 18:32:11,954 - Epoch: [168][ 480/ 1207] Overall Loss 0.238123 Objective Loss 0.238123 LR 0.000250 Time 0.020587 -2023-02-13 18:32:12,145 - Epoch: [168][ 490/ 1207] Overall Loss 0.237867 Objective Loss 0.237867 LR 0.000250 Time 0.020556 -2023-02-13 18:32:12,335 - Epoch: [168][ 500/ 1207] Overall Loss 0.237411 Objective Loss 0.237411 LR 0.000250 Time 0.020525 -2023-02-13 18:32:12,526 - Epoch: [168][ 510/ 1207] Overall Loss 0.237628 Objective Loss 0.237628 LR 0.000250 Time 0.020496 -2023-02-13 18:32:12,717 - Epoch: [168][ 520/ 1207] Overall Loss 0.237203 Objective Loss 0.237203 LR 0.000250 Time 0.020470 -2023-02-13 18:32:12,908 - Epoch: [168][ 530/ 1207] Overall Loss 0.236795 Objective Loss 0.236795 LR 0.000250 Time 0.020442 -2023-02-13 18:32:13,098 - Epoch: [168][ 540/ 1207] Overall Loss 0.236659 Objective Loss 0.236659 LR 0.000250 Time 0.020414 -2023-02-13 18:32:13,288 - Epoch: [168][ 550/ 1207] Overall Loss 0.236665 Objective Loss 0.236665 LR 0.000250 Time 0.020389 -2023-02-13 18:32:13,479 - Epoch: [168][ 560/ 1207] Overall Loss 0.237314 Objective Loss 0.237314 LR 0.000250 Time 0.020365 -2023-02-13 18:32:13,670 - Epoch: [168][ 570/ 1207] Overall Loss 0.237343 Objective Loss 0.237343 LR 0.000250 Time 0.020342 -2023-02-13 18:32:13,861 - Epoch: [168][ 580/ 1207] Overall Loss 0.237120 Objective Loss 0.237120 LR 0.000250 Time 0.020320 -2023-02-13 18:32:14,052 - Epoch: [168][ 590/ 1207] Overall Loss 0.237530 Objective Loss 0.237530 LR 0.000250 Time 0.020298 -2023-02-13 18:32:14,243 - Epoch: [168][ 600/ 1207] Overall Loss 0.237472 Objective Loss 0.237472 LR 0.000250 Time 0.020277 -2023-02-13 18:32:14,434 - Epoch: [168][ 610/ 1207] Overall Loss 0.237413 Objective Loss 0.237413 LR 0.000250 Time 0.020258 -2023-02-13 18:32:14,625 - Epoch: [168][ 620/ 1207] Overall Loss 0.237228 Objective Loss 0.237228 LR 0.000250 Time 0.020238 -2023-02-13 18:32:14,817 - Epoch: [168][ 630/ 1207] Overall Loss 0.237446 Objective Loss 0.237446 LR 0.000250 Time 0.020222 -2023-02-13 18:32:15,009 - Epoch: [168][ 640/ 1207] Overall Loss 0.237538 Objective Loss 0.237538 LR 0.000250 Time 0.020205 -2023-02-13 18:32:15,200 - Epoch: [168][ 650/ 1207] Overall Loss 0.237291 Objective Loss 0.237291 LR 0.000250 Time 0.020187 -2023-02-13 18:32:15,389 - Epoch: [168][ 660/ 1207] Overall Loss 0.237303 Objective Loss 0.237303 LR 0.000250 Time 0.020168 -2023-02-13 18:32:15,579 - Epoch: [168][ 670/ 1207] Overall Loss 0.237723 Objective Loss 0.237723 LR 0.000250 Time 0.020150 -2023-02-13 18:32:15,769 - Epoch: [168][ 680/ 1207] Overall Loss 0.238082 Objective Loss 0.238082 LR 0.000250 Time 0.020132 -2023-02-13 18:32:15,961 - Epoch: [168][ 690/ 1207] Overall Loss 0.237909 Objective Loss 0.237909 LR 0.000250 Time 0.020118 -2023-02-13 18:32:16,156 - Epoch: [168][ 700/ 1207] Overall Loss 0.237810 Objective Loss 0.237810 LR 0.000250 Time 0.020109 -2023-02-13 18:32:16,351 - Epoch: [168][ 710/ 1207] Overall Loss 0.237759 Objective Loss 0.237759 LR 0.000250 Time 0.020099 -2023-02-13 18:32:16,547 - Epoch: [168][ 720/ 1207] Overall Loss 0.237981 Objective Loss 0.237981 LR 0.000250 Time 0.020092 -2023-02-13 18:32:16,743 - Epoch: [168][ 730/ 1207] Overall Loss 0.237821 Objective Loss 0.237821 LR 0.000250 Time 0.020085 -2023-02-13 18:32:16,940 - Epoch: [168][ 740/ 1207] Overall Loss 0.237878 Objective Loss 0.237878 LR 0.000250 Time 0.020079 -2023-02-13 18:32:17,135 - Epoch: [168][ 750/ 1207] Overall Loss 0.237750 Objective Loss 0.237750 LR 0.000250 Time 0.020071 -2023-02-13 18:32:17,331 - Epoch: [168][ 760/ 1207] Overall Loss 0.237737 Objective Loss 0.237737 LR 0.000250 Time 0.020065 -2023-02-13 18:32:17,526 - Epoch: [168][ 770/ 1207] Overall Loss 0.237607 Objective Loss 0.237607 LR 0.000250 Time 0.020057 -2023-02-13 18:32:17,723 - Epoch: [168][ 780/ 1207] Overall Loss 0.237302 Objective Loss 0.237302 LR 0.000250 Time 0.020052 -2023-02-13 18:32:17,918 - Epoch: [168][ 790/ 1207] Overall Loss 0.237045 Objective Loss 0.237045 LR 0.000250 Time 0.020045 -2023-02-13 18:32:18,113 - Epoch: [168][ 800/ 1207] Overall Loss 0.236811 Objective Loss 0.236811 LR 0.000250 Time 0.020037 -2023-02-13 18:32:18,302 - Epoch: [168][ 810/ 1207] Overall Loss 0.236885 Objective Loss 0.236885 LR 0.000250 Time 0.020023 -2023-02-13 18:32:18,492 - Epoch: [168][ 820/ 1207] Overall Loss 0.236870 Objective Loss 0.236870 LR 0.000250 Time 0.020009 -2023-02-13 18:32:18,682 - Epoch: [168][ 830/ 1207] Overall Loss 0.236959 Objective Loss 0.236959 LR 0.000250 Time 0.019996 -2023-02-13 18:32:18,872 - Epoch: [168][ 840/ 1207] Overall Loss 0.236837 Objective Loss 0.236837 LR 0.000250 Time 0.019984 -2023-02-13 18:32:19,062 - Epoch: [168][ 850/ 1207] Overall Loss 0.236815 Objective Loss 0.236815 LR 0.000250 Time 0.019972 -2023-02-13 18:32:19,251 - Epoch: [168][ 860/ 1207] Overall Loss 0.236890 Objective Loss 0.236890 LR 0.000250 Time 0.019960 -2023-02-13 18:32:19,441 - Epoch: [168][ 870/ 1207] Overall Loss 0.236983 Objective Loss 0.236983 LR 0.000250 Time 0.019948 -2023-02-13 18:32:19,631 - Epoch: [168][ 880/ 1207] Overall Loss 0.236910 Objective Loss 0.236910 LR 0.000250 Time 0.019936 -2023-02-13 18:32:19,822 - Epoch: [168][ 890/ 1207] Overall Loss 0.236947 Objective Loss 0.236947 LR 0.000250 Time 0.019927 -2023-02-13 18:32:20,011 - Epoch: [168][ 900/ 1207] Overall Loss 0.236908 Objective Loss 0.236908 LR 0.000250 Time 0.019916 -2023-02-13 18:32:20,201 - Epoch: [168][ 910/ 1207] Overall Loss 0.236979 Objective Loss 0.236979 LR 0.000250 Time 0.019905 -2023-02-13 18:32:20,391 - Epoch: [168][ 920/ 1207] Overall Loss 0.236916 Objective Loss 0.236916 LR 0.000250 Time 0.019894 -2023-02-13 18:32:20,580 - Epoch: [168][ 930/ 1207] Overall Loss 0.236683 Objective Loss 0.236683 LR 0.000250 Time 0.019884 -2023-02-13 18:32:20,772 - Epoch: [168][ 940/ 1207] Overall Loss 0.237043 Objective Loss 0.237043 LR 0.000250 Time 0.019876 -2023-02-13 18:32:20,966 - Epoch: [168][ 950/ 1207] Overall Loss 0.236958 Objective Loss 0.236958 LR 0.000250 Time 0.019870 -2023-02-13 18:32:21,158 - Epoch: [168][ 960/ 1207] Overall Loss 0.236968 Objective Loss 0.236968 LR 0.000250 Time 0.019863 -2023-02-13 18:32:21,350 - Epoch: [168][ 970/ 1207] Overall Loss 0.237003 Objective Loss 0.237003 LR 0.000250 Time 0.019856 -2023-02-13 18:32:21,542 - Epoch: [168][ 980/ 1207] Overall Loss 0.237048 Objective Loss 0.237048 LR 0.000250 Time 0.019849 -2023-02-13 18:32:21,735 - Epoch: [168][ 990/ 1207] Overall Loss 0.237037 Objective Loss 0.237037 LR 0.000250 Time 0.019843 -2023-02-13 18:32:21,927 - Epoch: [168][ 1000/ 1207] Overall Loss 0.237053 Objective Loss 0.237053 LR 0.000250 Time 0.019836 -2023-02-13 18:32:22,120 - Epoch: [168][ 1010/ 1207] Overall Loss 0.236945 Objective Loss 0.236945 LR 0.000250 Time 0.019830 -2023-02-13 18:32:22,311 - Epoch: [168][ 1020/ 1207] Overall Loss 0.236963 Objective Loss 0.236963 LR 0.000250 Time 0.019823 -2023-02-13 18:32:22,504 - Epoch: [168][ 1030/ 1207] Overall Loss 0.236956 Objective Loss 0.236956 LR 0.000250 Time 0.019817 -2023-02-13 18:32:22,696 - Epoch: [168][ 1040/ 1207] Overall Loss 0.236946 Objective Loss 0.236946 LR 0.000250 Time 0.019811 -2023-02-13 18:32:22,889 - Epoch: [168][ 1050/ 1207] Overall Loss 0.236847 Objective Loss 0.236847 LR 0.000250 Time 0.019806 -2023-02-13 18:32:23,081 - Epoch: [168][ 1060/ 1207] Overall Loss 0.236896 Objective Loss 0.236896 LR 0.000250 Time 0.019799 -2023-02-13 18:32:23,274 - Epoch: [168][ 1070/ 1207] Overall Loss 0.236722 Objective Loss 0.236722 LR 0.000250 Time 0.019794 -2023-02-13 18:32:23,465 - Epoch: [168][ 1080/ 1207] Overall Loss 0.236804 Objective Loss 0.236804 LR 0.000250 Time 0.019788 -2023-02-13 18:32:23,659 - Epoch: [168][ 1090/ 1207] Overall Loss 0.236793 Objective Loss 0.236793 LR 0.000250 Time 0.019784 -2023-02-13 18:32:23,850 - Epoch: [168][ 1100/ 1207] Overall Loss 0.236943 Objective Loss 0.236943 LR 0.000250 Time 0.019778 -2023-02-13 18:32:24,040 - Epoch: [168][ 1110/ 1207] Overall Loss 0.237075 Objective Loss 0.237075 LR 0.000250 Time 0.019770 -2023-02-13 18:32:24,229 - Epoch: [168][ 1120/ 1207] Overall Loss 0.237125 Objective Loss 0.237125 LR 0.000250 Time 0.019762 -2023-02-13 18:32:24,419 - Epoch: [168][ 1130/ 1207] Overall Loss 0.237259 Objective Loss 0.237259 LR 0.000250 Time 0.019755 -2023-02-13 18:32:24,608 - Epoch: [168][ 1140/ 1207] Overall Loss 0.237362 Objective Loss 0.237362 LR 0.000250 Time 0.019747 -2023-02-13 18:32:24,798 - Epoch: [168][ 1150/ 1207] Overall Loss 0.237323 Objective Loss 0.237323 LR 0.000250 Time 0.019741 -2023-02-13 18:32:24,988 - Epoch: [168][ 1160/ 1207] Overall Loss 0.237310 Objective Loss 0.237310 LR 0.000250 Time 0.019733 -2023-02-13 18:32:25,177 - Epoch: [168][ 1170/ 1207] Overall Loss 0.237383 Objective Loss 0.237383 LR 0.000250 Time 0.019726 -2023-02-13 18:32:25,367 - Epoch: [168][ 1180/ 1207] Overall Loss 0.237139 Objective Loss 0.237139 LR 0.000250 Time 0.019720 -2023-02-13 18:32:25,556 - Epoch: [168][ 1190/ 1207] Overall Loss 0.237193 Objective Loss 0.237193 LR 0.000250 Time 0.019713 -2023-02-13 18:32:25,799 - Epoch: [168][ 1200/ 1207] Overall Loss 0.237280 Objective Loss 0.237280 LR 0.000250 Time 0.019750 -2023-02-13 18:32:25,915 - Epoch: [168][ 1207/ 1207] Overall Loss 0.237172 Objective Loss 0.237172 Top1 88.109756 Top5 98.780488 LR 0.000250 Time 0.019732 -2023-02-13 18:32:25,992 - --- validate (epoch=168)----------- -2023-02-13 18:32:25,992 - 34311 samples (256 per mini-batch) -2023-02-13 18:32:26,393 - Epoch: [168][ 10/ 135] Loss 0.313814 Top1 85.429688 Top5 97.812500 -2023-02-13 18:32:26,524 - Epoch: [168][ 20/ 135] Loss 0.293227 Top1 85.000000 Top5 97.832031 -2023-02-13 18:32:26,653 - Epoch: [168][ 30/ 135] Loss 0.313231 Top1 84.674479 Top5 97.695312 -2023-02-13 18:32:26,781 - Epoch: [168][ 40/ 135] Loss 0.310555 Top1 84.697266 Top5 97.705078 -2023-02-13 18:32:26,909 - Epoch: [168][ 50/ 135] Loss 0.309646 Top1 84.796875 Top5 97.679688 -2023-02-13 18:32:27,039 - Epoch: [168][ 60/ 135] Loss 0.309723 Top1 84.615885 Top5 97.617188 -2023-02-13 18:32:27,169 - Epoch: [168][ 70/ 135] Loss 0.307812 Top1 84.469866 Top5 97.661830 -2023-02-13 18:32:27,299 - Epoch: [168][ 80/ 135] Loss 0.306782 Top1 84.492188 Top5 97.651367 -2023-02-13 18:32:27,427 - Epoch: [168][ 90/ 135] Loss 0.304148 Top1 84.574653 Top5 97.699653 -2023-02-13 18:32:27,557 - Epoch: [168][ 100/ 135] Loss 0.305723 Top1 84.531250 Top5 97.703125 -2023-02-13 18:32:27,686 - Epoch: [168][ 110/ 135] Loss 0.308639 Top1 84.485085 Top5 97.634943 -2023-02-13 18:32:27,816 - Epoch: [168][ 120/ 135] Loss 0.311789 Top1 84.332682 Top5 97.613932 -2023-02-13 18:32:27,950 - Epoch: [168][ 130/ 135] Loss 0.312254 Top1 84.353966 Top5 97.584135 -2023-02-13 18:32:27,999 - Epoch: [168][ 135/ 135] Loss 0.316589 Top1 84.384017 Top5 97.589694 -2023-02-13 18:32:28,067 - ==> Top1: 84.384 Top5: 97.590 Loss: 0.317 - -2023-02-13 18:32:28,068 - ==> Confusion: -[[ 881 5 8 0 9 4 0 3 2 31 0 2 1 3 4 2 2 2 1 2 5] - [ 3 936 1 1 8 36 2 18 2 1 1 2 2 1 0 1 6 0 3 2 7] - [ 10 6 943 14 3 1 16 20 0 1 3 1 2 6 3 10 3 4 4 1 7] - [ 5 1 11 906 2 6 0 2 2 4 15 1 7 1 15 4 4 5 20 0 5] - [ 15 8 0 0 992 12 1 1 0 2 0 5 0 1 9 6 5 3 0 2 4] - [ 4 10 0 6 4 984 4 12 1 5 1 9 4 9 1 2 4 0 2 4 4] - [ 4 2 9 3 0 7 1036 7 0 1 3 1 3 1 0 4 3 3 2 5 5] - [ 3 7 9 1 0 26 2 940 0 1 0 7 3 1 0 0 1 2 12 7 2] - [ 18 4 1 2 1 0 1 2 916 34 5 3 0 6 9 2 1 0 4 0 0] - [ 90 1 3 0 8 0 0 2 30 850 0 1 0 14 3 2 1 3 1 1 2] - [ 1 4 3 5 0 2 1 6 16 3 985 2 1 7 2 0 1 1 9 0 2] - [ 4 4 0 0 0 11 0 6 1 2 1 921 25 5 0 6 1 11 1 4 2] - [ 0 0 1 8 2 4 0 0 5 0 0 23 886 0 0 4 3 16 0 1 6] - [ 6 2 1 0 5 16 0 1 17 15 11 5 1 918 4 8 4 4 0 0 6] - [ 14 1 0 13 2 7 0 2 23 7 1 2 1 1 995 0 4 5 8 0 6] - [ 2 2 5 1 4 2 1 2 0 0 0 5 3 3 0 977 11 15 1 7 5] - [ 2 5 1 2 8 3 0 2 1 0 0 2 1 3 1 11 1003 5 1 2 8] - [ 4 1 0 4 0 1 1 0 0 0 2 11 9 1 0 17 0 992 0 2 6] - [ 4 3 5 7 1 1 0 32 2 0 7 1 3 0 10 0 2 4 1000 2 2] - [ 2 4 0 1 0 4 6 9 0 0 0 19 3 2 1 6 4 4 0 1079 4] - [ 174 226 210 126 146 245 79 224 109 83 186 113 323 267 159 112 288 104 176 271 9813]] - -2023-02-13 18:32:28,070 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:32:28,070 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:32:28,076 - - -2023-02-13 18:32:28,076 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:32:29,084 - Epoch: [169][ 10/ 1207] Overall Loss 0.221606 Objective Loss 0.221606 LR 0.000250 Time 0.100697 -2023-02-13 18:32:29,277 - Epoch: [169][ 20/ 1207] Overall Loss 0.221496 Objective Loss 0.221496 LR 0.000250 Time 0.059992 -2023-02-13 18:32:29,472 - Epoch: [169][ 30/ 1207] Overall Loss 0.223340 Objective Loss 0.223340 LR 0.000250 Time 0.046495 -2023-02-13 18:32:29,665 - Epoch: [169][ 40/ 1207] Overall Loss 0.232601 Objective Loss 0.232601 LR 0.000250 Time 0.039693 -2023-02-13 18:32:29,861 - Epoch: [169][ 50/ 1207] Overall Loss 0.234439 Objective Loss 0.234439 LR 0.000250 Time 0.035661 -2023-02-13 18:32:30,055 - Epoch: [169][ 60/ 1207] Overall Loss 0.236998 Objective Loss 0.236998 LR 0.000250 Time 0.032939 -2023-02-13 18:32:30,250 - Epoch: [169][ 70/ 1207] Overall Loss 0.238061 Objective Loss 0.238061 LR 0.000250 Time 0.031015 -2023-02-13 18:32:30,443 - Epoch: [169][ 80/ 1207] Overall Loss 0.238464 Objective Loss 0.238464 LR 0.000250 Time 0.029545 -2023-02-13 18:32:30,638 - Epoch: [169][ 90/ 1207] Overall Loss 0.238024 Objective Loss 0.238024 LR 0.000250 Time 0.028427 -2023-02-13 18:32:30,832 - Epoch: [169][ 100/ 1207] Overall Loss 0.238015 Objective Loss 0.238015 LR 0.000250 Time 0.027520 -2023-02-13 18:32:31,027 - Epoch: [169][ 110/ 1207] Overall Loss 0.235654 Objective Loss 0.235654 LR 0.000250 Time 0.026787 -2023-02-13 18:32:31,220 - Epoch: [169][ 120/ 1207] Overall Loss 0.235226 Objective Loss 0.235226 LR 0.000250 Time 0.026162 -2023-02-13 18:32:31,415 - Epoch: [169][ 130/ 1207] Overall Loss 0.235654 Objective Loss 0.235654 LR 0.000250 Time 0.025645 -2023-02-13 18:32:31,608 - Epoch: [169][ 140/ 1207] Overall Loss 0.236285 Objective Loss 0.236285 LR 0.000250 Time 0.025188 -2023-02-13 18:32:31,803 - Epoch: [169][ 150/ 1207] Overall Loss 0.235294 Objective Loss 0.235294 LR 0.000250 Time 0.024811 -2023-02-13 18:32:31,997 - Epoch: [169][ 160/ 1207] Overall Loss 0.234098 Objective Loss 0.234098 LR 0.000250 Time 0.024467 -2023-02-13 18:32:32,192 - Epoch: [169][ 170/ 1207] Overall Loss 0.235556 Objective Loss 0.235556 LR 0.000250 Time 0.024172 -2023-02-13 18:32:32,385 - Epoch: [169][ 180/ 1207] Overall Loss 0.235289 Objective Loss 0.235289 LR 0.000250 Time 0.023903 -2023-02-13 18:32:32,581 - Epoch: [169][ 190/ 1207] Overall Loss 0.234793 Objective Loss 0.234793 LR 0.000250 Time 0.023672 -2023-02-13 18:32:32,776 - Epoch: [169][ 200/ 1207] Overall Loss 0.234979 Objective Loss 0.234979 LR 0.000250 Time 0.023463 -2023-02-13 18:32:32,971 - Epoch: [169][ 210/ 1207] Overall Loss 0.235432 Objective Loss 0.235432 LR 0.000250 Time 0.023273 -2023-02-13 18:32:33,165 - Epoch: [169][ 220/ 1207] Overall Loss 0.235701 Objective Loss 0.235701 LR 0.000250 Time 0.023092 -2023-02-13 18:32:33,360 - Epoch: [169][ 230/ 1207] Overall Loss 0.235961 Objective Loss 0.235961 LR 0.000250 Time 0.022936 -2023-02-13 18:32:33,554 - Epoch: [169][ 240/ 1207] Overall Loss 0.235892 Objective Loss 0.235892 LR 0.000250 Time 0.022787 -2023-02-13 18:32:33,749 - Epoch: [169][ 250/ 1207] Overall Loss 0.235976 Objective Loss 0.235976 LR 0.000250 Time 0.022655 -2023-02-13 18:32:33,943 - Epoch: [169][ 260/ 1207] Overall Loss 0.236514 Objective Loss 0.236514 LR 0.000250 Time 0.022528 -2023-02-13 18:32:34,138 - Epoch: [169][ 270/ 1207] Overall Loss 0.236723 Objective Loss 0.236723 LR 0.000250 Time 0.022416 -2023-02-13 18:32:34,333 - Epoch: [169][ 280/ 1207] Overall Loss 0.236676 Objective Loss 0.236676 LR 0.000250 Time 0.022307 -2023-02-13 18:32:34,528 - Epoch: [169][ 290/ 1207] Overall Loss 0.237252 Objective Loss 0.237252 LR 0.000250 Time 0.022211 -2023-02-13 18:32:34,722 - Epoch: [169][ 300/ 1207] Overall Loss 0.236758 Objective Loss 0.236758 LR 0.000250 Time 0.022116 -2023-02-13 18:32:34,918 - Epoch: [169][ 310/ 1207] Overall Loss 0.237319 Objective Loss 0.237319 LR 0.000250 Time 0.022033 -2023-02-13 18:32:35,112 - Epoch: [169][ 320/ 1207] Overall Loss 0.236715 Objective Loss 0.236715 LR 0.000250 Time 0.021949 -2023-02-13 18:32:35,307 - Epoch: [169][ 330/ 1207] Overall Loss 0.236017 Objective Loss 0.236017 LR 0.000250 Time 0.021874 -2023-02-13 18:32:35,501 - Epoch: [169][ 340/ 1207] Overall Loss 0.235972 Objective Loss 0.235972 LR 0.000250 Time 0.021800 -2023-02-13 18:32:35,697 - Epoch: [169][ 350/ 1207] Overall Loss 0.235433 Objective Loss 0.235433 LR 0.000250 Time 0.021737 -2023-02-13 18:32:35,892 - Epoch: [169][ 360/ 1207] Overall Loss 0.235203 Objective Loss 0.235203 LR 0.000250 Time 0.021673 -2023-02-13 18:32:36,088 - Epoch: [169][ 370/ 1207] Overall Loss 0.235385 Objective Loss 0.235385 LR 0.000250 Time 0.021616 -2023-02-13 18:32:36,281 - Epoch: [169][ 380/ 1207] Overall Loss 0.234427 Objective Loss 0.234427 LR 0.000250 Time 0.021556 -2023-02-13 18:32:36,477 - Epoch: [169][ 390/ 1207] Overall Loss 0.234298 Objective Loss 0.234298 LR 0.000250 Time 0.021504 -2023-02-13 18:32:36,670 - Epoch: [169][ 400/ 1207] Overall Loss 0.234025 Objective Loss 0.234025 LR 0.000250 Time 0.021449 -2023-02-13 18:32:36,866 - Epoch: [169][ 410/ 1207] Overall Loss 0.233796 Objective Loss 0.233796 LR 0.000250 Time 0.021403 -2023-02-13 18:32:37,061 - Epoch: [169][ 420/ 1207] Overall Loss 0.234224 Objective Loss 0.234224 LR 0.000250 Time 0.021357 -2023-02-13 18:32:37,257 - Epoch: [169][ 430/ 1207] Overall Loss 0.234081 Objective Loss 0.234081 LR 0.000250 Time 0.021313 -2023-02-13 18:32:37,453 - Epoch: [169][ 440/ 1207] Overall Loss 0.234371 Objective Loss 0.234371 LR 0.000250 Time 0.021273 -2023-02-13 18:32:37,645 - Epoch: [169][ 450/ 1207] Overall Loss 0.234910 Objective Loss 0.234910 LR 0.000250 Time 0.021228 -2023-02-13 18:32:37,842 - Epoch: [169][ 460/ 1207] Overall Loss 0.234821 Objective Loss 0.234821 LR 0.000250 Time 0.021194 -2023-02-13 18:32:38,035 - Epoch: [169][ 470/ 1207] Overall Loss 0.234569 Objective Loss 0.234569 LR 0.000250 Time 0.021151 -2023-02-13 18:32:38,230 - Epoch: [169][ 480/ 1207] Overall Loss 0.234336 Objective Loss 0.234336 LR 0.000250 Time 0.021118 -2023-02-13 18:32:38,423 - Epoch: [169][ 490/ 1207] Overall Loss 0.234321 Objective Loss 0.234321 LR 0.000250 Time 0.021080 -2023-02-13 18:32:38,619 - Epoch: [169][ 500/ 1207] Overall Loss 0.234627 Objective Loss 0.234627 LR 0.000250 Time 0.021048 -2023-02-13 18:32:38,811 - Epoch: [169][ 510/ 1207] Overall Loss 0.234311 Objective Loss 0.234311 LR 0.000250 Time 0.021012 -2023-02-13 18:32:39,005 - Epoch: [169][ 520/ 1207] Overall Loss 0.234406 Objective Loss 0.234406 LR 0.000250 Time 0.020980 -2023-02-13 18:32:39,199 - Epoch: [169][ 530/ 1207] Overall Loss 0.234105 Objective Loss 0.234105 LR 0.000250 Time 0.020950 -2023-02-13 18:32:39,393 - Epoch: [169][ 540/ 1207] Overall Loss 0.233726 Objective Loss 0.233726 LR 0.000250 Time 0.020921 -2023-02-13 18:32:39,589 - Epoch: [169][ 550/ 1207] Overall Loss 0.233802 Objective Loss 0.233802 LR 0.000250 Time 0.020895 -2023-02-13 18:32:39,783 - Epoch: [169][ 560/ 1207] Overall Loss 0.233660 Objective Loss 0.233660 LR 0.000250 Time 0.020869 -2023-02-13 18:32:39,978 - Epoch: [169][ 570/ 1207] Overall Loss 0.233695 Objective Loss 0.233695 LR 0.000250 Time 0.020844 -2023-02-13 18:32:40,172 - Epoch: [169][ 580/ 1207] Overall Loss 0.233321 Objective Loss 0.233321 LR 0.000250 Time 0.020818 -2023-02-13 18:32:40,367 - Epoch: [169][ 590/ 1207] Overall Loss 0.233824 Objective Loss 0.233824 LR 0.000250 Time 0.020795 -2023-02-13 18:32:40,561 - Epoch: [169][ 600/ 1207] Overall Loss 0.233908 Objective Loss 0.233908 LR 0.000250 Time 0.020771 -2023-02-13 18:32:40,756 - Epoch: [169][ 610/ 1207] Overall Loss 0.233933 Objective Loss 0.233933 LR 0.000250 Time 0.020751 -2023-02-13 18:32:40,951 - Epoch: [169][ 620/ 1207] Overall Loss 0.233915 Objective Loss 0.233915 LR 0.000250 Time 0.020730 -2023-02-13 18:32:41,147 - Epoch: [169][ 630/ 1207] Overall Loss 0.233687 Objective Loss 0.233687 LR 0.000250 Time 0.020711 -2023-02-13 18:32:41,342 - Epoch: [169][ 640/ 1207] Overall Loss 0.234000 Objective Loss 0.234000 LR 0.000250 Time 0.020691 -2023-02-13 18:32:41,537 - Epoch: [169][ 650/ 1207] Overall Loss 0.233966 Objective Loss 0.233966 LR 0.000250 Time 0.020672 -2023-02-13 18:32:41,732 - Epoch: [169][ 660/ 1207] Overall Loss 0.233884 Objective Loss 0.233884 LR 0.000250 Time 0.020654 -2023-02-13 18:32:41,928 - Epoch: [169][ 670/ 1207] Overall Loss 0.233451 Objective Loss 0.233451 LR 0.000250 Time 0.020639 -2023-02-13 18:32:42,122 - Epoch: [169][ 680/ 1207] Overall Loss 0.232969 Objective Loss 0.232969 LR 0.000250 Time 0.020620 -2023-02-13 18:32:42,318 - Epoch: [169][ 690/ 1207] Overall Loss 0.233041 Objective Loss 0.233041 LR 0.000250 Time 0.020604 -2023-02-13 18:32:42,512 - Epoch: [169][ 700/ 1207] Overall Loss 0.232862 Objective Loss 0.232862 LR 0.000250 Time 0.020586 -2023-02-13 18:32:42,707 - Epoch: [169][ 710/ 1207] Overall Loss 0.232817 Objective Loss 0.232817 LR 0.000250 Time 0.020571 -2023-02-13 18:32:42,902 - Epoch: [169][ 720/ 1207] Overall Loss 0.232862 Objective Loss 0.232862 LR 0.000250 Time 0.020555 -2023-02-13 18:32:43,097 - Epoch: [169][ 730/ 1207] Overall Loss 0.232870 Objective Loss 0.232870 LR 0.000250 Time 0.020541 -2023-02-13 18:32:43,291 - Epoch: [169][ 740/ 1207] Overall Loss 0.233351 Objective Loss 0.233351 LR 0.000250 Time 0.020525 -2023-02-13 18:32:43,487 - Epoch: [169][ 750/ 1207] Overall Loss 0.233518 Objective Loss 0.233518 LR 0.000250 Time 0.020511 -2023-02-13 18:32:43,680 - Epoch: [169][ 760/ 1207] Overall Loss 0.233282 Objective Loss 0.233282 LR 0.000250 Time 0.020495 -2023-02-13 18:32:43,877 - Epoch: [169][ 770/ 1207] Overall Loss 0.233335 Objective Loss 0.233335 LR 0.000250 Time 0.020483 -2023-02-13 18:32:44,070 - Epoch: [169][ 780/ 1207] Overall Loss 0.233418 Objective Loss 0.233418 LR 0.000250 Time 0.020468 -2023-02-13 18:32:44,265 - Epoch: [169][ 790/ 1207] Overall Loss 0.233250 Objective Loss 0.233250 LR 0.000250 Time 0.020455 -2023-02-13 18:32:44,455 - Epoch: [169][ 800/ 1207] Overall Loss 0.233486 Objective Loss 0.233486 LR 0.000250 Time 0.020437 -2023-02-13 18:32:44,649 - Epoch: [169][ 810/ 1207] Overall Loss 0.233441 Objective Loss 0.233441 LR 0.000250 Time 0.020424 -2023-02-13 18:32:44,843 - Epoch: [169][ 820/ 1207] Overall Loss 0.233459 Objective Loss 0.233459 LR 0.000250 Time 0.020410 -2023-02-13 18:32:45,036 - Epoch: [169][ 830/ 1207] Overall Loss 0.233735 Objective Loss 0.233735 LR 0.000250 Time 0.020397 -2023-02-13 18:32:45,229 - Epoch: [169][ 840/ 1207] Overall Loss 0.233788 Objective Loss 0.233788 LR 0.000250 Time 0.020383 -2023-02-13 18:32:45,422 - Epoch: [169][ 850/ 1207] Overall Loss 0.233901 Objective Loss 0.233901 LR 0.000250 Time 0.020370 -2023-02-13 18:32:45,613 - Epoch: [169][ 860/ 1207] Overall Loss 0.233881 Objective Loss 0.233881 LR 0.000250 Time 0.020356 -2023-02-13 18:32:45,809 - Epoch: [169][ 870/ 1207] Overall Loss 0.233933 Objective Loss 0.233933 LR 0.000250 Time 0.020346 -2023-02-13 18:32:46,003 - Epoch: [169][ 880/ 1207] Overall Loss 0.233837 Objective Loss 0.233837 LR 0.000250 Time 0.020335 -2023-02-13 18:32:46,197 - Epoch: [169][ 890/ 1207] Overall Loss 0.233733 Objective Loss 0.233733 LR 0.000250 Time 0.020324 -2023-02-13 18:32:46,390 - Epoch: [169][ 900/ 1207] Overall Loss 0.233887 Objective Loss 0.233887 LR 0.000250 Time 0.020312 -2023-02-13 18:32:46,583 - Epoch: [169][ 910/ 1207] Overall Loss 0.234105 Objective Loss 0.234105 LR 0.000250 Time 0.020301 -2023-02-13 18:32:46,777 - Epoch: [169][ 920/ 1207] Overall Loss 0.234125 Objective Loss 0.234125 LR 0.000250 Time 0.020290 -2023-02-13 18:32:46,972 - Epoch: [169][ 930/ 1207] Overall Loss 0.233986 Objective Loss 0.233986 LR 0.000250 Time 0.020281 -2023-02-13 18:32:47,164 - Epoch: [169][ 940/ 1207] Overall Loss 0.234187 Objective Loss 0.234187 LR 0.000250 Time 0.020270 -2023-02-13 18:32:47,358 - Epoch: [169][ 950/ 1207] Overall Loss 0.234402 Objective Loss 0.234402 LR 0.000250 Time 0.020260 -2023-02-13 18:32:47,549 - Epoch: [169][ 960/ 1207] Overall Loss 0.234488 Objective Loss 0.234488 LR 0.000250 Time 0.020248 -2023-02-13 18:32:47,744 - Epoch: [169][ 970/ 1207] Overall Loss 0.234799 Objective Loss 0.234799 LR 0.000250 Time 0.020239 -2023-02-13 18:32:47,937 - Epoch: [169][ 980/ 1207] Overall Loss 0.234511 Objective Loss 0.234511 LR 0.000250 Time 0.020230 -2023-02-13 18:32:48,131 - Epoch: [169][ 990/ 1207] Overall Loss 0.234298 Objective Loss 0.234298 LR 0.000250 Time 0.020221 -2023-02-13 18:32:48,323 - Epoch: [169][ 1000/ 1207] Overall Loss 0.234358 Objective Loss 0.234358 LR 0.000250 Time 0.020211 -2023-02-13 18:32:48,517 - Epoch: [169][ 1010/ 1207] Overall Loss 0.234186 Objective Loss 0.234186 LR 0.000250 Time 0.020202 -2023-02-13 18:32:48,711 - Epoch: [169][ 1020/ 1207] Overall Loss 0.234336 Objective Loss 0.234336 LR 0.000250 Time 0.020193 -2023-02-13 18:32:48,905 - Epoch: [169][ 1030/ 1207] Overall Loss 0.234634 Objective Loss 0.234634 LR 0.000250 Time 0.020186 -2023-02-13 18:32:49,097 - Epoch: [169][ 1040/ 1207] Overall Loss 0.234698 Objective Loss 0.234698 LR 0.000250 Time 0.020176 -2023-02-13 18:32:49,291 - Epoch: [169][ 1050/ 1207] Overall Loss 0.234693 Objective Loss 0.234693 LR 0.000250 Time 0.020168 -2023-02-13 18:32:49,483 - Epoch: [169][ 1060/ 1207] Overall Loss 0.234793 Objective Loss 0.234793 LR 0.000250 Time 0.020159 -2023-02-13 18:32:49,678 - Epoch: [169][ 1070/ 1207] Overall Loss 0.234742 Objective Loss 0.234742 LR 0.000250 Time 0.020152 -2023-02-13 18:32:49,870 - Epoch: [169][ 1080/ 1207] Overall Loss 0.234863 Objective Loss 0.234863 LR 0.000250 Time 0.020143 -2023-02-13 18:32:50,064 - Epoch: [169][ 1090/ 1207] Overall Loss 0.234899 Objective Loss 0.234899 LR 0.000250 Time 0.020136 -2023-02-13 18:32:50,257 - Epoch: [169][ 1100/ 1207] Overall Loss 0.234893 Objective Loss 0.234893 LR 0.000250 Time 0.020128 -2023-02-13 18:32:50,451 - Epoch: [169][ 1110/ 1207] Overall Loss 0.234783 Objective Loss 0.234783 LR 0.000250 Time 0.020121 -2023-02-13 18:32:50,645 - Epoch: [169][ 1120/ 1207] Overall Loss 0.234886 Objective Loss 0.234886 LR 0.000250 Time 0.020114 -2023-02-13 18:32:50,843 - Epoch: [169][ 1130/ 1207] Overall Loss 0.234931 Objective Loss 0.234931 LR 0.000250 Time 0.020111 -2023-02-13 18:32:51,038 - Epoch: [169][ 1140/ 1207] Overall Loss 0.235060 Objective Loss 0.235060 LR 0.000250 Time 0.020106 -2023-02-13 18:32:51,235 - Epoch: [169][ 1150/ 1207] Overall Loss 0.234929 Objective Loss 0.234929 LR 0.000250 Time 0.020102 -2023-02-13 18:32:51,430 - Epoch: [169][ 1160/ 1207] Overall Loss 0.235110 Objective Loss 0.235110 LR 0.000250 Time 0.020096 -2023-02-13 18:32:51,626 - Epoch: [169][ 1170/ 1207] Overall Loss 0.235219 Objective Loss 0.235219 LR 0.000250 Time 0.020092 -2023-02-13 18:32:51,821 - Epoch: [169][ 1180/ 1207] Overall Loss 0.235177 Objective Loss 0.235177 LR 0.000250 Time 0.020086 -2023-02-13 18:32:52,018 - Epoch: [169][ 1190/ 1207] Overall Loss 0.235261 Objective Loss 0.235261 LR 0.000250 Time 0.020083 -2023-02-13 18:32:52,268 - Epoch: [169][ 1200/ 1207] Overall Loss 0.235262 Objective Loss 0.235262 LR 0.000250 Time 0.020123 -2023-02-13 18:32:52,383 - Epoch: [169][ 1207/ 1207] Overall Loss 0.235462 Objective Loss 0.235462 Top1 84.146341 Top5 97.865854 LR 0.000250 Time 0.020101 -2023-02-13 18:32:52,459 - --- validate (epoch=169)----------- -2023-02-13 18:32:52,459 - 34311 samples (256 per mini-batch) -2023-02-13 18:32:52,854 - Epoch: [169][ 10/ 135] Loss 0.317455 Top1 85.625000 Top5 97.304688 -2023-02-13 18:32:52,980 - Epoch: [169][ 20/ 135] Loss 0.305219 Top1 85.039062 Top5 97.539062 -2023-02-13 18:32:53,105 - Epoch: [169][ 30/ 135] Loss 0.304705 Top1 84.895833 Top5 97.486979 -2023-02-13 18:32:53,230 - Epoch: [169][ 40/ 135] Loss 0.294684 Top1 85.097656 Top5 97.724609 -2023-02-13 18:32:53,353 - Epoch: [169][ 50/ 135] Loss 0.295535 Top1 84.968750 Top5 97.718750 -2023-02-13 18:32:53,477 - Epoch: [169][ 60/ 135] Loss 0.301403 Top1 84.824219 Top5 97.649740 -2023-02-13 18:32:53,602 - Epoch: [169][ 70/ 135] Loss 0.299394 Top1 84.899554 Top5 97.639509 -2023-02-13 18:32:53,726 - Epoch: [169][ 80/ 135] Loss 0.299154 Top1 84.936523 Top5 97.729492 -2023-02-13 18:32:53,853 - Epoch: [169][ 90/ 135] Loss 0.303237 Top1 84.839410 Top5 97.695312 -2023-02-13 18:32:53,977 - Epoch: [169][ 100/ 135] Loss 0.304073 Top1 84.765625 Top5 97.640625 -2023-02-13 18:32:54,102 - Epoch: [169][ 110/ 135] Loss 0.307179 Top1 84.701705 Top5 97.592330 -2023-02-13 18:32:54,226 - Epoch: [169][ 120/ 135] Loss 0.306158 Top1 84.710286 Top5 97.620443 -2023-02-13 18:32:54,353 - Epoch: [169][ 130/ 135] Loss 0.304110 Top1 84.768630 Top5 97.635216 -2023-02-13 18:32:54,397 - Epoch: [169][ 135/ 135] Loss 0.302164 Top1 84.800793 Top5 97.642156 -2023-02-13 18:32:54,476 - ==> Top1: 84.801 Top5: 97.642 Loss: 0.302 - -2023-02-13 18:32:54,477 - ==> Confusion: -[[ 856 2 10 1 6 3 0 2 6 51 0 3 0 3 5 2 4 2 2 1 8] - [ 4 954 1 1 8 29 2 12 2 1 1 1 1 0 0 2 3 1 3 1 6] - [ 7 7 960 11 5 2 15 16 1 1 2 2 1 4 5 4 0 4 1 2 8] - [ 5 0 15 909 2 6 1 1 1 2 14 1 8 1 18 3 2 6 15 0 6] - [ 8 7 0 0 1002 11 1 3 2 0 0 5 2 3 5 6 4 2 1 1 3] - [ 3 11 1 4 7 981 3 12 1 3 1 8 3 15 0 1 6 2 2 3 3] - [ 4 4 11 2 0 4 1049 5 0 3 0 1 2 2 0 2 2 3 1 1 3] - [ 2 6 11 0 3 30 2 930 0 2 1 9 2 1 0 0 0 2 11 8 4] - [ 20 2 0 1 1 0 1 1 911 33 8 2 0 11 11 2 1 1 2 0 1] - [ 72 0 4 0 5 1 0 1 30 866 0 0 0 19 4 2 0 2 2 0 4] - [ 3 0 1 4 1 1 5 4 10 2 996 1 1 7 2 0 2 2 5 0 4] - [ 4 2 1 0 1 11 0 6 1 1 0 925 17 6 0 7 2 12 2 6 1] - [ 1 0 1 7 1 4 0 1 1 0 0 25 875 0 5 9 4 16 0 0 9] - [ 5 2 2 0 6 11 1 3 8 14 8 5 2 936 3 3 4 4 0 1 6] - [ 7 2 1 12 4 4 0 2 26 8 3 0 2 2 994 0 2 6 7 0 10] - [ 2 1 6 1 3 4 3 3 0 0 0 7 5 1 1 975 11 9 2 7 5] - [ 2 3 0 1 6 2 0 0 1 0 0 1 1 3 1 8 1016 3 1 4 8] - [ 4 1 0 3 1 1 1 0 0 0 2 6 6 1 1 15 0 1005 0 0 4] - [ 4 5 6 8 0 1 0 24 6 0 8 3 4 0 12 1 0 3 998 3 0] - [ 2 3 1 1 0 5 8 10 1 0 0 18 1 0 1 5 5 5 0 1077 5] - [ 145 250 237 100 148 228 100 189 97 90 209 105 319 311 134 87 290 123 157 234 9881]] - -2023-02-13 18:32:54,478 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:32:54,478 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:32:54,484 - - -2023-02-13 18:32:54,484 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:32:55,377 - Epoch: [170][ 10/ 1207] Overall Loss 0.222430 Objective Loss 0.222430 LR 0.000250 Time 0.089265 -2023-02-13 18:32:55,576 - Epoch: [170][ 20/ 1207] Overall Loss 0.226664 Objective Loss 0.226664 LR 0.000250 Time 0.054539 -2023-02-13 18:32:55,769 - Epoch: [170][ 30/ 1207] Overall Loss 0.234223 Objective Loss 0.234223 LR 0.000250 Time 0.042791 -2023-02-13 18:32:55,968 - Epoch: [170][ 40/ 1207] Overall Loss 0.231464 Objective Loss 0.231464 LR 0.000250 Time 0.037043 -2023-02-13 18:32:56,160 - Epoch: [170][ 50/ 1207] Overall Loss 0.231984 Objective Loss 0.231984 LR 0.000250 Time 0.033481 -2023-02-13 18:32:56,356 - Epoch: [170][ 60/ 1207] Overall Loss 0.229646 Objective Loss 0.229646 LR 0.000250 Time 0.031158 -2023-02-13 18:32:56,549 - Epoch: [170][ 70/ 1207] Overall Loss 0.229131 Objective Loss 0.229131 LR 0.000250 Time 0.029452 -2023-02-13 18:32:56,745 - Epoch: [170][ 80/ 1207] Overall Loss 0.228123 Objective Loss 0.228123 LR 0.000250 Time 0.028221 -2023-02-13 18:32:56,938 - Epoch: [170][ 90/ 1207] Overall Loss 0.229586 Objective Loss 0.229586 LR 0.000250 Time 0.027230 -2023-02-13 18:32:57,135 - Epoch: [170][ 100/ 1207] Overall Loss 0.231087 Objective Loss 0.231087 LR 0.000250 Time 0.026467 -2023-02-13 18:32:57,328 - Epoch: [170][ 110/ 1207] Overall Loss 0.230002 Objective Loss 0.230002 LR 0.000250 Time 0.025811 -2023-02-13 18:32:57,524 - Epoch: [170][ 120/ 1207] Overall Loss 0.230492 Objective Loss 0.230492 LR 0.000250 Time 0.025291 -2023-02-13 18:32:57,716 - Epoch: [170][ 130/ 1207] Overall Loss 0.230585 Objective Loss 0.230585 LR 0.000250 Time 0.024823 -2023-02-13 18:32:57,913 - Epoch: [170][ 140/ 1207] Overall Loss 0.231723 Objective Loss 0.231723 LR 0.000250 Time 0.024449 -2023-02-13 18:32:58,105 - Epoch: [170][ 150/ 1207] Overall Loss 0.231563 Objective Loss 0.231563 LR 0.000250 Time 0.024100 -2023-02-13 18:32:58,301 - Epoch: [170][ 160/ 1207] Overall Loss 0.232651 Objective Loss 0.232651 LR 0.000250 Time 0.023815 -2023-02-13 18:32:58,493 - Epoch: [170][ 170/ 1207] Overall Loss 0.233225 Objective Loss 0.233225 LR 0.000250 Time 0.023542 -2023-02-13 18:32:58,689 - Epoch: [170][ 180/ 1207] Overall Loss 0.233036 Objective Loss 0.233036 LR 0.000250 Time 0.023321 -2023-02-13 18:32:58,882 - Epoch: [170][ 190/ 1207] Overall Loss 0.233206 Objective Loss 0.233206 LR 0.000250 Time 0.023110 -2023-02-13 18:32:59,078 - Epoch: [170][ 200/ 1207] Overall Loss 0.234400 Objective Loss 0.234400 LR 0.000250 Time 0.022930 -2023-02-13 18:32:59,270 - Epoch: [170][ 210/ 1207] Overall Loss 0.234443 Objective Loss 0.234443 LR 0.000250 Time 0.022750 -2023-02-13 18:32:59,466 - Epoch: [170][ 220/ 1207] Overall Loss 0.234467 Objective Loss 0.234467 LR 0.000250 Time 0.022605 -2023-02-13 18:32:59,658 - Epoch: [170][ 230/ 1207] Overall Loss 0.234384 Objective Loss 0.234384 LR 0.000250 Time 0.022458 -2023-02-13 18:32:59,865 - Epoch: [170][ 240/ 1207] Overall Loss 0.235579 Objective Loss 0.235579 LR 0.000250 Time 0.022383 -2023-02-13 18:33:00,072 - Epoch: [170][ 250/ 1207] Overall Loss 0.236571 Objective Loss 0.236571 LR 0.000250 Time 0.022314 -2023-02-13 18:33:00,280 - Epoch: [170][ 260/ 1207] Overall Loss 0.236352 Objective Loss 0.236352 LR 0.000250 Time 0.022255 -2023-02-13 18:33:00,487 - Epoch: [170][ 270/ 1207] Overall Loss 0.236476 Objective Loss 0.236476 LR 0.000250 Time 0.022195 -2023-02-13 18:33:00,693 - Epoch: [170][ 280/ 1207] Overall Loss 0.236990 Objective Loss 0.236990 LR 0.000250 Time 0.022137 -2023-02-13 18:33:00,901 - Epoch: [170][ 290/ 1207] Overall Loss 0.236601 Objective Loss 0.236601 LR 0.000250 Time 0.022088 -2023-02-13 18:33:01,107 - Epoch: [170][ 300/ 1207] Overall Loss 0.236882 Objective Loss 0.236882 LR 0.000250 Time 0.022038 -2023-02-13 18:33:01,314 - Epoch: [170][ 310/ 1207] Overall Loss 0.237059 Objective Loss 0.237059 LR 0.000250 Time 0.021994 -2023-02-13 18:33:01,521 - Epoch: [170][ 320/ 1207] Overall Loss 0.237562 Objective Loss 0.237562 LR 0.000250 Time 0.021951 -2023-02-13 18:33:01,729 - Epoch: [170][ 330/ 1207] Overall Loss 0.237214 Objective Loss 0.237214 LR 0.000250 Time 0.021914 -2023-02-13 18:33:01,936 - Epoch: [170][ 340/ 1207] Overall Loss 0.237155 Objective Loss 0.237155 LR 0.000250 Time 0.021877 -2023-02-13 18:33:02,143 - Epoch: [170][ 350/ 1207] Overall Loss 0.237219 Objective Loss 0.237219 LR 0.000250 Time 0.021843 -2023-02-13 18:33:02,348 - Epoch: [170][ 360/ 1207] Overall Loss 0.236252 Objective Loss 0.236252 LR 0.000250 Time 0.021806 -2023-02-13 18:33:02,557 - Epoch: [170][ 370/ 1207] Overall Loss 0.235592 Objective Loss 0.235592 LR 0.000250 Time 0.021779 -2023-02-13 18:33:02,765 - Epoch: [170][ 380/ 1207] Overall Loss 0.235996 Objective Loss 0.235996 LR 0.000250 Time 0.021753 -2023-02-13 18:33:02,973 - Epoch: [170][ 390/ 1207] Overall Loss 0.235931 Objective Loss 0.235931 LR 0.000250 Time 0.021727 -2023-02-13 18:33:03,180 - Epoch: [170][ 400/ 1207] Overall Loss 0.236201 Objective Loss 0.236201 LR 0.000250 Time 0.021701 -2023-02-13 18:33:03,387 - Epoch: [170][ 410/ 1207] Overall Loss 0.235809 Objective Loss 0.235809 LR 0.000250 Time 0.021676 -2023-02-13 18:33:03,594 - Epoch: [170][ 420/ 1207] Overall Loss 0.235199 Objective Loss 0.235199 LR 0.000250 Time 0.021650 -2023-02-13 18:33:03,801 - Epoch: [170][ 430/ 1207] Overall Loss 0.235456 Objective Loss 0.235456 LR 0.000250 Time 0.021628 -2023-02-13 18:33:04,009 - Epoch: [170][ 440/ 1207] Overall Loss 0.234945 Objective Loss 0.234945 LR 0.000250 Time 0.021607 -2023-02-13 18:33:04,216 - Epoch: [170][ 450/ 1207] Overall Loss 0.235398 Objective Loss 0.235398 LR 0.000250 Time 0.021588 -2023-02-13 18:33:04,422 - Epoch: [170][ 460/ 1207] Overall Loss 0.235127 Objective Loss 0.235127 LR 0.000250 Time 0.021564 -2023-02-13 18:33:04,632 - Epoch: [170][ 470/ 1207] Overall Loss 0.235369 Objective Loss 0.235369 LR 0.000250 Time 0.021552 -2023-02-13 18:33:04,836 - Epoch: [170][ 480/ 1207] Overall Loss 0.235020 Objective Loss 0.235020 LR 0.000250 Time 0.021526 -2023-02-13 18:33:05,042 - Epoch: [170][ 490/ 1207] Overall Loss 0.235066 Objective Loss 0.235066 LR 0.000250 Time 0.021507 -2023-02-13 18:33:05,243 - Epoch: [170][ 500/ 1207] Overall Loss 0.235344 Objective Loss 0.235344 LR 0.000250 Time 0.021478 -2023-02-13 18:33:05,448 - Epoch: [170][ 510/ 1207] Overall Loss 0.235713 Objective Loss 0.235713 LR 0.000250 Time 0.021459 -2023-02-13 18:33:05,650 - Epoch: [170][ 520/ 1207] Overall Loss 0.235608 Objective Loss 0.235608 LR 0.000250 Time 0.021434 -2023-02-13 18:33:05,856 - Epoch: [170][ 530/ 1207] Overall Loss 0.235261 Objective Loss 0.235261 LR 0.000250 Time 0.021417 -2023-02-13 18:33:06,058 - Epoch: [170][ 540/ 1207] Overall Loss 0.235193 Objective Loss 0.235193 LR 0.000250 Time 0.021393 -2023-02-13 18:33:06,263 - Epoch: [170][ 550/ 1207] Overall Loss 0.235377 Objective Loss 0.235377 LR 0.000250 Time 0.021376 -2023-02-13 18:33:06,465 - Epoch: [170][ 560/ 1207] Overall Loss 0.235171 Objective Loss 0.235171 LR 0.000250 Time 0.021354 -2023-02-13 18:33:06,670 - Epoch: [170][ 570/ 1207] Overall Loss 0.235377 Objective Loss 0.235377 LR 0.000250 Time 0.021339 -2023-02-13 18:33:06,872 - Epoch: [170][ 580/ 1207] Overall Loss 0.235453 Objective Loss 0.235453 LR 0.000250 Time 0.021318 -2023-02-13 18:33:07,077 - Epoch: [170][ 590/ 1207] Overall Loss 0.235165 Objective Loss 0.235165 LR 0.000250 Time 0.021304 -2023-02-13 18:33:07,278 - Epoch: [170][ 600/ 1207] Overall Loss 0.235026 Objective Loss 0.235026 LR 0.000250 Time 0.021284 -2023-02-13 18:33:07,483 - Epoch: [170][ 610/ 1207] Overall Loss 0.234928 Objective Loss 0.234928 LR 0.000250 Time 0.021271 -2023-02-13 18:33:07,684 - Epoch: [170][ 620/ 1207] Overall Loss 0.234644 Objective Loss 0.234644 LR 0.000250 Time 0.021251 -2023-02-13 18:33:07,891 - Epoch: [170][ 630/ 1207] Overall Loss 0.234279 Objective Loss 0.234279 LR 0.000250 Time 0.021241 -2023-02-13 18:33:08,091 - Epoch: [170][ 640/ 1207] Overall Loss 0.234319 Objective Loss 0.234319 LR 0.000250 Time 0.021222 -2023-02-13 18:33:08,297 - Epoch: [170][ 650/ 1207] Overall Loss 0.234409 Objective Loss 0.234409 LR 0.000250 Time 0.021211 -2023-02-13 18:33:08,498 - Epoch: [170][ 660/ 1207] Overall Loss 0.234553 Objective Loss 0.234553 LR 0.000250 Time 0.021193 -2023-02-13 18:33:08,704 - Epoch: [170][ 670/ 1207] Overall Loss 0.234584 Objective Loss 0.234584 LR 0.000250 Time 0.021184 -2023-02-13 18:33:08,906 - Epoch: [170][ 680/ 1207] Overall Loss 0.234662 Objective Loss 0.234662 LR 0.000250 Time 0.021168 -2023-02-13 18:33:09,110 - Epoch: [170][ 690/ 1207] Overall Loss 0.234747 Objective Loss 0.234747 LR 0.000250 Time 0.021157 -2023-02-13 18:33:09,311 - Epoch: [170][ 700/ 1207] Overall Loss 0.234873 Objective Loss 0.234873 LR 0.000250 Time 0.021142 -2023-02-13 18:33:09,517 - Epoch: [170][ 710/ 1207] Overall Loss 0.234826 Objective Loss 0.234826 LR 0.000250 Time 0.021133 -2023-02-13 18:33:09,717 - Epoch: [170][ 720/ 1207] Overall Loss 0.234610 Objective Loss 0.234610 LR 0.000250 Time 0.021117 -2023-02-13 18:33:09,924 - Epoch: [170][ 730/ 1207] Overall Loss 0.234426 Objective Loss 0.234426 LR 0.000250 Time 0.021110 -2023-02-13 18:33:10,124 - Epoch: [170][ 740/ 1207] Overall Loss 0.234327 Objective Loss 0.234327 LR 0.000250 Time 0.021095 -2023-02-13 18:33:10,329 - Epoch: [170][ 750/ 1207] Overall Loss 0.233905 Objective Loss 0.233905 LR 0.000250 Time 0.021087 -2023-02-13 18:33:10,530 - Epoch: [170][ 760/ 1207] Overall Loss 0.234041 Objective Loss 0.234041 LR 0.000250 Time 0.021073 -2023-02-13 18:33:10,736 - Epoch: [170][ 770/ 1207] Overall Loss 0.233809 Objective Loss 0.233809 LR 0.000250 Time 0.021067 -2023-02-13 18:33:10,940 - Epoch: [170][ 780/ 1207] Overall Loss 0.233855 Objective Loss 0.233855 LR 0.000250 Time 0.021057 -2023-02-13 18:33:11,145 - Epoch: [170][ 790/ 1207] Overall Loss 0.233889 Objective Loss 0.233889 LR 0.000250 Time 0.021050 -2023-02-13 18:33:11,345 - Epoch: [170][ 800/ 1207] Overall Loss 0.233999 Objective Loss 0.233999 LR 0.000250 Time 0.021036 -2023-02-13 18:33:11,551 - Epoch: [170][ 810/ 1207] Overall Loss 0.234126 Objective Loss 0.234126 LR 0.000250 Time 0.021031 -2023-02-13 18:33:11,752 - Epoch: [170][ 820/ 1207] Overall Loss 0.234118 Objective Loss 0.234118 LR 0.000250 Time 0.021018 -2023-02-13 18:33:11,958 - Epoch: [170][ 830/ 1207] Overall Loss 0.234381 Objective Loss 0.234381 LR 0.000250 Time 0.021013 -2023-02-13 18:33:12,160 - Epoch: [170][ 840/ 1207] Overall Loss 0.234242 Objective Loss 0.234242 LR 0.000250 Time 0.021002 -2023-02-13 18:33:12,365 - Epoch: [170][ 850/ 1207] Overall Loss 0.234748 Objective Loss 0.234748 LR 0.000250 Time 0.020996 -2023-02-13 18:33:12,565 - Epoch: [170][ 860/ 1207] Overall Loss 0.234585 Objective Loss 0.234585 LR 0.000250 Time 0.020985 -2023-02-13 18:33:12,771 - Epoch: [170][ 870/ 1207] Overall Loss 0.234344 Objective Loss 0.234344 LR 0.000250 Time 0.020979 -2023-02-13 18:33:12,971 - Epoch: [170][ 880/ 1207] Overall Loss 0.234214 Objective Loss 0.234214 LR 0.000250 Time 0.020968 -2023-02-13 18:33:13,177 - Epoch: [170][ 890/ 1207] Overall Loss 0.234339 Objective Loss 0.234339 LR 0.000250 Time 0.020963 -2023-02-13 18:33:13,378 - Epoch: [170][ 900/ 1207] Overall Loss 0.234191 Objective Loss 0.234191 LR 0.000250 Time 0.020953 -2023-02-13 18:33:13,583 - Epoch: [170][ 910/ 1207] Overall Loss 0.234145 Objective Loss 0.234145 LR 0.000250 Time 0.020948 -2023-02-13 18:33:13,785 - Epoch: [170][ 920/ 1207] Overall Loss 0.234150 Objective Loss 0.234150 LR 0.000250 Time 0.020939 -2023-02-13 18:33:13,991 - Epoch: [170][ 930/ 1207] Overall Loss 0.234294 Objective Loss 0.234294 LR 0.000250 Time 0.020935 -2023-02-13 18:33:14,191 - Epoch: [170][ 940/ 1207] Overall Loss 0.234049 Objective Loss 0.234049 LR 0.000250 Time 0.020925 -2023-02-13 18:33:14,396 - Epoch: [170][ 950/ 1207] Overall Loss 0.233803 Objective Loss 0.233803 LR 0.000250 Time 0.020919 -2023-02-13 18:33:14,596 - Epoch: [170][ 960/ 1207] Overall Loss 0.233788 Objective Loss 0.233788 LR 0.000250 Time 0.020910 -2023-02-13 18:33:14,801 - Epoch: [170][ 970/ 1207] Overall Loss 0.233867 Objective Loss 0.233867 LR 0.000250 Time 0.020905 -2023-02-13 18:33:15,002 - Epoch: [170][ 980/ 1207] Overall Loss 0.233731 Objective Loss 0.233731 LR 0.000250 Time 0.020897 -2023-02-13 18:33:15,207 - Epoch: [170][ 990/ 1207] Overall Loss 0.233706 Objective Loss 0.233706 LR 0.000250 Time 0.020892 -2023-02-13 18:33:15,407 - Epoch: [170][ 1000/ 1207] Overall Loss 0.233732 Objective Loss 0.233732 LR 0.000250 Time 0.020883 -2023-02-13 18:33:15,612 - Epoch: [170][ 1010/ 1207] Overall Loss 0.233822 Objective Loss 0.233822 LR 0.000250 Time 0.020879 -2023-02-13 18:33:15,816 - Epoch: [170][ 1020/ 1207] Overall Loss 0.233922 Objective Loss 0.233922 LR 0.000250 Time 0.020873 -2023-02-13 18:33:16,027 - Epoch: [170][ 1030/ 1207] Overall Loss 0.233732 Objective Loss 0.233732 LR 0.000250 Time 0.020876 -2023-02-13 18:33:16,232 - Epoch: [170][ 1040/ 1207] Overall Loss 0.233715 Objective Loss 0.233715 LR 0.000250 Time 0.020871 -2023-02-13 18:33:16,441 - Epoch: [170][ 1050/ 1207] Overall Loss 0.233674 Objective Loss 0.233674 LR 0.000250 Time 0.020871 -2023-02-13 18:33:16,646 - Epoch: [170][ 1060/ 1207] Overall Loss 0.233638 Objective Loss 0.233638 LR 0.000250 Time 0.020867 -2023-02-13 18:33:16,853 - Epoch: [170][ 1070/ 1207] Overall Loss 0.233586 Objective Loss 0.233586 LR 0.000250 Time 0.020866 -2023-02-13 18:33:17,058 - Epoch: [170][ 1080/ 1207] Overall Loss 0.233855 Objective Loss 0.233855 LR 0.000250 Time 0.020862 -2023-02-13 18:33:17,266 - Epoch: [170][ 1090/ 1207] Overall Loss 0.233985 Objective Loss 0.233985 LR 0.000250 Time 0.020861 -2023-02-13 18:33:17,472 - Epoch: [170][ 1100/ 1207] Overall Loss 0.234029 Objective Loss 0.234029 LR 0.000250 Time 0.020858 -2023-02-13 18:33:17,680 - Epoch: [170][ 1110/ 1207] Overall Loss 0.234084 Objective Loss 0.234084 LR 0.000250 Time 0.020857 -2023-02-13 18:33:17,885 - Epoch: [170][ 1120/ 1207] Overall Loss 0.233971 Objective Loss 0.233971 LR 0.000250 Time 0.020854 -2023-02-13 18:33:18,094 - Epoch: [170][ 1130/ 1207] Overall Loss 0.233875 Objective Loss 0.233875 LR 0.000250 Time 0.020854 -2023-02-13 18:33:18,299 - Epoch: [170][ 1140/ 1207] Overall Loss 0.233718 Objective Loss 0.233718 LR 0.000250 Time 0.020850 -2023-02-13 18:33:18,507 - Epoch: [170][ 1150/ 1207] Overall Loss 0.233722 Objective Loss 0.233722 LR 0.000250 Time 0.020849 -2023-02-13 18:33:18,712 - Epoch: [170][ 1160/ 1207] Overall Loss 0.233652 Objective Loss 0.233652 LR 0.000250 Time 0.020846 -2023-02-13 18:33:18,920 - Epoch: [170][ 1170/ 1207] Overall Loss 0.233411 Objective Loss 0.233411 LR 0.000250 Time 0.020845 -2023-02-13 18:33:19,123 - Epoch: [170][ 1180/ 1207] Overall Loss 0.233313 Objective Loss 0.233313 LR 0.000250 Time 0.020840 -2023-02-13 18:33:19,330 - Epoch: [170][ 1190/ 1207] Overall Loss 0.233185 Objective Loss 0.233185 LR 0.000250 Time 0.020839 -2023-02-13 18:33:19,587 - Epoch: [170][ 1200/ 1207] Overall Loss 0.233154 Objective Loss 0.233154 LR 0.000250 Time 0.020879 -2023-02-13 18:33:19,704 - Epoch: [170][ 1207/ 1207] Overall Loss 0.233111 Objective Loss 0.233111 Top1 89.329268 Top5 99.085366 LR 0.000250 Time 0.020854 -2023-02-13 18:33:19,777 - --- validate (epoch=170)----------- -2023-02-13 18:33:19,777 - 34311 samples (256 per mini-batch) -2023-02-13 18:33:20,181 - Epoch: [170][ 10/ 135] Loss 0.288324 Top1 84.648438 Top5 97.460938 -2023-02-13 18:33:20,312 - Epoch: [170][ 20/ 135] Loss 0.298303 Top1 84.863281 Top5 97.480469 -2023-02-13 18:33:20,441 - Epoch: [170][ 30/ 135] Loss 0.303401 Top1 84.895833 Top5 97.565104 -2023-02-13 18:33:20,572 - Epoch: [170][ 40/ 135] Loss 0.312391 Top1 84.541016 Top5 97.578125 -2023-02-13 18:33:20,705 - Epoch: [170][ 50/ 135] Loss 0.309897 Top1 84.625000 Top5 97.593750 -2023-02-13 18:33:20,836 - Epoch: [170][ 60/ 135] Loss 0.310379 Top1 84.407552 Top5 97.610677 -2023-02-13 18:33:20,965 - Epoch: [170][ 70/ 135] Loss 0.307477 Top1 84.536830 Top5 97.589286 -2023-02-13 18:33:21,101 - Epoch: [170][ 80/ 135] Loss 0.311406 Top1 84.409180 Top5 97.646484 -2023-02-13 18:33:21,234 - Epoch: [170][ 90/ 135] Loss 0.314363 Top1 84.370660 Top5 97.617188 -2023-02-13 18:33:21,370 - Epoch: [170][ 100/ 135] Loss 0.313921 Top1 84.300781 Top5 97.636719 -2023-02-13 18:33:21,502 - Epoch: [170][ 110/ 135] Loss 0.313741 Top1 84.208097 Top5 97.652699 -2023-02-13 18:33:21,634 - Epoch: [170][ 120/ 135] Loss 0.313407 Top1 84.293620 Top5 97.636719 -2023-02-13 18:33:21,765 - Epoch: [170][ 130/ 135] Loss 0.309149 Top1 84.465144 Top5 97.674279 -2023-02-13 18:33:21,809 - Epoch: [170][ 135/ 135] Loss 0.305425 Top1 84.468538 Top5 97.677130 -2023-02-13 18:33:21,893 - ==> Top1: 84.469 Top5: 97.677 Loss: 0.305 - -2023-02-13 18:33:21,894 - ==> Confusion: -[[ 858 3 7 2 5 2 0 1 5 51 0 3 0 5 6 4 3 2 2 2 6] - [ 4 959 1 2 7 18 1 12 3 0 1 2 2 0 0 1 7 1 5 1 6] - [ 3 5 962 11 4 0 12 11 3 1 1 2 3 7 5 8 2 3 4 3 8] - [ 7 0 21 891 1 4 0 2 4 2 15 0 9 0 20 2 5 5 19 0 9] - [ 11 10 1 0 990 8 1 2 2 0 0 6 2 5 9 8 3 3 1 1 3] - [ 3 22 1 3 5 954 3 18 6 2 2 9 4 15 0 3 6 2 2 4 6] - [ 2 4 17 2 0 4 1037 7 0 1 1 2 2 1 0 4 1 1 2 5 6] - [ 1 16 8 1 2 21 2 932 0 2 1 7 1 1 0 0 1 2 15 7 4] - [ 18 0 1 1 0 0 0 0 929 31 4 2 0 7 8 2 0 0 5 0 1] - [ 75 1 3 0 10 0 0 1 36 860 0 0 1 15 5 0 1 0 2 0 2] - [ 4 0 3 6 2 0 2 6 19 2 984 2 1 7 2 0 2 0 7 0 2] - [ 2 3 2 0 1 4 0 5 3 2 0 935 17 8 1 6 2 8 2 3 1] - [ 0 0 0 6 1 2 0 0 2 0 0 37 860 0 5 13 2 24 0 0 7] - [ 5 2 5 1 5 6 0 2 15 15 7 6 2 936 3 4 4 2 0 0 4] - [ 5 2 1 15 1 3 0 0 28 7 0 1 4 1 1004 0 4 5 3 0 8] - [ 4 0 5 0 4 3 3 2 3 0 0 8 5 3 0 975 11 8 1 6 5] - [ 2 5 1 1 8 0 0 0 2 0 0 1 1 4 2 9 1008 3 1 2 11] - [ 4 1 1 2 0 1 1 0 0 1 2 9 10 0 0 20 0 993 0 1 5] - [ 3 6 4 5 2 2 1 20 4 1 3 2 3 0 12 1 1 3 1011 0 2] - [ 0 4 1 0 0 2 3 13 1 0 0 22 4 3 1 9 9 5 0 1066 5] - [ 163 265 240 107 134 171 80 176 142 87 202 143 288 293 161 111 327 123 181 202 9838]] - -2023-02-13 18:33:21,896 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:33:21,896 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:33:21,901 - - -2023-02-13 18:33:21,902 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:33:22,786 - Epoch: [171][ 10/ 1207] Overall Loss 0.219913 Objective Loss 0.219913 LR 0.000250 Time 0.088394 -2023-02-13 18:33:22,978 - Epoch: [171][ 20/ 1207] Overall Loss 0.218308 Objective Loss 0.218308 LR 0.000250 Time 0.053788 -2023-02-13 18:33:23,166 - Epoch: [171][ 30/ 1207] Overall Loss 0.229609 Objective Loss 0.229609 LR 0.000250 Time 0.042099 -2023-02-13 18:33:23,353 - Epoch: [171][ 40/ 1207] Overall Loss 0.225145 Objective Loss 0.225145 LR 0.000250 Time 0.036233 -2023-02-13 18:33:23,543 - Epoch: [171][ 50/ 1207] Overall Loss 0.223633 Objective Loss 0.223633 LR 0.000250 Time 0.032791 -2023-02-13 18:33:23,732 - Epoch: [171][ 60/ 1207] Overall Loss 0.227280 Objective Loss 0.227280 LR 0.000250 Time 0.030472 -2023-02-13 18:33:23,924 - Epoch: [171][ 70/ 1207] Overall Loss 0.229897 Objective Loss 0.229897 LR 0.000250 Time 0.028844 -2023-02-13 18:33:24,113 - Epoch: [171][ 80/ 1207] Overall Loss 0.230553 Objective Loss 0.230553 LR 0.000250 Time 0.027599 -2023-02-13 18:33:24,303 - Epoch: [171][ 90/ 1207] Overall Loss 0.231763 Objective Loss 0.231763 LR 0.000250 Time 0.026641 -2023-02-13 18:33:24,493 - Epoch: [171][ 100/ 1207] Overall Loss 0.232749 Objective Loss 0.232749 LR 0.000250 Time 0.025871 -2023-02-13 18:33:24,683 - Epoch: [171][ 110/ 1207] Overall Loss 0.234468 Objective Loss 0.234468 LR 0.000250 Time 0.025247 -2023-02-13 18:33:24,873 - Epoch: [171][ 120/ 1207] Overall Loss 0.232870 Objective Loss 0.232870 LR 0.000250 Time 0.024722 -2023-02-13 18:33:25,064 - Epoch: [171][ 130/ 1207] Overall Loss 0.233260 Objective Loss 0.233260 LR 0.000250 Time 0.024289 -2023-02-13 18:33:25,253 - Epoch: [171][ 140/ 1207] Overall Loss 0.232616 Objective Loss 0.232616 LR 0.000250 Time 0.023899 -2023-02-13 18:33:25,444 - Epoch: [171][ 150/ 1207] Overall Loss 0.233156 Objective Loss 0.233156 LR 0.000250 Time 0.023578 -2023-02-13 18:33:25,633 - Epoch: [171][ 160/ 1207] Overall Loss 0.233314 Objective Loss 0.233314 LR 0.000250 Time 0.023284 -2023-02-13 18:33:25,824 - Epoch: [171][ 170/ 1207] Overall Loss 0.232878 Objective Loss 0.232878 LR 0.000250 Time 0.023032 -2023-02-13 18:33:26,014 - Epoch: [171][ 180/ 1207] Overall Loss 0.233298 Objective Loss 0.233298 LR 0.000250 Time 0.022811 -2023-02-13 18:33:26,205 - Epoch: [171][ 190/ 1207] Overall Loss 0.233246 Objective Loss 0.233246 LR 0.000250 Time 0.022609 -2023-02-13 18:33:26,394 - Epoch: [171][ 200/ 1207] Overall Loss 0.233281 Objective Loss 0.233281 LR 0.000250 Time 0.022424 -2023-02-13 18:33:26,585 - Epoch: [171][ 210/ 1207] Overall Loss 0.234312 Objective Loss 0.234312 LR 0.000250 Time 0.022264 -2023-02-13 18:33:26,774 - Epoch: [171][ 220/ 1207] Overall Loss 0.234501 Objective Loss 0.234501 LR 0.000250 Time 0.022112 -2023-02-13 18:33:26,966 - Epoch: [171][ 230/ 1207] Overall Loss 0.234759 Objective Loss 0.234759 LR 0.000250 Time 0.021983 -2023-02-13 18:33:27,156 - Epoch: [171][ 240/ 1207] Overall Loss 0.235509 Objective Loss 0.235509 LR 0.000250 Time 0.021857 -2023-02-13 18:33:27,349 - Epoch: [171][ 250/ 1207] Overall Loss 0.234680 Objective Loss 0.234680 LR 0.000250 Time 0.021750 -2023-02-13 18:33:27,539 - Epoch: [171][ 260/ 1207] Overall Loss 0.234544 Objective Loss 0.234544 LR 0.000250 Time 0.021644 -2023-02-13 18:33:27,730 - Epoch: [171][ 270/ 1207] Overall Loss 0.234276 Objective Loss 0.234276 LR 0.000250 Time 0.021551 -2023-02-13 18:33:27,921 - Epoch: [171][ 280/ 1207] Overall Loss 0.234802 Objective Loss 0.234802 LR 0.000250 Time 0.021460 -2023-02-13 18:33:28,113 - Epoch: [171][ 290/ 1207] Overall Loss 0.234360 Objective Loss 0.234360 LR 0.000250 Time 0.021380 -2023-02-13 18:33:28,302 - Epoch: [171][ 300/ 1207] Overall Loss 0.234143 Objective Loss 0.234143 LR 0.000250 Time 0.021297 -2023-02-13 18:33:28,493 - Epoch: [171][ 310/ 1207] Overall Loss 0.233334 Objective Loss 0.233334 LR 0.000250 Time 0.021224 -2023-02-13 18:33:28,683 - Epoch: [171][ 320/ 1207] Overall Loss 0.233159 Objective Loss 0.233159 LR 0.000250 Time 0.021153 -2023-02-13 18:33:28,873 - Epoch: [171][ 330/ 1207] Overall Loss 0.233445 Objective Loss 0.233445 LR 0.000250 Time 0.021089 -2023-02-13 18:33:29,063 - Epoch: [171][ 340/ 1207] Overall Loss 0.232994 Objective Loss 0.232994 LR 0.000250 Time 0.021027 -2023-02-13 18:33:29,255 - Epoch: [171][ 350/ 1207] Overall Loss 0.232989 Objective Loss 0.232989 LR 0.000250 Time 0.020973 -2023-02-13 18:33:29,444 - Epoch: [171][ 360/ 1207] Overall Loss 0.232304 Objective Loss 0.232304 LR 0.000250 Time 0.020915 -2023-02-13 18:33:29,637 - Epoch: [171][ 370/ 1207] Overall Loss 0.232689 Objective Loss 0.232689 LR 0.000250 Time 0.020868 -2023-02-13 18:33:29,826 - Epoch: [171][ 380/ 1207] Overall Loss 0.232298 Objective Loss 0.232298 LR 0.000250 Time 0.020818 -2023-02-13 18:33:30,018 - Epoch: [171][ 390/ 1207] Overall Loss 0.232126 Objective Loss 0.232126 LR 0.000250 Time 0.020774 -2023-02-13 18:33:30,208 - Epoch: [171][ 400/ 1207] Overall Loss 0.232329 Objective Loss 0.232329 LR 0.000250 Time 0.020730 -2023-02-13 18:33:30,400 - Epoch: [171][ 410/ 1207] Overall Loss 0.232437 Objective Loss 0.232437 LR 0.000250 Time 0.020690 -2023-02-13 18:33:30,590 - Epoch: [171][ 420/ 1207] Overall Loss 0.232148 Objective Loss 0.232148 LR 0.000250 Time 0.020649 -2023-02-13 18:33:30,781 - Epoch: [171][ 430/ 1207] Overall Loss 0.231937 Objective Loss 0.231937 LR 0.000250 Time 0.020613 -2023-02-13 18:33:30,972 - Epoch: [171][ 440/ 1207] Overall Loss 0.231555 Objective Loss 0.231555 LR 0.000250 Time 0.020578 -2023-02-13 18:33:31,164 - Epoch: [171][ 450/ 1207] Overall Loss 0.231654 Objective Loss 0.231654 LR 0.000250 Time 0.020546 -2023-02-13 18:33:31,355 - Epoch: [171][ 460/ 1207] Overall Loss 0.231484 Objective Loss 0.231484 LR 0.000250 Time 0.020514 -2023-02-13 18:33:31,547 - Epoch: [171][ 470/ 1207] Overall Loss 0.231214 Objective Loss 0.231214 LR 0.000250 Time 0.020486 -2023-02-13 18:33:31,739 - Epoch: [171][ 480/ 1207] Overall Loss 0.231197 Objective Loss 0.231197 LR 0.000250 Time 0.020457 -2023-02-13 18:33:31,932 - Epoch: [171][ 490/ 1207] Overall Loss 0.231119 Objective Loss 0.231119 LR 0.000250 Time 0.020433 -2023-02-13 18:33:32,123 - Epoch: [171][ 500/ 1207] Overall Loss 0.230791 Objective Loss 0.230791 LR 0.000250 Time 0.020406 -2023-02-13 18:33:32,315 - Epoch: [171][ 510/ 1207] Overall Loss 0.230665 Objective Loss 0.230665 LR 0.000250 Time 0.020381 -2023-02-13 18:33:32,505 - Epoch: [171][ 520/ 1207] Overall Loss 0.230676 Objective Loss 0.230676 LR 0.000250 Time 0.020354 -2023-02-13 18:33:32,697 - Epoch: [171][ 530/ 1207] Overall Loss 0.230720 Objective Loss 0.230720 LR 0.000250 Time 0.020331 -2023-02-13 18:33:32,887 - Epoch: [171][ 540/ 1207] Overall Loss 0.230740 Objective Loss 0.230740 LR 0.000250 Time 0.020307 -2023-02-13 18:33:33,079 - Epoch: [171][ 550/ 1207] Overall Loss 0.230399 Objective Loss 0.230399 LR 0.000250 Time 0.020286 -2023-02-13 18:33:33,269 - Epoch: [171][ 560/ 1207] Overall Loss 0.230594 Objective Loss 0.230594 LR 0.000250 Time 0.020263 -2023-02-13 18:33:33,461 - Epoch: [171][ 570/ 1207] Overall Loss 0.230968 Objective Loss 0.230968 LR 0.000250 Time 0.020243 -2023-02-13 18:33:33,651 - Epoch: [171][ 580/ 1207] Overall Loss 0.231043 Objective Loss 0.231043 LR 0.000250 Time 0.020221 -2023-02-13 18:33:33,843 - Epoch: [171][ 590/ 1207] Overall Loss 0.231002 Objective Loss 0.231002 LR 0.000250 Time 0.020204 -2023-02-13 18:33:34,034 - Epoch: [171][ 600/ 1207] Overall Loss 0.230810 Objective Loss 0.230810 LR 0.000250 Time 0.020184 -2023-02-13 18:33:34,226 - Epoch: [171][ 610/ 1207] Overall Loss 0.231080 Objective Loss 0.231080 LR 0.000250 Time 0.020167 -2023-02-13 18:33:34,417 - Epoch: [171][ 620/ 1207] Overall Loss 0.231213 Objective Loss 0.231213 LR 0.000250 Time 0.020149 -2023-02-13 18:33:34,608 - Epoch: [171][ 630/ 1207] Overall Loss 0.231251 Objective Loss 0.231251 LR 0.000250 Time 0.020132 -2023-02-13 18:33:34,798 - Epoch: [171][ 640/ 1207] Overall Loss 0.231522 Objective Loss 0.231522 LR 0.000250 Time 0.020114 -2023-02-13 18:33:34,990 - Epoch: [171][ 650/ 1207] Overall Loss 0.231322 Objective Loss 0.231322 LR 0.000250 Time 0.020099 -2023-02-13 18:33:35,181 - Epoch: [171][ 660/ 1207] Overall Loss 0.231201 Objective Loss 0.231201 LR 0.000250 Time 0.020083 -2023-02-13 18:33:35,372 - Epoch: [171][ 670/ 1207] Overall Loss 0.230792 Objective Loss 0.230792 LR 0.000250 Time 0.020068 -2023-02-13 18:33:35,564 - Epoch: [171][ 680/ 1207] Overall Loss 0.231003 Objective Loss 0.231003 LR 0.000250 Time 0.020055 -2023-02-13 18:33:35,755 - Epoch: [171][ 690/ 1207] Overall Loss 0.230561 Objective Loss 0.230561 LR 0.000250 Time 0.020042 -2023-02-13 18:33:35,947 - Epoch: [171][ 700/ 1207] Overall Loss 0.230656 Objective Loss 0.230656 LR 0.000250 Time 0.020029 -2023-02-13 18:33:36,139 - Epoch: [171][ 710/ 1207] Overall Loss 0.230499 Objective Loss 0.230499 LR 0.000250 Time 0.020017 -2023-02-13 18:33:36,330 - Epoch: [171][ 720/ 1207] Overall Loss 0.230521 Objective Loss 0.230521 LR 0.000250 Time 0.020002 -2023-02-13 18:33:36,521 - Epoch: [171][ 730/ 1207] Overall Loss 0.230697 Objective Loss 0.230697 LR 0.000250 Time 0.019990 -2023-02-13 18:33:36,711 - Epoch: [171][ 740/ 1207] Overall Loss 0.230425 Objective Loss 0.230425 LR 0.000250 Time 0.019976 -2023-02-13 18:33:36,903 - Epoch: [171][ 750/ 1207] Overall Loss 0.230608 Objective Loss 0.230608 LR 0.000250 Time 0.019965 -2023-02-13 18:33:37,094 - Epoch: [171][ 760/ 1207] Overall Loss 0.230397 Objective Loss 0.230397 LR 0.000250 Time 0.019954 -2023-02-13 18:33:37,285 - Epoch: [171][ 770/ 1207] Overall Loss 0.230126 Objective Loss 0.230126 LR 0.000250 Time 0.019943 -2023-02-13 18:33:37,476 - Epoch: [171][ 780/ 1207] Overall Loss 0.230128 Objective Loss 0.230128 LR 0.000250 Time 0.019930 -2023-02-13 18:33:37,667 - Epoch: [171][ 790/ 1207] Overall Loss 0.230256 Objective Loss 0.230256 LR 0.000250 Time 0.019920 -2023-02-13 18:33:37,858 - Epoch: [171][ 800/ 1207] Overall Loss 0.230317 Objective Loss 0.230317 LR 0.000250 Time 0.019908 -2023-02-13 18:33:38,050 - Epoch: [171][ 810/ 1207] Overall Loss 0.230455 Objective Loss 0.230455 LR 0.000250 Time 0.019900 -2023-02-13 18:33:38,240 - Epoch: [171][ 820/ 1207] Overall Loss 0.230608 Objective Loss 0.230608 LR 0.000250 Time 0.019888 -2023-02-13 18:33:38,432 - Epoch: [171][ 830/ 1207] Overall Loss 0.230441 Objective Loss 0.230441 LR 0.000250 Time 0.019879 -2023-02-13 18:33:38,622 - Epoch: [171][ 840/ 1207] Overall Loss 0.230316 Objective Loss 0.230316 LR 0.000250 Time 0.019868 -2023-02-13 18:33:38,813 - Epoch: [171][ 850/ 1207] Overall Loss 0.230204 Objective Loss 0.230204 LR 0.000250 Time 0.019859 -2023-02-13 18:33:39,004 - Epoch: [171][ 860/ 1207] Overall Loss 0.230221 Objective Loss 0.230221 LR 0.000250 Time 0.019850 -2023-02-13 18:33:39,196 - Epoch: [171][ 870/ 1207] Overall Loss 0.230243 Objective Loss 0.230243 LR 0.000250 Time 0.019843 -2023-02-13 18:33:39,387 - Epoch: [171][ 880/ 1207] Overall Loss 0.230312 Objective Loss 0.230312 LR 0.000250 Time 0.019833 -2023-02-13 18:33:39,579 - Epoch: [171][ 890/ 1207] Overall Loss 0.230132 Objective Loss 0.230132 LR 0.000250 Time 0.019826 -2023-02-13 18:33:39,770 - Epoch: [171][ 900/ 1207] Overall Loss 0.230230 Objective Loss 0.230230 LR 0.000250 Time 0.019817 -2023-02-13 18:33:39,961 - Epoch: [171][ 910/ 1207] Overall Loss 0.230294 Objective Loss 0.230294 LR 0.000250 Time 0.019809 -2023-02-13 18:33:40,152 - Epoch: [171][ 920/ 1207] Overall Loss 0.230613 Objective Loss 0.230613 LR 0.000250 Time 0.019801 -2023-02-13 18:33:40,344 - Epoch: [171][ 930/ 1207] Overall Loss 0.230539 Objective Loss 0.230539 LR 0.000250 Time 0.019794 -2023-02-13 18:33:40,534 - Epoch: [171][ 940/ 1207] Overall Loss 0.230371 Objective Loss 0.230371 LR 0.000250 Time 0.019785 -2023-02-13 18:33:40,725 - Epoch: [171][ 950/ 1207] Overall Loss 0.230382 Objective Loss 0.230382 LR 0.000250 Time 0.019778 -2023-02-13 18:33:40,916 - Epoch: [171][ 960/ 1207] Overall Loss 0.230169 Objective Loss 0.230169 LR 0.000250 Time 0.019770 -2023-02-13 18:33:41,109 - Epoch: [171][ 970/ 1207] Overall Loss 0.230326 Objective Loss 0.230326 LR 0.000250 Time 0.019765 -2023-02-13 18:33:41,299 - Epoch: [171][ 980/ 1207] Overall Loss 0.230279 Objective Loss 0.230279 LR 0.000250 Time 0.019757 -2023-02-13 18:33:41,491 - Epoch: [171][ 990/ 1207] Overall Loss 0.230269 Objective Loss 0.230269 LR 0.000250 Time 0.019751 -2023-02-13 18:33:41,682 - Epoch: [171][ 1000/ 1207] Overall Loss 0.230538 Objective Loss 0.230538 LR 0.000250 Time 0.019744 -2023-02-13 18:33:41,874 - Epoch: [171][ 1010/ 1207] Overall Loss 0.230471 Objective Loss 0.230471 LR 0.000250 Time 0.019739 -2023-02-13 18:33:42,066 - Epoch: [171][ 1020/ 1207] Overall Loss 0.230686 Objective Loss 0.230686 LR 0.000250 Time 0.019732 -2023-02-13 18:33:42,257 - Epoch: [171][ 1030/ 1207] Overall Loss 0.230706 Objective Loss 0.230706 LR 0.000250 Time 0.019726 -2023-02-13 18:33:42,447 - Epoch: [171][ 1040/ 1207] Overall Loss 0.230929 Objective Loss 0.230929 LR 0.000250 Time 0.019719 -2023-02-13 18:33:42,639 - Epoch: [171][ 1050/ 1207] Overall Loss 0.230775 Objective Loss 0.230775 LR 0.000250 Time 0.019713 -2023-02-13 18:33:42,830 - Epoch: [171][ 1060/ 1207] Overall Loss 0.230871 Objective Loss 0.230871 LR 0.000250 Time 0.019707 -2023-02-13 18:33:43,022 - Epoch: [171][ 1070/ 1207] Overall Loss 0.230841 Objective Loss 0.230841 LR 0.000250 Time 0.019702 -2023-02-13 18:33:43,213 - Epoch: [171][ 1080/ 1207] Overall Loss 0.230814 Objective Loss 0.230814 LR 0.000250 Time 0.019696 -2023-02-13 18:33:43,405 - Epoch: [171][ 1090/ 1207] Overall Loss 0.230771 Objective Loss 0.230771 LR 0.000250 Time 0.019692 -2023-02-13 18:33:43,596 - Epoch: [171][ 1100/ 1207] Overall Loss 0.230541 Objective Loss 0.230541 LR 0.000250 Time 0.019686 -2023-02-13 18:33:43,788 - Epoch: [171][ 1110/ 1207] Overall Loss 0.230426 Objective Loss 0.230426 LR 0.000250 Time 0.019681 -2023-02-13 18:33:43,978 - Epoch: [171][ 1120/ 1207] Overall Loss 0.230644 Objective Loss 0.230644 LR 0.000250 Time 0.019675 -2023-02-13 18:33:44,170 - Epoch: [171][ 1130/ 1207] Overall Loss 0.230639 Objective Loss 0.230639 LR 0.000250 Time 0.019670 -2023-02-13 18:33:44,361 - Epoch: [171][ 1140/ 1207] Overall Loss 0.230671 Objective Loss 0.230671 LR 0.000250 Time 0.019665 -2023-02-13 18:33:44,552 - Epoch: [171][ 1150/ 1207] Overall Loss 0.230511 Objective Loss 0.230511 LR 0.000250 Time 0.019660 -2023-02-13 18:33:44,743 - Epoch: [171][ 1160/ 1207] Overall Loss 0.230575 Objective Loss 0.230575 LR 0.000250 Time 0.019654 -2023-02-13 18:33:44,935 - Epoch: [171][ 1170/ 1207] Overall Loss 0.230619 Objective Loss 0.230619 LR 0.000250 Time 0.019650 -2023-02-13 18:33:45,126 - Epoch: [171][ 1180/ 1207] Overall Loss 0.230698 Objective Loss 0.230698 LR 0.000250 Time 0.019645 -2023-02-13 18:33:45,318 - Epoch: [171][ 1190/ 1207] Overall Loss 0.230674 Objective Loss 0.230674 LR 0.000250 Time 0.019641 -2023-02-13 18:33:45,559 - Epoch: [171][ 1200/ 1207] Overall Loss 0.230762 Objective Loss 0.230762 LR 0.000250 Time 0.019678 -2023-02-13 18:33:45,673 - Epoch: [171][ 1207/ 1207] Overall Loss 0.230706 Objective Loss 0.230706 Top1 84.756098 Top5 98.170732 LR 0.000250 Time 0.019658 -2023-02-13 18:33:45,745 - --- validate (epoch=171)----------- -2023-02-13 18:33:45,745 - 34311 samples (256 per mini-batch) -2023-02-13 18:33:46,154 - Epoch: [171][ 10/ 135] Loss 0.319408 Top1 84.257812 Top5 97.851562 -2023-02-13 18:33:46,282 - Epoch: [171][ 20/ 135] Loss 0.333047 Top1 83.867188 Top5 97.714844 -2023-02-13 18:33:46,409 - Epoch: [171][ 30/ 135] Loss 0.332440 Top1 83.776042 Top5 97.617188 -2023-02-13 18:33:46,533 - Epoch: [171][ 40/ 135] Loss 0.324106 Top1 84.208984 Top5 97.607422 -2023-02-13 18:33:46,661 - Epoch: [171][ 50/ 135] Loss 0.316009 Top1 84.312500 Top5 97.640625 -2023-02-13 18:33:46,786 - Epoch: [171][ 60/ 135] Loss 0.313149 Top1 84.492188 Top5 97.714844 -2023-02-13 18:33:46,914 - Epoch: [171][ 70/ 135] Loss 0.308938 Top1 84.681920 Top5 97.762277 -2023-02-13 18:33:47,044 - Epoch: [171][ 80/ 135] Loss 0.306689 Top1 84.560547 Top5 97.822266 -2023-02-13 18:33:47,176 - Epoch: [171][ 90/ 135] Loss 0.306639 Top1 84.500868 Top5 97.769097 -2023-02-13 18:33:47,307 - Epoch: [171][ 100/ 135] Loss 0.307787 Top1 84.398438 Top5 97.773438 -2023-02-13 18:33:47,438 - Epoch: [171][ 110/ 135] Loss 0.305928 Top1 84.375000 Top5 97.780540 -2023-02-13 18:33:47,570 - Epoch: [171][ 120/ 135] Loss 0.306699 Top1 84.427083 Top5 97.783203 -2023-02-13 18:33:47,703 - Epoch: [171][ 130/ 135] Loss 0.308805 Top1 84.435096 Top5 97.773438 -2023-02-13 18:33:47,750 - Epoch: [171][ 135/ 135] Loss 0.311845 Top1 84.389846 Top5 97.761651 -2023-02-13 18:33:47,824 - ==> Top1: 84.390 Top5: 97.762 Loss: 0.312 - -2023-02-13 18:33:47,825 - ==> Confusion: -[[ 879 4 5 3 10 3 0 1 4 29 2 2 0 3 6 1 2 3 1 3 6] - [ 3 948 1 2 8 25 2 18 3 0 1 2 0 0 0 2 3 0 5 3 7] - [ 5 4 968 9 6 1 12 17 0 1 3 4 2 5 3 6 2 0 4 3 3] - [ 6 0 23 907 1 6 1 2 0 3 10 0 5 0 21 2 3 2 18 0 6] - [ 10 10 1 0 999 9 1 2 2 0 0 2 4 1 5 5 5 2 2 2 4] - [ 1 10 0 5 9 978 3 17 2 4 1 7 2 12 2 1 6 1 3 2 4] - [ 3 2 14 2 0 8 1046 4 0 1 0 2 1 1 0 1 2 3 2 3 4] - [ 2 9 6 3 2 27 1 943 2 2 0 6 2 0 0 0 0 1 11 5 2] - [ 21 1 1 1 1 0 1 1 907 27 5 3 0 7 23 1 3 0 5 0 1] - [ 102 0 3 0 8 1 0 1 41 827 0 0 0 14 5 2 0 1 1 2 4] - [ 1 3 6 5 1 5 3 7 14 2 976 3 0 5 3 0 1 1 11 0 4] - [ 2 3 2 0 0 9 1 8 4 1 1 916 23 4 2 5 2 9 1 10 2] - [ 0 0 1 9 0 4 0 2 2 0 0 26 876 0 5 10 4 11 1 1 7] - [ 5 3 3 0 8 17 0 2 19 14 11 5 2 913 7 3 4 1 0 1 6] - [ 10 2 1 12 3 4 0 1 19 6 1 1 4 1 1005 0 3 4 8 0 7] - [ 5 3 8 1 4 2 5 0 0 0 0 7 5 3 0 968 10 7 2 7 9] - [ 5 6 2 1 5 1 0 2 0 1 0 1 1 1 2 10 1007 1 1 3 11] - [ 4 2 1 3 1 1 4 0 0 0 2 14 15 2 0 17 0 977 0 0 8] - [ 5 3 5 5 1 1 0 29 3 0 3 1 2 0 14 0 0 2 1009 2 1] - [ 0 3 0 1 1 10 9 12 2 1 0 14 1 2 1 5 4 2 0 1073 7] - [ 178 246 255 114 156 237 90 222 91 94 165 100 299 272 195 96 272 84 195 240 9833]] - -2023-02-13 18:33:47,826 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:33:47,826 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:33:47,832 - - -2023-02-13 18:33:47,832 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:33:48,829 - Epoch: [172][ 10/ 1207] Overall Loss 0.227794 Objective Loss 0.227794 LR 0.000250 Time 0.099654 -2023-02-13 18:33:49,027 - Epoch: [172][ 20/ 1207] Overall Loss 0.220521 Objective Loss 0.220521 LR 0.000250 Time 0.059660 -2023-02-13 18:33:49,216 - Epoch: [172][ 30/ 1207] Overall Loss 0.224212 Objective Loss 0.224212 LR 0.000250 Time 0.046068 -2023-02-13 18:33:49,407 - Epoch: [172][ 40/ 1207] Overall Loss 0.227493 Objective Loss 0.227493 LR 0.000250 Time 0.039316 -2023-02-13 18:33:49,597 - Epoch: [172][ 50/ 1207] Overall Loss 0.225308 Objective Loss 0.225308 LR 0.000250 Time 0.035244 -2023-02-13 18:33:49,787 - Epoch: [172][ 60/ 1207] Overall Loss 0.225635 Objective Loss 0.225635 LR 0.000250 Time 0.032543 -2023-02-13 18:33:49,977 - Epoch: [172][ 70/ 1207] Overall Loss 0.223302 Objective Loss 0.223302 LR 0.000250 Time 0.030604 -2023-02-13 18:33:50,169 - Epoch: [172][ 80/ 1207] Overall Loss 0.225827 Objective Loss 0.225827 LR 0.000250 Time 0.029173 -2023-02-13 18:33:50,359 - Epoch: [172][ 90/ 1207] Overall Loss 0.225671 Objective Loss 0.225671 LR 0.000250 Time 0.028035 -2023-02-13 18:33:50,550 - Epoch: [172][ 100/ 1207] Overall Loss 0.226909 Objective Loss 0.226909 LR 0.000250 Time 0.027134 -2023-02-13 18:33:50,741 - Epoch: [172][ 110/ 1207] Overall Loss 0.229401 Objective Loss 0.229401 LR 0.000250 Time 0.026399 -2023-02-13 18:33:50,933 - Epoch: [172][ 120/ 1207] Overall Loss 0.230479 Objective Loss 0.230479 LR 0.000250 Time 0.025797 -2023-02-13 18:33:51,123 - Epoch: [172][ 130/ 1207] Overall Loss 0.231090 Objective Loss 0.231090 LR 0.000250 Time 0.025277 -2023-02-13 18:33:51,313 - Epoch: [172][ 140/ 1207] Overall Loss 0.232673 Objective Loss 0.232673 LR 0.000250 Time 0.024823 -2023-02-13 18:33:51,502 - Epoch: [172][ 150/ 1207] Overall Loss 0.232769 Objective Loss 0.232769 LR 0.000250 Time 0.024428 -2023-02-13 18:33:51,691 - Epoch: [172][ 160/ 1207] Overall Loss 0.232813 Objective Loss 0.232813 LR 0.000250 Time 0.024080 -2023-02-13 18:33:51,881 - Epoch: [172][ 170/ 1207] Overall Loss 0.233559 Objective Loss 0.233559 LR 0.000250 Time 0.023777 -2023-02-13 18:33:52,071 - Epoch: [172][ 180/ 1207] Overall Loss 0.234053 Objective Loss 0.234053 LR 0.000250 Time 0.023512 -2023-02-13 18:33:52,261 - Epoch: [172][ 190/ 1207] Overall Loss 0.233961 Objective Loss 0.233961 LR 0.000250 Time 0.023268 -2023-02-13 18:33:52,449 - Epoch: [172][ 200/ 1207] Overall Loss 0.234013 Objective Loss 0.234013 LR 0.000250 Time 0.023045 -2023-02-13 18:33:52,638 - Epoch: [172][ 210/ 1207] Overall Loss 0.232980 Objective Loss 0.232980 LR 0.000250 Time 0.022847 -2023-02-13 18:33:52,828 - Epoch: [172][ 220/ 1207] Overall Loss 0.232180 Objective Loss 0.232180 LR 0.000250 Time 0.022669 -2023-02-13 18:33:53,018 - Epoch: [172][ 230/ 1207] Overall Loss 0.231872 Objective Loss 0.231872 LR 0.000250 Time 0.022506 -2023-02-13 18:33:53,207 - Epoch: [172][ 240/ 1207] Overall Loss 0.232055 Objective Loss 0.232055 LR 0.000250 Time 0.022356 -2023-02-13 18:33:53,396 - Epoch: [172][ 250/ 1207] Overall Loss 0.232348 Objective Loss 0.232348 LR 0.000250 Time 0.022218 -2023-02-13 18:33:53,586 - Epoch: [172][ 260/ 1207] Overall Loss 0.232136 Objective Loss 0.232136 LR 0.000250 Time 0.022092 -2023-02-13 18:33:53,775 - Epoch: [172][ 270/ 1207] Overall Loss 0.231430 Objective Loss 0.231430 LR 0.000250 Time 0.021973 -2023-02-13 18:33:53,966 - Epoch: [172][ 280/ 1207] Overall Loss 0.231894 Objective Loss 0.231894 LR 0.000250 Time 0.021867 -2023-02-13 18:33:54,155 - Epoch: [172][ 290/ 1207] Overall Loss 0.232065 Objective Loss 0.232065 LR 0.000250 Time 0.021765 -2023-02-13 18:33:54,351 - Epoch: [172][ 300/ 1207] Overall Loss 0.232711 Objective Loss 0.232711 LR 0.000250 Time 0.021690 -2023-02-13 18:33:54,540 - Epoch: [172][ 310/ 1207] Overall Loss 0.231651 Objective Loss 0.231651 LR 0.000250 Time 0.021599 -2023-02-13 18:33:54,729 - Epoch: [172][ 320/ 1207] Overall Loss 0.230854 Objective Loss 0.230854 LR 0.000250 Time 0.021514 -2023-02-13 18:33:54,919 - Epoch: [172][ 330/ 1207] Overall Loss 0.230835 Objective Loss 0.230835 LR 0.000250 Time 0.021436 -2023-02-13 18:33:55,109 - Epoch: [172][ 340/ 1207] Overall Loss 0.230078 Objective Loss 0.230078 LR 0.000250 Time 0.021364 -2023-02-13 18:33:55,298 - Epoch: [172][ 350/ 1207] Overall Loss 0.230100 Objective Loss 0.230100 LR 0.000250 Time 0.021294 -2023-02-13 18:33:55,487 - Epoch: [172][ 360/ 1207] Overall Loss 0.229839 Objective Loss 0.229839 LR 0.000250 Time 0.021226 -2023-02-13 18:33:55,677 - Epoch: [172][ 370/ 1207] Overall Loss 0.229802 Objective Loss 0.229802 LR 0.000250 Time 0.021164 -2023-02-13 18:33:55,866 - Epoch: [172][ 380/ 1207] Overall Loss 0.229863 Objective Loss 0.229863 LR 0.000250 Time 0.021104 -2023-02-13 18:33:56,060 - Epoch: [172][ 390/ 1207] Overall Loss 0.229801 Objective Loss 0.229801 LR 0.000250 Time 0.021059 -2023-02-13 18:33:56,251 - Epoch: [172][ 400/ 1207] Overall Loss 0.228966 Objective Loss 0.228966 LR 0.000250 Time 0.021009 -2023-02-13 18:33:56,442 - Epoch: [172][ 410/ 1207] Overall Loss 0.228904 Objective Loss 0.228904 LR 0.000250 Time 0.020962 -2023-02-13 18:33:56,633 - Epoch: [172][ 420/ 1207] Overall Loss 0.229198 Objective Loss 0.229198 LR 0.000250 Time 0.020917 -2023-02-13 18:33:56,825 - Epoch: [172][ 430/ 1207] Overall Loss 0.229368 Objective Loss 0.229368 LR 0.000250 Time 0.020876 -2023-02-13 18:33:57,017 - Epoch: [172][ 440/ 1207] Overall Loss 0.229176 Objective Loss 0.229176 LR 0.000250 Time 0.020836 -2023-02-13 18:33:57,207 - Epoch: [172][ 450/ 1207] Overall Loss 0.228709 Objective Loss 0.228709 LR 0.000250 Time 0.020795 -2023-02-13 18:33:57,397 - Epoch: [172][ 460/ 1207] Overall Loss 0.228765 Objective Loss 0.228765 LR 0.000250 Time 0.020754 -2023-02-13 18:33:57,586 - Epoch: [172][ 470/ 1207] Overall Loss 0.228731 Objective Loss 0.228731 LR 0.000250 Time 0.020715 -2023-02-13 18:33:57,775 - Epoch: [172][ 480/ 1207] Overall Loss 0.228508 Objective Loss 0.228508 LR 0.000250 Time 0.020677 -2023-02-13 18:33:57,965 - Epoch: [172][ 490/ 1207] Overall Loss 0.228548 Objective Loss 0.228548 LR 0.000250 Time 0.020641 -2023-02-13 18:33:58,155 - Epoch: [172][ 500/ 1207] Overall Loss 0.228357 Objective Loss 0.228357 LR 0.000250 Time 0.020608 -2023-02-13 18:33:58,345 - Epoch: [172][ 510/ 1207] Overall Loss 0.228478 Objective Loss 0.228478 LR 0.000250 Time 0.020575 -2023-02-13 18:33:58,533 - Epoch: [172][ 520/ 1207] Overall Loss 0.228540 Objective Loss 0.228540 LR 0.000250 Time 0.020541 -2023-02-13 18:33:58,723 - Epoch: [172][ 530/ 1207] Overall Loss 0.228716 Objective Loss 0.228716 LR 0.000250 Time 0.020510 -2023-02-13 18:33:58,911 - Epoch: [172][ 540/ 1207] Overall Loss 0.228989 Objective Loss 0.228989 LR 0.000250 Time 0.020479 -2023-02-13 18:33:59,102 - Epoch: [172][ 550/ 1207] Overall Loss 0.228981 Objective Loss 0.228981 LR 0.000250 Time 0.020452 -2023-02-13 18:33:59,291 - Epoch: [172][ 560/ 1207] Overall Loss 0.228824 Objective Loss 0.228824 LR 0.000250 Time 0.020423 -2023-02-13 18:33:59,480 - Epoch: [172][ 570/ 1207] Overall Loss 0.229024 Objective Loss 0.229024 LR 0.000250 Time 0.020396 -2023-02-13 18:33:59,669 - Epoch: [172][ 580/ 1207] Overall Loss 0.229014 Objective Loss 0.229014 LR 0.000250 Time 0.020370 -2023-02-13 18:33:59,861 - Epoch: [172][ 590/ 1207] Overall Loss 0.228888 Objective Loss 0.228888 LR 0.000250 Time 0.020351 -2023-02-13 18:34:00,055 - Epoch: [172][ 600/ 1207] Overall Loss 0.228859 Objective Loss 0.228859 LR 0.000250 Time 0.020334 -2023-02-13 18:34:00,253 - Epoch: [172][ 610/ 1207] Overall Loss 0.228970 Objective Loss 0.228970 LR 0.000250 Time 0.020324 -2023-02-13 18:34:00,446 - Epoch: [172][ 620/ 1207] Overall Loss 0.229006 Objective Loss 0.229006 LR 0.000250 Time 0.020308 -2023-02-13 18:34:00,643 - Epoch: [172][ 630/ 1207] Overall Loss 0.228992 Objective Loss 0.228992 LR 0.000250 Time 0.020298 -2023-02-13 18:34:00,837 - Epoch: [172][ 640/ 1207] Overall Loss 0.229071 Objective Loss 0.229071 LR 0.000250 Time 0.020283 -2023-02-13 18:34:01,036 - Epoch: [172][ 650/ 1207] Overall Loss 0.228946 Objective Loss 0.228946 LR 0.000250 Time 0.020275 -2023-02-13 18:34:01,229 - Epoch: [172][ 660/ 1207] Overall Loss 0.229297 Objective Loss 0.229297 LR 0.000250 Time 0.020261 -2023-02-13 18:34:01,426 - Epoch: [172][ 670/ 1207] Overall Loss 0.229167 Objective Loss 0.229167 LR 0.000250 Time 0.020252 -2023-02-13 18:34:01,620 - Epoch: [172][ 680/ 1207] Overall Loss 0.228944 Objective Loss 0.228944 LR 0.000250 Time 0.020239 -2023-02-13 18:34:01,817 - Epoch: [172][ 690/ 1207] Overall Loss 0.228806 Objective Loss 0.228806 LR 0.000250 Time 0.020230 -2023-02-13 18:34:02,012 - Epoch: [172][ 700/ 1207] Overall Loss 0.228984 Objective Loss 0.228984 LR 0.000250 Time 0.020218 -2023-02-13 18:34:02,209 - Epoch: [172][ 710/ 1207] Overall Loss 0.228908 Objective Loss 0.228908 LR 0.000250 Time 0.020212 -2023-02-13 18:34:02,403 - Epoch: [172][ 720/ 1207] Overall Loss 0.229025 Objective Loss 0.229025 LR 0.000250 Time 0.020199 -2023-02-13 18:34:02,600 - Epoch: [172][ 730/ 1207] Overall Loss 0.229409 Objective Loss 0.229409 LR 0.000250 Time 0.020191 -2023-02-13 18:34:02,794 - Epoch: [172][ 740/ 1207] Overall Loss 0.229518 Objective Loss 0.229518 LR 0.000250 Time 0.020180 -2023-02-13 18:34:02,991 - Epoch: [172][ 750/ 1207] Overall Loss 0.229305 Objective Loss 0.229305 LR 0.000250 Time 0.020174 -2023-02-13 18:34:03,185 - Epoch: [172][ 760/ 1207] Overall Loss 0.229004 Objective Loss 0.229004 LR 0.000250 Time 0.020163 -2023-02-13 18:34:03,383 - Epoch: [172][ 770/ 1207] Overall Loss 0.228728 Objective Loss 0.228728 LR 0.000250 Time 0.020157 -2023-02-13 18:34:03,576 - Epoch: [172][ 780/ 1207] Overall Loss 0.228755 Objective Loss 0.228755 LR 0.000250 Time 0.020146 -2023-02-13 18:34:03,771 - Epoch: [172][ 790/ 1207] Overall Loss 0.229193 Objective Loss 0.229193 LR 0.000250 Time 0.020137 -2023-02-13 18:34:03,965 - Epoch: [172][ 800/ 1207] Overall Loss 0.228958 Objective Loss 0.228958 LR 0.000250 Time 0.020128 -2023-02-13 18:34:04,159 - Epoch: [172][ 810/ 1207] Overall Loss 0.228961 Objective Loss 0.228961 LR 0.000250 Time 0.020118 -2023-02-13 18:34:04,353 - Epoch: [172][ 820/ 1207] Overall Loss 0.229067 Objective Loss 0.229067 LR 0.000250 Time 0.020109 -2023-02-13 18:34:04,546 - Epoch: [172][ 830/ 1207] Overall Loss 0.229249 Objective Loss 0.229249 LR 0.000250 Time 0.020099 -2023-02-13 18:34:04,740 - Epoch: [172][ 840/ 1207] Overall Loss 0.229332 Objective Loss 0.229332 LR 0.000250 Time 0.020090 -2023-02-13 18:34:04,933 - Epoch: [172][ 850/ 1207] Overall Loss 0.229631 Objective Loss 0.229631 LR 0.000250 Time 0.020080 -2023-02-13 18:34:05,128 - Epoch: [172][ 860/ 1207] Overall Loss 0.229534 Objective Loss 0.229534 LR 0.000250 Time 0.020074 -2023-02-13 18:34:05,321 - Epoch: [172][ 870/ 1207] Overall Loss 0.229211 Objective Loss 0.229211 LR 0.000250 Time 0.020065 -2023-02-13 18:34:05,516 - Epoch: [172][ 880/ 1207] Overall Loss 0.229275 Objective Loss 0.229275 LR 0.000250 Time 0.020057 -2023-02-13 18:34:05,709 - Epoch: [172][ 890/ 1207] Overall Loss 0.229587 Objective Loss 0.229587 LR 0.000250 Time 0.020048 -2023-02-13 18:34:05,904 - Epoch: [172][ 900/ 1207] Overall Loss 0.229852 Objective Loss 0.229852 LR 0.000250 Time 0.020042 -2023-02-13 18:34:06,099 - Epoch: [172][ 910/ 1207] Overall Loss 0.229716 Objective Loss 0.229716 LR 0.000250 Time 0.020036 -2023-02-13 18:34:06,294 - Epoch: [172][ 920/ 1207] Overall Loss 0.229514 Objective Loss 0.229514 LR 0.000250 Time 0.020029 -2023-02-13 18:34:06,487 - Epoch: [172][ 930/ 1207] Overall Loss 0.229472 Objective Loss 0.229472 LR 0.000250 Time 0.020021 -2023-02-13 18:34:06,681 - Epoch: [172][ 940/ 1207] Overall Loss 0.229116 Objective Loss 0.229116 LR 0.000250 Time 0.020014 -2023-02-13 18:34:06,874 - Epoch: [172][ 950/ 1207] Overall Loss 0.229215 Objective Loss 0.229215 LR 0.000250 Time 0.020006 -2023-02-13 18:34:07,070 - Epoch: [172][ 960/ 1207] Overall Loss 0.229187 Objective Loss 0.229187 LR 0.000250 Time 0.020001 -2023-02-13 18:34:07,263 - Epoch: [172][ 970/ 1207] Overall Loss 0.229297 Objective Loss 0.229297 LR 0.000250 Time 0.019994 -2023-02-13 18:34:07,458 - Epoch: [172][ 980/ 1207] Overall Loss 0.229322 Objective Loss 0.229322 LR 0.000250 Time 0.019989 -2023-02-13 18:34:07,651 - Epoch: [172][ 990/ 1207] Overall Loss 0.229210 Objective Loss 0.229210 LR 0.000250 Time 0.019981 -2023-02-13 18:34:07,846 - Epoch: [172][ 1000/ 1207] Overall Loss 0.229245 Objective Loss 0.229245 LR 0.000250 Time 0.019975 -2023-02-13 18:34:08,039 - Epoch: [172][ 1010/ 1207] Overall Loss 0.229680 Objective Loss 0.229680 LR 0.000250 Time 0.019968 -2023-02-13 18:34:08,234 - Epoch: [172][ 1020/ 1207] Overall Loss 0.229498 Objective Loss 0.229498 LR 0.000250 Time 0.019963 -2023-02-13 18:34:08,426 - Epoch: [172][ 1030/ 1207] Overall Loss 0.229456 Objective Loss 0.229456 LR 0.000250 Time 0.019956 -2023-02-13 18:34:08,621 - Epoch: [172][ 1040/ 1207] Overall Loss 0.229553 Objective Loss 0.229553 LR 0.000250 Time 0.019951 -2023-02-13 18:34:08,814 - Epoch: [172][ 1050/ 1207] Overall Loss 0.229601 Objective Loss 0.229601 LR 0.000250 Time 0.019945 -2023-02-13 18:34:09,009 - Epoch: [172][ 1060/ 1207] Overall Loss 0.229500 Objective Loss 0.229500 LR 0.000250 Time 0.019941 -2023-02-13 18:34:09,203 - Epoch: [172][ 1070/ 1207] Overall Loss 0.229303 Objective Loss 0.229303 LR 0.000250 Time 0.019935 -2023-02-13 18:34:09,398 - Epoch: [172][ 1080/ 1207] Overall Loss 0.229447 Objective Loss 0.229447 LR 0.000250 Time 0.019931 -2023-02-13 18:34:09,591 - Epoch: [172][ 1090/ 1207] Overall Loss 0.229473 Objective Loss 0.229473 LR 0.000250 Time 0.019925 -2023-02-13 18:34:09,786 - Epoch: [172][ 1100/ 1207] Overall Loss 0.229625 Objective Loss 0.229625 LR 0.000250 Time 0.019920 -2023-02-13 18:34:09,979 - Epoch: [172][ 1110/ 1207] Overall Loss 0.229680 Objective Loss 0.229680 LR 0.000250 Time 0.019914 -2023-02-13 18:34:10,175 - Epoch: [172][ 1120/ 1207] Overall Loss 0.229612 Objective Loss 0.229612 LR 0.000250 Time 0.019911 -2023-02-13 18:34:10,368 - Epoch: [172][ 1130/ 1207] Overall Loss 0.229653 Objective Loss 0.229653 LR 0.000250 Time 0.019906 -2023-02-13 18:34:10,563 - Epoch: [172][ 1140/ 1207] Overall Loss 0.229650 Objective Loss 0.229650 LR 0.000250 Time 0.019902 -2023-02-13 18:34:10,756 - Epoch: [172][ 1150/ 1207] Overall Loss 0.229587 Objective Loss 0.229587 LR 0.000250 Time 0.019896 -2023-02-13 18:34:10,953 - Epoch: [172][ 1160/ 1207] Overall Loss 0.229492 Objective Loss 0.229492 LR 0.000250 Time 0.019894 -2023-02-13 18:34:11,147 - Epoch: [172][ 1170/ 1207] Overall Loss 0.229667 Objective Loss 0.229667 LR 0.000250 Time 0.019890 -2023-02-13 18:34:11,343 - Epoch: [172][ 1180/ 1207] Overall Loss 0.229890 Objective Loss 0.229890 LR 0.000250 Time 0.019886 -2023-02-13 18:34:11,537 - Epoch: [172][ 1190/ 1207] Overall Loss 0.229851 Objective Loss 0.229851 LR 0.000250 Time 0.019882 -2023-02-13 18:34:11,788 - Epoch: [172][ 1200/ 1207] Overall Loss 0.229888 Objective Loss 0.229888 LR 0.000250 Time 0.019925 -2023-02-13 18:34:11,904 - Epoch: [172][ 1207/ 1207] Overall Loss 0.229967 Objective Loss 0.229967 Top1 85.670732 Top5 98.780488 LR 0.000250 Time 0.019906 -2023-02-13 18:34:11,977 - --- validate (epoch=172)----------- -2023-02-13 18:34:11,977 - 34311 samples (256 per mini-batch) -2023-02-13 18:34:12,375 - Epoch: [172][ 10/ 135] Loss 0.316590 Top1 84.023438 Top5 97.812500 -2023-02-13 18:34:12,507 - Epoch: [172][ 20/ 135] Loss 0.311511 Top1 84.492188 Top5 97.636719 -2023-02-13 18:34:12,639 - Epoch: [172][ 30/ 135] Loss 0.313108 Top1 84.401042 Top5 97.591146 -2023-02-13 18:34:12,764 - Epoch: [172][ 40/ 135] Loss 0.314689 Top1 84.326172 Top5 97.519531 -2023-02-13 18:34:12,891 - Epoch: [172][ 50/ 135] Loss 0.313798 Top1 84.414062 Top5 97.585938 -2023-02-13 18:34:13,019 - Epoch: [172][ 60/ 135] Loss 0.314602 Top1 84.348958 Top5 97.584635 -2023-02-13 18:34:13,148 - Epoch: [172][ 70/ 135] Loss 0.311407 Top1 84.469866 Top5 97.550223 -2023-02-13 18:34:13,274 - Epoch: [172][ 80/ 135] Loss 0.311544 Top1 84.550781 Top5 97.622070 -2023-02-13 18:34:13,400 - Epoch: [172][ 90/ 135] Loss 0.311837 Top1 84.509549 Top5 97.669271 -2023-02-13 18:34:13,526 - Epoch: [172][ 100/ 135] Loss 0.308560 Top1 84.656250 Top5 97.699219 -2023-02-13 18:34:13,653 - Epoch: [172][ 110/ 135] Loss 0.306915 Top1 84.719460 Top5 97.698864 -2023-02-13 18:34:13,779 - Epoch: [172][ 120/ 135] Loss 0.305235 Top1 84.820964 Top5 97.714844 -2023-02-13 18:34:13,912 - Epoch: [172][ 130/ 135] Loss 0.307421 Top1 84.810697 Top5 97.722356 -2023-02-13 18:34:13,958 - Epoch: [172][ 135/ 135] Loss 0.308114 Top1 84.847425 Top5 97.717933 -2023-02-13 18:34:14,044 - ==> Top1: 84.847 Top5: 97.718 Loss: 0.308 - -2023-02-13 18:34:14,045 - ==> Confusion: -[[ 884 4 7 2 5 2 0 3 4 29 0 2 1 4 6 3 4 1 0 1 5] - [ 2 954 1 2 13 22 1 16 4 0 1 0 2 0 1 0 4 0 2 3 5] - [ 7 5 964 13 7 1 11 13 0 1 3 4 0 4 4 9 3 0 3 2 4] - [ 6 0 19 916 1 3 1 2 2 2 8 0 8 0 20 3 4 5 10 0 6] - [ 17 7 0 0 989 10 1 1 2 0 0 7 2 2 6 5 6 1 1 3 6] - [ 3 11 0 3 4 970 3 17 1 7 1 13 3 13 1 3 6 0 0 6 5] - [ 3 3 15 3 0 8 1034 5 0 1 1 1 1 1 0 4 2 4 1 8 4] - [ 2 9 11 1 2 29 1 930 1 1 1 6 5 0 0 0 2 0 12 7 4] - [ 22 2 1 1 1 0 0 0 900 34 5 3 0 11 21 1 2 0 3 1 1] - [ 112 0 2 0 9 0 0 1 28 826 0 1 0 15 7 2 1 2 0 1 5] - [ 2 1 8 8 1 1 2 6 14 2 983 2 2 7 1 0 1 0 6 0 4] - [ 1 1 1 1 1 11 0 7 2 1 0 909 24 7 2 7 2 11 1 12 4] - [ 2 0 2 10 2 5 0 0 3 0 1 31 872 1 1 4 5 11 1 1 7] - [ 6 2 3 0 6 5 0 1 14 23 8 2 1 924 8 5 3 1 1 3 8] - [ 9 2 0 18 4 4 0 2 17 9 1 2 1 0 1004 0 2 5 4 0 8] - [ 6 2 7 0 3 1 3 0 0 0 0 6 4 4 1 974 8 12 1 7 7] - [ 5 4 0 1 5 2 0 0 3 0 0 1 1 1 2 8 1015 1 0 3 9] - [ 5 2 0 4 1 2 2 0 0 2 0 10 7 3 0 16 0 989 0 1 7] - [ 4 4 6 14 2 3 0 28 4 0 6 1 2 0 17 0 0 3 990 2 0] - [ 2 3 2 0 0 5 4 7 1 0 0 13 1 2 1 7 4 4 0 1084 8] - [ 177 217 243 128 134 190 70 200 86 75 159 99 290 277 180 99 283 97 165 264 10001]] - -2023-02-13 18:34:14,047 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:34:14,047 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:34:14,053 - - -2023-02-13 18:34:14,053 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:34:14,943 - Epoch: [173][ 10/ 1207] Overall Loss 0.215963 Objective Loss 0.215963 LR 0.000250 Time 0.088965 -2023-02-13 18:34:15,138 - Epoch: [173][ 20/ 1207] Overall Loss 0.216406 Objective Loss 0.216406 LR 0.000250 Time 0.054204 -2023-02-13 18:34:15,326 - Epoch: [173][ 30/ 1207] Overall Loss 0.214965 Objective Loss 0.214965 LR 0.000250 Time 0.042404 -2023-02-13 18:34:15,515 - Epoch: [173][ 40/ 1207] Overall Loss 0.216185 Objective Loss 0.216185 LR 0.000250 Time 0.036498 -2023-02-13 18:34:15,703 - Epoch: [173][ 50/ 1207] Overall Loss 0.218953 Objective Loss 0.218953 LR 0.000250 Time 0.032954 -2023-02-13 18:34:15,892 - Epoch: [173][ 60/ 1207] Overall Loss 0.221051 Objective Loss 0.221051 LR 0.000250 Time 0.030604 -2023-02-13 18:34:16,080 - Epoch: [173][ 70/ 1207] Overall Loss 0.224195 Objective Loss 0.224195 LR 0.000250 Time 0.028919 -2023-02-13 18:34:16,269 - Epoch: [173][ 80/ 1207] Overall Loss 0.225993 Objective Loss 0.225993 LR 0.000250 Time 0.027655 -2023-02-13 18:34:16,456 - Epoch: [173][ 90/ 1207] Overall Loss 0.224833 Objective Loss 0.224833 LR 0.000250 Time 0.026665 -2023-02-13 18:34:16,645 - Epoch: [173][ 100/ 1207] Overall Loss 0.224877 Objective Loss 0.224877 LR 0.000250 Time 0.025881 -2023-02-13 18:34:16,833 - Epoch: [173][ 110/ 1207] Overall Loss 0.224487 Objective Loss 0.224487 LR 0.000250 Time 0.025238 -2023-02-13 18:34:17,023 - Epoch: [173][ 120/ 1207] Overall Loss 0.223434 Objective Loss 0.223434 LR 0.000250 Time 0.024708 -2023-02-13 18:34:17,211 - Epoch: [173][ 130/ 1207] Overall Loss 0.223023 Objective Loss 0.223023 LR 0.000250 Time 0.024254 -2023-02-13 18:34:17,399 - Epoch: [173][ 140/ 1207] Overall Loss 0.222824 Objective Loss 0.222824 LR 0.000250 Time 0.023859 -2023-02-13 18:34:17,586 - Epoch: [173][ 150/ 1207] Overall Loss 0.222194 Objective Loss 0.222194 LR 0.000250 Time 0.023518 -2023-02-13 18:34:17,774 - Epoch: [173][ 160/ 1207] Overall Loss 0.222684 Objective Loss 0.222684 LR 0.000250 Time 0.023221 -2023-02-13 18:34:17,963 - Epoch: [173][ 170/ 1207] Overall Loss 0.224442 Objective Loss 0.224442 LR 0.000250 Time 0.022962 -2023-02-13 18:34:18,152 - Epoch: [173][ 180/ 1207] Overall Loss 0.224620 Objective Loss 0.224620 LR 0.000250 Time 0.022733 -2023-02-13 18:34:18,340 - Epoch: [173][ 190/ 1207] Overall Loss 0.224216 Objective Loss 0.224216 LR 0.000250 Time 0.022523 -2023-02-13 18:34:18,527 - Epoch: [173][ 200/ 1207] Overall Loss 0.224217 Objective Loss 0.224217 LR 0.000250 Time 0.022334 -2023-02-13 18:34:18,716 - Epoch: [173][ 210/ 1207] Overall Loss 0.225419 Objective Loss 0.225419 LR 0.000250 Time 0.022166 -2023-02-13 18:34:18,905 - Epoch: [173][ 220/ 1207] Overall Loss 0.225621 Objective Loss 0.225621 LR 0.000250 Time 0.022015 -2023-02-13 18:34:19,093 - Epoch: [173][ 230/ 1207] Overall Loss 0.224805 Objective Loss 0.224805 LR 0.000250 Time 0.021873 -2023-02-13 18:34:19,282 - Epoch: [173][ 240/ 1207] Overall Loss 0.225455 Objective Loss 0.225455 LR 0.000250 Time 0.021750 -2023-02-13 18:34:19,470 - Epoch: [173][ 250/ 1207] Overall Loss 0.225000 Objective Loss 0.225000 LR 0.000250 Time 0.021632 -2023-02-13 18:34:19,658 - Epoch: [173][ 260/ 1207] Overall Loss 0.225658 Objective Loss 0.225658 LR 0.000250 Time 0.021522 -2023-02-13 18:34:19,847 - Epoch: [173][ 270/ 1207] Overall Loss 0.225743 Objective Loss 0.225743 LR 0.000250 Time 0.021421 -2023-02-13 18:34:20,036 - Epoch: [173][ 280/ 1207] Overall Loss 0.225602 Objective Loss 0.225602 LR 0.000250 Time 0.021329 -2023-02-13 18:34:20,226 - Epoch: [173][ 290/ 1207] Overall Loss 0.225002 Objective Loss 0.225002 LR 0.000250 Time 0.021248 -2023-02-13 18:34:20,414 - Epoch: [173][ 300/ 1207] Overall Loss 0.225023 Objective Loss 0.225023 LR 0.000250 Time 0.021165 -2023-02-13 18:34:20,602 - Epoch: [173][ 310/ 1207] Overall Loss 0.225043 Objective Loss 0.225043 LR 0.000250 Time 0.021088 -2023-02-13 18:34:20,790 - Epoch: [173][ 320/ 1207] Overall Loss 0.224857 Objective Loss 0.224857 LR 0.000250 Time 0.021017 -2023-02-13 18:34:20,981 - Epoch: [173][ 330/ 1207] Overall Loss 0.224711 Objective Loss 0.224711 LR 0.000250 Time 0.020956 -2023-02-13 18:34:21,170 - Epoch: [173][ 340/ 1207] Overall Loss 0.225255 Objective Loss 0.225255 LR 0.000250 Time 0.020895 -2023-02-13 18:34:21,359 - Epoch: [173][ 350/ 1207] Overall Loss 0.224970 Objective Loss 0.224970 LR 0.000250 Time 0.020836 -2023-02-13 18:34:21,546 - Epoch: [173][ 360/ 1207] Overall Loss 0.225379 Objective Loss 0.225379 LR 0.000250 Time 0.020777 -2023-02-13 18:34:21,735 - Epoch: [173][ 370/ 1207] Overall Loss 0.225679 Objective Loss 0.225679 LR 0.000250 Time 0.020723 -2023-02-13 18:34:21,923 - Epoch: [173][ 380/ 1207] Overall Loss 0.225763 Objective Loss 0.225763 LR 0.000250 Time 0.020673 -2023-02-13 18:34:22,112 - Epoch: [173][ 390/ 1207] Overall Loss 0.226454 Objective Loss 0.226454 LR 0.000250 Time 0.020626 -2023-02-13 18:34:22,300 - Epoch: [173][ 400/ 1207] Overall Loss 0.226459 Objective Loss 0.226459 LR 0.000250 Time 0.020581 -2023-02-13 18:34:22,489 - Epoch: [173][ 410/ 1207] Overall Loss 0.226346 Objective Loss 0.226346 LR 0.000250 Time 0.020537 -2023-02-13 18:34:22,676 - Epoch: [173][ 420/ 1207] Overall Loss 0.226399 Objective Loss 0.226399 LR 0.000250 Time 0.020494 -2023-02-13 18:34:22,865 - Epoch: [173][ 430/ 1207] Overall Loss 0.227117 Objective Loss 0.227117 LR 0.000250 Time 0.020455 -2023-02-13 18:34:23,053 - Epoch: [173][ 440/ 1207] Overall Loss 0.226666 Objective Loss 0.226666 LR 0.000250 Time 0.020417 -2023-02-13 18:34:23,242 - Epoch: [173][ 450/ 1207] Overall Loss 0.226469 Objective Loss 0.226469 LR 0.000250 Time 0.020382 -2023-02-13 18:34:23,430 - Epoch: [173][ 460/ 1207] Overall Loss 0.226639 Objective Loss 0.226639 LR 0.000250 Time 0.020347 -2023-02-13 18:34:23,618 - Epoch: [173][ 470/ 1207] Overall Loss 0.227134 Objective Loss 0.227134 LR 0.000250 Time 0.020313 -2023-02-13 18:34:23,806 - Epoch: [173][ 480/ 1207] Overall Loss 0.227456 Objective Loss 0.227456 LR 0.000250 Time 0.020282 -2023-02-13 18:34:23,995 - Epoch: [173][ 490/ 1207] Overall Loss 0.227225 Objective Loss 0.227225 LR 0.000250 Time 0.020252 -2023-02-13 18:34:24,184 - Epoch: [173][ 500/ 1207] Overall Loss 0.226944 Objective Loss 0.226944 LR 0.000250 Time 0.020224 -2023-02-13 18:34:24,372 - Epoch: [173][ 510/ 1207] Overall Loss 0.226951 Objective Loss 0.226951 LR 0.000250 Time 0.020197 -2023-02-13 18:34:24,561 - Epoch: [173][ 520/ 1207] Overall Loss 0.227418 Objective Loss 0.227418 LR 0.000250 Time 0.020169 -2023-02-13 18:34:24,749 - Epoch: [173][ 530/ 1207] Overall Loss 0.227868 Objective Loss 0.227868 LR 0.000250 Time 0.020143 -2023-02-13 18:34:24,937 - Epoch: [173][ 540/ 1207] Overall Loss 0.227687 Objective Loss 0.227687 LR 0.000250 Time 0.020118 -2023-02-13 18:34:25,126 - Epoch: [173][ 550/ 1207] Overall Loss 0.228109 Objective Loss 0.228109 LR 0.000250 Time 0.020095 -2023-02-13 18:34:25,314 - Epoch: [173][ 560/ 1207] Overall Loss 0.227890 Objective Loss 0.227890 LR 0.000250 Time 0.020072 -2023-02-13 18:34:25,502 - Epoch: [173][ 570/ 1207] Overall Loss 0.227578 Objective Loss 0.227578 LR 0.000250 Time 0.020049 -2023-02-13 18:34:25,691 - Epoch: [173][ 580/ 1207] Overall Loss 0.227711 Objective Loss 0.227711 LR 0.000250 Time 0.020027 -2023-02-13 18:34:25,879 - Epoch: [173][ 590/ 1207] Overall Loss 0.227908 Objective Loss 0.227908 LR 0.000250 Time 0.020007 -2023-02-13 18:34:26,068 - Epoch: [173][ 600/ 1207] Overall Loss 0.227748 Objective Loss 0.227748 LR 0.000250 Time 0.019988 -2023-02-13 18:34:26,257 - Epoch: [173][ 610/ 1207] Overall Loss 0.227595 Objective Loss 0.227595 LR 0.000250 Time 0.019969 -2023-02-13 18:34:26,445 - Epoch: [173][ 620/ 1207] Overall Loss 0.227619 Objective Loss 0.227619 LR 0.000250 Time 0.019950 -2023-02-13 18:34:26,634 - Epoch: [173][ 630/ 1207] Overall Loss 0.227698 Objective Loss 0.227698 LR 0.000250 Time 0.019933 -2023-02-13 18:34:26,823 - Epoch: [173][ 640/ 1207] Overall Loss 0.227760 Objective Loss 0.227760 LR 0.000250 Time 0.019915 -2023-02-13 18:34:27,011 - Epoch: [173][ 650/ 1207] Overall Loss 0.228154 Objective Loss 0.228154 LR 0.000250 Time 0.019898 -2023-02-13 18:34:27,200 - Epoch: [173][ 660/ 1207] Overall Loss 0.228076 Objective Loss 0.228076 LR 0.000250 Time 0.019882 -2023-02-13 18:34:27,388 - Epoch: [173][ 670/ 1207] Overall Loss 0.228087 Objective Loss 0.228087 LR 0.000250 Time 0.019866 -2023-02-13 18:34:27,577 - Epoch: [173][ 680/ 1207] Overall Loss 0.227860 Objective Loss 0.227860 LR 0.000250 Time 0.019851 -2023-02-13 18:34:27,765 - Epoch: [173][ 690/ 1207] Overall Loss 0.227882 Objective Loss 0.227882 LR 0.000250 Time 0.019835 -2023-02-13 18:34:27,954 - Epoch: [173][ 700/ 1207] Overall Loss 0.228152 Objective Loss 0.228152 LR 0.000250 Time 0.019821 -2023-02-13 18:34:28,143 - Epoch: [173][ 710/ 1207] Overall Loss 0.228016 Objective Loss 0.228016 LR 0.000250 Time 0.019808 -2023-02-13 18:34:28,331 - Epoch: [173][ 720/ 1207] Overall Loss 0.227839 Objective Loss 0.227839 LR 0.000250 Time 0.019794 -2023-02-13 18:34:28,520 - Epoch: [173][ 730/ 1207] Overall Loss 0.227715 Objective Loss 0.227715 LR 0.000250 Time 0.019780 -2023-02-13 18:34:28,708 - Epoch: [173][ 740/ 1207] Overall Loss 0.227630 Objective Loss 0.227630 LR 0.000250 Time 0.019766 -2023-02-13 18:34:28,896 - Epoch: [173][ 750/ 1207] Overall Loss 0.227448 Objective Loss 0.227448 LR 0.000250 Time 0.019753 -2023-02-13 18:34:29,085 - Epoch: [173][ 760/ 1207] Overall Loss 0.227438 Objective Loss 0.227438 LR 0.000250 Time 0.019741 -2023-02-13 18:34:29,274 - Epoch: [173][ 770/ 1207] Overall Loss 0.227694 Objective Loss 0.227694 LR 0.000250 Time 0.019730 -2023-02-13 18:34:29,463 - Epoch: [173][ 780/ 1207] Overall Loss 0.227498 Objective Loss 0.227498 LR 0.000250 Time 0.019719 -2023-02-13 18:34:29,651 - Epoch: [173][ 790/ 1207] Overall Loss 0.227460 Objective Loss 0.227460 LR 0.000250 Time 0.019707 -2023-02-13 18:34:29,840 - Epoch: [173][ 800/ 1207] Overall Loss 0.227535 Objective Loss 0.227535 LR 0.000250 Time 0.019697 -2023-02-13 18:34:30,029 - Epoch: [173][ 810/ 1207] Overall Loss 0.227477 Objective Loss 0.227477 LR 0.000250 Time 0.019686 -2023-02-13 18:34:30,218 - Epoch: [173][ 820/ 1207] Overall Loss 0.227583 Objective Loss 0.227583 LR 0.000250 Time 0.019676 -2023-02-13 18:34:30,408 - Epoch: [173][ 830/ 1207] Overall Loss 0.227677 Objective Loss 0.227677 LR 0.000250 Time 0.019667 -2023-02-13 18:34:30,596 - Epoch: [173][ 840/ 1207] Overall Loss 0.227760 Objective Loss 0.227760 LR 0.000250 Time 0.019657 -2023-02-13 18:34:30,785 - Epoch: [173][ 850/ 1207] Overall Loss 0.227743 Objective Loss 0.227743 LR 0.000250 Time 0.019647 -2023-02-13 18:34:30,975 - Epoch: [173][ 860/ 1207] Overall Loss 0.227890 Objective Loss 0.227890 LR 0.000250 Time 0.019639 -2023-02-13 18:34:31,165 - Epoch: [173][ 870/ 1207] Overall Loss 0.227862 Objective Loss 0.227862 LR 0.000250 Time 0.019631 -2023-02-13 18:34:31,354 - Epoch: [173][ 880/ 1207] Overall Loss 0.227627 Objective Loss 0.227627 LR 0.000250 Time 0.019622 -2023-02-13 18:34:31,543 - Epoch: [173][ 890/ 1207] Overall Loss 0.227809 Objective Loss 0.227809 LR 0.000250 Time 0.019614 -2023-02-13 18:34:31,731 - Epoch: [173][ 900/ 1207] Overall Loss 0.227648 Objective Loss 0.227648 LR 0.000250 Time 0.019605 -2023-02-13 18:34:31,921 - Epoch: [173][ 910/ 1207] Overall Loss 0.227734 Objective Loss 0.227734 LR 0.000250 Time 0.019598 -2023-02-13 18:34:32,111 - Epoch: [173][ 920/ 1207] Overall Loss 0.227756 Objective Loss 0.227756 LR 0.000250 Time 0.019590 -2023-02-13 18:34:32,300 - Epoch: [173][ 930/ 1207] Overall Loss 0.227855 Objective Loss 0.227855 LR 0.000250 Time 0.019582 -2023-02-13 18:34:32,489 - Epoch: [173][ 940/ 1207] Overall Loss 0.227711 Objective Loss 0.227711 LR 0.000250 Time 0.019575 -2023-02-13 18:34:32,677 - Epoch: [173][ 950/ 1207] Overall Loss 0.227717 Objective Loss 0.227717 LR 0.000250 Time 0.019567 -2023-02-13 18:34:32,867 - Epoch: [173][ 960/ 1207] Overall Loss 0.227958 Objective Loss 0.227958 LR 0.000250 Time 0.019560 -2023-02-13 18:34:33,056 - Epoch: [173][ 970/ 1207] Overall Loss 0.227769 Objective Loss 0.227769 LR 0.000250 Time 0.019553 -2023-02-13 18:34:33,245 - Epoch: [173][ 980/ 1207] Overall Loss 0.227655 Objective Loss 0.227655 LR 0.000250 Time 0.019546 -2023-02-13 18:34:33,434 - Epoch: [173][ 990/ 1207] Overall Loss 0.227439 Objective Loss 0.227439 LR 0.000250 Time 0.019540 -2023-02-13 18:34:33,623 - Epoch: [173][ 1000/ 1207] Overall Loss 0.227503 Objective Loss 0.227503 LR 0.000250 Time 0.019533 -2023-02-13 18:34:33,812 - Epoch: [173][ 1010/ 1207] Overall Loss 0.227934 Objective Loss 0.227934 LR 0.000250 Time 0.019526 -2023-02-13 18:34:34,001 - Epoch: [173][ 1020/ 1207] Overall Loss 0.227776 Objective Loss 0.227776 LR 0.000250 Time 0.019519 -2023-02-13 18:34:34,190 - Epoch: [173][ 1030/ 1207] Overall Loss 0.227923 Objective Loss 0.227923 LR 0.000250 Time 0.019513 -2023-02-13 18:34:34,379 - Epoch: [173][ 1040/ 1207] Overall Loss 0.228197 Objective Loss 0.228197 LR 0.000250 Time 0.019506 -2023-02-13 18:34:34,568 - Epoch: [173][ 1050/ 1207] Overall Loss 0.228577 Objective Loss 0.228577 LR 0.000250 Time 0.019501 -2023-02-13 18:34:34,758 - Epoch: [173][ 1060/ 1207] Overall Loss 0.228458 Objective Loss 0.228458 LR 0.000250 Time 0.019495 -2023-02-13 18:34:34,947 - Epoch: [173][ 1070/ 1207] Overall Loss 0.228327 Objective Loss 0.228327 LR 0.000250 Time 0.019490 -2023-02-13 18:34:35,136 - Epoch: [173][ 1080/ 1207] Overall Loss 0.228243 Objective Loss 0.228243 LR 0.000250 Time 0.019484 -2023-02-13 18:34:35,325 - Epoch: [173][ 1090/ 1207] Overall Loss 0.228135 Objective Loss 0.228135 LR 0.000250 Time 0.019478 -2023-02-13 18:34:35,514 - Epoch: [173][ 1100/ 1207] Overall Loss 0.228167 Objective Loss 0.228167 LR 0.000250 Time 0.019473 -2023-02-13 18:34:35,704 - Epoch: [173][ 1110/ 1207] Overall Loss 0.228173 Objective Loss 0.228173 LR 0.000250 Time 0.019467 -2023-02-13 18:34:35,893 - Epoch: [173][ 1120/ 1207] Overall Loss 0.228094 Objective Loss 0.228094 LR 0.000250 Time 0.019462 -2023-02-13 18:34:36,082 - Epoch: [173][ 1130/ 1207] Overall Loss 0.227987 Objective Loss 0.227987 LR 0.000250 Time 0.019457 -2023-02-13 18:34:36,271 - Epoch: [173][ 1140/ 1207] Overall Loss 0.227995 Objective Loss 0.227995 LR 0.000250 Time 0.019452 -2023-02-13 18:34:36,461 - Epoch: [173][ 1150/ 1207] Overall Loss 0.228131 Objective Loss 0.228131 LR 0.000250 Time 0.019447 -2023-02-13 18:34:36,650 - Epoch: [173][ 1160/ 1207] Overall Loss 0.228074 Objective Loss 0.228074 LR 0.000250 Time 0.019443 -2023-02-13 18:34:36,839 - Epoch: [173][ 1170/ 1207] Overall Loss 0.227857 Objective Loss 0.227857 LR 0.000250 Time 0.019438 -2023-02-13 18:34:37,028 - Epoch: [173][ 1180/ 1207] Overall Loss 0.227944 Objective Loss 0.227944 LR 0.000250 Time 0.019433 -2023-02-13 18:34:37,217 - Epoch: [173][ 1190/ 1207] Overall Loss 0.228037 Objective Loss 0.228037 LR 0.000250 Time 0.019428 -2023-02-13 18:34:37,462 - Epoch: [173][ 1200/ 1207] Overall Loss 0.227741 Objective Loss 0.227741 LR 0.000250 Time 0.019470 -2023-02-13 18:34:37,578 - Epoch: [173][ 1207/ 1207] Overall Loss 0.227643 Objective Loss 0.227643 Top1 88.719512 Top5 99.085366 LR 0.000250 Time 0.019453 -2023-02-13 18:34:37,650 - --- validate (epoch=173)----------- -2023-02-13 18:34:37,650 - 34311 samples (256 per mini-batch) -2023-02-13 18:34:38,056 - Epoch: [173][ 10/ 135] Loss 0.274905 Top1 86.953125 Top5 98.046875 -2023-02-13 18:34:38,184 - Epoch: [173][ 20/ 135] Loss 0.277983 Top1 86.289062 Top5 97.851562 -2023-02-13 18:34:38,315 - Epoch: [173][ 30/ 135] Loss 0.289706 Top1 85.638021 Top5 97.812500 -2023-02-13 18:34:38,443 - Epoch: [173][ 40/ 135] Loss 0.300009 Top1 85.390625 Top5 97.802734 -2023-02-13 18:34:38,572 - Epoch: [173][ 50/ 135] Loss 0.299614 Top1 85.234375 Top5 97.796875 -2023-02-13 18:34:38,699 - Epoch: [173][ 60/ 135] Loss 0.295663 Top1 85.175781 Top5 97.858073 -2023-02-13 18:34:38,828 - Epoch: [173][ 70/ 135] Loss 0.296133 Top1 85.234375 Top5 97.890625 -2023-02-13 18:34:38,957 - Epoch: [173][ 80/ 135] Loss 0.302448 Top1 85.146484 Top5 97.890625 -2023-02-13 18:34:39,085 - Epoch: [173][ 90/ 135] Loss 0.300990 Top1 85.078125 Top5 97.890625 -2023-02-13 18:34:39,228 - Epoch: [173][ 100/ 135] Loss 0.301690 Top1 84.933594 Top5 97.851562 -2023-02-13 18:34:39,362 - Epoch: [173][ 110/ 135] Loss 0.300984 Top1 84.943182 Top5 97.858665 -2023-02-13 18:34:39,491 - Epoch: [173][ 120/ 135] Loss 0.301656 Top1 84.905599 Top5 97.815755 -2023-02-13 18:34:39,619 - Epoch: [173][ 130/ 135] Loss 0.300981 Top1 84.936899 Top5 97.803486 -2023-02-13 18:34:39,663 - Epoch: [173][ 135/ 135] Loss 0.310829 Top1 84.958177 Top5 97.782052 -2023-02-13 18:34:39,736 - ==> Top1: 84.958 Top5: 97.782 Loss: 0.311 - -2023-02-13 18:34:39,737 - ==> Confusion: -[[ 878 3 7 1 5 4 0 2 1 39 0 1 1 8 3 1 1 2 2 0 8] - [ 3 958 3 4 9 19 2 11 3 1 2 0 1 0 1 1 4 0 3 3 5] - [ 5 4 967 11 5 0 11 12 1 1 5 1 1 6 2 3 4 1 9 1 8] - [ 5 0 24 913 2 4 0 2 2 2 13 0 4 0 17 1 1 3 17 0 6] - [ 13 8 0 1 993 11 1 1 0 0 0 3 4 4 5 4 7 2 1 4 4] - [ 3 11 0 6 5 964 5 19 4 2 1 9 4 18 1 1 4 2 1 4 6] - [ 2 2 14 6 0 3 1040 7 0 3 1 1 3 2 0 3 1 1 1 7 2] - [ 4 9 10 1 0 23 4 928 1 1 0 4 3 1 0 0 2 1 19 8 5] - [ 14 1 1 1 1 1 1 3 905 39 7 2 0 12 14 0 2 0 3 0 2] - [ 83 0 3 0 11 0 0 3 28 851 1 1 0 15 5 1 1 2 1 1 5] - [ 2 1 3 6 1 0 2 4 13 1 993 0 1 9 2 0 1 1 6 0 5] - [ 3 4 1 1 1 15 1 5 1 1 0 903 23 8 0 6 2 8 3 16 3] - [ 1 0 2 12 2 4 0 2 3 1 0 28 866 0 1 7 5 11 3 2 9] - [ 2 2 2 0 9 6 1 4 11 12 10 3 2 939 4 4 3 2 1 1 6] - [ 5 2 1 20 5 2 0 1 21 6 2 1 2 2 1001 0 2 4 7 1 7] - [ 2 2 9 1 3 3 3 2 0 0 1 6 6 0 0 975 11 6 1 8 7] - [ 3 7 1 1 4 2 0 1 3 1 0 1 1 1 1 8 1008 1 1 5 11] - [ 5 2 1 5 0 1 2 0 0 1 2 12 10 0 0 10 2 987 0 2 9] - [ 4 4 6 11 2 2 1 20 3 0 7 2 2 0 10 0 0 2 1009 1 0] - [ 1 3 1 1 0 5 5 12 1 0 0 9 3 1 1 5 5 2 0 1089 4] - [ 160 237 251 144 132 210 85 184 94 80 210 93 306 303 148 87 221 85 168 253 9983]] - -2023-02-13 18:34:39,738 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:34:39,738 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:34:39,744 - - -2023-02-13 18:34:39,744 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:34:40,619 - Epoch: [174][ 10/ 1207] Overall Loss 0.243674 Objective Loss 0.243674 LR 0.000250 Time 0.087375 -2023-02-13 18:34:40,820 - Epoch: [174][ 20/ 1207] Overall Loss 0.234664 Objective Loss 0.234664 LR 0.000250 Time 0.053747 -2023-02-13 18:34:41,016 - Epoch: [174][ 30/ 1207] Overall Loss 0.232961 Objective Loss 0.232961 LR 0.000250 Time 0.042338 -2023-02-13 18:34:41,212 - Epoch: [174][ 40/ 1207] Overall Loss 0.228777 Objective Loss 0.228777 LR 0.000250 Time 0.036660 -2023-02-13 18:34:41,406 - Epoch: [174][ 50/ 1207] Overall Loss 0.226751 Objective Loss 0.226751 LR 0.000250 Time 0.033198 -2023-02-13 18:34:41,602 - Epoch: [174][ 60/ 1207] Overall Loss 0.225246 Objective Loss 0.225246 LR 0.000250 Time 0.030920 -2023-02-13 18:34:41,796 - Epoch: [174][ 70/ 1207] Overall Loss 0.225252 Objective Loss 0.225252 LR 0.000250 Time 0.029270 -2023-02-13 18:34:41,992 - Epoch: [174][ 80/ 1207] Overall Loss 0.226841 Objective Loss 0.226841 LR 0.000250 Time 0.028055 -2023-02-13 18:34:42,186 - Epoch: [174][ 90/ 1207] Overall Loss 0.225592 Objective Loss 0.225592 LR 0.000250 Time 0.027089 -2023-02-13 18:34:42,382 - Epoch: [174][ 100/ 1207] Overall Loss 0.224204 Objective Loss 0.224204 LR 0.000250 Time 0.026338 -2023-02-13 18:34:42,576 - Epoch: [174][ 110/ 1207] Overall Loss 0.224415 Objective Loss 0.224415 LR 0.000250 Time 0.025700 -2023-02-13 18:34:42,771 - Epoch: [174][ 120/ 1207] Overall Loss 0.223549 Objective Loss 0.223549 LR 0.000250 Time 0.025185 -2023-02-13 18:34:42,965 - Epoch: [174][ 130/ 1207] Overall Loss 0.222855 Objective Loss 0.222855 LR 0.000250 Time 0.024737 -2023-02-13 18:34:43,161 - Epoch: [174][ 140/ 1207] Overall Loss 0.222577 Objective Loss 0.222577 LR 0.000250 Time 0.024364 -2023-02-13 18:34:43,355 - Epoch: [174][ 150/ 1207] Overall Loss 0.222832 Objective Loss 0.222832 LR 0.000250 Time 0.024031 -2023-02-13 18:34:43,551 - Epoch: [174][ 160/ 1207] Overall Loss 0.223892 Objective Loss 0.223892 LR 0.000250 Time 0.023750 -2023-02-13 18:34:43,744 - Epoch: [174][ 170/ 1207] Overall Loss 0.224453 Objective Loss 0.224453 LR 0.000250 Time 0.023487 -2023-02-13 18:34:43,939 - Epoch: [174][ 180/ 1207] Overall Loss 0.224093 Objective Loss 0.224093 LR 0.000250 Time 0.023267 -2023-02-13 18:34:44,133 - Epoch: [174][ 190/ 1207] Overall Loss 0.224481 Objective Loss 0.224481 LR 0.000250 Time 0.023060 -2023-02-13 18:34:44,330 - Epoch: [174][ 200/ 1207] Overall Loss 0.223950 Objective Loss 0.223950 LR 0.000250 Time 0.022887 -2023-02-13 18:34:44,523 - Epoch: [174][ 210/ 1207] Overall Loss 0.223560 Objective Loss 0.223560 LR 0.000250 Time 0.022717 -2023-02-13 18:34:44,719 - Epoch: [174][ 220/ 1207] Overall Loss 0.223412 Objective Loss 0.223412 LR 0.000250 Time 0.022572 -2023-02-13 18:34:44,913 - Epoch: [174][ 230/ 1207] Overall Loss 0.223728 Objective Loss 0.223728 LR 0.000250 Time 0.022434 -2023-02-13 18:34:45,109 - Epoch: [174][ 240/ 1207] Overall Loss 0.223148 Objective Loss 0.223148 LR 0.000250 Time 0.022312 -2023-02-13 18:34:45,303 - Epoch: [174][ 250/ 1207] Overall Loss 0.223751 Objective Loss 0.223751 LR 0.000250 Time 0.022196 -2023-02-13 18:34:45,499 - Epoch: [174][ 260/ 1207] Overall Loss 0.223049 Objective Loss 0.223049 LR 0.000250 Time 0.022094 -2023-02-13 18:34:45,692 - Epoch: [174][ 270/ 1207] Overall Loss 0.223036 Objective Loss 0.223036 LR 0.000250 Time 0.021992 -2023-02-13 18:34:45,888 - Epoch: [174][ 280/ 1207] Overall Loss 0.223228 Objective Loss 0.223228 LR 0.000250 Time 0.021904 -2023-02-13 18:34:46,083 - Epoch: [174][ 290/ 1207] Overall Loss 0.223854 Objective Loss 0.223854 LR 0.000250 Time 0.021819 -2023-02-13 18:34:46,279 - Epoch: [174][ 300/ 1207] Overall Loss 0.223266 Objective Loss 0.223266 LR 0.000250 Time 0.021745 -2023-02-13 18:34:46,473 - Epoch: [174][ 310/ 1207] Overall Loss 0.223206 Objective Loss 0.223206 LR 0.000250 Time 0.021667 -2023-02-13 18:34:46,669 - Epoch: [174][ 320/ 1207] Overall Loss 0.224145 Objective Loss 0.224145 LR 0.000250 Time 0.021602 -2023-02-13 18:34:46,863 - Epoch: [174][ 330/ 1207] Overall Loss 0.224075 Objective Loss 0.224075 LR 0.000250 Time 0.021533 -2023-02-13 18:34:47,059 - Epoch: [174][ 340/ 1207] Overall Loss 0.224319 Objective Loss 0.224319 LR 0.000250 Time 0.021476 -2023-02-13 18:34:47,253 - Epoch: [174][ 350/ 1207] Overall Loss 0.224417 Objective Loss 0.224417 LR 0.000250 Time 0.021417 -2023-02-13 18:34:47,449 - Epoch: [174][ 360/ 1207] Overall Loss 0.225053 Objective Loss 0.225053 LR 0.000250 Time 0.021364 -2023-02-13 18:34:47,643 - Epoch: [174][ 370/ 1207] Overall Loss 0.225299 Objective Loss 0.225299 LR 0.000250 Time 0.021310 -2023-02-13 18:34:47,839 - Epoch: [174][ 380/ 1207] Overall Loss 0.225543 Objective Loss 0.225543 LR 0.000250 Time 0.021265 -2023-02-13 18:34:48,033 - Epoch: [174][ 390/ 1207] Overall Loss 0.225329 Objective Loss 0.225329 LR 0.000250 Time 0.021216 -2023-02-13 18:34:48,230 - Epoch: [174][ 400/ 1207] Overall Loss 0.224868 Objective Loss 0.224868 LR 0.000250 Time 0.021175 -2023-02-13 18:34:48,423 - Epoch: [174][ 410/ 1207] Overall Loss 0.224784 Objective Loss 0.224784 LR 0.000250 Time 0.021131 -2023-02-13 18:34:48,619 - Epoch: [174][ 420/ 1207] Overall Loss 0.225179 Objective Loss 0.225179 LR 0.000250 Time 0.021093 -2023-02-13 18:34:48,813 - Epoch: [174][ 430/ 1207] Overall Loss 0.225015 Objective Loss 0.225015 LR 0.000250 Time 0.021053 -2023-02-13 18:34:49,010 - Epoch: [174][ 440/ 1207] Overall Loss 0.224727 Objective Loss 0.224727 LR 0.000250 Time 0.021020 -2023-02-13 18:34:49,203 - Epoch: [174][ 450/ 1207] Overall Loss 0.224980 Objective Loss 0.224980 LR 0.000250 Time 0.020982 -2023-02-13 18:34:49,400 - Epoch: [174][ 460/ 1207] Overall Loss 0.225218 Objective Loss 0.225218 LR 0.000250 Time 0.020953 -2023-02-13 18:34:49,595 - Epoch: [174][ 470/ 1207] Overall Loss 0.225360 Objective Loss 0.225360 LR 0.000250 Time 0.020920 -2023-02-13 18:34:49,803 - Epoch: [174][ 480/ 1207] Overall Loss 0.225166 Objective Loss 0.225166 LR 0.000250 Time 0.020918 -2023-02-13 18:34:50,005 - Epoch: [174][ 490/ 1207] Overall Loss 0.225008 Objective Loss 0.225008 LR 0.000250 Time 0.020903 -2023-02-13 18:34:50,214 - Epoch: [174][ 500/ 1207] Overall Loss 0.225480 Objective Loss 0.225480 LR 0.000250 Time 0.020902 -2023-02-13 18:34:50,407 - Epoch: [174][ 510/ 1207] Overall Loss 0.226000 Objective Loss 0.226000 LR 0.000250 Time 0.020870 -2023-02-13 18:34:50,603 - Epoch: [174][ 520/ 1207] Overall Loss 0.225806 Objective Loss 0.225806 LR 0.000250 Time 0.020845 -2023-02-13 18:34:50,798 - Epoch: [174][ 530/ 1207] Overall Loss 0.226002 Objective Loss 0.226002 LR 0.000250 Time 0.020818 -2023-02-13 18:34:50,995 - Epoch: [174][ 540/ 1207] Overall Loss 0.225694 Objective Loss 0.225694 LR 0.000250 Time 0.020797 -2023-02-13 18:34:51,188 - Epoch: [174][ 550/ 1207] Overall Loss 0.225448 Objective Loss 0.225448 LR 0.000250 Time 0.020770 -2023-02-13 18:34:51,385 - Epoch: [174][ 560/ 1207] Overall Loss 0.225733 Objective Loss 0.225733 LR 0.000250 Time 0.020749 -2023-02-13 18:34:51,580 - Epoch: [174][ 570/ 1207] Overall Loss 0.225778 Objective Loss 0.225778 LR 0.000250 Time 0.020726 -2023-02-13 18:34:51,776 - Epoch: [174][ 580/ 1207] Overall Loss 0.225865 Objective Loss 0.225865 LR 0.000250 Time 0.020706 -2023-02-13 18:34:51,970 - Epoch: [174][ 590/ 1207] Overall Loss 0.225798 Objective Loss 0.225798 LR 0.000250 Time 0.020683 -2023-02-13 18:34:52,166 - Epoch: [174][ 600/ 1207] Overall Loss 0.225879 Objective Loss 0.225879 LR 0.000250 Time 0.020664 -2023-02-13 18:34:52,359 - Epoch: [174][ 610/ 1207] Overall Loss 0.225756 Objective Loss 0.225756 LR 0.000250 Time 0.020643 -2023-02-13 18:34:52,556 - Epoch: [174][ 620/ 1207] Overall Loss 0.225773 Objective Loss 0.225773 LR 0.000250 Time 0.020625 -2023-02-13 18:34:52,750 - Epoch: [174][ 630/ 1207] Overall Loss 0.225799 Objective Loss 0.225799 LR 0.000250 Time 0.020606 -2023-02-13 18:34:52,946 - Epoch: [174][ 640/ 1207] Overall Loss 0.225984 Objective Loss 0.225984 LR 0.000250 Time 0.020590 -2023-02-13 18:34:53,140 - Epoch: [174][ 650/ 1207] Overall Loss 0.225705 Objective Loss 0.225705 LR 0.000250 Time 0.020572 -2023-02-13 18:34:53,337 - Epoch: [174][ 660/ 1207] Overall Loss 0.225609 Objective Loss 0.225609 LR 0.000250 Time 0.020557 -2023-02-13 18:34:53,531 - Epoch: [174][ 670/ 1207] Overall Loss 0.225271 Objective Loss 0.225271 LR 0.000250 Time 0.020539 -2023-02-13 18:34:53,727 - Epoch: [174][ 680/ 1207] Overall Loss 0.225121 Objective Loss 0.225121 LR 0.000250 Time 0.020526 -2023-02-13 18:34:53,922 - Epoch: [174][ 690/ 1207] Overall Loss 0.225213 Objective Loss 0.225213 LR 0.000250 Time 0.020509 -2023-02-13 18:34:54,118 - Epoch: [174][ 700/ 1207] Overall Loss 0.225445 Objective Loss 0.225445 LR 0.000250 Time 0.020495 -2023-02-13 18:34:54,312 - Epoch: [174][ 710/ 1207] Overall Loss 0.225243 Objective Loss 0.225243 LR 0.000250 Time 0.020480 -2023-02-13 18:34:54,509 - Epoch: [174][ 720/ 1207] Overall Loss 0.225284 Objective Loss 0.225284 LR 0.000250 Time 0.020469 -2023-02-13 18:34:54,703 - Epoch: [174][ 730/ 1207] Overall Loss 0.225134 Objective Loss 0.225134 LR 0.000250 Time 0.020453 -2023-02-13 18:34:54,899 - Epoch: [174][ 740/ 1207] Overall Loss 0.225269 Objective Loss 0.225269 LR 0.000250 Time 0.020441 -2023-02-13 18:34:55,092 - Epoch: [174][ 750/ 1207] Overall Loss 0.225196 Objective Loss 0.225196 LR 0.000250 Time 0.020426 -2023-02-13 18:34:55,289 - Epoch: [174][ 760/ 1207] Overall Loss 0.225203 Objective Loss 0.225203 LR 0.000250 Time 0.020416 -2023-02-13 18:34:55,483 - Epoch: [174][ 770/ 1207] Overall Loss 0.225052 Objective Loss 0.225052 LR 0.000250 Time 0.020402 -2023-02-13 18:34:55,679 - Epoch: [174][ 780/ 1207] Overall Loss 0.225163 Objective Loss 0.225163 LR 0.000250 Time 0.020392 -2023-02-13 18:34:55,873 - Epoch: [174][ 790/ 1207] Overall Loss 0.225075 Objective Loss 0.225075 LR 0.000250 Time 0.020379 -2023-02-13 18:34:56,070 - Epoch: [174][ 800/ 1207] Overall Loss 0.225160 Objective Loss 0.225160 LR 0.000250 Time 0.020370 -2023-02-13 18:34:56,264 - Epoch: [174][ 810/ 1207] Overall Loss 0.225153 Objective Loss 0.225153 LR 0.000250 Time 0.020358 -2023-02-13 18:34:56,461 - Epoch: [174][ 820/ 1207] Overall Loss 0.225327 Objective Loss 0.225327 LR 0.000250 Time 0.020348 -2023-02-13 18:34:56,656 - Epoch: [174][ 830/ 1207] Overall Loss 0.225084 Objective Loss 0.225084 LR 0.000250 Time 0.020338 -2023-02-13 18:34:56,853 - Epoch: [174][ 840/ 1207] Overall Loss 0.224807 Objective Loss 0.224807 LR 0.000250 Time 0.020330 -2023-02-13 18:34:57,047 - Epoch: [174][ 850/ 1207] Overall Loss 0.224829 Objective Loss 0.224829 LR 0.000250 Time 0.020319 -2023-02-13 18:34:57,244 - Epoch: [174][ 860/ 1207] Overall Loss 0.224859 Objective Loss 0.224859 LR 0.000250 Time 0.020311 -2023-02-13 18:34:57,439 - Epoch: [174][ 870/ 1207] Overall Loss 0.224860 Objective Loss 0.224860 LR 0.000250 Time 0.020301 -2023-02-13 18:34:57,635 - Epoch: [174][ 880/ 1207] Overall Loss 0.224737 Objective Loss 0.224737 LR 0.000250 Time 0.020293 -2023-02-13 18:34:57,830 - Epoch: [174][ 890/ 1207] Overall Loss 0.224878 Objective Loss 0.224878 LR 0.000250 Time 0.020283 -2023-02-13 18:34:58,027 - Epoch: [174][ 900/ 1207] Overall Loss 0.224917 Objective Loss 0.224917 LR 0.000250 Time 0.020277 -2023-02-13 18:34:58,222 - Epoch: [174][ 910/ 1207] Overall Loss 0.224939 Objective Loss 0.224939 LR 0.000250 Time 0.020268 -2023-02-13 18:34:58,422 - Epoch: [174][ 920/ 1207] Overall Loss 0.225065 Objective Loss 0.225065 LR 0.000250 Time 0.020264 -2023-02-13 18:34:58,618 - Epoch: [174][ 930/ 1207] Overall Loss 0.224959 Objective Loss 0.224959 LR 0.000250 Time 0.020257 -2023-02-13 18:34:58,817 - Epoch: [174][ 940/ 1207] Overall Loss 0.224815 Objective Loss 0.224815 LR 0.000250 Time 0.020252 -2023-02-13 18:34:59,012 - Epoch: [174][ 950/ 1207] Overall Loss 0.224865 Objective Loss 0.224865 LR 0.000250 Time 0.020244 -2023-02-13 18:34:59,208 - Epoch: [174][ 960/ 1207] Overall Loss 0.225016 Objective Loss 0.225016 LR 0.000250 Time 0.020237 -2023-02-13 18:34:59,403 - Epoch: [174][ 970/ 1207] Overall Loss 0.225057 Objective Loss 0.225057 LR 0.000250 Time 0.020229 -2023-02-13 18:34:59,599 - Epoch: [174][ 980/ 1207] Overall Loss 0.225011 Objective Loss 0.225011 LR 0.000250 Time 0.020222 -2023-02-13 18:34:59,793 - Epoch: [174][ 990/ 1207] Overall Loss 0.225085 Objective Loss 0.225085 LR 0.000250 Time 0.020213 -2023-02-13 18:34:59,989 - Epoch: [174][ 1000/ 1207] Overall Loss 0.225111 Objective Loss 0.225111 LR 0.000250 Time 0.020208 -2023-02-13 18:35:00,183 - Epoch: [174][ 1010/ 1207] Overall Loss 0.225125 Objective Loss 0.225125 LR 0.000250 Time 0.020199 -2023-02-13 18:35:00,380 - Epoch: [174][ 1020/ 1207] Overall Loss 0.224983 Objective Loss 0.224983 LR 0.000250 Time 0.020194 -2023-02-13 18:35:00,575 - Epoch: [174][ 1030/ 1207] Overall Loss 0.224958 Objective Loss 0.224958 LR 0.000250 Time 0.020186 -2023-02-13 18:35:00,771 - Epoch: [174][ 1040/ 1207] Overall Loss 0.225011 Objective Loss 0.225011 LR 0.000250 Time 0.020180 -2023-02-13 18:35:00,966 - Epoch: [174][ 1050/ 1207] Overall Loss 0.224794 Objective Loss 0.224794 LR 0.000250 Time 0.020173 -2023-02-13 18:35:01,162 - Epoch: [174][ 1060/ 1207] Overall Loss 0.224807 Objective Loss 0.224807 LR 0.000250 Time 0.020168 -2023-02-13 18:35:01,357 - Epoch: [174][ 1070/ 1207] Overall Loss 0.224745 Objective Loss 0.224745 LR 0.000250 Time 0.020161 -2023-02-13 18:35:01,553 - Epoch: [174][ 1080/ 1207] Overall Loss 0.224781 Objective Loss 0.224781 LR 0.000250 Time 0.020156 -2023-02-13 18:35:01,747 - Epoch: [174][ 1090/ 1207] Overall Loss 0.224957 Objective Loss 0.224957 LR 0.000250 Time 0.020149 -2023-02-13 18:35:01,944 - Epoch: [174][ 1100/ 1207] Overall Loss 0.224841 Objective Loss 0.224841 LR 0.000250 Time 0.020144 -2023-02-13 18:35:02,139 - Epoch: [174][ 1110/ 1207] Overall Loss 0.224881 Objective Loss 0.224881 LR 0.000250 Time 0.020138 -2023-02-13 18:35:02,335 - Epoch: [174][ 1120/ 1207] Overall Loss 0.225061 Objective Loss 0.225061 LR 0.000250 Time 0.020133 -2023-02-13 18:35:02,529 - Epoch: [174][ 1130/ 1207] Overall Loss 0.225282 Objective Loss 0.225282 LR 0.000250 Time 0.020126 -2023-02-13 18:35:02,725 - Epoch: [174][ 1140/ 1207] Overall Loss 0.225394 Objective Loss 0.225394 LR 0.000250 Time 0.020121 -2023-02-13 18:35:02,920 - Epoch: [174][ 1150/ 1207] Overall Loss 0.225302 Objective Loss 0.225302 LR 0.000250 Time 0.020115 -2023-02-13 18:35:03,116 - Epoch: [174][ 1160/ 1207] Overall Loss 0.225457 Objective Loss 0.225457 LR 0.000250 Time 0.020111 -2023-02-13 18:35:03,310 - Epoch: [174][ 1170/ 1207] Overall Loss 0.225618 Objective Loss 0.225618 LR 0.000250 Time 0.020105 -2023-02-13 18:35:03,509 - Epoch: [174][ 1180/ 1207] Overall Loss 0.225730 Objective Loss 0.225730 LR 0.000250 Time 0.020102 -2023-02-13 18:35:03,705 - Epoch: [174][ 1190/ 1207] Overall Loss 0.225784 Objective Loss 0.225784 LR 0.000250 Time 0.020098 -2023-02-13 18:35:03,948 - Epoch: [174][ 1200/ 1207] Overall Loss 0.225921 Objective Loss 0.225921 LR 0.000250 Time 0.020132 -2023-02-13 18:35:04,061 - Epoch: [174][ 1207/ 1207] Overall Loss 0.225887 Objective Loss 0.225887 Top1 87.500000 Top5 98.780488 LR 0.000250 Time 0.020109 -2023-02-13 18:35:04,133 - --- validate (epoch=174)----------- -2023-02-13 18:35:04,134 - 34311 samples (256 per mini-batch) -2023-02-13 18:35:04,635 - Epoch: [174][ 10/ 135] Loss 0.339004 Top1 83.945312 Top5 97.773438 -2023-02-13 18:35:04,760 - Epoch: [174][ 20/ 135] Loss 0.316916 Top1 84.492188 Top5 98.007812 -2023-02-13 18:35:04,887 - Epoch: [174][ 30/ 135] Loss 0.311986 Top1 84.635417 Top5 97.929688 -2023-02-13 18:35:05,016 - Epoch: [174][ 40/ 135] Loss 0.308333 Top1 84.531250 Top5 97.900391 -2023-02-13 18:35:05,142 - Epoch: [174][ 50/ 135] Loss 0.305208 Top1 84.648438 Top5 97.851562 -2023-02-13 18:35:05,272 - Epoch: [174][ 60/ 135] Loss 0.302628 Top1 84.876302 Top5 97.825521 -2023-02-13 18:35:05,401 - Epoch: [174][ 70/ 135] Loss 0.302475 Top1 84.983259 Top5 97.834821 -2023-02-13 18:35:05,530 - Epoch: [174][ 80/ 135] Loss 0.299254 Top1 85.014648 Top5 97.856445 -2023-02-13 18:35:05,660 - Epoch: [174][ 90/ 135] Loss 0.299902 Top1 85.047743 Top5 97.786458 -2023-02-13 18:35:05,791 - Epoch: [174][ 100/ 135] Loss 0.303075 Top1 84.902344 Top5 97.687500 -2023-02-13 18:35:05,922 - Epoch: [174][ 110/ 135] Loss 0.305162 Top1 84.850852 Top5 97.649148 -2023-02-13 18:35:06,051 - Epoch: [174][ 120/ 135] Loss 0.302847 Top1 84.830729 Top5 97.695312 -2023-02-13 18:35:06,184 - Epoch: [174][ 130/ 135] Loss 0.300192 Top1 84.879808 Top5 97.698317 -2023-02-13 18:35:06,231 - Epoch: [174][ 135/ 135] Loss 0.300013 Top1 84.891143 Top5 97.709189 -2023-02-13 18:35:06,300 - ==> Top1: 84.891 Top5: 97.709 Loss: 0.300 - -2023-02-13 18:35:06,301 - ==> Confusion: -[[ 858 3 8 1 8 1 0 3 5 49 0 4 2 5 6 2 2 4 0 0 6] - [ 3 954 1 1 10 22 1 12 4 0 2 0 3 0 1 1 5 1 2 2 8] - [ 7 5 959 8 3 2 13 13 2 1 5 5 2 4 5 5 2 1 5 4 7] - [ 4 0 19 897 1 6 1 1 3 3 18 2 8 0 16 1 3 3 23 0 7] - [ 7 6 0 0 997 15 2 2 3 1 0 7 5 2 5 5 3 1 0 1 4] - [ 1 20 0 4 3 975 5 16 3 3 1 5 2 14 0 3 5 1 0 5 4] - [ 1 2 8 1 0 8 1048 3 0 2 0 2 2 1 0 4 2 2 2 6 5] - [ 2 13 8 0 1 28 2 925 0 1 1 9 1 2 0 0 1 1 18 7 4] - [ 10 1 0 1 1 1 1 1 930 32 5 2 1 6 11 0 1 0 5 0 0] - [ 56 1 2 0 8 3 0 3 37 867 0 1 3 18 5 0 2 1 1 0 4] - [ 1 2 3 5 1 2 2 4 22 2 982 2 2 8 2 0 1 1 6 0 3] - [ 2 2 0 0 2 14 1 4 4 2 0 915 25 10 0 4 3 9 1 6 1] - [ 0 0 2 7 1 4 0 0 2 0 0 19 894 1 1 8 4 9 2 1 4] - [ 5 2 1 0 6 10 0 2 11 14 8 4 2 941 5 4 4 1 0 0 4] - [ 4 0 1 11 4 3 0 2 26 7 3 2 3 2 1001 0 3 3 8 0 9] - [ 2 0 5 1 5 2 5 0 0 0 0 7 9 3 1 972 12 8 0 7 7] - [ 1 5 1 1 7 1 0 2 5 0 0 1 1 1 1 10 1007 3 2 2 10] - [ 3 3 0 4 0 2 2 0 2 2 1 11 15 2 0 16 1 979 0 3 5] - [ 4 4 4 7 0 1 0 21 6 0 6 0 5 0 12 1 0 3 1010 1 1] - [ 0 3 0 0 0 5 6 9 2 0 0 17 2 2 1 9 2 3 0 1080 7] - [ 143 248 210 91 144 229 84 158 117 84 201 116 315 317 157 86 285 83 169 261 9936]] - -2023-02-13 18:35:06,302 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:35:06,302 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:35:06,308 - - -2023-02-13 18:35:06,308 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:35:07,192 - Epoch: [175][ 10/ 1207] Overall Loss 0.226419 Objective Loss 0.226419 LR 0.000250 Time 0.088285 -2023-02-13 18:35:07,383 - Epoch: [175][ 20/ 1207] Overall Loss 0.230525 Objective Loss 0.230525 LR 0.000250 Time 0.053681 -2023-02-13 18:35:07,571 - Epoch: [175][ 30/ 1207] Overall Loss 0.227057 Objective Loss 0.227057 LR 0.000250 Time 0.042046 -2023-02-13 18:35:07,758 - Epoch: [175][ 40/ 1207] Overall Loss 0.226112 Objective Loss 0.226112 LR 0.000250 Time 0.036209 -2023-02-13 18:35:07,947 - Epoch: [175][ 50/ 1207] Overall Loss 0.225220 Objective Loss 0.225220 LR 0.000250 Time 0.032726 -2023-02-13 18:35:08,134 - Epoch: [175][ 60/ 1207] Overall Loss 0.225934 Objective Loss 0.225934 LR 0.000250 Time 0.030395 -2023-02-13 18:35:08,322 - Epoch: [175][ 70/ 1207] Overall Loss 0.227833 Objective Loss 0.227833 LR 0.000250 Time 0.028734 -2023-02-13 18:35:08,510 - Epoch: [175][ 80/ 1207] Overall Loss 0.225021 Objective Loss 0.225021 LR 0.000250 Time 0.027486 -2023-02-13 18:35:08,698 - Epoch: [175][ 90/ 1207] Overall Loss 0.225371 Objective Loss 0.225371 LR 0.000250 Time 0.026517 -2023-02-13 18:35:08,886 - Epoch: [175][ 100/ 1207] Overall Loss 0.223563 Objective Loss 0.223563 LR 0.000250 Time 0.025743 -2023-02-13 18:35:09,074 - Epoch: [175][ 110/ 1207] Overall Loss 0.222860 Objective Loss 0.222860 LR 0.000250 Time 0.025107 -2023-02-13 18:35:09,262 - Epoch: [175][ 120/ 1207] Overall Loss 0.224050 Objective Loss 0.224050 LR 0.000250 Time 0.024575 -2023-02-13 18:35:09,450 - Epoch: [175][ 130/ 1207] Overall Loss 0.225888 Objective Loss 0.225888 LR 0.000250 Time 0.024132 -2023-02-13 18:35:09,639 - Epoch: [175][ 140/ 1207] Overall Loss 0.226543 Objective Loss 0.226543 LR 0.000250 Time 0.023751 -2023-02-13 18:35:09,826 - Epoch: [175][ 150/ 1207] Overall Loss 0.226376 Objective Loss 0.226376 LR 0.000250 Time 0.023415 -2023-02-13 18:35:10,014 - Epoch: [175][ 160/ 1207] Overall Loss 0.226002 Objective Loss 0.226002 LR 0.000250 Time 0.023123 -2023-02-13 18:35:10,202 - Epoch: [175][ 170/ 1207] Overall Loss 0.226024 Objective Loss 0.226024 LR 0.000250 Time 0.022866 -2023-02-13 18:35:10,391 - Epoch: [175][ 180/ 1207] Overall Loss 0.226541 Objective Loss 0.226541 LR 0.000250 Time 0.022642 -2023-02-13 18:35:10,578 - Epoch: [175][ 190/ 1207] Overall Loss 0.226810 Objective Loss 0.226810 LR 0.000250 Time 0.022437 -2023-02-13 18:35:10,767 - Epoch: [175][ 200/ 1207] Overall Loss 0.227334 Objective Loss 0.227334 LR 0.000250 Time 0.022257 -2023-02-13 18:35:10,957 - Epoch: [175][ 210/ 1207] Overall Loss 0.226197 Objective Loss 0.226197 LR 0.000250 Time 0.022097 -2023-02-13 18:35:11,146 - Epoch: [175][ 220/ 1207] Overall Loss 0.226129 Objective Loss 0.226129 LR 0.000250 Time 0.021952 -2023-02-13 18:35:11,335 - Epoch: [175][ 230/ 1207] Overall Loss 0.225771 Objective Loss 0.225771 LR 0.000250 Time 0.021818 -2023-02-13 18:35:11,524 - Epoch: [175][ 240/ 1207] Overall Loss 0.225125 Objective Loss 0.225125 LR 0.000250 Time 0.021695 -2023-02-13 18:35:11,712 - Epoch: [175][ 250/ 1207] Overall Loss 0.225873 Objective Loss 0.225873 LR 0.000250 Time 0.021579 -2023-02-13 18:35:11,901 - Epoch: [175][ 260/ 1207] Overall Loss 0.226306 Objective Loss 0.226306 LR 0.000250 Time 0.021473 -2023-02-13 18:35:12,090 - Epoch: [175][ 270/ 1207] Overall Loss 0.226277 Objective Loss 0.226277 LR 0.000250 Time 0.021376 -2023-02-13 18:35:12,278 - Epoch: [175][ 280/ 1207] Overall Loss 0.226198 Objective Loss 0.226198 LR 0.000250 Time 0.021284 -2023-02-13 18:35:12,468 - Epoch: [175][ 290/ 1207] Overall Loss 0.226487 Objective Loss 0.226487 LR 0.000250 Time 0.021201 -2023-02-13 18:35:12,657 - Epoch: [175][ 300/ 1207] Overall Loss 0.226439 Objective Loss 0.226439 LR 0.000250 Time 0.021125 -2023-02-13 18:35:12,847 - Epoch: [175][ 310/ 1207] Overall Loss 0.226488 Objective Loss 0.226488 LR 0.000250 Time 0.021053 -2023-02-13 18:35:13,036 - Epoch: [175][ 320/ 1207] Overall Loss 0.226066 Objective Loss 0.226066 LR 0.000250 Time 0.020986 -2023-02-13 18:35:13,225 - Epoch: [175][ 330/ 1207] Overall Loss 0.224907 Objective Loss 0.224907 LR 0.000250 Time 0.020921 -2023-02-13 18:35:13,414 - Epoch: [175][ 340/ 1207] Overall Loss 0.224829 Objective Loss 0.224829 LR 0.000250 Time 0.020861 -2023-02-13 18:35:13,602 - Epoch: [175][ 350/ 1207] Overall Loss 0.224960 Objective Loss 0.224960 LR 0.000250 Time 0.020802 -2023-02-13 18:35:13,791 - Epoch: [175][ 360/ 1207] Overall Loss 0.224894 Objective Loss 0.224894 LR 0.000250 Time 0.020748 -2023-02-13 18:35:13,979 - Epoch: [175][ 370/ 1207] Overall Loss 0.224898 Objective Loss 0.224898 LR 0.000250 Time 0.020694 -2023-02-13 18:35:14,169 - Epoch: [175][ 380/ 1207] Overall Loss 0.224526 Objective Loss 0.224526 LR 0.000250 Time 0.020647 -2023-02-13 18:35:14,357 - Epoch: [175][ 390/ 1207] Overall Loss 0.224302 Objective Loss 0.224302 LR 0.000250 Time 0.020601 -2023-02-13 18:35:14,546 - Epoch: [175][ 400/ 1207] Overall Loss 0.224422 Objective Loss 0.224422 LR 0.000250 Time 0.020556 -2023-02-13 18:35:14,735 - Epoch: [175][ 410/ 1207] Overall Loss 0.224600 Objective Loss 0.224600 LR 0.000250 Time 0.020514 -2023-02-13 18:35:14,923 - Epoch: [175][ 420/ 1207] Overall Loss 0.224568 Objective Loss 0.224568 LR 0.000250 Time 0.020474 -2023-02-13 18:35:15,111 - Epoch: [175][ 430/ 1207] Overall Loss 0.224807 Objective Loss 0.224807 LR 0.000250 Time 0.020435 -2023-02-13 18:35:15,300 - Epoch: [175][ 440/ 1207] Overall Loss 0.224313 Objective Loss 0.224313 LR 0.000250 Time 0.020399 -2023-02-13 18:35:15,489 - Epoch: [175][ 450/ 1207] Overall Loss 0.224182 Objective Loss 0.224182 LR 0.000250 Time 0.020365 -2023-02-13 18:35:15,678 - Epoch: [175][ 460/ 1207] Overall Loss 0.224562 Objective Loss 0.224562 LR 0.000250 Time 0.020332 -2023-02-13 18:35:15,867 - Epoch: [175][ 470/ 1207] Overall Loss 0.224576 Objective Loss 0.224576 LR 0.000250 Time 0.020300 -2023-02-13 18:35:16,057 - Epoch: [175][ 480/ 1207] Overall Loss 0.224929 Objective Loss 0.224929 LR 0.000250 Time 0.020271 -2023-02-13 18:35:16,244 - Epoch: [175][ 490/ 1207] Overall Loss 0.224835 Objective Loss 0.224835 LR 0.000250 Time 0.020240 -2023-02-13 18:35:16,433 - Epoch: [175][ 500/ 1207] Overall Loss 0.225460 Objective Loss 0.225460 LR 0.000250 Time 0.020212 -2023-02-13 18:35:16,621 - Epoch: [175][ 510/ 1207] Overall Loss 0.225443 Objective Loss 0.225443 LR 0.000250 Time 0.020184 -2023-02-13 18:35:16,810 - Epoch: [175][ 520/ 1207] Overall Loss 0.225060 Objective Loss 0.225060 LR 0.000250 Time 0.020157 -2023-02-13 18:35:16,999 - Epoch: [175][ 530/ 1207] Overall Loss 0.224895 Objective Loss 0.224895 LR 0.000250 Time 0.020133 -2023-02-13 18:35:17,187 - Epoch: [175][ 540/ 1207] Overall Loss 0.225007 Objective Loss 0.225007 LR 0.000250 Time 0.020108 -2023-02-13 18:35:17,376 - Epoch: [175][ 550/ 1207] Overall Loss 0.225019 Objective Loss 0.225019 LR 0.000250 Time 0.020085 -2023-02-13 18:35:17,564 - Epoch: [175][ 560/ 1207] Overall Loss 0.225151 Objective Loss 0.225151 LR 0.000250 Time 0.020062 -2023-02-13 18:35:17,753 - Epoch: [175][ 570/ 1207] Overall Loss 0.225587 Objective Loss 0.225587 LR 0.000250 Time 0.020040 -2023-02-13 18:35:17,942 - Epoch: [175][ 580/ 1207] Overall Loss 0.225526 Objective Loss 0.225526 LR 0.000250 Time 0.020020 -2023-02-13 18:35:18,130 - Epoch: [175][ 590/ 1207] Overall Loss 0.225480 Objective Loss 0.225480 LR 0.000250 Time 0.019999 -2023-02-13 18:35:18,319 - Epoch: [175][ 600/ 1207] Overall Loss 0.225358 Objective Loss 0.225358 LR 0.000250 Time 0.019981 -2023-02-13 18:35:18,510 - Epoch: [175][ 610/ 1207] Overall Loss 0.224934 Objective Loss 0.224934 LR 0.000250 Time 0.019965 -2023-02-13 18:35:18,698 - Epoch: [175][ 620/ 1207] Overall Loss 0.224629 Objective Loss 0.224629 LR 0.000250 Time 0.019946 -2023-02-13 18:35:18,887 - Epoch: [175][ 630/ 1207] Overall Loss 0.224674 Objective Loss 0.224674 LR 0.000250 Time 0.019928 -2023-02-13 18:35:19,075 - Epoch: [175][ 640/ 1207] Overall Loss 0.224951 Objective Loss 0.224951 LR 0.000250 Time 0.019911 -2023-02-13 18:35:19,265 - Epoch: [175][ 650/ 1207] Overall Loss 0.224860 Objective Loss 0.224860 LR 0.000250 Time 0.019895 -2023-02-13 18:35:19,454 - Epoch: [175][ 660/ 1207] Overall Loss 0.224690 Objective Loss 0.224690 LR 0.000250 Time 0.019880 -2023-02-13 18:35:19,643 - Epoch: [175][ 670/ 1207] Overall Loss 0.224725 Objective Loss 0.224725 LR 0.000250 Time 0.019865 -2023-02-13 18:35:19,831 - Epoch: [175][ 680/ 1207] Overall Loss 0.224601 Objective Loss 0.224601 LR 0.000250 Time 0.019849 -2023-02-13 18:35:20,021 - Epoch: [175][ 690/ 1207] Overall Loss 0.224266 Objective Loss 0.224266 LR 0.000250 Time 0.019836 -2023-02-13 18:35:20,209 - Epoch: [175][ 700/ 1207] Overall Loss 0.224051 Objective Loss 0.224051 LR 0.000250 Time 0.019821 -2023-02-13 18:35:20,398 - Epoch: [175][ 710/ 1207] Overall Loss 0.224062 Objective Loss 0.224062 LR 0.000250 Time 0.019808 -2023-02-13 18:35:20,586 - Epoch: [175][ 720/ 1207] Overall Loss 0.224307 Objective Loss 0.224307 LR 0.000250 Time 0.019793 -2023-02-13 18:35:20,775 - Epoch: [175][ 730/ 1207] Overall Loss 0.224449 Objective Loss 0.224449 LR 0.000250 Time 0.019780 -2023-02-13 18:35:20,965 - Epoch: [175][ 740/ 1207] Overall Loss 0.224397 Objective Loss 0.224397 LR 0.000250 Time 0.019768 -2023-02-13 18:35:21,153 - Epoch: [175][ 750/ 1207] Overall Loss 0.224394 Objective Loss 0.224394 LR 0.000250 Time 0.019755 -2023-02-13 18:35:21,341 - Epoch: [175][ 760/ 1207] Overall Loss 0.224376 Objective Loss 0.224376 LR 0.000250 Time 0.019742 -2023-02-13 18:35:21,529 - Epoch: [175][ 770/ 1207] Overall Loss 0.224221 Objective Loss 0.224221 LR 0.000250 Time 0.019730 -2023-02-13 18:35:21,718 - Epoch: [175][ 780/ 1207] Overall Loss 0.224420 Objective Loss 0.224420 LR 0.000250 Time 0.019718 -2023-02-13 18:35:21,907 - Epoch: [175][ 790/ 1207] Overall Loss 0.224628 Objective Loss 0.224628 LR 0.000250 Time 0.019707 -2023-02-13 18:35:22,096 - Epoch: [175][ 800/ 1207] Overall Loss 0.224846 Objective Loss 0.224846 LR 0.000250 Time 0.019697 -2023-02-13 18:35:22,284 - Epoch: [175][ 810/ 1207] Overall Loss 0.225032 Objective Loss 0.225032 LR 0.000250 Time 0.019686 -2023-02-13 18:35:22,473 - Epoch: [175][ 820/ 1207] Overall Loss 0.225028 Objective Loss 0.225028 LR 0.000250 Time 0.019675 -2023-02-13 18:35:22,661 - Epoch: [175][ 830/ 1207] Overall Loss 0.225170 Objective Loss 0.225170 LR 0.000250 Time 0.019665 -2023-02-13 18:35:22,850 - Epoch: [175][ 840/ 1207] Overall Loss 0.225221 Objective Loss 0.225221 LR 0.000250 Time 0.019655 -2023-02-13 18:35:23,039 - Epoch: [175][ 850/ 1207] Overall Loss 0.225229 Objective Loss 0.225229 LR 0.000250 Time 0.019646 -2023-02-13 18:35:23,227 - Epoch: [175][ 860/ 1207] Overall Loss 0.225239 Objective Loss 0.225239 LR 0.000250 Time 0.019635 -2023-02-13 18:35:23,416 - Epoch: [175][ 870/ 1207] Overall Loss 0.225140 Objective Loss 0.225140 LR 0.000250 Time 0.019626 -2023-02-13 18:35:23,604 - Epoch: [175][ 880/ 1207] Overall Loss 0.225280 Objective Loss 0.225280 LR 0.000250 Time 0.019617 -2023-02-13 18:35:23,793 - Epoch: [175][ 890/ 1207] Overall Loss 0.225611 Objective Loss 0.225611 LR 0.000250 Time 0.019609 -2023-02-13 18:35:23,983 - Epoch: [175][ 900/ 1207] Overall Loss 0.225637 Objective Loss 0.225637 LR 0.000250 Time 0.019601 -2023-02-13 18:35:24,172 - Epoch: [175][ 910/ 1207] Overall Loss 0.225656 Objective Loss 0.225656 LR 0.000250 Time 0.019593 -2023-02-13 18:35:24,361 - Epoch: [175][ 920/ 1207] Overall Loss 0.225748 Objective Loss 0.225748 LR 0.000250 Time 0.019585 -2023-02-13 18:35:24,551 - Epoch: [175][ 930/ 1207] Overall Loss 0.225859 Objective Loss 0.225859 LR 0.000250 Time 0.019578 -2023-02-13 18:35:24,742 - Epoch: [175][ 940/ 1207] Overall Loss 0.225896 Objective Loss 0.225896 LR 0.000250 Time 0.019573 -2023-02-13 18:35:24,932 - Epoch: [175][ 950/ 1207] Overall Loss 0.225887 Objective Loss 0.225887 LR 0.000250 Time 0.019567 -2023-02-13 18:35:25,124 - Epoch: [175][ 960/ 1207] Overall Loss 0.225937 Objective Loss 0.225937 LR 0.000250 Time 0.019562 -2023-02-13 18:35:25,315 - Epoch: [175][ 970/ 1207] Overall Loss 0.225722 Objective Loss 0.225722 LR 0.000250 Time 0.019557 -2023-02-13 18:35:25,508 - Epoch: [175][ 980/ 1207] Overall Loss 0.225873 Objective Loss 0.225873 LR 0.000250 Time 0.019554 -2023-02-13 18:35:25,699 - Epoch: [175][ 990/ 1207] Overall Loss 0.225767 Objective Loss 0.225767 LR 0.000250 Time 0.019549 -2023-02-13 18:35:25,890 - Epoch: [175][ 1000/ 1207] Overall Loss 0.225875 Objective Loss 0.225875 LR 0.000250 Time 0.019544 -2023-02-13 18:35:26,081 - Epoch: [175][ 1010/ 1207] Overall Loss 0.225868 Objective Loss 0.225868 LR 0.000250 Time 0.019540 -2023-02-13 18:35:26,272 - Epoch: [175][ 1020/ 1207] Overall Loss 0.225676 Objective Loss 0.225676 LR 0.000250 Time 0.019534 -2023-02-13 18:35:26,464 - Epoch: [175][ 1030/ 1207] Overall Loss 0.225615 Objective Loss 0.225615 LR 0.000250 Time 0.019531 -2023-02-13 18:35:26,655 - Epoch: [175][ 1040/ 1207] Overall Loss 0.225683 Objective Loss 0.225683 LR 0.000250 Time 0.019526 -2023-02-13 18:35:26,845 - Epoch: [175][ 1050/ 1207] Overall Loss 0.225727 Objective Loss 0.225727 LR 0.000250 Time 0.019521 -2023-02-13 18:35:27,038 - Epoch: [175][ 1060/ 1207] Overall Loss 0.225744 Objective Loss 0.225744 LR 0.000250 Time 0.019518 -2023-02-13 18:35:27,228 - Epoch: [175][ 1070/ 1207] Overall Loss 0.225511 Objective Loss 0.225511 LR 0.000250 Time 0.019514 -2023-02-13 18:35:27,420 - Epoch: [175][ 1080/ 1207] Overall Loss 0.225406 Objective Loss 0.225406 LR 0.000250 Time 0.019511 -2023-02-13 18:35:27,612 - Epoch: [175][ 1090/ 1207] Overall Loss 0.225469 Objective Loss 0.225469 LR 0.000250 Time 0.019507 -2023-02-13 18:35:27,803 - Epoch: [175][ 1100/ 1207] Overall Loss 0.225429 Objective Loss 0.225429 LR 0.000250 Time 0.019503 -2023-02-13 18:35:27,995 - Epoch: [175][ 1110/ 1207] Overall Loss 0.225544 Objective Loss 0.225544 LR 0.000250 Time 0.019500 -2023-02-13 18:35:28,188 - Epoch: [175][ 1120/ 1207] Overall Loss 0.225417 Objective Loss 0.225417 LR 0.000250 Time 0.019498 -2023-02-13 18:35:28,382 - Epoch: [175][ 1130/ 1207] Overall Loss 0.225336 Objective Loss 0.225336 LR 0.000250 Time 0.019497 -2023-02-13 18:35:28,575 - Epoch: [175][ 1140/ 1207] Overall Loss 0.225368 Objective Loss 0.225368 LR 0.000250 Time 0.019494 -2023-02-13 18:35:28,768 - Epoch: [175][ 1150/ 1207] Overall Loss 0.225490 Objective Loss 0.225490 LR 0.000250 Time 0.019492 -2023-02-13 18:35:28,961 - Epoch: [175][ 1160/ 1207] Overall Loss 0.225469 Objective Loss 0.225469 LR 0.000250 Time 0.019491 -2023-02-13 18:35:29,155 - Epoch: [175][ 1170/ 1207] Overall Loss 0.225426 Objective Loss 0.225426 LR 0.000250 Time 0.019489 -2023-02-13 18:35:29,347 - Epoch: [175][ 1180/ 1207] Overall Loss 0.225306 Objective Loss 0.225306 LR 0.000250 Time 0.019487 -2023-02-13 18:35:29,541 - Epoch: [175][ 1190/ 1207] Overall Loss 0.225225 Objective Loss 0.225225 LR 0.000250 Time 0.019486 -2023-02-13 18:35:29,789 - Epoch: [175][ 1200/ 1207] Overall Loss 0.225214 Objective Loss 0.225214 LR 0.000250 Time 0.019530 -2023-02-13 18:35:29,903 - Epoch: [175][ 1207/ 1207] Overall Loss 0.225165 Objective Loss 0.225165 Top1 86.280488 Top5 99.085366 LR 0.000250 Time 0.019511 -2023-02-13 18:35:29,975 - --- validate (epoch=175)----------- -2023-02-13 18:35:29,976 - 34311 samples (256 per mini-batch) -2023-02-13 18:35:30,373 - Epoch: [175][ 10/ 135] Loss 0.311340 Top1 83.867188 Top5 98.242188 -2023-02-13 18:35:30,506 - Epoch: [175][ 20/ 135] Loss 0.290315 Top1 84.531250 Top5 98.125000 -2023-02-13 18:35:30,638 - Epoch: [175][ 30/ 135] Loss 0.288021 Top1 84.934896 Top5 97.942708 -2023-02-13 18:35:30,769 - Epoch: [175][ 40/ 135] Loss 0.295308 Top1 84.414062 Top5 97.949219 -2023-02-13 18:35:30,897 - Epoch: [175][ 50/ 135] Loss 0.291366 Top1 84.656250 Top5 97.921875 -2023-02-13 18:35:31,031 - Epoch: [175][ 60/ 135] Loss 0.294114 Top1 84.700521 Top5 97.962240 -2023-02-13 18:35:31,156 - Epoch: [175][ 70/ 135] Loss 0.295775 Top1 84.760045 Top5 98.013393 -2023-02-13 18:35:31,280 - Epoch: [175][ 80/ 135] Loss 0.294172 Top1 84.873047 Top5 98.037109 -2023-02-13 18:35:31,407 - Epoch: [175][ 90/ 135] Loss 0.296180 Top1 84.930556 Top5 97.977431 -2023-02-13 18:35:31,537 - Epoch: [175][ 100/ 135] Loss 0.296503 Top1 85.035156 Top5 97.929688 -2023-02-13 18:35:31,667 - Epoch: [175][ 110/ 135] Loss 0.299255 Top1 85.042614 Top5 97.950994 -2023-02-13 18:35:31,798 - Epoch: [175][ 120/ 135] Loss 0.301800 Top1 84.967448 Top5 97.916667 -2023-02-13 18:35:31,930 - Epoch: [175][ 130/ 135] Loss 0.302592 Top1 84.945913 Top5 97.914663 -2023-02-13 18:35:31,976 - Epoch: [175][ 135/ 135] Loss 0.300953 Top1 84.961091 Top5 97.895719 -2023-02-13 18:35:32,059 - ==> Top1: 84.961 Top5: 97.896 Loss: 0.301 - -2023-02-13 18:35:32,060 - ==> Confusion: -[[ 863 3 10 1 7 1 0 1 6 45 0 3 0 4 4 3 3 1 1 3 8] - [ 3 943 1 3 13 25 1 12 2 1 1 0 1 0 0 2 5 1 5 4 10] - [ 5 4 968 11 6 1 14 11 2 1 1 4 1 3 3 2 3 1 8 2 7] - [ 4 0 27 914 2 4 1 1 1 2 9 1 6 0 18 1 2 3 13 0 7] - [ 9 8 0 1 1000 6 1 2 2 2 0 6 3 5 8 6 2 1 0 2 2] - [ 3 18 2 2 4 965 4 16 1 6 2 5 3 17 0 3 4 1 4 5 5] - [ 3 3 19 2 0 3 1040 4 0 2 2 0 3 3 0 4 0 1 2 6 2] - [ 2 10 11 2 2 23 2 921 0 2 1 2 2 2 0 0 2 3 23 7 7] - [ 14 1 1 1 2 0 1 0 903 42 5 2 1 11 19 1 0 0 5 0 0] - [ 69 1 2 1 9 0 0 0 30 871 1 1 0 15 5 0 0 1 2 0 4] - [ 2 1 3 8 2 3 2 1 14 2 984 1 1 6 6 0 1 1 7 0 6] - [ 3 3 2 0 1 9 1 5 3 3 0 911 27 9 3 5 2 7 4 5 2] - [ 0 0 0 11 1 2 0 0 4 3 0 20 880 1 1 4 4 19 0 0 9] - [ 6 2 4 0 7 6 1 1 10 19 8 2 1 940 8 3 2 2 1 0 1] - [ 2 1 2 17 5 3 0 1 21 7 1 0 2 1 1009 0 2 4 6 0 8] - [ 6 2 9 1 6 1 5 1 0 0 0 7 4 3 1 965 11 10 0 6 8] - [ 3 7 2 3 7 1 0 0 2 1 0 0 0 1 2 9 1005 2 0 3 13] - [ 4 4 0 5 0 2 2 0 0 0 2 9 12 1 0 13 1 986 0 2 8] - [ 5 4 7 7 0 0 0 16 4 1 4 1 4 0 11 1 0 3 1014 2 2] - [ 0 3 0 2 1 4 8 11 1 0 0 14 2 1 1 8 3 3 0 1079 7] - [ 164 227 266 101 171 187 78 158 106 86 163 97 320 324 182 89 200 112 181 232 9990]] - -2023-02-13 18:35:32,062 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:35:32,062 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:35:32,067 - - -2023-02-13 18:35:32,068 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:35:33,059 - Epoch: [176][ 10/ 1207] Overall Loss 0.249280 Objective Loss 0.249280 LR 0.000250 Time 0.099097 -2023-02-13 18:35:33,249 - Epoch: [176][ 20/ 1207] Overall Loss 0.232922 Objective Loss 0.232922 LR 0.000250 Time 0.058999 -2023-02-13 18:35:33,438 - Epoch: [176][ 30/ 1207] Overall Loss 0.227124 Objective Loss 0.227124 LR 0.000250 Time 0.045620 -2023-02-13 18:35:33,628 - Epoch: [176][ 40/ 1207] Overall Loss 0.221928 Objective Loss 0.221928 LR 0.000250 Time 0.038956 -2023-02-13 18:35:33,818 - Epoch: [176][ 50/ 1207] Overall Loss 0.224864 Objective Loss 0.224864 LR 0.000250 Time 0.034957 -2023-02-13 18:35:34,008 - Epoch: [176][ 60/ 1207] Overall Loss 0.223362 Objective Loss 0.223362 LR 0.000250 Time 0.032299 -2023-02-13 18:35:34,198 - Epoch: [176][ 70/ 1207] Overall Loss 0.221242 Objective Loss 0.221242 LR 0.000250 Time 0.030390 -2023-02-13 18:35:34,388 - Epoch: [176][ 80/ 1207] Overall Loss 0.220921 Objective Loss 0.220921 LR 0.000250 Time 0.028969 -2023-02-13 18:35:34,578 - Epoch: [176][ 90/ 1207] Overall Loss 0.224519 Objective Loss 0.224519 LR 0.000250 Time 0.027855 -2023-02-13 18:35:34,769 - Epoch: [176][ 100/ 1207] Overall Loss 0.220537 Objective Loss 0.220537 LR 0.000250 Time 0.026971 -2023-02-13 18:35:34,959 - Epoch: [176][ 110/ 1207] Overall Loss 0.219207 Objective Loss 0.219207 LR 0.000250 Time 0.026244 -2023-02-13 18:35:35,148 - Epoch: [176][ 120/ 1207] Overall Loss 0.219217 Objective Loss 0.219217 LR 0.000250 Time 0.025634 -2023-02-13 18:35:35,338 - Epoch: [176][ 130/ 1207] Overall Loss 0.218918 Objective Loss 0.218918 LR 0.000250 Time 0.025121 -2023-02-13 18:35:35,529 - Epoch: [176][ 140/ 1207] Overall Loss 0.219570 Objective Loss 0.219570 LR 0.000250 Time 0.024685 -2023-02-13 18:35:35,719 - Epoch: [176][ 150/ 1207] Overall Loss 0.220027 Objective Loss 0.220027 LR 0.000250 Time 0.024305 -2023-02-13 18:35:35,911 - Epoch: [176][ 160/ 1207] Overall Loss 0.220526 Objective Loss 0.220526 LR 0.000250 Time 0.023983 -2023-02-13 18:35:36,102 - Epoch: [176][ 170/ 1207] Overall Loss 0.221158 Objective Loss 0.221158 LR 0.000250 Time 0.023692 -2023-02-13 18:35:36,292 - Epoch: [176][ 180/ 1207] Overall Loss 0.220254 Objective Loss 0.220254 LR 0.000250 Time 0.023429 -2023-02-13 18:35:36,483 - Epoch: [176][ 190/ 1207] Overall Loss 0.220095 Objective Loss 0.220095 LR 0.000250 Time 0.023198 -2023-02-13 18:35:36,673 - Epoch: [176][ 200/ 1207] Overall Loss 0.219470 Objective Loss 0.219470 LR 0.000250 Time 0.022990 -2023-02-13 18:35:36,863 - Epoch: [176][ 210/ 1207] Overall Loss 0.219272 Objective Loss 0.219272 LR 0.000250 Time 0.022796 -2023-02-13 18:35:37,054 - Epoch: [176][ 220/ 1207] Overall Loss 0.220082 Objective Loss 0.220082 LR 0.000250 Time 0.022625 -2023-02-13 18:35:37,243 - Epoch: [176][ 230/ 1207] Overall Loss 0.219376 Objective Loss 0.219376 LR 0.000250 Time 0.022464 -2023-02-13 18:35:37,434 - Epoch: [176][ 240/ 1207] Overall Loss 0.219671 Objective Loss 0.219671 LR 0.000250 Time 0.022323 -2023-02-13 18:35:37,625 - Epoch: [176][ 250/ 1207] Overall Loss 0.220165 Objective Loss 0.220165 LR 0.000250 Time 0.022191 -2023-02-13 18:35:37,817 - Epoch: [176][ 260/ 1207] Overall Loss 0.220002 Objective Loss 0.220002 LR 0.000250 Time 0.022075 -2023-02-13 18:35:38,006 - Epoch: [176][ 270/ 1207] Overall Loss 0.220237 Objective Loss 0.220237 LR 0.000250 Time 0.021957 -2023-02-13 18:35:38,196 - Epoch: [176][ 280/ 1207] Overall Loss 0.220475 Objective Loss 0.220475 LR 0.000250 Time 0.021851 -2023-02-13 18:35:38,387 - Epoch: [176][ 290/ 1207] Overall Loss 0.220655 Objective Loss 0.220655 LR 0.000250 Time 0.021752 -2023-02-13 18:35:38,578 - Epoch: [176][ 300/ 1207] Overall Loss 0.220257 Objective Loss 0.220257 LR 0.000250 Time 0.021664 -2023-02-13 18:35:38,768 - Epoch: [176][ 310/ 1207] Overall Loss 0.220741 Objective Loss 0.220741 LR 0.000250 Time 0.021576 -2023-02-13 18:35:38,959 - Epoch: [176][ 320/ 1207] Overall Loss 0.221666 Objective Loss 0.221666 LR 0.000250 Time 0.021497 -2023-02-13 18:35:39,150 - Epoch: [176][ 330/ 1207] Overall Loss 0.222534 Objective Loss 0.222534 LR 0.000250 Time 0.021424 -2023-02-13 18:35:39,341 - Epoch: [176][ 340/ 1207] Overall Loss 0.223095 Objective Loss 0.223095 LR 0.000250 Time 0.021354 -2023-02-13 18:35:39,532 - Epoch: [176][ 350/ 1207] Overall Loss 0.222716 Objective Loss 0.222716 LR 0.000250 Time 0.021289 -2023-02-13 18:35:39,723 - Epoch: [176][ 360/ 1207] Overall Loss 0.223111 Objective Loss 0.223111 LR 0.000250 Time 0.021229 -2023-02-13 18:35:39,914 - Epoch: [176][ 370/ 1207] Overall Loss 0.223196 Objective Loss 0.223196 LR 0.000250 Time 0.021169 -2023-02-13 18:35:40,105 - Epoch: [176][ 380/ 1207] Overall Loss 0.223155 Objective Loss 0.223155 LR 0.000250 Time 0.021115 -2023-02-13 18:35:40,296 - Epoch: [176][ 390/ 1207] Overall Loss 0.223761 Objective Loss 0.223761 LR 0.000250 Time 0.021061 -2023-02-13 18:35:40,488 - Epoch: [176][ 400/ 1207] Overall Loss 0.223444 Objective Loss 0.223444 LR 0.000250 Time 0.021014 -2023-02-13 18:35:40,679 - Epoch: [176][ 410/ 1207] Overall Loss 0.223339 Objective Loss 0.223339 LR 0.000250 Time 0.020965 -2023-02-13 18:35:40,870 - Epoch: [176][ 420/ 1207] Overall Loss 0.223753 Objective Loss 0.223753 LR 0.000250 Time 0.020922 -2023-02-13 18:35:41,062 - Epoch: [176][ 430/ 1207] Overall Loss 0.223424 Objective Loss 0.223424 LR 0.000250 Time 0.020879 -2023-02-13 18:35:41,252 - Epoch: [176][ 440/ 1207] Overall Loss 0.223294 Objective Loss 0.223294 LR 0.000250 Time 0.020837 -2023-02-13 18:35:41,443 - Epoch: [176][ 450/ 1207] Overall Loss 0.223445 Objective Loss 0.223445 LR 0.000250 Time 0.020796 -2023-02-13 18:35:41,633 - Epoch: [176][ 460/ 1207] Overall Loss 0.223234 Objective Loss 0.223234 LR 0.000250 Time 0.020758 -2023-02-13 18:35:41,823 - Epoch: [176][ 470/ 1207] Overall Loss 0.223202 Objective Loss 0.223202 LR 0.000250 Time 0.020720 -2023-02-13 18:35:42,016 - Epoch: [176][ 480/ 1207] Overall Loss 0.223469 Objective Loss 0.223469 LR 0.000250 Time 0.020689 -2023-02-13 18:35:42,207 - Epoch: [176][ 490/ 1207] Overall Loss 0.223327 Objective Loss 0.223327 LR 0.000250 Time 0.020656 -2023-02-13 18:35:42,398 - Epoch: [176][ 500/ 1207] Overall Loss 0.223398 Objective Loss 0.223398 LR 0.000250 Time 0.020623 -2023-02-13 18:35:42,588 - Epoch: [176][ 510/ 1207] Overall Loss 0.222905 Objective Loss 0.222905 LR 0.000250 Time 0.020591 -2023-02-13 18:35:42,779 - Epoch: [176][ 520/ 1207] Overall Loss 0.223175 Objective Loss 0.223175 LR 0.000250 Time 0.020562 -2023-02-13 18:35:42,970 - Epoch: [176][ 530/ 1207] Overall Loss 0.223453 Objective Loss 0.223453 LR 0.000250 Time 0.020534 -2023-02-13 18:35:43,161 - Epoch: [176][ 540/ 1207] Overall Loss 0.223415 Objective Loss 0.223415 LR 0.000250 Time 0.020507 -2023-02-13 18:35:43,353 - Epoch: [176][ 550/ 1207] Overall Loss 0.223816 Objective Loss 0.223816 LR 0.000250 Time 0.020481 -2023-02-13 18:35:43,544 - Epoch: [176][ 560/ 1207] Overall Loss 0.224075 Objective Loss 0.224075 LR 0.000250 Time 0.020456 -2023-02-13 18:35:43,734 - Epoch: [176][ 570/ 1207] Overall Loss 0.224281 Objective Loss 0.224281 LR 0.000250 Time 0.020431 -2023-02-13 18:35:43,925 - Epoch: [176][ 580/ 1207] Overall Loss 0.224369 Objective Loss 0.224369 LR 0.000250 Time 0.020407 -2023-02-13 18:35:44,116 - Epoch: [176][ 590/ 1207] Overall Loss 0.224860 Objective Loss 0.224860 LR 0.000250 Time 0.020384 -2023-02-13 18:35:44,306 - Epoch: [176][ 600/ 1207] Overall Loss 0.224892 Objective Loss 0.224892 LR 0.000250 Time 0.020361 -2023-02-13 18:35:44,498 - Epoch: [176][ 610/ 1207] Overall Loss 0.224543 Objective Loss 0.224543 LR 0.000250 Time 0.020341 -2023-02-13 18:35:44,690 - Epoch: [176][ 620/ 1207] Overall Loss 0.224466 Objective Loss 0.224466 LR 0.000250 Time 0.020322 -2023-02-13 18:35:44,882 - Epoch: [176][ 630/ 1207] Overall Loss 0.224391 Objective Loss 0.224391 LR 0.000250 Time 0.020302 -2023-02-13 18:35:45,074 - Epoch: [176][ 640/ 1207] Overall Loss 0.224380 Objective Loss 0.224380 LR 0.000250 Time 0.020286 -2023-02-13 18:35:45,266 - Epoch: [176][ 650/ 1207] Overall Loss 0.224313 Objective Loss 0.224313 LR 0.000250 Time 0.020268 -2023-02-13 18:35:45,459 - Epoch: [176][ 660/ 1207] Overall Loss 0.224319 Objective Loss 0.224319 LR 0.000250 Time 0.020253 -2023-02-13 18:35:45,651 - Epoch: [176][ 670/ 1207] Overall Loss 0.224476 Objective Loss 0.224476 LR 0.000250 Time 0.020236 -2023-02-13 18:35:45,843 - Epoch: [176][ 680/ 1207] Overall Loss 0.224353 Objective Loss 0.224353 LR 0.000250 Time 0.020221 -2023-02-13 18:35:46,035 - Epoch: [176][ 690/ 1207] Overall Loss 0.224501 Objective Loss 0.224501 LR 0.000250 Time 0.020205 -2023-02-13 18:35:46,227 - Epoch: [176][ 700/ 1207] Overall Loss 0.224234 Objective Loss 0.224234 LR 0.000250 Time 0.020191 -2023-02-13 18:35:46,419 - Epoch: [176][ 710/ 1207] Overall Loss 0.224523 Objective Loss 0.224523 LR 0.000250 Time 0.020176 -2023-02-13 18:35:46,611 - Epoch: [176][ 720/ 1207] Overall Loss 0.224792 Objective Loss 0.224792 LR 0.000250 Time 0.020162 -2023-02-13 18:35:46,802 - Epoch: [176][ 730/ 1207] Overall Loss 0.224692 Objective Loss 0.224692 LR 0.000250 Time 0.020147 -2023-02-13 18:35:46,995 - Epoch: [176][ 740/ 1207] Overall Loss 0.224398 Objective Loss 0.224398 LR 0.000250 Time 0.020134 -2023-02-13 18:35:47,186 - Epoch: [176][ 750/ 1207] Overall Loss 0.224268 Objective Loss 0.224268 LR 0.000250 Time 0.020120 -2023-02-13 18:35:47,378 - Epoch: [176][ 760/ 1207] Overall Loss 0.224241 Objective Loss 0.224241 LR 0.000250 Time 0.020107 -2023-02-13 18:35:47,568 - Epoch: [176][ 770/ 1207] Overall Loss 0.223826 Objective Loss 0.223826 LR 0.000250 Time 0.020094 -2023-02-13 18:35:47,761 - Epoch: [176][ 780/ 1207] Overall Loss 0.223734 Objective Loss 0.223734 LR 0.000250 Time 0.020082 -2023-02-13 18:35:47,952 - Epoch: [176][ 790/ 1207] Overall Loss 0.223667 Objective Loss 0.223667 LR 0.000250 Time 0.020069 -2023-02-13 18:35:48,144 - Epoch: [176][ 800/ 1207] Overall Loss 0.223450 Objective Loss 0.223450 LR 0.000250 Time 0.020059 -2023-02-13 18:35:48,335 - Epoch: [176][ 810/ 1207] Overall Loss 0.223443 Objective Loss 0.223443 LR 0.000250 Time 0.020046 -2023-02-13 18:35:48,527 - Epoch: [176][ 820/ 1207] Overall Loss 0.223412 Objective Loss 0.223412 LR 0.000250 Time 0.020036 -2023-02-13 18:35:48,718 - Epoch: [176][ 830/ 1207] Overall Loss 0.223430 Objective Loss 0.223430 LR 0.000250 Time 0.020024 -2023-02-13 18:35:48,911 - Epoch: [176][ 840/ 1207] Overall Loss 0.223742 Objective Loss 0.223742 LR 0.000250 Time 0.020014 -2023-02-13 18:35:49,102 - Epoch: [176][ 850/ 1207] Overall Loss 0.223588 Objective Loss 0.223588 LR 0.000250 Time 0.020003 -2023-02-13 18:35:49,294 - Epoch: [176][ 860/ 1207] Overall Loss 0.223403 Objective Loss 0.223403 LR 0.000250 Time 0.019994 -2023-02-13 18:35:49,485 - Epoch: [176][ 870/ 1207] Overall Loss 0.223457 Objective Loss 0.223457 LR 0.000250 Time 0.019983 -2023-02-13 18:35:49,678 - Epoch: [176][ 880/ 1207] Overall Loss 0.223406 Objective Loss 0.223406 LR 0.000250 Time 0.019975 -2023-02-13 18:35:49,868 - Epoch: [176][ 890/ 1207] Overall Loss 0.223464 Objective Loss 0.223464 LR 0.000250 Time 0.019963 -2023-02-13 18:35:50,057 - Epoch: [176][ 900/ 1207] Overall Loss 0.223280 Objective Loss 0.223280 LR 0.000250 Time 0.019951 -2023-02-13 18:35:50,247 - Epoch: [176][ 910/ 1207] Overall Loss 0.223305 Objective Loss 0.223305 LR 0.000250 Time 0.019940 -2023-02-13 18:35:50,436 - Epoch: [176][ 920/ 1207] Overall Loss 0.223009 Objective Loss 0.223009 LR 0.000250 Time 0.019929 -2023-02-13 18:35:50,625 - Epoch: [176][ 930/ 1207] Overall Loss 0.222987 Objective Loss 0.222987 LR 0.000250 Time 0.019917 -2023-02-13 18:35:50,814 - Epoch: [176][ 940/ 1207] Overall Loss 0.223035 Objective Loss 0.223035 LR 0.000250 Time 0.019906 -2023-02-13 18:35:51,004 - Epoch: [176][ 950/ 1207] Overall Loss 0.222964 Objective Loss 0.222964 LR 0.000250 Time 0.019896 -2023-02-13 18:35:51,195 - Epoch: [176][ 960/ 1207] Overall Loss 0.223004 Objective Loss 0.223004 LR 0.000250 Time 0.019887 -2023-02-13 18:35:51,385 - Epoch: [176][ 970/ 1207] Overall Loss 0.223095 Objective Loss 0.223095 LR 0.000250 Time 0.019877 -2023-02-13 18:35:51,575 - Epoch: [176][ 980/ 1207] Overall Loss 0.223239 Objective Loss 0.223239 LR 0.000250 Time 0.019868 -2023-02-13 18:35:51,764 - Epoch: [176][ 990/ 1207] Overall Loss 0.223289 Objective Loss 0.223289 LR 0.000250 Time 0.019858 -2023-02-13 18:35:51,953 - Epoch: [176][ 1000/ 1207] Overall Loss 0.223266 Objective Loss 0.223266 LR 0.000250 Time 0.019848 -2023-02-13 18:35:52,142 - Epoch: [176][ 1010/ 1207] Overall Loss 0.223322 Objective Loss 0.223322 LR 0.000250 Time 0.019838 -2023-02-13 18:35:52,331 - Epoch: [176][ 1020/ 1207] Overall Loss 0.223338 Objective Loss 0.223338 LR 0.000250 Time 0.019829 -2023-02-13 18:35:52,522 - Epoch: [176][ 1030/ 1207] Overall Loss 0.223256 Objective Loss 0.223256 LR 0.000250 Time 0.019821 -2023-02-13 18:35:52,711 - Epoch: [176][ 1040/ 1207] Overall Loss 0.223325 Objective Loss 0.223325 LR 0.000250 Time 0.019812 -2023-02-13 18:35:52,901 - Epoch: [176][ 1050/ 1207] Overall Loss 0.223191 Objective Loss 0.223191 LR 0.000250 Time 0.019804 -2023-02-13 18:35:53,089 - Epoch: [176][ 1060/ 1207] Overall Loss 0.223361 Objective Loss 0.223361 LR 0.000250 Time 0.019795 -2023-02-13 18:35:53,278 - Epoch: [176][ 1070/ 1207] Overall Loss 0.223473 Objective Loss 0.223473 LR 0.000250 Time 0.019785 -2023-02-13 18:35:53,468 - Epoch: [176][ 1080/ 1207] Overall Loss 0.223475 Objective Loss 0.223475 LR 0.000250 Time 0.019778 -2023-02-13 18:35:53,657 - Epoch: [176][ 1090/ 1207] Overall Loss 0.223538 Objective Loss 0.223538 LR 0.000250 Time 0.019770 -2023-02-13 18:35:53,847 - Epoch: [176][ 1100/ 1207] Overall Loss 0.223547 Objective Loss 0.223547 LR 0.000250 Time 0.019762 -2023-02-13 18:35:54,035 - Epoch: [176][ 1110/ 1207] Overall Loss 0.223601 Objective Loss 0.223601 LR 0.000250 Time 0.019754 -2023-02-13 18:35:54,224 - Epoch: [176][ 1120/ 1207] Overall Loss 0.223416 Objective Loss 0.223416 LR 0.000250 Time 0.019746 -2023-02-13 18:35:54,413 - Epoch: [176][ 1130/ 1207] Overall Loss 0.223480 Objective Loss 0.223480 LR 0.000250 Time 0.019738 -2023-02-13 18:35:54,603 - Epoch: [176][ 1140/ 1207] Overall Loss 0.223674 Objective Loss 0.223674 LR 0.000250 Time 0.019731 -2023-02-13 18:35:54,792 - Epoch: [176][ 1150/ 1207] Overall Loss 0.223581 Objective Loss 0.223581 LR 0.000250 Time 0.019723 -2023-02-13 18:35:54,982 - Epoch: [176][ 1160/ 1207] Overall Loss 0.223471 Objective Loss 0.223471 LR 0.000250 Time 0.019717 -2023-02-13 18:35:55,171 - Epoch: [176][ 1170/ 1207] Overall Loss 0.223427 Objective Loss 0.223427 LR 0.000250 Time 0.019709 -2023-02-13 18:35:55,361 - Epoch: [176][ 1180/ 1207] Overall Loss 0.223516 Objective Loss 0.223516 LR 0.000250 Time 0.019703 -2023-02-13 18:35:55,551 - Epoch: [176][ 1190/ 1207] Overall Loss 0.223675 Objective Loss 0.223675 LR 0.000250 Time 0.019696 -2023-02-13 18:35:55,796 - Epoch: [176][ 1200/ 1207] Overall Loss 0.223868 Objective Loss 0.223868 LR 0.000250 Time 0.019736 -2023-02-13 18:35:55,912 - Epoch: [176][ 1207/ 1207] Overall Loss 0.224046 Objective Loss 0.224046 Top1 88.414634 Top5 99.085366 LR 0.000250 Time 0.019718 -2023-02-13 18:35:55,985 - --- validate (epoch=176)----------- -2023-02-13 18:35:55,985 - 34311 samples (256 per mini-batch) -2023-02-13 18:35:56,391 - Epoch: [176][ 10/ 135] Loss 0.313944 Top1 84.726562 Top5 97.617188 -2023-02-13 18:35:56,522 - Epoch: [176][ 20/ 135] Loss 0.305530 Top1 84.394531 Top5 97.617188 -2023-02-13 18:35:56,650 - Epoch: [176][ 30/ 135] Loss 0.301618 Top1 84.791667 Top5 97.799479 -2023-02-13 18:35:56,775 - Epoch: [176][ 40/ 135] Loss 0.302631 Top1 84.589844 Top5 97.753906 -2023-02-13 18:35:56,904 - Epoch: [176][ 50/ 135] Loss 0.300168 Top1 84.679688 Top5 97.804688 -2023-02-13 18:35:57,034 - Epoch: [176][ 60/ 135] Loss 0.304644 Top1 84.648438 Top5 97.851562 -2023-02-13 18:35:57,163 - Epoch: [176][ 70/ 135] Loss 0.299807 Top1 84.899554 Top5 97.862723 -2023-02-13 18:35:57,293 - Epoch: [176][ 80/ 135] Loss 0.304679 Top1 84.770508 Top5 97.827148 -2023-02-13 18:35:57,424 - Epoch: [176][ 90/ 135] Loss 0.301474 Top1 84.748264 Top5 97.808160 -2023-02-13 18:35:57,553 - Epoch: [176][ 100/ 135] Loss 0.301350 Top1 84.609375 Top5 97.742188 -2023-02-13 18:35:57,683 - Epoch: [176][ 110/ 135] Loss 0.300645 Top1 84.595170 Top5 97.766335 -2023-02-13 18:35:57,812 - Epoch: [176][ 120/ 135] Loss 0.302509 Top1 84.508464 Top5 97.776693 -2023-02-13 18:35:57,947 - Epoch: [176][ 130/ 135] Loss 0.302467 Top1 84.573317 Top5 97.779447 -2023-02-13 18:35:57,995 - Epoch: [176][ 135/ 135] Loss 0.302456 Top1 84.544315 Top5 97.767480 -2023-02-13 18:35:58,064 - ==> Top1: 84.544 Top5: 97.767 Loss: 0.302 - -2023-02-13 18:35:58,065 - ==> Confusion: -[[ 867 3 9 1 9 4 0 0 4 37 0 4 1 3 6 1 1 6 1 3 7] - [ 2 952 2 1 7 31 1 12 1 2 2 1 1 0 0 1 5 1 4 1 6] - [ 4 8 961 12 1 1 12 11 1 2 3 4 2 4 3 6 1 1 7 2 12] - [ 2 0 21 913 4 5 1 1 2 3 11 0 6 0 13 1 1 5 19 0 8] - [ 10 9 0 1 997 14 1 2 0 2 0 6 1 2 6 4 4 3 0 2 2] - [ 1 12 1 4 5 973 4 13 2 5 1 12 2 13 1 3 6 2 2 4 4] - [ 1 4 22 2 1 8 1031 6 0 1 3 1 1 2 0 2 0 3 2 5 4] - [ 1 6 8 1 0 26 1 946 0 2 0 7 4 1 0 0 2 2 9 6 2] - [ 12 3 1 1 1 0 1 2 907 42 4 2 0 8 14 1 3 0 5 0 2] - [ 83 2 5 0 8 1 0 2 31 848 0 2 0 18 4 0 0 3 0 1 4] - [ 1 1 2 7 1 2 2 6 16 1 987 1 1 8 1 0 1 1 9 0 3] - [ 3 2 2 0 1 12 0 5 0 3 0 925 20 5 0 4 4 12 1 5 1] - [ 2 0 0 9 1 3 0 1 1 1 0 35 875 1 2 7 3 12 0 1 5] - [ 5 3 2 1 7 10 0 2 8 15 7 7 1 937 4 4 4 2 0 0 5] - [ 5 2 0 20 4 5 0 1 20 9 3 1 3 2 993 0 2 6 7 0 9] - [ 3 1 8 1 9 0 4 4 0 0 0 8 9 3 0 965 7 10 0 7 7] - [ 4 8 1 2 7 2 0 1 3 1 0 1 1 1 1 9 1003 4 0 3 9] - [ 4 2 0 5 1 1 2 0 0 0 3 10 16 0 0 14 1 986 0 1 5] - [ 4 5 3 5 0 2 1 24 2 0 6 2 3 0 13 1 0 2 1011 1 1] - [ 0 3 1 1 1 7 4 9 1 0 0 21 3 2 1 7 3 4 0 1072 8] - [ 143 256 245 135 127 261 65 201 96 80 198 116 311 298 158 84 240 119 185 257 9859]] - -2023-02-13 18:35:58,067 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:35:58,067 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:35:58,073 - - -2023-02-13 18:35:58,073 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:35:58,986 - Epoch: [177][ 10/ 1207] Overall Loss 0.234136 Objective Loss 0.234136 LR 0.000250 Time 0.091250 -2023-02-13 18:35:59,200 - Epoch: [177][ 20/ 1207] Overall Loss 0.223537 Objective Loss 0.223537 LR 0.000250 Time 0.056296 -2023-02-13 18:35:59,405 - Epoch: [177][ 30/ 1207] Overall Loss 0.223927 Objective Loss 0.223927 LR 0.000250 Time 0.044344 -2023-02-13 18:35:59,611 - Epoch: [177][ 40/ 1207] Overall Loss 0.223569 Objective Loss 0.223569 LR 0.000250 Time 0.038405 -2023-02-13 18:35:59,816 - Epoch: [177][ 50/ 1207] Overall Loss 0.222976 Objective Loss 0.222976 LR 0.000250 Time 0.034813 -2023-02-13 18:36:00,021 - Epoch: [177][ 60/ 1207] Overall Loss 0.223534 Objective Loss 0.223534 LR 0.000250 Time 0.032426 -2023-02-13 18:36:00,226 - Epoch: [177][ 70/ 1207] Overall Loss 0.219910 Objective Loss 0.219910 LR 0.000250 Time 0.030710 -2023-02-13 18:36:00,431 - Epoch: [177][ 80/ 1207] Overall Loss 0.221148 Objective Loss 0.221148 LR 0.000250 Time 0.029433 -2023-02-13 18:36:00,636 - Epoch: [177][ 90/ 1207] Overall Loss 0.221816 Objective Loss 0.221816 LR 0.000250 Time 0.028435 -2023-02-13 18:36:00,840 - Epoch: [177][ 100/ 1207] Overall Loss 0.222136 Objective Loss 0.222136 LR 0.000250 Time 0.027627 -2023-02-13 18:36:01,046 - Epoch: [177][ 110/ 1207] Overall Loss 0.221749 Objective Loss 0.221749 LR 0.000250 Time 0.026980 -2023-02-13 18:36:01,251 - Epoch: [177][ 120/ 1207] Overall Loss 0.222951 Objective Loss 0.222951 LR 0.000250 Time 0.026445 -2023-02-13 18:36:01,456 - Epoch: [177][ 130/ 1207] Overall Loss 0.223051 Objective Loss 0.223051 LR 0.000250 Time 0.025978 -2023-02-13 18:36:01,662 - Epoch: [177][ 140/ 1207] Overall Loss 0.223455 Objective Loss 0.223455 LR 0.000250 Time 0.025593 -2023-02-13 18:36:01,867 - Epoch: [177][ 150/ 1207] Overall Loss 0.224035 Objective Loss 0.224035 LR 0.000250 Time 0.025249 -2023-02-13 18:36:02,073 - Epoch: [177][ 160/ 1207] Overall Loss 0.223924 Objective Loss 0.223924 LR 0.000250 Time 0.024956 -2023-02-13 18:36:02,277 - Epoch: [177][ 170/ 1207] Overall Loss 0.224689 Objective Loss 0.224689 LR 0.000250 Time 0.024688 -2023-02-13 18:36:02,483 - Epoch: [177][ 180/ 1207] Overall Loss 0.224077 Objective Loss 0.224077 LR 0.000250 Time 0.024458 -2023-02-13 18:36:02,688 - Epoch: [177][ 190/ 1207] Overall Loss 0.224161 Objective Loss 0.224161 LR 0.000250 Time 0.024247 -2023-02-13 18:36:02,894 - Epoch: [177][ 200/ 1207] Overall Loss 0.223633 Objective Loss 0.223633 LR 0.000250 Time 0.024064 -2023-02-13 18:36:03,099 - Epoch: [177][ 210/ 1207] Overall Loss 0.224070 Objective Loss 0.224070 LR 0.000250 Time 0.023890 -2023-02-13 18:36:03,304 - Epoch: [177][ 220/ 1207] Overall Loss 0.223698 Objective Loss 0.223698 LR 0.000250 Time 0.023736 -2023-02-13 18:36:03,509 - Epoch: [177][ 230/ 1207] Overall Loss 0.224795 Objective Loss 0.224795 LR 0.000250 Time 0.023594 -2023-02-13 18:36:03,715 - Epoch: [177][ 240/ 1207] Overall Loss 0.225260 Objective Loss 0.225260 LR 0.000250 Time 0.023468 -2023-02-13 18:36:03,920 - Epoch: [177][ 250/ 1207] Overall Loss 0.225099 Objective Loss 0.225099 LR 0.000250 Time 0.023347 -2023-02-13 18:36:04,126 - Epoch: [177][ 260/ 1207] Overall Loss 0.225624 Objective Loss 0.225624 LR 0.000250 Time 0.023240 -2023-02-13 18:36:04,330 - Epoch: [177][ 270/ 1207] Overall Loss 0.224903 Objective Loss 0.224903 LR 0.000250 Time 0.023133 -2023-02-13 18:36:04,536 - Epoch: [177][ 280/ 1207] Overall Loss 0.224752 Objective Loss 0.224752 LR 0.000250 Time 0.023042 -2023-02-13 18:36:04,740 - Epoch: [177][ 290/ 1207] Overall Loss 0.224407 Objective Loss 0.224407 LR 0.000250 Time 0.022951 -2023-02-13 18:36:04,946 - Epoch: [177][ 300/ 1207] Overall Loss 0.224138 Objective Loss 0.224138 LR 0.000250 Time 0.022870 -2023-02-13 18:36:05,150 - Epoch: [177][ 310/ 1207] Overall Loss 0.223855 Objective Loss 0.223855 LR 0.000250 Time 0.022789 -2023-02-13 18:36:05,355 - Epoch: [177][ 320/ 1207] Overall Loss 0.223828 Objective Loss 0.223828 LR 0.000250 Time 0.022718 -2023-02-13 18:36:05,560 - Epoch: [177][ 330/ 1207] Overall Loss 0.223137 Objective Loss 0.223137 LR 0.000250 Time 0.022649 -2023-02-13 18:36:05,766 - Epoch: [177][ 340/ 1207] Overall Loss 0.222986 Objective Loss 0.222986 LR 0.000250 Time 0.022587 -2023-02-13 18:36:05,971 - Epoch: [177][ 350/ 1207] Overall Loss 0.222849 Objective Loss 0.222849 LR 0.000250 Time 0.022526 -2023-02-13 18:36:06,176 - Epoch: [177][ 360/ 1207] Overall Loss 0.222854 Objective Loss 0.222854 LR 0.000250 Time 0.022470 -2023-02-13 18:36:06,381 - Epoch: [177][ 370/ 1207] Overall Loss 0.223233 Objective Loss 0.223233 LR 0.000250 Time 0.022413 -2023-02-13 18:36:06,586 - Epoch: [177][ 380/ 1207] Overall Loss 0.223396 Objective Loss 0.223396 LR 0.000250 Time 0.022364 -2023-02-13 18:36:06,791 - Epoch: [177][ 390/ 1207] Overall Loss 0.223792 Objective Loss 0.223792 LR 0.000250 Time 0.022314 -2023-02-13 18:36:06,997 - Epoch: [177][ 400/ 1207] Overall Loss 0.223880 Objective Loss 0.223880 LR 0.000250 Time 0.022272 -2023-02-13 18:36:07,202 - Epoch: [177][ 410/ 1207] Overall Loss 0.223944 Objective Loss 0.223944 LR 0.000250 Time 0.022227 -2023-02-13 18:36:07,408 - Epoch: [177][ 420/ 1207] Overall Loss 0.223859 Objective Loss 0.223859 LR 0.000250 Time 0.022186 -2023-02-13 18:36:07,613 - Epoch: [177][ 430/ 1207] Overall Loss 0.223185 Objective Loss 0.223185 LR 0.000250 Time 0.022147 -2023-02-13 18:36:07,819 - Epoch: [177][ 440/ 1207] Overall Loss 0.223270 Objective Loss 0.223270 LR 0.000250 Time 0.022109 -2023-02-13 18:36:08,023 - Epoch: [177][ 450/ 1207] Overall Loss 0.222832 Objective Loss 0.222832 LR 0.000250 Time 0.022072 -2023-02-13 18:36:08,229 - Epoch: [177][ 460/ 1207] Overall Loss 0.223194 Objective Loss 0.223194 LR 0.000250 Time 0.022038 -2023-02-13 18:36:08,433 - Epoch: [177][ 470/ 1207] Overall Loss 0.223742 Objective Loss 0.223742 LR 0.000250 Time 0.022003 -2023-02-13 18:36:08,640 - Epoch: [177][ 480/ 1207] Overall Loss 0.223532 Objective Loss 0.223532 LR 0.000250 Time 0.021974 -2023-02-13 18:36:08,844 - Epoch: [177][ 490/ 1207] Overall Loss 0.223926 Objective Loss 0.223926 LR 0.000250 Time 0.021943 -2023-02-13 18:36:09,050 - Epoch: [177][ 500/ 1207] Overall Loss 0.223633 Objective Loss 0.223633 LR 0.000250 Time 0.021915 -2023-02-13 18:36:09,255 - Epoch: [177][ 510/ 1207] Overall Loss 0.223216 Objective Loss 0.223216 LR 0.000250 Time 0.021887 -2023-02-13 18:36:09,460 - Epoch: [177][ 520/ 1207] Overall Loss 0.223091 Objective Loss 0.223091 LR 0.000250 Time 0.021860 -2023-02-13 18:36:09,666 - Epoch: [177][ 530/ 1207] Overall Loss 0.222707 Objective Loss 0.222707 LR 0.000250 Time 0.021834 -2023-02-13 18:36:09,872 - Epoch: [177][ 540/ 1207] Overall Loss 0.222655 Objective Loss 0.222655 LR 0.000250 Time 0.021810 -2023-02-13 18:36:10,076 - Epoch: [177][ 550/ 1207] Overall Loss 0.222933 Objective Loss 0.222933 LR 0.000250 Time 0.021785 -2023-02-13 18:36:10,283 - Epoch: [177][ 560/ 1207] Overall Loss 0.223019 Objective Loss 0.223019 LR 0.000250 Time 0.021764 -2023-02-13 18:36:10,487 - Epoch: [177][ 570/ 1207] Overall Loss 0.222650 Objective Loss 0.222650 LR 0.000250 Time 0.021740 -2023-02-13 18:36:10,693 - Epoch: [177][ 580/ 1207] Overall Loss 0.222483 Objective Loss 0.222483 LR 0.000250 Time 0.021720 -2023-02-13 18:36:10,898 - Epoch: [177][ 590/ 1207] Overall Loss 0.222996 Objective Loss 0.222996 LR 0.000250 Time 0.021698 -2023-02-13 18:36:11,105 - Epoch: [177][ 600/ 1207] Overall Loss 0.223011 Objective Loss 0.223011 LR 0.000250 Time 0.021681 -2023-02-13 18:36:11,309 - Epoch: [177][ 610/ 1207] Overall Loss 0.222819 Objective Loss 0.222819 LR 0.000250 Time 0.021659 -2023-02-13 18:36:11,515 - Epoch: [177][ 620/ 1207] Overall Loss 0.222684 Objective Loss 0.222684 LR 0.000250 Time 0.021641 -2023-02-13 18:36:11,720 - Epoch: [177][ 630/ 1207] Overall Loss 0.222330 Objective Loss 0.222330 LR 0.000250 Time 0.021624 -2023-02-13 18:36:11,926 - Epoch: [177][ 640/ 1207] Overall Loss 0.222298 Objective Loss 0.222298 LR 0.000250 Time 0.021607 -2023-02-13 18:36:12,130 - Epoch: [177][ 650/ 1207] Overall Loss 0.222329 Objective Loss 0.222329 LR 0.000250 Time 0.021587 -2023-02-13 18:36:12,325 - Epoch: [177][ 660/ 1207] Overall Loss 0.222093 Objective Loss 0.222093 LR 0.000250 Time 0.021555 -2023-02-13 18:36:12,524 - Epoch: [177][ 670/ 1207] Overall Loss 0.222073 Objective Loss 0.222073 LR 0.000250 Time 0.021530 -2023-02-13 18:36:12,719 - Epoch: [177][ 680/ 1207] Overall Loss 0.222212 Objective Loss 0.222212 LR 0.000250 Time 0.021500 -2023-02-13 18:36:12,918 - Epoch: [177][ 690/ 1207] Overall Loss 0.222134 Objective Loss 0.222134 LR 0.000250 Time 0.021476 -2023-02-13 18:36:13,113 - Epoch: [177][ 700/ 1207] Overall Loss 0.222570 Objective Loss 0.222570 LR 0.000250 Time 0.021446 -2023-02-13 18:36:13,311 - Epoch: [177][ 710/ 1207] Overall Loss 0.222492 Objective Loss 0.222492 LR 0.000250 Time 0.021423 -2023-02-13 18:36:13,506 - Epoch: [177][ 720/ 1207] Overall Loss 0.222292 Objective Loss 0.222292 LR 0.000250 Time 0.021396 -2023-02-13 18:36:13,706 - Epoch: [177][ 730/ 1207] Overall Loss 0.222183 Objective Loss 0.222183 LR 0.000250 Time 0.021377 -2023-02-13 18:36:13,901 - Epoch: [177][ 740/ 1207] Overall Loss 0.222209 Objective Loss 0.222209 LR 0.000250 Time 0.021351 -2023-02-13 18:36:14,100 - Epoch: [177][ 750/ 1207] Overall Loss 0.222511 Objective Loss 0.222511 LR 0.000250 Time 0.021331 -2023-02-13 18:36:14,295 - Epoch: [177][ 760/ 1207] Overall Loss 0.222391 Objective Loss 0.222391 LR 0.000250 Time 0.021306 -2023-02-13 18:36:14,494 - Epoch: [177][ 770/ 1207] Overall Loss 0.222193 Objective Loss 0.222193 LR 0.000250 Time 0.021287 -2023-02-13 18:36:14,689 - Epoch: [177][ 780/ 1207] Overall Loss 0.222181 Objective Loss 0.222181 LR 0.000250 Time 0.021264 -2023-02-13 18:36:14,888 - Epoch: [177][ 790/ 1207] Overall Loss 0.222042 Objective Loss 0.222042 LR 0.000250 Time 0.021246 -2023-02-13 18:36:15,083 - Epoch: [177][ 800/ 1207] Overall Loss 0.222065 Objective Loss 0.222065 LR 0.000250 Time 0.021223 -2023-02-13 18:36:15,281 - Epoch: [177][ 810/ 1207] Overall Loss 0.222231 Objective Loss 0.222231 LR 0.000250 Time 0.021206 -2023-02-13 18:36:15,476 - Epoch: [177][ 820/ 1207] Overall Loss 0.222179 Objective Loss 0.222179 LR 0.000250 Time 0.021184 -2023-02-13 18:36:15,675 - Epoch: [177][ 830/ 1207] Overall Loss 0.222552 Objective Loss 0.222552 LR 0.000250 Time 0.021168 -2023-02-13 18:36:15,869 - Epoch: [177][ 840/ 1207] Overall Loss 0.222507 Objective Loss 0.222507 LR 0.000250 Time 0.021147 -2023-02-13 18:36:16,069 - Epoch: [177][ 850/ 1207] Overall Loss 0.222597 Objective Loss 0.222597 LR 0.000250 Time 0.021133 -2023-02-13 18:36:16,264 - Epoch: [177][ 860/ 1207] Overall Loss 0.222469 Objective Loss 0.222469 LR 0.000250 Time 0.021113 -2023-02-13 18:36:16,463 - Epoch: [177][ 870/ 1207] Overall Loss 0.222230 Objective Loss 0.222230 LR 0.000250 Time 0.021099 -2023-02-13 18:36:16,658 - Epoch: [177][ 880/ 1207] Overall Loss 0.222437 Objective Loss 0.222437 LR 0.000250 Time 0.021081 -2023-02-13 18:36:16,858 - Epoch: [177][ 890/ 1207] Overall Loss 0.222181 Objective Loss 0.222181 LR 0.000250 Time 0.021067 -2023-02-13 18:36:17,053 - Epoch: [177][ 900/ 1207] Overall Loss 0.222278 Objective Loss 0.222278 LR 0.000250 Time 0.021050 -2023-02-13 18:36:17,252 - Epoch: [177][ 910/ 1207] Overall Loss 0.222335 Objective Loss 0.222335 LR 0.000250 Time 0.021037 -2023-02-13 18:36:17,447 - Epoch: [177][ 920/ 1207] Overall Loss 0.222205 Objective Loss 0.222205 LR 0.000250 Time 0.021020 -2023-02-13 18:36:17,646 - Epoch: [177][ 930/ 1207] Overall Loss 0.222351 Objective Loss 0.222351 LR 0.000250 Time 0.021008 -2023-02-13 18:36:17,841 - Epoch: [177][ 940/ 1207] Overall Loss 0.222450 Objective Loss 0.222450 LR 0.000250 Time 0.020991 -2023-02-13 18:36:18,040 - Epoch: [177][ 950/ 1207] Overall Loss 0.222437 Objective Loss 0.222437 LR 0.000250 Time 0.020979 -2023-02-13 18:36:18,235 - Epoch: [177][ 960/ 1207] Overall Loss 0.222467 Objective Loss 0.222467 LR 0.000250 Time 0.020963 -2023-02-13 18:36:18,433 - Epoch: [177][ 970/ 1207] Overall Loss 0.222344 Objective Loss 0.222344 LR 0.000250 Time 0.020951 -2023-02-13 18:36:18,630 - Epoch: [177][ 980/ 1207] Overall Loss 0.222394 Objective Loss 0.222394 LR 0.000250 Time 0.020937 -2023-02-13 18:36:18,829 - Epoch: [177][ 990/ 1207] Overall Loss 0.222502 Objective Loss 0.222502 LR 0.000250 Time 0.020927 -2023-02-13 18:36:19,023 - Epoch: [177][ 1000/ 1207] Overall Loss 0.222585 Objective Loss 0.222585 LR 0.000250 Time 0.020912 -2023-02-13 18:36:19,222 - Epoch: [177][ 1010/ 1207] Overall Loss 0.222414 Objective Loss 0.222414 LR 0.000250 Time 0.020901 -2023-02-13 18:36:19,417 - Epoch: [177][ 1020/ 1207] Overall Loss 0.222509 Objective Loss 0.222509 LR 0.000250 Time 0.020887 -2023-02-13 18:36:19,616 - Epoch: [177][ 1030/ 1207] Overall Loss 0.222771 Objective Loss 0.222771 LR 0.000250 Time 0.020877 -2023-02-13 18:36:19,811 - Epoch: [177][ 1040/ 1207] Overall Loss 0.222965 Objective Loss 0.222965 LR 0.000250 Time 0.020863 -2023-02-13 18:36:20,010 - Epoch: [177][ 1050/ 1207] Overall Loss 0.223283 Objective Loss 0.223283 LR 0.000250 Time 0.020853 -2023-02-13 18:36:20,204 - Epoch: [177][ 1060/ 1207] Overall Loss 0.223210 Objective Loss 0.223210 LR 0.000250 Time 0.020840 -2023-02-13 18:36:20,403 - Epoch: [177][ 1070/ 1207] Overall Loss 0.223095 Objective Loss 0.223095 LR 0.000250 Time 0.020830 -2023-02-13 18:36:20,598 - Epoch: [177][ 1080/ 1207] Overall Loss 0.223100 Objective Loss 0.223100 LR 0.000250 Time 0.020818 -2023-02-13 18:36:20,797 - Epoch: [177][ 1090/ 1207] Overall Loss 0.223193 Objective Loss 0.223193 LR 0.000250 Time 0.020809 -2023-02-13 18:36:20,993 - Epoch: [177][ 1100/ 1207] Overall Loss 0.223058 Objective Loss 0.223058 LR 0.000250 Time 0.020798 -2023-02-13 18:36:21,192 - Epoch: [177][ 1110/ 1207] Overall Loss 0.222897 Objective Loss 0.222897 LR 0.000250 Time 0.020789 -2023-02-13 18:36:21,387 - Epoch: [177][ 1120/ 1207] Overall Loss 0.222969 Objective Loss 0.222969 LR 0.000250 Time 0.020777 -2023-02-13 18:36:21,586 - Epoch: [177][ 1130/ 1207] Overall Loss 0.223220 Objective Loss 0.223220 LR 0.000250 Time 0.020769 -2023-02-13 18:36:21,783 - Epoch: [177][ 1140/ 1207] Overall Loss 0.223317 Objective Loss 0.223317 LR 0.000250 Time 0.020759 -2023-02-13 18:36:21,982 - Epoch: [177][ 1150/ 1207] Overall Loss 0.223326 Objective Loss 0.223326 LR 0.000250 Time 0.020752 -2023-02-13 18:36:22,177 - Epoch: [177][ 1160/ 1207] Overall Loss 0.223441 Objective Loss 0.223441 LR 0.000250 Time 0.020741 -2023-02-13 18:36:22,376 - Epoch: [177][ 1170/ 1207] Overall Loss 0.223575 Objective Loss 0.223575 LR 0.000250 Time 0.020733 -2023-02-13 18:36:22,571 - Epoch: [177][ 1180/ 1207] Overall Loss 0.223806 Objective Loss 0.223806 LR 0.000250 Time 0.020723 -2023-02-13 18:36:22,770 - Epoch: [177][ 1190/ 1207] Overall Loss 0.223762 Objective Loss 0.223762 LR 0.000250 Time 0.020715 -2023-02-13 18:36:23,017 - Epoch: [177][ 1200/ 1207] Overall Loss 0.223673 Objective Loss 0.223673 LR 0.000250 Time 0.020748 -2023-02-13 18:36:23,132 - Epoch: [177][ 1207/ 1207] Overall Loss 0.223726 Objective Loss 0.223726 Top1 83.231707 Top5 98.780488 LR 0.000250 Time 0.020723 -2023-02-13 18:36:23,204 - --- validate (epoch=177)----------- -2023-02-13 18:36:23,205 - 34311 samples (256 per mini-batch) -2023-02-13 18:36:23,605 - Epoch: [177][ 10/ 135] Loss 0.286556 Top1 85.937500 Top5 97.773438 -2023-02-13 18:36:23,732 - Epoch: [177][ 20/ 135] Loss 0.302174 Top1 84.687500 Top5 97.617188 -2023-02-13 18:36:23,857 - Epoch: [177][ 30/ 135] Loss 0.301311 Top1 84.765625 Top5 97.578125 -2023-02-13 18:36:23,980 - Epoch: [177][ 40/ 135] Loss 0.299212 Top1 84.560547 Top5 97.548828 -2023-02-13 18:36:24,112 - Epoch: [177][ 50/ 135] Loss 0.304573 Top1 84.554688 Top5 97.523438 -2023-02-13 18:36:24,235 - Epoch: [177][ 60/ 135] Loss 0.304915 Top1 84.492188 Top5 97.506510 -2023-02-13 18:36:24,367 - Epoch: [177][ 70/ 135] Loss 0.304511 Top1 84.575893 Top5 97.527902 -2023-02-13 18:36:24,496 - Epoch: [177][ 80/ 135] Loss 0.302371 Top1 84.667969 Top5 97.583008 -2023-02-13 18:36:24,628 - Epoch: [177][ 90/ 135] Loss 0.302384 Top1 84.648438 Top5 97.656250 -2023-02-13 18:36:24,756 - Epoch: [177][ 100/ 135] Loss 0.303275 Top1 84.632812 Top5 97.671875 -2023-02-13 18:36:24,887 - Epoch: [177][ 110/ 135] Loss 0.303129 Top1 84.719460 Top5 97.674006 -2023-02-13 18:36:25,016 - Epoch: [177][ 120/ 135] Loss 0.300602 Top1 84.752604 Top5 97.646484 -2023-02-13 18:36:25,148 - Epoch: [177][ 130/ 135] Loss 0.299753 Top1 84.735577 Top5 97.650240 -2023-02-13 18:36:25,193 - Epoch: [177][ 135/ 135] Loss 0.299420 Top1 84.765819 Top5 97.653814 -2023-02-13 18:36:25,266 - ==> Top1: 84.766 Top5: 97.654 Loss: 0.299 - -2023-02-13 18:36:25,267 - ==> Confusion: -[[ 867 4 8 2 8 1 0 0 3 42 0 4 2 4 6 4 2 2 1 2 5] - [ 2 954 3 1 7 20 3 14 4 2 3 2 1 0 1 2 2 2 1 2 7] - [ 6 3 963 12 2 1 11 14 1 2 3 1 1 4 2 9 1 4 7 4 7] - [ 6 0 15 905 1 7 0 2 1 2 17 1 10 0 14 2 1 8 19 0 5] - [ 15 7 1 0 988 10 1 2 2 2 1 5 4 3 7 4 6 2 1 2 3] - [ 4 13 1 4 2 986 5 11 3 3 1 10 1 13 0 2 5 0 1 2 3] - [ 3 2 12 2 0 4 1040 4 0 1 4 0 3 3 0 4 2 3 2 6 4] - [ 1 6 9 0 0 24 4 933 1 2 1 8 3 2 0 0 0 1 17 7 5] - [ 11 3 1 1 2 0 0 0 924 34 7 3 1 8 7 1 1 0 5 0 0] - [ 75 0 1 0 6 0 0 3 35 861 0 2 0 17 3 1 0 3 1 1 3] - [ 2 0 2 4 2 3 2 5 11 2 999 1 0 6 1 0 2 1 5 0 3] - [ 2 1 1 0 4 7 2 4 1 2 0 935 16 5 1 3 1 11 1 7 1] - [ 1 0 0 4 1 2 0 0 1 1 0 34 881 2 1 5 3 11 3 1 8] - [ 6 2 1 0 5 8 0 0 14 19 9 6 4 929 3 5 3 2 1 0 7] - [ 3 1 2 16 4 2 0 1 32 10 3 2 2 3 978 0 3 6 17 0 7] - [ 5 0 4 0 2 1 3 1 0 0 0 7 9 3 0 980 7 8 0 10 6] - [ 2 7 0 1 4 3 0 1 2 0 1 1 3 2 1 12 1001 1 2 3 14] - [ 5 2 0 6 0 1 1 0 0 1 1 8 18 1 0 12 1 985 0 3 6] - [ 5 4 4 7 0 0 0 21 4 0 6 1 5 0 12 2 1 4 1008 1 1] - [ 1 3 0 1 1 3 3 11 0 0 0 17 1 2 1 6 3 5 1 1083 6] - [ 164 236 232 129 107 205 72 179 113 83 222 138 319 283 139 104 249 103 192 281 9884]] - -2023-02-13 18:36:25,269 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:36:25,269 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:36:25,275 - - -2023-02-13 18:36:25,275 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:36:26,161 - Epoch: [178][ 10/ 1207] Overall Loss 0.218895 Objective Loss 0.218895 LR 0.000250 Time 0.088511 -2023-02-13 18:36:26,355 - Epoch: [178][ 20/ 1207] Overall Loss 0.217004 Objective Loss 0.217004 LR 0.000250 Time 0.053927 -2023-02-13 18:36:26,543 - Epoch: [178][ 30/ 1207] Overall Loss 0.224643 Objective Loss 0.224643 LR 0.000250 Time 0.042233 -2023-02-13 18:36:26,732 - Epoch: [178][ 40/ 1207] Overall Loss 0.222344 Objective Loss 0.222344 LR 0.000250 Time 0.036387 -2023-02-13 18:36:26,921 - Epoch: [178][ 50/ 1207] Overall Loss 0.220763 Objective Loss 0.220763 LR 0.000250 Time 0.032883 -2023-02-13 18:36:27,111 - Epoch: [178][ 60/ 1207] Overall Loss 0.222676 Objective Loss 0.222676 LR 0.000250 Time 0.030552 -2023-02-13 18:36:27,301 - Epoch: [178][ 70/ 1207] Overall Loss 0.221632 Objective Loss 0.221632 LR 0.000250 Time 0.028897 -2023-02-13 18:36:27,490 - Epoch: [178][ 80/ 1207] Overall Loss 0.224092 Objective Loss 0.224092 LR 0.000250 Time 0.027648 -2023-02-13 18:36:27,680 - Epoch: [178][ 90/ 1207] Overall Loss 0.224182 Objective Loss 0.224182 LR 0.000250 Time 0.026685 -2023-02-13 18:36:27,869 - Epoch: [178][ 100/ 1207] Overall Loss 0.223892 Objective Loss 0.223892 LR 0.000250 Time 0.025900 -2023-02-13 18:36:28,058 - Epoch: [178][ 110/ 1207] Overall Loss 0.220669 Objective Loss 0.220669 LR 0.000250 Time 0.025256 -2023-02-13 18:36:28,246 - Epoch: [178][ 120/ 1207] Overall Loss 0.221264 Objective Loss 0.221264 LR 0.000250 Time 0.024720 -2023-02-13 18:36:28,435 - Epoch: [178][ 130/ 1207] Overall Loss 0.222021 Objective Loss 0.222021 LR 0.000250 Time 0.024267 -2023-02-13 18:36:28,624 - Epoch: [178][ 140/ 1207] Overall Loss 0.221094 Objective Loss 0.221094 LR 0.000250 Time 0.023881 -2023-02-13 18:36:28,813 - Epoch: [178][ 150/ 1207] Overall Loss 0.220093 Objective Loss 0.220093 LR 0.000250 Time 0.023545 -2023-02-13 18:36:29,001 - Epoch: [178][ 160/ 1207] Overall Loss 0.220501 Objective Loss 0.220501 LR 0.000250 Time 0.023251 -2023-02-13 18:36:29,191 - Epoch: [178][ 170/ 1207] Overall Loss 0.220470 Objective Loss 0.220470 LR 0.000250 Time 0.022993 -2023-02-13 18:36:29,380 - Epoch: [178][ 180/ 1207] Overall Loss 0.219783 Objective Loss 0.219783 LR 0.000250 Time 0.022765 -2023-02-13 18:36:29,569 - Epoch: [178][ 190/ 1207] Overall Loss 0.220681 Objective Loss 0.220681 LR 0.000250 Time 0.022561 -2023-02-13 18:36:29,759 - Epoch: [178][ 200/ 1207] Overall Loss 0.219730 Objective Loss 0.219730 LR 0.000250 Time 0.022381 -2023-02-13 18:36:29,949 - Epoch: [178][ 210/ 1207] Overall Loss 0.219924 Objective Loss 0.219924 LR 0.000250 Time 0.022218 -2023-02-13 18:36:30,139 - Epoch: [178][ 220/ 1207] Overall Loss 0.219575 Objective Loss 0.219575 LR 0.000250 Time 0.022071 -2023-02-13 18:36:30,329 - Epoch: [178][ 230/ 1207] Overall Loss 0.219829 Objective Loss 0.219829 LR 0.000250 Time 0.021937 -2023-02-13 18:36:30,519 - Epoch: [178][ 240/ 1207] Overall Loss 0.219602 Objective Loss 0.219602 LR 0.000250 Time 0.021812 -2023-02-13 18:36:30,710 - Epoch: [178][ 250/ 1207] Overall Loss 0.220655 Objective Loss 0.220655 LR 0.000250 Time 0.021702 -2023-02-13 18:36:30,901 - Epoch: [178][ 260/ 1207] Overall Loss 0.220571 Objective Loss 0.220571 LR 0.000250 Time 0.021600 -2023-02-13 18:36:31,093 - Epoch: [178][ 270/ 1207] Overall Loss 0.219838 Objective Loss 0.219838 LR 0.000250 Time 0.021510 -2023-02-13 18:36:31,284 - Epoch: [178][ 280/ 1207] Overall Loss 0.219987 Objective Loss 0.219987 LR 0.000250 Time 0.021422 -2023-02-13 18:36:31,474 - Epoch: [178][ 290/ 1207] Overall Loss 0.220649 Objective Loss 0.220649 LR 0.000250 Time 0.021339 -2023-02-13 18:36:31,666 - Epoch: [178][ 300/ 1207] Overall Loss 0.219700 Objective Loss 0.219700 LR 0.000250 Time 0.021264 -2023-02-13 18:36:31,857 - Epoch: [178][ 310/ 1207] Overall Loss 0.219857 Objective Loss 0.219857 LR 0.000250 Time 0.021193 -2023-02-13 18:36:32,048 - Epoch: [178][ 320/ 1207] Overall Loss 0.219618 Objective Loss 0.219618 LR 0.000250 Time 0.021128 -2023-02-13 18:36:32,239 - Epoch: [178][ 330/ 1207] Overall Loss 0.218830 Objective Loss 0.218830 LR 0.000250 Time 0.021064 -2023-02-13 18:36:32,430 - Epoch: [178][ 340/ 1207] Overall Loss 0.219416 Objective Loss 0.219416 LR 0.000250 Time 0.021006 -2023-02-13 18:36:32,620 - Epoch: [178][ 350/ 1207] Overall Loss 0.219425 Objective Loss 0.219425 LR 0.000250 Time 0.020949 -2023-02-13 18:36:32,811 - Epoch: [178][ 360/ 1207] Overall Loss 0.219554 Objective Loss 0.219554 LR 0.000250 Time 0.020896 -2023-02-13 18:36:33,002 - Epoch: [178][ 370/ 1207] Overall Loss 0.219738 Objective Loss 0.219738 LR 0.000250 Time 0.020845 -2023-02-13 18:36:33,194 - Epoch: [178][ 380/ 1207] Overall Loss 0.219715 Objective Loss 0.219715 LR 0.000250 Time 0.020800 -2023-02-13 18:36:33,384 - Epoch: [178][ 390/ 1207] Overall Loss 0.219759 Objective Loss 0.219759 LR 0.000250 Time 0.020755 -2023-02-13 18:36:33,575 - Epoch: [178][ 400/ 1207] Overall Loss 0.219861 Objective Loss 0.219861 LR 0.000250 Time 0.020712 -2023-02-13 18:36:33,769 - Epoch: [178][ 410/ 1207] Overall Loss 0.220992 Objective Loss 0.220992 LR 0.000250 Time 0.020678 -2023-02-13 18:36:33,960 - Epoch: [178][ 420/ 1207] Overall Loss 0.220639 Objective Loss 0.220639 LR 0.000250 Time 0.020639 -2023-02-13 18:36:34,150 - Epoch: [178][ 430/ 1207] Overall Loss 0.220229 Objective Loss 0.220229 LR 0.000250 Time 0.020601 -2023-02-13 18:36:34,341 - Epoch: [178][ 440/ 1207] Overall Loss 0.220618 Objective Loss 0.220618 LR 0.000250 Time 0.020566 -2023-02-13 18:36:34,531 - Epoch: [178][ 450/ 1207] Overall Loss 0.220548 Objective Loss 0.220548 LR 0.000250 Time 0.020531 -2023-02-13 18:36:34,723 - Epoch: [178][ 460/ 1207] Overall Loss 0.220160 Objective Loss 0.220160 LR 0.000250 Time 0.020502 -2023-02-13 18:36:34,914 - Epoch: [178][ 470/ 1207] Overall Loss 0.220119 Objective Loss 0.220119 LR 0.000250 Time 0.020471 -2023-02-13 18:36:35,105 - Epoch: [178][ 480/ 1207] Overall Loss 0.220051 Objective Loss 0.220051 LR 0.000250 Time 0.020442 -2023-02-13 18:36:35,297 - Epoch: [178][ 490/ 1207] Overall Loss 0.220492 Objective Loss 0.220492 LR 0.000250 Time 0.020414 -2023-02-13 18:36:35,487 - Epoch: [178][ 500/ 1207] Overall Loss 0.220739 Objective Loss 0.220739 LR 0.000250 Time 0.020387 -2023-02-13 18:36:35,679 - Epoch: [178][ 510/ 1207] Overall Loss 0.220661 Objective Loss 0.220661 LR 0.000250 Time 0.020362 -2023-02-13 18:36:35,870 - Epoch: [178][ 520/ 1207] Overall Loss 0.220684 Objective Loss 0.220684 LR 0.000250 Time 0.020337 -2023-02-13 18:36:36,063 - Epoch: [178][ 530/ 1207] Overall Loss 0.220849 Objective Loss 0.220849 LR 0.000250 Time 0.020316 -2023-02-13 18:36:36,253 - Epoch: [178][ 540/ 1207] Overall Loss 0.221256 Objective Loss 0.221256 LR 0.000250 Time 0.020291 -2023-02-13 18:36:36,443 - Epoch: [178][ 550/ 1207] Overall Loss 0.220947 Objective Loss 0.220947 LR 0.000250 Time 0.020268 -2023-02-13 18:36:36,636 - Epoch: [178][ 560/ 1207] Overall Loss 0.220823 Objective Loss 0.220823 LR 0.000250 Time 0.020248 -2023-02-13 18:36:36,826 - Epoch: [178][ 570/ 1207] Overall Loss 0.220897 Objective Loss 0.220897 LR 0.000250 Time 0.020227 -2023-02-13 18:36:37,018 - Epoch: [178][ 580/ 1207] Overall Loss 0.221106 Objective Loss 0.221106 LR 0.000250 Time 0.020208 -2023-02-13 18:36:37,209 - Epoch: [178][ 590/ 1207] Overall Loss 0.221184 Objective Loss 0.221184 LR 0.000250 Time 0.020189 -2023-02-13 18:36:37,400 - Epoch: [178][ 600/ 1207] Overall Loss 0.221432 Objective Loss 0.221432 LR 0.000250 Time 0.020171 -2023-02-13 18:36:37,591 - Epoch: [178][ 610/ 1207] Overall Loss 0.221834 Objective Loss 0.221834 LR 0.000250 Time 0.020152 -2023-02-13 18:36:37,783 - Epoch: [178][ 620/ 1207] Overall Loss 0.221898 Objective Loss 0.221898 LR 0.000250 Time 0.020137 -2023-02-13 18:36:37,974 - Epoch: [178][ 630/ 1207] Overall Loss 0.221828 Objective Loss 0.221828 LR 0.000250 Time 0.020119 -2023-02-13 18:36:38,165 - Epoch: [178][ 640/ 1207] Overall Loss 0.221856 Objective Loss 0.221856 LR 0.000250 Time 0.020102 -2023-02-13 18:36:38,355 - Epoch: [178][ 650/ 1207] Overall Loss 0.222187 Objective Loss 0.222187 LR 0.000250 Time 0.020085 -2023-02-13 18:36:38,546 - Epoch: [178][ 660/ 1207] Overall Loss 0.222435 Objective Loss 0.222435 LR 0.000250 Time 0.020070 -2023-02-13 18:36:38,738 - Epoch: [178][ 670/ 1207] Overall Loss 0.222335 Objective Loss 0.222335 LR 0.000250 Time 0.020056 -2023-02-13 18:36:38,929 - Epoch: [178][ 680/ 1207] Overall Loss 0.222590 Objective Loss 0.222590 LR 0.000250 Time 0.020041 -2023-02-13 18:36:39,120 - Epoch: [178][ 690/ 1207] Overall Loss 0.222355 Objective Loss 0.222355 LR 0.000250 Time 0.020027 -2023-02-13 18:36:39,312 - Epoch: [178][ 700/ 1207] Overall Loss 0.222525 Objective Loss 0.222525 LR 0.000250 Time 0.020015 -2023-02-13 18:36:39,503 - Epoch: [178][ 710/ 1207] Overall Loss 0.222226 Objective Loss 0.222226 LR 0.000250 Time 0.020001 -2023-02-13 18:36:39,696 - Epoch: [178][ 720/ 1207] Overall Loss 0.221905 Objective Loss 0.221905 LR 0.000250 Time 0.019991 -2023-02-13 18:36:39,887 - Epoch: [178][ 730/ 1207] Overall Loss 0.221844 Objective Loss 0.221844 LR 0.000250 Time 0.019979 -2023-02-13 18:36:40,078 - Epoch: [178][ 740/ 1207] Overall Loss 0.221527 Objective Loss 0.221527 LR 0.000250 Time 0.019966 -2023-02-13 18:36:40,270 - Epoch: [178][ 750/ 1207] Overall Loss 0.221797 Objective Loss 0.221797 LR 0.000250 Time 0.019955 -2023-02-13 18:36:40,461 - Epoch: [178][ 760/ 1207] Overall Loss 0.221849 Objective Loss 0.221849 LR 0.000250 Time 0.019943 -2023-02-13 18:36:40,653 - Epoch: [178][ 770/ 1207] Overall Loss 0.221762 Objective Loss 0.221762 LR 0.000250 Time 0.019933 -2023-02-13 18:36:40,844 - Epoch: [178][ 780/ 1207] Overall Loss 0.222241 Objective Loss 0.222241 LR 0.000250 Time 0.019922 -2023-02-13 18:36:41,037 - Epoch: [178][ 790/ 1207] Overall Loss 0.222123 Objective Loss 0.222123 LR 0.000250 Time 0.019913 -2023-02-13 18:36:41,228 - Epoch: [178][ 800/ 1207] Overall Loss 0.221931 Objective Loss 0.221931 LR 0.000250 Time 0.019903 -2023-02-13 18:36:41,419 - Epoch: [178][ 810/ 1207] Overall Loss 0.221988 Objective Loss 0.221988 LR 0.000250 Time 0.019893 -2023-02-13 18:36:41,609 - Epoch: [178][ 820/ 1207] Overall Loss 0.221838 Objective Loss 0.221838 LR 0.000250 Time 0.019882 -2023-02-13 18:36:41,801 - Epoch: [178][ 830/ 1207] Overall Loss 0.221855 Objective Loss 0.221855 LR 0.000250 Time 0.019873 -2023-02-13 18:36:41,992 - Epoch: [178][ 840/ 1207] Overall Loss 0.221572 Objective Loss 0.221572 LR 0.000250 Time 0.019864 -2023-02-13 18:36:42,184 - Epoch: [178][ 850/ 1207] Overall Loss 0.221927 Objective Loss 0.221927 LR 0.000250 Time 0.019855 -2023-02-13 18:36:42,375 - Epoch: [178][ 860/ 1207] Overall Loss 0.221875 Objective Loss 0.221875 LR 0.000250 Time 0.019846 -2023-02-13 18:36:42,567 - Epoch: [178][ 870/ 1207] Overall Loss 0.221985 Objective Loss 0.221985 LR 0.000250 Time 0.019838 -2023-02-13 18:36:42,760 - Epoch: [178][ 880/ 1207] Overall Loss 0.221841 Objective Loss 0.221841 LR 0.000250 Time 0.019831 -2023-02-13 18:36:42,951 - Epoch: [178][ 890/ 1207] Overall Loss 0.221416 Objective Loss 0.221416 LR 0.000250 Time 0.019823 -2023-02-13 18:36:43,143 - Epoch: [178][ 900/ 1207] Overall Loss 0.221511 Objective Loss 0.221511 LR 0.000250 Time 0.019815 -2023-02-13 18:36:43,334 - Epoch: [178][ 910/ 1207] Overall Loss 0.221505 Objective Loss 0.221505 LR 0.000250 Time 0.019807 -2023-02-13 18:36:43,525 - Epoch: [178][ 920/ 1207] Overall Loss 0.221656 Objective Loss 0.221656 LR 0.000250 Time 0.019799 -2023-02-13 18:36:43,717 - Epoch: [178][ 930/ 1207] Overall Loss 0.221713 Objective Loss 0.221713 LR 0.000250 Time 0.019792 -2023-02-13 18:36:43,909 - Epoch: [178][ 940/ 1207] Overall Loss 0.221513 Objective Loss 0.221513 LR 0.000250 Time 0.019785 -2023-02-13 18:36:44,100 - Epoch: [178][ 950/ 1207] Overall Loss 0.221750 Objective Loss 0.221750 LR 0.000250 Time 0.019777 -2023-02-13 18:36:44,291 - Epoch: [178][ 960/ 1207] Overall Loss 0.221748 Objective Loss 0.221748 LR 0.000250 Time 0.019771 -2023-02-13 18:36:44,482 - Epoch: [178][ 970/ 1207] Overall Loss 0.221735 Objective Loss 0.221735 LR 0.000250 Time 0.019763 -2023-02-13 18:36:44,675 - Epoch: [178][ 980/ 1207] Overall Loss 0.221448 Objective Loss 0.221448 LR 0.000250 Time 0.019758 -2023-02-13 18:36:44,866 - Epoch: [178][ 990/ 1207] Overall Loss 0.221602 Objective Loss 0.221602 LR 0.000250 Time 0.019751 -2023-02-13 18:36:45,057 - Epoch: [178][ 1000/ 1207] Overall Loss 0.221468 Objective Loss 0.221468 LR 0.000250 Time 0.019744 -2023-02-13 18:36:45,249 - Epoch: [178][ 1010/ 1207] Overall Loss 0.221381 Objective Loss 0.221381 LR 0.000250 Time 0.019739 -2023-02-13 18:36:45,440 - Epoch: [178][ 1020/ 1207] Overall Loss 0.221318 Objective Loss 0.221318 LR 0.000250 Time 0.019732 -2023-02-13 18:36:45,631 - Epoch: [178][ 1030/ 1207] Overall Loss 0.221385 Objective Loss 0.221385 LR 0.000250 Time 0.019725 -2023-02-13 18:36:45,824 - Epoch: [178][ 1040/ 1207] Overall Loss 0.221422 Objective Loss 0.221422 LR 0.000250 Time 0.019721 -2023-02-13 18:36:46,017 - Epoch: [178][ 1050/ 1207] Overall Loss 0.221274 Objective Loss 0.221274 LR 0.000250 Time 0.019716 -2023-02-13 18:36:46,208 - Epoch: [178][ 1060/ 1207] Overall Loss 0.221451 Objective Loss 0.221451 LR 0.000250 Time 0.019710 -2023-02-13 18:36:46,399 - Epoch: [178][ 1070/ 1207] Overall Loss 0.221351 Objective Loss 0.221351 LR 0.000250 Time 0.019705 -2023-02-13 18:36:46,590 - Epoch: [178][ 1080/ 1207] Overall Loss 0.221292 Objective Loss 0.221292 LR 0.000250 Time 0.019699 -2023-02-13 18:36:46,791 - Epoch: [178][ 1090/ 1207] Overall Loss 0.221241 Objective Loss 0.221241 LR 0.000250 Time 0.019701 -2023-02-13 18:36:46,989 - Epoch: [178][ 1100/ 1207] Overall Loss 0.221379 Objective Loss 0.221379 LR 0.000250 Time 0.019702 -2023-02-13 18:36:47,187 - Epoch: [178][ 1110/ 1207] Overall Loss 0.221409 Objective Loss 0.221409 LR 0.000250 Time 0.019703 -2023-02-13 18:36:47,384 - Epoch: [178][ 1120/ 1207] Overall Loss 0.221471 Objective Loss 0.221471 LR 0.000250 Time 0.019702 -2023-02-13 18:36:47,583 - Epoch: [178][ 1130/ 1207] Overall Loss 0.221488 Objective Loss 0.221488 LR 0.000250 Time 0.019704 -2023-02-13 18:36:47,781 - Epoch: [178][ 1140/ 1207] Overall Loss 0.221457 Objective Loss 0.221457 LR 0.000250 Time 0.019704 -2023-02-13 18:36:47,980 - Epoch: [178][ 1150/ 1207] Overall Loss 0.221616 Objective Loss 0.221616 LR 0.000250 Time 0.019706 -2023-02-13 18:36:48,178 - Epoch: [178][ 1160/ 1207] Overall Loss 0.221693 Objective Loss 0.221693 LR 0.000250 Time 0.019706 -2023-02-13 18:36:48,378 - Epoch: [178][ 1170/ 1207] Overall Loss 0.221669 Objective Loss 0.221669 LR 0.000250 Time 0.019708 -2023-02-13 18:36:48,575 - Epoch: [178][ 1180/ 1207] Overall Loss 0.221746 Objective Loss 0.221746 LR 0.000250 Time 0.019708 -2023-02-13 18:36:48,777 - Epoch: [178][ 1190/ 1207] Overall Loss 0.221710 Objective Loss 0.221710 LR 0.000250 Time 0.019712 -2023-02-13 18:36:49,031 - Epoch: [178][ 1200/ 1207] Overall Loss 0.221846 Objective Loss 0.221846 LR 0.000250 Time 0.019759 -2023-02-13 18:36:49,147 - Epoch: [178][ 1207/ 1207] Overall Loss 0.221702 Objective Loss 0.221702 Top1 85.975610 Top5 99.085366 LR 0.000250 Time 0.019740 -2023-02-13 18:36:49,218 - --- validate (epoch=178)----------- -2023-02-13 18:36:49,219 - 34311 samples (256 per mini-batch) -2023-02-13 18:36:49,615 - Epoch: [178][ 10/ 135] Loss 0.290735 Top1 84.843750 Top5 97.773438 -2023-02-13 18:36:49,758 - Epoch: [178][ 20/ 135] Loss 0.287425 Top1 84.941406 Top5 97.792969 -2023-02-13 18:36:49,898 - Epoch: [178][ 30/ 135] Loss 0.287065 Top1 84.791667 Top5 97.682292 -2023-02-13 18:36:50,041 - Epoch: [178][ 40/ 135] Loss 0.296260 Top1 84.775391 Top5 97.617188 -2023-02-13 18:36:50,181 - Epoch: [178][ 50/ 135] Loss 0.295649 Top1 84.539062 Top5 97.710938 -2023-02-13 18:36:50,324 - Epoch: [178][ 60/ 135] Loss 0.294495 Top1 84.537760 Top5 97.695312 -2023-02-13 18:36:50,453 - Epoch: [178][ 70/ 135] Loss 0.296427 Top1 84.453125 Top5 97.728795 -2023-02-13 18:36:50,591 - Epoch: [178][ 80/ 135] Loss 0.295964 Top1 84.531250 Top5 97.724609 -2023-02-13 18:36:50,724 - Epoch: [178][ 90/ 135] Loss 0.299549 Top1 84.496528 Top5 97.669271 -2023-02-13 18:36:50,849 - Epoch: [178][ 100/ 135] Loss 0.299498 Top1 84.503906 Top5 97.671875 -2023-02-13 18:36:50,989 - Epoch: [178][ 110/ 135] Loss 0.300562 Top1 84.513494 Top5 97.681108 -2023-02-13 18:36:51,123 - Epoch: [178][ 120/ 135] Loss 0.298943 Top1 84.544271 Top5 97.737630 -2023-02-13 18:36:51,258 - Epoch: [178][ 130/ 135] Loss 0.300380 Top1 84.519231 Top5 97.722356 -2023-02-13 18:36:51,303 - Epoch: [178][ 135/ 135] Loss 0.299892 Top1 84.518085 Top5 97.682959 -2023-02-13 18:36:51,374 - ==> Top1: 84.518 Top5: 97.683 Loss: 0.300 - -2023-02-13 18:36:51,375 - ==> Confusion: -[[ 874 4 10 1 9 2 0 1 2 34 0 2 0 5 8 2 2 2 0 3 6] - [ 3 949 1 1 11 23 3 16 3 0 2 0 2 0 0 1 4 2 3 2 7] - [ 5 5 970 16 3 1 11 12 1 1 2 4 0 4 3 8 2 1 5 2 2] - [ 4 0 20 915 0 4 0 1 1 1 14 0 9 1 17 3 3 3 15 2 3] - [ 9 10 0 2 997 11 1 1 0 2 0 3 4 4 7 6 4 1 0 2 2] - [ 3 15 0 4 5 972 4 16 2 1 1 10 2 13 1 4 6 2 1 3 5] - [ 4 4 14 1 0 3 1049 2 1 0 2 0 1 2 0 2 1 3 2 5 3] - [ 0 5 10 0 2 25 6 944 1 1 1 3 6 2 0 0 1 1 10 5 1] - [ 22 3 1 1 2 0 1 2 886 40 9 2 0 15 17 1 1 1 3 0 2] - [ 87 0 2 0 10 0 0 2 30 844 0 0 0 23 4 2 2 0 1 1 4] - [ 1 1 6 8 3 3 2 3 12 2 988 2 2 8 1 0 2 1 2 0 4] - [ 3 2 1 0 3 13 2 8 0 3 0 908 24 7 1 5 2 9 2 7 5] - [ 0 0 0 7 2 5 0 0 2 0 0 18 882 1 3 3 3 17 3 2 11] - [ 3 4 2 1 4 9 2 2 6 14 6 6 0 943 4 3 0 3 0 4 8] - [ 10 2 1 15 4 4 0 1 16 8 2 1 3 2 995 0 2 5 10 0 11] - [ 6 1 6 0 8 1 3 0 0 0 0 7 7 2 0 974 7 9 1 7 7] - [ 4 7 1 4 8 1 0 0 2 0 0 2 1 2 2 14 992 1 1 3 16] - [ 4 2 0 5 0 3 1 0 0 1 1 5 8 1 0 12 1 1000 0 1 6] - [ 6 4 6 8 0 0 0 26 4 0 5 1 3 0 10 1 1 4 1007 0 0] - [ 1 3 1 2 1 5 8 10 1 0 0 16 2 2 0 7 4 2 0 1076 7] - [ 161 246 274 151 143 232 98 183 74 79 191 90 285 318 148 124 269 107 166 261 9834]] - -2023-02-13 18:36:51,376 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:36:51,376 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:36:51,382 - - -2023-02-13 18:36:51,382 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:36:52,396 - Epoch: [179][ 10/ 1207] Overall Loss 0.219360 Objective Loss 0.219360 LR 0.000250 Time 0.101311 -2023-02-13 18:36:52,594 - Epoch: [179][ 20/ 1207] Overall Loss 0.217893 Objective Loss 0.217893 LR 0.000250 Time 0.060532 -2023-02-13 18:36:52,783 - Epoch: [179][ 30/ 1207] Overall Loss 0.218249 Objective Loss 0.218249 LR 0.000250 Time 0.046651 -2023-02-13 18:36:52,973 - Epoch: [179][ 40/ 1207] Overall Loss 0.217076 Objective Loss 0.217076 LR 0.000250 Time 0.039719 -2023-02-13 18:36:53,161 - Epoch: [179][ 50/ 1207] Overall Loss 0.221707 Objective Loss 0.221707 LR 0.000250 Time 0.035537 -2023-02-13 18:36:53,350 - Epoch: [179][ 60/ 1207] Overall Loss 0.220456 Objective Loss 0.220456 LR 0.000250 Time 0.032760 -2023-02-13 18:36:53,539 - Epoch: [179][ 70/ 1207] Overall Loss 0.218508 Objective Loss 0.218508 LR 0.000250 Time 0.030769 -2023-02-13 18:36:53,729 - Epoch: [179][ 80/ 1207] Overall Loss 0.220234 Objective Loss 0.220234 LR 0.000250 Time 0.029296 -2023-02-13 18:36:53,918 - Epoch: [179][ 90/ 1207] Overall Loss 0.219806 Objective Loss 0.219806 LR 0.000250 Time 0.028133 -2023-02-13 18:36:54,107 - Epoch: [179][ 100/ 1207] Overall Loss 0.222570 Objective Loss 0.222570 LR 0.000250 Time 0.027208 -2023-02-13 18:36:54,296 - Epoch: [179][ 110/ 1207] Overall Loss 0.221891 Objective Loss 0.221891 LR 0.000250 Time 0.026451 -2023-02-13 18:36:54,485 - Epoch: [179][ 120/ 1207] Overall Loss 0.220475 Objective Loss 0.220475 LR 0.000250 Time 0.025818 -2023-02-13 18:36:54,677 - Epoch: [179][ 130/ 1207] Overall Loss 0.220692 Objective Loss 0.220692 LR 0.000250 Time 0.025304 -2023-02-13 18:36:54,870 - Epoch: [179][ 140/ 1207] Overall Loss 0.219742 Objective Loss 0.219742 LR 0.000250 Time 0.024871 -2023-02-13 18:36:55,064 - Epoch: [179][ 150/ 1207] Overall Loss 0.219976 Objective Loss 0.219976 LR 0.000250 Time 0.024508 -2023-02-13 18:36:55,258 - Epoch: [179][ 160/ 1207] Overall Loss 0.219391 Objective Loss 0.219391 LR 0.000250 Time 0.024182 -2023-02-13 18:36:55,453 - Epoch: [179][ 170/ 1207] Overall Loss 0.218966 Objective Loss 0.218966 LR 0.000250 Time 0.023908 -2023-02-13 18:36:55,644 - Epoch: [179][ 180/ 1207] Overall Loss 0.218671 Objective Loss 0.218671 LR 0.000250 Time 0.023639 -2023-02-13 18:36:55,834 - Epoch: [179][ 190/ 1207] Overall Loss 0.217124 Objective Loss 0.217124 LR 0.000250 Time 0.023394 -2023-02-13 18:36:56,025 - Epoch: [179][ 200/ 1207] Overall Loss 0.217227 Objective Loss 0.217227 LR 0.000250 Time 0.023176 -2023-02-13 18:36:56,214 - Epoch: [179][ 210/ 1207] Overall Loss 0.217965 Objective Loss 0.217965 LR 0.000250 Time 0.022971 -2023-02-13 18:36:56,403 - Epoch: [179][ 220/ 1207] Overall Loss 0.218269 Objective Loss 0.218269 LR 0.000250 Time 0.022784 -2023-02-13 18:36:56,592 - Epoch: [179][ 230/ 1207] Overall Loss 0.218096 Objective Loss 0.218096 LR 0.000250 Time 0.022614 -2023-02-13 18:36:56,783 - Epoch: [179][ 240/ 1207] Overall Loss 0.217869 Objective Loss 0.217869 LR 0.000250 Time 0.022463 -2023-02-13 18:36:56,972 - Epoch: [179][ 250/ 1207] Overall Loss 0.217870 Objective Loss 0.217870 LR 0.000250 Time 0.022322 -2023-02-13 18:36:57,162 - Epoch: [179][ 260/ 1207] Overall Loss 0.217332 Objective Loss 0.217332 LR 0.000250 Time 0.022190 -2023-02-13 18:36:57,351 - Epoch: [179][ 270/ 1207] Overall Loss 0.217714 Objective Loss 0.217714 LR 0.000250 Time 0.022069 -2023-02-13 18:36:57,541 - Epoch: [179][ 280/ 1207] Overall Loss 0.218197 Objective Loss 0.218197 LR 0.000250 Time 0.021958 -2023-02-13 18:36:57,731 - Epoch: [179][ 290/ 1207] Overall Loss 0.217865 Objective Loss 0.217865 LR 0.000250 Time 0.021855 -2023-02-13 18:36:57,921 - Epoch: [179][ 300/ 1207] Overall Loss 0.218207 Objective Loss 0.218207 LR 0.000250 Time 0.021758 -2023-02-13 18:36:58,110 - Epoch: [179][ 310/ 1207] Overall Loss 0.218837 Objective Loss 0.218837 LR 0.000250 Time 0.021666 -2023-02-13 18:36:58,300 - Epoch: [179][ 320/ 1207] Overall Loss 0.219065 Objective Loss 0.219065 LR 0.000250 Time 0.021579 -2023-02-13 18:36:58,490 - Epoch: [179][ 330/ 1207] Overall Loss 0.219707 Objective Loss 0.219707 LR 0.000250 Time 0.021499 -2023-02-13 18:36:58,679 - Epoch: [179][ 340/ 1207] Overall Loss 0.219905 Objective Loss 0.219905 LR 0.000250 Time 0.021424 -2023-02-13 18:36:58,869 - Epoch: [179][ 350/ 1207] Overall Loss 0.219557 Objective Loss 0.219557 LR 0.000250 Time 0.021353 -2023-02-13 18:36:59,059 - Epoch: [179][ 360/ 1207] Overall Loss 0.219515 Objective Loss 0.219515 LR 0.000250 Time 0.021286 -2023-02-13 18:36:59,248 - Epoch: [179][ 370/ 1207] Overall Loss 0.219551 Objective Loss 0.219551 LR 0.000250 Time 0.021221 -2023-02-13 18:36:59,438 - Epoch: [179][ 380/ 1207] Overall Loss 0.218905 Objective Loss 0.218905 LR 0.000250 Time 0.021161 -2023-02-13 18:36:59,628 - Epoch: [179][ 390/ 1207] Overall Loss 0.219013 Objective Loss 0.219013 LR 0.000250 Time 0.021105 -2023-02-13 18:36:59,818 - Epoch: [179][ 400/ 1207] Overall Loss 0.219280 Objective Loss 0.219280 LR 0.000250 Time 0.021053 -2023-02-13 18:37:00,008 - Epoch: [179][ 410/ 1207] Overall Loss 0.219504 Objective Loss 0.219504 LR 0.000250 Time 0.021001 -2023-02-13 18:37:00,198 - Epoch: [179][ 420/ 1207] Overall Loss 0.220248 Objective Loss 0.220248 LR 0.000250 Time 0.020951 -2023-02-13 18:37:00,387 - Epoch: [179][ 430/ 1207] Overall Loss 0.220767 Objective Loss 0.220767 LR 0.000250 Time 0.020903 -2023-02-13 18:37:00,576 - Epoch: [179][ 440/ 1207] Overall Loss 0.220844 Objective Loss 0.220844 LR 0.000250 Time 0.020858 -2023-02-13 18:37:00,767 - Epoch: [179][ 450/ 1207] Overall Loss 0.220425 Objective Loss 0.220425 LR 0.000250 Time 0.020817 -2023-02-13 18:37:00,957 - Epoch: [179][ 460/ 1207] Overall Loss 0.219776 Objective Loss 0.219776 LR 0.000250 Time 0.020778 -2023-02-13 18:37:01,147 - Epoch: [179][ 470/ 1207] Overall Loss 0.219808 Objective Loss 0.219808 LR 0.000250 Time 0.020738 -2023-02-13 18:37:01,338 - Epoch: [179][ 480/ 1207] Overall Loss 0.219766 Objective Loss 0.219766 LR 0.000250 Time 0.020703 -2023-02-13 18:37:01,527 - Epoch: [179][ 490/ 1207] Overall Loss 0.219569 Objective Loss 0.219569 LR 0.000250 Time 0.020667 -2023-02-13 18:37:01,718 - Epoch: [179][ 500/ 1207] Overall Loss 0.219500 Objective Loss 0.219500 LR 0.000250 Time 0.020634 -2023-02-13 18:37:01,908 - Epoch: [179][ 510/ 1207] Overall Loss 0.219732 Objective Loss 0.219732 LR 0.000250 Time 0.020601 -2023-02-13 18:37:02,099 - Epoch: [179][ 520/ 1207] Overall Loss 0.220073 Objective Loss 0.220073 LR 0.000250 Time 0.020571 -2023-02-13 18:37:02,289 - Epoch: [179][ 530/ 1207] Overall Loss 0.220431 Objective Loss 0.220431 LR 0.000250 Time 0.020540 -2023-02-13 18:37:02,478 - Epoch: [179][ 540/ 1207] Overall Loss 0.220498 Objective Loss 0.220498 LR 0.000250 Time 0.020511 -2023-02-13 18:37:02,668 - Epoch: [179][ 550/ 1207] Overall Loss 0.220541 Objective Loss 0.220541 LR 0.000250 Time 0.020482 -2023-02-13 18:37:02,859 - Epoch: [179][ 560/ 1207] Overall Loss 0.220535 Objective Loss 0.220535 LR 0.000250 Time 0.020456 -2023-02-13 18:37:03,048 - Epoch: [179][ 570/ 1207] Overall Loss 0.220430 Objective Loss 0.220430 LR 0.000250 Time 0.020428 -2023-02-13 18:37:03,237 - Epoch: [179][ 580/ 1207] Overall Loss 0.220425 Objective Loss 0.220425 LR 0.000250 Time 0.020402 -2023-02-13 18:37:03,427 - Epoch: [179][ 590/ 1207] Overall Loss 0.220492 Objective Loss 0.220492 LR 0.000250 Time 0.020377 -2023-02-13 18:37:03,616 - Epoch: [179][ 600/ 1207] Overall Loss 0.220623 Objective Loss 0.220623 LR 0.000250 Time 0.020351 -2023-02-13 18:37:03,806 - Epoch: [179][ 610/ 1207] Overall Loss 0.220167 Objective Loss 0.220167 LR 0.000250 Time 0.020329 -2023-02-13 18:37:03,995 - Epoch: [179][ 620/ 1207] Overall Loss 0.220357 Objective Loss 0.220357 LR 0.000250 Time 0.020305 -2023-02-13 18:37:04,184 - Epoch: [179][ 630/ 1207] Overall Loss 0.220381 Objective Loss 0.220381 LR 0.000250 Time 0.020283 -2023-02-13 18:37:04,374 - Epoch: [179][ 640/ 1207] Overall Loss 0.220468 Objective Loss 0.220468 LR 0.000250 Time 0.020262 -2023-02-13 18:37:04,564 - Epoch: [179][ 650/ 1207] Overall Loss 0.220341 Objective Loss 0.220341 LR 0.000250 Time 0.020241 -2023-02-13 18:37:04,754 - Epoch: [179][ 660/ 1207] Overall Loss 0.220332 Objective Loss 0.220332 LR 0.000250 Time 0.020222 -2023-02-13 18:37:04,943 - Epoch: [179][ 670/ 1207] Overall Loss 0.220409 Objective Loss 0.220409 LR 0.000250 Time 0.020202 -2023-02-13 18:37:05,133 - Epoch: [179][ 680/ 1207] Overall Loss 0.220237 Objective Loss 0.220237 LR 0.000250 Time 0.020184 -2023-02-13 18:37:05,322 - Epoch: [179][ 690/ 1207] Overall Loss 0.220583 Objective Loss 0.220583 LR 0.000250 Time 0.020165 -2023-02-13 18:37:05,512 - Epoch: [179][ 700/ 1207] Overall Loss 0.220354 Objective Loss 0.220354 LR 0.000250 Time 0.020148 -2023-02-13 18:37:05,702 - Epoch: [179][ 710/ 1207] Overall Loss 0.220232 Objective Loss 0.220232 LR 0.000250 Time 0.020131 -2023-02-13 18:37:05,892 - Epoch: [179][ 720/ 1207] Overall Loss 0.220282 Objective Loss 0.220282 LR 0.000250 Time 0.020115 -2023-02-13 18:37:06,083 - Epoch: [179][ 730/ 1207] Overall Loss 0.220374 Objective Loss 0.220374 LR 0.000250 Time 0.020100 -2023-02-13 18:37:06,273 - Epoch: [179][ 740/ 1207] Overall Loss 0.220500 Objective Loss 0.220500 LR 0.000250 Time 0.020085 -2023-02-13 18:37:06,462 - Epoch: [179][ 750/ 1207] Overall Loss 0.220198 Objective Loss 0.220198 LR 0.000250 Time 0.020069 -2023-02-13 18:37:06,653 - Epoch: [179][ 760/ 1207] Overall Loss 0.219956 Objective Loss 0.219956 LR 0.000250 Time 0.020055 -2023-02-13 18:37:06,843 - Epoch: [179][ 770/ 1207] Overall Loss 0.220114 Objective Loss 0.220114 LR 0.000250 Time 0.020042 -2023-02-13 18:37:07,034 - Epoch: [179][ 780/ 1207] Overall Loss 0.220115 Objective Loss 0.220115 LR 0.000250 Time 0.020028 -2023-02-13 18:37:07,224 - Epoch: [179][ 790/ 1207] Overall Loss 0.220044 Objective Loss 0.220044 LR 0.000250 Time 0.020015 -2023-02-13 18:37:07,414 - Epoch: [179][ 800/ 1207] Overall Loss 0.219930 Objective Loss 0.219930 LR 0.000250 Time 0.020002 -2023-02-13 18:37:07,604 - Epoch: [179][ 810/ 1207] Overall Loss 0.220258 Objective Loss 0.220258 LR 0.000250 Time 0.019989 -2023-02-13 18:37:07,795 - Epoch: [179][ 820/ 1207] Overall Loss 0.220455 Objective Loss 0.220455 LR 0.000250 Time 0.019977 -2023-02-13 18:37:07,985 - Epoch: [179][ 830/ 1207] Overall Loss 0.220291 Objective Loss 0.220291 LR 0.000250 Time 0.019965 -2023-02-13 18:37:08,175 - Epoch: [179][ 840/ 1207] Overall Loss 0.220241 Objective Loss 0.220241 LR 0.000250 Time 0.019953 -2023-02-13 18:37:08,365 - Epoch: [179][ 850/ 1207] Overall Loss 0.220250 Objective Loss 0.220250 LR 0.000250 Time 0.019942 -2023-02-13 18:37:08,554 - Epoch: [179][ 860/ 1207] Overall Loss 0.220061 Objective Loss 0.220061 LR 0.000250 Time 0.019930 -2023-02-13 18:37:08,744 - Epoch: [179][ 870/ 1207] Overall Loss 0.220139 Objective Loss 0.220139 LR 0.000250 Time 0.019919 -2023-02-13 18:37:08,935 - Epoch: [179][ 880/ 1207] Overall Loss 0.220212 Objective Loss 0.220212 LR 0.000250 Time 0.019908 -2023-02-13 18:37:09,124 - Epoch: [179][ 890/ 1207] Overall Loss 0.220099 Objective Loss 0.220099 LR 0.000250 Time 0.019897 -2023-02-13 18:37:09,315 - Epoch: [179][ 900/ 1207] Overall Loss 0.220301 Objective Loss 0.220301 LR 0.000250 Time 0.019887 -2023-02-13 18:37:09,505 - Epoch: [179][ 910/ 1207] Overall Loss 0.220302 Objective Loss 0.220302 LR 0.000250 Time 0.019877 -2023-02-13 18:37:09,695 - Epoch: [179][ 920/ 1207] Overall Loss 0.220328 Objective Loss 0.220328 LR 0.000250 Time 0.019868 -2023-02-13 18:37:09,886 - Epoch: [179][ 930/ 1207] Overall Loss 0.220166 Objective Loss 0.220166 LR 0.000250 Time 0.019859 -2023-02-13 18:37:10,076 - Epoch: [179][ 940/ 1207] Overall Loss 0.220012 Objective Loss 0.220012 LR 0.000250 Time 0.019849 -2023-02-13 18:37:10,265 - Epoch: [179][ 950/ 1207] Overall Loss 0.219953 Objective Loss 0.219953 LR 0.000250 Time 0.019839 -2023-02-13 18:37:10,455 - Epoch: [179][ 960/ 1207] Overall Loss 0.219944 Objective Loss 0.219944 LR 0.000250 Time 0.019830 -2023-02-13 18:37:10,645 - Epoch: [179][ 970/ 1207] Overall Loss 0.220058 Objective Loss 0.220058 LR 0.000250 Time 0.019821 -2023-02-13 18:37:10,836 - Epoch: [179][ 980/ 1207] Overall Loss 0.220047 Objective Loss 0.220047 LR 0.000250 Time 0.019813 -2023-02-13 18:37:11,027 - Epoch: [179][ 990/ 1207] Overall Loss 0.220001 Objective Loss 0.220001 LR 0.000250 Time 0.019806 -2023-02-13 18:37:11,217 - Epoch: [179][ 1000/ 1207] Overall Loss 0.219934 Objective Loss 0.219934 LR 0.000250 Time 0.019797 -2023-02-13 18:37:11,407 - Epoch: [179][ 1010/ 1207] Overall Loss 0.220011 Objective Loss 0.220011 LR 0.000250 Time 0.019789 -2023-02-13 18:37:11,597 - Epoch: [179][ 1020/ 1207] Overall Loss 0.220262 Objective Loss 0.220262 LR 0.000250 Time 0.019781 -2023-02-13 18:37:11,788 - Epoch: [179][ 1030/ 1207] Overall Loss 0.220177 Objective Loss 0.220177 LR 0.000250 Time 0.019774 -2023-02-13 18:37:11,979 - Epoch: [179][ 1040/ 1207] Overall Loss 0.220329 Objective Loss 0.220329 LR 0.000250 Time 0.019767 -2023-02-13 18:37:12,169 - Epoch: [179][ 1050/ 1207] Overall Loss 0.220799 Objective Loss 0.220799 LR 0.000250 Time 0.019759 -2023-02-13 18:37:12,359 - Epoch: [179][ 1060/ 1207] Overall Loss 0.220804 Objective Loss 0.220804 LR 0.000250 Time 0.019752 -2023-02-13 18:37:12,549 - Epoch: [179][ 1070/ 1207] Overall Loss 0.220762 Objective Loss 0.220762 LR 0.000250 Time 0.019745 -2023-02-13 18:37:12,739 - Epoch: [179][ 1080/ 1207] Overall Loss 0.220628 Objective Loss 0.220628 LR 0.000250 Time 0.019738 -2023-02-13 18:37:12,930 - Epoch: [179][ 1090/ 1207] Overall Loss 0.220773 Objective Loss 0.220773 LR 0.000250 Time 0.019731 -2023-02-13 18:37:13,119 - Epoch: [179][ 1100/ 1207] Overall Loss 0.220629 Objective Loss 0.220629 LR 0.000250 Time 0.019723 -2023-02-13 18:37:13,309 - Epoch: [179][ 1110/ 1207] Overall Loss 0.220495 Objective Loss 0.220495 LR 0.000250 Time 0.019716 -2023-02-13 18:37:13,499 - Epoch: [179][ 1120/ 1207] Overall Loss 0.220497 Objective Loss 0.220497 LR 0.000250 Time 0.019710 -2023-02-13 18:37:13,689 - Epoch: [179][ 1130/ 1207] Overall Loss 0.220514 Objective Loss 0.220514 LR 0.000250 Time 0.019703 -2023-02-13 18:37:13,879 - Epoch: [179][ 1140/ 1207] Overall Loss 0.220481 Objective Loss 0.220481 LR 0.000250 Time 0.019697 -2023-02-13 18:37:14,069 - Epoch: [179][ 1150/ 1207] Overall Loss 0.220708 Objective Loss 0.220708 LR 0.000250 Time 0.019690 -2023-02-13 18:37:14,260 - Epoch: [179][ 1160/ 1207] Overall Loss 0.220866 Objective Loss 0.220866 LR 0.000250 Time 0.019684 -2023-02-13 18:37:14,450 - Epoch: [179][ 1170/ 1207] Overall Loss 0.220916 Objective Loss 0.220916 LR 0.000250 Time 0.019678 -2023-02-13 18:37:14,640 - Epoch: [179][ 1180/ 1207] Overall Loss 0.220918 Objective Loss 0.220918 LR 0.000250 Time 0.019672 -2023-02-13 18:37:14,830 - Epoch: [179][ 1190/ 1207] Overall Loss 0.220950 Objective Loss 0.220950 LR 0.000250 Time 0.019667 -2023-02-13 18:37:15,071 - Epoch: [179][ 1200/ 1207] Overall Loss 0.221061 Objective Loss 0.221061 LR 0.000250 Time 0.019703 -2023-02-13 18:37:15,187 - Epoch: [179][ 1207/ 1207] Overall Loss 0.221197 Objective Loss 0.221197 Top1 85.975610 Top5 97.865854 LR 0.000250 Time 0.019684 -2023-02-13 18:37:15,269 - --- validate (epoch=179)----------- -2023-02-13 18:37:15,269 - 34311 samples (256 per mini-batch) -2023-02-13 18:37:15,666 - Epoch: [179][ 10/ 135] Loss 0.308312 Top1 84.687500 Top5 97.382812 -2023-02-13 18:37:15,798 - Epoch: [179][ 20/ 135] Loss 0.301396 Top1 84.492188 Top5 97.656250 -2023-02-13 18:37:15,925 - Epoch: [179][ 30/ 135] Loss 0.295868 Top1 84.752604 Top5 97.669271 -2023-02-13 18:37:16,053 - Epoch: [179][ 40/ 135] Loss 0.298888 Top1 84.746094 Top5 97.578125 -2023-02-13 18:37:16,179 - Epoch: [179][ 50/ 135] Loss 0.299285 Top1 84.890625 Top5 97.601562 -2023-02-13 18:37:16,307 - Epoch: [179][ 60/ 135] Loss 0.303265 Top1 84.921875 Top5 97.636719 -2023-02-13 18:37:16,436 - Epoch: [179][ 70/ 135] Loss 0.301865 Top1 85.083705 Top5 97.672991 -2023-02-13 18:37:16,561 - Epoch: [179][ 80/ 135] Loss 0.304052 Top1 85.083008 Top5 97.636719 -2023-02-13 18:37:16,684 - Epoch: [179][ 90/ 135] Loss 0.302931 Top1 85.052083 Top5 97.638889 -2023-02-13 18:37:16,807 - Epoch: [179][ 100/ 135] Loss 0.303886 Top1 85.023438 Top5 97.625000 -2023-02-13 18:37:16,932 - Epoch: [179][ 110/ 135] Loss 0.303428 Top1 85.017756 Top5 97.638494 -2023-02-13 18:37:17,064 - Epoch: [179][ 120/ 135] Loss 0.301505 Top1 84.951172 Top5 97.620443 -2023-02-13 18:37:17,196 - Epoch: [179][ 130/ 135] Loss 0.300063 Top1 84.960938 Top5 97.653245 -2023-02-13 18:37:17,241 - Epoch: [179][ 135/ 135] Loss 0.303887 Top1 84.981493 Top5 97.650899 -2023-02-13 18:37:17,313 - ==> Top1: 84.981 Top5: 97.651 Loss: 0.304 - -2023-02-13 18:37:17,313 - ==> Confusion: -[[ 870 4 6 1 5 3 0 2 5 45 0 3 0 4 5 2 2 1 1 2 6] - [ 2 950 2 2 9 20 1 15 3 1 2 2 1 0 0 1 7 1 2 3 9] - [ 5 5 966 10 4 2 13 10 0 1 4 3 0 5 5 4 2 1 7 4 7] - [ 6 1 23 910 2 3 0 1 2 3 12 0 7 0 18 2 5 4 12 0 5] - [ 15 9 0 0 991 11 1 0 2 1 0 3 3 4 6 5 8 1 0 3 3] - [ 3 15 0 8 5 972 1 17 2 3 1 7 2 15 1 3 7 0 1 3 4] - [ 2 2 13 2 0 4 1042 7 0 1 2 0 2 3 1 1 1 3 1 7 5] - [ 3 7 9 0 2 27 5 936 0 2 1 5 3 2 0 1 0 3 10 5 3] - [ 16 3 1 1 1 0 1 1 925 31 5 4 0 9 5 2 1 0 3 0 0] - [ 82 1 4 0 8 0 0 1 35 854 0 1 1 14 3 0 1 1 1 1 4] - [ 3 0 6 7 1 2 3 3 14 1 987 0 1 7 3 0 1 2 4 0 6] - [ 2 1 2 0 4 13 1 6 4 1 0 913 16 7 0 7 2 14 2 9 1] - [ 0 0 1 11 1 5 0 0 2 0 0 24 873 2 1 5 3 21 1 2 7] - [ 5 1 0 0 3 10 0 2 10 18 5 4 4 938 3 4 6 4 1 1 5] - [ 4 5 2 17 5 3 0 1 22 6 4 1 3 2 995 0 3 5 5 0 9] - [ 4 2 8 0 4 0 4 1 0 0 0 7 7 3 0 967 12 14 1 6 6] - [ 1 7 0 1 7 3 0 0 5 1 0 0 2 2 2 10 1005 0 1 3 11] - [ 4 3 0 6 0 3 3 0 0 0 0 6 5 1 0 14 2 995 0 2 7] - [ 4 8 6 7 1 2 0 26 4 0 5 1 4 0 13 1 2 3 996 2 1] - [ 1 2 0 1 1 3 8 7 1 0 0 12 3 2 0 7 7 3 0 1084 6] - [ 165 234 270 122 135 215 88 160 106 76 182 95 303 256 148 110 273 117 156 234 9989]] - -2023-02-13 18:37:17,315 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:37:17,315 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:37:17,321 - - -2023-02-13 18:37:17,321 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:37:18,207 - Epoch: [180][ 10/ 1207] Overall Loss 0.241417 Objective Loss 0.241417 LR 0.000125 Time 0.088535 -2023-02-13 18:37:18,398 - Epoch: [180][ 20/ 1207] Overall Loss 0.236304 Objective Loss 0.236304 LR 0.000125 Time 0.053803 -2023-02-13 18:37:18,586 - Epoch: [180][ 30/ 1207] Overall Loss 0.230036 Objective Loss 0.230036 LR 0.000125 Time 0.042115 -2023-02-13 18:37:18,773 - Epoch: [180][ 40/ 1207] Overall Loss 0.225870 Objective Loss 0.225870 LR 0.000125 Time 0.036258 -2023-02-13 18:37:18,961 - Epoch: [180][ 50/ 1207] Overall Loss 0.222714 Objective Loss 0.222714 LR 0.000125 Time 0.032755 -2023-02-13 18:37:19,148 - Epoch: [180][ 60/ 1207] Overall Loss 0.220714 Objective Loss 0.220714 LR 0.000125 Time 0.030402 -2023-02-13 18:37:19,335 - Epoch: [180][ 70/ 1207] Overall Loss 0.217393 Objective Loss 0.217393 LR 0.000125 Time 0.028724 -2023-02-13 18:37:19,522 - Epoch: [180][ 80/ 1207] Overall Loss 0.213417 Objective Loss 0.213417 LR 0.000125 Time 0.027470 -2023-02-13 18:37:19,710 - Epoch: [180][ 90/ 1207] Overall Loss 0.212237 Objective Loss 0.212237 LR 0.000125 Time 0.026503 -2023-02-13 18:37:19,900 - Epoch: [180][ 100/ 1207] Overall Loss 0.211003 Objective Loss 0.211003 LR 0.000125 Time 0.025749 -2023-02-13 18:37:20,089 - Epoch: [180][ 110/ 1207] Overall Loss 0.211514 Objective Loss 0.211514 LR 0.000125 Time 0.025127 -2023-02-13 18:37:20,278 - Epoch: [180][ 120/ 1207] Overall Loss 0.210887 Objective Loss 0.210887 LR 0.000125 Time 0.024603 -2023-02-13 18:37:20,467 - Epoch: [180][ 130/ 1207] Overall Loss 0.212422 Objective Loss 0.212422 LR 0.000125 Time 0.024163 -2023-02-13 18:37:20,656 - Epoch: [180][ 140/ 1207] Overall Loss 0.209975 Objective Loss 0.209975 LR 0.000125 Time 0.023783 -2023-02-13 18:37:20,846 - Epoch: [180][ 150/ 1207] Overall Loss 0.210593 Objective Loss 0.210593 LR 0.000125 Time 0.023459 -2023-02-13 18:37:21,036 - Epoch: [180][ 160/ 1207] Overall Loss 0.210537 Objective Loss 0.210537 LR 0.000125 Time 0.023178 -2023-02-13 18:37:21,225 - Epoch: [180][ 170/ 1207] Overall Loss 0.210473 Objective Loss 0.210473 LR 0.000125 Time 0.022926 -2023-02-13 18:37:21,414 - Epoch: [180][ 180/ 1207] Overall Loss 0.209390 Objective Loss 0.209390 LR 0.000125 Time 0.022703 -2023-02-13 18:37:21,604 - Epoch: [180][ 190/ 1207] Overall Loss 0.209881 Objective Loss 0.209881 LR 0.000125 Time 0.022503 -2023-02-13 18:37:21,793 - Epoch: [180][ 200/ 1207] Overall Loss 0.209549 Objective Loss 0.209549 LR 0.000125 Time 0.022322 -2023-02-13 18:37:21,984 - Epoch: [180][ 210/ 1207] Overall Loss 0.210545 Objective Loss 0.210545 LR 0.000125 Time 0.022166 -2023-02-13 18:37:22,173 - Epoch: [180][ 220/ 1207] Overall Loss 0.210963 Objective Loss 0.210963 LR 0.000125 Time 0.022017 -2023-02-13 18:37:22,362 - Epoch: [180][ 230/ 1207] Overall Loss 0.211676 Objective Loss 0.211676 LR 0.000125 Time 0.021882 -2023-02-13 18:37:22,552 - Epoch: [180][ 240/ 1207] Overall Loss 0.212328 Objective Loss 0.212328 LR 0.000125 Time 0.021759 -2023-02-13 18:37:22,742 - Epoch: [180][ 250/ 1207] Overall Loss 0.212356 Objective Loss 0.212356 LR 0.000125 Time 0.021648 -2023-02-13 18:37:22,933 - Epoch: [180][ 260/ 1207] Overall Loss 0.211970 Objective Loss 0.211970 LR 0.000125 Time 0.021547 -2023-02-13 18:37:23,120 - Epoch: [180][ 270/ 1207] Overall Loss 0.211578 Objective Loss 0.211578 LR 0.000125 Time 0.021439 -2023-02-13 18:37:23,307 - Epoch: [180][ 280/ 1207] Overall Loss 0.212905 Objective Loss 0.212905 LR 0.000125 Time 0.021341 -2023-02-13 18:37:23,494 - Epoch: [180][ 290/ 1207] Overall Loss 0.213332 Objective Loss 0.213332 LR 0.000125 Time 0.021249 -2023-02-13 18:37:23,682 - Epoch: [180][ 300/ 1207] Overall Loss 0.212911 Objective Loss 0.212911 LR 0.000125 Time 0.021165 -2023-02-13 18:37:23,870 - Epoch: [180][ 310/ 1207] Overall Loss 0.212739 Objective Loss 0.212739 LR 0.000125 Time 0.021089 -2023-02-13 18:37:24,057 - Epoch: [180][ 320/ 1207] Overall Loss 0.212812 Objective Loss 0.212812 LR 0.000125 Time 0.021013 -2023-02-13 18:37:24,245 - Epoch: [180][ 330/ 1207] Overall Loss 0.212525 Objective Loss 0.212525 LR 0.000125 Time 0.020944 -2023-02-13 18:37:24,433 - Epoch: [180][ 340/ 1207] Overall Loss 0.212068 Objective Loss 0.212068 LR 0.000125 Time 0.020879 -2023-02-13 18:37:24,620 - Epoch: [180][ 350/ 1207] Overall Loss 0.212264 Objective Loss 0.212264 LR 0.000125 Time 0.020817 -2023-02-13 18:37:24,808 - Epoch: [180][ 360/ 1207] Overall Loss 0.212341 Objective Loss 0.212341 LR 0.000125 Time 0.020760 -2023-02-13 18:37:24,995 - Epoch: [180][ 370/ 1207] Overall Loss 0.212151 Objective Loss 0.212151 LR 0.000125 Time 0.020704 -2023-02-13 18:37:25,183 - Epoch: [180][ 380/ 1207] Overall Loss 0.212048 Objective Loss 0.212048 LR 0.000125 Time 0.020653 -2023-02-13 18:37:25,370 - Epoch: [180][ 390/ 1207] Overall Loss 0.212279 Objective Loss 0.212279 LR 0.000125 Time 0.020602 -2023-02-13 18:37:25,558 - Epoch: [180][ 400/ 1207] Overall Loss 0.212567 Objective Loss 0.212567 LR 0.000125 Time 0.020554 -2023-02-13 18:37:25,745 - Epoch: [180][ 410/ 1207] Overall Loss 0.213084 Objective Loss 0.213084 LR 0.000125 Time 0.020510 -2023-02-13 18:37:25,935 - Epoch: [180][ 420/ 1207] Overall Loss 0.212750 Objective Loss 0.212750 LR 0.000125 Time 0.020472 -2023-02-13 18:37:26,123 - Epoch: [180][ 430/ 1207] Overall Loss 0.212586 Objective Loss 0.212586 LR 0.000125 Time 0.020432 -2023-02-13 18:37:26,311 - Epoch: [180][ 440/ 1207] Overall Loss 0.212541 Objective Loss 0.212541 LR 0.000125 Time 0.020394 -2023-02-13 18:37:26,499 - Epoch: [180][ 450/ 1207] Overall Loss 0.212420 Objective Loss 0.212420 LR 0.000125 Time 0.020358 -2023-02-13 18:37:26,687 - Epoch: [180][ 460/ 1207] Overall Loss 0.212041 Objective Loss 0.212041 LR 0.000125 Time 0.020324 -2023-02-13 18:37:26,877 - Epoch: [180][ 470/ 1207] Overall Loss 0.211718 Objective Loss 0.211718 LR 0.000125 Time 0.020294 -2023-02-13 18:37:27,064 - Epoch: [180][ 480/ 1207] Overall Loss 0.212136 Objective Loss 0.212136 LR 0.000125 Time 0.020261 -2023-02-13 18:37:27,252 - Epoch: [180][ 490/ 1207] Overall Loss 0.211933 Objective Loss 0.211933 LR 0.000125 Time 0.020229 -2023-02-13 18:37:27,440 - Epoch: [180][ 500/ 1207] Overall Loss 0.211920 Objective Loss 0.211920 LR 0.000125 Time 0.020200 -2023-02-13 18:37:27,627 - Epoch: [180][ 510/ 1207] Overall Loss 0.211723 Objective Loss 0.211723 LR 0.000125 Time 0.020171 -2023-02-13 18:37:27,816 - Epoch: [180][ 520/ 1207] Overall Loss 0.212039 Objective Loss 0.212039 LR 0.000125 Time 0.020145 -2023-02-13 18:37:28,004 - Epoch: [180][ 530/ 1207] Overall Loss 0.212131 Objective Loss 0.212131 LR 0.000125 Time 0.020119 -2023-02-13 18:37:28,192 - Epoch: [180][ 540/ 1207] Overall Loss 0.212173 Objective Loss 0.212173 LR 0.000125 Time 0.020094 -2023-02-13 18:37:28,380 - Epoch: [180][ 550/ 1207] Overall Loss 0.212258 Objective Loss 0.212258 LR 0.000125 Time 0.020069 -2023-02-13 18:37:28,567 - Epoch: [180][ 560/ 1207] Overall Loss 0.212092 Objective Loss 0.212092 LR 0.000125 Time 0.020046 -2023-02-13 18:37:28,755 - Epoch: [180][ 570/ 1207] Overall Loss 0.212126 Objective Loss 0.212126 LR 0.000125 Time 0.020023 -2023-02-13 18:37:28,944 - Epoch: [180][ 580/ 1207] Overall Loss 0.211927 Objective Loss 0.211927 LR 0.000125 Time 0.020002 -2023-02-13 18:37:29,132 - Epoch: [180][ 590/ 1207] Overall Loss 0.211567 Objective Loss 0.211567 LR 0.000125 Time 0.019982 -2023-02-13 18:37:29,320 - Epoch: [180][ 600/ 1207] Overall Loss 0.211343 Objective Loss 0.211343 LR 0.000125 Time 0.019961 -2023-02-13 18:37:29,508 - Epoch: [180][ 610/ 1207] Overall Loss 0.211201 Objective Loss 0.211201 LR 0.000125 Time 0.019941 -2023-02-13 18:37:29,696 - Epoch: [180][ 620/ 1207] Overall Loss 0.211159 Objective Loss 0.211159 LR 0.000125 Time 0.019923 -2023-02-13 18:37:29,885 - Epoch: [180][ 630/ 1207] Overall Loss 0.211353 Objective Loss 0.211353 LR 0.000125 Time 0.019905 -2023-02-13 18:37:30,073 - Epoch: [180][ 640/ 1207] Overall Loss 0.211274 Objective Loss 0.211274 LR 0.000125 Time 0.019888 -2023-02-13 18:37:30,261 - Epoch: [180][ 650/ 1207] Overall Loss 0.211362 Objective Loss 0.211362 LR 0.000125 Time 0.019870 -2023-02-13 18:37:30,449 - Epoch: [180][ 660/ 1207] Overall Loss 0.211112 Objective Loss 0.211112 LR 0.000125 Time 0.019854 -2023-02-13 18:37:30,637 - Epoch: [180][ 670/ 1207] Overall Loss 0.211016 Objective Loss 0.211016 LR 0.000125 Time 0.019838 -2023-02-13 18:37:30,826 - Epoch: [180][ 680/ 1207] Overall Loss 0.211038 Objective Loss 0.211038 LR 0.000125 Time 0.019823 -2023-02-13 18:37:31,015 - Epoch: [180][ 690/ 1207] Overall Loss 0.211197 Objective Loss 0.211197 LR 0.000125 Time 0.019809 -2023-02-13 18:37:31,203 - Epoch: [180][ 700/ 1207] Overall Loss 0.211198 Objective Loss 0.211198 LR 0.000125 Time 0.019794 -2023-02-13 18:37:31,391 - Epoch: [180][ 710/ 1207] Overall Loss 0.211377 Objective Loss 0.211377 LR 0.000125 Time 0.019780 -2023-02-13 18:37:31,578 - Epoch: [180][ 720/ 1207] Overall Loss 0.211443 Objective Loss 0.211443 LR 0.000125 Time 0.019765 -2023-02-13 18:37:31,767 - Epoch: [180][ 730/ 1207] Overall Loss 0.211580 Objective Loss 0.211580 LR 0.000125 Time 0.019751 -2023-02-13 18:37:31,956 - Epoch: [180][ 740/ 1207] Overall Loss 0.211508 Objective Loss 0.211508 LR 0.000125 Time 0.019740 -2023-02-13 18:37:32,144 - Epoch: [180][ 750/ 1207] Overall Loss 0.211740 Objective Loss 0.211740 LR 0.000125 Time 0.019727 -2023-02-13 18:37:32,331 - Epoch: [180][ 760/ 1207] Overall Loss 0.211753 Objective Loss 0.211753 LR 0.000125 Time 0.019713 -2023-02-13 18:37:32,519 - Epoch: [180][ 770/ 1207] Overall Loss 0.211514 Objective Loss 0.211514 LR 0.000125 Time 0.019701 -2023-02-13 18:37:32,707 - Epoch: [180][ 780/ 1207] Overall Loss 0.211592 Objective Loss 0.211592 LR 0.000125 Time 0.019688 -2023-02-13 18:37:32,896 - Epoch: [180][ 790/ 1207] Overall Loss 0.211547 Objective Loss 0.211547 LR 0.000125 Time 0.019678 -2023-02-13 18:37:33,084 - Epoch: [180][ 800/ 1207] Overall Loss 0.211866 Objective Loss 0.211866 LR 0.000125 Time 0.019667 -2023-02-13 18:37:33,272 - Epoch: [180][ 810/ 1207] Overall Loss 0.211715 Objective Loss 0.211715 LR 0.000125 Time 0.019656 -2023-02-13 18:37:33,460 - Epoch: [180][ 820/ 1207] Overall Loss 0.211636 Objective Loss 0.211636 LR 0.000125 Time 0.019645 -2023-02-13 18:37:33,648 - Epoch: [180][ 830/ 1207] Overall Loss 0.211328 Objective Loss 0.211328 LR 0.000125 Time 0.019634 -2023-02-13 18:37:33,836 - Epoch: [180][ 840/ 1207] Overall Loss 0.211292 Objective Loss 0.211292 LR 0.000125 Time 0.019624 -2023-02-13 18:37:34,026 - Epoch: [180][ 850/ 1207] Overall Loss 0.210963 Objective Loss 0.210963 LR 0.000125 Time 0.019615 -2023-02-13 18:37:34,214 - Epoch: [180][ 860/ 1207] Overall Loss 0.210853 Objective Loss 0.210853 LR 0.000125 Time 0.019606 -2023-02-13 18:37:34,403 - Epoch: [180][ 870/ 1207] Overall Loss 0.210758 Objective Loss 0.210758 LR 0.000125 Time 0.019597 -2023-02-13 18:37:34,592 - Epoch: [180][ 880/ 1207] Overall Loss 0.210526 Objective Loss 0.210526 LR 0.000125 Time 0.019589 -2023-02-13 18:37:34,780 - Epoch: [180][ 890/ 1207] Overall Loss 0.210573 Objective Loss 0.210573 LR 0.000125 Time 0.019579 -2023-02-13 18:37:34,969 - Epoch: [180][ 900/ 1207] Overall Loss 0.210566 Objective Loss 0.210566 LR 0.000125 Time 0.019571 -2023-02-13 18:37:35,157 - Epoch: [180][ 910/ 1207] Overall Loss 0.210497 Objective Loss 0.210497 LR 0.000125 Time 0.019563 -2023-02-13 18:37:35,346 - Epoch: [180][ 920/ 1207] Overall Loss 0.210356 Objective Loss 0.210356 LR 0.000125 Time 0.019555 -2023-02-13 18:37:35,535 - Epoch: [180][ 930/ 1207] Overall Loss 0.210222 Objective Loss 0.210222 LR 0.000125 Time 0.019547 -2023-02-13 18:37:35,723 - Epoch: [180][ 940/ 1207] Overall Loss 0.210162 Objective Loss 0.210162 LR 0.000125 Time 0.019540 -2023-02-13 18:37:35,914 - Epoch: [180][ 950/ 1207] Overall Loss 0.210298 Objective Loss 0.210298 LR 0.000125 Time 0.019534 -2023-02-13 18:37:36,103 - Epoch: [180][ 960/ 1207] Overall Loss 0.210436 Objective Loss 0.210436 LR 0.000125 Time 0.019527 -2023-02-13 18:37:36,291 - Epoch: [180][ 970/ 1207] Overall Loss 0.210483 Objective Loss 0.210483 LR 0.000125 Time 0.019520 -2023-02-13 18:37:36,480 - Epoch: [180][ 980/ 1207] Overall Loss 0.210268 Objective Loss 0.210268 LR 0.000125 Time 0.019513 -2023-02-13 18:37:36,668 - Epoch: [180][ 990/ 1207] Overall Loss 0.210170 Objective Loss 0.210170 LR 0.000125 Time 0.019505 -2023-02-13 18:37:36,857 - Epoch: [180][ 1000/ 1207] Overall Loss 0.210226 Objective Loss 0.210226 LR 0.000125 Time 0.019499 -2023-02-13 18:37:37,046 - Epoch: [180][ 1010/ 1207] Overall Loss 0.210095 Objective Loss 0.210095 LR 0.000125 Time 0.019493 -2023-02-13 18:37:37,234 - Epoch: [180][ 1020/ 1207] Overall Loss 0.210097 Objective Loss 0.210097 LR 0.000125 Time 0.019486 -2023-02-13 18:37:37,423 - Epoch: [180][ 1030/ 1207] Overall Loss 0.210279 Objective Loss 0.210279 LR 0.000125 Time 0.019479 -2023-02-13 18:37:37,611 - Epoch: [180][ 1040/ 1207] Overall Loss 0.210219 Objective Loss 0.210219 LR 0.000125 Time 0.019472 -2023-02-13 18:37:37,799 - Epoch: [180][ 1050/ 1207] Overall Loss 0.210076 Objective Loss 0.210076 LR 0.000125 Time 0.019466 -2023-02-13 18:37:37,988 - Epoch: [180][ 1060/ 1207] Overall Loss 0.210360 Objective Loss 0.210360 LR 0.000125 Time 0.019460 -2023-02-13 18:37:38,176 - Epoch: [180][ 1070/ 1207] Overall Loss 0.210256 Objective Loss 0.210256 LR 0.000125 Time 0.019453 -2023-02-13 18:37:38,364 - Epoch: [180][ 1080/ 1207] Overall Loss 0.210388 Objective Loss 0.210388 LR 0.000125 Time 0.019447 -2023-02-13 18:37:38,552 - Epoch: [180][ 1090/ 1207] Overall Loss 0.210474 Objective Loss 0.210474 LR 0.000125 Time 0.019441 -2023-02-13 18:37:38,741 - Epoch: [180][ 1100/ 1207] Overall Loss 0.210564 Objective Loss 0.210564 LR 0.000125 Time 0.019435 -2023-02-13 18:37:38,930 - Epoch: [180][ 1110/ 1207] Overall Loss 0.210722 Objective Loss 0.210722 LR 0.000125 Time 0.019430 -2023-02-13 18:37:39,118 - Epoch: [180][ 1120/ 1207] Overall Loss 0.210616 Objective Loss 0.210616 LR 0.000125 Time 0.019425 -2023-02-13 18:37:39,307 - Epoch: [180][ 1130/ 1207] Overall Loss 0.210805 Objective Loss 0.210805 LR 0.000125 Time 0.019420 -2023-02-13 18:37:39,495 - Epoch: [180][ 1140/ 1207] Overall Loss 0.210780 Objective Loss 0.210780 LR 0.000125 Time 0.019414 -2023-02-13 18:37:39,683 - Epoch: [180][ 1150/ 1207] Overall Loss 0.210909 Objective Loss 0.210909 LR 0.000125 Time 0.019408 -2023-02-13 18:37:39,873 - Epoch: [180][ 1160/ 1207] Overall Loss 0.210972 Objective Loss 0.210972 LR 0.000125 Time 0.019404 -2023-02-13 18:37:40,062 - Epoch: [180][ 1170/ 1207] Overall Loss 0.210906 Objective Loss 0.210906 LR 0.000125 Time 0.019399 -2023-02-13 18:37:40,249 - Epoch: [180][ 1180/ 1207] Overall Loss 0.210975 Objective Loss 0.210975 LR 0.000125 Time 0.019394 -2023-02-13 18:37:40,438 - Epoch: [180][ 1190/ 1207] Overall Loss 0.210895 Objective Loss 0.210895 LR 0.000125 Time 0.019389 -2023-02-13 18:37:40,675 - Epoch: [180][ 1200/ 1207] Overall Loss 0.210983 Objective Loss 0.210983 LR 0.000125 Time 0.019425 -2023-02-13 18:37:40,789 - Epoch: [180][ 1207/ 1207] Overall Loss 0.211024 Objective Loss 0.211024 Top1 88.719512 Top5 98.780488 LR 0.000125 Time 0.019407 -2023-02-13 18:37:40,861 - --- validate (epoch=180)----------- -2023-02-13 18:37:40,861 - 34311 samples (256 per mini-batch) -2023-02-13 18:37:41,269 - Epoch: [180][ 10/ 135] Loss 0.281306 Top1 85.273438 Top5 97.734375 -2023-02-13 18:37:41,398 - Epoch: [180][ 20/ 135] Loss 0.291604 Top1 85.234375 Top5 97.773438 -2023-02-13 18:37:41,523 - Epoch: [180][ 30/ 135] Loss 0.300891 Top1 85.638021 Top5 97.656250 -2023-02-13 18:37:41,646 - Epoch: [180][ 40/ 135] Loss 0.295897 Top1 85.722656 Top5 97.734375 -2023-02-13 18:37:41,772 - Epoch: [180][ 50/ 135] Loss 0.299806 Top1 85.546875 Top5 97.789062 -2023-02-13 18:37:41,896 - Epoch: [180][ 60/ 135] Loss 0.295637 Top1 85.611979 Top5 97.799479 -2023-02-13 18:37:42,019 - Epoch: [180][ 70/ 135] Loss 0.294254 Top1 85.658482 Top5 97.779018 -2023-02-13 18:37:42,164 - Epoch: [180][ 80/ 135] Loss 0.288838 Top1 85.722656 Top5 97.856445 -2023-02-13 18:37:42,302 - Epoch: [180][ 90/ 135] Loss 0.291326 Top1 85.655382 Top5 97.842882 -2023-02-13 18:37:42,445 - Epoch: [180][ 100/ 135] Loss 0.292153 Top1 85.578125 Top5 97.855469 -2023-02-13 18:37:42,591 - Epoch: [180][ 110/ 135] Loss 0.293707 Top1 85.518466 Top5 97.897727 -2023-02-13 18:37:42,735 - Epoch: [180][ 120/ 135] Loss 0.293779 Top1 85.426432 Top5 97.890625 -2023-02-13 18:37:42,870 - Epoch: [180][ 130/ 135] Loss 0.294872 Top1 85.459736 Top5 97.875601 -2023-02-13 18:37:42,914 - Epoch: [180][ 135/ 135] Loss 0.292622 Top1 85.447816 Top5 97.872402 -2023-02-13 18:37:42,982 - ==> Top1: 85.448 Top5: 97.872 Loss: 0.293 - -2023-02-13 18:37:42,983 - ==> Confusion: -[[ 871 5 7 1 5 2 0 2 4 43 0 5 1 3 4 3 2 1 0 2 6] - [ 4 952 2 1 12 21 2 13 3 1 1 1 1 0 0 1 5 1 2 2 8] - [ 7 7 967 9 5 1 12 15 0 1 4 3 0 3 1 4 4 2 4 1 8] - [ 4 0 19 915 0 3 1 1 2 3 11 0 6 1 18 4 3 5 13 0 7] - [ 12 8 0 0 995 10 1 2 2 0 0 4 3 4 5 5 6 0 1 3 5] - [ 2 14 1 5 5 977 3 18 1 3 1 8 1 12 1 1 6 0 2 3 6] - [ 1 2 10 2 0 4 1049 13 0 2 1 0 1 2 0 2 1 1 1 4 3] - [ 5 10 7 0 1 32 4 937 0 2 0 4 1 1 0 0 1 3 8 4 4] - [ 12 2 0 2 2 0 0 2 907 47 4 3 0 8 13 2 1 0 4 0 0] - [ 82 0 2 1 8 0 0 2 24 863 1 0 0 14 3 2 1 2 1 1 5] - [ 3 0 5 7 2 2 3 7 16 1 980 0 1 7 3 0 2 1 6 0 5] - [ 3 3 2 0 1 12 1 5 0 2 0 925 19 6 0 6 2 12 1 5 0] - [ 0 0 1 9 1 3 0 0 2 1 0 25 877 1 2 8 2 16 3 1 7] - [ 5 2 2 1 4 10 0 4 9 21 5 5 2 931 6 5 4 1 0 1 6] - [ 4 2 1 16 8 3 0 1 24 8 4 1 1 2 996 0 2 4 5 0 10] - [ 4 1 8 0 6 1 4 2 0 0 0 8 6 3 0 969 12 8 0 7 7] - [ 1 7 0 2 7 1 0 0 1 1 0 0 2 2 2 9 1008 1 3 3 11] - [ 5 1 0 4 0 2 1 0 0 0 1 7 10 2 0 17 3 990 0 2 6] - [ 4 4 8 10 1 2 0 24 3 0 4 0 4 0 13 1 0 3 1003 1 1] - [ 2 4 3 0 0 6 8 12 1 0 0 14 2 3 1 8 5 3 0 1071 5] - [ 155 218 228 115 136 206 82 174 91 89 176 112 293 265 143 94 247 116 143 216 10135]] - -2023-02-13 18:37:42,985 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:37:42,985 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:37:42,990 - - -2023-02-13 18:37:42,990 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:37:43,871 - Epoch: [181][ 10/ 1207] Overall Loss 0.223498 Objective Loss 0.223498 LR 0.000125 Time 0.088012 -2023-02-13 18:37:44,065 - Epoch: [181][ 20/ 1207] Overall Loss 0.211144 Objective Loss 0.211144 LR 0.000125 Time 0.053691 -2023-02-13 18:37:44,261 - Epoch: [181][ 30/ 1207] Overall Loss 0.205754 Objective Loss 0.205754 LR 0.000125 Time 0.042315 -2023-02-13 18:37:44,454 - Epoch: [181][ 40/ 1207] Overall Loss 0.206325 Objective Loss 0.206325 LR 0.000125 Time 0.036542 -2023-02-13 18:37:44,650 - Epoch: [181][ 50/ 1207] Overall Loss 0.204646 Objective Loss 0.204646 LR 0.000125 Time 0.033145 -2023-02-13 18:37:44,843 - Epoch: [181][ 60/ 1207] Overall Loss 0.205865 Objective Loss 0.205865 LR 0.000125 Time 0.030834 -2023-02-13 18:37:45,039 - Epoch: [181][ 70/ 1207] Overall Loss 0.204963 Objective Loss 0.204963 LR 0.000125 Time 0.029222 -2023-02-13 18:37:45,231 - Epoch: [181][ 80/ 1207] Overall Loss 0.202979 Objective Loss 0.202979 LR 0.000125 Time 0.027970 -2023-02-13 18:37:45,427 - Epoch: [181][ 90/ 1207] Overall Loss 0.202474 Objective Loss 0.202474 LR 0.000125 Time 0.027037 -2023-02-13 18:37:45,621 - Epoch: [181][ 100/ 1207] Overall Loss 0.203068 Objective Loss 0.203068 LR 0.000125 Time 0.026263 -2023-02-13 18:37:45,817 - Epoch: [181][ 110/ 1207] Overall Loss 0.203106 Objective Loss 0.203106 LR 0.000125 Time 0.025657 -2023-02-13 18:37:46,012 - Epoch: [181][ 120/ 1207] Overall Loss 0.203488 Objective Loss 0.203488 LR 0.000125 Time 0.025141 -2023-02-13 18:37:46,208 - Epoch: [181][ 130/ 1207] Overall Loss 0.205385 Objective Loss 0.205385 LR 0.000125 Time 0.024707 -2023-02-13 18:37:46,400 - Epoch: [181][ 140/ 1207] Overall Loss 0.204263 Objective Loss 0.204263 LR 0.000125 Time 0.024316 -2023-02-13 18:37:46,596 - Epoch: [181][ 150/ 1207] Overall Loss 0.205572 Objective Loss 0.205572 LR 0.000125 Time 0.023996 -2023-02-13 18:37:46,789 - Epoch: [181][ 160/ 1207] Overall Loss 0.204915 Objective Loss 0.204915 LR 0.000125 Time 0.023699 -2023-02-13 18:37:46,986 - Epoch: [181][ 170/ 1207] Overall Loss 0.203464 Objective Loss 0.203464 LR 0.000125 Time 0.023464 -2023-02-13 18:37:47,179 - Epoch: [181][ 180/ 1207] Overall Loss 0.202604 Objective Loss 0.202604 LR 0.000125 Time 0.023230 -2023-02-13 18:37:47,375 - Epoch: [181][ 190/ 1207] Overall Loss 0.203095 Objective Loss 0.203095 LR 0.000125 Time 0.023038 -2023-02-13 18:37:47,568 - Epoch: [181][ 200/ 1207] Overall Loss 0.203294 Objective Loss 0.203294 LR 0.000125 Time 0.022849 -2023-02-13 18:37:47,766 - Epoch: [181][ 210/ 1207] Overall Loss 0.204374 Objective Loss 0.204374 LR 0.000125 Time 0.022703 -2023-02-13 18:37:47,961 - Epoch: [181][ 220/ 1207] Overall Loss 0.203831 Objective Loss 0.203831 LR 0.000125 Time 0.022556 -2023-02-13 18:37:48,160 - Epoch: [181][ 230/ 1207] Overall Loss 0.204332 Objective Loss 0.204332 LR 0.000125 Time 0.022436 -2023-02-13 18:37:48,355 - Epoch: [181][ 240/ 1207] Overall Loss 0.204622 Objective Loss 0.204622 LR 0.000125 Time 0.022313 -2023-02-13 18:37:48,554 - Epoch: [181][ 250/ 1207] Overall Loss 0.204671 Objective Loss 0.204671 LR 0.000125 Time 0.022214 -2023-02-13 18:37:48,748 - Epoch: [181][ 260/ 1207] Overall Loss 0.205129 Objective Loss 0.205129 LR 0.000125 Time 0.022106 -2023-02-13 18:37:48,940 - Epoch: [181][ 270/ 1207] Overall Loss 0.204264 Objective Loss 0.204264 LR 0.000125 Time 0.021997 -2023-02-13 18:37:49,131 - Epoch: [181][ 280/ 1207] Overall Loss 0.204415 Objective Loss 0.204415 LR 0.000125 Time 0.021891 -2023-02-13 18:37:49,322 - Epoch: [181][ 290/ 1207] Overall Loss 0.204839 Objective Loss 0.204839 LR 0.000125 Time 0.021795 -2023-02-13 18:37:49,512 - Epoch: [181][ 300/ 1207] Overall Loss 0.205096 Objective Loss 0.205096 LR 0.000125 Time 0.021701 -2023-02-13 18:37:49,704 - Epoch: [181][ 310/ 1207] Overall Loss 0.204590 Objective Loss 0.204590 LR 0.000125 Time 0.021619 -2023-02-13 18:37:49,895 - Epoch: [181][ 320/ 1207] Overall Loss 0.205561 Objective Loss 0.205561 LR 0.000125 Time 0.021538 -2023-02-13 18:37:50,086 - Epoch: [181][ 330/ 1207] Overall Loss 0.205805 Objective Loss 0.205805 LR 0.000125 Time 0.021464 -2023-02-13 18:37:50,276 - Epoch: [181][ 340/ 1207] Overall Loss 0.206115 Objective Loss 0.206115 LR 0.000125 Time 0.021389 -2023-02-13 18:37:50,467 - Epoch: [181][ 350/ 1207] Overall Loss 0.205961 Objective Loss 0.205961 LR 0.000125 Time 0.021323 -2023-02-13 18:37:50,657 - Epoch: [181][ 360/ 1207] Overall Loss 0.205852 Objective Loss 0.205852 LR 0.000125 Time 0.021258 -2023-02-13 18:37:50,849 - Epoch: [181][ 370/ 1207] Overall Loss 0.206276 Objective Loss 0.206276 LR 0.000125 Time 0.021201 -2023-02-13 18:37:51,041 - Epoch: [181][ 380/ 1207] Overall Loss 0.205941 Objective Loss 0.205941 LR 0.000125 Time 0.021146 -2023-02-13 18:37:51,232 - Epoch: [181][ 390/ 1207] Overall Loss 0.205608 Objective Loss 0.205608 LR 0.000125 Time 0.021095 -2023-02-13 18:37:51,423 - Epoch: [181][ 400/ 1207] Overall Loss 0.205232 Objective Loss 0.205232 LR 0.000125 Time 0.021042 -2023-02-13 18:37:51,614 - Epoch: [181][ 410/ 1207] Overall Loss 0.205144 Objective Loss 0.205144 LR 0.000125 Time 0.020995 -2023-02-13 18:37:51,804 - Epoch: [181][ 420/ 1207] Overall Loss 0.205167 Objective Loss 0.205167 LR 0.000125 Time 0.020948 -2023-02-13 18:37:51,997 - Epoch: [181][ 430/ 1207] Overall Loss 0.204728 Objective Loss 0.204728 LR 0.000125 Time 0.020907 -2023-02-13 18:37:52,187 - Epoch: [181][ 440/ 1207] Overall Loss 0.205368 Objective Loss 0.205368 LR 0.000125 Time 0.020863 -2023-02-13 18:37:52,378 - Epoch: [181][ 450/ 1207] Overall Loss 0.205351 Objective Loss 0.205351 LR 0.000125 Time 0.020824 -2023-02-13 18:37:52,570 - Epoch: [181][ 460/ 1207] Overall Loss 0.205106 Objective Loss 0.205106 LR 0.000125 Time 0.020786 -2023-02-13 18:37:52,761 - Epoch: [181][ 470/ 1207] Overall Loss 0.205413 Objective Loss 0.205413 LR 0.000125 Time 0.020751 -2023-02-13 18:37:52,954 - Epoch: [181][ 480/ 1207] Overall Loss 0.205611 Objective Loss 0.205611 LR 0.000125 Time 0.020719 -2023-02-13 18:37:53,145 - Epoch: [181][ 490/ 1207] Overall Loss 0.206087 Objective Loss 0.206087 LR 0.000125 Time 0.020686 -2023-02-13 18:37:53,336 - Epoch: [181][ 500/ 1207] Overall Loss 0.206011 Objective Loss 0.206011 LR 0.000125 Time 0.020652 -2023-02-13 18:37:53,528 - Epoch: [181][ 510/ 1207] Overall Loss 0.206234 Objective Loss 0.206234 LR 0.000125 Time 0.020623 -2023-02-13 18:37:53,717 - Epoch: [181][ 520/ 1207] Overall Loss 0.206174 Objective Loss 0.206174 LR 0.000125 Time 0.020590 -2023-02-13 18:37:53,910 - Epoch: [181][ 530/ 1207] Overall Loss 0.206190 Objective Loss 0.206190 LR 0.000125 Time 0.020564 -2023-02-13 18:37:54,101 - Epoch: [181][ 540/ 1207] Overall Loss 0.206045 Objective Loss 0.206045 LR 0.000125 Time 0.020537 -2023-02-13 18:37:54,292 - Epoch: [181][ 550/ 1207] Overall Loss 0.206319 Objective Loss 0.206319 LR 0.000125 Time 0.020510 -2023-02-13 18:37:54,482 - Epoch: [181][ 560/ 1207] Overall Loss 0.206429 Objective Loss 0.206429 LR 0.000125 Time 0.020483 -2023-02-13 18:37:54,674 - Epoch: [181][ 570/ 1207] Overall Loss 0.206293 Objective Loss 0.206293 LR 0.000125 Time 0.020459 -2023-02-13 18:37:54,865 - Epoch: [181][ 580/ 1207] Overall Loss 0.206425 Objective Loss 0.206425 LR 0.000125 Time 0.020435 -2023-02-13 18:37:55,054 - Epoch: [181][ 590/ 1207] Overall Loss 0.206391 Objective Loss 0.206391 LR 0.000125 Time 0.020409 -2023-02-13 18:37:55,242 - Epoch: [181][ 600/ 1207] Overall Loss 0.206282 Objective Loss 0.206282 LR 0.000125 Time 0.020382 -2023-02-13 18:37:55,431 - Epoch: [181][ 610/ 1207] Overall Loss 0.206116 Objective Loss 0.206116 LR 0.000125 Time 0.020356 -2023-02-13 18:37:55,618 - Epoch: [181][ 620/ 1207] Overall Loss 0.206333 Objective Loss 0.206333 LR 0.000125 Time 0.020329 -2023-02-13 18:37:55,807 - Epoch: [181][ 630/ 1207] Overall Loss 0.206445 Objective Loss 0.206445 LR 0.000125 Time 0.020305 -2023-02-13 18:37:55,997 - Epoch: [181][ 640/ 1207] Overall Loss 0.206530 Objective Loss 0.206530 LR 0.000125 Time 0.020285 -2023-02-13 18:37:56,186 - Epoch: [181][ 650/ 1207] Overall Loss 0.206579 Objective Loss 0.206579 LR 0.000125 Time 0.020262 -2023-02-13 18:37:56,373 - Epoch: [181][ 660/ 1207] Overall Loss 0.206839 Objective Loss 0.206839 LR 0.000125 Time 0.020239 -2023-02-13 18:37:56,563 - Epoch: [181][ 670/ 1207] Overall Loss 0.207182 Objective Loss 0.207182 LR 0.000125 Time 0.020219 -2023-02-13 18:37:56,751 - Epoch: [181][ 680/ 1207] Overall Loss 0.207302 Objective Loss 0.207302 LR 0.000125 Time 0.020198 -2023-02-13 18:37:56,941 - Epoch: [181][ 690/ 1207] Overall Loss 0.207294 Objective Loss 0.207294 LR 0.000125 Time 0.020180 -2023-02-13 18:37:57,130 - Epoch: [181][ 700/ 1207] Overall Loss 0.207498 Objective Loss 0.207498 LR 0.000125 Time 0.020161 -2023-02-13 18:37:57,319 - Epoch: [181][ 710/ 1207] Overall Loss 0.207636 Objective Loss 0.207636 LR 0.000125 Time 0.020143 -2023-02-13 18:37:57,508 - Epoch: [181][ 720/ 1207] Overall Loss 0.207763 Objective Loss 0.207763 LR 0.000125 Time 0.020125 -2023-02-13 18:37:57,697 - Epoch: [181][ 730/ 1207] Overall Loss 0.207516 Objective Loss 0.207516 LR 0.000125 Time 0.020107 -2023-02-13 18:37:57,886 - Epoch: [181][ 740/ 1207] Overall Loss 0.207291 Objective Loss 0.207291 LR 0.000125 Time 0.020091 -2023-02-13 18:37:58,076 - Epoch: [181][ 750/ 1207] Overall Loss 0.207218 Objective Loss 0.207218 LR 0.000125 Time 0.020076 -2023-02-13 18:37:58,264 - Epoch: [181][ 760/ 1207] Overall Loss 0.207125 Objective Loss 0.207125 LR 0.000125 Time 0.020059 -2023-02-13 18:37:58,453 - Epoch: [181][ 770/ 1207] Overall Loss 0.207046 Objective Loss 0.207046 LR 0.000125 Time 0.020043 -2023-02-13 18:37:58,641 - Epoch: [181][ 780/ 1207] Overall Loss 0.207237 Objective Loss 0.207237 LR 0.000125 Time 0.020027 -2023-02-13 18:37:58,829 - Epoch: [181][ 790/ 1207] Overall Loss 0.207350 Objective Loss 0.207350 LR 0.000125 Time 0.020012 -2023-02-13 18:37:59,019 - Epoch: [181][ 800/ 1207] Overall Loss 0.207223 Objective Loss 0.207223 LR 0.000125 Time 0.019998 -2023-02-13 18:37:59,208 - Epoch: [181][ 810/ 1207] Overall Loss 0.207261 Objective Loss 0.207261 LR 0.000125 Time 0.019983 -2023-02-13 18:37:59,396 - Epoch: [181][ 820/ 1207] Overall Loss 0.207370 Objective Loss 0.207370 LR 0.000125 Time 0.019969 -2023-02-13 18:37:59,585 - Epoch: [181][ 830/ 1207] Overall Loss 0.207388 Objective Loss 0.207388 LR 0.000125 Time 0.019955 -2023-02-13 18:37:59,773 - Epoch: [181][ 840/ 1207] Overall Loss 0.207695 Objective Loss 0.207695 LR 0.000125 Time 0.019941 -2023-02-13 18:37:59,963 - Epoch: [181][ 850/ 1207] Overall Loss 0.208003 Objective Loss 0.208003 LR 0.000125 Time 0.019929 -2023-02-13 18:38:00,151 - Epoch: [181][ 860/ 1207] Overall Loss 0.207870 Objective Loss 0.207870 LR 0.000125 Time 0.019916 -2023-02-13 18:38:00,339 - Epoch: [181][ 870/ 1207] Overall Loss 0.207804 Objective Loss 0.207804 LR 0.000125 Time 0.019903 -2023-02-13 18:38:00,528 - Epoch: [181][ 880/ 1207] Overall Loss 0.207981 Objective Loss 0.207981 LR 0.000125 Time 0.019891 -2023-02-13 18:38:00,716 - Epoch: [181][ 890/ 1207] Overall Loss 0.208041 Objective Loss 0.208041 LR 0.000125 Time 0.019879 -2023-02-13 18:38:00,906 - Epoch: [181][ 900/ 1207] Overall Loss 0.208331 Objective Loss 0.208331 LR 0.000125 Time 0.019869 -2023-02-13 18:38:01,095 - Epoch: [181][ 910/ 1207] Overall Loss 0.208576 Objective Loss 0.208576 LR 0.000125 Time 0.019858 -2023-02-13 18:38:01,283 - Epoch: [181][ 920/ 1207] Overall Loss 0.208548 Objective Loss 0.208548 LR 0.000125 Time 0.019846 -2023-02-13 18:38:01,473 - Epoch: [181][ 930/ 1207] Overall Loss 0.208518 Objective Loss 0.208518 LR 0.000125 Time 0.019835 -2023-02-13 18:38:01,661 - Epoch: [181][ 940/ 1207] Overall Loss 0.208372 Objective Loss 0.208372 LR 0.000125 Time 0.019825 -2023-02-13 18:38:01,850 - Epoch: [181][ 950/ 1207] Overall Loss 0.208396 Objective Loss 0.208396 LR 0.000125 Time 0.019814 -2023-02-13 18:38:02,039 - Epoch: [181][ 960/ 1207] Overall Loss 0.208428 Objective Loss 0.208428 LR 0.000125 Time 0.019805 -2023-02-13 18:38:02,228 - Epoch: [181][ 970/ 1207] Overall Loss 0.208510 Objective Loss 0.208510 LR 0.000125 Time 0.019795 -2023-02-13 18:38:02,417 - Epoch: [181][ 980/ 1207] Overall Loss 0.208582 Objective Loss 0.208582 LR 0.000125 Time 0.019785 -2023-02-13 18:38:02,605 - Epoch: [181][ 990/ 1207] Overall Loss 0.208633 Objective Loss 0.208633 LR 0.000125 Time 0.019775 -2023-02-13 18:38:02,793 - Epoch: [181][ 1000/ 1207] Overall Loss 0.208545 Objective Loss 0.208545 LR 0.000125 Time 0.019765 -2023-02-13 18:38:02,983 - Epoch: [181][ 1010/ 1207] Overall Loss 0.208502 Objective Loss 0.208502 LR 0.000125 Time 0.019756 -2023-02-13 18:38:03,171 - Epoch: [181][ 1020/ 1207] Overall Loss 0.208517 Objective Loss 0.208517 LR 0.000125 Time 0.019747 -2023-02-13 18:38:03,360 - Epoch: [181][ 1030/ 1207] Overall Loss 0.208318 Objective Loss 0.208318 LR 0.000125 Time 0.019738 -2023-02-13 18:38:03,549 - Epoch: [181][ 1040/ 1207] Overall Loss 0.208167 Objective Loss 0.208167 LR 0.000125 Time 0.019729 -2023-02-13 18:38:03,737 - Epoch: [181][ 1050/ 1207] Overall Loss 0.208238 Objective Loss 0.208238 LR 0.000125 Time 0.019721 -2023-02-13 18:38:03,926 - Epoch: [181][ 1060/ 1207] Overall Loss 0.208233 Objective Loss 0.208233 LR 0.000125 Time 0.019713 -2023-02-13 18:38:04,116 - Epoch: [181][ 1070/ 1207] Overall Loss 0.208525 Objective Loss 0.208525 LR 0.000125 Time 0.019705 -2023-02-13 18:38:04,304 - Epoch: [181][ 1080/ 1207] Overall Loss 0.208596 Objective Loss 0.208596 LR 0.000125 Time 0.019697 -2023-02-13 18:38:04,493 - Epoch: [181][ 1090/ 1207] Overall Loss 0.208658 Objective Loss 0.208658 LR 0.000125 Time 0.019689 -2023-02-13 18:38:04,682 - Epoch: [181][ 1100/ 1207] Overall Loss 0.208527 Objective Loss 0.208527 LR 0.000125 Time 0.019682 -2023-02-13 18:38:04,871 - Epoch: [181][ 1110/ 1207] Overall Loss 0.208404 Objective Loss 0.208404 LR 0.000125 Time 0.019675 -2023-02-13 18:38:05,063 - Epoch: [181][ 1120/ 1207] Overall Loss 0.208387 Objective Loss 0.208387 LR 0.000125 Time 0.019669 -2023-02-13 18:38:05,257 - Epoch: [181][ 1130/ 1207] Overall Loss 0.208292 Objective Loss 0.208292 LR 0.000125 Time 0.019667 -2023-02-13 18:38:05,446 - Epoch: [181][ 1140/ 1207] Overall Loss 0.208410 Objective Loss 0.208410 LR 0.000125 Time 0.019660 -2023-02-13 18:38:05,635 - Epoch: [181][ 1150/ 1207] Overall Loss 0.208501 Objective Loss 0.208501 LR 0.000125 Time 0.019653 -2023-02-13 18:38:05,823 - Epoch: [181][ 1160/ 1207] Overall Loss 0.208336 Objective Loss 0.208336 LR 0.000125 Time 0.019646 -2023-02-13 18:38:06,014 - Epoch: [181][ 1170/ 1207] Overall Loss 0.208551 Objective Loss 0.208551 LR 0.000125 Time 0.019640 -2023-02-13 18:38:06,203 - Epoch: [181][ 1180/ 1207] Overall Loss 0.208582 Objective Loss 0.208582 LR 0.000125 Time 0.019634 -2023-02-13 18:38:06,393 - Epoch: [181][ 1190/ 1207] Overall Loss 0.208549 Objective Loss 0.208549 LR 0.000125 Time 0.019628 -2023-02-13 18:38:06,639 - Epoch: [181][ 1200/ 1207] Overall Loss 0.208326 Objective Loss 0.208326 LR 0.000125 Time 0.019669 -2023-02-13 18:38:06,754 - Epoch: [181][ 1207/ 1207] Overall Loss 0.208370 Objective Loss 0.208370 Top1 86.280488 Top5 96.951220 LR 0.000125 Time 0.019650 -2023-02-13 18:38:06,827 - --- validate (epoch=181)----------- -2023-02-13 18:38:06,827 - 34311 samples (256 per mini-batch) -2023-02-13 18:38:07,337 - Epoch: [181][ 10/ 135] Loss 0.272564 Top1 85.820312 Top5 97.578125 -2023-02-13 18:38:07,461 - Epoch: [181][ 20/ 135] Loss 0.286015 Top1 85.644531 Top5 97.695312 -2023-02-13 18:38:07,586 - Epoch: [181][ 30/ 135] Loss 0.281733 Top1 85.768229 Top5 97.656250 -2023-02-13 18:38:07,708 - Epoch: [181][ 40/ 135] Loss 0.286502 Top1 85.537109 Top5 97.578125 -2023-02-13 18:38:07,833 - Epoch: [181][ 50/ 135] Loss 0.288017 Top1 85.250000 Top5 97.656250 -2023-02-13 18:38:08,006 - Epoch: [181][ 60/ 135] Loss 0.291818 Top1 85.065104 Top5 97.584635 -2023-02-13 18:38:08,135 - Epoch: [181][ 70/ 135] Loss 0.294003 Top1 85.039062 Top5 97.617188 -2023-02-13 18:38:08,278 - Epoch: [181][ 80/ 135] Loss 0.296054 Top1 84.995117 Top5 97.617188 -2023-02-13 18:38:08,415 - Epoch: [181][ 90/ 135] Loss 0.293144 Top1 85.130208 Top5 97.617188 -2023-02-13 18:38:08,550 - Epoch: [181][ 100/ 135] Loss 0.289979 Top1 85.191406 Top5 97.648438 -2023-02-13 18:38:08,684 - Epoch: [181][ 110/ 135] Loss 0.287535 Top1 85.223722 Top5 97.730824 -2023-02-13 18:38:08,807 - Epoch: [181][ 120/ 135] Loss 0.289972 Top1 85.195312 Top5 97.708333 -2023-02-13 18:38:08,935 - Epoch: [181][ 130/ 135] Loss 0.292258 Top1 85.216346 Top5 97.725361 -2023-02-13 18:38:08,979 - Epoch: [181][ 135/ 135] Loss 0.304025 Top1 85.278774 Top5 97.744164 -2023-02-13 18:38:09,047 - ==> Top1: 85.279 Top5: 97.744 Loss: 0.304 - -2023-02-13 18:38:09,048 - ==> Confusion: -[[ 863 4 7 2 11 3 0 2 3 42 0 2 2 4 7 1 3 2 0 2 7] - [ 3 948 1 1 9 23 2 15 2 1 2 0 2 0 0 3 5 1 4 5 6] - [ 3 6 967 11 3 1 14 14 0 2 4 2 0 4 5 6 3 3 4 4 2] - [ 4 1 17 910 2 4 0 1 2 3 14 0 7 1 16 3 5 4 15 1 6] - [ 6 9 0 0 995 11 1 2 3 0 1 7 2 5 6 7 4 0 1 3 3] - [ 2 16 0 5 4 971 3 10 2 2 1 11 4 17 2 4 7 2 1 3 3] - [ 1 2 10 1 0 6 1053 6 0 2 1 0 3 2 0 0 1 0 1 5 5] - [ 1 8 7 0 0 28 0 939 1 1 0 10 2 2 0 0 1 1 11 7 5] - [ 12 4 1 1 1 0 0 1 919 30 3 3 0 9 15 3 1 1 3 1 1] - [ 60 2 2 0 10 0 0 4 37 871 0 2 0 14 3 1 1 1 1 0 3] - [ 2 1 4 5 2 2 3 3 15 0 993 2 2 4 3 0 1 1 4 0 4] - [ 1 2 3 0 0 13 0 6 2 3 0 919 17 9 0 6 2 9 1 10 2] - [ 0 0 0 5 2 2 0 0 2 2 0 30 889 2 1 4 4 8 1 0 7] - [ 2 3 0 1 6 7 1 0 11 15 7 4 1 938 7 5 5 4 1 1 5] - [ 4 1 2 14 2 6 0 1 22 6 2 0 2 1 1013 1 1 4 5 0 5] - [ 3 3 7 0 4 0 3 0 0 0 0 8 9 3 0 973 11 8 1 7 6] - [ 1 6 0 1 9 1 0 0 2 0 0 1 3 2 2 11 1003 0 2 3 14] - [ 4 2 0 3 1 3 3 0 0 1 0 6 13 2 0 11 1 993 0 2 6] - [ 2 5 4 7 0 2 1 22 5 0 6 1 2 0 11 1 1 3 1011 1 1] - [ 0 3 0 0 2 3 5 6 0 0 0 16 2 1 1 5 1 3 1 1094 5] - [ 135 248 213 116 139 202 99 163 93 75 190 105 334 273 179 80 248 110 167 266 9999]] - -2023-02-13 18:38:09,049 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:38:09,049 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:38:09,055 - - -2023-02-13 18:38:09,055 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:38:09,997 - Epoch: [182][ 10/ 1207] Overall Loss 0.201208 Objective Loss 0.201208 LR 0.000125 Time 0.094103 -2023-02-13 18:38:10,201 - Epoch: [182][ 20/ 1207] Overall Loss 0.207040 Objective Loss 0.207040 LR 0.000125 Time 0.057248 -2023-02-13 18:38:10,395 - Epoch: [182][ 30/ 1207] Overall Loss 0.211441 Objective Loss 0.211441 LR 0.000125 Time 0.044607 -2023-02-13 18:38:10,590 - Epoch: [182][ 40/ 1207] Overall Loss 0.208978 Objective Loss 0.208978 LR 0.000125 Time 0.038329 -2023-02-13 18:38:10,783 - Epoch: [182][ 50/ 1207] Overall Loss 0.213254 Objective Loss 0.213254 LR 0.000125 Time 0.034519 -2023-02-13 18:38:10,980 - Epoch: [182][ 60/ 1207] Overall Loss 0.209718 Objective Loss 0.209718 LR 0.000125 Time 0.032039 -2023-02-13 18:38:11,173 - Epoch: [182][ 70/ 1207] Overall Loss 0.208835 Objective Loss 0.208835 LR 0.000125 Time 0.030216 -2023-02-13 18:38:11,369 - Epoch: [182][ 80/ 1207] Overall Loss 0.211340 Objective Loss 0.211340 LR 0.000125 Time 0.028878 -2023-02-13 18:38:11,562 - Epoch: [182][ 90/ 1207] Overall Loss 0.209164 Objective Loss 0.209164 LR 0.000125 Time 0.027812 -2023-02-13 18:38:11,758 - Epoch: [182][ 100/ 1207] Overall Loss 0.207683 Objective Loss 0.207683 LR 0.000125 Time 0.026983 -2023-02-13 18:38:11,952 - Epoch: [182][ 110/ 1207] Overall Loss 0.209131 Objective Loss 0.209131 LR 0.000125 Time 0.026290 -2023-02-13 18:38:12,148 - Epoch: [182][ 120/ 1207] Overall Loss 0.210146 Objective Loss 0.210146 LR 0.000125 Time 0.025737 -2023-02-13 18:38:12,342 - Epoch: [182][ 130/ 1207] Overall Loss 0.209960 Objective Loss 0.209960 LR 0.000125 Time 0.025242 -2023-02-13 18:38:12,537 - Epoch: [182][ 140/ 1207] Overall Loss 0.209368 Objective Loss 0.209368 LR 0.000125 Time 0.024834 -2023-02-13 18:38:12,731 - Epoch: [182][ 150/ 1207] Overall Loss 0.208853 Objective Loss 0.208853 LR 0.000125 Time 0.024466 -2023-02-13 18:38:12,927 - Epoch: [182][ 160/ 1207] Overall Loss 0.207227 Objective Loss 0.207227 LR 0.000125 Time 0.024162 -2023-02-13 18:38:13,122 - Epoch: [182][ 170/ 1207] Overall Loss 0.207161 Objective Loss 0.207161 LR 0.000125 Time 0.023886 -2023-02-13 18:38:13,318 - Epoch: [182][ 180/ 1207] Overall Loss 0.206114 Objective Loss 0.206114 LR 0.000125 Time 0.023641 -2023-02-13 18:38:13,511 - Epoch: [182][ 190/ 1207] Overall Loss 0.206206 Objective Loss 0.206206 LR 0.000125 Time 0.023411 -2023-02-13 18:38:13,706 - Epoch: [182][ 200/ 1207] Overall Loss 0.206481 Objective Loss 0.206481 LR 0.000125 Time 0.023216 -2023-02-13 18:38:13,900 - Epoch: [182][ 210/ 1207] Overall Loss 0.207323 Objective Loss 0.207323 LR 0.000125 Time 0.023032 -2023-02-13 18:38:14,097 - Epoch: [182][ 220/ 1207] Overall Loss 0.208279 Objective Loss 0.208279 LR 0.000125 Time 0.022878 -2023-02-13 18:38:14,290 - Epoch: [182][ 230/ 1207] Overall Loss 0.207547 Objective Loss 0.207547 LR 0.000125 Time 0.022722 -2023-02-13 18:38:14,486 - Epoch: [182][ 240/ 1207] Overall Loss 0.207737 Objective Loss 0.207737 LR 0.000125 Time 0.022590 -2023-02-13 18:38:14,679 - Epoch: [182][ 250/ 1207] Overall Loss 0.207013 Objective Loss 0.207013 LR 0.000125 Time 0.022457 -2023-02-13 18:38:14,875 - Epoch: [182][ 260/ 1207] Overall Loss 0.207006 Objective Loss 0.207006 LR 0.000125 Time 0.022346 -2023-02-13 18:38:15,069 - Epoch: [182][ 270/ 1207] Overall Loss 0.206408 Objective Loss 0.206408 LR 0.000125 Time 0.022235 -2023-02-13 18:38:15,265 - Epoch: [182][ 280/ 1207] Overall Loss 0.207322 Objective Loss 0.207322 LR 0.000125 Time 0.022140 -2023-02-13 18:38:15,458 - Epoch: [182][ 290/ 1207] Overall Loss 0.206456 Objective Loss 0.206456 LR 0.000125 Time 0.022042 -2023-02-13 18:38:15,654 - Epoch: [182][ 300/ 1207] Overall Loss 0.206211 Objective Loss 0.206211 LR 0.000125 Time 0.021958 -2023-02-13 18:38:15,847 - Epoch: [182][ 310/ 1207] Overall Loss 0.206743 Objective Loss 0.206743 LR 0.000125 Time 0.021872 -2023-02-13 18:38:16,045 - Epoch: [182][ 320/ 1207] Overall Loss 0.206888 Objective Loss 0.206888 LR 0.000125 Time 0.021806 -2023-02-13 18:38:16,239 - Epoch: [182][ 330/ 1207] Overall Loss 0.207355 Objective Loss 0.207355 LR 0.000125 Time 0.021730 -2023-02-13 18:38:16,435 - Epoch: [182][ 340/ 1207] Overall Loss 0.207431 Objective Loss 0.207431 LR 0.000125 Time 0.021666 -2023-02-13 18:38:16,628 - Epoch: [182][ 350/ 1207] Overall Loss 0.207312 Objective Loss 0.207312 LR 0.000125 Time 0.021599 -2023-02-13 18:38:16,824 - Epoch: [182][ 360/ 1207] Overall Loss 0.206936 Objective Loss 0.206936 LR 0.000125 Time 0.021542 -2023-02-13 18:38:17,018 - Epoch: [182][ 370/ 1207] Overall Loss 0.207357 Objective Loss 0.207357 LR 0.000125 Time 0.021484 -2023-02-13 18:38:17,214 - Epoch: [182][ 380/ 1207] Overall Loss 0.207619 Objective Loss 0.207619 LR 0.000125 Time 0.021434 -2023-02-13 18:38:17,408 - Epoch: [182][ 390/ 1207] Overall Loss 0.207513 Objective Loss 0.207513 LR 0.000125 Time 0.021379 -2023-02-13 18:38:17,604 - Epoch: [182][ 400/ 1207] Overall Loss 0.207841 Objective Loss 0.207841 LR 0.000125 Time 0.021334 -2023-02-13 18:38:17,797 - Epoch: [182][ 410/ 1207] Overall Loss 0.208143 Objective Loss 0.208143 LR 0.000125 Time 0.021284 -2023-02-13 18:38:17,994 - Epoch: [182][ 420/ 1207] Overall Loss 0.208148 Objective Loss 0.208148 LR 0.000125 Time 0.021245 -2023-02-13 18:38:18,188 - Epoch: [182][ 430/ 1207] Overall Loss 0.208077 Objective Loss 0.208077 LR 0.000125 Time 0.021202 -2023-02-13 18:38:18,384 - Epoch: [182][ 440/ 1207] Overall Loss 0.208274 Objective Loss 0.208274 LR 0.000125 Time 0.021165 -2023-02-13 18:38:18,578 - Epoch: [182][ 450/ 1207] Overall Loss 0.208442 Objective Loss 0.208442 LR 0.000125 Time 0.021124 -2023-02-13 18:38:18,774 - Epoch: [182][ 460/ 1207] Overall Loss 0.208200 Objective Loss 0.208200 LR 0.000125 Time 0.021090 -2023-02-13 18:38:18,968 - Epoch: [182][ 470/ 1207] Overall Loss 0.208310 Objective Loss 0.208310 LR 0.000125 Time 0.021053 -2023-02-13 18:38:19,165 - Epoch: [182][ 480/ 1207] Overall Loss 0.208287 Objective Loss 0.208287 LR 0.000125 Time 0.021024 -2023-02-13 18:38:19,358 - Epoch: [182][ 490/ 1207] Overall Loss 0.208359 Objective Loss 0.208359 LR 0.000125 Time 0.020989 -2023-02-13 18:38:19,555 - Epoch: [182][ 500/ 1207] Overall Loss 0.208156 Objective Loss 0.208156 LR 0.000125 Time 0.020961 -2023-02-13 18:38:19,748 - Epoch: [182][ 510/ 1207] Overall Loss 0.208196 Objective Loss 0.208196 LR 0.000125 Time 0.020928 -2023-02-13 18:38:19,944 - Epoch: [182][ 520/ 1207] Overall Loss 0.208421 Objective Loss 0.208421 LR 0.000125 Time 0.020901 -2023-02-13 18:38:20,139 - Epoch: [182][ 530/ 1207] Overall Loss 0.208338 Objective Loss 0.208338 LR 0.000125 Time 0.020874 -2023-02-13 18:38:20,335 - Epoch: [182][ 540/ 1207] Overall Loss 0.208556 Objective Loss 0.208556 LR 0.000125 Time 0.020850 -2023-02-13 18:38:20,528 - Epoch: [182][ 550/ 1207] Overall Loss 0.208582 Objective Loss 0.208582 LR 0.000125 Time 0.020823 -2023-02-13 18:38:20,725 - Epoch: [182][ 560/ 1207] Overall Loss 0.208398 Objective Loss 0.208398 LR 0.000125 Time 0.020800 -2023-02-13 18:38:20,919 - Epoch: [182][ 570/ 1207] Overall Loss 0.207932 Objective Loss 0.207932 LR 0.000125 Time 0.020776 -2023-02-13 18:38:21,116 - Epoch: [182][ 580/ 1207] Overall Loss 0.208072 Objective Loss 0.208072 LR 0.000125 Time 0.020757 -2023-02-13 18:38:21,310 - Epoch: [182][ 590/ 1207] Overall Loss 0.208100 Objective Loss 0.208100 LR 0.000125 Time 0.020733 -2023-02-13 18:38:21,507 - Epoch: [182][ 600/ 1207] Overall Loss 0.207955 Objective Loss 0.207955 LR 0.000125 Time 0.020715 -2023-02-13 18:38:21,703 - Epoch: [182][ 610/ 1207] Overall Loss 0.208247 Objective Loss 0.208247 LR 0.000125 Time 0.020697 -2023-02-13 18:38:21,897 - Epoch: [182][ 620/ 1207] Overall Loss 0.208151 Objective Loss 0.208151 LR 0.000125 Time 0.020675 -2023-02-13 18:38:22,095 - Epoch: [182][ 630/ 1207] Overall Loss 0.208022 Objective Loss 0.208022 LR 0.000125 Time 0.020660 -2023-02-13 18:38:22,288 - Epoch: [182][ 640/ 1207] Overall Loss 0.208282 Objective Loss 0.208282 LR 0.000125 Time 0.020638 -2023-02-13 18:38:22,484 - Epoch: [182][ 650/ 1207] Overall Loss 0.208072 Objective Loss 0.208072 LR 0.000125 Time 0.020622 -2023-02-13 18:38:22,678 - Epoch: [182][ 660/ 1207] Overall Loss 0.208204 Objective Loss 0.208204 LR 0.000125 Time 0.020603 -2023-02-13 18:38:22,875 - Epoch: [182][ 670/ 1207] Overall Loss 0.207976 Objective Loss 0.207976 LR 0.000125 Time 0.020588 -2023-02-13 18:38:23,068 - Epoch: [182][ 680/ 1207] Overall Loss 0.207877 Objective Loss 0.207877 LR 0.000125 Time 0.020570 -2023-02-13 18:38:23,265 - Epoch: [182][ 690/ 1207] Overall Loss 0.207548 Objective Loss 0.207548 LR 0.000125 Time 0.020556 -2023-02-13 18:38:23,458 - Epoch: [182][ 700/ 1207] Overall Loss 0.207446 Objective Loss 0.207446 LR 0.000125 Time 0.020538 -2023-02-13 18:38:23,654 - Epoch: [182][ 710/ 1207] Overall Loss 0.207533 Objective Loss 0.207533 LR 0.000125 Time 0.020524 -2023-02-13 18:38:23,847 - Epoch: [182][ 720/ 1207] Overall Loss 0.206898 Objective Loss 0.206898 LR 0.000125 Time 0.020507 -2023-02-13 18:38:24,044 - Epoch: [182][ 730/ 1207] Overall Loss 0.206425 Objective Loss 0.206425 LR 0.000125 Time 0.020495 -2023-02-13 18:38:24,238 - Epoch: [182][ 740/ 1207] Overall Loss 0.206651 Objective Loss 0.206651 LR 0.000125 Time 0.020480 -2023-02-13 18:38:24,435 - Epoch: [182][ 750/ 1207] Overall Loss 0.206778 Objective Loss 0.206778 LR 0.000125 Time 0.020468 -2023-02-13 18:38:24,627 - Epoch: [182][ 760/ 1207] Overall Loss 0.206750 Objective Loss 0.206750 LR 0.000125 Time 0.020452 -2023-02-13 18:38:24,824 - Epoch: [182][ 770/ 1207] Overall Loss 0.206842 Objective Loss 0.206842 LR 0.000125 Time 0.020441 -2023-02-13 18:38:25,018 - Epoch: [182][ 780/ 1207] Overall Loss 0.206777 Objective Loss 0.206777 LR 0.000125 Time 0.020427 -2023-02-13 18:38:25,214 - Epoch: [182][ 790/ 1207] Overall Loss 0.206811 Objective Loss 0.206811 LR 0.000125 Time 0.020417 -2023-02-13 18:38:25,408 - Epoch: [182][ 800/ 1207] Overall Loss 0.206887 Objective Loss 0.206887 LR 0.000125 Time 0.020404 -2023-02-13 18:38:25,606 - Epoch: [182][ 810/ 1207] Overall Loss 0.206945 Objective Loss 0.206945 LR 0.000125 Time 0.020395 -2023-02-13 18:38:25,800 - Epoch: [182][ 820/ 1207] Overall Loss 0.206900 Objective Loss 0.206900 LR 0.000125 Time 0.020382 -2023-02-13 18:38:25,998 - Epoch: [182][ 830/ 1207] Overall Loss 0.206909 Objective Loss 0.206909 LR 0.000125 Time 0.020375 -2023-02-13 18:38:26,191 - Epoch: [182][ 840/ 1207] Overall Loss 0.206968 Objective Loss 0.206968 LR 0.000125 Time 0.020362 -2023-02-13 18:38:26,388 - Epoch: [182][ 850/ 1207] Overall Loss 0.206832 Objective Loss 0.206832 LR 0.000125 Time 0.020354 -2023-02-13 18:38:26,582 - Epoch: [182][ 860/ 1207] Overall Loss 0.206612 Objective Loss 0.206612 LR 0.000125 Time 0.020343 -2023-02-13 18:38:26,779 - Epoch: [182][ 870/ 1207] Overall Loss 0.206493 Objective Loss 0.206493 LR 0.000125 Time 0.020334 -2023-02-13 18:38:26,974 - Epoch: [182][ 880/ 1207] Overall Loss 0.206553 Objective Loss 0.206553 LR 0.000125 Time 0.020324 -2023-02-13 18:38:27,171 - Epoch: [182][ 890/ 1207] Overall Loss 0.206317 Objective Loss 0.206317 LR 0.000125 Time 0.020317 -2023-02-13 18:38:27,365 - Epoch: [182][ 900/ 1207] Overall Loss 0.206407 Objective Loss 0.206407 LR 0.000125 Time 0.020306 -2023-02-13 18:38:27,562 - Epoch: [182][ 910/ 1207] Overall Loss 0.206720 Objective Loss 0.206720 LR 0.000125 Time 0.020299 -2023-02-13 18:38:27,756 - Epoch: [182][ 920/ 1207] Overall Loss 0.206628 Objective Loss 0.206628 LR 0.000125 Time 0.020290 -2023-02-13 18:38:27,954 - Epoch: [182][ 930/ 1207] Overall Loss 0.206554 Objective Loss 0.206554 LR 0.000125 Time 0.020283 -2023-02-13 18:38:28,149 - Epoch: [182][ 940/ 1207] Overall Loss 0.206509 Objective Loss 0.206509 LR 0.000125 Time 0.020275 -2023-02-13 18:38:28,346 - Epoch: [182][ 950/ 1207] Overall Loss 0.206464 Objective Loss 0.206464 LR 0.000125 Time 0.020268 -2023-02-13 18:38:28,540 - Epoch: [182][ 960/ 1207] Overall Loss 0.206178 Objective Loss 0.206178 LR 0.000125 Time 0.020259 -2023-02-13 18:38:28,737 - Epoch: [182][ 970/ 1207] Overall Loss 0.206233 Objective Loss 0.206233 LR 0.000125 Time 0.020253 -2023-02-13 18:38:28,931 - Epoch: [182][ 980/ 1207] Overall Loss 0.206258 Objective Loss 0.206258 LR 0.000125 Time 0.020243 -2023-02-13 18:38:29,128 - Epoch: [182][ 990/ 1207] Overall Loss 0.206419 Objective Loss 0.206419 LR 0.000125 Time 0.020238 -2023-02-13 18:38:29,322 - Epoch: [182][ 1000/ 1207] Overall Loss 0.206531 Objective Loss 0.206531 LR 0.000125 Time 0.020229 -2023-02-13 18:38:29,519 - Epoch: [182][ 1010/ 1207] Overall Loss 0.206488 Objective Loss 0.206488 LR 0.000125 Time 0.020223 -2023-02-13 18:38:29,713 - Epoch: [182][ 1020/ 1207] Overall Loss 0.206480 Objective Loss 0.206480 LR 0.000125 Time 0.020215 -2023-02-13 18:38:29,910 - Epoch: [182][ 1030/ 1207] Overall Loss 0.206652 Objective Loss 0.206652 LR 0.000125 Time 0.020210 -2023-02-13 18:38:30,105 - Epoch: [182][ 1040/ 1207] Overall Loss 0.206547 Objective Loss 0.206547 LR 0.000125 Time 0.020202 -2023-02-13 18:38:30,302 - Epoch: [182][ 1050/ 1207] Overall Loss 0.206655 Objective Loss 0.206655 LR 0.000125 Time 0.020197 -2023-02-13 18:38:30,497 - Epoch: [182][ 1060/ 1207] Overall Loss 0.206590 Objective Loss 0.206590 LR 0.000125 Time 0.020190 -2023-02-13 18:38:30,694 - Epoch: [182][ 1070/ 1207] Overall Loss 0.206525 Objective Loss 0.206525 LR 0.000125 Time 0.020185 -2023-02-13 18:38:30,888 - Epoch: [182][ 1080/ 1207] Overall Loss 0.206545 Objective Loss 0.206545 LR 0.000125 Time 0.020178 -2023-02-13 18:38:31,086 - Epoch: [182][ 1090/ 1207] Overall Loss 0.206627 Objective Loss 0.206627 LR 0.000125 Time 0.020174 -2023-02-13 18:38:31,281 - Epoch: [182][ 1100/ 1207] Overall Loss 0.206671 Objective Loss 0.206671 LR 0.000125 Time 0.020167 -2023-02-13 18:38:31,478 - Epoch: [182][ 1110/ 1207] Overall Loss 0.206580 Objective Loss 0.206580 LR 0.000125 Time 0.020163 -2023-02-13 18:38:31,672 - Epoch: [182][ 1120/ 1207] Overall Loss 0.206637 Objective Loss 0.206637 LR 0.000125 Time 0.020156 -2023-02-13 18:38:31,870 - Epoch: [182][ 1130/ 1207] Overall Loss 0.206702 Objective Loss 0.206702 LR 0.000125 Time 0.020152 -2023-02-13 18:38:32,065 - Epoch: [182][ 1140/ 1207] Overall Loss 0.206836 Objective Loss 0.206836 LR 0.000125 Time 0.020146 -2023-02-13 18:38:32,262 - Epoch: [182][ 1150/ 1207] Overall Loss 0.206811 Objective Loss 0.206811 LR 0.000125 Time 0.020142 -2023-02-13 18:38:32,456 - Epoch: [182][ 1160/ 1207] Overall Loss 0.206815 Objective Loss 0.206815 LR 0.000125 Time 0.020135 -2023-02-13 18:38:32,653 - Epoch: [182][ 1170/ 1207] Overall Loss 0.206969 Objective Loss 0.206969 LR 0.000125 Time 0.020131 -2023-02-13 18:38:32,846 - Epoch: [182][ 1180/ 1207] Overall Loss 0.207119 Objective Loss 0.207119 LR 0.000125 Time 0.020124 -2023-02-13 18:38:33,042 - Epoch: [182][ 1190/ 1207] Overall Loss 0.207018 Objective Loss 0.207018 LR 0.000125 Time 0.020120 -2023-02-13 18:38:33,292 - Epoch: [182][ 1200/ 1207] Overall Loss 0.207093 Objective Loss 0.207093 LR 0.000125 Time 0.020160 -2023-02-13 18:38:33,407 - Epoch: [182][ 1207/ 1207] Overall Loss 0.207266 Objective Loss 0.207266 Top1 90.243902 Top5 98.780488 LR 0.000125 Time 0.020137 -2023-02-13 18:38:33,497 - --- validate (epoch=182)----------- -2023-02-13 18:38:33,497 - 34311 samples (256 per mini-batch) -2023-02-13 18:38:33,898 - Epoch: [182][ 10/ 135] Loss 0.287538 Top1 84.921875 Top5 98.281250 -2023-02-13 18:38:34,030 - Epoch: [182][ 20/ 135] Loss 0.288724 Top1 85.410156 Top5 98.242188 -2023-02-13 18:38:34,152 - Epoch: [182][ 30/ 135] Loss 0.280581 Top1 85.859375 Top5 98.242188 -2023-02-13 18:38:34,277 - Epoch: [182][ 40/ 135] Loss 0.282083 Top1 85.869141 Top5 98.017578 -2023-02-13 18:38:34,405 - Epoch: [182][ 50/ 135] Loss 0.279695 Top1 85.906250 Top5 98.046875 -2023-02-13 18:38:34,531 - Epoch: [182][ 60/ 135] Loss 0.277865 Top1 85.859375 Top5 98.079427 -2023-02-13 18:38:34,656 - Epoch: [182][ 70/ 135] Loss 0.285286 Top1 85.887277 Top5 98.018973 -2023-02-13 18:38:34,780 - Epoch: [182][ 80/ 135] Loss 0.287659 Top1 85.825195 Top5 98.012695 -2023-02-13 18:38:34,919 - Epoch: [182][ 90/ 135] Loss 0.292324 Top1 85.646701 Top5 97.951389 -2023-02-13 18:38:35,063 - Epoch: [182][ 100/ 135] Loss 0.289983 Top1 85.664062 Top5 97.976562 -2023-02-13 18:38:35,203 - Epoch: [182][ 110/ 135] Loss 0.292881 Top1 85.482955 Top5 97.915483 -2023-02-13 18:38:35,349 - Epoch: [182][ 120/ 135] Loss 0.292922 Top1 85.488281 Top5 97.900391 -2023-02-13 18:38:35,485 - Epoch: [182][ 130/ 135] Loss 0.291248 Top1 85.549880 Top5 97.872596 -2023-02-13 18:38:35,530 - Epoch: [182][ 135/ 135] Loss 0.295276 Top1 85.514849 Top5 97.860744 -2023-02-13 18:38:35,599 - ==> Top1: 85.515 Top5: 97.861 Loss: 0.295 - -2023-02-13 18:38:35,599 - ==> Confusion: -[[ 860 5 5 1 7 1 0 2 4 49 0 6 1 4 8 1 4 1 1 1 6] - [ 2 951 2 1 10 21 1 12 2 1 2 1 2 0 0 2 6 1 3 4 9] - [ 10 8 958 9 2 2 10 12 1 2 2 2 2 2 4 3 5 2 7 4 11] - [ 4 1 18 910 0 3 0 2 2 3 11 0 7 1 23 0 4 4 17 1 5] - [ 10 7 0 0 992 11 1 1 2 1 0 6 3 2 11 4 5 2 1 1 6] - [ 2 13 1 4 4 973 2 18 2 2 2 15 1 12 1 5 6 1 0 2 4] - [ 3 2 16 1 0 7 1034 8 0 1 1 0 2 2 0 2 2 2 1 7 8] - [ 2 10 12 1 1 34 0 929 0 1 1 9 4 0 0 0 0 0 8 5 7] - [ 8 4 0 1 1 1 1 0 915 36 7 3 0 11 17 1 1 0 2 0 0] - [ 69 0 3 1 8 1 0 2 34 862 0 0 0 16 5 2 1 1 2 0 5] - [ 2 1 2 6 1 3 2 6 15 2 986 2 1 7 3 0 2 1 4 0 5] - [ 3 2 2 0 0 12 1 7 2 1 0 917 21 6 2 7 2 8 2 10 0] - [ 1 0 0 5 1 2 0 0 3 0 0 30 884 2 1 7 3 10 2 2 6] - [ 4 2 1 2 5 10 1 1 9 16 7 3 2 933 10 5 2 1 1 3 6] - [ 6 1 1 10 3 3 0 1 19 7 2 0 3 2 1013 0 2 5 6 0 8] - [ 4 2 9 0 5 1 2 1 0 0 0 6 6 2 0 976 11 9 0 7 5] - [ 0 6 0 1 10 3 0 1 3 1 0 1 2 2 1 9 1004 0 2 3 12] - [ 6 2 0 3 0 2 2 0 0 0 2 7 16 1 0 13 0 986 0 3 8] - [ 5 6 4 8 1 1 1 24 2 0 3 1 3 0 11 1 1 2 1009 1 2] - [ 0 3 1 0 1 7 5 9 0 0 0 16 3 1 1 5 4 2 1 1080 9] - [ 141 223 216 101 123 202 66 158 80 77 157 120 315 245 185 77 269 92 161 257 10169]] - -2023-02-13 18:38:35,601 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:38:35,601 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:38:35,607 - - -2023-02-13 18:38:35,607 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:38:36,601 - Epoch: [183][ 10/ 1207] Overall Loss 0.208754 Objective Loss 0.208754 LR 0.000125 Time 0.099302 -2023-02-13 18:38:36,792 - Epoch: [183][ 20/ 1207] Overall Loss 0.208293 Objective Loss 0.208293 LR 0.000125 Time 0.059196 -2023-02-13 18:38:36,983 - Epoch: [183][ 30/ 1207] Overall Loss 0.209557 Objective Loss 0.209557 LR 0.000125 Time 0.045792 -2023-02-13 18:38:37,172 - Epoch: [183][ 40/ 1207] Overall Loss 0.209266 Objective Loss 0.209266 LR 0.000125 Time 0.039075 -2023-02-13 18:38:37,361 - Epoch: [183][ 50/ 1207] Overall Loss 0.207654 Objective Loss 0.207654 LR 0.000125 Time 0.035029 -2023-02-13 18:38:37,550 - Epoch: [183][ 60/ 1207] Overall Loss 0.205606 Objective Loss 0.205606 LR 0.000125 Time 0.032338 -2023-02-13 18:38:37,740 - Epoch: [183][ 70/ 1207] Overall Loss 0.208349 Objective Loss 0.208349 LR 0.000125 Time 0.030427 -2023-02-13 18:38:37,931 - Epoch: [183][ 80/ 1207] Overall Loss 0.204926 Objective Loss 0.204926 LR 0.000125 Time 0.029000 -2023-02-13 18:38:38,122 - Epoch: [183][ 90/ 1207] Overall Loss 0.203424 Objective Loss 0.203424 LR 0.000125 Time 0.027904 -2023-02-13 18:38:38,313 - Epoch: [183][ 100/ 1207] Overall Loss 0.203358 Objective Loss 0.203358 LR 0.000125 Time 0.027019 -2023-02-13 18:38:38,504 - Epoch: [183][ 110/ 1207] Overall Loss 0.202784 Objective Loss 0.202784 LR 0.000125 Time 0.026291 -2023-02-13 18:38:38,695 - Epoch: [183][ 120/ 1207] Overall Loss 0.204025 Objective Loss 0.204025 LR 0.000125 Time 0.025689 -2023-02-13 18:38:38,885 - Epoch: [183][ 130/ 1207] Overall Loss 0.204208 Objective Loss 0.204208 LR 0.000125 Time 0.025177 -2023-02-13 18:38:39,078 - Epoch: [183][ 140/ 1207] Overall Loss 0.204235 Objective Loss 0.204235 LR 0.000125 Time 0.024752 -2023-02-13 18:38:39,269 - Epoch: [183][ 150/ 1207] Overall Loss 0.203103 Objective Loss 0.203103 LR 0.000125 Time 0.024373 -2023-02-13 18:38:39,461 - Epoch: [183][ 160/ 1207] Overall Loss 0.203153 Objective Loss 0.203153 LR 0.000125 Time 0.024049 -2023-02-13 18:38:39,653 - Epoch: [183][ 170/ 1207] Overall Loss 0.203126 Objective Loss 0.203126 LR 0.000125 Time 0.023756 -2023-02-13 18:38:39,844 - Epoch: [183][ 180/ 1207] Overall Loss 0.203031 Objective Loss 0.203031 LR 0.000125 Time 0.023498 -2023-02-13 18:38:40,036 - Epoch: [183][ 190/ 1207] Overall Loss 0.203480 Objective Loss 0.203480 LR 0.000125 Time 0.023267 -2023-02-13 18:38:40,227 - Epoch: [183][ 200/ 1207] Overall Loss 0.203754 Objective Loss 0.203754 LR 0.000125 Time 0.023061 -2023-02-13 18:38:40,418 - Epoch: [183][ 210/ 1207] Overall Loss 0.203543 Objective Loss 0.203543 LR 0.000125 Time 0.022869 -2023-02-13 18:38:40,609 - Epoch: [183][ 220/ 1207] Overall Loss 0.204065 Objective Loss 0.204065 LR 0.000125 Time 0.022697 -2023-02-13 18:38:40,801 - Epoch: [183][ 230/ 1207] Overall Loss 0.204351 Objective Loss 0.204351 LR 0.000125 Time 0.022541 -2023-02-13 18:38:40,993 - Epoch: [183][ 240/ 1207] Overall Loss 0.205890 Objective Loss 0.205890 LR 0.000125 Time 0.022401 -2023-02-13 18:38:41,183 - Epoch: [183][ 250/ 1207] Overall Loss 0.205890 Objective Loss 0.205890 LR 0.000125 Time 0.022266 -2023-02-13 18:38:41,374 - Epoch: [183][ 260/ 1207] Overall Loss 0.206250 Objective Loss 0.206250 LR 0.000125 Time 0.022141 -2023-02-13 18:38:41,566 - Epoch: [183][ 270/ 1207] Overall Loss 0.205815 Objective Loss 0.205815 LR 0.000125 Time 0.022031 -2023-02-13 18:38:41,757 - Epoch: [183][ 280/ 1207] Overall Loss 0.205522 Objective Loss 0.205522 LR 0.000125 Time 0.021925 -2023-02-13 18:38:41,948 - Epoch: [183][ 290/ 1207] Overall Loss 0.204929 Objective Loss 0.204929 LR 0.000125 Time 0.021827 -2023-02-13 18:38:42,141 - Epoch: [183][ 300/ 1207] Overall Loss 0.204855 Objective Loss 0.204855 LR 0.000125 Time 0.021740 -2023-02-13 18:38:42,331 - Epoch: [183][ 310/ 1207] Overall Loss 0.205701 Objective Loss 0.205701 LR 0.000125 Time 0.021652 -2023-02-13 18:38:42,524 - Epoch: [183][ 320/ 1207] Overall Loss 0.205675 Objective Loss 0.205675 LR 0.000125 Time 0.021577 -2023-02-13 18:38:42,715 - Epoch: [183][ 330/ 1207] Overall Loss 0.205913 Objective Loss 0.205913 LR 0.000125 Time 0.021502 -2023-02-13 18:38:42,908 - Epoch: [183][ 340/ 1207] Overall Loss 0.206220 Objective Loss 0.206220 LR 0.000125 Time 0.021433 -2023-02-13 18:38:43,099 - Epoch: [183][ 350/ 1207] Overall Loss 0.206309 Objective Loss 0.206309 LR 0.000125 Time 0.021368 -2023-02-13 18:38:43,291 - Epoch: [183][ 360/ 1207] Overall Loss 0.206817 Objective Loss 0.206817 LR 0.000125 Time 0.021306 -2023-02-13 18:38:43,483 - Epoch: [183][ 370/ 1207] Overall Loss 0.206442 Objective Loss 0.206442 LR 0.000125 Time 0.021247 -2023-02-13 18:38:43,673 - Epoch: [183][ 380/ 1207] Overall Loss 0.206290 Objective Loss 0.206290 LR 0.000125 Time 0.021188 -2023-02-13 18:38:43,864 - Epoch: [183][ 390/ 1207] Overall Loss 0.206474 Objective Loss 0.206474 LR 0.000125 Time 0.021133 -2023-02-13 18:38:44,057 - Epoch: [183][ 400/ 1207] Overall Loss 0.205919 Objective Loss 0.205919 LR 0.000125 Time 0.021086 -2023-02-13 18:38:44,248 - Epoch: [183][ 410/ 1207] Overall Loss 0.206562 Objective Loss 0.206562 LR 0.000125 Time 0.021037 -2023-02-13 18:38:44,439 - Epoch: [183][ 420/ 1207] Overall Loss 0.206567 Objective Loss 0.206567 LR 0.000125 Time 0.020990 -2023-02-13 18:38:44,630 - Epoch: [183][ 430/ 1207] Overall Loss 0.206231 Objective Loss 0.206231 LR 0.000125 Time 0.020945 -2023-02-13 18:38:44,821 - Epoch: [183][ 440/ 1207] Overall Loss 0.206120 Objective Loss 0.206120 LR 0.000125 Time 0.020903 -2023-02-13 18:38:45,012 - Epoch: [183][ 450/ 1207] Overall Loss 0.206462 Objective Loss 0.206462 LR 0.000125 Time 0.020862 -2023-02-13 18:38:45,202 - Epoch: [183][ 460/ 1207] Overall Loss 0.206510 Objective Loss 0.206510 LR 0.000125 Time 0.020821 -2023-02-13 18:38:45,392 - Epoch: [183][ 470/ 1207] Overall Loss 0.206322 Objective Loss 0.206322 LR 0.000125 Time 0.020781 -2023-02-13 18:38:45,583 - Epoch: [183][ 480/ 1207] Overall Loss 0.206386 Objective Loss 0.206386 LR 0.000125 Time 0.020745 -2023-02-13 18:38:45,773 - Epoch: [183][ 490/ 1207] Overall Loss 0.206294 Objective Loss 0.206294 LR 0.000125 Time 0.020708 -2023-02-13 18:38:45,963 - Epoch: [183][ 500/ 1207] Overall Loss 0.206538 Objective Loss 0.206538 LR 0.000125 Time 0.020675 -2023-02-13 18:38:46,155 - Epoch: [183][ 510/ 1207] Overall Loss 0.206671 Objective Loss 0.206671 LR 0.000125 Time 0.020643 -2023-02-13 18:38:46,345 - Epoch: [183][ 520/ 1207] Overall Loss 0.205818 Objective Loss 0.205818 LR 0.000125 Time 0.020611 -2023-02-13 18:38:46,535 - Epoch: [183][ 530/ 1207] Overall Loss 0.206315 Objective Loss 0.206315 LR 0.000125 Time 0.020580 -2023-02-13 18:38:46,725 - Epoch: [183][ 540/ 1207] Overall Loss 0.205955 Objective Loss 0.205955 LR 0.000125 Time 0.020550 -2023-02-13 18:38:46,915 - Epoch: [183][ 550/ 1207] Overall Loss 0.205845 Objective Loss 0.205845 LR 0.000125 Time 0.020522 -2023-02-13 18:38:47,106 - Epoch: [183][ 560/ 1207] Overall Loss 0.205746 Objective Loss 0.205746 LR 0.000125 Time 0.020495 -2023-02-13 18:38:47,296 - Epoch: [183][ 570/ 1207] Overall Loss 0.205478 Objective Loss 0.205478 LR 0.000125 Time 0.020469 -2023-02-13 18:38:47,487 - Epoch: [183][ 580/ 1207] Overall Loss 0.205401 Objective Loss 0.205401 LR 0.000125 Time 0.020444 -2023-02-13 18:38:47,677 - Epoch: [183][ 590/ 1207] Overall Loss 0.205431 Objective Loss 0.205431 LR 0.000125 Time 0.020419 -2023-02-13 18:38:47,866 - Epoch: [183][ 600/ 1207] Overall Loss 0.205751 Objective Loss 0.205751 LR 0.000125 Time 0.020394 -2023-02-13 18:38:48,056 - Epoch: [183][ 610/ 1207] Overall Loss 0.205601 Objective Loss 0.205601 LR 0.000125 Time 0.020371 -2023-02-13 18:38:48,247 - Epoch: [183][ 620/ 1207] Overall Loss 0.205406 Objective Loss 0.205406 LR 0.000125 Time 0.020349 -2023-02-13 18:38:48,440 - Epoch: [183][ 630/ 1207] Overall Loss 0.205495 Objective Loss 0.205495 LR 0.000125 Time 0.020332 -2023-02-13 18:38:48,631 - Epoch: [183][ 640/ 1207] Overall Loss 0.205623 Objective Loss 0.205623 LR 0.000125 Time 0.020311 -2023-02-13 18:38:48,821 - Epoch: [183][ 650/ 1207] Overall Loss 0.205775 Objective Loss 0.205775 LR 0.000125 Time 0.020291 -2023-02-13 18:38:49,011 - Epoch: [183][ 660/ 1207] Overall Loss 0.205871 Objective Loss 0.205871 LR 0.000125 Time 0.020271 -2023-02-13 18:38:49,202 - Epoch: [183][ 670/ 1207] Overall Loss 0.206019 Objective Loss 0.206019 LR 0.000125 Time 0.020252 -2023-02-13 18:38:49,391 - Epoch: [183][ 680/ 1207] Overall Loss 0.206057 Objective Loss 0.206057 LR 0.000125 Time 0.020233 -2023-02-13 18:38:49,582 - Epoch: [183][ 690/ 1207] Overall Loss 0.206150 Objective Loss 0.206150 LR 0.000125 Time 0.020215 -2023-02-13 18:38:49,772 - Epoch: [183][ 700/ 1207] Overall Loss 0.206126 Objective Loss 0.206126 LR 0.000125 Time 0.020198 -2023-02-13 18:38:49,962 - Epoch: [183][ 710/ 1207] Overall Loss 0.206074 Objective Loss 0.206074 LR 0.000125 Time 0.020181 -2023-02-13 18:38:50,153 - Epoch: [183][ 720/ 1207] Overall Loss 0.206152 Objective Loss 0.206152 LR 0.000125 Time 0.020164 -2023-02-13 18:38:50,343 - Epoch: [183][ 730/ 1207] Overall Loss 0.206311 Objective Loss 0.206311 LR 0.000125 Time 0.020149 -2023-02-13 18:38:50,533 - Epoch: [183][ 740/ 1207] Overall Loss 0.206237 Objective Loss 0.206237 LR 0.000125 Time 0.020132 -2023-02-13 18:38:50,723 - Epoch: [183][ 750/ 1207] Overall Loss 0.206472 Objective Loss 0.206472 LR 0.000125 Time 0.020117 -2023-02-13 18:38:50,913 - Epoch: [183][ 760/ 1207] Overall Loss 0.206318 Objective Loss 0.206318 LR 0.000125 Time 0.020102 -2023-02-13 18:38:51,105 - Epoch: [183][ 770/ 1207] Overall Loss 0.206066 Objective Loss 0.206066 LR 0.000125 Time 0.020089 -2023-02-13 18:38:51,295 - Epoch: [183][ 780/ 1207] Overall Loss 0.205883 Objective Loss 0.205883 LR 0.000125 Time 0.020074 -2023-02-13 18:38:51,487 - Epoch: [183][ 790/ 1207] Overall Loss 0.205999 Objective Loss 0.205999 LR 0.000125 Time 0.020063 -2023-02-13 18:38:51,677 - Epoch: [183][ 800/ 1207] Overall Loss 0.206224 Objective Loss 0.206224 LR 0.000125 Time 0.020050 -2023-02-13 18:38:51,867 - Epoch: [183][ 810/ 1207] Overall Loss 0.206226 Objective Loss 0.206226 LR 0.000125 Time 0.020037 -2023-02-13 18:38:52,058 - Epoch: [183][ 820/ 1207] Overall Loss 0.206463 Objective Loss 0.206463 LR 0.000125 Time 0.020025 -2023-02-13 18:38:52,250 - Epoch: [183][ 830/ 1207] Overall Loss 0.206593 Objective Loss 0.206593 LR 0.000125 Time 0.020014 -2023-02-13 18:38:52,440 - Epoch: [183][ 840/ 1207] Overall Loss 0.206689 Objective Loss 0.206689 LR 0.000125 Time 0.020002 -2023-02-13 18:38:52,631 - Epoch: [183][ 850/ 1207] Overall Loss 0.206886 Objective Loss 0.206886 LR 0.000125 Time 0.019990 -2023-02-13 18:38:52,823 - Epoch: [183][ 860/ 1207] Overall Loss 0.206815 Objective Loss 0.206815 LR 0.000125 Time 0.019980 -2023-02-13 18:38:53,014 - Epoch: [183][ 870/ 1207] Overall Loss 0.206701 Objective Loss 0.206701 LR 0.000125 Time 0.019970 -2023-02-13 18:38:53,207 - Epoch: [183][ 880/ 1207] Overall Loss 0.206668 Objective Loss 0.206668 LR 0.000125 Time 0.019962 -2023-02-13 18:38:53,398 - Epoch: [183][ 890/ 1207] Overall Loss 0.206498 Objective Loss 0.206498 LR 0.000125 Time 0.019952 -2023-02-13 18:38:53,590 - Epoch: [183][ 900/ 1207] Overall Loss 0.206418 Objective Loss 0.206418 LR 0.000125 Time 0.019943 -2023-02-13 18:38:53,781 - Epoch: [183][ 910/ 1207] Overall Loss 0.206369 Objective Loss 0.206369 LR 0.000125 Time 0.019934 -2023-02-13 18:38:53,974 - Epoch: [183][ 920/ 1207] Overall Loss 0.206430 Objective Loss 0.206430 LR 0.000125 Time 0.019926 -2023-02-13 18:38:54,166 - Epoch: [183][ 930/ 1207] Overall Loss 0.206500 Objective Loss 0.206500 LR 0.000125 Time 0.019918 -2023-02-13 18:38:54,356 - Epoch: [183][ 940/ 1207] Overall Loss 0.206222 Objective Loss 0.206222 LR 0.000125 Time 0.019908 -2023-02-13 18:38:54,547 - Epoch: [183][ 950/ 1207] Overall Loss 0.206266 Objective Loss 0.206266 LR 0.000125 Time 0.019899 -2023-02-13 18:38:54,737 - Epoch: [183][ 960/ 1207] Overall Loss 0.206437 Objective Loss 0.206437 LR 0.000125 Time 0.019889 -2023-02-13 18:38:54,928 - Epoch: [183][ 970/ 1207] Overall Loss 0.206429 Objective Loss 0.206429 LR 0.000125 Time 0.019881 -2023-02-13 18:38:55,119 - Epoch: [183][ 980/ 1207] Overall Loss 0.206291 Objective Loss 0.206291 LR 0.000125 Time 0.019872 -2023-02-13 18:38:55,310 - Epoch: [183][ 990/ 1207] Overall Loss 0.206334 Objective Loss 0.206334 LR 0.000125 Time 0.019864 -2023-02-13 18:38:55,499 - Epoch: [183][ 1000/ 1207] Overall Loss 0.206424 Objective Loss 0.206424 LR 0.000125 Time 0.019855 -2023-02-13 18:38:55,690 - Epoch: [183][ 1010/ 1207] Overall Loss 0.206433 Objective Loss 0.206433 LR 0.000125 Time 0.019846 -2023-02-13 18:38:55,880 - Epoch: [183][ 1020/ 1207] Overall Loss 0.206478 Objective Loss 0.206478 LR 0.000125 Time 0.019838 -2023-02-13 18:38:56,073 - Epoch: [183][ 1030/ 1207] Overall Loss 0.206705 Objective Loss 0.206705 LR 0.000125 Time 0.019832 -2023-02-13 18:38:56,265 - Epoch: [183][ 1040/ 1207] Overall Loss 0.206608 Objective Loss 0.206608 LR 0.000125 Time 0.019826 -2023-02-13 18:38:56,457 - Epoch: [183][ 1050/ 1207] Overall Loss 0.206514 Objective Loss 0.206514 LR 0.000125 Time 0.019819 -2023-02-13 18:38:56,649 - Epoch: [183][ 1060/ 1207] Overall Loss 0.206578 Objective Loss 0.206578 LR 0.000125 Time 0.019813 -2023-02-13 18:38:56,841 - Epoch: [183][ 1070/ 1207] Overall Loss 0.206676 Objective Loss 0.206676 LR 0.000125 Time 0.019807 -2023-02-13 18:38:57,034 - Epoch: [183][ 1080/ 1207] Overall Loss 0.206678 Objective Loss 0.206678 LR 0.000125 Time 0.019802 -2023-02-13 18:38:57,227 - Epoch: [183][ 1090/ 1207] Overall Loss 0.206660 Objective Loss 0.206660 LR 0.000125 Time 0.019797 -2023-02-13 18:38:57,418 - Epoch: [183][ 1100/ 1207] Overall Loss 0.206704 Objective Loss 0.206704 LR 0.000125 Time 0.019791 -2023-02-13 18:38:57,611 - Epoch: [183][ 1110/ 1207] Overall Loss 0.206811 Objective Loss 0.206811 LR 0.000125 Time 0.019785 -2023-02-13 18:38:57,803 - Epoch: [183][ 1120/ 1207] Overall Loss 0.206797 Objective Loss 0.206797 LR 0.000125 Time 0.019780 -2023-02-13 18:38:57,996 - Epoch: [183][ 1130/ 1207] Overall Loss 0.206900 Objective Loss 0.206900 LR 0.000125 Time 0.019775 -2023-02-13 18:38:58,188 - Epoch: [183][ 1140/ 1207] Overall Loss 0.206974 Objective Loss 0.206974 LR 0.000125 Time 0.019770 -2023-02-13 18:38:58,380 - Epoch: [183][ 1150/ 1207] Overall Loss 0.206840 Objective Loss 0.206840 LR 0.000125 Time 0.019765 -2023-02-13 18:38:58,572 - Epoch: [183][ 1160/ 1207] Overall Loss 0.206848 Objective Loss 0.206848 LR 0.000125 Time 0.019760 -2023-02-13 18:38:58,764 - Epoch: [183][ 1170/ 1207] Overall Loss 0.206666 Objective Loss 0.206666 LR 0.000125 Time 0.019754 -2023-02-13 18:38:58,956 - Epoch: [183][ 1180/ 1207] Overall Loss 0.206643 Objective Loss 0.206643 LR 0.000125 Time 0.019750 -2023-02-13 18:38:59,149 - Epoch: [183][ 1190/ 1207] Overall Loss 0.206896 Objective Loss 0.206896 LR 0.000125 Time 0.019745 -2023-02-13 18:38:59,390 - Epoch: [183][ 1200/ 1207] Overall Loss 0.206826 Objective Loss 0.206826 LR 0.000125 Time 0.019781 -2023-02-13 18:38:59,506 - Epoch: [183][ 1207/ 1207] Overall Loss 0.206853 Objective Loss 0.206853 Top1 88.719512 Top5 99.390244 LR 0.000125 Time 0.019763 -2023-02-13 18:38:59,578 - --- validate (epoch=183)----------- -2023-02-13 18:38:59,578 - 34311 samples (256 per mini-batch) -2023-02-13 18:38:59,971 - Epoch: [183][ 10/ 135] Loss 0.270011 Top1 86.484375 Top5 98.125000 -2023-02-13 18:39:00,111 - Epoch: [183][ 20/ 135] Loss 0.285078 Top1 85.820312 Top5 97.812500 -2023-02-13 18:39:00,251 - Epoch: [183][ 30/ 135] Loss 0.289403 Top1 85.546875 Top5 97.721354 -2023-02-13 18:39:00,397 - Epoch: [183][ 40/ 135] Loss 0.295073 Top1 85.341797 Top5 97.802734 -2023-02-13 18:39:00,537 - Epoch: [183][ 50/ 135] Loss 0.294677 Top1 85.445312 Top5 97.828125 -2023-02-13 18:39:00,684 - Epoch: [183][ 60/ 135] Loss 0.295964 Top1 85.488281 Top5 97.890625 -2023-02-13 18:39:00,821 - Epoch: [183][ 70/ 135] Loss 0.295945 Top1 85.468750 Top5 97.840402 -2023-02-13 18:39:00,962 - Epoch: [183][ 80/ 135] Loss 0.296612 Top1 85.410156 Top5 97.846680 -2023-02-13 18:39:01,089 - Epoch: [183][ 90/ 135] Loss 0.298398 Top1 85.451389 Top5 97.886285 -2023-02-13 18:39:01,222 - Epoch: [183][ 100/ 135] Loss 0.294966 Top1 85.507812 Top5 97.832031 -2023-02-13 18:39:01,354 - Epoch: [183][ 110/ 135] Loss 0.295086 Top1 85.507812 Top5 97.869318 -2023-02-13 18:39:01,485 - Epoch: [183][ 120/ 135] Loss 0.295186 Top1 85.507812 Top5 97.851562 -2023-02-13 18:39:01,620 - Epoch: [183][ 130/ 135] Loss 0.293216 Top1 85.483774 Top5 97.854567 -2023-02-13 18:39:01,664 - Epoch: [183][ 135/ 135] Loss 0.292452 Top1 85.433243 Top5 97.822856 -2023-02-13 18:39:01,735 - ==> Top1: 85.433 Top5: 97.823 Loss: 0.292 - -2023-02-13 18:39:01,736 - ==> Confusion: -[[ 862 5 7 2 8 2 0 3 7 41 0 4 1 3 5 2 2 1 0 1 11] - [ 3 966 3 1 8 16 1 9 2 1 1 1 2 0 0 1 4 0 2 3 9] - [ 4 4 962 12 3 0 13 12 1 2 5 2 1 6 4 4 4 1 4 5 9] - [ 3 0 18 912 1 6 1 2 1 2 17 0 6 0 14 2 4 4 15 1 7] - [ 9 8 0 0 998 8 1 2 3 0 0 4 4 3 5 6 6 1 1 3 4] - [ 1 21 0 4 4 969 4 19 0 2 2 11 2 9 1 4 6 1 1 2 7] - [ 1 3 18 1 0 5 1038 4 1 1 1 1 2 3 0 3 0 1 2 8 6] - [ 2 7 9 1 2 25 3 925 1 1 0 6 6 1 0 0 1 1 20 6 7] - [ 11 4 1 1 1 0 1 0 925 28 6 3 0 9 11 2 1 0 5 0 0] - [ 67 1 2 0 10 1 0 2 36 867 1 1 0 13 4 1 1 1 1 0 3] - [ 1 1 5 4 1 3 1 6 16 1 993 2 1 5 2 0 1 1 3 0 4] - [ 3 3 2 0 1 8 2 7 0 0 0 913 28 6 1 6 4 10 3 7 1] - [ 0 1 0 8 0 1 0 1 2 1 0 22 895 0 1 9 2 11 1 0 4] - [ 7 4 1 1 6 9 1 1 12 16 9 3 2 932 5 4 3 1 0 1 6] - [ 5 2 1 16 4 4 0 1 27 9 2 0 5 3 993 0 0 4 7 0 9] - [ 5 0 7 0 5 1 3 2 1 0 0 4 7 2 0 973 11 10 1 7 7] - [ 1 6 1 2 8 1 0 0 1 0 0 1 2 2 2 12 1003 2 2 3 12] - [ 4 1 0 6 0 3 1 0 0 0 1 6 14 1 0 17 0 988 0 2 7] - [ 4 8 4 7 0 2 0 19 3 0 3 0 2 0 10 1 0 3 1018 1 1] - [ 1 3 0 1 1 4 7 9 0 0 0 13 2 1 1 8 4 3 1 1083 6] - [ 130 271 210 100 138 183 73 165 103 79 225 100 325 244 145 95 243 94 167 246 10098]] - -2023-02-13 18:39:01,738 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:39:01,738 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:39:01,743 - - -2023-02-13 18:39:01,744 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:39:02,641 - Epoch: [184][ 10/ 1207] Overall Loss 0.207448 Objective Loss 0.207448 LR 0.000125 Time 0.089629 -2023-02-13 18:39:02,837 - Epoch: [184][ 20/ 1207] Overall Loss 0.213792 Objective Loss 0.213792 LR 0.000125 Time 0.054598 -2023-02-13 18:39:03,026 - Epoch: [184][ 30/ 1207] Overall Loss 0.211927 Objective Loss 0.211927 LR 0.000125 Time 0.042713 -2023-02-13 18:39:03,216 - Epoch: [184][ 40/ 1207] Overall Loss 0.210702 Objective Loss 0.210702 LR 0.000125 Time 0.036774 -2023-02-13 18:39:03,406 - Epoch: [184][ 50/ 1207] Overall Loss 0.211102 Objective Loss 0.211102 LR 0.000125 Time 0.033197 -2023-02-13 18:39:03,595 - Epoch: [184][ 60/ 1207] Overall Loss 0.205758 Objective Loss 0.205758 LR 0.000125 Time 0.030814 -2023-02-13 18:39:03,783 - Epoch: [184][ 70/ 1207] Overall Loss 0.204833 Objective Loss 0.204833 LR 0.000125 Time 0.029098 -2023-02-13 18:39:03,973 - Epoch: [184][ 80/ 1207] Overall Loss 0.204191 Objective Loss 0.204191 LR 0.000125 Time 0.027826 -2023-02-13 18:39:04,162 - Epoch: [184][ 90/ 1207] Overall Loss 0.202449 Objective Loss 0.202449 LR 0.000125 Time 0.026835 -2023-02-13 18:39:04,353 - Epoch: [184][ 100/ 1207] Overall Loss 0.200580 Objective Loss 0.200580 LR 0.000125 Time 0.026049 -2023-02-13 18:39:04,542 - Epoch: [184][ 110/ 1207] Overall Loss 0.199551 Objective Loss 0.199551 LR 0.000125 Time 0.025398 -2023-02-13 18:39:04,731 - Epoch: [184][ 120/ 1207] Overall Loss 0.200093 Objective Loss 0.200093 LR 0.000125 Time 0.024855 -2023-02-13 18:39:04,920 - Epoch: [184][ 130/ 1207] Overall Loss 0.200562 Objective Loss 0.200562 LR 0.000125 Time 0.024396 -2023-02-13 18:39:05,110 - Epoch: [184][ 140/ 1207] Overall Loss 0.200577 Objective Loss 0.200577 LR 0.000125 Time 0.024003 -2023-02-13 18:39:05,300 - Epoch: [184][ 150/ 1207] Overall Loss 0.201084 Objective Loss 0.201084 LR 0.000125 Time 0.023670 -2023-02-13 18:39:05,490 - Epoch: [184][ 160/ 1207] Overall Loss 0.200846 Objective Loss 0.200846 LR 0.000125 Time 0.023375 -2023-02-13 18:39:05,679 - Epoch: [184][ 170/ 1207] Overall Loss 0.203050 Objective Loss 0.203050 LR 0.000125 Time 0.023110 -2023-02-13 18:39:05,869 - Epoch: [184][ 180/ 1207] Overall Loss 0.202802 Objective Loss 0.202802 LR 0.000125 Time 0.022879 -2023-02-13 18:39:06,059 - Epoch: [184][ 190/ 1207] Overall Loss 0.203092 Objective Loss 0.203092 LR 0.000125 Time 0.022676 -2023-02-13 18:39:06,250 - Epoch: [184][ 200/ 1207] Overall Loss 0.203157 Objective Loss 0.203157 LR 0.000125 Time 0.022491 -2023-02-13 18:39:06,439 - Epoch: [184][ 210/ 1207] Overall Loss 0.203549 Objective Loss 0.203549 LR 0.000125 Time 0.022320 -2023-02-13 18:39:06,629 - Epoch: [184][ 220/ 1207] Overall Loss 0.203385 Objective Loss 0.203385 LR 0.000125 Time 0.022168 -2023-02-13 18:39:06,820 - Epoch: [184][ 230/ 1207] Overall Loss 0.203636 Objective Loss 0.203636 LR 0.000125 Time 0.022031 -2023-02-13 18:39:07,011 - Epoch: [184][ 240/ 1207] Overall Loss 0.204206 Objective Loss 0.204206 LR 0.000125 Time 0.021910 -2023-02-13 18:39:07,202 - Epoch: [184][ 250/ 1207] Overall Loss 0.204984 Objective Loss 0.204984 LR 0.000125 Time 0.021795 -2023-02-13 18:39:07,392 - Epoch: [184][ 260/ 1207] Overall Loss 0.204904 Objective Loss 0.204904 LR 0.000125 Time 0.021686 -2023-02-13 18:39:07,582 - Epoch: [184][ 270/ 1207] Overall Loss 0.204820 Objective Loss 0.204820 LR 0.000125 Time 0.021586 -2023-02-13 18:39:07,772 - Epoch: [184][ 280/ 1207] Overall Loss 0.204695 Objective Loss 0.204695 LR 0.000125 Time 0.021491 -2023-02-13 18:39:07,962 - Epoch: [184][ 290/ 1207] Overall Loss 0.205232 Objective Loss 0.205232 LR 0.000125 Time 0.021404 -2023-02-13 18:39:08,152 - Epoch: [184][ 300/ 1207] Overall Loss 0.206015 Objective Loss 0.206015 LR 0.000125 Time 0.021324 -2023-02-13 18:39:08,343 - Epoch: [184][ 310/ 1207] Overall Loss 0.205321 Objective Loss 0.205321 LR 0.000125 Time 0.021249 -2023-02-13 18:39:08,533 - Epoch: [184][ 320/ 1207] Overall Loss 0.204683 Objective Loss 0.204683 LR 0.000125 Time 0.021177 -2023-02-13 18:39:08,723 - Epoch: [184][ 330/ 1207] Overall Loss 0.204478 Objective Loss 0.204478 LR 0.000125 Time 0.021110 -2023-02-13 18:39:08,913 - Epoch: [184][ 340/ 1207] Overall Loss 0.204088 Objective Loss 0.204088 LR 0.000125 Time 0.021047 -2023-02-13 18:39:09,103 - Epoch: [184][ 350/ 1207] Overall Loss 0.204076 Objective Loss 0.204076 LR 0.000125 Time 0.020988 -2023-02-13 18:39:09,293 - Epoch: [184][ 360/ 1207] Overall Loss 0.203938 Objective Loss 0.203938 LR 0.000125 Time 0.020933 -2023-02-13 18:39:09,484 - Epoch: [184][ 370/ 1207] Overall Loss 0.203975 Objective Loss 0.203975 LR 0.000125 Time 0.020881 -2023-02-13 18:39:09,674 - Epoch: [184][ 380/ 1207] Overall Loss 0.204384 Objective Loss 0.204384 LR 0.000125 Time 0.020832 -2023-02-13 18:39:09,865 - Epoch: [184][ 390/ 1207] Overall Loss 0.204360 Objective Loss 0.204360 LR 0.000125 Time 0.020785 -2023-02-13 18:39:10,055 - Epoch: [184][ 400/ 1207] Overall Loss 0.204255 Objective Loss 0.204255 LR 0.000125 Time 0.020741 -2023-02-13 18:39:10,246 - Epoch: [184][ 410/ 1207] Overall Loss 0.204157 Objective Loss 0.204157 LR 0.000125 Time 0.020699 -2023-02-13 18:39:10,436 - Epoch: [184][ 420/ 1207] Overall Loss 0.204564 Objective Loss 0.204564 LR 0.000125 Time 0.020658 -2023-02-13 18:39:10,626 - Epoch: [184][ 430/ 1207] Overall Loss 0.204728 Objective Loss 0.204728 LR 0.000125 Time 0.020619 -2023-02-13 18:39:10,817 - Epoch: [184][ 440/ 1207] Overall Loss 0.204939 Objective Loss 0.204939 LR 0.000125 Time 0.020583 -2023-02-13 18:39:11,008 - Epoch: [184][ 450/ 1207] Overall Loss 0.204921 Objective Loss 0.204921 LR 0.000125 Time 0.020550 -2023-02-13 18:39:11,199 - Epoch: [184][ 460/ 1207] Overall Loss 0.204965 Objective Loss 0.204965 LR 0.000125 Time 0.020517 -2023-02-13 18:39:11,390 - Epoch: [184][ 470/ 1207] Overall Loss 0.205083 Objective Loss 0.205083 LR 0.000125 Time 0.020485 -2023-02-13 18:39:11,580 - Epoch: [184][ 480/ 1207] Overall Loss 0.205439 Objective Loss 0.205439 LR 0.000125 Time 0.020455 -2023-02-13 18:39:11,771 - Epoch: [184][ 490/ 1207] Overall Loss 0.205159 Objective Loss 0.205159 LR 0.000125 Time 0.020426 -2023-02-13 18:39:11,962 - Epoch: [184][ 500/ 1207] Overall Loss 0.205218 Objective Loss 0.205218 LR 0.000125 Time 0.020398 -2023-02-13 18:39:12,153 - Epoch: [184][ 510/ 1207] Overall Loss 0.205550 Objective Loss 0.205550 LR 0.000125 Time 0.020372 -2023-02-13 18:39:12,344 - Epoch: [184][ 520/ 1207] Overall Loss 0.205496 Objective Loss 0.205496 LR 0.000125 Time 0.020346 -2023-02-13 18:39:12,535 - Epoch: [184][ 530/ 1207] Overall Loss 0.206016 Objective Loss 0.206016 LR 0.000125 Time 0.020322 -2023-02-13 18:39:12,725 - Epoch: [184][ 540/ 1207] Overall Loss 0.206174 Objective Loss 0.206174 LR 0.000125 Time 0.020297 -2023-02-13 18:39:12,915 - Epoch: [184][ 550/ 1207] Overall Loss 0.206204 Objective Loss 0.206204 LR 0.000125 Time 0.020273 -2023-02-13 18:39:13,105 - Epoch: [184][ 560/ 1207] Overall Loss 0.206341 Objective Loss 0.206341 LR 0.000125 Time 0.020251 -2023-02-13 18:39:13,297 - Epoch: [184][ 570/ 1207] Overall Loss 0.206364 Objective Loss 0.206364 LR 0.000125 Time 0.020230 -2023-02-13 18:39:13,487 - Epoch: [184][ 580/ 1207] Overall Loss 0.206165 Objective Loss 0.206165 LR 0.000125 Time 0.020209 -2023-02-13 18:39:13,678 - Epoch: [184][ 590/ 1207] Overall Loss 0.206284 Objective Loss 0.206284 LR 0.000125 Time 0.020189 -2023-02-13 18:39:13,868 - Epoch: [184][ 600/ 1207] Overall Loss 0.205882 Objective Loss 0.205882 LR 0.000125 Time 0.020169 -2023-02-13 18:39:14,059 - Epoch: [184][ 610/ 1207] Overall Loss 0.206224 Objective Loss 0.206224 LR 0.000125 Time 0.020151 -2023-02-13 18:39:14,250 - Epoch: [184][ 620/ 1207] Overall Loss 0.206371 Objective Loss 0.206371 LR 0.000125 Time 0.020133 -2023-02-13 18:39:14,440 - Epoch: [184][ 630/ 1207] Overall Loss 0.206088 Objective Loss 0.206088 LR 0.000125 Time 0.020115 -2023-02-13 18:39:14,630 - Epoch: [184][ 640/ 1207] Overall Loss 0.206015 Objective Loss 0.206015 LR 0.000125 Time 0.020097 -2023-02-13 18:39:14,821 - Epoch: [184][ 650/ 1207] Overall Loss 0.206077 Objective Loss 0.206077 LR 0.000125 Time 0.020081 -2023-02-13 18:39:15,012 - Epoch: [184][ 660/ 1207] Overall Loss 0.206008 Objective Loss 0.206008 LR 0.000125 Time 0.020066 -2023-02-13 18:39:15,203 - Epoch: [184][ 670/ 1207] Overall Loss 0.206162 Objective Loss 0.206162 LR 0.000125 Time 0.020050 -2023-02-13 18:39:15,394 - Epoch: [184][ 680/ 1207] Overall Loss 0.206424 Objective Loss 0.206424 LR 0.000125 Time 0.020036 -2023-02-13 18:39:15,585 - Epoch: [184][ 690/ 1207] Overall Loss 0.206515 Objective Loss 0.206515 LR 0.000125 Time 0.020022 -2023-02-13 18:39:15,778 - Epoch: [184][ 700/ 1207] Overall Loss 0.206720 Objective Loss 0.206720 LR 0.000125 Time 0.020011 -2023-02-13 18:39:15,973 - Epoch: [184][ 710/ 1207] Overall Loss 0.207038 Objective Loss 0.207038 LR 0.000125 Time 0.020003 -2023-02-13 18:39:16,170 - Epoch: [184][ 720/ 1207] Overall Loss 0.207204 Objective Loss 0.207204 LR 0.000125 Time 0.019999 -2023-02-13 18:39:16,365 - Epoch: [184][ 730/ 1207] Overall Loss 0.206962 Objective Loss 0.206962 LR 0.000125 Time 0.019991 -2023-02-13 18:39:16,562 - Epoch: [184][ 740/ 1207] Overall Loss 0.206927 Objective Loss 0.206927 LR 0.000125 Time 0.019987 -2023-02-13 18:39:16,755 - Epoch: [184][ 750/ 1207] Overall Loss 0.207073 Objective Loss 0.207073 LR 0.000125 Time 0.019977 -2023-02-13 18:39:16,953 - Epoch: [184][ 760/ 1207] Overall Loss 0.207055 Objective Loss 0.207055 LR 0.000125 Time 0.019975 -2023-02-13 18:39:17,148 - Epoch: [184][ 770/ 1207] Overall Loss 0.207209 Objective Loss 0.207209 LR 0.000125 Time 0.019967 -2023-02-13 18:39:17,345 - Epoch: [184][ 780/ 1207] Overall Loss 0.207372 Objective Loss 0.207372 LR 0.000125 Time 0.019964 -2023-02-13 18:39:17,539 - Epoch: [184][ 790/ 1207] Overall Loss 0.207536 Objective Loss 0.207536 LR 0.000125 Time 0.019956 -2023-02-13 18:39:17,735 - Epoch: [184][ 800/ 1207] Overall Loss 0.207452 Objective Loss 0.207452 LR 0.000125 Time 0.019952 -2023-02-13 18:39:17,930 - Epoch: [184][ 810/ 1207] Overall Loss 0.207620 Objective Loss 0.207620 LR 0.000125 Time 0.019944 -2023-02-13 18:39:18,126 - Epoch: [184][ 820/ 1207] Overall Loss 0.207659 Objective Loss 0.207659 LR 0.000125 Time 0.019941 -2023-02-13 18:39:18,321 - Epoch: [184][ 830/ 1207] Overall Loss 0.207460 Objective Loss 0.207460 LR 0.000125 Time 0.019934 -2023-02-13 18:39:18,517 - Epoch: [184][ 840/ 1207] Overall Loss 0.207280 Objective Loss 0.207280 LR 0.000125 Time 0.019931 -2023-02-13 18:39:18,711 - Epoch: [184][ 850/ 1207] Overall Loss 0.207266 Objective Loss 0.207266 LR 0.000125 Time 0.019923 -2023-02-13 18:39:18,908 - Epoch: [184][ 860/ 1207] Overall Loss 0.207154 Objective Loss 0.207154 LR 0.000125 Time 0.019920 -2023-02-13 18:39:19,102 - Epoch: [184][ 870/ 1207] Overall Loss 0.207219 Objective Loss 0.207219 LR 0.000125 Time 0.019914 -2023-02-13 18:39:19,300 - Epoch: [184][ 880/ 1207] Overall Loss 0.207393 Objective Loss 0.207393 LR 0.000125 Time 0.019912 -2023-02-13 18:39:19,494 - Epoch: [184][ 890/ 1207] Overall Loss 0.207385 Objective Loss 0.207385 LR 0.000125 Time 0.019906 -2023-02-13 18:39:19,691 - Epoch: [184][ 900/ 1207] Overall Loss 0.207218 Objective Loss 0.207218 LR 0.000125 Time 0.019903 -2023-02-13 18:39:19,885 - Epoch: [184][ 910/ 1207] Overall Loss 0.207333 Objective Loss 0.207333 LR 0.000125 Time 0.019897 -2023-02-13 18:39:20,082 - Epoch: [184][ 920/ 1207] Overall Loss 0.207495 Objective Loss 0.207495 LR 0.000125 Time 0.019895 -2023-02-13 18:39:20,276 - Epoch: [184][ 930/ 1207] Overall Loss 0.207353 Objective Loss 0.207353 LR 0.000125 Time 0.019890 -2023-02-13 18:39:20,473 - Epoch: [184][ 940/ 1207] Overall Loss 0.207219 Objective Loss 0.207219 LR 0.000125 Time 0.019887 -2023-02-13 18:39:20,667 - Epoch: [184][ 950/ 1207] Overall Loss 0.207309 Objective Loss 0.207309 LR 0.000125 Time 0.019881 -2023-02-13 18:39:20,863 - Epoch: [184][ 960/ 1207] Overall Loss 0.207268 Objective Loss 0.207268 LR 0.000125 Time 0.019878 -2023-02-13 18:39:21,059 - Epoch: [184][ 970/ 1207] Overall Loss 0.207232 Objective Loss 0.207232 LR 0.000125 Time 0.019874 -2023-02-13 18:39:21,255 - Epoch: [184][ 980/ 1207] Overall Loss 0.207239 Objective Loss 0.207239 LR 0.000125 Time 0.019871 -2023-02-13 18:39:21,449 - Epoch: [184][ 990/ 1207] Overall Loss 0.207263 Objective Loss 0.207263 LR 0.000125 Time 0.019866 -2023-02-13 18:39:21,646 - Epoch: [184][ 1000/ 1207] Overall Loss 0.207194 Objective Loss 0.207194 LR 0.000125 Time 0.019864 -2023-02-13 18:39:21,839 - Epoch: [184][ 1010/ 1207] Overall Loss 0.207319 Objective Loss 0.207319 LR 0.000125 Time 0.019858 -2023-02-13 18:39:22,036 - Epoch: [184][ 1020/ 1207] Overall Loss 0.207035 Objective Loss 0.207035 LR 0.000125 Time 0.019856 -2023-02-13 18:39:22,230 - Epoch: [184][ 1030/ 1207] Overall Loss 0.206883 Objective Loss 0.206883 LR 0.000125 Time 0.019852 -2023-02-13 18:39:22,427 - Epoch: [184][ 1040/ 1207] Overall Loss 0.206809 Objective Loss 0.206809 LR 0.000125 Time 0.019849 -2023-02-13 18:39:22,621 - Epoch: [184][ 1050/ 1207] Overall Loss 0.206423 Objective Loss 0.206423 LR 0.000125 Time 0.019845 -2023-02-13 18:39:22,818 - Epoch: [184][ 1060/ 1207] Overall Loss 0.206425 Objective Loss 0.206425 LR 0.000125 Time 0.019843 -2023-02-13 18:39:23,012 - Epoch: [184][ 1070/ 1207] Overall Loss 0.206371 Objective Loss 0.206371 LR 0.000125 Time 0.019839 -2023-02-13 18:39:23,210 - Epoch: [184][ 1080/ 1207] Overall Loss 0.206399 Objective Loss 0.206399 LR 0.000125 Time 0.019838 -2023-02-13 18:39:23,404 - Epoch: [184][ 1090/ 1207] Overall Loss 0.206334 Objective Loss 0.206334 LR 0.000125 Time 0.019834 -2023-02-13 18:39:23,601 - Epoch: [184][ 1100/ 1207] Overall Loss 0.206399 Objective Loss 0.206399 LR 0.000125 Time 0.019832 -2023-02-13 18:39:23,794 - Epoch: [184][ 1110/ 1207] Overall Loss 0.206526 Objective Loss 0.206526 LR 0.000125 Time 0.019827 -2023-02-13 18:39:23,991 - Epoch: [184][ 1120/ 1207] Overall Loss 0.206575 Objective Loss 0.206575 LR 0.000125 Time 0.019826 -2023-02-13 18:39:24,185 - Epoch: [184][ 1130/ 1207] Overall Loss 0.206608 Objective Loss 0.206608 LR 0.000125 Time 0.019822 -2023-02-13 18:39:24,383 - Epoch: [184][ 1140/ 1207] Overall Loss 0.206427 Objective Loss 0.206427 LR 0.000125 Time 0.019821 -2023-02-13 18:39:24,577 - Epoch: [184][ 1150/ 1207] Overall Loss 0.206488 Objective Loss 0.206488 LR 0.000125 Time 0.019817 -2023-02-13 18:39:24,768 - Epoch: [184][ 1160/ 1207] Overall Loss 0.206434 Objective Loss 0.206434 LR 0.000125 Time 0.019810 -2023-02-13 18:39:24,958 - Epoch: [184][ 1170/ 1207] Overall Loss 0.206456 Objective Loss 0.206456 LR 0.000125 Time 0.019804 -2023-02-13 18:39:25,148 - Epoch: [184][ 1180/ 1207] Overall Loss 0.206662 Objective Loss 0.206662 LR 0.000125 Time 0.019796 -2023-02-13 18:39:25,339 - Epoch: [184][ 1190/ 1207] Overall Loss 0.206644 Objective Loss 0.206644 LR 0.000125 Time 0.019790 -2023-02-13 18:39:25,586 - Epoch: [184][ 1200/ 1207] Overall Loss 0.206783 Objective Loss 0.206783 LR 0.000125 Time 0.019831 -2023-02-13 18:39:25,701 - Epoch: [184][ 1207/ 1207] Overall Loss 0.206751 Objective Loss 0.206751 Top1 88.414634 Top5 98.475610 LR 0.000125 Time 0.019811 -2023-02-13 18:39:25,782 - --- validate (epoch=184)----------- -2023-02-13 18:39:25,783 - 34311 samples (256 per mini-batch) -2023-02-13 18:39:26,191 - Epoch: [184][ 10/ 135] Loss 0.294650 Top1 85.351562 Top5 97.890625 -2023-02-13 18:39:26,326 - Epoch: [184][ 20/ 135] Loss 0.283796 Top1 85.488281 Top5 97.871094 -2023-02-13 18:39:26,464 - Epoch: [184][ 30/ 135] Loss 0.290577 Top1 85.468750 Top5 97.942708 -2023-02-13 18:39:26,588 - Epoch: [184][ 40/ 135] Loss 0.291722 Top1 85.468750 Top5 98.007812 -2023-02-13 18:39:26,726 - Epoch: [184][ 50/ 135] Loss 0.289360 Top1 85.609375 Top5 97.968750 -2023-02-13 18:39:26,851 - Epoch: [184][ 60/ 135] Loss 0.287154 Top1 85.540365 Top5 97.936198 -2023-02-13 18:39:26,976 - Epoch: [184][ 70/ 135] Loss 0.291161 Top1 85.390625 Top5 97.929688 -2023-02-13 18:39:27,100 - Epoch: [184][ 80/ 135] Loss 0.290274 Top1 85.454102 Top5 97.929688 -2023-02-13 18:39:27,230 - Epoch: [184][ 90/ 135] Loss 0.292681 Top1 85.334201 Top5 97.942708 -2023-02-13 18:39:27,360 - Epoch: [184][ 100/ 135] Loss 0.294304 Top1 85.382812 Top5 97.902344 -2023-02-13 18:39:27,488 - Epoch: [184][ 110/ 135] Loss 0.296359 Top1 85.358665 Top5 97.897727 -2023-02-13 18:39:27,614 - Epoch: [184][ 120/ 135] Loss 0.297972 Top1 85.309245 Top5 97.871094 -2023-02-13 18:39:27,747 - Epoch: [184][ 130/ 135] Loss 0.297711 Top1 85.321514 Top5 97.821514 -2023-02-13 18:39:27,793 - Epoch: [184][ 135/ 135] Loss 0.299745 Top1 85.331235 Top5 97.831599 -2023-02-13 18:39:27,863 - ==> Top1: 85.331 Top5: 97.832 Loss: 0.300 - -2023-02-13 18:39:27,864 - ==> Confusion: -[[ 860 3 6 1 5 3 0 0 1 54 2 2 1 4 6 5 2 2 0 3 7] - [ 3 939 2 1 13 28 2 15 2 2 2 0 2 0 0 2 4 0 3 3 10] - [ 5 4 956 15 7 1 13 14 0 1 2 3 2 3 3 10 3 1 7 2 6] - [ 5 1 15 919 1 7 0 2 2 3 11 0 8 0 16 0 3 6 11 1 5] - [ 10 8 1 0 992 11 1 3 1 2 0 6 2 3 8 7 5 0 1 2 3] - [ 2 12 1 5 1 981 2 16 1 3 0 7 4 13 0 5 5 1 2 4 5] - [ 2 2 14 1 0 5 1046 4 0 1 2 1 3 2 0 5 0 1 2 4 4] - [ 4 3 10 2 1 29 6 924 0 1 0 5 3 2 0 0 0 1 19 9 5] - [ 11 4 1 3 1 0 0 1 912 42 4 2 0 12 9 1 1 0 3 0 2] - [ 67 0 2 0 12 0 0 1 31 871 0 0 0 13 3 4 1 2 1 1 3] - [ 1 1 2 10 2 3 2 6 16 2 978 2 1 9 4 1 3 0 4 1 3] - [ 4 3 2 0 1 15 1 5 1 2 0 906 27 7 1 7 1 10 1 8 3] - [ 0 0 1 7 0 2 0 1 2 1 0 24 889 1 0 9 1 13 1 0 7] - [ 8 3 2 0 6 7 1 1 6 16 6 3 2 944 6 6 2 0 0 0 5] - [ 6 1 1 23 6 3 0 1 25 8 2 2 2 2 993 0 1 5 3 0 8] - [ 4 2 7 0 6 1 4 0 1 0 0 4 4 3 0 985 5 8 0 6 6] - [ 5 7 1 2 6 2 0 1 1 0 0 0 1 2 2 15 994 1 2 3 16] - [ 5 3 0 5 1 3 1 0 0 1 0 6 12 1 1 16 0 991 0 0 5] - [ 4 5 5 9 1 2 0 22 3 0 4 0 2 0 14 3 1 2 1007 2 0] - [ 1 3 0 0 2 4 10 10 0 0 0 16 3 2 1 5 5 4 0 1075 7] - [ 144 178 216 128 123 213 83 169 94 109 180 103 312 278 152 123 205 105 167 236 10116]] - -2023-02-13 18:39:27,866 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:39:27,866 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:39:27,872 - - -2023-02-13 18:39:27,872 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:39:28,766 - Epoch: [185][ 10/ 1207] Overall Loss 0.213401 Objective Loss 0.213401 LR 0.000125 Time 0.089387 -2023-02-13 18:39:28,957 - Epoch: [185][ 20/ 1207] Overall Loss 0.203709 Objective Loss 0.203709 LR 0.000125 Time 0.054217 -2023-02-13 18:39:29,146 - Epoch: [185][ 30/ 1207] Overall Loss 0.207764 Objective Loss 0.207764 LR 0.000125 Time 0.042431 -2023-02-13 18:39:29,336 - Epoch: [185][ 40/ 1207] Overall Loss 0.208735 Objective Loss 0.208735 LR 0.000125 Time 0.036559 -2023-02-13 18:39:29,526 - Epoch: [185][ 50/ 1207] Overall Loss 0.207010 Objective Loss 0.207010 LR 0.000125 Time 0.033032 -2023-02-13 18:39:29,715 - Epoch: [185][ 60/ 1207] Overall Loss 0.204172 Objective Loss 0.204172 LR 0.000125 Time 0.030674 -2023-02-13 18:39:29,903 - Epoch: [185][ 70/ 1207] Overall Loss 0.206680 Objective Loss 0.206680 LR 0.000125 Time 0.028980 -2023-02-13 18:39:30,093 - Epoch: [185][ 80/ 1207] Overall Loss 0.206076 Objective Loss 0.206076 LR 0.000125 Time 0.027724 -2023-02-13 18:39:30,283 - Epoch: [185][ 90/ 1207] Overall Loss 0.207209 Objective Loss 0.207209 LR 0.000125 Time 0.026745 -2023-02-13 18:39:30,472 - Epoch: [185][ 100/ 1207] Overall Loss 0.206843 Objective Loss 0.206843 LR 0.000125 Time 0.025960 -2023-02-13 18:39:30,661 - Epoch: [185][ 110/ 1207] Overall Loss 0.205530 Objective Loss 0.205530 LR 0.000125 Time 0.025319 -2023-02-13 18:39:30,850 - Epoch: [185][ 120/ 1207] Overall Loss 0.206740 Objective Loss 0.206740 LR 0.000125 Time 0.024779 -2023-02-13 18:39:31,041 - Epoch: [185][ 130/ 1207] Overall Loss 0.207870 Objective Loss 0.207870 LR 0.000125 Time 0.024336 -2023-02-13 18:39:31,230 - Epoch: [185][ 140/ 1207] Overall Loss 0.206128 Objective Loss 0.206128 LR 0.000125 Time 0.023948 -2023-02-13 18:39:31,420 - Epoch: [185][ 150/ 1207] Overall Loss 0.206380 Objective Loss 0.206380 LR 0.000125 Time 0.023612 -2023-02-13 18:39:31,609 - Epoch: [185][ 160/ 1207] Overall Loss 0.208120 Objective Loss 0.208120 LR 0.000125 Time 0.023320 -2023-02-13 18:39:31,798 - Epoch: [185][ 170/ 1207] Overall Loss 0.207962 Objective Loss 0.207962 LR 0.000125 Time 0.023058 -2023-02-13 18:39:31,989 - Epoch: [185][ 180/ 1207] Overall Loss 0.207997 Objective Loss 0.207997 LR 0.000125 Time 0.022833 -2023-02-13 18:39:32,178 - Epoch: [185][ 190/ 1207] Overall Loss 0.208237 Objective Loss 0.208237 LR 0.000125 Time 0.022623 -2023-02-13 18:39:32,367 - Epoch: [185][ 200/ 1207] Overall Loss 0.208589 Objective Loss 0.208589 LR 0.000125 Time 0.022438 -2023-02-13 18:39:32,556 - Epoch: [185][ 210/ 1207] Overall Loss 0.207592 Objective Loss 0.207592 LR 0.000125 Time 0.022267 -2023-02-13 18:39:32,746 - Epoch: [185][ 220/ 1207] Overall Loss 0.207350 Objective Loss 0.207350 LR 0.000125 Time 0.022115 -2023-02-13 18:39:32,935 - Epoch: [185][ 230/ 1207] Overall Loss 0.206806 Objective Loss 0.206806 LR 0.000125 Time 0.021973 -2023-02-13 18:39:33,125 - Epoch: [185][ 240/ 1207] Overall Loss 0.207100 Objective Loss 0.207100 LR 0.000125 Time 0.021848 -2023-02-13 18:39:33,315 - Epoch: [185][ 250/ 1207] Overall Loss 0.207016 Objective Loss 0.207016 LR 0.000125 Time 0.021732 -2023-02-13 18:39:33,504 - Epoch: [185][ 260/ 1207] Overall Loss 0.208127 Objective Loss 0.208127 LR 0.000125 Time 0.021624 -2023-02-13 18:39:33,696 - Epoch: [185][ 270/ 1207] Overall Loss 0.209088 Objective Loss 0.209088 LR 0.000125 Time 0.021532 -2023-02-13 18:39:33,890 - Epoch: [185][ 280/ 1207] Overall Loss 0.209040 Objective Loss 0.209040 LR 0.000125 Time 0.021452 -2023-02-13 18:39:34,083 - Epoch: [185][ 290/ 1207] Overall Loss 0.208532 Objective Loss 0.208532 LR 0.000125 Time 0.021378 -2023-02-13 18:39:34,277 - Epoch: [185][ 300/ 1207] Overall Loss 0.209114 Objective Loss 0.209114 LR 0.000125 Time 0.021312 -2023-02-13 18:39:34,469 - Epoch: [185][ 310/ 1207] Overall Loss 0.209224 Objective Loss 0.209224 LR 0.000125 Time 0.021242 -2023-02-13 18:39:34,662 - Epoch: [185][ 320/ 1207] Overall Loss 0.209068 Objective Loss 0.209068 LR 0.000125 Time 0.021179 -2023-02-13 18:39:34,855 - Epoch: [185][ 330/ 1207] Overall Loss 0.208853 Objective Loss 0.208853 LR 0.000125 Time 0.021122 -2023-02-13 18:39:35,048 - Epoch: [185][ 340/ 1207] Overall Loss 0.209216 Objective Loss 0.209216 LR 0.000125 Time 0.021066 -2023-02-13 18:39:35,240 - Epoch: [185][ 350/ 1207] Overall Loss 0.209059 Objective Loss 0.209059 LR 0.000125 Time 0.021012 -2023-02-13 18:39:35,433 - Epoch: [185][ 360/ 1207] Overall Loss 0.208467 Objective Loss 0.208467 LR 0.000125 Time 0.020964 -2023-02-13 18:39:35,625 - Epoch: [185][ 370/ 1207] Overall Loss 0.208532 Objective Loss 0.208532 LR 0.000125 Time 0.020915 -2023-02-13 18:39:35,818 - Epoch: [185][ 380/ 1207] Overall Loss 0.208719 Objective Loss 0.208719 LR 0.000125 Time 0.020871 -2023-02-13 18:39:36,012 - Epoch: [185][ 390/ 1207] Overall Loss 0.208961 Objective Loss 0.208961 LR 0.000125 Time 0.020833 -2023-02-13 18:39:36,205 - Epoch: [185][ 400/ 1207] Overall Loss 0.208850 Objective Loss 0.208850 LR 0.000125 Time 0.020793 -2023-02-13 18:39:36,398 - Epoch: [185][ 410/ 1207] Overall Loss 0.208626 Objective Loss 0.208626 LR 0.000125 Time 0.020756 -2023-02-13 18:39:36,591 - Epoch: [185][ 420/ 1207] Overall Loss 0.208559 Objective Loss 0.208559 LR 0.000125 Time 0.020720 -2023-02-13 18:39:36,783 - Epoch: [185][ 430/ 1207] Overall Loss 0.208342 Objective Loss 0.208342 LR 0.000125 Time 0.020685 -2023-02-13 18:39:36,977 - Epoch: [185][ 440/ 1207] Overall Loss 0.208707 Objective Loss 0.208707 LR 0.000125 Time 0.020655 -2023-02-13 18:39:37,171 - Epoch: [185][ 450/ 1207] Overall Loss 0.209189 Objective Loss 0.209189 LR 0.000125 Time 0.020626 -2023-02-13 18:39:37,364 - Epoch: [185][ 460/ 1207] Overall Loss 0.209607 Objective Loss 0.209607 LR 0.000125 Time 0.020597 -2023-02-13 18:39:37,557 - Epoch: [185][ 470/ 1207] Overall Loss 0.209481 Objective Loss 0.209481 LR 0.000125 Time 0.020567 -2023-02-13 18:39:37,750 - Epoch: [185][ 480/ 1207] Overall Loss 0.209918 Objective Loss 0.209918 LR 0.000125 Time 0.020541 -2023-02-13 18:39:37,944 - Epoch: [185][ 490/ 1207] Overall Loss 0.209601 Objective Loss 0.209601 LR 0.000125 Time 0.020516 -2023-02-13 18:39:38,137 - Epoch: [185][ 500/ 1207] Overall Loss 0.209816 Objective Loss 0.209816 LR 0.000125 Time 0.020491 -2023-02-13 18:39:38,331 - Epoch: [185][ 510/ 1207] Overall Loss 0.209584 Objective Loss 0.209584 LR 0.000125 Time 0.020469 -2023-02-13 18:39:38,524 - Epoch: [185][ 520/ 1207] Overall Loss 0.209414 Objective Loss 0.209414 LR 0.000125 Time 0.020446 -2023-02-13 18:39:38,717 - Epoch: [185][ 530/ 1207] Overall Loss 0.209673 Objective Loss 0.209673 LR 0.000125 Time 0.020424 -2023-02-13 18:39:38,910 - Epoch: [185][ 540/ 1207] Overall Loss 0.209337 Objective Loss 0.209337 LR 0.000125 Time 0.020403 -2023-02-13 18:39:39,103 - Epoch: [185][ 550/ 1207] Overall Loss 0.209363 Objective Loss 0.209363 LR 0.000125 Time 0.020381 -2023-02-13 18:39:39,296 - Epoch: [185][ 560/ 1207] Overall Loss 0.209165 Objective Loss 0.209165 LR 0.000125 Time 0.020362 -2023-02-13 18:39:39,489 - Epoch: [185][ 570/ 1207] Overall Loss 0.209104 Objective Loss 0.209104 LR 0.000125 Time 0.020343 -2023-02-13 18:39:39,683 - Epoch: [185][ 580/ 1207] Overall Loss 0.208513 Objective Loss 0.208513 LR 0.000125 Time 0.020325 -2023-02-13 18:39:39,876 - Epoch: [185][ 590/ 1207] Overall Loss 0.208860 Objective Loss 0.208860 LR 0.000125 Time 0.020307 -2023-02-13 18:39:40,068 - Epoch: [185][ 600/ 1207] Overall Loss 0.209072 Objective Loss 0.209072 LR 0.000125 Time 0.020289 -2023-02-13 18:39:40,261 - Epoch: [185][ 610/ 1207] Overall Loss 0.209184 Objective Loss 0.209184 LR 0.000125 Time 0.020272 -2023-02-13 18:39:40,455 - Epoch: [185][ 620/ 1207] Overall Loss 0.209184 Objective Loss 0.209184 LR 0.000125 Time 0.020256 -2023-02-13 18:39:40,646 - Epoch: [185][ 630/ 1207] Overall Loss 0.209044 Objective Loss 0.209044 LR 0.000125 Time 0.020238 -2023-02-13 18:39:40,839 - Epoch: [185][ 640/ 1207] Overall Loss 0.209119 Objective Loss 0.209119 LR 0.000125 Time 0.020223 -2023-02-13 18:39:41,033 - Epoch: [185][ 650/ 1207] Overall Loss 0.208715 Objective Loss 0.208715 LR 0.000125 Time 0.020210 -2023-02-13 18:39:41,233 - Epoch: [185][ 660/ 1207] Overall Loss 0.208795 Objective Loss 0.208795 LR 0.000125 Time 0.020206 -2023-02-13 18:39:41,430 - Epoch: [185][ 670/ 1207] Overall Loss 0.208534 Objective Loss 0.208534 LR 0.000125 Time 0.020198 -2023-02-13 18:39:41,631 - Epoch: [185][ 680/ 1207] Overall Loss 0.208103 Objective Loss 0.208103 LR 0.000125 Time 0.020195 -2023-02-13 18:39:41,827 - Epoch: [185][ 690/ 1207] Overall Loss 0.208097 Objective Loss 0.208097 LR 0.000125 Time 0.020187 -2023-02-13 18:39:42,028 - Epoch: [185][ 700/ 1207] Overall Loss 0.208447 Objective Loss 0.208447 LR 0.000125 Time 0.020184 -2023-02-13 18:39:42,224 - Epoch: [185][ 710/ 1207] Overall Loss 0.208564 Objective Loss 0.208564 LR 0.000125 Time 0.020176 -2023-02-13 18:39:42,425 - Epoch: [185][ 720/ 1207] Overall Loss 0.208612 Objective Loss 0.208612 LR 0.000125 Time 0.020174 -2023-02-13 18:39:42,622 - Epoch: [185][ 730/ 1207] Overall Loss 0.208308 Objective Loss 0.208308 LR 0.000125 Time 0.020167 -2023-02-13 18:39:42,822 - Epoch: [185][ 740/ 1207] Overall Loss 0.208346 Objective Loss 0.208346 LR 0.000125 Time 0.020164 -2023-02-13 18:39:43,018 - Epoch: [185][ 750/ 1207] Overall Loss 0.208104 Objective Loss 0.208104 LR 0.000125 Time 0.020156 -2023-02-13 18:39:43,218 - Epoch: [185][ 760/ 1207] Overall Loss 0.208058 Objective Loss 0.208058 LR 0.000125 Time 0.020154 -2023-02-13 18:39:43,415 - Epoch: [185][ 770/ 1207] Overall Loss 0.208206 Objective Loss 0.208206 LR 0.000125 Time 0.020147 -2023-02-13 18:39:43,615 - Epoch: [185][ 780/ 1207] Overall Loss 0.207977 Objective Loss 0.207977 LR 0.000125 Time 0.020145 -2023-02-13 18:39:43,811 - Epoch: [185][ 790/ 1207] Overall Loss 0.207811 Objective Loss 0.207811 LR 0.000125 Time 0.020137 -2023-02-13 18:39:44,011 - Epoch: [185][ 800/ 1207] Overall Loss 0.207617 Objective Loss 0.207617 LR 0.000125 Time 0.020135 -2023-02-13 18:39:44,207 - Epoch: [185][ 810/ 1207] Overall Loss 0.207652 Objective Loss 0.207652 LR 0.000125 Time 0.020129 -2023-02-13 18:39:44,408 - Epoch: [185][ 820/ 1207] Overall Loss 0.207700 Objective Loss 0.207700 LR 0.000125 Time 0.020127 -2023-02-13 18:39:44,604 - Epoch: [185][ 830/ 1207] Overall Loss 0.207610 Objective Loss 0.207610 LR 0.000125 Time 0.020121 -2023-02-13 18:39:44,804 - Epoch: [185][ 840/ 1207] Overall Loss 0.207499 Objective Loss 0.207499 LR 0.000125 Time 0.020119 -2023-02-13 18:39:45,001 - Epoch: [185][ 850/ 1207] Overall Loss 0.207269 Objective Loss 0.207269 LR 0.000125 Time 0.020113 -2023-02-13 18:39:45,200 - Epoch: [185][ 860/ 1207] Overall Loss 0.207039 Objective Loss 0.207039 LR 0.000125 Time 0.020111 -2023-02-13 18:39:45,398 - Epoch: [185][ 870/ 1207] Overall Loss 0.207004 Objective Loss 0.207004 LR 0.000125 Time 0.020106 -2023-02-13 18:39:45,597 - Epoch: [185][ 880/ 1207] Overall Loss 0.207204 Objective Loss 0.207204 LR 0.000125 Time 0.020104 -2023-02-13 18:39:45,794 - Epoch: [185][ 890/ 1207] Overall Loss 0.207156 Objective Loss 0.207156 LR 0.000125 Time 0.020098 -2023-02-13 18:39:45,995 - Epoch: [185][ 900/ 1207] Overall Loss 0.207099 Objective Loss 0.207099 LR 0.000125 Time 0.020098 -2023-02-13 18:39:46,191 - Epoch: [185][ 910/ 1207] Overall Loss 0.206874 Objective Loss 0.206874 LR 0.000125 Time 0.020092 -2023-02-13 18:39:46,392 - Epoch: [185][ 920/ 1207] Overall Loss 0.206979 Objective Loss 0.206979 LR 0.000125 Time 0.020092 -2023-02-13 18:39:46,589 - Epoch: [185][ 930/ 1207] Overall Loss 0.206899 Objective Loss 0.206899 LR 0.000125 Time 0.020087 -2023-02-13 18:39:46,788 - Epoch: [185][ 940/ 1207] Overall Loss 0.206850 Objective Loss 0.206850 LR 0.000125 Time 0.020085 -2023-02-13 18:39:46,986 - Epoch: [185][ 950/ 1207] Overall Loss 0.206716 Objective Loss 0.206716 LR 0.000125 Time 0.020081 -2023-02-13 18:39:47,186 - Epoch: [185][ 960/ 1207] Overall Loss 0.206539 Objective Loss 0.206539 LR 0.000125 Time 0.020081 -2023-02-13 18:39:47,384 - Epoch: [185][ 970/ 1207] Overall Loss 0.206755 Objective Loss 0.206755 LR 0.000125 Time 0.020077 -2023-02-13 18:39:47,584 - Epoch: [185][ 980/ 1207] Overall Loss 0.206781 Objective Loss 0.206781 LR 0.000125 Time 0.020076 -2023-02-13 18:39:47,781 - Epoch: [185][ 990/ 1207] Overall Loss 0.206758 Objective Loss 0.206758 LR 0.000125 Time 0.020072 -2023-02-13 18:39:47,981 - Epoch: [185][ 1000/ 1207] Overall Loss 0.206626 Objective Loss 0.206626 LR 0.000125 Time 0.020070 -2023-02-13 18:39:48,176 - Epoch: [185][ 1010/ 1207] Overall Loss 0.206595 Objective Loss 0.206595 LR 0.000125 Time 0.020065 -2023-02-13 18:39:48,377 - Epoch: [185][ 1020/ 1207] Overall Loss 0.206560 Objective Loss 0.206560 LR 0.000125 Time 0.020064 -2023-02-13 18:39:48,573 - Epoch: [185][ 1030/ 1207] Overall Loss 0.206398 Objective Loss 0.206398 LR 0.000125 Time 0.020060 -2023-02-13 18:39:48,773 - Epoch: [185][ 1040/ 1207] Overall Loss 0.206599 Objective Loss 0.206599 LR 0.000125 Time 0.020059 -2023-02-13 18:39:48,967 - Epoch: [185][ 1050/ 1207] Overall Loss 0.206554 Objective Loss 0.206554 LR 0.000125 Time 0.020052 -2023-02-13 18:39:49,158 - Epoch: [185][ 1060/ 1207] Overall Loss 0.206601 Objective Loss 0.206601 LR 0.000125 Time 0.020043 -2023-02-13 18:39:49,349 - Epoch: [185][ 1070/ 1207] Overall Loss 0.206734 Objective Loss 0.206734 LR 0.000125 Time 0.020034 -2023-02-13 18:39:49,540 - Epoch: [185][ 1080/ 1207] Overall Loss 0.206688 Objective Loss 0.206688 LR 0.000125 Time 0.020024 -2023-02-13 18:39:49,730 - Epoch: [185][ 1090/ 1207] Overall Loss 0.206706 Objective Loss 0.206706 LR 0.000125 Time 0.020015 -2023-02-13 18:39:49,921 - Epoch: [185][ 1100/ 1207] Overall Loss 0.206917 Objective Loss 0.206917 LR 0.000125 Time 0.020006 -2023-02-13 18:39:50,112 - Epoch: [185][ 1110/ 1207] Overall Loss 0.206995 Objective Loss 0.206995 LR 0.000125 Time 0.019997 -2023-02-13 18:39:50,302 - Epoch: [185][ 1120/ 1207] Overall Loss 0.206733 Objective Loss 0.206733 LR 0.000125 Time 0.019989 -2023-02-13 18:39:50,493 - Epoch: [185][ 1130/ 1207] Overall Loss 0.206607 Objective Loss 0.206607 LR 0.000125 Time 0.019980 -2023-02-13 18:39:50,684 - Epoch: [185][ 1140/ 1207] Overall Loss 0.206437 Objective Loss 0.206437 LR 0.000125 Time 0.019972 -2023-02-13 18:39:50,875 - Epoch: [185][ 1150/ 1207] Overall Loss 0.206414 Objective Loss 0.206414 LR 0.000125 Time 0.019964 -2023-02-13 18:39:51,067 - Epoch: [185][ 1160/ 1207] Overall Loss 0.206363 Objective Loss 0.206363 LR 0.000125 Time 0.019957 -2023-02-13 18:39:51,258 - Epoch: [185][ 1170/ 1207] Overall Loss 0.206268 Objective Loss 0.206268 LR 0.000125 Time 0.019949 -2023-02-13 18:39:51,450 - Epoch: [185][ 1180/ 1207] Overall Loss 0.206280 Objective Loss 0.206280 LR 0.000125 Time 0.019943 -2023-02-13 18:39:51,640 - Epoch: [185][ 1190/ 1207] Overall Loss 0.206055 Objective Loss 0.206055 LR 0.000125 Time 0.019935 -2023-02-13 18:39:51,888 - Epoch: [185][ 1200/ 1207] Overall Loss 0.206135 Objective Loss 0.206135 LR 0.000125 Time 0.019975 -2023-02-13 18:39:52,005 - Epoch: [185][ 1207/ 1207] Overall Loss 0.206144 Objective Loss 0.206144 Top1 87.500000 Top5 98.475610 LR 0.000125 Time 0.019956 -2023-02-13 18:39:52,077 - --- validate (epoch=185)----------- -2023-02-13 18:39:52,078 - 34311 samples (256 per mini-batch) -2023-02-13 18:39:52,478 - Epoch: [185][ 10/ 135] Loss 0.303604 Top1 84.726562 Top5 97.812500 -2023-02-13 18:39:52,608 - Epoch: [185][ 20/ 135] Loss 0.308471 Top1 85.605469 Top5 97.617188 -2023-02-13 18:39:52,740 - Epoch: [185][ 30/ 135] Loss 0.308852 Top1 84.986979 Top5 97.682292 -2023-02-13 18:39:52,866 - Epoch: [185][ 40/ 135] Loss 0.307147 Top1 84.912109 Top5 97.597656 -2023-02-13 18:39:52,994 - Epoch: [185][ 50/ 135] Loss 0.299288 Top1 85.226562 Top5 97.703125 -2023-02-13 18:39:53,125 - Epoch: [185][ 60/ 135] Loss 0.300477 Top1 85.117188 Top5 97.708333 -2023-02-13 18:39:53,256 - Epoch: [185][ 70/ 135] Loss 0.298217 Top1 85.083705 Top5 97.767857 -2023-02-13 18:39:53,387 - Epoch: [185][ 80/ 135] Loss 0.293176 Top1 85.234375 Top5 97.792969 -2023-02-13 18:39:53,518 - Epoch: [185][ 90/ 135] Loss 0.295065 Top1 85.225694 Top5 97.799479 -2023-02-13 18:39:53,649 - Epoch: [185][ 100/ 135] Loss 0.294765 Top1 85.273438 Top5 97.824219 -2023-02-13 18:39:53,780 - Epoch: [185][ 110/ 135] Loss 0.294053 Top1 85.305398 Top5 97.826705 -2023-02-13 18:39:53,908 - Epoch: [185][ 120/ 135] Loss 0.295216 Top1 85.273438 Top5 97.815755 -2023-02-13 18:39:54,042 - Epoch: [185][ 130/ 135] Loss 0.294988 Top1 85.243389 Top5 97.827524 -2023-02-13 18:39:54,090 - Epoch: [185][ 135/ 135] Loss 0.294869 Top1 85.214654 Top5 97.846172 -2023-02-13 18:39:54,159 - ==> Top1: 85.215 Top5: 97.846 Loss: 0.295 - -2023-02-13 18:39:54,160 - ==> Confusion: -[[ 863 6 7 1 8 2 0 0 1 47 0 6 1 4 6 2 3 2 0 1 7] - [ 5 940 2 2 12 28 1 16 2 0 0 0 1 0 0 2 4 1 2 5 10] - [ 9 5 964 9 4 1 14 14 0 1 3 4 1 3 2 6 3 1 6 2 6] - [ 5 1 19 902 2 5 0 2 1 2 14 0 9 0 22 3 4 5 12 1 7] - [ 13 7 0 1 996 10 1 1 1 1 1 6 2 2 7 6 3 2 1 3 2] - [ 1 11 0 3 5 978 5 13 3 4 0 12 1 13 0 3 6 1 2 5 4] - [ 2 1 11 1 1 7 1048 4 0 1 2 1 1 1 0 3 1 2 1 5 6] - [ 3 7 13 0 1 29 2 936 0 1 0 8 4 1 0 0 1 0 8 7 3] - [ 15 3 0 1 2 1 1 0 904 44 4 4 0 9 13 2 2 0 4 0 0] - [ 75 0 4 0 8 1 0 1 29 867 0 0 0 15 2 2 1 2 2 0 3] - [ 3 0 1 4 3 3 3 7 18 2 982 1 1 7 1 0 1 1 5 0 8] - [ 3 3 3 0 2 6 1 5 0 1 0 934 11 5 2 6 2 9 0 11 1] - [ 0 0 1 7 2 2 0 0 2 0 0 33 881 1 1 6 3 12 1 1 6] - [ 3 2 1 0 8 11 1 1 8 22 6 7 1 925 6 5 3 2 0 5 7] - [ 8 1 2 12 7 3 0 1 22 8 2 1 3 2 997 1 4 5 4 0 9] - [ 4 0 9 0 3 1 2 0 0 0 0 9 7 2 0 972 10 10 0 10 7] - [ 2 5 1 0 8 2 0 0 2 0 0 2 3 3 2 10 1002 0 0 5 14] - [ 3 1 1 3 1 1 2 0 0 1 2 10 12 0 0 18 1 987 0 2 6] - [ 4 5 4 5 2 1 1 25 4 1 3 1 3 0 13 1 1 1 1008 2 1] - [ 2 3 0 0 1 2 8 6 0 0 0 16 2 2 0 6 3 3 0 1087 7] - [ 167 209 225 95 129 219 88 199 94 93 165 112 303 268 147 90 242 90 165 269 10065]] - -2023-02-13 18:39:54,162 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:39:54,162 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:39:54,168 - - -2023-02-13 18:39:54,168 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:39:55,161 - Epoch: [186][ 10/ 1207] Overall Loss 0.203684 Objective Loss 0.203684 LR 0.000125 Time 0.099186 -2023-02-13 18:39:55,358 - Epoch: [186][ 20/ 1207] Overall Loss 0.212864 Objective Loss 0.212864 LR 0.000125 Time 0.059440 -2023-02-13 18:39:55,547 - Epoch: [186][ 30/ 1207] Overall Loss 0.204869 Objective Loss 0.204869 LR 0.000125 Time 0.045927 -2023-02-13 18:39:55,736 - Epoch: [186][ 40/ 1207] Overall Loss 0.205357 Objective Loss 0.205357 LR 0.000125 Time 0.039156 -2023-02-13 18:39:55,927 - Epoch: [186][ 50/ 1207] Overall Loss 0.206641 Objective Loss 0.206641 LR 0.000125 Time 0.035131 -2023-02-13 18:39:56,116 - Epoch: [186][ 60/ 1207] Overall Loss 0.203385 Objective Loss 0.203385 LR 0.000125 Time 0.032418 -2023-02-13 18:39:56,305 - Epoch: [186][ 70/ 1207] Overall Loss 0.202850 Objective Loss 0.202850 LR 0.000125 Time 0.030481 -2023-02-13 18:39:56,493 - Epoch: [186][ 80/ 1207] Overall Loss 0.201737 Objective Loss 0.201737 LR 0.000125 Time 0.029023 -2023-02-13 18:39:56,683 - Epoch: [186][ 90/ 1207] Overall Loss 0.201712 Objective Loss 0.201712 LR 0.000125 Time 0.027897 -2023-02-13 18:39:56,871 - Epoch: [186][ 100/ 1207] Overall Loss 0.201524 Objective Loss 0.201524 LR 0.000125 Time 0.026989 -2023-02-13 18:39:57,060 - Epoch: [186][ 110/ 1207] Overall Loss 0.202399 Objective Loss 0.202399 LR 0.000125 Time 0.026252 -2023-02-13 18:39:57,249 - Epoch: [186][ 120/ 1207] Overall Loss 0.201884 Objective Loss 0.201884 LR 0.000125 Time 0.025631 -2023-02-13 18:39:57,438 - Epoch: [186][ 130/ 1207] Overall Loss 0.202174 Objective Loss 0.202174 LR 0.000125 Time 0.025112 -2023-02-13 18:39:57,627 - Epoch: [186][ 140/ 1207] Overall Loss 0.201494 Objective Loss 0.201494 LR 0.000125 Time 0.024665 -2023-02-13 18:39:57,815 - Epoch: [186][ 150/ 1207] Overall Loss 0.202636 Objective Loss 0.202636 LR 0.000125 Time 0.024275 -2023-02-13 18:39:58,005 - Epoch: [186][ 160/ 1207] Overall Loss 0.204491 Objective Loss 0.204491 LR 0.000125 Time 0.023939 -2023-02-13 18:39:58,195 - Epoch: [186][ 170/ 1207] Overall Loss 0.203394 Objective Loss 0.203394 LR 0.000125 Time 0.023647 -2023-02-13 18:39:58,385 - Epoch: [186][ 180/ 1207] Overall Loss 0.202939 Objective Loss 0.202939 LR 0.000125 Time 0.023388 -2023-02-13 18:39:58,575 - Epoch: [186][ 190/ 1207] Overall Loss 0.202741 Objective Loss 0.202741 LR 0.000125 Time 0.023155 -2023-02-13 18:39:58,764 - Epoch: [186][ 200/ 1207] Overall Loss 0.202704 Objective Loss 0.202704 LR 0.000125 Time 0.022942 -2023-02-13 18:39:58,954 - Epoch: [186][ 210/ 1207] Overall Loss 0.203081 Objective Loss 0.203081 LR 0.000125 Time 0.022752 -2023-02-13 18:39:59,144 - Epoch: [186][ 220/ 1207] Overall Loss 0.202037 Objective Loss 0.202037 LR 0.000125 Time 0.022579 -2023-02-13 18:39:59,334 - Epoch: [186][ 230/ 1207] Overall Loss 0.201599 Objective Loss 0.201599 LR 0.000125 Time 0.022420 -2023-02-13 18:39:59,524 - Epoch: [186][ 240/ 1207] Overall Loss 0.201856 Objective Loss 0.201856 LR 0.000125 Time 0.022278 -2023-02-13 18:39:59,714 - Epoch: [186][ 250/ 1207] Overall Loss 0.201688 Objective Loss 0.201688 LR 0.000125 Time 0.022143 -2023-02-13 18:39:59,904 - Epoch: [186][ 260/ 1207] Overall Loss 0.200494 Objective Loss 0.200494 LR 0.000125 Time 0.022021 -2023-02-13 18:40:00,093 - Epoch: [186][ 270/ 1207] Overall Loss 0.200190 Objective Loss 0.200190 LR 0.000125 Time 0.021907 -2023-02-13 18:40:00,283 - Epoch: [186][ 280/ 1207] Overall Loss 0.200299 Objective Loss 0.200299 LR 0.000125 Time 0.021802 -2023-02-13 18:40:00,473 - Epoch: [186][ 290/ 1207] Overall Loss 0.200733 Objective Loss 0.200733 LR 0.000125 Time 0.021701 -2023-02-13 18:40:00,663 - Epoch: [186][ 300/ 1207] Overall Loss 0.200281 Objective Loss 0.200281 LR 0.000125 Time 0.021610 -2023-02-13 18:40:00,853 - Epoch: [186][ 310/ 1207] Overall Loss 0.199888 Objective Loss 0.199888 LR 0.000125 Time 0.021524 -2023-02-13 18:40:01,044 - Epoch: [186][ 320/ 1207] Overall Loss 0.199821 Objective Loss 0.199821 LR 0.000125 Time 0.021448 -2023-02-13 18:40:01,234 - Epoch: [186][ 330/ 1207] Overall Loss 0.199860 Objective Loss 0.199860 LR 0.000125 Time 0.021374 -2023-02-13 18:40:01,425 - Epoch: [186][ 340/ 1207] Overall Loss 0.200113 Objective Loss 0.200113 LR 0.000125 Time 0.021305 -2023-02-13 18:40:01,614 - Epoch: [186][ 350/ 1207] Overall Loss 0.200080 Objective Loss 0.200080 LR 0.000125 Time 0.021236 -2023-02-13 18:40:01,804 - Epoch: [186][ 360/ 1207] Overall Loss 0.200584 Objective Loss 0.200584 LR 0.000125 Time 0.021172 -2023-02-13 18:40:01,994 - Epoch: [186][ 370/ 1207] Overall Loss 0.200707 Objective Loss 0.200707 LR 0.000125 Time 0.021114 -2023-02-13 18:40:02,184 - Epoch: [186][ 380/ 1207] Overall Loss 0.200565 Objective Loss 0.200565 LR 0.000125 Time 0.021057 -2023-02-13 18:40:02,374 - Epoch: [186][ 390/ 1207] Overall Loss 0.200291 Objective Loss 0.200291 LR 0.000125 Time 0.021002 -2023-02-13 18:40:02,564 - Epoch: [186][ 400/ 1207] Overall Loss 0.200958 Objective Loss 0.200958 LR 0.000125 Time 0.020952 -2023-02-13 18:40:02,754 - Epoch: [186][ 410/ 1207] Overall Loss 0.201434 Objective Loss 0.201434 LR 0.000125 Time 0.020902 -2023-02-13 18:40:02,944 - Epoch: [186][ 420/ 1207] Overall Loss 0.201544 Objective Loss 0.201544 LR 0.000125 Time 0.020856 -2023-02-13 18:40:03,134 - Epoch: [186][ 430/ 1207] Overall Loss 0.201097 Objective Loss 0.201097 LR 0.000125 Time 0.020811 -2023-02-13 18:40:03,323 - Epoch: [186][ 440/ 1207] Overall Loss 0.201264 Objective Loss 0.201264 LR 0.000125 Time 0.020769 -2023-02-13 18:40:03,515 - Epoch: [186][ 450/ 1207] Overall Loss 0.201442 Objective Loss 0.201442 LR 0.000125 Time 0.020731 -2023-02-13 18:40:03,704 - Epoch: [186][ 460/ 1207] Overall Loss 0.201193 Objective Loss 0.201193 LR 0.000125 Time 0.020693 -2023-02-13 18:40:03,894 - Epoch: [186][ 470/ 1207] Overall Loss 0.201237 Objective Loss 0.201237 LR 0.000125 Time 0.020655 -2023-02-13 18:40:04,085 - Epoch: [186][ 480/ 1207] Overall Loss 0.201091 Objective Loss 0.201091 LR 0.000125 Time 0.020621 -2023-02-13 18:40:04,274 - Epoch: [186][ 490/ 1207] Overall Loss 0.201242 Objective Loss 0.201242 LR 0.000125 Time 0.020587 -2023-02-13 18:40:04,465 - Epoch: [186][ 500/ 1207] Overall Loss 0.200901 Objective Loss 0.200901 LR 0.000125 Time 0.020555 -2023-02-13 18:40:04,655 - Epoch: [186][ 510/ 1207] Overall Loss 0.201104 Objective Loss 0.201104 LR 0.000125 Time 0.020524 -2023-02-13 18:40:04,845 - Epoch: [186][ 520/ 1207] Overall Loss 0.200882 Objective Loss 0.200882 LR 0.000125 Time 0.020495 -2023-02-13 18:40:05,036 - Epoch: [186][ 530/ 1207] Overall Loss 0.201142 Objective Loss 0.201142 LR 0.000125 Time 0.020467 -2023-02-13 18:40:05,225 - Epoch: [186][ 540/ 1207] Overall Loss 0.201456 Objective Loss 0.201456 LR 0.000125 Time 0.020438 -2023-02-13 18:40:05,415 - Epoch: [186][ 550/ 1207] Overall Loss 0.201134 Objective Loss 0.201134 LR 0.000125 Time 0.020411 -2023-02-13 18:40:05,606 - Epoch: [186][ 560/ 1207] Overall Loss 0.201441 Objective Loss 0.201441 LR 0.000125 Time 0.020386 -2023-02-13 18:40:05,796 - Epoch: [186][ 570/ 1207] Overall Loss 0.201251 Objective Loss 0.201251 LR 0.000125 Time 0.020362 -2023-02-13 18:40:05,986 - Epoch: [186][ 580/ 1207] Overall Loss 0.201398 Objective Loss 0.201398 LR 0.000125 Time 0.020338 -2023-02-13 18:40:06,177 - Epoch: [186][ 590/ 1207] Overall Loss 0.201435 Objective Loss 0.201435 LR 0.000125 Time 0.020315 -2023-02-13 18:40:06,367 - Epoch: [186][ 600/ 1207] Overall Loss 0.201200 Objective Loss 0.201200 LR 0.000125 Time 0.020294 -2023-02-13 18:40:06,558 - Epoch: [186][ 610/ 1207] Overall Loss 0.200860 Objective Loss 0.200860 LR 0.000125 Time 0.020273 -2023-02-13 18:40:06,748 - Epoch: [186][ 620/ 1207] Overall Loss 0.201004 Objective Loss 0.201004 LR 0.000125 Time 0.020253 -2023-02-13 18:40:06,939 - Epoch: [186][ 630/ 1207] Overall Loss 0.201226 Objective Loss 0.201226 LR 0.000125 Time 0.020233 -2023-02-13 18:40:07,129 - Epoch: [186][ 640/ 1207] Overall Loss 0.200986 Objective Loss 0.200986 LR 0.000125 Time 0.020214 -2023-02-13 18:40:07,320 - Epoch: [186][ 650/ 1207] Overall Loss 0.200661 Objective Loss 0.200661 LR 0.000125 Time 0.020196 -2023-02-13 18:40:07,511 - Epoch: [186][ 660/ 1207] Overall Loss 0.200982 Objective Loss 0.200982 LR 0.000125 Time 0.020179 -2023-02-13 18:40:07,702 - Epoch: [186][ 670/ 1207] Overall Loss 0.201077 Objective Loss 0.201077 LR 0.000125 Time 0.020162 -2023-02-13 18:40:07,892 - Epoch: [186][ 680/ 1207] Overall Loss 0.200993 Objective Loss 0.200993 LR 0.000125 Time 0.020144 -2023-02-13 18:40:08,082 - Epoch: [186][ 690/ 1207] Overall Loss 0.201132 Objective Loss 0.201132 LR 0.000125 Time 0.020127 -2023-02-13 18:40:08,272 - Epoch: [186][ 700/ 1207] Overall Loss 0.201326 Objective Loss 0.201326 LR 0.000125 Time 0.020111 -2023-02-13 18:40:08,463 - Epoch: [186][ 710/ 1207] Overall Loss 0.201180 Objective Loss 0.201180 LR 0.000125 Time 0.020095 -2023-02-13 18:40:08,653 - Epoch: [186][ 720/ 1207] Overall Loss 0.201114 Objective Loss 0.201114 LR 0.000125 Time 0.020080 -2023-02-13 18:40:08,845 - Epoch: [186][ 730/ 1207] Overall Loss 0.201107 Objective Loss 0.201107 LR 0.000125 Time 0.020067 -2023-02-13 18:40:09,035 - Epoch: [186][ 740/ 1207] Overall Loss 0.201121 Objective Loss 0.201121 LR 0.000125 Time 0.020053 -2023-02-13 18:40:09,226 - Epoch: [186][ 750/ 1207] Overall Loss 0.200960 Objective Loss 0.200960 LR 0.000125 Time 0.020039 -2023-02-13 18:40:09,417 - Epoch: [186][ 760/ 1207] Overall Loss 0.200597 Objective Loss 0.200597 LR 0.000125 Time 0.020026 -2023-02-13 18:40:09,608 - Epoch: [186][ 770/ 1207] Overall Loss 0.200504 Objective Loss 0.200504 LR 0.000125 Time 0.020013 -2023-02-13 18:40:09,799 - Epoch: [186][ 780/ 1207] Overall Loss 0.200303 Objective Loss 0.200303 LR 0.000125 Time 0.020001 -2023-02-13 18:40:09,989 - Epoch: [186][ 790/ 1207] Overall Loss 0.200517 Objective Loss 0.200517 LR 0.000125 Time 0.019988 -2023-02-13 18:40:10,180 - Epoch: [186][ 800/ 1207] Overall Loss 0.200538 Objective Loss 0.200538 LR 0.000125 Time 0.019977 -2023-02-13 18:40:10,370 - Epoch: [186][ 810/ 1207] Overall Loss 0.200699 Objective Loss 0.200699 LR 0.000125 Time 0.019965 -2023-02-13 18:40:10,561 - Epoch: [186][ 820/ 1207] Overall Loss 0.200899 Objective Loss 0.200899 LR 0.000125 Time 0.019953 -2023-02-13 18:40:10,752 - Epoch: [186][ 830/ 1207] Overall Loss 0.201211 Objective Loss 0.201211 LR 0.000125 Time 0.019942 -2023-02-13 18:40:10,943 - Epoch: [186][ 840/ 1207] Overall Loss 0.201618 Objective Loss 0.201618 LR 0.000125 Time 0.019932 -2023-02-13 18:40:11,133 - Epoch: [186][ 850/ 1207] Overall Loss 0.201535 Objective Loss 0.201535 LR 0.000125 Time 0.019921 -2023-02-13 18:40:11,323 - Epoch: [186][ 860/ 1207] Overall Loss 0.201717 Objective Loss 0.201717 LR 0.000125 Time 0.019910 -2023-02-13 18:40:11,514 - Epoch: [186][ 870/ 1207] Overall Loss 0.201716 Objective Loss 0.201716 LR 0.000125 Time 0.019900 -2023-02-13 18:40:11,709 - Epoch: [186][ 880/ 1207] Overall Loss 0.201646 Objective Loss 0.201646 LR 0.000125 Time 0.019895 -2023-02-13 18:40:11,903 - Epoch: [186][ 890/ 1207] Overall Loss 0.201824 Objective Loss 0.201824 LR 0.000125 Time 0.019889 -2023-02-13 18:40:12,101 - Epoch: [186][ 900/ 1207] Overall Loss 0.201674 Objective Loss 0.201674 LR 0.000125 Time 0.019888 -2023-02-13 18:40:12,295 - Epoch: [186][ 910/ 1207] Overall Loss 0.201780 Objective Loss 0.201780 LR 0.000125 Time 0.019882 -2023-02-13 18:40:12,493 - Epoch: [186][ 920/ 1207] Overall Loss 0.201860 Objective Loss 0.201860 LR 0.000125 Time 0.019880 -2023-02-13 18:40:12,687 - Epoch: [186][ 930/ 1207] Overall Loss 0.202030 Objective Loss 0.202030 LR 0.000125 Time 0.019875 -2023-02-13 18:40:12,884 - Epoch: [186][ 940/ 1207] Overall Loss 0.202006 Objective Loss 0.202006 LR 0.000125 Time 0.019872 -2023-02-13 18:40:13,078 - Epoch: [186][ 950/ 1207] Overall Loss 0.201988 Objective Loss 0.201988 LR 0.000125 Time 0.019867 -2023-02-13 18:40:13,276 - Epoch: [186][ 960/ 1207] Overall Loss 0.202193 Objective Loss 0.202193 LR 0.000125 Time 0.019865 -2023-02-13 18:40:13,470 - Epoch: [186][ 970/ 1207] Overall Loss 0.202369 Objective Loss 0.202369 LR 0.000125 Time 0.019861 -2023-02-13 18:40:13,667 - Epoch: [186][ 980/ 1207] Overall Loss 0.202243 Objective Loss 0.202243 LR 0.000125 Time 0.019859 -2023-02-13 18:40:13,862 - Epoch: [186][ 990/ 1207] Overall Loss 0.202284 Objective Loss 0.202284 LR 0.000125 Time 0.019854 -2023-02-13 18:40:14,058 - Epoch: [186][ 1000/ 1207] Overall Loss 0.202279 Objective Loss 0.202279 LR 0.000125 Time 0.019852 -2023-02-13 18:40:14,253 - Epoch: [186][ 1010/ 1207] Overall Loss 0.202410 Objective Loss 0.202410 LR 0.000125 Time 0.019848 -2023-02-13 18:40:14,450 - Epoch: [186][ 1020/ 1207] Overall Loss 0.202460 Objective Loss 0.202460 LR 0.000125 Time 0.019846 -2023-02-13 18:40:14,645 - Epoch: [186][ 1030/ 1207] Overall Loss 0.202404 Objective Loss 0.202404 LR 0.000125 Time 0.019842 -2023-02-13 18:40:14,841 - Epoch: [186][ 1040/ 1207] Overall Loss 0.202276 Objective Loss 0.202276 LR 0.000125 Time 0.019840 -2023-02-13 18:40:15,035 - Epoch: [186][ 1050/ 1207] Overall Loss 0.202344 Objective Loss 0.202344 LR 0.000125 Time 0.019835 -2023-02-13 18:40:15,231 - Epoch: [186][ 1060/ 1207] Overall Loss 0.202278 Objective Loss 0.202278 LR 0.000125 Time 0.019833 -2023-02-13 18:40:15,425 - Epoch: [186][ 1070/ 1207] Overall Loss 0.202214 Objective Loss 0.202214 LR 0.000125 Time 0.019829 -2023-02-13 18:40:15,623 - Epoch: [186][ 1080/ 1207] Overall Loss 0.202407 Objective Loss 0.202407 LR 0.000125 Time 0.019828 -2023-02-13 18:40:15,818 - Epoch: [186][ 1090/ 1207] Overall Loss 0.202358 Objective Loss 0.202358 LR 0.000125 Time 0.019824 -2023-02-13 18:40:16,015 - Epoch: [186][ 1100/ 1207] Overall Loss 0.202473 Objective Loss 0.202473 LR 0.000125 Time 0.019823 -2023-02-13 18:40:16,209 - Epoch: [186][ 1110/ 1207] Overall Loss 0.202521 Objective Loss 0.202521 LR 0.000125 Time 0.019819 -2023-02-13 18:40:16,406 - Epoch: [186][ 1120/ 1207] Overall Loss 0.202538 Objective Loss 0.202538 LR 0.000125 Time 0.019817 -2023-02-13 18:40:16,600 - Epoch: [186][ 1130/ 1207] Overall Loss 0.202877 Objective Loss 0.202877 LR 0.000125 Time 0.019813 -2023-02-13 18:40:16,798 - Epoch: [186][ 1140/ 1207] Overall Loss 0.202953 Objective Loss 0.202953 LR 0.000125 Time 0.019812 -2023-02-13 18:40:16,993 - Epoch: [186][ 1150/ 1207] Overall Loss 0.202822 Objective Loss 0.202822 LR 0.000125 Time 0.019809 -2023-02-13 18:40:17,189 - Epoch: [186][ 1160/ 1207] Overall Loss 0.202931 Objective Loss 0.202931 LR 0.000125 Time 0.019808 -2023-02-13 18:40:17,383 - Epoch: [186][ 1170/ 1207] Overall Loss 0.202970 Objective Loss 0.202970 LR 0.000125 Time 0.019804 -2023-02-13 18:40:17,581 - Epoch: [186][ 1180/ 1207] Overall Loss 0.202928 Objective Loss 0.202928 LR 0.000125 Time 0.019803 -2023-02-13 18:40:17,775 - Epoch: [186][ 1190/ 1207] Overall Loss 0.202948 Objective Loss 0.202948 LR 0.000125 Time 0.019800 -2023-02-13 18:40:18,025 - Epoch: [186][ 1200/ 1207] Overall Loss 0.203200 Objective Loss 0.203200 LR 0.000125 Time 0.019843 -2023-02-13 18:40:18,140 - Epoch: [186][ 1207/ 1207] Overall Loss 0.203294 Objective Loss 0.203294 Top1 87.804878 Top5 99.085366 LR 0.000125 Time 0.019823 -2023-02-13 18:40:18,212 - --- validate (epoch=186)----------- -2023-02-13 18:40:18,213 - 34311 samples (256 per mini-batch) -2023-02-13 18:40:18,613 - Epoch: [186][ 10/ 135] Loss 0.271115 Top1 86.953125 Top5 97.656250 -2023-02-13 18:40:18,756 - Epoch: [186][ 20/ 135] Loss 0.286245 Top1 86.015625 Top5 97.617188 -2023-02-13 18:40:18,881 - Epoch: [186][ 30/ 135] Loss 0.269601 Top1 86.562500 Top5 97.838542 -2023-02-13 18:40:19,017 - Epoch: [186][ 40/ 135] Loss 0.279030 Top1 86.171875 Top5 97.792969 -2023-02-13 18:40:19,160 - Epoch: [186][ 50/ 135] Loss 0.279424 Top1 85.953125 Top5 97.851562 -2023-02-13 18:40:19,289 - Epoch: [186][ 60/ 135] Loss 0.279671 Top1 85.852865 Top5 97.819010 -2023-02-13 18:40:19,424 - Epoch: [186][ 70/ 135] Loss 0.284651 Top1 85.675223 Top5 97.829241 -2023-02-13 18:40:19,562 - Epoch: [186][ 80/ 135] Loss 0.289693 Top1 85.625000 Top5 97.861328 -2023-02-13 18:40:19,702 - Epoch: [186][ 90/ 135] Loss 0.291205 Top1 85.633681 Top5 97.864583 -2023-02-13 18:40:19,843 - Epoch: [186][ 100/ 135] Loss 0.290987 Top1 85.652344 Top5 97.808594 -2023-02-13 18:40:19,983 - Epoch: [186][ 110/ 135] Loss 0.291202 Top1 85.706676 Top5 97.805398 -2023-02-13 18:40:20,124 - Epoch: [186][ 120/ 135] Loss 0.291322 Top1 85.641276 Top5 97.848307 -2023-02-13 18:40:20,260 - Epoch: [186][ 130/ 135] Loss 0.290770 Top1 85.640024 Top5 97.869591 -2023-02-13 18:40:20,304 - Epoch: [186][ 135/ 135] Loss 0.292391 Top1 85.599370 Top5 97.860744 -2023-02-13 18:40:20,387 - ==> Top1: 85.599 Top5: 97.861 Loss: 0.292 - -2023-02-13 18:40:20,388 - ==> Confusion: -[[ 865 4 9 1 5 1 0 0 4 46 0 4 1 2 6 4 3 1 0 4 7] - [ 2 962 2 1 7 20 1 12 3 0 0 2 1 1 0 1 4 1 3 3 7] - [ 6 3 968 10 3 1 11 13 0 1 4 2 1 4 3 6 4 2 5 2 9] - [ 2 0 19 910 1 3 0 1 3 3 13 0 7 1 24 2 2 4 13 1 7] - [ 9 8 1 0 997 11 1 2 1 2 0 2 3 3 9 6 4 1 1 3 2] - [ 1 13 0 4 3 974 2 15 2 4 1 9 5 14 0 5 7 1 1 1 8] - [ 2 2 20 2 0 7 1032 5 0 1 1 0 2 1 0 3 4 4 1 6 6] - [ 2 7 11 2 2 29 2 926 0 2 0 6 4 1 0 0 1 1 18 8 2] - [ 12 3 1 1 1 0 0 1 912 34 4 2 0 6 22 3 0 0 4 1 2] - [ 80 1 3 0 8 0 0 2 33 859 1 2 0 12 4 0 1 1 0 0 5] - [ 1 0 5 5 3 3 2 5 17 1 988 1 1 6 2 0 2 1 3 1 4] - [ 2 3 2 0 1 13 1 4 1 4 0 918 21 7 0 8 2 9 2 4 3] - [ 0 0 1 6 0 2 0 0 2 0 0 20 891 1 2 8 5 9 2 1 9] - [ 3 2 1 1 7 8 1 3 15 22 8 3 3 923 6 4 4 4 1 0 5] - [ 3 0 0 14 3 5 0 1 18 8 3 0 2 0 1021 0 1 4 5 0 4] - [ 3 4 8 0 3 1 1 1 1 0 0 7 5 2 0 974 10 10 1 8 7] - [ 1 8 1 2 6 2 0 0 2 0 0 1 1 2 5 12 1000 2 2 3 11] - [ 4 2 0 3 0 2 1 0 0 2 2 8 11 1 1 14 0 994 0 1 5] - [ 2 4 3 8 1 1 0 17 2 0 5 1 3 0 12 1 1 4 1019 1 1] - [ 0 3 1 1 2 5 7 9 0 0 0 18 3 0 2 8 5 2 0 1075 7] - [ 167 225 236 114 128 194 67 140 109 78 162 96 289 250 198 87 235 101 178 218 10162]] - -2023-02-13 18:40:20,390 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:40:20,390 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:40:20,396 - - -2023-02-13 18:40:20,396 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:40:21,293 - Epoch: [187][ 10/ 1207] Overall Loss 0.195052 Objective Loss 0.195052 LR 0.000125 Time 0.089639 -2023-02-13 18:40:21,494 - Epoch: [187][ 20/ 1207] Overall Loss 0.194297 Objective Loss 0.194297 LR 0.000125 Time 0.054833 -2023-02-13 18:40:21,687 - Epoch: [187][ 30/ 1207] Overall Loss 0.194342 Objective Loss 0.194342 LR 0.000125 Time 0.042990 -2023-02-13 18:40:21,882 - Epoch: [187][ 40/ 1207] Overall Loss 0.196241 Objective Loss 0.196241 LR 0.000125 Time 0.037114 -2023-02-13 18:40:22,075 - Epoch: [187][ 50/ 1207] Overall Loss 0.197081 Objective Loss 0.197081 LR 0.000125 Time 0.033550 -2023-02-13 18:40:22,271 - Epoch: [187][ 60/ 1207] Overall Loss 0.196999 Objective Loss 0.196999 LR 0.000125 Time 0.031210 -2023-02-13 18:40:22,464 - Epoch: [187][ 70/ 1207] Overall Loss 0.195788 Objective Loss 0.195788 LR 0.000125 Time 0.029506 -2023-02-13 18:40:22,660 - Epoch: [187][ 80/ 1207] Overall Loss 0.196161 Objective Loss 0.196161 LR 0.000125 Time 0.028259 -2023-02-13 18:40:22,853 - Epoch: [187][ 90/ 1207] Overall Loss 0.196705 Objective Loss 0.196705 LR 0.000125 Time 0.027256 -2023-02-13 18:40:23,051 - Epoch: [187][ 100/ 1207] Overall Loss 0.196617 Objective Loss 0.196617 LR 0.000125 Time 0.026509 -2023-02-13 18:40:23,245 - Epoch: [187][ 110/ 1207] Overall Loss 0.197417 Objective Loss 0.197417 LR 0.000125 Time 0.025864 -2023-02-13 18:40:23,440 - Epoch: [187][ 120/ 1207] Overall Loss 0.197647 Objective Loss 0.197647 LR 0.000125 Time 0.025325 -2023-02-13 18:40:23,630 - Epoch: [187][ 130/ 1207] Overall Loss 0.197272 Objective Loss 0.197272 LR 0.000125 Time 0.024838 -2023-02-13 18:40:23,819 - Epoch: [187][ 140/ 1207] Overall Loss 0.197647 Objective Loss 0.197647 LR 0.000125 Time 0.024415 -2023-02-13 18:40:24,009 - Epoch: [187][ 150/ 1207] Overall Loss 0.196468 Objective Loss 0.196468 LR 0.000125 Time 0.024049 -2023-02-13 18:40:24,199 - Epoch: [187][ 160/ 1207] Overall Loss 0.196210 Objective Loss 0.196210 LR 0.000125 Time 0.023729 -2023-02-13 18:40:24,388 - Epoch: [187][ 170/ 1207] Overall Loss 0.196071 Objective Loss 0.196071 LR 0.000125 Time 0.023444 -2023-02-13 18:40:24,578 - Epoch: [187][ 180/ 1207] Overall Loss 0.195246 Objective Loss 0.195246 LR 0.000125 Time 0.023193 -2023-02-13 18:40:24,767 - Epoch: [187][ 190/ 1207] Overall Loss 0.196278 Objective Loss 0.196278 LR 0.000125 Time 0.022968 -2023-02-13 18:40:24,957 - Epoch: [187][ 200/ 1207] Overall Loss 0.196937 Objective Loss 0.196937 LR 0.000125 Time 0.022766 -2023-02-13 18:40:25,147 - Epoch: [187][ 210/ 1207] Overall Loss 0.197845 Objective Loss 0.197845 LR 0.000125 Time 0.022587 -2023-02-13 18:40:25,337 - Epoch: [187][ 220/ 1207] Overall Loss 0.197455 Objective Loss 0.197455 LR 0.000125 Time 0.022420 -2023-02-13 18:40:25,527 - Epoch: [187][ 230/ 1207] Overall Loss 0.197752 Objective Loss 0.197752 LR 0.000125 Time 0.022271 -2023-02-13 18:40:25,717 - Epoch: [187][ 240/ 1207] Overall Loss 0.199064 Objective Loss 0.199064 LR 0.000125 Time 0.022131 -2023-02-13 18:40:25,907 - Epoch: [187][ 250/ 1207] Overall Loss 0.199242 Objective Loss 0.199242 LR 0.000125 Time 0.022005 -2023-02-13 18:40:26,098 - Epoch: [187][ 260/ 1207] Overall Loss 0.199107 Objective Loss 0.199107 LR 0.000125 Time 0.021891 -2023-02-13 18:40:26,287 - Epoch: [187][ 270/ 1207] Overall Loss 0.199334 Objective Loss 0.199334 LR 0.000125 Time 0.021781 -2023-02-13 18:40:26,477 - Epoch: [187][ 280/ 1207] Overall Loss 0.199998 Objective Loss 0.199998 LR 0.000125 Time 0.021680 -2023-02-13 18:40:26,667 - Epoch: [187][ 290/ 1207] Overall Loss 0.200788 Objective Loss 0.200788 LR 0.000125 Time 0.021585 -2023-02-13 18:40:26,857 - Epoch: [187][ 300/ 1207] Overall Loss 0.200795 Objective Loss 0.200795 LR 0.000125 Time 0.021500 -2023-02-13 18:40:27,047 - Epoch: [187][ 310/ 1207] Overall Loss 0.200514 Objective Loss 0.200514 LR 0.000125 Time 0.021418 -2023-02-13 18:40:27,237 - Epoch: [187][ 320/ 1207] Overall Loss 0.200533 Objective Loss 0.200533 LR 0.000125 Time 0.021340 -2023-02-13 18:40:27,426 - Epoch: [187][ 330/ 1207] Overall Loss 0.200084 Objective Loss 0.200084 LR 0.000125 Time 0.021266 -2023-02-13 18:40:27,617 - Epoch: [187][ 340/ 1207] Overall Loss 0.199829 Objective Loss 0.199829 LR 0.000125 Time 0.021200 -2023-02-13 18:40:27,808 - Epoch: [187][ 350/ 1207] Overall Loss 0.199613 Objective Loss 0.199613 LR 0.000125 Time 0.021138 -2023-02-13 18:40:27,997 - Epoch: [187][ 360/ 1207] Overall Loss 0.199383 Objective Loss 0.199383 LR 0.000125 Time 0.021078 -2023-02-13 18:40:28,187 - Epoch: [187][ 370/ 1207] Overall Loss 0.199017 Objective Loss 0.199017 LR 0.000125 Time 0.021020 -2023-02-13 18:40:28,377 - Epoch: [187][ 380/ 1207] Overall Loss 0.198526 Objective Loss 0.198526 LR 0.000125 Time 0.020965 -2023-02-13 18:40:28,567 - Epoch: [187][ 390/ 1207] Overall Loss 0.199340 Objective Loss 0.199340 LR 0.000125 Time 0.020914 -2023-02-13 18:40:28,757 - Epoch: [187][ 400/ 1207] Overall Loss 0.199734 Objective Loss 0.199734 LR 0.000125 Time 0.020866 -2023-02-13 18:40:28,947 - Epoch: [187][ 410/ 1207] Overall Loss 0.199560 Objective Loss 0.199560 LR 0.000125 Time 0.020818 -2023-02-13 18:40:29,136 - Epoch: [187][ 420/ 1207] Overall Loss 0.200048 Objective Loss 0.200048 LR 0.000125 Time 0.020773 -2023-02-13 18:40:29,326 - Epoch: [187][ 430/ 1207] Overall Loss 0.200601 Objective Loss 0.200601 LR 0.000125 Time 0.020730 -2023-02-13 18:40:29,517 - Epoch: [187][ 440/ 1207] Overall Loss 0.201118 Objective Loss 0.201118 LR 0.000125 Time 0.020691 -2023-02-13 18:40:29,707 - Epoch: [187][ 450/ 1207] Overall Loss 0.200564 Objective Loss 0.200564 LR 0.000125 Time 0.020653 -2023-02-13 18:40:29,896 - Epoch: [187][ 460/ 1207] Overall Loss 0.200408 Objective Loss 0.200408 LR 0.000125 Time 0.020615 -2023-02-13 18:40:30,086 - Epoch: [187][ 470/ 1207] Overall Loss 0.200617 Objective Loss 0.200617 LR 0.000125 Time 0.020580 -2023-02-13 18:40:30,276 - Epoch: [187][ 480/ 1207] Overall Loss 0.200501 Objective Loss 0.200501 LR 0.000125 Time 0.020546 -2023-02-13 18:40:30,466 - Epoch: [187][ 490/ 1207] Overall Loss 0.200530 Objective Loss 0.200530 LR 0.000125 Time 0.020514 -2023-02-13 18:40:30,657 - Epoch: [187][ 500/ 1207] Overall Loss 0.200732 Objective Loss 0.200732 LR 0.000125 Time 0.020484 -2023-02-13 18:40:30,847 - Epoch: [187][ 510/ 1207] Overall Loss 0.200669 Objective Loss 0.200669 LR 0.000125 Time 0.020455 -2023-02-13 18:40:31,038 - Epoch: [187][ 520/ 1207] Overall Loss 0.200874 Objective Loss 0.200874 LR 0.000125 Time 0.020428 -2023-02-13 18:40:31,227 - Epoch: [187][ 530/ 1207] Overall Loss 0.200721 Objective Loss 0.200721 LR 0.000125 Time 0.020399 -2023-02-13 18:40:31,417 - Epoch: [187][ 540/ 1207] Overall Loss 0.201088 Objective Loss 0.201088 LR 0.000125 Time 0.020372 -2023-02-13 18:40:31,607 - Epoch: [187][ 550/ 1207] Overall Loss 0.201067 Objective Loss 0.201067 LR 0.000125 Time 0.020346 -2023-02-13 18:40:31,797 - Epoch: [187][ 560/ 1207] Overall Loss 0.201832 Objective Loss 0.201832 LR 0.000125 Time 0.020321 -2023-02-13 18:40:31,987 - Epoch: [187][ 570/ 1207] Overall Loss 0.201856 Objective Loss 0.201856 LR 0.000125 Time 0.020298 -2023-02-13 18:40:32,177 - Epoch: [187][ 580/ 1207] Overall Loss 0.202013 Objective Loss 0.202013 LR 0.000125 Time 0.020275 -2023-02-13 18:40:32,367 - Epoch: [187][ 590/ 1207] Overall Loss 0.201984 Objective Loss 0.201984 LR 0.000125 Time 0.020253 -2023-02-13 18:40:32,558 - Epoch: [187][ 600/ 1207] Overall Loss 0.202111 Objective Loss 0.202111 LR 0.000125 Time 0.020232 -2023-02-13 18:40:32,748 - Epoch: [187][ 610/ 1207] Overall Loss 0.202457 Objective Loss 0.202457 LR 0.000125 Time 0.020211 -2023-02-13 18:40:32,938 - Epoch: [187][ 620/ 1207] Overall Loss 0.202063 Objective Loss 0.202063 LR 0.000125 Time 0.020192 -2023-02-13 18:40:33,128 - Epoch: [187][ 630/ 1207] Overall Loss 0.202195 Objective Loss 0.202195 LR 0.000125 Time 0.020173 -2023-02-13 18:40:33,318 - Epoch: [187][ 640/ 1207] Overall Loss 0.202039 Objective Loss 0.202039 LR 0.000125 Time 0.020154 -2023-02-13 18:40:33,509 - Epoch: [187][ 650/ 1207] Overall Loss 0.202240 Objective Loss 0.202240 LR 0.000125 Time 0.020136 -2023-02-13 18:40:33,699 - Epoch: [187][ 660/ 1207] Overall Loss 0.201997 Objective Loss 0.201997 LR 0.000125 Time 0.020119 -2023-02-13 18:40:33,890 - Epoch: [187][ 670/ 1207] Overall Loss 0.202594 Objective Loss 0.202594 LR 0.000125 Time 0.020102 -2023-02-13 18:40:34,081 - Epoch: [187][ 680/ 1207] Overall Loss 0.202899 Objective Loss 0.202899 LR 0.000125 Time 0.020087 -2023-02-13 18:40:34,271 - Epoch: [187][ 690/ 1207] Overall Loss 0.202762 Objective Loss 0.202762 LR 0.000125 Time 0.020071 -2023-02-13 18:40:34,461 - Epoch: [187][ 700/ 1207] Overall Loss 0.202801 Objective Loss 0.202801 LR 0.000125 Time 0.020056 -2023-02-13 18:40:34,651 - Epoch: [187][ 710/ 1207] Overall Loss 0.203023 Objective Loss 0.203023 LR 0.000125 Time 0.020041 -2023-02-13 18:40:34,841 - Epoch: [187][ 720/ 1207] Overall Loss 0.202770 Objective Loss 0.202770 LR 0.000125 Time 0.020026 -2023-02-13 18:40:35,031 - Epoch: [187][ 730/ 1207] Overall Loss 0.202727 Objective Loss 0.202727 LR 0.000125 Time 0.020011 -2023-02-13 18:40:35,221 - Epoch: [187][ 740/ 1207] Overall Loss 0.202691 Objective Loss 0.202691 LR 0.000125 Time 0.019997 -2023-02-13 18:40:35,411 - Epoch: [187][ 750/ 1207] Overall Loss 0.202244 Objective Loss 0.202244 LR 0.000125 Time 0.019983 -2023-02-13 18:40:35,602 - Epoch: [187][ 760/ 1207] Overall Loss 0.202394 Objective Loss 0.202394 LR 0.000125 Time 0.019970 -2023-02-13 18:40:35,792 - Epoch: [187][ 770/ 1207] Overall Loss 0.202503 Objective Loss 0.202503 LR 0.000125 Time 0.019957 -2023-02-13 18:40:35,983 - Epoch: [187][ 780/ 1207] Overall Loss 0.202584 Objective Loss 0.202584 LR 0.000125 Time 0.019946 -2023-02-13 18:40:36,173 - Epoch: [187][ 790/ 1207] Overall Loss 0.202654 Objective Loss 0.202654 LR 0.000125 Time 0.019933 -2023-02-13 18:40:36,362 - Epoch: [187][ 800/ 1207] Overall Loss 0.202603 Objective Loss 0.202603 LR 0.000125 Time 0.019920 -2023-02-13 18:40:36,553 - Epoch: [187][ 810/ 1207] Overall Loss 0.202784 Objective Loss 0.202784 LR 0.000125 Time 0.019910 -2023-02-13 18:40:36,744 - Epoch: [187][ 820/ 1207] Overall Loss 0.202711 Objective Loss 0.202711 LR 0.000125 Time 0.019899 -2023-02-13 18:40:36,934 - Epoch: [187][ 830/ 1207] Overall Loss 0.202717 Objective Loss 0.202717 LR 0.000125 Time 0.019889 -2023-02-13 18:40:37,125 - Epoch: [187][ 840/ 1207] Overall Loss 0.202614 Objective Loss 0.202614 LR 0.000125 Time 0.019878 -2023-02-13 18:40:37,315 - Epoch: [187][ 850/ 1207] Overall Loss 0.202640 Objective Loss 0.202640 LR 0.000125 Time 0.019867 -2023-02-13 18:40:37,505 - Epoch: [187][ 860/ 1207] Overall Loss 0.202666 Objective Loss 0.202666 LR 0.000125 Time 0.019857 -2023-02-13 18:40:37,695 - Epoch: [187][ 870/ 1207] Overall Loss 0.202748 Objective Loss 0.202748 LR 0.000125 Time 0.019847 -2023-02-13 18:40:37,885 - Epoch: [187][ 880/ 1207] Overall Loss 0.202998 Objective Loss 0.202998 LR 0.000125 Time 0.019837 -2023-02-13 18:40:38,076 - Epoch: [187][ 890/ 1207] Overall Loss 0.203378 Objective Loss 0.203378 LR 0.000125 Time 0.019827 -2023-02-13 18:40:38,266 - Epoch: [187][ 900/ 1207] Overall Loss 0.203391 Objective Loss 0.203391 LR 0.000125 Time 0.019818 -2023-02-13 18:40:38,456 - Epoch: [187][ 910/ 1207] Overall Loss 0.203575 Objective Loss 0.203575 LR 0.000125 Time 0.019808 -2023-02-13 18:40:38,646 - Epoch: [187][ 920/ 1207] Overall Loss 0.203345 Objective Loss 0.203345 LR 0.000125 Time 0.019800 -2023-02-13 18:40:38,836 - Epoch: [187][ 930/ 1207] Overall Loss 0.203222 Objective Loss 0.203222 LR 0.000125 Time 0.019791 -2023-02-13 18:40:39,026 - Epoch: [187][ 940/ 1207] Overall Loss 0.203335 Objective Loss 0.203335 LR 0.000125 Time 0.019782 -2023-02-13 18:40:39,216 - Epoch: [187][ 950/ 1207] Overall Loss 0.203281 Objective Loss 0.203281 LR 0.000125 Time 0.019773 -2023-02-13 18:40:39,406 - Epoch: [187][ 960/ 1207] Overall Loss 0.203162 Objective Loss 0.203162 LR 0.000125 Time 0.019764 -2023-02-13 18:40:39,596 - Epoch: [187][ 970/ 1207] Overall Loss 0.203049 Objective Loss 0.203049 LR 0.000125 Time 0.019756 -2023-02-13 18:40:39,786 - Epoch: [187][ 980/ 1207] Overall Loss 0.203236 Objective Loss 0.203236 LR 0.000125 Time 0.019748 -2023-02-13 18:40:39,976 - Epoch: [187][ 990/ 1207] Overall Loss 0.203283 Objective Loss 0.203283 LR 0.000125 Time 0.019740 -2023-02-13 18:40:40,165 - Epoch: [187][ 1000/ 1207] Overall Loss 0.203164 Objective Loss 0.203164 LR 0.000125 Time 0.019732 -2023-02-13 18:40:40,355 - Epoch: [187][ 1010/ 1207] Overall Loss 0.203313 Objective Loss 0.203313 LR 0.000125 Time 0.019724 -2023-02-13 18:40:40,545 - Epoch: [187][ 1020/ 1207] Overall Loss 0.203383 Objective Loss 0.203383 LR 0.000125 Time 0.019717 -2023-02-13 18:40:40,735 - Epoch: [187][ 1030/ 1207] Overall Loss 0.203532 Objective Loss 0.203532 LR 0.000125 Time 0.019710 -2023-02-13 18:40:40,927 - Epoch: [187][ 1040/ 1207] Overall Loss 0.203860 Objective Loss 0.203860 LR 0.000125 Time 0.019704 -2023-02-13 18:40:41,118 - Epoch: [187][ 1050/ 1207] Overall Loss 0.203585 Objective Loss 0.203585 LR 0.000125 Time 0.019697 -2023-02-13 18:40:41,308 - Epoch: [187][ 1060/ 1207] Overall Loss 0.203523 Objective Loss 0.203523 LR 0.000125 Time 0.019691 -2023-02-13 18:40:41,499 - Epoch: [187][ 1070/ 1207] Overall Loss 0.203549 Objective Loss 0.203549 LR 0.000125 Time 0.019685 -2023-02-13 18:40:41,689 - Epoch: [187][ 1080/ 1207] Overall Loss 0.203690 Objective Loss 0.203690 LR 0.000125 Time 0.019679 -2023-02-13 18:40:41,880 - Epoch: [187][ 1090/ 1207] Overall Loss 0.203874 Objective Loss 0.203874 LR 0.000125 Time 0.019672 -2023-02-13 18:40:42,070 - Epoch: [187][ 1100/ 1207] Overall Loss 0.203892 Objective Loss 0.203892 LR 0.000125 Time 0.019666 -2023-02-13 18:40:42,260 - Epoch: [187][ 1110/ 1207] Overall Loss 0.203956 Objective Loss 0.203956 LR 0.000125 Time 0.019660 -2023-02-13 18:40:42,450 - Epoch: [187][ 1120/ 1207] Overall Loss 0.203895 Objective Loss 0.203895 LR 0.000125 Time 0.019654 -2023-02-13 18:40:42,641 - Epoch: [187][ 1130/ 1207] Overall Loss 0.203910 Objective Loss 0.203910 LR 0.000125 Time 0.019648 -2023-02-13 18:40:42,831 - Epoch: [187][ 1140/ 1207] Overall Loss 0.203664 Objective Loss 0.203664 LR 0.000125 Time 0.019642 -2023-02-13 18:40:43,021 - Epoch: [187][ 1150/ 1207] Overall Loss 0.203701 Objective Loss 0.203701 LR 0.000125 Time 0.019637 -2023-02-13 18:40:43,212 - Epoch: [187][ 1160/ 1207] Overall Loss 0.203698 Objective Loss 0.203698 LR 0.000125 Time 0.019631 -2023-02-13 18:40:43,402 - Epoch: [187][ 1170/ 1207] Overall Loss 0.203815 Objective Loss 0.203815 LR 0.000125 Time 0.019625 -2023-02-13 18:40:43,592 - Epoch: [187][ 1180/ 1207] Overall Loss 0.203794 Objective Loss 0.203794 LR 0.000125 Time 0.019620 -2023-02-13 18:40:43,782 - Epoch: [187][ 1190/ 1207] Overall Loss 0.203903 Objective Loss 0.203903 LR 0.000125 Time 0.019615 -2023-02-13 18:40:44,029 - Epoch: [187][ 1200/ 1207] Overall Loss 0.203665 Objective Loss 0.203665 LR 0.000125 Time 0.019657 -2023-02-13 18:40:44,145 - Epoch: [187][ 1207/ 1207] Overall Loss 0.203496 Objective Loss 0.203496 Top1 88.414634 Top5 98.780488 LR 0.000125 Time 0.019639 -2023-02-13 18:40:44,217 - --- validate (epoch=187)----------- -2023-02-13 18:40:44,217 - 34311 samples (256 per mini-batch) -2023-02-13 18:40:44,611 - Epoch: [187][ 10/ 135] Loss 0.305882 Top1 84.765625 Top5 97.226562 -2023-02-13 18:40:44,740 - Epoch: [187][ 20/ 135] Loss 0.295628 Top1 84.921875 Top5 97.421875 -2023-02-13 18:40:44,867 - Epoch: [187][ 30/ 135] Loss 0.289713 Top1 85.338542 Top5 97.617188 -2023-02-13 18:40:44,997 - Epoch: [187][ 40/ 135] Loss 0.283671 Top1 85.537109 Top5 97.695312 -2023-02-13 18:40:45,127 - Epoch: [187][ 50/ 135] Loss 0.286321 Top1 85.593750 Top5 97.664062 -2023-02-13 18:40:45,256 - Epoch: [187][ 60/ 135] Loss 0.287717 Top1 85.546875 Top5 97.734375 -2023-02-13 18:40:45,384 - Epoch: [187][ 70/ 135] Loss 0.286042 Top1 85.569196 Top5 97.739955 -2023-02-13 18:40:45,513 - Epoch: [187][ 80/ 135] Loss 0.286931 Top1 85.615234 Top5 97.768555 -2023-02-13 18:40:45,643 - Epoch: [187][ 90/ 135] Loss 0.285798 Top1 85.729167 Top5 97.734375 -2023-02-13 18:40:45,768 - Epoch: [187][ 100/ 135] Loss 0.289185 Top1 85.472656 Top5 97.742188 -2023-02-13 18:40:45,891 - Epoch: [187][ 110/ 135] Loss 0.287427 Top1 85.433239 Top5 97.720170 -2023-02-13 18:40:46,024 - Epoch: [187][ 120/ 135] Loss 0.288568 Top1 85.400391 Top5 97.737630 -2023-02-13 18:40:46,157 - Epoch: [187][ 130/ 135] Loss 0.289695 Top1 85.348558 Top5 97.749399 -2023-02-13 18:40:46,203 - Epoch: [187][ 135/ 135] Loss 0.293894 Top1 85.319577 Top5 97.749993 -2023-02-13 18:40:46,273 - ==> Top1: 85.320 Top5: 97.750 Loss: 0.294 - -2023-02-13 18:40:46,273 - ==> Confusion: -[[ 867 5 5 1 7 2 0 0 5 43 0 4 1 4 5 3 1 3 0 3 8] - [ 2 963 2 2 9 14 2 12 2 1 0 1 2 0 0 2 6 0 2 1 10] - [ 6 2 967 11 4 0 12 13 0 1 3 4 2 2 3 6 4 4 3 2 9] - [ 5 1 25 905 0 6 1 2 2 0 11 0 6 1 19 3 4 7 15 0 3] - [ 15 10 1 0 997 6 1 2 1 1 1 5 1 4 7 5 3 1 1 2 2] - [ 0 17 1 4 6 963 2 22 2 2 2 11 3 17 0 4 6 1 1 2 4] - [ 1 3 8 1 0 5 1051 8 0 1 3 0 1 1 1 1 2 3 2 5 2] - [ 2 10 11 0 1 21 3 932 0 1 1 6 3 3 0 0 0 1 17 5 7] - [ 13 3 1 1 1 0 1 2 921 30 5 2 0 10 11 3 2 0 3 0 0] - [ 73 1 4 0 10 0 0 2 36 859 0 1 0 15 4 1 1 2 1 0 2] - [ 2 0 4 5 2 3 2 4 13 1 994 0 1 8 4 0 1 1 3 0 3] - [ 1 3 3 0 1 17 1 7 1 2 0 904 23 6 2 6 1 11 1 12 3] - [ 0 0 0 8 1 3 0 0 2 1 0 17 892 2 1 7 4 13 1 2 5] - [ 6 2 3 1 5 5 1 1 9 17 9 6 1 938 4 4 3 0 1 2 6] - [ 8 4 2 14 5 3 0 1 18 8 4 1 3 1 1001 1 3 5 4 0 6] - [ 2 2 8 0 4 1 2 0 1 0 0 9 6 2 0 973 11 12 0 7 6] - [ 3 5 0 3 6 1 0 0 1 1 0 0 2 2 2 11 1008 0 3 4 9] - [ 4 2 0 3 0 3 1 0 0 0 0 8 7 3 0 13 0 1003 0 0 4] - [ 3 4 7 8 1 2 0 22 3 0 4 1 2 0 10 1 0 2 1014 1 1] - [ 1 3 0 0 1 1 9 12 1 0 1 12 4 2 1 6 5 2 0 1081 6] - [ 152 226 235 114 134 167 83 182 96 84 201 98 305 263 154 82 276 106 173 262 10041]] - -2023-02-13 18:40:46,275 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:40:46,275 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:40:46,281 - - -2023-02-13 18:40:46,281 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:40:47,158 - Epoch: [188][ 10/ 1207] Overall Loss 0.191848 Objective Loss 0.191848 LR 0.000125 Time 0.087648 -2023-02-13 18:40:47,355 - Epoch: [188][ 20/ 1207] Overall Loss 0.201937 Objective Loss 0.201937 LR 0.000125 Time 0.053673 -2023-02-13 18:40:47,546 - Epoch: [188][ 30/ 1207] Overall Loss 0.201929 Objective Loss 0.201929 LR 0.000125 Time 0.042124 -2023-02-13 18:40:47,736 - Epoch: [188][ 40/ 1207] Overall Loss 0.202167 Objective Loss 0.202167 LR 0.000125 Time 0.036324 -2023-02-13 18:40:47,924 - Epoch: [188][ 50/ 1207] Overall Loss 0.204620 Objective Loss 0.204620 LR 0.000125 Time 0.032829 -2023-02-13 18:40:48,117 - Epoch: [188][ 60/ 1207] Overall Loss 0.200446 Objective Loss 0.200446 LR 0.000125 Time 0.030554 -2023-02-13 18:40:48,311 - Epoch: [188][ 70/ 1207] Overall Loss 0.200548 Objective Loss 0.200548 LR 0.000125 Time 0.028962 -2023-02-13 18:40:48,508 - Epoch: [188][ 80/ 1207] Overall Loss 0.200304 Objective Loss 0.200304 LR 0.000125 Time 0.027804 -2023-02-13 18:40:48,704 - Epoch: [188][ 90/ 1207] Overall Loss 0.201897 Objective Loss 0.201897 LR 0.000125 Time 0.026885 -2023-02-13 18:40:48,901 - Epoch: [188][ 100/ 1207] Overall Loss 0.200682 Objective Loss 0.200682 LR 0.000125 Time 0.026162 -2023-02-13 18:40:49,096 - Epoch: [188][ 110/ 1207] Overall Loss 0.200702 Objective Loss 0.200702 LR 0.000125 Time 0.025556 -2023-02-13 18:40:49,293 - Epoch: [188][ 120/ 1207] Overall Loss 0.201207 Objective Loss 0.201207 LR 0.000125 Time 0.025065 -2023-02-13 18:40:49,488 - Epoch: [188][ 130/ 1207] Overall Loss 0.202754 Objective Loss 0.202754 LR 0.000125 Time 0.024631 -2023-02-13 18:40:49,685 - Epoch: [188][ 140/ 1207] Overall Loss 0.201806 Objective Loss 0.201806 LR 0.000125 Time 0.024278 -2023-02-13 18:40:49,880 - Epoch: [188][ 150/ 1207] Overall Loss 0.202032 Objective Loss 0.202032 LR 0.000125 Time 0.023957 -2023-02-13 18:40:50,077 - Epoch: [188][ 160/ 1207] Overall Loss 0.202254 Objective Loss 0.202254 LR 0.000125 Time 0.023689 -2023-02-13 18:40:50,272 - Epoch: [188][ 170/ 1207] Overall Loss 0.202869 Objective Loss 0.202869 LR 0.000125 Time 0.023439 -2023-02-13 18:40:50,469 - Epoch: [188][ 180/ 1207] Overall Loss 0.203007 Objective Loss 0.203007 LR 0.000125 Time 0.023227 -2023-02-13 18:40:50,664 - Epoch: [188][ 190/ 1207] Overall Loss 0.203370 Objective Loss 0.203370 LR 0.000125 Time 0.023030 -2023-02-13 18:40:50,861 - Epoch: [188][ 200/ 1207] Overall Loss 0.204004 Objective Loss 0.204004 LR 0.000125 Time 0.022860 -2023-02-13 18:40:51,056 - Epoch: [188][ 210/ 1207] Overall Loss 0.205297 Objective Loss 0.205297 LR 0.000125 Time 0.022701 -2023-02-13 18:40:51,253 - Epoch: [188][ 220/ 1207] Overall Loss 0.205146 Objective Loss 0.205146 LR 0.000125 Time 0.022562 -2023-02-13 18:40:51,448 - Epoch: [188][ 230/ 1207] Overall Loss 0.204359 Objective Loss 0.204359 LR 0.000125 Time 0.022426 -2023-02-13 18:40:51,645 - Epoch: [188][ 240/ 1207] Overall Loss 0.204514 Objective Loss 0.204514 LR 0.000125 Time 0.022313 -2023-02-13 18:40:51,841 - Epoch: [188][ 250/ 1207] Overall Loss 0.204446 Objective Loss 0.204446 LR 0.000125 Time 0.022200 -2023-02-13 18:40:52,038 - Epoch: [188][ 260/ 1207] Overall Loss 0.204077 Objective Loss 0.204077 LR 0.000125 Time 0.022105 -2023-02-13 18:40:52,233 - Epoch: [188][ 270/ 1207] Overall Loss 0.203924 Objective Loss 0.203924 LR 0.000125 Time 0.022007 -2023-02-13 18:40:52,430 - Epoch: [188][ 280/ 1207] Overall Loss 0.203881 Objective Loss 0.203881 LR 0.000125 Time 0.021923 -2023-02-13 18:40:52,626 - Epoch: [188][ 290/ 1207] Overall Loss 0.203642 Objective Loss 0.203642 LR 0.000125 Time 0.021840 -2023-02-13 18:40:52,823 - Epoch: [188][ 300/ 1207] Overall Loss 0.203855 Objective Loss 0.203855 LR 0.000125 Time 0.021769 -2023-02-13 18:40:53,018 - Epoch: [188][ 310/ 1207] Overall Loss 0.204398 Objective Loss 0.204398 LR 0.000125 Time 0.021695 -2023-02-13 18:40:53,215 - Epoch: [188][ 320/ 1207] Overall Loss 0.204666 Objective Loss 0.204666 LR 0.000125 Time 0.021632 -2023-02-13 18:40:53,410 - Epoch: [188][ 330/ 1207] Overall Loss 0.204478 Objective Loss 0.204478 LR 0.000125 Time 0.021564 -2023-02-13 18:40:53,608 - Epoch: [188][ 340/ 1207] Overall Loss 0.204175 Objective Loss 0.204175 LR 0.000125 Time 0.021511 -2023-02-13 18:40:53,802 - Epoch: [188][ 350/ 1207] Overall Loss 0.204737 Objective Loss 0.204737 LR 0.000125 Time 0.021451 -2023-02-13 18:40:53,999 - Epoch: [188][ 360/ 1207] Overall Loss 0.205022 Objective Loss 0.205022 LR 0.000125 Time 0.021402 -2023-02-13 18:40:54,194 - Epoch: [188][ 370/ 1207] Overall Loss 0.204988 Objective Loss 0.204988 LR 0.000125 Time 0.021349 -2023-02-13 18:40:54,391 - Epoch: [188][ 380/ 1207] Overall Loss 0.205141 Objective Loss 0.205141 LR 0.000125 Time 0.021305 -2023-02-13 18:40:54,587 - Epoch: [188][ 390/ 1207] Overall Loss 0.205052 Objective Loss 0.205052 LR 0.000125 Time 0.021259 -2023-02-13 18:40:54,785 - Epoch: [188][ 400/ 1207] Overall Loss 0.205192 Objective Loss 0.205192 LR 0.000125 Time 0.021221 -2023-02-13 18:40:54,979 - Epoch: [188][ 410/ 1207] Overall Loss 0.204727 Objective Loss 0.204727 LR 0.000125 Time 0.021177 -2023-02-13 18:40:55,178 - Epoch: [188][ 420/ 1207] Overall Loss 0.204762 Objective Loss 0.204762 LR 0.000125 Time 0.021144 -2023-02-13 18:40:55,373 - Epoch: [188][ 430/ 1207] Overall Loss 0.204221 Objective Loss 0.204221 LR 0.000125 Time 0.021106 -2023-02-13 18:40:55,571 - Epoch: [188][ 440/ 1207] Overall Loss 0.204166 Objective Loss 0.204166 LR 0.000125 Time 0.021076 -2023-02-13 18:40:55,761 - Epoch: [188][ 450/ 1207] Overall Loss 0.204026 Objective Loss 0.204026 LR 0.000125 Time 0.021029 -2023-02-13 18:40:55,953 - Epoch: [188][ 460/ 1207] Overall Loss 0.204390 Objective Loss 0.204390 LR 0.000125 Time 0.020987 -2023-02-13 18:40:56,143 - Epoch: [188][ 470/ 1207] Overall Loss 0.204298 Objective Loss 0.204298 LR 0.000125 Time 0.020944 -2023-02-13 18:40:56,333 - Epoch: [188][ 480/ 1207] Overall Loss 0.204017 Objective Loss 0.204017 LR 0.000125 Time 0.020904 -2023-02-13 18:40:56,523 - Epoch: [188][ 490/ 1207] Overall Loss 0.204305 Objective Loss 0.204305 LR 0.000125 Time 0.020863 -2023-02-13 18:40:56,714 - Epoch: [188][ 500/ 1207] Overall Loss 0.204069 Objective Loss 0.204069 LR 0.000125 Time 0.020827 -2023-02-13 18:40:56,903 - Epoch: [188][ 510/ 1207] Overall Loss 0.204570 Objective Loss 0.204570 LR 0.000125 Time 0.020790 -2023-02-13 18:40:57,094 - Epoch: [188][ 520/ 1207] Overall Loss 0.204347 Objective Loss 0.204347 LR 0.000125 Time 0.020755 -2023-02-13 18:40:57,283 - Epoch: [188][ 530/ 1207] Overall Loss 0.203947 Objective Loss 0.203947 LR 0.000125 Time 0.020720 -2023-02-13 18:40:57,473 - Epoch: [188][ 540/ 1207] Overall Loss 0.203973 Objective Loss 0.203973 LR 0.000125 Time 0.020688 -2023-02-13 18:40:57,663 - Epoch: [188][ 550/ 1207] Overall Loss 0.203769 Objective Loss 0.203769 LR 0.000125 Time 0.020657 -2023-02-13 18:40:57,854 - Epoch: [188][ 560/ 1207] Overall Loss 0.203558 Objective Loss 0.203558 LR 0.000125 Time 0.020627 -2023-02-13 18:40:58,043 - Epoch: [188][ 570/ 1207] Overall Loss 0.203646 Objective Loss 0.203646 LR 0.000125 Time 0.020598 -2023-02-13 18:40:58,233 - Epoch: [188][ 580/ 1207] Overall Loss 0.203576 Objective Loss 0.203576 LR 0.000125 Time 0.020569 -2023-02-13 18:40:58,423 - Epoch: [188][ 590/ 1207] Overall Loss 0.203661 Objective Loss 0.203661 LR 0.000125 Time 0.020541 -2023-02-13 18:40:58,614 - Epoch: [188][ 600/ 1207] Overall Loss 0.203391 Objective Loss 0.203391 LR 0.000125 Time 0.020516 -2023-02-13 18:40:58,803 - Epoch: [188][ 610/ 1207] Overall Loss 0.203309 Objective Loss 0.203309 LR 0.000125 Time 0.020490 -2023-02-13 18:40:58,994 - Epoch: [188][ 620/ 1207] Overall Loss 0.203135 Objective Loss 0.203135 LR 0.000125 Time 0.020467 -2023-02-13 18:40:59,184 - Epoch: [188][ 630/ 1207] Overall Loss 0.203114 Objective Loss 0.203114 LR 0.000125 Time 0.020443 -2023-02-13 18:40:59,374 - Epoch: [188][ 640/ 1207] Overall Loss 0.202790 Objective Loss 0.202790 LR 0.000125 Time 0.020420 -2023-02-13 18:40:59,565 - Epoch: [188][ 650/ 1207] Overall Loss 0.202903 Objective Loss 0.202903 LR 0.000125 Time 0.020398 -2023-02-13 18:40:59,755 - Epoch: [188][ 660/ 1207] Overall Loss 0.202820 Objective Loss 0.202820 LR 0.000125 Time 0.020377 -2023-02-13 18:40:59,945 - Epoch: [188][ 670/ 1207] Overall Loss 0.202887 Objective Loss 0.202887 LR 0.000125 Time 0.020355 -2023-02-13 18:41:00,135 - Epoch: [188][ 680/ 1207] Overall Loss 0.202819 Objective Loss 0.202819 LR 0.000125 Time 0.020335 -2023-02-13 18:41:00,325 - Epoch: [188][ 690/ 1207] Overall Loss 0.202714 Objective Loss 0.202714 LR 0.000125 Time 0.020315 -2023-02-13 18:41:00,515 - Epoch: [188][ 700/ 1207] Overall Loss 0.202835 Objective Loss 0.202835 LR 0.000125 Time 0.020295 -2023-02-13 18:41:00,705 - Epoch: [188][ 710/ 1207] Overall Loss 0.202825 Objective Loss 0.202825 LR 0.000125 Time 0.020278 -2023-02-13 18:41:00,896 - Epoch: [188][ 720/ 1207] Overall Loss 0.202581 Objective Loss 0.202581 LR 0.000125 Time 0.020260 -2023-02-13 18:41:01,087 - Epoch: [188][ 730/ 1207] Overall Loss 0.202781 Objective Loss 0.202781 LR 0.000125 Time 0.020244 -2023-02-13 18:41:01,277 - Epoch: [188][ 740/ 1207] Overall Loss 0.202899 Objective Loss 0.202899 LR 0.000125 Time 0.020227 -2023-02-13 18:41:01,466 - Epoch: [188][ 750/ 1207] Overall Loss 0.203075 Objective Loss 0.203075 LR 0.000125 Time 0.020209 -2023-02-13 18:41:01,657 - Epoch: [188][ 760/ 1207] Overall Loss 0.203052 Objective Loss 0.203052 LR 0.000125 Time 0.020194 -2023-02-13 18:41:01,847 - Epoch: [188][ 770/ 1207] Overall Loss 0.203040 Objective Loss 0.203040 LR 0.000125 Time 0.020177 -2023-02-13 18:41:02,038 - Epoch: [188][ 780/ 1207] Overall Loss 0.202869 Objective Loss 0.202869 LR 0.000125 Time 0.020163 -2023-02-13 18:41:02,227 - Epoch: [188][ 790/ 1207] Overall Loss 0.202735 Objective Loss 0.202735 LR 0.000125 Time 0.020147 -2023-02-13 18:41:02,417 - Epoch: [188][ 800/ 1207] Overall Loss 0.202943 Objective Loss 0.202943 LR 0.000125 Time 0.020132 -2023-02-13 18:41:02,607 - Epoch: [188][ 810/ 1207] Overall Loss 0.202765 Objective Loss 0.202765 LR 0.000125 Time 0.020118 -2023-02-13 18:41:02,798 - Epoch: [188][ 820/ 1207] Overall Loss 0.202859 Objective Loss 0.202859 LR 0.000125 Time 0.020104 -2023-02-13 18:41:02,987 - Epoch: [188][ 830/ 1207] Overall Loss 0.202997 Objective Loss 0.202997 LR 0.000125 Time 0.020090 -2023-02-13 18:41:03,178 - Epoch: [188][ 840/ 1207] Overall Loss 0.203187 Objective Loss 0.203187 LR 0.000125 Time 0.020077 -2023-02-13 18:41:03,367 - Epoch: [188][ 850/ 1207] Overall Loss 0.203266 Objective Loss 0.203266 LR 0.000125 Time 0.020063 -2023-02-13 18:41:03,557 - Epoch: [188][ 860/ 1207] Overall Loss 0.203343 Objective Loss 0.203343 LR 0.000125 Time 0.020050 -2023-02-13 18:41:03,748 - Epoch: [188][ 870/ 1207] Overall Loss 0.203193 Objective Loss 0.203193 LR 0.000125 Time 0.020038 -2023-02-13 18:41:03,937 - Epoch: [188][ 880/ 1207] Overall Loss 0.203086 Objective Loss 0.203086 LR 0.000125 Time 0.020026 -2023-02-13 18:41:04,127 - Epoch: [188][ 890/ 1207] Overall Loss 0.203150 Objective Loss 0.203150 LR 0.000125 Time 0.020014 -2023-02-13 18:41:04,317 - Epoch: [188][ 900/ 1207] Overall Loss 0.203203 Objective Loss 0.203203 LR 0.000125 Time 0.020002 -2023-02-13 18:41:04,507 - Epoch: [188][ 910/ 1207] Overall Loss 0.203476 Objective Loss 0.203476 LR 0.000125 Time 0.019990 -2023-02-13 18:41:04,697 - Epoch: [188][ 920/ 1207] Overall Loss 0.203421 Objective Loss 0.203421 LR 0.000125 Time 0.019979 -2023-02-13 18:41:04,887 - Epoch: [188][ 930/ 1207] Overall Loss 0.203139 Objective Loss 0.203139 LR 0.000125 Time 0.019969 -2023-02-13 18:41:05,077 - Epoch: [188][ 940/ 1207] Overall Loss 0.203388 Objective Loss 0.203388 LR 0.000125 Time 0.019958 -2023-02-13 18:41:05,266 - Epoch: [188][ 950/ 1207] Overall Loss 0.203356 Objective Loss 0.203356 LR 0.000125 Time 0.019946 -2023-02-13 18:41:05,455 - Epoch: [188][ 960/ 1207] Overall Loss 0.203159 Objective Loss 0.203159 LR 0.000125 Time 0.019935 -2023-02-13 18:41:05,646 - Epoch: [188][ 970/ 1207] Overall Loss 0.203302 Objective Loss 0.203302 LR 0.000125 Time 0.019925 -2023-02-13 18:41:05,837 - Epoch: [188][ 980/ 1207] Overall Loss 0.203041 Objective Loss 0.203041 LR 0.000125 Time 0.019917 -2023-02-13 18:41:06,028 - Epoch: [188][ 990/ 1207] Overall Loss 0.203280 Objective Loss 0.203280 LR 0.000125 Time 0.019909 -2023-02-13 18:41:06,219 - Epoch: [188][ 1000/ 1207] Overall Loss 0.203331 Objective Loss 0.203331 LR 0.000125 Time 0.019900 -2023-02-13 18:41:06,409 - Epoch: [188][ 1010/ 1207] Overall Loss 0.203409 Objective Loss 0.203409 LR 0.000125 Time 0.019891 -2023-02-13 18:41:06,600 - Epoch: [188][ 1020/ 1207] Overall Loss 0.203355 Objective Loss 0.203355 LR 0.000125 Time 0.019882 -2023-02-13 18:41:06,791 - Epoch: [188][ 1030/ 1207] Overall Loss 0.203300 Objective Loss 0.203300 LR 0.000125 Time 0.019874 -2023-02-13 18:41:06,982 - Epoch: [188][ 1040/ 1207] Overall Loss 0.203297 Objective Loss 0.203297 LR 0.000125 Time 0.019866 -2023-02-13 18:41:07,172 - Epoch: [188][ 1050/ 1207] Overall Loss 0.203277 Objective Loss 0.203277 LR 0.000125 Time 0.019858 -2023-02-13 18:41:07,362 - Epoch: [188][ 1060/ 1207] Overall Loss 0.203612 Objective Loss 0.203612 LR 0.000125 Time 0.019850 -2023-02-13 18:41:07,551 - Epoch: [188][ 1070/ 1207] Overall Loss 0.203867 Objective Loss 0.203867 LR 0.000125 Time 0.019841 -2023-02-13 18:41:07,742 - Epoch: [188][ 1080/ 1207] Overall Loss 0.203799 Objective Loss 0.203799 LR 0.000125 Time 0.019834 -2023-02-13 18:41:07,933 - Epoch: [188][ 1090/ 1207] Overall Loss 0.203662 Objective Loss 0.203662 LR 0.000125 Time 0.019826 -2023-02-13 18:41:08,123 - Epoch: [188][ 1100/ 1207] Overall Loss 0.203592 Objective Loss 0.203592 LR 0.000125 Time 0.019818 -2023-02-13 18:41:08,312 - Epoch: [188][ 1110/ 1207] Overall Loss 0.203502 Objective Loss 0.203502 LR 0.000125 Time 0.019809 -2023-02-13 18:41:08,501 - Epoch: [188][ 1120/ 1207] Overall Loss 0.203534 Objective Loss 0.203534 LR 0.000125 Time 0.019801 -2023-02-13 18:41:08,692 - Epoch: [188][ 1130/ 1207] Overall Loss 0.203532 Objective Loss 0.203532 LR 0.000125 Time 0.019794 -2023-02-13 18:41:08,882 - Epoch: [188][ 1140/ 1207] Overall Loss 0.203598 Objective Loss 0.203598 LR 0.000125 Time 0.019787 -2023-02-13 18:41:09,072 - Epoch: [188][ 1150/ 1207] Overall Loss 0.203745 Objective Loss 0.203745 LR 0.000125 Time 0.019780 -2023-02-13 18:41:09,262 - Epoch: [188][ 1160/ 1207] Overall Loss 0.203591 Objective Loss 0.203591 LR 0.000125 Time 0.019773 -2023-02-13 18:41:09,452 - Epoch: [188][ 1170/ 1207] Overall Loss 0.203799 Objective Loss 0.203799 LR 0.000125 Time 0.019766 -2023-02-13 18:41:09,643 - Epoch: [188][ 1180/ 1207] Overall Loss 0.203648 Objective Loss 0.203648 LR 0.000125 Time 0.019760 -2023-02-13 18:41:09,833 - Epoch: [188][ 1190/ 1207] Overall Loss 0.203787 Objective Loss 0.203787 LR 0.000125 Time 0.019753 -2023-02-13 18:41:10,074 - Epoch: [188][ 1200/ 1207] Overall Loss 0.203751 Objective Loss 0.203751 LR 0.000125 Time 0.019789 -2023-02-13 18:41:10,190 - Epoch: [188][ 1207/ 1207] Overall Loss 0.203906 Objective Loss 0.203906 Top1 87.195122 Top5 98.475610 LR 0.000125 Time 0.019771 -2023-02-13 18:41:10,263 - --- validate (epoch=188)----------- -2023-02-13 18:41:10,263 - 34311 samples (256 per mini-batch) -2023-02-13 18:41:10,765 - Epoch: [188][ 10/ 135] Loss 0.281414 Top1 86.210938 Top5 98.281250 -2023-02-13 18:41:10,887 - Epoch: [188][ 20/ 135] Loss 0.290867 Top1 85.820312 Top5 97.988281 -2023-02-13 18:41:11,016 - Epoch: [188][ 30/ 135] Loss 0.287707 Top1 85.338542 Top5 98.072917 -2023-02-13 18:41:11,142 - Epoch: [188][ 40/ 135] Loss 0.292152 Top1 85.136719 Top5 97.968750 -2023-02-13 18:41:11,269 - Epoch: [188][ 50/ 135] Loss 0.296218 Top1 85.171875 Top5 97.812500 -2023-02-13 18:41:11,394 - Epoch: [188][ 60/ 135] Loss 0.292314 Top1 85.325521 Top5 97.890625 -2023-02-13 18:41:11,519 - Epoch: [188][ 70/ 135] Loss 0.290123 Top1 85.390625 Top5 97.924107 -2023-02-13 18:41:11,644 - Epoch: [188][ 80/ 135] Loss 0.289337 Top1 85.317383 Top5 97.934570 -2023-02-13 18:41:11,769 - Epoch: [188][ 90/ 135] Loss 0.290111 Top1 85.373264 Top5 97.942708 -2023-02-13 18:41:11,893 - Epoch: [188][ 100/ 135] Loss 0.290085 Top1 85.375000 Top5 97.894531 -2023-02-13 18:41:12,023 - Epoch: [188][ 110/ 135] Loss 0.292818 Top1 85.298295 Top5 97.851562 -2023-02-13 18:41:12,150 - Epoch: [188][ 120/ 135] Loss 0.294475 Top1 85.240885 Top5 97.838542 -2023-02-13 18:41:12,279 - Epoch: [188][ 130/ 135] Loss 0.293689 Top1 85.210337 Top5 97.839543 -2023-02-13 18:41:12,324 - Epoch: [188][ 135/ 135] Loss 0.292098 Top1 85.226312 Top5 97.828685 -2023-02-13 18:41:12,400 - ==> Top1: 85.226 Top5: 97.829 Loss: 0.292 - -2023-02-13 18:41:12,400 - ==> Confusion: -[[ 867 5 7 1 9 1 0 1 4 43 0 2 2 5 5 4 0 1 1 1 8] - [ 2 960 2 2 7 19 3 15 2 0 0 2 1 1 0 1 3 0 4 3 6] - [ 5 6 974 12 1 2 9 10 1 2 3 3 1 3 3 7 1 1 5 2 7] - [ 4 0 22 903 1 4 1 2 2 2 17 0 7 0 17 4 3 3 18 0 6] - [ 11 9 2 0 993 11 1 2 1 1 0 7 3 2 7 6 2 0 0 2 6] - [ 2 15 1 2 3 982 2 15 1 3 0 10 1 14 0 3 5 1 3 3 4] - [ 2 2 15 1 0 2 1046 7 0 3 3 1 2 3 0 3 1 0 1 3 4] - [ 1 6 11 0 2 31 3 926 0 2 0 4 3 1 0 0 2 0 21 7 4] - [ 18 3 0 1 1 0 0 1 901 28 11 4 0 11 23 1 1 0 4 1 0] - [ 78 0 3 0 16 2 0 2 29 846 0 1 1 17 5 2 2 1 2 1 4] - [ 3 1 3 4 3 1 2 4 11 2 991 1 0 10 2 0 2 1 7 0 3] - [ 2 2 3 0 3 13 1 5 1 1 0 911 23 6 3 8 1 8 1 8 5] - [ 1 0 1 11 1 5 0 1 1 1 0 22 878 1 1 4 4 15 3 1 8] - [ 4 2 3 0 9 3 0 3 6 12 7 5 1 950 4 2 5 2 0 0 6] - [ 5 2 1 12 4 5 0 1 12 7 5 0 2 1 1012 0 3 5 7 0 8] - [ 2 1 8 1 7 0 4 2 0 0 0 8 6 3 0 972 11 8 1 6 6] - [ 1 6 0 1 7 1 0 0 1 0 0 0 0 3 3 8 1011 2 2 4 11] - [ 4 1 0 6 1 1 2 0 1 0 2 7 13 1 0 18 0 985 0 2 7] - [ 1 4 4 7 2 2 0 22 4 1 7 1 3 0 10 0 1 2 1012 1 2] - [ 0 3 1 0 0 4 9 11 0 0 0 14 1 2 1 9 4 4 0 1078 7] - [ 142 252 235 111 126 209 72 162 82 72 203 111 276 294 177 87 267 94 192 226 10044]] - -2023-02-13 18:41:12,402 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:41:12,402 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:41:12,408 - - -2023-02-13 18:41:12,408 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:41:13,299 - Epoch: [189][ 10/ 1207] Overall Loss 0.239125 Objective Loss 0.239125 LR 0.000125 Time 0.088997 -2023-02-13 18:41:13,490 - Epoch: [189][ 20/ 1207] Overall Loss 0.215426 Objective Loss 0.215426 LR 0.000125 Time 0.054050 -2023-02-13 18:41:13,681 - Epoch: [189][ 30/ 1207] Overall Loss 0.204334 Objective Loss 0.204334 LR 0.000125 Time 0.042379 -2023-02-13 18:41:13,870 - Epoch: [189][ 40/ 1207] Overall Loss 0.209542 Objective Loss 0.209542 LR 0.000125 Time 0.036510 -2023-02-13 18:41:14,059 - Epoch: [189][ 50/ 1207] Overall Loss 0.203721 Objective Loss 0.203721 LR 0.000125 Time 0.032981 -2023-02-13 18:41:14,248 - Epoch: [189][ 60/ 1207] Overall Loss 0.202712 Objective Loss 0.202712 LR 0.000125 Time 0.030628 -2023-02-13 18:41:14,437 - Epoch: [189][ 70/ 1207] Overall Loss 0.202398 Objective Loss 0.202398 LR 0.000125 Time 0.028946 -2023-02-13 18:41:14,627 - Epoch: [189][ 80/ 1207] Overall Loss 0.204210 Objective Loss 0.204210 LR 0.000125 Time 0.027691 -2023-02-13 18:41:14,816 - Epoch: [189][ 90/ 1207] Overall Loss 0.205314 Objective Loss 0.205314 LR 0.000125 Time 0.026711 -2023-02-13 18:41:15,005 - Epoch: [189][ 100/ 1207] Overall Loss 0.204810 Objective Loss 0.204810 LR 0.000125 Time 0.025925 -2023-02-13 18:41:15,194 - Epoch: [189][ 110/ 1207] Overall Loss 0.206160 Objective Loss 0.206160 LR 0.000125 Time 0.025288 -2023-02-13 18:41:15,383 - Epoch: [189][ 120/ 1207] Overall Loss 0.204800 Objective Loss 0.204800 LR 0.000125 Time 0.024752 -2023-02-13 18:41:15,573 - Epoch: [189][ 130/ 1207] Overall Loss 0.204516 Objective Loss 0.204516 LR 0.000125 Time 0.024303 -2023-02-13 18:41:15,762 - Epoch: [189][ 140/ 1207] Overall Loss 0.204928 Objective Loss 0.204928 LR 0.000125 Time 0.023918 -2023-02-13 18:41:15,952 - Epoch: [189][ 150/ 1207] Overall Loss 0.203892 Objective Loss 0.203892 LR 0.000125 Time 0.023586 -2023-02-13 18:41:16,141 - Epoch: [189][ 160/ 1207] Overall Loss 0.202692 Objective Loss 0.202692 LR 0.000125 Time 0.023290 -2023-02-13 18:41:16,331 - Epoch: [189][ 170/ 1207] Overall Loss 0.204389 Objective Loss 0.204389 LR 0.000125 Time 0.023033 -2023-02-13 18:41:16,519 - Epoch: [189][ 180/ 1207] Overall Loss 0.203857 Objective Loss 0.203857 LR 0.000125 Time 0.022800 -2023-02-13 18:41:16,712 - Epoch: [189][ 190/ 1207] Overall Loss 0.202797 Objective Loss 0.202797 LR 0.000125 Time 0.022610 -2023-02-13 18:41:16,902 - Epoch: [189][ 200/ 1207] Overall Loss 0.203059 Objective Loss 0.203059 LR 0.000125 Time 0.022431 -2023-02-13 18:41:17,095 - Epoch: [189][ 210/ 1207] Overall Loss 0.203165 Objective Loss 0.203165 LR 0.000125 Time 0.022280 -2023-02-13 18:41:17,286 - Epoch: [189][ 220/ 1207] Overall Loss 0.203910 Objective Loss 0.203910 LR 0.000125 Time 0.022133 -2023-02-13 18:41:17,478 - Epoch: [189][ 230/ 1207] Overall Loss 0.203588 Objective Loss 0.203588 LR 0.000125 Time 0.022003 -2023-02-13 18:41:17,670 - Epoch: [189][ 240/ 1207] Overall Loss 0.203321 Objective Loss 0.203321 LR 0.000125 Time 0.021883 -2023-02-13 18:41:17,860 - Epoch: [189][ 250/ 1207] Overall Loss 0.203849 Objective Loss 0.203849 LR 0.000125 Time 0.021768 -2023-02-13 18:41:18,052 - Epoch: [189][ 260/ 1207] Overall Loss 0.203215 Objective Loss 0.203215 LR 0.000125 Time 0.021668 -2023-02-13 18:41:18,243 - Epoch: [189][ 270/ 1207] Overall Loss 0.203616 Objective Loss 0.203616 LR 0.000125 Time 0.021571 -2023-02-13 18:41:18,434 - Epoch: [189][ 280/ 1207] Overall Loss 0.203693 Objective Loss 0.203693 LR 0.000125 Time 0.021482 -2023-02-13 18:41:18,626 - Epoch: [189][ 290/ 1207] Overall Loss 0.203667 Objective Loss 0.203667 LR 0.000125 Time 0.021401 -2023-02-13 18:41:18,817 - Epoch: [189][ 300/ 1207] Overall Loss 0.203499 Objective Loss 0.203499 LR 0.000125 Time 0.021324 -2023-02-13 18:41:19,008 - Epoch: [189][ 310/ 1207] Overall Loss 0.203564 Objective Loss 0.203564 LR 0.000125 Time 0.021250 -2023-02-13 18:41:19,201 - Epoch: [189][ 320/ 1207] Overall Loss 0.203270 Objective Loss 0.203270 LR 0.000125 Time 0.021188 -2023-02-13 18:41:19,392 - Epoch: [189][ 330/ 1207] Overall Loss 0.203166 Objective Loss 0.203166 LR 0.000125 Time 0.021123 -2023-02-13 18:41:19,583 - Epoch: [189][ 340/ 1207] Overall Loss 0.202710 Objective Loss 0.202710 LR 0.000125 Time 0.021063 -2023-02-13 18:41:19,775 - Epoch: [189][ 350/ 1207] Overall Loss 0.202017 Objective Loss 0.202017 LR 0.000125 Time 0.021008 -2023-02-13 18:41:19,965 - Epoch: [189][ 360/ 1207] Overall Loss 0.201979 Objective Loss 0.201979 LR 0.000125 Time 0.020953 -2023-02-13 18:41:20,156 - Epoch: [189][ 370/ 1207] Overall Loss 0.202084 Objective Loss 0.202084 LR 0.000125 Time 0.020901 -2023-02-13 18:41:20,347 - Epoch: [189][ 380/ 1207] Overall Loss 0.201898 Objective Loss 0.201898 LR 0.000125 Time 0.020853 -2023-02-13 18:41:20,538 - Epoch: [189][ 390/ 1207] Overall Loss 0.202185 Objective Loss 0.202185 LR 0.000125 Time 0.020807 -2023-02-13 18:41:20,730 - Epoch: [189][ 400/ 1207] Overall Loss 0.202590 Objective Loss 0.202590 LR 0.000125 Time 0.020765 -2023-02-13 18:41:20,922 - Epoch: [189][ 410/ 1207] Overall Loss 0.202565 Objective Loss 0.202565 LR 0.000125 Time 0.020726 -2023-02-13 18:41:21,114 - Epoch: [189][ 420/ 1207] Overall Loss 0.202486 Objective Loss 0.202486 LR 0.000125 Time 0.020688 -2023-02-13 18:41:21,305 - Epoch: [189][ 430/ 1207] Overall Loss 0.202754 Objective Loss 0.202754 LR 0.000125 Time 0.020650 -2023-02-13 18:41:21,497 - Epoch: [189][ 440/ 1207] Overall Loss 0.202728 Objective Loss 0.202728 LR 0.000125 Time 0.020617 -2023-02-13 18:41:21,688 - Epoch: [189][ 450/ 1207] Overall Loss 0.202878 Objective Loss 0.202878 LR 0.000125 Time 0.020582 -2023-02-13 18:41:21,879 - Epoch: [189][ 460/ 1207] Overall Loss 0.202966 Objective Loss 0.202966 LR 0.000125 Time 0.020549 -2023-02-13 18:41:22,071 - Epoch: [189][ 470/ 1207] Overall Loss 0.203124 Objective Loss 0.203124 LR 0.000125 Time 0.020521 -2023-02-13 18:41:22,263 - Epoch: [189][ 480/ 1207] Overall Loss 0.203038 Objective Loss 0.203038 LR 0.000125 Time 0.020492 -2023-02-13 18:41:22,454 - Epoch: [189][ 490/ 1207] Overall Loss 0.203026 Objective Loss 0.203026 LR 0.000125 Time 0.020462 -2023-02-13 18:41:22,646 - Epoch: [189][ 500/ 1207] Overall Loss 0.203031 Objective Loss 0.203031 LR 0.000125 Time 0.020437 -2023-02-13 18:41:22,838 - Epoch: [189][ 510/ 1207] Overall Loss 0.202882 Objective Loss 0.202882 LR 0.000125 Time 0.020412 -2023-02-13 18:41:23,029 - Epoch: [189][ 520/ 1207] Overall Loss 0.202815 Objective Loss 0.202815 LR 0.000125 Time 0.020386 -2023-02-13 18:41:23,219 - Epoch: [189][ 530/ 1207] Overall Loss 0.203078 Objective Loss 0.203078 LR 0.000125 Time 0.020360 -2023-02-13 18:41:23,410 - Epoch: [189][ 540/ 1207] Overall Loss 0.203304 Objective Loss 0.203304 LR 0.000125 Time 0.020336 -2023-02-13 18:41:23,603 - Epoch: [189][ 550/ 1207] Overall Loss 0.203370 Objective Loss 0.203370 LR 0.000125 Time 0.020315 -2023-02-13 18:41:23,794 - Epoch: [189][ 560/ 1207] Overall Loss 0.203088 Objective Loss 0.203088 LR 0.000125 Time 0.020293 -2023-02-13 18:41:23,984 - Epoch: [189][ 570/ 1207] Overall Loss 0.203100 Objective Loss 0.203100 LR 0.000125 Time 0.020269 -2023-02-13 18:41:24,175 - Epoch: [189][ 580/ 1207] Overall Loss 0.203302 Objective Loss 0.203302 LR 0.000125 Time 0.020249 -2023-02-13 18:41:24,366 - Epoch: [189][ 590/ 1207] Overall Loss 0.203012 Objective Loss 0.203012 LR 0.000125 Time 0.020228 -2023-02-13 18:41:24,557 - Epoch: [189][ 600/ 1207] Overall Loss 0.202747 Objective Loss 0.202747 LR 0.000125 Time 0.020210 -2023-02-13 18:41:24,749 - Epoch: [189][ 610/ 1207] Overall Loss 0.202577 Objective Loss 0.202577 LR 0.000125 Time 0.020193 -2023-02-13 18:41:24,941 - Epoch: [189][ 620/ 1207] Overall Loss 0.202619 Objective Loss 0.202619 LR 0.000125 Time 0.020176 -2023-02-13 18:41:25,132 - Epoch: [189][ 630/ 1207] Overall Loss 0.202777 Objective Loss 0.202777 LR 0.000125 Time 0.020158 -2023-02-13 18:41:25,323 - Epoch: [189][ 640/ 1207] Overall Loss 0.202668 Objective Loss 0.202668 LR 0.000125 Time 0.020141 -2023-02-13 18:41:25,514 - Epoch: [189][ 650/ 1207] Overall Loss 0.202554 Objective Loss 0.202554 LR 0.000125 Time 0.020125 -2023-02-13 18:41:25,707 - Epoch: [189][ 660/ 1207] Overall Loss 0.202300 Objective Loss 0.202300 LR 0.000125 Time 0.020111 -2023-02-13 18:41:25,898 - Epoch: [189][ 670/ 1207] Overall Loss 0.202370 Objective Loss 0.202370 LR 0.000125 Time 0.020095 -2023-02-13 18:41:26,092 - Epoch: [189][ 680/ 1207] Overall Loss 0.202412 Objective Loss 0.202412 LR 0.000125 Time 0.020085 -2023-02-13 18:41:26,283 - Epoch: [189][ 690/ 1207] Overall Loss 0.202469 Objective Loss 0.202469 LR 0.000125 Time 0.020070 -2023-02-13 18:41:26,474 - Epoch: [189][ 700/ 1207] Overall Loss 0.202544 Objective Loss 0.202544 LR 0.000125 Time 0.020056 -2023-02-13 18:41:26,667 - Epoch: [189][ 710/ 1207] Overall Loss 0.202661 Objective Loss 0.202661 LR 0.000125 Time 0.020044 -2023-02-13 18:41:26,859 - Epoch: [189][ 720/ 1207] Overall Loss 0.202417 Objective Loss 0.202417 LR 0.000125 Time 0.020032 -2023-02-13 18:41:27,051 - Epoch: [189][ 730/ 1207] Overall Loss 0.202145 Objective Loss 0.202145 LR 0.000125 Time 0.020020 -2023-02-13 18:41:27,242 - Epoch: [189][ 740/ 1207] Overall Loss 0.202016 Objective Loss 0.202016 LR 0.000125 Time 0.020008 -2023-02-13 18:41:27,433 - Epoch: [189][ 750/ 1207] Overall Loss 0.202247 Objective Loss 0.202247 LR 0.000125 Time 0.019994 -2023-02-13 18:41:27,625 - Epoch: [189][ 760/ 1207] Overall Loss 0.202009 Objective Loss 0.202009 LR 0.000125 Time 0.019983 -2023-02-13 18:41:27,817 - Epoch: [189][ 770/ 1207] Overall Loss 0.202150 Objective Loss 0.202150 LR 0.000125 Time 0.019973 -2023-02-13 18:41:28,009 - Epoch: [189][ 780/ 1207] Overall Loss 0.202145 Objective Loss 0.202145 LR 0.000125 Time 0.019963 -2023-02-13 18:41:28,199 - Epoch: [189][ 790/ 1207] Overall Loss 0.202242 Objective Loss 0.202242 LR 0.000125 Time 0.019950 -2023-02-13 18:41:28,390 - Epoch: [189][ 800/ 1207] Overall Loss 0.202202 Objective Loss 0.202202 LR 0.000125 Time 0.019939 -2023-02-13 18:41:28,582 - Epoch: [189][ 810/ 1207] Overall Loss 0.202023 Objective Loss 0.202023 LR 0.000125 Time 0.019929 -2023-02-13 18:41:28,775 - Epoch: [189][ 820/ 1207] Overall Loss 0.202120 Objective Loss 0.202120 LR 0.000125 Time 0.019921 -2023-02-13 18:41:28,966 - Epoch: [189][ 830/ 1207] Overall Loss 0.201891 Objective Loss 0.201891 LR 0.000125 Time 0.019911 -2023-02-13 18:41:29,158 - Epoch: [189][ 840/ 1207] Overall Loss 0.202108 Objective Loss 0.202108 LR 0.000125 Time 0.019901 -2023-02-13 18:41:29,348 - Epoch: [189][ 850/ 1207] Overall Loss 0.202240 Objective Loss 0.202240 LR 0.000125 Time 0.019890 -2023-02-13 18:41:29,540 - Epoch: [189][ 860/ 1207] Overall Loss 0.202160 Objective Loss 0.202160 LR 0.000125 Time 0.019882 -2023-02-13 18:41:29,732 - Epoch: [189][ 870/ 1207] Overall Loss 0.202359 Objective Loss 0.202359 LR 0.000125 Time 0.019873 -2023-02-13 18:41:29,923 - Epoch: [189][ 880/ 1207] Overall Loss 0.202422 Objective Loss 0.202422 LR 0.000125 Time 0.019865 -2023-02-13 18:41:30,115 - Epoch: [189][ 890/ 1207] Overall Loss 0.202436 Objective Loss 0.202436 LR 0.000125 Time 0.019857 -2023-02-13 18:41:30,308 - Epoch: [189][ 900/ 1207] Overall Loss 0.202285 Objective Loss 0.202285 LR 0.000125 Time 0.019850 -2023-02-13 18:41:30,499 - Epoch: [189][ 910/ 1207] Overall Loss 0.202246 Objective Loss 0.202246 LR 0.000125 Time 0.019842 -2023-02-13 18:41:30,690 - Epoch: [189][ 920/ 1207] Overall Loss 0.202226 Objective Loss 0.202226 LR 0.000125 Time 0.019833 -2023-02-13 18:41:30,883 - Epoch: [189][ 930/ 1207] Overall Loss 0.202273 Objective Loss 0.202273 LR 0.000125 Time 0.019827 -2023-02-13 18:41:31,076 - Epoch: [189][ 940/ 1207] Overall Loss 0.202384 Objective Loss 0.202384 LR 0.000125 Time 0.019821 -2023-02-13 18:41:31,268 - Epoch: [189][ 950/ 1207] Overall Loss 0.202414 Objective Loss 0.202414 LR 0.000125 Time 0.019814 -2023-02-13 18:41:31,459 - Epoch: [189][ 960/ 1207] Overall Loss 0.202355 Objective Loss 0.202355 LR 0.000125 Time 0.019806 -2023-02-13 18:41:31,651 - Epoch: [189][ 970/ 1207] Overall Loss 0.202511 Objective Loss 0.202511 LR 0.000125 Time 0.019800 -2023-02-13 18:41:31,843 - Epoch: [189][ 980/ 1207] Overall Loss 0.202753 Objective Loss 0.202753 LR 0.000125 Time 0.019793 -2023-02-13 18:41:32,035 - Epoch: [189][ 990/ 1207] Overall Loss 0.202976 Objective Loss 0.202976 LR 0.000125 Time 0.019787 -2023-02-13 18:41:32,225 - Epoch: [189][ 1000/ 1207] Overall Loss 0.203099 Objective Loss 0.203099 LR 0.000125 Time 0.019779 -2023-02-13 18:41:32,415 - Epoch: [189][ 1010/ 1207] Overall Loss 0.203219 Objective Loss 0.203219 LR 0.000125 Time 0.019771 -2023-02-13 18:41:32,607 - Epoch: [189][ 1020/ 1207] Overall Loss 0.203319 Objective Loss 0.203319 LR 0.000125 Time 0.019764 -2023-02-13 18:41:32,798 - Epoch: [189][ 1030/ 1207] Overall Loss 0.203338 Objective Loss 0.203338 LR 0.000125 Time 0.019757 -2023-02-13 18:41:32,987 - Epoch: [189][ 1040/ 1207] Overall Loss 0.203251 Objective Loss 0.203251 LR 0.000125 Time 0.019749 -2023-02-13 18:41:33,175 - Epoch: [189][ 1050/ 1207] Overall Loss 0.203064 Objective Loss 0.203064 LR 0.000125 Time 0.019740 -2023-02-13 18:41:33,363 - Epoch: [189][ 1060/ 1207] Overall Loss 0.203308 Objective Loss 0.203308 LR 0.000125 Time 0.019730 -2023-02-13 18:41:33,552 - Epoch: [189][ 1070/ 1207] Overall Loss 0.202936 Objective Loss 0.202936 LR 0.000125 Time 0.019722 -2023-02-13 18:41:33,739 - Epoch: [189][ 1080/ 1207] Overall Loss 0.202907 Objective Loss 0.202907 LR 0.000125 Time 0.019713 -2023-02-13 18:41:33,929 - Epoch: [189][ 1090/ 1207] Overall Loss 0.202859 Objective Loss 0.202859 LR 0.000125 Time 0.019705 -2023-02-13 18:41:34,118 - Epoch: [189][ 1100/ 1207] Overall Loss 0.202916 Objective Loss 0.202916 LR 0.000125 Time 0.019698 -2023-02-13 18:41:34,306 - Epoch: [189][ 1110/ 1207] Overall Loss 0.203089 Objective Loss 0.203089 LR 0.000125 Time 0.019689 -2023-02-13 18:41:34,495 - Epoch: [189][ 1120/ 1207] Overall Loss 0.203019 Objective Loss 0.203019 LR 0.000125 Time 0.019682 -2023-02-13 18:41:34,684 - Epoch: [189][ 1130/ 1207] Overall Loss 0.202968 Objective Loss 0.202968 LR 0.000125 Time 0.019675 -2023-02-13 18:41:34,872 - Epoch: [189][ 1140/ 1207] Overall Loss 0.202797 Objective Loss 0.202797 LR 0.000125 Time 0.019667 -2023-02-13 18:41:35,061 - Epoch: [189][ 1150/ 1207] Overall Loss 0.202921 Objective Loss 0.202921 LR 0.000125 Time 0.019660 -2023-02-13 18:41:35,249 - Epoch: [189][ 1160/ 1207] Overall Loss 0.202996 Objective Loss 0.202996 LR 0.000125 Time 0.019652 -2023-02-13 18:41:35,437 - Epoch: [189][ 1170/ 1207] Overall Loss 0.202984 Objective Loss 0.202984 LR 0.000125 Time 0.019645 -2023-02-13 18:41:35,624 - Epoch: [189][ 1180/ 1207] Overall Loss 0.202873 Objective Loss 0.202873 LR 0.000125 Time 0.019636 -2023-02-13 18:41:35,813 - Epoch: [189][ 1190/ 1207] Overall Loss 0.202903 Objective Loss 0.202903 LR 0.000125 Time 0.019630 -2023-02-13 18:41:36,054 - Epoch: [189][ 1200/ 1207] Overall Loss 0.202880 Objective Loss 0.202880 LR 0.000125 Time 0.019667 -2023-02-13 18:41:36,171 - Epoch: [189][ 1207/ 1207] Overall Loss 0.202875 Objective Loss 0.202875 Top1 89.329268 Top5 99.695122 LR 0.000125 Time 0.019650 -2023-02-13 18:41:36,250 - --- validate (epoch=189)----------- -2023-02-13 18:41:36,250 - 34311 samples (256 per mini-batch) -2023-02-13 18:41:36,653 - Epoch: [189][ 10/ 135] Loss 0.287779 Top1 84.921875 Top5 98.046875 -2023-02-13 18:41:36,784 - Epoch: [189][ 20/ 135] Loss 0.287779 Top1 85.625000 Top5 98.164062 -2023-02-13 18:41:36,908 - Epoch: [189][ 30/ 135] Loss 0.294196 Top1 85.182292 Top5 97.890625 -2023-02-13 18:41:37,032 - Epoch: [189][ 40/ 135] Loss 0.298417 Top1 84.902344 Top5 97.890625 -2023-02-13 18:41:37,156 - Epoch: [189][ 50/ 135] Loss 0.295845 Top1 85.234375 Top5 97.921875 -2023-02-13 18:41:37,280 - Epoch: [189][ 60/ 135] Loss 0.296051 Top1 85.299479 Top5 97.897135 -2023-02-13 18:41:37,402 - Epoch: [189][ 70/ 135] Loss 0.293429 Top1 85.295759 Top5 97.912946 -2023-02-13 18:41:37,532 - Epoch: [189][ 80/ 135] Loss 0.289523 Top1 85.390625 Top5 97.954102 -2023-02-13 18:41:37,661 - Epoch: [189][ 90/ 135] Loss 0.288653 Top1 85.412326 Top5 98.012153 -2023-02-13 18:41:37,791 - Epoch: [189][ 100/ 135] Loss 0.287776 Top1 85.488281 Top5 97.976562 -2023-02-13 18:41:37,921 - Epoch: [189][ 110/ 135] Loss 0.290654 Top1 85.436790 Top5 97.936790 -2023-02-13 18:41:38,050 - Epoch: [189][ 120/ 135] Loss 0.289816 Top1 85.462240 Top5 97.923177 -2023-02-13 18:41:38,183 - Epoch: [189][ 130/ 135] Loss 0.291253 Top1 85.378606 Top5 97.905649 -2023-02-13 18:41:38,228 - Epoch: [189][ 135/ 135] Loss 0.290573 Top1 85.389525 Top5 97.927778 -2023-02-13 18:41:38,301 - ==> Top1: 85.390 Top5: 97.928 Loss: 0.291 - -2023-02-13 18:41:38,302 - ==> Confusion: -[[ 872 5 8 1 7 2 0 1 3 39 0 3 1 6 6 2 2 1 0 2 6] - [ 3 954 3 1 9 21 2 12 3 0 1 0 2 0 0 1 7 1 2 2 9] - [ 9 7 966 13 2 1 14 11 0 1 3 3 1 2 3 3 2 1 7 3 6] - [ 5 1 21 908 1 6 0 2 3 2 16 0 5 0 16 0 2 4 17 0 7] - [ 12 11 0 0 997 10 1 4 1 0 0 6 3 2 5 5 2 1 0 3 3] - [ 0 11 1 3 4 985 3 15 3 7 1 7 1 14 1 2 3 2 1 2 4] - [ 4 1 10 3 0 4 1052 7 0 1 0 0 2 1 0 1 3 1 1 3 5] - [ 2 10 13 3 0 24 2 939 0 2 0 6 2 3 0 0 2 0 8 6 2] - [ 14 2 0 1 1 0 1 0 929 29 5 3 0 8 8 2 1 0 5 0 0] - [ 77 1 3 0 8 0 0 3 36 856 1 0 0 15 3 1 2 0 1 0 5] - [ 2 0 1 5 1 1 1 5 17 1 993 1 1 6 3 0 2 0 5 0 6] - [ 3 3 2 0 0 10 1 6 0 3 0 926 21 8 1 4 1 8 1 5 2] - [ 0 0 1 8 1 4 0 0 3 0 0 32 874 0 1 6 4 12 3 1 9] - [ 4 3 5 2 7 7 1 1 8 18 10 5 0 933 5 4 3 0 2 1 5] - [ 7 2 1 15 7 5 0 1 22 6 3 0 4 1 1000 1 2 4 4 0 7] - [ 3 1 11 1 6 1 3 1 0 0 0 6 8 2 0 971 10 8 0 8 6] - [ 2 9 0 1 9 4 0 0 2 0 0 0 3 3 2 8 999 2 3 3 11] - [ 5 2 0 4 2 1 3 0 1 1 1 13 12 1 0 9 0 990 1 1 4] - [ 3 7 5 8 0 1 0 26 6 0 5 1 3 0 7 1 0 3 1005 1 4] - [ 1 3 1 0 1 5 7 13 1 0 0 13 3 2 1 7 5 3 0 1076 6] - [ 147 251 257 108 123 222 85 189 118 67 186 111 287 272 132 64 235 99 153 256 10072]] - -2023-02-13 18:41:38,303 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:41:38,303 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:41:38,309 - - -2023-02-13 18:41:38,309 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:41:39,303 - Epoch: [190][ 10/ 1207] Overall Loss 0.208325 Objective Loss 0.208325 LR 0.000063 Time 0.099286 -2023-02-13 18:41:39,503 - Epoch: [190][ 20/ 1207] Overall Loss 0.204616 Objective Loss 0.204616 LR 0.000063 Time 0.059621 -2023-02-13 18:41:39,695 - Epoch: [190][ 30/ 1207] Overall Loss 0.206760 Objective Loss 0.206760 LR 0.000063 Time 0.046131 -2023-02-13 18:41:39,890 - Epoch: [190][ 40/ 1207] Overall Loss 0.201323 Objective Loss 0.201323 LR 0.000063 Time 0.039461 -2023-02-13 18:41:40,080 - Epoch: [190][ 50/ 1207] Overall Loss 0.206289 Objective Loss 0.206289 LR 0.000063 Time 0.035377 -2023-02-13 18:41:40,273 - Epoch: [190][ 60/ 1207] Overall Loss 0.205426 Objective Loss 0.205426 LR 0.000063 Time 0.032691 -2023-02-13 18:41:40,464 - Epoch: [190][ 70/ 1207] Overall Loss 0.205478 Objective Loss 0.205478 LR 0.000063 Time 0.030747 -2023-02-13 18:41:40,657 - Epoch: [190][ 80/ 1207] Overall Loss 0.205228 Objective Loss 0.205228 LR 0.000063 Time 0.029311 -2023-02-13 18:41:40,849 - Epoch: [190][ 90/ 1207] Overall Loss 0.202358 Objective Loss 0.202358 LR 0.000063 Time 0.028178 -2023-02-13 18:41:41,043 - Epoch: [190][ 100/ 1207] Overall Loss 0.202658 Objective Loss 0.202658 LR 0.000063 Time 0.027302 -2023-02-13 18:41:41,238 - Epoch: [190][ 110/ 1207] Overall Loss 0.201164 Objective Loss 0.201164 LR 0.000063 Time 0.026588 -2023-02-13 18:41:41,434 - Epoch: [190][ 120/ 1207] Overall Loss 0.201286 Objective Loss 0.201286 LR 0.000063 Time 0.025996 -2023-02-13 18:41:41,629 - Epoch: [190][ 130/ 1207] Overall Loss 0.201589 Objective Loss 0.201589 LR 0.000063 Time 0.025495 -2023-02-13 18:41:41,825 - Epoch: [190][ 140/ 1207] Overall Loss 0.201813 Objective Loss 0.201813 LR 0.000063 Time 0.025070 -2023-02-13 18:41:42,021 - Epoch: [190][ 150/ 1207] Overall Loss 0.201379 Objective Loss 0.201379 LR 0.000063 Time 0.024704 -2023-02-13 18:41:42,216 - Epoch: [190][ 160/ 1207] Overall Loss 0.201124 Objective Loss 0.201124 LR 0.000063 Time 0.024379 -2023-02-13 18:41:42,412 - Epoch: [190][ 170/ 1207] Overall Loss 0.201138 Objective Loss 0.201138 LR 0.000063 Time 0.024092 -2023-02-13 18:41:42,607 - Epoch: [190][ 180/ 1207] Overall Loss 0.201209 Objective Loss 0.201209 LR 0.000063 Time 0.023836 -2023-02-13 18:41:42,804 - Epoch: [190][ 190/ 1207] Overall Loss 0.199518 Objective Loss 0.199518 LR 0.000063 Time 0.023618 -2023-02-13 18:41:43,000 - Epoch: [190][ 200/ 1207] Overall Loss 0.198979 Objective Loss 0.198979 LR 0.000063 Time 0.023414 -2023-02-13 18:41:43,195 - Epoch: [190][ 210/ 1207] Overall Loss 0.199648 Objective Loss 0.199648 LR 0.000063 Time 0.023226 -2023-02-13 18:41:43,391 - Epoch: [190][ 220/ 1207] Overall Loss 0.198399 Objective Loss 0.198399 LR 0.000063 Time 0.023059 -2023-02-13 18:41:43,586 - Epoch: [190][ 230/ 1207] Overall Loss 0.198431 Objective Loss 0.198431 LR 0.000063 Time 0.022903 -2023-02-13 18:41:43,779 - Epoch: [190][ 240/ 1207] Overall Loss 0.198015 Objective Loss 0.198015 LR 0.000063 Time 0.022754 -2023-02-13 18:41:43,970 - Epoch: [190][ 250/ 1207] Overall Loss 0.198040 Objective Loss 0.198040 LR 0.000063 Time 0.022604 -2023-02-13 18:41:44,159 - Epoch: [190][ 260/ 1207] Overall Loss 0.197310 Objective Loss 0.197310 LR 0.000063 Time 0.022461 -2023-02-13 18:41:44,350 - Epoch: [190][ 270/ 1207] Overall Loss 0.197207 Objective Loss 0.197207 LR 0.000063 Time 0.022334 -2023-02-13 18:41:44,539 - Epoch: [190][ 280/ 1207] Overall Loss 0.197930 Objective Loss 0.197930 LR 0.000063 Time 0.022211 -2023-02-13 18:41:44,730 - Epoch: [190][ 290/ 1207] Overall Loss 0.197317 Objective Loss 0.197317 LR 0.000063 Time 0.022104 -2023-02-13 18:41:44,921 - Epoch: [190][ 300/ 1207] Overall Loss 0.197298 Objective Loss 0.197298 LR 0.000063 Time 0.021999 -2023-02-13 18:41:45,111 - Epoch: [190][ 310/ 1207] Overall Loss 0.197995 Objective Loss 0.197995 LR 0.000063 Time 0.021904 -2023-02-13 18:41:45,301 - Epoch: [190][ 320/ 1207] Overall Loss 0.198419 Objective Loss 0.198419 LR 0.000063 Time 0.021812 -2023-02-13 18:41:45,492 - Epoch: [190][ 330/ 1207] Overall Loss 0.198272 Objective Loss 0.198272 LR 0.000063 Time 0.021728 -2023-02-13 18:41:45,682 - Epoch: [190][ 340/ 1207] Overall Loss 0.198673 Objective Loss 0.198673 LR 0.000063 Time 0.021646 -2023-02-13 18:41:45,874 - Epoch: [190][ 350/ 1207] Overall Loss 0.198366 Objective Loss 0.198366 LR 0.000063 Time 0.021575 -2023-02-13 18:41:46,065 - Epoch: [190][ 360/ 1207] Overall Loss 0.198175 Objective Loss 0.198175 LR 0.000063 Time 0.021505 -2023-02-13 18:41:46,256 - Epoch: [190][ 370/ 1207] Overall Loss 0.198773 Objective Loss 0.198773 LR 0.000063 Time 0.021440 -2023-02-13 18:41:46,446 - Epoch: [190][ 380/ 1207] Overall Loss 0.198572 Objective Loss 0.198572 LR 0.000063 Time 0.021373 -2023-02-13 18:41:46,637 - Epoch: [190][ 390/ 1207] Overall Loss 0.198885 Objective Loss 0.198885 LR 0.000063 Time 0.021314 -2023-02-13 18:41:46,827 - Epoch: [190][ 400/ 1207] Overall Loss 0.198833 Objective Loss 0.198833 LR 0.000063 Time 0.021257 -2023-02-13 18:41:47,019 - Epoch: [190][ 410/ 1207] Overall Loss 0.198507 Objective Loss 0.198507 LR 0.000063 Time 0.021205 -2023-02-13 18:41:47,209 - Epoch: [190][ 420/ 1207] Overall Loss 0.198601 Objective Loss 0.198601 LR 0.000063 Time 0.021151 -2023-02-13 18:41:47,399 - Epoch: [190][ 430/ 1207] Overall Loss 0.198386 Objective Loss 0.198386 LR 0.000063 Time 0.021101 -2023-02-13 18:41:47,590 - Epoch: [190][ 440/ 1207] Overall Loss 0.199049 Objective Loss 0.199049 LR 0.000063 Time 0.021055 -2023-02-13 18:41:47,781 - Epoch: [190][ 450/ 1207] Overall Loss 0.198966 Objective Loss 0.198966 LR 0.000063 Time 0.021010 -2023-02-13 18:41:47,971 - Epoch: [190][ 460/ 1207] Overall Loss 0.198846 Objective Loss 0.198846 LR 0.000063 Time 0.020966 -2023-02-13 18:41:48,160 - Epoch: [190][ 470/ 1207] Overall Loss 0.198526 Objective Loss 0.198526 LR 0.000063 Time 0.020920 -2023-02-13 18:41:48,348 - Epoch: [190][ 480/ 1207] Overall Loss 0.198462 Objective Loss 0.198462 LR 0.000063 Time 0.020876 -2023-02-13 18:41:48,536 - Epoch: [190][ 490/ 1207] Overall Loss 0.198674 Objective Loss 0.198674 LR 0.000063 Time 0.020833 -2023-02-13 18:41:48,725 - Epoch: [190][ 500/ 1207] Overall Loss 0.198510 Objective Loss 0.198510 LR 0.000063 Time 0.020793 -2023-02-13 18:41:48,915 - Epoch: [190][ 510/ 1207] Overall Loss 0.198726 Objective Loss 0.198726 LR 0.000063 Time 0.020758 -2023-02-13 18:41:49,104 - Epoch: [190][ 520/ 1207] Overall Loss 0.198464 Objective Loss 0.198464 LR 0.000063 Time 0.020721 -2023-02-13 18:41:49,293 - Epoch: [190][ 530/ 1207] Overall Loss 0.198817 Objective Loss 0.198817 LR 0.000063 Time 0.020687 -2023-02-13 18:41:49,481 - Epoch: [190][ 540/ 1207] Overall Loss 0.199048 Objective Loss 0.199048 LR 0.000063 Time 0.020651 -2023-02-13 18:41:49,670 - Epoch: [190][ 550/ 1207] Overall Loss 0.199075 Objective Loss 0.199075 LR 0.000063 Time 0.020619 -2023-02-13 18:41:49,859 - Epoch: [190][ 560/ 1207] Overall Loss 0.199177 Objective Loss 0.199177 LR 0.000063 Time 0.020587 -2023-02-13 18:41:50,047 - Epoch: [190][ 570/ 1207] Overall Loss 0.199246 Objective Loss 0.199246 LR 0.000063 Time 0.020555 -2023-02-13 18:41:50,236 - Epoch: [190][ 580/ 1207] Overall Loss 0.198990 Objective Loss 0.198990 LR 0.000063 Time 0.020525 -2023-02-13 18:41:50,424 - Epoch: [190][ 590/ 1207] Overall Loss 0.199048 Objective Loss 0.199048 LR 0.000063 Time 0.020496 -2023-02-13 18:41:50,613 - Epoch: [190][ 600/ 1207] Overall Loss 0.198988 Objective Loss 0.198988 LR 0.000063 Time 0.020468 -2023-02-13 18:41:50,802 - Epoch: [190][ 610/ 1207] Overall Loss 0.198876 Objective Loss 0.198876 LR 0.000063 Time 0.020442 -2023-02-13 18:41:50,991 - Epoch: [190][ 620/ 1207] Overall Loss 0.198843 Objective Loss 0.198843 LR 0.000063 Time 0.020417 -2023-02-13 18:41:51,179 - Epoch: [190][ 630/ 1207] Overall Loss 0.199104 Objective Loss 0.199104 LR 0.000063 Time 0.020390 -2023-02-13 18:41:51,366 - Epoch: [190][ 640/ 1207] Overall Loss 0.199130 Objective Loss 0.199130 LR 0.000063 Time 0.020364 -2023-02-13 18:41:51,554 - Epoch: [190][ 650/ 1207] Overall Loss 0.199272 Objective Loss 0.199272 LR 0.000063 Time 0.020339 -2023-02-13 18:41:51,742 - Epoch: [190][ 660/ 1207] Overall Loss 0.198829 Objective Loss 0.198829 LR 0.000063 Time 0.020315 -2023-02-13 18:41:51,932 - Epoch: [190][ 670/ 1207] Overall Loss 0.198941 Objective Loss 0.198941 LR 0.000063 Time 0.020295 -2023-02-13 18:41:52,121 - Epoch: [190][ 680/ 1207] Overall Loss 0.198620 Objective Loss 0.198620 LR 0.000063 Time 0.020273 -2023-02-13 18:41:52,309 - Epoch: [190][ 690/ 1207] Overall Loss 0.198484 Objective Loss 0.198484 LR 0.000063 Time 0.020252 -2023-02-13 18:41:52,497 - Epoch: [190][ 700/ 1207] Overall Loss 0.198681 Objective Loss 0.198681 LR 0.000063 Time 0.020231 -2023-02-13 18:41:52,685 - Epoch: [190][ 710/ 1207] Overall Loss 0.198726 Objective Loss 0.198726 LR 0.000063 Time 0.020210 -2023-02-13 18:41:52,875 - Epoch: [190][ 720/ 1207] Overall Loss 0.198768 Objective Loss 0.198768 LR 0.000063 Time 0.020192 -2023-02-13 18:41:53,063 - Epoch: [190][ 730/ 1207] Overall Loss 0.198643 Objective Loss 0.198643 LR 0.000063 Time 0.020173 -2023-02-13 18:41:53,252 - Epoch: [190][ 740/ 1207] Overall Loss 0.198620 Objective Loss 0.198620 LR 0.000063 Time 0.020155 -2023-02-13 18:41:53,440 - Epoch: [190][ 750/ 1207] Overall Loss 0.198667 Objective Loss 0.198667 LR 0.000063 Time 0.020136 -2023-02-13 18:41:53,628 - Epoch: [190][ 760/ 1207] Overall Loss 0.198690 Objective Loss 0.198690 LR 0.000063 Time 0.020118 -2023-02-13 18:41:53,816 - Epoch: [190][ 770/ 1207] Overall Loss 0.198412 Objective Loss 0.198412 LR 0.000063 Time 0.020101 -2023-02-13 18:41:54,004 - Epoch: [190][ 780/ 1207] Overall Loss 0.198386 Objective Loss 0.198386 LR 0.000063 Time 0.020084 -2023-02-13 18:41:54,193 - Epoch: [190][ 790/ 1207] Overall Loss 0.198411 Objective Loss 0.198411 LR 0.000063 Time 0.020068 -2023-02-13 18:41:54,381 - Epoch: [190][ 800/ 1207] Overall Loss 0.198385 Objective Loss 0.198385 LR 0.000063 Time 0.020052 -2023-02-13 18:41:54,569 - Epoch: [190][ 810/ 1207] Overall Loss 0.198365 Objective Loss 0.198365 LR 0.000063 Time 0.020036 -2023-02-13 18:41:54,758 - Epoch: [190][ 820/ 1207] Overall Loss 0.198584 Objective Loss 0.198584 LR 0.000063 Time 0.020021 -2023-02-13 18:41:54,947 - Epoch: [190][ 830/ 1207] Overall Loss 0.198622 Objective Loss 0.198622 LR 0.000063 Time 0.020008 -2023-02-13 18:41:55,136 - Epoch: [190][ 840/ 1207] Overall Loss 0.198683 Objective Loss 0.198683 LR 0.000063 Time 0.019994 -2023-02-13 18:41:55,325 - Epoch: [190][ 850/ 1207] Overall Loss 0.198805 Objective Loss 0.198805 LR 0.000063 Time 0.019980 -2023-02-13 18:41:55,513 - Epoch: [190][ 860/ 1207] Overall Loss 0.198768 Objective Loss 0.198768 LR 0.000063 Time 0.019967 -2023-02-13 18:41:55,701 - Epoch: [190][ 870/ 1207] Overall Loss 0.198640 Objective Loss 0.198640 LR 0.000063 Time 0.019953 -2023-02-13 18:41:55,891 - Epoch: [190][ 880/ 1207] Overall Loss 0.198382 Objective Loss 0.198382 LR 0.000063 Time 0.019941 -2023-02-13 18:41:56,081 - Epoch: [190][ 890/ 1207] Overall Loss 0.198182 Objective Loss 0.198182 LR 0.000063 Time 0.019931 -2023-02-13 18:41:56,270 - Epoch: [190][ 900/ 1207] Overall Loss 0.197983 Objective Loss 0.197983 LR 0.000063 Time 0.019918 -2023-02-13 18:41:56,458 - Epoch: [190][ 910/ 1207] Overall Loss 0.197607 Objective Loss 0.197607 LR 0.000063 Time 0.019906 -2023-02-13 18:41:56,646 - Epoch: [190][ 920/ 1207] Overall Loss 0.197771 Objective Loss 0.197771 LR 0.000063 Time 0.019894 -2023-02-13 18:41:56,836 - Epoch: [190][ 930/ 1207] Overall Loss 0.197714 Objective Loss 0.197714 LR 0.000063 Time 0.019884 -2023-02-13 18:41:57,025 - Epoch: [190][ 940/ 1207] Overall Loss 0.197878 Objective Loss 0.197878 LR 0.000063 Time 0.019873 -2023-02-13 18:41:57,213 - Epoch: [190][ 950/ 1207] Overall Loss 0.197949 Objective Loss 0.197949 LR 0.000063 Time 0.019862 -2023-02-13 18:41:57,402 - Epoch: [190][ 960/ 1207] Overall Loss 0.197910 Objective Loss 0.197910 LR 0.000063 Time 0.019850 -2023-02-13 18:41:57,591 - Epoch: [190][ 970/ 1207] Overall Loss 0.197625 Objective Loss 0.197625 LR 0.000063 Time 0.019840 -2023-02-13 18:41:57,779 - Epoch: [190][ 980/ 1207] Overall Loss 0.197462 Objective Loss 0.197462 LR 0.000063 Time 0.019830 -2023-02-13 18:41:57,968 - Epoch: [190][ 990/ 1207] Overall Loss 0.197303 Objective Loss 0.197303 LR 0.000063 Time 0.019820 -2023-02-13 18:41:58,157 - Epoch: [190][ 1000/ 1207] Overall Loss 0.197313 Objective Loss 0.197313 LR 0.000063 Time 0.019810 -2023-02-13 18:41:58,347 - Epoch: [190][ 1010/ 1207] Overall Loss 0.197369 Objective Loss 0.197369 LR 0.000063 Time 0.019801 -2023-02-13 18:41:58,535 - Epoch: [190][ 1020/ 1207] Overall Loss 0.197282 Objective Loss 0.197282 LR 0.000063 Time 0.019791 -2023-02-13 18:41:58,723 - Epoch: [190][ 1030/ 1207] Overall Loss 0.197093 Objective Loss 0.197093 LR 0.000063 Time 0.019782 -2023-02-13 18:41:58,912 - Epoch: [190][ 1040/ 1207] Overall Loss 0.197065 Objective Loss 0.197065 LR 0.000063 Time 0.019773 -2023-02-13 18:41:59,101 - Epoch: [190][ 1050/ 1207] Overall Loss 0.196976 Objective Loss 0.196976 LR 0.000063 Time 0.019764 -2023-02-13 18:41:59,289 - Epoch: [190][ 1060/ 1207] Overall Loss 0.196972 Objective Loss 0.196972 LR 0.000063 Time 0.019755 -2023-02-13 18:41:59,478 - Epoch: [190][ 1070/ 1207] Overall Loss 0.196947 Objective Loss 0.196947 LR 0.000063 Time 0.019746 -2023-02-13 18:41:59,666 - Epoch: [190][ 1080/ 1207] Overall Loss 0.196822 Objective Loss 0.196822 LR 0.000063 Time 0.019737 -2023-02-13 18:41:59,856 - Epoch: [190][ 1090/ 1207] Overall Loss 0.196882 Objective Loss 0.196882 LR 0.000063 Time 0.019730 -2023-02-13 18:42:00,045 - Epoch: [190][ 1100/ 1207] Overall Loss 0.196928 Objective Loss 0.196928 LR 0.000063 Time 0.019722 -2023-02-13 18:42:00,234 - Epoch: [190][ 1110/ 1207] Overall Loss 0.197063 Objective Loss 0.197063 LR 0.000063 Time 0.019714 -2023-02-13 18:42:00,422 - Epoch: [190][ 1120/ 1207] Overall Loss 0.197114 Objective Loss 0.197114 LR 0.000063 Time 0.019706 -2023-02-13 18:42:00,611 - Epoch: [190][ 1130/ 1207] Overall Loss 0.197056 Objective Loss 0.197056 LR 0.000063 Time 0.019698 -2023-02-13 18:42:00,799 - Epoch: [190][ 1140/ 1207] Overall Loss 0.197271 Objective Loss 0.197271 LR 0.000063 Time 0.019691 -2023-02-13 18:42:00,990 - Epoch: [190][ 1150/ 1207] Overall Loss 0.197344 Objective Loss 0.197344 LR 0.000063 Time 0.019684 -2023-02-13 18:42:01,179 - Epoch: [190][ 1160/ 1207] Overall Loss 0.197424 Objective Loss 0.197424 LR 0.000063 Time 0.019677 -2023-02-13 18:42:01,368 - Epoch: [190][ 1170/ 1207] Overall Loss 0.197610 Objective Loss 0.197610 LR 0.000063 Time 0.019671 -2023-02-13 18:42:01,556 - Epoch: [190][ 1180/ 1207] Overall Loss 0.197762 Objective Loss 0.197762 LR 0.000063 Time 0.019663 -2023-02-13 18:42:01,745 - Epoch: [190][ 1190/ 1207] Overall Loss 0.197719 Objective Loss 0.197719 LR 0.000063 Time 0.019656 -2023-02-13 18:42:01,986 - Epoch: [190][ 1200/ 1207] Overall Loss 0.197647 Objective Loss 0.197647 LR 0.000063 Time 0.019693 -2023-02-13 18:42:02,101 - Epoch: [190][ 1207/ 1207] Overall Loss 0.197618 Objective Loss 0.197618 Top1 88.414634 Top5 97.560976 LR 0.000063 Time 0.019674 -2023-02-13 18:42:02,173 - --- validate (epoch=190)----------- -2023-02-13 18:42:02,173 - 34311 samples (256 per mini-batch) -2023-02-13 18:42:02,576 - Epoch: [190][ 10/ 135] Loss 0.280989 Top1 86.054688 Top5 98.125000 -2023-02-13 18:42:02,708 - Epoch: [190][ 20/ 135] Loss 0.282033 Top1 86.328125 Top5 98.066406 -2023-02-13 18:42:02,837 - Epoch: [190][ 30/ 135] Loss 0.286389 Top1 86.145833 Top5 98.046875 -2023-02-13 18:42:02,966 - Epoch: [190][ 40/ 135] Loss 0.282267 Top1 86.240234 Top5 98.007812 -2023-02-13 18:42:03,097 - Epoch: [190][ 50/ 135] Loss 0.285208 Top1 85.953125 Top5 97.835938 -2023-02-13 18:42:03,221 - Epoch: [190][ 60/ 135] Loss 0.285885 Top1 85.950521 Top5 97.858073 -2023-02-13 18:42:03,348 - Epoch: [190][ 70/ 135] Loss 0.282918 Top1 85.965402 Top5 97.885045 -2023-02-13 18:42:03,476 - Epoch: [190][ 80/ 135] Loss 0.281877 Top1 85.883789 Top5 97.890625 -2023-02-13 18:42:03,603 - Epoch: [190][ 90/ 135] Loss 0.285478 Top1 85.833333 Top5 97.873264 -2023-02-13 18:42:03,729 - Epoch: [190][ 100/ 135] Loss 0.286683 Top1 85.812500 Top5 97.851562 -2023-02-13 18:42:03,855 - Epoch: [190][ 110/ 135] Loss 0.286029 Top1 85.784801 Top5 97.869318 -2023-02-13 18:42:03,982 - Epoch: [190][ 120/ 135] Loss 0.289040 Top1 85.794271 Top5 97.880859 -2023-02-13 18:42:04,110 - Epoch: [190][ 130/ 135] Loss 0.290590 Top1 85.718149 Top5 97.860577 -2023-02-13 18:42:04,156 - Epoch: [190][ 135/ 135] Loss 0.289870 Top1 85.698464 Top5 97.866573 -2023-02-13 18:42:04,224 - ==> Top1: 85.698 Top5: 97.867 Loss: 0.290 - -2023-02-13 18:42:04,225 - ==> Confusion: -[[ 872 5 6 1 9 1 0 2 3 39 0 2 1 4 8 3 1 2 0 2 6] - [ 3 949 2 3 13 20 2 16 2 0 2 0 1 0 0 2 4 1 2 3 8] - [ 7 5 969 12 1 2 16 12 2 1 2 3 0 2 4 4 3 3 5 0 5] - [ 6 1 21 905 0 6 0 2 4 2 12 0 7 0 22 0 3 5 14 0 6] - [ 12 10 0 0 997 9 1 2 1 0 0 5 1 2 6 5 5 1 2 4 3] - [ 0 11 0 6 4 985 3 11 2 5 2 9 2 13 0 3 7 1 2 1 3] - [ 3 2 13 1 0 6 1051 1 0 3 1 0 2 2 1 1 2 1 1 5 3] - [ 3 9 11 2 1 28 3 928 0 1 0 6 5 3 0 0 1 0 13 6 4] - [ 14 3 0 1 2 0 0 1 906 37 6 4 1 11 14 1 2 0 5 0 1] - [ 81 0 2 0 10 1 0 2 33 854 0 1 0 15 4 1 3 0 1 0 4] - [ 2 1 5 7 2 4 2 4 12 2 985 2 1 6 2 0 1 1 7 0 5] - [ 2 2 2 0 1 9 2 6 2 3 0 919 22 7 2 5 2 7 1 8 3] - [ 1 0 0 9 0 5 0 0 3 2 0 28 882 0 2 5 3 8 2 1 8] - [ 5 3 0 1 5 10 1 1 8 16 10 4 1 937 7 3 3 0 0 1 8] - [ 6 1 0 15 4 4 0 1 16 8 5 1 2 4 1007 0 2 3 6 0 7] - [ 5 2 9 1 4 0 4 2 1 0 0 2 9 3 0 971 13 7 0 7 6] - [ 0 6 0 2 9 2 0 0 2 0 0 1 1 3 3 8 1008 0 2 3 11] - [ 3 3 1 4 1 1 1 0 0 1 0 12 13 1 0 12 0 991 0 2 5] - [ 4 5 5 7 0 2 0 21 4 0 5 1 1 0 10 1 1 4 1010 3 2] - [ 2 3 0 0 0 4 6 9 1 0 0 12 3 1 1 7 3 3 0 1086 7] - [ 165 217 244 111 135 204 70 158 87 81 165 100 288 270 159 74 257 86 157 214 10192]] - -2023-02-13 18:42:04,227 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:42:04,227 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:42:04,232 - - -2023-02-13 18:42:04,233 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:42:05,141 - Epoch: [191][ 10/ 1207] Overall Loss 0.192077 Objective Loss 0.192077 LR 0.000063 Time 0.090763 -2023-02-13 18:42:05,344 - Epoch: [191][ 20/ 1207] Overall Loss 0.187550 Objective Loss 0.187550 LR 0.000063 Time 0.055538 -2023-02-13 18:42:05,537 - Epoch: [191][ 30/ 1207] Overall Loss 0.196385 Objective Loss 0.196385 LR 0.000063 Time 0.043431 -2023-02-13 18:42:05,730 - Epoch: [191][ 40/ 1207] Overall Loss 0.197023 Objective Loss 0.197023 LR 0.000063 Time 0.037400 -2023-02-13 18:42:05,924 - Epoch: [191][ 50/ 1207] Overall Loss 0.197306 Objective Loss 0.197306 LR 0.000063 Time 0.033786 -2023-02-13 18:42:06,119 - Epoch: [191][ 60/ 1207] Overall Loss 0.196737 Objective Loss 0.196737 LR 0.000063 Time 0.031396 -2023-02-13 18:42:06,311 - Epoch: [191][ 70/ 1207] Overall Loss 0.198309 Objective Loss 0.198309 LR 0.000063 Time 0.029650 -2023-02-13 18:42:06,505 - Epoch: [191][ 80/ 1207] Overall Loss 0.195831 Objective Loss 0.195831 LR 0.000063 Time 0.028364 -2023-02-13 18:42:06,697 - Epoch: [191][ 90/ 1207] Overall Loss 0.193674 Objective Loss 0.193674 LR 0.000063 Time 0.027346 -2023-02-13 18:42:06,893 - Epoch: [191][ 100/ 1207] Overall Loss 0.195916 Objective Loss 0.195916 LR 0.000063 Time 0.026566 -2023-02-13 18:42:07,088 - Epoch: [191][ 110/ 1207] Overall Loss 0.196281 Objective Loss 0.196281 LR 0.000063 Time 0.025923 -2023-02-13 18:42:07,285 - Epoch: [191][ 120/ 1207] Overall Loss 0.197339 Objective Loss 0.197339 LR 0.000063 Time 0.025396 -2023-02-13 18:42:07,480 - Epoch: [191][ 130/ 1207] Overall Loss 0.198447 Objective Loss 0.198447 LR 0.000063 Time 0.024940 -2023-02-13 18:42:07,675 - Epoch: [191][ 140/ 1207] Overall Loss 0.198899 Objective Loss 0.198899 LR 0.000063 Time 0.024553 -2023-02-13 18:42:07,871 - Epoch: [191][ 150/ 1207] Overall Loss 0.198332 Objective Loss 0.198332 LR 0.000063 Time 0.024215 -2023-02-13 18:42:08,066 - Epoch: [191][ 160/ 1207] Overall Loss 0.197700 Objective Loss 0.197700 LR 0.000063 Time 0.023921 -2023-02-13 18:42:08,261 - Epoch: [191][ 170/ 1207] Overall Loss 0.196891 Objective Loss 0.196891 LR 0.000063 Time 0.023656 -2023-02-13 18:42:08,456 - Epoch: [191][ 180/ 1207] Overall Loss 0.197130 Objective Loss 0.197130 LR 0.000063 Time 0.023426 -2023-02-13 18:42:08,651 - Epoch: [191][ 190/ 1207] Overall Loss 0.197916 Objective Loss 0.197916 LR 0.000063 Time 0.023215 -2023-02-13 18:42:08,847 - Epoch: [191][ 200/ 1207] Overall Loss 0.198362 Objective Loss 0.198362 LR 0.000063 Time 0.023034 -2023-02-13 18:42:09,042 - Epoch: [191][ 210/ 1207] Overall Loss 0.198168 Objective Loss 0.198168 LR 0.000063 Time 0.022862 -2023-02-13 18:42:09,237 - Epoch: [191][ 220/ 1207] Overall Loss 0.198964 Objective Loss 0.198964 LR 0.000063 Time 0.022709 -2023-02-13 18:42:09,431 - Epoch: [191][ 230/ 1207] Overall Loss 0.198242 Objective Loss 0.198242 LR 0.000063 Time 0.022564 -2023-02-13 18:42:09,626 - Epoch: [191][ 240/ 1207] Overall Loss 0.197154 Objective Loss 0.197154 LR 0.000063 Time 0.022435 -2023-02-13 18:42:09,823 - Epoch: [191][ 250/ 1207] Overall Loss 0.197451 Objective Loss 0.197451 LR 0.000063 Time 0.022324 -2023-02-13 18:42:10,017 - Epoch: [191][ 260/ 1207] Overall Loss 0.197715 Objective Loss 0.197715 LR 0.000063 Time 0.022212 -2023-02-13 18:42:10,214 - Epoch: [191][ 270/ 1207] Overall Loss 0.197958 Objective Loss 0.197958 LR 0.000063 Time 0.022116 -2023-02-13 18:42:10,407 - Epoch: [191][ 280/ 1207] Overall Loss 0.196975 Objective Loss 0.196975 LR 0.000063 Time 0.022015 -2023-02-13 18:42:10,604 - Epoch: [191][ 290/ 1207] Overall Loss 0.197676 Objective Loss 0.197676 LR 0.000063 Time 0.021933 -2023-02-13 18:42:10,798 - Epoch: [191][ 300/ 1207] Overall Loss 0.198378 Objective Loss 0.198378 LR 0.000063 Time 0.021846 -2023-02-13 18:42:10,997 - Epoch: [191][ 310/ 1207] Overall Loss 0.197849 Objective Loss 0.197849 LR 0.000063 Time 0.021782 -2023-02-13 18:42:11,191 - Epoch: [191][ 320/ 1207] Overall Loss 0.197902 Objective Loss 0.197902 LR 0.000063 Time 0.021707 -2023-02-13 18:42:11,387 - Epoch: [191][ 330/ 1207] Overall Loss 0.197700 Objective Loss 0.197700 LR 0.000063 Time 0.021643 -2023-02-13 18:42:11,581 - Epoch: [191][ 340/ 1207] Overall Loss 0.198057 Objective Loss 0.198057 LR 0.000063 Time 0.021576 -2023-02-13 18:42:11,778 - Epoch: [191][ 350/ 1207] Overall Loss 0.198062 Objective Loss 0.198062 LR 0.000063 Time 0.021520 -2023-02-13 18:42:11,973 - Epoch: [191][ 360/ 1207] Overall Loss 0.198214 Objective Loss 0.198214 LR 0.000063 Time 0.021463 -2023-02-13 18:42:12,169 - Epoch: [191][ 370/ 1207] Overall Loss 0.198161 Objective Loss 0.198161 LR 0.000063 Time 0.021413 -2023-02-13 18:42:12,363 - Epoch: [191][ 380/ 1207] Overall Loss 0.197825 Objective Loss 0.197825 LR 0.000063 Time 0.021360 -2023-02-13 18:42:12,560 - Epoch: [191][ 390/ 1207] Overall Loss 0.197426 Objective Loss 0.197426 LR 0.000063 Time 0.021314 -2023-02-13 18:42:12,754 - Epoch: [191][ 400/ 1207] Overall Loss 0.197260 Objective Loss 0.197260 LR 0.000063 Time 0.021265 -2023-02-13 18:42:12,952 - Epoch: [191][ 410/ 1207] Overall Loss 0.197197 Objective Loss 0.197197 LR 0.000063 Time 0.021229 -2023-02-13 18:42:13,146 - Epoch: [191][ 420/ 1207] Overall Loss 0.197455 Objective Loss 0.197455 LR 0.000063 Time 0.021185 -2023-02-13 18:42:13,343 - Epoch: [191][ 430/ 1207] Overall Loss 0.197159 Objective Loss 0.197159 LR 0.000063 Time 0.021148 -2023-02-13 18:42:13,536 - Epoch: [191][ 440/ 1207] Overall Loss 0.197433 Objective Loss 0.197433 LR 0.000063 Time 0.021107 -2023-02-13 18:42:13,733 - Epoch: [191][ 450/ 1207] Overall Loss 0.197649 Objective Loss 0.197649 LR 0.000063 Time 0.021075 -2023-02-13 18:42:13,928 - Epoch: [191][ 460/ 1207] Overall Loss 0.197564 Objective Loss 0.197564 LR 0.000063 Time 0.021040 -2023-02-13 18:42:14,125 - Epoch: [191][ 470/ 1207] Overall Loss 0.197355 Objective Loss 0.197355 LR 0.000063 Time 0.021011 -2023-02-13 18:42:14,319 - Epoch: [191][ 480/ 1207] Overall Loss 0.196965 Objective Loss 0.196965 LR 0.000063 Time 0.020976 -2023-02-13 18:42:14,516 - Epoch: [191][ 490/ 1207] Overall Loss 0.196781 Objective Loss 0.196781 LR 0.000063 Time 0.020949 -2023-02-13 18:42:14,709 - Epoch: [191][ 500/ 1207] Overall Loss 0.196926 Objective Loss 0.196926 LR 0.000063 Time 0.020916 -2023-02-13 18:42:14,907 - Epoch: [191][ 510/ 1207] Overall Loss 0.196975 Objective Loss 0.196975 LR 0.000063 Time 0.020892 -2023-02-13 18:42:15,101 - Epoch: [191][ 520/ 1207] Overall Loss 0.196933 Objective Loss 0.196933 LR 0.000063 Time 0.020863 -2023-02-13 18:42:15,297 - Epoch: [191][ 530/ 1207] Overall Loss 0.196931 Objective Loss 0.196931 LR 0.000063 Time 0.020839 -2023-02-13 18:42:15,490 - Epoch: [191][ 540/ 1207] Overall Loss 0.196750 Objective Loss 0.196750 LR 0.000063 Time 0.020811 -2023-02-13 18:42:15,688 - Epoch: [191][ 550/ 1207] Overall Loss 0.196780 Objective Loss 0.196780 LR 0.000063 Time 0.020790 -2023-02-13 18:42:15,882 - Epoch: [191][ 560/ 1207] Overall Loss 0.196766 Objective Loss 0.196766 LR 0.000063 Time 0.020765 -2023-02-13 18:42:16,080 - Epoch: [191][ 570/ 1207] Overall Loss 0.196861 Objective Loss 0.196861 LR 0.000063 Time 0.020747 -2023-02-13 18:42:16,273 - Epoch: [191][ 580/ 1207] Overall Loss 0.196464 Objective Loss 0.196464 LR 0.000063 Time 0.020723 -2023-02-13 18:42:16,471 - Epoch: [191][ 590/ 1207] Overall Loss 0.196807 Objective Loss 0.196807 LR 0.000063 Time 0.020705 -2023-02-13 18:42:16,665 - Epoch: [191][ 600/ 1207] Overall Loss 0.197050 Objective Loss 0.197050 LR 0.000063 Time 0.020683 -2023-02-13 18:42:16,861 - Epoch: [191][ 610/ 1207] Overall Loss 0.197249 Objective Loss 0.197249 LR 0.000063 Time 0.020665 -2023-02-13 18:42:17,056 - Epoch: [191][ 620/ 1207] Overall Loss 0.197047 Objective Loss 0.197047 LR 0.000063 Time 0.020645 -2023-02-13 18:42:17,252 - Epoch: [191][ 630/ 1207] Overall Loss 0.197031 Objective Loss 0.197031 LR 0.000063 Time 0.020629 -2023-02-13 18:42:17,446 - Epoch: [191][ 640/ 1207] Overall Loss 0.197064 Objective Loss 0.197064 LR 0.000063 Time 0.020609 -2023-02-13 18:42:17,642 - Epoch: [191][ 650/ 1207] Overall Loss 0.196905 Objective Loss 0.196905 LR 0.000063 Time 0.020593 -2023-02-13 18:42:17,835 - Epoch: [191][ 660/ 1207] Overall Loss 0.196869 Objective Loss 0.196869 LR 0.000063 Time 0.020572 -2023-02-13 18:42:18,031 - Epoch: [191][ 670/ 1207] Overall Loss 0.196852 Objective Loss 0.196852 LR 0.000063 Time 0.020557 -2023-02-13 18:42:18,224 - Epoch: [191][ 680/ 1207] Overall Loss 0.196895 Objective Loss 0.196895 LR 0.000063 Time 0.020538 -2023-02-13 18:42:18,419 - Epoch: [191][ 690/ 1207] Overall Loss 0.197005 Objective Loss 0.197005 LR 0.000063 Time 0.020523 -2023-02-13 18:42:18,611 - Epoch: [191][ 700/ 1207] Overall Loss 0.196927 Objective Loss 0.196927 LR 0.000063 Time 0.020504 -2023-02-13 18:42:18,807 - Epoch: [191][ 710/ 1207] Overall Loss 0.196999 Objective Loss 0.196999 LR 0.000063 Time 0.020490 -2023-02-13 18:42:19,001 - Epoch: [191][ 720/ 1207] Overall Loss 0.196999 Objective Loss 0.196999 LR 0.000063 Time 0.020475 -2023-02-13 18:42:19,197 - Epoch: [191][ 730/ 1207] Overall Loss 0.196992 Objective Loss 0.196992 LR 0.000063 Time 0.020462 -2023-02-13 18:42:19,390 - Epoch: [191][ 740/ 1207] Overall Loss 0.196997 Objective Loss 0.196997 LR 0.000063 Time 0.020446 -2023-02-13 18:42:19,585 - Epoch: [191][ 750/ 1207] Overall Loss 0.197213 Objective Loss 0.197213 LR 0.000063 Time 0.020433 -2023-02-13 18:42:19,778 - Epoch: [191][ 760/ 1207] Overall Loss 0.196999 Objective Loss 0.196999 LR 0.000063 Time 0.020417 -2023-02-13 18:42:19,974 - Epoch: [191][ 770/ 1207] Overall Loss 0.196940 Objective Loss 0.196940 LR 0.000063 Time 0.020405 -2023-02-13 18:42:20,166 - Epoch: [191][ 780/ 1207] Overall Loss 0.197043 Objective Loss 0.197043 LR 0.000063 Time 0.020391 -2023-02-13 18:42:20,361 - Epoch: [191][ 790/ 1207] Overall Loss 0.197085 Objective Loss 0.197085 LR 0.000063 Time 0.020378 -2023-02-13 18:42:20,554 - Epoch: [191][ 800/ 1207] Overall Loss 0.197192 Objective Loss 0.197192 LR 0.000063 Time 0.020365 -2023-02-13 18:42:20,749 - Epoch: [191][ 810/ 1207] Overall Loss 0.196975 Objective Loss 0.196975 LR 0.000063 Time 0.020354 -2023-02-13 18:42:20,944 - Epoch: [191][ 820/ 1207] Overall Loss 0.196798 Objective Loss 0.196798 LR 0.000063 Time 0.020342 -2023-02-13 18:42:21,140 - Epoch: [191][ 830/ 1207] Overall Loss 0.196749 Objective Loss 0.196749 LR 0.000063 Time 0.020333 -2023-02-13 18:42:21,333 - Epoch: [191][ 840/ 1207] Overall Loss 0.196939 Objective Loss 0.196939 LR 0.000063 Time 0.020320 -2023-02-13 18:42:21,528 - Epoch: [191][ 850/ 1207] Overall Loss 0.196793 Objective Loss 0.196793 LR 0.000063 Time 0.020311 -2023-02-13 18:42:21,735 - Epoch: [191][ 860/ 1207] Overall Loss 0.196636 Objective Loss 0.196636 LR 0.000063 Time 0.020315 -2023-02-13 18:42:21,942 - Epoch: [191][ 870/ 1207] Overall Loss 0.196528 Objective Loss 0.196528 LR 0.000063 Time 0.020318 -2023-02-13 18:42:22,149 - Epoch: [191][ 880/ 1207] Overall Loss 0.196454 Objective Loss 0.196454 LR 0.000063 Time 0.020323 -2023-02-13 18:42:22,355 - Epoch: [191][ 890/ 1207] Overall Loss 0.196424 Objective Loss 0.196424 LR 0.000063 Time 0.020325 -2023-02-13 18:42:22,561 - Epoch: [191][ 900/ 1207] Overall Loss 0.196679 Objective Loss 0.196679 LR 0.000063 Time 0.020328 -2023-02-13 18:42:22,767 - Epoch: [191][ 910/ 1207] Overall Loss 0.196432 Objective Loss 0.196432 LR 0.000063 Time 0.020330 -2023-02-13 18:42:22,974 - Epoch: [191][ 920/ 1207] Overall Loss 0.196483 Objective Loss 0.196483 LR 0.000063 Time 0.020334 -2023-02-13 18:42:23,180 - Epoch: [191][ 930/ 1207] Overall Loss 0.196455 Objective Loss 0.196455 LR 0.000063 Time 0.020336 -2023-02-13 18:42:23,388 - Epoch: [191][ 940/ 1207] Overall Loss 0.196494 Objective Loss 0.196494 LR 0.000063 Time 0.020340 -2023-02-13 18:42:23,593 - Epoch: [191][ 950/ 1207] Overall Loss 0.196484 Objective Loss 0.196484 LR 0.000063 Time 0.020342 -2023-02-13 18:42:23,800 - Epoch: [191][ 960/ 1207] Overall Loss 0.196258 Objective Loss 0.196258 LR 0.000063 Time 0.020345 -2023-02-13 18:42:24,005 - Epoch: [191][ 970/ 1207] Overall Loss 0.196073 Objective Loss 0.196073 LR 0.000063 Time 0.020346 -2023-02-13 18:42:24,213 - Epoch: [191][ 980/ 1207] Overall Loss 0.196213 Objective Loss 0.196213 LR 0.000063 Time 0.020351 -2023-02-13 18:42:24,418 - Epoch: [191][ 990/ 1207] Overall Loss 0.196090 Objective Loss 0.196090 LR 0.000063 Time 0.020352 -2023-02-13 18:42:24,624 - Epoch: [191][ 1000/ 1207] Overall Loss 0.196119 Objective Loss 0.196119 LR 0.000063 Time 0.020354 -2023-02-13 18:42:24,829 - Epoch: [191][ 1010/ 1207] Overall Loss 0.196343 Objective Loss 0.196343 LR 0.000063 Time 0.020355 -2023-02-13 18:42:25,037 - Epoch: [191][ 1020/ 1207] Overall Loss 0.196485 Objective Loss 0.196485 LR 0.000063 Time 0.020359 -2023-02-13 18:42:25,242 - Epoch: [191][ 1030/ 1207] Overall Loss 0.196560 Objective Loss 0.196560 LR 0.000063 Time 0.020360 -2023-02-13 18:42:25,451 - Epoch: [191][ 1040/ 1207] Overall Loss 0.196562 Objective Loss 0.196562 LR 0.000063 Time 0.020365 -2023-02-13 18:42:25,652 - Epoch: [191][ 1050/ 1207] Overall Loss 0.196619 Objective Loss 0.196619 LR 0.000063 Time 0.020362 -2023-02-13 18:42:25,857 - Epoch: [191][ 1060/ 1207] Overall Loss 0.196744 Objective Loss 0.196744 LR 0.000063 Time 0.020363 -2023-02-13 18:42:26,059 - Epoch: [191][ 1070/ 1207] Overall Loss 0.196682 Objective Loss 0.196682 LR 0.000063 Time 0.020361 -2023-02-13 18:42:26,264 - Epoch: [191][ 1080/ 1207] Overall Loss 0.196719 Objective Loss 0.196719 LR 0.000063 Time 0.020362 -2023-02-13 18:42:26,465 - Epoch: [191][ 1090/ 1207] Overall Loss 0.196675 Objective Loss 0.196675 LR 0.000063 Time 0.020359 -2023-02-13 18:42:26,671 - Epoch: [191][ 1100/ 1207] Overall Loss 0.196759 Objective Loss 0.196759 LR 0.000063 Time 0.020361 -2023-02-13 18:42:26,872 - Epoch: [191][ 1110/ 1207] Overall Loss 0.196761 Objective Loss 0.196761 LR 0.000063 Time 0.020358 -2023-02-13 18:42:27,078 - Epoch: [191][ 1120/ 1207] Overall Loss 0.196896 Objective Loss 0.196896 LR 0.000063 Time 0.020360 -2023-02-13 18:42:27,279 - Epoch: [191][ 1130/ 1207] Overall Loss 0.196989 Objective Loss 0.196989 LR 0.000063 Time 0.020358 -2023-02-13 18:42:27,485 - Epoch: [191][ 1140/ 1207] Overall Loss 0.197064 Objective Loss 0.197064 LR 0.000063 Time 0.020359 -2023-02-13 18:42:27,686 - Epoch: [191][ 1150/ 1207] Overall Loss 0.196990 Objective Loss 0.196990 LR 0.000063 Time 0.020357 -2023-02-13 18:42:27,891 - Epoch: [191][ 1160/ 1207] Overall Loss 0.197042 Objective Loss 0.197042 LR 0.000063 Time 0.020358 -2023-02-13 18:42:28,092 - Epoch: [191][ 1170/ 1207] Overall Loss 0.196962 Objective Loss 0.196962 LR 0.000063 Time 0.020355 -2023-02-13 18:42:28,297 - Epoch: [191][ 1180/ 1207] Overall Loss 0.197147 Objective Loss 0.197147 LR 0.000063 Time 0.020356 -2023-02-13 18:42:28,498 - Epoch: [191][ 1190/ 1207] Overall Loss 0.197300 Objective Loss 0.197300 LR 0.000063 Time 0.020354 -2023-02-13 18:42:28,756 - Epoch: [191][ 1200/ 1207] Overall Loss 0.197285 Objective Loss 0.197285 LR 0.000063 Time 0.020398 -2023-02-13 18:42:28,870 - Epoch: [191][ 1207/ 1207] Overall Loss 0.197256 Objective Loss 0.197256 Top1 89.634146 Top5 99.390244 LR 0.000063 Time 0.020375 -2023-02-13 18:42:28,951 - --- validate (epoch=191)----------- -2023-02-13 18:42:28,951 - 34311 samples (256 per mini-batch) -2023-02-13 18:42:29,355 - Epoch: [191][ 10/ 135] Loss 0.266995 Top1 86.484375 Top5 97.851562 -2023-02-13 18:42:29,488 - Epoch: [191][ 20/ 135] Loss 0.265037 Top1 86.171875 Top5 97.792969 -2023-02-13 18:42:29,619 - Epoch: [191][ 30/ 135] Loss 0.274141 Top1 85.924479 Top5 97.708333 -2023-02-13 18:42:29,751 - Epoch: [191][ 40/ 135] Loss 0.269651 Top1 85.976562 Top5 97.871094 -2023-02-13 18:42:29,883 - Epoch: [191][ 50/ 135] Loss 0.270671 Top1 85.937500 Top5 97.898438 -2023-02-13 18:42:30,012 - Epoch: [191][ 60/ 135] Loss 0.274180 Top1 85.937500 Top5 97.936198 -2023-02-13 18:42:30,144 - Epoch: [191][ 70/ 135] Loss 0.277173 Top1 85.948661 Top5 97.924107 -2023-02-13 18:42:30,275 - Epoch: [191][ 80/ 135] Loss 0.274959 Top1 85.976562 Top5 97.973633 -2023-02-13 18:42:30,407 - Epoch: [191][ 90/ 135] Loss 0.278514 Top1 85.881076 Top5 97.916667 -2023-02-13 18:42:30,539 - Epoch: [191][ 100/ 135] Loss 0.282198 Top1 85.812500 Top5 97.882812 -2023-02-13 18:42:30,671 - Epoch: [191][ 110/ 135] Loss 0.284682 Top1 85.784801 Top5 97.872869 -2023-02-13 18:42:30,803 - Epoch: [191][ 120/ 135] Loss 0.285790 Top1 85.813802 Top5 97.874349 -2023-02-13 18:42:30,931 - Epoch: [191][ 130/ 135] Loss 0.286438 Top1 85.766226 Top5 97.884615 -2023-02-13 18:42:30,977 - Epoch: [191][ 135/ 135] Loss 0.284138 Top1 85.768412 Top5 97.878231 -2023-02-13 18:42:31,045 - ==> Top1: 85.768 Top5: 97.878 Loss: 0.284 - -2023-02-13 18:42:31,045 - ==> Confusion: -[[ 868 4 7 1 7 2 0 2 4 42 0 3 2 3 5 3 1 2 0 2 9] - [ 4 965 1 1 11 15 2 11 2 0 1 1 2 0 0 1 5 1 2 1 7] - [ 4 7 964 14 2 1 13 14 1 1 3 2 1 5 5 5 3 2 2 2 7] - [ 5 1 21 920 0 5 0 1 3 2 12 0 5 0 17 1 2 4 12 0 5] - [ 13 7 0 0 998 8 1 1 1 0 0 7 4 3 6 5 3 2 1 2 4] - [ 1 17 1 5 4 973 2 15 3 4 2 9 0 12 2 3 8 1 2 1 5] - [ 3 3 14 3 0 5 1046 7 0 1 1 0 1 0 0 2 3 3 1 3 3] - [ 5 8 9 1 2 29 2 932 0 1 0 6 2 0 0 0 1 2 14 5 5] - [ 12 2 1 1 1 0 0 1 925 30 8 2 0 8 9 2 1 1 3 1 1] - [ 70 0 3 0 10 1 0 0 27 872 0 1 0 14 4 1 2 0 2 1 4] - [ 1 0 5 7 2 3 2 3 13 3 991 1 1 5 4 0 2 1 3 0 4] - [ 4 2 2 0 1 13 1 6 2 2 0 915 21 7 0 10 2 9 0 6 2] - [ 0 0 0 8 2 3 0 0 2 1 0 26 885 1 1 7 2 10 1 1 9] - [ 6 2 3 0 6 10 1 1 8 19 7 5 1 933 4 3 4 4 0 1 6] - [ 5 1 0 20 3 4 0 2 16 8 2 0 2 1 1011 0 2 3 4 0 8] - [ 4 1 8 1 5 0 4 2 0 1 0 6 8 1 0 974 11 9 0 6 5] - [ 2 7 0 1 5 1 0 0 2 0 0 0 3 3 3 10 1012 1 1 3 7] - [ 4 1 0 5 0 1 2 0 0 1 2 11 14 1 0 20 0 982 0 2 5] - [ 3 5 6 7 0 2 1 20 3 0 4 3 1 0 10 1 0 3 1012 3 2] - [ 0 4 0 0 0 5 8 9 0 0 0 13 4 4 0 6 7 2 0 1079 7] - [ 150 227 232 106 141 196 73 174 90 97 188 90 280 253 150 80 251 89 157 239 10171]] - -2023-02-13 18:42:31,047 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:42:31,047 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:42:31,053 - - -2023-02-13 18:42:31,054 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:42:31,944 - Epoch: [192][ 10/ 1207] Overall Loss 0.193578 Objective Loss 0.193578 LR 0.000063 Time 0.088999 -2023-02-13 18:42:32,136 - Epoch: [192][ 20/ 1207] Overall Loss 0.190666 Objective Loss 0.190666 LR 0.000063 Time 0.054046 -2023-02-13 18:42:32,325 - Epoch: [192][ 30/ 1207] Overall Loss 0.200743 Objective Loss 0.200743 LR 0.000063 Time 0.042311 -2023-02-13 18:42:32,512 - Epoch: [192][ 40/ 1207] Overall Loss 0.201885 Objective Loss 0.201885 LR 0.000063 Time 0.036420 -2023-02-13 18:42:32,700 - Epoch: [192][ 50/ 1207] Overall Loss 0.206503 Objective Loss 0.206503 LR 0.000063 Time 0.032879 -2023-02-13 18:42:32,887 - Epoch: [192][ 60/ 1207] Overall Loss 0.203408 Objective Loss 0.203408 LR 0.000063 Time 0.030520 -2023-02-13 18:42:33,076 - Epoch: [192][ 70/ 1207] Overall Loss 0.204083 Objective Loss 0.204083 LR 0.000063 Time 0.028852 -2023-02-13 18:42:33,264 - Epoch: [192][ 80/ 1207] Overall Loss 0.201714 Objective Loss 0.201714 LR 0.000063 Time 0.027586 -2023-02-13 18:42:33,452 - Epoch: [192][ 90/ 1207] Overall Loss 0.203352 Objective Loss 0.203352 LR 0.000063 Time 0.026603 -2023-02-13 18:42:33,639 - Epoch: [192][ 100/ 1207] Overall Loss 0.201080 Objective Loss 0.201080 LR 0.000063 Time 0.025808 -2023-02-13 18:42:33,826 - Epoch: [192][ 110/ 1207] Overall Loss 0.199516 Objective Loss 0.199516 LR 0.000063 Time 0.025165 -2023-02-13 18:42:34,015 - Epoch: [192][ 120/ 1207] Overall Loss 0.197372 Objective Loss 0.197372 LR 0.000063 Time 0.024638 -2023-02-13 18:42:34,203 - Epoch: [192][ 130/ 1207] Overall Loss 0.195938 Objective Loss 0.195938 LR 0.000063 Time 0.024185 -2023-02-13 18:42:34,390 - Epoch: [192][ 140/ 1207] Overall Loss 0.195533 Objective Loss 0.195533 LR 0.000063 Time 0.023790 -2023-02-13 18:42:34,577 - Epoch: [192][ 150/ 1207] Overall Loss 0.193850 Objective Loss 0.193850 LR 0.000063 Time 0.023450 -2023-02-13 18:42:34,765 - Epoch: [192][ 160/ 1207] Overall Loss 0.194680 Objective Loss 0.194680 LR 0.000063 Time 0.023155 -2023-02-13 18:42:34,954 - Epoch: [192][ 170/ 1207] Overall Loss 0.194455 Objective Loss 0.194455 LR 0.000063 Time 0.022902 -2023-02-13 18:42:35,141 - Epoch: [192][ 180/ 1207] Overall Loss 0.194772 Objective Loss 0.194772 LR 0.000063 Time 0.022670 -2023-02-13 18:42:35,329 - Epoch: [192][ 190/ 1207] Overall Loss 0.195755 Objective Loss 0.195755 LR 0.000063 Time 0.022461 -2023-02-13 18:42:35,516 - Epoch: [192][ 200/ 1207] Overall Loss 0.196092 Objective Loss 0.196092 LR 0.000063 Time 0.022271 -2023-02-13 18:42:35,703 - Epoch: [192][ 210/ 1207] Overall Loss 0.195352 Objective Loss 0.195352 LR 0.000063 Time 0.022101 -2023-02-13 18:42:35,890 - Epoch: [192][ 220/ 1207] Overall Loss 0.194670 Objective Loss 0.194670 LR 0.000063 Time 0.021946 -2023-02-13 18:42:36,079 - Epoch: [192][ 230/ 1207] Overall Loss 0.194207 Objective Loss 0.194207 LR 0.000063 Time 0.021812 -2023-02-13 18:42:36,267 - Epoch: [192][ 240/ 1207] Overall Loss 0.193206 Objective Loss 0.193206 LR 0.000063 Time 0.021682 -2023-02-13 18:42:36,454 - Epoch: [192][ 250/ 1207] Overall Loss 0.194388 Objective Loss 0.194388 LR 0.000063 Time 0.021562 -2023-02-13 18:42:36,641 - Epoch: [192][ 260/ 1207] Overall Loss 0.194289 Objective Loss 0.194289 LR 0.000063 Time 0.021451 -2023-02-13 18:42:36,829 - Epoch: [192][ 270/ 1207] Overall Loss 0.194395 Objective Loss 0.194395 LR 0.000063 Time 0.021350 -2023-02-13 18:42:37,017 - Epoch: [192][ 280/ 1207] Overall Loss 0.195348 Objective Loss 0.195348 LR 0.000063 Time 0.021260 -2023-02-13 18:42:37,204 - Epoch: [192][ 290/ 1207] Overall Loss 0.195538 Objective Loss 0.195538 LR 0.000063 Time 0.021171 -2023-02-13 18:42:37,391 - Epoch: [192][ 300/ 1207] Overall Loss 0.194788 Objective Loss 0.194788 LR 0.000063 Time 0.021086 -2023-02-13 18:42:37,578 - Epoch: [192][ 310/ 1207] Overall Loss 0.195403 Objective Loss 0.195403 LR 0.000063 Time 0.021008 -2023-02-13 18:42:37,765 - Epoch: [192][ 320/ 1207] Overall Loss 0.195845 Objective Loss 0.195845 LR 0.000063 Time 0.020935 -2023-02-13 18:42:37,954 - Epoch: [192][ 330/ 1207] Overall Loss 0.195837 Objective Loss 0.195837 LR 0.000063 Time 0.020872 -2023-02-13 18:42:38,141 - Epoch: [192][ 340/ 1207] Overall Loss 0.195714 Objective Loss 0.195714 LR 0.000063 Time 0.020809 -2023-02-13 18:42:38,329 - Epoch: [192][ 350/ 1207] Overall Loss 0.195632 Objective Loss 0.195632 LR 0.000063 Time 0.020748 -2023-02-13 18:42:38,515 - Epoch: [192][ 360/ 1207] Overall Loss 0.196236 Objective Loss 0.196236 LR 0.000063 Time 0.020689 -2023-02-13 18:42:38,703 - Epoch: [192][ 370/ 1207] Overall Loss 0.196546 Objective Loss 0.196546 LR 0.000063 Time 0.020636 -2023-02-13 18:42:38,891 - Epoch: [192][ 380/ 1207] Overall Loss 0.196691 Objective Loss 0.196691 LR 0.000063 Time 0.020585 -2023-02-13 18:42:39,079 - Epoch: [192][ 390/ 1207] Overall Loss 0.196902 Objective Loss 0.196902 LR 0.000063 Time 0.020539 -2023-02-13 18:42:39,266 - Epoch: [192][ 400/ 1207] Overall Loss 0.197148 Objective Loss 0.197148 LR 0.000063 Time 0.020493 -2023-02-13 18:42:39,453 - Epoch: [192][ 410/ 1207] Overall Loss 0.196837 Objective Loss 0.196837 LR 0.000063 Time 0.020448 -2023-02-13 18:42:39,640 - Epoch: [192][ 420/ 1207] Overall Loss 0.196741 Objective Loss 0.196741 LR 0.000063 Time 0.020405 -2023-02-13 18:42:39,827 - Epoch: [192][ 430/ 1207] Overall Loss 0.196776 Objective Loss 0.196776 LR 0.000063 Time 0.020365 -2023-02-13 18:42:40,015 - Epoch: [192][ 440/ 1207] Overall Loss 0.197260 Objective Loss 0.197260 LR 0.000063 Time 0.020328 -2023-02-13 18:42:40,202 - Epoch: [192][ 450/ 1207] Overall Loss 0.197533 Objective Loss 0.197533 LR 0.000063 Time 0.020293 -2023-02-13 18:42:40,390 - Epoch: [192][ 460/ 1207] Overall Loss 0.197744 Objective Loss 0.197744 LR 0.000063 Time 0.020258 -2023-02-13 18:42:40,578 - Epoch: [192][ 470/ 1207] Overall Loss 0.197724 Objective Loss 0.197724 LR 0.000063 Time 0.020226 -2023-02-13 18:42:40,765 - Epoch: [192][ 480/ 1207] Overall Loss 0.197761 Objective Loss 0.197761 LR 0.000063 Time 0.020194 -2023-02-13 18:42:40,953 - Epoch: [192][ 490/ 1207] Overall Loss 0.197345 Objective Loss 0.197345 LR 0.000063 Time 0.020166 -2023-02-13 18:42:41,141 - Epoch: [192][ 500/ 1207] Overall Loss 0.197141 Objective Loss 0.197141 LR 0.000063 Time 0.020138 -2023-02-13 18:42:41,329 - Epoch: [192][ 510/ 1207] Overall Loss 0.197072 Objective Loss 0.197072 LR 0.000063 Time 0.020110 -2023-02-13 18:42:41,517 - Epoch: [192][ 520/ 1207] Overall Loss 0.197388 Objective Loss 0.197388 LR 0.000063 Time 0.020084 -2023-02-13 18:42:41,705 - Epoch: [192][ 530/ 1207] Overall Loss 0.197248 Objective Loss 0.197248 LR 0.000063 Time 0.020059 -2023-02-13 18:42:41,892 - Epoch: [192][ 540/ 1207] Overall Loss 0.197077 Objective Loss 0.197077 LR 0.000063 Time 0.020034 -2023-02-13 18:42:42,081 - Epoch: [192][ 550/ 1207] Overall Loss 0.196917 Objective Loss 0.196917 LR 0.000063 Time 0.020013 -2023-02-13 18:42:42,270 - Epoch: [192][ 560/ 1207] Overall Loss 0.197034 Objective Loss 0.197034 LR 0.000063 Time 0.019991 -2023-02-13 18:42:42,457 - Epoch: [192][ 570/ 1207] Overall Loss 0.197162 Objective Loss 0.197162 LR 0.000063 Time 0.019969 -2023-02-13 18:42:42,645 - Epoch: [192][ 580/ 1207] Overall Loss 0.197156 Objective Loss 0.197156 LR 0.000063 Time 0.019947 -2023-02-13 18:42:42,832 - Epoch: [192][ 590/ 1207] Overall Loss 0.197441 Objective Loss 0.197441 LR 0.000063 Time 0.019926 -2023-02-13 18:42:43,021 - Epoch: [192][ 600/ 1207] Overall Loss 0.197348 Objective Loss 0.197348 LR 0.000063 Time 0.019908 -2023-02-13 18:42:43,209 - Epoch: [192][ 610/ 1207] Overall Loss 0.197307 Objective Loss 0.197307 LR 0.000063 Time 0.019889 -2023-02-13 18:42:43,397 - Epoch: [192][ 620/ 1207] Overall Loss 0.196912 Objective Loss 0.196912 LR 0.000063 Time 0.019871 -2023-02-13 18:42:43,584 - Epoch: [192][ 630/ 1207] Overall Loss 0.196743 Objective Loss 0.196743 LR 0.000063 Time 0.019853 -2023-02-13 18:42:43,772 - Epoch: [192][ 640/ 1207] Overall Loss 0.196583 Objective Loss 0.196583 LR 0.000063 Time 0.019834 -2023-02-13 18:42:43,960 - Epoch: [192][ 650/ 1207] Overall Loss 0.196594 Objective Loss 0.196594 LR 0.000063 Time 0.019818 -2023-02-13 18:42:44,148 - Epoch: [192][ 660/ 1207] Overall Loss 0.196779 Objective Loss 0.196779 LR 0.000063 Time 0.019803 -2023-02-13 18:42:44,337 - Epoch: [192][ 670/ 1207] Overall Loss 0.196804 Objective Loss 0.196804 LR 0.000063 Time 0.019788 -2023-02-13 18:42:44,524 - Epoch: [192][ 680/ 1207] Overall Loss 0.196779 Objective Loss 0.196779 LR 0.000063 Time 0.019772 -2023-02-13 18:42:44,712 - Epoch: [192][ 690/ 1207] Overall Loss 0.196781 Objective Loss 0.196781 LR 0.000063 Time 0.019758 -2023-02-13 18:42:44,900 - Epoch: [192][ 700/ 1207] Overall Loss 0.197046 Objective Loss 0.197046 LR 0.000063 Time 0.019743 -2023-02-13 18:42:45,089 - Epoch: [192][ 710/ 1207] Overall Loss 0.197386 Objective Loss 0.197386 LR 0.000063 Time 0.019731 -2023-02-13 18:42:45,278 - Epoch: [192][ 720/ 1207] Overall Loss 0.197414 Objective Loss 0.197414 LR 0.000063 Time 0.019719 -2023-02-13 18:42:45,467 - Epoch: [192][ 730/ 1207] Overall Loss 0.197515 Objective Loss 0.197515 LR 0.000063 Time 0.019707 -2023-02-13 18:42:45,656 - Epoch: [192][ 740/ 1207] Overall Loss 0.197647 Objective Loss 0.197647 LR 0.000063 Time 0.019695 -2023-02-13 18:42:45,845 - Epoch: [192][ 750/ 1207] Overall Loss 0.197884 Objective Loss 0.197884 LR 0.000063 Time 0.019684 -2023-02-13 18:42:46,035 - Epoch: [192][ 760/ 1207] Overall Loss 0.197843 Objective Loss 0.197843 LR 0.000063 Time 0.019675 -2023-02-13 18:42:46,224 - Epoch: [192][ 770/ 1207] Overall Loss 0.197996 Objective Loss 0.197996 LR 0.000063 Time 0.019665 -2023-02-13 18:42:46,413 - Epoch: [192][ 780/ 1207] Overall Loss 0.198183 Objective Loss 0.198183 LR 0.000063 Time 0.019654 -2023-02-13 18:42:46,602 - Epoch: [192][ 790/ 1207] Overall Loss 0.197961 Objective Loss 0.197961 LR 0.000063 Time 0.019644 -2023-02-13 18:42:46,791 - Epoch: [192][ 800/ 1207] Overall Loss 0.197834 Objective Loss 0.197834 LR 0.000063 Time 0.019634 -2023-02-13 18:42:46,980 - Epoch: [192][ 810/ 1207] Overall Loss 0.197711 Objective Loss 0.197711 LR 0.000063 Time 0.019625 -2023-02-13 18:42:47,172 - Epoch: [192][ 820/ 1207] Overall Loss 0.197797 Objective Loss 0.197797 LR 0.000063 Time 0.019619 -2023-02-13 18:42:47,362 - Epoch: [192][ 830/ 1207] Overall Loss 0.197665 Objective Loss 0.197665 LR 0.000063 Time 0.019612 -2023-02-13 18:42:47,554 - Epoch: [192][ 840/ 1207] Overall Loss 0.197661 Objective Loss 0.197661 LR 0.000063 Time 0.019606 -2023-02-13 18:42:47,744 - Epoch: [192][ 850/ 1207] Overall Loss 0.197495 Objective Loss 0.197495 LR 0.000063 Time 0.019599 -2023-02-13 18:42:47,936 - Epoch: [192][ 860/ 1207] Overall Loss 0.197684 Objective Loss 0.197684 LR 0.000063 Time 0.019593 -2023-02-13 18:42:48,127 - Epoch: [192][ 870/ 1207] Overall Loss 0.197714 Objective Loss 0.197714 LR 0.000063 Time 0.019588 -2023-02-13 18:42:48,319 - Epoch: [192][ 880/ 1207] Overall Loss 0.197804 Objective Loss 0.197804 LR 0.000063 Time 0.019582 -2023-02-13 18:42:48,509 - Epoch: [192][ 890/ 1207] Overall Loss 0.197763 Objective Loss 0.197763 LR 0.000063 Time 0.019576 -2023-02-13 18:42:48,700 - Epoch: [192][ 900/ 1207] Overall Loss 0.197841 Objective Loss 0.197841 LR 0.000063 Time 0.019570 -2023-02-13 18:42:48,891 - Epoch: [192][ 910/ 1207] Overall Loss 0.197851 Objective Loss 0.197851 LR 0.000063 Time 0.019564 -2023-02-13 18:42:49,084 - Epoch: [192][ 920/ 1207] Overall Loss 0.197959 Objective Loss 0.197959 LR 0.000063 Time 0.019561 -2023-02-13 18:42:49,273 - Epoch: [192][ 930/ 1207] Overall Loss 0.197693 Objective Loss 0.197693 LR 0.000063 Time 0.019554 -2023-02-13 18:42:49,462 - Epoch: [192][ 940/ 1207] Overall Loss 0.197544 Objective Loss 0.197544 LR 0.000063 Time 0.019546 -2023-02-13 18:42:49,651 - Epoch: [192][ 950/ 1207] Overall Loss 0.197732 Objective Loss 0.197732 LR 0.000063 Time 0.019539 -2023-02-13 18:42:49,840 - Epoch: [192][ 960/ 1207] Overall Loss 0.197702 Objective Loss 0.197702 LR 0.000063 Time 0.019532 -2023-02-13 18:42:50,030 - Epoch: [192][ 970/ 1207] Overall Loss 0.197883 Objective Loss 0.197883 LR 0.000063 Time 0.019526 -2023-02-13 18:42:50,220 - Epoch: [192][ 980/ 1207] Overall Loss 0.197779 Objective Loss 0.197779 LR 0.000063 Time 0.019520 -2023-02-13 18:42:50,409 - Epoch: [192][ 990/ 1207] Overall Loss 0.197728 Objective Loss 0.197728 LR 0.000063 Time 0.019513 -2023-02-13 18:42:50,597 - Epoch: [192][ 1000/ 1207] Overall Loss 0.197706 Objective Loss 0.197706 LR 0.000063 Time 0.019506 -2023-02-13 18:42:50,786 - Epoch: [192][ 1010/ 1207] Overall Loss 0.197783 Objective Loss 0.197783 LR 0.000063 Time 0.019500 -2023-02-13 18:42:50,976 - Epoch: [192][ 1020/ 1207] Overall Loss 0.197627 Objective Loss 0.197627 LR 0.000063 Time 0.019494 -2023-02-13 18:42:51,165 - Epoch: [192][ 1030/ 1207] Overall Loss 0.197488 Objective Loss 0.197488 LR 0.000063 Time 0.019489 -2023-02-13 18:42:51,354 - Epoch: [192][ 1040/ 1207] Overall Loss 0.197338 Objective Loss 0.197338 LR 0.000063 Time 0.019482 -2023-02-13 18:42:51,543 - Epoch: [192][ 1050/ 1207] Overall Loss 0.197206 Objective Loss 0.197206 LR 0.000063 Time 0.019477 -2023-02-13 18:42:51,732 - Epoch: [192][ 1060/ 1207] Overall Loss 0.197346 Objective Loss 0.197346 LR 0.000063 Time 0.019470 -2023-02-13 18:42:51,921 - Epoch: [192][ 1070/ 1207] Overall Loss 0.197484 Objective Loss 0.197484 LR 0.000063 Time 0.019465 -2023-02-13 18:42:52,112 - Epoch: [192][ 1080/ 1207] Overall Loss 0.197396 Objective Loss 0.197396 LR 0.000063 Time 0.019461 -2023-02-13 18:42:52,300 - Epoch: [192][ 1090/ 1207] Overall Loss 0.197381 Objective Loss 0.197381 LR 0.000063 Time 0.019455 -2023-02-13 18:42:52,489 - Epoch: [192][ 1100/ 1207] Overall Loss 0.197394 Objective Loss 0.197394 LR 0.000063 Time 0.019449 -2023-02-13 18:42:52,677 - Epoch: [192][ 1110/ 1207] Overall Loss 0.197327 Objective Loss 0.197327 LR 0.000063 Time 0.019443 -2023-02-13 18:42:52,865 - Epoch: [192][ 1120/ 1207] Overall Loss 0.197491 Objective Loss 0.197491 LR 0.000063 Time 0.019437 -2023-02-13 18:42:53,055 - Epoch: [192][ 1130/ 1207] Overall Loss 0.197424 Objective Loss 0.197424 LR 0.000063 Time 0.019433 -2023-02-13 18:42:53,243 - Epoch: [192][ 1140/ 1207] Overall Loss 0.197437 Objective Loss 0.197437 LR 0.000063 Time 0.019427 -2023-02-13 18:42:53,431 - Epoch: [192][ 1150/ 1207] Overall Loss 0.197548 Objective Loss 0.197548 LR 0.000063 Time 0.019421 -2023-02-13 18:42:53,620 - Epoch: [192][ 1160/ 1207] Overall Loss 0.197537 Objective Loss 0.197537 LR 0.000063 Time 0.019416 -2023-02-13 18:42:53,808 - Epoch: [192][ 1170/ 1207] Overall Loss 0.197506 Objective Loss 0.197506 LR 0.000063 Time 0.019411 -2023-02-13 18:42:53,997 - Epoch: [192][ 1180/ 1207] Overall Loss 0.197486 Objective Loss 0.197486 LR 0.000063 Time 0.019406 -2023-02-13 18:42:54,187 - Epoch: [192][ 1190/ 1207] Overall Loss 0.197674 Objective Loss 0.197674 LR 0.000063 Time 0.019402 -2023-02-13 18:42:54,426 - Epoch: [192][ 1200/ 1207] Overall Loss 0.197587 Objective Loss 0.197587 LR 0.000063 Time 0.019440 -2023-02-13 18:42:54,541 - Epoch: [192][ 1207/ 1207] Overall Loss 0.197444 Objective Loss 0.197444 Top1 87.804878 Top5 97.865854 LR 0.000063 Time 0.019422 -2023-02-13 18:42:54,613 - --- validate (epoch=192)----------- -2023-02-13 18:42:54,613 - 34311 samples (256 per mini-batch) -2023-02-13 18:42:55,015 - Epoch: [192][ 10/ 135] Loss 0.283128 Top1 84.921875 Top5 97.695312 -2023-02-13 18:42:55,140 - Epoch: [192][ 20/ 135] Loss 0.280913 Top1 84.863281 Top5 97.714844 -2023-02-13 18:42:55,267 - Epoch: [192][ 30/ 135] Loss 0.274404 Top1 85.091146 Top5 97.760417 -2023-02-13 18:42:55,397 - Epoch: [192][ 40/ 135] Loss 0.276619 Top1 85.205078 Top5 97.695312 -2023-02-13 18:42:55,523 - Epoch: [192][ 50/ 135] Loss 0.279578 Top1 85.343750 Top5 97.687500 -2023-02-13 18:42:55,653 - Epoch: [192][ 60/ 135] Loss 0.284415 Top1 85.208333 Top5 97.773438 -2023-02-13 18:42:55,782 - Epoch: [192][ 70/ 135] Loss 0.283941 Top1 85.239955 Top5 97.784598 -2023-02-13 18:42:55,913 - Epoch: [192][ 80/ 135] Loss 0.286952 Top1 85.209961 Top5 97.744141 -2023-02-13 18:42:56,040 - Epoch: [192][ 90/ 135] Loss 0.288841 Top1 85.125868 Top5 97.760417 -2023-02-13 18:42:56,166 - Epoch: [192][ 100/ 135] Loss 0.290004 Top1 85.070312 Top5 97.769531 -2023-02-13 18:42:56,293 - Epoch: [192][ 110/ 135] Loss 0.289739 Top1 85.113636 Top5 97.769886 -2023-02-13 18:42:56,420 - Epoch: [192][ 120/ 135] Loss 0.287716 Top1 85.179036 Top5 97.786458 -2023-02-13 18:42:56,548 - Epoch: [192][ 130/ 135] Loss 0.287332 Top1 85.240385 Top5 97.827524 -2023-02-13 18:42:56,592 - Epoch: [192][ 135/ 135] Loss 0.290348 Top1 85.208825 Top5 97.837428 -2023-02-13 18:42:56,663 - ==> Top1: 85.209 Top5: 97.837 Loss: 0.290 - -2023-02-13 18:42:56,664 - ==> Confusion: -[[ 878 6 7 1 8 1 0 1 4 30 0 3 1 5 6 3 3 2 0 1 7] - [ 2 955 1 1 10 21 1 12 3 0 1 1 4 0 0 1 5 0 3 4 8] - [ 4 2 970 12 2 1 11 14 1 1 2 4 0 3 4 7 5 1 6 4 4] - [ 4 1 21 909 1 4 0 2 3 2 9 0 8 0 20 3 3 5 16 0 5] - [ 14 7 0 0 996 13 1 0 1 1 0 4 4 3 8 5 3 2 0 2 2] - [ 2 9 2 6 5 984 2 13 2 2 0 9 0 15 0 3 7 2 2 2 3] - [ 2 2 15 2 0 5 1038 4 0 1 1 1 3 2 1 5 2 2 2 7 4] - [ 3 10 10 1 0 29 1 931 1 1 0 8 4 0 0 0 2 1 12 7 3] - [ 14 3 1 1 1 0 0 1 910 34 7 3 1 12 14 2 1 0 3 0 1] - [ 94 1 2 1 12 1 0 2 33 837 1 0 0 15 4 1 1 2 1 0 4] - [ 3 1 5 6 2 2 2 3 11 2 986 1 1 10 2 0 2 0 8 0 4] - [ 3 3 2 0 1 13 0 2 2 1 0 918 29 5 1 6 1 5 3 9 1] - [ 2 0 0 7 1 4 0 0 1 1 0 22 898 0 1 4 2 10 0 1 5] - [ 5 2 4 1 5 9 0 0 10 17 6 2 3 936 7 4 5 0 1 1 6] - [ 3 3 1 17 5 3 0 1 18 8 2 1 3 2 1009 0 2 4 4 0 6] - [ 2 2 5 0 2 1 2 2 0 0 0 5 8 2 1 983 10 6 0 8 7] - [ 2 4 2 1 5 1 0 0 1 0 0 1 5 2 2 11 1011 1 0 4 8] - [ 3 1 0 5 1 0 1 0 0 2 0 10 17 1 1 20 0 981 1 2 5] - [ 3 4 4 8 0 2 0 26 2 0 6 0 6 0 13 1 1 2 1004 1 3] - [ 0 2 0 0 0 5 7 9 0 0 0 14 4 3 1 7 3 2 0 1085 6] - [ 173 206 250 118 141 220 63 168 91 73 173 108 318 271 157 105 254 96 172 261 10016]] - -2023-02-13 18:42:56,666 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:42:56,666 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:42:56,672 - - -2023-02-13 18:42:56,672 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:42:57,679 - Epoch: [193][ 10/ 1207] Overall Loss 0.186029 Objective Loss 0.186029 LR 0.000063 Time 0.100646 -2023-02-13 18:42:57,877 - Epoch: [193][ 20/ 1207] Overall Loss 0.202693 Objective Loss 0.202693 LR 0.000063 Time 0.060206 -2023-02-13 18:42:58,073 - Epoch: [193][ 30/ 1207] Overall Loss 0.202184 Objective Loss 0.202184 LR 0.000063 Time 0.046670 -2023-02-13 18:42:58,269 - Epoch: [193][ 40/ 1207] Overall Loss 0.202027 Objective Loss 0.202027 LR 0.000063 Time 0.039876 -2023-02-13 18:42:58,464 - Epoch: [193][ 50/ 1207] Overall Loss 0.202564 Objective Loss 0.202564 LR 0.000063 Time 0.035790 -2023-02-13 18:42:58,659 - Epoch: [193][ 60/ 1207] Overall Loss 0.198974 Objective Loss 0.198974 LR 0.000063 Time 0.033070 -2023-02-13 18:42:58,853 - Epoch: [193][ 70/ 1207] Overall Loss 0.197131 Objective Loss 0.197131 LR 0.000063 Time 0.031122 -2023-02-13 18:42:59,049 - Epoch: [193][ 80/ 1207] Overall Loss 0.197295 Objective Loss 0.197295 LR 0.000063 Time 0.029677 -2023-02-13 18:42:59,244 - Epoch: [193][ 90/ 1207] Overall Loss 0.196797 Objective Loss 0.196797 LR 0.000063 Time 0.028543 -2023-02-13 18:42:59,439 - Epoch: [193][ 100/ 1207] Overall Loss 0.197458 Objective Loss 0.197458 LR 0.000063 Time 0.027633 -2023-02-13 18:42:59,634 - Epoch: [193][ 110/ 1207] Overall Loss 0.198849 Objective Loss 0.198849 LR 0.000063 Time 0.026886 -2023-02-13 18:42:59,829 - Epoch: [193][ 120/ 1207] Overall Loss 0.198801 Objective Loss 0.198801 LR 0.000063 Time 0.026268 -2023-02-13 18:43:00,024 - Epoch: [193][ 130/ 1207] Overall Loss 0.200514 Objective Loss 0.200514 LR 0.000063 Time 0.025748 -2023-02-13 18:43:00,220 - Epoch: [193][ 140/ 1207] Overall Loss 0.199094 Objective Loss 0.199094 LR 0.000063 Time 0.025304 -2023-02-13 18:43:00,415 - Epoch: [193][ 150/ 1207] Overall Loss 0.198244 Objective Loss 0.198244 LR 0.000063 Time 0.024911 -2023-02-13 18:43:00,610 - Epoch: [193][ 160/ 1207] Overall Loss 0.198042 Objective Loss 0.198042 LR 0.000063 Time 0.024570 -2023-02-13 18:43:00,804 - Epoch: [193][ 170/ 1207] Overall Loss 0.198084 Objective Loss 0.198084 LR 0.000063 Time 0.024268 -2023-02-13 18:43:01,003 - Epoch: [193][ 180/ 1207] Overall Loss 0.197990 Objective Loss 0.197990 LR 0.000063 Time 0.024021 -2023-02-13 18:43:01,198 - Epoch: [193][ 190/ 1207] Overall Loss 0.198206 Objective Loss 0.198206 LR 0.000063 Time 0.023782 -2023-02-13 18:43:01,393 - Epoch: [193][ 200/ 1207] Overall Loss 0.198381 Objective Loss 0.198381 LR 0.000063 Time 0.023566 -2023-02-13 18:43:01,588 - Epoch: [193][ 210/ 1207] Overall Loss 0.198839 Objective Loss 0.198839 LR 0.000063 Time 0.023372 -2023-02-13 18:43:01,784 - Epoch: [193][ 220/ 1207] Overall Loss 0.199009 Objective Loss 0.199009 LR 0.000063 Time 0.023195 -2023-02-13 18:43:01,980 - Epoch: [193][ 230/ 1207] Overall Loss 0.200177 Objective Loss 0.200177 LR 0.000063 Time 0.023037 -2023-02-13 18:43:02,176 - Epoch: [193][ 240/ 1207] Overall Loss 0.199866 Objective Loss 0.199866 LR 0.000063 Time 0.022894 -2023-02-13 18:43:02,371 - Epoch: [193][ 250/ 1207] Overall Loss 0.199446 Objective Loss 0.199446 LR 0.000063 Time 0.022756 -2023-02-13 18:43:02,566 - Epoch: [193][ 260/ 1207] Overall Loss 0.199816 Objective Loss 0.199816 LR 0.000063 Time 0.022628 -2023-02-13 18:43:02,761 - Epoch: [193][ 270/ 1207] Overall Loss 0.200153 Objective Loss 0.200153 LR 0.000063 Time 0.022511 -2023-02-13 18:43:02,957 - Epoch: [193][ 280/ 1207] Overall Loss 0.200096 Objective Loss 0.200096 LR 0.000063 Time 0.022405 -2023-02-13 18:43:03,152 - Epoch: [193][ 290/ 1207] Overall Loss 0.199534 Objective Loss 0.199534 LR 0.000063 Time 0.022307 -2023-02-13 18:43:03,348 - Epoch: [193][ 300/ 1207] Overall Loss 0.198812 Objective Loss 0.198812 LR 0.000063 Time 0.022214 -2023-02-13 18:43:03,543 - Epoch: [193][ 310/ 1207] Overall Loss 0.198422 Objective Loss 0.198422 LR 0.000063 Time 0.022125 -2023-02-13 18:43:03,738 - Epoch: [193][ 320/ 1207] Overall Loss 0.199223 Objective Loss 0.199223 LR 0.000063 Time 0.022042 -2023-02-13 18:43:03,933 - Epoch: [193][ 330/ 1207] Overall Loss 0.198690 Objective Loss 0.198690 LR 0.000063 Time 0.021964 -2023-02-13 18:43:04,130 - Epoch: [193][ 340/ 1207] Overall Loss 0.198705 Objective Loss 0.198705 LR 0.000063 Time 0.021895 -2023-02-13 18:43:04,325 - Epoch: [193][ 350/ 1207] Overall Loss 0.198535 Objective Loss 0.198535 LR 0.000063 Time 0.021827 -2023-02-13 18:43:04,519 - Epoch: [193][ 360/ 1207] Overall Loss 0.199051 Objective Loss 0.199051 LR 0.000063 Time 0.021758 -2023-02-13 18:43:04,708 - Epoch: [193][ 370/ 1207] Overall Loss 0.198954 Objective Loss 0.198954 LR 0.000063 Time 0.021680 -2023-02-13 18:43:04,897 - Epoch: [193][ 380/ 1207] Overall Loss 0.198888 Objective Loss 0.198888 LR 0.000063 Time 0.021606 -2023-02-13 18:43:05,087 - Epoch: [193][ 390/ 1207] Overall Loss 0.199023 Objective Loss 0.199023 LR 0.000063 Time 0.021537 -2023-02-13 18:43:05,279 - Epoch: [193][ 400/ 1207] Overall Loss 0.198945 Objective Loss 0.198945 LR 0.000063 Time 0.021479 -2023-02-13 18:43:05,475 - Epoch: [193][ 410/ 1207] Overall Loss 0.198337 Objective Loss 0.198337 LR 0.000063 Time 0.021431 -2023-02-13 18:43:05,678 - Epoch: [193][ 420/ 1207] Overall Loss 0.198430 Objective Loss 0.198430 LR 0.000063 Time 0.021404 -2023-02-13 18:43:05,875 - Epoch: [193][ 430/ 1207] Overall Loss 0.198306 Objective Loss 0.198306 LR 0.000063 Time 0.021362 -2023-02-13 18:43:06,073 - Epoch: [193][ 440/ 1207] Overall Loss 0.198027 Objective Loss 0.198027 LR 0.000063 Time 0.021327 -2023-02-13 18:43:06,270 - Epoch: [193][ 450/ 1207] Overall Loss 0.198248 Objective Loss 0.198248 LR 0.000063 Time 0.021290 -2023-02-13 18:43:06,467 - Epoch: [193][ 460/ 1207] Overall Loss 0.198053 Objective Loss 0.198053 LR 0.000063 Time 0.021253 -2023-02-13 18:43:06,664 - Epoch: [193][ 470/ 1207] Overall Loss 0.197758 Objective Loss 0.197758 LR 0.000063 Time 0.021219 -2023-02-13 18:43:06,860 - Epoch: [193][ 480/ 1207] Overall Loss 0.197714 Objective Loss 0.197714 LR 0.000063 Time 0.021186 -2023-02-13 18:43:07,058 - Epoch: [193][ 490/ 1207] Overall Loss 0.197695 Objective Loss 0.197695 LR 0.000063 Time 0.021156 -2023-02-13 18:43:07,255 - Epoch: [193][ 500/ 1207] Overall Loss 0.197182 Objective Loss 0.197182 LR 0.000063 Time 0.021126 -2023-02-13 18:43:07,453 - Epoch: [193][ 510/ 1207] Overall Loss 0.197147 Objective Loss 0.197147 LR 0.000063 Time 0.021100 -2023-02-13 18:43:07,652 - Epoch: [193][ 520/ 1207] Overall Loss 0.196640 Objective Loss 0.196640 LR 0.000063 Time 0.021076 -2023-02-13 18:43:07,854 - Epoch: [193][ 530/ 1207] Overall Loss 0.196340 Objective Loss 0.196340 LR 0.000063 Time 0.021059 -2023-02-13 18:43:08,053 - Epoch: [193][ 540/ 1207] Overall Loss 0.195834 Objective Loss 0.195834 LR 0.000063 Time 0.021037 -2023-02-13 18:43:08,250 - Epoch: [193][ 550/ 1207] Overall Loss 0.195902 Objective Loss 0.195902 LR 0.000063 Time 0.021012 -2023-02-13 18:43:08,444 - Epoch: [193][ 560/ 1207] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000063 Time 0.020981 -2023-02-13 18:43:08,640 - Epoch: [193][ 570/ 1207] Overall Loss 0.195610 Objective Loss 0.195610 LR 0.000063 Time 0.020957 -2023-02-13 18:43:08,833 - Epoch: [193][ 580/ 1207] Overall Loss 0.195596 Objective Loss 0.195596 LR 0.000063 Time 0.020928 -2023-02-13 18:43:09,030 - Epoch: [193][ 590/ 1207] Overall Loss 0.195749 Objective Loss 0.195749 LR 0.000063 Time 0.020906 -2023-02-13 18:43:09,224 - Epoch: [193][ 600/ 1207] Overall Loss 0.196043 Objective Loss 0.196043 LR 0.000063 Time 0.020881 -2023-02-13 18:43:09,421 - Epoch: [193][ 610/ 1207] Overall Loss 0.196228 Objective Loss 0.196228 LR 0.000063 Time 0.020861 -2023-02-13 18:43:09,615 - Epoch: [193][ 620/ 1207] Overall Loss 0.196473 Objective Loss 0.196473 LR 0.000063 Time 0.020836 -2023-02-13 18:43:09,811 - Epoch: [193][ 630/ 1207] Overall Loss 0.196477 Objective Loss 0.196477 LR 0.000063 Time 0.020816 -2023-02-13 18:43:10,005 - Epoch: [193][ 640/ 1207] Overall Loss 0.196566 Objective Loss 0.196566 LR 0.000063 Time 0.020793 -2023-02-13 18:43:10,202 - Epoch: [193][ 650/ 1207] Overall Loss 0.196298 Objective Loss 0.196298 LR 0.000063 Time 0.020776 -2023-02-13 18:43:10,397 - Epoch: [193][ 660/ 1207] Overall Loss 0.196038 Objective Loss 0.196038 LR 0.000063 Time 0.020756 -2023-02-13 18:43:10,594 - Epoch: [193][ 670/ 1207] Overall Loss 0.196086 Objective Loss 0.196086 LR 0.000063 Time 0.020740 -2023-02-13 18:43:10,787 - Epoch: [193][ 680/ 1207] Overall Loss 0.196159 Objective Loss 0.196159 LR 0.000063 Time 0.020718 -2023-02-13 18:43:10,985 - Epoch: [193][ 690/ 1207] Overall Loss 0.195951 Objective Loss 0.195951 LR 0.000063 Time 0.020704 -2023-02-13 18:43:11,179 - Epoch: [193][ 700/ 1207] Overall Loss 0.195868 Objective Loss 0.195868 LR 0.000063 Time 0.020685 -2023-02-13 18:43:11,377 - Epoch: [193][ 710/ 1207] Overall Loss 0.195838 Objective Loss 0.195838 LR 0.000063 Time 0.020672 -2023-02-13 18:43:11,571 - Epoch: [193][ 720/ 1207] Overall Loss 0.195761 Objective Loss 0.195761 LR 0.000063 Time 0.020653 -2023-02-13 18:43:11,767 - Epoch: [193][ 730/ 1207] Overall Loss 0.195755 Objective Loss 0.195755 LR 0.000063 Time 0.020640 -2023-02-13 18:43:11,962 - Epoch: [193][ 740/ 1207] Overall Loss 0.195489 Objective Loss 0.195489 LR 0.000063 Time 0.020623 -2023-02-13 18:43:12,159 - Epoch: [193][ 750/ 1207] Overall Loss 0.195541 Objective Loss 0.195541 LR 0.000063 Time 0.020611 -2023-02-13 18:43:12,353 - Epoch: [193][ 760/ 1207] Overall Loss 0.195752 Objective Loss 0.195752 LR 0.000063 Time 0.020593 -2023-02-13 18:43:12,550 - Epoch: [193][ 770/ 1207] Overall Loss 0.195943 Objective Loss 0.195943 LR 0.000063 Time 0.020582 -2023-02-13 18:43:12,744 - Epoch: [193][ 780/ 1207] Overall Loss 0.195842 Objective Loss 0.195842 LR 0.000063 Time 0.020566 -2023-02-13 18:43:12,941 - Epoch: [193][ 790/ 1207] Overall Loss 0.195940 Objective Loss 0.195940 LR 0.000063 Time 0.020555 -2023-02-13 18:43:13,135 - Epoch: [193][ 800/ 1207] Overall Loss 0.195880 Objective Loss 0.195880 LR 0.000063 Time 0.020540 -2023-02-13 18:43:13,333 - Epoch: [193][ 810/ 1207] Overall Loss 0.195835 Objective Loss 0.195835 LR 0.000063 Time 0.020530 -2023-02-13 18:43:13,527 - Epoch: [193][ 820/ 1207] Overall Loss 0.195818 Objective Loss 0.195818 LR 0.000063 Time 0.020516 -2023-02-13 18:43:13,724 - Epoch: [193][ 830/ 1207] Overall Loss 0.195756 Objective Loss 0.195756 LR 0.000063 Time 0.020505 -2023-02-13 18:43:13,917 - Epoch: [193][ 840/ 1207] Overall Loss 0.195766 Objective Loss 0.195766 LR 0.000063 Time 0.020491 -2023-02-13 18:43:14,115 - Epoch: [193][ 850/ 1207] Overall Loss 0.195654 Objective Loss 0.195654 LR 0.000063 Time 0.020483 -2023-02-13 18:43:14,309 - Epoch: [193][ 860/ 1207] Overall Loss 0.195513 Objective Loss 0.195513 LR 0.000063 Time 0.020469 -2023-02-13 18:43:14,506 - Epoch: [193][ 870/ 1207] Overall Loss 0.195580 Objective Loss 0.195580 LR 0.000063 Time 0.020460 -2023-02-13 18:43:14,699 - Epoch: [193][ 880/ 1207] Overall Loss 0.195550 Objective Loss 0.195550 LR 0.000063 Time 0.020447 -2023-02-13 18:43:14,896 - Epoch: [193][ 890/ 1207] Overall Loss 0.195494 Objective Loss 0.195494 LR 0.000063 Time 0.020438 -2023-02-13 18:43:15,091 - Epoch: [193][ 900/ 1207] Overall Loss 0.195709 Objective Loss 0.195709 LR 0.000063 Time 0.020426 -2023-02-13 18:43:15,288 - Epoch: [193][ 910/ 1207] Overall Loss 0.195509 Objective Loss 0.195509 LR 0.000063 Time 0.020418 -2023-02-13 18:43:15,481 - Epoch: [193][ 920/ 1207] Overall Loss 0.195597 Objective Loss 0.195597 LR 0.000063 Time 0.020406 -2023-02-13 18:43:15,678 - Epoch: [193][ 930/ 1207] Overall Loss 0.195615 Objective Loss 0.195615 LR 0.000063 Time 0.020398 -2023-02-13 18:43:15,872 - Epoch: [193][ 940/ 1207] Overall Loss 0.195616 Objective Loss 0.195616 LR 0.000063 Time 0.020386 -2023-02-13 18:43:16,070 - Epoch: [193][ 950/ 1207] Overall Loss 0.195753 Objective Loss 0.195753 LR 0.000063 Time 0.020381 -2023-02-13 18:43:16,264 - Epoch: [193][ 960/ 1207] Overall Loss 0.195713 Objective Loss 0.195713 LR 0.000063 Time 0.020370 -2023-02-13 18:43:16,461 - Epoch: [193][ 970/ 1207] Overall Loss 0.195791 Objective Loss 0.195791 LR 0.000063 Time 0.020362 -2023-02-13 18:43:16,655 - Epoch: [193][ 980/ 1207] Overall Loss 0.195611 Objective Loss 0.195611 LR 0.000063 Time 0.020352 -2023-02-13 18:43:16,852 - Epoch: [193][ 990/ 1207] Overall Loss 0.195659 Objective Loss 0.195659 LR 0.000063 Time 0.020345 -2023-02-13 18:43:17,046 - Epoch: [193][ 1000/ 1207] Overall Loss 0.195968 Objective Loss 0.195968 LR 0.000063 Time 0.020335 -2023-02-13 18:43:17,244 - Epoch: [193][ 1010/ 1207] Overall Loss 0.196126 Objective Loss 0.196126 LR 0.000063 Time 0.020330 -2023-02-13 18:43:17,438 - Epoch: [193][ 1020/ 1207] Overall Loss 0.196021 Objective Loss 0.196021 LR 0.000063 Time 0.020320 -2023-02-13 18:43:17,635 - Epoch: [193][ 1030/ 1207] Overall Loss 0.195972 Objective Loss 0.195972 LR 0.000063 Time 0.020313 -2023-02-13 18:43:17,829 - Epoch: [193][ 1040/ 1207] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000063 Time 0.020304 -2023-02-13 18:43:18,025 - Epoch: [193][ 1050/ 1207] Overall Loss 0.195959 Objective Loss 0.195959 LR 0.000063 Time 0.020298 -2023-02-13 18:43:18,220 - Epoch: [193][ 1060/ 1207] Overall Loss 0.195977 Objective Loss 0.195977 LR 0.000063 Time 0.020290 -2023-02-13 18:43:18,418 - Epoch: [193][ 1070/ 1207] Overall Loss 0.195665 Objective Loss 0.195665 LR 0.000063 Time 0.020284 -2023-02-13 18:43:18,611 - Epoch: [193][ 1080/ 1207] Overall Loss 0.195907 Objective Loss 0.195907 LR 0.000063 Time 0.020275 -2023-02-13 18:43:18,808 - Epoch: [193][ 1090/ 1207] Overall Loss 0.195657 Objective Loss 0.195657 LR 0.000063 Time 0.020269 -2023-02-13 18:43:19,003 - Epoch: [193][ 1100/ 1207] Overall Loss 0.195502 Objective Loss 0.195502 LR 0.000063 Time 0.020262 -2023-02-13 18:43:19,201 - Epoch: [193][ 1110/ 1207] Overall Loss 0.195543 Objective Loss 0.195543 LR 0.000063 Time 0.020258 -2023-02-13 18:43:19,393 - Epoch: [193][ 1120/ 1207] Overall Loss 0.195574 Objective Loss 0.195574 LR 0.000063 Time 0.020248 -2023-02-13 18:43:19,584 - Epoch: [193][ 1130/ 1207] Overall Loss 0.195662 Objective Loss 0.195662 LR 0.000063 Time 0.020238 -2023-02-13 18:43:19,775 - Epoch: [193][ 1140/ 1207] Overall Loss 0.195521 Objective Loss 0.195521 LR 0.000063 Time 0.020227 -2023-02-13 18:43:19,966 - Epoch: [193][ 1150/ 1207] Overall Loss 0.195640 Objective Loss 0.195640 LR 0.000063 Time 0.020217 -2023-02-13 18:43:20,158 - Epoch: [193][ 1160/ 1207] Overall Loss 0.195533 Objective Loss 0.195533 LR 0.000063 Time 0.020208 -2023-02-13 18:43:20,349 - Epoch: [193][ 1170/ 1207] Overall Loss 0.195314 Objective Loss 0.195314 LR 0.000063 Time 0.020198 -2023-02-13 18:43:20,540 - Epoch: [193][ 1180/ 1207] Overall Loss 0.195319 Objective Loss 0.195319 LR 0.000063 Time 0.020188 -2023-02-13 18:43:20,729 - Epoch: [193][ 1190/ 1207] Overall Loss 0.195296 Objective Loss 0.195296 LR 0.000063 Time 0.020178 -2023-02-13 18:43:20,970 - Epoch: [193][ 1200/ 1207] Overall Loss 0.195265 Objective Loss 0.195265 LR 0.000063 Time 0.020210 -2023-02-13 18:43:21,085 - Epoch: [193][ 1207/ 1207] Overall Loss 0.195245 Objective Loss 0.195245 Top1 85.365854 Top5 98.170732 LR 0.000063 Time 0.020187 -2023-02-13 18:43:21,157 - --- validate (epoch=193)----------- -2023-02-13 18:43:21,157 - 34311 samples (256 per mini-batch) -2023-02-13 18:43:21,568 - Epoch: [193][ 10/ 135] Loss 0.281112 Top1 85.312500 Top5 98.242188 -2023-02-13 18:43:21,714 - Epoch: [193][ 20/ 135] Loss 0.298219 Top1 85.214844 Top5 97.890625 -2023-02-13 18:43:21,855 - Epoch: [193][ 30/ 135] Loss 0.303304 Top1 85.351562 Top5 97.760417 -2023-02-13 18:43:21,995 - Epoch: [193][ 40/ 135] Loss 0.301983 Top1 85.468750 Top5 97.773438 -2023-02-13 18:43:22,123 - Epoch: [193][ 50/ 135] Loss 0.299231 Top1 85.523438 Top5 97.765625 -2023-02-13 18:43:22,251 - Epoch: [193][ 60/ 135] Loss 0.292842 Top1 85.618490 Top5 97.851562 -2023-02-13 18:43:22,375 - Epoch: [193][ 70/ 135] Loss 0.291409 Top1 85.680804 Top5 97.929688 -2023-02-13 18:43:22,513 - Epoch: [193][ 80/ 135] Loss 0.291594 Top1 85.742188 Top5 97.910156 -2023-02-13 18:43:22,654 - Epoch: [193][ 90/ 135] Loss 0.292472 Top1 85.655382 Top5 97.864583 -2023-02-13 18:43:22,800 - Epoch: [193][ 100/ 135] Loss 0.290456 Top1 85.703125 Top5 97.890625 -2023-02-13 18:43:22,926 - Epoch: [193][ 110/ 135] Loss 0.284695 Top1 85.681818 Top5 97.943892 -2023-02-13 18:43:23,055 - Epoch: [193][ 120/ 135] Loss 0.288238 Top1 85.651042 Top5 97.900391 -2023-02-13 18:43:23,190 - Epoch: [193][ 130/ 135] Loss 0.289536 Top1 85.552885 Top5 97.920673 -2023-02-13 18:43:23,237 - Epoch: [193][ 135/ 135] Loss 0.287929 Top1 85.587712 Top5 97.898633 -2023-02-13 18:43:23,307 - ==> Top1: 85.588 Top5: 97.899 Loss: 0.288 - -2023-02-13 18:43:23,307 - ==> Confusion: -[[ 863 5 7 0 10 1 0 0 4 48 0 3 1 4 7 2 1 1 0 3 7] - [ 5 960 1 1 12 14 2 12 2 1 2 0 0 0 0 2 5 0 2 4 8] - [ 9 7 962 12 3 1 13 12 1 1 4 2 1 3 3 4 4 2 4 3 7] - [ 5 2 23 899 2 4 0 1 4 3 16 0 7 0 18 1 4 4 14 0 9] - [ 12 8 0 0 1003 6 1 2 0 0 0 6 2 4 6 4 3 1 1 3 4] - [ 2 14 1 3 5 977 3 11 1 5 1 9 5 12 0 3 7 2 2 3 4] - [ 4 1 11 1 0 5 1053 4 0 1 1 0 2 3 1 1 2 1 1 4 3] - [ 5 7 10 2 1 27 1 937 1 1 0 4 2 2 0 0 2 1 13 3 5] - [ 12 2 1 1 1 0 0 1 917 42 5 3 0 8 8 2 1 0 3 1 1] - [ 63 0 2 0 8 0 0 0 27 883 0 0 0 15 3 2 2 0 3 0 4] - [ 2 0 2 4 3 1 2 3 19 1 990 1 2 7 3 0 1 1 5 0 4] - [ 3 2 3 0 2 11 0 4 1 4 0 914 21 7 0 8 3 12 1 7 2] - [ 1 0 0 9 2 4 0 0 2 1 0 30 876 2 3 4 4 14 1 1 5] - [ 6 2 2 0 5 3 1 1 10 24 10 3 1 939 5 4 3 0 0 0 5] - [ 5 2 0 11 4 5 0 1 23 13 2 0 2 2 1004 0 2 4 4 0 8] - [ 6 2 7 1 5 0 4 1 0 0 0 4 9 3 0 970 12 8 0 7 7] - [ 0 6 0 1 9 2 0 0 2 1 0 1 2 2 3 9 1005 3 1 3 11] - [ 4 2 0 4 1 1 2 0 0 0 1 12 11 2 0 16 0 987 0 1 7] - [ 7 5 3 5 2 3 0 20 4 1 5 1 2 0 14 1 1 1 1007 2 2] - [ 0 4 0 0 0 4 10 9 0 0 0 17 2 3 1 6 6 3 0 1078 5] - [ 172 234 219 92 144 172 88 157 116 106 184 95 268 274 163 78 243 98 152 237 10142]] - -2023-02-13 18:43:23,309 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:43:23,309 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:43:23,315 - - -2023-02-13 18:43:23,315 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:43:24,210 - Epoch: [194][ 10/ 1207] Overall Loss 0.215181 Objective Loss 0.215181 LR 0.000063 Time 0.089376 -2023-02-13 18:43:24,410 - Epoch: [194][ 20/ 1207] Overall Loss 0.212096 Objective Loss 0.212096 LR 0.000063 Time 0.054693 -2023-02-13 18:43:24,599 - Epoch: [194][ 30/ 1207] Overall Loss 0.206667 Objective Loss 0.206667 LR 0.000063 Time 0.042765 -2023-02-13 18:43:24,789 - Epoch: [194][ 40/ 1207] Overall Loss 0.205672 Objective Loss 0.205672 LR 0.000063 Time 0.036800 -2023-02-13 18:43:24,978 - Epoch: [194][ 50/ 1207] Overall Loss 0.211308 Objective Loss 0.211308 LR 0.000063 Time 0.033205 -2023-02-13 18:43:25,168 - Epoch: [194][ 60/ 1207] Overall Loss 0.209004 Objective Loss 0.209004 LR 0.000063 Time 0.030834 -2023-02-13 18:43:25,356 - Epoch: [194][ 70/ 1207] Overall Loss 0.210019 Objective Loss 0.210019 LR 0.000063 Time 0.029116 -2023-02-13 18:43:25,545 - Epoch: [194][ 80/ 1207] Overall Loss 0.207456 Objective Loss 0.207456 LR 0.000063 Time 0.027836 -2023-02-13 18:43:25,735 - Epoch: [194][ 90/ 1207] Overall Loss 0.207222 Objective Loss 0.207222 LR 0.000063 Time 0.026843 -2023-02-13 18:43:25,924 - Epoch: [194][ 100/ 1207] Overall Loss 0.204303 Objective Loss 0.204303 LR 0.000063 Time 0.026047 -2023-02-13 18:43:26,115 - Epoch: [194][ 110/ 1207] Overall Loss 0.204591 Objective Loss 0.204591 LR 0.000063 Time 0.025410 -2023-02-13 18:43:26,304 - Epoch: [194][ 120/ 1207] Overall Loss 0.204094 Objective Loss 0.204094 LR 0.000063 Time 0.024864 -2023-02-13 18:43:26,492 - Epoch: [194][ 130/ 1207] Overall Loss 0.203882 Objective Loss 0.203882 LR 0.000063 Time 0.024400 -2023-02-13 18:43:26,682 - Epoch: [194][ 140/ 1207] Overall Loss 0.203295 Objective Loss 0.203295 LR 0.000063 Time 0.024007 -2023-02-13 18:43:26,870 - Epoch: [194][ 150/ 1207] Overall Loss 0.202554 Objective Loss 0.202554 LR 0.000063 Time 0.023661 -2023-02-13 18:43:27,060 - Epoch: [194][ 160/ 1207] Overall Loss 0.202412 Objective Loss 0.202412 LR 0.000063 Time 0.023364 -2023-02-13 18:43:27,249 - Epoch: [194][ 170/ 1207] Overall Loss 0.201719 Objective Loss 0.201719 LR 0.000063 Time 0.023102 -2023-02-13 18:43:27,438 - Epoch: [194][ 180/ 1207] Overall Loss 0.202205 Objective Loss 0.202205 LR 0.000063 Time 0.022865 -2023-02-13 18:43:27,626 - Epoch: [194][ 190/ 1207] Overall Loss 0.201510 Objective Loss 0.201510 LR 0.000063 Time 0.022652 -2023-02-13 18:43:27,815 - Epoch: [194][ 200/ 1207] Overall Loss 0.203385 Objective Loss 0.203385 LR 0.000063 Time 0.022460 -2023-02-13 18:43:28,003 - Epoch: [194][ 210/ 1207] Overall Loss 0.203133 Objective Loss 0.203133 LR 0.000063 Time 0.022285 -2023-02-13 18:43:28,193 - Epoch: [194][ 220/ 1207] Overall Loss 0.203169 Objective Loss 0.203169 LR 0.000063 Time 0.022135 -2023-02-13 18:43:28,382 - Epoch: [194][ 230/ 1207] Overall Loss 0.202080 Objective Loss 0.202080 LR 0.000063 Time 0.021990 -2023-02-13 18:43:28,571 - Epoch: [194][ 240/ 1207] Overall Loss 0.201756 Objective Loss 0.201756 LR 0.000063 Time 0.021860 -2023-02-13 18:43:28,759 - Epoch: [194][ 250/ 1207] Overall Loss 0.201150 Objective Loss 0.201150 LR 0.000063 Time 0.021739 -2023-02-13 18:43:28,949 - Epoch: [194][ 260/ 1207] Overall Loss 0.200636 Objective Loss 0.200636 LR 0.000063 Time 0.021630 -2023-02-13 18:43:29,138 - Epoch: [194][ 270/ 1207] Overall Loss 0.200464 Objective Loss 0.200464 LR 0.000063 Time 0.021529 -2023-02-13 18:43:29,328 - Epoch: [194][ 280/ 1207] Overall Loss 0.200532 Objective Loss 0.200532 LR 0.000063 Time 0.021436 -2023-02-13 18:43:29,516 - Epoch: [194][ 290/ 1207] Overall Loss 0.200184 Objective Loss 0.200184 LR 0.000063 Time 0.021344 -2023-02-13 18:43:29,705 - Epoch: [194][ 300/ 1207] Overall Loss 0.200329 Objective Loss 0.200329 LR 0.000063 Time 0.021263 -2023-02-13 18:43:29,894 - Epoch: [194][ 310/ 1207] Overall Loss 0.200781 Objective Loss 0.200781 LR 0.000063 Time 0.021185 -2023-02-13 18:43:30,083 - Epoch: [194][ 320/ 1207] Overall Loss 0.200058 Objective Loss 0.200058 LR 0.000063 Time 0.021112 -2023-02-13 18:43:30,272 - Epoch: [194][ 330/ 1207] Overall Loss 0.199888 Objective Loss 0.199888 LR 0.000063 Time 0.021045 -2023-02-13 18:43:30,462 - Epoch: [194][ 340/ 1207] Overall Loss 0.199320 Objective Loss 0.199320 LR 0.000063 Time 0.020982 -2023-02-13 18:43:30,650 - Epoch: [194][ 350/ 1207] Overall Loss 0.198958 Objective Loss 0.198958 LR 0.000063 Time 0.020921 -2023-02-13 18:43:30,840 - Epoch: [194][ 360/ 1207] Overall Loss 0.198742 Objective Loss 0.198742 LR 0.000063 Time 0.020866 -2023-02-13 18:43:31,031 - Epoch: [194][ 370/ 1207] Overall Loss 0.198572 Objective Loss 0.198572 LR 0.000063 Time 0.020817 -2023-02-13 18:43:31,221 - Epoch: [194][ 380/ 1207] Overall Loss 0.197884 Objective Loss 0.197884 LR 0.000063 Time 0.020769 -2023-02-13 18:43:31,410 - Epoch: [194][ 390/ 1207] Overall Loss 0.197510 Objective Loss 0.197510 LR 0.000063 Time 0.020719 -2023-02-13 18:43:31,599 - Epoch: [194][ 400/ 1207] Overall Loss 0.197225 Objective Loss 0.197225 LR 0.000063 Time 0.020673 -2023-02-13 18:43:31,788 - Epoch: [194][ 410/ 1207] Overall Loss 0.197487 Objective Loss 0.197487 LR 0.000063 Time 0.020627 -2023-02-13 18:43:31,977 - Epoch: [194][ 420/ 1207] Overall Loss 0.197178 Objective Loss 0.197178 LR 0.000063 Time 0.020587 -2023-02-13 18:43:32,167 - Epoch: [194][ 430/ 1207] Overall Loss 0.196983 Objective Loss 0.196983 LR 0.000063 Time 0.020548 -2023-02-13 18:43:32,356 - Epoch: [194][ 440/ 1207] Overall Loss 0.197104 Objective Loss 0.197104 LR 0.000063 Time 0.020510 -2023-02-13 18:43:32,545 - Epoch: [194][ 450/ 1207] Overall Loss 0.197384 Objective Loss 0.197384 LR 0.000063 Time 0.020474 -2023-02-13 18:43:32,734 - Epoch: [194][ 460/ 1207] Overall Loss 0.197088 Objective Loss 0.197088 LR 0.000063 Time 0.020439 -2023-02-13 18:43:32,923 - Epoch: [194][ 470/ 1207] Overall Loss 0.196934 Objective Loss 0.196934 LR 0.000063 Time 0.020405 -2023-02-13 18:43:33,113 - Epoch: [194][ 480/ 1207] Overall Loss 0.196914 Objective Loss 0.196914 LR 0.000063 Time 0.020375 -2023-02-13 18:43:33,303 - Epoch: [194][ 490/ 1207] Overall Loss 0.196750 Objective Loss 0.196750 LR 0.000063 Time 0.020345 -2023-02-13 18:43:33,492 - Epoch: [194][ 500/ 1207] Overall Loss 0.196723 Objective Loss 0.196723 LR 0.000063 Time 0.020317 -2023-02-13 18:43:33,681 - Epoch: [194][ 510/ 1207] Overall Loss 0.196323 Objective Loss 0.196323 LR 0.000063 Time 0.020288 -2023-02-13 18:43:33,870 - Epoch: [194][ 520/ 1207] Overall Loss 0.196524 Objective Loss 0.196524 LR 0.000063 Time 0.020261 -2023-02-13 18:43:34,059 - Epoch: [194][ 530/ 1207] Overall Loss 0.196034 Objective Loss 0.196034 LR 0.000063 Time 0.020235 -2023-02-13 18:43:34,250 - Epoch: [194][ 540/ 1207] Overall Loss 0.196030 Objective Loss 0.196030 LR 0.000063 Time 0.020213 -2023-02-13 18:43:34,445 - Epoch: [194][ 550/ 1207] Overall Loss 0.196038 Objective Loss 0.196038 LR 0.000063 Time 0.020198 -2023-02-13 18:43:34,640 - Epoch: [194][ 560/ 1207] Overall Loss 0.195925 Objective Loss 0.195925 LR 0.000063 Time 0.020185 -2023-02-13 18:43:34,832 - Epoch: [194][ 570/ 1207] Overall Loss 0.195442 Objective Loss 0.195442 LR 0.000063 Time 0.020167 -2023-02-13 18:43:35,021 - Epoch: [194][ 580/ 1207] Overall Loss 0.195103 Objective Loss 0.195103 LR 0.000063 Time 0.020145 -2023-02-13 18:43:35,211 - Epoch: [194][ 590/ 1207] Overall Loss 0.195207 Objective Loss 0.195207 LR 0.000063 Time 0.020125 -2023-02-13 18:43:35,400 - Epoch: [194][ 600/ 1207] Overall Loss 0.195424 Objective Loss 0.195424 LR 0.000063 Time 0.020105 -2023-02-13 18:43:35,590 - Epoch: [194][ 610/ 1207] Overall Loss 0.195554 Objective Loss 0.195554 LR 0.000063 Time 0.020085 -2023-02-13 18:43:35,779 - Epoch: [194][ 620/ 1207] Overall Loss 0.195549 Objective Loss 0.195549 LR 0.000063 Time 0.020065 -2023-02-13 18:43:35,969 - Epoch: [194][ 630/ 1207] Overall Loss 0.195713 Objective Loss 0.195713 LR 0.000063 Time 0.020049 -2023-02-13 18:43:36,159 - Epoch: [194][ 640/ 1207] Overall Loss 0.195670 Objective Loss 0.195670 LR 0.000063 Time 0.020032 -2023-02-13 18:43:36,349 - Epoch: [194][ 650/ 1207] Overall Loss 0.195524 Objective Loss 0.195524 LR 0.000063 Time 0.020015 -2023-02-13 18:43:36,539 - Epoch: [194][ 660/ 1207] Overall Loss 0.195806 Objective Loss 0.195806 LR 0.000063 Time 0.019999 -2023-02-13 18:43:36,728 - Epoch: [194][ 670/ 1207] Overall Loss 0.195698 Objective Loss 0.195698 LR 0.000063 Time 0.019982 -2023-02-13 18:43:36,917 - Epoch: [194][ 680/ 1207] Overall Loss 0.195515 Objective Loss 0.195515 LR 0.000063 Time 0.019965 -2023-02-13 18:43:37,106 - Epoch: [194][ 690/ 1207] Overall Loss 0.195474 Objective Loss 0.195474 LR 0.000063 Time 0.019950 -2023-02-13 18:43:37,296 - Epoch: [194][ 700/ 1207] Overall Loss 0.195301 Objective Loss 0.195301 LR 0.000063 Time 0.019935 -2023-02-13 18:43:37,484 - Epoch: [194][ 710/ 1207] Overall Loss 0.195132 Objective Loss 0.195132 LR 0.000063 Time 0.019919 -2023-02-13 18:43:37,674 - Epoch: [194][ 720/ 1207] Overall Loss 0.195231 Objective Loss 0.195231 LR 0.000063 Time 0.019906 -2023-02-13 18:43:37,863 - Epoch: [194][ 730/ 1207] Overall Loss 0.195095 Objective Loss 0.195095 LR 0.000063 Time 0.019891 -2023-02-13 18:43:38,052 - Epoch: [194][ 740/ 1207] Overall Loss 0.194939 Objective Loss 0.194939 LR 0.000063 Time 0.019877 -2023-02-13 18:43:38,241 - Epoch: [194][ 750/ 1207] Overall Loss 0.194801 Objective Loss 0.194801 LR 0.000063 Time 0.019864 -2023-02-13 18:43:38,430 - Epoch: [194][ 760/ 1207] Overall Loss 0.194771 Objective Loss 0.194771 LR 0.000063 Time 0.019850 -2023-02-13 18:43:38,618 - Epoch: [194][ 770/ 1207] Overall Loss 0.195041 Objective Loss 0.195041 LR 0.000063 Time 0.019837 -2023-02-13 18:43:38,808 - Epoch: [194][ 780/ 1207] Overall Loss 0.195128 Objective Loss 0.195128 LR 0.000063 Time 0.019825 -2023-02-13 18:43:38,996 - Epoch: [194][ 790/ 1207] Overall Loss 0.195061 Objective Loss 0.195061 LR 0.000063 Time 0.019812 -2023-02-13 18:43:39,186 - Epoch: [194][ 800/ 1207] Overall Loss 0.195171 Objective Loss 0.195171 LR 0.000063 Time 0.019802 -2023-02-13 18:43:39,375 - Epoch: [194][ 810/ 1207] Overall Loss 0.195315 Objective Loss 0.195315 LR 0.000063 Time 0.019790 -2023-02-13 18:43:39,565 - Epoch: [194][ 820/ 1207] Overall Loss 0.195110 Objective Loss 0.195110 LR 0.000063 Time 0.019779 -2023-02-13 18:43:39,754 - Epoch: [194][ 830/ 1207] Overall Loss 0.194990 Objective Loss 0.194990 LR 0.000063 Time 0.019768 -2023-02-13 18:43:39,943 - Epoch: [194][ 840/ 1207] Overall Loss 0.194777 Objective Loss 0.194777 LR 0.000063 Time 0.019757 -2023-02-13 18:43:40,132 - Epoch: [194][ 850/ 1207] Overall Loss 0.195212 Objective Loss 0.195212 LR 0.000063 Time 0.019747 -2023-02-13 18:43:40,323 - Epoch: [194][ 860/ 1207] Overall Loss 0.195348 Objective Loss 0.195348 LR 0.000063 Time 0.019740 -2023-02-13 18:43:40,515 - Epoch: [194][ 870/ 1207] Overall Loss 0.195291 Objective Loss 0.195291 LR 0.000063 Time 0.019733 -2023-02-13 18:43:40,706 - Epoch: [194][ 880/ 1207] Overall Loss 0.195203 Objective Loss 0.195203 LR 0.000063 Time 0.019725 -2023-02-13 18:43:40,898 - Epoch: [194][ 890/ 1207] Overall Loss 0.195217 Objective Loss 0.195217 LR 0.000063 Time 0.019718 -2023-02-13 18:43:41,090 - Epoch: [194][ 900/ 1207] Overall Loss 0.195472 Objective Loss 0.195472 LR 0.000063 Time 0.019712 -2023-02-13 18:43:41,282 - Epoch: [194][ 910/ 1207] Overall Loss 0.195452 Objective Loss 0.195452 LR 0.000063 Time 0.019707 -2023-02-13 18:43:41,473 - Epoch: [194][ 920/ 1207] Overall Loss 0.195500 Objective Loss 0.195500 LR 0.000063 Time 0.019700 -2023-02-13 18:43:41,665 - Epoch: [194][ 930/ 1207] Overall Loss 0.195665 Objective Loss 0.195665 LR 0.000063 Time 0.019694 -2023-02-13 18:43:41,856 - Epoch: [194][ 940/ 1207] Overall Loss 0.195745 Objective Loss 0.195745 LR 0.000063 Time 0.019687 -2023-02-13 18:43:42,048 - Epoch: [194][ 950/ 1207] Overall Loss 0.195602 Objective Loss 0.195602 LR 0.000063 Time 0.019682 -2023-02-13 18:43:42,240 - Epoch: [194][ 960/ 1207] Overall Loss 0.195697 Objective Loss 0.195697 LR 0.000063 Time 0.019676 -2023-02-13 18:43:42,432 - Epoch: [194][ 970/ 1207] Overall Loss 0.195785 Objective Loss 0.195785 LR 0.000063 Time 0.019670 -2023-02-13 18:43:42,622 - Epoch: [194][ 980/ 1207] Overall Loss 0.195871 Objective Loss 0.195871 LR 0.000063 Time 0.019664 -2023-02-13 18:43:42,814 - Epoch: [194][ 990/ 1207] Overall Loss 0.195789 Objective Loss 0.195789 LR 0.000063 Time 0.019658 -2023-02-13 18:43:43,006 - Epoch: [194][ 1000/ 1207] Overall Loss 0.195804 Objective Loss 0.195804 LR 0.000063 Time 0.019654 -2023-02-13 18:43:43,199 - Epoch: [194][ 1010/ 1207] Overall Loss 0.195724 Objective Loss 0.195724 LR 0.000063 Time 0.019650 -2023-02-13 18:43:43,390 - Epoch: [194][ 1020/ 1207] Overall Loss 0.195802 Objective Loss 0.195802 LR 0.000063 Time 0.019644 -2023-02-13 18:43:43,584 - Epoch: [194][ 1030/ 1207] Overall Loss 0.195913 Objective Loss 0.195913 LR 0.000063 Time 0.019641 -2023-02-13 18:43:43,775 - Epoch: [194][ 1040/ 1207] Overall Loss 0.195934 Objective Loss 0.195934 LR 0.000063 Time 0.019635 -2023-02-13 18:43:43,967 - Epoch: [194][ 1050/ 1207] Overall Loss 0.195893 Objective Loss 0.195893 LR 0.000063 Time 0.019631 -2023-02-13 18:43:44,159 - Epoch: [194][ 1060/ 1207] Overall Loss 0.195946 Objective Loss 0.195946 LR 0.000063 Time 0.019627 -2023-02-13 18:43:44,351 - Epoch: [194][ 1070/ 1207] Overall Loss 0.195948 Objective Loss 0.195948 LR 0.000063 Time 0.019622 -2023-02-13 18:43:44,542 - Epoch: [194][ 1080/ 1207] Overall Loss 0.195848 Objective Loss 0.195848 LR 0.000063 Time 0.019617 -2023-02-13 18:43:44,734 - Epoch: [194][ 1090/ 1207] Overall Loss 0.195941 Objective Loss 0.195941 LR 0.000063 Time 0.019613 -2023-02-13 18:43:44,926 - Epoch: [194][ 1100/ 1207] Overall Loss 0.195844 Objective Loss 0.195844 LR 0.000063 Time 0.019608 -2023-02-13 18:43:45,118 - Epoch: [194][ 1110/ 1207] Overall Loss 0.195662 Objective Loss 0.195662 LR 0.000063 Time 0.019605 -2023-02-13 18:43:45,310 - Epoch: [194][ 1120/ 1207] Overall Loss 0.195483 Objective Loss 0.195483 LR 0.000063 Time 0.019601 -2023-02-13 18:43:45,502 - Epoch: [194][ 1130/ 1207] Overall Loss 0.195509 Objective Loss 0.195509 LR 0.000063 Time 0.019597 -2023-02-13 18:43:45,694 - Epoch: [194][ 1140/ 1207] Overall Loss 0.195694 Objective Loss 0.195694 LR 0.000063 Time 0.019593 -2023-02-13 18:43:45,886 - Epoch: [194][ 1150/ 1207] Overall Loss 0.195729 Objective Loss 0.195729 LR 0.000063 Time 0.019589 -2023-02-13 18:43:46,079 - Epoch: [194][ 1160/ 1207] Overall Loss 0.195792 Objective Loss 0.195792 LR 0.000063 Time 0.019586 -2023-02-13 18:43:46,269 - Epoch: [194][ 1170/ 1207] Overall Loss 0.195884 Objective Loss 0.195884 LR 0.000063 Time 0.019582 -2023-02-13 18:43:46,459 - Epoch: [194][ 1180/ 1207] Overall Loss 0.195741 Objective Loss 0.195741 LR 0.000063 Time 0.019576 -2023-02-13 18:43:46,649 - Epoch: [194][ 1190/ 1207] Overall Loss 0.195766 Objective Loss 0.195766 LR 0.000063 Time 0.019571 -2023-02-13 18:43:46,889 - Epoch: [194][ 1200/ 1207] Overall Loss 0.195676 Objective Loss 0.195676 LR 0.000063 Time 0.019608 -2023-02-13 18:43:47,006 - Epoch: [194][ 1207/ 1207] Overall Loss 0.195799 Objective Loss 0.195799 Top1 86.280488 Top5 98.780488 LR 0.000063 Time 0.019590 -2023-02-13 18:43:47,077 - --- validate (epoch=194)----------- -2023-02-13 18:43:47,077 - 34311 samples (256 per mini-batch) -2023-02-13 18:43:47,482 - Epoch: [194][ 10/ 135] Loss 0.312140 Top1 85.000000 Top5 98.085938 -2023-02-13 18:43:47,613 - Epoch: [194][ 20/ 135] Loss 0.313343 Top1 84.902344 Top5 97.910156 -2023-02-13 18:43:47,739 - Epoch: [194][ 30/ 135] Loss 0.301253 Top1 85.169271 Top5 97.877604 -2023-02-13 18:43:47,870 - Epoch: [194][ 40/ 135] Loss 0.300031 Top1 85.195312 Top5 97.880859 -2023-02-13 18:43:47,997 - Epoch: [194][ 50/ 135] Loss 0.295825 Top1 85.437500 Top5 97.882812 -2023-02-13 18:43:48,124 - Epoch: [194][ 60/ 135] Loss 0.290360 Top1 85.592448 Top5 97.838542 -2023-02-13 18:43:48,263 - Epoch: [194][ 70/ 135] Loss 0.289983 Top1 85.552455 Top5 97.901786 -2023-02-13 18:43:48,393 - Epoch: [194][ 80/ 135] Loss 0.290372 Top1 85.620117 Top5 97.915039 -2023-02-13 18:43:48,522 - Epoch: [194][ 90/ 135] Loss 0.289166 Top1 85.638021 Top5 97.855903 -2023-02-13 18:43:48,652 - Epoch: [194][ 100/ 135] Loss 0.288177 Top1 85.625000 Top5 97.890625 -2023-02-13 18:43:48,782 - Epoch: [194][ 110/ 135] Loss 0.287898 Top1 85.617898 Top5 97.894176 -2023-02-13 18:43:48,914 - Epoch: [194][ 120/ 135] Loss 0.286676 Top1 85.638021 Top5 97.890625 -2023-02-13 18:43:49,043 - Epoch: [194][ 130/ 135] Loss 0.285737 Top1 85.658053 Top5 97.875601 -2023-02-13 18:43:49,087 - Epoch: [194][ 135/ 135] Loss 0.285342 Top1 85.643088 Top5 97.869488 -2023-02-13 18:43:49,172 - ==> Top1: 85.643 Top5: 97.869 Loss: 0.285 - -2023-02-13 18:43:49,172 - ==> Confusion: -[[ 859 4 6 1 11 2 0 1 4 45 1 4 1 6 7 3 2 2 0 1 7] - [ 4 964 1 1 9 17 0 10 2 0 2 1 1 0 0 1 6 1 2 4 7] - [ 6 7 965 12 3 2 9 12 1 1 2 4 1 5 3 7 5 1 4 2 6] - [ 6 1 21 913 0 3 0 2 2 1 9 0 7 0 21 2 3 6 14 0 5] - [ 9 8 0 0 1003 11 1 1 1 0 0 2 4 3 5 6 5 1 0 2 4] - [ 2 17 1 4 3 975 2 18 1 3 2 10 1 12 0 4 7 1 2 1 4] - [ 4 3 13 2 0 4 1043 5 0 1 1 0 2 1 1 2 2 2 2 6 5] - [ 3 10 8 0 1 29 1 937 0 1 0 5 4 2 0 0 2 1 12 5 3] - [ 15 4 2 1 1 0 1 1 901 31 9 2 0 14 18 2 2 0 4 0 1] - [ 73 1 3 0 12 1 0 2 30 857 1 0 0 18 3 2 1 2 1 0 5] - [ 2 1 3 5 2 1 2 5 15 2 984 1 1 10 2 0 2 1 7 0 5] - [ 2 2 3 0 2 13 0 5 2 2 0 923 19 7 0 5 3 8 1 6 2] - [ 0 0 1 7 2 5 0 0 2 0 0 27 881 2 1 8 4 9 1 1 8] - [ 3 2 1 1 5 10 0 1 7 15 7 3 4 945 7 3 2 0 0 2 6] - [ 2 3 1 15 6 4 0 1 15 8 3 0 3 2 1010 1 3 4 1 0 10] - [ 4 1 5 1 3 0 3 1 0 0 0 6 8 3 0 973 13 10 0 7 8] - [ 1 4 0 1 7 1 0 0 1 0 0 0 2 3 3 8 1015 0 2 3 10] - [ 4 2 0 4 0 1 2 0 1 1 0 11 14 2 1 15 0 986 0 2 5] - [ 4 4 6 7 0 2 0 22 1 2 4 2 2 0 11 0 1 4 1011 2 1] - [ 1 3 0 0 0 3 8 6 1 0 0 13 1 2 1 8 4 3 0 1087 7] - [ 141 238 213 107 146 187 66 176 82 71 161 97 284 286 169 70 273 105 174 235 10153]] - -2023-02-13 18:43:49,174 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:43:49,174 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:43:49,180 - - -2023-02-13 18:43:49,180 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:43:50,059 - Epoch: [195][ 10/ 1207] Overall Loss 0.200014 Objective Loss 0.200014 LR 0.000031 Time 0.087880 -2023-02-13 18:43:50,254 - Epoch: [195][ 20/ 1207] Overall Loss 0.200799 Objective Loss 0.200799 LR 0.000031 Time 0.053656 -2023-02-13 18:43:50,443 - Epoch: [195][ 30/ 1207] Overall Loss 0.195963 Objective Loss 0.195963 LR 0.000031 Time 0.042052 -2023-02-13 18:43:50,631 - Epoch: [195][ 40/ 1207] Overall Loss 0.188812 Objective Loss 0.188812 LR 0.000031 Time 0.036241 -2023-02-13 18:43:50,820 - Epoch: [195][ 50/ 1207] Overall Loss 0.190987 Objective Loss 0.190987 LR 0.000031 Time 0.032750 -2023-02-13 18:43:51,008 - Epoch: [195][ 60/ 1207] Overall Loss 0.193309 Objective Loss 0.193309 LR 0.000031 Time 0.030428 -2023-02-13 18:43:51,198 - Epoch: [195][ 70/ 1207] Overall Loss 0.191462 Objective Loss 0.191462 LR 0.000031 Time 0.028782 -2023-02-13 18:43:51,386 - Epoch: [195][ 80/ 1207] Overall Loss 0.192772 Objective Loss 0.192772 LR 0.000031 Time 0.027527 -2023-02-13 18:43:51,574 - Epoch: [195][ 90/ 1207] Overall Loss 0.191913 Objective Loss 0.191913 LR 0.000031 Time 0.026555 -2023-02-13 18:43:51,762 - Epoch: [195][ 100/ 1207] Overall Loss 0.192464 Objective Loss 0.192464 LR 0.000031 Time 0.025779 -2023-02-13 18:43:51,951 - Epoch: [195][ 110/ 1207] Overall Loss 0.191567 Objective Loss 0.191567 LR 0.000031 Time 0.025147 -2023-02-13 18:43:52,140 - Epoch: [195][ 120/ 1207] Overall Loss 0.192804 Objective Loss 0.192804 LR 0.000031 Time 0.024625 -2023-02-13 18:43:52,329 - Epoch: [195][ 130/ 1207] Overall Loss 0.192966 Objective Loss 0.192966 LR 0.000031 Time 0.024184 -2023-02-13 18:43:52,517 - Epoch: [195][ 140/ 1207] Overall Loss 0.192367 Objective Loss 0.192367 LR 0.000031 Time 0.023797 -2023-02-13 18:43:52,706 - Epoch: [195][ 150/ 1207] Overall Loss 0.190863 Objective Loss 0.190863 LR 0.000031 Time 0.023467 -2023-02-13 18:43:52,896 - Epoch: [195][ 160/ 1207] Overall Loss 0.192063 Objective Loss 0.192063 LR 0.000031 Time 0.023184 -2023-02-13 18:43:53,085 - Epoch: [195][ 170/ 1207] Overall Loss 0.191636 Objective Loss 0.191636 LR 0.000031 Time 0.022931 -2023-02-13 18:43:53,273 - Epoch: [195][ 180/ 1207] Overall Loss 0.191927 Objective Loss 0.191927 LR 0.000031 Time 0.022702 -2023-02-13 18:43:53,463 - Epoch: [195][ 190/ 1207] Overall Loss 0.191968 Objective Loss 0.191968 LR 0.000031 Time 0.022504 -2023-02-13 18:43:53,651 - Epoch: [195][ 200/ 1207] Overall Loss 0.191982 Objective Loss 0.191982 LR 0.000031 Time 0.022317 -2023-02-13 18:43:53,840 - Epoch: [195][ 210/ 1207] Overall Loss 0.192042 Objective Loss 0.192042 LR 0.000031 Time 0.022153 -2023-02-13 18:43:54,029 - Epoch: [195][ 220/ 1207] Overall Loss 0.193167 Objective Loss 0.193167 LR 0.000031 Time 0.022002 -2023-02-13 18:43:54,218 - Epoch: [195][ 230/ 1207] Overall Loss 0.192869 Objective Loss 0.192869 LR 0.000031 Time 0.021867 -2023-02-13 18:43:54,407 - Epoch: [195][ 240/ 1207] Overall Loss 0.192249 Objective Loss 0.192249 LR 0.000031 Time 0.021739 -2023-02-13 18:43:54,595 - Epoch: [195][ 250/ 1207] Overall Loss 0.192166 Objective Loss 0.192166 LR 0.000031 Time 0.021622 -2023-02-13 18:43:54,783 - Epoch: [195][ 260/ 1207] Overall Loss 0.191755 Objective Loss 0.191755 LR 0.000031 Time 0.021513 -2023-02-13 18:43:54,973 - Epoch: [195][ 270/ 1207] Overall Loss 0.191097 Objective Loss 0.191097 LR 0.000031 Time 0.021417 -2023-02-13 18:43:55,161 - Epoch: [195][ 280/ 1207] Overall Loss 0.191124 Objective Loss 0.191124 LR 0.000031 Time 0.021322 -2023-02-13 18:43:55,351 - Epoch: [195][ 290/ 1207] Overall Loss 0.191001 Objective Loss 0.191001 LR 0.000031 Time 0.021239 -2023-02-13 18:43:55,539 - Epoch: [195][ 300/ 1207] Overall Loss 0.191338 Objective Loss 0.191338 LR 0.000031 Time 0.021159 -2023-02-13 18:43:55,728 - Epoch: [195][ 310/ 1207] Overall Loss 0.192292 Objective Loss 0.192292 LR 0.000031 Time 0.021083 -2023-02-13 18:43:55,917 - Epoch: [195][ 320/ 1207] Overall Loss 0.192408 Objective Loss 0.192408 LR 0.000031 Time 0.021014 -2023-02-13 18:43:56,107 - Epoch: [195][ 330/ 1207] Overall Loss 0.192543 Objective Loss 0.192543 LR 0.000031 Time 0.020951 -2023-02-13 18:43:56,295 - Epoch: [195][ 340/ 1207] Overall Loss 0.192144 Objective Loss 0.192144 LR 0.000031 Time 0.020889 -2023-02-13 18:43:56,484 - Epoch: [195][ 350/ 1207] Overall Loss 0.191521 Objective Loss 0.191521 LR 0.000031 Time 0.020829 -2023-02-13 18:43:56,672 - Epoch: [195][ 360/ 1207] Overall Loss 0.191442 Objective Loss 0.191442 LR 0.000031 Time 0.020774 -2023-02-13 18:43:56,862 - Epoch: [195][ 370/ 1207] Overall Loss 0.191897 Objective Loss 0.191897 LR 0.000031 Time 0.020723 -2023-02-13 18:43:57,051 - Epoch: [195][ 380/ 1207] Overall Loss 0.191348 Objective Loss 0.191348 LR 0.000031 Time 0.020674 -2023-02-13 18:43:57,240 - Epoch: [195][ 390/ 1207] Overall Loss 0.191348 Objective Loss 0.191348 LR 0.000031 Time 0.020628 -2023-02-13 18:43:57,428 - Epoch: [195][ 400/ 1207] Overall Loss 0.190946 Objective Loss 0.190946 LR 0.000031 Time 0.020582 -2023-02-13 18:43:57,616 - Epoch: [195][ 410/ 1207] Overall Loss 0.190924 Objective Loss 0.190924 LR 0.000031 Time 0.020538 -2023-02-13 18:43:57,805 - Epoch: [195][ 420/ 1207] Overall Loss 0.191734 Objective Loss 0.191734 LR 0.000031 Time 0.020498 -2023-02-13 18:43:57,993 - Epoch: [195][ 430/ 1207] Overall Loss 0.191584 Objective Loss 0.191584 LR 0.000031 Time 0.020457 -2023-02-13 18:43:58,182 - Epoch: [195][ 440/ 1207] Overall Loss 0.191617 Objective Loss 0.191617 LR 0.000031 Time 0.020421 -2023-02-13 18:43:58,371 - Epoch: [195][ 450/ 1207] Overall Loss 0.191414 Objective Loss 0.191414 LR 0.000031 Time 0.020386 -2023-02-13 18:43:58,560 - Epoch: [195][ 460/ 1207] Overall Loss 0.190980 Objective Loss 0.190980 LR 0.000031 Time 0.020352 -2023-02-13 18:43:58,748 - Epoch: [195][ 470/ 1207] Overall Loss 0.190819 Objective Loss 0.190819 LR 0.000031 Time 0.020318 -2023-02-13 18:43:58,935 - Epoch: [195][ 480/ 1207] Overall Loss 0.191020 Objective Loss 0.191020 LR 0.000031 Time 0.020285 -2023-02-13 18:43:59,124 - Epoch: [195][ 490/ 1207] Overall Loss 0.191265 Objective Loss 0.191265 LR 0.000031 Time 0.020255 -2023-02-13 18:43:59,312 - Epoch: [195][ 500/ 1207] Overall Loss 0.191373 Objective Loss 0.191373 LR 0.000031 Time 0.020226 -2023-02-13 18:43:59,500 - Epoch: [195][ 510/ 1207] Overall Loss 0.191659 Objective Loss 0.191659 LR 0.000031 Time 0.020197 -2023-02-13 18:43:59,688 - Epoch: [195][ 520/ 1207] Overall Loss 0.191804 Objective Loss 0.191804 LR 0.000031 Time 0.020169 -2023-02-13 18:43:59,875 - Epoch: [195][ 530/ 1207] Overall Loss 0.191642 Objective Loss 0.191642 LR 0.000031 Time 0.020141 -2023-02-13 18:44:00,062 - Epoch: [195][ 540/ 1207] Overall Loss 0.191720 Objective Loss 0.191720 LR 0.000031 Time 0.020114 -2023-02-13 18:44:00,250 - Epoch: [195][ 550/ 1207] Overall Loss 0.192073 Objective Loss 0.192073 LR 0.000031 Time 0.020089 -2023-02-13 18:44:00,438 - Epoch: [195][ 560/ 1207] Overall Loss 0.192217 Objective Loss 0.192217 LR 0.000031 Time 0.020065 -2023-02-13 18:44:00,625 - Epoch: [195][ 570/ 1207] Overall Loss 0.192743 Objective Loss 0.192743 LR 0.000031 Time 0.020041 -2023-02-13 18:44:00,812 - Epoch: [195][ 580/ 1207] Overall Loss 0.192663 Objective Loss 0.192663 LR 0.000031 Time 0.020017 -2023-02-13 18:44:01,001 - Epoch: [195][ 590/ 1207] Overall Loss 0.192839 Objective Loss 0.192839 LR 0.000031 Time 0.019997 -2023-02-13 18:44:01,189 - Epoch: [195][ 600/ 1207] Overall Loss 0.192521 Objective Loss 0.192521 LR 0.000031 Time 0.019977 -2023-02-13 18:44:01,378 - Epoch: [195][ 610/ 1207] Overall Loss 0.192713 Objective Loss 0.192713 LR 0.000031 Time 0.019958 -2023-02-13 18:44:01,566 - Epoch: [195][ 620/ 1207] Overall Loss 0.192581 Objective Loss 0.192581 LR 0.000031 Time 0.019939 -2023-02-13 18:44:01,754 - Epoch: [195][ 630/ 1207] Overall Loss 0.192302 Objective Loss 0.192302 LR 0.000031 Time 0.019921 -2023-02-13 18:44:01,943 - Epoch: [195][ 640/ 1207] Overall Loss 0.192168 Objective Loss 0.192168 LR 0.000031 Time 0.019904 -2023-02-13 18:44:02,132 - Epoch: [195][ 650/ 1207] Overall Loss 0.192257 Objective Loss 0.192257 LR 0.000031 Time 0.019888 -2023-02-13 18:44:02,322 - Epoch: [195][ 660/ 1207] Overall Loss 0.191924 Objective Loss 0.191924 LR 0.000031 Time 0.019873 -2023-02-13 18:44:02,509 - Epoch: [195][ 670/ 1207] Overall Loss 0.191834 Objective Loss 0.191834 LR 0.000031 Time 0.019857 -2023-02-13 18:44:02,698 - Epoch: [195][ 680/ 1207] Overall Loss 0.192015 Objective Loss 0.192015 LR 0.000031 Time 0.019841 -2023-02-13 18:44:02,887 - Epoch: [195][ 690/ 1207] Overall Loss 0.192015 Objective Loss 0.192015 LR 0.000031 Time 0.019827 -2023-02-13 18:44:03,075 - Epoch: [195][ 700/ 1207] Overall Loss 0.191737 Objective Loss 0.191737 LR 0.000031 Time 0.019812 -2023-02-13 18:44:03,264 - Epoch: [195][ 710/ 1207] Overall Loss 0.191969 Objective Loss 0.191969 LR 0.000031 Time 0.019799 -2023-02-13 18:44:03,453 - Epoch: [195][ 720/ 1207] Overall Loss 0.192219 Objective Loss 0.192219 LR 0.000031 Time 0.019786 -2023-02-13 18:44:03,641 - Epoch: [195][ 730/ 1207] Overall Loss 0.192313 Objective Loss 0.192313 LR 0.000031 Time 0.019772 -2023-02-13 18:44:03,830 - Epoch: [195][ 740/ 1207] Overall Loss 0.192116 Objective Loss 0.192116 LR 0.000031 Time 0.019760 -2023-02-13 18:44:04,019 - Epoch: [195][ 750/ 1207] Overall Loss 0.191908 Objective Loss 0.191908 LR 0.000031 Time 0.019747 -2023-02-13 18:44:04,208 - Epoch: [195][ 760/ 1207] Overall Loss 0.192022 Objective Loss 0.192022 LR 0.000031 Time 0.019735 -2023-02-13 18:44:04,397 - Epoch: [195][ 770/ 1207] Overall Loss 0.192286 Objective Loss 0.192286 LR 0.000031 Time 0.019724 -2023-02-13 18:44:04,586 - Epoch: [195][ 780/ 1207] Overall Loss 0.192126 Objective Loss 0.192126 LR 0.000031 Time 0.019713 -2023-02-13 18:44:04,774 - Epoch: [195][ 790/ 1207] Overall Loss 0.192279 Objective Loss 0.192279 LR 0.000031 Time 0.019702 -2023-02-13 18:44:04,963 - Epoch: [195][ 800/ 1207] Overall Loss 0.191931 Objective Loss 0.191931 LR 0.000031 Time 0.019691 -2023-02-13 18:44:05,152 - Epoch: [195][ 810/ 1207] Overall Loss 0.191978 Objective Loss 0.191978 LR 0.000031 Time 0.019680 -2023-02-13 18:44:05,341 - Epoch: [195][ 820/ 1207] Overall Loss 0.192398 Objective Loss 0.192398 LR 0.000031 Time 0.019671 -2023-02-13 18:44:05,530 - Epoch: [195][ 830/ 1207] Overall Loss 0.192211 Objective Loss 0.192211 LR 0.000031 Time 0.019661 -2023-02-13 18:44:05,719 - Epoch: [195][ 840/ 1207] Overall Loss 0.192377 Objective Loss 0.192377 LR 0.000031 Time 0.019651 -2023-02-13 18:44:05,908 - Epoch: [195][ 850/ 1207] Overall Loss 0.192397 Objective Loss 0.192397 LR 0.000031 Time 0.019642 -2023-02-13 18:44:06,098 - Epoch: [195][ 860/ 1207] Overall Loss 0.192527 Objective Loss 0.192527 LR 0.000031 Time 0.019634 -2023-02-13 18:44:06,288 - Epoch: [195][ 870/ 1207] Overall Loss 0.192596 Objective Loss 0.192596 LR 0.000031 Time 0.019626 -2023-02-13 18:44:06,477 - Epoch: [195][ 880/ 1207] Overall Loss 0.192749 Objective Loss 0.192749 LR 0.000031 Time 0.019618 -2023-02-13 18:44:06,666 - Epoch: [195][ 890/ 1207] Overall Loss 0.192819 Objective Loss 0.192819 LR 0.000031 Time 0.019609 -2023-02-13 18:44:06,856 - Epoch: [195][ 900/ 1207] Overall Loss 0.192928 Objective Loss 0.192928 LR 0.000031 Time 0.019602 -2023-02-13 18:44:07,046 - Epoch: [195][ 910/ 1207] Overall Loss 0.192993 Objective Loss 0.192993 LR 0.000031 Time 0.019595 -2023-02-13 18:44:07,236 - Epoch: [195][ 920/ 1207] Overall Loss 0.192979 Objective Loss 0.192979 LR 0.000031 Time 0.019588 -2023-02-13 18:44:07,425 - Epoch: [195][ 930/ 1207] Overall Loss 0.193095 Objective Loss 0.193095 LR 0.000031 Time 0.019580 -2023-02-13 18:44:07,614 - Epoch: [195][ 940/ 1207] Overall Loss 0.193100 Objective Loss 0.193100 LR 0.000031 Time 0.019572 -2023-02-13 18:44:07,803 - Epoch: [195][ 950/ 1207] Overall Loss 0.192994 Objective Loss 0.192994 LR 0.000031 Time 0.019565 -2023-02-13 18:44:07,992 - Epoch: [195][ 960/ 1207] Overall Loss 0.192877 Objective Loss 0.192877 LR 0.000031 Time 0.019557 -2023-02-13 18:44:08,181 - Epoch: [195][ 970/ 1207] Overall Loss 0.192907 Objective Loss 0.192907 LR 0.000031 Time 0.019550 -2023-02-13 18:44:08,371 - Epoch: [195][ 980/ 1207] Overall Loss 0.192858 Objective Loss 0.192858 LR 0.000031 Time 0.019544 -2023-02-13 18:44:08,559 - Epoch: [195][ 990/ 1207] Overall Loss 0.192784 Objective Loss 0.192784 LR 0.000031 Time 0.019536 -2023-02-13 18:44:08,747 - Epoch: [195][ 1000/ 1207] Overall Loss 0.192761 Objective Loss 0.192761 LR 0.000031 Time 0.019529 -2023-02-13 18:44:08,937 - Epoch: [195][ 1010/ 1207] Overall Loss 0.192716 Objective Loss 0.192716 LR 0.000031 Time 0.019523 -2023-02-13 18:44:09,126 - Epoch: [195][ 1020/ 1207] Overall Loss 0.192561 Objective Loss 0.192561 LR 0.000031 Time 0.019517 -2023-02-13 18:44:09,316 - Epoch: [195][ 1030/ 1207] Overall Loss 0.192601 Objective Loss 0.192601 LR 0.000031 Time 0.019512 -2023-02-13 18:44:09,505 - Epoch: [195][ 1040/ 1207] Overall Loss 0.192726 Objective Loss 0.192726 LR 0.000031 Time 0.019505 -2023-02-13 18:44:09,695 - Epoch: [195][ 1050/ 1207] Overall Loss 0.192732 Objective Loss 0.192732 LR 0.000031 Time 0.019500 -2023-02-13 18:44:09,884 - Epoch: [195][ 1060/ 1207] Overall Loss 0.192647 Objective Loss 0.192647 LR 0.000031 Time 0.019494 -2023-02-13 18:44:10,074 - Epoch: [195][ 1070/ 1207] Overall Loss 0.192706 Objective Loss 0.192706 LR 0.000031 Time 0.019489 -2023-02-13 18:44:10,264 - Epoch: [195][ 1080/ 1207] Overall Loss 0.192829 Objective Loss 0.192829 LR 0.000031 Time 0.019484 -2023-02-13 18:44:10,453 - Epoch: [195][ 1090/ 1207] Overall Loss 0.192863 Objective Loss 0.192863 LR 0.000031 Time 0.019479 -2023-02-13 18:44:10,643 - Epoch: [195][ 1100/ 1207] Overall Loss 0.192807 Objective Loss 0.192807 LR 0.000031 Time 0.019474 -2023-02-13 18:44:10,832 - Epoch: [195][ 1110/ 1207] Overall Loss 0.192766 Objective Loss 0.192766 LR 0.000031 Time 0.019468 -2023-02-13 18:44:11,021 - Epoch: [195][ 1120/ 1207] Overall Loss 0.192640 Objective Loss 0.192640 LR 0.000031 Time 0.019463 -2023-02-13 18:44:11,210 - Epoch: [195][ 1130/ 1207] Overall Loss 0.192606 Objective Loss 0.192606 LR 0.000031 Time 0.019457 -2023-02-13 18:44:11,400 - Epoch: [195][ 1140/ 1207] Overall Loss 0.192515 Objective Loss 0.192515 LR 0.000031 Time 0.019453 -2023-02-13 18:44:11,589 - Epoch: [195][ 1150/ 1207] Overall Loss 0.192411 Objective Loss 0.192411 LR 0.000031 Time 0.019448 -2023-02-13 18:44:11,778 - Epoch: [195][ 1160/ 1207] Overall Loss 0.192165 Objective Loss 0.192165 LR 0.000031 Time 0.019443 -2023-02-13 18:44:11,968 - Epoch: [195][ 1170/ 1207] Overall Loss 0.192213 Objective Loss 0.192213 LR 0.000031 Time 0.019438 -2023-02-13 18:44:12,157 - Epoch: [195][ 1180/ 1207] Overall Loss 0.192233 Objective Loss 0.192233 LR 0.000031 Time 0.019434 -2023-02-13 18:44:12,347 - Epoch: [195][ 1190/ 1207] Overall Loss 0.192071 Objective Loss 0.192071 LR 0.000031 Time 0.019430 -2023-02-13 18:44:12,585 - Epoch: [195][ 1200/ 1207] Overall Loss 0.192196 Objective Loss 0.192196 LR 0.000031 Time 0.019466 -2023-02-13 18:44:12,701 - Epoch: [195][ 1207/ 1207] Overall Loss 0.192005 Objective Loss 0.192005 Top1 86.890244 Top5 97.256098 LR 0.000031 Time 0.019449 -2023-02-13 18:44:12,773 - --- validate (epoch=195)----------- -2023-02-13 18:44:12,774 - 34311 samples (256 per mini-batch) -2023-02-13 18:44:13,277 - Epoch: [195][ 10/ 135] Loss 0.281025 Top1 85.625000 Top5 98.125000 -2023-02-13 18:44:13,419 - Epoch: [195][ 20/ 135] Loss 0.283030 Top1 85.996094 Top5 98.144531 -2023-02-13 18:44:13,547 - Epoch: [195][ 30/ 135] Loss 0.289642 Top1 85.690104 Top5 97.955729 -2023-02-13 18:44:13,679 - Epoch: [195][ 40/ 135] Loss 0.287281 Top1 85.947266 Top5 97.949219 -2023-02-13 18:44:13,811 - Epoch: [195][ 50/ 135] Loss 0.285770 Top1 86.015625 Top5 97.921875 -2023-02-13 18:44:13,941 - Epoch: [195][ 60/ 135] Loss 0.287582 Top1 85.898438 Top5 97.903646 -2023-02-13 18:44:14,072 - Epoch: [195][ 70/ 135] Loss 0.285798 Top1 85.920759 Top5 97.935268 -2023-02-13 18:44:14,204 - Epoch: [195][ 80/ 135] Loss 0.285913 Top1 85.913086 Top5 97.915039 -2023-02-13 18:44:14,341 - Epoch: [195][ 90/ 135] Loss 0.287764 Top1 85.872396 Top5 97.899306 -2023-02-13 18:44:14,482 - Epoch: [195][ 100/ 135] Loss 0.288732 Top1 85.839844 Top5 97.890625 -2023-02-13 18:44:14,617 - Epoch: [195][ 110/ 135] Loss 0.291169 Top1 85.830966 Top5 97.887074 -2023-02-13 18:44:14,748 - Epoch: [195][ 120/ 135] Loss 0.289214 Top1 85.895182 Top5 97.897135 -2023-02-13 18:44:14,880 - Epoch: [195][ 130/ 135] Loss 0.288810 Top1 85.934495 Top5 97.911659 -2023-02-13 18:44:14,926 - Epoch: [195][ 135/ 135] Loss 0.288096 Top1 85.931625 Top5 97.913206 -2023-02-13 18:44:14,996 - ==> Top1: 85.932 Top5: 97.913 Loss: 0.288 - -2023-02-13 18:44:14,997 - ==> Confusion: -[[ 869 3 7 2 9 2 0 1 3 39 0 3 2 4 5 3 1 2 1 3 8] - [ 4 958 1 1 11 17 2 17 2 0 1 1 1 0 0 1 6 0 2 2 6] - [ 6 4 969 10 2 0 12 13 0 1 2 3 2 3 2 5 5 2 5 3 9] - [ 6 1 22 903 1 6 0 1 1 2 13 0 5 0 17 1 4 5 21 0 7] - [ 11 8 1 0 997 10 1 0 2 1 0 6 3 2 7 4 4 0 1 3 5] - [ 2 15 1 3 3 973 3 18 0 3 2 9 2 14 0 4 6 2 2 3 5] - [ 1 3 9 2 0 5 1050 5 0 2 1 1 2 1 0 3 2 1 1 5 5] - [ 1 8 13 1 1 26 1 937 0 2 0 3 5 1 0 0 1 2 12 6 4] - [ 13 3 0 1 1 0 1 2 905 40 6 3 0 10 14 2 2 0 6 0 0] - [ 79 1 2 0 9 1 0 3 32 859 0 1 0 16 3 0 2 1 1 0 2] - [ 2 0 4 6 2 3 2 5 14 1 986 1 1 5 3 0 3 1 6 0 6] - [ 2 4 1 0 1 12 0 5 1 2 0 915 25 6 0 7 2 8 1 7 6] - [ 1 0 1 8 0 4 0 0 2 0 0 25 882 2 2 7 3 12 1 1 8] - [ 5 2 3 0 4 8 1 1 10 19 8 3 1 939 5 3 3 3 0 1 5] - [ 6 3 1 18 4 4 0 1 17 9 4 0 2 1 999 1 3 5 5 0 9] - [ 3 2 7 0 3 1 4 0 0 0 0 4 6 3 0 978 13 8 0 8 6] - [ 0 5 2 1 6 1 0 0 1 1 0 0 2 1 3 9 1013 0 1 4 11] - [ 5 2 0 5 0 1 2 0 0 1 0 9 13 2 0 16 0 988 0 1 6] - [ 4 3 6 7 0 1 0 19 1 2 3 2 1 0 10 0 1 3 1016 4 3] - [ 1 3 0 0 0 2 10 9 1 0 0 14 1 1 2 6 4 2 0 1085 7] - [ 143 225 244 98 122 186 70 188 84 79 158 98 281 245 145 77 244 96 161 227 10263]] - -2023-02-13 18:44:14,998 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:44:14,998 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:44:15,004 - - -2023-02-13 18:44:15,004 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:44:15,967 - Epoch: [196][ 10/ 1207] Overall Loss 0.181811 Objective Loss 0.181811 LR 0.000031 Time 0.096257 -2023-02-13 18:44:16,175 - Epoch: [196][ 20/ 1207] Overall Loss 0.185008 Objective Loss 0.185008 LR 0.000031 Time 0.058504 -2023-02-13 18:44:16,371 - Epoch: [196][ 30/ 1207] Overall Loss 0.189459 Objective Loss 0.189459 LR 0.000031 Time 0.045523 -2023-02-13 18:44:16,570 - Epoch: [196][ 40/ 1207] Overall Loss 0.183107 Objective Loss 0.183107 LR 0.000031 Time 0.039108 -2023-02-13 18:44:16,765 - Epoch: [196][ 50/ 1207] Overall Loss 0.181266 Objective Loss 0.181266 LR 0.000031 Time 0.035171 -2023-02-13 18:44:16,964 - Epoch: [196][ 60/ 1207] Overall Loss 0.179147 Objective Loss 0.179147 LR 0.000031 Time 0.032616 -2023-02-13 18:44:17,159 - Epoch: [196][ 70/ 1207] Overall Loss 0.181935 Objective Loss 0.181935 LR 0.000031 Time 0.030740 -2023-02-13 18:44:17,358 - Epoch: [196][ 80/ 1207] Overall Loss 0.183687 Objective Loss 0.183687 LR 0.000031 Time 0.029381 -2023-02-13 18:44:17,553 - Epoch: [196][ 90/ 1207] Overall Loss 0.184886 Objective Loss 0.184886 LR 0.000031 Time 0.028281 -2023-02-13 18:44:17,751 - Epoch: [196][ 100/ 1207] Overall Loss 0.184041 Objective Loss 0.184041 LR 0.000031 Time 0.027428 -2023-02-13 18:44:17,946 - Epoch: [196][ 110/ 1207] Overall Loss 0.186047 Objective Loss 0.186047 LR 0.000031 Time 0.026703 -2023-02-13 18:44:18,144 - Epoch: [196][ 120/ 1207] Overall Loss 0.186997 Objective Loss 0.186997 LR 0.000031 Time 0.026124 -2023-02-13 18:44:18,339 - Epoch: [196][ 130/ 1207] Overall Loss 0.185843 Objective Loss 0.185843 LR 0.000031 Time 0.025611 -2023-02-13 18:44:18,535 - Epoch: [196][ 140/ 1207] Overall Loss 0.186804 Objective Loss 0.186804 LR 0.000031 Time 0.025184 -2023-02-13 18:44:18,728 - Epoch: [196][ 150/ 1207] Overall Loss 0.186266 Objective Loss 0.186266 LR 0.000031 Time 0.024785 -2023-02-13 18:44:18,925 - Epoch: [196][ 160/ 1207] Overall Loss 0.185807 Objective Loss 0.185807 LR 0.000031 Time 0.024466 -2023-02-13 18:44:19,118 - Epoch: [196][ 170/ 1207] Overall Loss 0.185912 Objective Loss 0.185912 LR 0.000031 Time 0.024161 -2023-02-13 18:44:19,315 - Epoch: [196][ 180/ 1207] Overall Loss 0.186707 Objective Loss 0.186707 LR 0.000031 Time 0.023908 -2023-02-13 18:44:19,507 - Epoch: [196][ 190/ 1207] Overall Loss 0.186409 Objective Loss 0.186409 LR 0.000031 Time 0.023661 -2023-02-13 18:44:19,704 - Epoch: [196][ 200/ 1207] Overall Loss 0.187595 Objective Loss 0.187595 LR 0.000031 Time 0.023460 -2023-02-13 18:44:19,897 - Epoch: [196][ 210/ 1207] Overall Loss 0.187926 Objective Loss 0.187926 LR 0.000031 Time 0.023261 -2023-02-13 18:44:20,096 - Epoch: [196][ 220/ 1207] Overall Loss 0.188494 Objective Loss 0.188494 LR 0.000031 Time 0.023106 -2023-02-13 18:44:20,291 - Epoch: [196][ 230/ 1207] Overall Loss 0.187884 Objective Loss 0.187884 LR 0.000031 Time 0.022945 -2023-02-13 18:44:20,490 - Epoch: [196][ 240/ 1207] Overall Loss 0.188174 Objective Loss 0.188174 LR 0.000031 Time 0.022820 -2023-02-13 18:44:20,685 - Epoch: [196][ 250/ 1207] Overall Loss 0.188503 Objective Loss 0.188503 LR 0.000031 Time 0.022687 -2023-02-13 18:44:20,884 - Epoch: [196][ 260/ 1207] Overall Loss 0.188997 Objective Loss 0.188997 LR 0.000031 Time 0.022577 -2023-02-13 18:44:21,082 - Epoch: [196][ 270/ 1207] Overall Loss 0.188779 Objective Loss 0.188779 LR 0.000031 Time 0.022470 -2023-02-13 18:44:21,281 - Epoch: [196][ 280/ 1207] Overall Loss 0.188836 Objective Loss 0.188836 LR 0.000031 Time 0.022378 -2023-02-13 18:44:21,478 - Epoch: [196][ 290/ 1207] Overall Loss 0.188905 Objective Loss 0.188905 LR 0.000031 Time 0.022284 -2023-02-13 18:44:21,677 - Epoch: [196][ 300/ 1207] Overall Loss 0.188925 Objective Loss 0.188925 LR 0.000031 Time 0.022204 -2023-02-13 18:44:21,873 - Epoch: [196][ 310/ 1207] Overall Loss 0.189051 Objective Loss 0.189051 LR 0.000031 Time 0.022119 -2023-02-13 18:44:22,072 - Epoch: [196][ 320/ 1207] Overall Loss 0.189013 Objective Loss 0.189013 LR 0.000031 Time 0.022049 -2023-02-13 18:44:22,269 - Epoch: [196][ 330/ 1207] Overall Loss 0.188769 Objective Loss 0.188769 LR 0.000031 Time 0.021977 -2023-02-13 18:44:22,469 - Epoch: [196][ 340/ 1207] Overall Loss 0.188751 Objective Loss 0.188751 LR 0.000031 Time 0.021917 -2023-02-13 18:44:22,666 - Epoch: [196][ 350/ 1207] Overall Loss 0.189093 Objective Loss 0.189093 LR 0.000031 Time 0.021852 -2023-02-13 18:44:22,865 - Epoch: [196][ 360/ 1207] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000031 Time 0.021797 -2023-02-13 18:44:23,061 - Epoch: [196][ 370/ 1207] Overall Loss 0.189061 Objective Loss 0.189061 LR 0.000031 Time 0.021737 -2023-02-13 18:44:23,260 - Epoch: [196][ 380/ 1207] Overall Loss 0.189062 Objective Loss 0.189062 LR 0.000031 Time 0.021688 -2023-02-13 18:44:23,457 - Epoch: [196][ 390/ 1207] Overall Loss 0.189556 Objective Loss 0.189556 LR 0.000031 Time 0.021636 -2023-02-13 18:44:23,656 - Epoch: [196][ 400/ 1207] Overall Loss 0.189480 Objective Loss 0.189480 LR 0.000031 Time 0.021590 -2023-02-13 18:44:23,852 - Epoch: [196][ 410/ 1207] Overall Loss 0.189661 Objective Loss 0.189661 LR 0.000031 Time 0.021542 -2023-02-13 18:44:24,051 - Epoch: [196][ 420/ 1207] Overall Loss 0.189419 Objective Loss 0.189419 LR 0.000031 Time 0.021503 -2023-02-13 18:44:24,248 - Epoch: [196][ 430/ 1207] Overall Loss 0.189138 Objective Loss 0.189138 LR 0.000031 Time 0.021459 -2023-02-13 18:44:24,448 - Epoch: [196][ 440/ 1207] Overall Loss 0.189117 Objective Loss 0.189117 LR 0.000031 Time 0.021425 -2023-02-13 18:44:24,645 - Epoch: [196][ 450/ 1207] Overall Loss 0.189753 Objective Loss 0.189753 LR 0.000031 Time 0.021385 -2023-02-13 18:44:24,844 - Epoch: [196][ 460/ 1207] Overall Loss 0.190079 Objective Loss 0.190079 LR 0.000031 Time 0.021353 -2023-02-13 18:44:25,038 - Epoch: [196][ 470/ 1207] Overall Loss 0.190020 Objective Loss 0.190020 LR 0.000031 Time 0.021310 -2023-02-13 18:44:25,229 - Epoch: [196][ 480/ 1207] Overall Loss 0.189948 Objective Loss 0.189948 LR 0.000031 Time 0.021263 -2023-02-13 18:44:25,421 - Epoch: [196][ 490/ 1207] Overall Loss 0.189867 Objective Loss 0.189867 LR 0.000031 Time 0.021220 -2023-02-13 18:44:25,612 - Epoch: [196][ 500/ 1207] Overall Loss 0.190211 Objective Loss 0.190211 LR 0.000031 Time 0.021177 -2023-02-13 18:44:25,803 - Epoch: [196][ 510/ 1207] Overall Loss 0.189994 Objective Loss 0.189994 LR 0.000031 Time 0.021136 -2023-02-13 18:44:25,996 - Epoch: [196][ 520/ 1207] Overall Loss 0.190026 Objective Loss 0.190026 LR 0.000031 Time 0.021100 -2023-02-13 18:44:26,187 - Epoch: [196][ 530/ 1207] Overall Loss 0.190213 Objective Loss 0.190213 LR 0.000031 Time 0.021062 -2023-02-13 18:44:26,379 - Epoch: [196][ 540/ 1207] Overall Loss 0.189998 Objective Loss 0.189998 LR 0.000031 Time 0.021027 -2023-02-13 18:44:26,571 - Epoch: [196][ 550/ 1207] Overall Loss 0.190310 Objective Loss 0.190310 LR 0.000031 Time 0.020992 -2023-02-13 18:44:26,762 - Epoch: [196][ 560/ 1207] Overall Loss 0.189902 Objective Loss 0.189902 LR 0.000031 Time 0.020958 -2023-02-13 18:44:26,953 - Epoch: [196][ 570/ 1207] Overall Loss 0.189904 Objective Loss 0.189904 LR 0.000031 Time 0.020925 -2023-02-13 18:44:27,144 - Epoch: [196][ 580/ 1207] Overall Loss 0.189760 Objective Loss 0.189760 LR 0.000031 Time 0.020893 -2023-02-13 18:44:27,336 - Epoch: [196][ 590/ 1207] Overall Loss 0.189939 Objective Loss 0.189939 LR 0.000031 Time 0.020863 -2023-02-13 18:44:27,526 - Epoch: [196][ 600/ 1207] Overall Loss 0.190030 Objective Loss 0.190030 LR 0.000031 Time 0.020833 -2023-02-13 18:44:27,717 - Epoch: [196][ 610/ 1207] Overall Loss 0.189826 Objective Loss 0.189826 LR 0.000031 Time 0.020802 -2023-02-13 18:44:27,908 - Epoch: [196][ 620/ 1207] Overall Loss 0.189825 Objective Loss 0.189825 LR 0.000031 Time 0.020775 -2023-02-13 18:44:28,099 - Epoch: [196][ 630/ 1207] Overall Loss 0.190044 Objective Loss 0.190044 LR 0.000031 Time 0.020748 -2023-02-13 18:44:28,291 - Epoch: [196][ 640/ 1207] Overall Loss 0.190376 Objective Loss 0.190376 LR 0.000031 Time 0.020722 -2023-02-13 18:44:28,482 - Epoch: [196][ 650/ 1207] Overall Loss 0.190504 Objective Loss 0.190504 LR 0.000031 Time 0.020698 -2023-02-13 18:44:28,674 - Epoch: [196][ 660/ 1207] Overall Loss 0.190404 Objective Loss 0.190404 LR 0.000031 Time 0.020674 -2023-02-13 18:44:28,864 - Epoch: [196][ 670/ 1207] Overall Loss 0.190686 Objective Loss 0.190686 LR 0.000031 Time 0.020649 -2023-02-13 18:44:29,057 - Epoch: [196][ 680/ 1207] Overall Loss 0.190629 Objective Loss 0.190629 LR 0.000031 Time 0.020628 -2023-02-13 18:44:29,248 - Epoch: [196][ 690/ 1207] Overall Loss 0.190683 Objective Loss 0.190683 LR 0.000031 Time 0.020605 -2023-02-13 18:44:29,441 - Epoch: [196][ 700/ 1207] Overall Loss 0.190664 Objective Loss 0.190664 LR 0.000031 Time 0.020586 -2023-02-13 18:44:29,631 - Epoch: [196][ 710/ 1207] Overall Loss 0.191029 Objective Loss 0.191029 LR 0.000031 Time 0.020564 -2023-02-13 18:44:29,822 - Epoch: [196][ 720/ 1207] Overall Loss 0.191227 Objective Loss 0.191227 LR 0.000031 Time 0.020543 -2023-02-13 18:44:30,014 - Epoch: [196][ 730/ 1207] Overall Loss 0.190801 Objective Loss 0.190801 LR 0.000031 Time 0.020524 -2023-02-13 18:44:30,205 - Epoch: [196][ 740/ 1207] Overall Loss 0.190934 Objective Loss 0.190934 LR 0.000031 Time 0.020505 -2023-02-13 18:44:30,397 - Epoch: [196][ 750/ 1207] Overall Loss 0.191095 Objective Loss 0.191095 LR 0.000031 Time 0.020486 -2023-02-13 18:44:30,588 - Epoch: [196][ 760/ 1207] Overall Loss 0.191145 Objective Loss 0.191145 LR 0.000031 Time 0.020467 -2023-02-13 18:44:30,778 - Epoch: [196][ 770/ 1207] Overall Loss 0.191184 Objective Loss 0.191184 LR 0.000031 Time 0.020448 -2023-02-13 18:44:30,970 - Epoch: [196][ 780/ 1207] Overall Loss 0.190839 Objective Loss 0.190839 LR 0.000031 Time 0.020431 -2023-02-13 18:44:31,163 - Epoch: [196][ 790/ 1207] Overall Loss 0.191242 Objective Loss 0.191242 LR 0.000031 Time 0.020416 -2023-02-13 18:44:31,355 - Epoch: [196][ 800/ 1207] Overall Loss 0.191224 Objective Loss 0.191224 LR 0.000031 Time 0.020401 -2023-02-13 18:44:31,546 - Epoch: [196][ 810/ 1207] Overall Loss 0.191096 Objective Loss 0.191096 LR 0.000031 Time 0.020385 -2023-02-13 18:44:31,736 - Epoch: [196][ 820/ 1207] Overall Loss 0.191225 Objective Loss 0.191225 LR 0.000031 Time 0.020367 -2023-02-13 18:44:31,926 - Epoch: [196][ 830/ 1207] Overall Loss 0.191361 Objective Loss 0.191361 LR 0.000031 Time 0.020350 -2023-02-13 18:44:32,118 - Epoch: [196][ 840/ 1207] Overall Loss 0.191417 Objective Loss 0.191417 LR 0.000031 Time 0.020336 -2023-02-13 18:44:32,307 - Epoch: [196][ 850/ 1207] Overall Loss 0.191392 Objective Loss 0.191392 LR 0.000031 Time 0.020319 -2023-02-13 18:44:32,498 - Epoch: [196][ 860/ 1207] Overall Loss 0.191170 Objective Loss 0.191170 LR 0.000031 Time 0.020304 -2023-02-13 18:44:32,687 - Epoch: [196][ 870/ 1207] Overall Loss 0.191156 Objective Loss 0.191156 LR 0.000031 Time 0.020288 -2023-02-13 18:44:32,878 - Epoch: [196][ 880/ 1207] Overall Loss 0.191075 Objective Loss 0.191075 LR 0.000031 Time 0.020273 -2023-02-13 18:44:33,067 - Epoch: [196][ 890/ 1207] Overall Loss 0.190955 Objective Loss 0.190955 LR 0.000031 Time 0.020258 -2023-02-13 18:44:33,257 - Epoch: [196][ 900/ 1207] Overall Loss 0.190814 Objective Loss 0.190814 LR 0.000031 Time 0.020243 -2023-02-13 18:44:33,447 - Epoch: [196][ 910/ 1207] Overall Loss 0.190987 Objective Loss 0.190987 LR 0.000031 Time 0.020230 -2023-02-13 18:44:33,637 - Epoch: [196][ 920/ 1207] Overall Loss 0.191095 Objective Loss 0.191095 LR 0.000031 Time 0.020216 -2023-02-13 18:44:33,827 - Epoch: [196][ 930/ 1207] Overall Loss 0.191101 Objective Loss 0.191101 LR 0.000031 Time 0.020202 -2023-02-13 18:44:34,017 - Epoch: [196][ 940/ 1207] Overall Loss 0.191066 Objective Loss 0.191066 LR 0.000031 Time 0.020188 -2023-02-13 18:44:34,207 - Epoch: [196][ 950/ 1207] Overall Loss 0.191099 Objective Loss 0.191099 LR 0.000031 Time 0.020176 -2023-02-13 18:44:34,398 - Epoch: [196][ 960/ 1207] Overall Loss 0.191396 Objective Loss 0.191396 LR 0.000031 Time 0.020164 -2023-02-13 18:44:34,587 - Epoch: [196][ 970/ 1207] Overall Loss 0.191357 Objective Loss 0.191357 LR 0.000031 Time 0.020151 -2023-02-13 18:44:34,777 - Epoch: [196][ 980/ 1207] Overall Loss 0.191574 Objective Loss 0.191574 LR 0.000031 Time 0.020139 -2023-02-13 18:44:34,967 - Epoch: [196][ 990/ 1207] Overall Loss 0.191552 Objective Loss 0.191552 LR 0.000031 Time 0.020127 -2023-02-13 18:44:35,157 - Epoch: [196][ 1000/ 1207] Overall Loss 0.191870 Objective Loss 0.191870 LR 0.000031 Time 0.020115 -2023-02-13 18:44:35,347 - Epoch: [196][ 1010/ 1207] Overall Loss 0.191869 Objective Loss 0.191869 LR 0.000031 Time 0.020104 -2023-02-13 18:44:35,538 - Epoch: [196][ 1020/ 1207] Overall Loss 0.191769 Objective Loss 0.191769 LR 0.000031 Time 0.020094 -2023-02-13 18:44:35,728 - Epoch: [196][ 1030/ 1207] Overall Loss 0.191668 Objective Loss 0.191668 LR 0.000031 Time 0.020083 -2023-02-13 18:44:35,918 - Epoch: [196][ 1040/ 1207] Overall Loss 0.191585 Objective Loss 0.191585 LR 0.000031 Time 0.020072 -2023-02-13 18:44:36,109 - Epoch: [196][ 1050/ 1207] Overall Loss 0.191756 Objective Loss 0.191756 LR 0.000031 Time 0.020062 -2023-02-13 18:44:36,299 - Epoch: [196][ 1060/ 1207] Overall Loss 0.191896 Objective Loss 0.191896 LR 0.000031 Time 0.020052 -2023-02-13 18:44:36,489 - Epoch: [196][ 1070/ 1207] Overall Loss 0.191791 Objective Loss 0.191791 LR 0.000031 Time 0.020042 -2023-02-13 18:44:36,679 - Epoch: [196][ 1080/ 1207] Overall Loss 0.191712 Objective Loss 0.191712 LR 0.000031 Time 0.020032 -2023-02-13 18:44:36,869 - Epoch: [196][ 1090/ 1207] Overall Loss 0.191652 Objective Loss 0.191652 LR 0.000031 Time 0.020022 -2023-02-13 18:44:37,059 - Epoch: [196][ 1100/ 1207] Overall Loss 0.191588 Objective Loss 0.191588 LR 0.000031 Time 0.020013 -2023-02-13 18:44:37,250 - Epoch: [196][ 1110/ 1207] Overall Loss 0.191653 Objective Loss 0.191653 LR 0.000031 Time 0.020004 -2023-02-13 18:44:37,440 - Epoch: [196][ 1120/ 1207] Overall Loss 0.191694 Objective Loss 0.191694 LR 0.000031 Time 0.019995 -2023-02-13 18:44:37,630 - Epoch: [196][ 1130/ 1207] Overall Loss 0.191528 Objective Loss 0.191528 LR 0.000031 Time 0.019986 -2023-02-13 18:44:37,820 - Epoch: [196][ 1140/ 1207] Overall Loss 0.191705 Objective Loss 0.191705 LR 0.000031 Time 0.019977 -2023-02-13 18:44:38,010 - Epoch: [196][ 1150/ 1207] Overall Loss 0.191611 Objective Loss 0.191611 LR 0.000031 Time 0.019968 -2023-02-13 18:44:38,200 - Epoch: [196][ 1160/ 1207] Overall Loss 0.191814 Objective Loss 0.191814 LR 0.000031 Time 0.019959 -2023-02-13 18:44:38,391 - Epoch: [196][ 1170/ 1207] Overall Loss 0.191972 Objective Loss 0.191972 LR 0.000031 Time 0.019952 -2023-02-13 18:44:38,581 - Epoch: [196][ 1180/ 1207] Overall Loss 0.191841 Objective Loss 0.191841 LR 0.000031 Time 0.019943 -2023-02-13 18:44:38,771 - Epoch: [196][ 1190/ 1207] Overall Loss 0.191773 Objective Loss 0.191773 LR 0.000031 Time 0.019935 -2023-02-13 18:44:39,014 - Epoch: [196][ 1200/ 1207] Overall Loss 0.191757 Objective Loss 0.191757 LR 0.000031 Time 0.019971 -2023-02-13 18:44:39,131 - Epoch: [196][ 1207/ 1207] Overall Loss 0.191834 Objective Loss 0.191834 Top1 91.463415 Top5 99.695122 LR 0.000031 Time 0.019952 -2023-02-13 18:44:39,204 - --- validate (epoch=196)----------- -2023-02-13 18:44:39,204 - 34311 samples (256 per mini-batch) -2023-02-13 18:44:39,609 - Epoch: [196][ 10/ 135] Loss 0.273105 Top1 86.640625 Top5 97.851562 -2023-02-13 18:44:39,739 - Epoch: [196][ 20/ 135] Loss 0.272984 Top1 86.386719 Top5 97.988281 -2023-02-13 18:44:39,876 - Epoch: [196][ 30/ 135] Loss 0.277558 Top1 86.263021 Top5 97.942708 -2023-02-13 18:44:40,006 - Epoch: [196][ 40/ 135] Loss 0.284927 Top1 85.830078 Top5 97.832031 -2023-02-13 18:44:40,142 - Epoch: [196][ 50/ 135] Loss 0.279461 Top1 85.890625 Top5 97.937500 -2023-02-13 18:44:40,272 - Epoch: [196][ 60/ 135] Loss 0.278469 Top1 85.924479 Top5 97.903646 -2023-02-13 18:44:40,403 - Epoch: [196][ 70/ 135] Loss 0.280770 Top1 85.764509 Top5 97.885045 -2023-02-13 18:44:40,530 - Epoch: [196][ 80/ 135] Loss 0.283406 Top1 85.776367 Top5 97.841797 -2023-02-13 18:44:40,655 - Epoch: [196][ 90/ 135] Loss 0.284455 Top1 85.672743 Top5 97.838542 -2023-02-13 18:44:40,781 - Epoch: [196][ 100/ 135] Loss 0.285630 Top1 85.679688 Top5 97.871094 -2023-02-13 18:44:40,908 - Epoch: [196][ 110/ 135] Loss 0.288248 Top1 85.603693 Top5 97.883523 -2023-02-13 18:44:41,038 - Epoch: [196][ 120/ 135] Loss 0.286174 Top1 85.628255 Top5 97.916667 -2023-02-13 18:44:41,165 - Epoch: [196][ 130/ 135] Loss 0.287485 Top1 85.600962 Top5 97.914663 -2023-02-13 18:44:41,210 - Epoch: [196][ 135/ 135] Loss 0.284950 Top1 85.637259 Top5 97.907377 -2023-02-13 18:44:41,278 - ==> Top1: 85.637 Top5: 97.907 Loss: 0.285 - -2023-02-13 18:44:41,279 - ==> Confusion: -[[ 850 6 8 0 12 1 0 1 4 56 0 3 0 4 6 3 2 1 0 2 8] - [ 4 946 1 1 11 27 1 13 2 1 2 1 1 0 0 2 6 0 3 4 7] - [ 5 6 954 15 2 1 17 13 2 1 2 5 1 5 4 7 5 0 5 2 6] - [ 6 0 18 916 0 6 2 1 3 2 12 0 6 0 16 1 2 4 14 0 7] - [ 11 6 0 0 997 13 1 2 1 1 0 6 3 3 6 6 2 2 0 2 4] - [ 2 10 1 2 3 990 3 13 0 4 0 8 1 13 0 4 6 2 1 2 5] - [ 3 1 7 1 0 8 1052 5 0 1 1 1 1 2 0 3 1 2 1 3 6] - [ 2 7 10 0 1 34 4 928 0 3 2 5 3 0 0 0 2 1 11 8 3] - [ 11 3 1 1 1 0 1 1 899 45 5 2 0 15 14 2 2 0 5 0 1] - [ 66 0 2 0 8 0 0 1 28 873 0 0 0 19 3 4 1 1 1 0 5] - [ 2 1 0 7 2 1 2 3 16 1 988 1 1 11 3 0 1 1 6 0 4] - [ 3 2 2 0 2 13 0 7 3 1 0 914 25 7 0 8 2 8 1 5 2] - [ 1 0 0 7 2 5 0 0 2 2 0 27 881 1 2 6 1 12 3 1 6] - [ 4 2 0 2 6 6 1 1 6 17 8 4 1 947 5 3 2 2 0 1 6] - [ 3 2 1 16 4 5 0 1 14 9 3 0 2 3 1010 2 1 4 6 0 6] - [ 3 2 5 1 4 1 4 1 0 0 1 5 6 3 0 980 8 8 0 7 7] - [ 1 5 1 0 8 3 0 0 2 0 0 0 1 3 3 9 1003 3 1 3 15] - [ 4 2 0 5 1 2 2 0 0 2 0 10 8 3 0 15 0 991 0 1 5] - [ 1 5 6 5 1 2 0 23 2 2 6 1 2 0 11 0 0 4 1010 3 2] - [ 0 2 0 0 0 7 7 8 1 0 0 16 2 2 1 8 4 2 0 1079 9] - [ 127 214 215 113 120 224 74 155 83 89 179 101 281 300 150 91 225 105 170 243 10175]] - -2023-02-13 18:44:41,281 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:44:41,281 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:44:41,286 - - -2023-02-13 18:44:41,286 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:44:42,281 - Epoch: [197][ 10/ 1207] Overall Loss 0.181617 Objective Loss 0.181617 LR 0.000031 Time 0.099373 -2023-02-13 18:44:42,479 - Epoch: [197][ 20/ 1207] Overall Loss 0.194646 Objective Loss 0.194646 LR 0.000031 Time 0.059600 -2023-02-13 18:44:42,667 - Epoch: [197][ 30/ 1207] Overall Loss 0.194359 Objective Loss 0.194359 LR 0.000031 Time 0.045980 -2023-02-13 18:44:42,855 - Epoch: [197][ 40/ 1207] Overall Loss 0.195278 Objective Loss 0.195278 LR 0.000031 Time 0.039168 -2023-02-13 18:44:43,043 - Epoch: [197][ 50/ 1207] Overall Loss 0.193239 Objective Loss 0.193239 LR 0.000031 Time 0.035090 -2023-02-13 18:44:43,232 - Epoch: [197][ 60/ 1207] Overall Loss 0.194173 Objective Loss 0.194173 LR 0.000031 Time 0.032382 -2023-02-13 18:44:43,420 - Epoch: [197][ 70/ 1207] Overall Loss 0.192006 Objective Loss 0.192006 LR 0.000031 Time 0.030437 -2023-02-13 18:44:43,608 - Epoch: [197][ 80/ 1207] Overall Loss 0.190440 Objective Loss 0.190440 LR 0.000031 Time 0.028984 -2023-02-13 18:44:43,796 - Epoch: [197][ 90/ 1207] Overall Loss 0.190957 Objective Loss 0.190957 LR 0.000031 Time 0.027847 -2023-02-13 18:44:43,984 - Epoch: [197][ 100/ 1207] Overall Loss 0.192760 Objective Loss 0.192760 LR 0.000031 Time 0.026935 -2023-02-13 18:44:44,172 - Epoch: [197][ 110/ 1207] Overall Loss 0.193336 Objective Loss 0.193336 LR 0.000031 Time 0.026191 -2023-02-13 18:44:44,360 - Epoch: [197][ 120/ 1207] Overall Loss 0.196028 Objective Loss 0.196028 LR 0.000031 Time 0.025570 -2023-02-13 18:44:44,549 - Epoch: [197][ 130/ 1207] Overall Loss 0.194290 Objective Loss 0.194290 LR 0.000031 Time 0.025055 -2023-02-13 18:44:44,737 - Epoch: [197][ 140/ 1207] Overall Loss 0.192648 Objective Loss 0.192648 LR 0.000031 Time 0.024605 -2023-02-13 18:44:44,924 - Epoch: [197][ 150/ 1207] Overall Loss 0.192606 Objective Loss 0.192606 LR 0.000031 Time 0.024215 -2023-02-13 18:44:45,112 - Epoch: [197][ 160/ 1207] Overall Loss 0.191996 Objective Loss 0.191996 LR 0.000031 Time 0.023873 -2023-02-13 18:44:45,301 - Epoch: [197][ 170/ 1207] Overall Loss 0.191720 Objective Loss 0.191720 LR 0.000031 Time 0.023576 -2023-02-13 18:44:45,490 - Epoch: [197][ 180/ 1207] Overall Loss 0.192776 Objective Loss 0.192776 LR 0.000031 Time 0.023316 -2023-02-13 18:44:45,678 - Epoch: [197][ 190/ 1207] Overall Loss 0.193066 Objective Loss 0.193066 LR 0.000031 Time 0.023077 -2023-02-13 18:44:45,867 - Epoch: [197][ 200/ 1207] Overall Loss 0.193251 Objective Loss 0.193251 LR 0.000031 Time 0.022864 -2023-02-13 18:44:46,057 - Epoch: [197][ 210/ 1207] Overall Loss 0.192974 Objective Loss 0.192974 LR 0.000031 Time 0.022677 -2023-02-13 18:44:46,246 - Epoch: [197][ 220/ 1207] Overall Loss 0.193471 Objective Loss 0.193471 LR 0.000031 Time 0.022506 -2023-02-13 18:44:46,435 - Epoch: [197][ 230/ 1207] Overall Loss 0.192418 Objective Loss 0.192418 LR 0.000031 Time 0.022348 -2023-02-13 18:44:46,624 - Epoch: [197][ 240/ 1207] Overall Loss 0.192883 Objective Loss 0.192883 LR 0.000031 Time 0.022203 -2023-02-13 18:44:46,812 - Epoch: [197][ 250/ 1207] Overall Loss 0.192910 Objective Loss 0.192910 LR 0.000031 Time 0.022066 -2023-02-13 18:44:47,002 - Epoch: [197][ 260/ 1207] Overall Loss 0.193167 Objective Loss 0.193167 LR 0.000031 Time 0.021944 -2023-02-13 18:44:47,190 - Epoch: [197][ 270/ 1207] Overall Loss 0.192989 Objective Loss 0.192989 LR 0.000031 Time 0.021829 -2023-02-13 18:44:47,379 - Epoch: [197][ 280/ 1207] Overall Loss 0.194019 Objective Loss 0.194019 LR 0.000031 Time 0.021721 -2023-02-13 18:44:47,568 - Epoch: [197][ 290/ 1207] Overall Loss 0.193278 Objective Loss 0.193278 LR 0.000031 Time 0.021623 -2023-02-13 18:44:47,756 - Epoch: [197][ 300/ 1207] Overall Loss 0.193246 Objective Loss 0.193246 LR 0.000031 Time 0.021529 -2023-02-13 18:44:47,945 - Epoch: [197][ 310/ 1207] Overall Loss 0.193392 Objective Loss 0.193392 LR 0.000031 Time 0.021443 -2023-02-13 18:44:48,134 - Epoch: [197][ 320/ 1207] Overall Loss 0.193247 Objective Loss 0.193247 LR 0.000031 Time 0.021362 -2023-02-13 18:44:48,323 - Epoch: [197][ 330/ 1207] Overall Loss 0.192703 Objective Loss 0.192703 LR 0.000031 Time 0.021286 -2023-02-13 18:44:48,513 - Epoch: [197][ 340/ 1207] Overall Loss 0.192418 Objective Loss 0.192418 LR 0.000031 Time 0.021216 -2023-02-13 18:44:48,701 - Epoch: [197][ 350/ 1207] Overall Loss 0.192248 Objective Loss 0.192248 LR 0.000031 Time 0.021148 -2023-02-13 18:44:48,890 - Epoch: [197][ 360/ 1207] Overall Loss 0.192324 Objective Loss 0.192324 LR 0.000031 Time 0.021084 -2023-02-13 18:44:49,079 - Epoch: [197][ 370/ 1207] Overall Loss 0.191850 Objective Loss 0.191850 LR 0.000031 Time 0.021023 -2023-02-13 18:44:49,268 - Epoch: [197][ 380/ 1207] Overall Loss 0.192141 Objective Loss 0.192141 LR 0.000031 Time 0.020967 -2023-02-13 18:44:49,457 - Epoch: [197][ 390/ 1207] Overall Loss 0.192004 Objective Loss 0.192004 LR 0.000031 Time 0.020913 -2023-02-13 18:44:49,646 - Epoch: [197][ 400/ 1207] Overall Loss 0.191344 Objective Loss 0.191344 LR 0.000031 Time 0.020862 -2023-02-13 18:44:49,834 - Epoch: [197][ 410/ 1207] Overall Loss 0.191370 Objective Loss 0.191370 LR 0.000031 Time 0.020812 -2023-02-13 18:44:50,023 - Epoch: [197][ 420/ 1207] Overall Loss 0.191600 Objective Loss 0.191600 LR 0.000031 Time 0.020764 -2023-02-13 18:44:50,212 - Epoch: [197][ 430/ 1207] Overall Loss 0.191748 Objective Loss 0.191748 LR 0.000031 Time 0.020721 -2023-02-13 18:44:50,401 - Epoch: [197][ 440/ 1207] Overall Loss 0.191929 Objective Loss 0.191929 LR 0.000031 Time 0.020678 -2023-02-13 18:44:50,590 - Epoch: [197][ 450/ 1207] Overall Loss 0.192291 Objective Loss 0.192291 LR 0.000031 Time 0.020638 -2023-02-13 18:44:50,779 - Epoch: [197][ 460/ 1207] Overall Loss 0.191938 Objective Loss 0.191938 LR 0.000031 Time 0.020599 -2023-02-13 18:44:50,969 - Epoch: [197][ 470/ 1207] Overall Loss 0.192528 Objective Loss 0.192528 LR 0.000031 Time 0.020564 -2023-02-13 18:44:51,158 - Epoch: [197][ 480/ 1207] Overall Loss 0.192062 Objective Loss 0.192062 LR 0.000031 Time 0.020529 -2023-02-13 18:44:51,347 - Epoch: [197][ 490/ 1207] Overall Loss 0.192374 Objective Loss 0.192374 LR 0.000031 Time 0.020495 -2023-02-13 18:44:51,537 - Epoch: [197][ 500/ 1207] Overall Loss 0.192357 Objective Loss 0.192357 LR 0.000031 Time 0.020463 -2023-02-13 18:44:51,725 - Epoch: [197][ 510/ 1207] Overall Loss 0.192353 Objective Loss 0.192353 LR 0.000031 Time 0.020430 -2023-02-13 18:44:51,913 - Epoch: [197][ 520/ 1207] Overall Loss 0.192233 Objective Loss 0.192233 LR 0.000031 Time 0.020399 -2023-02-13 18:44:52,102 - Epoch: [197][ 530/ 1207] Overall Loss 0.192046 Objective Loss 0.192046 LR 0.000031 Time 0.020370 -2023-02-13 18:44:52,291 - Epoch: [197][ 540/ 1207] Overall Loss 0.192100 Objective Loss 0.192100 LR 0.000031 Time 0.020342 -2023-02-13 18:44:52,481 - Epoch: [197][ 550/ 1207] Overall Loss 0.191813 Objective Loss 0.191813 LR 0.000031 Time 0.020316 -2023-02-13 18:44:52,670 - Epoch: [197][ 560/ 1207] Overall Loss 0.191404 Objective Loss 0.191404 LR 0.000031 Time 0.020291 -2023-02-13 18:44:52,858 - Epoch: [197][ 570/ 1207] Overall Loss 0.191377 Objective Loss 0.191377 LR 0.000031 Time 0.020264 -2023-02-13 18:44:53,046 - Epoch: [197][ 580/ 1207] Overall Loss 0.191532 Objective Loss 0.191532 LR 0.000031 Time 0.020239 -2023-02-13 18:44:53,235 - Epoch: [197][ 590/ 1207] Overall Loss 0.191577 Objective Loss 0.191577 LR 0.000031 Time 0.020215 -2023-02-13 18:44:53,425 - Epoch: [197][ 600/ 1207] Overall Loss 0.191670 Objective Loss 0.191670 LR 0.000031 Time 0.020193 -2023-02-13 18:44:53,613 - Epoch: [197][ 610/ 1207] Overall Loss 0.191667 Objective Loss 0.191667 LR 0.000031 Time 0.020171 -2023-02-13 18:44:53,801 - Epoch: [197][ 620/ 1207] Overall Loss 0.191543 Objective Loss 0.191543 LR 0.000031 Time 0.020148 -2023-02-13 18:44:53,990 - Epoch: [197][ 630/ 1207] Overall Loss 0.191423 Objective Loss 0.191423 LR 0.000031 Time 0.020127 -2023-02-13 18:44:54,179 - Epoch: [197][ 640/ 1207] Overall Loss 0.191178 Objective Loss 0.191178 LR 0.000031 Time 0.020107 -2023-02-13 18:44:54,367 - Epoch: [197][ 650/ 1207] Overall Loss 0.191368 Objective Loss 0.191368 LR 0.000031 Time 0.020087 -2023-02-13 18:44:54,557 - Epoch: [197][ 660/ 1207] Overall Loss 0.191639 Objective Loss 0.191639 LR 0.000031 Time 0.020070 -2023-02-13 18:44:54,746 - Epoch: [197][ 670/ 1207] Overall Loss 0.191667 Objective Loss 0.191667 LR 0.000031 Time 0.020051 -2023-02-13 18:44:54,934 - Epoch: [197][ 680/ 1207] Overall Loss 0.191613 Objective Loss 0.191613 LR 0.000031 Time 0.020033 -2023-02-13 18:44:55,123 - Epoch: [197][ 690/ 1207] Overall Loss 0.191372 Objective Loss 0.191372 LR 0.000031 Time 0.020016 -2023-02-13 18:44:55,313 - Epoch: [197][ 700/ 1207] Overall Loss 0.191804 Objective Loss 0.191804 LR 0.000031 Time 0.020000 -2023-02-13 18:44:55,502 - Epoch: [197][ 710/ 1207] Overall Loss 0.191735 Objective Loss 0.191735 LR 0.000031 Time 0.019985 -2023-02-13 18:44:55,691 - Epoch: [197][ 720/ 1207] Overall Loss 0.191883 Objective Loss 0.191883 LR 0.000031 Time 0.019969 -2023-02-13 18:44:55,879 - Epoch: [197][ 730/ 1207] Overall Loss 0.191992 Objective Loss 0.191992 LR 0.000031 Time 0.019953 -2023-02-13 18:44:56,069 - Epoch: [197][ 740/ 1207] Overall Loss 0.191969 Objective Loss 0.191969 LR 0.000031 Time 0.019939 -2023-02-13 18:44:56,259 - Epoch: [197][ 750/ 1207] Overall Loss 0.191918 Objective Loss 0.191918 LR 0.000031 Time 0.019926 -2023-02-13 18:44:56,447 - Epoch: [197][ 760/ 1207] Overall Loss 0.191904 Objective Loss 0.191904 LR 0.000031 Time 0.019911 -2023-02-13 18:44:56,637 - Epoch: [197][ 770/ 1207] Overall Loss 0.192097 Objective Loss 0.192097 LR 0.000031 Time 0.019898 -2023-02-13 18:44:56,825 - Epoch: [197][ 780/ 1207] Overall Loss 0.192409 Objective Loss 0.192409 LR 0.000031 Time 0.019884 -2023-02-13 18:44:57,015 - Epoch: [197][ 790/ 1207] Overall Loss 0.192433 Objective Loss 0.192433 LR 0.000031 Time 0.019872 -2023-02-13 18:44:57,203 - Epoch: [197][ 800/ 1207] Overall Loss 0.192386 Objective Loss 0.192386 LR 0.000031 Time 0.019859 -2023-02-13 18:44:57,392 - Epoch: [197][ 810/ 1207] Overall Loss 0.192164 Objective Loss 0.192164 LR 0.000031 Time 0.019847 -2023-02-13 18:44:57,582 - Epoch: [197][ 820/ 1207] Overall Loss 0.191947 Objective Loss 0.191947 LR 0.000031 Time 0.019836 -2023-02-13 18:44:57,771 - Epoch: [197][ 830/ 1207] Overall Loss 0.191739 Objective Loss 0.191739 LR 0.000031 Time 0.019823 -2023-02-13 18:44:57,959 - Epoch: [197][ 840/ 1207] Overall Loss 0.191568 Objective Loss 0.191568 LR 0.000031 Time 0.019812 -2023-02-13 18:44:58,149 - Epoch: [197][ 850/ 1207] Overall Loss 0.191633 Objective Loss 0.191633 LR 0.000031 Time 0.019801 -2023-02-13 18:44:58,339 - Epoch: [197][ 860/ 1207] Overall Loss 0.191734 Objective Loss 0.191734 LR 0.000031 Time 0.019791 -2023-02-13 18:44:58,528 - Epoch: [197][ 870/ 1207] Overall Loss 0.191743 Objective Loss 0.191743 LR 0.000031 Time 0.019780 -2023-02-13 18:44:58,717 - Epoch: [197][ 880/ 1207] Overall Loss 0.191807 Objective Loss 0.191807 LR 0.000031 Time 0.019770 -2023-02-13 18:44:58,905 - Epoch: [197][ 890/ 1207] Overall Loss 0.191680 Objective Loss 0.191680 LR 0.000031 Time 0.019759 -2023-02-13 18:44:59,094 - Epoch: [197][ 900/ 1207] Overall Loss 0.191640 Objective Loss 0.191640 LR 0.000031 Time 0.019749 -2023-02-13 18:44:59,283 - Epoch: [197][ 910/ 1207] Overall Loss 0.191846 Objective Loss 0.191846 LR 0.000031 Time 0.019740 -2023-02-13 18:44:59,473 - Epoch: [197][ 920/ 1207] Overall Loss 0.191831 Objective Loss 0.191831 LR 0.000031 Time 0.019730 -2023-02-13 18:44:59,671 - Epoch: [197][ 930/ 1207] Overall Loss 0.191774 Objective Loss 0.191774 LR 0.000031 Time 0.019732 -2023-02-13 18:44:59,875 - Epoch: [197][ 940/ 1207] Overall Loss 0.191896 Objective Loss 0.191896 LR 0.000031 Time 0.019737 -2023-02-13 18:45:00,074 - Epoch: [197][ 950/ 1207] Overall Loss 0.191983 Objective Loss 0.191983 LR 0.000031 Time 0.019739 -2023-02-13 18:45:00,279 - Epoch: [197][ 960/ 1207] Overall Loss 0.192059 Objective Loss 0.192059 LR 0.000031 Time 0.019746 -2023-02-13 18:45:00,478 - Epoch: [197][ 970/ 1207] Overall Loss 0.192180 Objective Loss 0.192180 LR 0.000031 Time 0.019748 -2023-02-13 18:45:00,681 - Epoch: [197][ 980/ 1207] Overall Loss 0.192006 Objective Loss 0.192006 LR 0.000031 Time 0.019754 -2023-02-13 18:45:00,880 - Epoch: [197][ 990/ 1207] Overall Loss 0.192180 Objective Loss 0.192180 LR 0.000031 Time 0.019754 -2023-02-13 18:45:01,085 - Epoch: [197][ 1000/ 1207] Overall Loss 0.192271 Objective Loss 0.192271 LR 0.000031 Time 0.019761 -2023-02-13 18:45:01,284 - Epoch: [197][ 1010/ 1207] Overall Loss 0.192094 Objective Loss 0.192094 LR 0.000031 Time 0.019763 -2023-02-13 18:45:01,490 - Epoch: [197][ 1020/ 1207] Overall Loss 0.192052 Objective Loss 0.192052 LR 0.000031 Time 0.019770 -2023-02-13 18:45:01,689 - Epoch: [197][ 1030/ 1207] Overall Loss 0.192032 Objective Loss 0.192032 LR 0.000031 Time 0.019771 -2023-02-13 18:45:01,892 - Epoch: [197][ 1040/ 1207] Overall Loss 0.192104 Objective Loss 0.192104 LR 0.000031 Time 0.019776 -2023-02-13 18:45:02,092 - Epoch: [197][ 1050/ 1207] Overall Loss 0.192066 Objective Loss 0.192066 LR 0.000031 Time 0.019778 -2023-02-13 18:45:02,296 - Epoch: [197][ 1060/ 1207] Overall Loss 0.191977 Objective Loss 0.191977 LR 0.000031 Time 0.019783 -2023-02-13 18:45:02,496 - Epoch: [197][ 1070/ 1207] Overall Loss 0.191886 Objective Loss 0.191886 LR 0.000031 Time 0.019785 -2023-02-13 18:45:02,700 - Epoch: [197][ 1080/ 1207] Overall Loss 0.191872 Objective Loss 0.191872 LR 0.000031 Time 0.019790 -2023-02-13 18:45:02,899 - Epoch: [197][ 1090/ 1207] Overall Loss 0.191903 Objective Loss 0.191903 LR 0.000031 Time 0.019791 -2023-02-13 18:45:03,102 - Epoch: [197][ 1100/ 1207] Overall Loss 0.191950 Objective Loss 0.191950 LR 0.000031 Time 0.019795 -2023-02-13 18:45:03,302 - Epoch: [197][ 1110/ 1207] Overall Loss 0.192069 Objective Loss 0.192069 LR 0.000031 Time 0.019796 -2023-02-13 18:45:03,505 - Epoch: [197][ 1120/ 1207] Overall Loss 0.192177 Objective Loss 0.192177 LR 0.000031 Time 0.019801 -2023-02-13 18:45:03,704 - Epoch: [197][ 1130/ 1207] Overall Loss 0.192074 Objective Loss 0.192074 LR 0.000031 Time 0.019802 -2023-02-13 18:45:03,909 - Epoch: [197][ 1140/ 1207] Overall Loss 0.191924 Objective Loss 0.191924 LR 0.000031 Time 0.019807 -2023-02-13 18:45:04,108 - Epoch: [197][ 1150/ 1207] Overall Loss 0.191813 Objective Loss 0.191813 LR 0.000031 Time 0.019808 -2023-02-13 18:45:04,313 - Epoch: [197][ 1160/ 1207] Overall Loss 0.191901 Objective Loss 0.191901 LR 0.000031 Time 0.019813 -2023-02-13 18:45:04,512 - Epoch: [197][ 1170/ 1207] Overall Loss 0.191820 Objective Loss 0.191820 LR 0.000031 Time 0.019814 -2023-02-13 18:45:04,716 - Epoch: [197][ 1180/ 1207] Overall Loss 0.191884 Objective Loss 0.191884 LR 0.000031 Time 0.019818 -2023-02-13 18:45:04,915 - Epoch: [197][ 1190/ 1207] Overall Loss 0.191639 Objective Loss 0.191639 LR 0.000031 Time 0.019819 -2023-02-13 18:45:05,165 - Epoch: [197][ 1200/ 1207] Overall Loss 0.191500 Objective Loss 0.191500 LR 0.000031 Time 0.019862 -2023-02-13 18:45:05,281 - Epoch: [197][ 1207/ 1207] Overall Loss 0.191327 Objective Loss 0.191327 Top1 90.548780 Top5 100.000000 LR 0.000031 Time 0.019842 -2023-02-13 18:45:05,353 - --- validate (epoch=197)----------- -2023-02-13 18:45:05,354 - 34311 samples (256 per mini-batch) -2023-02-13 18:45:05,753 - Epoch: [197][ 10/ 135] Loss 0.281003 Top1 86.093750 Top5 97.929688 -2023-02-13 18:45:05,883 - Epoch: [197][ 20/ 135] Loss 0.276198 Top1 85.898438 Top5 98.105469 -2023-02-13 18:45:06,015 - Epoch: [197][ 30/ 135] Loss 0.281291 Top1 85.963542 Top5 97.994792 -2023-02-13 18:45:06,144 - Epoch: [197][ 40/ 135] Loss 0.268216 Top1 86.269531 Top5 97.890625 -2023-02-13 18:45:06,272 - Epoch: [197][ 50/ 135] Loss 0.275169 Top1 85.976562 Top5 97.820312 -2023-02-13 18:45:06,401 - Epoch: [197][ 60/ 135] Loss 0.276911 Top1 86.113281 Top5 97.825521 -2023-02-13 18:45:06,530 - Epoch: [197][ 70/ 135] Loss 0.277708 Top1 86.026786 Top5 97.812500 -2023-02-13 18:45:06,660 - Epoch: [197][ 80/ 135] Loss 0.282554 Top1 85.771484 Top5 97.778320 -2023-02-13 18:45:06,789 - Epoch: [197][ 90/ 135] Loss 0.285414 Top1 85.720486 Top5 97.782118 -2023-02-13 18:45:06,916 - Epoch: [197][ 100/ 135] Loss 0.284899 Top1 85.710938 Top5 97.808594 -2023-02-13 18:45:07,045 - Epoch: [197][ 110/ 135] Loss 0.282532 Top1 85.823864 Top5 97.830256 -2023-02-13 18:45:07,173 - Epoch: [197][ 120/ 135] Loss 0.283656 Top1 85.791016 Top5 97.809245 -2023-02-13 18:45:07,301 - Epoch: [197][ 130/ 135] Loss 0.287179 Top1 85.721154 Top5 97.776442 -2023-02-13 18:45:07,348 - Epoch: [197][ 135/ 135] Loss 0.285238 Top1 85.750925 Top5 97.805369 -2023-02-13 18:45:07,416 - ==> Top1: 85.751 Top5: 97.805 Loss: 0.285 - -2023-02-13 18:45:07,417 - ==> Confusion: -[[ 868 4 6 1 8 2 0 0 4 44 1 3 0 3 6 3 1 2 0 4 7] - [ 4 951 1 1 11 25 0 11 2 1 2 1 2 1 0 1 4 0 2 3 10] - [ 8 4 971 9 2 2 10 15 0 1 1 2 1 3 4 6 4 1 5 3 6] - [ 7 0 20 919 0 5 0 1 1 2 14 0 7 0 16 0 3 4 14 0 3] - [ 13 7 0 0 995 11 1 0 1 1 0 6 3 4 9 6 3 1 1 2 2] - [ 2 10 2 4 4 985 3 14 0 4 3 9 2 12 0 3 5 1 2 1 4] - [ 2 2 11 1 0 5 1047 7 0 1 2 1 2 2 1 2 2 1 3 3 4] - [ 2 9 11 0 1 27 1 941 1 1 0 6 1 1 0 0 1 1 12 6 2] - [ 12 2 2 1 1 0 1 1 916 36 5 2 0 11 14 2 1 0 1 0 1] - [ 79 1 2 0 11 1 0 1 31 853 1 0 0 16 3 2 2 2 1 1 5] - [ 1 0 5 4 2 2 2 4 14 1 994 0 1 7 5 0 2 1 3 0 3] - [ 3 3 3 0 1 11 0 5 1 3 0 924 21 6 0 6 2 10 1 3 2] - [ 1 0 1 8 2 3 0 1 2 1 0 28 878 2 2 5 3 14 2 1 5] - [ 5 2 2 1 3 9 0 0 7 17 9 6 1 941 6 4 2 1 1 1 6] - [ 0 2 1 16 3 5 0 1 16 8 4 0 3 1 1014 1 1 4 5 0 7] - [ 5 2 5 0 6 1 3 0 0 0 0 4 8 3 0 976 11 9 0 7 6] - [ 1 4 2 0 8 3 0 0 2 0 0 1 1 2 2 9 1010 2 0 4 10] - [ 4 2 0 3 0 1 2 0 0 2 1 13 7 1 0 11 0 998 0 1 5] - [ 3 4 5 6 1 1 0 25 5 1 5 1 3 0 15 0 1 3 1004 2 1] - [ 0 2 2 0 1 8 7 7 1 0 0 15 2 2 1 5 5 4 0 1081 5] - [ 160 215 240 110 133 220 58 190 99 78 161 105 286 274 153 80 236 104 153 223 10156]] - -2023-02-13 18:45:07,419 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:45:07,419 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:45:07,425 - - -2023-02-13 18:45:07,425 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:45:08,333 - Epoch: [198][ 10/ 1207] Overall Loss 0.181030 Objective Loss 0.181030 LR 0.000031 Time 0.090678 -2023-02-13 18:45:08,532 - Epoch: [198][ 20/ 1207] Overall Loss 0.193843 Objective Loss 0.193843 LR 0.000031 Time 0.055285 -2023-02-13 18:45:08,720 - Epoch: [198][ 30/ 1207] Overall Loss 0.193125 Objective Loss 0.193125 LR 0.000031 Time 0.043100 -2023-02-13 18:45:08,907 - Epoch: [198][ 40/ 1207] Overall Loss 0.195519 Objective Loss 0.195519 LR 0.000031 Time 0.037005 -2023-02-13 18:45:09,095 - Epoch: [198][ 50/ 1207] Overall Loss 0.192862 Objective Loss 0.192862 LR 0.000031 Time 0.033348 -2023-02-13 18:45:09,283 - Epoch: [198][ 60/ 1207] Overall Loss 0.188741 Objective Loss 0.188741 LR 0.000031 Time 0.030919 -2023-02-13 18:45:09,471 - Epoch: [198][ 70/ 1207] Overall Loss 0.192418 Objective Loss 0.192418 LR 0.000031 Time 0.029178 -2023-02-13 18:45:09,659 - Epoch: [198][ 80/ 1207] Overall Loss 0.193425 Objective Loss 0.193425 LR 0.000031 Time 0.027877 -2023-02-13 18:45:09,853 - Epoch: [198][ 90/ 1207] Overall Loss 0.193445 Objective Loss 0.193445 LR 0.000031 Time 0.026937 -2023-02-13 18:45:10,047 - Epoch: [198][ 100/ 1207] Overall Loss 0.192618 Objective Loss 0.192618 LR 0.000031 Time 0.026178 -2023-02-13 18:45:10,244 - Epoch: [198][ 110/ 1207] Overall Loss 0.190713 Objective Loss 0.190713 LR 0.000031 Time 0.025581 -2023-02-13 18:45:10,437 - Epoch: [198][ 120/ 1207] Overall Loss 0.190505 Objective Loss 0.190505 LR 0.000031 Time 0.025061 -2023-02-13 18:45:10,634 - Epoch: [198][ 130/ 1207] Overall Loss 0.189669 Objective Loss 0.189669 LR 0.000031 Time 0.024645 -2023-02-13 18:45:10,828 - Epoch: [198][ 140/ 1207] Overall Loss 0.189708 Objective Loss 0.189708 LR 0.000031 Time 0.024270 -2023-02-13 18:45:11,026 - Epoch: [198][ 150/ 1207] Overall Loss 0.188822 Objective Loss 0.188822 LR 0.000031 Time 0.023966 -2023-02-13 18:45:11,220 - Epoch: [198][ 160/ 1207] Overall Loss 0.189812 Objective Loss 0.189812 LR 0.000031 Time 0.023678 -2023-02-13 18:45:11,416 - Epoch: [198][ 170/ 1207] Overall Loss 0.189697 Objective Loss 0.189697 LR 0.000031 Time 0.023439 -2023-02-13 18:45:11,611 - Epoch: [198][ 180/ 1207] Overall Loss 0.188782 Objective Loss 0.188782 LR 0.000031 Time 0.023213 -2023-02-13 18:45:11,807 - Epoch: [198][ 190/ 1207] Overall Loss 0.188884 Objective Loss 0.188884 LR 0.000031 Time 0.023024 -2023-02-13 18:45:12,001 - Epoch: [198][ 200/ 1207] Overall Loss 0.188389 Objective Loss 0.188389 LR 0.000031 Time 0.022842 -2023-02-13 18:45:12,198 - Epoch: [198][ 210/ 1207] Overall Loss 0.188303 Objective Loss 0.188303 LR 0.000031 Time 0.022689 -2023-02-13 18:45:12,392 - Epoch: [198][ 220/ 1207] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000031 Time 0.022537 -2023-02-13 18:45:12,589 - Epoch: [198][ 230/ 1207] Overall Loss 0.188687 Objective Loss 0.188687 LR 0.000031 Time 0.022414 -2023-02-13 18:45:12,783 - Epoch: [198][ 240/ 1207] Overall Loss 0.188979 Objective Loss 0.188979 LR 0.000031 Time 0.022286 -2023-02-13 18:45:12,980 - Epoch: [198][ 250/ 1207] Overall Loss 0.187909 Objective Loss 0.187909 LR 0.000031 Time 0.022180 -2023-02-13 18:45:13,174 - Epoch: [198][ 260/ 1207] Overall Loss 0.187837 Objective Loss 0.187837 LR 0.000031 Time 0.022071 -2023-02-13 18:45:13,370 - Epoch: [198][ 270/ 1207] Overall Loss 0.188642 Objective Loss 0.188642 LR 0.000031 Time 0.021980 -2023-02-13 18:45:13,564 - Epoch: [198][ 280/ 1207] Overall Loss 0.188706 Objective Loss 0.188706 LR 0.000031 Time 0.021886 -2023-02-13 18:45:13,761 - Epoch: [198][ 290/ 1207] Overall Loss 0.188700 Objective Loss 0.188700 LR 0.000031 Time 0.021808 -2023-02-13 18:45:13,954 - Epoch: [198][ 300/ 1207] Overall Loss 0.188424 Objective Loss 0.188424 LR 0.000031 Time 0.021725 -2023-02-13 18:45:14,151 - Epoch: [198][ 310/ 1207] Overall Loss 0.188618 Objective Loss 0.188618 LR 0.000031 Time 0.021656 -2023-02-13 18:45:14,344 - Epoch: [198][ 320/ 1207] Overall Loss 0.188547 Objective Loss 0.188547 LR 0.000031 Time 0.021584 -2023-02-13 18:45:14,541 - Epoch: [198][ 330/ 1207] Overall Loss 0.187924 Objective Loss 0.187924 LR 0.000031 Time 0.021523 -2023-02-13 18:45:14,734 - Epoch: [198][ 340/ 1207] Overall Loss 0.188272 Objective Loss 0.188272 LR 0.000031 Time 0.021459 -2023-02-13 18:45:14,931 - Epoch: [198][ 350/ 1207] Overall Loss 0.188379 Objective Loss 0.188379 LR 0.000031 Time 0.021405 -2023-02-13 18:45:15,124 - Epoch: [198][ 360/ 1207] Overall Loss 0.188921 Objective Loss 0.188921 LR 0.000031 Time 0.021347 -2023-02-13 18:45:15,320 - Epoch: [198][ 370/ 1207] Overall Loss 0.189354 Objective Loss 0.189354 LR 0.000031 Time 0.021300 -2023-02-13 18:45:15,514 - Epoch: [198][ 380/ 1207] Overall Loss 0.189484 Objective Loss 0.189484 LR 0.000031 Time 0.021247 -2023-02-13 18:45:15,710 - Epoch: [198][ 390/ 1207] Overall Loss 0.189399 Objective Loss 0.189399 LR 0.000031 Time 0.021204 -2023-02-13 18:45:15,903 - Epoch: [198][ 400/ 1207] Overall Loss 0.190282 Objective Loss 0.190282 LR 0.000031 Time 0.021156 -2023-02-13 18:45:16,101 - Epoch: [198][ 410/ 1207] Overall Loss 0.190385 Objective Loss 0.190385 LR 0.000031 Time 0.021121 -2023-02-13 18:45:16,295 - Epoch: [198][ 420/ 1207] Overall Loss 0.190625 Objective Loss 0.190625 LR 0.000031 Time 0.021079 -2023-02-13 18:45:16,491 - Epoch: [198][ 430/ 1207] Overall Loss 0.190697 Objective Loss 0.190697 LR 0.000031 Time 0.021045 -2023-02-13 18:45:16,685 - Epoch: [198][ 440/ 1207] Overall Loss 0.190623 Objective Loss 0.190623 LR 0.000031 Time 0.021007 -2023-02-13 18:45:16,882 - Epoch: [198][ 450/ 1207] Overall Loss 0.190662 Objective Loss 0.190662 LR 0.000031 Time 0.020976 -2023-02-13 18:45:17,076 - Epoch: [198][ 460/ 1207] Overall Loss 0.190838 Objective Loss 0.190838 LR 0.000031 Time 0.020941 -2023-02-13 18:45:17,272 - Epoch: [198][ 470/ 1207] Overall Loss 0.190773 Objective Loss 0.190773 LR 0.000031 Time 0.020913 -2023-02-13 18:45:17,466 - Epoch: [198][ 480/ 1207] Overall Loss 0.190587 Objective Loss 0.190587 LR 0.000031 Time 0.020880 -2023-02-13 18:45:17,663 - Epoch: [198][ 490/ 1207] Overall Loss 0.190448 Objective Loss 0.190448 LR 0.000031 Time 0.020854 -2023-02-13 18:45:17,856 - Epoch: [198][ 500/ 1207] Overall Loss 0.190589 Objective Loss 0.190589 LR 0.000031 Time 0.020823 -2023-02-13 18:45:18,052 - Epoch: [198][ 510/ 1207] Overall Loss 0.190573 Objective Loss 0.190573 LR 0.000031 Time 0.020799 -2023-02-13 18:45:18,246 - Epoch: [198][ 520/ 1207] Overall Loss 0.190586 Objective Loss 0.190586 LR 0.000031 Time 0.020772 -2023-02-13 18:45:18,443 - Epoch: [198][ 530/ 1207] Overall Loss 0.190438 Objective Loss 0.190438 LR 0.000031 Time 0.020750 -2023-02-13 18:45:18,637 - Epoch: [198][ 540/ 1207] Overall Loss 0.190370 Objective Loss 0.190370 LR 0.000031 Time 0.020724 -2023-02-13 18:45:18,834 - Epoch: [198][ 550/ 1207] Overall Loss 0.190723 Objective Loss 0.190723 LR 0.000031 Time 0.020705 -2023-02-13 18:45:19,028 - Epoch: [198][ 560/ 1207] Overall Loss 0.190954 Objective Loss 0.190954 LR 0.000031 Time 0.020680 -2023-02-13 18:45:19,224 - Epoch: [198][ 570/ 1207] Overall Loss 0.191230 Objective Loss 0.191230 LR 0.000031 Time 0.020662 -2023-02-13 18:45:19,419 - Epoch: [198][ 580/ 1207] Overall Loss 0.191486 Objective Loss 0.191486 LR 0.000031 Time 0.020640 -2023-02-13 18:45:19,616 - Epoch: [198][ 590/ 1207] Overall Loss 0.191510 Objective Loss 0.191510 LR 0.000031 Time 0.020624 -2023-02-13 18:45:19,810 - Epoch: [198][ 600/ 1207] Overall Loss 0.191519 Objective Loss 0.191519 LR 0.000031 Time 0.020603 -2023-02-13 18:45:20,006 - Epoch: [198][ 610/ 1207] Overall Loss 0.191482 Objective Loss 0.191482 LR 0.000031 Time 0.020586 -2023-02-13 18:45:20,201 - Epoch: [198][ 620/ 1207] Overall Loss 0.191575 Objective Loss 0.191575 LR 0.000031 Time 0.020567 -2023-02-13 18:45:20,398 - Epoch: [198][ 630/ 1207] Overall Loss 0.191948 Objective Loss 0.191948 LR 0.000031 Time 0.020553 -2023-02-13 18:45:20,592 - Epoch: [198][ 640/ 1207] Overall Loss 0.192322 Objective Loss 0.192322 LR 0.000031 Time 0.020535 -2023-02-13 18:45:20,789 - Epoch: [198][ 650/ 1207] Overall Loss 0.192405 Objective Loss 0.192405 LR 0.000031 Time 0.020521 -2023-02-13 18:45:20,984 - Epoch: [198][ 660/ 1207] Overall Loss 0.192338 Objective Loss 0.192338 LR 0.000031 Time 0.020505 -2023-02-13 18:45:21,180 - Epoch: [198][ 670/ 1207] Overall Loss 0.192513 Objective Loss 0.192513 LR 0.000031 Time 0.020491 -2023-02-13 18:45:21,374 - Epoch: [198][ 680/ 1207] Overall Loss 0.192582 Objective Loss 0.192582 LR 0.000031 Time 0.020474 -2023-02-13 18:45:21,571 - Epoch: [198][ 690/ 1207] Overall Loss 0.192388 Objective Loss 0.192388 LR 0.000031 Time 0.020462 -2023-02-13 18:45:21,765 - Epoch: [198][ 700/ 1207] Overall Loss 0.192146 Objective Loss 0.192146 LR 0.000031 Time 0.020447 -2023-02-13 18:45:21,962 - Epoch: [198][ 710/ 1207] Overall Loss 0.192090 Objective Loss 0.192090 LR 0.000031 Time 0.020436 -2023-02-13 18:45:22,156 - Epoch: [198][ 720/ 1207] Overall Loss 0.192224 Objective Loss 0.192224 LR 0.000031 Time 0.020420 -2023-02-13 18:45:22,353 - Epoch: [198][ 730/ 1207] Overall Loss 0.192484 Objective Loss 0.192484 LR 0.000031 Time 0.020410 -2023-02-13 18:45:22,547 - Epoch: [198][ 740/ 1207] Overall Loss 0.192334 Objective Loss 0.192334 LR 0.000031 Time 0.020396 -2023-02-13 18:45:22,744 - Epoch: [198][ 750/ 1207] Overall Loss 0.192277 Objective Loss 0.192277 LR 0.000031 Time 0.020387 -2023-02-13 18:45:22,937 - Epoch: [198][ 760/ 1207] Overall Loss 0.192122 Objective Loss 0.192122 LR 0.000031 Time 0.020372 -2023-02-13 18:45:23,134 - Epoch: [198][ 770/ 1207] Overall Loss 0.191904 Objective Loss 0.191904 LR 0.000031 Time 0.020363 -2023-02-13 18:45:23,328 - Epoch: [198][ 780/ 1207] Overall Loss 0.192052 Objective Loss 0.192052 LR 0.000031 Time 0.020350 -2023-02-13 18:45:23,525 - Epoch: [198][ 790/ 1207] Overall Loss 0.191865 Objective Loss 0.191865 LR 0.000031 Time 0.020341 -2023-02-13 18:45:23,719 - Epoch: [198][ 800/ 1207] Overall Loss 0.191643 Objective Loss 0.191643 LR 0.000031 Time 0.020329 -2023-02-13 18:45:23,916 - Epoch: [198][ 810/ 1207] Overall Loss 0.191761 Objective Loss 0.191761 LR 0.000031 Time 0.020320 -2023-02-13 18:45:24,110 - Epoch: [198][ 820/ 1207] Overall Loss 0.191541 Objective Loss 0.191541 LR 0.000031 Time 0.020309 -2023-02-13 18:45:24,308 - Epoch: [198][ 830/ 1207] Overall Loss 0.191788 Objective Loss 0.191788 LR 0.000031 Time 0.020302 -2023-02-13 18:45:24,502 - Epoch: [198][ 840/ 1207] Overall Loss 0.191792 Objective Loss 0.191792 LR 0.000031 Time 0.020291 -2023-02-13 18:45:24,698 - Epoch: [198][ 850/ 1207] Overall Loss 0.191634 Objective Loss 0.191634 LR 0.000031 Time 0.020283 -2023-02-13 18:45:24,892 - Epoch: [198][ 860/ 1207] Overall Loss 0.191596 Objective Loss 0.191596 LR 0.000031 Time 0.020272 -2023-02-13 18:45:25,089 - Epoch: [198][ 870/ 1207] Overall Loss 0.191553 Objective Loss 0.191553 LR 0.000031 Time 0.020265 -2023-02-13 18:45:25,284 - Epoch: [198][ 880/ 1207] Overall Loss 0.191405 Objective Loss 0.191405 LR 0.000031 Time 0.020256 -2023-02-13 18:45:25,481 - Epoch: [198][ 890/ 1207] Overall Loss 0.191295 Objective Loss 0.191295 LR 0.000031 Time 0.020249 -2023-02-13 18:45:25,676 - Epoch: [198][ 900/ 1207] Overall Loss 0.191485 Objective Loss 0.191485 LR 0.000031 Time 0.020240 -2023-02-13 18:45:25,872 - Epoch: [198][ 910/ 1207] Overall Loss 0.191584 Objective Loss 0.191584 LR 0.000031 Time 0.020233 -2023-02-13 18:45:26,067 - Epoch: [198][ 920/ 1207] Overall Loss 0.191441 Objective Loss 0.191441 LR 0.000031 Time 0.020224 -2023-02-13 18:45:26,264 - Epoch: [198][ 930/ 1207] Overall Loss 0.191526 Objective Loss 0.191526 LR 0.000031 Time 0.020219 -2023-02-13 18:45:26,459 - Epoch: [198][ 940/ 1207] Overall Loss 0.191293 Objective Loss 0.191293 LR 0.000031 Time 0.020210 -2023-02-13 18:45:26,656 - Epoch: [198][ 950/ 1207] Overall Loss 0.191452 Objective Loss 0.191452 LR 0.000031 Time 0.020205 -2023-02-13 18:45:26,850 - Epoch: [198][ 960/ 1207] Overall Loss 0.191555 Objective Loss 0.191555 LR 0.000031 Time 0.020196 -2023-02-13 18:45:27,047 - Epoch: [198][ 970/ 1207] Overall Loss 0.191308 Objective Loss 0.191308 LR 0.000031 Time 0.020191 -2023-02-13 18:45:27,242 - Epoch: [198][ 980/ 1207] Overall Loss 0.191388 Objective Loss 0.191388 LR 0.000031 Time 0.020182 -2023-02-13 18:45:27,438 - Epoch: [198][ 990/ 1207] Overall Loss 0.191545 Objective Loss 0.191545 LR 0.000031 Time 0.020177 -2023-02-13 18:45:27,632 - Epoch: [198][ 1000/ 1207] Overall Loss 0.191505 Objective Loss 0.191505 LR 0.000031 Time 0.020168 -2023-02-13 18:45:27,829 - Epoch: [198][ 1010/ 1207] Overall Loss 0.191559 Objective Loss 0.191559 LR 0.000031 Time 0.020163 -2023-02-13 18:45:28,022 - Epoch: [198][ 1020/ 1207] Overall Loss 0.191596 Objective Loss 0.191596 LR 0.000031 Time 0.020155 -2023-02-13 18:45:28,219 - Epoch: [198][ 1030/ 1207] Overall Loss 0.191503 Objective Loss 0.191503 LR 0.000031 Time 0.020150 -2023-02-13 18:45:28,413 - Epoch: [198][ 1040/ 1207] Overall Loss 0.191548 Objective Loss 0.191548 LR 0.000031 Time 0.020142 -2023-02-13 18:45:28,610 - Epoch: [198][ 1050/ 1207] Overall Loss 0.191736 Objective Loss 0.191736 LR 0.000031 Time 0.020137 -2023-02-13 18:45:28,804 - Epoch: [198][ 1060/ 1207] Overall Loss 0.191618 Objective Loss 0.191618 LR 0.000031 Time 0.020130 -2023-02-13 18:45:29,001 - Epoch: [198][ 1070/ 1207] Overall Loss 0.191514 Objective Loss 0.191514 LR 0.000031 Time 0.020126 -2023-02-13 18:45:29,195 - Epoch: [198][ 1080/ 1207] Overall Loss 0.191460 Objective Loss 0.191460 LR 0.000031 Time 0.020119 -2023-02-13 18:45:29,393 - Epoch: [198][ 1090/ 1207] Overall Loss 0.191353 Objective Loss 0.191353 LR 0.000031 Time 0.020116 -2023-02-13 18:45:29,590 - Epoch: [198][ 1100/ 1207] Overall Loss 0.191292 Objective Loss 0.191292 LR 0.000031 Time 0.020111 -2023-02-13 18:45:29,788 - Epoch: [198][ 1110/ 1207] Overall Loss 0.191283 Objective Loss 0.191283 LR 0.000031 Time 0.020108 -2023-02-13 18:45:29,984 - Epoch: [198][ 1120/ 1207] Overall Loss 0.191468 Objective Loss 0.191468 LR 0.000031 Time 0.020103 -2023-02-13 18:45:30,182 - Epoch: [198][ 1130/ 1207] Overall Loss 0.191400 Objective Loss 0.191400 LR 0.000031 Time 0.020100 -2023-02-13 18:45:30,378 - Epoch: [198][ 1140/ 1207] Overall Loss 0.191407 Objective Loss 0.191407 LR 0.000031 Time 0.020095 -2023-02-13 18:45:30,577 - Epoch: [198][ 1150/ 1207] Overall Loss 0.191441 Objective Loss 0.191441 LR 0.000031 Time 0.020093 -2023-02-13 18:45:30,773 - Epoch: [198][ 1160/ 1207] Overall Loss 0.191560 Objective Loss 0.191560 LR 0.000031 Time 0.020089 -2023-02-13 18:45:30,972 - Epoch: [198][ 1170/ 1207] Overall Loss 0.191381 Objective Loss 0.191381 LR 0.000031 Time 0.020087 -2023-02-13 18:45:31,168 - Epoch: [198][ 1180/ 1207] Overall Loss 0.191320 Objective Loss 0.191320 LR 0.000031 Time 0.020083 -2023-02-13 18:45:31,367 - Epoch: [198][ 1190/ 1207] Overall Loss 0.191122 Objective Loss 0.191122 LR 0.000031 Time 0.020081 -2023-02-13 18:45:31,615 - Epoch: [198][ 1200/ 1207] Overall Loss 0.191039 Objective Loss 0.191039 LR 0.000031 Time 0.020119 -2023-02-13 18:45:31,730 - Epoch: [198][ 1207/ 1207] Overall Loss 0.190965 Objective Loss 0.190965 Top1 91.158537 Top5 99.390244 LR 0.000031 Time 0.020098 -2023-02-13 18:45:31,803 - --- validate (epoch=198)----------- -2023-02-13 18:45:31,803 - 34311 samples (256 per mini-batch) -2023-02-13 18:45:32,209 - Epoch: [198][ 10/ 135] Loss 0.283100 Top1 86.367188 Top5 97.812500 -2023-02-13 18:45:32,341 - Epoch: [198][ 20/ 135] Loss 0.299720 Top1 85.449219 Top5 97.695312 -2023-02-13 18:45:32,465 - Epoch: [198][ 30/ 135] Loss 0.282970 Top1 85.755208 Top5 97.916667 -2023-02-13 18:45:32,591 - Epoch: [198][ 40/ 135] Loss 0.286305 Top1 85.664062 Top5 97.919922 -2023-02-13 18:45:32,715 - Epoch: [198][ 50/ 135] Loss 0.290680 Top1 85.476562 Top5 97.820312 -2023-02-13 18:45:32,838 - Epoch: [198][ 60/ 135] Loss 0.283280 Top1 85.598958 Top5 97.825521 -2023-02-13 18:45:32,962 - Epoch: [198][ 70/ 135] Loss 0.284203 Top1 85.552455 Top5 97.767857 -2023-02-13 18:45:33,087 - Epoch: [198][ 80/ 135] Loss 0.284628 Top1 85.610352 Top5 97.753906 -2023-02-13 18:45:33,213 - Epoch: [198][ 90/ 135] Loss 0.285308 Top1 85.659722 Top5 97.790799 -2023-02-13 18:45:33,338 - Epoch: [198][ 100/ 135] Loss 0.284520 Top1 85.656250 Top5 97.769531 -2023-02-13 18:45:33,462 - Epoch: [198][ 110/ 135] Loss 0.286410 Top1 85.575284 Top5 97.784091 -2023-02-13 18:45:33,588 - Epoch: [198][ 120/ 135] Loss 0.288232 Top1 85.524089 Top5 97.822266 -2023-02-13 18:45:33,718 - Epoch: [198][ 130/ 135] Loss 0.286145 Top1 85.558894 Top5 97.836538 -2023-02-13 18:45:33,764 - Epoch: [198][ 135/ 135] Loss 0.283757 Top1 85.561482 Top5 97.837428 -2023-02-13 18:45:33,843 - ==> Top1: 85.561 Top5: 97.837 Loss: 0.284 - -2023-02-13 18:45:33,844 - ==> Confusion: -[[ 872 5 7 1 9 3 0 1 5 36 0 3 1 5 5 2 2 1 0 3 6] - [ 3 960 1 1 10 21 3 13 2 0 0 1 3 0 0 1 4 1 2 1 6] - [ 5 3 973 13 3 1 12 11 0 1 1 4 0 3 3 6 3 1 3 4 8] - [ 7 0 23 921 0 4 0 1 1 2 13 0 4 0 17 1 2 4 13 0 3] - [ 14 7 0 0 990 12 2 1 1 0 0 5 3 3 9 7 3 2 1 2 4] - [ 3 11 1 5 4 982 3 16 0 4 1 9 2 10 0 1 8 1 1 2 6] - [ 1 2 12 1 0 6 1050 5 0 1 3 0 2 1 1 2 2 2 1 3 4] - [ 4 5 11 1 1 27 3 931 0 1 0 8 4 1 0 0 1 0 18 4 4] - [ 15 4 0 1 1 0 0 1 907 36 7 4 0 12 12 3 1 0 5 0 0] - [ 84 2 3 0 10 1 0 3 29 851 0 1 0 14 5 1 2 1 2 0 3] - [ 3 1 4 8 2 2 2 1 13 2 992 2 1 5 4 0 2 1 3 0 3] - [ 3 3 2 0 2 14 0 6 0 3 0 918 26 5 0 6 1 8 0 5 3] - [ 1 0 0 7 1 4 0 0 2 1 0 24 889 1 1 4 4 9 3 1 7] - [ 5 2 1 0 7 10 1 2 9 16 10 5 1 936 6 4 2 1 0 1 5] - [ 5 1 2 19 3 4 0 1 14 8 3 1 3 3 1005 0 1 4 6 0 9] - [ 3 1 7 1 5 1 2 1 0 0 0 7 10 2 0 974 10 6 1 7 8] - [ 1 6 1 1 10 2 0 0 1 0 0 1 4 3 3 9 1004 1 0 4 10] - [ 4 2 0 4 0 3 2 0 1 1 2 14 12 1 0 15 0 984 1 0 5] - [ 2 5 5 5 1 2 1 16 1 1 5 1 1 0 13 0 0 5 1018 2 2] - [ 1 2 0 0 0 6 7 8 0 0 0 14 3 1 1 6 4 2 0 1087 6] - [ 163 224 250 127 125 212 71 171 84 70 188 95 304 288 152 83 214 96 178 226 10113]] - -2023-02-13 18:45:33,845 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:45:33,845 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:45:33,851 - - -2023-02-13 18:45:33,851 - Training epoch: 308808 samples (256 per mini-batch) -2023-02-13 18:45:34,744 - Epoch: [199][ 10/ 1207] Overall Loss 0.201677 Objective Loss 0.201677 LR 0.000031 Time 0.089277 -2023-02-13 18:45:34,939 - Epoch: [199][ 20/ 1207] Overall Loss 0.192999 Objective Loss 0.192999 LR 0.000031 Time 0.054368 -2023-02-13 18:45:35,133 - Epoch: [199][ 30/ 1207] Overall Loss 0.190046 Objective Loss 0.190046 LR 0.000031 Time 0.042701 -2023-02-13 18:45:35,328 - Epoch: [199][ 40/ 1207] Overall Loss 0.197314 Objective Loss 0.197314 LR 0.000031 Time 0.036878 -2023-02-13 18:45:35,522 - Epoch: [199][ 50/ 1207] Overall Loss 0.196877 Objective Loss 0.196877 LR 0.000031 Time 0.033372 -2023-02-13 18:45:35,716 - Epoch: [199][ 60/ 1207] Overall Loss 0.194572 Objective Loss 0.194572 LR 0.000031 Time 0.031043 -2023-02-13 18:45:35,910 - Epoch: [199][ 70/ 1207] Overall Loss 0.190696 Objective Loss 0.190696 LR 0.000031 Time 0.029372 -2023-02-13 18:45:36,105 - Epoch: [199][ 80/ 1207] Overall Loss 0.191232 Objective Loss 0.191232 LR 0.000031 Time 0.028142 -2023-02-13 18:45:36,300 - Epoch: [199][ 90/ 1207] Overall Loss 0.190585 Objective Loss 0.190585 LR 0.000031 Time 0.027170 -2023-02-13 18:45:36,493 - Epoch: [199][ 100/ 1207] Overall Loss 0.189429 Objective Loss 0.189429 LR 0.000031 Time 0.026386 -2023-02-13 18:45:36,687 - Epoch: [199][ 110/ 1207] Overall Loss 0.189756 Objective Loss 0.189756 LR 0.000031 Time 0.025739 -2023-02-13 18:45:36,874 - Epoch: [199][ 120/ 1207] Overall Loss 0.191933 Objective Loss 0.191933 LR 0.000031 Time 0.025156 -2023-02-13 18:45:37,067 - Epoch: [199][ 130/ 1207] Overall Loss 0.190558 Objective Loss 0.190558 LR 0.000031 Time 0.024697 -2023-02-13 18:45:37,259 - Epoch: [199][ 140/ 1207] Overall Loss 0.190706 Objective Loss 0.190706 LR 0.000031 Time 0.024308 -2023-02-13 18:45:37,455 - Epoch: [199][ 150/ 1207] Overall Loss 0.190271 Objective Loss 0.190271 LR 0.000031 Time 0.023989 -2023-02-13 18:45:37,648 - Epoch: [199][ 160/ 1207] Overall Loss 0.190356 Objective Loss 0.190356 LR 0.000031 Time 0.023692 -2023-02-13 18:45:37,844 - Epoch: [199][ 170/ 1207] Overall Loss 0.191521 Objective Loss 0.191521 LR 0.000031 Time 0.023449 -2023-02-13 18:45:38,036 - Epoch: [199][ 180/ 1207] Overall Loss 0.191088 Objective Loss 0.191088 LR 0.000031 Time 0.023212 -2023-02-13 18:45:38,232 - Epoch: [199][ 190/ 1207] Overall Loss 0.191538 Objective Loss 0.191538 LR 0.000031 Time 0.023019 -2023-02-13 18:45:38,424 - Epoch: [199][ 200/ 1207] Overall Loss 0.190774 Objective Loss 0.190774 LR 0.000031 Time 0.022829 -2023-02-13 18:45:38,620 - Epoch: [199][ 210/ 1207] Overall Loss 0.190724 Objective Loss 0.190724 LR 0.000031 Time 0.022673 -2023-02-13 18:45:38,812 - Epoch: [199][ 220/ 1207] Overall Loss 0.190580 Objective Loss 0.190580 LR 0.000031 Time 0.022512 -2023-02-13 18:45:39,008 - Epoch: [199][ 230/ 1207] Overall Loss 0.190865 Objective Loss 0.190865 LR 0.000031 Time 0.022384 -2023-02-13 18:45:39,200 - Epoch: [199][ 240/ 1207] Overall Loss 0.190902 Objective Loss 0.190902 LR 0.000031 Time 0.022251 -2023-02-13 18:45:39,397 - Epoch: [199][ 250/ 1207] Overall Loss 0.190895 Objective Loss 0.190895 LR 0.000031 Time 0.022145 -2023-02-13 18:45:39,589 - Epoch: [199][ 260/ 1207] Overall Loss 0.190531 Objective Loss 0.190531 LR 0.000031 Time 0.022032 -2023-02-13 18:45:39,785 - Epoch: [199][ 270/ 1207] Overall Loss 0.190720 Objective Loss 0.190720 LR 0.000031 Time 0.021939 -2023-02-13 18:45:39,977 - Epoch: [199][ 280/ 1207] Overall Loss 0.191695 Objective Loss 0.191695 LR 0.000031 Time 0.021841 -2023-02-13 18:45:40,173 - Epoch: [199][ 290/ 1207] Overall Loss 0.190989 Objective Loss 0.190989 LR 0.000031 Time 0.021761 -2023-02-13 18:45:40,366 - Epoch: [199][ 300/ 1207] Overall Loss 0.190890 Objective Loss 0.190890 LR 0.000031 Time 0.021678 -2023-02-13 18:45:40,562 - Epoch: [199][ 310/ 1207] Overall Loss 0.191260 Objective Loss 0.191260 LR 0.000031 Time 0.021609 -2023-02-13 18:45:40,754 - Epoch: [199][ 320/ 1207] Overall Loss 0.191171 Objective Loss 0.191171 LR 0.000031 Time 0.021535 -2023-02-13 18:45:40,951 - Epoch: [199][ 330/ 1207] Overall Loss 0.191034 Objective Loss 0.191034 LR 0.000031 Time 0.021477 -2023-02-13 18:45:41,143 - Epoch: [199][ 340/ 1207] Overall Loss 0.190425 Objective Loss 0.190425 LR 0.000031 Time 0.021411 -2023-02-13 18:45:41,339 - Epoch: [199][ 350/ 1207] Overall Loss 0.190505 Objective Loss 0.190505 LR 0.000031 Time 0.021357 -2023-02-13 18:45:41,531 - Epoch: [199][ 360/ 1207] Overall Loss 0.189649 Objective Loss 0.189649 LR 0.000031 Time 0.021297 -2023-02-13 18:45:41,727 - Epoch: [199][ 370/ 1207] Overall Loss 0.189402 Objective Loss 0.189402 LR 0.000031 Time 0.021249 -2023-02-13 18:45:41,919 - Epoch: [199][ 380/ 1207] Overall Loss 0.189220 Objective Loss 0.189220 LR 0.000031 Time 0.021195 -2023-02-13 18:45:42,116 - Epoch: [199][ 390/ 1207] Overall Loss 0.188916 Objective Loss 0.188916 LR 0.000031 Time 0.021154 -2023-02-13 18:45:42,308 - Epoch: [199][ 400/ 1207] Overall Loss 0.189342 Objective Loss 0.189342 LR 0.000031 Time 0.021105 -2023-02-13 18:45:42,504 - Epoch: [199][ 410/ 1207] Overall Loss 0.189237 Objective Loss 0.189237 LR 0.000031 Time 0.021067 -2023-02-13 18:45:42,696 - Epoch: [199][ 420/ 1207] Overall Loss 0.188898 Objective Loss 0.188898 LR 0.000031 Time 0.021023 -2023-02-13 18:45:42,891 - Epoch: [199][ 430/ 1207] Overall Loss 0.189296 Objective Loss 0.189296 LR 0.000031 Time 0.020986 -2023-02-13 18:45:43,083 - Epoch: [199][ 440/ 1207] Overall Loss 0.189321 Objective Loss 0.189321 LR 0.000031 Time 0.020945 -2023-02-13 18:45:43,279 - Epoch: [199][ 450/ 1207] Overall Loss 0.189492 Objective Loss 0.189492 LR 0.000031 Time 0.020914 -2023-02-13 18:45:43,472 - Epoch: [199][ 460/ 1207] Overall Loss 0.189733 Objective Loss 0.189733 LR 0.000031 Time 0.020878 -2023-02-13 18:45:43,668 - Epoch: [199][ 470/ 1207] Overall Loss 0.190250 Objective Loss 0.190250 LR 0.000031 Time 0.020850 -2023-02-13 18:45:43,861 - Epoch: [199][ 480/ 1207] Overall Loss 0.190718 Objective Loss 0.190718 LR 0.000031 Time 0.020817 -2023-02-13 18:45:44,056 - Epoch: [199][ 490/ 1207] Overall Loss 0.190770 Objective Loss 0.190770 LR 0.000031 Time 0.020789 -2023-02-13 18:45:44,249 - Epoch: [199][ 500/ 1207] Overall Loss 0.190392 Objective Loss 0.190392 LR 0.000031 Time 0.020758 -2023-02-13 18:45:44,444 - Epoch: [199][ 510/ 1207] Overall Loss 0.190108 Objective Loss 0.190108 LR 0.000031 Time 0.020734 -2023-02-13 18:45:44,637 - Epoch: [199][ 520/ 1207] Overall Loss 0.190092 Objective Loss 0.190092 LR 0.000031 Time 0.020704 -2023-02-13 18:45:44,832 - Epoch: [199][ 530/ 1207] Overall Loss 0.189868 Objective Loss 0.189868 LR 0.000031 Time 0.020682 -2023-02-13 18:45:45,024 - Epoch: [199][ 540/ 1207] Overall Loss 0.189989 Objective Loss 0.189989 LR 0.000031 Time 0.020654 -2023-02-13 18:45:45,220 - Epoch: [199][ 550/ 1207] Overall Loss 0.190090 Objective Loss 0.190090 LR 0.000031 Time 0.020634 -2023-02-13 18:45:45,412 - Epoch: [199][ 560/ 1207] Overall Loss 0.190557 Objective Loss 0.190557 LR 0.000031 Time 0.020608 -2023-02-13 18:45:45,607 - Epoch: [199][ 570/ 1207] Overall Loss 0.190702 Objective Loss 0.190702 LR 0.000031 Time 0.020588 -2023-02-13 18:45:45,800 - Epoch: [199][ 580/ 1207] Overall Loss 0.190366 Objective Loss 0.190366 LR 0.000031 Time 0.020564 -2023-02-13 18:45:45,996 - Epoch: [199][ 590/ 1207] Overall Loss 0.190441 Objective Loss 0.190441 LR 0.000031 Time 0.020547 -2023-02-13 18:45:46,188 - Epoch: [199][ 600/ 1207] Overall Loss 0.190166 Objective Loss 0.190166 LR 0.000031 Time 0.020524 -2023-02-13 18:45:46,383 - Epoch: [199][ 610/ 1207] Overall Loss 0.189934 Objective Loss 0.189934 LR 0.000031 Time 0.020507 -2023-02-13 18:45:46,575 - Epoch: [199][ 620/ 1207] Overall Loss 0.190154 Objective Loss 0.190154 LR 0.000031 Time 0.020485 -2023-02-13 18:45:46,771 - Epoch: [199][ 630/ 1207] Overall Loss 0.190277 Objective Loss 0.190277 LR 0.000031 Time 0.020470 -2023-02-13 18:45:46,964 - Epoch: [199][ 640/ 1207] Overall Loss 0.190211 Objective Loss 0.190211 LR 0.000031 Time 0.020452 -2023-02-13 18:45:47,160 - Epoch: [199][ 650/ 1207] Overall Loss 0.190204 Objective Loss 0.190204 LR 0.000031 Time 0.020438 -2023-02-13 18:45:47,353 - Epoch: [199][ 660/ 1207] Overall Loss 0.190115 Objective Loss 0.190115 LR 0.000031 Time 0.020420 -2023-02-13 18:45:47,548 - Epoch: [199][ 670/ 1207] Overall Loss 0.190316 Objective Loss 0.190316 LR 0.000031 Time 0.020406 -2023-02-13 18:45:47,745 - Epoch: [199][ 680/ 1207] Overall Loss 0.190426 Objective Loss 0.190426 LR 0.000031 Time 0.020396 -2023-02-13 18:45:47,950 - Epoch: [199][ 690/ 1207] Overall Loss 0.190253 Objective Loss 0.190253 LR 0.000031 Time 0.020396 -2023-02-13 18:45:48,154 - Epoch: [199][ 700/ 1207] Overall Loss 0.190186 Objective Loss 0.190186 LR 0.000031 Time 0.020396 -2023-02-13 18:45:48,358 - Epoch: [199][ 710/ 1207] Overall Loss 0.190505 Objective Loss 0.190505 LR 0.000031 Time 0.020395 -2023-02-13 18:45:48,562 - Epoch: [199][ 720/ 1207] Overall Loss 0.190390 Objective Loss 0.190390 LR 0.000031 Time 0.020395 -2023-02-13 18:45:48,767 - Epoch: [199][ 730/ 1207] Overall Loss 0.190415 Objective Loss 0.190415 LR 0.000031 Time 0.020395 -2023-02-13 18:45:48,971 - Epoch: [199][ 740/ 1207] Overall Loss 0.190408 Objective Loss 0.190408 LR 0.000031 Time 0.020395 -2023-02-13 18:45:49,175 - Epoch: [199][ 750/ 1207] Overall Loss 0.190544 Objective Loss 0.190544 LR 0.000031 Time 0.020395 -2023-02-13 18:45:49,376 - Epoch: [199][ 760/ 1207] Overall Loss 0.190647 Objective Loss 0.190647 LR 0.000031 Time 0.020390 -2023-02-13 18:45:49,571 - Epoch: [199][ 770/ 1207] Overall Loss 0.190867 Objective Loss 0.190867 LR 0.000031 Time 0.020378 -2023-02-13 18:45:49,763 - Epoch: [199][ 780/ 1207] Overall Loss 0.190792 Objective Loss 0.190792 LR 0.000031 Time 0.020363 -2023-02-13 18:45:49,959 - Epoch: [199][ 790/ 1207] Overall Loss 0.190679 Objective Loss 0.190679 LR 0.000031 Time 0.020353 -2023-02-13 18:45:50,151 - Epoch: [199][ 800/ 1207] Overall Loss 0.190874 Objective Loss 0.190874 LR 0.000031 Time 0.020338 -2023-02-13 18:45:50,357 - Epoch: [199][ 810/ 1207] Overall Loss 0.190970 Objective Loss 0.190970 LR 0.000031 Time 0.020341 -2023-02-13 18:45:50,561 - Epoch: [199][ 820/ 1207] Overall Loss 0.190849 Objective Loss 0.190849 LR 0.000031 Time 0.020340 -2023-02-13 18:45:50,763 - Epoch: [199][ 830/ 1207] Overall Loss 0.190567 Objective Loss 0.190567 LR 0.000031 Time 0.020339 -2023-02-13 18:45:50,957 - Epoch: [199][ 840/ 1207] Overall Loss 0.190483 Objective Loss 0.190483 LR 0.000031 Time 0.020327 -2023-02-13 18:45:51,154 - Epoch: [199][ 850/ 1207] Overall Loss 0.190326 Objective Loss 0.190326 LR 0.000031 Time 0.020319 -2023-02-13 18:45:51,347 - Epoch: [199][ 860/ 1207] Overall Loss 0.190158 Objective Loss 0.190158 LR 0.000031 Time 0.020307 -2023-02-13 18:45:51,543 - Epoch: [199][ 870/ 1207] Overall Loss 0.190145 Objective Loss 0.190145 LR 0.000031 Time 0.020299 -2023-02-13 18:45:51,738 - Epoch: [199][ 880/ 1207] Overall Loss 0.190383 Objective Loss 0.190383 LR 0.000031 Time 0.020289 -2023-02-13 18:45:51,934 - Epoch: [199][ 890/ 1207] Overall Loss 0.190631 Objective Loss 0.190631 LR 0.000031 Time 0.020281 -2023-02-13 18:45:52,128 - Epoch: [199][ 900/ 1207] Overall Loss 0.190480 Objective Loss 0.190480 LR 0.000031 Time 0.020271 -2023-02-13 18:45:52,325 - Epoch: [199][ 910/ 1207] Overall Loss 0.190389 Objective Loss 0.190389 LR 0.000031 Time 0.020264 -2023-02-13 18:45:52,517 - Epoch: [199][ 920/ 1207] Overall Loss 0.190256 Objective Loss 0.190256 LR 0.000031 Time 0.020252 -2023-02-13 18:45:52,714 - Epoch: [199][ 930/ 1207] Overall Loss 0.190082 Objective Loss 0.190082 LR 0.000031 Time 0.020245 -2023-02-13 18:45:52,906 - Epoch: [199][ 940/ 1207] Overall Loss 0.189856 Objective Loss 0.189856 LR 0.000031 Time 0.020235 -2023-02-13 18:45:53,102 - Epoch: [199][ 950/ 1207] Overall Loss 0.189835 Objective Loss 0.189835 LR 0.000031 Time 0.020227 -2023-02-13 18:45:53,295 - Epoch: [199][ 960/ 1207] Overall Loss 0.190184 Objective Loss 0.190184 LR 0.000031 Time 0.020217 -2023-02-13 18:45:53,490 - Epoch: [199][ 970/ 1207] Overall Loss 0.190236 Objective Loss 0.190236 LR 0.000031 Time 0.020210 -2023-02-13 18:45:53,685 - Epoch: [199][ 980/ 1207] Overall Loss 0.190253 Objective Loss 0.190253 LR 0.000031 Time 0.020201 -2023-02-13 18:45:53,880 - Epoch: [199][ 990/ 1207] Overall Loss 0.190053 Objective Loss 0.190053 LR 0.000031 Time 0.020194 -2023-02-13 18:45:54,073 - Epoch: [199][ 1000/ 1207] Overall Loss 0.190117 Objective Loss 0.190117 LR 0.000031 Time 0.020185 -2023-02-13 18:45:54,269 - Epoch: [199][ 1010/ 1207] Overall Loss 0.190131 Objective Loss 0.190131 LR 0.000031 Time 0.020179 -2023-02-13 18:45:54,462 - Epoch: [199][ 1020/ 1207] Overall Loss 0.190355 Objective Loss 0.190355 LR 0.000031 Time 0.020170 -2023-02-13 18:45:54,659 - Epoch: [199][ 1030/ 1207] Overall Loss 0.190229 Objective Loss 0.190229 LR 0.000031 Time 0.020164 -2023-02-13 18:45:54,852 - Epoch: [199][ 1040/ 1207] Overall Loss 0.190174 Objective Loss 0.190174 LR 0.000031 Time 0.020156 -2023-02-13 18:45:55,048 - Epoch: [199][ 1050/ 1207] Overall Loss 0.190104 Objective Loss 0.190104 LR 0.000031 Time 0.020151 -2023-02-13 18:45:55,241 - Epoch: [199][ 1060/ 1207] Overall Loss 0.190367 Objective Loss 0.190367 LR 0.000031 Time 0.020142 -2023-02-13 18:45:55,437 - Epoch: [199][ 1070/ 1207] Overall Loss 0.190360 Objective Loss 0.190360 LR 0.000031 Time 0.020137 -2023-02-13 18:45:55,630 - Epoch: [199][ 1080/ 1207] Overall Loss 0.190317 Objective Loss 0.190317 LR 0.000031 Time 0.020129 -2023-02-13 18:45:55,827 - Epoch: [199][ 1090/ 1207] Overall Loss 0.190316 Objective Loss 0.190316 LR 0.000031 Time 0.020124 -2023-02-13 18:45:56,021 - Epoch: [199][ 1100/ 1207] Overall Loss 0.190494 Objective Loss 0.190494 LR 0.000031 Time 0.020118 -2023-02-13 18:45:56,217 - Epoch: [199][ 1110/ 1207] Overall Loss 0.190558 Objective Loss 0.190558 LR 0.000031 Time 0.020112 -2023-02-13 18:45:56,410 - Epoch: [199][ 1120/ 1207] Overall Loss 0.190595 Objective Loss 0.190595 LR 0.000031 Time 0.020105 -2023-02-13 18:45:56,606 - Epoch: [199][ 1130/ 1207] Overall Loss 0.190742 Objective Loss 0.190742 LR 0.000031 Time 0.020100 -2023-02-13 18:45:56,799 - Epoch: [199][ 1140/ 1207] Overall Loss 0.190682 Objective Loss 0.190682 LR 0.000031 Time 0.020092 -2023-02-13 18:45:56,995 - Epoch: [199][ 1150/ 1207] Overall Loss 0.190474 Objective Loss 0.190474 LR 0.000031 Time 0.020088 -2023-02-13 18:45:57,188 - Epoch: [199][ 1160/ 1207] Overall Loss 0.190584 Objective Loss 0.190584 LR 0.000031 Time 0.020081 -2023-02-13 18:45:57,385 - Epoch: [199][ 1170/ 1207] Overall Loss 0.190834 Objective Loss 0.190834 LR 0.000031 Time 0.020077 -2023-02-13 18:45:57,577 - Epoch: [199][ 1180/ 1207] Overall Loss 0.190870 Objective Loss 0.190870 LR 0.000031 Time 0.020070 -2023-02-13 18:45:57,773 - Epoch: [199][ 1190/ 1207] Overall Loss 0.190875 Objective Loss 0.190875 LR 0.000031 Time 0.020066 -2023-02-13 18:45:58,023 - Epoch: [199][ 1200/ 1207] Overall Loss 0.190818 Objective Loss 0.190818 LR 0.000031 Time 0.020106 -2023-02-13 18:45:58,138 - Epoch: [199][ 1207/ 1207] Overall Loss 0.190660 Objective Loss 0.190660 Top1 89.634146 Top5 99.085366 LR 0.000031 Time 0.020085 -2023-02-13 18:45:58,211 - --- validate (epoch=199)----------- -2023-02-13 18:45:58,211 - 34311 samples (256 per mini-batch) -2023-02-13 18:45:58,626 - Epoch: [199][ 10/ 135] Loss 0.295969 Top1 85.742188 Top5 97.578125 -2023-02-13 18:45:58,763 - Epoch: [199][ 20/ 135] Loss 0.279681 Top1 86.191406 Top5 97.578125 -2023-02-13 18:45:58,889 - Epoch: [199][ 30/ 135] Loss 0.289953 Top1 86.002604 Top5 97.447917 -2023-02-13 18:45:59,014 - Epoch: [199][ 40/ 135] Loss 0.286746 Top1 86.113281 Top5 97.656250 -2023-02-13 18:45:59,142 - Epoch: [199][ 50/ 135] Loss 0.279385 Top1 86.351562 Top5 97.703125 -2023-02-13 18:45:59,270 - Epoch: [199][ 60/ 135] Loss 0.283012 Top1 86.178385 Top5 97.623698 -2023-02-13 18:45:59,401 - Epoch: [199][ 70/ 135] Loss 0.287326 Top1 86.065848 Top5 97.656250 -2023-02-13 18:45:59,530 - Epoch: [199][ 80/ 135] Loss 0.287253 Top1 85.966797 Top5 97.675781 -2023-02-13 18:45:59,660 - Epoch: [199][ 90/ 135] Loss 0.283980 Top1 85.972222 Top5 97.725694 -2023-02-13 18:45:59,785 - Epoch: [199][ 100/ 135] Loss 0.280509 Top1 86.015625 Top5 97.781250 -2023-02-13 18:45:59,919 - Epoch: [199][ 110/ 135] Loss 0.282776 Top1 86.012074 Top5 97.833807 -2023-02-13 18:46:00,059 - Epoch: [199][ 120/ 135] Loss 0.284276 Top1 85.914714 Top5 97.825521 -2023-02-13 18:46:00,184 - Epoch: [199][ 130/ 135] Loss 0.284261 Top1 85.925481 Top5 97.845553 -2023-02-13 18:46:00,228 - Epoch: [199][ 135/ 135] Loss 0.281216 Top1 85.925796 Top5 97.857830 -2023-02-13 18:46:00,303 - ==> Top1: 85.926 Top5: 97.858 Loss: 0.281 - -2023-02-13 18:46:00,304 - ==> Confusion: -[[ 863 6 8 1 9 1 0 1 3 45 0 5 0 4 8 2 2 1 0 2 6] - [ 4 961 1 1 11 16 3 9 3 0 3 1 2 1 0 1 5 0 2 3 6] - [ 6 4 972 10 2 1 10 12 1 1 2 2 0 4 4 6 4 1 4 2 10] - [ 7 0 19 907 0 4 0 1 2 2 16 0 9 0 21 0 1 7 15 1 4] - [ 10 7 0 0 998 10 1 1 1 0 0 6 4 5 7 5 4 1 1 2 3] - [ 0 15 1 3 4 979 2 12 3 4 1 11 1 15 0 4 5 2 1 1 6] - [ 2 2 16 2 0 6 1042 4 0 1 2 1 1 1 1 2 3 4 1 2 6] - [ 4 9 9 2 0 30 3 933 0 1 0 6 3 1 0 0 1 1 14 4 3] - [ 9 3 2 1 1 0 0 2 929 25 5 2 0 8 15 2 1 0 3 0 1] - [ 66 0 2 0 8 0 0 2 38 864 0 1 0 18 3 4 1 0 2 0 3] - [ 2 1 2 4 1 1 2 3 15 1 996 2 1 7 4 0 1 1 3 0 4] - [ 3 3 2 0 2 13 0 4 1 2 0 931 17 6 1 5 1 9 1 4 0] - [ 0 0 0 5 1 3 0 0 3 1 0 34 880 2 1 5 2 12 1 1 8] - [ 3 3 1 1 6 9 0 1 9 17 5 4 1 946 4 3 4 2 0 0 5] - [ 5 3 1 13 2 6 0 1 17 6 2 1 3 3 1012 0 2 3 4 0 8] - [ 6 1 6 0 2 2 3 1 0 0 1 7 6 3 0 972 12 9 0 7 8] - [ 0 5 0 1 9 2 0 0 2 0 0 2 2 3 3 8 1007 1 1 4 11] - [ 5 2 0 3 0 2 2 0 1 0 0 11 12 1 0 14 0 991 0 1 6] - [ 3 7 6 4 1 2 1 17 3 0 6 2 4 0 14 2 1 3 1009 1 0] - [ 1 3 0 0 0 3 7 8 1 0 0 15 2 3 1 7 4 2 0 1083 8] - [ 140 239 213 101 121 201 60 163 104 90 186 119 281 273 155 75 240 107 145 214 10207]] - -2023-02-13 18:46:00,305 - ==> Best [Top1: 86.095 Top5: 97.995 Sparsity:0.00 Params: 148928 on epoch: 149] -2023-02-13 18:46:00,306 - Saving checkpoint to: logs/2023.02.13-171954/qat_checkpoint.pth.tar -2023-02-13 18:46:00,311 - --- test --------------------- -2023-02-13 18:46:00,311 - 36234 samples (256 per mini-batch) -2023-02-13 18:46:00,712 - Test: [ 10/ 142] Loss 0.283413 Top1 86.171875 Top5 98.164062 -2023-02-13 18:46:00,838 - Test: [ 20/ 142] Loss 0.286099 Top1 85.957031 Top5 98.046875 -2023-02-13 18:46:00,966 - Test: [ 30/ 142] Loss 0.283828 Top1 86.067708 Top5 98.203125 -2023-02-13 18:46:01,093 - Test: [ 40/ 142] Loss 0.280440 Top1 86.562500 Top5 98.242188 -2023-02-13 18:46:01,219 - Test: [ 50/ 142] Loss 0.279370 Top1 86.468750 Top5 98.234375 -2023-02-13 18:46:01,347 - Test: [ 60/ 142] Loss 0.278438 Top1 86.549479 Top5 98.281250 -2023-02-13 18:46:01,473 - Test: [ 70/ 142] Loss 0.281202 Top1 86.523438 Top5 98.281250 -2023-02-13 18:46:01,598 - Test: [ 80/ 142] Loss 0.281989 Top1 86.411133 Top5 98.276367 -2023-02-13 18:46:01,733 - Test: [ 90/ 142] Loss 0.283541 Top1 86.401910 Top5 98.315972 -2023-02-13 18:46:01,877 - Test: [ 100/ 142] Loss 0.281960 Top1 86.519531 Top5 98.312500 -2023-02-13 18:46:02,008 - Test: [ 110/ 142] Loss 0.280234 Top1 86.534091 Top5 98.338068 -2023-02-13 18:46:02,140 - Test: [ 120/ 142] Loss 0.277140 Top1 86.559245 Top5 98.330078 -2023-02-13 18:46:02,271 - Test: [ 130/ 142] Loss 0.283737 Top1 86.427284 Top5 98.311298 -2023-02-13 18:46:02,400 - Test: [ 140/ 142] Loss 0.285620 Top1 86.400670 Top5 98.295201 -2023-02-13 18:46:02,420 - Test: [ 142/ 142] Loss 0.285232 Top1 86.427113 Top5 98.299939 -2023-02-13 18:46:02,507 - ==> Top1: 86.427 Top5: 98.300 Loss: 0.285 - -2023-02-13 18:46:02,508 - ==> Confusion: -[[ 913 1 3 0 7 3 0 1 11 42 0 1 1 0 5 0 3 4 1 0 3] - [ 2 1055 0 0 5 26 4 13 0 1 6 2 1 0 3 0 4 2 7 0 3] - [ 6 1 999 9 1 0 10 4 1 2 6 0 2 0 0 4 2 2 6 0 16] - [ 0 1 23 1006 1 2 0 4 1 0 9 1 5 1 16 4 1 2 14 0 7] - [ 9 8 0 0 999 6 0 5 0 5 0 3 0 3 4 5 7 2 0 0 6] - [ 2 32 3 2 5 1001 1 25 1 2 1 3 3 8 3 1 4 1 2 3 7] - [ 1 3 19 1 0 4 1097 8 1 0 3 3 1 2 0 3 0 1 0 8 12] - [ 3 16 9 3 0 41 0 995 3 0 5 3 1 0 2 0 0 2 29 7 3] - [ 9 1 0 1 0 5 0 3 944 24 6 0 0 8 21 0 0 4 0 1 2] - [ 67 0 2 0 2 5 2 2 29 925 0 2 0 13 10 4 0 1 0 1 6] - [ 0 8 6 8 0 5 1 3 7 0 1007 1 6 18 3 1 1 0 11 0 3] - [ 2 1 0 0 3 16 1 2 3 1 0 1018 14 3 2 2 2 5 0 12 2] - [ 1 0 0 2 2 1 0 3 0 0 2 35 931 0 3 8 5 24 0 4 8] - [ 1 3 1 1 2 6 1 4 20 10 7 2 5 986 3 2 7 4 1 2 3] - [ 9 3 2 14 4 4 0 0 25 4 0 2 3 2 1028 0 3 8 12 0 8] - [ 0 0 0 3 2 0 3 5 1 0 0 9 2 0 3 1055 9 10 0 4 4] - [ 0 1 0 2 2 5 1 3 4 0 0 5 5 0 4 20 1052 1 0 1 4] - [ 1 0 2 4 0 1 2 0 1 2 0 12 9 2 1 7 0 1062 0 1 0] - [ 4 9 6 7 1 1 1 15 5 1 4 0 2 0 11 3 0 0 977 3 3] - [ 0 5 1 0 4 2 7 13 1 0 0 8 7 1 0 12 6 1 1 1053 3] - [ 159 225 206 123 128 188 75 163 88 92 174 129 318 269 170 82 223 79 131 222 11213]] - -2023-02-13 18:46:02,629 - -2023-02-13 18:46:02,629 - Log file for this run: /data/ml/afshin/ai/kws20-enhancement/ai8x-training/logs/2023.02.13-171954/2023.02.13-171954.log +2023-10-05 20:42:41,531 - Log file for this run: /home/alicangok/Projects/AI8X/train_clean/logs/2023.10.05-204241/2023.10.05-204241.log +2023-10-05 20:42:43,645 - Optimizer Type: +2023-10-05 20:42:43,645 - Optimizer Args: {'lr': 0.001, 'betas': (0.9, 0.999), 'eps': 1e-08, 'weight_decay': 0.0, 'amsgrad': False} +2023-10-05 20:42:51,357 - Dataset sizes: + training=316395 + validation=29943 + test=33015 +2023-10-05 20:42:51,357 - Reading compression schedule from: policies/schedule_kws20.yaml +2023-10-05 20:42:51,359 - + +2023-10-05 20:42:51,360 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:42:52,624 - Epoch: [0][ 10/ 1236] Overall Loss 3.044521 Objective Loss 3.044521 LR 0.001000 Time 0.126254 +2023-10-05 20:42:52,789 - Epoch: [0][ 20/ 1236] Overall Loss 3.043474 Objective Loss 3.043474 LR 0.001000 Time 0.071376 +2023-10-05 20:42:52,951 - Epoch: [0][ 30/ 1236] Overall Loss 3.041399 Objective Loss 3.041399 LR 0.001000 Time 0.052958 +2023-10-05 20:42:53,112 - Epoch: [0][ 40/ 1236] Overall Loss 3.037595 Objective Loss 3.037595 LR 0.001000 Time 0.043737 +2023-10-05 20:42:53,273 - Epoch: [0][ 50/ 1236] Overall Loss 3.029175 Objective Loss 3.029175 LR 0.001000 Time 0.038199 +2023-10-05 20:42:53,434 - Epoch: [0][ 60/ 1236] Overall Loss 3.016609 Objective Loss 3.016609 LR 0.001000 Time 0.034520 +2023-10-05 20:42:53,595 - Epoch: [0][ 70/ 1236] Overall Loss 3.003995 Objective Loss 3.003995 LR 0.001000 Time 0.031882 +2023-10-05 20:42:53,756 - Epoch: [0][ 80/ 1236] Overall Loss 2.983447 Objective Loss 2.983447 LR 0.001000 Time 0.029899 +2023-10-05 20:42:53,916 - Epoch: [0][ 90/ 1236] Overall Loss 2.958151 Objective Loss 2.958151 LR 0.001000 Time 0.028355 +2023-10-05 20:42:54,077 - Epoch: [0][ 100/ 1236] Overall Loss 2.932277 Objective Loss 2.932277 LR 0.001000 Time 0.027125 +2023-10-05 20:42:54,237 - Epoch: [0][ 110/ 1236] Overall Loss 2.907136 Objective Loss 2.907136 LR 0.001000 Time 0.026110 +2023-10-05 20:42:54,397 - Epoch: [0][ 120/ 1236] Overall Loss 2.885512 Objective Loss 2.885512 LR 0.001000 Time 0.025270 +2023-10-05 20:42:54,557 - Epoch: [0][ 130/ 1236] Overall Loss 2.861067 Objective Loss 2.861067 LR 0.001000 Time 0.024554 +2023-10-05 20:42:54,718 - Epoch: [0][ 140/ 1236] Overall Loss 2.835966 Objective Loss 2.835966 LR 0.001000 Time 0.023944 +2023-10-05 20:42:54,878 - Epoch: [0][ 150/ 1236] Overall Loss 2.815935 Objective Loss 2.815935 LR 0.001000 Time 0.023413 +2023-10-05 20:42:55,038 - Epoch: [0][ 160/ 1236] Overall Loss 2.796071 Objective Loss 2.796071 LR 0.001000 Time 0.022951 +2023-10-05 20:42:55,198 - Epoch: [0][ 170/ 1236] Overall Loss 2.776232 Objective Loss 2.776232 LR 0.001000 Time 0.022540 +2023-10-05 20:42:55,359 - Epoch: [0][ 180/ 1236] Overall Loss 2.756368 Objective Loss 2.756368 LR 0.001000 Time 0.022178 +2023-10-05 20:42:55,519 - Epoch: [0][ 190/ 1236] Overall Loss 2.738630 Objective Loss 2.738630 LR 0.001000 Time 0.021852 +2023-10-05 20:42:55,680 - Epoch: [0][ 200/ 1236] Overall Loss 2.722901 Objective Loss 2.722901 LR 0.001000 Time 0.021562 +2023-10-05 20:42:55,841 - Epoch: [0][ 210/ 1236] Overall Loss 2.706463 Objective Loss 2.706463 LR 0.001000 Time 0.021302 +2023-10-05 20:42:56,003 - Epoch: [0][ 220/ 1236] Overall Loss 2.689623 Objective Loss 2.689623 LR 0.001000 Time 0.021067 +2023-10-05 20:42:56,164 - Epoch: [0][ 230/ 1236] Overall Loss 2.670562 Objective Loss 2.670562 LR 0.001000 Time 0.020850 +2023-10-05 20:42:56,325 - Epoch: [0][ 240/ 1236] Overall Loss 2.653258 Objective Loss 2.653258 LR 0.001000 Time 0.020652 +2023-10-05 20:42:56,487 - Epoch: [0][ 250/ 1236] Overall Loss 2.637443 Objective Loss 2.637443 LR 0.001000 Time 0.020471 +2023-10-05 20:42:56,648 - Epoch: [0][ 260/ 1236] Overall Loss 2.622395 Objective Loss 2.622395 LR 0.001000 Time 0.020304 +2023-10-05 20:42:56,809 - Epoch: [0][ 270/ 1236] Overall Loss 2.608909 Objective Loss 2.608909 LR 0.001000 Time 0.020147 +2023-10-05 20:42:56,971 - Epoch: [0][ 280/ 1236] Overall Loss 2.594065 Objective Loss 2.594065 LR 0.001000 Time 0.020003 +2023-10-05 20:42:57,132 - Epoch: [0][ 290/ 1236] Overall Loss 2.579045 Objective Loss 2.579045 LR 0.001000 Time 0.019868 +2023-10-05 20:42:57,293 - Epoch: [0][ 300/ 1236] Overall Loss 2.565607 Objective Loss 2.565607 LR 0.001000 Time 0.019742 +2023-10-05 20:42:57,454 - Epoch: [0][ 310/ 1236] Overall Loss 2.551466 Objective Loss 2.551466 LR 0.001000 Time 0.019623 +2023-10-05 20:42:57,616 - Epoch: [0][ 320/ 1236] Overall Loss 2.538432 Objective Loss 2.538432 LR 0.001000 Time 0.019515 +2023-10-05 20:42:57,777 - Epoch: [0][ 330/ 1236] Overall Loss 2.524052 Objective Loss 2.524052 LR 0.001000 Time 0.019410 +2023-10-05 20:42:57,938 - Epoch: [0][ 340/ 1236] Overall Loss 2.509787 Objective Loss 2.509787 LR 0.001000 Time 0.019313 +2023-10-05 20:42:58,099 - Epoch: [0][ 350/ 1236] Overall Loss 2.496098 Objective Loss 2.496098 LR 0.001000 Time 0.019221 +2023-10-05 20:42:58,261 - Epoch: [0][ 360/ 1236] Overall Loss 2.484683 Objective Loss 2.484683 LR 0.001000 Time 0.019136 +2023-10-05 20:42:58,423 - Epoch: [0][ 370/ 1236] Overall Loss 2.472530 Objective Loss 2.472530 LR 0.001000 Time 0.019055 +2023-10-05 20:42:58,584 - Epoch: [0][ 380/ 1236] Overall Loss 2.461318 Objective Loss 2.461318 LR 0.001000 Time 0.018977 +2023-10-05 20:42:58,745 - Epoch: [0][ 390/ 1236] Overall Loss 2.449790 Objective Loss 2.449790 LR 0.001000 Time 0.018902 +2023-10-05 20:42:58,907 - Epoch: [0][ 400/ 1236] Overall Loss 2.437974 Objective Loss 2.437974 LR 0.001000 Time 0.018834 +2023-10-05 20:42:59,068 - Epoch: [0][ 410/ 1236] Overall Loss 2.426250 Objective Loss 2.426250 LR 0.001000 Time 0.018766 +2023-10-05 20:42:59,230 - Epoch: [0][ 420/ 1236] Overall Loss 2.416008 Objective Loss 2.416008 LR 0.001000 Time 0.018704 +2023-10-05 20:42:59,391 - Epoch: [0][ 430/ 1236] Overall Loss 2.404747 Objective Loss 2.404747 LR 0.001000 Time 0.018643 +2023-10-05 20:42:59,553 - Epoch: [0][ 440/ 1236] Overall Loss 2.393470 Objective Loss 2.393470 LR 0.001000 Time 0.018586 +2023-10-05 20:42:59,714 - Epoch: [0][ 450/ 1236] Overall Loss 2.381681 Objective Loss 2.381681 LR 0.001000 Time 0.018531 +2023-10-05 20:42:59,876 - Epoch: [0][ 460/ 1236] Overall Loss 2.370645 Objective Loss 2.370645 LR 0.001000 Time 0.018480 +2023-10-05 20:43:00,038 - Epoch: [0][ 470/ 1236] Overall Loss 2.359651 Objective Loss 2.359651 LR 0.001000 Time 0.018430 +2023-10-05 20:43:00,200 - Epoch: [0][ 480/ 1236] Overall Loss 2.349306 Objective Loss 2.349306 LR 0.001000 Time 0.018383 +2023-10-05 20:43:00,361 - Epoch: [0][ 490/ 1236] Overall Loss 2.338880 Objective Loss 2.338880 LR 0.001000 Time 0.018337 +2023-10-05 20:43:00,523 - Epoch: [0][ 500/ 1236] Overall Loss 2.328520 Objective Loss 2.328520 LR 0.001000 Time 0.018294 +2023-10-05 20:43:00,684 - Epoch: [0][ 510/ 1236] Overall Loss 2.319047 Objective Loss 2.319047 LR 0.001000 Time 0.018250 +2023-10-05 20:43:00,847 - Epoch: [0][ 520/ 1236] Overall Loss 2.308662 Objective Loss 2.308662 LR 0.001000 Time 0.018211 +2023-10-05 20:43:01,009 - Epoch: [0][ 530/ 1236] Overall Loss 2.298026 Objective Loss 2.298026 LR 0.001000 Time 0.018172 +2023-10-05 20:43:01,170 - Epoch: [0][ 540/ 1236] Overall Loss 2.289064 Objective Loss 2.289064 LR 0.001000 Time 0.018134 +2023-10-05 20:43:01,332 - Epoch: [0][ 550/ 1236] Overall Loss 2.279539 Objective Loss 2.279539 LR 0.001000 Time 0.018098 +2023-10-05 20:43:01,494 - Epoch: [0][ 560/ 1236] Overall Loss 2.269625 Objective Loss 2.269625 LR 0.001000 Time 0.018063 +2023-10-05 20:43:01,655 - Epoch: [0][ 570/ 1236] Overall Loss 2.260846 Objective Loss 2.260846 LR 0.001000 Time 0.018029 +2023-10-05 20:43:01,817 - Epoch: [0][ 580/ 1236] Overall Loss 2.251048 Objective Loss 2.251048 LR 0.001000 Time 0.017997 +2023-10-05 20:43:01,979 - Epoch: [0][ 590/ 1236] Overall Loss 2.242042 Objective Loss 2.242042 LR 0.001000 Time 0.017965 +2023-10-05 20:43:02,141 - Epoch: [0][ 600/ 1236] Overall Loss 2.234683 Objective Loss 2.234683 LR 0.001000 Time 0.017935 +2023-10-05 20:43:02,302 - Epoch: [0][ 610/ 1236] Overall Loss 2.226055 Objective Loss 2.226055 LR 0.001000 Time 0.017905 +2023-10-05 20:43:02,464 - Epoch: [0][ 620/ 1236] Overall Loss 2.217549 Objective Loss 2.217549 LR 0.001000 Time 0.017878 +2023-10-05 20:43:02,626 - Epoch: [0][ 630/ 1236] Overall Loss 2.208980 Objective Loss 2.208980 LR 0.001000 Time 0.017850 +2023-10-05 20:43:02,788 - Epoch: [0][ 640/ 1236] Overall Loss 2.200253 Objective Loss 2.200253 LR 0.001000 Time 0.017824 +2023-10-05 20:43:02,950 - Epoch: [0][ 650/ 1236] Overall Loss 2.192506 Objective Loss 2.192506 LR 0.001000 Time 0.017798 +2023-10-05 20:43:03,112 - Epoch: [0][ 660/ 1236] Overall Loss 2.184375 Objective Loss 2.184375 LR 0.001000 Time 0.017774 +2023-10-05 20:43:03,274 - Epoch: [0][ 670/ 1236] Overall Loss 2.175788 Objective Loss 2.175788 LR 0.001000 Time 0.017750 +2023-10-05 20:43:03,436 - Epoch: [0][ 680/ 1236] Overall Loss 2.167705 Objective Loss 2.167705 LR 0.001000 Time 0.017727 +2023-10-05 20:43:03,598 - Epoch: [0][ 690/ 1236] Overall Loss 2.159969 Objective Loss 2.159969 LR 0.001000 Time 0.017703 +2023-10-05 20:43:03,760 - Epoch: [0][ 700/ 1236] Overall Loss 2.152331 Objective Loss 2.152331 LR 0.001000 Time 0.017682 +2023-10-05 20:43:03,921 - Epoch: [0][ 710/ 1236] Overall Loss 2.144959 Objective Loss 2.144959 LR 0.001000 Time 0.017659 +2023-10-05 20:43:04,083 - Epoch: [0][ 720/ 1236] Overall Loss 2.137323 Objective Loss 2.137323 LR 0.001000 Time 0.017639 +2023-10-05 20:43:04,245 - Epoch: [0][ 730/ 1236] Overall Loss 2.130174 Objective Loss 2.130174 LR 0.001000 Time 0.017618 +2023-10-05 20:43:04,407 - Epoch: [0][ 740/ 1236] Overall Loss 2.122998 Objective Loss 2.122998 LR 0.001000 Time 0.017599 +2023-10-05 20:43:04,569 - Epoch: [0][ 750/ 1236] Overall Loss 2.115688 Objective Loss 2.115688 LR 0.001000 Time 0.017580 +2023-10-05 20:43:04,731 - Epoch: [0][ 760/ 1236] Overall Loss 2.108823 Objective Loss 2.108823 LR 0.001000 Time 0.017561 +2023-10-05 20:43:04,892 - Epoch: [0][ 770/ 1236] Overall Loss 2.101356 Objective Loss 2.101356 LR 0.001000 Time 0.017542 +2023-10-05 20:43:05,054 - Epoch: [0][ 780/ 1236] Overall Loss 2.094202 Objective Loss 2.094202 LR 0.001000 Time 0.017525 +2023-10-05 20:43:05,216 - Epoch: [0][ 790/ 1236] Overall Loss 2.087874 Objective Loss 2.087874 LR 0.001000 Time 0.017507 +2023-10-05 20:43:05,377 - Epoch: [0][ 800/ 1236] Overall Loss 2.081260 Objective Loss 2.081260 LR 0.001000 Time 0.017489 +2023-10-05 20:43:05,538 - Epoch: [0][ 810/ 1236] Overall Loss 2.074659 Objective Loss 2.074659 LR 0.001000 Time 0.017472 +2023-10-05 20:43:05,700 - Epoch: [0][ 820/ 1236] Overall Loss 2.067529 Objective Loss 2.067529 LR 0.001000 Time 0.017456 +2023-10-05 20:43:05,861 - Epoch: [0][ 830/ 1236] Overall Loss 2.061074 Objective Loss 2.061074 LR 0.001000 Time 0.017439 +2023-10-05 20:43:06,022 - Epoch: [0][ 840/ 1236] Overall Loss 2.054614 Objective Loss 2.054614 LR 0.001000 Time 0.017423 +2023-10-05 20:43:06,183 - Epoch: [0][ 850/ 1236] Overall Loss 2.047830 Objective Loss 2.047830 LR 0.001000 Time 0.017407 +2023-10-05 20:43:06,345 - Epoch: [0][ 860/ 1236] Overall Loss 2.041490 Objective Loss 2.041490 LR 0.001000 Time 0.017393 +2023-10-05 20:43:06,507 - Epoch: [0][ 870/ 1236] Overall Loss 2.034894 Objective Loss 2.034894 LR 0.001000 Time 0.017378 +2023-10-05 20:43:06,669 - Epoch: [0][ 880/ 1236] Overall Loss 2.028470 Objective Loss 2.028470 LR 0.001000 Time 0.017364 +2023-10-05 20:43:06,830 - Epoch: [0][ 890/ 1236] Overall Loss 2.021690 Objective Loss 2.021690 LR 0.001000 Time 0.017350 +2023-10-05 20:43:06,992 - Epoch: [0][ 900/ 1236] Overall Loss 2.015243 Objective Loss 2.015243 LR 0.001000 Time 0.017337 +2023-10-05 20:43:07,153 - Epoch: [0][ 910/ 1236] Overall Loss 2.009380 Objective Loss 2.009380 LR 0.001000 Time 0.017323 +2023-10-05 20:43:07,315 - Epoch: [0][ 920/ 1236] Overall Loss 2.003216 Objective Loss 2.003216 LR 0.001000 Time 0.017311 +2023-10-05 20:43:07,476 - Epoch: [0][ 930/ 1236] Overall Loss 1.997177 Objective Loss 1.997177 LR 0.001000 Time 0.017297 +2023-10-05 20:43:07,638 - Epoch: [0][ 940/ 1236] Overall Loss 1.991647 Objective Loss 1.991647 LR 0.001000 Time 0.017285 +2023-10-05 20:43:07,800 - Epoch: [0][ 950/ 1236] Overall Loss 1.986270 Objective Loss 1.986270 LR 0.001000 Time 0.017273 +2023-10-05 20:43:07,962 - Epoch: [0][ 960/ 1236] Overall Loss 1.980641 Objective Loss 1.980641 LR 0.001000 Time 0.017262 +2023-10-05 20:43:08,123 - Epoch: [0][ 970/ 1236] Overall Loss 1.975123 Objective Loss 1.975123 LR 0.001000 Time 0.017249 +2023-10-05 20:43:08,285 - Epoch: [0][ 980/ 1236] Overall Loss 1.969148 Objective Loss 1.969148 LR 0.001000 Time 0.017238 +2023-10-05 20:43:08,446 - Epoch: [0][ 990/ 1236] Overall Loss 1.963434 Objective Loss 1.963434 LR 0.001000 Time 0.017227 +2023-10-05 20:43:08,609 - Epoch: [0][ 1000/ 1236] Overall Loss 1.957826 Objective Loss 1.957826 LR 0.001000 Time 0.017217 +2023-10-05 20:43:08,770 - Epoch: [0][ 1010/ 1236] Overall Loss 1.951921 Objective Loss 1.951921 LR 0.001000 Time 0.017206 +2023-10-05 20:43:08,933 - Epoch: [0][ 1020/ 1236] Overall Loss 1.946719 Objective Loss 1.946719 LR 0.001000 Time 0.017196 +2023-10-05 20:43:09,094 - Epoch: [0][ 1030/ 1236] Overall Loss 1.941507 Objective Loss 1.941507 LR 0.001000 Time 0.017186 +2023-10-05 20:43:09,256 - Epoch: [0][ 1040/ 1236] Overall Loss 1.936565 Objective Loss 1.936565 LR 0.001000 Time 0.017176 +2023-10-05 20:43:09,418 - Epoch: [0][ 1050/ 1236] Overall Loss 1.931181 Objective Loss 1.931181 LR 0.001000 Time 0.017166 +2023-10-05 20:43:09,579 - Epoch: [0][ 1060/ 1236] Overall Loss 1.926516 Objective Loss 1.926516 LR 0.001000 Time 0.017156 +2023-10-05 20:43:09,740 - Epoch: [0][ 1070/ 1236] Overall Loss 1.921710 Objective Loss 1.921710 LR 0.001000 Time 0.017146 +2023-10-05 20:43:09,903 - Epoch: [0][ 1080/ 1236] Overall Loss 1.917558 Objective Loss 1.917558 LR 0.001000 Time 0.017137 +2023-10-05 20:43:10,063 - Epoch: [0][ 1090/ 1236] Overall Loss 1.912682 Objective Loss 1.912682 LR 0.001000 Time 0.017127 +2023-10-05 20:43:10,225 - Epoch: [0][ 1100/ 1236] Overall Loss 1.908099 Objective Loss 1.908099 LR 0.001000 Time 0.017119 +2023-10-05 20:43:10,387 - Epoch: [0][ 1110/ 1236] Overall Loss 1.903168 Objective Loss 1.903168 LR 0.001000 Time 0.017110 +2023-10-05 20:43:10,549 - Epoch: [0][ 1120/ 1236] Overall Loss 1.897969 Objective Loss 1.897969 LR 0.001000 Time 0.017101 +2023-10-05 20:43:10,710 - Epoch: [0][ 1130/ 1236] Overall Loss 1.893202 Objective Loss 1.893202 LR 0.001000 Time 0.017092 +2023-10-05 20:43:10,872 - Epoch: [0][ 1140/ 1236] Overall Loss 1.888685 Objective Loss 1.888685 LR 0.001000 Time 0.017084 +2023-10-05 20:43:11,033 - Epoch: [0][ 1150/ 1236] Overall Loss 1.884159 Objective Loss 1.884159 LR 0.001000 Time 0.017075 +2023-10-05 20:43:11,195 - Epoch: [0][ 1160/ 1236] Overall Loss 1.879371 Objective Loss 1.879371 LR 0.001000 Time 0.017068 +2023-10-05 20:43:11,356 - Epoch: [0][ 1170/ 1236] Overall Loss 1.874747 Objective Loss 1.874747 LR 0.001000 Time 0.017059 +2023-10-05 20:43:11,518 - Epoch: [0][ 1180/ 1236] Overall Loss 1.869962 Objective Loss 1.869962 LR 0.001000 Time 0.017052 +2023-10-05 20:43:11,680 - Epoch: [0][ 1190/ 1236] Overall Loss 1.865491 Objective Loss 1.865491 LR 0.001000 Time 0.017044 +2023-10-05 20:43:11,844 - Epoch: [0][ 1200/ 1236] Overall Loss 1.860739 Objective Loss 1.860739 LR 0.001000 Time 0.017038 +2023-10-05 20:43:12,010 - Epoch: [0][ 1210/ 1236] Overall Loss 1.856510 Objective Loss 1.856510 LR 0.001000 Time 0.017035 +2023-10-05 20:43:12,176 - Epoch: [0][ 1220/ 1236] Overall Loss 1.852472 Objective Loss 1.852472 LR 0.001000 Time 0.017031 +2023-10-05 20:43:12,378 - Epoch: [0][ 1230/ 1236] Overall Loss 1.848307 Objective Loss 1.848307 LR 0.001000 Time 0.017056 +2023-10-05 20:43:12,468 - Epoch: [0][ 1236/ 1236] Overall Loss 1.845361 Objective Loss 1.845361 Top1 56.008147 Top5 87.780041 LR 0.001000 Time 0.017046 +2023-10-05 20:43:12,592 - --- validate (epoch=0)----------- +2023-10-05 20:43:12,592 - 29943 samples (256 per mini-batch) +2023-10-05 20:43:12,978 - Epoch: [0][ 10/ 117] Loss 1.293120 Top1 48.789062 Top5 85.468750 +2023-10-05 20:43:13,079 - Epoch: [0][ 20/ 117] Loss 1.297473 Top1 48.515625 Top5 85.605469 +2023-10-05 20:43:13,183 - Epoch: [0][ 30/ 117] Loss 1.295066 Top1 48.307292 Top5 85.325521 +2023-10-05 20:43:13,285 - Epoch: [0][ 40/ 117] Loss 1.288552 Top1 48.339844 Top5 85.380859 +2023-10-05 20:43:13,389 - Epoch: [0][ 50/ 117] Loss 1.278729 Top1 48.679688 Top5 85.679688 +2023-10-05 20:43:13,490 - Epoch: [0][ 60/ 117] Loss 1.283351 Top1 48.652344 Top5 85.696615 +2023-10-05 20:43:13,594 - Epoch: [0][ 70/ 117] Loss 1.280885 Top1 48.744420 Top5 85.664062 +2023-10-05 20:43:13,696 - Epoch: [0][ 80/ 117] Loss 1.280529 Top1 48.706055 Top5 85.688477 +2023-10-05 20:43:13,800 - Epoch: [0][ 90/ 117] Loss 1.281169 Top1 48.684896 Top5 85.711806 +2023-10-05 20:43:13,902 - Epoch: [0][ 100/ 117] Loss 1.280036 Top1 48.644531 Top5 85.738281 +2023-10-05 20:43:14,013 - Epoch: [0][ 110/ 117] Loss 1.281341 Top1 48.590199 Top5 85.791903 +2023-10-05 20:43:14,068 - Epoch: [0][ 117/ 117] Loss 1.279656 Top1 48.552249 Top5 85.739572 +2023-10-05 20:43:14,186 - ==> Top1: 48.552 Top5: 85.740 Loss: 1.280 + +2023-10-05 20:43:14,186 - ==> Confusion: +[[ 749 4 3 0 29 8 0 2 11 133 1 4 4 32 22 2 10 8 5 1 22] + [ 2 676 17 3 42 44 2 29 38 2 13 2 0 21 46 1 51 4 121 2 15] + [ 26 9 479 16 33 60 125 135 0 13 10 27 7 29 0 7 5 3 15 27 30] + [ 4 33 13 520 15 46 7 13 18 0 104 6 22 24 68 13 27 31 90 0 35] + [ 33 32 3 0 818 16 1 0 0 16 1 0 1 18 21 17 36 2 5 0 30] + [ 7 213 25 14 24 353 4 90 6 4 9 25 20 165 11 2 62 4 38 12 28] + [ 2 9 175 4 6 19 777 74 0 0 3 14 1 0 0 53 2 2 13 15 22] + [ 4 34 42 5 13 59 13 649 6 4 15 9 6 11 2 2 3 1 293 16 31] + [ 54 18 0 1 7 1 0 1 796 58 0 1 2 16 95 1 11 2 12 1 12] + [ 296 9 3 0 38 1 1 0 63 598 0 0 0 46 28 0 1 1 0 0 34] + [ 9 120 44 17 24 56 3 49 12 5 306 5 3 8 18 3 8 1 341 0 21] + [ 3 4 3 0 0 16 5 10 0 0 0 623 178 34 1 11 22 31 9 74 11] + [ 1 3 1 4 0 9 3 2 14 0 0 194 645 7 5 6 38 115 6 0 15] + [ 19 21 3 0 13 59 0 2 21 20 0 34 19 825 19 5 25 12 0 4 18] + [ 25 74 2 5 42 5 0 1 134 14 0 0 4 4 725 0 25 12 7 0 22] + [ 1 4 3 1 19 9 9 0 0 0 0 93 22 1 0 830 23 92 9 1 17] + [ 9 71 0 4 27 20 1 1 20 0 1 20 11 17 36 39 844 8 6 2 24] + [ 1 3 0 0 0 3 0 0 14 0 0 59 269 4 4 30 3 619 3 0 26] + [ 1 26 4 18 2 5 3 64 16 2 22 2 5 0 23 0 6 0 852 2 15] + [ 0 4 12 0 0 10 6 84 0 2 0 87 5 49 0 10 5 3 15 826 34] + [ 337 772 133 83 373 356 44 244 353 164 63 393 505 400 498 146 556 240 704 513 1028]] + +2023-10-05 20:43:14,187 - ==> Best [Top1: 48.552 Top5: 85.740 Sparsity:0.00 Params: 148928 on epoch: 0] +2023-10-05 20:43:14,187 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:43:14,194 - + +2023-10-05 20:43:14,194 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:43:15,119 - Epoch: [1][ 10/ 1236] Overall Loss 1.263910 Objective Loss 1.263910 LR 0.001000 Time 0.092444 +2023-10-05 20:43:15,283 - Epoch: [1][ 20/ 1236] Overall Loss 1.262766 Objective Loss 1.262766 LR 0.001000 Time 0.054431 +2023-10-05 20:43:15,447 - Epoch: [1][ 30/ 1236] Overall Loss 1.262380 Objective Loss 1.262380 LR 0.001000 Time 0.041730 +2023-10-05 20:43:15,612 - Epoch: [1][ 40/ 1236] Overall Loss 1.268612 Objective Loss 1.268612 LR 0.001000 Time 0.035407 +2023-10-05 20:43:15,776 - Epoch: [1][ 50/ 1236] Overall Loss 1.264179 Objective Loss 1.264179 LR 0.001000 Time 0.031603 +2023-10-05 20:43:15,941 - Epoch: [1][ 60/ 1236] Overall Loss 1.263519 Objective Loss 1.263519 LR 0.001000 Time 0.029091 +2023-10-05 20:43:16,107 - Epoch: [1][ 70/ 1236] Overall Loss 1.254394 Objective Loss 1.254394 LR 0.001000 Time 0.027296 +2023-10-05 20:43:16,270 - Epoch: [1][ 80/ 1236] Overall Loss 1.256971 Objective Loss 1.256971 LR 0.001000 Time 0.025922 +2023-10-05 20:43:16,432 - Epoch: [1][ 90/ 1236] Overall Loss 1.256098 Objective Loss 1.256098 LR 0.001000 Time 0.024840 +2023-10-05 20:43:16,596 - Epoch: [1][ 100/ 1236] Overall Loss 1.255957 Objective Loss 1.255957 LR 0.001000 Time 0.023987 +2023-10-05 20:43:16,759 - Epoch: [1][ 110/ 1236] Overall Loss 1.258032 Objective Loss 1.258032 LR 0.001000 Time 0.023286 +2023-10-05 20:43:16,922 - Epoch: [1][ 120/ 1236] Overall Loss 1.256621 Objective Loss 1.256621 LR 0.001000 Time 0.022703 +2023-10-05 20:43:17,086 - Epoch: [1][ 130/ 1236] Overall Loss 1.255508 Objective Loss 1.255508 LR 0.001000 Time 0.022215 +2023-10-05 20:43:17,250 - Epoch: [1][ 140/ 1236] Overall Loss 1.254479 Objective Loss 1.254479 LR 0.001000 Time 0.021798 +2023-10-05 20:43:17,413 - Epoch: [1][ 150/ 1236] Overall Loss 1.253472 Objective Loss 1.253472 LR 0.001000 Time 0.021433 +2023-10-05 20:43:17,577 - Epoch: [1][ 160/ 1236] Overall Loss 1.250434 Objective Loss 1.250434 LR 0.001000 Time 0.021114 +2023-10-05 20:43:17,741 - Epoch: [1][ 170/ 1236] Overall Loss 1.245630 Objective Loss 1.245630 LR 0.001000 Time 0.020834 +2023-10-05 20:43:17,905 - Epoch: [1][ 180/ 1236] Overall Loss 1.240868 Objective Loss 1.240868 LR 0.001000 Time 0.020586 +2023-10-05 20:43:18,069 - Epoch: [1][ 190/ 1236] Overall Loss 1.238619 Objective Loss 1.238619 LR 0.001000 Time 0.020363 +2023-10-05 20:43:18,233 - Epoch: [1][ 200/ 1236] Overall Loss 1.237315 Objective Loss 1.237315 LR 0.001000 Time 0.020163 +2023-10-05 20:43:18,396 - Epoch: [1][ 210/ 1236] Overall Loss 1.232630 Objective Loss 1.232630 LR 0.001000 Time 0.019979 +2023-10-05 20:43:18,560 - Epoch: [1][ 220/ 1236] Overall Loss 1.230966 Objective Loss 1.230966 LR 0.001000 Time 0.019815 +2023-10-05 20:43:18,724 - Epoch: [1][ 230/ 1236] Overall Loss 1.227625 Objective Loss 1.227625 LR 0.001000 Time 0.019664 +2023-10-05 20:43:18,888 - Epoch: [1][ 240/ 1236] Overall Loss 1.227138 Objective Loss 1.227138 LR 0.001000 Time 0.019527 +2023-10-05 20:43:19,052 - Epoch: [1][ 250/ 1236] Overall Loss 1.225388 Objective Loss 1.225388 LR 0.001000 Time 0.019401 +2023-10-05 20:43:19,216 - Epoch: [1][ 260/ 1236] Overall Loss 1.223170 Objective Loss 1.223170 LR 0.001000 Time 0.019285 +2023-10-05 20:43:19,380 - Epoch: [1][ 270/ 1236] Overall Loss 1.221813 Objective Loss 1.221813 LR 0.001000 Time 0.019176 +2023-10-05 20:43:19,543 - Epoch: [1][ 280/ 1236] Overall Loss 1.218934 Objective Loss 1.218934 LR 0.001000 Time 0.019074 +2023-10-05 20:43:19,707 - Epoch: [1][ 290/ 1236] Overall Loss 1.216454 Objective Loss 1.216454 LR 0.001000 Time 0.018980 +2023-10-05 20:43:19,871 - Epoch: [1][ 300/ 1236] Overall Loss 1.211745 Objective Loss 1.211745 LR 0.001000 Time 0.018893 +2023-10-05 20:43:20,035 - Epoch: [1][ 310/ 1236] Overall Loss 1.208931 Objective Loss 1.208931 LR 0.001000 Time 0.018813 +2023-10-05 20:43:20,199 - Epoch: [1][ 320/ 1236] Overall Loss 1.206073 Objective Loss 1.206073 LR 0.001000 Time 0.018736 +2023-10-05 20:43:20,362 - Epoch: [1][ 330/ 1236] Overall Loss 1.204419 Objective Loss 1.204419 LR 0.001000 Time 0.018660 +2023-10-05 20:43:20,525 - Epoch: [1][ 340/ 1236] Overall Loss 1.202358 Objective Loss 1.202358 LR 0.001000 Time 0.018591 +2023-10-05 20:43:20,688 - Epoch: [1][ 350/ 1236] Overall Loss 1.198783 Objective Loss 1.198783 LR 0.001000 Time 0.018524 +2023-10-05 20:43:20,852 - Epoch: [1][ 360/ 1236] Overall Loss 1.197180 Objective Loss 1.197180 LR 0.001000 Time 0.018463 +2023-10-05 20:43:21,014 - Epoch: [1][ 370/ 1236] Overall Loss 1.196596 Objective Loss 1.196596 LR 0.001000 Time 0.018403 +2023-10-05 20:43:21,178 - Epoch: [1][ 380/ 1236] Overall Loss 1.193284 Objective Loss 1.193284 LR 0.001000 Time 0.018349 +2023-10-05 20:43:21,341 - Epoch: [1][ 390/ 1236] Overall Loss 1.191474 Objective Loss 1.191474 LR 0.001000 Time 0.018297 +2023-10-05 20:43:21,505 - Epoch: [1][ 400/ 1236] Overall Loss 1.189063 Objective Loss 1.189063 LR 0.001000 Time 0.018247 +2023-10-05 20:43:21,667 - Epoch: [1][ 410/ 1236] Overall Loss 1.187302 Objective Loss 1.187302 LR 0.001000 Time 0.018198 +2023-10-05 20:43:21,831 - Epoch: [1][ 420/ 1236] Overall Loss 1.186103 Objective Loss 1.186103 LR 0.001000 Time 0.018153 +2023-10-05 20:43:21,993 - Epoch: [1][ 430/ 1236] Overall Loss 1.183771 Objective Loss 1.183771 LR 0.001000 Time 0.018108 +2023-10-05 20:43:22,157 - Epoch: [1][ 440/ 1236] Overall Loss 1.181915 Objective Loss 1.181915 LR 0.001000 Time 0.018068 +2023-10-05 20:43:22,320 - Epoch: [1][ 450/ 1236] Overall Loss 1.179702 Objective Loss 1.179702 LR 0.001000 Time 0.018028 +2023-10-05 20:43:22,485 - Epoch: [1][ 460/ 1236] Overall Loss 1.178397 Objective Loss 1.178397 LR 0.001000 Time 0.017993 +2023-10-05 20:43:22,648 - Epoch: [1][ 470/ 1236] Overall Loss 1.177545 Objective Loss 1.177545 LR 0.001000 Time 0.017957 +2023-10-05 20:43:22,812 - Epoch: [1][ 480/ 1236] Overall Loss 1.176476 Objective Loss 1.176476 LR 0.001000 Time 0.017924 +2023-10-05 20:43:22,976 - Epoch: [1][ 490/ 1236] Overall Loss 1.174562 Objective Loss 1.174562 LR 0.001000 Time 0.017892 +2023-10-05 20:43:23,140 - Epoch: [1][ 500/ 1236] Overall Loss 1.174162 Objective Loss 1.174162 LR 0.001000 Time 0.017861 +2023-10-05 20:43:23,303 - Epoch: [1][ 510/ 1236] Overall Loss 1.173772 Objective Loss 1.173772 LR 0.001000 Time 0.017831 +2023-10-05 20:43:23,467 - Epoch: [1][ 520/ 1236] Overall Loss 1.172661 Objective Loss 1.172661 LR 0.001000 Time 0.017803 +2023-10-05 20:43:23,631 - Epoch: [1][ 530/ 1236] Overall Loss 1.171815 Objective Loss 1.171815 LR 0.001000 Time 0.017775 +2023-10-05 20:43:23,795 - Epoch: [1][ 540/ 1236] Overall Loss 1.169831 Objective Loss 1.169831 LR 0.001000 Time 0.017749 +2023-10-05 20:43:23,959 - Epoch: [1][ 550/ 1236] Overall Loss 1.168186 Objective Loss 1.168186 LR 0.001000 Time 0.017724 +2023-10-05 20:43:24,123 - Epoch: [1][ 560/ 1236] Overall Loss 1.165801 Objective Loss 1.165801 LR 0.001000 Time 0.017700 +2023-10-05 20:43:24,287 - Epoch: [1][ 570/ 1236] Overall Loss 1.163164 Objective Loss 1.163164 LR 0.001000 Time 0.017677 +2023-10-05 20:43:24,451 - Epoch: [1][ 580/ 1236] Overall Loss 1.160795 Objective Loss 1.160795 LR 0.001000 Time 0.017654 +2023-10-05 20:43:24,615 - Epoch: [1][ 590/ 1236] Overall Loss 1.158260 Objective Loss 1.158260 LR 0.001000 Time 0.017632 +2023-10-05 20:43:24,780 - Epoch: [1][ 600/ 1236] Overall Loss 1.157482 Objective Loss 1.157482 LR 0.001000 Time 0.017613 +2023-10-05 20:43:24,944 - Epoch: [1][ 610/ 1236] Overall Loss 1.156290 Objective Loss 1.156290 LR 0.001000 Time 0.017593 +2023-10-05 20:43:25,108 - Epoch: [1][ 620/ 1236] Overall Loss 1.154499 Objective Loss 1.154499 LR 0.001000 Time 0.017573 +2023-10-05 20:43:25,272 - Epoch: [1][ 630/ 1236] Overall Loss 1.152716 Objective Loss 1.152716 LR 0.001000 Time 0.017554 +2023-10-05 20:43:25,436 - Epoch: [1][ 640/ 1236] Overall Loss 1.151071 Objective Loss 1.151071 LR 0.001000 Time 0.017536 +2023-10-05 20:43:25,601 - Epoch: [1][ 650/ 1236] Overall Loss 1.149357 Objective Loss 1.149357 LR 0.001000 Time 0.017519 +2023-10-05 20:43:25,765 - Epoch: [1][ 660/ 1236] Overall Loss 1.147554 Objective Loss 1.147554 LR 0.001000 Time 0.017502 +2023-10-05 20:43:25,929 - Epoch: [1][ 670/ 1236] Overall Loss 1.145441 Objective Loss 1.145441 LR 0.001000 Time 0.017485 +2023-10-05 20:43:26,094 - Epoch: [1][ 680/ 1236] Overall Loss 1.144184 Objective Loss 1.144184 LR 0.001000 Time 0.017470 +2023-10-05 20:43:26,258 - Epoch: [1][ 690/ 1236] Overall Loss 1.141963 Objective Loss 1.141963 LR 0.001000 Time 0.017454 +2023-10-05 20:43:26,422 - Epoch: [1][ 700/ 1236] Overall Loss 1.139810 Objective Loss 1.139810 LR 0.001000 Time 0.017439 +2023-10-05 20:43:26,587 - Epoch: [1][ 710/ 1236] Overall Loss 1.138664 Objective Loss 1.138664 LR 0.001000 Time 0.017424 +2023-10-05 20:43:26,751 - Epoch: [1][ 720/ 1236] Overall Loss 1.137887 Objective Loss 1.137887 LR 0.001000 Time 0.017410 +2023-10-05 20:43:26,915 - Epoch: [1][ 730/ 1236] Overall Loss 1.136205 Objective Loss 1.136205 LR 0.001000 Time 0.017396 +2023-10-05 20:43:27,079 - Epoch: [1][ 740/ 1236] Overall Loss 1.134571 Objective Loss 1.134571 LR 0.001000 Time 0.017382 +2023-10-05 20:43:27,244 - Epoch: [1][ 750/ 1236] Overall Loss 1.132918 Objective Loss 1.132918 LR 0.001000 Time 0.017370 +2023-10-05 20:43:27,408 - Epoch: [1][ 760/ 1236] Overall Loss 1.131168 Objective Loss 1.131168 LR 0.001000 Time 0.017357 +2023-10-05 20:43:27,573 - Epoch: [1][ 770/ 1236] Overall Loss 1.129161 Objective Loss 1.129161 LR 0.001000 Time 0.017345 +2023-10-05 20:43:27,737 - Epoch: [1][ 780/ 1236] Overall Loss 1.127868 Objective Loss 1.127868 LR 0.001000 Time 0.017333 +2023-10-05 20:43:27,902 - Epoch: [1][ 790/ 1236] Overall Loss 1.126835 Objective Loss 1.126835 LR 0.001000 Time 0.017321 +2023-10-05 20:43:28,066 - Epoch: [1][ 800/ 1236] Overall Loss 1.124755 Objective Loss 1.124755 LR 0.001000 Time 0.017310 +2023-10-05 20:43:28,231 - Epoch: [1][ 810/ 1236] Overall Loss 1.122866 Objective Loss 1.122866 LR 0.001000 Time 0.017299 +2023-10-05 20:43:28,395 - Epoch: [1][ 820/ 1236] Overall Loss 1.121423 Objective Loss 1.121423 LR 0.001000 Time 0.017289 +2023-10-05 20:43:28,560 - Epoch: [1][ 830/ 1236] Overall Loss 1.120500 Objective Loss 1.120500 LR 0.001000 Time 0.017278 +2023-10-05 20:43:28,724 - Epoch: [1][ 840/ 1236] Overall Loss 1.119002 Objective Loss 1.119002 LR 0.001000 Time 0.017268 +2023-10-05 20:43:28,889 - Epoch: [1][ 850/ 1236] Overall Loss 1.117731 Objective Loss 1.117731 LR 0.001000 Time 0.017258 +2023-10-05 20:43:29,053 - Epoch: [1][ 860/ 1236] Overall Loss 1.116187 Objective Loss 1.116187 LR 0.001000 Time 0.017248 +2023-10-05 20:43:29,218 - Epoch: [1][ 870/ 1236] Overall Loss 1.114637 Objective Loss 1.114637 LR 0.001000 Time 0.017239 +2023-10-05 20:43:29,382 - Epoch: [1][ 880/ 1236] Overall Loss 1.112833 Objective Loss 1.112833 LR 0.001000 Time 0.017229 +2023-10-05 20:43:29,547 - Epoch: [1][ 890/ 1236] Overall Loss 1.111425 Objective Loss 1.111425 LR 0.001000 Time 0.017220 +2023-10-05 20:43:29,711 - Epoch: [1][ 900/ 1236] Overall Loss 1.109503 Objective Loss 1.109503 LR 0.001000 Time 0.017211 +2023-10-05 20:43:29,875 - Epoch: [1][ 910/ 1236] Overall Loss 1.107631 Objective Loss 1.107631 LR 0.001000 Time 0.017202 +2023-10-05 20:43:30,040 - Epoch: [1][ 920/ 1236] Overall Loss 1.106378 Objective Loss 1.106378 LR 0.001000 Time 0.017194 +2023-10-05 20:43:30,204 - Epoch: [1][ 930/ 1236] Overall Loss 1.105217 Objective Loss 1.105217 LR 0.001000 Time 0.017185 +2023-10-05 20:43:30,369 - Epoch: [1][ 940/ 1236] Overall Loss 1.104197 Objective Loss 1.104197 LR 0.001000 Time 0.017177 +2023-10-05 20:43:30,533 - Epoch: [1][ 950/ 1236] Overall Loss 1.102753 Objective Loss 1.102753 LR 0.001000 Time 0.017169 +2023-10-05 20:43:30,698 - Epoch: [1][ 960/ 1236] Overall Loss 1.101630 Objective Loss 1.101630 LR 0.001000 Time 0.017162 +2023-10-05 20:43:30,863 - Epoch: [1][ 970/ 1236] Overall Loss 1.099652 Objective Loss 1.099652 LR 0.001000 Time 0.017154 +2023-10-05 20:43:31,028 - Epoch: [1][ 980/ 1236] Overall Loss 1.098277 Objective Loss 1.098277 LR 0.001000 Time 0.017148 +2023-10-05 20:43:31,193 - Epoch: [1][ 990/ 1236] Overall Loss 1.096332 Objective Loss 1.096332 LR 0.001000 Time 0.017141 +2023-10-05 20:43:31,358 - Epoch: [1][ 1000/ 1236] Overall Loss 1.095617 Objective Loss 1.095617 LR 0.001000 Time 0.017134 +2023-10-05 20:43:31,523 - Epoch: [1][ 1010/ 1236] Overall Loss 1.094439 Objective Loss 1.094439 LR 0.001000 Time 0.017127 +2023-10-05 20:43:31,687 - Epoch: [1][ 1020/ 1236] Overall Loss 1.093281 Objective Loss 1.093281 LR 0.001000 Time 0.017120 +2023-10-05 20:43:31,852 - Epoch: [1][ 1030/ 1236] Overall Loss 1.092340 Objective Loss 1.092340 LR 0.001000 Time 0.017114 +2023-10-05 20:43:32,017 - Epoch: [1][ 1040/ 1236] Overall Loss 1.090923 Objective Loss 1.090923 LR 0.001000 Time 0.017108 +2023-10-05 20:43:32,182 - Epoch: [1][ 1050/ 1236] Overall Loss 1.089888 Objective Loss 1.089888 LR 0.001000 Time 0.017101 +2023-10-05 20:43:32,346 - Epoch: [1][ 1060/ 1236] Overall Loss 1.088476 Objective Loss 1.088476 LR 0.001000 Time 0.017095 +2023-10-05 20:43:32,511 - Epoch: [1][ 1070/ 1236] Overall Loss 1.087340 Objective Loss 1.087340 LR 0.001000 Time 0.017089 +2023-10-05 20:43:32,676 - Epoch: [1][ 1080/ 1236] Overall Loss 1.085874 Objective Loss 1.085874 LR 0.001000 Time 0.017083 +2023-10-05 20:43:32,841 - Epoch: [1][ 1090/ 1236] Overall Loss 1.084385 Objective Loss 1.084385 LR 0.001000 Time 0.017077 +2023-10-05 20:43:33,006 - Epoch: [1][ 1100/ 1236] Overall Loss 1.083021 Objective Loss 1.083021 LR 0.001000 Time 0.017072 +2023-10-05 20:43:33,171 - Epoch: [1][ 1110/ 1236] Overall Loss 1.081771 Objective Loss 1.081771 LR 0.001000 Time 0.017066 +2023-10-05 20:43:33,336 - Epoch: [1][ 1120/ 1236] Overall Loss 1.080015 Objective Loss 1.080015 LR 0.001000 Time 0.017061 +2023-10-05 20:43:33,501 - Epoch: [1][ 1130/ 1236] Overall Loss 1.079057 Objective Loss 1.079057 LR 0.001000 Time 0.017056 +2023-10-05 20:43:33,666 - Epoch: [1][ 1140/ 1236] Overall Loss 1.078326 Objective Loss 1.078326 LR 0.001000 Time 0.017051 +2023-10-05 20:43:33,831 - Epoch: [1][ 1150/ 1236] Overall Loss 1.076835 Objective Loss 1.076835 LR 0.001000 Time 0.017046 +2023-10-05 20:43:33,996 - Epoch: [1][ 1160/ 1236] Overall Loss 1.075427 Objective Loss 1.075427 LR 0.001000 Time 0.017041 +2023-10-05 20:43:34,161 - Epoch: [1][ 1170/ 1236] Overall Loss 1.073884 Objective Loss 1.073884 LR 0.001000 Time 0.017036 +2023-10-05 20:43:34,326 - Epoch: [1][ 1180/ 1236] Overall Loss 1.072601 Objective Loss 1.072601 LR 0.001000 Time 0.017031 +2023-10-05 20:43:34,491 - Epoch: [1][ 1190/ 1236] Overall Loss 1.071452 Objective Loss 1.071452 LR 0.001000 Time 0.017026 +2023-10-05 20:43:34,656 - Epoch: [1][ 1200/ 1236] Overall Loss 1.070130 Objective Loss 1.070130 LR 0.001000 Time 0.017022 +2023-10-05 20:43:34,821 - Epoch: [1][ 1210/ 1236] Overall Loss 1.068736 Objective Loss 1.068736 LR 0.001000 Time 0.017017 +2023-10-05 20:43:34,986 - Epoch: [1][ 1220/ 1236] Overall Loss 1.067216 Objective Loss 1.067216 LR 0.001000 Time 0.017013 +2023-10-05 20:43:35,195 - Epoch: [1][ 1230/ 1236] Overall Loss 1.066096 Objective Loss 1.066096 LR 0.001000 Time 0.017044 +2023-10-05 20:43:35,285 - Epoch: [1][ 1236/ 1236] Overall Loss 1.065444 Objective Loss 1.065444 Top1 58.044807 Top5 89.613035 LR 0.001000 Time 0.017034 +2023-10-05 20:43:35,404 - --- validate (epoch=1)----------- +2023-10-05 20:43:35,404 - 29943 samples (256 per mini-batch) +2023-10-05 20:43:35,806 - Epoch: [1][ 10/ 117] Loss 0.842205 Top1 59.375000 Top5 89.140625 +2023-10-05 20:43:35,908 - Epoch: [1][ 20/ 117] Loss 0.836782 Top1 59.316406 Top5 89.042969 +2023-10-05 20:43:36,013 - Epoch: [1][ 30/ 117] Loss 0.838644 Top1 59.023438 Top5 89.231771 +2023-10-05 20:43:36,121 - Epoch: [1][ 40/ 117] Loss 0.854246 Top1 58.691406 Top5 89.101562 +2023-10-05 20:43:36,232 - Epoch: [1][ 50/ 117] Loss 0.858896 Top1 58.695312 Top5 89.031250 +2023-10-05 20:43:36,337 - Epoch: [1][ 60/ 117] Loss 0.859114 Top1 58.795573 Top5 89.134115 +2023-10-05 20:43:36,448 - Epoch: [1][ 70/ 117] Loss 0.855580 Top1 58.800223 Top5 89.023438 +2023-10-05 20:43:36,554 - Epoch: [1][ 80/ 117] Loss 0.866828 Top1 58.510742 Top5 88.974609 +2023-10-05 20:43:36,665 - Epoch: [1][ 90/ 117] Loss 0.868424 Top1 58.346354 Top5 88.823785 +2023-10-05 20:43:36,771 - Epoch: [1][ 100/ 117] Loss 0.867574 Top1 58.406250 Top5 88.902344 +2023-10-05 20:43:36,885 - Epoch: [1][ 110/ 117] Loss 0.868950 Top1 58.462358 Top5 88.888494 +2023-10-05 20:43:36,941 - Epoch: [1][ 117/ 117] Loss 0.866736 Top1 58.581304 Top5 88.928965 +2023-10-05 20:43:37,085 - ==> Top1: 58.581 Top5: 88.929 Loss: 0.867 + +2023-10-05 20:43:37,086 - ==> Confusion: +[[ 829 3 12 1 30 8 0 0 9 81 0 1 1 14 9 12 12 5 7 1 15] + [ 1 866 8 1 11 52 4 72 7 0 14 0 1 0 8 4 35 2 38 5 2] + [ 11 2 713 21 9 10 145 42 0 0 26 2 0 9 4 5 3 9 16 4 25] + [ 2 6 22 785 1 7 11 4 4 0 73 0 4 12 42 4 11 19 60 0 22] + [ 26 42 10 1 855 21 0 0 0 9 1 0 0 16 8 14 24 1 2 1 19] + [ 6 150 12 9 6 637 9 137 0 2 28 8 2 45 6 1 17 0 14 22 5] + [ 1 11 141 1 2 5 934 55 0 0 3 1 0 0 0 16 1 1 6 7 6] + [ 3 17 31 0 10 57 10 926 3 0 12 1 5 2 1 0 7 5 114 10 4] + [ 56 6 4 3 1 4 0 2 849 38 15 1 0 28 47 0 9 8 15 0 3] + [ 335 1 6 0 22 3 2 4 45 607 0 0 0 55 13 0 7 2 1 0 16] + [ 6 15 22 27 14 18 2 44 16 0 766 0 0 11 7 0 22 0 73 4 6] + [ 0 2 3 3 1 43 3 17 0 0 0 692 101 12 1 9 16 20 0 108 4] + [ 0 2 3 10 1 12 8 7 3 0 0 97 761 7 2 12 26 91 4 13 9] + [ 14 1 7 5 12 134 2 2 4 16 6 25 3 837 3 3 17 1 0 21 6] + [ 21 21 2 14 32 6 0 0 44 4 3 0 4 9 842 2 38 8 31 0 20] + [ 1 6 6 2 4 11 11 3 0 0 0 43 9 2 0 956 11 51 0 7 11] + [ 3 36 5 2 11 18 1 1 2 0 4 24 0 11 4 38 986 1 2 4 8] + [ 1 1 1 4 1 2 6 5 2 0 0 26 145 4 10 24 10 782 0 1 13] + [ 1 19 8 19 0 3 3 67 5 1 30 0 2 0 9 0 3 0 891 0 7] + [ 0 3 7 0 1 14 18 75 1 0 3 24 4 2 0 11 4 4 6 968 7] + [ 293 509 268 195 187 543 93 380 123 214 349 279 440 468 274 222 653 142 550 664 1059]] + +2023-10-05 20:43:37,087 - ==> Best [Top1: 58.581 Top5: 88.929 Sparsity:0.00 Params: 148928 on epoch: 1] +2023-10-05 20:43:37,087 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:43:37,100 - + +2023-10-05 20:43:37,100 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:43:38,033 - Epoch: [2][ 10/ 1236] Overall Loss 0.909773 Objective Loss 0.909773 LR 0.001000 Time 0.093255 +2023-10-05 20:43:38,198 - Epoch: [2][ 20/ 1236] Overall Loss 0.915271 Objective Loss 0.915271 LR 0.001000 Time 0.054842 +2023-10-05 20:43:38,361 - Epoch: [2][ 30/ 1236] Overall Loss 0.895023 Objective Loss 0.895023 LR 0.001000 Time 0.042005 +2023-10-05 20:43:38,524 - Epoch: [2][ 40/ 1236] Overall Loss 0.896631 Objective Loss 0.896631 LR 0.001000 Time 0.035568 +2023-10-05 20:43:38,687 - Epoch: [2][ 50/ 1236] Overall Loss 0.894038 Objective Loss 0.894038 LR 0.001000 Time 0.031712 +2023-10-05 20:43:38,850 - Epoch: [2][ 60/ 1236] Overall Loss 0.890588 Objective Loss 0.890588 LR 0.001000 Time 0.029133 +2023-10-05 20:43:39,015 - Epoch: [2][ 70/ 1236] Overall Loss 0.892190 Objective Loss 0.892190 LR 0.001000 Time 0.027322 +2023-10-05 20:43:39,179 - Epoch: [2][ 80/ 1236] Overall Loss 0.889800 Objective Loss 0.889800 LR 0.001000 Time 0.025954 +2023-10-05 20:43:39,343 - Epoch: [2][ 90/ 1236] Overall Loss 0.888088 Objective Loss 0.888088 LR 0.001000 Time 0.024887 +2023-10-05 20:43:39,506 - Epoch: [2][ 100/ 1236] Overall Loss 0.890611 Objective Loss 0.890611 LR 0.001000 Time 0.024025 +2023-10-05 20:43:39,669 - Epoch: [2][ 110/ 1236] Overall Loss 0.894247 Objective Loss 0.894247 LR 0.001000 Time 0.023317 +2023-10-05 20:43:39,832 - Epoch: [2][ 120/ 1236] Overall Loss 0.894434 Objective Loss 0.894434 LR 0.001000 Time 0.022734 +2023-10-05 20:43:39,994 - Epoch: [2][ 130/ 1236] Overall Loss 0.885914 Objective Loss 0.885914 LR 0.001000 Time 0.022232 +2023-10-05 20:43:40,157 - Epoch: [2][ 140/ 1236] Overall Loss 0.883246 Objective Loss 0.883246 LR 0.001000 Time 0.021805 +2023-10-05 20:43:40,320 - Epoch: [2][ 150/ 1236] Overall Loss 0.882310 Objective Loss 0.882310 LR 0.001000 Time 0.021437 +2023-10-05 20:43:40,483 - Epoch: [2][ 160/ 1236] Overall Loss 0.882592 Objective Loss 0.882592 LR 0.001000 Time 0.021114 +2023-10-05 20:43:40,647 - Epoch: [2][ 170/ 1236] Overall Loss 0.880383 Objective Loss 0.880383 LR 0.001000 Time 0.020831 +2023-10-05 20:43:40,809 - Epoch: [2][ 180/ 1236] Overall Loss 0.878107 Objective Loss 0.878107 LR 0.001000 Time 0.020575 +2023-10-05 20:43:40,973 - Epoch: [2][ 190/ 1236] Overall Loss 0.878155 Objective Loss 0.878155 LR 0.001000 Time 0.020351 +2023-10-05 20:43:41,136 - Epoch: [2][ 200/ 1236] Overall Loss 0.875960 Objective Loss 0.875960 LR 0.001000 Time 0.020149 +2023-10-05 20:43:41,299 - Epoch: [2][ 210/ 1236] Overall Loss 0.873474 Objective Loss 0.873474 LR 0.001000 Time 0.019962 +2023-10-05 20:43:41,462 - Epoch: [2][ 220/ 1236] Overall Loss 0.871627 Objective Loss 0.871627 LR 0.001000 Time 0.019796 +2023-10-05 20:43:41,625 - Epoch: [2][ 230/ 1236] Overall Loss 0.870045 Objective Loss 0.870045 LR 0.001000 Time 0.019641 +2023-10-05 20:43:41,788 - Epoch: [2][ 240/ 1236] Overall Loss 0.869190 Objective Loss 0.869190 LR 0.001000 Time 0.019503 +2023-10-05 20:43:41,951 - Epoch: [2][ 250/ 1236] Overall Loss 0.869315 Objective Loss 0.869315 LR 0.001000 Time 0.019375 +2023-10-05 20:43:42,115 - Epoch: [2][ 260/ 1236] Overall Loss 0.870225 Objective Loss 0.870225 LR 0.001000 Time 0.019256 +2023-10-05 20:43:42,277 - Epoch: [2][ 270/ 1236] Overall Loss 0.870964 Objective Loss 0.870964 LR 0.001000 Time 0.019145 +2023-10-05 20:43:42,441 - Epoch: [2][ 280/ 1236] Overall Loss 0.872657 Objective Loss 0.872657 LR 0.001000 Time 0.019044 +2023-10-05 20:43:42,604 - Epoch: [2][ 290/ 1236] Overall Loss 0.873168 Objective Loss 0.873168 LR 0.001000 Time 0.018948 +2023-10-05 20:43:42,768 - Epoch: [2][ 300/ 1236] Overall Loss 0.873693 Objective Loss 0.873693 LR 0.001000 Time 0.018862 +2023-10-05 20:43:42,931 - Epoch: [2][ 310/ 1236] Overall Loss 0.873844 Objective Loss 0.873844 LR 0.001000 Time 0.018780 +2023-10-05 20:43:43,095 - Epoch: [2][ 320/ 1236] Overall Loss 0.873055 Objective Loss 0.873055 LR 0.001000 Time 0.018705 +2023-10-05 20:43:43,258 - Epoch: [2][ 330/ 1236] Overall Loss 0.871727 Objective Loss 0.871727 LR 0.001000 Time 0.018631 +2023-10-05 20:43:43,422 - Epoch: [2][ 340/ 1236] Overall Loss 0.869611 Objective Loss 0.869611 LR 0.001000 Time 0.018563 +2023-10-05 20:43:43,585 - Epoch: [2][ 350/ 1236] Overall Loss 0.867278 Objective Loss 0.867278 LR 0.001000 Time 0.018498 +2023-10-05 20:43:43,748 - Epoch: [2][ 360/ 1236] Overall Loss 0.866408 Objective Loss 0.866408 LR 0.001000 Time 0.018437 +2023-10-05 20:43:43,912 - Epoch: [2][ 370/ 1236] Overall Loss 0.866494 Objective Loss 0.866494 LR 0.001000 Time 0.018381 +2023-10-05 20:43:44,076 - Epoch: [2][ 380/ 1236] Overall Loss 0.866838 Objective Loss 0.866838 LR 0.001000 Time 0.018327 +2023-10-05 20:43:44,240 - Epoch: [2][ 390/ 1236] Overall Loss 0.866435 Objective Loss 0.866435 LR 0.001000 Time 0.018276 +2023-10-05 20:43:44,403 - Epoch: [2][ 400/ 1236] Overall Loss 0.865564 Objective Loss 0.865564 LR 0.001000 Time 0.018227 +2023-10-05 20:43:44,567 - Epoch: [2][ 410/ 1236] Overall Loss 0.866730 Objective Loss 0.866730 LR 0.001000 Time 0.018181 +2023-10-05 20:43:44,731 - Epoch: [2][ 420/ 1236] Overall Loss 0.867084 Objective Loss 0.867084 LR 0.001000 Time 0.018138 +2023-10-05 20:43:44,894 - Epoch: [2][ 430/ 1236] Overall Loss 0.866165 Objective Loss 0.866165 LR 0.001000 Time 0.018095 +2023-10-05 20:43:45,058 - Epoch: [2][ 440/ 1236] Overall Loss 0.866226 Objective Loss 0.866226 LR 0.001000 Time 0.018055 +2023-10-05 20:43:45,221 - Epoch: [2][ 450/ 1236] Overall Loss 0.864983 Objective Loss 0.864983 LR 0.001000 Time 0.018017 +2023-10-05 20:43:45,385 - Epoch: [2][ 460/ 1236] Overall Loss 0.863879 Objective Loss 0.863879 LR 0.001000 Time 0.017981 +2023-10-05 20:43:45,549 - Epoch: [2][ 470/ 1236] Overall Loss 0.862143 Objective Loss 0.862143 LR 0.001000 Time 0.017947 +2023-10-05 20:43:45,713 - Epoch: [2][ 480/ 1236] Overall Loss 0.861751 Objective Loss 0.861751 LR 0.001000 Time 0.017913 +2023-10-05 20:43:45,877 - Epoch: [2][ 490/ 1236] Overall Loss 0.861226 Objective Loss 0.861226 LR 0.001000 Time 0.017881 +2023-10-05 20:43:46,040 - Epoch: [2][ 500/ 1236] Overall Loss 0.861506 Objective Loss 0.861506 LR 0.001000 Time 0.017850 +2023-10-05 20:43:46,204 - Epoch: [2][ 510/ 1236] Overall Loss 0.860458 Objective Loss 0.860458 LR 0.001000 Time 0.017821 +2023-10-05 20:43:46,368 - Epoch: [2][ 520/ 1236] Overall Loss 0.859060 Objective Loss 0.859060 LR 0.001000 Time 0.017793 +2023-10-05 20:43:46,532 - Epoch: [2][ 530/ 1236] Overall Loss 0.858550 Objective Loss 0.858550 LR 0.001000 Time 0.017766 +2023-10-05 20:43:46,696 - Epoch: [2][ 540/ 1236] Overall Loss 0.859200 Objective Loss 0.859200 LR 0.001000 Time 0.017740 +2023-10-05 20:43:46,859 - Epoch: [2][ 550/ 1236] Overall Loss 0.859014 Objective Loss 0.859014 LR 0.001000 Time 0.017714 +2023-10-05 20:43:47,024 - Epoch: [2][ 560/ 1236] Overall Loss 0.858591 Objective Loss 0.858591 LR 0.001000 Time 0.017690 +2023-10-05 20:43:47,187 - Epoch: [2][ 570/ 1236] Overall Loss 0.858545 Objective Loss 0.858545 LR 0.001000 Time 0.017666 +2023-10-05 20:43:47,351 - Epoch: [2][ 580/ 1236] Overall Loss 0.858316 Objective Loss 0.858316 LR 0.001000 Time 0.017644 +2023-10-05 20:43:47,515 - Epoch: [2][ 590/ 1236] Overall Loss 0.857483 Objective Loss 0.857483 LR 0.001000 Time 0.017622 +2023-10-05 20:43:47,679 - Epoch: [2][ 600/ 1236] Overall Loss 0.857718 Objective Loss 0.857718 LR 0.001000 Time 0.017601 +2023-10-05 20:43:47,843 - Epoch: [2][ 610/ 1236] Overall Loss 0.857123 Objective Loss 0.857123 LR 0.001000 Time 0.017580 +2023-10-05 20:43:48,006 - Epoch: [2][ 620/ 1236] Overall Loss 0.856304 Objective Loss 0.856304 LR 0.001000 Time 0.017560 +2023-10-05 20:43:48,171 - Epoch: [2][ 630/ 1236] Overall Loss 0.856258 Objective Loss 0.856258 LR 0.001000 Time 0.017542 +2023-10-05 20:43:48,335 - Epoch: [2][ 640/ 1236] Overall Loss 0.855553 Objective Loss 0.855553 LR 0.001000 Time 0.017524 +2023-10-05 20:43:48,499 - Epoch: [2][ 650/ 1236] Overall Loss 0.855070 Objective Loss 0.855070 LR 0.001000 Time 0.017506 +2023-10-05 20:43:48,663 - Epoch: [2][ 660/ 1236] Overall Loss 0.854038 Objective Loss 0.854038 LR 0.001000 Time 0.017489 +2023-10-05 20:43:48,827 - Epoch: [2][ 670/ 1236] Overall Loss 0.853841 Objective Loss 0.853841 LR 0.001000 Time 0.017472 +2023-10-05 20:43:48,991 - Epoch: [2][ 680/ 1236] Overall Loss 0.853507 Objective Loss 0.853507 LR 0.001000 Time 0.017456 +2023-10-05 20:43:49,155 - Epoch: [2][ 690/ 1236] Overall Loss 0.852675 Objective Loss 0.852675 LR 0.001000 Time 0.017441 +2023-10-05 20:43:49,319 - Epoch: [2][ 700/ 1236] Overall Loss 0.852128 Objective Loss 0.852128 LR 0.001000 Time 0.017426 +2023-10-05 20:43:49,483 - Epoch: [2][ 710/ 1236] Overall Loss 0.852421 Objective Loss 0.852421 LR 0.001000 Time 0.017411 +2023-10-05 20:43:49,648 - Epoch: [2][ 720/ 1236] Overall Loss 0.851874 Objective Loss 0.851874 LR 0.001000 Time 0.017397 +2023-10-05 20:43:49,812 - Epoch: [2][ 730/ 1236] Overall Loss 0.850485 Objective Loss 0.850485 LR 0.001000 Time 0.017382 +2023-10-05 20:43:49,976 - Epoch: [2][ 740/ 1236] Overall Loss 0.849461 Objective Loss 0.849461 LR 0.001000 Time 0.017369 +2023-10-05 20:43:50,140 - Epoch: [2][ 750/ 1236] Overall Loss 0.848503 Objective Loss 0.848503 LR 0.001000 Time 0.017356 +2023-10-05 20:43:50,303 - Epoch: [2][ 760/ 1236] Overall Loss 0.847647 Objective Loss 0.847647 LR 0.001000 Time 0.017342 +2023-10-05 20:43:50,467 - Epoch: [2][ 770/ 1236] Overall Loss 0.847026 Objective Loss 0.847026 LR 0.001000 Time 0.017330 +2023-10-05 20:43:50,632 - Epoch: [2][ 780/ 1236] Overall Loss 0.846105 Objective Loss 0.846105 LR 0.001000 Time 0.017318 +2023-10-05 20:43:50,796 - Epoch: [2][ 790/ 1236] Overall Loss 0.845890 Objective Loss 0.845890 LR 0.001000 Time 0.017306 +2023-10-05 20:43:50,960 - Epoch: [2][ 800/ 1236] Overall Loss 0.845350 Objective Loss 0.845350 LR 0.001000 Time 0.017294 +2023-10-05 20:43:51,124 - Epoch: [2][ 810/ 1236] Overall Loss 0.844518 Objective Loss 0.844518 LR 0.001000 Time 0.017283 +2023-10-05 20:43:51,288 - Epoch: [2][ 820/ 1236] Overall Loss 0.844249 Objective Loss 0.844249 LR 0.001000 Time 0.017272 +2023-10-05 20:43:51,452 - Epoch: [2][ 830/ 1236] Overall Loss 0.843672 Objective Loss 0.843672 LR 0.001000 Time 0.017261 +2023-10-05 20:43:51,616 - Epoch: [2][ 840/ 1236] Overall Loss 0.843675 Objective Loss 0.843675 LR 0.001000 Time 0.017250 +2023-10-05 20:43:51,780 - Epoch: [2][ 850/ 1236] Overall Loss 0.843074 Objective Loss 0.843074 LR 0.001000 Time 0.017239 +2023-10-05 20:43:51,944 - Epoch: [2][ 860/ 1236] Overall Loss 0.842456 Objective Loss 0.842456 LR 0.001000 Time 0.017230 +2023-10-05 20:43:52,108 - Epoch: [2][ 870/ 1236] Overall Loss 0.842414 Objective Loss 0.842414 LR 0.001000 Time 0.017220 +2023-10-05 20:43:52,272 - Epoch: [2][ 880/ 1236] Overall Loss 0.842434 Objective Loss 0.842434 LR 0.001000 Time 0.017210 +2023-10-05 20:43:52,436 - Epoch: [2][ 890/ 1236] Overall Loss 0.841914 Objective Loss 0.841914 LR 0.001000 Time 0.017200 +2023-10-05 20:43:52,600 - Epoch: [2][ 900/ 1236] Overall Loss 0.841130 Objective Loss 0.841130 LR 0.001000 Time 0.017191 +2023-10-05 20:43:52,763 - Epoch: [2][ 910/ 1236] Overall Loss 0.840944 Objective Loss 0.840944 LR 0.001000 Time 0.017182 +2023-10-05 20:43:52,927 - Epoch: [2][ 920/ 1236] Overall Loss 0.840375 Objective Loss 0.840375 LR 0.001000 Time 0.017173 +2023-10-05 20:43:53,091 - Epoch: [2][ 930/ 1236] Overall Loss 0.840254 Objective Loss 0.840254 LR 0.001000 Time 0.017164 +2023-10-05 20:43:53,255 - Epoch: [2][ 940/ 1236] Overall Loss 0.840015 Objective Loss 0.840015 LR 0.001000 Time 0.017156 +2023-10-05 20:43:53,419 - Epoch: [2][ 950/ 1236] Overall Loss 0.839551 Objective Loss 0.839551 LR 0.001000 Time 0.017147 +2023-10-05 20:43:53,583 - Epoch: [2][ 960/ 1236] Overall Loss 0.839158 Objective Loss 0.839158 LR 0.001000 Time 0.017139 +2023-10-05 20:43:53,747 - Epoch: [2][ 970/ 1236] Overall Loss 0.838453 Objective Loss 0.838453 LR 0.001000 Time 0.017131 +2023-10-05 20:43:53,911 - Epoch: [2][ 980/ 1236] Overall Loss 0.837295 Objective Loss 0.837295 LR 0.001000 Time 0.017123 +2023-10-05 20:43:54,075 - Epoch: [2][ 990/ 1236] Overall Loss 0.836907 Objective Loss 0.836907 LR 0.001000 Time 0.017116 +2023-10-05 20:43:54,239 - Epoch: [2][ 1000/ 1236] Overall Loss 0.836437 Objective Loss 0.836437 LR 0.001000 Time 0.017109 +2023-10-05 20:43:54,403 - Epoch: [2][ 1010/ 1236] Overall Loss 0.835463 Objective Loss 0.835463 LR 0.001000 Time 0.017101 +2023-10-05 20:43:54,567 - Epoch: [2][ 1020/ 1236] Overall Loss 0.834868 Objective Loss 0.834868 LR 0.001000 Time 0.017094 +2023-10-05 20:43:54,731 - Epoch: [2][ 1030/ 1236] Overall Loss 0.834378 Objective Loss 0.834378 LR 0.001000 Time 0.017087 +2023-10-05 20:43:54,895 - Epoch: [2][ 1040/ 1236] Overall Loss 0.833411 Objective Loss 0.833411 LR 0.001000 Time 0.017080 +2023-10-05 20:43:55,059 - Epoch: [2][ 1050/ 1236] Overall Loss 0.832497 Objective Loss 0.832497 LR 0.001000 Time 0.017073 +2023-10-05 20:43:55,223 - Epoch: [2][ 1060/ 1236] Overall Loss 0.831325 Objective Loss 0.831325 LR 0.001000 Time 0.017067 +2023-10-05 20:43:55,387 - Epoch: [2][ 1070/ 1236] Overall Loss 0.830760 Objective Loss 0.830760 LR 0.001000 Time 0.017060 +2023-10-05 20:43:55,552 - Epoch: [2][ 1080/ 1236] Overall Loss 0.830025 Objective Loss 0.830025 LR 0.001000 Time 0.017055 +2023-10-05 20:43:55,716 - Epoch: [2][ 1090/ 1236] Overall Loss 0.829832 Objective Loss 0.829832 LR 0.001000 Time 0.017048 +2023-10-05 20:43:55,880 - Epoch: [2][ 1100/ 1236] Overall Loss 0.829383 Objective Loss 0.829383 LR 0.001000 Time 0.017042 +2023-10-05 20:43:56,044 - Epoch: [2][ 1110/ 1236] Overall Loss 0.828624 Objective Loss 0.828624 LR 0.001000 Time 0.017036 +2023-10-05 20:43:56,208 - Epoch: [2][ 1120/ 1236] Overall Loss 0.827799 Objective Loss 0.827799 LR 0.001000 Time 0.017030 +2023-10-05 20:43:56,372 - Epoch: [2][ 1130/ 1236] Overall Loss 0.827024 Objective Loss 0.827024 LR 0.001000 Time 0.017024 +2023-10-05 20:43:56,536 - Epoch: [2][ 1140/ 1236] Overall Loss 0.826808 Objective Loss 0.826808 LR 0.001000 Time 0.017018 +2023-10-05 20:43:56,700 - Epoch: [2][ 1150/ 1236] Overall Loss 0.826198 Objective Loss 0.826198 LR 0.001000 Time 0.017013 +2023-10-05 20:43:56,864 - Epoch: [2][ 1160/ 1236] Overall Loss 0.825655 Objective Loss 0.825655 LR 0.001000 Time 0.017007 +2023-10-05 20:43:57,028 - Epoch: [2][ 1170/ 1236] Overall Loss 0.825022 Objective Loss 0.825022 LR 0.001000 Time 0.017002 +2023-10-05 20:43:57,192 - Epoch: [2][ 1180/ 1236] Overall Loss 0.824559 Objective Loss 0.824559 LR 0.001000 Time 0.016997 +2023-10-05 20:43:57,356 - Epoch: [2][ 1190/ 1236] Overall Loss 0.824389 Objective Loss 0.824389 LR 0.001000 Time 0.016992 +2023-10-05 20:43:57,521 - Epoch: [2][ 1200/ 1236] Overall Loss 0.823638 Objective Loss 0.823638 LR 0.001000 Time 0.016987 +2023-10-05 20:43:57,685 - Epoch: [2][ 1210/ 1236] Overall Loss 0.823372 Objective Loss 0.823372 LR 0.001000 Time 0.016982 +2023-10-05 20:43:57,849 - Epoch: [2][ 1220/ 1236] Overall Loss 0.822778 Objective Loss 0.822778 LR 0.001000 Time 0.016977 +2023-10-05 20:43:58,057 - Epoch: [2][ 1230/ 1236] Overall Loss 0.822311 Objective Loss 0.822311 LR 0.001000 Time 0.017008 +2023-10-05 20:43:58,147 - Epoch: [2][ 1236/ 1236] Overall Loss 0.821966 Objective Loss 0.821966 Top1 69.653768 Top5 95.723014 LR 0.001000 Time 0.016998 +2023-10-05 20:43:58,272 - --- validate (epoch=2)----------- +2023-10-05 20:43:58,272 - 29943 samples (256 per mini-batch) +2023-10-05 20:43:58,670 - Epoch: [2][ 10/ 117] Loss 0.669451 Top1 66.953125 Top5 94.375000 +2023-10-05 20:43:58,773 - Epoch: [2][ 20/ 117] Loss 0.675949 Top1 66.367188 Top5 94.316406 +2023-10-05 20:43:58,880 - Epoch: [2][ 30/ 117] Loss 0.686153 Top1 65.989583 Top5 94.088542 +2023-10-05 20:43:58,988 - Epoch: [2][ 40/ 117] Loss 0.687441 Top1 65.566406 Top5 94.169922 +2023-10-05 20:43:59,094 - Epoch: [2][ 50/ 117] Loss 0.691515 Top1 65.101562 Top5 94.132812 +2023-10-05 20:43:59,201 - Epoch: [2][ 60/ 117] Loss 0.694439 Top1 65.110677 Top5 94.088542 +2023-10-05 20:43:59,303 - Epoch: [2][ 70/ 117] Loss 0.691229 Top1 65.106027 Top5 94.017857 +2023-10-05 20:43:59,406 - Epoch: [2][ 80/ 117] Loss 0.695905 Top1 65.039062 Top5 93.950195 +2023-10-05 20:43:59,507 - Epoch: [2][ 90/ 117] Loss 0.697686 Top1 64.826389 Top5 93.919271 +2023-10-05 20:43:59,610 - Epoch: [2][ 100/ 117] Loss 0.699534 Top1 64.777344 Top5 93.964844 +2023-10-05 20:43:59,719 - Epoch: [2][ 110/ 117] Loss 0.699717 Top1 64.715909 Top5 93.948864 +2023-10-05 20:43:59,775 - Epoch: [2][ 117/ 117] Loss 0.700016 Top1 64.686237 Top5 93.921785 +2023-10-05 20:43:59,910 - ==> Top1: 64.686 Top5: 93.922 Loss: 0.700 + +2023-10-05 20:43:59,911 - ==> Confusion: +[[ 829 3 8 0 29 2 0 0 7 100 1 3 0 14 17 3 3 5 3 1 22] + [ 2 980 3 1 12 33 1 34 6 0 4 1 1 2 8 2 12 3 14 6 6] + [ 10 3 765 27 12 5 100 23 1 4 14 7 3 17 5 8 0 8 9 5 30] + [ 2 6 23 815 0 4 1 2 6 0 52 1 7 13 64 2 4 27 40 0 20] + [ 18 30 7 0 889 11 1 1 3 12 0 1 0 7 17 6 35 2 0 1 9] + [ 5 145 21 14 4 720 0 25 2 3 7 17 7 73 14 1 12 0 7 22 17] + [ 2 10 94 1 3 10 987 21 0 0 3 8 0 1 0 15 0 3 7 21 5] + [ 8 30 28 0 11 104 6 821 3 1 10 12 3 2 2 0 2 3 116 47 9] + [ 35 10 0 0 0 1 0 0 841 64 13 5 4 40 62 0 2 3 4 1 4] + [ 237 0 0 0 37 0 0 2 30 718 0 2 0 49 18 0 2 1 0 3 20] + [ 7 9 20 23 6 19 5 14 13 2 833 4 4 16 17 0 9 0 39 6 7] + [ 1 3 4 0 2 19 0 0 0 0 0 820 51 20 0 5 6 44 0 55 5] + [ 1 0 1 2 0 5 1 1 0 1 0 124 717 6 6 13 13 150 4 16 7] + [ 5 3 4 1 10 28 2 0 8 21 10 11 3 986 10 1 6 2 0 2 6] + [ 20 7 0 10 16 3 0 0 29 9 0 0 6 3 954 2 8 5 10 0 19] + [ 1 5 5 1 3 3 6 0 1 0 0 55 2 3 0 952 31 41 0 4 21] + [ 4 45 3 3 10 9 1 0 1 0 2 25 7 3 10 12 1002 2 2 7 13] + [ 1 0 0 1 0 0 0 0 2 0 0 27 42 1 5 13 3 934 1 0 8] + [ 0 14 8 15 0 7 0 38 7 4 24 1 4 0 23 0 2 0 891 14 16] + [ 0 11 2 0 0 6 5 18 0 0 2 42 1 11 1 2 6 1 1 1036 7] + [ 230 561 186 227 183 324 42 122 95 215 297 317 414 588 537 137 280 204 403 664 1879]] + +2023-10-05 20:43:59,912 - ==> Best [Top1: 64.686 Top5: 93.922 Sparsity:0.00 Params: 148928 on epoch: 2] +2023-10-05 20:43:59,912 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:43:59,926 - + +2023-10-05 20:43:59,926 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:44:00,876 - Epoch: [3][ 10/ 1236] Overall Loss 0.726998 Objective Loss 0.726998 LR 0.001000 Time 0.094955 +2023-10-05 20:44:01,042 - Epoch: [3][ 20/ 1236] Overall Loss 0.724968 Objective Loss 0.724968 LR 0.001000 Time 0.055778 +2023-10-05 20:44:01,208 - Epoch: [3][ 30/ 1236] Overall Loss 0.732357 Objective Loss 0.732357 LR 0.001000 Time 0.042702 +2023-10-05 20:44:01,375 - Epoch: [3][ 40/ 1236] Overall Loss 0.728486 Objective Loss 0.728486 LR 0.001000 Time 0.036205 +2023-10-05 20:44:01,543 - Epoch: [3][ 50/ 1236] Overall Loss 0.729948 Objective Loss 0.729948 LR 0.001000 Time 0.032315 +2023-10-05 20:44:01,711 - Epoch: [3][ 60/ 1236] Overall Loss 0.728560 Objective Loss 0.728560 LR 0.001000 Time 0.029712 +2023-10-05 20:44:01,878 - Epoch: [3][ 70/ 1236] Overall Loss 0.722866 Objective Loss 0.722866 LR 0.001000 Time 0.027859 +2023-10-05 20:44:02,046 - Epoch: [3][ 80/ 1236] Overall Loss 0.722283 Objective Loss 0.722283 LR 0.001000 Time 0.026467 +2023-10-05 20:44:02,213 - Epoch: [3][ 90/ 1236] Overall Loss 0.718268 Objective Loss 0.718268 LR 0.001000 Time 0.025380 +2023-10-05 20:44:02,380 - Epoch: [3][ 100/ 1236] Overall Loss 0.721028 Objective Loss 0.721028 LR 0.001000 Time 0.024505 +2023-10-05 20:44:02,546 - Epoch: [3][ 110/ 1236] Overall Loss 0.721223 Objective Loss 0.721223 LR 0.001000 Time 0.023790 +2023-10-05 20:44:02,713 - Epoch: [3][ 120/ 1236] Overall Loss 0.722846 Objective Loss 0.722846 LR 0.001000 Time 0.023195 +2023-10-05 20:44:02,880 - Epoch: [3][ 130/ 1236] Overall Loss 0.724970 Objective Loss 0.724970 LR 0.001000 Time 0.022691 +2023-10-05 20:44:03,047 - Epoch: [3][ 140/ 1236] Overall Loss 0.724698 Objective Loss 0.724698 LR 0.001000 Time 0.022260 +2023-10-05 20:44:03,213 - Epoch: [3][ 150/ 1236] Overall Loss 0.726539 Objective Loss 0.726539 LR 0.001000 Time 0.021886 +2023-10-05 20:44:03,380 - Epoch: [3][ 160/ 1236] Overall Loss 0.728349 Objective Loss 0.728349 LR 0.001000 Time 0.021559 +2023-10-05 20:44:03,546 - Epoch: [3][ 170/ 1236] Overall Loss 0.731017 Objective Loss 0.731017 LR 0.001000 Time 0.021265 +2023-10-05 20:44:03,713 - Epoch: [3][ 180/ 1236] Overall Loss 0.727862 Objective Loss 0.727862 LR 0.001000 Time 0.021010 +2023-10-05 20:44:03,879 - Epoch: [3][ 190/ 1236] Overall Loss 0.725271 Objective Loss 0.725271 LR 0.001000 Time 0.020776 +2023-10-05 20:44:04,046 - Epoch: [3][ 200/ 1236] Overall Loss 0.725504 Objective Loss 0.725504 LR 0.001000 Time 0.020568 +2023-10-05 20:44:04,212 - Epoch: [3][ 210/ 1236] Overall Loss 0.723165 Objective Loss 0.723165 LR 0.001000 Time 0.020379 +2023-10-05 20:44:04,378 - Epoch: [3][ 220/ 1236] Overall Loss 0.722939 Objective Loss 0.722939 LR 0.001000 Time 0.020207 +2023-10-05 20:44:04,544 - Epoch: [3][ 230/ 1236] Overall Loss 0.724953 Objective Loss 0.724953 LR 0.001000 Time 0.020049 +2023-10-05 20:44:04,710 - Epoch: [3][ 240/ 1236] Overall Loss 0.723071 Objective Loss 0.723071 LR 0.001000 Time 0.019905 +2023-10-05 20:44:04,877 - Epoch: [3][ 250/ 1236] Overall Loss 0.723288 Objective Loss 0.723288 LR 0.001000 Time 0.019775 +2023-10-05 20:44:05,043 - Epoch: [3][ 260/ 1236] Overall Loss 0.723509 Objective Loss 0.723509 LR 0.001000 Time 0.019652 +2023-10-05 20:44:05,209 - Epoch: [3][ 270/ 1236] Overall Loss 0.720935 Objective Loss 0.720935 LR 0.001000 Time 0.019537 +2023-10-05 20:44:05,375 - Epoch: [3][ 280/ 1236] Overall Loss 0.720134 Objective Loss 0.720134 LR 0.001000 Time 0.019432 +2023-10-05 20:44:05,541 - Epoch: [3][ 290/ 1236] Overall Loss 0.719149 Objective Loss 0.719149 LR 0.001000 Time 0.019332 +2023-10-05 20:44:05,708 - Epoch: [3][ 300/ 1236] Overall Loss 0.718417 Objective Loss 0.718417 LR 0.001000 Time 0.019242 +2023-10-05 20:44:05,874 - Epoch: [3][ 310/ 1236] Overall Loss 0.716961 Objective Loss 0.716961 LR 0.001000 Time 0.019157 +2023-10-05 20:44:06,042 - Epoch: [3][ 320/ 1236] Overall Loss 0.716611 Objective Loss 0.716611 LR 0.001000 Time 0.019082 +2023-10-05 20:44:06,208 - Epoch: [3][ 330/ 1236] Overall Loss 0.715821 Objective Loss 0.715821 LR 0.001000 Time 0.019007 +2023-10-05 20:44:06,377 - Epoch: [3][ 340/ 1236] Overall Loss 0.715747 Objective Loss 0.715747 LR 0.001000 Time 0.018943 +2023-10-05 20:44:06,544 - Epoch: [3][ 350/ 1236] Overall Loss 0.715968 Objective Loss 0.715968 LR 0.001000 Time 0.018877 +2023-10-05 20:44:06,713 - Epoch: [3][ 360/ 1236] Overall Loss 0.717256 Objective Loss 0.717256 LR 0.001000 Time 0.018822 +2023-10-05 20:44:06,880 - Epoch: [3][ 370/ 1236] Overall Loss 0.717813 Objective Loss 0.717813 LR 0.001000 Time 0.018764 +2023-10-05 20:44:07,049 - Epoch: [3][ 380/ 1236] Overall Loss 0.717179 Objective Loss 0.717179 LR 0.001000 Time 0.018715 +2023-10-05 20:44:07,216 - Epoch: [3][ 390/ 1236] Overall Loss 0.716999 Objective Loss 0.716999 LR 0.001000 Time 0.018664 +2023-10-05 20:44:07,386 - Epoch: [3][ 400/ 1236] Overall Loss 0.717014 Objective Loss 0.717014 LR 0.001000 Time 0.018620 +2023-10-05 20:44:07,553 - Epoch: [3][ 410/ 1236] Overall Loss 0.715754 Objective Loss 0.715754 LR 0.001000 Time 0.018572 +2023-10-05 20:44:07,722 - Epoch: [3][ 420/ 1236] Overall Loss 0.714973 Objective Loss 0.714973 LR 0.001000 Time 0.018532 +2023-10-05 20:44:07,889 - Epoch: [3][ 430/ 1236] Overall Loss 0.715304 Objective Loss 0.715304 LR 0.001000 Time 0.018488 +2023-10-05 20:44:08,057 - Epoch: [3][ 440/ 1236] Overall Loss 0.715997 Objective Loss 0.715997 LR 0.001000 Time 0.018450 +2023-10-05 20:44:08,224 - Epoch: [3][ 450/ 1236] Overall Loss 0.716330 Objective Loss 0.716330 LR 0.001000 Time 0.018410 +2023-10-05 20:44:08,395 - Epoch: [3][ 460/ 1236] Overall Loss 0.715251 Objective Loss 0.715251 LR 0.001000 Time 0.018380 +2023-10-05 20:44:08,568 - Epoch: [3][ 470/ 1236] Overall Loss 0.715673 Objective Loss 0.715673 LR 0.001000 Time 0.018357 +2023-10-05 20:44:08,738 - Epoch: [3][ 480/ 1236] Overall Loss 0.715590 Objective Loss 0.715590 LR 0.001000 Time 0.018329 +2023-10-05 20:44:08,911 - Epoch: [3][ 490/ 1236] Overall Loss 0.714795 Objective Loss 0.714795 LR 0.001000 Time 0.018308 +2023-10-05 20:44:09,082 - Epoch: [3][ 500/ 1236] Overall Loss 0.715228 Objective Loss 0.715228 LR 0.001000 Time 0.018283 +2023-10-05 20:44:09,256 - Epoch: [3][ 510/ 1236] Overall Loss 0.716035 Objective Loss 0.716035 LR 0.001000 Time 0.018263 +2023-10-05 20:44:09,426 - Epoch: [3][ 520/ 1236] Overall Loss 0.714978 Objective Loss 0.714978 LR 0.001000 Time 0.018240 +2023-10-05 20:44:09,599 - Epoch: [3][ 530/ 1236] Overall Loss 0.713495 Objective Loss 0.713495 LR 0.001000 Time 0.018221 +2023-10-05 20:44:09,770 - Epoch: [3][ 540/ 1236] Overall Loss 0.712783 Objective Loss 0.712783 LR 0.001000 Time 0.018200 +2023-10-05 20:44:09,944 - Epoch: [3][ 550/ 1236] Overall Loss 0.712342 Objective Loss 0.712342 LR 0.001000 Time 0.018184 +2023-10-05 20:44:10,115 - Epoch: [3][ 560/ 1236] Overall Loss 0.712156 Objective Loss 0.712156 LR 0.001000 Time 0.018165 +2023-10-05 20:44:10,288 - Epoch: [3][ 570/ 1236] Overall Loss 0.712210 Objective Loss 0.712210 LR 0.001000 Time 0.018148 +2023-10-05 20:44:10,459 - Epoch: [3][ 580/ 1236] Overall Loss 0.712786 Objective Loss 0.712786 LR 0.001000 Time 0.018131 +2023-10-05 20:44:10,633 - Epoch: [3][ 590/ 1236] Overall Loss 0.712266 Objective Loss 0.712266 LR 0.001000 Time 0.018117 +2023-10-05 20:44:10,804 - Epoch: [3][ 600/ 1236] Overall Loss 0.712394 Objective Loss 0.712394 LR 0.001000 Time 0.018100 +2023-10-05 20:44:10,977 - Epoch: [3][ 610/ 1236] Overall Loss 0.712005 Objective Loss 0.712005 LR 0.001000 Time 0.018087 +2023-10-05 20:44:11,148 - Epoch: [3][ 620/ 1236] Overall Loss 0.710229 Objective Loss 0.710229 LR 0.001000 Time 0.018070 +2023-10-05 20:44:11,322 - Epoch: [3][ 630/ 1236] Overall Loss 0.710182 Objective Loss 0.710182 LR 0.001000 Time 0.018058 +2023-10-05 20:44:11,494 - Epoch: [3][ 640/ 1236] Overall Loss 0.709041 Objective Loss 0.709041 LR 0.001000 Time 0.018045 +2023-10-05 20:44:11,665 - Epoch: [3][ 650/ 1236] Overall Loss 0.709733 Objective Loss 0.709733 LR 0.001000 Time 0.018030 +2023-10-05 20:44:11,834 - Epoch: [3][ 660/ 1236] Overall Loss 0.709219 Objective Loss 0.709219 LR 0.001000 Time 0.018012 +2023-10-05 20:44:12,005 - Epoch: [3][ 670/ 1236] Overall Loss 0.709155 Objective Loss 0.709155 LR 0.001000 Time 0.017998 +2023-10-05 20:44:12,174 - Epoch: [3][ 680/ 1236] Overall Loss 0.708499 Objective Loss 0.708499 LR 0.001000 Time 0.017982 +2023-10-05 20:44:12,345 - Epoch: [3][ 690/ 1236] Overall Loss 0.708658 Objective Loss 0.708658 LR 0.001000 Time 0.017968 +2023-10-05 20:44:12,514 - Epoch: [3][ 700/ 1236] Overall Loss 0.708527 Objective Loss 0.708527 LR 0.001000 Time 0.017953 +2023-10-05 20:44:12,685 - Epoch: [3][ 710/ 1236] Overall Loss 0.708956 Objective Loss 0.708956 LR 0.001000 Time 0.017940 +2023-10-05 20:44:12,854 - Epoch: [3][ 720/ 1236] Overall Loss 0.708774 Objective Loss 0.708774 LR 0.001000 Time 0.017925 +2023-10-05 20:44:13,025 - Epoch: [3][ 730/ 1236] Overall Loss 0.708035 Objective Loss 0.708035 LR 0.001000 Time 0.017913 +2023-10-05 20:44:13,193 - Epoch: [3][ 740/ 1236] Overall Loss 0.707751 Objective Loss 0.707751 LR 0.001000 Time 0.017899 +2023-10-05 20:44:13,365 - Epoch: [3][ 750/ 1236] Overall Loss 0.707286 Objective Loss 0.707286 LR 0.001000 Time 0.017889 +2023-10-05 20:44:13,534 - Epoch: [3][ 760/ 1236] Overall Loss 0.706548 Objective Loss 0.706548 LR 0.001000 Time 0.017876 +2023-10-05 20:44:13,705 - Epoch: [3][ 770/ 1236] Overall Loss 0.705769 Objective Loss 0.705769 LR 0.001000 Time 0.017865 +2023-10-05 20:44:13,874 - Epoch: [3][ 780/ 1236] Overall Loss 0.705356 Objective Loss 0.705356 LR 0.001000 Time 0.017853 +2023-10-05 20:44:14,046 - Epoch: [3][ 790/ 1236] Overall Loss 0.705079 Objective Loss 0.705079 LR 0.001000 Time 0.017843 +2023-10-05 20:44:14,215 - Epoch: [3][ 800/ 1236] Overall Loss 0.704817 Objective Loss 0.704817 LR 0.001000 Time 0.017831 +2023-10-05 20:44:14,386 - Epoch: [3][ 810/ 1236] Overall Loss 0.704649 Objective Loss 0.704649 LR 0.001000 Time 0.017822 +2023-10-05 20:44:14,555 - Epoch: [3][ 820/ 1236] Overall Loss 0.703834 Objective Loss 0.703834 LR 0.001000 Time 0.017811 +2023-10-05 20:44:14,726 - Epoch: [3][ 830/ 1236] Overall Loss 0.703473 Objective Loss 0.703473 LR 0.001000 Time 0.017802 +2023-10-05 20:44:14,894 - Epoch: [3][ 840/ 1236] Overall Loss 0.703003 Objective Loss 0.703003 LR 0.001000 Time 0.017789 +2023-10-05 20:44:15,062 - Epoch: [3][ 850/ 1236] Overall Loss 0.703079 Objective Loss 0.703079 LR 0.001000 Time 0.017777 +2023-10-05 20:44:15,232 - Epoch: [3][ 860/ 1236] Overall Loss 0.702824 Objective Loss 0.702824 LR 0.001000 Time 0.017767 +2023-10-05 20:44:15,403 - Epoch: [3][ 870/ 1236] Overall Loss 0.702765 Objective Loss 0.702765 LR 0.001000 Time 0.017760 +2023-10-05 20:44:15,572 - Epoch: [3][ 880/ 1236] Overall Loss 0.702204 Objective Loss 0.702204 LR 0.001000 Time 0.017750 +2023-10-05 20:44:15,743 - Epoch: [3][ 890/ 1236] Overall Loss 0.701767 Objective Loss 0.701767 LR 0.001000 Time 0.017742 +2023-10-05 20:44:15,913 - Epoch: [3][ 900/ 1236] Overall Loss 0.701039 Objective Loss 0.701039 LR 0.001000 Time 0.017733 +2023-10-05 20:44:16,084 - Epoch: [3][ 910/ 1236] Overall Loss 0.700810 Objective Loss 0.700810 LR 0.001000 Time 0.017726 +2023-10-05 20:44:16,253 - Epoch: [3][ 920/ 1236] Overall Loss 0.700289 Objective Loss 0.700289 LR 0.001000 Time 0.017717 +2023-10-05 20:44:16,424 - Epoch: [3][ 930/ 1236] Overall Loss 0.699791 Objective Loss 0.699791 LR 0.001000 Time 0.017710 +2023-10-05 20:44:16,593 - Epoch: [3][ 940/ 1236] Overall Loss 0.699246 Objective Loss 0.699246 LR 0.001000 Time 0.017701 +2023-10-05 20:44:16,765 - Epoch: [3][ 950/ 1236] Overall Loss 0.698932 Objective Loss 0.698932 LR 0.001000 Time 0.017695 +2023-10-05 20:44:16,934 - Epoch: [3][ 960/ 1236] Overall Loss 0.699045 Objective Loss 0.699045 LR 0.001000 Time 0.017687 +2023-10-05 20:44:17,105 - Epoch: [3][ 970/ 1236] Overall Loss 0.699077 Objective Loss 0.699077 LR 0.001000 Time 0.017680 +2023-10-05 20:44:17,274 - Epoch: [3][ 980/ 1236] Overall Loss 0.698721 Objective Loss 0.698721 LR 0.001000 Time 0.017672 +2023-10-05 20:44:17,446 - Epoch: [3][ 990/ 1236] Overall Loss 0.698481 Objective Loss 0.698481 LR 0.001000 Time 0.017666 +2023-10-05 20:44:17,615 - Epoch: [3][ 1000/ 1236] Overall Loss 0.697904 Objective Loss 0.697904 LR 0.001000 Time 0.017659 +2023-10-05 20:44:17,786 - Epoch: [3][ 1010/ 1236] Overall Loss 0.697503 Objective Loss 0.697503 LR 0.001000 Time 0.017653 +2023-10-05 20:44:17,955 - Epoch: [3][ 1020/ 1236] Overall Loss 0.697003 Objective Loss 0.697003 LR 0.001000 Time 0.017646 +2023-10-05 20:44:18,126 - Epoch: [3][ 1030/ 1236] Overall Loss 0.696743 Objective Loss 0.696743 LR 0.001000 Time 0.017640 +2023-10-05 20:44:18,296 - Epoch: [3][ 1040/ 1236] Overall Loss 0.696046 Objective Loss 0.696046 LR 0.001000 Time 0.017633 +2023-10-05 20:44:18,467 - Epoch: [3][ 1050/ 1236] Overall Loss 0.695878 Objective Loss 0.695878 LR 0.001000 Time 0.017628 +2023-10-05 20:44:18,637 - Epoch: [3][ 1060/ 1236] Overall Loss 0.696005 Objective Loss 0.696005 LR 0.001000 Time 0.017622 +2023-10-05 20:44:18,808 - Epoch: [3][ 1070/ 1236] Overall Loss 0.695706 Objective Loss 0.695706 LR 0.001000 Time 0.017617 +2023-10-05 20:44:18,978 - Epoch: [3][ 1080/ 1236] Overall Loss 0.695334 Objective Loss 0.695334 LR 0.001000 Time 0.017611 +2023-10-05 20:44:19,149 - Epoch: [3][ 1090/ 1236] Overall Loss 0.695607 Objective Loss 0.695607 LR 0.001000 Time 0.017606 +2023-10-05 20:44:19,318 - Epoch: [3][ 1100/ 1236] Overall Loss 0.694996 Objective Loss 0.694996 LR 0.001000 Time 0.017599 +2023-10-05 20:44:19,489 - Epoch: [3][ 1110/ 1236] Overall Loss 0.694466 Objective Loss 0.694466 LR 0.001000 Time 0.017595 +2023-10-05 20:44:19,659 - Epoch: [3][ 1120/ 1236] Overall Loss 0.694050 Objective Loss 0.694050 LR 0.001000 Time 0.017589 +2023-10-05 20:44:19,830 - Epoch: [3][ 1130/ 1236] Overall Loss 0.694058 Objective Loss 0.694058 LR 0.001000 Time 0.017584 +2023-10-05 20:44:19,999 - Epoch: [3][ 1140/ 1236] Overall Loss 0.694055 Objective Loss 0.694055 LR 0.001000 Time 0.017578 +2023-10-05 20:44:20,170 - Epoch: [3][ 1150/ 1236] Overall Loss 0.693546 Objective Loss 0.693546 LR 0.001000 Time 0.017573 +2023-10-05 20:44:20,340 - Epoch: [3][ 1160/ 1236] Overall Loss 0.693064 Objective Loss 0.693064 LR 0.001000 Time 0.017568 +2023-10-05 20:44:20,511 - Epoch: [3][ 1170/ 1236] Overall Loss 0.692325 Objective Loss 0.692325 LR 0.001000 Time 0.017564 +2023-10-05 20:44:20,681 - Epoch: [3][ 1180/ 1236] Overall Loss 0.691823 Objective Loss 0.691823 LR 0.001000 Time 0.017559 +2023-10-05 20:44:20,853 - Epoch: [3][ 1190/ 1236] Overall Loss 0.691234 Objective Loss 0.691234 LR 0.001000 Time 0.017555 +2023-10-05 20:44:21,023 - Epoch: [3][ 1200/ 1236] Overall Loss 0.690787 Objective Loss 0.690787 LR 0.001000 Time 0.017550 +2023-10-05 20:44:21,194 - Epoch: [3][ 1210/ 1236] Overall Loss 0.690655 Objective Loss 0.690655 LR 0.001000 Time 0.017547 +2023-10-05 20:44:21,363 - Epoch: [3][ 1220/ 1236] Overall Loss 0.690399 Objective Loss 0.690399 LR 0.001000 Time 0.017542 +2023-10-05 20:44:21,579 - Epoch: [3][ 1230/ 1236] Overall Loss 0.689940 Objective Loss 0.689940 LR 0.001000 Time 0.017574 +2023-10-05 20:44:21,670 - Epoch: [3][ 1236/ 1236] Overall Loss 0.689410 Objective Loss 0.689410 Top1 73.727088 Top5 95.723014 LR 0.001000 Time 0.017562 +2023-10-05 20:44:21,799 - --- validate (epoch=3)----------- +2023-10-05 20:44:21,799 - 29943 samples (256 per mini-batch) +2023-10-05 20:44:22,201 - Epoch: [3][ 10/ 117] Loss 0.653298 Top1 67.929688 Top5 96.093750 +2023-10-05 20:44:22,306 - Epoch: [3][ 20/ 117] Loss 0.622132 Top1 68.515625 Top5 95.917969 +2023-10-05 20:44:22,411 - Epoch: [3][ 30/ 117] Loss 0.614239 Top1 69.023438 Top5 96.184896 +2023-10-05 20:44:22,516 - Epoch: [3][ 40/ 117] Loss 0.608217 Top1 69.062500 Top5 96.279297 +2023-10-05 20:44:22,621 - Epoch: [3][ 50/ 117] Loss 0.611282 Top1 69.273438 Top5 96.328125 +2023-10-05 20:44:22,726 - Epoch: [3][ 60/ 117] Loss 0.613678 Top1 69.446615 Top5 96.263021 +2023-10-05 20:44:22,830 - Epoch: [3][ 70/ 117] Loss 0.613067 Top1 69.737723 Top5 96.283482 +2023-10-05 20:44:22,934 - Epoch: [3][ 80/ 117] Loss 0.610961 Top1 69.897461 Top5 96.235352 +2023-10-05 20:44:23,038 - Epoch: [3][ 90/ 117] Loss 0.612546 Top1 69.882812 Top5 96.184896 +2023-10-05 20:44:23,142 - Epoch: [3][ 100/ 117] Loss 0.613077 Top1 69.835938 Top5 96.179688 +2023-10-05 20:44:23,254 - Epoch: [3][ 110/ 117] Loss 0.615052 Top1 69.847301 Top5 96.182528 +2023-10-05 20:44:23,309 - Epoch: [3][ 117/ 117] Loss 0.616737 Top1 69.779247 Top5 96.142671 +2023-10-05 20:44:23,442 - ==> Top1: 69.779 Top5: 96.143 Loss: 0.617 + +2023-10-05 20:44:23,442 - ==> Confusion: +[[ 856 0 3 1 30 3 0 0 17 84 1 2 0 7 6 3 12 1 1 1 22] + [ 3 902 1 1 32 32 1 43 7 0 7 0 2 2 9 6 42 3 13 7 18] + [ 13 0 856 9 7 2 32 29 0 6 13 6 0 13 7 12 1 7 7 6 30] + [ 5 0 34 813 1 3 1 2 2 4 67 1 4 14 64 3 8 11 24 0 28] + [ 26 11 2 0 926 4 0 1 1 10 0 0 0 3 26 8 19 3 0 0 10] + [ 7 78 5 3 8 779 3 38 1 3 14 29 1 51 12 6 19 1 3 29 26] + [ 1 2 115 1 3 2 954 27 0 0 4 8 0 0 0 18 6 0 3 24 23] + [ 7 20 28 1 13 54 5 976 3 4 6 6 1 1 0 0 1 4 52 21 15] + [ 30 1 0 0 0 1 0 1 888 48 13 1 0 25 60 0 2 7 1 1 10] + [ 175 0 0 0 34 3 0 0 58 767 1 1 0 38 15 0 3 5 1 0 18] + [ 10 3 11 6 3 6 2 17 16 4 900 4 0 19 15 0 8 1 8 6 14] + [ 4 1 1 0 0 10 2 0 0 0 0 878 57 17 0 12 6 15 0 24 8] + [ 1 0 1 4 0 4 1 2 0 2 0 103 842 3 4 23 14 41 1 8 14] + [ 3 0 2 0 7 18 0 4 10 16 9 13 4 1002 6 4 5 1 0 2 13] + [ 20 3 0 9 21 2 0 1 27 8 2 0 4 3 956 0 9 3 1 0 32] + [ 3 1 4 1 2 0 1 1 0 0 0 30 4 0 0 1021 18 21 0 4 23] + [ 5 9 3 2 9 9 0 0 1 0 1 14 1 4 5 22 1059 1 0 4 12] + [ 1 0 1 3 0 0 4 0 3 0 0 28 72 3 6 50 3 845 1 2 16] + [ 1 8 12 21 1 1 1 55 5 9 20 1 2 0 25 0 3 0 873 3 27] + [ 0 1 0 0 0 3 11 18 2 0 0 28 3 2 1 10 13 6 0 1039 15] + [ 255 321 206 83 221 233 35 148 108 151 338 252 438 426 363 170 493 76 233 593 2762]] + +2023-10-05 20:44:23,443 - ==> Best [Top1: 69.779 Top5: 96.143 Sparsity:0.00 Params: 148928 on epoch: 3] +2023-10-05 20:44:23,443 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:44:23,457 - + +2023-10-05 20:44:23,457 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:44:24,526 - Epoch: [4][ 10/ 1236] Overall Loss 0.654018 Objective Loss 0.654018 LR 0.001000 Time 0.106872 +2023-10-05 20:44:24,693 - Epoch: [4][ 20/ 1236] Overall Loss 0.634802 Objective Loss 0.634802 LR 0.001000 Time 0.061755 +2023-10-05 20:44:24,859 - Epoch: [4][ 30/ 1236] Overall Loss 0.624193 Objective Loss 0.624193 LR 0.001000 Time 0.046690 +2023-10-05 20:44:25,026 - Epoch: [4][ 40/ 1236] Overall Loss 0.618380 Objective Loss 0.618380 LR 0.001000 Time 0.039193 +2023-10-05 20:44:25,191 - Epoch: [4][ 50/ 1236] Overall Loss 0.617750 Objective Loss 0.617750 LR 0.001000 Time 0.034657 +2023-10-05 20:44:25,358 - Epoch: [4][ 60/ 1236] Overall Loss 0.619369 Objective Loss 0.619369 LR 0.001000 Time 0.031656 +2023-10-05 20:44:25,524 - Epoch: [4][ 70/ 1236] Overall Loss 0.620463 Objective Loss 0.620463 LR 0.001000 Time 0.029491 +2023-10-05 20:44:25,690 - Epoch: [4][ 80/ 1236] Overall Loss 0.620028 Objective Loss 0.620028 LR 0.001000 Time 0.027880 +2023-10-05 20:44:25,854 - Epoch: [4][ 90/ 1236] Overall Loss 0.621607 Objective Loss 0.621607 LR 0.001000 Time 0.026604 +2023-10-05 20:44:26,020 - Epoch: [4][ 100/ 1236] Overall Loss 0.619771 Objective Loss 0.619771 LR 0.001000 Time 0.025599 +2023-10-05 20:44:26,184 - Epoch: [4][ 110/ 1236] Overall Loss 0.615589 Objective Loss 0.615589 LR 0.001000 Time 0.024762 +2023-10-05 20:44:26,350 - Epoch: [4][ 120/ 1236] Overall Loss 0.618045 Objective Loss 0.618045 LR 0.001000 Time 0.024077 +2023-10-05 20:44:26,514 - Epoch: [4][ 130/ 1236] Overall Loss 0.619587 Objective Loss 0.619587 LR 0.001000 Time 0.023487 +2023-10-05 20:44:26,680 - Epoch: [4][ 140/ 1236] Overall Loss 0.623875 Objective Loss 0.623875 LR 0.001000 Time 0.022987 +2023-10-05 20:44:26,843 - Epoch: [4][ 150/ 1236] Overall Loss 0.623807 Objective Loss 0.623807 LR 0.001000 Time 0.022545 +2023-10-05 20:44:27,009 - Epoch: [4][ 160/ 1236] Overall Loss 0.623896 Objective Loss 0.623896 LR 0.001000 Time 0.022170 +2023-10-05 20:44:27,173 - Epoch: [4][ 170/ 1236] Overall Loss 0.622574 Objective Loss 0.622574 LR 0.001000 Time 0.021829 +2023-10-05 20:44:27,339 - Epoch: [4][ 180/ 1236] Overall Loss 0.624180 Objective Loss 0.624180 LR 0.001000 Time 0.021535 +2023-10-05 20:44:27,503 - Epoch: [4][ 190/ 1236] Overall Loss 0.623306 Objective Loss 0.623306 LR 0.001000 Time 0.021265 +2023-10-05 20:44:27,669 - Epoch: [4][ 200/ 1236] Overall Loss 0.625597 Objective Loss 0.625597 LR 0.001000 Time 0.021028 +2023-10-05 20:44:27,832 - Epoch: [4][ 210/ 1236] Overall Loss 0.625803 Objective Loss 0.625803 LR 0.001000 Time 0.020805 +2023-10-05 20:44:27,998 - Epoch: [4][ 220/ 1236] Overall Loss 0.624109 Objective Loss 0.624109 LR 0.001000 Time 0.020610 +2023-10-05 20:44:28,162 - Epoch: [4][ 230/ 1236] Overall Loss 0.624417 Objective Loss 0.624417 LR 0.001000 Time 0.020427 +2023-10-05 20:44:28,328 - Epoch: [4][ 240/ 1236] Overall Loss 0.625188 Objective Loss 0.625188 LR 0.001000 Time 0.020265 +2023-10-05 20:44:28,492 - Epoch: [4][ 250/ 1236] Overall Loss 0.625392 Objective Loss 0.625392 LR 0.001000 Time 0.020111 +2023-10-05 20:44:28,658 - Epoch: [4][ 260/ 1236] Overall Loss 0.624093 Objective Loss 0.624093 LR 0.001000 Time 0.019974 +2023-10-05 20:44:28,824 - Epoch: [4][ 270/ 1236] Overall Loss 0.622324 Objective Loss 0.622324 LR 0.001000 Time 0.019848 +2023-10-05 20:44:28,992 - Epoch: [4][ 280/ 1236] Overall Loss 0.622974 Objective Loss 0.622974 LR 0.001000 Time 0.019737 +2023-10-05 20:44:29,159 - Epoch: [4][ 290/ 1236] Overall Loss 0.621807 Objective Loss 0.621807 LR 0.001000 Time 0.019632 +2023-10-05 20:44:29,327 - Epoch: [4][ 300/ 1236] Overall Loss 0.620135 Objective Loss 0.620135 LR 0.001000 Time 0.019536 +2023-10-05 20:44:29,492 - Epoch: [4][ 310/ 1236] Overall Loss 0.619459 Objective Loss 0.619459 LR 0.001000 Time 0.019438 +2023-10-05 20:44:29,659 - Epoch: [4][ 320/ 1236] Overall Loss 0.619925 Objective Loss 0.619925 LR 0.001000 Time 0.019350 +2023-10-05 20:44:29,826 - Epoch: [4][ 330/ 1236] Overall Loss 0.618898 Objective Loss 0.618898 LR 0.001000 Time 0.019269 +2023-10-05 20:44:29,994 - Epoch: [4][ 340/ 1236] Overall Loss 0.618774 Objective Loss 0.618774 LR 0.001000 Time 0.019196 +2023-10-05 20:44:30,161 - Epoch: [4][ 350/ 1236] Overall Loss 0.619627 Objective Loss 0.619627 LR 0.001000 Time 0.019125 +2023-10-05 20:44:30,329 - Epoch: [4][ 360/ 1236] Overall Loss 0.620254 Objective Loss 0.620254 LR 0.001000 Time 0.019059 +2023-10-05 20:44:30,496 - Epoch: [4][ 370/ 1236] Overall Loss 0.619865 Objective Loss 0.619865 LR 0.001000 Time 0.018994 +2023-10-05 20:44:30,664 - Epoch: [4][ 380/ 1236] Overall Loss 0.619481 Objective Loss 0.619481 LR 0.001000 Time 0.018935 +2023-10-05 20:44:30,831 - Epoch: [4][ 390/ 1236] Overall Loss 0.618094 Objective Loss 0.618094 LR 0.001000 Time 0.018878 +2023-10-05 20:44:30,999 - Epoch: [4][ 400/ 1236] Overall Loss 0.618482 Objective Loss 0.618482 LR 0.001000 Time 0.018825 +2023-10-05 20:44:31,167 - Epoch: [4][ 410/ 1236] Overall Loss 0.618760 Objective Loss 0.618760 LR 0.001000 Time 0.018774 +2023-10-05 20:44:31,335 - Epoch: [4][ 420/ 1236] Overall Loss 0.619200 Objective Loss 0.619200 LR 0.001000 Time 0.018726 +2023-10-05 20:44:31,502 - Epoch: [4][ 430/ 1236] Overall Loss 0.619768 Objective Loss 0.619768 LR 0.001000 Time 0.018679 +2023-10-05 20:44:31,670 - Epoch: [4][ 440/ 1236] Overall Loss 0.620091 Objective Loss 0.620091 LR 0.001000 Time 0.018636 +2023-10-05 20:44:31,835 - Epoch: [4][ 450/ 1236] Overall Loss 0.620105 Objective Loss 0.620105 LR 0.001000 Time 0.018586 +2023-10-05 20:44:32,000 - Epoch: [4][ 460/ 1236] Overall Loss 0.620881 Objective Loss 0.620881 LR 0.001000 Time 0.018542 +2023-10-05 20:44:32,164 - Epoch: [4][ 470/ 1236] Overall Loss 0.621449 Objective Loss 0.621449 LR 0.001000 Time 0.018496 +2023-10-05 20:44:32,331 - Epoch: [4][ 480/ 1236] Overall Loss 0.621669 Objective Loss 0.621669 LR 0.001000 Time 0.018456 +2023-10-05 20:44:32,496 - Epoch: [4][ 490/ 1236] Overall Loss 0.622800 Objective Loss 0.622800 LR 0.001000 Time 0.018416 +2023-10-05 20:44:32,661 - Epoch: [4][ 500/ 1236] Overall Loss 0.621498 Objective Loss 0.621498 LR 0.001000 Time 0.018378 +2023-10-05 20:44:32,825 - Epoch: [4][ 510/ 1236] Overall Loss 0.621574 Objective Loss 0.621574 LR 0.001000 Time 0.018339 +2023-10-05 20:44:32,991 - Epoch: [4][ 520/ 1236] Overall Loss 0.621827 Objective Loss 0.621827 LR 0.001000 Time 0.018305 +2023-10-05 20:44:33,155 - Epoch: [4][ 530/ 1236] Overall Loss 0.621471 Objective Loss 0.621471 LR 0.001000 Time 0.018269 +2023-10-05 20:44:33,321 - Epoch: [4][ 540/ 1236] Overall Loss 0.620969 Objective Loss 0.620969 LR 0.001000 Time 0.018237 +2023-10-05 20:44:33,486 - Epoch: [4][ 550/ 1236] Overall Loss 0.620730 Objective Loss 0.620730 LR 0.001000 Time 0.018203 +2023-10-05 20:44:33,652 - Epoch: [4][ 560/ 1236] Overall Loss 0.620521 Objective Loss 0.620521 LR 0.001000 Time 0.018174 +2023-10-05 20:44:33,816 - Epoch: [4][ 570/ 1236] Overall Loss 0.620130 Objective Loss 0.620130 LR 0.001000 Time 0.018143 +2023-10-05 20:44:33,982 - Epoch: [4][ 580/ 1236] Overall Loss 0.619598 Objective Loss 0.619598 LR 0.001000 Time 0.018117 +2023-10-05 20:44:34,147 - Epoch: [4][ 590/ 1236] Overall Loss 0.619196 Objective Loss 0.619196 LR 0.001000 Time 0.018088 +2023-10-05 20:44:34,312 - Epoch: [4][ 600/ 1236] Overall Loss 0.618093 Objective Loss 0.618093 LR 0.001000 Time 0.018062 +2023-10-05 20:44:34,477 - Epoch: [4][ 610/ 1236] Overall Loss 0.617532 Objective Loss 0.617532 LR 0.001000 Time 0.018035 +2023-10-05 20:44:34,642 - Epoch: [4][ 620/ 1236] Overall Loss 0.616849 Objective Loss 0.616849 LR 0.001000 Time 0.018011 +2023-10-05 20:44:34,807 - Epoch: [4][ 630/ 1236] Overall Loss 0.616354 Objective Loss 0.616354 LR 0.001000 Time 0.017985 +2023-10-05 20:44:34,973 - Epoch: [4][ 640/ 1236] Overall Loss 0.615882 Objective Loss 0.615882 LR 0.001000 Time 0.017964 +2023-10-05 20:44:35,137 - Epoch: [4][ 650/ 1236] Overall Loss 0.615417 Objective Loss 0.615417 LR 0.001000 Time 0.017939 +2023-10-05 20:44:35,303 - Epoch: [4][ 660/ 1236] Overall Loss 0.615251 Objective Loss 0.615251 LR 0.001000 Time 0.017919 +2023-10-05 20:44:35,468 - Epoch: [4][ 670/ 1236] Overall Loss 0.615074 Objective Loss 0.615074 LR 0.001000 Time 0.017896 +2023-10-05 20:44:35,634 - Epoch: [4][ 680/ 1236] Overall Loss 0.614774 Objective Loss 0.614774 LR 0.001000 Time 0.017877 +2023-10-05 20:44:35,799 - Epoch: [4][ 690/ 1236] Overall Loss 0.614018 Objective Loss 0.614018 LR 0.001000 Time 0.017856 +2023-10-05 20:44:35,965 - Epoch: [4][ 700/ 1236] Overall Loss 0.613997 Objective Loss 0.613997 LR 0.001000 Time 0.017838 +2023-10-05 20:44:36,129 - Epoch: [4][ 710/ 1236] Overall Loss 0.613750 Objective Loss 0.613750 LR 0.001000 Time 0.017817 +2023-10-05 20:44:36,295 - Epoch: [4][ 720/ 1236] Overall Loss 0.613556 Objective Loss 0.613556 LR 0.001000 Time 0.017800 +2023-10-05 20:44:36,465 - Epoch: [4][ 730/ 1236] Overall Loss 0.612487 Objective Loss 0.612487 LR 0.001000 Time 0.017789 +2023-10-05 20:44:36,635 - Epoch: [4][ 740/ 1236] Overall Loss 0.612184 Objective Loss 0.612184 LR 0.001000 Time 0.017778 +2023-10-05 20:44:36,805 - Epoch: [4][ 750/ 1236] Overall Loss 0.611711 Objective Loss 0.611711 LR 0.001000 Time 0.017767 +2023-10-05 20:44:36,974 - Epoch: [4][ 760/ 1236] Overall Loss 0.611311 Objective Loss 0.611311 LR 0.001000 Time 0.017755 +2023-10-05 20:44:37,144 - Epoch: [4][ 770/ 1236] Overall Loss 0.611280 Objective Loss 0.611280 LR 0.001000 Time 0.017745 +2023-10-05 20:44:37,314 - Epoch: [4][ 780/ 1236] Overall Loss 0.611304 Objective Loss 0.611304 LR 0.001000 Time 0.017735 +2023-10-05 20:44:37,484 - Epoch: [4][ 790/ 1236] Overall Loss 0.611009 Objective Loss 0.611009 LR 0.001000 Time 0.017725 +2023-10-05 20:44:37,654 - Epoch: [4][ 800/ 1236] Overall Loss 0.610951 Objective Loss 0.610951 LR 0.001000 Time 0.017716 +2023-10-05 20:44:37,824 - Epoch: [4][ 810/ 1236] Overall Loss 0.610094 Objective Loss 0.610094 LR 0.001000 Time 0.017707 +2023-10-05 20:44:37,994 - Epoch: [4][ 820/ 1236] Overall Loss 0.610282 Objective Loss 0.610282 LR 0.001000 Time 0.017698 +2023-10-05 20:44:38,164 - Epoch: [4][ 830/ 1236] Overall Loss 0.609953 Objective Loss 0.609953 LR 0.001000 Time 0.017689 +2023-10-05 20:44:38,334 - Epoch: [4][ 840/ 1236] Overall Loss 0.610039 Objective Loss 0.610039 LR 0.001000 Time 0.017680 +2023-10-05 20:44:38,504 - Epoch: [4][ 850/ 1236] Overall Loss 0.609673 Objective Loss 0.609673 LR 0.001000 Time 0.017672 +2023-10-05 20:44:38,673 - Epoch: [4][ 860/ 1236] Overall Loss 0.609751 Objective Loss 0.609751 LR 0.001000 Time 0.017663 +2023-10-05 20:44:38,843 - Epoch: [4][ 870/ 1236] Overall Loss 0.609739 Objective Loss 0.609739 LR 0.001000 Time 0.017655 +2023-10-05 20:44:39,012 - Epoch: [4][ 880/ 1236] Overall Loss 0.609796 Objective Loss 0.609796 LR 0.001000 Time 0.017646 +2023-10-05 20:44:39,182 - Epoch: [4][ 890/ 1236] Overall Loss 0.609709 Objective Loss 0.609709 LR 0.001000 Time 0.017638 +2023-10-05 20:44:39,352 - Epoch: [4][ 900/ 1236] Overall Loss 0.609706 Objective Loss 0.609706 LR 0.001000 Time 0.017631 +2023-10-05 20:44:39,522 - Epoch: [4][ 910/ 1236] Overall Loss 0.609075 Objective Loss 0.609075 LR 0.001000 Time 0.017623 +2023-10-05 20:44:39,692 - Epoch: [4][ 920/ 1236] Overall Loss 0.609296 Objective Loss 0.609296 LR 0.001000 Time 0.017616 +2023-10-05 20:44:39,861 - Epoch: [4][ 930/ 1236] Overall Loss 0.609101 Objective Loss 0.609101 LR 0.001000 Time 0.017609 +2023-10-05 20:44:40,031 - Epoch: [4][ 940/ 1236] Overall Loss 0.609220 Objective Loss 0.609220 LR 0.001000 Time 0.017601 +2023-10-05 20:44:40,201 - Epoch: [4][ 950/ 1236] Overall Loss 0.608634 Objective Loss 0.608634 LR 0.001000 Time 0.017595 +2023-10-05 20:44:40,371 - Epoch: [4][ 960/ 1236] Overall Loss 0.608543 Objective Loss 0.608543 LR 0.001000 Time 0.017588 +2023-10-05 20:44:40,541 - Epoch: [4][ 970/ 1236] Overall Loss 0.608429 Objective Loss 0.608429 LR 0.001000 Time 0.017582 +2023-10-05 20:44:40,711 - Epoch: [4][ 980/ 1236] Overall Loss 0.608797 Objective Loss 0.608797 LR 0.001000 Time 0.017576 +2023-10-05 20:44:40,880 - Epoch: [4][ 990/ 1236] Overall Loss 0.608520 Objective Loss 0.608520 LR 0.001000 Time 0.017569 +2023-10-05 20:44:41,050 - Epoch: [4][ 1000/ 1236] Overall Loss 0.608368 Objective Loss 0.608368 LR 0.001000 Time 0.017563 +2023-10-05 20:44:41,219 - Epoch: [4][ 1010/ 1236] Overall Loss 0.607849 Objective Loss 0.607849 LR 0.001000 Time 0.017556 +2023-10-05 20:44:41,389 - Epoch: [4][ 1020/ 1236] Overall Loss 0.607508 Objective Loss 0.607508 LR 0.001000 Time 0.017550 +2023-10-05 20:44:41,561 - Epoch: [4][ 1030/ 1236] Overall Loss 0.607242 Objective Loss 0.607242 LR 0.001000 Time 0.017543 +2023-10-05 20:44:41,731 - Epoch: [4][ 1040/ 1236] Overall Loss 0.607314 Objective Loss 0.607314 LR 0.001000 Time 0.017537 +2023-10-05 20:44:41,900 - Epoch: [4][ 1050/ 1236] Overall Loss 0.607281 Objective Loss 0.607281 LR 0.001000 Time 0.017531 +2023-10-05 20:44:42,070 - Epoch: [4][ 1060/ 1236] Overall Loss 0.607010 Objective Loss 0.607010 LR 0.001000 Time 0.017526 +2023-10-05 20:44:42,240 - Epoch: [4][ 1070/ 1236] Overall Loss 0.606830 Objective Loss 0.606830 LR 0.001000 Time 0.017521 +2023-10-05 20:44:42,410 - Epoch: [4][ 1080/ 1236] Overall Loss 0.606408 Objective Loss 0.606408 LR 0.001000 Time 0.017516 +2023-10-05 20:44:42,580 - Epoch: [4][ 1090/ 1236] Overall Loss 0.606215 Objective Loss 0.606215 LR 0.001000 Time 0.017511 +2023-10-05 20:44:42,750 - Epoch: [4][ 1100/ 1236] Overall Loss 0.605542 Objective Loss 0.605542 LR 0.001000 Time 0.017506 +2023-10-05 20:44:42,920 - Epoch: [4][ 1110/ 1236] Overall Loss 0.605187 Objective Loss 0.605187 LR 0.001000 Time 0.017500 +2023-10-05 20:44:43,089 - Epoch: [4][ 1120/ 1236] Overall Loss 0.604845 Objective Loss 0.604845 LR 0.001000 Time 0.017495 +2023-10-05 20:44:43,258 - Epoch: [4][ 1130/ 1236] Overall Loss 0.604283 Objective Loss 0.604283 LR 0.001000 Time 0.017490 +2023-10-05 20:44:43,428 - Epoch: [4][ 1140/ 1236] Overall Loss 0.603811 Objective Loss 0.603811 LR 0.001000 Time 0.017485 +2023-10-05 20:44:43,598 - Epoch: [4][ 1150/ 1236] Overall Loss 0.603274 Objective Loss 0.603274 LR 0.001000 Time 0.017481 +2023-10-05 20:44:43,768 - Epoch: [4][ 1160/ 1236] Overall Loss 0.603378 Objective Loss 0.603378 LR 0.001000 Time 0.017476 +2023-10-05 20:44:43,938 - Epoch: [4][ 1170/ 1236] Overall Loss 0.603341 Objective Loss 0.603341 LR 0.001000 Time 0.017472 +2023-10-05 20:44:44,108 - Epoch: [4][ 1180/ 1236] Overall Loss 0.603335 Objective Loss 0.603335 LR 0.001000 Time 0.017468 +2023-10-05 20:44:44,278 - Epoch: [4][ 1190/ 1236] Overall Loss 0.603680 Objective Loss 0.603680 LR 0.001000 Time 0.017464 +2023-10-05 20:44:44,448 - Epoch: [4][ 1200/ 1236] Overall Loss 0.604172 Objective Loss 0.604172 LR 0.001000 Time 0.017459 +2023-10-05 20:44:44,618 - Epoch: [4][ 1210/ 1236] Overall Loss 0.604506 Objective Loss 0.604506 LR 0.001000 Time 0.017455 +2023-10-05 20:44:44,787 - Epoch: [4][ 1220/ 1236] Overall Loss 0.604140 Objective Loss 0.604140 LR 0.001000 Time 0.017451 +2023-10-05 20:44:45,003 - Epoch: [4][ 1230/ 1236] Overall Loss 0.603788 Objective Loss 0.603788 LR 0.001000 Time 0.017484 +2023-10-05 20:44:45,095 - Epoch: [4][ 1236/ 1236] Overall Loss 0.603651 Objective Loss 0.603651 Top1 73.116090 Top5 95.315682 LR 0.001000 Time 0.017473 +2023-10-05 20:44:45,222 - --- validate (epoch=4)----------- +2023-10-05 20:44:45,222 - 29943 samples (256 per mini-batch) +2023-10-05 20:44:45,626 - Epoch: [4][ 10/ 117] Loss 0.523494 Top1 72.617188 Top5 96.953125 +2023-10-05 20:44:45,731 - Epoch: [4][ 20/ 117] Loss 0.533889 Top1 72.265625 Top5 96.699219 +2023-10-05 20:44:45,834 - Epoch: [4][ 30/ 117] Loss 0.548896 Top1 71.914062 Top5 96.549479 +2023-10-05 20:44:45,938 - Epoch: [4][ 40/ 117] Loss 0.544989 Top1 71.845703 Top5 96.455078 +2023-10-05 20:44:46,042 - Epoch: [4][ 50/ 117] Loss 0.535663 Top1 72.210938 Top5 96.632812 +2023-10-05 20:44:46,146 - Epoch: [4][ 60/ 117] Loss 0.531203 Top1 72.337240 Top5 96.673177 +2023-10-05 20:44:46,248 - Epoch: [4][ 70/ 117] Loss 0.531717 Top1 72.260045 Top5 96.618304 +2023-10-05 20:44:46,352 - Epoch: [4][ 80/ 117] Loss 0.529836 Top1 72.294922 Top5 96.621094 +2023-10-05 20:44:46,455 - Epoch: [4][ 90/ 117] Loss 0.529129 Top1 72.326389 Top5 96.675347 +2023-10-05 20:44:46,558 - Epoch: [4][ 100/ 117] Loss 0.533948 Top1 72.210938 Top5 96.691406 +2023-10-05 20:44:46,671 - Epoch: [4][ 110/ 117] Loss 0.531097 Top1 72.212358 Top5 96.725852 +2023-10-05 20:44:46,726 - Epoch: [4][ 117/ 117] Loss 0.530855 Top1 72.213873 Top5 96.727115 +2023-10-05 20:44:46,866 - ==> Top1: 72.214 Top5: 96.727 Loss: 0.531 + +2023-10-05 20:44:46,867 - ==> Confusion: +[[ 871 0 3 0 17 4 0 0 12 95 2 3 1 8 3 0 8 0 0 1 22] + [ 2 973 2 1 7 33 1 42 4 0 7 2 2 0 2 3 19 2 5 6 18] + [ 15 0 843 15 9 3 58 12 0 5 29 5 8 5 3 5 2 3 6 5 25] + [ 4 3 31 840 0 6 0 2 9 1 54 1 9 8 45 5 9 13 22 0 27] + [ 24 26 0 0 907 11 1 1 3 18 1 3 1 5 13 5 23 1 1 0 6] + [ 4 75 5 3 2 838 0 29 4 3 16 16 5 48 7 4 18 2 1 20 16] + [ 2 7 50 2 0 1 1042 13 1 0 7 7 2 1 0 22 3 1 2 14 14] + [ 8 15 24 0 2 63 9 971 4 0 7 16 1 0 0 0 0 2 59 25 12] + [ 31 1 0 2 0 0 0 1 936 53 20 5 0 19 7 1 2 5 1 0 5] + [ 169 0 0 0 15 3 0 3 49 815 1 3 0 43 3 0 2 1 0 0 12] + [ 7 2 9 3 4 3 0 5 16 2 960 4 0 12 4 1 5 0 4 1 11] + [ 2 1 0 0 1 16 2 0 1 0 0 900 52 8 0 5 5 16 0 20 6] + [ 1 0 4 4 0 5 0 2 0 0 0 82 905 0 2 10 13 27 0 10 3] + [ 5 0 1 0 6 18 0 2 13 16 19 4 3 1009 1 2 2 1 0 4 13] + [ 24 2 0 10 14 0 0 0 64 8 2 0 7 3 910 0 12 3 9 0 33] + [ 0 1 1 2 2 0 2 0 1 0 0 29 19 1 0 1010 23 21 0 2 20] + [ 0 11 1 2 8 9 0 0 1 0 1 13 3 2 1 13 1076 0 0 5 15] + [ 0 0 0 1 0 0 3 0 3 0 0 12 55 0 1 11 2 940 1 2 7] + [ 1 11 7 11 0 0 0 44 13 4 15 2 7 0 12 0 4 1 907 6 23] + [ 0 2 0 1 0 6 8 20 2 0 6 32 7 2 0 9 11 3 2 1022 19] + [ 268 348 115 109 169 273 42 120 141 160 333 193 627 430 205 147 500 81 230 466 2948]] + +2023-10-05 20:44:46,868 - ==> Best [Top1: 72.214 Top5: 96.727 Sparsity:0.00 Params: 148928 on epoch: 4] +2023-10-05 20:44:46,868 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:44:46,875 - + +2023-10-05 20:44:46,875 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:44:47,844 - Epoch: [5][ 10/ 1236] Overall Loss 0.544107 Objective Loss 0.544107 LR 0.001000 Time 0.096815 +2023-10-05 20:44:48,010 - Epoch: [5][ 20/ 1236] Overall Loss 0.555428 Objective Loss 0.555428 LR 0.001000 Time 0.056710 +2023-10-05 20:44:48,175 - Epoch: [5][ 30/ 1236] Overall Loss 0.548994 Objective Loss 0.548994 LR 0.001000 Time 0.043296 +2023-10-05 20:44:48,342 - Epoch: [5][ 40/ 1236] Overall Loss 0.555040 Objective Loss 0.555040 LR 0.001000 Time 0.036632 +2023-10-05 20:44:48,509 - Epoch: [5][ 50/ 1236] Overall Loss 0.556682 Objective Loss 0.556682 LR 0.001000 Time 0.032629 +2023-10-05 20:44:48,675 - Epoch: [5][ 60/ 1236] Overall Loss 0.562138 Objective Loss 0.562138 LR 0.001000 Time 0.029959 +2023-10-05 20:44:48,841 - Epoch: [5][ 70/ 1236] Overall Loss 0.559320 Objective Loss 0.559320 LR 0.001000 Time 0.028042 +2023-10-05 20:44:49,006 - Epoch: [5][ 80/ 1236] Overall Loss 0.556204 Objective Loss 0.556204 LR 0.001000 Time 0.026605 +2023-10-05 20:44:49,171 - Epoch: [5][ 90/ 1236] Overall Loss 0.555123 Objective Loss 0.555123 LR 0.001000 Time 0.025480 +2023-10-05 20:44:49,338 - Epoch: [5][ 100/ 1236] Overall Loss 0.556418 Objective Loss 0.556418 LR 0.001000 Time 0.024590 +2023-10-05 20:44:49,502 - Epoch: [5][ 110/ 1236] Overall Loss 0.554918 Objective Loss 0.554918 LR 0.001000 Time 0.023849 +2023-10-05 20:44:49,668 - Epoch: [5][ 120/ 1236] Overall Loss 0.553302 Objective Loss 0.553302 LR 0.001000 Time 0.023241 +2023-10-05 20:44:49,833 - Epoch: [5][ 130/ 1236] Overall Loss 0.555652 Objective Loss 0.555652 LR 0.001000 Time 0.022718 +2023-10-05 20:44:49,998 - Epoch: [5][ 140/ 1236] Overall Loss 0.556053 Objective Loss 0.556053 LR 0.001000 Time 0.022274 +2023-10-05 20:44:50,162 - Epoch: [5][ 150/ 1236] Overall Loss 0.558330 Objective Loss 0.558330 LR 0.001000 Time 0.021883 +2023-10-05 20:44:50,328 - Epoch: [5][ 160/ 1236] Overall Loss 0.554667 Objective Loss 0.554667 LR 0.001000 Time 0.021545 +2023-10-05 20:44:50,492 - Epoch: [5][ 170/ 1236] Overall Loss 0.555368 Objective Loss 0.555368 LR 0.001000 Time 0.021242 +2023-10-05 20:44:50,657 - Epoch: [5][ 180/ 1236] Overall Loss 0.556057 Objective Loss 0.556057 LR 0.001000 Time 0.020978 +2023-10-05 20:44:50,821 - Epoch: [5][ 190/ 1236] Overall Loss 0.557083 Objective Loss 0.557083 LR 0.001000 Time 0.020737 +2023-10-05 20:44:50,986 - Epoch: [5][ 200/ 1236] Overall Loss 0.556134 Objective Loss 0.556134 LR 0.001000 Time 0.020524 +2023-10-05 20:44:51,151 - Epoch: [5][ 210/ 1236] Overall Loss 0.556853 Objective Loss 0.556853 LR 0.001000 Time 0.020329 +2023-10-05 20:44:51,317 - Epoch: [5][ 220/ 1236] Overall Loss 0.556077 Objective Loss 0.556077 LR 0.001000 Time 0.020157 +2023-10-05 20:44:51,488 - Epoch: [5][ 230/ 1236] Overall Loss 0.556011 Objective Loss 0.556011 LR 0.001000 Time 0.020023 +2023-10-05 20:44:51,659 - Epoch: [5][ 240/ 1236] Overall Loss 0.555653 Objective Loss 0.555653 LR 0.001000 Time 0.019900 +2023-10-05 20:44:51,822 - Epoch: [5][ 250/ 1236] Overall Loss 0.557564 Objective Loss 0.557564 LR 0.001000 Time 0.019756 +2023-10-05 20:44:51,987 - Epoch: [5][ 260/ 1236] Overall Loss 0.558517 Objective Loss 0.558517 LR 0.001000 Time 0.019627 +2023-10-05 20:44:52,150 - Epoch: [5][ 270/ 1236] Overall Loss 0.558557 Objective Loss 0.558557 LR 0.001000 Time 0.019503 +2023-10-05 20:44:52,314 - Epoch: [5][ 280/ 1236] Overall Loss 0.559077 Objective Loss 0.559077 LR 0.001000 Time 0.019392 +2023-10-05 20:44:52,477 - Epoch: [5][ 290/ 1236] Overall Loss 0.559972 Objective Loss 0.559972 LR 0.001000 Time 0.019284 +2023-10-05 20:44:52,640 - Epoch: [5][ 300/ 1236] Overall Loss 0.560575 Objective Loss 0.560575 LR 0.001000 Time 0.019185 +2023-10-05 20:44:52,803 - Epoch: [5][ 310/ 1236] Overall Loss 0.560166 Objective Loss 0.560166 LR 0.001000 Time 0.019091 +2023-10-05 20:44:52,967 - Epoch: [5][ 320/ 1236] Overall Loss 0.561530 Objective Loss 0.561530 LR 0.001000 Time 0.019005 +2023-10-05 20:44:53,130 - Epoch: [5][ 330/ 1236] Overall Loss 0.562211 Objective Loss 0.562211 LR 0.001000 Time 0.018922 +2023-10-05 20:44:53,293 - Epoch: [5][ 340/ 1236] Overall Loss 0.562338 Objective Loss 0.562338 LR 0.001000 Time 0.018845 +2023-10-05 20:44:53,456 - Epoch: [5][ 350/ 1236] Overall Loss 0.562735 Objective Loss 0.562735 LR 0.001000 Time 0.018772 +2023-10-05 20:44:53,620 - Epoch: [5][ 360/ 1236] Overall Loss 0.564073 Objective Loss 0.564073 LR 0.001000 Time 0.018706 +2023-10-05 20:44:53,785 - Epoch: [5][ 370/ 1236] Overall Loss 0.562853 Objective Loss 0.562853 LR 0.001000 Time 0.018644 +2023-10-05 20:44:53,956 - Epoch: [5][ 380/ 1236] Overall Loss 0.562275 Objective Loss 0.562275 LR 0.001000 Time 0.018602 +2023-10-05 20:44:54,123 - Epoch: [5][ 390/ 1236] Overall Loss 0.561793 Objective Loss 0.561793 LR 0.001000 Time 0.018552 +2023-10-05 20:44:54,294 - Epoch: [5][ 400/ 1236] Overall Loss 0.561615 Objective Loss 0.561615 LR 0.001000 Time 0.018516 +2023-10-05 20:44:54,461 - Epoch: [5][ 410/ 1236] Overall Loss 0.561594 Objective Loss 0.561594 LR 0.001000 Time 0.018472 +2023-10-05 20:44:54,633 - Epoch: [5][ 420/ 1236] Overall Loss 0.560935 Objective Loss 0.560935 LR 0.001000 Time 0.018440 +2023-10-05 20:44:54,800 - Epoch: [5][ 430/ 1236] Overall Loss 0.561966 Objective Loss 0.561966 LR 0.001000 Time 0.018399 +2023-10-05 20:44:54,972 - Epoch: [5][ 440/ 1236] Overall Loss 0.562085 Objective Loss 0.562085 LR 0.001000 Time 0.018371 +2023-10-05 20:44:55,139 - Epoch: [5][ 450/ 1236] Overall Loss 0.561957 Objective Loss 0.561957 LR 0.001000 Time 0.018333 +2023-10-05 20:44:55,311 - Epoch: [5][ 460/ 1236] Overall Loss 0.561287 Objective Loss 0.561287 LR 0.001000 Time 0.018308 +2023-10-05 20:44:55,478 - Epoch: [5][ 470/ 1236] Overall Loss 0.561333 Objective Loss 0.561333 LR 0.001000 Time 0.018274 +2023-10-05 20:44:55,650 - Epoch: [5][ 480/ 1236] Overall Loss 0.560577 Objective Loss 0.560577 LR 0.001000 Time 0.018251 +2023-10-05 20:44:55,818 - Epoch: [5][ 490/ 1236] Overall Loss 0.559718 Objective Loss 0.559718 LR 0.001000 Time 0.018219 +2023-10-05 20:44:55,990 - Epoch: [5][ 500/ 1236] Overall Loss 0.559352 Objective Loss 0.559352 LR 0.001000 Time 0.018199 +2023-10-05 20:44:56,158 - Epoch: [5][ 510/ 1236] Overall Loss 0.558370 Objective Loss 0.558370 LR 0.001000 Time 0.018171 +2023-10-05 20:44:56,330 - Epoch: [5][ 520/ 1236] Overall Loss 0.558433 Objective Loss 0.558433 LR 0.001000 Time 0.018152 +2023-10-05 20:44:56,499 - Epoch: [5][ 530/ 1236] Overall Loss 0.558091 Objective Loss 0.558091 LR 0.001000 Time 0.018127 +2023-10-05 20:44:56,671 - Epoch: [5][ 540/ 1236] Overall Loss 0.558385 Objective Loss 0.558385 LR 0.001000 Time 0.018110 +2023-10-05 20:44:56,839 - Epoch: [5][ 550/ 1236] Overall Loss 0.557804 Objective Loss 0.557804 LR 0.001000 Time 0.018085 +2023-10-05 20:44:57,011 - Epoch: [5][ 560/ 1236] Overall Loss 0.557901 Objective Loss 0.557901 LR 0.001000 Time 0.018069 +2023-10-05 20:44:57,179 - Epoch: [5][ 570/ 1236] Overall Loss 0.557163 Objective Loss 0.557163 LR 0.001000 Time 0.018047 +2023-10-05 20:44:57,351 - Epoch: [5][ 580/ 1236] Overall Loss 0.556643 Objective Loss 0.556643 LR 0.001000 Time 0.018031 +2023-10-05 20:44:57,519 - Epoch: [5][ 590/ 1236] Overall Loss 0.556654 Objective Loss 0.556654 LR 0.001000 Time 0.018010 +2023-10-05 20:44:57,692 - Epoch: [5][ 600/ 1236] Overall Loss 0.556341 Objective Loss 0.556341 LR 0.001000 Time 0.017997 +2023-10-05 20:44:57,859 - Epoch: [5][ 610/ 1236] Overall Loss 0.556045 Objective Loss 0.556045 LR 0.001000 Time 0.017976 +2023-10-05 20:44:58,024 - Epoch: [5][ 620/ 1236] Overall Loss 0.555518 Objective Loss 0.555518 LR 0.001000 Time 0.017951 +2023-10-05 20:44:58,188 - Epoch: [5][ 630/ 1236] Overall Loss 0.554898 Objective Loss 0.554898 LR 0.001000 Time 0.017927 +2023-10-05 20:44:58,354 - Epoch: [5][ 640/ 1236] Overall Loss 0.554270 Objective Loss 0.554270 LR 0.001000 Time 0.017904 +2023-10-05 20:44:58,518 - Epoch: [5][ 650/ 1236] Overall Loss 0.554572 Objective Loss 0.554572 LR 0.001000 Time 0.017882 +2023-10-05 20:44:58,684 - Epoch: [5][ 660/ 1236] Overall Loss 0.554971 Objective Loss 0.554971 LR 0.001000 Time 0.017861 +2023-10-05 20:44:58,848 - Epoch: [5][ 670/ 1236] Overall Loss 0.555135 Objective Loss 0.555135 LR 0.001000 Time 0.017840 +2023-10-05 20:44:59,014 - Epoch: [5][ 680/ 1236] Overall Loss 0.555298 Objective Loss 0.555298 LR 0.001000 Time 0.017820 +2023-10-05 20:44:59,178 - Epoch: [5][ 690/ 1236] Overall Loss 0.555728 Objective Loss 0.555728 LR 0.001000 Time 0.017800 +2023-10-05 20:44:59,344 - Epoch: [5][ 700/ 1236] Overall Loss 0.556148 Objective Loss 0.556148 LR 0.001000 Time 0.017781 +2023-10-05 20:44:59,508 - Epoch: [5][ 710/ 1236] Overall Loss 0.556612 Objective Loss 0.556612 LR 0.001000 Time 0.017763 +2023-10-05 20:44:59,674 - Epoch: [5][ 720/ 1236] Overall Loss 0.555850 Objective Loss 0.555850 LR 0.001000 Time 0.017745 +2023-10-05 20:44:59,838 - Epoch: [5][ 730/ 1236] Overall Loss 0.556082 Objective Loss 0.556082 LR 0.001000 Time 0.017727 +2023-10-05 20:45:00,004 - Epoch: [5][ 740/ 1236] Overall Loss 0.555600 Objective Loss 0.555600 LR 0.001000 Time 0.017711 +2023-10-05 20:45:00,168 - Epoch: [5][ 750/ 1236] Overall Loss 0.554844 Objective Loss 0.554844 LR 0.001000 Time 0.017694 +2023-10-05 20:45:00,334 - Epoch: [5][ 760/ 1236] Overall Loss 0.554612 Objective Loss 0.554612 LR 0.001000 Time 0.017678 +2023-10-05 20:45:00,498 - Epoch: [5][ 770/ 1236] Overall Loss 0.554863 Objective Loss 0.554863 LR 0.001000 Time 0.017662 +2023-10-05 20:45:00,664 - Epoch: [5][ 780/ 1236] Overall Loss 0.554466 Objective Loss 0.554466 LR 0.001000 Time 0.017647 +2023-10-05 20:45:00,828 - Epoch: [5][ 790/ 1236] Overall Loss 0.554446 Objective Loss 0.554446 LR 0.001000 Time 0.017632 +2023-10-05 20:45:00,994 - Epoch: [5][ 800/ 1236] Overall Loss 0.554689 Objective Loss 0.554689 LR 0.001000 Time 0.017618 +2023-10-05 20:45:01,158 - Epoch: [5][ 810/ 1236] Overall Loss 0.554703 Objective Loss 0.554703 LR 0.001000 Time 0.017603 +2023-10-05 20:45:01,324 - Epoch: [5][ 820/ 1236] Overall Loss 0.554280 Objective Loss 0.554280 LR 0.001000 Time 0.017590 +2023-10-05 20:45:01,488 - Epoch: [5][ 830/ 1236] Overall Loss 0.553470 Objective Loss 0.553470 LR 0.001000 Time 0.017576 +2023-10-05 20:45:01,654 - Epoch: [5][ 840/ 1236] Overall Loss 0.553742 Objective Loss 0.553742 LR 0.001000 Time 0.017563 +2023-10-05 20:45:01,818 - Epoch: [5][ 850/ 1236] Overall Loss 0.553419 Objective Loss 0.553419 LR 0.001000 Time 0.017550 +2023-10-05 20:45:01,984 - Epoch: [5][ 860/ 1236] Overall Loss 0.553248 Objective Loss 0.553248 LR 0.001000 Time 0.017538 +2023-10-05 20:45:02,148 - Epoch: [5][ 870/ 1236] Overall Loss 0.552866 Objective Loss 0.552866 LR 0.001000 Time 0.017525 +2023-10-05 20:45:02,314 - Epoch: [5][ 880/ 1236] Overall Loss 0.552740 Objective Loss 0.552740 LR 0.001000 Time 0.017513 +2023-10-05 20:45:02,478 - Epoch: [5][ 890/ 1236] Overall Loss 0.552881 Objective Loss 0.552881 LR 0.001000 Time 0.017501 +2023-10-05 20:45:02,644 - Epoch: [5][ 900/ 1236] Overall Loss 0.552958 Objective Loss 0.552958 LR 0.001000 Time 0.017490 +2023-10-05 20:45:02,808 - Epoch: [5][ 910/ 1236] Overall Loss 0.553605 Objective Loss 0.553605 LR 0.001000 Time 0.017479 +2023-10-05 20:45:02,974 - Epoch: [5][ 920/ 1236] Overall Loss 0.553165 Objective Loss 0.553165 LR 0.001000 Time 0.017468 +2023-10-05 20:45:03,138 - Epoch: [5][ 930/ 1236] Overall Loss 0.553408 Objective Loss 0.553408 LR 0.001000 Time 0.017457 +2023-10-05 20:45:03,304 - Epoch: [5][ 940/ 1236] Overall Loss 0.553074 Objective Loss 0.553074 LR 0.001000 Time 0.017447 +2023-10-05 20:45:03,468 - Epoch: [5][ 950/ 1236] Overall Loss 0.553759 Objective Loss 0.553759 LR 0.001000 Time 0.017436 +2023-10-05 20:45:03,634 - Epoch: [5][ 960/ 1236] Overall Loss 0.553796 Objective Loss 0.553796 LR 0.001000 Time 0.017426 +2023-10-05 20:45:03,798 - Epoch: [5][ 970/ 1236] Overall Loss 0.553623 Objective Loss 0.553623 LR 0.001000 Time 0.017416 +2023-10-05 20:45:03,964 - Epoch: [5][ 980/ 1236] Overall Loss 0.553476 Objective Loss 0.553476 LR 0.001000 Time 0.017407 +2023-10-05 20:45:04,128 - Epoch: [5][ 990/ 1236] Overall Loss 0.553494 Objective Loss 0.553494 LR 0.001000 Time 0.017397 +2023-10-05 20:45:04,294 - Epoch: [5][ 1000/ 1236] Overall Loss 0.553739 Objective Loss 0.553739 LR 0.001000 Time 0.017388 +2023-10-05 20:45:04,458 - Epoch: [5][ 1010/ 1236] Overall Loss 0.553572 Objective Loss 0.553572 LR 0.001000 Time 0.017379 +2023-10-05 20:45:04,624 - Epoch: [5][ 1020/ 1236] Overall Loss 0.553886 Objective Loss 0.553886 LR 0.001000 Time 0.017370 +2023-10-05 20:45:04,788 - Epoch: [5][ 1030/ 1236] Overall Loss 0.553998 Objective Loss 0.553998 LR 0.001000 Time 0.017361 +2023-10-05 20:45:04,954 - Epoch: [5][ 1040/ 1236] Overall Loss 0.553948 Objective Loss 0.553948 LR 0.001000 Time 0.017353 +2023-10-05 20:45:05,118 - Epoch: [5][ 1050/ 1236] Overall Loss 0.553795 Objective Loss 0.553795 LR 0.001000 Time 0.017344 +2023-10-05 20:45:05,284 - Epoch: [5][ 1060/ 1236] Overall Loss 0.553816 Objective Loss 0.553816 LR 0.001000 Time 0.017336 +2023-10-05 20:45:05,449 - Epoch: [5][ 1070/ 1236] Overall Loss 0.553871 Objective Loss 0.553871 LR 0.001000 Time 0.017328 +2023-10-05 20:45:05,614 - Epoch: [5][ 1080/ 1236] Overall Loss 0.554047 Objective Loss 0.554047 LR 0.001000 Time 0.017320 +2023-10-05 20:45:05,778 - Epoch: [5][ 1090/ 1236] Overall Loss 0.553816 Objective Loss 0.553816 LR 0.001000 Time 0.017312 +2023-10-05 20:45:05,944 - Epoch: [5][ 1100/ 1236] Overall Loss 0.553348 Objective Loss 0.553348 LR 0.001000 Time 0.017305 +2023-10-05 20:45:06,108 - Epoch: [5][ 1110/ 1236] Overall Loss 0.553408 Objective Loss 0.553408 LR 0.001000 Time 0.017297 +2023-10-05 20:45:06,274 - Epoch: [5][ 1120/ 1236] Overall Loss 0.553092 Objective Loss 0.553092 LR 0.001000 Time 0.017290 +2023-10-05 20:45:06,439 - Epoch: [5][ 1130/ 1236] Overall Loss 0.552634 Objective Loss 0.552634 LR 0.001000 Time 0.017283 +2023-10-05 20:45:06,604 - Epoch: [5][ 1140/ 1236] Overall Loss 0.552433 Objective Loss 0.552433 LR 0.001000 Time 0.017276 +2023-10-05 20:45:06,769 - Epoch: [5][ 1150/ 1236] Overall Loss 0.552629 Objective Loss 0.552629 LR 0.001000 Time 0.017268 +2023-10-05 20:45:06,934 - Epoch: [5][ 1160/ 1236] Overall Loss 0.551992 Objective Loss 0.551992 LR 0.001000 Time 0.017262 +2023-10-05 20:45:07,098 - Epoch: [5][ 1170/ 1236] Overall Loss 0.551698 Objective Loss 0.551698 LR 0.001000 Time 0.017255 +2023-10-05 20:45:07,264 - Epoch: [5][ 1180/ 1236] Overall Loss 0.551517 Objective Loss 0.551517 LR 0.001000 Time 0.017248 +2023-10-05 20:45:07,429 - Epoch: [5][ 1190/ 1236] Overall Loss 0.551648 Objective Loss 0.551648 LR 0.001000 Time 0.017242 +2023-10-05 20:45:07,594 - Epoch: [5][ 1200/ 1236] Overall Loss 0.551574 Objective Loss 0.551574 LR 0.001000 Time 0.017235 +2023-10-05 20:45:07,758 - Epoch: [5][ 1210/ 1236] Overall Loss 0.551627 Objective Loss 0.551627 LR 0.001000 Time 0.017229 +2023-10-05 20:45:07,924 - Epoch: [5][ 1220/ 1236] Overall Loss 0.551275 Objective Loss 0.551275 LR 0.001000 Time 0.017223 +2023-10-05 20:45:08,135 - Epoch: [5][ 1230/ 1236] Overall Loss 0.551388 Objective Loss 0.551388 LR 0.001000 Time 0.017254 +2023-10-05 20:45:08,226 - Epoch: [5][ 1236/ 1236] Overall Loss 0.551284 Objective Loss 0.551284 Top1 76.985743 Top5 96.334012 LR 0.001000 Time 0.017244 +2023-10-05 20:45:08,348 - --- validate (epoch=5)----------- +2023-10-05 20:45:08,348 - 29943 samples (256 per mini-batch) +2023-10-05 20:45:08,772 - Epoch: [5][ 10/ 117] Loss 0.535498 Top1 72.304688 Top5 96.093750 +2023-10-05 20:45:08,886 - Epoch: [5][ 20/ 117] Loss 0.515058 Top1 73.417969 Top5 96.367188 +2023-10-05 20:45:08,999 - Epoch: [5][ 30/ 117] Loss 0.512442 Top1 73.125000 Top5 96.250000 +2023-10-05 20:45:09,111 - Epoch: [5][ 40/ 117] Loss 0.524621 Top1 73.007812 Top5 96.220703 +2023-10-05 20:45:09,223 - Epoch: [5][ 50/ 117] Loss 0.526551 Top1 72.867188 Top5 96.304688 +2023-10-05 20:45:09,336 - Epoch: [5][ 60/ 117] Loss 0.526087 Top1 73.020833 Top5 96.373698 +2023-10-05 20:45:09,448 - Epoch: [5][ 70/ 117] Loss 0.522773 Top1 73.158482 Top5 96.406250 +2023-10-05 20:45:09,553 - Epoch: [5][ 80/ 117] Loss 0.523580 Top1 73.076172 Top5 96.464844 +2023-10-05 20:45:09,656 - Epoch: [5][ 90/ 117] Loss 0.526001 Top1 73.007812 Top5 96.419271 +2023-10-05 20:45:09,760 - Epoch: [5][ 100/ 117] Loss 0.526064 Top1 73.031250 Top5 96.472656 +2023-10-05 20:45:09,877 - Epoch: [5][ 110/ 117] Loss 0.522466 Top1 73.128551 Top5 96.548295 +2023-10-05 20:45:09,932 - Epoch: [5][ 117/ 117] Loss 0.522083 Top1 73.148983 Top5 96.580169 +2023-10-05 20:45:10,051 - ==> Top1: 73.149 Top5: 96.580 Loss: 0.522 + +2023-10-05 20:45:10,052 - ==> Confusion: +[[ 859 1 2 1 8 5 0 0 19 117 0 0 1 8 2 1 6 3 1 1 15] + [ 2 1034 1 1 5 28 2 26 10 0 2 2 0 0 0 3 2 2 6 2 3] + [ 16 2 882 12 9 8 31 18 0 5 12 10 4 1 2 3 2 5 12 3 19] + [ 5 8 20 842 0 11 5 3 14 0 38 1 7 5 34 1 2 14 54 0 25] + [ 33 29 0 0 915 8 1 0 6 13 1 4 1 2 5 6 17 1 0 0 8] + [ 5 110 5 4 3 894 2 21 10 2 2 11 6 15 2 1 5 0 1 6 11] + [ 1 17 70 1 0 4 1027 36 0 0 1 5 1 0 0 9 0 1 5 5 8] + [ 9 33 15 1 1 64 4 970 7 5 6 13 0 0 0 1 0 2 66 10 11] + [ 25 4 0 2 0 2 0 0 969 52 9 3 0 11 3 6 0 1 1 1 0] + [ 109 1 0 0 10 2 0 0 57 890 0 5 0 27 3 1 0 0 1 0 13] + [ 6 8 7 8 4 9 2 15 30 5 885 3 2 13 7 5 0 1 30 1 12] + [ 1 3 0 0 0 37 0 0 1 0 0 846 89 3 0 4 5 19 0 19 8] + [ 2 4 0 4 0 8 0 2 0 0 1 49 929 0 1 8 8 38 2 3 9] + [ 5 1 0 0 3 48 0 0 21 16 5 15 7 969 1 3 7 1 0 2 15] + [ 18 7 0 8 14 0 0 0 124 20 1 0 6 2 848 0 2 6 12 0 33] + [ 1 3 4 2 6 5 3 0 1 2 0 21 13 0 0 1005 31 15 0 4 18] + [ 2 35 1 0 11 7 0 0 3 0 0 7 1 1 0 7 1063 1 0 5 17] + [ 1 0 1 0 0 0 2 0 5 0 0 12 56 0 1 9 1 939 1 0 10] + [ 2 25 10 5 2 4 1 30 17 3 6 0 4 0 4 0 1 1 939 1 13] + [ 0 11 1 0 1 12 9 26 1 0 0 31 4 0 0 1 15 4 1 1024 11] + [ 301 635 114 77 138 352 44 104 231 189 190 189 518 311 129 80 253 96 311 469 3174]] + +2023-10-05 20:45:10,053 - ==> Best [Top1: 73.149 Top5: 96.580 Sparsity:0.00 Params: 148928 on epoch: 5] +2023-10-05 20:45:10,053 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:45:10,066 - + +2023-10-05 20:45:10,066 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:45:11,008 - Epoch: [6][ 10/ 1236] Overall Loss 0.481077 Objective Loss 0.481077 LR 0.001000 Time 0.094170 +2023-10-05 20:45:11,178 - Epoch: [6][ 20/ 1236] Overall Loss 0.509634 Objective Loss 0.509634 LR 0.001000 Time 0.055526 +2023-10-05 20:45:11,345 - Epoch: [6][ 30/ 1236] Overall Loss 0.495766 Objective Loss 0.495766 LR 0.001000 Time 0.042595 +2023-10-05 20:45:11,511 - Epoch: [6][ 40/ 1236] Overall Loss 0.506478 Objective Loss 0.506478 LR 0.001000 Time 0.036086 +2023-10-05 20:45:11,675 - Epoch: [6][ 50/ 1236] Overall Loss 0.515040 Objective Loss 0.515040 LR 0.001000 Time 0.032143 +2023-10-05 20:45:11,841 - Epoch: [6][ 60/ 1236] Overall Loss 0.511437 Objective Loss 0.511437 LR 0.001000 Time 0.029550 +2023-10-05 20:45:12,005 - Epoch: [6][ 70/ 1236] Overall Loss 0.510906 Objective Loss 0.510906 LR 0.001000 Time 0.027665 +2023-10-05 20:45:12,171 - Epoch: [6][ 80/ 1236] Overall Loss 0.512215 Objective Loss 0.512215 LR 0.001000 Time 0.026275 +2023-10-05 20:45:12,335 - Epoch: [6][ 90/ 1236] Overall Loss 0.507937 Objective Loss 0.507937 LR 0.001000 Time 0.025181 +2023-10-05 20:45:12,501 - Epoch: [6][ 100/ 1236] Overall Loss 0.509655 Objective Loss 0.509655 LR 0.001000 Time 0.024315 +2023-10-05 20:45:12,665 - Epoch: [6][ 110/ 1236] Overall Loss 0.508797 Objective Loss 0.508797 LR 0.001000 Time 0.023594 +2023-10-05 20:45:12,831 - Epoch: [6][ 120/ 1236] Overall Loss 0.509617 Objective Loss 0.509617 LR 0.001000 Time 0.023007 +2023-10-05 20:45:12,995 - Epoch: [6][ 130/ 1236] Overall Loss 0.509071 Objective Loss 0.509071 LR 0.001000 Time 0.022498 +2023-10-05 20:45:13,161 - Epoch: [6][ 140/ 1236] Overall Loss 0.505611 Objective Loss 0.505611 LR 0.001000 Time 0.022073 +2023-10-05 20:45:13,325 - Epoch: [6][ 150/ 1236] Overall Loss 0.506679 Objective Loss 0.506679 LR 0.001000 Time 0.021690 +2023-10-05 20:45:13,490 - Epoch: [6][ 160/ 1236] Overall Loss 0.503838 Objective Loss 0.503838 LR 0.001000 Time 0.021364 +2023-10-05 20:45:13,654 - Epoch: [6][ 170/ 1236] Overall Loss 0.503384 Objective Loss 0.503384 LR 0.001000 Time 0.021071 +2023-10-05 20:45:13,819 - Epoch: [6][ 180/ 1236] Overall Loss 0.506007 Objective Loss 0.506007 LR 0.001000 Time 0.020818 +2023-10-05 20:45:13,983 - Epoch: [6][ 190/ 1236] Overall Loss 0.506019 Objective Loss 0.506019 LR 0.001000 Time 0.020584 +2023-10-05 20:45:14,149 - Epoch: [6][ 200/ 1236] Overall Loss 0.506471 Objective Loss 0.506471 LR 0.001000 Time 0.020383 +2023-10-05 20:45:14,313 - Epoch: [6][ 210/ 1236] Overall Loss 0.509590 Objective Loss 0.509590 LR 0.001000 Time 0.020191 +2023-10-05 20:45:14,479 - Epoch: [6][ 220/ 1236] Overall Loss 0.511157 Objective Loss 0.511157 LR 0.001000 Time 0.020024 +2023-10-05 20:45:14,643 - Epoch: [6][ 230/ 1236] Overall Loss 0.511001 Objective Loss 0.511001 LR 0.001000 Time 0.019866 +2023-10-05 20:45:14,808 - Epoch: [6][ 240/ 1236] Overall Loss 0.511773 Objective Loss 0.511773 LR 0.001000 Time 0.019727 +2023-10-05 20:45:14,973 - Epoch: [6][ 250/ 1236] Overall Loss 0.512428 Objective Loss 0.512428 LR 0.001000 Time 0.019594 +2023-10-05 20:45:15,138 - Epoch: [6][ 260/ 1236] Overall Loss 0.513105 Objective Loss 0.513105 LR 0.001000 Time 0.019476 +2023-10-05 20:45:15,301 - Epoch: [6][ 270/ 1236] Overall Loss 0.512656 Objective Loss 0.512656 LR 0.001000 Time 0.019358 +2023-10-05 20:45:15,466 - Epoch: [6][ 280/ 1236] Overall Loss 0.513978 Objective Loss 0.513978 LR 0.001000 Time 0.019254 +2023-10-05 20:45:15,630 - Epoch: [6][ 290/ 1236] Overall Loss 0.514240 Objective Loss 0.514240 LR 0.001000 Time 0.019153 +2023-10-05 20:45:15,794 - Epoch: [6][ 300/ 1236] Overall Loss 0.516326 Objective Loss 0.516326 LR 0.001000 Time 0.019062 +2023-10-05 20:45:15,958 - Epoch: [6][ 310/ 1236] Overall Loss 0.516243 Objective Loss 0.516243 LR 0.001000 Time 0.018973 +2023-10-05 20:45:16,123 - Epoch: [6][ 320/ 1236] Overall Loss 0.516861 Objective Loss 0.516861 LR 0.001000 Time 0.018897 +2023-10-05 20:45:16,287 - Epoch: [6][ 330/ 1236] Overall Loss 0.516849 Objective Loss 0.516849 LR 0.001000 Time 0.018818 +2023-10-05 20:45:16,451 - Epoch: [6][ 340/ 1236] Overall Loss 0.516364 Objective Loss 0.516364 LR 0.001000 Time 0.018748 +2023-10-05 20:45:16,615 - Epoch: [6][ 350/ 1236] Overall Loss 0.515812 Objective Loss 0.515812 LR 0.001000 Time 0.018678 +2023-10-05 20:45:16,779 - Epoch: [6][ 360/ 1236] Overall Loss 0.515210 Objective Loss 0.515210 LR 0.001000 Time 0.018616 +2023-10-05 20:45:16,943 - Epoch: [6][ 370/ 1236] Overall Loss 0.515258 Objective Loss 0.515258 LR 0.001000 Time 0.018554 +2023-10-05 20:45:17,108 - Epoch: [6][ 380/ 1236] Overall Loss 0.516275 Objective Loss 0.516275 LR 0.001000 Time 0.018499 +2023-10-05 20:45:17,272 - Epoch: [6][ 390/ 1236] Overall Loss 0.516708 Objective Loss 0.516708 LR 0.001000 Time 0.018444 +2023-10-05 20:45:17,437 - Epoch: [6][ 400/ 1236] Overall Loss 0.517017 Objective Loss 0.517017 LR 0.001000 Time 0.018395 +2023-10-05 20:45:17,601 - Epoch: [6][ 410/ 1236] Overall Loss 0.517182 Objective Loss 0.517182 LR 0.001000 Time 0.018345 +2023-10-05 20:45:17,766 - Epoch: [6][ 420/ 1236] Overall Loss 0.517519 Objective Loss 0.517519 LR 0.001000 Time 0.018301 +2023-10-05 20:45:17,929 - Epoch: [6][ 430/ 1236] Overall Loss 0.517278 Objective Loss 0.517278 LR 0.001000 Time 0.018255 +2023-10-05 20:45:18,095 - Epoch: [6][ 440/ 1236] Overall Loss 0.517137 Objective Loss 0.517137 LR 0.001000 Time 0.018215 +2023-10-05 20:45:18,259 - Epoch: [6][ 450/ 1236] Overall Loss 0.517430 Objective Loss 0.517430 LR 0.001000 Time 0.018174 +2023-10-05 20:45:18,424 - Epoch: [6][ 460/ 1236] Overall Loss 0.516890 Objective Loss 0.516890 LR 0.001000 Time 0.018138 +2023-10-05 20:45:18,589 - Epoch: [6][ 470/ 1236] Overall Loss 0.516913 Objective Loss 0.516913 LR 0.001000 Time 0.018101 +2023-10-05 20:45:18,754 - Epoch: [6][ 480/ 1236] Overall Loss 0.517074 Objective Loss 0.517074 LR 0.001000 Time 0.018069 +2023-10-05 20:45:18,919 - Epoch: [6][ 490/ 1236] Overall Loss 0.517321 Objective Loss 0.517321 LR 0.001000 Time 0.018034 +2023-10-05 20:45:19,084 - Epoch: [6][ 500/ 1236] Overall Loss 0.518203 Objective Loss 0.518203 LR 0.001000 Time 0.018005 +2023-10-05 20:45:19,249 - Epoch: [6][ 510/ 1236] Overall Loss 0.518308 Objective Loss 0.518308 LR 0.001000 Time 0.017973 +2023-10-05 20:45:19,414 - Epoch: [6][ 520/ 1236] Overall Loss 0.518104 Objective Loss 0.518104 LR 0.001000 Time 0.017946 +2023-10-05 20:45:19,579 - Epoch: [6][ 530/ 1236] Overall Loss 0.517895 Objective Loss 0.517895 LR 0.001000 Time 0.017916 +2023-10-05 20:45:19,744 - Epoch: [6][ 540/ 1236] Overall Loss 0.518315 Objective Loss 0.518315 LR 0.001000 Time 0.017891 +2023-10-05 20:45:19,908 - Epoch: [6][ 550/ 1236] Overall Loss 0.518215 Objective Loss 0.518215 LR 0.001000 Time 0.017863 +2023-10-05 20:45:20,074 - Epoch: [6][ 560/ 1236] Overall Loss 0.517782 Objective Loss 0.517782 LR 0.001000 Time 0.017840 +2023-10-05 20:45:20,238 - Epoch: [6][ 570/ 1236] Overall Loss 0.517378 Objective Loss 0.517378 LR 0.001000 Time 0.017814 +2023-10-05 20:45:20,404 - Epoch: [6][ 580/ 1236] Overall Loss 0.517549 Objective Loss 0.517549 LR 0.001000 Time 0.017793 +2023-10-05 20:45:20,569 - Epoch: [6][ 590/ 1236] Overall Loss 0.518524 Objective Loss 0.518524 LR 0.001000 Time 0.017769 +2023-10-05 20:45:20,734 - Epoch: [6][ 600/ 1236] Overall Loss 0.518791 Objective Loss 0.518791 LR 0.001000 Time 0.017749 +2023-10-05 20:45:20,899 - Epoch: [6][ 610/ 1236] Overall Loss 0.519252 Objective Loss 0.519252 LR 0.001000 Time 0.017726 +2023-10-05 20:45:21,064 - Epoch: [6][ 620/ 1236] Overall Loss 0.518876 Objective Loss 0.518876 LR 0.001000 Time 0.017707 +2023-10-05 20:45:21,228 - Epoch: [6][ 630/ 1236] Overall Loss 0.518872 Objective Loss 0.518872 LR 0.001000 Time 0.017686 +2023-10-05 20:45:21,394 - Epoch: [6][ 640/ 1236] Overall Loss 0.519212 Objective Loss 0.519212 LR 0.001000 Time 0.017669 +2023-10-05 20:45:21,559 - Epoch: [6][ 650/ 1236] Overall Loss 0.518534 Objective Loss 0.518534 LR 0.001000 Time 0.017649 +2023-10-05 20:45:21,724 - Epoch: [6][ 660/ 1236] Overall Loss 0.518345 Objective Loss 0.518345 LR 0.001000 Time 0.017632 +2023-10-05 20:45:21,889 - Epoch: [6][ 670/ 1236] Overall Loss 0.518263 Objective Loss 0.518263 LR 0.001000 Time 0.017614 +2023-10-05 20:45:22,054 - Epoch: [6][ 680/ 1236] Overall Loss 0.518517 Objective Loss 0.518517 LR 0.001000 Time 0.017598 +2023-10-05 20:45:22,218 - Epoch: [6][ 690/ 1236] Overall Loss 0.517807 Objective Loss 0.517807 LR 0.001000 Time 0.017580 +2023-10-05 20:45:22,384 - Epoch: [6][ 700/ 1236] Overall Loss 0.517437 Objective Loss 0.517437 LR 0.001000 Time 0.017566 +2023-10-05 20:45:22,549 - Epoch: [6][ 710/ 1236] Overall Loss 0.517390 Objective Loss 0.517390 LR 0.001000 Time 0.017549 +2023-10-05 20:45:22,714 - Epoch: [6][ 720/ 1236] Overall Loss 0.516841 Objective Loss 0.516841 LR 0.001000 Time 0.017535 +2023-10-05 20:45:22,879 - Epoch: [6][ 730/ 1236] Overall Loss 0.516551 Objective Loss 0.516551 LR 0.001000 Time 0.017520 +2023-10-05 20:45:23,044 - Epoch: [6][ 740/ 1236] Overall Loss 0.515845 Objective Loss 0.515845 LR 0.001000 Time 0.017506 +2023-10-05 20:45:23,208 - Epoch: [6][ 750/ 1236] Overall Loss 0.515981 Objective Loss 0.515981 LR 0.001000 Time 0.017491 +2023-10-05 20:45:23,374 - Epoch: [6][ 760/ 1236] Overall Loss 0.515417 Objective Loss 0.515417 LR 0.001000 Time 0.017479 +2023-10-05 20:45:23,539 - Epoch: [6][ 770/ 1236] Overall Loss 0.515382 Objective Loss 0.515382 LR 0.001000 Time 0.017465 +2023-10-05 20:45:23,704 - Epoch: [6][ 780/ 1236] Overall Loss 0.515081 Objective Loss 0.515081 LR 0.001000 Time 0.017453 +2023-10-05 20:45:23,869 - Epoch: [6][ 790/ 1236] Overall Loss 0.515507 Objective Loss 0.515507 LR 0.001000 Time 0.017440 +2023-10-05 20:45:24,034 - Epoch: [6][ 800/ 1236] Overall Loss 0.515758 Objective Loss 0.515758 LR 0.001000 Time 0.017429 +2023-10-05 20:45:24,199 - Epoch: [6][ 810/ 1236] Overall Loss 0.514916 Objective Loss 0.514916 LR 0.001000 Time 0.017416 +2023-10-05 20:45:24,364 - Epoch: [6][ 820/ 1236] Overall Loss 0.515107 Objective Loss 0.515107 LR 0.001000 Time 0.017405 +2023-10-05 20:45:24,529 - Epoch: [6][ 830/ 1236] Overall Loss 0.515241 Objective Loss 0.515241 LR 0.001000 Time 0.017393 +2023-10-05 20:45:24,694 - Epoch: [6][ 840/ 1236] Overall Loss 0.515578 Objective Loss 0.515578 LR 0.001000 Time 0.017383 +2023-10-05 20:45:24,859 - Epoch: [6][ 850/ 1236] Overall Loss 0.515248 Objective Loss 0.515248 LR 0.001000 Time 0.017371 +2023-10-05 20:45:25,024 - Epoch: [6][ 860/ 1236] Overall Loss 0.514907 Objective Loss 0.514907 LR 0.001000 Time 0.017362 +2023-10-05 20:45:25,189 - Epoch: [6][ 870/ 1236] Overall Loss 0.514873 Objective Loss 0.514873 LR 0.001000 Time 0.017351 +2023-10-05 20:45:25,354 - Epoch: [6][ 880/ 1236] Overall Loss 0.514718 Objective Loss 0.514718 LR 0.001000 Time 0.017341 +2023-10-05 20:45:25,519 - Epoch: [6][ 890/ 1236] Overall Loss 0.514659 Objective Loss 0.514659 LR 0.001000 Time 0.017331 +2023-10-05 20:45:25,684 - Epoch: [6][ 900/ 1236] Overall Loss 0.514567 Objective Loss 0.514567 LR 0.001000 Time 0.017322 +2023-10-05 20:45:25,849 - Epoch: [6][ 910/ 1236] Overall Loss 0.514207 Objective Loss 0.514207 LR 0.001000 Time 0.017312 +2023-10-05 20:45:26,014 - Epoch: [6][ 920/ 1236] Overall Loss 0.513714 Objective Loss 0.513714 LR 0.001000 Time 0.017303 +2023-10-05 20:45:26,179 - Epoch: [6][ 930/ 1236] Overall Loss 0.513557 Objective Loss 0.513557 LR 0.001000 Time 0.017294 +2023-10-05 20:45:26,344 - Epoch: [6][ 940/ 1236] Overall Loss 0.513254 Objective Loss 0.513254 LR 0.001000 Time 0.017286 +2023-10-05 20:45:26,509 - Epoch: [6][ 950/ 1236] Overall Loss 0.513259 Objective Loss 0.513259 LR 0.001000 Time 0.017276 +2023-10-05 20:45:26,674 - Epoch: [6][ 960/ 1236] Overall Loss 0.512748 Objective Loss 0.512748 LR 0.001000 Time 0.017269 +2023-10-05 20:45:26,839 - Epoch: [6][ 970/ 1236] Overall Loss 0.512485 Objective Loss 0.512485 LR 0.001000 Time 0.017260 +2023-10-05 20:45:27,010 - Epoch: [6][ 980/ 1236] Overall Loss 0.512333 Objective Loss 0.512333 LR 0.001000 Time 0.017258 +2023-10-05 20:45:27,179 - Epoch: [6][ 990/ 1236] Overall Loss 0.512430 Objective Loss 0.512430 LR 0.001000 Time 0.017254 +2023-10-05 20:45:27,353 - Epoch: [6][ 1000/ 1236] Overall Loss 0.512632 Objective Loss 0.512632 LR 0.001000 Time 0.017255 +2023-10-05 20:45:27,523 - Epoch: [6][ 1010/ 1236] Overall Loss 0.512021 Objective Loss 0.512021 LR 0.001000 Time 0.017253 +2023-10-05 20:45:27,689 - Epoch: [6][ 1020/ 1236] Overall Loss 0.511783 Objective Loss 0.511783 LR 0.001000 Time 0.017246 +2023-10-05 20:45:27,854 - Epoch: [6][ 1030/ 1236] Overall Loss 0.511670 Objective Loss 0.511670 LR 0.001000 Time 0.017238 +2023-10-05 20:45:28,019 - Epoch: [6][ 1040/ 1236] Overall Loss 0.511625 Objective Loss 0.511625 LR 0.001000 Time 0.017231 +2023-10-05 20:45:28,184 - Epoch: [6][ 1050/ 1236] Overall Loss 0.511647 Objective Loss 0.511647 LR 0.001000 Time 0.017223 +2023-10-05 20:45:28,349 - Epoch: [6][ 1060/ 1236] Overall Loss 0.511601 Objective Loss 0.511601 LR 0.001000 Time 0.017217 +2023-10-05 20:45:28,514 - Epoch: [6][ 1070/ 1236] Overall Loss 0.511325 Objective Loss 0.511325 LR 0.001000 Time 0.017209 +2023-10-05 20:45:28,680 - Epoch: [6][ 1080/ 1236] Overall Loss 0.511247 Objective Loss 0.511247 LR 0.001000 Time 0.017203 +2023-10-05 20:45:28,844 - Epoch: [6][ 1090/ 1236] Overall Loss 0.511349 Objective Loss 0.511349 LR 0.001000 Time 0.017195 +2023-10-05 20:45:29,009 - Epoch: [6][ 1100/ 1236] Overall Loss 0.511138 Objective Loss 0.511138 LR 0.001000 Time 0.017190 +2023-10-05 20:45:29,174 - Epoch: [6][ 1110/ 1236] Overall Loss 0.510638 Objective Loss 0.510638 LR 0.001000 Time 0.017182 +2023-10-05 20:45:29,340 - Epoch: [6][ 1120/ 1236] Overall Loss 0.510050 Objective Loss 0.510050 LR 0.001000 Time 0.017177 +2023-10-05 20:45:29,504 - Epoch: [6][ 1130/ 1236] Overall Loss 0.509677 Objective Loss 0.509677 LR 0.001000 Time 0.017170 +2023-10-05 20:45:29,670 - Epoch: [6][ 1140/ 1236] Overall Loss 0.509250 Objective Loss 0.509250 LR 0.001000 Time 0.017164 +2023-10-05 20:45:29,834 - Epoch: [6][ 1150/ 1236] Overall Loss 0.508921 Objective Loss 0.508921 LR 0.001000 Time 0.017157 +2023-10-05 20:45:30,000 - Epoch: [6][ 1160/ 1236] Overall Loss 0.508528 Objective Loss 0.508528 LR 0.001000 Time 0.017153 +2023-10-05 20:45:30,164 - Epoch: [6][ 1170/ 1236] Overall Loss 0.508779 Objective Loss 0.508779 LR 0.001000 Time 0.017146 +2023-10-05 20:45:30,330 - Epoch: [6][ 1180/ 1236] Overall Loss 0.508637 Objective Loss 0.508637 LR 0.001000 Time 0.017141 +2023-10-05 20:45:30,494 - Epoch: [6][ 1190/ 1236] Overall Loss 0.508678 Objective Loss 0.508678 LR 0.001000 Time 0.017134 +2023-10-05 20:45:30,660 - Epoch: [6][ 1200/ 1236] Overall Loss 0.508833 Objective Loss 0.508833 LR 0.001000 Time 0.017129 +2023-10-05 20:45:30,824 - Epoch: [6][ 1210/ 1236] Overall Loss 0.508831 Objective Loss 0.508831 LR 0.001000 Time 0.017123 +2023-10-05 20:45:30,990 - Epoch: [6][ 1220/ 1236] Overall Loss 0.508851 Objective Loss 0.508851 LR 0.001000 Time 0.017119 +2023-10-05 20:45:31,200 - Epoch: [6][ 1230/ 1236] Overall Loss 0.508725 Objective Loss 0.508725 LR 0.001000 Time 0.017150 +2023-10-05 20:45:31,290 - Epoch: [6][ 1236/ 1236] Overall Loss 0.508712 Objective Loss 0.508712 Top1 74.338086 Top5 96.334012 LR 0.001000 Time 0.017140 +2023-10-05 20:45:31,414 - --- validate (epoch=6)----------- +2023-10-05 20:45:31,414 - 29943 samples (256 per mini-batch) +2023-10-05 20:45:31,815 - Epoch: [6][ 10/ 117] Loss 0.487663 Top1 73.554688 Top5 96.015625 +2023-10-05 20:45:31,926 - Epoch: [6][ 20/ 117] Loss 0.496493 Top1 72.714844 Top5 96.113281 +2023-10-05 20:45:32,036 - Epoch: [6][ 30/ 117] Loss 0.489672 Top1 73.072917 Top5 96.197917 +2023-10-05 20:45:32,146 - Epoch: [6][ 40/ 117] Loss 0.490442 Top1 73.437500 Top5 96.015625 +2023-10-05 20:45:32,256 - Epoch: [6][ 50/ 117] Loss 0.491356 Top1 73.312500 Top5 96.101562 +2023-10-05 20:45:32,365 - Epoch: [6][ 60/ 117] Loss 0.489591 Top1 73.541667 Top5 96.184896 +2023-10-05 20:45:32,476 - Epoch: [6][ 70/ 117] Loss 0.486657 Top1 73.733259 Top5 96.216518 +2023-10-05 20:45:32,585 - Epoch: [6][ 80/ 117] Loss 0.487605 Top1 73.735352 Top5 96.230469 +2023-10-05 20:45:32,696 - Epoch: [6][ 90/ 117] Loss 0.486389 Top1 73.767361 Top5 96.228299 +2023-10-05 20:45:32,804 - Epoch: [6][ 100/ 117] Loss 0.486495 Top1 73.777344 Top5 96.191406 +2023-10-05 20:45:32,925 - Epoch: [6][ 110/ 117] Loss 0.482047 Top1 73.920455 Top5 96.242898 +2023-10-05 20:45:32,980 - Epoch: [6][ 117/ 117] Loss 0.480760 Top1 73.970544 Top5 96.276258 +2023-10-05 20:45:33,070 - ==> Top1: 73.971 Top5: 96.276 Loss: 0.481 + +2023-10-05 20:45:33,070 - ==> Confusion: +[[ 898 2 13 0 13 7 0 0 1 75 2 1 1 10 3 7 4 0 0 1 12] + [ 2 972 4 2 11 67 4 34 4 0 3 0 0 0 1 3 10 0 3 3 8] + [ 7 0 874 12 5 2 92 11 0 0 6 5 4 3 1 2 1 2 8 6 15] + [ 2 2 32 962 1 9 8 2 6 0 7 0 4 6 16 1 1 13 7 0 10] + [ 32 15 4 0 937 7 2 0 0 8 0 0 0 3 7 16 14 0 1 0 4] + [ 11 42 11 1 1 954 2 22 1 1 3 12 1 14 3 5 4 1 2 14 11] + [ 0 5 26 1 0 1 1127 11 0 0 0 1 0 0 0 6 0 1 2 8 2] + [ 10 12 28 1 6 80 18 987 2 1 3 3 1 0 0 4 0 3 38 15 6] + [ 33 5 0 0 2 6 2 0 868 82 21 1 1 35 21 2 1 3 2 1 3] + [ 146 1 1 0 15 4 1 1 21 856 1 0 0 48 2 2 1 0 2 2 15] + [ 5 5 22 30 4 5 6 6 9 2 925 1 1 4 4 5 0 1 6 2 10] + [ 2 2 0 0 0 29 2 0 0 0 0 891 37 3 0 12 1 18 0 32 6] + [ 1 0 2 11 2 11 2 4 0 1 0 51 917 0 0 22 3 26 0 8 7] + [ 7 0 5 1 4 38 2 2 2 10 17 10 2 976 1 6 3 1 0 17 15] + [ 23 6 3 43 19 1 0 0 23 7 3 0 5 2 923 1 5 4 11 0 22] + [ 0 2 3 1 6 3 8 0 0 1 0 12 5 0 0 1051 10 9 1 7 15] + [ 1 19 3 0 10 7 3 0 2 0 0 7 0 3 1 27 1051 0 1 6 20] + [ 0 0 0 9 1 1 11 0 2 0 0 8 41 0 1 20 2 936 0 3 3] + [ 2 13 18 35 1 6 1 49 4 1 5 0 5 0 10 0 2 0 898 3 15] + [ 2 3 0 0 1 9 25 8 0 0 0 13 5 0 0 5 4 4 0 1067 6] + [ 267 333 333 221 157 494 142 122 71 103 159 149 495 374 148 204 307 88 186 473 3079]] + +2023-10-05 20:45:33,071 - ==> Best [Top1: 73.971 Top5: 96.276 Sparsity:0.00 Params: 148928 on epoch: 6] +2023-10-05 20:45:33,072 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:45:33,085 - + +2023-10-05 20:45:33,085 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:45:34,015 - Epoch: [7][ 10/ 1236] Overall Loss 0.500898 Objective Loss 0.500898 LR 0.001000 Time 0.093019 +2023-10-05 20:45:34,180 - Epoch: [7][ 20/ 1236] Overall Loss 0.484873 Objective Loss 0.484873 LR 0.001000 Time 0.054732 +2023-10-05 20:45:34,343 - Epoch: [7][ 30/ 1236] Overall Loss 0.472571 Objective Loss 0.472571 LR 0.001000 Time 0.041900 +2023-10-05 20:45:34,505 - Epoch: [7][ 40/ 1236] Overall Loss 0.483425 Objective Loss 0.483425 LR 0.001000 Time 0.035482 +2023-10-05 20:45:34,669 - Epoch: [7][ 50/ 1236] Overall Loss 0.480169 Objective Loss 0.480169 LR 0.001000 Time 0.031652 +2023-10-05 20:45:34,833 - Epoch: [7][ 60/ 1236] Overall Loss 0.487163 Objective Loss 0.487163 LR 0.001000 Time 0.029097 +2023-10-05 20:45:34,995 - Epoch: [7][ 70/ 1236] Overall Loss 0.482336 Objective Loss 0.482336 LR 0.001000 Time 0.027255 +2023-10-05 20:45:35,159 - Epoch: [7][ 80/ 1236] Overall Loss 0.481089 Objective Loss 0.481089 LR 0.001000 Time 0.025892 +2023-10-05 20:45:35,321 - Epoch: [7][ 90/ 1236] Overall Loss 0.481413 Objective Loss 0.481413 LR 0.001000 Time 0.024812 +2023-10-05 20:45:35,484 - Epoch: [7][ 100/ 1236] Overall Loss 0.479652 Objective Loss 0.479652 LR 0.001000 Time 0.023959 +2023-10-05 20:45:35,646 - Epoch: [7][ 110/ 1236] Overall Loss 0.476215 Objective Loss 0.476215 LR 0.001000 Time 0.023250 +2023-10-05 20:45:35,810 - Epoch: [7][ 120/ 1236] Overall Loss 0.474344 Objective Loss 0.474344 LR 0.001000 Time 0.022679 +2023-10-05 20:45:35,973 - Epoch: [7][ 130/ 1236] Overall Loss 0.473492 Objective Loss 0.473492 LR 0.001000 Time 0.022184 +2023-10-05 20:45:36,136 - Epoch: [7][ 140/ 1236] Overall Loss 0.471967 Objective Loss 0.471967 LR 0.001000 Time 0.021762 +2023-10-05 20:45:36,298 - Epoch: [7][ 150/ 1236] Overall Loss 0.471116 Objective Loss 0.471116 LR 0.001000 Time 0.021389 +2023-10-05 20:45:36,461 - Epoch: [7][ 160/ 1236] Overall Loss 0.472696 Objective Loss 0.472696 LR 0.001000 Time 0.021068 +2023-10-05 20:45:36,623 - Epoch: [7][ 170/ 1236] Overall Loss 0.473620 Objective Loss 0.473620 LR 0.001000 Time 0.020780 +2023-10-05 20:45:36,786 - Epoch: [7][ 180/ 1236] Overall Loss 0.475262 Objective Loss 0.475262 LR 0.001000 Time 0.020529 +2023-10-05 20:45:36,948 - Epoch: [7][ 190/ 1236] Overall Loss 0.474138 Objective Loss 0.474138 LR 0.001000 Time 0.020300 +2023-10-05 20:45:37,111 - Epoch: [7][ 200/ 1236] Overall Loss 0.471726 Objective Loss 0.471726 LR 0.001000 Time 0.020099 +2023-10-05 20:45:37,273 - Epoch: [7][ 210/ 1236] Overall Loss 0.472080 Objective Loss 0.472080 LR 0.001000 Time 0.019912 +2023-10-05 20:45:37,436 - Epoch: [7][ 220/ 1236] Overall Loss 0.472213 Objective Loss 0.472213 LR 0.001000 Time 0.019747 +2023-10-05 20:45:37,598 - Epoch: [7][ 230/ 1236] Overall Loss 0.472963 Objective Loss 0.472963 LR 0.001000 Time 0.019592 +2023-10-05 20:45:37,762 - Epoch: [7][ 240/ 1236] Overall Loss 0.472114 Objective Loss 0.472114 LR 0.001000 Time 0.019455 +2023-10-05 20:45:37,924 - Epoch: [7][ 250/ 1236] Overall Loss 0.472745 Objective Loss 0.472745 LR 0.001000 Time 0.019325 +2023-10-05 20:45:38,087 - Epoch: [7][ 260/ 1236] Overall Loss 0.474433 Objective Loss 0.474433 LR 0.001000 Time 0.019208 +2023-10-05 20:45:38,249 - Epoch: [7][ 270/ 1236] Overall Loss 0.474101 Objective Loss 0.474101 LR 0.001000 Time 0.019097 +2023-10-05 20:45:38,414 - Epoch: [7][ 280/ 1236] Overall Loss 0.474073 Objective Loss 0.474073 LR 0.001000 Time 0.019002 +2023-10-05 20:45:38,577 - Epoch: [7][ 290/ 1236] Overall Loss 0.475255 Objective Loss 0.475255 LR 0.001000 Time 0.018907 +2023-10-05 20:45:38,741 - Epoch: [7][ 300/ 1236] Overall Loss 0.473970 Objective Loss 0.473970 LR 0.001000 Time 0.018824 +2023-10-05 20:45:38,905 - Epoch: [7][ 310/ 1236] Overall Loss 0.474714 Objective Loss 0.474714 LR 0.001000 Time 0.018743 +2023-10-05 20:45:39,069 - Epoch: [7][ 320/ 1236] Overall Loss 0.474156 Objective Loss 0.474156 LR 0.001000 Time 0.018668 +2023-10-05 20:45:39,231 - Epoch: [7][ 330/ 1236] Overall Loss 0.475206 Objective Loss 0.475206 LR 0.001000 Time 0.018594 +2023-10-05 20:45:39,396 - Epoch: [7][ 340/ 1236] Overall Loss 0.474178 Objective Loss 0.474178 LR 0.001000 Time 0.018531 +2023-10-05 20:45:39,559 - Epoch: [7][ 350/ 1236] Overall Loss 0.473292 Objective Loss 0.473292 LR 0.001000 Time 0.018466 +2023-10-05 20:45:39,723 - Epoch: [7][ 360/ 1236] Overall Loss 0.474175 Objective Loss 0.474175 LR 0.001000 Time 0.018409 +2023-10-05 20:45:39,886 - Epoch: [7][ 370/ 1236] Overall Loss 0.474117 Objective Loss 0.474117 LR 0.001000 Time 0.018351 +2023-10-05 20:45:40,051 - Epoch: [7][ 380/ 1236] Overall Loss 0.475094 Objective Loss 0.475094 LR 0.001000 Time 0.018301 +2023-10-05 20:45:40,214 - Epoch: [7][ 390/ 1236] Overall Loss 0.474315 Objective Loss 0.474315 LR 0.001000 Time 0.018249 +2023-10-05 20:45:40,379 - Epoch: [7][ 400/ 1236] Overall Loss 0.473795 Objective Loss 0.473795 LR 0.001000 Time 0.018204 +2023-10-05 20:45:40,542 - Epoch: [7][ 410/ 1236] Overall Loss 0.474014 Objective Loss 0.474014 LR 0.001000 Time 0.018158 +2023-10-05 20:45:40,707 - Epoch: [7][ 420/ 1236] Overall Loss 0.474958 Objective Loss 0.474958 LR 0.001000 Time 0.018118 +2023-10-05 20:45:40,871 - Epoch: [7][ 430/ 1236] Overall Loss 0.475119 Objective Loss 0.475119 LR 0.001000 Time 0.018075 +2023-10-05 20:45:41,035 - Epoch: [7][ 440/ 1236] Overall Loss 0.475338 Objective Loss 0.475338 LR 0.001000 Time 0.018038 +2023-10-05 20:45:41,198 - Epoch: [7][ 450/ 1236] Overall Loss 0.475312 Objective Loss 0.475312 LR 0.001000 Time 0.017998 +2023-10-05 20:45:41,362 - Epoch: [7][ 460/ 1236] Overall Loss 0.474864 Objective Loss 0.474864 LR 0.001000 Time 0.017963 +2023-10-05 20:45:41,525 - Epoch: [7][ 470/ 1236] Overall Loss 0.474140 Objective Loss 0.474140 LR 0.001000 Time 0.017926 +2023-10-05 20:45:41,688 - Epoch: [7][ 480/ 1236] Overall Loss 0.475197 Objective Loss 0.475197 LR 0.001000 Time 0.017893 +2023-10-05 20:45:41,851 - Epoch: [7][ 490/ 1236] Overall Loss 0.474751 Objective Loss 0.474751 LR 0.001000 Time 0.017859 +2023-10-05 20:45:42,014 - Epoch: [7][ 500/ 1236] Overall Loss 0.474965 Objective Loss 0.474965 LR 0.001000 Time 0.017827 +2023-10-05 20:45:42,176 - Epoch: [7][ 510/ 1236] Overall Loss 0.474883 Objective Loss 0.474883 LR 0.001000 Time 0.017796 +2023-10-05 20:45:42,340 - Epoch: [7][ 520/ 1236] Overall Loss 0.473872 Objective Loss 0.473872 LR 0.001000 Time 0.017767 +2023-10-05 20:45:42,502 - Epoch: [7][ 530/ 1236] Overall Loss 0.475580 Objective Loss 0.475580 LR 0.001000 Time 0.017737 +2023-10-05 20:45:42,666 - Epoch: [7][ 540/ 1236] Overall Loss 0.474977 Objective Loss 0.474977 LR 0.001000 Time 0.017712 +2023-10-05 20:45:42,830 - Epoch: [7][ 550/ 1236] Overall Loss 0.475530 Objective Loss 0.475530 LR 0.001000 Time 0.017688 +2023-10-05 20:45:42,996 - Epoch: [7][ 560/ 1236] Overall Loss 0.475889 Objective Loss 0.475889 LR 0.001000 Time 0.017667 +2023-10-05 20:45:43,161 - Epoch: [7][ 570/ 1236] Overall Loss 0.476890 Objective Loss 0.476890 LR 0.001000 Time 0.017645 +2023-10-05 20:45:43,326 - Epoch: [7][ 580/ 1236] Overall Loss 0.477517 Objective Loss 0.477517 LR 0.001000 Time 0.017626 +2023-10-05 20:45:43,491 - Epoch: [7][ 590/ 1236] Overall Loss 0.477945 Objective Loss 0.477945 LR 0.001000 Time 0.017606 +2023-10-05 20:45:43,656 - Epoch: [7][ 600/ 1236] Overall Loss 0.478023 Objective Loss 0.478023 LR 0.001000 Time 0.017588 +2023-10-05 20:45:43,821 - Epoch: [7][ 610/ 1236] Overall Loss 0.478090 Objective Loss 0.478090 LR 0.001000 Time 0.017569 +2023-10-05 20:45:43,986 - Epoch: [7][ 620/ 1236] Overall Loss 0.478382 Objective Loss 0.478382 LR 0.001000 Time 0.017552 +2023-10-05 20:45:44,151 - Epoch: [7][ 630/ 1236] Overall Loss 0.478127 Objective Loss 0.478127 LR 0.001000 Time 0.017534 +2023-10-05 20:45:44,316 - Epoch: [7][ 640/ 1236] Overall Loss 0.478479 Objective Loss 0.478479 LR 0.001000 Time 0.017518 +2023-10-05 20:45:44,481 - Epoch: [7][ 650/ 1236] Overall Loss 0.478683 Objective Loss 0.478683 LR 0.001000 Time 0.017501 +2023-10-05 20:45:44,646 - Epoch: [7][ 660/ 1236] Overall Loss 0.478511 Objective Loss 0.478511 LR 0.001000 Time 0.017486 +2023-10-05 20:45:44,811 - Epoch: [7][ 670/ 1236] Overall Loss 0.478315 Objective Loss 0.478315 LR 0.001000 Time 0.017470 +2023-10-05 20:45:44,976 - Epoch: [7][ 680/ 1236] Overall Loss 0.477758 Objective Loss 0.477758 LR 0.001000 Time 0.017456 +2023-10-05 20:45:45,141 - Epoch: [7][ 690/ 1236] Overall Loss 0.477443 Objective Loss 0.477443 LR 0.001000 Time 0.017441 +2023-10-05 20:45:45,306 - Epoch: [7][ 700/ 1236] Overall Loss 0.477456 Objective Loss 0.477456 LR 0.001000 Time 0.017428 +2023-10-05 20:45:45,471 - Epoch: [7][ 710/ 1236] Overall Loss 0.477580 Objective Loss 0.477580 LR 0.001000 Time 0.017414 +2023-10-05 20:45:45,636 - Epoch: [7][ 720/ 1236] Overall Loss 0.477779 Objective Loss 0.477779 LR 0.001000 Time 0.017402 +2023-10-05 20:45:45,801 - Epoch: [7][ 730/ 1236] Overall Loss 0.477451 Objective Loss 0.477451 LR 0.001000 Time 0.017388 +2023-10-05 20:45:45,966 - Epoch: [7][ 740/ 1236] Overall Loss 0.477443 Objective Loss 0.477443 LR 0.001000 Time 0.017376 +2023-10-05 20:45:46,131 - Epoch: [7][ 750/ 1236] Overall Loss 0.477578 Objective Loss 0.477578 LR 0.001000 Time 0.017363 +2023-10-05 20:45:46,296 - Epoch: [7][ 760/ 1236] Overall Loss 0.477404 Objective Loss 0.477404 LR 0.001000 Time 0.017352 +2023-10-05 20:45:46,461 - Epoch: [7][ 770/ 1236] Overall Loss 0.477761 Objective Loss 0.477761 LR 0.001000 Time 0.017340 +2023-10-05 20:45:46,626 - Epoch: [7][ 780/ 1236] Overall Loss 0.477733 Objective Loss 0.477733 LR 0.001000 Time 0.017329 +2023-10-05 20:45:46,790 - Epoch: [7][ 790/ 1236] Overall Loss 0.477675 Objective Loss 0.477675 LR 0.001000 Time 0.017318 +2023-10-05 20:45:46,956 - Epoch: [7][ 800/ 1236] Overall Loss 0.477699 Objective Loss 0.477699 LR 0.001000 Time 0.017308 +2023-10-05 20:45:47,121 - Epoch: [7][ 810/ 1236] Overall Loss 0.478041 Objective Loss 0.478041 LR 0.001000 Time 0.017297 +2023-10-05 20:45:47,286 - Epoch: [7][ 820/ 1236] Overall Loss 0.478399 Objective Loss 0.478399 LR 0.001000 Time 0.017287 +2023-10-05 20:45:47,451 - Epoch: [7][ 830/ 1236] Overall Loss 0.478778 Objective Loss 0.478778 LR 0.001000 Time 0.017277 +2023-10-05 20:45:47,616 - Epoch: [7][ 840/ 1236] Overall Loss 0.478968 Objective Loss 0.478968 LR 0.001000 Time 0.017268 +2023-10-05 20:45:47,781 - Epoch: [7][ 850/ 1236] Overall Loss 0.478555 Objective Loss 0.478555 LR 0.001000 Time 0.017258 +2023-10-05 20:45:47,946 - Epoch: [7][ 860/ 1236] Overall Loss 0.478616 Objective Loss 0.478616 LR 0.001000 Time 0.017250 +2023-10-05 20:45:48,111 - Epoch: [7][ 870/ 1236] Overall Loss 0.478593 Objective Loss 0.478593 LR 0.001000 Time 0.017240 +2023-10-05 20:45:48,276 - Epoch: [7][ 880/ 1236] Overall Loss 0.478979 Objective Loss 0.478979 LR 0.001000 Time 0.017232 +2023-10-05 20:45:48,441 - Epoch: [7][ 890/ 1236] Overall Loss 0.478914 Objective Loss 0.478914 LR 0.001000 Time 0.017223 +2023-10-05 20:45:48,606 - Epoch: [7][ 900/ 1236] Overall Loss 0.478968 Objective Loss 0.478968 LR 0.001000 Time 0.017215 +2023-10-05 20:45:48,770 - Epoch: [7][ 910/ 1236] Overall Loss 0.478862 Objective Loss 0.478862 LR 0.001000 Time 0.017206 +2023-10-05 20:45:48,936 - Epoch: [7][ 920/ 1236] Overall Loss 0.478920 Objective Loss 0.478920 LR 0.001000 Time 0.017199 +2023-10-05 20:45:49,101 - Epoch: [7][ 930/ 1236] Overall Loss 0.478950 Objective Loss 0.478950 LR 0.001000 Time 0.017190 +2023-10-05 20:45:49,266 - Epoch: [7][ 940/ 1236] Overall Loss 0.478923 Objective Loss 0.478923 LR 0.001000 Time 0.017183 +2023-10-05 20:45:49,431 - Epoch: [7][ 950/ 1236] Overall Loss 0.478798 Objective Loss 0.478798 LR 0.001000 Time 0.017175 +2023-10-05 20:45:49,596 - Epoch: [7][ 960/ 1236] Overall Loss 0.478890 Objective Loss 0.478890 LR 0.001000 Time 0.017168 +2023-10-05 20:45:49,760 - Epoch: [7][ 970/ 1236] Overall Loss 0.478635 Objective Loss 0.478635 LR 0.001000 Time 0.017160 +2023-10-05 20:45:49,926 - Epoch: [7][ 980/ 1236] Overall Loss 0.478809 Objective Loss 0.478809 LR 0.001000 Time 0.017154 +2023-10-05 20:45:50,091 - Epoch: [7][ 990/ 1236] Overall Loss 0.478448 Objective Loss 0.478448 LR 0.001000 Time 0.017146 +2023-10-05 20:45:50,256 - Epoch: [7][ 1000/ 1236] Overall Loss 0.478119 Objective Loss 0.478119 LR 0.001000 Time 0.017140 +2023-10-05 20:45:50,421 - Epoch: [7][ 1010/ 1236] Overall Loss 0.478132 Objective Loss 0.478132 LR 0.001000 Time 0.017133 +2023-10-05 20:45:50,586 - Epoch: [7][ 1020/ 1236] Overall Loss 0.478362 Objective Loss 0.478362 LR 0.001000 Time 0.017127 +2023-10-05 20:45:50,751 - Epoch: [7][ 1030/ 1236] Overall Loss 0.478669 Objective Loss 0.478669 LR 0.001000 Time 0.017120 +2023-10-05 20:45:50,916 - Epoch: [7][ 1040/ 1236] Overall Loss 0.478891 Objective Loss 0.478891 LR 0.001000 Time 0.017114 +2023-10-05 20:45:51,081 - Epoch: [7][ 1050/ 1236] Overall Loss 0.479292 Objective Loss 0.479292 LR 0.001000 Time 0.017108 +2023-10-05 20:45:51,246 - Epoch: [7][ 1060/ 1236] Overall Loss 0.479174 Objective Loss 0.479174 LR 0.001000 Time 0.017102 +2023-10-05 20:45:51,411 - Epoch: [7][ 1070/ 1236] Overall Loss 0.479321 Objective Loss 0.479321 LR 0.001000 Time 0.017096 +2023-10-05 20:45:51,576 - Epoch: [7][ 1080/ 1236] Overall Loss 0.479416 Objective Loss 0.479416 LR 0.001000 Time 0.017090 +2023-10-05 20:45:51,741 - Epoch: [7][ 1090/ 1236] Overall Loss 0.479146 Objective Loss 0.479146 LR 0.001000 Time 0.017084 +2023-10-05 20:45:51,906 - Epoch: [7][ 1100/ 1236] Overall Loss 0.478715 Objective Loss 0.478715 LR 0.001000 Time 0.017079 +2023-10-05 20:45:52,071 - Epoch: [7][ 1110/ 1236] Overall Loss 0.478428 Objective Loss 0.478428 LR 0.001000 Time 0.017074 +2023-10-05 20:45:52,236 - Epoch: [7][ 1120/ 1236] Overall Loss 0.478828 Objective Loss 0.478828 LR 0.001000 Time 0.017069 +2023-10-05 20:45:52,401 - Epoch: [7][ 1130/ 1236] Overall Loss 0.478764 Objective Loss 0.478764 LR 0.001000 Time 0.017063 +2023-10-05 20:45:52,568 - Epoch: [7][ 1140/ 1236] Overall Loss 0.478942 Objective Loss 0.478942 LR 0.001000 Time 0.017057 +2023-10-05 20:45:52,733 - Epoch: [7][ 1150/ 1236] Overall Loss 0.479111 Objective Loss 0.479111 LR 0.001000 Time 0.017051 +2023-10-05 20:45:52,898 - Epoch: [7][ 1160/ 1236] Overall Loss 0.479398 Objective Loss 0.479398 LR 0.001000 Time 0.017047 +2023-10-05 20:45:53,063 - Epoch: [7][ 1170/ 1236] Overall Loss 0.479370 Objective Loss 0.479370 LR 0.001000 Time 0.017042 +2023-10-05 20:45:53,229 - Epoch: [7][ 1180/ 1236] Overall Loss 0.479541 Objective Loss 0.479541 LR 0.001000 Time 0.017038 +2023-10-05 20:45:53,394 - Epoch: [7][ 1190/ 1236] Overall Loss 0.479332 Objective Loss 0.479332 LR 0.001000 Time 0.017032 +2023-10-05 20:45:53,559 - Epoch: [7][ 1200/ 1236] Overall Loss 0.479178 Objective Loss 0.479178 LR 0.001000 Time 0.017028 +2023-10-05 20:45:53,724 - Epoch: [7][ 1210/ 1236] Overall Loss 0.478834 Objective Loss 0.478834 LR 0.001000 Time 0.017023 +2023-10-05 20:45:53,889 - Epoch: [7][ 1220/ 1236] Overall Loss 0.478828 Objective Loss 0.478828 LR 0.001000 Time 0.017019 +2023-10-05 20:45:54,101 - Epoch: [7][ 1230/ 1236] Overall Loss 0.478946 Objective Loss 0.478946 LR 0.001000 Time 0.017053 +2023-10-05 20:45:54,192 - Epoch: [7][ 1236/ 1236] Overall Loss 0.478972 Objective Loss 0.478972 Top1 74.134420 Top5 95.723014 LR 0.001000 Time 0.017044 +2023-10-05 20:45:54,311 - --- validate (epoch=7)----------- +2023-10-05 20:45:54,311 - 29943 samples (256 per mini-batch) +2023-10-05 20:45:54,713 - Epoch: [7][ 10/ 117] Loss 0.448884 Top1 75.039062 Top5 96.914062 +2023-10-05 20:45:54,817 - Epoch: [7][ 20/ 117] Loss 0.438593 Top1 74.804688 Top5 96.503906 +2023-10-05 20:45:54,922 - Epoch: [7][ 30/ 117] Loss 0.444569 Top1 74.596354 Top5 96.263021 +2023-10-05 20:45:55,024 - Epoch: [7][ 40/ 117] Loss 0.445708 Top1 74.677734 Top5 96.152344 +2023-10-05 20:45:55,128 - Epoch: [7][ 50/ 117] Loss 0.450224 Top1 74.507812 Top5 96.265625 +2023-10-05 20:45:55,230 - Epoch: [7][ 60/ 117] Loss 0.446774 Top1 74.537760 Top5 96.263021 +2023-10-05 20:45:55,334 - Epoch: [7][ 70/ 117] Loss 0.447445 Top1 74.425223 Top5 96.205357 +2023-10-05 20:45:55,436 - Epoch: [7][ 80/ 117] Loss 0.441376 Top1 74.389648 Top5 96.235352 +2023-10-05 20:45:55,540 - Epoch: [7][ 90/ 117] Loss 0.444875 Top1 74.353299 Top5 96.236979 +2023-10-05 20:45:55,642 - Epoch: [7][ 100/ 117] Loss 0.449402 Top1 74.285156 Top5 96.250000 +2023-10-05 20:45:55,753 - Epoch: [7][ 110/ 117] Loss 0.448944 Top1 74.428267 Top5 96.214489 +2023-10-05 20:45:55,809 - Epoch: [7][ 117/ 117] Loss 0.448336 Top1 74.478175 Top5 96.236182 +2023-10-05 20:45:55,913 - ==> Top1: 74.478 Top5: 96.236 Loss: 0.448 + +2023-10-05 20:45:55,914 - ==> Confusion: +[[ 902 2 3 0 7 3 0 3 10 87 1 0 0 8 7 1 5 1 3 0 7] + [ 1 1020 2 1 5 21 1 43 4 1 4 0 0 0 0 4 5 1 13 4 1] + [ 9 3 914 13 5 1 35 26 0 6 7 3 2 1 3 0 4 3 8 4 9] + [ 1 4 32 917 1 4 2 1 7 1 20 0 9 3 27 2 2 13 37 0 6] + [ 37 24 0 0 908 10 0 1 1 16 1 2 0 4 11 5 24 0 1 0 5] + [ 10 82 5 0 1 872 1 58 3 8 3 15 2 29 5 1 2 3 1 7 8] + [ 0 10 44 1 0 2 1059 30 0 0 6 3 1 0 0 11 3 0 4 14 3] + [ 8 15 16 0 0 15 3 1080 5 4 6 9 0 0 0 6 0 1 41 6 3] + [ 24 4 0 1 0 0 0 0 954 69 7 2 0 9 6 3 1 2 5 1 1] + [ 141 1 0 0 3 0 0 1 43 893 0 1 0 21 2 3 0 0 1 2 7] + [ 5 6 5 12 2 0 1 6 17 4 954 0 0 7 12 4 1 0 9 1 7] + [ 4 1 2 0 3 11 0 5 0 1 0 918 22 9 0 2 3 20 0 32 2] + [ 4 1 3 6 0 2 1 4 1 1 2 78 913 1 0 7 5 26 3 4 6] + [ 6 0 0 0 2 4 1 3 19 25 14 8 1 1018 2 2 4 1 0 2 7] + [ 18 3 1 16 7 0 0 1 69 11 1 0 5 2 917 1 2 7 31 0 9] + [ 1 5 4 1 5 1 1 0 0 2 0 21 9 0 0 1016 37 14 1 6 10] + [ 4 27 2 0 5 9 1 0 5 0 3 8 0 1 2 8 1071 0 0 9 6] + [ 1 1 0 2 0 0 1 0 0 1 0 11 35 1 0 5 1 970 3 2 4] + [ 2 8 3 11 3 0 0 53 8 4 3 1 6 0 4 0 1 0 951 1 9] + [ 0 4 3 1 2 4 8 36 2 0 3 14 3 3 0 2 6 5 2 1047 7] + [ 302 526 169 113 150 221 42 229 225 158 241 187 514 382 214 77 248 108 337 455 3007]] + +2023-10-05 20:45:55,915 - ==> Best [Top1: 74.478 Top5: 96.236 Sparsity:0.00 Params: 148928 on epoch: 7] +2023-10-05 20:45:55,915 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:45:55,928 - + +2023-10-05 20:45:55,928 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:45:56,993 - Epoch: [8][ 10/ 1236] Overall Loss 0.465473 Objective Loss 0.465473 LR 0.001000 Time 0.106502 +2023-10-05 20:45:57,160 - Epoch: [8][ 20/ 1236] Overall Loss 0.473792 Objective Loss 0.473792 LR 0.001000 Time 0.061559 +2023-10-05 20:45:57,325 - Epoch: [8][ 30/ 1236] Overall Loss 0.475750 Objective Loss 0.475750 LR 0.001000 Time 0.046537 +2023-10-05 20:45:57,490 - Epoch: [8][ 40/ 1236] Overall Loss 0.478966 Objective Loss 0.478966 LR 0.001000 Time 0.039027 +2023-10-05 20:45:57,656 - Epoch: [8][ 50/ 1236] Overall Loss 0.471807 Objective Loss 0.471807 LR 0.001000 Time 0.034516 +2023-10-05 20:45:57,821 - Epoch: [8][ 60/ 1236] Overall Loss 0.471379 Objective Loss 0.471379 LR 0.001000 Time 0.031511 +2023-10-05 20:45:57,986 - Epoch: [8][ 70/ 1236] Overall Loss 0.466863 Objective Loss 0.466863 LR 0.001000 Time 0.029360 +2023-10-05 20:45:58,151 - Epoch: [8][ 80/ 1236] Overall Loss 0.465198 Objective Loss 0.465198 LR 0.001000 Time 0.027750 +2023-10-05 20:45:58,315 - Epoch: [8][ 90/ 1236] Overall Loss 0.458219 Objective Loss 0.458219 LR 0.001000 Time 0.026492 +2023-10-05 20:45:58,480 - Epoch: [8][ 100/ 1236] Overall Loss 0.455991 Objective Loss 0.455991 LR 0.001000 Time 0.025491 +2023-10-05 20:45:58,645 - Epoch: [8][ 110/ 1236] Overall Loss 0.454536 Objective Loss 0.454536 LR 0.001000 Time 0.024669 +2023-10-05 20:45:58,810 - Epoch: [8][ 120/ 1236] Overall Loss 0.453732 Objective Loss 0.453732 LR 0.001000 Time 0.023985 +2023-10-05 20:45:58,975 - Epoch: [8][ 130/ 1236] Overall Loss 0.453607 Objective Loss 0.453607 LR 0.001000 Time 0.023406 +2023-10-05 20:45:59,140 - Epoch: [8][ 140/ 1236] Overall Loss 0.452648 Objective Loss 0.452648 LR 0.001000 Time 0.022909 +2023-10-05 20:45:59,304 - Epoch: [8][ 150/ 1236] Overall Loss 0.453310 Objective Loss 0.453310 LR 0.001000 Time 0.022475 +2023-10-05 20:45:59,469 - Epoch: [8][ 160/ 1236] Overall Loss 0.450633 Objective Loss 0.450633 LR 0.001000 Time 0.022100 +2023-10-05 20:45:59,634 - Epoch: [8][ 170/ 1236] Overall Loss 0.452807 Objective Loss 0.452807 LR 0.001000 Time 0.021767 +2023-10-05 20:45:59,799 - Epoch: [8][ 180/ 1236] Overall Loss 0.450232 Objective Loss 0.450232 LR 0.001000 Time 0.021473 +2023-10-05 20:45:59,963 - Epoch: [8][ 190/ 1236] Overall Loss 0.449795 Objective Loss 0.449795 LR 0.001000 Time 0.021206 +2023-10-05 20:46:00,128 - Epoch: [8][ 200/ 1236] Overall Loss 0.450528 Objective Loss 0.450528 LR 0.001000 Time 0.020971 +2023-10-05 20:46:00,292 - Epoch: [8][ 210/ 1236] Overall Loss 0.448119 Objective Loss 0.448119 LR 0.001000 Time 0.020750 +2023-10-05 20:46:00,457 - Epoch: [8][ 220/ 1236] Overall Loss 0.448640 Objective Loss 0.448640 LR 0.001000 Time 0.020556 +2023-10-05 20:46:00,622 - Epoch: [8][ 230/ 1236] Overall Loss 0.448910 Objective Loss 0.448910 LR 0.001000 Time 0.020376 +2023-10-05 20:46:00,787 - Epoch: [8][ 240/ 1236] Overall Loss 0.449521 Objective Loss 0.449521 LR 0.001000 Time 0.020213 +2023-10-05 20:46:00,951 - Epoch: [8][ 250/ 1236] Overall Loss 0.449783 Objective Loss 0.449783 LR 0.001000 Time 0.020060 +2023-10-05 20:46:01,115 - Epoch: [8][ 260/ 1236] Overall Loss 0.449388 Objective Loss 0.449388 LR 0.001000 Time 0.019920 +2023-10-05 20:46:01,279 - Epoch: [8][ 270/ 1236] Overall Loss 0.450782 Objective Loss 0.450782 LR 0.001000 Time 0.019787 +2023-10-05 20:46:01,444 - Epoch: [8][ 280/ 1236] Overall Loss 0.450866 Objective Loss 0.450866 LR 0.001000 Time 0.019668 +2023-10-05 20:46:01,608 - Epoch: [8][ 290/ 1236] Overall Loss 0.451436 Objective Loss 0.451436 LR 0.001000 Time 0.019556 +2023-10-05 20:46:01,773 - Epoch: [8][ 300/ 1236] Overall Loss 0.451706 Objective Loss 0.451706 LR 0.001000 Time 0.019454 +2023-10-05 20:46:01,938 - Epoch: [8][ 310/ 1236] Overall Loss 0.452401 Objective Loss 0.452401 LR 0.001000 Time 0.019356 +2023-10-05 20:46:02,104 - Epoch: [8][ 320/ 1236] Overall Loss 0.453083 Objective Loss 0.453083 LR 0.001000 Time 0.019269 +2023-10-05 20:46:02,268 - Epoch: [8][ 330/ 1236] Overall Loss 0.453616 Objective Loss 0.453616 LR 0.001000 Time 0.019180 +2023-10-05 20:46:02,433 - Epoch: [8][ 340/ 1236] Overall Loss 0.453933 Objective Loss 0.453933 LR 0.001000 Time 0.019100 +2023-10-05 20:46:02,597 - Epoch: [8][ 350/ 1236] Overall Loss 0.454285 Objective Loss 0.454285 LR 0.001000 Time 0.019024 +2023-10-05 20:46:02,762 - Epoch: [8][ 360/ 1236] Overall Loss 0.455523 Objective Loss 0.455523 LR 0.001000 Time 0.018952 +2023-10-05 20:46:02,928 - Epoch: [8][ 370/ 1236] Overall Loss 0.455289 Objective Loss 0.455289 LR 0.001000 Time 0.018888 +2023-10-05 20:46:03,096 - Epoch: [8][ 380/ 1236] Overall Loss 0.455027 Objective Loss 0.455027 LR 0.001000 Time 0.018831 +2023-10-05 20:46:03,262 - Epoch: [8][ 390/ 1236] Overall Loss 0.454951 Objective Loss 0.454951 LR 0.001000 Time 0.018776 +2023-10-05 20:46:03,430 - Epoch: [8][ 400/ 1236] Overall Loss 0.455681 Objective Loss 0.455681 LR 0.001000 Time 0.018725 +2023-10-05 20:46:03,594 - Epoch: [8][ 410/ 1236] Overall Loss 0.455816 Objective Loss 0.455816 LR 0.001000 Time 0.018666 +2023-10-05 20:46:03,758 - Epoch: [8][ 420/ 1236] Overall Loss 0.455913 Objective Loss 0.455913 LR 0.001000 Time 0.018613 +2023-10-05 20:46:03,922 - Epoch: [8][ 430/ 1236] Overall Loss 0.456780 Objective Loss 0.456780 LR 0.001000 Time 0.018559 +2023-10-05 20:46:04,086 - Epoch: [8][ 440/ 1236] Overall Loss 0.456900 Objective Loss 0.456900 LR 0.001000 Time 0.018511 +2023-10-05 20:46:04,250 - Epoch: [8][ 450/ 1236] Overall Loss 0.458500 Objective Loss 0.458500 LR 0.001000 Time 0.018462 +2023-10-05 20:46:04,414 - Epoch: [8][ 460/ 1236] Overall Loss 0.458953 Objective Loss 0.458953 LR 0.001000 Time 0.018417 +2023-10-05 20:46:04,577 - Epoch: [8][ 470/ 1236] Overall Loss 0.458662 Objective Loss 0.458662 LR 0.001000 Time 0.018371 +2023-10-05 20:46:04,742 - Epoch: [8][ 480/ 1236] Overall Loss 0.458463 Objective Loss 0.458463 LR 0.001000 Time 0.018331 +2023-10-05 20:46:04,905 - Epoch: [8][ 490/ 1236] Overall Loss 0.458464 Objective Loss 0.458464 LR 0.001000 Time 0.018290 +2023-10-05 20:46:05,070 - Epoch: [8][ 500/ 1236] Overall Loss 0.457764 Objective Loss 0.457764 LR 0.001000 Time 0.018253 +2023-10-05 20:46:05,234 - Epoch: [8][ 510/ 1236] Overall Loss 0.458568 Objective Loss 0.458568 LR 0.001000 Time 0.018215 +2023-10-05 20:46:05,400 - Epoch: [8][ 520/ 1236] Overall Loss 0.458534 Objective Loss 0.458534 LR 0.001000 Time 0.018184 +2023-10-05 20:46:05,565 - Epoch: [8][ 530/ 1236] Overall Loss 0.458257 Objective Loss 0.458257 LR 0.001000 Time 0.018152 +2023-10-05 20:46:05,731 - Epoch: [8][ 540/ 1236] Overall Loss 0.458418 Objective Loss 0.458418 LR 0.001000 Time 0.018123 +2023-10-05 20:46:05,897 - Epoch: [8][ 550/ 1236] Overall Loss 0.458558 Objective Loss 0.458558 LR 0.001000 Time 0.018094 +2023-10-05 20:46:06,063 - Epoch: [8][ 560/ 1236] Overall Loss 0.458453 Objective Loss 0.458453 LR 0.001000 Time 0.018067 +2023-10-05 20:46:06,228 - Epoch: [8][ 570/ 1236] Overall Loss 0.457793 Objective Loss 0.457793 LR 0.001000 Time 0.018039 +2023-10-05 20:46:06,394 - Epoch: [8][ 580/ 1236] Overall Loss 0.457783 Objective Loss 0.457783 LR 0.001000 Time 0.018014 +2023-10-05 20:46:06,560 - Epoch: [8][ 590/ 1236] Overall Loss 0.458406 Objective Loss 0.458406 LR 0.001000 Time 0.017989 +2023-10-05 20:46:06,726 - Epoch: [8][ 600/ 1236] Overall Loss 0.458317 Objective Loss 0.458317 LR 0.001000 Time 0.017966 +2023-10-05 20:46:06,891 - Epoch: [8][ 610/ 1236] Overall Loss 0.458313 Objective Loss 0.458313 LR 0.001000 Time 0.017941 +2023-10-05 20:46:07,057 - Epoch: [8][ 620/ 1236] Overall Loss 0.458394 Objective Loss 0.458394 LR 0.001000 Time 0.017919 +2023-10-05 20:46:07,222 - Epoch: [8][ 630/ 1236] Overall Loss 0.458678 Objective Loss 0.458678 LR 0.001000 Time 0.017896 +2023-10-05 20:46:07,388 - Epoch: [8][ 640/ 1236] Overall Loss 0.458583 Objective Loss 0.458583 LR 0.001000 Time 0.017876 +2023-10-05 20:46:07,553 - Epoch: [8][ 650/ 1236] Overall Loss 0.458942 Objective Loss 0.458942 LR 0.001000 Time 0.017855 +2023-10-05 20:46:07,720 - Epoch: [8][ 660/ 1236] Overall Loss 0.459540 Objective Loss 0.459540 LR 0.001000 Time 0.017836 +2023-10-05 20:46:07,885 - Epoch: [8][ 670/ 1236] Overall Loss 0.459609 Objective Loss 0.459609 LR 0.001000 Time 0.017816 +2023-10-05 20:46:08,051 - Epoch: [8][ 680/ 1236] Overall Loss 0.459606 Objective Loss 0.459606 LR 0.001000 Time 0.017797 +2023-10-05 20:46:08,216 - Epoch: [8][ 690/ 1236] Overall Loss 0.459707 Objective Loss 0.459707 LR 0.001000 Time 0.017778 +2023-10-05 20:46:08,382 - Epoch: [8][ 700/ 1236] Overall Loss 0.459749 Objective Loss 0.459749 LR 0.001000 Time 0.017761 +2023-10-05 20:46:08,547 - Epoch: [8][ 710/ 1236] Overall Loss 0.459423 Objective Loss 0.459423 LR 0.001000 Time 0.017743 +2023-10-05 20:46:08,714 - Epoch: [8][ 720/ 1236] Overall Loss 0.458685 Objective Loss 0.458685 LR 0.001000 Time 0.017727 +2023-10-05 20:46:08,879 - Epoch: [8][ 730/ 1236] Overall Loss 0.458668 Objective Loss 0.458668 LR 0.001000 Time 0.017710 +2023-10-05 20:46:09,045 - Epoch: [8][ 740/ 1236] Overall Loss 0.458574 Objective Loss 0.458574 LR 0.001000 Time 0.017695 +2023-10-05 20:46:09,211 - Epoch: [8][ 750/ 1236] Overall Loss 0.458544 Objective Loss 0.458544 LR 0.001000 Time 0.017679 +2023-10-05 20:46:09,377 - Epoch: [8][ 760/ 1236] Overall Loss 0.458349 Objective Loss 0.458349 LR 0.001000 Time 0.017665 +2023-10-05 20:46:09,542 - Epoch: [8][ 770/ 1236] Overall Loss 0.458780 Objective Loss 0.458780 LR 0.001000 Time 0.017649 +2023-10-05 20:46:09,708 - Epoch: [8][ 780/ 1236] Overall Loss 0.458902 Objective Loss 0.458902 LR 0.001000 Time 0.017636 +2023-10-05 20:46:09,874 - Epoch: [8][ 790/ 1236] Overall Loss 0.458809 Objective Loss 0.458809 LR 0.001000 Time 0.017622 +2023-10-05 20:46:10,041 - Epoch: [8][ 800/ 1236] Overall Loss 0.458385 Objective Loss 0.458385 LR 0.001000 Time 0.017610 +2023-10-05 20:46:10,206 - Epoch: [8][ 810/ 1236] Overall Loss 0.458078 Objective Loss 0.458078 LR 0.001000 Time 0.017597 +2023-10-05 20:46:10,373 - Epoch: [8][ 820/ 1236] Overall Loss 0.457974 Objective Loss 0.457974 LR 0.001000 Time 0.017585 +2023-10-05 20:46:10,537 - Epoch: [8][ 830/ 1236] Overall Loss 0.457470 Objective Loss 0.457470 LR 0.001000 Time 0.017571 +2023-10-05 20:46:10,703 - Epoch: [8][ 840/ 1236] Overall Loss 0.456848 Objective Loss 0.456848 LR 0.001000 Time 0.017558 +2023-10-05 20:46:10,867 - Epoch: [8][ 850/ 1236] Overall Loss 0.456742 Objective Loss 0.456742 LR 0.001000 Time 0.017545 +2023-10-05 20:46:11,032 - Epoch: [8][ 860/ 1236] Overall Loss 0.456433 Objective Loss 0.456433 LR 0.001000 Time 0.017532 +2023-10-05 20:46:11,196 - Epoch: [8][ 870/ 1236] Overall Loss 0.456843 Objective Loss 0.456843 LR 0.001000 Time 0.017519 +2023-10-05 20:46:11,362 - Epoch: [8][ 880/ 1236] Overall Loss 0.457149 Objective Loss 0.457149 LR 0.001000 Time 0.017508 +2023-10-05 20:46:11,526 - Epoch: [8][ 890/ 1236] Overall Loss 0.457339 Objective Loss 0.457339 LR 0.001000 Time 0.017495 +2023-10-05 20:46:11,692 - Epoch: [8][ 900/ 1236] Overall Loss 0.457071 Objective Loss 0.457071 LR 0.001000 Time 0.017485 +2023-10-05 20:46:11,860 - Epoch: [8][ 910/ 1236] Overall Loss 0.456838 Objective Loss 0.456838 LR 0.001000 Time 0.017477 +2023-10-05 20:46:12,026 - Epoch: [8][ 920/ 1236] Overall Loss 0.457029 Objective Loss 0.457029 LR 0.001000 Time 0.017467 +2023-10-05 20:46:12,199 - Epoch: [8][ 930/ 1236] Overall Loss 0.456673 Objective Loss 0.456673 LR 0.001000 Time 0.017465 +2023-10-05 20:46:12,370 - Epoch: [8][ 940/ 1236] Overall Loss 0.456671 Objective Loss 0.456671 LR 0.001000 Time 0.017461 +2023-10-05 20:46:12,543 - Epoch: [8][ 950/ 1236] Overall Loss 0.456789 Objective Loss 0.456789 LR 0.001000 Time 0.017459 +2023-10-05 20:46:12,714 - Epoch: [8][ 960/ 1236] Overall Loss 0.456173 Objective Loss 0.456173 LR 0.001000 Time 0.017455 +2023-10-05 20:46:12,888 - Epoch: [8][ 970/ 1236] Overall Loss 0.455974 Objective Loss 0.455974 LR 0.001000 Time 0.017453 +2023-10-05 20:46:13,058 - Epoch: [8][ 980/ 1236] Overall Loss 0.455662 Objective Loss 0.455662 LR 0.001000 Time 0.017448 +2023-10-05 20:46:13,232 - Epoch: [8][ 990/ 1236] Overall Loss 0.455658 Objective Loss 0.455658 LR 0.001000 Time 0.017447 +2023-10-05 20:46:13,400 - Epoch: [8][ 1000/ 1236] Overall Loss 0.455382 Objective Loss 0.455382 LR 0.001000 Time 0.017441 +2023-10-05 20:46:13,567 - Epoch: [8][ 1010/ 1236] Overall Loss 0.455094 Objective Loss 0.455094 LR 0.001000 Time 0.017433 +2023-10-05 20:46:13,734 - Epoch: [8][ 1020/ 1236] Overall Loss 0.454969 Objective Loss 0.454969 LR 0.001000 Time 0.017426 +2023-10-05 20:46:13,901 - Epoch: [8][ 1030/ 1236] Overall Loss 0.455119 Objective Loss 0.455119 LR 0.001000 Time 0.017419 +2023-10-05 20:46:14,069 - Epoch: [8][ 1040/ 1236] Overall Loss 0.455002 Objective Loss 0.455002 LR 0.001000 Time 0.017412 +2023-10-05 20:46:14,236 - Epoch: [8][ 1050/ 1236] Overall Loss 0.454848 Objective Loss 0.454848 LR 0.001000 Time 0.017405 +2023-10-05 20:46:14,404 - Epoch: [8][ 1060/ 1236] Overall Loss 0.454863 Objective Loss 0.454863 LR 0.001000 Time 0.017399 +2023-10-05 20:46:14,571 - Epoch: [8][ 1070/ 1236] Overall Loss 0.455093 Objective Loss 0.455093 LR 0.001000 Time 0.017392 +2023-10-05 20:46:14,739 - Epoch: [8][ 1080/ 1236] Overall Loss 0.455193 Objective Loss 0.455193 LR 0.001000 Time 0.017386 +2023-10-05 20:46:14,906 - Epoch: [8][ 1090/ 1236] Overall Loss 0.455172 Objective Loss 0.455172 LR 0.001000 Time 0.017380 +2023-10-05 20:46:15,074 - Epoch: [8][ 1100/ 1236] Overall Loss 0.455126 Objective Loss 0.455126 LR 0.001000 Time 0.017374 +2023-10-05 20:46:15,240 - Epoch: [8][ 1110/ 1236] Overall Loss 0.455409 Objective Loss 0.455409 LR 0.001000 Time 0.017367 +2023-10-05 20:46:15,408 - Epoch: [8][ 1120/ 1236] Overall Loss 0.455251 Objective Loss 0.455251 LR 0.001000 Time 0.017361 +2023-10-05 20:46:15,575 - Epoch: [8][ 1130/ 1236] Overall Loss 0.455144 Objective Loss 0.455144 LR 0.001000 Time 0.017355 +2023-10-05 20:46:15,743 - Epoch: [8][ 1140/ 1236] Overall Loss 0.454851 Objective Loss 0.454851 LR 0.001000 Time 0.017350 +2023-10-05 20:46:15,911 - Epoch: [8][ 1150/ 1236] Overall Loss 0.454709 Objective Loss 0.454709 LR 0.001000 Time 0.017344 +2023-10-05 20:46:16,078 - Epoch: [8][ 1160/ 1236] Overall Loss 0.454766 Objective Loss 0.454766 LR 0.001000 Time 0.017339 +2023-10-05 20:46:16,245 - Epoch: [8][ 1170/ 1236] Overall Loss 0.454690 Objective Loss 0.454690 LR 0.001000 Time 0.017334 +2023-10-05 20:46:16,413 - Epoch: [8][ 1180/ 1236] Overall Loss 0.454899 Objective Loss 0.454899 LR 0.001000 Time 0.017328 +2023-10-05 20:46:16,580 - Epoch: [8][ 1190/ 1236] Overall Loss 0.454664 Objective Loss 0.454664 LR 0.001000 Time 0.017323 +2023-10-05 20:46:16,748 - Epoch: [8][ 1200/ 1236] Overall Loss 0.454420 Objective Loss 0.454420 LR 0.001000 Time 0.017318 +2023-10-05 20:46:16,916 - Epoch: [8][ 1210/ 1236] Overall Loss 0.454654 Objective Loss 0.454654 LR 0.001000 Time 0.017313 +2023-10-05 20:46:17,083 - Epoch: [8][ 1220/ 1236] Overall Loss 0.454978 Objective Loss 0.454978 LR 0.001000 Time 0.017308 +2023-10-05 20:46:17,296 - Epoch: [8][ 1230/ 1236] Overall Loss 0.455154 Objective Loss 0.455154 LR 0.001000 Time 0.017340 +2023-10-05 20:46:17,386 - Epoch: [8][ 1236/ 1236] Overall Loss 0.455095 Objective Loss 0.455095 Top1 75.967413 Top5 97.148676 LR 0.001000 Time 0.017329 +2023-10-05 20:46:17,510 - --- validate (epoch=8)----------- +2023-10-05 20:46:17,510 - 29943 samples (256 per mini-batch) +2023-10-05 20:46:17,925 - Epoch: [8][ 10/ 117] Loss 0.444104 Top1 75.976562 Top5 96.796875 +2023-10-05 20:46:18,028 - Epoch: [8][ 20/ 117] Loss 0.441446 Top1 75.996094 Top5 96.679688 +2023-10-05 20:46:18,128 - Epoch: [8][ 30/ 117] Loss 0.431993 Top1 75.950521 Top5 96.380208 +2023-10-05 20:46:18,230 - Epoch: [8][ 40/ 117] Loss 0.427945 Top1 75.839844 Top5 96.357422 +2023-10-05 20:46:18,329 - Epoch: [8][ 50/ 117] Loss 0.429365 Top1 75.687500 Top5 96.445312 +2023-10-05 20:46:18,430 - Epoch: [8][ 60/ 117] Loss 0.429572 Top1 75.651042 Top5 96.490885 +2023-10-05 20:46:18,529 - Epoch: [8][ 70/ 117] Loss 0.428009 Top1 75.580357 Top5 96.439732 +2023-10-05 20:46:18,632 - Epoch: [8][ 80/ 117] Loss 0.427365 Top1 75.727539 Top5 96.479492 +2023-10-05 20:46:18,734 - Epoch: [8][ 90/ 117] Loss 0.431497 Top1 75.559896 Top5 96.458333 +2023-10-05 20:46:18,837 - Epoch: [8][ 100/ 117] Loss 0.432843 Top1 75.507812 Top5 96.476562 +2023-10-05 20:46:18,950 - Epoch: [8][ 110/ 117] Loss 0.434922 Top1 75.465199 Top5 96.480824 +2023-10-05 20:46:19,005 - Epoch: [8][ 117/ 117] Loss 0.435377 Top1 75.416625 Top5 96.493337 +2023-10-05 20:46:19,142 - ==> Top1: 75.417 Top5: 96.493 Loss: 0.435 + +2023-10-05 20:46:19,143 - ==> Confusion: +[[ 881 5 10 1 14 6 0 0 3 77 1 3 0 4 8 10 7 1 1 1 17] + [ 0 986 0 4 8 65 3 23 2 0 4 1 0 0 2 5 5 1 16 2 4] + [ 5 1 921 27 7 2 24 12 0 0 7 4 7 4 0 7 0 4 7 1 16] + [ 2 2 34 955 0 8 0 1 3 0 10 0 8 5 26 3 0 13 3 2 14] + [ 20 16 3 0 947 14 2 1 0 6 1 3 2 4 11 6 6 0 3 2 3] + [ 5 47 4 3 2 977 1 14 1 2 5 9 0 15 5 4 2 0 3 8 9] + [ 0 13 83 3 0 7 1040 8 0 0 3 2 1 0 0 11 0 1 2 12 5] + [ 4 18 42 1 4 60 3 980 3 1 11 9 1 0 0 3 0 3 60 13 2] + [ 21 6 1 1 0 3 1 1 907 52 26 6 0 22 22 2 1 2 10 4 1] + [ 119 2 4 1 13 4 0 1 20 895 1 3 1 38 2 1 0 2 2 2 8] + [ 1 2 22 18 1 2 2 5 6 1 957 0 0 6 5 3 0 2 8 5 7] + [ 2 4 2 0 1 26 1 1 0 0 0 934 26 2 0 9 1 16 0 10 0] + [ 0 2 2 7 0 9 0 2 0 1 0 76 918 0 3 11 2 23 1 8 3] + [ 2 0 5 1 7 28 0 0 7 12 20 13 4 993 3 6 2 2 1 2 11] + [ 15 1 5 27 13 1 0 0 20 9 2 1 2 5 959 1 0 5 14 0 21] + [ 2 2 3 1 8 1 1 1 0 0 0 22 7 0 0 1052 6 10 0 8 10] + [ 1 18 2 0 13 14 0 0 1 0 0 10 0 2 2 19 1059 1 1 6 12] + [ 0 0 0 3 0 0 2 0 0 0 0 18 37 2 2 20 1 946 2 0 5] + [ 1 6 22 37 1 1 1 32 3 1 7 0 1 0 17 1 1 0 927 2 7] + [ 0 3 6 0 2 11 12 15 0 1 2 28 4 2 0 1 6 1 2 1046 10] + [ 170 344 334 195 112 429 39 104 90 120 294 230 435 394 217 199 190 94 230 383 3302]] + +2023-10-05 20:46:19,144 - ==> Best [Top1: 75.417 Top5: 96.493 Sparsity:0.00 Params: 148928 on epoch: 8] +2023-10-05 20:46:19,144 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:46:19,151 - + +2023-10-05 20:46:19,152 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:46:20,092 - Epoch: [9][ 10/ 1236] Overall Loss 0.464762 Objective Loss 0.464762 LR 0.001000 Time 0.093949 +2023-10-05 20:46:20,258 - Epoch: [9][ 20/ 1236] Overall Loss 0.455787 Objective Loss 0.455787 LR 0.001000 Time 0.055293 +2023-10-05 20:46:20,422 - Epoch: [9][ 30/ 1236] Overall Loss 0.438917 Objective Loss 0.438917 LR 0.001000 Time 0.042317 +2023-10-05 20:46:20,588 - Epoch: [9][ 40/ 1236] Overall Loss 0.430597 Objective Loss 0.430597 LR 0.001000 Time 0.035867 +2023-10-05 20:46:20,750 - Epoch: [9][ 50/ 1236] Overall Loss 0.433149 Objective Loss 0.433149 LR 0.001000 Time 0.031941 +2023-10-05 20:46:20,915 - Epoch: [9][ 60/ 1236] Overall Loss 0.427972 Objective Loss 0.427972 LR 0.001000 Time 0.029362 +2023-10-05 20:46:21,078 - Epoch: [9][ 70/ 1236] Overall Loss 0.426034 Objective Loss 0.426034 LR 0.001000 Time 0.027492 +2023-10-05 20:46:21,243 - Epoch: [9][ 80/ 1236] Overall Loss 0.426913 Objective Loss 0.426913 LR 0.001000 Time 0.026114 +2023-10-05 20:46:21,406 - Epoch: [9][ 90/ 1236] Overall Loss 0.426798 Objective Loss 0.426798 LR 0.001000 Time 0.025020 +2023-10-05 20:46:21,571 - Epoch: [9][ 100/ 1236] Overall Loss 0.428688 Objective Loss 0.428688 LR 0.001000 Time 0.024164 +2023-10-05 20:46:21,734 - Epoch: [9][ 110/ 1236] Overall Loss 0.426813 Objective Loss 0.426813 LR 0.001000 Time 0.023448 +2023-10-05 20:46:21,899 - Epoch: [9][ 120/ 1236] Overall Loss 0.425518 Objective Loss 0.425518 LR 0.001000 Time 0.022863 +2023-10-05 20:46:22,062 - Epoch: [9][ 130/ 1236] Overall Loss 0.428421 Objective Loss 0.428421 LR 0.001000 Time 0.022355 +2023-10-05 20:46:22,227 - Epoch: [9][ 140/ 1236] Overall Loss 0.427621 Objective Loss 0.427621 LR 0.001000 Time 0.021933 +2023-10-05 20:46:22,390 - Epoch: [9][ 150/ 1236] Overall Loss 0.429624 Objective Loss 0.429624 LR 0.001000 Time 0.021555 +2023-10-05 20:46:22,555 - Epoch: [9][ 160/ 1236] Overall Loss 0.432434 Objective Loss 0.432434 LR 0.001000 Time 0.021239 +2023-10-05 20:46:22,718 - Epoch: [9][ 170/ 1236] Overall Loss 0.432051 Objective Loss 0.432051 LR 0.001000 Time 0.020947 +2023-10-05 20:46:22,883 - Epoch: [9][ 180/ 1236] Overall Loss 0.430697 Objective Loss 0.430697 LR 0.001000 Time 0.020699 +2023-10-05 20:46:23,046 - Epoch: [9][ 190/ 1236] Overall Loss 0.430955 Objective Loss 0.430955 LR 0.001000 Time 0.020467 +2023-10-05 20:46:23,212 - Epoch: [9][ 200/ 1236] Overall Loss 0.430695 Objective Loss 0.430695 LR 0.001000 Time 0.020270 +2023-10-05 20:46:23,375 - Epoch: [9][ 210/ 1236] Overall Loss 0.432052 Objective Loss 0.432052 LR 0.001000 Time 0.020081 +2023-10-05 20:46:23,540 - Epoch: [9][ 220/ 1236] Overall Loss 0.432471 Objective Loss 0.432471 LR 0.001000 Time 0.019918 +2023-10-05 20:46:23,703 - Epoch: [9][ 230/ 1236] Overall Loss 0.431271 Objective Loss 0.431271 LR 0.001000 Time 0.019759 +2023-10-05 20:46:23,868 - Epoch: [9][ 240/ 1236] Overall Loss 0.430798 Objective Loss 0.430798 LR 0.001000 Time 0.019621 +2023-10-05 20:46:24,031 - Epoch: [9][ 250/ 1236] Overall Loss 0.431404 Objective Loss 0.431404 LR 0.001000 Time 0.019488 +2023-10-05 20:46:24,196 - Epoch: [9][ 260/ 1236] Overall Loss 0.430723 Objective Loss 0.430723 LR 0.001000 Time 0.019371 +2023-10-05 20:46:24,360 - Epoch: [9][ 270/ 1236] Overall Loss 0.430810 Objective Loss 0.430810 LR 0.001000 Time 0.019257 +2023-10-05 20:46:24,524 - Epoch: [9][ 280/ 1236] Overall Loss 0.430865 Objective Loss 0.430865 LR 0.001000 Time 0.019156 +2023-10-05 20:46:24,687 - Epoch: [9][ 290/ 1236] Overall Loss 0.430259 Objective Loss 0.430259 LR 0.001000 Time 0.019055 +2023-10-05 20:46:24,851 - Epoch: [9][ 300/ 1236] Overall Loss 0.430662 Objective Loss 0.430662 LR 0.001000 Time 0.018967 +2023-10-05 20:46:25,014 - Epoch: [9][ 310/ 1236] Overall Loss 0.430344 Objective Loss 0.430344 LR 0.001000 Time 0.018879 +2023-10-05 20:46:25,179 - Epoch: [9][ 320/ 1236] Overall Loss 0.430590 Objective Loss 0.430590 LR 0.001000 Time 0.018804 +2023-10-05 20:46:25,342 - Epoch: [9][ 330/ 1236] Overall Loss 0.430539 Objective Loss 0.430539 LR 0.001000 Time 0.018726 +2023-10-05 20:46:25,508 - Epoch: [9][ 340/ 1236] Overall Loss 0.430775 Objective Loss 0.430775 LR 0.001000 Time 0.018663 +2023-10-05 20:46:25,670 - Epoch: [9][ 350/ 1236] Overall Loss 0.430957 Objective Loss 0.430957 LR 0.001000 Time 0.018594 +2023-10-05 20:46:25,836 - Epoch: [9][ 360/ 1236] Overall Loss 0.431810 Objective Loss 0.431810 LR 0.001000 Time 0.018535 +2023-10-05 20:46:25,999 - Epoch: [9][ 370/ 1236] Overall Loss 0.431851 Objective Loss 0.431851 LR 0.001000 Time 0.018474 +2023-10-05 20:46:26,164 - Epoch: [9][ 380/ 1236] Overall Loss 0.431434 Objective Loss 0.431434 LR 0.001000 Time 0.018422 +2023-10-05 20:46:26,327 - Epoch: [9][ 390/ 1236] Overall Loss 0.431602 Objective Loss 0.431602 LR 0.001000 Time 0.018368 +2023-10-05 20:46:26,492 - Epoch: [9][ 400/ 1236] Overall Loss 0.431851 Objective Loss 0.431851 LR 0.001000 Time 0.018320 +2023-10-05 20:46:26,655 - Epoch: [9][ 410/ 1236] Overall Loss 0.432715 Objective Loss 0.432715 LR 0.001000 Time 0.018270 +2023-10-05 20:46:26,820 - Epoch: [9][ 420/ 1236] Overall Loss 0.433033 Objective Loss 0.433033 LR 0.001000 Time 0.018227 +2023-10-05 20:46:26,983 - Epoch: [9][ 430/ 1236] Overall Loss 0.432577 Objective Loss 0.432577 LR 0.001000 Time 0.018183 +2023-10-05 20:46:27,149 - Epoch: [9][ 440/ 1236] Overall Loss 0.431844 Objective Loss 0.431844 LR 0.001000 Time 0.018144 +2023-10-05 20:46:27,313 - Epoch: [9][ 450/ 1236] Overall Loss 0.431934 Objective Loss 0.431934 LR 0.001000 Time 0.018106 +2023-10-05 20:46:27,479 - Epoch: [9][ 460/ 1236] Overall Loss 0.431124 Objective Loss 0.431124 LR 0.001000 Time 0.018071 +2023-10-05 20:46:27,642 - Epoch: [9][ 470/ 1236] Overall Loss 0.431455 Objective Loss 0.431455 LR 0.001000 Time 0.018034 +2023-10-05 20:46:27,808 - Epoch: [9][ 480/ 1236] Overall Loss 0.432532 Objective Loss 0.432532 LR 0.001000 Time 0.018004 +2023-10-05 20:46:27,972 - Epoch: [9][ 490/ 1236] Overall Loss 0.432164 Objective Loss 0.432164 LR 0.001000 Time 0.017971 +2023-10-05 20:46:28,139 - Epoch: [9][ 500/ 1236] Overall Loss 0.432383 Objective Loss 0.432383 LR 0.001000 Time 0.017943 +2023-10-05 20:46:28,302 - Epoch: [9][ 510/ 1236] Overall Loss 0.431882 Objective Loss 0.431882 LR 0.001000 Time 0.017912 +2023-10-05 20:46:28,469 - Epoch: [9][ 520/ 1236] Overall Loss 0.431174 Objective Loss 0.431174 LR 0.001000 Time 0.017886 +2023-10-05 20:46:28,632 - Epoch: [9][ 530/ 1236] Overall Loss 0.431127 Objective Loss 0.431127 LR 0.001000 Time 0.017858 +2023-10-05 20:46:28,798 - Epoch: [9][ 540/ 1236] Overall Loss 0.431503 Objective Loss 0.431503 LR 0.001000 Time 0.017833 +2023-10-05 20:46:28,962 - Epoch: [9][ 550/ 1236] Overall Loss 0.431370 Objective Loss 0.431370 LR 0.001000 Time 0.017806 +2023-10-05 20:46:29,128 - Epoch: [9][ 560/ 1236] Overall Loss 0.431735 Objective Loss 0.431735 LR 0.001000 Time 0.017784 +2023-10-05 20:46:29,291 - Epoch: [9][ 570/ 1236] Overall Loss 0.432094 Objective Loss 0.432094 LR 0.001000 Time 0.017758 +2023-10-05 20:46:29,457 - Epoch: [9][ 580/ 1236] Overall Loss 0.432336 Objective Loss 0.432336 LR 0.001000 Time 0.017738 +2023-10-05 20:46:29,622 - Epoch: [9][ 590/ 1236] Overall Loss 0.432601 Objective Loss 0.432601 LR 0.001000 Time 0.017715 +2023-10-05 20:46:29,788 - Epoch: [9][ 600/ 1236] Overall Loss 0.432446 Objective Loss 0.432446 LR 0.001000 Time 0.017696 +2023-10-05 20:46:29,951 - Epoch: [9][ 610/ 1236] Overall Loss 0.432309 Objective Loss 0.432309 LR 0.001000 Time 0.017674 +2023-10-05 20:46:30,118 - Epoch: [9][ 620/ 1236] Overall Loss 0.431990 Objective Loss 0.431990 LR 0.001000 Time 0.017656 +2023-10-05 20:46:30,281 - Epoch: [9][ 630/ 1236] Overall Loss 0.432416 Objective Loss 0.432416 LR 0.001000 Time 0.017636 +2023-10-05 20:46:30,447 - Epoch: [9][ 640/ 1236] Overall Loss 0.432942 Objective Loss 0.432942 LR 0.001000 Time 0.017619 +2023-10-05 20:46:30,611 - Epoch: [9][ 650/ 1236] Overall Loss 0.433188 Objective Loss 0.433188 LR 0.001000 Time 0.017599 +2023-10-05 20:46:30,778 - Epoch: [9][ 660/ 1236] Overall Loss 0.432805 Objective Loss 0.432805 LR 0.001000 Time 0.017585 +2023-10-05 20:46:30,942 - Epoch: [9][ 670/ 1236] Overall Loss 0.432770 Objective Loss 0.432770 LR 0.001000 Time 0.017566 +2023-10-05 20:46:31,108 - Epoch: [9][ 680/ 1236] Overall Loss 0.432598 Objective Loss 0.432598 LR 0.001000 Time 0.017552 +2023-10-05 20:46:31,271 - Epoch: [9][ 690/ 1236] Overall Loss 0.432854 Objective Loss 0.432854 LR 0.001000 Time 0.017534 +2023-10-05 20:46:31,438 - Epoch: [9][ 700/ 1236] Overall Loss 0.433199 Objective Loss 0.433199 LR 0.001000 Time 0.017521 +2023-10-05 20:46:31,601 - Epoch: [9][ 710/ 1236] Overall Loss 0.433610 Objective Loss 0.433610 LR 0.001000 Time 0.017504 +2023-10-05 20:46:31,768 - Epoch: [9][ 720/ 1236] Overall Loss 0.433221 Objective Loss 0.433221 LR 0.001000 Time 0.017491 +2023-10-05 20:46:31,931 - Epoch: [9][ 730/ 1236] Overall Loss 0.433212 Objective Loss 0.433212 LR 0.001000 Time 0.017476 +2023-10-05 20:46:32,098 - Epoch: [9][ 740/ 1236] Overall Loss 0.433005 Objective Loss 0.433005 LR 0.001000 Time 0.017464 +2023-10-05 20:46:32,262 - Epoch: [9][ 750/ 1236] Overall Loss 0.432922 Objective Loss 0.432922 LR 0.001000 Time 0.017449 +2023-10-05 20:46:32,428 - Epoch: [9][ 760/ 1236] Overall Loss 0.432262 Objective Loss 0.432262 LR 0.001000 Time 0.017438 +2023-10-05 20:46:32,592 - Epoch: [9][ 770/ 1236] Overall Loss 0.432538 Objective Loss 0.432538 LR 0.001000 Time 0.017424 +2023-10-05 20:46:32,758 - Epoch: [9][ 780/ 1236] Overall Loss 0.431840 Objective Loss 0.431840 LR 0.001000 Time 0.017413 +2023-10-05 20:46:32,922 - Epoch: [9][ 790/ 1236] Overall Loss 0.431157 Objective Loss 0.431157 LR 0.001000 Time 0.017400 +2023-10-05 20:46:33,088 - Epoch: [9][ 800/ 1236] Overall Loss 0.431180 Objective Loss 0.431180 LR 0.001000 Time 0.017389 +2023-10-05 20:46:33,251 - Epoch: [9][ 810/ 1236] Overall Loss 0.431240 Objective Loss 0.431240 LR 0.001000 Time 0.017376 +2023-10-05 20:46:33,418 - Epoch: [9][ 820/ 1236] Overall Loss 0.431153 Objective Loss 0.431153 LR 0.001000 Time 0.017367 +2023-10-05 20:46:33,582 - Epoch: [9][ 830/ 1236] Overall Loss 0.431228 Objective Loss 0.431228 LR 0.001000 Time 0.017355 +2023-10-05 20:46:33,748 - Epoch: [9][ 840/ 1236] Overall Loss 0.431283 Objective Loss 0.431283 LR 0.001000 Time 0.017346 +2023-10-05 20:46:33,912 - Epoch: [9][ 850/ 1236] Overall Loss 0.431305 Objective Loss 0.431305 LR 0.001000 Time 0.017334 +2023-10-05 20:46:34,077 - Epoch: [9][ 860/ 1236] Overall Loss 0.431596 Objective Loss 0.431596 LR 0.001000 Time 0.017325 +2023-10-05 20:46:34,241 - Epoch: [9][ 870/ 1236] Overall Loss 0.431965 Objective Loss 0.431965 LR 0.001000 Time 0.017314 +2023-10-05 20:46:34,408 - Epoch: [9][ 880/ 1236] Overall Loss 0.431519 Objective Loss 0.431519 LR 0.001000 Time 0.017306 +2023-10-05 20:46:34,572 - Epoch: [9][ 890/ 1236] Overall Loss 0.431894 Objective Loss 0.431894 LR 0.001000 Time 0.017295 +2023-10-05 20:46:34,738 - Epoch: [9][ 900/ 1236] Overall Loss 0.432101 Objective Loss 0.432101 LR 0.001000 Time 0.017287 +2023-10-05 20:46:34,902 - Epoch: [9][ 910/ 1236] Overall Loss 0.432236 Objective Loss 0.432236 LR 0.001000 Time 0.017277 +2023-10-05 20:46:35,068 - Epoch: [9][ 920/ 1236] Overall Loss 0.432342 Objective Loss 0.432342 LR 0.001000 Time 0.017270 +2023-10-05 20:46:35,231 - Epoch: [9][ 930/ 1236] Overall Loss 0.432307 Objective Loss 0.432307 LR 0.001000 Time 0.017259 +2023-10-05 20:46:35,398 - Epoch: [9][ 940/ 1236] Overall Loss 0.432331 Objective Loss 0.432331 LR 0.001000 Time 0.017253 +2023-10-05 20:46:35,565 - Epoch: [9][ 950/ 1236] Overall Loss 0.431826 Objective Loss 0.431826 LR 0.001000 Time 0.017247 +2023-10-05 20:46:35,731 - Epoch: [9][ 960/ 1236] Overall Loss 0.431907 Objective Loss 0.431907 LR 0.001000 Time 0.017239 +2023-10-05 20:46:35,894 - Epoch: [9][ 970/ 1236] Overall Loss 0.431808 Objective Loss 0.431808 LR 0.001000 Time 0.017230 +2023-10-05 20:46:36,061 - Epoch: [9][ 980/ 1236] Overall Loss 0.431734 Objective Loss 0.431734 LR 0.001000 Time 0.017224 +2023-10-05 20:46:36,225 - Epoch: [9][ 990/ 1236] Overall Loss 0.431724 Objective Loss 0.431724 LR 0.001000 Time 0.017215 +2023-10-05 20:46:36,391 - Epoch: [9][ 1000/ 1236] Overall Loss 0.431324 Objective Loss 0.431324 LR 0.001000 Time 0.017209 +2023-10-05 20:46:36,555 - Epoch: [9][ 1010/ 1236] Overall Loss 0.431335 Objective Loss 0.431335 LR 0.001000 Time 0.017200 +2023-10-05 20:46:36,721 - Epoch: [9][ 1020/ 1236] Overall Loss 0.431289 Objective Loss 0.431289 LR 0.001000 Time 0.017194 +2023-10-05 20:46:36,884 - Epoch: [9][ 1030/ 1236] Overall Loss 0.431798 Objective Loss 0.431798 LR 0.001000 Time 0.017186 +2023-10-05 20:46:37,051 - Epoch: [9][ 1040/ 1236] Overall Loss 0.431893 Objective Loss 0.431893 LR 0.001000 Time 0.017180 +2023-10-05 20:46:37,215 - Epoch: [9][ 1050/ 1236] Overall Loss 0.431669 Objective Loss 0.431669 LR 0.001000 Time 0.017172 +2023-10-05 20:46:37,381 - Epoch: [9][ 1060/ 1236] Overall Loss 0.431923 Objective Loss 0.431923 LR 0.001000 Time 0.017167 +2023-10-05 20:46:37,545 - Epoch: [9][ 1070/ 1236] Overall Loss 0.432075 Objective Loss 0.432075 LR 0.001000 Time 0.017159 +2023-10-05 20:46:37,711 - Epoch: [9][ 1080/ 1236] Overall Loss 0.431944 Objective Loss 0.431944 LR 0.001000 Time 0.017154 +2023-10-05 20:46:37,875 - Epoch: [9][ 1090/ 1236] Overall Loss 0.432098 Objective Loss 0.432098 LR 0.001000 Time 0.017147 +2023-10-05 20:46:38,041 - Epoch: [9][ 1100/ 1236] Overall Loss 0.432063 Objective Loss 0.432063 LR 0.001000 Time 0.017142 +2023-10-05 20:46:38,205 - Epoch: [9][ 1110/ 1236] Overall Loss 0.432072 Objective Loss 0.432072 LR 0.001000 Time 0.017134 +2023-10-05 20:46:38,371 - Epoch: [9][ 1120/ 1236] Overall Loss 0.432001 Objective Loss 0.432001 LR 0.001000 Time 0.017130 +2023-10-05 20:46:38,535 - Epoch: [9][ 1130/ 1236] Overall Loss 0.432095 Objective Loss 0.432095 LR 0.001000 Time 0.017123 +2023-10-05 20:46:38,701 - Epoch: [9][ 1140/ 1236] Overall Loss 0.432218 Objective Loss 0.432218 LR 0.001000 Time 0.017118 +2023-10-05 20:46:38,865 - Epoch: [9][ 1150/ 1236] Overall Loss 0.432224 Objective Loss 0.432224 LR 0.001000 Time 0.017111 +2023-10-05 20:46:39,031 - Epoch: [9][ 1160/ 1236] Overall Loss 0.432118 Objective Loss 0.432118 LR 0.001000 Time 0.017107 +2023-10-05 20:46:39,195 - Epoch: [9][ 1170/ 1236] Overall Loss 0.431972 Objective Loss 0.431972 LR 0.001000 Time 0.017100 +2023-10-05 20:46:39,361 - Epoch: [9][ 1180/ 1236] Overall Loss 0.432089 Objective Loss 0.432089 LR 0.001000 Time 0.017096 +2023-10-05 20:46:39,525 - Epoch: [9][ 1190/ 1236] Overall Loss 0.432125 Objective Loss 0.432125 LR 0.001000 Time 0.017090 +2023-10-05 20:46:39,691 - Epoch: [9][ 1200/ 1236] Overall Loss 0.432295 Objective Loss 0.432295 LR 0.001000 Time 0.017086 +2023-10-05 20:46:39,855 - Epoch: [9][ 1210/ 1236] Overall Loss 0.432266 Objective Loss 0.432266 LR 0.001000 Time 0.017080 +2023-10-05 20:46:40,021 - Epoch: [9][ 1220/ 1236] Overall Loss 0.432154 Objective Loss 0.432154 LR 0.001000 Time 0.017076 +2023-10-05 20:46:40,233 - Epoch: [9][ 1230/ 1236] Overall Loss 0.431771 Objective Loss 0.431771 LR 0.001000 Time 0.017109 +2023-10-05 20:46:40,324 - Epoch: [9][ 1236/ 1236] Overall Loss 0.431903 Objective Loss 0.431903 Top1 77.189409 Top5 96.537678 LR 0.001000 Time 0.017100 +2023-10-05 20:46:40,464 - --- validate (epoch=9)----------- +2023-10-05 20:46:40,465 - 29943 samples (256 per mini-batch) +2023-10-05 20:46:40,870 - Epoch: [9][ 10/ 117] Loss 0.417960 Top1 77.539062 Top5 97.187500 +2023-10-05 20:46:40,978 - Epoch: [9][ 20/ 117] Loss 0.434650 Top1 76.757812 Top5 97.187500 +2023-10-05 20:46:41,083 - Epoch: [9][ 30/ 117] Loss 0.429899 Top1 76.888021 Top5 97.304688 +2023-10-05 20:46:41,190 - Epoch: [9][ 40/ 117] Loss 0.427970 Top1 76.943359 Top5 97.285156 +2023-10-05 20:46:41,294 - Epoch: [9][ 50/ 117] Loss 0.422940 Top1 76.992188 Top5 97.328125 +2023-10-05 20:46:41,398 - Epoch: [9][ 60/ 117] Loss 0.425695 Top1 77.044271 Top5 97.252604 +2023-10-05 20:46:41,502 - Epoch: [9][ 70/ 117] Loss 0.423980 Top1 77.287946 Top5 97.271205 +2023-10-05 20:46:41,607 - Epoch: [9][ 80/ 117] Loss 0.422394 Top1 77.089844 Top5 97.265625 +2023-10-05 20:46:41,711 - Epoch: [9][ 90/ 117] Loss 0.421956 Top1 77.061632 Top5 97.204861 +2023-10-05 20:46:41,815 - Epoch: [9][ 100/ 117] Loss 0.425954 Top1 76.984375 Top5 97.199219 +2023-10-05 20:46:41,928 - Epoch: [9][ 110/ 117] Loss 0.426486 Top1 77.038352 Top5 97.251420 +2023-10-05 20:46:41,984 - Epoch: [9][ 117/ 117] Loss 0.425574 Top1 77.043048 Top5 97.264803 +2023-10-05 20:46:42,113 - ==> Top1: 77.043 Top5: 97.265 Loss: 0.426 + +2023-10-05 20:46:42,114 - ==> Confusion: +[[ 884 5 0 1 21 5 0 1 3 96 0 0 0 4 6 2 6 1 1 0 14] + [ 2 1033 0 0 6 35 2 31 1 3 1 1 0 0 0 3 4 0 4 2 3] + [ 8 2 830 11 8 3 90 31 0 5 10 6 5 2 7 2 3 2 10 6 15] + [ 3 4 18 863 2 7 4 3 7 1 28 0 6 7 61 2 4 17 38 1 13] + [ 19 23 0 0 967 8 1 0 0 13 1 2 0 1 5 2 3 0 1 1 3] + [ 5 55 1 0 6 953 4 29 2 6 5 8 1 15 4 0 3 0 3 7 9] + [ 1 9 14 1 0 4 1111 10 0 1 1 6 0 0 0 12 0 1 2 14 4] + [ 7 24 11 0 0 38 7 1064 0 5 4 9 0 0 0 1 2 0 27 12 7] + [ 18 9 0 0 0 7 0 0 941 60 11 4 0 19 11 0 1 0 4 4 0] + [ 105 0 0 0 11 5 0 0 36 918 0 0 0 25 2 1 0 1 1 4 10] + [ 4 7 7 4 0 2 4 6 17 5 954 0 0 18 4 1 0 1 7 2 10] + [ 3 1 0 0 2 22 1 0 0 0 0 935 22 4 0 4 1 19 0 14 7] + [ 2 2 2 1 0 3 1 2 1 4 0 84 897 0 3 10 3 35 1 7 10] + [ 2 1 1 1 12 33 0 0 12 17 7 11 2 990 1 3 1 1 0 5 19] + [ 13 2 0 6 13 0 0 0 43 17 3 0 1 3 962 0 4 5 15 0 14] + [ 1 3 0 1 6 3 0 0 0 2 0 20 3 0 0 1041 21 13 0 9 11] + [ 0 31 1 0 17 5 0 0 0 0 0 7 0 1 1 10 1060 0 1 12 15] + [ 0 1 0 2 0 0 1 0 1 0 0 11 19 0 2 14 1 975 1 3 7] + [ 1 19 4 7 1 4 1 53 4 2 6 0 1 0 15 1 3 1 936 1 8] + [ 0 6 0 0 2 6 8 9 0 0 0 23 2 0 0 2 3 3 3 1078 7] + [ 219 442 98 44 183 292 84 186 128 147 178 228 366 362 189 124 300 70 207 381 3677]] + +2023-10-05 20:46:42,115 - ==> Best [Top1: 77.043 Top5: 97.265 Sparsity:0.00 Params: 148928 on epoch: 9] +2023-10-05 20:46:42,115 - Saving checkpoint to: logs/2023.10.05-204241/checkpoint.pth.tar +2023-10-05 20:46:42,128 - + +2023-10-05 20:46:42,128 - Initiating quantization aware training (QAT)... +2023-10-05 20:46:42,139 - + +2023-10-05 20:46:42,140 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:46:43,133 - Epoch: [10][ 10/ 1236] Overall Loss 0.614252 Objective Loss 0.614252 LR 0.001000 Time 0.099305 +2023-10-05 20:46:43,338 - Epoch: [10][ 20/ 1236] Overall Loss 0.580908 Objective Loss 0.580908 LR 0.001000 Time 0.059872 +2023-10-05 20:46:43,540 - Epoch: [10][ 30/ 1236] Overall Loss 0.560634 Objective Loss 0.560634 LR 0.001000 Time 0.046652 +2023-10-05 20:46:43,745 - Epoch: [10][ 40/ 1236] Overall Loss 0.551464 Objective Loss 0.551464 LR 0.001000 Time 0.040112 +2023-10-05 20:46:43,948 - Epoch: [10][ 50/ 1236] Overall Loss 0.546530 Objective Loss 0.546530 LR 0.001000 Time 0.036132 +2023-10-05 20:46:44,153 - Epoch: [10][ 60/ 1236] Overall Loss 0.537666 Objective Loss 0.537666 LR 0.001000 Time 0.033523 +2023-10-05 20:46:44,355 - Epoch: [10][ 70/ 1236] Overall Loss 0.524834 Objective Loss 0.524834 LR 0.001000 Time 0.031612 +2023-10-05 20:46:44,560 - Epoch: [10][ 80/ 1236] Overall Loss 0.517410 Objective Loss 0.517410 LR 0.001000 Time 0.030218 +2023-10-05 20:46:44,762 - Epoch: [10][ 90/ 1236] Overall Loss 0.513204 Objective Loss 0.513204 LR 0.001000 Time 0.029101 +2023-10-05 20:46:44,967 - Epoch: [10][ 100/ 1236] Overall Loss 0.508549 Objective Loss 0.508549 LR 0.001000 Time 0.028238 +2023-10-05 20:46:45,168 - Epoch: [10][ 110/ 1236] Overall Loss 0.507345 Objective Loss 0.507345 LR 0.001000 Time 0.027497 +2023-10-05 20:46:45,372 - Epoch: [10][ 120/ 1236] Overall Loss 0.504287 Objective Loss 0.504287 LR 0.001000 Time 0.026901 +2023-10-05 20:46:45,573 - Epoch: [10][ 130/ 1236] Overall Loss 0.502218 Objective Loss 0.502218 LR 0.001000 Time 0.026376 +2023-10-05 20:46:45,777 - Epoch: [10][ 140/ 1236] Overall Loss 0.498513 Objective Loss 0.498513 LR 0.001000 Time 0.025945 +2023-10-05 20:46:45,978 - Epoch: [10][ 150/ 1236] Overall Loss 0.496729 Objective Loss 0.496729 LR 0.001000 Time 0.025554 +2023-10-05 20:46:46,182 - Epoch: [10][ 160/ 1236] Overall Loss 0.496494 Objective Loss 0.496494 LR 0.001000 Time 0.025231 +2023-10-05 20:46:46,383 - Epoch: [10][ 170/ 1236] Overall Loss 0.495457 Objective Loss 0.495457 LR 0.001000 Time 0.024930 +2023-10-05 20:46:46,588 - Epoch: [10][ 180/ 1236] Overall Loss 0.494276 Objective Loss 0.494276 LR 0.001000 Time 0.024680 +2023-10-05 20:46:46,789 - Epoch: [10][ 190/ 1236] Overall Loss 0.492471 Objective Loss 0.492471 LR 0.001000 Time 0.024437 +2023-10-05 20:46:46,993 - Epoch: [10][ 200/ 1236] Overall Loss 0.492338 Objective Loss 0.492338 LR 0.001000 Time 0.024234 +2023-10-05 20:46:47,194 - Epoch: [10][ 210/ 1236] Overall Loss 0.490472 Objective Loss 0.490472 LR 0.001000 Time 0.024035 +2023-10-05 20:46:47,398 - Epoch: [10][ 220/ 1236] Overall Loss 0.490100 Objective Loss 0.490100 LR 0.001000 Time 0.023869 +2023-10-05 20:46:47,599 - Epoch: [10][ 230/ 1236] Overall Loss 0.486847 Objective Loss 0.486847 LR 0.001000 Time 0.023705 +2023-10-05 20:46:47,804 - Epoch: [10][ 240/ 1236] Overall Loss 0.485788 Objective Loss 0.485788 LR 0.001000 Time 0.023568 +2023-10-05 20:46:48,005 - Epoch: [10][ 250/ 1236] Overall Loss 0.485646 Objective Loss 0.485646 LR 0.001000 Time 0.023428 +2023-10-05 20:46:48,210 - Epoch: [10][ 260/ 1236] Overall Loss 0.483998 Objective Loss 0.483998 LR 0.001000 Time 0.023313 +2023-10-05 20:46:48,411 - Epoch: [10][ 270/ 1236] Overall Loss 0.482399 Objective Loss 0.482399 LR 0.001000 Time 0.023194 +2023-10-05 20:46:48,615 - Epoch: [10][ 280/ 1236] Overall Loss 0.481804 Objective Loss 0.481804 LR 0.001000 Time 0.023092 +2023-10-05 20:46:48,815 - Epoch: [10][ 290/ 1236] Overall Loss 0.480790 Objective Loss 0.480790 LR 0.001000 Time 0.022987 +2023-10-05 20:46:49,020 - Epoch: [10][ 300/ 1236] Overall Loss 0.479596 Objective Loss 0.479596 LR 0.001000 Time 0.022901 +2023-10-05 20:46:49,222 - Epoch: [10][ 310/ 1236] Overall Loss 0.478752 Objective Loss 0.478752 LR 0.001000 Time 0.022812 +2023-10-05 20:46:49,427 - Epoch: [10][ 320/ 1236] Overall Loss 0.477485 Objective Loss 0.477485 LR 0.001000 Time 0.022738 +2023-10-05 20:46:49,628 - Epoch: [10][ 330/ 1236] Overall Loss 0.475617 Objective Loss 0.475617 LR 0.001000 Time 0.022658 +2023-10-05 20:46:49,833 - Epoch: [10][ 340/ 1236] Overall Loss 0.473808 Objective Loss 0.473808 LR 0.001000 Time 0.022593 +2023-10-05 20:46:50,034 - Epoch: [10][ 350/ 1236] Overall Loss 0.472398 Objective Loss 0.472398 LR 0.001000 Time 0.022523 +2023-10-05 20:46:50,239 - Epoch: [10][ 360/ 1236] Overall Loss 0.471784 Objective Loss 0.471784 LR 0.001000 Time 0.022465 +2023-10-05 20:46:50,441 - Epoch: [10][ 370/ 1236] Overall Loss 0.470787 Objective Loss 0.470787 LR 0.001000 Time 0.022403 +2023-10-05 20:46:50,646 - Epoch: [10][ 380/ 1236] Overall Loss 0.469718 Objective Loss 0.469718 LR 0.001000 Time 0.022353 +2023-10-05 20:46:50,849 - Epoch: [10][ 390/ 1236] Overall Loss 0.469613 Objective Loss 0.469613 LR 0.001000 Time 0.022298 +2023-10-05 20:46:51,054 - Epoch: [10][ 400/ 1236] Overall Loss 0.469197 Objective Loss 0.469197 LR 0.001000 Time 0.022253 +2023-10-05 20:46:51,256 - Epoch: [10][ 410/ 1236] Overall Loss 0.468961 Objective Loss 0.468961 LR 0.001000 Time 0.022203 +2023-10-05 20:46:51,461 - Epoch: [10][ 420/ 1236] Overall Loss 0.468170 Objective Loss 0.468170 LR 0.001000 Time 0.022161 +2023-10-05 20:46:51,664 - Epoch: [10][ 430/ 1236] Overall Loss 0.467497 Objective Loss 0.467497 LR 0.001000 Time 0.022116 +2023-10-05 20:46:51,869 - Epoch: [10][ 440/ 1236] Overall Loss 0.467586 Objective Loss 0.467586 LR 0.001000 Time 0.022078 +2023-10-05 20:46:52,071 - Epoch: [10][ 450/ 1236] Overall Loss 0.467705 Objective Loss 0.467705 LR 0.001000 Time 0.022038 +2023-10-05 20:46:52,276 - Epoch: [10][ 460/ 1236] Overall Loss 0.467348 Objective Loss 0.467348 LR 0.001000 Time 0.022003 +2023-10-05 20:46:52,479 - Epoch: [10][ 470/ 1236] Overall Loss 0.467355 Objective Loss 0.467355 LR 0.001000 Time 0.021966 +2023-10-05 20:46:52,684 - Epoch: [10][ 480/ 1236] Overall Loss 0.466673 Objective Loss 0.466673 LR 0.001000 Time 0.021935 +2023-10-05 20:46:52,887 - Epoch: [10][ 490/ 1236] Overall Loss 0.466020 Objective Loss 0.466020 LR 0.001000 Time 0.021900 +2023-10-05 20:46:53,092 - Epoch: [10][ 500/ 1236] Overall Loss 0.465215 Objective Loss 0.465215 LR 0.001000 Time 0.021871 +2023-10-05 20:46:53,294 - Epoch: [10][ 510/ 1236] Overall Loss 0.465175 Objective Loss 0.465175 LR 0.001000 Time 0.021839 +2023-10-05 20:46:53,499 - Epoch: [10][ 520/ 1236] Overall Loss 0.464380 Objective Loss 0.464380 LR 0.001000 Time 0.021812 +2023-10-05 20:46:53,701 - Epoch: [10][ 530/ 1236] Overall Loss 0.464667 Objective Loss 0.464667 LR 0.001000 Time 0.021782 +2023-10-05 20:46:53,906 - Epoch: [10][ 540/ 1236] Overall Loss 0.464208 Objective Loss 0.464208 LR 0.001000 Time 0.021758 +2023-10-05 20:46:54,109 - Epoch: [10][ 550/ 1236] Overall Loss 0.463381 Objective Loss 0.463381 LR 0.001000 Time 0.021730 +2023-10-05 20:46:54,315 - Epoch: [10][ 560/ 1236] Overall Loss 0.463017 Objective Loss 0.463017 LR 0.001000 Time 0.021709 +2023-10-05 20:46:54,519 - Epoch: [10][ 570/ 1236] Overall Loss 0.462333 Objective Loss 0.462333 LR 0.001000 Time 0.021686 +2023-10-05 20:46:54,724 - Epoch: [10][ 580/ 1236] Overall Loss 0.462301 Objective Loss 0.462301 LR 0.001000 Time 0.021664 +2023-10-05 20:46:54,928 - Epoch: [10][ 590/ 1236] Overall Loss 0.461586 Objective Loss 0.461586 LR 0.001000 Time 0.021643 +2023-10-05 20:46:55,133 - Epoch: [10][ 600/ 1236] Overall Loss 0.461512 Objective Loss 0.461512 LR 0.001000 Time 0.021624 +2023-10-05 20:46:55,338 - Epoch: [10][ 610/ 1236] Overall Loss 0.460705 Objective Loss 0.460705 LR 0.001000 Time 0.021605 +2023-10-05 20:46:55,542 - Epoch: [10][ 620/ 1236] Overall Loss 0.460051 Objective Loss 0.460051 LR 0.001000 Time 0.021585 +2023-10-05 20:46:55,747 - Epoch: [10][ 630/ 1236] Overall Loss 0.459309 Objective Loss 0.459309 LR 0.001000 Time 0.021567 +2023-10-05 20:46:55,952 - Epoch: [10][ 640/ 1236] Overall Loss 0.459189 Objective Loss 0.459189 LR 0.001000 Time 0.021550 +2023-10-05 20:46:56,157 - Epoch: [10][ 650/ 1236] Overall Loss 0.458930 Objective Loss 0.458930 LR 0.001000 Time 0.021533 +2023-10-05 20:46:56,362 - Epoch: [10][ 660/ 1236] Overall Loss 0.458523 Objective Loss 0.458523 LR 0.001000 Time 0.021516 +2023-10-05 20:46:56,566 - Epoch: [10][ 670/ 1236] Overall Loss 0.457974 Objective Loss 0.457974 LR 0.001000 Time 0.021501 +2023-10-05 20:46:56,771 - Epoch: [10][ 680/ 1236] Overall Loss 0.457921 Objective Loss 0.457921 LR 0.001000 Time 0.021485 +2023-10-05 20:46:56,976 - Epoch: [10][ 690/ 1236] Overall Loss 0.458008 Objective Loss 0.458008 LR 0.001000 Time 0.021470 +2023-10-05 20:46:57,181 - Epoch: [10][ 700/ 1236] Overall Loss 0.457618 Objective Loss 0.457618 LR 0.001000 Time 0.021455 +2023-10-05 20:46:57,385 - Epoch: [10][ 710/ 1236] Overall Loss 0.457445 Objective Loss 0.457445 LR 0.001000 Time 0.021441 +2023-10-05 20:46:57,590 - Epoch: [10][ 720/ 1236] Overall Loss 0.457096 Objective Loss 0.457096 LR 0.001000 Time 0.021427 +2023-10-05 20:46:57,795 - Epoch: [10][ 730/ 1236] Overall Loss 0.456943 Objective Loss 0.456943 LR 0.001000 Time 0.021414 +2023-10-05 20:46:58,000 - Epoch: [10][ 740/ 1236] Overall Loss 0.456698 Objective Loss 0.456698 LR 0.001000 Time 0.021400 +2023-10-05 20:46:58,204 - Epoch: [10][ 750/ 1236] Overall Loss 0.456381 Objective Loss 0.456381 LR 0.001000 Time 0.021388 +2023-10-05 20:46:58,409 - Epoch: [10][ 760/ 1236] Overall Loss 0.455636 Objective Loss 0.455636 LR 0.001000 Time 0.021375 +2023-10-05 20:46:58,613 - Epoch: [10][ 770/ 1236] Overall Loss 0.456008 Objective Loss 0.456008 LR 0.001000 Time 0.021362 +2023-10-05 20:46:58,819 - Epoch: [10][ 780/ 1236] Overall Loss 0.455951 Objective Loss 0.455951 LR 0.001000 Time 0.021351 +2023-10-05 20:46:59,023 - Epoch: [10][ 790/ 1236] Overall Loss 0.455652 Objective Loss 0.455652 LR 0.001000 Time 0.021340 +2023-10-05 20:46:59,228 - Epoch: [10][ 800/ 1236] Overall Loss 0.455577 Objective Loss 0.455577 LR 0.001000 Time 0.021329 +2023-10-05 20:46:59,433 - Epoch: [10][ 810/ 1236] Overall Loss 0.455060 Objective Loss 0.455060 LR 0.001000 Time 0.021317 +2023-10-05 20:46:59,637 - Epoch: [10][ 820/ 1236] Overall Loss 0.455030 Objective Loss 0.455030 LR 0.001000 Time 0.021307 +2023-10-05 20:46:59,842 - Epoch: [10][ 830/ 1236] Overall Loss 0.454760 Objective Loss 0.454760 LR 0.001000 Time 0.021296 +2023-10-05 20:47:00,046 - Epoch: [10][ 840/ 1236] Overall Loss 0.454580 Objective Loss 0.454580 LR 0.001000 Time 0.021286 +2023-10-05 20:47:00,251 - Epoch: [10][ 850/ 1236] Overall Loss 0.454420 Objective Loss 0.454420 LR 0.001000 Time 0.021276 +2023-10-05 20:47:00,456 - Epoch: [10][ 860/ 1236] Overall Loss 0.454482 Objective Loss 0.454482 LR 0.001000 Time 0.021266 +2023-10-05 20:47:00,661 - Epoch: [10][ 870/ 1236] Overall Loss 0.454390 Objective Loss 0.454390 LR 0.001000 Time 0.021257 +2023-10-05 20:47:00,865 - Epoch: [10][ 880/ 1236] Overall Loss 0.453749 Objective Loss 0.453749 LR 0.001000 Time 0.021247 +2023-10-05 20:47:01,069 - Epoch: [10][ 890/ 1236] Overall Loss 0.454047 Objective Loss 0.454047 LR 0.001000 Time 0.021237 +2023-10-05 20:47:01,274 - Epoch: [10][ 900/ 1236] Overall Loss 0.453393 Objective Loss 0.453393 LR 0.001000 Time 0.021229 +2023-10-05 20:47:01,479 - Epoch: [10][ 910/ 1236] Overall Loss 0.453273 Objective Loss 0.453273 LR 0.001000 Time 0.021220 +2023-10-05 20:47:01,684 - Epoch: [10][ 920/ 1236] Overall Loss 0.453142 Objective Loss 0.453142 LR 0.001000 Time 0.021212 +2023-10-05 20:47:01,889 - Epoch: [10][ 930/ 1236] Overall Loss 0.453010 Objective Loss 0.453010 LR 0.001000 Time 0.021204 +2023-10-05 20:47:02,093 - Epoch: [10][ 940/ 1236] Overall Loss 0.452973 Objective Loss 0.452973 LR 0.001000 Time 0.021196 +2023-10-05 20:47:02,298 - Epoch: [10][ 950/ 1236] Overall Loss 0.452462 Objective Loss 0.452462 LR 0.001000 Time 0.021188 +2023-10-05 20:47:02,503 - Epoch: [10][ 960/ 1236] Overall Loss 0.452374 Objective Loss 0.452374 LR 0.001000 Time 0.021180 +2023-10-05 20:47:02,708 - Epoch: [10][ 970/ 1236] Overall Loss 0.451926 Objective Loss 0.451926 LR 0.001000 Time 0.021172 +2023-10-05 20:47:02,913 - Epoch: [10][ 980/ 1236] Overall Loss 0.451480 Objective Loss 0.451480 LR 0.001000 Time 0.021165 +2023-10-05 20:47:03,117 - Epoch: [10][ 990/ 1236] Overall Loss 0.451183 Objective Loss 0.451183 LR 0.001000 Time 0.021157 +2023-10-05 20:47:03,322 - Epoch: [10][ 1000/ 1236] Overall Loss 0.451206 Objective Loss 0.451206 LR 0.001000 Time 0.021151 +2023-10-05 20:47:03,527 - Epoch: [10][ 1010/ 1236] Overall Loss 0.451022 Objective Loss 0.451022 LR 0.001000 Time 0.021143 +2023-10-05 20:47:03,731 - Epoch: [10][ 1020/ 1236] Overall Loss 0.451060 Objective Loss 0.451060 LR 0.001000 Time 0.021136 +2023-10-05 20:47:03,936 - Epoch: [10][ 1030/ 1236] Overall Loss 0.451220 Objective Loss 0.451220 LR 0.001000 Time 0.021129 +2023-10-05 20:47:04,140 - Epoch: [10][ 1040/ 1236] Overall Loss 0.450987 Objective Loss 0.450987 LR 0.001000 Time 0.021123 +2023-10-05 20:47:04,345 - Epoch: [10][ 1050/ 1236] Overall Loss 0.450877 Objective Loss 0.450877 LR 0.001000 Time 0.021116 +2023-10-05 20:47:04,550 - Epoch: [10][ 1060/ 1236] Overall Loss 0.450521 Objective Loss 0.450521 LR 0.001000 Time 0.021110 +2023-10-05 20:47:04,754 - Epoch: [10][ 1070/ 1236] Overall Loss 0.450483 Objective Loss 0.450483 LR 0.001000 Time 0.021103 +2023-10-05 20:47:04,959 - Epoch: [10][ 1080/ 1236] Overall Loss 0.450861 Objective Loss 0.450861 LR 0.001000 Time 0.021098 +2023-10-05 20:47:05,164 - Epoch: [10][ 1090/ 1236] Overall Loss 0.450584 Objective Loss 0.450584 LR 0.001000 Time 0.021092 +2023-10-05 20:47:05,369 - Epoch: [10][ 1100/ 1236] Overall Loss 0.450318 Objective Loss 0.450318 LR 0.001000 Time 0.021085 +2023-10-05 20:47:05,573 - Epoch: [10][ 1110/ 1236] Overall Loss 0.450152 Objective Loss 0.450152 LR 0.001000 Time 0.021080 +2023-10-05 20:47:05,778 - Epoch: [10][ 1120/ 1236] Overall Loss 0.450181 Objective Loss 0.450181 LR 0.001000 Time 0.021074 +2023-10-05 20:47:05,983 - Epoch: [10][ 1130/ 1236] Overall Loss 0.450213 Objective Loss 0.450213 LR 0.001000 Time 0.021068 +2023-10-05 20:47:06,187 - Epoch: [10][ 1140/ 1236] Overall Loss 0.449958 Objective Loss 0.449958 LR 0.001000 Time 0.021063 +2023-10-05 20:47:06,392 - Epoch: [10][ 1150/ 1236] Overall Loss 0.449747 Objective Loss 0.449747 LR 0.001000 Time 0.021058 +2023-10-05 20:47:06,597 - Epoch: [10][ 1160/ 1236] Overall Loss 0.449737 Objective Loss 0.449737 LR 0.001000 Time 0.021052 +2023-10-05 20:47:06,802 - Epoch: [10][ 1170/ 1236] Overall Loss 0.449810 Objective Loss 0.449810 LR 0.001000 Time 0.021047 +2023-10-05 20:47:07,007 - Epoch: [10][ 1180/ 1236] Overall Loss 0.449479 Objective Loss 0.449479 LR 0.001000 Time 0.021042 +2023-10-05 20:47:07,212 - Epoch: [10][ 1190/ 1236] Overall Loss 0.449174 Objective Loss 0.449174 LR 0.001000 Time 0.021037 +2023-10-05 20:47:07,416 - Epoch: [10][ 1200/ 1236] Overall Loss 0.449359 Objective Loss 0.449359 LR 0.001000 Time 0.021032 +2023-10-05 20:47:07,621 - Epoch: [10][ 1210/ 1236] Overall Loss 0.449061 Objective Loss 0.449061 LR 0.001000 Time 0.021028 +2023-10-05 20:47:07,826 - Epoch: [10][ 1220/ 1236] Overall Loss 0.448731 Objective Loss 0.448731 LR 0.001000 Time 0.021023 +2023-10-05 20:47:08,085 - Epoch: [10][ 1230/ 1236] Overall Loss 0.448593 Objective Loss 0.448593 LR 0.001000 Time 0.021062 +2023-10-05 20:47:08,204 - Epoch: [10][ 1236/ 1236] Overall Loss 0.448471 Objective Loss 0.448471 Top1 78.818737 Top5 97.352342 LR 0.001000 Time 0.021056 +2023-10-05 20:47:08,346 - --- validate (epoch=10)----------- +2023-10-05 20:47:08,346 - 29943 samples (256 per mini-batch) +2023-10-05 20:47:08,799 - Epoch: [10][ 10/ 117] Loss 0.429634 Top1 76.523438 Top5 96.992188 +2023-10-05 20:47:08,948 - Epoch: [10][ 20/ 117] Loss 0.425545 Top1 75.996094 Top5 96.718750 +2023-10-05 20:47:09,093 - Epoch: [10][ 30/ 117] Loss 0.425781 Top1 76.276042 Top5 96.718750 +2023-10-05 20:47:09,240 - Epoch: [10][ 40/ 117] Loss 0.421235 Top1 76.513672 Top5 96.699219 +2023-10-05 20:47:09,386 - Epoch: [10][ 50/ 117] Loss 0.414346 Top1 76.328125 Top5 96.796875 +2023-10-05 20:47:09,532 - Epoch: [10][ 60/ 117] Loss 0.412725 Top1 76.412760 Top5 96.783854 +2023-10-05 20:47:09,678 - Epoch: [10][ 70/ 117] Loss 0.412591 Top1 76.551339 Top5 96.796875 +2023-10-05 20:47:09,824 - Epoch: [10][ 80/ 117] Loss 0.414145 Top1 76.528320 Top5 96.772461 +2023-10-05 20:47:09,969 - Epoch: [10][ 90/ 117] Loss 0.415894 Top1 76.375868 Top5 96.744792 +2023-10-05 20:47:10,115 - Epoch: [10][ 100/ 117] Loss 0.412792 Top1 76.582031 Top5 96.703125 +2023-10-05 20:47:10,268 - Epoch: [10][ 110/ 117] Loss 0.411975 Top1 76.576705 Top5 96.626420 +2023-10-05 20:47:10,352 - Epoch: [10][ 117/ 117] Loss 0.411240 Top1 76.562135 Top5 96.643623 +2023-10-05 20:47:10,478 - ==> Top1: 76.562 Top5: 96.644 Loss: 0.411 + +2023-10-05 20:47:10,479 - ==> Confusion: +[[ 917 2 9 1 10 4 0 1 9 75 1 2 1 2 1 2 2 0 0 0 11] + [ 3 1022 5 2 10 28 1 30 2 0 7 1 0 0 1 3 3 1 8 1 3] + [ 8 0 963 9 2 1 27 5 0 0 3 5 9 2 3 0 1 3 4 1 10] + [ 5 1 35 931 2 4 0 2 9 1 21 0 9 7 25 2 0 9 11 0 15] + [ 32 12 3 1 945 5 1 0 0 14 3 5 0 2 7 4 5 2 1 2 6] + [ 10 61 5 0 3 946 0 22 4 8 4 2 2 26 5 1 3 2 2 4 6] + [ 1 10 45 1 0 0 1085 10 0 0 10 4 2 0 0 3 0 1 0 13 6] + [ 6 17 36 0 0 43 5 1027 1 3 10 11 1 0 0 1 0 2 38 11 6] + [ 24 2 0 1 1 0 0 0 968 46 10 4 3 12 13 1 1 1 1 0 1] + [ 116 0 1 1 7 1 0 2 34 913 1 4 0 21 4 0 0 0 1 3 10] + [ 5 2 18 4 1 1 0 5 15 1 974 6 0 7 5 1 1 0 1 0 6] + [ 3 0 3 0 0 23 0 0 0 0 1 888 66 5 0 5 1 26 0 7 7] + [ 1 0 7 9 0 2 0 1 0 1 2 33 958 0 5 4 1 31 0 4 9] + [ 1 0 4 0 8 7 0 1 24 23 15 9 0 1009 5 2 0 0 0 1 10] + [ 16 3 3 21 6 1 0 0 43 7 3 0 4 3 965 0 0 5 11 0 10] + [ 2 4 9 2 8 2 7 0 0 0 0 10 9 2 0 1028 15 20 0 6 10] + [ 6 22 5 0 15 5 1 0 5 0 0 7 1 1 2 11 1061 0 0 3 16] + [ 2 0 5 3 0 0 2 0 1 0 0 7 26 0 1 6 1 980 1 0 3] + [ 3 7 18 19 0 0 0 27 8 0 12 2 6 0 6 0 1 1 947 0 11] + [ 1 3 9 0 2 10 9 8 0 0 5 35 8 1 0 2 12 1 1 1034 11] + [ 249 347 385 97 187 259 69 107 190 159 359 179 547 364 185 91 171 121 167 308 3364]] + +2023-10-05 20:47:10,480 - ==> Best [Top1: 76.562 Top5: 96.644 Sparsity:0.00 Params: 148928 on epoch: 10] +2023-10-05 20:47:10,480 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:47:10,486 - + +2023-10-05 20:47:10,487 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:47:11,464 - Epoch: [11][ 10/ 1236] Overall Loss 0.391190 Objective Loss 0.391190 LR 0.001000 Time 0.097644 +2023-10-05 20:47:11,667 - Epoch: [11][ 20/ 1236] Overall Loss 0.417701 Objective Loss 0.417701 LR 0.001000 Time 0.058976 +2023-10-05 20:47:11,869 - Epoch: [11][ 30/ 1236] Overall Loss 0.422722 Objective Loss 0.422722 LR 0.001000 Time 0.046024 +2023-10-05 20:47:12,072 - Epoch: [11][ 40/ 1236] Overall Loss 0.423963 Objective Loss 0.423963 LR 0.001000 Time 0.039598 +2023-10-05 20:47:12,274 - Epoch: [11][ 50/ 1236] Overall Loss 0.430239 Objective Loss 0.430239 LR 0.001000 Time 0.035703 +2023-10-05 20:47:12,477 - Epoch: [11][ 60/ 1236] Overall Loss 0.428440 Objective Loss 0.428440 LR 0.001000 Time 0.033138 +2023-10-05 20:47:12,679 - Epoch: [11][ 70/ 1236] Overall Loss 0.424648 Objective Loss 0.424648 LR 0.001000 Time 0.031278 +2023-10-05 20:47:12,882 - Epoch: [11][ 80/ 1236] Overall Loss 0.421533 Objective Loss 0.421533 LR 0.001000 Time 0.029906 +2023-10-05 20:47:13,085 - Epoch: [11][ 90/ 1236] Overall Loss 0.422560 Objective Loss 0.422560 LR 0.001000 Time 0.028838 +2023-10-05 20:47:13,289 - Epoch: [11][ 100/ 1236] Overall Loss 0.422044 Objective Loss 0.422044 LR 0.001000 Time 0.027989 +2023-10-05 20:47:13,492 - Epoch: [11][ 110/ 1236] Overall Loss 0.423404 Objective Loss 0.423404 LR 0.001000 Time 0.027289 +2023-10-05 20:47:13,696 - Epoch: [11][ 120/ 1236] Overall Loss 0.422009 Objective Loss 0.422009 LR 0.001000 Time 0.026712 +2023-10-05 20:47:13,900 - Epoch: [11][ 130/ 1236] Overall Loss 0.420811 Objective Loss 0.420811 LR 0.001000 Time 0.026220 +2023-10-05 20:47:14,104 - Epoch: [11][ 140/ 1236] Overall Loss 0.417419 Objective Loss 0.417419 LR 0.001000 Time 0.025804 +2023-10-05 20:47:14,307 - Epoch: [11][ 150/ 1236] Overall Loss 0.416854 Objective Loss 0.416854 LR 0.001000 Time 0.025436 +2023-10-05 20:47:14,511 - Epoch: [11][ 160/ 1236] Overall Loss 0.418005 Objective Loss 0.418005 LR 0.001000 Time 0.025119 +2023-10-05 20:47:14,715 - Epoch: [11][ 170/ 1236] Overall Loss 0.417779 Objective Loss 0.417779 LR 0.001000 Time 0.024837 +2023-10-05 20:47:14,919 - Epoch: [11][ 180/ 1236] Overall Loss 0.416728 Objective Loss 0.416728 LR 0.001000 Time 0.024589 +2023-10-05 20:47:15,122 - Epoch: [11][ 190/ 1236] Overall Loss 0.415536 Objective Loss 0.415536 LR 0.001000 Time 0.024363 +2023-10-05 20:47:15,326 - Epoch: [11][ 200/ 1236] Overall Loss 0.415492 Objective Loss 0.415492 LR 0.001000 Time 0.024162 +2023-10-05 20:47:15,529 - Epoch: [11][ 210/ 1236] Overall Loss 0.415645 Objective Loss 0.415645 LR 0.001000 Time 0.023977 +2023-10-05 20:47:15,732 - Epoch: [11][ 220/ 1236] Overall Loss 0.417379 Objective Loss 0.417379 LR 0.001000 Time 0.023810 +2023-10-05 20:47:15,936 - Epoch: [11][ 230/ 1236] Overall Loss 0.416876 Objective Loss 0.416876 LR 0.001000 Time 0.023658 +2023-10-05 20:47:16,140 - Epoch: [11][ 240/ 1236] Overall Loss 0.415547 Objective Loss 0.415547 LR 0.001000 Time 0.023520 +2023-10-05 20:47:16,343 - Epoch: [11][ 250/ 1236] Overall Loss 0.417872 Objective Loss 0.417872 LR 0.001000 Time 0.023391 +2023-10-05 20:47:16,547 - Epoch: [11][ 260/ 1236] Overall Loss 0.418266 Objective Loss 0.418266 LR 0.001000 Time 0.023275 +2023-10-05 20:47:16,750 - Epoch: [11][ 270/ 1236] Overall Loss 0.417562 Objective Loss 0.417562 LR 0.001000 Time 0.023165 +2023-10-05 20:47:16,954 - Epoch: [11][ 280/ 1236] Overall Loss 0.418620 Objective Loss 0.418620 LR 0.001000 Time 0.023064 +2023-10-05 20:47:17,157 - Epoch: [11][ 290/ 1236] Overall Loss 0.418363 Objective Loss 0.418363 LR 0.001000 Time 0.022969 +2023-10-05 20:47:17,362 - Epoch: [11][ 300/ 1236] Overall Loss 0.418380 Objective Loss 0.418380 LR 0.001000 Time 0.022885 +2023-10-05 20:47:17,565 - Epoch: [11][ 310/ 1236] Overall Loss 0.418979 Objective Loss 0.418979 LR 0.001000 Time 0.022801 +2023-10-05 20:47:17,770 - Epoch: [11][ 320/ 1236] Overall Loss 0.418764 Objective Loss 0.418764 LR 0.001000 Time 0.022728 +2023-10-05 20:47:17,974 - Epoch: [11][ 330/ 1236] Overall Loss 0.420028 Objective Loss 0.420028 LR 0.001000 Time 0.022656 +2023-10-05 20:47:18,179 - Epoch: [11][ 340/ 1236] Overall Loss 0.419671 Objective Loss 0.419671 LR 0.001000 Time 0.022590 +2023-10-05 20:47:18,382 - Epoch: [11][ 350/ 1236] Overall Loss 0.419666 Objective Loss 0.419666 LR 0.001000 Time 0.022526 +2023-10-05 20:47:18,587 - Epoch: [11][ 360/ 1236] Overall Loss 0.418891 Objective Loss 0.418891 LR 0.001000 Time 0.022467 +2023-10-05 20:47:18,789 - Epoch: [11][ 370/ 1236] Overall Loss 0.418251 Objective Loss 0.418251 LR 0.001000 Time 0.022406 +2023-10-05 20:47:18,993 - Epoch: [11][ 380/ 1236] Overall Loss 0.418314 Objective Loss 0.418314 LR 0.001000 Time 0.022350 +2023-10-05 20:47:19,195 - Epoch: [11][ 390/ 1236] Overall Loss 0.417582 Objective Loss 0.417582 LR 0.001000 Time 0.022295 +2023-10-05 20:47:19,398 - Epoch: [11][ 400/ 1236] Overall Loss 0.417424 Objective Loss 0.417424 LR 0.001000 Time 0.022245 +2023-10-05 20:47:19,600 - Epoch: [11][ 410/ 1236] Overall Loss 0.417074 Objective Loss 0.417074 LR 0.001000 Time 0.022195 +2023-10-05 20:47:19,804 - Epoch: [11][ 420/ 1236] Overall Loss 0.417774 Objective Loss 0.417774 LR 0.001000 Time 0.022150 +2023-10-05 20:47:20,006 - Epoch: [11][ 430/ 1236] Overall Loss 0.417296 Objective Loss 0.417296 LR 0.001000 Time 0.022105 +2023-10-05 20:47:20,209 - Epoch: [11][ 440/ 1236] Overall Loss 0.417294 Objective Loss 0.417294 LR 0.001000 Time 0.022064 +2023-10-05 20:47:20,411 - Epoch: [11][ 450/ 1236] Overall Loss 0.417594 Objective Loss 0.417594 LR 0.001000 Time 0.022021 +2023-10-05 20:47:20,618 - Epoch: [11][ 460/ 1236] Overall Loss 0.417445 Objective Loss 0.417445 LR 0.001000 Time 0.021981 +2023-10-05 20:47:20,819 - Epoch: [11][ 470/ 1236] Overall Loss 0.417615 Objective Loss 0.417615 LR 0.001000 Time 0.021942 +2023-10-05 20:47:21,023 - Epoch: [11][ 480/ 1236] Overall Loss 0.417027 Objective Loss 0.417027 LR 0.001000 Time 0.021908 +2023-10-05 20:47:21,224 - Epoch: [11][ 490/ 1236] Overall Loss 0.416679 Objective Loss 0.416679 LR 0.001000 Time 0.021871 +2023-10-05 20:47:21,428 - Epoch: [11][ 500/ 1236] Overall Loss 0.416887 Objective Loss 0.416887 LR 0.001000 Time 0.021840 +2023-10-05 20:47:21,631 - Epoch: [11][ 510/ 1236] Overall Loss 0.417088 Objective Loss 0.417088 LR 0.001000 Time 0.021811 +2023-10-05 20:47:21,835 - Epoch: [11][ 520/ 1236] Overall Loss 0.416969 Objective Loss 0.416969 LR 0.001000 Time 0.021783 +2023-10-05 20:47:22,037 - Epoch: [11][ 530/ 1236] Overall Loss 0.416537 Objective Loss 0.416537 LR 0.001000 Time 0.021752 +2023-10-05 20:47:22,241 - Epoch: [11][ 540/ 1236] Overall Loss 0.416304 Objective Loss 0.416304 LR 0.001000 Time 0.021725 +2023-10-05 20:47:22,443 - Epoch: [11][ 550/ 1236] Overall Loss 0.415980 Objective Loss 0.415980 LR 0.001000 Time 0.021697 +2023-10-05 20:47:22,647 - Epoch: [11][ 560/ 1236] Overall Loss 0.416422 Objective Loss 0.416422 LR 0.001000 Time 0.021673 +2023-10-05 20:47:22,849 - Epoch: [11][ 570/ 1236] Overall Loss 0.416194 Objective Loss 0.416194 LR 0.001000 Time 0.021647 +2023-10-05 20:47:23,053 - Epoch: [11][ 580/ 1236] Overall Loss 0.415944 Objective Loss 0.415944 LR 0.001000 Time 0.021624 +2023-10-05 20:47:23,255 - Epoch: [11][ 590/ 1236] Overall Loss 0.415324 Objective Loss 0.415324 LR 0.001000 Time 0.021600 +2023-10-05 20:47:23,459 - Epoch: [11][ 600/ 1236] Overall Loss 0.415358 Objective Loss 0.415358 LR 0.001000 Time 0.021579 +2023-10-05 20:47:23,661 - Epoch: [11][ 610/ 1236] Overall Loss 0.414683 Objective Loss 0.414683 LR 0.001000 Time 0.021557 +2023-10-05 20:47:23,865 - Epoch: [11][ 620/ 1236] Overall Loss 0.414550 Objective Loss 0.414550 LR 0.001000 Time 0.021537 +2023-10-05 20:47:24,067 - Epoch: [11][ 630/ 1236] Overall Loss 0.414175 Objective Loss 0.414175 LR 0.001000 Time 0.021515 +2023-10-05 20:47:24,271 - Epoch: [11][ 640/ 1236] Overall Loss 0.413606 Objective Loss 0.413606 LR 0.001000 Time 0.021497 +2023-10-05 20:47:24,473 - Epoch: [11][ 650/ 1236] Overall Loss 0.413800 Objective Loss 0.413800 LR 0.001000 Time 0.021477 +2023-10-05 20:47:24,677 - Epoch: [11][ 660/ 1236] Overall Loss 0.413657 Objective Loss 0.413657 LR 0.001000 Time 0.021460 +2023-10-05 20:47:24,879 - Epoch: [11][ 670/ 1236] Overall Loss 0.413671 Objective Loss 0.413671 LR 0.001000 Time 0.021440 +2023-10-05 20:47:25,083 - Epoch: [11][ 680/ 1236] Overall Loss 0.413916 Objective Loss 0.413916 LR 0.001000 Time 0.021425 +2023-10-05 20:47:25,285 - Epoch: [11][ 690/ 1236] Overall Loss 0.413876 Objective Loss 0.413876 LR 0.001000 Time 0.021407 +2023-10-05 20:47:25,489 - Epoch: [11][ 700/ 1236] Overall Loss 0.414569 Objective Loss 0.414569 LR 0.001000 Time 0.021392 +2023-10-05 20:47:25,691 - Epoch: [11][ 710/ 1236] Overall Loss 0.414861 Objective Loss 0.414861 LR 0.001000 Time 0.021374 +2023-10-05 20:47:25,895 - Epoch: [11][ 720/ 1236] Overall Loss 0.414467 Objective Loss 0.414467 LR 0.001000 Time 0.021360 +2023-10-05 20:47:26,097 - Epoch: [11][ 730/ 1236] Overall Loss 0.414159 Objective Loss 0.414159 LR 0.001000 Time 0.021344 +2023-10-05 20:47:26,300 - Epoch: [11][ 740/ 1236] Overall Loss 0.414085 Objective Loss 0.414085 LR 0.001000 Time 0.021330 +2023-10-05 20:47:26,502 - Epoch: [11][ 750/ 1236] Overall Loss 0.413940 Objective Loss 0.413940 LR 0.001000 Time 0.021314 +2023-10-05 20:47:26,706 - Epoch: [11][ 760/ 1236] Overall Loss 0.413654 Objective Loss 0.413654 LR 0.001000 Time 0.021302 +2023-10-05 20:47:26,908 - Epoch: [11][ 770/ 1236] Overall Loss 0.413486 Objective Loss 0.413486 LR 0.001000 Time 0.021287 +2023-10-05 20:47:27,112 - Epoch: [11][ 780/ 1236] Overall Loss 0.413710 Objective Loss 0.413710 LR 0.001000 Time 0.021274 +2023-10-05 20:47:27,314 - Epoch: [11][ 790/ 1236] Overall Loss 0.413795 Objective Loss 0.413795 LR 0.001000 Time 0.021261 +2023-10-05 20:47:27,518 - Epoch: [11][ 800/ 1236] Overall Loss 0.413540 Objective Loss 0.413540 LR 0.001000 Time 0.021249 +2023-10-05 20:47:27,720 - Epoch: [11][ 810/ 1236] Overall Loss 0.413930 Objective Loss 0.413930 LR 0.001000 Time 0.021236 +2023-10-05 20:47:27,924 - Epoch: [11][ 820/ 1236] Overall Loss 0.414058 Objective Loss 0.414058 LR 0.001000 Time 0.021225 +2023-10-05 20:47:28,127 - Epoch: [11][ 830/ 1236] Overall Loss 0.414060 Objective Loss 0.414060 LR 0.001000 Time 0.021213 +2023-10-05 20:47:28,330 - Epoch: [11][ 840/ 1236] Overall Loss 0.413955 Objective Loss 0.413955 LR 0.001000 Time 0.021203 +2023-10-05 20:47:28,532 - Epoch: [11][ 850/ 1236] Overall Loss 0.414019 Objective Loss 0.414019 LR 0.001000 Time 0.021191 +2023-10-05 20:47:28,736 - Epoch: [11][ 860/ 1236] Overall Loss 0.414515 Objective Loss 0.414515 LR 0.001000 Time 0.021181 +2023-10-05 20:47:28,938 - Epoch: [11][ 870/ 1236] Overall Loss 0.414601 Objective Loss 0.414601 LR 0.001000 Time 0.021170 +2023-10-05 20:47:29,142 - Epoch: [11][ 880/ 1236] Overall Loss 0.414378 Objective Loss 0.414378 LR 0.001000 Time 0.021160 +2023-10-05 20:47:29,344 - Epoch: [11][ 890/ 1236] Overall Loss 0.414430 Objective Loss 0.414430 LR 0.001000 Time 0.021149 +2023-10-05 20:47:29,548 - Epoch: [11][ 900/ 1236] Overall Loss 0.414069 Objective Loss 0.414069 LR 0.001000 Time 0.021140 +2023-10-05 20:47:29,750 - Epoch: [11][ 910/ 1236] Overall Loss 0.414173 Objective Loss 0.414173 LR 0.001000 Time 0.021130 +2023-10-05 20:47:29,954 - Epoch: [11][ 920/ 1236] Overall Loss 0.414098 Objective Loss 0.414098 LR 0.001000 Time 0.021121 +2023-10-05 20:47:30,156 - Epoch: [11][ 930/ 1236] Overall Loss 0.414292 Objective Loss 0.414292 LR 0.001000 Time 0.021111 +2023-10-05 20:47:30,360 - Epoch: [11][ 940/ 1236] Overall Loss 0.414373 Objective Loss 0.414373 LR 0.001000 Time 0.021103 +2023-10-05 20:47:30,562 - Epoch: [11][ 950/ 1236] Overall Loss 0.414317 Objective Loss 0.414317 LR 0.001000 Time 0.021093 +2023-10-05 20:47:30,766 - Epoch: [11][ 960/ 1236] Overall Loss 0.414252 Objective Loss 0.414252 LR 0.001000 Time 0.021086 +2023-10-05 20:47:30,968 - Epoch: [11][ 970/ 1236] Overall Loss 0.414095 Objective Loss 0.414095 LR 0.001000 Time 0.021076 +2023-10-05 20:47:31,172 - Epoch: [11][ 980/ 1236] Overall Loss 0.413948 Objective Loss 0.413948 LR 0.001000 Time 0.021069 +2023-10-05 20:47:31,374 - Epoch: [11][ 990/ 1236] Overall Loss 0.413749 Objective Loss 0.413749 LR 0.001000 Time 0.021060 +2023-10-05 20:47:31,578 - Epoch: [11][ 1000/ 1236] Overall Loss 0.413679 Objective Loss 0.413679 LR 0.001000 Time 0.021053 +2023-10-05 20:47:31,780 - Epoch: [11][ 1010/ 1236] Overall Loss 0.413563 Objective Loss 0.413563 LR 0.001000 Time 0.021044 +2023-10-05 20:47:31,985 - Epoch: [11][ 1020/ 1236] Overall Loss 0.413687 Objective Loss 0.413687 LR 0.001000 Time 0.021038 +2023-10-05 20:47:32,187 - Epoch: [11][ 1030/ 1236] Overall Loss 0.413774 Objective Loss 0.413774 LR 0.001000 Time 0.021030 +2023-10-05 20:47:32,391 - Epoch: [11][ 1040/ 1236] Overall Loss 0.413786 Objective Loss 0.413786 LR 0.001000 Time 0.021023 +2023-10-05 20:47:32,593 - Epoch: [11][ 1050/ 1236] Overall Loss 0.413696 Objective Loss 0.413696 LR 0.001000 Time 0.021015 +2023-10-05 20:47:32,797 - Epoch: [11][ 1060/ 1236] Overall Loss 0.413705 Objective Loss 0.413705 LR 0.001000 Time 0.021009 +2023-10-05 20:47:32,999 - Epoch: [11][ 1070/ 1236] Overall Loss 0.413777 Objective Loss 0.413777 LR 0.001000 Time 0.021001 +2023-10-05 20:47:33,203 - Epoch: [11][ 1080/ 1236] Overall Loss 0.413701 Objective Loss 0.413701 LR 0.001000 Time 0.020995 +2023-10-05 20:47:33,405 - Epoch: [11][ 1090/ 1236] Overall Loss 0.413761 Objective Loss 0.413761 LR 0.001000 Time 0.020987 +2023-10-05 20:47:33,610 - Epoch: [11][ 1100/ 1236] Overall Loss 0.413554 Objective Loss 0.413554 LR 0.001000 Time 0.020983 +2023-10-05 20:47:33,812 - Epoch: [11][ 1110/ 1236] Overall Loss 0.413883 Objective Loss 0.413883 LR 0.001000 Time 0.020976 +2023-10-05 20:47:34,016 - Epoch: [11][ 1120/ 1236] Overall Loss 0.413975 Objective Loss 0.413975 LR 0.001000 Time 0.020970 +2023-10-05 20:47:34,218 - Epoch: [11][ 1130/ 1236] Overall Loss 0.413975 Objective Loss 0.413975 LR 0.001000 Time 0.020963 +2023-10-05 20:47:34,422 - Epoch: [11][ 1140/ 1236] Overall Loss 0.414161 Objective Loss 0.414161 LR 0.001000 Time 0.020958 +2023-10-05 20:47:34,624 - Epoch: [11][ 1150/ 1236] Overall Loss 0.414325 Objective Loss 0.414325 LR 0.001000 Time 0.020951 +2023-10-05 20:47:34,828 - Epoch: [11][ 1160/ 1236] Overall Loss 0.414023 Objective Loss 0.414023 LR 0.001000 Time 0.020946 +2023-10-05 20:47:35,030 - Epoch: [11][ 1170/ 1236] Overall Loss 0.413814 Objective Loss 0.413814 LR 0.001000 Time 0.020939 +2023-10-05 20:47:35,234 - Epoch: [11][ 1180/ 1236] Overall Loss 0.414491 Objective Loss 0.414491 LR 0.001000 Time 0.020934 +2023-10-05 20:47:35,437 - Epoch: [11][ 1190/ 1236] Overall Loss 0.414668 Objective Loss 0.414668 LR 0.001000 Time 0.020928 +2023-10-05 20:47:35,641 - Epoch: [11][ 1200/ 1236] Overall Loss 0.414878 Objective Loss 0.414878 LR 0.001000 Time 0.020924 +2023-10-05 20:47:35,844 - Epoch: [11][ 1210/ 1236] Overall Loss 0.415139 Objective Loss 0.415139 LR 0.001000 Time 0.020919 +2023-10-05 20:47:36,048 - Epoch: [11][ 1220/ 1236] Overall Loss 0.415150 Objective Loss 0.415150 LR 0.001000 Time 0.020914 +2023-10-05 20:47:36,305 - Epoch: [11][ 1230/ 1236] Overall Loss 0.415292 Objective Loss 0.415292 LR 0.001000 Time 0.020952 +2023-10-05 20:47:36,422 - Epoch: [11][ 1236/ 1236] Overall Loss 0.415468 Objective Loss 0.415468 Top1 80.244399 Top5 95.723014 LR 0.001000 Time 0.020945 +2023-10-05 20:47:36,564 - --- validate (epoch=11)----------- +2023-10-05 20:47:36,564 - 29943 samples (256 per mini-batch) +2023-10-05 20:47:37,017 - Epoch: [11][ 10/ 117] Loss 0.377211 Top1 78.984375 Top5 97.031250 +2023-10-05 20:47:37,164 - Epoch: [11][ 20/ 117] Loss 0.399264 Top1 78.906250 Top5 96.972656 +2023-10-05 20:47:37,313 - Epoch: [11][ 30/ 117] Loss 0.417423 Top1 78.059896 Top5 96.731771 +2023-10-05 20:47:37,460 - Epoch: [11][ 40/ 117] Loss 0.411052 Top1 78.212891 Top5 96.923828 +2023-10-05 20:47:37,607 - Epoch: [11][ 50/ 117] Loss 0.403371 Top1 78.351562 Top5 96.945312 +2023-10-05 20:47:37,754 - Epoch: [11][ 60/ 117] Loss 0.407047 Top1 78.346354 Top5 96.920573 +2023-10-05 20:47:37,900 - Epoch: [11][ 70/ 117] Loss 0.408096 Top1 78.443080 Top5 96.902902 +2023-10-05 20:47:38,048 - Epoch: [11][ 80/ 117] Loss 0.409869 Top1 78.256836 Top5 96.870117 +2023-10-05 20:47:38,196 - Epoch: [11][ 90/ 117] Loss 0.407811 Top1 78.376736 Top5 96.879340 +2023-10-05 20:47:38,343 - Epoch: [11][ 100/ 117] Loss 0.410076 Top1 78.292969 Top5 96.929688 +2023-10-05 20:47:38,497 - Epoch: [11][ 110/ 117] Loss 0.410635 Top1 78.256392 Top5 96.853693 +2023-10-05 20:47:38,581 - Epoch: [11][ 117/ 117] Loss 0.409920 Top1 78.245333 Top5 96.870721 +2023-10-05 20:47:38,696 - ==> Top1: 78.245 Top5: 96.871 Loss: 0.410 + +2023-10-05 20:47:38,697 - ==> Confusion: +[[ 928 2 9 1 14 4 0 0 0 56 1 0 0 4 9 6 1 0 0 0 15] + [ 2 981 2 2 13 66 5 25 2 0 4 2 0 1 1 3 7 1 8 3 3] + [ 7 1 919 11 4 2 48 14 0 2 2 5 3 3 1 6 0 5 6 2 15] + [ 1 1 32 942 2 5 4 1 0 3 16 0 8 1 25 5 2 8 12 2 19] + [ 21 9 3 0 972 9 3 0 0 7 1 3 0 1 4 5 7 0 1 0 4] + [ 13 22 4 2 6 975 2 23 0 4 1 17 2 18 4 0 4 0 2 6 11] + [ 2 6 22 0 0 2 1115 15 0 1 1 2 1 1 0 10 1 0 0 7 5] + [ 9 13 23 0 0 39 6 1052 1 3 5 9 2 1 0 4 1 4 29 13 4] + [ 29 3 0 1 1 6 0 0 870 52 29 4 0 33 48 1 0 0 5 0 7] + [ 114 0 4 0 13 3 1 1 22 890 3 1 0 48 3 5 1 1 0 1 8] + [ 10 2 15 6 1 1 3 4 9 2 963 1 1 8 6 3 0 2 1 2 13] + [ 3 0 1 0 1 15 1 3 0 0 0 923 30 6 1 9 2 17 0 16 7] + [ 1 0 3 5 1 6 0 2 0 3 0 41 959 1 3 12 4 14 0 7 6] + [ 3 0 1 0 6 21 0 2 5 12 6 2 4 1023 3 4 4 1 0 8 14] + [ 15 0 5 28 15 0 0 0 4 6 4 0 3 1 984 0 3 2 8 0 23] + [ 1 3 5 1 4 1 2 1 0 1 0 9 6 1 0 1062 16 6 0 8 7] + [ 2 12 3 0 13 10 1 0 2 0 0 4 0 2 2 23 1070 2 0 4 11] + [ 5 0 3 4 1 0 0 0 0 0 0 11 47 3 1 25 2 928 1 1 6] + [ 4 12 15 20 2 0 1 58 1 1 8 0 0 0 20 2 0 4 903 1 16] + [ 3 2 5 0 1 9 14 16 0 1 1 17 4 1 0 5 5 3 1 1054 10] + [ 217 241 337 93 183 348 84 163 40 90 201 131 475 403 132 179 167 53 108 344 3916]] + +2023-10-05 20:47:38,698 - ==> Best [Top1: 78.245 Top5: 96.871 Sparsity:0.00 Params: 148928 on epoch: 11] +2023-10-05 20:47:38,698 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:47:38,711 - + +2023-10-05 20:47:38,711 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:47:39,678 - Epoch: [12][ 10/ 1236] Overall Loss 0.381080 Objective Loss 0.381080 LR 0.001000 Time 0.096603 +2023-10-05 20:47:39,880 - Epoch: [12][ 20/ 1236] Overall Loss 0.381897 Objective Loss 0.381897 LR 0.001000 Time 0.058391 +2023-10-05 20:47:40,080 - Epoch: [12][ 30/ 1236] Overall Loss 0.387779 Objective Loss 0.387779 LR 0.001000 Time 0.045581 +2023-10-05 20:47:40,282 - Epoch: [12][ 40/ 1236] Overall Loss 0.391301 Objective Loss 0.391301 LR 0.001000 Time 0.039234 +2023-10-05 20:47:40,482 - Epoch: [12][ 50/ 1236] Overall Loss 0.386686 Objective Loss 0.386686 LR 0.001000 Time 0.035379 +2023-10-05 20:47:40,684 - Epoch: [12][ 60/ 1236] Overall Loss 0.397744 Objective Loss 0.397744 LR 0.001000 Time 0.032852 +2023-10-05 20:47:40,884 - Epoch: [12][ 70/ 1236] Overall Loss 0.398336 Objective Loss 0.398336 LR 0.001000 Time 0.031011 +2023-10-05 20:47:41,087 - Epoch: [12][ 80/ 1236] Overall Loss 0.400259 Objective Loss 0.400259 LR 0.001000 Time 0.029658 +2023-10-05 20:47:41,286 - Epoch: [12][ 90/ 1236] Overall Loss 0.405645 Objective Loss 0.405645 LR 0.001000 Time 0.028571 +2023-10-05 20:47:41,487 - Epoch: [12][ 100/ 1236] Overall Loss 0.401197 Objective Loss 0.401197 LR 0.001000 Time 0.027726 +2023-10-05 20:47:41,687 - Epoch: [12][ 110/ 1236] Overall Loss 0.401304 Objective Loss 0.401304 LR 0.001000 Time 0.027016 +2023-10-05 20:47:41,888 - Epoch: [12][ 120/ 1236] Overall Loss 0.400672 Objective Loss 0.400672 LR 0.001000 Time 0.026442 +2023-10-05 20:47:42,089 - Epoch: [12][ 130/ 1236] Overall Loss 0.398157 Objective Loss 0.398157 LR 0.001000 Time 0.025945 +2023-10-05 20:47:42,289 - Epoch: [12][ 140/ 1236] Overall Loss 0.400014 Objective Loss 0.400014 LR 0.001000 Time 0.025524 +2023-10-05 20:47:42,490 - Epoch: [12][ 150/ 1236] Overall Loss 0.400860 Objective Loss 0.400860 LR 0.001000 Time 0.025154 +2023-10-05 20:47:42,691 - Epoch: [12][ 160/ 1236] Overall Loss 0.399706 Objective Loss 0.399706 LR 0.001000 Time 0.024836 +2023-10-05 20:47:42,891 - Epoch: [12][ 170/ 1236] Overall Loss 0.400811 Objective Loss 0.400811 LR 0.001000 Time 0.024550 +2023-10-05 20:47:43,092 - Epoch: [12][ 180/ 1236] Overall Loss 0.403447 Objective Loss 0.403447 LR 0.001000 Time 0.024300 +2023-10-05 20:47:43,292 - Epoch: [12][ 190/ 1236] Overall Loss 0.403847 Objective Loss 0.403847 LR 0.001000 Time 0.024073 +2023-10-05 20:47:43,493 - Epoch: [12][ 200/ 1236] Overall Loss 0.405295 Objective Loss 0.405295 LR 0.001000 Time 0.023872 +2023-10-05 20:47:43,693 - Epoch: [12][ 210/ 1236] Overall Loss 0.406363 Objective Loss 0.406363 LR 0.001000 Time 0.023686 +2023-10-05 20:47:43,894 - Epoch: [12][ 220/ 1236] Overall Loss 0.404307 Objective Loss 0.404307 LR 0.001000 Time 0.023522 +2023-10-05 20:47:44,093 - Epoch: [12][ 230/ 1236] Overall Loss 0.403891 Objective Loss 0.403891 LR 0.001000 Time 0.023366 +2023-10-05 20:47:44,295 - Epoch: [12][ 240/ 1236] Overall Loss 0.403804 Objective Loss 0.403804 LR 0.001000 Time 0.023231 +2023-10-05 20:47:44,495 - Epoch: [12][ 250/ 1236] Overall Loss 0.404563 Objective Loss 0.404563 LR 0.001000 Time 0.023100 +2023-10-05 20:47:44,696 - Epoch: [12][ 260/ 1236] Overall Loss 0.402705 Objective Loss 0.402705 LR 0.001000 Time 0.022984 +2023-10-05 20:47:44,896 - Epoch: [12][ 270/ 1236] Overall Loss 0.402823 Objective Loss 0.402823 LR 0.001000 Time 0.022871 +2023-10-05 20:47:45,097 - Epoch: [12][ 280/ 1236] Overall Loss 0.403096 Objective Loss 0.403096 LR 0.001000 Time 0.022771 +2023-10-05 20:47:45,297 - Epoch: [12][ 290/ 1236] Overall Loss 0.404315 Objective Loss 0.404315 LR 0.001000 Time 0.022675 +2023-10-05 20:47:45,498 - Epoch: [12][ 300/ 1236] Overall Loss 0.403559 Objective Loss 0.403559 LR 0.001000 Time 0.022589 +2023-10-05 20:47:45,699 - Epoch: [12][ 310/ 1236] Overall Loss 0.403591 Objective Loss 0.403591 LR 0.001000 Time 0.022504 +2023-10-05 20:47:45,900 - Epoch: [12][ 320/ 1236] Overall Loss 0.404666 Objective Loss 0.404666 LR 0.001000 Time 0.022429 +2023-10-05 20:47:46,101 - Epoch: [12][ 330/ 1236] Overall Loss 0.404342 Objective Loss 0.404342 LR 0.001000 Time 0.022357 +2023-10-05 20:47:46,303 - Epoch: [12][ 340/ 1236] Overall Loss 0.405468 Objective Loss 0.405468 LR 0.001000 Time 0.022295 +2023-10-05 20:47:46,504 - Epoch: [12][ 350/ 1236] Overall Loss 0.404932 Objective Loss 0.404932 LR 0.001000 Time 0.022230 +2023-10-05 20:47:46,707 - Epoch: [12][ 360/ 1236] Overall Loss 0.405115 Objective Loss 0.405115 LR 0.001000 Time 0.022175 +2023-10-05 20:47:46,908 - Epoch: [12][ 370/ 1236] Overall Loss 0.404596 Objective Loss 0.404596 LR 0.001000 Time 0.022117 +2023-10-05 20:47:47,110 - Epoch: [12][ 380/ 1236] Overall Loss 0.404294 Objective Loss 0.404294 LR 0.001000 Time 0.022067 +2023-10-05 20:47:47,311 - Epoch: [12][ 390/ 1236] Overall Loss 0.404073 Objective Loss 0.404073 LR 0.001000 Time 0.022015 +2023-10-05 20:47:47,514 - Epoch: [12][ 400/ 1236] Overall Loss 0.403777 Objective Loss 0.403777 LR 0.001000 Time 0.021971 +2023-10-05 20:47:47,714 - Epoch: [12][ 410/ 1236] Overall Loss 0.403271 Objective Loss 0.403271 LR 0.001000 Time 0.021923 +2023-10-05 20:47:47,917 - Epoch: [12][ 420/ 1236] Overall Loss 0.403034 Objective Loss 0.403034 LR 0.001000 Time 0.021883 +2023-10-05 20:47:48,118 - Epoch: [12][ 430/ 1236] Overall Loss 0.402636 Objective Loss 0.402636 LR 0.001000 Time 0.021840 +2023-10-05 20:47:48,320 - Epoch: [12][ 440/ 1236] Overall Loss 0.403216 Objective Loss 0.403216 LR 0.001000 Time 0.021803 +2023-10-05 20:47:48,521 - Epoch: [12][ 450/ 1236] Overall Loss 0.403530 Objective Loss 0.403530 LR 0.001000 Time 0.021764 +2023-10-05 20:47:48,724 - Epoch: [12][ 460/ 1236] Overall Loss 0.403653 Objective Loss 0.403653 LR 0.001000 Time 0.021731 +2023-10-05 20:47:48,924 - Epoch: [12][ 470/ 1236] Overall Loss 0.403321 Objective Loss 0.403321 LR 0.001000 Time 0.021694 +2023-10-05 20:47:49,127 - Epoch: [12][ 480/ 1236] Overall Loss 0.403085 Objective Loss 0.403085 LR 0.001000 Time 0.021663 +2023-10-05 20:47:49,327 - Epoch: [12][ 490/ 1236] Overall Loss 0.402300 Objective Loss 0.402300 LR 0.001000 Time 0.021630 +2023-10-05 20:47:49,530 - Epoch: [12][ 500/ 1236] Overall Loss 0.402548 Objective Loss 0.402548 LR 0.001000 Time 0.021602 +2023-10-05 20:47:49,731 - Epoch: [12][ 510/ 1236] Overall Loss 0.402162 Objective Loss 0.402162 LR 0.001000 Time 0.021571 +2023-10-05 20:47:49,933 - Epoch: [12][ 520/ 1236] Overall Loss 0.402107 Objective Loss 0.402107 LR 0.001000 Time 0.021545 +2023-10-05 20:47:50,134 - Epoch: [12][ 530/ 1236] Overall Loss 0.403125 Objective Loss 0.403125 LR 0.001000 Time 0.021517 +2023-10-05 20:47:50,337 - Epoch: [12][ 540/ 1236] Overall Loss 0.402636 Objective Loss 0.402636 LR 0.001000 Time 0.021493 +2023-10-05 20:47:50,537 - Epoch: [12][ 550/ 1236] Overall Loss 0.402507 Objective Loss 0.402507 LR 0.001000 Time 0.021466 +2023-10-05 20:47:50,740 - Epoch: [12][ 560/ 1236] Overall Loss 0.402677 Objective Loss 0.402677 LR 0.001000 Time 0.021444 +2023-10-05 20:47:50,941 - Epoch: [12][ 570/ 1236] Overall Loss 0.403326 Objective Loss 0.403326 LR 0.001000 Time 0.021419 +2023-10-05 20:47:51,143 - Epoch: [12][ 580/ 1236] Overall Loss 0.403722 Objective Loss 0.403722 LR 0.001000 Time 0.021399 +2023-10-05 20:47:51,344 - Epoch: [12][ 590/ 1236] Overall Loss 0.404047 Objective Loss 0.404047 LR 0.001000 Time 0.021376 +2023-10-05 20:47:51,546 - Epoch: [12][ 600/ 1236] Overall Loss 0.403922 Objective Loss 0.403922 LR 0.001000 Time 0.021356 +2023-10-05 20:47:51,747 - Epoch: [12][ 610/ 1236] Overall Loss 0.403417 Objective Loss 0.403417 LR 0.001000 Time 0.021335 +2023-10-05 20:47:51,950 - Epoch: [12][ 620/ 1236] Overall Loss 0.403276 Objective Loss 0.403276 LR 0.001000 Time 0.021317 +2023-10-05 20:47:52,150 - Epoch: [12][ 630/ 1236] Overall Loss 0.403401 Objective Loss 0.403401 LR 0.001000 Time 0.021296 +2023-10-05 20:47:52,353 - Epoch: [12][ 640/ 1236] Overall Loss 0.403853 Objective Loss 0.403853 LR 0.001000 Time 0.021280 +2023-10-05 20:47:52,554 - Epoch: [12][ 650/ 1236] Overall Loss 0.404038 Objective Loss 0.404038 LR 0.001000 Time 0.021261 +2023-10-05 20:47:52,757 - Epoch: [12][ 660/ 1236] Overall Loss 0.403572 Objective Loss 0.403572 LR 0.001000 Time 0.021246 +2023-10-05 20:47:52,957 - Epoch: [12][ 670/ 1236] Overall Loss 0.403523 Objective Loss 0.403523 LR 0.001000 Time 0.021228 +2023-10-05 20:47:53,160 - Epoch: [12][ 680/ 1236] Overall Loss 0.403433 Objective Loss 0.403433 LR 0.001000 Time 0.021213 +2023-10-05 20:47:53,361 - Epoch: [12][ 690/ 1236] Overall Loss 0.403310 Objective Loss 0.403310 LR 0.001000 Time 0.021196 +2023-10-05 20:47:53,564 - Epoch: [12][ 700/ 1236] Overall Loss 0.403516 Objective Loss 0.403516 LR 0.001000 Time 0.021182 +2023-10-05 20:47:53,765 - Epoch: [12][ 710/ 1236] Overall Loss 0.403719 Objective Loss 0.403719 LR 0.001000 Time 0.021166 +2023-10-05 20:47:53,967 - Epoch: [12][ 720/ 1236] Overall Loss 0.404203 Objective Loss 0.404203 LR 0.001000 Time 0.021153 +2023-10-05 20:47:54,168 - Epoch: [12][ 730/ 1236] Overall Loss 0.404852 Objective Loss 0.404852 LR 0.001000 Time 0.021138 +2023-10-05 20:47:54,371 - Epoch: [12][ 740/ 1236] Overall Loss 0.405097 Objective Loss 0.405097 LR 0.001000 Time 0.021126 +2023-10-05 20:47:54,571 - Epoch: [12][ 750/ 1236] Overall Loss 0.405347 Objective Loss 0.405347 LR 0.001000 Time 0.021111 +2023-10-05 20:47:54,774 - Epoch: [12][ 760/ 1236] Overall Loss 0.405604 Objective Loss 0.405604 LR 0.001000 Time 0.021100 +2023-10-05 20:47:54,975 - Epoch: [12][ 770/ 1236] Overall Loss 0.405442 Objective Loss 0.405442 LR 0.001000 Time 0.021086 +2023-10-05 20:47:55,178 - Epoch: [12][ 780/ 1236] Overall Loss 0.405227 Objective Loss 0.405227 LR 0.001000 Time 0.021075 +2023-10-05 20:47:55,378 - Epoch: [12][ 790/ 1236] Overall Loss 0.405178 Objective Loss 0.405178 LR 0.001000 Time 0.021062 +2023-10-05 20:47:55,581 - Epoch: [12][ 800/ 1236] Overall Loss 0.404703 Objective Loss 0.404703 LR 0.001000 Time 0.021052 +2023-10-05 20:47:55,782 - Epoch: [12][ 810/ 1236] Overall Loss 0.404883 Objective Loss 0.404883 LR 0.001000 Time 0.021039 +2023-10-05 20:47:55,984 - Epoch: [12][ 820/ 1236] Overall Loss 0.405093 Objective Loss 0.405093 LR 0.001000 Time 0.021029 +2023-10-05 20:47:56,185 - Epoch: [12][ 830/ 1236] Overall Loss 0.405479 Objective Loss 0.405479 LR 0.001000 Time 0.021017 +2023-10-05 20:47:56,388 - Epoch: [12][ 840/ 1236] Overall Loss 0.405680 Objective Loss 0.405680 LR 0.001000 Time 0.021008 +2023-10-05 20:47:56,588 - Epoch: [12][ 850/ 1236] Overall Loss 0.405802 Objective Loss 0.405802 LR 0.001000 Time 0.020996 +2023-10-05 20:47:56,791 - Epoch: [12][ 860/ 1236] Overall Loss 0.405099 Objective Loss 0.405099 LR 0.001000 Time 0.020987 +2023-10-05 20:47:56,992 - Epoch: [12][ 870/ 1236] Overall Loss 0.405063 Objective Loss 0.405063 LR 0.001000 Time 0.020976 +2023-10-05 20:47:57,195 - Epoch: [12][ 880/ 1236] Overall Loss 0.405291 Objective Loss 0.405291 LR 0.001000 Time 0.020968 +2023-10-05 20:47:57,396 - Epoch: [12][ 890/ 1236] Overall Loss 0.407296 Objective Loss 0.407296 LR 0.001000 Time 0.020958 +2023-10-05 20:47:57,599 - Epoch: [12][ 900/ 1236] Overall Loss 0.409659 Objective Loss 0.409659 LR 0.001000 Time 0.020951 +2023-10-05 20:47:57,800 - Epoch: [12][ 910/ 1236] Overall Loss 0.411379 Objective Loss 0.411379 LR 0.001000 Time 0.020941 +2023-10-05 20:47:58,003 - Epoch: [12][ 920/ 1236] Overall Loss 0.412993 Objective Loss 0.412993 LR 0.001000 Time 0.020934 +2023-10-05 20:47:58,205 - Epoch: [12][ 930/ 1236] Overall Loss 0.414399 Objective Loss 0.414399 LR 0.001000 Time 0.020925 +2023-10-05 20:47:58,408 - Epoch: [12][ 940/ 1236] Overall Loss 0.415348 Objective Loss 0.415348 LR 0.001000 Time 0.020918 +2023-10-05 20:47:58,609 - Epoch: [12][ 950/ 1236] Overall Loss 0.416485 Objective Loss 0.416485 LR 0.001000 Time 0.020909 +2023-10-05 20:47:58,813 - Epoch: [12][ 960/ 1236] Overall Loss 0.417624 Objective Loss 0.417624 LR 0.001000 Time 0.020903 +2023-10-05 20:47:59,014 - Epoch: [12][ 970/ 1236] Overall Loss 0.418339 Objective Loss 0.418339 LR 0.001000 Time 0.020895 +2023-10-05 20:47:59,217 - Epoch: [12][ 980/ 1236] Overall Loss 0.419114 Objective Loss 0.419114 LR 0.001000 Time 0.020888 +2023-10-05 20:47:59,418 - Epoch: [12][ 990/ 1236] Overall Loss 0.419698 Objective Loss 0.419698 LR 0.001000 Time 0.020880 +2023-10-05 20:47:59,622 - Epoch: [12][ 1000/ 1236] Overall Loss 0.420481 Objective Loss 0.420481 LR 0.001000 Time 0.020874 +2023-10-05 20:47:59,823 - Epoch: [12][ 1010/ 1236] Overall Loss 0.421226 Objective Loss 0.421226 LR 0.001000 Time 0.020866 +2023-10-05 20:48:00,026 - Epoch: [12][ 1020/ 1236] Overall Loss 0.421933 Objective Loss 0.421933 LR 0.001000 Time 0.020861 +2023-10-05 20:48:00,228 - Epoch: [12][ 1030/ 1236] Overall Loss 0.422541 Objective Loss 0.422541 LR 0.001000 Time 0.020854 +2023-10-05 20:48:00,431 - Epoch: [12][ 1040/ 1236] Overall Loss 0.422900 Objective Loss 0.422900 LR 0.001000 Time 0.020848 +2023-10-05 20:48:00,632 - Epoch: [12][ 1050/ 1236] Overall Loss 0.423175 Objective Loss 0.423175 LR 0.001000 Time 0.020841 +2023-10-05 20:48:00,835 - Epoch: [12][ 1060/ 1236] Overall Loss 0.423384 Objective Loss 0.423384 LR 0.001000 Time 0.020835 +2023-10-05 20:48:01,036 - Epoch: [12][ 1070/ 1236] Overall Loss 0.423790 Objective Loss 0.423790 LR 0.001000 Time 0.020828 +2023-10-05 20:48:01,239 - Epoch: [12][ 1080/ 1236] Overall Loss 0.424059 Objective Loss 0.424059 LR 0.001000 Time 0.020823 +2023-10-05 20:48:01,441 - Epoch: [12][ 1090/ 1236] Overall Loss 0.424205 Objective Loss 0.424205 LR 0.001000 Time 0.020817 +2023-10-05 20:48:01,644 - Epoch: [12][ 1100/ 1236] Overall Loss 0.424458 Objective Loss 0.424458 LR 0.001000 Time 0.020812 +2023-10-05 20:48:01,845 - Epoch: [12][ 1110/ 1236] Overall Loss 0.424851 Objective Loss 0.424851 LR 0.001000 Time 0.020806 +2023-10-05 20:48:02,049 - Epoch: [12][ 1120/ 1236] Overall Loss 0.425035 Objective Loss 0.425035 LR 0.001000 Time 0.020801 +2023-10-05 20:48:02,250 - Epoch: [12][ 1130/ 1236] Overall Loss 0.425213 Objective Loss 0.425213 LR 0.001000 Time 0.020795 +2023-10-05 20:48:02,454 - Epoch: [12][ 1140/ 1236] Overall Loss 0.425442 Objective Loss 0.425442 LR 0.001000 Time 0.020790 +2023-10-05 20:48:02,655 - Epoch: [12][ 1150/ 1236] Overall Loss 0.425667 Objective Loss 0.425667 LR 0.001000 Time 0.020784 +2023-10-05 20:48:02,858 - Epoch: [12][ 1160/ 1236] Overall Loss 0.425799 Objective Loss 0.425799 LR 0.001000 Time 0.020780 +2023-10-05 20:48:03,059 - Epoch: [12][ 1170/ 1236] Overall Loss 0.425984 Objective Loss 0.425984 LR 0.001000 Time 0.020774 +2023-10-05 20:48:03,263 - Epoch: [12][ 1180/ 1236] Overall Loss 0.426453 Objective Loss 0.426453 LR 0.001000 Time 0.020770 +2023-10-05 20:48:03,465 - Epoch: [12][ 1190/ 1236] Overall Loss 0.426677 Objective Loss 0.426677 LR 0.001000 Time 0.020765 +2023-10-05 20:48:03,668 - Epoch: [12][ 1200/ 1236] Overall Loss 0.426604 Objective Loss 0.426604 LR 0.001000 Time 0.020761 +2023-10-05 20:48:03,869 - Epoch: [12][ 1210/ 1236] Overall Loss 0.426752 Objective Loss 0.426752 LR 0.001000 Time 0.020755 +2023-10-05 20:48:04,073 - Epoch: [12][ 1220/ 1236] Overall Loss 0.427133 Objective Loss 0.427133 LR 0.001000 Time 0.020752 +2023-10-05 20:48:04,327 - Epoch: [12][ 1230/ 1236] Overall Loss 0.427393 Objective Loss 0.427393 LR 0.001000 Time 0.020790 +2023-10-05 20:48:04,445 - Epoch: [12][ 1236/ 1236] Overall Loss 0.427542 Objective Loss 0.427542 Top1 82.892057 Top5 97.148676 LR 0.001000 Time 0.020784 +2023-10-05 20:48:04,600 - --- validate (epoch=12)----------- +2023-10-05 20:48:04,600 - 29943 samples (256 per mini-batch) +2023-10-05 20:48:05,047 - Epoch: [12][ 10/ 117] Loss 0.456102 Top1 76.835938 Top5 97.265625 +2023-10-05 20:48:05,194 - Epoch: [12][ 20/ 117] Loss 0.461306 Top1 77.031250 Top5 97.128906 +2023-10-05 20:48:05,340 - Epoch: [12][ 30/ 117] Loss 0.451764 Top1 77.109375 Top5 96.796875 +2023-10-05 20:48:05,485 - Epoch: [12][ 40/ 117] Loss 0.450218 Top1 77.099609 Top5 96.728516 +2023-10-05 20:48:05,631 - Epoch: [12][ 50/ 117] Loss 0.446325 Top1 77.007812 Top5 96.773438 +2023-10-05 20:48:05,777 - Epoch: [12][ 60/ 117] Loss 0.445298 Top1 77.233073 Top5 96.875000 +2023-10-05 20:48:05,924 - Epoch: [12][ 70/ 117] Loss 0.450150 Top1 77.142857 Top5 96.813616 +2023-10-05 20:48:06,072 - Epoch: [12][ 80/ 117] Loss 0.446916 Top1 77.260742 Top5 96.865234 +2023-10-05 20:48:06,219 - Epoch: [12][ 90/ 117] Loss 0.447435 Top1 77.191840 Top5 96.909722 +2023-10-05 20:48:06,366 - Epoch: [12][ 100/ 117] Loss 0.446026 Top1 77.148438 Top5 96.882812 +2023-10-05 20:48:06,519 - Epoch: [12][ 110/ 117] Loss 0.446856 Top1 77.159091 Top5 96.899858 +2023-10-05 20:48:06,604 - Epoch: [12][ 117/ 117] Loss 0.446967 Top1 77.086464 Top5 96.914137 +2023-10-05 20:48:06,725 - ==> Top1: 77.086 Top5: 96.914 Loss: 0.447 + +2023-10-05 20:48:06,725 - ==> Confusion: +[[ 930 2 5 2 9 4 0 0 9 62 2 0 0 3 4 1 7 0 0 1 9] + [ 4 988 3 0 4 37 1 41 2 0 1 1 0 1 1 3 15 2 19 4 4] + [ 8 0 933 15 3 0 33 13 0 0 1 5 3 4 2 3 4 1 12 5 11] + [ 3 2 25 934 2 5 2 0 8 0 5 0 1 7 28 6 2 12 29 1 17] + [ 38 13 1 0 942 4 1 3 2 5 0 4 1 2 4 5 16 3 0 0 6] + [ 8 35 2 0 3 943 1 34 3 4 0 15 3 29 5 1 9 0 6 6 9] + [ 1 9 37 1 0 2 1096 14 0 0 1 4 0 0 0 3 1 1 3 11 7] + [ 6 10 21 0 0 24 7 1060 2 1 3 7 1 0 1 0 0 4 53 14 4] + [ 23 3 0 0 0 0 1 0 956 45 7 2 0 19 8 1 3 5 15 1 0] + [ 151 0 1 0 3 1 0 2 28 877 0 2 0 32 1 1 2 3 2 2 11] + [ 5 6 13 13 0 0 5 6 23 3 915 3 0 16 7 0 3 0 17 4 14] + [ 3 0 1 0 0 16 0 3 1 0 0 938 17 4 0 4 2 19 1 23 3] + [ 0 1 1 5 0 1 0 2 0 0 2 60 900 0 3 9 6 52 3 15 8] + [ 5 0 1 0 4 11 0 1 17 17 4 11 1 1022 3 3 0 1 0 5 13] + [ 21 1 2 8 5 0 0 1 30 7 0 0 2 2 968 0 5 5 29 0 15] + [ 2 2 6 2 3 1 1 1 0 1 0 13 3 2 0 1033 20 22 0 11 11] + [ 3 16 1 0 8 5 2 1 4 0 0 7 0 2 2 9 1084 1 0 4 12] + [ 2 0 0 3 1 0 1 0 0 0 0 7 7 2 1 5 1 1002 1 3 2] + [ 0 2 5 12 2 0 0 45 3 1 0 0 3 0 4 0 2 0 975 6 8] + [ 1 4 2 0 1 5 9 18 0 0 1 22 1 0 0 2 6 3 4 1068 5] + [ 286 278 267 80 139 224 63 214 140 91 143 210 410 361 186 75 325 117 340 438 3518]] + +2023-10-05 20:48:06,727 - ==> Best [Top1: 78.245 Top5: 96.871 Sparsity:0.00 Params: 148928 on epoch: 11] +2023-10-05 20:48:06,727 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:48:06,733 - + +2023-10-05 20:48:06,733 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:48:07,830 - Epoch: [13][ 10/ 1236] Overall Loss 0.418644 Objective Loss 0.418644 LR 0.001000 Time 0.109651 +2023-10-05 20:48:08,031 - Epoch: [13][ 20/ 1236] Overall Loss 0.426998 Objective Loss 0.426998 LR 0.001000 Time 0.064860 +2023-10-05 20:48:08,231 - Epoch: [13][ 30/ 1236] Overall Loss 0.421851 Objective Loss 0.421851 LR 0.001000 Time 0.049886 +2023-10-05 20:48:08,433 - Epoch: [13][ 40/ 1236] Overall Loss 0.426973 Objective Loss 0.426973 LR 0.001000 Time 0.042464 +2023-10-05 20:48:08,632 - Epoch: [13][ 50/ 1236] Overall Loss 0.425932 Objective Loss 0.425932 LR 0.001000 Time 0.037946 +2023-10-05 20:48:08,834 - Epoch: [13][ 60/ 1236] Overall Loss 0.422317 Objective Loss 0.422317 LR 0.001000 Time 0.034985 +2023-10-05 20:48:09,034 - Epoch: [13][ 70/ 1236] Overall Loss 0.424947 Objective Loss 0.424947 LR 0.001000 Time 0.032832 +2023-10-05 20:48:09,236 - Epoch: [13][ 80/ 1236] Overall Loss 0.419252 Objective Loss 0.419252 LR 0.001000 Time 0.031250 +2023-10-05 20:48:09,436 - Epoch: [13][ 90/ 1236] Overall Loss 0.424356 Objective Loss 0.424356 LR 0.001000 Time 0.030001 +2023-10-05 20:48:09,644 - Epoch: [13][ 100/ 1236] Overall Loss 0.425826 Objective Loss 0.425826 LR 0.001000 Time 0.029075 +2023-10-05 20:48:09,848 - Epoch: [13][ 110/ 1236] Overall Loss 0.425914 Objective Loss 0.425914 LR 0.001000 Time 0.028284 +2023-10-05 20:48:10,052 - Epoch: [13][ 120/ 1236] Overall Loss 0.426174 Objective Loss 0.426174 LR 0.001000 Time 0.027628 +2023-10-05 20:48:10,256 - Epoch: [13][ 130/ 1236] Overall Loss 0.427008 Objective Loss 0.427008 LR 0.001000 Time 0.027063 +2023-10-05 20:48:10,461 - Epoch: [13][ 140/ 1236] Overall Loss 0.426987 Objective Loss 0.426987 LR 0.001000 Time 0.026592 +2023-10-05 20:48:10,665 - Epoch: [13][ 150/ 1236] Overall Loss 0.427540 Objective Loss 0.427540 LR 0.001000 Time 0.026181 +2023-10-05 20:48:10,866 - Epoch: [13][ 160/ 1236] Overall Loss 0.427418 Objective Loss 0.427418 LR 0.001000 Time 0.025798 +2023-10-05 20:48:11,068 - Epoch: [13][ 170/ 1236] Overall Loss 0.429597 Objective Loss 0.429597 LR 0.001000 Time 0.025468 +2023-10-05 20:48:11,269 - Epoch: [13][ 180/ 1236] Overall Loss 0.431082 Objective Loss 0.431082 LR 0.001000 Time 0.025166 +2023-10-05 20:48:11,471 - Epoch: [13][ 190/ 1236] Overall Loss 0.433028 Objective Loss 0.433028 LR 0.001000 Time 0.024902 +2023-10-05 20:48:11,671 - Epoch: [13][ 200/ 1236] Overall Loss 0.434095 Objective Loss 0.434095 LR 0.001000 Time 0.024658 +2023-10-05 20:48:11,874 - Epoch: [13][ 210/ 1236] Overall Loss 0.434367 Objective Loss 0.434367 LR 0.001000 Time 0.024446 +2023-10-05 20:48:12,075 - Epoch: [13][ 220/ 1236] Overall Loss 0.435224 Objective Loss 0.435224 LR 0.001000 Time 0.024247 +2023-10-05 20:48:12,276 - Epoch: [13][ 230/ 1236] Overall Loss 0.436243 Objective Loss 0.436243 LR 0.001000 Time 0.024067 +2023-10-05 20:48:12,478 - Epoch: [13][ 240/ 1236] Overall Loss 0.437185 Objective Loss 0.437185 LR 0.001000 Time 0.023903 +2023-10-05 20:48:12,680 - Epoch: [13][ 250/ 1236] Overall Loss 0.437277 Objective Loss 0.437277 LR 0.001000 Time 0.023752 +2023-10-05 20:48:12,881 - Epoch: [13][ 260/ 1236] Overall Loss 0.437034 Objective Loss 0.437034 LR 0.001000 Time 0.023612 +2023-10-05 20:48:13,083 - Epoch: [13][ 270/ 1236] Overall Loss 0.438361 Objective Loss 0.438361 LR 0.001000 Time 0.023483 +2023-10-05 20:48:13,284 - Epoch: [13][ 280/ 1236] Overall Loss 0.438451 Objective Loss 0.438451 LR 0.001000 Time 0.023362 +2023-10-05 20:48:13,486 - Epoch: [13][ 290/ 1236] Overall Loss 0.438521 Objective Loss 0.438521 LR 0.001000 Time 0.023251 +2023-10-05 20:48:13,687 - Epoch: [13][ 300/ 1236] Overall Loss 0.438296 Objective Loss 0.438296 LR 0.001000 Time 0.023146 +2023-10-05 20:48:13,889 - Epoch: [13][ 310/ 1236] Overall Loss 0.437339 Objective Loss 0.437339 LR 0.001000 Time 0.023049 +2023-10-05 20:48:14,090 - Epoch: [13][ 320/ 1236] Overall Loss 0.437639 Objective Loss 0.437639 LR 0.001000 Time 0.022957 +2023-10-05 20:48:14,297 - Epoch: [13][ 330/ 1236] Overall Loss 0.436708 Objective Loss 0.436708 LR 0.001000 Time 0.022888 +2023-10-05 20:48:14,510 - Epoch: [13][ 340/ 1236] Overall Loss 0.437236 Objective Loss 0.437236 LR 0.001000 Time 0.022841 +2023-10-05 20:48:14,727 - Epoch: [13][ 350/ 1236] Overall Loss 0.436821 Objective Loss 0.436821 LR 0.001000 Time 0.022806 +2023-10-05 20:48:14,940 - Epoch: [13][ 360/ 1236] Overall Loss 0.436030 Objective Loss 0.436030 LR 0.001000 Time 0.022762 +2023-10-05 20:48:15,154 - Epoch: [13][ 370/ 1236] Overall Loss 0.435889 Objective Loss 0.435889 LR 0.001000 Time 0.022726 +2023-10-05 20:48:15,364 - Epoch: [13][ 380/ 1236] Overall Loss 0.435325 Objective Loss 0.435325 LR 0.001000 Time 0.022679 +2023-10-05 20:48:15,578 - Epoch: [13][ 390/ 1236] Overall Loss 0.434173 Objective Loss 0.434173 LR 0.001000 Time 0.022646 +2023-10-05 20:48:15,788 - Epoch: [13][ 400/ 1236] Overall Loss 0.433509 Objective Loss 0.433509 LR 0.001000 Time 0.022604 +2023-10-05 20:48:16,002 - Epoch: [13][ 410/ 1236] Overall Loss 0.433578 Objective Loss 0.433578 LR 0.001000 Time 0.022574 +2023-10-05 20:48:16,212 - Epoch: [13][ 420/ 1236] Overall Loss 0.432894 Objective Loss 0.432894 LR 0.001000 Time 0.022536 +2023-10-05 20:48:16,427 - Epoch: [13][ 430/ 1236] Overall Loss 0.432392 Objective Loss 0.432392 LR 0.001000 Time 0.022511 +2023-10-05 20:48:16,637 - Epoch: [13][ 440/ 1236] Overall Loss 0.432006 Objective Loss 0.432006 LR 0.001000 Time 0.022475 +2023-10-05 20:48:16,848 - Epoch: [13][ 450/ 1236] Overall Loss 0.432307 Objective Loss 0.432307 LR 0.001000 Time 0.022444 +2023-10-05 20:48:17,051 - Epoch: [13][ 460/ 1236] Overall Loss 0.432388 Objective Loss 0.432388 LR 0.001000 Time 0.022397 +2023-10-05 20:48:17,256 - Epoch: [13][ 470/ 1236] Overall Loss 0.432625 Objective Loss 0.432625 LR 0.001000 Time 0.022356 +2023-10-05 20:48:17,460 - Epoch: [13][ 480/ 1236] Overall Loss 0.431874 Objective Loss 0.431874 LR 0.001000 Time 0.022313 +2023-10-05 20:48:17,664 - Epoch: [13][ 490/ 1236] Overall Loss 0.431835 Objective Loss 0.431835 LR 0.001000 Time 0.022275 +2023-10-05 20:48:17,868 - Epoch: [13][ 500/ 1236] Overall Loss 0.431721 Objective Loss 0.431721 LR 0.001000 Time 0.022236 +2023-10-05 20:48:18,073 - Epoch: [13][ 510/ 1236] Overall Loss 0.431530 Objective Loss 0.431530 LR 0.001000 Time 0.022201 +2023-10-05 20:48:18,276 - Epoch: [13][ 520/ 1236] Overall Loss 0.432166 Objective Loss 0.432166 LR 0.001000 Time 0.022164 +2023-10-05 20:48:18,481 - Epoch: [13][ 530/ 1236] Overall Loss 0.432695 Objective Loss 0.432695 LR 0.001000 Time 0.022132 +2023-10-05 20:48:18,684 - Epoch: [13][ 540/ 1236] Overall Loss 0.432862 Objective Loss 0.432862 LR 0.001000 Time 0.022097 +2023-10-05 20:48:18,889 - Epoch: [13][ 550/ 1236] Overall Loss 0.432814 Objective Loss 0.432814 LR 0.001000 Time 0.022068 +2023-10-05 20:48:19,093 - Epoch: [13][ 560/ 1236] Overall Loss 0.432688 Objective Loss 0.432688 LR 0.001000 Time 0.022036 +2023-10-05 20:48:19,298 - Epoch: [13][ 570/ 1236] Overall Loss 0.432150 Objective Loss 0.432150 LR 0.001000 Time 0.022009 +2023-10-05 20:48:19,501 - Epoch: [13][ 580/ 1236] Overall Loss 0.432314 Objective Loss 0.432314 LR 0.001000 Time 0.021979 +2023-10-05 20:48:19,706 - Epoch: [13][ 590/ 1236] Overall Loss 0.432428 Objective Loss 0.432428 LR 0.001000 Time 0.021954 +2023-10-05 20:48:19,909 - Epoch: [13][ 600/ 1236] Overall Loss 0.432267 Objective Loss 0.432267 LR 0.001000 Time 0.021926 +2023-10-05 20:48:20,114 - Epoch: [13][ 610/ 1236] Overall Loss 0.432524 Objective Loss 0.432524 LR 0.001000 Time 0.021903 +2023-10-05 20:48:20,318 - Epoch: [13][ 620/ 1236] Overall Loss 0.432651 Objective Loss 0.432651 LR 0.001000 Time 0.021877 +2023-10-05 20:48:20,523 - Epoch: [13][ 630/ 1236] Overall Loss 0.432772 Objective Loss 0.432772 LR 0.001000 Time 0.021855 +2023-10-05 20:48:20,726 - Epoch: [13][ 640/ 1236] Overall Loss 0.433245 Objective Loss 0.433245 LR 0.001000 Time 0.021830 +2023-10-05 20:48:20,931 - Epoch: [13][ 650/ 1236] Overall Loss 0.432852 Objective Loss 0.432852 LR 0.001000 Time 0.021809 +2023-10-05 20:48:21,135 - Epoch: [13][ 660/ 1236] Overall Loss 0.432848 Objective Loss 0.432848 LR 0.001000 Time 0.021787 +2023-10-05 20:48:21,340 - Epoch: [13][ 670/ 1236] Overall Loss 0.432884 Objective Loss 0.432884 LR 0.001000 Time 0.021767 +2023-10-05 20:48:21,543 - Epoch: [13][ 680/ 1236] Overall Loss 0.432998 Objective Loss 0.432998 LR 0.001000 Time 0.021746 +2023-10-05 20:48:21,748 - Epoch: [13][ 690/ 1236] Overall Loss 0.433152 Objective Loss 0.433152 LR 0.001000 Time 0.021727 +2023-10-05 20:48:21,952 - Epoch: [13][ 700/ 1236] Overall Loss 0.432266 Objective Loss 0.432266 LR 0.001000 Time 0.021707 +2023-10-05 20:48:22,157 - Epoch: [13][ 710/ 1236] Overall Loss 0.432680 Objective Loss 0.432680 LR 0.001000 Time 0.021689 +2023-10-05 20:48:22,360 - Epoch: [13][ 720/ 1236] Overall Loss 0.432434 Objective Loss 0.432434 LR 0.001000 Time 0.021670 +2023-10-05 20:48:22,565 - Epoch: [13][ 730/ 1236] Overall Loss 0.432373 Objective Loss 0.432373 LR 0.001000 Time 0.021654 +2023-10-05 20:48:22,769 - Epoch: [13][ 740/ 1236] Overall Loss 0.432233 Objective Loss 0.432233 LR 0.001000 Time 0.021635 +2023-10-05 20:48:22,974 - Epoch: [13][ 750/ 1236] Overall Loss 0.432158 Objective Loss 0.432158 LR 0.001000 Time 0.021620 +2023-10-05 20:48:23,177 - Epoch: [13][ 760/ 1236] Overall Loss 0.431936 Objective Loss 0.431936 LR 0.001000 Time 0.021603 +2023-10-05 20:48:23,382 - Epoch: [13][ 770/ 1236] Overall Loss 0.431457 Objective Loss 0.431457 LR 0.001000 Time 0.021588 +2023-10-05 20:48:23,586 - Epoch: [13][ 780/ 1236] Overall Loss 0.431795 Objective Loss 0.431795 LR 0.001000 Time 0.021572 +2023-10-05 20:48:23,791 - Epoch: [13][ 790/ 1236] Overall Loss 0.431760 Objective Loss 0.431760 LR 0.001000 Time 0.021558 +2023-10-05 20:48:23,995 - Epoch: [13][ 800/ 1236] Overall Loss 0.431829 Objective Loss 0.431829 LR 0.001000 Time 0.021543 +2023-10-05 20:48:24,200 - Epoch: [13][ 810/ 1236] Overall Loss 0.431873 Objective Loss 0.431873 LR 0.001000 Time 0.021529 +2023-10-05 20:48:24,403 - Epoch: [13][ 820/ 1236] Overall Loss 0.432318 Objective Loss 0.432318 LR 0.001000 Time 0.021515 +2023-10-05 20:48:24,608 - Epoch: [13][ 830/ 1236] Overall Loss 0.432633 Objective Loss 0.432633 LR 0.001000 Time 0.021502 +2023-10-05 20:48:24,812 - Epoch: [13][ 840/ 1236] Overall Loss 0.432752 Objective Loss 0.432752 LR 0.001000 Time 0.021488 +2023-10-05 20:48:25,017 - Epoch: [13][ 850/ 1236] Overall Loss 0.432571 Objective Loss 0.432571 LR 0.001000 Time 0.021476 +2023-10-05 20:48:25,221 - Epoch: [13][ 860/ 1236] Overall Loss 0.432538 Objective Loss 0.432538 LR 0.001000 Time 0.021463 +2023-10-05 20:48:25,426 - Epoch: [13][ 870/ 1236] Overall Loss 0.432527 Objective Loss 0.432527 LR 0.001000 Time 0.021451 +2023-10-05 20:48:25,629 - Epoch: [13][ 880/ 1236] Overall Loss 0.432328 Objective Loss 0.432328 LR 0.001000 Time 0.021438 +2023-10-05 20:48:25,834 - Epoch: [13][ 890/ 1236] Overall Loss 0.432335 Objective Loss 0.432335 LR 0.001000 Time 0.021427 +2023-10-05 20:48:26,037 - Epoch: [13][ 900/ 1236] Overall Loss 0.431947 Objective Loss 0.431947 LR 0.001000 Time 0.021414 +2023-10-05 20:48:26,242 - Epoch: [13][ 910/ 1236] Overall Loss 0.431833 Objective Loss 0.431833 LR 0.001000 Time 0.021404 +2023-10-05 20:48:26,446 - Epoch: [13][ 920/ 1236] Overall Loss 0.431938 Objective Loss 0.431938 LR 0.001000 Time 0.021392 +2023-10-05 20:48:26,651 - Epoch: [13][ 930/ 1236] Overall Loss 0.431743 Objective Loss 0.431743 LR 0.001000 Time 0.021382 +2023-10-05 20:48:26,854 - Epoch: [13][ 940/ 1236] Overall Loss 0.431555 Objective Loss 0.431555 LR 0.001000 Time 0.021371 +2023-10-05 20:48:27,059 - Epoch: [13][ 950/ 1236] Overall Loss 0.431847 Objective Loss 0.431847 LR 0.001000 Time 0.021361 +2023-10-05 20:48:27,263 - Epoch: [13][ 960/ 1236] Overall Loss 0.431689 Objective Loss 0.431689 LR 0.001000 Time 0.021350 +2023-10-05 20:48:27,468 - Epoch: [13][ 970/ 1236] Overall Loss 0.431414 Objective Loss 0.431414 LR 0.001000 Time 0.021341 +2023-10-05 20:48:27,671 - Epoch: [13][ 980/ 1236] Overall Loss 0.431407 Objective Loss 0.431407 LR 0.001000 Time 0.021331 +2023-10-05 20:48:27,877 - Epoch: [13][ 990/ 1236] Overall Loss 0.431764 Objective Loss 0.431764 LR 0.001000 Time 0.021322 +2023-10-05 20:48:28,080 - Epoch: [13][ 1000/ 1236] Overall Loss 0.431614 Objective Loss 0.431614 LR 0.001000 Time 0.021312 +2023-10-05 20:48:28,285 - Epoch: [13][ 1010/ 1236] Overall Loss 0.431900 Objective Loss 0.431900 LR 0.001000 Time 0.021304 +2023-10-05 20:48:28,489 - Epoch: [13][ 1020/ 1236] Overall Loss 0.431865 Objective Loss 0.431865 LR 0.001000 Time 0.021295 +2023-10-05 20:48:28,694 - Epoch: [13][ 1030/ 1236] Overall Loss 0.431360 Objective Loss 0.431360 LR 0.001000 Time 0.021286 +2023-10-05 20:48:28,898 - Epoch: [13][ 1040/ 1236] Overall Loss 0.431306 Objective Loss 0.431306 LR 0.001000 Time 0.021277 +2023-10-05 20:48:29,103 - Epoch: [13][ 1050/ 1236] Overall Loss 0.431433 Objective Loss 0.431433 LR 0.001000 Time 0.021270 +2023-10-05 20:48:29,306 - Epoch: [13][ 1060/ 1236] Overall Loss 0.431387 Objective Loss 0.431387 LR 0.001000 Time 0.021261 +2023-10-05 20:48:29,511 - Epoch: [13][ 1070/ 1236] Overall Loss 0.431491 Objective Loss 0.431491 LR 0.001000 Time 0.021253 +2023-10-05 20:48:29,715 - Epoch: [13][ 1080/ 1236] Overall Loss 0.431644 Objective Loss 0.431644 LR 0.001000 Time 0.021245 +2023-10-05 20:48:29,920 - Epoch: [13][ 1090/ 1236] Overall Loss 0.431505 Objective Loss 0.431505 LR 0.001000 Time 0.021238 +2023-10-05 20:48:30,124 - Epoch: [13][ 1100/ 1236] Overall Loss 0.431645 Objective Loss 0.431645 LR 0.001000 Time 0.021229 +2023-10-05 20:48:30,329 - Epoch: [13][ 1110/ 1236] Overall Loss 0.431961 Objective Loss 0.431961 LR 0.001000 Time 0.021223 +2023-10-05 20:48:30,532 - Epoch: [13][ 1120/ 1236] Overall Loss 0.432221 Objective Loss 0.432221 LR 0.001000 Time 0.021215 +2023-10-05 20:48:30,738 - Epoch: [13][ 1130/ 1236] Overall Loss 0.432468 Objective Loss 0.432468 LR 0.001000 Time 0.021208 +2023-10-05 20:48:30,941 - Epoch: [13][ 1140/ 1236] Overall Loss 0.432545 Objective Loss 0.432545 LR 0.001000 Time 0.021200 +2023-10-05 20:48:31,146 - Epoch: [13][ 1150/ 1236] Overall Loss 0.432302 Objective Loss 0.432302 LR 0.001000 Time 0.021194 +2023-10-05 20:48:31,349 - Epoch: [13][ 1160/ 1236] Overall Loss 0.432611 Objective Loss 0.432611 LR 0.001000 Time 0.021186 +2023-10-05 20:48:31,553 - Epoch: [13][ 1170/ 1236] Overall Loss 0.432187 Objective Loss 0.432187 LR 0.001000 Time 0.021179 +2023-10-05 20:48:31,756 - Epoch: [13][ 1180/ 1236] Overall Loss 0.431692 Objective Loss 0.431692 LR 0.001000 Time 0.021171 +2023-10-05 20:48:31,961 - Epoch: [13][ 1190/ 1236] Overall Loss 0.431757 Objective Loss 0.431757 LR 0.001000 Time 0.021165 +2023-10-05 20:48:32,164 - Epoch: [13][ 1200/ 1236] Overall Loss 0.431638 Objective Loss 0.431638 LR 0.001000 Time 0.021157 +2023-10-05 20:48:32,368 - Epoch: [13][ 1210/ 1236] Overall Loss 0.431509 Objective Loss 0.431509 LR 0.001000 Time 0.021151 +2023-10-05 20:48:32,571 - Epoch: [13][ 1220/ 1236] Overall Loss 0.431638 Objective Loss 0.431638 LR 0.001000 Time 0.021144 +2023-10-05 20:48:32,830 - Epoch: [13][ 1230/ 1236] Overall Loss 0.431744 Objective Loss 0.431744 LR 0.001000 Time 0.021182 +2023-10-05 20:48:32,948 - Epoch: [13][ 1236/ 1236] Overall Loss 0.431623 Objective Loss 0.431623 Top1 79.633401 Top5 96.334012 LR 0.001000 Time 0.021175 +2023-10-05 20:48:33,091 - --- validate (epoch=13)----------- +2023-10-05 20:48:33,091 - 29943 samples (256 per mini-batch) +2023-10-05 20:48:33,547 - Epoch: [13][ 10/ 117] Loss 0.448211 Top1 77.421875 Top5 96.679688 +2023-10-05 20:48:33,698 - Epoch: [13][ 20/ 117] Loss 0.452814 Top1 77.480469 Top5 96.406250 +2023-10-05 20:48:33,849 - Epoch: [13][ 30/ 117] Loss 0.450983 Top1 77.382812 Top5 96.510417 +2023-10-05 20:48:33,999 - Epoch: [13][ 40/ 117] Loss 0.447280 Top1 77.490234 Top5 96.416016 +2023-10-05 20:48:34,148 - Epoch: [13][ 50/ 117] Loss 0.449445 Top1 77.367188 Top5 96.437500 +2023-10-05 20:48:34,300 - Epoch: [13][ 60/ 117] Loss 0.445588 Top1 77.532552 Top5 96.458333 +2023-10-05 20:48:34,454 - Epoch: [13][ 70/ 117] Loss 0.447578 Top1 77.421875 Top5 96.445312 +2023-10-05 20:48:34,610 - Epoch: [13][ 80/ 117] Loss 0.443785 Top1 77.519531 Top5 96.469727 +2023-10-05 20:48:34,765 - Epoch: [13][ 90/ 117] Loss 0.445265 Top1 77.317708 Top5 96.440972 +2023-10-05 20:48:34,921 - Epoch: [13][ 100/ 117] Loss 0.444749 Top1 77.316406 Top5 96.437500 +2023-10-05 20:48:35,084 - Epoch: [13][ 110/ 117] Loss 0.446325 Top1 77.201705 Top5 96.424006 +2023-10-05 20:48:35,170 - Epoch: [13][ 117/ 117] Loss 0.445220 Top1 77.256788 Top5 96.413185 +2023-10-05 20:48:35,303 - ==> Top1: 77.257 Top5: 96.413 Loss: 0.445 + +2023-10-05 20:48:35,304 - ==> Confusion: +[[ 918 0 5 1 16 1 1 0 14 47 2 3 0 1 12 1 7 1 2 1 17] + [ 0 1033 3 0 8 24 2 19 1 0 6 0 0 0 5 4 9 2 8 2 5] + [ 6 1 907 37 4 2 35 10 0 0 2 4 7 1 4 4 3 6 7 1 15] + [ 2 2 9 947 1 4 1 1 2 0 17 0 2 1 63 4 2 7 13 0 11] + [ 30 17 2 1 937 6 2 0 1 7 1 0 1 1 27 2 8 1 1 1 4] + [ 5 62 0 5 6 949 1 15 4 0 6 13 1 21 15 0 3 1 1 0 8] + [ 0 10 34 3 0 0 1095 6 0 0 3 3 1 0 0 9 1 2 4 10 10] + [ 8 37 18 1 1 45 6 977 2 1 7 7 7 1 1 1 0 2 79 6 11] + [ 28 8 0 0 0 1 1 0 935 25 20 3 0 11 48 1 1 4 3 0 0] + [ 161 1 1 1 13 5 3 1 53 822 1 1 0 28 10 1 3 3 2 1 8] + [ 1 5 7 9 1 3 2 3 15 0 970 0 0 5 12 0 2 2 9 0 7] + [ 1 2 2 0 0 26 0 1 0 0 1 891 56 5 2 6 2 22 1 12 5] + [ 1 1 7 10 0 4 1 3 0 0 0 54 909 1 6 9 6 42 2 6 6] + [ 4 1 0 5 5 21 0 1 23 12 13 4 8 989 6 2 4 1 1 6 13] + [ 17 2 1 7 1 0 0 0 7 2 3 0 1 1 1027 0 3 3 8 0 18] + [ 0 2 3 3 3 2 5 0 0 0 0 12 7 1 1 1026 37 20 1 5 6] + [ 1 24 1 1 11 3 2 0 2 0 0 4 0 2 5 9 1081 1 0 5 9] + [ 0 2 1 4 1 0 1 0 1 0 0 11 15 1 3 3 0 989 1 0 5] + [ 2 6 7 22 1 0 2 31 2 0 7 1 1 0 16 1 1 2 957 0 9] + [ 0 4 3 5 1 9 19 18 0 0 4 29 5 0 0 3 14 4 7 1010 17] + [ 223 354 162 119 128 215 57 85 108 59 230 164 448 326 395 82 317 117 262 290 3764]] + +2023-10-05 20:48:35,305 - ==> Best [Top1: 78.245 Top5: 96.871 Sparsity:0.00 Params: 148928 on epoch: 11] +2023-10-05 20:48:35,305 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:48:35,311 - + +2023-10-05 20:48:35,311 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:48:36,314 - Epoch: [14][ 10/ 1236] Overall Loss 0.455402 Objective Loss 0.455402 LR 0.001000 Time 0.100311 +2023-10-05 20:48:36,517 - Epoch: [14][ 20/ 1236] Overall Loss 0.451603 Objective Loss 0.451603 LR 0.001000 Time 0.060290 +2023-10-05 20:48:36,720 - Epoch: [14][ 30/ 1236] Overall Loss 0.439872 Objective Loss 0.439872 LR 0.001000 Time 0.046925 +2023-10-05 20:48:36,923 - Epoch: [14][ 40/ 1236] Overall Loss 0.435962 Objective Loss 0.435962 LR 0.001000 Time 0.040258 +2023-10-05 20:48:37,125 - Epoch: [14][ 50/ 1236] Overall Loss 0.439568 Objective Loss 0.439568 LR 0.001000 Time 0.036244 +2023-10-05 20:48:37,328 - Epoch: [14][ 60/ 1236] Overall Loss 0.441443 Objective Loss 0.441443 LR 0.001000 Time 0.033581 +2023-10-05 20:48:37,529 - Epoch: [14][ 70/ 1236] Overall Loss 0.437400 Objective Loss 0.437400 LR 0.001000 Time 0.031651 +2023-10-05 20:48:37,728 - Epoch: [14][ 80/ 1236] Overall Loss 0.434387 Objective Loss 0.434387 LR 0.001000 Time 0.030179 +2023-10-05 20:48:37,928 - Epoch: [14][ 90/ 1236] Overall Loss 0.433040 Objective Loss 0.433040 LR 0.001000 Time 0.029047 +2023-10-05 20:48:38,130 - Epoch: [14][ 100/ 1236] Overall Loss 0.437303 Objective Loss 0.437303 LR 0.001000 Time 0.028156 +2023-10-05 20:48:38,331 - Epoch: [14][ 110/ 1236] Overall Loss 0.443482 Objective Loss 0.443482 LR 0.001000 Time 0.027423 +2023-10-05 20:48:38,534 - Epoch: [14][ 120/ 1236] Overall Loss 0.441757 Objective Loss 0.441757 LR 0.001000 Time 0.026822 +2023-10-05 20:48:38,735 - Epoch: [14][ 130/ 1236] Overall Loss 0.439401 Objective Loss 0.439401 LR 0.001000 Time 0.026301 +2023-10-05 20:48:38,937 - Epoch: [14][ 140/ 1236] Overall Loss 0.440373 Objective Loss 0.440373 LR 0.001000 Time 0.025869 +2023-10-05 20:48:39,139 - Epoch: [14][ 150/ 1236] Overall Loss 0.440007 Objective Loss 0.440007 LR 0.001000 Time 0.025486 +2023-10-05 20:48:39,342 - Epoch: [14][ 160/ 1236] Overall Loss 0.438698 Objective Loss 0.438698 LR 0.001000 Time 0.025158 +2023-10-05 20:48:39,542 - Epoch: [14][ 170/ 1236] Overall Loss 0.440349 Objective Loss 0.440349 LR 0.001000 Time 0.024856 +2023-10-05 20:48:39,745 - Epoch: [14][ 180/ 1236] Overall Loss 0.440800 Objective Loss 0.440800 LR 0.001000 Time 0.024600 +2023-10-05 20:48:39,946 - Epoch: [14][ 190/ 1236] Overall Loss 0.440238 Objective Loss 0.440238 LR 0.001000 Time 0.024361 +2023-10-05 20:48:40,146 - Epoch: [14][ 200/ 1236] Overall Loss 0.441161 Objective Loss 0.441161 LR 0.001000 Time 0.024140 +2023-10-05 20:48:40,346 - Epoch: [14][ 210/ 1236] Overall Loss 0.443081 Objective Loss 0.443081 LR 0.001000 Time 0.023942 +2023-10-05 20:48:40,551 - Epoch: [14][ 220/ 1236] Overall Loss 0.444304 Objective Loss 0.444304 LR 0.001000 Time 0.023784 +2023-10-05 20:48:40,755 - Epoch: [14][ 230/ 1236] Overall Loss 0.443905 Objective Loss 0.443905 LR 0.001000 Time 0.023634 +2023-10-05 20:48:40,959 - Epoch: [14][ 240/ 1236] Overall Loss 0.443583 Objective Loss 0.443583 LR 0.001000 Time 0.023500 +2023-10-05 20:48:41,163 - Epoch: [14][ 250/ 1236] Overall Loss 0.443640 Objective Loss 0.443640 LR 0.001000 Time 0.023374 +2023-10-05 20:48:41,368 - Epoch: [14][ 260/ 1236] Overall Loss 0.443031 Objective Loss 0.443031 LR 0.001000 Time 0.023261 +2023-10-05 20:48:41,572 - Epoch: [14][ 270/ 1236] Overall Loss 0.442823 Objective Loss 0.442823 LR 0.001000 Time 0.023153 +2023-10-05 20:48:41,776 - Epoch: [14][ 280/ 1236] Overall Loss 0.441918 Objective Loss 0.441918 LR 0.001000 Time 0.023055 +2023-10-05 20:48:41,980 - Epoch: [14][ 290/ 1236] Overall Loss 0.440574 Objective Loss 0.440574 LR 0.001000 Time 0.022962 +2023-10-05 20:48:42,184 - Epoch: [14][ 300/ 1236] Overall Loss 0.440808 Objective Loss 0.440808 LR 0.001000 Time 0.022878 +2023-10-05 20:48:42,389 - Epoch: [14][ 310/ 1236] Overall Loss 0.440735 Objective Loss 0.440735 LR 0.001000 Time 0.022797 +2023-10-05 20:48:42,592 - Epoch: [14][ 320/ 1236] Overall Loss 0.439647 Objective Loss 0.439647 LR 0.001000 Time 0.022718 +2023-10-05 20:48:42,792 - Epoch: [14][ 330/ 1236] Overall Loss 0.438443 Objective Loss 0.438443 LR 0.001000 Time 0.022636 +2023-10-05 20:48:42,995 - Epoch: [14][ 340/ 1236] Overall Loss 0.437314 Objective Loss 0.437314 LR 0.001000 Time 0.022566 +2023-10-05 20:48:43,196 - Epoch: [14][ 350/ 1236] Overall Loss 0.437598 Objective Loss 0.437598 LR 0.001000 Time 0.022493 +2023-10-05 20:48:43,398 - Epoch: [14][ 360/ 1236] Overall Loss 0.437174 Objective Loss 0.437174 LR 0.001000 Time 0.022432 +2023-10-05 20:48:43,599 - Epoch: [14][ 370/ 1236] Overall Loss 0.436278 Objective Loss 0.436278 LR 0.001000 Time 0.022367 +2023-10-05 20:48:43,802 - Epoch: [14][ 380/ 1236] Overall Loss 0.435809 Objective Loss 0.435809 LR 0.001000 Time 0.022311 +2023-10-05 20:48:44,002 - Epoch: [14][ 390/ 1236] Overall Loss 0.434860 Objective Loss 0.434860 LR 0.001000 Time 0.022252 +2023-10-05 20:48:44,205 - Epoch: [14][ 400/ 1236] Overall Loss 0.434618 Objective Loss 0.434618 LR 0.001000 Time 0.022202 +2023-10-05 20:48:44,406 - Epoch: [14][ 410/ 1236] Overall Loss 0.435525 Objective Loss 0.435525 LR 0.001000 Time 0.022149 +2023-10-05 20:48:44,609 - Epoch: [14][ 420/ 1236] Overall Loss 0.435238 Objective Loss 0.435238 LR 0.001000 Time 0.022105 +2023-10-05 20:48:44,809 - Epoch: [14][ 430/ 1236] Overall Loss 0.434086 Objective Loss 0.434086 LR 0.001000 Time 0.022056 +2023-10-05 20:48:45,012 - Epoch: [14][ 440/ 1236] Overall Loss 0.432488 Objective Loss 0.432488 LR 0.001000 Time 0.022015 +2023-10-05 20:48:45,212 - Epoch: [14][ 450/ 1236] Overall Loss 0.431343 Objective Loss 0.431343 LR 0.001000 Time 0.021970 +2023-10-05 20:48:45,415 - Epoch: [14][ 460/ 1236] Overall Loss 0.431228 Objective Loss 0.431228 LR 0.001000 Time 0.021933 +2023-10-05 20:48:45,616 - Epoch: [14][ 470/ 1236] Overall Loss 0.430925 Objective Loss 0.430925 LR 0.001000 Time 0.021892 +2023-10-05 20:48:45,819 - Epoch: [14][ 480/ 1236] Overall Loss 0.429283 Objective Loss 0.429283 LR 0.001000 Time 0.021859 +2023-10-05 20:48:46,019 - Epoch: [14][ 490/ 1236] Overall Loss 0.427800 Objective Loss 0.427800 LR 0.001000 Time 0.021820 +2023-10-05 20:48:46,222 - Epoch: [14][ 500/ 1236] Overall Loss 0.427211 Objective Loss 0.427211 LR 0.001000 Time 0.021789 +2023-10-05 20:48:46,423 - Epoch: [14][ 510/ 1236] Overall Loss 0.426273 Objective Loss 0.426273 LR 0.001000 Time 0.021755 +2023-10-05 20:48:46,625 - Epoch: [14][ 520/ 1236] Overall Loss 0.426558 Objective Loss 0.426558 LR 0.001000 Time 0.021724 +2023-10-05 20:48:46,826 - Epoch: [14][ 530/ 1236] Overall Loss 0.426313 Objective Loss 0.426313 LR 0.001000 Time 0.021691 +2023-10-05 20:48:47,028 - Epoch: [14][ 540/ 1236] Overall Loss 0.426466 Objective Loss 0.426466 LR 0.001000 Time 0.021665 +2023-10-05 20:48:47,229 - Epoch: [14][ 550/ 1236] Overall Loss 0.426709 Objective Loss 0.426709 LR 0.001000 Time 0.021635 +2023-10-05 20:48:47,432 - Epoch: [14][ 560/ 1236] Overall Loss 0.425607 Objective Loss 0.425607 LR 0.001000 Time 0.021610 +2023-10-05 20:48:47,636 - Epoch: [14][ 570/ 1236] Overall Loss 0.425061 Objective Loss 0.425061 LR 0.001000 Time 0.021589 +2023-10-05 20:48:47,848 - Epoch: [14][ 580/ 1236] Overall Loss 0.424896 Objective Loss 0.424896 LR 0.001000 Time 0.021582 +2023-10-05 20:48:48,056 - Epoch: [14][ 590/ 1236] Overall Loss 0.424570 Objective Loss 0.424570 LR 0.001000 Time 0.021568 +2023-10-05 20:48:48,269 - Epoch: [14][ 600/ 1236] Overall Loss 0.424056 Objective Loss 0.424056 LR 0.001000 Time 0.021562 +2023-10-05 20:48:48,476 - Epoch: [14][ 610/ 1236] Overall Loss 0.424560 Objective Loss 0.424560 LR 0.001000 Time 0.021548 +2023-10-05 20:48:48,689 - Epoch: [14][ 620/ 1236] Overall Loss 0.424648 Objective Loss 0.424648 LR 0.001000 Time 0.021543 +2023-10-05 20:48:48,896 - Epoch: [14][ 630/ 1236] Overall Loss 0.424250 Objective Loss 0.424250 LR 0.001000 Time 0.021530 +2023-10-05 20:48:49,109 - Epoch: [14][ 640/ 1236] Overall Loss 0.423861 Objective Loss 0.423861 LR 0.001000 Time 0.021525 +2023-10-05 20:48:49,316 - Epoch: [14][ 650/ 1236] Overall Loss 0.423973 Objective Loss 0.423973 LR 0.001000 Time 0.021513 +2023-10-05 20:48:49,529 - Epoch: [14][ 660/ 1236] Overall Loss 0.423806 Objective Loss 0.423806 LR 0.001000 Time 0.021508 +2023-10-05 20:48:49,736 - Epoch: [14][ 670/ 1236] Overall Loss 0.424036 Objective Loss 0.424036 LR 0.001000 Time 0.021496 +2023-10-05 20:48:49,949 - Epoch: [14][ 680/ 1236] Overall Loss 0.423906 Objective Loss 0.423906 LR 0.001000 Time 0.021492 +2023-10-05 20:48:50,156 - Epoch: [14][ 690/ 1236] Overall Loss 0.424040 Objective Loss 0.424040 LR 0.001000 Time 0.021481 +2023-10-05 20:48:50,369 - Epoch: [14][ 700/ 1236] Overall Loss 0.423487 Objective Loss 0.423487 LR 0.001000 Time 0.021477 +2023-10-05 20:48:50,576 - Epoch: [14][ 710/ 1236] Overall Loss 0.423566 Objective Loss 0.423566 LR 0.001000 Time 0.021467 +2023-10-05 20:48:50,789 - Epoch: [14][ 720/ 1236] Overall Loss 0.423845 Objective Loss 0.423845 LR 0.001000 Time 0.021463 +2023-10-05 20:48:50,996 - Epoch: [14][ 730/ 1236] Overall Loss 0.423827 Objective Loss 0.423827 LR 0.001000 Time 0.021453 +2023-10-05 20:48:51,209 - Epoch: [14][ 740/ 1236] Overall Loss 0.423758 Objective Loss 0.423758 LR 0.001000 Time 0.021449 +2023-10-05 20:48:51,416 - Epoch: [14][ 750/ 1236] Overall Loss 0.423830 Objective Loss 0.423830 LR 0.001000 Time 0.021440 +2023-10-05 20:48:51,629 - Epoch: [14][ 760/ 1236] Overall Loss 0.423731 Objective Loss 0.423731 LR 0.001000 Time 0.021437 +2023-10-05 20:48:51,836 - Epoch: [14][ 770/ 1236] Overall Loss 0.423263 Objective Loss 0.423263 LR 0.001000 Time 0.021427 +2023-10-05 20:48:52,049 - Epoch: [14][ 780/ 1236] Overall Loss 0.423107 Objective Loss 0.423107 LR 0.001000 Time 0.021425 +2023-10-05 20:48:52,256 - Epoch: [14][ 790/ 1236] Overall Loss 0.423389 Objective Loss 0.423389 LR 0.001000 Time 0.021416 +2023-10-05 20:48:52,469 - Epoch: [14][ 800/ 1236] Overall Loss 0.422771 Objective Loss 0.422771 LR 0.001000 Time 0.021413 +2023-10-05 20:48:52,676 - Epoch: [14][ 810/ 1236] Overall Loss 0.422504 Objective Loss 0.422504 LR 0.001000 Time 0.021405 +2023-10-05 20:48:52,889 - Epoch: [14][ 820/ 1236] Overall Loss 0.421732 Objective Loss 0.421732 LR 0.001000 Time 0.021403 +2023-10-05 20:48:53,096 - Epoch: [14][ 830/ 1236] Overall Loss 0.421811 Objective Loss 0.421811 LR 0.001000 Time 0.021394 +2023-10-05 20:48:53,309 - Epoch: [14][ 840/ 1236] Overall Loss 0.421632 Objective Loss 0.421632 LR 0.001000 Time 0.021392 +2023-10-05 20:48:53,516 - Epoch: [14][ 850/ 1236] Overall Loss 0.421288 Objective Loss 0.421288 LR 0.001000 Time 0.021384 +2023-10-05 20:48:53,729 - Epoch: [14][ 860/ 1236] Overall Loss 0.421433 Objective Loss 0.421433 LR 0.001000 Time 0.021382 +2023-10-05 20:48:53,936 - Epoch: [14][ 870/ 1236] Overall Loss 0.420989 Objective Loss 0.420989 LR 0.001000 Time 0.021374 +2023-10-05 20:48:54,149 - Epoch: [14][ 880/ 1236] Overall Loss 0.420377 Objective Loss 0.420377 LR 0.001000 Time 0.021373 +2023-10-05 20:48:54,356 - Epoch: [14][ 890/ 1236] Overall Loss 0.420010 Objective Loss 0.420010 LR 0.001000 Time 0.021365 +2023-10-05 20:48:54,569 - Epoch: [14][ 900/ 1236] Overall Loss 0.419422 Objective Loss 0.419422 LR 0.001000 Time 0.021364 +2023-10-05 20:48:54,776 - Epoch: [14][ 910/ 1236] Overall Loss 0.419000 Objective Loss 0.419000 LR 0.001000 Time 0.021356 +2023-10-05 20:48:54,989 - Epoch: [14][ 920/ 1236] Overall Loss 0.418951 Objective Loss 0.418951 LR 0.001000 Time 0.021355 +2023-10-05 20:48:55,196 - Epoch: [14][ 930/ 1236] Overall Loss 0.419308 Objective Loss 0.419308 LR 0.001000 Time 0.021348 +2023-10-05 20:48:55,409 - Epoch: [14][ 940/ 1236] Overall Loss 0.419071 Objective Loss 0.419071 LR 0.001000 Time 0.021347 +2023-10-05 20:48:55,616 - Epoch: [14][ 950/ 1236] Overall Loss 0.418910 Objective Loss 0.418910 LR 0.001000 Time 0.021340 +2023-10-05 20:48:55,829 - Epoch: [14][ 960/ 1236] Overall Loss 0.418652 Objective Loss 0.418652 LR 0.001000 Time 0.021339 +2023-10-05 20:48:56,036 - Epoch: [14][ 970/ 1236] Overall Loss 0.418206 Objective Loss 0.418206 LR 0.001000 Time 0.021332 +2023-10-05 20:48:56,249 - Epoch: [14][ 980/ 1236] Overall Loss 0.418199 Objective Loss 0.418199 LR 0.001000 Time 0.021331 +2023-10-05 20:48:56,456 - Epoch: [14][ 990/ 1236] Overall Loss 0.417948 Objective Loss 0.417948 LR 0.001000 Time 0.021325 +2023-10-05 20:48:56,669 - Epoch: [14][ 1000/ 1236] Overall Loss 0.418068 Objective Loss 0.418068 LR 0.001000 Time 0.021324 +2023-10-05 20:48:56,876 - Epoch: [14][ 1010/ 1236] Overall Loss 0.417967 Objective Loss 0.417967 LR 0.001000 Time 0.021318 +2023-10-05 20:48:57,089 - Epoch: [14][ 1020/ 1236] Overall Loss 0.417704 Objective Loss 0.417704 LR 0.001000 Time 0.021317 +2023-10-05 20:48:57,296 - Epoch: [14][ 1030/ 1236] Overall Loss 0.418042 Objective Loss 0.418042 LR 0.001000 Time 0.021311 +2023-10-05 20:48:57,509 - Epoch: [14][ 1040/ 1236] Overall Loss 0.417822 Objective Loss 0.417822 LR 0.001000 Time 0.021311 +2023-10-05 20:48:57,716 - Epoch: [14][ 1050/ 1236] Overall Loss 0.417648 Objective Loss 0.417648 LR 0.001000 Time 0.021305 +2023-10-05 20:48:57,929 - Epoch: [14][ 1060/ 1236] Overall Loss 0.417927 Objective Loss 0.417927 LR 0.001000 Time 0.021304 +2023-10-05 20:48:58,136 - Epoch: [14][ 1070/ 1236] Overall Loss 0.417811 Objective Loss 0.417811 LR 0.001000 Time 0.021298 +2023-10-05 20:48:58,349 - Epoch: [14][ 1080/ 1236] Overall Loss 0.418038 Objective Loss 0.418038 LR 0.001000 Time 0.021298 +2023-10-05 20:48:58,557 - Epoch: [14][ 1090/ 1236] Overall Loss 0.418137 Objective Loss 0.418137 LR 0.001000 Time 0.021293 +2023-10-05 20:48:58,769 - Epoch: [14][ 1100/ 1236] Overall Loss 0.418261 Objective Loss 0.418261 LR 0.001000 Time 0.021292 +2023-10-05 20:48:58,977 - Epoch: [14][ 1110/ 1236] Overall Loss 0.418533 Objective Loss 0.418533 LR 0.001000 Time 0.021287 +2023-10-05 20:48:59,189 - Epoch: [14][ 1120/ 1236] Overall Loss 0.418515 Objective Loss 0.418515 LR 0.001000 Time 0.021287 +2023-10-05 20:48:59,397 - Epoch: [14][ 1130/ 1236] Overall Loss 0.418578 Objective Loss 0.418578 LR 0.001000 Time 0.021281 +2023-10-05 20:48:59,609 - Epoch: [14][ 1140/ 1236] Overall Loss 0.418855 Objective Loss 0.418855 LR 0.001000 Time 0.021281 +2023-10-05 20:48:59,817 - Epoch: [14][ 1150/ 1236] Overall Loss 0.418717 Objective Loss 0.418717 LR 0.001000 Time 0.021276 +2023-10-05 20:49:00,029 - Epoch: [14][ 1160/ 1236] Overall Loss 0.418616 Objective Loss 0.418616 LR 0.001000 Time 0.021276 +2023-10-05 20:49:00,237 - Epoch: [14][ 1170/ 1236] Overall Loss 0.418816 Objective Loss 0.418816 LR 0.001000 Time 0.021271 +2023-10-05 20:49:00,449 - Epoch: [14][ 1180/ 1236] Overall Loss 0.418829 Objective Loss 0.418829 LR 0.001000 Time 0.021271 +2023-10-05 20:49:00,657 - Epoch: [14][ 1190/ 1236] Overall Loss 0.419050 Objective Loss 0.419050 LR 0.001000 Time 0.021266 +2023-10-05 20:49:00,869 - Epoch: [14][ 1200/ 1236] Overall Loss 0.418982 Objective Loss 0.418982 LR 0.001000 Time 0.021266 +2023-10-05 20:49:01,077 - Epoch: [14][ 1210/ 1236] Overall Loss 0.418860 Objective Loss 0.418860 LR 0.001000 Time 0.021261 +2023-10-05 20:49:01,289 - Epoch: [14][ 1220/ 1236] Overall Loss 0.418653 Objective Loss 0.418653 LR 0.001000 Time 0.021261 +2023-10-05 20:49:01,551 - Epoch: [14][ 1230/ 1236] Overall Loss 0.418422 Objective Loss 0.418422 LR 0.001000 Time 0.021300 +2023-10-05 20:49:01,669 - Epoch: [14][ 1236/ 1236] Overall Loss 0.418374 Objective Loss 0.418374 Top1 83.503055 Top5 98.370672 LR 0.001000 Time 0.021292 +2023-10-05 20:49:01,800 - --- validate (epoch=14)----------- +2023-10-05 20:49:01,800 - 29943 samples (256 per mini-batch) +2023-10-05 20:49:02,257 - Epoch: [14][ 10/ 117] Loss 0.410995 Top1 78.281250 Top5 97.265625 +2023-10-05 20:49:02,404 - Epoch: [14][ 20/ 117] Loss 0.420208 Top1 78.261719 Top5 96.933594 +2023-10-05 20:49:02,551 - Epoch: [14][ 30/ 117] Loss 0.423436 Top1 78.281250 Top5 97.018229 +2023-10-05 20:49:02,696 - Epoch: [14][ 40/ 117] Loss 0.415683 Top1 78.564453 Top5 97.158203 +2023-10-05 20:49:02,843 - Epoch: [14][ 50/ 117] Loss 0.405063 Top1 78.921875 Top5 97.320312 +2023-10-05 20:49:02,989 - Epoch: [14][ 60/ 117] Loss 0.406509 Top1 78.880208 Top5 97.311198 +2023-10-05 20:49:03,135 - Epoch: [14][ 70/ 117] Loss 0.402473 Top1 78.967634 Top5 97.332589 +2023-10-05 20:49:03,282 - Epoch: [14][ 80/ 117] Loss 0.404531 Top1 79.038086 Top5 97.348633 +2023-10-05 20:49:03,426 - Epoch: [14][ 90/ 117] Loss 0.408077 Top1 78.936632 Top5 97.269965 +2023-10-05 20:49:03,572 - Epoch: [14][ 100/ 117] Loss 0.407150 Top1 78.886719 Top5 97.257812 +2023-10-05 20:49:03,723 - Epoch: [14][ 110/ 117] Loss 0.410387 Top1 78.902699 Top5 97.265625 +2023-10-05 20:49:03,808 - Epoch: [14][ 117/ 117] Loss 0.412757 Top1 78.849815 Top5 97.218048 +2023-10-05 20:49:03,946 - ==> Top1: 78.850 Top5: 97.218 Loss: 0.413 + +2023-10-05 20:49:03,946 - ==> Confusion: +[[ 867 2 11 2 33 1 0 0 4 77 1 1 2 7 7 8 3 0 0 0 24] + [ 2 1025 2 1 10 27 7 20 0 0 4 2 0 1 1 5 7 1 4 5 7] + [ 4 1 943 18 4 2 37 6 0 1 2 4 7 1 1 5 1 1 4 3 11] + [ 2 1 27 951 2 5 4 0 1 0 9 0 6 7 32 6 1 9 11 0 15] + [ 12 10 0 1 981 5 1 0 0 4 1 2 2 5 5 6 6 1 0 1 7] + [ 5 39 2 3 6 969 5 23 0 1 3 7 1 22 7 6 2 0 0 7 8] + [ 0 2 35 0 0 0 1124 1 0 0 2 5 1 0 0 12 0 0 1 4 4] + [ 1 16 33 0 4 37 12 1020 2 1 7 10 2 0 0 6 0 0 52 8 7] + [ 19 14 2 1 2 2 0 0 856 33 32 5 3 49 61 2 0 0 4 1 3] + [ 106 0 5 0 12 6 2 0 35 851 0 0 0 59 17 8 2 2 0 4 10] + [ 2 1 14 12 1 0 5 4 6 0 959 2 0 17 7 6 1 1 3 0 12] + [ 2 0 1 1 1 16 0 0 0 0 0 927 32 4 1 5 1 16 0 14 14] + [ 0 1 4 5 2 1 1 3 0 1 1 43 960 0 0 9 2 17 1 3 14] + [ 1 0 1 2 7 15 0 0 8 3 10 9 2 1031 4 3 1 0 0 4 18] + [ 8 3 2 21 24 0 0 0 7 3 5 0 4 3 974 0 4 8 8 0 27] + [ 1 2 3 2 3 1 1 0 0 0 1 11 11 0 0 1053 10 12 0 7 16] + [ 0 12 0 0 13 8 2 0 0 0 0 5 0 3 2 21 1067 1 1 2 24] + [ 0 0 1 2 1 0 6 0 0 0 0 8 26 2 2 12 1 967 0 1 9] + [ 2 13 11 21 3 0 4 27 3 0 5 0 6 0 14 0 1 0 948 1 9] + [ 0 1 6 0 1 6 24 13 0 0 1 16 6 3 0 5 6 1 2 1051 10] + [ 108 277 321 100 181 267 106 101 51 62 214 129 467 437 124 130 218 58 173 295 4086]] + +2023-10-05 20:49:03,947 - ==> Best [Top1: 78.850 Top5: 97.218 Sparsity:0.00 Params: 148928 on epoch: 14] +2023-10-05 20:49:03,947 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:49:03,954 - + +2023-10-05 20:49:03,954 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:49:04,932 - Epoch: [15][ 10/ 1236] Overall Loss 0.375984 Objective Loss 0.375984 LR 0.001000 Time 0.097689 +2023-10-05 20:49:05,137 - Epoch: [15][ 20/ 1236] Overall Loss 0.364825 Objective Loss 0.364825 LR 0.001000 Time 0.059080 +2023-10-05 20:49:05,340 - Epoch: [15][ 30/ 1236] Overall Loss 0.372213 Objective Loss 0.372213 LR 0.001000 Time 0.046156 +2023-10-05 20:49:05,543 - Epoch: [15][ 40/ 1236] Overall Loss 0.365874 Objective Loss 0.365874 LR 0.001000 Time 0.039674 +2023-10-05 20:49:05,744 - Epoch: [15][ 50/ 1236] Overall Loss 0.363189 Objective Loss 0.363189 LR 0.001000 Time 0.035756 +2023-10-05 20:49:05,946 - Epoch: [15][ 60/ 1236] Overall Loss 0.370642 Objective Loss 0.370642 LR 0.001000 Time 0.033162 +2023-10-05 20:49:06,147 - Epoch: [15][ 70/ 1236] Overall Loss 0.370655 Objective Loss 0.370655 LR 0.001000 Time 0.031291 +2023-10-05 20:49:06,350 - Epoch: [15][ 80/ 1236] Overall Loss 0.369470 Objective Loss 0.369470 LR 0.001000 Time 0.029911 +2023-10-05 20:49:06,550 - Epoch: [15][ 90/ 1236] Overall Loss 0.370656 Objective Loss 0.370656 LR 0.001000 Time 0.028805 +2023-10-05 20:49:06,750 - Epoch: [15][ 100/ 1236] Overall Loss 0.369395 Objective Loss 0.369395 LR 0.001000 Time 0.027923 +2023-10-05 20:49:06,951 - Epoch: [15][ 110/ 1236] Overall Loss 0.372736 Objective Loss 0.372736 LR 0.001000 Time 0.027202 +2023-10-05 20:49:07,151 - Epoch: [15][ 120/ 1236] Overall Loss 0.371319 Objective Loss 0.371319 LR 0.001000 Time 0.026600 +2023-10-05 20:49:07,351 - Epoch: [15][ 130/ 1236] Overall Loss 0.374814 Objective Loss 0.374814 LR 0.001000 Time 0.026090 +2023-10-05 20:49:07,552 - Epoch: [15][ 140/ 1236] Overall Loss 0.376544 Objective Loss 0.376544 LR 0.001000 Time 0.025658 +2023-10-05 20:49:07,752 - Epoch: [15][ 150/ 1236] Overall Loss 0.380087 Objective Loss 0.380087 LR 0.001000 Time 0.025282 +2023-10-05 20:49:07,953 - Epoch: [15][ 160/ 1236] Overall Loss 0.380969 Objective Loss 0.380969 LR 0.001000 Time 0.024954 +2023-10-05 20:49:08,153 - Epoch: [15][ 170/ 1236] Overall Loss 0.382461 Objective Loss 0.382461 LR 0.001000 Time 0.024663 +2023-10-05 20:49:08,354 - Epoch: [15][ 180/ 1236] Overall Loss 0.383772 Objective Loss 0.383772 LR 0.001000 Time 0.024404 +2023-10-05 20:49:08,554 - Epoch: [15][ 190/ 1236] Overall Loss 0.385144 Objective Loss 0.385144 LR 0.001000 Time 0.024173 +2023-10-05 20:49:08,755 - Epoch: [15][ 200/ 1236] Overall Loss 0.385159 Objective Loss 0.385159 LR 0.001000 Time 0.023965 +2023-10-05 20:49:08,955 - Epoch: [15][ 210/ 1236] Overall Loss 0.385798 Objective Loss 0.385798 LR 0.001000 Time 0.023776 +2023-10-05 20:49:09,155 - Epoch: [15][ 220/ 1236] Overall Loss 0.385581 Objective Loss 0.385581 LR 0.001000 Time 0.023605 +2023-10-05 20:49:09,356 - Epoch: [15][ 230/ 1236] Overall Loss 0.384269 Objective Loss 0.384269 LR 0.001000 Time 0.023451 +2023-10-05 20:49:09,558 - Epoch: [15][ 240/ 1236] Overall Loss 0.384930 Objective Loss 0.384930 LR 0.001000 Time 0.023310 +2023-10-05 20:49:09,758 - Epoch: [15][ 250/ 1236] Overall Loss 0.386032 Objective Loss 0.386032 LR 0.001000 Time 0.023177 +2023-10-05 20:49:09,958 - Epoch: [15][ 260/ 1236] Overall Loss 0.386185 Objective Loss 0.386185 LR 0.001000 Time 0.023055 +2023-10-05 20:49:10,158 - Epoch: [15][ 270/ 1236] Overall Loss 0.387235 Objective Loss 0.387235 LR 0.001000 Time 0.022940 +2023-10-05 20:49:10,359 - Epoch: [15][ 280/ 1236] Overall Loss 0.387536 Objective Loss 0.387536 LR 0.001000 Time 0.022836 +2023-10-05 20:49:10,559 - Epoch: [15][ 290/ 1236] Overall Loss 0.387431 Objective Loss 0.387431 LR 0.001000 Time 0.022740 +2023-10-05 20:49:10,760 - Epoch: [15][ 300/ 1236] Overall Loss 0.387804 Objective Loss 0.387804 LR 0.001000 Time 0.022649 +2023-10-05 20:49:10,960 - Epoch: [15][ 310/ 1236] Overall Loss 0.387291 Objective Loss 0.387291 LR 0.001000 Time 0.022563 +2023-10-05 20:49:11,161 - Epoch: [15][ 320/ 1236] Overall Loss 0.388284 Objective Loss 0.388284 LR 0.001000 Time 0.022484 +2023-10-05 20:49:11,361 - Epoch: [15][ 330/ 1236] Overall Loss 0.388095 Objective Loss 0.388095 LR 0.001000 Time 0.022408 +2023-10-05 20:49:11,562 - Epoch: [15][ 340/ 1236] Overall Loss 0.388142 Objective Loss 0.388142 LR 0.001000 Time 0.022339 +2023-10-05 20:49:11,762 - Epoch: [15][ 350/ 1236] Overall Loss 0.388058 Objective Loss 0.388058 LR 0.001000 Time 0.022273 +2023-10-05 20:49:11,963 - Epoch: [15][ 360/ 1236] Overall Loss 0.388437 Objective Loss 0.388437 LR 0.001000 Time 0.022211 +2023-10-05 20:49:12,164 - Epoch: [15][ 370/ 1236] Overall Loss 0.387998 Objective Loss 0.387998 LR 0.001000 Time 0.022154 +2023-10-05 20:49:12,367 - Epoch: [15][ 380/ 1236] Overall Loss 0.388212 Objective Loss 0.388212 LR 0.001000 Time 0.022103 +2023-10-05 20:49:12,569 - Epoch: [15][ 390/ 1236] Overall Loss 0.387771 Objective Loss 0.387771 LR 0.001000 Time 0.022054 +2023-10-05 20:49:12,772 - Epoch: [15][ 400/ 1236] Overall Loss 0.387756 Objective Loss 0.387756 LR 0.001000 Time 0.022009 +2023-10-05 20:49:12,974 - Epoch: [15][ 410/ 1236] Overall Loss 0.387762 Objective Loss 0.387762 LR 0.001000 Time 0.021964 +2023-10-05 20:49:13,177 - Epoch: [15][ 420/ 1236] Overall Loss 0.387876 Objective Loss 0.387876 LR 0.001000 Time 0.021924 +2023-10-05 20:49:13,379 - Epoch: [15][ 430/ 1236] Overall Loss 0.388333 Objective Loss 0.388333 LR 0.001000 Time 0.021882 +2023-10-05 20:49:13,582 - Epoch: [15][ 440/ 1236] Overall Loss 0.389509 Objective Loss 0.389509 LR 0.001000 Time 0.021846 +2023-10-05 20:49:13,784 - Epoch: [15][ 450/ 1236] Overall Loss 0.389998 Objective Loss 0.389998 LR 0.001000 Time 0.021809 +2023-10-05 20:49:13,987 - Epoch: [15][ 460/ 1236] Overall Loss 0.390177 Objective Loss 0.390177 LR 0.001000 Time 0.021775 +2023-10-05 20:49:14,189 - Epoch: [15][ 470/ 1236] Overall Loss 0.390547 Objective Loss 0.390547 LR 0.001000 Time 0.021740 +2023-10-05 20:49:14,392 - Epoch: [15][ 480/ 1236] Overall Loss 0.389629 Objective Loss 0.389629 LR 0.001000 Time 0.021710 +2023-10-05 20:49:14,594 - Epoch: [15][ 490/ 1236] Overall Loss 0.389205 Objective Loss 0.389205 LR 0.001000 Time 0.021677 +2023-10-05 20:49:14,797 - Epoch: [15][ 500/ 1236] Overall Loss 0.388615 Objective Loss 0.388615 LR 0.001000 Time 0.021650 +2023-10-05 20:49:14,998 - Epoch: [15][ 510/ 1236] Overall Loss 0.388970 Objective Loss 0.388970 LR 0.001000 Time 0.021619 +2023-10-05 20:49:15,201 - Epoch: [15][ 520/ 1236] Overall Loss 0.388465 Objective Loss 0.388465 LR 0.001000 Time 0.021593 +2023-10-05 20:49:15,403 - Epoch: [15][ 530/ 1236] Overall Loss 0.389036 Objective Loss 0.389036 LR 0.001000 Time 0.021564 +2023-10-05 20:49:15,605 - Epoch: [15][ 540/ 1236] Overall Loss 0.389298 Objective Loss 0.389298 LR 0.001000 Time 0.021540 +2023-10-05 20:49:15,807 - Epoch: [15][ 550/ 1236] Overall Loss 0.389800 Objective Loss 0.389800 LR 0.001000 Time 0.021514 +2023-10-05 20:49:16,010 - Epoch: [15][ 560/ 1236] Overall Loss 0.389737 Objective Loss 0.389737 LR 0.001000 Time 0.021492 +2023-10-05 20:49:16,212 - Epoch: [15][ 570/ 1236] Overall Loss 0.389942 Objective Loss 0.389942 LR 0.001000 Time 0.021468 +2023-10-05 20:49:16,415 - Epoch: [15][ 580/ 1236] Overall Loss 0.390312 Objective Loss 0.390312 LR 0.001000 Time 0.021448 +2023-10-05 20:49:16,617 - Epoch: [15][ 590/ 1236] Overall Loss 0.389863 Objective Loss 0.389863 LR 0.001000 Time 0.021426 +2023-10-05 20:49:16,820 - Epoch: [15][ 600/ 1236] Overall Loss 0.389645 Objective Loss 0.389645 LR 0.001000 Time 0.021407 +2023-10-05 20:49:17,022 - Epoch: [15][ 610/ 1236] Overall Loss 0.389365 Objective Loss 0.389365 LR 0.001000 Time 0.021386 +2023-10-05 20:49:17,225 - Epoch: [15][ 620/ 1236] Overall Loss 0.389228 Objective Loss 0.389228 LR 0.001000 Time 0.021369 +2023-10-05 20:49:17,427 - Epoch: [15][ 630/ 1236] Overall Loss 0.389516 Objective Loss 0.389516 LR 0.001000 Time 0.021348 +2023-10-05 20:49:17,630 - Epoch: [15][ 640/ 1236] Overall Loss 0.389190 Objective Loss 0.389190 LR 0.001000 Time 0.021332 +2023-10-05 20:49:17,832 - Epoch: [15][ 650/ 1236] Overall Loss 0.389369 Objective Loss 0.389369 LR 0.001000 Time 0.021314 +2023-10-05 20:49:18,035 - Epoch: [15][ 660/ 1236] Overall Loss 0.389184 Objective Loss 0.389184 LR 0.001000 Time 0.021298 +2023-10-05 20:49:18,237 - Epoch: [15][ 670/ 1236] Overall Loss 0.389188 Objective Loss 0.389188 LR 0.001000 Time 0.021281 +2023-10-05 20:49:18,441 - Epoch: [15][ 680/ 1236] Overall Loss 0.389076 Objective Loss 0.389076 LR 0.001000 Time 0.021267 +2023-10-05 20:49:18,642 - Epoch: [15][ 690/ 1236] Overall Loss 0.389603 Objective Loss 0.389603 LR 0.001000 Time 0.021250 +2023-10-05 20:49:18,845 - Epoch: [15][ 700/ 1236] Overall Loss 0.389769 Objective Loss 0.389769 LR 0.001000 Time 0.021236 +2023-10-05 20:49:19,047 - Epoch: [15][ 710/ 1236] Overall Loss 0.389787 Objective Loss 0.389787 LR 0.001000 Time 0.021221 +2023-10-05 20:49:19,250 - Epoch: [15][ 720/ 1236] Overall Loss 0.389757 Objective Loss 0.389757 LR 0.001000 Time 0.021208 +2023-10-05 20:49:19,452 - Epoch: [15][ 730/ 1236] Overall Loss 0.390288 Objective Loss 0.390288 LR 0.001000 Time 0.021194 +2023-10-05 20:49:19,655 - Epoch: [15][ 740/ 1236] Overall Loss 0.390021 Objective Loss 0.390021 LR 0.001000 Time 0.021181 +2023-10-05 20:49:19,857 - Epoch: [15][ 750/ 1236] Overall Loss 0.389889 Objective Loss 0.389889 LR 0.001000 Time 0.021167 +2023-10-05 20:49:20,060 - Epoch: [15][ 760/ 1236] Overall Loss 0.390306 Objective Loss 0.390306 LR 0.001000 Time 0.021156 +2023-10-05 20:49:20,263 - Epoch: [15][ 770/ 1236] Overall Loss 0.390244 Objective Loss 0.390244 LR 0.001000 Time 0.021143 +2023-10-05 20:49:20,464 - Epoch: [15][ 780/ 1236] Overall Loss 0.390861 Objective Loss 0.390861 LR 0.001000 Time 0.021129 +2023-10-05 20:49:20,663 - Epoch: [15][ 790/ 1236] Overall Loss 0.391325 Objective Loss 0.391325 LR 0.001000 Time 0.021114 +2023-10-05 20:49:20,865 - Epoch: [15][ 800/ 1236] Overall Loss 0.391770 Objective Loss 0.391770 LR 0.001000 Time 0.021102 +2023-10-05 20:49:21,064 - Epoch: [15][ 810/ 1236] Overall Loss 0.391948 Objective Loss 0.391948 LR 0.001000 Time 0.021087 +2023-10-05 20:49:21,266 - Epoch: [15][ 820/ 1236] Overall Loss 0.392137 Objective Loss 0.392137 LR 0.001000 Time 0.021075 +2023-10-05 20:49:21,466 - Epoch: [15][ 830/ 1236] Overall Loss 0.392344 Objective Loss 0.392344 LR 0.001000 Time 0.021062 +2023-10-05 20:49:21,667 - Epoch: [15][ 840/ 1236] Overall Loss 0.392469 Objective Loss 0.392469 LR 0.001000 Time 0.021050 +2023-10-05 20:49:21,866 - Epoch: [15][ 850/ 1236] Overall Loss 0.392068 Objective Loss 0.392068 LR 0.001000 Time 0.021037 +2023-10-05 20:49:22,068 - Epoch: [15][ 860/ 1236] Overall Loss 0.392018 Objective Loss 0.392018 LR 0.001000 Time 0.021026 +2023-10-05 20:49:22,268 - Epoch: [15][ 870/ 1236] Overall Loss 0.391645 Objective Loss 0.391645 LR 0.001000 Time 0.021014 +2023-10-05 20:49:22,471 - Epoch: [15][ 880/ 1236] Overall Loss 0.391812 Objective Loss 0.391812 LR 0.001000 Time 0.021005 +2023-10-05 20:49:22,670 - Epoch: [15][ 890/ 1236] Overall Loss 0.391762 Objective Loss 0.391762 LR 0.001000 Time 0.020993 +2023-10-05 20:49:22,872 - Epoch: [15][ 900/ 1236] Overall Loss 0.391734 Objective Loss 0.391734 LR 0.001000 Time 0.020983 +2023-10-05 20:49:23,072 - Epoch: [15][ 910/ 1236] Overall Loss 0.391808 Objective Loss 0.391808 LR 0.001000 Time 0.020971 +2023-10-05 20:49:23,273 - Epoch: [15][ 920/ 1236] Overall Loss 0.391810 Objective Loss 0.391810 LR 0.001000 Time 0.020962 +2023-10-05 20:49:23,473 - Epoch: [15][ 930/ 1236] Overall Loss 0.391719 Objective Loss 0.391719 LR 0.001000 Time 0.020952 +2023-10-05 20:49:23,675 - Epoch: [15][ 940/ 1236] Overall Loss 0.391395 Objective Loss 0.391395 LR 0.001000 Time 0.020943 +2023-10-05 20:49:23,874 - Epoch: [15][ 950/ 1236] Overall Loss 0.391434 Objective Loss 0.391434 LR 0.001000 Time 0.020932 +2023-10-05 20:49:24,076 - Epoch: [15][ 960/ 1236] Overall Loss 0.391489 Objective Loss 0.391489 LR 0.001000 Time 0.020923 +2023-10-05 20:49:24,275 - Epoch: [15][ 970/ 1236] Overall Loss 0.391336 Objective Loss 0.391336 LR 0.001000 Time 0.020913 +2023-10-05 20:49:24,477 - Epoch: [15][ 980/ 1236] Overall Loss 0.391108 Objective Loss 0.391108 LR 0.001000 Time 0.020905 +2023-10-05 20:49:24,677 - Epoch: [15][ 990/ 1236] Overall Loss 0.391314 Objective Loss 0.391314 LR 0.001000 Time 0.020895 +2023-10-05 20:49:24,878 - Epoch: [15][ 1000/ 1236] Overall Loss 0.391755 Objective Loss 0.391755 LR 0.001000 Time 0.020887 +2023-10-05 20:49:25,078 - Epoch: [15][ 1010/ 1236] Overall Loss 0.391787 Objective Loss 0.391787 LR 0.001000 Time 0.020877 +2023-10-05 20:49:25,279 - Epoch: [15][ 1020/ 1236] Overall Loss 0.391685 Objective Loss 0.391685 LR 0.001000 Time 0.020869 +2023-10-05 20:49:25,479 - Epoch: [15][ 1030/ 1236] Overall Loss 0.391944 Objective Loss 0.391944 LR 0.001000 Time 0.020860 +2023-10-05 20:49:25,680 - Epoch: [15][ 1040/ 1236] Overall Loss 0.391664 Objective Loss 0.391664 LR 0.001000 Time 0.020853 +2023-10-05 20:49:25,880 - Epoch: [15][ 1050/ 1236] Overall Loss 0.391608 Objective Loss 0.391608 LR 0.001000 Time 0.020844 +2023-10-05 20:49:26,082 - Epoch: [15][ 1060/ 1236] Overall Loss 0.391711 Objective Loss 0.391711 LR 0.001000 Time 0.020838 +2023-10-05 20:49:26,281 - Epoch: [15][ 1070/ 1236] Overall Loss 0.391705 Objective Loss 0.391705 LR 0.001000 Time 0.020829 +2023-10-05 20:49:26,483 - Epoch: [15][ 1080/ 1236] Overall Loss 0.391911 Objective Loss 0.391911 LR 0.001000 Time 0.020823 +2023-10-05 20:49:26,683 - Epoch: [15][ 1090/ 1236] Overall Loss 0.391713 Objective Loss 0.391713 LR 0.001000 Time 0.020815 +2023-10-05 20:49:26,884 - Epoch: [15][ 1100/ 1236] Overall Loss 0.391694 Objective Loss 0.391694 LR 0.001000 Time 0.020808 +2023-10-05 20:49:27,084 - Epoch: [15][ 1110/ 1236] Overall Loss 0.391610 Objective Loss 0.391610 LR 0.001000 Time 0.020800 +2023-10-05 20:49:27,285 - Epoch: [15][ 1120/ 1236] Overall Loss 0.391476 Objective Loss 0.391476 LR 0.001000 Time 0.020794 +2023-10-05 20:49:27,485 - Epoch: [15][ 1130/ 1236] Overall Loss 0.391260 Objective Loss 0.391260 LR 0.001000 Time 0.020786 +2023-10-05 20:49:27,687 - Epoch: [15][ 1140/ 1236] Overall Loss 0.391536 Objective Loss 0.391536 LR 0.001000 Time 0.020780 +2023-10-05 20:49:27,886 - Epoch: [15][ 1150/ 1236] Overall Loss 0.391615 Objective Loss 0.391615 LR 0.001000 Time 0.020773 +2023-10-05 20:49:28,088 - Epoch: [15][ 1160/ 1236] Overall Loss 0.391681 Objective Loss 0.391681 LR 0.001000 Time 0.020768 +2023-10-05 20:49:28,288 - Epoch: [15][ 1170/ 1236] Overall Loss 0.391960 Objective Loss 0.391960 LR 0.001000 Time 0.020761 +2023-10-05 20:49:28,490 - Epoch: [15][ 1180/ 1236] Overall Loss 0.391916 Objective Loss 0.391916 LR 0.001000 Time 0.020756 +2023-10-05 20:49:28,690 - Epoch: [15][ 1190/ 1236] Overall Loss 0.391980 Objective Loss 0.391980 LR 0.001000 Time 0.020749 +2023-10-05 20:49:28,892 - Epoch: [15][ 1200/ 1236] Overall Loss 0.391675 Objective Loss 0.391675 LR 0.001000 Time 0.020744 +2023-10-05 20:49:29,092 - Epoch: [15][ 1210/ 1236] Overall Loss 0.391505 Objective Loss 0.391505 LR 0.001000 Time 0.020738 +2023-10-05 20:49:29,294 - Epoch: [15][ 1220/ 1236] Overall Loss 0.391634 Objective Loss 0.391634 LR 0.001000 Time 0.020733 +2023-10-05 20:49:29,546 - Epoch: [15][ 1230/ 1236] Overall Loss 0.391626 Objective Loss 0.391626 LR 0.001000 Time 0.020768 +2023-10-05 20:49:29,663 - Epoch: [15][ 1236/ 1236] Overall Loss 0.391397 Objective Loss 0.391397 Top1 84.114053 Top5 98.574338 LR 0.001000 Time 0.020762 +2023-10-05 20:49:29,788 - --- validate (epoch=15)----------- +2023-10-05 20:49:29,789 - 29943 samples (256 per mini-batch) +2023-10-05 20:49:30,235 - Epoch: [15][ 10/ 117] Loss 0.401358 Top1 79.140625 Top5 96.914062 +2023-10-05 20:49:30,384 - Epoch: [15][ 20/ 117] Loss 0.389587 Top1 78.066406 Top5 97.031250 +2023-10-05 20:49:30,534 - Epoch: [15][ 30/ 117] Loss 0.389827 Top1 77.955729 Top5 97.161458 +2023-10-05 20:49:30,682 - Epoch: [15][ 40/ 117] Loss 0.388304 Top1 78.144531 Top5 97.314453 +2023-10-05 20:49:30,832 - Epoch: [15][ 50/ 117] Loss 0.387233 Top1 78.187500 Top5 97.187500 +2023-10-05 20:49:30,983 - Epoch: [15][ 60/ 117] Loss 0.388659 Top1 78.216146 Top5 97.246094 +2023-10-05 20:49:31,139 - Epoch: [15][ 70/ 117] Loss 0.389665 Top1 78.180804 Top5 97.248884 +2023-10-05 20:49:31,292 - Epoch: [15][ 80/ 117] Loss 0.394321 Top1 78.090820 Top5 97.143555 +2023-10-05 20:49:31,446 - Epoch: [15][ 90/ 117] Loss 0.392475 Top1 78.159722 Top5 97.113715 +2023-10-05 20:49:31,600 - Epoch: [15][ 100/ 117] Loss 0.389308 Top1 78.250000 Top5 97.160156 +2023-10-05 20:49:31,755 - Epoch: [15][ 110/ 117] Loss 0.390931 Top1 78.199574 Top5 97.098722 +2023-10-05 20:49:31,840 - Epoch: [15][ 117/ 117] Loss 0.389682 Top1 78.332164 Top5 97.104499 +2023-10-05 20:49:31,955 - ==> Top1: 78.332 Top5: 97.104 Loss: 0.390 + +2023-10-05 20:49:31,956 - ==> Confusion: +[[ 937 1 2 2 3 7 0 1 13 63 0 1 0 5 1 1 3 0 1 1 8] + [ 1 1018 0 0 5 33 2 33 2 0 7 2 0 0 2 3 8 3 8 2 2] + [ 12 1 917 21 2 0 38 10 0 2 6 6 5 1 1 2 4 2 11 3 12] + [ 5 1 22 936 0 8 2 1 12 0 3 2 8 5 32 3 0 10 21 2 16] + [ 44 12 1 0 935 8 0 0 5 14 0 2 0 2 7 3 12 1 0 0 4] + [ 9 44 0 4 6 948 1 27 4 7 6 9 2 27 3 0 6 0 2 3 8] + [ 1 7 20 0 0 0 1114 21 0 0 2 5 2 0 0 6 0 2 0 6 5] + [ 6 18 12 0 0 28 4 1083 3 2 3 12 2 0 0 1 0 0 29 6 9] + [ 21 2 0 0 0 0 0 1 964 43 14 6 4 14 9 0 0 3 7 1 0] + [ 142 1 0 0 0 0 0 1 49 888 1 2 0 12 6 1 1 1 2 3 9] + [ 8 3 8 8 0 2 2 5 29 3 950 3 0 15 2 1 1 0 3 0 10] + [ 3 1 1 0 0 17 0 0 0 0 0 958 15 7 1 3 1 13 0 11 4] + [ 0 0 3 7 0 0 0 4 2 0 1 78 916 1 3 4 5 33 1 5 5] + [ 3 0 1 1 4 10 0 0 29 25 6 10 2 1015 2 1 0 1 0 2 7] + [ 16 3 2 11 3 0 0 0 48 5 0 0 4 2 969 0 4 6 16 0 12] + [ 2 2 2 3 4 1 2 1 0 0 0 17 8 1 0 1042 13 14 0 12 10] + [ 2 14 3 0 9 9 0 0 0 0 0 9 0 2 3 12 1080 1 0 5 12] + [ 3 0 0 3 0 0 1 0 0 0 0 14 13 1 0 10 1 986 1 1 4] + [ 0 5 9 18 1 2 1 47 9 1 3 0 5 0 4 0 1 0 953 3 6] + [ 0 5 4 0 1 6 12 13 0 0 1 24 2 0 0 1 7 0 1 1071 4] + [ 371 279 177 75 100 226 54 150 213 119 212 230 427 397 156 77 268 107 159 333 3775]] + +2023-10-05 20:49:31,957 - ==> Best [Top1: 78.850 Top5: 97.218 Sparsity:0.00 Params: 148928 on epoch: 14] +2023-10-05 20:49:31,957 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:49:31,963 - + +2023-10-05 20:49:31,963 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:49:33,048 - Epoch: [16][ 10/ 1236] Overall Loss 0.388647 Objective Loss 0.388647 LR 0.001000 Time 0.108384 +2023-10-05 20:49:33,249 - Epoch: [16][ 20/ 1236] Overall Loss 0.379587 Objective Loss 0.379587 LR 0.001000 Time 0.064223 +2023-10-05 20:49:33,448 - Epoch: [16][ 30/ 1236] Overall Loss 0.378519 Objective Loss 0.378519 LR 0.001000 Time 0.049447 +2023-10-05 20:49:33,651 - Epoch: [16][ 40/ 1236] Overall Loss 0.384574 Objective Loss 0.384574 LR 0.001000 Time 0.042149 +2023-10-05 20:49:33,850 - Epoch: [16][ 50/ 1236] Overall Loss 0.383818 Objective Loss 0.383818 LR 0.001000 Time 0.037706 +2023-10-05 20:49:34,052 - Epoch: [16][ 60/ 1236] Overall Loss 0.370304 Objective Loss 0.370304 LR 0.001000 Time 0.034785 +2023-10-05 20:49:34,252 - Epoch: [16][ 70/ 1236] Overall Loss 0.372787 Objective Loss 0.372787 LR 0.001000 Time 0.032659 +2023-10-05 20:49:34,454 - Epoch: [16][ 80/ 1236] Overall Loss 0.368710 Objective Loss 0.368710 LR 0.001000 Time 0.031101 +2023-10-05 20:49:34,654 - Epoch: [16][ 90/ 1236] Overall Loss 0.368268 Objective Loss 0.368268 LR 0.001000 Time 0.029859 +2023-10-05 20:49:34,857 - Epoch: [16][ 100/ 1236] Overall Loss 0.365514 Objective Loss 0.365514 LR 0.001000 Time 0.028899 +2023-10-05 20:49:35,056 - Epoch: [16][ 110/ 1236] Overall Loss 0.364220 Objective Loss 0.364220 LR 0.001000 Time 0.028078 +2023-10-05 20:49:35,258 - Epoch: [16][ 120/ 1236] Overall Loss 0.364857 Objective Loss 0.364857 LR 0.001000 Time 0.027422 +2023-10-05 20:49:35,458 - Epoch: [16][ 130/ 1236] Overall Loss 0.368204 Objective Loss 0.368204 LR 0.001000 Time 0.026844 +2023-10-05 20:49:35,660 - Epoch: [16][ 140/ 1236] Overall Loss 0.369143 Objective Loss 0.369143 LR 0.001000 Time 0.026371 +2023-10-05 20:49:35,859 - Epoch: [16][ 150/ 1236] Overall Loss 0.370412 Objective Loss 0.370412 LR 0.001000 Time 0.025937 +2023-10-05 20:49:36,062 - Epoch: [16][ 160/ 1236] Overall Loss 0.370608 Objective Loss 0.370608 LR 0.001000 Time 0.025580 +2023-10-05 20:49:36,261 - Epoch: [16][ 170/ 1236] Overall Loss 0.371320 Objective Loss 0.371320 LR 0.001000 Time 0.025243 +2023-10-05 20:49:36,463 - Epoch: [16][ 180/ 1236] Overall Loss 0.370706 Objective Loss 0.370706 LR 0.001000 Time 0.024963 +2023-10-05 20:49:36,663 - Epoch: [16][ 190/ 1236] Overall Loss 0.370198 Objective Loss 0.370198 LR 0.001000 Time 0.024698 +2023-10-05 20:49:36,865 - Epoch: [16][ 200/ 1236] Overall Loss 0.370828 Objective Loss 0.370828 LR 0.001000 Time 0.024473 +2023-10-05 20:49:37,068 - Epoch: [16][ 210/ 1236] Overall Loss 0.371314 Objective Loss 0.371314 LR 0.001000 Time 0.024272 +2023-10-05 20:49:37,274 - Epoch: [16][ 220/ 1236] Overall Loss 0.370704 Objective Loss 0.370704 LR 0.001000 Time 0.024103 +2023-10-05 20:49:37,477 - Epoch: [16][ 230/ 1236] Overall Loss 0.371854 Objective Loss 0.371854 LR 0.001000 Time 0.023937 +2023-10-05 20:49:37,683 - Epoch: [16][ 240/ 1236] Overall Loss 0.372103 Objective Loss 0.372103 LR 0.001000 Time 0.023797 +2023-10-05 20:49:37,886 - Epoch: [16][ 250/ 1236] Overall Loss 0.370358 Objective Loss 0.370358 LR 0.001000 Time 0.023653 +2023-10-05 20:49:38,091 - Epoch: [16][ 260/ 1236] Overall Loss 0.370732 Objective Loss 0.370732 LR 0.001000 Time 0.023532 +2023-10-05 20:49:38,290 - Epoch: [16][ 270/ 1236] Overall Loss 0.370699 Objective Loss 0.370699 LR 0.001000 Time 0.023395 +2023-10-05 20:49:38,488 - Epoch: [16][ 280/ 1236] Overall Loss 0.370747 Objective Loss 0.370747 LR 0.001000 Time 0.023266 +2023-10-05 20:49:38,687 - Epoch: [16][ 290/ 1236] Overall Loss 0.370813 Objective Loss 0.370813 LR 0.001000 Time 0.023150 +2023-10-05 20:49:38,885 - Epoch: [16][ 300/ 1236] Overall Loss 0.370308 Objective Loss 0.370308 LR 0.001000 Time 0.023036 +2023-10-05 20:49:39,084 - Epoch: [16][ 310/ 1236] Overall Loss 0.370178 Objective Loss 0.370178 LR 0.001000 Time 0.022935 +2023-10-05 20:49:39,282 - Epoch: [16][ 320/ 1236] Overall Loss 0.370637 Objective Loss 0.370637 LR 0.001000 Time 0.022836 +2023-10-05 20:49:39,481 - Epoch: [16][ 330/ 1236] Overall Loss 0.370178 Objective Loss 0.370178 LR 0.001000 Time 0.022747 +2023-10-05 20:49:39,679 - Epoch: [16][ 340/ 1236] Overall Loss 0.370584 Objective Loss 0.370584 LR 0.001000 Time 0.022659 +2023-10-05 20:49:39,882 - Epoch: [16][ 350/ 1236] Overall Loss 0.370202 Objective Loss 0.370202 LR 0.001000 Time 0.022593 +2023-10-05 20:49:40,078 - Epoch: [16][ 360/ 1236] Overall Loss 0.371101 Objective Loss 0.371101 LR 0.001000 Time 0.022509 +2023-10-05 20:49:40,277 - Epoch: [16][ 370/ 1236] Overall Loss 0.370912 Objective Loss 0.370912 LR 0.001000 Time 0.022435 +2023-10-05 20:49:40,476 - Epoch: [16][ 380/ 1236] Overall Loss 0.371595 Objective Loss 0.371595 LR 0.001000 Time 0.022370 +2023-10-05 20:49:40,675 - Epoch: [16][ 390/ 1236] Overall Loss 0.372262 Objective Loss 0.372262 LR 0.001000 Time 0.022306 +2023-10-05 20:49:40,876 - Epoch: [16][ 400/ 1236] Overall Loss 0.372277 Objective Loss 0.372277 LR 0.001000 Time 0.022248 +2023-10-05 20:49:41,074 - Epoch: [16][ 410/ 1236] Overall Loss 0.371631 Objective Loss 0.371631 LR 0.001000 Time 0.022188 +2023-10-05 20:49:41,274 - Epoch: [16][ 420/ 1236] Overall Loss 0.372113 Objective Loss 0.372113 LR 0.001000 Time 0.022136 +2023-10-05 20:49:41,472 - Epoch: [16][ 430/ 1236] Overall Loss 0.372160 Objective Loss 0.372160 LR 0.001000 Time 0.022082 +2023-10-05 20:49:41,672 - Epoch: [16][ 440/ 1236] Overall Loss 0.372827 Objective Loss 0.372827 LR 0.001000 Time 0.022033 +2023-10-05 20:49:41,871 - Epoch: [16][ 450/ 1236] Overall Loss 0.373018 Objective Loss 0.373018 LR 0.001000 Time 0.021985 +2023-10-05 20:49:42,071 - Epoch: [16][ 460/ 1236] Overall Loss 0.372667 Objective Loss 0.372667 LR 0.001000 Time 0.021941 +2023-10-05 20:49:42,270 - Epoch: [16][ 470/ 1236] Overall Loss 0.372903 Objective Loss 0.372903 LR 0.001000 Time 0.021897 +2023-10-05 20:49:42,470 - Epoch: [16][ 480/ 1236] Overall Loss 0.373407 Objective Loss 0.373407 LR 0.001000 Time 0.021857 +2023-10-05 20:49:42,668 - Epoch: [16][ 490/ 1236] Overall Loss 0.373182 Objective Loss 0.373182 LR 0.001000 Time 0.021815 +2023-10-05 20:49:42,868 - Epoch: [16][ 500/ 1236] Overall Loss 0.372803 Objective Loss 0.372803 LR 0.001000 Time 0.021778 +2023-10-05 20:49:43,067 - Epoch: [16][ 510/ 1236] Overall Loss 0.372779 Objective Loss 0.372779 LR 0.001000 Time 0.021740 +2023-10-05 20:49:43,266 - Epoch: [16][ 520/ 1236] Overall Loss 0.372437 Objective Loss 0.372437 LR 0.001000 Time 0.021705 +2023-10-05 20:49:43,465 - Epoch: [16][ 530/ 1236] Overall Loss 0.372728 Objective Loss 0.372728 LR 0.001000 Time 0.021670 +2023-10-05 20:49:43,665 - Epoch: [16][ 540/ 1236] Overall Loss 0.372258 Objective Loss 0.372258 LR 0.001000 Time 0.021638 +2023-10-05 20:49:43,864 - Epoch: [16][ 550/ 1236] Overall Loss 0.372685 Objective Loss 0.372685 LR 0.001000 Time 0.021606 +2023-10-05 20:49:44,064 - Epoch: [16][ 560/ 1236] Overall Loss 0.372919 Objective Loss 0.372919 LR 0.001000 Time 0.021577 +2023-10-05 20:49:44,262 - Epoch: [16][ 570/ 1236] Overall Loss 0.372781 Objective Loss 0.372781 LR 0.001000 Time 0.021546 +2023-10-05 20:49:44,462 - Epoch: [16][ 580/ 1236] Overall Loss 0.373483 Objective Loss 0.373483 LR 0.001000 Time 0.021518 +2023-10-05 20:49:44,661 - Epoch: [16][ 590/ 1236] Overall Loss 0.373880 Objective Loss 0.373880 LR 0.001000 Time 0.021491 +2023-10-05 20:49:44,861 - Epoch: [16][ 600/ 1236] Overall Loss 0.373946 Objective Loss 0.373946 LR 0.001000 Time 0.021465 +2023-10-05 20:49:45,060 - Epoch: [16][ 610/ 1236] Overall Loss 0.374165 Objective Loss 0.374165 LR 0.001000 Time 0.021439 +2023-10-05 20:49:45,260 - Epoch: [16][ 620/ 1236] Overall Loss 0.374449 Objective Loss 0.374449 LR 0.001000 Time 0.021415 +2023-10-05 20:49:45,459 - Epoch: [16][ 630/ 1236] Overall Loss 0.375031 Objective Loss 0.375031 LR 0.001000 Time 0.021391 +2023-10-05 20:49:45,659 - Epoch: [16][ 640/ 1236] Overall Loss 0.374614 Objective Loss 0.374614 LR 0.001000 Time 0.021368 +2023-10-05 20:49:45,857 - Epoch: [16][ 650/ 1236] Overall Loss 0.375072 Objective Loss 0.375072 LR 0.001000 Time 0.021345 +2023-10-05 20:49:46,057 - Epoch: [16][ 660/ 1236] Overall Loss 0.374853 Objective Loss 0.374853 LR 0.001000 Time 0.021323 +2023-10-05 20:49:46,256 - Epoch: [16][ 670/ 1236] Overall Loss 0.375003 Objective Loss 0.375003 LR 0.001000 Time 0.021301 +2023-10-05 20:49:46,456 - Epoch: [16][ 680/ 1236] Overall Loss 0.375190 Objective Loss 0.375190 LR 0.001000 Time 0.021282 +2023-10-05 20:49:46,655 - Epoch: [16][ 690/ 1236] Overall Loss 0.375398 Objective Loss 0.375398 LR 0.001000 Time 0.021262 +2023-10-05 20:49:46,855 - Epoch: [16][ 700/ 1236] Overall Loss 0.374975 Objective Loss 0.374975 LR 0.001000 Time 0.021243 +2023-10-05 20:49:47,053 - Epoch: [16][ 710/ 1236] Overall Loss 0.375038 Objective Loss 0.375038 LR 0.001000 Time 0.021223 +2023-10-05 20:49:47,253 - Epoch: [16][ 720/ 1236] Overall Loss 0.375525 Objective Loss 0.375525 LR 0.001000 Time 0.021205 +2023-10-05 20:49:47,452 - Epoch: [16][ 730/ 1236] Overall Loss 0.374959 Objective Loss 0.374959 LR 0.001000 Time 0.021187 +2023-10-05 20:49:47,652 - Epoch: [16][ 740/ 1236] Overall Loss 0.375247 Objective Loss 0.375247 LR 0.001000 Time 0.021171 +2023-10-05 20:49:47,851 - Epoch: [16][ 750/ 1236] Overall Loss 0.375170 Objective Loss 0.375170 LR 0.001000 Time 0.021153 +2023-10-05 20:49:48,051 - Epoch: [16][ 760/ 1236] Overall Loss 0.374978 Objective Loss 0.374978 LR 0.001000 Time 0.021138 +2023-10-05 20:49:48,249 - Epoch: [16][ 770/ 1236] Overall Loss 0.375109 Objective Loss 0.375109 LR 0.001000 Time 0.021121 +2023-10-05 20:49:48,449 - Epoch: [16][ 780/ 1236] Overall Loss 0.375534 Objective Loss 0.375534 LR 0.001000 Time 0.021106 +2023-10-05 20:49:48,649 - Epoch: [16][ 790/ 1236] Overall Loss 0.375589 Objective Loss 0.375589 LR 0.001000 Time 0.021091 +2023-10-05 20:49:48,848 - Epoch: [16][ 800/ 1236] Overall Loss 0.375981 Objective Loss 0.375981 LR 0.001000 Time 0.021076 +2023-10-05 20:49:49,047 - Epoch: [16][ 810/ 1236] Overall Loss 0.376097 Objective Loss 0.376097 LR 0.001000 Time 0.021061 +2023-10-05 20:49:49,247 - Epoch: [16][ 820/ 1236] Overall Loss 0.376665 Objective Loss 0.376665 LR 0.001000 Time 0.021047 +2023-10-05 20:49:49,446 - Epoch: [16][ 830/ 1236] Overall Loss 0.377027 Objective Loss 0.377027 LR 0.001000 Time 0.021033 +2023-10-05 20:49:49,646 - Epoch: [16][ 840/ 1236] Overall Loss 0.377012 Objective Loss 0.377012 LR 0.001000 Time 0.021021 +2023-10-05 20:49:49,845 - Epoch: [16][ 850/ 1236] Overall Loss 0.377683 Objective Loss 0.377683 LR 0.001000 Time 0.021007 +2023-10-05 20:49:50,044 - Epoch: [16][ 860/ 1236] Overall Loss 0.377688 Objective Loss 0.377688 LR 0.001000 Time 0.020994 +2023-10-05 20:49:50,243 - Epoch: [16][ 870/ 1236] Overall Loss 0.378054 Objective Loss 0.378054 LR 0.001000 Time 0.020981 +2023-10-05 20:49:50,443 - Epoch: [16][ 880/ 1236] Overall Loss 0.378088 Objective Loss 0.378088 LR 0.001000 Time 0.020970 +2023-10-05 20:49:50,642 - Epoch: [16][ 890/ 1236] Overall Loss 0.378494 Objective Loss 0.378494 LR 0.001000 Time 0.020957 +2023-10-05 20:49:50,842 - Epoch: [16][ 900/ 1236] Overall Loss 0.378531 Objective Loss 0.378531 LR 0.001000 Time 0.020946 +2023-10-05 20:49:51,041 - Epoch: [16][ 910/ 1236] Overall Loss 0.378540 Objective Loss 0.378540 LR 0.001000 Time 0.020934 +2023-10-05 20:49:51,240 - Epoch: [16][ 920/ 1236] Overall Loss 0.378802 Objective Loss 0.378802 LR 0.001000 Time 0.020923 +2023-10-05 20:49:51,439 - Epoch: [16][ 930/ 1236] Overall Loss 0.378950 Objective Loss 0.378950 LR 0.001000 Time 0.020912 +2023-10-05 20:49:51,639 - Epoch: [16][ 940/ 1236] Overall Loss 0.379436 Objective Loss 0.379436 LR 0.001000 Time 0.020902 +2023-10-05 20:49:51,838 - Epoch: [16][ 950/ 1236] Overall Loss 0.379462 Objective Loss 0.379462 LR 0.001000 Time 0.020891 +2023-10-05 20:49:52,038 - Epoch: [16][ 960/ 1236] Overall Loss 0.379717 Objective Loss 0.379717 LR 0.001000 Time 0.020881 +2023-10-05 20:49:52,237 - Epoch: [16][ 970/ 1236] Overall Loss 0.379958 Objective Loss 0.379958 LR 0.001000 Time 0.020871 +2023-10-05 20:49:52,437 - Epoch: [16][ 980/ 1236] Overall Loss 0.380250 Objective Loss 0.380250 LR 0.001000 Time 0.020862 +2023-10-05 20:49:52,636 - Epoch: [16][ 990/ 1236] Overall Loss 0.380246 Objective Loss 0.380246 LR 0.001000 Time 0.020851 +2023-10-05 20:49:52,836 - Epoch: [16][ 1000/ 1236] Overall Loss 0.380145 Objective Loss 0.380145 LR 0.001000 Time 0.020843 +2023-10-05 20:49:53,034 - Epoch: [16][ 1010/ 1236] Overall Loss 0.380243 Objective Loss 0.380243 LR 0.001000 Time 0.020833 +2023-10-05 20:49:53,234 - Epoch: [16][ 1020/ 1236] Overall Loss 0.380613 Objective Loss 0.380613 LR 0.001000 Time 0.020824 +2023-10-05 20:49:53,433 - Epoch: [16][ 1030/ 1236] Overall Loss 0.380463 Objective Loss 0.380463 LR 0.001000 Time 0.020815 +2023-10-05 20:49:53,633 - Epoch: [16][ 1040/ 1236] Overall Loss 0.380237 Objective Loss 0.380237 LR 0.001000 Time 0.020807 +2023-10-05 20:49:53,832 - Epoch: [16][ 1050/ 1236] Overall Loss 0.380015 Objective Loss 0.380015 LR 0.001000 Time 0.020798 +2023-10-05 20:49:54,032 - Epoch: [16][ 1060/ 1236] Overall Loss 0.380180 Objective Loss 0.380180 LR 0.001000 Time 0.020790 +2023-10-05 20:49:54,231 - Epoch: [16][ 1070/ 1236] Overall Loss 0.380269 Objective Loss 0.380269 LR 0.001000 Time 0.020781 +2023-10-05 20:49:54,431 - Epoch: [16][ 1080/ 1236] Overall Loss 0.380219 Objective Loss 0.380219 LR 0.001000 Time 0.020774 +2023-10-05 20:49:54,630 - Epoch: [16][ 1090/ 1236] Overall Loss 0.380061 Objective Loss 0.380061 LR 0.001000 Time 0.020765 +2023-10-05 20:49:54,830 - Epoch: [16][ 1100/ 1236] Overall Loss 0.380203 Objective Loss 0.380203 LR 0.001000 Time 0.020758 +2023-10-05 20:49:55,028 - Epoch: [16][ 1110/ 1236] Overall Loss 0.380234 Objective Loss 0.380234 LR 0.001000 Time 0.020750 +2023-10-05 20:49:55,228 - Epoch: [16][ 1120/ 1236] Overall Loss 0.380263 Objective Loss 0.380263 LR 0.001000 Time 0.020743 +2023-10-05 20:49:55,427 - Epoch: [16][ 1130/ 1236] Overall Loss 0.380397 Objective Loss 0.380397 LR 0.001000 Time 0.020735 +2023-10-05 20:49:55,627 - Epoch: [16][ 1140/ 1236] Overall Loss 0.381074 Objective Loss 0.381074 LR 0.001000 Time 0.020728 +2023-10-05 20:49:55,826 - Epoch: [16][ 1150/ 1236] Overall Loss 0.381130 Objective Loss 0.381130 LR 0.001000 Time 0.020721 +2023-10-05 20:49:56,025 - Epoch: [16][ 1160/ 1236] Overall Loss 0.381278 Objective Loss 0.381278 LR 0.001000 Time 0.020714 +2023-10-05 20:49:56,224 - Epoch: [16][ 1170/ 1236] Overall Loss 0.381349 Objective Loss 0.381349 LR 0.001000 Time 0.020706 +2023-10-05 20:49:56,424 - Epoch: [16][ 1180/ 1236] Overall Loss 0.381530 Objective Loss 0.381530 LR 0.001000 Time 0.020700 +2023-10-05 20:49:56,623 - Epoch: [16][ 1190/ 1236] Overall Loss 0.381351 Objective Loss 0.381351 LR 0.001000 Time 0.020693 +2023-10-05 20:49:56,823 - Epoch: [16][ 1200/ 1236] Overall Loss 0.381073 Objective Loss 0.381073 LR 0.001000 Time 0.020687 +2023-10-05 20:49:57,022 - Epoch: [16][ 1210/ 1236] Overall Loss 0.381268 Objective Loss 0.381268 LR 0.001000 Time 0.020680 +2023-10-05 20:49:57,221 - Epoch: [16][ 1220/ 1236] Overall Loss 0.381242 Objective Loss 0.381242 LR 0.001000 Time 0.020674 +2023-10-05 20:49:57,474 - Epoch: [16][ 1230/ 1236] Overall Loss 0.381266 Objective Loss 0.381266 LR 0.001000 Time 0.020711 +2023-10-05 20:49:57,593 - Epoch: [16][ 1236/ 1236] Overall Loss 0.381054 Objective Loss 0.381054 Top1 75.967413 Top5 97.148676 LR 0.001000 Time 0.020706 +2023-10-05 20:49:57,722 - --- validate (epoch=16)----------- +2023-10-05 20:49:57,722 - 29943 samples (256 per mini-batch) +2023-10-05 20:49:58,181 - Epoch: [16][ 10/ 117] Loss 0.371627 Top1 78.593750 Top5 97.265625 +2023-10-05 20:49:58,340 - Epoch: [16][ 20/ 117] Loss 0.396257 Top1 77.285156 Top5 97.128906 +2023-10-05 20:49:58,490 - Epoch: [16][ 30/ 117] Loss 0.404545 Top1 77.187500 Top5 96.992188 +2023-10-05 20:49:58,640 - Epoch: [16][ 40/ 117] Loss 0.403663 Top1 77.167969 Top5 97.001953 +2023-10-05 20:49:58,787 - Epoch: [16][ 50/ 117] Loss 0.404065 Top1 77.187500 Top5 96.953125 +2023-10-05 20:49:58,935 - Epoch: [16][ 60/ 117] Loss 0.400917 Top1 77.063802 Top5 97.031250 +2023-10-05 20:49:59,082 - Epoch: [16][ 70/ 117] Loss 0.397637 Top1 77.014509 Top5 96.986607 +2023-10-05 20:49:59,230 - Epoch: [16][ 80/ 117] Loss 0.394952 Top1 77.182617 Top5 96.918945 +2023-10-05 20:49:59,377 - Epoch: [16][ 90/ 117] Loss 0.392633 Top1 77.178819 Top5 96.909722 +2023-10-05 20:49:59,525 - Epoch: [16][ 100/ 117] Loss 0.395151 Top1 77.242188 Top5 96.945312 +2023-10-05 20:49:59,678 - Epoch: [16][ 110/ 117] Loss 0.392555 Top1 77.350852 Top5 96.942472 +2023-10-05 20:49:59,762 - Epoch: [16][ 117/ 117] Loss 0.390916 Top1 77.393715 Top5 96.934175 +2023-10-05 20:49:59,900 - ==> Top1: 77.394 Top5: 96.934 Loss: 0.391 + +2023-10-05 20:49:59,901 - ==> Confusion: +[[ 936 3 9 2 12 5 0 0 6 48 3 0 0 3 6 1 2 1 4 0 9] + [ 1 1048 1 1 7 16 4 21 0 1 2 0 0 0 1 3 5 0 13 2 5] + [ 8 1 919 15 6 0 38 10 0 0 5 5 4 2 2 3 4 0 16 5 13] + [ 7 3 40 910 0 3 3 1 6 1 15 0 9 4 28 1 1 9 29 1 18] + [ 37 19 1 1 952 5 1 1 2 6 0 0 0 1 6 4 11 1 0 1 1] + [ 8 62 2 1 5 919 3 50 2 1 7 6 4 21 7 0 5 0 1 5 7] + [ 0 2 29 0 0 0 1131 5 0 0 4 2 2 0 0 2 0 0 5 7 2] + [ 7 30 12 0 2 31 7 1027 2 0 2 7 0 0 0 3 0 3 71 8 6] + [ 23 5 0 2 2 1 0 0 954 38 17 3 0 15 15 5 1 2 4 2 0] + [ 181 1 3 0 8 0 1 0 39 838 0 1 0 21 9 2 0 3 1 4 7] + [ 5 5 11 4 2 0 3 6 9 1 973 0 0 8 4 3 2 0 8 0 9] + [ 3 1 0 0 0 14 0 7 0 0 0 918 40 4 0 1 5 12 0 26 4] + [ 0 0 6 8 0 0 0 3 1 1 1 43 952 4 5 5 7 17 2 8 5] + [ 2 0 3 1 9 11 1 0 15 10 13 3 4 1023 5 4 1 1 0 4 9] + [ 19 4 0 11 11 1 0 0 26 2 6 0 2 1 985 0 1 4 14 0 14] + [ 2 3 6 2 1 1 2 0 0 0 0 14 12 0 0 1031 23 14 0 13 10] + [ 2 20 0 0 10 0 2 0 1 0 1 6 1 2 3 9 1093 1 0 7 3] + [ 0 0 0 4 0 0 1 0 1 0 0 22 41 1 1 9 1 952 0 1 4] + [ 1 12 11 10 1 0 2 24 7 1 2 2 0 0 7 0 2 0 974 1 11] + [ 0 5 1 0 2 3 10 11 0 0 2 15 4 2 0 5 9 1 4 1074 4] + [ 286 441 186 66 191 192 81 131 101 92 195 198 473 393 214 85 315 69 262 369 3565]] + +2023-10-05 20:49:59,902 - ==> Best [Top1: 78.850 Top5: 97.218 Sparsity:0.00 Params: 148928 on epoch: 14] +2023-10-05 20:49:59,902 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:49:59,908 - + +2023-10-05 20:49:59,908 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:50:00,885 - Epoch: [17][ 10/ 1236] Overall Loss 0.357491 Objective Loss 0.357491 LR 0.001000 Time 0.097689 +2023-10-05 20:50:01,090 - Epoch: [17][ 20/ 1236] Overall Loss 0.383187 Objective Loss 0.383187 LR 0.001000 Time 0.059057 +2023-10-05 20:50:01,292 - Epoch: [17][ 30/ 1236] Overall Loss 0.378760 Objective Loss 0.378760 LR 0.001000 Time 0.046107 +2023-10-05 20:50:01,497 - Epoch: [17][ 40/ 1236] Overall Loss 0.372874 Objective Loss 0.372874 LR 0.001000 Time 0.039691 +2023-10-05 20:50:01,700 - Epoch: [17][ 50/ 1236] Overall Loss 0.377870 Objective Loss 0.377870 LR 0.001000 Time 0.035800 +2023-10-05 20:50:01,904 - Epoch: [17][ 60/ 1236] Overall Loss 0.372321 Objective Loss 0.372321 LR 0.001000 Time 0.033238 +2023-10-05 20:50:02,107 - Epoch: [17][ 70/ 1236] Overall Loss 0.375964 Objective Loss 0.375964 LR 0.001000 Time 0.031385 +2023-10-05 20:50:02,312 - Epoch: [17][ 80/ 1236] Overall Loss 0.375739 Objective Loss 0.375739 LR 0.001000 Time 0.030016 +2023-10-05 20:50:02,515 - Epoch: [17][ 90/ 1236] Overall Loss 0.375082 Objective Loss 0.375082 LR 0.001000 Time 0.028930 +2023-10-05 20:50:02,719 - Epoch: [17][ 100/ 1236] Overall Loss 0.369142 Objective Loss 0.369142 LR 0.001000 Time 0.028077 +2023-10-05 20:50:02,922 - Epoch: [17][ 110/ 1236] Overall Loss 0.367148 Objective Loss 0.367148 LR 0.001000 Time 0.027369 +2023-10-05 20:50:03,126 - Epoch: [17][ 120/ 1236] Overall Loss 0.368230 Objective Loss 0.368230 LR 0.001000 Time 0.026786 +2023-10-05 20:50:03,330 - Epoch: [17][ 130/ 1236] Overall Loss 0.368204 Objective Loss 0.368204 LR 0.001000 Time 0.026286 +2023-10-05 20:50:03,533 - Epoch: [17][ 140/ 1236] Overall Loss 0.368951 Objective Loss 0.368951 LR 0.001000 Time 0.025861 +2023-10-05 20:50:03,737 - Epoch: [17][ 150/ 1236] Overall Loss 0.369070 Objective Loss 0.369070 LR 0.001000 Time 0.025492 +2023-10-05 20:50:03,941 - Epoch: [17][ 160/ 1236] Overall Loss 0.368081 Objective Loss 0.368081 LR 0.001000 Time 0.025174 +2023-10-05 20:50:04,144 - Epoch: [17][ 170/ 1236] Overall Loss 0.368969 Objective Loss 0.368969 LR 0.001000 Time 0.024887 +2023-10-05 20:50:04,349 - Epoch: [17][ 180/ 1236] Overall Loss 0.367403 Objective Loss 0.367403 LR 0.001000 Time 0.024636 +2023-10-05 20:50:04,552 - Epoch: [17][ 190/ 1236] Overall Loss 0.367796 Objective Loss 0.367796 LR 0.001000 Time 0.024409 +2023-10-05 20:50:04,756 - Epoch: [17][ 200/ 1236] Overall Loss 0.369416 Objective Loss 0.369416 LR 0.001000 Time 0.024208 +2023-10-05 20:50:04,960 - Epoch: [17][ 210/ 1236] Overall Loss 0.370862 Objective Loss 0.370862 LR 0.001000 Time 0.024023 +2023-10-05 20:50:05,164 - Epoch: [17][ 220/ 1236] Overall Loss 0.371048 Objective Loss 0.371048 LR 0.001000 Time 0.023858 +2023-10-05 20:50:05,367 - Epoch: [17][ 230/ 1236] Overall Loss 0.371792 Objective Loss 0.371792 LR 0.001000 Time 0.023703 +2023-10-05 20:50:05,572 - Epoch: [17][ 240/ 1236] Overall Loss 0.372426 Objective Loss 0.372426 LR 0.001000 Time 0.023565 +2023-10-05 20:50:05,775 - Epoch: [17][ 250/ 1236] Overall Loss 0.374194 Objective Loss 0.374194 LR 0.001000 Time 0.023435 +2023-10-05 20:50:05,979 - Epoch: [17][ 260/ 1236] Overall Loss 0.375193 Objective Loss 0.375193 LR 0.001000 Time 0.023318 +2023-10-05 20:50:06,182 - Epoch: [17][ 270/ 1236] Overall Loss 0.374890 Objective Loss 0.374890 LR 0.001000 Time 0.023205 +2023-10-05 20:50:06,387 - Epoch: [17][ 280/ 1236] Overall Loss 0.375110 Objective Loss 0.375110 LR 0.001000 Time 0.023105 +2023-10-05 20:50:06,590 - Epoch: [17][ 290/ 1236] Overall Loss 0.374361 Objective Loss 0.374361 LR 0.001000 Time 0.023008 +2023-10-05 20:50:06,794 - Epoch: [17][ 300/ 1236] Overall Loss 0.373225 Objective Loss 0.373225 LR 0.001000 Time 0.022921 +2023-10-05 20:50:06,998 - Epoch: [17][ 310/ 1236] Overall Loss 0.371796 Objective Loss 0.371796 LR 0.001000 Time 0.022838 +2023-10-05 20:50:07,202 - Epoch: [17][ 320/ 1236] Overall Loss 0.371992 Objective Loss 0.371992 LR 0.001000 Time 0.022762 +2023-10-05 20:50:07,406 - Epoch: [17][ 330/ 1236] Overall Loss 0.372577 Objective Loss 0.372577 LR 0.001000 Time 0.022688 +2023-10-05 20:50:07,610 - Epoch: [17][ 340/ 1236] Overall Loss 0.372926 Objective Loss 0.372926 LR 0.001000 Time 0.022621 +2023-10-05 20:50:07,814 - Epoch: [17][ 350/ 1236] Overall Loss 0.374594 Objective Loss 0.374594 LR 0.001000 Time 0.022555 +2023-10-05 20:50:08,018 - Epoch: [17][ 360/ 1236] Overall Loss 0.374418 Objective Loss 0.374418 LR 0.001000 Time 0.022496 +2023-10-05 20:50:08,222 - Epoch: [17][ 370/ 1236] Overall Loss 0.374134 Objective Loss 0.374134 LR 0.001000 Time 0.022437 +2023-10-05 20:50:08,427 - Epoch: [17][ 380/ 1236] Overall Loss 0.373779 Objective Loss 0.373779 LR 0.001000 Time 0.022384 +2023-10-05 20:50:08,630 - Epoch: [17][ 390/ 1236] Overall Loss 0.372995 Objective Loss 0.372995 LR 0.001000 Time 0.022331 +2023-10-05 20:50:08,835 - Epoch: [17][ 400/ 1236] Overall Loss 0.373086 Objective Loss 0.373086 LR 0.001000 Time 0.022285 +2023-10-05 20:50:09,039 - Epoch: [17][ 410/ 1236] Overall Loss 0.372357 Objective Loss 0.372357 LR 0.001000 Time 0.022237 +2023-10-05 20:50:09,243 - Epoch: [17][ 420/ 1236] Overall Loss 0.372147 Objective Loss 0.372147 LR 0.001000 Time 0.022194 +2023-10-05 20:50:09,447 - Epoch: [17][ 430/ 1236] Overall Loss 0.372832 Objective Loss 0.372832 LR 0.001000 Time 0.022149 +2023-10-05 20:50:09,652 - Epoch: [17][ 440/ 1236] Overall Loss 0.372429 Objective Loss 0.372429 LR 0.001000 Time 0.022112 +2023-10-05 20:50:09,856 - Epoch: [17][ 450/ 1236] Overall Loss 0.372667 Objective Loss 0.372667 LR 0.001000 Time 0.022073 +2023-10-05 20:50:10,061 - Epoch: [17][ 460/ 1236] Overall Loss 0.373342 Objective Loss 0.373342 LR 0.001000 Time 0.022039 +2023-10-05 20:50:10,265 - Epoch: [17][ 470/ 1236] Overall Loss 0.372913 Objective Loss 0.372913 LR 0.001000 Time 0.022003 +2023-10-05 20:50:10,470 - Epoch: [17][ 480/ 1236] Overall Loss 0.372272 Objective Loss 0.372272 LR 0.001000 Time 0.021970 +2023-10-05 20:50:10,673 - Epoch: [17][ 490/ 1236] Overall Loss 0.372600 Objective Loss 0.372600 LR 0.001000 Time 0.021937 +2023-10-05 20:50:10,878 - Epoch: [17][ 500/ 1236] Overall Loss 0.372367 Objective Loss 0.372367 LR 0.001000 Time 0.021907 +2023-10-05 20:50:11,081 - Epoch: [17][ 510/ 1236] Overall Loss 0.372002 Objective Loss 0.372002 LR 0.001000 Time 0.021875 +2023-10-05 20:50:11,285 - Epoch: [17][ 520/ 1236] Overall Loss 0.373088 Objective Loss 0.373088 LR 0.001000 Time 0.021846 +2023-10-05 20:50:11,489 - Epoch: [17][ 530/ 1236] Overall Loss 0.372986 Objective Loss 0.372986 LR 0.001000 Time 0.021818 +2023-10-05 20:50:11,693 - Epoch: [17][ 540/ 1236] Overall Loss 0.372907 Objective Loss 0.372907 LR 0.001000 Time 0.021791 +2023-10-05 20:50:11,895 - Epoch: [17][ 550/ 1236] Overall Loss 0.372197 Objective Loss 0.372197 LR 0.001000 Time 0.021761 +2023-10-05 20:50:12,098 - Epoch: [17][ 560/ 1236] Overall Loss 0.372866 Objective Loss 0.372866 LR 0.001000 Time 0.021735 +2023-10-05 20:50:12,300 - Epoch: [17][ 570/ 1236] Overall Loss 0.372231 Objective Loss 0.372231 LR 0.001000 Time 0.021707 +2023-10-05 20:50:12,503 - Epoch: [17][ 580/ 1236] Overall Loss 0.371940 Objective Loss 0.371940 LR 0.001000 Time 0.021682 +2023-10-05 20:50:12,705 - Epoch: [17][ 590/ 1236] Overall Loss 0.372773 Objective Loss 0.372773 LR 0.001000 Time 0.021656 +2023-10-05 20:50:12,908 - Epoch: [17][ 600/ 1236] Overall Loss 0.372691 Objective Loss 0.372691 LR 0.001000 Time 0.021633 +2023-10-05 20:50:13,110 - Epoch: [17][ 610/ 1236] Overall Loss 0.372500 Objective Loss 0.372500 LR 0.001000 Time 0.021609 +2023-10-05 20:50:13,313 - Epoch: [17][ 620/ 1236] Overall Loss 0.371938 Objective Loss 0.371938 LR 0.001000 Time 0.021587 +2023-10-05 20:50:13,515 - Epoch: [17][ 630/ 1236] Overall Loss 0.371281 Objective Loss 0.371281 LR 0.001000 Time 0.021565 +2023-10-05 20:50:13,718 - Epoch: [17][ 640/ 1236] Overall Loss 0.370881 Objective Loss 0.370881 LR 0.001000 Time 0.021545 +2023-10-05 20:50:13,920 - Epoch: [17][ 650/ 1236] Overall Loss 0.371472 Objective Loss 0.371472 LR 0.001000 Time 0.021524 +2023-10-05 20:50:14,123 - Epoch: [17][ 660/ 1236] Overall Loss 0.371063 Objective Loss 0.371063 LR 0.001000 Time 0.021504 +2023-10-05 20:50:14,325 - Epoch: [17][ 670/ 1236] Overall Loss 0.371654 Objective Loss 0.371654 LR 0.001000 Time 0.021485 +2023-10-05 20:50:14,528 - Epoch: [17][ 680/ 1236] Overall Loss 0.371708 Objective Loss 0.371708 LR 0.001000 Time 0.021467 +2023-10-05 20:50:14,730 - Epoch: [17][ 690/ 1236] Overall Loss 0.371430 Objective Loss 0.371430 LR 0.001000 Time 0.021447 +2023-10-05 20:50:14,933 - Epoch: [17][ 700/ 1236] Overall Loss 0.371542 Objective Loss 0.371542 LR 0.001000 Time 0.021431 +2023-10-05 20:50:15,135 - Epoch: [17][ 710/ 1236] Overall Loss 0.371632 Objective Loss 0.371632 LR 0.001000 Time 0.021413 +2023-10-05 20:50:15,338 - Epoch: [17][ 720/ 1236] Overall Loss 0.371931 Objective Loss 0.371931 LR 0.001000 Time 0.021397 +2023-10-05 20:50:15,540 - Epoch: [17][ 730/ 1236] Overall Loss 0.372067 Objective Loss 0.372067 LR 0.001000 Time 0.021380 +2023-10-05 20:50:15,743 - Epoch: [17][ 740/ 1236] Overall Loss 0.371896 Objective Loss 0.371896 LR 0.001000 Time 0.021365 +2023-10-05 20:50:15,946 - Epoch: [17][ 750/ 1236] Overall Loss 0.372083 Objective Loss 0.372083 LR 0.001000 Time 0.021350 +2023-10-05 20:50:16,148 - Epoch: [17][ 760/ 1236] Overall Loss 0.372203 Objective Loss 0.372203 LR 0.001000 Time 0.021335 +2023-10-05 20:50:16,350 - Epoch: [17][ 770/ 1236] Overall Loss 0.371934 Objective Loss 0.371934 LR 0.001000 Time 0.021320 +2023-10-05 20:50:16,553 - Epoch: [17][ 780/ 1236] Overall Loss 0.372234 Objective Loss 0.372234 LR 0.001000 Time 0.021306 +2023-10-05 20:50:16,755 - Epoch: [17][ 790/ 1236] Overall Loss 0.371930 Objective Loss 0.371930 LR 0.001000 Time 0.021292 +2023-10-05 20:50:16,958 - Epoch: [17][ 800/ 1236] Overall Loss 0.372162 Objective Loss 0.372162 LR 0.001000 Time 0.021279 +2023-10-05 20:50:17,160 - Epoch: [17][ 810/ 1236] Overall Loss 0.372005 Objective Loss 0.372005 LR 0.001000 Time 0.021265 +2023-10-05 20:50:17,363 - Epoch: [17][ 820/ 1236] Overall Loss 0.372103 Objective Loss 0.372103 LR 0.001000 Time 0.021253 +2023-10-05 20:50:17,565 - Epoch: [17][ 830/ 1236] Overall Loss 0.371875 Objective Loss 0.371875 LR 0.001000 Time 0.021240 +2023-10-05 20:50:17,768 - Epoch: [17][ 840/ 1236] Overall Loss 0.371470 Objective Loss 0.371470 LR 0.001000 Time 0.021228 +2023-10-05 20:50:17,970 - Epoch: [17][ 850/ 1236] Overall Loss 0.371278 Objective Loss 0.371278 LR 0.001000 Time 0.021215 +2023-10-05 20:50:18,173 - Epoch: [17][ 860/ 1236] Overall Loss 0.371694 Objective Loss 0.371694 LR 0.001000 Time 0.021204 +2023-10-05 20:50:18,375 - Epoch: [17][ 870/ 1236] Overall Loss 0.372114 Objective Loss 0.372114 LR 0.001000 Time 0.021192 +2023-10-05 20:50:18,578 - Epoch: [17][ 880/ 1236] Overall Loss 0.372105 Objective Loss 0.372105 LR 0.001000 Time 0.021182 +2023-10-05 20:50:18,780 - Epoch: [17][ 890/ 1236] Overall Loss 0.372439 Objective Loss 0.372439 LR 0.001000 Time 0.021171 +2023-10-05 20:50:18,983 - Epoch: [17][ 900/ 1236] Overall Loss 0.372387 Objective Loss 0.372387 LR 0.001000 Time 0.021161 +2023-10-05 20:50:19,185 - Epoch: [17][ 910/ 1236] Overall Loss 0.372422 Objective Loss 0.372422 LR 0.001000 Time 0.021150 +2023-10-05 20:50:19,388 - Epoch: [17][ 920/ 1236] Overall Loss 0.371989 Objective Loss 0.371989 LR 0.001000 Time 0.021140 +2023-10-05 20:50:19,590 - Epoch: [17][ 930/ 1236] Overall Loss 0.372263 Objective Loss 0.372263 LR 0.001000 Time 0.021130 +2023-10-05 20:50:19,793 - Epoch: [17][ 940/ 1236] Overall Loss 0.372292 Objective Loss 0.372292 LR 0.001000 Time 0.021120 +2023-10-05 20:50:19,995 - Epoch: [17][ 950/ 1236] Overall Loss 0.372262 Objective Loss 0.372262 LR 0.001000 Time 0.021110 +2023-10-05 20:50:20,198 - Epoch: [17][ 960/ 1236] Overall Loss 0.372254 Objective Loss 0.372254 LR 0.001000 Time 0.021102 +2023-10-05 20:50:20,400 - Epoch: [17][ 970/ 1236] Overall Loss 0.372227 Objective Loss 0.372227 LR 0.001000 Time 0.021092 +2023-10-05 20:50:20,603 - Epoch: [17][ 980/ 1236] Overall Loss 0.372078 Objective Loss 0.372078 LR 0.001000 Time 0.021083 +2023-10-05 20:50:20,805 - Epoch: [17][ 990/ 1236] Overall Loss 0.372019 Objective Loss 0.372019 LR 0.001000 Time 0.021074 +2023-10-05 20:50:21,008 - Epoch: [17][ 1000/ 1236] Overall Loss 0.371955 Objective Loss 0.371955 LR 0.001000 Time 0.021066 +2023-10-05 20:50:21,210 - Epoch: [17][ 1010/ 1236] Overall Loss 0.371638 Objective Loss 0.371638 LR 0.001000 Time 0.021057 +2023-10-05 20:50:21,414 - Epoch: [17][ 1020/ 1236] Overall Loss 0.371698 Objective Loss 0.371698 LR 0.001000 Time 0.021050 +2023-10-05 20:50:21,615 - Epoch: [17][ 1030/ 1236] Overall Loss 0.371782 Objective Loss 0.371782 LR 0.001000 Time 0.021041 +2023-10-05 20:50:21,819 - Epoch: [17][ 1040/ 1236] Overall Loss 0.372263 Objective Loss 0.372263 LR 0.001000 Time 0.021034 +2023-10-05 20:50:22,020 - Epoch: [17][ 1050/ 1236] Overall Loss 0.371960 Objective Loss 0.371960 LR 0.001000 Time 0.021025 +2023-10-05 20:50:22,224 - Epoch: [17][ 1060/ 1236] Overall Loss 0.371955 Objective Loss 0.371955 LR 0.001000 Time 0.021019 +2023-10-05 20:50:22,426 - Epoch: [17][ 1070/ 1236] Overall Loss 0.371907 Objective Loss 0.371907 LR 0.001000 Time 0.021011 +2023-10-05 20:50:22,629 - Epoch: [17][ 1080/ 1236] Overall Loss 0.372540 Objective Loss 0.372540 LR 0.001000 Time 0.021004 +2023-10-05 20:50:22,831 - Epoch: [17][ 1090/ 1236] Overall Loss 0.372726 Objective Loss 0.372726 LR 0.001000 Time 0.020996 +2023-10-05 20:50:23,034 - Epoch: [17][ 1100/ 1236] Overall Loss 0.372591 Objective Loss 0.372591 LR 0.001000 Time 0.020990 +2023-10-05 20:50:23,236 - Epoch: [17][ 1110/ 1236] Overall Loss 0.372566 Objective Loss 0.372566 LR 0.001000 Time 0.020982 +2023-10-05 20:50:23,439 - Epoch: [17][ 1120/ 1236] Overall Loss 0.372588 Objective Loss 0.372588 LR 0.001000 Time 0.020976 +2023-10-05 20:50:23,641 - Epoch: [17][ 1130/ 1236] Overall Loss 0.372509 Objective Loss 0.372509 LR 0.001000 Time 0.020968 +2023-10-05 20:50:23,844 - Epoch: [17][ 1140/ 1236] Overall Loss 0.372422 Objective Loss 0.372422 LR 0.001000 Time 0.020962 +2023-10-05 20:50:24,047 - Epoch: [17][ 1150/ 1236] Overall Loss 0.372411 Objective Loss 0.372411 LR 0.001000 Time 0.020957 +2023-10-05 20:50:24,252 - Epoch: [17][ 1160/ 1236] Overall Loss 0.372236 Objective Loss 0.372236 LR 0.001000 Time 0.020952 +2023-10-05 20:50:24,456 - Epoch: [17][ 1170/ 1236] Overall Loss 0.372177 Objective Loss 0.372177 LR 0.001000 Time 0.020947 +2023-10-05 20:50:24,661 - Epoch: [17][ 1180/ 1236] Overall Loss 0.372107 Objective Loss 0.372107 LR 0.001000 Time 0.020943 +2023-10-05 20:50:24,864 - Epoch: [17][ 1190/ 1236] Overall Loss 0.372579 Objective Loss 0.372579 LR 0.001000 Time 0.020938 +2023-10-05 20:50:25,070 - Epoch: [17][ 1200/ 1236] Overall Loss 0.372376 Objective Loss 0.372376 LR 0.001000 Time 0.020934 +2023-10-05 20:50:25,273 - Epoch: [17][ 1210/ 1236] Overall Loss 0.372010 Objective Loss 0.372010 LR 0.001000 Time 0.020928 +2023-10-05 20:50:25,478 - Epoch: [17][ 1220/ 1236] Overall Loss 0.372112 Objective Loss 0.372112 LR 0.001000 Time 0.020925 +2023-10-05 20:50:25,734 - Epoch: [17][ 1230/ 1236] Overall Loss 0.372289 Objective Loss 0.372289 LR 0.001000 Time 0.020963 +2023-10-05 20:50:25,853 - Epoch: [17][ 1236/ 1236] Overall Loss 0.372543 Objective Loss 0.372543 Top1 79.022403 Top5 97.352342 LR 0.001000 Time 0.020957 +2023-10-05 20:50:25,989 - --- validate (epoch=17)----------- +2023-10-05 20:50:25,989 - 29943 samples (256 per mini-batch) +2023-10-05 20:50:26,442 - Epoch: [17][ 10/ 117] Loss 0.382205 Top1 80.390625 Top5 97.031250 +2023-10-05 20:50:26,586 - Epoch: [17][ 20/ 117] Loss 0.363599 Top1 80.449219 Top5 97.324219 +2023-10-05 20:50:26,728 - Epoch: [17][ 30/ 117] Loss 0.354231 Top1 80.911458 Top5 97.408854 +2023-10-05 20:50:26,871 - Epoch: [17][ 40/ 117] Loss 0.353014 Top1 80.761719 Top5 97.451172 +2023-10-05 20:50:27,014 - Epoch: [17][ 50/ 117] Loss 0.363085 Top1 80.515625 Top5 97.421875 +2023-10-05 20:50:27,156 - Epoch: [17][ 60/ 117] Loss 0.367360 Top1 80.305990 Top5 97.402344 +2023-10-05 20:50:27,297 - Epoch: [17][ 70/ 117] Loss 0.367646 Top1 80.306920 Top5 97.388393 +2023-10-05 20:50:27,439 - Epoch: [17][ 80/ 117] Loss 0.369691 Top1 80.166016 Top5 97.348633 +2023-10-05 20:50:27,581 - Epoch: [17][ 90/ 117] Loss 0.373874 Top1 80.000000 Top5 97.317708 +2023-10-05 20:50:27,723 - Epoch: [17][ 100/ 117] Loss 0.379854 Top1 79.773438 Top5 97.238281 +2023-10-05 20:50:27,872 - Epoch: [17][ 110/ 117] Loss 0.378400 Top1 79.850852 Top5 97.276278 +2023-10-05 20:50:27,957 - Epoch: [17][ 117/ 117] Loss 0.378072 Top1 79.848379 Top5 97.258124 +2023-10-05 20:50:28,090 - ==> Top1: 79.848 Top5: 97.258 Loss: 0.378 + +2023-10-05 20:50:28,091 - ==> Confusion: +[[ 935 2 5 1 5 3 0 1 6 64 2 0 0 3 6 0 1 3 2 0 11] + [ 3 1065 1 2 5 8 2 22 0 0 1 0 0 0 1 4 4 1 8 1 3] + [ 10 1 952 10 3 0 21 7 0 2 3 1 8 2 3 1 3 3 9 2 15] + [ 6 2 15 937 1 1 2 0 3 0 8 1 11 4 48 5 0 7 27 1 10] + [ 31 18 1 1 962 2 0 0 1 2 2 0 3 1 6 3 6 3 2 0 6] + [ 5 85 0 4 9 923 1 15 2 2 3 12 1 29 5 0 2 4 6 1 7] + [ 0 5 78 1 0 0 1057 17 0 0 5 4 1 0 1 8 1 0 4 4 5] + [ 11 23 14 0 0 25 3 1069 1 2 2 5 1 0 1 2 0 1 49 3 6] + [ 24 3 1 0 0 1 1 0 940 41 13 1 2 15 35 3 0 2 7 0 0] + [ 140 1 1 0 4 7 0 1 26 881 0 3 2 28 8 3 2 1 1 2 8] + [ 2 6 13 6 0 0 0 4 11 0 973 3 0 5 10 0 0 1 9 1 9] + [ 2 2 1 1 1 13 2 3 0 0 0 956 20 4 1 3 1 17 0 5 3] + [ 2 0 3 7 0 1 0 3 0 0 1 48 950 4 3 4 2 22 7 5 6] + [ 1 1 2 1 4 6 0 0 14 13 14 8 2 1026 5 1 0 2 0 2 17] + [ 13 5 1 5 5 0 0 0 12 5 1 0 0 1 1018 0 0 4 18 0 13] + [ 3 3 6 3 5 1 0 0 0 0 1 13 11 2 1 1032 15 20 0 6 12] + [ 0 38 1 1 12 0 0 0 3 0 0 3 2 2 3 5 1069 0 0 8 14] + [ 1 1 1 5 0 0 1 0 0 0 0 3 16 2 1 8 1 992 2 0 4] + [ 3 7 11 14 0 0 0 34 5 0 0 1 0 0 11 0 1 0 973 0 8] + [ 1 6 4 0 3 3 10 23 0 0 3 19 10 4 0 1 6 3 5 1038 13] + [ 246 450 213 90 107 151 42 102 114 82 214 137 468 341 204 66 116 94 273 234 4161]] + +2023-10-05 20:50:28,092 - ==> Best [Top1: 79.848 Top5: 97.258 Sparsity:0.00 Params: 148928 on epoch: 17] +2023-10-05 20:50:28,092 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:50:28,105 - + +2023-10-05 20:50:28,105 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:50:29,225 - Epoch: [18][ 10/ 1236] Overall Loss 0.348354 Objective Loss 0.348354 LR 0.001000 Time 0.111928 +2023-10-05 20:50:29,428 - Epoch: [18][ 20/ 1236] Overall Loss 0.355777 Objective Loss 0.355777 LR 0.001000 Time 0.066072 +2023-10-05 20:50:29,630 - Epoch: [18][ 30/ 1236] Overall Loss 0.356927 Objective Loss 0.356927 LR 0.001000 Time 0.050773 +2023-10-05 20:50:29,832 - Epoch: [18][ 40/ 1236] Overall Loss 0.356034 Objective Loss 0.356034 LR 0.001000 Time 0.043140 +2023-10-05 20:50:30,035 - Epoch: [18][ 50/ 1236] Overall Loss 0.355149 Objective Loss 0.355149 LR 0.001000 Time 0.038565 +2023-10-05 20:50:30,237 - Epoch: [18][ 60/ 1236] Overall Loss 0.360492 Objective Loss 0.360492 LR 0.001000 Time 0.035501 +2023-10-05 20:50:30,440 - Epoch: [18][ 70/ 1236] Overall Loss 0.359230 Objective Loss 0.359230 LR 0.001000 Time 0.033321 +2023-10-05 20:50:30,643 - Epoch: [18][ 80/ 1236] Overall Loss 0.356415 Objective Loss 0.356415 LR 0.001000 Time 0.031681 +2023-10-05 20:50:30,845 - Epoch: [18][ 90/ 1236] Overall Loss 0.358529 Objective Loss 0.358529 LR 0.001000 Time 0.030409 +2023-10-05 20:50:31,047 - Epoch: [18][ 100/ 1236] Overall Loss 0.360451 Objective Loss 0.360451 LR 0.001000 Time 0.029387 +2023-10-05 20:50:31,250 - Epoch: [18][ 110/ 1236] Overall Loss 0.360993 Objective Loss 0.360993 LR 0.001000 Time 0.028557 +2023-10-05 20:50:31,453 - Epoch: [18][ 120/ 1236] Overall Loss 0.363593 Objective Loss 0.363593 LR 0.001000 Time 0.027860 +2023-10-05 20:50:31,656 - Epoch: [18][ 130/ 1236] Overall Loss 0.362295 Objective Loss 0.362295 LR 0.001000 Time 0.027281 +2023-10-05 20:50:31,861 - Epoch: [18][ 140/ 1236] Overall Loss 0.362050 Objective Loss 0.362050 LR 0.001000 Time 0.026794 +2023-10-05 20:50:32,065 - Epoch: [18][ 150/ 1236] Overall Loss 0.361787 Objective Loss 0.361787 LR 0.001000 Time 0.026364 +2023-10-05 20:50:32,266 - Epoch: [18][ 160/ 1236] Overall Loss 0.362173 Objective Loss 0.362173 LR 0.001000 Time 0.025971 +2023-10-05 20:50:32,467 - Epoch: [18][ 170/ 1236] Overall Loss 0.362066 Objective Loss 0.362066 LR 0.001000 Time 0.025623 +2023-10-05 20:50:32,670 - Epoch: [18][ 180/ 1236] Overall Loss 0.362097 Objective Loss 0.362097 LR 0.001000 Time 0.025325 +2023-10-05 20:50:32,871 - Epoch: [18][ 190/ 1236] Overall Loss 0.362468 Objective Loss 0.362468 LR 0.001000 Time 0.025050 +2023-10-05 20:50:33,072 - Epoch: [18][ 200/ 1236] Overall Loss 0.362185 Objective Loss 0.362185 LR 0.001000 Time 0.024797 +2023-10-05 20:50:33,272 - Epoch: [18][ 210/ 1236] Overall Loss 0.362178 Objective Loss 0.362178 LR 0.001000 Time 0.024570 +2023-10-05 20:50:33,471 - Epoch: [18][ 220/ 1236] Overall Loss 0.362704 Objective Loss 0.362704 LR 0.001000 Time 0.024353 +2023-10-05 20:50:33,671 - Epoch: [18][ 230/ 1236] Overall Loss 0.363886 Objective Loss 0.363886 LR 0.001000 Time 0.024163 +2023-10-05 20:50:33,869 - Epoch: [18][ 240/ 1236] Overall Loss 0.362575 Objective Loss 0.362575 LR 0.001000 Time 0.023982 +2023-10-05 20:50:34,070 - Epoch: [18][ 250/ 1236] Overall Loss 0.364434 Objective Loss 0.364434 LR 0.001000 Time 0.023823 +2023-10-05 20:50:34,269 - Epoch: [18][ 260/ 1236] Overall Loss 0.363839 Objective Loss 0.363839 LR 0.001000 Time 0.023670 +2023-10-05 20:50:34,469 - Epoch: [18][ 270/ 1236] Overall Loss 0.363450 Objective Loss 0.363450 LR 0.001000 Time 0.023535 +2023-10-05 20:50:34,667 - Epoch: [18][ 280/ 1236] Overall Loss 0.361660 Objective Loss 0.361660 LR 0.001000 Time 0.023401 +2023-10-05 20:50:34,869 - Epoch: [18][ 290/ 1236] Overall Loss 0.360961 Objective Loss 0.360961 LR 0.001000 Time 0.023286 +2023-10-05 20:50:35,067 - Epoch: [18][ 300/ 1236] Overall Loss 0.361179 Objective Loss 0.361179 LR 0.001000 Time 0.023171 +2023-10-05 20:50:35,268 - Epoch: [18][ 310/ 1236] Overall Loss 0.360297 Objective Loss 0.360297 LR 0.001000 Time 0.023070 +2023-10-05 20:50:35,467 - Epoch: [18][ 320/ 1236] Overall Loss 0.360874 Objective Loss 0.360874 LR 0.001000 Time 0.022970 +2023-10-05 20:50:35,667 - Epoch: [18][ 330/ 1236] Overall Loss 0.361761 Objective Loss 0.361761 LR 0.001000 Time 0.022880 +2023-10-05 20:50:35,866 - Epoch: [18][ 340/ 1236] Overall Loss 0.361136 Objective Loss 0.361136 LR 0.001000 Time 0.022791 +2023-10-05 20:50:36,066 - Epoch: [18][ 350/ 1236] Overall Loss 0.362076 Objective Loss 0.362076 LR 0.001000 Time 0.022711 +2023-10-05 20:50:36,265 - Epoch: [18][ 360/ 1236] Overall Loss 0.361784 Objective Loss 0.361784 LR 0.001000 Time 0.022631 +2023-10-05 20:50:36,466 - Epoch: [18][ 370/ 1236] Overall Loss 0.361264 Objective Loss 0.361264 LR 0.001000 Time 0.022560 +2023-10-05 20:50:36,669 - Epoch: [18][ 380/ 1236] Overall Loss 0.360598 Objective Loss 0.360598 LR 0.001000 Time 0.022501 +2023-10-05 20:50:36,881 - Epoch: [18][ 390/ 1236] Overall Loss 0.360018 Objective Loss 0.360018 LR 0.001000 Time 0.022465 +2023-10-05 20:50:37,088 - Epoch: [18][ 400/ 1236] Overall Loss 0.359964 Objective Loss 0.359964 LR 0.001000 Time 0.022421 +2023-10-05 20:50:37,299 - Epoch: [18][ 410/ 1236] Overall Loss 0.361024 Objective Loss 0.361024 LR 0.001000 Time 0.022388 +2023-10-05 20:50:37,506 - Epoch: [18][ 420/ 1236] Overall Loss 0.360577 Objective Loss 0.360577 LR 0.001000 Time 0.022348 +2023-10-05 20:50:37,717 - Epoch: [18][ 430/ 1236] Overall Loss 0.360427 Objective Loss 0.360427 LR 0.001000 Time 0.022318 +2023-10-05 20:50:37,924 - Epoch: [18][ 440/ 1236] Overall Loss 0.360701 Objective Loss 0.360701 LR 0.001000 Time 0.022281 +2023-10-05 20:50:38,133 - Epoch: [18][ 450/ 1236] Overall Loss 0.360643 Objective Loss 0.360643 LR 0.001000 Time 0.022250 +2023-10-05 20:50:38,334 - Epoch: [18][ 460/ 1236] Overall Loss 0.360437 Objective Loss 0.360437 LR 0.001000 Time 0.022201 +2023-10-05 20:50:38,536 - Epoch: [18][ 470/ 1236] Overall Loss 0.360410 Objective Loss 0.360410 LR 0.001000 Time 0.022157 +2023-10-05 20:50:38,736 - Epoch: [18][ 480/ 1236] Overall Loss 0.360348 Objective Loss 0.360348 LR 0.001000 Time 0.022112 +2023-10-05 20:50:38,938 - Epoch: [18][ 490/ 1236] Overall Loss 0.360693 Objective Loss 0.360693 LR 0.001000 Time 0.022071 +2023-10-05 20:50:39,137 - Epoch: [18][ 500/ 1236] Overall Loss 0.360514 Objective Loss 0.360514 LR 0.001000 Time 0.022029 +2023-10-05 20:50:39,339 - Epoch: [18][ 510/ 1236] Overall Loss 0.360121 Objective Loss 0.360121 LR 0.001000 Time 0.021991 +2023-10-05 20:50:39,539 - Epoch: [18][ 520/ 1236] Overall Loss 0.359546 Objective Loss 0.359546 LR 0.001000 Time 0.021952 +2023-10-05 20:50:39,741 - Epoch: [18][ 530/ 1236] Overall Loss 0.359539 Objective Loss 0.359539 LR 0.001000 Time 0.021918 +2023-10-05 20:50:39,941 - Epoch: [18][ 540/ 1236] Overall Loss 0.359725 Objective Loss 0.359725 LR 0.001000 Time 0.021882 +2023-10-05 20:50:40,142 - Epoch: [18][ 550/ 1236] Overall Loss 0.359656 Objective Loss 0.359656 LR 0.001000 Time 0.021850 +2023-10-05 20:50:40,342 - Epoch: [18][ 560/ 1236] Overall Loss 0.359375 Objective Loss 0.359375 LR 0.001000 Time 0.021815 +2023-10-05 20:50:40,544 - Epoch: [18][ 570/ 1236] Overall Loss 0.359537 Objective Loss 0.359537 LR 0.001000 Time 0.021786 +2023-10-05 20:50:40,744 - Epoch: [18][ 580/ 1236] Overall Loss 0.359534 Objective Loss 0.359534 LR 0.001000 Time 0.021754 +2023-10-05 20:50:40,946 - Epoch: [18][ 590/ 1236] Overall Loss 0.359362 Objective Loss 0.359362 LR 0.001000 Time 0.021728 +2023-10-05 20:50:41,146 - Epoch: [18][ 600/ 1236] Overall Loss 0.359733 Objective Loss 0.359733 LR 0.001000 Time 0.021699 +2023-10-05 20:50:41,348 - Epoch: [18][ 610/ 1236] Overall Loss 0.359721 Objective Loss 0.359721 LR 0.001000 Time 0.021673 +2023-10-05 20:50:41,548 - Epoch: [18][ 620/ 1236] Overall Loss 0.358862 Objective Loss 0.358862 LR 0.001000 Time 0.021646 +2023-10-05 20:50:41,750 - Epoch: [18][ 630/ 1236] Overall Loss 0.359024 Objective Loss 0.359024 LR 0.001000 Time 0.021622 +2023-10-05 20:50:41,950 - Epoch: [18][ 640/ 1236] Overall Loss 0.359212 Objective Loss 0.359212 LR 0.001000 Time 0.021596 +2023-10-05 20:50:42,152 - Epoch: [18][ 650/ 1236] Overall Loss 0.359475 Objective Loss 0.359475 LR 0.001000 Time 0.021574 +2023-10-05 20:50:42,352 - Epoch: [18][ 660/ 1236] Overall Loss 0.359719 Objective Loss 0.359719 LR 0.001000 Time 0.021549 +2023-10-05 20:50:42,554 - Epoch: [18][ 670/ 1236] Overall Loss 0.359694 Objective Loss 0.359694 LR 0.001000 Time 0.021528 +2023-10-05 20:50:42,753 - Epoch: [18][ 680/ 1236] Overall Loss 0.359910 Objective Loss 0.359910 LR 0.001000 Time 0.021505 +2023-10-05 20:50:42,956 - Epoch: [18][ 690/ 1236] Overall Loss 0.359846 Objective Loss 0.359846 LR 0.001000 Time 0.021485 +2023-10-05 20:50:43,155 - Epoch: [18][ 700/ 1236] Overall Loss 0.359941 Objective Loss 0.359941 LR 0.001000 Time 0.021463 +2023-10-05 20:50:43,357 - Epoch: [18][ 710/ 1236] Overall Loss 0.359801 Objective Loss 0.359801 LR 0.001000 Time 0.021445 +2023-10-05 20:50:43,557 - Epoch: [18][ 720/ 1236] Overall Loss 0.360250 Objective Loss 0.360250 LR 0.001000 Time 0.021424 +2023-10-05 20:50:43,759 - Epoch: [18][ 730/ 1236] Overall Loss 0.359569 Objective Loss 0.359569 LR 0.001000 Time 0.021407 +2023-10-05 20:50:43,959 - Epoch: [18][ 740/ 1236] Overall Loss 0.359659 Objective Loss 0.359659 LR 0.001000 Time 0.021387 +2023-10-05 20:50:44,161 - Epoch: [18][ 750/ 1236] Overall Loss 0.360047 Objective Loss 0.360047 LR 0.001000 Time 0.021370 +2023-10-05 20:50:44,361 - Epoch: [18][ 760/ 1236] Overall Loss 0.360558 Objective Loss 0.360558 LR 0.001000 Time 0.021352 +2023-10-05 20:50:44,563 - Epoch: [18][ 770/ 1236] Overall Loss 0.360432 Objective Loss 0.360432 LR 0.001000 Time 0.021336 +2023-10-05 20:50:44,762 - Epoch: [18][ 780/ 1236] Overall Loss 0.360543 Objective Loss 0.360543 LR 0.001000 Time 0.021318 +2023-10-05 20:50:44,964 - Epoch: [18][ 790/ 1236] Overall Loss 0.361115 Objective Loss 0.361115 LR 0.001000 Time 0.021303 +2023-10-05 20:50:45,164 - Epoch: [18][ 800/ 1236] Overall Loss 0.361301 Objective Loss 0.361301 LR 0.001000 Time 0.021286 +2023-10-05 20:50:45,366 - Epoch: [18][ 810/ 1236] Overall Loss 0.361226 Objective Loss 0.361226 LR 0.001000 Time 0.021272 +2023-10-05 20:50:45,566 - Epoch: [18][ 820/ 1236] Overall Loss 0.361151 Objective Loss 0.361151 LR 0.001000 Time 0.021256 +2023-10-05 20:50:45,768 - Epoch: [18][ 830/ 1236] Overall Loss 0.361067 Objective Loss 0.361067 LR 0.001000 Time 0.021243 +2023-10-05 20:50:45,968 - Epoch: [18][ 840/ 1236] Overall Loss 0.361009 Objective Loss 0.361009 LR 0.001000 Time 0.021228 +2023-10-05 20:50:46,170 - Epoch: [18][ 850/ 1236] Overall Loss 0.360681 Objective Loss 0.360681 LR 0.001000 Time 0.021216 +2023-10-05 20:50:46,370 - Epoch: [18][ 860/ 1236] Overall Loss 0.360643 Objective Loss 0.360643 LR 0.001000 Time 0.021201 +2023-10-05 20:50:46,572 - Epoch: [18][ 870/ 1236] Overall Loss 0.360830 Objective Loss 0.360830 LR 0.001000 Time 0.021189 +2023-10-05 20:50:46,772 - Epoch: [18][ 880/ 1236] Overall Loss 0.360994 Objective Loss 0.360994 LR 0.001000 Time 0.021175 +2023-10-05 20:50:46,974 - Epoch: [18][ 890/ 1236] Overall Loss 0.360708 Objective Loss 0.360708 LR 0.001000 Time 0.021163 +2023-10-05 20:50:47,175 - Epoch: [18][ 900/ 1236] Overall Loss 0.360927 Objective Loss 0.360927 LR 0.001000 Time 0.021151 +2023-10-05 20:50:47,377 - Epoch: [18][ 910/ 1236] Overall Loss 0.361065 Objective Loss 0.361065 LR 0.001000 Time 0.021140 +2023-10-05 20:50:47,577 - Epoch: [18][ 920/ 1236] Overall Loss 0.360830 Objective Loss 0.360830 LR 0.001000 Time 0.021127 +2023-10-05 20:50:47,779 - Epoch: [18][ 930/ 1236] Overall Loss 0.360627 Objective Loss 0.360627 LR 0.001000 Time 0.021117 +2023-10-05 20:50:47,979 - Epoch: [18][ 940/ 1236] Overall Loss 0.360695 Objective Loss 0.360695 LR 0.001000 Time 0.021104 +2023-10-05 20:50:48,181 - Epoch: [18][ 950/ 1236] Overall Loss 0.360760 Objective Loss 0.360760 LR 0.001000 Time 0.021094 +2023-10-05 20:50:48,381 - Epoch: [18][ 960/ 1236] Overall Loss 0.360972 Objective Loss 0.360972 LR 0.001000 Time 0.021083 +2023-10-05 20:50:48,583 - Epoch: [18][ 970/ 1236] Overall Loss 0.361345 Objective Loss 0.361345 LR 0.001000 Time 0.021073 +2023-10-05 20:50:48,783 - Epoch: [18][ 980/ 1236] Overall Loss 0.361453 Objective Loss 0.361453 LR 0.001000 Time 0.021062 +2023-10-05 20:50:48,985 - Epoch: [18][ 990/ 1236] Overall Loss 0.361324 Objective Loss 0.361324 LR 0.001000 Time 0.021053 +2023-10-05 20:50:49,185 - Epoch: [18][ 1000/ 1236] Overall Loss 0.361744 Objective Loss 0.361744 LR 0.001000 Time 0.021042 +2023-10-05 20:50:49,387 - Epoch: [18][ 1010/ 1236] Overall Loss 0.361776 Objective Loss 0.361776 LR 0.001000 Time 0.021033 +2023-10-05 20:50:49,587 - Epoch: [18][ 1020/ 1236] Overall Loss 0.361745 Objective Loss 0.361745 LR 0.001000 Time 0.021023 +2023-10-05 20:50:49,789 - Epoch: [18][ 1030/ 1236] Overall Loss 0.361817 Objective Loss 0.361817 LR 0.001000 Time 0.021014 +2023-10-05 20:50:49,989 - Epoch: [18][ 1040/ 1236] Overall Loss 0.362123 Objective Loss 0.362123 LR 0.001000 Time 0.021004 +2023-10-05 20:50:50,191 - Epoch: [18][ 1050/ 1236] Overall Loss 0.362095 Objective Loss 0.362095 LR 0.001000 Time 0.020996 +2023-10-05 20:50:50,391 - Epoch: [18][ 1060/ 1236] Overall Loss 0.362577 Objective Loss 0.362577 LR 0.001000 Time 0.020986 +2023-10-05 20:50:50,593 - Epoch: [18][ 1070/ 1236] Overall Loss 0.362801 Objective Loss 0.362801 LR 0.001000 Time 0.020979 +2023-10-05 20:50:50,793 - Epoch: [18][ 1080/ 1236] Overall Loss 0.362855 Objective Loss 0.362855 LR 0.001000 Time 0.020969 +2023-10-05 20:50:50,995 - Epoch: [18][ 1090/ 1236] Overall Loss 0.362811 Objective Loss 0.362811 LR 0.001000 Time 0.020962 +2023-10-05 20:50:51,196 - Epoch: [18][ 1100/ 1236] Overall Loss 0.362948 Objective Loss 0.362948 LR 0.001000 Time 0.020953 +2023-10-05 20:50:51,398 - Epoch: [18][ 1110/ 1236] Overall Loss 0.362923 Objective Loss 0.362923 LR 0.001000 Time 0.020946 +2023-10-05 20:50:51,598 - Epoch: [18][ 1120/ 1236] Overall Loss 0.362957 Objective Loss 0.362957 LR 0.001000 Time 0.020937 +2023-10-05 20:50:51,800 - Epoch: [18][ 1130/ 1236] Overall Loss 0.362624 Objective Loss 0.362624 LR 0.001000 Time 0.020930 +2023-10-05 20:50:52,000 - Epoch: [18][ 1140/ 1236] Overall Loss 0.362238 Objective Loss 0.362238 LR 0.001000 Time 0.020922 +2023-10-05 20:50:52,202 - Epoch: [18][ 1150/ 1236] Overall Loss 0.362451 Objective Loss 0.362451 LR 0.001000 Time 0.020916 +2023-10-05 20:50:52,402 - Epoch: [18][ 1160/ 1236] Overall Loss 0.362958 Objective Loss 0.362958 LR 0.001000 Time 0.020907 +2023-10-05 20:50:52,604 - Epoch: [18][ 1170/ 1236] Overall Loss 0.363014 Objective Loss 0.363014 LR 0.001000 Time 0.020901 +2023-10-05 20:50:52,805 - Epoch: [18][ 1180/ 1236] Overall Loss 0.362879 Objective Loss 0.362879 LR 0.001000 Time 0.020894 +2023-10-05 20:50:53,007 - Epoch: [18][ 1190/ 1236] Overall Loss 0.363183 Objective Loss 0.363183 LR 0.001000 Time 0.020888 +2023-10-05 20:50:53,207 - Epoch: [18][ 1200/ 1236] Overall Loss 0.363298 Objective Loss 0.363298 LR 0.001000 Time 0.020880 +2023-10-05 20:50:53,410 - Epoch: [18][ 1210/ 1236] Overall Loss 0.363535 Objective Loss 0.363535 LR 0.001000 Time 0.020874 +2023-10-05 20:50:53,609 - Epoch: [18][ 1220/ 1236] Overall Loss 0.363572 Objective Loss 0.363572 LR 0.001000 Time 0.020867 +2023-10-05 20:50:53,863 - Epoch: [18][ 1230/ 1236] Overall Loss 0.363667 Objective Loss 0.363667 LR 0.001000 Time 0.020903 +2023-10-05 20:50:53,981 - Epoch: [18][ 1236/ 1236] Overall Loss 0.363871 Objective Loss 0.363871 Top1 80.040733 Top5 96.130346 LR 0.001000 Time 0.020897 +2023-10-05 20:50:54,115 - --- validate (epoch=18)----------- +2023-10-05 20:50:54,116 - 29943 samples (256 per mini-batch) +2023-10-05 20:50:54,571 - Epoch: [18][ 10/ 117] Loss 0.392466 Top1 79.335938 Top5 97.070312 +2023-10-05 20:50:54,728 - Epoch: [18][ 20/ 117] Loss 0.380779 Top1 79.238281 Top5 97.148438 +2023-10-05 20:50:54,882 - Epoch: [18][ 30/ 117] Loss 0.369721 Top1 79.466146 Top5 97.070312 +2023-10-05 20:50:55,034 - Epoch: [18][ 40/ 117] Loss 0.370402 Top1 79.326172 Top5 97.099609 +2023-10-05 20:50:55,188 - Epoch: [18][ 50/ 117] Loss 0.374188 Top1 78.953125 Top5 96.968750 +2023-10-05 20:50:55,344 - Epoch: [18][ 60/ 117] Loss 0.370948 Top1 78.893229 Top5 96.875000 +2023-10-05 20:50:55,494 - Epoch: [18][ 70/ 117] Loss 0.372415 Top1 78.911830 Top5 96.880580 +2023-10-05 20:50:55,650 - Epoch: [18][ 80/ 117] Loss 0.370090 Top1 79.028320 Top5 96.943359 +2023-10-05 20:50:55,807 - Epoch: [18][ 90/ 117] Loss 0.370106 Top1 78.971354 Top5 96.974826 +2023-10-05 20:50:55,968 - Epoch: [18][ 100/ 117] Loss 0.369335 Top1 78.863281 Top5 96.968750 +2023-10-05 20:50:56,129 - Epoch: [18][ 110/ 117] Loss 0.370785 Top1 78.799716 Top5 96.967330 +2023-10-05 20:50:56,214 - Epoch: [18][ 117/ 117] Loss 0.371533 Top1 78.806399 Top5 96.944194 +2023-10-05 20:50:56,346 - ==> Top1: 78.806 Top5: 96.944 Loss: 0.372 + +2023-10-05 20:50:56,347 - ==> Confusion: +[[ 934 2 8 2 5 4 0 2 8 59 0 0 2 6 5 3 2 0 1 1 6] + [ 2 1013 1 1 2 40 4 38 1 0 2 1 0 0 0 3 5 0 8 2 8] + [ 8 0 896 21 3 0 62 15 0 2 3 4 9 2 2 4 3 0 10 4 8] + [ 3 3 18 965 0 5 3 1 2 0 5 0 15 1 20 4 0 7 25 1 11] + [ 36 16 3 0 943 10 1 1 0 3 0 2 2 1 11 3 13 1 1 0 3] + [ 8 26 2 1 3 982 2 35 2 3 2 9 3 14 4 0 4 0 6 6 4] + [ 0 4 21 0 0 1 1124 10 0 0 3 1 1 0 0 7 1 1 2 11 4] + [ 6 11 6 0 0 23 7 1083 0 3 5 7 3 0 0 2 0 0 42 13 7] + [ 24 2 0 0 0 2 0 1 941 47 17 2 2 16 20 4 2 1 6 2 0] + [ 130 0 2 0 3 1 0 2 35 902 0 0 2 17 9 4 2 0 0 2 8] + [ 7 0 10 8 2 1 3 5 15 2 964 2 0 10 5 2 0 0 9 0 8] + [ 0 0 1 1 2 10 2 2 0 0 0 913 59 4 1 2 2 18 0 14 4] + [ 2 0 7 7 0 1 1 4 0 1 2 28 983 1 1 4 1 10 2 8 5] + [ 4 0 3 0 3 11 0 1 22 14 8 8 4 1023 6 3 1 0 0 2 6] + [ 14 2 2 15 2 1 0 1 19 3 4 1 3 0 1002 0 1 2 21 0 8] + [ 3 3 4 3 3 0 1 0 0 0 1 8 15 2 0 1038 21 17 0 7 8] + [ 0 16 1 1 9 8 1 0 2 0 0 4 6 2 3 9 1081 0 1 9 8] + [ 3 0 0 6 0 1 1 0 1 0 0 2 38 1 0 7 1 970 1 1 5] + [ 2 3 9 12 0 0 0 35 1 0 1 1 6 0 13 0 3 0 973 3 6] + [ 1 1 2 1 1 8 11 20 0 0 1 18 8 3 0 5 6 0 2 1058 6] + [ 215 268 201 125 105 271 78 169 118 92 204 116 531 377 243 71 284 59 233 336 3809]] + +2023-10-05 20:50:56,348 - ==> Best [Top1: 79.848 Top5: 97.258 Sparsity:0.00 Params: 148928 on epoch: 17] +2023-10-05 20:50:56,348 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:50:56,354 - + +2023-10-05 20:50:56,354 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:50:57,339 - Epoch: [19][ 10/ 1236] Overall Loss 0.373172 Objective Loss 0.373172 LR 0.001000 Time 0.098439 +2023-10-05 20:50:57,540 - Epoch: [19][ 20/ 1236] Overall Loss 0.353476 Objective Loss 0.353476 LR 0.001000 Time 0.059240 +2023-10-05 20:50:57,737 - Epoch: [19][ 30/ 1236] Overall Loss 0.360149 Objective Loss 0.360149 LR 0.001000 Time 0.046080 +2023-10-05 20:50:57,938 - Epoch: [19][ 40/ 1236] Overall Loss 0.363102 Objective Loss 0.363102 LR 0.001000 Time 0.039570 +2023-10-05 20:50:58,136 - Epoch: [19][ 50/ 1236] Overall Loss 0.357557 Objective Loss 0.357557 LR 0.001000 Time 0.035615 +2023-10-05 20:50:58,337 - Epoch: [19][ 60/ 1236] Overall Loss 0.363139 Objective Loss 0.363139 LR 0.001000 Time 0.033018 +2023-10-05 20:50:58,536 - Epoch: [19][ 70/ 1236] Overall Loss 0.363392 Objective Loss 0.363392 LR 0.001000 Time 0.031132 +2023-10-05 20:50:58,736 - Epoch: [19][ 80/ 1236] Overall Loss 0.363309 Objective Loss 0.363309 LR 0.001000 Time 0.029741 +2023-10-05 20:50:58,935 - Epoch: [19][ 90/ 1236] Overall Loss 0.367762 Objective Loss 0.367762 LR 0.001000 Time 0.028637 +2023-10-05 20:50:59,134 - Epoch: [19][ 100/ 1236] Overall Loss 0.368614 Objective Loss 0.368614 LR 0.001000 Time 0.027766 +2023-10-05 20:50:59,332 - Epoch: [19][ 110/ 1236] Overall Loss 0.370634 Objective Loss 0.370634 LR 0.001000 Time 0.027035 +2023-10-05 20:50:59,532 - Epoch: [19][ 120/ 1236] Overall Loss 0.370641 Objective Loss 0.370641 LR 0.001000 Time 0.026448 +2023-10-05 20:50:59,731 - Epoch: [19][ 130/ 1236] Overall Loss 0.369284 Objective Loss 0.369284 LR 0.001000 Time 0.025941 +2023-10-05 20:50:59,932 - Epoch: [19][ 140/ 1236] Overall Loss 0.367901 Objective Loss 0.367901 LR 0.001000 Time 0.025518 +2023-10-05 20:51:00,131 - Epoch: [19][ 150/ 1236] Overall Loss 0.366760 Objective Loss 0.366760 LR 0.001000 Time 0.025144 +2023-10-05 20:51:00,331 - Epoch: [19][ 160/ 1236] Overall Loss 0.363771 Objective Loss 0.363771 LR 0.001000 Time 0.024823 +2023-10-05 20:51:00,531 - Epoch: [19][ 170/ 1236] Overall Loss 0.364459 Objective Loss 0.364459 LR 0.001000 Time 0.024532 +2023-10-05 20:51:00,731 - Epoch: [19][ 180/ 1236] Overall Loss 0.364242 Objective Loss 0.364242 LR 0.001000 Time 0.024278 +2023-10-05 20:51:00,929 - Epoch: [19][ 190/ 1236] Overall Loss 0.365888 Objective Loss 0.365888 LR 0.001000 Time 0.024041 +2023-10-05 20:51:01,129 - Epoch: [19][ 200/ 1236] Overall Loss 0.365708 Objective Loss 0.365708 LR 0.001000 Time 0.023838 +2023-10-05 20:51:01,327 - Epoch: [19][ 210/ 1236] Overall Loss 0.365887 Objective Loss 0.365887 LR 0.001000 Time 0.023647 +2023-10-05 20:51:01,527 - Epoch: [19][ 220/ 1236] Overall Loss 0.366409 Objective Loss 0.366409 LR 0.001000 Time 0.023476 +2023-10-05 20:51:01,725 - Epoch: [19][ 230/ 1236] Overall Loss 0.366607 Objective Loss 0.366607 LR 0.001000 Time 0.023317 +2023-10-05 20:51:01,926 - Epoch: [19][ 240/ 1236] Overall Loss 0.367310 Objective Loss 0.367310 LR 0.001000 Time 0.023181 +2023-10-05 20:51:02,125 - Epoch: [19][ 250/ 1236] Overall Loss 0.367530 Objective Loss 0.367530 LR 0.001000 Time 0.023046 +2023-10-05 20:51:02,325 - Epoch: [19][ 260/ 1236] Overall Loss 0.366813 Objective Loss 0.366813 LR 0.001000 Time 0.022930 +2023-10-05 20:51:02,524 - Epoch: [19][ 270/ 1236] Overall Loss 0.366670 Objective Loss 0.366670 LR 0.001000 Time 0.022816 +2023-10-05 20:51:02,725 - Epoch: [19][ 280/ 1236] Overall Loss 0.366229 Objective Loss 0.366229 LR 0.001000 Time 0.022716 +2023-10-05 20:51:02,924 - Epoch: [19][ 290/ 1236] Overall Loss 0.366567 Objective Loss 0.366567 LR 0.001000 Time 0.022617 +2023-10-05 20:51:03,125 - Epoch: [19][ 300/ 1236] Overall Loss 0.367094 Objective Loss 0.367094 LR 0.001000 Time 0.022533 +2023-10-05 20:51:03,324 - Epoch: [19][ 310/ 1236] Overall Loss 0.366786 Objective Loss 0.366786 LR 0.001000 Time 0.022447 +2023-10-05 20:51:03,524 - Epoch: [19][ 320/ 1236] Overall Loss 0.366410 Objective Loss 0.366410 LR 0.001000 Time 0.022369 +2023-10-05 20:51:03,723 - Epoch: [19][ 330/ 1236] Overall Loss 0.366626 Objective Loss 0.366626 LR 0.001000 Time 0.022295 +2023-10-05 20:51:03,924 - Epoch: [19][ 340/ 1236] Overall Loss 0.367473 Objective Loss 0.367473 LR 0.001000 Time 0.022229 +2023-10-05 20:51:04,123 - Epoch: [19][ 350/ 1236] Overall Loss 0.366576 Objective Loss 0.366576 LR 0.001000 Time 0.022161 +2023-10-05 20:51:04,324 - Epoch: [19][ 360/ 1236] Overall Loss 0.366777 Objective Loss 0.366777 LR 0.001000 Time 0.022103 +2023-10-05 20:51:04,523 - Epoch: [19][ 370/ 1236] Overall Loss 0.366460 Objective Loss 0.366460 LR 0.001000 Time 0.022041 +2023-10-05 20:51:04,724 - Epoch: [19][ 380/ 1236] Overall Loss 0.365875 Objective Loss 0.365875 LR 0.001000 Time 0.021989 +2023-10-05 20:51:04,923 - Epoch: [19][ 390/ 1236] Overall Loss 0.365721 Objective Loss 0.365721 LR 0.001000 Time 0.021934 +2023-10-05 20:51:05,123 - Epoch: [19][ 400/ 1236] Overall Loss 0.364415 Objective Loss 0.364415 LR 0.001000 Time 0.021887 +2023-10-05 20:51:05,322 - Epoch: [19][ 410/ 1236] Overall Loss 0.363994 Objective Loss 0.363994 LR 0.001000 Time 0.021838 +2023-10-05 20:51:05,523 - Epoch: [19][ 420/ 1236] Overall Loss 0.363786 Objective Loss 0.363786 LR 0.001000 Time 0.021795 +2023-10-05 20:51:05,722 - Epoch: [19][ 430/ 1236] Overall Loss 0.363622 Objective Loss 0.363622 LR 0.001000 Time 0.021749 +2023-10-05 20:51:05,923 - Epoch: [19][ 440/ 1236] Overall Loss 0.362900 Objective Loss 0.362900 LR 0.001000 Time 0.021710 +2023-10-05 20:51:06,122 - Epoch: [19][ 450/ 1236] Overall Loss 0.363439 Objective Loss 0.363439 LR 0.001000 Time 0.021669 +2023-10-05 20:51:06,322 - Epoch: [19][ 460/ 1236] Overall Loss 0.363885 Objective Loss 0.363885 LR 0.001000 Time 0.021633 +2023-10-05 20:51:06,521 - Epoch: [19][ 470/ 1236] Overall Loss 0.362967 Objective Loss 0.362967 LR 0.001000 Time 0.021595 +2023-10-05 20:51:06,722 - Epoch: [19][ 480/ 1236] Overall Loss 0.362955 Objective Loss 0.362955 LR 0.001000 Time 0.021563 +2023-10-05 20:51:06,921 - Epoch: [19][ 490/ 1236] Overall Loss 0.362103 Objective Loss 0.362103 LR 0.001000 Time 0.021528 +2023-10-05 20:51:07,122 - Epoch: [19][ 500/ 1236] Overall Loss 0.362888 Objective Loss 0.362888 LR 0.001000 Time 0.021499 +2023-10-05 20:51:07,321 - Epoch: [19][ 510/ 1236] Overall Loss 0.362196 Objective Loss 0.362196 LR 0.001000 Time 0.021467 +2023-10-05 20:51:07,521 - Epoch: [19][ 520/ 1236] Overall Loss 0.362039 Objective Loss 0.362039 LR 0.001000 Time 0.021439 +2023-10-05 20:51:07,720 - Epoch: [19][ 530/ 1236] Overall Loss 0.362252 Objective Loss 0.362252 LR 0.001000 Time 0.021409 +2023-10-05 20:51:07,921 - Epoch: [19][ 540/ 1236] Overall Loss 0.362353 Objective Loss 0.362353 LR 0.001000 Time 0.021384 +2023-10-05 20:51:08,120 - Epoch: [19][ 550/ 1236] Overall Loss 0.362397 Objective Loss 0.362397 LR 0.001000 Time 0.021357 +2023-10-05 20:51:08,321 - Epoch: [19][ 560/ 1236] Overall Loss 0.362378 Objective Loss 0.362378 LR 0.001000 Time 0.021334 +2023-10-05 20:51:08,520 - Epoch: [19][ 570/ 1236] Overall Loss 0.362267 Objective Loss 0.362267 LR 0.001000 Time 0.021308 +2023-10-05 20:51:08,721 - Epoch: [19][ 580/ 1236] Overall Loss 0.362570 Objective Loss 0.362570 LR 0.001000 Time 0.021286 +2023-10-05 20:51:08,921 - Epoch: [19][ 590/ 1236] Overall Loss 0.362206 Objective Loss 0.362206 LR 0.001000 Time 0.021262 +2023-10-05 20:51:09,122 - Epoch: [19][ 600/ 1236] Overall Loss 0.362290 Objective Loss 0.362290 LR 0.001000 Time 0.021243 +2023-10-05 20:51:09,321 - Epoch: [19][ 610/ 1236] Overall Loss 0.362416 Objective Loss 0.362416 LR 0.001000 Time 0.021220 +2023-10-05 20:51:09,522 - Epoch: [19][ 620/ 1236] Overall Loss 0.362264 Objective Loss 0.362264 LR 0.001000 Time 0.021201 +2023-10-05 20:51:09,720 - Epoch: [19][ 630/ 1236] Overall Loss 0.362165 Objective Loss 0.362165 LR 0.001000 Time 0.021180 +2023-10-05 20:51:09,921 - Epoch: [19][ 640/ 1236] Overall Loss 0.362674 Objective Loss 0.362674 LR 0.001000 Time 0.021162 +2023-10-05 20:51:10,120 - Epoch: [19][ 650/ 1236] Overall Loss 0.362414 Objective Loss 0.362414 LR 0.001000 Time 0.021141 +2023-10-05 20:51:10,321 - Epoch: [19][ 660/ 1236] Overall Loss 0.362426 Objective Loss 0.362426 LR 0.001000 Time 0.021125 +2023-10-05 20:51:10,520 - Epoch: [19][ 670/ 1236] Overall Loss 0.362213 Objective Loss 0.362213 LR 0.001000 Time 0.021107 +2023-10-05 20:51:10,721 - Epoch: [19][ 680/ 1236] Overall Loss 0.361985 Objective Loss 0.361985 LR 0.001000 Time 0.021091 +2023-10-05 20:51:10,920 - Epoch: [19][ 690/ 1236] Overall Loss 0.362251 Objective Loss 0.362251 LR 0.001000 Time 0.021074 +2023-10-05 20:51:11,121 - Epoch: [19][ 700/ 1236] Overall Loss 0.362211 Objective Loss 0.362211 LR 0.001000 Time 0.021059 +2023-10-05 20:51:11,320 - Epoch: [19][ 710/ 1236] Overall Loss 0.362131 Objective Loss 0.362131 LR 0.001000 Time 0.021042 +2023-10-05 20:51:11,521 - Epoch: [19][ 720/ 1236] Overall Loss 0.362357 Objective Loss 0.362357 LR 0.001000 Time 0.021029 +2023-10-05 20:51:11,720 - Epoch: [19][ 730/ 1236] Overall Loss 0.362011 Objective Loss 0.362011 LR 0.001000 Time 0.021013 +2023-10-05 20:51:11,921 - Epoch: [19][ 740/ 1236] Overall Loss 0.362083 Objective Loss 0.362083 LR 0.001000 Time 0.021000 +2023-10-05 20:51:12,121 - Epoch: [19][ 750/ 1236] Overall Loss 0.362188 Objective Loss 0.362188 LR 0.001000 Time 0.020986 +2023-10-05 20:51:12,321 - Epoch: [19][ 760/ 1236] Overall Loss 0.362155 Objective Loss 0.362155 LR 0.001000 Time 0.020973 +2023-10-05 20:51:12,521 - Epoch: [19][ 770/ 1236] Overall Loss 0.361768 Objective Loss 0.361768 LR 0.001000 Time 0.020959 +2023-10-05 20:51:12,721 - Epoch: [19][ 780/ 1236] Overall Loss 0.362047 Objective Loss 0.362047 LR 0.001000 Time 0.020946 +2023-10-05 20:51:12,920 - Epoch: [19][ 790/ 1236] Overall Loss 0.361745 Objective Loss 0.361745 LR 0.001000 Time 0.020932 +2023-10-05 20:51:13,120 - Epoch: [19][ 800/ 1236] Overall Loss 0.361740 Objective Loss 0.361740 LR 0.001000 Time 0.020921 +2023-10-05 20:51:13,319 - Epoch: [19][ 810/ 1236] Overall Loss 0.361527 Objective Loss 0.361527 LR 0.001000 Time 0.020907 +2023-10-05 20:51:13,519 - Epoch: [19][ 820/ 1236] Overall Loss 0.361709 Objective Loss 0.361709 LR 0.001000 Time 0.020896 +2023-10-05 20:51:13,718 - Epoch: [19][ 830/ 1236] Overall Loss 0.362160 Objective Loss 0.362160 LR 0.001000 Time 0.020884 +2023-10-05 20:51:13,919 - Epoch: [19][ 840/ 1236] Overall Loss 0.362531 Objective Loss 0.362531 LR 0.001000 Time 0.020874 +2023-10-05 20:51:14,119 - Epoch: [19][ 850/ 1236] Overall Loss 0.362474 Objective Loss 0.362474 LR 0.001000 Time 0.020863 +2023-10-05 20:51:14,319 - Epoch: [19][ 860/ 1236] Overall Loss 0.362521 Objective Loss 0.362521 LR 0.001000 Time 0.020853 +2023-10-05 20:51:14,519 - Epoch: [19][ 870/ 1236] Overall Loss 0.362734 Objective Loss 0.362734 LR 0.001000 Time 0.020842 +2023-10-05 20:51:14,719 - Epoch: [19][ 880/ 1236] Overall Loss 0.362527 Objective Loss 0.362527 LR 0.001000 Time 0.020833 +2023-10-05 20:51:14,919 - Epoch: [19][ 890/ 1236] Overall Loss 0.362169 Objective Loss 0.362169 LR 0.001000 Time 0.020822 +2023-10-05 20:51:15,119 - Epoch: [19][ 900/ 1236] Overall Loss 0.362106 Objective Loss 0.362106 LR 0.001000 Time 0.020813 +2023-10-05 20:51:15,319 - Epoch: [19][ 910/ 1236] Overall Loss 0.362115 Objective Loss 0.362115 LR 0.001000 Time 0.020803 +2023-10-05 20:51:15,520 - Epoch: [19][ 920/ 1236] Overall Loss 0.362102 Objective Loss 0.362102 LR 0.001000 Time 0.020796 +2023-10-05 20:51:15,720 - Epoch: [19][ 930/ 1236] Overall Loss 0.362098 Objective Loss 0.362098 LR 0.001000 Time 0.020787 +2023-10-05 20:51:15,921 - Epoch: [19][ 940/ 1236] Overall Loss 0.361870 Objective Loss 0.361870 LR 0.001000 Time 0.020779 +2023-10-05 20:51:16,120 - Epoch: [19][ 950/ 1236] Overall Loss 0.361742 Objective Loss 0.361742 LR 0.001000 Time 0.020769 +2023-10-05 20:51:16,321 - Epoch: [19][ 960/ 1236] Overall Loss 0.361585 Objective Loss 0.361585 LR 0.001000 Time 0.020762 +2023-10-05 20:51:16,520 - Epoch: [19][ 970/ 1236] Overall Loss 0.361924 Objective Loss 0.361924 LR 0.001000 Time 0.020753 +2023-10-05 20:51:16,721 - Epoch: [19][ 980/ 1236] Overall Loss 0.361453 Objective Loss 0.361453 LR 0.001000 Time 0.020746 +2023-10-05 20:51:16,921 - Epoch: [19][ 990/ 1236] Overall Loss 0.361521 Objective Loss 0.361521 LR 0.001000 Time 0.020738 +2023-10-05 20:51:17,121 - Epoch: [19][ 1000/ 1236] Overall Loss 0.361770 Objective Loss 0.361770 LR 0.001000 Time 0.020730 +2023-10-05 20:51:17,321 - Epoch: [19][ 1010/ 1236] Overall Loss 0.361997 Objective Loss 0.361997 LR 0.001000 Time 0.020722 +2023-10-05 20:51:17,521 - Epoch: [19][ 1020/ 1236] Overall Loss 0.361698 Objective Loss 0.361698 LR 0.001000 Time 0.020716 +2023-10-05 20:51:17,721 - Epoch: [19][ 1030/ 1236] Overall Loss 0.362015 Objective Loss 0.362015 LR 0.001000 Time 0.020708 +2023-10-05 20:51:17,921 - Epoch: [19][ 1040/ 1236] Overall Loss 0.361745 Objective Loss 0.361745 LR 0.001000 Time 0.020701 +2023-10-05 20:51:18,121 - Epoch: [19][ 1050/ 1236] Overall Loss 0.361304 Objective Loss 0.361304 LR 0.001000 Time 0.020694 +2023-10-05 20:51:18,321 - Epoch: [19][ 1060/ 1236] Overall Loss 0.361150 Objective Loss 0.361150 LR 0.001000 Time 0.020687 +2023-10-05 20:51:18,519 - Epoch: [19][ 1070/ 1236] Overall Loss 0.361153 Objective Loss 0.361153 LR 0.001000 Time 0.020679 +2023-10-05 20:51:18,720 - Epoch: [19][ 1080/ 1236] Overall Loss 0.361147 Objective Loss 0.361147 LR 0.001000 Time 0.020673 +2023-10-05 20:51:18,920 - Epoch: [19][ 1090/ 1236] Overall Loss 0.361219 Objective Loss 0.361219 LR 0.001000 Time 0.020666 +2023-10-05 20:51:19,121 - Epoch: [19][ 1100/ 1236] Overall Loss 0.361158 Objective Loss 0.361158 LR 0.001000 Time 0.020660 +2023-10-05 20:51:19,320 - Epoch: [19][ 1110/ 1236] Overall Loss 0.361411 Objective Loss 0.361411 LR 0.001000 Time 0.020654 +2023-10-05 20:51:19,521 - Epoch: [19][ 1120/ 1236] Overall Loss 0.361778 Objective Loss 0.361778 LR 0.001000 Time 0.020648 +2023-10-05 20:51:19,720 - Epoch: [19][ 1130/ 1236] Overall Loss 0.361935 Objective Loss 0.361935 LR 0.001000 Time 0.020642 +2023-10-05 20:51:19,921 - Epoch: [19][ 1140/ 1236] Overall Loss 0.361876 Objective Loss 0.361876 LR 0.001000 Time 0.020636 +2023-10-05 20:51:20,120 - Epoch: [19][ 1150/ 1236] Overall Loss 0.361768 Objective Loss 0.361768 LR 0.001000 Time 0.020630 +2023-10-05 20:51:20,320 - Epoch: [19][ 1160/ 1236] Overall Loss 0.361803 Objective Loss 0.361803 LR 0.001000 Time 0.020624 +2023-10-05 20:51:20,519 - Epoch: [19][ 1170/ 1236] Overall Loss 0.361596 Objective Loss 0.361596 LR 0.001000 Time 0.020617 +2023-10-05 20:51:20,720 - Epoch: [19][ 1180/ 1236] Overall Loss 0.361679 Objective Loss 0.361679 LR 0.001000 Time 0.020612 +2023-10-05 20:51:20,919 - Epoch: [19][ 1190/ 1236] Overall Loss 0.361713 Objective Loss 0.361713 LR 0.001000 Time 0.020606 +2023-10-05 20:51:21,120 - Epoch: [19][ 1200/ 1236] Overall Loss 0.361688 Objective Loss 0.361688 LR 0.001000 Time 0.020601 +2023-10-05 20:51:21,319 - Epoch: [19][ 1210/ 1236] Overall Loss 0.361553 Objective Loss 0.361553 LR 0.001000 Time 0.020595 +2023-10-05 20:51:21,520 - Epoch: [19][ 1220/ 1236] Overall Loss 0.361881 Objective Loss 0.361881 LR 0.001000 Time 0.020591 +2023-10-05 20:51:21,772 - Epoch: [19][ 1230/ 1236] Overall Loss 0.361916 Objective Loss 0.361916 LR 0.001000 Time 0.020628 +2023-10-05 20:51:21,889 - Epoch: [19][ 1236/ 1236] Overall Loss 0.361988 Objective Loss 0.361988 Top1 80.040733 Top5 96.334012 LR 0.001000 Time 0.020623 +2023-10-05 20:51:22,020 - --- validate (epoch=19)----------- +2023-10-05 20:51:22,021 - 29943 samples (256 per mini-batch) +2023-10-05 20:51:22,472 - Epoch: [19][ 10/ 117] Loss 0.398803 Top1 78.437500 Top5 97.304688 +2023-10-05 20:51:22,623 - Epoch: [19][ 20/ 117] Loss 0.371890 Top1 79.335938 Top5 97.324219 +2023-10-05 20:51:22,771 - Epoch: [19][ 30/ 117] Loss 0.378598 Top1 79.335938 Top5 97.239583 +2023-10-05 20:51:22,922 - Epoch: [19][ 40/ 117] Loss 0.377768 Top1 79.785156 Top5 97.285156 +2023-10-05 20:51:23,070 - Epoch: [19][ 50/ 117] Loss 0.379665 Top1 79.789062 Top5 97.218750 +2023-10-05 20:51:23,220 - Epoch: [19][ 60/ 117] Loss 0.377201 Top1 79.830729 Top5 97.239583 +2023-10-05 20:51:23,375 - Epoch: [19][ 70/ 117] Loss 0.375201 Top1 80.044643 Top5 97.304688 +2023-10-05 20:51:23,530 - Epoch: [19][ 80/ 117] Loss 0.376106 Top1 80.024414 Top5 97.329102 +2023-10-05 20:51:23,685 - Epoch: [19][ 90/ 117] Loss 0.376091 Top1 79.991319 Top5 97.369792 +2023-10-05 20:51:23,838 - Epoch: [19][ 100/ 117] Loss 0.374164 Top1 80.105469 Top5 97.371094 +2023-10-05 20:51:23,998 - Epoch: [19][ 110/ 117] Loss 0.374267 Top1 80.081676 Top5 97.368608 +2023-10-05 20:51:24,083 - Epoch: [19][ 117/ 117] Loss 0.375946 Top1 80.075477 Top5 97.288181 +2023-10-05 20:51:24,216 - ==> Top1: 80.075 Top5: 97.288 Loss: 0.376 + +2023-10-05 20:51:24,217 - ==> Confusion: +[[ 907 3 1 0 8 2 0 0 15 76 0 0 1 3 6 3 9 1 0 1 14] + [ 3 1014 3 2 10 33 2 23 1 0 6 1 0 0 1 4 8 0 10 3 7] + [ 8 0 942 7 2 0 39 5 0 1 5 3 8 2 2 6 1 2 8 3 12] + [ 6 1 32 909 0 3 4 0 13 0 11 0 17 1 36 5 2 15 20 3 11] + [ 37 10 2 1 955 5 0 0 2 8 0 0 0 1 7 6 10 1 1 1 3] + [ 8 37 0 2 10 952 2 14 5 8 5 11 0 25 6 2 4 0 3 9 13] + [ 0 5 25 1 0 0 1114 7 0 0 6 4 1 0 0 11 0 2 2 11 2] + [ 8 16 16 0 0 37 3 1026 2 5 7 8 1 0 0 2 0 1 68 12 6] + [ 18 1 0 0 0 0 1 0 973 47 9 1 2 7 14 3 2 1 6 4 0] + [ 117 0 2 0 5 1 2 0 31 926 1 1 0 16 2 3 2 0 2 4 4] + [ 4 2 10 4 3 0 2 4 20 2 969 1 0 8 4 5 1 1 6 1 6] + [ 2 0 1 0 2 11 0 0 0 1 0 934 34 5 0 2 2 21 0 17 3] + [ 0 1 3 5 1 0 0 0 4 0 0 40 955 1 1 8 1 30 0 7 11] + [ 2 0 1 0 9 2 0 0 22 16 13 1 4 1021 5 2 0 2 0 3 16] + [ 19 2 2 9 4 0 0 0 47 9 3 0 2 1 966 0 1 9 12 0 15] + [ 3 0 1 0 4 1 0 0 1 0 0 8 7 3 0 1060 12 17 1 9 7] + [ 2 14 1 3 7 5 0 0 2 0 1 1 3 1 5 9 1087 0 0 8 12] + [ 1 0 0 0 0 0 0 0 2 1 0 7 18 0 0 6 1 1000 0 0 2] + [ 4 8 9 14 0 1 0 16 5 1 4 2 7 0 11 1 2 0 973 0 10] + [ 0 0 5 0 1 6 9 13 0 0 4 18 5 1 0 5 8 2 2 1065 8] + [ 223 201 190 80 114 199 92 78 191 135 219 137 432 294 161 139 189 137 179 286 4229]] + +2023-10-05 20:51:24,218 - ==> Best [Top1: 80.075 Top5: 97.288 Sparsity:0.00 Params: 148928 on epoch: 19] +2023-10-05 20:51:24,218 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:51:24,232 - + +2023-10-05 20:51:24,232 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:51:25,210 - Epoch: [20][ 10/ 1236] Overall Loss 0.363354 Objective Loss 0.363354 LR 0.001000 Time 0.097736 +2023-10-05 20:51:25,413 - Epoch: [20][ 20/ 1236] Overall Loss 0.347617 Objective Loss 0.347617 LR 0.001000 Time 0.059022 +2023-10-05 20:51:25,614 - Epoch: [20][ 30/ 1236] Overall Loss 0.342190 Objective Loss 0.342190 LR 0.001000 Time 0.046021 +2023-10-05 20:51:25,817 - Epoch: [20][ 40/ 1236] Overall Loss 0.351360 Objective Loss 0.351360 LR 0.001000 Time 0.039592 +2023-10-05 20:51:26,018 - Epoch: [20][ 50/ 1236] Overall Loss 0.352313 Objective Loss 0.352313 LR 0.001000 Time 0.035679 +2023-10-05 20:51:26,221 - Epoch: [20][ 60/ 1236] Overall Loss 0.358023 Objective Loss 0.358023 LR 0.001000 Time 0.033115 +2023-10-05 20:51:26,421 - Epoch: [20][ 70/ 1236] Overall Loss 0.356071 Objective Loss 0.356071 LR 0.001000 Time 0.031243 +2023-10-05 20:51:26,625 - Epoch: [20][ 80/ 1236] Overall Loss 0.354118 Objective Loss 0.354118 LR 0.001000 Time 0.029872 +2023-10-05 20:51:26,825 - Epoch: [20][ 90/ 1236] Overall Loss 0.355698 Objective Loss 0.355698 LR 0.001000 Time 0.028780 +2023-10-05 20:51:27,026 - Epoch: [20][ 100/ 1236] Overall Loss 0.355965 Objective Loss 0.355965 LR 0.001000 Time 0.027901 +2023-10-05 20:51:27,224 - Epoch: [20][ 110/ 1236] Overall Loss 0.352481 Objective Loss 0.352481 LR 0.001000 Time 0.027165 +2023-10-05 20:51:27,424 - Epoch: [20][ 120/ 1236] Overall Loss 0.353038 Objective Loss 0.353038 LR 0.001000 Time 0.026568 +2023-10-05 20:51:27,624 - Epoch: [20][ 130/ 1236] Overall Loss 0.350104 Objective Loss 0.350104 LR 0.001000 Time 0.026059 +2023-10-05 20:51:27,824 - Epoch: [20][ 140/ 1236] Overall Loss 0.349691 Objective Loss 0.349691 LR 0.001000 Time 0.025622 +2023-10-05 20:51:28,023 - Epoch: [20][ 150/ 1236] Overall Loss 0.349876 Objective Loss 0.349876 LR 0.001000 Time 0.025237 +2023-10-05 20:51:28,224 - Epoch: [20][ 160/ 1236] Overall Loss 0.350996 Objective Loss 0.350996 LR 0.001000 Time 0.024912 +2023-10-05 20:51:28,422 - Epoch: [20][ 170/ 1236] Overall Loss 0.351596 Objective Loss 0.351596 LR 0.001000 Time 0.024613 +2023-10-05 20:51:28,621 - Epoch: [20][ 180/ 1236] Overall Loss 0.352320 Objective Loss 0.352320 LR 0.001000 Time 0.024348 +2023-10-05 20:51:28,819 - Epoch: [20][ 190/ 1236] Overall Loss 0.353587 Objective Loss 0.353587 LR 0.001000 Time 0.024109 +2023-10-05 20:51:29,021 - Epoch: [20][ 200/ 1236] Overall Loss 0.353806 Objective Loss 0.353806 LR 0.001000 Time 0.023912 +2023-10-05 20:51:29,224 - Epoch: [20][ 210/ 1236] Overall Loss 0.352713 Objective Loss 0.352713 LR 0.001000 Time 0.023735 +2023-10-05 20:51:29,423 - Epoch: [20][ 220/ 1236] Overall Loss 0.352130 Objective Loss 0.352130 LR 0.001000 Time 0.023561 +2023-10-05 20:51:29,624 - Epoch: [20][ 230/ 1236] Overall Loss 0.353800 Objective Loss 0.353800 LR 0.001000 Time 0.023407 +2023-10-05 20:51:29,823 - Epoch: [20][ 240/ 1236] Overall Loss 0.353090 Objective Loss 0.353090 LR 0.001000 Time 0.023261 +2023-10-05 20:51:30,024 - Epoch: [20][ 250/ 1236] Overall Loss 0.352705 Objective Loss 0.352705 LR 0.001000 Time 0.023131 +2023-10-05 20:51:30,223 - Epoch: [20][ 260/ 1236] Overall Loss 0.352914 Objective Loss 0.352914 LR 0.001000 Time 0.023008 +2023-10-05 20:51:30,424 - Epoch: [20][ 270/ 1236] Overall Loss 0.352574 Objective Loss 0.352574 LR 0.001000 Time 0.022897 +2023-10-05 20:51:30,623 - Epoch: [20][ 280/ 1236] Overall Loss 0.353125 Objective Loss 0.353125 LR 0.001000 Time 0.022791 +2023-10-05 20:51:30,824 - Epoch: [20][ 290/ 1236] Overall Loss 0.352776 Objective Loss 0.352776 LR 0.001000 Time 0.022696 +2023-10-05 20:51:31,024 - Epoch: [20][ 300/ 1236] Overall Loss 0.352711 Objective Loss 0.352711 LR 0.001000 Time 0.022605 +2023-10-05 20:51:31,225 - Epoch: [20][ 310/ 1236] Overall Loss 0.352851 Objective Loss 0.352851 LR 0.001000 Time 0.022523 +2023-10-05 20:51:31,425 - Epoch: [20][ 320/ 1236] Overall Loss 0.352791 Objective Loss 0.352791 LR 0.001000 Time 0.022443 +2023-10-05 20:51:31,625 - Epoch: [20][ 330/ 1236] Overall Loss 0.352328 Objective Loss 0.352328 LR 0.001000 Time 0.022370 +2023-10-05 20:51:31,826 - Epoch: [20][ 340/ 1236] Overall Loss 0.351189 Objective Loss 0.351189 LR 0.001000 Time 0.022300 +2023-10-05 20:51:32,026 - Epoch: [20][ 350/ 1236] Overall Loss 0.351959 Objective Loss 0.351959 LR 0.001000 Time 0.022233 +2023-10-05 20:51:32,226 - Epoch: [20][ 360/ 1236] Overall Loss 0.351657 Objective Loss 0.351657 LR 0.001000 Time 0.022171 +2023-10-05 20:51:32,428 - Epoch: [20][ 370/ 1236] Overall Loss 0.352594 Objective Loss 0.352594 LR 0.001000 Time 0.022117 +2023-10-05 20:51:32,630 - Epoch: [20][ 380/ 1236] Overall Loss 0.353140 Objective Loss 0.353140 LR 0.001000 Time 0.022065 +2023-10-05 20:51:32,833 - Epoch: [20][ 390/ 1236] Overall Loss 0.352814 Objective Loss 0.352814 LR 0.001000 Time 0.022020 +2023-10-05 20:51:33,035 - Epoch: [20][ 400/ 1236] Overall Loss 0.352700 Objective Loss 0.352700 LR 0.001000 Time 0.021972 +2023-10-05 20:51:33,238 - Epoch: [20][ 410/ 1236] Overall Loss 0.354022 Objective Loss 0.354022 LR 0.001000 Time 0.021932 +2023-10-05 20:51:33,440 - Epoch: [20][ 420/ 1236] Overall Loss 0.354099 Objective Loss 0.354099 LR 0.001000 Time 0.021889 +2023-10-05 20:51:33,643 - Epoch: [20][ 430/ 1236] Overall Loss 0.354420 Objective Loss 0.354420 LR 0.001000 Time 0.021852 +2023-10-05 20:51:33,845 - Epoch: [20][ 440/ 1236] Overall Loss 0.353595 Objective Loss 0.353595 LR 0.001000 Time 0.021813 +2023-10-05 20:51:34,048 - Epoch: [20][ 450/ 1236] Overall Loss 0.353907 Objective Loss 0.353907 LR 0.001000 Time 0.021780 +2023-10-05 20:51:34,250 - Epoch: [20][ 460/ 1236] Overall Loss 0.353699 Objective Loss 0.353699 LR 0.001000 Time 0.021744 +2023-10-05 20:51:34,453 - Epoch: [20][ 470/ 1236] Overall Loss 0.353168 Objective Loss 0.353168 LR 0.001000 Time 0.021713 +2023-10-05 20:51:34,655 - Epoch: [20][ 480/ 1236] Overall Loss 0.353011 Objective Loss 0.353011 LR 0.001000 Time 0.021680 +2023-10-05 20:51:34,858 - Epoch: [20][ 490/ 1236] Overall Loss 0.353043 Objective Loss 0.353043 LR 0.001000 Time 0.021652 +2023-10-05 20:51:35,060 - Epoch: [20][ 500/ 1236] Overall Loss 0.353258 Objective Loss 0.353258 LR 0.001000 Time 0.021622 +2023-10-05 20:51:35,263 - Epoch: [20][ 510/ 1236] Overall Loss 0.353220 Objective Loss 0.353220 LR 0.001000 Time 0.021596 +2023-10-05 20:51:35,465 - Epoch: [20][ 520/ 1236] Overall Loss 0.353861 Objective Loss 0.353861 LR 0.001000 Time 0.021568 +2023-10-05 20:51:35,668 - Epoch: [20][ 530/ 1236] Overall Loss 0.353095 Objective Loss 0.353095 LR 0.001000 Time 0.021544 +2023-10-05 20:51:35,870 - Epoch: [20][ 540/ 1236] Overall Loss 0.352691 Objective Loss 0.352691 LR 0.001000 Time 0.021518 +2023-10-05 20:51:36,074 - Epoch: [20][ 550/ 1236] Overall Loss 0.353010 Objective Loss 0.353010 LR 0.001000 Time 0.021498 +2023-10-05 20:51:36,276 - Epoch: [20][ 560/ 1236] Overall Loss 0.352679 Objective Loss 0.352679 LR 0.001000 Time 0.021474 +2023-10-05 20:51:36,479 - Epoch: [20][ 570/ 1236] Overall Loss 0.352300 Objective Loss 0.352300 LR 0.001000 Time 0.021453 +2023-10-05 20:51:36,682 - Epoch: [20][ 580/ 1236] Overall Loss 0.352378 Objective Loss 0.352378 LR 0.001000 Time 0.021431 +2023-10-05 20:51:36,885 - Epoch: [20][ 590/ 1236] Overall Loss 0.352935 Objective Loss 0.352935 LR 0.001000 Time 0.021411 +2023-10-05 20:51:37,086 - Epoch: [20][ 600/ 1236] Overall Loss 0.353534 Objective Loss 0.353534 LR 0.001000 Time 0.021390 +2023-10-05 20:51:37,289 - Epoch: [20][ 610/ 1236] Overall Loss 0.353964 Objective Loss 0.353964 LR 0.001000 Time 0.021372 +2023-10-05 20:51:37,491 - Epoch: [20][ 620/ 1236] Overall Loss 0.353661 Objective Loss 0.353661 LR 0.001000 Time 0.021352 +2023-10-05 20:51:37,695 - Epoch: [20][ 630/ 1236] Overall Loss 0.353714 Objective Loss 0.353714 LR 0.001000 Time 0.021335 +2023-10-05 20:51:37,896 - Epoch: [20][ 640/ 1236] Overall Loss 0.353888 Objective Loss 0.353888 LR 0.001000 Time 0.021317 +2023-10-05 20:51:38,100 - Epoch: [20][ 650/ 1236] Overall Loss 0.354443 Objective Loss 0.354443 LR 0.001000 Time 0.021302 +2023-10-05 20:51:38,302 - Epoch: [20][ 660/ 1236] Overall Loss 0.354956 Objective Loss 0.354956 LR 0.001000 Time 0.021284 +2023-10-05 20:51:38,505 - Epoch: [20][ 670/ 1236] Overall Loss 0.354909 Objective Loss 0.354909 LR 0.001000 Time 0.021269 +2023-10-05 20:51:38,707 - Epoch: [20][ 680/ 1236] Overall Loss 0.355132 Objective Loss 0.355132 LR 0.001000 Time 0.021252 +2023-10-05 20:51:38,910 - Epoch: [20][ 690/ 1236] Overall Loss 0.355980 Objective Loss 0.355980 LR 0.001000 Time 0.021239 +2023-10-05 20:51:39,112 - Epoch: [20][ 700/ 1236] Overall Loss 0.356290 Objective Loss 0.356290 LR 0.001000 Time 0.021223 +2023-10-05 20:51:39,315 - Epoch: [20][ 710/ 1236] Overall Loss 0.356034 Objective Loss 0.356034 LR 0.001000 Time 0.021210 +2023-10-05 20:51:39,517 - Epoch: [20][ 720/ 1236] Overall Loss 0.356290 Objective Loss 0.356290 LR 0.001000 Time 0.021195 +2023-10-05 20:51:39,720 - Epoch: [20][ 730/ 1236] Overall Loss 0.356592 Objective Loss 0.356592 LR 0.001000 Time 0.021183 +2023-10-05 20:51:39,922 - Epoch: [20][ 740/ 1236] Overall Loss 0.357004 Objective Loss 0.357004 LR 0.001000 Time 0.021168 +2023-10-05 20:51:40,125 - Epoch: [20][ 750/ 1236] Overall Loss 0.357707 Objective Loss 0.357707 LR 0.001000 Time 0.021157 +2023-10-05 20:51:40,333 - Epoch: [20][ 760/ 1236] Overall Loss 0.358319 Objective Loss 0.358319 LR 0.001000 Time 0.021151 +2023-10-05 20:51:40,546 - Epoch: [20][ 770/ 1236] Overall Loss 0.358807 Objective Loss 0.358807 LR 0.001000 Time 0.021153 +2023-10-05 20:51:40,754 - Epoch: [20][ 780/ 1236] Overall Loss 0.359063 Objective Loss 0.359063 LR 0.001000 Time 0.021147 +2023-10-05 20:51:40,967 - Epoch: [20][ 790/ 1236] Overall Loss 0.359046 Objective Loss 0.359046 LR 0.001000 Time 0.021149 +2023-10-05 20:51:41,174 - Epoch: [20][ 800/ 1236] Overall Loss 0.359125 Objective Loss 0.359125 LR 0.001000 Time 0.021144 +2023-10-05 20:51:41,388 - Epoch: [20][ 810/ 1236] Overall Loss 0.358841 Objective Loss 0.358841 LR 0.001000 Time 0.021145 +2023-10-05 20:51:41,595 - Epoch: [20][ 820/ 1236] Overall Loss 0.359085 Objective Loss 0.359085 LR 0.001000 Time 0.021141 +2023-10-05 20:51:41,808 - Epoch: [20][ 830/ 1236] Overall Loss 0.359572 Objective Loss 0.359572 LR 0.001000 Time 0.021142 +2023-10-05 20:51:42,017 - Epoch: [20][ 840/ 1236] Overall Loss 0.359347 Objective Loss 0.359347 LR 0.001000 Time 0.021138 +2023-10-05 20:51:42,230 - Epoch: [20][ 850/ 1236] Overall Loss 0.359651 Objective Loss 0.359651 LR 0.001000 Time 0.021140 +2023-10-05 20:51:42,438 - Epoch: [20][ 860/ 1236] Overall Loss 0.359570 Objective Loss 0.359570 LR 0.001000 Time 0.021135 +2023-10-05 20:51:42,651 - Epoch: [20][ 870/ 1236] Overall Loss 0.359231 Objective Loss 0.359231 LR 0.001000 Time 0.021137 +2023-10-05 20:51:42,859 - Epoch: [20][ 880/ 1236] Overall Loss 0.358943 Objective Loss 0.358943 LR 0.001000 Time 0.021133 +2023-10-05 20:51:43,072 - Epoch: [20][ 890/ 1236] Overall Loss 0.359246 Objective Loss 0.359246 LR 0.001000 Time 0.021134 +2023-10-05 20:51:43,280 - Epoch: [20][ 900/ 1236] Overall Loss 0.359046 Objective Loss 0.359046 LR 0.001000 Time 0.021131 +2023-10-05 20:51:43,493 - Epoch: [20][ 910/ 1236] Overall Loss 0.359453 Objective Loss 0.359453 LR 0.001000 Time 0.021132 +2023-10-05 20:51:43,701 - Epoch: [20][ 920/ 1236] Overall Loss 0.359439 Objective Loss 0.359439 LR 0.001000 Time 0.021128 +2023-10-05 20:51:43,914 - Epoch: [20][ 930/ 1236] Overall Loss 0.359492 Objective Loss 0.359492 LR 0.001000 Time 0.021130 +2023-10-05 20:51:44,122 - Epoch: [20][ 940/ 1236] Overall Loss 0.359728 Objective Loss 0.359728 LR 0.001000 Time 0.021125 +2023-10-05 20:51:44,334 - Epoch: [20][ 950/ 1236] Overall Loss 0.359818 Objective Loss 0.359818 LR 0.001000 Time 0.021126 +2023-10-05 20:51:44,543 - Epoch: [20][ 960/ 1236] Overall Loss 0.359892 Objective Loss 0.359892 LR 0.001000 Time 0.021123 +2023-10-05 20:51:44,756 - Epoch: [20][ 970/ 1236] Overall Loss 0.359838 Objective Loss 0.359838 LR 0.001000 Time 0.021125 +2023-10-05 20:51:44,964 - Epoch: [20][ 980/ 1236] Overall Loss 0.359999 Objective Loss 0.359999 LR 0.001000 Time 0.021121 +2023-10-05 20:51:45,177 - Epoch: [20][ 990/ 1236] Overall Loss 0.360494 Objective Loss 0.360494 LR 0.001000 Time 0.021122 +2023-10-05 20:51:45,386 - Epoch: [20][ 1000/ 1236] Overall Loss 0.360999 Objective Loss 0.360999 LR 0.001000 Time 0.021119 +2023-10-05 20:51:45,598 - Epoch: [20][ 1010/ 1236] Overall Loss 0.360963 Objective Loss 0.360963 LR 0.001000 Time 0.021121 +2023-10-05 20:51:45,806 - Epoch: [20][ 1020/ 1236] Overall Loss 0.361037 Objective Loss 0.361037 LR 0.001000 Time 0.021117 +2023-10-05 20:51:46,019 - Epoch: [20][ 1030/ 1236] Overall Loss 0.361376 Objective Loss 0.361376 LR 0.001000 Time 0.021118 +2023-10-05 20:51:46,226 - Epoch: [20][ 1040/ 1236] Overall Loss 0.361547 Objective Loss 0.361547 LR 0.001000 Time 0.021114 +2023-10-05 20:51:46,439 - Epoch: [20][ 1050/ 1236] Overall Loss 0.361884 Objective Loss 0.361884 LR 0.001000 Time 0.021115 +2023-10-05 20:51:46,647 - Epoch: [20][ 1060/ 1236] Overall Loss 0.362151 Objective Loss 0.362151 LR 0.001000 Time 0.021112 +2023-10-05 20:51:46,859 - Epoch: [20][ 1070/ 1236] Overall Loss 0.362385 Objective Loss 0.362385 LR 0.001000 Time 0.021113 +2023-10-05 20:51:47,068 - Epoch: [20][ 1080/ 1236] Overall Loss 0.362054 Objective Loss 0.362054 LR 0.001000 Time 0.021111 +2023-10-05 20:51:47,281 - Epoch: [20][ 1090/ 1236] Overall Loss 0.362054 Objective Loss 0.362054 LR 0.001000 Time 0.021112 +2023-10-05 20:51:47,489 - Epoch: [20][ 1100/ 1236] Overall Loss 0.362085 Objective Loss 0.362085 LR 0.001000 Time 0.021109 +2023-10-05 20:51:47,702 - Epoch: [20][ 1110/ 1236] Overall Loss 0.362059 Objective Loss 0.362059 LR 0.001000 Time 0.021110 +2023-10-05 20:51:47,910 - Epoch: [20][ 1120/ 1236] Overall Loss 0.361965 Objective Loss 0.361965 LR 0.001000 Time 0.021107 +2023-10-05 20:51:48,123 - Epoch: [20][ 1130/ 1236] Overall Loss 0.362090 Objective Loss 0.362090 LR 0.001000 Time 0.021108 +2023-10-05 20:51:48,331 - Epoch: [20][ 1140/ 1236] Overall Loss 0.362350 Objective Loss 0.362350 LR 0.001000 Time 0.021105 +2023-10-05 20:51:48,543 - Epoch: [20][ 1150/ 1236] Overall Loss 0.362192 Objective Loss 0.362192 LR 0.001000 Time 0.021106 +2023-10-05 20:51:48,752 - Epoch: [20][ 1160/ 1236] Overall Loss 0.362267 Objective Loss 0.362267 LR 0.001000 Time 0.021104 +2023-10-05 20:51:48,964 - Epoch: [20][ 1170/ 1236] Overall Loss 0.362186 Objective Loss 0.362186 LR 0.001000 Time 0.021105 +2023-10-05 20:51:49,173 - Epoch: [20][ 1180/ 1236] Overall Loss 0.362231 Objective Loss 0.362231 LR 0.001000 Time 0.021102 +2023-10-05 20:51:49,385 - Epoch: [20][ 1190/ 1236] Overall Loss 0.362043 Objective Loss 0.362043 LR 0.001000 Time 0.021103 +2023-10-05 20:51:49,593 - Epoch: [20][ 1200/ 1236] Overall Loss 0.361937 Objective Loss 0.361937 LR 0.001000 Time 0.021101 +2023-10-05 20:51:49,806 - Epoch: [20][ 1210/ 1236] Overall Loss 0.362003 Objective Loss 0.362003 LR 0.001000 Time 0.021102 +2023-10-05 20:51:50,014 - Epoch: [20][ 1220/ 1236] Overall Loss 0.361881 Objective Loss 0.361881 LR 0.001000 Time 0.021099 +2023-10-05 20:51:50,273 - Epoch: [20][ 1230/ 1236] Overall Loss 0.362065 Objective Loss 0.362065 LR 0.001000 Time 0.021137 +2023-10-05 20:51:50,390 - Epoch: [20][ 1236/ 1236] Overall Loss 0.362190 Objective Loss 0.362190 Top1 83.095723 Top5 97.556008 LR 0.001000 Time 0.021130 +2023-10-05 20:51:50,504 - --- validate (epoch=20)----------- +2023-10-05 20:51:50,504 - 29943 samples (256 per mini-batch) +2023-10-05 20:51:50,958 - Epoch: [20][ 10/ 117] Loss 0.391900 Top1 80.468750 Top5 97.148438 +2023-10-05 20:51:51,109 - Epoch: [20][ 20/ 117] Loss 0.374163 Top1 79.902344 Top5 97.070312 +2023-10-05 20:51:51,257 - Epoch: [20][ 30/ 117] Loss 0.376798 Top1 80.078125 Top5 97.369792 +2023-10-05 20:51:51,407 - Epoch: [20][ 40/ 117] Loss 0.379200 Top1 80.048828 Top5 97.246094 +2023-10-05 20:51:51,555 - Epoch: [20][ 50/ 117] Loss 0.373247 Top1 80.375000 Top5 97.343750 +2023-10-05 20:51:51,704 - Epoch: [20][ 60/ 117] Loss 0.376563 Top1 80.305990 Top5 97.324219 +2023-10-05 20:51:51,851 - Epoch: [20][ 70/ 117] Loss 0.374179 Top1 80.357143 Top5 97.332589 +2023-10-05 20:51:52,001 - Epoch: [20][ 80/ 117] Loss 0.376412 Top1 80.185547 Top5 97.280273 +2023-10-05 20:51:52,148 - Epoch: [20][ 90/ 117] Loss 0.376295 Top1 80.247396 Top5 97.343750 +2023-10-05 20:51:52,297 - Epoch: [20][ 100/ 117] Loss 0.380858 Top1 80.066406 Top5 97.265625 +2023-10-05 20:51:52,452 - Epoch: [20][ 110/ 117] Loss 0.379320 Top1 80.163352 Top5 97.265625 +2023-10-05 20:51:52,537 - Epoch: [20][ 117/ 117] Loss 0.381538 Top1 80.115553 Top5 97.221387 +2023-10-05 20:51:52,683 - ==> Top1: 80.116 Top5: 97.221 Loss: 0.382 + +2023-10-05 20:51:52,684 - ==> Confusion: +[[ 883 5 5 0 14 3 0 0 7 98 3 0 1 5 3 4 3 2 2 1 11] + [ 1 1022 0 0 9 34 3 31 2 1 4 0 0 1 1 5 3 0 9 2 3] + [ 2 2 938 8 2 0 46 13 0 1 5 2 4 5 2 4 4 1 4 3 10] + [ 6 6 26 940 1 1 7 1 6 0 12 0 9 1 28 4 0 6 14 2 19] + [ 28 9 1 0 956 4 1 1 2 8 0 1 0 2 8 6 14 1 0 1 7] + [ 4 29 1 0 5 975 7 33 2 3 2 5 7 15 10 1 4 1 1 3 8] + [ 0 8 26 0 1 1 1111 8 0 0 3 2 2 0 1 13 0 3 0 7 5] + [ 2 22 13 1 0 33 10 1070 0 1 4 7 1 0 1 2 1 0 29 12 9] + [ 15 5 0 0 0 2 1 0 952 58 9 1 1 17 14 3 2 1 5 0 3] + [ 86 2 1 0 2 7 1 1 37 928 1 0 0 32 1 5 3 2 1 2 7] + [ 4 4 11 4 1 0 7 6 12 1 964 1 0 16 6 3 1 0 5 2 5] + [ 2 0 1 0 2 19 2 0 0 0 0 908 52 6 0 6 1 21 0 8 7] + [ 0 0 5 3 2 1 0 3 1 0 2 31 960 4 4 12 3 22 1 0 14] + [ 1 0 0 0 2 15 0 1 16 15 9 7 2 1029 5 5 1 0 0 3 8] + [ 13 3 3 11 5 0 0 0 37 13 2 0 2 0 991 0 1 2 7 0 11] + [ 0 3 1 3 5 0 1 0 0 0 1 5 7 2 0 1065 17 12 0 8 4] + [ 1 16 1 0 5 8 0 0 4 0 0 3 1 2 3 15 1081 1 0 4 16] + [ 1 0 0 6 0 0 1 0 1 1 0 4 18 0 7 10 1 981 1 1 5] + [ 0 13 6 22 1 2 3 45 4 0 4 1 1 0 10 0 0 0 942 4 10] + [ 0 4 4 1 2 7 16 11 0 0 2 24 6 4 0 6 7 0 1 1048 9] + [ 172 270 177 55 161 213 81 123 126 151 231 137 440 377 188 109 149 84 133 283 4245]] + +2023-10-05 20:51:52,685 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:51:52,685 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:51:52,698 - + +2023-10-05 20:51:52,699 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:51:53,676 - Epoch: [21][ 10/ 1236] Overall Loss 0.347893 Objective Loss 0.347893 LR 0.001000 Time 0.097738 +2023-10-05 20:51:53,879 - Epoch: [21][ 20/ 1236] Overall Loss 0.357204 Objective Loss 0.357204 LR 0.001000 Time 0.058957 +2023-10-05 20:51:54,079 - Epoch: [21][ 30/ 1236] Overall Loss 0.354500 Objective Loss 0.354500 LR 0.001000 Time 0.045987 +2023-10-05 20:51:54,282 - Epoch: [21][ 40/ 1236] Overall Loss 0.353404 Objective Loss 0.353404 LR 0.001000 Time 0.039536 +2023-10-05 20:51:54,482 - Epoch: [21][ 50/ 1236] Overall Loss 0.345924 Objective Loss 0.345924 LR 0.001000 Time 0.035639 +2023-10-05 20:51:54,684 - Epoch: [21][ 60/ 1236] Overall Loss 0.341566 Objective Loss 0.341566 LR 0.001000 Time 0.033058 +2023-10-05 20:51:54,885 - Epoch: [21][ 70/ 1236] Overall Loss 0.345164 Objective Loss 0.345164 LR 0.001000 Time 0.031202 +2023-10-05 20:51:55,088 - Epoch: [21][ 80/ 1236] Overall Loss 0.342858 Objective Loss 0.342858 LR 0.001000 Time 0.029834 +2023-10-05 20:51:55,291 - Epoch: [21][ 90/ 1236] Overall Loss 0.343550 Objective Loss 0.343550 LR 0.001000 Time 0.028771 +2023-10-05 20:51:55,495 - Epoch: [21][ 100/ 1236] Overall Loss 0.345192 Objective Loss 0.345192 LR 0.001000 Time 0.027923 +2023-10-05 20:51:55,695 - Epoch: [21][ 110/ 1236] Overall Loss 0.346871 Objective Loss 0.346871 LR 0.001000 Time 0.027208 +2023-10-05 20:51:55,898 - Epoch: [21][ 120/ 1236] Overall Loss 0.347775 Objective Loss 0.347775 LR 0.001000 Time 0.026626 +2023-10-05 20:51:56,101 - Epoch: [21][ 130/ 1236] Overall Loss 0.351185 Objective Loss 0.351185 LR 0.001000 Time 0.026139 +2023-10-05 20:51:56,302 - Epoch: [21][ 140/ 1236] Overall Loss 0.350689 Objective Loss 0.350689 LR 0.001000 Time 0.025704 +2023-10-05 20:51:56,502 - Epoch: [21][ 150/ 1236] Overall Loss 0.352414 Objective Loss 0.352414 LR 0.001000 Time 0.025318 +2023-10-05 20:51:56,702 - Epoch: [21][ 160/ 1236] Overall Loss 0.353311 Objective Loss 0.353311 LR 0.001000 Time 0.024983 +2023-10-05 20:51:56,902 - Epoch: [21][ 170/ 1236] Overall Loss 0.353624 Objective Loss 0.353624 LR 0.001000 Time 0.024692 +2023-10-05 20:51:57,101 - Epoch: [21][ 180/ 1236] Overall Loss 0.354053 Objective Loss 0.354053 LR 0.001000 Time 0.024425 +2023-10-05 20:51:57,302 - Epoch: [21][ 190/ 1236] Overall Loss 0.354731 Objective Loss 0.354731 LR 0.001000 Time 0.024194 +2023-10-05 20:51:57,501 - Epoch: [21][ 200/ 1236] Overall Loss 0.355102 Objective Loss 0.355102 LR 0.001000 Time 0.023977 +2023-10-05 20:51:57,701 - Epoch: [21][ 210/ 1236] Overall Loss 0.353810 Objective Loss 0.353810 LR 0.001000 Time 0.023789 +2023-10-05 20:51:57,902 - Epoch: [21][ 220/ 1236] Overall Loss 0.353107 Objective Loss 0.353107 LR 0.001000 Time 0.023618 +2023-10-05 20:51:58,103 - Epoch: [21][ 230/ 1236] Overall Loss 0.353257 Objective Loss 0.353257 LR 0.001000 Time 0.023462 +2023-10-05 20:51:58,303 - Epoch: [21][ 240/ 1236] Overall Loss 0.353449 Objective Loss 0.353449 LR 0.001000 Time 0.023318 +2023-10-05 20:51:58,504 - Epoch: [21][ 250/ 1236] Overall Loss 0.351749 Objective Loss 0.351749 LR 0.001000 Time 0.023187 +2023-10-05 20:51:58,705 - Epoch: [21][ 260/ 1236] Overall Loss 0.351357 Objective Loss 0.351357 LR 0.001000 Time 0.023066 +2023-10-05 20:51:58,904 - Epoch: [21][ 270/ 1236] Overall Loss 0.351050 Objective Loss 0.351050 LR 0.001000 Time 0.022947 +2023-10-05 20:51:59,107 - Epoch: [21][ 280/ 1236] Overall Loss 0.351398 Objective Loss 0.351398 LR 0.001000 Time 0.022854 +2023-10-05 20:51:59,309 - Epoch: [21][ 290/ 1236] Overall Loss 0.352231 Objective Loss 0.352231 LR 0.001000 Time 0.022759 +2023-10-05 20:51:59,512 - Epoch: [21][ 300/ 1236] Overall Loss 0.351487 Objective Loss 0.351487 LR 0.001000 Time 0.022678 +2023-10-05 20:51:59,713 - Epoch: [21][ 310/ 1236] Overall Loss 0.350492 Objective Loss 0.350492 LR 0.001000 Time 0.022594 +2023-10-05 20:51:59,916 - Epoch: [21][ 320/ 1236] Overall Loss 0.351600 Objective Loss 0.351600 LR 0.001000 Time 0.022521 +2023-10-05 20:52:00,117 - Epoch: [21][ 330/ 1236] Overall Loss 0.351135 Objective Loss 0.351135 LR 0.001000 Time 0.022447 +2023-10-05 20:52:00,318 - Epoch: [21][ 340/ 1236] Overall Loss 0.350763 Objective Loss 0.350763 LR 0.001000 Time 0.022375 +2023-10-05 20:52:00,519 - Epoch: [21][ 350/ 1236] Overall Loss 0.351654 Objective Loss 0.351654 LR 0.001000 Time 0.022309 +2023-10-05 20:52:00,720 - Epoch: [21][ 360/ 1236] Overall Loss 0.351492 Objective Loss 0.351492 LR 0.001000 Time 0.022246 +2023-10-05 20:52:00,921 - Epoch: [21][ 370/ 1236] Overall Loss 0.352530 Objective Loss 0.352530 LR 0.001000 Time 0.022188 +2023-10-05 20:52:01,123 - Epoch: [21][ 380/ 1236] Overall Loss 0.352927 Objective Loss 0.352927 LR 0.001000 Time 0.022136 +2023-10-05 20:52:01,330 - Epoch: [21][ 390/ 1236] Overall Loss 0.353064 Objective Loss 0.353064 LR 0.001000 Time 0.022098 +2023-10-05 20:52:01,533 - Epoch: [21][ 400/ 1236] Overall Loss 0.353561 Objective Loss 0.353561 LR 0.001000 Time 0.022052 +2023-10-05 20:52:01,740 - Epoch: [21][ 410/ 1236] Overall Loss 0.353814 Objective Loss 0.353814 LR 0.001000 Time 0.022017 +2023-10-05 20:52:01,943 - Epoch: [21][ 420/ 1236] Overall Loss 0.353930 Objective Loss 0.353930 LR 0.001000 Time 0.021976 +2023-10-05 20:52:02,150 - Epoch: [21][ 430/ 1236] Overall Loss 0.353759 Objective Loss 0.353759 LR 0.001000 Time 0.021945 +2023-10-05 20:52:02,353 - Epoch: [21][ 440/ 1236] Overall Loss 0.354267 Objective Loss 0.354267 LR 0.001000 Time 0.021908 +2023-10-05 20:52:02,560 - Epoch: [21][ 450/ 1236] Overall Loss 0.354280 Objective Loss 0.354280 LR 0.001000 Time 0.021879 +2023-10-05 20:52:02,763 - Epoch: [21][ 460/ 1236] Overall Loss 0.354542 Objective Loss 0.354542 LR 0.001000 Time 0.021845 +2023-10-05 20:52:02,971 - Epoch: [21][ 470/ 1236] Overall Loss 0.354161 Objective Loss 0.354161 LR 0.001000 Time 0.021820 +2023-10-05 20:52:03,173 - Epoch: [21][ 480/ 1236] Overall Loss 0.353931 Objective Loss 0.353931 LR 0.001000 Time 0.021787 +2023-10-05 20:52:03,380 - Epoch: [21][ 490/ 1236] Overall Loss 0.354054 Objective Loss 0.354054 LR 0.001000 Time 0.021764 +2023-10-05 20:52:03,583 - Epoch: [21][ 500/ 1236] Overall Loss 0.353480 Objective Loss 0.353480 LR 0.001000 Time 0.021734 +2023-10-05 20:52:03,790 - Epoch: [21][ 510/ 1236] Overall Loss 0.353313 Objective Loss 0.353313 LR 0.001000 Time 0.021713 +2023-10-05 20:52:03,993 - Epoch: [21][ 520/ 1236] Overall Loss 0.352657 Objective Loss 0.352657 LR 0.001000 Time 0.021686 +2023-10-05 20:52:04,200 - Epoch: [21][ 530/ 1236] Overall Loss 0.352843 Objective Loss 0.352843 LR 0.001000 Time 0.021666 +2023-10-05 20:52:04,403 - Epoch: [21][ 540/ 1236] Overall Loss 0.352907 Objective Loss 0.352907 LR 0.001000 Time 0.021640 +2023-10-05 20:52:04,610 - Epoch: [21][ 550/ 1236] Overall Loss 0.352947 Objective Loss 0.352947 LR 0.001000 Time 0.021622 +2023-10-05 20:52:04,813 - Epoch: [21][ 560/ 1236] Overall Loss 0.353189 Objective Loss 0.353189 LR 0.001000 Time 0.021598 +2023-10-05 20:52:05,020 - Epoch: [21][ 570/ 1236] Overall Loss 0.353111 Objective Loss 0.353111 LR 0.001000 Time 0.021581 +2023-10-05 20:52:05,223 - Epoch: [21][ 580/ 1236] Overall Loss 0.353139 Objective Loss 0.353139 LR 0.001000 Time 0.021559 +2023-10-05 20:52:05,431 - Epoch: [21][ 590/ 1236] Overall Loss 0.353142 Objective Loss 0.353142 LR 0.001000 Time 0.021544 +2023-10-05 20:52:05,633 - Epoch: [21][ 600/ 1236] Overall Loss 0.352764 Objective Loss 0.352764 LR 0.001000 Time 0.021522 +2023-10-05 20:52:05,840 - Epoch: [21][ 610/ 1236] Overall Loss 0.352991 Objective Loss 0.352991 LR 0.001000 Time 0.021508 +2023-10-05 20:52:06,043 - Epoch: [21][ 620/ 1236] Overall Loss 0.352546 Objective Loss 0.352546 LR 0.001000 Time 0.021488 +2023-10-05 20:52:06,250 - Epoch: [21][ 630/ 1236] Overall Loss 0.352734 Objective Loss 0.352734 LR 0.001000 Time 0.021475 +2023-10-05 20:52:06,453 - Epoch: [21][ 640/ 1236] Overall Loss 0.353116 Objective Loss 0.353116 LR 0.001000 Time 0.021456 +2023-10-05 20:52:06,661 - Epoch: [21][ 650/ 1236] Overall Loss 0.353017 Objective Loss 0.353017 LR 0.001000 Time 0.021441 +2023-10-05 20:52:06,864 - Epoch: [21][ 660/ 1236] Overall Loss 0.352369 Objective Loss 0.352369 LR 0.001000 Time 0.021423 +2023-10-05 20:52:07,070 - Epoch: [21][ 670/ 1236] Overall Loss 0.352141 Objective Loss 0.352141 LR 0.001000 Time 0.021411 +2023-10-05 20:52:07,273 - Epoch: [21][ 680/ 1236] Overall Loss 0.352130 Objective Loss 0.352130 LR 0.001000 Time 0.021395 +2023-10-05 20:52:07,480 - Epoch: [21][ 690/ 1236] Overall Loss 0.351674 Objective Loss 0.351674 LR 0.001000 Time 0.021384 +2023-10-05 20:52:07,683 - Epoch: [21][ 700/ 1236] Overall Loss 0.351580 Objective Loss 0.351580 LR 0.001000 Time 0.021368 +2023-10-05 20:52:07,891 - Epoch: [21][ 710/ 1236] Overall Loss 0.351928 Objective Loss 0.351928 LR 0.001000 Time 0.021358 +2023-10-05 20:52:08,093 - Epoch: [21][ 720/ 1236] Overall Loss 0.351864 Objective Loss 0.351864 LR 0.001000 Time 0.021343 +2023-10-05 20:52:08,300 - Epoch: [21][ 730/ 1236] Overall Loss 0.352272 Objective Loss 0.352272 LR 0.001000 Time 0.021333 +2023-10-05 20:52:08,503 - Epoch: [21][ 740/ 1236] Overall Loss 0.352828 Objective Loss 0.352828 LR 0.001000 Time 0.021319 +2023-10-05 20:52:08,710 - Epoch: [21][ 750/ 1236] Overall Loss 0.352671 Objective Loss 0.352671 LR 0.001000 Time 0.021310 +2023-10-05 20:52:08,916 - Epoch: [21][ 760/ 1236] Overall Loss 0.352602 Objective Loss 0.352602 LR 0.001000 Time 0.021300 +2023-10-05 20:52:09,129 - Epoch: [21][ 770/ 1236] Overall Loss 0.352376 Objective Loss 0.352376 LR 0.001000 Time 0.021300 +2023-10-05 20:52:09,338 - Epoch: [21][ 780/ 1236] Overall Loss 0.352419 Objective Loss 0.352419 LR 0.001000 Time 0.021294 +2023-10-05 20:52:09,551 - Epoch: [21][ 790/ 1236] Overall Loss 0.352705 Objective Loss 0.352705 LR 0.001000 Time 0.021294 +2023-10-05 20:52:09,760 - Epoch: [21][ 800/ 1236] Overall Loss 0.352957 Objective Loss 0.352957 LR 0.001000 Time 0.021289 +2023-10-05 20:52:09,973 - Epoch: [21][ 810/ 1236] Overall Loss 0.353125 Objective Loss 0.353125 LR 0.001000 Time 0.021289 +2023-10-05 20:52:10,182 - Epoch: [21][ 820/ 1236] Overall Loss 0.353058 Objective Loss 0.353058 LR 0.001000 Time 0.021284 +2023-10-05 20:52:10,395 - Epoch: [21][ 830/ 1236] Overall Loss 0.352894 Objective Loss 0.352894 LR 0.001000 Time 0.021283 +2023-10-05 20:52:10,605 - Epoch: [21][ 840/ 1236] Overall Loss 0.353010 Objective Loss 0.353010 LR 0.001000 Time 0.021279 +2023-10-05 20:52:10,817 - Epoch: [21][ 850/ 1236] Overall Loss 0.353094 Objective Loss 0.353094 LR 0.001000 Time 0.021278 +2023-10-05 20:52:11,026 - Epoch: [21][ 860/ 1236] Overall Loss 0.353621 Objective Loss 0.353621 LR 0.001000 Time 0.021274 +2023-10-05 20:52:11,240 - Epoch: [21][ 870/ 1236] Overall Loss 0.353855 Objective Loss 0.353855 LR 0.001000 Time 0.021274 +2023-10-05 20:52:11,449 - Epoch: [21][ 880/ 1236] Overall Loss 0.353354 Objective Loss 0.353354 LR 0.001000 Time 0.021269 +2023-10-05 20:52:11,662 - Epoch: [21][ 890/ 1236] Overall Loss 0.353685 Objective Loss 0.353685 LR 0.001000 Time 0.021270 +2023-10-05 20:52:11,871 - Epoch: [21][ 900/ 1236] Overall Loss 0.353676 Objective Loss 0.353676 LR 0.001000 Time 0.021265 +2023-10-05 20:52:12,084 - Epoch: [21][ 910/ 1236] Overall Loss 0.353660 Objective Loss 0.353660 LR 0.001000 Time 0.021265 +2023-10-05 20:52:12,294 - Epoch: [21][ 920/ 1236] Overall Loss 0.353200 Objective Loss 0.353200 LR 0.001000 Time 0.021261 +2023-10-05 20:52:12,507 - Epoch: [21][ 930/ 1236] Overall Loss 0.353692 Objective Loss 0.353692 LR 0.001000 Time 0.021262 +2023-10-05 20:52:12,716 - Epoch: [21][ 940/ 1236] Overall Loss 0.353854 Objective Loss 0.353854 LR 0.001000 Time 0.021258 +2023-10-05 20:52:12,929 - Epoch: [21][ 950/ 1236] Overall Loss 0.353889 Objective Loss 0.353889 LR 0.001000 Time 0.021258 +2023-10-05 20:52:13,138 - Epoch: [21][ 960/ 1236] Overall Loss 0.353846 Objective Loss 0.353846 LR 0.001000 Time 0.021254 +2023-10-05 20:52:13,352 - Epoch: [21][ 970/ 1236] Overall Loss 0.353838 Objective Loss 0.353838 LR 0.001000 Time 0.021255 +2023-10-05 20:52:13,561 - Epoch: [21][ 980/ 1236] Overall Loss 0.353885 Objective Loss 0.353885 LR 0.001000 Time 0.021251 +2023-10-05 20:52:13,773 - Epoch: [21][ 990/ 1236] Overall Loss 0.354171 Objective Loss 0.354171 LR 0.001000 Time 0.021250 +2023-10-05 20:52:13,976 - Epoch: [21][ 1000/ 1236] Overall Loss 0.354052 Objective Loss 0.354052 LR 0.001000 Time 0.021240 +2023-10-05 20:52:14,180 - Epoch: [21][ 1010/ 1236] Overall Loss 0.353840 Objective Loss 0.353840 LR 0.001000 Time 0.021232 +2023-10-05 20:52:14,382 - Epoch: [21][ 1020/ 1236] Overall Loss 0.353761 Objective Loss 0.353761 LR 0.001000 Time 0.021221 +2023-10-05 20:52:14,586 - Epoch: [21][ 1030/ 1236] Overall Loss 0.353706 Objective Loss 0.353706 LR 0.001000 Time 0.021213 +2023-10-05 20:52:14,788 - Epoch: [21][ 1040/ 1236] Overall Loss 0.353844 Objective Loss 0.353844 LR 0.001000 Time 0.021203 +2023-10-05 20:52:14,992 - Epoch: [21][ 1050/ 1236] Overall Loss 0.353690 Objective Loss 0.353690 LR 0.001000 Time 0.021194 +2023-10-05 20:52:15,193 - Epoch: [21][ 1060/ 1236] Overall Loss 0.353885 Objective Loss 0.353885 LR 0.001000 Time 0.021183 +2023-10-05 20:52:15,396 - Epoch: [21][ 1070/ 1236] Overall Loss 0.354287 Objective Loss 0.354287 LR 0.001000 Time 0.021175 +2023-10-05 20:52:15,596 - Epoch: [21][ 1080/ 1236] Overall Loss 0.353714 Objective Loss 0.353714 LR 0.001000 Time 0.021164 +2023-10-05 20:52:15,799 - Epoch: [21][ 1090/ 1236] Overall Loss 0.353740 Objective Loss 0.353740 LR 0.001000 Time 0.021156 +2023-10-05 20:52:16,000 - Epoch: [21][ 1100/ 1236] Overall Loss 0.354081 Objective Loss 0.354081 LR 0.001000 Time 0.021145 +2023-10-05 20:52:16,202 - Epoch: [21][ 1110/ 1236] Overall Loss 0.354063 Objective Loss 0.354063 LR 0.001000 Time 0.021137 +2023-10-05 20:52:16,404 - Epoch: [21][ 1120/ 1236] Overall Loss 0.353984 Objective Loss 0.353984 LR 0.001000 Time 0.021128 +2023-10-05 20:52:16,606 - Epoch: [21][ 1130/ 1236] Overall Loss 0.353799 Objective Loss 0.353799 LR 0.001000 Time 0.021120 +2023-10-05 20:52:16,807 - Epoch: [21][ 1140/ 1236] Overall Loss 0.353511 Objective Loss 0.353511 LR 0.001000 Time 0.021110 +2023-10-05 20:52:17,011 - Epoch: [21][ 1150/ 1236] Overall Loss 0.353533 Objective Loss 0.353533 LR 0.001000 Time 0.021104 +2023-10-05 20:52:17,213 - Epoch: [21][ 1160/ 1236] Overall Loss 0.353626 Objective Loss 0.353626 LR 0.001000 Time 0.021096 +2023-10-05 20:52:17,424 - Epoch: [21][ 1170/ 1236] Overall Loss 0.353541 Objective Loss 0.353541 LR 0.001000 Time 0.021095 +2023-10-05 20:52:17,633 - Epoch: [21][ 1180/ 1236] Overall Loss 0.353590 Objective Loss 0.353590 LR 0.001000 Time 0.021094 +2023-10-05 20:52:17,847 - Epoch: [21][ 1190/ 1236] Overall Loss 0.353595 Objective Loss 0.353595 LR 0.001000 Time 0.021096 +2023-10-05 20:52:18,057 - Epoch: [21][ 1200/ 1236] Overall Loss 0.353331 Objective Loss 0.353331 LR 0.001000 Time 0.021095 +2023-10-05 20:52:18,271 - Epoch: [21][ 1210/ 1236] Overall Loss 0.353451 Objective Loss 0.353451 LR 0.001000 Time 0.021097 +2023-10-05 20:52:18,480 - Epoch: [21][ 1220/ 1236] Overall Loss 0.353491 Objective Loss 0.353491 LR 0.001000 Time 0.021095 +2023-10-05 20:52:18,744 - Epoch: [21][ 1230/ 1236] Overall Loss 0.353642 Objective Loss 0.353642 LR 0.001000 Time 0.021138 +2023-10-05 20:52:18,863 - Epoch: [21][ 1236/ 1236] Overall Loss 0.353554 Objective Loss 0.353554 Top1 84.521385 Top5 97.148676 LR 0.001000 Time 0.021132 +2023-10-05 20:52:18,982 - --- validate (epoch=21)----------- +2023-10-05 20:52:18,982 - 29943 samples (256 per mini-batch) +2023-10-05 20:52:19,435 - Epoch: [21][ 10/ 117] Loss 0.379669 Top1 78.164062 Top5 96.835938 +2023-10-05 20:52:19,587 - Epoch: [21][ 20/ 117] Loss 0.398073 Top1 77.792969 Top5 96.855469 +2023-10-05 20:52:19,735 - Epoch: [21][ 30/ 117] Loss 0.375694 Top1 78.541667 Top5 96.888021 +2023-10-05 20:52:19,885 - Epoch: [21][ 40/ 117] Loss 0.376953 Top1 78.798828 Top5 96.884766 +2023-10-05 20:52:20,034 - Epoch: [21][ 50/ 117] Loss 0.371592 Top1 79.054688 Top5 96.945312 +2023-10-05 20:52:20,182 - Epoch: [21][ 60/ 117] Loss 0.370622 Top1 78.951823 Top5 96.979167 +2023-10-05 20:52:20,332 - Epoch: [21][ 70/ 117] Loss 0.375922 Top1 78.722098 Top5 96.914062 +2023-10-05 20:52:20,484 - Epoch: [21][ 80/ 117] Loss 0.375943 Top1 78.696289 Top5 96.811523 +2023-10-05 20:52:20,632 - Epoch: [21][ 90/ 117] Loss 0.377035 Top1 78.628472 Top5 96.783854 +2023-10-05 20:52:20,780 - Epoch: [21][ 100/ 117] Loss 0.382006 Top1 78.574219 Top5 96.699219 +2023-10-05 20:52:20,935 - Epoch: [21][ 110/ 117] Loss 0.379431 Top1 78.725142 Top5 96.754261 +2023-10-05 20:52:21,021 - Epoch: [21][ 117/ 117] Loss 0.381469 Top1 78.686170 Top5 96.757172 +2023-10-05 20:52:21,125 - ==> Top1: 78.686 Top5: 96.757 Loss: 0.381 + +2023-10-05 20:52:21,126 - ==> Confusion: +[[ 888 0 6 1 15 1 0 0 2 96 0 1 1 7 6 3 11 2 0 2 8] + [ 2 1025 1 0 11 15 3 23 4 2 5 3 0 0 2 5 14 2 8 2 4] + [ 1 0 924 5 4 1 52 13 0 1 5 7 7 5 2 3 3 2 5 4 12] + [ 5 1 32 899 3 4 2 0 9 0 17 1 11 4 47 5 3 15 14 2 15] + [ 12 8 0 0 981 5 1 0 1 10 1 3 0 1 9 3 10 2 1 2 0] + [ 8 51 1 1 13 953 2 17 4 2 6 18 2 12 4 2 8 0 1 1 10] + [ 0 6 20 0 0 0 1123 5 0 0 6 1 1 0 0 11 0 2 4 7 5] + [ 4 33 24 0 6 32 9 998 2 3 6 21 1 1 1 4 1 0 44 21 7] + [ 17 2 1 0 0 2 0 0 958 54 10 3 1 18 12 7 0 2 1 1 0] + [ 86 0 1 0 6 1 0 1 29 942 0 4 1 30 6 2 4 0 0 3 3] + [ 3 4 13 4 1 2 2 3 15 2 960 2 0 17 6 4 1 0 4 3 7] + [ 0 0 1 1 2 9 0 0 0 0 0 940 45 5 0 3 1 19 0 8 1] + [ 0 1 4 3 1 1 0 2 0 0 1 52 944 5 4 11 4 25 1 6 3] + [ 2 0 0 1 6 10 0 1 13 13 13 8 2 1029 3 4 2 0 0 1 11] + [ 19 3 4 6 6 0 0 0 32 9 3 0 2 2 991 1 1 5 8 0 9] + [ 0 2 5 0 5 1 0 0 0 0 0 8 12 2 0 1052 18 11 0 7 11] + [ 0 13 1 1 12 2 0 1 0 0 0 10 3 2 3 13 1093 0 0 4 3] + [ 1 0 0 0 0 0 1 0 1 0 0 5 27 0 3 10 1 988 1 0 0] + [ 0 11 14 7 3 1 0 32 11 0 5 8 8 0 25 0 2 0 930 2 9] + [ 0 3 2 0 2 7 9 13 0 1 2 17 10 1 0 8 6 1 3 1055 12] + [ 242 275 244 42 205 183 58 79 156 147 289 172 500 374 201 108 269 84 125 264 3888]] + +2023-10-05 20:52:21,127 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:52:21,127 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:52:21,133 - + +2023-10-05 20:52:21,133 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:52:22,247 - Epoch: [22][ 10/ 1236] Overall Loss 0.359896 Objective Loss 0.359896 LR 0.001000 Time 0.111312 +2023-10-05 20:52:22,448 - Epoch: [22][ 20/ 1236] Overall Loss 0.345101 Objective Loss 0.345101 LR 0.001000 Time 0.065668 +2023-10-05 20:52:22,646 - Epoch: [22][ 30/ 1236] Overall Loss 0.343758 Objective Loss 0.343758 LR 0.001000 Time 0.050390 +2023-10-05 20:52:22,847 - Epoch: [22][ 40/ 1236] Overall Loss 0.338695 Objective Loss 0.338695 LR 0.001000 Time 0.042802 +2023-10-05 20:52:23,044 - Epoch: [22][ 50/ 1236] Overall Loss 0.332546 Objective Loss 0.332546 LR 0.001000 Time 0.038171 +2023-10-05 20:52:23,245 - Epoch: [22][ 60/ 1236] Overall Loss 0.336772 Objective Loss 0.336772 LR 0.001000 Time 0.035155 +2023-10-05 20:52:23,444 - Epoch: [22][ 70/ 1236] Overall Loss 0.331439 Objective Loss 0.331439 LR 0.001000 Time 0.032966 +2023-10-05 20:52:23,643 - Epoch: [22][ 80/ 1236] Overall Loss 0.331316 Objective Loss 0.331316 LR 0.001000 Time 0.031337 +2023-10-05 20:52:23,841 - Epoch: [22][ 90/ 1236] Overall Loss 0.329449 Objective Loss 0.329449 LR 0.001000 Time 0.030050 +2023-10-05 20:52:24,042 - Epoch: [22][ 100/ 1236] Overall Loss 0.329550 Objective Loss 0.329550 LR 0.001000 Time 0.029049 +2023-10-05 20:52:24,239 - Epoch: [22][ 110/ 1236] Overall Loss 0.331166 Objective Loss 0.331166 LR 0.001000 Time 0.028202 +2023-10-05 20:52:24,440 - Epoch: [22][ 120/ 1236] Overall Loss 0.329639 Objective Loss 0.329639 LR 0.001000 Time 0.027520 +2023-10-05 20:52:24,639 - Epoch: [22][ 130/ 1236] Overall Loss 0.331370 Objective Loss 0.331370 LR 0.001000 Time 0.026928 +2023-10-05 20:52:24,840 - Epoch: [22][ 140/ 1236] Overall Loss 0.329185 Objective Loss 0.329185 LR 0.001000 Time 0.026437 +2023-10-05 20:52:25,038 - Epoch: [22][ 150/ 1236] Overall Loss 0.328414 Objective Loss 0.328414 LR 0.001000 Time 0.025994 +2023-10-05 20:52:25,239 - Epoch: [22][ 160/ 1236] Overall Loss 0.327008 Objective Loss 0.327008 LR 0.001000 Time 0.025624 +2023-10-05 20:52:25,437 - Epoch: [22][ 170/ 1236] Overall Loss 0.326969 Objective Loss 0.326969 LR 0.001000 Time 0.025282 +2023-10-05 20:52:25,638 - Epoch: [22][ 180/ 1236] Overall Loss 0.326746 Objective Loss 0.326746 LR 0.001000 Time 0.024991 +2023-10-05 20:52:25,837 - Epoch: [22][ 190/ 1236] Overall Loss 0.327370 Objective Loss 0.327370 LR 0.001000 Time 0.024719 +2023-10-05 20:52:26,038 - Epoch: [22][ 200/ 1236] Overall Loss 0.328425 Objective Loss 0.328425 LR 0.001000 Time 0.024485 +2023-10-05 20:52:26,237 - Epoch: [22][ 210/ 1236] Overall Loss 0.328050 Objective Loss 0.328050 LR 0.001000 Time 0.024265 +2023-10-05 20:52:26,437 - Epoch: [22][ 220/ 1236] Overall Loss 0.327653 Objective Loss 0.327653 LR 0.001000 Time 0.024073 +2023-10-05 20:52:26,636 - Epoch: [22][ 230/ 1236] Overall Loss 0.327454 Objective Loss 0.327454 LR 0.001000 Time 0.023888 +2023-10-05 20:52:26,837 - Epoch: [22][ 240/ 1236] Overall Loss 0.328391 Objective Loss 0.328391 LR 0.001000 Time 0.023728 +2023-10-05 20:52:27,035 - Epoch: [22][ 250/ 1236] Overall Loss 0.328758 Objective Loss 0.328758 LR 0.001000 Time 0.023572 +2023-10-05 20:52:27,236 - Epoch: [22][ 260/ 1236] Overall Loss 0.328208 Objective Loss 0.328208 LR 0.001000 Time 0.023435 +2023-10-05 20:52:27,435 - Epoch: [22][ 270/ 1236] Overall Loss 0.329870 Objective Loss 0.329870 LR 0.001000 Time 0.023302 +2023-10-05 20:52:27,635 - Epoch: [22][ 280/ 1236] Overall Loss 0.329807 Objective Loss 0.329807 LR 0.001000 Time 0.023185 +2023-10-05 20:52:27,834 - Epoch: [22][ 290/ 1236] Overall Loss 0.331403 Objective Loss 0.331403 LR 0.001000 Time 0.023069 +2023-10-05 20:52:28,035 - Epoch: [22][ 300/ 1236] Overall Loss 0.331133 Objective Loss 0.331133 LR 0.001000 Time 0.022967 +2023-10-05 20:52:28,233 - Epoch: [22][ 310/ 1236] Overall Loss 0.331499 Objective Loss 0.331499 LR 0.001000 Time 0.022866 +2023-10-05 20:52:28,434 - Epoch: [22][ 320/ 1236] Overall Loss 0.331157 Objective Loss 0.331157 LR 0.001000 Time 0.022778 +2023-10-05 20:52:28,633 - Epoch: [22][ 330/ 1236] Overall Loss 0.332030 Objective Loss 0.332030 LR 0.001000 Time 0.022689 +2023-10-05 20:52:28,834 - Epoch: [22][ 340/ 1236] Overall Loss 0.332399 Objective Loss 0.332399 LR 0.001000 Time 0.022611 +2023-10-05 20:52:29,033 - Epoch: [22][ 350/ 1236] Overall Loss 0.331464 Objective Loss 0.331464 LR 0.001000 Time 0.022532 +2023-10-05 20:52:29,234 - Epoch: [22][ 360/ 1236] Overall Loss 0.330962 Objective Loss 0.330962 LR 0.001000 Time 0.022464 +2023-10-05 20:52:29,432 - Epoch: [22][ 370/ 1236] Overall Loss 0.331467 Objective Loss 0.331467 LR 0.001000 Time 0.022393 +2023-10-05 20:52:29,632 - Epoch: [22][ 380/ 1236] Overall Loss 0.332215 Objective Loss 0.332215 LR 0.001000 Time 0.022328 +2023-10-05 20:52:29,831 - Epoch: [22][ 390/ 1236] Overall Loss 0.332334 Objective Loss 0.332334 LR 0.001000 Time 0.022264 +2023-10-05 20:52:30,032 - Epoch: [22][ 400/ 1236] Overall Loss 0.332645 Objective Loss 0.332645 LR 0.001000 Time 0.022210 +2023-10-05 20:52:30,231 - Epoch: [22][ 410/ 1236] Overall Loss 0.333345 Objective Loss 0.333345 LR 0.001000 Time 0.022153 +2023-10-05 20:52:30,432 - Epoch: [22][ 420/ 1236] Overall Loss 0.333259 Objective Loss 0.333259 LR 0.001000 Time 0.022103 +2023-10-05 20:52:30,631 - Epoch: [22][ 430/ 1236] Overall Loss 0.333922 Objective Loss 0.333922 LR 0.001000 Time 0.022051 +2023-10-05 20:52:30,830 - Epoch: [22][ 440/ 1236] Overall Loss 0.334261 Objective Loss 0.334261 LR 0.001000 Time 0.022002 +2023-10-05 20:52:31,029 - Epoch: [22][ 450/ 1236] Overall Loss 0.335217 Objective Loss 0.335217 LR 0.001000 Time 0.021954 +2023-10-05 20:52:31,230 - Epoch: [22][ 460/ 1236] Overall Loss 0.335171 Objective Loss 0.335171 LR 0.001000 Time 0.021913 +2023-10-05 20:52:31,427 - Epoch: [22][ 470/ 1236] Overall Loss 0.335689 Objective Loss 0.335689 LR 0.001000 Time 0.021866 +2023-10-05 20:52:31,628 - Epoch: [22][ 480/ 1236] Overall Loss 0.336109 Objective Loss 0.336109 LR 0.001000 Time 0.021828 +2023-10-05 20:52:31,827 - Epoch: [22][ 490/ 1236] Overall Loss 0.335937 Objective Loss 0.335937 LR 0.001000 Time 0.021788 +2023-10-05 20:52:32,028 - Epoch: [22][ 500/ 1236] Overall Loss 0.335533 Objective Loss 0.335533 LR 0.001000 Time 0.021754 +2023-10-05 20:52:32,226 - Epoch: [22][ 510/ 1236] Overall Loss 0.335785 Objective Loss 0.335785 LR 0.001000 Time 0.021715 +2023-10-05 20:52:32,427 - Epoch: [22][ 520/ 1236] Overall Loss 0.335823 Objective Loss 0.335823 LR 0.001000 Time 0.021683 +2023-10-05 20:52:32,626 - Epoch: [22][ 530/ 1236] Overall Loss 0.335695 Objective Loss 0.335695 LR 0.001000 Time 0.021648 +2023-10-05 20:52:32,825 - Epoch: [22][ 540/ 1236] Overall Loss 0.335391 Objective Loss 0.335391 LR 0.001000 Time 0.021616 +2023-10-05 20:52:33,024 - Epoch: [22][ 550/ 1236] Overall Loss 0.335679 Objective Loss 0.335679 LR 0.001000 Time 0.021584 +2023-10-05 20:52:33,225 - Epoch: [22][ 560/ 1236] Overall Loss 0.335742 Objective Loss 0.335742 LR 0.001000 Time 0.021557 +2023-10-05 20:52:33,424 - Epoch: [22][ 570/ 1236] Overall Loss 0.335648 Objective Loss 0.335648 LR 0.001000 Time 0.021527 +2023-10-05 20:52:33,625 - Epoch: [22][ 580/ 1236] Overall Loss 0.335362 Objective Loss 0.335362 LR 0.001000 Time 0.021502 +2023-10-05 20:52:33,824 - Epoch: [22][ 590/ 1236] Overall Loss 0.335373 Objective Loss 0.335373 LR 0.001000 Time 0.021474 +2023-10-05 20:52:34,025 - Epoch: [22][ 600/ 1236] Overall Loss 0.335688 Objective Loss 0.335688 LR 0.001000 Time 0.021451 +2023-10-05 20:52:34,224 - Epoch: [22][ 610/ 1236] Overall Loss 0.335436 Objective Loss 0.335436 LR 0.001000 Time 0.021425 +2023-10-05 20:52:34,425 - Epoch: [22][ 620/ 1236] Overall Loss 0.335888 Objective Loss 0.335888 LR 0.001000 Time 0.021403 +2023-10-05 20:52:34,624 - Epoch: [22][ 630/ 1236] Overall Loss 0.335844 Objective Loss 0.335844 LR 0.001000 Time 0.021379 +2023-10-05 20:52:34,825 - Epoch: [22][ 640/ 1236] Overall Loss 0.336810 Objective Loss 0.336810 LR 0.001000 Time 0.021358 +2023-10-05 20:52:35,024 - Epoch: [22][ 650/ 1236] Overall Loss 0.336928 Objective Loss 0.336928 LR 0.001000 Time 0.021335 +2023-10-05 20:52:35,224 - Epoch: [22][ 660/ 1236] Overall Loss 0.337358 Objective Loss 0.337358 LR 0.001000 Time 0.021314 +2023-10-05 20:52:35,423 - Epoch: [22][ 670/ 1236] Overall Loss 0.337546 Objective Loss 0.337546 LR 0.001000 Time 0.021292 +2023-10-05 20:52:35,624 - Epoch: [22][ 680/ 1236] Overall Loss 0.338307 Objective Loss 0.338307 LR 0.001000 Time 0.021275 +2023-10-05 20:52:35,823 - Epoch: [22][ 690/ 1236] Overall Loss 0.338911 Objective Loss 0.338911 LR 0.001000 Time 0.021254 +2023-10-05 20:52:36,024 - Epoch: [22][ 700/ 1236] Overall Loss 0.339588 Objective Loss 0.339588 LR 0.001000 Time 0.021237 +2023-10-05 20:52:36,223 - Epoch: [22][ 710/ 1236] Overall Loss 0.339723 Objective Loss 0.339723 LR 0.001000 Time 0.021218 +2023-10-05 20:52:36,424 - Epoch: [22][ 720/ 1236] Overall Loss 0.340064 Objective Loss 0.340064 LR 0.001000 Time 0.021202 +2023-10-05 20:52:36,623 - Epoch: [22][ 730/ 1236] Overall Loss 0.340032 Objective Loss 0.340032 LR 0.001000 Time 0.021183 +2023-10-05 20:52:36,821 - Epoch: [22][ 740/ 1236] Overall Loss 0.339869 Objective Loss 0.339869 LR 0.001000 Time 0.021165 +2023-10-05 20:52:37,020 - Epoch: [22][ 750/ 1236] Overall Loss 0.340382 Objective Loss 0.340382 LR 0.001000 Time 0.021147 +2023-10-05 20:52:37,220 - Epoch: [22][ 760/ 1236] Overall Loss 0.340611 Objective Loss 0.340611 LR 0.001000 Time 0.021131 +2023-10-05 20:52:37,419 - Epoch: [22][ 770/ 1236] Overall Loss 0.341336 Objective Loss 0.341336 LR 0.001000 Time 0.021115 +2023-10-05 20:52:37,620 - Epoch: [22][ 780/ 1236] Overall Loss 0.341768 Objective Loss 0.341768 LR 0.001000 Time 0.021102 +2023-10-05 20:52:37,819 - Epoch: [22][ 790/ 1236] Overall Loss 0.341875 Objective Loss 0.341875 LR 0.001000 Time 0.021086 +2023-10-05 20:52:38,018 - Epoch: [22][ 800/ 1236] Overall Loss 0.342163 Objective Loss 0.342163 LR 0.001000 Time 0.021071 +2023-10-05 20:52:38,217 - Epoch: [22][ 810/ 1236] Overall Loss 0.342497 Objective Loss 0.342497 LR 0.001000 Time 0.021056 +2023-10-05 20:52:38,418 - Epoch: [22][ 820/ 1236] Overall Loss 0.342750 Objective Loss 0.342750 LR 0.001000 Time 0.021045 +2023-10-05 20:52:38,615 - Epoch: [22][ 830/ 1236] Overall Loss 0.342803 Objective Loss 0.342803 LR 0.001000 Time 0.021028 +2023-10-05 20:52:38,816 - Epoch: [22][ 840/ 1236] Overall Loss 0.342834 Objective Loss 0.342834 LR 0.001000 Time 0.021017 +2023-10-05 20:52:39,016 - Epoch: [22][ 850/ 1236] Overall Loss 0.343186 Objective Loss 0.343186 LR 0.001000 Time 0.021003 +2023-10-05 20:52:39,216 - Epoch: [22][ 860/ 1236] Overall Loss 0.343785 Objective Loss 0.343785 LR 0.001000 Time 0.020991 +2023-10-05 20:52:39,415 - Epoch: [22][ 870/ 1236] Overall Loss 0.344310 Objective Loss 0.344310 LR 0.001000 Time 0.020979 +2023-10-05 20:52:39,616 - Epoch: [22][ 880/ 1236] Overall Loss 0.344395 Objective Loss 0.344395 LR 0.001000 Time 0.020968 +2023-10-05 20:52:39,813 - Epoch: [22][ 890/ 1236] Overall Loss 0.344343 Objective Loss 0.344343 LR 0.001000 Time 0.020954 +2023-10-05 20:52:40,014 - Epoch: [22][ 900/ 1236] Overall Loss 0.344665 Objective Loss 0.344665 LR 0.001000 Time 0.020944 +2023-10-05 20:52:40,211 - Epoch: [22][ 910/ 1236] Overall Loss 0.344757 Objective Loss 0.344757 LR 0.001000 Time 0.020931 +2023-10-05 20:52:40,413 - Epoch: [22][ 920/ 1236] Overall Loss 0.344648 Objective Loss 0.344648 LR 0.001000 Time 0.020922 +2023-10-05 20:52:40,611 - Epoch: [22][ 930/ 1236] Overall Loss 0.344624 Objective Loss 0.344624 LR 0.001000 Time 0.020910 +2023-10-05 20:52:40,812 - Epoch: [22][ 940/ 1236] Overall Loss 0.344930 Objective Loss 0.344930 LR 0.001000 Time 0.020900 +2023-10-05 20:52:41,010 - Epoch: [22][ 950/ 1236] Overall Loss 0.345108 Objective Loss 0.345108 LR 0.001000 Time 0.020889 +2023-10-05 20:52:41,212 - Epoch: [22][ 960/ 1236] Overall Loss 0.345203 Objective Loss 0.345203 LR 0.001000 Time 0.020881 +2023-10-05 20:52:41,411 - Epoch: [22][ 970/ 1236] Overall Loss 0.345453 Objective Loss 0.345453 LR 0.001000 Time 0.020870 +2023-10-05 20:52:41,612 - Epoch: [22][ 980/ 1236] Overall Loss 0.345614 Objective Loss 0.345614 LR 0.001000 Time 0.020862 +2023-10-05 20:52:41,811 - Epoch: [22][ 990/ 1236] Overall Loss 0.345404 Objective Loss 0.345404 LR 0.001000 Time 0.020852 +2023-10-05 20:52:42,012 - Epoch: [22][ 1000/ 1236] Overall Loss 0.345662 Objective Loss 0.345662 LR 0.001000 Time 0.020845 +2023-10-05 20:52:42,209 - Epoch: [22][ 1010/ 1236] Overall Loss 0.345764 Objective Loss 0.345764 LR 0.001000 Time 0.020833 +2023-10-05 20:52:42,410 - Epoch: [22][ 1020/ 1236] Overall Loss 0.346026 Objective Loss 0.346026 LR 0.001000 Time 0.020826 +2023-10-05 20:52:42,609 - Epoch: [22][ 1030/ 1236] Overall Loss 0.346487 Objective Loss 0.346487 LR 0.001000 Time 0.020816 +2023-10-05 20:52:42,810 - Epoch: [22][ 1040/ 1236] Overall Loss 0.346335 Objective Loss 0.346335 LR 0.001000 Time 0.020809 +2023-10-05 20:52:43,007 - Epoch: [22][ 1050/ 1236] Overall Loss 0.346388 Objective Loss 0.346388 LR 0.001000 Time 0.020798 +2023-10-05 20:52:43,209 - Epoch: [22][ 1060/ 1236] Overall Loss 0.346458 Objective Loss 0.346458 LR 0.001000 Time 0.020792 +2023-10-05 20:52:43,408 - Epoch: [22][ 1070/ 1236] Overall Loss 0.346512 Objective Loss 0.346512 LR 0.001000 Time 0.020783 +2023-10-05 20:52:43,609 - Epoch: [22][ 1080/ 1236] Overall Loss 0.346584 Objective Loss 0.346584 LR 0.001000 Time 0.020777 +2023-10-05 20:52:43,808 - Epoch: [22][ 1090/ 1236] Overall Loss 0.346523 Objective Loss 0.346523 LR 0.001000 Time 0.020768 +2023-10-05 20:52:44,009 - Epoch: [22][ 1100/ 1236] Overall Loss 0.346913 Objective Loss 0.346913 LR 0.001000 Time 0.020762 +2023-10-05 20:52:44,208 - Epoch: [22][ 1110/ 1236] Overall Loss 0.347016 Objective Loss 0.347016 LR 0.001000 Time 0.020754 +2023-10-05 20:52:44,409 - Epoch: [22][ 1120/ 1236] Overall Loss 0.347374 Objective Loss 0.347374 LR 0.001000 Time 0.020748 +2023-10-05 20:52:44,608 - Epoch: [22][ 1130/ 1236] Overall Loss 0.348009 Objective Loss 0.348009 LR 0.001000 Time 0.020740 +2023-10-05 20:52:44,809 - Epoch: [22][ 1140/ 1236] Overall Loss 0.348070 Objective Loss 0.348070 LR 0.001000 Time 0.020734 +2023-10-05 20:52:45,008 - Epoch: [22][ 1150/ 1236] Overall Loss 0.348376 Objective Loss 0.348376 LR 0.001000 Time 0.020727 +2023-10-05 20:52:45,208 - Epoch: [22][ 1160/ 1236] Overall Loss 0.348177 Objective Loss 0.348177 LR 0.001000 Time 0.020720 +2023-10-05 20:52:45,407 - Epoch: [22][ 1170/ 1236] Overall Loss 0.348346 Objective Loss 0.348346 LR 0.001000 Time 0.020713 +2023-10-05 20:52:45,607 - Epoch: [22][ 1180/ 1236] Overall Loss 0.348190 Objective Loss 0.348190 LR 0.001000 Time 0.020706 +2023-10-05 20:52:45,805 - Epoch: [22][ 1190/ 1236] Overall Loss 0.348009 Objective Loss 0.348009 LR 0.001000 Time 0.020698 +2023-10-05 20:52:46,004 - Epoch: [22][ 1200/ 1236] Overall Loss 0.348142 Objective Loss 0.348142 LR 0.001000 Time 0.020692 +2023-10-05 20:52:46,204 - Epoch: [22][ 1210/ 1236] Overall Loss 0.348234 Objective Loss 0.348234 LR 0.001000 Time 0.020685 +2023-10-05 20:52:46,405 - Epoch: [22][ 1220/ 1236] Overall Loss 0.348257 Objective Loss 0.348257 LR 0.001000 Time 0.020680 +2023-10-05 20:52:46,657 - Epoch: [22][ 1230/ 1236] Overall Loss 0.348408 Objective Loss 0.348408 LR 0.001000 Time 0.020717 +2023-10-05 20:52:46,774 - Epoch: [22][ 1236/ 1236] Overall Loss 0.348422 Objective Loss 0.348422 Top1 82.688391 Top5 96.741344 LR 0.001000 Time 0.020711 +2023-10-05 20:52:46,895 - --- validate (epoch=22)----------- +2023-10-05 20:52:46,896 - 29943 samples (256 per mini-batch) +2023-10-05 20:52:47,356 - Epoch: [22][ 10/ 117] Loss 0.376143 Top1 80.312500 Top5 97.226562 +2023-10-05 20:52:47,508 - Epoch: [22][ 20/ 117] Loss 0.375844 Top1 79.804688 Top5 97.226562 +2023-10-05 20:52:47,657 - Epoch: [22][ 30/ 117] Loss 0.378896 Top1 80.065104 Top5 97.304688 +2023-10-05 20:52:47,808 - Epoch: [22][ 40/ 117] Loss 0.372862 Top1 79.960938 Top5 97.246094 +2023-10-05 20:52:47,957 - Epoch: [22][ 50/ 117] Loss 0.373535 Top1 79.750000 Top5 97.273438 +2023-10-05 20:52:48,108 - Epoch: [22][ 60/ 117] Loss 0.373051 Top1 79.746094 Top5 97.278646 +2023-10-05 20:52:48,257 - Epoch: [22][ 70/ 117] Loss 0.367825 Top1 79.949777 Top5 97.349330 +2023-10-05 20:52:48,406 - Epoch: [22][ 80/ 117] Loss 0.368150 Top1 80.004883 Top5 97.285156 +2023-10-05 20:52:48,556 - Epoch: [22][ 90/ 117] Loss 0.368346 Top1 80.112847 Top5 97.287326 +2023-10-05 20:52:48,704 - Epoch: [22][ 100/ 117] Loss 0.367524 Top1 80.160156 Top5 97.304688 +2023-10-05 20:52:48,861 - Epoch: [22][ 110/ 117] Loss 0.369808 Top1 80.078125 Top5 97.340199 +2023-10-05 20:52:48,947 - Epoch: [22][ 117/ 117] Loss 0.370230 Top1 80.028721 Top5 97.368333 +2023-10-05 20:52:49,086 - ==> Top1: 80.029 Top5: 97.368 Loss: 0.370 + +2023-10-05 20:52:49,087 - ==> Confusion: +[[ 953 1 3 0 5 4 0 0 6 41 1 0 1 7 6 2 6 3 0 0 11] + [ 2 1018 2 1 4 35 1 25 2 0 1 1 0 1 1 4 12 0 16 2 3] + [ 8 1 921 25 0 0 28 7 1 1 5 4 4 3 3 2 4 5 12 6 16] + [ 4 1 16 945 0 5 2 1 9 1 8 0 7 3 33 5 2 13 21 0 13] + [ 33 10 1 0 955 6 0 2 0 9 1 1 1 2 5 4 12 3 0 0 5] + [ 8 37 1 3 3 967 3 32 0 0 4 11 2 20 5 1 8 0 4 1 6] + [ 1 9 35 0 0 2 1111 8 0 0 5 0 2 1 0 4 1 1 0 5 6] + [ 7 19 16 3 0 44 7 1014 1 2 5 9 4 0 0 1 0 1 62 11 12] + [ 25 4 0 0 1 4 0 1 960 41 12 1 1 18 11 3 3 0 3 0 1] + [ 161 1 0 0 4 3 0 0 27 856 0 1 1 32 9 6 2 1 0 3 12] + [ 6 3 9 6 1 0 4 3 6 1 970 2 0 11 8 3 1 2 7 0 10] + [ 2 1 2 0 0 14 1 2 0 0 0 960 13 7 0 2 1 21 1 5 3] + [ 0 0 6 5 0 3 1 0 0 0 1 43 943 4 3 10 5 23 5 7 9] + [ 0 0 0 1 2 17 1 0 9 10 14 7 1 1034 5 2 1 0 0 2 13] + [ 21 2 0 15 2 0 0 0 23 6 4 0 3 0 995 0 1 4 9 0 16] + [ 0 2 2 3 2 0 0 2 0 1 0 8 11 1 0 1058 17 9 0 8 10] + [ 4 11 1 2 6 7 1 0 3 0 0 5 2 3 4 17 1083 0 1 3 8] + [ 0 0 0 5 0 0 3 0 0 1 1 6 28 1 0 1 2 987 1 0 2] + [ 2 4 6 15 2 1 1 23 6 0 3 3 2 0 16 0 0 0 973 2 9] + [ 2 0 0 1 1 10 11 11 0 0 0 20 4 2 0 5 10 2 0 1062 11] + [ 269 266 163 106 135 214 64 97 94 77 214 168 433 367 172 80 213 114 183 278 4198]] + +2023-10-05 20:52:49,088 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:52:49,088 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:52:49,094 - + +2023-10-05 20:52:49,094 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:52:50,093 - Epoch: [23][ 10/ 1236] Overall Loss 0.296769 Objective Loss 0.296769 LR 0.001000 Time 0.099828 +2023-10-05 20:52:50,295 - Epoch: [23][ 20/ 1236] Overall Loss 0.315602 Objective Loss 0.315602 LR 0.001000 Time 0.059977 +2023-10-05 20:52:50,494 - Epoch: [23][ 30/ 1236] Overall Loss 0.327254 Objective Loss 0.327254 LR 0.001000 Time 0.046633 +2023-10-05 20:52:50,696 - Epoch: [23][ 40/ 1236] Overall Loss 0.331384 Objective Loss 0.331384 LR 0.001000 Time 0.040009 +2023-10-05 20:52:50,896 - Epoch: [23][ 50/ 1236] Overall Loss 0.336084 Objective Loss 0.336084 LR 0.001000 Time 0.036000 +2023-10-05 20:52:51,098 - Epoch: [23][ 60/ 1236] Overall Loss 0.330464 Objective Loss 0.330464 LR 0.001000 Time 0.033353 +2023-10-05 20:52:51,298 - Epoch: [23][ 70/ 1236] Overall Loss 0.332297 Objective Loss 0.332297 LR 0.001000 Time 0.031440 +2023-10-05 20:52:51,499 - Epoch: [23][ 80/ 1236] Overall Loss 0.334367 Objective Loss 0.334367 LR 0.001000 Time 0.030024 +2023-10-05 20:52:51,699 - Epoch: [23][ 90/ 1236] Overall Loss 0.334098 Objective Loss 0.334098 LR 0.001000 Time 0.028904 +2023-10-05 20:52:51,900 - Epoch: [23][ 100/ 1236] Overall Loss 0.333737 Objective Loss 0.333737 LR 0.001000 Time 0.028025 +2023-10-05 20:52:52,106 - Epoch: [23][ 110/ 1236] Overall Loss 0.331745 Objective Loss 0.331745 LR 0.001000 Time 0.027342 +2023-10-05 20:52:52,313 - Epoch: [23][ 120/ 1236] Overall Loss 0.333926 Objective Loss 0.333926 LR 0.001000 Time 0.026783 +2023-10-05 20:52:52,515 - Epoch: [23][ 130/ 1236] Overall Loss 0.334237 Objective Loss 0.334237 LR 0.001000 Time 0.026278 +2023-10-05 20:52:52,722 - Epoch: [23][ 140/ 1236] Overall Loss 0.332383 Objective Loss 0.332383 LR 0.001000 Time 0.025874 +2023-10-05 20:52:52,928 - Epoch: [23][ 150/ 1236] Overall Loss 0.333446 Objective Loss 0.333446 LR 0.001000 Time 0.025516 +2023-10-05 20:52:53,134 - Epoch: [23][ 160/ 1236] Overall Loss 0.334220 Objective Loss 0.334220 LR 0.001000 Time 0.025211 +2023-10-05 20:52:53,340 - Epoch: [23][ 170/ 1236] Overall Loss 0.333490 Objective Loss 0.333490 LR 0.001000 Time 0.024935 +2023-10-05 20:52:53,547 - Epoch: [23][ 180/ 1236] Overall Loss 0.335931 Objective Loss 0.335931 LR 0.001000 Time 0.024696 +2023-10-05 20:52:53,753 - Epoch: [23][ 190/ 1236] Overall Loss 0.335250 Objective Loss 0.335250 LR 0.001000 Time 0.024477 +2023-10-05 20:52:53,960 - Epoch: [23][ 200/ 1236] Overall Loss 0.334142 Objective Loss 0.334142 LR 0.001000 Time 0.024285 +2023-10-05 20:52:54,165 - Epoch: [23][ 210/ 1236] Overall Loss 0.334061 Objective Loss 0.334061 LR 0.001000 Time 0.024106 +2023-10-05 20:52:54,372 - Epoch: [23][ 220/ 1236] Overall Loss 0.333683 Objective Loss 0.333683 LR 0.001000 Time 0.023948 +2023-10-05 20:52:54,578 - Epoch: [23][ 230/ 1236] Overall Loss 0.333548 Objective Loss 0.333548 LR 0.001000 Time 0.023799 +2023-10-05 20:52:54,785 - Epoch: [23][ 240/ 1236] Overall Loss 0.333871 Objective Loss 0.333871 LR 0.001000 Time 0.023667 +2023-10-05 20:52:54,990 - Epoch: [23][ 250/ 1236] Overall Loss 0.333679 Objective Loss 0.333679 LR 0.001000 Time 0.023540 +2023-10-05 20:52:55,197 - Epoch: [23][ 260/ 1236] Overall Loss 0.334244 Objective Loss 0.334244 LR 0.001000 Time 0.023428 +2023-10-05 20:52:55,403 - Epoch: [23][ 270/ 1236] Overall Loss 0.334734 Objective Loss 0.334734 LR 0.001000 Time 0.023321 +2023-10-05 20:52:55,610 - Epoch: [23][ 280/ 1236] Overall Loss 0.335350 Objective Loss 0.335350 LR 0.001000 Time 0.023225 +2023-10-05 20:52:55,815 - Epoch: [23][ 290/ 1236] Overall Loss 0.334451 Objective Loss 0.334451 LR 0.001000 Time 0.023131 +2023-10-05 20:52:56,022 - Epoch: [23][ 300/ 1236] Overall Loss 0.335592 Objective Loss 0.335592 LR 0.001000 Time 0.023048 +2023-10-05 20:52:56,228 - Epoch: [23][ 310/ 1236] Overall Loss 0.335520 Objective Loss 0.335520 LR 0.001000 Time 0.022966 +2023-10-05 20:52:56,435 - Epoch: [23][ 320/ 1236] Overall Loss 0.336386 Objective Loss 0.336386 LR 0.001000 Time 0.022893 +2023-10-05 20:52:56,641 - Epoch: [23][ 330/ 1236] Overall Loss 0.336181 Objective Loss 0.336181 LR 0.001000 Time 0.022823 +2023-10-05 20:52:56,848 - Epoch: [23][ 340/ 1236] Overall Loss 0.337352 Objective Loss 0.337352 LR 0.001000 Time 0.022758 +2023-10-05 20:52:57,053 - Epoch: [23][ 350/ 1236] Overall Loss 0.338290 Objective Loss 0.338290 LR 0.001000 Time 0.022695 +2023-10-05 20:52:57,260 - Epoch: [23][ 360/ 1236] Overall Loss 0.338742 Objective Loss 0.338742 LR 0.001000 Time 0.022638 +2023-10-05 20:52:57,463 - Epoch: [23][ 370/ 1236] Overall Loss 0.338953 Objective Loss 0.338953 LR 0.001000 Time 0.022572 +2023-10-05 20:52:57,665 - Epoch: [23][ 380/ 1236] Overall Loss 0.340608 Objective Loss 0.340608 LR 0.001000 Time 0.022509 +2023-10-05 20:52:57,865 - Epoch: [23][ 390/ 1236] Overall Loss 0.340528 Objective Loss 0.340528 LR 0.001000 Time 0.022445 +2023-10-05 20:52:58,067 - Epoch: [23][ 400/ 1236] Overall Loss 0.340798 Objective Loss 0.340798 LR 0.001000 Time 0.022388 +2023-10-05 20:52:58,267 - Epoch: [23][ 410/ 1236] Overall Loss 0.340992 Objective Loss 0.340992 LR 0.001000 Time 0.022329 +2023-10-05 20:52:58,470 - Epoch: [23][ 420/ 1236] Overall Loss 0.340831 Objective Loss 0.340831 LR 0.001000 Time 0.022279 +2023-10-05 20:52:58,670 - Epoch: [23][ 430/ 1236] Overall Loss 0.340065 Objective Loss 0.340065 LR 0.001000 Time 0.022225 +2023-10-05 20:52:58,872 - Epoch: [23][ 440/ 1236] Overall Loss 0.339677 Objective Loss 0.339677 LR 0.001000 Time 0.022178 +2023-10-05 20:52:59,072 - Epoch: [23][ 450/ 1236] Overall Loss 0.339830 Objective Loss 0.339830 LR 0.001000 Time 0.022130 +2023-10-05 20:52:59,274 - Epoch: [23][ 460/ 1236] Overall Loss 0.339860 Objective Loss 0.339860 LR 0.001000 Time 0.022086 +2023-10-05 20:52:59,474 - Epoch: [23][ 470/ 1236] Overall Loss 0.339426 Objective Loss 0.339426 LR 0.001000 Time 0.022042 +2023-10-05 20:52:59,676 - Epoch: [23][ 480/ 1236] Overall Loss 0.339134 Objective Loss 0.339134 LR 0.001000 Time 0.022003 +2023-10-05 20:52:59,877 - Epoch: [23][ 490/ 1236] Overall Loss 0.338803 Objective Loss 0.338803 LR 0.001000 Time 0.021962 +2023-10-05 20:53:00,079 - Epoch: [23][ 500/ 1236] Overall Loss 0.338353 Objective Loss 0.338353 LR 0.001000 Time 0.021926 +2023-10-05 20:53:00,279 - Epoch: [23][ 510/ 1236] Overall Loss 0.339385 Objective Loss 0.339385 LR 0.001000 Time 0.021888 +2023-10-05 20:53:00,481 - Epoch: [23][ 520/ 1236] Overall Loss 0.339046 Objective Loss 0.339046 LR 0.001000 Time 0.021855 +2023-10-05 20:53:00,681 - Epoch: [23][ 530/ 1236] Overall Loss 0.339337 Objective Loss 0.339337 LR 0.001000 Time 0.021820 +2023-10-05 20:53:00,883 - Epoch: [23][ 540/ 1236] Overall Loss 0.339449 Objective Loss 0.339449 LR 0.001000 Time 0.021789 +2023-10-05 20:53:01,083 - Epoch: [23][ 550/ 1236] Overall Loss 0.339478 Objective Loss 0.339478 LR 0.001000 Time 0.021756 +2023-10-05 20:53:01,285 - Epoch: [23][ 560/ 1236] Overall Loss 0.339964 Objective Loss 0.339964 LR 0.001000 Time 0.021728 +2023-10-05 20:53:01,486 - Epoch: [23][ 570/ 1236] Overall Loss 0.340048 Objective Loss 0.340048 LR 0.001000 Time 0.021698 +2023-10-05 20:53:01,688 - Epoch: [23][ 580/ 1236] Overall Loss 0.339864 Objective Loss 0.339864 LR 0.001000 Time 0.021671 +2023-10-05 20:53:01,888 - Epoch: [23][ 590/ 1236] Overall Loss 0.339705 Objective Loss 0.339705 LR 0.001000 Time 0.021643 +2023-10-05 20:53:02,090 - Epoch: [23][ 600/ 1236] Overall Loss 0.339445 Objective Loss 0.339445 LR 0.001000 Time 0.021618 +2023-10-05 20:53:02,290 - Epoch: [23][ 610/ 1236] Overall Loss 0.339628 Objective Loss 0.339628 LR 0.001000 Time 0.021591 +2023-10-05 20:53:02,492 - Epoch: [23][ 620/ 1236] Overall Loss 0.339597 Objective Loss 0.339597 LR 0.001000 Time 0.021569 +2023-10-05 20:53:02,692 - Epoch: [23][ 630/ 1236] Overall Loss 0.339963 Objective Loss 0.339963 LR 0.001000 Time 0.021543 +2023-10-05 20:53:02,894 - Epoch: [23][ 640/ 1236] Overall Loss 0.340753 Objective Loss 0.340753 LR 0.001000 Time 0.021522 +2023-10-05 20:53:03,095 - Epoch: [23][ 650/ 1236] Overall Loss 0.341151 Objective Loss 0.341151 LR 0.001000 Time 0.021499 +2023-10-05 20:53:03,297 - Epoch: [23][ 660/ 1236] Overall Loss 0.340843 Objective Loss 0.340843 LR 0.001000 Time 0.021478 +2023-10-05 20:53:03,497 - Epoch: [23][ 670/ 1236] Overall Loss 0.341169 Objective Loss 0.341169 LR 0.001000 Time 0.021456 +2023-10-05 20:53:03,699 - Epoch: [23][ 680/ 1236] Overall Loss 0.341640 Objective Loss 0.341640 LR 0.001000 Time 0.021437 +2023-10-05 20:53:03,899 - Epoch: [23][ 690/ 1236] Overall Loss 0.341690 Objective Loss 0.341690 LR 0.001000 Time 0.021416 +2023-10-05 20:53:04,101 - Epoch: [23][ 700/ 1236] Overall Loss 0.342070 Objective Loss 0.342070 LR 0.001000 Time 0.021399 +2023-10-05 20:53:04,302 - Epoch: [23][ 710/ 1236] Overall Loss 0.341786 Objective Loss 0.341786 LR 0.001000 Time 0.021379 +2023-10-05 20:53:04,504 - Epoch: [23][ 720/ 1236] Overall Loss 0.341952 Objective Loss 0.341952 LR 0.001000 Time 0.021362 +2023-10-05 20:53:04,704 - Epoch: [23][ 730/ 1236] Overall Loss 0.342454 Objective Loss 0.342454 LR 0.001000 Time 0.021343 +2023-10-05 20:53:04,906 - Epoch: [23][ 740/ 1236] Overall Loss 0.342586 Objective Loss 0.342586 LR 0.001000 Time 0.021327 +2023-10-05 20:53:05,106 - Epoch: [23][ 750/ 1236] Overall Loss 0.343094 Objective Loss 0.343094 LR 0.001000 Time 0.021309 +2023-10-05 20:53:05,308 - Epoch: [23][ 760/ 1236] Overall Loss 0.342998 Objective Loss 0.342998 LR 0.001000 Time 0.021294 +2023-10-05 20:53:05,509 - Epoch: [23][ 770/ 1236] Overall Loss 0.343657 Objective Loss 0.343657 LR 0.001000 Time 0.021278 +2023-10-05 20:53:05,711 - Epoch: [23][ 780/ 1236] Overall Loss 0.343935 Objective Loss 0.343935 LR 0.001000 Time 0.021263 +2023-10-05 20:53:05,911 - Epoch: [23][ 790/ 1236] Overall Loss 0.344140 Objective Loss 0.344140 LR 0.001000 Time 0.021247 +2023-10-05 20:53:06,113 - Epoch: [23][ 800/ 1236] Overall Loss 0.344006 Objective Loss 0.344006 LR 0.001000 Time 0.021234 +2023-10-05 20:53:06,313 - Epoch: [23][ 810/ 1236] Overall Loss 0.343974 Objective Loss 0.343974 LR 0.001000 Time 0.021218 +2023-10-05 20:53:06,515 - Epoch: [23][ 820/ 1236] Overall Loss 0.343697 Objective Loss 0.343697 LR 0.001000 Time 0.021206 +2023-10-05 20:53:06,715 - Epoch: [23][ 830/ 1236] Overall Loss 0.343997 Objective Loss 0.343997 LR 0.001000 Time 0.021191 +2023-10-05 20:53:06,917 - Epoch: [23][ 840/ 1236] Overall Loss 0.344016 Objective Loss 0.344016 LR 0.001000 Time 0.021179 +2023-10-05 20:53:07,118 - Epoch: [23][ 850/ 1236] Overall Loss 0.343754 Objective Loss 0.343754 LR 0.001000 Time 0.021165 +2023-10-05 20:53:07,320 - Epoch: [23][ 860/ 1236] Overall Loss 0.343888 Objective Loss 0.343888 LR 0.001000 Time 0.021153 +2023-10-05 20:53:07,520 - Epoch: [23][ 870/ 1236] Overall Loss 0.344302 Objective Loss 0.344302 LR 0.001000 Time 0.021140 +2023-10-05 20:53:07,722 - Epoch: [23][ 880/ 1236] Overall Loss 0.344195 Objective Loss 0.344195 LR 0.001000 Time 0.021129 +2023-10-05 20:53:07,922 - Epoch: [23][ 890/ 1236] Overall Loss 0.344206 Objective Loss 0.344206 LR 0.001000 Time 0.021116 +2023-10-05 20:53:08,124 - Epoch: [23][ 900/ 1236] Overall Loss 0.343876 Objective Loss 0.343876 LR 0.001000 Time 0.021105 +2023-10-05 20:53:08,325 - Epoch: [23][ 910/ 1236] Overall Loss 0.345467 Objective Loss 0.345467 LR 0.001000 Time 0.021093 +2023-10-05 20:53:08,527 - Epoch: [23][ 920/ 1236] Overall Loss 0.347219 Objective Loss 0.347219 LR 0.001000 Time 0.021084 +2023-10-05 20:53:08,727 - Epoch: [23][ 930/ 1236] Overall Loss 0.348724 Objective Loss 0.348724 LR 0.001000 Time 0.021072 +2023-10-05 20:53:08,929 - Epoch: [23][ 940/ 1236] Overall Loss 0.349739 Objective Loss 0.349739 LR 0.001000 Time 0.021063 +2023-10-05 20:53:09,130 - Epoch: [23][ 950/ 1236] Overall Loss 0.350894 Objective Loss 0.350894 LR 0.001000 Time 0.021051 +2023-10-05 20:53:09,332 - Epoch: [23][ 960/ 1236] Overall Loss 0.351900 Objective Loss 0.351900 LR 0.001000 Time 0.021042 +2023-10-05 20:53:09,532 - Epoch: [23][ 970/ 1236] Overall Loss 0.352799 Objective Loss 0.352799 LR 0.001000 Time 0.021031 +2023-10-05 20:53:09,734 - Epoch: [23][ 980/ 1236] Overall Loss 0.353540 Objective Loss 0.353540 LR 0.001000 Time 0.021022 +2023-10-05 20:53:09,934 - Epoch: [23][ 990/ 1236] Overall Loss 0.354583 Objective Loss 0.354583 LR 0.001000 Time 0.021012 +2023-10-05 20:53:10,136 - Epoch: [23][ 1000/ 1236] Overall Loss 0.355311 Objective Loss 0.355311 LR 0.001000 Time 0.021004 +2023-10-05 20:53:10,337 - Epoch: [23][ 1010/ 1236] Overall Loss 0.356180 Objective Loss 0.356180 LR 0.001000 Time 0.020994 +2023-10-05 20:53:10,539 - Epoch: [23][ 1020/ 1236] Overall Loss 0.356878 Objective Loss 0.356878 LR 0.001000 Time 0.020986 +2023-10-05 20:53:10,739 - Epoch: [23][ 1030/ 1236] Overall Loss 0.357527 Objective Loss 0.357527 LR 0.001000 Time 0.020976 +2023-10-05 20:53:10,941 - Epoch: [23][ 1040/ 1236] Overall Loss 0.357835 Objective Loss 0.357835 LR 0.001000 Time 0.020969 +2023-10-05 20:53:11,142 - Epoch: [23][ 1050/ 1236] Overall Loss 0.358270 Objective Loss 0.358270 LR 0.001000 Time 0.020959 +2023-10-05 20:53:11,344 - Epoch: [23][ 1060/ 1236] Overall Loss 0.358679 Objective Loss 0.358679 LR 0.001000 Time 0.020952 +2023-10-05 20:53:11,544 - Epoch: [23][ 1070/ 1236] Overall Loss 0.359417 Objective Loss 0.359417 LR 0.001000 Time 0.020943 +2023-10-05 20:53:11,746 - Epoch: [23][ 1080/ 1236] Overall Loss 0.359749 Objective Loss 0.359749 LR 0.001000 Time 0.020936 +2023-10-05 20:53:11,946 - Epoch: [23][ 1090/ 1236] Overall Loss 0.360547 Objective Loss 0.360547 LR 0.001000 Time 0.020927 +2023-10-05 20:53:12,148 - Epoch: [23][ 1100/ 1236] Overall Loss 0.360984 Objective Loss 0.360984 LR 0.001000 Time 0.020920 +2023-10-05 20:53:12,348 - Epoch: [23][ 1110/ 1236] Overall Loss 0.361403 Objective Loss 0.361403 LR 0.001000 Time 0.020912 +2023-10-05 20:53:12,550 - Epoch: [23][ 1120/ 1236] Overall Loss 0.361616 Objective Loss 0.361616 LR 0.001000 Time 0.020905 +2023-10-05 20:53:12,751 - Epoch: [23][ 1130/ 1236] Overall Loss 0.361941 Objective Loss 0.361941 LR 0.001000 Time 0.020897 +2023-10-05 20:53:12,953 - Epoch: [23][ 1140/ 1236] Overall Loss 0.362483 Objective Loss 0.362483 LR 0.001000 Time 0.020891 +2023-10-05 20:53:13,153 - Epoch: [23][ 1150/ 1236] Overall Loss 0.362830 Objective Loss 0.362830 LR 0.001000 Time 0.020883 +2023-10-05 20:53:13,355 - Epoch: [23][ 1160/ 1236] Overall Loss 0.363085 Objective Loss 0.363085 LR 0.001000 Time 0.020877 +2023-10-05 20:53:13,555 - Epoch: [23][ 1170/ 1236] Overall Loss 0.363160 Objective Loss 0.363160 LR 0.001000 Time 0.020869 +2023-10-05 20:53:13,758 - Epoch: [23][ 1180/ 1236] Overall Loss 0.363306 Objective Loss 0.363306 LR 0.001000 Time 0.020863 +2023-10-05 20:53:13,958 - Epoch: [23][ 1190/ 1236] Overall Loss 0.363427 Objective Loss 0.363427 LR 0.001000 Time 0.020856 +2023-10-05 20:53:14,160 - Epoch: [23][ 1200/ 1236] Overall Loss 0.363628 Objective Loss 0.363628 LR 0.001000 Time 0.020850 +2023-10-05 20:53:14,361 - Epoch: [23][ 1210/ 1236] Overall Loss 0.363857 Objective Loss 0.363857 LR 0.001000 Time 0.020844 +2023-10-05 20:53:14,563 - Epoch: [23][ 1220/ 1236] Overall Loss 0.363919 Objective Loss 0.363919 LR 0.001000 Time 0.020838 +2023-10-05 20:53:14,818 - Epoch: [23][ 1230/ 1236] Overall Loss 0.364210 Objective Loss 0.364210 LR 0.001000 Time 0.020876 +2023-10-05 20:53:14,936 - Epoch: [23][ 1236/ 1236] Overall Loss 0.364246 Objective Loss 0.364246 Top1 80.244399 Top5 97.759674 LR 0.001000 Time 0.020870 +2023-10-05 20:53:15,064 - --- validate (epoch=23)----------- +2023-10-05 20:53:15,064 - 29943 samples (256 per mini-batch) +2023-10-05 20:53:15,517 - Epoch: [23][ 10/ 117] Loss 0.395013 Top1 78.632812 Top5 97.421875 +2023-10-05 20:53:15,668 - Epoch: [23][ 20/ 117] Loss 0.400459 Top1 78.593750 Top5 96.992188 +2023-10-05 20:53:15,821 - Epoch: [23][ 30/ 117] Loss 0.394697 Top1 78.776042 Top5 97.122396 +2023-10-05 20:53:15,971 - Epoch: [23][ 40/ 117] Loss 0.406082 Top1 78.574219 Top5 97.236328 +2023-10-05 20:53:16,123 - Epoch: [23][ 50/ 117] Loss 0.402958 Top1 78.906250 Top5 97.218750 +2023-10-05 20:53:16,272 - Epoch: [23][ 60/ 117] Loss 0.408292 Top1 78.802083 Top5 97.141927 +2023-10-05 20:53:16,422 - Epoch: [23][ 70/ 117] Loss 0.407049 Top1 78.766741 Top5 97.176339 +2023-10-05 20:53:16,571 - Epoch: [23][ 80/ 117] Loss 0.403765 Top1 78.774414 Top5 97.211914 +2023-10-05 20:53:16,719 - Epoch: [23][ 90/ 117] Loss 0.405227 Top1 78.710938 Top5 97.200521 +2023-10-05 20:53:16,869 - Epoch: [23][ 100/ 117] Loss 0.405344 Top1 78.753906 Top5 97.203125 +2023-10-05 20:53:17,025 - Epoch: [23][ 110/ 117] Loss 0.402880 Top1 78.792614 Top5 97.233665 +2023-10-05 20:53:17,111 - Epoch: [23][ 117/ 117] Loss 0.401240 Top1 78.879872 Top5 97.231406 +2023-10-05 20:53:17,220 - ==> Top1: 78.880 Top5: 97.231 Loss: 0.401 + +2023-10-05 20:53:17,221 - ==> Confusion: +[[ 955 1 5 1 4 2 0 0 10 36 1 0 0 6 7 2 6 0 0 1 13] + [ 4 1024 3 2 9 22 2 23 4 0 3 2 0 0 1 4 14 0 8 1 5] + [ 10 0 932 21 0 0 33 8 0 2 2 7 5 4 4 3 7 1 4 6 7] + [ 5 0 15 959 0 1 1 1 8 0 4 0 5 5 35 3 4 11 17 3 12] + [ 44 7 1 0 952 3 1 0 3 2 2 3 0 4 5 5 12 1 1 2 2] + [ 13 35 0 2 4 963 4 20 4 0 6 11 2 22 5 2 10 0 2 3 8] + [ 0 8 25 0 0 1 1118 8 0 0 2 2 1 0 0 6 2 2 1 10 5] + [ 13 17 21 0 2 28 3 1044 0 4 4 8 0 2 0 2 0 0 48 13 9] + [ 30 3 1 0 1 1 0 0 987 24 3 2 0 19 9 3 2 0 1 2 1] + [ 176 0 0 0 3 1 2 1 44 837 0 1 0 31 4 4 2 1 0 3 9] + [ 6 3 12 7 0 2 2 4 14 2 953 3 0 19 6 1 1 0 7 2 9] + [ 2 0 1 0 0 15 0 1 0 0 0 956 18 6 0 3 2 19 1 7 4] + [ 1 0 3 7 0 2 2 0 0 1 1 50 932 3 2 11 7 33 0 6 7] + [ 3 1 1 3 3 4 0 0 13 11 7 6 1 1051 2 1 1 2 0 2 7] + [ 20 0 0 11 1 0 0 0 29 0 1 0 3 2 1000 0 2 5 7 0 20] + [ 1 5 4 2 1 1 0 0 0 0 1 7 8 2 0 1058 14 16 0 12 2] + [ 3 5 1 1 7 5 0 0 1 0 0 4 0 3 3 11 1107 0 0 4 6] + [ 0 0 1 1 0 0 0 0 1 1 0 7 8 0 1 3 1 1011 0 1 2] + [ 1 5 11 21 1 0 4 30 12 0 1 1 1 0 14 1 2 0 955 2 6] + [ 1 2 1 2 2 6 10 12 0 0 1 12 3 0 0 6 8 4 1 1075 6] + [ 271 222 232 98 147 182 77 79 150 88 229 186 459 459 222 101 283 135 212 323 3750]] + +2023-10-05 20:53:17,222 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:53:17,222 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:53:17,228 - + +2023-10-05 20:53:17,228 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:53:18,224 - Epoch: [24][ 10/ 1236] Overall Loss 0.384894 Objective Loss 0.384894 LR 0.001000 Time 0.099588 +2023-10-05 20:53:18,426 - Epoch: [24][ 20/ 1236] Overall Loss 0.385989 Objective Loss 0.385989 LR 0.001000 Time 0.059865 +2023-10-05 20:53:18,626 - Epoch: [24][ 30/ 1236] Overall Loss 0.382865 Objective Loss 0.382865 LR 0.001000 Time 0.046555 +2023-10-05 20:53:18,827 - Epoch: [24][ 40/ 1236] Overall Loss 0.376458 Objective Loss 0.376458 LR 0.001000 Time 0.039941 +2023-10-05 20:53:19,027 - Epoch: [24][ 50/ 1236] Overall Loss 0.373917 Objective Loss 0.373917 LR 0.001000 Time 0.035944 +2023-10-05 20:53:19,228 - Epoch: [24][ 60/ 1236] Overall Loss 0.374242 Objective Loss 0.374242 LR 0.001000 Time 0.033305 +2023-10-05 20:53:19,429 - Epoch: [24][ 70/ 1236] Overall Loss 0.373580 Objective Loss 0.373580 LR 0.001000 Time 0.031406 +2023-10-05 20:53:19,630 - Epoch: [24][ 80/ 1236] Overall Loss 0.371382 Objective Loss 0.371382 LR 0.001000 Time 0.029991 +2023-10-05 20:53:19,830 - Epoch: [24][ 90/ 1236] Overall Loss 0.367529 Objective Loss 0.367529 LR 0.001000 Time 0.028874 +2023-10-05 20:53:20,031 - Epoch: [24][ 100/ 1236] Overall Loss 0.364944 Objective Loss 0.364944 LR 0.001000 Time 0.027994 +2023-10-05 20:53:20,230 - Epoch: [24][ 110/ 1236] Overall Loss 0.363599 Objective Loss 0.363599 LR 0.001000 Time 0.027254 +2023-10-05 20:53:20,431 - Epoch: [24][ 120/ 1236] Overall Loss 0.365313 Objective Loss 0.365313 LR 0.001000 Time 0.026653 +2023-10-05 20:53:20,629 - Epoch: [24][ 130/ 1236] Overall Loss 0.366396 Objective Loss 0.366396 LR 0.001000 Time 0.026126 +2023-10-05 20:53:20,830 - Epoch: [24][ 140/ 1236] Overall Loss 0.367572 Objective Loss 0.367572 LR 0.001000 Time 0.025690 +2023-10-05 20:53:21,028 - Epoch: [24][ 150/ 1236] Overall Loss 0.366396 Objective Loss 0.366396 LR 0.001000 Time 0.025298 +2023-10-05 20:53:21,229 - Epoch: [24][ 160/ 1236] Overall Loss 0.366139 Objective Loss 0.366139 LR 0.001000 Time 0.024968 +2023-10-05 20:53:21,427 - Epoch: [24][ 170/ 1236] Overall Loss 0.365757 Objective Loss 0.365757 LR 0.001000 Time 0.024666 +2023-10-05 20:53:21,628 - Epoch: [24][ 180/ 1236] Overall Loss 0.366292 Objective Loss 0.366292 LR 0.001000 Time 0.024408 +2023-10-05 20:53:21,826 - Epoch: [24][ 190/ 1236] Overall Loss 0.367670 Objective Loss 0.367670 LR 0.001000 Time 0.024167 +2023-10-05 20:53:22,027 - Epoch: [24][ 200/ 1236] Overall Loss 0.367138 Objective Loss 0.367138 LR 0.001000 Time 0.023959 +2023-10-05 20:53:22,225 - Epoch: [24][ 210/ 1236] Overall Loss 0.368021 Objective Loss 0.368021 LR 0.001000 Time 0.023762 +2023-10-05 20:53:22,426 - Epoch: [24][ 220/ 1236] Overall Loss 0.369187 Objective Loss 0.369187 LR 0.001000 Time 0.023593 +2023-10-05 20:53:22,625 - Epoch: [24][ 230/ 1236] Overall Loss 0.369115 Objective Loss 0.369115 LR 0.001000 Time 0.023428 +2023-10-05 20:53:22,825 - Epoch: [24][ 240/ 1236] Overall Loss 0.370193 Objective Loss 0.370193 LR 0.001000 Time 0.023286 +2023-10-05 20:53:23,024 - Epoch: [24][ 250/ 1236] Overall Loss 0.369445 Objective Loss 0.369445 LR 0.001000 Time 0.023148 +2023-10-05 20:53:23,224 - Epoch: [24][ 260/ 1236] Overall Loss 0.368729 Objective Loss 0.368729 LR 0.001000 Time 0.023028 +2023-10-05 20:53:23,423 - Epoch: [24][ 270/ 1236] Overall Loss 0.368092 Objective Loss 0.368092 LR 0.001000 Time 0.022910 +2023-10-05 20:53:23,622 - Epoch: [24][ 280/ 1236] Overall Loss 0.368184 Objective Loss 0.368184 LR 0.001000 Time 0.022800 +2023-10-05 20:53:23,821 - Epoch: [24][ 290/ 1236] Overall Loss 0.367455 Objective Loss 0.367455 LR 0.001000 Time 0.022699 +2023-10-05 20:53:24,021 - Epoch: [24][ 300/ 1236] Overall Loss 0.367903 Objective Loss 0.367903 LR 0.001000 Time 0.022610 +2023-10-05 20:53:24,220 - Epoch: [24][ 310/ 1236] Overall Loss 0.368810 Objective Loss 0.368810 LR 0.001000 Time 0.022520 +2023-10-05 20:53:24,421 - Epoch: [24][ 320/ 1236] Overall Loss 0.367629 Objective Loss 0.367629 LR 0.001000 Time 0.022442 +2023-10-05 20:53:24,619 - Epoch: [24][ 330/ 1236] Overall Loss 0.367534 Objective Loss 0.367534 LR 0.001000 Time 0.022363 +2023-10-05 20:53:24,820 - Epoch: [24][ 340/ 1236] Overall Loss 0.366957 Objective Loss 0.366957 LR 0.001000 Time 0.022295 +2023-10-05 20:53:25,019 - Epoch: [24][ 350/ 1236] Overall Loss 0.366834 Objective Loss 0.366834 LR 0.001000 Time 0.022224 +2023-10-05 20:53:25,220 - Epoch: [24][ 360/ 1236] Overall Loss 0.366569 Objective Loss 0.366569 LR 0.001000 Time 0.022164 +2023-10-05 20:53:25,419 - Epoch: [24][ 370/ 1236] Overall Loss 0.366907 Objective Loss 0.366907 LR 0.001000 Time 0.022102 +2023-10-05 20:53:25,621 - Epoch: [24][ 380/ 1236] Overall Loss 0.365936 Objective Loss 0.365936 LR 0.001000 Time 0.022051 +2023-10-05 20:53:25,821 - Epoch: [24][ 390/ 1236] Overall Loss 0.366423 Objective Loss 0.366423 LR 0.001000 Time 0.021998 +2023-10-05 20:53:26,023 - Epoch: [24][ 400/ 1236] Overall Loss 0.366520 Objective Loss 0.366520 LR 0.001000 Time 0.021952 +2023-10-05 20:53:26,223 - Epoch: [24][ 410/ 1236] Overall Loss 0.367135 Objective Loss 0.367135 LR 0.001000 Time 0.021903 +2023-10-05 20:53:26,424 - Epoch: [24][ 420/ 1236] Overall Loss 0.367242 Objective Loss 0.367242 LR 0.001000 Time 0.021861 +2023-10-05 20:53:26,624 - Epoch: [24][ 430/ 1236] Overall Loss 0.367458 Objective Loss 0.367458 LR 0.001000 Time 0.021817 +2023-10-05 20:53:26,826 - Epoch: [24][ 440/ 1236] Overall Loss 0.367945 Objective Loss 0.367945 LR 0.001000 Time 0.021780 +2023-10-05 20:53:27,026 - Epoch: [24][ 450/ 1236] Overall Loss 0.367334 Objective Loss 0.367334 LR 0.001000 Time 0.021740 +2023-10-05 20:53:27,228 - Epoch: [24][ 460/ 1236] Overall Loss 0.366971 Objective Loss 0.366971 LR 0.001000 Time 0.021705 +2023-10-05 20:53:27,429 - Epoch: [24][ 470/ 1236] Overall Loss 0.367202 Objective Loss 0.367202 LR 0.001000 Time 0.021669 +2023-10-05 20:53:27,630 - Epoch: [24][ 480/ 1236] Overall Loss 0.367804 Objective Loss 0.367804 LR 0.001000 Time 0.021637 +2023-10-05 20:53:27,831 - Epoch: [24][ 490/ 1236] Overall Loss 0.367254 Objective Loss 0.367254 LR 0.001000 Time 0.021603 +2023-10-05 20:53:28,032 - Epoch: [24][ 500/ 1236] Overall Loss 0.367234 Objective Loss 0.367234 LR 0.001000 Time 0.021574 +2023-10-05 20:53:28,232 - Epoch: [24][ 510/ 1236] Overall Loss 0.366898 Objective Loss 0.366898 LR 0.001000 Time 0.021543 +2023-10-05 20:53:28,434 - Epoch: [24][ 520/ 1236] Overall Loss 0.367116 Objective Loss 0.367116 LR 0.001000 Time 0.021516 +2023-10-05 20:53:28,634 - Epoch: [24][ 530/ 1236] Overall Loss 0.367551 Objective Loss 0.367551 LR 0.001000 Time 0.021487 +2023-10-05 20:53:28,836 - Epoch: [24][ 540/ 1236] Overall Loss 0.367222 Objective Loss 0.367222 LR 0.001000 Time 0.021462 +2023-10-05 20:53:29,036 - Epoch: [24][ 550/ 1236] Overall Loss 0.367393 Objective Loss 0.367393 LR 0.001000 Time 0.021435 +2023-10-05 20:53:29,238 - Epoch: [24][ 560/ 1236] Overall Loss 0.367632 Objective Loss 0.367632 LR 0.001000 Time 0.021412 +2023-10-05 20:53:29,438 - Epoch: [24][ 570/ 1236] Overall Loss 0.367190 Objective Loss 0.367190 LR 0.001000 Time 0.021387 +2023-10-05 20:53:29,640 - Epoch: [24][ 580/ 1236] Overall Loss 0.367462 Objective Loss 0.367462 LR 0.001000 Time 0.021366 +2023-10-05 20:53:29,840 - Epoch: [24][ 590/ 1236] Overall Loss 0.367532 Objective Loss 0.367532 LR 0.001000 Time 0.021343 +2023-10-05 20:53:30,042 - Epoch: [24][ 600/ 1236] Overall Loss 0.367464 Objective Loss 0.367464 LR 0.001000 Time 0.021322 +2023-10-05 20:53:30,242 - Epoch: [24][ 610/ 1236] Overall Loss 0.367064 Objective Loss 0.367064 LR 0.001000 Time 0.021300 +2023-10-05 20:53:30,444 - Epoch: [24][ 620/ 1236] Overall Loss 0.366959 Objective Loss 0.366959 LR 0.001000 Time 0.021282 +2023-10-05 20:53:30,644 - Epoch: [24][ 630/ 1236] Overall Loss 0.367073 Objective Loss 0.367073 LR 0.001000 Time 0.021261 +2023-10-05 20:53:30,846 - Epoch: [24][ 640/ 1236] Overall Loss 0.366824 Objective Loss 0.366824 LR 0.001000 Time 0.021244 +2023-10-05 20:53:31,046 - Epoch: [24][ 650/ 1236] Overall Loss 0.366865 Objective Loss 0.366865 LR 0.001000 Time 0.021225 +2023-10-05 20:53:31,248 - Epoch: [24][ 660/ 1236] Overall Loss 0.367120 Objective Loss 0.367120 LR 0.001000 Time 0.021208 +2023-10-05 20:53:31,448 - Epoch: [24][ 670/ 1236] Overall Loss 0.367022 Objective Loss 0.367022 LR 0.001000 Time 0.021190 +2023-10-05 20:53:31,650 - Epoch: [24][ 680/ 1236] Overall Loss 0.367162 Objective Loss 0.367162 LR 0.001000 Time 0.021175 +2023-10-05 20:53:31,851 - Epoch: [24][ 690/ 1236] Overall Loss 0.367219 Objective Loss 0.367219 LR 0.001000 Time 0.021158 +2023-10-05 20:53:32,053 - Epoch: [24][ 700/ 1236] Overall Loss 0.366879 Objective Loss 0.366879 LR 0.001000 Time 0.021144 +2023-10-05 20:53:32,253 - Epoch: [24][ 710/ 1236] Overall Loss 0.366745 Objective Loss 0.366745 LR 0.001000 Time 0.021128 +2023-10-05 20:53:32,455 - Epoch: [24][ 720/ 1236] Overall Loss 0.366627 Objective Loss 0.366627 LR 0.001000 Time 0.021114 +2023-10-05 20:53:32,655 - Epoch: [24][ 730/ 1236] Overall Loss 0.366892 Objective Loss 0.366892 LR 0.001000 Time 0.021099 +2023-10-05 20:53:32,857 - Epoch: [24][ 740/ 1236] Overall Loss 0.367269 Objective Loss 0.367269 LR 0.001000 Time 0.021086 +2023-10-05 20:53:33,057 - Epoch: [24][ 750/ 1236] Overall Loss 0.367295 Objective Loss 0.367295 LR 0.001000 Time 0.021072 +2023-10-05 20:53:33,259 - Epoch: [24][ 760/ 1236] Overall Loss 0.367223 Objective Loss 0.367223 LR 0.001000 Time 0.021059 +2023-10-05 20:53:33,459 - Epoch: [24][ 770/ 1236] Overall Loss 0.366999 Objective Loss 0.366999 LR 0.001000 Time 0.021046 +2023-10-05 20:53:33,661 - Epoch: [24][ 780/ 1236] Overall Loss 0.366874 Objective Loss 0.366874 LR 0.001000 Time 0.021034 +2023-10-05 20:53:33,862 - Epoch: [24][ 790/ 1236] Overall Loss 0.366736 Objective Loss 0.366736 LR 0.001000 Time 0.021021 +2023-10-05 20:53:34,064 - Epoch: [24][ 800/ 1236] Overall Loss 0.366883 Objective Loss 0.366883 LR 0.001000 Time 0.021010 +2023-10-05 20:53:34,264 - Epoch: [24][ 810/ 1236] Overall Loss 0.366893 Objective Loss 0.366893 LR 0.001000 Time 0.020998 +2023-10-05 20:53:34,466 - Epoch: [24][ 820/ 1236] Overall Loss 0.366889 Objective Loss 0.366889 LR 0.001000 Time 0.020988 +2023-10-05 20:53:34,666 - Epoch: [24][ 830/ 1236] Overall Loss 0.367194 Objective Loss 0.367194 LR 0.001000 Time 0.020976 +2023-10-05 20:53:34,868 - Epoch: [24][ 840/ 1236] Overall Loss 0.366597 Objective Loss 0.366597 LR 0.001000 Time 0.020966 +2023-10-05 20:53:35,069 - Epoch: [24][ 850/ 1236] Overall Loss 0.366623 Objective Loss 0.366623 LR 0.001000 Time 0.020955 +2023-10-05 20:53:35,270 - Epoch: [24][ 860/ 1236] Overall Loss 0.366521 Objective Loss 0.366521 LR 0.001000 Time 0.020945 +2023-10-05 20:53:35,471 - Epoch: [24][ 870/ 1236] Overall Loss 0.366743 Objective Loss 0.366743 LR 0.001000 Time 0.020935 +2023-10-05 20:53:35,673 - Epoch: [24][ 880/ 1236] Overall Loss 0.366553 Objective Loss 0.366553 LR 0.001000 Time 0.020926 +2023-10-05 20:53:35,873 - Epoch: [24][ 890/ 1236] Overall Loss 0.366516 Objective Loss 0.366516 LR 0.001000 Time 0.020916 +2023-10-05 20:53:36,075 - Epoch: [24][ 900/ 1236] Overall Loss 0.366642 Objective Loss 0.366642 LR 0.001000 Time 0.020908 +2023-10-05 20:53:36,276 - Epoch: [24][ 910/ 1236] Overall Loss 0.366796 Objective Loss 0.366796 LR 0.001000 Time 0.020897 +2023-10-05 20:53:36,478 - Epoch: [24][ 920/ 1236] Overall Loss 0.366767 Objective Loss 0.366767 LR 0.001000 Time 0.020889 +2023-10-05 20:53:36,678 - Epoch: [24][ 930/ 1236] Overall Loss 0.366817 Objective Loss 0.366817 LR 0.001000 Time 0.020880 +2023-10-05 20:53:36,880 - Epoch: [24][ 940/ 1236] Overall Loss 0.366839 Objective Loss 0.366839 LR 0.001000 Time 0.020872 +2023-10-05 20:53:37,080 - Epoch: [24][ 950/ 1236] Overall Loss 0.366823 Objective Loss 0.366823 LR 0.001000 Time 0.020863 +2023-10-05 20:53:37,282 - Epoch: [24][ 960/ 1236] Overall Loss 0.366678 Objective Loss 0.366678 LR 0.001000 Time 0.020856 +2023-10-05 20:53:37,483 - Epoch: [24][ 970/ 1236] Overall Loss 0.366469 Objective Loss 0.366469 LR 0.001000 Time 0.020847 +2023-10-05 20:53:37,685 - Epoch: [24][ 980/ 1236] Overall Loss 0.366552 Objective Loss 0.366552 LR 0.001000 Time 0.020840 +2023-10-05 20:53:37,885 - Epoch: [24][ 990/ 1236] Overall Loss 0.366326 Objective Loss 0.366326 LR 0.001000 Time 0.020832 +2023-10-05 20:53:38,087 - Epoch: [24][ 1000/ 1236] Overall Loss 0.365996 Objective Loss 0.365996 LR 0.001000 Time 0.020825 +2023-10-05 20:53:38,287 - Epoch: [24][ 1010/ 1236] Overall Loss 0.365922 Objective Loss 0.365922 LR 0.001000 Time 0.020816 +2023-10-05 20:53:38,489 - Epoch: [24][ 1020/ 1236] Overall Loss 0.366040 Objective Loss 0.366040 LR 0.001000 Time 0.020810 +2023-10-05 20:53:38,689 - Epoch: [24][ 1030/ 1236] Overall Loss 0.365824 Objective Loss 0.365824 LR 0.001000 Time 0.020802 +2023-10-05 20:53:38,891 - Epoch: [24][ 1040/ 1236] Overall Loss 0.365914 Objective Loss 0.365914 LR 0.001000 Time 0.020796 +2023-10-05 20:53:39,092 - Epoch: [24][ 1050/ 1236] Overall Loss 0.365899 Objective Loss 0.365899 LR 0.001000 Time 0.020788 +2023-10-05 20:53:39,294 - Epoch: [24][ 1060/ 1236] Overall Loss 0.366350 Objective Loss 0.366350 LR 0.001000 Time 0.020783 +2023-10-05 20:53:39,494 - Epoch: [24][ 1070/ 1236] Overall Loss 0.366534 Objective Loss 0.366534 LR 0.001000 Time 0.020775 +2023-10-05 20:53:39,696 - Epoch: [24][ 1080/ 1236] Overall Loss 0.367042 Objective Loss 0.367042 LR 0.001000 Time 0.020770 +2023-10-05 20:53:39,896 - Epoch: [24][ 1090/ 1236] Overall Loss 0.366923 Objective Loss 0.366923 LR 0.001000 Time 0.020762 +2023-10-05 20:53:40,098 - Epoch: [24][ 1100/ 1236] Overall Loss 0.366997 Objective Loss 0.366997 LR 0.001000 Time 0.020757 +2023-10-05 20:53:40,299 - Epoch: [24][ 1110/ 1236] Overall Loss 0.367249 Objective Loss 0.367249 LR 0.001000 Time 0.020750 +2023-10-05 20:53:40,501 - Epoch: [24][ 1120/ 1236] Overall Loss 0.367156 Objective Loss 0.367156 LR 0.001000 Time 0.020745 +2023-10-05 20:53:40,701 - Epoch: [24][ 1130/ 1236] Overall Loss 0.367078 Objective Loss 0.367078 LR 0.001000 Time 0.020738 +2023-10-05 20:53:40,903 - Epoch: [24][ 1140/ 1236] Overall Loss 0.367141 Objective Loss 0.367141 LR 0.001000 Time 0.020733 +2023-10-05 20:53:41,103 - Epoch: [24][ 1150/ 1236] Overall Loss 0.367078 Objective Loss 0.367078 LR 0.001000 Time 0.020726 +2023-10-05 20:53:41,305 - Epoch: [24][ 1160/ 1236] Overall Loss 0.366898 Objective Loss 0.366898 LR 0.001000 Time 0.020721 +2023-10-05 20:53:41,505 - Epoch: [24][ 1170/ 1236] Overall Loss 0.366896 Objective Loss 0.366896 LR 0.001000 Time 0.020715 +2023-10-05 20:53:41,707 - Epoch: [24][ 1180/ 1236] Overall Loss 0.367237 Objective Loss 0.367237 LR 0.001000 Time 0.020711 +2023-10-05 20:53:41,907 - Epoch: [24][ 1190/ 1236] Overall Loss 0.367597 Objective Loss 0.367597 LR 0.001000 Time 0.020705 +2023-10-05 20:53:42,109 - Epoch: [24][ 1200/ 1236] Overall Loss 0.367763 Objective Loss 0.367763 LR 0.001000 Time 0.020700 +2023-10-05 20:53:42,310 - Epoch: [24][ 1210/ 1236] Overall Loss 0.367701 Objective Loss 0.367701 LR 0.001000 Time 0.020694 +2023-10-05 20:53:42,511 - Epoch: [24][ 1220/ 1236] Overall Loss 0.367796 Objective Loss 0.367796 LR 0.001000 Time 0.020690 +2023-10-05 20:53:42,765 - Epoch: [24][ 1230/ 1236] Overall Loss 0.367968 Objective Loss 0.367968 LR 0.001000 Time 0.020727 +2023-10-05 20:53:42,882 - Epoch: [24][ 1236/ 1236] Overall Loss 0.367964 Objective Loss 0.367964 Top1 83.299389 Top5 96.537678 LR 0.001000 Time 0.020722 +2023-10-05 20:53:43,024 - --- validate (epoch=24)----------- +2023-10-05 20:53:43,025 - 29943 samples (256 per mini-batch) +2023-10-05 20:53:43,482 - Epoch: [24][ 10/ 117] Loss 0.410575 Top1 78.046875 Top5 96.328125 +2023-10-05 20:53:43,631 - Epoch: [24][ 20/ 117] Loss 0.411062 Top1 78.730469 Top5 96.679688 +2023-10-05 20:53:43,778 - Epoch: [24][ 30/ 117] Loss 0.398576 Top1 79.179688 Top5 97.135417 +2023-10-05 20:53:43,924 - Epoch: [24][ 40/ 117] Loss 0.393530 Top1 79.130859 Top5 97.138672 +2023-10-05 20:53:44,071 - Epoch: [24][ 50/ 117] Loss 0.393124 Top1 79.328125 Top5 97.164062 +2023-10-05 20:53:44,217 - Epoch: [24][ 60/ 117] Loss 0.394793 Top1 79.348958 Top5 97.076823 +2023-10-05 20:53:44,364 - Epoch: [24][ 70/ 117] Loss 0.399822 Top1 79.268973 Top5 97.070312 +2023-10-05 20:53:44,513 - Epoch: [24][ 80/ 117] Loss 0.397149 Top1 79.326172 Top5 96.987305 +2023-10-05 20:53:44,662 - Epoch: [24][ 90/ 117] Loss 0.393765 Top1 79.357639 Top5 97.005208 +2023-10-05 20:53:44,809 - Epoch: [24][ 100/ 117] Loss 0.393255 Top1 79.453125 Top5 97.007812 +2023-10-05 20:53:44,962 - Epoch: [24][ 110/ 117] Loss 0.393355 Top1 79.474432 Top5 97.009943 +2023-10-05 20:53:45,048 - Epoch: [24][ 117/ 117] Loss 0.391916 Top1 79.514411 Top5 97.034365 +2023-10-05 20:53:45,183 - ==> Top1: 79.514 Top5: 97.034 Loss: 0.392 + +2023-10-05 20:53:45,184 - ==> Confusion: +[[ 921 1 3 2 18 3 0 1 3 57 2 0 0 5 5 3 5 1 0 1 19] + [ 0 1040 0 1 8 27 2 23 2 0 4 2 0 1 2 5 4 0 5 2 3] + [ 3 0 910 21 1 2 53 13 0 1 8 2 8 0 2 2 4 2 10 2 12] + [ 1 2 18 945 0 6 6 0 1 0 12 1 7 0 36 5 1 12 20 2 14] + [ 21 18 2 0 979 5 1 0 0 5 1 1 0 2 2 2 5 1 0 2 3] + [ 8 48 2 2 7 962 2 24 0 1 10 7 0 24 3 0 5 0 2 2 7] + [ 0 7 17 0 0 1 1131 10 0 0 1 2 2 0 0 7 0 0 3 5 5] + [ 2 24 18 0 2 32 5 1059 0 1 3 10 0 1 1 2 0 2 40 9 7] + [ 33 8 1 0 1 3 0 2 905 33 22 1 0 37 29 2 1 0 5 4 2] + [ 133 1 0 0 11 4 1 0 24 850 1 0 0 65 5 6 1 2 0 5 10] + [ 3 3 13 9 0 2 3 8 7 1 972 3 0 8 4 1 1 0 6 2 7] + [ 1 2 3 1 0 14 0 1 1 0 0 956 18 4 0 2 1 15 0 13 3] + [ 0 0 4 3 1 0 2 2 0 0 2 43 952 6 3 9 6 10 7 8 10] + [ 2 1 0 0 5 7 1 1 3 10 15 5 3 1043 2 4 2 1 0 4 10] + [ 15 5 1 11 9 0 0 0 17 3 3 0 2 0 995 0 1 5 13 0 21] + [ 0 2 3 2 3 1 0 1 0 0 1 11 9 2 0 1052 9 19 0 13 6] + [ 3 14 2 0 15 6 1 1 1 0 0 5 2 3 0 15 1073 0 1 6 13] + [ 1 1 1 0 0 1 2 0 0 0 0 6 31 4 1 5 1 976 1 0 7] + [ 0 15 8 16 3 0 0 39 5 0 8 0 2 0 8 0 1 0 951 5 7] + [ 0 3 3 0 2 7 13 16 0 0 1 13 5 2 0 4 5 0 1 1074 3] + [ 164 319 209 82 181 210 111 139 73 87 275 129 499 415 172 79 159 75 176 288 4063]] + +2023-10-05 20:53:45,185 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:53:45,185 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:53:45,191 - + +2023-10-05 20:53:45,191 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:53:46,174 - Epoch: [25][ 10/ 1236] Overall Loss 0.373511 Objective Loss 0.373511 LR 0.001000 Time 0.098207 +2023-10-05 20:53:46,374 - Epoch: [25][ 20/ 1236] Overall Loss 0.367251 Objective Loss 0.367251 LR 0.001000 Time 0.059100 +2023-10-05 20:53:46,573 - Epoch: [25][ 30/ 1236] Overall Loss 0.367703 Objective Loss 0.367703 LR 0.001000 Time 0.046008 +2023-10-05 20:53:46,773 - Epoch: [25][ 40/ 1236] Overall Loss 0.365422 Objective Loss 0.365422 LR 0.001000 Time 0.039509 +2023-10-05 20:53:46,972 - Epoch: [25][ 50/ 1236] Overall Loss 0.364637 Objective Loss 0.364637 LR 0.001000 Time 0.035570 +2023-10-05 20:53:47,172 - Epoch: [25][ 60/ 1236] Overall Loss 0.361747 Objective Loss 0.361747 LR 0.001000 Time 0.032980 +2023-10-05 20:53:47,371 - Epoch: [25][ 70/ 1236] Overall Loss 0.361427 Objective Loss 0.361427 LR 0.001000 Time 0.031100 +2023-10-05 20:53:47,572 - Epoch: [25][ 80/ 1236] Overall Loss 0.363273 Objective Loss 0.363273 LR 0.001000 Time 0.029719 +2023-10-05 20:53:47,770 - Epoch: [25][ 90/ 1236] Overall Loss 0.364904 Objective Loss 0.364904 LR 0.001000 Time 0.028617 +2023-10-05 20:53:47,971 - Epoch: [25][ 100/ 1236] Overall Loss 0.364536 Objective Loss 0.364536 LR 0.001000 Time 0.027758 +2023-10-05 20:53:48,169 - Epoch: [25][ 110/ 1236] Overall Loss 0.362867 Objective Loss 0.362867 LR 0.001000 Time 0.027037 +2023-10-05 20:53:48,370 - Epoch: [25][ 120/ 1236] Overall Loss 0.366507 Objective Loss 0.366507 LR 0.001000 Time 0.026452 +2023-10-05 20:53:48,569 - Epoch: [25][ 130/ 1236] Overall Loss 0.366226 Objective Loss 0.366226 LR 0.001000 Time 0.025943 +2023-10-05 20:53:48,769 - Epoch: [25][ 140/ 1236] Overall Loss 0.363669 Objective Loss 0.363669 LR 0.001000 Time 0.025521 +2023-10-05 20:53:48,968 - Epoch: [25][ 150/ 1236] Overall Loss 0.363250 Objective Loss 0.363250 LR 0.001000 Time 0.025142 +2023-10-05 20:53:49,169 - Epoch: [25][ 160/ 1236] Overall Loss 0.365471 Objective Loss 0.365471 LR 0.001000 Time 0.024822 +2023-10-05 20:53:49,368 - Epoch: [25][ 170/ 1236] Overall Loss 0.364072 Objective Loss 0.364072 LR 0.001000 Time 0.024530 +2023-10-05 20:53:49,568 - Epoch: [25][ 180/ 1236] Overall Loss 0.364173 Objective Loss 0.364173 LR 0.001000 Time 0.024280 +2023-10-05 20:53:49,767 - Epoch: [25][ 190/ 1236] Overall Loss 0.362817 Objective Loss 0.362817 LR 0.001000 Time 0.024046 +2023-10-05 20:53:49,968 - Epoch: [25][ 200/ 1236] Overall Loss 0.363679 Objective Loss 0.363679 LR 0.001000 Time 0.023845 +2023-10-05 20:53:50,166 - Epoch: [25][ 210/ 1236] Overall Loss 0.364618 Objective Loss 0.364618 LR 0.001000 Time 0.023654 +2023-10-05 20:53:50,367 - Epoch: [25][ 220/ 1236] Overall Loss 0.364480 Objective Loss 0.364480 LR 0.001000 Time 0.023489 +2023-10-05 20:53:50,566 - Epoch: [25][ 230/ 1236] Overall Loss 0.365483 Objective Loss 0.365483 LR 0.001000 Time 0.023330 +2023-10-05 20:53:50,767 - Epoch: [25][ 240/ 1236] Overall Loss 0.365719 Objective Loss 0.365719 LR 0.001000 Time 0.023193 +2023-10-05 20:53:50,965 - Epoch: [25][ 250/ 1236] Overall Loss 0.366160 Objective Loss 0.366160 LR 0.001000 Time 0.023059 +2023-10-05 20:53:51,166 - Epoch: [25][ 260/ 1236] Overall Loss 0.365074 Objective Loss 0.365074 LR 0.001000 Time 0.022944 +2023-10-05 20:53:51,365 - Epoch: [25][ 270/ 1236] Overall Loss 0.365181 Objective Loss 0.365181 LR 0.001000 Time 0.022829 +2023-10-05 20:53:51,566 - Epoch: [25][ 280/ 1236] Overall Loss 0.364221 Objective Loss 0.364221 LR 0.001000 Time 0.022729 +2023-10-05 20:53:51,764 - Epoch: [25][ 290/ 1236] Overall Loss 0.364579 Objective Loss 0.364579 LR 0.001000 Time 0.022629 +2023-10-05 20:53:51,965 - Epoch: [25][ 300/ 1236] Overall Loss 0.365100 Objective Loss 0.365100 LR 0.001000 Time 0.022542 +2023-10-05 20:53:52,164 - Epoch: [25][ 310/ 1236] Overall Loss 0.366021 Objective Loss 0.366021 LR 0.001000 Time 0.022454 +2023-10-05 20:53:52,365 - Epoch: [25][ 320/ 1236] Overall Loss 0.365023 Objective Loss 0.365023 LR 0.001000 Time 0.022380 +2023-10-05 20:53:52,564 - Epoch: [25][ 330/ 1236] Overall Loss 0.364888 Objective Loss 0.364888 LR 0.001000 Time 0.022305 +2023-10-05 20:53:52,765 - Epoch: [25][ 340/ 1236] Overall Loss 0.364257 Objective Loss 0.364257 LR 0.001000 Time 0.022238 +2023-10-05 20:53:52,963 - Epoch: [25][ 350/ 1236] Overall Loss 0.363652 Objective Loss 0.363652 LR 0.001000 Time 0.022169 +2023-10-05 20:53:53,164 - Epoch: [25][ 360/ 1236] Overall Loss 0.363849 Objective Loss 0.363849 LR 0.001000 Time 0.022110 +2023-10-05 20:53:53,363 - Epoch: [25][ 370/ 1236] Overall Loss 0.363185 Objective Loss 0.363185 LR 0.001000 Time 0.022049 +2023-10-05 20:53:53,564 - Epoch: [25][ 380/ 1236] Overall Loss 0.363559 Objective Loss 0.363559 LR 0.001000 Time 0.021996 +2023-10-05 20:53:53,763 - Epoch: [25][ 390/ 1236] Overall Loss 0.363209 Objective Loss 0.363209 LR 0.001000 Time 0.021942 +2023-10-05 20:53:53,964 - Epoch: [25][ 400/ 1236] Overall Loss 0.362886 Objective Loss 0.362886 LR 0.001000 Time 0.021894 +2023-10-05 20:53:54,163 - Epoch: [25][ 410/ 1236] Overall Loss 0.362828 Objective Loss 0.362828 LR 0.001000 Time 0.021845 +2023-10-05 20:53:54,363 - Epoch: [25][ 420/ 1236] Overall Loss 0.363513 Objective Loss 0.363513 LR 0.001000 Time 0.021802 +2023-10-05 20:53:54,562 - Epoch: [25][ 430/ 1236] Overall Loss 0.363226 Objective Loss 0.363226 LR 0.001000 Time 0.021757 +2023-10-05 20:53:54,763 - Epoch: [25][ 440/ 1236] Overall Loss 0.363472 Objective Loss 0.363472 LR 0.001000 Time 0.021718 +2023-10-05 20:53:54,963 - Epoch: [25][ 450/ 1236] Overall Loss 0.362948 Objective Loss 0.362948 LR 0.001000 Time 0.021679 +2023-10-05 20:53:55,165 - Epoch: [25][ 460/ 1236] Overall Loss 0.363277 Objective Loss 0.363277 LR 0.001000 Time 0.021647 +2023-10-05 20:53:55,366 - Epoch: [25][ 470/ 1236] Overall Loss 0.363796 Objective Loss 0.363796 LR 0.001000 Time 0.021612 +2023-10-05 20:53:55,569 - Epoch: [25][ 480/ 1236] Overall Loss 0.364043 Objective Loss 0.364043 LR 0.001000 Time 0.021583 +2023-10-05 20:53:55,769 - Epoch: [25][ 490/ 1236] Overall Loss 0.364264 Objective Loss 0.364264 LR 0.001000 Time 0.021551 +2023-10-05 20:53:55,977 - Epoch: [25][ 500/ 1236] Overall Loss 0.363938 Objective Loss 0.363938 LR 0.001000 Time 0.021536 +2023-10-05 20:53:56,180 - Epoch: [25][ 510/ 1236] Overall Loss 0.363925 Objective Loss 0.363925 LR 0.001000 Time 0.021510 +2023-10-05 20:53:56,385 - Epoch: [25][ 520/ 1236] Overall Loss 0.364229 Objective Loss 0.364229 LR 0.001000 Time 0.021491 +2023-10-05 20:53:56,588 - Epoch: [25][ 530/ 1236] Overall Loss 0.363788 Objective Loss 0.363788 LR 0.001000 Time 0.021467 +2023-10-05 20:53:56,794 - Epoch: [25][ 540/ 1236] Overall Loss 0.364030 Objective Loss 0.364030 LR 0.001000 Time 0.021450 +2023-10-05 20:53:56,996 - Epoch: [25][ 550/ 1236] Overall Loss 0.363196 Objective Loss 0.363196 LR 0.001000 Time 0.021427 +2023-10-05 20:53:57,201 - Epoch: [25][ 560/ 1236] Overall Loss 0.362882 Objective Loss 0.362882 LR 0.001000 Time 0.021411 +2023-10-05 20:53:57,404 - Epoch: [25][ 570/ 1236] Overall Loss 0.363370 Objective Loss 0.363370 LR 0.001000 Time 0.021389 +2023-10-05 20:53:57,610 - Epoch: [25][ 580/ 1236] Overall Loss 0.363169 Objective Loss 0.363169 LR 0.001000 Time 0.021375 +2023-10-05 20:53:57,812 - Epoch: [25][ 590/ 1236] Overall Loss 0.363234 Objective Loss 0.363234 LR 0.001000 Time 0.021355 +2023-10-05 20:53:58,017 - Epoch: [25][ 600/ 1236] Overall Loss 0.363044 Objective Loss 0.363044 LR 0.001000 Time 0.021341 +2023-10-05 20:53:58,220 - Epoch: [25][ 610/ 1236] Overall Loss 0.362981 Objective Loss 0.362981 LR 0.001000 Time 0.021322 +2023-10-05 20:53:58,426 - Epoch: [25][ 620/ 1236] Overall Loss 0.362505 Objective Loss 0.362505 LR 0.001000 Time 0.021309 +2023-10-05 20:53:58,628 - Epoch: [25][ 630/ 1236] Overall Loss 0.362547 Objective Loss 0.362547 LR 0.001000 Time 0.021293 +2023-10-05 20:53:58,834 - Epoch: [25][ 640/ 1236] Overall Loss 0.362954 Objective Loss 0.362954 LR 0.001000 Time 0.021280 +2023-10-05 20:53:59,036 - Epoch: [25][ 650/ 1236] Overall Loss 0.362904 Objective Loss 0.362904 LR 0.001000 Time 0.021264 +2023-10-05 20:53:59,242 - Epoch: [25][ 660/ 1236] Overall Loss 0.363065 Objective Loss 0.363065 LR 0.001000 Time 0.021253 +2023-10-05 20:53:59,445 - Epoch: [25][ 670/ 1236] Overall Loss 0.363541 Objective Loss 0.363541 LR 0.001000 Time 0.021238 +2023-10-05 20:53:59,649 - Epoch: [25][ 680/ 1236] Overall Loss 0.363880 Objective Loss 0.363880 LR 0.001000 Time 0.021225 +2023-10-05 20:53:59,849 - Epoch: [25][ 690/ 1236] Overall Loss 0.363550 Objective Loss 0.363550 LR 0.001000 Time 0.021208 +2023-10-05 20:54:00,052 - Epoch: [25][ 700/ 1236] Overall Loss 0.363259 Objective Loss 0.363259 LR 0.001000 Time 0.021194 +2023-10-05 20:54:00,253 - Epoch: [25][ 710/ 1236] Overall Loss 0.363446 Objective Loss 0.363446 LR 0.001000 Time 0.021177 +2023-10-05 20:54:00,455 - Epoch: [25][ 720/ 1236] Overall Loss 0.363195 Objective Loss 0.363195 LR 0.001000 Time 0.021164 +2023-10-05 20:54:00,656 - Epoch: [25][ 730/ 1236] Overall Loss 0.363157 Objective Loss 0.363157 LR 0.001000 Time 0.021149 +2023-10-05 20:54:00,859 - Epoch: [25][ 740/ 1236] Overall Loss 0.363673 Objective Loss 0.363673 LR 0.001000 Time 0.021136 +2023-10-05 20:54:01,060 - Epoch: [25][ 750/ 1236] Overall Loss 0.363847 Objective Loss 0.363847 LR 0.001000 Time 0.021122 +2023-10-05 20:54:01,262 - Epoch: [25][ 760/ 1236] Overall Loss 0.363887 Objective Loss 0.363887 LR 0.001000 Time 0.021110 +2023-10-05 20:54:01,463 - Epoch: [25][ 770/ 1236] Overall Loss 0.364023 Objective Loss 0.364023 LR 0.001000 Time 0.021096 +2023-10-05 20:54:01,666 - Epoch: [25][ 780/ 1236] Overall Loss 0.363745 Objective Loss 0.363745 LR 0.001000 Time 0.021085 +2023-10-05 20:54:01,867 - Epoch: [25][ 790/ 1236] Overall Loss 0.364391 Objective Loss 0.364391 LR 0.001000 Time 0.021072 +2023-10-05 20:54:02,070 - Epoch: [25][ 800/ 1236] Overall Loss 0.365024 Objective Loss 0.365024 LR 0.001000 Time 0.021062 +2023-10-05 20:54:02,270 - Epoch: [25][ 810/ 1236] Overall Loss 0.365520 Objective Loss 0.365520 LR 0.001000 Time 0.021049 +2023-10-05 20:54:02,473 - Epoch: [25][ 820/ 1236] Overall Loss 0.365451 Objective Loss 0.365451 LR 0.001000 Time 0.021039 +2023-10-05 20:54:02,674 - Epoch: [25][ 830/ 1236] Overall Loss 0.365490 Objective Loss 0.365490 LR 0.001000 Time 0.021027 +2023-10-05 20:54:02,877 - Epoch: [25][ 840/ 1236] Overall Loss 0.365168 Objective Loss 0.365168 LR 0.001000 Time 0.021018 +2023-10-05 20:54:03,078 - Epoch: [25][ 850/ 1236] Overall Loss 0.365570 Objective Loss 0.365570 LR 0.001000 Time 0.021007 +2023-10-05 20:54:03,281 - Epoch: [25][ 860/ 1236] Overall Loss 0.366097 Objective Loss 0.366097 LR 0.001000 Time 0.020998 +2023-10-05 20:54:03,486 - Epoch: [25][ 870/ 1236] Overall Loss 0.366101 Objective Loss 0.366101 LR 0.001000 Time 0.020992 +2023-10-05 20:54:03,687 - Epoch: [25][ 880/ 1236] Overall Loss 0.365994 Objective Loss 0.365994 LR 0.001000 Time 0.020982 +2023-10-05 20:54:03,887 - Epoch: [25][ 890/ 1236] Overall Loss 0.366287 Objective Loss 0.366287 LR 0.001000 Time 0.020971 +2023-10-05 20:54:04,090 - Epoch: [25][ 900/ 1236] Overall Loss 0.366466 Objective Loss 0.366466 LR 0.001000 Time 0.020962 +2023-10-05 20:54:04,290 - Epoch: [25][ 910/ 1236] Overall Loss 0.366285 Objective Loss 0.366285 LR 0.001000 Time 0.020951 +2023-10-05 20:54:04,492 - Epoch: [25][ 920/ 1236] Overall Loss 0.366260 Objective Loss 0.366260 LR 0.001000 Time 0.020942 +2023-10-05 20:54:04,692 - Epoch: [25][ 930/ 1236] Overall Loss 0.366330 Objective Loss 0.366330 LR 0.001000 Time 0.020932 +2023-10-05 20:54:04,894 - Epoch: [25][ 940/ 1236] Overall Loss 0.366206 Objective Loss 0.366206 LR 0.001000 Time 0.020924 +2023-10-05 20:54:05,094 - Epoch: [25][ 950/ 1236] Overall Loss 0.365993 Objective Loss 0.365993 LR 0.001000 Time 0.020914 +2023-10-05 20:54:05,296 - Epoch: [25][ 960/ 1236] Overall Loss 0.366067 Objective Loss 0.366067 LR 0.001000 Time 0.020906 +2023-10-05 20:54:05,496 - Epoch: [25][ 970/ 1236] Overall Loss 0.366230 Objective Loss 0.366230 LR 0.001000 Time 0.020896 +2023-10-05 20:54:05,698 - Epoch: [25][ 980/ 1236] Overall Loss 0.366009 Objective Loss 0.366009 LR 0.001000 Time 0.020889 +2023-10-05 20:54:05,899 - Epoch: [25][ 990/ 1236] Overall Loss 0.366167 Objective Loss 0.366167 LR 0.001000 Time 0.020880 +2023-10-05 20:54:06,101 - Epoch: [25][ 1000/ 1236] Overall Loss 0.366128 Objective Loss 0.366128 LR 0.001000 Time 0.020873 +2023-10-05 20:54:06,301 - Epoch: [25][ 1010/ 1236] Overall Loss 0.366029 Objective Loss 0.366029 LR 0.001000 Time 0.020864 +2023-10-05 20:54:06,503 - Epoch: [25][ 1020/ 1236] Overall Loss 0.366240 Objective Loss 0.366240 LR 0.001000 Time 0.020858 +2023-10-05 20:54:06,704 - Epoch: [25][ 1030/ 1236] Overall Loss 0.366351 Objective Loss 0.366351 LR 0.001000 Time 0.020849 +2023-10-05 20:54:06,906 - Epoch: [25][ 1040/ 1236] Overall Loss 0.366884 Objective Loss 0.366884 LR 0.001000 Time 0.020843 +2023-10-05 20:54:07,106 - Epoch: [25][ 1050/ 1236] Overall Loss 0.366886 Objective Loss 0.366886 LR 0.001000 Time 0.020835 +2023-10-05 20:54:07,308 - Epoch: [25][ 1060/ 1236] Overall Loss 0.366766 Objective Loss 0.366766 LR 0.001000 Time 0.020828 +2023-10-05 20:54:07,509 - Epoch: [25][ 1070/ 1236] Overall Loss 0.366652 Objective Loss 0.366652 LR 0.001000 Time 0.020821 +2023-10-05 20:54:07,711 - Epoch: [25][ 1080/ 1236] Overall Loss 0.366626 Objective Loss 0.366626 LR 0.001000 Time 0.020815 +2023-10-05 20:54:07,911 - Epoch: [25][ 1090/ 1236] Overall Loss 0.366605 Objective Loss 0.366605 LR 0.001000 Time 0.020807 +2023-10-05 20:54:08,113 - Epoch: [25][ 1100/ 1236] Overall Loss 0.367090 Objective Loss 0.367090 LR 0.001000 Time 0.020802 +2023-10-05 20:54:08,314 - Epoch: [25][ 1110/ 1236] Overall Loss 0.367029 Objective Loss 0.367029 LR 0.001000 Time 0.020794 +2023-10-05 20:54:08,516 - Epoch: [25][ 1120/ 1236] Overall Loss 0.366686 Objective Loss 0.366686 LR 0.001000 Time 0.020789 +2023-10-05 20:54:08,716 - Epoch: [25][ 1130/ 1236] Overall Loss 0.366776 Objective Loss 0.366776 LR 0.001000 Time 0.020782 +2023-10-05 20:54:08,918 - Epoch: [25][ 1140/ 1236] Overall Loss 0.366613 Objective Loss 0.366613 LR 0.001000 Time 0.020777 +2023-10-05 20:54:09,119 - Epoch: [25][ 1150/ 1236] Overall Loss 0.366371 Objective Loss 0.366371 LR 0.001000 Time 0.020770 +2023-10-05 20:54:09,321 - Epoch: [25][ 1160/ 1236] Overall Loss 0.366238 Objective Loss 0.366238 LR 0.001000 Time 0.020765 +2023-10-05 20:54:09,521 - Epoch: [25][ 1170/ 1236] Overall Loss 0.366250 Objective Loss 0.366250 LR 0.001000 Time 0.020758 +2023-10-05 20:54:09,723 - Epoch: [25][ 1180/ 1236] Overall Loss 0.366141 Objective Loss 0.366141 LR 0.001000 Time 0.020753 +2023-10-05 20:54:09,923 - Epoch: [25][ 1190/ 1236] Overall Loss 0.365979 Objective Loss 0.365979 LR 0.001000 Time 0.020747 +2023-10-05 20:54:10,125 - Epoch: [25][ 1200/ 1236] Overall Loss 0.366093 Objective Loss 0.366093 LR 0.001000 Time 0.020742 +2023-10-05 20:54:10,325 - Epoch: [25][ 1210/ 1236] Overall Loss 0.365883 Objective Loss 0.365883 LR 0.001000 Time 0.020735 +2023-10-05 20:54:10,527 - Epoch: [25][ 1220/ 1236] Overall Loss 0.365744 Objective Loss 0.365744 LR 0.001000 Time 0.020731 +2023-10-05 20:54:10,784 - Epoch: [25][ 1230/ 1236] Overall Loss 0.365750 Objective Loss 0.365750 LR 0.001000 Time 0.020770 +2023-10-05 20:54:10,902 - Epoch: [25][ 1236/ 1236] Overall Loss 0.365765 Objective Loss 0.365765 Top1 84.725051 Top5 97.556008 LR 0.001000 Time 0.020765 +2023-10-05 20:54:11,039 - --- validate (epoch=25)----------- +2023-10-05 20:54:11,039 - 29943 samples (256 per mini-batch) +2023-10-05 20:54:11,502 - Epoch: [25][ 10/ 117] Loss 0.371288 Top1 79.726562 Top5 97.070312 +2023-10-05 20:54:11,655 - Epoch: [25][ 20/ 117] Loss 0.375102 Top1 79.238281 Top5 97.050781 +2023-10-05 20:54:11,805 - Epoch: [25][ 30/ 117] Loss 0.369697 Top1 79.361979 Top5 97.161458 +2023-10-05 20:54:11,957 - Epoch: [25][ 40/ 117] Loss 0.369082 Top1 79.238281 Top5 97.236328 +2023-10-05 20:54:12,109 - Epoch: [25][ 50/ 117] Loss 0.370521 Top1 79.226562 Top5 97.218750 +2023-10-05 20:54:12,261 - Epoch: [25][ 60/ 117] Loss 0.376585 Top1 79.173177 Top5 97.213542 +2023-10-05 20:54:12,415 - Epoch: [25][ 70/ 117] Loss 0.377804 Top1 79.174107 Top5 97.154018 +2023-10-05 20:54:12,567 - Epoch: [25][ 80/ 117] Loss 0.376411 Top1 79.262695 Top5 97.202148 +2023-10-05 20:54:12,716 - Epoch: [25][ 90/ 117] Loss 0.378253 Top1 79.275174 Top5 97.157118 +2023-10-05 20:54:12,866 - Epoch: [25][ 100/ 117] Loss 0.380668 Top1 79.164062 Top5 97.117188 +2023-10-05 20:54:13,021 - Epoch: [25][ 110/ 117] Loss 0.381362 Top1 79.122869 Top5 97.098722 +2023-10-05 20:54:13,107 - Epoch: [25][ 117/ 117] Loss 0.379515 Top1 79.167084 Top5 97.087800 +2023-10-05 20:54:13,238 - ==> Top1: 79.167 Top5: 97.088 Loss: 0.380 + +2023-10-05 20:54:13,239 - ==> Confusion: +[[ 911 1 6 0 2 2 0 0 6 86 1 2 1 6 4 4 6 0 0 1 11] + [ 3 1037 2 0 6 18 4 35 2 0 4 1 0 1 0 3 3 0 7 3 2] + [ 3 2 917 21 2 0 38 10 0 3 7 4 7 4 3 4 5 1 11 4 10] + [ 5 2 14 927 0 6 3 2 6 0 8 0 12 2 50 5 1 9 22 3 12] + [ 42 11 3 0 904 8 1 0 3 12 2 4 0 3 12 6 22 2 0 2 13] + [ 7 44 0 1 1 980 2 22 1 0 3 8 1 21 4 2 5 0 2 4 8] + [ 0 9 21 0 0 4 1109 7 0 0 3 2 2 0 0 20 1 1 1 9 2] + [ 7 11 21 1 0 38 4 1070 1 2 3 10 0 0 0 3 0 1 32 9 5] + [ 14 1 0 0 0 2 0 1 979 39 14 6 0 14 12 4 0 0 3 0 0] + [ 97 2 0 0 2 2 1 0 44 918 2 4 0 30 4 5 1 0 0 1 6] + [ 3 4 8 7 0 1 5 4 25 1 949 3 0 11 7 2 0 1 12 1 9] + [ 1 1 0 1 0 25 1 0 0 0 0 941 27 4 0 6 2 14 0 9 3] + [ 0 0 1 10 0 2 1 0 1 0 1 50 936 2 2 16 3 22 3 8 10] + [ 2 1 0 0 1 12 1 0 14 16 5 7 3 1031 5 5 1 2 0 5 8] + [ 16 3 1 10 5 1 0 0 30 7 1 0 1 1 991 0 1 6 16 0 11] + [ 0 1 1 1 0 1 1 1 0 0 1 13 8 1 0 1070 12 5 0 10 8] + [ 4 17 2 0 4 6 0 0 5 0 1 4 3 2 1 18 1081 0 0 6 7] + [ 1 0 0 2 0 0 2 0 2 2 0 10 35 1 3 13 0 963 1 1 2] + [ 0 10 11 21 0 1 0 40 8 1 3 1 4 0 15 0 1 0 943 1 8] + [ 0 3 2 0 2 8 14 16 0 1 2 16 7 1 0 3 6 0 2 1069 0] + [ 201 331 199 78 69 223 67 104 154 111 202 144 480 356 201 118 259 59 201 369 3979]] + +2023-10-05 20:54:13,240 - ==> Best [Top1: 80.116 Top5: 97.221 Sparsity:0.00 Params: 148928 on epoch: 20] +2023-10-05 20:54:13,240 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:54:13,246 - + +2023-10-05 20:54:13,246 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:54:14,376 - Epoch: [26][ 10/ 1236] Overall Loss 0.349012 Objective Loss 0.349012 LR 0.001000 Time 0.112960 +2023-10-05 20:54:14,578 - Epoch: [26][ 20/ 1236] Overall Loss 0.347494 Objective Loss 0.347494 LR 0.001000 Time 0.066550 +2023-10-05 20:54:14,778 - Epoch: [26][ 30/ 1236] Overall Loss 0.347768 Objective Loss 0.347768 LR 0.001000 Time 0.051000 +2023-10-05 20:54:14,980 - Epoch: [26][ 40/ 1236] Overall Loss 0.352944 Objective Loss 0.352944 LR 0.001000 Time 0.043291 +2023-10-05 20:54:15,179 - Epoch: [26][ 50/ 1236] Overall Loss 0.346139 Objective Loss 0.346139 LR 0.001000 Time 0.038618 +2023-10-05 20:54:15,381 - Epoch: [26][ 60/ 1236] Overall Loss 0.350551 Objective Loss 0.350551 LR 0.001000 Time 0.035542 +2023-10-05 20:54:15,581 - Epoch: [26][ 70/ 1236] Overall Loss 0.345716 Objective Loss 0.345716 LR 0.001000 Time 0.033318 +2023-10-05 20:54:15,783 - Epoch: [26][ 80/ 1236] Overall Loss 0.341781 Objective Loss 0.341781 LR 0.001000 Time 0.031672 +2023-10-05 20:54:15,983 - Epoch: [26][ 90/ 1236] Overall Loss 0.341544 Objective Loss 0.341544 LR 0.001000 Time 0.030369 +2023-10-05 20:54:16,185 - Epoch: [26][ 100/ 1236] Overall Loss 0.339360 Objective Loss 0.339360 LR 0.001000 Time 0.029350 +2023-10-05 20:54:16,385 - Epoch: [26][ 110/ 1236] Overall Loss 0.339514 Objective Loss 0.339514 LR 0.001000 Time 0.028498 +2023-10-05 20:54:16,587 - Epoch: [26][ 120/ 1236] Overall Loss 0.339268 Objective Loss 0.339268 LR 0.001000 Time 0.027802 +2023-10-05 20:54:16,788 - Epoch: [26][ 130/ 1236] Overall Loss 0.339595 Objective Loss 0.339595 LR 0.001000 Time 0.027207 +2023-10-05 20:54:16,989 - Epoch: [26][ 140/ 1236] Overall Loss 0.341261 Objective Loss 0.341261 LR 0.001000 Time 0.026696 +2023-10-05 20:54:17,189 - Epoch: [26][ 150/ 1236] Overall Loss 0.343414 Objective Loss 0.343414 LR 0.001000 Time 0.026249 +2023-10-05 20:54:17,391 - Epoch: [26][ 160/ 1236] Overall Loss 0.344551 Objective Loss 0.344551 LR 0.001000 Time 0.025865 +2023-10-05 20:54:17,591 - Epoch: [26][ 170/ 1236] Overall Loss 0.344544 Objective Loss 0.344544 LR 0.001000 Time 0.025521 +2023-10-05 20:54:17,793 - Epoch: [26][ 180/ 1236] Overall Loss 0.344289 Objective Loss 0.344289 LR 0.001000 Time 0.025222 +2023-10-05 20:54:17,994 - Epoch: [26][ 190/ 1236] Overall Loss 0.342980 Objective Loss 0.342980 LR 0.001000 Time 0.024949 +2023-10-05 20:54:18,195 - Epoch: [26][ 200/ 1236] Overall Loss 0.341937 Objective Loss 0.341937 LR 0.001000 Time 0.024705 +2023-10-05 20:54:18,395 - Epoch: [26][ 210/ 1236] Overall Loss 0.341284 Objective Loss 0.341284 LR 0.001000 Time 0.024482 +2023-10-05 20:54:18,597 - Epoch: [26][ 220/ 1236] Overall Loss 0.342067 Objective Loss 0.342067 LR 0.001000 Time 0.024283 +2023-10-05 20:54:18,797 - Epoch: [26][ 230/ 1236] Overall Loss 0.341886 Objective Loss 0.341886 LR 0.001000 Time 0.024097 +2023-10-05 20:54:18,999 - Epoch: [26][ 240/ 1236] Overall Loss 0.343344 Objective Loss 0.343344 LR 0.001000 Time 0.023931 +2023-10-05 20:54:19,199 - Epoch: [26][ 250/ 1236] Overall Loss 0.342635 Objective Loss 0.342635 LR 0.001000 Time 0.023776 +2023-10-05 20:54:19,400 - Epoch: [26][ 260/ 1236] Overall Loss 0.342353 Objective Loss 0.342353 LR 0.001000 Time 0.023631 +2023-10-05 20:54:19,601 - Epoch: [26][ 270/ 1236] Overall Loss 0.343768 Objective Loss 0.343768 LR 0.001000 Time 0.023498 +2023-10-05 20:54:19,800 - Epoch: [26][ 280/ 1236] Overall Loss 0.343323 Objective Loss 0.343323 LR 0.001000 Time 0.023371 +2023-10-05 20:54:20,001 - Epoch: [26][ 290/ 1236] Overall Loss 0.342900 Objective Loss 0.342900 LR 0.001000 Time 0.023256 +2023-10-05 20:54:20,201 - Epoch: [26][ 300/ 1236] Overall Loss 0.343469 Objective Loss 0.343469 LR 0.001000 Time 0.023148 +2023-10-05 20:54:20,402 - Epoch: [26][ 310/ 1236] Overall Loss 0.343099 Objective Loss 0.343099 LR 0.001000 Time 0.023047 +2023-10-05 20:54:20,602 - Epoch: [26][ 320/ 1236] Overall Loss 0.343610 Objective Loss 0.343610 LR 0.001000 Time 0.022951 +2023-10-05 20:54:20,805 - Epoch: [26][ 330/ 1236] Overall Loss 0.343138 Objective Loss 0.343138 LR 0.001000 Time 0.022869 +2023-10-05 20:54:21,009 - Epoch: [26][ 340/ 1236] Overall Loss 0.343493 Objective Loss 0.343493 LR 0.001000 Time 0.022797 +2023-10-05 20:54:21,222 - Epoch: [26][ 350/ 1236] Overall Loss 0.343822 Objective Loss 0.343822 LR 0.001000 Time 0.022751 +2023-10-05 20:54:21,429 - Epoch: [26][ 360/ 1236] Overall Loss 0.343982 Objective Loss 0.343982 LR 0.001000 Time 0.022695 +2023-10-05 20:54:21,641 - Epoch: [26][ 370/ 1236] Overall Loss 0.345056 Objective Loss 0.345056 LR 0.001000 Time 0.022654 +2023-10-05 20:54:21,849 - Epoch: [26][ 380/ 1236] Overall Loss 0.345238 Objective Loss 0.345238 LR 0.001000 Time 0.022604 +2023-10-05 20:54:22,061 - Epoch: [26][ 390/ 1236] Overall Loss 0.345495 Objective Loss 0.345495 LR 0.001000 Time 0.022568 +2023-10-05 20:54:22,269 - Epoch: [26][ 400/ 1236] Overall Loss 0.345875 Objective Loss 0.345875 LR 0.001000 Time 0.022522 +2023-10-05 20:54:22,481 - Epoch: [26][ 410/ 1236] Overall Loss 0.345940 Objective Loss 0.345940 LR 0.001000 Time 0.022490 +2023-10-05 20:54:22,689 - Epoch: [26][ 420/ 1236] Overall Loss 0.346088 Objective Loss 0.346088 LR 0.001000 Time 0.022449 +2023-10-05 20:54:22,901 - Epoch: [26][ 430/ 1236] Overall Loss 0.346679 Objective Loss 0.346679 LR 0.001000 Time 0.022420 +2023-10-05 20:54:23,109 - Epoch: [26][ 440/ 1236] Overall Loss 0.346843 Objective Loss 0.346843 LR 0.001000 Time 0.022381 +2023-10-05 20:54:23,321 - Epoch: [26][ 450/ 1236] Overall Loss 0.347352 Objective Loss 0.347352 LR 0.001000 Time 0.022355 +2023-10-05 20:54:23,529 - Epoch: [26][ 460/ 1236] Overall Loss 0.347627 Objective Loss 0.347627 LR 0.001000 Time 0.022320 +2023-10-05 20:54:23,742 - Epoch: [26][ 470/ 1236] Overall Loss 0.347566 Objective Loss 0.347566 LR 0.001000 Time 0.022297 +2023-10-05 20:54:23,949 - Epoch: [26][ 480/ 1236] Overall Loss 0.347282 Objective Loss 0.347282 LR 0.001000 Time 0.022264 +2023-10-05 20:54:24,161 - Epoch: [26][ 490/ 1236] Overall Loss 0.347226 Objective Loss 0.347226 LR 0.001000 Time 0.022242 +2023-10-05 20:54:24,369 - Epoch: [26][ 500/ 1236] Overall Loss 0.347672 Objective Loss 0.347672 LR 0.001000 Time 0.022212 +2023-10-05 20:54:24,581 - Epoch: [26][ 510/ 1236] Overall Loss 0.348160 Objective Loss 0.348160 LR 0.001000 Time 0.022192 +2023-10-05 20:54:24,789 - Epoch: [26][ 520/ 1236] Overall Loss 0.348630 Objective Loss 0.348630 LR 0.001000 Time 0.022164 +2023-10-05 20:54:25,001 - Epoch: [26][ 530/ 1236] Overall Loss 0.348642 Objective Loss 0.348642 LR 0.001000 Time 0.022146 +2023-10-05 20:54:25,209 - Epoch: [26][ 540/ 1236] Overall Loss 0.348613 Objective Loss 0.348613 LR 0.001000 Time 0.022120 +2023-10-05 20:54:25,421 - Epoch: [26][ 550/ 1236] Overall Loss 0.348747 Objective Loss 0.348747 LR 0.001000 Time 0.022103 +2023-10-05 20:54:25,629 - Epoch: [26][ 560/ 1236] Overall Loss 0.348567 Objective Loss 0.348567 LR 0.001000 Time 0.022079 +2023-10-05 20:54:25,841 - Epoch: [26][ 570/ 1236] Overall Loss 0.348177 Objective Loss 0.348177 LR 0.001000 Time 0.022064 +2023-10-05 20:54:26,049 - Epoch: [26][ 580/ 1236] Overall Loss 0.348929 Objective Loss 0.348929 LR 0.001000 Time 0.022041 +2023-10-05 20:54:26,261 - Epoch: [26][ 590/ 1236] Overall Loss 0.349346 Objective Loss 0.349346 LR 0.001000 Time 0.022027 +2023-10-05 20:54:26,469 - Epoch: [26][ 600/ 1236] Overall Loss 0.349805 Objective Loss 0.349805 LR 0.001000 Time 0.022005 +2023-10-05 20:54:26,682 - Epoch: [26][ 610/ 1236] Overall Loss 0.349933 Objective Loss 0.349933 LR 0.001000 Time 0.021992 +2023-10-05 20:54:26,889 - Epoch: [26][ 620/ 1236] Overall Loss 0.349837 Objective Loss 0.349837 LR 0.001000 Time 0.021972 +2023-10-05 20:54:27,101 - Epoch: [26][ 630/ 1236] Overall Loss 0.349425 Objective Loss 0.349425 LR 0.001000 Time 0.021960 +2023-10-05 20:54:27,309 - Epoch: [26][ 640/ 1236] Overall Loss 0.350039 Objective Loss 0.350039 LR 0.001000 Time 0.021941 +2023-10-05 20:54:27,522 - Epoch: [26][ 650/ 1236] Overall Loss 0.350298 Objective Loss 0.350298 LR 0.001000 Time 0.021929 +2023-10-05 20:54:27,729 - Epoch: [26][ 660/ 1236] Overall Loss 0.350243 Objective Loss 0.350243 LR 0.001000 Time 0.021911 +2023-10-05 20:54:27,942 - Epoch: [26][ 670/ 1236] Overall Loss 0.350231 Objective Loss 0.350231 LR 0.001000 Time 0.021901 +2023-10-05 20:54:28,149 - Epoch: [26][ 680/ 1236] Overall Loss 0.350455 Objective Loss 0.350455 LR 0.001000 Time 0.021884 +2023-10-05 20:54:28,362 - Epoch: [26][ 690/ 1236] Overall Loss 0.351189 Objective Loss 0.351189 LR 0.001000 Time 0.021874 +2023-10-05 20:54:28,570 - Epoch: [26][ 700/ 1236] Overall Loss 0.351239 Objective Loss 0.351239 LR 0.001000 Time 0.021858 +2023-10-05 20:54:28,782 - Epoch: [26][ 710/ 1236] Overall Loss 0.351844 Objective Loss 0.351844 LR 0.001000 Time 0.021848 +2023-10-05 20:54:28,990 - Epoch: [26][ 720/ 1236] Overall Loss 0.351976 Objective Loss 0.351976 LR 0.001000 Time 0.021833 +2023-10-05 20:54:29,202 - Epoch: [26][ 730/ 1236] Overall Loss 0.351572 Objective Loss 0.351572 LR 0.001000 Time 0.021824 +2023-10-05 20:54:29,409 - Epoch: [26][ 740/ 1236] Overall Loss 0.351559 Objective Loss 0.351559 LR 0.001000 Time 0.021810 +2023-10-05 20:54:29,621 - Epoch: [26][ 750/ 1236] Overall Loss 0.351734 Objective Loss 0.351734 LR 0.001000 Time 0.021801 +2023-10-05 20:54:29,830 - Epoch: [26][ 760/ 1236] Overall Loss 0.351423 Objective Loss 0.351423 LR 0.001000 Time 0.021788 +2023-10-05 20:54:30,042 - Epoch: [26][ 770/ 1236] Overall Loss 0.351011 Objective Loss 0.351011 LR 0.001000 Time 0.021780 +2023-10-05 20:54:30,250 - Epoch: [26][ 780/ 1236] Overall Loss 0.350945 Objective Loss 0.350945 LR 0.001000 Time 0.021767 +2023-10-05 20:54:30,462 - Epoch: [26][ 790/ 1236] Overall Loss 0.351295 Objective Loss 0.351295 LR 0.001000 Time 0.021759 +2023-10-05 20:54:30,669 - Epoch: [26][ 800/ 1236] Overall Loss 0.351309 Objective Loss 0.351309 LR 0.001000 Time 0.021747 +2023-10-05 20:54:30,882 - Epoch: [26][ 810/ 1236] Overall Loss 0.351173 Objective Loss 0.351173 LR 0.001000 Time 0.021740 +2023-10-05 20:54:31,089 - Epoch: [26][ 820/ 1236] Overall Loss 0.351084 Objective Loss 0.351084 LR 0.001000 Time 0.021728 +2023-10-05 20:54:31,302 - Epoch: [26][ 830/ 1236] Overall Loss 0.350644 Objective Loss 0.350644 LR 0.001000 Time 0.021721 +2023-10-05 20:54:31,509 - Epoch: [26][ 840/ 1236] Overall Loss 0.350632 Objective Loss 0.350632 LR 0.001000 Time 0.021710 +2023-10-05 20:54:31,722 - Epoch: [26][ 850/ 1236] Overall Loss 0.350736 Objective Loss 0.350736 LR 0.001000 Time 0.021704 +2023-10-05 20:54:31,929 - Epoch: [26][ 860/ 1236] Overall Loss 0.350888 Objective Loss 0.350888 LR 0.001000 Time 0.021693 +2023-10-05 20:54:32,142 - Epoch: [26][ 870/ 1236] Overall Loss 0.351318 Objective Loss 0.351318 LR 0.001000 Time 0.021687 +2023-10-05 20:54:32,349 - Epoch: [26][ 880/ 1236] Overall Loss 0.351609 Objective Loss 0.351609 LR 0.001000 Time 0.021676 +2023-10-05 20:54:32,562 - Epoch: [26][ 890/ 1236] Overall Loss 0.351505 Objective Loss 0.351505 LR 0.001000 Time 0.021671 +2023-10-05 20:54:32,770 - Epoch: [26][ 900/ 1236] Overall Loss 0.351532 Objective Loss 0.351532 LR 0.001000 Time 0.021661 +2023-10-05 20:54:32,982 - Epoch: [26][ 910/ 1236] Overall Loss 0.351483 Objective Loss 0.351483 LR 0.001000 Time 0.021655 +2023-10-05 20:54:33,190 - Epoch: [26][ 920/ 1236] Overall Loss 0.351473 Objective Loss 0.351473 LR 0.001000 Time 0.021646 +2023-10-05 20:54:33,402 - Epoch: [26][ 930/ 1236] Overall Loss 0.351333 Objective Loss 0.351333 LR 0.001000 Time 0.021641 +2023-10-05 20:54:33,610 - Epoch: [26][ 940/ 1236] Overall Loss 0.351125 Objective Loss 0.351125 LR 0.001000 Time 0.021631 +2023-10-05 20:54:33,822 - Epoch: [26][ 950/ 1236] Overall Loss 0.351237 Objective Loss 0.351237 LR 0.001000 Time 0.021627 +2023-10-05 20:54:34,030 - Epoch: [26][ 960/ 1236] Overall Loss 0.351316 Objective Loss 0.351316 LR 0.001000 Time 0.021618 +2023-10-05 20:54:34,242 - Epoch: [26][ 970/ 1236] Overall Loss 0.351363 Objective Loss 0.351363 LR 0.001000 Time 0.021613 +2023-10-05 20:54:34,450 - Epoch: [26][ 980/ 1236] Overall Loss 0.351552 Objective Loss 0.351552 LR 0.001000 Time 0.021605 +2023-10-05 20:54:34,662 - Epoch: [26][ 990/ 1236] Overall Loss 0.351590 Objective Loss 0.351590 LR 0.001000 Time 0.021600 +2023-10-05 20:54:34,870 - Epoch: [26][ 1000/ 1236] Overall Loss 0.351510 Objective Loss 0.351510 LR 0.001000 Time 0.021592 +2023-10-05 20:54:35,082 - Epoch: [26][ 1010/ 1236] Overall Loss 0.351708 Objective Loss 0.351708 LR 0.001000 Time 0.021588 +2023-10-05 20:54:35,289 - Epoch: [26][ 1020/ 1236] Overall Loss 0.351991 Objective Loss 0.351991 LR 0.001000 Time 0.021579 +2023-10-05 20:54:35,502 - Epoch: [26][ 1030/ 1236] Overall Loss 0.352176 Objective Loss 0.352176 LR 0.001000 Time 0.021576 +2023-10-05 20:54:35,710 - Epoch: [26][ 1040/ 1236] Overall Loss 0.352090 Objective Loss 0.352090 LR 0.001000 Time 0.021568 +2023-10-05 20:54:35,922 - Epoch: [26][ 1050/ 1236] Overall Loss 0.351899 Objective Loss 0.351899 LR 0.001000 Time 0.021564 +2023-10-05 20:54:36,130 - Epoch: [26][ 1060/ 1236] Overall Loss 0.351928 Objective Loss 0.351928 LR 0.001000 Time 0.021556 +2023-10-05 20:54:36,342 - Epoch: [26][ 1070/ 1236] Overall Loss 0.351847 Objective Loss 0.351847 LR 0.001000 Time 0.021553 +2023-10-05 20:54:36,549 - Epoch: [26][ 1080/ 1236] Overall Loss 0.351815 Objective Loss 0.351815 LR 0.001000 Time 0.021545 +2023-10-05 20:54:36,762 - Epoch: [26][ 1090/ 1236] Overall Loss 0.352011 Objective Loss 0.352011 LR 0.001000 Time 0.021542 +2023-10-05 20:54:36,970 - Epoch: [26][ 1100/ 1236] Overall Loss 0.352084 Objective Loss 0.352084 LR 0.001000 Time 0.021535 +2023-10-05 20:54:37,182 - Epoch: [26][ 1110/ 1236] Overall Loss 0.352039 Objective Loss 0.352039 LR 0.001000 Time 0.021532 +2023-10-05 20:54:37,389 - Epoch: [26][ 1120/ 1236] Overall Loss 0.351745 Objective Loss 0.351745 LR 0.001000 Time 0.021525 +2023-10-05 20:54:37,602 - Epoch: [26][ 1130/ 1236] Overall Loss 0.351589 Objective Loss 0.351589 LR 0.001000 Time 0.021522 +2023-10-05 20:54:37,806 - Epoch: [26][ 1140/ 1236] Overall Loss 0.351527 Objective Loss 0.351527 LR 0.001000 Time 0.021512 +2023-10-05 20:54:38,008 - Epoch: [26][ 1150/ 1236] Overall Loss 0.351672 Objective Loss 0.351672 LR 0.001000 Time 0.021501 +2023-10-05 20:54:38,210 - Epoch: [26][ 1160/ 1236] Overall Loss 0.351663 Objective Loss 0.351663 LR 0.001000 Time 0.021489 +2023-10-05 20:54:38,412 - Epoch: [26][ 1170/ 1236] Overall Loss 0.351541 Objective Loss 0.351541 LR 0.001000 Time 0.021478 +2023-10-05 20:54:38,613 - Epoch: [26][ 1180/ 1236] Overall Loss 0.351794 Objective Loss 0.351794 LR 0.001000 Time 0.021466 +2023-10-05 20:54:38,816 - Epoch: [26][ 1190/ 1236] Overall Loss 0.351950 Objective Loss 0.351950 LR 0.001000 Time 0.021455 +2023-10-05 20:54:39,017 - Epoch: [26][ 1200/ 1236] Overall Loss 0.351889 Objective Loss 0.351889 LR 0.001000 Time 0.021444 +2023-10-05 20:54:39,219 - Epoch: [26][ 1210/ 1236] Overall Loss 0.351898 Objective Loss 0.351898 LR 0.001000 Time 0.021434 +2023-10-05 20:54:39,421 - Epoch: [26][ 1220/ 1236] Overall Loss 0.351809 Objective Loss 0.351809 LR 0.001000 Time 0.021423 +2023-10-05 20:54:39,675 - Epoch: [26][ 1230/ 1236] Overall Loss 0.351797 Objective Loss 0.351797 LR 0.001000 Time 0.021456 +2023-10-05 20:54:39,792 - Epoch: [26][ 1236/ 1236] Overall Loss 0.351589 Objective Loss 0.351589 Top1 85.947047 Top5 97.352342 LR 0.001000 Time 0.021446 +2023-10-05 20:54:39,914 - --- validate (epoch=26)----------- +2023-10-05 20:54:39,915 - 29943 samples (256 per mini-batch) +2023-10-05 20:54:40,370 - Epoch: [26][ 10/ 117] Loss 0.379081 Top1 81.210938 Top5 97.460938 +2023-10-05 20:54:40,521 - Epoch: [26][ 20/ 117] Loss 0.370015 Top1 81.679688 Top5 97.617188 +2023-10-05 20:54:40,672 - Epoch: [26][ 30/ 117] Loss 0.374868 Top1 81.276042 Top5 97.617188 +2023-10-05 20:54:40,824 - Epoch: [26][ 40/ 117] Loss 0.379425 Top1 81.113281 Top5 97.568359 +2023-10-05 20:54:40,974 - Epoch: [26][ 50/ 117] Loss 0.374803 Top1 80.953125 Top5 97.531250 +2023-10-05 20:54:41,125 - Epoch: [26][ 60/ 117] Loss 0.384143 Top1 80.592448 Top5 97.493490 +2023-10-05 20:54:41,276 - Epoch: [26][ 70/ 117] Loss 0.385070 Top1 80.563616 Top5 97.427455 +2023-10-05 20:54:41,425 - Epoch: [26][ 80/ 117] Loss 0.381598 Top1 80.566406 Top5 97.465820 +2023-10-05 20:54:41,573 - Epoch: [26][ 90/ 117] Loss 0.379340 Top1 80.499132 Top5 97.408854 +2023-10-05 20:54:41,721 - Epoch: [26][ 100/ 117] Loss 0.378102 Top1 80.390625 Top5 97.417969 +2023-10-05 20:54:41,876 - Epoch: [26][ 110/ 117] Loss 0.379900 Top1 80.230824 Top5 97.432528 +2023-10-05 20:54:41,961 - Epoch: [26][ 117/ 117] Loss 0.382225 Top1 80.172327 Top5 97.405070 +2023-10-05 20:54:42,096 - ==> Top1: 80.172 Top5: 97.405 Loss: 0.382 + +2023-10-05 20:54:42,097 - ==> Confusion: +[[ 943 2 2 2 16 4 0 0 13 42 1 0 1 1 5 0 7 1 0 0 10] + [ 2 1035 0 0 7 31 2 25 1 0 0 4 0 1 1 3 6 0 8 1 4] + [ 8 0 935 25 4 0 26 10 0 1 6 3 6 3 1 4 1 2 5 2 14] + [ 4 1 13 924 1 10 3 1 4 0 15 0 8 2 45 6 2 11 16 2 21] + [ 37 11 0 0 957 5 0 0 1 5 0 1 1 2 6 6 10 1 0 1 6] + [ 7 48 0 2 3 992 1 19 1 0 3 9 1 18 3 1 2 0 1 2 3] + [ 0 7 33 0 1 0 1093 11 0 0 4 6 1 1 0 15 0 1 0 8 10] + [ 6 20 18 0 0 32 3 1059 0 0 2 14 2 2 0 1 1 0 41 7 10] + [ 19 7 0 0 0 4 0 0 947 41 15 4 1 22 19 4 0 0 5 1 0] + [ 152 1 0 0 12 6 0 0 35 852 0 0 0 34 5 9 2 1 0 3 7] + [ 6 4 8 3 1 3 2 6 12 0 965 5 0 14 5 2 0 0 8 0 9] + [ 2 1 0 0 3 18 0 1 0 0 0 962 18 4 0 1 3 14 0 4 4] + [ 1 0 1 4 0 4 0 0 4 0 1 63 930 3 2 7 10 24 0 5 9] + [ 4 2 0 2 5 12 1 1 6 7 8 6 1 1044 3 3 1 1 0 1 11] + [ 14 3 0 6 8 2 0 0 30 5 0 0 4 2 983 0 3 7 13 0 21] + [ 0 4 1 1 4 1 0 0 0 0 1 11 9 3 1 1047 24 12 0 7 8] + [ 1 17 0 1 8 8 2 0 3 0 0 3 2 2 2 8 1090 0 0 6 8] + [ 1 0 1 0 2 0 0 0 0 0 0 5 36 2 1 3 0 980 1 0 6] + [ 2 16 7 11 1 1 1 40 5 0 4 3 5 0 13 1 2 0 944 2 10] + [ 0 1 7 1 1 9 7 12 0 0 1 22 8 2 0 1 19 2 2 1050 7] + [ 197 328 134 54 127 280 42 125 126 54 183 168 470 341 141 75 245 69 190 282 4274]] + +2023-10-05 20:54:42,098 - ==> Best [Top1: 80.172 Top5: 97.405 Sparsity:0.00 Params: 148928 on epoch: 26] +2023-10-05 20:54:42,098 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:54:42,105 - + +2023-10-05 20:54:42,105 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:54:43,095 - Epoch: [27][ 10/ 1236] Overall Loss 0.346628 Objective Loss 0.346628 LR 0.001000 Time 0.098994 +2023-10-05 20:54:43,298 - Epoch: [27][ 20/ 1236] Overall Loss 0.343533 Objective Loss 0.343533 LR 0.001000 Time 0.059618 +2023-10-05 20:54:43,500 - Epoch: [27][ 30/ 1236] Overall Loss 0.342954 Objective Loss 0.342954 LR 0.001000 Time 0.046476 +2023-10-05 20:54:43,703 - Epoch: [27][ 40/ 1236] Overall Loss 0.342337 Objective Loss 0.342337 LR 0.001000 Time 0.039912 +2023-10-05 20:54:43,905 - Epoch: [27][ 50/ 1236] Overall Loss 0.348101 Objective Loss 0.348101 LR 0.001000 Time 0.035964 +2023-10-05 20:54:44,107 - Epoch: [27][ 60/ 1236] Overall Loss 0.350648 Objective Loss 0.350648 LR 0.001000 Time 0.033331 +2023-10-05 20:54:44,310 - Epoch: [27][ 70/ 1236] Overall Loss 0.346120 Objective Loss 0.346120 LR 0.001000 Time 0.031459 +2023-10-05 20:54:44,512 - Epoch: [27][ 80/ 1236] Overall Loss 0.347852 Objective Loss 0.347852 LR 0.001000 Time 0.030053 +2023-10-05 20:54:44,715 - Epoch: [27][ 90/ 1236] Overall Loss 0.347949 Objective Loss 0.347949 LR 0.001000 Time 0.028961 +2023-10-05 20:54:44,917 - Epoch: [27][ 100/ 1236] Overall Loss 0.345471 Objective Loss 0.345471 LR 0.001000 Time 0.028087 +2023-10-05 20:54:45,120 - Epoch: [27][ 110/ 1236] Overall Loss 0.345137 Objective Loss 0.345137 LR 0.001000 Time 0.027371 +2023-10-05 20:54:45,320 - Epoch: [27][ 120/ 1236] Overall Loss 0.343861 Objective Loss 0.343861 LR 0.001000 Time 0.026758 +2023-10-05 20:54:45,520 - Epoch: [27][ 130/ 1236] Overall Loss 0.342779 Objective Loss 0.342779 LR 0.001000 Time 0.026234 +2023-10-05 20:54:45,720 - Epoch: [27][ 140/ 1236] Overall Loss 0.344087 Objective Loss 0.344087 LR 0.001000 Time 0.025788 +2023-10-05 20:54:45,921 - Epoch: [27][ 150/ 1236] Overall Loss 0.342876 Objective Loss 0.342876 LR 0.001000 Time 0.025403 +2023-10-05 20:54:46,125 - Epoch: [27][ 160/ 1236] Overall Loss 0.343480 Objective Loss 0.343480 LR 0.001000 Time 0.025090 +2023-10-05 20:54:46,326 - Epoch: [27][ 170/ 1236] Overall Loss 0.342855 Objective Loss 0.342855 LR 0.001000 Time 0.024796 +2023-10-05 20:54:46,529 - Epoch: [27][ 180/ 1236] Overall Loss 0.343874 Objective Loss 0.343874 LR 0.001000 Time 0.024543 +2023-10-05 20:54:46,729 - Epoch: [27][ 190/ 1236] Overall Loss 0.342553 Objective Loss 0.342553 LR 0.001000 Time 0.024301 +2023-10-05 20:54:46,929 - Epoch: [27][ 200/ 1236] Overall Loss 0.341481 Objective Loss 0.341481 LR 0.001000 Time 0.024086 +2023-10-05 20:54:47,129 - Epoch: [27][ 210/ 1236] Overall Loss 0.341541 Objective Loss 0.341541 LR 0.001000 Time 0.023889 +2023-10-05 20:54:47,329 - Epoch: [27][ 220/ 1236] Overall Loss 0.343515 Objective Loss 0.343515 LR 0.001000 Time 0.023711 +2023-10-05 20:54:47,529 - Epoch: [27][ 230/ 1236] Overall Loss 0.342663 Objective Loss 0.342663 LR 0.001000 Time 0.023547 +2023-10-05 20:54:47,731 - Epoch: [27][ 240/ 1236] Overall Loss 0.343820 Objective Loss 0.343820 LR 0.001000 Time 0.023406 +2023-10-05 20:54:47,932 - Epoch: [27][ 250/ 1236] Overall Loss 0.344681 Objective Loss 0.344681 LR 0.001000 Time 0.023275 +2023-10-05 20:54:48,137 - Epoch: [27][ 260/ 1236] Overall Loss 0.344964 Objective Loss 0.344964 LR 0.001000 Time 0.023167 +2023-10-05 20:54:48,338 - Epoch: [27][ 270/ 1236] Overall Loss 0.345709 Objective Loss 0.345709 LR 0.001000 Time 0.023050 +2023-10-05 20:54:48,538 - Epoch: [27][ 280/ 1236] Overall Loss 0.345291 Objective Loss 0.345291 LR 0.001000 Time 0.022939 +2023-10-05 20:54:48,738 - Epoch: [27][ 290/ 1236] Overall Loss 0.345189 Objective Loss 0.345189 LR 0.001000 Time 0.022837 +2023-10-05 20:54:48,938 - Epoch: [27][ 300/ 1236] Overall Loss 0.344097 Objective Loss 0.344097 LR 0.001000 Time 0.022743 +2023-10-05 20:54:49,138 - Epoch: [27][ 310/ 1236] Overall Loss 0.344837 Objective Loss 0.344837 LR 0.001000 Time 0.022652 +2023-10-05 20:54:49,339 - Epoch: [27][ 320/ 1236] Overall Loss 0.344746 Objective Loss 0.344746 LR 0.001000 Time 0.022571 +2023-10-05 20:54:49,539 - Epoch: [27][ 330/ 1236] Overall Loss 0.344388 Objective Loss 0.344388 LR 0.001000 Time 0.022492 +2023-10-05 20:54:49,739 - Epoch: [27][ 340/ 1236] Overall Loss 0.344865 Objective Loss 0.344865 LR 0.001000 Time 0.022419 +2023-10-05 20:54:49,940 - Epoch: [27][ 350/ 1236] Overall Loss 0.345521 Objective Loss 0.345521 LR 0.001000 Time 0.022351 +2023-10-05 20:54:50,140 - Epoch: [27][ 360/ 1236] Overall Loss 0.345330 Objective Loss 0.345330 LR 0.001000 Time 0.022285 +2023-10-05 20:54:50,342 - Epoch: [27][ 370/ 1236] Overall Loss 0.345341 Objective Loss 0.345341 LR 0.001000 Time 0.022228 +2023-10-05 20:54:50,545 - Epoch: [27][ 380/ 1236] Overall Loss 0.344720 Objective Loss 0.344720 LR 0.001000 Time 0.022175 +2023-10-05 20:54:50,747 - Epoch: [27][ 390/ 1236] Overall Loss 0.345164 Objective Loss 0.345164 LR 0.001000 Time 0.022125 +2023-10-05 20:54:50,950 - Epoch: [27][ 400/ 1236] Overall Loss 0.344852 Objective Loss 0.344852 LR 0.001000 Time 0.022077 +2023-10-05 20:54:51,152 - Epoch: [27][ 410/ 1236] Overall Loss 0.344925 Objective Loss 0.344925 LR 0.001000 Time 0.022031 +2023-10-05 20:54:51,354 - Epoch: [27][ 420/ 1236] Overall Loss 0.345325 Objective Loss 0.345325 LR 0.001000 Time 0.021988 +2023-10-05 20:54:51,557 - Epoch: [27][ 430/ 1236] Overall Loss 0.345695 Objective Loss 0.345695 LR 0.001000 Time 0.021946 +2023-10-05 20:54:51,759 - Epoch: [27][ 440/ 1236] Overall Loss 0.345125 Objective Loss 0.345125 LR 0.001000 Time 0.021907 +2023-10-05 20:54:51,962 - Epoch: [27][ 450/ 1236] Overall Loss 0.344750 Objective Loss 0.344750 LR 0.001000 Time 0.021870 +2023-10-05 20:54:52,164 - Epoch: [27][ 460/ 1236] Overall Loss 0.345003 Objective Loss 0.345003 LR 0.001000 Time 0.021834 +2023-10-05 20:54:52,367 - Epoch: [27][ 470/ 1236] Overall Loss 0.345694 Objective Loss 0.345694 LR 0.001000 Time 0.021799 +2023-10-05 20:54:52,569 - Epoch: [27][ 480/ 1236] Overall Loss 0.346182 Objective Loss 0.346182 LR 0.001000 Time 0.021766 +2023-10-05 20:54:52,772 - Epoch: [27][ 490/ 1236] Overall Loss 0.346162 Objective Loss 0.346162 LR 0.001000 Time 0.021735 +2023-10-05 20:54:52,974 - Epoch: [27][ 500/ 1236] Overall Loss 0.346315 Objective Loss 0.346315 LR 0.001000 Time 0.021704 +2023-10-05 20:54:53,177 - Epoch: [27][ 510/ 1236] Overall Loss 0.346266 Objective Loss 0.346266 LR 0.001000 Time 0.021675 +2023-10-05 20:54:53,379 - Epoch: [27][ 520/ 1236] Overall Loss 0.346480 Objective Loss 0.346480 LR 0.001000 Time 0.021647 +2023-10-05 20:54:53,582 - Epoch: [27][ 530/ 1236] Overall Loss 0.346968 Objective Loss 0.346968 LR 0.001000 Time 0.021620 +2023-10-05 20:54:53,784 - Epoch: [27][ 540/ 1236] Overall Loss 0.347349 Objective Loss 0.347349 LR 0.001000 Time 0.021594 +2023-10-05 20:54:53,986 - Epoch: [27][ 550/ 1236] Overall Loss 0.347552 Objective Loss 0.347552 LR 0.001000 Time 0.021568 +2023-10-05 20:54:54,189 - Epoch: [27][ 560/ 1236] Overall Loss 0.348084 Objective Loss 0.348084 LR 0.001000 Time 0.021544 +2023-10-05 20:54:54,391 - Epoch: [27][ 570/ 1236] Overall Loss 0.347756 Objective Loss 0.347756 LR 0.001000 Time 0.021518 +2023-10-05 20:54:54,593 - Epoch: [27][ 580/ 1236] Overall Loss 0.347823 Objective Loss 0.347823 LR 0.001000 Time 0.021496 +2023-10-05 20:54:54,796 - Epoch: [27][ 590/ 1236] Overall Loss 0.347539 Objective Loss 0.347539 LR 0.001000 Time 0.021474 +2023-10-05 20:54:54,998 - Epoch: [27][ 600/ 1236] Overall Loss 0.347127 Objective Loss 0.347127 LR 0.001000 Time 0.021453 +2023-10-05 20:54:55,201 - Epoch: [27][ 610/ 1236] Overall Loss 0.347076 Objective Loss 0.347076 LR 0.001000 Time 0.021433 +2023-10-05 20:54:55,403 - Epoch: [27][ 620/ 1236] Overall Loss 0.346997 Objective Loss 0.346997 LR 0.001000 Time 0.021413 +2023-10-05 20:54:55,606 - Epoch: [27][ 630/ 1236] Overall Loss 0.346944 Objective Loss 0.346944 LR 0.001000 Time 0.021394 +2023-10-05 20:54:55,808 - Epoch: [27][ 640/ 1236] Overall Loss 0.346534 Objective Loss 0.346534 LR 0.001000 Time 0.021376 +2023-10-05 20:54:56,011 - Epoch: [27][ 650/ 1236] Overall Loss 0.346823 Objective Loss 0.346823 LR 0.001000 Time 0.021358 +2023-10-05 20:54:56,213 - Epoch: [27][ 660/ 1236] Overall Loss 0.346518 Objective Loss 0.346518 LR 0.001000 Time 0.021341 +2023-10-05 20:54:56,416 - Epoch: [27][ 670/ 1236] Overall Loss 0.346060 Objective Loss 0.346060 LR 0.001000 Time 0.021324 +2023-10-05 20:54:56,618 - Epoch: [27][ 680/ 1236] Overall Loss 0.346374 Objective Loss 0.346374 LR 0.001000 Time 0.021308 +2023-10-05 20:54:56,821 - Epoch: [27][ 690/ 1236] Overall Loss 0.346576 Objective Loss 0.346576 LR 0.001000 Time 0.021292 +2023-10-05 20:54:57,023 - Epoch: [27][ 700/ 1236] Overall Loss 0.347015 Objective Loss 0.347015 LR 0.001000 Time 0.021277 +2023-10-05 20:54:57,225 - Epoch: [27][ 710/ 1236] Overall Loss 0.347137 Objective Loss 0.347137 LR 0.001000 Time 0.021260 +2023-10-05 20:54:57,426 - Epoch: [27][ 720/ 1236] Overall Loss 0.346951 Objective Loss 0.346951 LR 0.001000 Time 0.021244 +2023-10-05 20:54:57,628 - Epoch: [27][ 730/ 1236] Overall Loss 0.347181 Objective Loss 0.347181 LR 0.001000 Time 0.021229 +2023-10-05 20:54:57,830 - Epoch: [27][ 740/ 1236] Overall Loss 0.346944 Objective Loss 0.346944 LR 0.001000 Time 0.021215 +2023-10-05 20:54:58,033 - Epoch: [27][ 750/ 1236] Overall Loss 0.346632 Objective Loss 0.346632 LR 0.001000 Time 0.021202 +2023-10-05 20:54:58,236 - Epoch: [27][ 760/ 1236] Overall Loss 0.346776 Objective Loss 0.346776 LR 0.001000 Time 0.021189 +2023-10-05 20:54:58,438 - Epoch: [27][ 770/ 1236] Overall Loss 0.346787 Objective Loss 0.346787 LR 0.001000 Time 0.021177 +2023-10-05 20:54:58,641 - Epoch: [27][ 780/ 1236] Overall Loss 0.346771 Objective Loss 0.346771 LR 0.001000 Time 0.021164 +2023-10-05 20:54:58,843 - Epoch: [27][ 790/ 1236] Overall Loss 0.346993 Objective Loss 0.346993 LR 0.001000 Time 0.021152 +2023-10-05 20:54:59,045 - Epoch: [27][ 800/ 1236] Overall Loss 0.346906 Objective Loss 0.346906 LR 0.001000 Time 0.021140 +2023-10-05 20:54:59,248 - Epoch: [27][ 810/ 1236] Overall Loss 0.346785 Objective Loss 0.346785 LR 0.001000 Time 0.021129 +2023-10-05 20:54:59,451 - Epoch: [27][ 820/ 1236] Overall Loss 0.346690 Objective Loss 0.346690 LR 0.001000 Time 0.021118 +2023-10-05 20:54:59,653 - Epoch: [27][ 830/ 1236] Overall Loss 0.346774 Objective Loss 0.346774 LR 0.001000 Time 0.021107 +2023-10-05 20:54:59,856 - Epoch: [27][ 840/ 1236] Overall Loss 0.346711 Objective Loss 0.346711 LR 0.001000 Time 0.021097 +2023-10-05 20:55:00,058 - Epoch: [27][ 850/ 1236] Overall Loss 0.346814 Objective Loss 0.346814 LR 0.001000 Time 0.021087 +2023-10-05 20:55:00,261 - Epoch: [27][ 860/ 1236] Overall Loss 0.346872 Objective Loss 0.346872 LR 0.001000 Time 0.021076 +2023-10-05 20:55:00,463 - Epoch: [27][ 870/ 1236] Overall Loss 0.347153 Objective Loss 0.347153 LR 0.001000 Time 0.021066 +2023-10-05 20:55:00,666 - Epoch: [27][ 880/ 1236] Overall Loss 0.347277 Objective Loss 0.347277 LR 0.001000 Time 0.021057 +2023-10-05 20:55:00,868 - Epoch: [27][ 890/ 1236] Overall Loss 0.347814 Objective Loss 0.347814 LR 0.001000 Time 0.021047 +2023-10-05 20:55:01,071 - Epoch: [27][ 900/ 1236] Overall Loss 0.347848 Objective Loss 0.347848 LR 0.001000 Time 0.021038 +2023-10-05 20:55:01,273 - Epoch: [27][ 910/ 1236] Overall Loss 0.347695 Objective Loss 0.347695 LR 0.001000 Time 0.021029 +2023-10-05 20:55:01,476 - Epoch: [27][ 920/ 1236] Overall Loss 0.347908 Objective Loss 0.347908 LR 0.001000 Time 0.021020 +2023-10-05 20:55:01,678 - Epoch: [27][ 930/ 1236] Overall Loss 0.348197 Objective Loss 0.348197 LR 0.001000 Time 0.021012 +2023-10-05 20:55:01,881 - Epoch: [27][ 940/ 1236] Overall Loss 0.348323 Objective Loss 0.348323 LR 0.001000 Time 0.021004 +2023-10-05 20:55:02,083 - Epoch: [27][ 950/ 1236] Overall Loss 0.347965 Objective Loss 0.347965 LR 0.001000 Time 0.020995 +2023-10-05 20:55:02,286 - Epoch: [27][ 960/ 1236] Overall Loss 0.347739 Objective Loss 0.347739 LR 0.001000 Time 0.020987 +2023-10-05 20:55:02,488 - Epoch: [27][ 970/ 1236] Overall Loss 0.347540 Objective Loss 0.347540 LR 0.001000 Time 0.020979 +2023-10-05 20:55:02,690 - Epoch: [27][ 980/ 1236] Overall Loss 0.347330 Objective Loss 0.347330 LR 0.001000 Time 0.020971 +2023-10-05 20:55:02,893 - Epoch: [27][ 990/ 1236] Overall Loss 0.347304 Objective Loss 0.347304 LR 0.001000 Time 0.020964 +2023-10-05 20:55:03,096 - Epoch: [27][ 1000/ 1236] Overall Loss 0.347470 Objective Loss 0.347470 LR 0.001000 Time 0.020956 +2023-10-05 20:55:03,298 - Epoch: [27][ 1010/ 1236] Overall Loss 0.347724 Objective Loss 0.347724 LR 0.001000 Time 0.020949 +2023-10-05 20:55:03,501 - Epoch: [27][ 1020/ 1236] Overall Loss 0.348118 Objective Loss 0.348118 LR 0.001000 Time 0.020942 +2023-10-05 20:55:03,703 - Epoch: [27][ 1030/ 1236] Overall Loss 0.348145 Objective Loss 0.348145 LR 0.001000 Time 0.020935 +2023-10-05 20:55:03,906 - Epoch: [27][ 1040/ 1236] Overall Loss 0.348310 Objective Loss 0.348310 LR 0.001000 Time 0.020928 +2023-10-05 20:55:04,108 - Epoch: [27][ 1050/ 1236] Overall Loss 0.348424 Objective Loss 0.348424 LR 0.001000 Time 0.020921 +2023-10-05 20:55:04,310 - Epoch: [27][ 1060/ 1236] Overall Loss 0.347960 Objective Loss 0.347960 LR 0.001000 Time 0.020914 +2023-10-05 20:55:04,513 - Epoch: [27][ 1070/ 1236] Overall Loss 0.347969 Objective Loss 0.347969 LR 0.001000 Time 0.020908 +2023-10-05 20:55:04,716 - Epoch: [27][ 1080/ 1236] Overall Loss 0.347830 Objective Loss 0.347830 LR 0.001000 Time 0.020901 +2023-10-05 20:55:04,918 - Epoch: [27][ 1090/ 1236] Overall Loss 0.347919 Objective Loss 0.347919 LR 0.001000 Time 0.020895 +2023-10-05 20:55:05,121 - Epoch: [27][ 1100/ 1236] Overall Loss 0.348086 Objective Loss 0.348086 LR 0.001000 Time 0.020889 +2023-10-05 20:55:05,323 - Epoch: [27][ 1110/ 1236] Overall Loss 0.348284 Objective Loss 0.348284 LR 0.001000 Time 0.020883 +2023-10-05 20:55:05,526 - Epoch: [27][ 1120/ 1236] Overall Loss 0.348102 Objective Loss 0.348102 LR 0.001000 Time 0.020877 +2023-10-05 20:55:05,728 - Epoch: [27][ 1130/ 1236] Overall Loss 0.348278 Objective Loss 0.348278 LR 0.001000 Time 0.020871 +2023-10-05 20:55:05,931 - Epoch: [27][ 1140/ 1236] Overall Loss 0.348407 Objective Loss 0.348407 LR 0.001000 Time 0.020865 +2023-10-05 20:55:06,133 - Epoch: [27][ 1150/ 1236] Overall Loss 0.348441 Objective Loss 0.348441 LR 0.001000 Time 0.020860 +2023-10-05 20:55:06,336 - Epoch: [27][ 1160/ 1236] Overall Loss 0.348607 Objective Loss 0.348607 LR 0.001000 Time 0.020854 +2023-10-05 20:55:06,538 - Epoch: [27][ 1170/ 1236] Overall Loss 0.348464 Objective Loss 0.348464 LR 0.001000 Time 0.020849 +2023-10-05 20:55:06,741 - Epoch: [27][ 1180/ 1236] Overall Loss 0.348418 Objective Loss 0.348418 LR 0.001000 Time 0.020843 +2023-10-05 20:55:06,944 - Epoch: [27][ 1190/ 1236] Overall Loss 0.348555 Objective Loss 0.348555 LR 0.001000 Time 0.020839 +2023-10-05 20:55:07,147 - Epoch: [27][ 1200/ 1236] Overall Loss 0.348622 Objective Loss 0.348622 LR 0.001000 Time 0.020834 +2023-10-05 20:55:07,351 - Epoch: [27][ 1210/ 1236] Overall Loss 0.348653 Objective Loss 0.348653 LR 0.001000 Time 0.020830 +2023-10-05 20:55:07,555 - Epoch: [27][ 1220/ 1236] Overall Loss 0.348307 Objective Loss 0.348307 LR 0.001000 Time 0.020826 +2023-10-05 20:55:07,810 - Epoch: [27][ 1230/ 1236] Overall Loss 0.348082 Objective Loss 0.348082 LR 0.001000 Time 0.020864 +2023-10-05 20:55:07,927 - Epoch: [27][ 1236/ 1236] Overall Loss 0.347938 Objective Loss 0.347938 Top1 82.077393 Top5 96.537678 LR 0.001000 Time 0.020857 +2023-10-05 20:55:08,061 - --- validate (epoch=27)----------- +2023-10-05 20:55:08,062 - 29943 samples (256 per mini-batch) +2023-10-05 20:55:08,515 - Epoch: [27][ 10/ 117] Loss 0.368173 Top1 79.218750 Top5 97.070312 +2023-10-05 20:55:08,672 - Epoch: [27][ 20/ 117] Loss 0.364717 Top1 79.843750 Top5 97.265625 +2023-10-05 20:55:08,826 - Epoch: [27][ 30/ 117] Loss 0.368630 Top1 79.661458 Top5 97.135417 +2023-10-05 20:55:08,981 - Epoch: [27][ 40/ 117] Loss 0.371440 Top1 79.960938 Top5 97.197266 +2023-10-05 20:55:09,134 - Epoch: [27][ 50/ 117] Loss 0.378159 Top1 79.890625 Top5 97.273438 +2023-10-05 20:55:09,290 - Epoch: [27][ 60/ 117] Loss 0.377913 Top1 79.661458 Top5 97.285156 +2023-10-05 20:55:09,445 - Epoch: [27][ 70/ 117] Loss 0.375822 Top1 79.665179 Top5 97.343750 +2023-10-05 20:55:09,601 - Epoch: [27][ 80/ 117] Loss 0.377553 Top1 79.570312 Top5 97.314453 +2023-10-05 20:55:09,755 - Epoch: [27][ 90/ 117] Loss 0.381005 Top1 79.474826 Top5 97.313368 +2023-10-05 20:55:09,911 - Epoch: [27][ 100/ 117] Loss 0.378565 Top1 79.609375 Top5 97.343750 +2023-10-05 20:55:10,073 - Epoch: [27][ 110/ 117] Loss 0.378515 Top1 79.648438 Top5 97.311790 +2023-10-05 20:55:10,157 - Epoch: [27][ 117/ 117] Loss 0.378597 Top1 79.634639 Top5 97.308219 +2023-10-05 20:55:10,295 - ==> Top1: 79.635 Top5: 97.308 Loss: 0.379 + +2023-10-05 20:55:10,295 - ==> Confusion: +[[ 873 1 2 0 15 6 0 1 12 108 1 0 0 2 8 1 6 1 1 0 12] + [ 2 1040 0 0 21 14 2 27 3 0 1 2 0 1 1 2 7 0 5 1 2] + [ 4 2 918 11 6 0 37 10 0 4 9 10 5 2 2 5 3 1 10 1 16] + [ 2 3 23 919 4 8 2 4 10 1 13 1 2 0 45 2 2 10 22 1 15] + [ 15 8 0 0 983 1 0 0 1 14 0 4 1 2 5 5 7 1 1 0 2] + [ 3 73 0 0 15 939 1 24 2 5 4 11 1 16 6 0 5 0 0 2 9] + [ 0 11 16 0 0 2 1108 16 0 0 5 7 0 0 0 6 0 0 1 10 9] + [ 4 38 10 0 4 32 0 1059 2 0 4 11 0 0 0 0 0 0 31 16 7] + [ 19 4 0 0 0 0 0 1 996 39 12 3 0 8 6 0 0 0 1 0 0] + [ 61 0 0 0 9 0 1 0 43 972 0 3 1 13 2 2 0 1 0 2 9] + [ 5 6 11 2 4 3 2 5 22 3 953 4 0 10 7 0 1 0 4 2 9] + [ 0 0 0 0 1 20 0 0 0 0 1 945 28 3 0 1 3 14 0 17 2] + [ 0 1 1 3 0 1 1 1 2 0 2 55 955 1 2 4 8 12 3 8 8] + [ 2 1 1 0 7 15 0 1 25 19 14 5 1 997 4 2 3 1 0 3 18] + [ 13 1 3 5 10 0 0 0 50 7 2 1 3 1 973 0 4 4 7 0 17] + [ 0 5 1 1 2 1 0 0 1 0 0 13 7 0 0 1053 18 16 0 8 8] + [ 0 15 0 0 16 4 1 0 2 0 0 3 0 2 2 11 1089 1 0 5 10] + [ 2 0 1 1 0 1 1 0 3 1 0 8 21 0 0 6 1 985 3 0 4] + [ 1 16 6 9 3 0 0 39 7 0 7 3 4 0 12 0 2 0 947 0 12] + [ 0 1 4 1 3 5 5 20 0 1 2 18 2 0 0 4 11 1 1 1061 12] + [ 171 306 144 42 269 178 50 127 181 155 186 145 455 236 137 68 469 82 138 286 4080]] + +2023-10-05 20:55:10,297 - ==> Best [Top1: 80.172 Top5: 97.405 Sparsity:0.00 Params: 148928 on epoch: 26] +2023-10-05 20:55:10,297 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:55:10,302 - + +2023-10-05 20:55:10,302 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:55:11,281 - Epoch: [28][ 10/ 1236] Overall Loss 0.305386 Objective Loss 0.305386 LR 0.001000 Time 0.097813 +2023-10-05 20:55:11,483 - Epoch: [28][ 20/ 1236] Overall Loss 0.324008 Objective Loss 0.324008 LR 0.001000 Time 0.058987 +2023-10-05 20:55:11,687 - Epoch: [28][ 30/ 1236] Overall Loss 0.317385 Objective Loss 0.317385 LR 0.001000 Time 0.046006 +2023-10-05 20:55:11,889 - Epoch: [28][ 40/ 1236] Overall Loss 0.318938 Objective Loss 0.318938 LR 0.001000 Time 0.039538 +2023-10-05 20:55:12,090 - Epoch: [28][ 50/ 1236] Overall Loss 0.321738 Objective Loss 0.321738 LR 0.001000 Time 0.035646 +2023-10-05 20:55:12,292 - Epoch: [28][ 60/ 1236] Overall Loss 0.323204 Objective Loss 0.323204 LR 0.001000 Time 0.033064 +2023-10-05 20:55:12,493 - Epoch: [28][ 70/ 1236] Overall Loss 0.323034 Objective Loss 0.323034 LR 0.001000 Time 0.031214 +2023-10-05 20:55:12,695 - Epoch: [28][ 80/ 1236] Overall Loss 0.327373 Objective Loss 0.327373 LR 0.001000 Time 0.029825 +2023-10-05 20:55:12,896 - Epoch: [28][ 90/ 1236] Overall Loss 0.332009 Objective Loss 0.332009 LR 0.001000 Time 0.028740 +2023-10-05 20:55:13,096 - Epoch: [28][ 100/ 1236] Overall Loss 0.333694 Objective Loss 0.333694 LR 0.001000 Time 0.027867 +2023-10-05 20:55:13,299 - Epoch: [28][ 110/ 1236] Overall Loss 0.330628 Objective Loss 0.330628 LR 0.001000 Time 0.027177 +2023-10-05 20:55:13,503 - Epoch: [28][ 120/ 1236] Overall Loss 0.329833 Objective Loss 0.329833 LR 0.001000 Time 0.026612 +2023-10-05 20:55:13,706 - Epoch: [28][ 130/ 1236] Overall Loss 0.330342 Objective Loss 0.330342 LR 0.001000 Time 0.026114 +2023-10-05 20:55:13,910 - Epoch: [28][ 140/ 1236] Overall Loss 0.331276 Objective Loss 0.331276 LR 0.001000 Time 0.025704 +2023-10-05 20:55:14,115 - Epoch: [28][ 150/ 1236] Overall Loss 0.331436 Objective Loss 0.331436 LR 0.001000 Time 0.025350 +2023-10-05 20:55:14,318 - Epoch: [28][ 160/ 1236] Overall Loss 0.330134 Objective Loss 0.330134 LR 0.001000 Time 0.025036 +2023-10-05 20:55:14,523 - Epoch: [28][ 170/ 1236] Overall Loss 0.330982 Objective Loss 0.330982 LR 0.001000 Time 0.024765 +2023-10-05 20:55:14,727 - Epoch: [28][ 180/ 1236] Overall Loss 0.331820 Objective Loss 0.331820 LR 0.001000 Time 0.024521 +2023-10-05 20:55:14,931 - Epoch: [28][ 190/ 1236] Overall Loss 0.332962 Objective Loss 0.332962 LR 0.001000 Time 0.024303 +2023-10-05 20:55:15,135 - Epoch: [28][ 200/ 1236] Overall Loss 0.332797 Objective Loss 0.332797 LR 0.001000 Time 0.024109 +2023-10-05 20:55:15,339 - Epoch: [28][ 210/ 1236] Overall Loss 0.334354 Objective Loss 0.334354 LR 0.001000 Time 0.023923 +2023-10-05 20:55:15,543 - Epoch: [28][ 220/ 1236] Overall Loss 0.334228 Objective Loss 0.334228 LR 0.001000 Time 0.023762 +2023-10-05 20:55:15,748 - Epoch: [28][ 230/ 1236] Overall Loss 0.334352 Objective Loss 0.334352 LR 0.001000 Time 0.023616 +2023-10-05 20:55:15,952 - Epoch: [28][ 240/ 1236] Overall Loss 0.335278 Objective Loss 0.335278 LR 0.001000 Time 0.023481 +2023-10-05 20:55:16,155 - Epoch: [28][ 250/ 1236] Overall Loss 0.335095 Objective Loss 0.335095 LR 0.001000 Time 0.023354 +2023-10-05 20:55:16,359 - Epoch: [28][ 260/ 1236] Overall Loss 0.334871 Objective Loss 0.334871 LR 0.001000 Time 0.023238 +2023-10-05 20:55:16,563 - Epoch: [28][ 270/ 1236] Overall Loss 0.334935 Objective Loss 0.334935 LR 0.001000 Time 0.023127 +2023-10-05 20:55:16,767 - Epoch: [28][ 280/ 1236] Overall Loss 0.334372 Objective Loss 0.334372 LR 0.001000 Time 0.023028 +2023-10-05 20:55:16,971 - Epoch: [28][ 290/ 1236] Overall Loss 0.334583 Objective Loss 0.334583 LR 0.001000 Time 0.022936 +2023-10-05 20:55:17,175 - Epoch: [28][ 300/ 1236] Overall Loss 0.335172 Objective Loss 0.335172 LR 0.001000 Time 0.022851 +2023-10-05 20:55:17,379 - Epoch: [28][ 310/ 1236] Overall Loss 0.334239 Objective Loss 0.334239 LR 0.001000 Time 0.022766 +2023-10-05 20:55:17,583 - Epoch: [28][ 320/ 1236] Overall Loss 0.333862 Objective Loss 0.333862 LR 0.001000 Time 0.022691 +2023-10-05 20:55:17,786 - Epoch: [28][ 330/ 1236] Overall Loss 0.334470 Objective Loss 0.334470 LR 0.001000 Time 0.022620 +2023-10-05 20:55:17,991 - Epoch: [28][ 340/ 1236] Overall Loss 0.335806 Objective Loss 0.335806 LR 0.001000 Time 0.022556 +2023-10-05 20:55:18,195 - Epoch: [28][ 350/ 1236] Overall Loss 0.336395 Objective Loss 0.336395 LR 0.001000 Time 0.022489 +2023-10-05 20:55:18,400 - Epoch: [28][ 360/ 1236] Overall Loss 0.337398 Objective Loss 0.337398 LR 0.001000 Time 0.022432 +2023-10-05 20:55:18,602 - Epoch: [28][ 370/ 1236] Overall Loss 0.337960 Objective Loss 0.337960 LR 0.001000 Time 0.022369 +2023-10-05 20:55:18,804 - Epoch: [28][ 380/ 1236] Overall Loss 0.338403 Objective Loss 0.338403 LR 0.001000 Time 0.022311 +2023-10-05 20:55:19,006 - Epoch: [28][ 390/ 1236] Overall Loss 0.338072 Objective Loss 0.338072 LR 0.001000 Time 0.022255 +2023-10-05 20:55:19,208 - Epoch: [28][ 400/ 1236] Overall Loss 0.338677 Objective Loss 0.338677 LR 0.001000 Time 0.022203 +2023-10-05 20:55:19,409 - Epoch: [28][ 410/ 1236] Overall Loss 0.337541 Objective Loss 0.337541 LR 0.001000 Time 0.022152 +2023-10-05 20:55:19,611 - Epoch: [28][ 420/ 1236] Overall Loss 0.337623 Objective Loss 0.337623 LR 0.001000 Time 0.022104 +2023-10-05 20:55:19,812 - Epoch: [28][ 430/ 1236] Overall Loss 0.337258 Objective Loss 0.337258 LR 0.001000 Time 0.022058 +2023-10-05 20:55:20,015 - Epoch: [28][ 440/ 1236] Overall Loss 0.338170 Objective Loss 0.338170 LR 0.001000 Time 0.022015 +2023-10-05 20:55:20,216 - Epoch: [28][ 450/ 1236] Overall Loss 0.337939 Objective Loss 0.337939 LR 0.001000 Time 0.021972 +2023-10-05 20:55:20,418 - Epoch: [28][ 460/ 1236] Overall Loss 0.337337 Objective Loss 0.337337 LR 0.001000 Time 0.021933 +2023-10-05 20:55:20,619 - Epoch: [28][ 470/ 1236] Overall Loss 0.337779 Objective Loss 0.337779 LR 0.001000 Time 0.021894 +2023-10-05 20:55:20,821 - Epoch: [28][ 480/ 1236] Overall Loss 0.338016 Objective Loss 0.338016 LR 0.001000 Time 0.021858 +2023-10-05 20:55:21,022 - Epoch: [28][ 490/ 1236] Overall Loss 0.337696 Objective Loss 0.337696 LR 0.001000 Time 0.021822 +2023-10-05 20:55:21,224 - Epoch: [28][ 500/ 1236] Overall Loss 0.337801 Objective Loss 0.337801 LR 0.001000 Time 0.021789 +2023-10-05 20:55:21,426 - Epoch: [28][ 510/ 1236] Overall Loss 0.337843 Objective Loss 0.337843 LR 0.001000 Time 0.021756 +2023-10-05 20:55:21,628 - Epoch: [28][ 520/ 1236] Overall Loss 0.338424 Objective Loss 0.338424 LR 0.001000 Time 0.021725 +2023-10-05 20:55:21,829 - Epoch: [28][ 530/ 1236] Overall Loss 0.338169 Objective Loss 0.338169 LR 0.001000 Time 0.021695 +2023-10-05 20:55:22,031 - Epoch: [28][ 540/ 1236] Overall Loss 0.337840 Objective Loss 0.337840 LR 0.001000 Time 0.021666 +2023-10-05 20:55:22,232 - Epoch: [28][ 550/ 1236] Overall Loss 0.338915 Objective Loss 0.338915 LR 0.001000 Time 0.021638 +2023-10-05 20:55:22,434 - Epoch: [28][ 560/ 1236] Overall Loss 0.339140 Objective Loss 0.339140 LR 0.001000 Time 0.021612 +2023-10-05 20:55:22,636 - Epoch: [28][ 570/ 1236] Overall Loss 0.339676 Objective Loss 0.339676 LR 0.001000 Time 0.021585 +2023-10-05 20:55:22,838 - Epoch: [28][ 580/ 1236] Overall Loss 0.340450 Objective Loss 0.340450 LR 0.001000 Time 0.021561 +2023-10-05 20:55:23,039 - Epoch: [28][ 590/ 1236] Overall Loss 0.340628 Objective Loss 0.340628 LR 0.001000 Time 0.021536 +2023-10-05 20:55:23,241 - Epoch: [28][ 600/ 1236] Overall Loss 0.340840 Objective Loss 0.340840 LR 0.001000 Time 0.021513 +2023-10-05 20:55:23,442 - Epoch: [28][ 610/ 1236] Overall Loss 0.340578 Objective Loss 0.340578 LR 0.001000 Time 0.021490 +2023-10-05 20:55:23,644 - Epoch: [28][ 620/ 1236] Overall Loss 0.340532 Objective Loss 0.340532 LR 0.001000 Time 0.021469 +2023-10-05 20:55:23,845 - Epoch: [28][ 630/ 1236] Overall Loss 0.340446 Objective Loss 0.340446 LR 0.001000 Time 0.021447 +2023-10-05 20:55:24,048 - Epoch: [28][ 640/ 1236] Overall Loss 0.340116 Objective Loss 0.340116 LR 0.001000 Time 0.021427 +2023-10-05 20:55:24,249 - Epoch: [28][ 650/ 1236] Overall Loss 0.339953 Objective Loss 0.339953 LR 0.001000 Time 0.021407 +2023-10-05 20:55:24,451 - Epoch: [28][ 660/ 1236] Overall Loss 0.340362 Objective Loss 0.340362 LR 0.001000 Time 0.021388 +2023-10-05 20:55:24,652 - Epoch: [28][ 670/ 1236] Overall Loss 0.340565 Objective Loss 0.340565 LR 0.001000 Time 0.021369 +2023-10-05 20:55:24,855 - Epoch: [28][ 680/ 1236] Overall Loss 0.340944 Objective Loss 0.340944 LR 0.001000 Time 0.021352 +2023-10-05 20:55:25,056 - Epoch: [28][ 690/ 1236] Overall Loss 0.341384 Objective Loss 0.341384 LR 0.001000 Time 0.021334 +2023-10-05 20:55:25,258 - Epoch: [28][ 700/ 1236] Overall Loss 0.341999 Objective Loss 0.341999 LR 0.001000 Time 0.021317 +2023-10-05 20:55:25,459 - Epoch: [28][ 710/ 1236] Overall Loss 0.342652 Objective Loss 0.342652 LR 0.001000 Time 0.021300 +2023-10-05 20:55:25,661 - Epoch: [28][ 720/ 1236] Overall Loss 0.342679 Objective Loss 0.342679 LR 0.001000 Time 0.021284 +2023-10-05 20:55:25,863 - Epoch: [28][ 730/ 1236] Overall Loss 0.342550 Objective Loss 0.342550 LR 0.001000 Time 0.021268 +2023-10-05 20:55:26,065 - Epoch: [28][ 740/ 1236] Overall Loss 0.342813 Objective Loss 0.342813 LR 0.001000 Time 0.021253 +2023-10-05 20:55:26,266 - Epoch: [28][ 750/ 1236] Overall Loss 0.342684 Objective Loss 0.342684 LR 0.001000 Time 0.021238 +2023-10-05 20:55:26,468 - Epoch: [28][ 760/ 1236] Overall Loss 0.342786 Objective Loss 0.342786 LR 0.001000 Time 0.021224 +2023-10-05 20:55:26,670 - Epoch: [28][ 770/ 1236] Overall Loss 0.342380 Objective Loss 0.342380 LR 0.001000 Time 0.021209 +2023-10-05 20:55:26,872 - Epoch: [28][ 780/ 1236] Overall Loss 0.342728 Objective Loss 0.342728 LR 0.001000 Time 0.021196 +2023-10-05 20:55:27,073 - Epoch: [28][ 790/ 1236] Overall Loss 0.342957 Objective Loss 0.342957 LR 0.001000 Time 0.021183 +2023-10-05 20:55:27,275 - Epoch: [28][ 800/ 1236] Overall Loss 0.343006 Objective Loss 0.343006 LR 0.001000 Time 0.021170 +2023-10-05 20:55:27,476 - Epoch: [28][ 810/ 1236] Overall Loss 0.343165 Objective Loss 0.343165 LR 0.001000 Time 0.021157 +2023-10-05 20:55:27,679 - Epoch: [28][ 820/ 1236] Overall Loss 0.343082 Objective Loss 0.343082 LR 0.001000 Time 0.021145 +2023-10-05 20:55:27,880 - Epoch: [28][ 830/ 1236] Overall Loss 0.343379 Objective Loss 0.343379 LR 0.001000 Time 0.021132 +2023-10-05 20:55:28,082 - Epoch: [28][ 840/ 1236] Overall Loss 0.343460 Objective Loss 0.343460 LR 0.001000 Time 0.021120 +2023-10-05 20:55:28,282 - Epoch: [28][ 850/ 1236] Overall Loss 0.343703 Objective Loss 0.343703 LR 0.001000 Time 0.021108 +2023-10-05 20:55:28,485 - Epoch: [28][ 860/ 1236] Overall Loss 0.343694 Objective Loss 0.343694 LR 0.001000 Time 0.021097 +2023-10-05 20:55:28,686 - Epoch: [28][ 870/ 1236] Overall Loss 0.344078 Objective Loss 0.344078 LR 0.001000 Time 0.021086 +2023-10-05 20:55:28,888 - Epoch: [28][ 880/ 1236] Overall Loss 0.343936 Objective Loss 0.343936 LR 0.001000 Time 0.021075 +2023-10-05 20:55:29,089 - Epoch: [28][ 890/ 1236] Overall Loss 0.343886 Objective Loss 0.343886 LR 0.001000 Time 0.021064 +2023-10-05 20:55:29,291 - Epoch: [28][ 900/ 1236] Overall Loss 0.343515 Objective Loss 0.343515 LR 0.001000 Time 0.021054 +2023-10-05 20:55:29,492 - Epoch: [28][ 910/ 1236] Overall Loss 0.343601 Objective Loss 0.343601 LR 0.001000 Time 0.021043 +2023-10-05 20:55:29,694 - Epoch: [28][ 920/ 1236] Overall Loss 0.343663 Objective Loss 0.343663 LR 0.001000 Time 0.021034 +2023-10-05 20:55:29,895 - Epoch: [28][ 930/ 1236] Overall Loss 0.343861 Objective Loss 0.343861 LR 0.001000 Time 0.021024 +2023-10-05 20:55:30,098 - Epoch: [28][ 940/ 1236] Overall Loss 0.344004 Objective Loss 0.344004 LR 0.001000 Time 0.021015 +2023-10-05 20:55:30,300 - Epoch: [28][ 950/ 1236] Overall Loss 0.344238 Objective Loss 0.344238 LR 0.001000 Time 0.021007 +2023-10-05 20:55:30,503 - Epoch: [28][ 960/ 1236] Overall Loss 0.344162 Objective Loss 0.344162 LR 0.001000 Time 0.020998 +2023-10-05 20:55:30,704 - Epoch: [28][ 970/ 1236] Overall Loss 0.344403 Objective Loss 0.344403 LR 0.001000 Time 0.020989 +2023-10-05 20:55:30,906 - Epoch: [28][ 980/ 1236] Overall Loss 0.344485 Objective Loss 0.344485 LR 0.001000 Time 0.020981 +2023-10-05 20:55:31,107 - Epoch: [28][ 990/ 1236] Overall Loss 0.344883 Objective Loss 0.344883 LR 0.001000 Time 0.020972 +2023-10-05 20:55:31,310 - Epoch: [28][ 1000/ 1236] Overall Loss 0.344724 Objective Loss 0.344724 LR 0.001000 Time 0.020964 +2023-10-05 20:55:31,511 - Epoch: [28][ 1010/ 1236] Overall Loss 0.344924 Objective Loss 0.344924 LR 0.001000 Time 0.020956 +2023-10-05 20:55:31,713 - Epoch: [28][ 1020/ 1236] Overall Loss 0.345032 Objective Loss 0.345032 LR 0.001000 Time 0.020948 +2023-10-05 20:55:31,915 - Epoch: [28][ 1030/ 1236] Overall Loss 0.345151 Objective Loss 0.345151 LR 0.001000 Time 0.020940 +2023-10-05 20:55:32,117 - Epoch: [28][ 1040/ 1236] Overall Loss 0.345248 Objective Loss 0.345248 LR 0.001000 Time 0.020932 +2023-10-05 20:55:32,318 - Epoch: [28][ 1050/ 1236] Overall Loss 0.345285 Objective Loss 0.345285 LR 0.001000 Time 0.020925 +2023-10-05 20:55:32,520 - Epoch: [28][ 1060/ 1236] Overall Loss 0.345347 Objective Loss 0.345347 LR 0.001000 Time 0.020917 +2023-10-05 20:55:32,721 - Epoch: [28][ 1070/ 1236] Overall Loss 0.345347 Objective Loss 0.345347 LR 0.001000 Time 0.020910 +2023-10-05 20:55:32,923 - Epoch: [28][ 1080/ 1236] Overall Loss 0.345261 Objective Loss 0.345261 LR 0.001000 Time 0.020903 +2023-10-05 20:55:33,124 - Epoch: [28][ 1090/ 1236] Overall Loss 0.344916 Objective Loss 0.344916 LR 0.001000 Time 0.020895 +2023-10-05 20:55:33,326 - Epoch: [28][ 1100/ 1236] Overall Loss 0.344953 Objective Loss 0.344953 LR 0.001000 Time 0.020888 +2023-10-05 20:55:33,528 - Epoch: [28][ 1110/ 1236] Overall Loss 0.345174 Objective Loss 0.345174 LR 0.001000 Time 0.020882 +2023-10-05 20:55:33,730 - Epoch: [28][ 1120/ 1236] Overall Loss 0.345186 Objective Loss 0.345186 LR 0.001000 Time 0.020875 +2023-10-05 20:55:33,931 - Epoch: [28][ 1130/ 1236] Overall Loss 0.345099 Objective Loss 0.345099 LR 0.001000 Time 0.020869 +2023-10-05 20:55:34,133 - Epoch: [28][ 1140/ 1236] Overall Loss 0.344823 Objective Loss 0.344823 LR 0.001000 Time 0.020862 +2023-10-05 20:55:34,335 - Epoch: [28][ 1150/ 1236] Overall Loss 0.344596 Objective Loss 0.344596 LR 0.001000 Time 0.020856 +2023-10-05 20:55:34,537 - Epoch: [28][ 1160/ 1236] Overall Loss 0.344216 Objective Loss 0.344216 LR 0.001000 Time 0.020850 +2023-10-05 20:55:34,739 - Epoch: [28][ 1170/ 1236] Overall Loss 0.344205 Objective Loss 0.344205 LR 0.001000 Time 0.020844 +2023-10-05 20:55:34,940 - Epoch: [28][ 1180/ 1236] Overall Loss 0.344367 Objective Loss 0.344367 LR 0.001000 Time 0.020838 +2023-10-05 20:55:35,142 - Epoch: [28][ 1190/ 1236] Overall Loss 0.344130 Objective Loss 0.344130 LR 0.001000 Time 0.020832 +2023-10-05 20:55:35,344 - Epoch: [28][ 1200/ 1236] Overall Loss 0.344235 Objective Loss 0.344235 LR 0.001000 Time 0.020826 +2023-10-05 20:55:35,546 - Epoch: [28][ 1210/ 1236] Overall Loss 0.344340 Objective Loss 0.344340 LR 0.001000 Time 0.020821 +2023-10-05 20:55:35,748 - Epoch: [28][ 1220/ 1236] Overall Loss 0.344313 Objective Loss 0.344313 LR 0.001000 Time 0.020815 +2023-10-05 20:55:36,003 - Epoch: [28][ 1230/ 1236] Overall Loss 0.344460 Objective Loss 0.344460 LR 0.001000 Time 0.020853 +2023-10-05 20:55:36,120 - Epoch: [28][ 1236/ 1236] Overall Loss 0.344386 Objective Loss 0.344386 Top1 82.484725 Top5 98.167006 LR 0.001000 Time 0.020847 +2023-10-05 20:55:36,239 - --- validate (epoch=28)----------- +2023-10-05 20:55:36,239 - 29943 samples (256 per mini-batch) +2023-10-05 20:55:36,696 - Epoch: [28][ 10/ 117] Loss 0.358033 Top1 82.070312 Top5 97.343750 +2023-10-05 20:55:36,847 - Epoch: [28][ 20/ 117] Loss 0.353862 Top1 81.582031 Top5 97.363281 +2023-10-05 20:55:36,997 - Epoch: [28][ 30/ 117] Loss 0.366056 Top1 81.419271 Top5 97.304688 +2023-10-05 20:55:37,148 - Epoch: [28][ 40/ 117] Loss 0.375802 Top1 80.947266 Top5 97.265625 +2023-10-05 20:55:37,299 - Epoch: [28][ 50/ 117] Loss 0.366943 Top1 80.984375 Top5 97.296875 +2023-10-05 20:55:37,448 - Epoch: [28][ 60/ 117] Loss 0.376113 Top1 80.605469 Top5 97.265625 +2023-10-05 20:55:37,595 - Epoch: [28][ 70/ 117] Loss 0.378182 Top1 80.396205 Top5 97.243304 +2023-10-05 20:55:37,743 - Epoch: [28][ 80/ 117] Loss 0.380064 Top1 80.366211 Top5 97.285156 +2023-10-05 20:55:37,891 - Epoch: [28][ 90/ 117] Loss 0.377255 Top1 80.329861 Top5 97.300347 +2023-10-05 20:55:38,039 - Epoch: [28][ 100/ 117] Loss 0.376194 Top1 80.414062 Top5 97.316406 +2023-10-05 20:55:38,192 - Epoch: [28][ 110/ 117] Loss 0.374520 Top1 80.497159 Top5 97.311790 +2023-10-05 20:55:38,276 - Epoch: [28][ 117/ 117] Loss 0.377294 Top1 80.452860 Top5 97.314898 +2023-10-05 20:55:38,401 - ==> Top1: 80.453 Top5: 97.315 Loss: 0.377 + +2023-10-05 20:55:38,402 - ==> Confusion: +[[ 908 1 5 2 13 1 0 0 3 78 1 0 0 5 4 5 3 1 0 1 19] + [ 0 1030 3 1 11 12 1 28 3 0 1 4 1 1 0 7 6 1 9 3 9] + [ 4 0 947 26 1 1 17 8 0 1 3 2 8 3 2 11 0 2 2 2 16] + [ 4 2 20 947 1 3 2 0 3 0 5 0 15 2 29 6 0 13 19 3 15] + [ 24 6 2 0 962 3 1 0 1 6 0 4 0 4 6 9 16 2 0 0 4] + [ 6 52 1 4 3 945 0 14 1 4 5 18 8 20 6 1 4 0 3 9 12] + [ 0 8 39 0 0 0 1091 6 0 0 0 3 3 0 0 20 0 3 3 4 11] + [ 2 22 25 0 2 36 0 1030 1 0 4 16 5 0 1 2 0 1 51 14 6] + [ 17 3 1 0 4 2 0 0 939 55 20 5 4 14 18 4 1 0 1 1 0] + [ 90 0 1 0 9 3 3 0 33 915 1 4 0 30 5 7 1 2 1 3 11] + [ 1 3 13 9 1 0 3 5 16 2 953 5 0 11 5 2 2 3 7 3 9] + [ 1 0 3 0 0 11 0 1 0 0 0 938 30 3 0 6 3 20 1 17 1] + [ 0 0 2 2 0 0 0 0 0 0 3 40 955 0 0 14 3 27 2 8 12] + [ 2 0 4 1 2 11 1 0 12 13 14 9 2 1026 3 4 2 1 0 3 9] + [ 13 4 4 15 6 0 0 0 31 6 2 0 5 6 972 0 2 7 15 0 13] + [ 0 1 2 2 0 0 2 0 0 0 1 11 10 2 0 1067 9 14 1 8 4] + [ 2 8 3 2 2 4 1 0 1 0 2 6 5 2 5 17 1079 1 0 6 15] + [ 0 0 0 2 0 0 1 0 0 0 0 10 21 2 0 9 0 993 0 0 0] + [ 0 9 7 19 2 0 2 24 7 0 4 1 5 0 9 4 1 2 966 0 6] + [ 0 1 10 2 2 5 7 10 0 0 0 15 7 1 0 8 7 1 2 1066 8] + [ 133 220 213 129 137 132 37 101 110 85 215 158 546 330 154 111 126 118 194 295 4361]] + +2023-10-05 20:55:38,403 - ==> Best [Top1: 80.453 Top5: 97.315 Sparsity:0.00 Params: 148928 on epoch: 28] +2023-10-05 20:55:38,403 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:55:38,416 - + +2023-10-05 20:55:38,416 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:55:39,391 - Epoch: [29][ 10/ 1236] Overall Loss 0.317273 Objective Loss 0.317273 LR 0.001000 Time 0.097444 +2023-10-05 20:55:39,594 - Epoch: [29][ 20/ 1236] Overall Loss 0.325384 Objective Loss 0.325384 LR 0.001000 Time 0.058835 +2023-10-05 20:55:39,796 - Epoch: [29][ 30/ 1236] Overall Loss 0.314826 Objective Loss 0.314826 LR 0.001000 Time 0.045958 +2023-10-05 20:55:39,999 - Epoch: [29][ 40/ 1236] Overall Loss 0.315929 Objective Loss 0.315929 LR 0.001000 Time 0.039520 +2023-10-05 20:55:40,202 - Epoch: [29][ 50/ 1236] Overall Loss 0.323676 Objective Loss 0.323676 LR 0.001000 Time 0.035668 +2023-10-05 20:55:40,403 - Epoch: [29][ 60/ 1236] Overall Loss 0.317984 Objective Loss 0.317984 LR 0.001000 Time 0.033080 +2023-10-05 20:55:40,606 - Epoch: [29][ 70/ 1236] Overall Loss 0.320360 Objective Loss 0.320360 LR 0.001000 Time 0.031241 +2023-10-05 20:55:40,807 - Epoch: [29][ 80/ 1236] Overall Loss 0.322612 Objective Loss 0.322612 LR 0.001000 Time 0.029853 +2023-10-05 20:55:41,010 - Epoch: [29][ 90/ 1236] Overall Loss 0.324704 Objective Loss 0.324704 LR 0.001000 Time 0.028781 +2023-10-05 20:55:41,212 - Epoch: [29][ 100/ 1236] Overall Loss 0.322559 Objective Loss 0.322559 LR 0.001000 Time 0.027919 +2023-10-05 20:55:41,412 - Epoch: [29][ 110/ 1236] Overall Loss 0.323289 Objective Loss 0.323289 LR 0.001000 Time 0.027201 +2023-10-05 20:55:41,611 - Epoch: [29][ 120/ 1236] Overall Loss 0.324093 Objective Loss 0.324093 LR 0.001000 Time 0.026588 +2023-10-05 20:55:41,811 - Epoch: [29][ 130/ 1236] Overall Loss 0.324631 Objective Loss 0.324631 LR 0.001000 Time 0.026082 +2023-10-05 20:55:42,011 - Epoch: [29][ 140/ 1236] Overall Loss 0.325578 Objective Loss 0.325578 LR 0.001000 Time 0.025644 +2023-10-05 20:55:42,212 - Epoch: [29][ 150/ 1236] Overall Loss 0.326861 Objective Loss 0.326861 LR 0.001000 Time 0.025267 +2023-10-05 20:55:42,411 - Epoch: [29][ 160/ 1236] Overall Loss 0.326494 Objective Loss 0.326494 LR 0.001000 Time 0.024934 +2023-10-05 20:55:42,611 - Epoch: [29][ 170/ 1236] Overall Loss 0.326505 Objective Loss 0.326505 LR 0.001000 Time 0.024643 +2023-10-05 20:55:42,811 - Epoch: [29][ 180/ 1236] Overall Loss 0.328127 Objective Loss 0.328127 LR 0.001000 Time 0.024382 +2023-10-05 20:55:43,012 - Epoch: [29][ 190/ 1236] Overall Loss 0.329120 Objective Loss 0.329120 LR 0.001000 Time 0.024151 +2023-10-05 20:55:43,211 - Epoch: [29][ 200/ 1236] Overall Loss 0.329028 Objective Loss 0.329028 LR 0.001000 Time 0.023941 +2023-10-05 20:55:43,412 - Epoch: [29][ 210/ 1236] Overall Loss 0.329937 Objective Loss 0.329937 LR 0.001000 Time 0.023753 +2023-10-05 20:55:43,611 - Epoch: [29][ 220/ 1236] Overall Loss 0.329897 Objective Loss 0.329897 LR 0.001000 Time 0.023579 +2023-10-05 20:55:43,812 - Epoch: [29][ 230/ 1236] Overall Loss 0.331503 Objective Loss 0.331503 LR 0.001000 Time 0.023423 +2023-10-05 20:55:44,011 - Epoch: [29][ 240/ 1236] Overall Loss 0.331628 Objective Loss 0.331628 LR 0.001000 Time 0.023278 +2023-10-05 20:55:44,212 - Epoch: [29][ 250/ 1236] Overall Loss 0.332256 Objective Loss 0.332256 LR 0.001000 Time 0.023147 +2023-10-05 20:55:44,411 - Epoch: [29][ 260/ 1236] Overall Loss 0.332845 Objective Loss 0.332845 LR 0.001000 Time 0.023024 +2023-10-05 20:55:44,612 - Epoch: [29][ 270/ 1236] Overall Loss 0.333290 Objective Loss 0.333290 LR 0.001000 Time 0.022911 +2023-10-05 20:55:44,811 - Epoch: [29][ 280/ 1236] Overall Loss 0.333073 Objective Loss 0.333073 LR 0.001000 Time 0.022805 +2023-10-05 20:55:45,011 - Epoch: [29][ 290/ 1236] Overall Loss 0.332150 Objective Loss 0.332150 LR 0.001000 Time 0.022708 +2023-10-05 20:55:45,211 - Epoch: [29][ 300/ 1236] Overall Loss 0.332480 Objective Loss 0.332480 LR 0.001000 Time 0.022616 +2023-10-05 20:55:45,412 - Epoch: [29][ 310/ 1236] Overall Loss 0.333075 Objective Loss 0.333075 LR 0.001000 Time 0.022531 +2023-10-05 20:55:45,612 - Epoch: [29][ 320/ 1236] Overall Loss 0.333616 Objective Loss 0.333616 LR 0.001000 Time 0.022451 +2023-10-05 20:55:45,812 - Epoch: [29][ 330/ 1236] Overall Loss 0.334148 Objective Loss 0.334148 LR 0.001000 Time 0.022376 +2023-10-05 20:55:46,012 - Epoch: [29][ 340/ 1236] Overall Loss 0.334127 Objective Loss 0.334127 LR 0.001000 Time 0.022305 +2023-10-05 20:55:46,212 - Epoch: [29][ 350/ 1236] Overall Loss 0.333967 Objective Loss 0.333967 LR 0.001000 Time 0.022239 +2023-10-05 20:55:46,412 - Epoch: [29][ 360/ 1236] Overall Loss 0.334087 Objective Loss 0.334087 LR 0.001000 Time 0.022177 +2023-10-05 20:55:46,614 - Epoch: [29][ 370/ 1236] Overall Loss 0.334451 Objective Loss 0.334451 LR 0.001000 Time 0.022121 +2023-10-05 20:55:46,816 - Epoch: [29][ 380/ 1236] Overall Loss 0.334707 Objective Loss 0.334707 LR 0.001000 Time 0.022070 +2023-10-05 20:55:47,019 - Epoch: [29][ 390/ 1236] Overall Loss 0.334840 Objective Loss 0.334840 LR 0.001000 Time 0.022023 +2023-10-05 20:55:47,221 - Epoch: [29][ 400/ 1236] Overall Loss 0.335107 Objective Loss 0.335107 LR 0.001000 Time 0.021977 +2023-10-05 20:55:47,424 - Epoch: [29][ 410/ 1236] Overall Loss 0.335198 Objective Loss 0.335198 LR 0.001000 Time 0.021936 +2023-10-05 20:55:47,626 - Epoch: [29][ 420/ 1236] Overall Loss 0.336017 Objective Loss 0.336017 LR 0.001000 Time 0.021894 +2023-10-05 20:55:47,829 - Epoch: [29][ 430/ 1236] Overall Loss 0.335312 Objective Loss 0.335312 LR 0.001000 Time 0.021856 +2023-10-05 20:55:48,031 - Epoch: [29][ 440/ 1236] Overall Loss 0.335847 Objective Loss 0.335847 LR 0.001000 Time 0.021817 +2023-10-05 20:55:48,234 - Epoch: [29][ 450/ 1236] Overall Loss 0.336397 Objective Loss 0.336397 LR 0.001000 Time 0.021783 +2023-10-05 20:55:48,436 - Epoch: [29][ 460/ 1236] Overall Loss 0.336841 Objective Loss 0.336841 LR 0.001000 Time 0.021748 +2023-10-05 20:55:48,639 - Epoch: [29][ 470/ 1236] Overall Loss 0.336175 Objective Loss 0.336175 LR 0.001000 Time 0.021716 +2023-10-05 20:55:48,841 - Epoch: [29][ 480/ 1236] Overall Loss 0.335794 Objective Loss 0.335794 LR 0.001000 Time 0.021684 +2023-10-05 20:55:49,044 - Epoch: [29][ 490/ 1236] Overall Loss 0.335324 Objective Loss 0.335324 LR 0.001000 Time 0.021655 +2023-10-05 20:55:49,246 - Epoch: [29][ 500/ 1236] Overall Loss 0.335438 Objective Loss 0.335438 LR 0.001000 Time 0.021626 +2023-10-05 20:55:49,448 - Epoch: [29][ 510/ 1236] Overall Loss 0.335526 Objective Loss 0.335526 LR 0.001000 Time 0.021598 +2023-10-05 20:55:49,651 - Epoch: [29][ 520/ 1236] Overall Loss 0.335658 Objective Loss 0.335658 LR 0.001000 Time 0.021571 +2023-10-05 20:55:49,854 - Epoch: [29][ 530/ 1236] Overall Loss 0.335120 Objective Loss 0.335120 LR 0.001000 Time 0.021546 +2023-10-05 20:55:50,056 - Epoch: [29][ 540/ 1236] Overall Loss 0.335495 Objective Loss 0.335495 LR 0.001000 Time 0.021521 +2023-10-05 20:55:50,259 - Epoch: [29][ 550/ 1236] Overall Loss 0.335252 Objective Loss 0.335252 LR 0.001000 Time 0.021498 +2023-10-05 20:55:50,461 - Epoch: [29][ 560/ 1236] Overall Loss 0.335170 Objective Loss 0.335170 LR 0.001000 Time 0.021475 +2023-10-05 20:55:50,664 - Epoch: [29][ 570/ 1236] Overall Loss 0.335797 Objective Loss 0.335797 LR 0.001000 Time 0.021453 +2023-10-05 20:55:50,866 - Epoch: [29][ 580/ 1236] Overall Loss 0.335960 Objective Loss 0.335960 LR 0.001000 Time 0.021431 +2023-10-05 20:55:51,069 - Epoch: [29][ 590/ 1236] Overall Loss 0.335944 Objective Loss 0.335944 LR 0.001000 Time 0.021411 +2023-10-05 20:55:51,271 - Epoch: [29][ 600/ 1236] Overall Loss 0.335579 Objective Loss 0.335579 LR 0.001000 Time 0.021391 +2023-10-05 20:55:51,474 - Epoch: [29][ 610/ 1236] Overall Loss 0.335596 Objective Loss 0.335596 LR 0.001000 Time 0.021372 +2023-10-05 20:55:51,676 - Epoch: [29][ 620/ 1236] Overall Loss 0.335395 Objective Loss 0.335395 LR 0.001000 Time 0.021353 +2023-10-05 20:55:51,878 - Epoch: [29][ 630/ 1236] Overall Loss 0.335337 Objective Loss 0.335337 LR 0.001000 Time 0.021335 +2023-10-05 20:55:52,081 - Epoch: [29][ 640/ 1236] Overall Loss 0.335580 Objective Loss 0.335580 LR 0.001000 Time 0.021317 +2023-10-05 20:55:52,284 - Epoch: [29][ 650/ 1236] Overall Loss 0.335252 Objective Loss 0.335252 LR 0.001000 Time 0.021301 +2023-10-05 20:55:52,486 - Epoch: [29][ 660/ 1236] Overall Loss 0.335890 Objective Loss 0.335890 LR 0.001000 Time 0.021284 +2023-10-05 20:55:52,689 - Epoch: [29][ 670/ 1236] Overall Loss 0.336080 Objective Loss 0.336080 LR 0.001000 Time 0.021269 +2023-10-05 20:55:52,891 - Epoch: [29][ 680/ 1236] Overall Loss 0.336510 Objective Loss 0.336510 LR 0.001000 Time 0.021253 +2023-10-05 20:55:53,094 - Epoch: [29][ 690/ 1236] Overall Loss 0.336435 Objective Loss 0.336435 LR 0.001000 Time 0.021238 +2023-10-05 20:55:53,296 - Epoch: [29][ 700/ 1236] Overall Loss 0.336665 Objective Loss 0.336665 LR 0.001000 Time 0.021223 +2023-10-05 20:55:53,499 - Epoch: [29][ 710/ 1236] Overall Loss 0.336872 Objective Loss 0.336872 LR 0.001000 Time 0.021210 +2023-10-05 20:55:53,701 - Epoch: [29][ 720/ 1236] Overall Loss 0.336728 Objective Loss 0.336728 LR 0.001000 Time 0.021195 +2023-10-05 20:55:53,904 - Epoch: [29][ 730/ 1236] Overall Loss 0.336619 Objective Loss 0.336619 LR 0.001000 Time 0.021182 +2023-10-05 20:55:54,106 - Epoch: [29][ 740/ 1236] Overall Loss 0.336635 Objective Loss 0.336635 LR 0.001000 Time 0.021169 +2023-10-05 20:55:54,309 - Epoch: [29][ 750/ 1236] Overall Loss 0.336668 Objective Loss 0.336668 LR 0.001000 Time 0.021157 +2023-10-05 20:55:54,511 - Epoch: [29][ 760/ 1236] Overall Loss 0.337010 Objective Loss 0.337010 LR 0.001000 Time 0.021144 +2023-10-05 20:55:54,714 - Epoch: [29][ 770/ 1236] Overall Loss 0.337544 Objective Loss 0.337544 LR 0.001000 Time 0.021132 +2023-10-05 20:55:54,916 - Epoch: [29][ 780/ 1236] Overall Loss 0.337709 Objective Loss 0.337709 LR 0.001000 Time 0.021120 +2023-10-05 20:55:55,119 - Epoch: [29][ 790/ 1236] Overall Loss 0.337879 Objective Loss 0.337879 LR 0.001000 Time 0.021109 +2023-10-05 20:55:55,321 - Epoch: [29][ 800/ 1236] Overall Loss 0.337542 Objective Loss 0.337542 LR 0.001000 Time 0.021098 +2023-10-05 20:55:55,524 - Epoch: [29][ 810/ 1236] Overall Loss 0.337464 Objective Loss 0.337464 LR 0.001000 Time 0.021087 +2023-10-05 20:55:55,726 - Epoch: [29][ 820/ 1236] Overall Loss 0.337329 Objective Loss 0.337329 LR 0.001000 Time 0.021076 +2023-10-05 20:55:55,929 - Epoch: [29][ 830/ 1236] Overall Loss 0.337067 Objective Loss 0.337067 LR 0.001000 Time 0.021066 +2023-10-05 20:55:56,131 - Epoch: [29][ 840/ 1236] Overall Loss 0.337165 Objective Loss 0.337165 LR 0.001000 Time 0.021056 +2023-10-05 20:55:56,334 - Epoch: [29][ 850/ 1236] Overall Loss 0.337222 Objective Loss 0.337222 LR 0.001000 Time 0.021046 +2023-10-05 20:55:56,536 - Epoch: [29][ 860/ 1236] Overall Loss 0.337375 Objective Loss 0.337375 LR 0.001000 Time 0.021036 +2023-10-05 20:55:56,739 - Epoch: [29][ 870/ 1236] Overall Loss 0.337277 Objective Loss 0.337277 LR 0.001000 Time 0.021027 +2023-10-05 20:55:56,941 - Epoch: [29][ 880/ 1236] Overall Loss 0.337157 Objective Loss 0.337157 LR 0.001000 Time 0.021018 +2023-10-05 20:55:57,144 - Epoch: [29][ 890/ 1236] Overall Loss 0.337299 Objective Loss 0.337299 LR 0.001000 Time 0.021009 +2023-10-05 20:55:57,346 - Epoch: [29][ 900/ 1236] Overall Loss 0.337021 Objective Loss 0.337021 LR 0.001000 Time 0.021000 +2023-10-05 20:55:57,549 - Epoch: [29][ 910/ 1236] Overall Loss 0.337042 Objective Loss 0.337042 LR 0.001000 Time 0.020992 +2023-10-05 20:55:57,751 - Epoch: [29][ 920/ 1236] Overall Loss 0.337518 Objective Loss 0.337518 LR 0.001000 Time 0.020983 +2023-10-05 20:55:57,954 - Epoch: [29][ 930/ 1236] Overall Loss 0.337530 Objective Loss 0.337530 LR 0.001000 Time 0.020975 +2023-10-05 20:55:58,156 - Epoch: [29][ 940/ 1236] Overall Loss 0.337559 Objective Loss 0.337559 LR 0.001000 Time 0.020967 +2023-10-05 20:55:58,359 - Epoch: [29][ 950/ 1236] Overall Loss 0.337399 Objective Loss 0.337399 LR 0.001000 Time 0.020959 +2023-10-05 20:55:58,561 - Epoch: [29][ 960/ 1236] Overall Loss 0.337287 Objective Loss 0.337287 LR 0.001000 Time 0.020951 +2023-10-05 20:55:58,764 - Epoch: [29][ 970/ 1236] Overall Loss 0.337215 Objective Loss 0.337215 LR 0.001000 Time 0.020944 +2023-10-05 20:55:58,966 - Epoch: [29][ 980/ 1236] Overall Loss 0.337387 Objective Loss 0.337387 LR 0.001000 Time 0.020936 +2023-10-05 20:55:59,169 - Epoch: [29][ 990/ 1236] Overall Loss 0.337467 Objective Loss 0.337467 LR 0.001000 Time 0.020929 +2023-10-05 20:55:59,371 - Epoch: [29][ 1000/ 1236] Overall Loss 0.337242 Objective Loss 0.337242 LR 0.001000 Time 0.020922 +2023-10-05 20:55:59,574 - Epoch: [29][ 1010/ 1236] Overall Loss 0.337106 Objective Loss 0.337106 LR 0.001000 Time 0.020915 +2023-10-05 20:55:59,776 - Epoch: [29][ 1020/ 1236] Overall Loss 0.337230 Objective Loss 0.337230 LR 0.001000 Time 0.020908 +2023-10-05 20:55:59,979 - Epoch: [29][ 1030/ 1236] Overall Loss 0.337075 Objective Loss 0.337075 LR 0.001000 Time 0.020901 +2023-10-05 20:56:00,181 - Epoch: [29][ 1040/ 1236] Overall Loss 0.337132 Objective Loss 0.337132 LR 0.001000 Time 0.020895 +2023-10-05 20:56:00,384 - Epoch: [29][ 1050/ 1236] Overall Loss 0.337008 Objective Loss 0.337008 LR 0.001000 Time 0.020888 +2023-10-05 20:56:00,586 - Epoch: [29][ 1060/ 1236] Overall Loss 0.336801 Objective Loss 0.336801 LR 0.001000 Time 0.020882 +2023-10-05 20:56:00,789 - Epoch: [29][ 1070/ 1236] Overall Loss 0.336743 Objective Loss 0.336743 LR 0.001000 Time 0.020876 +2023-10-05 20:56:00,991 - Epoch: [29][ 1080/ 1236] Overall Loss 0.336562 Objective Loss 0.336562 LR 0.001000 Time 0.020869 +2023-10-05 20:56:01,194 - Epoch: [29][ 1090/ 1236] Overall Loss 0.336182 Objective Loss 0.336182 LR 0.001000 Time 0.020864 +2023-10-05 20:56:01,396 - Epoch: [29][ 1100/ 1236] Overall Loss 0.335952 Objective Loss 0.335952 LR 0.001000 Time 0.020858 +2023-10-05 20:56:01,599 - Epoch: [29][ 1110/ 1236] Overall Loss 0.336006 Objective Loss 0.336006 LR 0.001000 Time 0.020852 +2023-10-05 20:56:01,801 - Epoch: [29][ 1120/ 1236] Overall Loss 0.336113 Objective Loss 0.336113 LR 0.001000 Time 0.020846 +2023-10-05 20:56:02,004 - Epoch: [29][ 1130/ 1236] Overall Loss 0.336263 Objective Loss 0.336263 LR 0.001000 Time 0.020841 +2023-10-05 20:56:02,206 - Epoch: [29][ 1140/ 1236] Overall Loss 0.336743 Objective Loss 0.336743 LR 0.001000 Time 0.020835 +2023-10-05 20:56:02,409 - Epoch: [29][ 1150/ 1236] Overall Loss 0.337017 Objective Loss 0.337017 LR 0.001000 Time 0.020830 +2023-10-05 20:56:02,611 - Epoch: [29][ 1160/ 1236] Overall Loss 0.336886 Objective Loss 0.336886 LR 0.001000 Time 0.020825 +2023-10-05 20:56:02,814 - Epoch: [29][ 1170/ 1236] Overall Loss 0.337438 Objective Loss 0.337438 LR 0.001000 Time 0.020820 +2023-10-05 20:56:03,016 - Epoch: [29][ 1180/ 1236] Overall Loss 0.337277 Objective Loss 0.337277 LR 0.001000 Time 0.020814 +2023-10-05 20:56:03,220 - Epoch: [29][ 1190/ 1236] Overall Loss 0.337553 Objective Loss 0.337553 LR 0.001000 Time 0.020810 +2023-10-05 20:56:03,422 - Epoch: [29][ 1200/ 1236] Overall Loss 0.337331 Objective Loss 0.337331 LR 0.001000 Time 0.020805 +2023-10-05 20:56:03,625 - Epoch: [29][ 1210/ 1236] Overall Loss 0.337297 Objective Loss 0.337297 LR 0.001000 Time 0.020800 +2023-10-05 20:56:03,827 - Epoch: [29][ 1220/ 1236] Overall Loss 0.337717 Objective Loss 0.337717 LR 0.001000 Time 0.020796 +2023-10-05 20:56:04,083 - Epoch: [29][ 1230/ 1236] Overall Loss 0.337760 Objective Loss 0.337760 LR 0.001000 Time 0.020834 +2023-10-05 20:56:04,200 - Epoch: [29][ 1236/ 1236] Overall Loss 0.337746 Objective Loss 0.337746 Top1 82.077393 Top5 97.556008 LR 0.001000 Time 0.020828 +2023-10-05 20:56:04,327 - --- validate (epoch=29)----------- +2023-10-05 20:56:04,327 - 29943 samples (256 per mini-batch) +2023-10-05 20:56:04,790 - Epoch: [29][ 10/ 117] Loss 0.376245 Top1 80.976562 Top5 97.382812 +2023-10-05 20:56:04,947 - Epoch: [29][ 20/ 117] Loss 0.364513 Top1 81.425781 Top5 97.675781 +2023-10-05 20:56:05,100 - Epoch: [29][ 30/ 117] Loss 0.367337 Top1 81.158854 Top5 97.460938 +2023-10-05 20:56:05,260 - Epoch: [29][ 40/ 117] Loss 0.371150 Top1 80.996094 Top5 97.382812 +2023-10-05 20:56:05,413 - Epoch: [29][ 50/ 117] Loss 0.371381 Top1 80.968750 Top5 97.289062 +2023-10-05 20:56:05,571 - Epoch: [29][ 60/ 117] Loss 0.372145 Top1 80.950521 Top5 97.311198 +2023-10-05 20:56:05,725 - Epoch: [29][ 70/ 117] Loss 0.369909 Top1 80.965402 Top5 97.304688 +2023-10-05 20:56:05,883 - Epoch: [29][ 80/ 117] Loss 0.369951 Top1 81.049805 Top5 97.319336 +2023-10-05 20:56:06,035 - Epoch: [29][ 90/ 117] Loss 0.371457 Top1 81.098090 Top5 97.313368 +2023-10-05 20:56:06,193 - Epoch: [29][ 100/ 117] Loss 0.369783 Top1 81.097656 Top5 97.386719 +2023-10-05 20:56:06,352 - Epoch: [29][ 110/ 117] Loss 0.374895 Top1 81.019176 Top5 97.397017 +2023-10-05 20:56:06,436 - Epoch: [29][ 117/ 117] Loss 0.373458 Top1 80.987209 Top5 97.425108 +2023-10-05 20:56:06,530 - ==> Top1: 80.987 Top5: 97.425 Loss: 0.373 + +2023-10-05 20:56:06,530 - ==> Confusion: +[[ 907 3 7 1 23 3 0 1 2 63 1 0 0 6 10 2 7 0 2 1 11] + [ 2 1066 1 1 7 13 2 25 0 0 0 0 0 0 1 3 3 1 2 3 1] + [ 4 1 944 22 5 1 31 5 0 1 2 4 8 2 1 3 1 2 7 1 11] + [ 2 2 17 971 1 10 5 0 0 0 1 0 2 2 30 2 1 6 21 0 16] + [ 13 14 2 0 979 5 1 0 0 5 1 1 0 2 8 2 9 1 1 1 5] + [ 2 59 1 0 5 962 2 28 3 1 2 7 2 16 11 1 2 0 1 1 10] + [ 0 13 29 0 0 0 1114 11 0 0 3 2 1 0 0 7 0 2 0 5 4] + [ 5 33 23 0 1 26 7 1056 1 2 4 7 1 0 0 3 1 0 35 4 9] + [ 20 11 4 0 3 0 0 1 926 56 11 0 1 17 23 3 3 0 9 0 1] + [ 120 0 4 1 16 6 4 2 24 891 1 0 0 23 9 4 1 0 1 2 10] + [ 1 11 20 7 2 2 2 7 10 1 944 5 0 16 1 1 4 1 7 1 10] + [ 1 0 4 0 2 19 1 1 0 0 0 919 41 3 0 5 4 15 0 13 7] + [ 1 1 12 8 1 2 1 2 0 0 2 32 957 2 6 9 3 13 0 2 14] + [ 2 1 1 2 6 16 1 1 8 19 3 3 5 1027 4 2 6 1 0 3 8] + [ 16 6 1 13 12 0 0 0 7 6 1 0 4 0 1005 0 1 3 12 0 14] + [ 0 2 2 4 4 1 2 2 0 0 0 6 6 0 0 1066 15 9 0 6 9] + [ 0 19 1 1 13 4 0 0 1 0 0 2 1 1 4 11 1088 0 0 4 11] + [ 1 0 0 4 4 1 1 0 0 0 0 5 32 1 3 11 1 966 2 0 6] + [ 0 17 9 13 1 0 0 41 3 1 0 1 1 0 13 0 2 0 959 0 7] + [ 0 2 9 0 2 9 16 31 0 0 0 20 6 2 1 8 11 0 2 1020 13] + [ 113 397 226 133 157 182 81 132 62 71 123 104 416 309 167 90 225 67 188 180 4482]] + +2023-10-05 20:56:06,532 - ==> Best [Top1: 80.987 Top5: 97.425 Sparsity:0.00 Params: 148928 on epoch: 29] +2023-10-05 20:56:06,532 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:56:06,539 - + +2023-10-05 20:56:06,539 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:56:07,638 - Epoch: [30][ 10/ 1236] Overall Loss 0.331663 Objective Loss 0.331663 LR 0.001000 Time 0.109835 +2023-10-05 20:56:07,841 - Epoch: [30][ 20/ 1236] Overall Loss 0.333949 Objective Loss 0.333949 LR 0.001000 Time 0.065057 +2023-10-05 20:56:08,043 - Epoch: [30][ 30/ 1236] Overall Loss 0.328515 Objective Loss 0.328515 LR 0.001000 Time 0.050097 +2023-10-05 20:56:08,245 - Epoch: [30][ 40/ 1236] Overall Loss 0.324086 Objective Loss 0.324086 LR 0.001000 Time 0.042625 +2023-10-05 20:56:08,448 - Epoch: [30][ 50/ 1236] Overall Loss 0.332243 Objective Loss 0.332243 LR 0.001000 Time 0.038144 +2023-10-05 20:56:08,650 - Epoch: [30][ 60/ 1236] Overall Loss 0.331077 Objective Loss 0.331077 LR 0.001000 Time 0.035158 +2023-10-05 20:56:08,853 - Epoch: [30][ 70/ 1236] Overall Loss 0.329124 Objective Loss 0.329124 LR 0.001000 Time 0.033022 +2023-10-05 20:56:09,054 - Epoch: [30][ 80/ 1236] Overall Loss 0.331557 Objective Loss 0.331557 LR 0.001000 Time 0.031411 +2023-10-05 20:56:09,254 - Epoch: [30][ 90/ 1236] Overall Loss 0.332199 Objective Loss 0.332199 LR 0.001000 Time 0.030138 +2023-10-05 20:56:09,455 - Epoch: [30][ 100/ 1236] Overall Loss 0.331681 Objective Loss 0.331681 LR 0.001000 Time 0.029129 +2023-10-05 20:56:09,657 - Epoch: [30][ 110/ 1236] Overall Loss 0.331262 Objective Loss 0.331262 LR 0.001000 Time 0.028313 +2023-10-05 20:56:09,862 - Epoch: [30][ 120/ 1236] Overall Loss 0.330901 Objective Loss 0.330901 LR 0.001000 Time 0.027659 +2023-10-05 20:56:10,064 - Epoch: [30][ 130/ 1236] Overall Loss 0.330204 Objective Loss 0.330204 LR 0.001000 Time 0.027075 +2023-10-05 20:56:10,268 - Epoch: [30][ 140/ 1236] Overall Loss 0.329808 Objective Loss 0.329808 LR 0.001000 Time 0.026595 +2023-10-05 20:56:10,472 - Epoch: [30][ 150/ 1236] Overall Loss 0.331760 Objective Loss 0.331760 LR 0.001000 Time 0.026170 +2023-10-05 20:56:10,674 - Epoch: [30][ 160/ 1236] Overall Loss 0.331760 Objective Loss 0.331760 LR 0.001000 Time 0.025793 +2023-10-05 20:56:10,874 - Epoch: [30][ 170/ 1236] Overall Loss 0.332904 Objective Loss 0.332904 LR 0.001000 Time 0.025452 +2023-10-05 20:56:11,074 - Epoch: [30][ 180/ 1236] Overall Loss 0.333699 Objective Loss 0.333699 LR 0.001000 Time 0.025150 +2023-10-05 20:56:11,275 - Epoch: [30][ 190/ 1236] Overall Loss 0.333331 Objective Loss 0.333331 LR 0.001000 Time 0.024880 +2023-10-05 20:56:11,476 - Epoch: [30][ 200/ 1236] Overall Loss 0.331023 Objective Loss 0.331023 LR 0.001000 Time 0.024638 +2023-10-05 20:56:11,676 - Epoch: [30][ 210/ 1236] Overall Loss 0.331138 Objective Loss 0.331138 LR 0.001000 Time 0.024419 +2023-10-05 20:56:11,877 - Epoch: [30][ 220/ 1236] Overall Loss 0.331824 Objective Loss 0.331824 LR 0.001000 Time 0.024220 +2023-10-05 20:56:12,078 - Epoch: [30][ 230/ 1236] Overall Loss 0.330633 Objective Loss 0.330633 LR 0.001000 Time 0.024037 +2023-10-05 20:56:12,278 - Epoch: [30][ 240/ 1236] Overall Loss 0.330420 Objective Loss 0.330420 LR 0.001000 Time 0.023869 +2023-10-05 20:56:12,479 - Epoch: [30][ 250/ 1236] Overall Loss 0.330301 Objective Loss 0.330301 LR 0.001000 Time 0.023715 +2023-10-05 20:56:12,679 - Epoch: [30][ 260/ 1236] Overall Loss 0.329783 Objective Loss 0.329783 LR 0.001000 Time 0.023573 +2023-10-05 20:56:12,880 - Epoch: [30][ 270/ 1236] Overall Loss 0.329778 Objective Loss 0.329778 LR 0.001000 Time 0.023441 +2023-10-05 20:56:13,080 - Epoch: [30][ 280/ 1236] Overall Loss 0.330177 Objective Loss 0.330177 LR 0.001000 Time 0.023319 +2023-10-05 20:56:13,281 - Epoch: [30][ 290/ 1236] Overall Loss 0.329502 Objective Loss 0.329502 LR 0.001000 Time 0.023205 +2023-10-05 20:56:13,481 - Epoch: [30][ 300/ 1236] Overall Loss 0.330035 Objective Loss 0.330035 LR 0.001000 Time 0.023098 +2023-10-05 20:56:13,682 - Epoch: [30][ 310/ 1236] Overall Loss 0.328342 Objective Loss 0.328342 LR 0.001000 Time 0.023000 +2023-10-05 20:56:13,884 - Epoch: [30][ 320/ 1236] Overall Loss 0.328031 Objective Loss 0.328031 LR 0.001000 Time 0.022911 +2023-10-05 20:56:14,086 - Epoch: [30][ 330/ 1236] Overall Loss 0.327757 Objective Loss 0.327757 LR 0.001000 Time 0.022828 +2023-10-05 20:56:14,287 - Epoch: [30][ 340/ 1236] Overall Loss 0.327537 Objective Loss 0.327537 LR 0.001000 Time 0.022748 +2023-10-05 20:56:14,489 - Epoch: [30][ 350/ 1236] Overall Loss 0.326974 Objective Loss 0.326974 LR 0.001000 Time 0.022674 +2023-10-05 20:56:14,691 - Epoch: [30][ 360/ 1236] Overall Loss 0.327309 Objective Loss 0.327309 LR 0.001000 Time 0.022604 +2023-10-05 20:56:14,893 - Epoch: [30][ 370/ 1236] Overall Loss 0.326416 Objective Loss 0.326416 LR 0.001000 Time 0.022539 +2023-10-05 20:56:15,096 - Epoch: [30][ 380/ 1236] Overall Loss 0.326162 Objective Loss 0.326162 LR 0.001000 Time 0.022477 +2023-10-05 20:56:15,298 - Epoch: [30][ 390/ 1236] Overall Loss 0.325830 Objective Loss 0.325830 LR 0.001000 Time 0.022419 +2023-10-05 20:56:15,500 - Epoch: [30][ 400/ 1236] Overall Loss 0.325665 Objective Loss 0.325665 LR 0.001000 Time 0.022364 +2023-10-05 20:56:15,703 - Epoch: [30][ 410/ 1236] Overall Loss 0.326031 Objective Loss 0.326031 LR 0.001000 Time 0.022312 +2023-10-05 20:56:15,905 - Epoch: [30][ 420/ 1236] Overall Loss 0.326247 Objective Loss 0.326247 LR 0.001000 Time 0.022262 +2023-10-05 20:56:16,108 - Epoch: [30][ 430/ 1236] Overall Loss 0.326873 Objective Loss 0.326873 LR 0.001000 Time 0.022214 +2023-10-05 20:56:16,311 - Epoch: [30][ 440/ 1236] Overall Loss 0.327336 Objective Loss 0.327336 LR 0.001000 Time 0.022169 +2023-10-05 20:56:16,513 - Epoch: [30][ 450/ 1236] Overall Loss 0.327599 Objective Loss 0.327599 LR 0.001000 Time 0.022126 +2023-10-05 20:56:16,715 - Epoch: [30][ 460/ 1236] Overall Loss 0.326966 Objective Loss 0.326966 LR 0.001000 Time 0.022084 +2023-10-05 20:56:16,918 - Epoch: [30][ 470/ 1236] Overall Loss 0.327286 Objective Loss 0.327286 LR 0.001000 Time 0.022044 +2023-10-05 20:56:17,121 - Epoch: [30][ 480/ 1236] Overall Loss 0.327726 Objective Loss 0.327726 LR 0.001000 Time 0.022006 +2023-10-05 20:56:17,323 - Epoch: [30][ 490/ 1236] Overall Loss 0.327613 Objective Loss 0.327613 LR 0.001000 Time 0.021970 +2023-10-05 20:56:17,526 - Epoch: [30][ 500/ 1236] Overall Loss 0.327242 Objective Loss 0.327242 LR 0.001000 Time 0.021935 +2023-10-05 20:56:17,728 - Epoch: [30][ 510/ 1236] Overall Loss 0.326597 Objective Loss 0.326597 LR 0.001000 Time 0.021901 +2023-10-05 20:56:17,930 - Epoch: [30][ 520/ 1236] Overall Loss 0.326627 Objective Loss 0.326627 LR 0.001000 Time 0.021869 +2023-10-05 20:56:18,133 - Epoch: [30][ 530/ 1236] Overall Loss 0.326121 Objective Loss 0.326121 LR 0.001000 Time 0.021838 +2023-10-05 20:56:18,336 - Epoch: [30][ 540/ 1236] Overall Loss 0.326200 Objective Loss 0.326200 LR 0.001000 Time 0.021808 +2023-10-05 20:56:18,538 - Epoch: [30][ 550/ 1236] Overall Loss 0.326406 Objective Loss 0.326406 LR 0.001000 Time 0.021779 +2023-10-05 20:56:18,740 - Epoch: [30][ 560/ 1236] Overall Loss 0.326727 Objective Loss 0.326727 LR 0.001000 Time 0.021751 +2023-10-05 20:56:18,943 - Epoch: [30][ 570/ 1236] Overall Loss 0.327481 Objective Loss 0.327481 LR 0.001000 Time 0.021724 +2023-10-05 20:56:19,146 - Epoch: [30][ 580/ 1236] Overall Loss 0.327545 Objective Loss 0.327545 LR 0.001000 Time 0.021699 +2023-10-05 20:56:19,351 - Epoch: [30][ 590/ 1236] Overall Loss 0.328217 Objective Loss 0.328217 LR 0.001000 Time 0.021678 +2023-10-05 20:56:19,556 - Epoch: [30][ 600/ 1236] Overall Loss 0.328273 Objective Loss 0.328273 LR 0.001000 Time 0.021658 +2023-10-05 20:56:19,761 - Epoch: [30][ 610/ 1236] Overall Loss 0.329215 Objective Loss 0.329215 LR 0.001000 Time 0.021638 +2023-10-05 20:56:19,966 - Epoch: [30][ 620/ 1236] Overall Loss 0.329071 Objective Loss 0.329071 LR 0.001000 Time 0.021619 +2023-10-05 20:56:20,171 - Epoch: [30][ 630/ 1236] Overall Loss 0.329744 Objective Loss 0.329744 LR 0.001000 Time 0.021600 +2023-10-05 20:56:20,376 - Epoch: [30][ 640/ 1236] Overall Loss 0.330163 Objective Loss 0.330163 LR 0.001000 Time 0.021583 +2023-10-05 20:56:20,581 - Epoch: [30][ 650/ 1236] Overall Loss 0.330290 Objective Loss 0.330290 LR 0.001000 Time 0.021566 +2023-10-05 20:56:20,786 - Epoch: [30][ 660/ 1236] Overall Loss 0.330168 Objective Loss 0.330168 LR 0.001000 Time 0.021549 +2023-10-05 20:56:20,990 - Epoch: [30][ 670/ 1236] Overall Loss 0.330686 Objective Loss 0.330686 LR 0.001000 Time 0.021532 +2023-10-05 20:56:21,196 - Epoch: [30][ 680/ 1236] Overall Loss 0.330754 Objective Loss 0.330754 LR 0.001000 Time 0.021517 +2023-10-05 20:56:21,403 - Epoch: [30][ 690/ 1236] Overall Loss 0.330874 Objective Loss 0.330874 LR 0.001000 Time 0.021505 +2023-10-05 20:56:21,610 - Epoch: [30][ 700/ 1236] Overall Loss 0.331302 Objective Loss 0.331302 LR 0.001000 Time 0.021493 +2023-10-05 20:56:21,814 - Epoch: [30][ 710/ 1236] Overall Loss 0.331654 Objective Loss 0.331654 LR 0.001000 Time 0.021477 +2023-10-05 20:56:22,019 - Epoch: [30][ 720/ 1236] Overall Loss 0.331147 Objective Loss 0.331147 LR 0.001000 Time 0.021462 +2023-10-05 20:56:22,218 - Epoch: [30][ 730/ 1236] Overall Loss 0.331740 Objective Loss 0.331740 LR 0.001000 Time 0.021440 +2023-10-05 20:56:22,415 - Epoch: [30][ 740/ 1236] Overall Loss 0.331838 Objective Loss 0.331838 LR 0.001000 Time 0.021416 +2023-10-05 20:56:22,612 - Epoch: [30][ 750/ 1236] Overall Loss 0.331621 Objective Loss 0.331621 LR 0.001000 Time 0.021393 +2023-10-05 20:56:22,810 - Epoch: [30][ 760/ 1236] Overall Loss 0.331700 Objective Loss 0.331700 LR 0.001000 Time 0.021372 +2023-10-05 20:56:23,007 - Epoch: [30][ 770/ 1236] Overall Loss 0.331748 Objective Loss 0.331748 LR 0.001000 Time 0.021350 +2023-10-05 20:56:23,205 - Epoch: [30][ 780/ 1236] Overall Loss 0.331379 Objective Loss 0.331379 LR 0.001000 Time 0.021329 +2023-10-05 20:56:23,403 - Epoch: [30][ 790/ 1236] Overall Loss 0.331909 Objective Loss 0.331909 LR 0.001000 Time 0.021309 +2023-10-05 20:56:23,600 - Epoch: [30][ 800/ 1236] Overall Loss 0.331836 Objective Loss 0.331836 LR 0.001000 Time 0.021289 +2023-10-05 20:56:23,797 - Epoch: [30][ 810/ 1236] Overall Loss 0.332177 Objective Loss 0.332177 LR 0.001000 Time 0.021270 +2023-10-05 20:56:23,995 - Epoch: [30][ 820/ 1236] Overall Loss 0.331917 Objective Loss 0.331917 LR 0.001000 Time 0.021251 +2023-10-05 20:56:24,192 - Epoch: [30][ 830/ 1236] Overall Loss 0.331796 Objective Loss 0.331796 LR 0.001000 Time 0.021232 +2023-10-05 20:56:24,389 - Epoch: [30][ 840/ 1236] Overall Loss 0.331961 Objective Loss 0.331961 LR 0.001000 Time 0.021213 +2023-10-05 20:56:24,586 - Epoch: [30][ 850/ 1236] Overall Loss 0.332003 Objective Loss 0.332003 LR 0.001000 Time 0.021196 +2023-10-05 20:56:24,784 - Epoch: [30][ 860/ 1236] Overall Loss 0.332007 Objective Loss 0.332007 LR 0.001000 Time 0.021179 +2023-10-05 20:56:24,981 - Epoch: [30][ 870/ 1236] Overall Loss 0.332282 Objective Loss 0.332282 LR 0.001000 Time 0.021162 +2023-10-05 20:56:25,179 - Epoch: [30][ 880/ 1236] Overall Loss 0.332345 Objective Loss 0.332345 LR 0.001000 Time 0.021145 +2023-10-05 20:56:25,376 - Epoch: [30][ 890/ 1236] Overall Loss 0.332495 Objective Loss 0.332495 LR 0.001000 Time 0.021129 +2023-10-05 20:56:25,574 - Epoch: [30][ 900/ 1236] Overall Loss 0.332799 Objective Loss 0.332799 LR 0.001000 Time 0.021114 +2023-10-05 20:56:25,772 - Epoch: [30][ 910/ 1236] Overall Loss 0.332785 Objective Loss 0.332785 LR 0.001000 Time 0.021099 +2023-10-05 20:56:25,969 - Epoch: [30][ 920/ 1236] Overall Loss 0.333273 Objective Loss 0.333273 LR 0.001000 Time 0.021083 +2023-10-05 20:56:26,166 - Epoch: [30][ 930/ 1236] Overall Loss 0.333896 Objective Loss 0.333896 LR 0.001000 Time 0.021068 +2023-10-05 20:56:26,363 - Epoch: [30][ 940/ 1236] Overall Loss 0.334166 Objective Loss 0.334166 LR 0.001000 Time 0.021054 +2023-10-05 20:56:26,561 - Epoch: [30][ 950/ 1236] Overall Loss 0.334270 Objective Loss 0.334270 LR 0.001000 Time 0.021040 +2023-10-05 20:56:26,758 - Epoch: [30][ 960/ 1236] Overall Loss 0.334605 Objective Loss 0.334605 LR 0.001000 Time 0.021026 +2023-10-05 20:56:26,955 - Epoch: [30][ 970/ 1236] Overall Loss 0.334697 Objective Loss 0.334697 LR 0.001000 Time 0.021012 +2023-10-05 20:56:27,153 - Epoch: [30][ 980/ 1236] Overall Loss 0.334529 Objective Loss 0.334529 LR 0.001000 Time 0.020999 +2023-10-05 20:56:27,350 - Epoch: [30][ 990/ 1236] Overall Loss 0.334768 Objective Loss 0.334768 LR 0.001000 Time 0.020986 +2023-10-05 20:56:27,548 - Epoch: [30][ 1000/ 1236] Overall Loss 0.334526 Objective Loss 0.334526 LR 0.001000 Time 0.020973 +2023-10-05 20:56:27,745 - Epoch: [30][ 1010/ 1236] Overall Loss 0.334517 Objective Loss 0.334517 LR 0.001000 Time 0.020961 +2023-10-05 20:56:27,943 - Epoch: [30][ 1020/ 1236] Overall Loss 0.334680 Objective Loss 0.334680 LR 0.001000 Time 0.020949 +2023-10-05 20:56:28,140 - Epoch: [30][ 1030/ 1236] Overall Loss 0.335084 Objective Loss 0.335084 LR 0.001000 Time 0.020937 +2023-10-05 20:56:28,337 - Epoch: [30][ 1040/ 1236] Overall Loss 0.335466 Objective Loss 0.335466 LR 0.001000 Time 0.020925 +2023-10-05 20:56:28,535 - Epoch: [30][ 1050/ 1236] Overall Loss 0.335561 Objective Loss 0.335561 LR 0.001000 Time 0.020914 +2023-10-05 20:56:28,733 - Epoch: [30][ 1060/ 1236] Overall Loss 0.335275 Objective Loss 0.335275 LR 0.001000 Time 0.020902 +2023-10-05 20:56:28,930 - Epoch: [30][ 1070/ 1236] Overall Loss 0.335320 Objective Loss 0.335320 LR 0.001000 Time 0.020891 +2023-10-05 20:56:29,128 - Epoch: [30][ 1080/ 1236] Overall Loss 0.335367 Objective Loss 0.335367 LR 0.001000 Time 0.020881 +2023-10-05 20:56:29,325 - Epoch: [30][ 1090/ 1236] Overall Loss 0.335355 Objective Loss 0.335355 LR 0.001000 Time 0.020870 +2023-10-05 20:56:29,522 - Epoch: [30][ 1100/ 1236] Overall Loss 0.335242 Objective Loss 0.335242 LR 0.001000 Time 0.020859 +2023-10-05 20:56:29,720 - Epoch: [30][ 1110/ 1236] Overall Loss 0.335283 Objective Loss 0.335283 LR 0.001000 Time 0.020849 +2023-10-05 20:56:29,917 - Epoch: [30][ 1120/ 1236] Overall Loss 0.335327 Objective Loss 0.335327 LR 0.001000 Time 0.020838 +2023-10-05 20:56:30,114 - Epoch: [30][ 1130/ 1236] Overall Loss 0.335274 Objective Loss 0.335274 LR 0.001000 Time 0.020828 +2023-10-05 20:56:30,312 - Epoch: [30][ 1140/ 1236] Overall Loss 0.334934 Objective Loss 0.334934 LR 0.001000 Time 0.020819 +2023-10-05 20:56:30,509 - Epoch: [30][ 1150/ 1236] Overall Loss 0.335523 Objective Loss 0.335523 LR 0.001000 Time 0.020809 +2023-10-05 20:56:30,707 - Epoch: [30][ 1160/ 1236] Overall Loss 0.335888 Objective Loss 0.335888 LR 0.001000 Time 0.020800 +2023-10-05 20:56:30,904 - Epoch: [30][ 1170/ 1236] Overall Loss 0.336070 Objective Loss 0.336070 LR 0.001000 Time 0.020790 +2023-10-05 20:56:31,101 - Epoch: [30][ 1180/ 1236] Overall Loss 0.336223 Objective Loss 0.336223 LR 0.001000 Time 0.020781 +2023-10-05 20:56:31,299 - Epoch: [30][ 1190/ 1236] Overall Loss 0.336541 Objective Loss 0.336541 LR 0.001000 Time 0.020772 +2023-10-05 20:56:31,496 - Epoch: [30][ 1200/ 1236] Overall Loss 0.336654 Objective Loss 0.336654 LR 0.001000 Time 0.020763 +2023-10-05 20:56:31,694 - Epoch: [30][ 1210/ 1236] Overall Loss 0.336745 Objective Loss 0.336745 LR 0.001000 Time 0.020753 +2023-10-05 20:56:31,892 - Epoch: [30][ 1220/ 1236] Overall Loss 0.336528 Objective Loss 0.336528 LR 0.001000 Time 0.020745 +2023-10-05 20:56:32,146 - Epoch: [30][ 1230/ 1236] Overall Loss 0.336347 Objective Loss 0.336347 LR 0.001000 Time 0.020782 +2023-10-05 20:56:32,265 - Epoch: [30][ 1236/ 1236] Overall Loss 0.336409 Objective Loss 0.336409 Top1 83.706721 Top5 97.352342 LR 0.001000 Time 0.020777 +2023-10-05 20:56:32,381 - --- validate (epoch=30)----------- +2023-10-05 20:56:32,382 - 29943 samples (256 per mini-batch) +2023-10-05 20:56:32,834 - Epoch: [30][ 10/ 117] Loss 0.365168 Top1 78.671875 Top5 96.992188 +2023-10-05 20:56:32,985 - Epoch: [30][ 20/ 117] Loss 0.367268 Top1 79.277344 Top5 97.070312 +2023-10-05 20:56:33,136 - Epoch: [30][ 30/ 117] Loss 0.360340 Top1 79.609375 Top5 97.070312 +2023-10-05 20:56:33,286 - Epoch: [30][ 40/ 117] Loss 0.361379 Top1 79.511719 Top5 96.953125 +2023-10-05 20:56:33,436 - Epoch: [30][ 50/ 117] Loss 0.358777 Top1 79.664062 Top5 97.085938 +2023-10-05 20:56:33,586 - Epoch: [30][ 60/ 117] Loss 0.361486 Top1 79.570312 Top5 97.096354 +2023-10-05 20:56:33,740 - Epoch: [30][ 70/ 117] Loss 0.363574 Top1 79.609375 Top5 97.070312 +2023-10-05 20:56:33,898 - Epoch: [30][ 80/ 117] Loss 0.366061 Top1 79.638672 Top5 97.075195 +2023-10-05 20:56:34,055 - Epoch: [30][ 90/ 117] Loss 0.362400 Top1 79.696181 Top5 97.096354 +2023-10-05 20:56:34,213 - Epoch: [30][ 100/ 117] Loss 0.367061 Top1 79.597656 Top5 97.062500 +2023-10-05 20:56:34,378 - Epoch: [30][ 110/ 117] Loss 0.366299 Top1 79.563210 Top5 97.059659 +2023-10-05 20:56:34,464 - Epoch: [30][ 117/ 117] Loss 0.364305 Top1 79.541128 Top5 97.054403 +2023-10-05 20:56:34,568 - ==> Top1: 79.541 Top5: 97.054 Loss: 0.364 + +2023-10-05 20:56:34,569 - ==> Confusion: +[[ 926 0 6 1 5 3 0 0 4 72 1 1 0 2 10 1 10 0 0 1 7] + [ 1 1011 3 0 10 28 1 30 6 0 5 2 0 1 4 3 6 0 11 1 8] + [ 6 0 953 17 1 0 20 5 0 2 10 4 7 4 3 1 5 1 5 3 9] + [ 3 0 17 960 0 3 0 1 1 0 15 0 2 2 42 5 4 6 16 0 12] + [ 24 9 3 0 950 1 0 1 0 8 2 4 0 3 13 9 20 1 0 1 1] + [ 6 26 3 3 5 963 0 26 5 3 6 7 1 27 6 2 7 0 4 5 11] + [ 1 8 52 2 0 1 1083 8 0 0 10 4 1 0 0 6 1 1 2 8 3] + [ 5 10 25 1 2 31 5 1064 1 2 6 14 2 3 0 2 0 0 35 5 5] + [ 28 5 0 0 1 6 0 0 954 35 13 3 0 12 21 5 1 0 5 0 0] + [ 111 0 1 1 6 3 0 1 31 918 1 4 0 20 7 6 0 0 0 1 8] + [ 2 2 14 8 0 0 1 5 17 1 975 6 0 7 2 3 0 1 2 0 7] + [ 0 0 1 0 0 13 0 2 0 0 0 949 29 9 0 5 0 15 0 9 3] + [ 0 0 5 6 1 1 0 2 0 0 0 48 940 4 4 11 5 34 1 4 2] + [ 1 0 1 0 4 2 0 0 9 19 20 2 1 1044 4 2 4 1 0 0 5] + [ 15 0 3 18 4 0 0 0 17 7 5 0 2 1 1004 1 3 4 6 0 11] + [ 0 1 2 5 2 1 0 0 0 0 0 9 9 5 0 1051 19 17 0 9 4] + [ 2 4 1 0 6 9 0 0 4 0 1 3 1 1 4 8 1109 0 0 5 3] + [ 0 0 0 1 0 0 1 1 0 1 0 6 8 0 1 10 4 1002 1 0 2] + [ 0 6 10 13 3 0 0 43 3 0 3 1 2 2 15 0 1 0 958 2 6] + [ 0 0 4 2 0 8 11 15 0 0 8 23 3 3 0 5 13 0 4 1046 7] + [ 189 183 249 127 139 200 47 124 104 114 260 177 461 353 206 85 406 83 175 266 3957]] + +2023-10-05 20:56:34,570 - ==> Best [Top1: 80.987 Top5: 97.425 Sparsity:0.00 Params: 148928 on epoch: 29] +2023-10-05 20:56:34,570 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:56:34,576 - + +2023-10-05 20:56:34,576 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:56:35,586 - Epoch: [31][ 10/ 1236] Overall Loss 0.318888 Objective Loss 0.318888 LR 0.001000 Time 0.100944 +2023-10-05 20:56:35,788 - Epoch: [31][ 20/ 1236] Overall Loss 0.323524 Objective Loss 0.323524 LR 0.001000 Time 0.060553 +2023-10-05 20:56:35,989 - Epoch: [31][ 30/ 1236] Overall Loss 0.320105 Objective Loss 0.320105 LR 0.001000 Time 0.047044 +2023-10-05 20:56:36,191 - Epoch: [31][ 40/ 1236] Overall Loss 0.319954 Objective Loss 0.319954 LR 0.001000 Time 0.040334 +2023-10-05 20:56:36,392 - Epoch: [31][ 50/ 1236] Overall Loss 0.316802 Objective Loss 0.316802 LR 0.001000 Time 0.036273 +2023-10-05 20:56:36,605 - Epoch: [31][ 60/ 1236] Overall Loss 0.321473 Objective Loss 0.321473 LR 0.001000 Time 0.033767 +2023-10-05 20:56:36,813 - Epoch: [31][ 70/ 1236] Overall Loss 0.323006 Objective Loss 0.323006 LR 0.001000 Time 0.031918 +2023-10-05 20:56:37,026 - Epoch: [31][ 80/ 1236] Overall Loss 0.324153 Objective Loss 0.324153 LR 0.001000 Time 0.030587 +2023-10-05 20:56:37,235 - Epoch: [31][ 90/ 1236] Overall Loss 0.327681 Objective Loss 0.327681 LR 0.001000 Time 0.029504 +2023-10-05 20:56:37,449 - Epoch: [31][ 100/ 1236] Overall Loss 0.324372 Objective Loss 0.324372 LR 0.001000 Time 0.028686 +2023-10-05 20:56:37,652 - Epoch: [31][ 110/ 1236] Overall Loss 0.324951 Objective Loss 0.324951 LR 0.001000 Time 0.027924 +2023-10-05 20:56:37,855 - Epoch: [31][ 120/ 1236] Overall Loss 0.326832 Objective Loss 0.326832 LR 0.001000 Time 0.027281 +2023-10-05 20:56:38,056 - Epoch: [31][ 130/ 1236] Overall Loss 0.329706 Objective Loss 0.329706 LR 0.001000 Time 0.026726 +2023-10-05 20:56:38,258 - Epoch: [31][ 140/ 1236] Overall Loss 0.328148 Objective Loss 0.328148 LR 0.001000 Time 0.026261 +2023-10-05 20:56:38,459 - Epoch: [31][ 150/ 1236] Overall Loss 0.325854 Objective Loss 0.325854 LR 0.001000 Time 0.025846 +2023-10-05 20:56:38,662 - Epoch: [31][ 160/ 1236] Overall Loss 0.326769 Objective Loss 0.326769 LR 0.001000 Time 0.025498 +2023-10-05 20:56:38,862 - Epoch: [31][ 170/ 1236] Overall Loss 0.327087 Objective Loss 0.327087 LR 0.001000 Time 0.025174 +2023-10-05 20:56:39,066 - Epoch: [31][ 180/ 1236] Overall Loss 0.328143 Objective Loss 0.328143 LR 0.001000 Time 0.024902 +2023-10-05 20:56:39,266 - Epoch: [31][ 190/ 1236] Overall Loss 0.327914 Objective Loss 0.327914 LR 0.001000 Time 0.024646 +2023-10-05 20:56:39,469 - Epoch: [31][ 200/ 1236] Overall Loss 0.327708 Objective Loss 0.327708 LR 0.001000 Time 0.024426 +2023-10-05 20:56:39,670 - Epoch: [31][ 210/ 1236] Overall Loss 0.329212 Objective Loss 0.329212 LR 0.001000 Time 0.024219 +2023-10-05 20:56:39,873 - Epoch: [31][ 220/ 1236] Overall Loss 0.328122 Objective Loss 0.328122 LR 0.001000 Time 0.024037 +2023-10-05 20:56:40,074 - Epoch: [31][ 230/ 1236] Overall Loss 0.327804 Objective Loss 0.327804 LR 0.001000 Time 0.023864 +2023-10-05 20:56:40,276 - Epoch: [31][ 240/ 1236] Overall Loss 0.328408 Objective Loss 0.328408 LR 0.001000 Time 0.023713 +2023-10-05 20:56:40,477 - Epoch: [31][ 250/ 1236] Overall Loss 0.327216 Objective Loss 0.327216 LR 0.001000 Time 0.023566 +2023-10-05 20:56:40,680 - Epoch: [31][ 260/ 1236] Overall Loss 0.327277 Objective Loss 0.327277 LR 0.001000 Time 0.023438 +2023-10-05 20:56:40,881 - Epoch: [31][ 270/ 1236] Overall Loss 0.326397 Objective Loss 0.326397 LR 0.001000 Time 0.023314 +2023-10-05 20:56:41,084 - Epoch: [31][ 280/ 1236] Overall Loss 0.327286 Objective Loss 0.327286 LR 0.001000 Time 0.023205 +2023-10-05 20:56:41,285 - Epoch: [31][ 290/ 1236] Overall Loss 0.328308 Objective Loss 0.328308 LR 0.001000 Time 0.023097 +2023-10-05 20:56:41,488 - Epoch: [31][ 300/ 1236] Overall Loss 0.327858 Objective Loss 0.327858 LR 0.001000 Time 0.023003 +2023-10-05 20:56:41,690 - Epoch: [31][ 310/ 1236] Overall Loss 0.327732 Objective Loss 0.327732 LR 0.001000 Time 0.022910 +2023-10-05 20:56:41,893 - Epoch: [31][ 320/ 1236] Overall Loss 0.327195 Objective Loss 0.327195 LR 0.001000 Time 0.022827 +2023-10-05 20:56:42,094 - Epoch: [31][ 330/ 1236] Overall Loss 0.327678 Objective Loss 0.327678 LR 0.001000 Time 0.022743 +2023-10-05 20:56:42,297 - Epoch: [31][ 340/ 1236] Overall Loss 0.328656 Objective Loss 0.328656 LR 0.001000 Time 0.022669 +2023-10-05 20:56:42,498 - Epoch: [31][ 350/ 1236] Overall Loss 0.328766 Objective Loss 0.328766 LR 0.001000 Time 0.022595 +2023-10-05 20:56:42,701 - Epoch: [31][ 360/ 1236] Overall Loss 0.328581 Objective Loss 0.328581 LR 0.001000 Time 0.022530 +2023-10-05 20:56:42,901 - Epoch: [31][ 370/ 1236] Overall Loss 0.328627 Objective Loss 0.328627 LR 0.001000 Time 0.022463 +2023-10-05 20:56:43,104 - Epoch: [31][ 380/ 1236] Overall Loss 0.329706 Objective Loss 0.329706 LR 0.001000 Time 0.022405 +2023-10-05 20:56:43,306 - Epoch: [31][ 390/ 1236] Overall Loss 0.329855 Objective Loss 0.329855 LR 0.001000 Time 0.022345 +2023-10-05 20:56:43,509 - Epoch: [31][ 400/ 1236] Overall Loss 0.329326 Objective Loss 0.329326 LR 0.001000 Time 0.022294 +2023-10-05 20:56:43,711 - Epoch: [31][ 410/ 1236] Overall Loss 0.329517 Objective Loss 0.329517 LR 0.001000 Time 0.022243 +2023-10-05 20:56:43,917 - Epoch: [31][ 420/ 1236] Overall Loss 0.329229 Objective Loss 0.329229 LR 0.001000 Time 0.022201 +2023-10-05 20:56:44,120 - Epoch: [31][ 430/ 1236] Overall Loss 0.329408 Objective Loss 0.329408 LR 0.001000 Time 0.022156 +2023-10-05 20:56:44,325 - Epoch: [31][ 440/ 1236] Overall Loss 0.329686 Objective Loss 0.329686 LR 0.001000 Time 0.022119 +2023-10-05 20:56:44,528 - Epoch: [31][ 450/ 1236] Overall Loss 0.329252 Objective Loss 0.329252 LR 0.001000 Time 0.022076 +2023-10-05 20:56:44,733 - Epoch: [31][ 460/ 1236] Overall Loss 0.329358 Objective Loss 0.329358 LR 0.001000 Time 0.022042 +2023-10-05 20:56:44,936 - Epoch: [31][ 470/ 1236] Overall Loss 0.330262 Objective Loss 0.330262 LR 0.001000 Time 0.022004 +2023-10-05 20:56:45,141 - Epoch: [31][ 480/ 1236] Overall Loss 0.330016 Objective Loss 0.330016 LR 0.001000 Time 0.021973 +2023-10-05 20:56:45,344 - Epoch: [31][ 490/ 1236] Overall Loss 0.330324 Objective Loss 0.330324 LR 0.001000 Time 0.021938 +2023-10-05 20:56:45,550 - Epoch: [31][ 500/ 1236] Overall Loss 0.330912 Objective Loss 0.330912 LR 0.001000 Time 0.021911 +2023-10-05 20:56:45,753 - Epoch: [31][ 510/ 1236] Overall Loss 0.331069 Objective Loss 0.331069 LR 0.001000 Time 0.021878 +2023-10-05 20:56:45,959 - Epoch: [31][ 520/ 1236] Overall Loss 0.331183 Objective Loss 0.331183 LR 0.001000 Time 0.021853 +2023-10-05 20:56:46,162 - Epoch: [31][ 530/ 1236] Overall Loss 0.331730 Objective Loss 0.331730 LR 0.001000 Time 0.021822 +2023-10-05 20:56:46,367 - Epoch: [31][ 540/ 1236] Overall Loss 0.331608 Objective Loss 0.331608 LR 0.001000 Time 0.021798 +2023-10-05 20:56:46,570 - Epoch: [31][ 550/ 1236] Overall Loss 0.331988 Objective Loss 0.331988 LR 0.001000 Time 0.021770 +2023-10-05 20:56:46,776 - Epoch: [31][ 560/ 1236] Overall Loss 0.332168 Objective Loss 0.332168 LR 0.001000 Time 0.021748 +2023-10-05 20:56:46,979 - Epoch: [31][ 570/ 1236] Overall Loss 0.332460 Objective Loss 0.332460 LR 0.001000 Time 0.021721 +2023-10-05 20:56:47,185 - Epoch: [31][ 580/ 1236] Overall Loss 0.331780 Objective Loss 0.331780 LR 0.001000 Time 0.021701 +2023-10-05 20:56:47,387 - Epoch: [31][ 590/ 1236] Overall Loss 0.332182 Objective Loss 0.332182 LR 0.001000 Time 0.021676 +2023-10-05 20:56:47,593 - Epoch: [31][ 600/ 1236] Overall Loss 0.331812 Objective Loss 0.331812 LR 0.001000 Time 0.021657 +2023-10-05 20:56:47,796 - Epoch: [31][ 610/ 1236] Overall Loss 0.332591 Objective Loss 0.332591 LR 0.001000 Time 0.021634 +2023-10-05 20:56:48,001 - Epoch: [31][ 620/ 1236] Overall Loss 0.332932 Objective Loss 0.332932 LR 0.001000 Time 0.021616 +2023-10-05 20:56:48,204 - Epoch: [31][ 630/ 1236] Overall Loss 0.332635 Objective Loss 0.332635 LR 0.001000 Time 0.021594 +2023-10-05 20:56:48,410 - Epoch: [31][ 640/ 1236] Overall Loss 0.332619 Objective Loss 0.332619 LR 0.001000 Time 0.021578 +2023-10-05 20:56:48,613 - Epoch: [31][ 650/ 1236] Overall Loss 0.332934 Objective Loss 0.332934 LR 0.001000 Time 0.021557 +2023-10-05 20:56:48,819 - Epoch: [31][ 660/ 1236] Overall Loss 0.333240 Objective Loss 0.333240 LR 0.001000 Time 0.021542 +2023-10-05 20:56:49,021 - Epoch: [31][ 670/ 1236] Overall Loss 0.333358 Objective Loss 0.333358 LR 0.001000 Time 0.021523 +2023-10-05 20:56:49,227 - Epoch: [31][ 680/ 1236] Overall Loss 0.333182 Objective Loss 0.333182 LR 0.001000 Time 0.021508 +2023-10-05 20:56:49,430 - Epoch: [31][ 690/ 1236] Overall Loss 0.333624 Objective Loss 0.333624 LR 0.001000 Time 0.021490 +2023-10-05 20:56:49,636 - Epoch: [31][ 700/ 1236] Overall Loss 0.333963 Objective Loss 0.333963 LR 0.001000 Time 0.021476 +2023-10-05 20:56:49,839 - Epoch: [31][ 710/ 1236] Overall Loss 0.334004 Objective Loss 0.334004 LR 0.001000 Time 0.021459 +2023-10-05 20:56:50,046 - Epoch: [31][ 720/ 1236] Overall Loss 0.334483 Objective Loss 0.334483 LR 0.001000 Time 0.021449 +2023-10-05 20:56:50,250 - Epoch: [31][ 730/ 1236] Overall Loss 0.334706 Objective Loss 0.334706 LR 0.001000 Time 0.021434 +2023-10-05 20:56:50,458 - Epoch: [31][ 740/ 1236] Overall Loss 0.334960 Objective Loss 0.334960 LR 0.001000 Time 0.021424 +2023-10-05 20:56:50,662 - Epoch: [31][ 750/ 1236] Overall Loss 0.334980 Objective Loss 0.334980 LR 0.001000 Time 0.021410 +2023-10-05 20:56:50,869 - Epoch: [31][ 760/ 1236] Overall Loss 0.334748 Objective Loss 0.334748 LR 0.001000 Time 0.021401 +2023-10-05 20:56:51,073 - Epoch: [31][ 770/ 1236] Overall Loss 0.335261 Objective Loss 0.335261 LR 0.001000 Time 0.021388 +2023-10-05 20:56:51,281 - Epoch: [31][ 780/ 1236] Overall Loss 0.335323 Objective Loss 0.335323 LR 0.001000 Time 0.021379 +2023-10-05 20:56:51,485 - Epoch: [31][ 790/ 1236] Overall Loss 0.335094 Objective Loss 0.335094 LR 0.001000 Time 0.021366 +2023-10-05 20:56:51,692 - Epoch: [31][ 800/ 1236] Overall Loss 0.335065 Objective Loss 0.335065 LR 0.001000 Time 0.021358 +2023-10-05 20:56:51,896 - Epoch: [31][ 810/ 1236] Overall Loss 0.334961 Objective Loss 0.334961 LR 0.001000 Time 0.021346 +2023-10-05 20:56:52,103 - Epoch: [31][ 820/ 1236] Overall Loss 0.334871 Objective Loss 0.334871 LR 0.001000 Time 0.021338 +2023-10-05 20:56:52,305 - Epoch: [31][ 830/ 1236] Overall Loss 0.335296 Objective Loss 0.335296 LR 0.001000 Time 0.021324 +2023-10-05 20:56:52,511 - Epoch: [31][ 840/ 1236] Overall Loss 0.335683 Objective Loss 0.335683 LR 0.001000 Time 0.021314 +2023-10-05 20:56:52,713 - Epoch: [31][ 850/ 1236] Overall Loss 0.336109 Objective Loss 0.336109 LR 0.001000 Time 0.021300 +2023-10-05 20:56:52,917 - Epoch: [31][ 860/ 1236] Overall Loss 0.336375 Objective Loss 0.336375 LR 0.001000 Time 0.021290 +2023-10-05 20:56:53,119 - Epoch: [31][ 870/ 1236] Overall Loss 0.337068 Objective Loss 0.337068 LR 0.001000 Time 0.021277 +2023-10-05 20:56:53,324 - Epoch: [31][ 880/ 1236] Overall Loss 0.336905 Objective Loss 0.336905 LR 0.001000 Time 0.021268 +2023-10-05 20:56:53,527 - Epoch: [31][ 890/ 1236] Overall Loss 0.336605 Objective Loss 0.336605 LR 0.001000 Time 0.021256 +2023-10-05 20:56:53,732 - Epoch: [31][ 900/ 1236] Overall Loss 0.336404 Objective Loss 0.336404 LR 0.001000 Time 0.021247 +2023-10-05 20:56:53,934 - Epoch: [31][ 910/ 1236] Overall Loss 0.336436 Objective Loss 0.336436 LR 0.001000 Time 0.021236 +2023-10-05 20:56:54,139 - Epoch: [31][ 920/ 1236] Overall Loss 0.336329 Objective Loss 0.336329 LR 0.001000 Time 0.021227 +2023-10-05 20:56:54,341 - Epoch: [31][ 930/ 1236] Overall Loss 0.336431 Objective Loss 0.336431 LR 0.001000 Time 0.021216 +2023-10-05 20:56:54,546 - Epoch: [31][ 940/ 1236] Overall Loss 0.336716 Objective Loss 0.336716 LR 0.001000 Time 0.021208 +2023-10-05 20:56:54,748 - Epoch: [31][ 950/ 1236] Overall Loss 0.336914 Objective Loss 0.336914 LR 0.001000 Time 0.021197 +2023-10-05 20:56:54,953 - Epoch: [31][ 960/ 1236] Overall Loss 0.336943 Objective Loss 0.336943 LR 0.001000 Time 0.021189 +2023-10-05 20:56:55,155 - Epoch: [31][ 970/ 1236] Overall Loss 0.337002 Objective Loss 0.337002 LR 0.001000 Time 0.021179 +2023-10-05 20:56:55,361 - Epoch: [31][ 980/ 1236] Overall Loss 0.337006 Objective Loss 0.337006 LR 0.001000 Time 0.021172 +2023-10-05 20:56:55,564 - Epoch: [31][ 990/ 1236] Overall Loss 0.336514 Objective Loss 0.336514 LR 0.001000 Time 0.021163 +2023-10-05 20:56:55,768 - Epoch: [31][ 1000/ 1236] Overall Loss 0.336431 Objective Loss 0.336431 LR 0.001000 Time 0.021155 +2023-10-05 20:56:55,970 - Epoch: [31][ 1010/ 1236] Overall Loss 0.336535 Objective Loss 0.336535 LR 0.001000 Time 0.021146 +2023-10-05 20:56:56,175 - Epoch: [31][ 1020/ 1236] Overall Loss 0.336390 Objective Loss 0.336390 LR 0.001000 Time 0.021139 +2023-10-05 20:56:56,377 - Epoch: [31][ 1030/ 1236] Overall Loss 0.336225 Objective Loss 0.336225 LR 0.001000 Time 0.021130 +2023-10-05 20:56:56,582 - Epoch: [31][ 1040/ 1236] Overall Loss 0.335997 Objective Loss 0.335997 LR 0.001000 Time 0.021123 +2023-10-05 20:56:56,784 - Epoch: [31][ 1050/ 1236] Overall Loss 0.335812 Objective Loss 0.335812 LR 0.001000 Time 0.021114 +2023-10-05 20:56:56,989 - Epoch: [31][ 1060/ 1236] Overall Loss 0.335937 Objective Loss 0.335937 LR 0.001000 Time 0.021108 +2023-10-05 20:56:57,191 - Epoch: [31][ 1070/ 1236] Overall Loss 0.335939 Objective Loss 0.335939 LR 0.001000 Time 0.021099 +2023-10-05 20:56:57,396 - Epoch: [31][ 1080/ 1236] Overall Loss 0.335783 Objective Loss 0.335783 LR 0.001000 Time 0.021093 +2023-10-05 20:56:57,598 - Epoch: [31][ 1090/ 1236] Overall Loss 0.335983 Objective Loss 0.335983 LR 0.001000 Time 0.021085 +2023-10-05 20:56:57,803 - Epoch: [31][ 1100/ 1236] Overall Loss 0.336154 Objective Loss 0.336154 LR 0.001000 Time 0.021079 +2023-10-05 20:56:58,006 - Epoch: [31][ 1110/ 1236] Overall Loss 0.335824 Objective Loss 0.335824 LR 0.001000 Time 0.021071 +2023-10-05 20:56:58,210 - Epoch: [31][ 1120/ 1236] Overall Loss 0.335634 Objective Loss 0.335634 LR 0.001000 Time 0.021065 +2023-10-05 20:56:58,413 - Epoch: [31][ 1130/ 1236] Overall Loss 0.335982 Objective Loss 0.335982 LR 0.001000 Time 0.021058 +2023-10-05 20:56:58,617 - Epoch: [31][ 1140/ 1236] Overall Loss 0.336259 Objective Loss 0.336259 LR 0.001000 Time 0.021052 +2023-10-05 20:56:58,819 - Epoch: [31][ 1150/ 1236] Overall Loss 0.336019 Objective Loss 0.336019 LR 0.001000 Time 0.021045 +2023-10-05 20:56:59,024 - Epoch: [31][ 1160/ 1236] Overall Loss 0.335907 Objective Loss 0.335907 LR 0.001000 Time 0.021040 +2023-10-05 20:56:59,227 - Epoch: [31][ 1170/ 1236] Overall Loss 0.335485 Objective Loss 0.335485 LR 0.001000 Time 0.021032 +2023-10-05 20:56:59,432 - Epoch: [31][ 1180/ 1236] Overall Loss 0.335371 Objective Loss 0.335371 LR 0.001000 Time 0.021027 +2023-10-05 20:56:59,634 - Epoch: [31][ 1190/ 1236] Overall Loss 0.335016 Objective Loss 0.335016 LR 0.001000 Time 0.021021 +2023-10-05 20:56:59,839 - Epoch: [31][ 1200/ 1236] Overall Loss 0.334898 Objective Loss 0.334898 LR 0.001000 Time 0.021016 +2023-10-05 20:57:00,041 - Epoch: [31][ 1210/ 1236] Overall Loss 0.334823 Objective Loss 0.334823 LR 0.001000 Time 0.021009 +2023-10-05 20:57:00,246 - Epoch: [31][ 1220/ 1236] Overall Loss 0.334684 Objective Loss 0.334684 LR 0.001000 Time 0.021004 +2023-10-05 20:57:00,504 - Epoch: [31][ 1230/ 1236] Overall Loss 0.334607 Objective Loss 0.334607 LR 0.001000 Time 0.021043 +2023-10-05 20:57:00,624 - Epoch: [31][ 1236/ 1236] Overall Loss 0.334557 Objective Loss 0.334557 Top1 82.688391 Top5 98.167006 LR 0.001000 Time 0.021038 +2023-10-05 20:57:00,763 - --- validate (epoch=31)----------- +2023-10-05 20:57:00,764 - 29943 samples (256 per mini-batch) +2023-10-05 20:57:01,234 - Epoch: [31][ 10/ 117] Loss 0.355400 Top1 81.015625 Top5 97.578125 +2023-10-05 20:57:01,379 - Epoch: [31][ 20/ 117] Loss 0.362299 Top1 81.582031 Top5 97.695312 +2023-10-05 20:57:01,525 - Epoch: [31][ 30/ 117] Loss 0.358145 Top1 81.809896 Top5 97.630208 +2023-10-05 20:57:01,670 - Epoch: [31][ 40/ 117] Loss 0.353315 Top1 82.011719 Top5 97.666016 +2023-10-05 20:57:01,816 - Epoch: [31][ 50/ 117] Loss 0.353638 Top1 82.015625 Top5 97.695312 +2023-10-05 20:57:01,961 - Epoch: [31][ 60/ 117] Loss 0.358138 Top1 81.894531 Top5 97.682292 +2023-10-05 20:57:02,104 - Epoch: [31][ 70/ 117] Loss 0.364876 Top1 81.674107 Top5 97.628348 +2023-10-05 20:57:02,247 - Epoch: [31][ 80/ 117] Loss 0.369626 Top1 81.455078 Top5 97.607422 +2023-10-05 20:57:02,392 - Epoch: [31][ 90/ 117] Loss 0.371221 Top1 81.345486 Top5 97.569444 +2023-10-05 20:57:02,535 - Epoch: [31][ 100/ 117] Loss 0.368618 Top1 81.378906 Top5 97.593750 +2023-10-05 20:57:02,687 - Epoch: [31][ 110/ 117] Loss 0.367005 Top1 81.434659 Top5 97.592330 +2023-10-05 20:57:02,773 - Epoch: [31][ 117/ 117] Loss 0.367059 Top1 81.428047 Top5 97.605450 +2023-10-05 20:57:02,894 - ==> Top1: 81.428 Top5: 97.605 Loss: 0.367 + +2023-10-05 20:57:02,895 - ==> Confusion: +[[ 921 3 5 0 5 4 0 0 3 82 1 1 0 3 12 1 3 0 0 0 6] + [ 6 1042 2 1 6 23 0 25 2 0 3 1 0 0 2 4 2 0 5 1 6] + [ 12 0 928 31 1 0 26 9 0 0 3 3 8 1 2 3 2 1 13 3 10] + [ 3 0 8 976 0 6 1 0 4 0 2 0 4 3 34 3 0 8 19 5 13] + [ 40 8 2 0 937 5 0 0 2 14 3 2 1 1 15 3 6 3 1 0 7] + [ 9 51 2 2 5 966 1 23 0 1 5 6 2 16 5 0 4 0 3 3 12] + [ 0 8 26 0 1 1 1102 12 0 1 5 4 1 0 1 10 1 2 3 8 5] + [ 17 26 13 2 3 36 2 1036 0 0 5 8 1 0 0 0 0 0 52 11 6] + [ 26 2 0 0 1 4 0 0 963 37 11 4 0 8 20 4 1 0 4 0 4] + [ 106 0 3 1 4 4 0 0 38 910 2 1 0 23 7 5 0 1 0 3 11] + [ 6 2 7 17 0 1 2 6 12 2 955 3 0 5 12 1 0 1 12 1 8] + [ 2 0 2 0 1 21 0 5 0 0 0 944 17 8 0 2 2 17 0 9 5] + [ 0 0 6 16 0 0 0 2 1 2 1 50 940 1 1 7 2 23 2 1 13] + [ 5 2 2 0 3 17 0 0 13 10 10 7 2 1029 3 1 5 1 0 2 7] + [ 20 3 4 27 0 0 0 0 28 5 0 0 0 1 991 0 1 3 8 0 10] + [ 2 1 3 1 1 0 3 1 0 0 0 11 5 1 0 1059 16 16 2 5 7] + [ 3 13 4 1 3 6 0 0 2 0 0 7 0 2 3 12 1081 0 1 5 18] + [ 1 0 1 4 0 0 1 0 2 0 0 6 12 1 2 4 0 1003 1 0 0] + [ 1 11 14 26 3 0 0 17 7 0 1 2 0 0 19 0 0 0 957 0 10] + [ 0 1 4 2 2 7 7 19 1 0 0 27 3 0 1 6 6 1 3 1052 10] + [ 264 240 149 121 81 174 38 95 118 96 172 161 360 301 201 75 140 93 226 210 4590]] + +2023-10-05 20:57:02,896 - ==> Best [Top1: 81.428 Top5: 97.605 Sparsity:0.00 Params: 148928 on epoch: 31] +2023-10-05 20:57:02,896 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:57:02,910 - + +2023-10-05 20:57:02,910 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:57:03,918 - Epoch: [32][ 10/ 1236] Overall Loss 0.282312 Objective Loss 0.282312 LR 0.001000 Time 0.100726 +2023-10-05 20:57:04,121 - Epoch: [32][ 20/ 1236] Overall Loss 0.304424 Objective Loss 0.304424 LR 0.001000 Time 0.060491 +2023-10-05 20:57:04,323 - Epoch: [32][ 30/ 1236] Overall Loss 0.310997 Objective Loss 0.310997 LR 0.001000 Time 0.047063 +2023-10-05 20:57:04,526 - Epoch: [32][ 40/ 1236] Overall Loss 0.315206 Objective Loss 0.315206 LR 0.001000 Time 0.040360 +2023-10-05 20:57:04,728 - Epoch: [32][ 50/ 1236] Overall Loss 0.315525 Objective Loss 0.315525 LR 0.001000 Time 0.036321 +2023-10-05 20:57:04,931 - Epoch: [32][ 60/ 1236] Overall Loss 0.309434 Objective Loss 0.309434 LR 0.001000 Time 0.033647 +2023-10-05 20:57:05,133 - Epoch: [32][ 70/ 1236] Overall Loss 0.309894 Objective Loss 0.309894 LR 0.001000 Time 0.031723 +2023-10-05 20:57:05,336 - Epoch: [32][ 80/ 1236] Overall Loss 0.311279 Objective Loss 0.311279 LR 0.001000 Time 0.030289 +2023-10-05 20:57:05,538 - Epoch: [32][ 90/ 1236] Overall Loss 0.313923 Objective Loss 0.313923 LR 0.001000 Time 0.029164 +2023-10-05 20:57:05,741 - Epoch: [32][ 100/ 1236] Overall Loss 0.311967 Objective Loss 0.311967 LR 0.001000 Time 0.028272 +2023-10-05 20:57:05,943 - Epoch: [32][ 110/ 1236] Overall Loss 0.315532 Objective Loss 0.315532 LR 0.001000 Time 0.027538 +2023-10-05 20:57:06,146 - Epoch: [32][ 120/ 1236] Overall Loss 0.319129 Objective Loss 0.319129 LR 0.001000 Time 0.026930 +2023-10-05 20:57:06,348 - Epoch: [32][ 130/ 1236] Overall Loss 0.320166 Objective Loss 0.320166 LR 0.001000 Time 0.026412 +2023-10-05 20:57:06,551 - Epoch: [32][ 140/ 1236] Overall Loss 0.318050 Objective Loss 0.318050 LR 0.001000 Time 0.025972 +2023-10-05 20:57:06,753 - Epoch: [32][ 150/ 1236] Overall Loss 0.318897 Objective Loss 0.318897 LR 0.001000 Time 0.025584 +2023-10-05 20:57:06,956 - Epoch: [32][ 160/ 1236] Overall Loss 0.317776 Objective Loss 0.317776 LR 0.001000 Time 0.025251 +2023-10-05 20:57:07,158 - Epoch: [32][ 170/ 1236] Overall Loss 0.317675 Objective Loss 0.317675 LR 0.001000 Time 0.024952 +2023-10-05 20:57:07,361 - Epoch: [32][ 180/ 1236] Overall Loss 0.318626 Objective Loss 0.318626 LR 0.001000 Time 0.024692 +2023-10-05 20:57:07,563 - Epoch: [32][ 190/ 1236] Overall Loss 0.317824 Objective Loss 0.317824 LR 0.001000 Time 0.024452 +2023-10-05 20:57:07,766 - Epoch: [32][ 200/ 1236] Overall Loss 0.318196 Objective Loss 0.318196 LR 0.001000 Time 0.024243 +2023-10-05 20:57:07,968 - Epoch: [32][ 210/ 1236] Overall Loss 0.318763 Objective Loss 0.318763 LR 0.001000 Time 0.024050 +2023-10-05 20:57:08,171 - Epoch: [32][ 220/ 1236] Overall Loss 0.319974 Objective Loss 0.319974 LR 0.001000 Time 0.023878 +2023-10-05 20:57:08,373 - Epoch: [32][ 230/ 1236] Overall Loss 0.321831 Objective Loss 0.321831 LR 0.001000 Time 0.023717 +2023-10-05 20:57:08,576 - Epoch: [32][ 240/ 1236] Overall Loss 0.321344 Objective Loss 0.321344 LR 0.001000 Time 0.023573 +2023-10-05 20:57:08,778 - Epoch: [32][ 250/ 1236] Overall Loss 0.321013 Objective Loss 0.321013 LR 0.001000 Time 0.023437 +2023-10-05 20:57:08,981 - Epoch: [32][ 260/ 1236] Overall Loss 0.321453 Objective Loss 0.321453 LR 0.001000 Time 0.023314 +2023-10-05 20:57:09,183 - Epoch: [32][ 270/ 1236] Overall Loss 0.321459 Objective Loss 0.321459 LR 0.001000 Time 0.023197 +2023-10-05 20:57:09,386 - Epoch: [32][ 280/ 1236] Overall Loss 0.321931 Objective Loss 0.321931 LR 0.001000 Time 0.023093 +2023-10-05 20:57:09,588 - Epoch: [32][ 290/ 1236] Overall Loss 0.322184 Objective Loss 0.322184 LR 0.001000 Time 0.022992 +2023-10-05 20:57:09,790 - Epoch: [32][ 300/ 1236] Overall Loss 0.323667 Objective Loss 0.323667 LR 0.001000 Time 0.022896 +2023-10-05 20:57:09,992 - Epoch: [32][ 310/ 1236] Overall Loss 0.323040 Objective Loss 0.323040 LR 0.001000 Time 0.022808 +2023-10-05 20:57:10,194 - Epoch: [32][ 320/ 1236] Overall Loss 0.322514 Objective Loss 0.322514 LR 0.001000 Time 0.022727 +2023-10-05 20:57:10,397 - Epoch: [32][ 330/ 1236] Overall Loss 0.323875 Objective Loss 0.323875 LR 0.001000 Time 0.022652 +2023-10-05 20:57:10,600 - Epoch: [32][ 340/ 1236] Overall Loss 0.324123 Objective Loss 0.324123 LR 0.001000 Time 0.022582 +2023-10-05 20:57:10,803 - Epoch: [32][ 350/ 1236] Overall Loss 0.324428 Objective Loss 0.324428 LR 0.001000 Time 0.022514 +2023-10-05 20:57:11,006 - Epoch: [32][ 360/ 1236] Overall Loss 0.324716 Objective Loss 0.324716 LR 0.001000 Time 0.022451 +2023-10-05 20:57:11,208 - Epoch: [32][ 370/ 1236] Overall Loss 0.324948 Objective Loss 0.324948 LR 0.001000 Time 0.022390 +2023-10-05 20:57:11,410 - Epoch: [32][ 380/ 1236] Overall Loss 0.324845 Objective Loss 0.324845 LR 0.001000 Time 0.022333 +2023-10-05 20:57:11,613 - Epoch: [32][ 390/ 1236] Overall Loss 0.324848 Objective Loss 0.324848 LR 0.001000 Time 0.022279 +2023-10-05 20:57:11,816 - Epoch: [32][ 400/ 1236] Overall Loss 0.324133 Objective Loss 0.324133 LR 0.001000 Time 0.022229 +2023-10-05 20:57:12,019 - Epoch: [32][ 410/ 1236] Overall Loss 0.324237 Objective Loss 0.324237 LR 0.001000 Time 0.022180 +2023-10-05 20:57:12,221 - Epoch: [32][ 420/ 1236] Overall Loss 0.324439 Objective Loss 0.324439 LR 0.001000 Time 0.022134 +2023-10-05 20:57:12,424 - Epoch: [32][ 430/ 1236] Overall Loss 0.325030 Objective Loss 0.325030 LR 0.001000 Time 0.022090 +2023-10-05 20:57:12,627 - Epoch: [32][ 440/ 1236] Overall Loss 0.324670 Objective Loss 0.324670 LR 0.001000 Time 0.022049 +2023-10-05 20:57:12,830 - Epoch: [32][ 450/ 1236] Overall Loss 0.324698 Objective Loss 0.324698 LR 0.001000 Time 0.022008 +2023-10-05 20:57:13,032 - Epoch: [32][ 460/ 1236] Overall Loss 0.324709 Objective Loss 0.324709 LR 0.001000 Time 0.021969 +2023-10-05 20:57:13,235 - Epoch: [32][ 470/ 1236] Overall Loss 0.325073 Objective Loss 0.325073 LR 0.001000 Time 0.021932 +2023-10-05 20:57:13,438 - Epoch: [32][ 480/ 1236] Overall Loss 0.324822 Objective Loss 0.324822 LR 0.001000 Time 0.021898 +2023-10-05 20:57:13,641 - Epoch: [32][ 490/ 1236] Overall Loss 0.324689 Objective Loss 0.324689 LR 0.001000 Time 0.021864 +2023-10-05 20:57:13,843 - Epoch: [32][ 500/ 1236] Overall Loss 0.324773 Objective Loss 0.324773 LR 0.001000 Time 0.021831 +2023-10-05 20:57:14,046 - Epoch: [32][ 510/ 1236] Overall Loss 0.325605 Objective Loss 0.325605 LR 0.001000 Time 0.021799 +2023-10-05 20:57:14,249 - Epoch: [32][ 520/ 1236] Overall Loss 0.325504 Objective Loss 0.325504 LR 0.001000 Time 0.021769 +2023-10-05 20:57:14,452 - Epoch: [32][ 530/ 1236] Overall Loss 0.325517 Objective Loss 0.325517 LR 0.001000 Time 0.021741 +2023-10-05 20:57:14,654 - Epoch: [32][ 540/ 1236] Overall Loss 0.326178 Objective Loss 0.326178 LR 0.001000 Time 0.021713 +2023-10-05 20:57:14,857 - Epoch: [32][ 550/ 1236] Overall Loss 0.326773 Objective Loss 0.326773 LR 0.001000 Time 0.021686 +2023-10-05 20:57:15,059 - Epoch: [32][ 560/ 1236] Overall Loss 0.326575 Objective Loss 0.326575 LR 0.001000 Time 0.021660 +2023-10-05 20:57:15,262 - Epoch: [32][ 570/ 1236] Overall Loss 0.326780 Objective Loss 0.326780 LR 0.001000 Time 0.021634 +2023-10-05 20:57:15,465 - Epoch: [32][ 580/ 1236] Overall Loss 0.326858 Objective Loss 0.326858 LR 0.001000 Time 0.021610 +2023-10-05 20:57:15,667 - Epoch: [32][ 590/ 1236] Overall Loss 0.327307 Objective Loss 0.327307 LR 0.001000 Time 0.021586 +2023-10-05 20:57:15,869 - Epoch: [32][ 600/ 1236] Overall Loss 0.326667 Objective Loss 0.326667 LR 0.001000 Time 0.021564 +2023-10-05 20:57:16,073 - Epoch: [32][ 610/ 1236] Overall Loss 0.326315 Objective Loss 0.326315 LR 0.001000 Time 0.021543 +2023-10-05 20:57:16,275 - Epoch: [32][ 620/ 1236] Overall Loss 0.326878 Objective Loss 0.326878 LR 0.001000 Time 0.021522 +2023-10-05 20:57:16,478 - Epoch: [32][ 630/ 1236] Overall Loss 0.327266 Objective Loss 0.327266 LR 0.001000 Time 0.021501 +2023-10-05 20:57:16,681 - Epoch: [32][ 640/ 1236] Overall Loss 0.327467 Objective Loss 0.327467 LR 0.001000 Time 0.021481 +2023-10-05 20:57:16,883 - Epoch: [32][ 650/ 1236] Overall Loss 0.327826 Objective Loss 0.327826 LR 0.001000 Time 0.021462 +2023-10-05 20:57:17,087 - Epoch: [32][ 660/ 1236] Overall Loss 0.327442 Objective Loss 0.327442 LR 0.001000 Time 0.021444 +2023-10-05 20:57:17,289 - Epoch: [32][ 670/ 1236] Overall Loss 0.327939 Objective Loss 0.327939 LR 0.001000 Time 0.021426 +2023-10-05 20:57:17,493 - Epoch: [32][ 680/ 1236] Overall Loss 0.327995 Objective Loss 0.327995 LR 0.001000 Time 0.021410 +2023-10-05 20:57:17,696 - Epoch: [32][ 690/ 1236] Overall Loss 0.327819 Objective Loss 0.327819 LR 0.001000 Time 0.021393 +2023-10-05 20:57:17,899 - Epoch: [32][ 700/ 1236] Overall Loss 0.328438 Objective Loss 0.328438 LR 0.001000 Time 0.021377 +2023-10-05 20:57:18,102 - Epoch: [32][ 710/ 1236] Overall Loss 0.328470 Objective Loss 0.328470 LR 0.001000 Time 0.021361 +2023-10-05 20:57:18,305 - Epoch: [32][ 720/ 1236] Overall Loss 0.328597 Objective Loss 0.328597 LR 0.001000 Time 0.021346 +2023-10-05 20:57:18,507 - Epoch: [32][ 730/ 1236] Overall Loss 0.328904 Objective Loss 0.328904 LR 0.001000 Time 0.021331 +2023-10-05 20:57:18,711 - Epoch: [32][ 740/ 1236] Overall Loss 0.328902 Objective Loss 0.328902 LR 0.001000 Time 0.021317 +2023-10-05 20:57:18,913 - Epoch: [32][ 750/ 1236] Overall Loss 0.328669 Objective Loss 0.328669 LR 0.001000 Time 0.021302 +2023-10-05 20:57:19,116 - Epoch: [32][ 760/ 1236] Overall Loss 0.328671 Objective Loss 0.328671 LR 0.001000 Time 0.021289 +2023-10-05 20:57:19,319 - Epoch: [32][ 770/ 1236] Overall Loss 0.328435 Objective Loss 0.328435 LR 0.001000 Time 0.021275 +2023-10-05 20:57:19,523 - Epoch: [32][ 780/ 1236] Overall Loss 0.328585 Objective Loss 0.328585 LR 0.001000 Time 0.021263 +2023-10-05 20:57:19,725 - Epoch: [32][ 790/ 1236] Overall Loss 0.328226 Objective Loss 0.328226 LR 0.001000 Time 0.021249 +2023-10-05 20:57:19,928 - Epoch: [32][ 800/ 1236] Overall Loss 0.328418 Objective Loss 0.328418 LR 0.001000 Time 0.021237 +2023-10-05 20:57:20,131 - Epoch: [32][ 810/ 1236] Overall Loss 0.328249 Objective Loss 0.328249 LR 0.001000 Time 0.021225 +2023-10-05 20:57:20,334 - Epoch: [32][ 820/ 1236] Overall Loss 0.328323 Objective Loss 0.328323 LR 0.001000 Time 0.021213 +2023-10-05 20:57:20,536 - Epoch: [32][ 830/ 1236] Overall Loss 0.328518 Objective Loss 0.328518 LR 0.001000 Time 0.021201 +2023-10-05 20:57:20,740 - Epoch: [32][ 840/ 1236] Overall Loss 0.328410 Objective Loss 0.328410 LR 0.001000 Time 0.021190 +2023-10-05 20:57:20,942 - Epoch: [32][ 850/ 1236] Overall Loss 0.328546 Objective Loss 0.328546 LR 0.001000 Time 0.021178 +2023-10-05 20:57:21,145 - Epoch: [32][ 860/ 1236] Overall Loss 0.328628 Objective Loss 0.328628 LR 0.001000 Time 0.021168 +2023-10-05 20:57:21,347 - Epoch: [32][ 870/ 1236] Overall Loss 0.328827 Objective Loss 0.328827 LR 0.001000 Time 0.021156 +2023-10-05 20:57:21,550 - Epoch: [32][ 880/ 1236] Overall Loss 0.329212 Objective Loss 0.329212 LR 0.001000 Time 0.021147 +2023-10-05 20:57:21,753 - Epoch: [32][ 890/ 1236] Overall Loss 0.329664 Objective Loss 0.329664 LR 0.001000 Time 0.021137 +2023-10-05 20:57:21,956 - Epoch: [32][ 900/ 1236] Overall Loss 0.329857 Objective Loss 0.329857 LR 0.001000 Time 0.021127 +2023-10-05 20:57:22,159 - Epoch: [32][ 910/ 1236] Overall Loss 0.330197 Objective Loss 0.330197 LR 0.001000 Time 0.021117 +2023-10-05 20:57:22,362 - Epoch: [32][ 920/ 1236] Overall Loss 0.330457 Objective Loss 0.330457 LR 0.001000 Time 0.021108 +2023-10-05 20:57:22,565 - Epoch: [32][ 930/ 1236] Overall Loss 0.330653 Objective Loss 0.330653 LR 0.001000 Time 0.021099 +2023-10-05 20:57:22,768 - Epoch: [32][ 940/ 1236] Overall Loss 0.330796 Objective Loss 0.330796 LR 0.001000 Time 0.021090 +2023-10-05 20:57:22,970 - Epoch: [32][ 950/ 1236] Overall Loss 0.330601 Objective Loss 0.330601 LR 0.001000 Time 0.021080 +2023-10-05 20:57:23,173 - Epoch: [32][ 960/ 1236] Overall Loss 0.330418 Objective Loss 0.330418 LR 0.001000 Time 0.021072 +2023-10-05 20:57:23,376 - Epoch: [32][ 970/ 1236] Overall Loss 0.330376 Objective Loss 0.330376 LR 0.001000 Time 0.021064 +2023-10-05 20:57:23,579 - Epoch: [32][ 980/ 1236] Overall Loss 0.330568 Objective Loss 0.330568 LR 0.001000 Time 0.021055 +2023-10-05 20:57:23,782 - Epoch: [32][ 990/ 1236] Overall Loss 0.330723 Objective Loss 0.330723 LR 0.001000 Time 0.021047 +2023-10-05 20:57:23,985 - Epoch: [32][ 1000/ 1236] Overall Loss 0.330788 Objective Loss 0.330788 LR 0.001000 Time 0.021040 +2023-10-05 20:57:24,188 - Epoch: [32][ 1010/ 1236] Overall Loss 0.330652 Objective Loss 0.330652 LR 0.001000 Time 0.021032 +2023-10-05 20:57:24,392 - Epoch: [32][ 1020/ 1236] Overall Loss 0.330507 Objective Loss 0.330507 LR 0.001000 Time 0.021025 +2023-10-05 20:57:24,594 - Epoch: [32][ 1030/ 1236] Overall Loss 0.330663 Objective Loss 0.330663 LR 0.001000 Time 0.021017 +2023-10-05 20:57:24,797 - Epoch: [32][ 1040/ 1236] Overall Loss 0.330515 Objective Loss 0.330515 LR 0.001000 Time 0.021010 +2023-10-05 20:57:25,000 - Epoch: [32][ 1050/ 1236] Overall Loss 0.330705 Objective Loss 0.330705 LR 0.001000 Time 0.021003 +2023-10-05 20:57:25,203 - Epoch: [32][ 1060/ 1236] Overall Loss 0.330569 Objective Loss 0.330569 LR 0.001000 Time 0.020996 +2023-10-05 20:57:25,406 - Epoch: [32][ 1070/ 1236] Overall Loss 0.330611 Objective Loss 0.330611 LR 0.001000 Time 0.020989 +2023-10-05 20:57:25,609 - Epoch: [32][ 1080/ 1236] Overall Loss 0.330739 Objective Loss 0.330739 LR 0.001000 Time 0.020983 +2023-10-05 20:57:25,812 - Epoch: [32][ 1090/ 1236] Overall Loss 0.330905 Objective Loss 0.330905 LR 0.001000 Time 0.020976 +2023-10-05 20:57:26,015 - Epoch: [32][ 1100/ 1236] Overall Loss 0.331116 Objective Loss 0.331116 LR 0.001000 Time 0.020969 +2023-10-05 20:57:26,218 - Epoch: [32][ 1110/ 1236] Overall Loss 0.331490 Objective Loss 0.331490 LR 0.001000 Time 0.020963 +2023-10-05 20:57:26,421 - Epoch: [32][ 1120/ 1236] Overall Loss 0.331505 Objective Loss 0.331505 LR 0.001000 Time 0.020957 +2023-10-05 20:57:26,624 - Epoch: [32][ 1130/ 1236] Overall Loss 0.331570 Objective Loss 0.331570 LR 0.001000 Time 0.020951 +2023-10-05 20:57:26,827 - Epoch: [32][ 1140/ 1236] Overall Loss 0.331673 Objective Loss 0.331673 LR 0.001000 Time 0.020945 +2023-10-05 20:57:27,030 - Epoch: [32][ 1150/ 1236] Overall Loss 0.331945 Objective Loss 0.331945 LR 0.001000 Time 0.020938 +2023-10-05 20:57:27,233 - Epoch: [32][ 1160/ 1236] Overall Loss 0.332122 Objective Loss 0.332122 LR 0.001000 Time 0.020933 +2023-10-05 20:57:27,436 - Epoch: [32][ 1170/ 1236] Overall Loss 0.332154 Objective Loss 0.332154 LR 0.001000 Time 0.020927 +2023-10-05 20:57:27,639 - Epoch: [32][ 1180/ 1236] Overall Loss 0.332425 Objective Loss 0.332425 LR 0.001000 Time 0.020922 +2023-10-05 20:57:27,842 - Epoch: [32][ 1190/ 1236] Overall Loss 0.332433 Objective Loss 0.332433 LR 0.001000 Time 0.020916 +2023-10-05 20:57:28,044 - Epoch: [32][ 1200/ 1236] Overall Loss 0.332558 Objective Loss 0.332558 LR 0.001000 Time 0.020910 +2023-10-05 20:57:28,247 - Epoch: [32][ 1210/ 1236] Overall Loss 0.332464 Objective Loss 0.332464 LR 0.001000 Time 0.020905 +2023-10-05 20:57:28,450 - Epoch: [32][ 1220/ 1236] Overall Loss 0.332495 Objective Loss 0.332495 LR 0.001000 Time 0.020900 +2023-10-05 20:57:28,707 - Epoch: [32][ 1230/ 1236] Overall Loss 0.332314 Objective Loss 0.332314 LR 0.001000 Time 0.020938 +2023-10-05 20:57:28,826 - Epoch: [32][ 1236/ 1236] Overall Loss 0.332276 Objective Loss 0.332276 Top1 83.299389 Top5 96.945010 LR 0.001000 Time 0.020933 +2023-10-05 20:57:28,950 - --- validate (epoch=32)----------- +2023-10-05 20:57:28,950 - 29943 samples (256 per mini-batch) +2023-10-05 20:57:29,398 - Epoch: [32][ 10/ 117] Loss 0.371542 Top1 81.406250 Top5 97.343750 +2023-10-05 20:57:29,548 - Epoch: [32][ 20/ 117] Loss 0.392337 Top1 80.527344 Top5 97.070312 +2023-10-05 20:57:29,698 - Epoch: [32][ 30/ 117] Loss 0.386213 Top1 80.611979 Top5 97.239583 +2023-10-05 20:57:29,848 - Epoch: [32][ 40/ 117] Loss 0.380024 Top1 80.537109 Top5 97.314453 +2023-10-05 20:57:29,998 - Epoch: [32][ 50/ 117] Loss 0.383668 Top1 80.539062 Top5 97.171875 +2023-10-05 20:57:30,151 - Epoch: [32][ 60/ 117] Loss 0.381056 Top1 80.533854 Top5 97.246094 +2023-10-05 20:57:30,306 - Epoch: [32][ 70/ 117] Loss 0.377485 Top1 80.708705 Top5 97.282366 +2023-10-05 20:57:30,462 - Epoch: [32][ 80/ 117] Loss 0.376573 Top1 80.571289 Top5 97.207031 +2023-10-05 20:57:30,617 - Epoch: [32][ 90/ 117] Loss 0.373354 Top1 80.677083 Top5 97.222222 +2023-10-05 20:57:30,772 - Epoch: [32][ 100/ 117] Loss 0.374534 Top1 80.652344 Top5 97.183594 +2023-10-05 20:57:30,935 - Epoch: [32][ 110/ 117] Loss 0.373060 Top1 80.813210 Top5 97.251420 +2023-10-05 20:57:31,020 - Epoch: [32][ 117/ 117] Loss 0.374766 Top1 80.786828 Top5 97.244765 +2023-10-05 20:57:31,175 - ==> Top1: 80.787 Top5: 97.245 Loss: 0.375 + +2023-10-05 20:57:31,176 - ==> Confusion: +[[ 910 3 3 1 9 3 0 0 6 81 1 0 0 4 4 2 6 3 1 1 12] + [ 2 1036 2 0 8 27 2 32 1 0 1 3 1 1 0 2 6 0 6 1 0] + [ 11 1 922 21 3 0 41 9 0 3 3 5 8 3 4 2 1 1 6 2 10] + [ 6 3 24 926 1 5 2 0 10 1 6 3 10 6 29 4 0 11 23 3 16] + [ 27 5 1 0 953 6 0 0 4 14 0 1 3 3 7 7 10 2 0 0 7] + [ 6 36 1 0 8 966 2 27 7 4 2 9 6 17 3 1 4 1 0 5 11] + [ 1 8 24 0 0 0 1123 7 0 0 1 5 1 0 0 6 2 1 3 6 3] + [ 5 21 13 0 3 31 7 1036 2 3 1 10 6 1 0 1 1 0 38 29 10] + [ 15 6 0 0 0 1 0 1 989 34 7 3 2 9 6 4 1 3 7 1 0] + [ 95 2 0 0 11 1 0 0 48 929 0 2 0 14 1 6 3 1 0 1 5] + [ 8 6 14 7 0 3 7 3 37 1 933 3 1 14 4 1 1 0 4 1 5] + [ 1 0 2 0 0 12 0 1 1 1 0 921 63 5 0 3 2 15 2 5 1] + [ 1 0 3 4 0 2 0 1 0 0 0 36 974 4 0 8 2 17 2 7 7] + [ 5 0 0 0 3 12 0 1 24 28 4 9 1 1020 2 2 1 1 0 2 4] + [ 15 2 2 13 11 0 0 0 57 5 3 0 4 3 961 0 0 3 13 0 9] + [ 1 3 1 1 4 1 0 0 0 0 0 11 13 2 1 1049 14 13 0 10 10] + [ 2 17 3 0 9 6 0 0 2 0 0 6 3 2 3 9 1082 0 0 5 12] + [ 0 0 0 4 0 1 1 0 2 1 0 7 28 0 0 3 0 987 1 0 3] + [ 1 7 8 14 2 1 0 42 9 0 0 1 9 1 8 0 0 0 948 9 8] + [ 0 1 5 1 1 5 8 10 1 0 1 21 10 1 0 1 7 1 1 1069 8] + [ 202 206 148 40 155 173 60 122 212 106 153 164 490 317 137 64 230 75 153 242 4456]] + +2023-10-05 20:57:31,177 - ==> Best [Top1: 81.428 Top5: 97.605 Sparsity:0.00 Params: 148928 on epoch: 31] +2023-10-05 20:57:31,177 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:57:31,183 - + +2023-10-05 20:57:31,183 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:57:32,182 - Epoch: [33][ 10/ 1236] Overall Loss 0.326256 Objective Loss 0.326256 LR 0.001000 Time 0.099880 +2023-10-05 20:57:32,387 - Epoch: [33][ 20/ 1236] Overall Loss 0.318539 Objective Loss 0.318539 LR 0.001000 Time 0.060181 +2023-10-05 20:57:32,590 - Epoch: [33][ 30/ 1236] Overall Loss 0.324584 Objective Loss 0.324584 LR 0.001000 Time 0.046856 +2023-10-05 20:57:32,795 - Epoch: [33][ 40/ 1236] Overall Loss 0.329009 Objective Loss 0.329009 LR 0.001000 Time 0.040264 +2023-10-05 20:57:32,997 - Epoch: [33][ 50/ 1236] Overall Loss 0.327791 Objective Loss 0.327791 LR 0.001000 Time 0.036249 +2023-10-05 20:57:33,203 - Epoch: [33][ 60/ 1236] Overall Loss 0.331625 Objective Loss 0.331625 LR 0.001000 Time 0.033625 +2023-10-05 20:57:33,405 - Epoch: [33][ 70/ 1236] Overall Loss 0.332637 Objective Loss 0.332637 LR 0.001000 Time 0.031708 +2023-10-05 20:57:33,611 - Epoch: [33][ 80/ 1236] Overall Loss 0.334525 Objective Loss 0.334525 LR 0.001000 Time 0.030310 +2023-10-05 20:57:33,812 - Epoch: [33][ 90/ 1236] Overall Loss 0.332816 Objective Loss 0.332816 LR 0.001000 Time 0.029175 +2023-10-05 20:57:34,015 - Epoch: [33][ 100/ 1236] Overall Loss 0.332976 Objective Loss 0.332976 LR 0.001000 Time 0.028288 +2023-10-05 20:57:34,216 - Epoch: [33][ 110/ 1236] Overall Loss 0.332231 Objective Loss 0.332231 LR 0.001000 Time 0.027539 +2023-10-05 20:57:34,418 - Epoch: [33][ 120/ 1236] Overall Loss 0.328240 Objective Loss 0.328240 LR 0.001000 Time 0.026926 +2023-10-05 20:57:34,619 - Epoch: [33][ 130/ 1236] Overall Loss 0.327263 Objective Loss 0.327263 LR 0.001000 Time 0.026397 +2023-10-05 20:57:34,821 - Epoch: [33][ 140/ 1236] Overall Loss 0.325735 Objective Loss 0.325735 LR 0.001000 Time 0.025951 +2023-10-05 20:57:35,021 - Epoch: [33][ 150/ 1236] Overall Loss 0.326765 Objective Loss 0.326765 LR 0.001000 Time 0.025554 +2023-10-05 20:57:35,224 - Epoch: [33][ 160/ 1236] Overall Loss 0.328288 Objective Loss 0.328288 LR 0.001000 Time 0.025220 +2023-10-05 20:57:35,424 - Epoch: [33][ 170/ 1236] Overall Loss 0.328820 Objective Loss 0.328820 LR 0.001000 Time 0.024914 +2023-10-05 20:57:35,626 - Epoch: [33][ 180/ 1236] Overall Loss 0.330050 Objective Loss 0.330050 LR 0.001000 Time 0.024652 +2023-10-05 20:57:35,827 - Epoch: [33][ 190/ 1236] Overall Loss 0.330404 Objective Loss 0.330404 LR 0.001000 Time 0.024408 +2023-10-05 20:57:36,028 - Epoch: [33][ 200/ 1236] Overall Loss 0.331848 Objective Loss 0.331848 LR 0.001000 Time 0.024193 +2023-10-05 20:57:36,229 - Epoch: [33][ 210/ 1236] Overall Loss 0.331110 Objective Loss 0.331110 LR 0.001000 Time 0.023995 +2023-10-05 20:57:36,431 - Epoch: [33][ 220/ 1236] Overall Loss 0.332260 Objective Loss 0.332260 LR 0.001000 Time 0.023822 +2023-10-05 20:57:36,632 - Epoch: [33][ 230/ 1236] Overall Loss 0.331359 Objective Loss 0.331359 LR 0.001000 Time 0.023658 +2023-10-05 20:57:36,834 - Epoch: [33][ 240/ 1236] Overall Loss 0.330553 Objective Loss 0.330553 LR 0.001000 Time 0.023512 +2023-10-05 20:57:37,034 - Epoch: [33][ 250/ 1236] Overall Loss 0.332041 Objective Loss 0.332041 LR 0.001000 Time 0.023371 +2023-10-05 20:57:37,237 - Epoch: [33][ 260/ 1236] Overall Loss 0.332770 Objective Loss 0.332770 LR 0.001000 Time 0.023251 +2023-10-05 20:57:37,437 - Epoch: [33][ 270/ 1236] Overall Loss 0.331998 Objective Loss 0.331998 LR 0.001000 Time 0.023130 +2023-10-05 20:57:37,640 - Epoch: [33][ 280/ 1236] Overall Loss 0.331062 Objective Loss 0.331062 LR 0.001000 Time 0.023027 +2023-10-05 20:57:37,840 - Epoch: [33][ 290/ 1236] Overall Loss 0.331479 Objective Loss 0.331479 LR 0.001000 Time 0.022922 +2023-10-05 20:57:38,043 - Epoch: [33][ 300/ 1236] Overall Loss 0.330901 Objective Loss 0.330901 LR 0.001000 Time 0.022832 +2023-10-05 20:57:38,243 - Epoch: [33][ 310/ 1236] Overall Loss 0.331383 Objective Loss 0.331383 LR 0.001000 Time 0.022741 +2023-10-05 20:57:38,446 - Epoch: [33][ 320/ 1236] Overall Loss 0.331387 Objective Loss 0.331387 LR 0.001000 Time 0.022663 +2023-10-05 20:57:38,646 - Epoch: [33][ 330/ 1236] Overall Loss 0.331871 Objective Loss 0.331871 LR 0.001000 Time 0.022582 +2023-10-05 20:57:38,852 - Epoch: [33][ 340/ 1236] Overall Loss 0.331265 Objective Loss 0.331265 LR 0.001000 Time 0.022522 +2023-10-05 20:57:39,055 - Epoch: [33][ 350/ 1236] Overall Loss 0.331681 Objective Loss 0.331681 LR 0.001000 Time 0.022457 +2023-10-05 20:57:39,260 - Epoch: [33][ 360/ 1236] Overall Loss 0.331155 Objective Loss 0.331155 LR 0.001000 Time 0.022404 +2023-10-05 20:57:39,464 - Epoch: [33][ 370/ 1236] Overall Loss 0.330694 Objective Loss 0.330694 LR 0.001000 Time 0.022347 +2023-10-05 20:57:39,670 - Epoch: [33][ 380/ 1236] Overall Loss 0.330248 Objective Loss 0.330248 LR 0.001000 Time 0.022300 +2023-10-05 20:57:39,873 - Epoch: [33][ 390/ 1236] Overall Loss 0.330423 Objective Loss 0.330423 LR 0.001000 Time 0.022247 +2023-10-05 20:57:40,078 - Epoch: [33][ 400/ 1236] Overall Loss 0.330451 Objective Loss 0.330451 LR 0.001000 Time 0.022205 +2023-10-05 20:57:40,281 - Epoch: [33][ 410/ 1236] Overall Loss 0.329664 Objective Loss 0.329664 LR 0.001000 Time 0.022157 +2023-10-05 20:57:40,487 - Epoch: [33][ 420/ 1236] Overall Loss 0.330042 Objective Loss 0.330042 LR 0.001000 Time 0.022118 +2023-10-05 20:57:40,690 - Epoch: [33][ 430/ 1236] Overall Loss 0.330398 Objective Loss 0.330398 LR 0.001000 Time 0.022075 +2023-10-05 20:57:40,896 - Epoch: [33][ 440/ 1236] Overall Loss 0.330589 Objective Loss 0.330589 LR 0.001000 Time 0.022041 +2023-10-05 20:57:41,099 - Epoch: [33][ 450/ 1236] Overall Loss 0.329485 Objective Loss 0.329485 LR 0.001000 Time 0.022001 +2023-10-05 20:57:41,305 - Epoch: [33][ 460/ 1236] Overall Loss 0.330036 Objective Loss 0.330036 LR 0.001000 Time 0.021970 +2023-10-05 20:57:41,508 - Epoch: [33][ 470/ 1236] Overall Loss 0.329346 Objective Loss 0.329346 LR 0.001000 Time 0.021935 +2023-10-05 20:57:41,715 - Epoch: [33][ 480/ 1236] Overall Loss 0.328916 Objective Loss 0.328916 LR 0.001000 Time 0.021908 +2023-10-05 20:57:41,918 - Epoch: [33][ 490/ 1236] Overall Loss 0.329572 Objective Loss 0.329572 LR 0.001000 Time 0.021875 +2023-10-05 20:57:42,125 - Epoch: [33][ 500/ 1236] Overall Loss 0.329562 Objective Loss 0.329562 LR 0.001000 Time 0.021850 +2023-10-05 20:57:42,328 - Epoch: [33][ 510/ 1236] Overall Loss 0.330703 Objective Loss 0.330703 LR 0.001000 Time 0.021819 +2023-10-05 20:57:42,534 - Epoch: [33][ 520/ 1236] Overall Loss 0.330501 Objective Loss 0.330501 LR 0.001000 Time 0.021795 +2023-10-05 20:57:42,737 - Epoch: [33][ 530/ 1236] Overall Loss 0.330545 Objective Loss 0.330545 LR 0.001000 Time 0.021767 +2023-10-05 20:57:42,944 - Epoch: [33][ 540/ 1236] Overall Loss 0.330300 Objective Loss 0.330300 LR 0.001000 Time 0.021746 +2023-10-05 20:57:43,147 - Epoch: [33][ 550/ 1236] Overall Loss 0.330448 Objective Loss 0.330448 LR 0.001000 Time 0.021719 +2023-10-05 20:57:43,353 - Epoch: [33][ 560/ 1236] Overall Loss 0.329964 Objective Loss 0.329964 LR 0.001000 Time 0.021699 +2023-10-05 20:57:43,557 - Epoch: [33][ 570/ 1236] Overall Loss 0.329766 Objective Loss 0.329766 LR 0.001000 Time 0.021675 +2023-10-05 20:57:43,763 - Epoch: [33][ 580/ 1236] Overall Loss 0.329309 Objective Loss 0.329309 LR 0.001000 Time 0.021656 +2023-10-05 20:57:43,966 - Epoch: [33][ 590/ 1236] Overall Loss 0.329170 Objective Loss 0.329170 LR 0.001000 Time 0.021633 +2023-10-05 20:57:44,172 - Epoch: [33][ 600/ 1236] Overall Loss 0.328997 Objective Loss 0.328997 LR 0.001000 Time 0.021616 +2023-10-05 20:57:44,376 - Epoch: [33][ 610/ 1236] Overall Loss 0.329490 Objective Loss 0.329490 LR 0.001000 Time 0.021594 +2023-10-05 20:57:44,582 - Epoch: [33][ 620/ 1236] Overall Loss 0.329626 Objective Loss 0.329626 LR 0.001000 Time 0.021578 +2023-10-05 20:57:44,785 - Epoch: [33][ 630/ 1236] Overall Loss 0.329359 Objective Loss 0.329359 LR 0.001000 Time 0.021558 +2023-10-05 20:57:44,992 - Epoch: [33][ 640/ 1236] Overall Loss 0.329442 Objective Loss 0.329442 LR 0.001000 Time 0.021543 +2023-10-05 20:57:45,195 - Epoch: [33][ 650/ 1236] Overall Loss 0.329553 Objective Loss 0.329553 LR 0.001000 Time 0.021523 +2023-10-05 20:57:45,402 - Epoch: [33][ 660/ 1236] Overall Loss 0.330006 Objective Loss 0.330006 LR 0.001000 Time 0.021510 +2023-10-05 20:57:45,605 - Epoch: [33][ 670/ 1236] Overall Loss 0.330412 Objective Loss 0.330412 LR 0.001000 Time 0.021492 +2023-10-05 20:57:45,812 - Epoch: [33][ 680/ 1236] Overall Loss 0.330588 Objective Loss 0.330588 LR 0.001000 Time 0.021479 +2023-10-05 20:57:46,015 - Epoch: [33][ 690/ 1236] Overall Loss 0.330323 Objective Loss 0.330323 LR 0.001000 Time 0.021462 +2023-10-05 20:57:46,221 - Epoch: [33][ 700/ 1236] Overall Loss 0.330377 Objective Loss 0.330377 LR 0.001000 Time 0.021450 +2023-10-05 20:57:46,425 - Epoch: [33][ 710/ 1236] Overall Loss 0.330124 Objective Loss 0.330124 LR 0.001000 Time 0.021434 +2023-10-05 20:57:46,631 - Epoch: [33][ 720/ 1236] Overall Loss 0.330181 Objective Loss 0.330181 LR 0.001000 Time 0.021422 +2023-10-05 20:57:46,834 - Epoch: [33][ 730/ 1236] Overall Loss 0.330226 Objective Loss 0.330226 LR 0.001000 Time 0.021407 +2023-10-05 20:57:47,041 - Epoch: [33][ 740/ 1236] Overall Loss 0.330432 Objective Loss 0.330432 LR 0.001000 Time 0.021396 +2023-10-05 20:57:47,244 - Epoch: [33][ 750/ 1236] Overall Loss 0.330005 Objective Loss 0.330005 LR 0.001000 Time 0.021381 +2023-10-05 20:57:47,451 - Epoch: [33][ 760/ 1236] Overall Loss 0.330229 Objective Loss 0.330229 LR 0.001000 Time 0.021372 +2023-10-05 20:57:47,654 - Epoch: [33][ 770/ 1236] Overall Loss 0.330258 Objective Loss 0.330258 LR 0.001000 Time 0.021358 +2023-10-05 20:57:47,861 - Epoch: [33][ 780/ 1236] Overall Loss 0.330052 Objective Loss 0.330052 LR 0.001000 Time 0.021348 +2023-10-05 20:57:48,064 - Epoch: [33][ 790/ 1236] Overall Loss 0.329855 Objective Loss 0.329855 LR 0.001000 Time 0.021335 +2023-10-05 20:57:48,271 - Epoch: [33][ 800/ 1236] Overall Loss 0.329422 Objective Loss 0.329422 LR 0.001000 Time 0.021326 +2023-10-05 20:57:48,474 - Epoch: [33][ 810/ 1236] Overall Loss 0.329414 Objective Loss 0.329414 LR 0.001000 Time 0.021313 +2023-10-05 20:57:48,680 - Epoch: [33][ 820/ 1236] Overall Loss 0.329207 Objective Loss 0.329207 LR 0.001000 Time 0.021304 +2023-10-05 20:57:48,883 - Epoch: [33][ 830/ 1236] Overall Loss 0.328988 Objective Loss 0.328988 LR 0.001000 Time 0.021292 +2023-10-05 20:57:49,090 - Epoch: [33][ 840/ 1236] Overall Loss 0.329151 Objective Loss 0.329151 LR 0.001000 Time 0.021284 +2023-10-05 20:57:49,293 - Epoch: [33][ 850/ 1236] Overall Loss 0.328923 Objective Loss 0.328923 LR 0.001000 Time 0.021273 +2023-10-05 20:57:49,499 - Epoch: [33][ 860/ 1236] Overall Loss 0.329329 Objective Loss 0.329329 LR 0.001000 Time 0.021265 +2023-10-05 20:57:49,702 - Epoch: [33][ 870/ 1236] Overall Loss 0.329586 Objective Loss 0.329586 LR 0.001000 Time 0.021253 +2023-10-05 20:57:49,909 - Epoch: [33][ 880/ 1236] Overall Loss 0.329990 Objective Loss 0.329990 LR 0.001000 Time 0.021246 +2023-10-05 20:57:50,112 - Epoch: [33][ 890/ 1236] Overall Loss 0.329919 Objective Loss 0.329919 LR 0.001000 Time 0.021236 +2023-10-05 20:57:50,319 - Epoch: [33][ 900/ 1236] Overall Loss 0.330074 Objective Loss 0.330074 LR 0.001000 Time 0.021229 +2023-10-05 20:57:50,522 - Epoch: [33][ 910/ 1236] Overall Loss 0.330178 Objective Loss 0.330178 LR 0.001000 Time 0.021219 +2023-10-05 20:57:50,729 - Epoch: [33][ 920/ 1236] Overall Loss 0.330120 Objective Loss 0.330120 LR 0.001000 Time 0.021212 +2023-10-05 20:57:50,932 - Epoch: [33][ 930/ 1236] Overall Loss 0.330469 Objective Loss 0.330469 LR 0.001000 Time 0.021202 +2023-10-05 20:57:51,139 - Epoch: [33][ 940/ 1236] Overall Loss 0.330244 Objective Loss 0.330244 LR 0.001000 Time 0.021196 +2023-10-05 20:57:51,342 - Epoch: [33][ 950/ 1236] Overall Loss 0.330545 Objective Loss 0.330545 LR 0.001000 Time 0.021187 +2023-10-05 20:57:51,549 - Epoch: [33][ 960/ 1236] Overall Loss 0.330278 Objective Loss 0.330278 LR 0.001000 Time 0.021181 +2023-10-05 20:57:51,752 - Epoch: [33][ 970/ 1236] Overall Loss 0.330678 Objective Loss 0.330678 LR 0.001000 Time 0.021172 +2023-10-05 20:57:51,959 - Epoch: [33][ 980/ 1236] Overall Loss 0.330822 Objective Loss 0.330822 LR 0.001000 Time 0.021166 +2023-10-05 20:57:52,162 - Epoch: [33][ 990/ 1236] Overall Loss 0.331081 Objective Loss 0.331081 LR 0.001000 Time 0.021158 +2023-10-05 20:57:52,369 - Epoch: [33][ 1000/ 1236] Overall Loss 0.331162 Objective Loss 0.331162 LR 0.001000 Time 0.021152 +2023-10-05 20:57:52,572 - Epoch: [33][ 1010/ 1236] Overall Loss 0.331159 Objective Loss 0.331159 LR 0.001000 Time 0.021144 +2023-10-05 20:57:52,779 - Epoch: [33][ 1020/ 1236] Overall Loss 0.331023 Objective Loss 0.331023 LR 0.001000 Time 0.021139 +2023-10-05 20:57:52,982 - Epoch: [33][ 1030/ 1236] Overall Loss 0.331106 Objective Loss 0.331106 LR 0.001000 Time 0.021131 +2023-10-05 20:57:53,189 - Epoch: [33][ 1040/ 1236] Overall Loss 0.331218 Objective Loss 0.331218 LR 0.001000 Time 0.021126 +2023-10-05 20:57:53,392 - Epoch: [33][ 1050/ 1236] Overall Loss 0.331296 Objective Loss 0.331296 LR 0.001000 Time 0.021118 +2023-10-05 20:57:53,598 - Epoch: [33][ 1060/ 1236] Overall Loss 0.331719 Objective Loss 0.331719 LR 0.001000 Time 0.021113 +2023-10-05 20:57:53,802 - Epoch: [33][ 1070/ 1236] Overall Loss 0.331698 Objective Loss 0.331698 LR 0.001000 Time 0.021106 +2023-10-05 20:57:54,008 - Epoch: [33][ 1080/ 1236] Overall Loss 0.331722 Objective Loss 0.331722 LR 0.001000 Time 0.021101 +2023-10-05 20:57:54,212 - Epoch: [33][ 1090/ 1236] Overall Loss 0.331715 Objective Loss 0.331715 LR 0.001000 Time 0.021094 +2023-10-05 20:57:54,418 - Epoch: [33][ 1100/ 1236] Overall Loss 0.331870 Objective Loss 0.331870 LR 0.001000 Time 0.021090 +2023-10-05 20:57:54,622 - Epoch: [33][ 1110/ 1236] Overall Loss 0.332033 Objective Loss 0.332033 LR 0.001000 Time 0.021083 +2023-10-05 20:57:54,828 - Epoch: [33][ 1120/ 1236] Overall Loss 0.331912 Objective Loss 0.331912 LR 0.001000 Time 0.021078 +2023-10-05 20:57:55,032 - Epoch: [33][ 1130/ 1236] Overall Loss 0.331612 Objective Loss 0.331612 LR 0.001000 Time 0.021072 +2023-10-05 20:57:55,238 - Epoch: [33][ 1140/ 1236] Overall Loss 0.331467 Objective Loss 0.331467 LR 0.001000 Time 0.021068 +2023-10-05 20:57:55,442 - Epoch: [33][ 1150/ 1236] Overall Loss 0.331433 Objective Loss 0.331433 LR 0.001000 Time 0.021061 +2023-10-05 20:57:55,648 - Epoch: [33][ 1160/ 1236] Overall Loss 0.331396 Objective Loss 0.331396 LR 0.001000 Time 0.021057 +2023-10-05 20:57:55,852 - Epoch: [33][ 1170/ 1236] Overall Loss 0.331342 Objective Loss 0.331342 LR 0.001000 Time 0.021051 +2023-10-05 20:57:56,058 - Epoch: [33][ 1180/ 1236] Overall Loss 0.331362 Objective Loss 0.331362 LR 0.001000 Time 0.021047 +2023-10-05 20:57:56,261 - Epoch: [33][ 1190/ 1236] Overall Loss 0.331624 Objective Loss 0.331624 LR 0.001000 Time 0.021041 +2023-10-05 20:57:56,468 - Epoch: [33][ 1200/ 1236] Overall Loss 0.331552 Objective Loss 0.331552 LR 0.001000 Time 0.021038 +2023-10-05 20:57:56,671 - Epoch: [33][ 1210/ 1236] Overall Loss 0.331695 Objective Loss 0.331695 LR 0.001000 Time 0.021032 +2023-10-05 20:57:56,878 - Epoch: [33][ 1220/ 1236] Overall Loss 0.331754 Objective Loss 0.331754 LR 0.001000 Time 0.021028 +2023-10-05 20:57:57,137 - Epoch: [33][ 1230/ 1236] Overall Loss 0.331918 Objective Loss 0.331918 LR 0.001000 Time 0.021067 +2023-10-05 20:57:57,256 - Epoch: [33][ 1236/ 1236] Overall Loss 0.332200 Objective Loss 0.332200 Top1 82.892057 Top5 97.148676 LR 0.001000 Time 0.021061 +2023-10-05 20:57:57,397 - --- validate (epoch=33)----------- +2023-10-05 20:57:57,397 - 29943 samples (256 per mini-batch) +2023-10-05 20:57:57,857 - Epoch: [33][ 10/ 117] Loss 0.347020 Top1 80.742188 Top5 97.500000 +2023-10-05 20:57:58,008 - Epoch: [33][ 20/ 117] Loss 0.351611 Top1 80.839844 Top5 97.558594 +2023-10-05 20:57:58,159 - Epoch: [33][ 30/ 117] Loss 0.356269 Top1 80.976562 Top5 97.526042 +2023-10-05 20:57:58,310 - Epoch: [33][ 40/ 117] Loss 0.358577 Top1 80.976562 Top5 97.480469 +2023-10-05 20:57:58,461 - Epoch: [33][ 50/ 117] Loss 0.365206 Top1 80.671875 Top5 97.492188 +2023-10-05 20:57:58,610 - Epoch: [33][ 60/ 117] Loss 0.360338 Top1 80.722656 Top5 97.408854 +2023-10-05 20:57:58,758 - Epoch: [33][ 70/ 117] Loss 0.365219 Top1 80.597098 Top5 97.382812 +2023-10-05 20:57:58,907 - Epoch: [33][ 80/ 117] Loss 0.364878 Top1 80.571289 Top5 97.358398 +2023-10-05 20:57:59,055 - Epoch: [33][ 90/ 117] Loss 0.366297 Top1 80.633681 Top5 97.404514 +2023-10-05 20:57:59,205 - Epoch: [33][ 100/ 117] Loss 0.367080 Top1 80.617188 Top5 97.453125 +2023-10-05 20:57:59,360 - Epoch: [33][ 110/ 117] Loss 0.365845 Top1 80.628551 Top5 97.514205 +2023-10-05 20:57:59,446 - Epoch: [33][ 117/ 117] Loss 0.366020 Top1 80.669940 Top5 97.518619 +2023-10-05 20:57:59,554 - ==> Top1: 80.670 Top5: 97.519 Loss: 0.366 + +2023-10-05 20:57:59,554 - ==> Confusion: +[[ 911 1 7 0 3 4 0 1 8 91 1 0 0 2 6 0 5 0 1 0 9] + [ 2 1022 2 0 0 32 2 42 2 1 0 0 0 0 1 4 6 0 8 2 5] + [ 8 0 927 10 1 1 50 14 0 2 6 1 6 6 1 3 1 3 2 2 12] + [ 6 1 19 937 0 4 3 1 5 2 11 0 13 5 27 2 2 7 28 2 14] + [ 29 8 2 0 947 8 0 1 2 19 1 1 1 1 5 3 14 2 1 1 4] + [ 5 30 1 5 4 988 1 22 7 9 3 6 3 11 1 1 4 0 5 3 7] + [ 0 6 22 0 0 1 1125 11 0 0 2 2 0 0 1 1 0 2 2 9 7] + [ 5 14 14 0 0 35 4 1072 0 4 3 6 1 0 0 3 1 0 38 13 5] + [ 20 4 1 0 0 3 0 2 972 46 6 2 1 10 12 2 0 1 7 0 0] + [ 87 2 3 0 3 1 2 0 40 938 0 2 0 21 2 4 1 0 0 4 9] + [ 5 2 15 2 0 1 4 9 19 3 953 1 0 22 3 3 1 0 5 1 4] + [ 1 0 3 0 0 15 0 2 1 1 0 925 34 2 0 4 4 20 1 17 5] + [ 3 1 6 6 0 1 0 2 2 0 0 41 928 5 2 6 2 32 2 14 15] + [ 1 0 0 0 5 12 0 0 20 20 5 4 1 1027 2 1 3 2 0 4 12] + [ 21 5 3 15 6 0 0 0 35 8 1 0 0 5 978 0 0 2 13 0 9] + [ 0 1 7 3 2 1 3 2 0 0 0 6 6 1 0 1037 25 16 1 12 11] + [ 3 17 4 0 7 4 0 0 3 0 0 3 1 0 4 7 1085 0 3 7 13] + [ 0 0 1 2 0 0 2 1 1 0 0 3 15 0 2 7 0 995 4 1 4] + [ 0 4 9 15 1 0 1 32 2 0 1 2 3 1 15 0 1 0 969 3 9] + [ 0 1 2 0 2 6 11 17 0 0 0 15 4 2 0 4 8 0 3 1069 8] + [ 221 269 219 72 62 218 71 139 207 127 166 106 385 342 149 49 155 86 206 306 4350]] + +2023-10-05 20:57:59,555 - ==> Best [Top1: 81.428 Top5: 97.605 Sparsity:0.00 Params: 148928 on epoch: 31] +2023-10-05 20:57:59,555 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:57:59,561 - + +2023-10-05 20:57:59,561 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:58:00,563 - Epoch: [34][ 10/ 1236] Overall Loss 0.317944 Objective Loss 0.317944 LR 0.001000 Time 0.100108 +2023-10-05 20:58:00,767 - Epoch: [34][ 20/ 1236] Overall Loss 0.324216 Objective Loss 0.324216 LR 0.001000 Time 0.060218 +2023-10-05 20:58:00,968 - Epoch: [34][ 30/ 1236] Overall Loss 0.312327 Objective Loss 0.312327 LR 0.001000 Time 0.046833 +2023-10-05 20:58:01,171 - Epoch: [34][ 40/ 1236] Overall Loss 0.309812 Objective Loss 0.309812 LR 0.001000 Time 0.040203 +2023-10-05 20:58:01,372 - Epoch: [34][ 50/ 1236] Overall Loss 0.303075 Objective Loss 0.303075 LR 0.001000 Time 0.036185 +2023-10-05 20:58:01,576 - Epoch: [34][ 60/ 1236] Overall Loss 0.302949 Objective Loss 0.302949 LR 0.001000 Time 0.033542 +2023-10-05 20:58:01,778 - Epoch: [34][ 70/ 1236] Overall Loss 0.301745 Objective Loss 0.301745 LR 0.001000 Time 0.031625 +2023-10-05 20:58:01,981 - Epoch: [34][ 80/ 1236] Overall Loss 0.303016 Objective Loss 0.303016 LR 0.001000 Time 0.030208 +2023-10-05 20:58:02,181 - Epoch: [34][ 90/ 1236] Overall Loss 0.308267 Objective Loss 0.308267 LR 0.001000 Time 0.029076 +2023-10-05 20:58:02,384 - Epoch: [34][ 100/ 1236] Overall Loss 0.311462 Objective Loss 0.311462 LR 0.001000 Time 0.028187 +2023-10-05 20:58:02,584 - Epoch: [34][ 110/ 1236] Overall Loss 0.311383 Objective Loss 0.311383 LR 0.001000 Time 0.027447 +2023-10-05 20:58:02,787 - Epoch: [34][ 120/ 1236] Overall Loss 0.311540 Objective Loss 0.311540 LR 0.001000 Time 0.026843 +2023-10-05 20:58:02,987 - Epoch: [34][ 130/ 1236] Overall Loss 0.312190 Objective Loss 0.312190 LR 0.001000 Time 0.026317 +2023-10-05 20:58:03,189 - Epoch: [34][ 140/ 1236] Overall Loss 0.311966 Objective Loss 0.311966 LR 0.001000 Time 0.025878 +2023-10-05 20:58:03,389 - Epoch: [34][ 150/ 1236] Overall Loss 0.313870 Objective Loss 0.313870 LR 0.001000 Time 0.025487 +2023-10-05 20:58:03,592 - Epoch: [34][ 160/ 1236] Overall Loss 0.313642 Objective Loss 0.313642 LR 0.001000 Time 0.025156 +2023-10-05 20:58:03,792 - Epoch: [34][ 170/ 1236] Overall Loss 0.313743 Objective Loss 0.313743 LR 0.001000 Time 0.024854 +2023-10-05 20:58:03,995 - Epoch: [34][ 180/ 1236] Overall Loss 0.313110 Objective Loss 0.313110 LR 0.001000 Time 0.024595 +2023-10-05 20:58:04,195 - Epoch: [34][ 190/ 1236] Overall Loss 0.313602 Objective Loss 0.313602 LR 0.001000 Time 0.024354 +2023-10-05 20:58:04,398 - Epoch: [34][ 200/ 1236] Overall Loss 0.313705 Objective Loss 0.313705 LR 0.001000 Time 0.024146 +2023-10-05 20:58:04,597 - Epoch: [34][ 210/ 1236] Overall Loss 0.316045 Objective Loss 0.316045 LR 0.001000 Time 0.023946 +2023-10-05 20:58:04,800 - Epoch: [34][ 220/ 1236] Overall Loss 0.316810 Objective Loss 0.316810 LR 0.001000 Time 0.023777 +2023-10-05 20:58:05,000 - Epoch: [34][ 230/ 1236] Overall Loss 0.317305 Objective Loss 0.317305 LR 0.001000 Time 0.023613 +2023-10-05 20:58:05,203 - Epoch: [34][ 240/ 1236] Overall Loss 0.317741 Objective Loss 0.317741 LR 0.001000 Time 0.023471 +2023-10-05 20:58:05,403 - Epoch: [34][ 250/ 1236] Overall Loss 0.319263 Objective Loss 0.319263 LR 0.001000 Time 0.023333 +2023-10-05 20:58:05,605 - Epoch: [34][ 260/ 1236] Overall Loss 0.320004 Objective Loss 0.320004 LR 0.001000 Time 0.023211 +2023-10-05 20:58:05,805 - Epoch: [34][ 270/ 1236] Overall Loss 0.320489 Objective Loss 0.320489 LR 0.001000 Time 0.023092 +2023-10-05 20:58:06,008 - Epoch: [34][ 280/ 1236] Overall Loss 0.320900 Objective Loss 0.320900 LR 0.001000 Time 0.022989 +2023-10-05 20:58:06,208 - Epoch: [34][ 290/ 1236] Overall Loss 0.321741 Objective Loss 0.321741 LR 0.001000 Time 0.022886 +2023-10-05 20:58:06,411 - Epoch: [34][ 300/ 1236] Overall Loss 0.322343 Objective Loss 0.322343 LR 0.001000 Time 0.022797 +2023-10-05 20:58:06,612 - Epoch: [34][ 310/ 1236] Overall Loss 0.322266 Objective Loss 0.322266 LR 0.001000 Time 0.022709 +2023-10-05 20:58:06,815 - Epoch: [34][ 320/ 1236] Overall Loss 0.323208 Objective Loss 0.323208 LR 0.001000 Time 0.022632 +2023-10-05 20:58:07,017 - Epoch: [34][ 330/ 1236] Overall Loss 0.324054 Objective Loss 0.324054 LR 0.001000 Time 0.022558 +2023-10-05 20:58:07,221 - Epoch: [34][ 340/ 1236] Overall Loss 0.324565 Objective Loss 0.324565 LR 0.001000 Time 0.022493 +2023-10-05 20:58:07,422 - Epoch: [34][ 350/ 1236] Overall Loss 0.324685 Objective Loss 0.324685 LR 0.001000 Time 0.022425 +2023-10-05 20:58:07,626 - Epoch: [34][ 360/ 1236] Overall Loss 0.324720 Objective Loss 0.324720 LR 0.001000 Time 0.022368 +2023-10-05 20:58:07,828 - Epoch: [34][ 370/ 1236] Overall Loss 0.325353 Objective Loss 0.325353 LR 0.001000 Time 0.022308 +2023-10-05 20:58:08,032 - Epoch: [34][ 380/ 1236] Overall Loss 0.324745 Objective Loss 0.324745 LR 0.001000 Time 0.022257 +2023-10-05 20:58:08,234 - Epoch: [34][ 390/ 1236] Overall Loss 0.324092 Objective Loss 0.324092 LR 0.001000 Time 0.022203 +2023-10-05 20:58:08,438 - Epoch: [34][ 400/ 1236] Overall Loss 0.323686 Objective Loss 0.323686 LR 0.001000 Time 0.022157 +2023-10-05 20:58:08,640 - Epoch: [34][ 410/ 1236] Overall Loss 0.323757 Objective Loss 0.323757 LR 0.001000 Time 0.022107 +2023-10-05 20:58:08,844 - Epoch: [34][ 420/ 1236] Overall Loss 0.322761 Objective Loss 0.322761 LR 0.001000 Time 0.022065 +2023-10-05 20:58:09,045 - Epoch: [34][ 430/ 1236] Overall Loss 0.322558 Objective Loss 0.322558 LR 0.001000 Time 0.022021 +2023-10-05 20:58:09,249 - Epoch: [34][ 440/ 1236] Overall Loss 0.323135 Objective Loss 0.323135 LR 0.001000 Time 0.021983 +2023-10-05 20:58:09,451 - Epoch: [34][ 450/ 1236] Overall Loss 0.323355 Objective Loss 0.323355 LR 0.001000 Time 0.021943 +2023-10-05 20:58:09,655 - Epoch: [34][ 460/ 1236] Overall Loss 0.323229 Objective Loss 0.323229 LR 0.001000 Time 0.021909 +2023-10-05 20:58:09,857 - Epoch: [34][ 470/ 1236] Overall Loss 0.324048 Objective Loss 0.324048 LR 0.001000 Time 0.021871 +2023-10-05 20:58:10,062 - Epoch: [34][ 480/ 1236] Overall Loss 0.324523 Objective Loss 0.324523 LR 0.001000 Time 0.021841 +2023-10-05 20:58:10,263 - Epoch: [34][ 490/ 1236] Overall Loss 0.324196 Objective Loss 0.324196 LR 0.001000 Time 0.021805 +2023-10-05 20:58:10,467 - Epoch: [34][ 500/ 1236] Overall Loss 0.325166 Objective Loss 0.325166 LR 0.001000 Time 0.021776 +2023-10-05 20:58:10,678 - Epoch: [34][ 510/ 1236] Overall Loss 0.324974 Objective Loss 0.324974 LR 0.001000 Time 0.021762 +2023-10-05 20:58:10,890 - Epoch: [34][ 520/ 1236] Overall Loss 0.325467 Objective Loss 0.325467 LR 0.001000 Time 0.021750 +2023-10-05 20:58:11,105 - Epoch: [34][ 530/ 1236] Overall Loss 0.325726 Objective Loss 0.325726 LR 0.001000 Time 0.021744 +2023-10-05 20:58:11,316 - Epoch: [34][ 540/ 1236] Overall Loss 0.325588 Objective Loss 0.325588 LR 0.001000 Time 0.021733 +2023-10-05 20:58:11,531 - Epoch: [34][ 550/ 1236] Overall Loss 0.325750 Objective Loss 0.325750 LR 0.001000 Time 0.021728 +2023-10-05 20:58:11,743 - Epoch: [34][ 560/ 1236] Overall Loss 0.325954 Objective Loss 0.325954 LR 0.001000 Time 0.021717 +2023-10-05 20:58:11,958 - Epoch: [34][ 570/ 1236] Overall Loss 0.325782 Objective Loss 0.325782 LR 0.001000 Time 0.021713 +2023-10-05 20:58:12,170 - Epoch: [34][ 580/ 1236] Overall Loss 0.326103 Objective Loss 0.326103 LR 0.001000 Time 0.021703 +2023-10-05 20:58:12,386 - Epoch: [34][ 590/ 1236] Overall Loss 0.326240 Objective Loss 0.326240 LR 0.001000 Time 0.021700 +2023-10-05 20:58:12,597 - Epoch: [34][ 600/ 1236] Overall Loss 0.326734 Objective Loss 0.326734 LR 0.001000 Time 0.021690 +2023-10-05 20:58:12,808 - Epoch: [34][ 610/ 1236] Overall Loss 0.327132 Objective Loss 0.327132 LR 0.001000 Time 0.021680 +2023-10-05 20:58:13,017 - Epoch: [34][ 620/ 1236] Overall Loss 0.327543 Objective Loss 0.327543 LR 0.001000 Time 0.021667 +2023-10-05 20:58:13,228 - Epoch: [34][ 630/ 1236] Overall Loss 0.327945 Objective Loss 0.327945 LR 0.001000 Time 0.021658 +2023-10-05 20:58:13,437 - Epoch: [34][ 640/ 1236] Overall Loss 0.327579 Objective Loss 0.327579 LR 0.001000 Time 0.021646 +2023-10-05 20:58:13,645 - Epoch: [34][ 650/ 1236] Overall Loss 0.327916 Objective Loss 0.327916 LR 0.001000 Time 0.021632 +2023-10-05 20:58:13,850 - Epoch: [34][ 660/ 1236] Overall Loss 0.328506 Objective Loss 0.328506 LR 0.001000 Time 0.021615 +2023-10-05 20:58:14,061 - Epoch: [34][ 670/ 1236] Overall Loss 0.328352 Objective Loss 0.328352 LR 0.001000 Time 0.021607 +2023-10-05 20:58:14,270 - Epoch: [34][ 680/ 1236] Overall Loss 0.328274 Objective Loss 0.328274 LR 0.001000 Time 0.021595 +2023-10-05 20:58:14,481 - Epoch: [34][ 690/ 1236] Overall Loss 0.328301 Objective Loss 0.328301 LR 0.001000 Time 0.021588 +2023-10-05 20:58:14,690 - Epoch: [34][ 700/ 1236] Overall Loss 0.328249 Objective Loss 0.328249 LR 0.001000 Time 0.021577 +2023-10-05 20:58:14,901 - Epoch: [34][ 710/ 1236] Overall Loss 0.328087 Objective Loss 0.328087 LR 0.001000 Time 0.021571 +2023-10-05 20:58:15,110 - Epoch: [34][ 720/ 1236] Overall Loss 0.328498 Objective Loss 0.328498 LR 0.001000 Time 0.021560 +2023-10-05 20:58:15,321 - Epoch: [34][ 730/ 1236] Overall Loss 0.328181 Objective Loss 0.328181 LR 0.001000 Time 0.021554 +2023-10-05 20:58:15,530 - Epoch: [34][ 740/ 1236] Overall Loss 0.328302 Objective Loss 0.328302 LR 0.001000 Time 0.021544 +2023-10-05 20:58:15,741 - Epoch: [34][ 750/ 1236] Overall Loss 0.328401 Objective Loss 0.328401 LR 0.001000 Time 0.021539 +2023-10-05 20:58:15,950 - Epoch: [34][ 760/ 1236] Overall Loss 0.328268 Objective Loss 0.328268 LR 0.001000 Time 0.021530 +2023-10-05 20:58:16,162 - Epoch: [34][ 770/ 1236] Overall Loss 0.328103 Objective Loss 0.328103 LR 0.001000 Time 0.021525 +2023-10-05 20:58:16,370 - Epoch: [34][ 780/ 1236] Overall Loss 0.327721 Objective Loss 0.327721 LR 0.001000 Time 0.021516 +2023-10-05 20:58:16,578 - Epoch: [34][ 790/ 1236] Overall Loss 0.328374 Objective Loss 0.328374 LR 0.001000 Time 0.021506 +2023-10-05 20:58:16,780 - Epoch: [34][ 800/ 1236] Overall Loss 0.328666 Objective Loss 0.328666 LR 0.001000 Time 0.021489 +2023-10-05 20:58:16,981 - Epoch: [34][ 810/ 1236] Overall Loss 0.328751 Objective Loss 0.328751 LR 0.001000 Time 0.021472 +2023-10-05 20:58:17,183 - Epoch: [34][ 820/ 1236] Overall Loss 0.329004 Objective Loss 0.329004 LR 0.001000 Time 0.021456 +2023-10-05 20:58:17,385 - Epoch: [34][ 830/ 1236] Overall Loss 0.329176 Objective Loss 0.329176 LR 0.001000 Time 0.021440 +2023-10-05 20:58:17,587 - Epoch: [34][ 840/ 1236] Overall Loss 0.329575 Objective Loss 0.329575 LR 0.001000 Time 0.021425 +2023-10-05 20:58:17,789 - Epoch: [34][ 850/ 1236] Overall Loss 0.329738 Objective Loss 0.329738 LR 0.001000 Time 0.021410 +2023-10-05 20:58:17,991 - Epoch: [34][ 860/ 1236] Overall Loss 0.329776 Objective Loss 0.329776 LR 0.001000 Time 0.021396 +2023-10-05 20:58:18,192 - Epoch: [34][ 870/ 1236] Overall Loss 0.329662 Objective Loss 0.329662 LR 0.001000 Time 0.021381 +2023-10-05 20:58:18,394 - Epoch: [34][ 880/ 1236] Overall Loss 0.329864 Objective Loss 0.329864 LR 0.001000 Time 0.021367 +2023-10-05 20:58:18,595 - Epoch: [34][ 890/ 1236] Overall Loss 0.329779 Objective Loss 0.329779 LR 0.001000 Time 0.021352 +2023-10-05 20:58:18,796 - Epoch: [34][ 900/ 1236] Overall Loss 0.329959 Objective Loss 0.329959 LR 0.001000 Time 0.021338 +2023-10-05 20:58:18,997 - Epoch: [34][ 910/ 1236] Overall Loss 0.329729 Objective Loss 0.329729 LR 0.001000 Time 0.021324 +2023-10-05 20:58:19,198 - Epoch: [34][ 920/ 1236] Overall Loss 0.329791 Objective Loss 0.329791 LR 0.001000 Time 0.021310 +2023-10-05 20:58:19,399 - Epoch: [34][ 930/ 1236] Overall Loss 0.329965 Objective Loss 0.329965 LR 0.001000 Time 0.021296 +2023-10-05 20:58:19,600 - Epoch: [34][ 940/ 1236] Overall Loss 0.329855 Objective Loss 0.329855 LR 0.001000 Time 0.021283 +2023-10-05 20:58:19,801 - Epoch: [34][ 950/ 1236] Overall Loss 0.330011 Objective Loss 0.330011 LR 0.001000 Time 0.021270 +2023-10-05 20:58:20,002 - Epoch: [34][ 960/ 1236] Overall Loss 0.330152 Objective Loss 0.330152 LR 0.001000 Time 0.021258 +2023-10-05 20:58:20,203 - Epoch: [34][ 970/ 1236] Overall Loss 0.330341 Objective Loss 0.330341 LR 0.001000 Time 0.021246 +2023-10-05 20:58:20,404 - Epoch: [34][ 980/ 1236] Overall Loss 0.330624 Objective Loss 0.330624 LR 0.001000 Time 0.021233 +2023-10-05 20:58:20,605 - Epoch: [34][ 990/ 1236] Overall Loss 0.330283 Objective Loss 0.330283 LR 0.001000 Time 0.021222 +2023-10-05 20:58:20,805 - Epoch: [34][ 1000/ 1236] Overall Loss 0.330471 Objective Loss 0.330471 LR 0.001000 Time 0.021210 +2023-10-05 20:58:21,007 - Epoch: [34][ 1010/ 1236] Overall Loss 0.330561 Objective Loss 0.330561 LR 0.001000 Time 0.021199 +2023-10-05 20:58:21,208 - Epoch: [34][ 1020/ 1236] Overall Loss 0.330757 Objective Loss 0.330757 LR 0.001000 Time 0.021188 +2023-10-05 20:58:21,409 - Epoch: [34][ 1030/ 1236] Overall Loss 0.330635 Objective Loss 0.330635 LR 0.001000 Time 0.021177 +2023-10-05 20:58:21,609 - Epoch: [34][ 1040/ 1236] Overall Loss 0.330784 Objective Loss 0.330784 LR 0.001000 Time 0.021166 +2023-10-05 20:58:21,811 - Epoch: [34][ 1050/ 1236] Overall Loss 0.330580 Objective Loss 0.330580 LR 0.001000 Time 0.021156 +2023-10-05 20:58:22,011 - Epoch: [34][ 1060/ 1236] Overall Loss 0.330633 Objective Loss 0.330633 LR 0.001000 Time 0.021145 +2023-10-05 20:58:22,213 - Epoch: [34][ 1070/ 1236] Overall Loss 0.330962 Objective Loss 0.330962 LR 0.001000 Time 0.021135 +2023-10-05 20:58:22,413 - Epoch: [34][ 1080/ 1236] Overall Loss 0.330994 Objective Loss 0.330994 LR 0.001000 Time 0.021125 +2023-10-05 20:58:22,615 - Epoch: [34][ 1090/ 1236] Overall Loss 0.331196 Objective Loss 0.331196 LR 0.001000 Time 0.021116 +2023-10-05 20:58:22,815 - Epoch: [34][ 1100/ 1236] Overall Loss 0.331119 Objective Loss 0.331119 LR 0.001000 Time 0.021106 +2023-10-05 20:58:23,017 - Epoch: [34][ 1110/ 1236] Overall Loss 0.331168 Objective Loss 0.331168 LR 0.001000 Time 0.021097 +2023-10-05 20:58:23,218 - Epoch: [34][ 1120/ 1236] Overall Loss 0.331103 Objective Loss 0.331103 LR 0.001000 Time 0.021086 +2023-10-05 20:58:23,419 - Epoch: [34][ 1130/ 1236] Overall Loss 0.330990 Objective Loss 0.330990 LR 0.001000 Time 0.021078 +2023-10-05 20:58:23,621 - Epoch: [34][ 1140/ 1236] Overall Loss 0.331160 Objective Loss 0.331160 LR 0.001000 Time 0.021068 +2023-10-05 20:58:23,822 - Epoch: [34][ 1150/ 1236] Overall Loss 0.330993 Objective Loss 0.330993 LR 0.001000 Time 0.021059 +2023-10-05 20:58:24,022 - Epoch: [34][ 1160/ 1236] Overall Loss 0.331128 Objective Loss 0.331128 LR 0.001000 Time 0.021049 +2023-10-05 20:58:24,224 - Epoch: [34][ 1170/ 1236] Overall Loss 0.330826 Objective Loss 0.330826 LR 0.001000 Time 0.021042 +2023-10-05 20:58:24,425 - Epoch: [34][ 1180/ 1236] Overall Loss 0.330544 Objective Loss 0.330544 LR 0.001000 Time 0.021033 +2023-10-05 20:58:24,627 - Epoch: [34][ 1190/ 1236] Overall Loss 0.330359 Objective Loss 0.330359 LR 0.001000 Time 0.021025 +2023-10-05 20:58:24,827 - Epoch: [34][ 1200/ 1236] Overall Loss 0.330394 Objective Loss 0.330394 LR 0.001000 Time 0.021016 +2023-10-05 20:58:25,029 - Epoch: [34][ 1210/ 1236] Overall Loss 0.330660 Objective Loss 0.330660 LR 0.001000 Time 0.021009 +2023-10-05 20:58:25,230 - Epoch: [34][ 1220/ 1236] Overall Loss 0.330650 Objective Loss 0.330650 LR 0.001000 Time 0.021000 +2023-10-05 20:58:25,483 - Epoch: [34][ 1230/ 1236] Overall Loss 0.330697 Objective Loss 0.330697 LR 0.001000 Time 0.021035 +2023-10-05 20:58:25,600 - Epoch: [34][ 1236/ 1236] Overall Loss 0.330756 Objective Loss 0.330756 Top1 83.503055 Top5 98.167006 LR 0.001000 Time 0.021027 +2023-10-05 20:58:25,724 - --- validate (epoch=34)----------- +2023-10-05 20:58:25,724 - 29943 samples (256 per mini-batch) +2023-10-05 20:58:26,184 - Epoch: [34][ 10/ 117] Loss 0.337085 Top1 80.859375 Top5 97.460938 +2023-10-05 20:58:26,332 - Epoch: [34][ 20/ 117] Loss 0.335167 Top1 80.898438 Top5 97.363281 +2023-10-05 20:58:26,480 - Epoch: [34][ 30/ 117] Loss 0.332067 Top1 81.184896 Top5 97.330729 +2023-10-05 20:58:26,627 - Epoch: [34][ 40/ 117] Loss 0.337128 Top1 81.259766 Top5 97.451172 +2023-10-05 20:58:26,775 - Epoch: [34][ 50/ 117] Loss 0.343824 Top1 81.164062 Top5 97.437500 +2023-10-05 20:58:26,923 - Epoch: [34][ 60/ 117] Loss 0.345588 Top1 81.100260 Top5 97.376302 +2023-10-05 20:58:27,069 - Epoch: [34][ 70/ 117] Loss 0.351721 Top1 80.853795 Top5 97.304688 +2023-10-05 20:58:27,217 - Epoch: [34][ 80/ 117] Loss 0.352111 Top1 80.878906 Top5 97.285156 +2023-10-05 20:58:27,365 - Epoch: [34][ 90/ 117] Loss 0.359851 Top1 80.516493 Top5 97.217882 +2023-10-05 20:58:27,519 - Epoch: [34][ 100/ 117] Loss 0.362493 Top1 80.492188 Top5 97.121094 +2023-10-05 20:58:27,677 - Epoch: [34][ 110/ 117] Loss 0.363083 Top1 80.472301 Top5 97.116477 +2023-10-05 20:58:27,762 - Epoch: [34][ 117/ 117] Loss 0.361321 Top1 80.536352 Top5 97.151254 +2023-10-05 20:58:27,912 - ==> Top1: 80.536 Top5: 97.151 Loss: 0.361 + +2023-10-05 20:58:27,912 - ==> Confusion: +[[ 908 6 6 2 14 1 1 0 5 66 1 0 1 2 17 2 6 2 0 0 10] + [ 1 1053 1 0 3 19 2 27 1 0 3 0 0 1 4 4 3 0 6 0 3] + [ 3 1 956 8 2 1 26 7 0 3 3 3 7 3 2 5 2 2 8 5 9] + [ 3 1 23 952 0 5 3 3 1 1 6 0 4 3 32 4 0 7 25 0 16] + [ 21 6 1 0 955 6 1 1 0 11 2 2 1 1 15 5 13 2 0 1 6] + [ 2 46 4 2 5 949 1 34 3 7 3 4 2 19 7 1 5 2 5 3 12] + [ 0 7 38 0 0 0 1114 10 0 0 1 2 1 0 1 2 0 0 2 6 7] + [ 6 15 18 0 4 23 9 1058 1 5 4 6 5 1 1 2 0 2 47 7 4] + [ 16 7 0 2 2 2 1 1 934 53 11 1 1 10 33 4 0 1 9 0 1] + [ 102 2 4 2 6 1 1 0 47 902 0 2 0 28 8 2 1 0 0 4 7] + [ 4 4 16 5 0 1 5 3 7 2 956 0 0 22 6 2 0 3 9 1 7] + [ 1 0 3 0 0 22 1 0 1 0 0 932 26 11 0 4 5 14 0 12 3] + [ 1 1 9 10 0 3 3 0 0 0 1 37 952 2 1 6 2 18 6 6 10] + [ 2 0 2 0 5 9 0 1 9 11 11 2 1 1047 8 1 0 1 0 0 9] + [ 13 2 3 15 1 0 0 0 8 7 0 0 0 0 1026 0 1 2 11 0 12] + [ 0 3 4 1 3 1 4 0 0 0 0 8 7 2 0 1033 25 15 0 16 12] + [ 0 22 3 1 8 1 1 0 1 0 0 1 2 1 4 9 1087 0 2 6 12] + [ 0 0 1 2 1 0 4 0 0 2 0 8 15 1 4 7 0 983 0 2 8] + [ 1 9 8 20 2 1 1 34 3 0 3 0 2 0 19 0 0 0 958 1 6] + [ 1 1 8 2 2 9 7 18 0 0 1 12 5 5 0 2 8 0 2 1061 8] + [ 152 304 219 103 109 158 45 133 95 127 237 131 416 304 244 51 202 77 215 284 4299]] + +2023-10-05 20:58:27,914 - ==> Best [Top1: 81.428 Top5: 97.605 Sparsity:0.00 Params: 148928 on epoch: 31] +2023-10-05 20:58:27,914 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:58:27,919 - + +2023-10-05 20:58:27,919 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:58:29,017 - Epoch: [35][ 10/ 1236] Overall Loss 0.344749 Objective Loss 0.344749 LR 0.001000 Time 0.109658 +2023-10-05 20:58:29,220 - Epoch: [35][ 20/ 1236] Overall Loss 0.330253 Objective Loss 0.330253 LR 0.001000 Time 0.064987 +2023-10-05 20:58:29,425 - Epoch: [35][ 30/ 1236] Overall Loss 0.325265 Objective Loss 0.325265 LR 0.001000 Time 0.050157 +2023-10-05 20:58:29,629 - Epoch: [35][ 40/ 1236] Overall Loss 0.318186 Objective Loss 0.318186 LR 0.001000 Time 0.042696 +2023-10-05 20:58:29,833 - Epoch: [35][ 50/ 1236] Overall Loss 0.325943 Objective Loss 0.325943 LR 0.001000 Time 0.038233 +2023-10-05 20:58:30,037 - Epoch: [35][ 60/ 1236] Overall Loss 0.329674 Objective Loss 0.329674 LR 0.001000 Time 0.035250 +2023-10-05 20:58:30,241 - Epoch: [35][ 70/ 1236] Overall Loss 0.331735 Objective Loss 0.331735 LR 0.001000 Time 0.033131 +2023-10-05 20:58:30,443 - Epoch: [35][ 80/ 1236] Overall Loss 0.328503 Objective Loss 0.328503 LR 0.001000 Time 0.031509 +2023-10-05 20:58:30,646 - Epoch: [35][ 90/ 1236] Overall Loss 0.328577 Objective Loss 0.328577 LR 0.001000 Time 0.030257 +2023-10-05 20:58:30,848 - Epoch: [35][ 100/ 1236] Overall Loss 0.324702 Objective Loss 0.324702 LR 0.001000 Time 0.029253 +2023-10-05 20:58:31,051 - Epoch: [35][ 110/ 1236] Overall Loss 0.324329 Objective Loss 0.324329 LR 0.001000 Time 0.028435 +2023-10-05 20:58:31,254 - Epoch: [35][ 120/ 1236] Overall Loss 0.323997 Objective Loss 0.323997 LR 0.001000 Time 0.027751 +2023-10-05 20:58:31,457 - Epoch: [35][ 130/ 1236] Overall Loss 0.323352 Objective Loss 0.323352 LR 0.001000 Time 0.027175 +2023-10-05 20:58:31,659 - Epoch: [35][ 140/ 1236] Overall Loss 0.321735 Objective Loss 0.321735 LR 0.001000 Time 0.026678 +2023-10-05 20:58:31,861 - Epoch: [35][ 150/ 1236] Overall Loss 0.323229 Objective Loss 0.323229 LR 0.001000 Time 0.026244 +2023-10-05 20:58:32,064 - Epoch: [35][ 160/ 1236] Overall Loss 0.323142 Objective Loss 0.323142 LR 0.001000 Time 0.025872 +2023-10-05 20:58:32,267 - Epoch: [35][ 170/ 1236] Overall Loss 0.323258 Objective Loss 0.323258 LR 0.001000 Time 0.025540 +2023-10-05 20:58:32,470 - Epoch: [35][ 180/ 1236] Overall Loss 0.322999 Objective Loss 0.322999 LR 0.001000 Time 0.025248 +2023-10-05 20:58:32,673 - Epoch: [35][ 190/ 1236] Overall Loss 0.321438 Objective Loss 0.321438 LR 0.001000 Time 0.024977 +2023-10-05 20:58:32,875 - Epoch: [35][ 200/ 1236] Overall Loss 0.321688 Objective Loss 0.321688 LR 0.001000 Time 0.024739 +2023-10-05 20:58:33,078 - Epoch: [35][ 210/ 1236] Overall Loss 0.322530 Objective Loss 0.322530 LR 0.001000 Time 0.024525 +2023-10-05 20:58:33,281 - Epoch: [35][ 220/ 1236] Overall Loss 0.322586 Objective Loss 0.322586 LR 0.001000 Time 0.024329 +2023-10-05 20:58:33,484 - Epoch: [35][ 230/ 1236] Overall Loss 0.322648 Objective Loss 0.322648 LR 0.001000 Time 0.024154 +2023-10-05 20:58:33,687 - Epoch: [35][ 240/ 1236] Overall Loss 0.321796 Objective Loss 0.321796 LR 0.001000 Time 0.023991 +2023-10-05 20:58:33,889 - Epoch: [35][ 250/ 1236] Overall Loss 0.321467 Objective Loss 0.321467 LR 0.001000 Time 0.023841 +2023-10-05 20:58:34,092 - Epoch: [35][ 260/ 1236] Overall Loss 0.321255 Objective Loss 0.321255 LR 0.001000 Time 0.023702 +2023-10-05 20:58:34,294 - Epoch: [35][ 270/ 1236] Overall Loss 0.320789 Objective Loss 0.320789 LR 0.001000 Time 0.023573 +2023-10-05 20:58:34,497 - Epoch: [35][ 280/ 1236] Overall Loss 0.321914 Objective Loss 0.321914 LR 0.001000 Time 0.023454 +2023-10-05 20:58:34,700 - Epoch: [35][ 290/ 1236] Overall Loss 0.322464 Objective Loss 0.322464 LR 0.001000 Time 0.023340 +2023-10-05 20:58:34,903 - Epoch: [35][ 300/ 1236] Overall Loss 0.322586 Objective Loss 0.322586 LR 0.001000 Time 0.023236 +2023-10-05 20:58:35,105 - Epoch: [35][ 310/ 1236] Overall Loss 0.321793 Objective Loss 0.321793 LR 0.001000 Time 0.023139 +2023-10-05 20:58:35,307 - Epoch: [35][ 320/ 1236] Overall Loss 0.321278 Objective Loss 0.321278 LR 0.001000 Time 0.023045 +2023-10-05 20:58:35,508 - Epoch: [35][ 330/ 1236] Overall Loss 0.320788 Objective Loss 0.320788 LR 0.001000 Time 0.022954 +2023-10-05 20:58:35,710 - Epoch: [35][ 340/ 1236] Overall Loss 0.319846 Objective Loss 0.319846 LR 0.001000 Time 0.022872 +2023-10-05 20:58:35,910 - Epoch: [35][ 350/ 1236] Overall Loss 0.320468 Objective Loss 0.320468 LR 0.001000 Time 0.022791 +2023-10-05 20:58:36,113 - Epoch: [35][ 360/ 1236] Overall Loss 0.320585 Objective Loss 0.320585 LR 0.001000 Time 0.022718 +2023-10-05 20:58:36,313 - Epoch: [35][ 370/ 1236] Overall Loss 0.321301 Objective Loss 0.321301 LR 0.001000 Time 0.022646 +2023-10-05 20:58:36,516 - Epoch: [35][ 380/ 1236] Overall Loss 0.322290 Objective Loss 0.322290 LR 0.001000 Time 0.022581 +2023-10-05 20:58:36,717 - Epoch: [35][ 390/ 1236] Overall Loss 0.322376 Objective Loss 0.322376 LR 0.001000 Time 0.022516 +2023-10-05 20:58:36,919 - Epoch: [35][ 400/ 1236] Overall Loss 0.322263 Objective Loss 0.322263 LR 0.001000 Time 0.022458 +2023-10-05 20:58:37,119 - Epoch: [35][ 410/ 1236] Overall Loss 0.322529 Objective Loss 0.322529 LR 0.001000 Time 0.022398 +2023-10-05 20:58:37,322 - Epoch: [35][ 420/ 1236] Overall Loss 0.322732 Objective Loss 0.322732 LR 0.001000 Time 0.022345 +2023-10-05 20:58:37,522 - Epoch: [35][ 430/ 1236] Overall Loss 0.322505 Objective Loss 0.322505 LR 0.001000 Time 0.022290 +2023-10-05 20:58:37,724 - Epoch: [35][ 440/ 1236] Overall Loss 0.322817 Objective Loss 0.322817 LR 0.001000 Time 0.022243 +2023-10-05 20:58:37,925 - Epoch: [35][ 450/ 1236] Overall Loss 0.323311 Objective Loss 0.323311 LR 0.001000 Time 0.022193 +2023-10-05 20:58:38,127 - Epoch: [35][ 460/ 1236] Overall Loss 0.324032 Objective Loss 0.324032 LR 0.001000 Time 0.022149 +2023-10-05 20:58:38,327 - Epoch: [35][ 470/ 1236] Overall Loss 0.324383 Objective Loss 0.324383 LR 0.001000 Time 0.022103 +2023-10-05 20:58:38,529 - Epoch: [35][ 480/ 1236] Overall Loss 0.324877 Objective Loss 0.324877 LR 0.001000 Time 0.022063 +2023-10-05 20:58:38,730 - Epoch: [35][ 490/ 1236] Overall Loss 0.324622 Objective Loss 0.324622 LR 0.001000 Time 0.022022 +2023-10-05 20:58:38,932 - Epoch: [35][ 500/ 1236] Overall Loss 0.325101 Objective Loss 0.325101 LR 0.001000 Time 0.021985 +2023-10-05 20:58:39,132 - Epoch: [35][ 510/ 1236] Overall Loss 0.325575 Objective Loss 0.325575 LR 0.001000 Time 0.021947 +2023-10-05 20:58:39,335 - Epoch: [35][ 520/ 1236] Overall Loss 0.325452 Objective Loss 0.325452 LR 0.001000 Time 0.021913 +2023-10-05 20:58:39,535 - Epoch: [35][ 530/ 1236] Overall Loss 0.325254 Objective Loss 0.325254 LR 0.001000 Time 0.021877 +2023-10-05 20:58:39,737 - Epoch: [35][ 540/ 1236] Overall Loss 0.325500 Objective Loss 0.325500 LR 0.001000 Time 0.021845 +2023-10-05 20:58:39,937 - Epoch: [35][ 550/ 1236] Overall Loss 0.325363 Objective Loss 0.325363 LR 0.001000 Time 0.021812 +2023-10-05 20:58:40,140 - Epoch: [35][ 560/ 1236] Overall Loss 0.325370 Objective Loss 0.325370 LR 0.001000 Time 0.021782 +2023-10-05 20:58:40,340 - Epoch: [35][ 570/ 1236] Overall Loss 0.325579 Objective Loss 0.325579 LR 0.001000 Time 0.021751 +2023-10-05 20:58:40,542 - Epoch: [35][ 580/ 1236] Overall Loss 0.325868 Objective Loss 0.325868 LR 0.001000 Time 0.021724 +2023-10-05 20:58:40,743 - Epoch: [35][ 590/ 1236] Overall Loss 0.325782 Objective Loss 0.325782 LR 0.001000 Time 0.021695 +2023-10-05 20:58:40,945 - Epoch: [35][ 600/ 1236] Overall Loss 0.325604 Objective Loss 0.325604 LR 0.001000 Time 0.021670 +2023-10-05 20:58:41,145 - Epoch: [35][ 610/ 1236] Overall Loss 0.325187 Objective Loss 0.325187 LR 0.001000 Time 0.021642 +2023-10-05 20:58:41,347 - Epoch: [35][ 620/ 1236] Overall Loss 0.324975 Objective Loss 0.324975 LR 0.001000 Time 0.021619 +2023-10-05 20:58:41,548 - Epoch: [35][ 630/ 1236] Overall Loss 0.325140 Objective Loss 0.325140 LR 0.001000 Time 0.021593 +2023-10-05 20:58:41,750 - Epoch: [35][ 640/ 1236] Overall Loss 0.325079 Objective Loss 0.325079 LR 0.001000 Time 0.021571 +2023-10-05 20:58:41,950 - Epoch: [35][ 650/ 1236] Overall Loss 0.324413 Objective Loss 0.324413 LR 0.001000 Time 0.021547 +2023-10-05 20:58:42,152 - Epoch: [35][ 660/ 1236] Overall Loss 0.324963 Objective Loss 0.324963 LR 0.001000 Time 0.021526 +2023-10-05 20:58:42,353 - Epoch: [35][ 670/ 1236] Overall Loss 0.324926 Objective Loss 0.324926 LR 0.001000 Time 0.021503 +2023-10-05 20:58:42,555 - Epoch: [35][ 680/ 1236] Overall Loss 0.325101 Objective Loss 0.325101 LR 0.001000 Time 0.021484 +2023-10-05 20:58:42,755 - Epoch: [35][ 690/ 1236] Overall Loss 0.325233 Objective Loss 0.325233 LR 0.001000 Time 0.021463 +2023-10-05 20:58:42,957 - Epoch: [35][ 700/ 1236] Overall Loss 0.325265 Objective Loss 0.325265 LR 0.001000 Time 0.021444 +2023-10-05 20:58:43,157 - Epoch: [35][ 710/ 1236] Overall Loss 0.325055 Objective Loss 0.325055 LR 0.001000 Time 0.021424 +2023-10-05 20:58:43,360 - Epoch: [35][ 720/ 1236] Overall Loss 0.326074 Objective Loss 0.326074 LR 0.001000 Time 0.021407 +2023-10-05 20:58:43,560 - Epoch: [35][ 730/ 1236] Overall Loss 0.326390 Objective Loss 0.326390 LR 0.001000 Time 0.021387 +2023-10-05 20:58:43,762 - Epoch: [35][ 740/ 1236] Overall Loss 0.326531 Objective Loss 0.326531 LR 0.001000 Time 0.021371 +2023-10-05 20:58:43,963 - Epoch: [35][ 750/ 1236] Overall Loss 0.326645 Objective Loss 0.326645 LR 0.001000 Time 0.021353 +2023-10-05 20:58:44,165 - Epoch: [35][ 760/ 1236] Overall Loss 0.326708 Objective Loss 0.326708 LR 0.001000 Time 0.021337 +2023-10-05 20:58:44,365 - Epoch: [35][ 770/ 1236] Overall Loss 0.326928 Objective Loss 0.326928 LR 0.001000 Time 0.021320 +2023-10-05 20:58:44,567 - Epoch: [35][ 780/ 1236] Overall Loss 0.327356 Objective Loss 0.327356 LR 0.001000 Time 0.021305 +2023-10-05 20:58:44,768 - Epoch: [35][ 790/ 1236] Overall Loss 0.327450 Objective Loss 0.327450 LR 0.001000 Time 0.021289 +2023-10-05 20:58:44,970 - Epoch: [35][ 800/ 1236] Overall Loss 0.327360 Objective Loss 0.327360 LR 0.001000 Time 0.021275 +2023-10-05 20:58:45,170 - Epoch: [35][ 810/ 1236] Overall Loss 0.327196 Objective Loss 0.327196 LR 0.001000 Time 0.021259 +2023-10-05 20:58:45,372 - Epoch: [35][ 820/ 1236] Overall Loss 0.327288 Objective Loss 0.327288 LR 0.001000 Time 0.021246 +2023-10-05 20:58:45,573 - Epoch: [35][ 830/ 1236] Overall Loss 0.327357 Objective Loss 0.327357 LR 0.001000 Time 0.021231 +2023-10-05 20:58:45,775 - Epoch: [35][ 840/ 1236] Overall Loss 0.327541 Objective Loss 0.327541 LR 0.001000 Time 0.021219 +2023-10-05 20:58:45,975 - Epoch: [35][ 850/ 1236] Overall Loss 0.327535 Objective Loss 0.327535 LR 0.001000 Time 0.021204 +2023-10-05 20:58:46,177 - Epoch: [35][ 860/ 1236] Overall Loss 0.327454 Objective Loss 0.327454 LR 0.001000 Time 0.021193 +2023-10-05 20:58:46,378 - Epoch: [35][ 870/ 1236] Overall Loss 0.327639 Objective Loss 0.327639 LR 0.001000 Time 0.021179 +2023-10-05 20:58:46,580 - Epoch: [35][ 880/ 1236] Overall Loss 0.327886 Objective Loss 0.327886 LR 0.001000 Time 0.021167 +2023-10-05 20:58:46,780 - Epoch: [35][ 890/ 1236] Overall Loss 0.327949 Objective Loss 0.327949 LR 0.001000 Time 0.021155 +2023-10-05 20:58:46,983 - Epoch: [35][ 900/ 1236] Overall Loss 0.328158 Objective Loss 0.328158 LR 0.001000 Time 0.021144 +2023-10-05 20:58:47,183 - Epoch: [35][ 910/ 1236] Overall Loss 0.327963 Objective Loss 0.327963 LR 0.001000 Time 0.021132 +2023-10-05 20:58:47,385 - Epoch: [35][ 920/ 1236] Overall Loss 0.328275 Objective Loss 0.328275 LR 0.001000 Time 0.021121 +2023-10-05 20:58:47,586 - Epoch: [35][ 930/ 1236] Overall Loss 0.328059 Objective Loss 0.328059 LR 0.001000 Time 0.021109 +2023-10-05 20:58:47,788 - Epoch: [35][ 940/ 1236] Overall Loss 0.328203 Objective Loss 0.328203 LR 0.001000 Time 0.021100 +2023-10-05 20:58:47,989 - Epoch: [35][ 950/ 1236] Overall Loss 0.328059 Objective Loss 0.328059 LR 0.001000 Time 0.021088 +2023-10-05 20:58:48,191 - Epoch: [35][ 960/ 1236] Overall Loss 0.327954 Objective Loss 0.327954 LR 0.001000 Time 0.021079 +2023-10-05 20:58:48,391 - Epoch: [35][ 970/ 1236] Overall Loss 0.327910 Objective Loss 0.327910 LR 0.001000 Time 0.021068 +2023-10-05 20:58:48,593 - Epoch: [35][ 980/ 1236] Overall Loss 0.328097 Objective Loss 0.328097 LR 0.001000 Time 0.021059 +2023-10-05 20:58:48,794 - Epoch: [35][ 990/ 1236] Overall Loss 0.328070 Objective Loss 0.328070 LR 0.001000 Time 0.021048 +2023-10-05 20:58:48,996 - Epoch: [35][ 1000/ 1236] Overall Loss 0.327994 Objective Loss 0.327994 LR 0.001000 Time 0.021040 +2023-10-05 20:58:49,197 - Epoch: [35][ 1010/ 1236] Overall Loss 0.328060 Objective Loss 0.328060 LR 0.001000 Time 0.021029 +2023-10-05 20:58:49,399 - Epoch: [35][ 1020/ 1236] Overall Loss 0.328075 Objective Loss 0.328075 LR 0.001000 Time 0.021021 +2023-10-05 20:58:49,599 - Epoch: [35][ 1030/ 1236] Overall Loss 0.328101 Objective Loss 0.328101 LR 0.001000 Time 0.021011 +2023-10-05 20:58:49,802 - Epoch: [35][ 1040/ 1236] Overall Loss 0.328097 Objective Loss 0.328097 LR 0.001000 Time 0.021003 +2023-10-05 20:58:50,002 - Epoch: [35][ 1050/ 1236] Overall Loss 0.328166 Objective Loss 0.328166 LR 0.001000 Time 0.020994 +2023-10-05 20:58:50,204 - Epoch: [35][ 1060/ 1236] Overall Loss 0.328524 Objective Loss 0.328524 LR 0.001000 Time 0.020986 +2023-10-05 20:58:50,405 - Epoch: [35][ 1070/ 1236] Overall Loss 0.328522 Objective Loss 0.328522 LR 0.001000 Time 0.020977 +2023-10-05 20:58:50,607 - Epoch: [35][ 1080/ 1236] Overall Loss 0.328443 Objective Loss 0.328443 LR 0.001000 Time 0.020970 +2023-10-05 20:58:50,807 - Epoch: [35][ 1090/ 1236] Overall Loss 0.328499 Objective Loss 0.328499 LR 0.001000 Time 0.020961 +2023-10-05 20:58:51,009 - Epoch: [35][ 1100/ 1236] Overall Loss 0.328590 Objective Loss 0.328590 LR 0.001000 Time 0.020954 +2023-10-05 20:58:51,210 - Epoch: [35][ 1110/ 1236] Overall Loss 0.328791 Objective Loss 0.328791 LR 0.001000 Time 0.020945 +2023-10-05 20:58:51,412 - Epoch: [35][ 1120/ 1236] Overall Loss 0.328793 Objective Loss 0.328793 LR 0.001000 Time 0.020938 +2023-10-05 20:58:51,612 - Epoch: [35][ 1130/ 1236] Overall Loss 0.329200 Objective Loss 0.329200 LR 0.001000 Time 0.020930 +2023-10-05 20:58:51,814 - Epoch: [35][ 1140/ 1236] Overall Loss 0.329236 Objective Loss 0.329236 LR 0.001000 Time 0.020923 +2023-10-05 20:58:52,015 - Epoch: [35][ 1150/ 1236] Overall Loss 0.329373 Objective Loss 0.329373 LR 0.001000 Time 0.020915 +2023-10-05 20:58:52,217 - Epoch: [35][ 1160/ 1236] Overall Loss 0.329423 Objective Loss 0.329423 LR 0.001000 Time 0.020909 +2023-10-05 20:58:52,417 - Epoch: [35][ 1170/ 1236] Overall Loss 0.330153 Objective Loss 0.330153 LR 0.001000 Time 0.020901 +2023-10-05 20:58:52,619 - Epoch: [35][ 1180/ 1236] Overall Loss 0.330357 Objective Loss 0.330357 LR 0.001000 Time 0.020895 +2023-10-05 20:58:52,820 - Epoch: [35][ 1190/ 1236] Overall Loss 0.330245 Objective Loss 0.330245 LR 0.001000 Time 0.020888 +2023-10-05 20:58:53,022 - Epoch: [35][ 1200/ 1236] Overall Loss 0.330352 Objective Loss 0.330352 LR 0.001000 Time 0.020882 +2023-10-05 20:58:53,223 - Epoch: [35][ 1210/ 1236] Overall Loss 0.330229 Objective Loss 0.330229 LR 0.001000 Time 0.020875 +2023-10-05 20:58:53,425 - Epoch: [35][ 1220/ 1236] Overall Loss 0.330499 Objective Loss 0.330499 LR 0.001000 Time 0.020869 +2023-10-05 20:58:53,679 - Epoch: [35][ 1230/ 1236] Overall Loss 0.330589 Objective Loss 0.330589 LR 0.001000 Time 0.020906 +2023-10-05 20:58:53,796 - Epoch: [35][ 1236/ 1236] Overall Loss 0.330637 Objective Loss 0.330637 Top1 83.299389 Top5 97.148676 LR 0.001000 Time 0.020899 +2023-10-05 20:58:53,932 - --- validate (epoch=35)----------- +2023-10-05 20:58:53,932 - 29943 samples (256 per mini-batch) +2023-10-05 20:58:54,382 - Epoch: [35][ 10/ 117] Loss 0.379885 Top1 80.312500 Top5 96.796875 +2023-10-05 20:58:54,531 - Epoch: [35][ 20/ 117] Loss 0.355570 Top1 80.996094 Top5 96.835938 +2023-10-05 20:58:54,679 - Epoch: [35][ 30/ 117] Loss 0.351763 Top1 80.755208 Top5 96.992188 +2023-10-05 20:58:54,827 - Epoch: [35][ 40/ 117] Loss 0.356612 Top1 80.878906 Top5 96.962891 +2023-10-05 20:58:54,974 - Epoch: [35][ 50/ 117] Loss 0.354871 Top1 80.976562 Top5 97.085938 +2023-10-05 20:58:55,122 - Epoch: [35][ 60/ 117] Loss 0.356822 Top1 80.963542 Top5 97.128906 +2023-10-05 20:58:55,269 - Epoch: [35][ 70/ 117] Loss 0.356748 Top1 81.032366 Top5 97.187500 +2023-10-05 20:58:55,424 - Epoch: [35][ 80/ 117] Loss 0.359430 Top1 80.991211 Top5 97.163086 +2023-10-05 20:58:55,579 - Epoch: [35][ 90/ 117] Loss 0.359622 Top1 80.902778 Top5 97.126736 +2023-10-05 20:58:55,735 - Epoch: [35][ 100/ 117] Loss 0.361557 Top1 80.937500 Top5 97.089844 +2023-10-05 20:58:55,897 - Epoch: [35][ 110/ 117] Loss 0.364895 Top1 80.898438 Top5 97.038352 +2023-10-05 20:58:55,982 - Epoch: [35][ 117/ 117] Loss 0.362778 Top1 80.910396 Top5 97.031026 +2023-10-05 20:58:56,087 - ==> Top1: 80.910 Top5: 97.031 Loss: 0.363 + +2023-10-05 20:58:56,088 - ==> Confusion: +[[ 905 2 4 3 17 1 0 1 6 79 1 0 1 6 5 4 3 0 3 0 9] + [ 0 1051 2 2 7 14 1 24 0 0 6 3 0 1 0 3 5 0 7 0 5] + [ 8 1 924 11 4 1 35 16 0 4 8 4 6 4 2 4 3 2 10 0 9] + [ 4 4 19 942 2 4 1 1 5 1 10 0 4 0 33 4 1 8 28 3 15] + [ 18 5 0 0 975 4 0 2 1 10 2 3 1 1 8 4 10 2 2 0 2] + [ 5 49 0 0 5 922 1 34 2 3 5 19 4 28 4 0 5 1 4 8 17] + [ 0 6 22 0 0 0 1113 7 0 0 8 3 4 0 0 12 0 1 4 5 6] + [ 5 15 9 0 2 26 2 1064 0 5 6 9 3 1 1 1 0 1 54 7 7] + [ 17 4 1 2 0 2 0 2 955 44 12 1 2 24 13 2 1 0 5 1 1] + [ 102 2 0 1 3 2 0 1 33 927 1 1 0 25 3 8 0 0 1 2 7] + [ 3 1 13 6 0 1 3 3 13 0 974 2 3 8 4 1 0 0 12 1 5] + [ 1 0 0 1 2 8 0 1 0 0 0 958 28 5 0 4 2 17 1 5 2] + [ 1 0 6 10 0 3 0 1 0 2 2 48 955 1 2 5 1 17 2 1 11] + [ 1 0 3 1 4 5 0 1 7 22 9 8 3 1038 3 0 2 2 0 3 7] + [ 13 2 3 15 4 0 0 0 32 9 6 0 2 1 986 0 0 4 11 0 13] + [ 0 3 2 1 3 1 2 0 0 0 0 12 5 1 1 1061 10 20 2 6 4] + [ 4 10 0 1 9 3 1 1 2 0 1 5 2 1 1 14 1086 0 0 5 15] + [ 0 0 0 4 0 0 1 0 1 0 0 8 12 0 1 9 0 997 2 0 3] + [ 1 7 9 16 1 0 0 28 6 0 4 2 0 0 12 0 1 0 968 1 12] + [ 0 4 3 1 3 3 12 14 0 1 4 20 5 0 0 10 7 0 3 1050 12] + [ 204 304 167 78 143 142 63 117 94 146 210 197 397 360 139 95 163 92 180 238 4376]] + +2023-10-05 20:58:56,089 - ==> Best [Top1: 81.428 Top5: 97.605 Sparsity:0.00 Params: 148928 on epoch: 31] +2023-10-05 20:58:56,089 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:58:56,095 - + +2023-10-05 20:58:56,095 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:58:57,073 - Epoch: [36][ 10/ 1236] Overall Loss 0.360137 Objective Loss 0.360137 LR 0.001000 Time 0.097724 +2023-10-05 20:58:57,278 - Epoch: [36][ 20/ 1236] Overall Loss 0.340365 Objective Loss 0.340365 LR 0.001000 Time 0.059088 +2023-10-05 20:58:57,481 - Epoch: [36][ 30/ 1236] Overall Loss 0.327068 Objective Loss 0.327068 LR 0.001000 Time 0.046153 +2023-10-05 20:58:57,686 - Epoch: [36][ 40/ 1236] Overall Loss 0.330747 Objective Loss 0.330747 LR 0.001000 Time 0.039728 +2023-10-05 20:58:57,889 - Epoch: [36][ 50/ 1236] Overall Loss 0.330066 Objective Loss 0.330066 LR 0.001000 Time 0.035844 +2023-10-05 20:58:58,094 - Epoch: [36][ 60/ 1236] Overall Loss 0.327682 Objective Loss 0.327682 LR 0.001000 Time 0.033276 +2023-10-05 20:58:58,297 - Epoch: [36][ 70/ 1236] Overall Loss 0.322135 Objective Loss 0.322135 LR 0.001000 Time 0.031423 +2023-10-05 20:58:58,501 - Epoch: [36][ 80/ 1236] Overall Loss 0.321290 Objective Loss 0.321290 LR 0.001000 Time 0.030040 +2023-10-05 20:58:58,704 - Epoch: [36][ 90/ 1236] Overall Loss 0.319760 Objective Loss 0.319760 LR 0.001000 Time 0.028953 +2023-10-05 20:58:58,909 - Epoch: [36][ 100/ 1236] Overall Loss 0.322792 Objective Loss 0.322792 LR 0.001000 Time 0.028101 +2023-10-05 20:58:59,112 - Epoch: [36][ 110/ 1236] Overall Loss 0.319485 Objective Loss 0.319485 LR 0.001000 Time 0.027376 +2023-10-05 20:58:59,316 - Epoch: [36][ 120/ 1236] Overall Loss 0.319890 Objective Loss 0.319890 LR 0.001000 Time 0.026798 +2023-10-05 20:58:59,521 - Epoch: [36][ 130/ 1236] Overall Loss 0.320240 Objective Loss 0.320240 LR 0.001000 Time 0.026306 +2023-10-05 20:58:59,726 - Epoch: [36][ 140/ 1236] Overall Loss 0.321484 Objective Loss 0.321484 LR 0.001000 Time 0.025890 +2023-10-05 20:58:59,931 - Epoch: [36][ 150/ 1236] Overall Loss 0.321960 Objective Loss 0.321960 LR 0.001000 Time 0.025525 +2023-10-05 20:59:00,136 - Epoch: [36][ 160/ 1236] Overall Loss 0.321163 Objective Loss 0.321163 LR 0.001000 Time 0.025211 +2023-10-05 20:59:00,338 - Epoch: [36][ 170/ 1236] Overall Loss 0.320527 Objective Loss 0.320527 LR 0.001000 Time 0.024913 +2023-10-05 20:59:00,538 - Epoch: [36][ 180/ 1236] Overall Loss 0.323461 Objective Loss 0.323461 LR 0.001000 Time 0.024640 +2023-10-05 20:59:00,738 - Epoch: [36][ 190/ 1236] Overall Loss 0.323412 Objective Loss 0.323412 LR 0.001000 Time 0.024391 +2023-10-05 20:59:00,939 - Epoch: [36][ 200/ 1236] Overall Loss 0.322150 Objective Loss 0.322150 LR 0.001000 Time 0.024176 +2023-10-05 20:59:01,139 - Epoch: [36][ 210/ 1236] Overall Loss 0.321153 Objective Loss 0.321153 LR 0.001000 Time 0.023976 +2023-10-05 20:59:01,339 - Epoch: [36][ 220/ 1236] Overall Loss 0.321675 Objective Loss 0.321675 LR 0.001000 Time 0.023795 +2023-10-05 20:59:01,539 - Epoch: [36][ 230/ 1236] Overall Loss 0.321104 Objective Loss 0.321104 LR 0.001000 Time 0.023626 +2023-10-05 20:59:01,740 - Epoch: [36][ 240/ 1236] Overall Loss 0.320053 Objective Loss 0.320053 LR 0.001000 Time 0.023477 +2023-10-05 20:59:01,940 - Epoch: [36][ 250/ 1236] Overall Loss 0.318790 Objective Loss 0.318790 LR 0.001000 Time 0.023337 +2023-10-05 20:59:02,140 - Epoch: [36][ 260/ 1236] Overall Loss 0.319483 Objective Loss 0.319483 LR 0.001000 Time 0.023210 +2023-10-05 20:59:02,341 - Epoch: [36][ 270/ 1236] Overall Loss 0.317982 Objective Loss 0.317982 LR 0.001000 Time 0.023091 +2023-10-05 20:59:02,541 - Epoch: [36][ 280/ 1236] Overall Loss 0.319284 Objective Loss 0.319284 LR 0.001000 Time 0.022980 +2023-10-05 20:59:02,740 - Epoch: [36][ 290/ 1236] Overall Loss 0.319056 Objective Loss 0.319056 LR 0.001000 Time 0.022872 +2023-10-05 20:59:02,940 - Epoch: [36][ 300/ 1236] Overall Loss 0.318663 Objective Loss 0.318663 LR 0.001000 Time 0.022776 +2023-10-05 20:59:03,140 - Epoch: [36][ 310/ 1236] Overall Loss 0.318914 Objective Loss 0.318914 LR 0.001000 Time 0.022684 +2023-10-05 20:59:03,340 - Epoch: [36][ 320/ 1236] Overall Loss 0.319012 Objective Loss 0.319012 LR 0.001000 Time 0.022600 +2023-10-05 20:59:03,540 - Epoch: [36][ 330/ 1236] Overall Loss 0.319292 Objective Loss 0.319292 LR 0.001000 Time 0.022520 +2023-10-05 20:59:03,742 - Epoch: [36][ 340/ 1236] Overall Loss 0.319019 Objective Loss 0.319019 LR 0.001000 Time 0.022450 +2023-10-05 20:59:03,942 - Epoch: [36][ 350/ 1236] Overall Loss 0.319321 Objective Loss 0.319321 LR 0.001000 Time 0.022379 +2023-10-05 20:59:04,143 - Epoch: [36][ 360/ 1236] Overall Loss 0.318006 Objective Loss 0.318006 LR 0.001000 Time 0.022316 +2023-10-05 20:59:04,343 - Epoch: [36][ 370/ 1236] Overall Loss 0.318353 Objective Loss 0.318353 LR 0.001000 Time 0.022252 +2023-10-05 20:59:04,545 - Epoch: [36][ 380/ 1236] Overall Loss 0.318373 Objective Loss 0.318373 LR 0.001000 Time 0.022198 +2023-10-05 20:59:04,745 - Epoch: [36][ 390/ 1236] Overall Loss 0.318850 Objective Loss 0.318850 LR 0.001000 Time 0.022140 +2023-10-05 20:59:04,947 - Epoch: [36][ 400/ 1236] Overall Loss 0.318503 Objective Loss 0.318503 LR 0.001000 Time 0.022090 +2023-10-05 20:59:05,147 - Epoch: [36][ 410/ 1236] Overall Loss 0.319220 Objective Loss 0.319220 LR 0.001000 Time 0.022039 +2023-10-05 20:59:05,356 - Epoch: [36][ 420/ 1236] Overall Loss 0.319065 Objective Loss 0.319065 LR 0.001000 Time 0.022010 +2023-10-05 20:59:05,563 - Epoch: [36][ 430/ 1236] Overall Loss 0.319442 Objective Loss 0.319442 LR 0.001000 Time 0.021978 +2023-10-05 20:59:05,775 - Epoch: [36][ 440/ 1236] Overall Loss 0.320006 Objective Loss 0.320006 LR 0.001000 Time 0.021960 +2023-10-05 20:59:05,981 - Epoch: [36][ 450/ 1236] Overall Loss 0.320208 Objective Loss 0.320208 LR 0.001000 Time 0.021929 +2023-10-05 20:59:06,193 - Epoch: [36][ 460/ 1236] Overall Loss 0.320358 Objective Loss 0.320358 LR 0.001000 Time 0.021913 +2023-10-05 20:59:06,399 - Epoch: [36][ 470/ 1236] Overall Loss 0.320653 Objective Loss 0.320653 LR 0.001000 Time 0.021885 +2023-10-05 20:59:06,611 - Epoch: [36][ 480/ 1236] Overall Loss 0.320216 Objective Loss 0.320216 LR 0.001000 Time 0.021870 +2023-10-05 20:59:06,818 - Epoch: [36][ 490/ 1236] Overall Loss 0.320324 Objective Loss 0.320324 LR 0.001000 Time 0.021844 +2023-10-05 20:59:07,030 - Epoch: [36][ 500/ 1236] Overall Loss 0.321009 Objective Loss 0.321009 LR 0.001000 Time 0.021831 +2023-10-05 20:59:07,236 - Epoch: [36][ 510/ 1236] Overall Loss 0.320998 Objective Loss 0.320998 LR 0.001000 Time 0.021807 +2023-10-05 20:59:07,448 - Epoch: [36][ 520/ 1236] Overall Loss 0.321635 Objective Loss 0.321635 LR 0.001000 Time 0.021795 +2023-10-05 20:59:07,654 - Epoch: [36][ 530/ 1236] Overall Loss 0.321614 Objective Loss 0.321614 LR 0.001000 Time 0.021772 +2023-10-05 20:59:07,866 - Epoch: [36][ 540/ 1236] Overall Loss 0.321457 Objective Loss 0.321457 LR 0.001000 Time 0.021761 +2023-10-05 20:59:08,073 - Epoch: [36][ 550/ 1236] Overall Loss 0.321301 Objective Loss 0.321301 LR 0.001000 Time 0.021740 +2023-10-05 20:59:08,285 - Epoch: [36][ 560/ 1236] Overall Loss 0.320523 Objective Loss 0.320523 LR 0.001000 Time 0.021730 +2023-10-05 20:59:08,491 - Epoch: [36][ 570/ 1236] Overall Loss 0.320425 Objective Loss 0.320425 LR 0.001000 Time 0.021710 +2023-10-05 20:59:08,703 - Epoch: [36][ 580/ 1236] Overall Loss 0.320575 Objective Loss 0.320575 LR 0.001000 Time 0.021700 +2023-10-05 20:59:08,903 - Epoch: [36][ 590/ 1236] Overall Loss 0.321053 Objective Loss 0.321053 LR 0.001000 Time 0.021670 +2023-10-05 20:59:09,105 - Epoch: [36][ 600/ 1236] Overall Loss 0.320962 Objective Loss 0.320962 LR 0.001000 Time 0.021645 +2023-10-05 20:59:09,305 - Epoch: [36][ 610/ 1236] Overall Loss 0.320865 Objective Loss 0.320865 LR 0.001000 Time 0.021618 +2023-10-05 20:59:09,506 - Epoch: [36][ 620/ 1236] Overall Loss 0.320673 Objective Loss 0.320673 LR 0.001000 Time 0.021594 +2023-10-05 20:59:09,706 - Epoch: [36][ 630/ 1236] Overall Loss 0.320999 Objective Loss 0.320999 LR 0.001000 Time 0.021567 +2023-10-05 20:59:09,908 - Epoch: [36][ 640/ 1236] Overall Loss 0.320357 Objective Loss 0.320357 LR 0.001000 Time 0.021545 +2023-10-05 20:59:10,108 - Epoch: [36][ 650/ 1236] Overall Loss 0.320569 Objective Loss 0.320569 LR 0.001000 Time 0.021521 +2023-10-05 20:59:10,310 - Epoch: [36][ 660/ 1236] Overall Loss 0.321371 Objective Loss 0.321371 LR 0.001000 Time 0.021500 +2023-10-05 20:59:10,511 - Epoch: [36][ 670/ 1236] Overall Loss 0.321665 Objective Loss 0.321665 LR 0.001000 Time 0.021479 +2023-10-05 20:59:10,712 - Epoch: [36][ 680/ 1236] Overall Loss 0.321618 Objective Loss 0.321618 LR 0.001000 Time 0.021459 +2023-10-05 20:59:10,913 - Epoch: [36][ 690/ 1236] Overall Loss 0.320927 Objective Loss 0.320927 LR 0.001000 Time 0.021438 +2023-10-05 20:59:11,114 - Epoch: [36][ 700/ 1236] Overall Loss 0.320338 Objective Loss 0.320338 LR 0.001000 Time 0.021419 +2023-10-05 20:59:11,314 - Epoch: [36][ 710/ 1236] Overall Loss 0.320404 Objective Loss 0.320404 LR 0.001000 Time 0.021398 +2023-10-05 20:59:11,516 - Epoch: [36][ 720/ 1236] Overall Loss 0.320204 Objective Loss 0.320204 LR 0.001000 Time 0.021381 +2023-10-05 20:59:11,716 - Epoch: [36][ 730/ 1236] Overall Loss 0.320418 Objective Loss 0.320418 LR 0.001000 Time 0.021362 +2023-10-05 20:59:11,918 - Epoch: [36][ 740/ 1236] Overall Loss 0.319965 Objective Loss 0.319965 LR 0.001000 Time 0.021345 +2023-10-05 20:59:12,118 - Epoch: [36][ 750/ 1236] Overall Loss 0.320291 Objective Loss 0.320291 LR 0.001000 Time 0.021327 +2023-10-05 20:59:12,321 - Epoch: [36][ 760/ 1236] Overall Loss 0.319758 Objective Loss 0.319758 LR 0.001000 Time 0.021313 +2023-10-05 20:59:12,522 - Epoch: [36][ 770/ 1236] Overall Loss 0.319630 Objective Loss 0.319630 LR 0.001000 Time 0.021297 +2023-10-05 20:59:12,726 - Epoch: [36][ 780/ 1236] Overall Loss 0.319946 Objective Loss 0.319946 LR 0.001000 Time 0.021284 +2023-10-05 20:59:12,927 - Epoch: [36][ 790/ 1236] Overall Loss 0.319621 Objective Loss 0.319621 LR 0.001000 Time 0.021269 +2023-10-05 20:59:13,131 - Epoch: [36][ 800/ 1236] Overall Loss 0.319977 Objective Loss 0.319977 LR 0.001000 Time 0.021259 +2023-10-05 20:59:13,333 - Epoch: [36][ 810/ 1236] Overall Loss 0.320443 Objective Loss 0.320443 LR 0.001000 Time 0.021245 +2023-10-05 20:59:13,538 - Epoch: [36][ 820/ 1236] Overall Loss 0.320506 Objective Loss 0.320506 LR 0.001000 Time 0.021234 +2023-10-05 20:59:13,739 - Epoch: [36][ 830/ 1236] Overall Loss 0.320860 Objective Loss 0.320860 LR 0.001000 Time 0.021221 +2023-10-05 20:59:13,943 - Epoch: [36][ 840/ 1236] Overall Loss 0.320743 Objective Loss 0.320743 LR 0.001000 Time 0.021211 +2023-10-05 20:59:14,145 - Epoch: [36][ 850/ 1236] Overall Loss 0.320716 Objective Loss 0.320716 LR 0.001000 Time 0.021199 +2023-10-05 20:59:14,350 - Epoch: [36][ 860/ 1236] Overall Loss 0.320830 Objective Loss 0.320830 LR 0.001000 Time 0.021190 +2023-10-05 20:59:14,551 - Epoch: [36][ 870/ 1236] Overall Loss 0.321115 Objective Loss 0.321115 LR 0.001000 Time 0.021177 +2023-10-05 20:59:14,755 - Epoch: [36][ 880/ 1236] Overall Loss 0.320992 Objective Loss 0.320992 LR 0.001000 Time 0.021168 +2023-10-05 20:59:14,957 - Epoch: [36][ 890/ 1236] Overall Loss 0.321096 Objective Loss 0.321096 LR 0.001000 Time 0.021157 +2023-10-05 20:59:15,162 - Epoch: [36][ 900/ 1236] Overall Loss 0.320912 Objective Loss 0.320912 LR 0.001000 Time 0.021149 +2023-10-05 20:59:15,363 - Epoch: [36][ 910/ 1236] Overall Loss 0.321583 Objective Loss 0.321583 LR 0.001000 Time 0.021137 +2023-10-05 20:59:15,567 - Epoch: [36][ 920/ 1236] Overall Loss 0.321718 Objective Loss 0.321718 LR 0.001000 Time 0.021129 +2023-10-05 20:59:15,769 - Epoch: [36][ 930/ 1236] Overall Loss 0.322055 Objective Loss 0.322055 LR 0.001000 Time 0.021118 +2023-10-05 20:59:15,973 - Epoch: [36][ 940/ 1236] Overall Loss 0.322061 Objective Loss 0.322061 LR 0.001000 Time 0.021111 +2023-10-05 20:59:16,175 - Epoch: [36][ 950/ 1236] Overall Loss 0.321905 Objective Loss 0.321905 LR 0.001000 Time 0.021100 +2023-10-05 20:59:16,379 - Epoch: [36][ 960/ 1236] Overall Loss 0.321868 Objective Loss 0.321868 LR 0.001000 Time 0.021093 +2023-10-05 20:59:16,581 - Epoch: [36][ 970/ 1236] Overall Loss 0.321553 Objective Loss 0.321553 LR 0.001000 Time 0.021083 +2023-10-05 20:59:16,785 - Epoch: [36][ 980/ 1236] Overall Loss 0.321642 Objective Loss 0.321642 LR 0.001000 Time 0.021076 +2023-10-05 20:59:16,987 - Epoch: [36][ 990/ 1236] Overall Loss 0.321252 Objective Loss 0.321252 LR 0.001000 Time 0.021066 +2023-10-05 20:59:17,191 - Epoch: [36][ 1000/ 1236] Overall Loss 0.321520 Objective Loss 0.321520 LR 0.001000 Time 0.021060 +2023-10-05 20:59:17,392 - Epoch: [36][ 1010/ 1236] Overall Loss 0.321684 Objective Loss 0.321684 LR 0.001000 Time 0.021050 +2023-10-05 20:59:17,597 - Epoch: [36][ 1020/ 1236] Overall Loss 0.322087 Objective Loss 0.322087 LR 0.001000 Time 0.021044 +2023-10-05 20:59:17,798 - Epoch: [36][ 1030/ 1236] Overall Loss 0.322239 Objective Loss 0.322239 LR 0.001000 Time 0.021035 +2023-10-05 20:59:18,002 - Epoch: [36][ 1040/ 1236] Overall Loss 0.321948 Objective Loss 0.321948 LR 0.001000 Time 0.021029 +2023-10-05 20:59:18,204 - Epoch: [36][ 1050/ 1236] Overall Loss 0.321821 Objective Loss 0.321821 LR 0.001000 Time 0.021020 +2023-10-05 20:59:18,408 - Epoch: [36][ 1060/ 1236] Overall Loss 0.321625 Objective Loss 0.321625 LR 0.001000 Time 0.021014 +2023-10-05 20:59:18,609 - Epoch: [36][ 1070/ 1236] Overall Loss 0.321654 Objective Loss 0.321654 LR 0.001000 Time 0.021005 +2023-10-05 20:59:18,814 - Epoch: [36][ 1080/ 1236] Overall Loss 0.321594 Objective Loss 0.321594 LR 0.001000 Time 0.021000 +2023-10-05 20:59:19,015 - Epoch: [36][ 1090/ 1236] Overall Loss 0.321829 Objective Loss 0.321829 LR 0.001000 Time 0.020992 +2023-10-05 20:59:19,219 - Epoch: [36][ 1100/ 1236] Overall Loss 0.321838 Objective Loss 0.321838 LR 0.001000 Time 0.020986 +2023-10-05 20:59:19,421 - Epoch: [36][ 1110/ 1236] Overall Loss 0.321953 Objective Loss 0.321953 LR 0.001000 Time 0.020978 +2023-10-05 20:59:19,625 - Epoch: [36][ 1120/ 1236] Overall Loss 0.321915 Objective Loss 0.321915 LR 0.001000 Time 0.020973 +2023-10-05 20:59:19,827 - Epoch: [36][ 1130/ 1236] Overall Loss 0.321879 Objective Loss 0.321879 LR 0.001000 Time 0.020965 +2023-10-05 20:59:20,031 - Epoch: [36][ 1140/ 1236] Overall Loss 0.321817 Objective Loss 0.321817 LR 0.001000 Time 0.020961 +2023-10-05 20:59:20,232 - Epoch: [36][ 1150/ 1236] Overall Loss 0.321870 Objective Loss 0.321870 LR 0.001000 Time 0.020953 +2023-10-05 20:59:20,437 - Epoch: [36][ 1160/ 1236] Overall Loss 0.321880 Objective Loss 0.321880 LR 0.001000 Time 0.020948 +2023-10-05 20:59:20,638 - Epoch: [36][ 1170/ 1236] Overall Loss 0.322012 Objective Loss 0.322012 LR 0.001000 Time 0.020941 +2023-10-05 20:59:20,843 - Epoch: [36][ 1180/ 1236] Overall Loss 0.322104 Objective Loss 0.322104 LR 0.001000 Time 0.020937 +2023-10-05 20:59:21,044 - Epoch: [36][ 1190/ 1236] Overall Loss 0.321834 Objective Loss 0.321834 LR 0.001000 Time 0.020929 +2023-10-05 20:59:21,248 - Epoch: [36][ 1200/ 1236] Overall Loss 0.321653 Objective Loss 0.321653 LR 0.001000 Time 0.020925 +2023-10-05 20:59:21,449 - Epoch: [36][ 1210/ 1236] Overall Loss 0.321607 Objective Loss 0.321607 LR 0.001000 Time 0.020918 +2023-10-05 20:59:21,654 - Epoch: [36][ 1220/ 1236] Overall Loss 0.321599 Objective Loss 0.321599 LR 0.001000 Time 0.020914 +2023-10-05 20:59:21,907 - Epoch: [36][ 1230/ 1236] Overall Loss 0.321820 Objective Loss 0.321820 LR 0.001000 Time 0.020950 +2023-10-05 20:59:22,025 - Epoch: [36][ 1236/ 1236] Overall Loss 0.321948 Objective Loss 0.321948 Top1 83.503055 Top5 97.148676 LR 0.001000 Time 0.020943 +2023-10-05 20:59:22,168 - --- validate (epoch=36)----------- +2023-10-05 20:59:22,168 - 29943 samples (256 per mini-batch) +2023-10-05 20:59:22,625 - Epoch: [36][ 10/ 117] Loss 0.357453 Top1 81.992188 Top5 97.968750 +2023-10-05 20:59:22,781 - Epoch: [36][ 20/ 117] Loss 0.352508 Top1 81.777344 Top5 97.910156 +2023-10-05 20:59:22,936 - Epoch: [36][ 30/ 117] Loss 0.377101 Top1 81.328125 Top5 97.656250 +2023-10-05 20:59:23,092 - Epoch: [36][ 40/ 117] Loss 0.379155 Top1 81.269531 Top5 97.529297 +2023-10-05 20:59:23,246 - Epoch: [36][ 50/ 117] Loss 0.388580 Top1 81.093750 Top5 97.531250 +2023-10-05 20:59:23,402 - Epoch: [36][ 60/ 117] Loss 0.385929 Top1 81.165365 Top5 97.558594 +2023-10-05 20:59:23,556 - Epoch: [36][ 70/ 117] Loss 0.378886 Top1 81.389509 Top5 97.606027 +2023-10-05 20:59:23,712 - Epoch: [36][ 80/ 117] Loss 0.376724 Top1 81.557617 Top5 97.583008 +2023-10-05 20:59:23,866 - Epoch: [36][ 90/ 117] Loss 0.372632 Top1 81.666667 Top5 97.621528 +2023-10-05 20:59:24,024 - Epoch: [36][ 100/ 117] Loss 0.374301 Top1 81.574219 Top5 97.597656 +2023-10-05 20:59:24,187 - Epoch: [36][ 110/ 117] Loss 0.371894 Top1 81.640625 Top5 97.581676 +2023-10-05 20:59:24,271 - Epoch: [36][ 117/ 117] Loss 0.370817 Top1 81.741976 Top5 97.612130 +2023-10-05 20:59:24,411 - ==> Top1: 81.742 Top5: 97.612 Loss: 0.371 + +2023-10-05 20:59:24,412 - ==> Confusion: +[[ 919 2 4 1 10 1 0 0 6 71 2 0 0 0 8 2 6 1 0 1 16] + [ 1 1052 1 2 6 17 2 14 3 0 8 2 0 1 1 4 6 0 5 1 5] + [ 9 0 938 14 3 0 28 7 0 3 12 3 7 4 4 0 4 3 3 5 9] + [ 4 2 19 949 1 2 3 0 5 0 17 0 5 4 37 9 2 6 11 3 10] + [ 25 8 1 0 972 4 0 0 0 9 0 1 3 1 4 5 9 3 0 0 5] + [ 4 43 3 1 11 923 1 23 6 3 13 7 5 25 6 3 2 3 2 13 19] + [ 0 8 38 1 1 0 1104 5 0 0 5 2 3 0 0 8 0 2 1 7 6] + [ 3 27 21 0 2 35 3 1006 0 3 15 8 4 2 0 4 0 0 39 29 17] + [ 17 2 0 0 1 0 0 0 972 50 10 0 2 9 13 4 1 2 1 3 2] + [ 88 0 0 0 3 0 0 0 32 938 2 2 0 29 2 7 1 1 0 2 12] + [ 4 2 12 4 1 3 3 0 11 0 987 2 1 7 2 1 1 1 2 2 7] + [ 0 0 1 0 0 9 0 0 0 1 0 906 71 6 0 5 1 17 0 10 8] + [ 1 0 8 3 0 0 0 0 0 0 1 20 989 1 3 11 2 8 4 5 12] + [ 4 0 1 1 5 7 0 0 11 15 15 6 6 1027 3 1 3 0 0 0 14] + [ 11 1 3 14 7 0 0 0 31 8 8 1 3 1 987 0 4 2 4 0 16] + [ 1 5 2 1 2 0 0 0 0 0 0 11 8 0 0 1072 7 11 1 8 5] + [ 3 14 0 0 9 5 0 0 3 0 3 5 1 2 3 18 1072 0 0 10 13] + [ 0 0 0 3 0 0 2 0 2 0 0 5 23 0 2 6 0 990 0 0 5] + [ 1 14 11 15 1 0 1 19 5 0 9 0 6 0 19 1 1 0 945 5 15] + [ 0 1 6 2 2 2 6 10 0 0 4 16 9 1 0 9 7 0 1 1063 13] + [ 177 242 190 76 111 119 63 55 122 106 260 77 508 272 128 103 209 81 122 219 4665]] + +2023-10-05 20:59:24,413 - ==> Best [Top1: 81.742 Top5: 97.612 Sparsity:0.00 Params: 148928 on epoch: 36] +2023-10-05 20:59:24,413 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:59:24,426 - + +2023-10-05 20:59:24,426 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:59:25,397 - Epoch: [37][ 10/ 1236] Overall Loss 0.297647 Objective Loss 0.297647 LR 0.001000 Time 0.097029 +2023-10-05 20:59:25,596 - Epoch: [37][ 20/ 1236] Overall Loss 0.310999 Objective Loss 0.310999 LR 0.001000 Time 0.058468 +2023-10-05 20:59:25,795 - Epoch: [37][ 30/ 1236] Overall Loss 0.309863 Objective Loss 0.309863 LR 0.001000 Time 0.045583 +2023-10-05 20:59:25,994 - Epoch: [37][ 40/ 1236] Overall Loss 0.304685 Objective Loss 0.304685 LR 0.001000 Time 0.039155 +2023-10-05 20:59:26,193 - Epoch: [37][ 50/ 1236] Overall Loss 0.312516 Objective Loss 0.312516 LR 0.001000 Time 0.035296 +2023-10-05 20:59:26,392 - Epoch: [37][ 60/ 1236] Overall Loss 0.315267 Objective Loss 0.315267 LR 0.001000 Time 0.032732 +2023-10-05 20:59:26,591 - Epoch: [37][ 70/ 1236] Overall Loss 0.310915 Objective Loss 0.310915 LR 0.001000 Time 0.030893 +2023-10-05 20:59:26,791 - Epoch: [37][ 80/ 1236] Overall Loss 0.311168 Objective Loss 0.311168 LR 0.001000 Time 0.029523 +2023-10-05 20:59:26,989 - Epoch: [37][ 90/ 1236] Overall Loss 0.310972 Objective Loss 0.310972 LR 0.001000 Time 0.028447 +2023-10-05 20:59:27,189 - Epoch: [37][ 100/ 1236] Overall Loss 0.310550 Objective Loss 0.310550 LR 0.001000 Time 0.027596 +2023-10-05 20:59:27,388 - Epoch: [37][ 110/ 1236] Overall Loss 0.311830 Objective Loss 0.311830 LR 0.001000 Time 0.026893 +2023-10-05 20:59:27,587 - Epoch: [37][ 120/ 1236] Overall Loss 0.309540 Objective Loss 0.309540 LR 0.001000 Time 0.026309 +2023-10-05 20:59:27,786 - Epoch: [37][ 130/ 1236] Overall Loss 0.308369 Objective Loss 0.308369 LR 0.001000 Time 0.025812 +2023-10-05 20:59:27,985 - Epoch: [37][ 140/ 1236] Overall Loss 0.311509 Objective Loss 0.311509 LR 0.001000 Time 0.025389 +2023-10-05 20:59:28,184 - Epoch: [37][ 150/ 1236] Overall Loss 0.313441 Objective Loss 0.313441 LR 0.001000 Time 0.025019 +2023-10-05 20:59:28,384 - Epoch: [37][ 160/ 1236] Overall Loss 0.317941 Objective Loss 0.317941 LR 0.001000 Time 0.024699 +2023-10-05 20:59:28,583 - Epoch: [37][ 170/ 1236] Overall Loss 0.318438 Objective Loss 0.318438 LR 0.001000 Time 0.024418 +2023-10-05 20:59:28,782 - Epoch: [37][ 180/ 1236] Overall Loss 0.318611 Objective Loss 0.318611 LR 0.001000 Time 0.024166 +2023-10-05 20:59:28,982 - Epoch: [37][ 190/ 1236] Overall Loss 0.318167 Objective Loss 0.318167 LR 0.001000 Time 0.023941 +2023-10-05 20:59:29,181 - Epoch: [37][ 200/ 1236] Overall Loss 0.318046 Objective Loss 0.318046 LR 0.001000 Time 0.023740 +2023-10-05 20:59:29,379 - Epoch: [37][ 210/ 1236] Overall Loss 0.317052 Objective Loss 0.317052 LR 0.001000 Time 0.023551 +2023-10-05 20:59:29,578 - Epoch: [37][ 220/ 1236] Overall Loss 0.318134 Objective Loss 0.318134 LR 0.001000 Time 0.023385 +2023-10-05 20:59:29,777 - Epoch: [37][ 230/ 1236] Overall Loss 0.319438 Objective Loss 0.319438 LR 0.001000 Time 0.023230 +2023-10-05 20:59:29,976 - Epoch: [37][ 240/ 1236] Overall Loss 0.319712 Objective Loss 0.319712 LR 0.001000 Time 0.023091 +2023-10-05 20:59:30,175 - Epoch: [37][ 250/ 1236] Overall Loss 0.319584 Objective Loss 0.319584 LR 0.001000 Time 0.022962 +2023-10-05 20:59:30,375 - Epoch: [37][ 260/ 1236] Overall Loss 0.320482 Objective Loss 0.320482 LR 0.001000 Time 0.022844 +2023-10-05 20:59:30,573 - Epoch: [37][ 270/ 1236] Overall Loss 0.320263 Objective Loss 0.320263 LR 0.001000 Time 0.022733 +2023-10-05 20:59:30,773 - Epoch: [37][ 280/ 1236] Overall Loss 0.319322 Objective Loss 0.319322 LR 0.001000 Time 0.022633 +2023-10-05 20:59:30,977 - Epoch: [37][ 290/ 1236] Overall Loss 0.319287 Objective Loss 0.319287 LR 0.001000 Time 0.022554 +2023-10-05 20:59:31,186 - Epoch: [37][ 300/ 1236] Overall Loss 0.318472 Objective Loss 0.318472 LR 0.001000 Time 0.022498 +2023-10-05 20:59:31,390 - Epoch: [37][ 310/ 1236] Overall Loss 0.318198 Objective Loss 0.318198 LR 0.001000 Time 0.022431 +2023-10-05 20:59:31,593 - Epoch: [37][ 320/ 1236] Overall Loss 0.318155 Objective Loss 0.318155 LR 0.001000 Time 0.022362 +2023-10-05 20:59:31,796 - Epoch: [37][ 330/ 1236] Overall Loss 0.317206 Objective Loss 0.317206 LR 0.001000 Time 0.022300 +2023-10-05 20:59:32,001 - Epoch: [37][ 340/ 1236] Overall Loss 0.316262 Objective Loss 0.316262 LR 0.001000 Time 0.022245 +2023-10-05 20:59:32,207 - Epoch: [37][ 350/ 1236] Overall Loss 0.316137 Objective Loss 0.316137 LR 0.001000 Time 0.022195 +2023-10-05 20:59:32,412 - Epoch: [37][ 360/ 1236] Overall Loss 0.316375 Objective Loss 0.316375 LR 0.001000 Time 0.022148 +2023-10-05 20:59:32,617 - Epoch: [37][ 370/ 1236] Overall Loss 0.317113 Objective Loss 0.317113 LR 0.001000 Time 0.022103 +2023-10-05 20:59:32,822 - Epoch: [37][ 380/ 1236] Overall Loss 0.317318 Objective Loss 0.317318 LR 0.001000 Time 0.022058 +2023-10-05 20:59:33,027 - Epoch: [37][ 390/ 1236] Overall Loss 0.318077 Objective Loss 0.318077 LR 0.001000 Time 0.022018 +2023-10-05 20:59:33,233 - Epoch: [37][ 400/ 1236] Overall Loss 0.318561 Objective Loss 0.318561 LR 0.001000 Time 0.021981 +2023-10-05 20:59:33,438 - Epoch: [37][ 410/ 1236] Overall Loss 0.318209 Objective Loss 0.318209 LR 0.001000 Time 0.021944 +2023-10-05 20:59:33,642 - Epoch: [37][ 420/ 1236] Overall Loss 0.318101 Objective Loss 0.318101 LR 0.001000 Time 0.021907 +2023-10-05 20:59:33,847 - Epoch: [37][ 430/ 1236] Overall Loss 0.317301 Objective Loss 0.317301 LR 0.001000 Time 0.021875 +2023-10-05 20:59:34,052 - Epoch: [37][ 440/ 1236] Overall Loss 0.316947 Objective Loss 0.316947 LR 0.001000 Time 0.021842 +2023-10-05 20:59:34,258 - Epoch: [37][ 450/ 1236] Overall Loss 0.317131 Objective Loss 0.317131 LR 0.001000 Time 0.021812 +2023-10-05 20:59:34,462 - Epoch: [37][ 460/ 1236] Overall Loss 0.316809 Objective Loss 0.316809 LR 0.001000 Time 0.021782 +2023-10-05 20:59:34,668 - Epoch: [37][ 470/ 1236] Overall Loss 0.317621 Objective Loss 0.317621 LR 0.001000 Time 0.021755 +2023-10-05 20:59:34,873 - Epoch: [37][ 480/ 1236] Overall Loss 0.317935 Objective Loss 0.317935 LR 0.001000 Time 0.021728 +2023-10-05 20:59:35,078 - Epoch: [37][ 490/ 1236] Overall Loss 0.317489 Objective Loss 0.317489 LR 0.001000 Time 0.021702 +2023-10-05 20:59:35,282 - Epoch: [37][ 500/ 1236] Overall Loss 0.318183 Objective Loss 0.318183 LR 0.001000 Time 0.021676 +2023-10-05 20:59:35,488 - Epoch: [37][ 510/ 1236] Overall Loss 0.318808 Objective Loss 0.318808 LR 0.001000 Time 0.021653 +2023-10-05 20:59:35,693 - Epoch: [37][ 520/ 1236] Overall Loss 0.318960 Objective Loss 0.318960 LR 0.001000 Time 0.021631 +2023-10-05 20:59:35,898 - Epoch: [37][ 530/ 1236] Overall Loss 0.318359 Objective Loss 0.318359 LR 0.001000 Time 0.021608 +2023-10-05 20:59:36,103 - Epoch: [37][ 540/ 1236] Overall Loss 0.318567 Objective Loss 0.318567 LR 0.001000 Time 0.021586 +2023-10-05 20:59:36,308 - Epoch: [37][ 550/ 1236] Overall Loss 0.318384 Objective Loss 0.318384 LR 0.001000 Time 0.021567 +2023-10-05 20:59:36,513 - Epoch: [37][ 560/ 1236] Overall Loss 0.318070 Objective Loss 0.318070 LR 0.001000 Time 0.021546 +2023-10-05 20:59:36,718 - Epoch: [37][ 570/ 1236] Overall Loss 0.317938 Objective Loss 0.317938 LR 0.001000 Time 0.021528 +2023-10-05 20:59:36,923 - Epoch: [37][ 580/ 1236] Overall Loss 0.317851 Objective Loss 0.317851 LR 0.001000 Time 0.021509 +2023-10-05 20:59:37,129 - Epoch: [37][ 590/ 1236] Overall Loss 0.317858 Objective Loss 0.317858 LR 0.001000 Time 0.021493 +2023-10-05 20:59:37,333 - Epoch: [37][ 600/ 1236] Overall Loss 0.318177 Objective Loss 0.318177 LR 0.001000 Time 0.021475 +2023-10-05 20:59:37,539 - Epoch: [37][ 610/ 1236] Overall Loss 0.318434 Objective Loss 0.318434 LR 0.001000 Time 0.021459 +2023-10-05 20:59:37,744 - Epoch: [37][ 620/ 1236] Overall Loss 0.318151 Objective Loss 0.318151 LR 0.001000 Time 0.021444 +2023-10-05 20:59:37,949 - Epoch: [37][ 630/ 1236] Overall Loss 0.318288 Objective Loss 0.318288 LR 0.001000 Time 0.021428 +2023-10-05 20:59:38,154 - Epoch: [37][ 640/ 1236] Overall Loss 0.318375 Objective Loss 0.318375 LR 0.001000 Time 0.021413 +2023-10-05 20:59:38,360 - Epoch: [37][ 650/ 1236] Overall Loss 0.318693 Objective Loss 0.318693 LR 0.001000 Time 0.021399 +2023-10-05 20:59:38,565 - Epoch: [37][ 660/ 1236] Overall Loss 0.318286 Objective Loss 0.318286 LR 0.001000 Time 0.021385 +2023-10-05 20:59:38,770 - Epoch: [37][ 670/ 1236] Overall Loss 0.318339 Objective Loss 0.318339 LR 0.001000 Time 0.021372 +2023-10-05 20:59:38,975 - Epoch: [37][ 680/ 1236] Overall Loss 0.318466 Objective Loss 0.318466 LR 0.001000 Time 0.021358 +2023-10-05 20:59:39,181 - Epoch: [37][ 690/ 1236] Overall Loss 0.318539 Objective Loss 0.318539 LR 0.001000 Time 0.021346 +2023-10-05 20:59:39,385 - Epoch: [37][ 700/ 1236] Overall Loss 0.318531 Objective Loss 0.318531 LR 0.001000 Time 0.021333 +2023-10-05 20:59:39,591 - Epoch: [37][ 710/ 1236] Overall Loss 0.318821 Objective Loss 0.318821 LR 0.001000 Time 0.021321 +2023-10-05 20:59:39,796 - Epoch: [37][ 720/ 1236] Overall Loss 0.319182 Objective Loss 0.319182 LR 0.001000 Time 0.021309 +2023-10-05 20:59:40,001 - Epoch: [37][ 730/ 1236] Overall Loss 0.319219 Objective Loss 0.319219 LR 0.001000 Time 0.021298 +2023-10-05 20:59:40,205 - Epoch: [37][ 740/ 1236] Overall Loss 0.319677 Objective Loss 0.319677 LR 0.001000 Time 0.021286 +2023-10-05 20:59:40,411 - Epoch: [37][ 750/ 1236] Overall Loss 0.319840 Objective Loss 0.319840 LR 0.001000 Time 0.021275 +2023-10-05 20:59:40,615 - Epoch: [37][ 760/ 1236] Overall Loss 0.320140 Objective Loss 0.320140 LR 0.001000 Time 0.021264 +2023-10-05 20:59:40,821 - Epoch: [37][ 770/ 1236] Overall Loss 0.320864 Objective Loss 0.320864 LR 0.001000 Time 0.021254 +2023-10-05 20:59:41,026 - Epoch: [37][ 780/ 1236] Overall Loss 0.320944 Objective Loss 0.320944 LR 0.001000 Time 0.021244 +2023-10-05 20:59:41,231 - Epoch: [37][ 790/ 1236] Overall Loss 0.320775 Objective Loss 0.320775 LR 0.001000 Time 0.021234 +2023-10-05 20:59:41,436 - Epoch: [37][ 800/ 1236] Overall Loss 0.320662 Objective Loss 0.320662 LR 0.001000 Time 0.021225 +2023-10-05 20:59:41,642 - Epoch: [37][ 810/ 1236] Overall Loss 0.320906 Objective Loss 0.320906 LR 0.001000 Time 0.021216 +2023-10-05 20:59:41,846 - Epoch: [37][ 820/ 1236] Overall Loss 0.320809 Objective Loss 0.320809 LR 0.001000 Time 0.021207 +2023-10-05 20:59:42,052 - Epoch: [37][ 830/ 1236] Overall Loss 0.320758 Objective Loss 0.320758 LR 0.001000 Time 0.021198 +2023-10-05 20:59:42,257 - Epoch: [37][ 840/ 1236] Overall Loss 0.320797 Objective Loss 0.320797 LR 0.001000 Time 0.021189 +2023-10-05 20:59:42,462 - Epoch: [37][ 850/ 1236] Overall Loss 0.320994 Objective Loss 0.320994 LR 0.001000 Time 0.021181 +2023-10-05 20:59:42,667 - Epoch: [37][ 860/ 1236] Overall Loss 0.320928 Objective Loss 0.320928 LR 0.001000 Time 0.021173 +2023-10-05 20:59:42,873 - Epoch: [37][ 870/ 1236] Overall Loss 0.320840 Objective Loss 0.320840 LR 0.001000 Time 0.021165 +2023-10-05 20:59:43,078 - Epoch: [37][ 880/ 1236] Overall Loss 0.321131 Objective Loss 0.321131 LR 0.001000 Time 0.021158 +2023-10-05 20:59:43,283 - Epoch: [37][ 890/ 1236] Overall Loss 0.321631 Objective Loss 0.321631 LR 0.001000 Time 0.021150 +2023-10-05 20:59:43,488 - Epoch: [37][ 900/ 1236] Overall Loss 0.321547 Objective Loss 0.321547 LR 0.001000 Time 0.021142 +2023-10-05 20:59:43,694 - Epoch: [37][ 910/ 1236] Overall Loss 0.321604 Objective Loss 0.321604 LR 0.001000 Time 0.021135 +2023-10-05 20:59:43,899 - Epoch: [37][ 920/ 1236] Overall Loss 0.321829 Objective Loss 0.321829 LR 0.001000 Time 0.021129 +2023-10-05 20:59:44,105 - Epoch: [37][ 930/ 1236] Overall Loss 0.322118 Objective Loss 0.322118 LR 0.001000 Time 0.021122 +2023-10-05 20:59:44,310 - Epoch: [37][ 940/ 1236] Overall Loss 0.322230 Objective Loss 0.322230 LR 0.001000 Time 0.021115 +2023-10-05 20:59:44,515 - Epoch: [37][ 950/ 1236] Overall Loss 0.322231 Objective Loss 0.322231 LR 0.001000 Time 0.021108 +2023-10-05 20:59:44,720 - Epoch: [37][ 960/ 1236] Overall Loss 0.322307 Objective Loss 0.322307 LR 0.001000 Time 0.021101 +2023-10-05 20:59:44,925 - Epoch: [37][ 970/ 1236] Overall Loss 0.322300 Objective Loss 0.322300 LR 0.001000 Time 0.021095 +2023-10-05 20:59:45,130 - Epoch: [37][ 980/ 1236] Overall Loss 0.322619 Objective Loss 0.322619 LR 0.001000 Time 0.021088 +2023-10-05 20:59:45,329 - Epoch: [37][ 990/ 1236] Overall Loss 0.323059 Objective Loss 0.323059 LR 0.001000 Time 0.021076 +2023-10-05 20:59:45,529 - Epoch: [37][ 1000/ 1236] Overall Loss 0.323225 Objective Loss 0.323225 LR 0.001000 Time 0.021065 +2023-10-05 20:59:45,729 - Epoch: [37][ 1010/ 1236] Overall Loss 0.322985 Objective Loss 0.322985 LR 0.001000 Time 0.021054 +2023-10-05 20:59:45,929 - Epoch: [37][ 1020/ 1236] Overall Loss 0.323104 Objective Loss 0.323104 LR 0.001000 Time 0.021043 +2023-10-05 20:59:46,128 - Epoch: [37][ 1030/ 1236] Overall Loss 0.323278 Objective Loss 0.323278 LR 0.001000 Time 0.021031 +2023-10-05 20:59:46,328 - Epoch: [37][ 1040/ 1236] Overall Loss 0.323273 Objective Loss 0.323273 LR 0.001000 Time 0.021021 +2023-10-05 20:59:46,527 - Epoch: [37][ 1050/ 1236] Overall Loss 0.323148 Objective Loss 0.323148 LR 0.001000 Time 0.021010 +2023-10-05 20:59:46,728 - Epoch: [37][ 1060/ 1236] Overall Loss 0.323049 Objective Loss 0.323049 LR 0.001000 Time 0.021001 +2023-10-05 20:59:46,927 - Epoch: [37][ 1070/ 1236] Overall Loss 0.323120 Objective Loss 0.323120 LR 0.001000 Time 0.020991 +2023-10-05 20:59:47,127 - Epoch: [37][ 1080/ 1236] Overall Loss 0.322918 Objective Loss 0.322918 LR 0.001000 Time 0.020981 +2023-10-05 20:59:47,326 - Epoch: [37][ 1090/ 1236] Overall Loss 0.322751 Objective Loss 0.322751 LR 0.001000 Time 0.020971 +2023-10-05 20:59:47,526 - Epoch: [37][ 1100/ 1236] Overall Loss 0.322645 Objective Loss 0.322645 LR 0.001000 Time 0.020962 +2023-10-05 20:59:47,725 - Epoch: [37][ 1110/ 1236] Overall Loss 0.322762 Objective Loss 0.322762 LR 0.001000 Time 0.020952 +2023-10-05 20:59:47,926 - Epoch: [37][ 1120/ 1236] Overall Loss 0.322976 Objective Loss 0.322976 LR 0.001000 Time 0.020944 +2023-10-05 20:59:48,125 - Epoch: [37][ 1130/ 1236] Overall Loss 0.323155 Objective Loss 0.323155 LR 0.001000 Time 0.020934 +2023-10-05 20:59:48,325 - Epoch: [37][ 1140/ 1236] Overall Loss 0.323088 Objective Loss 0.323088 LR 0.001000 Time 0.020926 +2023-10-05 20:59:48,524 - Epoch: [37][ 1150/ 1236] Overall Loss 0.322917 Objective Loss 0.322917 LR 0.001000 Time 0.020917 +2023-10-05 20:59:48,724 - Epoch: [37][ 1160/ 1236] Overall Loss 0.323166 Objective Loss 0.323166 LR 0.001000 Time 0.020909 +2023-10-05 20:59:48,923 - Epoch: [37][ 1170/ 1236] Overall Loss 0.323140 Objective Loss 0.323140 LR 0.001000 Time 0.020900 +2023-10-05 20:59:49,124 - Epoch: [37][ 1180/ 1236] Overall Loss 0.323169 Objective Loss 0.323169 LR 0.001000 Time 0.020893 +2023-10-05 20:59:49,323 - Epoch: [37][ 1190/ 1236] Overall Loss 0.323325 Objective Loss 0.323325 LR 0.001000 Time 0.020884 +2023-10-05 20:59:49,523 - Epoch: [37][ 1200/ 1236] Overall Loss 0.323387 Objective Loss 0.323387 LR 0.001000 Time 0.020877 +2023-10-05 20:59:49,723 - Epoch: [37][ 1210/ 1236] Overall Loss 0.323564 Objective Loss 0.323564 LR 0.001000 Time 0.020868 +2023-10-05 20:59:49,923 - Epoch: [37][ 1220/ 1236] Overall Loss 0.323490 Objective Loss 0.323490 LR 0.001000 Time 0.020861 +2023-10-05 20:59:50,175 - Epoch: [37][ 1230/ 1236] Overall Loss 0.323413 Objective Loss 0.323413 LR 0.001000 Time 0.020896 +2023-10-05 20:59:50,292 - Epoch: [37][ 1236/ 1236] Overall Loss 0.323629 Objective Loss 0.323629 Top1 83.706721 Top5 98.370672 LR 0.001000 Time 0.020889 +2023-10-05 20:59:50,395 - --- validate (epoch=37)----------- +2023-10-05 20:59:50,395 - 29943 samples (256 per mini-batch) +2023-10-05 20:59:50,846 - Epoch: [37][ 10/ 117] Loss 0.344390 Top1 82.773438 Top5 97.539062 +2023-10-05 20:59:50,997 - Epoch: [37][ 20/ 117] Loss 0.372644 Top1 81.328125 Top5 97.265625 +2023-10-05 20:59:51,147 - Epoch: [37][ 30/ 117] Loss 0.368421 Top1 81.171875 Top5 97.408854 +2023-10-05 20:59:51,298 - Epoch: [37][ 40/ 117] Loss 0.369602 Top1 80.986328 Top5 97.373047 +2023-10-05 20:59:51,448 - Epoch: [37][ 50/ 117] Loss 0.375471 Top1 80.828125 Top5 97.343750 +2023-10-05 20:59:51,599 - Epoch: [37][ 60/ 117] Loss 0.379489 Top1 80.644531 Top5 97.369792 +2023-10-05 20:59:51,747 - Epoch: [37][ 70/ 117] Loss 0.378273 Top1 80.652902 Top5 97.421875 +2023-10-05 20:59:51,893 - Epoch: [37][ 80/ 117] Loss 0.378203 Top1 80.644531 Top5 97.397461 +2023-10-05 20:59:52,039 - Epoch: [37][ 90/ 117] Loss 0.377853 Top1 80.672743 Top5 97.404514 +2023-10-05 20:59:52,186 - Epoch: [37][ 100/ 117] Loss 0.378407 Top1 80.597656 Top5 97.371094 +2023-10-05 20:59:52,337 - Epoch: [37][ 110/ 117] Loss 0.381432 Top1 80.518466 Top5 97.357955 +2023-10-05 20:59:52,421 - Epoch: [37][ 117/ 117] Loss 0.382128 Top1 80.439502 Top5 97.364993 +2023-10-05 20:59:52,559 - ==> Top1: 80.440 Top5: 97.365 Loss: 0.382 + +2023-10-05 20:59:52,560 - ==> Confusion: +[[ 916 6 7 2 19 0 0 0 4 57 1 0 0 3 15 4 6 0 1 0 9] + [ 2 1062 2 0 17 16 1 13 2 0 1 1 0 0 2 3 3 0 6 0 0] + [ 1 1 920 22 5 0 44 7 0 1 6 4 4 4 0 3 2 1 16 5 10] + [ 3 1 18 939 3 6 1 0 2 0 10 0 1 2 38 4 2 7 35 2 15] + [ 28 8 1 0 971 3 1 0 1 11 2 1 0 0 7 6 7 1 0 1 1] + [ 5 55 2 0 5 936 6 26 0 4 4 4 2 17 12 2 5 1 3 8 19] + [ 0 11 32 0 1 0 1115 5 0 0 1 1 1 0 1 9 0 0 3 7 4] + [ 4 55 19 1 6 33 6 982 0 1 3 9 1 0 1 2 0 0 77 11 7] + [ 23 7 0 0 2 2 0 0 937 54 17 1 0 10 25 5 0 0 1 1 4] + [ 115 1 3 0 9 4 2 0 19 921 0 3 0 16 9 4 3 1 1 2 6] + [ 5 7 11 4 3 0 7 3 6 1 972 4 0 7 7 1 1 0 8 2 4] + [ 1 1 4 1 3 18 0 3 0 0 1 922 36 4 0 4 8 17 0 9 3] + [ 1 4 5 2 0 0 1 0 0 1 1 39 950 0 3 13 1 18 6 8 15] + [ 1 2 2 2 10 11 0 2 12 22 8 4 2 1008 6 2 5 2 0 6 12] + [ 13 2 3 14 9 0 0 0 18 7 5 0 0 1 1001 0 2 2 13 0 11] + [ 0 3 2 2 5 1 2 1 0 1 0 7 5 1 0 1054 19 13 1 10 7] + [ 0 17 1 0 17 0 1 0 0 0 0 1 1 2 2 9 1101 0 0 3 6] + [ 0 0 0 2 0 0 3 0 0 1 0 7 12 1 9 7 1 989 1 1 4] + [ 2 22 11 13 1 1 1 11 1 0 1 0 1 0 18 2 2 0 970 1 10] + [ 0 6 6 3 3 8 17 13 0 0 4 11 5 2 1 8 16 1 6 1033 9] + [ 150 289 173 61 181 162 62 76 94 116 194 94 383 291 215 93 342 78 199 265 4387]] + +2023-10-05 20:59:52,561 - ==> Best [Top1: 81.742 Top5: 97.612 Sparsity:0.00 Params: 148928 on epoch: 36] +2023-10-05 20:59:52,561 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 20:59:52,567 - + +2023-10-05 20:59:52,567 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 20:59:53,659 - Epoch: [38][ 10/ 1236] Overall Loss 0.321159 Objective Loss 0.321159 LR 0.001000 Time 0.109201 +2023-10-05 20:59:53,861 - Epoch: [38][ 20/ 1236] Overall Loss 0.325465 Objective Loss 0.325465 LR 0.001000 Time 0.064688 +2023-10-05 20:59:54,062 - Epoch: [38][ 30/ 1236] Overall Loss 0.326642 Objective Loss 0.326642 LR 0.001000 Time 0.049786 +2023-10-05 20:59:54,264 - Epoch: [38][ 40/ 1236] Overall Loss 0.320241 Objective Loss 0.320241 LR 0.001000 Time 0.042382 +2023-10-05 20:59:54,464 - Epoch: [38][ 50/ 1236] Overall Loss 0.320418 Objective Loss 0.320418 LR 0.001000 Time 0.037913 +2023-10-05 20:59:54,666 - Epoch: [38][ 60/ 1236] Overall Loss 0.324063 Objective Loss 0.324063 LR 0.001000 Time 0.034944 +2023-10-05 20:59:54,866 - Epoch: [38][ 70/ 1236] Overall Loss 0.320732 Objective Loss 0.320732 LR 0.001000 Time 0.032814 +2023-10-05 20:59:55,069 - Epoch: [38][ 80/ 1236] Overall Loss 0.318386 Objective Loss 0.318386 LR 0.001000 Time 0.031245 +2023-10-05 20:59:55,271 - Epoch: [38][ 90/ 1236] Overall Loss 0.318970 Objective Loss 0.318970 LR 0.001000 Time 0.030009 +2023-10-05 20:59:55,473 - Epoch: [38][ 100/ 1236] Overall Loss 0.321181 Objective Loss 0.321181 LR 0.001000 Time 0.029029 +2023-10-05 20:59:55,675 - Epoch: [38][ 110/ 1236] Overall Loss 0.320256 Objective Loss 0.320256 LR 0.001000 Time 0.028226 +2023-10-05 20:59:55,878 - Epoch: [38][ 120/ 1236] Overall Loss 0.321998 Objective Loss 0.321998 LR 0.001000 Time 0.027563 +2023-10-05 20:59:56,081 - Epoch: [38][ 130/ 1236] Overall Loss 0.324099 Objective Loss 0.324099 LR 0.001000 Time 0.026997 +2023-10-05 20:59:56,284 - Epoch: [38][ 140/ 1236] Overall Loss 0.322355 Objective Loss 0.322355 LR 0.001000 Time 0.026518 +2023-10-05 20:59:56,486 - Epoch: [38][ 150/ 1236] Overall Loss 0.322668 Objective Loss 0.322668 LR 0.001000 Time 0.026095 +2023-10-05 20:59:56,689 - Epoch: [38][ 160/ 1236] Overall Loss 0.325016 Objective Loss 0.325016 LR 0.001000 Time 0.025731 +2023-10-05 20:59:56,891 - Epoch: [38][ 170/ 1236] Overall Loss 0.325150 Objective Loss 0.325150 LR 0.001000 Time 0.025404 +2023-10-05 20:59:57,094 - Epoch: [38][ 180/ 1236] Overall Loss 0.322921 Objective Loss 0.322921 LR 0.001000 Time 0.025118 +2023-10-05 20:59:57,297 - Epoch: [38][ 190/ 1236] Overall Loss 0.322981 Objective Loss 0.322981 LR 0.001000 Time 0.024861 +2023-10-05 20:59:57,500 - Epoch: [38][ 200/ 1236] Overall Loss 0.323314 Objective Loss 0.323314 LR 0.001000 Time 0.024632 +2023-10-05 20:59:57,702 - Epoch: [38][ 210/ 1236] Overall Loss 0.323840 Objective Loss 0.323840 LR 0.001000 Time 0.024421 +2023-10-05 20:59:57,905 - Epoch: [38][ 220/ 1236] Overall Loss 0.324295 Objective Loss 0.324295 LR 0.001000 Time 0.024232 +2023-10-05 20:59:58,108 - Epoch: [38][ 230/ 1236] Overall Loss 0.326033 Objective Loss 0.326033 LR 0.001000 Time 0.024056 +2023-10-05 20:59:58,311 - Epoch: [38][ 240/ 1236] Overall Loss 0.326050 Objective Loss 0.326050 LR 0.001000 Time 0.023899 +2023-10-05 20:59:58,513 - Epoch: [38][ 250/ 1236] Overall Loss 0.324315 Objective Loss 0.324315 LR 0.001000 Time 0.023752 +2023-10-05 20:59:58,716 - Epoch: [38][ 260/ 1236] Overall Loss 0.324460 Objective Loss 0.324460 LR 0.001000 Time 0.023618 +2023-10-05 20:59:58,917 - Epoch: [38][ 270/ 1236] Overall Loss 0.322970 Objective Loss 0.322970 LR 0.001000 Time 0.023487 +2023-10-05 20:59:59,120 - Epoch: [38][ 280/ 1236] Overall Loss 0.323022 Objective Loss 0.323022 LR 0.001000 Time 0.023371 +2023-10-05 20:59:59,322 - Epoch: [38][ 290/ 1236] Overall Loss 0.323037 Objective Loss 0.323037 LR 0.001000 Time 0.023261 +2023-10-05 20:59:59,525 - Epoch: [38][ 300/ 1236] Overall Loss 0.322185 Objective Loss 0.322185 LR 0.001000 Time 0.023161 +2023-10-05 20:59:59,727 - Epoch: [38][ 310/ 1236] Overall Loss 0.322527 Objective Loss 0.322527 LR 0.001000 Time 0.023065 +2023-10-05 20:59:59,931 - Epoch: [38][ 320/ 1236] Overall Loss 0.322414 Objective Loss 0.322414 LR 0.001000 Time 0.022978 +2023-10-05 21:00:00,132 - Epoch: [38][ 330/ 1236] Overall Loss 0.323147 Objective Loss 0.323147 LR 0.001000 Time 0.022893 +2023-10-05 21:00:00,332 - Epoch: [38][ 340/ 1236] Overall Loss 0.323370 Objective Loss 0.323370 LR 0.001000 Time 0.022806 +2023-10-05 21:00:00,532 - Epoch: [38][ 350/ 1236] Overall Loss 0.323515 Objective Loss 0.323515 LR 0.001000 Time 0.022724 +2023-10-05 21:00:00,734 - Epoch: [38][ 360/ 1236] Overall Loss 0.323028 Objective Loss 0.323028 LR 0.001000 Time 0.022652 +2023-10-05 21:00:00,934 - Epoch: [38][ 370/ 1236] Overall Loss 0.322169 Objective Loss 0.322169 LR 0.001000 Time 0.022581 +2023-10-05 21:00:01,135 - Epoch: [38][ 380/ 1236] Overall Loss 0.320751 Objective Loss 0.320751 LR 0.001000 Time 0.022513 +2023-10-05 21:00:01,335 - Epoch: [38][ 390/ 1236] Overall Loss 0.320845 Objective Loss 0.320845 LR 0.001000 Time 0.022449 +2023-10-05 21:00:01,536 - Epoch: [38][ 400/ 1236] Overall Loss 0.320873 Objective Loss 0.320873 LR 0.001000 Time 0.022389 +2023-10-05 21:00:01,736 - Epoch: [38][ 410/ 1236] Overall Loss 0.321029 Objective Loss 0.321029 LR 0.001000 Time 0.022330 +2023-10-05 21:00:01,937 - Epoch: [38][ 420/ 1236] Overall Loss 0.321150 Objective Loss 0.321150 LR 0.001000 Time 0.022276 +2023-10-05 21:00:02,137 - Epoch: [38][ 430/ 1236] Overall Loss 0.320923 Objective Loss 0.320923 LR 0.001000 Time 0.022224 +2023-10-05 21:00:02,340 - Epoch: [38][ 440/ 1236] Overall Loss 0.320706 Objective Loss 0.320706 LR 0.001000 Time 0.022178 +2023-10-05 21:00:02,545 - Epoch: [38][ 450/ 1236] Overall Loss 0.320454 Objective Loss 0.320454 LR 0.001000 Time 0.022139 +2023-10-05 21:00:02,749 - Epoch: [38][ 460/ 1236] Overall Loss 0.320281 Objective Loss 0.320281 LR 0.001000 Time 0.022102 +2023-10-05 21:00:02,954 - Epoch: [38][ 470/ 1236] Overall Loss 0.319948 Objective Loss 0.319948 LR 0.001000 Time 0.022066 +2023-10-05 21:00:03,159 - Epoch: [38][ 480/ 1236] Overall Loss 0.319509 Objective Loss 0.319509 LR 0.001000 Time 0.022033 +2023-10-05 21:00:03,364 - Epoch: [38][ 490/ 1236] Overall Loss 0.319555 Objective Loss 0.319555 LR 0.001000 Time 0.022001 +2023-10-05 21:00:03,568 - Epoch: [38][ 500/ 1236] Overall Loss 0.319730 Objective Loss 0.319730 LR 0.001000 Time 0.021970 +2023-10-05 21:00:03,773 - Epoch: [38][ 510/ 1236] Overall Loss 0.319024 Objective Loss 0.319024 LR 0.001000 Time 0.021939 +2023-10-05 21:00:03,973 - Epoch: [38][ 520/ 1236] Overall Loss 0.318613 Objective Loss 0.318613 LR 0.001000 Time 0.021900 +2023-10-05 21:00:04,172 - Epoch: [38][ 530/ 1236] Overall Loss 0.318612 Objective Loss 0.318612 LR 0.001000 Time 0.021863 +2023-10-05 21:00:04,372 - Epoch: [38][ 540/ 1236] Overall Loss 0.318072 Objective Loss 0.318072 LR 0.001000 Time 0.021828 +2023-10-05 21:00:04,572 - Epoch: [38][ 550/ 1236] Overall Loss 0.318002 Objective Loss 0.318002 LR 0.001000 Time 0.021795 +2023-10-05 21:00:04,775 - Epoch: [38][ 560/ 1236] Overall Loss 0.317760 Objective Loss 0.317760 LR 0.001000 Time 0.021767 +2023-10-05 21:00:04,979 - Epoch: [38][ 570/ 1236] Overall Loss 0.317977 Objective Loss 0.317977 LR 0.001000 Time 0.021743 +2023-10-05 21:00:05,184 - Epoch: [38][ 580/ 1236] Overall Loss 0.317860 Objective Loss 0.317860 LR 0.001000 Time 0.021720 +2023-10-05 21:00:05,389 - Epoch: [38][ 590/ 1236] Overall Loss 0.317581 Objective Loss 0.317581 LR 0.001000 Time 0.021699 +2023-10-05 21:00:05,594 - Epoch: [38][ 600/ 1236] Overall Loss 0.317706 Objective Loss 0.317706 LR 0.001000 Time 0.021678 +2023-10-05 21:00:05,798 - Epoch: [38][ 610/ 1236] Overall Loss 0.317727 Objective Loss 0.317727 LR 0.001000 Time 0.021657 +2023-10-05 21:00:06,003 - Epoch: [38][ 620/ 1236] Overall Loss 0.317301 Objective Loss 0.317301 LR 0.001000 Time 0.021638 +2023-10-05 21:00:06,208 - Epoch: [38][ 630/ 1236] Overall Loss 0.317171 Objective Loss 0.317171 LR 0.001000 Time 0.021619 +2023-10-05 21:00:06,412 - Epoch: [38][ 640/ 1236] Overall Loss 0.317242 Objective Loss 0.317242 LR 0.001000 Time 0.021600 +2023-10-05 21:00:06,617 - Epoch: [38][ 650/ 1236] Overall Loss 0.317396 Objective Loss 0.317396 LR 0.001000 Time 0.021582 +2023-10-05 21:00:06,822 - Epoch: [38][ 660/ 1236] Overall Loss 0.317185 Objective Loss 0.317185 LR 0.001000 Time 0.021565 +2023-10-05 21:00:07,026 - Epoch: [38][ 670/ 1236] Overall Loss 0.317373 Objective Loss 0.317373 LR 0.001000 Time 0.021547 +2023-10-05 21:00:07,231 - Epoch: [38][ 680/ 1236] Overall Loss 0.317945 Objective Loss 0.317945 LR 0.001000 Time 0.021531 +2023-10-05 21:00:07,435 - Epoch: [38][ 690/ 1236] Overall Loss 0.317795 Objective Loss 0.317795 LR 0.001000 Time 0.021515 +2023-10-05 21:00:07,637 - Epoch: [38][ 700/ 1236] Overall Loss 0.318295 Objective Loss 0.318295 LR 0.001000 Time 0.021495 +2023-10-05 21:00:07,837 - Epoch: [38][ 710/ 1236] Overall Loss 0.318506 Objective Loss 0.318506 LR 0.001000 Time 0.021474 +2023-10-05 21:00:08,037 - Epoch: [38][ 720/ 1236] Overall Loss 0.318585 Objective Loss 0.318585 LR 0.001000 Time 0.021452 +2023-10-05 21:00:08,236 - Epoch: [38][ 730/ 1236] Overall Loss 0.317946 Objective Loss 0.317946 LR 0.001000 Time 0.021430 +2023-10-05 21:00:08,437 - Epoch: [38][ 740/ 1236] Overall Loss 0.318138 Objective Loss 0.318138 LR 0.001000 Time 0.021412 +2023-10-05 21:00:08,640 - Epoch: [38][ 750/ 1236] Overall Loss 0.318648 Objective Loss 0.318648 LR 0.001000 Time 0.021398 +2023-10-05 21:00:08,845 - Epoch: [38][ 760/ 1236] Overall Loss 0.318737 Objective Loss 0.318737 LR 0.001000 Time 0.021385 +2023-10-05 21:00:09,050 - Epoch: [38][ 770/ 1236] Overall Loss 0.319099 Objective Loss 0.319099 LR 0.001000 Time 0.021373 +2023-10-05 21:00:09,255 - Epoch: [38][ 780/ 1236] Overall Loss 0.319305 Objective Loss 0.319305 LR 0.001000 Time 0.021362 +2023-10-05 21:00:09,460 - Epoch: [38][ 790/ 1236] Overall Loss 0.319250 Objective Loss 0.319250 LR 0.001000 Time 0.021350 +2023-10-05 21:00:09,665 - Epoch: [38][ 800/ 1236] Overall Loss 0.319090 Objective Loss 0.319090 LR 0.001000 Time 0.021339 +2023-10-05 21:00:09,869 - Epoch: [38][ 810/ 1236] Overall Loss 0.319078 Objective Loss 0.319078 LR 0.001000 Time 0.021328 +2023-10-05 21:00:10,074 - Epoch: [38][ 820/ 1236] Overall Loss 0.319154 Objective Loss 0.319154 LR 0.001000 Time 0.021317 +2023-10-05 21:00:10,279 - Epoch: [38][ 830/ 1236] Overall Loss 0.319305 Objective Loss 0.319305 LR 0.001000 Time 0.021306 +2023-10-05 21:00:10,484 - Epoch: [38][ 840/ 1236] Overall Loss 0.319577 Objective Loss 0.319577 LR 0.001000 Time 0.021296 +2023-10-05 21:00:10,689 - Epoch: [38][ 850/ 1236] Overall Loss 0.319533 Objective Loss 0.319533 LR 0.001000 Time 0.021286 +2023-10-05 21:00:10,893 - Epoch: [38][ 860/ 1236] Overall Loss 0.320086 Objective Loss 0.320086 LR 0.001000 Time 0.021276 +2023-10-05 21:00:11,098 - Epoch: [38][ 870/ 1236] Overall Loss 0.320309 Objective Loss 0.320309 LR 0.001000 Time 0.021266 +2023-10-05 21:00:11,303 - Epoch: [38][ 880/ 1236] Overall Loss 0.320604 Objective Loss 0.320604 LR 0.001000 Time 0.021257 +2023-10-05 21:00:11,507 - Epoch: [38][ 890/ 1236] Overall Loss 0.321185 Objective Loss 0.321185 LR 0.001000 Time 0.021248 +2023-10-05 21:00:11,712 - Epoch: [38][ 900/ 1236] Overall Loss 0.321264 Objective Loss 0.321264 LR 0.001000 Time 0.021238 +2023-10-05 21:00:11,916 - Epoch: [38][ 910/ 1236] Overall Loss 0.321281 Objective Loss 0.321281 LR 0.001000 Time 0.021230 +2023-10-05 21:00:12,121 - Epoch: [38][ 920/ 1236] Overall Loss 0.321221 Objective Loss 0.321221 LR 0.001000 Time 0.021221 +2023-10-05 21:00:12,326 - Epoch: [38][ 930/ 1236] Overall Loss 0.321394 Objective Loss 0.321394 LR 0.001000 Time 0.021212 +2023-10-05 21:00:12,531 - Epoch: [38][ 940/ 1236] Overall Loss 0.321614 Objective Loss 0.321614 LR 0.001000 Time 0.021204 +2023-10-05 21:00:12,735 - Epoch: [38][ 950/ 1236] Overall Loss 0.321521 Objective Loss 0.321521 LR 0.001000 Time 0.021196 +2023-10-05 21:00:12,940 - Epoch: [38][ 960/ 1236] Overall Loss 0.322013 Objective Loss 0.322013 LR 0.001000 Time 0.021188 +2023-10-05 21:00:13,145 - Epoch: [38][ 970/ 1236] Overall Loss 0.321872 Objective Loss 0.321872 LR 0.001000 Time 0.021181 +2023-10-05 21:00:13,350 - Epoch: [38][ 980/ 1236] Overall Loss 0.321847 Objective Loss 0.321847 LR 0.001000 Time 0.021173 +2023-10-05 21:00:13,555 - Epoch: [38][ 990/ 1236] Overall Loss 0.322066 Objective Loss 0.322066 LR 0.001000 Time 0.021166 +2023-10-05 21:00:13,760 - Epoch: [38][ 1000/ 1236] Overall Loss 0.322262 Objective Loss 0.322262 LR 0.001000 Time 0.021159 +2023-10-05 21:00:13,965 - Epoch: [38][ 1010/ 1236] Overall Loss 0.322122 Objective Loss 0.322122 LR 0.001000 Time 0.021152 +2023-10-05 21:00:14,170 - Epoch: [38][ 1020/ 1236] Overall Loss 0.321979 Objective Loss 0.321979 LR 0.001000 Time 0.021145 +2023-10-05 21:00:14,375 - Epoch: [38][ 1030/ 1236] Overall Loss 0.321982 Objective Loss 0.321982 LR 0.001000 Time 0.021139 +2023-10-05 21:00:14,579 - Epoch: [38][ 1040/ 1236] Overall Loss 0.321909 Objective Loss 0.321909 LR 0.001000 Time 0.021132 +2023-10-05 21:00:14,784 - Epoch: [38][ 1050/ 1236] Overall Loss 0.322032 Objective Loss 0.322032 LR 0.001000 Time 0.021125 +2023-10-05 21:00:14,989 - Epoch: [38][ 1060/ 1236] Overall Loss 0.322104 Objective Loss 0.322104 LR 0.001000 Time 0.021119 +2023-10-05 21:00:15,194 - Epoch: [38][ 1070/ 1236] Overall Loss 0.322250 Objective Loss 0.322250 LR 0.001000 Time 0.021113 +2023-10-05 21:00:15,399 - Epoch: [38][ 1080/ 1236] Overall Loss 0.322688 Objective Loss 0.322688 LR 0.001000 Time 0.021107 +2023-10-05 21:00:15,604 - Epoch: [38][ 1090/ 1236] Overall Loss 0.322627 Objective Loss 0.322627 LR 0.001000 Time 0.021101 +2023-10-05 21:00:15,809 - Epoch: [38][ 1100/ 1236] Overall Loss 0.323135 Objective Loss 0.323135 LR 0.001000 Time 0.021095 +2023-10-05 21:00:16,014 - Epoch: [38][ 1110/ 1236] Overall Loss 0.323010 Objective Loss 0.323010 LR 0.001000 Time 0.021089 +2023-10-05 21:00:16,219 - Epoch: [38][ 1120/ 1236] Overall Loss 0.322843 Objective Loss 0.322843 LR 0.001000 Time 0.021084 +2023-10-05 21:00:16,424 - Epoch: [38][ 1130/ 1236] Overall Loss 0.322817 Objective Loss 0.322817 LR 0.001000 Time 0.021078 +2023-10-05 21:00:16,629 - Epoch: [38][ 1140/ 1236] Overall Loss 0.323058 Objective Loss 0.323058 LR 0.001000 Time 0.021073 +2023-10-05 21:00:16,834 - Epoch: [38][ 1150/ 1236] Overall Loss 0.323113 Objective Loss 0.323113 LR 0.001000 Time 0.021068 +2023-10-05 21:00:17,039 - Epoch: [38][ 1160/ 1236] Overall Loss 0.323320 Objective Loss 0.323320 LR 0.001000 Time 0.021062 +2023-10-05 21:00:17,244 - Epoch: [38][ 1170/ 1236] Overall Loss 0.323216 Objective Loss 0.323216 LR 0.001000 Time 0.021057 +2023-10-05 21:00:17,449 - Epoch: [38][ 1180/ 1236] Overall Loss 0.323029 Objective Loss 0.323029 LR 0.001000 Time 0.021052 +2023-10-05 21:00:17,654 - Epoch: [38][ 1190/ 1236] Overall Loss 0.323078 Objective Loss 0.323078 LR 0.001000 Time 0.021047 +2023-10-05 21:00:17,859 - Epoch: [38][ 1200/ 1236] Overall Loss 0.323206 Objective Loss 0.323206 LR 0.001000 Time 0.021043 +2023-10-05 21:00:18,064 - Epoch: [38][ 1210/ 1236] Overall Loss 0.323526 Objective Loss 0.323526 LR 0.001000 Time 0.021038 +2023-10-05 21:00:18,269 - Epoch: [38][ 1220/ 1236] Overall Loss 0.323434 Objective Loss 0.323434 LR 0.001000 Time 0.021033 +2023-10-05 21:00:18,525 - Epoch: [38][ 1230/ 1236] Overall Loss 0.323552 Objective Loss 0.323552 LR 0.001000 Time 0.021070 +2023-10-05 21:00:18,642 - Epoch: [38][ 1236/ 1236] Overall Loss 0.323691 Objective Loss 0.323691 Top1 84.317719 Top5 97.148676 LR 0.001000 Time 0.021062 +2023-10-05 21:00:18,780 - --- validate (epoch=38)----------- +2023-10-05 21:00:18,780 - 29943 samples (256 per mini-batch) +2023-10-05 21:00:19,233 - Epoch: [38][ 10/ 117] Loss 0.351041 Top1 82.031250 Top5 97.773438 +2023-10-05 21:00:19,381 - Epoch: [38][ 20/ 117] Loss 0.350146 Top1 82.910156 Top5 97.695312 +2023-10-05 21:00:19,528 - Epoch: [38][ 30/ 117] Loss 0.352728 Top1 82.591146 Top5 97.565104 +2023-10-05 21:00:19,676 - Epoch: [38][ 40/ 117] Loss 0.358903 Top1 82.646484 Top5 97.587891 +2023-10-05 21:00:19,826 - Epoch: [38][ 50/ 117] Loss 0.352685 Top1 82.609375 Top5 97.648438 +2023-10-05 21:00:19,976 - Epoch: [38][ 60/ 117] Loss 0.360372 Top1 82.513021 Top5 97.591146 +2023-10-05 21:00:20,130 - Epoch: [38][ 70/ 117] Loss 0.362226 Top1 82.360491 Top5 97.628348 +2023-10-05 21:00:20,285 - Epoch: [38][ 80/ 117] Loss 0.367796 Top1 82.138672 Top5 97.553711 +2023-10-05 21:00:20,440 - Epoch: [38][ 90/ 117] Loss 0.366441 Top1 82.135417 Top5 97.526042 +2023-10-05 21:00:20,595 - Epoch: [38][ 100/ 117] Loss 0.363668 Top1 82.175781 Top5 97.562500 +2023-10-05 21:00:20,758 - Epoch: [38][ 110/ 117] Loss 0.359847 Top1 82.173295 Top5 97.585227 +2023-10-05 21:00:20,844 - Epoch: [38][ 117/ 117] Loss 0.362480 Top1 82.116020 Top5 97.562035 +2023-10-05 21:00:20,984 - ==> Top1: 82.116 Top5: 97.562 Loss: 0.362 + +2023-10-05 21:00:20,985 - ==> Confusion: +[[ 913 5 3 3 14 1 1 0 10 61 3 0 0 2 9 5 2 1 2 1 14] + [ 3 1043 0 1 16 15 2 25 0 0 4 1 0 0 4 3 6 0 5 0 3] + [ 4 1 920 26 2 0 28 10 0 1 13 5 7 3 0 5 0 1 8 3 19] + [ 5 2 12 956 2 5 2 0 1 0 6 0 8 2 37 5 1 6 24 1 14] + [ 25 10 1 0 971 2 0 0 2 5 1 3 1 0 4 3 12 3 1 3 3] + [ 3 64 3 1 8 945 3 25 4 5 4 8 2 21 2 0 4 1 2 3 8] + [ 1 8 14 2 0 0 1123 9 0 0 10 3 2 1 0 9 0 0 1 6 2] + [ 3 30 11 1 4 32 3 1061 0 5 2 10 2 1 0 4 0 1 29 8 11] + [ 14 9 0 1 2 3 0 0 930 54 7 1 2 7 38 4 1 0 13 3 0] + [ 107 0 1 0 10 2 2 0 21 921 0 2 0 27 7 5 1 1 1 2 9] + [ 2 9 8 4 3 3 6 1 15 1 971 4 0 7 4 1 1 0 4 0 9] + [ 1 0 1 1 1 21 0 0 0 0 0 956 14 3 1 5 5 14 0 6 6] + [ 0 1 2 5 0 2 1 1 0 0 0 50 949 2 2 14 3 17 4 4 11] + [ 1 0 1 0 7 5 0 1 11 21 11 7 1 1029 3 3 1 1 0 3 13] + [ 15 3 3 16 9 0 0 0 26 5 2 1 0 0 993 0 1 2 6 0 19] + [ 0 3 2 3 3 1 4 0 0 0 0 8 4 1 0 1060 14 12 1 10 8] + [ 1 21 0 0 3 4 1 0 2 0 0 4 1 1 4 15 1088 0 0 4 12] + [ 0 0 0 1 0 1 2 0 0 0 0 7 28 1 4 6 1 979 1 0 7] + [ 1 17 10 19 1 1 0 26 3 0 4 0 4 0 16 0 1 0 954 1 10] + [ 0 0 3 1 1 10 10 18 0 1 5 14 5 4 0 7 10 1 2 1050 10] + [ 127 250 125 91 140 158 54 92 110 113 177 137 398 288 182 95 142 81 184 185 4776]] + +2023-10-05 21:00:20,986 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:00:20,986 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:00:20,993 - + +2023-10-05 21:00:20,993 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:00:21,994 - Epoch: [39][ 10/ 1236] Overall Loss 0.345607 Objective Loss 0.345607 LR 0.001000 Time 0.100023 +2023-10-05 21:00:22,200 - Epoch: [39][ 20/ 1236] Overall Loss 0.323087 Objective Loss 0.323087 LR 0.001000 Time 0.060270 +2023-10-05 21:00:22,402 - Epoch: [39][ 30/ 1236] Overall Loss 0.324463 Objective Loss 0.324463 LR 0.001000 Time 0.046909 +2023-10-05 21:00:22,607 - Epoch: [39][ 40/ 1236] Overall Loss 0.322609 Objective Loss 0.322609 LR 0.001000 Time 0.040311 +2023-10-05 21:00:22,809 - Epoch: [39][ 50/ 1236] Overall Loss 0.321149 Objective Loss 0.321149 LR 0.001000 Time 0.036275 +2023-10-05 21:00:23,015 - Epoch: [39][ 60/ 1236] Overall Loss 0.319576 Objective Loss 0.319576 LR 0.001000 Time 0.033652 +2023-10-05 21:00:23,217 - Epoch: [39][ 70/ 1236] Overall Loss 0.324169 Objective Loss 0.324169 LR 0.001000 Time 0.031725 +2023-10-05 21:00:23,422 - Epoch: [39][ 80/ 1236] Overall Loss 0.324992 Objective Loss 0.324992 LR 0.001000 Time 0.030317 +2023-10-05 21:00:23,624 - Epoch: [39][ 90/ 1236] Overall Loss 0.329694 Objective Loss 0.329694 LR 0.001000 Time 0.029190 +2023-10-05 21:00:23,829 - Epoch: [39][ 100/ 1236] Overall Loss 0.328454 Objective Loss 0.328454 LR 0.001000 Time 0.028318 +2023-10-05 21:00:24,031 - Epoch: [39][ 110/ 1236] Overall Loss 0.331255 Objective Loss 0.331255 LR 0.001000 Time 0.027578 +2023-10-05 21:00:24,236 - Epoch: [39][ 120/ 1236] Overall Loss 0.326056 Objective Loss 0.326056 LR 0.001000 Time 0.026988 +2023-10-05 21:00:24,439 - Epoch: [39][ 130/ 1236] Overall Loss 0.325228 Objective Loss 0.325228 LR 0.001000 Time 0.026468 +2023-10-05 21:00:24,643 - Epoch: [39][ 140/ 1236] Overall Loss 0.323327 Objective Loss 0.323327 LR 0.001000 Time 0.026031 +2023-10-05 21:00:24,845 - Epoch: [39][ 150/ 1236] Overall Loss 0.321431 Objective Loss 0.321431 LR 0.001000 Time 0.025641 +2023-10-05 21:00:25,049 - Epoch: [39][ 160/ 1236] Overall Loss 0.319826 Objective Loss 0.319826 LR 0.001000 Time 0.025315 +2023-10-05 21:00:25,252 - Epoch: [39][ 170/ 1236] Overall Loss 0.319466 Objective Loss 0.319466 LR 0.001000 Time 0.025014 +2023-10-05 21:00:25,456 - Epoch: [39][ 180/ 1236] Overall Loss 0.320263 Objective Loss 0.320263 LR 0.001000 Time 0.024758 +2023-10-05 21:00:25,659 - Epoch: [39][ 190/ 1236] Overall Loss 0.321448 Objective Loss 0.321448 LR 0.001000 Time 0.024520 +2023-10-05 21:00:25,862 - Epoch: [39][ 200/ 1236] Overall Loss 0.321892 Objective Loss 0.321892 LR 0.001000 Time 0.024307 +2023-10-05 21:00:26,064 - Epoch: [39][ 210/ 1236] Overall Loss 0.319704 Objective Loss 0.319704 LR 0.001000 Time 0.024112 +2023-10-05 21:00:26,269 - Epoch: [39][ 220/ 1236] Overall Loss 0.318822 Objective Loss 0.318822 LR 0.001000 Time 0.023946 +2023-10-05 21:00:26,471 - Epoch: [39][ 230/ 1236] Overall Loss 0.318303 Objective Loss 0.318303 LR 0.001000 Time 0.023781 +2023-10-05 21:00:26,676 - Epoch: [39][ 240/ 1236] Overall Loss 0.318687 Objective Loss 0.318687 LR 0.001000 Time 0.023643 +2023-10-05 21:00:26,878 - Epoch: [39][ 250/ 1236] Overall Loss 0.319134 Objective Loss 0.319134 LR 0.001000 Time 0.023505 +2023-10-05 21:00:27,083 - Epoch: [39][ 260/ 1236] Overall Loss 0.317126 Objective Loss 0.317126 LR 0.001000 Time 0.023387 +2023-10-05 21:00:27,285 - Epoch: [39][ 270/ 1236] Overall Loss 0.316918 Objective Loss 0.316918 LR 0.001000 Time 0.023269 +2023-10-05 21:00:27,490 - Epoch: [39][ 280/ 1236] Overall Loss 0.316163 Objective Loss 0.316163 LR 0.001000 Time 0.023168 +2023-10-05 21:00:27,692 - Epoch: [39][ 290/ 1236] Overall Loss 0.316299 Objective Loss 0.316299 LR 0.001000 Time 0.023065 +2023-10-05 21:00:27,897 - Epoch: [39][ 300/ 1236] Overall Loss 0.316874 Objective Loss 0.316874 LR 0.001000 Time 0.022977 +2023-10-05 21:00:28,100 - Epoch: [39][ 310/ 1236] Overall Loss 0.317066 Objective Loss 0.317066 LR 0.001000 Time 0.022889 +2023-10-05 21:00:28,304 - Epoch: [39][ 320/ 1236] Overall Loss 0.317893 Objective Loss 0.317893 LR 0.001000 Time 0.022812 +2023-10-05 21:00:28,507 - Epoch: [39][ 330/ 1236] Overall Loss 0.318113 Objective Loss 0.318113 LR 0.001000 Time 0.022733 +2023-10-05 21:00:28,713 - Epoch: [39][ 340/ 1236] Overall Loss 0.317788 Objective Loss 0.317788 LR 0.001000 Time 0.022670 +2023-10-05 21:00:28,915 - Epoch: [39][ 350/ 1236] Overall Loss 0.318229 Objective Loss 0.318229 LR 0.001000 Time 0.022600 +2023-10-05 21:00:29,121 - Epoch: [39][ 360/ 1236] Overall Loss 0.318257 Objective Loss 0.318257 LR 0.001000 Time 0.022543 +2023-10-05 21:00:29,324 - Epoch: [39][ 370/ 1236] Overall Loss 0.318359 Objective Loss 0.318359 LR 0.001000 Time 0.022481 +2023-10-05 21:00:29,530 - Epoch: [39][ 380/ 1236] Overall Loss 0.319243 Objective Loss 0.319243 LR 0.001000 Time 0.022431 +2023-10-05 21:00:29,733 - Epoch: [39][ 390/ 1236] Overall Loss 0.318776 Objective Loss 0.318776 LR 0.001000 Time 0.022374 +2023-10-05 21:00:29,939 - Epoch: [39][ 400/ 1236] Overall Loss 0.319661 Objective Loss 0.319661 LR 0.001000 Time 0.022329 +2023-10-05 21:00:30,141 - Epoch: [39][ 410/ 1236] Overall Loss 0.319212 Objective Loss 0.319212 LR 0.001000 Time 0.022277 +2023-10-05 21:00:30,347 - Epoch: [39][ 420/ 1236] Overall Loss 0.318679 Objective Loss 0.318679 LR 0.001000 Time 0.022237 +2023-10-05 21:00:30,550 - Epoch: [39][ 430/ 1236] Overall Loss 0.318913 Objective Loss 0.318913 LR 0.001000 Time 0.022190 +2023-10-05 21:00:30,756 - Epoch: [39][ 440/ 1236] Overall Loss 0.319439 Objective Loss 0.319439 LR 0.001000 Time 0.022154 +2023-10-05 21:00:30,959 - Epoch: [39][ 450/ 1236] Overall Loss 0.319869 Objective Loss 0.319869 LR 0.001000 Time 0.022111 +2023-10-05 21:00:31,165 - Epoch: [39][ 460/ 1236] Overall Loss 0.319953 Objective Loss 0.319953 LR 0.001000 Time 0.022078 +2023-10-05 21:00:31,368 - Epoch: [39][ 470/ 1236] Overall Loss 0.320611 Objective Loss 0.320611 LR 0.001000 Time 0.022038 +2023-10-05 21:00:31,574 - Epoch: [39][ 480/ 1236] Overall Loss 0.320795 Objective Loss 0.320795 LR 0.001000 Time 0.022008 +2023-10-05 21:00:31,776 - Epoch: [39][ 490/ 1236] Overall Loss 0.320737 Objective Loss 0.320737 LR 0.001000 Time 0.021971 +2023-10-05 21:00:31,983 - Epoch: [39][ 500/ 1236] Overall Loss 0.320912 Objective Loss 0.320912 LR 0.001000 Time 0.021943 +2023-10-05 21:00:32,185 - Epoch: [39][ 510/ 1236] Overall Loss 0.321140 Objective Loss 0.321140 LR 0.001000 Time 0.021909 +2023-10-05 21:00:32,391 - Epoch: [39][ 520/ 1236] Overall Loss 0.320786 Objective Loss 0.320786 LR 0.001000 Time 0.021883 +2023-10-05 21:00:32,594 - Epoch: [39][ 530/ 1236] Overall Loss 0.321274 Objective Loss 0.321274 LR 0.001000 Time 0.021852 +2023-10-05 21:00:32,800 - Epoch: [39][ 540/ 1236] Overall Loss 0.321516 Objective Loss 0.321516 LR 0.001000 Time 0.021828 +2023-10-05 21:00:33,002 - Epoch: [39][ 550/ 1236] Overall Loss 0.321402 Objective Loss 0.321402 LR 0.001000 Time 0.021799 +2023-10-05 21:00:33,209 - Epoch: [39][ 560/ 1236] Overall Loss 0.321304 Objective Loss 0.321304 LR 0.001000 Time 0.021777 +2023-10-05 21:00:33,411 - Epoch: [39][ 570/ 1236] Overall Loss 0.320963 Objective Loss 0.320963 LR 0.001000 Time 0.021750 +2023-10-05 21:00:33,617 - Epoch: [39][ 580/ 1236] Overall Loss 0.320784 Objective Loss 0.320784 LR 0.001000 Time 0.021729 +2023-10-05 21:00:33,820 - Epoch: [39][ 590/ 1236] Overall Loss 0.320979 Objective Loss 0.320979 LR 0.001000 Time 0.021704 +2023-10-05 21:00:34,026 - Epoch: [39][ 600/ 1236] Overall Loss 0.321513 Objective Loss 0.321513 LR 0.001000 Time 0.021686 +2023-10-05 21:00:34,229 - Epoch: [39][ 610/ 1236] Overall Loss 0.321292 Objective Loss 0.321292 LR 0.001000 Time 0.021662 +2023-10-05 21:00:34,435 - Epoch: [39][ 620/ 1236] Overall Loss 0.321479 Objective Loss 0.321479 LR 0.001000 Time 0.021644 +2023-10-05 21:00:34,637 - Epoch: [39][ 630/ 1236] Overall Loss 0.321099 Objective Loss 0.321099 LR 0.001000 Time 0.021621 +2023-10-05 21:00:34,843 - Epoch: [39][ 640/ 1236] Overall Loss 0.320986 Objective Loss 0.320986 LR 0.001000 Time 0.021605 +2023-10-05 21:00:35,046 - Epoch: [39][ 650/ 1236] Overall Loss 0.321195 Objective Loss 0.321195 LR 0.001000 Time 0.021584 +2023-10-05 21:00:35,253 - Epoch: [39][ 660/ 1236] Overall Loss 0.320936 Objective Loss 0.320936 LR 0.001000 Time 0.021569 +2023-10-05 21:00:35,455 - Epoch: [39][ 670/ 1236] Overall Loss 0.321328 Objective Loss 0.321328 LR 0.001000 Time 0.021549 +2023-10-05 21:00:35,661 - Epoch: [39][ 680/ 1236] Overall Loss 0.321186 Objective Loss 0.321186 LR 0.001000 Time 0.021535 +2023-10-05 21:00:35,864 - Epoch: [39][ 690/ 1236] Overall Loss 0.320651 Objective Loss 0.320651 LR 0.001000 Time 0.021516 +2023-10-05 21:00:36,070 - Epoch: [39][ 700/ 1236] Overall Loss 0.321138 Objective Loss 0.321138 LR 0.001000 Time 0.021502 +2023-10-05 21:00:36,273 - Epoch: [39][ 710/ 1236] Overall Loss 0.320957 Objective Loss 0.320957 LR 0.001000 Time 0.021485 +2023-10-05 21:00:36,479 - Epoch: [39][ 720/ 1236] Overall Loss 0.320935 Objective Loss 0.320935 LR 0.001000 Time 0.021472 +2023-10-05 21:00:36,682 - Epoch: [39][ 730/ 1236] Overall Loss 0.321043 Objective Loss 0.321043 LR 0.001000 Time 0.021455 +2023-10-05 21:00:36,888 - Epoch: [39][ 740/ 1236] Overall Loss 0.321637 Objective Loss 0.321637 LR 0.001000 Time 0.021443 +2023-10-05 21:00:37,090 - Epoch: [39][ 750/ 1236] Overall Loss 0.322101 Objective Loss 0.322101 LR 0.001000 Time 0.021427 +2023-10-05 21:00:37,297 - Epoch: [39][ 760/ 1236] Overall Loss 0.322136 Objective Loss 0.322136 LR 0.001000 Time 0.021416 +2023-10-05 21:00:37,499 - Epoch: [39][ 770/ 1236] Overall Loss 0.322165 Objective Loss 0.322165 LR 0.001000 Time 0.021400 +2023-10-05 21:00:37,705 - Epoch: [39][ 780/ 1236] Overall Loss 0.322200 Objective Loss 0.322200 LR 0.001000 Time 0.021390 +2023-10-05 21:00:37,908 - Epoch: [39][ 790/ 1236] Overall Loss 0.322340 Objective Loss 0.322340 LR 0.001000 Time 0.021375 +2023-10-05 21:00:38,114 - Epoch: [39][ 800/ 1236] Overall Loss 0.322092 Objective Loss 0.322092 LR 0.001000 Time 0.021365 +2023-10-05 21:00:38,317 - Epoch: [39][ 810/ 1236] Overall Loss 0.322297 Objective Loss 0.322297 LR 0.001000 Time 0.021351 +2023-10-05 21:00:38,523 - Epoch: [39][ 820/ 1236] Overall Loss 0.322336 Objective Loss 0.322336 LR 0.001000 Time 0.021342 +2023-10-05 21:00:38,726 - Epoch: [39][ 830/ 1236] Overall Loss 0.322194 Objective Loss 0.322194 LR 0.001000 Time 0.021329 +2023-10-05 21:00:38,933 - Epoch: [39][ 840/ 1236] Overall Loss 0.322530 Objective Loss 0.322530 LR 0.001000 Time 0.021320 +2023-10-05 21:00:39,135 - Epoch: [39][ 850/ 1236] Overall Loss 0.322976 Objective Loss 0.322976 LR 0.001000 Time 0.021307 +2023-10-05 21:00:39,342 - Epoch: [39][ 860/ 1236] Overall Loss 0.323124 Objective Loss 0.323124 LR 0.001000 Time 0.021299 +2023-10-05 21:00:39,544 - Epoch: [39][ 870/ 1236] Overall Loss 0.323000 Objective Loss 0.323000 LR 0.001000 Time 0.021287 +2023-10-05 21:00:39,751 - Epoch: [39][ 880/ 1236] Overall Loss 0.323179 Objective Loss 0.323179 LR 0.001000 Time 0.021279 +2023-10-05 21:00:39,953 - Epoch: [39][ 890/ 1236] Overall Loss 0.323554 Objective Loss 0.323554 LR 0.001000 Time 0.021267 +2023-10-05 21:00:40,160 - Epoch: [39][ 900/ 1236] Overall Loss 0.323551 Objective Loss 0.323551 LR 0.001000 Time 0.021260 +2023-10-05 21:00:40,362 - Epoch: [39][ 910/ 1236] Overall Loss 0.323798 Objective Loss 0.323798 LR 0.001000 Time 0.021248 +2023-10-05 21:00:40,568 - Epoch: [39][ 920/ 1236] Overall Loss 0.323666 Objective Loss 0.323666 LR 0.001000 Time 0.021241 +2023-10-05 21:00:40,771 - Epoch: [39][ 930/ 1236] Overall Loss 0.323959 Objective Loss 0.323959 LR 0.001000 Time 0.021230 +2023-10-05 21:00:40,977 - Epoch: [39][ 940/ 1236] Overall Loss 0.324089 Objective Loss 0.324089 LR 0.001000 Time 0.021223 +2023-10-05 21:00:41,180 - Epoch: [39][ 950/ 1236] Overall Loss 0.324629 Objective Loss 0.324629 LR 0.001000 Time 0.021213 +2023-10-05 21:00:41,386 - Epoch: [39][ 960/ 1236] Overall Loss 0.324455 Objective Loss 0.324455 LR 0.001000 Time 0.021206 +2023-10-05 21:00:41,589 - Epoch: [39][ 970/ 1236] Overall Loss 0.324427 Objective Loss 0.324427 LR 0.001000 Time 0.021196 +2023-10-05 21:00:41,795 - Epoch: [39][ 980/ 1236] Overall Loss 0.324464 Objective Loss 0.324464 LR 0.001000 Time 0.021190 +2023-10-05 21:00:41,998 - Epoch: [39][ 990/ 1236] Overall Loss 0.324403 Objective Loss 0.324403 LR 0.001000 Time 0.021180 +2023-10-05 21:00:42,204 - Epoch: [39][ 1000/ 1236] Overall Loss 0.324900 Objective Loss 0.324900 LR 0.001000 Time 0.021174 +2023-10-05 21:00:42,406 - Epoch: [39][ 1010/ 1236] Overall Loss 0.325231 Objective Loss 0.325231 LR 0.001000 Time 0.021164 +2023-10-05 21:00:42,612 - Epoch: [39][ 1020/ 1236] Overall Loss 0.325190 Objective Loss 0.325190 LR 0.001000 Time 0.021159 +2023-10-05 21:00:42,815 - Epoch: [39][ 1030/ 1236] Overall Loss 0.324971 Objective Loss 0.324971 LR 0.001000 Time 0.021150 +2023-10-05 21:00:43,021 - Epoch: [39][ 1040/ 1236] Overall Loss 0.325189 Objective Loss 0.325189 LR 0.001000 Time 0.021144 +2023-10-05 21:00:43,224 - Epoch: [39][ 1050/ 1236] Overall Loss 0.325430 Objective Loss 0.325430 LR 0.001000 Time 0.021136 +2023-10-05 21:00:43,430 - Epoch: [39][ 1060/ 1236] Overall Loss 0.325780 Objective Loss 0.325780 LR 0.001000 Time 0.021130 +2023-10-05 21:00:43,632 - Epoch: [39][ 1070/ 1236] Overall Loss 0.325734 Objective Loss 0.325734 LR 0.001000 Time 0.021122 +2023-10-05 21:00:43,838 - Epoch: [39][ 1080/ 1236] Overall Loss 0.325740 Objective Loss 0.325740 LR 0.001000 Time 0.021117 +2023-10-05 21:00:44,041 - Epoch: [39][ 1090/ 1236] Overall Loss 0.325920 Objective Loss 0.325920 LR 0.001000 Time 0.021108 +2023-10-05 21:00:44,247 - Epoch: [39][ 1100/ 1236] Overall Loss 0.325867 Objective Loss 0.325867 LR 0.001000 Time 0.021103 +2023-10-05 21:00:44,450 - Epoch: [39][ 1110/ 1236] Overall Loss 0.326021 Objective Loss 0.326021 LR 0.001000 Time 0.021095 +2023-10-05 21:00:44,654 - Epoch: [39][ 1120/ 1236] Overall Loss 0.325724 Objective Loss 0.325724 LR 0.001000 Time 0.021089 +2023-10-05 21:00:44,852 - Epoch: [39][ 1130/ 1236] Overall Loss 0.325698 Objective Loss 0.325698 LR 0.001000 Time 0.021078 +2023-10-05 21:00:45,054 - Epoch: [39][ 1140/ 1236] Overall Loss 0.325656 Objective Loss 0.325656 LR 0.001000 Time 0.021070 +2023-10-05 21:00:45,254 - Epoch: [39][ 1150/ 1236] Overall Loss 0.326011 Objective Loss 0.326011 LR 0.001000 Time 0.021060 +2023-10-05 21:00:45,456 - Epoch: [39][ 1160/ 1236] Overall Loss 0.325820 Objective Loss 0.325820 LR 0.001000 Time 0.021052 +2023-10-05 21:00:45,656 - Epoch: [39][ 1170/ 1236] Overall Loss 0.325850 Objective Loss 0.325850 LR 0.001000 Time 0.021043 +2023-10-05 21:00:45,857 - Epoch: [39][ 1180/ 1236] Overall Loss 0.325754 Objective Loss 0.325754 LR 0.001000 Time 0.021035 +2023-10-05 21:00:46,057 - Epoch: [39][ 1190/ 1236] Overall Loss 0.326068 Objective Loss 0.326068 LR 0.001000 Time 0.021026 +2023-10-05 21:00:46,258 - Epoch: [39][ 1200/ 1236] Overall Loss 0.325918 Objective Loss 0.325918 LR 0.001000 Time 0.021017 +2023-10-05 21:00:46,458 - Epoch: [39][ 1210/ 1236] Overall Loss 0.326145 Objective Loss 0.326145 LR 0.001000 Time 0.021009 +2023-10-05 21:00:46,659 - Epoch: [39][ 1220/ 1236] Overall Loss 0.325957 Objective Loss 0.325957 LR 0.001000 Time 0.021001 +2023-10-05 21:00:46,913 - Epoch: [39][ 1230/ 1236] Overall Loss 0.326125 Objective Loss 0.326125 LR 0.001000 Time 0.021036 +2023-10-05 21:00:47,031 - Epoch: [39][ 1236/ 1236] Overall Loss 0.326174 Objective Loss 0.326174 Top1 83.503055 Top5 97.352342 LR 0.001000 Time 0.021030 +2023-10-05 21:00:47,158 - --- validate (epoch=39)----------- +2023-10-05 21:00:47,158 - 29943 samples (256 per mini-batch) +2023-10-05 21:00:47,614 - Epoch: [39][ 10/ 117] Loss 0.359437 Top1 82.109375 Top5 98.085938 +2023-10-05 21:00:47,759 - Epoch: [39][ 20/ 117] Loss 0.378828 Top1 81.523438 Top5 97.871094 +2023-10-05 21:00:47,901 - Epoch: [39][ 30/ 117] Loss 0.374534 Top1 81.432292 Top5 97.786458 +2023-10-05 21:00:48,046 - Epoch: [39][ 40/ 117] Loss 0.375446 Top1 81.503906 Top5 97.792969 +2023-10-05 21:00:48,188 - Epoch: [39][ 50/ 117] Loss 0.375819 Top1 81.539062 Top5 97.703125 +2023-10-05 21:00:48,331 - Epoch: [39][ 60/ 117] Loss 0.379299 Top1 81.464844 Top5 97.714844 +2023-10-05 21:00:48,474 - Epoch: [39][ 70/ 117] Loss 0.379086 Top1 81.696429 Top5 97.689732 +2023-10-05 21:00:48,618 - Epoch: [39][ 80/ 117] Loss 0.377357 Top1 81.694336 Top5 97.709961 +2023-10-05 21:00:48,763 - Epoch: [39][ 90/ 117] Loss 0.373732 Top1 81.749132 Top5 97.712674 +2023-10-05 21:00:48,906 - Epoch: [39][ 100/ 117] Loss 0.373481 Top1 81.746094 Top5 97.628906 +2023-10-05 21:00:49,058 - Epoch: [39][ 110/ 117] Loss 0.374781 Top1 81.725852 Top5 97.666903 +2023-10-05 21:00:49,143 - Epoch: [39][ 117/ 117] Loss 0.372624 Top1 81.728618 Top5 97.682263 +2023-10-05 21:00:49,277 - ==> Top1: 81.729 Top5: 97.682 Loss: 0.373 + +2023-10-05 21:00:49,277 - ==> Confusion: +[[ 943 2 3 1 7 4 0 0 3 61 1 0 0 2 3 1 6 1 2 0 10] + [ 3 1014 2 0 13 43 5 27 5 0 1 0 0 0 1 1 1 2 3 2 8] + [ 10 0 932 14 4 3 27 18 0 3 2 5 8 3 1 3 0 2 3 4 14] + [ 7 0 15 952 0 10 5 0 4 2 1 0 11 4 33 2 2 8 12 2 19] + [ 33 9 0 0 959 7 1 0 3 9 1 3 0 2 5 4 4 3 1 1 5] + [ 3 35 2 0 4 986 0 27 3 1 3 6 1 16 3 1 2 1 5 7 10] + [ 1 7 21 1 2 2 1112 18 0 0 1 3 2 0 0 2 1 3 3 10 2] + [ 6 20 10 2 2 48 4 1061 1 4 1 11 6 3 0 0 0 0 19 10 10] + [ 23 1 0 0 0 2 0 0 943 58 9 4 4 15 18 1 1 1 4 1 4] + [ 113 0 3 0 4 3 1 0 28 922 0 2 1 29 2 2 0 0 1 2 6] + [ 6 4 13 8 0 1 8 6 22 3 925 4 1 23 4 2 1 2 4 0 16] + [ 1 0 0 0 0 19 0 0 0 0 0 949 26 7 0 3 1 15 0 11 3] + [ 0 0 4 3 0 0 0 0 0 0 0 43 963 2 2 9 1 17 2 7 15] + [ 2 0 0 0 1 17 0 1 7 18 5 6 2 1039 2 1 1 3 0 2 12] + [ 21 2 3 13 5 0 0 0 21 9 1 1 2 3 986 1 1 5 6 0 21] + [ 1 3 4 3 2 1 3 0 0 0 0 13 8 1 0 1049 13 14 1 7 11] + [ 4 13 2 0 13 11 0 0 0 0 0 7 3 4 3 12 1057 1 0 12 19] + [ 0 0 0 4 0 0 1 0 0 0 0 10 23 0 4 6 0 988 1 1 0] + [ 1 17 9 24 2 2 0 55 8 2 0 1 5 0 11 1 0 0 917 1 12] + [ 0 1 4 0 1 8 3 11 0 0 0 13 9 3 0 3 5 1 1 1081 8] + [ 200 234 173 74 110 282 60 127 99 124 114 127 394 322 132 49 125 89 124 252 4694]] + +2023-10-05 21:00:49,279 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:00:49,279 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:00:49,285 - + +2023-10-05 21:00:49,285 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:00:50,409 - Epoch: [40][ 10/ 1236] Overall Loss 0.322683 Objective Loss 0.322683 LR 0.001000 Time 0.112363 +2023-10-05 21:00:50,613 - Epoch: [40][ 20/ 1236] Overall Loss 0.307073 Objective Loss 0.307073 LR 0.001000 Time 0.066351 +2023-10-05 21:00:50,814 - Epoch: [40][ 30/ 1236] Overall Loss 0.311763 Objective Loss 0.311763 LR 0.001000 Time 0.050929 +2023-10-05 21:00:51,018 - Epoch: [40][ 40/ 1236] Overall Loss 0.318254 Objective Loss 0.318254 LR 0.001000 Time 0.043285 +2023-10-05 21:00:51,218 - Epoch: [40][ 50/ 1236] Overall Loss 0.310706 Objective Loss 0.310706 LR 0.001000 Time 0.038634 +2023-10-05 21:00:51,423 - Epoch: [40][ 60/ 1236] Overall Loss 0.309356 Objective Loss 0.309356 LR 0.001000 Time 0.035595 +2023-10-05 21:00:51,623 - Epoch: [40][ 70/ 1236] Overall Loss 0.309284 Objective Loss 0.309284 LR 0.001000 Time 0.033370 +2023-10-05 21:00:51,828 - Epoch: [40][ 80/ 1236] Overall Loss 0.319148 Objective Loss 0.319148 LR 0.001000 Time 0.031748 +2023-10-05 21:00:52,028 - Epoch: [40][ 90/ 1236] Overall Loss 0.326824 Objective Loss 0.326824 LR 0.001000 Time 0.030447 +2023-10-05 21:00:52,233 - Epoch: [40][ 100/ 1236] Overall Loss 0.329076 Objective Loss 0.329076 LR 0.001000 Time 0.029442 +2023-10-05 21:00:52,433 - Epoch: [40][ 110/ 1236] Overall Loss 0.330251 Objective Loss 0.330251 LR 0.001000 Time 0.028588 +2023-10-05 21:00:52,637 - Epoch: [40][ 120/ 1236] Overall Loss 0.326090 Objective Loss 0.326090 LR 0.001000 Time 0.027901 +2023-10-05 21:00:52,838 - Epoch: [40][ 130/ 1236] Overall Loss 0.325491 Objective Loss 0.325491 LR 0.001000 Time 0.027298 +2023-10-05 21:00:53,042 - Epoch: [40][ 140/ 1236] Overall Loss 0.325376 Objective Loss 0.325376 LR 0.001000 Time 0.026803 +2023-10-05 21:00:53,243 - Epoch: [40][ 150/ 1236] Overall Loss 0.324031 Objective Loss 0.324031 LR 0.001000 Time 0.026354 +2023-10-05 21:00:53,447 - Epoch: [40][ 160/ 1236] Overall Loss 0.324971 Objective Loss 0.324971 LR 0.001000 Time 0.025979 +2023-10-05 21:00:53,648 - Epoch: [40][ 170/ 1236] Overall Loss 0.323262 Objective Loss 0.323262 LR 0.001000 Time 0.025632 +2023-10-05 21:00:53,853 - Epoch: [40][ 180/ 1236] Overall Loss 0.322498 Objective Loss 0.322498 LR 0.001000 Time 0.025340 +2023-10-05 21:00:54,054 - Epoch: [40][ 190/ 1236] Overall Loss 0.321423 Objective Loss 0.321423 LR 0.001000 Time 0.025064 +2023-10-05 21:00:54,258 - Epoch: [40][ 200/ 1236] Overall Loss 0.320871 Objective Loss 0.320871 LR 0.001000 Time 0.024829 +2023-10-05 21:00:54,460 - Epoch: [40][ 210/ 1236] Overall Loss 0.320267 Objective Loss 0.320267 LR 0.001000 Time 0.024606 +2023-10-05 21:00:54,663 - Epoch: [40][ 220/ 1236] Overall Loss 0.320986 Objective Loss 0.320986 LR 0.001000 Time 0.024411 +2023-10-05 21:00:54,865 - Epoch: [40][ 230/ 1236] Overall Loss 0.321801 Objective Loss 0.321801 LR 0.001000 Time 0.024226 +2023-10-05 21:00:55,068 - Epoch: [40][ 240/ 1236] Overall Loss 0.321415 Objective Loss 0.321415 LR 0.001000 Time 0.024062 +2023-10-05 21:00:55,271 - Epoch: [40][ 250/ 1236] Overall Loss 0.320963 Objective Loss 0.320963 LR 0.001000 Time 0.023908 +2023-10-05 21:00:55,474 - Epoch: [40][ 260/ 1236] Overall Loss 0.320599 Objective Loss 0.320599 LR 0.001000 Time 0.023770 +2023-10-05 21:00:55,676 - Epoch: [40][ 270/ 1236] Overall Loss 0.320573 Objective Loss 0.320573 LR 0.001000 Time 0.023638 +2023-10-05 21:00:55,880 - Epoch: [40][ 280/ 1236] Overall Loss 0.320364 Objective Loss 0.320364 LR 0.001000 Time 0.023519 +2023-10-05 21:00:56,081 - Epoch: [40][ 290/ 1236] Overall Loss 0.320977 Objective Loss 0.320977 LR 0.001000 Time 0.023400 +2023-10-05 21:00:56,284 - Epoch: [40][ 300/ 1236] Overall Loss 0.321244 Objective Loss 0.321244 LR 0.001000 Time 0.023297 +2023-10-05 21:00:56,487 - Epoch: [40][ 310/ 1236] Overall Loss 0.322397 Objective Loss 0.322397 LR 0.001000 Time 0.023197 +2023-10-05 21:00:56,688 - Epoch: [40][ 320/ 1236] Overall Loss 0.322508 Objective Loss 0.322508 LR 0.001000 Time 0.023100 +2023-10-05 21:00:56,889 - Epoch: [40][ 330/ 1236] Overall Loss 0.322201 Objective Loss 0.322201 LR 0.001000 Time 0.023010 +2023-10-05 21:00:57,090 - Epoch: [40][ 340/ 1236] Overall Loss 0.321847 Objective Loss 0.321847 LR 0.001000 Time 0.022922 +2023-10-05 21:00:57,292 - Epoch: [40][ 350/ 1236] Overall Loss 0.321325 Objective Loss 0.321325 LR 0.001000 Time 0.022842 +2023-10-05 21:00:57,494 - Epoch: [40][ 360/ 1236] Overall Loss 0.321102 Objective Loss 0.321102 LR 0.001000 Time 0.022764 +2023-10-05 21:00:57,703 - Epoch: [40][ 370/ 1236] Overall Loss 0.321077 Objective Loss 0.321077 LR 0.001000 Time 0.022713 +2023-10-05 21:00:57,911 - Epoch: [40][ 380/ 1236] Overall Loss 0.321480 Objective Loss 0.321480 LR 0.001000 Time 0.022662 +2023-10-05 21:00:58,121 - Epoch: [40][ 390/ 1236] Overall Loss 0.321487 Objective Loss 0.321487 LR 0.001000 Time 0.022620 +2023-10-05 21:00:58,327 - Epoch: [40][ 400/ 1236] Overall Loss 0.322022 Objective Loss 0.322022 LR 0.001000 Time 0.022568 +2023-10-05 21:00:58,528 - Epoch: [40][ 410/ 1236] Overall Loss 0.322716 Objective Loss 0.322716 LR 0.001000 Time 0.022507 +2023-10-05 21:00:58,729 - Epoch: [40][ 420/ 1236] Overall Loss 0.323459 Objective Loss 0.323459 LR 0.001000 Time 0.022449 +2023-10-05 21:00:58,931 - Epoch: [40][ 430/ 1236] Overall Loss 0.323439 Objective Loss 0.323439 LR 0.001000 Time 0.022395 +2023-10-05 21:00:59,132 - Epoch: [40][ 440/ 1236] Overall Loss 0.323529 Objective Loss 0.323529 LR 0.001000 Time 0.022342 +2023-10-05 21:00:59,333 - Epoch: [40][ 450/ 1236] Overall Loss 0.323420 Objective Loss 0.323420 LR 0.001000 Time 0.022293 +2023-10-05 21:00:59,534 - Epoch: [40][ 460/ 1236] Overall Loss 0.323695 Objective Loss 0.323695 LR 0.001000 Time 0.022244 +2023-10-05 21:00:59,735 - Epoch: [40][ 470/ 1236] Overall Loss 0.323830 Objective Loss 0.323830 LR 0.001000 Time 0.022198 +2023-10-05 21:00:59,937 - Epoch: [40][ 480/ 1236] Overall Loss 0.323524 Objective Loss 0.323524 LR 0.001000 Time 0.022155 +2023-10-05 21:01:00,138 - Epoch: [40][ 490/ 1236] Overall Loss 0.323634 Objective Loss 0.323634 LR 0.001000 Time 0.022113 +2023-10-05 21:01:00,340 - Epoch: [40][ 500/ 1236] Overall Loss 0.322758 Objective Loss 0.322758 LR 0.001000 Time 0.022073 +2023-10-05 21:01:00,541 - Epoch: [40][ 510/ 1236] Overall Loss 0.322566 Objective Loss 0.322566 LR 0.001000 Time 0.022035 +2023-10-05 21:01:00,742 - Epoch: [40][ 520/ 1236] Overall Loss 0.322019 Objective Loss 0.322019 LR 0.001000 Time 0.021997 +2023-10-05 21:01:00,943 - Epoch: [40][ 530/ 1236] Overall Loss 0.321165 Objective Loss 0.321165 LR 0.001000 Time 0.021961 +2023-10-05 21:01:01,145 - Epoch: [40][ 540/ 1236] Overall Loss 0.321540 Objective Loss 0.321540 LR 0.001000 Time 0.021926 +2023-10-05 21:01:01,346 - Epoch: [40][ 550/ 1236] Overall Loss 0.321300 Objective Loss 0.321300 LR 0.001000 Time 0.021893 +2023-10-05 21:01:01,547 - Epoch: [40][ 560/ 1236] Overall Loss 0.321772 Objective Loss 0.321772 LR 0.001000 Time 0.021861 +2023-10-05 21:01:01,749 - Epoch: [40][ 570/ 1236] Overall Loss 0.322302 Objective Loss 0.322302 LR 0.001000 Time 0.021830 +2023-10-05 21:01:01,950 - Epoch: [40][ 580/ 1236] Overall Loss 0.321688 Objective Loss 0.321688 LR 0.001000 Time 0.021800 +2023-10-05 21:01:02,151 - Epoch: [40][ 590/ 1236] Overall Loss 0.321793 Objective Loss 0.321793 LR 0.001000 Time 0.021771 +2023-10-05 21:01:02,352 - Epoch: [40][ 600/ 1236] Overall Loss 0.321899 Objective Loss 0.321899 LR 0.001000 Time 0.021743 +2023-10-05 21:01:02,553 - Epoch: [40][ 610/ 1236] Overall Loss 0.322055 Objective Loss 0.322055 LR 0.001000 Time 0.021716 +2023-10-05 21:01:02,755 - Epoch: [40][ 620/ 1236] Overall Loss 0.321975 Objective Loss 0.321975 LR 0.001000 Time 0.021689 +2023-10-05 21:01:02,956 - Epoch: [40][ 630/ 1236] Overall Loss 0.322018 Objective Loss 0.322018 LR 0.001000 Time 0.021664 +2023-10-05 21:01:03,157 - Epoch: [40][ 640/ 1236] Overall Loss 0.321811 Objective Loss 0.321811 LR 0.001000 Time 0.021640 +2023-10-05 21:01:03,360 - Epoch: [40][ 650/ 1236] Overall Loss 0.322577 Objective Loss 0.322577 LR 0.001000 Time 0.021619 +2023-10-05 21:01:03,564 - Epoch: [40][ 660/ 1236] Overall Loss 0.322015 Objective Loss 0.322015 LR 0.001000 Time 0.021599 +2023-10-05 21:01:03,768 - Epoch: [40][ 670/ 1236] Overall Loss 0.321860 Objective Loss 0.321860 LR 0.001000 Time 0.021580 +2023-10-05 21:01:03,971 - Epoch: [40][ 680/ 1236] Overall Loss 0.321476 Objective Loss 0.321476 LR 0.001000 Time 0.021561 +2023-10-05 21:01:04,175 - Epoch: [40][ 690/ 1236] Overall Loss 0.321519 Objective Loss 0.321519 LR 0.001000 Time 0.021544 +2023-10-05 21:01:04,379 - Epoch: [40][ 700/ 1236] Overall Loss 0.321443 Objective Loss 0.321443 LR 0.001000 Time 0.021526 +2023-10-05 21:01:04,583 - Epoch: [40][ 710/ 1236] Overall Loss 0.321762 Objective Loss 0.321762 LR 0.001000 Time 0.021510 +2023-10-05 21:01:04,786 - Epoch: [40][ 720/ 1236] Overall Loss 0.321840 Objective Loss 0.321840 LR 0.001000 Time 0.021493 +2023-10-05 21:01:04,991 - Epoch: [40][ 730/ 1236] Overall Loss 0.322314 Objective Loss 0.322314 LR 0.001000 Time 0.021478 +2023-10-05 21:01:05,194 - Epoch: [40][ 740/ 1236] Overall Loss 0.322468 Objective Loss 0.322468 LR 0.001000 Time 0.021462 +2023-10-05 21:01:05,398 - Epoch: [40][ 750/ 1236] Overall Loss 0.322154 Objective Loss 0.322154 LR 0.001000 Time 0.021447 +2023-10-05 21:01:05,601 - Epoch: [40][ 760/ 1236] Overall Loss 0.322241 Objective Loss 0.322241 LR 0.001000 Time 0.021432 +2023-10-05 21:01:05,805 - Epoch: [40][ 770/ 1236] Overall Loss 0.321682 Objective Loss 0.321682 LR 0.001000 Time 0.021418 +2023-10-05 21:01:06,009 - Epoch: [40][ 780/ 1236] Overall Loss 0.321764 Objective Loss 0.321764 LR 0.001000 Time 0.021404 +2023-10-05 21:01:06,213 - Epoch: [40][ 790/ 1236] Overall Loss 0.321632 Objective Loss 0.321632 LR 0.001000 Time 0.021391 +2023-10-05 21:01:06,416 - Epoch: [40][ 800/ 1236] Overall Loss 0.321871 Objective Loss 0.321871 LR 0.001000 Time 0.021377 +2023-10-05 21:01:06,621 - Epoch: [40][ 810/ 1236] Overall Loss 0.321865 Objective Loss 0.321865 LR 0.001000 Time 0.021365 +2023-10-05 21:01:06,824 - Epoch: [40][ 820/ 1236] Overall Loss 0.321665 Objective Loss 0.321665 LR 0.001000 Time 0.021352 +2023-10-05 21:01:07,026 - Epoch: [40][ 830/ 1236] Overall Loss 0.321773 Objective Loss 0.321773 LR 0.001000 Time 0.021338 +2023-10-05 21:01:07,228 - Epoch: [40][ 840/ 1236] Overall Loss 0.321720 Objective Loss 0.321720 LR 0.001000 Time 0.021324 +2023-10-05 21:01:07,428 - Epoch: [40][ 850/ 1236] Overall Loss 0.322082 Objective Loss 0.322082 LR 0.001000 Time 0.021308 +2023-10-05 21:01:07,629 - Epoch: [40][ 860/ 1236] Overall Loss 0.321638 Objective Loss 0.321638 LR 0.001000 Time 0.021294 +2023-10-05 21:01:07,829 - Epoch: [40][ 870/ 1236] Overall Loss 0.321787 Objective Loss 0.321787 LR 0.001000 Time 0.021278 +2023-10-05 21:01:08,030 - Epoch: [40][ 880/ 1236] Overall Loss 0.321813 Objective Loss 0.321813 LR 0.001000 Time 0.021264 +2023-10-05 21:01:08,230 - Epoch: [40][ 890/ 1236] Overall Loss 0.321869 Objective Loss 0.321869 LR 0.001000 Time 0.021249 +2023-10-05 21:01:08,431 - Epoch: [40][ 900/ 1236] Overall Loss 0.322108 Objective Loss 0.322108 LR 0.001000 Time 0.021237 +2023-10-05 21:01:08,631 - Epoch: [40][ 910/ 1236] Overall Loss 0.322699 Objective Loss 0.322699 LR 0.001000 Time 0.021222 +2023-10-05 21:01:08,833 - Epoch: [40][ 920/ 1236] Overall Loss 0.322666 Objective Loss 0.322666 LR 0.001000 Time 0.021210 +2023-10-05 21:01:09,033 - Epoch: [40][ 930/ 1236] Overall Loss 0.322347 Objective Loss 0.322347 LR 0.001000 Time 0.021197 +2023-10-05 21:01:09,234 - Epoch: [40][ 940/ 1236] Overall Loss 0.322719 Objective Loss 0.322719 LR 0.001000 Time 0.021185 +2023-10-05 21:01:09,434 - Epoch: [40][ 950/ 1236] Overall Loss 0.322657 Objective Loss 0.322657 LR 0.001000 Time 0.021172 +2023-10-05 21:01:09,634 - Epoch: [40][ 960/ 1236] Overall Loss 0.323180 Objective Loss 0.323180 LR 0.001000 Time 0.021159 +2023-10-05 21:01:09,833 - Epoch: [40][ 970/ 1236] Overall Loss 0.323259 Objective Loss 0.323259 LR 0.001000 Time 0.021147 +2023-10-05 21:01:10,035 - Epoch: [40][ 980/ 1236] Overall Loss 0.323454 Objective Loss 0.323454 LR 0.001000 Time 0.021136 +2023-10-05 21:01:10,232 - Epoch: [40][ 990/ 1236] Overall Loss 0.323400 Objective Loss 0.323400 LR 0.001000 Time 0.021122 +2023-10-05 21:01:10,434 - Epoch: [40][ 1000/ 1236] Overall Loss 0.323222 Objective Loss 0.323222 LR 0.001000 Time 0.021113 +2023-10-05 21:01:10,634 - Epoch: [40][ 1010/ 1236] Overall Loss 0.323232 Objective Loss 0.323232 LR 0.001000 Time 0.021101 +2023-10-05 21:01:10,836 - Epoch: [40][ 1020/ 1236] Overall Loss 0.323046 Objective Loss 0.323046 LR 0.001000 Time 0.021091 +2023-10-05 21:01:11,035 - Epoch: [40][ 1030/ 1236] Overall Loss 0.323023 Objective Loss 0.323023 LR 0.001000 Time 0.021080 +2023-10-05 21:01:11,236 - Epoch: [40][ 1040/ 1236] Overall Loss 0.322815 Objective Loss 0.322815 LR 0.001000 Time 0.021070 +2023-10-05 21:01:11,436 - Epoch: [40][ 1050/ 1236] Overall Loss 0.323062 Objective Loss 0.323062 LR 0.001000 Time 0.021060 +2023-10-05 21:01:11,638 - Epoch: [40][ 1060/ 1236] Overall Loss 0.322926 Objective Loss 0.322926 LR 0.001000 Time 0.021051 +2023-10-05 21:01:11,836 - Epoch: [40][ 1070/ 1236] Overall Loss 0.323335 Objective Loss 0.323335 LR 0.001000 Time 0.021039 +2023-10-05 21:01:12,038 - Epoch: [40][ 1080/ 1236] Overall Loss 0.323298 Objective Loss 0.323298 LR 0.001000 Time 0.021031 +2023-10-05 21:01:12,236 - Epoch: [40][ 1090/ 1236] Overall Loss 0.323216 Objective Loss 0.323216 LR 0.001000 Time 0.021019 +2023-10-05 21:01:12,437 - Epoch: [40][ 1100/ 1236] Overall Loss 0.323651 Objective Loss 0.323651 LR 0.001000 Time 0.021011 +2023-10-05 21:01:12,637 - Epoch: [40][ 1110/ 1236] Overall Loss 0.323612 Objective Loss 0.323612 LR 0.001000 Time 0.021002 +2023-10-05 21:01:12,839 - Epoch: [40][ 1120/ 1236] Overall Loss 0.323883 Objective Loss 0.323883 LR 0.001000 Time 0.020994 +2023-10-05 21:01:13,037 - Epoch: [40][ 1130/ 1236] Overall Loss 0.323816 Objective Loss 0.323816 LR 0.001000 Time 0.020983 +2023-10-05 21:01:13,238 - Epoch: [40][ 1140/ 1236] Overall Loss 0.324202 Objective Loss 0.324202 LR 0.001000 Time 0.020975 +2023-10-05 21:01:13,438 - Epoch: [40][ 1150/ 1236] Overall Loss 0.324419 Objective Loss 0.324419 LR 0.001000 Time 0.020966 +2023-10-05 21:01:13,639 - Epoch: [40][ 1160/ 1236] Overall Loss 0.324537 Objective Loss 0.324537 LR 0.001000 Time 0.020959 +2023-10-05 21:01:13,839 - Epoch: [40][ 1170/ 1236] Overall Loss 0.324726 Objective Loss 0.324726 LR 0.001000 Time 0.020950 +2023-10-05 21:01:14,040 - Epoch: [40][ 1180/ 1236] Overall Loss 0.324952 Objective Loss 0.324952 LR 0.001000 Time 0.020943 +2023-10-05 21:01:14,240 - Epoch: [40][ 1190/ 1236] Overall Loss 0.324913 Objective Loss 0.324913 LR 0.001000 Time 0.020934 +2023-10-05 21:01:14,441 - Epoch: [40][ 1200/ 1236] Overall Loss 0.325106 Objective Loss 0.325106 LR 0.001000 Time 0.020927 +2023-10-05 21:01:14,641 - Epoch: [40][ 1210/ 1236] Overall Loss 0.325139 Objective Loss 0.325139 LR 0.001000 Time 0.020919 +2023-10-05 21:01:14,842 - Epoch: [40][ 1220/ 1236] Overall Loss 0.325255 Objective Loss 0.325255 LR 0.001000 Time 0.020912 +2023-10-05 21:01:15,096 - Epoch: [40][ 1230/ 1236] Overall Loss 0.325335 Objective Loss 0.325335 LR 0.001000 Time 0.020948 +2023-10-05 21:01:15,214 - Epoch: [40][ 1236/ 1236] Overall Loss 0.325503 Objective Loss 0.325503 Top1 80.855397 Top5 97.148676 LR 0.001000 Time 0.020942 +2023-10-05 21:01:15,357 - --- validate (epoch=40)----------- +2023-10-05 21:01:15,357 - 29943 samples (256 per mini-batch) +2023-10-05 21:01:15,809 - Epoch: [40][ 10/ 117] Loss 0.373815 Top1 81.132812 Top5 96.992188 +2023-10-05 21:01:15,959 - Epoch: [40][ 20/ 117] Loss 0.367634 Top1 81.367188 Top5 97.031250 +2023-10-05 21:01:16,105 - Epoch: [40][ 30/ 117] Loss 0.383534 Top1 80.651042 Top5 96.940104 +2023-10-05 21:01:16,254 - Epoch: [40][ 40/ 117] Loss 0.383901 Top1 80.849609 Top5 97.148438 +2023-10-05 21:01:16,401 - Epoch: [40][ 50/ 117] Loss 0.382137 Top1 80.867188 Top5 97.179688 +2023-10-05 21:01:16,550 - Epoch: [40][ 60/ 117] Loss 0.381468 Top1 80.787760 Top5 97.109375 +2023-10-05 21:01:16,695 - Epoch: [40][ 70/ 117] Loss 0.381527 Top1 80.625000 Top5 97.165179 +2023-10-05 21:01:16,842 - Epoch: [40][ 80/ 117] Loss 0.382628 Top1 80.585938 Top5 97.192383 +2023-10-05 21:01:16,988 - Epoch: [40][ 90/ 117] Loss 0.379262 Top1 80.607639 Top5 97.222222 +2023-10-05 21:01:17,136 - Epoch: [40][ 100/ 117] Loss 0.382078 Top1 80.605469 Top5 97.183594 +2023-10-05 21:01:17,289 - Epoch: [40][ 110/ 117] Loss 0.383759 Top1 80.479403 Top5 97.162642 +2023-10-05 21:01:17,375 - Epoch: [40][ 117/ 117] Loss 0.381422 Top1 80.466219 Top5 97.147914 +2023-10-05 21:01:17,521 - ==> Top1: 80.466 Top5: 97.148 Loss: 0.381 + +2023-10-05 21:01:17,522 - ==> Confusion: +[[ 956 2 8 3 10 3 1 0 3 34 1 1 1 1 5 7 2 2 0 0 10] + [ 1 1029 0 0 6 25 3 38 2 1 2 1 0 0 0 4 3 0 11 1 4] + [ 6 0 951 11 3 1 37 11 0 0 2 5 6 1 2 2 2 1 3 2 10] + [ 3 1 24 926 1 9 4 3 2 0 6 0 4 2 34 5 3 8 35 0 19] + [ 31 9 0 0 961 7 1 3 1 3 2 2 1 0 6 6 7 2 4 0 4] + [ 3 25 1 0 4 972 2 37 3 2 4 7 1 13 5 4 11 3 3 6 10] + [ 0 11 35 0 0 4 1107 11 0 0 0 2 1 0 1 8 1 0 1 4 5] + [ 1 19 19 0 2 28 2 1064 0 4 3 10 3 0 1 4 2 1 42 8 5] + [ 25 4 0 0 0 3 0 1 940 40 11 4 3 9 26 2 1 2 16 0 2] + [ 195 1 1 0 11 2 1 2 29 829 0 1 1 20 6 5 2 1 1 3 8] + [ 3 8 19 9 2 2 3 5 17 1 943 3 1 8 2 1 5 1 8 2 10] + [ 1 2 5 0 0 16 1 9 1 0 0 925 38 0 0 5 5 14 1 10 2] + [ 0 1 8 5 0 0 1 2 0 0 0 60 939 0 4 14 2 17 3 7 5] + [ 1 0 3 0 9 12 1 3 13 14 8 8 4 1005 4 6 5 3 0 7 13] + [ 15 5 2 12 7 0 0 1 20 6 2 1 0 0 993 0 3 3 20 0 11] + [ 0 1 3 0 3 1 5 0 0 0 0 11 4 0 0 1070 12 11 1 7 5] + [ 3 14 2 1 5 4 1 2 0 0 0 5 0 1 1 12 1088 1 1 6 14] + [ 0 0 1 1 1 0 2 0 0 1 0 6 24 0 2 7 0 987 3 1 2] + [ 0 11 12 11 2 0 0 43 1 1 1 1 7 0 10 0 2 0 954 0 12] + [ 0 4 5 1 2 9 14 29 0 0 0 16 7 1 0 4 8 0 3 1042 7] + [ 185 261 208 77 103 191 70 174 103 54 136 149 441 267 138 89 323 71 181 271 4413]] + +2023-10-05 21:01:17,523 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:01:17,523 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:01:17,529 - + +2023-10-05 21:01:17,529 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:01:18,540 - Epoch: [41][ 10/ 1236] Overall Loss 0.328795 Objective Loss 0.328795 LR 0.001000 Time 0.101093 +2023-10-05 21:01:18,745 - Epoch: [41][ 20/ 1236] Overall Loss 0.313907 Objective Loss 0.313907 LR 0.001000 Time 0.060740 +2023-10-05 21:01:18,946 - Epoch: [41][ 30/ 1236] Overall Loss 0.324980 Objective Loss 0.324980 LR 0.001000 Time 0.047204 +2023-10-05 21:01:19,151 - Epoch: [41][ 40/ 1236] Overall Loss 0.314437 Objective Loss 0.314437 LR 0.001000 Time 0.040507 +2023-10-05 21:01:19,353 - Epoch: [41][ 50/ 1236] Overall Loss 0.316473 Objective Loss 0.316473 LR 0.001000 Time 0.036437 +2023-10-05 21:01:19,557 - Epoch: [41][ 60/ 1236] Overall Loss 0.313878 Objective Loss 0.313878 LR 0.001000 Time 0.033767 +2023-10-05 21:01:19,758 - Epoch: [41][ 70/ 1236] Overall Loss 0.317748 Objective Loss 0.317748 LR 0.001000 Time 0.031812 +2023-10-05 21:01:19,963 - Epoch: [41][ 80/ 1236] Overall Loss 0.316570 Objective Loss 0.316570 LR 0.001000 Time 0.030389 +2023-10-05 21:01:20,165 - Epoch: [41][ 90/ 1236] Overall Loss 0.319429 Objective Loss 0.319429 LR 0.001000 Time 0.029251 +2023-10-05 21:01:20,370 - Epoch: [41][ 100/ 1236] Overall Loss 0.321195 Objective Loss 0.321195 LR 0.001000 Time 0.028373 +2023-10-05 21:01:20,571 - Epoch: [41][ 110/ 1236] Overall Loss 0.322215 Objective Loss 0.322215 LR 0.001000 Time 0.027621 +2023-10-05 21:01:20,776 - Epoch: [41][ 120/ 1236] Overall Loss 0.321699 Objective Loss 0.321699 LR 0.001000 Time 0.027026 +2023-10-05 21:01:20,977 - Epoch: [41][ 130/ 1236] Overall Loss 0.319994 Objective Loss 0.319994 LR 0.001000 Time 0.026491 +2023-10-05 21:01:21,182 - Epoch: [41][ 140/ 1236] Overall Loss 0.317792 Objective Loss 0.317792 LR 0.001000 Time 0.026061 +2023-10-05 21:01:21,384 - Epoch: [41][ 150/ 1236] Overall Loss 0.318138 Objective Loss 0.318138 LR 0.001000 Time 0.025666 +2023-10-05 21:01:21,589 - Epoch: [41][ 160/ 1236] Overall Loss 0.317141 Objective Loss 0.317141 LR 0.001000 Time 0.025341 +2023-10-05 21:01:21,790 - Epoch: [41][ 170/ 1236] Overall Loss 0.316434 Objective Loss 0.316434 LR 0.001000 Time 0.025031 +2023-10-05 21:01:21,995 - Epoch: [41][ 180/ 1236] Overall Loss 0.318567 Objective Loss 0.318567 LR 0.001000 Time 0.024778 +2023-10-05 21:01:22,197 - Epoch: [41][ 190/ 1236] Overall Loss 0.316317 Objective Loss 0.316317 LR 0.001000 Time 0.024532 +2023-10-05 21:01:22,402 - Epoch: [41][ 200/ 1236] Overall Loss 0.315076 Objective Loss 0.315076 LR 0.001000 Time 0.024332 +2023-10-05 21:01:22,601 - Epoch: [41][ 210/ 1236] Overall Loss 0.314595 Objective Loss 0.314595 LR 0.001000 Time 0.024118 +2023-10-05 21:01:22,806 - Epoch: [41][ 220/ 1236] Overall Loss 0.313398 Objective Loss 0.313398 LR 0.001000 Time 0.023952 +2023-10-05 21:01:23,007 - Epoch: [41][ 230/ 1236] Overall Loss 0.314101 Objective Loss 0.314101 LR 0.001000 Time 0.023784 +2023-10-05 21:01:23,212 - Epoch: [41][ 240/ 1236] Overall Loss 0.313648 Objective Loss 0.313648 LR 0.001000 Time 0.023647 +2023-10-05 21:01:23,414 - Epoch: [41][ 250/ 1236] Overall Loss 0.314703 Objective Loss 0.314703 LR 0.001000 Time 0.023506 +2023-10-05 21:01:23,619 - Epoch: [41][ 260/ 1236] Overall Loss 0.315369 Objective Loss 0.315369 LR 0.001000 Time 0.023388 +2023-10-05 21:01:23,820 - Epoch: [41][ 270/ 1236] Overall Loss 0.315632 Objective Loss 0.315632 LR 0.001000 Time 0.023267 +2023-10-05 21:01:24,026 - Epoch: [41][ 280/ 1236] Overall Loss 0.315318 Objective Loss 0.315318 LR 0.001000 Time 0.023167 +2023-10-05 21:01:24,227 - Epoch: [41][ 290/ 1236] Overall Loss 0.314602 Objective Loss 0.314602 LR 0.001000 Time 0.023063 +2023-10-05 21:01:24,429 - Epoch: [41][ 300/ 1236] Overall Loss 0.315732 Objective Loss 0.315732 LR 0.001000 Time 0.022966 +2023-10-05 21:01:24,629 - Epoch: [41][ 310/ 1236] Overall Loss 0.316186 Objective Loss 0.316186 LR 0.001000 Time 0.022869 +2023-10-05 21:01:24,831 - Epoch: [41][ 320/ 1236] Overall Loss 0.317460 Objective Loss 0.317460 LR 0.001000 Time 0.022785 +2023-10-05 21:01:25,032 - Epoch: [41][ 330/ 1236] Overall Loss 0.318170 Objective Loss 0.318170 LR 0.001000 Time 0.022702 +2023-10-05 21:01:25,237 - Epoch: [41][ 340/ 1236] Overall Loss 0.317784 Objective Loss 0.317784 LR 0.001000 Time 0.022637 +2023-10-05 21:01:25,440 - Epoch: [41][ 350/ 1236] Overall Loss 0.317186 Objective Loss 0.317186 LR 0.001000 Time 0.022567 +2023-10-05 21:01:25,644 - Epoch: [41][ 360/ 1236] Overall Loss 0.316720 Objective Loss 0.316720 LR 0.001000 Time 0.022508 +2023-10-05 21:01:25,846 - Epoch: [41][ 370/ 1236] Overall Loss 0.317151 Objective Loss 0.317151 LR 0.001000 Time 0.022445 +2023-10-05 21:01:26,051 - Epoch: [41][ 380/ 1236] Overall Loss 0.317517 Objective Loss 0.317517 LR 0.001000 Time 0.022391 +2023-10-05 21:01:26,253 - Epoch: [41][ 390/ 1236] Overall Loss 0.316563 Objective Loss 0.316563 LR 0.001000 Time 0.022334 +2023-10-05 21:01:26,458 - Epoch: [41][ 400/ 1236] Overall Loss 0.316242 Objective Loss 0.316242 LR 0.001000 Time 0.022287 +2023-10-05 21:01:26,660 - Epoch: [41][ 410/ 1236] Overall Loss 0.316982 Objective Loss 0.316982 LR 0.001000 Time 0.022235 +2023-10-05 21:01:26,864 - Epoch: [41][ 420/ 1236] Overall Loss 0.317269 Objective Loss 0.317269 LR 0.001000 Time 0.022192 +2023-10-05 21:01:27,066 - Epoch: [41][ 430/ 1236] Overall Loss 0.317973 Objective Loss 0.317973 LR 0.001000 Time 0.022145 +2023-10-05 21:01:27,271 - Epoch: [41][ 440/ 1236] Overall Loss 0.317757 Objective Loss 0.317757 LR 0.001000 Time 0.022107 +2023-10-05 21:01:27,473 - Epoch: [41][ 450/ 1236] Overall Loss 0.318101 Objective Loss 0.318101 LR 0.001000 Time 0.022064 +2023-10-05 21:01:27,678 - Epoch: [41][ 460/ 1236] Overall Loss 0.317632 Objective Loss 0.317632 LR 0.001000 Time 0.022028 +2023-10-05 21:01:27,880 - Epoch: [41][ 470/ 1236] Overall Loss 0.317522 Objective Loss 0.317522 LR 0.001000 Time 0.021989 +2023-10-05 21:01:28,085 - Epoch: [41][ 480/ 1236] Overall Loss 0.317844 Objective Loss 0.317844 LR 0.001000 Time 0.021957 +2023-10-05 21:01:28,286 - Epoch: [41][ 490/ 1236] Overall Loss 0.318352 Objective Loss 0.318352 LR 0.001000 Time 0.021919 +2023-10-05 21:01:28,491 - Epoch: [41][ 500/ 1236] Overall Loss 0.318556 Objective Loss 0.318556 LR 0.001000 Time 0.021889 +2023-10-05 21:01:28,693 - Epoch: [41][ 510/ 1236] Overall Loss 0.318626 Objective Loss 0.318626 LR 0.001000 Time 0.021856 +2023-10-05 21:01:28,898 - Epoch: [41][ 520/ 1236] Overall Loss 0.318424 Objective Loss 0.318424 LR 0.001000 Time 0.021828 +2023-10-05 21:01:29,100 - Epoch: [41][ 530/ 1236] Overall Loss 0.318313 Objective Loss 0.318313 LR 0.001000 Time 0.021797 +2023-10-05 21:01:29,305 - Epoch: [41][ 540/ 1236] Overall Loss 0.317845 Objective Loss 0.317845 LR 0.001000 Time 0.021772 +2023-10-05 21:01:29,507 - Epoch: [41][ 550/ 1236] Overall Loss 0.318193 Objective Loss 0.318193 LR 0.001000 Time 0.021743 +2023-10-05 21:01:29,711 - Epoch: [41][ 560/ 1236] Overall Loss 0.318789 Objective Loss 0.318789 LR 0.001000 Time 0.021719 +2023-10-05 21:01:29,914 - Epoch: [41][ 570/ 1236] Overall Loss 0.319256 Objective Loss 0.319256 LR 0.001000 Time 0.021692 +2023-10-05 21:01:30,118 - Epoch: [41][ 580/ 1236] Overall Loss 0.319314 Objective Loss 0.319314 LR 0.001000 Time 0.021671 +2023-10-05 21:01:30,320 - Epoch: [41][ 590/ 1236] Overall Loss 0.319359 Objective Loss 0.319359 LR 0.001000 Time 0.021645 +2023-10-05 21:01:30,525 - Epoch: [41][ 600/ 1236] Overall Loss 0.319243 Objective Loss 0.319243 LR 0.001000 Time 0.021625 +2023-10-05 21:01:30,727 - Epoch: [41][ 610/ 1236] Overall Loss 0.319145 Objective Loss 0.319145 LR 0.001000 Time 0.021602 +2023-10-05 21:01:30,932 - Epoch: [41][ 620/ 1236] Overall Loss 0.319489 Objective Loss 0.319489 LR 0.001000 Time 0.021583 +2023-10-05 21:01:31,134 - Epoch: [41][ 630/ 1236] Overall Loss 0.319831 Objective Loss 0.319831 LR 0.001000 Time 0.021561 +2023-10-05 21:01:31,339 - Epoch: [41][ 640/ 1236] Overall Loss 0.320532 Objective Loss 0.320532 LR 0.001000 Time 0.021544 +2023-10-05 21:01:31,541 - Epoch: [41][ 650/ 1236] Overall Loss 0.320875 Objective Loss 0.320875 LR 0.001000 Time 0.021523 +2023-10-05 21:01:31,746 - Epoch: [41][ 660/ 1236] Overall Loss 0.320980 Objective Loss 0.320980 LR 0.001000 Time 0.021506 +2023-10-05 21:01:31,948 - Epoch: [41][ 670/ 1236] Overall Loss 0.320926 Objective Loss 0.320926 LR 0.001000 Time 0.021487 +2023-10-05 21:01:32,153 - Epoch: [41][ 680/ 1236] Overall Loss 0.321254 Objective Loss 0.321254 LR 0.001000 Time 0.021471 +2023-10-05 21:01:32,355 - Epoch: [41][ 690/ 1236] Overall Loss 0.321149 Objective Loss 0.321149 LR 0.001000 Time 0.021453 +2023-10-05 21:01:32,571 - Epoch: [41][ 700/ 1236] Overall Loss 0.321545 Objective Loss 0.321545 LR 0.001000 Time 0.021453 +2023-10-05 21:01:32,779 - Epoch: [41][ 710/ 1236] Overall Loss 0.321358 Objective Loss 0.321358 LR 0.001000 Time 0.021445 +2023-10-05 21:01:32,996 - Epoch: [41][ 720/ 1236] Overall Loss 0.321433 Objective Loss 0.321433 LR 0.001000 Time 0.021446 +2023-10-05 21:01:33,204 - Epoch: [41][ 730/ 1236] Overall Loss 0.321259 Objective Loss 0.321259 LR 0.001000 Time 0.021438 +2023-10-05 21:01:33,421 - Epoch: [41][ 740/ 1236] Overall Loss 0.321024 Objective Loss 0.321024 LR 0.001000 Time 0.021440 +2023-10-05 21:01:33,629 - Epoch: [41][ 750/ 1236] Overall Loss 0.320769 Objective Loss 0.320769 LR 0.001000 Time 0.021432 +2023-10-05 21:01:33,845 - Epoch: [41][ 760/ 1236] Overall Loss 0.320757 Objective Loss 0.320757 LR 0.001000 Time 0.021433 +2023-10-05 21:01:34,054 - Epoch: [41][ 770/ 1236] Overall Loss 0.320760 Objective Loss 0.320760 LR 0.001000 Time 0.021426 +2023-10-05 21:01:34,271 - Epoch: [41][ 780/ 1236] Overall Loss 0.320749 Objective Loss 0.320749 LR 0.001000 Time 0.021428 +2023-10-05 21:01:34,476 - Epoch: [41][ 790/ 1236] Overall Loss 0.320756 Objective Loss 0.320756 LR 0.001000 Time 0.021416 +2023-10-05 21:01:34,689 - Epoch: [41][ 800/ 1236] Overall Loss 0.320819 Objective Loss 0.320819 LR 0.001000 Time 0.021414 +2023-10-05 21:01:34,898 - Epoch: [41][ 810/ 1236] Overall Loss 0.320879 Objective Loss 0.320879 LR 0.001000 Time 0.021408 +2023-10-05 21:01:35,105 - Epoch: [41][ 820/ 1236] Overall Loss 0.321173 Objective Loss 0.321173 LR 0.001000 Time 0.021398 +2023-10-05 21:01:35,310 - Epoch: [41][ 830/ 1236] Overall Loss 0.321249 Objective Loss 0.321249 LR 0.001000 Time 0.021387 +2023-10-05 21:01:35,517 - Epoch: [41][ 840/ 1236] Overall Loss 0.321684 Objective Loss 0.321684 LR 0.001000 Time 0.021379 +2023-10-05 21:01:35,722 - Epoch: [41][ 850/ 1236] Overall Loss 0.321546 Objective Loss 0.321546 LR 0.001000 Time 0.021369 +2023-10-05 21:01:35,929 - Epoch: [41][ 860/ 1236] Overall Loss 0.321470 Objective Loss 0.321470 LR 0.001000 Time 0.021360 +2023-10-05 21:01:36,135 - Epoch: [41][ 870/ 1236] Overall Loss 0.321697 Objective Loss 0.321697 LR 0.001000 Time 0.021351 +2023-10-05 21:01:36,341 - Epoch: [41][ 880/ 1236] Overall Loss 0.321779 Objective Loss 0.321779 LR 0.001000 Time 0.021343 +2023-10-05 21:01:36,547 - Epoch: [41][ 890/ 1236] Overall Loss 0.321715 Objective Loss 0.321715 LR 0.001000 Time 0.021333 +2023-10-05 21:01:36,754 - Epoch: [41][ 900/ 1236] Overall Loss 0.321960 Objective Loss 0.321960 LR 0.001000 Time 0.021326 +2023-10-05 21:01:36,959 - Epoch: [41][ 910/ 1236] Overall Loss 0.321716 Objective Loss 0.321716 LR 0.001000 Time 0.021317 +2023-10-05 21:01:37,169 - Epoch: [41][ 920/ 1236] Overall Loss 0.322000 Objective Loss 0.322000 LR 0.001000 Time 0.021312 +2023-10-05 21:01:37,374 - Epoch: [41][ 930/ 1236] Overall Loss 0.322307 Objective Loss 0.322307 LR 0.001000 Time 0.021304 +2023-10-05 21:01:37,581 - Epoch: [41][ 940/ 1236] Overall Loss 0.322237 Objective Loss 0.322237 LR 0.001000 Time 0.021297 +2023-10-05 21:01:37,787 - Epoch: [41][ 950/ 1236] Overall Loss 0.322055 Objective Loss 0.322055 LR 0.001000 Time 0.021289 +2023-10-05 21:01:37,994 - Epoch: [41][ 960/ 1236] Overall Loss 0.321889 Objective Loss 0.321889 LR 0.001000 Time 0.021283 +2023-10-05 21:01:38,199 - Epoch: [41][ 970/ 1236] Overall Loss 0.321885 Objective Loss 0.321885 LR 0.001000 Time 0.021275 +2023-10-05 21:01:38,407 - Epoch: [41][ 980/ 1236] Overall Loss 0.321775 Objective Loss 0.321775 LR 0.001000 Time 0.021269 +2023-10-05 21:01:38,612 - Epoch: [41][ 990/ 1236] Overall Loss 0.321759 Objective Loss 0.321759 LR 0.001000 Time 0.021261 +2023-10-05 21:01:38,819 - Epoch: [41][ 1000/ 1236] Overall Loss 0.322089 Objective Loss 0.322089 LR 0.001000 Time 0.021255 +2023-10-05 21:01:39,024 - Epoch: [41][ 1010/ 1236] Overall Loss 0.322434 Objective Loss 0.322434 LR 0.001000 Time 0.021247 +2023-10-05 21:01:39,231 - Epoch: [41][ 1020/ 1236] Overall Loss 0.322276 Objective Loss 0.322276 LR 0.001000 Time 0.021241 +2023-10-05 21:01:39,436 - Epoch: [41][ 1030/ 1236] Overall Loss 0.322556 Objective Loss 0.322556 LR 0.001000 Time 0.021234 +2023-10-05 21:01:39,651 - Epoch: [41][ 1040/ 1236] Overall Loss 0.322760 Objective Loss 0.322760 LR 0.001000 Time 0.021236 +2023-10-05 21:01:39,868 - Epoch: [41][ 1050/ 1236] Overall Loss 0.322820 Objective Loss 0.322820 LR 0.001000 Time 0.021240 +2023-10-05 21:01:40,078 - Epoch: [41][ 1060/ 1236] Overall Loss 0.322787 Objective Loss 0.322787 LR 0.001000 Time 0.021238 +2023-10-05 21:01:40,284 - Epoch: [41][ 1070/ 1236] Overall Loss 0.323117 Objective Loss 0.323117 LR 0.001000 Time 0.021231 +2023-10-05 21:01:40,490 - Epoch: [41][ 1080/ 1236] Overall Loss 0.323181 Objective Loss 0.323181 LR 0.001000 Time 0.021226 +2023-10-05 21:01:40,696 - Epoch: [41][ 1090/ 1236] Overall Loss 0.323354 Objective Loss 0.323354 LR 0.001000 Time 0.021219 +2023-10-05 21:01:40,903 - Epoch: [41][ 1100/ 1236] Overall Loss 0.323529 Objective Loss 0.323529 LR 0.001000 Time 0.021214 +2023-10-05 21:01:41,108 - Epoch: [41][ 1110/ 1236] Overall Loss 0.323608 Objective Loss 0.323608 LR 0.001000 Time 0.021208 +2023-10-05 21:01:41,313 - Epoch: [41][ 1120/ 1236] Overall Loss 0.323860 Objective Loss 0.323860 LR 0.001000 Time 0.021201 +2023-10-05 21:01:41,517 - Epoch: [41][ 1130/ 1236] Overall Loss 0.323961 Objective Loss 0.323961 LR 0.001000 Time 0.021194 +2023-10-05 21:01:41,722 - Epoch: [41][ 1140/ 1236] Overall Loss 0.324134 Objective Loss 0.324134 LR 0.001000 Time 0.021187 +2023-10-05 21:01:41,926 - Epoch: [41][ 1150/ 1236] Overall Loss 0.324195 Objective Loss 0.324195 LR 0.001000 Time 0.021180 +2023-10-05 21:01:42,130 - Epoch: [41][ 1160/ 1236] Overall Loss 0.324165 Objective Loss 0.324165 LR 0.001000 Time 0.021173 +2023-10-05 21:01:42,336 - Epoch: [41][ 1170/ 1236] Overall Loss 0.324086 Objective Loss 0.324086 LR 0.001000 Time 0.021168 +2023-10-05 21:01:42,543 - Epoch: [41][ 1180/ 1236] Overall Loss 0.323871 Objective Loss 0.323871 LR 0.001000 Time 0.021163 +2023-10-05 21:01:42,748 - Epoch: [41][ 1190/ 1236] Overall Loss 0.324052 Objective Loss 0.324052 LR 0.001000 Time 0.021158 +2023-10-05 21:01:42,955 - Epoch: [41][ 1200/ 1236] Overall Loss 0.324094 Objective Loss 0.324094 LR 0.001000 Time 0.021154 +2023-10-05 21:01:43,160 - Epoch: [41][ 1210/ 1236] Overall Loss 0.324203 Objective Loss 0.324203 LR 0.001000 Time 0.021148 +2023-10-05 21:01:43,367 - Epoch: [41][ 1220/ 1236] Overall Loss 0.324208 Objective Loss 0.324208 LR 0.001000 Time 0.021145 +2023-10-05 21:01:43,627 - Epoch: [41][ 1230/ 1236] Overall Loss 0.324167 Objective Loss 0.324167 LR 0.001000 Time 0.021184 +2023-10-05 21:01:43,747 - Epoch: [41][ 1236/ 1236] Overall Loss 0.324355 Objective Loss 0.324355 Top1 85.336049 Top5 97.556008 LR 0.001000 Time 0.021177 +2023-10-05 21:01:43,867 - --- validate (epoch=41)----------- +2023-10-05 21:01:43,867 - 29943 samples (256 per mini-batch) +2023-10-05 21:01:44,321 - Epoch: [41][ 10/ 117] Loss 0.354782 Top1 79.921875 Top5 97.148438 +2023-10-05 21:01:44,469 - Epoch: [41][ 20/ 117] Loss 0.349792 Top1 80.546875 Top5 97.324219 +2023-10-05 21:01:44,614 - Epoch: [41][ 30/ 117] Loss 0.350792 Top1 80.794271 Top5 97.226562 +2023-10-05 21:01:44,762 - Epoch: [41][ 40/ 117] Loss 0.359398 Top1 80.546875 Top5 97.285156 +2023-10-05 21:01:44,907 - Epoch: [41][ 50/ 117] Loss 0.366541 Top1 80.500000 Top5 97.250000 +2023-10-05 21:01:45,056 - Epoch: [41][ 60/ 117] Loss 0.369850 Top1 80.332031 Top5 97.233073 +2023-10-05 21:01:45,200 - Epoch: [41][ 70/ 117] Loss 0.370708 Top1 80.251116 Top5 97.193080 +2023-10-05 21:01:45,350 - Epoch: [41][ 80/ 117] Loss 0.374461 Top1 80.273438 Top5 97.202148 +2023-10-05 21:01:45,496 - Epoch: [41][ 90/ 117] Loss 0.372880 Top1 80.403646 Top5 97.204861 +2023-10-05 21:01:45,646 - Epoch: [41][ 100/ 117] Loss 0.374944 Top1 80.253906 Top5 97.144531 +2023-10-05 21:01:45,802 - Epoch: [41][ 110/ 117] Loss 0.374585 Top1 80.163352 Top5 97.155540 +2023-10-05 21:01:45,888 - Epoch: [41][ 117/ 117] Loss 0.375755 Top1 80.135591 Top5 97.151254 +2023-10-05 21:01:46,018 - ==> Top1: 80.136 Top5: 97.151 Loss: 0.376 + +2023-10-05 21:01:46,019 - ==> Confusion: +[[ 933 4 6 4 8 3 0 0 0 63 2 0 0 2 4 5 4 1 3 0 8] + [ 0 1034 1 2 3 35 3 28 1 1 2 0 0 0 1 3 3 0 9 2 3] + [ 7 0 913 21 3 0 42 23 0 2 5 3 7 3 0 3 3 1 5 7 8] + [ 2 0 15 958 1 7 3 1 3 0 3 0 10 5 27 4 0 8 28 4 10] + [ 25 6 1 1 970 9 0 0 0 13 0 2 1 1 7 4 5 2 1 2 0] + [ 5 40 1 2 4 972 6 28 1 1 2 11 1 15 4 1 3 1 3 8 7] + [ 0 3 18 1 0 1 1109 14 0 0 5 5 3 1 1 16 0 1 1 8 4] + [ 3 20 13 0 4 30 7 1064 0 5 4 10 0 1 1 2 1 1 38 11 3] + [ 25 6 0 0 1 3 1 1 934 49 9 2 0 15 28 3 2 2 8 0 0] + [ 134 1 0 1 6 7 1 2 25 893 1 0 0 23 4 8 1 1 0 7 4] + [ 3 6 15 6 0 1 6 5 15 1 955 3 1 10 7 0 0 1 10 0 8] + [ 1 0 2 0 0 14 1 4 1 0 1 949 24 6 0 2 1 18 3 6 2] + [ 2 1 5 7 0 1 2 1 3 0 1 61 926 1 4 8 2 26 5 7 5] + [ 3 1 0 0 5 11 0 1 14 22 12 5 2 1021 2 3 2 2 0 2 11] + [ 14 4 3 11 8 0 0 1 24 6 3 0 1 1 991 0 2 5 20 0 7] + [ 0 3 3 2 3 1 2 0 0 1 0 11 9 0 0 1064 8 16 0 6 5] + [ 4 19 3 2 13 9 1 0 1 0 0 9 3 3 1 12 1061 1 1 10 8] + [ 2 1 0 4 0 1 2 0 0 1 0 10 17 2 3 6 0 987 1 0 1] + [ 2 12 2 16 0 0 1 39 1 0 2 0 3 0 6 1 1 0 976 1 5] + [ 0 2 2 2 2 8 9 16 0 0 2 15 6 1 1 5 5 1 2 1067 6] + [ 225 237 151 123 156 269 77 134 95 97 190 162 404 364 147 81 142 128 230 275 4218]] + +2023-10-05 21:01:46,020 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:01:46,020 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:01:46,026 - + +2023-10-05 21:01:46,026 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:01:47,040 - Epoch: [42][ 10/ 1236] Overall Loss 0.324531 Objective Loss 0.324531 LR 0.001000 Time 0.101337 +2023-10-05 21:01:47,244 - Epoch: [42][ 20/ 1236] Overall Loss 0.314552 Objective Loss 0.314552 LR 0.001000 Time 0.060865 +2023-10-05 21:01:47,446 - Epoch: [42][ 30/ 1236] Overall Loss 0.316866 Objective Loss 0.316866 LR 0.001000 Time 0.047276 +2023-10-05 21:01:47,650 - Epoch: [42][ 40/ 1236] Overall Loss 0.312356 Objective Loss 0.312356 LR 0.001000 Time 0.040562 +2023-10-05 21:01:47,852 - Epoch: [42][ 50/ 1236] Overall Loss 0.317912 Objective Loss 0.317912 LR 0.001000 Time 0.036476 +2023-10-05 21:01:48,056 - Epoch: [42][ 60/ 1236] Overall Loss 0.316723 Objective Loss 0.316723 LR 0.001000 Time 0.033797 +2023-10-05 21:01:48,258 - Epoch: [42][ 70/ 1236] Overall Loss 0.316497 Objective Loss 0.316497 LR 0.001000 Time 0.031844 +2023-10-05 21:01:48,461 - Epoch: [42][ 80/ 1236] Overall Loss 0.316124 Objective Loss 0.316124 LR 0.001000 Time 0.030401 +2023-10-05 21:01:48,663 - Epoch: [42][ 90/ 1236] Overall Loss 0.315940 Objective Loss 0.315940 LR 0.001000 Time 0.029260 +2023-10-05 21:01:48,866 - Epoch: [42][ 100/ 1236] Overall Loss 0.316441 Objective Loss 0.316441 LR 0.001000 Time 0.028361 +2023-10-05 21:01:49,067 - Epoch: [42][ 110/ 1236] Overall Loss 0.314671 Objective Loss 0.314671 LR 0.001000 Time 0.027610 +2023-10-05 21:01:49,270 - Epoch: [42][ 120/ 1236] Overall Loss 0.314679 Objective Loss 0.314679 LR 0.001000 Time 0.026998 +2023-10-05 21:01:49,472 - Epoch: [42][ 130/ 1236] Overall Loss 0.313236 Objective Loss 0.313236 LR 0.001000 Time 0.026470 +2023-10-05 21:01:49,675 - Epoch: [42][ 140/ 1236] Overall Loss 0.312222 Objective Loss 0.312222 LR 0.001000 Time 0.026027 +2023-10-05 21:01:49,876 - Epoch: [42][ 150/ 1236] Overall Loss 0.312272 Objective Loss 0.312272 LR 0.001000 Time 0.025628 +2023-10-05 21:01:50,079 - Epoch: [42][ 160/ 1236] Overall Loss 0.313249 Objective Loss 0.313249 LR 0.001000 Time 0.025294 +2023-10-05 21:01:50,280 - Epoch: [42][ 170/ 1236] Overall Loss 0.312522 Objective Loss 0.312522 LR 0.001000 Time 0.024987 +2023-10-05 21:01:50,483 - Epoch: [42][ 180/ 1236] Overall Loss 0.313089 Objective Loss 0.313089 LR 0.001000 Time 0.024725 +2023-10-05 21:01:50,685 - Epoch: [42][ 190/ 1236] Overall Loss 0.314108 Objective Loss 0.314108 LR 0.001000 Time 0.024483 +2023-10-05 21:01:50,889 - Epoch: [42][ 200/ 1236] Overall Loss 0.315173 Objective Loss 0.315173 LR 0.001000 Time 0.024277 +2023-10-05 21:01:51,090 - Epoch: [42][ 210/ 1236] Overall Loss 0.314001 Objective Loss 0.314001 LR 0.001000 Time 0.024076 +2023-10-05 21:01:51,294 - Epoch: [42][ 220/ 1236] Overall Loss 0.312357 Objective Loss 0.312357 LR 0.001000 Time 0.023907 +2023-10-05 21:01:51,495 - Epoch: [42][ 230/ 1236] Overall Loss 0.311563 Objective Loss 0.311563 LR 0.001000 Time 0.023740 +2023-10-05 21:01:51,699 - Epoch: [42][ 240/ 1236] Overall Loss 0.311397 Objective Loss 0.311397 LR 0.001000 Time 0.023599 +2023-10-05 21:01:51,900 - Epoch: [42][ 250/ 1236] Overall Loss 0.313232 Objective Loss 0.313232 LR 0.001000 Time 0.023459 +2023-10-05 21:01:52,104 - Epoch: [42][ 260/ 1236] Overall Loss 0.312810 Objective Loss 0.312810 LR 0.001000 Time 0.023339 +2023-10-05 21:01:52,305 - Epoch: [42][ 270/ 1236] Overall Loss 0.312485 Objective Loss 0.312485 LR 0.001000 Time 0.023219 +2023-10-05 21:01:52,509 - Epoch: [42][ 280/ 1236] Overall Loss 0.313061 Objective Loss 0.313061 LR 0.001000 Time 0.023116 +2023-10-05 21:01:52,710 - Epoch: [42][ 290/ 1236] Overall Loss 0.313649 Objective Loss 0.313649 LR 0.001000 Time 0.023012 +2023-10-05 21:01:52,916 - Epoch: [42][ 300/ 1236] Overall Loss 0.313906 Objective Loss 0.313906 LR 0.001000 Time 0.022929 +2023-10-05 21:01:53,122 - Epoch: [42][ 310/ 1236] Overall Loss 0.313928 Objective Loss 0.313928 LR 0.001000 Time 0.022855 +2023-10-05 21:01:53,334 - Epoch: [42][ 320/ 1236] Overall Loss 0.313477 Objective Loss 0.313477 LR 0.001000 Time 0.022802 +2023-10-05 21:01:53,540 - Epoch: [42][ 330/ 1236] Overall Loss 0.314313 Objective Loss 0.314313 LR 0.001000 Time 0.022735 +2023-10-05 21:01:53,748 - Epoch: [42][ 340/ 1236] Overall Loss 0.313853 Objective Loss 0.313853 LR 0.001000 Time 0.022675 +2023-10-05 21:01:53,948 - Epoch: [42][ 350/ 1236] Overall Loss 0.314657 Objective Loss 0.314657 LR 0.001000 Time 0.022598 +2023-10-05 21:01:54,149 - Epoch: [42][ 360/ 1236] Overall Loss 0.315975 Objective Loss 0.315975 LR 0.001000 Time 0.022530 +2023-10-05 21:01:54,350 - Epoch: [42][ 370/ 1236] Overall Loss 0.315662 Objective Loss 0.315662 LR 0.001000 Time 0.022461 +2023-10-05 21:01:54,553 - Epoch: [42][ 380/ 1236] Overall Loss 0.315517 Objective Loss 0.315517 LR 0.001000 Time 0.022403 +2023-10-05 21:01:54,753 - Epoch: [42][ 390/ 1236] Overall Loss 0.315364 Objective Loss 0.315364 LR 0.001000 Time 0.022342 +2023-10-05 21:01:54,956 - Epoch: [42][ 400/ 1236] Overall Loss 0.315049 Objective Loss 0.315049 LR 0.001000 Time 0.022289 +2023-10-05 21:01:55,157 - Epoch: [42][ 410/ 1236] Overall Loss 0.315290 Objective Loss 0.315290 LR 0.001000 Time 0.022234 +2023-10-05 21:01:55,359 - Epoch: [42][ 420/ 1236] Overall Loss 0.315000 Objective Loss 0.315000 LR 0.001000 Time 0.022187 +2023-10-05 21:01:55,560 - Epoch: [42][ 430/ 1236] Overall Loss 0.314754 Objective Loss 0.314754 LR 0.001000 Time 0.022137 +2023-10-05 21:01:55,763 - Epoch: [42][ 440/ 1236] Overall Loss 0.314847 Objective Loss 0.314847 LR 0.001000 Time 0.022093 +2023-10-05 21:01:55,963 - Epoch: [42][ 450/ 1236] Overall Loss 0.315556 Objective Loss 0.315556 LR 0.001000 Time 0.022048 +2023-10-05 21:01:56,166 - Epoch: [42][ 460/ 1236] Overall Loss 0.315169 Objective Loss 0.315169 LR 0.001000 Time 0.022009 +2023-10-05 21:01:56,367 - Epoch: [42][ 470/ 1236] Overall Loss 0.315552 Objective Loss 0.315552 LR 0.001000 Time 0.021966 +2023-10-05 21:01:56,569 - Epoch: [42][ 480/ 1236] Overall Loss 0.315410 Objective Loss 0.315410 LR 0.001000 Time 0.021930 +2023-10-05 21:01:56,770 - Epoch: [42][ 490/ 1236] Overall Loss 0.315121 Objective Loss 0.315121 LR 0.001000 Time 0.021891 +2023-10-05 21:01:56,973 - Epoch: [42][ 500/ 1236] Overall Loss 0.315002 Objective Loss 0.315002 LR 0.001000 Time 0.021859 +2023-10-05 21:01:57,174 - Epoch: [42][ 510/ 1236] Overall Loss 0.315198 Objective Loss 0.315198 LR 0.001000 Time 0.021823 +2023-10-05 21:01:57,377 - Epoch: [42][ 520/ 1236] Overall Loss 0.315137 Objective Loss 0.315137 LR 0.001000 Time 0.021793 +2023-10-05 21:01:57,577 - Epoch: [42][ 530/ 1236] Overall Loss 0.314249 Objective Loss 0.314249 LR 0.001000 Time 0.021759 +2023-10-05 21:01:57,780 - Epoch: [42][ 540/ 1236] Overall Loss 0.313995 Objective Loss 0.313995 LR 0.001000 Time 0.021731 +2023-10-05 21:01:57,981 - Epoch: [42][ 550/ 1236] Overall Loss 0.314041 Objective Loss 0.314041 LR 0.001000 Time 0.021700 +2023-10-05 21:01:58,187 - Epoch: [42][ 560/ 1236] Overall Loss 0.314146 Objective Loss 0.314146 LR 0.001000 Time 0.021681 +2023-10-05 21:01:58,396 - Epoch: [42][ 570/ 1236] Overall Loss 0.314349 Objective Loss 0.314349 LR 0.001000 Time 0.021666 +2023-10-05 21:01:58,609 - Epoch: [42][ 580/ 1236] Overall Loss 0.314160 Objective Loss 0.314160 LR 0.001000 Time 0.021660 +2023-10-05 21:01:58,818 - Epoch: [42][ 590/ 1236] Overall Loss 0.314055 Objective Loss 0.314055 LR 0.001000 Time 0.021646 +2023-10-05 21:01:59,032 - Epoch: [42][ 600/ 1236] Overall Loss 0.314109 Objective Loss 0.314109 LR 0.001000 Time 0.021641 +2023-10-05 21:01:59,241 - Epoch: [42][ 610/ 1236] Overall Loss 0.314478 Objective Loss 0.314478 LR 0.001000 Time 0.021629 +2023-10-05 21:01:59,455 - Epoch: [42][ 620/ 1236] Overall Loss 0.314526 Objective Loss 0.314526 LR 0.001000 Time 0.021625 +2023-10-05 21:01:59,659 - Epoch: [42][ 630/ 1236] Overall Loss 0.314063 Objective Loss 0.314063 LR 0.001000 Time 0.021605 +2023-10-05 21:01:59,863 - Epoch: [42][ 640/ 1236] Overall Loss 0.313506 Objective Loss 0.313506 LR 0.001000 Time 0.021585 +2023-10-05 21:02:00,064 - Epoch: [42][ 650/ 1236] Overall Loss 0.313079 Objective Loss 0.313079 LR 0.001000 Time 0.021562 +2023-10-05 21:02:00,266 - Epoch: [42][ 660/ 1236] Overall Loss 0.313426 Objective Loss 0.313426 LR 0.001000 Time 0.021541 +2023-10-05 21:02:00,468 - Epoch: [42][ 670/ 1236] Overall Loss 0.313372 Objective Loss 0.313372 LR 0.001000 Time 0.021520 +2023-10-05 21:02:00,671 - Epoch: [42][ 680/ 1236] Overall Loss 0.313426 Objective Loss 0.313426 LR 0.001000 Time 0.021502 +2023-10-05 21:02:00,873 - Epoch: [42][ 690/ 1236] Overall Loss 0.312953 Objective Loss 0.312953 LR 0.001000 Time 0.021482 +2023-10-05 21:02:01,076 - Epoch: [42][ 700/ 1236] Overall Loss 0.312938 Objective Loss 0.312938 LR 0.001000 Time 0.021465 +2023-10-05 21:02:01,277 - Epoch: [42][ 710/ 1236] Overall Loss 0.313022 Objective Loss 0.313022 LR 0.001000 Time 0.021446 +2023-10-05 21:02:01,481 - Epoch: [42][ 720/ 1236] Overall Loss 0.313213 Objective Loss 0.313213 LR 0.001000 Time 0.021430 +2023-10-05 21:02:01,682 - Epoch: [42][ 730/ 1236] Overall Loss 0.313630 Objective Loss 0.313630 LR 0.001000 Time 0.021412 +2023-10-05 21:02:01,885 - Epoch: [42][ 740/ 1236] Overall Loss 0.313850 Objective Loss 0.313850 LR 0.001000 Time 0.021396 +2023-10-05 21:02:02,086 - Epoch: [42][ 750/ 1236] Overall Loss 0.314072 Objective Loss 0.314072 LR 0.001000 Time 0.021379 +2023-10-05 21:02:02,289 - Epoch: [42][ 760/ 1236] Overall Loss 0.314526 Objective Loss 0.314526 LR 0.001000 Time 0.021364 +2023-10-05 21:02:02,490 - Epoch: [42][ 770/ 1236] Overall Loss 0.314764 Objective Loss 0.314764 LR 0.001000 Time 0.021347 +2023-10-05 21:02:02,694 - Epoch: [42][ 780/ 1236] Overall Loss 0.315144 Objective Loss 0.315144 LR 0.001000 Time 0.021334 +2023-10-05 21:02:02,895 - Epoch: [42][ 790/ 1236] Overall Loss 0.315537 Objective Loss 0.315537 LR 0.001000 Time 0.021318 +2023-10-05 21:02:03,098 - Epoch: [42][ 800/ 1236] Overall Loss 0.315668 Objective Loss 0.315668 LR 0.001000 Time 0.021305 +2023-10-05 21:02:03,299 - Epoch: [42][ 810/ 1236] Overall Loss 0.315767 Objective Loss 0.315767 LR 0.001000 Time 0.021290 +2023-10-05 21:02:03,503 - Epoch: [42][ 820/ 1236] Overall Loss 0.315812 Objective Loss 0.315812 LR 0.001000 Time 0.021278 +2023-10-05 21:02:03,704 - Epoch: [42][ 830/ 1236] Overall Loss 0.316365 Objective Loss 0.316365 LR 0.001000 Time 0.021263 +2023-10-05 21:02:03,907 - Epoch: [42][ 840/ 1236] Overall Loss 0.316434 Objective Loss 0.316434 LR 0.001000 Time 0.021252 +2023-10-05 21:02:04,108 - Epoch: [42][ 850/ 1236] Overall Loss 0.316272 Objective Loss 0.316272 LR 0.001000 Time 0.021238 +2023-10-05 21:02:04,312 - Epoch: [42][ 860/ 1236] Overall Loss 0.316639 Objective Loss 0.316639 LR 0.001000 Time 0.021227 +2023-10-05 21:02:04,513 - Epoch: [42][ 870/ 1236] Overall Loss 0.316911 Objective Loss 0.316911 LR 0.001000 Time 0.021214 +2023-10-05 21:02:04,716 - Epoch: [42][ 880/ 1236] Overall Loss 0.316702 Objective Loss 0.316702 LR 0.001000 Time 0.021204 +2023-10-05 21:02:04,917 - Epoch: [42][ 890/ 1236] Overall Loss 0.316923 Objective Loss 0.316923 LR 0.001000 Time 0.021191 +2023-10-05 21:02:05,121 - Epoch: [42][ 900/ 1236] Overall Loss 0.317247 Objective Loss 0.317247 LR 0.001000 Time 0.021182 +2023-10-05 21:02:05,321 - Epoch: [42][ 910/ 1236] Overall Loss 0.317408 Objective Loss 0.317408 LR 0.001000 Time 0.021169 +2023-10-05 21:02:05,526 - Epoch: [42][ 920/ 1236] Overall Loss 0.317410 Objective Loss 0.317410 LR 0.001000 Time 0.021160 +2023-10-05 21:02:05,726 - Epoch: [42][ 930/ 1236] Overall Loss 0.317577 Objective Loss 0.317577 LR 0.001000 Time 0.021148 +2023-10-05 21:02:05,927 - Epoch: [42][ 940/ 1236] Overall Loss 0.317815 Objective Loss 0.317815 LR 0.001000 Time 0.021136 +2023-10-05 21:02:06,126 - Epoch: [42][ 950/ 1236] Overall Loss 0.317930 Objective Loss 0.317930 LR 0.001000 Time 0.021122 +2023-10-05 21:02:06,326 - Epoch: [42][ 960/ 1236] Overall Loss 0.317918 Objective Loss 0.317918 LR 0.001000 Time 0.021111 +2023-10-05 21:02:06,525 - Epoch: [42][ 970/ 1236] Overall Loss 0.318141 Objective Loss 0.318141 LR 0.001000 Time 0.021097 +2023-10-05 21:02:06,725 - Epoch: [42][ 980/ 1236] Overall Loss 0.318044 Objective Loss 0.318044 LR 0.001000 Time 0.021086 +2023-10-05 21:02:06,924 - Epoch: [42][ 990/ 1236] Overall Loss 0.317813 Objective Loss 0.317813 LR 0.001000 Time 0.021074 +2023-10-05 21:02:07,124 - Epoch: [42][ 1000/ 1236] Overall Loss 0.317776 Objective Loss 0.317776 LR 0.001000 Time 0.021063 +2023-10-05 21:02:07,323 - Epoch: [42][ 1010/ 1236] Overall Loss 0.317900 Objective Loss 0.317900 LR 0.001000 Time 0.021051 +2023-10-05 21:02:07,524 - Epoch: [42][ 1020/ 1236] Overall Loss 0.317638 Objective Loss 0.317638 LR 0.001000 Time 0.021041 +2023-10-05 21:02:07,722 - Epoch: [42][ 1030/ 1236] Overall Loss 0.317892 Objective Loss 0.317892 LR 0.001000 Time 0.021030 +2023-10-05 21:02:07,923 - Epoch: [42][ 1040/ 1236] Overall Loss 0.318168 Objective Loss 0.318168 LR 0.001000 Time 0.021020 +2023-10-05 21:02:08,121 - Epoch: [42][ 1050/ 1236] Overall Loss 0.318124 Objective Loss 0.318124 LR 0.001000 Time 0.021009 +2023-10-05 21:02:08,322 - Epoch: [42][ 1060/ 1236] Overall Loss 0.318389 Objective Loss 0.318389 LR 0.001000 Time 0.020999 +2023-10-05 21:02:08,520 - Epoch: [42][ 1070/ 1236] Overall Loss 0.318734 Objective Loss 0.318734 LR 0.001000 Time 0.020988 +2023-10-05 21:02:08,721 - Epoch: [42][ 1080/ 1236] Overall Loss 0.318992 Objective Loss 0.318992 LR 0.001000 Time 0.020979 +2023-10-05 21:02:08,919 - Epoch: [42][ 1090/ 1236] Overall Loss 0.319044 Objective Loss 0.319044 LR 0.001000 Time 0.020969 +2023-10-05 21:02:09,120 - Epoch: [42][ 1100/ 1236] Overall Loss 0.318881 Objective Loss 0.318881 LR 0.001000 Time 0.020960 +2023-10-05 21:02:09,319 - Epoch: [42][ 1110/ 1236] Overall Loss 0.318755 Objective Loss 0.318755 LR 0.001000 Time 0.020950 +2023-10-05 21:02:09,516 - Epoch: [42][ 1120/ 1236] Overall Loss 0.318536 Objective Loss 0.318536 LR 0.001000 Time 0.020939 +2023-10-05 21:02:09,711 - Epoch: [42][ 1130/ 1236] Overall Loss 0.318509 Objective Loss 0.318509 LR 0.001000 Time 0.020926 +2023-10-05 21:02:09,906 - Epoch: [42][ 1140/ 1236] Overall Loss 0.318653 Objective Loss 0.318653 LR 0.001000 Time 0.020913 +2023-10-05 21:02:10,101 - Epoch: [42][ 1150/ 1236] Overall Loss 0.318778 Objective Loss 0.318778 LR 0.001000 Time 0.020901 +2023-10-05 21:02:10,296 - Epoch: [42][ 1160/ 1236] Overall Loss 0.318726 Objective Loss 0.318726 LR 0.001000 Time 0.020888 +2023-10-05 21:02:10,491 - Epoch: [42][ 1170/ 1236] Overall Loss 0.318621 Objective Loss 0.318621 LR 0.001000 Time 0.020876 +2023-10-05 21:02:10,686 - Epoch: [42][ 1180/ 1236] Overall Loss 0.318517 Objective Loss 0.318517 LR 0.001000 Time 0.020864 +2023-10-05 21:02:10,881 - Epoch: [42][ 1190/ 1236] Overall Loss 0.318504 Objective Loss 0.318504 LR 0.001000 Time 0.020852 +2023-10-05 21:02:11,075 - Epoch: [42][ 1200/ 1236] Overall Loss 0.318330 Objective Loss 0.318330 LR 0.001000 Time 0.020841 +2023-10-05 21:02:11,271 - Epoch: [42][ 1210/ 1236] Overall Loss 0.318483 Objective Loss 0.318483 LR 0.001000 Time 0.020830 +2023-10-05 21:02:11,466 - Epoch: [42][ 1220/ 1236] Overall Loss 0.318477 Objective Loss 0.318477 LR 0.001000 Time 0.020818 +2023-10-05 21:02:11,712 - Epoch: [42][ 1230/ 1236] Overall Loss 0.318205 Objective Loss 0.318205 LR 0.001000 Time 0.020849 +2023-10-05 21:02:11,829 - Epoch: [42][ 1236/ 1236] Overall Loss 0.318244 Objective Loss 0.318244 Top1 82.281059 Top5 97.963340 LR 0.001000 Time 0.020843 +2023-10-05 21:02:11,955 - --- validate (epoch=42)----------- +2023-10-05 21:02:11,955 - 29943 samples (256 per mini-batch) +2023-10-05 21:02:12,417 - Epoch: [42][ 10/ 117] Loss 0.416183 Top1 81.406250 Top5 97.070312 +2023-10-05 21:02:12,569 - Epoch: [42][ 20/ 117] Loss 0.387809 Top1 81.640625 Top5 97.480469 +2023-10-05 21:02:12,719 - Epoch: [42][ 30/ 117] Loss 0.376311 Top1 81.432292 Top5 97.421875 +2023-10-05 21:02:12,870 - Epoch: [42][ 40/ 117] Loss 0.374949 Top1 81.269531 Top5 97.460938 +2023-10-05 21:02:13,020 - Epoch: [42][ 50/ 117] Loss 0.370530 Top1 81.468750 Top5 97.609375 +2023-10-05 21:02:13,172 - Epoch: [42][ 60/ 117] Loss 0.369587 Top1 81.477865 Top5 97.688802 +2023-10-05 21:02:13,323 - Epoch: [42][ 70/ 117] Loss 0.366762 Top1 81.601562 Top5 97.739955 +2023-10-05 21:02:13,480 - Epoch: [42][ 80/ 117] Loss 0.364693 Top1 81.713867 Top5 97.753906 +2023-10-05 21:02:13,629 - Epoch: [42][ 90/ 117] Loss 0.364098 Top1 81.940104 Top5 97.734375 +2023-10-05 21:02:13,776 - Epoch: [42][ 100/ 117] Loss 0.366426 Top1 81.937500 Top5 97.761719 +2023-10-05 21:02:13,934 - Epoch: [42][ 110/ 117] Loss 0.361611 Top1 82.095170 Top5 97.780540 +2023-10-05 21:02:14,020 - Epoch: [42][ 117/ 117] Loss 0.364943 Top1 81.918979 Top5 97.795812 +2023-10-05 21:02:14,131 - ==> Top1: 81.919 Top5: 97.796 Loss: 0.365 + +2023-10-05 21:02:14,131 - ==> Confusion: +[[ 931 0 10 1 7 1 0 2 6 66 1 1 0 3 4 4 2 0 0 0 11] + [ 1 1049 3 2 3 18 1 23 5 0 2 1 0 0 1 4 4 0 10 0 4] + [ 7 1 941 18 1 1 38 6 0 1 3 5 3 3 5 4 2 2 7 0 8] + [ 5 2 28 947 2 1 2 0 3 0 4 0 6 4 32 3 0 11 25 1 13] + [ 40 5 0 1 943 6 0 0 0 13 0 3 1 1 8 4 10 2 2 1 10] + [ 7 53 1 3 6 966 1 20 3 5 4 5 2 15 6 1 1 1 2 1 13] + [ 0 6 31 0 0 2 1126 3 0 0 1 2 2 0 1 1 0 0 3 4 9] + [ 1 21 22 0 2 23 6 1056 0 4 5 10 4 0 2 2 2 0 44 6 8] + [ 22 5 0 2 2 2 0 2 952 37 11 2 2 12 20 3 3 0 9 0 3] + [ 131 0 2 0 4 1 0 1 38 891 1 0 0 20 9 9 0 0 0 2 10] + [ 1 3 16 9 2 2 7 4 11 4 947 2 0 10 9 0 0 0 19 0 7] + [ 1 1 3 1 0 13 0 5 1 0 0 948 28 4 0 1 3 18 0 7 1] + [ 0 1 6 8 0 4 1 1 1 0 0 39 946 6 3 9 3 27 4 4 5] + [ 1 0 2 3 4 9 0 0 12 11 10 8 1 1038 2 3 3 0 1 1 10] + [ 21 2 4 17 7 0 0 0 25 6 1 1 0 1 989 0 2 1 19 0 5] + [ 1 3 1 2 3 2 5 0 0 0 0 15 8 1 1 1043 20 12 1 5 11] + [ 2 26 4 1 8 2 0 1 3 0 0 5 1 2 2 7 1078 0 1 6 12] + [ 0 0 0 0 1 0 0 0 0 1 0 2 20 0 4 6 0 996 1 0 7] + [ 0 7 13 19 0 0 2 26 2 0 0 2 1 0 12 0 1 0 972 0 11] + [ 0 1 6 1 1 7 12 19 0 0 1 11 10 3 0 7 5 1 6 1045 16] + [ 149 247 206 79 88 133 47 101 126 82 156 130 417 339 155 72 141 82 245 185 4725]] + +2023-10-05 21:02:14,133 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:02:14,133 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:02:14,138 - + +2023-10-05 21:02:14,139 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:02:15,140 - Epoch: [43][ 10/ 1236] Overall Loss 0.301677 Objective Loss 0.301677 LR 0.001000 Time 0.100087 +2023-10-05 21:02:15,344 - Epoch: [43][ 20/ 1236] Overall Loss 0.302172 Objective Loss 0.302172 LR 0.001000 Time 0.060232 +2023-10-05 21:02:15,546 - Epoch: [43][ 30/ 1236] Overall Loss 0.302147 Objective Loss 0.302147 LR 0.001000 Time 0.046878 +2023-10-05 21:02:15,751 - Epoch: [43][ 40/ 1236] Overall Loss 0.305765 Objective Loss 0.305765 LR 0.001000 Time 0.040264 +2023-10-05 21:02:15,953 - Epoch: [43][ 50/ 1236] Overall Loss 0.306962 Objective Loss 0.306962 LR 0.001000 Time 0.036244 +2023-10-05 21:02:16,157 - Epoch: [43][ 60/ 1236] Overall Loss 0.307530 Objective Loss 0.307530 LR 0.001000 Time 0.033604 +2023-10-05 21:02:16,359 - Epoch: [43][ 70/ 1236] Overall Loss 0.309574 Objective Loss 0.309574 LR 0.001000 Time 0.031679 +2023-10-05 21:02:16,563 - Epoch: [43][ 80/ 1236] Overall Loss 0.307964 Objective Loss 0.307964 LR 0.001000 Time 0.030270 +2023-10-05 21:02:16,764 - Epoch: [43][ 90/ 1236] Overall Loss 0.310556 Objective Loss 0.310556 LR 0.001000 Time 0.029134 +2023-10-05 21:02:16,966 - Epoch: [43][ 100/ 1236] Overall Loss 0.309757 Objective Loss 0.309757 LR 0.001000 Time 0.028242 +2023-10-05 21:02:17,166 - Epoch: [43][ 110/ 1236] Overall Loss 0.309068 Objective Loss 0.309068 LR 0.001000 Time 0.027488 +2023-10-05 21:02:17,368 - Epoch: [43][ 120/ 1236] Overall Loss 0.308177 Objective Loss 0.308177 LR 0.001000 Time 0.026879 +2023-10-05 21:02:17,568 - Epoch: [43][ 130/ 1236] Overall Loss 0.306656 Objective Loss 0.306656 LR 0.001000 Time 0.026347 +2023-10-05 21:02:17,771 - Epoch: [43][ 140/ 1236] Overall Loss 0.306563 Objective Loss 0.306563 LR 0.001000 Time 0.025906 +2023-10-05 21:02:17,971 - Epoch: [43][ 150/ 1236] Overall Loss 0.308346 Objective Loss 0.308346 LR 0.001000 Time 0.025514 +2023-10-05 21:02:18,173 - Epoch: [43][ 160/ 1236] Overall Loss 0.308311 Objective Loss 0.308311 LR 0.001000 Time 0.025179 +2023-10-05 21:02:18,373 - Epoch: [43][ 170/ 1236] Overall Loss 0.306492 Objective Loss 0.306492 LR 0.001000 Time 0.024872 +2023-10-05 21:02:18,576 - Epoch: [43][ 180/ 1236] Overall Loss 0.306042 Objective Loss 0.306042 LR 0.001000 Time 0.024614 +2023-10-05 21:02:18,776 - Epoch: [43][ 190/ 1236] Overall Loss 0.306220 Objective Loss 0.306220 LR 0.001000 Time 0.024372 +2023-10-05 21:02:18,982 - Epoch: [43][ 200/ 1236] Overall Loss 0.305318 Objective Loss 0.305318 LR 0.001000 Time 0.024180 +2023-10-05 21:02:19,184 - Epoch: [43][ 210/ 1236] Overall Loss 0.303648 Objective Loss 0.303648 LR 0.001000 Time 0.023988 +2023-10-05 21:02:19,389 - Epoch: [43][ 220/ 1236] Overall Loss 0.304030 Objective Loss 0.304030 LR 0.001000 Time 0.023829 +2023-10-05 21:02:19,591 - Epoch: [43][ 230/ 1236] Overall Loss 0.303747 Objective Loss 0.303747 LR 0.001000 Time 0.023670 +2023-10-05 21:02:19,797 - Epoch: [43][ 240/ 1236] Overall Loss 0.304407 Objective Loss 0.304407 LR 0.001000 Time 0.023539 +2023-10-05 21:02:19,998 - Epoch: [43][ 250/ 1236] Overall Loss 0.304109 Objective Loss 0.304109 LR 0.001000 Time 0.023403 +2023-10-05 21:02:20,204 - Epoch: [43][ 260/ 1236] Overall Loss 0.305179 Objective Loss 0.305179 LR 0.001000 Time 0.023290 +2023-10-05 21:02:20,406 - Epoch: [43][ 270/ 1236] Overall Loss 0.305630 Objective Loss 0.305630 LR 0.001000 Time 0.023177 +2023-10-05 21:02:20,612 - Epoch: [43][ 280/ 1236] Overall Loss 0.304889 Objective Loss 0.304889 LR 0.001000 Time 0.023082 +2023-10-05 21:02:20,814 - Epoch: [43][ 290/ 1236] Overall Loss 0.304767 Objective Loss 0.304767 LR 0.001000 Time 0.022982 +2023-10-05 21:02:21,019 - Epoch: [43][ 300/ 1236] Overall Loss 0.305655 Objective Loss 0.305655 LR 0.001000 Time 0.022898 +2023-10-05 21:02:21,221 - Epoch: [43][ 310/ 1236] Overall Loss 0.305657 Objective Loss 0.305657 LR 0.001000 Time 0.022811 +2023-10-05 21:02:21,425 - Epoch: [43][ 320/ 1236] Overall Loss 0.305724 Objective Loss 0.305724 LR 0.001000 Time 0.022733 +2023-10-05 21:02:21,627 - Epoch: [43][ 330/ 1236] Overall Loss 0.305151 Objective Loss 0.305151 LR 0.001000 Time 0.022655 +2023-10-05 21:02:21,831 - Epoch: [43][ 340/ 1236] Overall Loss 0.304629 Objective Loss 0.304629 LR 0.001000 Time 0.022587 +2023-10-05 21:02:22,033 - Epoch: [43][ 350/ 1236] Overall Loss 0.304797 Objective Loss 0.304797 LR 0.001000 Time 0.022517 +2023-10-05 21:02:22,236 - Epoch: [43][ 360/ 1236] Overall Loss 0.304844 Objective Loss 0.304844 LR 0.001000 Time 0.022456 +2023-10-05 21:02:22,438 - Epoch: [43][ 370/ 1236] Overall Loss 0.305413 Objective Loss 0.305413 LR 0.001000 Time 0.022393 +2023-10-05 21:02:22,643 - Epoch: [43][ 380/ 1236] Overall Loss 0.305787 Objective Loss 0.305787 LR 0.001000 Time 0.022343 +2023-10-05 21:02:22,845 - Epoch: [43][ 390/ 1236] Overall Loss 0.304869 Objective Loss 0.304869 LR 0.001000 Time 0.022288 +2023-10-05 21:02:23,050 - Epoch: [43][ 400/ 1236] Overall Loss 0.305587 Objective Loss 0.305587 LR 0.001000 Time 0.022242 +2023-10-05 21:02:23,253 - Epoch: [43][ 410/ 1236] Overall Loss 0.306040 Objective Loss 0.306040 LR 0.001000 Time 0.022192 +2023-10-05 21:02:23,458 - Epoch: [43][ 420/ 1236] Overall Loss 0.306663 Objective Loss 0.306663 LR 0.001000 Time 0.022152 +2023-10-05 21:02:23,660 - Epoch: [43][ 430/ 1236] Overall Loss 0.307143 Objective Loss 0.307143 LR 0.001000 Time 0.022106 +2023-10-05 21:02:23,865 - Epoch: [43][ 440/ 1236] Overall Loss 0.308529 Objective Loss 0.308529 LR 0.001000 Time 0.022068 +2023-10-05 21:02:24,067 - Epoch: [43][ 450/ 1236] Overall Loss 0.308739 Objective Loss 0.308739 LR 0.001000 Time 0.022026 +2023-10-05 21:02:24,272 - Epoch: [43][ 460/ 1236] Overall Loss 0.309531 Objective Loss 0.309531 LR 0.001000 Time 0.021993 +2023-10-05 21:02:24,474 - Epoch: [43][ 470/ 1236] Overall Loss 0.310177 Objective Loss 0.310177 LR 0.001000 Time 0.021954 +2023-10-05 21:02:24,680 - Epoch: [43][ 480/ 1236] Overall Loss 0.311093 Objective Loss 0.311093 LR 0.001000 Time 0.021924 +2023-10-05 21:02:24,882 - Epoch: [43][ 490/ 1236] Overall Loss 0.312008 Objective Loss 0.312008 LR 0.001000 Time 0.021888 +2023-10-05 21:02:25,087 - Epoch: [43][ 500/ 1236] Overall Loss 0.312583 Objective Loss 0.312583 LR 0.001000 Time 0.021860 +2023-10-05 21:02:25,290 - Epoch: [43][ 510/ 1236] Overall Loss 0.313002 Objective Loss 0.313002 LR 0.001000 Time 0.021828 +2023-10-05 21:02:25,495 - Epoch: [43][ 520/ 1236] Overall Loss 0.313319 Objective Loss 0.313319 LR 0.001000 Time 0.021802 +2023-10-05 21:02:25,698 - Epoch: [43][ 530/ 1236] Overall Loss 0.313824 Objective Loss 0.313824 LR 0.001000 Time 0.021773 +2023-10-05 21:02:25,903 - Epoch: [43][ 540/ 1236] Overall Loss 0.313924 Objective Loss 0.313924 LR 0.001000 Time 0.021749 +2023-10-05 21:02:26,105 - Epoch: [43][ 550/ 1236] Overall Loss 0.314700 Objective Loss 0.314700 LR 0.001000 Time 0.021721 +2023-10-05 21:02:26,310 - Epoch: [43][ 560/ 1236] Overall Loss 0.315205 Objective Loss 0.315205 LR 0.001000 Time 0.021698 +2023-10-05 21:02:26,513 - Epoch: [43][ 570/ 1236] Overall Loss 0.315704 Objective Loss 0.315704 LR 0.001000 Time 0.021672 +2023-10-05 21:02:26,718 - Epoch: [43][ 580/ 1236] Overall Loss 0.315861 Objective Loss 0.315861 LR 0.001000 Time 0.021652 +2023-10-05 21:02:26,920 - Epoch: [43][ 590/ 1236] Overall Loss 0.316367 Objective Loss 0.316367 LR 0.001000 Time 0.021627 +2023-10-05 21:02:27,126 - Epoch: [43][ 600/ 1236] Overall Loss 0.316222 Objective Loss 0.316222 LR 0.001000 Time 0.021609 +2023-10-05 21:02:27,327 - Epoch: [43][ 610/ 1236] Overall Loss 0.316108 Objective Loss 0.316108 LR 0.001000 Time 0.021584 +2023-10-05 21:02:27,533 - Epoch: [43][ 620/ 1236] Overall Loss 0.315990 Objective Loss 0.315990 LR 0.001000 Time 0.021566 +2023-10-05 21:02:27,735 - Epoch: [43][ 630/ 1236] Overall Loss 0.316537 Objective Loss 0.316537 LR 0.001000 Time 0.021545 +2023-10-05 21:02:27,941 - Epoch: [43][ 640/ 1236] Overall Loss 0.316637 Objective Loss 0.316637 LR 0.001000 Time 0.021529 +2023-10-05 21:02:28,143 - Epoch: [43][ 650/ 1236] Overall Loss 0.316238 Objective Loss 0.316238 LR 0.001000 Time 0.021508 +2023-10-05 21:02:28,348 - Epoch: [43][ 660/ 1236] Overall Loss 0.316423 Objective Loss 0.316423 LR 0.001000 Time 0.021493 +2023-10-05 21:02:28,550 - Epoch: [43][ 670/ 1236] Overall Loss 0.316729 Objective Loss 0.316729 LR 0.001000 Time 0.021473 +2023-10-05 21:02:28,756 - Epoch: [43][ 680/ 1236] Overall Loss 0.317294 Objective Loss 0.317294 LR 0.001000 Time 0.021459 +2023-10-05 21:02:28,958 - Epoch: [43][ 690/ 1236] Overall Loss 0.316813 Objective Loss 0.316813 LR 0.001000 Time 0.021441 +2023-10-05 21:02:29,164 - Epoch: [43][ 700/ 1236] Overall Loss 0.317160 Objective Loss 0.317160 LR 0.001000 Time 0.021427 +2023-10-05 21:02:29,366 - Epoch: [43][ 710/ 1236] Overall Loss 0.317830 Objective Loss 0.317830 LR 0.001000 Time 0.021410 +2023-10-05 21:02:29,571 - Epoch: [43][ 720/ 1236] Overall Loss 0.317769 Objective Loss 0.317769 LR 0.001000 Time 0.021397 +2023-10-05 21:02:29,774 - Epoch: [43][ 730/ 1236] Overall Loss 0.317853 Objective Loss 0.317853 LR 0.001000 Time 0.021381 +2023-10-05 21:02:29,979 - Epoch: [43][ 740/ 1236] Overall Loss 0.318309 Objective Loss 0.318309 LR 0.001000 Time 0.021369 +2023-10-05 21:02:30,181 - Epoch: [43][ 750/ 1236] Overall Loss 0.318546 Objective Loss 0.318546 LR 0.001000 Time 0.021353 +2023-10-05 21:02:30,386 - Epoch: [43][ 760/ 1236] Overall Loss 0.318256 Objective Loss 0.318256 LR 0.001000 Time 0.021341 +2023-10-05 21:02:30,588 - Epoch: [43][ 770/ 1236] Overall Loss 0.317975 Objective Loss 0.317975 LR 0.001000 Time 0.021326 +2023-10-05 21:02:30,793 - Epoch: [43][ 780/ 1236] Overall Loss 0.318087 Objective Loss 0.318087 LR 0.001000 Time 0.021315 +2023-10-05 21:02:30,994 - Epoch: [43][ 790/ 1236] Overall Loss 0.317939 Objective Loss 0.317939 LR 0.001000 Time 0.021297 +2023-10-05 21:02:31,196 - Epoch: [43][ 800/ 1236] Overall Loss 0.318153 Objective Loss 0.318153 LR 0.001000 Time 0.021283 +2023-10-05 21:02:31,397 - Epoch: [43][ 810/ 1236] Overall Loss 0.318461 Objective Loss 0.318461 LR 0.001000 Time 0.021268 +2023-10-05 21:02:31,600 - Epoch: [43][ 820/ 1236] Overall Loss 0.318708 Objective Loss 0.318708 LR 0.001000 Time 0.021256 +2023-10-05 21:02:31,801 - Epoch: [43][ 830/ 1236] Overall Loss 0.318578 Objective Loss 0.318578 LR 0.001000 Time 0.021241 +2023-10-05 21:02:32,003 - Epoch: [43][ 840/ 1236] Overall Loss 0.318824 Objective Loss 0.318824 LR 0.001000 Time 0.021229 +2023-10-05 21:02:32,204 - Epoch: [43][ 850/ 1236] Overall Loss 0.319107 Objective Loss 0.319107 LR 0.001000 Time 0.021215 +2023-10-05 21:02:32,406 - Epoch: [43][ 860/ 1236] Overall Loss 0.319059 Objective Loss 0.319059 LR 0.001000 Time 0.021203 +2023-10-05 21:02:32,607 - Epoch: [43][ 870/ 1236] Overall Loss 0.319108 Objective Loss 0.319108 LR 0.001000 Time 0.021190 +2023-10-05 21:02:32,810 - Epoch: [43][ 880/ 1236] Overall Loss 0.319273 Objective Loss 0.319273 LR 0.001000 Time 0.021179 +2023-10-05 21:02:33,010 - Epoch: [43][ 890/ 1236] Overall Loss 0.319404 Objective Loss 0.319404 LR 0.001000 Time 0.021166 +2023-10-05 21:02:33,213 - Epoch: [43][ 900/ 1236] Overall Loss 0.319489 Objective Loss 0.319489 LR 0.001000 Time 0.021155 +2023-10-05 21:02:33,414 - Epoch: [43][ 910/ 1236] Overall Loss 0.319409 Objective Loss 0.319409 LR 0.001000 Time 0.021143 +2023-10-05 21:02:33,616 - Epoch: [43][ 920/ 1236] Overall Loss 0.319691 Objective Loss 0.319691 LR 0.001000 Time 0.021133 +2023-10-05 21:02:33,817 - Epoch: [43][ 930/ 1236] Overall Loss 0.319832 Objective Loss 0.319832 LR 0.001000 Time 0.021121 +2023-10-05 21:02:34,020 - Epoch: [43][ 940/ 1236] Overall Loss 0.319945 Objective Loss 0.319945 LR 0.001000 Time 0.021112 +2023-10-05 21:02:34,220 - Epoch: [43][ 950/ 1236] Overall Loss 0.319701 Objective Loss 0.319701 LR 0.001000 Time 0.021101 +2023-10-05 21:02:34,424 - Epoch: [43][ 960/ 1236] Overall Loss 0.319473 Objective Loss 0.319473 LR 0.001000 Time 0.021092 +2023-10-05 21:02:34,624 - Epoch: [43][ 970/ 1236] Overall Loss 0.319798 Objective Loss 0.319798 LR 0.001000 Time 0.021081 +2023-10-05 21:02:34,827 - Epoch: [43][ 980/ 1236] Overall Loss 0.319925 Objective Loss 0.319925 LR 0.001000 Time 0.021072 +2023-10-05 21:02:35,028 - Epoch: [43][ 990/ 1236] Overall Loss 0.320051 Objective Loss 0.320051 LR 0.001000 Time 0.021062 +2023-10-05 21:02:35,230 - Epoch: [43][ 1000/ 1236] Overall Loss 0.320210 Objective Loss 0.320210 LR 0.001000 Time 0.021053 +2023-10-05 21:02:35,430 - Epoch: [43][ 1010/ 1236] Overall Loss 0.320249 Objective Loss 0.320249 LR 0.001000 Time 0.021043 +2023-10-05 21:02:35,634 - Epoch: [43][ 1020/ 1236] Overall Loss 0.320184 Objective Loss 0.320184 LR 0.001000 Time 0.021035 +2023-10-05 21:02:35,834 - Epoch: [43][ 1030/ 1236] Overall Loss 0.320340 Objective Loss 0.320340 LR 0.001000 Time 0.021025 +2023-10-05 21:02:36,037 - Epoch: [43][ 1040/ 1236] Overall Loss 0.320259 Objective Loss 0.320259 LR 0.001000 Time 0.021018 +2023-10-05 21:02:36,237 - Epoch: [43][ 1050/ 1236] Overall Loss 0.320181 Objective Loss 0.320181 LR 0.001000 Time 0.021008 +2023-10-05 21:02:36,440 - Epoch: [43][ 1060/ 1236] Overall Loss 0.319994 Objective Loss 0.319994 LR 0.001000 Time 0.021001 +2023-10-05 21:02:36,640 - Epoch: [43][ 1070/ 1236] Overall Loss 0.320746 Objective Loss 0.320746 LR 0.001000 Time 0.020992 +2023-10-05 21:02:36,843 - Epoch: [43][ 1080/ 1236] Overall Loss 0.321064 Objective Loss 0.321064 LR 0.001000 Time 0.020984 +2023-10-05 21:02:37,043 - Epoch: [43][ 1090/ 1236] Overall Loss 0.321066 Objective Loss 0.321066 LR 0.001000 Time 0.020975 +2023-10-05 21:02:37,246 - Epoch: [43][ 1100/ 1236] Overall Loss 0.321109 Objective Loss 0.321109 LR 0.001000 Time 0.020969 +2023-10-05 21:02:37,446 - Epoch: [43][ 1110/ 1236] Overall Loss 0.321338 Objective Loss 0.321338 LR 0.001000 Time 0.020960 +2023-10-05 21:02:37,649 - Epoch: [43][ 1120/ 1236] Overall Loss 0.321357 Objective Loss 0.321357 LR 0.001000 Time 0.020953 +2023-10-05 21:02:37,849 - Epoch: [43][ 1130/ 1236] Overall Loss 0.321410 Objective Loss 0.321410 LR 0.001000 Time 0.020945 +2023-10-05 21:02:38,052 - Epoch: [43][ 1140/ 1236] Overall Loss 0.321476 Objective Loss 0.321476 LR 0.001000 Time 0.020938 +2023-10-05 21:02:38,253 - Epoch: [43][ 1150/ 1236] Overall Loss 0.321557 Objective Loss 0.321557 LR 0.001000 Time 0.020930 +2023-10-05 21:02:38,455 - Epoch: [43][ 1160/ 1236] Overall Loss 0.321580 Objective Loss 0.321580 LR 0.001000 Time 0.020924 +2023-10-05 21:02:38,657 - Epoch: [43][ 1170/ 1236] Overall Loss 0.321515 Objective Loss 0.321515 LR 0.001000 Time 0.020917 +2023-10-05 21:02:38,859 - Epoch: [43][ 1180/ 1236] Overall Loss 0.321522 Objective Loss 0.321522 LR 0.001000 Time 0.020911 +2023-10-05 21:02:39,059 - Epoch: [43][ 1190/ 1236] Overall Loss 0.321522 Objective Loss 0.321522 LR 0.001000 Time 0.020904 +2023-10-05 21:02:39,262 - Epoch: [43][ 1200/ 1236] Overall Loss 0.321837 Objective Loss 0.321837 LR 0.001000 Time 0.020898 +2023-10-05 21:02:39,463 - Epoch: [43][ 1210/ 1236] Overall Loss 0.321902 Objective Loss 0.321902 LR 0.001000 Time 0.020891 +2023-10-05 21:02:39,666 - Epoch: [43][ 1220/ 1236] Overall Loss 0.321939 Objective Loss 0.321939 LR 0.001000 Time 0.020886 +2023-10-05 21:02:39,921 - Epoch: [43][ 1230/ 1236] Overall Loss 0.321835 Objective Loss 0.321835 LR 0.001000 Time 0.020923 +2023-10-05 21:02:40,039 - Epoch: [43][ 1236/ 1236] Overall Loss 0.321837 Objective Loss 0.321837 Top1 84.114053 Top5 97.759674 LR 0.001000 Time 0.020917 +2023-10-05 21:02:40,160 - --- validate (epoch=43)----------- +2023-10-05 21:02:40,161 - 29943 samples (256 per mini-batch) +2023-10-05 21:02:40,620 - Epoch: [43][ 10/ 117] Loss 0.361969 Top1 81.054688 Top5 97.695312 +2023-10-05 21:02:40,766 - Epoch: [43][ 20/ 117] Loss 0.367051 Top1 81.269531 Top5 97.910156 +2023-10-05 21:02:40,912 - Epoch: [43][ 30/ 117] Loss 0.381627 Top1 81.080729 Top5 97.526042 +2023-10-05 21:02:41,059 - Epoch: [43][ 40/ 117] Loss 0.380389 Top1 81.142578 Top5 97.646484 +2023-10-05 21:02:41,205 - Epoch: [43][ 50/ 117] Loss 0.375452 Top1 81.218750 Top5 97.679688 +2023-10-05 21:02:41,352 - Epoch: [43][ 60/ 117] Loss 0.380281 Top1 81.191406 Top5 97.584635 +2023-10-05 21:02:41,499 - Epoch: [43][ 70/ 117] Loss 0.382616 Top1 81.099330 Top5 97.477679 +2023-10-05 21:02:41,646 - Epoch: [43][ 80/ 117] Loss 0.384127 Top1 81.123047 Top5 97.509766 +2023-10-05 21:02:41,793 - Epoch: [43][ 90/ 117] Loss 0.382446 Top1 81.236979 Top5 97.513021 +2023-10-05 21:02:41,940 - Epoch: [43][ 100/ 117] Loss 0.379477 Top1 81.343750 Top5 97.562500 +2023-10-05 21:02:42,093 - Epoch: [43][ 110/ 117] Loss 0.377859 Top1 81.416903 Top5 97.581676 +2023-10-05 21:02:42,178 - Epoch: [43][ 117/ 117] Loss 0.377874 Top1 81.468123 Top5 97.572054 +2023-10-05 21:02:42,309 - ==> Top1: 81.468 Top5: 97.572 Loss: 0.378 + +2023-10-05 21:02:42,310 - ==> Confusion: +[[ 939 3 5 1 7 1 0 1 5 49 1 1 2 4 5 3 3 3 0 0 17] + [ 1 1020 1 0 4 36 1 30 2 0 4 2 0 3 2 5 6 0 10 0 4] + [ 6 0 925 22 3 1 27 5 0 2 11 4 10 4 1 6 3 2 7 3 14] + [ 4 0 22 933 2 5 0 0 4 1 15 0 9 5 32 8 5 4 17 0 23] + [ 37 5 1 1 955 9 0 1 0 8 0 1 1 1 7 5 10 2 0 0 6] + [ 3 36 0 1 6 985 2 26 0 0 6 14 3 13 1 3 5 1 0 0 11] + [ 0 7 41 0 1 1 1089 19 0 0 5 3 3 0 0 7 0 0 2 5 8] + [ 5 21 18 0 3 31 3 1054 0 2 6 12 0 3 0 2 0 0 45 4 9] + [ 21 3 0 1 1 6 0 2 949 35 13 3 3 18 16 5 3 0 3 0 7] + [ 144 1 0 0 6 5 0 0 26 871 0 1 0 34 6 12 2 1 0 0 10] + [ 4 0 11 8 1 3 3 5 21 2 947 2 0 19 4 4 1 0 6 1 11] + [ 1 1 0 0 1 12 0 1 0 0 0 970 19 5 0 2 2 12 0 4 5] + [ 0 1 5 6 0 3 0 1 2 0 0 74 932 1 1 13 2 10 1 0 16] + [ 3 0 0 1 4 14 0 1 10 7 3 9 3 1044 2 3 4 1 0 2 8] + [ 16 2 3 8 6 0 0 0 34 7 1 1 5 0 988 0 1 2 5 0 22] + [ 2 5 1 0 3 0 2 0 0 0 0 11 9 0 1 1073 12 7 0 4 4] + [ 2 12 1 0 7 6 0 0 0 0 0 2 1 1 1 15 1094 0 0 3 16] + [ 0 0 0 0 1 0 1 0 0 0 0 20 39 2 2 15 0 953 1 1 3] + [ 1 8 9 14 0 1 0 35 3 0 3 4 8 1 17 1 2 0 944 0 17] + [ 0 6 3 0 1 7 5 18 0 0 0 27 7 1 1 11 11 1 3 1027 23] + [ 203 232 164 57 102 210 50 127 102 72 192 191 353 331 121 97 227 70 126 176 4702]] + +2023-10-05 21:02:42,311 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:02:42,311 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:02:42,317 - + +2023-10-05 21:02:42,317 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:02:43,434 - Epoch: [44][ 10/ 1236] Overall Loss 0.311019 Objective Loss 0.311019 LR 0.001000 Time 0.111579 +2023-10-05 21:02:43,637 - Epoch: [44][ 20/ 1236] Overall Loss 0.306970 Objective Loss 0.306970 LR 0.001000 Time 0.065962 +2023-10-05 21:02:43,840 - Epoch: [44][ 30/ 1236] Overall Loss 0.312220 Objective Loss 0.312220 LR 0.001000 Time 0.050735 +2023-10-05 21:02:44,044 - Epoch: [44][ 40/ 1236] Overall Loss 0.311695 Objective Loss 0.311695 LR 0.001000 Time 0.043139 +2023-10-05 21:02:44,248 - Epoch: [44][ 50/ 1236] Overall Loss 0.308013 Objective Loss 0.308013 LR 0.001000 Time 0.038574 +2023-10-05 21:02:44,452 - Epoch: [44][ 60/ 1236] Overall Loss 0.310294 Objective Loss 0.310294 LR 0.001000 Time 0.035538 +2023-10-05 21:02:44,655 - Epoch: [44][ 70/ 1236] Overall Loss 0.311093 Objective Loss 0.311093 LR 0.001000 Time 0.033366 +2023-10-05 21:02:44,859 - Epoch: [44][ 80/ 1236] Overall Loss 0.308713 Objective Loss 0.308713 LR 0.001000 Time 0.031741 +2023-10-05 21:02:45,063 - Epoch: [44][ 90/ 1236] Overall Loss 0.310898 Objective Loss 0.310898 LR 0.001000 Time 0.030471 +2023-10-05 21:02:45,267 - Epoch: [44][ 100/ 1236] Overall Loss 0.309492 Objective Loss 0.309492 LR 0.001000 Time 0.029459 +2023-10-05 21:02:45,468 - Epoch: [44][ 110/ 1236] Overall Loss 0.312035 Objective Loss 0.312035 LR 0.001000 Time 0.028611 +2023-10-05 21:02:45,669 - Epoch: [44][ 120/ 1236] Overall Loss 0.312930 Objective Loss 0.312930 LR 0.001000 Time 0.027901 +2023-10-05 21:02:45,870 - Epoch: [44][ 130/ 1236] Overall Loss 0.314447 Objective Loss 0.314447 LR 0.001000 Time 0.027292 +2023-10-05 21:02:46,071 - Epoch: [44][ 140/ 1236] Overall Loss 0.314922 Objective Loss 0.314922 LR 0.001000 Time 0.026779 +2023-10-05 21:02:46,270 - Epoch: [44][ 150/ 1236] Overall Loss 0.313922 Objective Loss 0.313922 LR 0.001000 Time 0.026320 +2023-10-05 21:02:46,472 - Epoch: [44][ 160/ 1236] Overall Loss 0.315031 Objective Loss 0.315031 LR 0.001000 Time 0.025932 +2023-10-05 21:02:46,671 - Epoch: [44][ 170/ 1236] Overall Loss 0.316917 Objective Loss 0.316917 LR 0.001000 Time 0.025578 +2023-10-05 21:02:46,873 - Epoch: [44][ 180/ 1236] Overall Loss 0.315405 Objective Loss 0.315405 LR 0.001000 Time 0.025274 +2023-10-05 21:02:47,073 - Epoch: [44][ 190/ 1236] Overall Loss 0.315147 Objective Loss 0.315147 LR 0.001000 Time 0.024993 +2023-10-05 21:02:47,274 - Epoch: [44][ 200/ 1236] Overall Loss 0.314235 Objective Loss 0.314235 LR 0.001000 Time 0.024748 +2023-10-05 21:02:47,474 - Epoch: [44][ 210/ 1236] Overall Loss 0.313890 Objective Loss 0.313890 LR 0.001000 Time 0.024518 +2023-10-05 21:02:47,675 - Epoch: [44][ 220/ 1236] Overall Loss 0.314748 Objective Loss 0.314748 LR 0.001000 Time 0.024318 +2023-10-05 21:02:47,875 - Epoch: [44][ 230/ 1236] Overall Loss 0.314837 Objective Loss 0.314837 LR 0.001000 Time 0.024129 +2023-10-05 21:02:48,077 - Epoch: [44][ 240/ 1236] Overall Loss 0.314479 Objective Loss 0.314479 LR 0.001000 Time 0.023962 +2023-10-05 21:02:48,276 - Epoch: [44][ 250/ 1236] Overall Loss 0.314407 Objective Loss 0.314407 LR 0.001000 Time 0.023801 +2023-10-05 21:02:48,478 - Epoch: [44][ 260/ 1236] Overall Loss 0.314774 Objective Loss 0.314774 LR 0.001000 Time 0.023658 +2023-10-05 21:02:48,678 - Epoch: [44][ 270/ 1236] Overall Loss 0.314801 Objective Loss 0.314801 LR 0.001000 Time 0.023521 +2023-10-05 21:02:48,879 - Epoch: [44][ 280/ 1236] Overall Loss 0.314349 Objective Loss 0.314349 LR 0.001000 Time 0.023400 +2023-10-05 21:02:49,079 - Epoch: [44][ 290/ 1236] Overall Loss 0.314782 Objective Loss 0.314782 LR 0.001000 Time 0.023282 +2023-10-05 21:02:49,281 - Epoch: [44][ 300/ 1236] Overall Loss 0.315114 Objective Loss 0.315114 LR 0.001000 Time 0.023176 +2023-10-05 21:02:49,481 - Epoch: [44][ 310/ 1236] Overall Loss 0.313898 Objective Loss 0.313898 LR 0.001000 Time 0.023072 +2023-10-05 21:02:49,683 - Epoch: [44][ 320/ 1236] Overall Loss 0.314143 Objective Loss 0.314143 LR 0.001000 Time 0.022982 +2023-10-05 21:02:49,883 - Epoch: [44][ 330/ 1236] Overall Loss 0.313995 Objective Loss 0.313995 LR 0.001000 Time 0.022890 +2023-10-05 21:02:50,084 - Epoch: [44][ 340/ 1236] Overall Loss 0.313581 Objective Loss 0.313581 LR 0.001000 Time 0.022809 +2023-10-05 21:02:50,284 - Epoch: [44][ 350/ 1236] Overall Loss 0.313858 Objective Loss 0.313858 LR 0.001000 Time 0.022727 +2023-10-05 21:02:50,486 - Epoch: [44][ 360/ 1236] Overall Loss 0.313854 Objective Loss 0.313854 LR 0.001000 Time 0.022655 +2023-10-05 21:02:50,686 - Epoch: [44][ 370/ 1236] Overall Loss 0.314162 Objective Loss 0.314162 LR 0.001000 Time 0.022584 +2023-10-05 21:02:50,888 - Epoch: [44][ 380/ 1236] Overall Loss 0.313932 Objective Loss 0.313932 LR 0.001000 Time 0.022519 +2023-10-05 21:02:51,088 - Epoch: [44][ 390/ 1236] Overall Loss 0.313962 Objective Loss 0.313962 LR 0.001000 Time 0.022453 +2023-10-05 21:02:51,290 - Epoch: [44][ 400/ 1236] Overall Loss 0.312995 Objective Loss 0.312995 LR 0.001000 Time 0.022395 +2023-10-05 21:02:51,490 - Epoch: [44][ 410/ 1236] Overall Loss 0.313924 Objective Loss 0.313924 LR 0.001000 Time 0.022336 +2023-10-05 21:02:51,691 - Epoch: [44][ 420/ 1236] Overall Loss 0.314673 Objective Loss 0.314673 LR 0.001000 Time 0.022282 +2023-10-05 21:02:51,891 - Epoch: [44][ 430/ 1236] Overall Loss 0.314032 Objective Loss 0.314032 LR 0.001000 Time 0.022229 +2023-10-05 21:02:52,093 - Epoch: [44][ 440/ 1236] Overall Loss 0.314206 Objective Loss 0.314206 LR 0.001000 Time 0.022181 +2023-10-05 21:02:52,293 - Epoch: [44][ 450/ 1236] Overall Loss 0.314694 Objective Loss 0.314694 LR 0.001000 Time 0.022131 +2023-10-05 21:02:52,494 - Epoch: [44][ 460/ 1236] Overall Loss 0.314896 Objective Loss 0.314896 LR 0.001000 Time 0.022088 +2023-10-05 21:02:52,694 - Epoch: [44][ 470/ 1236] Overall Loss 0.314662 Objective Loss 0.314662 LR 0.001000 Time 0.022043 +2023-10-05 21:02:52,896 - Epoch: [44][ 480/ 1236] Overall Loss 0.314944 Objective Loss 0.314944 LR 0.001000 Time 0.022003 +2023-10-05 21:02:53,096 - Epoch: [44][ 490/ 1236] Overall Loss 0.314684 Objective Loss 0.314684 LR 0.001000 Time 0.021961 +2023-10-05 21:02:53,298 - Epoch: [44][ 500/ 1236] Overall Loss 0.315028 Objective Loss 0.315028 LR 0.001000 Time 0.021925 +2023-10-05 21:02:53,498 - Epoch: [44][ 510/ 1236] Overall Loss 0.316300 Objective Loss 0.316300 LR 0.001000 Time 0.021886 +2023-10-05 21:02:53,699 - Epoch: [44][ 520/ 1236] Overall Loss 0.316201 Objective Loss 0.316201 LR 0.001000 Time 0.021853 +2023-10-05 21:02:53,899 - Epoch: [44][ 530/ 1236] Overall Loss 0.316125 Objective Loss 0.316125 LR 0.001000 Time 0.021817 +2023-10-05 21:02:54,101 - Epoch: [44][ 540/ 1236] Overall Loss 0.316087 Objective Loss 0.316087 LR 0.001000 Time 0.021786 +2023-10-05 21:02:54,301 - Epoch: [44][ 550/ 1236] Overall Loss 0.316440 Objective Loss 0.316440 LR 0.001000 Time 0.021753 +2023-10-05 21:02:54,503 - Epoch: [44][ 560/ 1236] Overall Loss 0.316126 Objective Loss 0.316126 LR 0.001000 Time 0.021724 +2023-10-05 21:02:54,703 - Epoch: [44][ 570/ 1236] Overall Loss 0.316743 Objective Loss 0.316743 LR 0.001000 Time 0.021693 +2023-10-05 21:02:54,904 - Epoch: [44][ 580/ 1236] Overall Loss 0.316677 Objective Loss 0.316677 LR 0.001000 Time 0.021666 +2023-10-05 21:02:55,104 - Epoch: [44][ 590/ 1236] Overall Loss 0.316813 Objective Loss 0.316813 LR 0.001000 Time 0.021637 +2023-10-05 21:02:55,306 - Epoch: [44][ 600/ 1236] Overall Loss 0.316992 Objective Loss 0.316992 LR 0.001000 Time 0.021612 +2023-10-05 21:02:55,506 - Epoch: [44][ 610/ 1236] Overall Loss 0.317130 Objective Loss 0.317130 LR 0.001000 Time 0.021585 +2023-10-05 21:02:55,708 - Epoch: [44][ 620/ 1236] Overall Loss 0.317176 Objective Loss 0.317176 LR 0.001000 Time 0.021562 +2023-10-05 21:02:55,908 - Epoch: [44][ 630/ 1236] Overall Loss 0.317086 Objective Loss 0.317086 LR 0.001000 Time 0.021536 +2023-10-05 21:02:56,109 - Epoch: [44][ 640/ 1236] Overall Loss 0.317582 Objective Loss 0.317582 LR 0.001000 Time 0.021513 +2023-10-05 21:02:56,308 - Epoch: [44][ 650/ 1236] Overall Loss 0.317792 Objective Loss 0.317792 LR 0.001000 Time 0.021489 +2023-10-05 21:02:56,510 - Epoch: [44][ 660/ 1236] Overall Loss 0.317466 Objective Loss 0.317466 LR 0.001000 Time 0.021468 +2023-10-05 21:02:56,710 - Epoch: [44][ 670/ 1236] Overall Loss 0.317334 Objective Loss 0.317334 LR 0.001000 Time 0.021446 +2023-10-05 21:02:56,912 - Epoch: [44][ 680/ 1236] Overall Loss 0.317733 Objective Loss 0.317733 LR 0.001000 Time 0.021427 +2023-10-05 21:02:57,112 - Epoch: [44][ 690/ 1236] Overall Loss 0.317781 Objective Loss 0.317781 LR 0.001000 Time 0.021405 +2023-10-05 21:02:57,313 - Epoch: [44][ 700/ 1236] Overall Loss 0.317999 Objective Loss 0.317999 LR 0.001000 Time 0.021387 +2023-10-05 21:02:57,513 - Epoch: [44][ 710/ 1236] Overall Loss 0.317961 Objective Loss 0.317961 LR 0.001000 Time 0.021367 +2023-10-05 21:02:57,714 - Epoch: [44][ 720/ 1236] Overall Loss 0.318214 Objective Loss 0.318214 LR 0.001000 Time 0.021348 +2023-10-05 21:02:57,913 - Epoch: [44][ 730/ 1236] Overall Loss 0.318409 Objective Loss 0.318409 LR 0.001000 Time 0.021329 +2023-10-05 21:02:58,115 - Epoch: [44][ 740/ 1236] Overall Loss 0.318650 Objective Loss 0.318650 LR 0.001000 Time 0.021313 +2023-10-05 21:02:58,315 - Epoch: [44][ 750/ 1236] Overall Loss 0.319051 Objective Loss 0.319051 LR 0.001000 Time 0.021295 +2023-10-05 21:02:58,517 - Epoch: [44][ 760/ 1236] Overall Loss 0.319693 Objective Loss 0.319693 LR 0.001000 Time 0.021279 +2023-10-05 21:02:58,717 - Epoch: [44][ 770/ 1236] Overall Loss 0.320094 Objective Loss 0.320094 LR 0.001000 Time 0.021262 +2023-10-05 21:02:58,919 - Epoch: [44][ 780/ 1236] Overall Loss 0.319934 Objective Loss 0.319934 LR 0.001000 Time 0.021248 +2023-10-05 21:02:59,118 - Epoch: [44][ 790/ 1236] Overall Loss 0.320189 Objective Loss 0.320189 LR 0.001000 Time 0.021231 +2023-10-05 21:02:59,320 - Epoch: [44][ 800/ 1236] Overall Loss 0.319934 Objective Loss 0.319934 LR 0.001000 Time 0.021218 +2023-10-05 21:02:59,519 - Epoch: [44][ 810/ 1236] Overall Loss 0.320307 Objective Loss 0.320307 LR 0.001000 Time 0.021201 +2023-10-05 21:02:59,721 - Epoch: [44][ 820/ 1236] Overall Loss 0.320808 Objective Loss 0.320808 LR 0.001000 Time 0.021188 +2023-10-05 21:02:59,919 - Epoch: [44][ 830/ 1236] Overall Loss 0.320427 Objective Loss 0.320427 LR 0.001000 Time 0.021172 +2023-10-05 21:03:00,121 - Epoch: [44][ 840/ 1236] Overall Loss 0.320714 Objective Loss 0.320714 LR 0.001000 Time 0.021159 +2023-10-05 21:03:00,320 - Epoch: [44][ 850/ 1236] Overall Loss 0.320856 Objective Loss 0.320856 LR 0.001000 Time 0.021144 +2023-10-05 21:03:00,522 - Epoch: [44][ 860/ 1236] Overall Loss 0.320403 Objective Loss 0.320403 LR 0.001000 Time 0.021133 +2023-10-05 21:03:00,722 - Epoch: [44][ 870/ 1236] Overall Loss 0.320327 Objective Loss 0.320327 LR 0.001000 Time 0.021120 +2023-10-05 21:03:00,924 - Epoch: [44][ 880/ 1236] Overall Loss 0.320638 Objective Loss 0.320638 LR 0.001000 Time 0.021108 +2023-10-05 21:03:01,124 - Epoch: [44][ 890/ 1236] Overall Loss 0.320762 Objective Loss 0.320762 LR 0.001000 Time 0.021095 +2023-10-05 21:03:01,325 - Epoch: [44][ 900/ 1236] Overall Loss 0.320952 Objective Loss 0.320952 LR 0.001000 Time 0.021084 +2023-10-05 21:03:01,524 - Epoch: [44][ 910/ 1236] Overall Loss 0.320772 Objective Loss 0.320772 LR 0.001000 Time 0.021071 +2023-10-05 21:03:01,727 - Epoch: [44][ 920/ 1236] Overall Loss 0.320789 Objective Loss 0.320789 LR 0.001000 Time 0.021061 +2023-10-05 21:03:01,926 - Epoch: [44][ 930/ 1236] Overall Loss 0.320847 Objective Loss 0.320847 LR 0.001000 Time 0.021049 +2023-10-05 21:03:02,128 - Epoch: [44][ 940/ 1236] Overall Loss 0.320786 Objective Loss 0.320786 LR 0.001000 Time 0.021039 +2023-10-05 21:03:02,328 - Epoch: [44][ 950/ 1236] Overall Loss 0.320859 Objective Loss 0.320859 LR 0.001000 Time 0.021028 +2023-10-05 21:03:02,529 - Epoch: [44][ 960/ 1236] Overall Loss 0.320818 Objective Loss 0.320818 LR 0.001000 Time 0.021018 +2023-10-05 21:03:02,728 - Epoch: [44][ 970/ 1236] Overall Loss 0.320928 Objective Loss 0.320928 LR 0.001000 Time 0.021007 +2023-10-05 21:03:02,930 - Epoch: [44][ 980/ 1236] Overall Loss 0.320620 Objective Loss 0.320620 LR 0.001000 Time 0.020998 +2023-10-05 21:03:03,130 - Epoch: [44][ 990/ 1236] Overall Loss 0.320483 Objective Loss 0.320483 LR 0.001000 Time 0.020987 +2023-10-05 21:03:03,332 - Epoch: [44][ 1000/ 1236] Overall Loss 0.320685 Objective Loss 0.320685 LR 0.001000 Time 0.020979 +2023-10-05 21:03:03,532 - Epoch: [44][ 1010/ 1236] Overall Loss 0.320775 Objective Loss 0.320775 LR 0.001000 Time 0.020969 +2023-10-05 21:03:03,734 - Epoch: [44][ 1020/ 1236] Overall Loss 0.320947 Objective Loss 0.320947 LR 0.001000 Time 0.020961 +2023-10-05 21:03:03,933 - Epoch: [44][ 1030/ 1236] Overall Loss 0.321188 Objective Loss 0.321188 LR 0.001000 Time 0.020951 +2023-10-05 21:03:04,135 - Epoch: [44][ 1040/ 1236] Overall Loss 0.321454 Objective Loss 0.321454 LR 0.001000 Time 0.020943 +2023-10-05 21:03:04,343 - Epoch: [44][ 1050/ 1236] Overall Loss 0.321251 Objective Loss 0.321251 LR 0.001000 Time 0.020941 +2023-10-05 21:03:04,555 - Epoch: [44][ 1060/ 1236] Overall Loss 0.321207 Objective Loss 0.321207 LR 0.001000 Time 0.020943 +2023-10-05 21:03:04,763 - Epoch: [44][ 1070/ 1236] Overall Loss 0.321122 Objective Loss 0.321122 LR 0.001000 Time 0.020942 +2023-10-05 21:03:04,975 - Epoch: [44][ 1080/ 1236] Overall Loss 0.321153 Objective Loss 0.321153 LR 0.001000 Time 0.020944 +2023-10-05 21:03:05,183 - Epoch: [44][ 1090/ 1236] Overall Loss 0.321481 Objective Loss 0.321481 LR 0.001000 Time 0.020942 +2023-10-05 21:03:05,394 - Epoch: [44][ 1100/ 1236] Overall Loss 0.321194 Objective Loss 0.321194 LR 0.001000 Time 0.020943 +2023-10-05 21:03:05,602 - Epoch: [44][ 1110/ 1236] Overall Loss 0.321215 Objective Loss 0.321215 LR 0.001000 Time 0.020941 +2023-10-05 21:03:05,814 - Epoch: [44][ 1120/ 1236] Overall Loss 0.321201 Objective Loss 0.321201 LR 0.001000 Time 0.020944 +2023-10-05 21:03:06,022 - Epoch: [44][ 1130/ 1236] Overall Loss 0.321351 Objective Loss 0.321351 LR 0.001000 Time 0.020942 +2023-10-05 21:03:06,234 - Epoch: [44][ 1140/ 1236] Overall Loss 0.321151 Objective Loss 0.321151 LR 0.001000 Time 0.020944 +2023-10-05 21:03:06,442 - Epoch: [44][ 1150/ 1236] Overall Loss 0.321247 Objective Loss 0.321247 LR 0.001000 Time 0.020942 +2023-10-05 21:03:06,654 - Epoch: [44][ 1160/ 1236] Overall Loss 0.321142 Objective Loss 0.321142 LR 0.001000 Time 0.020944 +2023-10-05 21:03:06,862 - Epoch: [44][ 1170/ 1236] Overall Loss 0.321313 Objective Loss 0.321313 LR 0.001000 Time 0.020943 +2023-10-05 21:03:07,074 - Epoch: [44][ 1180/ 1236] Overall Loss 0.321272 Objective Loss 0.321272 LR 0.001000 Time 0.020944 +2023-10-05 21:03:07,282 - Epoch: [44][ 1190/ 1236] Overall Loss 0.321701 Objective Loss 0.321701 LR 0.001000 Time 0.020943 +2023-10-05 21:03:07,494 - Epoch: [44][ 1200/ 1236] Overall Loss 0.321616 Objective Loss 0.321616 LR 0.001000 Time 0.020945 +2023-10-05 21:03:07,702 - Epoch: [44][ 1210/ 1236] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.020943 +2023-10-05 21:03:07,914 - Epoch: [44][ 1220/ 1236] Overall Loss 0.321860 Objective Loss 0.321860 LR 0.001000 Time 0.020945 +2023-10-05 21:03:08,172 - Epoch: [44][ 1230/ 1236] Overall Loss 0.321856 Objective Loss 0.321856 LR 0.001000 Time 0.020985 +2023-10-05 21:03:08,291 - Epoch: [44][ 1236/ 1236] Overall Loss 0.321974 Objective Loss 0.321974 Top1 84.928717 Top5 97.963340 LR 0.001000 Time 0.020979 +2023-10-05 21:03:08,429 - --- validate (epoch=44)----------- +2023-10-05 21:03:08,429 - 29943 samples (256 per mini-batch) +2023-10-05 21:03:08,886 - Epoch: [44][ 10/ 117] Loss 0.351843 Top1 80.703125 Top5 97.187500 +2023-10-05 21:03:09,034 - Epoch: [44][ 20/ 117] Loss 0.369155 Top1 80.078125 Top5 97.031250 +2023-10-05 21:03:09,182 - Epoch: [44][ 30/ 117] Loss 0.372620 Top1 79.934896 Top5 97.135417 +2023-10-05 21:03:09,330 - Epoch: [44][ 40/ 117] Loss 0.367334 Top1 80.195312 Top5 97.216797 +2023-10-05 21:03:09,478 - Epoch: [44][ 50/ 117] Loss 0.367926 Top1 80.203125 Top5 97.296875 +2023-10-05 21:03:09,627 - Epoch: [44][ 60/ 117] Loss 0.370631 Top1 80.286458 Top5 97.226562 +2023-10-05 21:03:09,774 - Epoch: [44][ 70/ 117] Loss 0.375129 Top1 80.089286 Top5 97.215402 +2023-10-05 21:03:09,923 - Epoch: [44][ 80/ 117] Loss 0.377582 Top1 79.926758 Top5 97.221680 +2023-10-05 21:03:10,070 - Epoch: [44][ 90/ 117] Loss 0.377198 Top1 79.882812 Top5 97.226562 +2023-10-05 21:03:10,218 - Epoch: [44][ 100/ 117] Loss 0.378347 Top1 79.921875 Top5 97.238281 +2023-10-05 21:03:10,372 - Epoch: [44][ 110/ 117] Loss 0.379409 Top1 79.790483 Top5 97.194602 +2023-10-05 21:03:10,458 - Epoch: [44][ 117/ 117] Loss 0.379310 Top1 79.814982 Top5 97.221387 +2023-10-05 21:03:10,571 - ==> Top1: 79.815 Top5: 97.221 Loss: 0.379 + +2023-10-05 21:03:10,572 - ==> Confusion: +[[ 959 3 2 2 16 1 0 0 4 35 1 0 1 4 6 2 4 1 0 1 8] + [ 3 1058 4 0 5 16 3 21 1 0 3 0 0 0 1 3 7 1 1 0 4] + [ 13 2 925 11 12 1 38 15 0 1 2 4 5 4 1 0 1 1 7 7 6] + [ 3 1 27 930 2 6 3 0 5 0 4 0 8 7 43 4 7 7 16 4 12] + [ 26 8 3 1 974 3 0 0 0 7 0 1 0 0 7 2 11 3 0 1 3] + [ 7 49 0 0 6 969 6 28 0 1 1 11 2 9 6 3 5 0 0 4 9] + [ 0 5 28 2 1 0 1114 10 0 0 3 3 2 0 0 5 0 1 2 10 5] + [ 9 27 19 1 9 47 4 1027 0 2 5 7 1 2 1 0 2 2 41 6 6] + [ 39 1 0 1 1 13 2 1 900 55 10 2 1 22 28 2 4 1 3 0 3] + [ 169 1 1 1 12 6 0 0 15 855 0 2 0 39 2 7 2 0 0 1 6] + [ 6 9 13 2 2 0 4 4 13 2 950 0 0 15 7 6 2 1 5 2 10] + [ 1 1 2 0 1 20 0 1 1 0 0 935 29 3 0 4 4 16 0 14 3] + [ 0 2 3 7 1 6 1 5 1 0 0 45 942 0 1 18 4 14 2 9 7] + [ 2 0 1 1 6 22 0 0 9 15 5 6 1 1034 2 3 2 0 0 4 6] + [ 26 2 2 15 14 0 0 0 27 5 2 1 0 2 979 0 5 3 9 0 9] + [ 1 4 1 1 4 1 3 1 0 1 0 10 8 0 1 1058 15 14 0 7 4] + [ 2 17 3 0 10 4 0 1 1 0 0 3 3 1 3 8 1091 1 0 3 10] + [ 2 1 0 2 1 1 1 0 0 3 0 8 26 1 3 13 0 969 1 0 6] + [ 2 16 16 15 2 1 3 33 3 0 5 0 1 0 23 0 1 0 932 5 10] + [ 0 1 4 0 1 8 11 17 0 0 1 22 2 0 0 7 12 2 1 1055 8] + [ 263 346 189 55 190 226 62 130 85 80 204 132 368 369 130 101 293 70 113 256 4243]] + +2023-10-05 21:03:10,573 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:03:10,573 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:03:10,579 - + +2023-10-05 21:03:10,579 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:03:11,592 - Epoch: [45][ 10/ 1236] Overall Loss 0.299079 Objective Loss 0.299079 LR 0.001000 Time 0.101180 +2023-10-05 21:03:11,794 - Epoch: [45][ 20/ 1236] Overall Loss 0.307502 Objective Loss 0.307502 LR 0.001000 Time 0.060686 +2023-10-05 21:03:11,993 - Epoch: [45][ 30/ 1236] Overall Loss 0.303646 Objective Loss 0.303646 LR 0.001000 Time 0.047101 +2023-10-05 21:03:12,195 - Epoch: [45][ 40/ 1236] Overall Loss 0.299083 Objective Loss 0.299083 LR 0.001000 Time 0.040359 +2023-10-05 21:03:12,395 - Epoch: [45][ 50/ 1236] Overall Loss 0.296916 Objective Loss 0.296916 LR 0.001000 Time 0.036283 +2023-10-05 21:03:12,597 - Epoch: [45][ 60/ 1236] Overall Loss 0.301218 Objective Loss 0.301218 LR 0.001000 Time 0.033589 +2023-10-05 21:03:12,798 - Epoch: [45][ 70/ 1236] Overall Loss 0.302548 Objective Loss 0.302548 LR 0.001000 Time 0.031655 +2023-10-05 21:03:13,002 - Epoch: [45][ 80/ 1236] Overall Loss 0.302967 Objective Loss 0.302967 LR 0.001000 Time 0.030249 +2023-10-05 21:03:13,203 - Epoch: [45][ 90/ 1236] Overall Loss 0.308884 Objective Loss 0.308884 LR 0.001000 Time 0.029118 +2023-10-05 21:03:13,408 - Epoch: [45][ 100/ 1236] Overall Loss 0.307645 Objective Loss 0.307645 LR 0.001000 Time 0.028244 +2023-10-05 21:03:13,609 - Epoch: [45][ 110/ 1236] Overall Loss 0.308843 Objective Loss 0.308843 LR 0.001000 Time 0.027501 +2023-10-05 21:03:13,813 - Epoch: [45][ 120/ 1236] Overall Loss 0.310756 Objective Loss 0.310756 LR 0.001000 Time 0.026907 +2023-10-05 21:03:14,013 - Epoch: [45][ 130/ 1236] Overall Loss 0.309564 Objective Loss 0.309564 LR 0.001000 Time 0.026378 +2023-10-05 21:03:14,218 - Epoch: [45][ 140/ 1236] Overall Loss 0.312882 Objective Loss 0.312882 LR 0.001000 Time 0.025950 +2023-10-05 21:03:14,418 - Epoch: [45][ 150/ 1236] Overall Loss 0.313732 Objective Loss 0.313732 LR 0.001000 Time 0.025556 +2023-10-05 21:03:14,623 - Epoch: [45][ 160/ 1236] Overall Loss 0.315843 Objective Loss 0.315843 LR 0.001000 Time 0.025234 +2023-10-05 21:03:14,822 - Epoch: [45][ 170/ 1236] Overall Loss 0.313898 Objective Loss 0.313898 LR 0.001000 Time 0.024922 +2023-10-05 21:03:15,024 - Epoch: [45][ 180/ 1236] Overall Loss 0.313872 Objective Loss 0.313872 LR 0.001000 Time 0.024654 +2023-10-05 21:03:15,223 - Epoch: [45][ 190/ 1236] Overall Loss 0.314029 Objective Loss 0.314029 LR 0.001000 Time 0.024404 +2023-10-05 21:03:15,424 - Epoch: [45][ 200/ 1236] Overall Loss 0.315342 Objective Loss 0.315342 LR 0.001000 Time 0.024188 +2023-10-05 21:03:15,624 - Epoch: [45][ 210/ 1236] Overall Loss 0.314627 Objective Loss 0.314627 LR 0.001000 Time 0.023987 +2023-10-05 21:03:15,826 - Epoch: [45][ 220/ 1236] Overall Loss 0.314914 Objective Loss 0.314914 LR 0.001000 Time 0.023811 +2023-10-05 21:03:16,027 - Epoch: [45][ 230/ 1236] Overall Loss 0.315642 Objective Loss 0.315642 LR 0.001000 Time 0.023649 +2023-10-05 21:03:16,229 - Epoch: [45][ 240/ 1236] Overall Loss 0.316140 Objective Loss 0.316140 LR 0.001000 Time 0.023504 +2023-10-05 21:03:16,432 - Epoch: [45][ 250/ 1236] Overall Loss 0.318402 Objective Loss 0.318402 LR 0.001000 Time 0.023375 +2023-10-05 21:03:16,634 - Epoch: [45][ 260/ 1236] Overall Loss 0.319566 Objective Loss 0.319566 LR 0.001000 Time 0.023252 +2023-10-05 21:03:16,837 - Epoch: [45][ 270/ 1236] Overall Loss 0.319861 Objective Loss 0.319861 LR 0.001000 Time 0.023139 +2023-10-05 21:03:17,039 - Epoch: [45][ 280/ 1236] Overall Loss 0.319248 Objective Loss 0.319248 LR 0.001000 Time 0.023033 +2023-10-05 21:03:17,241 - Epoch: [45][ 290/ 1236] Overall Loss 0.318781 Objective Loss 0.318781 LR 0.001000 Time 0.022936 +2023-10-05 21:03:17,443 - Epoch: [45][ 300/ 1236] Overall Loss 0.318262 Objective Loss 0.318262 LR 0.001000 Time 0.022843 +2023-10-05 21:03:17,646 - Epoch: [45][ 310/ 1236] Overall Loss 0.318087 Objective Loss 0.318087 LR 0.001000 Time 0.022759 +2023-10-05 21:03:17,848 - Epoch: [45][ 320/ 1236] Overall Loss 0.318289 Objective Loss 0.318289 LR 0.001000 Time 0.022677 +2023-10-05 21:03:18,050 - Epoch: [45][ 330/ 1236] Overall Loss 0.317522 Objective Loss 0.317522 LR 0.001000 Time 0.022604 +2023-10-05 21:03:18,252 - Epoch: [45][ 340/ 1236] Overall Loss 0.317593 Objective Loss 0.317593 LR 0.001000 Time 0.022532 +2023-10-05 21:03:18,455 - Epoch: [45][ 350/ 1236] Overall Loss 0.317661 Objective Loss 0.317661 LR 0.001000 Time 0.022467 +2023-10-05 21:03:18,657 - Epoch: [45][ 360/ 1236] Overall Loss 0.318028 Objective Loss 0.318028 LR 0.001000 Time 0.022403 +2023-10-05 21:03:18,861 - Epoch: [45][ 370/ 1236] Overall Loss 0.318227 Objective Loss 0.318227 LR 0.001000 Time 0.022347 +2023-10-05 21:03:19,063 - Epoch: [45][ 380/ 1236] Overall Loss 0.317769 Objective Loss 0.317769 LR 0.001000 Time 0.022290 +2023-10-05 21:03:19,267 - Epoch: [45][ 390/ 1236] Overall Loss 0.317135 Objective Loss 0.317135 LR 0.001000 Time 0.022240 +2023-10-05 21:03:19,469 - Epoch: [45][ 400/ 1236] Overall Loss 0.316905 Objective Loss 0.316905 LR 0.001000 Time 0.022189 +2023-10-05 21:03:19,672 - Epoch: [45][ 410/ 1236] Overall Loss 0.316284 Objective Loss 0.316284 LR 0.001000 Time 0.022143 +2023-10-05 21:03:19,875 - Epoch: [45][ 420/ 1236] Overall Loss 0.315838 Objective Loss 0.315838 LR 0.001000 Time 0.022097 +2023-10-05 21:03:20,078 - Epoch: [45][ 430/ 1236] Overall Loss 0.316301 Objective Loss 0.316301 LR 0.001000 Time 0.022055 +2023-10-05 21:03:20,280 - Epoch: [45][ 440/ 1236] Overall Loss 0.316301 Objective Loss 0.316301 LR 0.001000 Time 0.022012 +2023-10-05 21:03:20,484 - Epoch: [45][ 450/ 1236] Overall Loss 0.316108 Objective Loss 0.316108 LR 0.001000 Time 0.021974 +2023-10-05 21:03:20,686 - Epoch: [45][ 460/ 1236] Overall Loss 0.316341 Objective Loss 0.316341 LR 0.001000 Time 0.021935 +2023-10-05 21:03:20,889 - Epoch: [45][ 470/ 1236] Overall Loss 0.316646 Objective Loss 0.316646 LR 0.001000 Time 0.021901 +2023-10-05 21:03:21,092 - Epoch: [45][ 480/ 1236] Overall Loss 0.316326 Objective Loss 0.316326 LR 0.001000 Time 0.021866 +2023-10-05 21:03:21,295 - Epoch: [45][ 490/ 1236] Overall Loss 0.316242 Objective Loss 0.316242 LR 0.001000 Time 0.021834 +2023-10-05 21:03:21,498 - Epoch: [45][ 500/ 1236] Overall Loss 0.316429 Objective Loss 0.316429 LR 0.001000 Time 0.021803 +2023-10-05 21:03:21,703 - Epoch: [45][ 510/ 1236] Overall Loss 0.317057 Objective Loss 0.317057 LR 0.001000 Time 0.021775 +2023-10-05 21:03:21,903 - Epoch: [45][ 520/ 1236] Overall Loss 0.317246 Objective Loss 0.317246 LR 0.001000 Time 0.021741 +2023-10-05 21:03:22,106 - Epoch: [45][ 530/ 1236] Overall Loss 0.317188 Objective Loss 0.317188 LR 0.001000 Time 0.021713 +2023-10-05 21:03:22,305 - Epoch: [45][ 540/ 1236] Overall Loss 0.317320 Objective Loss 0.317320 LR 0.001000 Time 0.021680 +2023-10-05 21:03:22,509 - Epoch: [45][ 550/ 1236] Overall Loss 0.317905 Objective Loss 0.317905 LR 0.001000 Time 0.021655 +2023-10-05 21:03:22,709 - Epoch: [45][ 560/ 1236] Overall Loss 0.318161 Objective Loss 0.318161 LR 0.001000 Time 0.021625 +2023-10-05 21:03:22,912 - Epoch: [45][ 570/ 1236] Overall Loss 0.318464 Objective Loss 0.318464 LR 0.001000 Time 0.021601 +2023-10-05 21:03:23,112 - Epoch: [45][ 580/ 1236] Overall Loss 0.318412 Objective Loss 0.318412 LR 0.001000 Time 0.021573 +2023-10-05 21:03:23,315 - Epoch: [45][ 590/ 1236] Overall Loss 0.318308 Objective Loss 0.318308 LR 0.001000 Time 0.021550 +2023-10-05 21:03:23,515 - Epoch: [45][ 600/ 1236] Overall Loss 0.318165 Objective Loss 0.318165 LR 0.001000 Time 0.021525 +2023-10-05 21:03:23,718 - Epoch: [45][ 610/ 1236] Overall Loss 0.317994 Objective Loss 0.317994 LR 0.001000 Time 0.021504 +2023-10-05 21:03:23,919 - Epoch: [45][ 620/ 1236] Overall Loss 0.317736 Objective Loss 0.317736 LR 0.001000 Time 0.021480 +2023-10-05 21:03:24,121 - Epoch: [45][ 630/ 1236] Overall Loss 0.317386 Objective Loss 0.317386 LR 0.001000 Time 0.021460 +2023-10-05 21:03:24,322 - Epoch: [45][ 640/ 1236] Overall Loss 0.317434 Objective Loss 0.317434 LR 0.001000 Time 0.021437 +2023-10-05 21:03:24,524 - Epoch: [45][ 650/ 1236] Overall Loss 0.316913 Objective Loss 0.316913 LR 0.001000 Time 0.021419 +2023-10-05 21:03:24,725 - Epoch: [45][ 660/ 1236] Overall Loss 0.316486 Objective Loss 0.316486 LR 0.001000 Time 0.021398 +2023-10-05 21:03:24,928 - Epoch: [45][ 670/ 1236] Overall Loss 0.316933 Objective Loss 0.316933 LR 0.001000 Time 0.021381 +2023-10-05 21:03:25,129 - Epoch: [45][ 680/ 1236] Overall Loss 0.317101 Objective Loss 0.317101 LR 0.001000 Time 0.021361 +2023-10-05 21:03:25,332 - Epoch: [45][ 690/ 1236] Overall Loss 0.317458 Objective Loss 0.317458 LR 0.001000 Time 0.021346 +2023-10-05 21:03:25,532 - Epoch: [45][ 700/ 1236] Overall Loss 0.317986 Objective Loss 0.317986 LR 0.001000 Time 0.021327 +2023-10-05 21:03:25,735 - Epoch: [45][ 710/ 1236] Overall Loss 0.317804 Objective Loss 0.317804 LR 0.001000 Time 0.021311 +2023-10-05 21:03:25,935 - Epoch: [45][ 720/ 1236] Overall Loss 0.318564 Objective Loss 0.318564 LR 0.001000 Time 0.021293 +2023-10-05 21:03:26,138 - Epoch: [45][ 730/ 1236] Overall Loss 0.318538 Objective Loss 0.318538 LR 0.001000 Time 0.021279 +2023-10-05 21:03:26,338 - Epoch: [45][ 740/ 1236] Overall Loss 0.318839 Objective Loss 0.318839 LR 0.001000 Time 0.021261 +2023-10-05 21:03:26,541 - Epoch: [45][ 750/ 1236] Overall Loss 0.319013 Objective Loss 0.319013 LR 0.001000 Time 0.021248 +2023-10-05 21:03:26,741 - Epoch: [45][ 760/ 1236] Overall Loss 0.319905 Objective Loss 0.319905 LR 0.001000 Time 0.021231 +2023-10-05 21:03:26,944 - Epoch: [45][ 770/ 1236] Overall Loss 0.319940 Objective Loss 0.319940 LR 0.001000 Time 0.021218 +2023-10-05 21:03:27,144 - Epoch: [45][ 780/ 1236] Overall Loss 0.320169 Objective Loss 0.320169 LR 0.001000 Time 0.021202 +2023-10-05 21:03:27,347 - Epoch: [45][ 790/ 1236] Overall Loss 0.319976 Objective Loss 0.319976 LR 0.001000 Time 0.021190 +2023-10-05 21:03:27,547 - Epoch: [45][ 800/ 1236] Overall Loss 0.320006 Objective Loss 0.320006 LR 0.001000 Time 0.021175 +2023-10-05 21:03:27,750 - Epoch: [45][ 810/ 1236] Overall Loss 0.319863 Objective Loss 0.319863 LR 0.001000 Time 0.021164 +2023-10-05 21:03:27,950 - Epoch: [45][ 820/ 1236] Overall Loss 0.320152 Objective Loss 0.320152 LR 0.001000 Time 0.021149 +2023-10-05 21:03:28,153 - Epoch: [45][ 830/ 1236] Overall Loss 0.320188 Objective Loss 0.320188 LR 0.001000 Time 0.021139 +2023-10-05 21:03:28,353 - Epoch: [45][ 840/ 1236] Overall Loss 0.320378 Objective Loss 0.320378 LR 0.001000 Time 0.021124 +2023-10-05 21:03:28,556 - Epoch: [45][ 850/ 1236] Overall Loss 0.320595 Objective Loss 0.320595 LR 0.001000 Time 0.021114 +2023-10-05 21:03:28,756 - Epoch: [45][ 860/ 1236] Overall Loss 0.320516 Objective Loss 0.320516 LR 0.001000 Time 0.021101 +2023-10-05 21:03:28,959 - Epoch: [45][ 870/ 1236] Overall Loss 0.320717 Objective Loss 0.320717 LR 0.001000 Time 0.021092 +2023-10-05 21:03:29,159 - Epoch: [45][ 880/ 1236] Overall Loss 0.320964 Objective Loss 0.320964 LR 0.001000 Time 0.021079 +2023-10-05 21:03:29,362 - Epoch: [45][ 890/ 1236] Overall Loss 0.320701 Objective Loss 0.320701 LR 0.001000 Time 0.021070 +2023-10-05 21:03:29,562 - Epoch: [45][ 900/ 1236] Overall Loss 0.320877 Objective Loss 0.320877 LR 0.001000 Time 0.021058 +2023-10-05 21:03:29,766 - Epoch: [45][ 910/ 1236] Overall Loss 0.320667 Objective Loss 0.320667 LR 0.001000 Time 0.021049 +2023-10-05 21:03:29,965 - Epoch: [45][ 920/ 1236] Overall Loss 0.320860 Objective Loss 0.320860 LR 0.001000 Time 0.021037 +2023-10-05 21:03:30,169 - Epoch: [45][ 930/ 1236] Overall Loss 0.321023 Objective Loss 0.321023 LR 0.001000 Time 0.021029 +2023-10-05 21:03:30,369 - Epoch: [45][ 940/ 1236] Overall Loss 0.321319 Objective Loss 0.321319 LR 0.001000 Time 0.021018 +2023-10-05 21:03:30,572 - Epoch: [45][ 950/ 1236] Overall Loss 0.321384 Objective Loss 0.321384 LR 0.001000 Time 0.021010 +2023-10-05 21:03:30,772 - Epoch: [45][ 960/ 1236] Overall Loss 0.321289 Objective Loss 0.321289 LR 0.001000 Time 0.020999 +2023-10-05 21:03:30,976 - Epoch: [45][ 970/ 1236] Overall Loss 0.321004 Objective Loss 0.321004 LR 0.001000 Time 0.020993 +2023-10-05 21:03:31,177 - Epoch: [45][ 980/ 1236] Overall Loss 0.321091 Objective Loss 0.321091 LR 0.001000 Time 0.020983 +2023-10-05 21:03:31,380 - Epoch: [45][ 990/ 1236] Overall Loss 0.321353 Objective Loss 0.321353 LR 0.001000 Time 0.020976 +2023-10-05 21:03:31,581 - Epoch: [45][ 1000/ 1236] Overall Loss 0.321240 Objective Loss 0.321240 LR 0.001000 Time 0.020967 +2023-10-05 21:03:31,784 - Epoch: [45][ 1010/ 1236] Overall Loss 0.320961 Objective Loss 0.320961 LR 0.001000 Time 0.020960 +2023-10-05 21:03:31,985 - Epoch: [45][ 1020/ 1236] Overall Loss 0.320835 Objective Loss 0.320835 LR 0.001000 Time 0.020951 +2023-10-05 21:03:32,188 - Epoch: [45][ 1030/ 1236] Overall Loss 0.320969 Objective Loss 0.320969 LR 0.001000 Time 0.020945 +2023-10-05 21:03:32,389 - Epoch: [45][ 1040/ 1236] Overall Loss 0.321089 Objective Loss 0.321089 LR 0.001000 Time 0.020936 +2023-10-05 21:03:32,592 - Epoch: [45][ 1050/ 1236] Overall Loss 0.321340 Objective Loss 0.321340 LR 0.001000 Time 0.020930 +2023-10-05 21:03:32,793 - Epoch: [45][ 1060/ 1236] Overall Loss 0.321403 Objective Loss 0.321403 LR 0.001000 Time 0.020922 +2023-10-05 21:03:32,997 - Epoch: [45][ 1070/ 1236] Overall Loss 0.321644 Objective Loss 0.321644 LR 0.001000 Time 0.020917 +2023-10-05 21:03:33,198 - Epoch: [45][ 1080/ 1236] Overall Loss 0.322146 Objective Loss 0.322146 LR 0.001000 Time 0.020909 +2023-10-05 21:03:33,402 - Epoch: [45][ 1090/ 1236] Overall Loss 0.322385 Objective Loss 0.322385 LR 0.001000 Time 0.020903 +2023-10-05 21:03:33,602 - Epoch: [45][ 1100/ 1236] Overall Loss 0.322209 Objective Loss 0.322209 LR 0.001000 Time 0.020895 +2023-10-05 21:03:33,806 - Epoch: [45][ 1110/ 1236] Overall Loss 0.322457 Objective Loss 0.322457 LR 0.001000 Time 0.020890 +2023-10-05 21:03:34,007 - Epoch: [45][ 1120/ 1236] Overall Loss 0.322396 Objective Loss 0.322396 LR 0.001000 Time 0.020882 +2023-10-05 21:03:34,210 - Epoch: [45][ 1130/ 1236] Overall Loss 0.322465 Objective Loss 0.322465 LR 0.001000 Time 0.020877 +2023-10-05 21:03:34,411 - Epoch: [45][ 1140/ 1236] Overall Loss 0.322495 Objective Loss 0.322495 LR 0.001000 Time 0.020870 +2023-10-05 21:03:34,614 - Epoch: [45][ 1150/ 1236] Overall Loss 0.322174 Objective Loss 0.322174 LR 0.001000 Time 0.020865 +2023-10-05 21:03:34,815 - Epoch: [45][ 1160/ 1236] Overall Loss 0.322357 Objective Loss 0.322357 LR 0.001000 Time 0.020858 +2023-10-05 21:03:35,018 - Epoch: [45][ 1170/ 1236] Overall Loss 0.322232 Objective Loss 0.322232 LR 0.001000 Time 0.020853 +2023-10-05 21:03:35,219 - Epoch: [45][ 1180/ 1236] Overall Loss 0.322418 Objective Loss 0.322418 LR 0.001000 Time 0.020846 +2023-10-05 21:03:35,422 - Epoch: [45][ 1190/ 1236] Overall Loss 0.322799 Objective Loss 0.322799 LR 0.001000 Time 0.020842 +2023-10-05 21:03:35,623 - Epoch: [45][ 1200/ 1236] Overall Loss 0.322869 Objective Loss 0.322869 LR 0.001000 Time 0.020835 +2023-10-05 21:03:35,827 - Epoch: [45][ 1210/ 1236] Overall Loss 0.322922 Objective Loss 0.322922 LR 0.001000 Time 0.020831 +2023-10-05 21:03:36,028 - Epoch: [45][ 1220/ 1236] Overall Loss 0.322976 Objective Loss 0.322976 LR 0.001000 Time 0.020824 +2023-10-05 21:03:36,284 - Epoch: [45][ 1230/ 1236] Overall Loss 0.323253 Objective Loss 0.323253 LR 0.001000 Time 0.020864 +2023-10-05 21:03:36,403 - Epoch: [45][ 1236/ 1236] Overall Loss 0.323437 Objective Loss 0.323437 Top1 82.688391 Top5 97.148676 LR 0.001000 Time 0.020858 +2023-10-05 21:03:36,533 - --- validate (epoch=45)----------- +2023-10-05 21:03:36,534 - 29943 samples (256 per mini-batch) +2023-10-05 21:03:36,986 - Epoch: [45][ 10/ 117] Loss 0.387668 Top1 80.546875 Top5 97.148438 +2023-10-05 21:03:37,128 - Epoch: [45][ 20/ 117] Loss 0.361273 Top1 81.230469 Top5 97.246094 +2023-10-05 21:03:37,270 - Epoch: [45][ 30/ 117] Loss 0.374558 Top1 80.963542 Top5 97.291667 +2023-10-05 21:03:37,411 - Epoch: [45][ 40/ 117] Loss 0.368373 Top1 81.064453 Top5 97.343750 +2023-10-05 21:03:37,553 - Epoch: [45][ 50/ 117] Loss 0.374103 Top1 81.046875 Top5 97.273438 +2023-10-05 21:03:37,701 - Epoch: [45][ 60/ 117] Loss 0.369446 Top1 80.911458 Top5 97.180990 +2023-10-05 21:03:37,843 - Epoch: [45][ 70/ 117] Loss 0.368379 Top1 80.876116 Top5 97.304688 +2023-10-05 21:03:37,985 - Epoch: [45][ 80/ 117] Loss 0.367119 Top1 80.952148 Top5 97.275391 +2023-10-05 21:03:38,127 - Epoch: [45][ 90/ 117] Loss 0.362256 Top1 81.032986 Top5 97.278646 +2023-10-05 21:03:38,270 - Epoch: [45][ 100/ 117] Loss 0.364960 Top1 81.070312 Top5 97.257812 +2023-10-05 21:03:38,420 - Epoch: [45][ 110/ 117] Loss 0.364576 Top1 81.154119 Top5 97.244318 +2023-10-05 21:03:38,506 - Epoch: [45][ 117/ 117] Loss 0.365999 Top1 81.177571 Top5 97.264803 +2023-10-05 21:03:38,641 - ==> Top1: 81.178 Top5: 97.265 Loss: 0.366 + +2023-10-05 21:03:38,642 - ==> Confusion: +[[ 911 2 2 2 7 3 0 0 9 80 1 0 2 7 6 3 3 2 0 1 9] + [ 0 1016 0 0 5 39 0 32 3 0 5 1 0 1 2 4 4 0 11 3 5] + [ 8 0 920 26 2 0 29 10 0 0 7 3 6 8 2 6 2 1 9 6 11] + [ 1 1 16 946 2 3 0 2 4 2 9 0 5 9 39 4 1 6 27 2 10] + [ 33 10 1 1 946 4 0 1 1 11 0 5 1 2 12 5 8 2 1 1 5] + [ 6 29 0 4 2 972 1 29 3 2 4 5 1 24 6 2 3 2 5 4 12] + [ 0 4 34 0 0 2 1105 8 0 0 5 3 2 0 1 7 1 1 3 9 6] + [ 8 22 12 0 0 35 2 1032 0 2 10 13 2 3 1 1 2 1 53 10 9] + [ 15 3 0 0 0 1 0 1 957 51 8 3 2 16 17 3 1 0 9 1 1] + [ 95 1 0 0 3 1 0 0 27 942 0 2 1 21 7 7 2 0 1 2 7] + [ 1 2 14 11 0 1 1 4 24 2 941 4 0 22 4 1 1 4 7 2 7] + [ 1 0 1 0 0 17 0 0 0 0 0 950 21 9 0 4 2 12 1 14 3] + [ 2 1 0 8 1 4 0 4 1 1 1 42 951 6 3 10 1 18 2 2 10] + [ 2 0 0 0 4 7 0 2 12 16 5 6 3 1048 3 1 0 1 0 0 9] + [ 18 2 3 13 2 0 0 0 27 6 4 0 4 5 986 0 2 5 14 1 9] + [ 1 3 0 2 2 1 2 0 0 0 0 8 12 1 1 1059 15 15 0 6 6] + [ 1 19 1 0 10 11 0 0 1 0 0 10 1 2 1 14 1074 1 0 6 9] + [ 0 0 0 3 0 0 1 0 0 2 0 12 21 2 4 11 0 979 3 0 0] + [ 1 5 5 14 1 2 0 36 7 0 4 3 6 1 18 0 1 0 958 1 5] + [ 0 2 1 1 1 7 3 13 0 0 2 19 2 7 2 7 7 3 1 1065 9] + [ 175 189 146 89 63 178 45 137 147 109 207 151 355 364 168 82 190 93 220 248 4549]] + +2023-10-05 21:03:38,643 - ==> Best [Top1: 82.116 Top5: 97.562 Sparsity:0.00 Params: 148928 on epoch: 38] +2023-10-05 21:03:38,643 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:03:38,649 - + +2023-10-05 21:03:38,649 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:03:39,655 - Epoch: [46][ 10/ 1236] Overall Loss 0.315416 Objective Loss 0.315416 LR 0.001000 Time 0.100548 +2023-10-05 21:03:39,856 - Epoch: [46][ 20/ 1236] Overall Loss 0.320255 Objective Loss 0.320255 LR 0.001000 Time 0.060287 +2023-10-05 21:03:40,055 - Epoch: [46][ 30/ 1236] Overall Loss 0.317671 Objective Loss 0.317671 LR 0.001000 Time 0.046834 +2023-10-05 21:03:40,257 - Epoch: [46][ 40/ 1236] Overall Loss 0.312709 Objective Loss 0.312709 LR 0.001000 Time 0.040160 +2023-10-05 21:03:40,459 - Epoch: [46][ 50/ 1236] Overall Loss 0.314274 Objective Loss 0.314274 LR 0.001000 Time 0.036154 +2023-10-05 21:03:40,662 - Epoch: [46][ 60/ 1236] Overall Loss 0.310616 Objective Loss 0.310616 LR 0.001000 Time 0.033507 +2023-10-05 21:03:40,862 - Epoch: [46][ 70/ 1236] Overall Loss 0.303900 Objective Loss 0.303900 LR 0.001000 Time 0.031576 +2023-10-05 21:03:41,065 - Epoch: [46][ 80/ 1236] Overall Loss 0.299790 Objective Loss 0.299790 LR 0.001000 Time 0.030165 +2023-10-05 21:03:41,266 - Epoch: [46][ 90/ 1236] Overall Loss 0.302259 Objective Loss 0.302259 LR 0.001000 Time 0.029035 +2023-10-05 21:03:41,468 - Epoch: [46][ 100/ 1236] Overall Loss 0.301875 Objective Loss 0.301875 LR 0.001000 Time 0.028154 +2023-10-05 21:03:41,669 - Epoch: [46][ 110/ 1236] Overall Loss 0.302798 Objective Loss 0.302798 LR 0.001000 Time 0.027413 +2023-10-05 21:03:41,871 - Epoch: [46][ 120/ 1236] Overall Loss 0.304499 Objective Loss 0.304499 LR 0.001000 Time 0.026814 +2023-10-05 21:03:42,072 - Epoch: [46][ 130/ 1236] Overall Loss 0.304197 Objective Loss 0.304197 LR 0.001000 Time 0.026291 +2023-10-05 21:03:42,275 - Epoch: [46][ 140/ 1236] Overall Loss 0.305747 Objective Loss 0.305747 LR 0.001000 Time 0.025859 +2023-10-05 21:03:42,475 - Epoch: [46][ 150/ 1236] Overall Loss 0.306867 Objective Loss 0.306867 LR 0.001000 Time 0.025468 +2023-10-05 21:03:42,678 - Epoch: [46][ 160/ 1236] Overall Loss 0.305641 Objective Loss 0.305641 LR 0.001000 Time 0.025144 +2023-10-05 21:03:42,879 - Epoch: [46][ 170/ 1236] Overall Loss 0.306704 Objective Loss 0.306704 LR 0.001000 Time 0.024843 +2023-10-05 21:03:43,081 - Epoch: [46][ 180/ 1236] Overall Loss 0.307302 Objective Loss 0.307302 LR 0.001000 Time 0.024586 +2023-10-05 21:03:43,282 - Epoch: [46][ 190/ 1236] Overall Loss 0.308972 Objective Loss 0.308972 LR 0.001000 Time 0.024346 +2023-10-05 21:03:43,485 - Epoch: [46][ 200/ 1236] Overall Loss 0.309070 Objective Loss 0.309070 LR 0.001000 Time 0.024141 +2023-10-05 21:03:43,686 - Epoch: [46][ 210/ 1236] Overall Loss 0.310904 Objective Loss 0.310904 LR 0.001000 Time 0.023946 +2023-10-05 21:03:43,888 - Epoch: [46][ 220/ 1236] Overall Loss 0.310632 Objective Loss 0.310632 LR 0.001000 Time 0.023775 +2023-10-05 21:03:44,089 - Epoch: [46][ 230/ 1236] Overall Loss 0.309971 Objective Loss 0.309971 LR 0.001000 Time 0.023612 +2023-10-05 21:03:44,291 - Epoch: [46][ 240/ 1236] Overall Loss 0.310791 Objective Loss 0.310791 LR 0.001000 Time 0.023472 +2023-10-05 21:03:44,492 - Epoch: [46][ 250/ 1236] Overall Loss 0.311896 Objective Loss 0.311896 LR 0.001000 Time 0.023334 +2023-10-05 21:03:44,695 - Epoch: [46][ 260/ 1236] Overall Loss 0.310948 Objective Loss 0.310948 LR 0.001000 Time 0.023215 +2023-10-05 21:03:44,896 - Epoch: [46][ 270/ 1236] Overall Loss 0.310231 Objective Loss 0.310231 LR 0.001000 Time 0.023098 +2023-10-05 21:03:45,098 - Epoch: [46][ 280/ 1236] Overall Loss 0.309502 Objective Loss 0.309502 LR 0.001000 Time 0.022995 +2023-10-05 21:03:45,299 - Epoch: [46][ 290/ 1236] Overall Loss 0.309722 Objective Loss 0.309722 LR 0.001000 Time 0.022892 +2023-10-05 21:03:45,502 - Epoch: [46][ 300/ 1236] Overall Loss 0.309904 Objective Loss 0.309904 LR 0.001000 Time 0.022805 +2023-10-05 21:03:45,703 - Epoch: [46][ 310/ 1236] Overall Loss 0.309963 Objective Loss 0.309963 LR 0.001000 Time 0.022718 +2023-10-05 21:03:45,906 - Epoch: [46][ 320/ 1236] Overall Loss 0.310094 Objective Loss 0.310094 LR 0.001000 Time 0.022640 +2023-10-05 21:03:46,107 - Epoch: [46][ 330/ 1236] Overall Loss 0.311419 Objective Loss 0.311419 LR 0.001000 Time 0.022562 +2023-10-05 21:03:46,310 - Epoch: [46][ 340/ 1236] Overall Loss 0.311414 Objective Loss 0.311414 LR 0.001000 Time 0.022493 +2023-10-05 21:03:46,510 - Epoch: [46][ 350/ 1236] Overall Loss 0.311917 Objective Loss 0.311917 LR 0.001000 Time 0.022422 +2023-10-05 21:03:46,713 - Epoch: [46][ 360/ 1236] Overall Loss 0.311805 Objective Loss 0.311805 LR 0.001000 Time 0.022362 +2023-10-05 21:03:46,916 - Epoch: [46][ 370/ 1236] Overall Loss 0.311475 Objective Loss 0.311475 LR 0.001000 Time 0.022305 +2023-10-05 21:03:47,121 - Epoch: [46][ 380/ 1236] Overall Loss 0.311006 Objective Loss 0.311006 LR 0.001000 Time 0.022256 +2023-10-05 21:03:47,321 - Epoch: [46][ 390/ 1236] Overall Loss 0.311665 Objective Loss 0.311665 LR 0.001000 Time 0.022199 +2023-10-05 21:03:47,523 - Epoch: [46][ 400/ 1236] Overall Loss 0.311695 Objective Loss 0.311695 LR 0.001000 Time 0.022148 +2023-10-05 21:03:47,724 - Epoch: [46][ 410/ 1236] Overall Loss 0.311859 Objective Loss 0.311859 LR 0.001000 Time 0.022097 +2023-10-05 21:03:47,926 - Epoch: [46][ 420/ 1236] Overall Loss 0.311808 Objective Loss 0.311808 LR 0.001000 Time 0.022052 +2023-10-05 21:03:48,127 - Epoch: [46][ 430/ 1236] Overall Loss 0.312765 Objective Loss 0.312765 LR 0.001000 Time 0.022004 +2023-10-05 21:03:48,329 - Epoch: [46][ 440/ 1236] Overall Loss 0.312044 Objective Loss 0.312044 LR 0.001000 Time 0.021963 +2023-10-05 21:03:48,529 - Epoch: [46][ 450/ 1236] Overall Loss 0.312504 Objective Loss 0.312504 LR 0.001000 Time 0.021919 +2023-10-05 21:03:48,732 - Epoch: [46][ 460/ 1236] Overall Loss 0.313013 Objective Loss 0.313013 LR 0.001000 Time 0.021882 +2023-10-05 21:03:48,932 - Epoch: [46][ 470/ 1236] Overall Loss 0.313194 Objective Loss 0.313194 LR 0.001000 Time 0.021842 +2023-10-05 21:03:49,135 - Epoch: [46][ 480/ 1236] Overall Loss 0.312902 Objective Loss 0.312902 LR 0.001000 Time 0.021808 +2023-10-05 21:03:49,335 - Epoch: [46][ 490/ 1236] Overall Loss 0.313214 Objective Loss 0.313214 LR 0.001000 Time 0.021771 +2023-10-05 21:03:49,537 - Epoch: [46][ 500/ 1236] Overall Loss 0.312904 Objective Loss 0.312904 LR 0.001000 Time 0.021740 +2023-10-05 21:03:49,738 - Epoch: [46][ 510/ 1236] Overall Loss 0.312793 Objective Loss 0.312793 LR 0.001000 Time 0.021706 +2023-10-05 21:03:49,940 - Epoch: [46][ 520/ 1236] Overall Loss 0.313080 Objective Loss 0.313080 LR 0.001000 Time 0.021677 +2023-10-05 21:03:50,141 - Epoch: [46][ 530/ 1236] Overall Loss 0.314315 Objective Loss 0.314315 LR 0.001000 Time 0.021646 +2023-10-05 21:03:50,343 - Epoch: [46][ 540/ 1236] Overall Loss 0.315440 Objective Loss 0.315440 LR 0.001000 Time 0.021620 +2023-10-05 21:03:50,544 - Epoch: [46][ 550/ 1236] Overall Loss 0.316176 Objective Loss 0.316176 LR 0.001000 Time 0.021590 +2023-10-05 21:03:50,746 - Epoch: [46][ 560/ 1236] Overall Loss 0.316593 Objective Loss 0.316593 LR 0.001000 Time 0.021566 +2023-10-05 21:03:50,947 - Epoch: [46][ 570/ 1236] Overall Loss 0.316095 Objective Loss 0.316095 LR 0.001000 Time 0.021539 +2023-10-05 21:03:51,149 - Epoch: [46][ 580/ 1236] Overall Loss 0.315796 Objective Loss 0.315796 LR 0.001000 Time 0.021516 +2023-10-05 21:03:51,350 - Epoch: [46][ 590/ 1236] Overall Loss 0.316090 Objective Loss 0.316090 LR 0.001000 Time 0.021491 +2023-10-05 21:03:51,552 - Epoch: [46][ 600/ 1236] Overall Loss 0.316285 Objective Loss 0.316285 LR 0.001000 Time 0.021469 +2023-10-05 21:03:51,752 - Epoch: [46][ 610/ 1236] Overall Loss 0.315940 Objective Loss 0.315940 LR 0.001000 Time 0.021445 +2023-10-05 21:03:51,955 - Epoch: [46][ 620/ 1236] Overall Loss 0.316042 Objective Loss 0.316042 LR 0.001000 Time 0.021425 +2023-10-05 21:03:52,155 - Epoch: [46][ 630/ 1236] Overall Loss 0.315590 Objective Loss 0.315590 LR 0.001000 Time 0.021403 +2023-10-05 21:03:52,358 - Epoch: [46][ 640/ 1236] Overall Loss 0.314865 Objective Loss 0.314865 LR 0.001000 Time 0.021384 +2023-10-05 21:03:52,558 - Epoch: [46][ 650/ 1236] Overall Loss 0.314782 Objective Loss 0.314782 LR 0.001000 Time 0.021363 +2023-10-05 21:03:52,764 - Epoch: [46][ 660/ 1236] Overall Loss 0.314655 Objective Loss 0.314655 LR 0.001000 Time 0.021344 +2023-10-05 21:03:52,964 - Epoch: [46][ 670/ 1236] Overall Loss 0.314674 Objective Loss 0.314674 LR 0.001000 Time 0.021324 +2023-10-05 21:03:53,167 - Epoch: [46][ 680/ 1236] Overall Loss 0.314355 Objective Loss 0.314355 LR 0.001000 Time 0.021308 +2023-10-05 21:03:53,368 - Epoch: [46][ 690/ 1236] Overall Loss 0.314234 Objective Loss 0.314234 LR 0.001000 Time 0.021289 +2023-10-05 21:03:53,570 - Epoch: [46][ 700/ 1236] Overall Loss 0.313743 Objective Loss 0.313743 LR 0.001000 Time 0.021274 +2023-10-05 21:03:53,771 - Epoch: [46][ 710/ 1236] Overall Loss 0.313961 Objective Loss 0.313961 LR 0.001000 Time 0.021256 +2023-10-05 21:03:53,973 - Epoch: [46][ 720/ 1236] Overall Loss 0.314407 Objective Loss 0.314407 LR 0.001000 Time 0.021242 +2023-10-05 21:03:54,174 - Epoch: [46][ 730/ 1236] Overall Loss 0.314152 Objective Loss 0.314152 LR 0.001000 Time 0.021225 +2023-10-05 21:03:54,376 - Epoch: [46][ 740/ 1236] Overall Loss 0.314607 Objective Loss 0.314607 LR 0.001000 Time 0.021211 +2023-10-05 21:03:54,577 - Epoch: [46][ 750/ 1236] Overall Loss 0.314419 Objective Loss 0.314419 LR 0.001000 Time 0.021196 +2023-10-05 21:03:54,779 - Epoch: [46][ 760/ 1236] Overall Loss 0.315023 Objective Loss 0.315023 LR 0.001000 Time 0.021183 +2023-10-05 21:03:54,980 - Epoch: [46][ 770/ 1236] Overall Loss 0.315371 Objective Loss 0.315371 LR 0.001000 Time 0.021168 +2023-10-05 21:03:55,182 - Epoch: [46][ 780/ 1236] Overall Loss 0.315344 Objective Loss 0.315344 LR 0.001000 Time 0.021155 +2023-10-05 21:03:55,382 - Epoch: [46][ 790/ 1236] Overall Loss 0.315539 Objective Loss 0.315539 LR 0.001000 Time 0.021141 +2023-10-05 21:03:55,585 - Epoch: [46][ 800/ 1236] Overall Loss 0.315908 Objective Loss 0.315908 LR 0.001000 Time 0.021129 +2023-10-05 21:03:55,785 - Epoch: [46][ 810/ 1236] Overall Loss 0.315940 Objective Loss 0.315940 LR 0.001000 Time 0.021115 +2023-10-05 21:03:55,988 - Epoch: [46][ 820/ 1236] Overall Loss 0.315933 Objective Loss 0.315933 LR 0.001000 Time 0.021104 +2023-10-05 21:03:56,188 - Epoch: [46][ 830/ 1236] Overall Loss 0.316396 Objective Loss 0.316396 LR 0.001000 Time 0.021091 +2023-10-05 21:03:56,391 - Epoch: [46][ 840/ 1236] Overall Loss 0.316569 Objective Loss 0.316569 LR 0.001000 Time 0.021081 +2023-10-05 21:03:56,591 - Epoch: [46][ 850/ 1236] Overall Loss 0.316707 Objective Loss 0.316707 LR 0.001000 Time 0.021068 +2023-10-05 21:03:56,794 - Epoch: [46][ 860/ 1236] Overall Loss 0.316715 Objective Loss 0.316715 LR 0.001000 Time 0.021059 +2023-10-05 21:03:56,994 - Epoch: [46][ 870/ 1236] Overall Loss 0.317042 Objective Loss 0.317042 LR 0.001000 Time 0.021047 +2023-10-05 21:03:57,197 - Epoch: [46][ 880/ 1236] Overall Loss 0.316923 Objective Loss 0.316923 LR 0.001000 Time 0.021038 +2023-10-05 21:03:57,398 - Epoch: [46][ 890/ 1236] Overall Loss 0.316796 Objective Loss 0.316796 LR 0.001000 Time 0.021026 +2023-10-05 21:03:57,600 - Epoch: [46][ 900/ 1236] Overall Loss 0.317117 Objective Loss 0.317117 LR 0.001000 Time 0.021017 +2023-10-05 21:03:57,805 - Epoch: [46][ 910/ 1236] Overall Loss 0.317427 Objective Loss 0.317427 LR 0.001000 Time 0.021011 +2023-10-05 21:03:58,020 - Epoch: [46][ 920/ 1236] Overall Loss 0.317476 Objective Loss 0.317476 LR 0.001000 Time 0.021015 +2023-10-05 21:03:58,230 - Epoch: [46][ 930/ 1236] Overall Loss 0.317217 Objective Loss 0.317217 LR 0.001000 Time 0.021015 +2023-10-05 21:03:58,444 - Epoch: [46][ 940/ 1236] Overall Loss 0.317659 Objective Loss 0.317659 LR 0.001000 Time 0.021019 +2023-10-05 21:03:58,654 - Epoch: [46][ 950/ 1236] Overall Loss 0.317462 Objective Loss 0.317462 LR 0.001000 Time 0.021018 +2023-10-05 21:03:58,868 - Epoch: [46][ 960/ 1236] Overall Loss 0.317457 Objective Loss 0.317457 LR 0.001000 Time 0.021022 +2023-10-05 21:03:59,078 - Epoch: [46][ 970/ 1236] Overall Loss 0.317479 Objective Loss 0.317479 LR 0.001000 Time 0.021021 +2023-10-05 21:03:59,293 - Epoch: [46][ 980/ 1236] Overall Loss 0.317602 Objective Loss 0.317602 LR 0.001000 Time 0.021026 +2023-10-05 21:03:59,502 - Epoch: [46][ 990/ 1236] Overall Loss 0.317779 Objective Loss 0.317779 LR 0.001000 Time 0.021025 +2023-10-05 21:03:59,717 - Epoch: [46][ 1000/ 1236] Overall Loss 0.317913 Objective Loss 0.317913 LR 0.001000 Time 0.021029 +2023-10-05 21:03:59,927 - Epoch: [46][ 1010/ 1236] Overall Loss 0.317998 Objective Loss 0.317998 LR 0.001000 Time 0.021028 +2023-10-05 21:04:00,142 - Epoch: [46][ 1020/ 1236] Overall Loss 0.318166 Objective Loss 0.318166 LR 0.001000 Time 0.021032 +2023-10-05 21:04:00,352 - Epoch: [46][ 1030/ 1236] Overall Loss 0.318254 Objective Loss 0.318254 LR 0.001000 Time 0.021032 +2023-10-05 21:04:00,567 - Epoch: [46][ 1040/ 1236] Overall Loss 0.318025 Objective Loss 0.318025 LR 0.001000 Time 0.021036 +2023-10-05 21:04:00,777 - Epoch: [46][ 1050/ 1236] Overall Loss 0.318094 Objective Loss 0.318094 LR 0.001000 Time 0.021035 +2023-10-05 21:04:00,992 - Epoch: [46][ 1060/ 1236] Overall Loss 0.318018 Objective Loss 0.318018 LR 0.001000 Time 0.021039 +2023-10-05 21:04:01,202 - Epoch: [46][ 1070/ 1236] Overall Loss 0.317812 Objective Loss 0.317812 LR 0.001000 Time 0.021039 +2023-10-05 21:04:01,417 - Epoch: [46][ 1080/ 1236] Overall Loss 0.317677 Objective Loss 0.317677 LR 0.001000 Time 0.021042 +2023-10-05 21:04:01,627 - Epoch: [46][ 1090/ 1236] Overall Loss 0.317647 Objective Loss 0.317647 LR 0.001000 Time 0.021042 +2023-10-05 21:04:01,842 - Epoch: [46][ 1100/ 1236] Overall Loss 0.317455 Objective Loss 0.317455 LR 0.001000 Time 0.021046 +2023-10-05 21:04:02,051 - Epoch: [46][ 1110/ 1236] Overall Loss 0.317832 Objective Loss 0.317832 LR 0.001000 Time 0.021044 +2023-10-05 21:04:02,266 - Epoch: [46][ 1120/ 1236] Overall Loss 0.318202 Objective Loss 0.318202 LR 0.001000 Time 0.021048 +2023-10-05 21:04:02,475 - Epoch: [46][ 1130/ 1236] Overall Loss 0.318081 Objective Loss 0.318081 LR 0.001000 Time 0.021046 +2023-10-05 21:04:02,690 - Epoch: [46][ 1140/ 1236] Overall Loss 0.317858 Objective Loss 0.317858 LR 0.001000 Time 0.021050 +2023-10-05 21:04:02,900 - Epoch: [46][ 1150/ 1236] Overall Loss 0.318138 Objective Loss 0.318138 LR 0.001000 Time 0.021049 +2023-10-05 21:04:03,115 - Epoch: [46][ 1160/ 1236] Overall Loss 0.318249 Objective Loss 0.318249 LR 0.001000 Time 0.021053 +2023-10-05 21:04:03,325 - Epoch: [46][ 1170/ 1236] Overall Loss 0.318298 Objective Loss 0.318298 LR 0.001000 Time 0.021052 +2023-10-05 21:04:03,540 - Epoch: [46][ 1180/ 1236] Overall Loss 0.318407 Objective Loss 0.318407 LR 0.001000 Time 0.021055 +2023-10-05 21:04:03,741 - Epoch: [46][ 1190/ 1236] Overall Loss 0.318472 Objective Loss 0.318472 LR 0.001000 Time 0.021047 +2023-10-05 21:04:03,943 - Epoch: [46][ 1200/ 1236] Overall Loss 0.318381 Objective Loss 0.318381 LR 0.001000 Time 0.021040 +2023-10-05 21:04:04,144 - Epoch: [46][ 1210/ 1236] Overall Loss 0.318491 Objective Loss 0.318491 LR 0.001000 Time 0.021032 +2023-10-05 21:04:04,347 - Epoch: [46][ 1220/ 1236] Overall Loss 0.318429 Objective Loss 0.318429 LR 0.001000 Time 0.021025 +2023-10-05 21:04:04,602 - Epoch: [46][ 1230/ 1236] Overall Loss 0.318532 Objective Loss 0.318532 LR 0.001000 Time 0.021062 +2023-10-05 21:04:04,721 - Epoch: [46][ 1236/ 1236] Overall Loss 0.318246 Objective Loss 0.318246 Top1 86.965377 Top5 97.963340 LR 0.001000 Time 0.021055 +2023-10-05 21:04:04,851 - --- validate (epoch=46)----------- +2023-10-05 21:04:04,852 - 29943 samples (256 per mini-batch) +2023-10-05 21:04:05,306 - Epoch: [46][ 10/ 117] Loss 0.338330 Top1 82.382812 Top5 98.242188 +2023-10-05 21:04:05,455 - Epoch: [46][ 20/ 117] Loss 0.368825 Top1 81.757812 Top5 97.812500 +2023-10-05 21:04:05,603 - Epoch: [46][ 30/ 117] Loss 0.365465 Top1 82.200521 Top5 97.929688 +2023-10-05 21:04:05,750 - Epoch: [46][ 40/ 117] Loss 0.366140 Top1 82.158203 Top5 97.851562 +2023-10-05 21:04:05,899 - Epoch: [46][ 50/ 117] Loss 0.362069 Top1 82.140625 Top5 97.828125 +2023-10-05 21:04:06,047 - Epoch: [46][ 60/ 117] Loss 0.365852 Top1 82.148438 Top5 97.753906 +2023-10-05 21:04:06,196 - Epoch: [46][ 70/ 117] Loss 0.366362 Top1 82.120536 Top5 97.801339 +2023-10-05 21:04:06,344 - Epoch: [46][ 80/ 117] Loss 0.363085 Top1 82.202148 Top5 97.817383 +2023-10-05 21:04:06,492 - Epoch: [46][ 90/ 117] Loss 0.361573 Top1 82.213542 Top5 97.764757 +2023-10-05 21:04:06,640 - Epoch: [46][ 100/ 117] Loss 0.361466 Top1 82.246094 Top5 97.730469 +2023-10-05 21:04:06,795 - Epoch: [46][ 110/ 117] Loss 0.363382 Top1 82.198153 Top5 97.702415 +2023-10-05 21:04:06,881 - Epoch: [46][ 117/ 117] Loss 0.361610 Top1 82.189493 Top5 97.729018 +2023-10-05 21:04:07,015 - ==> Top1: 82.189 Top5: 97.729 Loss: 0.362 + +2023-10-05 21:04:07,016 - ==> Confusion: +[[ 932 2 10 1 8 1 1 0 5 52 0 1 1 3 10 3 4 1 2 0 13] + [ 2 1032 1 0 5 22 2 31 3 0 2 2 0 0 1 4 10 0 7 2 5] + [ 3 0 947 19 1 0 30 6 0 0 2 3 7 1 3 7 2 0 9 2 14] + [ 3 0 28 927 1 6 2 0 1 0 3 0 14 4 32 2 0 5 39 1 21] + [ 29 6 1 1 961 7 0 1 0 6 0 3 0 1 9 2 11 3 1 0 8] + [ 4 36 1 0 1 976 0 25 1 0 2 10 4 17 6 0 4 0 8 3 18] + [ 0 7 24 0 1 1 1117 10 0 0 3 3 2 0 2 5 1 1 1 8 5] + [ 7 17 21 1 0 40 4 1032 0 2 8 10 6 2 0 1 1 2 42 11 11] + [ 19 3 1 2 0 3 0 1 935 38 22 2 4 17 28 4 1 0 7 0 2] + [ 116 3 2 0 6 5 2 0 38 875 0 0 0 30 10 7 2 4 0 3 16] + [ 3 4 17 13 2 2 7 4 3 2 953 4 0 8 9 2 0 2 11 0 7] + [ 1 0 3 0 0 14 1 1 2 0 0 937 35 4 0 5 0 15 0 14 3] + [ 2 0 5 5 0 2 0 1 1 0 1 34 964 1 7 9 3 16 2 5 10] + [ 1 0 1 2 3 8 0 0 8 9 13 2 6 1046 3 2 0 2 0 2 11] + [ 13 1 3 16 3 0 0 0 18 5 5 1 4 0 992 0 1 3 22 0 14] + [ 1 2 4 3 4 1 3 0 0 0 0 11 10 1 0 1039 14 12 1 9 19] + [ 0 12 2 0 7 4 0 0 0 0 0 6 1 1 1 11 1094 1 0 4 17] + [ 0 0 0 3 1 0 0 0 0 0 0 7 21 2 2 11 0 983 1 0 7] + [ 0 5 7 16 3 0 2 28 5 0 2 0 6 0 7 0 2 1 975 0 9] + [ 0 3 4 1 2 5 7 10 0 0 2 17 5 1 0 3 11 1 2 1067 11] + [ 106 203 205 63 73 159 56 87 97 63 167 116 434 326 187 75 151 85 197 229 4826]] + +2023-10-05 21:04:07,017 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:04:07,017 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:04:07,030 - + +2023-10-05 21:04:07,030 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:04:08,030 - Epoch: [47][ 10/ 1236] Overall Loss 0.342754 Objective Loss 0.342754 LR 0.001000 Time 0.099956 +2023-10-05 21:04:08,231 - Epoch: [47][ 20/ 1236] Overall Loss 0.316654 Objective Loss 0.316654 LR 0.001000 Time 0.060010 +2023-10-05 21:04:08,430 - Epoch: [47][ 30/ 1236] Overall Loss 0.318199 Objective Loss 0.318199 LR 0.001000 Time 0.046629 +2023-10-05 21:04:08,631 - Epoch: [47][ 40/ 1236] Overall Loss 0.317959 Objective Loss 0.317959 LR 0.001000 Time 0.039992 +2023-10-05 21:04:08,830 - Epoch: [47][ 50/ 1236] Overall Loss 0.323236 Objective Loss 0.323236 LR 0.001000 Time 0.035963 +2023-10-05 21:04:09,031 - Epoch: [47][ 60/ 1236] Overall Loss 0.322765 Objective Loss 0.322765 LR 0.001000 Time 0.033315 +2023-10-05 21:04:09,230 - Epoch: [47][ 70/ 1236] Overall Loss 0.319999 Objective Loss 0.319999 LR 0.001000 Time 0.031393 +2023-10-05 21:04:09,429 - Epoch: [47][ 80/ 1236] Overall Loss 0.318937 Objective Loss 0.318937 LR 0.001000 Time 0.029955 +2023-10-05 21:04:09,628 - Epoch: [47][ 90/ 1236] Overall Loss 0.319536 Objective Loss 0.319536 LR 0.001000 Time 0.028836 +2023-10-05 21:04:09,830 - Epoch: [47][ 100/ 1236] Overall Loss 0.316394 Objective Loss 0.316394 LR 0.001000 Time 0.027961 +2023-10-05 21:04:10,028 - Epoch: [47][ 110/ 1236] Overall Loss 0.316059 Objective Loss 0.316059 LR 0.001000 Time 0.027224 +2023-10-05 21:04:10,229 - Epoch: [47][ 120/ 1236] Overall Loss 0.314893 Objective Loss 0.314893 LR 0.001000 Time 0.026628 +2023-10-05 21:04:10,432 - Epoch: [47][ 130/ 1236] Overall Loss 0.315319 Objective Loss 0.315319 LR 0.001000 Time 0.026132 +2023-10-05 21:04:10,631 - Epoch: [47][ 140/ 1236] Overall Loss 0.314633 Objective Loss 0.314633 LR 0.001000 Time 0.025689 +2023-10-05 21:04:10,832 - Epoch: [47][ 150/ 1236] Overall Loss 0.313387 Objective Loss 0.313387 LR 0.001000 Time 0.025312 +2023-10-05 21:04:11,032 - Epoch: [47][ 160/ 1236] Overall Loss 0.312417 Objective Loss 0.312417 LR 0.001000 Time 0.024978 +2023-10-05 21:04:11,233 - Epoch: [47][ 170/ 1236] Overall Loss 0.313353 Objective Loss 0.313353 LR 0.001000 Time 0.024687 +2023-10-05 21:04:11,433 - Epoch: [47][ 180/ 1236] Overall Loss 0.315077 Objective Loss 0.315077 LR 0.001000 Time 0.024425 +2023-10-05 21:04:11,633 - Epoch: [47][ 190/ 1236] Overall Loss 0.317203 Objective Loss 0.317203 LR 0.001000 Time 0.024194 +2023-10-05 21:04:11,833 - Epoch: [47][ 200/ 1236] Overall Loss 0.316005 Objective Loss 0.316005 LR 0.001000 Time 0.023982 +2023-10-05 21:04:12,034 - Epoch: [47][ 210/ 1236] Overall Loss 0.315825 Objective Loss 0.315825 LR 0.001000 Time 0.023794 +2023-10-05 21:04:12,234 - Epoch: [47][ 220/ 1236] Overall Loss 0.316232 Objective Loss 0.316232 LR 0.001000 Time 0.023620 +2023-10-05 21:04:12,435 - Epoch: [47][ 230/ 1236] Overall Loss 0.315831 Objective Loss 0.315831 LR 0.001000 Time 0.023465 +2023-10-05 21:04:12,634 - Epoch: [47][ 240/ 1236] Overall Loss 0.315705 Objective Loss 0.315705 LR 0.001000 Time 0.023318 +2023-10-05 21:04:12,835 - Epoch: [47][ 250/ 1236] Overall Loss 0.315802 Objective Loss 0.315802 LR 0.001000 Time 0.023186 +2023-10-05 21:04:13,035 - Epoch: [47][ 260/ 1236] Overall Loss 0.316228 Objective Loss 0.316228 LR 0.001000 Time 0.023061 +2023-10-05 21:04:13,235 - Epoch: [47][ 270/ 1236] Overall Loss 0.315829 Objective Loss 0.315829 LR 0.001000 Time 0.022948 +2023-10-05 21:04:13,435 - Epoch: [47][ 280/ 1236] Overall Loss 0.316843 Objective Loss 0.316843 LR 0.001000 Time 0.022842 +2023-10-05 21:04:13,636 - Epoch: [47][ 290/ 1236] Overall Loss 0.317017 Objective Loss 0.317017 LR 0.001000 Time 0.022745 +2023-10-05 21:04:13,836 - Epoch: [47][ 300/ 1236] Overall Loss 0.316152 Objective Loss 0.316152 LR 0.001000 Time 0.022652 +2023-10-05 21:04:14,036 - Epoch: [47][ 310/ 1236] Overall Loss 0.316363 Objective Loss 0.316363 LR 0.001000 Time 0.022566 +2023-10-05 21:04:14,236 - Epoch: [47][ 320/ 1236] Overall Loss 0.316840 Objective Loss 0.316840 LR 0.001000 Time 0.022485 +2023-10-05 21:04:14,437 - Epoch: [47][ 330/ 1236] Overall Loss 0.316341 Objective Loss 0.316341 LR 0.001000 Time 0.022412 +2023-10-05 21:04:14,636 - Epoch: [47][ 340/ 1236] Overall Loss 0.315526 Objective Loss 0.315526 LR 0.001000 Time 0.022338 +2023-10-05 21:04:14,837 - Epoch: [47][ 350/ 1236] Overall Loss 0.315596 Objective Loss 0.315596 LR 0.001000 Time 0.022273 +2023-10-05 21:04:15,037 - Epoch: [47][ 360/ 1236] Overall Loss 0.315441 Objective Loss 0.315441 LR 0.001000 Time 0.022209 +2023-10-05 21:04:15,239 - Epoch: [47][ 370/ 1236] Overall Loss 0.316092 Objective Loss 0.316092 LR 0.001000 Time 0.022153 +2023-10-05 21:04:15,439 - Epoch: [47][ 380/ 1236] Overall Loss 0.315956 Objective Loss 0.315956 LR 0.001000 Time 0.022096 +2023-10-05 21:04:15,641 - Epoch: [47][ 390/ 1236] Overall Loss 0.316484 Objective Loss 0.316484 LR 0.001000 Time 0.022045 +2023-10-05 21:04:15,840 - Epoch: [47][ 400/ 1236] Overall Loss 0.317553 Objective Loss 0.317553 LR 0.001000 Time 0.021991 +2023-10-05 21:04:16,041 - Epoch: [47][ 410/ 1236] Overall Loss 0.317843 Objective Loss 0.317843 LR 0.001000 Time 0.021945 +2023-10-05 21:04:16,240 - Epoch: [47][ 420/ 1236] Overall Loss 0.317809 Objective Loss 0.317809 LR 0.001000 Time 0.021895 +2023-10-05 21:04:16,441 - Epoch: [47][ 430/ 1236] Overall Loss 0.317775 Objective Loss 0.317775 LR 0.001000 Time 0.021853 +2023-10-05 21:04:16,641 - Epoch: [47][ 440/ 1236] Overall Loss 0.317266 Objective Loss 0.317266 LR 0.001000 Time 0.021809 +2023-10-05 21:04:16,842 - Epoch: [47][ 450/ 1236] Overall Loss 0.316680 Objective Loss 0.316680 LR 0.001000 Time 0.021771 +2023-10-05 21:04:17,042 - Epoch: [47][ 460/ 1236] Overall Loss 0.316585 Objective Loss 0.316585 LR 0.001000 Time 0.021731 +2023-10-05 21:04:17,244 - Epoch: [47][ 470/ 1236] Overall Loss 0.317188 Objective Loss 0.317188 LR 0.001000 Time 0.021698 +2023-10-05 21:04:17,444 - Epoch: [47][ 480/ 1236] Overall Loss 0.316760 Objective Loss 0.316760 LR 0.001000 Time 0.021662 +2023-10-05 21:04:17,646 - Epoch: [47][ 490/ 1236] Overall Loss 0.316968 Objective Loss 0.316968 LR 0.001000 Time 0.021631 +2023-10-05 21:04:17,845 - Epoch: [47][ 500/ 1236] Overall Loss 0.317591 Objective Loss 0.317591 LR 0.001000 Time 0.021595 +2023-10-05 21:04:18,046 - Epoch: [47][ 510/ 1236] Overall Loss 0.317712 Objective Loss 0.317712 LR 0.001000 Time 0.021566 +2023-10-05 21:04:18,245 - Epoch: [47][ 520/ 1236] Overall Loss 0.318042 Objective Loss 0.318042 LR 0.001000 Time 0.021533 +2023-10-05 21:04:18,446 - Epoch: [47][ 530/ 1236] Overall Loss 0.317719 Objective Loss 0.317719 LR 0.001000 Time 0.021506 +2023-10-05 21:04:18,645 - Epoch: [47][ 540/ 1236] Overall Loss 0.317613 Objective Loss 0.317613 LR 0.001000 Time 0.021475 +2023-10-05 21:04:18,847 - Epoch: [47][ 550/ 1236] Overall Loss 0.317970 Objective Loss 0.317970 LR 0.001000 Time 0.021451 +2023-10-05 21:04:19,046 - Epoch: [47][ 560/ 1236] Overall Loss 0.317124 Objective Loss 0.317124 LR 0.001000 Time 0.021422 +2023-10-05 21:04:19,247 - Epoch: [47][ 570/ 1236] Overall Loss 0.317461 Objective Loss 0.317461 LR 0.001000 Time 0.021398 +2023-10-05 21:04:19,446 - Epoch: [47][ 580/ 1236] Overall Loss 0.317573 Objective Loss 0.317573 LR 0.001000 Time 0.021372 +2023-10-05 21:04:19,647 - Epoch: [47][ 590/ 1236] Overall Loss 0.317471 Objective Loss 0.317471 LR 0.001000 Time 0.021350 +2023-10-05 21:04:19,846 - Epoch: [47][ 600/ 1236] Overall Loss 0.317649 Objective Loss 0.317649 LR 0.001000 Time 0.021325 +2023-10-05 21:04:20,047 - Epoch: [47][ 610/ 1236] Overall Loss 0.317780 Objective Loss 0.317780 LR 0.001000 Time 0.021305 +2023-10-05 21:04:20,247 - Epoch: [47][ 620/ 1236] Overall Loss 0.317730 Objective Loss 0.317730 LR 0.001000 Time 0.021282 +2023-10-05 21:04:20,449 - Epoch: [47][ 630/ 1236] Overall Loss 0.317514 Objective Loss 0.317514 LR 0.001000 Time 0.021265 +2023-10-05 21:04:20,648 - Epoch: [47][ 640/ 1236] Overall Loss 0.316850 Objective Loss 0.316850 LR 0.001000 Time 0.021243 +2023-10-05 21:04:20,849 - Epoch: [47][ 650/ 1236] Overall Loss 0.316658 Objective Loss 0.316658 LR 0.001000 Time 0.021226 +2023-10-05 21:04:21,048 - Epoch: [47][ 660/ 1236] Overall Loss 0.316273 Objective Loss 0.316273 LR 0.001000 Time 0.021205 +2023-10-05 21:04:21,250 - Epoch: [47][ 670/ 1236] Overall Loss 0.316335 Objective Loss 0.316335 LR 0.001000 Time 0.021189 +2023-10-05 21:04:21,449 - Epoch: [47][ 680/ 1236] Overall Loss 0.316453 Objective Loss 0.316453 LR 0.001000 Time 0.021169 +2023-10-05 21:04:21,651 - Epoch: [47][ 690/ 1236] Overall Loss 0.316532 Objective Loss 0.316532 LR 0.001000 Time 0.021154 +2023-10-05 21:04:21,850 - Epoch: [47][ 700/ 1236] Overall Loss 0.316364 Objective Loss 0.316364 LR 0.001000 Time 0.021136 +2023-10-05 21:04:22,051 - Epoch: [47][ 710/ 1236] Overall Loss 0.316226 Objective Loss 0.316226 LR 0.001000 Time 0.021121 +2023-10-05 21:04:22,250 - Epoch: [47][ 720/ 1236] Overall Loss 0.316349 Objective Loss 0.316349 LR 0.001000 Time 0.021104 +2023-10-05 21:04:22,451 - Epoch: [47][ 730/ 1236] Overall Loss 0.316117 Objective Loss 0.316117 LR 0.001000 Time 0.021090 +2023-10-05 21:04:22,651 - Epoch: [47][ 740/ 1236] Overall Loss 0.315944 Objective Loss 0.315944 LR 0.001000 Time 0.021074 +2023-10-05 21:04:22,852 - Epoch: [47][ 750/ 1236] Overall Loss 0.316126 Objective Loss 0.316126 LR 0.001000 Time 0.021061 +2023-10-05 21:04:23,051 - Epoch: [47][ 760/ 1236] Overall Loss 0.316305 Objective Loss 0.316305 LR 0.001000 Time 0.021045 +2023-10-05 21:04:23,252 - Epoch: [47][ 770/ 1236] Overall Loss 0.316487 Objective Loss 0.316487 LR 0.001000 Time 0.021032 +2023-10-05 21:04:23,451 - Epoch: [47][ 780/ 1236] Overall Loss 0.316450 Objective Loss 0.316450 LR 0.001000 Time 0.021017 +2023-10-05 21:04:23,652 - Epoch: [47][ 790/ 1236] Overall Loss 0.316671 Objective Loss 0.316671 LR 0.001000 Time 0.021005 +2023-10-05 21:04:23,851 - Epoch: [47][ 800/ 1236] Overall Loss 0.316328 Objective Loss 0.316328 LR 0.001000 Time 0.020991 +2023-10-05 21:04:24,052 - Epoch: [47][ 810/ 1236] Overall Loss 0.316774 Objective Loss 0.316774 LR 0.001000 Time 0.020980 +2023-10-05 21:04:24,251 - Epoch: [47][ 820/ 1236] Overall Loss 0.316491 Objective Loss 0.316491 LR 0.001000 Time 0.020966 +2023-10-05 21:04:24,452 - Epoch: [47][ 830/ 1236] Overall Loss 0.316898 Objective Loss 0.316898 LR 0.001000 Time 0.020955 +2023-10-05 21:04:24,652 - Epoch: [47][ 840/ 1236] Overall Loss 0.316822 Objective Loss 0.316822 LR 0.001000 Time 0.020944 +2023-10-05 21:04:24,854 - Epoch: [47][ 850/ 1236] Overall Loss 0.316720 Objective Loss 0.316720 LR 0.001000 Time 0.020934 +2023-10-05 21:04:25,053 - Epoch: [47][ 860/ 1236] Overall Loss 0.316963 Objective Loss 0.316963 LR 0.001000 Time 0.020922 +2023-10-05 21:04:25,255 - Epoch: [47][ 870/ 1236] Overall Loss 0.316907 Objective Loss 0.316907 LR 0.001000 Time 0.020913 +2023-10-05 21:04:25,454 - Epoch: [47][ 880/ 1236] Overall Loss 0.316642 Objective Loss 0.316642 LR 0.001000 Time 0.020901 +2023-10-05 21:04:25,655 - Epoch: [47][ 890/ 1236] Overall Loss 0.316643 Objective Loss 0.316643 LR 0.001000 Time 0.020892 +2023-10-05 21:04:25,854 - Epoch: [47][ 900/ 1236] Overall Loss 0.316960 Objective Loss 0.316960 LR 0.001000 Time 0.020880 +2023-10-05 21:04:26,056 - Epoch: [47][ 910/ 1236] Overall Loss 0.316753 Objective Loss 0.316753 LR 0.001000 Time 0.020872 +2023-10-05 21:04:26,254 - Epoch: [47][ 920/ 1236] Overall Loss 0.316601 Objective Loss 0.316601 LR 0.001000 Time 0.020860 +2023-10-05 21:04:26,456 - Epoch: [47][ 930/ 1236] Overall Loss 0.316574 Objective Loss 0.316574 LR 0.001000 Time 0.020852 +2023-10-05 21:04:26,654 - Epoch: [47][ 940/ 1236] Overall Loss 0.316558 Objective Loss 0.316558 LR 0.001000 Time 0.020841 +2023-10-05 21:04:26,856 - Epoch: [47][ 950/ 1236] Overall Loss 0.316815 Objective Loss 0.316815 LR 0.001000 Time 0.020833 +2023-10-05 21:04:27,054 - Epoch: [47][ 960/ 1236] Overall Loss 0.316984 Objective Loss 0.316984 LR 0.001000 Time 0.020823 +2023-10-05 21:04:27,256 - Epoch: [47][ 970/ 1236] Overall Loss 0.317444 Objective Loss 0.317444 LR 0.001000 Time 0.020816 +2023-10-05 21:04:27,456 - Epoch: [47][ 980/ 1236] Overall Loss 0.317617 Objective Loss 0.317617 LR 0.001000 Time 0.020807 +2023-10-05 21:04:27,661 - Epoch: [47][ 990/ 1236] Overall Loss 0.317453 Objective Loss 0.317453 LR 0.001000 Time 0.020804 +2023-10-05 21:04:27,862 - Epoch: [47][ 1000/ 1236] Overall Loss 0.317494 Objective Loss 0.317494 LR 0.001000 Time 0.020796 +2023-10-05 21:04:28,066 - Epoch: [47][ 1010/ 1236] Overall Loss 0.317114 Objective Loss 0.317114 LR 0.001000 Time 0.020792 +2023-10-05 21:04:28,267 - Epoch: [47][ 1020/ 1236] Overall Loss 0.317122 Objective Loss 0.317122 LR 0.001000 Time 0.020785 +2023-10-05 21:04:28,471 - Epoch: [47][ 1030/ 1236] Overall Loss 0.317630 Objective Loss 0.317630 LR 0.001000 Time 0.020781 +2023-10-05 21:04:28,672 - Epoch: [47][ 1040/ 1236] Overall Loss 0.317635 Objective Loss 0.317635 LR 0.001000 Time 0.020774 +2023-10-05 21:04:28,876 - Epoch: [47][ 1050/ 1236] Overall Loss 0.317631 Objective Loss 0.317631 LR 0.001000 Time 0.020770 +2023-10-05 21:04:29,077 - Epoch: [47][ 1060/ 1236] Overall Loss 0.317723 Objective Loss 0.317723 LR 0.001000 Time 0.020763 +2023-10-05 21:04:29,281 - Epoch: [47][ 1070/ 1236] Overall Loss 0.317693 Objective Loss 0.317693 LR 0.001000 Time 0.020760 +2023-10-05 21:04:29,482 - Epoch: [47][ 1080/ 1236] Overall Loss 0.317430 Objective Loss 0.317430 LR 0.001000 Time 0.020753 +2023-10-05 21:04:29,686 - Epoch: [47][ 1090/ 1236] Overall Loss 0.317260 Objective Loss 0.317260 LR 0.001000 Time 0.020749 +2023-10-05 21:04:29,887 - Epoch: [47][ 1100/ 1236] Overall Loss 0.317192 Objective Loss 0.317192 LR 0.001000 Time 0.020743 +2023-10-05 21:04:30,091 - Epoch: [47][ 1110/ 1236] Overall Loss 0.317109 Objective Loss 0.317109 LR 0.001000 Time 0.020740 +2023-10-05 21:04:30,292 - Epoch: [47][ 1120/ 1236] Overall Loss 0.317392 Objective Loss 0.317392 LR 0.001000 Time 0.020734 +2023-10-05 21:04:30,496 - Epoch: [47][ 1130/ 1236] Overall Loss 0.317525 Objective Loss 0.317525 LR 0.001000 Time 0.020730 +2023-10-05 21:04:30,697 - Epoch: [47][ 1140/ 1236] Overall Loss 0.317621 Objective Loss 0.317621 LR 0.001000 Time 0.020725 +2023-10-05 21:04:30,901 - Epoch: [47][ 1150/ 1236] Overall Loss 0.317486 Objective Loss 0.317486 LR 0.001000 Time 0.020722 +2023-10-05 21:04:31,102 - Epoch: [47][ 1160/ 1236] Overall Loss 0.317409 Objective Loss 0.317409 LR 0.001000 Time 0.020716 +2023-10-05 21:04:31,306 - Epoch: [47][ 1170/ 1236] Overall Loss 0.317253 Objective Loss 0.317253 LR 0.001000 Time 0.020713 +2023-10-05 21:04:31,507 - Epoch: [47][ 1180/ 1236] Overall Loss 0.317422 Objective Loss 0.317422 LR 0.001000 Time 0.020707 +2023-10-05 21:04:31,711 - Epoch: [47][ 1190/ 1236] Overall Loss 0.317421 Objective Loss 0.317421 LR 0.001000 Time 0.020704 +2023-10-05 21:04:31,912 - Epoch: [47][ 1200/ 1236] Overall Loss 0.317430 Objective Loss 0.317430 LR 0.001000 Time 0.020699 +2023-10-05 21:04:32,115 - Epoch: [47][ 1210/ 1236] Overall Loss 0.317216 Objective Loss 0.317216 LR 0.001000 Time 0.020696 +2023-10-05 21:04:32,316 - Epoch: [47][ 1220/ 1236] Overall Loss 0.317312 Objective Loss 0.317312 LR 0.001000 Time 0.020691 +2023-10-05 21:04:32,568 - Epoch: [47][ 1230/ 1236] Overall Loss 0.317472 Objective Loss 0.317472 LR 0.001000 Time 0.020727 +2023-10-05 21:04:32,685 - Epoch: [47][ 1236/ 1236] Overall Loss 0.317493 Objective Loss 0.317493 Top1 80.855397 Top5 96.334012 LR 0.001000 Time 0.020721 +2023-10-05 21:04:32,818 - --- validate (epoch=47)----------- +2023-10-05 21:04:32,818 - 29943 samples (256 per mini-batch) +2023-10-05 21:04:33,267 - Epoch: [47][ 10/ 117] Loss 0.369359 Top1 79.140625 Top5 97.187500 +2023-10-05 21:04:33,418 - Epoch: [47][ 20/ 117] Loss 0.353150 Top1 80.273438 Top5 97.304688 +2023-10-05 21:04:33,568 - Epoch: [47][ 30/ 117] Loss 0.363044 Top1 80.481771 Top5 97.434896 +2023-10-05 21:04:33,717 - Epoch: [47][ 40/ 117] Loss 0.357742 Top1 80.712891 Top5 97.597656 +2023-10-05 21:04:33,866 - Epoch: [47][ 50/ 117] Loss 0.366965 Top1 80.437500 Top5 97.546875 +2023-10-05 21:04:34,017 - Epoch: [47][ 60/ 117] Loss 0.374163 Top1 80.351562 Top5 97.467448 +2023-10-05 21:04:34,170 - Epoch: [47][ 70/ 117] Loss 0.378294 Top1 80.267857 Top5 97.416295 +2023-10-05 21:04:34,325 - Epoch: [47][ 80/ 117] Loss 0.376956 Top1 80.375977 Top5 97.387695 +2023-10-05 21:04:34,479 - Epoch: [47][ 90/ 117] Loss 0.374310 Top1 80.455729 Top5 97.413194 +2023-10-05 21:04:34,634 - Epoch: [47][ 100/ 117] Loss 0.375596 Top1 80.527344 Top5 97.406250 +2023-10-05 21:04:34,795 - Epoch: [47][ 110/ 117] Loss 0.371580 Top1 80.607244 Top5 97.460938 +2023-10-05 21:04:34,881 - Epoch: [47][ 117/ 117] Loss 0.373343 Top1 80.519654 Top5 97.431787 +2023-10-05 21:04:35,022 - ==> Top1: 80.520 Top5: 97.432 Loss: 0.373 + +2023-10-05 21:04:35,022 - ==> Confusion: +[[ 929 4 8 0 5 3 0 0 7 64 2 0 2 2 3 3 8 1 1 0 8] + [ 2 1030 1 1 9 43 1 17 0 0 2 1 0 0 3 3 9 0 1 2 6] + [ 10 2 936 17 1 3 38 7 0 2 1 2 7 2 1 7 2 0 8 2 8] + [ 6 2 31 949 1 6 4 0 2 0 13 0 5 4 24 4 3 8 13 1 13] + [ 37 4 2 1 940 8 1 0 2 10 2 7 1 0 7 7 16 0 0 0 5] + [ 5 32 1 0 4 990 2 16 1 0 5 7 1 14 5 4 8 1 2 8 10] + [ 0 6 35 0 0 5 1106 6 0 0 2 2 1 0 0 9 1 1 1 8 8] + [ 8 28 30 0 4 59 1 1018 2 1 1 11 5 1 0 2 0 0 22 20 5] + [ 26 4 1 1 1 5 0 0 952 45 12 3 4 10 14 2 3 0 4 0 2] + [ 134 3 0 1 5 3 1 1 28 904 0 2 0 17 4 4 4 2 0 2 4] + [ 1 7 19 4 2 1 5 6 22 2 947 3 0 12 2 6 1 1 6 0 6] + [ 3 0 3 1 0 23 0 1 0 0 0 933 26 4 0 6 2 12 0 19 2] + [ 1 0 7 7 0 6 0 0 2 0 1 42 936 2 2 15 4 28 5 3 7] + [ 4 0 2 0 5 14 0 0 10 14 8 7 4 1025 1 3 4 2 0 7 9] + [ 22 4 3 21 7 0 0 0 30 8 4 1 1 1 970 0 7 3 7 0 12] + [ 0 3 2 2 2 1 1 0 0 0 0 8 6 0 0 1067 17 12 1 8 4] + [ 0 6 3 0 3 10 0 0 0 0 0 2 2 0 3 12 1112 0 1 4 3] + [ 0 1 1 1 0 0 0 0 1 0 0 6 25 0 3 12 4 980 2 0 2] + [ 1 11 7 21 2 1 0 39 8 1 3 2 4 0 17 1 5 0 933 1 11] + [ 0 2 10 0 1 16 2 8 0 0 0 14 8 0 0 4 12 0 0 1070 5] + [ 197 204 206 75 91 269 60 99 99 85 168 134 386 318 136 103 435 97 85 275 4383]] + +2023-10-05 21:04:35,024 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:04:35,024 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:04:35,029 - + +2023-10-05 21:04:35,030 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:04:36,135 - Epoch: [48][ 10/ 1236] Overall Loss 0.313607 Objective Loss 0.313607 LR 0.001000 Time 0.110498 +2023-10-05 21:04:36,332 - Epoch: [48][ 20/ 1236] Overall Loss 0.305225 Objective Loss 0.305225 LR 0.001000 Time 0.065098 +2023-10-05 21:04:36,531 - Epoch: [48][ 30/ 1236] Overall Loss 0.313340 Objective Loss 0.313340 LR 0.001000 Time 0.050022 +2023-10-05 21:04:36,729 - Epoch: [48][ 40/ 1236] Overall Loss 0.312727 Objective Loss 0.312727 LR 0.001000 Time 0.042444 +2023-10-05 21:04:36,928 - Epoch: [48][ 50/ 1236] Overall Loss 0.309059 Objective Loss 0.309059 LR 0.001000 Time 0.037928 +2023-10-05 21:04:37,125 - Epoch: [48][ 60/ 1236] Overall Loss 0.308051 Objective Loss 0.308051 LR 0.001000 Time 0.034898 +2023-10-05 21:04:37,324 - Epoch: [48][ 70/ 1236] Overall Loss 0.312379 Objective Loss 0.312379 LR 0.001000 Time 0.032752 +2023-10-05 21:04:37,522 - Epoch: [48][ 80/ 1236] Overall Loss 0.306389 Objective Loss 0.306389 LR 0.001000 Time 0.031124 +2023-10-05 21:04:37,721 - Epoch: [48][ 90/ 1236] Overall Loss 0.307320 Objective Loss 0.307320 LR 0.001000 Time 0.029874 +2023-10-05 21:04:37,920 - Epoch: [48][ 100/ 1236] Overall Loss 0.307136 Objective Loss 0.307136 LR 0.001000 Time 0.028869 +2023-10-05 21:04:38,115 - Epoch: [48][ 110/ 1236] Overall Loss 0.304427 Objective Loss 0.304427 LR 0.001000 Time 0.028015 +2023-10-05 21:04:38,311 - Epoch: [48][ 120/ 1236] Overall Loss 0.301992 Objective Loss 0.301992 LR 0.001000 Time 0.027311 +2023-10-05 21:04:38,507 - Epoch: [48][ 130/ 1236] Overall Loss 0.301534 Objective Loss 0.301534 LR 0.001000 Time 0.026715 +2023-10-05 21:04:38,702 - Epoch: [48][ 140/ 1236] Overall Loss 0.303745 Objective Loss 0.303745 LR 0.001000 Time 0.026204 +2023-10-05 21:04:38,898 - Epoch: [48][ 150/ 1236] Overall Loss 0.301751 Objective Loss 0.301751 LR 0.001000 Time 0.025761 +2023-10-05 21:04:39,094 - Epoch: [48][ 160/ 1236] Overall Loss 0.301764 Objective Loss 0.301764 LR 0.001000 Time 0.025371 +2023-10-05 21:04:39,290 - Epoch: [48][ 170/ 1236] Overall Loss 0.301904 Objective Loss 0.301904 LR 0.001000 Time 0.025029 +2023-10-05 21:04:39,485 - Epoch: [48][ 180/ 1236] Overall Loss 0.300863 Objective Loss 0.300863 LR 0.001000 Time 0.024721 +2023-10-05 21:04:39,681 - Epoch: [48][ 190/ 1236] Overall Loss 0.301432 Objective Loss 0.301432 LR 0.001000 Time 0.024450 +2023-10-05 21:04:39,876 - Epoch: [48][ 200/ 1236] Overall Loss 0.301685 Objective Loss 0.301685 LR 0.001000 Time 0.024205 +2023-10-05 21:04:40,072 - Epoch: [48][ 210/ 1236] Overall Loss 0.301265 Objective Loss 0.301265 LR 0.001000 Time 0.023980 +2023-10-05 21:04:40,268 - Epoch: [48][ 220/ 1236] Overall Loss 0.301965 Objective Loss 0.301965 LR 0.001000 Time 0.023781 +2023-10-05 21:04:40,463 - Epoch: [48][ 230/ 1236] Overall Loss 0.301604 Objective Loss 0.301604 LR 0.001000 Time 0.023596 +2023-10-05 21:04:40,659 - Epoch: [48][ 240/ 1236] Overall Loss 0.302451 Objective Loss 0.302451 LR 0.001000 Time 0.023429 +2023-10-05 21:04:40,855 - Epoch: [48][ 250/ 1236] Overall Loss 0.303606 Objective Loss 0.303606 LR 0.001000 Time 0.023271 +2023-10-05 21:04:41,051 - Epoch: [48][ 260/ 1236] Overall Loss 0.304280 Objective Loss 0.304280 LR 0.001000 Time 0.023129 +2023-10-05 21:04:41,246 - Epoch: [48][ 270/ 1236] Overall Loss 0.303540 Objective Loss 0.303540 LR 0.001000 Time 0.022997 +2023-10-05 21:04:41,442 - Epoch: [48][ 280/ 1236] Overall Loss 0.303017 Objective Loss 0.303017 LR 0.001000 Time 0.022873 +2023-10-05 21:04:41,638 - Epoch: [48][ 290/ 1236] Overall Loss 0.303061 Objective Loss 0.303061 LR 0.001000 Time 0.022758 +2023-10-05 21:04:41,834 - Epoch: [48][ 300/ 1236] Overall Loss 0.303338 Objective Loss 0.303338 LR 0.001000 Time 0.022652 +2023-10-05 21:04:42,029 - Epoch: [48][ 310/ 1236] Overall Loss 0.303769 Objective Loss 0.303769 LR 0.001000 Time 0.022552 +2023-10-05 21:04:42,227 - Epoch: [48][ 320/ 1236] Overall Loss 0.302985 Objective Loss 0.302985 LR 0.001000 Time 0.022462 +2023-10-05 21:04:42,428 - Epoch: [48][ 330/ 1236] Overall Loss 0.304182 Objective Loss 0.304182 LR 0.001000 Time 0.022390 +2023-10-05 21:04:42,627 - Epoch: [48][ 340/ 1236] Overall Loss 0.303811 Objective Loss 0.303811 LR 0.001000 Time 0.022316 +2023-10-05 21:04:42,828 - Epoch: [48][ 350/ 1236] Overall Loss 0.303657 Objective Loss 0.303657 LR 0.001000 Time 0.022252 +2023-10-05 21:04:43,027 - Epoch: [48][ 360/ 1236] Overall Loss 0.303805 Objective Loss 0.303805 LR 0.001000 Time 0.022187 +2023-10-05 21:04:43,229 - Epoch: [48][ 370/ 1236] Overall Loss 0.305232 Objective Loss 0.305232 LR 0.001000 Time 0.022132 +2023-10-05 21:04:43,431 - Epoch: [48][ 380/ 1236] Overall Loss 0.304726 Objective Loss 0.304726 LR 0.001000 Time 0.022079 +2023-10-05 21:04:43,629 - Epoch: [48][ 390/ 1236] Overall Loss 0.304184 Objective Loss 0.304184 LR 0.001000 Time 0.022022 +2023-10-05 21:04:43,832 - Epoch: [48][ 400/ 1236] Overall Loss 0.304261 Objective Loss 0.304261 LR 0.001000 Time 0.021976 +2023-10-05 21:04:44,034 - Epoch: [48][ 410/ 1236] Overall Loss 0.305352 Objective Loss 0.305352 LR 0.001000 Time 0.021934 +2023-10-05 21:04:44,236 - Epoch: [48][ 420/ 1236] Overall Loss 0.304508 Objective Loss 0.304508 LR 0.001000 Time 0.021891 +2023-10-05 21:04:44,439 - Epoch: [48][ 430/ 1236] Overall Loss 0.304833 Objective Loss 0.304833 LR 0.001000 Time 0.021853 +2023-10-05 21:04:44,641 - Epoch: [48][ 440/ 1236] Overall Loss 0.305229 Objective Loss 0.305229 LR 0.001000 Time 0.021815 +2023-10-05 21:04:44,844 - Epoch: [48][ 450/ 1236] Overall Loss 0.305139 Objective Loss 0.305139 LR 0.001000 Time 0.021780 +2023-10-05 21:04:45,046 - Epoch: [48][ 460/ 1236] Overall Loss 0.305445 Objective Loss 0.305445 LR 0.001000 Time 0.021745 +2023-10-05 21:04:45,249 - Epoch: [48][ 470/ 1236] Overall Loss 0.304566 Objective Loss 0.304566 LR 0.001000 Time 0.021714 +2023-10-05 21:04:45,451 - Epoch: [48][ 480/ 1236] Overall Loss 0.304545 Objective Loss 0.304545 LR 0.001000 Time 0.021682 +2023-10-05 21:04:45,654 - Epoch: [48][ 490/ 1236] Overall Loss 0.304505 Objective Loss 0.304505 LR 0.001000 Time 0.021653 +2023-10-05 21:04:45,856 - Epoch: [48][ 500/ 1236] Overall Loss 0.304737 Objective Loss 0.304737 LR 0.001000 Time 0.021623 +2023-10-05 21:04:46,059 - Epoch: [48][ 510/ 1236] Overall Loss 0.305703 Objective Loss 0.305703 LR 0.001000 Time 0.021597 +2023-10-05 21:04:46,261 - Epoch: [48][ 520/ 1236] Overall Loss 0.305697 Objective Loss 0.305697 LR 0.001000 Time 0.021569 +2023-10-05 21:04:46,464 - Epoch: [48][ 530/ 1236] Overall Loss 0.306173 Objective Loss 0.306173 LR 0.001000 Time 0.021545 +2023-10-05 21:04:46,666 - Epoch: [48][ 540/ 1236] Overall Loss 0.306210 Objective Loss 0.306210 LR 0.001000 Time 0.021519 +2023-10-05 21:04:46,869 - Epoch: [48][ 550/ 1236] Overall Loss 0.306516 Objective Loss 0.306516 LR 0.001000 Time 0.021497 +2023-10-05 21:04:47,071 - Epoch: [48][ 560/ 1236] Overall Loss 0.306784 Objective Loss 0.306784 LR 0.001000 Time 0.021473 +2023-10-05 21:04:47,274 - Epoch: [48][ 570/ 1236] Overall Loss 0.306783 Objective Loss 0.306783 LR 0.001000 Time 0.021452 +2023-10-05 21:04:47,476 - Epoch: [48][ 580/ 1236] Overall Loss 0.307673 Objective Loss 0.307673 LR 0.001000 Time 0.021429 +2023-10-05 21:04:47,679 - Epoch: [48][ 590/ 1236] Overall Loss 0.307733 Objective Loss 0.307733 LR 0.001000 Time 0.021410 +2023-10-05 21:04:47,881 - Epoch: [48][ 600/ 1236] Overall Loss 0.308185 Objective Loss 0.308185 LR 0.001000 Time 0.021389 +2023-10-05 21:04:48,084 - Epoch: [48][ 610/ 1236] Overall Loss 0.309310 Objective Loss 0.309310 LR 0.001000 Time 0.021371 +2023-10-05 21:04:48,286 - Epoch: [48][ 620/ 1236] Overall Loss 0.309712 Objective Loss 0.309712 LR 0.001000 Time 0.021351 +2023-10-05 21:04:48,489 - Epoch: [48][ 630/ 1236] Overall Loss 0.310433 Objective Loss 0.310433 LR 0.001000 Time 0.021335 +2023-10-05 21:04:48,691 - Epoch: [48][ 640/ 1236] Overall Loss 0.311052 Objective Loss 0.311052 LR 0.001000 Time 0.021317 +2023-10-05 21:04:48,894 - Epoch: [48][ 650/ 1236] Overall Loss 0.311136 Objective Loss 0.311136 LR 0.001000 Time 0.021300 +2023-10-05 21:04:49,096 - Epoch: [48][ 660/ 1236] Overall Loss 0.311066 Objective Loss 0.311066 LR 0.001000 Time 0.021283 +2023-10-05 21:04:49,299 - Epoch: [48][ 670/ 1236] Overall Loss 0.311682 Objective Loss 0.311682 LR 0.001000 Time 0.021268 +2023-10-05 21:04:49,501 - Epoch: [48][ 680/ 1236] Overall Loss 0.311635 Objective Loss 0.311635 LR 0.001000 Time 0.021252 +2023-10-05 21:04:49,704 - Epoch: [48][ 690/ 1236] Overall Loss 0.312009 Objective Loss 0.312009 LR 0.001000 Time 0.021237 +2023-10-05 21:04:49,906 - Epoch: [48][ 700/ 1236] Overall Loss 0.312568 Objective Loss 0.312568 LR 0.001000 Time 0.021222 +2023-10-05 21:04:50,109 - Epoch: [48][ 710/ 1236] Overall Loss 0.312632 Objective Loss 0.312632 LR 0.001000 Time 0.021209 +2023-10-05 21:04:50,311 - Epoch: [48][ 720/ 1236] Overall Loss 0.312592 Objective Loss 0.312592 LR 0.001000 Time 0.021194 +2023-10-05 21:04:50,514 - Epoch: [48][ 730/ 1236] Overall Loss 0.312466 Objective Loss 0.312466 LR 0.001000 Time 0.021182 +2023-10-05 21:04:50,716 - Epoch: [48][ 740/ 1236] Overall Loss 0.312863 Objective Loss 0.312863 LR 0.001000 Time 0.021168 +2023-10-05 21:04:50,919 - Epoch: [48][ 750/ 1236] Overall Loss 0.312872 Objective Loss 0.312872 LR 0.001000 Time 0.021156 +2023-10-05 21:04:51,121 - Epoch: [48][ 760/ 1236] Overall Loss 0.312816 Objective Loss 0.312816 LR 0.001000 Time 0.021143 +2023-10-05 21:04:51,324 - Epoch: [48][ 770/ 1236] Overall Loss 0.312745 Objective Loss 0.312745 LR 0.001000 Time 0.021131 +2023-10-05 21:04:51,526 - Epoch: [48][ 780/ 1236] Overall Loss 0.312801 Objective Loss 0.312801 LR 0.001000 Time 0.021119 +2023-10-05 21:04:51,729 - Epoch: [48][ 790/ 1236] Overall Loss 0.313051 Objective Loss 0.313051 LR 0.001000 Time 0.021108 +2023-10-05 21:04:51,931 - Epoch: [48][ 800/ 1236] Overall Loss 0.313148 Objective Loss 0.313148 LR 0.001000 Time 0.021097 +2023-10-05 21:04:52,134 - Epoch: [48][ 810/ 1236] Overall Loss 0.313409 Objective Loss 0.313409 LR 0.001000 Time 0.021086 +2023-10-05 21:04:52,336 - Epoch: [48][ 820/ 1236] Overall Loss 0.313597 Objective Loss 0.313597 LR 0.001000 Time 0.021076 +2023-10-05 21:04:52,539 - Epoch: [48][ 830/ 1236] Overall Loss 0.313729 Objective Loss 0.313729 LR 0.001000 Time 0.021066 +2023-10-05 21:04:52,741 - Epoch: [48][ 840/ 1236] Overall Loss 0.313573 Objective Loss 0.313573 LR 0.001000 Time 0.021055 +2023-10-05 21:04:52,944 - Epoch: [48][ 850/ 1236] Overall Loss 0.314196 Objective Loss 0.314196 LR 0.001000 Time 0.021046 +2023-10-05 21:04:53,146 - Epoch: [48][ 860/ 1236] Overall Loss 0.314265 Objective Loss 0.314265 LR 0.001000 Time 0.021035 +2023-10-05 21:04:53,349 - Epoch: [48][ 870/ 1236] Overall Loss 0.314658 Objective Loss 0.314658 LR 0.001000 Time 0.021027 +2023-10-05 21:04:53,551 - Epoch: [48][ 880/ 1236] Overall Loss 0.314528 Objective Loss 0.314528 LR 0.001000 Time 0.021017 +2023-10-05 21:04:53,754 - Epoch: [48][ 890/ 1236] Overall Loss 0.314609 Objective Loss 0.314609 LR 0.001000 Time 0.021008 +2023-10-05 21:04:53,956 - Epoch: [48][ 900/ 1236] Overall Loss 0.315029 Objective Loss 0.315029 LR 0.001000 Time 0.020999 +2023-10-05 21:04:54,159 - Epoch: [48][ 910/ 1236] Overall Loss 0.315073 Objective Loss 0.315073 LR 0.001000 Time 0.020991 +2023-10-05 21:04:54,361 - Epoch: [48][ 920/ 1236] Overall Loss 0.315644 Objective Loss 0.315644 LR 0.001000 Time 0.020982 +2023-10-05 21:04:54,564 - Epoch: [48][ 930/ 1236] Overall Loss 0.315948 Objective Loss 0.315948 LR 0.001000 Time 0.020975 +2023-10-05 21:04:54,766 - Epoch: [48][ 940/ 1236] Overall Loss 0.315648 Objective Loss 0.315648 LR 0.001000 Time 0.020966 +2023-10-05 21:04:54,969 - Epoch: [48][ 950/ 1236] Overall Loss 0.315742 Objective Loss 0.315742 LR 0.001000 Time 0.020958 +2023-10-05 21:04:55,171 - Epoch: [48][ 960/ 1236] Overall Loss 0.315695 Objective Loss 0.315695 LR 0.001000 Time 0.020950 +2023-10-05 21:04:55,374 - Epoch: [48][ 970/ 1236] Overall Loss 0.316026 Objective Loss 0.316026 LR 0.001000 Time 0.020943 +2023-10-05 21:04:55,576 - Epoch: [48][ 980/ 1236] Overall Loss 0.316542 Objective Loss 0.316542 LR 0.001000 Time 0.020935 +2023-10-05 21:04:55,779 - Epoch: [48][ 990/ 1236] Overall Loss 0.316599 Objective Loss 0.316599 LR 0.001000 Time 0.020929 +2023-10-05 21:04:55,981 - Epoch: [48][ 1000/ 1236] Overall Loss 0.317081 Objective Loss 0.317081 LR 0.001000 Time 0.020921 +2023-10-05 21:04:56,184 - Epoch: [48][ 1010/ 1236] Overall Loss 0.317309 Objective Loss 0.317309 LR 0.001000 Time 0.020915 +2023-10-05 21:04:56,386 - Epoch: [48][ 1020/ 1236] Overall Loss 0.317471 Objective Loss 0.317471 LR 0.001000 Time 0.020907 +2023-10-05 21:04:56,589 - Epoch: [48][ 1030/ 1236] Overall Loss 0.317713 Objective Loss 0.317713 LR 0.001000 Time 0.020901 +2023-10-05 21:04:56,791 - Epoch: [48][ 1040/ 1236] Overall Loss 0.317727 Objective Loss 0.317727 LR 0.001000 Time 0.020894 +2023-10-05 21:04:56,994 - Epoch: [48][ 1050/ 1236] Overall Loss 0.317747 Objective Loss 0.317747 LR 0.001000 Time 0.020888 +2023-10-05 21:04:57,196 - Epoch: [48][ 1060/ 1236] Overall Loss 0.317598 Objective Loss 0.317598 LR 0.001000 Time 0.020882 +2023-10-05 21:04:57,399 - Epoch: [48][ 1070/ 1236] Overall Loss 0.317789 Objective Loss 0.317789 LR 0.001000 Time 0.020876 +2023-10-05 21:04:57,601 - Epoch: [48][ 1080/ 1236] Overall Loss 0.317545 Objective Loss 0.317545 LR 0.001000 Time 0.020869 +2023-10-05 21:04:57,808 - Epoch: [48][ 1090/ 1236] Overall Loss 0.317479 Objective Loss 0.317479 LR 0.001000 Time 0.020862 +2023-10-05 21:04:58,010 - Epoch: [48][ 1100/ 1236] Overall Loss 0.317417 Objective Loss 0.317417 LR 0.001000 Time 0.020856 +2023-10-05 21:04:58,213 - Epoch: [48][ 1110/ 1236] Overall Loss 0.317568 Objective Loss 0.317568 LR 0.001000 Time 0.020850 +2023-10-05 21:04:58,415 - Epoch: [48][ 1120/ 1236] Overall Loss 0.317525 Objective Loss 0.317525 LR 0.001000 Time 0.020845 +2023-10-05 21:04:58,617 - Epoch: [48][ 1130/ 1236] Overall Loss 0.317426 Objective Loss 0.317426 LR 0.001000 Time 0.020838 +2023-10-05 21:04:58,819 - Epoch: [48][ 1140/ 1236] Overall Loss 0.317227 Objective Loss 0.317227 LR 0.001000 Time 0.020833 +2023-10-05 21:04:59,022 - Epoch: [48][ 1150/ 1236] Overall Loss 0.317244 Objective Loss 0.317244 LR 0.001000 Time 0.020828 +2023-10-05 21:04:59,224 - Epoch: [48][ 1160/ 1236] Overall Loss 0.317527 Objective Loss 0.317527 LR 0.001000 Time 0.020822 +2023-10-05 21:04:59,427 - Epoch: [48][ 1170/ 1236] Overall Loss 0.317483 Objective Loss 0.317483 LR 0.001000 Time 0.020817 +2023-10-05 21:04:59,629 - Epoch: [48][ 1180/ 1236] Overall Loss 0.317598 Objective Loss 0.317598 LR 0.001000 Time 0.020812 +2023-10-05 21:04:59,832 - Epoch: [48][ 1190/ 1236] Overall Loss 0.317915 Objective Loss 0.317915 LR 0.001000 Time 0.020807 +2023-10-05 21:05:00,035 - Epoch: [48][ 1200/ 1236] Overall Loss 0.317982 Objective Loss 0.317982 LR 0.001000 Time 0.020802 +2023-10-05 21:05:00,237 - Epoch: [48][ 1210/ 1236] Overall Loss 0.317974 Objective Loss 0.317974 LR 0.001000 Time 0.020797 +2023-10-05 21:05:00,439 - Epoch: [48][ 1220/ 1236] Overall Loss 0.318169 Objective Loss 0.318169 LR 0.001000 Time 0.020793 +2023-10-05 21:05:00,697 - Epoch: [48][ 1230/ 1236] Overall Loss 0.318469 Objective Loss 0.318469 LR 0.001000 Time 0.020833 +2023-10-05 21:05:00,815 - Epoch: [48][ 1236/ 1236] Overall Loss 0.318647 Objective Loss 0.318647 Top1 81.466395 Top5 97.963340 LR 0.001000 Time 0.020827 +2023-10-05 21:05:00,943 - --- validate (epoch=48)----------- +2023-10-05 21:05:00,943 - 29943 samples (256 per mini-batch) +2023-10-05 21:05:01,395 - Epoch: [48][ 10/ 117] Loss 0.368039 Top1 82.460938 Top5 97.773438 +2023-10-05 21:05:01,542 - Epoch: [48][ 20/ 117] Loss 0.381658 Top1 81.777344 Top5 97.695312 +2023-10-05 21:05:01,691 - Epoch: [48][ 30/ 117] Loss 0.376839 Top1 81.979167 Top5 97.799479 +2023-10-05 21:05:01,839 - Epoch: [48][ 40/ 117] Loss 0.376720 Top1 82.080078 Top5 97.783203 +2023-10-05 21:05:01,987 - Epoch: [48][ 50/ 117] Loss 0.373486 Top1 81.960938 Top5 97.734375 +2023-10-05 21:05:02,135 - Epoch: [48][ 60/ 117] Loss 0.372577 Top1 82.063802 Top5 97.727865 +2023-10-05 21:05:02,283 - Epoch: [48][ 70/ 117] Loss 0.370298 Top1 82.047991 Top5 97.745536 +2023-10-05 21:05:02,429 - Epoch: [48][ 80/ 117] Loss 0.372115 Top1 81.914062 Top5 97.768555 +2023-10-05 21:05:02,575 - Epoch: [48][ 90/ 117] Loss 0.374922 Top1 81.844618 Top5 97.712674 +2023-10-05 21:05:02,721 - Epoch: [48][ 100/ 117] Loss 0.374339 Top1 81.882812 Top5 97.671875 +2023-10-05 21:05:02,876 - Epoch: [48][ 110/ 117] Loss 0.370509 Top1 81.992188 Top5 97.656250 +2023-10-05 21:05:02,961 - Epoch: [48][ 117/ 117] Loss 0.371769 Top1 82.035868 Top5 97.635507 +2023-10-05 21:05:03,096 - ==> Top1: 82.036 Top5: 97.636 Loss: 0.372 + +2023-10-05 21:05:03,097 - ==> Confusion: +[[ 930 2 7 0 7 2 0 1 2 65 0 0 0 5 4 3 7 2 1 0 12] + [ 0 1017 3 0 19 17 1 24 3 1 7 2 0 0 1 2 8 2 12 0 12] + [ 7 1 946 18 4 0 28 8 0 1 4 3 7 1 2 5 1 0 6 3 11] + [ 2 3 19 959 2 3 4 1 2 0 7 1 5 3 28 3 2 10 18 2 15] + [ 30 3 2 0 967 4 1 1 0 7 0 2 0 3 5 6 12 2 0 0 5] + [ 7 47 1 6 12 920 5 32 4 2 3 10 4 18 9 1 2 1 4 5 23] + [ 1 7 29 0 1 0 1109 7 0 0 2 3 1 1 1 11 2 2 1 7 6] + [ 6 17 21 0 2 34 7 1035 3 4 6 11 4 0 1 3 1 1 39 10 13] + [ 21 1 0 1 0 2 0 1 951 52 15 1 3 15 12 1 1 3 3 1 5] + [ 103 0 1 0 8 2 2 0 17 933 2 1 1 22 8 2 1 3 0 1 12] + [ 2 3 15 13 1 0 14 2 23 0 939 3 0 13 4 0 2 3 10 0 6] + [ 1 0 2 0 2 8 0 1 2 1 0 937 40 2 0 1 1 16 0 17 4] + [ 2 0 2 4 0 0 0 1 1 0 1 39 967 1 5 10 1 17 3 2 12] + [ 2 0 1 0 6 6 2 0 17 21 16 9 3 1007 4 2 3 1 0 1 18] + [ 14 0 3 18 12 0 0 0 29 10 4 1 2 4 966 1 2 5 9 0 21] + [ 1 3 1 4 4 0 0 0 0 0 0 14 10 1 0 1056 17 12 0 5 6] + [ 0 11 1 1 5 4 1 1 1 0 0 2 4 1 1 14 1095 1 1 6 11] + [ 0 0 0 1 0 0 1 0 0 0 0 6 30 0 4 5 0 982 2 2 5] + [ 2 4 9 17 1 0 2 31 11 0 4 2 4 0 15 1 1 0 950 0 14] + [ 0 3 8 0 4 5 11 10 0 1 0 13 5 0 0 10 14 1 1 1048 18] + [ 147 160 172 102 149 103 61 107 100 100 169 131 394 265 135 76 198 85 177 224 4850]] + +2023-10-05 21:05:03,098 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:05:03,098 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:05:03,104 - + +2023-10-05 21:05:03,104 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:05:04,090 - Epoch: [49][ 10/ 1236] Overall Loss 0.317868 Objective Loss 0.317868 LR 0.001000 Time 0.098539 +2023-10-05 21:05:04,292 - Epoch: [49][ 20/ 1236] Overall Loss 0.319518 Objective Loss 0.319518 LR 0.001000 Time 0.059329 +2023-10-05 21:05:04,492 - Epoch: [49][ 30/ 1236] Overall Loss 0.309728 Objective Loss 0.309728 LR 0.001000 Time 0.046212 +2023-10-05 21:05:04,693 - Epoch: [49][ 40/ 1236] Overall Loss 0.306042 Objective Loss 0.306042 LR 0.001000 Time 0.039695 +2023-10-05 21:05:04,894 - Epoch: [49][ 50/ 1236] Overall Loss 0.306038 Objective Loss 0.306038 LR 0.001000 Time 0.035753 +2023-10-05 21:05:05,095 - Epoch: [49][ 60/ 1236] Overall Loss 0.309294 Objective Loss 0.309294 LR 0.001000 Time 0.033151 +2023-10-05 21:05:05,296 - Epoch: [49][ 70/ 1236] Overall Loss 0.307173 Objective Loss 0.307173 LR 0.001000 Time 0.031269 +2023-10-05 21:05:05,498 - Epoch: [49][ 80/ 1236] Overall Loss 0.305774 Objective Loss 0.305774 LR 0.001000 Time 0.029880 +2023-10-05 21:05:05,697 - Epoch: [49][ 90/ 1236] Overall Loss 0.302982 Objective Loss 0.302982 LR 0.001000 Time 0.028773 +2023-10-05 21:05:05,897 - Epoch: [49][ 100/ 1236] Overall Loss 0.305122 Objective Loss 0.305122 LR 0.001000 Time 0.027894 +2023-10-05 21:05:06,097 - Epoch: [49][ 110/ 1236] Overall Loss 0.305718 Objective Loss 0.305718 LR 0.001000 Time 0.027168 +2023-10-05 21:05:06,297 - Epoch: [49][ 120/ 1236] Overall Loss 0.304118 Objective Loss 0.304118 LR 0.001000 Time 0.026573 +2023-10-05 21:05:06,497 - Epoch: [49][ 130/ 1236] Overall Loss 0.303511 Objective Loss 0.303511 LR 0.001000 Time 0.026064 +2023-10-05 21:05:06,697 - Epoch: [49][ 140/ 1236] Overall Loss 0.301965 Objective Loss 0.301965 LR 0.001000 Time 0.025631 +2023-10-05 21:05:06,897 - Epoch: [49][ 150/ 1236] Overall Loss 0.304941 Objective Loss 0.304941 LR 0.001000 Time 0.025251 +2023-10-05 21:05:07,097 - Epoch: [49][ 160/ 1236] Overall Loss 0.306566 Objective Loss 0.306566 LR 0.001000 Time 0.024922 +2023-10-05 21:05:07,297 - Epoch: [49][ 170/ 1236] Overall Loss 0.305795 Objective Loss 0.305795 LR 0.001000 Time 0.024629 +2023-10-05 21:05:07,497 - Epoch: [49][ 180/ 1236] Overall Loss 0.305587 Objective Loss 0.305587 LR 0.001000 Time 0.024372 +2023-10-05 21:05:07,697 - Epoch: [49][ 190/ 1236] Overall Loss 0.306309 Objective Loss 0.306309 LR 0.001000 Time 0.024139 +2023-10-05 21:05:07,897 - Epoch: [49][ 200/ 1236] Overall Loss 0.308186 Objective Loss 0.308186 LR 0.001000 Time 0.023931 +2023-10-05 21:05:08,097 - Epoch: [49][ 210/ 1236] Overall Loss 0.307769 Objective Loss 0.307769 LR 0.001000 Time 0.023741 +2023-10-05 21:05:08,297 - Epoch: [49][ 220/ 1236] Overall Loss 0.306450 Objective Loss 0.306450 LR 0.001000 Time 0.023571 +2023-10-05 21:05:08,497 - Epoch: [49][ 230/ 1236] Overall Loss 0.305933 Objective Loss 0.305933 LR 0.001000 Time 0.023413 +2023-10-05 21:05:08,697 - Epoch: [49][ 240/ 1236] Overall Loss 0.306737 Objective Loss 0.306737 LR 0.001000 Time 0.023271 +2023-10-05 21:05:08,897 - Epoch: [49][ 250/ 1236] Overall Loss 0.307579 Objective Loss 0.307579 LR 0.001000 Time 0.023137 +2023-10-05 21:05:09,097 - Epoch: [49][ 260/ 1236] Overall Loss 0.307944 Objective Loss 0.307944 LR 0.001000 Time 0.023017 +2023-10-05 21:05:09,297 - Epoch: [49][ 270/ 1236] Overall Loss 0.308404 Objective Loss 0.308404 LR 0.001000 Time 0.022903 +2023-10-05 21:05:09,497 - Epoch: [49][ 280/ 1236] Overall Loss 0.307263 Objective Loss 0.307263 LR 0.001000 Time 0.022799 +2023-10-05 21:05:09,697 - Epoch: [49][ 290/ 1236] Overall Loss 0.306506 Objective Loss 0.306506 LR 0.001000 Time 0.022700 +2023-10-05 21:05:09,897 - Epoch: [49][ 300/ 1236] Overall Loss 0.306518 Objective Loss 0.306518 LR 0.001000 Time 0.022610 +2023-10-05 21:05:10,097 - Epoch: [49][ 310/ 1236] Overall Loss 0.307337 Objective Loss 0.307337 LR 0.001000 Time 0.022524 +2023-10-05 21:05:10,298 - Epoch: [49][ 320/ 1236] Overall Loss 0.308202 Objective Loss 0.308202 LR 0.001000 Time 0.022445 +2023-10-05 21:05:10,498 - Epoch: [49][ 330/ 1236] Overall Loss 0.308728 Objective Loss 0.308728 LR 0.001000 Time 0.022372 +2023-10-05 21:05:10,698 - Epoch: [49][ 340/ 1236] Overall Loss 0.308060 Objective Loss 0.308060 LR 0.001000 Time 0.022302 +2023-10-05 21:05:10,898 - Epoch: [49][ 350/ 1236] Overall Loss 0.306787 Objective Loss 0.306787 LR 0.001000 Time 0.022235 +2023-10-05 21:05:11,098 - Epoch: [49][ 360/ 1236] Overall Loss 0.305649 Objective Loss 0.305649 LR 0.001000 Time 0.022172 +2023-10-05 21:05:11,298 - Epoch: [49][ 370/ 1236] Overall Loss 0.305746 Objective Loss 0.305746 LR 0.001000 Time 0.022112 +2023-10-05 21:05:11,500 - Epoch: [49][ 380/ 1236] Overall Loss 0.305758 Objective Loss 0.305758 LR 0.001000 Time 0.022061 +2023-10-05 21:05:11,701 - Epoch: [49][ 390/ 1236] Overall Loss 0.306168 Objective Loss 0.306168 LR 0.001000 Time 0.022008 +2023-10-05 21:05:11,903 - Epoch: [49][ 400/ 1236] Overall Loss 0.305964 Objective Loss 0.305964 LR 0.001000 Time 0.021962 +2023-10-05 21:05:12,103 - Epoch: [49][ 410/ 1236] Overall Loss 0.306109 Objective Loss 0.306109 LR 0.001000 Time 0.021914 +2023-10-05 21:05:12,305 - Epoch: [49][ 420/ 1236] Overall Loss 0.306550 Objective Loss 0.306550 LR 0.001000 Time 0.021872 +2023-10-05 21:05:12,505 - Epoch: [49][ 430/ 1236] Overall Loss 0.306732 Objective Loss 0.306732 LR 0.001000 Time 0.021829 +2023-10-05 21:05:12,708 - Epoch: [49][ 440/ 1236] Overall Loss 0.306591 Objective Loss 0.306591 LR 0.001000 Time 0.021792 +2023-10-05 21:05:12,908 - Epoch: [49][ 450/ 1236] Overall Loss 0.306800 Objective Loss 0.306800 LR 0.001000 Time 0.021753 +2023-10-05 21:05:13,110 - Epoch: [49][ 460/ 1236] Overall Loss 0.306838 Objective Loss 0.306838 LR 0.001000 Time 0.021718 +2023-10-05 21:05:13,311 - Epoch: [49][ 470/ 1236] Overall Loss 0.307026 Objective Loss 0.307026 LR 0.001000 Time 0.021682 +2023-10-05 21:05:13,513 - Epoch: [49][ 480/ 1236] Overall Loss 0.306823 Objective Loss 0.306823 LR 0.001000 Time 0.021651 +2023-10-05 21:05:13,713 - Epoch: [49][ 490/ 1236] Overall Loss 0.306997 Objective Loss 0.306997 LR 0.001000 Time 0.021617 +2023-10-05 21:05:13,915 - Epoch: [49][ 500/ 1236] Overall Loss 0.306868 Objective Loss 0.306868 LR 0.001000 Time 0.021588 +2023-10-05 21:05:14,116 - Epoch: [49][ 510/ 1236] Overall Loss 0.307395 Objective Loss 0.307395 LR 0.001000 Time 0.021557 +2023-10-05 21:05:14,318 - Epoch: [49][ 520/ 1236] Overall Loss 0.308135 Objective Loss 0.308135 LR 0.001000 Time 0.021530 +2023-10-05 21:05:14,518 - Epoch: [49][ 530/ 1236] Overall Loss 0.308969 Objective Loss 0.308969 LR 0.001000 Time 0.021501 +2023-10-05 21:05:14,720 - Epoch: [49][ 540/ 1236] Overall Loss 0.308968 Objective Loss 0.308968 LR 0.001000 Time 0.021477 +2023-10-05 21:05:14,921 - Epoch: [49][ 550/ 1236] Overall Loss 0.308951 Objective Loss 0.308951 LR 0.001000 Time 0.021450 +2023-10-05 21:05:15,123 - Epoch: [49][ 560/ 1236] Overall Loss 0.309465 Objective Loss 0.309465 LR 0.001000 Time 0.021427 +2023-10-05 21:05:15,323 - Epoch: [49][ 570/ 1236] Overall Loss 0.309497 Objective Loss 0.309497 LR 0.001000 Time 0.021402 +2023-10-05 21:05:15,525 - Epoch: [49][ 580/ 1236] Overall Loss 0.309870 Objective Loss 0.309870 LR 0.001000 Time 0.021381 +2023-10-05 21:05:15,726 - Epoch: [49][ 590/ 1236] Overall Loss 0.310244 Objective Loss 0.310244 LR 0.001000 Time 0.021358 +2023-10-05 21:05:15,928 - Epoch: [49][ 600/ 1236] Overall Loss 0.310364 Objective Loss 0.310364 LR 0.001000 Time 0.021338 +2023-10-05 21:05:16,128 - Epoch: [49][ 610/ 1236] Overall Loss 0.309950 Objective Loss 0.309950 LR 0.001000 Time 0.021317 +2023-10-05 21:05:16,330 - Epoch: [49][ 620/ 1236] Overall Loss 0.310670 Objective Loss 0.310670 LR 0.001000 Time 0.021298 +2023-10-05 21:05:16,531 - Epoch: [49][ 630/ 1236] Overall Loss 0.310703 Objective Loss 0.310703 LR 0.001000 Time 0.021278 +2023-10-05 21:05:16,733 - Epoch: [49][ 640/ 1236] Overall Loss 0.311543 Objective Loss 0.311543 LR 0.001000 Time 0.021260 +2023-10-05 21:05:16,933 - Epoch: [49][ 650/ 1236] Overall Loss 0.311811 Objective Loss 0.311811 LR 0.001000 Time 0.021241 +2023-10-05 21:05:17,135 - Epoch: [49][ 660/ 1236] Overall Loss 0.311694 Objective Loss 0.311694 LR 0.001000 Time 0.021225 +2023-10-05 21:05:17,336 - Epoch: [49][ 670/ 1236] Overall Loss 0.312232 Objective Loss 0.312232 LR 0.001000 Time 0.021207 +2023-10-05 21:05:17,538 - Epoch: [49][ 680/ 1236] Overall Loss 0.312280 Objective Loss 0.312280 LR 0.001000 Time 0.021192 +2023-10-05 21:05:17,738 - Epoch: [49][ 690/ 1236] Overall Loss 0.312516 Objective Loss 0.312516 LR 0.001000 Time 0.021174 +2023-10-05 21:05:17,940 - Epoch: [49][ 700/ 1236] Overall Loss 0.312515 Objective Loss 0.312515 LR 0.001000 Time 0.021160 +2023-10-05 21:05:18,141 - Epoch: [49][ 710/ 1236] Overall Loss 0.312237 Objective Loss 0.312237 LR 0.001000 Time 0.021144 +2023-10-05 21:05:18,343 - Epoch: [49][ 720/ 1236] Overall Loss 0.312307 Objective Loss 0.312307 LR 0.001000 Time 0.021130 +2023-10-05 21:05:18,543 - Epoch: [49][ 730/ 1236] Overall Loss 0.312156 Objective Loss 0.312156 LR 0.001000 Time 0.021115 +2023-10-05 21:05:18,745 - Epoch: [49][ 740/ 1236] Overall Loss 0.312109 Objective Loss 0.312109 LR 0.001000 Time 0.021102 +2023-10-05 21:05:18,946 - Epoch: [49][ 750/ 1236] Overall Loss 0.311924 Objective Loss 0.311924 LR 0.001000 Time 0.021088 +2023-10-05 21:05:19,148 - Epoch: [49][ 760/ 1236] Overall Loss 0.312070 Objective Loss 0.312070 LR 0.001000 Time 0.021076 +2023-10-05 21:05:19,348 - Epoch: [49][ 770/ 1236] Overall Loss 0.312259 Objective Loss 0.312259 LR 0.001000 Time 0.021062 +2023-10-05 21:05:19,550 - Epoch: [49][ 780/ 1236] Overall Loss 0.312609 Objective Loss 0.312609 LR 0.001000 Time 0.021050 +2023-10-05 21:05:19,751 - Epoch: [49][ 790/ 1236] Overall Loss 0.312952 Objective Loss 0.312952 LR 0.001000 Time 0.021037 +2023-10-05 21:05:19,953 - Epoch: [49][ 800/ 1236] Overall Loss 0.313178 Objective Loss 0.313178 LR 0.001000 Time 0.021026 +2023-10-05 21:05:20,153 - Epoch: [49][ 810/ 1236] Overall Loss 0.313368 Objective Loss 0.313368 LR 0.001000 Time 0.021014 +2023-10-05 21:05:20,355 - Epoch: [49][ 820/ 1236] Overall Loss 0.313651 Objective Loss 0.313651 LR 0.001000 Time 0.021004 +2023-10-05 21:05:20,556 - Epoch: [49][ 830/ 1236] Overall Loss 0.313922 Objective Loss 0.313922 LR 0.001000 Time 0.020992 +2023-10-05 21:05:20,758 - Epoch: [49][ 840/ 1236] Overall Loss 0.313778 Objective Loss 0.313778 LR 0.001000 Time 0.020982 +2023-10-05 21:05:20,959 - Epoch: [49][ 850/ 1236] Overall Loss 0.313602 Objective Loss 0.313602 LR 0.001000 Time 0.020971 +2023-10-05 21:05:21,160 - Epoch: [49][ 860/ 1236] Overall Loss 0.313641 Objective Loss 0.313641 LR 0.001000 Time 0.020961 +2023-10-05 21:05:21,361 - Epoch: [49][ 870/ 1236] Overall Loss 0.313635 Objective Loss 0.313635 LR 0.001000 Time 0.020950 +2023-10-05 21:05:21,563 - Epoch: [49][ 880/ 1236] Overall Loss 0.313498 Objective Loss 0.313498 LR 0.001000 Time 0.020941 +2023-10-05 21:05:21,763 - Epoch: [49][ 890/ 1236] Overall Loss 0.313849 Objective Loss 0.313849 LR 0.001000 Time 0.020931 +2023-10-05 21:05:21,966 - Epoch: [49][ 900/ 1236] Overall Loss 0.314217 Objective Loss 0.314217 LR 0.001000 Time 0.020922 +2023-10-05 21:05:22,166 - Epoch: [49][ 910/ 1236] Overall Loss 0.314492 Objective Loss 0.314492 LR 0.001000 Time 0.020913 +2023-10-05 21:05:22,368 - Epoch: [49][ 920/ 1236] Overall Loss 0.314927 Objective Loss 0.314927 LR 0.001000 Time 0.020905 +2023-10-05 21:05:22,569 - Epoch: [49][ 930/ 1236] Overall Loss 0.315298 Objective Loss 0.315298 LR 0.001000 Time 0.020895 +2023-10-05 21:05:22,771 - Epoch: [49][ 940/ 1236] Overall Loss 0.315546 Objective Loss 0.315546 LR 0.001000 Time 0.020888 +2023-10-05 21:05:22,971 - Epoch: [49][ 950/ 1236] Overall Loss 0.315898 Objective Loss 0.315898 LR 0.001000 Time 0.020878 +2023-10-05 21:05:23,174 - Epoch: [49][ 960/ 1236] Overall Loss 0.316062 Objective Loss 0.316062 LR 0.001000 Time 0.020871 +2023-10-05 21:05:23,374 - Epoch: [49][ 970/ 1236] Overall Loss 0.315987 Objective Loss 0.315987 LR 0.001000 Time 0.020862 +2023-10-05 21:05:23,576 - Epoch: [49][ 980/ 1236] Overall Loss 0.315970 Objective Loss 0.315970 LR 0.001000 Time 0.020855 +2023-10-05 21:05:23,776 - Epoch: [49][ 990/ 1236] Overall Loss 0.315766 Objective Loss 0.315766 LR 0.001000 Time 0.020846 +2023-10-05 21:05:23,978 - Epoch: [49][ 1000/ 1236] Overall Loss 0.315501 Objective Loss 0.315501 LR 0.001000 Time 0.020840 +2023-10-05 21:05:24,179 - Epoch: [49][ 1010/ 1236] Overall Loss 0.315527 Objective Loss 0.315527 LR 0.001000 Time 0.020832 +2023-10-05 21:05:24,381 - Epoch: [49][ 1020/ 1236] Overall Loss 0.315557 Objective Loss 0.315557 LR 0.001000 Time 0.020825 +2023-10-05 21:05:24,582 - Epoch: [49][ 1030/ 1236] Overall Loss 0.315627 Objective Loss 0.315627 LR 0.001000 Time 0.020818 +2023-10-05 21:05:24,784 - Epoch: [49][ 1040/ 1236] Overall Loss 0.315636 Objective Loss 0.315636 LR 0.001000 Time 0.020811 +2023-10-05 21:05:24,984 - Epoch: [49][ 1050/ 1236] Overall Loss 0.315607 Objective Loss 0.315607 LR 0.001000 Time 0.020804 +2023-10-05 21:05:25,186 - Epoch: [49][ 1060/ 1236] Overall Loss 0.315724 Objective Loss 0.315724 LR 0.001000 Time 0.020798 +2023-10-05 21:05:25,387 - Epoch: [49][ 1070/ 1236] Overall Loss 0.315749 Objective Loss 0.315749 LR 0.001000 Time 0.020790 +2023-10-05 21:05:25,589 - Epoch: [49][ 1080/ 1236] Overall Loss 0.316000 Objective Loss 0.316000 LR 0.001000 Time 0.020785 +2023-10-05 21:05:25,790 - Epoch: [49][ 1090/ 1236] Overall Loss 0.316178 Objective Loss 0.316178 LR 0.001000 Time 0.020778 +2023-10-05 21:05:26,003 - Epoch: [49][ 1100/ 1236] Overall Loss 0.316501 Objective Loss 0.316501 LR 0.001000 Time 0.020783 +2023-10-05 21:05:26,212 - Epoch: [49][ 1110/ 1236] Overall Loss 0.316476 Objective Loss 0.316476 LR 0.001000 Time 0.020783 +2023-10-05 21:05:26,424 - Epoch: [49][ 1120/ 1236] Overall Loss 0.316848 Objective Loss 0.316848 LR 0.001000 Time 0.020787 +2023-10-05 21:05:26,633 - Epoch: [49][ 1130/ 1236] Overall Loss 0.317063 Objective Loss 0.317063 LR 0.001000 Time 0.020787 +2023-10-05 21:05:26,846 - Epoch: [49][ 1140/ 1236] Overall Loss 0.316702 Objective Loss 0.316702 LR 0.001000 Time 0.020791 +2023-10-05 21:05:27,054 - Epoch: [49][ 1150/ 1236] Overall Loss 0.316668 Objective Loss 0.316668 LR 0.001000 Time 0.020791 +2023-10-05 21:05:27,267 - Epoch: [49][ 1160/ 1236] Overall Loss 0.316995 Objective Loss 0.316995 LR 0.001000 Time 0.020795 +2023-10-05 21:05:27,475 - Epoch: [49][ 1170/ 1236] Overall Loss 0.317277 Objective Loss 0.317277 LR 0.001000 Time 0.020795 +2023-10-05 21:05:27,688 - Epoch: [49][ 1180/ 1236] Overall Loss 0.317382 Objective Loss 0.317382 LR 0.001000 Time 0.020799 +2023-10-05 21:05:27,896 - Epoch: [49][ 1190/ 1236] Overall Loss 0.317304 Objective Loss 0.317304 LR 0.001000 Time 0.020799 +2023-10-05 21:05:28,109 - Epoch: [49][ 1200/ 1236] Overall Loss 0.317420 Objective Loss 0.317420 LR 0.001000 Time 0.020803 +2023-10-05 21:05:28,317 - Epoch: [49][ 1210/ 1236] Overall Loss 0.317410 Objective Loss 0.317410 LR 0.001000 Time 0.020803 +2023-10-05 21:05:28,530 - Epoch: [49][ 1220/ 1236] Overall Loss 0.317263 Objective Loss 0.317263 LR 0.001000 Time 0.020807 +2023-10-05 21:05:28,791 - Epoch: [49][ 1230/ 1236] Overall Loss 0.317353 Objective Loss 0.317353 LR 0.001000 Time 0.020849 +2023-10-05 21:05:28,909 - Epoch: [49][ 1236/ 1236] Overall Loss 0.317556 Objective Loss 0.317556 Top1 82.688391 Top5 96.334012 LR 0.001000 Time 0.020843 +2023-10-05 21:05:29,047 - --- validate (epoch=49)----------- +2023-10-05 21:05:29,047 - 29943 samples (256 per mini-batch) +2023-10-05 21:05:29,496 - Epoch: [49][ 10/ 117] Loss 0.369935 Top1 80.117188 Top5 96.992188 +2023-10-05 21:05:29,646 - Epoch: [49][ 20/ 117] Loss 0.367473 Top1 80.429688 Top5 96.992188 +2023-10-05 21:05:29,795 - Epoch: [49][ 30/ 117] Loss 0.373102 Top1 80.390625 Top5 97.070312 +2023-10-05 21:05:29,944 - Epoch: [49][ 40/ 117] Loss 0.369588 Top1 80.615234 Top5 97.246094 +2023-10-05 21:05:30,093 - Epoch: [49][ 50/ 117] Loss 0.366378 Top1 80.671875 Top5 97.335938 +2023-10-05 21:05:30,242 - Epoch: [49][ 60/ 117] Loss 0.370823 Top1 80.618490 Top5 97.298177 +2023-10-05 21:05:30,390 - Epoch: [49][ 70/ 117] Loss 0.372227 Top1 80.479911 Top5 97.315848 +2023-10-05 21:05:30,539 - Epoch: [49][ 80/ 117] Loss 0.370952 Top1 80.498047 Top5 97.309570 +2023-10-05 21:05:30,687 - Epoch: [49][ 90/ 117] Loss 0.372532 Top1 80.546875 Top5 97.300347 +2023-10-05 21:05:30,833 - Epoch: [49][ 100/ 117] Loss 0.368021 Top1 80.578125 Top5 97.289062 +2023-10-05 21:05:30,985 - Epoch: [49][ 110/ 117] Loss 0.366444 Top1 80.529119 Top5 97.286932 +2023-10-05 21:05:31,071 - Epoch: [49][ 117/ 117] Loss 0.364970 Top1 80.599806 Top5 97.294860 +2023-10-05 21:05:31,203 - ==> Top1: 80.600 Top5: 97.295 Loss: 0.365 + +2023-10-05 21:05:31,204 - ==> Confusion: +[[ 945 1 4 1 7 3 0 0 1 61 0 0 2 3 4 3 6 2 1 0 6] + [ 1 1042 0 1 12 31 2 19 1 0 6 1 0 1 0 4 4 0 3 0 3] + [ 9 2 946 17 3 2 30 9 0 0 2 3 5 2 1 6 3 0 6 5 5] + [ 4 1 25 962 0 5 0 1 3 0 7 0 0 4 25 4 2 6 28 3 9] + [ 34 6 1 3 961 8 1 1 0 9 1 2 0 1 6 4 9 2 0 0 1] + [ 6 29 0 2 5 991 3 21 0 2 3 8 2 14 3 1 4 1 4 6 11] + [ 1 10 34 0 1 1 1096 14 0 0 2 1 3 1 0 10 1 1 1 10 4] + [ 4 30 21 0 3 41 5 1027 1 2 6 6 1 1 2 1 1 0 42 14 10] + [ 26 2 0 1 1 11 2 0 920 64 16 1 0 20 18 3 0 0 2 2 0] + [ 122 1 4 1 10 8 1 0 19 912 2 1 0 24 1 3 3 1 0 1 5] + [ 5 7 23 4 1 1 8 4 15 0 943 1 0 18 4 1 1 1 6 1 9] + [ 1 1 3 1 4 18 0 2 1 0 0 948 20 3 0 2 2 13 0 13 3] + [ 1 1 6 7 2 1 1 3 3 0 1 50 937 3 2 13 4 8 4 8 13] + [ 2 1 1 1 5 15 0 2 7 16 6 3 2 1034 2 2 1 1 0 2 16] + [ 21 1 3 21 7 0 0 0 23 11 3 1 0 1 985 0 1 2 12 0 9] + [ 3 1 5 1 4 1 4 0 0 0 0 14 7 2 0 1051 25 10 1 3 2] + [ 2 13 1 1 9 4 0 1 1 0 0 3 1 1 4 14 1096 0 0 3 7] + [ 1 0 1 3 0 0 0 0 1 0 0 20 39 3 3 18 1 942 1 1 4] + [ 5 9 7 17 2 1 0 34 2 1 2 2 2 0 11 0 1 0 961 1 10] + [ 0 4 6 0 2 11 5 9 1 1 2 9 5 2 0 8 10 1 1 1069 6] + [ 233 229 225 92 163 212 82 112 86 101 158 132 352 306 129 115 338 55 156 263 4366]] + +2023-10-05 21:05:31,205 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:05:31,205 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:05:31,211 - + +2023-10-05 21:05:31,211 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:05:32,201 - Epoch: [50][ 10/ 1236] Overall Loss 0.294107 Objective Loss 0.294107 LR 0.001000 Time 0.098922 +2023-10-05 21:05:32,408 - Epoch: [50][ 20/ 1236] Overall Loss 0.299450 Objective Loss 0.299450 LR 0.001000 Time 0.059825 +2023-10-05 21:05:32,608 - Epoch: [50][ 30/ 1236] Overall Loss 0.302650 Objective Loss 0.302650 LR 0.001000 Time 0.046524 +2023-10-05 21:05:32,809 - Epoch: [50][ 40/ 1236] Overall Loss 0.315965 Objective Loss 0.315965 LR 0.001000 Time 0.039916 +2023-10-05 21:05:33,009 - Epoch: [50][ 50/ 1236] Overall Loss 0.319056 Objective Loss 0.319056 LR 0.001000 Time 0.035922 +2023-10-05 21:05:33,210 - Epoch: [50][ 60/ 1236] Overall Loss 0.322423 Objective Loss 0.322423 LR 0.001000 Time 0.033278 +2023-10-05 21:05:33,410 - Epoch: [50][ 70/ 1236] Overall Loss 0.321760 Objective Loss 0.321760 LR 0.001000 Time 0.031375 +2023-10-05 21:05:33,611 - Epoch: [50][ 80/ 1236] Overall Loss 0.319822 Objective Loss 0.319822 LR 0.001000 Time 0.029959 +2023-10-05 21:05:33,810 - Epoch: [50][ 90/ 1236] Overall Loss 0.319642 Objective Loss 0.319642 LR 0.001000 Time 0.028841 +2023-10-05 21:05:34,010 - Epoch: [50][ 100/ 1236] Overall Loss 0.323261 Objective Loss 0.323261 LR 0.001000 Time 0.027955 +2023-10-05 21:05:34,208 - Epoch: [50][ 110/ 1236] Overall Loss 0.321347 Objective Loss 0.321347 LR 0.001000 Time 0.027207 +2023-10-05 21:05:34,407 - Epoch: [50][ 120/ 1236] Overall Loss 0.322866 Objective Loss 0.322866 LR 0.001000 Time 0.026596 +2023-10-05 21:05:34,604 - Epoch: [50][ 130/ 1236] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.026067 +2023-10-05 21:05:34,804 - Epoch: [50][ 140/ 1236] Overall Loss 0.323804 Objective Loss 0.323804 LR 0.001000 Time 0.025626 +2023-10-05 21:05:35,001 - Epoch: [50][ 150/ 1236] Overall Loss 0.322626 Objective Loss 0.322626 LR 0.001000 Time 0.025233 +2023-10-05 21:05:35,200 - Epoch: [50][ 160/ 1236] Overall Loss 0.324015 Objective Loss 0.324015 LR 0.001000 Time 0.024898 +2023-10-05 21:05:35,398 - Epoch: [50][ 170/ 1236] Overall Loss 0.323126 Objective Loss 0.323126 LR 0.001000 Time 0.024594 +2023-10-05 21:05:35,597 - Epoch: [50][ 180/ 1236] Overall Loss 0.321886 Objective Loss 0.321886 LR 0.001000 Time 0.024333 +2023-10-05 21:05:35,795 - Epoch: [50][ 190/ 1236] Overall Loss 0.322461 Objective Loss 0.322461 LR 0.001000 Time 0.024090 +2023-10-05 21:05:35,994 - Epoch: [50][ 200/ 1236] Overall Loss 0.319957 Objective Loss 0.319957 LR 0.001000 Time 0.023880 +2023-10-05 21:05:36,192 - Epoch: [50][ 210/ 1236] Overall Loss 0.321667 Objective Loss 0.321667 LR 0.001000 Time 0.023682 +2023-10-05 21:05:36,391 - Epoch: [50][ 220/ 1236] Overall Loss 0.321433 Objective Loss 0.321433 LR 0.001000 Time 0.023510 +2023-10-05 21:05:36,588 - Epoch: [50][ 230/ 1236] Overall Loss 0.321119 Objective Loss 0.321119 LR 0.001000 Time 0.023345 +2023-10-05 21:05:36,787 - Epoch: [50][ 240/ 1236] Overall Loss 0.320143 Objective Loss 0.320143 LR 0.001000 Time 0.023201 +2023-10-05 21:05:36,984 - Epoch: [50][ 250/ 1236] Overall Loss 0.318765 Objective Loss 0.318765 LR 0.001000 Time 0.023056 +2023-10-05 21:05:37,183 - Epoch: [50][ 260/ 1236] Overall Loss 0.318113 Objective Loss 0.318113 LR 0.001000 Time 0.022935 +2023-10-05 21:05:37,381 - Epoch: [50][ 270/ 1236] Overall Loss 0.318241 Objective Loss 0.318241 LR 0.001000 Time 0.022816 +2023-10-05 21:05:37,580 - Epoch: [50][ 280/ 1236] Overall Loss 0.318277 Objective Loss 0.318277 LR 0.001000 Time 0.022712 +2023-10-05 21:05:37,777 - Epoch: [50][ 290/ 1236] Overall Loss 0.318598 Objective Loss 0.318598 LR 0.001000 Time 0.022609 +2023-10-05 21:05:37,977 - Epoch: [50][ 300/ 1236] Overall Loss 0.319025 Objective Loss 0.319025 LR 0.001000 Time 0.022518 +2023-10-05 21:05:38,174 - Epoch: [50][ 310/ 1236] Overall Loss 0.318395 Objective Loss 0.318395 LR 0.001000 Time 0.022428 +2023-10-05 21:05:38,374 - Epoch: [50][ 320/ 1236] Overall Loss 0.317046 Objective Loss 0.317046 LR 0.001000 Time 0.022349 +2023-10-05 21:05:38,571 - Epoch: [50][ 330/ 1236] Overall Loss 0.317815 Objective Loss 0.317815 LR 0.001000 Time 0.022271 +2023-10-05 21:05:38,771 - Epoch: [50][ 340/ 1236] Overall Loss 0.318466 Objective Loss 0.318466 LR 0.001000 Time 0.022201 +2023-10-05 21:05:38,969 - Epoch: [50][ 350/ 1236] Overall Loss 0.318214 Objective Loss 0.318214 LR 0.001000 Time 0.022132 +2023-10-05 21:05:39,168 - Epoch: [50][ 360/ 1236] Overall Loss 0.317125 Objective Loss 0.317125 LR 0.001000 Time 0.022070 +2023-10-05 21:05:39,366 - Epoch: [50][ 370/ 1236] Overall Loss 0.316789 Objective Loss 0.316789 LR 0.001000 Time 0.022008 +2023-10-05 21:05:39,565 - Epoch: [50][ 380/ 1236] Overall Loss 0.317413 Objective Loss 0.317413 LR 0.001000 Time 0.021952 +2023-10-05 21:05:39,763 - Epoch: [50][ 390/ 1236] Overall Loss 0.317824 Objective Loss 0.317824 LR 0.001000 Time 0.021894 +2023-10-05 21:05:39,962 - Epoch: [50][ 400/ 1236] Overall Loss 0.317970 Objective Loss 0.317970 LR 0.001000 Time 0.021844 +2023-10-05 21:05:40,160 - Epoch: [50][ 410/ 1236] Overall Loss 0.317599 Objective Loss 0.317599 LR 0.001000 Time 0.021793 +2023-10-05 21:05:40,359 - Epoch: [50][ 420/ 1236] Overall Loss 0.318353 Objective Loss 0.318353 LR 0.001000 Time 0.021748 +2023-10-05 21:05:40,557 - Epoch: [50][ 430/ 1236] Overall Loss 0.318115 Objective Loss 0.318115 LR 0.001000 Time 0.021701 +2023-10-05 21:05:40,757 - Epoch: [50][ 440/ 1236] Overall Loss 0.318108 Objective Loss 0.318108 LR 0.001000 Time 0.021661 +2023-10-05 21:05:40,954 - Epoch: [50][ 450/ 1236] Overall Loss 0.319060 Objective Loss 0.319060 LR 0.001000 Time 0.021618 +2023-10-05 21:05:41,154 - Epoch: [50][ 460/ 1236] Overall Loss 0.318865 Objective Loss 0.318865 LR 0.001000 Time 0.021581 +2023-10-05 21:05:41,352 - Epoch: [50][ 470/ 1236] Overall Loss 0.319206 Objective Loss 0.319206 LR 0.001000 Time 0.021543 +2023-10-05 21:05:41,551 - Epoch: [50][ 480/ 1236] Overall Loss 0.319645 Objective Loss 0.319645 LR 0.001000 Time 0.021509 +2023-10-05 21:05:41,749 - Epoch: [50][ 490/ 1236] Overall Loss 0.320397 Objective Loss 0.320397 LR 0.001000 Time 0.021473 +2023-10-05 21:05:41,948 - Epoch: [50][ 500/ 1236] Overall Loss 0.320664 Objective Loss 0.320664 LR 0.001000 Time 0.021441 +2023-10-05 21:05:42,146 - Epoch: [50][ 510/ 1236] Overall Loss 0.320241 Objective Loss 0.320241 LR 0.001000 Time 0.021408 +2023-10-05 21:05:42,345 - Epoch: [50][ 520/ 1236] Overall Loss 0.320419 Objective Loss 0.320419 LR 0.001000 Time 0.021379 +2023-10-05 21:05:42,543 - Epoch: [50][ 530/ 1236] Overall Loss 0.320284 Objective Loss 0.320284 LR 0.001000 Time 0.021349 +2023-10-05 21:05:42,743 - Epoch: [50][ 540/ 1236] Overall Loss 0.320019 Objective Loss 0.320019 LR 0.001000 Time 0.021322 +2023-10-05 21:05:42,941 - Epoch: [50][ 550/ 1236] Overall Loss 0.320142 Objective Loss 0.320142 LR 0.001000 Time 0.021293 +2023-10-05 21:05:43,139 - Epoch: [50][ 560/ 1236] Overall Loss 0.320123 Objective Loss 0.320123 LR 0.001000 Time 0.021267 +2023-10-05 21:05:43,337 - Epoch: [50][ 570/ 1236] Overall Loss 0.320251 Objective Loss 0.320251 LR 0.001000 Time 0.021241 +2023-10-05 21:05:43,537 - Epoch: [50][ 580/ 1236] Overall Loss 0.319918 Objective Loss 0.319918 LR 0.001000 Time 0.021218 +2023-10-05 21:05:43,735 - Epoch: [50][ 590/ 1236] Overall Loss 0.319426 Objective Loss 0.319426 LR 0.001000 Time 0.021193 +2023-10-05 21:05:43,934 - Epoch: [50][ 600/ 1236] Overall Loss 0.319238 Objective Loss 0.319238 LR 0.001000 Time 0.021172 +2023-10-05 21:05:44,132 - Epoch: [50][ 610/ 1236] Overall Loss 0.318752 Objective Loss 0.318752 LR 0.001000 Time 0.021149 +2023-10-05 21:05:44,332 - Epoch: [50][ 620/ 1236] Overall Loss 0.318535 Objective Loss 0.318535 LR 0.001000 Time 0.021129 +2023-10-05 21:05:44,530 - Epoch: [50][ 630/ 1236] Overall Loss 0.317829 Objective Loss 0.317829 LR 0.001000 Time 0.021107 +2023-10-05 21:05:44,728 - Epoch: [50][ 640/ 1236] Overall Loss 0.317519 Objective Loss 0.317519 LR 0.001000 Time 0.021088 +2023-10-05 21:05:44,927 - Epoch: [50][ 650/ 1236] Overall Loss 0.317194 Objective Loss 0.317194 LR 0.001000 Time 0.021067 +2023-10-05 21:05:45,126 - Epoch: [50][ 660/ 1236] Overall Loss 0.316623 Objective Loss 0.316623 LR 0.001000 Time 0.021050 +2023-10-05 21:05:45,323 - Epoch: [50][ 670/ 1236] Overall Loss 0.317257 Objective Loss 0.317257 LR 0.001000 Time 0.021030 +2023-10-05 21:05:45,522 - Epoch: [50][ 680/ 1236] Overall Loss 0.317412 Objective Loss 0.317412 LR 0.001000 Time 0.021013 +2023-10-05 21:05:45,719 - Epoch: [50][ 690/ 1236] Overall Loss 0.317023 Objective Loss 0.317023 LR 0.001000 Time 0.020993 +2023-10-05 21:05:45,919 - Epoch: [50][ 700/ 1236] Overall Loss 0.316918 Objective Loss 0.316918 LR 0.001000 Time 0.020978 +2023-10-05 21:05:46,116 - Epoch: [50][ 710/ 1236] Overall Loss 0.316903 Objective Loss 0.316903 LR 0.001000 Time 0.020960 +2023-10-05 21:05:46,316 - Epoch: [50][ 720/ 1236] Overall Loss 0.316965 Objective Loss 0.316965 LR 0.001000 Time 0.020945 +2023-10-05 21:05:46,513 - Epoch: [50][ 730/ 1236] Overall Loss 0.317146 Objective Loss 0.317146 LR 0.001000 Time 0.020929 +2023-10-05 21:05:46,712 - Epoch: [50][ 740/ 1236] Overall Loss 0.316840 Objective Loss 0.316840 LR 0.001000 Time 0.020914 +2023-10-05 21:05:46,910 - Epoch: [50][ 750/ 1236] Overall Loss 0.316806 Objective Loss 0.316806 LR 0.001000 Time 0.020899 +2023-10-05 21:05:47,110 - Epoch: [50][ 760/ 1236] Overall Loss 0.316938 Objective Loss 0.316938 LR 0.001000 Time 0.020886 +2023-10-05 21:05:47,308 - Epoch: [50][ 770/ 1236] Overall Loss 0.317122 Objective Loss 0.317122 LR 0.001000 Time 0.020871 +2023-10-05 21:05:47,507 - Epoch: [50][ 780/ 1236] Overall Loss 0.317343 Objective Loss 0.317343 LR 0.001000 Time 0.020859 +2023-10-05 21:05:47,705 - Epoch: [50][ 790/ 1236] Overall Loss 0.317504 Objective Loss 0.317504 LR 0.001000 Time 0.020845 +2023-10-05 21:05:47,904 - Epoch: [50][ 800/ 1236] Overall Loss 0.317379 Objective Loss 0.317379 LR 0.001000 Time 0.020833 +2023-10-05 21:05:48,102 - Epoch: [50][ 810/ 1236] Overall Loss 0.317457 Objective Loss 0.317457 LR 0.001000 Time 0.020819 +2023-10-05 21:05:48,301 - Epoch: [50][ 820/ 1236] Overall Loss 0.317357 Objective Loss 0.317357 LR 0.001000 Time 0.020808 +2023-10-05 21:05:48,498 - Epoch: [50][ 830/ 1236] Overall Loss 0.317518 Objective Loss 0.317518 LR 0.001000 Time 0.020795 +2023-10-05 21:05:48,697 - Epoch: [50][ 840/ 1236] Overall Loss 0.317466 Objective Loss 0.317466 LR 0.001000 Time 0.020783 +2023-10-05 21:05:48,895 - Epoch: [50][ 850/ 1236] Overall Loss 0.317683 Objective Loss 0.317683 LR 0.001000 Time 0.020771 +2023-10-05 21:05:49,094 - Epoch: [50][ 860/ 1236] Overall Loss 0.317907 Objective Loss 0.317907 LR 0.001000 Time 0.020761 +2023-10-05 21:05:49,292 - Epoch: [50][ 870/ 1236] Overall Loss 0.317597 Objective Loss 0.317597 LR 0.001000 Time 0.020749 +2023-10-05 21:05:49,491 - Epoch: [50][ 880/ 1236] Overall Loss 0.318038 Objective Loss 0.318038 LR 0.001000 Time 0.020739 +2023-10-05 21:05:49,688 - Epoch: [50][ 890/ 1236] Overall Loss 0.318399 Objective Loss 0.318399 LR 0.001000 Time 0.020727 +2023-10-05 21:05:49,887 - Epoch: [50][ 900/ 1236] Overall Loss 0.318010 Objective Loss 0.318010 LR 0.001000 Time 0.020718 +2023-10-05 21:05:50,085 - Epoch: [50][ 910/ 1236] Overall Loss 0.318382 Objective Loss 0.318382 LR 0.001000 Time 0.020708 +2023-10-05 21:05:50,284 - Epoch: [50][ 920/ 1236] Overall Loss 0.318444 Objective Loss 0.318444 LR 0.001000 Time 0.020699 +2023-10-05 21:05:50,483 - Epoch: [50][ 930/ 1236] Overall Loss 0.318796 Objective Loss 0.318796 LR 0.001000 Time 0.020689 +2023-10-05 21:05:50,682 - Epoch: [50][ 940/ 1236] Overall Loss 0.318807 Objective Loss 0.318807 LR 0.001000 Time 0.020681 +2023-10-05 21:05:50,880 - Epoch: [50][ 950/ 1236] Overall Loss 0.319075 Objective Loss 0.319075 LR 0.001000 Time 0.020671 +2023-10-05 21:05:51,079 - Epoch: [50][ 960/ 1236] Overall Loss 0.318901 Objective Loss 0.318901 LR 0.001000 Time 0.020663 +2023-10-05 21:05:51,277 - Epoch: [50][ 970/ 1236] Overall Loss 0.319048 Objective Loss 0.319048 LR 0.001000 Time 0.020654 +2023-10-05 21:05:51,476 - Epoch: [50][ 980/ 1236] Overall Loss 0.318738 Objective Loss 0.318738 LR 0.001000 Time 0.020646 +2023-10-05 21:05:51,674 - Epoch: [50][ 990/ 1236] Overall Loss 0.318820 Objective Loss 0.318820 LR 0.001000 Time 0.020637 +2023-10-05 21:05:51,873 - Epoch: [50][ 1000/ 1236] Overall Loss 0.318868 Objective Loss 0.318868 LR 0.001000 Time 0.020629 +2023-10-05 21:05:52,070 - Epoch: [50][ 1010/ 1236] Overall Loss 0.318881 Objective Loss 0.318881 LR 0.001000 Time 0.020620 +2023-10-05 21:05:52,270 - Epoch: [50][ 1020/ 1236] Overall Loss 0.318894 Objective Loss 0.318894 LR 0.001000 Time 0.020613 +2023-10-05 21:05:52,468 - Epoch: [50][ 1030/ 1236] Overall Loss 0.319091 Objective Loss 0.319091 LR 0.001000 Time 0.020605 +2023-10-05 21:05:52,667 - Epoch: [50][ 1040/ 1236] Overall Loss 0.319222 Objective Loss 0.319222 LR 0.001000 Time 0.020598 +2023-10-05 21:05:52,865 - Epoch: [50][ 1050/ 1236] Overall Loss 0.319177 Objective Loss 0.319177 LR 0.001000 Time 0.020590 +2023-10-05 21:05:53,064 - Epoch: [50][ 1060/ 1236] Overall Loss 0.319428 Objective Loss 0.319428 LR 0.001000 Time 0.020583 +2023-10-05 21:05:53,262 - Epoch: [50][ 1070/ 1236] Overall Loss 0.319624 Objective Loss 0.319624 LR 0.001000 Time 0.020576 +2023-10-05 21:05:53,461 - Epoch: [50][ 1080/ 1236] Overall Loss 0.319795 Objective Loss 0.319795 LR 0.001000 Time 0.020569 +2023-10-05 21:05:53,659 - Epoch: [50][ 1090/ 1236] Overall Loss 0.319890 Objective Loss 0.319890 LR 0.001000 Time 0.020562 +2023-10-05 21:05:53,858 - Epoch: [50][ 1100/ 1236] Overall Loss 0.320210 Objective Loss 0.320210 LR 0.001000 Time 0.020555 +2023-10-05 21:05:54,055 - Epoch: [50][ 1110/ 1236] Overall Loss 0.320074 Objective Loss 0.320074 LR 0.001000 Time 0.020548 +2023-10-05 21:05:54,255 - Epoch: [50][ 1120/ 1236] Overall Loss 0.320187 Objective Loss 0.320187 LR 0.001000 Time 0.020542 +2023-10-05 21:05:54,453 - Epoch: [50][ 1130/ 1236] Overall Loss 0.320559 Objective Loss 0.320559 LR 0.001000 Time 0.020535 +2023-10-05 21:05:54,653 - Epoch: [50][ 1140/ 1236] Overall Loss 0.320607 Objective Loss 0.320607 LR 0.001000 Time 0.020530 +2023-10-05 21:05:54,851 - Epoch: [50][ 1150/ 1236] Overall Loss 0.320755 Objective Loss 0.320755 LR 0.001000 Time 0.020523 +2023-10-05 21:05:55,050 - Epoch: [50][ 1160/ 1236] Overall Loss 0.321086 Objective Loss 0.321086 LR 0.001000 Time 0.020518 +2023-10-05 21:05:55,248 - Epoch: [50][ 1170/ 1236] Overall Loss 0.321451 Objective Loss 0.321451 LR 0.001000 Time 0.020512 +2023-10-05 21:05:55,448 - Epoch: [50][ 1180/ 1236] Overall Loss 0.321424 Objective Loss 0.321424 LR 0.001000 Time 0.020507 +2023-10-05 21:05:55,644 - Epoch: [50][ 1190/ 1236] Overall Loss 0.321548 Objective Loss 0.321548 LR 0.001000 Time 0.020499 +2023-10-05 21:05:55,844 - Epoch: [50][ 1200/ 1236] Overall Loss 0.321963 Objective Loss 0.321963 LR 0.001000 Time 0.020495 +2023-10-05 21:05:56,042 - Epoch: [50][ 1210/ 1236] Overall Loss 0.322027 Objective Loss 0.322027 LR 0.001000 Time 0.020488 +2023-10-05 21:05:56,241 - Epoch: [50][ 1220/ 1236] Overall Loss 0.321741 Objective Loss 0.321741 LR 0.001000 Time 0.020484 +2023-10-05 21:05:56,492 - Epoch: [50][ 1230/ 1236] Overall Loss 0.321534 Objective Loss 0.321534 LR 0.001000 Time 0.020521 +2023-10-05 21:05:56,608 - Epoch: [50][ 1236/ 1236] Overall Loss 0.321600 Objective Loss 0.321600 Top1 82.077393 Top5 96.334012 LR 0.001000 Time 0.020515 +2023-10-05 21:05:56,731 - --- validate (epoch=50)----------- +2023-10-05 21:05:56,732 - 29943 samples (256 per mini-batch) +2023-10-05 21:05:57,192 - Epoch: [50][ 10/ 117] Loss 0.370886 Top1 80.703125 Top5 97.851562 +2023-10-05 21:05:57,349 - Epoch: [50][ 20/ 117] Loss 0.367942 Top1 80.292969 Top5 97.480469 +2023-10-05 21:05:57,510 - Epoch: [50][ 30/ 117] Loss 0.368239 Top1 80.690104 Top5 97.526042 +2023-10-05 21:05:57,666 - Epoch: [50][ 40/ 117] Loss 0.368298 Top1 80.810547 Top5 97.373047 +2023-10-05 21:05:57,825 - Epoch: [50][ 50/ 117] Loss 0.371908 Top1 80.835938 Top5 97.390625 +2023-10-05 21:05:57,983 - Epoch: [50][ 60/ 117] Loss 0.369554 Top1 81.028646 Top5 97.441406 +2023-10-05 21:05:58,145 - Epoch: [50][ 70/ 117] Loss 0.374499 Top1 80.831473 Top5 97.427455 +2023-10-05 21:05:58,304 - Epoch: [50][ 80/ 117] Loss 0.371560 Top1 80.903320 Top5 97.421875 +2023-10-05 21:05:58,463 - Epoch: [50][ 90/ 117] Loss 0.370465 Top1 81.028646 Top5 97.387153 +2023-10-05 21:05:58,615 - Epoch: [50][ 100/ 117] Loss 0.366529 Top1 81.117188 Top5 97.425781 +2023-10-05 21:05:58,773 - Epoch: [50][ 110/ 117] Loss 0.363836 Top1 81.168324 Top5 97.443182 +2023-10-05 21:05:58,858 - Epoch: [50][ 117/ 117] Loss 0.362527 Top1 81.147514 Top5 97.445146 +2023-10-05 21:05:58,955 - ==> Top1: 81.148 Top5: 97.445 Loss: 0.363 + +2023-10-05 21:05:58,956 - ==> Confusion: +[[ 937 1 3 1 18 3 0 0 4 46 1 0 0 6 8 3 6 1 0 0 12] + [ 1 1030 1 0 13 32 1 18 3 0 3 2 0 0 1 3 8 0 8 2 5] + [ 5 1 919 27 4 0 36 8 0 2 8 3 6 1 2 2 2 2 7 4 17] + [ 3 0 14 966 1 9 2 0 3 0 4 2 3 5 30 3 5 3 23 3 10] + [ 22 10 0 0 967 5 1 0 0 4 0 2 0 1 12 3 15 2 1 2 3] + [ 4 32 1 2 7 967 2 23 3 1 1 12 2 23 7 0 9 0 2 4 14] + [ 0 10 21 1 2 1 1115 8 0 0 3 3 1 0 1 6 2 1 1 5 10] + [ 2 22 9 0 3 30 1 1054 0 6 6 7 3 4 0 2 3 0 42 9 15] + [ 22 1 0 2 0 4 1 0 945 36 19 2 0 22 23 6 4 0 2 0 0] + [ 119 1 2 0 11 6 0 0 37 880 3 2 0 36 10 2 2 0 1 2 5] + [ 3 3 13 11 2 0 4 7 11 2 950 1 0 22 9 2 3 0 3 1 6] + [ 1 1 2 0 1 14 0 3 0 0 0 961 20 2 0 6 3 12 0 5 4] + [ 2 0 5 6 1 1 0 2 0 2 1 50 941 3 6 19 1 10 3 3 12] + [ 3 0 2 0 4 8 0 0 10 5 9 3 3 1048 2 4 1 1 0 3 13] + [ 15 1 2 8 8 0 0 0 17 7 3 1 2 2 1010 0 3 1 10 0 11] + [ 2 1 4 4 3 1 2 0 0 0 0 13 4 1 1 1058 22 8 1 5 4] + [ 1 14 1 0 3 4 0 0 3 0 0 4 1 1 1 14 1105 0 0 3 6] + [ 3 0 2 5 0 0 1 0 1 1 0 9 18 3 3 15 2 971 2 0 2] + [ 4 8 12 15 0 1 0 36 7 1 8 0 2 0 10 0 3 0 952 0 9] + [ 0 4 4 1 3 10 3 14 1 1 1 24 6 4 1 6 20 1 1 1029 18] + [ 156 205 147 83 103 188 55 98 84 81 197 164 380 395 183 78 373 69 166 207 4493]] + +2023-10-05 21:05:58,957 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:05:58,957 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:05:58,963 - + +2023-10-05 21:05:58,963 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:05:59,939 - Epoch: [51][ 10/ 1236] Overall Loss 0.309471 Objective Loss 0.309471 LR 0.001000 Time 0.097556 +2023-10-05 21:06:00,140 - Epoch: [51][ 20/ 1236] Overall Loss 0.322329 Objective Loss 0.322329 LR 0.001000 Time 0.058767 +2023-10-05 21:06:00,338 - Epoch: [51][ 30/ 1236] Overall Loss 0.326328 Objective Loss 0.326328 LR 0.001000 Time 0.045796 +2023-10-05 21:06:00,539 - Epoch: [51][ 40/ 1236] Overall Loss 0.317622 Objective Loss 0.317622 LR 0.001000 Time 0.039347 +2023-10-05 21:06:00,738 - Epoch: [51][ 50/ 1236] Overall Loss 0.318053 Objective Loss 0.318053 LR 0.001000 Time 0.035460 +2023-10-05 21:06:00,938 - Epoch: [51][ 60/ 1236] Overall Loss 0.311897 Objective Loss 0.311897 LR 0.001000 Time 0.032868 +2023-10-05 21:06:01,137 - Epoch: [51][ 70/ 1236] Overall Loss 0.307706 Objective Loss 0.307706 LR 0.001000 Time 0.031012 +2023-10-05 21:06:01,337 - Epoch: [51][ 80/ 1236] Overall Loss 0.308481 Objective Loss 0.308481 LR 0.001000 Time 0.029630 +2023-10-05 21:06:01,536 - Epoch: [51][ 90/ 1236] Overall Loss 0.313466 Objective Loss 0.313466 LR 0.001000 Time 0.028552 +2023-10-05 21:06:01,736 - Epoch: [51][ 100/ 1236] Overall Loss 0.314291 Objective Loss 0.314291 LR 0.001000 Time 0.027692 +2023-10-05 21:06:01,936 - Epoch: [51][ 110/ 1236] Overall Loss 0.316443 Objective Loss 0.316443 LR 0.001000 Time 0.026986 +2023-10-05 21:06:02,135 - Epoch: [51][ 120/ 1236] Overall Loss 0.315464 Objective Loss 0.315464 LR 0.001000 Time 0.026399 +2023-10-05 21:06:02,334 - Epoch: [51][ 130/ 1236] Overall Loss 0.315537 Objective Loss 0.315537 LR 0.001000 Time 0.025894 +2023-10-05 21:06:02,533 - Epoch: [51][ 140/ 1236] Overall Loss 0.318069 Objective Loss 0.318069 LR 0.001000 Time 0.025464 +2023-10-05 21:06:02,731 - Epoch: [51][ 150/ 1236] Overall Loss 0.318021 Objective Loss 0.318021 LR 0.001000 Time 0.025082 +2023-10-05 21:06:02,942 - Epoch: [51][ 160/ 1236] Overall Loss 0.318278 Objective Loss 0.318278 LR 0.001000 Time 0.024829 +2023-10-05 21:06:03,154 - Epoch: [51][ 170/ 1236] Overall Loss 0.317172 Objective Loss 0.317172 LR 0.001000 Time 0.024614 +2023-10-05 21:06:03,369 - Epoch: [51][ 180/ 1236] Overall Loss 0.316276 Objective Loss 0.316276 LR 0.001000 Time 0.024439 +2023-10-05 21:06:03,581 - Epoch: [51][ 190/ 1236] Overall Loss 0.315182 Objective Loss 0.315182 LR 0.001000 Time 0.024268 +2023-10-05 21:06:03,797 - Epoch: [51][ 200/ 1236] Overall Loss 0.315312 Objective Loss 0.315312 LR 0.001000 Time 0.024129 +2023-10-05 21:06:04,009 - Epoch: [51][ 210/ 1236] Overall Loss 0.314300 Objective Loss 0.314300 LR 0.001000 Time 0.023988 +2023-10-05 21:06:04,224 - Epoch: [51][ 220/ 1236] Overall Loss 0.315350 Objective Loss 0.315350 LR 0.001000 Time 0.023873 +2023-10-05 21:06:04,436 - Epoch: [51][ 230/ 1236] Overall Loss 0.315064 Objective Loss 0.315064 LR 0.001000 Time 0.023755 +2023-10-05 21:06:04,652 - Epoch: [51][ 240/ 1236] Overall Loss 0.315472 Objective Loss 0.315472 LR 0.001000 Time 0.023665 +2023-10-05 21:06:04,865 - Epoch: [51][ 250/ 1236] Overall Loss 0.316219 Objective Loss 0.316219 LR 0.001000 Time 0.023566 +2023-10-05 21:06:05,080 - Epoch: [51][ 260/ 1236] Overall Loss 0.315687 Objective Loss 0.315687 LR 0.001000 Time 0.023485 +2023-10-05 21:06:05,292 - Epoch: [51][ 270/ 1236] Overall Loss 0.315838 Objective Loss 0.315838 LR 0.001000 Time 0.023400 +2023-10-05 21:06:05,508 - Epoch: [51][ 280/ 1236] Overall Loss 0.315165 Objective Loss 0.315165 LR 0.001000 Time 0.023331 +2023-10-05 21:06:05,720 - Epoch: [51][ 290/ 1236] Overall Loss 0.315479 Objective Loss 0.315479 LR 0.001000 Time 0.023257 +2023-10-05 21:06:05,935 - Epoch: [51][ 300/ 1236] Overall Loss 0.315621 Objective Loss 0.315621 LR 0.001000 Time 0.023197 +2023-10-05 21:06:06,147 - Epoch: [51][ 310/ 1236] Overall Loss 0.314387 Objective Loss 0.314387 LR 0.001000 Time 0.023132 +2023-10-05 21:06:06,363 - Epoch: [51][ 320/ 1236] Overall Loss 0.313367 Objective Loss 0.313367 LR 0.001000 Time 0.023082 +2023-10-05 21:06:06,577 - Epoch: [51][ 330/ 1236] Overall Loss 0.313275 Objective Loss 0.313275 LR 0.001000 Time 0.023028 +2023-10-05 21:06:06,792 - Epoch: [51][ 340/ 1236] Overall Loss 0.311985 Objective Loss 0.311985 LR 0.001000 Time 0.022982 +2023-10-05 21:06:07,004 - Epoch: [51][ 350/ 1236] Overall Loss 0.311633 Objective Loss 0.311633 LR 0.001000 Time 0.022930 +2023-10-05 21:06:07,215 - Epoch: [51][ 360/ 1236] Overall Loss 0.311389 Objective Loss 0.311389 LR 0.001000 Time 0.022877 +2023-10-05 21:06:07,421 - Epoch: [51][ 370/ 1236] Overall Loss 0.310347 Objective Loss 0.310347 LR 0.001000 Time 0.022816 +2023-10-05 21:06:07,632 - Epoch: [51][ 380/ 1236] Overall Loss 0.310914 Objective Loss 0.310914 LR 0.001000 Time 0.022770 +2023-10-05 21:06:07,839 - Epoch: [51][ 390/ 1236] Overall Loss 0.310498 Objective Loss 0.310498 LR 0.001000 Time 0.022715 +2023-10-05 21:06:08,050 - Epoch: [51][ 400/ 1236] Overall Loss 0.310281 Objective Loss 0.310281 LR 0.001000 Time 0.022673 +2023-10-05 21:06:08,256 - Epoch: [51][ 410/ 1236] Overall Loss 0.310965 Objective Loss 0.310965 LR 0.001000 Time 0.022624 +2023-10-05 21:06:08,467 - Epoch: [51][ 420/ 1236] Overall Loss 0.311134 Objective Loss 0.311134 LR 0.001000 Time 0.022586 +2023-10-05 21:06:08,674 - Epoch: [51][ 430/ 1236] Overall Loss 0.311486 Objective Loss 0.311486 LR 0.001000 Time 0.022540 +2023-10-05 21:06:08,885 - Epoch: [51][ 440/ 1236] Overall Loss 0.311866 Objective Loss 0.311866 LR 0.001000 Time 0.022507 +2023-10-05 21:06:09,091 - Epoch: [51][ 450/ 1236] Overall Loss 0.311314 Objective Loss 0.311314 LR 0.001000 Time 0.022465 +2023-10-05 21:06:09,296 - Epoch: [51][ 460/ 1236] Overall Loss 0.311344 Objective Loss 0.311344 LR 0.001000 Time 0.022421 +2023-10-05 21:06:09,496 - Epoch: [51][ 470/ 1236] Overall Loss 0.311881 Objective Loss 0.311881 LR 0.001000 Time 0.022366 +2023-10-05 21:06:09,696 - Epoch: [51][ 480/ 1236] Overall Loss 0.312173 Objective Loss 0.312173 LR 0.001000 Time 0.022316 +2023-10-05 21:06:09,897 - Epoch: [51][ 490/ 1236] Overall Loss 0.311790 Objective Loss 0.311790 LR 0.001000 Time 0.022268 +2023-10-05 21:06:10,098 - Epoch: [51][ 500/ 1236] Overall Loss 0.312225 Objective Loss 0.312225 LR 0.001000 Time 0.022224 +2023-10-05 21:06:10,299 - Epoch: [51][ 510/ 1236] Overall Loss 0.312010 Objective Loss 0.312010 LR 0.001000 Time 0.022178 +2023-10-05 21:06:10,501 - Epoch: [51][ 520/ 1236] Overall Loss 0.312123 Objective Loss 0.312123 LR 0.001000 Time 0.022140 +2023-10-05 21:06:10,703 - Epoch: [51][ 530/ 1236] Overall Loss 0.312398 Objective Loss 0.312398 LR 0.001000 Time 0.022100 +2023-10-05 21:06:10,905 - Epoch: [51][ 540/ 1236] Overall Loss 0.312492 Objective Loss 0.312492 LR 0.001000 Time 0.022065 +2023-10-05 21:06:11,108 - Epoch: [51][ 550/ 1236] Overall Loss 0.312783 Objective Loss 0.312783 LR 0.001000 Time 0.022029 +2023-10-05 21:06:11,310 - Epoch: [51][ 560/ 1236] Overall Loss 0.312788 Objective Loss 0.312788 LR 0.001000 Time 0.021997 +2023-10-05 21:06:11,512 - Epoch: [51][ 570/ 1236] Overall Loss 0.313367 Objective Loss 0.313367 LR 0.001000 Time 0.021965 +2023-10-05 21:06:11,715 - Epoch: [51][ 580/ 1236] Overall Loss 0.313880 Objective Loss 0.313880 LR 0.001000 Time 0.021935 +2023-10-05 21:06:11,917 - Epoch: [51][ 590/ 1236] Overall Loss 0.314464 Objective Loss 0.314464 LR 0.001000 Time 0.021902 +2023-10-05 21:06:12,119 - Epoch: [51][ 600/ 1236] Overall Loss 0.314473 Objective Loss 0.314473 LR 0.001000 Time 0.021875 +2023-10-05 21:06:12,322 - Epoch: [51][ 610/ 1236] Overall Loss 0.314211 Objective Loss 0.314211 LR 0.001000 Time 0.021845 +2023-10-05 21:06:12,524 - Epoch: [51][ 620/ 1236] Overall Loss 0.314186 Objective Loss 0.314186 LR 0.001000 Time 0.021819 +2023-10-05 21:06:12,727 - Epoch: [51][ 630/ 1236] Overall Loss 0.314662 Objective Loss 0.314662 LR 0.001000 Time 0.021791 +2023-10-05 21:06:12,929 - Epoch: [51][ 640/ 1236] Overall Loss 0.314786 Objective Loss 0.314786 LR 0.001000 Time 0.021767 +2023-10-05 21:06:13,131 - Epoch: [51][ 650/ 1236] Overall Loss 0.315072 Objective Loss 0.315072 LR 0.001000 Time 0.021742 +2023-10-05 21:06:13,334 - Epoch: [51][ 660/ 1236] Overall Loss 0.315483 Objective Loss 0.315483 LR 0.001000 Time 0.021719 +2023-10-05 21:06:13,536 - Epoch: [51][ 670/ 1236] Overall Loss 0.316125 Objective Loss 0.316125 LR 0.001000 Time 0.021696 +2023-10-05 21:06:13,738 - Epoch: [51][ 680/ 1236] Overall Loss 0.316112 Objective Loss 0.316112 LR 0.001000 Time 0.021674 +2023-10-05 21:06:13,940 - Epoch: [51][ 690/ 1236] Overall Loss 0.316096 Objective Loss 0.316096 LR 0.001000 Time 0.021651 +2023-10-05 21:06:14,142 - Epoch: [51][ 700/ 1236] Overall Loss 0.316153 Objective Loss 0.316153 LR 0.001000 Time 0.021631 +2023-10-05 21:06:14,344 - Epoch: [51][ 710/ 1236] Overall Loss 0.315983 Objective Loss 0.315983 LR 0.001000 Time 0.021610 +2023-10-05 21:06:14,547 - Epoch: [51][ 720/ 1236] Overall Loss 0.316127 Objective Loss 0.316127 LR 0.001000 Time 0.021591 +2023-10-05 21:06:14,748 - Epoch: [51][ 730/ 1236] Overall Loss 0.316709 Objective Loss 0.316709 LR 0.001000 Time 0.021571 +2023-10-05 21:06:14,952 - Epoch: [51][ 740/ 1236] Overall Loss 0.316303 Objective Loss 0.316303 LR 0.001000 Time 0.021553 +2023-10-05 21:06:15,154 - Epoch: [51][ 750/ 1236] Overall Loss 0.316253 Objective Loss 0.316253 LR 0.001000 Time 0.021535 +2023-10-05 21:06:15,356 - Epoch: [51][ 760/ 1236] Overall Loss 0.316217 Objective Loss 0.316217 LR 0.001000 Time 0.021517 +2023-10-05 21:06:15,556 - Epoch: [51][ 770/ 1236] Overall Loss 0.316747 Objective Loss 0.316747 LR 0.001000 Time 0.021497 +2023-10-05 21:06:15,757 - Epoch: [51][ 780/ 1236] Overall Loss 0.317632 Objective Loss 0.317632 LR 0.001000 Time 0.021479 +2023-10-05 21:06:15,957 - Epoch: [51][ 790/ 1236] Overall Loss 0.318331 Objective Loss 0.318331 LR 0.001000 Time 0.021458 +2023-10-05 21:06:16,157 - Epoch: [51][ 800/ 1236] Overall Loss 0.318673 Objective Loss 0.318673 LR 0.001000 Time 0.021439 +2023-10-05 21:06:16,357 - Epoch: [51][ 810/ 1236] Overall Loss 0.318632 Objective Loss 0.318632 LR 0.001000 Time 0.021421 +2023-10-05 21:06:16,558 - Epoch: [51][ 820/ 1236] Overall Loss 0.318563 Objective Loss 0.318563 LR 0.001000 Time 0.021404 +2023-10-05 21:06:16,758 - Epoch: [51][ 830/ 1236] Overall Loss 0.318355 Objective Loss 0.318355 LR 0.001000 Time 0.021387 +2023-10-05 21:06:16,958 - Epoch: [51][ 840/ 1236] Overall Loss 0.318335 Objective Loss 0.318335 LR 0.001000 Time 0.021370 +2023-10-05 21:06:17,159 - Epoch: [51][ 850/ 1236] Overall Loss 0.318596 Objective Loss 0.318596 LR 0.001000 Time 0.021354 +2023-10-05 21:06:17,359 - Epoch: [51][ 860/ 1236] Overall Loss 0.318606 Objective Loss 0.318606 LR 0.001000 Time 0.021339 +2023-10-05 21:06:17,559 - Epoch: [51][ 870/ 1236] Overall Loss 0.318586 Objective Loss 0.318586 LR 0.001000 Time 0.021321 +2023-10-05 21:06:17,760 - Epoch: [51][ 880/ 1236] Overall Loss 0.318716 Objective Loss 0.318716 LR 0.001000 Time 0.021307 +2023-10-05 21:06:17,960 - Epoch: [51][ 890/ 1236] Overall Loss 0.318632 Objective Loss 0.318632 LR 0.001000 Time 0.021290 +2023-10-05 21:06:18,160 - Epoch: [51][ 900/ 1236] Overall Loss 0.318752 Objective Loss 0.318752 LR 0.001000 Time 0.021276 +2023-10-05 21:06:18,361 - Epoch: [51][ 910/ 1236] Overall Loss 0.318818 Objective Loss 0.318818 LR 0.001000 Time 0.021262 +2023-10-05 21:06:18,561 - Epoch: [51][ 920/ 1236] Overall Loss 0.318907 Objective Loss 0.318907 LR 0.001000 Time 0.021248 +2023-10-05 21:06:18,761 - Epoch: [51][ 930/ 1236] Overall Loss 0.318827 Objective Loss 0.318827 LR 0.001000 Time 0.021234 +2023-10-05 21:06:18,961 - Epoch: [51][ 940/ 1236] Overall Loss 0.319058 Objective Loss 0.319058 LR 0.001000 Time 0.021221 +2023-10-05 21:06:19,162 - Epoch: [51][ 950/ 1236] Overall Loss 0.319013 Objective Loss 0.319013 LR 0.001000 Time 0.021209 +2023-10-05 21:06:19,362 - Epoch: [51][ 960/ 1236] Overall Loss 0.319110 Objective Loss 0.319110 LR 0.001000 Time 0.021196 +2023-10-05 21:06:19,562 - Epoch: [51][ 970/ 1236] Overall Loss 0.319372 Objective Loss 0.319372 LR 0.001000 Time 0.021182 +2023-10-05 21:06:19,763 - Epoch: [51][ 980/ 1236] Overall Loss 0.319894 Objective Loss 0.319894 LR 0.001000 Time 0.021170 +2023-10-05 21:06:19,963 - Epoch: [51][ 990/ 1236] Overall Loss 0.319754 Objective Loss 0.319754 LR 0.001000 Time 0.021158 +2023-10-05 21:06:20,163 - Epoch: [51][ 1000/ 1236] Overall Loss 0.319803 Objective Loss 0.319803 LR 0.001000 Time 0.021146 +2023-10-05 21:06:20,363 - Epoch: [51][ 1010/ 1236] Overall Loss 0.319510 Objective Loss 0.319510 LR 0.001000 Time 0.021134 +2023-10-05 21:06:20,564 - Epoch: [51][ 1020/ 1236] Overall Loss 0.319689 Objective Loss 0.319689 LR 0.001000 Time 0.021123 +2023-10-05 21:06:20,764 - Epoch: [51][ 1030/ 1236] Overall Loss 0.319690 Objective Loss 0.319690 LR 0.001000 Time 0.021110 +2023-10-05 21:06:20,964 - Epoch: [51][ 1040/ 1236] Overall Loss 0.319349 Objective Loss 0.319349 LR 0.001000 Time 0.021099 +2023-10-05 21:06:21,164 - Epoch: [51][ 1050/ 1236] Overall Loss 0.319088 Objective Loss 0.319088 LR 0.001000 Time 0.021088 +2023-10-05 21:06:21,365 - Epoch: [51][ 1060/ 1236] Overall Loss 0.319013 Objective Loss 0.319013 LR 0.001000 Time 0.021078 +2023-10-05 21:06:21,565 - Epoch: [51][ 1070/ 1236] Overall Loss 0.319165 Objective Loss 0.319165 LR 0.001000 Time 0.021068 +2023-10-05 21:06:21,766 - Epoch: [51][ 1080/ 1236] Overall Loss 0.319373 Objective Loss 0.319373 LR 0.001000 Time 0.021058 +2023-10-05 21:06:21,967 - Epoch: [51][ 1090/ 1236] Overall Loss 0.319277 Objective Loss 0.319277 LR 0.001000 Time 0.021048 +2023-10-05 21:06:22,168 - Epoch: [51][ 1100/ 1236] Overall Loss 0.318805 Objective Loss 0.318805 LR 0.001000 Time 0.021038 +2023-10-05 21:06:22,368 - Epoch: [51][ 1110/ 1236] Overall Loss 0.318671 Objective Loss 0.318671 LR 0.001000 Time 0.021029 +2023-10-05 21:06:22,569 - Epoch: [51][ 1120/ 1236] Overall Loss 0.318731 Objective Loss 0.318731 LR 0.001000 Time 0.021021 +2023-10-05 21:06:22,770 - Epoch: [51][ 1130/ 1236] Overall Loss 0.318916 Objective Loss 0.318916 LR 0.001000 Time 0.021012 +2023-10-05 21:06:22,971 - Epoch: [51][ 1140/ 1236] Overall Loss 0.318801 Objective Loss 0.318801 LR 0.001000 Time 0.021004 +2023-10-05 21:06:23,172 - Epoch: [51][ 1150/ 1236] Overall Loss 0.319120 Objective Loss 0.319120 LR 0.001000 Time 0.020995 +2023-10-05 21:06:23,373 - Epoch: [51][ 1160/ 1236] Overall Loss 0.318858 Objective Loss 0.318858 LR 0.001000 Time 0.020988 +2023-10-05 21:06:23,574 - Epoch: [51][ 1170/ 1236] Overall Loss 0.318896 Objective Loss 0.318896 LR 0.001000 Time 0.020980 +2023-10-05 21:06:23,775 - Epoch: [51][ 1180/ 1236] Overall Loss 0.318841 Objective Loss 0.318841 LR 0.001000 Time 0.020972 +2023-10-05 21:06:23,976 - Epoch: [51][ 1190/ 1236] Overall Loss 0.319004 Objective Loss 0.319004 LR 0.001000 Time 0.020964 +2023-10-05 21:06:24,177 - Epoch: [51][ 1200/ 1236] Overall Loss 0.319190 Objective Loss 0.319190 LR 0.001000 Time 0.020957 +2023-10-05 21:06:24,378 - Epoch: [51][ 1210/ 1236] Overall Loss 0.319302 Objective Loss 0.319302 LR 0.001000 Time 0.020949 +2023-10-05 21:06:24,579 - Epoch: [51][ 1220/ 1236] Overall Loss 0.319247 Objective Loss 0.319247 LR 0.001000 Time 0.020942 +2023-10-05 21:06:24,831 - Epoch: [51][ 1230/ 1236] Overall Loss 0.319326 Objective Loss 0.319326 LR 0.001000 Time 0.020975 +2023-10-05 21:06:24,948 - Epoch: [51][ 1236/ 1236] Overall Loss 0.319210 Objective Loss 0.319210 Top1 85.336049 Top5 96.537678 LR 0.001000 Time 0.020968 +2023-10-05 21:06:25,066 - --- validate (epoch=51)----------- +2023-10-05 21:06:25,067 - 29943 samples (256 per mini-batch) +2023-10-05 21:06:25,516 - Epoch: [51][ 10/ 117] Loss 0.372163 Top1 81.484375 Top5 97.851562 +2023-10-05 21:06:25,665 - Epoch: [51][ 20/ 117] Loss 0.366663 Top1 80.820312 Top5 97.421875 +2023-10-05 21:06:25,810 - Epoch: [51][ 30/ 117] Loss 0.374017 Top1 80.755208 Top5 97.460938 +2023-10-05 21:06:25,959 - Epoch: [51][ 40/ 117] Loss 0.383032 Top1 80.693359 Top5 97.392578 +2023-10-05 21:06:26,104 - Epoch: [51][ 50/ 117] Loss 0.381545 Top1 80.726562 Top5 97.437500 +2023-10-05 21:06:26,253 - Epoch: [51][ 60/ 117] Loss 0.378697 Top1 80.898438 Top5 97.408854 +2023-10-05 21:06:26,397 - Epoch: [51][ 70/ 117] Loss 0.372763 Top1 80.915179 Top5 97.438616 +2023-10-05 21:06:26,545 - Epoch: [51][ 80/ 117] Loss 0.370484 Top1 80.908203 Top5 97.402344 +2023-10-05 21:06:26,691 - Epoch: [51][ 90/ 117] Loss 0.370887 Top1 80.950521 Top5 97.361111 +2023-10-05 21:06:26,841 - Epoch: [51][ 100/ 117] Loss 0.369060 Top1 81.042969 Top5 97.343750 +2023-10-05 21:06:26,995 - Epoch: [51][ 110/ 117] Loss 0.370432 Top1 81.029830 Top5 97.279830 +2023-10-05 21:06:27,080 - Epoch: [51][ 117/ 117] Loss 0.368873 Top1 81.104098 Top5 97.308219 +2023-10-05 21:06:27,230 - ==> Top1: 81.104 Top5: 97.308 Loss: 0.369 + +2023-10-05 21:06:27,230 - ==> Confusion: +[[ 932 3 3 1 10 2 0 0 4 67 1 0 1 3 5 5 2 1 1 1 8] + [ 0 999 2 1 9 34 1 42 5 0 4 2 0 2 1 5 8 0 7 4 5] + [ 7 1 918 16 4 0 40 11 0 3 3 5 5 5 1 10 2 2 6 4 13] + [ 4 1 25 930 2 5 1 1 4 0 11 0 10 6 23 5 2 12 24 5 18] + [ 31 4 1 0 956 6 3 0 0 16 2 1 2 4 9 2 10 0 1 0 2] + [ 7 35 1 0 5 954 3 36 1 1 3 20 2 18 3 0 5 2 4 6 10] + [ 0 7 29 0 0 2 1107 11 0 0 4 3 2 0 0 11 0 2 1 6 6] + [ 5 14 8 0 2 29 5 1079 1 3 0 11 4 1 0 0 0 0 39 8 9] + [ 15 1 1 0 0 1 0 0 945 50 9 5 5 26 19 0 0 1 10 0 1] + [ 107 2 5 0 6 4 0 0 30 912 0 1 0 33 12 2 1 0 0 1 3] + [ 4 4 11 8 0 2 6 5 22 2 933 1 1 22 7 0 1 1 12 0 11] + [ 1 0 0 0 0 10 1 5 0 0 0 955 37 1 1 1 4 13 0 6 0] + [ 2 3 3 2 0 2 0 1 0 1 0 47 956 1 3 7 2 18 3 11 6] + [ 0 0 1 1 3 8 0 0 7 15 5 13 8 1046 3 2 0 0 0 4 3] + [ 19 0 3 7 2 0 0 0 19 9 3 1 5 5 992 1 1 2 18 0 14] + [ 0 0 1 1 5 0 0 0 0 0 0 13 13 0 0 1048 18 14 3 12 6] + [ 1 14 2 1 5 7 0 0 2 0 0 5 4 3 4 13 1082 0 0 7 11] + [ 0 0 0 0 0 0 0 0 0 1 0 12 27 1 4 7 0 983 3 0 0] + [ 2 4 10 10 1 0 1 48 3 0 1 0 9 0 9 0 1 1 958 1 9] + [ 0 1 1 0 2 5 7 13 0 0 2 19 7 2 1 8 6 1 2 1068 7] + [ 163 163 151 51 100 183 42 170 97 113 125 178 413 456 137 84 177 77 194 299 4532]] + +2023-10-05 21:06:27,232 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:06:27,232 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:06:27,237 - + +2023-10-05 21:06:27,237 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:06:28,337 - Epoch: [52][ 10/ 1236] Overall Loss 0.297667 Objective Loss 0.297667 LR 0.001000 Time 0.109941 +2023-10-05 21:06:28,537 - Epoch: [52][ 20/ 1236] Overall Loss 0.316244 Objective Loss 0.316244 LR 0.001000 Time 0.064944 +2023-10-05 21:06:28,736 - Epoch: [52][ 30/ 1236] Overall Loss 0.376936 Objective Loss 0.376936 LR 0.001000 Time 0.049899 +2023-10-05 21:06:28,936 - Epoch: [52][ 40/ 1236] Overall Loss 0.403382 Objective Loss 0.403382 LR 0.001000 Time 0.042414 +2023-10-05 21:06:29,134 - Epoch: [52][ 50/ 1236] Overall Loss 0.406249 Objective Loss 0.406249 LR 0.001000 Time 0.037901 +2023-10-05 21:06:29,334 - Epoch: [52][ 60/ 1236] Overall Loss 0.409425 Objective Loss 0.409425 LR 0.001000 Time 0.034912 +2023-10-05 21:06:29,533 - Epoch: [52][ 70/ 1236] Overall Loss 0.413365 Objective Loss 0.413365 LR 0.001000 Time 0.032764 +2023-10-05 21:06:29,733 - Epoch: [52][ 80/ 1236] Overall Loss 0.408490 Objective Loss 0.408490 LR 0.001000 Time 0.031157 +2023-10-05 21:06:29,932 - Epoch: [52][ 90/ 1236] Overall Loss 0.406038 Objective Loss 0.406038 LR 0.001000 Time 0.029900 +2023-10-05 21:06:30,132 - Epoch: [52][ 100/ 1236] Overall Loss 0.404394 Objective Loss 0.404394 LR 0.001000 Time 0.028904 +2023-10-05 21:06:30,330 - Epoch: [52][ 110/ 1236] Overall Loss 0.403097 Objective Loss 0.403097 LR 0.001000 Time 0.028082 +2023-10-05 21:06:30,530 - Epoch: [52][ 120/ 1236] Overall Loss 0.402099 Objective Loss 0.402099 LR 0.001000 Time 0.027405 +2023-10-05 21:06:30,728 - Epoch: [52][ 130/ 1236] Overall Loss 0.402278 Objective Loss 0.402278 LR 0.001000 Time 0.026818 +2023-10-05 21:06:30,928 - Epoch: [52][ 140/ 1236] Overall Loss 0.401635 Objective Loss 0.401635 LR 0.001000 Time 0.026323 +2023-10-05 21:06:31,127 - Epoch: [52][ 150/ 1236] Overall Loss 0.400292 Objective Loss 0.400292 LR 0.001000 Time 0.025893 +2023-10-05 21:06:31,325 - Epoch: [52][ 160/ 1236] Overall Loss 0.400493 Objective Loss 0.400493 LR 0.001000 Time 0.025511 +2023-10-05 21:06:31,524 - Epoch: [52][ 170/ 1236] Overall Loss 0.401178 Objective Loss 0.401178 LR 0.001000 Time 0.025178 +2023-10-05 21:06:31,723 - Epoch: [52][ 180/ 1236] Overall Loss 0.399392 Objective Loss 0.399392 LR 0.001000 Time 0.024885 +2023-10-05 21:06:31,921 - Epoch: [52][ 190/ 1236] Overall Loss 0.398181 Objective Loss 0.398181 LR 0.001000 Time 0.024617 +2023-10-05 21:06:32,120 - Epoch: [52][ 200/ 1236] Overall Loss 0.398793 Objective Loss 0.398793 LR 0.001000 Time 0.024376 +2023-10-05 21:06:32,318 - Epoch: [52][ 210/ 1236] Overall Loss 0.397778 Objective Loss 0.397778 LR 0.001000 Time 0.024157 +2023-10-05 21:06:32,517 - Epoch: [52][ 220/ 1236] Overall Loss 0.397398 Objective Loss 0.397398 LR 0.001000 Time 0.023964 +2023-10-05 21:06:32,718 - Epoch: [52][ 230/ 1236] Overall Loss 0.395593 Objective Loss 0.395593 LR 0.001000 Time 0.023792 +2023-10-05 21:06:32,919 - Epoch: [52][ 240/ 1236] Overall Loss 0.394926 Objective Loss 0.394926 LR 0.001000 Time 0.023639 +2023-10-05 21:06:33,122 - Epoch: [52][ 250/ 1236] Overall Loss 0.394181 Objective Loss 0.394181 LR 0.001000 Time 0.023504 +2023-10-05 21:06:33,323 - Epoch: [52][ 260/ 1236] Overall Loss 0.392498 Objective Loss 0.392498 LR 0.001000 Time 0.023373 +2023-10-05 21:06:33,525 - Epoch: [52][ 270/ 1236] Overall Loss 0.391862 Objective Loss 0.391862 LR 0.001000 Time 0.023254 +2023-10-05 21:06:33,727 - Epoch: [52][ 280/ 1236] Overall Loss 0.390369 Objective Loss 0.390369 LR 0.001000 Time 0.023142 +2023-10-05 21:06:33,929 - Epoch: [52][ 290/ 1236] Overall Loss 0.389956 Objective Loss 0.389956 LR 0.001000 Time 0.023040 +2023-10-05 21:06:34,130 - Epoch: [52][ 300/ 1236] Overall Loss 0.388583 Objective Loss 0.388583 LR 0.001000 Time 0.022942 +2023-10-05 21:06:34,333 - Epoch: [52][ 310/ 1236] Overall Loss 0.388854 Objective Loss 0.388854 LR 0.001000 Time 0.022854 +2023-10-05 21:06:34,535 - Epoch: [52][ 320/ 1236] Overall Loss 0.387782 Objective Loss 0.387782 LR 0.001000 Time 0.022771 +2023-10-05 21:06:34,737 - Epoch: [52][ 330/ 1236] Overall Loss 0.386701 Objective Loss 0.386701 LR 0.001000 Time 0.022693 +2023-10-05 21:06:34,939 - Epoch: [52][ 340/ 1236] Overall Loss 0.385801 Objective Loss 0.385801 LR 0.001000 Time 0.022618 +2023-10-05 21:06:35,142 - Epoch: [52][ 350/ 1236] Overall Loss 0.385651 Objective Loss 0.385651 LR 0.001000 Time 0.022549 +2023-10-05 21:06:35,343 - Epoch: [52][ 360/ 1236] Overall Loss 0.384805 Objective Loss 0.384805 LR 0.001000 Time 0.022482 +2023-10-05 21:06:35,546 - Epoch: [52][ 370/ 1236] Overall Loss 0.384066 Objective Loss 0.384066 LR 0.001000 Time 0.022421 +2023-10-05 21:06:35,748 - Epoch: [52][ 380/ 1236] Overall Loss 0.383186 Objective Loss 0.383186 LR 0.001000 Time 0.022363 +2023-10-05 21:06:35,951 - Epoch: [52][ 390/ 1236] Overall Loss 0.382688 Objective Loss 0.382688 LR 0.001000 Time 0.022307 +2023-10-05 21:06:36,152 - Epoch: [52][ 400/ 1236] Overall Loss 0.381619 Objective Loss 0.381619 LR 0.001000 Time 0.022253 +2023-10-05 21:06:36,354 - Epoch: [52][ 410/ 1236] Overall Loss 0.381650 Objective Loss 0.381650 LR 0.001000 Time 0.022202 +2023-10-05 21:06:36,556 - Epoch: [52][ 420/ 1236] Overall Loss 0.380506 Objective Loss 0.380506 LR 0.001000 Time 0.022154 +2023-10-05 21:06:36,759 - Epoch: [52][ 430/ 1236] Overall Loss 0.379628 Objective Loss 0.379628 LR 0.001000 Time 0.022109 +2023-10-05 21:06:36,961 - Epoch: [52][ 440/ 1236] Overall Loss 0.379136 Objective Loss 0.379136 LR 0.001000 Time 0.022066 +2023-10-05 21:06:37,164 - Epoch: [52][ 450/ 1236] Overall Loss 0.378599 Objective Loss 0.378599 LR 0.001000 Time 0.022025 +2023-10-05 21:06:37,367 - Epoch: [52][ 460/ 1236] Overall Loss 0.378418 Objective Loss 0.378418 LR 0.001000 Time 0.021986 +2023-10-05 21:06:37,569 - Epoch: [52][ 470/ 1236] Overall Loss 0.377981 Objective Loss 0.377981 LR 0.001000 Time 0.021948 +2023-10-05 21:06:37,772 - Epoch: [52][ 480/ 1236] Overall Loss 0.377473 Objective Loss 0.377473 LR 0.001000 Time 0.021912 +2023-10-05 21:06:37,974 - Epoch: [52][ 490/ 1236] Overall Loss 0.376464 Objective Loss 0.376464 LR 0.001000 Time 0.021877 +2023-10-05 21:06:38,177 - Epoch: [52][ 500/ 1236] Overall Loss 0.375937 Objective Loss 0.375937 LR 0.001000 Time 0.021844 +2023-10-05 21:06:38,379 - Epoch: [52][ 510/ 1236] Overall Loss 0.375650 Objective Loss 0.375650 LR 0.001000 Time 0.021812 +2023-10-05 21:06:38,582 - Epoch: [52][ 520/ 1236] Overall Loss 0.374772 Objective Loss 0.374772 LR 0.001000 Time 0.021781 +2023-10-05 21:06:38,784 - Epoch: [52][ 530/ 1236] Overall Loss 0.374126 Objective Loss 0.374126 LR 0.001000 Time 0.021752 +2023-10-05 21:06:38,987 - Epoch: [52][ 540/ 1236] Overall Loss 0.373424 Objective Loss 0.373424 LR 0.001000 Time 0.021723 +2023-10-05 21:06:39,189 - Epoch: [52][ 550/ 1236] Overall Loss 0.372868 Objective Loss 0.372868 LR 0.001000 Time 0.021696 +2023-10-05 21:06:39,392 - Epoch: [52][ 560/ 1236] Overall Loss 0.372694 Objective Loss 0.372694 LR 0.001000 Time 0.021669 +2023-10-05 21:06:39,594 - Epoch: [52][ 570/ 1236] Overall Loss 0.372370 Objective Loss 0.372370 LR 0.001000 Time 0.021644 +2023-10-05 21:06:39,797 - Epoch: [52][ 580/ 1236] Overall Loss 0.371772 Objective Loss 0.371772 LR 0.001000 Time 0.021619 +2023-10-05 21:06:39,999 - Epoch: [52][ 590/ 1236] Overall Loss 0.371369 Objective Loss 0.371369 LR 0.001000 Time 0.021595 +2023-10-05 21:06:40,202 - Epoch: [52][ 600/ 1236] Overall Loss 0.371229 Objective Loss 0.371229 LR 0.001000 Time 0.021572 +2023-10-05 21:06:40,404 - Epoch: [52][ 610/ 1236] Overall Loss 0.370543 Objective Loss 0.370543 LR 0.001000 Time 0.021550 +2023-10-05 21:06:40,607 - Epoch: [52][ 620/ 1236] Overall Loss 0.370268 Objective Loss 0.370268 LR 0.001000 Time 0.021529 +2023-10-05 21:06:40,810 - Epoch: [52][ 630/ 1236] Overall Loss 0.370088 Objective Loss 0.370088 LR 0.001000 Time 0.021508 +2023-10-05 21:06:41,012 - Epoch: [52][ 640/ 1236] Overall Loss 0.370121 Objective Loss 0.370121 LR 0.001000 Time 0.021489 +2023-10-05 21:06:41,215 - Epoch: [52][ 650/ 1236] Overall Loss 0.370256 Objective Loss 0.370256 LR 0.001000 Time 0.021469 +2023-10-05 21:06:41,418 - Epoch: [52][ 660/ 1236] Overall Loss 0.370021 Objective Loss 0.370021 LR 0.001000 Time 0.021451 +2023-10-05 21:06:41,620 - Epoch: [52][ 670/ 1236] Overall Loss 0.370062 Objective Loss 0.370062 LR 0.001000 Time 0.021432 +2023-10-05 21:06:41,823 - Epoch: [52][ 680/ 1236] Overall Loss 0.369750 Objective Loss 0.369750 LR 0.001000 Time 0.021414 +2023-10-05 21:06:42,025 - Epoch: [52][ 690/ 1236] Overall Loss 0.369314 Objective Loss 0.369314 LR 0.001000 Time 0.021397 +2023-10-05 21:06:42,228 - Epoch: [52][ 700/ 1236] Overall Loss 0.368854 Objective Loss 0.368854 LR 0.001000 Time 0.021380 +2023-10-05 21:06:42,431 - Epoch: [52][ 710/ 1236] Overall Loss 0.368676 Objective Loss 0.368676 LR 0.001000 Time 0.021364 +2023-10-05 21:06:42,633 - Epoch: [52][ 720/ 1236] Overall Loss 0.368162 Objective Loss 0.368162 LR 0.001000 Time 0.021348 +2023-10-05 21:06:42,835 - Epoch: [52][ 730/ 1236] Overall Loss 0.368330 Objective Loss 0.368330 LR 0.001000 Time 0.021332 +2023-10-05 21:06:43,038 - Epoch: [52][ 740/ 1236] Overall Loss 0.367998 Objective Loss 0.367998 LR 0.001000 Time 0.021317 +2023-10-05 21:06:43,241 - Epoch: [52][ 750/ 1236] Overall Loss 0.367894 Objective Loss 0.367894 LR 0.001000 Time 0.021303 +2023-10-05 21:06:43,443 - Epoch: [52][ 760/ 1236] Overall Loss 0.367535 Objective Loss 0.367535 LR 0.001000 Time 0.021288 +2023-10-05 21:06:43,645 - Epoch: [52][ 770/ 1236] Overall Loss 0.367134 Objective Loss 0.367134 LR 0.001000 Time 0.021274 +2023-10-05 21:06:43,848 - Epoch: [52][ 780/ 1236] Overall Loss 0.366954 Objective Loss 0.366954 LR 0.001000 Time 0.021261 +2023-10-05 21:06:44,051 - Epoch: [52][ 790/ 1236] Overall Loss 0.366852 Objective Loss 0.366852 LR 0.001000 Time 0.021248 +2023-10-05 21:06:44,253 - Epoch: [52][ 800/ 1236] Overall Loss 0.366497 Objective Loss 0.366497 LR 0.001000 Time 0.021235 +2023-10-05 21:06:44,455 - Epoch: [52][ 810/ 1236] Overall Loss 0.365673 Objective Loss 0.365673 LR 0.001000 Time 0.021222 +2023-10-05 21:06:44,658 - Epoch: [52][ 820/ 1236] Overall Loss 0.365413 Objective Loss 0.365413 LR 0.001000 Time 0.021210 +2023-10-05 21:06:44,861 - Epoch: [52][ 830/ 1236] Overall Loss 0.365078 Objective Loss 0.365078 LR 0.001000 Time 0.021198 +2023-10-05 21:06:45,063 - Epoch: [52][ 840/ 1236] Overall Loss 0.364845 Objective Loss 0.364845 LR 0.001000 Time 0.021186 +2023-10-05 21:06:45,266 - Epoch: [52][ 850/ 1236] Overall Loss 0.364693 Objective Loss 0.364693 LR 0.001000 Time 0.021175 +2023-10-05 21:06:45,468 - Epoch: [52][ 860/ 1236] Overall Loss 0.364573 Objective Loss 0.364573 LR 0.001000 Time 0.021164 +2023-10-05 21:06:45,671 - Epoch: [52][ 870/ 1236] Overall Loss 0.364507 Objective Loss 0.364507 LR 0.001000 Time 0.021153 +2023-10-05 21:06:45,874 - Epoch: [52][ 880/ 1236] Overall Loss 0.364396 Objective Loss 0.364396 LR 0.001000 Time 0.021143 +2023-10-05 21:06:46,076 - Epoch: [52][ 890/ 1236] Overall Loss 0.364086 Objective Loss 0.364086 LR 0.001000 Time 0.021132 +2023-10-05 21:06:46,279 - Epoch: [52][ 900/ 1236] Overall Loss 0.363795 Objective Loss 0.363795 LR 0.001000 Time 0.021122 +2023-10-05 21:06:46,481 - Epoch: [52][ 910/ 1236] Overall Loss 0.363792 Objective Loss 0.363792 LR 0.001000 Time 0.021112 +2023-10-05 21:06:46,684 - Epoch: [52][ 920/ 1236] Overall Loss 0.363757 Objective Loss 0.363757 LR 0.001000 Time 0.021103 +2023-10-05 21:06:46,887 - Epoch: [52][ 930/ 1236] Overall Loss 0.363820 Objective Loss 0.363820 LR 0.001000 Time 0.021093 +2023-10-05 21:06:47,089 - Epoch: [52][ 940/ 1236] Overall Loss 0.363636 Objective Loss 0.363636 LR 0.001000 Time 0.021084 +2023-10-05 21:06:47,291 - Epoch: [52][ 950/ 1236] Overall Loss 0.363530 Objective Loss 0.363530 LR 0.001000 Time 0.021075 +2023-10-05 21:06:47,494 - Epoch: [52][ 960/ 1236] Overall Loss 0.363705 Objective Loss 0.363705 LR 0.001000 Time 0.021066 +2023-10-05 21:06:47,697 - Epoch: [52][ 970/ 1236] Overall Loss 0.363612 Objective Loss 0.363612 LR 0.001000 Time 0.021057 +2023-10-05 21:06:47,899 - Epoch: [52][ 980/ 1236] Overall Loss 0.363354 Objective Loss 0.363354 LR 0.001000 Time 0.021048 +2023-10-05 21:06:48,102 - Epoch: [52][ 990/ 1236] Overall Loss 0.363172 Objective Loss 0.363172 LR 0.001000 Time 0.021040 +2023-10-05 21:06:48,304 - Epoch: [52][ 1000/ 1236] Overall Loss 0.362967 Objective Loss 0.362967 LR 0.001000 Time 0.021032 +2023-10-05 21:06:48,507 - Epoch: [52][ 1010/ 1236] Overall Loss 0.362763 Objective Loss 0.362763 LR 0.001000 Time 0.021024 +2023-10-05 21:06:48,709 - Epoch: [52][ 1020/ 1236] Overall Loss 0.362614 Objective Loss 0.362614 LR 0.001000 Time 0.021016 +2023-10-05 21:06:48,911 - Epoch: [52][ 1030/ 1236] Overall Loss 0.362418 Objective Loss 0.362418 LR 0.001000 Time 0.021008 +2023-10-05 21:06:49,114 - Epoch: [52][ 1040/ 1236] Overall Loss 0.362237 Objective Loss 0.362237 LR 0.001000 Time 0.021000 +2023-10-05 21:06:49,317 - Epoch: [52][ 1050/ 1236] Overall Loss 0.362033 Objective Loss 0.362033 LR 0.001000 Time 0.020993 +2023-10-05 21:06:49,519 - Epoch: [52][ 1060/ 1236] Overall Loss 0.362075 Objective Loss 0.362075 LR 0.001000 Time 0.020986 +2023-10-05 21:06:49,722 - Epoch: [52][ 1070/ 1236] Overall Loss 0.361957 Objective Loss 0.361957 LR 0.001000 Time 0.020978 +2023-10-05 21:06:49,924 - Epoch: [52][ 1080/ 1236] Overall Loss 0.361796 Objective Loss 0.361796 LR 0.001000 Time 0.020971 +2023-10-05 21:06:50,127 - Epoch: [52][ 1090/ 1236] Overall Loss 0.361805 Objective Loss 0.361805 LR 0.001000 Time 0.020965 +2023-10-05 21:06:50,330 - Epoch: [52][ 1100/ 1236] Overall Loss 0.361790 Objective Loss 0.361790 LR 0.001000 Time 0.020958 +2023-10-05 21:06:50,532 - Epoch: [52][ 1110/ 1236] Overall Loss 0.361574 Objective Loss 0.361574 LR 0.001000 Time 0.020951 +2023-10-05 21:06:50,735 - Epoch: [52][ 1120/ 1236] Overall Loss 0.361666 Objective Loss 0.361666 LR 0.001000 Time 0.020945 +2023-10-05 21:06:50,937 - Epoch: [52][ 1130/ 1236] Overall Loss 0.361500 Objective Loss 0.361500 LR 0.001000 Time 0.020938 +2023-10-05 21:06:51,140 - Epoch: [52][ 1140/ 1236] Overall Loss 0.361186 Objective Loss 0.361186 LR 0.001000 Time 0.020932 +2023-10-05 21:06:51,342 - Epoch: [52][ 1150/ 1236] Overall Loss 0.361264 Objective Loss 0.361264 LR 0.001000 Time 0.020926 +2023-10-05 21:06:51,545 - Epoch: [52][ 1160/ 1236] Overall Loss 0.361141 Objective Loss 0.361141 LR 0.001000 Time 0.020920 +2023-10-05 21:06:51,748 - Epoch: [52][ 1170/ 1236] Overall Loss 0.361177 Objective Loss 0.361177 LR 0.001000 Time 0.020914 +2023-10-05 21:06:51,951 - Epoch: [52][ 1180/ 1236] Overall Loss 0.360982 Objective Loss 0.360982 LR 0.001000 Time 0.020909 +2023-10-05 21:06:52,153 - Epoch: [52][ 1190/ 1236] Overall Loss 0.361000 Objective Loss 0.361000 LR 0.001000 Time 0.020903 +2023-10-05 21:06:52,356 - Epoch: [52][ 1200/ 1236] Overall Loss 0.361065 Objective Loss 0.361065 LR 0.001000 Time 0.020897 +2023-10-05 21:06:52,559 - Epoch: [52][ 1210/ 1236] Overall Loss 0.360989 Objective Loss 0.360989 LR 0.001000 Time 0.020892 +2023-10-05 21:06:52,761 - Epoch: [52][ 1220/ 1236] Overall Loss 0.361029 Objective Loss 0.361029 LR 0.001000 Time 0.020886 +2023-10-05 21:06:53,017 - Epoch: [52][ 1230/ 1236] Overall Loss 0.361402 Objective Loss 0.361402 LR 0.001000 Time 0.020924 +2023-10-05 21:06:53,136 - Epoch: [52][ 1236/ 1236] Overall Loss 0.361189 Objective Loss 0.361189 Top1 81.262729 Top5 98.778004 LR 0.001000 Time 0.020919 +2023-10-05 21:06:53,264 - --- validate (epoch=52)----------- +2023-10-05 21:06:53,264 - 29943 samples (256 per mini-batch) +2023-10-05 21:06:53,723 - Epoch: [52][ 10/ 117] Loss 0.353040 Top1 80.117188 Top5 97.109375 +2023-10-05 21:06:53,875 - Epoch: [52][ 20/ 117] Loss 0.369364 Top1 79.550781 Top5 97.031250 +2023-10-05 21:06:54,025 - Epoch: [52][ 30/ 117] Loss 0.373655 Top1 79.531250 Top5 97.044271 +2023-10-05 21:06:54,177 - Epoch: [52][ 40/ 117] Loss 0.376769 Top1 79.589844 Top5 97.070312 +2023-10-05 21:06:54,327 - Epoch: [52][ 50/ 117] Loss 0.369797 Top1 79.648438 Top5 97.218750 +2023-10-05 21:06:54,477 - Epoch: [52][ 60/ 117] Loss 0.372327 Top1 79.550781 Top5 97.161458 +2023-10-05 21:06:54,625 - Epoch: [52][ 70/ 117] Loss 0.371071 Top1 79.748884 Top5 97.131696 +2023-10-05 21:06:54,779 - Epoch: [52][ 80/ 117] Loss 0.371093 Top1 79.907227 Top5 97.158203 +2023-10-05 21:06:54,932 - Epoch: [52][ 90/ 117] Loss 0.369716 Top1 79.895833 Top5 97.100694 +2023-10-05 21:06:55,081 - Epoch: [52][ 100/ 117] Loss 0.369011 Top1 79.894531 Top5 97.066406 +2023-10-05 21:06:55,236 - Epoch: [52][ 110/ 117] Loss 0.371103 Top1 79.960938 Top5 97.059659 +2023-10-05 21:06:55,322 - Epoch: [52][ 117/ 117] Loss 0.370946 Top1 79.958588 Top5 97.047724 +2023-10-05 21:06:55,461 - ==> Top1: 79.959 Top5: 97.048 Loss: 0.371 + +2023-10-05 21:06:55,462 - ==> Confusion: +[[ 924 4 3 3 7 1 0 0 6 68 1 0 1 5 6 3 5 3 0 0 10] + [ 1 1032 2 1 13 26 1 21 3 0 6 1 1 0 1 4 8 0 9 0 1] + [ 0 0 945 28 5 1 21 3 0 0 8 3 7 6 0 4 5 5 6 2 7] + [ 4 0 20 969 1 4 0 1 2 0 6 0 1 9 32 3 4 9 15 1 8] + [ 23 7 0 0 966 4 0 1 0 11 4 1 1 4 13 2 11 1 0 0 1] + [ 4 41 1 2 4 975 2 18 1 1 6 5 1 19 10 0 9 2 3 3 9] + [ 0 7 37 1 1 1 1104 10 0 0 8 1 1 0 1 2 0 4 1 7 5] + [ 2 23 23 0 2 38 3 1038 4 2 8 10 4 4 0 1 1 0 43 6 6] + [ 22 4 0 0 0 3 0 0 920 47 28 1 1 18 35 2 3 0 3 2 0] + [ 102 1 4 0 8 3 1 0 28 917 2 1 1 29 8 4 2 0 0 0 8] + [ 3 5 14 8 0 1 4 4 11 2 952 3 0 22 7 1 3 0 5 0 8] + [ 1 0 4 0 0 14 0 1 1 0 0 953 26 5 0 4 0 19 1 5 1] + [ 1 1 6 9 1 1 0 0 6 0 2 33 955 2 5 7 7 24 0 4 4] + [ 1 1 1 0 3 16 0 0 10 14 5 3 1 1049 2 1 2 1 0 0 9] + [ 18 1 3 9 4 1 0 0 11 5 0 0 0 2 1031 0 1 2 7 0 6] + [ 0 2 3 2 3 0 2 0 1 0 0 10 11 1 1 1055 23 13 0 6 1] + [ 1 8 3 0 11 6 0 0 2 0 0 3 1 1 2 7 1105 0 0 4 7] + [ 0 0 0 1 0 0 0 0 2 0 1 1 20 2 6 5 0 999 0 0 1] + [ 0 5 10 23 2 3 0 31 6 0 6 1 0 0 13 0 0 0 960 0 8] + [ 0 4 5 1 1 9 6 11 0 0 1 24 5 5 1 5 13 1 2 1054 4] + [ 162 245 251 114 147 200 55 78 101 99 266 113 509 382 276 72 277 116 150 253 4039]] + +2023-10-05 21:06:55,463 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:06:55,463 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:06:55,469 - + +2023-10-05 21:06:55,469 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:06:56,477 - Epoch: [53][ 10/ 1236] Overall Loss 0.357594 Objective Loss 0.357594 LR 0.001000 Time 0.100754 +2023-10-05 21:06:56,680 - Epoch: [53][ 20/ 1236] Overall Loss 0.351295 Objective Loss 0.351295 LR 0.001000 Time 0.060488 +2023-10-05 21:06:56,882 - Epoch: [53][ 30/ 1236] Overall Loss 0.344862 Objective Loss 0.344862 LR 0.001000 Time 0.047041 +2023-10-05 21:06:57,083 - Epoch: [53][ 40/ 1236] Overall Loss 0.346567 Objective Loss 0.346567 LR 0.001000 Time 0.040315 +2023-10-05 21:06:57,285 - Epoch: [53][ 50/ 1236] Overall Loss 0.343103 Objective Loss 0.343103 LR 0.001000 Time 0.036284 +2023-10-05 21:06:57,487 - Epoch: [53][ 60/ 1236] Overall Loss 0.345861 Objective Loss 0.345861 LR 0.001000 Time 0.033594 +2023-10-05 21:06:57,689 - Epoch: [53][ 70/ 1236] Overall Loss 0.347602 Objective Loss 0.347602 LR 0.001000 Time 0.031678 +2023-10-05 21:06:57,891 - Epoch: [53][ 80/ 1236] Overall Loss 0.349589 Objective Loss 0.349589 LR 0.001000 Time 0.030226 +2023-10-05 21:06:58,093 - Epoch: [53][ 90/ 1236] Overall Loss 0.345061 Objective Loss 0.345061 LR 0.001000 Time 0.029104 +2023-10-05 21:06:58,294 - Epoch: [53][ 100/ 1236] Overall Loss 0.343797 Objective Loss 0.343797 LR 0.001000 Time 0.028204 +2023-10-05 21:06:58,507 - Epoch: [53][ 110/ 1236] Overall Loss 0.343240 Objective Loss 0.343240 LR 0.001000 Time 0.027571 +2023-10-05 21:06:58,724 - Epoch: [53][ 120/ 1236] Overall Loss 0.344174 Objective Loss 0.344174 LR 0.001000 Time 0.027073 +2023-10-05 21:06:58,937 - Epoch: [53][ 130/ 1236] Overall Loss 0.342514 Objective Loss 0.342514 LR 0.001000 Time 0.026626 +2023-10-05 21:06:59,152 - Epoch: [53][ 140/ 1236] Overall Loss 0.340155 Objective Loss 0.340155 LR 0.001000 Time 0.026254 +2023-10-05 21:06:59,360 - Epoch: [53][ 150/ 1236] Overall Loss 0.340262 Objective Loss 0.340262 LR 0.001000 Time 0.025891 +2023-10-05 21:06:59,572 - Epoch: [53][ 160/ 1236] Overall Loss 0.339556 Objective Loss 0.339556 LR 0.001000 Time 0.025594 +2023-10-05 21:06:59,780 - Epoch: [53][ 170/ 1236] Overall Loss 0.339177 Objective Loss 0.339177 LR 0.001000 Time 0.025311 +2023-10-05 21:06:59,992 - Epoch: [53][ 180/ 1236] Overall Loss 0.338821 Objective Loss 0.338821 LR 0.001000 Time 0.025080 +2023-10-05 21:07:00,200 - Epoch: [53][ 190/ 1236] Overall Loss 0.338757 Objective Loss 0.338757 LR 0.001000 Time 0.024856 +2023-10-05 21:07:00,412 - Epoch: [53][ 200/ 1236] Overall Loss 0.338797 Objective Loss 0.338797 LR 0.001000 Time 0.024671 +2023-10-05 21:07:00,621 - Epoch: [53][ 210/ 1236] Overall Loss 0.337766 Objective Loss 0.337766 LR 0.001000 Time 0.024488 +2023-10-05 21:07:00,833 - Epoch: [53][ 220/ 1236] Overall Loss 0.337569 Objective Loss 0.337569 LR 0.001000 Time 0.024337 +2023-10-05 21:07:01,041 - Epoch: [53][ 230/ 1236] Overall Loss 0.337296 Objective Loss 0.337296 LR 0.001000 Time 0.024183 +2023-10-05 21:07:01,253 - Epoch: [53][ 240/ 1236] Overall Loss 0.338214 Objective Loss 0.338214 LR 0.001000 Time 0.024056 +2023-10-05 21:07:01,461 - Epoch: [53][ 250/ 1236] Overall Loss 0.338507 Objective Loss 0.338507 LR 0.001000 Time 0.023925 +2023-10-05 21:07:01,666 - Epoch: [53][ 260/ 1236] Overall Loss 0.337106 Objective Loss 0.337106 LR 0.001000 Time 0.023794 +2023-10-05 21:07:01,869 - Epoch: [53][ 270/ 1236] Overall Loss 0.337149 Objective Loss 0.337149 LR 0.001000 Time 0.023661 +2023-10-05 21:07:02,071 - Epoch: [53][ 280/ 1236] Overall Loss 0.336287 Objective Loss 0.336287 LR 0.001000 Time 0.023538 +2023-10-05 21:07:02,274 - Epoch: [53][ 290/ 1236] Overall Loss 0.335953 Objective Loss 0.335953 LR 0.001000 Time 0.023423 +2023-10-05 21:07:02,476 - Epoch: [53][ 300/ 1236] Overall Loss 0.335561 Objective Loss 0.335561 LR 0.001000 Time 0.023316 +2023-10-05 21:07:02,679 - Epoch: [53][ 310/ 1236] Overall Loss 0.335732 Objective Loss 0.335732 LR 0.001000 Time 0.023216 +2023-10-05 21:07:02,881 - Epoch: [53][ 320/ 1236] Overall Loss 0.335661 Objective Loss 0.335661 LR 0.001000 Time 0.023123 +2023-10-05 21:07:03,084 - Epoch: [53][ 330/ 1236] Overall Loss 0.334919 Objective Loss 0.334919 LR 0.001000 Time 0.023034 +2023-10-05 21:07:03,286 - Epoch: [53][ 340/ 1236] Overall Loss 0.334243 Objective Loss 0.334243 LR 0.001000 Time 0.022952 +2023-10-05 21:07:03,489 - Epoch: [53][ 350/ 1236] Overall Loss 0.333666 Objective Loss 0.333666 LR 0.001000 Time 0.022875 +2023-10-05 21:07:03,694 - Epoch: [53][ 360/ 1236] Overall Loss 0.333387 Objective Loss 0.333387 LR 0.001000 Time 0.022806 +2023-10-05 21:07:03,897 - Epoch: [53][ 370/ 1236] Overall Loss 0.334157 Objective Loss 0.334157 LR 0.001000 Time 0.022740 +2023-10-05 21:07:04,101 - Epoch: [53][ 380/ 1236] Overall Loss 0.333955 Objective Loss 0.333955 LR 0.001000 Time 0.022676 +2023-10-05 21:07:04,305 - Epoch: [53][ 390/ 1236] Overall Loss 0.333821 Objective Loss 0.333821 LR 0.001000 Time 0.022616 +2023-10-05 21:07:04,508 - Epoch: [53][ 400/ 1236] Overall Loss 0.334343 Objective Loss 0.334343 LR 0.001000 Time 0.022559 +2023-10-05 21:07:04,712 - Epoch: [53][ 410/ 1236] Overall Loss 0.333947 Objective Loss 0.333947 LR 0.001000 Time 0.022504 +2023-10-05 21:07:04,915 - Epoch: [53][ 420/ 1236] Overall Loss 0.334465 Objective Loss 0.334465 LR 0.001000 Time 0.022452 +2023-10-05 21:07:05,119 - Epoch: [53][ 430/ 1236] Overall Loss 0.334731 Objective Loss 0.334731 LR 0.001000 Time 0.022402 +2023-10-05 21:07:05,323 - Epoch: [53][ 440/ 1236] Overall Loss 0.334746 Objective Loss 0.334746 LR 0.001000 Time 0.022355 +2023-10-05 21:07:05,526 - Epoch: [53][ 450/ 1236] Overall Loss 0.334845 Objective Loss 0.334845 LR 0.001000 Time 0.022310 +2023-10-05 21:07:05,730 - Epoch: [53][ 460/ 1236] Overall Loss 0.335109 Objective Loss 0.335109 LR 0.001000 Time 0.022267 +2023-10-05 21:07:05,934 - Epoch: [53][ 470/ 1236] Overall Loss 0.335277 Objective Loss 0.335277 LR 0.001000 Time 0.022226 +2023-10-05 21:07:06,137 - Epoch: [53][ 480/ 1236] Overall Loss 0.335639 Objective Loss 0.335639 LR 0.001000 Time 0.022186 +2023-10-05 21:07:06,341 - Epoch: [53][ 490/ 1236] Overall Loss 0.335681 Objective Loss 0.335681 LR 0.001000 Time 0.022149 +2023-10-05 21:07:06,545 - Epoch: [53][ 500/ 1236] Overall Loss 0.336022 Objective Loss 0.336022 LR 0.001000 Time 0.022113 +2023-10-05 21:07:06,748 - Epoch: [53][ 510/ 1236] Overall Loss 0.336778 Objective Loss 0.336778 LR 0.001000 Time 0.022078 +2023-10-05 21:07:06,952 - Epoch: [53][ 520/ 1236] Overall Loss 0.336711 Objective Loss 0.336711 LR 0.001000 Time 0.022045 +2023-10-05 21:07:07,156 - Epoch: [53][ 530/ 1236] Overall Loss 0.336792 Objective Loss 0.336792 LR 0.001000 Time 0.022012 +2023-10-05 21:07:07,360 - Epoch: [53][ 540/ 1236] Overall Loss 0.336540 Objective Loss 0.336540 LR 0.001000 Time 0.021982 +2023-10-05 21:07:07,564 - Epoch: [53][ 550/ 1236] Overall Loss 0.336012 Objective Loss 0.336012 LR 0.001000 Time 0.021952 +2023-10-05 21:07:07,768 - Epoch: [53][ 560/ 1236] Overall Loss 0.336004 Objective Loss 0.336004 LR 0.001000 Time 0.021923 +2023-10-05 21:07:07,971 - Epoch: [53][ 570/ 1236] Overall Loss 0.335885 Objective Loss 0.335885 LR 0.001000 Time 0.021895 +2023-10-05 21:07:08,175 - Epoch: [53][ 580/ 1236] Overall Loss 0.335690 Objective Loss 0.335690 LR 0.001000 Time 0.021868 +2023-10-05 21:07:08,378 - Epoch: [53][ 590/ 1236] Overall Loss 0.335711 Objective Loss 0.335711 LR 0.001000 Time 0.021842 +2023-10-05 21:07:08,582 - Epoch: [53][ 600/ 1236] Overall Loss 0.335854 Objective Loss 0.335854 LR 0.001000 Time 0.021817 +2023-10-05 21:07:08,786 - Epoch: [53][ 610/ 1236] Overall Loss 0.336218 Objective Loss 0.336218 LR 0.001000 Time 0.021793 +2023-10-05 21:07:08,990 - Epoch: [53][ 620/ 1236] Overall Loss 0.336357 Objective Loss 0.336357 LR 0.001000 Time 0.021769 +2023-10-05 21:07:09,194 - Epoch: [53][ 630/ 1236] Overall Loss 0.336293 Objective Loss 0.336293 LR 0.001000 Time 0.021747 +2023-10-05 21:07:09,397 - Epoch: [53][ 640/ 1236] Overall Loss 0.336408 Objective Loss 0.336408 LR 0.001000 Time 0.021725 +2023-10-05 21:07:09,601 - Epoch: [53][ 650/ 1236] Overall Loss 0.336155 Objective Loss 0.336155 LR 0.001000 Time 0.021703 +2023-10-05 21:07:09,804 - Epoch: [53][ 660/ 1236] Overall Loss 0.335926 Objective Loss 0.335926 LR 0.001000 Time 0.021682 +2023-10-05 21:07:10,008 - Epoch: [53][ 670/ 1236] Overall Loss 0.336110 Objective Loss 0.336110 LR 0.001000 Time 0.021662 +2023-10-05 21:07:10,212 - Epoch: [53][ 680/ 1236] Overall Loss 0.336409 Objective Loss 0.336409 LR 0.001000 Time 0.021642 +2023-10-05 21:07:10,416 - Epoch: [53][ 690/ 1236] Overall Loss 0.336375 Objective Loss 0.336375 LR 0.001000 Time 0.021624 +2023-10-05 21:07:10,619 - Epoch: [53][ 700/ 1236] Overall Loss 0.336627 Objective Loss 0.336627 LR 0.001000 Time 0.021605 +2023-10-05 21:07:10,823 - Epoch: [53][ 710/ 1236] Overall Loss 0.337050 Objective Loss 0.337050 LR 0.001000 Time 0.021587 +2023-10-05 21:07:11,027 - Epoch: [53][ 720/ 1236] Overall Loss 0.337754 Objective Loss 0.337754 LR 0.001000 Time 0.021570 +2023-10-05 21:07:11,231 - Epoch: [53][ 730/ 1236] Overall Loss 0.337723 Objective Loss 0.337723 LR 0.001000 Time 0.021554 +2023-10-05 21:07:11,434 - Epoch: [53][ 740/ 1236] Overall Loss 0.337824 Objective Loss 0.337824 LR 0.001000 Time 0.021537 +2023-10-05 21:07:11,638 - Epoch: [53][ 750/ 1236] Overall Loss 0.337722 Objective Loss 0.337722 LR 0.001000 Time 0.021521 +2023-10-05 21:07:11,842 - Epoch: [53][ 760/ 1236] Overall Loss 0.338136 Objective Loss 0.338136 LR 0.001000 Time 0.021506 +2023-10-05 21:07:12,047 - Epoch: [53][ 770/ 1236] Overall Loss 0.338059 Objective Loss 0.338059 LR 0.001000 Time 0.021492 +2023-10-05 21:07:12,250 - Epoch: [53][ 780/ 1236] Overall Loss 0.337820 Objective Loss 0.337820 LR 0.001000 Time 0.021477 +2023-10-05 21:07:12,454 - Epoch: [53][ 790/ 1236] Overall Loss 0.337788 Objective Loss 0.337788 LR 0.001000 Time 0.021462 +2023-10-05 21:07:12,658 - Epoch: [53][ 800/ 1236] Overall Loss 0.337546 Objective Loss 0.337546 LR 0.001000 Time 0.021448 +2023-10-05 21:07:12,862 - Epoch: [53][ 810/ 1236] Overall Loss 0.338056 Objective Loss 0.338056 LR 0.001000 Time 0.021435 +2023-10-05 21:07:13,065 - Epoch: [53][ 820/ 1236] Overall Loss 0.338225 Objective Loss 0.338225 LR 0.001000 Time 0.021421 +2023-10-05 21:07:13,269 - Epoch: [53][ 830/ 1236] Overall Loss 0.338224 Objective Loss 0.338224 LR 0.001000 Time 0.021408 +2023-10-05 21:07:13,473 - Epoch: [53][ 840/ 1236] Overall Loss 0.338333 Objective Loss 0.338333 LR 0.001000 Time 0.021395 +2023-10-05 21:07:13,676 - Epoch: [53][ 850/ 1236] Overall Loss 0.338183 Objective Loss 0.338183 LR 0.001000 Time 0.021383 +2023-10-05 21:07:13,880 - Epoch: [53][ 860/ 1236] Overall Loss 0.338002 Objective Loss 0.338002 LR 0.001000 Time 0.021370 +2023-10-05 21:07:14,084 - Epoch: [53][ 870/ 1236] Overall Loss 0.338273 Objective Loss 0.338273 LR 0.001000 Time 0.021359 +2023-10-05 21:07:14,288 - Epoch: [53][ 880/ 1236] Overall Loss 0.338015 Objective Loss 0.338015 LR 0.001000 Time 0.021348 +2023-10-05 21:07:14,492 - Epoch: [53][ 890/ 1236] Overall Loss 0.338115 Objective Loss 0.338115 LR 0.001000 Time 0.021336 +2023-10-05 21:07:14,696 - Epoch: [53][ 900/ 1236] Overall Loss 0.338154 Objective Loss 0.338154 LR 0.001000 Time 0.021325 +2023-10-05 21:07:14,899 - Epoch: [53][ 910/ 1236] Overall Loss 0.338072 Objective Loss 0.338072 LR 0.001000 Time 0.021314 +2023-10-05 21:07:15,103 - Epoch: [53][ 920/ 1236] Overall Loss 0.338007 Objective Loss 0.338007 LR 0.001000 Time 0.021304 +2023-10-05 21:07:15,307 - Epoch: [53][ 930/ 1236] Overall Loss 0.337992 Objective Loss 0.337992 LR 0.001000 Time 0.021293 +2023-10-05 21:07:15,510 - Epoch: [53][ 940/ 1236] Overall Loss 0.338172 Objective Loss 0.338172 LR 0.001000 Time 0.021283 +2023-10-05 21:07:15,714 - Epoch: [53][ 950/ 1236] Overall Loss 0.338195 Objective Loss 0.338195 LR 0.001000 Time 0.021273 +2023-10-05 21:07:15,918 - Epoch: [53][ 960/ 1236] Overall Loss 0.338257 Objective Loss 0.338257 LR 0.001000 Time 0.021263 +2023-10-05 21:07:16,122 - Epoch: [53][ 970/ 1236] Overall Loss 0.338352 Objective Loss 0.338352 LR 0.001000 Time 0.021254 +2023-10-05 21:07:16,325 - Epoch: [53][ 980/ 1236] Overall Loss 0.338403 Objective Loss 0.338403 LR 0.001000 Time 0.021245 +2023-10-05 21:07:16,529 - Epoch: [53][ 990/ 1236] Overall Loss 0.338669 Objective Loss 0.338669 LR 0.001000 Time 0.021235 +2023-10-05 21:07:16,733 - Epoch: [53][ 1000/ 1236] Overall Loss 0.338820 Objective Loss 0.338820 LR 0.001000 Time 0.021226 +2023-10-05 21:07:16,936 - Epoch: [53][ 1010/ 1236] Overall Loss 0.338834 Objective Loss 0.338834 LR 0.001000 Time 0.021218 +2023-10-05 21:07:17,140 - Epoch: [53][ 1020/ 1236] Overall Loss 0.339103 Objective Loss 0.339103 LR 0.001000 Time 0.021209 +2023-10-05 21:07:17,344 - Epoch: [53][ 1030/ 1236] Overall Loss 0.339021 Objective Loss 0.339021 LR 0.001000 Time 0.021200 +2023-10-05 21:07:17,547 - Epoch: [53][ 1040/ 1236] Overall Loss 0.339007 Objective Loss 0.339007 LR 0.001000 Time 0.021192 +2023-10-05 21:07:17,751 - Epoch: [53][ 1050/ 1236] Overall Loss 0.339064 Objective Loss 0.339064 LR 0.001000 Time 0.021184 +2023-10-05 21:07:17,955 - Epoch: [53][ 1060/ 1236] Overall Loss 0.339355 Objective Loss 0.339355 LR 0.001000 Time 0.021176 +2023-10-05 21:07:18,159 - Epoch: [53][ 1070/ 1236] Overall Loss 0.339183 Objective Loss 0.339183 LR 0.001000 Time 0.021168 +2023-10-05 21:07:18,362 - Epoch: [53][ 1080/ 1236] Overall Loss 0.339355 Objective Loss 0.339355 LR 0.001000 Time 0.021160 +2023-10-05 21:07:18,566 - Epoch: [53][ 1090/ 1236] Overall Loss 0.339167 Objective Loss 0.339167 LR 0.001000 Time 0.021153 +2023-10-05 21:07:18,770 - Epoch: [53][ 1100/ 1236] Overall Loss 0.338906 Objective Loss 0.338906 LR 0.001000 Time 0.021146 +2023-10-05 21:07:18,974 - Epoch: [53][ 1110/ 1236] Overall Loss 0.338886 Objective Loss 0.338886 LR 0.001000 Time 0.021139 +2023-10-05 21:07:19,178 - Epoch: [53][ 1120/ 1236] Overall Loss 0.338994 Objective Loss 0.338994 LR 0.001000 Time 0.021132 +2023-10-05 21:07:19,382 - Epoch: [53][ 1130/ 1236] Overall Loss 0.339182 Objective Loss 0.339182 LR 0.001000 Time 0.021125 +2023-10-05 21:07:19,586 - Epoch: [53][ 1140/ 1236] Overall Loss 0.339043 Objective Loss 0.339043 LR 0.001000 Time 0.021118 +2023-10-05 21:07:19,790 - Epoch: [53][ 1150/ 1236] Overall Loss 0.338927 Objective Loss 0.338927 LR 0.001000 Time 0.021111 +2023-10-05 21:07:19,993 - Epoch: [53][ 1160/ 1236] Overall Loss 0.338909 Objective Loss 0.338909 LR 0.001000 Time 0.021105 +2023-10-05 21:07:20,197 - Epoch: [53][ 1170/ 1236] Overall Loss 0.338946 Objective Loss 0.338946 LR 0.001000 Time 0.021098 +2023-10-05 21:07:20,401 - Epoch: [53][ 1180/ 1236] Overall Loss 0.338825 Objective Loss 0.338825 LR 0.001000 Time 0.021092 +2023-10-05 21:07:20,605 - Epoch: [53][ 1190/ 1236] Overall Loss 0.338827 Objective Loss 0.338827 LR 0.001000 Time 0.021086 +2023-10-05 21:07:20,809 - Epoch: [53][ 1200/ 1236] Overall Loss 0.338810 Objective Loss 0.338810 LR 0.001000 Time 0.021079 +2023-10-05 21:07:21,012 - Epoch: [53][ 1210/ 1236] Overall Loss 0.338639 Objective Loss 0.338639 LR 0.001000 Time 0.021073 +2023-10-05 21:07:21,216 - Epoch: [53][ 1220/ 1236] Overall Loss 0.338600 Objective Loss 0.338600 LR 0.001000 Time 0.021067 +2023-10-05 21:07:21,477 - Epoch: [53][ 1230/ 1236] Overall Loss 0.338561 Objective Loss 0.338561 LR 0.001000 Time 0.021108 +2023-10-05 21:07:21,595 - Epoch: [53][ 1236/ 1236] Overall Loss 0.338357 Objective Loss 0.338357 Top1 84.725051 Top5 97.556008 LR 0.001000 Time 0.021100 +2023-10-05 21:07:21,726 - --- validate (epoch=53)----------- +2023-10-05 21:07:21,727 - 29943 samples (256 per mini-batch) +2023-10-05 21:07:22,178 - Epoch: [53][ 10/ 117] Loss 0.402307 Top1 80.937500 Top5 97.304688 +2023-10-05 21:07:22,327 - Epoch: [53][ 20/ 117] Loss 0.373271 Top1 82.011719 Top5 97.617188 +2023-10-05 21:07:22,474 - Epoch: [53][ 30/ 117] Loss 0.369717 Top1 82.096354 Top5 97.656250 +2023-10-05 21:07:22,622 - Epoch: [53][ 40/ 117] Loss 0.369899 Top1 82.119141 Top5 97.617188 +2023-10-05 21:07:22,771 - Epoch: [53][ 50/ 117] Loss 0.363745 Top1 82.312500 Top5 97.695312 +2023-10-05 21:07:22,920 - Epoch: [53][ 60/ 117] Loss 0.363039 Top1 82.115885 Top5 97.714844 +2023-10-05 21:07:23,069 - Epoch: [53][ 70/ 117] Loss 0.365855 Top1 82.064732 Top5 97.689732 +2023-10-05 21:07:23,216 - Epoch: [53][ 80/ 117] Loss 0.365308 Top1 82.065430 Top5 97.700195 +2023-10-05 21:07:23,364 - Epoch: [53][ 90/ 117] Loss 0.363785 Top1 82.122396 Top5 97.738715 +2023-10-05 21:07:23,512 - Epoch: [53][ 100/ 117] Loss 0.362744 Top1 82.066406 Top5 97.730469 +2023-10-05 21:07:23,666 - Epoch: [53][ 110/ 117] Loss 0.361365 Top1 82.080966 Top5 97.666903 +2023-10-05 21:07:23,751 - Epoch: [53][ 117/ 117] Loss 0.361532 Top1 82.082624 Top5 97.675584 +2023-10-05 21:07:23,881 - ==> Top1: 82.083 Top5: 97.676 Loss: 0.362 + +2023-10-05 21:07:23,881 - ==> Confusion: +[[ 924 2 7 3 8 5 0 1 5 58 0 1 1 5 9 1 4 1 0 0 15] + [ 1 1026 1 1 8 27 1 31 6 0 3 4 2 1 1 2 3 0 8 0 5] + [ 1 1 934 19 2 1 33 9 0 1 5 3 7 6 1 3 2 1 4 5 18] + [ 2 1 23 961 0 6 0 0 2 1 4 0 5 4 28 4 2 7 17 4 18] + [ 32 7 1 0 957 4 0 1 0 7 0 0 1 1 8 4 16 1 1 3 6] + [ 4 28 2 0 3 988 1 18 0 0 5 5 2 17 6 1 7 0 5 5 19] + [ 0 8 28 0 0 1 1114 11 0 0 4 1 2 0 1 5 0 1 1 6 8] + [ 3 12 18 0 1 51 3 1058 1 5 2 7 0 2 0 1 0 1 36 8 9] + [ 15 4 1 1 0 2 1 2 951 33 15 0 1 24 26 1 2 0 5 3 2] + [ 95 1 2 1 8 3 0 0 33 924 0 1 1 27 6 0 0 0 1 5 11] + [ 4 5 9 10 1 1 3 7 17 0 950 0 0 19 3 1 2 1 6 2 12] + [ 0 0 2 0 0 13 1 3 0 0 0 947 35 4 0 4 0 18 0 5 3] + [ 0 0 5 5 0 2 0 2 2 0 1 32 966 3 1 10 2 17 3 2 15] + [ 1 0 0 1 3 10 0 0 9 14 6 4 1 1047 2 1 0 1 0 1 18] + [ 13 0 4 14 3 0 0 0 15 4 3 1 2 1 1013 0 1 1 10 0 16] + [ 1 0 6 2 1 0 1 0 0 0 0 7 13 2 1 1049 19 10 1 11 10] + [ 1 13 2 0 5 7 0 0 1 0 1 4 2 2 4 13 1087 0 1 2 16] + [ 0 0 0 2 0 0 2 0 1 0 0 7 12 1 5 12 0 988 1 0 7] + [ 0 7 4 19 1 0 2 44 2 1 2 0 2 0 12 0 1 0 959 2 10] + [ 0 4 4 1 1 3 6 16 0 0 3 13 5 4 0 5 6 0 2 1064 15] + [ 119 183 212 74 108 184 56 123 94 84 170 120 427 355 177 62 162 83 190 251 4671]] + +2023-10-05 21:07:23,883 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:07:23,883 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:07:23,888 - + +2023-10-05 21:07:23,888 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:07:24,875 - Epoch: [54][ 10/ 1236] Overall Loss 0.343780 Objective Loss 0.343780 LR 0.001000 Time 0.098631 +2023-10-05 21:07:25,076 - Epoch: [54][ 20/ 1236] Overall Loss 0.330039 Objective Loss 0.330039 LR 0.001000 Time 0.059339 +2023-10-05 21:07:25,275 - Epoch: [54][ 30/ 1236] Overall Loss 0.343025 Objective Loss 0.343025 LR 0.001000 Time 0.046190 +2023-10-05 21:07:25,476 - Epoch: [54][ 40/ 1236] Overall Loss 0.340193 Objective Loss 0.340193 LR 0.001000 Time 0.039660 +2023-10-05 21:07:25,675 - Epoch: [54][ 50/ 1236] Overall Loss 0.341924 Objective Loss 0.341924 LR 0.001000 Time 0.035701 +2023-10-05 21:07:25,877 - Epoch: [54][ 60/ 1236] Overall Loss 0.337024 Objective Loss 0.337024 LR 0.001000 Time 0.033105 +2023-10-05 21:07:26,076 - Epoch: [54][ 70/ 1236] Overall Loss 0.333333 Objective Loss 0.333333 LR 0.001000 Time 0.031223 +2023-10-05 21:07:26,278 - Epoch: [54][ 80/ 1236] Overall Loss 0.334911 Objective Loss 0.334911 LR 0.001000 Time 0.029832 +2023-10-05 21:07:26,477 - Epoch: [54][ 90/ 1236] Overall Loss 0.334787 Objective Loss 0.334787 LR 0.001000 Time 0.028731 +2023-10-05 21:07:26,679 - Epoch: [54][ 100/ 1236] Overall Loss 0.335927 Objective Loss 0.335927 LR 0.001000 Time 0.027877 +2023-10-05 21:07:26,879 - Epoch: [54][ 110/ 1236] Overall Loss 0.337788 Objective Loss 0.337788 LR 0.001000 Time 0.027155 +2023-10-05 21:07:27,080 - Epoch: [54][ 120/ 1236] Overall Loss 0.337372 Objective Loss 0.337372 LR 0.001000 Time 0.026563 +2023-10-05 21:07:27,280 - Epoch: [54][ 130/ 1236] Overall Loss 0.337412 Objective Loss 0.337412 LR 0.001000 Time 0.026055 +2023-10-05 21:07:27,481 - Epoch: [54][ 140/ 1236] Overall Loss 0.335626 Objective Loss 0.335626 LR 0.001000 Time 0.025627 +2023-10-05 21:07:27,680 - Epoch: [54][ 150/ 1236] Overall Loss 0.335213 Objective Loss 0.335213 LR 0.001000 Time 0.025245 +2023-10-05 21:07:27,881 - Epoch: [54][ 160/ 1236] Overall Loss 0.335422 Objective Loss 0.335422 LR 0.001000 Time 0.024920 +2023-10-05 21:07:28,081 - Epoch: [54][ 170/ 1236] Overall Loss 0.336429 Objective Loss 0.336429 LR 0.001000 Time 0.024629 +2023-10-05 21:07:28,282 - Epoch: [54][ 180/ 1236] Overall Loss 0.338109 Objective Loss 0.338109 LR 0.001000 Time 0.024375 +2023-10-05 21:07:28,481 - Epoch: [54][ 190/ 1236] Overall Loss 0.336827 Objective Loss 0.336827 LR 0.001000 Time 0.024139 +2023-10-05 21:07:28,682 - Epoch: [54][ 200/ 1236] Overall Loss 0.336344 Objective Loss 0.336344 LR 0.001000 Time 0.023935 +2023-10-05 21:07:28,882 - Epoch: [54][ 210/ 1236] Overall Loss 0.336654 Objective Loss 0.336654 LR 0.001000 Time 0.023745 +2023-10-05 21:07:29,083 - Epoch: [54][ 220/ 1236] Overall Loss 0.335796 Objective Loss 0.335796 LR 0.001000 Time 0.023578 +2023-10-05 21:07:29,283 - Epoch: [54][ 230/ 1236] Overall Loss 0.336476 Objective Loss 0.336476 LR 0.001000 Time 0.023419 +2023-10-05 21:07:29,484 - Epoch: [54][ 240/ 1236] Overall Loss 0.336912 Objective Loss 0.336912 LR 0.001000 Time 0.023281 +2023-10-05 21:07:29,683 - Epoch: [54][ 250/ 1236] Overall Loss 0.337341 Objective Loss 0.337341 LR 0.001000 Time 0.023147 +2023-10-05 21:07:29,885 - Epoch: [54][ 260/ 1236] Overall Loss 0.336053 Objective Loss 0.336053 LR 0.001000 Time 0.023030 +2023-10-05 21:07:30,085 - Epoch: [54][ 270/ 1236] Overall Loss 0.335467 Objective Loss 0.335467 LR 0.001000 Time 0.022917 +2023-10-05 21:07:30,286 - Epoch: [54][ 280/ 1236] Overall Loss 0.336514 Objective Loss 0.336514 LR 0.001000 Time 0.022815 +2023-10-05 21:07:30,485 - Epoch: [54][ 290/ 1236] Overall Loss 0.337102 Objective Loss 0.337102 LR 0.001000 Time 0.022715 +2023-10-05 21:07:30,686 - Epoch: [54][ 300/ 1236] Overall Loss 0.336951 Objective Loss 0.336951 LR 0.001000 Time 0.022625 +2023-10-05 21:07:30,885 - Epoch: [54][ 310/ 1236] Overall Loss 0.337476 Objective Loss 0.337476 LR 0.001000 Time 0.022537 +2023-10-05 21:07:31,086 - Epoch: [54][ 320/ 1236] Overall Loss 0.337018 Objective Loss 0.337018 LR 0.001000 Time 0.022461 +2023-10-05 21:07:31,287 - Epoch: [54][ 330/ 1236] Overall Loss 0.336166 Objective Loss 0.336166 LR 0.001000 Time 0.022386 +2023-10-05 21:07:31,486 - Epoch: [54][ 340/ 1236] Overall Loss 0.336132 Objective Loss 0.336132 LR 0.001000 Time 0.022314 +2023-10-05 21:07:31,684 - Epoch: [54][ 350/ 1236] Overall Loss 0.336623 Objective Loss 0.336623 LR 0.001000 Time 0.022241 +2023-10-05 21:07:31,885 - Epoch: [54][ 360/ 1236] Overall Loss 0.336825 Objective Loss 0.336825 LR 0.001000 Time 0.022179 +2023-10-05 21:07:32,084 - Epoch: [54][ 370/ 1236] Overall Loss 0.336580 Objective Loss 0.336580 LR 0.001000 Time 0.022117 +2023-10-05 21:07:32,284 - Epoch: [54][ 380/ 1236] Overall Loss 0.336258 Objective Loss 0.336258 LR 0.001000 Time 0.022060 +2023-10-05 21:07:32,483 - Epoch: [54][ 390/ 1236] Overall Loss 0.336020 Objective Loss 0.336020 LR 0.001000 Time 0.022004 +2023-10-05 21:07:32,683 - Epoch: [54][ 400/ 1236] Overall Loss 0.336022 Objective Loss 0.336022 LR 0.001000 Time 0.021954 +2023-10-05 21:07:32,882 - Epoch: [54][ 410/ 1236] Overall Loss 0.335731 Objective Loss 0.335731 LR 0.001000 Time 0.021902 +2023-10-05 21:07:33,082 - Epoch: [54][ 420/ 1236] Overall Loss 0.335909 Objective Loss 0.335909 LR 0.001000 Time 0.021856 +2023-10-05 21:07:33,281 - Epoch: [54][ 430/ 1236] Overall Loss 0.335498 Objective Loss 0.335498 LR 0.001000 Time 0.021809 +2023-10-05 21:07:33,481 - Epoch: [54][ 440/ 1236] Overall Loss 0.335809 Objective Loss 0.335809 LR 0.001000 Time 0.021767 +2023-10-05 21:07:33,680 - Epoch: [54][ 450/ 1236] Overall Loss 0.335928 Objective Loss 0.335928 LR 0.001000 Time 0.021724 +2023-10-05 21:07:33,880 - Epoch: [54][ 460/ 1236] Overall Loss 0.336179 Objective Loss 0.336179 LR 0.001000 Time 0.021687 +2023-10-05 21:07:34,079 - Epoch: [54][ 470/ 1236] Overall Loss 0.336149 Objective Loss 0.336149 LR 0.001000 Time 0.021648 +2023-10-05 21:07:34,279 - Epoch: [54][ 480/ 1236] Overall Loss 0.335921 Objective Loss 0.335921 LR 0.001000 Time 0.021613 +2023-10-05 21:07:34,478 - Epoch: [54][ 490/ 1236] Overall Loss 0.335306 Objective Loss 0.335306 LR 0.001000 Time 0.021577 +2023-10-05 21:07:34,678 - Epoch: [54][ 500/ 1236] Overall Loss 0.334729 Objective Loss 0.334729 LR 0.001000 Time 0.021545 +2023-10-05 21:07:34,877 - Epoch: [54][ 510/ 1236] Overall Loss 0.334617 Objective Loss 0.334617 LR 0.001000 Time 0.021512 +2023-10-05 21:07:35,077 - Epoch: [54][ 520/ 1236] Overall Loss 0.334733 Objective Loss 0.334733 LR 0.001000 Time 0.021482 +2023-10-05 21:07:35,276 - Epoch: [54][ 530/ 1236] Overall Loss 0.334336 Objective Loss 0.334336 LR 0.001000 Time 0.021451 +2023-10-05 21:07:35,476 - Epoch: [54][ 540/ 1236] Overall Loss 0.334545 Objective Loss 0.334545 LR 0.001000 Time 0.021424 +2023-10-05 21:07:35,675 - Epoch: [54][ 550/ 1236] Overall Loss 0.334705 Objective Loss 0.334705 LR 0.001000 Time 0.021395 +2023-10-05 21:07:35,875 - Epoch: [54][ 560/ 1236] Overall Loss 0.335000 Objective Loss 0.335000 LR 0.001000 Time 0.021370 +2023-10-05 21:07:36,074 - Epoch: [54][ 570/ 1236] Overall Loss 0.334420 Objective Loss 0.334420 LR 0.001000 Time 0.021343 +2023-10-05 21:07:36,274 - Epoch: [54][ 580/ 1236] Overall Loss 0.334589 Objective Loss 0.334589 LR 0.001000 Time 0.021320 +2023-10-05 21:07:36,473 - Epoch: [54][ 590/ 1236] Overall Loss 0.334845 Objective Loss 0.334845 LR 0.001000 Time 0.021295 +2023-10-05 21:07:36,673 - Epoch: [54][ 600/ 1236] Overall Loss 0.334588 Objective Loss 0.334588 LR 0.001000 Time 0.021272 +2023-10-05 21:07:36,872 - Epoch: [54][ 610/ 1236] Overall Loss 0.334664 Objective Loss 0.334664 LR 0.001000 Time 0.021249 +2023-10-05 21:07:37,073 - Epoch: [54][ 620/ 1236] Overall Loss 0.334807 Objective Loss 0.334807 LR 0.001000 Time 0.021229 +2023-10-05 21:07:37,271 - Epoch: [54][ 630/ 1236] Overall Loss 0.335144 Objective Loss 0.335144 LR 0.001000 Time 0.021207 +2023-10-05 21:07:37,472 - Epoch: [54][ 640/ 1236] Overall Loss 0.335139 Objective Loss 0.335139 LR 0.001000 Time 0.021188 +2023-10-05 21:07:37,670 - Epoch: [54][ 650/ 1236] Overall Loss 0.335520 Objective Loss 0.335520 LR 0.001000 Time 0.021167 +2023-10-05 21:07:37,871 - Epoch: [54][ 660/ 1236] Overall Loss 0.335380 Objective Loss 0.335380 LR 0.001000 Time 0.021150 +2023-10-05 21:07:38,069 - Epoch: [54][ 670/ 1236] Overall Loss 0.335354 Objective Loss 0.335354 LR 0.001000 Time 0.021130 +2023-10-05 21:07:38,270 - Epoch: [54][ 680/ 1236] Overall Loss 0.335327 Objective Loss 0.335327 LR 0.001000 Time 0.021113 +2023-10-05 21:07:38,468 - Epoch: [54][ 690/ 1236] Overall Loss 0.335300 Objective Loss 0.335300 LR 0.001000 Time 0.021094 +2023-10-05 21:07:38,669 - Epoch: [54][ 700/ 1236] Overall Loss 0.335438 Objective Loss 0.335438 LR 0.001000 Time 0.021079 +2023-10-05 21:07:38,867 - Epoch: [54][ 710/ 1236] Overall Loss 0.335565 Objective Loss 0.335565 LR 0.001000 Time 0.021061 +2023-10-05 21:07:39,068 - Epoch: [54][ 720/ 1236] Overall Loss 0.335456 Objective Loss 0.335456 LR 0.001000 Time 0.021046 +2023-10-05 21:07:39,266 - Epoch: [54][ 730/ 1236] Overall Loss 0.335507 Objective Loss 0.335507 LR 0.001000 Time 0.021030 +2023-10-05 21:07:39,467 - Epoch: [54][ 740/ 1236] Overall Loss 0.335976 Objective Loss 0.335976 LR 0.001000 Time 0.021016 +2023-10-05 21:07:39,665 - Epoch: [54][ 750/ 1236] Overall Loss 0.336459 Objective Loss 0.336459 LR 0.001000 Time 0.021000 +2023-10-05 21:07:39,866 - Epoch: [54][ 760/ 1236] Overall Loss 0.336378 Objective Loss 0.336378 LR 0.001000 Time 0.020987 +2023-10-05 21:07:40,064 - Epoch: [54][ 770/ 1236] Overall Loss 0.336636 Objective Loss 0.336636 LR 0.001000 Time 0.020972 +2023-10-05 21:07:40,265 - Epoch: [54][ 780/ 1236] Overall Loss 0.336879 Objective Loss 0.336879 LR 0.001000 Time 0.020959 +2023-10-05 21:07:40,464 - Epoch: [54][ 790/ 1236] Overall Loss 0.337095 Objective Loss 0.337095 LR 0.001000 Time 0.020945 +2023-10-05 21:07:40,664 - Epoch: [54][ 800/ 1236] Overall Loss 0.337486 Objective Loss 0.337486 LR 0.001000 Time 0.020934 +2023-10-05 21:07:40,863 - Epoch: [54][ 810/ 1236] Overall Loss 0.338018 Objective Loss 0.338018 LR 0.001000 Time 0.020920 +2023-10-05 21:07:41,063 - Epoch: [54][ 820/ 1236] Overall Loss 0.337599 Objective Loss 0.337599 LR 0.001000 Time 0.020909 +2023-10-05 21:07:41,262 - Epoch: [54][ 830/ 1236] Overall Loss 0.337615 Objective Loss 0.337615 LR 0.001000 Time 0.020896 +2023-10-05 21:07:41,462 - Epoch: [54][ 840/ 1236] Overall Loss 0.337615 Objective Loss 0.337615 LR 0.001000 Time 0.020885 +2023-10-05 21:07:41,661 - Epoch: [54][ 850/ 1236] Overall Loss 0.337960 Objective Loss 0.337960 LR 0.001000 Time 0.020872 +2023-10-05 21:07:41,861 - Epoch: [54][ 860/ 1236] Overall Loss 0.337859 Objective Loss 0.337859 LR 0.001000 Time 0.020862 +2023-10-05 21:07:42,060 - Epoch: [54][ 870/ 1236] Overall Loss 0.337937 Objective Loss 0.337937 LR 0.001000 Time 0.020850 +2023-10-05 21:07:42,260 - Epoch: [54][ 880/ 1236] Overall Loss 0.338162 Objective Loss 0.338162 LR 0.001000 Time 0.020841 +2023-10-05 21:07:42,459 - Epoch: [54][ 890/ 1236] Overall Loss 0.338282 Objective Loss 0.338282 LR 0.001000 Time 0.020830 +2023-10-05 21:07:42,660 - Epoch: [54][ 900/ 1236] Overall Loss 0.338106 Objective Loss 0.338106 LR 0.001000 Time 0.020820 +2023-10-05 21:07:42,858 - Epoch: [54][ 910/ 1236] Overall Loss 0.338128 Objective Loss 0.338128 LR 0.001000 Time 0.020809 +2023-10-05 21:07:43,058 - Epoch: [54][ 920/ 1236] Overall Loss 0.338197 Objective Loss 0.338197 LR 0.001000 Time 0.020800 +2023-10-05 21:07:43,257 - Epoch: [54][ 930/ 1236] Overall Loss 0.338535 Objective Loss 0.338535 LR 0.001000 Time 0.020790 +2023-10-05 21:07:43,458 - Epoch: [54][ 940/ 1236] Overall Loss 0.338668 Objective Loss 0.338668 LR 0.001000 Time 0.020782 +2023-10-05 21:07:43,656 - Epoch: [54][ 950/ 1236] Overall Loss 0.338688 Objective Loss 0.338688 LR 0.001000 Time 0.020772 +2023-10-05 21:07:43,857 - Epoch: [54][ 960/ 1236] Overall Loss 0.338393 Objective Loss 0.338393 LR 0.001000 Time 0.020764 +2023-10-05 21:07:44,055 - Epoch: [54][ 970/ 1236] Overall Loss 0.338187 Objective Loss 0.338187 LR 0.001000 Time 0.020754 +2023-10-05 21:07:44,256 - Epoch: [54][ 980/ 1236] Overall Loss 0.338207 Objective Loss 0.338207 LR 0.001000 Time 0.020746 +2023-10-05 21:07:44,454 - Epoch: [54][ 990/ 1236] Overall Loss 0.338175 Objective Loss 0.338175 LR 0.001000 Time 0.020737 +2023-10-05 21:07:44,654 - Epoch: [54][ 1000/ 1236] Overall Loss 0.337946 Objective Loss 0.337946 LR 0.001000 Time 0.020729 +2023-10-05 21:07:44,853 - Epoch: [54][ 1010/ 1236] Overall Loss 0.338300 Objective Loss 0.338300 LR 0.001000 Time 0.020720 +2023-10-05 21:07:45,054 - Epoch: [54][ 1020/ 1236] Overall Loss 0.338354 Objective Loss 0.338354 LR 0.001000 Time 0.020713 +2023-10-05 21:07:45,252 - Epoch: [54][ 1030/ 1236] Overall Loss 0.338638 Objective Loss 0.338638 LR 0.001000 Time 0.020705 +2023-10-05 21:07:45,453 - Epoch: [54][ 1040/ 1236] Overall Loss 0.338876 Objective Loss 0.338876 LR 0.001000 Time 0.020698 +2023-10-05 21:07:45,652 - Epoch: [54][ 1050/ 1236] Overall Loss 0.338713 Objective Loss 0.338713 LR 0.001000 Time 0.020691 +2023-10-05 21:07:45,853 - Epoch: [54][ 1060/ 1236] Overall Loss 0.339068 Objective Loss 0.339068 LR 0.001000 Time 0.020684 +2023-10-05 21:07:46,051 - Epoch: [54][ 1070/ 1236] Overall Loss 0.339368 Objective Loss 0.339368 LR 0.001000 Time 0.020676 +2023-10-05 21:07:46,252 - Epoch: [54][ 1080/ 1236] Overall Loss 0.339376 Objective Loss 0.339376 LR 0.001000 Time 0.020670 +2023-10-05 21:07:46,451 - Epoch: [54][ 1090/ 1236] Overall Loss 0.339534 Objective Loss 0.339534 LR 0.001000 Time 0.020663 +2023-10-05 21:07:46,652 - Epoch: [54][ 1100/ 1236] Overall Loss 0.339502 Objective Loss 0.339502 LR 0.001000 Time 0.020657 +2023-10-05 21:07:46,850 - Epoch: [54][ 1110/ 1236] Overall Loss 0.339526 Objective Loss 0.339526 LR 0.001000 Time 0.020649 +2023-10-05 21:07:47,051 - Epoch: [54][ 1120/ 1236] Overall Loss 0.339499 Objective Loss 0.339499 LR 0.001000 Time 0.020644 +2023-10-05 21:07:47,249 - Epoch: [54][ 1130/ 1236] Overall Loss 0.339703 Objective Loss 0.339703 LR 0.001000 Time 0.020636 +2023-10-05 21:07:47,450 - Epoch: [54][ 1140/ 1236] Overall Loss 0.339684 Objective Loss 0.339684 LR 0.001000 Time 0.020631 +2023-10-05 21:07:47,648 - Epoch: [54][ 1150/ 1236] Overall Loss 0.339528 Objective Loss 0.339528 LR 0.001000 Time 0.020624 +2023-10-05 21:07:47,849 - Epoch: [54][ 1160/ 1236] Overall Loss 0.339667 Objective Loss 0.339667 LR 0.001000 Time 0.020619 +2023-10-05 21:07:48,048 - Epoch: [54][ 1170/ 1236] Overall Loss 0.339850 Objective Loss 0.339850 LR 0.001000 Time 0.020612 +2023-10-05 21:07:48,249 - Epoch: [54][ 1180/ 1236] Overall Loss 0.339765 Objective Loss 0.339765 LR 0.001000 Time 0.020608 +2023-10-05 21:07:48,448 - Epoch: [54][ 1190/ 1236] Overall Loss 0.339776 Objective Loss 0.339776 LR 0.001000 Time 0.020601 +2023-10-05 21:07:48,648 - Epoch: [54][ 1200/ 1236] Overall Loss 0.339789 Objective Loss 0.339789 LR 0.001000 Time 0.020596 +2023-10-05 21:07:48,847 - Epoch: [54][ 1210/ 1236] Overall Loss 0.339658 Objective Loss 0.339658 LR 0.001000 Time 0.020590 +2023-10-05 21:07:49,048 - Epoch: [54][ 1220/ 1236] Overall Loss 0.339512 Objective Loss 0.339512 LR 0.001000 Time 0.020586 +2023-10-05 21:07:49,301 - Epoch: [54][ 1230/ 1236] Overall Loss 0.339410 Objective Loss 0.339410 LR 0.001000 Time 0.020624 +2023-10-05 21:07:49,418 - Epoch: [54][ 1236/ 1236] Overall Loss 0.339201 Objective Loss 0.339201 Top1 82.281059 Top5 97.556008 LR 0.001000 Time 0.020618 +2023-10-05 21:07:49,559 - --- validate (epoch=54)----------- +2023-10-05 21:07:49,559 - 29943 samples (256 per mini-batch) +2023-10-05 21:07:50,010 - Epoch: [54][ 10/ 117] Loss 0.391676 Top1 81.289062 Top5 96.875000 +2023-10-05 21:07:50,158 - Epoch: [54][ 20/ 117] Loss 0.370002 Top1 81.289062 Top5 96.855469 +2023-10-05 21:07:50,305 - Epoch: [54][ 30/ 117] Loss 0.370957 Top1 81.223958 Top5 96.979167 +2023-10-05 21:07:50,451 - Epoch: [54][ 40/ 117] Loss 0.372653 Top1 81.171875 Top5 97.021484 +2023-10-05 21:07:50,598 - Epoch: [54][ 50/ 117] Loss 0.365041 Top1 81.203125 Top5 97.078125 +2023-10-05 21:07:50,744 - Epoch: [54][ 60/ 117] Loss 0.369058 Top1 81.028646 Top5 97.122396 +2023-10-05 21:07:50,890 - Epoch: [54][ 70/ 117] Loss 0.364282 Top1 81.060268 Top5 97.209821 +2023-10-05 21:07:51,037 - Epoch: [54][ 80/ 117] Loss 0.362664 Top1 81.254883 Top5 97.211914 +2023-10-05 21:07:51,183 - Epoch: [54][ 90/ 117] Loss 0.363163 Top1 81.241319 Top5 97.239583 +2023-10-05 21:07:51,330 - Epoch: [54][ 100/ 117] Loss 0.362426 Top1 81.289062 Top5 97.253906 +2023-10-05 21:07:51,482 - Epoch: [54][ 110/ 117] Loss 0.361695 Top1 81.296165 Top5 97.230114 +2023-10-05 21:07:51,567 - Epoch: [54][ 117/ 117] Loss 0.359514 Top1 81.357913 Top5 97.234746 +2023-10-05 21:07:51,686 - ==> Top1: 81.358 Top5: 97.235 Loss: 0.360 + +2023-10-05 21:07:51,687 - ==> Confusion: +[[ 932 2 4 0 16 3 0 1 1 60 1 3 1 1 7 1 5 0 1 0 11] + [ 0 1042 2 1 19 21 1 26 0 0 0 3 0 0 0 3 1 1 6 1 4] + [ 6 0 911 20 7 0 48 7 0 3 5 6 8 1 2 5 2 2 8 4 11] + [ 1 2 15 953 2 8 1 1 2 0 5 0 6 4 46 4 2 9 14 3 11] + [ 17 4 0 0 990 1 0 0 1 7 0 1 1 0 5 5 12 1 0 1 4] + [ 4 39 0 1 9 987 2 21 1 0 2 9 3 10 7 1 3 2 3 4 8] + [ 0 8 22 1 3 2 1125 5 0 0 3 2 1 0 1 7 0 2 2 4 3] + [ 5 32 13 0 8 47 6 1036 0 1 4 7 0 0 1 2 1 0 38 10 7] + [ 20 4 0 0 0 6 0 1 942 43 11 3 1 13 31 6 3 0 3 1 1] + [ 123 1 0 0 10 5 1 1 26 908 1 0 0 14 8 9 1 0 0 2 9] + [ 1 7 14 3 1 0 9 5 18 4 942 2 0 8 10 0 3 2 13 2 9] + [ 3 0 2 0 0 13 0 0 0 0 1 956 25 0 0 2 2 17 0 11 3] + [ 1 2 4 5 0 2 2 0 0 1 0 31 968 0 2 10 3 15 4 7 11] + [ 1 0 2 1 6 13 0 0 22 21 4 6 3 1009 7 6 0 2 0 6 10] + [ 16 1 0 6 8 0 0 0 20 4 2 2 2 1 1023 0 3 0 5 0 8] + [ 1 2 2 1 4 0 0 0 0 0 0 9 8 0 1 1062 11 16 2 8 7] + [ 0 22 1 0 11 5 0 0 1 0 0 4 0 1 3 17 1086 0 1 5 4] + [ 0 0 0 1 0 0 3 0 2 0 0 2 20 0 8 7 0 990 2 1 2] + [ 1 10 9 17 2 1 0 34 3 0 2 0 2 0 14 1 0 0 962 3 7] + [ 0 4 3 2 2 6 7 8 0 0 0 15 3 0 0 13 10 0 0 1075 4] + [ 135 266 150 66 178 195 66 96 98 91 200 129 445 289 201 94 217 93 149 285 4462]] + +2023-10-05 21:07:51,688 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:07:51,688 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:07:51,694 - + +2023-10-05 21:07:51,694 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:07:52,673 - Epoch: [55][ 10/ 1236] Overall Loss 0.331765 Objective Loss 0.331765 LR 0.001000 Time 0.097839 +2023-10-05 21:07:52,872 - Epoch: [55][ 20/ 1236] Overall Loss 0.323216 Objective Loss 0.323216 LR 0.001000 Time 0.058889 +2023-10-05 21:07:53,071 - Epoch: [55][ 30/ 1236] Overall Loss 0.328056 Objective Loss 0.328056 LR 0.001000 Time 0.045864 +2023-10-05 21:07:53,271 - Epoch: [55][ 40/ 1236] Overall Loss 0.326718 Objective Loss 0.326718 LR 0.001000 Time 0.039382 +2023-10-05 21:07:53,469 - Epoch: [55][ 50/ 1236] Overall Loss 0.326804 Objective Loss 0.326804 LR 0.001000 Time 0.035473 +2023-10-05 21:07:53,669 - Epoch: [55][ 60/ 1236] Overall Loss 0.325714 Objective Loss 0.325714 LR 0.001000 Time 0.032882 +2023-10-05 21:07:53,868 - Epoch: [55][ 70/ 1236] Overall Loss 0.326822 Objective Loss 0.326822 LR 0.001000 Time 0.031024 +2023-10-05 21:07:54,067 - Epoch: [55][ 80/ 1236] Overall Loss 0.328061 Objective Loss 0.328061 LR 0.001000 Time 0.029635 +2023-10-05 21:07:54,266 - Epoch: [55][ 90/ 1236] Overall Loss 0.325691 Objective Loss 0.325691 LR 0.001000 Time 0.028547 +2023-10-05 21:07:54,465 - Epoch: [55][ 100/ 1236] Overall Loss 0.322619 Objective Loss 0.322619 LR 0.001000 Time 0.027682 +2023-10-05 21:07:54,664 - Epoch: [55][ 110/ 1236] Overall Loss 0.319181 Objective Loss 0.319181 LR 0.001000 Time 0.026968 +2023-10-05 21:07:54,864 - Epoch: [55][ 120/ 1236] Overall Loss 0.319683 Objective Loss 0.319683 LR 0.001000 Time 0.026380 +2023-10-05 21:07:55,062 - Epoch: [55][ 130/ 1236] Overall Loss 0.318192 Objective Loss 0.318192 LR 0.001000 Time 0.025874 +2023-10-05 21:07:55,261 - Epoch: [55][ 140/ 1236] Overall Loss 0.320300 Objective Loss 0.320300 LR 0.001000 Time 0.025448 +2023-10-05 21:07:55,460 - Epoch: [55][ 150/ 1236] Overall Loss 0.320935 Objective Loss 0.320935 LR 0.001000 Time 0.025073 +2023-10-05 21:07:55,659 - Epoch: [55][ 160/ 1236] Overall Loss 0.320567 Objective Loss 0.320567 LR 0.001000 Time 0.024749 +2023-10-05 21:07:55,858 - Epoch: [55][ 170/ 1236] Overall Loss 0.321709 Objective Loss 0.321709 LR 0.001000 Time 0.024458 +2023-10-05 21:07:56,057 - Epoch: [55][ 180/ 1236] Overall Loss 0.323515 Objective Loss 0.323515 LR 0.001000 Time 0.024206 +2023-10-05 21:07:56,255 - Epoch: [55][ 190/ 1236] Overall Loss 0.324375 Objective Loss 0.324375 LR 0.001000 Time 0.023973 +2023-10-05 21:07:56,455 - Epoch: [55][ 200/ 1236] Overall Loss 0.321408 Objective Loss 0.321408 LR 0.001000 Time 0.023770 +2023-10-05 21:07:56,653 - Epoch: [55][ 210/ 1236] Overall Loss 0.321075 Objective Loss 0.321075 LR 0.001000 Time 0.023581 +2023-10-05 21:07:56,853 - Epoch: [55][ 220/ 1236] Overall Loss 0.321922 Objective Loss 0.321922 LR 0.001000 Time 0.023413 +2023-10-05 21:07:57,051 - Epoch: [55][ 230/ 1236] Overall Loss 0.322494 Objective Loss 0.322494 LR 0.001000 Time 0.023256 +2023-10-05 21:07:57,250 - Epoch: [55][ 240/ 1236] Overall Loss 0.322825 Objective Loss 0.322825 LR 0.001000 Time 0.023116 +2023-10-05 21:07:57,448 - Epoch: [55][ 250/ 1236] Overall Loss 0.321918 Objective Loss 0.321918 LR 0.001000 Time 0.022983 +2023-10-05 21:07:57,648 - Epoch: [55][ 260/ 1236] Overall Loss 0.322842 Objective Loss 0.322842 LR 0.001000 Time 0.022866 +2023-10-05 21:07:57,846 - Epoch: [55][ 270/ 1236] Overall Loss 0.323493 Objective Loss 0.323493 LR 0.001000 Time 0.022752 +2023-10-05 21:07:58,046 - Epoch: [55][ 280/ 1236] Overall Loss 0.323817 Objective Loss 0.323817 LR 0.001000 Time 0.022649 +2023-10-05 21:07:58,244 - Epoch: [55][ 290/ 1236] Overall Loss 0.324649 Objective Loss 0.324649 LR 0.001000 Time 0.022551 +2023-10-05 21:07:58,444 - Epoch: [55][ 300/ 1236] Overall Loss 0.324754 Objective Loss 0.324754 LR 0.001000 Time 0.022463 +2023-10-05 21:07:58,642 - Epoch: [55][ 310/ 1236] Overall Loss 0.324826 Objective Loss 0.324826 LR 0.001000 Time 0.022378 +2023-10-05 21:07:58,842 - Epoch: [55][ 320/ 1236] Overall Loss 0.325050 Objective Loss 0.325050 LR 0.001000 Time 0.022301 +2023-10-05 21:07:59,040 - Epoch: [55][ 330/ 1236] Overall Loss 0.325305 Objective Loss 0.325305 LR 0.001000 Time 0.022224 +2023-10-05 21:07:59,240 - Epoch: [55][ 340/ 1236] Overall Loss 0.324994 Objective Loss 0.324994 LR 0.001000 Time 0.022158 +2023-10-05 21:07:59,438 - Epoch: [55][ 350/ 1236] Overall Loss 0.324894 Objective Loss 0.324894 LR 0.001000 Time 0.022092 +2023-10-05 21:07:59,638 - Epoch: [55][ 360/ 1236] Overall Loss 0.324509 Objective Loss 0.324509 LR 0.001000 Time 0.022031 +2023-10-05 21:07:59,837 - Epoch: [55][ 370/ 1236] Overall Loss 0.324808 Objective Loss 0.324808 LR 0.001000 Time 0.021972 +2023-10-05 21:08:00,037 - Epoch: [55][ 380/ 1236] Overall Loss 0.324012 Objective Loss 0.324012 LR 0.001000 Time 0.021919 +2023-10-05 21:08:00,235 - Epoch: [55][ 390/ 1236] Overall Loss 0.324243 Objective Loss 0.324243 LR 0.001000 Time 0.021866 +2023-10-05 21:08:00,435 - Epoch: [55][ 400/ 1236] Overall Loss 0.325255 Objective Loss 0.325255 LR 0.001000 Time 0.021817 +2023-10-05 21:08:00,634 - Epoch: [55][ 410/ 1236] Overall Loss 0.324995 Objective Loss 0.324995 LR 0.001000 Time 0.021769 +2023-10-05 21:08:00,834 - Epoch: [55][ 420/ 1236] Overall Loss 0.324736 Objective Loss 0.324736 LR 0.001000 Time 0.021726 +2023-10-05 21:08:01,032 - Epoch: [55][ 430/ 1236] Overall Loss 0.324734 Objective Loss 0.324734 LR 0.001000 Time 0.021682 +2023-10-05 21:08:01,232 - Epoch: [55][ 440/ 1236] Overall Loss 0.324873 Objective Loss 0.324873 LR 0.001000 Time 0.021642 +2023-10-05 21:08:01,430 - Epoch: [55][ 450/ 1236] Overall Loss 0.324362 Objective Loss 0.324362 LR 0.001000 Time 0.021600 +2023-10-05 21:08:01,629 - Epoch: [55][ 460/ 1236] Overall Loss 0.324334 Objective Loss 0.324334 LR 0.001000 Time 0.021563 +2023-10-05 21:08:01,827 - Epoch: [55][ 470/ 1236] Overall Loss 0.324615 Objective Loss 0.324615 LR 0.001000 Time 0.021524 +2023-10-05 21:08:02,026 - Epoch: [55][ 480/ 1236] Overall Loss 0.324562 Objective Loss 0.324562 LR 0.001000 Time 0.021490 +2023-10-05 21:08:02,224 - Epoch: [55][ 490/ 1236] Overall Loss 0.324861 Objective Loss 0.324861 LR 0.001000 Time 0.021454 +2023-10-05 21:08:02,423 - Epoch: [55][ 500/ 1236] Overall Loss 0.324501 Objective Loss 0.324501 LR 0.001000 Time 0.021423 +2023-10-05 21:08:02,621 - Epoch: [55][ 510/ 1236] Overall Loss 0.325461 Objective Loss 0.325461 LR 0.001000 Time 0.021389 +2023-10-05 21:08:02,820 - Epoch: [55][ 520/ 1236] Overall Loss 0.325168 Objective Loss 0.325168 LR 0.001000 Time 0.021360 +2023-10-05 21:08:03,017 - Epoch: [55][ 530/ 1236] Overall Loss 0.325902 Objective Loss 0.325902 LR 0.001000 Time 0.021330 +2023-10-05 21:08:03,216 - Epoch: [55][ 540/ 1236] Overall Loss 0.325148 Objective Loss 0.325148 LR 0.001000 Time 0.021303 +2023-10-05 21:08:03,414 - Epoch: [55][ 550/ 1236] Overall Loss 0.325472 Objective Loss 0.325472 LR 0.001000 Time 0.021274 +2023-10-05 21:08:03,613 - Epoch: [55][ 560/ 1236] Overall Loss 0.325111 Objective Loss 0.325111 LR 0.001000 Time 0.021249 +2023-10-05 21:08:03,811 - Epoch: [55][ 570/ 1236] Overall Loss 0.324964 Objective Loss 0.324964 LR 0.001000 Time 0.021223 +2023-10-05 21:08:04,010 - Epoch: [55][ 580/ 1236] Overall Loss 0.324703 Objective Loss 0.324703 LR 0.001000 Time 0.021200 +2023-10-05 21:08:04,208 - Epoch: [55][ 590/ 1236] Overall Loss 0.324172 Objective Loss 0.324172 LR 0.001000 Time 0.021175 +2023-10-05 21:08:04,407 - Epoch: [55][ 600/ 1236] Overall Loss 0.324373 Objective Loss 0.324373 LR 0.001000 Time 0.021153 +2023-10-05 21:08:04,605 - Epoch: [55][ 610/ 1236] Overall Loss 0.324036 Objective Loss 0.324036 LR 0.001000 Time 0.021131 +2023-10-05 21:08:04,804 - Epoch: [55][ 620/ 1236] Overall Loss 0.324201 Objective Loss 0.324201 LR 0.001000 Time 0.021110 +2023-10-05 21:08:05,002 - Epoch: [55][ 630/ 1236] Overall Loss 0.324426 Objective Loss 0.324426 LR 0.001000 Time 0.021089 +2023-10-05 21:08:05,201 - Epoch: [55][ 640/ 1236] Overall Loss 0.324946 Objective Loss 0.324946 LR 0.001000 Time 0.021070 +2023-10-05 21:08:05,399 - Epoch: [55][ 650/ 1236] Overall Loss 0.325491 Objective Loss 0.325491 LR 0.001000 Time 0.021049 +2023-10-05 21:08:05,598 - Epoch: [55][ 660/ 1236] Overall Loss 0.325057 Objective Loss 0.325057 LR 0.001000 Time 0.021032 +2023-10-05 21:08:05,796 - Epoch: [55][ 670/ 1236] Overall Loss 0.325278 Objective Loss 0.325278 LR 0.001000 Time 0.021013 +2023-10-05 21:08:05,995 - Epoch: [55][ 680/ 1236] Overall Loss 0.325472 Objective Loss 0.325472 LR 0.001000 Time 0.020996 +2023-10-05 21:08:06,193 - Epoch: [55][ 690/ 1236] Overall Loss 0.325091 Objective Loss 0.325091 LR 0.001000 Time 0.020978 +2023-10-05 21:08:06,392 - Epoch: [55][ 700/ 1236] Overall Loss 0.325410 Objective Loss 0.325410 LR 0.001000 Time 0.020962 +2023-10-05 21:08:06,590 - Epoch: [55][ 710/ 1236] Overall Loss 0.325637 Objective Loss 0.325637 LR 0.001000 Time 0.020945 +2023-10-05 21:08:06,789 - Epoch: [55][ 720/ 1236] Overall Loss 0.326006 Objective Loss 0.326006 LR 0.001000 Time 0.020930 +2023-10-05 21:08:06,986 - Epoch: [55][ 730/ 1236] Overall Loss 0.326436 Objective Loss 0.326436 LR 0.001000 Time 0.020914 +2023-10-05 21:08:07,185 - Epoch: [55][ 740/ 1236] Overall Loss 0.326932 Objective Loss 0.326932 LR 0.001000 Time 0.020900 +2023-10-05 21:08:07,383 - Epoch: [55][ 750/ 1236] Overall Loss 0.326922 Objective Loss 0.326922 LR 0.001000 Time 0.020884 +2023-10-05 21:08:07,582 - Epoch: [55][ 760/ 1236] Overall Loss 0.327353 Objective Loss 0.327353 LR 0.001000 Time 0.020871 +2023-10-05 21:08:07,780 - Epoch: [55][ 770/ 1236] Overall Loss 0.327460 Objective Loss 0.327460 LR 0.001000 Time 0.020856 +2023-10-05 21:08:07,979 - Epoch: [55][ 780/ 1236] Overall Loss 0.327301 Objective Loss 0.327301 LR 0.001000 Time 0.020843 +2023-10-05 21:08:08,177 - Epoch: [55][ 790/ 1236] Overall Loss 0.327548 Objective Loss 0.327548 LR 0.001000 Time 0.020829 +2023-10-05 21:08:08,376 - Epoch: [55][ 800/ 1236] Overall Loss 0.327375 Objective Loss 0.327375 LR 0.001000 Time 0.020818 +2023-10-05 21:08:08,574 - Epoch: [55][ 810/ 1236] Overall Loss 0.327181 Objective Loss 0.327181 LR 0.001000 Time 0.020805 +2023-10-05 21:08:08,773 - Epoch: [55][ 820/ 1236] Overall Loss 0.327331 Objective Loss 0.327331 LR 0.001000 Time 0.020793 +2023-10-05 21:08:08,971 - Epoch: [55][ 830/ 1236] Overall Loss 0.327884 Objective Loss 0.327884 LR 0.001000 Time 0.020781 +2023-10-05 21:08:09,170 - Epoch: [55][ 840/ 1236] Overall Loss 0.327518 Objective Loss 0.327518 LR 0.001000 Time 0.020770 +2023-10-05 21:08:09,368 - Epoch: [55][ 850/ 1236] Overall Loss 0.327561 Objective Loss 0.327561 LR 0.001000 Time 0.020758 +2023-10-05 21:08:09,567 - Epoch: [55][ 860/ 1236] Overall Loss 0.327456 Objective Loss 0.327456 LR 0.001000 Time 0.020748 +2023-10-05 21:08:09,765 - Epoch: [55][ 870/ 1236] Overall Loss 0.328056 Objective Loss 0.328056 LR 0.001000 Time 0.020736 +2023-10-05 21:08:09,964 - Epoch: [55][ 880/ 1236] Overall Loss 0.328079 Objective Loss 0.328079 LR 0.001000 Time 0.020727 +2023-10-05 21:08:10,162 - Epoch: [55][ 890/ 1236] Overall Loss 0.328195 Objective Loss 0.328195 LR 0.001000 Time 0.020716 +2023-10-05 21:08:10,361 - Epoch: [55][ 900/ 1236] Overall Loss 0.328285 Objective Loss 0.328285 LR 0.001000 Time 0.020707 +2023-10-05 21:08:10,563 - Epoch: [55][ 910/ 1236] Overall Loss 0.328325 Objective Loss 0.328325 LR 0.001000 Time 0.020700 +2023-10-05 21:08:10,763 - Epoch: [55][ 920/ 1236] Overall Loss 0.328600 Objective Loss 0.328600 LR 0.001000 Time 0.020693 +2023-10-05 21:08:10,962 - Epoch: [55][ 930/ 1236] Overall Loss 0.328768 Objective Loss 0.328768 LR 0.001000 Time 0.020684 +2023-10-05 21:08:11,163 - Epoch: [55][ 940/ 1236] Overall Loss 0.329390 Objective Loss 0.329390 LR 0.001000 Time 0.020677 +2023-10-05 21:08:11,362 - Epoch: [55][ 950/ 1236] Overall Loss 0.329376 Objective Loss 0.329376 LR 0.001000 Time 0.020668 +2023-10-05 21:08:11,562 - Epoch: [55][ 960/ 1236] Overall Loss 0.329777 Objective Loss 0.329777 LR 0.001000 Time 0.020661 +2023-10-05 21:08:11,761 - Epoch: [55][ 970/ 1236] Overall Loss 0.329860 Objective Loss 0.329860 LR 0.001000 Time 0.020653 +2023-10-05 21:08:11,961 - Epoch: [55][ 980/ 1236] Overall Loss 0.329928 Objective Loss 0.329928 LR 0.001000 Time 0.020646 +2023-10-05 21:08:12,160 - Epoch: [55][ 990/ 1236] Overall Loss 0.329916 Objective Loss 0.329916 LR 0.001000 Time 0.020638 +2023-10-05 21:08:12,360 - Epoch: [55][ 1000/ 1236] Overall Loss 0.329933 Objective Loss 0.329933 LR 0.001000 Time 0.020632 +2023-10-05 21:08:12,559 - Epoch: [55][ 1010/ 1236] Overall Loss 0.329747 Objective Loss 0.329747 LR 0.001000 Time 0.020624 +2023-10-05 21:08:12,760 - Epoch: [55][ 1020/ 1236] Overall Loss 0.329473 Objective Loss 0.329473 LR 0.001000 Time 0.020618 +2023-10-05 21:08:12,958 - Epoch: [55][ 1030/ 1236] Overall Loss 0.329139 Objective Loss 0.329139 LR 0.001000 Time 0.020610 +2023-10-05 21:08:13,159 - Epoch: [55][ 1040/ 1236] Overall Loss 0.329130 Objective Loss 0.329130 LR 0.001000 Time 0.020604 +2023-10-05 21:08:13,357 - Epoch: [55][ 1050/ 1236] Overall Loss 0.329106 Objective Loss 0.329106 LR 0.001000 Time 0.020597 +2023-10-05 21:08:13,558 - Epoch: [55][ 1060/ 1236] Overall Loss 0.329028 Objective Loss 0.329028 LR 0.001000 Time 0.020591 +2023-10-05 21:08:13,756 - Epoch: [55][ 1070/ 1236] Overall Loss 0.329185 Objective Loss 0.329185 LR 0.001000 Time 0.020584 +2023-10-05 21:08:13,957 - Epoch: [55][ 1080/ 1236] Overall Loss 0.329370 Objective Loss 0.329370 LR 0.001000 Time 0.020579 +2023-10-05 21:08:14,156 - Epoch: [55][ 1090/ 1236] Overall Loss 0.329301 Objective Loss 0.329301 LR 0.001000 Time 0.020572 +2023-10-05 21:08:14,356 - Epoch: [55][ 1100/ 1236] Overall Loss 0.329168 Objective Loss 0.329168 LR 0.001000 Time 0.020567 +2023-10-05 21:08:14,555 - Epoch: [55][ 1110/ 1236] Overall Loss 0.329050 Objective Loss 0.329050 LR 0.001000 Time 0.020561 +2023-10-05 21:08:14,755 - Epoch: [55][ 1120/ 1236] Overall Loss 0.329062 Objective Loss 0.329062 LR 0.001000 Time 0.020556 +2023-10-05 21:08:14,954 - Epoch: [55][ 1130/ 1236] Overall Loss 0.329004 Objective Loss 0.329004 LR 0.001000 Time 0.020550 +2023-10-05 21:08:15,155 - Epoch: [55][ 1140/ 1236] Overall Loss 0.328725 Objective Loss 0.328725 LR 0.001000 Time 0.020545 +2023-10-05 21:08:15,354 - Epoch: [55][ 1150/ 1236] Overall Loss 0.328814 Objective Loss 0.328814 LR 0.001000 Time 0.020539 +2023-10-05 21:08:15,554 - Epoch: [55][ 1160/ 1236] Overall Loss 0.328951 Objective Loss 0.328951 LR 0.001000 Time 0.020534 +2023-10-05 21:08:15,753 - Epoch: [55][ 1170/ 1236] Overall Loss 0.328857 Objective Loss 0.328857 LR 0.001000 Time 0.020528 +2023-10-05 21:08:15,953 - Epoch: [55][ 1180/ 1236] Overall Loss 0.328658 Objective Loss 0.328658 LR 0.001000 Time 0.020524 +2023-10-05 21:08:16,152 - Epoch: [55][ 1190/ 1236] Overall Loss 0.328902 Objective Loss 0.328902 LR 0.001000 Time 0.020518 +2023-10-05 21:08:16,352 - Epoch: [55][ 1200/ 1236] Overall Loss 0.328957 Objective Loss 0.328957 LR 0.001000 Time 0.020514 +2023-10-05 21:08:16,552 - Epoch: [55][ 1210/ 1236] Overall Loss 0.329189 Objective Loss 0.329189 LR 0.001000 Time 0.020509 +2023-10-05 21:08:16,752 - Epoch: [55][ 1220/ 1236] Overall Loss 0.329109 Objective Loss 0.329109 LR 0.001000 Time 0.020505 +2023-10-05 21:08:17,005 - Epoch: [55][ 1230/ 1236] Overall Loss 0.329075 Objective Loss 0.329075 LR 0.001000 Time 0.020543 +2023-10-05 21:08:17,123 - Epoch: [55][ 1236/ 1236] Overall Loss 0.329166 Objective Loss 0.329166 Top1 84.725051 Top5 97.352342 LR 0.001000 Time 0.020539 +2023-10-05 21:08:17,251 - --- validate (epoch=55)----------- +2023-10-05 21:08:17,252 - 29943 samples (256 per mini-batch) +2023-10-05 21:08:17,702 - Epoch: [55][ 10/ 117] Loss 0.380087 Top1 81.289062 Top5 96.835938 +2023-10-05 21:08:17,853 - Epoch: [55][ 20/ 117] Loss 0.364643 Top1 81.406250 Top5 97.109375 +2023-10-05 21:08:18,002 - Epoch: [55][ 30/ 117] Loss 0.371252 Top1 80.937500 Top5 97.135417 +2023-10-05 21:08:18,151 - Epoch: [55][ 40/ 117] Loss 0.366633 Top1 80.859375 Top5 97.138672 +2023-10-05 21:08:18,300 - Epoch: [55][ 50/ 117] Loss 0.366248 Top1 80.609375 Top5 97.078125 +2023-10-05 21:08:18,452 - Epoch: [55][ 60/ 117] Loss 0.362244 Top1 80.846354 Top5 97.037760 +2023-10-05 21:08:18,606 - Epoch: [55][ 70/ 117] Loss 0.360649 Top1 80.814732 Top5 97.070312 +2023-10-05 21:08:18,764 - Epoch: [55][ 80/ 117] Loss 0.362353 Top1 80.942383 Top5 97.124023 +2023-10-05 21:08:18,917 - Epoch: [55][ 90/ 117] Loss 0.364620 Top1 80.781250 Top5 97.070312 +2023-10-05 21:08:19,075 - Epoch: [55][ 100/ 117] Loss 0.364833 Top1 80.699219 Top5 97.074219 +2023-10-05 21:08:19,235 - Epoch: [55][ 110/ 117] Loss 0.363642 Top1 80.763494 Top5 97.116477 +2023-10-05 21:08:19,320 - Epoch: [55][ 117/ 117] Loss 0.364555 Top1 80.790168 Top5 97.114518 +2023-10-05 21:08:19,453 - ==> Top1: 80.790 Top5: 97.115 Loss: 0.365 + +2023-10-05 21:08:19,454 - ==> Confusion: +[[ 895 3 3 1 20 2 0 0 8 79 1 2 0 5 7 5 8 1 0 0 10] + [ 0 1058 1 1 9 9 2 19 1 0 1 4 0 2 0 4 8 0 4 1 7] + [ 2 1 931 15 3 0 31 7 0 1 4 5 7 4 2 9 1 0 12 8 13] + [ 0 4 17 958 3 2 1 1 1 1 10 1 1 5 31 5 4 9 19 4 12] + [ 15 5 0 0 977 1 0 0 1 6 1 2 3 2 8 9 13 1 0 4 2] + [ 3 65 1 1 7 930 0 27 1 0 6 12 2 21 5 2 8 2 1 10 12] + [ 0 12 25 1 0 0 1099 10 0 0 5 3 1 0 1 14 0 1 1 11 7] + [ 5 40 13 0 5 32 5 1026 0 2 4 11 5 1 0 1 1 0 36 23 8] + [ 19 0 0 0 0 3 0 0 953 30 18 3 1 19 18 6 4 1 4 3 7] + [ 83 2 1 0 8 1 2 1 31 914 2 1 0 44 7 9 1 0 0 3 9] + [ 0 4 13 7 0 0 5 2 17 0 965 2 0 15 5 4 1 0 7 0 6] + [ 1 0 0 0 0 9 1 1 0 0 0 964 16 5 0 4 1 16 0 16 1] + [ 0 0 5 1 0 3 0 1 0 0 1 40 969 1 1 10 2 12 1 10 11] + [ 1 0 1 0 2 5 0 1 12 8 7 8 2 1048 2 5 0 1 0 6 10] + [ 16 3 2 10 5 0 0 0 30 1 6 2 2 3 992 1 4 2 13 0 9] + [ 0 0 1 5 1 0 2 1 0 0 0 8 11 0 0 1066 12 13 0 12 2] + [ 0 19 1 0 11 1 1 2 1 0 0 3 1 0 2 13 1096 0 0 5 5] + [ 0 0 1 4 0 0 2 0 2 0 0 8 30 1 2 12 1 970 0 1 4] + [ 4 17 9 15 2 0 0 30 6 0 3 0 6 0 16 1 1 0 947 1 10] + [ 0 1 1 2 3 5 5 12 0 0 3 8 7 2 0 9 13 0 1 1072 8] + [ 115 288 169 82 197 124 40 94 89 64 255 141 419 327 174 116 354 61 141 294 4361]] + +2023-10-05 21:08:19,455 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:08:19,455 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:08:19,461 - + +2023-10-05 21:08:19,461 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:08:20,442 - Epoch: [56][ 10/ 1236] Overall Loss 0.288495 Objective Loss 0.288495 LR 0.001000 Time 0.098014 +2023-10-05 21:08:20,641 - Epoch: [56][ 20/ 1236] Overall Loss 0.293385 Objective Loss 0.293385 LR 0.001000 Time 0.058964 +2023-10-05 21:08:20,840 - Epoch: [56][ 30/ 1236] Overall Loss 0.308306 Objective Loss 0.308306 LR 0.001000 Time 0.045917 +2023-10-05 21:08:21,039 - Epoch: [56][ 40/ 1236] Overall Loss 0.310712 Objective Loss 0.310712 LR 0.001000 Time 0.039419 +2023-10-05 21:08:21,238 - Epoch: [56][ 50/ 1236] Overall Loss 0.314472 Objective Loss 0.314472 LR 0.001000 Time 0.035507 +2023-10-05 21:08:21,438 - Epoch: [56][ 60/ 1236] Overall Loss 0.313817 Objective Loss 0.313817 LR 0.001000 Time 0.032914 +2023-10-05 21:08:21,637 - Epoch: [56][ 70/ 1236] Overall Loss 0.313160 Objective Loss 0.313160 LR 0.001000 Time 0.031045 +2023-10-05 21:08:21,837 - Epoch: [56][ 80/ 1236] Overall Loss 0.314872 Objective Loss 0.314872 LR 0.001000 Time 0.029659 +2023-10-05 21:08:22,035 - Epoch: [56][ 90/ 1236] Overall Loss 0.316094 Objective Loss 0.316094 LR 0.001000 Time 0.028568 +2023-10-05 21:08:22,235 - Epoch: [56][ 100/ 1236] Overall Loss 0.315240 Objective Loss 0.315240 LR 0.001000 Time 0.027703 +2023-10-05 21:08:22,434 - Epoch: [56][ 110/ 1236] Overall Loss 0.314486 Objective Loss 0.314486 LR 0.001000 Time 0.026989 +2023-10-05 21:08:22,633 - Epoch: [56][ 120/ 1236] Overall Loss 0.315674 Objective Loss 0.315674 LR 0.001000 Time 0.026400 +2023-10-05 21:08:22,832 - Epoch: [56][ 130/ 1236] Overall Loss 0.313924 Objective Loss 0.313924 LR 0.001000 Time 0.025893 +2023-10-05 21:08:23,031 - Epoch: [56][ 140/ 1236] Overall Loss 0.312987 Objective Loss 0.312987 LR 0.001000 Time 0.025466 +2023-10-05 21:08:23,230 - Epoch: [56][ 150/ 1236] Overall Loss 0.312067 Objective Loss 0.312067 LR 0.001000 Time 0.025090 +2023-10-05 21:08:23,430 - Epoch: [56][ 160/ 1236] Overall Loss 0.313427 Objective Loss 0.313427 LR 0.001000 Time 0.024769 +2023-10-05 21:08:23,628 - Epoch: [56][ 170/ 1236] Overall Loss 0.313913 Objective Loss 0.313913 LR 0.001000 Time 0.024479 +2023-10-05 21:08:23,828 - Epoch: [56][ 180/ 1236] Overall Loss 0.313608 Objective Loss 0.313608 LR 0.001000 Time 0.024226 +2023-10-05 21:08:24,027 - Epoch: [56][ 190/ 1236] Overall Loss 0.314615 Objective Loss 0.314615 LR 0.001000 Time 0.023995 +2023-10-05 21:08:24,226 - Epoch: [56][ 200/ 1236] Overall Loss 0.314572 Objective Loss 0.314572 LR 0.001000 Time 0.023790 +2023-10-05 21:08:24,425 - Epoch: [56][ 210/ 1236] Overall Loss 0.315597 Objective Loss 0.315597 LR 0.001000 Time 0.023602 +2023-10-05 21:08:24,624 - Epoch: [56][ 220/ 1236] Overall Loss 0.315789 Objective Loss 0.315789 LR 0.001000 Time 0.023435 +2023-10-05 21:08:24,823 - Epoch: [56][ 230/ 1236] Overall Loss 0.316454 Objective Loss 0.316454 LR 0.001000 Time 0.023277 +2023-10-05 21:08:25,023 - Epoch: [56][ 240/ 1236] Overall Loss 0.315813 Objective Loss 0.315813 LR 0.001000 Time 0.023138 +2023-10-05 21:08:25,221 - Epoch: [56][ 250/ 1236] Overall Loss 0.315855 Objective Loss 0.315855 LR 0.001000 Time 0.023005 +2023-10-05 21:08:25,421 - Epoch: [56][ 260/ 1236] Overall Loss 0.315112 Objective Loss 0.315112 LR 0.001000 Time 0.022887 +2023-10-05 21:08:25,620 - Epoch: [56][ 270/ 1236] Overall Loss 0.315469 Objective Loss 0.315469 LR 0.001000 Time 0.022775 +2023-10-05 21:08:25,819 - Epoch: [56][ 280/ 1236] Overall Loss 0.315492 Objective Loss 0.315492 LR 0.001000 Time 0.022673 +2023-10-05 21:08:26,018 - Epoch: [56][ 290/ 1236] Overall Loss 0.315165 Objective Loss 0.315165 LR 0.001000 Time 0.022577 +2023-10-05 21:08:26,218 - Epoch: [56][ 300/ 1236] Overall Loss 0.315099 Objective Loss 0.315099 LR 0.001000 Time 0.022489 +2023-10-05 21:08:26,417 - Epoch: [56][ 310/ 1236] Overall Loss 0.314054 Objective Loss 0.314054 LR 0.001000 Time 0.022404 +2023-10-05 21:08:26,617 - Epoch: [56][ 320/ 1236] Overall Loss 0.314144 Objective Loss 0.314144 LR 0.001000 Time 0.022327 +2023-10-05 21:08:26,816 - Epoch: [56][ 330/ 1236] Overall Loss 0.314880 Objective Loss 0.314880 LR 0.001000 Time 0.022253 +2023-10-05 21:08:27,016 - Epoch: [56][ 340/ 1236] Overall Loss 0.315240 Objective Loss 0.315240 LR 0.001000 Time 0.022186 +2023-10-05 21:08:27,215 - Epoch: [56][ 350/ 1236] Overall Loss 0.315053 Objective Loss 0.315053 LR 0.001000 Time 0.022120 +2023-10-05 21:08:27,415 - Epoch: [56][ 360/ 1236] Overall Loss 0.315123 Objective Loss 0.315123 LR 0.001000 Time 0.022060 +2023-10-05 21:08:27,614 - Epoch: [56][ 370/ 1236] Overall Loss 0.314669 Objective Loss 0.314669 LR 0.001000 Time 0.022001 +2023-10-05 21:08:27,814 - Epoch: [56][ 380/ 1236] Overall Loss 0.315192 Objective Loss 0.315192 LR 0.001000 Time 0.021947 +2023-10-05 21:08:28,013 - Epoch: [56][ 390/ 1236] Overall Loss 0.315195 Objective Loss 0.315195 LR 0.001000 Time 0.021893 +2023-10-05 21:08:28,213 - Epoch: [56][ 400/ 1236] Overall Loss 0.315472 Objective Loss 0.315472 LR 0.001000 Time 0.021845 +2023-10-05 21:08:28,412 - Epoch: [56][ 410/ 1236] Overall Loss 0.315701 Objective Loss 0.315701 LR 0.001000 Time 0.021796 +2023-10-05 21:08:28,612 - Epoch: [56][ 420/ 1236] Overall Loss 0.315968 Objective Loss 0.315968 LR 0.001000 Time 0.021752 +2023-10-05 21:08:28,811 - Epoch: [56][ 430/ 1236] Overall Loss 0.316330 Objective Loss 0.316330 LR 0.001000 Time 0.021709 +2023-10-05 21:08:29,011 - Epoch: [56][ 440/ 1236] Overall Loss 0.316422 Objective Loss 0.316422 LR 0.001000 Time 0.021669 +2023-10-05 21:08:29,210 - Epoch: [56][ 450/ 1236] Overall Loss 0.317131 Objective Loss 0.317131 LR 0.001000 Time 0.021629 +2023-10-05 21:08:29,409 - Epoch: [56][ 460/ 1236] Overall Loss 0.317707 Objective Loss 0.317707 LR 0.001000 Time 0.021592 +2023-10-05 21:08:29,608 - Epoch: [56][ 470/ 1236] Overall Loss 0.318740 Objective Loss 0.318740 LR 0.001000 Time 0.021554 +2023-10-05 21:08:29,808 - Epoch: [56][ 480/ 1236] Overall Loss 0.318796 Objective Loss 0.318796 LR 0.001000 Time 0.021521 +2023-10-05 21:08:30,006 - Epoch: [56][ 490/ 1236] Overall Loss 0.319447 Objective Loss 0.319447 LR 0.001000 Time 0.021486 +2023-10-05 21:08:30,206 - Epoch: [56][ 500/ 1236] Overall Loss 0.319005 Objective Loss 0.319005 LR 0.001000 Time 0.021455 +2023-10-05 21:08:30,405 - Epoch: [56][ 510/ 1236] Overall Loss 0.320310 Objective Loss 0.320310 LR 0.001000 Time 0.021423 +2023-10-05 21:08:30,604 - Epoch: [56][ 520/ 1236] Overall Loss 0.320038 Objective Loss 0.320038 LR 0.001000 Time 0.021395 +2023-10-05 21:08:30,803 - Epoch: [56][ 530/ 1236] Overall Loss 0.320140 Objective Loss 0.320140 LR 0.001000 Time 0.021366 +2023-10-05 21:08:31,003 - Epoch: [56][ 540/ 1236] Overall Loss 0.319988 Objective Loss 0.319988 LR 0.001000 Time 0.021339 +2023-10-05 21:08:31,202 - Epoch: [56][ 550/ 1236] Overall Loss 0.320104 Objective Loss 0.320104 LR 0.001000 Time 0.021312 +2023-10-05 21:08:31,402 - Epoch: [56][ 560/ 1236] Overall Loss 0.319904 Objective Loss 0.319904 LR 0.001000 Time 0.021288 +2023-10-05 21:08:31,600 - Epoch: [56][ 570/ 1236] Overall Loss 0.319485 Objective Loss 0.319485 LR 0.001000 Time 0.021262 +2023-10-05 21:08:31,801 - Epoch: [56][ 580/ 1236] Overall Loss 0.320380 Objective Loss 0.320380 LR 0.001000 Time 0.021241 +2023-10-05 21:08:32,002 - Epoch: [56][ 590/ 1236] Overall Loss 0.320621 Objective Loss 0.320621 LR 0.001000 Time 0.021221 +2023-10-05 21:08:32,203 - Epoch: [56][ 600/ 1236] Overall Loss 0.321111 Objective Loss 0.321111 LR 0.001000 Time 0.021202 +2023-10-05 21:08:32,403 - Epoch: [56][ 610/ 1236] Overall Loss 0.321344 Objective Loss 0.321344 LR 0.001000 Time 0.021182 +2023-10-05 21:08:32,604 - Epoch: [56][ 620/ 1236] Overall Loss 0.320826 Objective Loss 0.320826 LR 0.001000 Time 0.021164 +2023-10-05 21:08:32,804 - Epoch: [56][ 630/ 1236] Overall Loss 0.320881 Objective Loss 0.320881 LR 0.001000 Time 0.021145 +2023-10-05 21:08:33,005 - Epoch: [56][ 640/ 1236] Overall Loss 0.320966 Objective Loss 0.320966 LR 0.001000 Time 0.021128 +2023-10-05 21:08:33,205 - Epoch: [56][ 650/ 1236] Overall Loss 0.321633 Objective Loss 0.321633 LR 0.001000 Time 0.021111 +2023-10-05 21:08:33,407 - Epoch: [56][ 660/ 1236] Overall Loss 0.321877 Objective Loss 0.321877 LR 0.001000 Time 0.021095 +2023-10-05 21:08:33,607 - Epoch: [56][ 670/ 1236] Overall Loss 0.321389 Objective Loss 0.321389 LR 0.001000 Time 0.021079 +2023-10-05 21:08:33,808 - Epoch: [56][ 680/ 1236] Overall Loss 0.321031 Objective Loss 0.321031 LR 0.001000 Time 0.021064 +2023-10-05 21:08:34,008 - Epoch: [56][ 690/ 1236] Overall Loss 0.321034 Objective Loss 0.321034 LR 0.001000 Time 0.021048 +2023-10-05 21:08:34,208 - Epoch: [56][ 700/ 1236] Overall Loss 0.321363 Objective Loss 0.321363 LR 0.001000 Time 0.021033 +2023-10-05 21:08:34,407 - Epoch: [56][ 710/ 1236] Overall Loss 0.321825 Objective Loss 0.321825 LR 0.001000 Time 0.021017 +2023-10-05 21:08:34,609 - Epoch: [56][ 720/ 1236] Overall Loss 0.321882 Objective Loss 0.321882 LR 0.001000 Time 0.021004 +2023-10-05 21:08:34,809 - Epoch: [56][ 730/ 1236] Overall Loss 0.321727 Objective Loss 0.321727 LR 0.001000 Time 0.020991 +2023-10-05 21:08:35,010 - Epoch: [56][ 740/ 1236] Overall Loss 0.321509 Objective Loss 0.321509 LR 0.001000 Time 0.020978 +2023-10-05 21:08:35,210 - Epoch: [56][ 750/ 1236] Overall Loss 0.321543 Objective Loss 0.321543 LR 0.001000 Time 0.020965 +2023-10-05 21:08:35,411 - Epoch: [56][ 760/ 1236] Overall Loss 0.321809 Objective Loss 0.321809 LR 0.001000 Time 0.020953 +2023-10-05 21:08:35,612 - Epoch: [56][ 770/ 1236] Overall Loss 0.321818 Objective Loss 0.321818 LR 0.001000 Time 0.020941 +2023-10-05 21:08:35,813 - Epoch: [56][ 780/ 1236] Overall Loss 0.321684 Objective Loss 0.321684 LR 0.001000 Time 0.020930 +2023-10-05 21:08:36,013 - Epoch: [56][ 790/ 1236] Overall Loss 0.321489 Objective Loss 0.321489 LR 0.001000 Time 0.020918 +2023-10-05 21:08:36,214 - Epoch: [56][ 800/ 1236] Overall Loss 0.321240 Objective Loss 0.321240 LR 0.001000 Time 0.020907 +2023-10-05 21:08:36,414 - Epoch: [56][ 810/ 1236] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.020896 +2023-10-05 21:08:36,615 - Epoch: [56][ 820/ 1236] Overall Loss 0.321545 Objective Loss 0.321545 LR 0.001000 Time 0.020886 +2023-10-05 21:08:36,816 - Epoch: [56][ 830/ 1236] Overall Loss 0.321263 Objective Loss 0.321263 LR 0.001000 Time 0.020876 +2023-10-05 21:08:37,017 - Epoch: [56][ 840/ 1236] Overall Loss 0.321466 Objective Loss 0.321466 LR 0.001000 Time 0.020866 +2023-10-05 21:08:37,218 - Epoch: [56][ 850/ 1236] Overall Loss 0.321749 Objective Loss 0.321749 LR 0.001000 Time 0.020857 +2023-10-05 21:08:37,419 - Epoch: [56][ 860/ 1236] Overall Loss 0.321893 Objective Loss 0.321893 LR 0.001000 Time 0.020847 +2023-10-05 21:08:37,619 - Epoch: [56][ 870/ 1236] Overall Loss 0.321905 Objective Loss 0.321905 LR 0.001000 Time 0.020838 +2023-10-05 21:08:37,821 - Epoch: [56][ 880/ 1236] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.020830 +2023-10-05 21:08:38,022 - Epoch: [56][ 890/ 1236] Overall Loss 0.322391 Objective Loss 0.322391 LR 0.001000 Time 0.020821 +2023-10-05 21:08:38,223 - Epoch: [56][ 900/ 1236] Overall Loss 0.322604 Objective Loss 0.322604 LR 0.001000 Time 0.020813 +2023-10-05 21:08:38,423 - Epoch: [56][ 910/ 1236] Overall Loss 0.322869 Objective Loss 0.322869 LR 0.001000 Time 0.020804 +2023-10-05 21:08:38,624 - Epoch: [56][ 920/ 1236] Overall Loss 0.323082 Objective Loss 0.323082 LR 0.001000 Time 0.020795 +2023-10-05 21:08:38,825 - Epoch: [56][ 930/ 1236] Overall Loss 0.322989 Objective Loss 0.322989 LR 0.001000 Time 0.020788 +2023-10-05 21:08:39,026 - Epoch: [56][ 940/ 1236] Overall Loss 0.323150 Objective Loss 0.323150 LR 0.001000 Time 0.020780 +2023-10-05 21:08:39,226 - Epoch: [56][ 950/ 1236] Overall Loss 0.323378 Objective Loss 0.323378 LR 0.001000 Time 0.020772 +2023-10-05 21:08:39,427 - Epoch: [56][ 960/ 1236] Overall Loss 0.323976 Objective Loss 0.323976 LR 0.001000 Time 0.020764 +2023-10-05 21:08:39,627 - Epoch: [56][ 970/ 1236] Overall Loss 0.323989 Objective Loss 0.323989 LR 0.001000 Time 0.020756 +2023-10-05 21:08:39,828 - Epoch: [56][ 980/ 1236] Overall Loss 0.324179 Objective Loss 0.324179 LR 0.001000 Time 0.020749 +2023-10-05 21:08:40,029 - Epoch: [56][ 990/ 1236] Overall Loss 0.324342 Objective Loss 0.324342 LR 0.001000 Time 0.020742 +2023-10-05 21:08:40,230 - Epoch: [56][ 1000/ 1236] Overall Loss 0.324335 Objective Loss 0.324335 LR 0.001000 Time 0.020735 +2023-10-05 21:08:40,430 - Epoch: [56][ 1010/ 1236] Overall Loss 0.324536 Objective Loss 0.324536 LR 0.001000 Time 0.020728 +2023-10-05 21:08:40,631 - Epoch: [56][ 1020/ 1236] Overall Loss 0.324621 Objective Loss 0.324621 LR 0.001000 Time 0.020722 +2023-10-05 21:08:40,832 - Epoch: [56][ 1030/ 1236] Overall Loss 0.325353 Objective Loss 0.325353 LR 0.001000 Time 0.020715 +2023-10-05 21:08:41,033 - Epoch: [56][ 1040/ 1236] Overall Loss 0.325269 Objective Loss 0.325269 LR 0.001000 Time 0.020709 +2023-10-05 21:08:41,234 - Epoch: [56][ 1050/ 1236] Overall Loss 0.325185 Objective Loss 0.325185 LR 0.001000 Time 0.020702 +2023-10-05 21:08:41,435 - Epoch: [56][ 1060/ 1236] Overall Loss 0.325180 Objective Loss 0.325180 LR 0.001000 Time 0.020696 +2023-10-05 21:08:41,635 - Epoch: [56][ 1070/ 1236] Overall Loss 0.325310 Objective Loss 0.325310 LR 0.001000 Time 0.020689 +2023-10-05 21:08:41,836 - Epoch: [56][ 1080/ 1236] Overall Loss 0.325599 Objective Loss 0.325599 LR 0.001000 Time 0.020684 +2023-10-05 21:08:42,037 - Epoch: [56][ 1090/ 1236] Overall Loss 0.325456 Objective Loss 0.325456 LR 0.001000 Time 0.020678 +2023-10-05 21:08:42,238 - Epoch: [56][ 1100/ 1236] Overall Loss 0.325385 Objective Loss 0.325385 LR 0.001000 Time 0.020672 +2023-10-05 21:08:42,438 - Epoch: [56][ 1110/ 1236] Overall Loss 0.325351 Objective Loss 0.325351 LR 0.001000 Time 0.020667 +2023-10-05 21:08:42,639 - Epoch: [56][ 1120/ 1236] Overall Loss 0.325413 Objective Loss 0.325413 LR 0.001000 Time 0.020661 +2023-10-05 21:08:42,839 - Epoch: [56][ 1130/ 1236] Overall Loss 0.325855 Objective Loss 0.325855 LR 0.001000 Time 0.020655 +2023-10-05 21:08:43,040 - Epoch: [56][ 1140/ 1236] Overall Loss 0.325896 Objective Loss 0.325896 LR 0.001000 Time 0.020650 +2023-10-05 21:08:43,241 - Epoch: [56][ 1150/ 1236] Overall Loss 0.326017 Objective Loss 0.326017 LR 0.001000 Time 0.020645 +2023-10-05 21:08:43,442 - Epoch: [56][ 1160/ 1236] Overall Loss 0.326171 Objective Loss 0.326171 LR 0.001000 Time 0.020640 +2023-10-05 21:08:43,643 - Epoch: [56][ 1170/ 1236] Overall Loss 0.326277 Objective Loss 0.326277 LR 0.001000 Time 0.020635 +2023-10-05 21:08:43,844 - Epoch: [56][ 1180/ 1236] Overall Loss 0.326563 Objective Loss 0.326563 LR 0.001000 Time 0.020630 +2023-10-05 21:08:44,045 - Epoch: [56][ 1190/ 1236] Overall Loss 0.326676 Objective Loss 0.326676 LR 0.001000 Time 0.020625 +2023-10-05 21:08:44,245 - Epoch: [56][ 1200/ 1236] Overall Loss 0.326808 Objective Loss 0.326808 LR 0.001000 Time 0.020620 +2023-10-05 21:08:44,446 - Epoch: [56][ 1210/ 1236] Overall Loss 0.326927 Objective Loss 0.326927 LR 0.001000 Time 0.020615 +2023-10-05 21:08:44,647 - Epoch: [56][ 1220/ 1236] Overall Loss 0.327006 Objective Loss 0.327006 LR 0.001000 Time 0.020610 +2023-10-05 21:08:44,899 - Epoch: [56][ 1230/ 1236] Overall Loss 0.327189 Objective Loss 0.327189 LR 0.001000 Time 0.020648 +2023-10-05 21:08:45,016 - Epoch: [56][ 1236/ 1236] Overall Loss 0.327044 Objective Loss 0.327044 Top1 85.336049 Top5 97.556008 LR 0.001000 Time 0.020642 +2023-10-05 21:08:45,157 - --- validate (epoch=56)----------- +2023-10-05 21:08:45,157 - 29943 samples (256 per mini-batch) +2023-10-05 21:08:45,612 - Epoch: [56][ 10/ 117] Loss 0.376884 Top1 80.937500 Top5 97.617188 +2023-10-05 21:08:45,768 - Epoch: [56][ 20/ 117] Loss 0.378477 Top1 81.035156 Top5 97.441406 +2023-10-05 21:08:45,920 - Epoch: [56][ 30/ 117] Loss 0.364258 Top1 81.835938 Top5 97.591146 +2023-10-05 21:08:46,075 - Epoch: [56][ 40/ 117] Loss 0.369982 Top1 81.875000 Top5 97.607422 +2023-10-05 21:08:46,229 - Epoch: [56][ 50/ 117] Loss 0.376207 Top1 81.601562 Top5 97.539062 +2023-10-05 21:08:46,384 - Epoch: [56][ 60/ 117] Loss 0.378350 Top1 81.562500 Top5 97.493490 +2023-10-05 21:08:46,532 - Epoch: [56][ 70/ 117] Loss 0.377548 Top1 81.450893 Top5 97.527902 +2023-10-05 21:08:46,679 - Epoch: [56][ 80/ 117] Loss 0.375370 Top1 81.372070 Top5 97.460938 +2023-10-05 21:08:46,826 - Epoch: [56][ 90/ 117] Loss 0.374406 Top1 81.497396 Top5 97.443576 +2023-10-05 21:08:46,972 - Epoch: [56][ 100/ 117] Loss 0.371118 Top1 81.484375 Top5 97.433594 +2023-10-05 21:08:47,124 - Epoch: [56][ 110/ 117] Loss 0.374587 Top1 81.278409 Top5 97.414773 +2023-10-05 21:08:47,209 - Epoch: [56][ 117/ 117] Loss 0.373110 Top1 81.357913 Top5 97.461844 +2023-10-05 21:08:47,341 - ==> Top1: 81.358 Top5: 97.462 Loss: 0.373 + +2023-10-05 21:08:47,342 - ==> Confusion: +[[ 930 4 4 4 16 6 0 1 3 51 0 0 0 6 6 3 4 0 2 0 10] + [ 3 1038 2 0 10 24 1 19 2 0 1 3 0 0 0 4 7 0 7 2 8] + [ 6 0 931 18 4 1 31 5 0 0 8 3 6 4 3 4 3 1 12 4 12] + [ 6 2 23 948 2 3 2 0 4 0 4 0 8 2 24 5 6 11 22 4 13] + [ 23 5 0 0 986 5 0 0 0 5 0 0 1 1 6 3 7 2 1 1 4] + [ 6 46 1 1 10 959 0 23 4 0 3 7 6 15 4 1 3 2 4 8 13] + [ 1 7 35 1 0 0 1104 8 0 0 8 2 1 1 1 6 0 2 2 7 5] + [ 5 31 14 0 1 46 5 999 1 2 7 7 3 3 2 1 0 0 65 19 7] + [ 29 2 0 1 2 4 0 1 936 43 15 2 1 19 17 3 2 0 12 0 0] + [ 142 4 2 1 8 4 1 0 20 860 1 2 0 49 6 4 1 0 2 3 9] + [ 4 6 15 4 0 0 4 2 16 0 957 1 0 14 5 3 2 3 7 0 10] + [ 1 0 1 0 0 11 1 1 1 0 0 969 22 3 0 3 2 14 0 4 2] + [ 2 1 2 3 1 2 1 2 3 1 0 52 954 1 2 12 4 14 1 2 8] + [ 4 1 3 0 4 10 1 0 4 9 8 8 1 1041 3 4 4 0 1 4 9] + [ 15 2 4 12 9 0 0 0 26 2 1 1 2 2 1001 2 3 2 10 0 7] + [ 0 1 1 0 4 0 1 0 1 0 0 12 9 0 0 1064 16 9 1 8 7] + [ 0 18 1 0 7 4 0 1 2 0 0 8 1 1 2 14 1089 1 0 1 11] + [ 4 1 0 2 1 0 0 0 2 0 0 16 19 0 5 4 1 977 2 0 4] + [ 1 12 7 18 2 1 1 26 7 0 6 2 1 0 10 2 1 0 961 2 8] + [ 0 3 1 0 2 4 4 8 0 0 3 22 7 3 0 11 12 0 1 1058 13] + [ 194 244 129 88 172 172 45 78 111 55 190 160 375 314 153 93 270 70 177 216 4599]] + +2023-10-05 21:08:47,343 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:08:47,343 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:08:47,349 - + +2023-10-05 21:08:47,349 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:08:48,444 - Epoch: [57][ 10/ 1236] Overall Loss 0.324808 Objective Loss 0.324808 LR 0.001000 Time 0.109474 +2023-10-05 21:08:48,645 - Epoch: [57][ 20/ 1236] Overall Loss 0.317223 Objective Loss 0.317223 LR 0.001000 Time 0.064781 +2023-10-05 21:08:48,846 - Epoch: [57][ 30/ 1236] Overall Loss 0.316504 Objective Loss 0.316504 LR 0.001000 Time 0.049863 +2023-10-05 21:08:49,047 - Epoch: [57][ 40/ 1236] Overall Loss 0.314614 Objective Loss 0.314614 LR 0.001000 Time 0.042423 +2023-10-05 21:08:49,248 - Epoch: [57][ 50/ 1236] Overall Loss 0.320226 Objective Loss 0.320226 LR 0.001000 Time 0.037955 +2023-10-05 21:08:49,449 - Epoch: [57][ 60/ 1236] Overall Loss 0.317472 Objective Loss 0.317472 LR 0.001000 Time 0.034973 +2023-10-05 21:08:49,651 - Epoch: [57][ 70/ 1236] Overall Loss 0.320023 Objective Loss 0.320023 LR 0.001000 Time 0.032852 +2023-10-05 21:08:49,852 - Epoch: [57][ 80/ 1236] Overall Loss 0.321104 Objective Loss 0.321104 LR 0.001000 Time 0.031253 +2023-10-05 21:08:50,053 - Epoch: [57][ 90/ 1236] Overall Loss 0.320186 Objective Loss 0.320186 LR 0.001000 Time 0.030012 +2023-10-05 21:08:50,254 - Epoch: [57][ 100/ 1236] Overall Loss 0.321420 Objective Loss 0.321420 LR 0.001000 Time 0.029017 +2023-10-05 21:08:50,453 - Epoch: [57][ 110/ 1236] Overall Loss 0.322297 Objective Loss 0.322297 LR 0.001000 Time 0.028183 +2023-10-05 21:08:50,653 - Epoch: [57][ 120/ 1236] Overall Loss 0.322118 Objective Loss 0.322118 LR 0.001000 Time 0.027501 +2023-10-05 21:08:50,853 - Epoch: [57][ 130/ 1236] Overall Loss 0.320533 Objective Loss 0.320533 LR 0.001000 Time 0.026925 +2023-10-05 21:08:51,054 - Epoch: [57][ 140/ 1236] Overall Loss 0.319622 Objective Loss 0.319622 LR 0.001000 Time 0.026430 +2023-10-05 21:08:51,255 - Epoch: [57][ 150/ 1236] Overall Loss 0.319757 Objective Loss 0.319757 LR 0.001000 Time 0.026004 +2023-10-05 21:08:51,455 - Epoch: [57][ 160/ 1236] Overall Loss 0.318880 Objective Loss 0.318880 LR 0.001000 Time 0.025629 +2023-10-05 21:08:51,655 - Epoch: [57][ 170/ 1236] Overall Loss 0.319539 Objective Loss 0.319539 LR 0.001000 Time 0.025294 +2023-10-05 21:08:51,856 - Epoch: [57][ 180/ 1236] Overall Loss 0.320917 Objective Loss 0.320917 LR 0.001000 Time 0.025009 +2023-10-05 21:08:52,057 - Epoch: [57][ 190/ 1236] Overall Loss 0.320950 Objective Loss 0.320950 LR 0.001000 Time 0.024745 +2023-10-05 21:08:52,258 - Epoch: [57][ 200/ 1236] Overall Loss 0.320967 Objective Loss 0.320967 LR 0.001000 Time 0.024511 +2023-10-05 21:08:52,458 - Epoch: [57][ 210/ 1236] Overall Loss 0.322321 Objective Loss 0.322321 LR 0.001000 Time 0.024294 +2023-10-05 21:08:52,658 - Epoch: [57][ 220/ 1236] Overall Loss 0.322202 Objective Loss 0.322202 LR 0.001000 Time 0.024098 +2023-10-05 21:08:52,858 - Epoch: [57][ 230/ 1236] Overall Loss 0.320497 Objective Loss 0.320497 LR 0.001000 Time 0.023918 +2023-10-05 21:08:53,058 - Epoch: [57][ 240/ 1236] Overall Loss 0.320498 Objective Loss 0.320498 LR 0.001000 Time 0.023754 +2023-10-05 21:08:53,257 - Epoch: [57][ 250/ 1236] Overall Loss 0.320387 Objective Loss 0.320387 LR 0.001000 Time 0.023601 +2023-10-05 21:08:53,458 - Epoch: [57][ 260/ 1236] Overall Loss 0.321726 Objective Loss 0.321726 LR 0.001000 Time 0.023463 +2023-10-05 21:08:53,658 - Epoch: [57][ 270/ 1236] Overall Loss 0.320377 Objective Loss 0.320377 LR 0.001000 Time 0.023333 +2023-10-05 21:08:53,858 - Epoch: [57][ 280/ 1236] Overall Loss 0.320112 Objective Loss 0.320112 LR 0.001000 Time 0.023213 +2023-10-05 21:08:54,058 - Epoch: [57][ 290/ 1236] Overall Loss 0.320865 Objective Loss 0.320865 LR 0.001000 Time 0.023101 +2023-10-05 21:08:54,258 - Epoch: [57][ 300/ 1236] Overall Loss 0.319998 Objective Loss 0.319998 LR 0.001000 Time 0.022997 +2023-10-05 21:08:54,458 - Epoch: [57][ 310/ 1236] Overall Loss 0.319732 Objective Loss 0.319732 LR 0.001000 Time 0.022899 +2023-10-05 21:08:54,658 - Epoch: [57][ 320/ 1236] Overall Loss 0.318859 Objective Loss 0.318859 LR 0.001000 Time 0.022808 +2023-10-05 21:08:54,858 - Epoch: [57][ 330/ 1236] Overall Loss 0.318370 Objective Loss 0.318370 LR 0.001000 Time 0.022722 +2023-10-05 21:08:55,059 - Epoch: [57][ 340/ 1236] Overall Loss 0.318203 Objective Loss 0.318203 LR 0.001000 Time 0.022643 +2023-10-05 21:08:55,259 - Epoch: [57][ 350/ 1236] Overall Loss 0.318115 Objective Loss 0.318115 LR 0.001000 Time 0.022567 +2023-10-05 21:08:55,459 - Epoch: [57][ 360/ 1236] Overall Loss 0.317051 Objective Loss 0.317051 LR 0.001000 Time 0.022495 +2023-10-05 21:08:55,659 - Epoch: [57][ 370/ 1236] Overall Loss 0.316895 Objective Loss 0.316895 LR 0.001000 Time 0.022427 +2023-10-05 21:08:55,861 - Epoch: [57][ 380/ 1236] Overall Loss 0.316928 Objective Loss 0.316928 LR 0.001000 Time 0.022368 +2023-10-05 21:08:56,062 - Epoch: [57][ 390/ 1236] Overall Loss 0.316803 Objective Loss 0.316803 LR 0.001000 Time 0.022308 +2023-10-05 21:08:56,264 - Epoch: [57][ 400/ 1236] Overall Loss 0.317564 Objective Loss 0.317564 LR 0.001000 Time 0.022254 +2023-10-05 21:08:56,464 - Epoch: [57][ 410/ 1236] Overall Loss 0.317482 Objective Loss 0.317482 LR 0.001000 Time 0.022200 +2023-10-05 21:08:56,666 - Epoch: [57][ 420/ 1236] Overall Loss 0.317438 Objective Loss 0.317438 LR 0.001000 Time 0.022151 +2023-10-05 21:08:56,867 - Epoch: [57][ 430/ 1236] Overall Loss 0.317752 Objective Loss 0.317752 LR 0.001000 Time 0.022102 +2023-10-05 21:08:57,069 - Epoch: [57][ 440/ 1236] Overall Loss 0.317482 Objective Loss 0.317482 LR 0.001000 Time 0.022058 +2023-10-05 21:08:57,269 - Epoch: [57][ 450/ 1236] Overall Loss 0.317131 Objective Loss 0.317131 LR 0.001000 Time 0.022013 +2023-10-05 21:08:57,471 - Epoch: [57][ 460/ 1236] Overall Loss 0.316668 Objective Loss 0.316668 LR 0.001000 Time 0.021972 +2023-10-05 21:08:57,672 - Epoch: [57][ 470/ 1236] Overall Loss 0.317420 Objective Loss 0.317420 LR 0.001000 Time 0.021930 +2023-10-05 21:08:57,874 - Epoch: [57][ 480/ 1236] Overall Loss 0.317722 Objective Loss 0.317722 LR 0.001000 Time 0.021894 +2023-10-05 21:08:58,074 - Epoch: [57][ 490/ 1236] Overall Loss 0.318226 Objective Loss 0.318226 LR 0.001000 Time 0.021855 +2023-10-05 21:08:58,276 - Epoch: [57][ 500/ 1236] Overall Loss 0.319576 Objective Loss 0.319576 LR 0.001000 Time 0.021822 +2023-10-05 21:08:58,477 - Epoch: [57][ 510/ 1236] Overall Loss 0.319370 Objective Loss 0.319370 LR 0.001000 Time 0.021786 +2023-10-05 21:08:58,679 - Epoch: [57][ 520/ 1236] Overall Loss 0.319044 Objective Loss 0.319044 LR 0.001000 Time 0.021755 +2023-10-05 21:08:58,879 - Epoch: [57][ 530/ 1236] Overall Loss 0.319359 Objective Loss 0.319359 LR 0.001000 Time 0.021722 +2023-10-05 21:08:59,081 - Epoch: [57][ 540/ 1236] Overall Loss 0.318567 Objective Loss 0.318567 LR 0.001000 Time 0.021693 +2023-10-05 21:08:59,282 - Epoch: [57][ 550/ 1236] Overall Loss 0.318618 Objective Loss 0.318618 LR 0.001000 Time 0.021663 +2023-10-05 21:08:59,484 - Epoch: [57][ 560/ 1236] Overall Loss 0.318715 Objective Loss 0.318715 LR 0.001000 Time 0.021636 +2023-10-05 21:08:59,684 - Epoch: [57][ 570/ 1236] Overall Loss 0.319145 Objective Loss 0.319145 LR 0.001000 Time 0.021608 +2023-10-05 21:08:59,886 - Epoch: [57][ 580/ 1236] Overall Loss 0.319244 Objective Loss 0.319244 LR 0.001000 Time 0.021583 +2023-10-05 21:09:00,087 - Epoch: [57][ 590/ 1236] Overall Loss 0.319281 Objective Loss 0.319281 LR 0.001000 Time 0.021557 +2023-10-05 21:09:00,289 - Epoch: [57][ 600/ 1236] Overall Loss 0.318919 Objective Loss 0.318919 LR 0.001000 Time 0.021534 +2023-10-05 21:09:00,490 - Epoch: [57][ 610/ 1236] Overall Loss 0.318954 Objective Loss 0.318954 LR 0.001000 Time 0.021509 +2023-10-05 21:09:00,692 - Epoch: [57][ 620/ 1236] Overall Loss 0.319015 Objective Loss 0.319015 LR 0.001000 Time 0.021487 +2023-10-05 21:09:00,892 - Epoch: [57][ 630/ 1236] Overall Loss 0.319231 Objective Loss 0.319231 LR 0.001000 Time 0.021464 +2023-10-05 21:09:01,094 - Epoch: [57][ 640/ 1236] Overall Loss 0.319422 Objective Loss 0.319422 LR 0.001000 Time 0.021444 +2023-10-05 21:09:01,295 - Epoch: [57][ 650/ 1236] Overall Loss 0.319056 Objective Loss 0.319056 LR 0.001000 Time 0.021422 +2023-10-05 21:09:01,497 - Epoch: [57][ 660/ 1236] Overall Loss 0.318689 Objective Loss 0.318689 LR 0.001000 Time 0.021403 +2023-10-05 21:09:01,697 - Epoch: [57][ 670/ 1236] Overall Loss 0.318940 Objective Loss 0.318940 LR 0.001000 Time 0.021382 +2023-10-05 21:09:01,899 - Epoch: [57][ 680/ 1236] Overall Loss 0.319041 Objective Loss 0.319041 LR 0.001000 Time 0.021365 +2023-10-05 21:09:02,100 - Epoch: [57][ 690/ 1236] Overall Loss 0.318917 Objective Loss 0.318917 LR 0.001000 Time 0.021345 +2023-10-05 21:09:02,302 - Epoch: [57][ 700/ 1236] Overall Loss 0.319171 Objective Loss 0.319171 LR 0.001000 Time 0.021328 +2023-10-05 21:09:02,502 - Epoch: [57][ 710/ 1236] Overall Loss 0.319183 Objective Loss 0.319183 LR 0.001000 Time 0.021309 +2023-10-05 21:09:02,704 - Epoch: [57][ 720/ 1236] Overall Loss 0.319113 Objective Loss 0.319113 LR 0.001000 Time 0.021294 +2023-10-05 21:09:02,905 - Epoch: [57][ 730/ 1236] Overall Loss 0.319374 Objective Loss 0.319374 LR 0.001000 Time 0.021276 +2023-10-05 21:09:03,107 - Epoch: [57][ 740/ 1236] Overall Loss 0.319792 Objective Loss 0.319792 LR 0.001000 Time 0.021261 +2023-10-05 21:09:03,307 - Epoch: [57][ 750/ 1236] Overall Loss 0.319812 Objective Loss 0.319812 LR 0.001000 Time 0.021245 +2023-10-05 21:09:03,509 - Epoch: [57][ 760/ 1236] Overall Loss 0.319969 Objective Loss 0.319969 LR 0.001000 Time 0.021230 +2023-10-05 21:09:03,710 - Epoch: [57][ 770/ 1236] Overall Loss 0.320600 Objective Loss 0.320600 LR 0.001000 Time 0.021214 +2023-10-05 21:09:03,912 - Epoch: [57][ 780/ 1236] Overall Loss 0.321244 Objective Loss 0.321244 LR 0.001000 Time 0.021201 +2023-10-05 21:09:04,112 - Epoch: [57][ 790/ 1236] Overall Loss 0.321656 Objective Loss 0.321656 LR 0.001000 Time 0.021186 +2023-10-05 21:09:04,314 - Epoch: [57][ 800/ 1236] Overall Loss 0.321911 Objective Loss 0.321911 LR 0.001000 Time 0.021173 +2023-10-05 21:09:04,515 - Epoch: [57][ 810/ 1236] Overall Loss 0.321926 Objective Loss 0.321926 LR 0.001000 Time 0.021159 +2023-10-05 21:09:04,717 - Epoch: [57][ 820/ 1236] Overall Loss 0.322274 Objective Loss 0.322274 LR 0.001000 Time 0.021147 +2023-10-05 21:09:04,917 - Epoch: [57][ 830/ 1236] Overall Loss 0.322263 Objective Loss 0.322263 LR 0.001000 Time 0.021133 +2023-10-05 21:09:05,119 - Epoch: [57][ 840/ 1236] Overall Loss 0.322246 Objective Loss 0.322246 LR 0.001000 Time 0.021122 +2023-10-05 21:09:05,320 - Epoch: [57][ 850/ 1236] Overall Loss 0.322032 Objective Loss 0.322032 LR 0.001000 Time 0.021109 +2023-10-05 21:09:05,522 - Epoch: [57][ 860/ 1236] Overall Loss 0.322014 Objective Loss 0.322014 LR 0.001000 Time 0.021098 +2023-10-05 21:09:05,723 - Epoch: [57][ 870/ 1236] Overall Loss 0.322512 Objective Loss 0.322512 LR 0.001000 Time 0.021086 +2023-10-05 21:09:05,924 - Epoch: [57][ 880/ 1236] Overall Loss 0.322297 Objective Loss 0.322297 LR 0.001000 Time 0.021075 +2023-10-05 21:09:06,125 - Epoch: [57][ 890/ 1236] Overall Loss 0.322021 Objective Loss 0.322021 LR 0.001000 Time 0.021063 +2023-10-05 21:09:06,327 - Epoch: [57][ 900/ 1236] Overall Loss 0.322047 Objective Loss 0.322047 LR 0.001000 Time 0.021053 +2023-10-05 21:09:06,528 - Epoch: [57][ 910/ 1236] Overall Loss 0.321961 Objective Loss 0.321961 LR 0.001000 Time 0.021042 +2023-10-05 21:09:06,729 - Epoch: [57][ 920/ 1236] Overall Loss 0.322268 Objective Loss 0.322268 LR 0.001000 Time 0.021032 +2023-10-05 21:09:06,930 - Epoch: [57][ 930/ 1236] Overall Loss 0.322265 Objective Loss 0.322265 LR 0.001000 Time 0.021021 +2023-10-05 21:09:07,132 - Epoch: [57][ 940/ 1236] Overall Loss 0.322120 Objective Loss 0.322120 LR 0.001000 Time 0.021012 +2023-10-05 21:09:07,333 - Epoch: [57][ 950/ 1236] Overall Loss 0.322259 Objective Loss 0.322259 LR 0.001000 Time 0.021002 +2023-10-05 21:09:07,534 - Epoch: [57][ 960/ 1236] Overall Loss 0.322505 Objective Loss 0.322505 LR 0.001000 Time 0.020993 +2023-10-05 21:09:07,735 - Epoch: [57][ 970/ 1236] Overall Loss 0.322780 Objective Loss 0.322780 LR 0.001000 Time 0.020983 +2023-10-05 21:09:07,937 - Epoch: [57][ 980/ 1236] Overall Loss 0.322835 Objective Loss 0.322835 LR 0.001000 Time 0.020975 +2023-10-05 21:09:08,138 - Epoch: [57][ 990/ 1236] Overall Loss 0.322847 Objective Loss 0.322847 LR 0.001000 Time 0.020965 +2023-10-05 21:09:08,339 - Epoch: [57][ 1000/ 1236] Overall Loss 0.322830 Objective Loss 0.322830 LR 0.001000 Time 0.020957 +2023-10-05 21:09:08,540 - Epoch: [57][ 1010/ 1236] Overall Loss 0.322628 Objective Loss 0.322628 LR 0.001000 Time 0.020948 +2023-10-05 21:09:08,742 - Epoch: [57][ 1020/ 1236] Overall Loss 0.322480 Objective Loss 0.322480 LR 0.001000 Time 0.020940 +2023-10-05 21:09:08,943 - Epoch: [57][ 1030/ 1236] Overall Loss 0.322709 Objective Loss 0.322709 LR 0.001000 Time 0.020931 +2023-10-05 21:09:09,145 - Epoch: [57][ 1040/ 1236] Overall Loss 0.322528 Objective Loss 0.322528 LR 0.001000 Time 0.020924 +2023-10-05 21:09:09,345 - Epoch: [57][ 1050/ 1236] Overall Loss 0.322531 Objective Loss 0.322531 LR 0.001000 Time 0.020915 +2023-10-05 21:09:09,547 - Epoch: [57][ 1060/ 1236] Overall Loss 0.322845 Objective Loss 0.322845 LR 0.001000 Time 0.020908 +2023-10-05 21:09:09,748 - Epoch: [57][ 1070/ 1236] Overall Loss 0.322941 Objective Loss 0.322941 LR 0.001000 Time 0.020900 +2023-10-05 21:09:09,950 - Epoch: [57][ 1080/ 1236] Overall Loss 0.323086 Objective Loss 0.323086 LR 0.001000 Time 0.020893 +2023-10-05 21:09:10,151 - Epoch: [57][ 1090/ 1236] Overall Loss 0.323075 Objective Loss 0.323075 LR 0.001000 Time 0.020886 +2023-10-05 21:09:10,353 - Epoch: [57][ 1100/ 1236] Overall Loss 0.323050 Objective Loss 0.323050 LR 0.001000 Time 0.020879 +2023-10-05 21:09:10,554 - Epoch: [57][ 1110/ 1236] Overall Loss 0.323113 Objective Loss 0.323113 LR 0.001000 Time 0.020872 +2023-10-05 21:09:10,756 - Epoch: [57][ 1120/ 1236] Overall Loss 0.323238 Objective Loss 0.323238 LR 0.001000 Time 0.020865 +2023-10-05 21:09:10,957 - Epoch: [57][ 1130/ 1236] Overall Loss 0.322991 Objective Loss 0.322991 LR 0.001000 Time 0.020858 +2023-10-05 21:09:11,158 - Epoch: [57][ 1140/ 1236] Overall Loss 0.322935 Objective Loss 0.322935 LR 0.001000 Time 0.020852 +2023-10-05 21:09:11,359 - Epoch: [57][ 1150/ 1236] Overall Loss 0.323025 Objective Loss 0.323025 LR 0.001000 Time 0.020845 +2023-10-05 21:09:11,561 - Epoch: [57][ 1160/ 1236] Overall Loss 0.322958 Objective Loss 0.322958 LR 0.001000 Time 0.020839 +2023-10-05 21:09:11,762 - Epoch: [57][ 1170/ 1236] Overall Loss 0.323264 Objective Loss 0.323264 LR 0.001000 Time 0.020832 +2023-10-05 21:09:11,963 - Epoch: [57][ 1180/ 1236] Overall Loss 0.323338 Objective Loss 0.323338 LR 0.001000 Time 0.020826 +2023-10-05 21:09:12,164 - Epoch: [57][ 1190/ 1236] Overall Loss 0.323285 Objective Loss 0.323285 LR 0.001000 Time 0.020819 +2023-10-05 21:09:12,366 - Epoch: [57][ 1200/ 1236] Overall Loss 0.323349 Objective Loss 0.323349 LR 0.001000 Time 0.020814 +2023-10-05 21:09:12,567 - Epoch: [57][ 1210/ 1236] Overall Loss 0.323433 Objective Loss 0.323433 LR 0.001000 Time 0.020807 +2023-10-05 21:09:12,768 - Epoch: [57][ 1220/ 1236] Overall Loss 0.323245 Objective Loss 0.323245 LR 0.001000 Time 0.020802 +2023-10-05 21:09:13,022 - Epoch: [57][ 1230/ 1236] Overall Loss 0.323284 Objective Loss 0.323284 LR 0.001000 Time 0.020838 +2023-10-05 21:09:13,139 - Epoch: [57][ 1236/ 1236] Overall Loss 0.323378 Objective Loss 0.323378 Top1 85.947047 Top5 98.778004 LR 0.001000 Time 0.020832 +2023-10-05 21:09:13,271 - --- validate (epoch=57)----------- +2023-10-05 21:09:13,271 - 29943 samples (256 per mini-batch) +2023-10-05 21:09:13,724 - Epoch: [57][ 10/ 117] Loss 0.350468 Top1 82.734375 Top5 97.617188 +2023-10-05 21:09:13,871 - Epoch: [57][ 20/ 117] Loss 0.355184 Top1 82.128906 Top5 97.656250 +2023-10-05 21:09:14,018 - Epoch: [57][ 30/ 117] Loss 0.362493 Top1 82.265625 Top5 97.591146 +2023-10-05 21:09:14,165 - Epoch: [57][ 40/ 117] Loss 0.354147 Top1 82.373047 Top5 97.714844 +2023-10-05 21:09:14,312 - Epoch: [57][ 50/ 117] Loss 0.359593 Top1 82.164062 Top5 97.617188 +2023-10-05 21:09:14,461 - Epoch: [57][ 60/ 117] Loss 0.360697 Top1 82.018229 Top5 97.578125 +2023-10-05 21:09:14,610 - Epoch: [57][ 70/ 117] Loss 0.361971 Top1 82.003348 Top5 97.516741 +2023-10-05 21:09:14,761 - Epoch: [57][ 80/ 117] Loss 0.358092 Top1 82.148438 Top5 97.558594 +2023-10-05 21:09:14,910 - Epoch: [57][ 90/ 117] Loss 0.361136 Top1 82.135417 Top5 97.495660 +2023-10-05 21:09:15,058 - Epoch: [57][ 100/ 117] Loss 0.360936 Top1 82.105469 Top5 97.503906 +2023-10-05 21:09:15,212 - Epoch: [57][ 110/ 117] Loss 0.358974 Top1 82.070312 Top5 97.531960 +2023-10-05 21:09:15,297 - Epoch: [57][ 117/ 117] Loss 0.361728 Top1 82.032528 Top5 97.535317 +2023-10-05 21:09:15,426 - ==> Top1: 82.033 Top5: 97.535 Loss: 0.362 + +2023-10-05 21:09:15,427 - ==> Confusion: +[[ 939 1 8 1 10 0 0 0 4 56 1 1 1 0 5 2 7 2 0 1 11] + [ 2 1042 1 1 4 15 2 26 2 0 4 2 0 0 0 4 7 0 16 0 3] + [ 3 1 922 24 3 0 38 14 0 0 7 5 3 0 1 7 2 2 8 3 13] + [ 2 2 20 953 3 3 2 0 1 1 6 0 6 4 28 5 1 10 24 0 18] + [ 33 10 1 0 967 1 0 2 1 5 1 1 1 1 2 3 12 2 3 0 4] + [ 5 72 0 0 5 934 2 32 2 1 8 14 1 12 4 1 4 1 3 7 8] + [ 0 6 21 2 1 0 1129 5 0 0 3 2 0 0 1 7 1 1 2 5 5] + [ 4 20 18 1 2 28 4 1050 0 3 6 10 2 0 0 2 0 0 55 4 9] + [ 21 5 1 0 4 1 0 2 943 41 25 2 6 7 11 2 2 0 12 0 4] + [ 120 1 1 0 8 0 1 0 26 905 1 1 1 24 8 4 0 1 1 4 12] + [ 2 2 9 10 0 2 7 2 8 0 969 2 0 7 6 1 3 2 10 1 10] + [ 1 0 3 0 1 12 1 3 0 1 0 952 25 3 0 2 2 17 0 7 5] + [ 1 2 5 4 1 2 1 2 0 0 1 35 959 0 2 7 1 23 0 6 16] + [ 2 2 4 0 10 13 1 0 13 21 12 12 5 986 5 3 3 2 0 3 22] + [ 16 2 1 12 9 0 0 0 25 3 4 1 3 1 987 0 5 5 10 0 17] + [ 2 3 2 0 5 0 0 1 0 0 0 9 9 0 0 1047 18 22 0 6 10] + [ 1 20 1 2 6 1 0 0 0 0 0 5 2 1 4 7 1097 1 0 4 9] + [ 0 0 0 1 0 0 1 0 2 0 0 6 16 1 1 5 0 999 1 1 4] + [ 1 7 9 19 2 0 0 30 1 0 0 0 8 0 6 0 3 0 974 0 8] + [ 0 3 2 0 3 5 10 13 0 0 1 21 7 0 0 5 11 2 3 1051 15] + [ 197 277 163 72 145 126 57 121 94 79 200 133 358 228 113 70 215 103 222 174 4758]] + +2023-10-05 21:09:15,428 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:09:15,428 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:09:15,434 - + +2023-10-05 21:09:15,434 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:09:16,418 - Epoch: [58][ 10/ 1236] Overall Loss 0.313781 Objective Loss 0.313781 LR 0.001000 Time 0.098300 +2023-10-05 21:09:16,619 - Epoch: [58][ 20/ 1236] Overall Loss 0.308328 Objective Loss 0.308328 LR 0.001000 Time 0.059212 +2023-10-05 21:09:16,819 - Epoch: [58][ 30/ 1236] Overall Loss 0.311585 Objective Loss 0.311585 LR 0.001000 Time 0.046132 +2023-10-05 21:09:17,021 - Epoch: [58][ 40/ 1236] Overall Loss 0.316422 Objective Loss 0.316422 LR 0.001000 Time 0.039638 +2023-10-05 21:09:17,222 - Epoch: [58][ 50/ 1236] Overall Loss 0.323969 Objective Loss 0.323969 LR 0.001000 Time 0.035712 +2023-10-05 21:09:17,423 - Epoch: [58][ 60/ 1236] Overall Loss 0.325043 Objective Loss 0.325043 LR 0.001000 Time 0.033114 +2023-10-05 21:09:17,624 - Epoch: [58][ 70/ 1236] Overall Loss 0.321534 Objective Loss 0.321534 LR 0.001000 Time 0.031245 +2023-10-05 21:09:17,826 - Epoch: [58][ 80/ 1236] Overall Loss 0.322146 Objective Loss 0.322146 LR 0.001000 Time 0.029860 +2023-10-05 21:09:18,027 - Epoch: [58][ 90/ 1236] Overall Loss 0.321423 Objective Loss 0.321423 LR 0.001000 Time 0.028770 +2023-10-05 21:09:18,228 - Epoch: [58][ 100/ 1236] Overall Loss 0.320068 Objective Loss 0.320068 LR 0.001000 Time 0.027897 +2023-10-05 21:09:18,428 - Epoch: [58][ 110/ 1236] Overall Loss 0.321975 Objective Loss 0.321975 LR 0.001000 Time 0.027178 +2023-10-05 21:09:18,628 - Epoch: [58][ 120/ 1236] Overall Loss 0.320173 Objective Loss 0.320173 LR 0.001000 Time 0.026580 +2023-10-05 21:09:18,828 - Epoch: [58][ 130/ 1236] Overall Loss 0.321608 Objective Loss 0.321608 LR 0.001000 Time 0.026070 +2023-10-05 21:09:19,028 - Epoch: [58][ 140/ 1236] Overall Loss 0.320725 Objective Loss 0.320725 LR 0.001000 Time 0.025635 +2023-10-05 21:09:19,228 - Epoch: [58][ 150/ 1236] Overall Loss 0.321491 Objective Loss 0.321491 LR 0.001000 Time 0.025256 +2023-10-05 21:09:19,428 - Epoch: [58][ 160/ 1236] Overall Loss 0.319867 Objective Loss 0.319867 LR 0.001000 Time 0.024927 +2023-10-05 21:09:19,628 - Epoch: [58][ 170/ 1236] Overall Loss 0.319003 Objective Loss 0.319003 LR 0.001000 Time 0.024634 +2023-10-05 21:09:19,828 - Epoch: [58][ 180/ 1236] Overall Loss 0.317762 Objective Loss 0.317762 LR 0.001000 Time 0.024375 +2023-10-05 21:09:20,027 - Epoch: [58][ 190/ 1236] Overall Loss 0.316517 Objective Loss 0.316517 LR 0.001000 Time 0.024138 +2023-10-05 21:09:20,227 - Epoch: [58][ 200/ 1236] Overall Loss 0.316654 Objective Loss 0.316654 LR 0.001000 Time 0.023930 +2023-10-05 21:09:20,427 - Epoch: [58][ 210/ 1236] Overall Loss 0.316001 Objective Loss 0.316001 LR 0.001000 Time 0.023740 +2023-10-05 21:09:20,627 - Epoch: [58][ 220/ 1236] Overall Loss 0.316642 Objective Loss 0.316642 LR 0.001000 Time 0.023570 +2023-10-05 21:09:20,827 - Epoch: [58][ 230/ 1236] Overall Loss 0.318387 Objective Loss 0.318387 LR 0.001000 Time 0.023412 +2023-10-05 21:09:21,027 - Epoch: [58][ 240/ 1236] Overall Loss 0.317764 Objective Loss 0.317764 LR 0.001000 Time 0.023270 +2023-10-05 21:09:21,227 - Epoch: [58][ 250/ 1236] Overall Loss 0.318375 Objective Loss 0.318375 LR 0.001000 Time 0.023137 +2023-10-05 21:09:21,427 - Epoch: [58][ 260/ 1236] Overall Loss 0.319184 Objective Loss 0.319184 LR 0.001000 Time 0.023016 +2023-10-05 21:09:21,627 - Epoch: [58][ 270/ 1236] Overall Loss 0.319862 Objective Loss 0.319862 LR 0.001000 Time 0.022902 +2023-10-05 21:09:21,827 - Epoch: [58][ 280/ 1236] Overall Loss 0.320794 Objective Loss 0.320794 LR 0.001000 Time 0.022798 +2023-10-05 21:09:22,027 - Epoch: [58][ 290/ 1236] Overall Loss 0.320683 Objective Loss 0.320683 LR 0.001000 Time 0.022700 +2023-10-05 21:09:22,228 - Epoch: [58][ 300/ 1236] Overall Loss 0.321118 Objective Loss 0.321118 LR 0.001000 Time 0.022612 +2023-10-05 21:09:22,433 - Epoch: [58][ 310/ 1236] Overall Loss 0.320685 Objective Loss 0.320685 LR 0.001000 Time 0.022542 +2023-10-05 21:09:22,636 - Epoch: [58][ 320/ 1236] Overall Loss 0.320281 Objective Loss 0.320281 LR 0.001000 Time 0.022473 +2023-10-05 21:09:22,841 - Epoch: [58][ 330/ 1236] Overall Loss 0.320090 Objective Loss 0.320090 LR 0.001000 Time 0.022406 +2023-10-05 21:09:23,045 - Epoch: [58][ 340/ 1236] Overall Loss 0.319843 Objective Loss 0.319843 LR 0.001000 Time 0.022347 +2023-10-05 21:09:23,249 - Epoch: [58][ 350/ 1236] Overall Loss 0.319107 Objective Loss 0.319107 LR 0.001000 Time 0.022288 +2023-10-05 21:09:23,454 - Epoch: [58][ 360/ 1236] Overall Loss 0.320386 Objective Loss 0.320386 LR 0.001000 Time 0.022237 +2023-10-05 21:09:23,658 - Epoch: [58][ 370/ 1236] Overall Loss 0.320898 Objective Loss 0.320898 LR 0.001000 Time 0.022181 +2023-10-05 21:09:23,859 - Epoch: [58][ 380/ 1236] Overall Loss 0.320920 Objective Loss 0.320920 LR 0.001000 Time 0.022126 +2023-10-05 21:09:24,060 - Epoch: [58][ 390/ 1236] Overall Loss 0.321298 Objective Loss 0.321298 LR 0.001000 Time 0.022073 +2023-10-05 21:09:24,261 - Epoch: [58][ 400/ 1236] Overall Loss 0.321133 Objective Loss 0.321133 LR 0.001000 Time 0.022024 +2023-10-05 21:09:24,462 - Epoch: [58][ 410/ 1236] Overall Loss 0.321579 Objective Loss 0.321579 LR 0.001000 Time 0.021977 +2023-10-05 21:09:24,663 - Epoch: [58][ 420/ 1236] Overall Loss 0.321940 Objective Loss 0.321940 LR 0.001000 Time 0.021932 +2023-10-05 21:09:24,865 - Epoch: [58][ 430/ 1236] Overall Loss 0.322979 Objective Loss 0.322979 LR 0.001000 Time 0.021889 +2023-10-05 21:09:25,066 - Epoch: [58][ 440/ 1236] Overall Loss 0.322972 Objective Loss 0.322972 LR 0.001000 Time 0.021848 +2023-10-05 21:09:25,267 - Epoch: [58][ 450/ 1236] Overall Loss 0.323521 Objective Loss 0.323521 LR 0.001000 Time 0.021810 +2023-10-05 21:09:25,469 - Epoch: [58][ 460/ 1236] Overall Loss 0.322908 Objective Loss 0.322908 LR 0.001000 Time 0.021772 +2023-10-05 21:09:25,670 - Epoch: [58][ 470/ 1236] Overall Loss 0.323929 Objective Loss 0.323929 LR 0.001000 Time 0.021736 +2023-10-05 21:09:25,871 - Epoch: [58][ 480/ 1236] Overall Loss 0.323427 Objective Loss 0.323427 LR 0.001000 Time 0.021702 +2023-10-05 21:09:26,072 - Epoch: [58][ 490/ 1236] Overall Loss 0.322890 Objective Loss 0.322890 LR 0.001000 Time 0.021670 +2023-10-05 21:09:26,273 - Epoch: [58][ 500/ 1236] Overall Loss 0.322815 Objective Loss 0.322815 LR 0.001000 Time 0.021637 +2023-10-05 21:09:26,475 - Epoch: [58][ 510/ 1236] Overall Loss 0.323789 Objective Loss 0.323789 LR 0.001000 Time 0.021607 +2023-10-05 21:09:26,676 - Epoch: [58][ 520/ 1236] Overall Loss 0.323748 Objective Loss 0.323748 LR 0.001000 Time 0.021578 +2023-10-05 21:09:26,877 - Epoch: [58][ 530/ 1236] Overall Loss 0.324163 Objective Loss 0.324163 LR 0.001000 Time 0.021551 +2023-10-05 21:09:27,079 - Epoch: [58][ 540/ 1236] Overall Loss 0.324456 Objective Loss 0.324456 LR 0.001000 Time 0.021524 +2023-10-05 21:09:27,280 - Epoch: [58][ 550/ 1236] Overall Loss 0.324803 Objective Loss 0.324803 LR 0.001000 Time 0.021497 +2023-10-05 21:09:27,481 - Epoch: [58][ 560/ 1236] Overall Loss 0.324610 Objective Loss 0.324610 LR 0.001000 Time 0.021473 +2023-10-05 21:09:27,683 - Epoch: [58][ 570/ 1236] Overall Loss 0.324949 Objective Loss 0.324949 LR 0.001000 Time 0.021449 +2023-10-05 21:09:27,884 - Epoch: [58][ 580/ 1236] Overall Loss 0.324866 Objective Loss 0.324866 LR 0.001000 Time 0.021425 +2023-10-05 21:09:28,085 - Epoch: [58][ 590/ 1236] Overall Loss 0.324174 Objective Loss 0.324174 LR 0.001000 Time 0.021403 +2023-10-05 21:09:28,287 - Epoch: [58][ 600/ 1236] Overall Loss 0.323900 Objective Loss 0.323900 LR 0.001000 Time 0.021382 +2023-10-05 21:09:28,488 - Epoch: [58][ 610/ 1236] Overall Loss 0.323482 Objective Loss 0.323482 LR 0.001000 Time 0.021361 +2023-10-05 21:09:28,689 - Epoch: [58][ 620/ 1236] Overall Loss 0.322888 Objective Loss 0.322888 LR 0.001000 Time 0.021340 +2023-10-05 21:09:28,891 - Epoch: [58][ 630/ 1236] Overall Loss 0.323022 Objective Loss 0.323022 LR 0.001000 Time 0.021321 +2023-10-05 21:09:29,092 - Epoch: [58][ 640/ 1236] Overall Loss 0.322747 Objective Loss 0.322747 LR 0.001000 Time 0.021301 +2023-10-05 21:09:29,293 - Epoch: [58][ 650/ 1236] Overall Loss 0.323082 Objective Loss 0.323082 LR 0.001000 Time 0.021283 +2023-10-05 21:09:29,495 - Epoch: [58][ 660/ 1236] Overall Loss 0.322847 Objective Loss 0.322847 LR 0.001000 Time 0.021265 +2023-10-05 21:09:29,696 - Epoch: [58][ 670/ 1236] Overall Loss 0.323208 Objective Loss 0.323208 LR 0.001000 Time 0.021248 +2023-10-05 21:09:29,897 - Epoch: [58][ 680/ 1236] Overall Loss 0.323827 Objective Loss 0.323827 LR 0.001000 Time 0.021230 +2023-10-05 21:09:30,099 - Epoch: [58][ 690/ 1236] Overall Loss 0.323721 Objective Loss 0.323721 LR 0.001000 Time 0.021215 +2023-10-05 21:09:30,300 - Epoch: [58][ 700/ 1236] Overall Loss 0.323822 Objective Loss 0.323822 LR 0.001000 Time 0.021199 +2023-10-05 21:09:30,502 - Epoch: [58][ 710/ 1236] Overall Loss 0.324134 Objective Loss 0.324134 LR 0.001000 Time 0.021184 +2023-10-05 21:09:30,703 - Epoch: [58][ 720/ 1236] Overall Loss 0.324101 Objective Loss 0.324101 LR 0.001000 Time 0.021168 +2023-10-05 21:09:30,905 - Epoch: [58][ 730/ 1236] Overall Loss 0.324064 Objective Loss 0.324064 LR 0.001000 Time 0.021154 +2023-10-05 21:09:31,105 - Epoch: [58][ 740/ 1236] Overall Loss 0.323653 Objective Loss 0.323653 LR 0.001000 Time 0.021139 +2023-10-05 21:09:31,307 - Epoch: [58][ 750/ 1236] Overall Loss 0.323096 Objective Loss 0.323096 LR 0.001000 Time 0.021125 +2023-10-05 21:09:31,508 - Epoch: [58][ 760/ 1236] Overall Loss 0.322977 Objective Loss 0.322977 LR 0.001000 Time 0.021112 +2023-10-05 21:09:31,709 - Epoch: [58][ 770/ 1236] Overall Loss 0.323044 Objective Loss 0.323044 LR 0.001000 Time 0.021099 +2023-10-05 21:09:31,911 - Epoch: [58][ 780/ 1236] Overall Loss 0.323040 Objective Loss 0.323040 LR 0.001000 Time 0.021086 +2023-10-05 21:09:32,112 - Epoch: [58][ 790/ 1236] Overall Loss 0.322872 Objective Loss 0.322872 LR 0.001000 Time 0.021074 +2023-10-05 21:09:32,313 - Epoch: [58][ 800/ 1236] Overall Loss 0.322641 Objective Loss 0.322641 LR 0.001000 Time 0.021061 +2023-10-05 21:09:32,515 - Epoch: [58][ 810/ 1236] Overall Loss 0.322679 Objective Loss 0.322679 LR 0.001000 Time 0.021050 +2023-10-05 21:09:32,717 - Epoch: [58][ 820/ 1236] Overall Loss 0.322761 Objective Loss 0.322761 LR 0.001000 Time 0.021039 +2023-10-05 21:09:32,918 - Epoch: [58][ 830/ 1236] Overall Loss 0.322681 Objective Loss 0.322681 LR 0.001000 Time 0.021028 +2023-10-05 21:09:33,120 - Epoch: [58][ 840/ 1236] Overall Loss 0.322690 Objective Loss 0.322690 LR 0.001000 Time 0.021017 +2023-10-05 21:09:33,324 - Epoch: [58][ 850/ 1236] Overall Loss 0.322979 Objective Loss 0.322979 LR 0.001000 Time 0.021010 +2023-10-05 21:09:33,535 - Epoch: [58][ 860/ 1236] Overall Loss 0.323198 Objective Loss 0.323198 LR 0.001000 Time 0.021010 +2023-10-05 21:09:33,744 - Epoch: [58][ 870/ 1236] Overall Loss 0.323054 Objective Loss 0.323054 LR 0.001000 Time 0.021008 +2023-10-05 21:09:33,954 - Epoch: [58][ 880/ 1236] Overall Loss 0.322833 Objective Loss 0.322833 LR 0.001000 Time 0.021009 +2023-10-05 21:09:34,163 - Epoch: [58][ 890/ 1236] Overall Loss 0.322506 Objective Loss 0.322506 LR 0.001000 Time 0.021006 +2023-10-05 21:09:34,373 - Epoch: [58][ 900/ 1236] Overall Loss 0.322503 Objective Loss 0.322503 LR 0.001000 Time 0.021006 +2023-10-05 21:09:34,582 - Epoch: [58][ 910/ 1236] Overall Loss 0.322434 Objective Loss 0.322434 LR 0.001000 Time 0.021004 +2023-10-05 21:09:34,793 - Epoch: [58][ 920/ 1236] Overall Loss 0.322225 Objective Loss 0.322225 LR 0.001000 Time 0.021005 +2023-10-05 21:09:35,001 - Epoch: [58][ 930/ 1236] Overall Loss 0.322195 Objective Loss 0.322195 LR 0.001000 Time 0.021002 +2023-10-05 21:09:35,212 - Epoch: [58][ 940/ 1236] Overall Loss 0.322037 Objective Loss 0.322037 LR 0.001000 Time 0.021003 +2023-10-05 21:09:35,421 - Epoch: [58][ 950/ 1236] Overall Loss 0.321792 Objective Loss 0.321792 LR 0.001000 Time 0.021002 +2023-10-05 21:09:35,631 - Epoch: [58][ 960/ 1236] Overall Loss 0.321492 Objective Loss 0.321492 LR 0.001000 Time 0.021002 +2023-10-05 21:09:35,840 - Epoch: [58][ 970/ 1236] Overall Loss 0.321513 Objective Loss 0.321513 LR 0.001000 Time 0.021000 +2023-10-05 21:09:36,051 - Epoch: [58][ 980/ 1236] Overall Loss 0.321680 Objective Loss 0.321680 LR 0.001000 Time 0.021001 +2023-10-05 21:09:36,259 - Epoch: [58][ 990/ 1236] Overall Loss 0.321558 Objective Loss 0.321558 LR 0.001000 Time 0.020999 +2023-10-05 21:09:36,470 - Epoch: [58][ 1000/ 1236] Overall Loss 0.321605 Objective Loss 0.321605 LR 0.001000 Time 0.020999 +2023-10-05 21:09:36,679 - Epoch: [58][ 1010/ 1236] Overall Loss 0.321488 Objective Loss 0.321488 LR 0.001000 Time 0.020997 +2023-10-05 21:09:36,890 - Epoch: [58][ 1020/ 1236] Overall Loss 0.321403 Objective Loss 0.321403 LR 0.001000 Time 0.020998 +2023-10-05 21:09:37,098 - Epoch: [58][ 1030/ 1236] Overall Loss 0.321613 Objective Loss 0.321613 LR 0.001000 Time 0.020996 +2023-10-05 21:09:37,309 - Epoch: [58][ 1040/ 1236] Overall Loss 0.321512 Objective Loss 0.321512 LR 0.001000 Time 0.020997 +2023-10-05 21:09:37,517 - Epoch: [58][ 1050/ 1236] Overall Loss 0.321620 Objective Loss 0.321620 LR 0.001000 Time 0.020995 +2023-10-05 21:09:37,728 - Epoch: [58][ 1060/ 1236] Overall Loss 0.321708 Objective Loss 0.321708 LR 0.001000 Time 0.020996 +2023-10-05 21:09:37,937 - Epoch: [58][ 1070/ 1236] Overall Loss 0.321729 Objective Loss 0.321729 LR 0.001000 Time 0.020994 +2023-10-05 21:09:38,148 - Epoch: [58][ 1080/ 1236] Overall Loss 0.321433 Objective Loss 0.321433 LR 0.001000 Time 0.020995 +2023-10-05 21:09:38,356 - Epoch: [58][ 1090/ 1236] Overall Loss 0.321290 Objective Loss 0.321290 LR 0.001000 Time 0.020993 +2023-10-05 21:09:38,567 - Epoch: [58][ 1100/ 1236] Overall Loss 0.321600 Objective Loss 0.321600 LR 0.001000 Time 0.020994 +2023-10-05 21:09:38,776 - Epoch: [58][ 1110/ 1236] Overall Loss 0.321930 Objective Loss 0.321930 LR 0.001000 Time 0.020992 +2023-10-05 21:09:38,986 - Epoch: [58][ 1120/ 1236] Overall Loss 0.321944 Objective Loss 0.321944 LR 0.001000 Time 0.020993 +2023-10-05 21:09:39,195 - Epoch: [58][ 1130/ 1236] Overall Loss 0.322032 Objective Loss 0.322032 LR 0.001000 Time 0.020991 +2023-10-05 21:09:39,406 - Epoch: [58][ 1140/ 1236] Overall Loss 0.322124 Objective Loss 0.322124 LR 0.001000 Time 0.020992 +2023-10-05 21:09:39,614 - Epoch: [58][ 1150/ 1236] Overall Loss 0.322169 Objective Loss 0.322169 LR 0.001000 Time 0.020990 +2023-10-05 21:09:39,825 - Epoch: [58][ 1160/ 1236] Overall Loss 0.321887 Objective Loss 0.321887 LR 0.001000 Time 0.020990 +2023-10-05 21:09:40,033 - Epoch: [58][ 1170/ 1236] Overall Loss 0.322183 Objective Loss 0.322183 LR 0.001000 Time 0.020989 +2023-10-05 21:09:40,244 - Epoch: [58][ 1180/ 1236] Overall Loss 0.322338 Objective Loss 0.322338 LR 0.001000 Time 0.020989 +2023-10-05 21:09:40,453 - Epoch: [58][ 1190/ 1236] Overall Loss 0.322162 Objective Loss 0.322162 LR 0.001000 Time 0.020988 +2023-10-05 21:09:40,663 - Epoch: [58][ 1200/ 1236] Overall Loss 0.322101 Objective Loss 0.322101 LR 0.001000 Time 0.020989 +2023-10-05 21:09:40,872 - Epoch: [58][ 1210/ 1236] Overall Loss 0.322026 Objective Loss 0.322026 LR 0.001000 Time 0.020987 +2023-10-05 21:09:41,083 - Epoch: [58][ 1220/ 1236] Overall Loss 0.322586 Objective Loss 0.322586 LR 0.001000 Time 0.020988 +2023-10-05 21:09:41,344 - Epoch: [58][ 1230/ 1236] Overall Loss 0.322743 Objective Loss 0.322743 LR 0.001000 Time 0.021029 +2023-10-05 21:09:41,461 - Epoch: [58][ 1236/ 1236] Overall Loss 0.322778 Objective Loss 0.322778 Top1 83.299389 Top5 97.556008 LR 0.001000 Time 0.021022 +2023-10-05 21:09:41,584 - --- validate (epoch=58)----------- +2023-10-05 21:09:41,584 - 29943 samples (256 per mini-batch) +2023-10-05 21:09:42,029 - Epoch: [58][ 10/ 117] Loss 0.371382 Top1 81.250000 Top5 97.226562 +2023-10-05 21:09:42,177 - Epoch: [58][ 20/ 117] Loss 0.345236 Top1 81.757812 Top5 97.167969 +2023-10-05 21:09:42,325 - Epoch: [58][ 30/ 117] Loss 0.351681 Top1 81.705729 Top5 97.291667 +2023-10-05 21:09:42,473 - Epoch: [58][ 40/ 117] Loss 0.351850 Top1 81.806641 Top5 97.285156 +2023-10-05 21:09:42,621 - Epoch: [58][ 50/ 117] Loss 0.355806 Top1 81.507812 Top5 97.218750 +2023-10-05 21:09:42,768 - Epoch: [58][ 60/ 117] Loss 0.361410 Top1 81.367188 Top5 97.213542 +2023-10-05 21:09:42,914 - Epoch: [58][ 70/ 117] Loss 0.360555 Top1 81.311384 Top5 97.226562 +2023-10-05 21:09:43,062 - Epoch: [58][ 80/ 117] Loss 0.359676 Top1 81.445312 Top5 97.255859 +2023-10-05 21:09:43,209 - Epoch: [58][ 90/ 117] Loss 0.363268 Top1 81.345486 Top5 97.204861 +2023-10-05 21:09:43,356 - Epoch: [58][ 100/ 117] Loss 0.366507 Top1 81.257812 Top5 97.167969 +2023-10-05 21:09:43,509 - Epoch: [58][ 110/ 117] Loss 0.368596 Top1 81.193182 Top5 97.144886 +2023-10-05 21:09:43,593 - Epoch: [58][ 117/ 117] Loss 0.365403 Top1 81.231006 Top5 97.161273 +2023-10-05 21:09:43,738 - ==> Top1: 81.231 Top5: 97.161 Loss: 0.365 + +2023-10-05 21:09:43,739 - ==> Confusion: +[[ 941 2 1 1 8 5 0 0 5 60 1 1 1 2 3 3 7 1 2 0 6] + [ 0 1039 3 0 12 22 0 16 4 0 2 4 0 0 0 3 6 1 11 3 5] + [ 5 2 947 12 2 1 21 7 0 1 6 4 7 2 0 4 3 1 11 9 11] + [ 6 2 30 936 2 4 0 0 0 0 12 0 1 6 22 5 2 9 31 5 16] + [ 28 4 0 0 968 4 0 0 0 9 0 1 0 2 6 4 20 2 1 1 0] + [ 5 45 1 1 1 981 2 20 2 4 1 15 1 10 5 1 7 0 2 4 8] + [ 1 9 39 0 0 0 1090 8 0 0 3 3 3 0 1 8 0 0 3 13 10] + [ 7 31 24 0 3 38 3 1006 1 2 2 18 4 2 1 3 0 0 56 11 6] + [ 29 4 0 0 1 4 0 0 946 41 17 4 2 15 6 2 6 0 9 0 3] + [ 126 0 0 0 8 4 1 0 28 897 1 1 0 26 3 5 4 1 1 3 10] + [ 5 5 12 6 0 2 3 4 12 2 952 3 1 12 5 1 0 1 15 1 11] + [ 1 0 0 0 1 10 0 0 2 0 0 962 29 4 0 4 3 11 0 8 0] + [ 1 0 2 5 0 3 0 0 1 0 1 51 967 3 0 3 4 7 4 7 9] + [ 3 1 0 0 5 20 0 3 20 15 6 10 2 1016 2 3 2 1 0 4 6] + [ 16 3 2 15 11 0 0 0 25 7 5 1 4 1 980 0 5 2 17 0 7] + [ 1 3 2 2 2 0 0 0 0 0 0 11 14 2 0 1054 17 11 1 9 5] + [ 1 10 1 0 5 5 0 1 2 0 0 7 1 0 1 10 1106 0 1 5 5] + [ 1 0 1 3 0 0 1 0 0 1 0 5 29 1 1 5 0 984 2 0 4] + [ 2 14 10 14 2 0 0 25 3 0 4 1 2 1 8 1 3 0 971 0 7] + [ 0 3 1 0 2 4 2 10 0 0 2 18 11 1 0 2 15 0 2 1068 11] + [ 237 244 214 58 115 190 25 100 123 95 183 152 437 287 114 69 267 63 179 241 4512]] + +2023-10-05 21:09:43,740 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:09:43,740 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:09:43,746 - + +2023-10-05 21:09:43,746 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:09:44,723 - Epoch: [59][ 10/ 1236] Overall Loss 0.314996 Objective Loss 0.314996 LR 0.001000 Time 0.097664 +2023-10-05 21:09:44,926 - Epoch: [59][ 20/ 1236] Overall Loss 0.319485 Objective Loss 0.319485 LR 0.001000 Time 0.058967 +2023-10-05 21:09:45,128 - Epoch: [59][ 30/ 1236] Overall Loss 0.317926 Objective Loss 0.317926 LR 0.001000 Time 0.046020 +2023-10-05 21:09:45,331 - Epoch: [59][ 40/ 1236] Overall Loss 0.322123 Objective Loss 0.322123 LR 0.001000 Time 0.039585 +2023-10-05 21:09:45,533 - Epoch: [59][ 50/ 1236] Overall Loss 0.319587 Objective Loss 0.319587 LR 0.001000 Time 0.035702 +2023-10-05 21:09:45,736 - Epoch: [59][ 60/ 1236] Overall Loss 0.323256 Objective Loss 0.323256 LR 0.001000 Time 0.033131 +2023-10-05 21:09:45,938 - Epoch: [59][ 70/ 1236] Overall Loss 0.322499 Objective Loss 0.322499 LR 0.001000 Time 0.031277 +2023-10-05 21:09:46,141 - Epoch: [59][ 80/ 1236] Overall Loss 0.316407 Objective Loss 0.316407 LR 0.001000 Time 0.029904 +2023-10-05 21:09:46,343 - Epoch: [59][ 90/ 1236] Overall Loss 0.316593 Objective Loss 0.316593 LR 0.001000 Time 0.028818 +2023-10-05 21:09:46,546 - Epoch: [59][ 100/ 1236] Overall Loss 0.314447 Objective Loss 0.314447 LR 0.001000 Time 0.027964 +2023-10-05 21:09:46,746 - Epoch: [59][ 110/ 1236] Overall Loss 0.314978 Objective Loss 0.314978 LR 0.001000 Time 0.027236 +2023-10-05 21:09:46,948 - Epoch: [59][ 120/ 1236] Overall Loss 0.312940 Objective Loss 0.312940 LR 0.001000 Time 0.026645 +2023-10-05 21:09:47,148 - Epoch: [59][ 130/ 1236] Overall Loss 0.313662 Objective Loss 0.313662 LR 0.001000 Time 0.026135 +2023-10-05 21:09:47,349 - Epoch: [59][ 140/ 1236] Overall Loss 0.313529 Objective Loss 0.313529 LR 0.001000 Time 0.025701 +2023-10-05 21:09:47,549 - Epoch: [59][ 150/ 1236] Overall Loss 0.314275 Objective Loss 0.314275 LR 0.001000 Time 0.025318 +2023-10-05 21:09:47,750 - Epoch: [59][ 160/ 1236] Overall Loss 0.313627 Objective Loss 0.313627 LR 0.001000 Time 0.024987 +2023-10-05 21:09:47,950 - Epoch: [59][ 170/ 1236] Overall Loss 0.313076 Objective Loss 0.313076 LR 0.001000 Time 0.024692 +2023-10-05 21:09:48,150 - Epoch: [59][ 180/ 1236] Overall Loss 0.313109 Objective Loss 0.313109 LR 0.001000 Time 0.024430 +2023-10-05 21:09:48,349 - Epoch: [59][ 190/ 1236] Overall Loss 0.311382 Objective Loss 0.311382 LR 0.001000 Time 0.024193 +2023-10-05 21:09:48,550 - Epoch: [59][ 200/ 1236] Overall Loss 0.310922 Objective Loss 0.310922 LR 0.001000 Time 0.023984 +2023-10-05 21:09:48,751 - Epoch: [59][ 210/ 1236] Overall Loss 0.310622 Objective Loss 0.310622 LR 0.001000 Time 0.023797 +2023-10-05 21:09:48,951 - Epoch: [59][ 220/ 1236] Overall Loss 0.309807 Objective Loss 0.309807 LR 0.001000 Time 0.023623 +2023-10-05 21:09:49,150 - Epoch: [59][ 230/ 1236] Overall Loss 0.310111 Objective Loss 0.310111 LR 0.001000 Time 0.023462 +2023-10-05 21:09:49,351 - Epoch: [59][ 240/ 1236] Overall Loss 0.310074 Objective Loss 0.310074 LR 0.001000 Time 0.023318 +2023-10-05 21:09:49,550 - Epoch: [59][ 250/ 1236] Overall Loss 0.310185 Objective Loss 0.310185 LR 0.001000 Time 0.023183 +2023-10-05 21:09:49,749 - Epoch: [59][ 260/ 1236] Overall Loss 0.311449 Objective Loss 0.311449 LR 0.001000 Time 0.023054 +2023-10-05 21:09:49,948 - Epoch: [59][ 270/ 1236] Overall Loss 0.311147 Objective Loss 0.311147 LR 0.001000 Time 0.022937 +2023-10-05 21:09:50,149 - Epoch: [59][ 280/ 1236] Overall Loss 0.311961 Objective Loss 0.311961 LR 0.001000 Time 0.022832 +2023-10-05 21:09:50,348 - Epoch: [59][ 290/ 1236] Overall Loss 0.312405 Objective Loss 0.312405 LR 0.001000 Time 0.022733 +2023-10-05 21:09:50,549 - Epoch: [59][ 300/ 1236] Overall Loss 0.312106 Objective Loss 0.312106 LR 0.001000 Time 0.022641 +2023-10-05 21:09:50,749 - Epoch: [59][ 310/ 1236] Overall Loss 0.312404 Objective Loss 0.312404 LR 0.001000 Time 0.022555 +2023-10-05 21:09:50,949 - Epoch: [59][ 320/ 1236] Overall Loss 0.312696 Objective Loss 0.312696 LR 0.001000 Time 0.022475 +2023-10-05 21:09:51,150 - Epoch: [59][ 330/ 1236] Overall Loss 0.313828 Objective Loss 0.313828 LR 0.001000 Time 0.022402 +2023-10-05 21:09:51,351 - Epoch: [59][ 340/ 1236] Overall Loss 0.315458 Objective Loss 0.315458 LR 0.001000 Time 0.022333 +2023-10-05 21:09:51,550 - Epoch: [59][ 350/ 1236] Overall Loss 0.316727 Objective Loss 0.316727 LR 0.001000 Time 0.022265 +2023-10-05 21:09:51,751 - Epoch: [59][ 360/ 1236] Overall Loss 0.316629 Objective Loss 0.316629 LR 0.001000 Time 0.022202 +2023-10-05 21:09:51,951 - Epoch: [59][ 370/ 1236] Overall Loss 0.316339 Objective Loss 0.316339 LR 0.001000 Time 0.022143 +2023-10-05 21:09:52,153 - Epoch: [59][ 380/ 1236] Overall Loss 0.315923 Objective Loss 0.315923 LR 0.001000 Time 0.022089 +2023-10-05 21:09:52,353 - Epoch: [59][ 390/ 1236] Overall Loss 0.316110 Objective Loss 0.316110 LR 0.001000 Time 0.022036 +2023-10-05 21:09:52,554 - Epoch: [59][ 400/ 1236] Overall Loss 0.316282 Objective Loss 0.316282 LR 0.001000 Time 0.021987 +2023-10-05 21:09:52,755 - Epoch: [59][ 410/ 1236] Overall Loss 0.317179 Objective Loss 0.317179 LR 0.001000 Time 0.021939 +2023-10-05 21:09:52,956 - Epoch: [59][ 420/ 1236] Overall Loss 0.317646 Objective Loss 0.317646 LR 0.001000 Time 0.021895 +2023-10-05 21:09:53,156 - Epoch: [59][ 430/ 1236] Overall Loss 0.317324 Objective Loss 0.317324 LR 0.001000 Time 0.021851 +2023-10-05 21:09:53,358 - Epoch: [59][ 440/ 1236] Overall Loss 0.317287 Objective Loss 0.317287 LR 0.001000 Time 0.021811 +2023-10-05 21:09:53,558 - Epoch: [59][ 450/ 1236] Overall Loss 0.317600 Objective Loss 0.317600 LR 0.001000 Time 0.021772 +2023-10-05 21:09:53,760 - Epoch: [59][ 460/ 1236] Overall Loss 0.317509 Objective Loss 0.317509 LR 0.001000 Time 0.021736 +2023-10-05 21:09:53,960 - Epoch: [59][ 470/ 1236] Overall Loss 0.318022 Objective Loss 0.318022 LR 0.001000 Time 0.021699 +2023-10-05 21:09:54,161 - Epoch: [59][ 480/ 1236] Overall Loss 0.318536 Objective Loss 0.318536 LR 0.001000 Time 0.021665 +2023-10-05 21:09:54,361 - Epoch: [59][ 490/ 1236] Overall Loss 0.318678 Objective Loss 0.318678 LR 0.001000 Time 0.021629 +2023-10-05 21:09:54,562 - Epoch: [59][ 500/ 1236] Overall Loss 0.318318 Objective Loss 0.318318 LR 0.001000 Time 0.021598 +2023-10-05 21:09:54,762 - Epoch: [59][ 510/ 1236] Overall Loss 0.318560 Objective Loss 0.318560 LR 0.001000 Time 0.021567 +2023-10-05 21:09:54,964 - Epoch: [59][ 520/ 1236] Overall Loss 0.319109 Objective Loss 0.319109 LR 0.001000 Time 0.021539 +2023-10-05 21:09:55,164 - Epoch: [59][ 530/ 1236] Overall Loss 0.319076 Objective Loss 0.319076 LR 0.001000 Time 0.021510 +2023-10-05 21:09:55,366 - Epoch: [59][ 540/ 1236] Overall Loss 0.319019 Objective Loss 0.319019 LR 0.001000 Time 0.021485 +2023-10-05 21:09:55,566 - Epoch: [59][ 550/ 1236] Overall Loss 0.318699 Objective Loss 0.318699 LR 0.001000 Time 0.021458 +2023-10-05 21:09:55,768 - Epoch: [59][ 560/ 1236] Overall Loss 0.319065 Objective Loss 0.319065 LR 0.001000 Time 0.021435 +2023-10-05 21:09:55,969 - Epoch: [59][ 570/ 1236] Overall Loss 0.319071 Objective Loss 0.319071 LR 0.001000 Time 0.021410 +2023-10-05 21:09:56,171 - Epoch: [59][ 580/ 1236] Overall Loss 0.319094 Objective Loss 0.319094 LR 0.001000 Time 0.021389 +2023-10-05 21:09:56,371 - Epoch: [59][ 590/ 1236] Overall Loss 0.319265 Objective Loss 0.319265 LR 0.001000 Time 0.021365 +2023-10-05 21:09:56,574 - Epoch: [59][ 600/ 1236] Overall Loss 0.319366 Objective Loss 0.319366 LR 0.001000 Time 0.021346 +2023-10-05 21:09:56,774 - Epoch: [59][ 610/ 1236] Overall Loss 0.319540 Objective Loss 0.319540 LR 0.001000 Time 0.021324 +2023-10-05 21:09:56,976 - Epoch: [59][ 620/ 1236] Overall Loss 0.319360 Objective Loss 0.319360 LR 0.001000 Time 0.021306 +2023-10-05 21:09:57,176 - Epoch: [59][ 630/ 1236] Overall Loss 0.319132 Objective Loss 0.319132 LR 0.001000 Time 0.021284 +2023-10-05 21:09:57,379 - Epoch: [59][ 640/ 1236] Overall Loss 0.319523 Objective Loss 0.319523 LR 0.001000 Time 0.021268 +2023-10-05 21:09:57,579 - Epoch: [59][ 650/ 1236] Overall Loss 0.319624 Objective Loss 0.319624 LR 0.001000 Time 0.021248 +2023-10-05 21:09:57,781 - Epoch: [59][ 660/ 1236] Overall Loss 0.319808 Objective Loss 0.319808 LR 0.001000 Time 0.021232 +2023-10-05 21:09:57,981 - Epoch: [59][ 670/ 1236] Overall Loss 0.319872 Objective Loss 0.319872 LR 0.001000 Time 0.021213 +2023-10-05 21:09:58,183 - Epoch: [59][ 680/ 1236] Overall Loss 0.319953 Objective Loss 0.319953 LR 0.001000 Time 0.021197 +2023-10-05 21:09:58,383 - Epoch: [59][ 690/ 1236] Overall Loss 0.320056 Objective Loss 0.320056 LR 0.001000 Time 0.021179 +2023-10-05 21:09:58,585 - Epoch: [59][ 700/ 1236] Overall Loss 0.320132 Objective Loss 0.320132 LR 0.001000 Time 0.021165 +2023-10-05 21:09:58,785 - Epoch: [59][ 710/ 1236] Overall Loss 0.320240 Objective Loss 0.320240 LR 0.001000 Time 0.021148 +2023-10-05 21:09:58,987 - Epoch: [59][ 720/ 1236] Overall Loss 0.320879 Objective Loss 0.320879 LR 0.001000 Time 0.021135 +2023-10-05 21:09:59,185 - Epoch: [59][ 730/ 1236] Overall Loss 0.320979 Objective Loss 0.320979 LR 0.001000 Time 0.021116 +2023-10-05 21:09:59,387 - Epoch: [59][ 740/ 1236] Overall Loss 0.320810 Objective Loss 0.320810 LR 0.001000 Time 0.021103 +2023-10-05 21:09:59,587 - Epoch: [59][ 750/ 1236] Overall Loss 0.321203 Objective Loss 0.321203 LR 0.001000 Time 0.021088 +2023-10-05 21:09:59,789 - Epoch: [59][ 760/ 1236] Overall Loss 0.321065 Objective Loss 0.321065 LR 0.001000 Time 0.021076 +2023-10-05 21:09:59,989 - Epoch: [59][ 770/ 1236] Overall Loss 0.321174 Objective Loss 0.321174 LR 0.001000 Time 0.021061 +2023-10-05 21:10:00,191 - Epoch: [59][ 780/ 1236] Overall Loss 0.321654 Objective Loss 0.321654 LR 0.001000 Time 0.021050 +2023-10-05 21:10:00,391 - Epoch: [59][ 790/ 1236] Overall Loss 0.322065 Objective Loss 0.322065 LR 0.001000 Time 0.021036 +2023-10-05 21:10:00,593 - Epoch: [59][ 800/ 1236] Overall Loss 0.322383 Objective Loss 0.322383 LR 0.001000 Time 0.021025 +2023-10-05 21:10:00,793 - Epoch: [59][ 810/ 1236] Overall Loss 0.322164 Objective Loss 0.322164 LR 0.001000 Time 0.021012 +2023-10-05 21:10:00,996 - Epoch: [59][ 820/ 1236] Overall Loss 0.322155 Objective Loss 0.322155 LR 0.001000 Time 0.021003 +2023-10-05 21:10:01,196 - Epoch: [59][ 830/ 1236] Overall Loss 0.322159 Objective Loss 0.322159 LR 0.001000 Time 0.020990 +2023-10-05 21:10:01,398 - Epoch: [59][ 840/ 1236] Overall Loss 0.322429 Objective Loss 0.322429 LR 0.001000 Time 0.020981 +2023-10-05 21:10:01,598 - Epoch: [59][ 850/ 1236] Overall Loss 0.322104 Objective Loss 0.322104 LR 0.001000 Time 0.020969 +2023-10-05 21:10:01,800 - Epoch: [59][ 860/ 1236] Overall Loss 0.322046 Objective Loss 0.322046 LR 0.001000 Time 0.020959 +2023-10-05 21:10:02,000 - Epoch: [59][ 870/ 1236] Overall Loss 0.322360 Objective Loss 0.322360 LR 0.001000 Time 0.020948 +2023-10-05 21:10:02,202 - Epoch: [59][ 880/ 1236] Overall Loss 0.322384 Objective Loss 0.322384 LR 0.001000 Time 0.020939 +2023-10-05 21:10:02,402 - Epoch: [59][ 890/ 1236] Overall Loss 0.322170 Objective Loss 0.322170 LR 0.001000 Time 0.020928 +2023-10-05 21:10:02,604 - Epoch: [59][ 900/ 1236] Overall Loss 0.322096 Objective Loss 0.322096 LR 0.001000 Time 0.020920 +2023-10-05 21:10:02,801 - Epoch: [59][ 910/ 1236] Overall Loss 0.322150 Objective Loss 0.322150 LR 0.001000 Time 0.020906 +2023-10-05 21:10:03,004 - Epoch: [59][ 920/ 1236] Overall Loss 0.322373 Objective Loss 0.322373 LR 0.001000 Time 0.020899 +2023-10-05 21:10:03,204 - Epoch: [59][ 930/ 1236] Overall Loss 0.321955 Objective Loss 0.321955 LR 0.001000 Time 0.020889 +2023-10-05 21:10:03,406 - Epoch: [59][ 940/ 1236] Overall Loss 0.321824 Objective Loss 0.321824 LR 0.001000 Time 0.020881 +2023-10-05 21:10:03,606 - Epoch: [59][ 950/ 1236] Overall Loss 0.322075 Objective Loss 0.322075 LR 0.001000 Time 0.020872 +2023-10-05 21:10:03,809 - Epoch: [59][ 960/ 1236] Overall Loss 0.321887 Objective Loss 0.321887 LR 0.001000 Time 0.020865 +2023-10-05 21:10:04,009 - Epoch: [59][ 970/ 1236] Overall Loss 0.321866 Objective Loss 0.321866 LR 0.001000 Time 0.020856 +2023-10-05 21:10:04,211 - Epoch: [59][ 980/ 1236] Overall Loss 0.321634 Objective Loss 0.321634 LR 0.001000 Time 0.020849 +2023-10-05 21:10:04,411 - Epoch: [59][ 990/ 1236] Overall Loss 0.321792 Objective Loss 0.321792 LR 0.001000 Time 0.020840 +2023-10-05 21:10:04,613 - Epoch: [59][ 1000/ 1236] Overall Loss 0.322012 Objective Loss 0.322012 LR 0.001000 Time 0.020834 +2023-10-05 21:10:04,813 - Epoch: [59][ 1010/ 1236] Overall Loss 0.321946 Objective Loss 0.321946 LR 0.001000 Time 0.020825 +2023-10-05 21:10:05,016 - Epoch: [59][ 1020/ 1236] Overall Loss 0.321712 Objective Loss 0.321712 LR 0.001000 Time 0.020819 +2023-10-05 21:10:05,216 - Epoch: [59][ 1030/ 1236] Overall Loss 0.322026 Objective Loss 0.322026 LR 0.001000 Time 0.020811 +2023-10-05 21:10:05,418 - Epoch: [59][ 1040/ 1236] Overall Loss 0.322142 Objective Loss 0.322142 LR 0.001000 Time 0.020805 +2023-10-05 21:10:05,618 - Epoch: [59][ 1050/ 1236] Overall Loss 0.322223 Objective Loss 0.322223 LR 0.001000 Time 0.020797 +2023-10-05 21:10:05,820 - Epoch: [59][ 1060/ 1236] Overall Loss 0.322429 Objective Loss 0.322429 LR 0.001000 Time 0.020791 +2023-10-05 21:10:06,019 - Epoch: [59][ 1070/ 1236] Overall Loss 0.322372 Objective Loss 0.322372 LR 0.001000 Time 0.020783 +2023-10-05 21:10:06,222 - Epoch: [59][ 1080/ 1236] Overall Loss 0.322381 Objective Loss 0.322381 LR 0.001000 Time 0.020777 +2023-10-05 21:10:06,422 - Epoch: [59][ 1090/ 1236] Overall Loss 0.322667 Objective Loss 0.322667 LR 0.001000 Time 0.020770 +2023-10-05 21:10:06,624 - Epoch: [59][ 1100/ 1236] Overall Loss 0.322635 Objective Loss 0.322635 LR 0.001000 Time 0.020764 +2023-10-05 21:10:06,824 - Epoch: [59][ 1110/ 1236] Overall Loss 0.322757 Objective Loss 0.322757 LR 0.001000 Time 0.020757 +2023-10-05 21:10:07,026 - Epoch: [59][ 1120/ 1236] Overall Loss 0.322744 Objective Loss 0.322744 LR 0.001000 Time 0.020752 +2023-10-05 21:10:07,226 - Epoch: [59][ 1130/ 1236] Overall Loss 0.322749 Objective Loss 0.322749 LR 0.001000 Time 0.020745 +2023-10-05 21:10:07,428 - Epoch: [59][ 1140/ 1236] Overall Loss 0.322774 Objective Loss 0.322774 LR 0.001000 Time 0.020740 +2023-10-05 21:10:07,628 - Epoch: [59][ 1150/ 1236] Overall Loss 0.322855 Objective Loss 0.322855 LR 0.001000 Time 0.020733 +2023-10-05 21:10:07,830 - Epoch: [59][ 1160/ 1236] Overall Loss 0.323022 Objective Loss 0.323022 LR 0.001000 Time 0.020728 +2023-10-05 21:10:08,030 - Epoch: [59][ 1170/ 1236] Overall Loss 0.323005 Objective Loss 0.323005 LR 0.001000 Time 0.020722 +2023-10-05 21:10:08,232 - Epoch: [59][ 1180/ 1236] Overall Loss 0.323208 Objective Loss 0.323208 LR 0.001000 Time 0.020717 +2023-10-05 21:10:08,432 - Epoch: [59][ 1190/ 1236] Overall Loss 0.323353 Objective Loss 0.323353 LR 0.001000 Time 0.020711 +2023-10-05 21:10:08,635 - Epoch: [59][ 1200/ 1236] Overall Loss 0.323384 Objective Loss 0.323384 LR 0.001000 Time 0.020707 +2023-10-05 21:10:08,835 - Epoch: [59][ 1210/ 1236] Overall Loss 0.323412 Objective Loss 0.323412 LR 0.001000 Time 0.020701 +2023-10-05 21:10:09,039 - Epoch: [59][ 1220/ 1236] Overall Loss 0.323368 Objective Loss 0.323368 LR 0.001000 Time 0.020695 +2023-10-05 21:10:09,292 - Epoch: [59][ 1230/ 1236] Overall Loss 0.323424 Objective Loss 0.323424 LR 0.001000 Time 0.020733 +2023-10-05 21:10:09,409 - Epoch: [59][ 1236/ 1236] Overall Loss 0.323329 Objective Loss 0.323329 Top1 84.317719 Top5 97.963340 LR 0.001000 Time 0.020727 +2023-10-05 21:10:09,540 - --- validate (epoch=59)----------- +2023-10-05 21:10:09,540 - 29943 samples (256 per mini-batch) +2023-10-05 21:10:09,987 - Epoch: [59][ 10/ 117] Loss 0.419110 Top1 80.546875 Top5 97.187500 +2023-10-05 21:10:10,134 - Epoch: [59][ 20/ 117] Loss 0.405004 Top1 80.722656 Top5 97.304688 +2023-10-05 21:10:10,278 - Epoch: [59][ 30/ 117] Loss 0.406172 Top1 80.911458 Top5 97.187500 +2023-10-05 21:10:10,424 - Epoch: [59][ 40/ 117] Loss 0.398481 Top1 80.898438 Top5 97.275391 +2023-10-05 21:10:10,567 - Epoch: [59][ 50/ 117] Loss 0.385026 Top1 81.257812 Top5 97.304688 +2023-10-05 21:10:10,712 - Epoch: [59][ 60/ 117] Loss 0.385184 Top1 81.145833 Top5 97.350260 +2023-10-05 21:10:10,856 - Epoch: [59][ 70/ 117] Loss 0.380905 Top1 81.049107 Top5 97.360491 +2023-10-05 21:10:11,002 - Epoch: [59][ 80/ 117] Loss 0.380442 Top1 81.103516 Top5 97.387695 +2023-10-05 21:10:11,147 - Epoch: [59][ 90/ 117] Loss 0.376951 Top1 81.341146 Top5 97.417535 +2023-10-05 21:10:11,293 - Epoch: [59][ 100/ 117] Loss 0.376309 Top1 81.320312 Top5 97.433594 +2023-10-05 21:10:11,446 - Epoch: [59][ 110/ 117] Loss 0.374089 Top1 81.445312 Top5 97.425426 +2023-10-05 21:10:11,531 - Epoch: [59][ 117/ 117] Loss 0.373304 Top1 81.481481 Top5 97.435127 +2023-10-05 21:10:11,645 - ==> Top1: 81.481 Top5: 97.435 Loss: 0.373 + +2023-10-05 21:10:11,645 - ==> Confusion: +[[ 921 3 5 2 13 3 0 0 4 68 2 0 1 1 3 7 5 0 0 0 12] + [ 1 1032 2 0 12 19 2 28 4 0 4 1 0 2 0 3 8 0 9 1 3] + [ 5 2 914 19 3 0 49 10 0 1 3 6 6 2 1 6 4 0 3 8 14] + [ 3 1 27 931 1 7 6 1 5 0 5 0 3 6 25 4 3 10 30 5 16] + [ 21 9 1 0 966 8 0 0 0 6 0 2 0 1 6 6 14 2 1 0 7] + [ 3 48 1 1 3 978 3 22 0 2 1 11 1 7 5 1 5 3 3 8 10] + [ 0 8 18 1 0 1 1122 3 0 0 2 4 2 0 1 12 0 0 1 6 10] + [ 6 20 17 0 4 38 9 1051 0 4 0 10 0 2 1 1 0 1 28 17 9] + [ 28 1 0 0 1 4 0 2 940 60 10 3 5 12 13 3 1 1 2 0 3] + [ 115 0 0 0 14 5 0 1 17 911 2 1 0 24 2 9 1 2 0 2 13] + [ 3 5 12 6 0 3 16 8 14 1 943 3 0 15 6 0 1 0 6 1 10] + [ 1 0 0 0 0 10 1 0 0 0 0 955 30 4 0 3 2 13 0 16 0] + [ 0 0 8 4 1 6 0 2 1 0 0 56 942 0 1 13 2 11 1 8 12] + [ 2 0 2 0 8 25 0 0 9 17 5 6 2 1022 2 1 2 1 0 6 9] + [ 13 3 3 13 11 0 0 0 26 8 0 1 3 3 976 0 5 3 13 0 20] + [ 1 0 1 0 4 0 1 0 0 0 0 9 9 0 0 1070 10 6 1 14 8] + [ 0 13 1 0 9 6 1 0 0 0 0 5 0 0 1 14 1088 1 0 9 13] + [ 0 0 0 2 0 0 2 1 1 0 0 8 23 1 2 24 0 965 1 1 7] + [ 0 5 10 11 1 3 4 40 5 0 10 3 2 0 14 0 0 0 942 5 13] + [ 0 5 2 0 1 4 6 12 0 0 2 21 2 0 0 3 9 0 1 1080 4] + [ 151 241 179 44 137 236 50 113 94 86 176 187 310 327 123 90 223 62 151 276 4649]] + +2023-10-05 21:10:11,647 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:10:11,647 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:10:11,652 - + +2023-10-05 21:10:11,652 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:10:12,744 - Epoch: [60][ 10/ 1236] Overall Loss 0.335638 Objective Loss 0.335638 LR 0.001000 Time 0.109093 +2023-10-05 21:10:12,944 - Epoch: [60][ 20/ 1236] Overall Loss 0.323941 Objective Loss 0.323941 LR 0.001000 Time 0.064536 +2023-10-05 21:10:13,142 - Epoch: [60][ 30/ 1236] Overall Loss 0.323524 Objective Loss 0.323524 LR 0.001000 Time 0.049628 +2023-10-05 21:10:13,342 - Epoch: [60][ 40/ 1236] Overall Loss 0.321398 Objective Loss 0.321398 LR 0.001000 Time 0.042211 +2023-10-05 21:10:13,541 - Epoch: [60][ 50/ 1236] Overall Loss 0.317372 Objective Loss 0.317372 LR 0.001000 Time 0.037735 +2023-10-05 21:10:13,741 - Epoch: [60][ 60/ 1236] Overall Loss 0.317872 Objective Loss 0.317872 LR 0.001000 Time 0.034777 +2023-10-05 21:10:13,940 - Epoch: [60][ 70/ 1236] Overall Loss 0.320133 Objective Loss 0.320133 LR 0.001000 Time 0.032642 +2023-10-05 21:10:14,140 - Epoch: [60][ 80/ 1236] Overall Loss 0.321586 Objective Loss 0.321586 LR 0.001000 Time 0.031057 +2023-10-05 21:10:14,338 - Epoch: [60][ 90/ 1236] Overall Loss 0.321376 Objective Loss 0.321376 LR 0.001000 Time 0.029809 +2023-10-05 21:10:14,538 - Epoch: [60][ 100/ 1236] Overall Loss 0.322948 Objective Loss 0.322948 LR 0.001000 Time 0.028823 +2023-10-05 21:10:14,737 - Epoch: [60][ 110/ 1236] Overall Loss 0.323333 Objective Loss 0.323333 LR 0.001000 Time 0.028004 +2023-10-05 21:10:14,937 - Epoch: [60][ 120/ 1236] Overall Loss 0.321503 Objective Loss 0.321503 LR 0.001000 Time 0.027337 +2023-10-05 21:10:15,135 - Epoch: [60][ 130/ 1236] Overall Loss 0.318963 Objective Loss 0.318963 LR 0.001000 Time 0.026757 +2023-10-05 21:10:15,335 - Epoch: [60][ 140/ 1236] Overall Loss 0.317291 Objective Loss 0.317291 LR 0.001000 Time 0.026271 +2023-10-05 21:10:15,534 - Epoch: [60][ 150/ 1236] Overall Loss 0.316225 Objective Loss 0.316225 LR 0.001000 Time 0.025844 +2023-10-05 21:10:15,734 - Epoch: [60][ 160/ 1236] Overall Loss 0.315077 Objective Loss 0.315077 LR 0.001000 Time 0.025476 +2023-10-05 21:10:15,932 - Epoch: [60][ 170/ 1236] Overall Loss 0.314818 Objective Loss 0.314818 LR 0.001000 Time 0.025140 +2023-10-05 21:10:16,131 - Epoch: [60][ 180/ 1236] Overall Loss 0.314870 Objective Loss 0.314870 LR 0.001000 Time 0.024849 +2023-10-05 21:10:16,330 - Epoch: [60][ 190/ 1236] Overall Loss 0.314233 Objective Loss 0.314233 LR 0.001000 Time 0.024586 +2023-10-05 21:10:16,530 - Epoch: [60][ 200/ 1236] Overall Loss 0.312564 Objective Loss 0.312564 LR 0.001000 Time 0.024355 +2023-10-05 21:10:16,729 - Epoch: [60][ 210/ 1236] Overall Loss 0.312148 Objective Loss 0.312148 LR 0.001000 Time 0.024139 +2023-10-05 21:10:16,928 - Epoch: [60][ 220/ 1236] Overall Loss 0.311653 Objective Loss 0.311653 LR 0.001000 Time 0.023946 +2023-10-05 21:10:17,127 - Epoch: [60][ 230/ 1236] Overall Loss 0.314317 Objective Loss 0.314317 LR 0.001000 Time 0.023768 +2023-10-05 21:10:17,327 - Epoch: [60][ 240/ 1236] Overall Loss 0.313615 Objective Loss 0.313615 LR 0.001000 Time 0.023611 +2023-10-05 21:10:17,527 - Epoch: [60][ 250/ 1236] Overall Loss 0.313981 Objective Loss 0.313981 LR 0.001000 Time 0.023463 +2023-10-05 21:10:17,727 - Epoch: [60][ 260/ 1236] Overall Loss 0.313475 Objective Loss 0.313475 LR 0.001000 Time 0.023329 +2023-10-05 21:10:17,925 - Epoch: [60][ 270/ 1236] Overall Loss 0.312766 Objective Loss 0.312766 LR 0.001000 Time 0.023199 +2023-10-05 21:10:18,125 - Epoch: [60][ 280/ 1236] Overall Loss 0.311409 Objective Loss 0.311409 LR 0.001000 Time 0.023082 +2023-10-05 21:10:18,324 - Epoch: [60][ 290/ 1236] Overall Loss 0.311816 Objective Loss 0.311816 LR 0.001000 Time 0.022970 +2023-10-05 21:10:18,524 - Epoch: [60][ 300/ 1236] Overall Loss 0.312227 Objective Loss 0.312227 LR 0.001000 Time 0.022871 +2023-10-05 21:10:18,723 - Epoch: [60][ 310/ 1236] Overall Loss 0.311619 Objective Loss 0.311619 LR 0.001000 Time 0.022774 +2023-10-05 21:10:18,923 - Epoch: [60][ 320/ 1236] Overall Loss 0.311362 Objective Loss 0.311362 LR 0.001000 Time 0.022687 +2023-10-05 21:10:19,123 - Epoch: [60][ 330/ 1236] Overall Loss 0.311597 Objective Loss 0.311597 LR 0.001000 Time 0.022603 +2023-10-05 21:10:19,323 - Epoch: [60][ 340/ 1236] Overall Loss 0.312069 Objective Loss 0.312069 LR 0.001000 Time 0.022526 +2023-10-05 21:10:19,522 - Epoch: [60][ 350/ 1236] Overall Loss 0.312069 Objective Loss 0.312069 LR 0.001000 Time 0.022449 +2023-10-05 21:10:19,722 - Epoch: [60][ 360/ 1236] Overall Loss 0.312307 Objective Loss 0.312307 LR 0.001000 Time 0.022381 +2023-10-05 21:10:19,921 - Epoch: [60][ 370/ 1236] Overall Loss 0.312116 Objective Loss 0.312116 LR 0.001000 Time 0.022313 +2023-10-05 21:10:20,121 - Epoch: [60][ 380/ 1236] Overall Loss 0.312177 Objective Loss 0.312177 LR 0.001000 Time 0.022251 +2023-10-05 21:10:20,320 - Epoch: [60][ 390/ 1236] Overall Loss 0.312216 Objective Loss 0.312216 LR 0.001000 Time 0.022190 +2023-10-05 21:10:20,520 - Epoch: [60][ 400/ 1236] Overall Loss 0.311559 Objective Loss 0.311559 LR 0.001000 Time 0.022134 +2023-10-05 21:10:20,719 - Epoch: [60][ 410/ 1236] Overall Loss 0.311723 Objective Loss 0.311723 LR 0.001000 Time 0.022079 +2023-10-05 21:10:20,919 - Epoch: [60][ 420/ 1236] Overall Loss 0.311562 Objective Loss 0.311562 LR 0.001000 Time 0.022029 +2023-10-05 21:10:21,118 - Epoch: [60][ 430/ 1236] Overall Loss 0.311002 Objective Loss 0.311002 LR 0.001000 Time 0.021978 +2023-10-05 21:10:21,318 - Epoch: [60][ 440/ 1236] Overall Loss 0.311297 Objective Loss 0.311297 LR 0.001000 Time 0.021933 +2023-10-05 21:10:21,517 - Epoch: [60][ 450/ 1236] Overall Loss 0.311388 Objective Loss 0.311388 LR 0.001000 Time 0.021887 +2023-10-05 21:10:21,717 - Epoch: [60][ 460/ 1236] Overall Loss 0.311466 Objective Loss 0.311466 LR 0.001000 Time 0.021845 +2023-10-05 21:10:21,916 - Epoch: [60][ 470/ 1236] Overall Loss 0.311466 Objective Loss 0.311466 LR 0.001000 Time 0.021803 +2023-10-05 21:10:22,116 - Epoch: [60][ 480/ 1236] Overall Loss 0.312009 Objective Loss 0.312009 LR 0.001000 Time 0.021765 +2023-10-05 21:10:22,315 - Epoch: [60][ 490/ 1236] Overall Loss 0.312003 Objective Loss 0.312003 LR 0.001000 Time 0.021727 +2023-10-05 21:10:22,515 - Epoch: [60][ 500/ 1236] Overall Loss 0.311746 Objective Loss 0.311746 LR 0.001000 Time 0.021692 +2023-10-05 21:10:22,714 - Epoch: [60][ 510/ 1236] Overall Loss 0.311383 Objective Loss 0.311383 LR 0.001000 Time 0.021656 +2023-10-05 21:10:22,915 - Epoch: [60][ 520/ 1236] Overall Loss 0.311727 Objective Loss 0.311727 LR 0.001000 Time 0.021624 +2023-10-05 21:10:23,114 - Epoch: [60][ 530/ 1236] Overall Loss 0.311548 Objective Loss 0.311548 LR 0.001000 Time 0.021591 +2023-10-05 21:10:23,314 - Epoch: [60][ 540/ 1236] Overall Loss 0.312110 Objective Loss 0.312110 LR 0.001000 Time 0.021561 +2023-10-05 21:10:23,513 - Epoch: [60][ 550/ 1236] Overall Loss 0.312256 Objective Loss 0.312256 LR 0.001000 Time 0.021530 +2023-10-05 21:10:23,713 - Epoch: [60][ 560/ 1236] Overall Loss 0.312531 Objective Loss 0.312531 LR 0.001000 Time 0.021502 +2023-10-05 21:10:23,912 - Epoch: [60][ 570/ 1236] Overall Loss 0.312487 Objective Loss 0.312487 LR 0.001000 Time 0.021474 +2023-10-05 21:10:24,112 - Epoch: [60][ 580/ 1236] Overall Loss 0.312628 Objective Loss 0.312628 LR 0.001000 Time 0.021448 +2023-10-05 21:10:24,311 - Epoch: [60][ 590/ 1236] Overall Loss 0.312887 Objective Loss 0.312887 LR 0.001000 Time 0.021421 +2023-10-05 21:10:24,511 - Epoch: [60][ 600/ 1236] Overall Loss 0.312952 Objective Loss 0.312952 LR 0.001000 Time 0.021397 +2023-10-05 21:10:24,710 - Epoch: [60][ 610/ 1236] Overall Loss 0.313059 Objective Loss 0.313059 LR 0.001000 Time 0.021372 +2023-10-05 21:10:24,910 - Epoch: [60][ 620/ 1236] Overall Loss 0.313179 Objective Loss 0.313179 LR 0.001000 Time 0.021349 +2023-10-05 21:10:25,109 - Epoch: [60][ 630/ 1236] Overall Loss 0.313225 Objective Loss 0.313225 LR 0.001000 Time 0.021326 +2023-10-05 21:10:25,310 - Epoch: [60][ 640/ 1236] Overall Loss 0.312973 Objective Loss 0.312973 LR 0.001000 Time 0.021305 +2023-10-05 21:10:25,509 - Epoch: [60][ 650/ 1236] Overall Loss 0.313176 Objective Loss 0.313176 LR 0.001000 Time 0.021283 +2023-10-05 21:10:25,709 - Epoch: [60][ 660/ 1236] Overall Loss 0.313344 Objective Loss 0.313344 LR 0.001000 Time 0.021263 +2023-10-05 21:10:25,908 - Epoch: [60][ 670/ 1236] Overall Loss 0.312883 Objective Loss 0.312883 LR 0.001000 Time 0.021242 +2023-10-05 21:10:26,108 - Epoch: [60][ 680/ 1236] Overall Loss 0.313304 Objective Loss 0.313304 LR 0.001000 Time 0.021223 +2023-10-05 21:10:26,307 - Epoch: [60][ 690/ 1236] Overall Loss 0.313623 Objective Loss 0.313623 LR 0.001000 Time 0.021204 +2023-10-05 21:10:26,507 - Epoch: [60][ 700/ 1236] Overall Loss 0.314349 Objective Loss 0.314349 LR 0.001000 Time 0.021186 +2023-10-05 21:10:26,706 - Epoch: [60][ 710/ 1236] Overall Loss 0.314938 Objective Loss 0.314938 LR 0.001000 Time 0.021168 +2023-10-05 21:10:26,906 - Epoch: [60][ 720/ 1236] Overall Loss 0.314882 Objective Loss 0.314882 LR 0.001000 Time 0.021151 +2023-10-05 21:10:27,105 - Epoch: [60][ 730/ 1236] Overall Loss 0.314623 Objective Loss 0.314623 LR 0.001000 Time 0.021133 +2023-10-05 21:10:27,305 - Epoch: [60][ 740/ 1236] Overall Loss 0.314499 Objective Loss 0.314499 LR 0.001000 Time 0.021118 +2023-10-05 21:10:27,504 - Epoch: [60][ 750/ 1236] Overall Loss 0.314944 Objective Loss 0.314944 LR 0.001000 Time 0.021101 +2023-10-05 21:10:27,705 - Epoch: [60][ 760/ 1236] Overall Loss 0.314979 Objective Loss 0.314979 LR 0.001000 Time 0.021086 +2023-10-05 21:10:27,904 - Epoch: [60][ 770/ 1236] Overall Loss 0.315140 Objective Loss 0.315140 LR 0.001000 Time 0.021071 +2023-10-05 21:10:28,104 - Epoch: [60][ 780/ 1236] Overall Loss 0.315093 Objective Loss 0.315093 LR 0.001000 Time 0.021057 +2023-10-05 21:10:28,303 - Epoch: [60][ 790/ 1236] Overall Loss 0.315348 Objective Loss 0.315348 LR 0.001000 Time 0.021042 +2023-10-05 21:10:28,503 - Epoch: [60][ 800/ 1236] Overall Loss 0.314895 Objective Loss 0.314895 LR 0.001000 Time 0.021029 +2023-10-05 21:10:28,702 - Epoch: [60][ 810/ 1236] Overall Loss 0.315076 Objective Loss 0.315076 LR 0.001000 Time 0.021014 +2023-10-05 21:10:28,902 - Epoch: [60][ 820/ 1236] Overall Loss 0.315091 Objective Loss 0.315091 LR 0.001000 Time 0.021002 +2023-10-05 21:10:29,101 - Epoch: [60][ 830/ 1236] Overall Loss 0.315055 Objective Loss 0.315055 LR 0.001000 Time 0.020988 +2023-10-05 21:10:29,302 - Epoch: [60][ 840/ 1236] Overall Loss 0.315229 Objective Loss 0.315229 LR 0.001000 Time 0.020976 +2023-10-05 21:10:29,501 - Epoch: [60][ 850/ 1236] Overall Loss 0.315326 Objective Loss 0.315326 LR 0.001000 Time 0.020964 +2023-10-05 21:10:29,701 - Epoch: [60][ 860/ 1236] Overall Loss 0.315214 Objective Loss 0.315214 LR 0.001000 Time 0.020952 +2023-10-05 21:10:29,900 - Epoch: [60][ 870/ 1236] Overall Loss 0.314990 Objective Loss 0.314990 LR 0.001000 Time 0.020940 +2023-10-05 21:10:30,101 - Epoch: [60][ 880/ 1236] Overall Loss 0.315297 Objective Loss 0.315297 LR 0.001000 Time 0.020929 +2023-10-05 21:10:30,300 - Epoch: [60][ 890/ 1236] Overall Loss 0.315721 Objective Loss 0.315721 LR 0.001000 Time 0.020918 +2023-10-05 21:10:30,500 - Epoch: [60][ 900/ 1236] Overall Loss 0.316093 Objective Loss 0.316093 LR 0.001000 Time 0.020908 +2023-10-05 21:10:30,699 - Epoch: [60][ 910/ 1236] Overall Loss 0.315858 Objective Loss 0.315858 LR 0.001000 Time 0.020896 +2023-10-05 21:10:30,899 - Epoch: [60][ 920/ 1236] Overall Loss 0.315791 Objective Loss 0.315791 LR 0.001000 Time 0.020886 +2023-10-05 21:10:31,098 - Epoch: [60][ 930/ 1236] Overall Loss 0.316110 Objective Loss 0.316110 LR 0.001000 Time 0.020875 +2023-10-05 21:10:31,299 - Epoch: [60][ 940/ 1236] Overall Loss 0.316142 Objective Loss 0.316142 LR 0.001000 Time 0.020865 +2023-10-05 21:10:31,498 - Epoch: [60][ 950/ 1236] Overall Loss 0.315947 Objective Loss 0.315947 LR 0.001000 Time 0.020855 +2023-10-05 21:10:31,698 - Epoch: [60][ 960/ 1236] Overall Loss 0.316001 Objective Loss 0.316001 LR 0.001000 Time 0.020846 +2023-10-05 21:10:31,897 - Epoch: [60][ 970/ 1236] Overall Loss 0.315850 Objective Loss 0.315850 LR 0.001000 Time 0.020836 +2023-10-05 21:10:32,097 - Epoch: [60][ 980/ 1236] Overall Loss 0.315855 Objective Loss 0.315855 LR 0.001000 Time 0.020827 +2023-10-05 21:10:32,296 - Epoch: [60][ 990/ 1236] Overall Loss 0.315837 Objective Loss 0.315837 LR 0.001000 Time 0.020817 +2023-10-05 21:10:32,496 - Epoch: [60][ 1000/ 1236] Overall Loss 0.316070 Objective Loss 0.316070 LR 0.001000 Time 0.020809 +2023-10-05 21:10:32,695 - Epoch: [60][ 1010/ 1236] Overall Loss 0.316149 Objective Loss 0.316149 LR 0.001000 Time 0.020800 +2023-10-05 21:10:32,895 - Epoch: [60][ 1020/ 1236] Overall Loss 0.316010 Objective Loss 0.316010 LR 0.001000 Time 0.020791 +2023-10-05 21:10:33,094 - Epoch: [60][ 1030/ 1236] Overall Loss 0.316260 Objective Loss 0.316260 LR 0.001000 Time 0.020782 +2023-10-05 21:10:33,294 - Epoch: [60][ 1040/ 1236] Overall Loss 0.316626 Objective Loss 0.316626 LR 0.001000 Time 0.020775 +2023-10-05 21:10:33,493 - Epoch: [60][ 1050/ 1236] Overall Loss 0.316838 Objective Loss 0.316838 LR 0.001000 Time 0.020766 +2023-10-05 21:10:33,693 - Epoch: [60][ 1060/ 1236] Overall Loss 0.317102 Objective Loss 0.317102 LR 0.001000 Time 0.020758 +2023-10-05 21:10:33,892 - Epoch: [60][ 1070/ 1236] Overall Loss 0.317341 Objective Loss 0.317341 LR 0.001000 Time 0.020750 +2023-10-05 21:10:34,092 - Epoch: [60][ 1080/ 1236] Overall Loss 0.317692 Objective Loss 0.317692 LR 0.001000 Time 0.020743 +2023-10-05 21:10:34,292 - Epoch: [60][ 1090/ 1236] Overall Loss 0.318038 Objective Loss 0.318038 LR 0.001000 Time 0.020736 +2023-10-05 21:10:34,492 - Epoch: [60][ 1100/ 1236] Overall Loss 0.318180 Objective Loss 0.318180 LR 0.001000 Time 0.020729 +2023-10-05 21:10:34,692 - Epoch: [60][ 1110/ 1236] Overall Loss 0.318148 Objective Loss 0.318148 LR 0.001000 Time 0.020721 +2023-10-05 21:10:34,892 - Epoch: [60][ 1120/ 1236] Overall Loss 0.318150 Objective Loss 0.318150 LR 0.001000 Time 0.020715 +2023-10-05 21:10:35,091 - Epoch: [60][ 1130/ 1236] Overall Loss 0.318172 Objective Loss 0.318172 LR 0.001000 Time 0.020707 +2023-10-05 21:10:35,291 - Epoch: [60][ 1140/ 1236] Overall Loss 0.317877 Objective Loss 0.317877 LR 0.001000 Time 0.020701 +2023-10-05 21:10:35,490 - Epoch: [60][ 1150/ 1236] Overall Loss 0.318013 Objective Loss 0.318013 LR 0.001000 Time 0.020694 +2023-10-05 21:10:35,690 - Epoch: [60][ 1160/ 1236] Overall Loss 0.317926 Objective Loss 0.317926 LR 0.001000 Time 0.020688 +2023-10-05 21:10:35,889 - Epoch: [60][ 1170/ 1236] Overall Loss 0.318167 Objective Loss 0.318167 LR 0.001000 Time 0.020681 +2023-10-05 21:10:36,090 - Epoch: [60][ 1180/ 1236] Overall Loss 0.318171 Objective Loss 0.318171 LR 0.001000 Time 0.020675 +2023-10-05 21:10:36,289 - Epoch: [60][ 1190/ 1236] Overall Loss 0.318164 Objective Loss 0.318164 LR 0.001000 Time 0.020668 +2023-10-05 21:10:36,489 - Epoch: [60][ 1200/ 1236] Overall Loss 0.318132 Objective Loss 0.318132 LR 0.001000 Time 0.020662 +2023-10-05 21:10:36,688 - Epoch: [60][ 1210/ 1236] Overall Loss 0.318236 Objective Loss 0.318236 LR 0.001000 Time 0.020656 +2023-10-05 21:10:36,889 - Epoch: [60][ 1220/ 1236] Overall Loss 0.318193 Objective Loss 0.318193 LR 0.001000 Time 0.020651 +2023-10-05 21:10:37,142 - Epoch: [60][ 1230/ 1236] Overall Loss 0.318145 Objective Loss 0.318145 LR 0.001000 Time 0.020688 +2023-10-05 21:10:37,259 - Epoch: [60][ 1236/ 1236] Overall Loss 0.318244 Objective Loss 0.318244 Top1 83.910387 Top5 98.167006 LR 0.001000 Time 0.020683 +2023-10-05 21:10:37,380 - --- validate (epoch=60)----------- +2023-10-05 21:10:37,380 - 29943 samples (256 per mini-batch) +2023-10-05 21:10:37,834 - Epoch: [60][ 10/ 117] Loss 0.371656 Top1 80.468750 Top5 97.578125 +2023-10-05 21:10:37,981 - Epoch: [60][ 20/ 117] Loss 0.363884 Top1 80.898438 Top5 97.167969 +2023-10-05 21:10:38,128 - Epoch: [60][ 30/ 117] Loss 0.353295 Top1 81.393229 Top5 97.265625 +2023-10-05 21:10:38,273 - Epoch: [60][ 40/ 117] Loss 0.356323 Top1 81.357422 Top5 97.392578 +2023-10-05 21:10:38,419 - Epoch: [60][ 50/ 117] Loss 0.358687 Top1 81.375000 Top5 97.343750 +2023-10-05 21:10:38,565 - Epoch: [60][ 60/ 117] Loss 0.359126 Top1 81.354167 Top5 97.330729 +2023-10-05 21:10:38,712 - Epoch: [60][ 70/ 117] Loss 0.354124 Top1 81.395089 Top5 97.382812 +2023-10-05 21:10:38,858 - Epoch: [60][ 80/ 117] Loss 0.353160 Top1 81.372070 Top5 97.377930 +2023-10-05 21:10:39,008 - Epoch: [60][ 90/ 117] Loss 0.354680 Top1 81.462674 Top5 97.439236 +2023-10-05 21:10:39,155 - Epoch: [60][ 100/ 117] Loss 0.354347 Top1 81.390625 Top5 97.457031 +2023-10-05 21:10:39,306 - Epoch: [60][ 110/ 117] Loss 0.353490 Top1 81.448864 Top5 97.404119 +2023-10-05 21:10:39,391 - Epoch: [60][ 117/ 117] Loss 0.355711 Top1 81.337875 Top5 97.351635 +2023-10-05 21:10:39,525 - ==> Top1: 81.338 Top5: 97.352 Loss: 0.356 + +2023-10-05 21:10:39,526 - ==> Confusion: +[[ 923 7 2 4 5 5 0 0 5 72 1 0 2 2 4 4 4 0 1 0 9] + [ 2 1043 1 0 10 26 1 22 2 2 5 2 2 0 0 2 2 0 6 0 3] + [ 6 0 938 21 4 1 24 8 0 3 9 4 7 3 3 3 4 1 5 3 9] + [ 4 2 14 956 1 8 2 2 5 0 9 0 6 6 37 5 2 4 15 1 10] + [ 30 8 0 1 953 9 0 1 1 10 1 1 1 2 13 3 11 2 1 1 1] + [ 6 41 0 2 3 996 1 14 1 0 6 5 3 15 7 2 4 0 2 3 5] + [ 0 11 28 1 1 0 1103 9 0 0 7 3 1 0 1 10 0 1 4 6 5] + [ 11 34 9 0 1 58 3 1015 0 4 8 8 4 3 0 0 1 0 40 9 10] + [ 17 1 0 4 1 10 1 0 929 46 25 3 3 20 20 0 0 1 6 0 2] + [ 111 0 3 1 5 2 0 0 17 911 0 1 1 37 9 6 1 0 0 3 11] + [ 5 7 10 10 1 2 3 3 6 0 975 1 0 11 3 1 2 0 6 0 7] + [ 2 1 3 0 0 14 0 2 0 0 0 951 31 3 0 1 0 13 1 7 6] + [ 1 3 6 5 0 5 1 1 2 0 2 39 966 1 4 6 1 12 3 2 8] + [ 1 0 0 1 2 10 0 2 14 14 8 9 2 1041 3 3 1 0 0 3 5] + [ 19 2 2 7 1 0 0 0 26 6 1 2 1 1 1001 0 4 0 15 0 13] + [ 3 1 1 1 3 0 2 0 0 0 0 13 10 0 0 1047 18 14 1 10 10] + [ 0 19 4 0 4 5 0 0 3 0 0 4 1 1 3 11 1091 0 1 6 8] + [ 1 1 1 1 0 0 0 0 2 1 0 8 26 3 3 4 2 980 1 0 4] + [ 1 13 8 20 1 3 0 34 2 0 4 0 1 1 21 0 1 0 948 1 9] + [ 0 4 1 3 3 8 4 16 0 0 1 24 5 2 1 4 6 0 2 1058 10] + [ 145 281 174 84 82 209 38 86 80 84 228 154 446 384 161 53 210 75 170 231 4530]] + +2023-10-05 21:10:39,527 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:10:39,527 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:10:39,533 - + +2023-10-05 21:10:39,533 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:10:40,509 - Epoch: [61][ 10/ 1236] Overall Loss 0.310832 Objective Loss 0.310832 LR 0.001000 Time 0.097598 +2023-10-05 21:10:40,709 - Epoch: [61][ 20/ 1236] Overall Loss 0.316505 Objective Loss 0.316505 LR 0.001000 Time 0.058764 +2023-10-05 21:10:40,907 - Epoch: [61][ 30/ 1236] Overall Loss 0.316396 Objective Loss 0.316396 LR 0.001000 Time 0.045775 +2023-10-05 21:10:41,107 - Epoch: [61][ 40/ 1236] Overall Loss 0.314448 Objective Loss 0.314448 LR 0.001000 Time 0.039320 +2023-10-05 21:10:41,306 - Epoch: [61][ 50/ 1236] Overall Loss 0.309056 Objective Loss 0.309056 LR 0.001000 Time 0.035421 +2023-10-05 21:10:41,505 - Epoch: [61][ 60/ 1236] Overall Loss 0.302109 Objective Loss 0.302109 LR 0.001000 Time 0.032840 +2023-10-05 21:10:41,704 - Epoch: [61][ 70/ 1236] Overall Loss 0.300281 Objective Loss 0.300281 LR 0.001000 Time 0.030981 +2023-10-05 21:10:41,904 - Epoch: [61][ 80/ 1236] Overall Loss 0.299461 Objective Loss 0.299461 LR 0.001000 Time 0.029599 +2023-10-05 21:10:42,102 - Epoch: [61][ 90/ 1236] Overall Loss 0.300124 Objective Loss 0.300124 LR 0.001000 Time 0.028513 +2023-10-05 21:10:42,302 - Epoch: [61][ 100/ 1236] Overall Loss 0.304858 Objective Loss 0.304858 LR 0.001000 Time 0.027656 +2023-10-05 21:10:42,501 - Epoch: [61][ 110/ 1236] Overall Loss 0.306081 Objective Loss 0.306081 LR 0.001000 Time 0.026946 +2023-10-05 21:10:42,701 - Epoch: [61][ 120/ 1236] Overall Loss 0.306024 Objective Loss 0.306024 LR 0.001000 Time 0.026362 +2023-10-05 21:10:42,899 - Epoch: [61][ 130/ 1236] Overall Loss 0.307602 Objective Loss 0.307602 LR 0.001000 Time 0.025860 +2023-10-05 21:10:43,099 - Epoch: [61][ 140/ 1236] Overall Loss 0.308468 Objective Loss 0.308468 LR 0.001000 Time 0.025437 +2023-10-05 21:10:43,298 - Epoch: [61][ 150/ 1236] Overall Loss 0.307632 Objective Loss 0.307632 LR 0.001000 Time 0.025062 +2023-10-05 21:10:43,497 - Epoch: [61][ 160/ 1236] Overall Loss 0.307240 Objective Loss 0.307240 LR 0.001000 Time 0.024742 +2023-10-05 21:10:43,696 - Epoch: [61][ 170/ 1236] Overall Loss 0.310072 Objective Loss 0.310072 LR 0.001000 Time 0.024452 +2023-10-05 21:10:43,896 - Epoch: [61][ 180/ 1236] Overall Loss 0.311268 Objective Loss 0.311268 LR 0.001000 Time 0.024203 +2023-10-05 21:10:44,094 - Epoch: [61][ 190/ 1236] Overall Loss 0.312254 Objective Loss 0.312254 LR 0.001000 Time 0.023973 +2023-10-05 21:10:44,294 - Epoch: [61][ 200/ 1236] Overall Loss 0.311686 Objective Loss 0.311686 LR 0.001000 Time 0.023771 +2023-10-05 21:10:44,493 - Epoch: [61][ 210/ 1236] Overall Loss 0.312591 Objective Loss 0.312591 LR 0.001000 Time 0.023583 +2023-10-05 21:10:44,693 - Epoch: [61][ 220/ 1236] Overall Loss 0.313225 Objective Loss 0.313225 LR 0.001000 Time 0.023418 +2023-10-05 21:10:44,891 - Epoch: [61][ 230/ 1236] Overall Loss 0.312605 Objective Loss 0.312605 LR 0.001000 Time 0.023262 +2023-10-05 21:10:45,091 - Epoch: [61][ 240/ 1236] Overall Loss 0.313737 Objective Loss 0.313737 LR 0.001000 Time 0.023125 +2023-10-05 21:10:45,290 - Epoch: [61][ 250/ 1236] Overall Loss 0.314139 Objective Loss 0.314139 LR 0.001000 Time 0.022995 +2023-10-05 21:10:45,490 - Epoch: [61][ 260/ 1236] Overall Loss 0.314789 Objective Loss 0.314789 LR 0.001000 Time 0.022877 +2023-10-05 21:10:45,689 - Epoch: [61][ 270/ 1236] Overall Loss 0.315162 Objective Loss 0.315162 LR 0.001000 Time 0.022765 +2023-10-05 21:10:45,889 - Epoch: [61][ 280/ 1236] Overall Loss 0.316426 Objective Loss 0.316426 LR 0.001000 Time 0.022664 +2023-10-05 21:10:46,087 - Epoch: [61][ 290/ 1236] Overall Loss 0.317602 Objective Loss 0.317602 LR 0.001000 Time 0.022566 +2023-10-05 21:10:46,287 - Epoch: [61][ 300/ 1236] Overall Loss 0.318532 Objective Loss 0.318532 LR 0.001000 Time 0.022480 +2023-10-05 21:10:46,486 - Epoch: [61][ 310/ 1236] Overall Loss 0.318290 Objective Loss 0.318290 LR 0.001000 Time 0.022395 +2023-10-05 21:10:46,686 - Epoch: [61][ 320/ 1236] Overall Loss 0.317909 Objective Loss 0.317909 LR 0.001000 Time 0.022319 +2023-10-05 21:10:46,885 - Epoch: [61][ 330/ 1236] Overall Loss 0.317833 Objective Loss 0.317833 LR 0.001000 Time 0.022244 +2023-10-05 21:10:47,085 - Epoch: [61][ 340/ 1236] Overall Loss 0.317871 Objective Loss 0.317871 LR 0.001000 Time 0.022177 +2023-10-05 21:10:47,284 - Epoch: [61][ 350/ 1236] Overall Loss 0.317874 Objective Loss 0.317874 LR 0.001000 Time 0.022111 +2023-10-05 21:10:47,484 - Epoch: [61][ 360/ 1236] Overall Loss 0.318374 Objective Loss 0.318374 LR 0.001000 Time 0.022052 +2023-10-05 21:10:47,683 - Epoch: [61][ 370/ 1236] Overall Loss 0.318128 Objective Loss 0.318128 LR 0.001000 Time 0.021993 +2023-10-05 21:10:47,884 - Epoch: [61][ 380/ 1236] Overall Loss 0.318367 Objective Loss 0.318367 LR 0.001000 Time 0.021940 +2023-10-05 21:10:48,083 - Epoch: [61][ 390/ 1236] Overall Loss 0.318811 Objective Loss 0.318811 LR 0.001000 Time 0.021888 +2023-10-05 21:10:48,283 - Epoch: [61][ 400/ 1236] Overall Loss 0.319188 Objective Loss 0.319188 LR 0.001000 Time 0.021840 +2023-10-05 21:10:48,482 - Epoch: [61][ 410/ 1236] Overall Loss 0.319490 Objective Loss 0.319490 LR 0.001000 Time 0.021792 +2023-10-05 21:10:48,682 - Epoch: [61][ 420/ 1236] Overall Loss 0.319314 Objective Loss 0.319314 LR 0.001000 Time 0.021749 +2023-10-05 21:10:48,881 - Epoch: [61][ 430/ 1236] Overall Loss 0.319498 Objective Loss 0.319498 LR 0.001000 Time 0.021705 +2023-10-05 21:10:49,081 - Epoch: [61][ 440/ 1236] Overall Loss 0.319330 Objective Loss 0.319330 LR 0.001000 Time 0.021666 +2023-10-05 21:10:49,280 - Epoch: [61][ 450/ 1236] Overall Loss 0.319090 Objective Loss 0.319090 LR 0.001000 Time 0.021625 +2023-10-05 21:10:49,481 - Epoch: [61][ 460/ 1236] Overall Loss 0.319450 Objective Loss 0.319450 LR 0.001000 Time 0.021590 +2023-10-05 21:10:49,680 - Epoch: [61][ 470/ 1236] Overall Loss 0.319455 Objective Loss 0.319455 LR 0.001000 Time 0.021554 +2023-10-05 21:10:49,880 - Epoch: [61][ 480/ 1236] Overall Loss 0.319502 Objective Loss 0.319502 LR 0.001000 Time 0.021522 +2023-10-05 21:10:50,079 - Epoch: [61][ 490/ 1236] Overall Loss 0.320415 Objective Loss 0.320415 LR 0.001000 Time 0.021488 +2023-10-05 21:10:50,280 - Epoch: [61][ 500/ 1236] Overall Loss 0.320700 Objective Loss 0.320700 LR 0.001000 Time 0.021458 +2023-10-05 21:10:50,479 - Epoch: [61][ 510/ 1236] Overall Loss 0.319800 Objective Loss 0.319800 LR 0.001000 Time 0.021427 +2023-10-05 21:10:50,679 - Epoch: [61][ 520/ 1236] Overall Loss 0.319809 Objective Loss 0.319809 LR 0.001000 Time 0.021399 +2023-10-05 21:10:50,878 - Epoch: [61][ 530/ 1236] Overall Loss 0.319754 Objective Loss 0.319754 LR 0.001000 Time 0.021371 +2023-10-05 21:10:51,078 - Epoch: [61][ 540/ 1236] Overall Loss 0.320178 Objective Loss 0.320178 LR 0.001000 Time 0.021344 +2023-10-05 21:10:51,277 - Epoch: [61][ 550/ 1236] Overall Loss 0.320367 Objective Loss 0.320367 LR 0.001000 Time 0.021318 +2023-10-05 21:10:51,477 - Epoch: [61][ 560/ 1236] Overall Loss 0.320601 Objective Loss 0.320601 LR 0.001000 Time 0.021293 +2023-10-05 21:10:51,676 - Epoch: [61][ 570/ 1236] Overall Loss 0.320853 Objective Loss 0.320853 LR 0.001000 Time 0.021269 +2023-10-05 21:10:51,877 - Epoch: [61][ 580/ 1236] Overall Loss 0.320683 Objective Loss 0.320683 LR 0.001000 Time 0.021247 +2023-10-05 21:10:52,076 - Epoch: [61][ 590/ 1236] Overall Loss 0.320903 Objective Loss 0.320903 LR 0.001000 Time 0.021225 +2023-10-05 21:10:52,276 - Epoch: [61][ 600/ 1236] Overall Loss 0.321010 Objective Loss 0.321010 LR 0.001000 Time 0.021204 +2023-10-05 21:10:52,475 - Epoch: [61][ 610/ 1236] Overall Loss 0.320399 Objective Loss 0.320399 LR 0.001000 Time 0.021182 +2023-10-05 21:10:52,676 - Epoch: [61][ 620/ 1236] Overall Loss 0.320478 Objective Loss 0.320478 LR 0.001000 Time 0.021163 +2023-10-05 21:10:52,875 - Epoch: [61][ 630/ 1236] Overall Loss 0.320774 Objective Loss 0.320774 LR 0.001000 Time 0.021143 +2023-10-05 21:10:53,076 - Epoch: [61][ 640/ 1236] Overall Loss 0.320393 Objective Loss 0.320393 LR 0.001000 Time 0.021125 +2023-10-05 21:10:53,275 - Epoch: [61][ 650/ 1236] Overall Loss 0.320498 Objective Loss 0.320498 LR 0.001000 Time 0.021106 +2023-10-05 21:10:53,475 - Epoch: [61][ 660/ 1236] Overall Loss 0.320666 Objective Loss 0.320666 LR 0.001000 Time 0.021089 +2023-10-05 21:10:53,674 - Epoch: [61][ 670/ 1236] Overall Loss 0.320346 Objective Loss 0.320346 LR 0.001000 Time 0.021071 +2023-10-05 21:10:53,874 - Epoch: [61][ 680/ 1236] Overall Loss 0.320459 Objective Loss 0.320459 LR 0.001000 Time 0.021055 +2023-10-05 21:10:54,074 - Epoch: [61][ 690/ 1236] Overall Loss 0.320736 Objective Loss 0.320736 LR 0.001000 Time 0.021036 +2023-10-05 21:10:54,275 - Epoch: [61][ 700/ 1236] Overall Loss 0.320010 Objective Loss 0.320010 LR 0.001000 Time 0.021021 +2023-10-05 21:10:54,474 - Epoch: [61][ 710/ 1236] Overall Loss 0.320380 Objective Loss 0.320380 LR 0.001000 Time 0.021005 +2023-10-05 21:10:54,674 - Epoch: [61][ 720/ 1236] Overall Loss 0.320554 Objective Loss 0.320554 LR 0.001000 Time 0.020991 +2023-10-05 21:10:54,873 - Epoch: [61][ 730/ 1236] Overall Loss 0.319901 Objective Loss 0.319901 LR 0.001000 Time 0.020975 +2023-10-05 21:10:55,073 - Epoch: [61][ 740/ 1236] Overall Loss 0.319702 Objective Loss 0.319702 LR 0.001000 Time 0.020962 +2023-10-05 21:10:55,272 - Epoch: [61][ 750/ 1236] Overall Loss 0.319678 Objective Loss 0.319678 LR 0.001000 Time 0.020947 +2023-10-05 21:10:55,472 - Epoch: [61][ 760/ 1236] Overall Loss 0.319810 Objective Loss 0.319810 LR 0.001000 Time 0.020934 +2023-10-05 21:10:55,671 - Epoch: [61][ 770/ 1236] Overall Loss 0.319718 Objective Loss 0.319718 LR 0.001000 Time 0.020921 +2023-10-05 21:10:55,872 - Epoch: [61][ 780/ 1236] Overall Loss 0.320050 Objective Loss 0.320050 LR 0.001000 Time 0.020909 +2023-10-05 21:10:56,071 - Epoch: [61][ 790/ 1236] Overall Loss 0.320251 Objective Loss 0.320251 LR 0.001000 Time 0.020896 +2023-10-05 21:10:56,271 - Epoch: [61][ 800/ 1236] Overall Loss 0.319972 Objective Loss 0.319972 LR 0.001000 Time 0.020884 +2023-10-05 21:10:56,470 - Epoch: [61][ 810/ 1236] Overall Loss 0.319850 Objective Loss 0.319850 LR 0.001000 Time 0.020872 +2023-10-05 21:10:56,671 - Epoch: [61][ 820/ 1236] Overall Loss 0.320039 Objective Loss 0.320039 LR 0.001000 Time 0.020862 +2023-10-05 21:10:56,870 - Epoch: [61][ 830/ 1236] Overall Loss 0.320300 Objective Loss 0.320300 LR 0.001000 Time 0.020850 +2023-10-05 21:10:57,071 - Epoch: [61][ 840/ 1236] Overall Loss 0.320017 Objective Loss 0.320017 LR 0.001000 Time 0.020840 +2023-10-05 21:10:57,270 - Epoch: [61][ 850/ 1236] Overall Loss 0.319783 Objective Loss 0.319783 LR 0.001000 Time 0.020830 +2023-10-05 21:10:57,470 - Epoch: [61][ 860/ 1236] Overall Loss 0.319774 Objective Loss 0.319774 LR 0.001000 Time 0.020820 +2023-10-05 21:10:57,670 - Epoch: [61][ 870/ 1236] Overall Loss 0.319706 Objective Loss 0.319706 LR 0.001000 Time 0.020809 +2023-10-05 21:10:57,869 - Epoch: [61][ 880/ 1236] Overall Loss 0.319734 Objective Loss 0.319734 LR 0.001000 Time 0.020799 +2023-10-05 21:10:58,068 - Epoch: [61][ 890/ 1236] Overall Loss 0.319700 Objective Loss 0.319700 LR 0.001000 Time 0.020789 +2023-10-05 21:10:58,269 - Epoch: [61][ 900/ 1236] Overall Loss 0.320118 Objective Loss 0.320118 LR 0.001000 Time 0.020780 +2023-10-05 21:10:58,468 - Epoch: [61][ 910/ 1236] Overall Loss 0.320107 Objective Loss 0.320107 LR 0.001000 Time 0.020771 +2023-10-05 21:10:58,669 - Epoch: [61][ 920/ 1236] Overall Loss 0.320211 Objective Loss 0.320211 LR 0.001000 Time 0.020762 +2023-10-05 21:10:58,868 - Epoch: [61][ 930/ 1236] Overall Loss 0.320067 Objective Loss 0.320067 LR 0.001000 Time 0.020753 +2023-10-05 21:10:59,069 - Epoch: [61][ 940/ 1236] Overall Loss 0.320128 Objective Loss 0.320128 LR 0.001000 Time 0.020745 +2023-10-05 21:10:59,268 - Epoch: [61][ 950/ 1236] Overall Loss 0.320532 Objective Loss 0.320532 LR 0.001000 Time 0.020737 +2023-10-05 21:10:59,469 - Epoch: [61][ 960/ 1236] Overall Loss 0.320654 Objective Loss 0.320654 LR 0.001000 Time 0.020729 +2023-10-05 21:10:59,668 - Epoch: [61][ 970/ 1236] Overall Loss 0.320388 Objective Loss 0.320388 LR 0.001000 Time 0.020721 +2023-10-05 21:10:59,869 - Epoch: [61][ 980/ 1236] Overall Loss 0.320609 Objective Loss 0.320609 LR 0.001000 Time 0.020714 +2023-10-05 21:11:00,068 - Epoch: [61][ 990/ 1236] Overall Loss 0.320583 Objective Loss 0.320583 LR 0.001000 Time 0.020706 +2023-10-05 21:11:00,269 - Epoch: [61][ 1000/ 1236] Overall Loss 0.320301 Objective Loss 0.320301 LR 0.001000 Time 0.020699 +2023-10-05 21:11:00,468 - Epoch: [61][ 1010/ 1236] Overall Loss 0.320371 Objective Loss 0.320371 LR 0.001000 Time 0.020691 +2023-10-05 21:11:00,669 - Epoch: [61][ 1020/ 1236] Overall Loss 0.320212 Objective Loss 0.320212 LR 0.001000 Time 0.020684 +2023-10-05 21:11:00,868 - Epoch: [61][ 1030/ 1236] Overall Loss 0.320373 Objective Loss 0.320373 LR 0.001000 Time 0.020676 +2023-10-05 21:11:01,068 - Epoch: [61][ 1040/ 1236] Overall Loss 0.320257 Objective Loss 0.320257 LR 0.001000 Time 0.020670 +2023-10-05 21:11:01,267 - Epoch: [61][ 1050/ 1236] Overall Loss 0.320185 Objective Loss 0.320185 LR 0.001000 Time 0.020662 +2023-10-05 21:11:01,468 - Epoch: [61][ 1060/ 1236] Overall Loss 0.320240 Objective Loss 0.320240 LR 0.001000 Time 0.020656 +2023-10-05 21:11:01,667 - Epoch: [61][ 1070/ 1236] Overall Loss 0.320197 Objective Loss 0.320197 LR 0.001000 Time 0.020649 +2023-10-05 21:11:01,868 - Epoch: [61][ 1080/ 1236] Overall Loss 0.320377 Objective Loss 0.320377 LR 0.001000 Time 0.020643 +2023-10-05 21:11:02,067 - Epoch: [61][ 1090/ 1236] Overall Loss 0.320365 Objective Loss 0.320365 LR 0.001000 Time 0.020637 +2023-10-05 21:11:02,268 - Epoch: [61][ 1100/ 1236] Overall Loss 0.320587 Objective Loss 0.320587 LR 0.001000 Time 0.020631 +2023-10-05 21:11:02,467 - Epoch: [61][ 1110/ 1236] Overall Loss 0.320654 Objective Loss 0.320654 LR 0.001000 Time 0.020625 +2023-10-05 21:11:02,668 - Epoch: [61][ 1120/ 1236] Overall Loss 0.320892 Objective Loss 0.320892 LR 0.001000 Time 0.020619 +2023-10-05 21:11:02,867 - Epoch: [61][ 1130/ 1236] Overall Loss 0.320644 Objective Loss 0.320644 LR 0.001000 Time 0.020613 +2023-10-05 21:11:03,068 - Epoch: [61][ 1140/ 1236] Overall Loss 0.320643 Objective Loss 0.320643 LR 0.001000 Time 0.020608 +2023-10-05 21:11:03,267 - Epoch: [61][ 1150/ 1236] Overall Loss 0.320793 Objective Loss 0.320793 LR 0.001000 Time 0.020602 +2023-10-05 21:11:03,467 - Epoch: [61][ 1160/ 1236] Overall Loss 0.320733 Objective Loss 0.320733 LR 0.001000 Time 0.020596 +2023-10-05 21:11:03,667 - Epoch: [61][ 1170/ 1236] Overall Loss 0.320859 Objective Loss 0.320859 LR 0.001000 Time 0.020590 +2023-10-05 21:11:03,867 - Epoch: [61][ 1180/ 1236] Overall Loss 0.320981 Objective Loss 0.320981 LR 0.001000 Time 0.020585 +2023-10-05 21:11:04,066 - Epoch: [61][ 1190/ 1236] Overall Loss 0.320974 Objective Loss 0.320974 LR 0.001000 Time 0.020580 +2023-10-05 21:11:04,267 - Epoch: [61][ 1200/ 1236] Overall Loss 0.321227 Objective Loss 0.321227 LR 0.001000 Time 0.020575 +2023-10-05 21:11:04,466 - Epoch: [61][ 1210/ 1236] Overall Loss 0.321138 Objective Loss 0.321138 LR 0.001000 Time 0.020570 +2023-10-05 21:11:04,667 - Epoch: [61][ 1220/ 1236] Overall Loss 0.321216 Objective Loss 0.321216 LR 0.001000 Time 0.020565 +2023-10-05 21:11:04,919 - Epoch: [61][ 1230/ 1236] Overall Loss 0.321176 Objective Loss 0.321176 LR 0.001000 Time 0.020602 +2023-10-05 21:11:05,036 - Epoch: [61][ 1236/ 1236] Overall Loss 0.321229 Objective Loss 0.321229 Top1 85.336049 Top5 98.370672 LR 0.001000 Time 0.020597 +2023-10-05 21:11:05,167 - --- validate (epoch=61)----------- +2023-10-05 21:11:05,167 - 29943 samples (256 per mini-batch) +2023-10-05 21:11:05,625 - Epoch: [61][ 10/ 117] Loss 0.374736 Top1 81.093750 Top5 97.500000 +2023-10-05 21:11:05,773 - Epoch: [61][ 20/ 117] Loss 0.380489 Top1 80.781250 Top5 97.363281 +2023-10-05 21:11:05,921 - Epoch: [61][ 30/ 117] Loss 0.358585 Top1 80.664062 Top5 97.343750 +2023-10-05 21:11:06,070 - Epoch: [61][ 40/ 117] Loss 0.367652 Top1 80.478516 Top5 97.343750 +2023-10-05 21:11:06,217 - Epoch: [61][ 50/ 117] Loss 0.362858 Top1 80.859375 Top5 97.437500 +2023-10-05 21:11:06,365 - Epoch: [61][ 60/ 117] Loss 0.364654 Top1 80.657552 Top5 97.382812 +2023-10-05 21:11:06,513 - Epoch: [61][ 70/ 117] Loss 0.365705 Top1 80.602679 Top5 97.310268 +2023-10-05 21:11:06,664 - Epoch: [61][ 80/ 117] Loss 0.363892 Top1 80.864258 Top5 97.324219 +2023-10-05 21:11:06,809 - Epoch: [61][ 90/ 117] Loss 0.364672 Top1 80.894097 Top5 97.352431 +2023-10-05 21:11:06,956 - Epoch: [61][ 100/ 117] Loss 0.364985 Top1 80.929688 Top5 97.328125 +2023-10-05 21:11:07,109 - Epoch: [61][ 110/ 117] Loss 0.366516 Top1 80.855824 Top5 97.286932 +2023-10-05 21:11:07,194 - Epoch: [61][ 117/ 117] Loss 0.366009 Top1 80.780149 Top5 97.258124 +2023-10-05 21:11:07,303 - ==> Top1: 80.780 Top5: 97.258 Loss: 0.366 + +2023-10-05 21:11:07,304 - ==> Confusion: +[[ 941 3 3 0 7 2 0 0 8 61 1 1 2 2 3 2 5 0 1 0 8] + [ 2 1032 0 0 21 16 3 24 4 0 5 2 0 1 1 4 5 0 10 0 1] + [ 10 1 928 23 5 1 26 8 0 4 5 4 6 6 9 3 2 1 6 5 3] + [ 4 1 11 920 1 4 1 0 6 0 12 0 8 6 58 5 0 8 27 3 14] + [ 28 5 0 0 970 3 0 1 1 8 0 0 1 1 12 4 10 1 0 0 5] + [ 5 62 2 0 7 927 0 36 4 0 1 14 2 19 7 1 7 1 4 5 12] + [ 0 7 45 0 0 0 1102 8 0 0 4 4 0 0 1 3 0 2 3 7 5] + [ 5 24 13 0 5 22 2 1056 1 2 2 11 4 3 1 2 0 0 44 7 14] + [ 19 0 0 0 1 0 0 0 974 41 8 1 3 11 17 4 3 1 6 0 0] + [ 111 1 1 0 7 0 1 0 25 928 0 1 0 20 5 9 0 0 1 1 8] + [ 2 2 7 5 1 2 5 5 27 1 941 3 1 18 12 2 5 0 6 0 8] + [ 2 0 0 0 0 9 2 3 0 2 0 943 23 9 0 2 1 18 0 17 4] + [ 1 0 2 2 1 2 1 3 3 0 2 48 964 4 5 3 2 9 2 7 7] + [ 2 2 1 0 5 7 1 0 18 10 5 6 0 1045 6 0 1 0 0 2 8] + [ 16 2 2 3 4 0 0 0 40 7 2 2 1 2 992 0 1 2 16 0 9] + [ 1 2 2 2 3 0 1 1 0 0 0 10 7 3 0 1054 24 12 1 5 6] + [ 1 15 2 0 10 3 0 1 4 0 0 6 1 1 6 9 1092 0 0 4 6] + [ 0 0 0 3 0 0 1 0 1 0 0 5 12 0 5 8 0 998 2 0 3] + [ 2 13 6 17 3 0 0 24 5 0 3 3 4 0 17 0 1 1 960 0 9] + [ 0 2 1 0 1 5 8 15 0 0 2 12 6 2 1 10 12 1 1 1063 10] + [ 222 263 204 63 163 95 44 111 179 87 200 139 383 331 238 61 236 77 213 238 4358]] + +2023-10-05 21:11:07,305 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:11:07,305 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:11:07,311 - + +2023-10-05 21:11:07,311 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:11:08,416 - Epoch: [62][ 10/ 1236] Overall Loss 0.298074 Objective Loss 0.298074 LR 0.001000 Time 0.110492 +2023-10-05 21:11:08,616 - Epoch: [62][ 20/ 1236] Overall Loss 0.284270 Objective Loss 0.284270 LR 0.001000 Time 0.065223 +2023-10-05 21:11:08,821 - Epoch: [62][ 30/ 1236] Overall Loss 0.280884 Objective Loss 0.280884 LR 0.001000 Time 0.050308 +2023-10-05 21:11:09,031 - Epoch: [62][ 40/ 1236] Overall Loss 0.282911 Objective Loss 0.282911 LR 0.001000 Time 0.042966 +2023-10-05 21:11:09,236 - Epoch: [62][ 50/ 1236] Overall Loss 0.293047 Objective Loss 0.293047 LR 0.001000 Time 0.038473 +2023-10-05 21:11:09,446 - Epoch: [62][ 60/ 1236] Overall Loss 0.298676 Objective Loss 0.298676 LR 0.001000 Time 0.035546 +2023-10-05 21:11:09,651 - Epoch: [62][ 70/ 1236] Overall Loss 0.298938 Objective Loss 0.298938 LR 0.001000 Time 0.033399 +2023-10-05 21:11:09,859 - Epoch: [62][ 80/ 1236] Overall Loss 0.301740 Objective Loss 0.301740 LR 0.001000 Time 0.031812 +2023-10-05 21:11:10,059 - Epoch: [62][ 90/ 1236] Overall Loss 0.305241 Objective Loss 0.305241 LR 0.001000 Time 0.030494 +2023-10-05 21:11:10,261 - Epoch: [62][ 100/ 1236] Overall Loss 0.306692 Objective Loss 0.306692 LR 0.001000 Time 0.029461 +2023-10-05 21:11:10,460 - Epoch: [62][ 110/ 1236] Overall Loss 0.308764 Objective Loss 0.308764 LR 0.001000 Time 0.028592 +2023-10-05 21:11:10,662 - Epoch: [62][ 120/ 1236] Overall Loss 0.309036 Objective Loss 0.309036 LR 0.001000 Time 0.027887 +2023-10-05 21:11:10,861 - Epoch: [62][ 130/ 1236] Overall Loss 0.313191 Objective Loss 0.313191 LR 0.001000 Time 0.027275 +2023-10-05 21:11:11,063 - Epoch: [62][ 140/ 1236] Overall Loss 0.312352 Objective Loss 0.312352 LR 0.001000 Time 0.026766 +2023-10-05 21:11:11,263 - Epoch: [62][ 150/ 1236] Overall Loss 0.312542 Objective Loss 0.312542 LR 0.001000 Time 0.026311 +2023-10-05 21:11:11,465 - Epoch: [62][ 160/ 1236] Overall Loss 0.311100 Objective Loss 0.311100 LR 0.001000 Time 0.025926 +2023-10-05 21:11:11,664 - Epoch: [62][ 170/ 1236] Overall Loss 0.311656 Objective Loss 0.311656 LR 0.001000 Time 0.025574 +2023-10-05 21:11:11,866 - Epoch: [62][ 180/ 1236] Overall Loss 0.311786 Objective Loss 0.311786 LR 0.001000 Time 0.025273 +2023-10-05 21:11:12,066 - Epoch: [62][ 190/ 1236] Overall Loss 0.310835 Objective Loss 0.310835 LR 0.001000 Time 0.024991 +2023-10-05 21:11:12,268 - Epoch: [62][ 200/ 1236] Overall Loss 0.312079 Objective Loss 0.312079 LR 0.001000 Time 0.024748 +2023-10-05 21:11:12,468 - Epoch: [62][ 210/ 1236] Overall Loss 0.312743 Objective Loss 0.312743 LR 0.001000 Time 0.024521 +2023-10-05 21:11:12,669 - Epoch: [62][ 220/ 1236] Overall Loss 0.312376 Objective Loss 0.312376 LR 0.001000 Time 0.024322 +2023-10-05 21:11:12,869 - Epoch: [62][ 230/ 1236] Overall Loss 0.314415 Objective Loss 0.314415 LR 0.001000 Time 0.024131 +2023-10-05 21:11:13,069 - Epoch: [62][ 240/ 1236] Overall Loss 0.314087 Objective Loss 0.314087 LR 0.001000 Time 0.023958 +2023-10-05 21:11:13,269 - Epoch: [62][ 250/ 1236] Overall Loss 0.315017 Objective Loss 0.315017 LR 0.001000 Time 0.023798 +2023-10-05 21:11:13,471 - Epoch: [62][ 260/ 1236] Overall Loss 0.314517 Objective Loss 0.314517 LR 0.001000 Time 0.023657 +2023-10-05 21:11:13,671 - Epoch: [62][ 270/ 1236] Overall Loss 0.315489 Objective Loss 0.315489 LR 0.001000 Time 0.023520 +2023-10-05 21:11:13,873 - Epoch: [62][ 280/ 1236] Overall Loss 0.316067 Objective Loss 0.316067 LR 0.001000 Time 0.023399 +2023-10-05 21:11:14,073 - Epoch: [62][ 290/ 1236] Overall Loss 0.316351 Objective Loss 0.316351 LR 0.001000 Time 0.023281 +2023-10-05 21:11:14,274 - Epoch: [62][ 300/ 1236] Overall Loss 0.317961 Objective Loss 0.317961 LR 0.001000 Time 0.023176 +2023-10-05 21:11:14,474 - Epoch: [62][ 310/ 1236] Overall Loss 0.318207 Objective Loss 0.318207 LR 0.001000 Time 0.023072 +2023-10-05 21:11:14,676 - Epoch: [62][ 320/ 1236] Overall Loss 0.317142 Objective Loss 0.317142 LR 0.001000 Time 0.022979 +2023-10-05 21:11:14,876 - Epoch: [62][ 330/ 1236] Overall Loss 0.317958 Objective Loss 0.317958 LR 0.001000 Time 0.022889 +2023-10-05 21:11:15,078 - Epoch: [62][ 340/ 1236] Overall Loss 0.317476 Objective Loss 0.317476 LR 0.001000 Time 0.022809 +2023-10-05 21:11:15,278 - Epoch: [62][ 350/ 1236] Overall Loss 0.317433 Objective Loss 0.317433 LR 0.001000 Time 0.022729 +2023-10-05 21:11:15,480 - Epoch: [62][ 360/ 1236] Overall Loss 0.317332 Objective Loss 0.317332 LR 0.001000 Time 0.022658 +2023-10-05 21:11:15,680 - Epoch: [62][ 370/ 1236] Overall Loss 0.317292 Objective Loss 0.317292 LR 0.001000 Time 0.022585 +2023-10-05 21:11:15,887 - Epoch: [62][ 380/ 1236] Overall Loss 0.317313 Objective Loss 0.317313 LR 0.001000 Time 0.022534 +2023-10-05 21:11:16,093 - Epoch: [62][ 390/ 1236] Overall Loss 0.316878 Objective Loss 0.316878 LR 0.001000 Time 0.022482 +2023-10-05 21:11:16,303 - Epoch: [62][ 400/ 1236] Overall Loss 0.316488 Objective Loss 0.316488 LR 0.001000 Time 0.022444 +2023-10-05 21:11:16,508 - Epoch: [62][ 410/ 1236] Overall Loss 0.316620 Objective Loss 0.316620 LR 0.001000 Time 0.022396 +2023-10-05 21:11:16,717 - Epoch: [62][ 420/ 1236] Overall Loss 0.316747 Objective Loss 0.316747 LR 0.001000 Time 0.022362 +2023-10-05 21:11:16,923 - Epoch: [62][ 430/ 1236] Overall Loss 0.316907 Objective Loss 0.316907 LR 0.001000 Time 0.022318 +2023-10-05 21:11:17,132 - Epoch: [62][ 440/ 1236] Overall Loss 0.316241 Objective Loss 0.316241 LR 0.001000 Time 0.022287 +2023-10-05 21:11:17,338 - Epoch: [62][ 450/ 1236] Overall Loss 0.316167 Objective Loss 0.316167 LR 0.001000 Time 0.022247 +2023-10-05 21:11:17,548 - Epoch: [62][ 460/ 1236] Overall Loss 0.316723 Objective Loss 0.316723 LR 0.001000 Time 0.022219 +2023-10-05 21:11:17,753 - Epoch: [62][ 470/ 1236] Overall Loss 0.315991 Objective Loss 0.315991 LR 0.001000 Time 0.022182 +2023-10-05 21:11:17,963 - Epoch: [62][ 480/ 1236] Overall Loss 0.315567 Objective Loss 0.315567 LR 0.001000 Time 0.022157 +2023-10-05 21:11:18,168 - Epoch: [62][ 490/ 1236] Overall Loss 0.315428 Objective Loss 0.315428 LR 0.001000 Time 0.022122 +2023-10-05 21:11:18,378 - Epoch: [62][ 500/ 1236] Overall Loss 0.314873 Objective Loss 0.314873 LR 0.001000 Time 0.022099 +2023-10-05 21:11:18,583 - Epoch: [62][ 510/ 1236] Overall Loss 0.314730 Objective Loss 0.314730 LR 0.001000 Time 0.022067 +2023-10-05 21:11:18,793 - Epoch: [62][ 520/ 1236] Overall Loss 0.315131 Objective Loss 0.315131 LR 0.001000 Time 0.022046 +2023-10-05 21:11:18,998 - Epoch: [62][ 530/ 1236] Overall Loss 0.314924 Objective Loss 0.314924 LR 0.001000 Time 0.022016 +2023-10-05 21:11:19,207 - Epoch: [62][ 540/ 1236] Overall Loss 0.315062 Objective Loss 0.315062 LR 0.001000 Time 0.021996 +2023-10-05 21:11:19,413 - Epoch: [62][ 550/ 1236] Overall Loss 0.314412 Objective Loss 0.314412 LR 0.001000 Time 0.021969 +2023-10-05 21:11:19,622 - Epoch: [62][ 560/ 1236] Overall Loss 0.315192 Objective Loss 0.315192 LR 0.001000 Time 0.021950 +2023-10-05 21:11:19,828 - Epoch: [62][ 570/ 1236] Overall Loss 0.314921 Objective Loss 0.314921 LR 0.001000 Time 0.021924 +2023-10-05 21:11:20,037 - Epoch: [62][ 580/ 1236] Overall Loss 0.315155 Objective Loss 0.315155 LR 0.001000 Time 0.021908 +2023-10-05 21:11:20,243 - Epoch: [62][ 590/ 1236] Overall Loss 0.315454 Objective Loss 0.315454 LR 0.001000 Time 0.021883 +2023-10-05 21:11:20,452 - Epoch: [62][ 600/ 1236] Overall Loss 0.315328 Objective Loss 0.315328 LR 0.001000 Time 0.021868 +2023-10-05 21:11:20,658 - Epoch: [62][ 610/ 1236] Overall Loss 0.315186 Objective Loss 0.315186 LR 0.001000 Time 0.021845 +2023-10-05 21:11:20,868 - Epoch: [62][ 620/ 1236] Overall Loss 0.315113 Objective Loss 0.315113 LR 0.001000 Time 0.021831 +2023-10-05 21:11:21,073 - Epoch: [62][ 630/ 1236] Overall Loss 0.314913 Objective Loss 0.314913 LR 0.001000 Time 0.021809 +2023-10-05 21:11:21,283 - Epoch: [62][ 640/ 1236] Overall Loss 0.315024 Objective Loss 0.315024 LR 0.001000 Time 0.021796 +2023-10-05 21:11:21,488 - Epoch: [62][ 650/ 1236] Overall Loss 0.314656 Objective Loss 0.314656 LR 0.001000 Time 0.021776 +2023-10-05 21:11:21,698 - Epoch: [62][ 660/ 1236] Overall Loss 0.314983 Objective Loss 0.314983 LR 0.001000 Time 0.021763 +2023-10-05 21:11:21,903 - Epoch: [62][ 670/ 1236] Overall Loss 0.315212 Objective Loss 0.315212 LR 0.001000 Time 0.021744 +2023-10-05 21:11:22,113 - Epoch: [62][ 680/ 1236] Overall Loss 0.315963 Objective Loss 0.315963 LR 0.001000 Time 0.021733 +2023-10-05 21:11:22,318 - Epoch: [62][ 690/ 1236] Overall Loss 0.316527 Objective Loss 0.316527 LR 0.001000 Time 0.021715 +2023-10-05 21:11:22,528 - Epoch: [62][ 700/ 1236] Overall Loss 0.316869 Objective Loss 0.316869 LR 0.001000 Time 0.021704 +2023-10-05 21:11:22,734 - Epoch: [62][ 710/ 1236] Overall Loss 0.316838 Objective Loss 0.316838 LR 0.001000 Time 0.021688 +2023-10-05 21:11:22,944 - Epoch: [62][ 720/ 1236] Overall Loss 0.316707 Objective Loss 0.316707 LR 0.001000 Time 0.021678 +2023-10-05 21:11:23,149 - Epoch: [62][ 730/ 1236] Overall Loss 0.316938 Objective Loss 0.316938 LR 0.001000 Time 0.021661 +2023-10-05 21:11:23,359 - Epoch: [62][ 740/ 1236] Overall Loss 0.316980 Objective Loss 0.316980 LR 0.001000 Time 0.021652 +2023-10-05 21:11:23,564 - Epoch: [62][ 750/ 1236] Overall Loss 0.316569 Objective Loss 0.316569 LR 0.001000 Time 0.021636 +2023-10-05 21:11:23,774 - Epoch: [62][ 760/ 1236] Overall Loss 0.316836 Objective Loss 0.316836 LR 0.001000 Time 0.021627 +2023-10-05 21:11:23,979 - Epoch: [62][ 770/ 1236] Overall Loss 0.317228 Objective Loss 0.317228 LR 0.001000 Time 0.021612 +2023-10-05 21:11:24,189 - Epoch: [62][ 780/ 1236] Overall Loss 0.317294 Objective Loss 0.317294 LR 0.001000 Time 0.021604 +2023-10-05 21:11:24,394 - Epoch: [62][ 790/ 1236] Overall Loss 0.317537 Objective Loss 0.317537 LR 0.001000 Time 0.021589 +2023-10-05 21:11:24,604 - Epoch: [62][ 800/ 1236] Overall Loss 0.317494 Objective Loss 0.317494 LR 0.001000 Time 0.021581 +2023-10-05 21:11:24,809 - Epoch: [62][ 810/ 1236] Overall Loss 0.317844 Objective Loss 0.317844 LR 0.001000 Time 0.021568 +2023-10-05 21:11:25,019 - Epoch: [62][ 820/ 1236] Overall Loss 0.317884 Objective Loss 0.317884 LR 0.001000 Time 0.021560 +2023-10-05 21:11:25,221 - Epoch: [62][ 830/ 1236] Overall Loss 0.317914 Objective Loss 0.317914 LR 0.001000 Time 0.021543 +2023-10-05 21:11:25,424 - Epoch: [62][ 840/ 1236] Overall Loss 0.317529 Objective Loss 0.317529 LR 0.001000 Time 0.021528 +2023-10-05 21:11:25,624 - Epoch: [62][ 850/ 1236] Overall Loss 0.317286 Objective Loss 0.317286 LR 0.001000 Time 0.021510 +2023-10-05 21:11:25,826 - Epoch: [62][ 860/ 1236] Overall Loss 0.317175 Objective Loss 0.317175 LR 0.001000 Time 0.021494 +2023-10-05 21:11:26,025 - Epoch: [62][ 870/ 1236] Overall Loss 0.317014 Objective Loss 0.317014 LR 0.001000 Time 0.021476 +2023-10-05 21:11:26,235 - Epoch: [62][ 880/ 1236] Overall Loss 0.316315 Objective Loss 0.316315 LR 0.001000 Time 0.021470 +2023-10-05 21:11:26,440 - Epoch: [62][ 890/ 1236] Overall Loss 0.316269 Objective Loss 0.316269 LR 0.001000 Time 0.021458 +2023-10-05 21:11:26,650 - Epoch: [62][ 900/ 1236] Overall Loss 0.316078 Objective Loss 0.316078 LR 0.001000 Time 0.021453 +2023-10-05 21:11:26,855 - Epoch: [62][ 910/ 1236] Overall Loss 0.316069 Objective Loss 0.316069 LR 0.001000 Time 0.021442 +2023-10-05 21:11:27,065 - Epoch: [62][ 920/ 1236] Overall Loss 0.316083 Objective Loss 0.316083 LR 0.001000 Time 0.021437 +2023-10-05 21:11:27,270 - Epoch: [62][ 930/ 1236] Overall Loss 0.315767 Objective Loss 0.315767 LR 0.001000 Time 0.021426 +2023-10-05 21:11:27,480 - Epoch: [62][ 940/ 1236] Overall Loss 0.316080 Objective Loss 0.316080 LR 0.001000 Time 0.021422 +2023-10-05 21:11:27,678 - Epoch: [62][ 950/ 1236] Overall Loss 0.316046 Objective Loss 0.316046 LR 0.001000 Time 0.021405 +2023-10-05 21:11:27,878 - Epoch: [62][ 960/ 1236] Overall Loss 0.316019 Objective Loss 0.316019 LR 0.001000 Time 0.021389 +2023-10-05 21:11:28,077 - Epoch: [62][ 970/ 1236] Overall Loss 0.315704 Objective Loss 0.315704 LR 0.001000 Time 0.021374 +2023-10-05 21:11:28,277 - Epoch: [62][ 980/ 1236] Overall Loss 0.315894 Objective Loss 0.315894 LR 0.001000 Time 0.021359 +2023-10-05 21:11:28,475 - Epoch: [62][ 990/ 1236] Overall Loss 0.315771 Objective Loss 0.315771 LR 0.001000 Time 0.021343 +2023-10-05 21:11:28,674 - Epoch: [62][ 1000/ 1236] Overall Loss 0.315784 Objective Loss 0.315784 LR 0.001000 Time 0.021329 +2023-10-05 21:11:28,873 - Epoch: [62][ 1010/ 1236] Overall Loss 0.315723 Objective Loss 0.315723 LR 0.001000 Time 0.021313 +2023-10-05 21:11:29,073 - Epoch: [62][ 1020/ 1236] Overall Loss 0.315479 Objective Loss 0.315479 LR 0.001000 Time 0.021299 +2023-10-05 21:11:29,272 - Epoch: [62][ 1030/ 1236] Overall Loss 0.315557 Objective Loss 0.315557 LR 0.001000 Time 0.021285 +2023-10-05 21:11:29,472 - Epoch: [62][ 1040/ 1236] Overall Loss 0.315368 Objective Loss 0.315368 LR 0.001000 Time 0.021272 +2023-10-05 21:11:29,670 - Epoch: [62][ 1050/ 1236] Overall Loss 0.315336 Objective Loss 0.315336 LR 0.001000 Time 0.021259 +2023-10-05 21:11:29,870 - Epoch: [62][ 1060/ 1236] Overall Loss 0.315383 Objective Loss 0.315383 LR 0.001000 Time 0.021246 +2023-10-05 21:11:30,068 - Epoch: [62][ 1070/ 1236] Overall Loss 0.315389 Objective Loss 0.315389 LR 0.001000 Time 0.021232 +2023-10-05 21:11:30,268 - Epoch: [62][ 1080/ 1236] Overall Loss 0.315526 Objective Loss 0.315526 LR 0.001000 Time 0.021220 +2023-10-05 21:11:30,467 - Epoch: [62][ 1090/ 1236] Overall Loss 0.315651 Objective Loss 0.315651 LR 0.001000 Time 0.021208 +2023-10-05 21:11:30,667 - Epoch: [62][ 1100/ 1236] Overall Loss 0.315655 Objective Loss 0.315655 LR 0.001000 Time 0.021197 +2023-10-05 21:11:30,866 - Epoch: [62][ 1110/ 1236] Overall Loss 0.315892 Objective Loss 0.315892 LR 0.001000 Time 0.021185 +2023-10-05 21:11:31,064 - Epoch: [62][ 1120/ 1236] Overall Loss 0.316256 Objective Loss 0.316256 LR 0.001000 Time 0.021173 +2023-10-05 21:11:31,263 - Epoch: [62][ 1130/ 1236] Overall Loss 0.316557 Objective Loss 0.316557 LR 0.001000 Time 0.021161 +2023-10-05 21:11:31,463 - Epoch: [62][ 1140/ 1236] Overall Loss 0.316414 Objective Loss 0.316414 LR 0.001000 Time 0.021150 +2023-10-05 21:11:31,663 - Epoch: [62][ 1150/ 1236] Overall Loss 0.316340 Objective Loss 0.316340 LR 0.001000 Time 0.021139 +2023-10-05 21:11:31,865 - Epoch: [62][ 1160/ 1236] Overall Loss 0.316682 Objective Loss 0.316682 LR 0.001000 Time 0.021130 +2023-10-05 21:11:32,065 - Epoch: [62][ 1170/ 1236] Overall Loss 0.316752 Objective Loss 0.316752 LR 0.001000 Time 0.021120 +2023-10-05 21:11:32,267 - Epoch: [62][ 1180/ 1236] Overall Loss 0.317046 Objective Loss 0.317046 LR 0.001000 Time 0.021112 +2023-10-05 21:11:32,465 - Epoch: [62][ 1190/ 1236] Overall Loss 0.317149 Objective Loss 0.317149 LR 0.001000 Time 0.021101 +2023-10-05 21:11:32,670 - Epoch: [62][ 1200/ 1236] Overall Loss 0.317524 Objective Loss 0.317524 LR 0.001000 Time 0.021095 +2023-10-05 21:11:32,870 - Epoch: [62][ 1210/ 1236] Overall Loss 0.317618 Objective Loss 0.317618 LR 0.001000 Time 0.021086 +2023-10-05 21:11:33,070 - Epoch: [62][ 1220/ 1236] Overall Loss 0.317802 Objective Loss 0.317802 LR 0.001000 Time 0.021077 +2023-10-05 21:11:33,322 - Epoch: [62][ 1230/ 1236] Overall Loss 0.317878 Objective Loss 0.317878 LR 0.001000 Time 0.021110 +2023-10-05 21:11:33,440 - Epoch: [62][ 1236/ 1236] Overall Loss 0.317968 Objective Loss 0.317968 Top1 83.910387 Top5 97.148676 LR 0.001000 Time 0.021103 +2023-10-05 21:11:33,581 - --- validate (epoch=62)----------- +2023-10-05 21:11:33,582 - 29943 samples (256 per mini-batch) +2023-10-05 21:11:34,029 - Epoch: [62][ 10/ 117] Loss 0.384307 Top1 81.328125 Top5 97.382812 +2023-10-05 21:11:34,176 - Epoch: [62][ 20/ 117] Loss 0.362168 Top1 82.382812 Top5 97.578125 +2023-10-05 21:11:34,322 - Epoch: [62][ 30/ 117] Loss 0.355146 Top1 82.786458 Top5 97.721354 +2023-10-05 21:11:34,467 - Epoch: [62][ 40/ 117] Loss 0.360810 Top1 82.246094 Top5 97.626953 +2023-10-05 21:11:34,612 - Epoch: [62][ 50/ 117] Loss 0.363678 Top1 82.101562 Top5 97.625000 +2023-10-05 21:11:34,758 - Epoch: [62][ 60/ 117] Loss 0.360759 Top1 82.096354 Top5 97.656250 +2023-10-05 21:11:34,903 - Epoch: [62][ 70/ 117] Loss 0.355015 Top1 82.137277 Top5 97.656250 +2023-10-05 21:11:35,047 - Epoch: [62][ 80/ 117] Loss 0.357596 Top1 82.075195 Top5 97.646484 +2023-10-05 21:11:35,194 - Epoch: [62][ 90/ 117] Loss 0.358540 Top1 82.031250 Top5 97.677951 +2023-10-05 21:11:35,339 - Epoch: [62][ 100/ 117] Loss 0.359539 Top1 81.980469 Top5 97.648438 +2023-10-05 21:11:35,491 - Epoch: [62][ 110/ 117] Loss 0.364780 Top1 81.814631 Top5 97.606534 +2023-10-05 21:11:35,576 - Epoch: [62][ 117/ 117] Loss 0.364138 Top1 81.838827 Top5 97.602111 +2023-10-05 21:11:35,696 - ==> Top1: 81.839 Top5: 97.602 Loss: 0.364 + +2023-10-05 21:11:35,696 - ==> Confusion: +[[ 903 3 5 3 9 2 1 0 8 79 1 0 0 2 8 5 6 1 0 1 13] + [ 0 1031 2 0 9 31 2 24 1 0 2 2 0 0 1 2 3 0 16 2 3] + [ 3 1 927 27 3 2 43 9 0 0 4 5 5 1 0 2 1 3 7 5 8] + [ 2 1 12 977 2 6 1 1 4 0 4 0 4 4 25 2 0 9 21 3 11] + [ 21 3 1 0 972 5 0 0 1 7 0 3 1 2 12 7 7 1 0 2 5] + [ 3 32 2 3 2 965 1 27 1 1 4 14 3 23 4 1 2 1 5 5 17] + [ 0 8 29 2 0 3 1106 9 0 0 6 1 1 1 1 5 0 2 1 10 6] + [ 4 15 23 3 5 42 6 1045 3 4 4 2 6 0 1 2 0 0 36 12 5] + [ 25 2 0 1 1 4 0 3 954 28 15 2 3 10 25 6 0 1 6 1 2] + [ 87 2 2 1 8 3 0 1 48 895 1 2 1 36 11 5 1 3 0 3 9] + [ 1 2 13 14 0 0 6 2 9 1 953 1 0 17 6 2 2 2 12 3 7] + [ 0 0 1 0 0 11 0 1 1 0 0 956 28 4 0 2 3 15 1 9 3] + [ 0 1 3 7 1 4 1 2 1 0 5 39 969 0 6 5 1 12 1 6 4] + [ 1 0 1 1 4 5 0 0 18 10 10 8 1 1031 3 2 1 1 0 1 21] + [ 13 1 2 18 7 0 0 0 27 4 2 2 4 2 988 0 1 2 14 0 14] + [ 0 1 1 5 1 0 0 1 0 0 0 12 13 1 0 1054 15 15 1 8 6] + [ 0 17 4 1 5 6 1 1 2 0 0 4 4 1 3 13 1078 1 0 5 15] + [ 0 0 1 1 0 0 4 0 0 0 0 8 22 1 3 5 0 985 1 0 7] + [ 1 8 13 17 2 1 0 29 5 0 2 1 7 0 13 2 0 0 960 0 7] + [ 0 3 5 1 1 2 5 13 0 0 0 13 5 7 0 5 6 2 2 1071 11] + [ 141 227 175 114 87 190 44 112 120 73 194 143 417 313 190 69 103 95 205 208 4685]] + +2023-10-05 21:11:35,697 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:11:35,697 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:11:35,703 - + +2023-10-05 21:11:35,703 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:11:36,684 - Epoch: [63][ 10/ 1236] Overall Loss 0.326026 Objective Loss 0.326026 LR 0.001000 Time 0.097993 +2023-10-05 21:11:36,884 - Epoch: [63][ 20/ 1236] Overall Loss 0.314851 Objective Loss 0.314851 LR 0.001000 Time 0.058989 +2023-10-05 21:11:37,082 - Epoch: [63][ 30/ 1236] Overall Loss 0.325493 Objective Loss 0.325493 LR 0.001000 Time 0.045933 +2023-10-05 21:11:37,282 - Epoch: [63][ 40/ 1236] Overall Loss 0.324558 Objective Loss 0.324558 LR 0.001000 Time 0.039440 +2023-10-05 21:11:37,482 - Epoch: [63][ 50/ 1236] Overall Loss 0.316392 Objective Loss 0.316392 LR 0.001000 Time 0.035536 +2023-10-05 21:11:37,681 - Epoch: [63][ 60/ 1236] Overall Loss 0.317471 Objective Loss 0.317471 LR 0.001000 Time 0.032933 +2023-10-05 21:11:37,881 - Epoch: [63][ 70/ 1236] Overall Loss 0.314202 Objective Loss 0.314202 LR 0.001000 Time 0.031074 +2023-10-05 21:11:38,081 - Epoch: [63][ 80/ 1236] Overall Loss 0.311912 Objective Loss 0.311912 LR 0.001000 Time 0.029683 +2023-10-05 21:11:38,280 - Epoch: [63][ 90/ 1236] Overall Loss 0.312238 Objective Loss 0.312238 LR 0.001000 Time 0.028592 +2023-10-05 21:11:38,480 - Epoch: [63][ 100/ 1236] Overall Loss 0.312519 Objective Loss 0.312519 LR 0.001000 Time 0.027731 +2023-10-05 21:11:38,679 - Epoch: [63][ 110/ 1236] Overall Loss 0.315217 Objective Loss 0.315217 LR 0.001000 Time 0.027016 +2023-10-05 21:11:38,878 - Epoch: [63][ 120/ 1236] Overall Loss 0.314665 Objective Loss 0.314665 LR 0.001000 Time 0.026424 +2023-10-05 21:11:39,077 - Epoch: [63][ 130/ 1236] Overall Loss 0.311738 Objective Loss 0.311738 LR 0.001000 Time 0.025920 +2023-10-05 21:11:39,286 - Epoch: [63][ 140/ 1236] Overall Loss 0.308883 Objective Loss 0.308883 LR 0.001000 Time 0.025560 +2023-10-05 21:11:39,484 - Epoch: [63][ 150/ 1236] Overall Loss 0.310009 Objective Loss 0.310009 LR 0.001000 Time 0.025171 +2023-10-05 21:11:39,682 - Epoch: [63][ 160/ 1236] Overall Loss 0.310388 Objective Loss 0.310388 LR 0.001000 Time 0.024836 +2023-10-05 21:11:39,879 - Epoch: [63][ 170/ 1236] Overall Loss 0.311178 Objective Loss 0.311178 LR 0.001000 Time 0.024532 +2023-10-05 21:11:40,078 - Epoch: [63][ 180/ 1236] Overall Loss 0.310787 Objective Loss 0.310787 LR 0.001000 Time 0.024270 +2023-10-05 21:11:40,275 - Epoch: [63][ 190/ 1236] Overall Loss 0.310808 Objective Loss 0.310808 LR 0.001000 Time 0.024027 +2023-10-05 21:11:40,473 - Epoch: [63][ 200/ 1236] Overall Loss 0.312219 Objective Loss 0.312219 LR 0.001000 Time 0.023815 +2023-10-05 21:11:40,670 - Epoch: [63][ 210/ 1236] Overall Loss 0.310500 Objective Loss 0.310500 LR 0.001000 Time 0.023619 +2023-10-05 21:11:40,869 - Epoch: [63][ 220/ 1236] Overall Loss 0.310604 Objective Loss 0.310604 LR 0.001000 Time 0.023446 +2023-10-05 21:11:41,066 - Epoch: [63][ 230/ 1236] Overall Loss 0.310837 Objective Loss 0.310837 LR 0.001000 Time 0.023283 +2023-10-05 21:11:41,265 - Epoch: [63][ 240/ 1236] Overall Loss 0.312465 Objective Loss 0.312465 LR 0.001000 Time 0.023139 +2023-10-05 21:11:41,461 - Epoch: [63][ 250/ 1236] Overall Loss 0.313226 Objective Loss 0.313226 LR 0.001000 Time 0.022997 +2023-10-05 21:11:41,659 - Epoch: [63][ 260/ 1236] Overall Loss 0.311722 Objective Loss 0.311722 LR 0.001000 Time 0.022871 +2023-10-05 21:11:41,856 - Epoch: [63][ 270/ 1236] Overall Loss 0.312034 Objective Loss 0.312034 LR 0.001000 Time 0.022755 +2023-10-05 21:11:42,053 - Epoch: [63][ 280/ 1236] Overall Loss 0.312610 Objective Loss 0.312610 LR 0.001000 Time 0.022644 +2023-10-05 21:11:42,249 - Epoch: [63][ 290/ 1236] Overall Loss 0.314414 Objective Loss 0.314414 LR 0.001000 Time 0.022537 +2023-10-05 21:11:42,447 - Epoch: [63][ 300/ 1236] Overall Loss 0.315339 Objective Loss 0.315339 LR 0.001000 Time 0.022446 +2023-10-05 21:11:42,644 - Epoch: [63][ 310/ 1236] Overall Loss 0.316365 Objective Loss 0.316365 LR 0.001000 Time 0.022357 +2023-10-05 21:11:42,843 - Epoch: [63][ 320/ 1236] Overall Loss 0.316906 Objective Loss 0.316906 LR 0.001000 Time 0.022278 +2023-10-05 21:11:43,041 - Epoch: [63][ 330/ 1236] Overall Loss 0.315886 Objective Loss 0.315886 LR 0.001000 Time 0.022200 +2023-10-05 21:11:43,239 - Epoch: [63][ 340/ 1236] Overall Loss 0.316027 Objective Loss 0.316027 LR 0.001000 Time 0.022130 +2023-10-05 21:11:43,437 - Epoch: [63][ 350/ 1236] Overall Loss 0.316189 Objective Loss 0.316189 LR 0.001000 Time 0.022062 +2023-10-05 21:11:43,636 - Epoch: [63][ 360/ 1236] Overall Loss 0.315800 Objective Loss 0.315800 LR 0.001000 Time 0.022000 +2023-10-05 21:11:43,833 - Epoch: [63][ 370/ 1236] Overall Loss 0.315387 Objective Loss 0.315387 LR 0.001000 Time 0.021938 +2023-10-05 21:11:44,032 - Epoch: [63][ 380/ 1236] Overall Loss 0.315544 Objective Loss 0.315544 LR 0.001000 Time 0.021883 +2023-10-05 21:11:44,230 - Epoch: [63][ 390/ 1236] Overall Loss 0.315492 Objective Loss 0.315492 LR 0.001000 Time 0.021828 +2023-10-05 21:11:44,428 - Epoch: [63][ 400/ 1236] Overall Loss 0.315856 Objective Loss 0.315856 LR 0.001000 Time 0.021778 +2023-10-05 21:11:44,626 - Epoch: [63][ 410/ 1236] Overall Loss 0.315602 Objective Loss 0.315602 LR 0.001000 Time 0.021729 +2023-10-05 21:11:44,825 - Epoch: [63][ 420/ 1236] Overall Loss 0.315621 Objective Loss 0.315621 LR 0.001000 Time 0.021684 +2023-10-05 21:11:45,023 - Epoch: [63][ 430/ 1236] Overall Loss 0.315017 Objective Loss 0.315017 LR 0.001000 Time 0.021638 +2023-10-05 21:11:45,221 - Epoch: [63][ 440/ 1236] Overall Loss 0.315210 Objective Loss 0.315210 LR 0.001000 Time 0.021598 +2023-10-05 21:11:45,420 - Epoch: [63][ 450/ 1236] Overall Loss 0.314359 Objective Loss 0.314359 LR 0.001000 Time 0.021558 +2023-10-05 21:11:45,618 - Epoch: [63][ 460/ 1236] Overall Loss 0.313531 Objective Loss 0.313531 LR 0.001000 Time 0.021520 +2023-10-05 21:11:45,816 - Epoch: [63][ 470/ 1236] Overall Loss 0.312916 Objective Loss 0.312916 LR 0.001000 Time 0.021481 +2023-10-05 21:11:46,014 - Epoch: [63][ 480/ 1236] Overall Loss 0.312605 Objective Loss 0.312605 LR 0.001000 Time 0.021447 +2023-10-05 21:11:46,212 - Epoch: [63][ 490/ 1236] Overall Loss 0.312085 Objective Loss 0.312085 LR 0.001000 Time 0.021411 +2023-10-05 21:11:46,411 - Epoch: [63][ 500/ 1236] Overall Loss 0.311464 Objective Loss 0.311464 LR 0.001000 Time 0.021380 +2023-10-05 21:11:46,608 - Epoch: [63][ 510/ 1236] Overall Loss 0.311370 Objective Loss 0.311370 LR 0.001000 Time 0.021348 +2023-10-05 21:11:46,807 - Epoch: [63][ 520/ 1236] Overall Loss 0.311189 Objective Loss 0.311189 LR 0.001000 Time 0.021319 +2023-10-05 21:11:47,005 - Epoch: [63][ 530/ 1236] Overall Loss 0.310747 Objective Loss 0.310747 LR 0.001000 Time 0.021290 +2023-10-05 21:11:47,204 - Epoch: [63][ 540/ 1236] Overall Loss 0.310997 Objective Loss 0.310997 LR 0.001000 Time 0.021263 +2023-10-05 21:11:47,402 - Epoch: [63][ 550/ 1236] Overall Loss 0.311108 Objective Loss 0.311108 LR 0.001000 Time 0.021235 +2023-10-05 21:11:47,600 - Epoch: [63][ 560/ 1236] Overall Loss 0.310794 Objective Loss 0.310794 LR 0.001000 Time 0.021210 +2023-10-05 21:11:47,798 - Epoch: [63][ 570/ 1236] Overall Loss 0.311308 Objective Loss 0.311308 LR 0.001000 Time 0.021184 +2023-10-05 21:11:47,997 - Epoch: [63][ 580/ 1236] Overall Loss 0.311583 Objective Loss 0.311583 LR 0.001000 Time 0.021161 +2023-10-05 21:11:48,194 - Epoch: [63][ 590/ 1236] Overall Loss 0.311905 Objective Loss 0.311905 LR 0.001000 Time 0.021137 +2023-10-05 21:11:48,393 - Epoch: [63][ 600/ 1236] Overall Loss 0.311651 Objective Loss 0.311651 LR 0.001000 Time 0.021115 +2023-10-05 21:11:48,591 - Epoch: [63][ 610/ 1236] Overall Loss 0.311940 Objective Loss 0.311940 LR 0.001000 Time 0.021093 +2023-10-05 21:11:48,790 - Epoch: [63][ 620/ 1236] Overall Loss 0.311790 Objective Loss 0.311790 LR 0.001000 Time 0.021072 +2023-10-05 21:11:48,988 - Epoch: [63][ 630/ 1236] Overall Loss 0.311867 Objective Loss 0.311867 LR 0.001000 Time 0.021052 +2023-10-05 21:11:49,186 - Epoch: [63][ 640/ 1236] Overall Loss 0.312047 Objective Loss 0.312047 LR 0.001000 Time 0.021032 +2023-10-05 21:11:49,384 - Epoch: [63][ 650/ 1236] Overall Loss 0.312530 Objective Loss 0.312530 LR 0.001000 Time 0.021013 +2023-10-05 21:11:49,583 - Epoch: [63][ 660/ 1236] Overall Loss 0.312803 Objective Loss 0.312803 LR 0.001000 Time 0.020995 +2023-10-05 21:11:49,780 - Epoch: [63][ 670/ 1236] Overall Loss 0.312959 Objective Loss 0.312959 LR 0.001000 Time 0.020976 +2023-10-05 21:11:49,979 - Epoch: [63][ 680/ 1236] Overall Loss 0.313413 Objective Loss 0.313413 LR 0.001000 Time 0.020960 +2023-10-05 21:11:50,178 - Epoch: [63][ 690/ 1236] Overall Loss 0.313420 Objective Loss 0.313420 LR 0.001000 Time 0.020943 +2023-10-05 21:11:50,376 - Epoch: [63][ 700/ 1236] Overall Loss 0.313589 Objective Loss 0.313589 LR 0.001000 Time 0.020927 +2023-10-05 21:11:50,574 - Epoch: [63][ 710/ 1236] Overall Loss 0.313980 Objective Loss 0.313980 LR 0.001000 Time 0.020910 +2023-10-05 21:11:50,773 - Epoch: [63][ 720/ 1236] Overall Loss 0.313872 Objective Loss 0.313872 LR 0.001000 Time 0.020895 +2023-10-05 21:11:50,971 - Epoch: [63][ 730/ 1236] Overall Loss 0.313688 Objective Loss 0.313688 LR 0.001000 Time 0.020880 +2023-10-05 21:11:51,169 - Epoch: [63][ 740/ 1236] Overall Loss 0.313682 Objective Loss 0.313682 LR 0.001000 Time 0.020865 +2023-10-05 21:11:51,367 - Epoch: [63][ 750/ 1236] Overall Loss 0.314055 Objective Loss 0.314055 LR 0.001000 Time 0.020850 +2023-10-05 21:11:51,566 - Epoch: [63][ 760/ 1236] Overall Loss 0.314308 Objective Loss 0.314308 LR 0.001000 Time 0.020837 +2023-10-05 21:11:51,763 - Epoch: [63][ 770/ 1236] Overall Loss 0.314373 Objective Loss 0.314373 LR 0.001000 Time 0.020823 +2023-10-05 21:11:51,962 - Epoch: [63][ 780/ 1236] Overall Loss 0.314460 Objective Loss 0.314460 LR 0.001000 Time 0.020810 +2023-10-05 21:11:52,160 - Epoch: [63][ 790/ 1236] Overall Loss 0.314740 Objective Loss 0.314740 LR 0.001000 Time 0.020796 +2023-10-05 21:11:52,359 - Epoch: [63][ 800/ 1236] Overall Loss 0.315209 Objective Loss 0.315209 LR 0.001000 Time 0.020785 +2023-10-05 21:11:52,557 - Epoch: [63][ 810/ 1236] Overall Loss 0.315360 Objective Loss 0.315360 LR 0.001000 Time 0.020773 +2023-10-05 21:11:52,755 - Epoch: [63][ 820/ 1236] Overall Loss 0.315260 Objective Loss 0.315260 LR 0.001000 Time 0.020761 +2023-10-05 21:11:52,953 - Epoch: [63][ 830/ 1236] Overall Loss 0.314908 Objective Loss 0.314908 LR 0.001000 Time 0.020748 +2023-10-05 21:11:53,152 - Epoch: [63][ 840/ 1236] Overall Loss 0.315302 Objective Loss 0.315302 LR 0.001000 Time 0.020738 +2023-10-05 21:11:53,350 - Epoch: [63][ 850/ 1236] Overall Loss 0.315808 Objective Loss 0.315808 LR 0.001000 Time 0.020726 +2023-10-05 21:11:53,549 - Epoch: [63][ 860/ 1236] Overall Loss 0.316204 Objective Loss 0.316204 LR 0.001000 Time 0.020716 +2023-10-05 21:11:53,746 - Epoch: [63][ 870/ 1236] Overall Loss 0.316234 Objective Loss 0.316234 LR 0.001000 Time 0.020705 +2023-10-05 21:11:53,945 - Epoch: [63][ 880/ 1236] Overall Loss 0.316508 Objective Loss 0.316508 LR 0.001000 Time 0.020695 +2023-10-05 21:11:54,144 - Epoch: [63][ 890/ 1236] Overall Loss 0.316592 Objective Loss 0.316592 LR 0.001000 Time 0.020685 +2023-10-05 21:11:54,342 - Epoch: [63][ 900/ 1236] Overall Loss 0.316623 Objective Loss 0.316623 LR 0.001000 Time 0.020675 +2023-10-05 21:11:54,540 - Epoch: [63][ 910/ 1236] Overall Loss 0.316634 Objective Loss 0.316634 LR 0.001000 Time 0.020665 +2023-10-05 21:11:54,738 - Epoch: [63][ 920/ 1236] Overall Loss 0.316184 Objective Loss 0.316184 LR 0.001000 Time 0.020656 +2023-10-05 21:11:54,936 - Epoch: [63][ 930/ 1236] Overall Loss 0.316027 Objective Loss 0.316027 LR 0.001000 Time 0.020646 +2023-10-05 21:11:55,135 - Epoch: [63][ 940/ 1236] Overall Loss 0.315910 Objective Loss 0.315910 LR 0.001000 Time 0.020638 +2023-10-05 21:11:55,333 - Epoch: [63][ 950/ 1236] Overall Loss 0.315998 Objective Loss 0.315998 LR 0.001000 Time 0.020628 +2023-10-05 21:11:55,532 - Epoch: [63][ 960/ 1236] Overall Loss 0.316427 Objective Loss 0.316427 LR 0.001000 Time 0.020620 +2023-10-05 21:11:55,730 - Epoch: [63][ 970/ 1236] Overall Loss 0.316435 Objective Loss 0.316435 LR 0.001000 Time 0.020611 +2023-10-05 21:11:55,929 - Epoch: [63][ 980/ 1236] Overall Loss 0.316669 Objective Loss 0.316669 LR 0.001000 Time 0.020604 +2023-10-05 21:11:56,127 - Epoch: [63][ 990/ 1236] Overall Loss 0.316690 Objective Loss 0.316690 LR 0.001000 Time 0.020595 +2023-10-05 21:11:56,325 - Epoch: [63][ 1000/ 1236] Overall Loss 0.316773 Objective Loss 0.316773 LR 0.001000 Time 0.020587 +2023-10-05 21:11:56,523 - Epoch: [63][ 1010/ 1236] Overall Loss 0.316996 Objective Loss 0.316996 LR 0.001000 Time 0.020579 +2023-10-05 21:11:56,722 - Epoch: [63][ 1020/ 1236] Overall Loss 0.317112 Objective Loss 0.317112 LR 0.001000 Time 0.020572 +2023-10-05 21:11:56,920 - Epoch: [63][ 1030/ 1236] Overall Loss 0.317130 Objective Loss 0.317130 LR 0.001000 Time 0.020564 +2023-10-05 21:11:57,119 - Epoch: [63][ 1040/ 1236] Overall Loss 0.317051 Objective Loss 0.317051 LR 0.001000 Time 0.020557 +2023-10-05 21:11:57,317 - Epoch: [63][ 1050/ 1236] Overall Loss 0.317623 Objective Loss 0.317623 LR 0.001000 Time 0.020550 +2023-10-05 21:11:57,515 - Epoch: [63][ 1060/ 1236] Overall Loss 0.317689 Objective Loss 0.317689 LR 0.001000 Time 0.020543 +2023-10-05 21:11:57,713 - Epoch: [63][ 1070/ 1236] Overall Loss 0.317690 Objective Loss 0.317690 LR 0.001000 Time 0.020536 +2023-10-05 21:11:57,912 - Epoch: [63][ 1080/ 1236] Overall Loss 0.318193 Objective Loss 0.318193 LR 0.001000 Time 0.020529 +2023-10-05 21:11:58,110 - Epoch: [63][ 1090/ 1236] Overall Loss 0.317947 Objective Loss 0.317947 LR 0.001000 Time 0.020522 +2023-10-05 21:11:58,309 - Epoch: [63][ 1100/ 1236] Overall Loss 0.318000 Objective Loss 0.318000 LR 0.001000 Time 0.020516 +2023-10-05 21:11:58,507 - Epoch: [63][ 1110/ 1236] Overall Loss 0.318098 Objective Loss 0.318098 LR 0.001000 Time 0.020509 +2023-10-05 21:11:58,705 - Epoch: [63][ 1120/ 1236] Overall Loss 0.317711 Objective Loss 0.317711 LR 0.001000 Time 0.020503 +2023-10-05 21:11:58,904 - Epoch: [63][ 1130/ 1236] Overall Loss 0.317440 Objective Loss 0.317440 LR 0.001000 Time 0.020497 +2023-10-05 21:11:59,102 - Epoch: [63][ 1140/ 1236] Overall Loss 0.317489 Objective Loss 0.317489 LR 0.001000 Time 0.020491 +2023-10-05 21:11:59,300 - Epoch: [63][ 1150/ 1236] Overall Loss 0.317884 Objective Loss 0.317884 LR 0.001000 Time 0.020485 +2023-10-05 21:11:59,499 - Epoch: [63][ 1160/ 1236] Overall Loss 0.317933 Objective Loss 0.317933 LR 0.001000 Time 0.020479 +2023-10-05 21:11:59,697 - Epoch: [63][ 1170/ 1236] Overall Loss 0.318148 Objective Loss 0.318148 LR 0.001000 Time 0.020473 +2023-10-05 21:11:59,896 - Epoch: [63][ 1180/ 1236] Overall Loss 0.318674 Objective Loss 0.318674 LR 0.001000 Time 0.020468 +2023-10-05 21:12:00,094 - Epoch: [63][ 1190/ 1236] Overall Loss 0.318831 Objective Loss 0.318831 LR 0.001000 Time 0.020462 +2023-10-05 21:12:00,293 - Epoch: [63][ 1200/ 1236] Overall Loss 0.318840 Objective Loss 0.318840 LR 0.001000 Time 0.020457 +2023-10-05 21:12:00,490 - Epoch: [63][ 1210/ 1236] Overall Loss 0.318566 Objective Loss 0.318566 LR 0.001000 Time 0.020451 +2023-10-05 21:12:00,689 - Epoch: [63][ 1220/ 1236] Overall Loss 0.318621 Objective Loss 0.318621 LR 0.001000 Time 0.020446 +2023-10-05 21:12:00,938 - Epoch: [63][ 1230/ 1236] Overall Loss 0.318989 Objective Loss 0.318989 LR 0.001000 Time 0.020481 +2023-10-05 21:12:01,054 - Epoch: [63][ 1236/ 1236] Overall Loss 0.319130 Objective Loss 0.319130 Top1 85.336049 Top5 98.370672 LR 0.001000 Time 0.020476 +2023-10-05 21:12:01,173 - --- validate (epoch=63)----------- +2023-10-05 21:12:01,174 - 29943 samples (256 per mini-batch) +2023-10-05 21:12:01,633 - Epoch: [63][ 10/ 117] Loss 0.370043 Top1 80.898438 Top5 97.109375 +2023-10-05 21:12:01,792 - Epoch: [63][ 20/ 117] Loss 0.389549 Top1 80.429688 Top5 96.972656 +2023-10-05 21:12:01,951 - Epoch: [63][ 30/ 117] Loss 0.388206 Top1 80.364583 Top5 97.148438 +2023-10-05 21:12:02,108 - Epoch: [63][ 40/ 117] Loss 0.382572 Top1 80.458984 Top5 97.119141 +2023-10-05 21:12:02,267 - Epoch: [63][ 50/ 117] Loss 0.373494 Top1 80.632812 Top5 97.203125 +2023-10-05 21:12:02,424 - Epoch: [63][ 60/ 117] Loss 0.367432 Top1 80.748698 Top5 97.213542 +2023-10-05 21:12:02,581 - Epoch: [63][ 70/ 117] Loss 0.367562 Top1 80.876116 Top5 97.260045 +2023-10-05 21:12:02,738 - Epoch: [63][ 80/ 117] Loss 0.366986 Top1 80.927734 Top5 97.250977 +2023-10-05 21:12:02,901 - Epoch: [63][ 90/ 117] Loss 0.364573 Top1 80.985243 Top5 97.278646 +2023-10-05 21:12:03,058 - Epoch: [63][ 100/ 117] Loss 0.361586 Top1 81.109375 Top5 97.285156 +2023-10-05 21:12:03,219 - Epoch: [63][ 110/ 117] Loss 0.361584 Top1 81.136364 Top5 97.276278 +2023-10-05 21:12:03,304 - Epoch: [63][ 117/ 117] Loss 0.362357 Top1 81.080720 Top5 97.271482 +2023-10-05 21:12:03,432 - ==> Top1: 81.081 Top5: 97.271 Loss: 0.362 + +2023-10-05 21:12:03,433 - ==> Confusion: +[[ 940 1 2 2 16 3 0 1 5 52 1 0 2 2 1 4 2 5 1 0 10] + [ 1 1042 2 0 9 19 2 33 2 0 2 4 0 0 0 1 1 0 9 2 2] + [ 9 2 932 11 3 1 37 15 0 1 4 6 5 1 2 4 1 2 8 5 7] + [ 2 3 16 960 3 8 3 2 3 0 7 0 2 3 19 4 0 7 29 4 14] + [ 26 10 0 1 969 6 0 2 0 6 0 2 1 0 6 3 10 8 0 0 0] + [ 7 48 0 0 6 957 1 34 3 1 5 6 4 16 5 1 3 2 2 6 9] + [ 1 6 30 0 0 1 1116 10 0 0 3 1 2 0 0 5 1 5 2 5 3] + [ 2 20 14 0 1 29 6 1072 0 5 2 13 2 2 0 2 3 2 30 8 5] + [ 30 6 0 0 0 1 0 2 945 50 8 3 1 9 14 2 1 4 6 1 6] + [ 152 1 5 0 13 3 1 0 24 881 0 2 0 19 0 1 1 4 0 2 10] + [ 2 4 10 3 2 0 2 10 18 1 961 4 0 12 2 0 2 3 7 0 10] + [ 2 0 0 0 0 13 0 1 0 0 0 947 36 4 0 3 2 15 0 8 4] + [ 1 0 0 7 0 2 0 5 0 0 2 46 950 2 0 4 0 27 2 5 15] + [ 5 1 1 1 2 10 0 0 20 15 7 8 2 1019 3 7 7 4 0 0 7] + [ 17 6 1 13 22 0 0 0 30 4 0 1 1 2 961 1 4 5 21 0 12] + [ 0 2 1 0 3 1 1 0 0 0 0 8 7 1 0 1068 9 21 2 5 5] + [ 2 18 2 0 6 4 0 1 4 0 0 8 3 0 1 14 1079 1 0 9 9] + [ 0 1 0 1 0 0 0 0 0 0 0 5 11 1 1 7 0 1005 1 1 4] + [ 0 13 7 16 0 3 0 34 3 0 1 2 1 0 4 0 1 1 970 6 6] + [ 0 6 4 0 2 7 3 19 0 0 2 18 11 2 0 4 8 1 0 1061 4] + [ 185 311 172 82 137 173 78 185 120 75 219 152 381 298 97 69 177 123 174 254 4443]] + +2023-10-05 21:12:03,434 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:12:03,435 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:12:03,440 - + +2023-10-05 21:12:03,440 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:12:04,425 - Epoch: [64][ 10/ 1236] Overall Loss 0.319937 Objective Loss 0.319937 LR 0.001000 Time 0.098413 +2023-10-05 21:12:04,625 - Epoch: [64][ 20/ 1236] Overall Loss 0.337419 Objective Loss 0.337419 LR 0.001000 Time 0.059163 +2023-10-05 21:12:04,823 - Epoch: [64][ 30/ 1236] Overall Loss 0.343818 Objective Loss 0.343818 LR 0.001000 Time 0.046054 +2023-10-05 21:12:05,024 - Epoch: [64][ 40/ 1236] Overall Loss 0.333618 Objective Loss 0.333618 LR 0.001000 Time 0.039541 +2023-10-05 21:12:05,222 - Epoch: [64][ 50/ 1236] Overall Loss 0.332901 Objective Loss 0.332901 LR 0.001000 Time 0.035593 +2023-10-05 21:12:05,422 - Epoch: [64][ 60/ 1236] Overall Loss 0.324568 Objective Loss 0.324568 LR 0.001000 Time 0.032986 +2023-10-05 21:12:05,620 - Epoch: [64][ 70/ 1236] Overall Loss 0.323815 Objective Loss 0.323815 LR 0.001000 Time 0.031105 +2023-10-05 21:12:05,820 - Epoch: [64][ 80/ 1236] Overall Loss 0.324259 Objective Loss 0.324259 LR 0.001000 Time 0.029715 +2023-10-05 21:12:06,019 - Epoch: [64][ 90/ 1236] Overall Loss 0.322915 Objective Loss 0.322915 LR 0.001000 Time 0.028615 +2023-10-05 21:12:06,219 - Epoch: [64][ 100/ 1236] Overall Loss 0.322394 Objective Loss 0.322394 LR 0.001000 Time 0.027748 +2023-10-05 21:12:06,417 - Epoch: [64][ 110/ 1236] Overall Loss 0.320715 Objective Loss 0.320715 LR 0.001000 Time 0.027025 +2023-10-05 21:12:06,617 - Epoch: [64][ 120/ 1236] Overall Loss 0.320442 Objective Loss 0.320442 LR 0.001000 Time 0.026434 +2023-10-05 21:12:06,815 - Epoch: [64][ 130/ 1236] Overall Loss 0.323962 Objective Loss 0.323962 LR 0.001000 Time 0.025926 +2023-10-05 21:12:07,015 - Epoch: [64][ 140/ 1236] Overall Loss 0.325509 Objective Loss 0.325509 LR 0.001000 Time 0.025499 +2023-10-05 21:12:07,213 - Epoch: [64][ 150/ 1236] Overall Loss 0.324887 Objective Loss 0.324887 LR 0.001000 Time 0.025119 +2023-10-05 21:12:07,413 - Epoch: [64][ 160/ 1236] Overall Loss 0.323238 Objective Loss 0.323238 LR 0.001000 Time 0.024796 +2023-10-05 21:12:07,612 - Epoch: [64][ 170/ 1236] Overall Loss 0.321924 Objective Loss 0.321924 LR 0.001000 Time 0.024504 +2023-10-05 21:12:07,812 - Epoch: [64][ 180/ 1236] Overall Loss 0.321036 Objective Loss 0.321036 LR 0.001000 Time 0.024250 +2023-10-05 21:12:08,010 - Epoch: [64][ 190/ 1236] Overall Loss 0.319588 Objective Loss 0.319588 LR 0.001000 Time 0.024016 +2023-10-05 21:12:08,210 - Epoch: [64][ 200/ 1236] Overall Loss 0.319995 Objective Loss 0.319995 LR 0.001000 Time 0.023811 +2023-10-05 21:12:08,408 - Epoch: [64][ 210/ 1236] Overall Loss 0.319569 Objective Loss 0.319569 LR 0.001000 Time 0.023621 +2023-10-05 21:12:08,608 - Epoch: [64][ 220/ 1236] Overall Loss 0.319568 Objective Loss 0.319568 LR 0.001000 Time 0.023455 +2023-10-05 21:12:08,807 - Epoch: [64][ 230/ 1236] Overall Loss 0.319496 Objective Loss 0.319496 LR 0.001000 Time 0.023298 +2023-10-05 21:12:09,007 - Epoch: [64][ 240/ 1236] Overall Loss 0.318826 Objective Loss 0.318826 LR 0.001000 Time 0.023158 +2023-10-05 21:12:09,205 - Epoch: [64][ 250/ 1236] Overall Loss 0.318197 Objective Loss 0.318197 LR 0.001000 Time 0.023025 +2023-10-05 21:12:09,406 - Epoch: [64][ 260/ 1236] Overall Loss 0.317365 Objective Loss 0.317365 LR 0.001000 Time 0.022908 +2023-10-05 21:12:09,604 - Epoch: [64][ 270/ 1236] Overall Loss 0.317951 Objective Loss 0.317951 LR 0.001000 Time 0.022794 +2023-10-05 21:12:09,804 - Epoch: [64][ 280/ 1236] Overall Loss 0.317368 Objective Loss 0.317368 LR 0.001000 Time 0.022692 +2023-10-05 21:12:10,004 - Epoch: [64][ 290/ 1236] Overall Loss 0.317918 Objective Loss 0.317918 LR 0.001000 Time 0.022598 +2023-10-05 21:12:10,204 - Epoch: [64][ 300/ 1236] Overall Loss 0.318234 Objective Loss 0.318234 LR 0.001000 Time 0.022510 +2023-10-05 21:12:10,403 - Epoch: [64][ 310/ 1236] Overall Loss 0.317566 Objective Loss 0.317566 LR 0.001000 Time 0.022425 +2023-10-05 21:12:10,603 - Epoch: [64][ 320/ 1236] Overall Loss 0.318061 Objective Loss 0.318061 LR 0.001000 Time 0.022349 +2023-10-05 21:12:10,802 - Epoch: [64][ 330/ 1236] Overall Loss 0.317974 Objective Loss 0.317974 LR 0.001000 Time 0.022273 +2023-10-05 21:12:11,002 - Epoch: [64][ 340/ 1236] Overall Loss 0.318511 Objective Loss 0.318511 LR 0.001000 Time 0.022205 +2023-10-05 21:12:11,201 - Epoch: [64][ 350/ 1236] Overall Loss 0.317647 Objective Loss 0.317647 LR 0.001000 Time 0.022138 +2023-10-05 21:12:11,401 - Epoch: [64][ 360/ 1236] Overall Loss 0.318041 Objective Loss 0.318041 LR 0.001000 Time 0.022078 +2023-10-05 21:12:11,600 - Epoch: [64][ 370/ 1236] Overall Loss 0.318009 Objective Loss 0.318009 LR 0.001000 Time 0.022017 +2023-10-05 21:12:11,800 - Epoch: [64][ 380/ 1236] Overall Loss 0.318546 Objective Loss 0.318546 LR 0.001000 Time 0.021964 +2023-10-05 21:12:11,999 - Epoch: [64][ 390/ 1236] Overall Loss 0.318123 Objective Loss 0.318123 LR 0.001000 Time 0.021910 +2023-10-05 21:12:12,199 - Epoch: [64][ 400/ 1236] Overall Loss 0.317963 Objective Loss 0.317963 LR 0.001000 Time 0.021861 +2023-10-05 21:12:12,398 - Epoch: [64][ 410/ 1236] Overall Loss 0.318524 Objective Loss 0.318524 LR 0.001000 Time 0.021812 +2023-10-05 21:12:12,598 - Epoch: [64][ 420/ 1236] Overall Loss 0.319617 Objective Loss 0.319617 LR 0.001000 Time 0.021769 +2023-10-05 21:12:12,797 - Epoch: [64][ 430/ 1236] Overall Loss 0.318697 Objective Loss 0.318697 LR 0.001000 Time 0.021724 +2023-10-05 21:12:12,997 - Epoch: [64][ 440/ 1236] Overall Loss 0.319389 Objective Loss 0.319389 LR 0.001000 Time 0.021683 +2023-10-05 21:12:13,196 - Epoch: [64][ 450/ 1236] Overall Loss 0.319091 Objective Loss 0.319091 LR 0.001000 Time 0.021643 +2023-10-05 21:12:13,396 - Epoch: [64][ 460/ 1236] Overall Loss 0.318215 Objective Loss 0.318215 LR 0.001000 Time 0.021607 +2023-10-05 21:12:13,595 - Epoch: [64][ 470/ 1236] Overall Loss 0.317670 Objective Loss 0.317670 LR 0.001000 Time 0.021571 +2023-10-05 21:12:13,795 - Epoch: [64][ 480/ 1236] Overall Loss 0.317870 Objective Loss 0.317870 LR 0.001000 Time 0.021537 +2023-10-05 21:12:13,994 - Epoch: [64][ 490/ 1236] Overall Loss 0.318158 Objective Loss 0.318158 LR 0.001000 Time 0.021503 +2023-10-05 21:12:14,195 - Epoch: [64][ 500/ 1236] Overall Loss 0.317893 Objective Loss 0.317893 LR 0.001000 Time 0.021473 +2023-10-05 21:12:14,393 - Epoch: [64][ 510/ 1236] Overall Loss 0.318118 Objective Loss 0.318118 LR 0.001000 Time 0.021441 +2023-10-05 21:12:14,593 - Epoch: [64][ 520/ 1236] Overall Loss 0.318585 Objective Loss 0.318585 LR 0.001000 Time 0.021412 +2023-10-05 21:12:14,792 - Epoch: [64][ 530/ 1236] Overall Loss 0.318471 Objective Loss 0.318471 LR 0.001000 Time 0.021383 +2023-10-05 21:12:14,993 - Epoch: [64][ 540/ 1236] Overall Loss 0.318923 Objective Loss 0.318923 LR 0.001000 Time 0.021357 +2023-10-05 21:12:15,192 - Epoch: [64][ 550/ 1236] Overall Loss 0.319193 Objective Loss 0.319193 LR 0.001000 Time 0.021331 +2023-10-05 21:12:15,392 - Epoch: [64][ 560/ 1236] Overall Loss 0.318939 Objective Loss 0.318939 LR 0.001000 Time 0.021307 +2023-10-05 21:12:15,591 - Epoch: [64][ 570/ 1236] Overall Loss 0.318610 Objective Loss 0.318610 LR 0.001000 Time 0.021282 +2023-10-05 21:12:15,791 - Epoch: [64][ 580/ 1236] Overall Loss 0.318905 Objective Loss 0.318905 LR 0.001000 Time 0.021259 +2023-10-05 21:12:15,990 - Epoch: [64][ 590/ 1236] Overall Loss 0.318896 Objective Loss 0.318896 LR 0.001000 Time 0.021235 +2023-10-05 21:12:16,191 - Epoch: [64][ 600/ 1236] Overall Loss 0.318805 Objective Loss 0.318805 LR 0.001000 Time 0.021215 +2023-10-05 21:12:16,390 - Epoch: [64][ 610/ 1236] Overall Loss 0.319197 Objective Loss 0.319197 LR 0.001000 Time 0.021193 +2023-10-05 21:12:16,590 - Epoch: [64][ 620/ 1236] Overall Loss 0.319665 Objective Loss 0.319665 LR 0.001000 Time 0.021173 +2023-10-05 21:12:16,789 - Epoch: [64][ 630/ 1236] Overall Loss 0.319222 Objective Loss 0.319222 LR 0.001000 Time 0.021153 +2023-10-05 21:12:16,989 - Epoch: [64][ 640/ 1236] Overall Loss 0.319421 Objective Loss 0.319421 LR 0.001000 Time 0.021135 +2023-10-05 21:12:17,188 - Epoch: [64][ 650/ 1236] Overall Loss 0.319283 Objective Loss 0.319283 LR 0.001000 Time 0.021115 +2023-10-05 21:12:17,389 - Epoch: [64][ 660/ 1236] Overall Loss 0.320159 Objective Loss 0.320159 LR 0.001000 Time 0.021098 +2023-10-05 21:12:17,588 - Epoch: [64][ 670/ 1236] Overall Loss 0.320400 Objective Loss 0.320400 LR 0.001000 Time 0.021080 +2023-10-05 21:12:17,788 - Epoch: [64][ 680/ 1236] Overall Loss 0.320336 Objective Loss 0.320336 LR 0.001000 Time 0.021063 +2023-10-05 21:12:17,987 - Epoch: [64][ 690/ 1236] Overall Loss 0.320363 Objective Loss 0.320363 LR 0.001000 Time 0.021046 +2023-10-05 21:12:18,188 - Epoch: [64][ 700/ 1236] Overall Loss 0.320604 Objective Loss 0.320604 LR 0.001000 Time 0.021032 +2023-10-05 21:12:18,387 - Epoch: [64][ 710/ 1236] Overall Loss 0.320519 Objective Loss 0.320519 LR 0.001000 Time 0.021016 +2023-10-05 21:12:18,587 - Epoch: [64][ 720/ 1236] Overall Loss 0.320638 Objective Loss 0.320638 LR 0.001000 Time 0.021002 +2023-10-05 21:12:18,786 - Epoch: [64][ 730/ 1236] Overall Loss 0.320636 Objective Loss 0.320636 LR 0.001000 Time 0.020986 +2023-10-05 21:12:18,986 - Epoch: [64][ 740/ 1236] Overall Loss 0.320128 Objective Loss 0.320128 LR 0.001000 Time 0.020972 +2023-10-05 21:12:19,185 - Epoch: [64][ 750/ 1236] Overall Loss 0.319877 Objective Loss 0.319877 LR 0.001000 Time 0.020957 +2023-10-05 21:12:19,385 - Epoch: [64][ 760/ 1236] Overall Loss 0.319764 Objective Loss 0.319764 LR 0.001000 Time 0.020944 +2023-10-05 21:12:19,584 - Epoch: [64][ 770/ 1236] Overall Loss 0.319706 Objective Loss 0.319706 LR 0.001000 Time 0.020930 +2023-10-05 21:12:19,784 - Epoch: [64][ 780/ 1236] Overall Loss 0.319590 Objective Loss 0.319590 LR 0.001000 Time 0.020918 +2023-10-05 21:12:19,983 - Epoch: [64][ 790/ 1236] Overall Loss 0.319253 Objective Loss 0.319253 LR 0.001000 Time 0.020905 +2023-10-05 21:12:20,184 - Epoch: [64][ 800/ 1236] Overall Loss 0.318831 Objective Loss 0.318831 LR 0.001000 Time 0.020893 +2023-10-05 21:12:20,383 - Epoch: [64][ 810/ 1236] Overall Loss 0.318857 Objective Loss 0.318857 LR 0.001000 Time 0.020881 +2023-10-05 21:12:20,584 - Epoch: [64][ 820/ 1236] Overall Loss 0.319477 Objective Loss 0.319477 LR 0.001000 Time 0.020871 +2023-10-05 21:12:20,782 - Epoch: [64][ 830/ 1236] Overall Loss 0.319584 Objective Loss 0.319584 LR 0.001000 Time 0.020858 +2023-10-05 21:12:20,983 - Epoch: [64][ 840/ 1236] Overall Loss 0.319407 Objective Loss 0.319407 LR 0.001000 Time 0.020848 +2023-10-05 21:12:21,182 - Epoch: [64][ 850/ 1236] Overall Loss 0.319371 Objective Loss 0.319371 LR 0.001000 Time 0.020836 +2023-10-05 21:12:21,382 - Epoch: [64][ 860/ 1236] Overall Loss 0.318913 Objective Loss 0.318913 LR 0.001000 Time 0.020826 +2023-10-05 21:12:21,581 - Epoch: [64][ 870/ 1236] Overall Loss 0.318846 Objective Loss 0.318846 LR 0.001000 Time 0.020815 +2023-10-05 21:12:21,781 - Epoch: [64][ 880/ 1236] Overall Loss 0.318904 Objective Loss 0.318904 LR 0.001000 Time 0.020806 +2023-10-05 21:12:21,980 - Epoch: [64][ 890/ 1236] Overall Loss 0.318797 Objective Loss 0.318797 LR 0.001000 Time 0.020795 +2023-10-05 21:12:22,181 - Epoch: [64][ 900/ 1236] Overall Loss 0.318522 Objective Loss 0.318522 LR 0.001000 Time 0.020787 +2023-10-05 21:12:22,379 - Epoch: [64][ 910/ 1236] Overall Loss 0.318498 Objective Loss 0.318498 LR 0.001000 Time 0.020776 +2023-10-05 21:12:22,580 - Epoch: [64][ 920/ 1236] Overall Loss 0.318288 Objective Loss 0.318288 LR 0.001000 Time 0.020768 +2023-10-05 21:12:22,779 - Epoch: [64][ 930/ 1236] Overall Loss 0.318435 Objective Loss 0.318435 LR 0.001000 Time 0.020759 +2023-10-05 21:12:22,980 - Epoch: [64][ 940/ 1236] Overall Loss 0.318395 Objective Loss 0.318395 LR 0.001000 Time 0.020751 +2023-10-05 21:12:23,178 - Epoch: [64][ 950/ 1236] Overall Loss 0.318417 Objective Loss 0.318417 LR 0.001000 Time 0.020741 +2023-10-05 21:12:23,379 - Epoch: [64][ 960/ 1236] Overall Loss 0.318201 Objective Loss 0.318201 LR 0.001000 Time 0.020733 +2023-10-05 21:12:23,577 - Epoch: [64][ 970/ 1236] Overall Loss 0.318236 Objective Loss 0.318236 LR 0.001000 Time 0.020724 +2023-10-05 21:12:23,777 - Epoch: [64][ 980/ 1236] Overall Loss 0.317807 Objective Loss 0.317807 LR 0.001000 Time 0.020716 +2023-10-05 21:12:23,976 - Epoch: [64][ 990/ 1236] Overall Loss 0.317711 Objective Loss 0.317711 LR 0.001000 Time 0.020707 +2023-10-05 21:12:24,177 - Epoch: [64][ 1000/ 1236] Overall Loss 0.317802 Objective Loss 0.317802 LR 0.001000 Time 0.020700 +2023-10-05 21:12:24,376 - Epoch: [64][ 1010/ 1236] Overall Loss 0.317942 Objective Loss 0.317942 LR 0.001000 Time 0.020692 +2023-10-05 21:12:24,576 - Epoch: [64][ 1020/ 1236] Overall Loss 0.317924 Objective Loss 0.317924 LR 0.001000 Time 0.020686 +2023-10-05 21:12:24,775 - Epoch: [64][ 1030/ 1236] Overall Loss 0.318200 Objective Loss 0.318200 LR 0.001000 Time 0.020677 +2023-10-05 21:12:24,975 - Epoch: [64][ 1040/ 1236] Overall Loss 0.318110 Objective Loss 0.318110 LR 0.001000 Time 0.020671 +2023-10-05 21:12:25,174 - Epoch: [64][ 1050/ 1236] Overall Loss 0.318160 Objective Loss 0.318160 LR 0.001000 Time 0.020663 +2023-10-05 21:12:25,375 - Epoch: [64][ 1060/ 1236] Overall Loss 0.318085 Objective Loss 0.318085 LR 0.001000 Time 0.020657 +2023-10-05 21:12:25,573 - Epoch: [64][ 1070/ 1236] Overall Loss 0.318313 Objective Loss 0.318313 LR 0.001000 Time 0.020649 +2023-10-05 21:12:25,774 - Epoch: [64][ 1080/ 1236] Overall Loss 0.318655 Objective Loss 0.318655 LR 0.001000 Time 0.020643 +2023-10-05 21:12:25,972 - Epoch: [64][ 1090/ 1236] Overall Loss 0.318609 Objective Loss 0.318609 LR 0.001000 Time 0.020636 +2023-10-05 21:12:26,173 - Epoch: [64][ 1100/ 1236] Overall Loss 0.318545 Objective Loss 0.318545 LR 0.001000 Time 0.020630 +2023-10-05 21:12:26,372 - Epoch: [64][ 1110/ 1236] Overall Loss 0.318977 Objective Loss 0.318977 LR 0.001000 Time 0.020623 +2023-10-05 21:12:26,572 - Epoch: [64][ 1120/ 1236] Overall Loss 0.319183 Objective Loss 0.319183 LR 0.001000 Time 0.020617 +2023-10-05 21:12:26,771 - Epoch: [64][ 1130/ 1236] Overall Loss 0.319264 Objective Loss 0.319264 LR 0.001000 Time 0.020611 +2023-10-05 21:12:26,972 - Epoch: [64][ 1140/ 1236] Overall Loss 0.319576 Objective Loss 0.319576 LR 0.001000 Time 0.020606 +2023-10-05 21:12:27,171 - Epoch: [64][ 1150/ 1236] Overall Loss 0.319635 Objective Loss 0.319635 LR 0.001000 Time 0.020599 +2023-10-05 21:12:27,371 - Epoch: [64][ 1160/ 1236] Overall Loss 0.319390 Objective Loss 0.319390 LR 0.001000 Time 0.020594 +2023-10-05 21:12:27,570 - Epoch: [64][ 1170/ 1236] Overall Loss 0.319721 Objective Loss 0.319721 LR 0.001000 Time 0.020588 +2023-10-05 21:12:27,771 - Epoch: [64][ 1180/ 1236] Overall Loss 0.319721 Objective Loss 0.319721 LR 0.001000 Time 0.020583 +2023-10-05 21:12:27,970 - Epoch: [64][ 1190/ 1236] Overall Loss 0.319912 Objective Loss 0.319912 LR 0.001000 Time 0.020577 +2023-10-05 21:12:28,170 - Epoch: [64][ 1200/ 1236] Overall Loss 0.319850 Objective Loss 0.319850 LR 0.001000 Time 0.020572 +2023-10-05 21:12:28,369 - Epoch: [64][ 1210/ 1236] Overall Loss 0.319855 Objective Loss 0.319855 LR 0.001000 Time 0.020566 +2023-10-05 21:12:28,570 - Epoch: [64][ 1220/ 1236] Overall Loss 0.320181 Objective Loss 0.320181 LR 0.001000 Time 0.020562 +2023-10-05 21:12:28,824 - Epoch: [64][ 1230/ 1236] Overall Loss 0.320061 Objective Loss 0.320061 LR 0.001000 Time 0.020601 +2023-10-05 21:12:28,941 - Epoch: [64][ 1236/ 1236] Overall Loss 0.320318 Objective Loss 0.320318 Top1 82.892057 Top5 97.963340 LR 0.001000 Time 0.020596 +2023-10-05 21:12:29,064 - --- validate (epoch=64)----------- +2023-10-05 21:12:29,064 - 29943 samples (256 per mini-batch) +2023-10-05 21:12:29,528 - Epoch: [64][ 10/ 117] Loss 0.361996 Top1 80.546875 Top5 97.343750 +2023-10-05 21:12:29,686 - Epoch: [64][ 20/ 117] Loss 0.372091 Top1 79.941406 Top5 97.460938 +2023-10-05 21:12:29,845 - Epoch: [64][ 30/ 117] Loss 0.361877 Top1 80.611979 Top5 97.239583 +2023-10-05 21:12:30,002 - Epoch: [64][ 40/ 117] Loss 0.357595 Top1 80.478516 Top5 97.158203 +2023-10-05 21:12:30,160 - Epoch: [64][ 50/ 117] Loss 0.357558 Top1 80.546875 Top5 97.117188 +2023-10-05 21:12:30,310 - Epoch: [64][ 60/ 117] Loss 0.358446 Top1 80.690104 Top5 97.148438 +2023-10-05 21:12:30,458 - Epoch: [64][ 70/ 117] Loss 0.359365 Top1 80.597098 Top5 97.154018 +2023-10-05 21:12:30,605 - Epoch: [64][ 80/ 117] Loss 0.360827 Top1 80.537109 Top5 97.099609 +2023-10-05 21:12:30,751 - Epoch: [64][ 90/ 117] Loss 0.358394 Top1 80.625000 Top5 97.131076 +2023-10-05 21:12:30,898 - Epoch: [64][ 100/ 117] Loss 0.364022 Top1 80.398438 Top5 97.089844 +2023-10-05 21:12:31,051 - Epoch: [64][ 110/ 117] Loss 0.362975 Top1 80.362216 Top5 97.127131 +2023-10-05 21:12:31,136 - Epoch: [64][ 117/ 117] Loss 0.363740 Top1 80.345991 Top5 97.107838 +2023-10-05 21:12:31,274 - ==> Top1: 80.346 Top5: 97.108 Loss: 0.364 + +2023-10-05 21:12:31,274 - ==> Confusion: +[[ 944 4 7 0 3 4 0 0 6 58 2 0 0 1 7 2 3 1 0 0 8] + [ 3 1034 4 0 3 20 1 31 4 0 2 1 0 1 1 4 5 0 10 1 6] + [ 13 1 957 12 3 0 14 12 0 1 4 2 10 4 3 5 2 0 8 2 3] + [ 6 2 19 951 1 5 1 1 2 1 10 0 5 4 34 3 2 4 27 1 10] + [ 51 11 0 0 932 9 0 0 0 5 1 0 0 1 20 5 11 1 0 0 3] + [ 9 48 2 3 1 945 1 49 2 0 4 5 1 15 6 1 5 2 2 9 6] + [ 0 12 44 0 0 1 1096 11 0 0 3 2 1 0 1 4 0 2 4 9 1] + [ 8 12 13 0 1 22 6 1050 2 6 2 8 4 3 0 1 1 0 66 5 8] + [ 26 4 2 0 0 2 0 2 958 40 13 1 0 15 23 2 0 0 1 0 0] + [ 130 1 1 0 4 7 1 0 36 887 0 1 0 27 8 4 2 0 0 1 9] + [ 5 5 16 9 2 3 3 3 17 4 941 1 0 20 6 1 2 0 7 0 8] + [ 1 1 4 0 1 10 0 1 0 1 0 959 24 6 0 3 0 14 0 6 4] + [ 4 1 2 4 0 5 0 1 5 0 4 48 955 0 2 10 2 13 1 4 7] + [ 2 0 3 2 2 12 0 1 12 17 5 6 3 1032 6 5 2 1 0 3 5] + [ 20 1 1 11 1 0 0 0 30 2 2 0 3 0 1006 1 2 3 10 0 8] + [ 2 2 2 1 4 0 3 0 0 0 0 6 9 0 0 1066 17 10 2 7 3] + [ 2 21 1 1 7 2 0 0 1 0 1 2 2 0 3 13 1095 0 1 4 5] + [ 4 0 0 0 0 0 1 0 0 1 0 9 18 1 2 14 2 981 1 0 4] + [ 4 6 14 16 1 1 0 29 10 0 3 0 2 0 16 0 1 0 959 0 6] + [ 1 4 8 0 1 5 5 16 0 0 0 10 8 1 0 5 12 0 0 1073 3] + [ 264 264 215 86 86 171 50 128 133 97 243 159 345 344 222 71 235 79 220 256 4237]] + +2023-10-05 21:12:31,276 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:12:31,276 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:12:31,281 - + +2023-10-05 21:12:31,282 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:12:32,272 - Epoch: [65][ 10/ 1236] Overall Loss 0.334599 Objective Loss 0.334599 LR 0.001000 Time 0.098952 +2023-10-05 21:12:32,470 - Epoch: [65][ 20/ 1236] Overall Loss 0.314988 Objective Loss 0.314988 LR 0.001000 Time 0.059398 +2023-10-05 21:12:32,668 - Epoch: [65][ 30/ 1236] Overall Loss 0.307272 Objective Loss 0.307272 LR 0.001000 Time 0.046178 +2023-10-05 21:12:32,867 - Epoch: [65][ 40/ 1236] Overall Loss 0.305794 Objective Loss 0.305794 LR 0.001000 Time 0.039600 +2023-10-05 21:12:33,065 - Epoch: [65][ 50/ 1236] Overall Loss 0.303550 Objective Loss 0.303550 LR 0.001000 Time 0.035624 +2023-10-05 21:12:33,264 - Epoch: [65][ 60/ 1236] Overall Loss 0.298165 Objective Loss 0.298165 LR 0.001000 Time 0.033000 +2023-10-05 21:12:33,461 - Epoch: [65][ 70/ 1236] Overall Loss 0.303190 Objective Loss 0.303190 LR 0.001000 Time 0.031103 +2023-10-05 21:12:33,660 - Epoch: [65][ 80/ 1236] Overall Loss 0.304800 Objective Loss 0.304800 LR 0.001000 Time 0.029698 +2023-10-05 21:12:33,858 - Epoch: [65][ 90/ 1236] Overall Loss 0.305130 Objective Loss 0.305130 LR 0.001000 Time 0.028588 +2023-10-05 21:12:34,057 - Epoch: [65][ 100/ 1236] Overall Loss 0.306742 Objective Loss 0.306742 LR 0.001000 Time 0.027716 +2023-10-05 21:12:34,254 - Epoch: [65][ 110/ 1236] Overall Loss 0.308567 Objective Loss 0.308567 LR 0.001000 Time 0.026988 +2023-10-05 21:12:34,452 - Epoch: [65][ 120/ 1236] Overall Loss 0.308071 Objective Loss 0.308071 LR 0.001000 Time 0.026390 +2023-10-05 21:12:34,650 - Epoch: [65][ 130/ 1236] Overall Loss 0.308871 Objective Loss 0.308871 LR 0.001000 Time 0.025877 +2023-10-05 21:12:34,849 - Epoch: [65][ 140/ 1236] Overall Loss 0.306815 Objective Loss 0.306815 LR 0.001000 Time 0.025448 +2023-10-05 21:12:35,046 - Epoch: [65][ 150/ 1236] Overall Loss 0.308446 Objective Loss 0.308446 LR 0.001000 Time 0.025065 +2023-10-05 21:12:35,244 - Epoch: [65][ 160/ 1236] Overall Loss 0.310476 Objective Loss 0.310476 LR 0.001000 Time 0.024733 +2023-10-05 21:12:35,442 - Epoch: [65][ 170/ 1236] Overall Loss 0.311423 Objective Loss 0.311423 LR 0.001000 Time 0.024437 +2023-10-05 21:12:35,640 - Epoch: [65][ 180/ 1236] Overall Loss 0.311955 Objective Loss 0.311955 LR 0.001000 Time 0.024179 +2023-10-05 21:12:35,836 - Epoch: [65][ 190/ 1236] Overall Loss 0.312608 Objective Loss 0.312608 LR 0.001000 Time 0.023936 +2023-10-05 21:12:36,034 - Epoch: [65][ 200/ 1236] Overall Loss 0.313057 Objective Loss 0.313057 LR 0.001000 Time 0.023731 +2023-10-05 21:12:36,231 - Epoch: [65][ 210/ 1236] Overall Loss 0.313218 Objective Loss 0.313218 LR 0.001000 Time 0.023537 +2023-10-05 21:12:36,430 - Epoch: [65][ 220/ 1236] Overall Loss 0.314505 Objective Loss 0.314505 LR 0.001000 Time 0.023370 +2023-10-05 21:12:36,628 - Epoch: [65][ 230/ 1236] Overall Loss 0.316644 Objective Loss 0.316644 LR 0.001000 Time 0.023211 +2023-10-05 21:12:36,827 - Epoch: [65][ 240/ 1236] Overall Loss 0.316782 Objective Loss 0.316782 LR 0.001000 Time 0.023073 +2023-10-05 21:12:37,023 - Epoch: [65][ 250/ 1236] Overall Loss 0.318040 Objective Loss 0.318040 LR 0.001000 Time 0.022934 +2023-10-05 21:12:37,223 - Epoch: [65][ 260/ 1236] Overall Loss 0.318912 Objective Loss 0.318912 LR 0.001000 Time 0.022817 +2023-10-05 21:12:37,420 - Epoch: [65][ 270/ 1236] Overall Loss 0.318507 Objective Loss 0.318507 LR 0.001000 Time 0.022702 +2023-10-05 21:12:37,619 - Epoch: [65][ 280/ 1236] Overall Loss 0.318766 Objective Loss 0.318766 LR 0.001000 Time 0.022601 +2023-10-05 21:12:37,817 - Epoch: [65][ 290/ 1236] Overall Loss 0.318967 Objective Loss 0.318967 LR 0.001000 Time 0.022502 +2023-10-05 21:12:38,015 - Epoch: [65][ 300/ 1236] Overall Loss 0.318913 Objective Loss 0.318913 LR 0.001000 Time 0.022412 +2023-10-05 21:12:38,212 - Epoch: [65][ 310/ 1236] Overall Loss 0.319206 Objective Loss 0.319206 LR 0.001000 Time 0.022323 +2023-10-05 21:12:38,411 - Epoch: [65][ 320/ 1236] Overall Loss 0.318950 Objective Loss 0.318950 LR 0.001000 Time 0.022247 +2023-10-05 21:12:38,609 - Epoch: [65][ 330/ 1236] Overall Loss 0.319460 Objective Loss 0.319460 LR 0.001000 Time 0.022171 +2023-10-05 21:12:38,808 - Epoch: [65][ 340/ 1236] Overall Loss 0.318665 Objective Loss 0.318665 LR 0.001000 Time 0.022104 +2023-10-05 21:12:39,006 - Epoch: [65][ 350/ 1236] Overall Loss 0.319177 Objective Loss 0.319177 LR 0.001000 Time 0.022037 +2023-10-05 21:12:39,205 - Epoch: [65][ 360/ 1236] Overall Loss 0.319711 Objective Loss 0.319711 LR 0.001000 Time 0.021977 +2023-10-05 21:12:39,403 - Epoch: [65][ 370/ 1236] Overall Loss 0.319549 Objective Loss 0.319549 LR 0.001000 Time 0.021917 +2023-10-05 21:12:39,602 - Epoch: [65][ 380/ 1236] Overall Loss 0.318509 Objective Loss 0.318509 LR 0.001000 Time 0.021863 +2023-10-05 21:12:39,800 - Epoch: [65][ 390/ 1236] Overall Loss 0.318742 Objective Loss 0.318742 LR 0.001000 Time 0.021809 +2023-10-05 21:12:39,999 - Epoch: [65][ 400/ 1236] Overall Loss 0.318136 Objective Loss 0.318136 LR 0.001000 Time 0.021761 +2023-10-05 21:12:40,197 - Epoch: [65][ 410/ 1236] Overall Loss 0.317976 Objective Loss 0.317976 LR 0.001000 Time 0.021712 +2023-10-05 21:12:40,396 - Epoch: [65][ 420/ 1236] Overall Loss 0.317995 Objective Loss 0.317995 LR 0.001000 Time 0.021667 +2023-10-05 21:12:40,593 - Epoch: [65][ 430/ 1236] Overall Loss 0.317451 Objective Loss 0.317451 LR 0.001000 Time 0.021622 +2023-10-05 21:12:40,792 - Epoch: [65][ 440/ 1236] Overall Loss 0.317824 Objective Loss 0.317824 LR 0.001000 Time 0.021583 +2023-10-05 21:12:40,990 - Epoch: [65][ 450/ 1236] Overall Loss 0.317923 Objective Loss 0.317923 LR 0.001000 Time 0.021542 +2023-10-05 21:12:41,189 - Epoch: [65][ 460/ 1236] Overall Loss 0.317903 Objective Loss 0.317903 LR 0.001000 Time 0.021506 +2023-10-05 21:12:41,387 - Epoch: [65][ 470/ 1236] Overall Loss 0.318035 Objective Loss 0.318035 LR 0.001000 Time 0.021468 +2023-10-05 21:12:41,586 - Epoch: [65][ 480/ 1236] Overall Loss 0.318574 Objective Loss 0.318574 LR 0.001000 Time 0.021435 +2023-10-05 21:12:41,784 - Epoch: [65][ 490/ 1236] Overall Loss 0.318765 Objective Loss 0.318765 LR 0.001000 Time 0.021401 +2023-10-05 21:12:41,983 - Epoch: [65][ 500/ 1236] Overall Loss 0.318780 Objective Loss 0.318780 LR 0.001000 Time 0.021371 +2023-10-05 21:12:42,181 - Epoch: [65][ 510/ 1236] Overall Loss 0.318446 Objective Loss 0.318446 LR 0.001000 Time 0.021339 +2023-10-05 21:12:42,385 - Epoch: [65][ 520/ 1236] Overall Loss 0.318810 Objective Loss 0.318810 LR 0.001000 Time 0.021321 +2023-10-05 21:12:42,591 - Epoch: [65][ 530/ 1236] Overall Loss 0.318910 Objective Loss 0.318910 LR 0.001000 Time 0.021305 +2023-10-05 21:12:42,794 - Epoch: [65][ 540/ 1236] Overall Loss 0.319230 Objective Loss 0.319230 LR 0.001000 Time 0.021287 +2023-10-05 21:12:42,999 - Epoch: [65][ 550/ 1236] Overall Loss 0.319872 Objective Loss 0.319872 LR 0.001000 Time 0.021271 +2023-10-05 21:12:43,202 - Epoch: [65][ 560/ 1236] Overall Loss 0.319811 Objective Loss 0.319811 LR 0.001000 Time 0.021254 +2023-10-05 21:12:43,407 - Epoch: [65][ 570/ 1236] Overall Loss 0.320295 Objective Loss 0.320295 LR 0.001000 Time 0.021241 +2023-10-05 21:12:43,611 - Epoch: [65][ 580/ 1236] Overall Loss 0.320550 Objective Loss 0.320550 LR 0.001000 Time 0.021225 +2023-10-05 21:12:43,816 - Epoch: [65][ 590/ 1236] Overall Loss 0.320899 Objective Loss 0.320899 LR 0.001000 Time 0.021212 +2023-10-05 21:12:44,020 - Epoch: [65][ 600/ 1236] Overall Loss 0.320805 Objective Loss 0.320805 LR 0.001000 Time 0.021197 +2023-10-05 21:12:44,225 - Epoch: [65][ 610/ 1236] Overall Loss 0.320961 Objective Loss 0.320961 LR 0.001000 Time 0.021186 +2023-10-05 21:12:44,429 - Epoch: [65][ 620/ 1236] Overall Loss 0.320880 Objective Loss 0.320880 LR 0.001000 Time 0.021172 +2023-10-05 21:12:44,634 - Epoch: [65][ 630/ 1236] Overall Loss 0.320614 Objective Loss 0.320614 LR 0.001000 Time 0.021161 +2023-10-05 21:12:44,837 - Epoch: [65][ 640/ 1236] Overall Loss 0.320953 Objective Loss 0.320953 LR 0.001000 Time 0.021148 +2023-10-05 21:12:45,043 - Epoch: [65][ 650/ 1236] Overall Loss 0.321050 Objective Loss 0.321050 LR 0.001000 Time 0.021138 +2023-10-05 21:12:45,246 - Epoch: [65][ 660/ 1236] Overall Loss 0.321475 Objective Loss 0.321475 LR 0.001000 Time 0.021125 +2023-10-05 21:12:45,451 - Epoch: [65][ 670/ 1236] Overall Loss 0.321181 Objective Loss 0.321181 LR 0.001000 Time 0.021115 +2023-10-05 21:12:45,655 - Epoch: [65][ 680/ 1236] Overall Loss 0.321497 Objective Loss 0.321497 LR 0.001000 Time 0.021104 +2023-10-05 21:12:45,860 - Epoch: [65][ 690/ 1236] Overall Loss 0.321460 Objective Loss 0.321460 LR 0.001000 Time 0.021095 +2023-10-05 21:12:46,064 - Epoch: [65][ 700/ 1236] Overall Loss 0.321585 Objective Loss 0.321585 LR 0.001000 Time 0.021083 +2023-10-05 21:12:46,267 - Epoch: [65][ 710/ 1236] Overall Loss 0.321484 Objective Loss 0.321484 LR 0.001000 Time 0.021073 +2023-10-05 21:12:46,471 - Epoch: [65][ 720/ 1236] Overall Loss 0.321317 Objective Loss 0.321317 LR 0.001000 Time 0.021062 +2023-10-05 21:12:46,676 - Epoch: [65][ 730/ 1236] Overall Loss 0.321317 Objective Loss 0.321317 LR 0.001000 Time 0.021055 +2023-10-05 21:12:46,880 - Epoch: [65][ 740/ 1236] Overall Loss 0.321624 Objective Loss 0.321624 LR 0.001000 Time 0.021045 +2023-10-05 21:12:47,084 - Epoch: [65][ 750/ 1236] Overall Loss 0.321463 Objective Loss 0.321463 LR 0.001000 Time 0.021037 +2023-10-05 21:12:47,286 - Epoch: [65][ 760/ 1236] Overall Loss 0.321243 Objective Loss 0.321243 LR 0.001000 Time 0.021025 +2023-10-05 21:12:47,491 - Epoch: [65][ 770/ 1236] Overall Loss 0.321430 Objective Loss 0.321430 LR 0.001000 Time 0.021018 +2023-10-05 21:12:47,695 - Epoch: [65][ 780/ 1236] Overall Loss 0.321735 Objective Loss 0.321735 LR 0.001000 Time 0.021009 +2023-10-05 21:12:47,900 - Epoch: [65][ 790/ 1236] Overall Loss 0.321915 Objective Loss 0.321915 LR 0.001000 Time 0.021002 +2023-10-05 21:12:48,102 - Epoch: [65][ 800/ 1236] Overall Loss 0.322056 Objective Loss 0.322056 LR 0.001000 Time 0.020992 +2023-10-05 21:12:48,307 - Epoch: [65][ 810/ 1236] Overall Loss 0.321891 Objective Loss 0.321891 LR 0.001000 Time 0.020985 +2023-10-05 21:12:48,511 - Epoch: [65][ 820/ 1236] Overall Loss 0.321969 Objective Loss 0.321969 LR 0.001000 Time 0.020977 +2023-10-05 21:12:48,716 - Epoch: [65][ 830/ 1236] Overall Loss 0.322228 Objective Loss 0.322228 LR 0.001000 Time 0.020971 +2023-10-05 21:12:48,919 - Epoch: [65][ 840/ 1236] Overall Loss 0.322239 Objective Loss 0.322239 LR 0.001000 Time 0.020963 +2023-10-05 21:12:49,125 - Epoch: [65][ 850/ 1236] Overall Loss 0.322233 Objective Loss 0.322233 LR 0.001000 Time 0.020958 +2023-10-05 21:12:49,328 - Epoch: [65][ 860/ 1236] Overall Loss 0.322501 Objective Loss 0.322501 LR 0.001000 Time 0.020950 +2023-10-05 21:12:49,533 - Epoch: [65][ 870/ 1236] Overall Loss 0.322723 Objective Loss 0.322723 LR 0.001000 Time 0.020945 +2023-10-05 21:12:49,737 - Epoch: [65][ 880/ 1236] Overall Loss 0.323065 Objective Loss 0.323065 LR 0.001000 Time 0.020937 +2023-10-05 21:12:49,942 - Epoch: [65][ 890/ 1236] Overall Loss 0.323527 Objective Loss 0.323527 LR 0.001000 Time 0.020932 +2023-10-05 21:12:50,146 - Epoch: [65][ 900/ 1236] Overall Loss 0.323855 Objective Loss 0.323855 LR 0.001000 Time 0.020926 +2023-10-05 21:12:50,351 - Epoch: [65][ 910/ 1236] Overall Loss 0.324137 Objective Loss 0.324137 LR 0.001000 Time 0.020921 +2023-10-05 21:12:50,555 - Epoch: [65][ 920/ 1236] Overall Loss 0.324323 Objective Loss 0.324323 LR 0.001000 Time 0.020915 +2023-10-05 21:12:50,760 - Epoch: [65][ 930/ 1236] Overall Loss 0.324317 Objective Loss 0.324317 LR 0.001000 Time 0.020910 +2023-10-05 21:12:50,964 - Epoch: [65][ 940/ 1236] Overall Loss 0.324292 Objective Loss 0.324292 LR 0.001000 Time 0.020904 +2023-10-05 21:12:51,169 - Epoch: [65][ 950/ 1236] Overall Loss 0.324165 Objective Loss 0.324165 LR 0.001000 Time 0.020900 +2023-10-05 21:12:51,373 - Epoch: [65][ 960/ 1236] Overall Loss 0.323851 Objective Loss 0.323851 LR 0.001000 Time 0.020894 +2023-10-05 21:12:51,578 - Epoch: [65][ 970/ 1236] Overall Loss 0.324015 Objective Loss 0.324015 LR 0.001000 Time 0.020890 +2023-10-05 21:12:51,782 - Epoch: [65][ 980/ 1236] Overall Loss 0.323833 Objective Loss 0.323833 LR 0.001000 Time 0.020884 +2023-10-05 21:12:51,987 - Epoch: [65][ 990/ 1236] Overall Loss 0.323940 Objective Loss 0.323940 LR 0.001000 Time 0.020880 +2023-10-05 21:12:52,191 - Epoch: [65][ 1000/ 1236] Overall Loss 0.323795 Objective Loss 0.323795 LR 0.001000 Time 0.020874 +2023-10-05 21:12:52,396 - Epoch: [65][ 1010/ 1236] Overall Loss 0.323696 Objective Loss 0.323696 LR 0.001000 Time 0.020871 +2023-10-05 21:12:52,600 - Epoch: [65][ 1020/ 1236] Overall Loss 0.323725 Objective Loss 0.323725 LR 0.001000 Time 0.020865 +2023-10-05 21:12:52,805 - Epoch: [65][ 1030/ 1236] Overall Loss 0.323776 Objective Loss 0.323776 LR 0.001000 Time 0.020861 +2023-10-05 21:12:53,008 - Epoch: [65][ 1040/ 1236] Overall Loss 0.323696 Objective Loss 0.323696 LR 0.001000 Time 0.020856 +2023-10-05 21:12:53,213 - Epoch: [65][ 1050/ 1236] Overall Loss 0.323641 Objective Loss 0.323641 LR 0.001000 Time 0.020853 +2023-10-05 21:12:53,417 - Epoch: [65][ 1060/ 1236] Overall Loss 0.323549 Objective Loss 0.323549 LR 0.001000 Time 0.020847 +2023-10-05 21:12:53,622 - Epoch: [65][ 1070/ 1236] Overall Loss 0.323868 Objective Loss 0.323868 LR 0.001000 Time 0.020844 +2023-10-05 21:12:53,826 - Epoch: [65][ 1080/ 1236] Overall Loss 0.323854 Objective Loss 0.323854 LR 0.001000 Time 0.020839 +2023-10-05 21:12:54,031 - Epoch: [65][ 1090/ 1236] Overall Loss 0.324353 Objective Loss 0.324353 LR 0.001000 Time 0.020836 +2023-10-05 21:12:54,235 - Epoch: [65][ 1100/ 1236] Overall Loss 0.324702 Objective Loss 0.324702 LR 0.001000 Time 0.020831 +2023-10-05 21:12:54,440 - Epoch: [65][ 1110/ 1236] Overall Loss 0.325074 Objective Loss 0.325074 LR 0.001000 Time 0.020828 +2023-10-05 21:12:54,643 - Epoch: [65][ 1120/ 1236] Overall Loss 0.325156 Objective Loss 0.325156 LR 0.001000 Time 0.020824 +2023-10-05 21:12:54,848 - Epoch: [65][ 1130/ 1236] Overall Loss 0.325357 Objective Loss 0.325357 LR 0.001000 Time 0.020820 +2023-10-05 21:12:55,052 - Epoch: [65][ 1140/ 1236] Overall Loss 0.325814 Objective Loss 0.325814 LR 0.001000 Time 0.020816 +2023-10-05 21:12:55,258 - Epoch: [65][ 1150/ 1236] Overall Loss 0.325985 Objective Loss 0.325985 LR 0.001000 Time 0.020814 +2023-10-05 21:12:55,461 - Epoch: [65][ 1160/ 1236] Overall Loss 0.325940 Objective Loss 0.325940 LR 0.001000 Time 0.020809 +2023-10-05 21:12:55,666 - Epoch: [65][ 1170/ 1236] Overall Loss 0.325948 Objective Loss 0.325948 LR 0.001000 Time 0.020806 +2023-10-05 21:12:55,870 - Epoch: [65][ 1180/ 1236] Overall Loss 0.326063 Objective Loss 0.326063 LR 0.001000 Time 0.020802 +2023-10-05 21:12:56,075 - Epoch: [65][ 1190/ 1236] Overall Loss 0.325797 Objective Loss 0.325797 LR 0.001000 Time 0.020800 +2023-10-05 21:12:56,279 - Epoch: [65][ 1200/ 1236] Overall Loss 0.325926 Objective Loss 0.325926 LR 0.001000 Time 0.020796 +2023-10-05 21:12:56,484 - Epoch: [65][ 1210/ 1236] Overall Loss 0.325883 Objective Loss 0.325883 LR 0.001000 Time 0.020793 +2023-10-05 21:12:56,688 - Epoch: [65][ 1220/ 1236] Overall Loss 0.325838 Objective Loss 0.325838 LR 0.001000 Time 0.020789 +2023-10-05 21:12:56,945 - Epoch: [65][ 1230/ 1236] Overall Loss 0.326076 Objective Loss 0.326076 LR 0.001000 Time 0.020830 +2023-10-05 21:12:57,064 - Epoch: [65][ 1236/ 1236] Overall Loss 0.326065 Objective Loss 0.326065 Top1 86.965377 Top5 98.167006 LR 0.001000 Time 0.020824 +2023-10-05 21:12:57,196 - --- validate (epoch=65)----------- +2023-10-05 21:12:57,196 - 29943 samples (256 per mini-batch) +2023-10-05 21:12:57,664 - Epoch: [65][ 10/ 117] Loss 0.382917 Top1 80.468750 Top5 97.031250 +2023-10-05 21:12:57,814 - Epoch: [65][ 20/ 117] Loss 0.380881 Top1 80.371094 Top5 97.031250 +2023-10-05 21:12:57,965 - Epoch: [65][ 30/ 117] Loss 0.378409 Top1 80.546875 Top5 97.005208 +2023-10-05 21:12:58,112 - Epoch: [65][ 40/ 117] Loss 0.381409 Top1 80.654297 Top5 97.119141 +2023-10-05 21:12:58,261 - Epoch: [65][ 50/ 117] Loss 0.377257 Top1 80.781250 Top5 97.210938 +2023-10-05 21:12:58,407 - Epoch: [65][ 60/ 117] Loss 0.373359 Top1 81.009115 Top5 97.167969 +2023-10-05 21:12:58,556 - Epoch: [65][ 70/ 117] Loss 0.370196 Top1 81.054688 Top5 97.187500 +2023-10-05 21:12:58,704 - Epoch: [65][ 80/ 117] Loss 0.372182 Top1 80.932617 Top5 97.128906 +2023-10-05 21:12:58,851 - Epoch: [65][ 90/ 117] Loss 0.372435 Top1 80.798611 Top5 97.100694 +2023-10-05 21:12:58,998 - Epoch: [65][ 100/ 117] Loss 0.375124 Top1 80.742188 Top5 97.097656 +2023-10-05 21:12:59,153 - Epoch: [65][ 110/ 117] Loss 0.372732 Top1 80.717330 Top5 97.162642 +2023-10-05 21:12:59,238 - Epoch: [65][ 117/ 117] Loss 0.373960 Top1 80.629863 Top5 97.164613 +2023-10-05 21:12:59,338 - ==> Top1: 80.630 Top5: 97.165 Loss: 0.374 + +2023-10-05 21:12:59,339 - ==> Confusion: +[[ 915 2 9 6 3 4 0 0 12 76 1 0 1 2 3 2 3 1 2 0 8] + [ 1 1009 4 1 7 38 2 24 6 0 1 1 0 3 0 4 7 1 13 0 9] + [ 3 1 977 18 0 0 15 3 0 2 5 2 8 0 1 3 3 2 6 1 6] + [ 4 0 35 959 1 4 0 1 4 0 15 0 5 4 21 3 0 6 13 1 13] + [ 39 5 3 0 929 12 0 0 2 13 3 1 1 2 13 5 15 2 0 0 5] + [ 5 30 3 3 0 979 1 31 5 3 7 10 0 11 6 0 4 1 1 4 12] + [ 0 7 67 0 0 0 1084 6 0 0 7 3 1 0 1 3 0 2 3 5 2] + [ 4 16 29 0 0 27 3 1056 3 4 10 6 4 5 0 0 2 0 35 6 8] + [ 24 0 1 0 0 3 1 1 992 40 7 0 0 5 8 2 2 1 2 0 0] + [ 112 0 3 2 1 1 1 1 43 915 0 0 0 24 1 5 1 1 1 1 6] + [ 3 5 23 5 0 2 1 3 24 0 955 0 1 13 2 1 2 1 5 1 6] + [ 2 0 1 0 0 17 0 0 0 0 2 971 10 5 0 2 1 13 0 8 3] + [ 1 0 7 7 0 2 1 2 4 0 1 58 945 2 1 1 4 16 3 4 9] + [ 2 1 2 0 3 13 0 1 35 19 9 7 3 1002 3 0 4 0 0 3 12] + [ 17 4 3 28 3 1 0 0 47 7 4 0 3 0 956 0 1 5 14 0 8] + [ 1 1 7 1 2 0 3 0 1 0 0 15 13 0 1 1033 23 21 0 3 9] + [ 3 14 3 0 5 7 0 0 2 0 0 8 1 1 6 4 1085 1 0 4 17] + [ 1 1 1 3 1 1 1 0 2 1 2 9 23 0 1 2 0 983 2 0 4] + [ 0 3 16 19 0 0 0 31 4 0 4 0 1 0 7 0 2 0 974 0 7] + [ 0 3 10 1 1 6 10 19 0 0 1 18 5 0 1 4 6 1 1 1061 4] + [ 201 198 313 104 77 219 44 124 209 113 228 171 410 294 149 36 156 89 169 238 4363]] + +2023-10-05 21:12:59,340 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:12:59,340 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:12:59,346 - + +2023-10-05 21:12:59,346 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:13:00,441 - Epoch: [66][ 10/ 1236] Overall Loss 0.319138 Objective Loss 0.319138 LR 0.001000 Time 0.109403 +2023-10-05 21:13:00,640 - Epoch: [66][ 20/ 1236] Overall Loss 0.309809 Objective Loss 0.309809 LR 0.001000 Time 0.064626 +2023-10-05 21:13:00,838 - Epoch: [66][ 30/ 1236] Overall Loss 0.294804 Objective Loss 0.294804 LR 0.001000 Time 0.049690 +2023-10-05 21:13:01,038 - Epoch: [66][ 40/ 1236] Overall Loss 0.297340 Objective Loss 0.297340 LR 0.001000 Time 0.042247 +2023-10-05 21:13:01,236 - Epoch: [66][ 50/ 1236] Overall Loss 0.305510 Objective Loss 0.305510 LR 0.001000 Time 0.037761 +2023-10-05 21:13:01,436 - Epoch: [66][ 60/ 1236] Overall Loss 0.314512 Objective Loss 0.314512 LR 0.001000 Time 0.034789 +2023-10-05 21:13:01,635 - Epoch: [66][ 70/ 1236] Overall Loss 0.314245 Objective Loss 0.314245 LR 0.001000 Time 0.032660 +2023-10-05 21:13:01,834 - Epoch: [66][ 80/ 1236] Overall Loss 0.314298 Objective Loss 0.314298 LR 0.001000 Time 0.031063 +2023-10-05 21:13:02,033 - Epoch: [66][ 90/ 1236] Overall Loss 0.316269 Objective Loss 0.316269 LR 0.001000 Time 0.029815 +2023-10-05 21:13:02,232 - Epoch: [66][ 100/ 1236] Overall Loss 0.311528 Objective Loss 0.311528 LR 0.001000 Time 0.028823 +2023-10-05 21:13:02,431 - Epoch: [66][ 110/ 1236] Overall Loss 0.310198 Objective Loss 0.310198 LR 0.001000 Time 0.028006 +2023-10-05 21:13:02,630 - Epoch: [66][ 120/ 1236] Overall Loss 0.309476 Objective Loss 0.309476 LR 0.001000 Time 0.027330 +2023-10-05 21:13:02,829 - Epoch: [66][ 130/ 1236] Overall Loss 0.307998 Objective Loss 0.307998 LR 0.001000 Time 0.026754 +2023-10-05 21:13:03,028 - Epoch: [66][ 140/ 1236] Overall Loss 0.307425 Objective Loss 0.307425 LR 0.001000 Time 0.026262 +2023-10-05 21:13:03,227 - Epoch: [66][ 150/ 1236] Overall Loss 0.309608 Objective Loss 0.309608 LR 0.001000 Time 0.025835 +2023-10-05 21:13:03,426 - Epoch: [66][ 160/ 1236] Overall Loss 0.308496 Objective Loss 0.308496 LR 0.001000 Time 0.025465 +2023-10-05 21:13:03,625 - Epoch: [66][ 170/ 1236] Overall Loss 0.308810 Objective Loss 0.308810 LR 0.001000 Time 0.025134 +2023-10-05 21:13:03,825 - Epoch: [66][ 180/ 1236] Overall Loss 0.309590 Objective Loss 0.309590 LR 0.001000 Time 0.024844 +2023-10-05 21:13:04,024 - Epoch: [66][ 190/ 1236] Overall Loss 0.310125 Objective Loss 0.310125 LR 0.001000 Time 0.024584 +2023-10-05 21:13:04,224 - Epoch: [66][ 200/ 1236] Overall Loss 0.310346 Objective Loss 0.310346 LR 0.001000 Time 0.024353 +2023-10-05 21:13:04,423 - Epoch: [66][ 210/ 1236] Overall Loss 0.310888 Objective Loss 0.310888 LR 0.001000 Time 0.024140 +2023-10-05 21:13:04,623 - Epoch: [66][ 220/ 1236] Overall Loss 0.310286 Objective Loss 0.310286 LR 0.001000 Time 0.023950 +2023-10-05 21:13:04,822 - Epoch: [66][ 230/ 1236] Overall Loss 0.309645 Objective Loss 0.309645 LR 0.001000 Time 0.023774 +2023-10-05 21:13:05,022 - Epoch: [66][ 240/ 1236] Overall Loss 0.308421 Objective Loss 0.308421 LR 0.001000 Time 0.023614 +2023-10-05 21:13:05,221 - Epoch: [66][ 250/ 1236] Overall Loss 0.307836 Objective Loss 0.307836 LR 0.001000 Time 0.023464 +2023-10-05 21:13:05,421 - Epoch: [66][ 260/ 1236] Overall Loss 0.307987 Objective Loss 0.307987 LR 0.001000 Time 0.023331 +2023-10-05 21:13:05,621 - Epoch: [66][ 270/ 1236] Overall Loss 0.308867 Objective Loss 0.308867 LR 0.001000 Time 0.023205 +2023-10-05 21:13:05,821 - Epoch: [66][ 280/ 1236] Overall Loss 0.309724 Objective Loss 0.309724 LR 0.001000 Time 0.023089 +2023-10-05 21:13:06,020 - Epoch: [66][ 290/ 1236] Overall Loss 0.310702 Objective Loss 0.310702 LR 0.001000 Time 0.022979 +2023-10-05 21:13:06,220 - Epoch: [66][ 300/ 1236] Overall Loss 0.311401 Objective Loss 0.311401 LR 0.001000 Time 0.022879 +2023-10-05 21:13:06,420 - Epoch: [66][ 310/ 1236] Overall Loss 0.312773 Objective Loss 0.312773 LR 0.001000 Time 0.022784 +2023-10-05 21:13:06,620 - Epoch: [66][ 320/ 1236] Overall Loss 0.313322 Objective Loss 0.313322 LR 0.001000 Time 0.022697 +2023-10-05 21:13:06,820 - Epoch: [66][ 330/ 1236] Overall Loss 0.312871 Objective Loss 0.312871 LR 0.001000 Time 0.022613 +2023-10-05 21:13:07,020 - Epoch: [66][ 340/ 1236] Overall Loss 0.311365 Objective Loss 0.311365 LR 0.001000 Time 0.022536 +2023-10-05 21:13:07,220 - Epoch: [66][ 350/ 1236] Overall Loss 0.311404 Objective Loss 0.311404 LR 0.001000 Time 0.022461 +2023-10-05 21:13:07,420 - Epoch: [66][ 360/ 1236] Overall Loss 0.312531 Objective Loss 0.312531 LR 0.001000 Time 0.022392 +2023-10-05 21:13:07,619 - Epoch: [66][ 370/ 1236] Overall Loss 0.313565 Objective Loss 0.313565 LR 0.001000 Time 0.022325 +2023-10-05 21:13:07,819 - Epoch: [66][ 380/ 1236] Overall Loss 0.313911 Objective Loss 0.313911 LR 0.001000 Time 0.022263 +2023-10-05 21:13:08,019 - Epoch: [66][ 390/ 1236] Overall Loss 0.313871 Objective Loss 0.313871 LR 0.001000 Time 0.022204 +2023-10-05 21:13:08,219 - Epoch: [66][ 400/ 1236] Overall Loss 0.313496 Objective Loss 0.313496 LR 0.001000 Time 0.022148 +2023-10-05 21:13:08,419 - Epoch: [66][ 410/ 1236] Overall Loss 0.313296 Objective Loss 0.313296 LR 0.001000 Time 0.022094 +2023-10-05 21:13:08,619 - Epoch: [66][ 420/ 1236] Overall Loss 0.313258 Objective Loss 0.313258 LR 0.001000 Time 0.022044 +2023-10-05 21:13:08,818 - Epoch: [66][ 430/ 1236] Overall Loss 0.313274 Objective Loss 0.313274 LR 0.001000 Time 0.021994 +2023-10-05 21:13:09,018 - Epoch: [66][ 440/ 1236] Overall Loss 0.314104 Objective Loss 0.314104 LR 0.001000 Time 0.021948 +2023-10-05 21:13:09,218 - Epoch: [66][ 450/ 1236] Overall Loss 0.314141 Objective Loss 0.314141 LR 0.001000 Time 0.021903 +2023-10-05 21:13:09,418 - Epoch: [66][ 460/ 1236] Overall Loss 0.314353 Objective Loss 0.314353 LR 0.001000 Time 0.021861 +2023-10-05 21:13:09,617 - Epoch: [66][ 470/ 1236] Overall Loss 0.314619 Objective Loss 0.314619 LR 0.001000 Time 0.021819 +2023-10-05 21:13:09,817 - Epoch: [66][ 480/ 1236] Overall Loss 0.314768 Objective Loss 0.314768 LR 0.001000 Time 0.021780 +2023-10-05 21:13:10,017 - Epoch: [66][ 490/ 1236] Overall Loss 0.314702 Objective Loss 0.314702 LR 0.001000 Time 0.021744 +2023-10-05 21:13:10,217 - Epoch: [66][ 500/ 1236] Overall Loss 0.314599 Objective Loss 0.314599 LR 0.001000 Time 0.021708 +2023-10-05 21:13:10,422 - Epoch: [66][ 510/ 1236] Overall Loss 0.314782 Objective Loss 0.314782 LR 0.001000 Time 0.021684 +2023-10-05 21:13:10,624 - Epoch: [66][ 520/ 1236] Overall Loss 0.314509 Objective Loss 0.314509 LR 0.001000 Time 0.021653 +2023-10-05 21:13:10,826 - Epoch: [66][ 530/ 1236] Overall Loss 0.314430 Objective Loss 0.314430 LR 0.001000 Time 0.021626 +2023-10-05 21:13:11,028 - Epoch: [66][ 540/ 1236] Overall Loss 0.315048 Objective Loss 0.315048 LR 0.001000 Time 0.021600 +2023-10-05 21:13:11,231 - Epoch: [66][ 550/ 1236] Overall Loss 0.314415 Objective Loss 0.314415 LR 0.001000 Time 0.021575 +2023-10-05 21:13:11,434 - Epoch: [66][ 560/ 1236] Overall Loss 0.314526 Objective Loss 0.314526 LR 0.001000 Time 0.021550 +2023-10-05 21:13:11,636 - Epoch: [66][ 570/ 1236] Overall Loss 0.314462 Objective Loss 0.314462 LR 0.001000 Time 0.021527 +2023-10-05 21:13:11,839 - Epoch: [66][ 580/ 1236] Overall Loss 0.314912 Objective Loss 0.314912 LR 0.001000 Time 0.021505 +2023-10-05 21:13:12,041 - Epoch: [66][ 590/ 1236] Overall Loss 0.315267 Objective Loss 0.315267 LR 0.001000 Time 0.021483 +2023-10-05 21:13:12,244 - Epoch: [66][ 600/ 1236] Overall Loss 0.315689 Objective Loss 0.315689 LR 0.001000 Time 0.021461 +2023-10-05 21:13:12,446 - Epoch: [66][ 610/ 1236] Overall Loss 0.315248 Objective Loss 0.315248 LR 0.001000 Time 0.021442 +2023-10-05 21:13:12,649 - Epoch: [66][ 620/ 1236] Overall Loss 0.314816 Objective Loss 0.314816 LR 0.001000 Time 0.021421 +2023-10-05 21:13:12,851 - Epoch: [66][ 630/ 1236] Overall Loss 0.315168 Objective Loss 0.315168 LR 0.001000 Time 0.021403 +2023-10-05 21:13:13,054 - Epoch: [66][ 640/ 1236] Overall Loss 0.315197 Objective Loss 0.315197 LR 0.001000 Time 0.021384 +2023-10-05 21:13:13,257 - Epoch: [66][ 650/ 1236] Overall Loss 0.315808 Objective Loss 0.315808 LR 0.001000 Time 0.021366 +2023-10-05 21:13:13,459 - Epoch: [66][ 660/ 1236] Overall Loss 0.316364 Objective Loss 0.316364 LR 0.001000 Time 0.021349 +2023-10-05 21:13:13,661 - Epoch: [66][ 670/ 1236] Overall Loss 0.316911 Objective Loss 0.316911 LR 0.001000 Time 0.021332 +2023-10-05 21:13:13,864 - Epoch: [66][ 680/ 1236] Overall Loss 0.317165 Objective Loss 0.317165 LR 0.001000 Time 0.021315 +2023-10-05 21:13:14,066 - Epoch: [66][ 690/ 1236] Overall Loss 0.317497 Objective Loss 0.317497 LR 0.001000 Time 0.021300 +2023-10-05 21:13:14,269 - Epoch: [66][ 700/ 1236] Overall Loss 0.317541 Objective Loss 0.317541 LR 0.001000 Time 0.021284 +2023-10-05 21:13:14,471 - Epoch: [66][ 710/ 1236] Overall Loss 0.317494 Objective Loss 0.317494 LR 0.001000 Time 0.021269 +2023-10-05 21:13:14,674 - Epoch: [66][ 720/ 1236] Overall Loss 0.317690 Objective Loss 0.317690 LR 0.001000 Time 0.021254 +2023-10-05 21:13:14,876 - Epoch: [66][ 730/ 1236] Overall Loss 0.317701 Objective Loss 0.317701 LR 0.001000 Time 0.021240 +2023-10-05 21:13:15,079 - Epoch: [66][ 740/ 1236] Overall Loss 0.317971 Objective Loss 0.317971 LR 0.001000 Time 0.021226 +2023-10-05 21:13:15,281 - Epoch: [66][ 750/ 1236] Overall Loss 0.318356 Objective Loss 0.318356 LR 0.001000 Time 0.021213 +2023-10-05 21:13:15,484 - Epoch: [66][ 760/ 1236] Overall Loss 0.318530 Objective Loss 0.318530 LR 0.001000 Time 0.021200 +2023-10-05 21:13:15,687 - Epoch: [66][ 770/ 1236] Overall Loss 0.318572 Objective Loss 0.318572 LR 0.001000 Time 0.021187 +2023-10-05 21:13:15,889 - Epoch: [66][ 780/ 1236] Overall Loss 0.318568 Objective Loss 0.318568 LR 0.001000 Time 0.021174 +2023-10-05 21:13:16,091 - Epoch: [66][ 790/ 1236] Overall Loss 0.318462 Objective Loss 0.318462 LR 0.001000 Time 0.021163 +2023-10-05 21:13:16,294 - Epoch: [66][ 800/ 1236] Overall Loss 0.318259 Objective Loss 0.318259 LR 0.001000 Time 0.021150 +2023-10-05 21:13:16,496 - Epoch: [66][ 810/ 1236] Overall Loss 0.318037 Objective Loss 0.318037 LR 0.001000 Time 0.021139 +2023-10-05 21:13:16,699 - Epoch: [66][ 820/ 1236] Overall Loss 0.318312 Objective Loss 0.318312 LR 0.001000 Time 0.021128 +2023-10-05 21:13:16,901 - Epoch: [66][ 830/ 1236] Overall Loss 0.318400 Objective Loss 0.318400 LR 0.001000 Time 0.021117 +2023-10-05 21:13:17,104 - Epoch: [66][ 840/ 1236] Overall Loss 0.318618 Objective Loss 0.318618 LR 0.001000 Time 0.021106 +2023-10-05 21:13:17,306 - Epoch: [66][ 850/ 1236] Overall Loss 0.318637 Objective Loss 0.318637 LR 0.001000 Time 0.021096 +2023-10-05 21:13:17,509 - Epoch: [66][ 860/ 1236] Overall Loss 0.318479 Objective Loss 0.318479 LR 0.001000 Time 0.021085 +2023-10-05 21:13:17,711 - Epoch: [66][ 870/ 1236] Overall Loss 0.318615 Objective Loss 0.318615 LR 0.001000 Time 0.021075 +2023-10-05 21:13:17,914 - Epoch: [66][ 880/ 1236] Overall Loss 0.318836 Objective Loss 0.318836 LR 0.001000 Time 0.021066 +2023-10-05 21:13:18,117 - Epoch: [66][ 890/ 1236] Overall Loss 0.318844 Objective Loss 0.318844 LR 0.001000 Time 0.021056 +2023-10-05 21:13:18,319 - Epoch: [66][ 900/ 1236] Overall Loss 0.318434 Objective Loss 0.318434 LR 0.001000 Time 0.021047 +2023-10-05 21:13:18,522 - Epoch: [66][ 910/ 1236] Overall Loss 0.318342 Objective Loss 0.318342 LR 0.001000 Time 0.021038 +2023-10-05 21:13:18,724 - Epoch: [66][ 920/ 1236] Overall Loss 0.318431 Objective Loss 0.318431 LR 0.001000 Time 0.021029 +2023-10-05 21:13:18,927 - Epoch: [66][ 930/ 1236] Overall Loss 0.318477 Objective Loss 0.318477 LR 0.001000 Time 0.021020 +2023-10-05 21:13:19,129 - Epoch: [66][ 940/ 1236] Overall Loss 0.318565 Objective Loss 0.318565 LR 0.001000 Time 0.021011 +2023-10-05 21:13:19,331 - Epoch: [66][ 950/ 1236] Overall Loss 0.319063 Objective Loss 0.319063 LR 0.001000 Time 0.021003 +2023-10-05 21:13:19,534 - Epoch: [66][ 960/ 1236] Overall Loss 0.318999 Objective Loss 0.318999 LR 0.001000 Time 0.020995 +2023-10-05 21:13:19,737 - Epoch: [66][ 970/ 1236] Overall Loss 0.319177 Objective Loss 0.319177 LR 0.001000 Time 0.020987 +2023-10-05 21:13:19,939 - Epoch: [66][ 980/ 1236] Overall Loss 0.319210 Objective Loss 0.319210 LR 0.001000 Time 0.020979 +2023-10-05 21:13:20,142 - Epoch: [66][ 990/ 1236] Overall Loss 0.319467 Objective Loss 0.319467 LR 0.001000 Time 0.020971 +2023-10-05 21:13:20,344 - Epoch: [66][ 1000/ 1236] Overall Loss 0.319320 Objective Loss 0.319320 LR 0.001000 Time 0.020963 +2023-10-05 21:13:20,547 - Epoch: [66][ 1010/ 1236] Overall Loss 0.319632 Objective Loss 0.319632 LR 0.001000 Time 0.020957 +2023-10-05 21:13:20,749 - Epoch: [66][ 1020/ 1236] Overall Loss 0.319991 Objective Loss 0.319991 LR 0.001000 Time 0.020949 +2023-10-05 21:13:20,952 - Epoch: [66][ 1030/ 1236] Overall Loss 0.320257 Objective Loss 0.320257 LR 0.001000 Time 0.020942 +2023-10-05 21:13:21,154 - Epoch: [66][ 1040/ 1236] Overall Loss 0.320602 Objective Loss 0.320602 LR 0.001000 Time 0.020935 +2023-10-05 21:13:21,357 - Epoch: [66][ 1050/ 1236] Overall Loss 0.321014 Objective Loss 0.321014 LR 0.001000 Time 0.020928 +2023-10-05 21:13:21,559 - Epoch: [66][ 1060/ 1236] Overall Loss 0.320855 Objective Loss 0.320855 LR 0.001000 Time 0.020921 +2023-10-05 21:13:21,762 - Epoch: [66][ 1070/ 1236] Overall Loss 0.320870 Objective Loss 0.320870 LR 0.001000 Time 0.020915 +2023-10-05 21:13:21,964 - Epoch: [66][ 1080/ 1236] Overall Loss 0.320815 Objective Loss 0.320815 LR 0.001000 Time 0.020908 +2023-10-05 21:13:22,167 - Epoch: [66][ 1090/ 1236] Overall Loss 0.320778 Objective Loss 0.320778 LR 0.001000 Time 0.020902 +2023-10-05 21:13:22,369 - Epoch: [66][ 1100/ 1236] Overall Loss 0.320811 Objective Loss 0.320811 LR 0.001000 Time 0.020896 +2023-10-05 21:13:22,571 - Epoch: [66][ 1110/ 1236] Overall Loss 0.320588 Objective Loss 0.320588 LR 0.001000 Time 0.020890 +2023-10-05 21:13:22,774 - Epoch: [66][ 1120/ 1236] Overall Loss 0.320509 Objective Loss 0.320509 LR 0.001000 Time 0.020884 +2023-10-05 21:13:22,977 - Epoch: [66][ 1130/ 1236] Overall Loss 0.320816 Objective Loss 0.320816 LR 0.001000 Time 0.020878 +2023-10-05 21:13:23,179 - Epoch: [66][ 1140/ 1236] Overall Loss 0.321103 Objective Loss 0.321103 LR 0.001000 Time 0.020872 +2023-10-05 21:13:23,382 - Epoch: [66][ 1150/ 1236] Overall Loss 0.321083 Objective Loss 0.321083 LR 0.001000 Time 0.020866 +2023-10-05 21:13:23,584 - Epoch: [66][ 1160/ 1236] Overall Loss 0.321228 Objective Loss 0.321228 LR 0.001000 Time 0.020861 +2023-10-05 21:13:23,787 - Epoch: [66][ 1170/ 1236] Overall Loss 0.321201 Objective Loss 0.321201 LR 0.001000 Time 0.020855 +2023-10-05 21:13:23,989 - Epoch: [66][ 1180/ 1236] Overall Loss 0.321454 Objective Loss 0.321454 LR 0.001000 Time 0.020850 +2023-10-05 21:13:24,192 - Epoch: [66][ 1190/ 1236] Overall Loss 0.321400 Objective Loss 0.321400 LR 0.001000 Time 0.020845 +2023-10-05 21:13:24,394 - Epoch: [66][ 1200/ 1236] Overall Loss 0.321388 Objective Loss 0.321388 LR 0.001000 Time 0.020839 +2023-10-05 21:13:24,597 - Epoch: [66][ 1210/ 1236] Overall Loss 0.321373 Objective Loss 0.321373 LR 0.001000 Time 0.020834 +2023-10-05 21:13:24,799 - Epoch: [66][ 1220/ 1236] Overall Loss 0.321297 Objective Loss 0.321297 LR 0.001000 Time 0.020829 +2023-10-05 21:13:25,056 - Epoch: [66][ 1230/ 1236] Overall Loss 0.321262 Objective Loss 0.321262 LR 0.001000 Time 0.020869 +2023-10-05 21:13:25,175 - Epoch: [66][ 1236/ 1236] Overall Loss 0.321576 Objective Loss 0.321576 Top1 83.299389 Top5 97.352342 LR 0.001000 Time 0.020863 +2023-10-05 21:13:25,297 - --- validate (epoch=66)----------- +2023-10-05 21:13:25,297 - 29943 samples (256 per mini-batch) +2023-10-05 21:13:25,752 - Epoch: [66][ 10/ 117] Loss 0.370463 Top1 81.679688 Top5 97.070312 +2023-10-05 21:13:25,902 - Epoch: [66][ 20/ 117] Loss 0.372699 Top1 81.328125 Top5 97.500000 +2023-10-05 21:13:26,049 - Epoch: [66][ 30/ 117] Loss 0.376713 Top1 81.158854 Top5 97.395833 +2023-10-05 21:13:26,198 - Epoch: [66][ 40/ 117] Loss 0.380361 Top1 81.445312 Top5 97.470703 +2023-10-05 21:13:26,345 - Epoch: [66][ 50/ 117] Loss 0.375965 Top1 81.460938 Top5 97.476562 +2023-10-05 21:13:26,494 - Epoch: [66][ 60/ 117] Loss 0.367313 Top1 81.490885 Top5 97.480469 +2023-10-05 21:13:26,642 - Epoch: [66][ 70/ 117] Loss 0.362231 Top1 81.523438 Top5 97.533482 +2023-10-05 21:13:26,792 - Epoch: [66][ 80/ 117] Loss 0.358628 Top1 81.591797 Top5 97.558594 +2023-10-05 21:13:26,940 - Epoch: [66][ 90/ 117] Loss 0.358477 Top1 81.653646 Top5 97.552083 +2023-10-05 21:13:27,091 - Epoch: [66][ 100/ 117] Loss 0.357719 Top1 81.671875 Top5 97.550781 +2023-10-05 21:13:27,247 - Epoch: [66][ 110/ 117] Loss 0.358306 Top1 81.633523 Top5 97.549716 +2023-10-05 21:13:27,333 - Epoch: [66][ 117/ 117] Loss 0.357057 Top1 81.661824 Top5 97.548676 +2023-10-05 21:13:27,469 - ==> Top1: 81.662 Top5: 97.549 Loss: 0.357 + +2023-10-05 21:13:27,470 - ==> Confusion: +[[ 912 2 5 2 9 4 0 1 4 76 1 1 0 3 8 6 2 1 0 0 13] + [ 2 1031 4 0 7 23 4 26 3 1 1 3 0 0 1 2 7 0 8 0 8] + [ 2 1 953 19 3 0 25 8 0 3 1 4 4 4 0 5 2 2 5 3 12] + [ 4 0 24 945 2 6 0 2 1 0 8 0 3 5 29 6 3 5 30 0 16] + [ 28 6 0 0 951 6 0 0 0 7 1 1 1 2 11 3 21 3 1 0 8] + [ 9 34 2 2 7 949 3 25 3 0 1 18 5 19 5 0 4 1 3 9 17] + [ 0 5 42 0 0 0 1103 9 0 0 4 1 3 0 1 5 0 1 0 8 9] + [ 7 19 21 0 1 33 9 1065 1 1 6 6 3 2 1 2 0 0 25 10 6] + [ 22 5 0 0 0 3 0 0 957 32 6 2 2 16 34 2 1 1 4 0 2] + [ 106 2 2 1 7 5 0 0 28 914 0 1 0 30 7 3 1 3 0 1 8] + [ 5 6 19 8 1 1 5 3 18 2 946 0 0 16 4 0 1 1 8 0 9] + [ 0 1 1 0 0 11 0 0 0 1 0 965 19 5 0 2 3 14 2 10 1] + [ 3 0 5 4 0 4 1 1 1 0 1 49 944 0 3 8 2 26 3 3 10] + [ 2 1 2 1 5 10 0 1 17 14 5 7 4 1027 5 1 4 1 0 2 10] + [ 18 1 4 10 7 0 0 0 20 6 1 0 3 1 1003 0 2 2 10 0 13] + [ 2 2 7 1 1 0 2 0 0 0 0 9 9 2 1 1054 15 14 1 9 5] + [ 2 8 1 2 4 3 0 1 2 0 0 11 1 0 3 11 1094 0 0 6 12] + [ 2 0 3 1 0 0 1 1 0 0 0 2 13 0 2 10 1 993 1 2 6] + [ 2 6 9 13 2 2 0 27 2 1 1 2 8 1 15 0 0 0 964 1 12] + [ 1 1 4 0 1 5 7 17 0 0 2 21 4 1 3 8 9 1 2 1057 8] + [ 149 189 231 68 116 152 45 105 104 97 166 154 395 351 155 61 240 77 173 252 4625]] + +2023-10-05 21:13:27,471 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:13:27,471 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:13:27,477 - + +2023-10-05 21:13:27,477 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:13:28,505 - Epoch: [67][ 10/ 1236] Overall Loss 0.305532 Objective Loss 0.305532 LR 0.001000 Time 0.102733 +2023-10-05 21:13:28,710 - Epoch: [67][ 20/ 1236] Overall Loss 0.317279 Objective Loss 0.317279 LR 0.001000 Time 0.061581 +2023-10-05 21:13:28,915 - Epoch: [67][ 30/ 1236] Overall Loss 0.307057 Objective Loss 0.307057 LR 0.001000 Time 0.047888 +2023-10-05 21:13:29,120 - Epoch: [67][ 40/ 1236] Overall Loss 0.310133 Objective Loss 0.310133 LR 0.001000 Time 0.041023 +2023-10-05 21:13:29,325 - Epoch: [67][ 50/ 1236] Overall Loss 0.305582 Objective Loss 0.305582 LR 0.001000 Time 0.036926 +2023-10-05 21:13:29,530 - Epoch: [67][ 60/ 1236] Overall Loss 0.310190 Objective Loss 0.310190 LR 0.001000 Time 0.034174 +2023-10-05 21:13:29,735 - Epoch: [67][ 70/ 1236] Overall Loss 0.308216 Objective Loss 0.308216 LR 0.001000 Time 0.032219 +2023-10-05 21:13:29,939 - Epoch: [67][ 80/ 1236] Overall Loss 0.306628 Objective Loss 0.306628 LR 0.001000 Time 0.030744 +2023-10-05 21:13:30,145 - Epoch: [67][ 90/ 1236] Overall Loss 0.309253 Objective Loss 0.309253 LR 0.001000 Time 0.029609 +2023-10-05 21:13:30,350 - Epoch: [67][ 100/ 1236] Overall Loss 0.308759 Objective Loss 0.308759 LR 0.001000 Time 0.028690 +2023-10-05 21:13:30,556 - Epoch: [67][ 110/ 1236] Overall Loss 0.310525 Objective Loss 0.310525 LR 0.001000 Time 0.027958 +2023-10-05 21:13:30,763 - Epoch: [67][ 120/ 1236] Overall Loss 0.311362 Objective Loss 0.311362 LR 0.001000 Time 0.027344 +2023-10-05 21:13:30,969 - Epoch: [67][ 130/ 1236] Overall Loss 0.312799 Objective Loss 0.312799 LR 0.001000 Time 0.026830 +2023-10-05 21:13:31,176 - Epoch: [67][ 140/ 1236] Overall Loss 0.312132 Objective Loss 0.312132 LR 0.001000 Time 0.026385 +2023-10-05 21:13:31,382 - Epoch: [67][ 150/ 1236] Overall Loss 0.312092 Objective Loss 0.312092 LR 0.001000 Time 0.025998 +2023-10-05 21:13:31,586 - Epoch: [67][ 160/ 1236] Overall Loss 0.312650 Objective Loss 0.312650 LR 0.001000 Time 0.025644 +2023-10-05 21:13:31,791 - Epoch: [67][ 170/ 1236] Overall Loss 0.311186 Objective Loss 0.311186 LR 0.001000 Time 0.025339 +2023-10-05 21:13:31,994 - Epoch: [67][ 180/ 1236] Overall Loss 0.310585 Objective Loss 0.310585 LR 0.001000 Time 0.025059 +2023-10-05 21:13:32,199 - Epoch: [67][ 190/ 1236] Overall Loss 0.311897 Objective Loss 0.311897 LR 0.001000 Time 0.024816 +2023-10-05 21:13:32,402 - Epoch: [67][ 200/ 1236] Overall Loss 0.310498 Objective Loss 0.310498 LR 0.001000 Time 0.024590 +2023-10-05 21:13:32,606 - Epoch: [67][ 210/ 1236] Overall Loss 0.309959 Objective Loss 0.309959 LR 0.001000 Time 0.024392 +2023-10-05 21:13:32,810 - Epoch: [67][ 220/ 1236] Overall Loss 0.310280 Objective Loss 0.310280 LR 0.001000 Time 0.024207 +2023-10-05 21:13:33,015 - Epoch: [67][ 230/ 1236] Overall Loss 0.308839 Objective Loss 0.308839 LR 0.001000 Time 0.024044 +2023-10-05 21:13:33,218 - Epoch: [67][ 240/ 1236] Overall Loss 0.309724 Objective Loss 0.309724 LR 0.001000 Time 0.023888 +2023-10-05 21:13:33,423 - Epoch: [67][ 250/ 1236] Overall Loss 0.307546 Objective Loss 0.307546 LR 0.001000 Time 0.023750 +2023-10-05 21:13:33,627 - Epoch: [67][ 260/ 1236] Overall Loss 0.309530 Objective Loss 0.309530 LR 0.001000 Time 0.023618 +2023-10-05 21:13:33,832 - Epoch: [67][ 270/ 1236] Overall Loss 0.309887 Objective Loss 0.309887 LR 0.001000 Time 0.023501 +2023-10-05 21:13:34,035 - Epoch: [67][ 280/ 1236] Overall Loss 0.309723 Objective Loss 0.309723 LR 0.001000 Time 0.023388 +2023-10-05 21:13:34,240 - Epoch: [67][ 290/ 1236] Overall Loss 0.308861 Objective Loss 0.308861 LR 0.001000 Time 0.023286 +2023-10-05 21:13:34,443 - Epoch: [67][ 300/ 1236] Overall Loss 0.308155 Objective Loss 0.308155 LR 0.001000 Time 0.023187 +2023-10-05 21:13:34,648 - Epoch: [67][ 310/ 1236] Overall Loss 0.308783 Objective Loss 0.308783 LR 0.001000 Time 0.023098 +2023-10-05 21:13:34,852 - Epoch: [67][ 320/ 1236] Overall Loss 0.308058 Objective Loss 0.308058 LR 0.001000 Time 0.023012 +2023-10-05 21:13:35,057 - Epoch: [67][ 330/ 1236] Overall Loss 0.308470 Objective Loss 0.308470 LR 0.001000 Time 0.022935 +2023-10-05 21:13:35,261 - Epoch: [67][ 340/ 1236] Overall Loss 0.308912 Objective Loss 0.308912 LR 0.001000 Time 0.022858 +2023-10-05 21:13:35,465 - Epoch: [67][ 350/ 1236] Overall Loss 0.309302 Objective Loss 0.309302 LR 0.001000 Time 0.022790 +2023-10-05 21:13:35,669 - Epoch: [67][ 360/ 1236] Overall Loss 0.309231 Objective Loss 0.309231 LR 0.001000 Time 0.022722 +2023-10-05 21:13:35,875 - Epoch: [67][ 370/ 1236] Overall Loss 0.309161 Objective Loss 0.309161 LR 0.001000 Time 0.022664 +2023-10-05 21:13:36,080 - Epoch: [67][ 380/ 1236] Overall Loss 0.309509 Objective Loss 0.309509 LR 0.001000 Time 0.022605 +2023-10-05 21:13:36,285 - Epoch: [67][ 390/ 1236] Overall Loss 0.309469 Objective Loss 0.309469 LR 0.001000 Time 0.022552 +2023-10-05 21:13:36,490 - Epoch: [67][ 400/ 1236] Overall Loss 0.309278 Objective Loss 0.309278 LR 0.001000 Time 0.022498 +2023-10-05 21:13:36,695 - Epoch: [67][ 410/ 1236] Overall Loss 0.308885 Objective Loss 0.308885 LR 0.001000 Time 0.022450 +2023-10-05 21:13:36,900 - Epoch: [67][ 420/ 1236] Overall Loss 0.309247 Objective Loss 0.309247 LR 0.001000 Time 0.022401 +2023-10-05 21:13:37,105 - Epoch: [67][ 430/ 1236] Overall Loss 0.309552 Objective Loss 0.309552 LR 0.001000 Time 0.022358 +2023-10-05 21:13:37,310 - Epoch: [67][ 440/ 1236] Overall Loss 0.309566 Objective Loss 0.309566 LR 0.001000 Time 0.022313 +2023-10-05 21:13:37,515 - Epoch: [67][ 450/ 1236] Overall Loss 0.309843 Objective Loss 0.309843 LR 0.001000 Time 0.022273 +2023-10-05 21:13:37,720 - Epoch: [67][ 460/ 1236] Overall Loss 0.309544 Objective Loss 0.309544 LR 0.001000 Time 0.022233 +2023-10-05 21:13:37,925 - Epoch: [67][ 470/ 1236] Overall Loss 0.310617 Objective Loss 0.310617 LR 0.001000 Time 0.022197 +2023-10-05 21:13:38,130 - Epoch: [67][ 480/ 1236] Overall Loss 0.311290 Objective Loss 0.311290 LR 0.001000 Time 0.022159 +2023-10-05 21:13:38,335 - Epoch: [67][ 490/ 1236] Overall Loss 0.311641 Objective Loss 0.311641 LR 0.001000 Time 0.022126 +2023-10-05 21:13:38,540 - Epoch: [67][ 500/ 1236] Overall Loss 0.311522 Objective Loss 0.311522 LR 0.001000 Time 0.022092 +2023-10-05 21:13:38,745 - Epoch: [67][ 510/ 1236] Overall Loss 0.310960 Objective Loss 0.310960 LR 0.001000 Time 0.022061 +2023-10-05 21:13:38,950 - Epoch: [67][ 520/ 1236] Overall Loss 0.311172 Objective Loss 0.311172 LR 0.001000 Time 0.022029 +2023-10-05 21:13:39,155 - Epoch: [67][ 530/ 1236] Overall Loss 0.311373 Objective Loss 0.311373 LR 0.001000 Time 0.022001 +2023-10-05 21:13:39,360 - Epoch: [67][ 540/ 1236] Overall Loss 0.311390 Objective Loss 0.311390 LR 0.001000 Time 0.021971 +2023-10-05 21:13:39,565 - Epoch: [67][ 550/ 1236] Overall Loss 0.311303 Objective Loss 0.311303 LR 0.001000 Time 0.021945 +2023-10-05 21:13:39,770 - Epoch: [67][ 560/ 1236] Overall Loss 0.311774 Objective Loss 0.311774 LR 0.001000 Time 0.021918 +2023-10-05 21:13:39,975 - Epoch: [67][ 570/ 1236] Overall Loss 0.312261 Objective Loss 0.312261 LR 0.001000 Time 0.021893 +2023-10-05 21:13:40,180 - Epoch: [67][ 580/ 1236] Overall Loss 0.312891 Objective Loss 0.312891 LR 0.001000 Time 0.021868 +2023-10-05 21:13:40,385 - Epoch: [67][ 590/ 1236] Overall Loss 0.313080 Objective Loss 0.313080 LR 0.001000 Time 0.021845 +2023-10-05 21:13:40,590 - Epoch: [67][ 600/ 1236] Overall Loss 0.312992 Objective Loss 0.312992 LR 0.001000 Time 0.021821 +2023-10-05 21:13:40,795 - Epoch: [67][ 610/ 1236] Overall Loss 0.313055 Objective Loss 0.313055 LR 0.001000 Time 0.021799 +2023-10-05 21:13:41,000 - Epoch: [67][ 620/ 1236] Overall Loss 0.313372 Objective Loss 0.313372 LR 0.001000 Time 0.021777 +2023-10-05 21:13:41,205 - Epoch: [67][ 630/ 1236] Overall Loss 0.313823 Objective Loss 0.313823 LR 0.001000 Time 0.021757 +2023-10-05 21:13:41,410 - Epoch: [67][ 640/ 1236] Overall Loss 0.314026 Objective Loss 0.314026 LR 0.001000 Time 0.021736 +2023-10-05 21:13:41,616 - Epoch: [67][ 650/ 1236] Overall Loss 0.314383 Objective Loss 0.314383 LR 0.001000 Time 0.021718 +2023-10-05 21:13:41,820 - Epoch: [67][ 660/ 1236] Overall Loss 0.315093 Objective Loss 0.315093 LR 0.001000 Time 0.021698 +2023-10-05 21:13:42,026 - Epoch: [67][ 670/ 1236] Overall Loss 0.315528 Objective Loss 0.315528 LR 0.001000 Time 0.021681 +2023-10-05 21:13:42,230 - Epoch: [67][ 680/ 1236] Overall Loss 0.315801 Objective Loss 0.315801 LR 0.001000 Time 0.021662 +2023-10-05 21:13:42,436 - Epoch: [67][ 690/ 1236] Overall Loss 0.316358 Objective Loss 0.316358 LR 0.001000 Time 0.021645 +2023-10-05 21:13:42,640 - Epoch: [67][ 700/ 1236] Overall Loss 0.316128 Objective Loss 0.316128 LR 0.001000 Time 0.021628 +2023-10-05 21:13:42,846 - Epoch: [67][ 710/ 1236] Overall Loss 0.316003 Objective Loss 0.316003 LR 0.001000 Time 0.021612 +2023-10-05 21:13:43,050 - Epoch: [67][ 720/ 1236] Overall Loss 0.316016 Objective Loss 0.316016 LR 0.001000 Time 0.021595 +2023-10-05 21:13:43,256 - Epoch: [67][ 730/ 1236] Overall Loss 0.316121 Objective Loss 0.316121 LR 0.001000 Time 0.021581 +2023-10-05 21:13:43,460 - Epoch: [67][ 740/ 1236] Overall Loss 0.315776 Objective Loss 0.315776 LR 0.001000 Time 0.021565 +2023-10-05 21:13:43,665 - Epoch: [67][ 750/ 1236] Overall Loss 0.315921 Objective Loss 0.315921 LR 0.001000 Time 0.021551 +2023-10-05 21:13:43,870 - Epoch: [67][ 760/ 1236] Overall Loss 0.316266 Objective Loss 0.316266 LR 0.001000 Time 0.021536 +2023-10-05 21:13:44,075 - Epoch: [67][ 770/ 1236] Overall Loss 0.316529 Objective Loss 0.316529 LR 0.001000 Time 0.021523 +2023-10-05 21:13:44,280 - Epoch: [67][ 780/ 1236] Overall Loss 0.316609 Objective Loss 0.316609 LR 0.001000 Time 0.021509 +2023-10-05 21:13:44,486 - Epoch: [67][ 790/ 1236] Overall Loss 0.316890 Objective Loss 0.316890 LR 0.001000 Time 0.021496 +2023-10-05 21:13:44,690 - Epoch: [67][ 800/ 1236] Overall Loss 0.316910 Objective Loss 0.316910 LR 0.001000 Time 0.021483 +2023-10-05 21:13:44,896 - Epoch: [67][ 810/ 1236] Overall Loss 0.316869 Objective Loss 0.316869 LR 0.001000 Time 0.021471 +2023-10-05 21:13:45,100 - Epoch: [67][ 820/ 1236] Overall Loss 0.317254 Objective Loss 0.317254 LR 0.001000 Time 0.021458 +2023-10-05 21:13:45,305 - Epoch: [67][ 830/ 1236] Overall Loss 0.317214 Objective Loss 0.317214 LR 0.001000 Time 0.021446 +2023-10-05 21:13:45,510 - Epoch: [67][ 840/ 1236] Overall Loss 0.317416 Objective Loss 0.317416 LR 0.001000 Time 0.021434 +2023-10-05 21:13:45,715 - Epoch: [67][ 850/ 1236] Overall Loss 0.317485 Objective Loss 0.317485 LR 0.001000 Time 0.021424 +2023-10-05 21:13:45,920 - Epoch: [67][ 860/ 1236] Overall Loss 0.317542 Objective Loss 0.317542 LR 0.001000 Time 0.021412 +2023-10-05 21:13:46,125 - Epoch: [67][ 870/ 1236] Overall Loss 0.317129 Objective Loss 0.317129 LR 0.001000 Time 0.021401 +2023-10-05 21:13:46,330 - Epoch: [67][ 880/ 1236] Overall Loss 0.317157 Objective Loss 0.317157 LR 0.001000 Time 0.021390 +2023-10-05 21:13:46,535 - Epoch: [67][ 890/ 1236] Overall Loss 0.317641 Objective Loss 0.317641 LR 0.001000 Time 0.021380 +2023-10-05 21:13:46,740 - Epoch: [67][ 900/ 1236] Overall Loss 0.317322 Objective Loss 0.317322 LR 0.001000 Time 0.021370 +2023-10-05 21:13:46,945 - Epoch: [67][ 910/ 1236] Overall Loss 0.317518 Objective Loss 0.317518 LR 0.001000 Time 0.021360 +2023-10-05 21:13:47,150 - Epoch: [67][ 920/ 1236] Overall Loss 0.317415 Objective Loss 0.317415 LR 0.001000 Time 0.021350 +2023-10-05 21:13:47,356 - Epoch: [67][ 930/ 1236] Overall Loss 0.317449 Objective Loss 0.317449 LR 0.001000 Time 0.021341 +2023-10-05 21:13:47,560 - Epoch: [67][ 940/ 1236] Overall Loss 0.317790 Objective Loss 0.317790 LR 0.001000 Time 0.021331 +2023-10-05 21:13:47,765 - Epoch: [67][ 950/ 1236] Overall Loss 0.317621 Objective Loss 0.317621 LR 0.001000 Time 0.021322 +2023-10-05 21:13:47,969 - Epoch: [67][ 960/ 1236] Overall Loss 0.317307 Objective Loss 0.317307 LR 0.001000 Time 0.021312 +2023-10-05 21:13:48,171 - Epoch: [67][ 970/ 1236] Overall Loss 0.317243 Objective Loss 0.317243 LR 0.001000 Time 0.021301 +2023-10-05 21:13:48,376 - Epoch: [67][ 980/ 1236] Overall Loss 0.317426 Objective Loss 0.317426 LR 0.001000 Time 0.021292 +2023-10-05 21:13:48,578 - Epoch: [67][ 990/ 1236] Overall Loss 0.317519 Objective Loss 0.317519 LR 0.001000 Time 0.021281 +2023-10-05 21:13:48,783 - Epoch: [67][ 1000/ 1236] Overall Loss 0.317665 Objective Loss 0.317665 LR 0.001000 Time 0.021272 +2023-10-05 21:13:48,985 - Epoch: [67][ 1010/ 1236] Overall Loss 0.317766 Objective Loss 0.317766 LR 0.001000 Time 0.021261 +2023-10-05 21:13:49,189 - Epoch: [67][ 1020/ 1236] Overall Loss 0.318024 Objective Loss 0.318024 LR 0.001000 Time 0.021253 +2023-10-05 21:13:49,391 - Epoch: [67][ 1030/ 1236] Overall Loss 0.318189 Objective Loss 0.318189 LR 0.001000 Time 0.021242 +2023-10-05 21:13:49,596 - Epoch: [67][ 1040/ 1236] Overall Loss 0.318503 Objective Loss 0.318503 LR 0.001000 Time 0.021235 +2023-10-05 21:13:49,798 - Epoch: [67][ 1050/ 1236] Overall Loss 0.318676 Objective Loss 0.318676 LR 0.001000 Time 0.021224 +2023-10-05 21:13:50,002 - Epoch: [67][ 1060/ 1236] Overall Loss 0.319091 Objective Loss 0.319091 LR 0.001000 Time 0.021217 +2023-10-05 21:13:50,205 - Epoch: [67][ 1070/ 1236] Overall Loss 0.319026 Objective Loss 0.319026 LR 0.001000 Time 0.021207 +2023-10-05 21:13:50,409 - Epoch: [67][ 1080/ 1236] Overall Loss 0.319002 Objective Loss 0.319002 LR 0.001000 Time 0.021200 +2023-10-05 21:13:50,611 - Epoch: [67][ 1090/ 1236] Overall Loss 0.318883 Objective Loss 0.318883 LR 0.001000 Time 0.021190 +2023-10-05 21:13:50,816 - Epoch: [67][ 1100/ 1236] Overall Loss 0.318596 Objective Loss 0.318596 LR 0.001000 Time 0.021183 +2023-10-05 21:13:51,018 - Epoch: [67][ 1110/ 1236] Overall Loss 0.318299 Objective Loss 0.318299 LR 0.001000 Time 0.021174 +2023-10-05 21:13:51,222 - Epoch: [67][ 1120/ 1236] Overall Loss 0.318546 Objective Loss 0.318546 LR 0.001000 Time 0.021167 +2023-10-05 21:13:51,425 - Epoch: [67][ 1130/ 1236] Overall Loss 0.318257 Objective Loss 0.318257 LR 0.001000 Time 0.021159 +2023-10-05 21:13:51,630 - Epoch: [67][ 1140/ 1236] Overall Loss 0.318106 Objective Loss 0.318106 LR 0.001000 Time 0.021152 +2023-10-05 21:13:51,832 - Epoch: [67][ 1150/ 1236] Overall Loss 0.318221 Objective Loss 0.318221 LR 0.001000 Time 0.021144 +2023-10-05 21:13:52,036 - Epoch: [67][ 1160/ 1236] Overall Loss 0.318246 Objective Loss 0.318246 LR 0.001000 Time 0.021138 +2023-10-05 21:13:52,238 - Epoch: [67][ 1170/ 1236] Overall Loss 0.318372 Objective Loss 0.318372 LR 0.001000 Time 0.021130 +2023-10-05 21:13:52,443 - Epoch: [67][ 1180/ 1236] Overall Loss 0.318698 Objective Loss 0.318698 LR 0.001000 Time 0.021124 +2023-10-05 21:13:52,645 - Epoch: [67][ 1190/ 1236] Overall Loss 0.318770 Objective Loss 0.318770 LR 0.001000 Time 0.021116 +2023-10-05 21:13:52,850 - Epoch: [67][ 1200/ 1236] Overall Loss 0.318864 Objective Loss 0.318864 LR 0.001000 Time 0.021110 +2023-10-05 21:13:53,052 - Epoch: [67][ 1210/ 1236] Overall Loss 0.318828 Objective Loss 0.318828 LR 0.001000 Time 0.021102 +2023-10-05 21:13:53,256 - Epoch: [67][ 1220/ 1236] Overall Loss 0.318896 Objective Loss 0.318896 LR 0.001000 Time 0.021096 +2023-10-05 21:13:53,514 - Epoch: [67][ 1230/ 1236] Overall Loss 0.319142 Objective Loss 0.319142 LR 0.001000 Time 0.021134 +2023-10-05 21:13:53,634 - Epoch: [67][ 1236/ 1236] Overall Loss 0.319222 Objective Loss 0.319222 Top1 84.317719 Top5 97.759674 LR 0.001000 Time 0.021128 +2023-10-05 21:13:53,774 - --- validate (epoch=67)----------- +2023-10-05 21:13:53,774 - 29943 samples (256 per mini-batch) +2023-10-05 21:13:54,241 - Epoch: [67][ 10/ 117] Loss 0.357926 Top1 81.406250 Top5 96.601562 +2023-10-05 21:13:54,390 - Epoch: [67][ 20/ 117] Loss 0.367707 Top1 81.152344 Top5 97.207031 +2023-10-05 21:13:54,539 - Epoch: [67][ 30/ 117] Loss 0.363021 Top1 80.768229 Top5 97.226562 +2023-10-05 21:13:54,688 - Epoch: [67][ 40/ 117] Loss 0.366749 Top1 80.810547 Top5 97.207031 +2023-10-05 21:13:54,837 - Epoch: [67][ 50/ 117] Loss 0.358607 Top1 81.179688 Top5 97.257812 +2023-10-05 21:13:54,985 - Epoch: [67][ 60/ 117] Loss 0.361763 Top1 80.904948 Top5 97.167969 +2023-10-05 21:13:55,133 - Epoch: [67][ 70/ 117] Loss 0.357798 Top1 80.920759 Top5 97.220982 +2023-10-05 21:13:55,281 - Epoch: [67][ 80/ 117] Loss 0.356826 Top1 80.981445 Top5 97.280273 +2023-10-05 21:13:55,429 - Epoch: [67][ 90/ 117] Loss 0.354376 Top1 81.024306 Top5 97.300347 +2023-10-05 21:13:55,577 - Epoch: [67][ 100/ 117] Loss 0.354686 Top1 81.007812 Top5 97.332031 +2023-10-05 21:13:55,731 - Epoch: [67][ 110/ 117] Loss 0.360997 Top1 80.855824 Top5 97.301136 +2023-10-05 21:13:55,818 - Epoch: [67][ 117/ 117] Loss 0.361169 Top1 80.836923 Top5 97.304879 +2023-10-05 21:13:55,953 - ==> Top1: 80.837 Top5: 97.305 Loss: 0.361 + +2023-10-05 21:13:55,954 - ==> Confusion: +[[ 920 3 3 2 9 3 0 0 2 84 0 1 0 1 6 5 1 2 0 0 8] + [ 1 1064 6 0 10 17 1 14 1 0 3 0 0 0 3 4 0 0 5 0 2] + [ 9 2 940 20 6 0 34 5 0 3 3 1 6 1 3 5 1 0 10 1 6] + [ 5 1 21 932 2 2 2 0 2 0 5 0 2 6 59 5 1 6 24 1 13] + [ 27 3 3 0 966 4 0 1 1 12 0 1 0 0 13 4 8 1 0 1 5] + [ 7 48 1 2 11 944 2 27 3 6 7 6 2 17 4 0 6 1 2 11 9] + [ 0 7 24 0 0 1 1115 9 0 0 3 3 1 0 1 8 3 2 3 7 4] + [ 8 30 18 1 3 33 6 1024 1 5 4 5 2 5 1 2 0 2 50 9 9] + [ 21 3 0 3 2 1 0 1 937 69 12 1 1 7 22 2 2 0 4 0 1] + [ 84 1 1 1 7 1 0 0 18 967 0 1 0 15 10 4 1 1 0 0 7] + [ 5 3 16 7 2 2 4 3 15 3 949 2 0 13 9 2 1 0 3 2 12] + [ 0 0 4 0 1 11 1 3 1 2 0 932 39 5 0 3 1 8 0 18 6] + [ 4 1 10 10 1 0 1 1 3 0 2 22 950 7 10 8 1 17 4 6 10] + [ 1 0 3 2 8 5 0 0 19 38 7 3 4 997 5 5 2 2 0 4 14] + [ 20 0 2 8 5 0 0 0 29 9 5 0 1 0 1004 0 1 2 8 0 7] + [ 2 2 1 3 6 0 1 0 0 0 0 7 6 1 2 1064 11 8 0 9 11] + [ 3 12 1 3 11 4 0 0 0 1 0 4 1 1 1 19 1086 0 1 5 8] + [ 2 0 1 1 0 0 1 0 1 0 0 6 21 1 6 7 0 989 0 1 1] + [ 2 4 14 16 2 0 1 24 6 3 2 0 3 0 21 0 2 0 960 0 8] + [ 0 4 6 2 3 5 9 7 0 1 2 11 6 2 0 8 6 1 2 1073 4] + [ 166 251 194 93 143 127 57 105 141 141 198 121 341 319 243 80 213 94 204 282 4392]] + +2023-10-05 21:13:55,955 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:13:55,955 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:13:55,961 - + +2023-10-05 21:13:55,961 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:13:56,972 - Epoch: [68][ 10/ 1236] Overall Loss 0.287784 Objective Loss 0.287784 LR 0.001000 Time 0.101056 +2023-10-05 21:13:57,176 - Epoch: [68][ 20/ 1236] Overall Loss 0.293700 Objective Loss 0.293700 LR 0.001000 Time 0.060693 +2023-10-05 21:13:57,378 - Epoch: [68][ 30/ 1236] Overall Loss 0.301860 Objective Loss 0.301860 LR 0.001000 Time 0.047184 +2023-10-05 21:13:57,582 - Epoch: [68][ 40/ 1236] Overall Loss 0.311330 Objective Loss 0.311330 LR 0.001000 Time 0.040480 +2023-10-05 21:13:57,784 - Epoch: [68][ 50/ 1236] Overall Loss 0.312763 Objective Loss 0.312763 LR 0.001000 Time 0.036416 +2023-10-05 21:13:57,988 - Epoch: [68][ 60/ 1236] Overall Loss 0.317231 Objective Loss 0.317231 LR 0.001000 Time 0.033737 +2023-10-05 21:13:58,190 - Epoch: [68][ 70/ 1236] Overall Loss 0.317010 Objective Loss 0.317010 LR 0.001000 Time 0.031803 +2023-10-05 21:13:58,394 - Epoch: [68][ 80/ 1236] Overall Loss 0.315896 Objective Loss 0.315896 LR 0.001000 Time 0.030365 +2023-10-05 21:13:58,596 - Epoch: [68][ 90/ 1236] Overall Loss 0.312084 Objective Loss 0.312084 LR 0.001000 Time 0.029233 +2023-10-05 21:13:58,800 - Epoch: [68][ 100/ 1236] Overall Loss 0.316417 Objective Loss 0.316417 LR 0.001000 Time 0.028344 +2023-10-05 21:13:59,000 - Epoch: [68][ 110/ 1236] Overall Loss 0.313721 Objective Loss 0.313721 LR 0.001000 Time 0.027589 +2023-10-05 21:13:59,204 - Epoch: [68][ 120/ 1236] Overall Loss 0.311934 Objective Loss 0.311934 LR 0.001000 Time 0.026980 +2023-10-05 21:13:59,404 - Epoch: [68][ 130/ 1236] Overall Loss 0.312701 Objective Loss 0.312701 LR 0.001000 Time 0.026446 +2023-10-05 21:13:59,607 - Epoch: [68][ 140/ 1236] Overall Loss 0.313540 Objective Loss 0.313540 LR 0.001000 Time 0.026002 +2023-10-05 21:13:59,808 - Epoch: [68][ 150/ 1236] Overall Loss 0.312970 Objective Loss 0.312970 LR 0.001000 Time 0.025606 +2023-10-05 21:14:00,011 - Epoch: [68][ 160/ 1236] Overall Loss 0.312376 Objective Loss 0.312376 LR 0.001000 Time 0.025273 +2023-10-05 21:14:00,212 - Epoch: [68][ 170/ 1236] Overall Loss 0.311532 Objective Loss 0.311532 LR 0.001000 Time 0.024966 +2023-10-05 21:14:00,415 - Epoch: [68][ 180/ 1236] Overall Loss 0.310891 Objective Loss 0.310891 LR 0.001000 Time 0.024706 +2023-10-05 21:14:00,617 - Epoch: [68][ 190/ 1236] Overall Loss 0.310174 Objective Loss 0.310174 LR 0.001000 Time 0.024466 +2023-10-05 21:14:00,819 - Epoch: [68][ 200/ 1236] Overall Loss 0.310669 Objective Loss 0.310669 LR 0.001000 Time 0.024251 +2023-10-05 21:14:01,021 - Epoch: [68][ 210/ 1236] Overall Loss 0.311195 Objective Loss 0.311195 LR 0.001000 Time 0.024056 +2023-10-05 21:14:01,223 - Epoch: [68][ 220/ 1236] Overall Loss 0.309504 Objective Loss 0.309504 LR 0.001000 Time 0.023880 +2023-10-05 21:14:01,425 - Epoch: [68][ 230/ 1236] Overall Loss 0.310724 Objective Loss 0.310724 LR 0.001000 Time 0.023718 +2023-10-05 21:14:01,627 - Epoch: [68][ 240/ 1236] Overall Loss 0.310674 Objective Loss 0.310674 LR 0.001000 Time 0.023570 +2023-10-05 21:14:01,829 - Epoch: [68][ 250/ 1236] Overall Loss 0.310303 Objective Loss 0.310303 LR 0.001000 Time 0.023433 +2023-10-05 21:14:02,031 - Epoch: [68][ 260/ 1236] Overall Loss 0.309861 Objective Loss 0.309861 LR 0.001000 Time 0.023308 +2023-10-05 21:14:02,231 - Epoch: [68][ 270/ 1236] Overall Loss 0.309632 Objective Loss 0.309632 LR 0.001000 Time 0.023186 +2023-10-05 21:14:02,434 - Epoch: [68][ 280/ 1236] Overall Loss 0.311072 Objective Loss 0.311072 LR 0.001000 Time 0.023079 +2023-10-05 21:14:02,634 - Epoch: [68][ 290/ 1236] Overall Loss 0.311357 Objective Loss 0.311357 LR 0.001000 Time 0.022972 +2023-10-05 21:14:02,836 - Epoch: [68][ 300/ 1236] Overall Loss 0.311413 Objective Loss 0.311413 LR 0.001000 Time 0.022879 +2023-10-05 21:14:03,037 - Epoch: [68][ 310/ 1236] Overall Loss 0.311604 Objective Loss 0.311604 LR 0.001000 Time 0.022786 +2023-10-05 21:14:03,239 - Epoch: [68][ 320/ 1236] Overall Loss 0.311416 Objective Loss 0.311416 LR 0.001000 Time 0.022705 +2023-10-05 21:14:03,439 - Epoch: [68][ 330/ 1236] Overall Loss 0.312290 Objective Loss 0.312290 LR 0.001000 Time 0.022624 +2023-10-05 21:14:03,641 - Epoch: [68][ 340/ 1236] Overall Loss 0.312626 Objective Loss 0.312626 LR 0.001000 Time 0.022552 +2023-10-05 21:14:03,842 - Epoch: [68][ 350/ 1236] Overall Loss 0.312719 Objective Loss 0.312719 LR 0.001000 Time 0.022479 +2023-10-05 21:14:04,044 - Epoch: [68][ 360/ 1236] Overall Loss 0.313224 Objective Loss 0.313224 LR 0.001000 Time 0.022416 +2023-10-05 21:14:04,245 - Epoch: [68][ 370/ 1236] Overall Loss 0.313843 Objective Loss 0.313843 LR 0.001000 Time 0.022351 +2023-10-05 21:14:04,447 - Epoch: [68][ 380/ 1236] Overall Loss 0.314391 Objective Loss 0.314391 LR 0.001000 Time 0.022294 +2023-10-05 21:14:04,647 - Epoch: [68][ 390/ 1236] Overall Loss 0.314793 Objective Loss 0.314793 LR 0.001000 Time 0.022235 +2023-10-05 21:14:04,849 - Epoch: [68][ 400/ 1236] Overall Loss 0.314426 Objective Loss 0.314426 LR 0.001000 Time 0.022184 +2023-10-05 21:14:05,050 - Epoch: [68][ 410/ 1236] Overall Loss 0.314993 Objective Loss 0.314993 LR 0.001000 Time 0.022130 +2023-10-05 21:14:05,252 - Epoch: [68][ 420/ 1236] Overall Loss 0.314345 Objective Loss 0.314345 LR 0.001000 Time 0.022085 +2023-10-05 21:14:05,452 - Epoch: [68][ 430/ 1236] Overall Loss 0.313637 Objective Loss 0.313637 LR 0.001000 Time 0.022036 +2023-10-05 21:14:05,654 - Epoch: [68][ 440/ 1236] Overall Loss 0.313372 Objective Loss 0.313372 LR 0.001000 Time 0.021994 +2023-10-05 21:14:05,855 - Epoch: [68][ 450/ 1236] Overall Loss 0.313630 Objective Loss 0.313630 LR 0.001000 Time 0.021949 +2023-10-05 21:14:06,057 - Epoch: [68][ 460/ 1236] Overall Loss 0.313329 Objective Loss 0.313329 LR 0.001000 Time 0.021912 +2023-10-05 21:14:06,257 - Epoch: [68][ 470/ 1236] Overall Loss 0.313062 Objective Loss 0.313062 LR 0.001000 Time 0.021870 +2023-10-05 21:14:06,459 - Epoch: [68][ 480/ 1236] Overall Loss 0.313709 Objective Loss 0.313709 LR 0.001000 Time 0.021835 +2023-10-05 21:14:06,659 - Epoch: [68][ 490/ 1236] Overall Loss 0.314622 Objective Loss 0.314622 LR 0.001000 Time 0.021797 +2023-10-05 21:14:06,862 - Epoch: [68][ 500/ 1236] Overall Loss 0.314732 Objective Loss 0.314732 LR 0.001000 Time 0.021766 +2023-10-05 21:14:07,062 - Epoch: [68][ 510/ 1236] Overall Loss 0.315457 Objective Loss 0.315457 LR 0.001000 Time 0.021731 +2023-10-05 21:14:07,265 - Epoch: [68][ 520/ 1236] Overall Loss 0.315605 Objective Loss 0.315605 LR 0.001000 Time 0.021702 +2023-10-05 21:14:07,466 - Epoch: [68][ 530/ 1236] Overall Loss 0.315000 Objective Loss 0.315000 LR 0.001000 Time 0.021671 +2023-10-05 21:14:07,667 - Epoch: [68][ 540/ 1236] Overall Loss 0.315764 Objective Loss 0.315764 LR 0.001000 Time 0.021642 +2023-10-05 21:14:07,868 - Epoch: [68][ 550/ 1236] Overall Loss 0.315667 Objective Loss 0.315667 LR 0.001000 Time 0.021614 +2023-10-05 21:14:08,071 - Epoch: [68][ 560/ 1236] Overall Loss 0.315915 Objective Loss 0.315915 LR 0.001000 Time 0.021589 +2023-10-05 21:14:08,272 - Epoch: [68][ 570/ 1236] Overall Loss 0.315082 Objective Loss 0.315082 LR 0.001000 Time 0.021563 +2023-10-05 21:14:08,475 - Epoch: [68][ 580/ 1236] Overall Loss 0.315167 Objective Loss 0.315167 LR 0.001000 Time 0.021540 +2023-10-05 21:14:08,676 - Epoch: [68][ 590/ 1236] Overall Loss 0.315059 Objective Loss 0.315059 LR 0.001000 Time 0.021515 +2023-10-05 21:14:08,880 - Epoch: [68][ 600/ 1236] Overall Loss 0.316011 Objective Loss 0.316011 LR 0.001000 Time 0.021495 +2023-10-05 21:14:09,081 - Epoch: [68][ 610/ 1236] Overall Loss 0.316564 Objective Loss 0.316564 LR 0.001000 Time 0.021473 +2023-10-05 21:14:09,285 - Epoch: [68][ 620/ 1236] Overall Loss 0.316938 Objective Loss 0.316938 LR 0.001000 Time 0.021454 +2023-10-05 21:14:09,486 - Epoch: [68][ 630/ 1236] Overall Loss 0.317296 Objective Loss 0.317296 LR 0.001000 Time 0.021432 +2023-10-05 21:14:09,689 - Epoch: [68][ 640/ 1236] Overall Loss 0.317710 Objective Loss 0.317710 LR 0.001000 Time 0.021414 +2023-10-05 21:14:09,890 - Epoch: [68][ 650/ 1236] Overall Loss 0.317524 Objective Loss 0.317524 LR 0.001000 Time 0.021394 +2023-10-05 21:14:10,094 - Epoch: [68][ 660/ 1236] Overall Loss 0.317341 Objective Loss 0.317341 LR 0.001000 Time 0.021377 +2023-10-05 21:14:10,295 - Epoch: [68][ 670/ 1236] Overall Loss 0.317815 Objective Loss 0.317815 LR 0.001000 Time 0.021358 +2023-10-05 21:14:10,498 - Epoch: [68][ 680/ 1236] Overall Loss 0.318022 Objective Loss 0.318022 LR 0.001000 Time 0.021343 +2023-10-05 21:14:10,700 - Epoch: [68][ 690/ 1236] Overall Loss 0.317292 Objective Loss 0.317292 LR 0.001000 Time 0.021325 +2023-10-05 21:14:10,903 - Epoch: [68][ 700/ 1236] Overall Loss 0.317376 Objective Loss 0.317376 LR 0.001000 Time 0.021310 +2023-10-05 21:14:11,105 - Epoch: [68][ 710/ 1236] Overall Loss 0.317245 Objective Loss 0.317245 LR 0.001000 Time 0.021293 +2023-10-05 21:14:11,308 - Epoch: [68][ 720/ 1236] Overall Loss 0.317648 Objective Loss 0.317648 LR 0.001000 Time 0.021280 +2023-10-05 21:14:11,509 - Epoch: [68][ 730/ 1236] Overall Loss 0.317959 Objective Loss 0.317959 LR 0.001000 Time 0.021264 +2023-10-05 21:14:11,713 - Epoch: [68][ 740/ 1236] Overall Loss 0.317869 Objective Loss 0.317869 LR 0.001000 Time 0.021250 +2023-10-05 21:14:11,914 - Epoch: [68][ 750/ 1236] Overall Loss 0.318038 Objective Loss 0.318038 LR 0.001000 Time 0.021235 +2023-10-05 21:14:12,117 - Epoch: [68][ 760/ 1236] Overall Loss 0.318348 Objective Loss 0.318348 LR 0.001000 Time 0.021223 +2023-10-05 21:14:12,318 - Epoch: [68][ 770/ 1236] Overall Loss 0.318616 Objective Loss 0.318616 LR 0.001000 Time 0.021208 +2023-10-05 21:14:12,522 - Epoch: [68][ 780/ 1236] Overall Loss 0.318485 Objective Loss 0.318485 LR 0.001000 Time 0.021197 +2023-10-05 21:14:12,723 - Epoch: [68][ 790/ 1236] Overall Loss 0.318830 Objective Loss 0.318830 LR 0.001000 Time 0.021182 +2023-10-05 21:14:12,926 - Epoch: [68][ 800/ 1236] Overall Loss 0.318819 Objective Loss 0.318819 LR 0.001000 Time 0.021171 +2023-10-05 21:14:13,128 - Epoch: [68][ 810/ 1236] Overall Loss 0.318865 Objective Loss 0.318865 LR 0.001000 Time 0.021158 +2023-10-05 21:14:13,331 - Epoch: [68][ 820/ 1236] Overall Loss 0.318922 Objective Loss 0.318922 LR 0.001000 Time 0.021148 +2023-10-05 21:14:13,532 - Epoch: [68][ 830/ 1236] Overall Loss 0.318892 Objective Loss 0.318892 LR 0.001000 Time 0.021135 +2023-10-05 21:14:13,736 - Epoch: [68][ 840/ 1236] Overall Loss 0.318877 Objective Loss 0.318877 LR 0.001000 Time 0.021125 +2023-10-05 21:14:13,937 - Epoch: [68][ 850/ 1236] Overall Loss 0.318952 Objective Loss 0.318952 LR 0.001000 Time 0.021114 +2023-10-05 21:14:14,141 - Epoch: [68][ 860/ 1236] Overall Loss 0.318735 Objective Loss 0.318735 LR 0.001000 Time 0.021104 +2023-10-05 21:14:14,342 - Epoch: [68][ 870/ 1236] Overall Loss 0.318498 Objective Loss 0.318498 LR 0.001000 Time 0.021093 +2023-10-05 21:14:14,545 - Epoch: [68][ 880/ 1236] Overall Loss 0.318698 Objective Loss 0.318698 LR 0.001000 Time 0.021083 +2023-10-05 21:14:14,747 - Epoch: [68][ 890/ 1236] Overall Loss 0.318777 Objective Loss 0.318777 LR 0.001000 Time 0.021072 +2023-10-05 21:14:14,950 - Epoch: [68][ 900/ 1236] Overall Loss 0.318844 Objective Loss 0.318844 LR 0.001000 Time 0.021064 +2023-10-05 21:14:15,152 - Epoch: [68][ 910/ 1236] Overall Loss 0.318625 Objective Loss 0.318625 LR 0.001000 Time 0.021054 +2023-10-05 21:14:15,355 - Epoch: [68][ 920/ 1236] Overall Loss 0.318915 Objective Loss 0.318915 LR 0.001000 Time 0.021046 +2023-10-05 21:14:15,556 - Epoch: [68][ 930/ 1236] Overall Loss 0.318895 Objective Loss 0.318895 LR 0.001000 Time 0.021035 +2023-10-05 21:14:15,760 - Epoch: [68][ 940/ 1236] Overall Loss 0.319066 Objective Loss 0.319066 LR 0.001000 Time 0.021028 +2023-10-05 21:14:15,961 - Epoch: [68][ 950/ 1236] Overall Loss 0.318858 Objective Loss 0.318858 LR 0.001000 Time 0.021018 +2023-10-05 21:14:16,165 - Epoch: [68][ 960/ 1236] Overall Loss 0.319295 Objective Loss 0.319295 LR 0.001000 Time 0.021011 +2023-10-05 21:14:16,367 - Epoch: [68][ 970/ 1236] Overall Loss 0.319374 Objective Loss 0.319374 LR 0.001000 Time 0.021002 +2023-10-05 21:14:16,571 - Epoch: [68][ 980/ 1236] Overall Loss 0.319187 Objective Loss 0.319187 LR 0.001000 Time 0.020996 +2023-10-05 21:14:16,774 - Epoch: [68][ 990/ 1236] Overall Loss 0.319245 Objective Loss 0.319245 LR 0.001000 Time 0.020988 +2023-10-05 21:14:16,980 - Epoch: [68][ 1000/ 1236] Overall Loss 0.319107 Objective Loss 0.319107 LR 0.001000 Time 0.020984 +2023-10-05 21:14:17,182 - Epoch: [68][ 1010/ 1236] Overall Loss 0.319356 Objective Loss 0.319356 LR 0.001000 Time 0.020976 +2023-10-05 21:14:17,388 - Epoch: [68][ 1020/ 1236] Overall Loss 0.319389 Objective Loss 0.319389 LR 0.001000 Time 0.020972 +2023-10-05 21:14:17,590 - Epoch: [68][ 1030/ 1236] Overall Loss 0.319362 Objective Loss 0.319362 LR 0.001000 Time 0.020965 +2023-10-05 21:14:17,796 - Epoch: [68][ 1040/ 1236] Overall Loss 0.319302 Objective Loss 0.319302 LR 0.001000 Time 0.020960 +2023-10-05 21:14:17,999 - Epoch: [68][ 1050/ 1236] Overall Loss 0.319402 Objective Loss 0.319402 LR 0.001000 Time 0.020954 +2023-10-05 21:14:18,204 - Epoch: [68][ 1060/ 1236] Overall Loss 0.319451 Objective Loss 0.319451 LR 0.001000 Time 0.020950 +2023-10-05 21:14:18,407 - Epoch: [68][ 1070/ 1236] Overall Loss 0.319591 Objective Loss 0.319591 LR 0.001000 Time 0.020943 +2023-10-05 21:14:18,612 - Epoch: [68][ 1080/ 1236] Overall Loss 0.319254 Objective Loss 0.319254 LR 0.001000 Time 0.020938 +2023-10-05 21:14:18,815 - Epoch: [68][ 1090/ 1236] Overall Loss 0.319057 Objective Loss 0.319057 LR 0.001000 Time 0.020932 +2023-10-05 21:14:19,020 - Epoch: [68][ 1100/ 1236] Overall Loss 0.319080 Objective Loss 0.319080 LR 0.001000 Time 0.020928 +2023-10-05 21:14:19,223 - Epoch: [68][ 1110/ 1236] Overall Loss 0.318816 Objective Loss 0.318816 LR 0.001000 Time 0.020922 +2023-10-05 21:14:19,428 - Epoch: [68][ 1120/ 1236] Overall Loss 0.318993 Objective Loss 0.318993 LR 0.001000 Time 0.020918 +2023-10-05 21:14:19,631 - Epoch: [68][ 1130/ 1236] Overall Loss 0.318755 Objective Loss 0.318755 LR 0.001000 Time 0.020912 +2023-10-05 21:14:19,837 - Epoch: [68][ 1140/ 1236] Overall Loss 0.318567 Objective Loss 0.318567 LR 0.001000 Time 0.020909 +2023-10-05 21:14:20,040 - Epoch: [68][ 1150/ 1236] Overall Loss 0.318526 Objective Loss 0.318526 LR 0.001000 Time 0.020903 +2023-10-05 21:14:20,245 - Epoch: [68][ 1160/ 1236] Overall Loss 0.318475 Objective Loss 0.318475 LR 0.001000 Time 0.020900 +2023-10-05 21:14:20,448 - Epoch: [68][ 1170/ 1236] Overall Loss 0.318600 Objective Loss 0.318600 LR 0.001000 Time 0.020894 +2023-10-05 21:14:20,653 - Epoch: [68][ 1180/ 1236] Overall Loss 0.318569 Objective Loss 0.318569 LR 0.001000 Time 0.020891 +2023-10-05 21:14:20,855 - Epoch: [68][ 1190/ 1236] Overall Loss 0.318154 Objective Loss 0.318154 LR 0.001000 Time 0.020885 +2023-10-05 21:14:21,061 - Epoch: [68][ 1200/ 1236] Overall Loss 0.318398 Objective Loss 0.318398 LR 0.001000 Time 0.020882 +2023-10-05 21:14:21,264 - Epoch: [68][ 1210/ 1236] Overall Loss 0.318339 Objective Loss 0.318339 LR 0.001000 Time 0.020877 +2023-10-05 21:14:21,469 - Epoch: [68][ 1220/ 1236] Overall Loss 0.318490 Objective Loss 0.318490 LR 0.001000 Time 0.020873 +2023-10-05 21:14:21,725 - Epoch: [68][ 1230/ 1236] Overall Loss 0.318452 Objective Loss 0.318452 LR 0.001000 Time 0.020911 +2023-10-05 21:14:21,844 - Epoch: [68][ 1236/ 1236] Overall Loss 0.318645 Objective Loss 0.318645 Top1 85.539715 Top5 98.574338 LR 0.001000 Time 0.020906 +2023-10-05 21:14:21,974 - --- validate (epoch=68)----------- +2023-10-05 21:14:21,975 - 29943 samples (256 per mini-batch) +2023-10-05 21:14:22,434 - Epoch: [68][ 10/ 117] Loss 0.349418 Top1 81.093750 Top5 97.773438 +2023-10-05 21:14:22,583 - Epoch: [68][ 20/ 117] Loss 0.385691 Top1 80.292969 Top5 97.324219 +2023-10-05 21:14:22,731 - Epoch: [68][ 30/ 117] Loss 0.381102 Top1 80.598958 Top5 97.317708 +2023-10-05 21:14:22,878 - Epoch: [68][ 40/ 117] Loss 0.373687 Top1 80.878906 Top5 97.412109 +2023-10-05 21:14:23,026 - Epoch: [68][ 50/ 117] Loss 0.365778 Top1 81.140625 Top5 97.562500 +2023-10-05 21:14:23,173 - Epoch: [68][ 60/ 117] Loss 0.362140 Top1 81.093750 Top5 97.584635 +2023-10-05 21:14:23,323 - Epoch: [68][ 70/ 117] Loss 0.361410 Top1 81.210938 Top5 97.611607 +2023-10-05 21:14:23,471 - Epoch: [68][ 80/ 117] Loss 0.359128 Top1 81.250000 Top5 97.553711 +2023-10-05 21:14:23,621 - Epoch: [68][ 90/ 117] Loss 0.361659 Top1 81.076389 Top5 97.543403 +2023-10-05 21:14:23,769 - Epoch: [68][ 100/ 117] Loss 0.361162 Top1 81.035156 Top5 97.488281 +2023-10-05 21:14:23,924 - Epoch: [68][ 110/ 117] Loss 0.360005 Top1 81.086648 Top5 97.485795 +2023-10-05 21:14:24,010 - Epoch: [68][ 117/ 117] Loss 0.358850 Top1 81.084060 Top5 97.478543 +2023-10-05 21:14:24,136 - ==> Top1: 81.084 Top5: 97.479 Loss: 0.359 + +2023-10-05 21:14:24,136 - ==> Confusion: +[[ 907 1 6 2 15 3 0 2 2 82 1 1 0 3 4 5 3 0 0 0 13] + [ 0 1043 4 0 11 17 2 22 4 0 5 0 0 4 1 6 4 0 4 1 3] + [ 2 2 949 13 5 0 20 12 0 3 7 1 6 5 1 9 1 0 7 3 10] + [ 3 2 27 927 1 5 2 1 2 2 19 0 5 9 38 5 3 6 19 0 13] + [ 20 7 0 0 972 4 0 1 0 8 5 1 1 3 6 4 11 1 0 1 5] + [ 2 47 0 1 8 949 0 25 4 2 4 7 2 32 6 1 5 1 3 5 12] + [ 0 5 35 0 1 1 1102 9 0 0 6 0 2 0 0 10 1 1 0 12 6] + [ 4 24 20 0 2 36 5 1039 1 5 8 7 1 3 1 6 1 0 33 13 9] + [ 17 2 0 0 1 1 0 0 961 41 11 2 1 23 17 7 0 0 4 0 1] + [ 105 2 3 0 9 2 0 0 28 924 1 1 0 27 2 4 2 0 0 1 8] + [ 2 3 11 0 0 1 3 2 19 1 962 0 0 29 3 1 0 0 3 3 10] + [ 1 0 2 0 1 19 0 3 2 2 1 951 19 4 0 2 3 13 1 7 4] + [ 2 4 3 7 1 3 0 2 3 0 3 41 950 6 2 6 3 14 1 6 11] + [ 2 0 3 0 1 3 0 0 14 22 3 4 1 1057 2 1 0 0 0 0 6] + [ 9 4 1 11 12 0 0 1 27 5 0 0 2 1 1004 1 3 2 9 0 9] + [ 0 2 1 2 2 1 1 0 0 0 0 8 9 1 0 1065 17 8 0 9 8] + [ 1 14 2 0 8 4 1 2 2 0 1 5 3 5 2 12 1081 0 0 10 8] + [ 0 0 2 0 0 0 2 0 1 0 0 5 23 3 2 10 0 983 2 2 3] + [ 1 16 5 13 2 0 0 28 7 0 4 1 4 1 15 0 0 0 960 0 11] + [ 0 4 2 0 2 6 6 14 0 0 3 10 6 9 0 4 13 1 0 1066 6] + [ 144 289 211 50 146 151 42 120 122 127 213 134 374 503 142 70 178 61 122 279 4427]] + +2023-10-05 21:14:24,137 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:14:24,137 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:14:24,143 - + +2023-10-05 21:14:24,143 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:14:25,144 - Epoch: [69][ 10/ 1236] Overall Loss 0.307711 Objective Loss 0.307711 LR 0.001000 Time 0.100040 +2023-10-05 21:14:25,347 - Epoch: [69][ 20/ 1236] Overall Loss 0.305318 Objective Loss 0.305318 LR 0.001000 Time 0.060134 +2023-10-05 21:14:25,548 - Epoch: [69][ 30/ 1236] Overall Loss 0.306245 Objective Loss 0.306245 LR 0.001000 Time 0.046774 +2023-10-05 21:14:25,751 - Epoch: [69][ 40/ 1236] Overall Loss 0.310863 Objective Loss 0.310863 LR 0.001000 Time 0.040149 +2023-10-05 21:14:25,952 - Epoch: [69][ 50/ 1236] Overall Loss 0.311622 Objective Loss 0.311622 LR 0.001000 Time 0.036131 +2023-10-05 21:14:26,155 - Epoch: [69][ 60/ 1236] Overall Loss 0.308970 Objective Loss 0.308970 LR 0.001000 Time 0.033489 +2023-10-05 21:14:26,356 - Epoch: [69][ 70/ 1236] Overall Loss 0.312611 Objective Loss 0.312611 LR 0.001000 Time 0.031571 +2023-10-05 21:14:26,559 - Epoch: [69][ 80/ 1236] Overall Loss 0.309604 Objective Loss 0.309604 LR 0.001000 Time 0.030159 +2023-10-05 21:14:26,760 - Epoch: [69][ 90/ 1236] Overall Loss 0.310390 Objective Loss 0.310390 LR 0.001000 Time 0.029035 +2023-10-05 21:14:26,963 - Epoch: [69][ 100/ 1236] Overall Loss 0.308636 Objective Loss 0.308636 LR 0.001000 Time 0.028160 +2023-10-05 21:14:27,164 - Epoch: [69][ 110/ 1236] Overall Loss 0.309704 Objective Loss 0.309704 LR 0.001000 Time 0.027426 +2023-10-05 21:14:27,367 - Epoch: [69][ 120/ 1236] Overall Loss 0.308272 Objective Loss 0.308272 LR 0.001000 Time 0.026824 +2023-10-05 21:14:27,568 - Epoch: [69][ 130/ 1236] Overall Loss 0.310609 Objective Loss 0.310609 LR 0.001000 Time 0.026306 +2023-10-05 21:14:27,772 - Epoch: [69][ 140/ 1236] Overall Loss 0.309800 Objective Loss 0.309800 LR 0.001000 Time 0.025881 +2023-10-05 21:14:27,976 - Epoch: [69][ 150/ 1236] Overall Loss 0.308664 Objective Loss 0.308664 LR 0.001000 Time 0.025513 +2023-10-05 21:14:28,178 - Epoch: [69][ 160/ 1236] Overall Loss 0.310292 Objective Loss 0.310292 LR 0.001000 Time 0.025178 +2023-10-05 21:14:28,379 - Epoch: [69][ 170/ 1236] Overall Loss 0.309371 Objective Loss 0.309371 LR 0.001000 Time 0.024877 +2023-10-05 21:14:28,581 - Epoch: [69][ 180/ 1236] Overall Loss 0.309860 Objective Loss 0.309860 LR 0.001000 Time 0.024617 +2023-10-05 21:14:28,783 - Epoch: [69][ 190/ 1236] Overall Loss 0.310582 Objective Loss 0.310582 LR 0.001000 Time 0.024381 +2023-10-05 21:14:28,984 - Epoch: [69][ 200/ 1236] Overall Loss 0.309700 Objective Loss 0.309700 LR 0.001000 Time 0.024168 +2023-10-05 21:14:29,185 - Epoch: [69][ 210/ 1236] Overall Loss 0.311549 Objective Loss 0.311549 LR 0.001000 Time 0.023970 +2023-10-05 21:14:29,387 - Epoch: [69][ 220/ 1236] Overall Loss 0.311806 Objective Loss 0.311806 LR 0.001000 Time 0.023797 +2023-10-05 21:14:29,588 - Epoch: [69][ 230/ 1236] Overall Loss 0.311095 Objective Loss 0.311095 LR 0.001000 Time 0.023637 +2023-10-05 21:14:29,789 - Epoch: [69][ 240/ 1236] Overall Loss 0.312109 Objective Loss 0.312109 LR 0.001000 Time 0.023489 +2023-10-05 21:14:29,990 - Epoch: [69][ 250/ 1236] Overall Loss 0.313133 Objective Loss 0.313133 LR 0.001000 Time 0.023350 +2023-10-05 21:14:30,192 - Epoch: [69][ 260/ 1236] Overall Loss 0.312111 Objective Loss 0.312111 LR 0.001000 Time 0.023228 +2023-10-05 21:14:30,392 - Epoch: [69][ 270/ 1236] Overall Loss 0.312063 Objective Loss 0.312063 LR 0.001000 Time 0.023108 +2023-10-05 21:14:30,594 - Epoch: [69][ 280/ 1236] Overall Loss 0.311435 Objective Loss 0.311435 LR 0.001000 Time 0.023003 +2023-10-05 21:14:30,795 - Epoch: [69][ 290/ 1236] Overall Loss 0.311535 Objective Loss 0.311535 LR 0.001000 Time 0.022900 +2023-10-05 21:14:30,997 - Epoch: [69][ 300/ 1236] Overall Loss 0.311772 Objective Loss 0.311772 LR 0.001000 Time 0.022809 +2023-10-05 21:14:31,198 - Epoch: [69][ 310/ 1236] Overall Loss 0.312011 Objective Loss 0.312011 LR 0.001000 Time 0.022720 +2023-10-05 21:14:31,399 - Epoch: [69][ 320/ 1236] Overall Loss 0.311845 Objective Loss 0.311845 LR 0.001000 Time 0.022637 +2023-10-05 21:14:31,599 - Epoch: [69][ 330/ 1236] Overall Loss 0.312445 Objective Loss 0.312445 LR 0.001000 Time 0.022557 +2023-10-05 21:14:31,801 - Epoch: [69][ 340/ 1236] Overall Loss 0.312680 Objective Loss 0.312680 LR 0.001000 Time 0.022487 +2023-10-05 21:14:32,002 - Epoch: [69][ 350/ 1236] Overall Loss 0.312230 Objective Loss 0.312230 LR 0.001000 Time 0.022416 +2023-10-05 21:14:32,204 - Epoch: [69][ 360/ 1236] Overall Loss 0.311760 Objective Loss 0.311760 LR 0.001000 Time 0.022355 +2023-10-05 21:14:32,406 - Epoch: [69][ 370/ 1236] Overall Loss 0.312322 Objective Loss 0.312322 LR 0.001000 Time 0.022294 +2023-10-05 21:14:32,609 - Epoch: [69][ 380/ 1236] Overall Loss 0.311721 Objective Loss 0.311721 LR 0.001000 Time 0.022241 +2023-10-05 21:14:32,810 - Epoch: [69][ 390/ 1236] Overall Loss 0.311681 Objective Loss 0.311681 LR 0.001000 Time 0.022187 +2023-10-05 21:14:33,013 - Epoch: [69][ 400/ 1236] Overall Loss 0.311045 Objective Loss 0.311045 LR 0.001000 Time 0.022139 +2023-10-05 21:14:33,215 - Epoch: [69][ 410/ 1236] Overall Loss 0.310290 Objective Loss 0.310290 LR 0.001000 Time 0.022089 +2023-10-05 21:14:33,418 - Epoch: [69][ 420/ 1236] Overall Loss 0.310503 Objective Loss 0.310503 LR 0.001000 Time 0.022046 +2023-10-05 21:14:33,620 - Epoch: [69][ 430/ 1236] Overall Loss 0.311125 Objective Loss 0.311125 LR 0.001000 Time 0.022001 +2023-10-05 21:14:33,822 - Epoch: [69][ 440/ 1236] Overall Loss 0.311993 Objective Loss 0.311993 LR 0.001000 Time 0.021961 +2023-10-05 21:14:34,024 - Epoch: [69][ 450/ 1236] Overall Loss 0.311774 Objective Loss 0.311774 LR 0.001000 Time 0.021920 +2023-10-05 21:14:34,227 - Epoch: [69][ 460/ 1236] Overall Loss 0.311950 Objective Loss 0.311950 LR 0.001000 Time 0.021884 +2023-10-05 21:14:34,429 - Epoch: [69][ 470/ 1236] Overall Loss 0.311651 Objective Loss 0.311651 LR 0.001000 Time 0.021847 +2023-10-05 21:14:34,632 - Epoch: [69][ 480/ 1236] Overall Loss 0.311480 Objective Loss 0.311480 LR 0.001000 Time 0.021814 +2023-10-05 21:14:34,834 - Epoch: [69][ 490/ 1236] Overall Loss 0.310766 Objective Loss 0.310766 LR 0.001000 Time 0.021780 +2023-10-05 21:14:35,037 - Epoch: [69][ 500/ 1236] Overall Loss 0.310758 Objective Loss 0.310758 LR 0.001000 Time 0.021750 +2023-10-05 21:14:35,240 - Epoch: [69][ 510/ 1236] Overall Loss 0.310902 Objective Loss 0.310902 LR 0.001000 Time 0.021721 +2023-10-05 21:14:35,444 - Epoch: [69][ 520/ 1236] Overall Loss 0.311284 Objective Loss 0.311284 LR 0.001000 Time 0.021694 +2023-10-05 21:14:35,646 - Epoch: [69][ 530/ 1236] Overall Loss 0.311000 Objective Loss 0.311000 LR 0.001000 Time 0.021666 +2023-10-05 21:14:35,850 - Epoch: [69][ 540/ 1236] Overall Loss 0.311059 Objective Loss 0.311059 LR 0.001000 Time 0.021641 +2023-10-05 21:14:36,051 - Epoch: [69][ 550/ 1236] Overall Loss 0.311153 Objective Loss 0.311153 LR 0.001000 Time 0.021613 +2023-10-05 21:14:36,255 - Epoch: [69][ 560/ 1236] Overall Loss 0.311258 Objective Loss 0.311258 LR 0.001000 Time 0.021591 +2023-10-05 21:14:36,457 - Epoch: [69][ 570/ 1236] Overall Loss 0.311454 Objective Loss 0.311454 LR 0.001000 Time 0.021565 +2023-10-05 21:14:36,660 - Epoch: [69][ 580/ 1236] Overall Loss 0.311427 Objective Loss 0.311427 LR 0.001000 Time 0.021543 +2023-10-05 21:14:36,862 - Epoch: [69][ 590/ 1236] Overall Loss 0.311719 Objective Loss 0.311719 LR 0.001000 Time 0.021519 +2023-10-05 21:14:37,066 - Epoch: [69][ 600/ 1236] Overall Loss 0.311619 Objective Loss 0.311619 LR 0.001000 Time 0.021499 +2023-10-05 21:14:37,267 - Epoch: [69][ 610/ 1236] Overall Loss 0.311462 Objective Loss 0.311462 LR 0.001000 Time 0.021477 +2023-10-05 21:14:37,472 - Epoch: [69][ 620/ 1236] Overall Loss 0.311029 Objective Loss 0.311029 LR 0.001000 Time 0.021460 +2023-10-05 21:14:37,675 - Epoch: [69][ 630/ 1236] Overall Loss 0.311064 Objective Loss 0.311064 LR 0.001000 Time 0.021441 +2023-10-05 21:14:37,882 - Epoch: [69][ 640/ 1236] Overall Loss 0.311148 Objective Loss 0.311148 LR 0.001000 Time 0.021429 +2023-10-05 21:14:38,083 - Epoch: [69][ 650/ 1236] Overall Loss 0.311374 Objective Loss 0.311374 LR 0.001000 Time 0.021408 +2023-10-05 21:14:38,288 - Epoch: [69][ 660/ 1236] Overall Loss 0.312270 Objective Loss 0.312270 LR 0.001000 Time 0.021393 +2023-10-05 21:14:38,493 - Epoch: [69][ 670/ 1236] Overall Loss 0.312599 Objective Loss 0.312599 LR 0.001000 Time 0.021380 +2023-10-05 21:14:38,696 - Epoch: [69][ 680/ 1236] Overall Loss 0.313147 Objective Loss 0.313147 LR 0.001000 Time 0.021364 +2023-10-05 21:14:38,901 - Epoch: [69][ 690/ 1236] Overall Loss 0.313627 Objective Loss 0.313627 LR 0.001000 Time 0.021350 +2023-10-05 21:14:39,105 - Epoch: [69][ 700/ 1236] Overall Loss 0.313702 Objective Loss 0.313702 LR 0.001000 Time 0.021335 +2023-10-05 21:14:39,306 - Epoch: [69][ 710/ 1236] Overall Loss 0.313841 Objective Loss 0.313841 LR 0.001000 Time 0.021318 +2023-10-05 21:14:39,511 - Epoch: [69][ 720/ 1236] Overall Loss 0.313905 Objective Loss 0.313905 LR 0.001000 Time 0.021305 +2023-10-05 21:14:39,712 - Epoch: [69][ 730/ 1236] Overall Loss 0.314124 Objective Loss 0.314124 LR 0.001000 Time 0.021289 +2023-10-05 21:14:39,916 - Epoch: [69][ 740/ 1236] Overall Loss 0.313648 Objective Loss 0.313648 LR 0.001000 Time 0.021277 +2023-10-05 21:14:40,118 - Epoch: [69][ 750/ 1236] Overall Loss 0.313982 Objective Loss 0.313982 LR 0.001000 Time 0.021261 +2023-10-05 21:14:40,322 - Epoch: [69][ 760/ 1236] Overall Loss 0.314264 Objective Loss 0.314264 LR 0.001000 Time 0.021249 +2023-10-05 21:14:40,523 - Epoch: [69][ 770/ 1236] Overall Loss 0.314407 Objective Loss 0.314407 LR 0.001000 Time 0.021234 +2023-10-05 21:14:40,728 - Epoch: [69][ 780/ 1236] Overall Loss 0.314313 Objective Loss 0.314313 LR 0.001000 Time 0.021224 +2023-10-05 21:14:40,929 - Epoch: [69][ 790/ 1236] Overall Loss 0.314295 Objective Loss 0.314295 LR 0.001000 Time 0.021209 +2023-10-05 21:14:41,134 - Epoch: [69][ 800/ 1236] Overall Loss 0.314265 Objective Loss 0.314265 LR 0.001000 Time 0.021200 +2023-10-05 21:14:41,335 - Epoch: [69][ 810/ 1236] Overall Loss 0.314629 Objective Loss 0.314629 LR 0.001000 Time 0.021186 +2023-10-05 21:14:41,539 - Epoch: [69][ 820/ 1236] Overall Loss 0.314625 Objective Loss 0.314625 LR 0.001000 Time 0.021176 +2023-10-05 21:14:41,740 - Epoch: [69][ 830/ 1236] Overall Loss 0.314566 Objective Loss 0.314566 LR 0.001000 Time 0.021163 +2023-10-05 21:14:41,945 - Epoch: [69][ 840/ 1236] Overall Loss 0.314327 Objective Loss 0.314327 LR 0.001000 Time 0.021154 +2023-10-05 21:14:42,146 - Epoch: [69][ 850/ 1236] Overall Loss 0.314272 Objective Loss 0.314272 LR 0.001000 Time 0.021141 +2023-10-05 21:14:42,351 - Epoch: [69][ 860/ 1236] Overall Loss 0.314300 Objective Loss 0.314300 LR 0.001000 Time 0.021133 +2023-10-05 21:14:42,553 - Epoch: [69][ 870/ 1236] Overall Loss 0.314128 Objective Loss 0.314128 LR 0.001000 Time 0.021122 +2023-10-05 21:14:42,757 - Epoch: [69][ 880/ 1236] Overall Loss 0.313729 Objective Loss 0.313729 LR 0.001000 Time 0.021113 +2023-10-05 21:14:42,958 - Epoch: [69][ 890/ 1236] Overall Loss 0.313543 Objective Loss 0.313543 LR 0.001000 Time 0.021102 +2023-10-05 21:14:43,163 - Epoch: [69][ 900/ 1236] Overall Loss 0.313999 Objective Loss 0.313999 LR 0.001000 Time 0.021094 +2023-10-05 21:14:43,364 - Epoch: [69][ 910/ 1236] Overall Loss 0.314300 Objective Loss 0.314300 LR 0.001000 Time 0.021084 +2023-10-05 21:14:43,568 - Epoch: [69][ 920/ 1236] Overall Loss 0.314243 Objective Loss 0.314243 LR 0.001000 Time 0.021076 +2023-10-05 21:14:43,770 - Epoch: [69][ 930/ 1236] Overall Loss 0.314333 Objective Loss 0.314333 LR 0.001000 Time 0.021066 +2023-10-05 21:14:43,974 - Epoch: [69][ 940/ 1236] Overall Loss 0.314062 Objective Loss 0.314062 LR 0.001000 Time 0.021058 +2023-10-05 21:14:44,175 - Epoch: [69][ 950/ 1236] Overall Loss 0.314018 Objective Loss 0.314018 LR 0.001000 Time 0.021048 +2023-10-05 21:14:44,379 - Epoch: [69][ 960/ 1236] Overall Loss 0.314380 Objective Loss 0.314380 LR 0.001000 Time 0.021041 +2023-10-05 21:14:44,580 - Epoch: [69][ 970/ 1236] Overall Loss 0.314428 Objective Loss 0.314428 LR 0.001000 Time 0.021031 +2023-10-05 21:14:44,784 - Epoch: [69][ 980/ 1236] Overall Loss 0.314973 Objective Loss 0.314973 LR 0.001000 Time 0.021024 +2023-10-05 21:14:44,986 - Epoch: [69][ 990/ 1236] Overall Loss 0.314994 Objective Loss 0.314994 LR 0.001000 Time 0.021015 +2023-10-05 21:14:45,191 - Epoch: [69][ 1000/ 1236] Overall Loss 0.315009 Objective Loss 0.315009 LR 0.001000 Time 0.021009 +2023-10-05 21:14:45,392 - Epoch: [69][ 1010/ 1236] Overall Loss 0.315263 Objective Loss 0.315263 LR 0.001000 Time 0.021000 +2023-10-05 21:14:45,597 - Epoch: [69][ 1020/ 1236] Overall Loss 0.315712 Objective Loss 0.315712 LR 0.001000 Time 0.020995 +2023-10-05 21:14:45,798 - Epoch: [69][ 1030/ 1236] Overall Loss 0.315879 Objective Loss 0.315879 LR 0.001000 Time 0.020986 +2023-10-05 21:14:46,002 - Epoch: [69][ 1040/ 1236] Overall Loss 0.315959 Objective Loss 0.315959 LR 0.001000 Time 0.020980 +2023-10-05 21:14:46,205 - Epoch: [69][ 1050/ 1236] Overall Loss 0.316309 Objective Loss 0.316309 LR 0.001000 Time 0.020973 +2023-10-05 21:14:46,409 - Epoch: [69][ 1060/ 1236] Overall Loss 0.316598 Objective Loss 0.316598 LR 0.001000 Time 0.020967 +2023-10-05 21:14:46,610 - Epoch: [69][ 1070/ 1236] Overall Loss 0.316458 Objective Loss 0.316458 LR 0.001000 Time 0.020959 +2023-10-05 21:14:46,814 - Epoch: [69][ 1080/ 1236] Overall Loss 0.316577 Objective Loss 0.316577 LR 0.001000 Time 0.020953 +2023-10-05 21:14:47,016 - Epoch: [69][ 1090/ 1236] Overall Loss 0.316671 Objective Loss 0.316671 LR 0.001000 Time 0.020946 +2023-10-05 21:14:47,220 - Epoch: [69][ 1100/ 1236] Overall Loss 0.316532 Objective Loss 0.316532 LR 0.001000 Time 0.020941 +2023-10-05 21:14:47,421 - Epoch: [69][ 1110/ 1236] Overall Loss 0.316666 Objective Loss 0.316666 LR 0.001000 Time 0.020933 +2023-10-05 21:14:47,625 - Epoch: [69][ 1120/ 1236] Overall Loss 0.316968 Objective Loss 0.316968 LR 0.001000 Time 0.020928 +2023-10-05 21:14:47,827 - Epoch: [69][ 1130/ 1236] Overall Loss 0.317200 Objective Loss 0.317200 LR 0.001000 Time 0.020921 +2023-10-05 21:14:48,031 - Epoch: [69][ 1140/ 1236] Overall Loss 0.316838 Objective Loss 0.316838 LR 0.001000 Time 0.020916 +2023-10-05 21:14:48,233 - Epoch: [69][ 1150/ 1236] Overall Loss 0.316642 Objective Loss 0.316642 LR 0.001000 Time 0.020910 +2023-10-05 21:14:48,437 - Epoch: [69][ 1160/ 1236] Overall Loss 0.316514 Objective Loss 0.316514 LR 0.001000 Time 0.020905 +2023-10-05 21:14:48,639 - Epoch: [69][ 1170/ 1236] Overall Loss 0.316430 Objective Loss 0.316430 LR 0.001000 Time 0.020898 +2023-10-05 21:14:48,843 - Epoch: [69][ 1180/ 1236] Overall Loss 0.316349 Objective Loss 0.316349 LR 0.001000 Time 0.020894 +2023-10-05 21:14:49,044 - Epoch: [69][ 1190/ 1236] Overall Loss 0.316337 Objective Loss 0.316337 LR 0.001000 Time 0.020887 +2023-10-05 21:14:49,249 - Epoch: [69][ 1200/ 1236] Overall Loss 0.316300 Objective Loss 0.316300 LR 0.001000 Time 0.020883 +2023-10-05 21:14:49,450 - Epoch: [69][ 1210/ 1236] Overall Loss 0.316235 Objective Loss 0.316235 LR 0.001000 Time 0.020877 +2023-10-05 21:14:49,654 - Epoch: [69][ 1220/ 1236] Overall Loss 0.316271 Objective Loss 0.316271 LR 0.001000 Time 0.020872 +2023-10-05 21:14:49,910 - Epoch: [69][ 1230/ 1236] Overall Loss 0.316180 Objective Loss 0.316180 LR 0.001000 Time 0.020910 +2023-10-05 21:14:50,029 - Epoch: [69][ 1236/ 1236] Overall Loss 0.316210 Objective Loss 0.316210 Top1 82.892057 Top5 97.759674 LR 0.001000 Time 0.020905 +2023-10-05 21:14:50,162 - --- validate (epoch=69)----------- +2023-10-05 21:14:50,162 - 29943 samples (256 per mini-batch) +2023-10-05 21:14:50,623 - Epoch: [69][ 10/ 117] Loss 0.366252 Top1 80.742188 Top5 96.562500 +2023-10-05 21:14:50,777 - Epoch: [69][ 20/ 117] Loss 0.355720 Top1 80.937500 Top5 97.109375 +2023-10-05 21:14:50,929 - Epoch: [69][ 30/ 117] Loss 0.348953 Top1 81.302083 Top5 97.369792 +2023-10-05 21:14:51,080 - Epoch: [69][ 40/ 117] Loss 0.358814 Top1 80.986328 Top5 97.236328 +2023-10-05 21:14:51,232 - Epoch: [69][ 50/ 117] Loss 0.354057 Top1 81.023438 Top5 97.289062 +2023-10-05 21:14:51,384 - Epoch: [69][ 60/ 117] Loss 0.354930 Top1 80.976562 Top5 97.220052 +2023-10-05 21:14:51,540 - Epoch: [69][ 70/ 117] Loss 0.354204 Top1 80.926339 Top5 97.276786 +2023-10-05 21:14:51,697 - Epoch: [69][ 80/ 117] Loss 0.355871 Top1 80.937500 Top5 97.260742 +2023-10-05 21:14:51,853 - Epoch: [69][ 90/ 117] Loss 0.357657 Top1 80.989583 Top5 97.256944 +2023-10-05 21:14:52,011 - Epoch: [69][ 100/ 117] Loss 0.357224 Top1 80.980469 Top5 97.261719 +2023-10-05 21:14:52,176 - Epoch: [69][ 110/ 117] Loss 0.356868 Top1 81.004972 Top5 97.251420 +2023-10-05 21:14:52,262 - Epoch: [69][ 117/ 117] Loss 0.351675 Top1 81.167552 Top5 97.311559 +2023-10-05 21:14:52,368 - ==> Top1: 81.168 Top5: 97.312 Loss: 0.352 + +2023-10-05 21:14:52,369 - ==> Confusion: +[[ 920 5 3 2 17 3 0 0 4 69 0 0 0 3 7 4 2 0 0 0 11] + [ 1 1044 3 2 15 15 0 28 2 0 5 2 0 0 2 4 0 0 5 1 2] + [ 8 0 959 19 4 0 17 8 0 2 4 3 7 1 2 7 1 0 5 3 6] + [ 1 0 21 938 1 3 0 1 2 2 9 0 8 6 55 6 0 4 21 1 10] + [ 15 6 1 0 985 3 0 1 0 11 0 1 3 0 7 6 8 2 0 0 1] + [ 6 48 2 0 6 973 1 23 1 4 6 7 0 10 7 2 2 0 2 4 12] + [ 0 6 53 0 0 0 1090 9 0 0 6 2 1 0 1 7 0 1 2 5 8] + [ 4 24 19 0 4 40 5 1051 0 1 5 7 3 2 0 6 2 0 34 5 6] + [ 22 2 0 0 1 3 0 1 933 54 11 2 1 16 31 5 0 0 3 0 4] + [ 92 0 1 0 11 5 1 1 26 930 1 1 0 30 5 10 0 0 0 0 5] + [ 3 7 15 8 3 0 4 3 15 3 960 4 0 10 3 2 1 0 4 0 8] + [ 1 0 1 0 2 17 0 2 1 0 0 944 22 3 1 6 5 17 0 11 2] + [ 5 1 4 9 3 4 0 2 1 0 0 31 964 2 8 8 1 12 4 0 9] + [ 2 0 1 0 8 11 0 0 9 15 7 7 2 1043 3 2 0 0 0 1 8] + [ 16 0 2 5 13 0 0 0 12 4 2 1 0 1 1025 0 1 0 9 1 9] + [ 2 3 3 3 4 0 1 0 0 0 0 7 9 1 0 1063 14 9 0 8 7] + [ 2 18 3 1 14 4 0 1 1 0 0 7 3 3 2 17 1076 0 1 2 6] + [ 1 0 2 0 0 1 0 0 0 0 0 3 27 3 2 9 0 984 3 1 2] + [ 2 9 8 15 3 2 0 27 4 0 4 2 2 0 23 0 1 0 957 1 8] + [ 0 5 4 3 4 6 7 14 0 0 1 19 6 3 1 7 8 3 1 1053 7] + [ 194 239 185 103 192 178 44 95 97 97 198 134 408 349 226 118 146 71 204 216 4411]] + +2023-10-05 21:14:52,370 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:14:52,370 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:14:52,376 - + +2023-10-05 21:14:52,376 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:14:53,498 - Epoch: [70][ 10/ 1236] Overall Loss 0.297418 Objective Loss 0.297418 LR 0.001000 Time 0.112162 +2023-10-05 21:14:53,700 - Epoch: [70][ 20/ 1236] Overall Loss 0.306751 Objective Loss 0.306751 LR 0.001000 Time 0.066181 +2023-10-05 21:14:53,900 - Epoch: [70][ 30/ 1236] Overall Loss 0.309873 Objective Loss 0.309873 LR 0.001000 Time 0.050781 +2023-10-05 21:14:54,103 - Epoch: [70][ 40/ 1236] Overall Loss 0.311333 Objective Loss 0.311333 LR 0.001000 Time 0.043136 +2023-10-05 21:14:54,303 - Epoch: [70][ 50/ 1236] Overall Loss 0.315646 Objective Loss 0.315646 LR 0.001000 Time 0.038507 +2023-10-05 21:14:54,505 - Epoch: [70][ 60/ 1236] Overall Loss 0.310535 Objective Loss 0.310535 LR 0.001000 Time 0.035455 +2023-10-05 21:14:54,705 - Epoch: [70][ 70/ 1236] Overall Loss 0.308529 Objective Loss 0.308529 LR 0.001000 Time 0.033244 +2023-10-05 21:14:54,908 - Epoch: [70][ 80/ 1236] Overall Loss 0.308472 Objective Loss 0.308472 LR 0.001000 Time 0.031611 +2023-10-05 21:14:55,108 - Epoch: [70][ 90/ 1236] Overall Loss 0.307770 Objective Loss 0.307770 LR 0.001000 Time 0.030324 +2023-10-05 21:14:55,311 - Epoch: [70][ 100/ 1236] Overall Loss 0.305294 Objective Loss 0.305294 LR 0.001000 Time 0.029313 +2023-10-05 21:14:55,511 - Epoch: [70][ 110/ 1236] Overall Loss 0.305181 Objective Loss 0.305181 LR 0.001000 Time 0.028464 +2023-10-05 21:14:55,713 - Epoch: [70][ 120/ 1236] Overall Loss 0.307913 Objective Loss 0.307913 LR 0.001000 Time 0.027771 +2023-10-05 21:14:55,913 - Epoch: [70][ 130/ 1236] Overall Loss 0.306821 Objective Loss 0.306821 LR 0.001000 Time 0.027173 +2023-10-05 21:14:56,115 - Epoch: [70][ 140/ 1236] Overall Loss 0.309410 Objective Loss 0.309410 LR 0.001000 Time 0.026674 +2023-10-05 21:14:56,316 - Epoch: [70][ 150/ 1236] Overall Loss 0.308971 Objective Loss 0.308971 LR 0.001000 Time 0.026229 +2023-10-05 21:14:56,518 - Epoch: [70][ 160/ 1236] Overall Loss 0.308229 Objective Loss 0.308229 LR 0.001000 Time 0.025851 +2023-10-05 21:14:56,718 - Epoch: [70][ 170/ 1236] Overall Loss 0.307630 Objective Loss 0.307630 LR 0.001000 Time 0.025506 +2023-10-05 21:14:56,920 - Epoch: [70][ 180/ 1236] Overall Loss 0.307801 Objective Loss 0.307801 LR 0.001000 Time 0.025211 +2023-10-05 21:14:57,121 - Epoch: [70][ 190/ 1236] Overall Loss 0.308053 Objective Loss 0.308053 LR 0.001000 Time 0.024938 +2023-10-05 21:14:57,323 - Epoch: [70][ 200/ 1236] Overall Loss 0.309274 Objective Loss 0.309274 LR 0.001000 Time 0.024699 +2023-10-05 21:14:57,523 - Epoch: [70][ 210/ 1236] Overall Loss 0.310866 Objective Loss 0.310866 LR 0.001000 Time 0.024475 +2023-10-05 21:14:57,725 - Epoch: [70][ 220/ 1236] Overall Loss 0.309851 Objective Loss 0.309851 LR 0.001000 Time 0.024280 +2023-10-05 21:14:57,926 - Epoch: [70][ 230/ 1236] Overall Loss 0.310395 Objective Loss 0.310395 LR 0.001000 Time 0.024095 +2023-10-05 21:14:58,129 - Epoch: [70][ 240/ 1236] Overall Loss 0.310092 Objective Loss 0.310092 LR 0.001000 Time 0.023937 +2023-10-05 21:14:58,329 - Epoch: [70][ 250/ 1236] Overall Loss 0.309912 Objective Loss 0.309912 LR 0.001000 Time 0.023776 +2023-10-05 21:14:58,532 - Epoch: [70][ 260/ 1236] Overall Loss 0.310410 Objective Loss 0.310410 LR 0.001000 Time 0.023642 +2023-10-05 21:14:58,736 - Epoch: [70][ 270/ 1236] Overall Loss 0.310287 Objective Loss 0.310287 LR 0.001000 Time 0.023522 +2023-10-05 21:14:58,940 - Epoch: [70][ 280/ 1236] Overall Loss 0.310574 Objective Loss 0.310574 LR 0.001000 Time 0.023409 +2023-10-05 21:14:59,141 - Epoch: [70][ 290/ 1236] Overall Loss 0.311485 Objective Loss 0.311485 LR 0.001000 Time 0.023293 +2023-10-05 21:14:59,346 - Epoch: [70][ 300/ 1236] Overall Loss 0.311951 Objective Loss 0.311951 LR 0.001000 Time 0.023198 +2023-10-05 21:14:59,549 - Epoch: [70][ 310/ 1236] Overall Loss 0.312714 Objective Loss 0.312714 LR 0.001000 Time 0.023105 +2023-10-05 21:14:59,754 - Epoch: [70][ 320/ 1236] Overall Loss 0.313964 Objective Loss 0.313964 LR 0.001000 Time 0.023022 +2023-10-05 21:14:59,958 - Epoch: [70][ 330/ 1236] Overall Loss 0.313404 Objective Loss 0.313404 LR 0.001000 Time 0.022941 +2023-10-05 21:15:00,159 - Epoch: [70][ 340/ 1236] Overall Loss 0.315432 Objective Loss 0.315432 LR 0.001000 Time 0.022857 +2023-10-05 21:15:00,358 - Epoch: [70][ 350/ 1236] Overall Loss 0.316502 Objective Loss 0.316502 LR 0.001000 Time 0.022771 +2023-10-05 21:15:00,558 - Epoch: [70][ 360/ 1236] Overall Loss 0.316817 Objective Loss 0.316817 LR 0.001000 Time 0.022694 +2023-10-05 21:15:00,759 - Epoch: [70][ 370/ 1236] Overall Loss 0.317219 Objective Loss 0.317219 LR 0.001000 Time 0.022624 +2023-10-05 21:15:00,963 - Epoch: [70][ 380/ 1236] Overall Loss 0.316402 Objective Loss 0.316402 LR 0.001000 Time 0.022564 +2023-10-05 21:15:01,166 - Epoch: [70][ 390/ 1236] Overall Loss 0.316283 Objective Loss 0.316283 LR 0.001000 Time 0.022506 +2023-10-05 21:15:01,369 - Epoch: [70][ 400/ 1236] Overall Loss 0.316383 Objective Loss 0.316383 LR 0.001000 Time 0.022450 +2023-10-05 21:15:01,573 - Epoch: [70][ 410/ 1236] Overall Loss 0.316493 Objective Loss 0.316493 LR 0.001000 Time 0.022398 +2023-10-05 21:15:01,776 - Epoch: [70][ 420/ 1236] Overall Loss 0.315444 Objective Loss 0.315444 LR 0.001000 Time 0.022349 +2023-10-05 21:15:01,979 - Epoch: [70][ 430/ 1236] Overall Loss 0.315943 Objective Loss 0.315943 LR 0.001000 Time 0.022299 +2023-10-05 21:15:02,181 - Epoch: [70][ 440/ 1236] Overall Loss 0.316045 Objective Loss 0.316045 LR 0.001000 Time 0.022252 +2023-10-05 21:15:02,384 - Epoch: [70][ 450/ 1236] Overall Loss 0.316706 Objective Loss 0.316706 LR 0.001000 Time 0.022207 +2023-10-05 21:15:02,587 - Epoch: [70][ 460/ 1236] Overall Loss 0.317489 Objective Loss 0.317489 LR 0.001000 Time 0.022164 +2023-10-05 21:15:02,789 - Epoch: [70][ 470/ 1236] Overall Loss 0.318499 Objective Loss 0.318499 LR 0.001000 Time 0.022123 +2023-10-05 21:15:02,992 - Epoch: [70][ 480/ 1236] Overall Loss 0.318872 Objective Loss 0.318872 LR 0.001000 Time 0.022084 +2023-10-05 21:15:03,194 - Epoch: [70][ 490/ 1236] Overall Loss 0.318871 Objective Loss 0.318871 LR 0.001000 Time 0.022045 +2023-10-05 21:15:03,396 - Epoch: [70][ 500/ 1236] Overall Loss 0.318589 Objective Loss 0.318589 LR 0.001000 Time 0.022009 +2023-10-05 21:15:03,599 - Epoch: [70][ 510/ 1236] Overall Loss 0.318731 Objective Loss 0.318731 LR 0.001000 Time 0.021974 +2023-10-05 21:15:03,802 - Epoch: [70][ 520/ 1236] Overall Loss 0.318742 Objective Loss 0.318742 LR 0.001000 Time 0.021940 +2023-10-05 21:15:04,004 - Epoch: [70][ 530/ 1236] Overall Loss 0.319267 Objective Loss 0.319267 LR 0.001000 Time 0.021908 +2023-10-05 21:15:04,207 - Epoch: [70][ 540/ 1236] Overall Loss 0.319483 Objective Loss 0.319483 LR 0.001000 Time 0.021877 +2023-10-05 21:15:04,409 - Epoch: [70][ 550/ 1236] Overall Loss 0.319529 Objective Loss 0.319529 LR 0.001000 Time 0.021847 +2023-10-05 21:15:04,612 - Epoch: [70][ 560/ 1236] Overall Loss 0.320222 Objective Loss 0.320222 LR 0.001000 Time 0.021819 +2023-10-05 21:15:04,814 - Epoch: [70][ 570/ 1236] Overall Loss 0.319784 Objective Loss 0.319784 LR 0.001000 Time 0.021790 +2023-10-05 21:15:05,017 - Epoch: [70][ 580/ 1236] Overall Loss 0.319421 Objective Loss 0.319421 LR 0.001000 Time 0.021764 +2023-10-05 21:15:05,220 - Epoch: [70][ 590/ 1236] Overall Loss 0.319378 Objective Loss 0.319378 LR 0.001000 Time 0.021738 +2023-10-05 21:15:05,422 - Epoch: [70][ 600/ 1236] Overall Loss 0.319117 Objective Loss 0.319117 LR 0.001000 Time 0.021713 +2023-10-05 21:15:05,625 - Epoch: [70][ 610/ 1236] Overall Loss 0.318416 Objective Loss 0.318416 LR 0.001000 Time 0.021688 +2023-10-05 21:15:05,828 - Epoch: [70][ 620/ 1236] Overall Loss 0.318678 Objective Loss 0.318678 LR 0.001000 Time 0.021665 +2023-10-05 21:15:06,030 - Epoch: [70][ 630/ 1236] Overall Loss 0.318751 Objective Loss 0.318751 LR 0.001000 Time 0.021642 +2023-10-05 21:15:06,233 - Epoch: [70][ 640/ 1236] Overall Loss 0.318674 Objective Loss 0.318674 LR 0.001000 Time 0.021620 +2023-10-05 21:15:06,435 - Epoch: [70][ 650/ 1236] Overall Loss 0.318530 Objective Loss 0.318530 LR 0.001000 Time 0.021599 +2023-10-05 21:15:06,638 - Epoch: [70][ 660/ 1236] Overall Loss 0.318081 Objective Loss 0.318081 LR 0.001000 Time 0.021578 +2023-10-05 21:15:06,841 - Epoch: [70][ 670/ 1236] Overall Loss 0.318147 Objective Loss 0.318147 LR 0.001000 Time 0.021558 +2023-10-05 21:15:07,044 - Epoch: [70][ 680/ 1236] Overall Loss 0.317454 Objective Loss 0.317454 LR 0.001000 Time 0.021539 +2023-10-05 21:15:07,246 - Epoch: [70][ 690/ 1236] Overall Loss 0.316859 Objective Loss 0.316859 LR 0.001000 Time 0.021520 +2023-10-05 21:15:07,449 - Epoch: [70][ 700/ 1236] Overall Loss 0.316894 Objective Loss 0.316894 LR 0.001000 Time 0.021502 +2023-10-05 21:15:07,652 - Epoch: [70][ 710/ 1236] Overall Loss 0.316349 Objective Loss 0.316349 LR 0.001000 Time 0.021484 +2023-10-05 21:15:07,855 - Epoch: [70][ 720/ 1236] Overall Loss 0.316089 Objective Loss 0.316089 LR 0.001000 Time 0.021467 +2023-10-05 21:15:08,057 - Epoch: [70][ 730/ 1236] Overall Loss 0.316056 Objective Loss 0.316056 LR 0.001000 Time 0.021450 +2023-10-05 21:15:08,260 - Epoch: [70][ 740/ 1236] Overall Loss 0.315811 Objective Loss 0.315811 LR 0.001000 Time 0.021434 +2023-10-05 21:15:08,462 - Epoch: [70][ 750/ 1236] Overall Loss 0.315526 Objective Loss 0.315526 LR 0.001000 Time 0.021417 +2023-10-05 21:15:08,664 - Epoch: [70][ 760/ 1236] Overall Loss 0.315830 Objective Loss 0.315830 LR 0.001000 Time 0.021402 +2023-10-05 21:15:08,867 - Epoch: [70][ 770/ 1236] Overall Loss 0.315769 Objective Loss 0.315769 LR 0.001000 Time 0.021386 +2023-10-05 21:15:09,070 - Epoch: [70][ 780/ 1236] Overall Loss 0.315513 Objective Loss 0.315513 LR 0.001000 Time 0.021372 +2023-10-05 21:15:09,272 - Epoch: [70][ 790/ 1236] Overall Loss 0.315477 Objective Loss 0.315477 LR 0.001000 Time 0.021357 +2023-10-05 21:15:09,475 - Epoch: [70][ 800/ 1236] Overall Loss 0.315565 Objective Loss 0.315565 LR 0.001000 Time 0.021343 +2023-10-05 21:15:09,677 - Epoch: [70][ 810/ 1236] Overall Loss 0.315554 Objective Loss 0.315554 LR 0.001000 Time 0.021329 +2023-10-05 21:15:09,880 - Epoch: [70][ 820/ 1236] Overall Loss 0.316049 Objective Loss 0.316049 LR 0.001000 Time 0.021316 +2023-10-05 21:15:10,082 - Epoch: [70][ 830/ 1236] Overall Loss 0.316623 Objective Loss 0.316623 LR 0.001000 Time 0.021302 +2023-10-05 21:15:10,285 - Epoch: [70][ 840/ 1236] Overall Loss 0.316438 Objective Loss 0.316438 LR 0.001000 Time 0.021290 +2023-10-05 21:15:10,487 - Epoch: [70][ 850/ 1236] Overall Loss 0.316352 Objective Loss 0.316352 LR 0.001000 Time 0.021277 +2023-10-05 21:15:10,690 - Epoch: [70][ 860/ 1236] Overall Loss 0.316423 Objective Loss 0.316423 LR 0.001000 Time 0.021265 +2023-10-05 21:15:10,892 - Epoch: [70][ 870/ 1236] Overall Loss 0.316578 Objective Loss 0.316578 LR 0.001000 Time 0.021253 +2023-10-05 21:15:11,095 - Epoch: [70][ 880/ 1236] Overall Loss 0.316595 Objective Loss 0.316595 LR 0.001000 Time 0.021242 +2023-10-05 21:15:11,298 - Epoch: [70][ 890/ 1236] Overall Loss 0.316543 Objective Loss 0.316543 LR 0.001000 Time 0.021230 +2023-10-05 21:15:11,501 - Epoch: [70][ 900/ 1236] Overall Loss 0.316820 Objective Loss 0.316820 LR 0.001000 Time 0.021219 +2023-10-05 21:15:11,703 - Epoch: [70][ 910/ 1236] Overall Loss 0.316743 Objective Loss 0.316743 LR 0.001000 Time 0.021208 +2023-10-05 21:15:11,906 - Epoch: [70][ 920/ 1236] Overall Loss 0.317101 Objective Loss 0.317101 LR 0.001000 Time 0.021198 +2023-10-05 21:15:12,108 - Epoch: [70][ 930/ 1236] Overall Loss 0.316966 Objective Loss 0.316966 LR 0.001000 Time 0.021187 +2023-10-05 21:15:12,311 - Epoch: [70][ 940/ 1236] Overall Loss 0.317124 Objective Loss 0.317124 LR 0.001000 Time 0.021178 +2023-10-05 21:15:12,514 - Epoch: [70][ 950/ 1236] Overall Loss 0.317180 Objective Loss 0.317180 LR 0.001000 Time 0.021168 +2023-10-05 21:15:12,717 - Epoch: [70][ 960/ 1236] Overall Loss 0.317043 Objective Loss 0.317043 LR 0.001000 Time 0.021158 +2023-10-05 21:15:12,919 - Epoch: [70][ 970/ 1236] Overall Loss 0.317094 Objective Loss 0.317094 LR 0.001000 Time 0.021148 +2023-10-05 21:15:13,122 - Epoch: [70][ 980/ 1236] Overall Loss 0.317250 Objective Loss 0.317250 LR 0.001000 Time 0.021140 +2023-10-05 21:15:13,325 - Epoch: [70][ 990/ 1236] Overall Loss 0.317401 Objective Loss 0.317401 LR 0.001000 Time 0.021130 +2023-10-05 21:15:13,528 - Epoch: [70][ 1000/ 1236] Overall Loss 0.317565 Objective Loss 0.317565 LR 0.001000 Time 0.021122 +2023-10-05 21:15:13,731 - Epoch: [70][ 1010/ 1236] Overall Loss 0.317752 Objective Loss 0.317752 LR 0.001000 Time 0.021113 +2023-10-05 21:15:13,934 - Epoch: [70][ 1020/ 1236] Overall Loss 0.317732 Objective Loss 0.317732 LR 0.001000 Time 0.021105 +2023-10-05 21:15:14,136 - Epoch: [70][ 1030/ 1236] Overall Loss 0.317855 Objective Loss 0.317855 LR 0.001000 Time 0.021097 +2023-10-05 21:15:14,339 - Epoch: [70][ 1040/ 1236] Overall Loss 0.318196 Objective Loss 0.318196 LR 0.001000 Time 0.021089 +2023-10-05 21:15:14,542 - Epoch: [70][ 1050/ 1236] Overall Loss 0.318255 Objective Loss 0.318255 LR 0.001000 Time 0.021080 +2023-10-05 21:15:14,745 - Epoch: [70][ 1060/ 1236] Overall Loss 0.318410 Objective Loss 0.318410 LR 0.001000 Time 0.021073 +2023-10-05 21:15:14,947 - Epoch: [70][ 1070/ 1236] Overall Loss 0.318280 Objective Loss 0.318280 LR 0.001000 Time 0.021065 +2023-10-05 21:15:15,150 - Epoch: [70][ 1080/ 1236] Overall Loss 0.318376 Objective Loss 0.318376 LR 0.001000 Time 0.021057 +2023-10-05 21:15:15,353 - Epoch: [70][ 1090/ 1236] Overall Loss 0.318579 Objective Loss 0.318579 LR 0.001000 Time 0.021050 +2023-10-05 21:15:15,556 - Epoch: [70][ 1100/ 1236] Overall Loss 0.318666 Objective Loss 0.318666 LR 0.001000 Time 0.021043 +2023-10-05 21:15:15,758 - Epoch: [70][ 1110/ 1236] Overall Loss 0.318994 Objective Loss 0.318994 LR 0.001000 Time 0.021035 +2023-10-05 21:15:15,961 - Epoch: [70][ 1120/ 1236] Overall Loss 0.318984 Objective Loss 0.318984 LR 0.001000 Time 0.021028 +2023-10-05 21:15:16,163 - Epoch: [70][ 1130/ 1236] Overall Loss 0.319230 Objective Loss 0.319230 LR 0.001000 Time 0.021021 +2023-10-05 21:15:16,366 - Epoch: [70][ 1140/ 1236] Overall Loss 0.319164 Objective Loss 0.319164 LR 0.001000 Time 0.021014 +2023-10-05 21:15:16,569 - Epoch: [70][ 1150/ 1236] Overall Loss 0.318768 Objective Loss 0.318768 LR 0.001000 Time 0.021007 +2023-10-05 21:15:16,772 - Epoch: [70][ 1160/ 1236] Overall Loss 0.318606 Objective Loss 0.318606 LR 0.001000 Time 0.021001 +2023-10-05 21:15:16,974 - Epoch: [70][ 1170/ 1236] Overall Loss 0.318527 Objective Loss 0.318527 LR 0.001000 Time 0.020994 +2023-10-05 21:15:17,177 - Epoch: [70][ 1180/ 1236] Overall Loss 0.318558 Objective Loss 0.318558 LR 0.001000 Time 0.020988 +2023-10-05 21:15:17,379 - Epoch: [70][ 1190/ 1236] Overall Loss 0.318725 Objective Loss 0.318725 LR 0.001000 Time 0.020981 +2023-10-05 21:15:17,582 - Epoch: [70][ 1200/ 1236] Overall Loss 0.318601 Objective Loss 0.318601 LR 0.001000 Time 0.020975 +2023-10-05 21:15:17,784 - Epoch: [70][ 1210/ 1236] Overall Loss 0.318624 Objective Loss 0.318624 LR 0.001000 Time 0.020969 +2023-10-05 21:15:17,987 - Epoch: [70][ 1220/ 1236] Overall Loss 0.318607 Objective Loss 0.318607 LR 0.001000 Time 0.020963 +2023-10-05 21:15:18,241 - Epoch: [70][ 1230/ 1236] Overall Loss 0.318964 Objective Loss 0.318964 LR 0.001000 Time 0.020999 +2023-10-05 21:15:18,359 - Epoch: [70][ 1236/ 1236] Overall Loss 0.318978 Objective Loss 0.318978 Top1 82.077393 Top5 97.352342 LR 0.001000 Time 0.020992 +2023-10-05 21:15:18,491 - --- validate (epoch=70)----------- +2023-10-05 21:15:18,491 - 29943 samples (256 per mini-batch) +2023-10-05 21:15:18,946 - Epoch: [70][ 10/ 117] Loss 0.371399 Top1 79.296875 Top5 97.187500 +2023-10-05 21:15:19,095 - Epoch: [70][ 20/ 117] Loss 0.374728 Top1 79.921875 Top5 97.089844 +2023-10-05 21:15:19,242 - Epoch: [70][ 30/ 117] Loss 0.367867 Top1 80.377604 Top5 97.135417 +2023-10-05 21:15:19,390 - Epoch: [70][ 40/ 117] Loss 0.365211 Top1 80.781250 Top5 97.148438 +2023-10-05 21:15:19,538 - Epoch: [70][ 50/ 117] Loss 0.364713 Top1 81.039062 Top5 97.234375 +2023-10-05 21:15:19,689 - Epoch: [70][ 60/ 117] Loss 0.363313 Top1 81.126302 Top5 97.207031 +2023-10-05 21:15:19,838 - Epoch: [70][ 70/ 117] Loss 0.365629 Top1 81.127232 Top5 97.220982 +2023-10-05 21:15:19,990 - Epoch: [70][ 80/ 117] Loss 0.368062 Top1 81.069336 Top5 97.255859 +2023-10-05 21:15:20,140 - Epoch: [70][ 90/ 117] Loss 0.364961 Top1 81.171875 Top5 97.317708 +2023-10-05 21:15:20,292 - Epoch: [70][ 100/ 117] Loss 0.366637 Top1 81.023438 Top5 97.292969 +2023-10-05 21:15:20,449 - Epoch: [70][ 110/ 117] Loss 0.363195 Top1 81.104403 Top5 97.343750 +2023-10-05 21:15:20,535 - Epoch: [70][ 117/ 117] Loss 0.364113 Top1 81.147514 Top5 97.341616 +2023-10-05 21:15:20,653 - ==> Top1: 81.148 Top5: 97.342 Loss: 0.364 + +2023-10-05 21:15:20,654 - ==> Confusion: +[[ 890 4 4 1 12 3 0 1 7 82 1 0 0 3 15 3 10 1 1 0 12] + [ 1 1046 3 0 6 17 1 24 1 1 3 0 0 0 1 4 2 0 16 2 3] + [ 3 2 955 9 4 0 25 9 0 0 2 2 6 2 0 1 3 1 15 6 11] + [ 2 1 25 945 1 1 2 0 4 3 6 1 10 5 26 3 2 6 34 1 11] + [ 19 7 1 0 963 6 0 1 2 8 0 0 0 2 13 5 18 1 0 2 2] + [ 6 51 1 1 5 930 1 49 3 4 1 9 0 17 6 4 4 1 6 9 8] + [ 0 8 38 0 0 0 1100 11 0 0 3 1 1 1 1 9 0 2 2 10 4] + [ 1 20 17 0 2 26 5 1053 0 1 4 4 1 1 2 5 0 4 57 7 8] + [ 13 7 1 0 0 2 2 0 955 42 13 0 1 10 28 3 1 0 10 1 0] + [ 76 2 2 1 7 7 1 0 31 925 1 2 0 26 21 4 0 0 2 4 7] + [ 3 5 16 4 2 0 5 4 15 2 951 2 0 14 9 1 1 0 6 4 9] + [ 1 1 0 0 0 11 0 2 0 2 1 934 32 5 0 4 0 14 0 23 5] + [ 1 2 5 3 0 2 3 3 0 0 1 32 956 4 2 10 5 14 6 6 13] + [ 2 0 3 0 3 5 0 0 7 19 7 4 2 1048 5 1 1 1 0 2 9] + [ 9 5 4 18 2 1 0 0 18 5 4 0 3 1 1001 0 3 2 18 0 7] + [ 0 4 3 2 4 0 3 0 0 0 0 5 9 0 0 1061 19 11 0 7 6] + [ 1 20 2 0 7 6 0 0 4 0 0 3 1 2 3 9 1092 0 1 3 7] + [ 0 0 2 3 0 0 2 0 1 0 1 6 19 3 2 13 1 976 2 0 7] + [ 1 6 7 13 0 1 0 25 1 0 3 0 0 0 13 0 0 0 986 1 11] + [ 0 3 4 0 2 4 6 21 0 0 0 14 3 4 1 6 12 0 1 1067 4] + [ 105 333 186 78 129 167 59 147 109 99 180 110 344 366 215 67 197 69 240 241 4464]] + +2023-10-05 21:15:20,655 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:15:20,655 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:15:20,661 - + +2023-10-05 21:15:20,661 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:15:21,670 - Epoch: [71][ 10/ 1236] Overall Loss 0.283323 Objective Loss 0.283323 LR 0.001000 Time 0.100882 +2023-10-05 21:15:21,875 - Epoch: [71][ 20/ 1236] Overall Loss 0.282649 Objective Loss 0.282649 LR 0.001000 Time 0.060645 +2023-10-05 21:15:22,080 - Epoch: [71][ 30/ 1236] Overall Loss 0.295697 Objective Loss 0.295697 LR 0.001000 Time 0.047263 +2023-10-05 21:15:22,285 - Epoch: [71][ 40/ 1236] Overall Loss 0.310978 Objective Loss 0.310978 LR 0.001000 Time 0.040544 +2023-10-05 21:15:22,490 - Epoch: [71][ 50/ 1236] Overall Loss 0.311634 Objective Loss 0.311634 LR 0.001000 Time 0.036543 +2023-10-05 21:15:22,694 - Epoch: [71][ 60/ 1236] Overall Loss 0.313822 Objective Loss 0.313822 LR 0.001000 Time 0.033848 +2023-10-05 21:15:22,900 - Epoch: [71][ 70/ 1236] Overall Loss 0.308483 Objective Loss 0.308483 LR 0.001000 Time 0.031938 +2023-10-05 21:15:23,104 - Epoch: [71][ 80/ 1236] Overall Loss 0.311158 Objective Loss 0.311158 LR 0.001000 Time 0.030500 +2023-10-05 21:15:23,310 - Epoch: [71][ 90/ 1236] Overall Loss 0.310175 Objective Loss 0.310175 LR 0.001000 Time 0.029388 +2023-10-05 21:15:23,515 - Epoch: [71][ 100/ 1236] Overall Loss 0.309310 Objective Loss 0.309310 LR 0.001000 Time 0.028494 +2023-10-05 21:15:23,724 - Epoch: [71][ 110/ 1236] Overall Loss 0.307838 Objective Loss 0.307838 LR 0.001000 Time 0.027800 +2023-10-05 21:15:23,932 - Epoch: [71][ 120/ 1236] Overall Loss 0.308389 Objective Loss 0.308389 LR 0.001000 Time 0.027216 +2023-10-05 21:15:24,141 - Epoch: [71][ 130/ 1236] Overall Loss 0.308469 Objective Loss 0.308469 LR 0.001000 Time 0.026726 +2023-10-05 21:15:24,349 - Epoch: [71][ 140/ 1236] Overall Loss 0.308314 Objective Loss 0.308314 LR 0.001000 Time 0.026299 +2023-10-05 21:15:24,557 - Epoch: [71][ 150/ 1236] Overall Loss 0.307724 Objective Loss 0.307724 LR 0.001000 Time 0.025926 +2023-10-05 21:15:24,761 - Epoch: [71][ 160/ 1236] Overall Loss 0.306813 Objective Loss 0.306813 LR 0.001000 Time 0.025580 +2023-10-05 21:15:24,965 - Epoch: [71][ 170/ 1236] Overall Loss 0.307572 Objective Loss 0.307572 LR 0.001000 Time 0.025272 +2023-10-05 21:15:25,167 - Epoch: [71][ 180/ 1236] Overall Loss 0.309656 Objective Loss 0.309656 LR 0.001000 Time 0.024991 +2023-10-05 21:15:25,371 - Epoch: [71][ 190/ 1236] Overall Loss 0.309500 Objective Loss 0.309500 LR 0.001000 Time 0.024746 +2023-10-05 21:15:25,573 - Epoch: [71][ 200/ 1236] Overall Loss 0.310615 Objective Loss 0.310615 LR 0.001000 Time 0.024519 +2023-10-05 21:15:25,777 - Epoch: [71][ 210/ 1236] Overall Loss 0.311326 Objective Loss 0.311326 LR 0.001000 Time 0.024321 +2023-10-05 21:15:25,980 - Epoch: [71][ 220/ 1236] Overall Loss 0.311126 Objective Loss 0.311126 LR 0.001000 Time 0.024134 +2023-10-05 21:15:26,182 - Epoch: [71][ 230/ 1236] Overall Loss 0.310046 Objective Loss 0.310046 LR 0.001000 Time 0.023964 +2023-10-05 21:15:26,385 - Epoch: [71][ 240/ 1236] Overall Loss 0.309868 Objective Loss 0.309868 LR 0.001000 Time 0.023808 +2023-10-05 21:15:26,588 - Epoch: [71][ 250/ 1236] Overall Loss 0.309016 Objective Loss 0.309016 LR 0.001000 Time 0.023669 +2023-10-05 21:15:26,791 - Epoch: [71][ 260/ 1236] Overall Loss 0.308600 Objective Loss 0.308600 LR 0.001000 Time 0.023537 +2023-10-05 21:15:26,995 - Epoch: [71][ 270/ 1236] Overall Loss 0.308270 Objective Loss 0.308270 LR 0.001000 Time 0.023419 +2023-10-05 21:15:27,196 - Epoch: [71][ 280/ 1236] Overall Loss 0.308741 Objective Loss 0.308741 LR 0.001000 Time 0.023300 +2023-10-05 21:15:27,400 - Epoch: [71][ 290/ 1236] Overall Loss 0.310178 Objective Loss 0.310178 LR 0.001000 Time 0.023198 +2023-10-05 21:15:27,603 - Epoch: [71][ 300/ 1236] Overall Loss 0.310818 Objective Loss 0.310818 LR 0.001000 Time 0.023100 +2023-10-05 21:15:27,807 - Epoch: [71][ 310/ 1236] Overall Loss 0.310652 Objective Loss 0.310652 LR 0.001000 Time 0.023012 +2023-10-05 21:15:28,009 - Epoch: [71][ 320/ 1236] Overall Loss 0.311100 Objective Loss 0.311100 LR 0.001000 Time 0.022925 +2023-10-05 21:15:28,213 - Epoch: [71][ 330/ 1236] Overall Loss 0.311908 Objective Loss 0.311908 LR 0.001000 Time 0.022848 +2023-10-05 21:15:28,416 - Epoch: [71][ 340/ 1236] Overall Loss 0.311912 Objective Loss 0.311912 LR 0.001000 Time 0.022771 +2023-10-05 21:15:28,621 - Epoch: [71][ 350/ 1236] Overall Loss 0.312439 Objective Loss 0.312439 LR 0.001000 Time 0.022704 +2023-10-05 21:15:28,824 - Epoch: [71][ 360/ 1236] Overall Loss 0.312317 Objective Loss 0.312317 LR 0.001000 Time 0.022637 +2023-10-05 21:15:29,029 - Epoch: [71][ 370/ 1236] Overall Loss 0.312636 Objective Loss 0.312636 LR 0.001000 Time 0.022579 +2023-10-05 21:15:29,234 - Epoch: [71][ 380/ 1236] Overall Loss 0.312547 Objective Loss 0.312547 LR 0.001000 Time 0.022523 +2023-10-05 21:15:29,439 - Epoch: [71][ 390/ 1236] Overall Loss 0.312023 Objective Loss 0.312023 LR 0.001000 Time 0.022470 +2023-10-05 21:15:29,644 - Epoch: [71][ 400/ 1236] Overall Loss 0.311745 Objective Loss 0.311745 LR 0.001000 Time 0.022420 +2023-10-05 21:15:29,849 - Epoch: [71][ 410/ 1236] Overall Loss 0.311897 Objective Loss 0.311897 LR 0.001000 Time 0.022372 +2023-10-05 21:15:30,054 - Epoch: [71][ 420/ 1236] Overall Loss 0.312367 Objective Loss 0.312367 LR 0.001000 Time 0.022326 +2023-10-05 21:15:30,260 - Epoch: [71][ 430/ 1236] Overall Loss 0.312311 Objective Loss 0.312311 LR 0.001000 Time 0.022284 +2023-10-05 21:15:30,464 - Epoch: [71][ 440/ 1236] Overall Loss 0.313067 Objective Loss 0.313067 LR 0.001000 Time 0.022242 +2023-10-05 21:15:30,670 - Epoch: [71][ 450/ 1236] Overall Loss 0.313143 Objective Loss 0.313143 LR 0.001000 Time 0.022203 +2023-10-05 21:15:30,874 - Epoch: [71][ 460/ 1236] Overall Loss 0.313039 Objective Loss 0.313039 LR 0.001000 Time 0.022165 +2023-10-05 21:15:31,080 - Epoch: [71][ 470/ 1236] Overall Loss 0.313554 Objective Loss 0.313554 LR 0.001000 Time 0.022129 +2023-10-05 21:15:31,285 - Epoch: [71][ 480/ 1236] Overall Loss 0.313835 Objective Loss 0.313835 LR 0.001000 Time 0.022094 +2023-10-05 21:15:31,490 - Epoch: [71][ 490/ 1236] Overall Loss 0.314276 Objective Loss 0.314276 LR 0.001000 Time 0.022061 +2023-10-05 21:15:31,695 - Epoch: [71][ 500/ 1236] Overall Loss 0.313633 Objective Loss 0.313633 LR 0.001000 Time 0.022029 +2023-10-05 21:15:31,900 - Epoch: [71][ 510/ 1236] Overall Loss 0.313684 Objective Loss 0.313684 LR 0.001000 Time 0.021998 +2023-10-05 21:15:32,105 - Epoch: [71][ 520/ 1236] Overall Loss 0.313641 Objective Loss 0.313641 LR 0.001000 Time 0.021968 +2023-10-05 21:15:32,310 - Epoch: [71][ 530/ 1236] Overall Loss 0.313777 Objective Loss 0.313777 LR 0.001000 Time 0.021940 +2023-10-05 21:15:32,515 - Epoch: [71][ 540/ 1236] Overall Loss 0.313813 Objective Loss 0.313813 LR 0.001000 Time 0.021913 +2023-10-05 21:15:32,720 - Epoch: [71][ 550/ 1236] Overall Loss 0.313971 Objective Loss 0.313971 LR 0.001000 Time 0.021886 +2023-10-05 21:15:32,925 - Epoch: [71][ 560/ 1236] Overall Loss 0.314089 Objective Loss 0.314089 LR 0.001000 Time 0.021861 +2023-10-05 21:15:33,130 - Epoch: [71][ 570/ 1236] Overall Loss 0.315031 Objective Loss 0.315031 LR 0.001000 Time 0.021836 +2023-10-05 21:15:33,334 - Epoch: [71][ 580/ 1236] Overall Loss 0.315482 Objective Loss 0.315482 LR 0.001000 Time 0.021812 +2023-10-05 21:15:33,537 - Epoch: [71][ 590/ 1236] Overall Loss 0.315990 Objective Loss 0.315990 LR 0.001000 Time 0.021786 +2023-10-05 21:15:33,740 - Epoch: [71][ 600/ 1236] Overall Loss 0.315904 Objective Loss 0.315904 LR 0.001000 Time 0.021759 +2023-10-05 21:15:33,940 - Epoch: [71][ 610/ 1236] Overall Loss 0.315949 Objective Loss 0.315949 LR 0.001000 Time 0.021731 +2023-10-05 21:15:34,143 - Epoch: [71][ 620/ 1236] Overall Loss 0.315603 Objective Loss 0.315603 LR 0.001000 Time 0.021707 +2023-10-05 21:15:34,344 - Epoch: [71][ 630/ 1236] Overall Loss 0.314963 Objective Loss 0.314963 LR 0.001000 Time 0.021681 +2023-10-05 21:15:34,547 - Epoch: [71][ 640/ 1236] Overall Loss 0.315258 Objective Loss 0.315258 LR 0.001000 Time 0.021657 +2023-10-05 21:15:34,747 - Epoch: [71][ 650/ 1236] Overall Loss 0.315393 Objective Loss 0.315393 LR 0.001000 Time 0.021633 +2023-10-05 21:15:34,950 - Epoch: [71][ 660/ 1236] Overall Loss 0.315604 Objective Loss 0.315604 LR 0.001000 Time 0.021612 +2023-10-05 21:15:35,151 - Epoch: [71][ 670/ 1236] Overall Loss 0.315606 Objective Loss 0.315606 LR 0.001000 Time 0.021588 +2023-10-05 21:15:35,354 - Epoch: [71][ 680/ 1236] Overall Loss 0.315392 Objective Loss 0.315392 LR 0.001000 Time 0.021568 +2023-10-05 21:15:35,555 - Epoch: [71][ 690/ 1236] Overall Loss 0.315597 Objective Loss 0.315597 LR 0.001000 Time 0.021546 +2023-10-05 21:15:35,757 - Epoch: [71][ 700/ 1236] Overall Loss 0.315483 Objective Loss 0.315483 LR 0.001000 Time 0.021527 +2023-10-05 21:15:35,958 - Epoch: [71][ 710/ 1236] Overall Loss 0.315386 Objective Loss 0.315386 LR 0.001000 Time 0.021507 +2023-10-05 21:15:36,161 - Epoch: [71][ 720/ 1236] Overall Loss 0.315388 Objective Loss 0.315388 LR 0.001000 Time 0.021488 +2023-10-05 21:15:36,362 - Epoch: [71][ 730/ 1236] Overall Loss 0.315326 Objective Loss 0.315326 LR 0.001000 Time 0.021469 +2023-10-05 21:15:36,565 - Epoch: [71][ 740/ 1236] Overall Loss 0.315145 Objective Loss 0.315145 LR 0.001000 Time 0.021453 +2023-10-05 21:15:36,765 - Epoch: [71][ 750/ 1236] Overall Loss 0.315192 Objective Loss 0.315192 LR 0.001000 Time 0.021434 +2023-10-05 21:15:36,968 - Epoch: [71][ 760/ 1236] Overall Loss 0.315312 Objective Loss 0.315312 LR 0.001000 Time 0.021418 +2023-10-05 21:15:37,169 - Epoch: [71][ 770/ 1236] Overall Loss 0.315279 Objective Loss 0.315279 LR 0.001000 Time 0.021400 +2023-10-05 21:15:37,372 - Epoch: [71][ 780/ 1236] Overall Loss 0.315032 Objective Loss 0.315032 LR 0.001000 Time 0.021385 +2023-10-05 21:15:37,572 - Epoch: [71][ 790/ 1236] Overall Loss 0.315159 Objective Loss 0.315159 LR 0.001000 Time 0.021368 +2023-10-05 21:15:37,775 - Epoch: [71][ 800/ 1236] Overall Loss 0.315725 Objective Loss 0.315725 LR 0.001000 Time 0.021354 +2023-10-05 21:15:37,976 - Epoch: [71][ 810/ 1236] Overall Loss 0.315599 Objective Loss 0.315599 LR 0.001000 Time 0.021338 +2023-10-05 21:15:38,179 - Epoch: [71][ 820/ 1236] Overall Loss 0.315874 Objective Loss 0.315874 LR 0.001000 Time 0.021324 +2023-10-05 21:15:38,380 - Epoch: [71][ 830/ 1236] Overall Loss 0.315673 Objective Loss 0.315673 LR 0.001000 Time 0.021309 +2023-10-05 21:15:38,582 - Epoch: [71][ 840/ 1236] Overall Loss 0.315850 Objective Loss 0.315850 LR 0.001000 Time 0.021296 +2023-10-05 21:15:38,783 - Epoch: [71][ 850/ 1236] Overall Loss 0.316146 Objective Loss 0.316146 LR 0.001000 Time 0.021281 +2023-10-05 21:15:38,985 - Epoch: [71][ 860/ 1236] Overall Loss 0.316062 Objective Loss 0.316062 LR 0.001000 Time 0.021269 +2023-10-05 21:15:39,186 - Epoch: [71][ 870/ 1236] Overall Loss 0.315836 Objective Loss 0.315836 LR 0.001000 Time 0.021255 +2023-10-05 21:15:39,389 - Epoch: [71][ 880/ 1236] Overall Loss 0.315973 Objective Loss 0.315973 LR 0.001000 Time 0.021243 +2023-10-05 21:15:39,590 - Epoch: [71][ 890/ 1236] Overall Loss 0.316266 Objective Loss 0.316266 LR 0.001000 Time 0.021230 +2023-10-05 21:15:39,793 - Epoch: [71][ 900/ 1236] Overall Loss 0.316199 Objective Loss 0.316199 LR 0.001000 Time 0.021219 +2023-10-05 21:15:39,994 - Epoch: [71][ 910/ 1236] Overall Loss 0.316058 Objective Loss 0.316058 LR 0.001000 Time 0.021206 +2023-10-05 21:15:40,196 - Epoch: [71][ 920/ 1236] Overall Loss 0.316092 Objective Loss 0.316092 LR 0.001000 Time 0.021195 +2023-10-05 21:15:40,397 - Epoch: [71][ 930/ 1236] Overall Loss 0.315857 Objective Loss 0.315857 LR 0.001000 Time 0.021183 +2023-10-05 21:15:40,601 - Epoch: [71][ 940/ 1236] Overall Loss 0.315947 Objective Loss 0.315947 LR 0.001000 Time 0.021174 +2023-10-05 21:15:40,802 - Epoch: [71][ 950/ 1236] Overall Loss 0.315800 Objective Loss 0.315800 LR 0.001000 Time 0.021163 +2023-10-05 21:15:41,005 - Epoch: [71][ 960/ 1236] Overall Loss 0.316131 Objective Loss 0.316131 LR 0.001000 Time 0.021153 +2023-10-05 21:15:41,206 - Epoch: [71][ 970/ 1236] Overall Loss 0.315960 Objective Loss 0.315960 LR 0.001000 Time 0.021142 +2023-10-05 21:15:41,408 - Epoch: [71][ 980/ 1236] Overall Loss 0.316097 Objective Loss 0.316097 LR 0.001000 Time 0.021132 +2023-10-05 21:15:41,609 - Epoch: [71][ 990/ 1236] Overall Loss 0.315686 Objective Loss 0.315686 LR 0.001000 Time 0.021121 +2023-10-05 21:15:41,812 - Epoch: [71][ 1000/ 1236] Overall Loss 0.315515 Objective Loss 0.315515 LR 0.001000 Time 0.021112 +2023-10-05 21:15:42,013 - Epoch: [71][ 1010/ 1236] Overall Loss 0.315442 Objective Loss 0.315442 LR 0.001000 Time 0.021101 +2023-10-05 21:15:42,215 - Epoch: [71][ 1020/ 1236] Overall Loss 0.315578 Objective Loss 0.315578 LR 0.001000 Time 0.021093 +2023-10-05 21:15:42,416 - Epoch: [71][ 1030/ 1236] Overall Loss 0.315654 Objective Loss 0.315654 LR 0.001000 Time 0.021083 +2023-10-05 21:15:42,619 - Epoch: [71][ 1040/ 1236] Overall Loss 0.315770 Objective Loss 0.315770 LR 0.001000 Time 0.021074 +2023-10-05 21:15:42,820 - Epoch: [71][ 1050/ 1236] Overall Loss 0.315833 Objective Loss 0.315833 LR 0.001000 Time 0.021065 +2023-10-05 21:15:43,022 - Epoch: [71][ 1060/ 1236] Overall Loss 0.315497 Objective Loss 0.315497 LR 0.001000 Time 0.021057 +2023-10-05 21:15:43,223 - Epoch: [71][ 1070/ 1236] Overall Loss 0.315858 Objective Loss 0.315858 LR 0.001000 Time 0.021047 +2023-10-05 21:15:43,426 - Epoch: [71][ 1080/ 1236] Overall Loss 0.316200 Objective Loss 0.316200 LR 0.001000 Time 0.021040 +2023-10-05 21:15:43,634 - Epoch: [71][ 1090/ 1236] Overall Loss 0.316042 Objective Loss 0.316042 LR 0.001000 Time 0.021037 +2023-10-05 21:15:43,846 - Epoch: [71][ 1100/ 1236] Overall Loss 0.315908 Objective Loss 0.315908 LR 0.001000 Time 0.021039 +2023-10-05 21:15:44,054 - Epoch: [71][ 1110/ 1236] Overall Loss 0.315821 Objective Loss 0.315821 LR 0.001000 Time 0.021036 +2023-10-05 21:15:44,260 - Epoch: [71][ 1120/ 1236] Overall Loss 0.315998 Objective Loss 0.315998 LR 0.001000 Time 0.021032 +2023-10-05 21:15:44,461 - Epoch: [71][ 1130/ 1236] Overall Loss 0.316274 Objective Loss 0.316274 LR 0.001000 Time 0.021023 +2023-10-05 21:15:44,663 - Epoch: [71][ 1140/ 1236] Overall Loss 0.316425 Objective Loss 0.316425 LR 0.001000 Time 0.021016 +2023-10-05 21:15:44,864 - Epoch: [71][ 1150/ 1236] Overall Loss 0.316445 Objective Loss 0.316445 LR 0.001000 Time 0.021007 +2023-10-05 21:15:45,066 - Epoch: [71][ 1160/ 1236] Overall Loss 0.316224 Objective Loss 0.316224 LR 0.001000 Time 0.021000 +2023-10-05 21:15:45,267 - Epoch: [71][ 1170/ 1236] Overall Loss 0.316333 Objective Loss 0.316333 LR 0.001000 Time 0.020992 +2023-10-05 21:15:45,470 - Epoch: [71][ 1180/ 1236] Overall Loss 0.316121 Objective Loss 0.316121 LR 0.001000 Time 0.020985 +2023-10-05 21:15:45,670 - Epoch: [71][ 1190/ 1236] Overall Loss 0.316095 Objective Loss 0.316095 LR 0.001000 Time 0.020977 +2023-10-05 21:15:45,872 - Epoch: [71][ 1200/ 1236] Overall Loss 0.315988 Objective Loss 0.315988 LR 0.001000 Time 0.020970 +2023-10-05 21:15:46,072 - Epoch: [71][ 1210/ 1236] Overall Loss 0.315892 Objective Loss 0.315892 LR 0.001000 Time 0.020962 +2023-10-05 21:15:46,275 - Epoch: [71][ 1220/ 1236] Overall Loss 0.316386 Objective Loss 0.316386 LR 0.001000 Time 0.020956 +2023-10-05 21:15:46,528 - Epoch: [71][ 1230/ 1236] Overall Loss 0.316457 Objective Loss 0.316457 LR 0.001000 Time 0.020992 +2023-10-05 21:15:46,646 - Epoch: [71][ 1236/ 1236] Overall Loss 0.316456 Objective Loss 0.316456 Top1 85.539715 Top5 98.981670 LR 0.001000 Time 0.020985 +2023-10-05 21:15:46,774 - --- validate (epoch=71)----------- +2023-10-05 21:15:46,774 - 29943 samples (256 per mini-batch) +2023-10-05 21:15:47,241 - Epoch: [71][ 10/ 117] Loss 0.359453 Top1 80.937500 Top5 97.343750 +2023-10-05 21:15:47,402 - Epoch: [71][ 20/ 117] Loss 0.348068 Top1 80.703125 Top5 97.265625 +2023-10-05 21:15:47,556 - Epoch: [71][ 30/ 117] Loss 0.358068 Top1 80.859375 Top5 97.187500 +2023-10-05 21:15:47,705 - Epoch: [71][ 40/ 117] Loss 0.364966 Top1 80.644531 Top5 97.197266 +2023-10-05 21:15:47,852 - Epoch: [71][ 50/ 117] Loss 0.358499 Top1 81.085938 Top5 97.281250 +2023-10-05 21:15:48,001 - Epoch: [71][ 60/ 117] Loss 0.355860 Top1 81.243490 Top5 97.330729 +2023-10-05 21:15:48,147 - Epoch: [71][ 70/ 117] Loss 0.355781 Top1 81.400670 Top5 97.338170 +2023-10-05 21:15:48,296 - Epoch: [71][ 80/ 117] Loss 0.360757 Top1 81.367188 Top5 97.392578 +2023-10-05 21:15:48,442 - Epoch: [71][ 90/ 117] Loss 0.358881 Top1 81.414931 Top5 97.426215 +2023-10-05 21:15:48,591 - Epoch: [71][ 100/ 117] Loss 0.356891 Top1 81.503906 Top5 97.414062 +2023-10-05 21:15:48,745 - Epoch: [71][ 110/ 117] Loss 0.355285 Top1 81.537642 Top5 97.453835 +2023-10-05 21:15:48,831 - Epoch: [71][ 117/ 117] Loss 0.354929 Top1 81.554954 Top5 97.445146 +2023-10-05 21:15:48,973 - ==> Top1: 81.555 Top5: 97.445 Loss: 0.355 + +2023-10-05 21:15:48,974 - ==> Confusion: +[[ 916 2 1 1 17 2 0 0 6 75 1 0 1 2 5 5 6 1 0 1 8] + [ 1 1053 2 0 7 19 2 12 2 0 4 3 0 0 2 3 8 1 5 2 5] + [ 4 2 915 16 4 1 44 13 0 2 10 4 9 3 1 4 2 1 3 2 16] + [ 2 4 10 965 2 3 0 2 3 0 11 0 4 6 37 4 0 10 12 3 11] + [ 15 7 0 0 971 4 0 0 1 11 1 2 1 2 8 2 21 1 0 1 2] + [ 4 38 0 2 1 971 2 16 1 3 3 16 3 19 9 1 4 1 2 6 14] + [ 0 6 30 0 0 0 1118 5 0 0 8 3 1 0 1 4 1 2 2 5 5] + [ 5 39 19 1 2 56 10 1009 4 4 5 5 0 2 0 0 1 0 30 18 8] + [ 13 5 0 1 1 2 0 1 958 46 7 0 1 20 23 2 1 1 4 0 3] + [ 90 1 2 0 10 8 1 0 25 939 0 0 2 19 8 1 1 0 0 2 10] + [ 1 10 11 7 1 0 6 2 14 1 957 2 0 14 7 1 0 2 7 2 8] + [ 0 0 0 0 2 20 1 2 1 0 0 919 44 5 0 4 3 19 0 11 4] + [ 0 2 3 5 1 3 1 0 2 1 1 31 970 3 0 4 4 21 1 3 12] + [ 2 0 1 0 4 10 0 0 16 20 5 6 4 1026 5 4 2 1 0 3 10] + [ 10 4 2 8 6 1 0 0 20 6 3 1 2 2 1012 0 1 2 9 0 12] + [ 0 2 1 0 3 0 2 0 0 0 0 9 10 2 0 1060 14 18 0 5 8] + [ 1 17 1 2 6 5 1 0 1 0 0 1 3 1 5 9 1092 0 0 7 9] + [ 0 1 0 1 0 0 1 0 0 0 0 2 18 1 2 5 0 997 1 1 8] + [ 1 14 4 14 3 0 1 30 6 2 7 1 1 0 14 0 1 0 956 2 11] + [ 0 1 2 3 2 3 9 10 1 1 1 8 6 6 0 8 11 0 0 1076 4] + [ 127 289 116 93 126 181 73 104 108 103 207 110 419 319 195 61 200 88 153 293 4540]] + +2023-10-05 21:15:48,975 - ==> Best [Top1: 82.189 Top5: 97.729 Sparsity:0.00 Params: 148928 on epoch: 46] +2023-10-05 21:15:48,975 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:15:48,981 - + +2023-10-05 21:15:48,981 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:15:49,980 - Epoch: [72][ 10/ 1236] Overall Loss 0.321633 Objective Loss 0.321633 LR 0.001000 Time 0.099847 +2023-10-05 21:15:50,181 - Epoch: [72][ 20/ 1236] Overall Loss 0.298241 Objective Loss 0.298241 LR 0.001000 Time 0.059950 +2023-10-05 21:15:50,380 - Epoch: [72][ 30/ 1236] Overall Loss 0.298796 Objective Loss 0.298796 LR 0.001000 Time 0.046589 +2023-10-05 21:15:50,580 - Epoch: [72][ 40/ 1236] Overall Loss 0.302286 Objective Loss 0.302286 LR 0.001000 Time 0.039930 +2023-10-05 21:15:50,778 - Epoch: [72][ 50/ 1236] Overall Loss 0.301326 Objective Loss 0.301326 LR 0.001000 Time 0.035914 +2023-10-05 21:15:50,979 - Epoch: [72][ 60/ 1236] Overall Loss 0.303878 Objective Loss 0.303878 LR 0.001000 Time 0.033272 +2023-10-05 21:15:51,178 - Epoch: [72][ 70/ 1236] Overall Loss 0.304110 Objective Loss 0.304110 LR 0.001000 Time 0.031353 +2023-10-05 21:15:51,379 - Epoch: [72][ 80/ 1236] Overall Loss 0.301133 Objective Loss 0.301133 LR 0.001000 Time 0.029940 +2023-10-05 21:15:51,578 - Epoch: [72][ 90/ 1236] Overall Loss 0.304166 Objective Loss 0.304166 LR 0.001000 Time 0.028820 +2023-10-05 21:15:51,779 - Epoch: [72][ 100/ 1236] Overall Loss 0.306842 Objective Loss 0.306842 LR 0.001000 Time 0.027945 +2023-10-05 21:15:51,976 - Epoch: [72][ 110/ 1236] Overall Loss 0.305901 Objective Loss 0.305901 LR 0.001000 Time 0.027197 +2023-10-05 21:15:52,177 - Epoch: [72][ 120/ 1236] Overall Loss 0.306686 Objective Loss 0.306686 LR 0.001000 Time 0.026597 +2023-10-05 21:15:52,376 - Epoch: [72][ 130/ 1236] Overall Loss 0.307295 Objective Loss 0.307295 LR 0.001000 Time 0.026078 +2023-10-05 21:15:52,576 - Epoch: [72][ 140/ 1236] Overall Loss 0.307614 Objective Loss 0.307614 LR 0.001000 Time 0.025647 +2023-10-05 21:15:52,775 - Epoch: [72][ 150/ 1236] Overall Loss 0.311363 Objective Loss 0.311363 LR 0.001000 Time 0.025261 +2023-10-05 21:15:52,976 - Epoch: [72][ 160/ 1236] Overall Loss 0.310474 Objective Loss 0.310474 LR 0.001000 Time 0.024935 +2023-10-05 21:15:53,175 - Epoch: [72][ 170/ 1236] Overall Loss 0.310778 Objective Loss 0.310778 LR 0.001000 Time 0.024636 +2023-10-05 21:15:53,376 - Epoch: [72][ 180/ 1236] Overall Loss 0.310951 Objective Loss 0.310951 LR 0.001000 Time 0.024382 +2023-10-05 21:15:53,575 - Epoch: [72][ 190/ 1236] Overall Loss 0.309499 Objective Loss 0.309499 LR 0.001000 Time 0.024144 +2023-10-05 21:15:53,776 - Epoch: [72][ 200/ 1236] Overall Loss 0.308938 Objective Loss 0.308938 LR 0.001000 Time 0.023940 +2023-10-05 21:15:53,975 - Epoch: [72][ 210/ 1236] Overall Loss 0.309033 Objective Loss 0.309033 LR 0.001000 Time 0.023746 +2023-10-05 21:15:54,176 - Epoch: [72][ 220/ 1236] Overall Loss 0.307916 Objective Loss 0.307916 LR 0.001000 Time 0.023578 +2023-10-05 21:15:54,375 - Epoch: [72][ 230/ 1236] Overall Loss 0.307361 Objective Loss 0.307361 LR 0.001000 Time 0.023416 +2023-10-05 21:15:54,576 - Epoch: [72][ 240/ 1236] Overall Loss 0.307235 Objective Loss 0.307235 LR 0.001000 Time 0.023277 +2023-10-05 21:15:54,775 - Epoch: [72][ 250/ 1236] Overall Loss 0.307453 Objective Loss 0.307453 LR 0.001000 Time 0.023140 +2023-10-05 21:15:54,976 - Epoch: [72][ 260/ 1236] Overall Loss 0.307424 Objective Loss 0.307424 LR 0.001000 Time 0.023022 +2023-10-05 21:15:55,175 - Epoch: [72][ 270/ 1236] Overall Loss 0.307800 Objective Loss 0.307800 LR 0.001000 Time 0.022907 +2023-10-05 21:15:55,376 - Epoch: [72][ 280/ 1236] Overall Loss 0.306590 Objective Loss 0.306590 LR 0.001000 Time 0.022805 +2023-10-05 21:15:55,576 - Epoch: [72][ 290/ 1236] Overall Loss 0.306048 Objective Loss 0.306048 LR 0.001000 Time 0.022707 +2023-10-05 21:15:55,777 - Epoch: [72][ 300/ 1236] Overall Loss 0.306342 Objective Loss 0.306342 LR 0.001000 Time 0.022619 +2023-10-05 21:15:55,977 - Epoch: [72][ 310/ 1236] Overall Loss 0.307107 Objective Loss 0.307107 LR 0.001000 Time 0.022534 +2023-10-05 21:15:56,178 - Epoch: [72][ 320/ 1236] Overall Loss 0.306597 Objective Loss 0.306597 LR 0.001000 Time 0.022456 +2023-10-05 21:15:56,378 - Epoch: [72][ 330/ 1236] Overall Loss 0.306921 Objective Loss 0.306921 LR 0.001000 Time 0.022381 +2023-10-05 21:15:56,579 - Epoch: [72][ 340/ 1236] Overall Loss 0.307332 Objective Loss 0.307332 LR 0.001000 Time 0.022313 +2023-10-05 21:15:56,779 - Epoch: [72][ 350/ 1236] Overall Loss 0.306969 Objective Loss 0.306969 LR 0.001000 Time 0.022245 +2023-10-05 21:15:56,980 - Epoch: [72][ 360/ 1236] Overall Loss 0.307781 Objective Loss 0.307781 LR 0.001000 Time 0.022185 +2023-10-05 21:15:57,180 - Epoch: [72][ 370/ 1236] Overall Loss 0.307860 Objective Loss 0.307860 LR 0.001000 Time 0.022124 +2023-10-05 21:15:57,381 - Epoch: [72][ 380/ 1236] Overall Loss 0.308961 Objective Loss 0.308961 LR 0.001000 Time 0.022071 +2023-10-05 21:15:57,581 - Epoch: [72][ 390/ 1236] Overall Loss 0.309119 Objective Loss 0.309119 LR 0.001000 Time 0.022015 +2023-10-05 21:15:57,781 - Epoch: [72][ 400/ 1236] Overall Loss 0.309592 Objective Loss 0.309592 LR 0.001000 Time 0.021964 +2023-10-05 21:15:57,980 - Epoch: [72][ 410/ 1236] Overall Loss 0.308871 Objective Loss 0.308871 LR 0.001000 Time 0.021914 +2023-10-05 21:15:58,182 - Epoch: [72][ 420/ 1236] Overall Loss 0.309432 Objective Loss 0.309432 LR 0.001000 Time 0.021871 +2023-10-05 21:15:58,381 - Epoch: [72][ 430/ 1236] Overall Loss 0.309423 Objective Loss 0.309423 LR 0.001000 Time 0.021826 +2023-10-05 21:15:58,581 - Epoch: [72][ 440/ 1236] Overall Loss 0.309245 Objective Loss 0.309245 LR 0.001000 Time 0.021783 +2023-10-05 21:15:58,779 - Epoch: [72][ 450/ 1236] Overall Loss 0.309512 Objective Loss 0.309512 LR 0.001000 Time 0.021738 +2023-10-05 21:15:58,981 - Epoch: [72][ 460/ 1236] Overall Loss 0.309606 Objective Loss 0.309606 LR 0.001000 Time 0.021703 +2023-10-05 21:15:59,181 - Epoch: [72][ 470/ 1236] Overall Loss 0.310188 Objective Loss 0.310188 LR 0.001000 Time 0.021665 +2023-10-05 21:15:59,382 - Epoch: [72][ 480/ 1236] Overall Loss 0.310703 Objective Loss 0.310703 LR 0.001000 Time 0.021633 +2023-10-05 21:15:59,580 - Epoch: [72][ 490/ 1236] Overall Loss 0.311104 Objective Loss 0.311104 LR 0.001000 Time 0.021596 +2023-10-05 21:15:59,782 - Epoch: [72][ 500/ 1236] Overall Loss 0.312175 Objective Loss 0.312175 LR 0.001000 Time 0.021565 +2023-10-05 21:15:59,980 - Epoch: [72][ 510/ 1236] Overall Loss 0.312870 Objective Loss 0.312870 LR 0.001000 Time 0.021530 +2023-10-05 21:16:00,181 - Epoch: [72][ 520/ 1236] Overall Loss 0.313018 Objective Loss 0.313018 LR 0.001000 Time 0.021503 +2023-10-05 21:16:00,380 - Epoch: [72][ 530/ 1236] Overall Loss 0.312958 Objective Loss 0.312958 LR 0.001000 Time 0.021472 +2023-10-05 21:16:00,581 - Epoch: [72][ 540/ 1236] Overall Loss 0.313046 Objective Loss 0.313046 LR 0.001000 Time 0.021446 +2023-10-05 21:16:00,780 - Epoch: [72][ 550/ 1236] Overall Loss 0.313602 Objective Loss 0.313602 LR 0.001000 Time 0.021416 +2023-10-05 21:16:00,980 - Epoch: [72][ 560/ 1236] Overall Loss 0.313777 Objective Loss 0.313777 LR 0.001000 Time 0.021391 +2023-10-05 21:16:01,179 - Epoch: [72][ 570/ 1236] Overall Loss 0.313495 Objective Loss 0.313495 LR 0.001000 Time 0.021364 +2023-10-05 21:16:01,380 - Epoch: [72][ 580/ 1236] Overall Loss 0.313470 Objective Loss 0.313470 LR 0.001000 Time 0.021341 +2023-10-05 21:16:01,578 - Epoch: [72][ 590/ 1236] Overall Loss 0.313511 Objective Loss 0.313511 LR 0.001000 Time 0.021316 +2023-10-05 21:16:01,779 - Epoch: [72][ 600/ 1236] Overall Loss 0.313661 Objective Loss 0.313661 LR 0.001000 Time 0.021294 +2023-10-05 21:16:01,976 - Epoch: [72][ 610/ 1236] Overall Loss 0.313808 Objective Loss 0.313808 LR 0.001000 Time 0.021267 +2023-10-05 21:16:02,176 - Epoch: [72][ 620/ 1236] Overall Loss 0.313917 Objective Loss 0.313917 LR 0.001000 Time 0.021247 +2023-10-05 21:16:02,374 - Epoch: [72][ 630/ 1236] Overall Loss 0.313959 Objective Loss 0.313959 LR 0.001000 Time 0.021223 +2023-10-05 21:16:02,574 - Epoch: [72][ 640/ 1236] Overall Loss 0.314112 Objective Loss 0.314112 LR 0.001000 Time 0.021204 +2023-10-05 21:16:02,773 - Epoch: [72][ 650/ 1236] Overall Loss 0.313677 Objective Loss 0.313677 LR 0.001000 Time 0.021182 +2023-10-05 21:16:02,973 - Epoch: [72][ 660/ 1236] Overall Loss 0.313419 Objective Loss 0.313419 LR 0.001000 Time 0.021165 +2023-10-05 21:16:03,170 - Epoch: [72][ 670/ 1236] Overall Loss 0.313451 Objective Loss 0.313451 LR 0.001000 Time 0.021142 +2023-10-05 21:16:03,371 - Epoch: [72][ 680/ 1236] Overall Loss 0.314165 Objective Loss 0.314165 LR 0.001000 Time 0.021126 +2023-10-05 21:16:03,569 - Epoch: [72][ 690/ 1236] Overall Loss 0.314196 Objective Loss 0.314196 LR 0.001000 Time 0.021107 +2023-10-05 21:16:03,769 - Epoch: [72][ 700/ 1236] Overall Loss 0.314160 Objective Loss 0.314160 LR 0.001000 Time 0.021090 +2023-10-05 21:16:03,967 - Epoch: [72][ 710/ 1236] Overall Loss 0.313985 Objective Loss 0.313985 LR 0.001000 Time 0.021072 +2023-10-05 21:16:04,168 - Epoch: [72][ 720/ 1236] Overall Loss 0.314046 Objective Loss 0.314046 LR 0.001000 Time 0.021057 +2023-10-05 21:16:04,366 - Epoch: [72][ 730/ 1236] Overall Loss 0.314272 Objective Loss 0.314272 LR 0.001000 Time 0.021040 +2023-10-05 21:16:04,567 - Epoch: [72][ 740/ 1236] Overall Loss 0.314367 Objective Loss 0.314367 LR 0.001000 Time 0.021026 +2023-10-05 21:16:04,765 - Epoch: [72][ 750/ 1236] Overall Loss 0.314198 Objective Loss 0.314198 LR 0.001000 Time 0.021009 +2023-10-05 21:16:04,965 - Epoch: [72][ 760/ 1236] Overall Loss 0.314291 Objective Loss 0.314291 LR 0.001000 Time 0.020997 +2023-10-05 21:16:05,164 - Epoch: [72][ 770/ 1236] Overall Loss 0.314644 Objective Loss 0.314644 LR 0.001000 Time 0.020981 +2023-10-05 21:16:05,364 - Epoch: [72][ 780/ 1236] Overall Loss 0.314897 Objective Loss 0.314897 LR 0.001000 Time 0.020969 +2023-10-05 21:16:05,563 - Epoch: [72][ 790/ 1236] Overall Loss 0.315128 Objective Loss 0.315128 LR 0.001000 Time 0.020954 +2023-10-05 21:16:05,763 - Epoch: [72][ 800/ 1236] Overall Loss 0.315024 Objective Loss 0.315024 LR 0.001000 Time 0.020942 +2023-10-05 21:16:05,960 - Epoch: [72][ 810/ 1236] Overall Loss 0.315370 Objective Loss 0.315370 LR 0.001000 Time 0.020926 +2023-10-05 21:16:06,159 - Epoch: [72][ 820/ 1236] Overall Loss 0.315123 Objective Loss 0.315123 LR 0.001000 Time 0.020913 +2023-10-05 21:16:06,357 - Epoch: [72][ 830/ 1236] Overall Loss 0.315309 Objective Loss 0.315309 LR 0.001000 Time 0.020900 +2023-10-05 21:16:06,556 - Epoch: [72][ 840/ 1236] Overall Loss 0.315220 Objective Loss 0.315220 LR 0.001000 Time 0.020887 +2023-10-05 21:16:06,755 - Epoch: [72][ 850/ 1236] Overall Loss 0.315436 Objective Loss 0.315436 LR 0.001000 Time 0.020875 +2023-10-05 21:16:06,955 - Epoch: [72][ 860/ 1236] Overall Loss 0.315506 Objective Loss 0.315506 LR 0.001000 Time 0.020865 +2023-10-05 21:16:07,154 - Epoch: [72][ 870/ 1236] Overall Loss 0.315329 Objective Loss 0.315329 LR 0.001000 Time 0.020853 +2023-10-05 21:16:07,352 - Epoch: [72][ 880/ 1236] Overall Loss 0.315287 Objective Loss 0.315287 LR 0.001000 Time 0.020841 +2023-10-05 21:16:07,550 - Epoch: [72][ 890/ 1236] Overall Loss 0.315051 Objective Loss 0.315051 LR 0.001000 Time 0.020829 +2023-10-05 21:16:07,751 - Epoch: [72][ 900/ 1236] Overall Loss 0.314701 Objective Loss 0.314701 LR 0.001000 Time 0.020820 +2023-10-05 21:16:07,947 - Epoch: [72][ 910/ 1236] Overall Loss 0.314978 Objective Loss 0.314978 LR 0.001000 Time 0.020807 +2023-10-05 21:16:08,148 - Epoch: [72][ 920/ 1236] Overall Loss 0.314555 Objective Loss 0.314555 LR 0.001000 Time 0.020798 +2023-10-05 21:16:08,346 - Epoch: [72][ 930/ 1236] Overall Loss 0.314653 Objective Loss 0.314653 LR 0.001000 Time 0.020788 +2023-10-05 21:16:08,547 - Epoch: [72][ 940/ 1236] Overall Loss 0.314712 Objective Loss 0.314712 LR 0.001000 Time 0.020779 +2023-10-05 21:16:08,743 - Epoch: [72][ 950/ 1236] Overall Loss 0.314766 Objective Loss 0.314766 LR 0.001000 Time 0.020767 +2023-10-05 21:16:08,944 - Epoch: [72][ 960/ 1236] Overall Loss 0.314932 Objective Loss 0.314932 LR 0.001000 Time 0.020759 +2023-10-05 21:16:09,142 - Epoch: [72][ 970/ 1236] Overall Loss 0.315226 Objective Loss 0.315226 LR 0.001000 Time 0.020750 +2023-10-05 21:16:09,342 - Epoch: [72][ 980/ 1236] Overall Loss 0.314973 Objective Loss 0.314973 LR 0.001000 Time 0.020742 +2023-10-05 21:16:09,539 - Epoch: [72][ 990/ 1236] Overall Loss 0.314811 Objective Loss 0.314811 LR 0.001000 Time 0.020731 +2023-10-05 21:16:09,740 - Epoch: [72][ 1000/ 1236] Overall Loss 0.314768 Objective Loss 0.314768 LR 0.001000 Time 0.020724 +2023-10-05 21:16:09,938 - Epoch: [72][ 1010/ 1236] Overall Loss 0.314735 Objective Loss 0.314735 LR 0.001000 Time 0.020714 +2023-10-05 21:16:10,137 - Epoch: [72][ 1020/ 1236] Overall Loss 0.314994 Objective Loss 0.314994 LR 0.001000 Time 0.020706 +2023-10-05 21:16:10,335 - Epoch: [72][ 1030/ 1236] Overall Loss 0.315017 Objective Loss 0.315017 LR 0.001000 Time 0.020697 +2023-10-05 21:16:10,536 - Epoch: [72][ 1040/ 1236] Overall Loss 0.315088 Objective Loss 0.315088 LR 0.001000 Time 0.020691 +2023-10-05 21:16:10,734 - Epoch: [72][ 1050/ 1236] Overall Loss 0.315021 Objective Loss 0.315021 LR 0.001000 Time 0.020682 +2023-10-05 21:16:10,933 - Epoch: [72][ 1060/ 1236] Overall Loss 0.314837 Objective Loss 0.314837 LR 0.001000 Time 0.020675 +2023-10-05 21:16:11,131 - Epoch: [72][ 1070/ 1236] Overall Loss 0.314375 Objective Loss 0.314375 LR 0.001000 Time 0.020667 +2023-10-05 21:16:11,332 - Epoch: [72][ 1080/ 1236] Overall Loss 0.314415 Objective Loss 0.314415 LR 0.001000 Time 0.020661 +2023-10-05 21:16:11,530 - Epoch: [72][ 1090/ 1236] Overall Loss 0.314553 Objective Loss 0.314553 LR 0.001000 Time 0.020653 +2023-10-05 21:16:11,729 - Epoch: [72][ 1100/ 1236] Overall Loss 0.314732 Objective Loss 0.314732 LR 0.001000 Time 0.020646 +2023-10-05 21:16:11,926 - Epoch: [72][ 1110/ 1236] Overall Loss 0.315166 Objective Loss 0.315166 LR 0.001000 Time 0.020637 +2023-10-05 21:16:12,125 - Epoch: [72][ 1120/ 1236] Overall Loss 0.315526 Objective Loss 0.315526 LR 0.001000 Time 0.020630 +2023-10-05 21:16:12,324 - Epoch: [72][ 1130/ 1236] Overall Loss 0.315583 Objective Loss 0.315583 LR 0.001000 Time 0.020623 +2023-10-05 21:16:12,524 - Epoch: [72][ 1140/ 1236] Overall Loss 0.315513 Objective Loss 0.315513 LR 0.001000 Time 0.020617 +2023-10-05 21:16:12,723 - Epoch: [72][ 1150/ 1236] Overall Loss 0.315322 Objective Loss 0.315322 LR 0.001000 Time 0.020610 +2023-10-05 21:16:12,923 - Epoch: [72][ 1160/ 1236] Overall Loss 0.315506 Objective Loss 0.315506 LR 0.001000 Time 0.020605 +2023-10-05 21:16:13,122 - Epoch: [72][ 1170/ 1236] Overall Loss 0.315731 Objective Loss 0.315731 LR 0.001000 Time 0.020598 +2023-10-05 21:16:13,322 - Epoch: [72][ 1180/ 1236] Overall Loss 0.316135 Objective Loss 0.316135 LR 0.001000 Time 0.020593 +2023-10-05 21:16:13,521 - Epoch: [72][ 1190/ 1236] Overall Loss 0.316211 Objective Loss 0.316211 LR 0.001000 Time 0.020587 +2023-10-05 21:16:13,720 - Epoch: [72][ 1200/ 1236] Overall Loss 0.316451 Objective Loss 0.316451 LR 0.001000 Time 0.020581 +2023-10-05 21:16:13,918 - Epoch: [72][ 1210/ 1236] Overall Loss 0.316673 Objective Loss 0.316673 LR 0.001000 Time 0.020575 +2023-10-05 21:16:14,117 - Epoch: [72][ 1220/ 1236] Overall Loss 0.316490 Objective Loss 0.316490 LR 0.001000 Time 0.020569 +2023-10-05 21:16:14,370 - Epoch: [72][ 1230/ 1236] Overall Loss 0.316575 Objective Loss 0.316575 LR 0.001000 Time 0.020607 +2023-10-05 21:16:14,488 - Epoch: [72][ 1236/ 1236] Overall Loss 0.316753 Objective Loss 0.316753 Top1 85.947047 Top5 98.370672 LR 0.001000 Time 0.020602 +2023-10-05 21:16:14,631 - --- validate (epoch=72)----------- +2023-10-05 21:16:14,631 - 29943 samples (256 per mini-batch) +2023-10-05 21:16:15,081 - Epoch: [72][ 10/ 117] Loss 0.354125 Top1 82.226562 Top5 97.812500 +2023-10-05 21:16:15,229 - Epoch: [72][ 20/ 117] Loss 0.374975 Top1 82.597656 Top5 97.675781 +2023-10-05 21:16:15,376 - Epoch: [72][ 30/ 117] Loss 0.377699 Top1 82.890625 Top5 97.669271 +2023-10-05 21:16:15,522 - Epoch: [72][ 40/ 117] Loss 0.376936 Top1 83.066406 Top5 97.675781 +2023-10-05 21:16:15,670 - Epoch: [72][ 50/ 117] Loss 0.377220 Top1 83.093750 Top5 97.703125 +2023-10-05 21:16:15,818 - Epoch: [72][ 60/ 117] Loss 0.374840 Top1 83.105469 Top5 97.701823 +2023-10-05 21:16:15,966 - Epoch: [72][ 70/ 117] Loss 0.370927 Top1 83.091518 Top5 97.700893 +2023-10-05 21:16:16,114 - Epoch: [72][ 80/ 117] Loss 0.373519 Top1 83.071289 Top5 97.680664 +2023-10-05 21:16:16,263 - Epoch: [72][ 90/ 117] Loss 0.375537 Top1 82.925347 Top5 97.725694 +2023-10-05 21:16:16,411 - Epoch: [72][ 100/ 117] Loss 0.375113 Top1 82.765625 Top5 97.757812 +2023-10-05 21:16:16,566 - Epoch: [72][ 110/ 117] Loss 0.374367 Top1 82.691761 Top5 97.716619 +2023-10-05 21:16:16,652 - Epoch: [72][ 117/ 117] Loss 0.373611 Top1 82.713823 Top5 97.702301 +2023-10-05 21:16:16,750 - ==> Top1: 82.714 Top5: 97.702 Loss: 0.374 + +2023-10-05 21:16:16,751 - ==> Confusion: +[[ 953 3 4 3 6 3 0 0 10 32 1 0 0 5 6 1 9 0 0 0 14] + [ 1 1016 3 0 9 28 2 30 3 0 3 1 0 0 2 2 9 2 7 1 12] + [ 4 0 922 21 3 1 44 9 0 1 4 0 9 2 3 1 8 0 6 4 14] + [ 3 1 15 966 3 5 4 1 2 0 4 0 2 2 25 4 1 6 24 3 18] + [ 30 8 1 0 962 2 0 0 1 6 2 2 0 1 5 5 12 3 0 0 10] + [ 3 30 2 0 8 960 3 33 0 0 6 5 5 12 8 3 6 1 2 7 22] + [ 1 4 30 1 2 0 1120 7 0 0 3 2 1 0 1 4 0 1 4 4 6] + [ 1 21 20 3 2 36 5 1040 0 0 7 8 3 1 5 0 1 0 47 9 9] + [ 20 4 2 0 2 2 0 2 957 30 15 2 1 14 16 3 4 1 6 1 7] + [ 154 4 1 0 10 6 2 1 39 833 0 0 1 31 12 4 1 3 0 0 17] + [ 1 2 8 6 0 2 3 3 13 0 971 0 1 11 4 1 4 2 9 2 10] + [ 1 1 2 0 2 19 0 4 0 0 0 925 29 5 0 3 3 16 0 13 12] + [ 0 0 6 7 0 2 4 3 1 0 5 28 950 2 4 10 3 15 1 6 21] + [ 2 0 2 1 5 12 2 0 13 12 10 3 1 1024 3 2 2 1 0 7 17] + [ 14 0 2 12 6 0 0 0 28 4 3 0 6 3 989 0 2 5 11 0 16] + [ 1 1 5 2 5 0 2 1 0 0 0 5 10 1 0 1059 11 13 0 9 9] + [ 1 10 4 1 4 6 0 0 1 1 0 2 1 2 3 12 1098 0 0 4 11] + [ 0 0 1 1 1 0 0 0 1 0 0 8 24 0 1 6 0 987 1 2 5] + [ 0 6 12 14 1 0 0 32 3 0 1 0 6 0 12 0 1 1 964 1 14] + [ 0 4 4 0 3 5 18 15 1 0 1 8 5 3 2 10 12 0 3 1038 20] + [ 164 191 150 97 98 171 66 100 90 37 162 100 292 285 153 61 197 77 166 215 5033]] + +2023-10-05 21:16:16,752 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:16:16,752 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:16:16,765 - + +2023-10-05 21:16:16,766 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:16:17,760 - Epoch: [73][ 10/ 1236] Overall Loss 0.303913 Objective Loss 0.303913 LR 0.001000 Time 0.099347 +2023-10-05 21:16:17,964 - Epoch: [73][ 20/ 1236] Overall Loss 0.302651 Objective Loss 0.302651 LR 0.001000 Time 0.059852 +2023-10-05 21:16:18,165 - Epoch: [73][ 30/ 1236] Overall Loss 0.314539 Objective Loss 0.314539 LR 0.001000 Time 0.046598 +2023-10-05 21:16:18,369 - Epoch: [73][ 40/ 1236] Overall Loss 0.317470 Objective Loss 0.317470 LR 0.001000 Time 0.040033 +2023-10-05 21:16:18,569 - Epoch: [73][ 50/ 1236] Overall Loss 0.313320 Objective Loss 0.313320 LR 0.001000 Time 0.036037 +2023-10-05 21:16:18,774 - Epoch: [73][ 60/ 1236] Overall Loss 0.313858 Objective Loss 0.313858 LR 0.001000 Time 0.033427 +2023-10-05 21:16:18,974 - Epoch: [73][ 70/ 1236] Overall Loss 0.317577 Objective Loss 0.317577 LR 0.001000 Time 0.031513 +2023-10-05 21:16:19,178 - Epoch: [73][ 80/ 1236] Overall Loss 0.318514 Objective Loss 0.318514 LR 0.001000 Time 0.030122 +2023-10-05 21:16:19,381 - Epoch: [73][ 90/ 1236] Overall Loss 0.320724 Objective Loss 0.320724 LR 0.001000 Time 0.028995 +2023-10-05 21:16:19,586 - Epoch: [73][ 100/ 1236] Overall Loss 0.322226 Objective Loss 0.322226 LR 0.001000 Time 0.028135 +2023-10-05 21:16:19,787 - Epoch: [73][ 110/ 1236] Overall Loss 0.322531 Objective Loss 0.322531 LR 0.001000 Time 0.027401 +2023-10-05 21:16:19,990 - Epoch: [73][ 120/ 1236] Overall Loss 0.324156 Objective Loss 0.324156 LR 0.001000 Time 0.026812 +2023-10-05 21:16:20,191 - Epoch: [73][ 130/ 1236] Overall Loss 0.326879 Objective Loss 0.326879 LR 0.001000 Time 0.026294 +2023-10-05 21:16:20,395 - Epoch: [73][ 140/ 1236] Overall Loss 0.329063 Objective Loss 0.329063 LR 0.001000 Time 0.025870 +2023-10-05 21:16:20,596 - Epoch: [73][ 150/ 1236] Overall Loss 0.328164 Objective Loss 0.328164 LR 0.001000 Time 0.025483 +2023-10-05 21:16:20,800 - Epoch: [73][ 160/ 1236] Overall Loss 0.328319 Objective Loss 0.328319 LR 0.001000 Time 0.025163 +2023-10-05 21:16:21,001 - Epoch: [73][ 170/ 1236] Overall Loss 0.329562 Objective Loss 0.329562 LR 0.001000 Time 0.024863 +2023-10-05 21:16:21,205 - Epoch: [73][ 180/ 1236] Overall Loss 0.330267 Objective Loss 0.330267 LR 0.001000 Time 0.024613 +2023-10-05 21:16:21,406 - Epoch: [73][ 190/ 1236] Overall Loss 0.328721 Objective Loss 0.328721 LR 0.001000 Time 0.024374 +2023-10-05 21:16:21,610 - Epoch: [73][ 200/ 1236] Overall Loss 0.327971 Objective Loss 0.327971 LR 0.001000 Time 0.024173 +2023-10-05 21:16:21,811 - Epoch: [73][ 210/ 1236] Overall Loss 0.329134 Objective Loss 0.329134 LR 0.001000 Time 0.023977 +2023-10-05 21:16:22,015 - Epoch: [73][ 220/ 1236] Overall Loss 0.328341 Objective Loss 0.328341 LR 0.001000 Time 0.023813 +2023-10-05 21:16:22,216 - Epoch: [73][ 230/ 1236] Overall Loss 0.329231 Objective Loss 0.329231 LR 0.001000 Time 0.023650 +2023-10-05 21:16:22,420 - Epoch: [73][ 240/ 1236] Overall Loss 0.330043 Objective Loss 0.330043 LR 0.001000 Time 0.023513 +2023-10-05 21:16:22,621 - Epoch: [73][ 250/ 1236] Overall Loss 0.329162 Objective Loss 0.329162 LR 0.001000 Time 0.023376 +2023-10-05 21:16:22,825 - Epoch: [73][ 260/ 1236] Overall Loss 0.329723 Objective Loss 0.329723 LR 0.001000 Time 0.023260 +2023-10-05 21:16:23,026 - Epoch: [73][ 270/ 1236] Overall Loss 0.329390 Objective Loss 0.329390 LR 0.001000 Time 0.023141 +2023-10-05 21:16:23,230 - Epoch: [73][ 280/ 1236] Overall Loss 0.329685 Objective Loss 0.329685 LR 0.001000 Time 0.023042 +2023-10-05 21:16:23,431 - Epoch: [73][ 290/ 1236] Overall Loss 0.329172 Objective Loss 0.329172 LR 0.001000 Time 0.022940 +2023-10-05 21:16:23,635 - Epoch: [73][ 300/ 1236] Overall Loss 0.329979 Objective Loss 0.329979 LR 0.001000 Time 0.022854 +2023-10-05 21:16:23,836 - Epoch: [73][ 310/ 1236] Overall Loss 0.329964 Objective Loss 0.329964 LR 0.001000 Time 0.022764 +2023-10-05 21:16:24,040 - Epoch: [73][ 320/ 1236] Overall Loss 0.330168 Objective Loss 0.330168 LR 0.001000 Time 0.022689 +2023-10-05 21:16:24,242 - Epoch: [73][ 330/ 1236] Overall Loss 0.330338 Objective Loss 0.330338 LR 0.001000 Time 0.022610 +2023-10-05 21:16:24,445 - Epoch: [73][ 340/ 1236] Overall Loss 0.330190 Objective Loss 0.330190 LR 0.001000 Time 0.022543 +2023-10-05 21:16:24,647 - Epoch: [73][ 350/ 1236] Overall Loss 0.330095 Objective Loss 0.330095 LR 0.001000 Time 0.022473 +2023-10-05 21:16:24,849 - Epoch: [73][ 360/ 1236] Overall Loss 0.329809 Objective Loss 0.329809 LR 0.001000 Time 0.022409 +2023-10-05 21:16:25,049 - Epoch: [73][ 370/ 1236] Overall Loss 0.329194 Objective Loss 0.329194 LR 0.001000 Time 0.022344 +2023-10-05 21:16:25,251 - Epoch: [73][ 380/ 1236] Overall Loss 0.328394 Objective Loss 0.328394 LR 0.001000 Time 0.022287 +2023-10-05 21:16:25,451 - Epoch: [73][ 390/ 1236] Overall Loss 0.328564 Objective Loss 0.328564 LR 0.001000 Time 0.022228 +2023-10-05 21:16:25,654 - Epoch: [73][ 400/ 1236] Overall Loss 0.329482 Objective Loss 0.329482 LR 0.001000 Time 0.022177 +2023-10-05 21:16:25,854 - Epoch: [73][ 410/ 1236] Overall Loss 0.329278 Objective Loss 0.329278 LR 0.001000 Time 0.022123 +2023-10-05 21:16:26,056 - Epoch: [73][ 420/ 1236] Overall Loss 0.329597 Objective Loss 0.329597 LR 0.001000 Time 0.022077 +2023-10-05 21:16:26,256 - Epoch: [73][ 430/ 1236] Overall Loss 0.329794 Objective Loss 0.329794 LR 0.001000 Time 0.022028 +2023-10-05 21:16:26,458 - Epoch: [73][ 440/ 1236] Overall Loss 0.330050 Objective Loss 0.330050 LR 0.001000 Time 0.021986 +2023-10-05 21:16:26,658 - Epoch: [73][ 450/ 1236] Overall Loss 0.330369 Objective Loss 0.330369 LR 0.001000 Time 0.021941 +2023-10-05 21:16:26,861 - Epoch: [73][ 460/ 1236] Overall Loss 0.330560 Objective Loss 0.330560 LR 0.001000 Time 0.021903 +2023-10-05 21:16:27,061 - Epoch: [73][ 470/ 1236] Overall Loss 0.331281 Objective Loss 0.331281 LR 0.001000 Time 0.021863 +2023-10-05 21:16:27,263 - Epoch: [73][ 480/ 1236] Overall Loss 0.331541 Objective Loss 0.331541 LR 0.001000 Time 0.021827 +2023-10-05 21:16:27,463 - Epoch: [73][ 490/ 1236] Overall Loss 0.331199 Objective Loss 0.331199 LR 0.001000 Time 0.021789 +2023-10-05 21:16:27,665 - Epoch: [73][ 500/ 1236] Overall Loss 0.331489 Objective Loss 0.331489 LR 0.001000 Time 0.021756 +2023-10-05 21:16:27,865 - Epoch: [73][ 510/ 1236] Overall Loss 0.331435 Objective Loss 0.331435 LR 0.001000 Time 0.021722 +2023-10-05 21:16:28,067 - Epoch: [73][ 520/ 1236] Overall Loss 0.331701 Objective Loss 0.331701 LR 0.001000 Time 0.021692 +2023-10-05 21:16:28,268 - Epoch: [73][ 530/ 1236] Overall Loss 0.332161 Objective Loss 0.332161 LR 0.001000 Time 0.021660 +2023-10-05 21:16:28,470 - Epoch: [73][ 540/ 1236] Overall Loss 0.331867 Objective Loss 0.331867 LR 0.001000 Time 0.021632 +2023-10-05 21:16:28,670 - Epoch: [73][ 550/ 1236] Overall Loss 0.332147 Objective Loss 0.332147 LR 0.001000 Time 0.021603 +2023-10-05 21:16:28,872 - Epoch: [73][ 560/ 1236] Overall Loss 0.332246 Objective Loss 0.332246 LR 0.001000 Time 0.021578 +2023-10-05 21:16:29,072 - Epoch: [73][ 570/ 1236] Overall Loss 0.332214 Objective Loss 0.332214 LR 0.001000 Time 0.021549 +2023-10-05 21:16:29,275 - Epoch: [73][ 580/ 1236] Overall Loss 0.331426 Objective Loss 0.331426 LR 0.001000 Time 0.021526 +2023-10-05 21:16:29,475 - Epoch: [73][ 590/ 1236] Overall Loss 0.331514 Objective Loss 0.331514 LR 0.001000 Time 0.021500 +2023-10-05 21:16:29,677 - Epoch: [73][ 600/ 1236] Overall Loss 0.331340 Objective Loss 0.331340 LR 0.001000 Time 0.021478 +2023-10-05 21:16:29,877 - Epoch: [73][ 610/ 1236] Overall Loss 0.331162 Objective Loss 0.331162 LR 0.001000 Time 0.021454 +2023-10-05 21:16:30,079 - Epoch: [73][ 620/ 1236] Overall Loss 0.331295 Objective Loss 0.331295 LR 0.001000 Time 0.021433 +2023-10-05 21:16:30,280 - Epoch: [73][ 630/ 1236] Overall Loss 0.331412 Objective Loss 0.331412 LR 0.001000 Time 0.021410 +2023-10-05 21:16:30,482 - Epoch: [73][ 640/ 1236] Overall Loss 0.331784 Objective Loss 0.331784 LR 0.001000 Time 0.021391 +2023-10-05 21:16:30,682 - Epoch: [73][ 650/ 1236] Overall Loss 0.331691 Objective Loss 0.331691 LR 0.001000 Time 0.021369 +2023-10-05 21:16:30,884 - Epoch: [73][ 660/ 1236] Overall Loss 0.331756 Objective Loss 0.331756 LR 0.001000 Time 0.021351 +2023-10-05 21:16:31,084 - Epoch: [73][ 670/ 1236] Overall Loss 0.331544 Objective Loss 0.331544 LR 0.001000 Time 0.021331 +2023-10-05 21:16:31,287 - Epoch: [73][ 680/ 1236] Overall Loss 0.331354 Objective Loss 0.331354 LR 0.001000 Time 0.021314 +2023-10-05 21:16:31,487 - Epoch: [73][ 690/ 1236] Overall Loss 0.330955 Objective Loss 0.330955 LR 0.001000 Time 0.021295 +2023-10-05 21:16:31,689 - Epoch: [73][ 700/ 1236] Overall Loss 0.331041 Objective Loss 0.331041 LR 0.001000 Time 0.021279 +2023-10-05 21:16:31,889 - Epoch: [73][ 710/ 1236] Overall Loss 0.330491 Objective Loss 0.330491 LR 0.001000 Time 0.021260 +2023-10-05 21:16:32,092 - Epoch: [73][ 720/ 1236] Overall Loss 0.330665 Objective Loss 0.330665 LR 0.001000 Time 0.021246 +2023-10-05 21:16:32,292 - Epoch: [73][ 730/ 1236] Overall Loss 0.330905 Objective Loss 0.330905 LR 0.001000 Time 0.021228 +2023-10-05 21:16:32,494 - Epoch: [73][ 740/ 1236] Overall Loss 0.330732 Objective Loss 0.330732 LR 0.001000 Time 0.021214 +2023-10-05 21:16:32,694 - Epoch: [73][ 750/ 1236] Overall Loss 0.330178 Objective Loss 0.330178 LR 0.001000 Time 0.021197 +2023-10-05 21:16:32,896 - Epoch: [73][ 760/ 1236] Overall Loss 0.330202 Objective Loss 0.330202 LR 0.001000 Time 0.021184 +2023-10-05 21:16:33,097 - Epoch: [73][ 770/ 1236] Overall Loss 0.329452 Objective Loss 0.329452 LR 0.001000 Time 0.021169 +2023-10-05 21:16:33,299 - Epoch: [73][ 780/ 1236] Overall Loss 0.329611 Objective Loss 0.329611 LR 0.001000 Time 0.021157 +2023-10-05 21:16:33,499 - Epoch: [73][ 790/ 1236] Overall Loss 0.329462 Objective Loss 0.329462 LR 0.001000 Time 0.021142 +2023-10-05 21:16:33,702 - Epoch: [73][ 800/ 1236] Overall Loss 0.329850 Objective Loss 0.329850 LR 0.001000 Time 0.021130 +2023-10-05 21:16:33,902 - Epoch: [73][ 810/ 1236] Overall Loss 0.330154 Objective Loss 0.330154 LR 0.001000 Time 0.021116 +2023-10-05 21:16:34,104 - Epoch: [73][ 820/ 1236] Overall Loss 0.330426 Objective Loss 0.330426 LR 0.001000 Time 0.021105 +2023-10-05 21:16:34,305 - Epoch: [73][ 830/ 1236] Overall Loss 0.331419 Objective Loss 0.331419 LR 0.001000 Time 0.021091 +2023-10-05 21:16:34,507 - Epoch: [73][ 840/ 1236] Overall Loss 0.331418 Objective Loss 0.331418 LR 0.001000 Time 0.021080 +2023-10-05 21:16:34,707 - Epoch: [73][ 850/ 1236] Overall Loss 0.330760 Objective Loss 0.330760 LR 0.001000 Time 0.021068 +2023-10-05 21:16:34,909 - Epoch: [73][ 860/ 1236] Overall Loss 0.330881 Objective Loss 0.330881 LR 0.001000 Time 0.021058 +2023-10-05 21:16:35,110 - Epoch: [73][ 870/ 1236] Overall Loss 0.330308 Objective Loss 0.330308 LR 0.001000 Time 0.021046 +2023-10-05 21:16:35,312 - Epoch: [73][ 880/ 1236] Overall Loss 0.330126 Objective Loss 0.330126 LR 0.001000 Time 0.021036 +2023-10-05 21:16:35,512 - Epoch: [73][ 890/ 1236] Overall Loss 0.329839 Objective Loss 0.329839 LR 0.001000 Time 0.021024 +2023-10-05 21:16:35,715 - Epoch: [73][ 900/ 1236] Overall Loss 0.329536 Objective Loss 0.329536 LR 0.001000 Time 0.021015 +2023-10-05 21:16:35,915 - Epoch: [73][ 910/ 1236] Overall Loss 0.329603 Objective Loss 0.329603 LR 0.001000 Time 0.021003 +2023-10-05 21:16:36,117 - Epoch: [73][ 920/ 1236] Overall Loss 0.329630 Objective Loss 0.329630 LR 0.001000 Time 0.020995 +2023-10-05 21:16:36,318 - Epoch: [73][ 930/ 1236] Overall Loss 0.329923 Objective Loss 0.329923 LR 0.001000 Time 0.020984 +2023-10-05 21:16:36,520 - Epoch: [73][ 940/ 1236] Overall Loss 0.330297 Objective Loss 0.330297 LR 0.001000 Time 0.020976 +2023-10-05 21:16:36,721 - Epoch: [73][ 950/ 1236] Overall Loss 0.330298 Objective Loss 0.330298 LR 0.001000 Time 0.020966 +2023-10-05 21:16:36,923 - Epoch: [73][ 960/ 1236] Overall Loss 0.330424 Objective Loss 0.330424 LR 0.001000 Time 0.020957 +2023-10-05 21:16:37,123 - Epoch: [73][ 970/ 1236] Overall Loss 0.330818 Objective Loss 0.330818 LR 0.001000 Time 0.020947 +2023-10-05 21:16:37,325 - Epoch: [73][ 980/ 1236] Overall Loss 0.331193 Objective Loss 0.331193 LR 0.001000 Time 0.020940 +2023-10-05 21:16:37,526 - Epoch: [73][ 990/ 1236] Overall Loss 0.331143 Objective Loss 0.331143 LR 0.001000 Time 0.020930 +2023-10-05 21:16:37,728 - Epoch: [73][ 1000/ 1236] Overall Loss 0.331212 Objective Loss 0.331212 LR 0.001000 Time 0.020922 +2023-10-05 21:16:37,928 - Epoch: [73][ 1010/ 1236] Overall Loss 0.331158 Objective Loss 0.331158 LR 0.001000 Time 0.020913 +2023-10-05 21:16:38,130 - Epoch: [73][ 1020/ 1236] Overall Loss 0.331365 Objective Loss 0.331365 LR 0.001000 Time 0.020906 +2023-10-05 21:16:38,331 - Epoch: [73][ 1030/ 1236] Overall Loss 0.331428 Objective Loss 0.331428 LR 0.001000 Time 0.020897 +2023-10-05 21:16:38,533 - Epoch: [73][ 1040/ 1236] Overall Loss 0.331439 Objective Loss 0.331439 LR 0.001000 Time 0.020890 +2023-10-05 21:16:38,733 - Epoch: [73][ 1050/ 1236] Overall Loss 0.331240 Objective Loss 0.331240 LR 0.001000 Time 0.020882 +2023-10-05 21:16:38,936 - Epoch: [73][ 1060/ 1236] Overall Loss 0.331128 Objective Loss 0.331128 LR 0.001000 Time 0.020876 +2023-10-05 21:16:39,136 - Epoch: [73][ 1070/ 1236] Overall Loss 0.330782 Objective Loss 0.330782 LR 0.001000 Time 0.020867 +2023-10-05 21:16:39,338 - Epoch: [73][ 1080/ 1236] Overall Loss 0.330573 Objective Loss 0.330573 LR 0.001000 Time 0.020861 +2023-10-05 21:16:39,538 - Epoch: [73][ 1090/ 1236] Overall Loss 0.330635 Objective Loss 0.330635 LR 0.001000 Time 0.020853 +2023-10-05 21:16:39,740 - Epoch: [73][ 1100/ 1236] Overall Loss 0.330637 Objective Loss 0.330637 LR 0.001000 Time 0.020846 +2023-10-05 21:16:39,940 - Epoch: [73][ 1110/ 1236] Overall Loss 0.330581 Objective Loss 0.330581 LR 0.001000 Time 0.020839 +2023-10-05 21:16:40,142 - Epoch: [73][ 1120/ 1236] Overall Loss 0.330670 Objective Loss 0.330670 LR 0.001000 Time 0.020833 +2023-10-05 21:16:40,343 - Epoch: [73][ 1130/ 1236] Overall Loss 0.330827 Objective Loss 0.330827 LR 0.001000 Time 0.020825 +2023-10-05 21:16:40,545 - Epoch: [73][ 1140/ 1236] Overall Loss 0.331010 Objective Loss 0.331010 LR 0.001000 Time 0.020820 +2023-10-05 21:16:40,745 - Epoch: [73][ 1150/ 1236] Overall Loss 0.331298 Objective Loss 0.331298 LR 0.001000 Time 0.020813 +2023-10-05 21:16:40,947 - Epoch: [73][ 1160/ 1236] Overall Loss 0.331151 Objective Loss 0.331151 LR 0.001000 Time 0.020807 +2023-10-05 21:16:41,148 - Epoch: [73][ 1170/ 1236] Overall Loss 0.331026 Objective Loss 0.331026 LR 0.001000 Time 0.020800 +2023-10-05 21:16:41,350 - Epoch: [73][ 1180/ 1236] Overall Loss 0.330959 Objective Loss 0.330959 LR 0.001000 Time 0.020795 +2023-10-05 21:16:41,550 - Epoch: [73][ 1190/ 1236] Overall Loss 0.330960 Objective Loss 0.330960 LR 0.001000 Time 0.020788 +2023-10-05 21:16:41,752 - Epoch: [73][ 1200/ 1236] Overall Loss 0.331099 Objective Loss 0.331099 LR 0.001000 Time 0.020783 +2023-10-05 21:16:41,957 - Epoch: [73][ 1210/ 1236] Overall Loss 0.331411 Objective Loss 0.331411 LR 0.001000 Time 0.020780 +2023-10-05 21:16:42,159 - Epoch: [73][ 1220/ 1236] Overall Loss 0.331363 Objective Loss 0.331363 LR 0.001000 Time 0.020775 +2023-10-05 21:16:42,414 - Epoch: [73][ 1230/ 1236] Overall Loss 0.331044 Objective Loss 0.331044 LR 0.001000 Time 0.020813 +2023-10-05 21:16:42,534 - Epoch: [73][ 1236/ 1236] Overall Loss 0.331079 Objective Loss 0.331079 Top1 85.336049 Top5 97.759674 LR 0.001000 Time 0.020809 +2023-10-05 21:16:42,660 - --- validate (epoch=73)----------- +2023-10-05 21:16:42,660 - 29943 samples (256 per mini-batch) +2023-10-05 21:16:43,133 - Epoch: [73][ 10/ 117] Loss 0.344213 Top1 81.289062 Top5 97.734375 +2023-10-05 21:16:43,281 - Epoch: [73][ 20/ 117] Loss 0.347618 Top1 81.542969 Top5 97.773438 +2023-10-05 21:16:43,429 - Epoch: [73][ 30/ 117] Loss 0.344305 Top1 81.770833 Top5 97.591146 +2023-10-05 21:16:43,576 - Epoch: [73][ 40/ 117] Loss 0.352168 Top1 81.367188 Top5 97.470703 +2023-10-05 21:16:43,731 - Epoch: [73][ 50/ 117] Loss 0.351455 Top1 81.421875 Top5 97.531250 +2023-10-05 21:16:43,882 - Epoch: [73][ 60/ 117] Loss 0.355862 Top1 81.282552 Top5 97.382812 +2023-10-05 21:16:44,030 - Epoch: [73][ 70/ 117] Loss 0.357208 Top1 81.344866 Top5 97.427455 +2023-10-05 21:16:44,177 - Epoch: [73][ 80/ 117] Loss 0.358214 Top1 81.396484 Top5 97.436523 +2023-10-05 21:16:44,325 - Epoch: [73][ 90/ 117] Loss 0.361947 Top1 81.293403 Top5 97.426215 +2023-10-05 21:16:44,471 - Epoch: [73][ 100/ 117] Loss 0.362434 Top1 81.316406 Top5 97.417969 +2023-10-05 21:16:44,626 - Epoch: [73][ 110/ 117] Loss 0.362695 Top1 81.253551 Top5 97.450284 +2023-10-05 21:16:44,712 - Epoch: [73][ 117/ 117] Loss 0.363958 Top1 81.254383 Top5 97.441806 +2023-10-05 21:16:44,846 - ==> Top1: 81.254 Top5: 97.442 Loss: 0.364 + +2023-10-05 21:16:44,847 - ==> Confusion: +[[ 936 1 2 2 9 9 0 1 5 56 2 0 1 4 1 4 2 1 0 0 14] + [ 2 1048 0 0 7 17 5 26 1 0 3 1 0 1 0 3 3 0 7 4 3] + [ 5 1 926 11 5 3 36 17 0 0 10 4 5 3 1 7 3 1 3 5 10] + [ 1 3 25 933 4 6 1 0 7 0 13 0 10 4 21 5 1 10 25 3 17] + [ 31 8 1 1 969 5 1 2 0 4 0 3 1 2 5 2 9 2 0 0 4] + [ 5 42 0 0 3 982 2 21 1 1 4 9 3 18 4 0 1 0 3 8 9] + [ 0 6 27 0 1 0 1115 7 0 0 4 5 1 0 1 4 1 3 3 10 3] + [ 6 20 14 0 3 40 3 1048 0 4 4 9 2 3 0 1 1 1 38 15 6] + [ 23 5 0 0 1 7 0 1 929 45 20 4 3 21 13 1 3 2 8 0 3] + [ 133 1 1 0 8 5 1 1 25 893 1 2 0 20 5 5 4 1 0 2 11] + [ 5 6 7 4 1 1 4 4 12 0 972 2 0 13 2 0 1 0 6 3 10] + [ 1 1 0 0 1 14 0 1 0 0 0 925 50 7 0 0 3 16 0 11 5] + [ 0 1 1 3 1 2 0 3 1 0 2 48 970 2 0 6 1 11 2 8 6] + [ 3 1 2 0 3 7 0 0 8 16 18 8 1 1031 3 0 0 0 0 6 12] + [ 16 3 4 13 11 2 0 0 29 4 6 1 6 2 963 0 3 1 15 0 22] + [ 2 4 1 1 6 1 2 0 0 0 0 8 8 2 0 1058 12 13 0 11 5] + [ 2 22 3 2 15 5 1 0 1 0 0 6 2 1 2 7 1074 0 0 8 10] + [ 0 1 1 2 0 0 0 1 1 0 1 7 18 3 2 10 1 983 2 1 4] + [ 0 7 12 12 3 1 0 40 5 0 9 1 4 0 8 0 1 0 953 0 12] + [ 0 2 2 1 1 5 7 12 0 0 4 20 8 4 0 3 6 0 0 1071 6] + [ 159 302 139 58 120 229 66 122 97 78 205 133 401 349 101 85 194 60 184 272 4551]] + +2023-10-05 21:16:44,848 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:16:44,848 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:16:44,854 - + +2023-10-05 21:16:44,854 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:16:45,992 - Epoch: [74][ 10/ 1236] Overall Loss 0.335262 Objective Loss 0.335262 LR 0.001000 Time 0.113669 +2023-10-05 21:16:46,196 - Epoch: [74][ 20/ 1236] Overall Loss 0.333249 Objective Loss 0.333249 LR 0.001000 Time 0.067046 +2023-10-05 21:16:46,403 - Epoch: [74][ 30/ 1236] Overall Loss 0.326515 Objective Loss 0.326515 LR 0.001000 Time 0.051575 +2023-10-05 21:16:46,608 - Epoch: [74][ 40/ 1236] Overall Loss 0.327679 Objective Loss 0.327679 LR 0.001000 Time 0.043803 +2023-10-05 21:16:46,816 - Epoch: [74][ 50/ 1236] Overall Loss 0.320267 Objective Loss 0.320267 LR 0.001000 Time 0.039186 +2023-10-05 21:16:47,022 - Epoch: [74][ 60/ 1236] Overall Loss 0.313327 Objective Loss 0.313327 LR 0.001000 Time 0.036082 +2023-10-05 21:16:47,229 - Epoch: [74][ 70/ 1236] Overall Loss 0.319694 Objective Loss 0.319694 LR 0.001000 Time 0.033885 +2023-10-05 21:16:47,451 - Epoch: [74][ 80/ 1236] Overall Loss 0.322330 Objective Loss 0.322330 LR 0.001000 Time 0.032418 +2023-10-05 21:16:47,673 - Epoch: [74][ 90/ 1236] Overall Loss 0.323049 Objective Loss 0.323049 LR 0.001000 Time 0.031276 +2023-10-05 21:16:47,894 - Epoch: [74][ 100/ 1236] Overall Loss 0.319263 Objective Loss 0.319263 LR 0.001000 Time 0.030357 +2023-10-05 21:16:48,113 - Epoch: [74][ 110/ 1236] Overall Loss 0.319161 Objective Loss 0.319161 LR 0.001000 Time 0.029579 +2023-10-05 21:16:48,336 - Epoch: [74][ 120/ 1236] Overall Loss 0.321218 Objective Loss 0.321218 LR 0.001000 Time 0.028969 +2023-10-05 21:16:48,549 - Epoch: [74][ 130/ 1236] Overall Loss 0.321630 Objective Loss 0.321630 LR 0.001000 Time 0.028381 +2023-10-05 21:16:48,768 - Epoch: [74][ 140/ 1236] Overall Loss 0.321470 Objective Loss 0.321470 LR 0.001000 Time 0.027910 +2023-10-05 21:16:48,975 - Epoch: [74][ 150/ 1236] Overall Loss 0.320569 Objective Loss 0.320569 LR 0.001000 Time 0.027427 +2023-10-05 21:16:49,191 - Epoch: [74][ 160/ 1236] Overall Loss 0.320365 Objective Loss 0.320365 LR 0.001000 Time 0.027062 +2023-10-05 21:16:49,402 - Epoch: [74][ 170/ 1236] Overall Loss 0.319526 Objective Loss 0.319526 LR 0.001000 Time 0.026705 +2023-10-05 21:16:49,611 - Epoch: [74][ 180/ 1236] Overall Loss 0.319704 Objective Loss 0.319704 LR 0.001000 Time 0.026382 +2023-10-05 21:16:49,821 - Epoch: [74][ 190/ 1236] Overall Loss 0.318792 Objective Loss 0.318792 LR 0.001000 Time 0.026093 +2023-10-05 21:16:50,029 - Epoch: [74][ 200/ 1236] Overall Loss 0.318664 Objective Loss 0.318664 LR 0.001000 Time 0.025825 +2023-10-05 21:16:50,242 - Epoch: [74][ 210/ 1236] Overall Loss 0.317636 Objective Loss 0.317636 LR 0.001000 Time 0.025606 +2023-10-05 21:16:50,461 - Epoch: [74][ 220/ 1236] Overall Loss 0.317570 Objective Loss 0.317570 LR 0.001000 Time 0.025436 +2023-10-05 21:16:50,689 - Epoch: [74][ 230/ 1236] Overall Loss 0.317939 Objective Loss 0.317939 LR 0.001000 Time 0.025322 +2023-10-05 21:16:50,915 - Epoch: [74][ 240/ 1236] Overall Loss 0.319256 Objective Loss 0.319256 LR 0.001000 Time 0.025207 +2023-10-05 21:16:51,122 - Epoch: [74][ 250/ 1236] Overall Loss 0.318483 Objective Loss 0.318483 LR 0.001000 Time 0.025026 +2023-10-05 21:16:51,326 - Epoch: [74][ 260/ 1236] Overall Loss 0.318478 Objective Loss 0.318478 LR 0.001000 Time 0.024844 +2023-10-05 21:16:51,530 - Epoch: [74][ 270/ 1236] Overall Loss 0.320055 Objective Loss 0.320055 LR 0.001000 Time 0.024677 +2023-10-05 21:16:51,733 - Epoch: [74][ 280/ 1236] Overall Loss 0.320042 Objective Loss 0.320042 LR 0.001000 Time 0.024522 +2023-10-05 21:16:51,937 - Epoch: [74][ 290/ 1236] Overall Loss 0.319060 Objective Loss 0.319060 LR 0.001000 Time 0.024378 +2023-10-05 21:16:52,142 - Epoch: [74][ 300/ 1236] Overall Loss 0.318358 Objective Loss 0.318358 LR 0.001000 Time 0.024245 +2023-10-05 21:16:52,345 - Epoch: [74][ 310/ 1236] Overall Loss 0.318249 Objective Loss 0.318249 LR 0.001000 Time 0.024119 +2023-10-05 21:16:52,549 - Epoch: [74][ 320/ 1236] Overall Loss 0.319399 Objective Loss 0.319399 LR 0.001000 Time 0.024001 +2023-10-05 21:16:52,753 - Epoch: [74][ 330/ 1236] Overall Loss 0.319169 Objective Loss 0.319169 LR 0.001000 Time 0.023891 +2023-10-05 21:16:52,957 - Epoch: [74][ 340/ 1236] Overall Loss 0.319947 Objective Loss 0.319947 LR 0.001000 Time 0.023787 +2023-10-05 21:16:53,164 - Epoch: [74][ 350/ 1236] Overall Loss 0.319335 Objective Loss 0.319335 LR 0.001000 Time 0.023698 +2023-10-05 21:16:53,374 - Epoch: [74][ 360/ 1236] Overall Loss 0.319163 Objective Loss 0.319163 LR 0.001000 Time 0.023623 +2023-10-05 21:16:53,588 - Epoch: [74][ 370/ 1236] Overall Loss 0.318789 Objective Loss 0.318789 LR 0.001000 Time 0.023561 +2023-10-05 21:16:53,800 - Epoch: [74][ 380/ 1236] Overall Loss 0.318906 Objective Loss 0.318906 LR 0.001000 Time 0.023497 +2023-10-05 21:16:54,016 - Epoch: [74][ 390/ 1236] Overall Loss 0.320250 Objective Loss 0.320250 LR 0.001000 Time 0.023447 +2023-10-05 21:16:54,220 - Epoch: [74][ 400/ 1236] Overall Loss 0.320312 Objective Loss 0.320312 LR 0.001000 Time 0.023370 +2023-10-05 21:16:54,426 - Epoch: [74][ 410/ 1236] Overall Loss 0.320429 Objective Loss 0.320429 LR 0.001000 Time 0.023301 +2023-10-05 21:16:54,629 - Epoch: [74][ 420/ 1236] Overall Loss 0.321170 Objective Loss 0.321170 LR 0.001000 Time 0.023230 +2023-10-05 21:16:54,834 - Epoch: [74][ 430/ 1236] Overall Loss 0.321675 Objective Loss 0.321675 LR 0.001000 Time 0.023164 +2023-10-05 21:16:55,040 - Epoch: [74][ 440/ 1236] Overall Loss 0.322422 Objective Loss 0.322422 LR 0.001000 Time 0.023106 +2023-10-05 21:16:55,250 - Epoch: [74][ 450/ 1236] Overall Loss 0.322439 Objective Loss 0.322439 LR 0.001000 Time 0.023058 +2023-10-05 21:16:55,471 - Epoch: [74][ 460/ 1236] Overall Loss 0.323272 Objective Loss 0.323272 LR 0.001000 Time 0.023037 +2023-10-05 21:16:55,709 - Epoch: [74][ 470/ 1236] Overall Loss 0.323383 Objective Loss 0.323383 LR 0.001000 Time 0.023053 +2023-10-05 21:16:55,941 - Epoch: [74][ 480/ 1236] Overall Loss 0.323660 Objective Loss 0.323660 LR 0.001000 Time 0.023055 +2023-10-05 21:16:56,168 - Epoch: [74][ 490/ 1236] Overall Loss 0.324211 Objective Loss 0.324211 LR 0.001000 Time 0.023047 +2023-10-05 21:16:56,385 - Epoch: [74][ 500/ 1236] Overall Loss 0.323863 Objective Loss 0.323863 LR 0.001000 Time 0.023019 +2023-10-05 21:16:56,595 - Epoch: [74][ 510/ 1236] Overall Loss 0.323907 Objective Loss 0.323907 LR 0.001000 Time 0.022979 +2023-10-05 21:16:56,805 - Epoch: [74][ 520/ 1236] Overall Loss 0.324321 Objective Loss 0.324321 LR 0.001000 Time 0.022939 +2023-10-05 21:16:57,006 - Epoch: [74][ 530/ 1236] Overall Loss 0.324866 Objective Loss 0.324866 LR 0.001000 Time 0.022886 +2023-10-05 21:16:57,209 - Epoch: [74][ 540/ 1236] Overall Loss 0.324183 Objective Loss 0.324183 LR 0.001000 Time 0.022837 +2023-10-05 21:16:57,411 - Epoch: [74][ 550/ 1236] Overall Loss 0.323505 Objective Loss 0.323505 LR 0.001000 Time 0.022790 +2023-10-05 21:16:57,613 - Epoch: [74][ 560/ 1236] Overall Loss 0.323087 Objective Loss 0.323087 LR 0.001000 Time 0.022742 +2023-10-05 21:16:57,815 - Epoch: [74][ 570/ 1236] Overall Loss 0.323390 Objective Loss 0.323390 LR 0.001000 Time 0.022697 +2023-10-05 21:16:58,017 - Epoch: [74][ 580/ 1236] Overall Loss 0.322836 Objective Loss 0.322836 LR 0.001000 Time 0.022654 +2023-10-05 21:16:58,220 - Epoch: [74][ 590/ 1236] Overall Loss 0.322673 Objective Loss 0.322673 LR 0.001000 Time 0.022613 +2023-10-05 21:16:58,422 - Epoch: [74][ 600/ 1236] Overall Loss 0.322596 Objective Loss 0.322596 LR 0.001000 Time 0.022573 +2023-10-05 21:16:58,625 - Epoch: [74][ 610/ 1236] Overall Loss 0.322081 Objective Loss 0.322081 LR 0.001000 Time 0.022535 +2023-10-05 21:16:58,829 - Epoch: [74][ 620/ 1236] Overall Loss 0.321974 Objective Loss 0.321974 LR 0.001000 Time 0.022500 +2023-10-05 21:16:59,031 - Epoch: [74][ 630/ 1236] Overall Loss 0.322062 Objective Loss 0.322062 LR 0.001000 Time 0.022464 +2023-10-05 21:16:59,235 - Epoch: [74][ 640/ 1236] Overall Loss 0.321857 Objective Loss 0.321857 LR 0.001000 Time 0.022431 +2023-10-05 21:16:59,437 - Epoch: [74][ 650/ 1236] Overall Loss 0.321798 Objective Loss 0.321798 LR 0.001000 Time 0.022396 +2023-10-05 21:16:59,640 - Epoch: [74][ 660/ 1236] Overall Loss 0.321959 Objective Loss 0.321959 LR 0.001000 Time 0.022363 +2023-10-05 21:16:59,842 - Epoch: [74][ 670/ 1236] Overall Loss 0.321653 Objective Loss 0.321653 LR 0.001000 Time 0.022330 +2023-10-05 21:17:00,044 - Epoch: [74][ 680/ 1236] Overall Loss 0.321703 Objective Loss 0.321703 LR 0.001000 Time 0.022299 +2023-10-05 21:17:00,245 - Epoch: [74][ 690/ 1236] Overall Loss 0.322175 Objective Loss 0.322175 LR 0.001000 Time 0.022267 +2023-10-05 21:17:00,445 - Epoch: [74][ 700/ 1236] Overall Loss 0.321964 Objective Loss 0.321964 LR 0.001000 Time 0.022234 +2023-10-05 21:17:00,645 - Epoch: [74][ 710/ 1236] Overall Loss 0.321926 Objective Loss 0.321926 LR 0.001000 Time 0.022202 +2023-10-05 21:17:00,845 - Epoch: [74][ 720/ 1236] Overall Loss 0.322517 Objective Loss 0.322517 LR 0.001000 Time 0.022171 +2023-10-05 21:17:01,045 - Epoch: [74][ 730/ 1236] Overall Loss 0.322149 Objective Loss 0.322149 LR 0.001000 Time 0.022141 +2023-10-05 21:17:01,245 - Epoch: [74][ 740/ 1236] Overall Loss 0.321678 Objective Loss 0.321678 LR 0.001000 Time 0.022112 +2023-10-05 21:17:01,445 - Epoch: [74][ 750/ 1236] Overall Loss 0.321478 Objective Loss 0.321478 LR 0.001000 Time 0.022084 +2023-10-05 21:17:01,646 - Epoch: [74][ 760/ 1236] Overall Loss 0.321423 Objective Loss 0.321423 LR 0.001000 Time 0.022057 +2023-10-05 21:17:01,849 - Epoch: [74][ 770/ 1236] Overall Loss 0.321749 Objective Loss 0.321749 LR 0.001000 Time 0.022034 +2023-10-05 21:17:02,052 - Epoch: [74][ 780/ 1236] Overall Loss 0.321325 Objective Loss 0.321325 LR 0.001000 Time 0.022011 +2023-10-05 21:17:02,255 - Epoch: [74][ 790/ 1236] Overall Loss 0.321821 Objective Loss 0.321821 LR 0.001000 Time 0.021989 +2023-10-05 21:17:02,456 - Epoch: [74][ 800/ 1236] Overall Loss 0.322124 Objective Loss 0.322124 LR 0.001000 Time 0.021966 +2023-10-05 21:17:02,659 - Epoch: [74][ 810/ 1236] Overall Loss 0.322136 Objective Loss 0.322136 LR 0.001000 Time 0.021945 +2023-10-05 21:17:02,861 - Epoch: [74][ 820/ 1236] Overall Loss 0.321889 Objective Loss 0.321889 LR 0.001000 Time 0.021922 +2023-10-05 21:17:03,063 - Epoch: [74][ 830/ 1236] Overall Loss 0.321900 Objective Loss 0.321900 LR 0.001000 Time 0.021902 +2023-10-05 21:17:03,264 - Epoch: [74][ 840/ 1236] Overall Loss 0.321482 Objective Loss 0.321482 LR 0.001000 Time 0.021880 +2023-10-05 21:17:03,464 - Epoch: [74][ 850/ 1236] Overall Loss 0.321043 Objective Loss 0.321043 LR 0.001000 Time 0.021857 +2023-10-05 21:17:03,665 - Epoch: [74][ 860/ 1236] Overall Loss 0.320897 Objective Loss 0.320897 LR 0.001000 Time 0.021836 +2023-10-05 21:17:03,865 - Epoch: [74][ 870/ 1236] Overall Loss 0.321270 Objective Loss 0.321270 LR 0.001000 Time 0.021815 +2023-10-05 21:17:04,065 - Epoch: [74][ 880/ 1236] Overall Loss 0.322033 Objective Loss 0.322033 LR 0.001000 Time 0.021794 +2023-10-05 21:17:04,264 - Epoch: [74][ 890/ 1236] Overall Loss 0.322065 Objective Loss 0.322065 LR 0.001000 Time 0.021773 +2023-10-05 21:17:04,465 - Epoch: [74][ 900/ 1236] Overall Loss 0.321980 Objective Loss 0.321980 LR 0.001000 Time 0.021753 +2023-10-05 21:17:04,664 - Epoch: [74][ 910/ 1236] Overall Loss 0.321680 Objective Loss 0.321680 LR 0.001000 Time 0.021733 +2023-10-05 21:17:04,865 - Epoch: [74][ 920/ 1236] Overall Loss 0.321913 Objective Loss 0.321913 LR 0.001000 Time 0.021715 +2023-10-05 21:17:05,064 - Epoch: [74][ 930/ 1236] Overall Loss 0.321649 Objective Loss 0.321649 LR 0.001000 Time 0.021695 +2023-10-05 21:17:05,265 - Epoch: [74][ 940/ 1236] Overall Loss 0.321694 Objective Loss 0.321694 LR 0.001000 Time 0.021678 +2023-10-05 21:17:05,464 - Epoch: [74][ 950/ 1236] Overall Loss 0.321603 Objective Loss 0.321603 LR 0.001000 Time 0.021659 +2023-10-05 21:17:05,665 - Epoch: [74][ 960/ 1236] Overall Loss 0.321806 Objective Loss 0.321806 LR 0.001000 Time 0.021642 +2023-10-05 21:17:05,864 - Epoch: [74][ 970/ 1236] Overall Loss 0.321924 Objective Loss 0.321924 LR 0.001000 Time 0.021625 +2023-10-05 21:17:06,065 - Epoch: [74][ 980/ 1236] Overall Loss 0.322203 Objective Loss 0.322203 LR 0.001000 Time 0.021608 +2023-10-05 21:17:06,264 - Epoch: [74][ 990/ 1236] Overall Loss 0.322599 Objective Loss 0.322599 LR 0.001000 Time 0.021591 +2023-10-05 21:17:06,465 - Epoch: [74][ 1000/ 1236] Overall Loss 0.322911 Objective Loss 0.322911 LR 0.001000 Time 0.021576 +2023-10-05 21:17:06,665 - Epoch: [74][ 1010/ 1236] Overall Loss 0.323305 Objective Loss 0.323305 LR 0.001000 Time 0.021560 +2023-10-05 21:17:06,866 - Epoch: [74][ 1020/ 1236] Overall Loss 0.323417 Objective Loss 0.323417 LR 0.001000 Time 0.021545 +2023-10-05 21:17:07,065 - Epoch: [74][ 1030/ 1236] Overall Loss 0.323593 Objective Loss 0.323593 LR 0.001000 Time 0.021529 +2023-10-05 21:17:07,266 - Epoch: [74][ 1040/ 1236] Overall Loss 0.323721 Objective Loss 0.323721 LR 0.001000 Time 0.021515 +2023-10-05 21:17:07,467 - Epoch: [74][ 1050/ 1236] Overall Loss 0.323741 Objective Loss 0.323741 LR 0.001000 Time 0.021502 +2023-10-05 21:17:07,667 - Epoch: [74][ 1060/ 1236] Overall Loss 0.323763 Objective Loss 0.323763 LR 0.001000 Time 0.021487 +2023-10-05 21:17:07,871 - Epoch: [74][ 1070/ 1236] Overall Loss 0.323867 Objective Loss 0.323867 LR 0.001000 Time 0.021476 +2023-10-05 21:17:08,077 - Epoch: [74][ 1080/ 1236] Overall Loss 0.323954 Objective Loss 0.323954 LR 0.001000 Time 0.021468 +2023-10-05 21:17:08,279 - Epoch: [74][ 1090/ 1236] Overall Loss 0.324156 Objective Loss 0.324156 LR 0.001000 Time 0.021456 +2023-10-05 21:17:08,480 - Epoch: [74][ 1100/ 1236] Overall Loss 0.324379 Objective Loss 0.324379 LR 0.001000 Time 0.021443 +2023-10-05 21:17:08,679 - Epoch: [74][ 1110/ 1236] Overall Loss 0.324847 Objective Loss 0.324847 LR 0.001000 Time 0.021430 +2023-10-05 21:17:08,879 - Epoch: [74][ 1120/ 1236] Overall Loss 0.324987 Objective Loss 0.324987 LR 0.001000 Time 0.021417 +2023-10-05 21:17:09,079 - Epoch: [74][ 1130/ 1236] Overall Loss 0.325067 Objective Loss 0.325067 LR 0.001000 Time 0.021404 +2023-10-05 21:17:09,279 - Epoch: [74][ 1140/ 1236] Overall Loss 0.325365 Objective Loss 0.325365 LR 0.001000 Time 0.021392 +2023-10-05 21:17:09,479 - Epoch: [74][ 1150/ 1236] Overall Loss 0.325670 Objective Loss 0.325670 LR 0.001000 Time 0.021379 +2023-10-05 21:17:09,679 - Epoch: [74][ 1160/ 1236] Overall Loss 0.325944 Objective Loss 0.325944 LR 0.001000 Time 0.021367 +2023-10-05 21:17:09,879 - Epoch: [74][ 1170/ 1236] Overall Loss 0.325948 Objective Loss 0.325948 LR 0.001000 Time 0.021355 +2023-10-05 21:17:10,080 - Epoch: [74][ 1180/ 1236] Overall Loss 0.325826 Objective Loss 0.325826 LR 0.001000 Time 0.021344 +2023-10-05 21:17:10,281 - Epoch: [74][ 1190/ 1236] Overall Loss 0.325698 Objective Loss 0.325698 LR 0.001000 Time 0.021333 +2023-10-05 21:17:10,480 - Epoch: [74][ 1200/ 1236] Overall Loss 0.325803 Objective Loss 0.325803 LR 0.001000 Time 0.021321 +2023-10-05 21:17:10,680 - Epoch: [74][ 1210/ 1236] Overall Loss 0.325747 Objective Loss 0.325747 LR 0.001000 Time 0.021310 +2023-10-05 21:17:10,880 - Epoch: [74][ 1220/ 1236] Overall Loss 0.325881 Objective Loss 0.325881 LR 0.001000 Time 0.021299 +2023-10-05 21:17:11,137 - Epoch: [74][ 1230/ 1236] Overall Loss 0.326049 Objective Loss 0.326049 LR 0.001000 Time 0.021335 +2023-10-05 21:17:11,254 - Epoch: [74][ 1236/ 1236] Overall Loss 0.325906 Objective Loss 0.325906 Top1 84.725051 Top5 98.167006 LR 0.001000 Time 0.021325 +2023-10-05 21:17:11,385 - --- validate (epoch=74)----------- +2023-10-05 21:17:11,385 - 29943 samples (256 per mini-batch) +2023-10-05 21:17:11,852 - Epoch: [74][ 10/ 117] Loss 0.382815 Top1 81.250000 Top5 97.304688 +2023-10-05 21:17:11,998 - Epoch: [74][ 20/ 117] Loss 0.361905 Top1 81.660156 Top5 97.363281 +2023-10-05 21:17:12,145 - Epoch: [74][ 30/ 117] Loss 0.369423 Top1 81.223958 Top5 97.343750 +2023-10-05 21:17:12,296 - Epoch: [74][ 40/ 117] Loss 0.365402 Top1 81.367188 Top5 97.392578 +2023-10-05 21:17:12,442 - Epoch: [74][ 50/ 117] Loss 0.371538 Top1 81.164062 Top5 97.515625 +2023-10-05 21:17:12,589 - Epoch: [74][ 60/ 117] Loss 0.372309 Top1 81.061198 Top5 97.447917 +2023-10-05 21:17:12,737 - Epoch: [74][ 70/ 117] Loss 0.370706 Top1 81.121652 Top5 97.438616 +2023-10-05 21:17:12,884 - Epoch: [74][ 80/ 117] Loss 0.372350 Top1 80.859375 Top5 97.407227 +2023-10-05 21:17:13,031 - Epoch: [74][ 90/ 117] Loss 0.367901 Top1 81.015625 Top5 97.482639 +2023-10-05 21:17:13,179 - Epoch: [74][ 100/ 117] Loss 0.367588 Top1 81.085938 Top5 97.500000 +2023-10-05 21:17:13,332 - Epoch: [74][ 110/ 117] Loss 0.369973 Top1 81.093750 Top5 97.500000 +2023-10-05 21:17:13,416 - Epoch: [74][ 117/ 117] Loss 0.372995 Top1 81.077380 Top5 97.491901 +2023-10-05 21:17:13,551 - ==> Top1: 81.077 Top5: 97.492 Loss: 0.373 + +2023-10-05 21:17:13,551 - ==> Confusion: +[[ 932 1 8 2 14 0 0 0 4 54 3 2 1 5 4 3 4 4 0 1 8] + [ 1 1026 2 0 10 23 1 33 1 0 2 0 1 0 1 5 7 0 7 3 8] + [ 1 1 963 10 3 1 13 10 0 0 4 3 6 3 1 9 5 2 4 7 10] + [ 0 1 28 944 1 6 1 0 2 0 5 0 18 6 19 5 2 9 25 3 14] + [ 20 7 0 0 964 11 0 1 1 6 1 1 0 0 6 5 20 1 0 5 1] + [ 3 35 0 0 1 958 3 29 0 1 7 16 1 19 3 2 5 0 5 11 17] + [ 0 7 36 0 2 1 1099 7 0 0 5 3 1 0 1 7 0 1 0 12 9] + [ 2 21 16 0 1 47 4 1051 2 1 4 11 2 4 0 3 2 0 29 11 7] + [ 24 6 0 1 0 3 0 2 932 39 17 2 3 28 22 1 2 0 2 3 2] + [ 121 0 3 0 6 8 1 0 30 883 0 2 1 38 2 7 1 3 1 3 9] + [ 3 2 12 12 1 3 3 6 18 1 940 5 0 25 3 1 3 0 5 1 9] + [ 1 0 1 0 0 11 0 1 0 0 0 983 13 3 1 1 0 9 0 9 2] + [ 1 0 5 3 0 3 1 2 0 0 2 63 956 2 3 6 2 4 1 6 8] + [ 2 0 1 0 2 13 1 0 6 8 3 8 3 1045 3 6 2 1 0 7 8] + [ 20 4 3 17 5 0 0 0 28 5 7 2 2 4 961 1 4 5 15 0 18] + [ 0 0 2 0 4 1 0 0 0 0 0 12 9 0 0 1067 13 6 2 12 6] + [ 2 9 2 0 6 8 0 0 2 0 0 5 2 1 2 16 1095 0 1 2 8] + [ 0 0 2 0 0 0 0 0 0 0 0 24 40 3 2 9 0 947 1 2 8] + [ 3 1 9 14 0 2 0 40 5 0 2 2 4 0 12 0 3 1 952 2 16] + [ 0 3 2 0 2 3 6 6 0 0 0 16 5 3 0 10 12 0 3 1076 5] + [ 146 203 186 55 115 216 35 125 67 79 174 214 405 415 118 90 238 49 138 334 4503]] + +2023-10-05 21:17:13,553 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:17:13,553 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:17:13,559 - + +2023-10-05 21:17:13,559 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:17:14,553 - Epoch: [75][ 10/ 1236] Overall Loss 0.338025 Objective Loss 0.338025 LR 0.001000 Time 0.099368 +2023-10-05 21:17:14,756 - Epoch: [75][ 20/ 1236] Overall Loss 0.330751 Objective Loss 0.330751 LR 0.001000 Time 0.059817 +2023-10-05 21:17:14,960 - Epoch: [75][ 30/ 1236] Overall Loss 0.336720 Objective Loss 0.336720 LR 0.001000 Time 0.046655 +2023-10-05 21:17:15,163 - Epoch: [75][ 40/ 1236] Overall Loss 0.334840 Objective Loss 0.334840 LR 0.001000 Time 0.040059 +2023-10-05 21:17:15,367 - Epoch: [75][ 50/ 1236] Overall Loss 0.333702 Objective Loss 0.333702 LR 0.001000 Time 0.036120 +2023-10-05 21:17:15,569 - Epoch: [75][ 60/ 1236] Overall Loss 0.328797 Objective Loss 0.328797 LR 0.001000 Time 0.033474 +2023-10-05 21:17:15,773 - Epoch: [75][ 70/ 1236] Overall Loss 0.332151 Objective Loss 0.332151 LR 0.001000 Time 0.031598 +2023-10-05 21:17:15,976 - Epoch: [75][ 80/ 1236] Overall Loss 0.328971 Objective Loss 0.328971 LR 0.001000 Time 0.030181 +2023-10-05 21:17:16,180 - Epoch: [75][ 90/ 1236] Overall Loss 0.327575 Objective Loss 0.327575 LR 0.001000 Time 0.029091 +2023-10-05 21:17:16,384 - Epoch: [75][ 100/ 1236] Overall Loss 0.328452 Objective Loss 0.328452 LR 0.001000 Time 0.028215 +2023-10-05 21:17:16,588 - Epoch: [75][ 110/ 1236] Overall Loss 0.326578 Objective Loss 0.326578 LR 0.001000 Time 0.027505 +2023-10-05 21:17:16,790 - Epoch: [75][ 120/ 1236] Overall Loss 0.325570 Objective Loss 0.325570 LR 0.001000 Time 0.026888 +2023-10-05 21:17:16,989 - Epoch: [75][ 130/ 1236] Overall Loss 0.328106 Objective Loss 0.328106 LR 0.001000 Time 0.026354 +2023-10-05 21:17:17,191 - Epoch: [75][ 140/ 1236] Overall Loss 0.325280 Objective Loss 0.325280 LR 0.001000 Time 0.025908 +2023-10-05 21:17:17,390 - Epoch: [75][ 150/ 1236] Overall Loss 0.325503 Objective Loss 0.325503 LR 0.001000 Time 0.025509 +2023-10-05 21:17:17,594 - Epoch: [75][ 160/ 1236] Overall Loss 0.324302 Objective Loss 0.324302 LR 0.001000 Time 0.025187 +2023-10-05 21:17:17,797 - Epoch: [75][ 170/ 1236] Overall Loss 0.323430 Objective Loss 0.323430 LR 0.001000 Time 0.024897 +2023-10-05 21:17:18,001 - Epoch: [75][ 180/ 1236] Overall Loss 0.323899 Objective Loss 0.323899 LR 0.001000 Time 0.024642 +2023-10-05 21:17:18,205 - Epoch: [75][ 190/ 1236] Overall Loss 0.323579 Objective Loss 0.323579 LR 0.001000 Time 0.024421 +2023-10-05 21:17:18,409 - Epoch: [75][ 200/ 1236] Overall Loss 0.323873 Objective Loss 0.323873 LR 0.001000 Time 0.024214 +2023-10-05 21:17:18,612 - Epoch: [75][ 210/ 1236] Overall Loss 0.323043 Objective Loss 0.323043 LR 0.001000 Time 0.024027 +2023-10-05 21:17:18,815 - Epoch: [75][ 220/ 1236] Overall Loss 0.320960 Objective Loss 0.320960 LR 0.001000 Time 0.023859 +2023-10-05 21:17:19,018 - Epoch: [75][ 230/ 1236] Overall Loss 0.320468 Objective Loss 0.320468 LR 0.001000 Time 0.023701 +2023-10-05 21:17:19,221 - Epoch: [75][ 240/ 1236] Overall Loss 0.320762 Objective Loss 0.320762 LR 0.001000 Time 0.023560 +2023-10-05 21:17:19,424 - Epoch: [75][ 250/ 1236] Overall Loss 0.320609 Objective Loss 0.320609 LR 0.001000 Time 0.023427 +2023-10-05 21:17:19,628 - Epoch: [75][ 260/ 1236] Overall Loss 0.320760 Objective Loss 0.320760 LR 0.001000 Time 0.023307 +2023-10-05 21:17:19,830 - Epoch: [75][ 270/ 1236] Overall Loss 0.320199 Objective Loss 0.320199 LR 0.001000 Time 0.023194 +2023-10-05 21:17:20,034 - Epoch: [75][ 280/ 1236] Overall Loss 0.319986 Objective Loss 0.319986 LR 0.001000 Time 0.023090 +2023-10-05 21:17:20,237 - Epoch: [75][ 290/ 1236] Overall Loss 0.319798 Objective Loss 0.319798 LR 0.001000 Time 0.022995 +2023-10-05 21:17:20,441 - Epoch: [75][ 300/ 1236] Overall Loss 0.319465 Objective Loss 0.319465 LR 0.001000 Time 0.022906 +2023-10-05 21:17:20,643 - Epoch: [75][ 310/ 1236] Overall Loss 0.319058 Objective Loss 0.319058 LR 0.001000 Time 0.022820 +2023-10-05 21:17:20,847 - Epoch: [75][ 320/ 1236] Overall Loss 0.319130 Objective Loss 0.319130 LR 0.001000 Time 0.022743 +2023-10-05 21:17:21,050 - Epoch: [75][ 330/ 1236] Overall Loss 0.317828 Objective Loss 0.317828 LR 0.001000 Time 0.022668 +2023-10-05 21:17:21,254 - Epoch: [75][ 340/ 1236] Overall Loss 0.318432 Objective Loss 0.318432 LR 0.001000 Time 0.022600 +2023-10-05 21:17:21,458 - Epoch: [75][ 350/ 1236] Overall Loss 0.318727 Objective Loss 0.318727 LR 0.001000 Time 0.022534 +2023-10-05 21:17:21,662 - Epoch: [75][ 360/ 1236] Overall Loss 0.318988 Objective Loss 0.318988 LR 0.001000 Time 0.022474 +2023-10-05 21:17:21,865 - Epoch: [75][ 370/ 1236] Overall Loss 0.319030 Objective Loss 0.319030 LR 0.001000 Time 0.022415 +2023-10-05 21:17:22,068 - Epoch: [75][ 380/ 1236] Overall Loss 0.318288 Objective Loss 0.318288 LR 0.001000 Time 0.022360 +2023-10-05 21:17:22,272 - Epoch: [75][ 390/ 1236] Overall Loss 0.318131 Objective Loss 0.318131 LR 0.001000 Time 0.022308 +2023-10-05 21:17:22,476 - Epoch: [75][ 400/ 1236] Overall Loss 0.317872 Objective Loss 0.317872 LR 0.001000 Time 0.022259 +2023-10-05 21:17:22,680 - Epoch: [75][ 410/ 1236] Overall Loss 0.317676 Objective Loss 0.317676 LR 0.001000 Time 0.022212 +2023-10-05 21:17:22,884 - Epoch: [75][ 420/ 1236] Overall Loss 0.317882 Objective Loss 0.317882 LR 0.001000 Time 0.022168 +2023-10-05 21:17:23,087 - Epoch: [75][ 430/ 1236] Overall Loss 0.317434 Objective Loss 0.317434 LR 0.001000 Time 0.022125 +2023-10-05 21:17:23,291 - Epoch: [75][ 440/ 1236] Overall Loss 0.317965 Objective Loss 0.317965 LR 0.001000 Time 0.022084 +2023-10-05 21:17:23,495 - Epoch: [75][ 450/ 1236] Overall Loss 0.317303 Objective Loss 0.317303 LR 0.001000 Time 0.022046 +2023-10-05 21:17:23,700 - Epoch: [75][ 460/ 1236] Overall Loss 0.316861 Objective Loss 0.316861 LR 0.001000 Time 0.022011 +2023-10-05 21:17:23,905 - Epoch: [75][ 470/ 1236] Overall Loss 0.317104 Objective Loss 0.317104 LR 0.001000 Time 0.021978 +2023-10-05 21:17:24,109 - Epoch: [75][ 480/ 1236] Overall Loss 0.317183 Objective Loss 0.317183 LR 0.001000 Time 0.021945 +2023-10-05 21:17:24,313 - Epoch: [75][ 490/ 1236] Overall Loss 0.317744 Objective Loss 0.317744 LR 0.001000 Time 0.021913 +2023-10-05 21:17:24,517 - Epoch: [75][ 500/ 1236] Overall Loss 0.317972 Objective Loss 0.317972 LR 0.001000 Time 0.021882 +2023-10-05 21:17:24,721 - Epoch: [75][ 510/ 1236] Overall Loss 0.318173 Objective Loss 0.318173 LR 0.001000 Time 0.021853 +2023-10-05 21:17:24,926 - Epoch: [75][ 520/ 1236] Overall Loss 0.318406 Objective Loss 0.318406 LR 0.001000 Time 0.021825 +2023-10-05 21:17:25,131 - Epoch: [75][ 530/ 1236] Overall Loss 0.318644 Objective Loss 0.318644 LR 0.001000 Time 0.021799 +2023-10-05 21:17:25,335 - Epoch: [75][ 540/ 1236] Overall Loss 0.319032 Objective Loss 0.319032 LR 0.001000 Time 0.021773 +2023-10-05 21:17:25,540 - Epoch: [75][ 550/ 1236] Overall Loss 0.319751 Objective Loss 0.319751 LR 0.001000 Time 0.021749 +2023-10-05 21:17:25,744 - Epoch: [75][ 560/ 1236] Overall Loss 0.319872 Objective Loss 0.319872 LR 0.001000 Time 0.021725 +2023-10-05 21:17:25,949 - Epoch: [75][ 570/ 1236] Overall Loss 0.319451 Objective Loss 0.319451 LR 0.001000 Time 0.021702 +2023-10-05 21:17:26,153 - Epoch: [75][ 580/ 1236] Overall Loss 0.318895 Objective Loss 0.318895 LR 0.001000 Time 0.021680 +2023-10-05 21:17:26,358 - Epoch: [75][ 590/ 1236] Overall Loss 0.319304 Objective Loss 0.319304 LR 0.001000 Time 0.021660 +2023-10-05 21:17:26,563 - Epoch: [75][ 600/ 1236] Overall Loss 0.319586 Objective Loss 0.319586 LR 0.001000 Time 0.021639 +2023-10-05 21:17:26,768 - Epoch: [75][ 610/ 1236] Overall Loss 0.319643 Objective Loss 0.319643 LR 0.001000 Time 0.021620 +2023-10-05 21:17:26,972 - Epoch: [75][ 620/ 1236] Overall Loss 0.319406 Objective Loss 0.319406 LR 0.001000 Time 0.021600 +2023-10-05 21:17:27,177 - Epoch: [75][ 630/ 1236] Overall Loss 0.319058 Objective Loss 0.319058 LR 0.001000 Time 0.021582 +2023-10-05 21:17:27,382 - Epoch: [75][ 640/ 1236] Overall Loss 0.319147 Objective Loss 0.319147 LR 0.001000 Time 0.021564 +2023-10-05 21:17:27,587 - Epoch: [75][ 650/ 1236] Overall Loss 0.319589 Objective Loss 0.319589 LR 0.001000 Time 0.021547 +2023-10-05 21:17:27,797 - Epoch: [75][ 660/ 1236] Overall Loss 0.319695 Objective Loss 0.319695 LR 0.001000 Time 0.021538 +2023-10-05 21:17:28,001 - Epoch: [75][ 670/ 1236] Overall Loss 0.320331 Objective Loss 0.320331 LR 0.001000 Time 0.021521 +2023-10-05 21:17:28,207 - Epoch: [75][ 680/ 1236] Overall Loss 0.320481 Objective Loss 0.320481 LR 0.001000 Time 0.021506 +2023-10-05 21:17:28,409 - Epoch: [75][ 690/ 1236] Overall Loss 0.320587 Objective Loss 0.320587 LR 0.001000 Time 0.021488 +2023-10-05 21:17:28,621 - Epoch: [75][ 700/ 1236] Overall Loss 0.320403 Objective Loss 0.320403 LR 0.001000 Time 0.021482 +2023-10-05 21:17:28,824 - Epoch: [75][ 710/ 1236] Overall Loss 0.321099 Objective Loss 0.321099 LR 0.001000 Time 0.021465 +2023-10-05 21:17:29,027 - Epoch: [75][ 720/ 1236] Overall Loss 0.320415 Objective Loss 0.320415 LR 0.001000 Time 0.021448 +2023-10-05 21:17:29,236 - Epoch: [75][ 730/ 1236] Overall Loss 0.320352 Objective Loss 0.320352 LR 0.001000 Time 0.021440 +2023-10-05 21:17:29,439 - Epoch: [75][ 740/ 1236] Overall Loss 0.319774 Objective Loss 0.319774 LR 0.001000 Time 0.021424 +2023-10-05 21:17:29,641 - Epoch: [75][ 750/ 1236] Overall Loss 0.319927 Objective Loss 0.319927 LR 0.001000 Time 0.021407 +2023-10-05 21:17:29,844 - Epoch: [75][ 760/ 1236] Overall Loss 0.320234 Objective Loss 0.320234 LR 0.001000 Time 0.021392 +2023-10-05 21:17:30,047 - Epoch: [75][ 770/ 1236] Overall Loss 0.320174 Objective Loss 0.320174 LR 0.001000 Time 0.021377 +2023-10-05 21:17:30,249 - Epoch: [75][ 780/ 1236] Overall Loss 0.320191 Objective Loss 0.320191 LR 0.001000 Time 0.021362 +2023-10-05 21:17:30,452 - Epoch: [75][ 790/ 1236] Overall Loss 0.319963 Objective Loss 0.319963 LR 0.001000 Time 0.021348 +2023-10-05 21:17:30,654 - Epoch: [75][ 800/ 1236] Overall Loss 0.320010 Objective Loss 0.320010 LR 0.001000 Time 0.021333 +2023-10-05 21:17:30,857 - Epoch: [75][ 810/ 1236] Overall Loss 0.320063 Objective Loss 0.320063 LR 0.001000 Time 0.021320 +2023-10-05 21:17:31,059 - Epoch: [75][ 820/ 1236] Overall Loss 0.320371 Objective Loss 0.320371 LR 0.001000 Time 0.021306 +2023-10-05 21:17:31,262 - Epoch: [75][ 830/ 1236] Overall Loss 0.320816 Objective Loss 0.320816 LR 0.001000 Time 0.021293 +2023-10-05 21:17:31,465 - Epoch: [75][ 840/ 1236] Overall Loss 0.320436 Objective Loss 0.320436 LR 0.001000 Time 0.021281 +2023-10-05 21:17:31,668 - Epoch: [75][ 850/ 1236] Overall Loss 0.320440 Objective Loss 0.320440 LR 0.001000 Time 0.021269 +2023-10-05 21:17:31,871 - Epoch: [75][ 860/ 1236] Overall Loss 0.320738 Objective Loss 0.320738 LR 0.001000 Time 0.021257 +2023-10-05 21:17:32,074 - Epoch: [75][ 870/ 1236] Overall Loss 0.320977 Objective Loss 0.320977 LR 0.001000 Time 0.021245 +2023-10-05 21:17:32,277 - Epoch: [75][ 880/ 1236] Overall Loss 0.321289 Objective Loss 0.321289 LR 0.001000 Time 0.021235 +2023-10-05 21:17:32,480 - Epoch: [75][ 890/ 1236] Overall Loss 0.321477 Objective Loss 0.321477 LR 0.001000 Time 0.021224 +2023-10-05 21:17:32,683 - Epoch: [75][ 900/ 1236] Overall Loss 0.321620 Objective Loss 0.321620 LR 0.001000 Time 0.021213 +2023-10-05 21:17:32,886 - Epoch: [75][ 910/ 1236] Overall Loss 0.321873 Objective Loss 0.321873 LR 0.001000 Time 0.021202 +2023-10-05 21:17:33,088 - Epoch: [75][ 920/ 1236] Overall Loss 0.321557 Objective Loss 0.321557 LR 0.001000 Time 0.021191 +2023-10-05 21:17:33,291 - Epoch: [75][ 930/ 1236] Overall Loss 0.321491 Objective Loss 0.321491 LR 0.001000 Time 0.021181 +2023-10-05 21:17:33,494 - Epoch: [75][ 940/ 1236] Overall Loss 0.321310 Objective Loss 0.321310 LR 0.001000 Time 0.021171 +2023-10-05 21:17:33,697 - Epoch: [75][ 950/ 1236] Overall Loss 0.321633 Objective Loss 0.321633 LR 0.001000 Time 0.021161 +2023-10-05 21:17:33,900 - Epoch: [75][ 960/ 1236] Overall Loss 0.321516 Objective Loss 0.321516 LR 0.001000 Time 0.021152 +2023-10-05 21:17:34,103 - Epoch: [75][ 970/ 1236] Overall Loss 0.321615 Objective Loss 0.321615 LR 0.001000 Time 0.021143 +2023-10-05 21:17:34,305 - Epoch: [75][ 980/ 1236] Overall Loss 0.321665 Objective Loss 0.321665 LR 0.001000 Time 0.021133 +2023-10-05 21:17:34,509 - Epoch: [75][ 990/ 1236] Overall Loss 0.321468 Objective Loss 0.321468 LR 0.001000 Time 0.021125 +2023-10-05 21:17:34,711 - Epoch: [75][ 1000/ 1236] Overall Loss 0.321779 Objective Loss 0.321779 LR 0.001000 Time 0.021115 +2023-10-05 21:17:34,913 - Epoch: [75][ 1010/ 1236] Overall Loss 0.321642 Objective Loss 0.321642 LR 0.001000 Time 0.021107 +2023-10-05 21:17:35,116 - Epoch: [75][ 1020/ 1236] Overall Loss 0.321668 Objective Loss 0.321668 LR 0.001000 Time 0.021098 +2023-10-05 21:17:35,320 - Epoch: [75][ 1030/ 1236] Overall Loss 0.321440 Objective Loss 0.321440 LR 0.001000 Time 0.021090 +2023-10-05 21:17:35,522 - Epoch: [75][ 1040/ 1236] Overall Loss 0.321404 Objective Loss 0.321404 LR 0.001000 Time 0.021082 +2023-10-05 21:17:35,725 - Epoch: [75][ 1050/ 1236] Overall Loss 0.321265 Objective Loss 0.321265 LR 0.001000 Time 0.021074 +2023-10-05 21:17:35,928 - Epoch: [75][ 1060/ 1236] Overall Loss 0.321053 Objective Loss 0.321053 LR 0.001000 Time 0.021066 +2023-10-05 21:17:36,131 - Epoch: [75][ 1070/ 1236] Overall Loss 0.321079 Objective Loss 0.321079 LR 0.001000 Time 0.021059 +2023-10-05 21:17:36,334 - Epoch: [75][ 1080/ 1236] Overall Loss 0.321073 Objective Loss 0.321073 LR 0.001000 Time 0.021051 +2023-10-05 21:17:36,537 - Epoch: [75][ 1090/ 1236] Overall Loss 0.320770 Objective Loss 0.320770 LR 0.001000 Time 0.021044 +2023-10-05 21:17:36,740 - Epoch: [75][ 1100/ 1236] Overall Loss 0.320610 Objective Loss 0.320610 LR 0.001000 Time 0.021037 +2023-10-05 21:17:36,942 - Epoch: [75][ 1110/ 1236] Overall Loss 0.320667 Objective Loss 0.320667 LR 0.001000 Time 0.021029 +2023-10-05 21:17:37,145 - Epoch: [75][ 1120/ 1236] Overall Loss 0.321039 Objective Loss 0.321039 LR 0.001000 Time 0.021022 +2023-10-05 21:17:37,349 - Epoch: [75][ 1130/ 1236] Overall Loss 0.321114 Objective Loss 0.321114 LR 0.001000 Time 0.021016 +2023-10-05 21:17:37,551 - Epoch: [75][ 1140/ 1236] Overall Loss 0.320777 Objective Loss 0.320777 LR 0.001000 Time 0.021009 +2023-10-05 21:17:37,754 - Epoch: [75][ 1150/ 1236] Overall Loss 0.320763 Objective Loss 0.320763 LR 0.001000 Time 0.021003 +2023-10-05 21:17:37,957 - Epoch: [75][ 1160/ 1236] Overall Loss 0.320625 Objective Loss 0.320625 LR 0.001000 Time 0.020996 +2023-10-05 21:17:38,161 - Epoch: [75][ 1170/ 1236] Overall Loss 0.320660 Objective Loss 0.320660 LR 0.001000 Time 0.020990 +2023-10-05 21:17:38,363 - Epoch: [75][ 1180/ 1236] Overall Loss 0.321067 Objective Loss 0.321067 LR 0.001000 Time 0.020984 +2023-10-05 21:17:38,566 - Epoch: [75][ 1190/ 1236] Overall Loss 0.321111 Objective Loss 0.321111 LR 0.001000 Time 0.020977 +2023-10-05 21:17:38,769 - Epoch: [75][ 1200/ 1236] Overall Loss 0.321368 Objective Loss 0.321368 LR 0.001000 Time 0.020971 +2023-10-05 21:17:38,972 - Epoch: [75][ 1210/ 1236] Overall Loss 0.321481 Objective Loss 0.321481 LR 0.001000 Time 0.020965 +2023-10-05 21:17:39,175 - Epoch: [75][ 1220/ 1236] Overall Loss 0.321403 Objective Loss 0.321403 LR 0.001000 Time 0.020960 +2023-10-05 21:17:39,430 - Epoch: [75][ 1230/ 1236] Overall Loss 0.321379 Objective Loss 0.321379 LR 0.001000 Time 0.020996 +2023-10-05 21:17:39,548 - Epoch: [75][ 1236/ 1236] Overall Loss 0.321464 Objective Loss 0.321464 Top1 78.004073 Top5 96.537678 LR 0.001000 Time 0.020990 +2023-10-05 21:17:39,678 - --- validate (epoch=75)----------- +2023-10-05 21:17:39,678 - 29943 samples (256 per mini-batch) +2023-10-05 21:17:40,129 - Epoch: [75][ 10/ 117] Loss 0.405273 Top1 79.765625 Top5 96.796875 +2023-10-05 21:17:40,276 - Epoch: [75][ 20/ 117] Loss 0.378922 Top1 80.820312 Top5 96.875000 +2023-10-05 21:17:40,424 - Epoch: [75][ 30/ 117] Loss 0.375777 Top1 81.002604 Top5 96.927083 +2023-10-05 21:17:40,569 - Epoch: [75][ 40/ 117] Loss 0.373223 Top1 81.210938 Top5 97.011719 +2023-10-05 21:17:40,717 - Epoch: [75][ 50/ 117] Loss 0.372440 Top1 81.117188 Top5 97.070312 +2023-10-05 21:17:40,863 - Epoch: [75][ 60/ 117] Loss 0.369734 Top1 81.204427 Top5 97.115885 +2023-10-05 21:17:41,010 - Epoch: [75][ 70/ 117] Loss 0.369522 Top1 81.171875 Top5 97.142857 +2023-10-05 21:17:41,158 - Epoch: [75][ 80/ 117] Loss 0.371287 Top1 81.098633 Top5 97.143555 +2023-10-05 21:17:41,306 - Epoch: [75][ 90/ 117] Loss 0.375094 Top1 80.920139 Top5 97.113715 +2023-10-05 21:17:41,453 - Epoch: [75][ 100/ 117] Loss 0.375380 Top1 80.812500 Top5 97.089844 +2023-10-05 21:17:41,606 - Epoch: [75][ 110/ 117] Loss 0.375992 Top1 80.752841 Top5 97.095170 +2023-10-05 21:17:41,691 - Epoch: [75][ 117/ 117] Loss 0.373888 Top1 80.813546 Top5 97.131216 +2023-10-05 21:17:41,814 - ==> Top1: 80.814 Top5: 97.131 Loss: 0.374 + +2023-10-05 21:17:41,814 - ==> Confusion: +[[ 930 2 4 2 8 6 0 1 4 63 1 2 0 3 4 2 5 1 0 0 12] + [ 1 1026 4 0 12 42 2 16 1 0 1 1 0 0 0 4 8 1 8 1 3] + [ 5 1 943 12 3 2 35 10 0 0 6 2 6 1 0 5 6 1 7 4 7] + [ 1 1 21 951 2 7 2 1 1 1 9 0 14 2 24 10 2 3 20 2 15] + [ 16 3 2 0 972 11 0 1 0 8 0 7 0 1 5 9 11 1 0 0 3] + [ 4 30 0 2 3 994 4 19 1 2 3 6 3 10 3 2 5 1 2 9 13] + [ 0 4 27 0 0 3 1118 7 0 0 2 4 1 1 0 9 0 1 0 10 4] + [ 3 21 14 0 4 61 9 1019 0 3 5 10 5 2 0 4 0 0 39 9 10] + [ 27 6 0 2 0 5 0 2 924 51 15 3 0 19 17 4 2 1 6 0 5] + [ 125 2 3 0 5 16 2 1 18 892 0 2 0 32 1 6 0 1 0 1 12] + [ 3 4 16 5 1 4 8 1 14 0 951 2 1 18 3 4 4 0 4 0 10] + [ 1 0 0 0 0 16 0 6 0 0 0 955 17 3 0 1 4 13 0 16 3] + [ 0 0 9 5 1 6 3 5 1 0 2 48 945 2 2 9 2 9 1 7 11] + [ 1 0 3 1 3 20 1 0 7 13 5 8 2 1030 2 4 1 0 0 6 12] + [ 18 3 4 18 11 1 1 0 27 5 8 1 2 3 961 1 3 2 21 0 11] + [ 0 3 1 2 6 2 0 0 0 0 0 10 10 1 0 1070 6 8 1 11 3] + [ 0 14 3 2 7 6 0 0 1 0 1 4 1 2 2 16 1091 0 1 3 7] + [ 0 2 2 0 0 0 1 0 1 0 0 13 29 2 3 14 2 958 3 3 5] + [ 0 15 16 12 2 4 0 37 2 0 3 0 6 0 11 2 1 0 947 1 9] + [ 0 4 6 0 1 7 12 10 0 0 0 10 7 4 0 5 10 1 0 1072 3] + [ 129 196 222 93 105 334 73 121 71 106 173 162 389 320 128 95 194 57 183 305 4449]] + +2023-10-05 21:17:41,816 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:17:41,816 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:17:41,822 - + +2023-10-05 21:17:41,822 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:17:42,827 - Epoch: [76][ 10/ 1236] Overall Loss 0.326654 Objective Loss 0.326654 LR 0.001000 Time 0.100491 +2023-10-05 21:17:43,031 - Epoch: [76][ 20/ 1236] Overall Loss 0.338901 Objective Loss 0.338901 LR 0.001000 Time 0.060396 +2023-10-05 21:17:43,232 - Epoch: [76][ 30/ 1236] Overall Loss 0.338607 Objective Loss 0.338607 LR 0.001000 Time 0.046973 +2023-10-05 21:17:43,435 - Epoch: [76][ 40/ 1236] Overall Loss 0.335776 Objective Loss 0.335776 LR 0.001000 Time 0.040299 +2023-10-05 21:17:43,637 - Epoch: [76][ 50/ 1236] Overall Loss 0.330203 Objective Loss 0.330203 LR 0.001000 Time 0.036273 +2023-10-05 21:17:43,840 - Epoch: [76][ 60/ 1236] Overall Loss 0.327089 Objective Loss 0.327089 LR 0.001000 Time 0.033607 +2023-10-05 21:17:44,042 - Epoch: [76][ 70/ 1236] Overall Loss 0.322606 Objective Loss 0.322606 LR 0.001000 Time 0.031683 +2023-10-05 21:17:44,245 - Epoch: [76][ 80/ 1236] Overall Loss 0.315963 Objective Loss 0.315963 LR 0.001000 Time 0.030260 +2023-10-05 21:17:44,446 - Epoch: [76][ 90/ 1236] Overall Loss 0.314671 Objective Loss 0.314671 LR 0.001000 Time 0.029127 +2023-10-05 21:17:44,650 - Epoch: [76][ 100/ 1236] Overall Loss 0.315628 Objective Loss 0.315628 LR 0.001000 Time 0.028244 +2023-10-05 21:17:44,855 - Epoch: [76][ 110/ 1236] Overall Loss 0.314082 Objective Loss 0.314082 LR 0.001000 Time 0.027535 +2023-10-05 21:17:45,059 - Epoch: [76][ 120/ 1236] Overall Loss 0.312094 Objective Loss 0.312094 LR 0.001000 Time 0.026945 +2023-10-05 21:17:45,264 - Epoch: [76][ 130/ 1236] Overall Loss 0.310668 Objective Loss 0.310668 LR 0.001000 Time 0.026443 +2023-10-05 21:17:45,469 - Epoch: [76][ 140/ 1236] Overall Loss 0.310367 Objective Loss 0.310367 LR 0.001000 Time 0.026020 +2023-10-05 21:17:45,674 - Epoch: [76][ 150/ 1236] Overall Loss 0.310204 Objective Loss 0.310204 LR 0.001000 Time 0.025646 +2023-10-05 21:17:45,877 - Epoch: [76][ 160/ 1236] Overall Loss 0.311344 Objective Loss 0.311344 LR 0.001000 Time 0.025310 +2023-10-05 21:17:46,079 - Epoch: [76][ 170/ 1236] Overall Loss 0.312264 Objective Loss 0.312264 LR 0.001000 Time 0.025009 +2023-10-05 21:17:46,283 - Epoch: [76][ 180/ 1236] Overall Loss 0.312267 Objective Loss 0.312267 LR 0.001000 Time 0.024750 +2023-10-05 21:17:46,486 - Epoch: [76][ 190/ 1236] Overall Loss 0.312369 Objective Loss 0.312369 LR 0.001000 Time 0.024516 +2023-10-05 21:17:46,690 - Epoch: [76][ 200/ 1236] Overall Loss 0.313067 Objective Loss 0.313067 LR 0.001000 Time 0.024305 +2023-10-05 21:17:46,893 - Epoch: [76][ 210/ 1236] Overall Loss 0.313295 Objective Loss 0.313295 LR 0.001000 Time 0.024114 +2023-10-05 21:17:47,096 - Epoch: [76][ 220/ 1236] Overall Loss 0.311931 Objective Loss 0.311931 LR 0.001000 Time 0.023942 +2023-10-05 21:17:47,300 - Epoch: [76][ 230/ 1236] Overall Loss 0.312214 Objective Loss 0.312214 LR 0.001000 Time 0.023784 +2023-10-05 21:17:47,503 - Epoch: [76][ 240/ 1236] Overall Loss 0.313414 Objective Loss 0.313414 LR 0.001000 Time 0.023639 +2023-10-05 21:17:47,707 - Epoch: [76][ 250/ 1236] Overall Loss 0.313422 Objective Loss 0.313422 LR 0.001000 Time 0.023507 +2023-10-05 21:17:47,910 - Epoch: [76][ 260/ 1236] Overall Loss 0.313818 Objective Loss 0.313818 LR 0.001000 Time 0.023384 +2023-10-05 21:17:48,113 - Epoch: [76][ 270/ 1236] Overall Loss 0.315124 Objective Loss 0.315124 LR 0.001000 Time 0.023266 +2023-10-05 21:17:48,317 - Epoch: [76][ 280/ 1236] Overall Loss 0.317267 Objective Loss 0.317267 LR 0.001000 Time 0.023162 +2023-10-05 21:17:48,520 - Epoch: [76][ 290/ 1236] Overall Loss 0.316205 Objective Loss 0.316205 LR 0.001000 Time 0.023063 +2023-10-05 21:17:48,723 - Epoch: [76][ 300/ 1236] Overall Loss 0.316608 Objective Loss 0.316608 LR 0.001000 Time 0.022971 +2023-10-05 21:17:48,927 - Epoch: [76][ 310/ 1236] Overall Loss 0.316227 Objective Loss 0.316227 LR 0.001000 Time 0.022885 +2023-10-05 21:17:49,130 - Epoch: [76][ 320/ 1236] Overall Loss 0.317166 Objective Loss 0.317166 LR 0.001000 Time 0.022806 +2023-10-05 21:17:49,333 - Epoch: [76][ 330/ 1236] Overall Loss 0.317756 Objective Loss 0.317756 LR 0.001000 Time 0.022729 +2023-10-05 21:17:49,536 - Epoch: [76][ 340/ 1236] Overall Loss 0.317489 Objective Loss 0.317489 LR 0.001000 Time 0.022657 +2023-10-05 21:17:49,740 - Epoch: [76][ 350/ 1236] Overall Loss 0.317880 Objective Loss 0.317880 LR 0.001000 Time 0.022590 +2023-10-05 21:17:49,944 - Epoch: [76][ 360/ 1236] Overall Loss 0.317778 Objective Loss 0.317778 LR 0.001000 Time 0.022527 +2023-10-05 21:17:50,146 - Epoch: [76][ 370/ 1236] Overall Loss 0.318559 Objective Loss 0.318559 LR 0.001000 Time 0.022466 +2023-10-05 21:17:50,352 - Epoch: [76][ 380/ 1236] Overall Loss 0.317505 Objective Loss 0.317505 LR 0.001000 Time 0.022414 +2023-10-05 21:17:50,555 - Epoch: [76][ 390/ 1236] Overall Loss 0.317104 Objective Loss 0.317104 LR 0.001000 Time 0.022360 +2023-10-05 21:17:50,761 - Epoch: [76][ 400/ 1236] Overall Loss 0.317538 Objective Loss 0.317538 LR 0.001000 Time 0.022313 +2023-10-05 21:17:50,964 - Epoch: [76][ 410/ 1236] Overall Loss 0.317663 Objective Loss 0.317663 LR 0.001000 Time 0.022265 +2023-10-05 21:17:51,170 - Epoch: [76][ 420/ 1236] Overall Loss 0.317113 Objective Loss 0.317113 LR 0.001000 Time 0.022223 +2023-10-05 21:17:51,372 - Epoch: [76][ 430/ 1236] Overall Loss 0.317402 Objective Loss 0.317402 LR 0.001000 Time 0.022177 +2023-10-05 21:17:51,578 - Epoch: [76][ 440/ 1236] Overall Loss 0.317793 Objective Loss 0.317793 LR 0.001000 Time 0.022138 +2023-10-05 21:17:51,781 - Epoch: [76][ 450/ 1236] Overall Loss 0.317473 Objective Loss 0.317473 LR 0.001000 Time 0.022098 +2023-10-05 21:17:51,986 - Epoch: [76][ 460/ 1236] Overall Loss 0.317348 Objective Loss 0.317348 LR 0.001000 Time 0.022062 +2023-10-05 21:17:52,189 - Epoch: [76][ 470/ 1236] Overall Loss 0.317567 Objective Loss 0.317567 LR 0.001000 Time 0.022024 +2023-10-05 21:17:52,395 - Epoch: [76][ 480/ 1236] Overall Loss 0.317014 Objective Loss 0.317014 LR 0.001000 Time 0.021993 +2023-10-05 21:17:52,599 - Epoch: [76][ 490/ 1236] Overall Loss 0.316776 Objective Loss 0.316776 LR 0.001000 Time 0.021959 +2023-10-05 21:17:52,804 - Epoch: [76][ 500/ 1236] Overall Loss 0.316990 Objective Loss 0.316990 LR 0.001000 Time 0.021930 +2023-10-05 21:17:53,007 - Epoch: [76][ 510/ 1236] Overall Loss 0.316596 Objective Loss 0.316596 LR 0.001000 Time 0.021898 +2023-10-05 21:17:53,213 - Epoch: [76][ 520/ 1236] Overall Loss 0.316104 Objective Loss 0.316104 LR 0.001000 Time 0.021871 +2023-10-05 21:17:53,416 - Epoch: [76][ 530/ 1236] Overall Loss 0.316164 Objective Loss 0.316164 LR 0.001000 Time 0.021840 +2023-10-05 21:17:53,621 - Epoch: [76][ 540/ 1236] Overall Loss 0.316353 Objective Loss 0.316353 LR 0.001000 Time 0.021814 +2023-10-05 21:17:53,824 - Epoch: [76][ 550/ 1236] Overall Loss 0.316154 Objective Loss 0.316154 LR 0.001000 Time 0.021785 +2023-10-05 21:17:54,030 - Epoch: [76][ 560/ 1236] Overall Loss 0.316436 Objective Loss 0.316436 LR 0.001000 Time 0.021763 +2023-10-05 21:17:54,233 - Epoch: [76][ 570/ 1236] Overall Loss 0.316482 Objective Loss 0.316482 LR 0.001000 Time 0.021735 +2023-10-05 21:17:54,438 - Epoch: [76][ 580/ 1236] Overall Loss 0.316685 Objective Loss 0.316685 LR 0.001000 Time 0.021713 +2023-10-05 21:17:54,641 - Epoch: [76][ 590/ 1236] Overall Loss 0.316881 Objective Loss 0.316881 LR 0.001000 Time 0.021687 +2023-10-05 21:17:54,847 - Epoch: [76][ 600/ 1236] Overall Loss 0.317110 Objective Loss 0.317110 LR 0.001000 Time 0.021667 +2023-10-05 21:17:55,050 - Epoch: [76][ 610/ 1236] Overall Loss 0.317251 Objective Loss 0.317251 LR 0.001000 Time 0.021642 +2023-10-05 21:17:55,255 - Epoch: [76][ 620/ 1236] Overall Loss 0.317603 Objective Loss 0.317603 LR 0.001000 Time 0.021624 +2023-10-05 21:17:55,458 - Epoch: [76][ 630/ 1236] Overall Loss 0.318053 Objective Loss 0.318053 LR 0.001000 Time 0.021601 +2023-10-05 21:17:55,664 - Epoch: [76][ 640/ 1236] Overall Loss 0.317485 Objective Loss 0.317485 LR 0.001000 Time 0.021583 +2023-10-05 21:17:55,867 - Epoch: [76][ 650/ 1236] Overall Loss 0.317558 Objective Loss 0.317558 LR 0.001000 Time 0.021561 +2023-10-05 21:17:56,072 - Epoch: [76][ 660/ 1236] Overall Loss 0.317643 Objective Loss 0.317643 LR 0.001000 Time 0.021545 +2023-10-05 21:17:56,275 - Epoch: [76][ 670/ 1236] Overall Loss 0.317611 Objective Loss 0.317611 LR 0.001000 Time 0.021524 +2023-10-05 21:17:56,481 - Epoch: [76][ 680/ 1236] Overall Loss 0.317918 Objective Loss 0.317918 LR 0.001000 Time 0.021510 +2023-10-05 21:17:56,684 - Epoch: [76][ 690/ 1236] Overall Loss 0.317986 Objective Loss 0.317986 LR 0.001000 Time 0.021491 +2023-10-05 21:17:56,889 - Epoch: [76][ 700/ 1236] Overall Loss 0.317911 Objective Loss 0.317911 LR 0.001000 Time 0.021476 +2023-10-05 21:17:57,092 - Epoch: [76][ 710/ 1236] Overall Loss 0.317655 Objective Loss 0.317655 LR 0.001000 Time 0.021459 +2023-10-05 21:17:57,298 - Epoch: [76][ 720/ 1236] Overall Loss 0.317528 Objective Loss 0.317528 LR 0.001000 Time 0.021446 +2023-10-05 21:17:57,502 - Epoch: [76][ 730/ 1236] Overall Loss 0.317119 Objective Loss 0.317119 LR 0.001000 Time 0.021431 +2023-10-05 21:17:57,707 - Epoch: [76][ 740/ 1236] Overall Loss 0.316762 Objective Loss 0.316762 LR 0.001000 Time 0.021418 +2023-10-05 21:17:57,910 - Epoch: [76][ 750/ 1236] Overall Loss 0.316937 Objective Loss 0.316937 LR 0.001000 Time 0.021401 +2023-10-05 21:17:58,116 - Epoch: [76][ 760/ 1236] Overall Loss 0.316729 Objective Loss 0.316729 LR 0.001000 Time 0.021389 +2023-10-05 21:17:58,318 - Epoch: [76][ 770/ 1236] Overall Loss 0.316471 Objective Loss 0.316471 LR 0.001000 Time 0.021373 +2023-10-05 21:17:58,524 - Epoch: [76][ 780/ 1236] Overall Loss 0.315895 Objective Loss 0.315895 LR 0.001000 Time 0.021361 +2023-10-05 21:17:58,727 - Epoch: [76][ 790/ 1236] Overall Loss 0.315781 Objective Loss 0.315781 LR 0.001000 Time 0.021348 +2023-10-05 21:17:58,932 - Epoch: [76][ 800/ 1236] Overall Loss 0.315830 Objective Loss 0.315830 LR 0.001000 Time 0.021337 +2023-10-05 21:17:59,135 - Epoch: [76][ 810/ 1236] Overall Loss 0.315482 Objective Loss 0.315482 LR 0.001000 Time 0.021322 +2023-10-05 21:17:59,341 - Epoch: [76][ 820/ 1236] Overall Loss 0.315991 Objective Loss 0.315991 LR 0.001000 Time 0.021313 +2023-10-05 21:17:59,544 - Epoch: [76][ 830/ 1236] Overall Loss 0.315853 Objective Loss 0.315853 LR 0.001000 Time 0.021300 +2023-10-05 21:17:59,749 - Epoch: [76][ 840/ 1236] Overall Loss 0.316060 Objective Loss 0.316060 LR 0.001000 Time 0.021291 +2023-10-05 21:17:59,953 - Epoch: [76][ 850/ 1236] Overall Loss 0.315859 Objective Loss 0.315859 LR 0.001000 Time 0.021278 +2023-10-05 21:18:00,160 - Epoch: [76][ 860/ 1236] Overall Loss 0.316029 Objective Loss 0.316029 LR 0.001000 Time 0.021271 +2023-10-05 21:18:00,365 - Epoch: [76][ 870/ 1236] Overall Loss 0.315974 Objective Loss 0.315974 LR 0.001000 Time 0.021261 +2023-10-05 21:18:00,573 - Epoch: [76][ 880/ 1236] Overall Loss 0.315639 Objective Loss 0.315639 LR 0.001000 Time 0.021256 +2023-10-05 21:18:00,778 - Epoch: [76][ 890/ 1236] Overall Loss 0.315562 Objective Loss 0.315562 LR 0.001000 Time 0.021247 +2023-10-05 21:18:00,986 - Epoch: [76][ 900/ 1236] Overall Loss 0.315194 Objective Loss 0.315194 LR 0.001000 Time 0.021241 +2023-10-05 21:18:01,190 - Epoch: [76][ 910/ 1236] Overall Loss 0.314908 Objective Loss 0.314908 LR 0.001000 Time 0.021233 +2023-10-05 21:18:01,398 - Epoch: [76][ 920/ 1236] Overall Loss 0.314953 Objective Loss 0.314953 LR 0.001000 Time 0.021227 +2023-10-05 21:18:01,603 - Epoch: [76][ 930/ 1236] Overall Loss 0.314979 Objective Loss 0.314979 LR 0.001000 Time 0.021218 +2023-10-05 21:18:01,811 - Epoch: [76][ 940/ 1236] Overall Loss 0.314916 Objective Loss 0.314916 LR 0.001000 Time 0.021214 +2023-10-05 21:18:02,016 - Epoch: [76][ 950/ 1236] Overall Loss 0.314976 Objective Loss 0.314976 LR 0.001000 Time 0.021205 +2023-10-05 21:18:02,223 - Epoch: [76][ 960/ 1236] Overall Loss 0.314898 Objective Loss 0.314898 LR 0.001000 Time 0.021200 +2023-10-05 21:18:02,428 - Epoch: [76][ 970/ 1236] Overall Loss 0.314575 Objective Loss 0.314575 LR 0.001000 Time 0.021193 +2023-10-05 21:18:02,636 - Epoch: [76][ 980/ 1236] Overall Loss 0.314588 Objective Loss 0.314588 LR 0.001000 Time 0.021188 +2023-10-05 21:18:02,841 - Epoch: [76][ 990/ 1236] Overall Loss 0.314244 Objective Loss 0.314244 LR 0.001000 Time 0.021181 +2023-10-05 21:18:03,049 - Epoch: [76][ 1000/ 1236] Overall Loss 0.314356 Objective Loss 0.314356 LR 0.001000 Time 0.021176 +2023-10-05 21:18:03,254 - Epoch: [76][ 1010/ 1236] Overall Loss 0.314600 Objective Loss 0.314600 LR 0.001000 Time 0.021170 +2023-10-05 21:18:03,462 - Epoch: [76][ 1020/ 1236] Overall Loss 0.314884 Objective Loss 0.314884 LR 0.001000 Time 0.021166 +2023-10-05 21:18:03,667 - Epoch: [76][ 1030/ 1236] Overall Loss 0.315107 Objective Loss 0.315107 LR 0.001000 Time 0.021159 +2023-10-05 21:18:03,875 - Epoch: [76][ 1040/ 1236] Overall Loss 0.315103 Objective Loss 0.315103 LR 0.001000 Time 0.021155 +2023-10-05 21:18:04,080 - Epoch: [76][ 1050/ 1236] Overall Loss 0.314913 Objective Loss 0.314913 LR 0.001000 Time 0.021148 +2023-10-05 21:18:04,288 - Epoch: [76][ 1060/ 1236] Overall Loss 0.315311 Objective Loss 0.315311 LR 0.001000 Time 0.021145 +2023-10-05 21:18:04,492 - Epoch: [76][ 1070/ 1236] Overall Loss 0.315309 Objective Loss 0.315309 LR 0.001000 Time 0.021138 +2023-10-05 21:18:04,700 - Epoch: [76][ 1080/ 1236] Overall Loss 0.315628 Objective Loss 0.315628 LR 0.001000 Time 0.021134 +2023-10-05 21:18:04,905 - Epoch: [76][ 1090/ 1236] Overall Loss 0.315934 Objective Loss 0.315934 LR 0.001000 Time 0.021128 +2023-10-05 21:18:05,113 - Epoch: [76][ 1100/ 1236] Overall Loss 0.315809 Objective Loss 0.315809 LR 0.001000 Time 0.021124 +2023-10-05 21:18:05,317 - Epoch: [76][ 1110/ 1236] Overall Loss 0.315815 Objective Loss 0.315815 LR 0.001000 Time 0.021118 +2023-10-05 21:18:05,525 - Epoch: [76][ 1120/ 1236] Overall Loss 0.315997 Objective Loss 0.315997 LR 0.001000 Time 0.021115 +2023-10-05 21:18:05,730 - Epoch: [76][ 1130/ 1236] Overall Loss 0.315762 Objective Loss 0.315762 LR 0.001000 Time 0.021109 +2023-10-05 21:18:05,938 - Epoch: [76][ 1140/ 1236] Overall Loss 0.316011 Objective Loss 0.316011 LR 0.001000 Time 0.021106 +2023-10-05 21:18:06,143 - Epoch: [76][ 1150/ 1236] Overall Loss 0.316154 Objective Loss 0.316154 LR 0.001000 Time 0.021100 +2023-10-05 21:18:06,350 - Epoch: [76][ 1160/ 1236] Overall Loss 0.316280 Objective Loss 0.316280 LR 0.001000 Time 0.021097 +2023-10-05 21:18:06,555 - Epoch: [76][ 1170/ 1236] Overall Loss 0.316191 Objective Loss 0.316191 LR 0.001000 Time 0.021091 +2023-10-05 21:18:06,763 - Epoch: [76][ 1180/ 1236] Overall Loss 0.316268 Objective Loss 0.316268 LR 0.001000 Time 0.021088 +2023-10-05 21:18:06,968 - Epoch: [76][ 1190/ 1236] Overall Loss 0.316536 Objective Loss 0.316536 LR 0.001000 Time 0.021083 +2023-10-05 21:18:07,175 - Epoch: [76][ 1200/ 1236] Overall Loss 0.316675 Objective Loss 0.316675 LR 0.001000 Time 0.021080 +2023-10-05 21:18:07,380 - Epoch: [76][ 1210/ 1236] Overall Loss 0.317060 Objective Loss 0.317060 LR 0.001000 Time 0.021074 +2023-10-05 21:18:07,588 - Epoch: [76][ 1220/ 1236] Overall Loss 0.317061 Objective Loss 0.317061 LR 0.001000 Time 0.021072 +2023-10-05 21:18:07,847 - Epoch: [76][ 1230/ 1236] Overall Loss 0.317130 Objective Loss 0.317130 LR 0.001000 Time 0.021111 +2023-10-05 21:18:07,965 - Epoch: [76][ 1236/ 1236] Overall Loss 0.317384 Objective Loss 0.317384 Top1 82.892057 Top5 98.167006 LR 0.001000 Time 0.021103 +2023-10-05 21:18:08,101 - --- validate (epoch=76)----------- +2023-10-05 21:18:08,102 - 29943 samples (256 per mini-batch) +2023-10-05 21:18:08,571 - Epoch: [76][ 10/ 117] Loss 0.395569 Top1 81.640625 Top5 97.421875 +2023-10-05 21:18:08,723 - Epoch: [76][ 20/ 117] Loss 0.393448 Top1 80.898438 Top5 97.578125 +2023-10-05 21:18:08,873 - Epoch: [76][ 30/ 117] Loss 0.376067 Top1 81.432292 Top5 97.617188 +2023-10-05 21:18:09,023 - Epoch: [76][ 40/ 117] Loss 0.379465 Top1 81.513672 Top5 97.529297 +2023-10-05 21:18:09,174 - Epoch: [76][ 50/ 117] Loss 0.379333 Top1 81.398438 Top5 97.578125 +2023-10-05 21:18:09,326 - Epoch: [76][ 60/ 117] Loss 0.377973 Top1 81.477865 Top5 97.558594 +2023-10-05 21:18:09,482 - Epoch: [76][ 70/ 117] Loss 0.379191 Top1 81.417411 Top5 97.500000 +2023-10-05 21:18:09,633 - Epoch: [76][ 80/ 117] Loss 0.378519 Top1 81.381836 Top5 97.465820 +2023-10-05 21:18:09,779 - Epoch: [76][ 90/ 117] Loss 0.374772 Top1 81.467014 Top5 97.539062 +2023-10-05 21:18:09,927 - Epoch: [76][ 100/ 117] Loss 0.377875 Top1 81.535156 Top5 97.492188 +2023-10-05 21:18:10,079 - Epoch: [76][ 110/ 117] Loss 0.377544 Top1 81.530540 Top5 97.478693 +2023-10-05 21:18:10,164 - Epoch: [76][ 117/ 117] Loss 0.375618 Top1 81.558294 Top5 97.458504 +2023-10-05 21:18:10,292 - ==> Top1: 81.558 Top5: 97.459 Loss: 0.376 + +2023-10-05 21:18:10,293 - ==> Confusion: +[[ 867 4 5 1 21 3 0 0 8 107 0 1 1 4 7 3 4 1 0 0 13] + [ 0 1031 1 0 12 28 2 22 2 0 2 4 1 1 3 3 7 0 3 1 8] + [ 2 1 948 6 3 1 27 16 0 0 8 0 8 4 3 6 3 2 4 6 8] + [ 3 1 21 922 2 8 1 1 8 0 12 0 10 4 38 5 2 3 28 1 19] + [ 15 5 0 0 982 11 0 1 1 8 2 1 0 1 3 4 10 1 0 1 4] + [ 3 43 1 0 4 961 0 33 5 0 4 14 0 18 6 2 2 0 3 4 13] + [ 1 7 26 0 0 2 1107 13 0 0 7 4 1 0 1 3 0 1 2 7 9] + [ 3 19 9 0 4 32 2 1068 3 1 6 13 4 2 2 2 0 0 32 8 8] + [ 15 2 2 0 4 4 0 0 962 52 8 2 3 12 15 2 2 0 1 1 2] + [ 69 0 3 1 13 5 1 0 21 944 0 0 0 39 8 4 1 0 0 1 9] + [ 4 3 11 4 0 5 3 6 18 0 949 2 0 18 6 2 2 0 7 0 13] + [ 0 0 0 0 0 13 0 1 1 1 0 960 21 11 0 2 3 12 0 3 7] + [ 3 1 6 3 1 3 0 4 2 0 1 63 944 3 1 4 1 12 1 3 12] + [ 3 1 0 0 3 15 0 0 16 10 6 9 2 1036 2 2 1 1 0 4 8] + [ 10 5 3 8 9 1 0 0 34 9 2 1 2 2 988 1 2 1 11 0 12] + [ 0 4 3 2 4 1 3 1 0 0 0 12 8 1 0 1059 15 9 0 5 7] + [ 0 16 2 0 16 10 0 0 1 0 1 7 1 3 4 17 1072 0 0 4 7] + [ 1 0 2 1 0 1 1 0 0 0 0 12 37 3 2 10 0 957 2 5 4] + [ 0 5 10 10 2 0 1 36 7 0 7 0 3 3 15 1 2 0 955 1 10] + [ 0 3 1 0 3 10 5 26 1 0 1 28 6 10 1 7 10 0 2 1024 14] + [ 115 261 160 46 124 205 57 153 140 129 188 139 363 359 157 64 175 58 156 171 4685]] + +2023-10-05 21:18:10,294 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:18:10,294 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:18:10,300 - + +2023-10-05 21:18:10,300 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:18:11,297 - Epoch: [77][ 10/ 1236] Overall Loss 0.333546 Objective Loss 0.333546 LR 0.001000 Time 0.099587 +2023-10-05 21:18:11,504 - Epoch: [77][ 20/ 1236] Overall Loss 0.344543 Objective Loss 0.344543 LR 0.001000 Time 0.060152 +2023-10-05 21:18:11,705 - Epoch: [77][ 30/ 1236] Overall Loss 0.348384 Objective Loss 0.348384 LR 0.001000 Time 0.046787 +2023-10-05 21:18:11,909 - Epoch: [77][ 40/ 1236] Overall Loss 0.339791 Objective Loss 0.339791 LR 0.001000 Time 0.040179 +2023-10-05 21:18:12,110 - Epoch: [77][ 50/ 1236] Overall Loss 0.328865 Objective Loss 0.328865 LR 0.001000 Time 0.036159 +2023-10-05 21:18:12,314 - Epoch: [77][ 60/ 1236] Overall Loss 0.327188 Objective Loss 0.327188 LR 0.001000 Time 0.033528 +2023-10-05 21:18:12,514 - Epoch: [77][ 70/ 1236] Overall Loss 0.321515 Objective Loss 0.321515 LR 0.001000 Time 0.031587 +2023-10-05 21:18:12,715 - Epoch: [77][ 80/ 1236] Overall Loss 0.321058 Objective Loss 0.321058 LR 0.001000 Time 0.030148 +2023-10-05 21:18:12,915 - Epoch: [77][ 90/ 1236] Overall Loss 0.319245 Objective Loss 0.319245 LR 0.001000 Time 0.029018 +2023-10-05 21:18:13,118 - Epoch: [77][ 100/ 1236] Overall Loss 0.320381 Objective Loss 0.320381 LR 0.001000 Time 0.028133 +2023-10-05 21:18:13,318 - Epoch: [77][ 110/ 1236] Overall Loss 0.320690 Objective Loss 0.320690 LR 0.001000 Time 0.027397 +2023-10-05 21:18:13,521 - Epoch: [77][ 120/ 1236] Overall Loss 0.322107 Objective Loss 0.322107 LR 0.001000 Time 0.026796 +2023-10-05 21:18:13,723 - Epoch: [77][ 130/ 1236] Overall Loss 0.319826 Objective Loss 0.319826 LR 0.001000 Time 0.026292 +2023-10-05 21:18:13,926 - Epoch: [77][ 140/ 1236] Overall Loss 0.318440 Objective Loss 0.318440 LR 0.001000 Time 0.025858 +2023-10-05 21:18:14,130 - Epoch: [77][ 150/ 1236] Overall Loss 0.317824 Objective Loss 0.317824 LR 0.001000 Time 0.025493 +2023-10-05 21:18:14,335 - Epoch: [77][ 160/ 1236] Overall Loss 0.315236 Objective Loss 0.315236 LR 0.001000 Time 0.025178 +2023-10-05 21:18:14,540 - Epoch: [77][ 170/ 1236] Overall Loss 0.313179 Objective Loss 0.313179 LR 0.001000 Time 0.024902 +2023-10-05 21:18:14,745 - Epoch: [77][ 180/ 1236] Overall Loss 0.313159 Objective Loss 0.313159 LR 0.001000 Time 0.024655 +2023-10-05 21:18:14,950 - Epoch: [77][ 190/ 1236] Overall Loss 0.314530 Objective Loss 0.314530 LR 0.001000 Time 0.024435 +2023-10-05 21:18:15,156 - Epoch: [77][ 200/ 1236] Overall Loss 0.314435 Objective Loss 0.314435 LR 0.001000 Time 0.024242 +2023-10-05 21:18:15,358 - Epoch: [77][ 210/ 1236] Overall Loss 0.313142 Objective Loss 0.313142 LR 0.001000 Time 0.024046 +2023-10-05 21:18:15,560 - Epoch: [77][ 220/ 1236] Overall Loss 0.311869 Objective Loss 0.311869 LR 0.001000 Time 0.023867 +2023-10-05 21:18:15,761 - Epoch: [77][ 230/ 1236] Overall Loss 0.311305 Objective Loss 0.311305 LR 0.001000 Time 0.023704 +2023-10-05 21:18:15,963 - Epoch: [77][ 240/ 1236] Overall Loss 0.310324 Objective Loss 0.310324 LR 0.001000 Time 0.023554 +2023-10-05 21:18:16,164 - Epoch: [77][ 250/ 1236] Overall Loss 0.308871 Objective Loss 0.308871 LR 0.001000 Time 0.023416 +2023-10-05 21:18:16,366 - Epoch: [77][ 260/ 1236] Overall Loss 0.309608 Objective Loss 0.309608 LR 0.001000 Time 0.023289 +2023-10-05 21:18:16,567 - Epoch: [77][ 270/ 1236] Overall Loss 0.309443 Objective Loss 0.309443 LR 0.001000 Time 0.023171 +2023-10-05 21:18:16,768 - Epoch: [77][ 280/ 1236] Overall Loss 0.309605 Objective Loss 0.309605 LR 0.001000 Time 0.023060 +2023-10-05 21:18:16,970 - Epoch: [77][ 290/ 1236] Overall Loss 0.310187 Objective Loss 0.310187 LR 0.001000 Time 0.022959 +2023-10-05 21:18:17,171 - Epoch: [77][ 300/ 1236] Overall Loss 0.311045 Objective Loss 0.311045 LR 0.001000 Time 0.022863 +2023-10-05 21:18:17,373 - Epoch: [77][ 310/ 1236] Overall Loss 0.311706 Objective Loss 0.311706 LR 0.001000 Time 0.022775 +2023-10-05 21:18:17,575 - Epoch: [77][ 320/ 1236] Overall Loss 0.312422 Objective Loss 0.312422 LR 0.001000 Time 0.022692 +2023-10-05 21:18:17,777 - Epoch: [77][ 330/ 1236] Overall Loss 0.313470 Objective Loss 0.313470 LR 0.001000 Time 0.022616 +2023-10-05 21:18:17,978 - Epoch: [77][ 340/ 1236] Overall Loss 0.313491 Objective Loss 0.313491 LR 0.001000 Time 0.022542 +2023-10-05 21:18:18,180 - Epoch: [77][ 350/ 1236] Overall Loss 0.313422 Objective Loss 0.313422 LR 0.001000 Time 0.022474 +2023-10-05 21:18:18,382 - Epoch: [77][ 360/ 1236] Overall Loss 0.313120 Objective Loss 0.313120 LR 0.001000 Time 0.022409 +2023-10-05 21:18:18,585 - Epoch: [77][ 370/ 1236] Overall Loss 0.313850 Objective Loss 0.313850 LR 0.001000 Time 0.022350 +2023-10-05 21:18:18,787 - Epoch: [77][ 380/ 1236] Overall Loss 0.314107 Objective Loss 0.314107 LR 0.001000 Time 0.022293 +2023-10-05 21:18:18,990 - Epoch: [77][ 390/ 1236] Overall Loss 0.314785 Objective Loss 0.314785 LR 0.001000 Time 0.022240 +2023-10-05 21:18:19,192 - Epoch: [77][ 400/ 1236] Overall Loss 0.314886 Objective Loss 0.314886 LR 0.001000 Time 0.022189 +2023-10-05 21:18:19,395 - Epoch: [77][ 410/ 1236] Overall Loss 0.316172 Objective Loss 0.316172 LR 0.001000 Time 0.022142 +2023-10-05 21:18:19,598 - Epoch: [77][ 420/ 1236] Overall Loss 0.316695 Objective Loss 0.316695 LR 0.001000 Time 0.022096 +2023-10-05 21:18:19,800 - Epoch: [77][ 430/ 1236] Overall Loss 0.316756 Objective Loss 0.316756 LR 0.001000 Time 0.022053 +2023-10-05 21:18:20,003 - Epoch: [77][ 440/ 1236] Overall Loss 0.317257 Objective Loss 0.317257 LR 0.001000 Time 0.022011 +2023-10-05 21:18:20,206 - Epoch: [77][ 450/ 1236] Overall Loss 0.317470 Objective Loss 0.317470 LR 0.001000 Time 0.021972 +2023-10-05 21:18:20,408 - Epoch: [77][ 460/ 1236] Overall Loss 0.318128 Objective Loss 0.318128 LR 0.001000 Time 0.021934 +2023-10-05 21:18:20,611 - Epoch: [77][ 470/ 1236] Overall Loss 0.318155 Objective Loss 0.318155 LR 0.001000 Time 0.021897 +2023-10-05 21:18:20,813 - Epoch: [77][ 480/ 1236] Overall Loss 0.318422 Objective Loss 0.318422 LR 0.001000 Time 0.021861 +2023-10-05 21:18:21,016 - Epoch: [77][ 490/ 1236] Overall Loss 0.318429 Objective Loss 0.318429 LR 0.001000 Time 0.021828 +2023-10-05 21:18:21,218 - Epoch: [77][ 500/ 1236] Overall Loss 0.318433 Objective Loss 0.318433 LR 0.001000 Time 0.021795 +2023-10-05 21:18:21,421 - Epoch: [77][ 510/ 1236] Overall Loss 0.318919 Objective Loss 0.318919 LR 0.001000 Time 0.021765 +2023-10-05 21:18:21,624 - Epoch: [77][ 520/ 1236] Overall Loss 0.318631 Objective Loss 0.318631 LR 0.001000 Time 0.021735 +2023-10-05 21:18:21,827 - Epoch: [77][ 530/ 1236] Overall Loss 0.318266 Objective Loss 0.318266 LR 0.001000 Time 0.021709 +2023-10-05 21:18:22,030 - Epoch: [77][ 540/ 1236] Overall Loss 0.318543 Objective Loss 0.318543 LR 0.001000 Time 0.021681 +2023-10-05 21:18:22,233 - Epoch: [77][ 550/ 1236] Overall Loss 0.318754 Objective Loss 0.318754 LR 0.001000 Time 0.021655 +2023-10-05 21:18:22,436 - Epoch: [77][ 560/ 1236] Overall Loss 0.318771 Objective Loss 0.318771 LR 0.001000 Time 0.021630 +2023-10-05 21:18:22,639 - Epoch: [77][ 570/ 1236] Overall Loss 0.318644 Objective Loss 0.318644 LR 0.001000 Time 0.021606 +2023-10-05 21:18:22,842 - Epoch: [77][ 580/ 1236] Overall Loss 0.319632 Objective Loss 0.319632 LR 0.001000 Time 0.021582 +2023-10-05 21:18:23,045 - Epoch: [77][ 590/ 1236] Overall Loss 0.319815 Objective Loss 0.319815 LR 0.001000 Time 0.021560 +2023-10-05 21:18:23,248 - Epoch: [77][ 600/ 1236] Overall Loss 0.319713 Objective Loss 0.319713 LR 0.001000 Time 0.021538 +2023-10-05 21:18:23,452 - Epoch: [77][ 610/ 1236] Overall Loss 0.320581 Objective Loss 0.320581 LR 0.001000 Time 0.021519 +2023-10-05 21:18:23,654 - Epoch: [77][ 620/ 1236] Overall Loss 0.320853 Objective Loss 0.320853 LR 0.001000 Time 0.021498 +2023-10-05 21:18:23,858 - Epoch: [77][ 630/ 1236] Overall Loss 0.320348 Objective Loss 0.320348 LR 0.001000 Time 0.021478 +2023-10-05 21:18:24,060 - Epoch: [77][ 640/ 1236] Overall Loss 0.320599 Objective Loss 0.320599 LR 0.001000 Time 0.021459 +2023-10-05 21:18:24,264 - Epoch: [77][ 650/ 1236] Overall Loss 0.320535 Objective Loss 0.320535 LR 0.001000 Time 0.021441 +2023-10-05 21:18:24,466 - Epoch: [77][ 660/ 1236] Overall Loss 0.320716 Objective Loss 0.320716 LR 0.001000 Time 0.021422 +2023-10-05 21:18:24,669 - Epoch: [77][ 670/ 1236] Overall Loss 0.320555 Objective Loss 0.320555 LR 0.001000 Time 0.021405 +2023-10-05 21:18:24,873 - Epoch: [77][ 680/ 1236] Overall Loss 0.320090 Objective Loss 0.320090 LR 0.001000 Time 0.021389 +2023-10-05 21:18:25,078 - Epoch: [77][ 690/ 1236] Overall Loss 0.320574 Objective Loss 0.320574 LR 0.001000 Time 0.021376 +2023-10-05 21:18:25,282 - Epoch: [77][ 700/ 1236] Overall Loss 0.320746 Objective Loss 0.320746 LR 0.001000 Time 0.021361 +2023-10-05 21:18:25,484 - Epoch: [77][ 710/ 1236] Overall Loss 0.320836 Objective Loss 0.320836 LR 0.001000 Time 0.021344 +2023-10-05 21:18:25,689 - Epoch: [77][ 720/ 1236] Overall Loss 0.320849 Objective Loss 0.320849 LR 0.001000 Time 0.021332 +2023-10-05 21:18:25,892 - Epoch: [77][ 730/ 1236] Overall Loss 0.320926 Objective Loss 0.320926 LR 0.001000 Time 0.021317 +2023-10-05 21:18:26,094 - Epoch: [77][ 740/ 1236] Overall Loss 0.321091 Objective Loss 0.321091 LR 0.001000 Time 0.021302 +2023-10-05 21:18:26,297 - Epoch: [77][ 750/ 1236] Overall Loss 0.321132 Objective Loss 0.321132 LR 0.001000 Time 0.021288 +2023-10-05 21:18:26,500 - Epoch: [77][ 760/ 1236] Overall Loss 0.321073 Objective Loss 0.321073 LR 0.001000 Time 0.021274 +2023-10-05 21:18:26,702 - Epoch: [77][ 770/ 1236] Overall Loss 0.321115 Objective Loss 0.321115 LR 0.001000 Time 0.021260 +2023-10-05 21:18:26,905 - Epoch: [77][ 780/ 1236] Overall Loss 0.321348 Objective Loss 0.321348 LR 0.001000 Time 0.021247 +2023-10-05 21:18:27,108 - Epoch: [77][ 790/ 1236] Overall Loss 0.321342 Objective Loss 0.321342 LR 0.001000 Time 0.021235 +2023-10-05 21:18:27,311 - Epoch: [77][ 800/ 1236] Overall Loss 0.321444 Objective Loss 0.321444 LR 0.001000 Time 0.021222 +2023-10-05 21:18:27,513 - Epoch: [77][ 810/ 1236] Overall Loss 0.321420 Objective Loss 0.321420 LR 0.001000 Time 0.021210 +2023-10-05 21:18:27,716 - Epoch: [77][ 820/ 1236] Overall Loss 0.321732 Objective Loss 0.321732 LR 0.001000 Time 0.021198 +2023-10-05 21:18:27,919 - Epoch: [77][ 830/ 1236] Overall Loss 0.322212 Objective Loss 0.322212 LR 0.001000 Time 0.021187 +2023-10-05 21:18:28,122 - Epoch: [77][ 840/ 1236] Overall Loss 0.322637 Objective Loss 0.322637 LR 0.001000 Time 0.021175 +2023-10-05 21:18:28,324 - Epoch: [77][ 850/ 1236] Overall Loss 0.322860 Objective Loss 0.322860 LR 0.001000 Time 0.021164 +2023-10-05 21:18:28,527 - Epoch: [77][ 860/ 1236] Overall Loss 0.322959 Objective Loss 0.322959 LR 0.001000 Time 0.021154 +2023-10-05 21:18:28,729 - Epoch: [77][ 870/ 1236] Overall Loss 0.322838 Objective Loss 0.322838 LR 0.001000 Time 0.021142 +2023-10-05 21:18:28,932 - Epoch: [77][ 880/ 1236] Overall Loss 0.322985 Objective Loss 0.322985 LR 0.001000 Time 0.021132 +2023-10-05 21:18:29,134 - Epoch: [77][ 890/ 1236] Overall Loss 0.322669 Objective Loss 0.322669 LR 0.001000 Time 0.021122 +2023-10-05 21:18:29,338 - Epoch: [77][ 900/ 1236] Overall Loss 0.323151 Objective Loss 0.323151 LR 0.001000 Time 0.021113 +2023-10-05 21:18:29,540 - Epoch: [77][ 910/ 1236] Overall Loss 0.322803 Objective Loss 0.322803 LR 0.001000 Time 0.021103 +2023-10-05 21:18:29,743 - Epoch: [77][ 920/ 1236] Overall Loss 0.322829 Objective Loss 0.322829 LR 0.001000 Time 0.021094 +2023-10-05 21:18:29,946 - Epoch: [77][ 930/ 1236] Overall Loss 0.323080 Objective Loss 0.323080 LR 0.001000 Time 0.021084 +2023-10-05 21:18:30,149 - Epoch: [77][ 940/ 1236] Overall Loss 0.323232 Objective Loss 0.323232 LR 0.001000 Time 0.021076 +2023-10-05 21:18:30,351 - Epoch: [77][ 950/ 1236] Overall Loss 0.323109 Objective Loss 0.323109 LR 0.001000 Time 0.021067 +2023-10-05 21:18:30,554 - Epoch: [77][ 960/ 1236] Overall Loss 0.323181 Objective Loss 0.323181 LR 0.001000 Time 0.021058 +2023-10-05 21:18:30,757 - Epoch: [77][ 970/ 1236] Overall Loss 0.323173 Objective Loss 0.323173 LR 0.001000 Time 0.021049 +2023-10-05 21:18:30,960 - Epoch: [77][ 980/ 1236] Overall Loss 0.323167 Objective Loss 0.323167 LR 0.001000 Time 0.021041 +2023-10-05 21:18:31,162 - Epoch: [77][ 990/ 1236] Overall Loss 0.323272 Objective Loss 0.323272 LR 0.001000 Time 0.021033 +2023-10-05 21:18:31,365 - Epoch: [77][ 1000/ 1236] Overall Loss 0.323474 Objective Loss 0.323474 LR 0.001000 Time 0.021025 +2023-10-05 21:18:31,567 - Epoch: [77][ 1010/ 1236] Overall Loss 0.323393 Objective Loss 0.323393 LR 0.001000 Time 0.021017 +2023-10-05 21:18:31,770 - Epoch: [77][ 1020/ 1236] Overall Loss 0.323414 Objective Loss 0.323414 LR 0.001000 Time 0.021009 +2023-10-05 21:18:31,972 - Epoch: [77][ 1030/ 1236] Overall Loss 0.323940 Objective Loss 0.323940 LR 0.001000 Time 0.021001 +2023-10-05 21:18:32,175 - Epoch: [77][ 1040/ 1236] Overall Loss 0.324044 Objective Loss 0.324044 LR 0.001000 Time 0.020994 +2023-10-05 21:18:32,377 - Epoch: [77][ 1050/ 1236] Overall Loss 0.323997 Objective Loss 0.323997 LR 0.001000 Time 0.020986 +2023-10-05 21:18:32,580 - Epoch: [77][ 1060/ 1236] Overall Loss 0.323606 Objective Loss 0.323606 LR 0.001000 Time 0.020979 +2023-10-05 21:18:32,783 - Epoch: [77][ 1070/ 1236] Overall Loss 0.323681 Objective Loss 0.323681 LR 0.001000 Time 0.020972 +2023-10-05 21:18:32,986 - Epoch: [77][ 1080/ 1236] Overall Loss 0.323886 Objective Loss 0.323886 LR 0.001000 Time 0.020966 +2023-10-05 21:18:33,188 - Epoch: [77][ 1090/ 1236] Overall Loss 0.323971 Objective Loss 0.323971 LR 0.001000 Time 0.020958 +2023-10-05 21:18:33,391 - Epoch: [77][ 1100/ 1236] Overall Loss 0.324089 Objective Loss 0.324089 LR 0.001000 Time 0.020952 +2023-10-05 21:18:33,593 - Epoch: [77][ 1110/ 1236] Overall Loss 0.324022 Objective Loss 0.324022 LR 0.001000 Time 0.020945 +2023-10-05 21:18:33,796 - Epoch: [77][ 1120/ 1236] Overall Loss 0.323989 Objective Loss 0.323989 LR 0.001000 Time 0.020939 +2023-10-05 21:18:33,998 - Epoch: [77][ 1130/ 1236] Overall Loss 0.323747 Objective Loss 0.323747 LR 0.001000 Time 0.020933 +2023-10-05 21:18:34,201 - Epoch: [77][ 1140/ 1236] Overall Loss 0.323636 Objective Loss 0.323636 LR 0.001000 Time 0.020927 +2023-10-05 21:18:34,403 - Epoch: [77][ 1150/ 1236] Overall Loss 0.323684 Objective Loss 0.323684 LR 0.001000 Time 0.020920 +2023-10-05 21:18:34,606 - Epoch: [77][ 1160/ 1236] Overall Loss 0.323647 Objective Loss 0.323647 LR 0.001000 Time 0.020914 +2023-10-05 21:18:34,808 - Epoch: [77][ 1170/ 1236] Overall Loss 0.323861 Objective Loss 0.323861 LR 0.001000 Time 0.020908 +2023-10-05 21:18:35,011 - Epoch: [77][ 1180/ 1236] Overall Loss 0.323871 Objective Loss 0.323871 LR 0.001000 Time 0.020902 +2023-10-05 21:18:35,213 - Epoch: [77][ 1190/ 1236] Overall Loss 0.323883 Objective Loss 0.323883 LR 0.001000 Time 0.020896 +2023-10-05 21:18:35,416 - Epoch: [77][ 1200/ 1236] Overall Loss 0.323762 Objective Loss 0.323762 LR 0.001000 Time 0.020891 +2023-10-05 21:18:35,619 - Epoch: [77][ 1210/ 1236] Overall Loss 0.323938 Objective Loss 0.323938 LR 0.001000 Time 0.020886 +2023-10-05 21:18:35,822 - Epoch: [77][ 1220/ 1236] Overall Loss 0.323923 Objective Loss 0.323923 LR 0.001000 Time 0.020881 +2023-10-05 21:18:36,080 - Epoch: [77][ 1230/ 1236] Overall Loss 0.323702 Objective Loss 0.323702 LR 0.001000 Time 0.020920 +2023-10-05 21:18:36,198 - Epoch: [77][ 1236/ 1236] Overall Loss 0.323907 Objective Loss 0.323907 Top1 84.928717 Top5 97.352342 LR 0.001000 Time 0.020914 +2023-10-05 21:18:36,338 - --- validate (epoch=77)----------- +2023-10-05 21:18:36,339 - 29943 samples (256 per mini-batch) +2023-10-05 21:18:36,794 - Epoch: [77][ 10/ 117] Loss 0.342349 Top1 82.148438 Top5 97.773438 +2023-10-05 21:18:36,952 - Epoch: [77][ 20/ 117] Loss 0.353400 Top1 82.148438 Top5 97.558594 +2023-10-05 21:18:37,108 - Epoch: [77][ 30/ 117] Loss 0.356790 Top1 82.044271 Top5 97.669271 +2023-10-05 21:18:37,260 - Epoch: [77][ 40/ 117] Loss 0.367179 Top1 81.816406 Top5 97.656250 +2023-10-05 21:18:37,410 - Epoch: [77][ 50/ 117] Loss 0.369741 Top1 81.851562 Top5 97.609375 +2023-10-05 21:18:37,558 - Epoch: [77][ 60/ 117] Loss 0.365451 Top1 81.875000 Top5 97.669271 +2023-10-05 21:18:37,710 - Epoch: [77][ 70/ 117] Loss 0.366754 Top1 81.875000 Top5 97.656250 +2023-10-05 21:18:37,859 - Epoch: [77][ 80/ 117] Loss 0.370403 Top1 81.835938 Top5 97.587891 +2023-10-05 21:18:38,012 - Epoch: [77][ 90/ 117] Loss 0.370036 Top1 81.957465 Top5 97.582465 +2023-10-05 21:18:38,160 - Epoch: [77][ 100/ 117] Loss 0.367965 Top1 82.011719 Top5 97.597656 +2023-10-05 21:18:38,322 - Epoch: [77][ 110/ 117] Loss 0.363781 Top1 82.045455 Top5 97.560369 +2023-10-05 21:18:38,408 - Epoch: [77][ 117/ 117] Loss 0.362740 Top1 82.049227 Top5 97.548676 +2023-10-05 21:18:38,525 - ==> Top1: 82.049 Top5: 97.549 Loss: 0.363 + +2023-10-05 21:18:38,525 - ==> Confusion: +[[ 916 1 0 2 11 1 0 2 6 71 0 1 2 3 11 2 4 1 2 0 14] + [ 0 1036 0 0 19 23 0 23 2 0 3 1 0 0 3 4 5 0 6 1 5] + [ 4 3 950 14 1 1 19 10 0 0 9 2 7 0 0 7 5 2 6 5 11] + [ 2 1 23 953 1 5 2 0 4 2 9 0 5 5 31 5 2 7 19 0 13] + [ 25 8 0 0 972 3 0 3 0 7 2 2 0 3 13 1 8 1 0 0 2] + [ 7 43 0 0 2 955 2 27 2 5 5 8 1 19 11 2 4 0 5 6 12] + [ 1 6 36 0 1 1 1089 16 0 0 5 4 2 0 1 10 1 1 1 5 11] + [ 4 24 11 0 6 31 4 1050 2 2 6 8 1 1 0 3 1 0 42 7 15] + [ 15 4 2 1 0 3 0 0 963 38 11 0 2 13 21 3 3 1 4 0 5] + [ 98 1 1 1 9 5 1 0 29 923 0 0 0 25 10 4 2 0 1 1 8] + [ 5 1 11 11 1 4 2 3 13 0 956 0 0 14 6 1 1 0 10 6 8] + [ 0 0 3 0 0 8 0 2 0 3 0 929 46 9 0 3 2 13 0 15 2] + [ 1 2 4 6 1 3 1 2 1 0 1 31 962 6 4 10 2 18 1 4 8] + [ 4 0 2 1 6 7 0 0 18 17 8 3 2 1032 3 0 2 0 0 4 10] + [ 11 2 3 12 3 0 0 0 30 6 2 2 1 1 1012 0 0 1 5 0 10] + [ 0 2 1 2 5 1 0 0 0 0 0 9 9 1 0 1056 20 13 1 8 6] + [ 0 14 1 1 9 4 0 1 0 0 0 2 1 1 3 13 1102 0 1 2 6] + [ 0 1 2 1 0 0 0 0 4 0 0 4 16 1 0 8 0 987 3 3 8] + [ 3 10 5 15 2 0 1 25 5 0 7 0 1 1 8 0 0 0 972 0 13] + [ 0 5 4 0 1 4 5 24 0 0 0 15 8 2 0 8 12 0 4 1050 10] + [ 139 219 134 83 119 173 30 112 97 108 185 114 374 358 207 67 222 76 194 191 4703]] + +2023-10-05 21:18:38,527 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:18:38,527 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:18:38,533 - + +2023-10-05 21:18:38,533 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:18:39,528 - Epoch: [78][ 10/ 1236] Overall Loss 0.325601 Objective Loss 0.325601 LR 0.001000 Time 0.099466 +2023-10-05 21:18:39,731 - Epoch: [78][ 20/ 1236] Overall Loss 0.323934 Objective Loss 0.323934 LR 0.001000 Time 0.059886 +2023-10-05 21:18:39,934 - Epoch: [78][ 30/ 1236] Overall Loss 0.324073 Objective Loss 0.324073 LR 0.001000 Time 0.046658 +2023-10-05 21:18:40,137 - Epoch: [78][ 40/ 1236] Overall Loss 0.311208 Objective Loss 0.311208 LR 0.001000 Time 0.040077 +2023-10-05 21:18:40,339 - Epoch: [78][ 50/ 1236] Overall Loss 0.306763 Objective Loss 0.306763 LR 0.001000 Time 0.036085 +2023-10-05 21:18:40,542 - Epoch: [78][ 60/ 1236] Overall Loss 0.309634 Objective Loss 0.309634 LR 0.001000 Time 0.033455 +2023-10-05 21:18:40,744 - Epoch: [78][ 70/ 1236] Overall Loss 0.311694 Objective Loss 0.311694 LR 0.001000 Time 0.031551 +2023-10-05 21:18:40,947 - Epoch: [78][ 80/ 1236] Overall Loss 0.317868 Objective Loss 0.317868 LR 0.001000 Time 0.030141 +2023-10-05 21:18:41,149 - Epoch: [78][ 90/ 1236] Overall Loss 0.319925 Objective Loss 0.319925 LR 0.001000 Time 0.029030 +2023-10-05 21:18:41,351 - Epoch: [78][ 100/ 1236] Overall Loss 0.318807 Objective Loss 0.318807 LR 0.001000 Time 0.028146 +2023-10-05 21:18:41,554 - Epoch: [78][ 110/ 1236] Overall Loss 0.318890 Objective Loss 0.318890 LR 0.001000 Time 0.027428 +2023-10-05 21:18:41,759 - Epoch: [78][ 120/ 1236] Overall Loss 0.319335 Objective Loss 0.319335 LR 0.001000 Time 0.026850 +2023-10-05 21:18:41,963 - Epoch: [78][ 130/ 1236] Overall Loss 0.321463 Objective Loss 0.321463 LR 0.001000 Time 0.026344 +2023-10-05 21:18:42,169 - Epoch: [78][ 140/ 1236] Overall Loss 0.322637 Objective Loss 0.322637 LR 0.001000 Time 0.025925 +2023-10-05 21:18:42,372 - Epoch: [78][ 150/ 1236] Overall Loss 0.320727 Objective Loss 0.320727 LR 0.001000 Time 0.025551 +2023-10-05 21:18:42,578 - Epoch: [78][ 160/ 1236] Overall Loss 0.321533 Objective Loss 0.321533 LR 0.001000 Time 0.025236 +2023-10-05 21:18:42,784 - Epoch: [78][ 170/ 1236] Overall Loss 0.320811 Objective Loss 0.320811 LR 0.001000 Time 0.024961 +2023-10-05 21:18:42,989 - Epoch: [78][ 180/ 1236] Overall Loss 0.322150 Objective Loss 0.322150 LR 0.001000 Time 0.024716 +2023-10-05 21:18:43,194 - Epoch: [78][ 190/ 1236] Overall Loss 0.320291 Objective Loss 0.320291 LR 0.001000 Time 0.024491 +2023-10-05 21:18:43,400 - Epoch: [78][ 200/ 1236] Overall Loss 0.319457 Objective Loss 0.319457 LR 0.001000 Time 0.024293 +2023-10-05 21:18:43,607 - Epoch: [78][ 210/ 1236] Overall Loss 0.318558 Objective Loss 0.318558 LR 0.001000 Time 0.024124 +2023-10-05 21:18:43,813 - Epoch: [78][ 220/ 1236] Overall Loss 0.319166 Objective Loss 0.319166 LR 0.001000 Time 0.023961 +2023-10-05 21:18:44,018 - Epoch: [78][ 230/ 1236] Overall Loss 0.319168 Objective Loss 0.319168 LR 0.001000 Time 0.023802 +2023-10-05 21:18:44,223 - Epoch: [78][ 240/ 1236] Overall Loss 0.318974 Objective Loss 0.318974 LR 0.001000 Time 0.023663 +2023-10-05 21:18:44,427 - Epoch: [78][ 250/ 1236] Overall Loss 0.318549 Objective Loss 0.318549 LR 0.001000 Time 0.023533 +2023-10-05 21:18:44,633 - Epoch: [78][ 260/ 1236] Overall Loss 0.317104 Objective Loss 0.317104 LR 0.001000 Time 0.023416 +2023-10-05 21:18:44,838 - Epoch: [78][ 270/ 1236] Overall Loss 0.317693 Objective Loss 0.317693 LR 0.001000 Time 0.023307 +2023-10-05 21:18:45,043 - Epoch: [78][ 280/ 1236] Overall Loss 0.318365 Objective Loss 0.318365 LR 0.001000 Time 0.023205 +2023-10-05 21:18:45,247 - Epoch: [78][ 290/ 1236] Overall Loss 0.318794 Objective Loss 0.318794 LR 0.001000 Time 0.023106 +2023-10-05 21:18:45,453 - Epoch: [78][ 300/ 1236] Overall Loss 0.318575 Objective Loss 0.318575 LR 0.001000 Time 0.023019 +2023-10-05 21:18:45,659 - Epoch: [78][ 310/ 1236] Overall Loss 0.318952 Objective Loss 0.318952 LR 0.001000 Time 0.022939 +2023-10-05 21:18:45,869 - Epoch: [78][ 320/ 1236] Overall Loss 0.318500 Objective Loss 0.318500 LR 0.001000 Time 0.022881 +2023-10-05 21:18:46,084 - Epoch: [78][ 330/ 1236] Overall Loss 0.318003 Objective Loss 0.318003 LR 0.001000 Time 0.022836 +2023-10-05 21:18:46,290 - Epoch: [78][ 340/ 1236] Overall Loss 0.318252 Objective Loss 0.318252 LR 0.001000 Time 0.022771 +2023-10-05 21:18:46,496 - Epoch: [78][ 350/ 1236] Overall Loss 0.318447 Objective Loss 0.318447 LR 0.001000 Time 0.022705 +2023-10-05 21:18:46,701 - Epoch: [78][ 360/ 1236] Overall Loss 0.318603 Objective Loss 0.318603 LR 0.001000 Time 0.022644 +2023-10-05 21:18:46,907 - Epoch: [78][ 370/ 1236] Overall Loss 0.318532 Objective Loss 0.318532 LR 0.001000 Time 0.022589 +2023-10-05 21:18:47,115 - Epoch: [78][ 380/ 1236] Overall Loss 0.318496 Objective Loss 0.318496 LR 0.001000 Time 0.022536 +2023-10-05 21:18:47,330 - Epoch: [78][ 390/ 1236] Overall Loss 0.317859 Objective Loss 0.317859 LR 0.001000 Time 0.022509 +2023-10-05 21:18:47,541 - Epoch: [78][ 400/ 1236] Overall Loss 0.318322 Objective Loss 0.318322 LR 0.001000 Time 0.022473 +2023-10-05 21:18:47,754 - Epoch: [78][ 410/ 1236] Overall Loss 0.318421 Objective Loss 0.318421 LR 0.001000 Time 0.022444 +2023-10-05 21:18:47,966 - Epoch: [78][ 420/ 1236] Overall Loss 0.317577 Objective Loss 0.317577 LR 0.001000 Time 0.022413 +2023-10-05 21:18:48,178 - Epoch: [78][ 430/ 1236] Overall Loss 0.317821 Objective Loss 0.317821 LR 0.001000 Time 0.022385 +2023-10-05 21:18:48,390 - Epoch: [78][ 440/ 1236] Overall Loss 0.317803 Objective Loss 0.317803 LR 0.001000 Time 0.022356 +2023-10-05 21:18:48,609 - Epoch: [78][ 450/ 1236] Overall Loss 0.317305 Objective Loss 0.317305 LR 0.001000 Time 0.022345 +2023-10-05 21:18:48,815 - Epoch: [78][ 460/ 1236] Overall Loss 0.317242 Objective Loss 0.317242 LR 0.001000 Time 0.022306 +2023-10-05 21:18:49,017 - Epoch: [78][ 470/ 1236] Overall Loss 0.316381 Objective Loss 0.316381 LR 0.001000 Time 0.022261 +2023-10-05 21:18:49,223 - Epoch: [78][ 480/ 1236] Overall Loss 0.316110 Objective Loss 0.316110 LR 0.001000 Time 0.022225 +2023-10-05 21:18:49,426 - Epoch: [78][ 490/ 1236] Overall Loss 0.316683 Objective Loss 0.316683 LR 0.001000 Time 0.022185 +2023-10-05 21:18:49,631 - Epoch: [78][ 500/ 1236] Overall Loss 0.316848 Objective Loss 0.316848 LR 0.001000 Time 0.022152 +2023-10-05 21:18:49,835 - Epoch: [78][ 510/ 1236] Overall Loss 0.317266 Objective Loss 0.317266 LR 0.001000 Time 0.022116 +2023-10-05 21:18:50,040 - Epoch: [78][ 520/ 1236] Overall Loss 0.316902 Objective Loss 0.316902 LR 0.001000 Time 0.022084 +2023-10-05 21:18:50,244 - Epoch: [78][ 530/ 1236] Overall Loss 0.317244 Objective Loss 0.317244 LR 0.001000 Time 0.022051 +2023-10-05 21:18:50,449 - Epoch: [78][ 540/ 1236] Overall Loss 0.317506 Objective Loss 0.317506 LR 0.001000 Time 0.022022 +2023-10-05 21:18:50,653 - Epoch: [78][ 550/ 1236] Overall Loss 0.317224 Objective Loss 0.317224 LR 0.001000 Time 0.021992 +2023-10-05 21:18:50,858 - Epoch: [78][ 560/ 1236] Overall Loss 0.317554 Objective Loss 0.317554 LR 0.001000 Time 0.021965 +2023-10-05 21:18:51,061 - Epoch: [78][ 570/ 1236] Overall Loss 0.317055 Objective Loss 0.317055 LR 0.001000 Time 0.021936 +2023-10-05 21:18:51,266 - Epoch: [78][ 580/ 1236] Overall Loss 0.316717 Objective Loss 0.316717 LR 0.001000 Time 0.021910 +2023-10-05 21:18:51,470 - Epoch: [78][ 590/ 1236] Overall Loss 0.316404 Objective Loss 0.316404 LR 0.001000 Time 0.021884 +2023-10-05 21:18:51,675 - Epoch: [78][ 600/ 1236] Overall Loss 0.315694 Objective Loss 0.315694 LR 0.001000 Time 0.021860 +2023-10-05 21:18:51,879 - Epoch: [78][ 610/ 1236] Overall Loss 0.315208 Objective Loss 0.315208 LR 0.001000 Time 0.021836 +2023-10-05 21:18:52,085 - Epoch: [78][ 620/ 1236] Overall Loss 0.315386 Objective Loss 0.315386 LR 0.001000 Time 0.021815 +2023-10-05 21:18:52,289 - Epoch: [78][ 630/ 1236] Overall Loss 0.315646 Objective Loss 0.315646 LR 0.001000 Time 0.021791 +2023-10-05 21:18:52,494 - Epoch: [78][ 640/ 1236] Overall Loss 0.315115 Objective Loss 0.315115 LR 0.001000 Time 0.021771 +2023-10-05 21:18:52,708 - Epoch: [78][ 650/ 1236] Overall Loss 0.314625 Objective Loss 0.314625 LR 0.001000 Time 0.021765 +2023-10-05 21:18:52,921 - Epoch: [78][ 660/ 1236] Overall Loss 0.315225 Objective Loss 0.315225 LR 0.001000 Time 0.021757 +2023-10-05 21:18:53,134 - Epoch: [78][ 670/ 1236] Overall Loss 0.315366 Objective Loss 0.315366 LR 0.001000 Time 0.021750 +2023-10-05 21:18:53,346 - Epoch: [78][ 680/ 1236] Overall Loss 0.315341 Objective Loss 0.315341 LR 0.001000 Time 0.021742 +2023-10-05 21:18:53,560 - Epoch: [78][ 690/ 1236] Overall Loss 0.315747 Objective Loss 0.315747 LR 0.001000 Time 0.021736 +2023-10-05 21:18:53,772 - Epoch: [78][ 700/ 1236] Overall Loss 0.315448 Objective Loss 0.315448 LR 0.001000 Time 0.021728 +2023-10-05 21:18:53,985 - Epoch: [78][ 710/ 1236] Overall Loss 0.314554 Objective Loss 0.314554 LR 0.001000 Time 0.021722 +2023-10-05 21:18:54,198 - Epoch: [78][ 720/ 1236] Overall Loss 0.314765 Objective Loss 0.314765 LR 0.001000 Time 0.021714 +2023-10-05 21:18:54,411 - Epoch: [78][ 730/ 1236] Overall Loss 0.314307 Objective Loss 0.314307 LR 0.001000 Time 0.021709 +2023-10-05 21:18:54,623 - Epoch: [78][ 740/ 1236] Overall Loss 0.314252 Objective Loss 0.314252 LR 0.001000 Time 0.021702 +2023-10-05 21:18:54,837 - Epoch: [78][ 750/ 1236] Overall Loss 0.314154 Objective Loss 0.314154 LR 0.001000 Time 0.021697 +2023-10-05 21:18:55,049 - Epoch: [78][ 760/ 1236] Overall Loss 0.313953 Objective Loss 0.313953 LR 0.001000 Time 0.021690 +2023-10-05 21:18:55,262 - Epoch: [78][ 770/ 1236] Overall Loss 0.313540 Objective Loss 0.313540 LR 0.001000 Time 0.021685 +2023-10-05 21:18:55,474 - Epoch: [78][ 780/ 1236] Overall Loss 0.313861 Objective Loss 0.313861 LR 0.001000 Time 0.021678 +2023-10-05 21:18:55,688 - Epoch: [78][ 790/ 1236] Overall Loss 0.313869 Objective Loss 0.313869 LR 0.001000 Time 0.021674 +2023-10-05 21:18:55,900 - Epoch: [78][ 800/ 1236] Overall Loss 0.313891 Objective Loss 0.313891 LR 0.001000 Time 0.021668 +2023-10-05 21:18:56,114 - Epoch: [78][ 810/ 1236] Overall Loss 0.314072 Objective Loss 0.314072 LR 0.001000 Time 0.021664 +2023-10-05 21:18:56,326 - Epoch: [78][ 820/ 1236] Overall Loss 0.314142 Objective Loss 0.314142 LR 0.001000 Time 0.021658 +2023-10-05 21:18:56,539 - Epoch: [78][ 830/ 1236] Overall Loss 0.314280 Objective Loss 0.314280 LR 0.001000 Time 0.021654 +2023-10-05 21:18:56,752 - Epoch: [78][ 840/ 1236] Overall Loss 0.314597 Objective Loss 0.314597 LR 0.001000 Time 0.021649 +2023-10-05 21:18:56,965 - Epoch: [78][ 850/ 1236] Overall Loss 0.314365 Objective Loss 0.314365 LR 0.001000 Time 0.021645 +2023-10-05 21:18:57,177 - Epoch: [78][ 860/ 1236] Overall Loss 0.314022 Objective Loss 0.314022 LR 0.001000 Time 0.021639 +2023-10-05 21:18:57,391 - Epoch: [78][ 870/ 1236] Overall Loss 0.314107 Objective Loss 0.314107 LR 0.001000 Time 0.021636 +2023-10-05 21:18:57,604 - Epoch: [78][ 880/ 1236] Overall Loss 0.314304 Objective Loss 0.314304 LR 0.001000 Time 0.021631 +2023-10-05 21:18:57,817 - Epoch: [78][ 890/ 1236] Overall Loss 0.314592 Objective Loss 0.314592 LR 0.001000 Time 0.021628 +2023-10-05 21:18:58,029 - Epoch: [78][ 900/ 1236] Overall Loss 0.314498 Objective Loss 0.314498 LR 0.001000 Time 0.021623 +2023-10-05 21:18:58,243 - Epoch: [78][ 910/ 1236] Overall Loss 0.314133 Objective Loss 0.314133 LR 0.001000 Time 0.021620 +2023-10-05 21:18:58,456 - Epoch: [78][ 920/ 1236] Overall Loss 0.314376 Objective Loss 0.314376 LR 0.001000 Time 0.021616 +2023-10-05 21:18:58,670 - Epoch: [78][ 930/ 1236] Overall Loss 0.314338 Objective Loss 0.314338 LR 0.001000 Time 0.021613 +2023-10-05 21:18:58,883 - Epoch: [78][ 940/ 1236] Overall Loss 0.314475 Objective Loss 0.314475 LR 0.001000 Time 0.021610 +2023-10-05 21:18:59,096 - Epoch: [78][ 950/ 1236] Overall Loss 0.314402 Objective Loss 0.314402 LR 0.001000 Time 0.021606 +2023-10-05 21:18:59,309 - Epoch: [78][ 960/ 1236] Overall Loss 0.314488 Objective Loss 0.314488 LR 0.001000 Time 0.021602 +2023-10-05 21:18:59,522 - Epoch: [78][ 970/ 1236] Overall Loss 0.314205 Objective Loss 0.314205 LR 0.001000 Time 0.021599 +2023-10-05 21:18:59,734 - Epoch: [78][ 980/ 1236] Overall Loss 0.314512 Objective Loss 0.314512 LR 0.001000 Time 0.021595 +2023-10-05 21:18:59,948 - Epoch: [78][ 990/ 1236] Overall Loss 0.314888 Objective Loss 0.314888 LR 0.001000 Time 0.021592 +2023-10-05 21:19:00,160 - Epoch: [78][ 1000/ 1236] Overall Loss 0.315175 Objective Loss 0.315175 LR 0.001000 Time 0.021588 +2023-10-05 21:19:00,374 - Epoch: [78][ 1010/ 1236] Overall Loss 0.314774 Objective Loss 0.314774 LR 0.001000 Time 0.021586 +2023-10-05 21:19:00,588 - Epoch: [78][ 1020/ 1236] Overall Loss 0.314888 Objective Loss 0.314888 LR 0.001000 Time 0.021584 +2023-10-05 21:19:00,802 - Epoch: [78][ 1030/ 1236] Overall Loss 0.314869 Objective Loss 0.314869 LR 0.001000 Time 0.021582 +2023-10-05 21:19:01,014 - Epoch: [78][ 1040/ 1236] Overall Loss 0.314648 Objective Loss 0.314648 LR 0.001000 Time 0.021578 +2023-10-05 21:19:01,228 - Epoch: [78][ 1050/ 1236] Overall Loss 0.314630 Objective Loss 0.314630 LR 0.001000 Time 0.021576 +2023-10-05 21:19:01,440 - Epoch: [78][ 1060/ 1236] Overall Loss 0.314605 Objective Loss 0.314605 LR 0.001000 Time 0.021572 +2023-10-05 21:19:01,654 - Epoch: [78][ 1070/ 1236] Overall Loss 0.314499 Objective Loss 0.314499 LR 0.001000 Time 0.021570 +2023-10-05 21:19:01,866 - Epoch: [78][ 1080/ 1236] Overall Loss 0.314465 Objective Loss 0.314465 LR 0.001000 Time 0.021566 +2023-10-05 21:19:02,080 - Epoch: [78][ 1090/ 1236] Overall Loss 0.314367 Objective Loss 0.314367 LR 0.001000 Time 0.021564 +2023-10-05 21:19:02,289 - Epoch: [78][ 1100/ 1236] Overall Loss 0.314270 Objective Loss 0.314270 LR 0.001000 Time 0.021558 +2023-10-05 21:19:02,492 - Epoch: [78][ 1110/ 1236] Overall Loss 0.314348 Objective Loss 0.314348 LR 0.001000 Time 0.021547 +2023-10-05 21:19:02,697 - Epoch: [78][ 1120/ 1236] Overall Loss 0.314470 Objective Loss 0.314470 LR 0.001000 Time 0.021537 +2023-10-05 21:19:02,901 - Epoch: [78][ 1130/ 1236] Overall Loss 0.314502 Objective Loss 0.314502 LR 0.001000 Time 0.021526 +2023-10-05 21:19:03,109 - Epoch: [78][ 1140/ 1236] Overall Loss 0.314492 Objective Loss 0.314492 LR 0.001000 Time 0.021520 +2023-10-05 21:19:03,312 - Epoch: [78][ 1150/ 1236] Overall Loss 0.314364 Objective Loss 0.314364 LR 0.001000 Time 0.021509 +2023-10-05 21:19:03,514 - Epoch: [78][ 1160/ 1236] Overall Loss 0.314182 Objective Loss 0.314182 LR 0.001000 Time 0.021498 +2023-10-05 21:19:03,716 - Epoch: [78][ 1170/ 1236] Overall Loss 0.314310 Objective Loss 0.314310 LR 0.001000 Time 0.021486 +2023-10-05 21:19:03,920 - Epoch: [78][ 1180/ 1236] Overall Loss 0.314500 Objective Loss 0.314500 LR 0.001000 Time 0.021476 +2023-10-05 21:19:04,121 - Epoch: [78][ 1190/ 1236] Overall Loss 0.314590 Objective Loss 0.314590 LR 0.001000 Time 0.021465 +2023-10-05 21:19:04,324 - Epoch: [78][ 1200/ 1236] Overall Loss 0.314614 Objective Loss 0.314614 LR 0.001000 Time 0.021455 +2023-10-05 21:19:04,526 - Epoch: [78][ 1210/ 1236] Overall Loss 0.314398 Objective Loss 0.314398 LR 0.001000 Time 0.021444 +2023-10-05 21:19:04,729 - Epoch: [78][ 1220/ 1236] Overall Loss 0.314635 Objective Loss 0.314635 LR 0.001000 Time 0.021434 +2023-10-05 21:19:04,981 - Epoch: [78][ 1230/ 1236] Overall Loss 0.314649 Objective Loss 0.314649 LR 0.001000 Time 0.021464 +2023-10-05 21:19:05,098 - Epoch: [78][ 1236/ 1236] Overall Loss 0.314678 Objective Loss 0.314678 Top1 85.539715 Top5 97.963340 LR 0.001000 Time 0.021455 +2023-10-05 21:19:05,225 - --- validate (epoch=78)----------- +2023-10-05 21:19:05,225 - 29943 samples (256 per mini-batch) +2023-10-05 21:19:05,675 - Epoch: [78][ 10/ 117] Loss 0.327754 Top1 82.539062 Top5 97.929688 +2023-10-05 21:19:05,822 - Epoch: [78][ 20/ 117] Loss 0.334674 Top1 82.695312 Top5 97.812500 +2023-10-05 21:19:05,967 - Epoch: [78][ 30/ 117] Loss 0.351613 Top1 82.408854 Top5 97.721354 +2023-10-05 21:19:06,113 - Epoch: [78][ 40/ 117] Loss 0.357768 Top1 82.539062 Top5 97.753906 +2023-10-05 21:19:06,258 - Epoch: [78][ 50/ 117] Loss 0.368329 Top1 82.078125 Top5 97.617188 +2023-10-05 21:19:06,405 - Epoch: [78][ 60/ 117] Loss 0.365062 Top1 82.109375 Top5 97.597656 +2023-10-05 21:19:06,551 - Epoch: [78][ 70/ 117] Loss 0.365390 Top1 82.025670 Top5 97.594866 +2023-10-05 21:19:06,698 - Epoch: [78][ 80/ 117] Loss 0.367340 Top1 81.958008 Top5 97.558594 +2023-10-05 21:19:06,843 - Epoch: [78][ 90/ 117] Loss 0.363743 Top1 81.979167 Top5 97.569444 +2023-10-05 21:19:06,990 - Epoch: [78][ 100/ 117] Loss 0.365334 Top1 81.980469 Top5 97.593750 +2023-10-05 21:19:07,144 - Epoch: [78][ 110/ 117] Loss 0.367041 Top1 81.995739 Top5 97.556818 +2023-10-05 21:19:07,228 - Epoch: [78][ 117/ 117] Loss 0.364809 Top1 82.042548 Top5 97.612130 +2023-10-05 21:19:07,339 - ==> Top1: 82.043 Top5: 97.612 Loss: 0.365 + +2023-10-05 21:19:07,340 - ==> Confusion: +[[ 909 1 9 1 14 2 0 3 6 71 0 0 0 4 9 2 3 3 0 0 13] + [ 2 1029 1 0 9 25 5 28 1 0 2 1 0 1 0 3 7 0 10 3 4] + [ 2 3 958 12 1 0 24 11 0 0 5 1 8 3 2 5 3 1 5 3 9] + [ 1 1 34 931 3 5 5 1 1 0 12 1 2 8 28 5 2 6 26 2 15] + [ 20 7 0 0 980 4 0 1 0 8 1 1 0 2 9 2 8 1 2 0 4] + [ 7 33 2 0 4 964 1 32 2 5 6 8 3 8 4 3 6 1 7 9 11] + [ 0 6 23 0 0 0 1123 12 0 0 6 3 2 0 1 5 0 1 0 3 6] + [ 3 13 24 0 4 48 6 1044 1 1 2 12 4 1 0 0 0 0 39 9 7] + [ 20 2 2 0 4 0 0 2 944 51 11 2 1 16 19 2 1 1 9 1 1] + [ 101 0 6 0 7 3 0 0 23 918 0 2 0 29 14 3 1 0 0 0 12] + [ 3 6 14 8 0 3 4 4 18 2 942 2 0 21 4 1 1 1 6 5 8] + [ 2 1 1 0 0 13 1 1 0 2 0 935 28 5 1 4 0 20 0 13 8] + [ 2 0 4 4 1 2 1 0 1 0 2 45 935 2 3 9 3 28 5 3 18] + [ 2 0 2 0 3 9 0 0 6 19 1 11 1 1051 2 1 2 1 0 0 8] + [ 14 3 5 13 8 0 0 0 15 4 1 1 0 1 1003 0 0 4 15 0 14] + [ 1 2 2 1 5 2 1 1 0 0 0 5 6 1 1 1056 21 11 1 11 6] + [ 0 8 3 1 10 8 1 0 1 0 0 4 0 1 2 12 1097 1 2 2 8] + [ 0 0 1 0 0 0 0 0 0 0 0 3 13 2 2 7 1 1002 1 1 5] + [ 1 5 12 11 1 1 1 27 3 1 2 3 4 0 12 0 1 0 972 1 10] + [ 0 3 5 0 1 7 6 15 0 0 0 19 6 2 0 7 11 1 1 1059 9] + [ 144 200 213 57 115 184 56 125 103 104 180 155 325 348 142 81 237 70 165 187 4714]] + +2023-10-05 21:19:07,341 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:19:07,341 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:19:07,347 - + +2023-10-05 21:19:07,347 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:19:08,452 - Epoch: [79][ 10/ 1236] Overall Loss 0.301897 Objective Loss 0.301897 LR 0.001000 Time 0.110469 +2023-10-05 21:19:08,657 - Epoch: [79][ 20/ 1236] Overall Loss 0.310134 Objective Loss 0.310134 LR 0.001000 Time 0.065428 +2023-10-05 21:19:08,858 - Epoch: [79][ 30/ 1236] Overall Loss 0.307909 Objective Loss 0.307909 LR 0.001000 Time 0.050332 +2023-10-05 21:19:09,061 - Epoch: [79][ 40/ 1236] Overall Loss 0.314236 Objective Loss 0.314236 LR 0.001000 Time 0.042800 +2023-10-05 21:19:09,262 - Epoch: [79][ 50/ 1236] Overall Loss 0.310491 Objective Loss 0.310491 LR 0.001000 Time 0.038258 +2023-10-05 21:19:09,464 - Epoch: [79][ 60/ 1236] Overall Loss 0.306772 Objective Loss 0.306772 LR 0.001000 Time 0.035245 +2023-10-05 21:19:09,666 - Epoch: [79][ 70/ 1236] Overall Loss 0.309454 Objective Loss 0.309454 LR 0.001000 Time 0.033081 +2023-10-05 21:19:09,868 - Epoch: [79][ 80/ 1236] Overall Loss 0.309410 Objective Loss 0.309410 LR 0.001000 Time 0.031467 +2023-10-05 21:19:10,069 - Epoch: [79][ 90/ 1236] Overall Loss 0.309705 Objective Loss 0.309705 LR 0.001000 Time 0.030205 +2023-10-05 21:19:10,271 - Epoch: [79][ 100/ 1236] Overall Loss 0.306314 Objective Loss 0.306314 LR 0.001000 Time 0.029201 +2023-10-05 21:19:10,472 - Epoch: [79][ 110/ 1236] Overall Loss 0.306239 Objective Loss 0.306239 LR 0.001000 Time 0.028369 +2023-10-05 21:19:10,673 - Epoch: [79][ 120/ 1236] Overall Loss 0.303701 Objective Loss 0.303701 LR 0.001000 Time 0.027678 +2023-10-05 21:19:10,873 - Epoch: [79][ 130/ 1236] Overall Loss 0.300571 Objective Loss 0.300571 LR 0.001000 Time 0.027083 +2023-10-05 21:19:11,075 - Epoch: [79][ 140/ 1236] Overall Loss 0.298722 Objective Loss 0.298722 LR 0.001000 Time 0.026591 +2023-10-05 21:19:11,278 - Epoch: [79][ 150/ 1236] Overall Loss 0.298470 Objective Loss 0.298470 LR 0.001000 Time 0.026168 +2023-10-05 21:19:11,483 - Epoch: [79][ 160/ 1236] Overall Loss 0.299818 Objective Loss 0.299818 LR 0.001000 Time 0.025810 +2023-10-05 21:19:11,689 - Epoch: [79][ 170/ 1236] Overall Loss 0.299003 Objective Loss 0.299003 LR 0.001000 Time 0.025491 +2023-10-05 21:19:11,894 - Epoch: [79][ 180/ 1236] Overall Loss 0.301028 Objective Loss 0.301028 LR 0.001000 Time 0.025216 +2023-10-05 21:19:12,105 - Epoch: [79][ 190/ 1236] Overall Loss 0.300376 Objective Loss 0.300376 LR 0.001000 Time 0.024989 +2023-10-05 21:19:12,315 - Epoch: [79][ 200/ 1236] Overall Loss 0.300638 Objective Loss 0.300638 LR 0.001000 Time 0.024781 +2023-10-05 21:19:12,524 - Epoch: [79][ 210/ 1236] Overall Loss 0.303295 Objective Loss 0.303295 LR 0.001000 Time 0.024590 +2023-10-05 21:19:12,733 - Epoch: [79][ 220/ 1236] Overall Loss 0.303301 Objective Loss 0.303301 LR 0.001000 Time 0.024416 +2023-10-05 21:19:12,943 - Epoch: [79][ 230/ 1236] Overall Loss 0.302471 Objective Loss 0.302471 LR 0.001000 Time 0.024262 +2023-10-05 21:19:13,146 - Epoch: [79][ 240/ 1236] Overall Loss 0.302506 Objective Loss 0.302506 LR 0.001000 Time 0.024098 +2023-10-05 21:19:13,347 - Epoch: [79][ 250/ 1236] Overall Loss 0.303568 Objective Loss 0.303568 LR 0.001000 Time 0.023934 +2023-10-05 21:19:13,548 - Epoch: [79][ 260/ 1236] Overall Loss 0.302780 Objective Loss 0.302780 LR 0.001000 Time 0.023786 +2023-10-05 21:19:13,751 - Epoch: [79][ 270/ 1236] Overall Loss 0.302214 Objective Loss 0.302214 LR 0.001000 Time 0.023658 +2023-10-05 21:19:13,953 - Epoch: [79][ 280/ 1236] Overall Loss 0.302695 Objective Loss 0.302695 LR 0.001000 Time 0.023530 +2023-10-05 21:19:14,154 - Epoch: [79][ 290/ 1236] Overall Loss 0.303117 Objective Loss 0.303117 LR 0.001000 Time 0.023411 +2023-10-05 21:19:14,355 - Epoch: [79][ 300/ 1236] Overall Loss 0.303076 Objective Loss 0.303076 LR 0.001000 Time 0.023301 +2023-10-05 21:19:14,556 - Epoch: [79][ 310/ 1236] Overall Loss 0.303802 Objective Loss 0.303802 LR 0.001000 Time 0.023197 +2023-10-05 21:19:14,758 - Epoch: [79][ 320/ 1236] Overall Loss 0.303471 Objective Loss 0.303471 LR 0.001000 Time 0.023100 +2023-10-05 21:19:14,959 - Epoch: [79][ 330/ 1236] Overall Loss 0.303784 Objective Loss 0.303784 LR 0.001000 Time 0.023008 +2023-10-05 21:19:15,160 - Epoch: [79][ 340/ 1236] Overall Loss 0.302690 Objective Loss 0.302690 LR 0.001000 Time 0.022923 +2023-10-05 21:19:15,361 - Epoch: [79][ 350/ 1236] Overall Loss 0.302834 Objective Loss 0.302834 LR 0.001000 Time 0.022841 +2023-10-05 21:19:15,562 - Epoch: [79][ 360/ 1236] Overall Loss 0.302867 Objective Loss 0.302867 LR 0.001000 Time 0.022765 +2023-10-05 21:19:15,767 - Epoch: [79][ 370/ 1236] Overall Loss 0.302667 Objective Loss 0.302667 LR 0.001000 Time 0.022700 +2023-10-05 21:19:15,970 - Epoch: [79][ 380/ 1236] Overall Loss 0.302996 Objective Loss 0.302996 LR 0.001000 Time 0.022638 +2023-10-05 21:19:16,173 - Epoch: [79][ 390/ 1236] Overall Loss 0.302909 Objective Loss 0.302909 LR 0.001000 Time 0.022577 +2023-10-05 21:19:16,379 - Epoch: [79][ 400/ 1236] Overall Loss 0.302930 Objective Loss 0.302930 LR 0.001000 Time 0.022526 +2023-10-05 21:19:16,582 - Epoch: [79][ 410/ 1236] Overall Loss 0.303326 Objective Loss 0.303326 LR 0.001000 Time 0.022472 +2023-10-05 21:19:16,785 - Epoch: [79][ 420/ 1236] Overall Loss 0.303632 Objective Loss 0.303632 LR 0.001000 Time 0.022419 +2023-10-05 21:19:16,988 - Epoch: [79][ 430/ 1236] Overall Loss 0.303797 Objective Loss 0.303797 LR 0.001000 Time 0.022368 +2023-10-05 21:19:17,190 - Epoch: [79][ 440/ 1236] Overall Loss 0.303224 Objective Loss 0.303224 LR 0.001000 Time 0.022318 +2023-10-05 21:19:17,393 - Epoch: [79][ 450/ 1236] Overall Loss 0.303260 Objective Loss 0.303260 LR 0.001000 Time 0.022272 +2023-10-05 21:19:17,594 - Epoch: [79][ 460/ 1236] Overall Loss 0.303152 Objective Loss 0.303152 LR 0.001000 Time 0.022225 +2023-10-05 21:19:17,797 - Epoch: [79][ 470/ 1236] Overall Loss 0.303476 Objective Loss 0.303476 LR 0.001000 Time 0.022182 +2023-10-05 21:19:17,999 - Epoch: [79][ 480/ 1236] Overall Loss 0.304053 Objective Loss 0.304053 LR 0.001000 Time 0.022141 +2023-10-05 21:19:18,202 - Epoch: [79][ 490/ 1236] Overall Loss 0.304133 Objective Loss 0.304133 LR 0.001000 Time 0.022102 +2023-10-05 21:19:18,404 - Epoch: [79][ 500/ 1236] Overall Loss 0.304851 Objective Loss 0.304851 LR 0.001000 Time 0.022064 +2023-10-05 21:19:18,607 - Epoch: [79][ 510/ 1236] Overall Loss 0.305333 Objective Loss 0.305333 LR 0.001000 Time 0.022027 +2023-10-05 21:19:18,810 - Epoch: [79][ 520/ 1236] Overall Loss 0.305571 Objective Loss 0.305571 LR 0.001000 Time 0.021994 +2023-10-05 21:19:19,013 - Epoch: [79][ 530/ 1236] Overall Loss 0.305617 Objective Loss 0.305617 LR 0.001000 Time 0.021962 +2023-10-05 21:19:19,215 - Epoch: [79][ 540/ 1236] Overall Loss 0.305678 Objective Loss 0.305678 LR 0.001000 Time 0.021928 +2023-10-05 21:19:19,417 - Epoch: [79][ 550/ 1236] Overall Loss 0.305730 Objective Loss 0.305730 LR 0.001000 Time 0.021897 +2023-10-05 21:19:19,619 - Epoch: [79][ 560/ 1236] Overall Loss 0.305517 Objective Loss 0.305517 LR 0.001000 Time 0.021866 +2023-10-05 21:19:19,822 - Epoch: [79][ 570/ 1236] Overall Loss 0.306096 Objective Loss 0.306096 LR 0.001000 Time 0.021837 +2023-10-05 21:19:20,024 - Epoch: [79][ 580/ 1236] Overall Loss 0.306261 Objective Loss 0.306261 LR 0.001000 Time 0.021808 +2023-10-05 21:19:20,227 - Epoch: [79][ 590/ 1236] Overall Loss 0.306182 Objective Loss 0.306182 LR 0.001000 Time 0.021782 +2023-10-05 21:19:20,429 - Epoch: [79][ 600/ 1236] Overall Loss 0.306691 Objective Loss 0.306691 LR 0.001000 Time 0.021755 +2023-10-05 21:19:20,632 - Epoch: [79][ 610/ 1236] Overall Loss 0.307260 Objective Loss 0.307260 LR 0.001000 Time 0.021730 +2023-10-05 21:19:20,834 - Epoch: [79][ 620/ 1236] Overall Loss 0.308036 Objective Loss 0.308036 LR 0.001000 Time 0.021705 +2023-10-05 21:19:21,037 - Epoch: [79][ 630/ 1236] Overall Loss 0.309275 Objective Loss 0.309275 LR 0.001000 Time 0.021682 +2023-10-05 21:19:21,239 - Epoch: [79][ 640/ 1236] Overall Loss 0.309488 Objective Loss 0.309488 LR 0.001000 Time 0.021659 +2023-10-05 21:19:21,442 - Epoch: [79][ 650/ 1236] Overall Loss 0.309934 Objective Loss 0.309934 LR 0.001000 Time 0.021637 +2023-10-05 21:19:21,644 - Epoch: [79][ 660/ 1236] Overall Loss 0.310336 Objective Loss 0.310336 LR 0.001000 Time 0.021615 +2023-10-05 21:19:21,847 - Epoch: [79][ 670/ 1236] Overall Loss 0.310719 Objective Loss 0.310719 LR 0.001000 Time 0.021594 +2023-10-05 21:19:22,049 - Epoch: [79][ 680/ 1236] Overall Loss 0.311401 Objective Loss 0.311401 LR 0.001000 Time 0.021574 +2023-10-05 21:19:22,252 - Epoch: [79][ 690/ 1236] Overall Loss 0.311758 Objective Loss 0.311758 LR 0.001000 Time 0.021554 +2023-10-05 21:19:22,454 - Epoch: [79][ 700/ 1236] Overall Loss 0.312037 Objective Loss 0.312037 LR 0.001000 Time 0.021535 +2023-10-05 21:19:22,656 - Epoch: [79][ 710/ 1236] Overall Loss 0.311623 Objective Loss 0.311623 LR 0.001000 Time 0.021515 +2023-10-05 21:19:22,858 - Epoch: [79][ 720/ 1236] Overall Loss 0.311750 Objective Loss 0.311750 LR 0.001000 Time 0.021497 +2023-10-05 21:19:23,062 - Epoch: [79][ 730/ 1236] Overall Loss 0.311537 Objective Loss 0.311537 LR 0.001000 Time 0.021481 +2023-10-05 21:19:23,264 - Epoch: [79][ 740/ 1236] Overall Loss 0.311065 Objective Loss 0.311065 LR 0.001000 Time 0.021464 +2023-10-05 21:19:23,467 - Epoch: [79][ 750/ 1236] Overall Loss 0.310844 Objective Loss 0.310844 LR 0.001000 Time 0.021447 +2023-10-05 21:19:23,669 - Epoch: [79][ 760/ 1236] Overall Loss 0.311192 Objective Loss 0.311192 LR 0.001000 Time 0.021431 +2023-10-05 21:19:23,872 - Epoch: [79][ 770/ 1236] Overall Loss 0.311228 Objective Loss 0.311228 LR 0.001000 Time 0.021415 +2023-10-05 21:19:24,074 - Epoch: [79][ 780/ 1236] Overall Loss 0.311545 Objective Loss 0.311545 LR 0.001000 Time 0.021399 +2023-10-05 21:19:24,277 - Epoch: [79][ 790/ 1236] Overall Loss 0.311710 Objective Loss 0.311710 LR 0.001000 Time 0.021385 +2023-10-05 21:19:24,479 - Epoch: [79][ 800/ 1236] Overall Loss 0.311894 Objective Loss 0.311894 LR 0.001000 Time 0.021370 +2023-10-05 21:19:24,685 - Epoch: [79][ 810/ 1236] Overall Loss 0.311933 Objective Loss 0.311933 LR 0.001000 Time 0.021360 +2023-10-05 21:19:24,895 - Epoch: [79][ 820/ 1236] Overall Loss 0.311736 Objective Loss 0.311736 LR 0.001000 Time 0.021356 +2023-10-05 21:19:25,110 - Epoch: [79][ 830/ 1236] Overall Loss 0.311902 Objective Loss 0.311902 LR 0.001000 Time 0.021356 +2023-10-05 21:19:25,319 - Epoch: [79][ 840/ 1236] Overall Loss 0.312467 Objective Loss 0.312467 LR 0.001000 Time 0.021351 +2023-10-05 21:19:25,533 - Epoch: [79][ 850/ 1236] Overall Loss 0.312703 Objective Loss 0.312703 LR 0.001000 Time 0.021350 +2023-10-05 21:19:25,743 - Epoch: [79][ 860/ 1236] Overall Loss 0.312610 Objective Loss 0.312610 LR 0.001000 Time 0.021346 +2023-10-05 21:19:25,957 - Epoch: [79][ 870/ 1236] Overall Loss 0.313097 Objective Loss 0.313097 LR 0.001000 Time 0.021346 +2023-10-05 21:19:26,166 - Epoch: [79][ 880/ 1236] Overall Loss 0.312963 Objective Loss 0.312963 LR 0.001000 Time 0.021341 +2023-10-05 21:19:26,380 - Epoch: [79][ 890/ 1236] Overall Loss 0.312360 Objective Loss 0.312360 LR 0.001000 Time 0.021341 +2023-10-05 21:19:26,590 - Epoch: [79][ 900/ 1236] Overall Loss 0.313011 Objective Loss 0.313011 LR 0.001000 Time 0.021336 +2023-10-05 21:19:26,803 - Epoch: [79][ 910/ 1236] Overall Loss 0.313365 Objective Loss 0.313365 LR 0.001000 Time 0.021336 +2023-10-05 21:19:27,013 - Epoch: [79][ 920/ 1236] Overall Loss 0.313442 Objective Loss 0.313442 LR 0.001000 Time 0.021332 +2023-10-05 21:19:27,226 - Epoch: [79][ 930/ 1236] Overall Loss 0.313514 Objective Loss 0.313514 LR 0.001000 Time 0.021331 +2023-10-05 21:19:27,437 - Epoch: [79][ 940/ 1236] Overall Loss 0.313217 Objective Loss 0.313217 LR 0.001000 Time 0.021328 +2023-10-05 21:19:27,652 - Epoch: [79][ 950/ 1236] Overall Loss 0.313030 Objective Loss 0.313030 LR 0.001000 Time 0.021329 +2023-10-05 21:19:27,857 - Epoch: [79][ 960/ 1236] Overall Loss 0.312805 Objective Loss 0.312805 LR 0.001000 Time 0.021320 +2023-10-05 21:19:28,065 - Epoch: [79][ 970/ 1236] Overall Loss 0.312638 Objective Loss 0.312638 LR 0.001000 Time 0.021314 +2023-10-05 21:19:28,267 - Epoch: [79][ 980/ 1236] Overall Loss 0.312761 Objective Loss 0.312761 LR 0.001000 Time 0.021303 +2023-10-05 21:19:28,471 - Epoch: [79][ 990/ 1236] Overall Loss 0.312877 Objective Loss 0.312877 LR 0.001000 Time 0.021293 +2023-10-05 21:19:28,674 - Epoch: [79][ 1000/ 1236] Overall Loss 0.312962 Objective Loss 0.312962 LR 0.001000 Time 0.021282 +2023-10-05 21:19:28,877 - Epoch: [79][ 1010/ 1236] Overall Loss 0.312795 Objective Loss 0.312795 LR 0.001000 Time 0.021272 +2023-10-05 21:19:29,080 - Epoch: [79][ 1020/ 1236] Overall Loss 0.312694 Objective Loss 0.312694 LR 0.001000 Time 0.021262 +2023-10-05 21:19:29,283 - Epoch: [79][ 1030/ 1236] Overall Loss 0.312454 Objective Loss 0.312454 LR 0.001000 Time 0.021252 +2023-10-05 21:19:29,485 - Epoch: [79][ 1040/ 1236] Overall Loss 0.312731 Objective Loss 0.312731 LR 0.001000 Time 0.021242 +2023-10-05 21:19:29,688 - Epoch: [79][ 1050/ 1236] Overall Loss 0.312442 Objective Loss 0.312442 LR 0.001000 Time 0.021232 +2023-10-05 21:19:29,890 - Epoch: [79][ 1060/ 1236] Overall Loss 0.312555 Objective Loss 0.312555 LR 0.001000 Time 0.021222 +2023-10-05 21:19:30,093 - Epoch: [79][ 1070/ 1236] Overall Loss 0.312722 Objective Loss 0.312722 LR 0.001000 Time 0.021213 +2023-10-05 21:19:30,295 - Epoch: [79][ 1080/ 1236] Overall Loss 0.312771 Objective Loss 0.312771 LR 0.001000 Time 0.021204 +2023-10-05 21:19:30,498 - Epoch: [79][ 1090/ 1236] Overall Loss 0.312787 Objective Loss 0.312787 LR 0.001000 Time 0.021195 +2023-10-05 21:19:30,700 - Epoch: [79][ 1100/ 1236] Overall Loss 0.313224 Objective Loss 0.313224 LR 0.001000 Time 0.021186 +2023-10-05 21:19:30,903 - Epoch: [79][ 1110/ 1236] Overall Loss 0.313295 Objective Loss 0.313295 LR 0.001000 Time 0.021177 +2023-10-05 21:19:31,107 - Epoch: [79][ 1120/ 1236] Overall Loss 0.313416 Objective Loss 0.313416 LR 0.001000 Time 0.021169 +2023-10-05 21:19:31,309 - Epoch: [79][ 1130/ 1236] Overall Loss 0.313639 Objective Loss 0.313639 LR 0.001000 Time 0.021161 +2023-10-05 21:19:31,512 - Epoch: [79][ 1140/ 1236] Overall Loss 0.313773 Objective Loss 0.313773 LR 0.001000 Time 0.021153 +2023-10-05 21:19:31,715 - Epoch: [79][ 1150/ 1236] Overall Loss 0.313648 Objective Loss 0.313648 LR 0.001000 Time 0.021145 +2023-10-05 21:19:31,917 - Epoch: [79][ 1160/ 1236] Overall Loss 0.313673 Objective Loss 0.313673 LR 0.001000 Time 0.021137 +2023-10-05 21:19:32,121 - Epoch: [79][ 1170/ 1236] Overall Loss 0.313551 Objective Loss 0.313551 LR 0.001000 Time 0.021130 +2023-10-05 21:19:32,323 - Epoch: [79][ 1180/ 1236] Overall Loss 0.313611 Objective Loss 0.313611 LR 0.001000 Time 0.021122 +2023-10-05 21:19:32,527 - Epoch: [79][ 1190/ 1236] Overall Loss 0.313599 Objective Loss 0.313599 LR 0.001000 Time 0.021115 +2023-10-05 21:19:32,729 - Epoch: [79][ 1200/ 1236] Overall Loss 0.313611 Objective Loss 0.313611 LR 0.001000 Time 0.021108 +2023-10-05 21:19:32,932 - Epoch: [79][ 1210/ 1236] Overall Loss 0.313578 Objective Loss 0.313578 LR 0.001000 Time 0.021101 +2023-10-05 21:19:33,135 - Epoch: [79][ 1220/ 1236] Overall Loss 0.313515 Objective Loss 0.313515 LR 0.001000 Time 0.021094 +2023-10-05 21:19:33,389 - Epoch: [79][ 1230/ 1236] Overall Loss 0.313142 Objective Loss 0.313142 LR 0.001000 Time 0.021128 +2023-10-05 21:19:33,506 - Epoch: [79][ 1236/ 1236] Overall Loss 0.313304 Objective Loss 0.313304 Top1 84.114053 Top5 97.963340 LR 0.001000 Time 0.021120 +2023-10-05 21:19:33,629 - --- validate (epoch=79)----------- +2023-10-05 21:19:33,630 - 29943 samples (256 per mini-batch) +2023-10-05 21:19:34,084 - Epoch: [79][ 10/ 117] Loss 0.339940 Top1 80.859375 Top5 97.343750 +2023-10-05 21:19:34,227 - Epoch: [79][ 20/ 117] Loss 0.326625 Top1 81.523438 Top5 97.636719 +2023-10-05 21:19:34,369 - Epoch: [79][ 30/ 117] Loss 0.343603 Top1 81.523438 Top5 97.513021 +2023-10-05 21:19:34,512 - Epoch: [79][ 40/ 117] Loss 0.347917 Top1 81.708984 Top5 97.500000 +2023-10-05 21:19:34,655 - Epoch: [79][ 50/ 117] Loss 0.343826 Top1 81.765625 Top5 97.476562 +2023-10-05 21:19:34,797 - Epoch: [79][ 60/ 117] Loss 0.349082 Top1 81.692708 Top5 97.467448 +2023-10-05 21:19:34,939 - Epoch: [79][ 70/ 117] Loss 0.354836 Top1 81.523438 Top5 97.460938 +2023-10-05 21:19:35,084 - Epoch: [79][ 80/ 117] Loss 0.354462 Top1 81.630859 Top5 97.456055 +2023-10-05 21:19:35,228 - Epoch: [79][ 90/ 117] Loss 0.355951 Top1 81.575521 Top5 97.430556 +2023-10-05 21:19:35,372 - Epoch: [79][ 100/ 117] Loss 0.354698 Top1 81.664062 Top5 97.425781 +2023-10-05 21:19:35,523 - Epoch: [79][ 110/ 117] Loss 0.355879 Top1 81.700994 Top5 97.375710 +2023-10-05 21:19:35,608 - Epoch: [79][ 117/ 117] Loss 0.356134 Top1 81.661824 Top5 97.378352 +2023-10-05 21:19:35,731 - ==> Top1: 81.662 Top5: 97.378 Loss: 0.356 + +2023-10-05 21:19:35,732 - ==> Confusion: +[[ 916 2 3 2 7 1 0 2 5 83 1 1 1 4 3 2 2 1 1 0 13] + [ 0 1042 2 0 8 25 0 32 0 0 1 0 0 1 1 5 2 1 8 1 2] + [ 1 3 949 16 2 2 33 10 0 0 2 2 7 0 2 6 1 1 6 3 10] + [ 5 1 29 938 3 4 1 2 3 1 13 0 6 4 16 6 1 9 29 1 17] + [ 23 12 1 1 966 3 0 1 0 10 0 2 0 3 8 4 10 1 1 0 4] + [ 5 43 1 1 3 961 2 34 1 3 3 9 1 17 6 3 2 0 3 6 12] + [ 0 8 30 0 0 1 1111 19 0 0 2 3 0 0 1 4 0 3 2 2 5] + [ 4 20 13 1 3 29 9 1060 2 5 3 8 1 0 0 4 0 0 45 5 6] + [ 23 1 0 0 2 4 0 3 962 50 11 2 0 7 12 2 4 1 5 0 0] + [ 99 0 5 1 1 3 1 1 22 948 0 1 0 17 3 5 2 2 0 0 8] + [ 3 5 18 7 2 3 6 4 17 2 950 1 0 8 3 4 2 2 8 0 8] + [ 0 0 1 0 0 9 2 5 0 0 0 958 12 3 0 1 1 17 0 23 3] + [ 1 0 10 6 0 2 0 1 0 1 0 65 939 1 2 4 0 20 1 2 13] + [ 0 0 1 0 1 16 1 1 20 18 12 7 1 1014 3 3 0 3 0 7 11] + [ 22 9 3 11 9 0 0 0 28 4 4 2 2 2 979 0 1 5 10 0 10] + [ 0 2 3 0 3 1 2 0 0 0 0 9 8 1 0 1055 18 14 3 6 9] + [ 0 18 2 1 10 8 0 1 4 0 1 7 1 2 2 10 1077 0 2 5 10] + [ 0 0 0 0 0 0 2 0 0 0 0 11 14 0 1 4 0 999 2 0 5] + [ 0 7 10 12 0 1 1 32 2 1 1 0 2 0 10 0 1 1 976 1 10] + [ 0 5 4 2 1 8 14 22 0 0 2 16 3 2 0 3 8 1 2 1050 9] + [ 148 304 200 62 96 212 89 158 135 121 158 162 297 339 95 70 127 84 211 235 4602]] + +2023-10-05 21:19:35,733 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:19:35,733 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:19:35,739 - + +2023-10-05 21:19:35,739 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:19:36,725 - Epoch: [80][ 10/ 1236] Overall Loss 0.298617 Objective Loss 0.298617 LR 0.001000 Time 0.098512 +2023-10-05 21:19:36,927 - Epoch: [80][ 20/ 1236] Overall Loss 0.305814 Objective Loss 0.305814 LR 0.001000 Time 0.059369 +2023-10-05 21:19:37,129 - Epoch: [80][ 30/ 1236] Overall Loss 0.307857 Objective Loss 0.307857 LR 0.001000 Time 0.046284 +2023-10-05 21:19:37,330 - Epoch: [80][ 40/ 1236] Overall Loss 0.305822 Objective Loss 0.305822 LR 0.001000 Time 0.039746 +2023-10-05 21:19:37,532 - Epoch: [80][ 50/ 1236] Overall Loss 0.295829 Objective Loss 0.295829 LR 0.001000 Time 0.035829 +2023-10-05 21:19:37,734 - Epoch: [80][ 60/ 1236] Overall Loss 0.291578 Objective Loss 0.291578 LR 0.001000 Time 0.033219 +2023-10-05 21:19:37,936 - Epoch: [80][ 70/ 1236] Overall Loss 0.296546 Objective Loss 0.296546 LR 0.001000 Time 0.031352 +2023-10-05 21:19:38,138 - Epoch: [80][ 80/ 1236] Overall Loss 0.301807 Objective Loss 0.301807 LR 0.001000 Time 0.029953 +2023-10-05 21:19:38,340 - Epoch: [80][ 90/ 1236] Overall Loss 0.298645 Objective Loss 0.298645 LR 0.001000 Time 0.028862 +2023-10-05 21:19:38,542 - Epoch: [80][ 100/ 1236] Overall Loss 0.302674 Objective Loss 0.302674 LR 0.001000 Time 0.027991 +2023-10-05 21:19:38,743 - Epoch: [80][ 110/ 1236] Overall Loss 0.304282 Objective Loss 0.304282 LR 0.001000 Time 0.027269 +2023-10-05 21:19:38,943 - Epoch: [80][ 120/ 1236] Overall Loss 0.303710 Objective Loss 0.303710 LR 0.001000 Time 0.026663 +2023-10-05 21:19:39,148 - Epoch: [80][ 130/ 1236] Overall Loss 0.302786 Objective Loss 0.302786 LR 0.001000 Time 0.026189 +2023-10-05 21:19:39,353 - Epoch: [80][ 140/ 1236] Overall Loss 0.303662 Objective Loss 0.303662 LR 0.001000 Time 0.025777 +2023-10-05 21:19:39,557 - Epoch: [80][ 150/ 1236] Overall Loss 0.303173 Objective Loss 0.303173 LR 0.001000 Time 0.025419 +2023-10-05 21:19:39,760 - Epoch: [80][ 160/ 1236] Overall Loss 0.300589 Objective Loss 0.300589 LR 0.001000 Time 0.025094 +2023-10-05 21:19:39,964 - Epoch: [80][ 170/ 1236] Overall Loss 0.300874 Objective Loss 0.300874 LR 0.001000 Time 0.024817 +2023-10-05 21:19:40,166 - Epoch: [80][ 180/ 1236] Overall Loss 0.300627 Objective Loss 0.300627 LR 0.001000 Time 0.024559 +2023-10-05 21:19:40,370 - Epoch: [80][ 190/ 1236] Overall Loss 0.299834 Objective Loss 0.299834 LR 0.001000 Time 0.024338 +2023-10-05 21:19:40,573 - Epoch: [80][ 200/ 1236] Overall Loss 0.299811 Objective Loss 0.299811 LR 0.001000 Time 0.024132 +2023-10-05 21:19:40,777 - Epoch: [80][ 210/ 1236] Overall Loss 0.300318 Objective Loss 0.300318 LR 0.001000 Time 0.023956 +2023-10-05 21:19:40,980 - Epoch: [80][ 220/ 1236] Overall Loss 0.299605 Objective Loss 0.299605 LR 0.001000 Time 0.023786 +2023-10-05 21:19:41,180 - Epoch: [80][ 230/ 1236] Overall Loss 0.300375 Objective Loss 0.300375 LR 0.001000 Time 0.023622 +2023-10-05 21:19:41,381 - Epoch: [80][ 240/ 1236] Overall Loss 0.299457 Objective Loss 0.299457 LR 0.001000 Time 0.023472 +2023-10-05 21:19:41,582 - Epoch: [80][ 250/ 1236] Overall Loss 0.300142 Objective Loss 0.300142 LR 0.001000 Time 0.023335 +2023-10-05 21:19:41,783 - Epoch: [80][ 260/ 1236] Overall Loss 0.299681 Objective Loss 0.299681 LR 0.001000 Time 0.023210 +2023-10-05 21:19:41,984 - Epoch: [80][ 270/ 1236] Overall Loss 0.300894 Objective Loss 0.300894 LR 0.001000 Time 0.023092 +2023-10-05 21:19:42,184 - Epoch: [80][ 280/ 1236] Overall Loss 0.300096 Objective Loss 0.300096 LR 0.001000 Time 0.022983 +2023-10-05 21:19:42,385 - Epoch: [80][ 290/ 1236] Overall Loss 0.299261 Objective Loss 0.299261 LR 0.001000 Time 0.022881 +2023-10-05 21:19:42,587 - Epoch: [80][ 300/ 1236] Overall Loss 0.301202 Objective Loss 0.301202 LR 0.001000 Time 0.022789 +2023-10-05 21:19:42,789 - Epoch: [80][ 310/ 1236] Overall Loss 0.302108 Objective Loss 0.302108 LR 0.001000 Time 0.022706 +2023-10-05 21:19:42,992 - Epoch: [80][ 320/ 1236] Overall Loss 0.301609 Objective Loss 0.301609 LR 0.001000 Time 0.022628 +2023-10-05 21:19:43,195 - Epoch: [80][ 330/ 1236] Overall Loss 0.302146 Objective Loss 0.302146 LR 0.001000 Time 0.022559 +2023-10-05 21:19:43,401 - Epoch: [80][ 340/ 1236] Overall Loss 0.302424 Objective Loss 0.302424 LR 0.001000 Time 0.022499 +2023-10-05 21:19:43,605 - Epoch: [80][ 350/ 1236] Overall Loss 0.302694 Objective Loss 0.302694 LR 0.001000 Time 0.022437 +2023-10-05 21:19:43,808 - Epoch: [80][ 360/ 1236] Overall Loss 0.302456 Objective Loss 0.302456 LR 0.001000 Time 0.022377 +2023-10-05 21:19:44,012 - Epoch: [80][ 370/ 1236] Overall Loss 0.302873 Objective Loss 0.302873 LR 0.001000 Time 0.022324 +2023-10-05 21:19:44,216 - Epoch: [80][ 380/ 1236] Overall Loss 0.303256 Objective Loss 0.303256 LR 0.001000 Time 0.022272 +2023-10-05 21:19:44,421 - Epoch: [80][ 390/ 1236] Overall Loss 0.303566 Objective Loss 0.303566 LR 0.001000 Time 0.022226 +2023-10-05 21:19:44,625 - Epoch: [80][ 400/ 1236] Overall Loss 0.303684 Objective Loss 0.303684 LR 0.001000 Time 0.022179 +2023-10-05 21:19:44,830 - Epoch: [80][ 410/ 1236] Overall Loss 0.304423 Objective Loss 0.304423 LR 0.001000 Time 0.022137 +2023-10-05 21:19:45,033 - Epoch: [80][ 420/ 1236] Overall Loss 0.304012 Objective Loss 0.304012 LR 0.001000 Time 0.022092 +2023-10-05 21:19:45,238 - Epoch: [80][ 430/ 1236] Overall Loss 0.303114 Objective Loss 0.303114 LR 0.001000 Time 0.022054 +2023-10-05 21:19:45,441 - Epoch: [80][ 440/ 1236] Overall Loss 0.303334 Objective Loss 0.303334 LR 0.001000 Time 0.022015 +2023-10-05 21:19:45,649 - Epoch: [80][ 450/ 1236] Overall Loss 0.302914 Objective Loss 0.302914 LR 0.001000 Time 0.021986 +2023-10-05 21:19:45,851 - Epoch: [80][ 460/ 1236] Overall Loss 0.302510 Objective Loss 0.302510 LR 0.001000 Time 0.021948 +2023-10-05 21:19:46,054 - Epoch: [80][ 470/ 1236] Overall Loss 0.302409 Objective Loss 0.302409 LR 0.001000 Time 0.021911 +2023-10-05 21:19:46,256 - Epoch: [80][ 480/ 1236] Overall Loss 0.302129 Objective Loss 0.302129 LR 0.001000 Time 0.021876 +2023-10-05 21:19:46,465 - Epoch: [80][ 490/ 1236] Overall Loss 0.301669 Objective Loss 0.301669 LR 0.001000 Time 0.021855 +2023-10-05 21:19:46,677 - Epoch: [80][ 500/ 1236] Overall Loss 0.301499 Objective Loss 0.301499 LR 0.001000 Time 0.021841 +2023-10-05 21:19:46,885 - Epoch: [80][ 510/ 1236] Overall Loss 0.301464 Objective Loss 0.301464 LR 0.001000 Time 0.021820 +2023-10-05 21:19:47,097 - Epoch: [80][ 520/ 1236] Overall Loss 0.301442 Objective Loss 0.301442 LR 0.001000 Time 0.021806 +2023-10-05 21:19:47,303 - Epoch: [80][ 530/ 1236] Overall Loss 0.301658 Objective Loss 0.301658 LR 0.001000 Time 0.021784 +2023-10-05 21:19:47,515 - Epoch: [80][ 540/ 1236] Overall Loss 0.301038 Objective Loss 0.301038 LR 0.001000 Time 0.021773 +2023-10-05 21:19:47,722 - Epoch: [80][ 550/ 1236] Overall Loss 0.300952 Objective Loss 0.300952 LR 0.001000 Time 0.021752 +2023-10-05 21:19:47,934 - Epoch: [80][ 560/ 1236] Overall Loss 0.301520 Objective Loss 0.301520 LR 0.001000 Time 0.021742 +2023-10-05 21:19:48,142 - Epoch: [80][ 570/ 1236] Overall Loss 0.301886 Objective Loss 0.301886 LR 0.001000 Time 0.021724 +2023-10-05 21:19:48,350 - Epoch: [80][ 580/ 1236] Overall Loss 0.302326 Objective Loss 0.302326 LR 0.001000 Time 0.021707 +2023-10-05 21:19:48,550 - Epoch: [80][ 590/ 1236] Overall Loss 0.302795 Objective Loss 0.302795 LR 0.001000 Time 0.021678 +2023-10-05 21:19:48,752 - Epoch: [80][ 600/ 1236] Overall Loss 0.303587 Objective Loss 0.303587 LR 0.001000 Time 0.021653 +2023-10-05 21:19:48,952 - Epoch: [80][ 610/ 1236] Overall Loss 0.303983 Objective Loss 0.303983 LR 0.001000 Time 0.021626 +2023-10-05 21:19:49,163 - Epoch: [80][ 620/ 1236] Overall Loss 0.304412 Objective Loss 0.304412 LR 0.001000 Time 0.021617 +2023-10-05 21:19:49,370 - Epoch: [80][ 630/ 1236] Overall Loss 0.304617 Objective Loss 0.304617 LR 0.001000 Time 0.021601 +2023-10-05 21:19:49,582 - Epoch: [80][ 640/ 1236] Overall Loss 0.304301 Objective Loss 0.304301 LR 0.001000 Time 0.021594 +2023-10-05 21:19:49,788 - Epoch: [80][ 650/ 1236] Overall Loss 0.304913 Objective Loss 0.304913 LR 0.001000 Time 0.021579 +2023-10-05 21:19:49,999 - Epoch: [80][ 660/ 1236] Overall Loss 0.304891 Objective Loss 0.304891 LR 0.001000 Time 0.021571 +2023-10-05 21:19:50,205 - Epoch: [80][ 670/ 1236] Overall Loss 0.305362 Objective Loss 0.305362 LR 0.001000 Time 0.021556 +2023-10-05 21:19:50,415 - Epoch: [80][ 680/ 1236] Overall Loss 0.305968 Objective Loss 0.305968 LR 0.001000 Time 0.021547 +2023-10-05 21:19:50,615 - Epoch: [80][ 690/ 1236] Overall Loss 0.305851 Objective Loss 0.305851 LR 0.001000 Time 0.021524 +2023-10-05 21:19:50,817 - Epoch: [80][ 700/ 1236] Overall Loss 0.306075 Objective Loss 0.306075 LR 0.001000 Time 0.021504 +2023-10-05 21:19:51,017 - Epoch: [80][ 710/ 1236] Overall Loss 0.306461 Objective Loss 0.306461 LR 0.001000 Time 0.021483 +2023-10-05 21:19:51,218 - Epoch: [80][ 720/ 1236] Overall Loss 0.307134 Objective Loss 0.307134 LR 0.001000 Time 0.021463 +2023-10-05 21:19:51,418 - Epoch: [80][ 730/ 1236] Overall Loss 0.307188 Objective Loss 0.307188 LR 0.001000 Time 0.021443 +2023-10-05 21:19:51,619 - Epoch: [80][ 740/ 1236] Overall Loss 0.306739 Objective Loss 0.306739 LR 0.001000 Time 0.021424 +2023-10-05 21:19:51,819 - Epoch: [80][ 750/ 1236] Overall Loss 0.306860 Objective Loss 0.306860 LR 0.001000 Time 0.021405 +2023-10-05 21:19:52,020 - Epoch: [80][ 760/ 1236] Overall Loss 0.306869 Objective Loss 0.306869 LR 0.001000 Time 0.021387 +2023-10-05 21:19:52,223 - Epoch: [80][ 770/ 1236] Overall Loss 0.306553 Objective Loss 0.306553 LR 0.001000 Time 0.021372 +2023-10-05 21:19:52,435 - Epoch: [80][ 780/ 1236] Overall Loss 0.307100 Objective Loss 0.307100 LR 0.001000 Time 0.021369 +2023-10-05 21:19:52,641 - Epoch: [80][ 790/ 1236] Overall Loss 0.307125 Objective Loss 0.307125 LR 0.001000 Time 0.021360 +2023-10-05 21:19:52,852 - Epoch: [80][ 800/ 1236] Overall Loss 0.306846 Objective Loss 0.306846 LR 0.001000 Time 0.021356 +2023-10-05 21:19:53,059 - Epoch: [80][ 810/ 1236] Overall Loss 0.306850 Objective Loss 0.306850 LR 0.001000 Time 0.021348 +2023-10-05 21:19:53,275 - Epoch: [80][ 820/ 1236] Overall Loss 0.307137 Objective Loss 0.307137 LR 0.001000 Time 0.021350 +2023-10-05 21:19:53,482 - Epoch: [80][ 830/ 1236] Overall Loss 0.307293 Objective Loss 0.307293 LR 0.001000 Time 0.021341 +2023-10-05 21:19:53,692 - Epoch: [80][ 840/ 1236] Overall Loss 0.306842 Objective Loss 0.306842 LR 0.001000 Time 0.021338 +2023-10-05 21:19:53,899 - Epoch: [80][ 850/ 1236] Overall Loss 0.307290 Objective Loss 0.307290 LR 0.001000 Time 0.021330 +2023-10-05 21:19:54,110 - Epoch: [80][ 860/ 1236] Overall Loss 0.307662 Objective Loss 0.307662 LR 0.001000 Time 0.021327 +2023-10-05 21:19:54,317 - Epoch: [80][ 870/ 1236] Overall Loss 0.307958 Objective Loss 0.307958 LR 0.001000 Time 0.021318 +2023-10-05 21:19:54,528 - Epoch: [80][ 880/ 1236] Overall Loss 0.308361 Objective Loss 0.308361 LR 0.001000 Time 0.021316 +2023-10-05 21:19:54,735 - Epoch: [80][ 890/ 1236] Overall Loss 0.308570 Objective Loss 0.308570 LR 0.001000 Time 0.021309 +2023-10-05 21:19:54,946 - Epoch: [80][ 900/ 1236] Overall Loss 0.309148 Objective Loss 0.309148 LR 0.001000 Time 0.021306 +2023-10-05 21:19:55,145 - Epoch: [80][ 910/ 1236] Overall Loss 0.309025 Objective Loss 0.309025 LR 0.001000 Time 0.021291 +2023-10-05 21:19:55,346 - Epoch: [80][ 920/ 1236] Overall Loss 0.308991 Objective Loss 0.308991 LR 0.001000 Time 0.021277 +2023-10-05 21:19:55,550 - Epoch: [80][ 930/ 1236] Overall Loss 0.309052 Objective Loss 0.309052 LR 0.001000 Time 0.021267 +2023-10-05 21:19:55,752 - Epoch: [80][ 940/ 1236] Overall Loss 0.309026 Objective Loss 0.309026 LR 0.001000 Time 0.021255 +2023-10-05 21:19:55,952 - Epoch: [80][ 950/ 1236] Overall Loss 0.309113 Objective Loss 0.309113 LR 0.001000 Time 0.021242 +2023-10-05 21:19:56,153 - Epoch: [80][ 960/ 1236] Overall Loss 0.309284 Objective Loss 0.309284 LR 0.001000 Time 0.021229 +2023-10-05 21:19:56,353 - Epoch: [80][ 970/ 1236] Overall Loss 0.309487 Objective Loss 0.309487 LR 0.001000 Time 0.021216 +2023-10-05 21:19:56,553 - Epoch: [80][ 980/ 1236] Overall Loss 0.309586 Objective Loss 0.309586 LR 0.001000 Time 0.021203 +2023-10-05 21:19:56,752 - Epoch: [80][ 990/ 1236] Overall Loss 0.309889 Objective Loss 0.309889 LR 0.001000 Time 0.021190 +2023-10-05 21:19:56,954 - Epoch: [80][ 1000/ 1236] Overall Loss 0.309840 Objective Loss 0.309840 LR 0.001000 Time 0.021179 +2023-10-05 21:19:57,153 - Epoch: [80][ 1010/ 1236] Overall Loss 0.310162 Objective Loss 0.310162 LR 0.001000 Time 0.021167 +2023-10-05 21:19:57,354 - Epoch: [80][ 1020/ 1236] Overall Loss 0.310352 Objective Loss 0.310352 LR 0.001000 Time 0.021156 +2023-10-05 21:19:57,554 - Epoch: [80][ 1030/ 1236] Overall Loss 0.310417 Objective Loss 0.310417 LR 0.001000 Time 0.021144 +2023-10-05 21:19:57,754 - Epoch: [80][ 1040/ 1236] Overall Loss 0.310628 Objective Loss 0.310628 LR 0.001000 Time 0.021133 +2023-10-05 21:19:57,954 - Epoch: [80][ 1050/ 1236] Overall Loss 0.310627 Objective Loss 0.310627 LR 0.001000 Time 0.021122 +2023-10-05 21:19:58,155 - Epoch: [80][ 1060/ 1236] Overall Loss 0.310214 Objective Loss 0.310214 LR 0.001000 Time 0.021112 +2023-10-05 21:19:58,355 - Epoch: [80][ 1070/ 1236] Overall Loss 0.309959 Objective Loss 0.309959 LR 0.001000 Time 0.021101 +2023-10-05 21:19:58,556 - Epoch: [80][ 1080/ 1236] Overall Loss 0.310135 Objective Loss 0.310135 LR 0.001000 Time 0.021092 +2023-10-05 21:19:58,756 - Epoch: [80][ 1090/ 1236] Overall Loss 0.310535 Objective Loss 0.310535 LR 0.001000 Time 0.021081 +2023-10-05 21:19:58,957 - Epoch: [80][ 1100/ 1236] Overall Loss 0.310823 Objective Loss 0.310823 LR 0.001000 Time 0.021072 +2023-10-05 21:19:59,157 - Epoch: [80][ 1110/ 1236] Overall Loss 0.310722 Objective Loss 0.310722 LR 0.001000 Time 0.021062 +2023-10-05 21:19:59,358 - Epoch: [80][ 1120/ 1236] Overall Loss 0.310553 Objective Loss 0.310553 LR 0.001000 Time 0.021053 +2023-10-05 21:19:59,558 - Epoch: [80][ 1130/ 1236] Overall Loss 0.310581 Objective Loss 0.310581 LR 0.001000 Time 0.021043 +2023-10-05 21:19:59,764 - Epoch: [80][ 1140/ 1236] Overall Loss 0.310512 Objective Loss 0.310512 LR 0.001000 Time 0.021039 +2023-10-05 21:19:59,964 - Epoch: [80][ 1150/ 1236] Overall Loss 0.310703 Objective Loss 0.310703 LR 0.001000 Time 0.021030 +2023-10-05 21:20:00,168 - Epoch: [80][ 1160/ 1236] Overall Loss 0.310854 Objective Loss 0.310854 LR 0.001000 Time 0.021024 +2023-10-05 21:20:00,368 - Epoch: [80][ 1170/ 1236] Overall Loss 0.311195 Objective Loss 0.311195 LR 0.001000 Time 0.021015 +2023-10-05 21:20:00,570 - Epoch: [80][ 1180/ 1236] Overall Loss 0.311581 Objective Loss 0.311581 LR 0.001000 Time 0.021008 +2023-10-05 21:20:00,770 - Epoch: [80][ 1190/ 1236] Overall Loss 0.311731 Objective Loss 0.311731 LR 0.001000 Time 0.020999 +2023-10-05 21:20:00,972 - Epoch: [80][ 1200/ 1236] Overall Loss 0.311876 Objective Loss 0.311876 LR 0.001000 Time 0.020992 +2023-10-05 21:20:01,172 - Epoch: [80][ 1210/ 1236] Overall Loss 0.312159 Objective Loss 0.312159 LR 0.001000 Time 0.020983 +2023-10-05 21:20:01,374 - Epoch: [80][ 1220/ 1236] Overall Loss 0.312302 Objective Loss 0.312302 LR 0.001000 Time 0.020977 +2023-10-05 21:20:01,627 - Epoch: [80][ 1230/ 1236] Overall Loss 0.312716 Objective Loss 0.312716 LR 0.001000 Time 0.021012 +2023-10-05 21:20:01,744 - Epoch: [80][ 1236/ 1236] Overall Loss 0.312973 Objective Loss 0.312973 Top1 81.466395 Top5 96.741344 LR 0.001000 Time 0.021004 +2023-10-05 21:20:01,884 - --- validate (epoch=80)----------- +2023-10-05 21:20:01,885 - 29943 samples (256 per mini-batch) +2023-10-05 21:20:02,341 - Epoch: [80][ 10/ 117] Loss 0.379502 Top1 81.289062 Top5 97.617188 +2023-10-05 21:20:02,491 - Epoch: [80][ 20/ 117] Loss 0.377109 Top1 81.406250 Top5 97.597656 +2023-10-05 21:20:02,640 - Epoch: [80][ 30/ 117] Loss 0.374732 Top1 81.601562 Top5 97.526042 +2023-10-05 21:20:02,790 - Epoch: [80][ 40/ 117] Loss 0.376415 Top1 81.513672 Top5 97.500000 +2023-10-05 21:20:02,939 - Epoch: [80][ 50/ 117] Loss 0.380310 Top1 81.328125 Top5 97.390625 +2023-10-05 21:20:03,088 - Epoch: [80][ 60/ 117] Loss 0.379498 Top1 81.210938 Top5 97.291667 +2023-10-05 21:20:03,235 - Epoch: [80][ 70/ 117] Loss 0.380173 Top1 81.093750 Top5 97.243304 +2023-10-05 21:20:03,384 - Epoch: [80][ 80/ 117] Loss 0.380946 Top1 81.069336 Top5 97.246094 +2023-10-05 21:20:03,531 - Epoch: [80][ 90/ 117] Loss 0.375665 Top1 81.080729 Top5 97.296007 +2023-10-05 21:20:03,681 - Epoch: [80][ 100/ 117] Loss 0.375780 Top1 81.125000 Top5 97.300781 +2023-10-05 21:20:03,836 - Epoch: [80][ 110/ 117] Loss 0.374825 Top1 81.058239 Top5 97.325994 +2023-10-05 21:20:03,920 - Epoch: [80][ 117/ 117] Loss 0.375208 Top1 81.057342 Top5 97.304879 +2023-10-05 21:20:04,032 - ==> Top1: 81.057 Top5: 97.305 Loss: 0.375 + +2023-10-05 21:20:04,032 - ==> Confusion: +[[ 921 4 4 1 5 4 0 1 5 68 0 2 0 5 8 1 4 0 0 1 16] + [ 1 1046 1 1 5 14 1 22 2 0 7 0 1 0 0 4 6 1 16 0 3] + [ 5 0 931 15 2 0 27 15 0 0 8 2 7 1 7 3 4 1 6 5 17] + [ 2 4 24 908 0 6 3 2 2 0 27 0 6 8 48 4 3 4 23 2 13] + [ 26 11 1 0 959 7 0 2 1 6 1 1 0 1 13 5 9 1 0 0 6] + [ 4 67 0 1 3 932 2 26 2 2 4 13 1 25 5 4 4 2 2 6 11] + [ 0 4 32 0 0 0 1118 10 0 0 4 4 2 0 1 5 0 1 0 2 8] + [ 7 25 13 0 3 36 9 1037 2 1 10 10 5 2 1 2 0 2 40 6 7] + [ 18 0 1 0 1 1 0 0 954 44 15 1 5 15 24 0 6 2 1 0 1] + [ 133 0 3 0 11 3 2 0 41 862 1 1 1 27 13 5 2 4 1 1 8] + [ 5 4 8 2 0 4 5 6 18 2 958 3 0 11 6 5 2 1 4 1 8] + [ 2 2 1 0 0 15 0 3 0 0 0 955 23 7 0 3 2 11 0 9 2] + [ 2 0 2 3 0 2 0 5 0 1 1 51 953 8 0 7 3 15 2 6 7] + [ 2 0 2 1 1 14 0 0 17 15 6 5 3 1030 5 1 1 1 0 1 14] + [ 20 4 2 4 5 0 0 0 26 1 3 1 0 2 996 0 2 3 11 1 20] + [ 0 2 1 1 7 1 2 0 0 0 0 11 11 1 1 1055 13 13 0 7 8] + [ 0 13 4 0 8 5 0 0 1 0 0 10 1 2 3 10 1088 1 1 4 10] + [ 0 0 2 1 0 0 0 0 0 0 0 6 30 0 1 7 1 986 1 0 3] + [ 0 6 9 12 1 1 2 32 7 0 2 2 4 0 12 0 1 0 961 1 15] + [ 0 2 3 0 1 11 8 11 0 0 3 15 6 3 0 7 13 2 5 1049 13] + [ 149 269 150 80 83 152 61 106 122 79 257 146 438 331 184 72 172 82 170 230 4572]] + +2023-10-05 21:20:04,034 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:20:04,034 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:20:04,039 - + +2023-10-05 21:20:04,040 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:20:05,024 - Epoch: [81][ 10/ 1236] Overall Loss 0.351964 Objective Loss 0.351964 LR 0.001000 Time 0.098410 +2023-10-05 21:20:05,226 - Epoch: [81][ 20/ 1236] Overall Loss 0.339372 Objective Loss 0.339372 LR 0.001000 Time 0.059279 +2023-10-05 21:20:05,425 - Epoch: [81][ 30/ 1236] Overall Loss 0.333569 Objective Loss 0.333569 LR 0.001000 Time 0.046157 +2023-10-05 21:20:05,629 - Epoch: [81][ 40/ 1236] Overall Loss 0.338720 Objective Loss 0.338720 LR 0.001000 Time 0.039689 +2023-10-05 21:20:05,829 - Epoch: [81][ 50/ 1236] Overall Loss 0.340073 Objective Loss 0.340073 LR 0.001000 Time 0.035758 +2023-10-05 21:20:06,032 - Epoch: [81][ 60/ 1236] Overall Loss 0.340150 Objective Loss 0.340150 LR 0.001000 Time 0.033176 +2023-10-05 21:20:06,233 - Epoch: [81][ 70/ 1236] Overall Loss 0.345427 Objective Loss 0.345427 LR 0.001000 Time 0.031296 +2023-10-05 21:20:06,436 - Epoch: [81][ 80/ 1236] Overall Loss 0.343010 Objective Loss 0.343010 LR 0.001000 Time 0.029918 +2023-10-05 21:20:06,636 - Epoch: [81][ 90/ 1236] Overall Loss 0.342701 Objective Loss 0.342701 LR 0.001000 Time 0.028818 +2023-10-05 21:20:06,839 - Epoch: [81][ 100/ 1236] Overall Loss 0.343126 Objective Loss 0.343126 LR 0.001000 Time 0.027963 +2023-10-05 21:20:07,038 - Epoch: [81][ 110/ 1236] Overall Loss 0.343377 Objective Loss 0.343377 LR 0.001000 Time 0.027222 +2023-10-05 21:20:07,239 - Epoch: [81][ 120/ 1236] Overall Loss 0.343014 Objective Loss 0.343014 LR 0.001000 Time 0.026625 +2023-10-05 21:20:07,445 - Epoch: [81][ 130/ 1236] Overall Loss 0.341777 Objective Loss 0.341777 LR 0.001000 Time 0.026160 +2023-10-05 21:20:07,657 - Epoch: [81][ 140/ 1236] Overall Loss 0.341420 Objective Loss 0.341420 LR 0.001000 Time 0.025806 +2023-10-05 21:20:07,863 - Epoch: [81][ 150/ 1236] Overall Loss 0.343996 Objective Loss 0.343996 LR 0.001000 Time 0.025454 +2023-10-05 21:20:08,072 - Epoch: [81][ 160/ 1236] Overall Loss 0.345715 Objective Loss 0.345715 LR 0.001000 Time 0.025166 +2023-10-05 21:20:08,277 - Epoch: [81][ 170/ 1236] Overall Loss 0.345900 Objective Loss 0.345900 LR 0.001000 Time 0.024891 +2023-10-05 21:20:08,486 - Epoch: [81][ 180/ 1236] Overall Loss 0.345673 Objective Loss 0.345673 LR 0.001000 Time 0.024666 +2023-10-05 21:20:08,691 - Epoch: [81][ 190/ 1236] Overall Loss 0.346252 Objective Loss 0.346252 LR 0.001000 Time 0.024446 +2023-10-05 21:20:08,900 - Epoch: [81][ 200/ 1236] Overall Loss 0.345170 Objective Loss 0.345170 LR 0.001000 Time 0.024267 +2023-10-05 21:20:09,105 - Epoch: [81][ 210/ 1236] Overall Loss 0.344337 Objective Loss 0.344337 LR 0.001000 Time 0.024087 +2023-10-05 21:20:09,315 - Epoch: [81][ 220/ 1236] Overall Loss 0.341605 Objective Loss 0.341605 LR 0.001000 Time 0.023943 +2023-10-05 21:20:09,525 - Epoch: [81][ 230/ 1236] Overall Loss 0.342038 Objective Loss 0.342038 LR 0.001000 Time 0.023816 +2023-10-05 21:20:09,736 - Epoch: [81][ 240/ 1236] Overall Loss 0.341392 Objective Loss 0.341392 LR 0.001000 Time 0.023699 +2023-10-05 21:20:09,941 - Epoch: [81][ 250/ 1236] Overall Loss 0.340861 Objective Loss 0.340861 LR 0.001000 Time 0.023572 +2023-10-05 21:20:10,142 - Epoch: [81][ 260/ 1236] Overall Loss 0.340699 Objective Loss 0.340699 LR 0.001000 Time 0.023434 +2023-10-05 21:20:10,340 - Epoch: [81][ 270/ 1236] Overall Loss 0.340748 Objective Loss 0.340748 LR 0.001000 Time 0.023300 +2023-10-05 21:20:10,541 - Epoch: [81][ 280/ 1236] Overall Loss 0.341346 Objective Loss 0.341346 LR 0.001000 Time 0.023183 +2023-10-05 21:20:10,739 - Epoch: [81][ 290/ 1236] Overall Loss 0.340193 Objective Loss 0.340193 LR 0.001000 Time 0.023067 +2023-10-05 21:20:10,939 - Epoch: [81][ 300/ 1236] Overall Loss 0.338840 Objective Loss 0.338840 LR 0.001000 Time 0.022963 +2023-10-05 21:20:11,138 - Epoch: [81][ 310/ 1236] Overall Loss 0.337882 Objective Loss 0.337882 LR 0.001000 Time 0.022863 +2023-10-05 21:20:11,338 - Epoch: [81][ 320/ 1236] Overall Loss 0.337286 Objective Loss 0.337286 LR 0.001000 Time 0.022772 +2023-10-05 21:20:11,536 - Epoch: [81][ 330/ 1236] Overall Loss 0.336686 Objective Loss 0.336686 LR 0.001000 Time 0.022682 +2023-10-05 21:20:11,736 - Epoch: [81][ 340/ 1236] Overall Loss 0.335452 Objective Loss 0.335452 LR 0.001000 Time 0.022600 +2023-10-05 21:20:11,945 - Epoch: [81][ 350/ 1236] Overall Loss 0.336027 Objective Loss 0.336027 LR 0.001000 Time 0.022550 +2023-10-05 21:20:12,154 - Epoch: [81][ 360/ 1236] Overall Loss 0.336273 Objective Loss 0.336273 LR 0.001000 Time 0.022505 +2023-10-05 21:20:12,361 - Epoch: [81][ 370/ 1236] Overall Loss 0.336012 Objective Loss 0.336012 LR 0.001000 Time 0.022455 +2023-10-05 21:20:12,574 - Epoch: [81][ 380/ 1236] Overall Loss 0.336566 Objective Loss 0.336566 LR 0.001000 Time 0.022423 +2023-10-05 21:20:12,777 - Epoch: [81][ 390/ 1236] Overall Loss 0.334686 Objective Loss 0.334686 LR 0.001000 Time 0.022369 +2023-10-05 21:20:12,984 - Epoch: [81][ 400/ 1236] Overall Loss 0.335993 Objective Loss 0.335993 LR 0.001000 Time 0.022326 +2023-10-05 21:20:13,190 - Epoch: [81][ 410/ 1236] Overall Loss 0.336818 Objective Loss 0.336818 LR 0.001000 Time 0.022281 +2023-10-05 21:20:13,394 - Epoch: [81][ 420/ 1236] Overall Loss 0.337084 Objective Loss 0.337084 LR 0.001000 Time 0.022238 +2023-10-05 21:20:13,600 - Epoch: [81][ 430/ 1236] Overall Loss 0.336567 Objective Loss 0.336567 LR 0.001000 Time 0.022198 +2023-10-05 21:20:13,805 - Epoch: [81][ 440/ 1236] Overall Loss 0.336146 Objective Loss 0.336146 LR 0.001000 Time 0.022159 +2023-10-05 21:20:14,010 - Epoch: [81][ 450/ 1236] Overall Loss 0.336427 Objective Loss 0.336427 LR 0.001000 Time 0.022122 +2023-10-05 21:20:14,215 - Epoch: [81][ 460/ 1236] Overall Loss 0.335839 Objective Loss 0.335839 LR 0.001000 Time 0.022086 +2023-10-05 21:20:14,421 - Epoch: [81][ 470/ 1236] Overall Loss 0.335360 Objective Loss 0.335360 LR 0.001000 Time 0.022052 +2023-10-05 21:20:14,627 - Epoch: [81][ 480/ 1236] Overall Loss 0.335561 Objective Loss 0.335561 LR 0.001000 Time 0.022022 +2023-10-05 21:20:14,834 - Epoch: [81][ 490/ 1236] Overall Loss 0.336068 Objective Loss 0.336068 LR 0.001000 Time 0.021994 +2023-10-05 21:20:15,042 - Epoch: [81][ 500/ 1236] Overall Loss 0.335997 Objective Loss 0.335997 LR 0.001000 Time 0.021970 +2023-10-05 21:20:15,250 - Epoch: [81][ 510/ 1236] Overall Loss 0.335605 Objective Loss 0.335605 LR 0.001000 Time 0.021946 +2023-10-05 21:20:15,457 - Epoch: [81][ 520/ 1236] Overall Loss 0.335819 Objective Loss 0.335819 LR 0.001000 Time 0.021921 +2023-10-05 21:20:15,666 - Epoch: [81][ 530/ 1236] Overall Loss 0.335426 Objective Loss 0.335426 LR 0.001000 Time 0.021900 +2023-10-05 21:20:15,874 - Epoch: [81][ 540/ 1236] Overall Loss 0.336046 Objective Loss 0.336046 LR 0.001000 Time 0.021881 +2023-10-05 21:20:16,082 - Epoch: [81][ 550/ 1236] Overall Loss 0.335746 Objective Loss 0.335746 LR 0.001000 Time 0.021860 +2023-10-05 21:20:16,291 - Epoch: [81][ 560/ 1236] Overall Loss 0.335522 Objective Loss 0.335522 LR 0.001000 Time 0.021841 +2023-10-05 21:20:16,499 - Epoch: [81][ 570/ 1236] Overall Loss 0.335494 Objective Loss 0.335494 LR 0.001000 Time 0.021823 +2023-10-05 21:20:16,707 - Epoch: [81][ 580/ 1236] Overall Loss 0.335290 Objective Loss 0.335290 LR 0.001000 Time 0.021805 +2023-10-05 21:20:16,916 - Epoch: [81][ 590/ 1236] Overall Loss 0.335545 Objective Loss 0.335545 LR 0.001000 Time 0.021789 +2023-10-05 21:20:17,125 - Epoch: [81][ 600/ 1236] Overall Loss 0.335963 Objective Loss 0.335963 LR 0.001000 Time 0.021772 +2023-10-05 21:20:17,341 - Epoch: [81][ 610/ 1236] Overall Loss 0.335855 Objective Loss 0.335855 LR 0.001000 Time 0.021769 +2023-10-05 21:20:17,550 - Epoch: [81][ 620/ 1236] Overall Loss 0.335940 Objective Loss 0.335940 LR 0.001000 Time 0.021755 +2023-10-05 21:20:17,756 - Epoch: [81][ 630/ 1236] Overall Loss 0.335791 Objective Loss 0.335791 LR 0.001000 Time 0.021737 +2023-10-05 21:20:17,965 - Epoch: [81][ 640/ 1236] Overall Loss 0.335490 Objective Loss 0.335490 LR 0.001000 Time 0.021723 +2023-10-05 21:20:18,172 - Epoch: [81][ 650/ 1236] Overall Loss 0.335254 Objective Loss 0.335254 LR 0.001000 Time 0.021707 +2023-10-05 21:20:18,381 - Epoch: [81][ 660/ 1236] Overall Loss 0.335384 Objective Loss 0.335384 LR 0.001000 Time 0.021694 +2023-10-05 21:20:18,588 - Epoch: [81][ 670/ 1236] Overall Loss 0.334761 Objective Loss 0.334761 LR 0.001000 Time 0.021679 +2023-10-05 21:20:18,797 - Epoch: [81][ 680/ 1236] Overall Loss 0.334681 Objective Loss 0.334681 LR 0.001000 Time 0.021667 +2023-10-05 21:20:19,005 - Epoch: [81][ 690/ 1236] Overall Loss 0.334714 Objective Loss 0.334714 LR 0.001000 Time 0.021653 +2023-10-05 21:20:19,214 - Epoch: [81][ 700/ 1236] Overall Loss 0.334263 Objective Loss 0.334263 LR 0.001000 Time 0.021642 +2023-10-05 21:20:19,422 - Epoch: [81][ 710/ 1236] Overall Loss 0.333956 Objective Loss 0.333956 LR 0.001000 Time 0.021629 +2023-10-05 21:20:19,630 - Epoch: [81][ 720/ 1236] Overall Loss 0.334305 Objective Loss 0.334305 LR 0.001000 Time 0.021618 +2023-10-05 21:20:19,838 - Epoch: [81][ 730/ 1236] Overall Loss 0.334745 Objective Loss 0.334745 LR 0.001000 Time 0.021606 +2023-10-05 21:20:20,048 - Epoch: [81][ 740/ 1236] Overall Loss 0.334178 Objective Loss 0.334178 LR 0.001000 Time 0.021597 +2023-10-05 21:20:20,256 - Epoch: [81][ 750/ 1236] Overall Loss 0.333902 Objective Loss 0.333902 LR 0.001000 Time 0.021586 +2023-10-05 21:20:20,465 - Epoch: [81][ 760/ 1236] Overall Loss 0.333526 Objective Loss 0.333526 LR 0.001000 Time 0.021577 +2023-10-05 21:20:20,680 - Epoch: [81][ 770/ 1236] Overall Loss 0.333332 Objective Loss 0.333332 LR 0.001000 Time 0.021576 +2023-10-05 21:20:20,893 - Epoch: [81][ 780/ 1236] Overall Loss 0.333143 Objective Loss 0.333143 LR 0.001000 Time 0.021571 +2023-10-05 21:20:21,106 - Epoch: [81][ 790/ 1236] Overall Loss 0.332529 Objective Loss 0.332529 LR 0.001000 Time 0.021567 +2023-10-05 21:20:21,315 - Epoch: [81][ 800/ 1236] Overall Loss 0.332221 Objective Loss 0.332221 LR 0.001000 Time 0.021558 +2023-10-05 21:20:21,522 - Epoch: [81][ 810/ 1236] Overall Loss 0.331487 Objective Loss 0.331487 LR 0.001000 Time 0.021548 +2023-10-05 21:20:21,732 - Epoch: [81][ 820/ 1236] Overall Loss 0.331038 Objective Loss 0.331038 LR 0.001000 Time 0.021540 +2023-10-05 21:20:21,939 - Epoch: [81][ 830/ 1236] Overall Loss 0.330965 Objective Loss 0.330965 LR 0.001000 Time 0.021529 +2023-10-05 21:20:22,148 - Epoch: [81][ 840/ 1236] Overall Loss 0.330970 Objective Loss 0.330970 LR 0.001000 Time 0.021521 +2023-10-05 21:20:22,355 - Epoch: [81][ 850/ 1236] Overall Loss 0.330349 Objective Loss 0.330349 LR 0.001000 Time 0.021512 +2023-10-05 21:20:22,565 - Epoch: [81][ 860/ 1236] Overall Loss 0.329896 Objective Loss 0.329896 LR 0.001000 Time 0.021505 +2023-10-05 21:20:22,778 - Epoch: [81][ 870/ 1236] Overall Loss 0.329858 Objective Loss 0.329858 LR 0.001000 Time 0.021503 +2023-10-05 21:20:22,991 - Epoch: [81][ 880/ 1236] Overall Loss 0.329514 Objective Loss 0.329514 LR 0.001000 Time 0.021500 +2023-10-05 21:20:23,198 - Epoch: [81][ 890/ 1236] Overall Loss 0.329465 Objective Loss 0.329465 LR 0.001000 Time 0.021490 +2023-10-05 21:20:23,407 - Epoch: [81][ 900/ 1236] Overall Loss 0.329548 Objective Loss 0.329548 LR 0.001000 Time 0.021483 +2023-10-05 21:20:23,614 - Epoch: [81][ 910/ 1236] Overall Loss 0.329521 Objective Loss 0.329521 LR 0.001000 Time 0.021474 +2023-10-05 21:20:23,823 - Epoch: [81][ 920/ 1236] Overall Loss 0.329139 Objective Loss 0.329139 LR 0.001000 Time 0.021467 +2023-10-05 21:20:24,030 - Epoch: [81][ 930/ 1236] Overall Loss 0.328975 Objective Loss 0.328975 LR 0.001000 Time 0.021459 +2023-10-05 21:20:24,239 - Epoch: [81][ 940/ 1236] Overall Loss 0.328551 Objective Loss 0.328551 LR 0.001000 Time 0.021453 +2023-10-05 21:20:24,446 - Epoch: [81][ 950/ 1236] Overall Loss 0.328355 Objective Loss 0.328355 LR 0.001000 Time 0.021445 +2023-10-05 21:20:24,655 - Epoch: [81][ 960/ 1236] Overall Loss 0.328403 Objective Loss 0.328403 LR 0.001000 Time 0.021438 +2023-10-05 21:20:24,862 - Epoch: [81][ 970/ 1236] Overall Loss 0.328376 Objective Loss 0.328376 LR 0.001000 Time 0.021430 +2023-10-05 21:20:25,079 - Epoch: [81][ 980/ 1236] Overall Loss 0.328193 Objective Loss 0.328193 LR 0.001000 Time 0.021433 +2023-10-05 21:20:25,298 - Epoch: [81][ 990/ 1236] Overall Loss 0.327851 Objective Loss 0.327851 LR 0.001000 Time 0.021437 +2023-10-05 21:20:25,516 - Epoch: [81][ 1000/ 1236] Overall Loss 0.327957 Objective Loss 0.327957 LR 0.001000 Time 0.021440 +2023-10-05 21:20:25,731 - Epoch: [81][ 1010/ 1236] Overall Loss 0.328087 Objective Loss 0.328087 LR 0.001000 Time 0.021441 +2023-10-05 21:20:25,946 - Epoch: [81][ 1020/ 1236] Overall Loss 0.327985 Objective Loss 0.327985 LR 0.001000 Time 0.021441 +2023-10-05 21:20:26,161 - Epoch: [81][ 1030/ 1236] Overall Loss 0.327563 Objective Loss 0.327563 LR 0.001000 Time 0.021441 +2023-10-05 21:20:26,375 - Epoch: [81][ 1040/ 1236] Overall Loss 0.327373 Objective Loss 0.327373 LR 0.001000 Time 0.021440 +2023-10-05 21:20:26,592 - Epoch: [81][ 1050/ 1236] Overall Loss 0.327031 Objective Loss 0.327031 LR 0.001000 Time 0.021442 +2023-10-05 21:20:26,807 - Epoch: [81][ 1060/ 1236] Overall Loss 0.326636 Objective Loss 0.326636 LR 0.001000 Time 0.021442 +2023-10-05 21:20:27,022 - Epoch: [81][ 1070/ 1236] Overall Loss 0.326452 Objective Loss 0.326452 LR 0.001000 Time 0.021443 +2023-10-05 21:20:27,235 - Epoch: [81][ 1080/ 1236] Overall Loss 0.326219 Objective Loss 0.326219 LR 0.001000 Time 0.021441 +2023-10-05 21:20:27,449 - Epoch: [81][ 1090/ 1236] Overall Loss 0.325912 Objective Loss 0.325912 LR 0.001000 Time 0.021441 +2023-10-05 21:20:27,656 - Epoch: [81][ 1100/ 1236] Overall Loss 0.325970 Objective Loss 0.325970 LR 0.001000 Time 0.021434 +2023-10-05 21:20:27,861 - Epoch: [81][ 1110/ 1236] Overall Loss 0.325824 Objective Loss 0.325824 LR 0.001000 Time 0.021425 +2023-10-05 21:20:28,070 - Epoch: [81][ 1120/ 1236] Overall Loss 0.325811 Objective Loss 0.325811 LR 0.001000 Time 0.021420 +2023-10-05 21:20:28,276 - Epoch: [81][ 1130/ 1236] Overall Loss 0.325996 Objective Loss 0.325996 LR 0.001000 Time 0.021413 +2023-10-05 21:20:28,487 - Epoch: [81][ 1140/ 1236] Overall Loss 0.325651 Objective Loss 0.325651 LR 0.001000 Time 0.021409 +2023-10-05 21:20:28,693 - Epoch: [81][ 1150/ 1236] Overall Loss 0.325769 Objective Loss 0.325769 LR 0.001000 Time 0.021402 +2023-10-05 21:20:28,902 - Epoch: [81][ 1160/ 1236] Overall Loss 0.325908 Objective Loss 0.325908 LR 0.001000 Time 0.021397 +2023-10-05 21:20:29,109 - Epoch: [81][ 1170/ 1236] Overall Loss 0.325835 Objective Loss 0.325835 LR 0.001000 Time 0.021391 +2023-10-05 21:20:29,318 - Epoch: [81][ 1180/ 1236] Overall Loss 0.325474 Objective Loss 0.325474 LR 0.001000 Time 0.021386 +2023-10-05 21:20:29,526 - Epoch: [81][ 1190/ 1236] Overall Loss 0.325686 Objective Loss 0.325686 LR 0.001000 Time 0.021381 +2023-10-05 21:20:29,735 - Epoch: [81][ 1200/ 1236] Overall Loss 0.325740 Objective Loss 0.325740 LR 0.001000 Time 0.021377 +2023-10-05 21:20:29,942 - Epoch: [81][ 1210/ 1236] Overall Loss 0.325865 Objective Loss 0.325865 LR 0.001000 Time 0.021371 +2023-10-05 21:20:30,152 - Epoch: [81][ 1220/ 1236] Overall Loss 0.325961 Objective Loss 0.325961 LR 0.001000 Time 0.021367 +2023-10-05 21:20:30,418 - Epoch: [81][ 1230/ 1236] Overall Loss 0.325815 Objective Loss 0.325815 LR 0.001000 Time 0.021410 +2023-10-05 21:20:30,538 - Epoch: [81][ 1236/ 1236] Overall Loss 0.325769 Objective Loss 0.325769 Top1 82.892057 Top5 98.370672 LR 0.001000 Time 0.021403 +2023-10-05 21:20:30,655 - --- validate (epoch=81)----------- +2023-10-05 21:20:30,656 - 29943 samples (256 per mini-batch) +2023-10-05 21:20:31,110 - Epoch: [81][ 10/ 117] Loss 0.324333 Top1 83.593750 Top5 97.890625 +2023-10-05 21:20:31,256 - Epoch: [81][ 20/ 117] Loss 0.340050 Top1 82.988281 Top5 97.617188 +2023-10-05 21:20:31,402 - Epoch: [81][ 30/ 117] Loss 0.352852 Top1 82.799479 Top5 97.460938 +2023-10-05 21:20:31,549 - Epoch: [81][ 40/ 117] Loss 0.364273 Top1 82.265625 Top5 97.490234 +2023-10-05 21:20:31,695 - Epoch: [81][ 50/ 117] Loss 0.364495 Top1 82.085938 Top5 97.484375 +2023-10-05 21:20:31,841 - Epoch: [81][ 60/ 117] Loss 0.361082 Top1 82.213542 Top5 97.558594 +2023-10-05 21:20:31,992 - Epoch: [81][ 70/ 117] Loss 0.364874 Top1 82.209821 Top5 97.539062 +2023-10-05 21:20:32,153 - Epoch: [81][ 80/ 117] Loss 0.369776 Top1 82.011719 Top5 97.490234 +2023-10-05 21:20:32,310 - Epoch: [81][ 90/ 117] Loss 0.370098 Top1 81.970486 Top5 97.430556 +2023-10-05 21:20:32,458 - Epoch: [81][ 100/ 117] Loss 0.370371 Top1 81.996094 Top5 97.441406 +2023-10-05 21:20:32,619 - Epoch: [81][ 110/ 117] Loss 0.369579 Top1 81.992188 Top5 97.500000 +2023-10-05 21:20:32,706 - Epoch: [81][ 117/ 117] Loss 0.369858 Top1 82.112681 Top5 97.498581 +2023-10-05 21:20:32,820 - ==> Top1: 82.113 Top5: 97.499 Loss: 0.370 + +2023-10-05 21:20:32,821 - ==> Confusion: +[[ 918 4 14 4 6 4 0 2 4 61 1 0 0 2 4 5 2 1 1 0 17] + [ 2 1026 2 1 4 29 2 30 1 0 4 3 1 1 0 2 0 1 9 1 12] + [ 2 2 957 15 1 2 29 8 0 0 4 2 6 1 0 5 1 1 4 6 10] + [ 2 1 29 966 3 7 5 3 4 0 2 0 5 3 19 5 1 6 12 2 14] + [ 32 17 3 0 952 8 2 1 0 8 1 2 1 1 3 4 7 1 1 0 6] + [ 6 41 0 2 3 986 4 21 3 1 2 3 3 10 1 0 5 0 3 9 13] + [ 0 2 38 0 0 1 1115 9 0 0 3 1 0 0 1 5 0 0 2 8 6] + [ 2 18 16 0 5 29 8 1077 0 3 4 6 4 1 0 1 2 1 31 3 7] + [ 18 3 1 1 1 7 2 1 924 55 14 2 4 19 10 2 4 3 13 0 5] + [ 128 2 0 2 4 10 1 3 22 905 0 0 0 18 3 6 2 2 0 2 9] + [ 4 3 16 7 2 1 5 10 8 3 948 2 0 15 5 1 0 1 10 1 11] + [ 1 0 4 0 0 18 0 3 1 0 1 922 28 7 0 1 1 20 0 24 4] + [ 3 1 9 8 0 5 0 4 0 1 1 35 947 1 0 7 0 26 4 2 14] + [ 6 0 3 2 2 19 0 0 13 15 10 4 1 1013 4 2 2 1 0 11 11] + [ 14 1 7 40 3 0 0 1 22 8 0 1 2 4 947 0 3 4 24 0 20] + [ 0 1 5 2 3 1 1 1 0 0 0 8 8 2 1 1062 14 11 3 4 7] + [ 0 15 3 1 9 12 0 0 1 0 0 8 0 1 3 10 1065 0 0 13 20] + [ 0 1 1 4 0 0 1 1 0 1 0 3 21 1 3 6 0 986 1 1 7] + [ 2 4 13 18 0 0 1 39 1 0 5 1 2 0 8 0 4 1 955 1 13] + [ 0 3 5 0 1 7 9 16 1 0 0 12 4 3 0 7 4 2 3 1068 7] + [ 128 175 225 94 86 195 78 164 75 125 142 111 330 282 106 85 112 94 182 268 4848]] + +2023-10-05 21:20:32,822 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:20:32,822 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:20:32,828 - + +2023-10-05 21:20:32,828 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:20:33,942 - Epoch: [82][ 10/ 1236] Overall Loss 0.327483 Objective Loss 0.327483 LR 0.001000 Time 0.111350 +2023-10-05 21:20:34,146 - Epoch: [82][ 20/ 1236] Overall Loss 0.322949 Objective Loss 0.322949 LR 0.001000 Time 0.065838 +2023-10-05 21:20:34,355 - Epoch: [82][ 30/ 1236] Overall Loss 0.316271 Objective Loss 0.316271 LR 0.001000 Time 0.050828 +2023-10-05 21:20:34,563 - Epoch: [82][ 40/ 1236] Overall Loss 0.309172 Objective Loss 0.309172 LR 0.001000 Time 0.043329 +2023-10-05 21:20:34,773 - Epoch: [82][ 50/ 1236] Overall Loss 0.312254 Objective Loss 0.312254 LR 0.001000 Time 0.038841 +2023-10-05 21:20:34,976 - Epoch: [82][ 60/ 1236] Overall Loss 0.308426 Objective Loss 0.308426 LR 0.001000 Time 0.035754 +2023-10-05 21:20:35,181 - Epoch: [82][ 70/ 1236] Overall Loss 0.307015 Objective Loss 0.307015 LR 0.001000 Time 0.033565 +2023-10-05 21:20:35,385 - Epoch: [82][ 80/ 1236] Overall Loss 0.309196 Objective Loss 0.309196 LR 0.001000 Time 0.031923 +2023-10-05 21:20:35,590 - Epoch: [82][ 90/ 1236] Overall Loss 0.314811 Objective Loss 0.314811 LR 0.001000 Time 0.030644 +2023-10-05 21:20:35,793 - Epoch: [82][ 100/ 1236] Overall Loss 0.320140 Objective Loss 0.320140 LR 0.001000 Time 0.029608 +2023-10-05 21:20:35,998 - Epoch: [82][ 110/ 1236] Overall Loss 0.320689 Objective Loss 0.320689 LR 0.001000 Time 0.028772 +2023-10-05 21:20:36,201 - Epoch: [82][ 120/ 1236] Overall Loss 0.322201 Objective Loss 0.322201 LR 0.001000 Time 0.028065 +2023-10-05 21:20:36,403 - Epoch: [82][ 130/ 1236] Overall Loss 0.319473 Objective Loss 0.319473 LR 0.001000 Time 0.027460 +2023-10-05 21:20:36,606 - Epoch: [82][ 140/ 1236] Overall Loss 0.317975 Objective Loss 0.317975 LR 0.001000 Time 0.026949 +2023-10-05 21:20:36,810 - Epoch: [82][ 150/ 1236] Overall Loss 0.319164 Objective Loss 0.319164 LR 0.001000 Time 0.026505 +2023-10-05 21:20:37,014 - Epoch: [82][ 160/ 1236] Overall Loss 0.319202 Objective Loss 0.319202 LR 0.001000 Time 0.026121 +2023-10-05 21:20:37,216 - Epoch: [82][ 170/ 1236] Overall Loss 0.320112 Objective Loss 0.320112 LR 0.001000 Time 0.025772 +2023-10-05 21:20:37,431 - Epoch: [82][ 180/ 1236] Overall Loss 0.322433 Objective Loss 0.322433 LR 0.001000 Time 0.025534 +2023-10-05 21:20:37,642 - Epoch: [82][ 190/ 1236] Overall Loss 0.322653 Objective Loss 0.322653 LR 0.001000 Time 0.025297 +2023-10-05 21:20:37,851 - Epoch: [82][ 200/ 1236] Overall Loss 0.324939 Objective Loss 0.324939 LR 0.001000 Time 0.025077 +2023-10-05 21:20:38,054 - Epoch: [82][ 210/ 1236] Overall Loss 0.324920 Objective Loss 0.324920 LR 0.001000 Time 0.024847 +2023-10-05 21:20:38,257 - Epoch: [82][ 220/ 1236] Overall Loss 0.325417 Objective Loss 0.325417 LR 0.001000 Time 0.024641 +2023-10-05 21:20:38,460 - Epoch: [82][ 230/ 1236] Overall Loss 0.325194 Objective Loss 0.325194 LR 0.001000 Time 0.024451 +2023-10-05 21:20:38,663 - Epoch: [82][ 240/ 1236] Overall Loss 0.326935 Objective Loss 0.326935 LR 0.001000 Time 0.024276 +2023-10-05 21:20:38,866 - Epoch: [82][ 250/ 1236] Overall Loss 0.327235 Objective Loss 0.327235 LR 0.001000 Time 0.024113 +2023-10-05 21:20:39,068 - Epoch: [82][ 260/ 1236] Overall Loss 0.326972 Objective Loss 0.326972 LR 0.001000 Time 0.023963 +2023-10-05 21:20:39,279 - Epoch: [82][ 270/ 1236] Overall Loss 0.327559 Objective Loss 0.327559 LR 0.001000 Time 0.023855 +2023-10-05 21:20:39,482 - Epoch: [82][ 280/ 1236] Overall Loss 0.327973 Objective Loss 0.327973 LR 0.001000 Time 0.023728 +2023-10-05 21:20:39,685 - Epoch: [82][ 290/ 1236] Overall Loss 0.327847 Objective Loss 0.327847 LR 0.001000 Time 0.023609 +2023-10-05 21:20:39,889 - Epoch: [82][ 300/ 1236] Overall Loss 0.328154 Objective Loss 0.328154 LR 0.001000 Time 0.023500 +2023-10-05 21:20:40,092 - Epoch: [82][ 310/ 1236] Overall Loss 0.329073 Objective Loss 0.329073 LR 0.001000 Time 0.023394 +2023-10-05 21:20:40,295 - Epoch: [82][ 320/ 1236] Overall Loss 0.329523 Objective Loss 0.329523 LR 0.001000 Time 0.023298 +2023-10-05 21:20:40,499 - Epoch: [82][ 330/ 1236] Overall Loss 0.329859 Objective Loss 0.329859 LR 0.001000 Time 0.023209 +2023-10-05 21:20:40,704 - Epoch: [82][ 340/ 1236] Overall Loss 0.329042 Objective Loss 0.329042 LR 0.001000 Time 0.023127 +2023-10-05 21:20:40,907 - Epoch: [82][ 350/ 1236] Overall Loss 0.329025 Objective Loss 0.329025 LR 0.001000 Time 0.023045 +2023-10-05 21:20:41,110 - Epoch: [82][ 360/ 1236] Overall Loss 0.329608 Objective Loss 0.329608 LR 0.001000 Time 0.022970 +2023-10-05 21:20:41,316 - Epoch: [82][ 370/ 1236] Overall Loss 0.329129 Objective Loss 0.329129 LR 0.001000 Time 0.022903 +2023-10-05 21:20:41,522 - Epoch: [82][ 380/ 1236] Overall Loss 0.329950 Objective Loss 0.329950 LR 0.001000 Time 0.022843 +2023-10-05 21:20:41,727 - Epoch: [82][ 390/ 1236] Overall Loss 0.329429 Objective Loss 0.329429 LR 0.001000 Time 0.022781 +2023-10-05 21:20:41,931 - Epoch: [82][ 400/ 1236] Overall Loss 0.329418 Objective Loss 0.329418 LR 0.001000 Time 0.022720 +2023-10-05 21:20:42,135 - Epoch: [82][ 410/ 1236] Overall Loss 0.329975 Objective Loss 0.329975 LR 0.001000 Time 0.022664 +2023-10-05 21:20:42,339 - Epoch: [82][ 420/ 1236] Overall Loss 0.330236 Objective Loss 0.330236 LR 0.001000 Time 0.022610 +2023-10-05 21:20:42,543 - Epoch: [82][ 430/ 1236] Overall Loss 0.330206 Objective Loss 0.330206 LR 0.001000 Time 0.022557 +2023-10-05 21:20:42,746 - Epoch: [82][ 440/ 1236] Overall Loss 0.330535 Objective Loss 0.330535 LR 0.001000 Time 0.022506 +2023-10-05 21:20:42,949 - Epoch: [82][ 450/ 1236] Overall Loss 0.330988 Objective Loss 0.330988 LR 0.001000 Time 0.022455 +2023-10-05 21:20:43,152 - Epoch: [82][ 460/ 1236] Overall Loss 0.331427 Objective Loss 0.331427 LR 0.001000 Time 0.022408 +2023-10-05 21:20:43,356 - Epoch: [82][ 470/ 1236] Overall Loss 0.332184 Objective Loss 0.332184 LR 0.001000 Time 0.022364 +2023-10-05 21:20:43,566 - Epoch: [82][ 480/ 1236] Overall Loss 0.331878 Objective Loss 0.331878 LR 0.001000 Time 0.022334 +2023-10-05 21:20:43,769 - Epoch: [82][ 490/ 1236] Overall Loss 0.331849 Objective Loss 0.331849 LR 0.001000 Time 0.022292 +2023-10-05 21:20:43,972 - Epoch: [82][ 500/ 1236] Overall Loss 0.331842 Objective Loss 0.331842 LR 0.001000 Time 0.022253 +2023-10-05 21:20:44,175 - Epoch: [82][ 510/ 1236] Overall Loss 0.331612 Objective Loss 0.331612 LR 0.001000 Time 0.022213 +2023-10-05 21:20:44,379 - Epoch: [82][ 520/ 1236] Overall Loss 0.332535 Objective Loss 0.332535 LR 0.001000 Time 0.022177 +2023-10-05 21:20:44,581 - Epoch: [82][ 530/ 1236] Overall Loss 0.332803 Objective Loss 0.332803 LR 0.001000 Time 0.022141 +2023-10-05 21:20:44,785 - Epoch: [82][ 540/ 1236] Overall Loss 0.332514 Objective Loss 0.332514 LR 0.001000 Time 0.022107 +2023-10-05 21:20:44,988 - Epoch: [82][ 550/ 1236] Overall Loss 0.332757 Objective Loss 0.332757 LR 0.001000 Time 0.022073 +2023-10-05 21:20:45,192 - Epoch: [82][ 560/ 1236] Overall Loss 0.332574 Objective Loss 0.332574 LR 0.001000 Time 0.022042 +2023-10-05 21:20:45,402 - Epoch: [82][ 570/ 1236] Overall Loss 0.332632 Objective Loss 0.332632 LR 0.001000 Time 0.022023 +2023-10-05 21:20:45,607 - Epoch: [82][ 580/ 1236] Overall Loss 0.332969 Objective Loss 0.332969 LR 0.001000 Time 0.021997 +2023-10-05 21:20:45,810 - Epoch: [82][ 590/ 1236] Overall Loss 0.332659 Objective Loss 0.332659 LR 0.001000 Time 0.021967 +2023-10-05 21:20:46,014 - Epoch: [82][ 600/ 1236] Overall Loss 0.332898 Objective Loss 0.332898 LR 0.001000 Time 0.021940 +2023-10-05 21:20:46,217 - Epoch: [82][ 610/ 1236] Overall Loss 0.332937 Objective Loss 0.332937 LR 0.001000 Time 0.021913 +2023-10-05 21:20:46,421 - Epoch: [82][ 620/ 1236] Overall Loss 0.332440 Objective Loss 0.332440 LR 0.001000 Time 0.021888 +2023-10-05 21:20:46,623 - Epoch: [82][ 630/ 1236] Overall Loss 0.332655 Objective Loss 0.332655 LR 0.001000 Time 0.021862 +2023-10-05 21:20:46,827 - Epoch: [82][ 640/ 1236] Overall Loss 0.333411 Objective Loss 0.333411 LR 0.001000 Time 0.021837 +2023-10-05 21:20:47,030 - Epoch: [82][ 650/ 1236] Overall Loss 0.332977 Objective Loss 0.332977 LR 0.001000 Time 0.021814 +2023-10-05 21:20:47,234 - Epoch: [82][ 660/ 1236] Overall Loss 0.333501 Objective Loss 0.333501 LR 0.001000 Time 0.021792 +2023-10-05 21:20:47,445 - Epoch: [82][ 670/ 1236] Overall Loss 0.333817 Objective Loss 0.333817 LR 0.001000 Time 0.021780 +2023-10-05 21:20:47,649 - Epoch: [82][ 680/ 1236] Overall Loss 0.334043 Objective Loss 0.334043 LR 0.001000 Time 0.021760 +2023-10-05 21:20:47,857 - Epoch: [82][ 690/ 1236] Overall Loss 0.333935 Objective Loss 0.333935 LR 0.001000 Time 0.021745 +2023-10-05 21:20:48,071 - Epoch: [82][ 700/ 1236] Overall Loss 0.333906 Objective Loss 0.333906 LR 0.001000 Time 0.021740 +2023-10-05 21:20:48,281 - Epoch: [82][ 710/ 1236] Overall Loss 0.333449 Objective Loss 0.333449 LR 0.001000 Time 0.021730 +2023-10-05 21:20:48,496 - Epoch: [82][ 720/ 1236] Overall Loss 0.333525 Objective Loss 0.333525 LR 0.001000 Time 0.021725 +2023-10-05 21:20:48,706 - Epoch: [82][ 730/ 1236] Overall Loss 0.333916 Objective Loss 0.333916 LR 0.001000 Time 0.021716 +2023-10-05 21:20:48,921 - Epoch: [82][ 740/ 1236] Overall Loss 0.333482 Objective Loss 0.333482 LR 0.001000 Time 0.021712 +2023-10-05 21:20:49,137 - Epoch: [82][ 750/ 1236] Overall Loss 0.333268 Objective Loss 0.333268 LR 0.001000 Time 0.021710 +2023-10-05 21:20:49,338 - Epoch: [82][ 760/ 1236] Overall Loss 0.332982 Objective Loss 0.332982 LR 0.001000 Time 0.021688 +2023-10-05 21:20:49,540 - Epoch: [82][ 770/ 1236] Overall Loss 0.332288 Objective Loss 0.332288 LR 0.001000 Time 0.021669 +2023-10-05 21:20:49,741 - Epoch: [82][ 780/ 1236] Overall Loss 0.331937 Objective Loss 0.331937 LR 0.001000 Time 0.021648 +2023-10-05 21:20:49,943 - Epoch: [82][ 790/ 1236] Overall Loss 0.331915 Objective Loss 0.331915 LR 0.001000 Time 0.021630 +2023-10-05 21:20:50,145 - Epoch: [82][ 800/ 1236] Overall Loss 0.332174 Objective Loss 0.332174 LR 0.001000 Time 0.021611 +2023-10-05 21:20:50,347 - Epoch: [82][ 810/ 1236] Overall Loss 0.332250 Objective Loss 0.332250 LR 0.001000 Time 0.021593 +2023-10-05 21:20:50,549 - Epoch: [82][ 820/ 1236] Overall Loss 0.332459 Objective Loss 0.332459 LR 0.001000 Time 0.021575 +2023-10-05 21:20:50,754 - Epoch: [82][ 830/ 1236] Overall Loss 0.332146 Objective Loss 0.332146 LR 0.001000 Time 0.021562 +2023-10-05 21:20:50,957 - Epoch: [82][ 840/ 1236] Overall Loss 0.332395 Objective Loss 0.332395 LR 0.001000 Time 0.021547 +2023-10-05 21:20:51,160 - Epoch: [82][ 850/ 1236] Overall Loss 0.332400 Objective Loss 0.332400 LR 0.001000 Time 0.021532 +2023-10-05 21:20:51,362 - Epoch: [82][ 860/ 1236] Overall Loss 0.332387 Objective Loss 0.332387 LR 0.001000 Time 0.021516 +2023-10-05 21:20:51,565 - Epoch: [82][ 870/ 1236] Overall Loss 0.332619 Objective Loss 0.332619 LR 0.001000 Time 0.021501 +2023-10-05 21:20:51,766 - Epoch: [82][ 880/ 1236] Overall Loss 0.332082 Objective Loss 0.332082 LR 0.001000 Time 0.021485 +2023-10-05 21:20:51,968 - Epoch: [82][ 890/ 1236] Overall Loss 0.332130 Objective Loss 0.332130 LR 0.001000 Time 0.021471 +2023-10-05 21:20:52,171 - Epoch: [82][ 900/ 1236] Overall Loss 0.331575 Objective Loss 0.331575 LR 0.001000 Time 0.021457 +2023-10-05 21:20:52,379 - Epoch: [82][ 910/ 1236] Overall Loss 0.331466 Objective Loss 0.331466 LR 0.001000 Time 0.021449 +2023-10-05 21:20:52,587 - Epoch: [82][ 920/ 1236] Overall Loss 0.331150 Objective Loss 0.331150 LR 0.001000 Time 0.021442 +2023-10-05 21:20:52,802 - Epoch: [82][ 930/ 1236] Overall Loss 0.330808 Objective Loss 0.330808 LR 0.001000 Time 0.021442 +2023-10-05 21:20:53,012 - Epoch: [82][ 940/ 1236] Overall Loss 0.330700 Objective Loss 0.330700 LR 0.001000 Time 0.021437 +2023-10-05 21:20:53,216 - Epoch: [82][ 950/ 1236] Overall Loss 0.330552 Objective Loss 0.330552 LR 0.001000 Time 0.021426 +2023-10-05 21:20:53,424 - Epoch: [82][ 960/ 1236] Overall Loss 0.330700 Objective Loss 0.330700 LR 0.001000 Time 0.021419 +2023-10-05 21:20:53,628 - Epoch: [82][ 970/ 1236] Overall Loss 0.330496 Objective Loss 0.330496 LR 0.001000 Time 0.021408 +2023-10-05 21:20:53,835 - Epoch: [82][ 980/ 1236] Overall Loss 0.330461 Objective Loss 0.330461 LR 0.001000 Time 0.021400 +2023-10-05 21:20:54,038 - Epoch: [82][ 990/ 1236] Overall Loss 0.330667 Objective Loss 0.330667 LR 0.001000 Time 0.021389 +2023-10-05 21:20:54,241 - Epoch: [82][ 1000/ 1236] Overall Loss 0.330714 Objective Loss 0.330714 LR 0.001000 Time 0.021378 +2023-10-05 21:20:54,444 - Epoch: [82][ 1010/ 1236] Overall Loss 0.330827 Objective Loss 0.330827 LR 0.001000 Time 0.021367 +2023-10-05 21:20:54,648 - Epoch: [82][ 1020/ 1236] Overall Loss 0.331144 Objective Loss 0.331144 LR 0.001000 Time 0.021357 +2023-10-05 21:20:54,851 - Epoch: [82][ 1030/ 1236] Overall Loss 0.331049 Objective Loss 0.331049 LR 0.001000 Time 0.021346 +2023-10-05 21:20:55,055 - Epoch: [82][ 1040/ 1236] Overall Loss 0.330925 Objective Loss 0.330925 LR 0.001000 Time 0.021337 +2023-10-05 21:20:55,257 - Epoch: [82][ 1050/ 1236] Overall Loss 0.331015 Objective Loss 0.331015 LR 0.001000 Time 0.021326 +2023-10-05 21:20:55,464 - Epoch: [82][ 1060/ 1236] Overall Loss 0.330702 Objective Loss 0.330702 LR 0.001000 Time 0.021319 +2023-10-05 21:20:55,667 - Epoch: [82][ 1070/ 1236] Overall Loss 0.330136 Objective Loss 0.330136 LR 0.001000 Time 0.021309 +2023-10-05 21:20:55,873 - Epoch: [82][ 1080/ 1236] Overall Loss 0.330121 Objective Loss 0.330121 LR 0.001000 Time 0.021302 +2023-10-05 21:20:56,076 - Epoch: [82][ 1090/ 1236] Overall Loss 0.330035 Objective Loss 0.330035 LR 0.001000 Time 0.021293 +2023-10-05 21:20:56,281 - Epoch: [82][ 1100/ 1236] Overall Loss 0.329615 Objective Loss 0.329615 LR 0.001000 Time 0.021285 +2023-10-05 21:20:56,484 - Epoch: [82][ 1110/ 1236] Overall Loss 0.329479 Objective Loss 0.329479 LR 0.001000 Time 0.021277 +2023-10-05 21:20:56,689 - Epoch: [82][ 1120/ 1236] Overall Loss 0.329457 Objective Loss 0.329457 LR 0.001000 Time 0.021269 +2023-10-05 21:20:56,892 - Epoch: [82][ 1130/ 1236] Overall Loss 0.330001 Objective Loss 0.330001 LR 0.001000 Time 0.021260 +2023-10-05 21:20:57,097 - Epoch: [82][ 1140/ 1236] Overall Loss 0.330121 Objective Loss 0.330121 LR 0.001000 Time 0.021253 +2023-10-05 21:20:57,301 - Epoch: [82][ 1150/ 1236] Overall Loss 0.329880 Objective Loss 0.329880 LR 0.001000 Time 0.021245 +2023-10-05 21:20:57,506 - Epoch: [82][ 1160/ 1236] Overall Loss 0.330150 Objective Loss 0.330150 LR 0.001000 Time 0.021239 +2023-10-05 21:20:57,709 - Epoch: [82][ 1170/ 1236] Overall Loss 0.330232 Objective Loss 0.330232 LR 0.001000 Time 0.021230 +2023-10-05 21:20:57,913 - Epoch: [82][ 1180/ 1236] Overall Loss 0.330634 Objective Loss 0.330634 LR 0.001000 Time 0.021223 +2023-10-05 21:20:58,117 - Epoch: [82][ 1190/ 1236] Overall Loss 0.331062 Objective Loss 0.331062 LR 0.001000 Time 0.021216 +2023-10-05 21:20:58,322 - Epoch: [82][ 1200/ 1236] Overall Loss 0.331266 Objective Loss 0.331266 LR 0.001000 Time 0.021210 +2023-10-05 21:20:58,526 - Epoch: [82][ 1210/ 1236] Overall Loss 0.331205 Objective Loss 0.331205 LR 0.001000 Time 0.021202 +2023-10-05 21:20:58,731 - Epoch: [82][ 1220/ 1236] Overall Loss 0.331329 Objective Loss 0.331329 LR 0.001000 Time 0.021197 +2023-10-05 21:20:58,989 - Epoch: [82][ 1230/ 1236] Overall Loss 0.331312 Objective Loss 0.331312 LR 0.001000 Time 0.021234 +2023-10-05 21:20:59,109 - Epoch: [82][ 1236/ 1236] Overall Loss 0.331174 Objective Loss 0.331174 Top1 82.688391 Top5 97.963340 LR 0.001000 Time 0.021227 +2023-10-05 21:20:59,253 - --- validate (epoch=82)----------- +2023-10-05 21:20:59,253 - 29943 samples (256 per mini-batch) +2023-10-05 21:20:59,715 - Epoch: [82][ 10/ 117] Loss 0.347360 Top1 80.156250 Top5 96.992188 +2023-10-05 21:20:59,869 - Epoch: [82][ 20/ 117] Loss 0.364226 Top1 80.449219 Top5 96.972656 +2023-10-05 21:21:00,022 - Epoch: [82][ 30/ 117] Loss 0.359438 Top1 80.794271 Top5 97.109375 +2023-10-05 21:21:00,174 - Epoch: [82][ 40/ 117] Loss 0.367201 Top1 80.566406 Top5 97.099609 +2023-10-05 21:21:00,327 - Epoch: [82][ 50/ 117] Loss 0.366531 Top1 80.546875 Top5 97.000000 +2023-10-05 21:21:00,479 - Epoch: [82][ 60/ 117] Loss 0.368865 Top1 80.533854 Top5 97.005208 +2023-10-05 21:21:00,630 - Epoch: [82][ 70/ 117] Loss 0.366658 Top1 80.753348 Top5 97.003348 +2023-10-05 21:21:00,780 - Epoch: [82][ 80/ 117] Loss 0.364460 Top1 80.839844 Top5 97.094727 +2023-10-05 21:21:00,927 - Epoch: [82][ 90/ 117] Loss 0.366683 Top1 80.872396 Top5 97.131076 +2023-10-05 21:21:01,077 - Epoch: [82][ 100/ 117] Loss 0.363228 Top1 81.042969 Top5 97.160156 +2023-10-05 21:21:01,234 - Epoch: [82][ 110/ 117] Loss 0.365077 Top1 81.019176 Top5 97.137784 +2023-10-05 21:21:01,320 - Epoch: [82][ 117/ 117] Loss 0.363495 Top1 81.007247 Top5 97.141235 +2023-10-05 21:21:01,436 - ==> Top1: 81.007 Top5: 97.141 Loss: 0.363 + +2023-10-05 21:21:01,437 - ==> Confusion: +[[ 915 1 3 3 18 0 0 0 7 74 1 2 1 3 3 1 3 0 1 0 14] + [ 1 1040 2 1 10 17 1 22 1 0 3 2 0 0 0 3 10 1 10 0 7] + [ 9 1 949 6 7 0 30 6 0 1 2 3 9 4 3 3 1 2 9 2 9] + [ 4 2 21 956 3 3 3 2 4 0 14 0 5 3 26 3 1 8 19 0 12] + [ 19 7 2 0 982 1 1 1 0 6 0 1 1 2 6 6 13 1 0 0 1] + [ 7 48 1 1 8 956 0 31 1 0 4 9 5 13 2 2 6 1 5 8 8] + [ 0 4 50 0 0 0 1105 4 0 0 3 3 1 1 1 7 0 1 1 6 4] + [ 2 24 26 0 7 30 10 1024 0 2 5 9 7 1 1 4 1 1 47 10 7] + [ 20 4 1 0 1 4 0 0 960 49 10 1 3 7 12 3 3 2 8 0 1] + [ 100 1 2 0 7 2 1 0 28 926 0 1 0 25 5 2 5 3 1 3 7] + [ 1 4 16 4 2 1 5 5 7 0 955 3 1 15 3 1 2 3 13 2 10] + [ 1 1 3 0 2 9 0 2 1 2 0 927 44 7 0 1 1 20 1 8 5] + [ 0 3 6 6 0 1 1 4 2 0 0 48 948 2 4 4 2 25 2 3 7] + [ 2 0 1 0 2 15 0 1 20 11 6 9 6 1019 4 3 6 3 0 3 8] + [ 21 1 5 14 11 0 1 0 19 10 1 1 1 1 985 0 4 1 14 0 11] + [ 0 4 4 4 6 1 1 0 0 0 0 7 11 0 0 1047 21 14 0 8 6] + [ 1 7 3 0 15 5 0 2 1 0 0 7 1 1 4 10 1084 1 3 4 12] + [ 0 0 0 1 0 0 0 0 1 1 0 4 19 1 2 5 0 998 3 2 1] + [ 1 7 11 11 0 1 0 29 1 2 1 4 6 0 5 0 2 1 979 1 6] + [ 0 5 6 2 2 3 10 11 0 0 1 11 7 7 0 3 13 1 2 1063 5] + [ 194 218 205 103 174 138 62 115 137 85 177 128 403 306 135 53 199 111 226 298 4438]] + +2023-10-05 21:21:01,438 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:21:01,438 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:21:01,444 - + +2023-10-05 21:21:01,444 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:21:02,444 - Epoch: [83][ 10/ 1236] Overall Loss 0.317370 Objective Loss 0.317370 LR 0.001000 Time 0.099947 +2023-10-05 21:21:02,648 - Epoch: [83][ 20/ 1236] Overall Loss 0.325070 Objective Loss 0.325070 LR 0.001000 Time 0.060126 +2023-10-05 21:21:02,856 - Epoch: [83][ 30/ 1236] Overall Loss 0.320906 Objective Loss 0.320906 LR 0.001000 Time 0.046992 +2023-10-05 21:21:03,060 - Epoch: [83][ 40/ 1236] Overall Loss 0.321463 Objective Loss 0.321463 LR 0.001000 Time 0.040337 +2023-10-05 21:21:03,265 - Epoch: [83][ 50/ 1236] Overall Loss 0.321731 Objective Loss 0.321731 LR 0.001000 Time 0.036357 +2023-10-05 21:21:03,469 - Epoch: [83][ 60/ 1236] Overall Loss 0.325260 Objective Loss 0.325260 LR 0.001000 Time 0.033699 +2023-10-05 21:21:03,674 - Epoch: [83][ 70/ 1236] Overall Loss 0.322416 Objective Loss 0.322416 LR 0.001000 Time 0.031805 +2023-10-05 21:21:03,877 - Epoch: [83][ 80/ 1236] Overall Loss 0.324860 Objective Loss 0.324860 LR 0.001000 Time 0.030370 +2023-10-05 21:21:04,083 - Epoch: [83][ 90/ 1236] Overall Loss 0.324728 Objective Loss 0.324728 LR 0.001000 Time 0.029270 +2023-10-05 21:21:04,286 - Epoch: [83][ 100/ 1236] Overall Loss 0.325498 Objective Loss 0.325498 LR 0.001000 Time 0.028376 +2023-10-05 21:21:04,491 - Epoch: [83][ 110/ 1236] Overall Loss 0.327453 Objective Loss 0.327453 LR 0.001000 Time 0.027657 +2023-10-05 21:21:04,695 - Epoch: [83][ 120/ 1236] Overall Loss 0.327679 Objective Loss 0.327679 LR 0.001000 Time 0.027046 +2023-10-05 21:21:04,897 - Epoch: [83][ 130/ 1236] Overall Loss 0.327113 Objective Loss 0.327113 LR 0.001000 Time 0.026516 +2023-10-05 21:21:05,101 - Epoch: [83][ 140/ 1236] Overall Loss 0.325739 Objective Loss 0.325739 LR 0.001000 Time 0.026074 +2023-10-05 21:21:05,304 - Epoch: [83][ 150/ 1236] Overall Loss 0.327803 Objective Loss 0.327803 LR 0.001000 Time 0.025691 +2023-10-05 21:21:05,507 - Epoch: [83][ 160/ 1236] Overall Loss 0.327309 Objective Loss 0.327309 LR 0.001000 Time 0.025352 +2023-10-05 21:21:05,707 - Epoch: [83][ 170/ 1236] Overall Loss 0.328200 Objective Loss 0.328200 LR 0.001000 Time 0.025035 +2023-10-05 21:21:05,908 - Epoch: [83][ 180/ 1236] Overall Loss 0.327193 Objective Loss 0.327193 LR 0.001000 Time 0.024757 +2023-10-05 21:21:06,107 - Epoch: [83][ 190/ 1236] Overall Loss 0.328268 Objective Loss 0.328268 LR 0.001000 Time 0.024498 +2023-10-05 21:21:06,307 - Epoch: [83][ 200/ 1236] Overall Loss 0.327744 Objective Loss 0.327744 LR 0.001000 Time 0.024275 +2023-10-05 21:21:06,505 - Epoch: [83][ 210/ 1236] Overall Loss 0.328207 Objective Loss 0.328207 LR 0.001000 Time 0.024060 +2023-10-05 21:21:06,706 - Epoch: [83][ 220/ 1236] Overall Loss 0.329177 Objective Loss 0.329177 LR 0.001000 Time 0.023877 +2023-10-05 21:21:06,904 - Epoch: [83][ 230/ 1236] Overall Loss 0.328627 Objective Loss 0.328627 LR 0.001000 Time 0.023698 +2023-10-05 21:21:07,105 - Epoch: [83][ 240/ 1236] Overall Loss 0.329008 Objective Loss 0.329008 LR 0.001000 Time 0.023547 +2023-10-05 21:21:07,304 - Epoch: [83][ 250/ 1236] Overall Loss 0.329006 Objective Loss 0.329006 LR 0.001000 Time 0.023400 +2023-10-05 21:21:07,504 - Epoch: [83][ 260/ 1236] Overall Loss 0.327766 Objective Loss 0.327766 LR 0.001000 Time 0.023269 +2023-10-05 21:21:07,703 - Epoch: [83][ 270/ 1236] Overall Loss 0.327476 Objective Loss 0.327476 LR 0.001000 Time 0.023142 +2023-10-05 21:21:07,904 - Epoch: [83][ 280/ 1236] Overall Loss 0.326571 Objective Loss 0.326571 LR 0.001000 Time 0.023030 +2023-10-05 21:21:08,102 - Epoch: [83][ 290/ 1236] Overall Loss 0.326477 Objective Loss 0.326477 LR 0.001000 Time 0.022921 +2023-10-05 21:21:08,303 - Epoch: [83][ 300/ 1236] Overall Loss 0.326019 Objective Loss 0.326019 LR 0.001000 Time 0.022825 +2023-10-05 21:21:08,502 - Epoch: [83][ 310/ 1236] Overall Loss 0.325375 Objective Loss 0.325375 LR 0.001000 Time 0.022730 +2023-10-05 21:21:08,703 - Epoch: [83][ 320/ 1236] Overall Loss 0.325299 Objective Loss 0.325299 LR 0.001000 Time 0.022646 +2023-10-05 21:21:08,905 - Epoch: [83][ 330/ 1236] Overall Loss 0.324436 Objective Loss 0.324436 LR 0.001000 Time 0.022571 +2023-10-05 21:21:09,108 - Epoch: [83][ 340/ 1236] Overall Loss 0.323917 Objective Loss 0.323917 LR 0.001000 Time 0.022502 +2023-10-05 21:21:09,307 - Epoch: [83][ 350/ 1236] Overall Loss 0.322878 Objective Loss 0.322878 LR 0.001000 Time 0.022426 +2023-10-05 21:21:09,507 - Epoch: [83][ 360/ 1236] Overall Loss 0.322825 Objective Loss 0.322825 LR 0.001000 Time 0.022359 +2023-10-05 21:21:09,712 - Epoch: [83][ 370/ 1236] Overall Loss 0.322641 Objective Loss 0.322641 LR 0.001000 Time 0.022307 +2023-10-05 21:21:09,924 - Epoch: [83][ 380/ 1236] Overall Loss 0.322694 Objective Loss 0.322694 LR 0.001000 Time 0.022278 +2023-10-05 21:21:10,134 - Epoch: [83][ 390/ 1236] Overall Loss 0.323891 Objective Loss 0.323891 LR 0.001000 Time 0.022243 +2023-10-05 21:21:10,346 - Epoch: [83][ 400/ 1236] Overall Loss 0.324459 Objective Loss 0.324459 LR 0.001000 Time 0.022217 +2023-10-05 21:21:10,550 - Epoch: [83][ 410/ 1236] Overall Loss 0.324639 Objective Loss 0.324639 LR 0.001000 Time 0.022173 +2023-10-05 21:21:10,753 - Epoch: [83][ 420/ 1236] Overall Loss 0.324769 Objective Loss 0.324769 LR 0.001000 Time 0.022126 +2023-10-05 21:21:10,955 - Epoch: [83][ 430/ 1236] Overall Loss 0.324375 Objective Loss 0.324375 LR 0.001000 Time 0.022082 +2023-10-05 21:21:11,158 - Epoch: [83][ 440/ 1236] Overall Loss 0.324068 Objective Loss 0.324068 LR 0.001000 Time 0.022039 +2023-10-05 21:21:11,360 - Epoch: [83][ 450/ 1236] Overall Loss 0.324815 Objective Loss 0.324815 LR 0.001000 Time 0.021999 +2023-10-05 21:21:11,563 - Epoch: [83][ 460/ 1236] Overall Loss 0.325437 Objective Loss 0.325437 LR 0.001000 Time 0.021960 +2023-10-05 21:21:11,765 - Epoch: [83][ 470/ 1236] Overall Loss 0.324671 Objective Loss 0.324671 LR 0.001000 Time 0.021923 +2023-10-05 21:21:11,968 - Epoch: [83][ 480/ 1236] Overall Loss 0.324210 Objective Loss 0.324210 LR 0.001000 Time 0.021888 +2023-10-05 21:21:12,170 - Epoch: [83][ 490/ 1236] Overall Loss 0.324345 Objective Loss 0.324345 LR 0.001000 Time 0.021853 +2023-10-05 21:21:12,373 - Epoch: [83][ 500/ 1236] Overall Loss 0.324377 Objective Loss 0.324377 LR 0.001000 Time 0.021821 +2023-10-05 21:21:12,575 - Epoch: [83][ 510/ 1236] Overall Loss 0.324408 Objective Loss 0.324408 LR 0.001000 Time 0.021789 +2023-10-05 21:21:12,778 - Epoch: [83][ 520/ 1236] Overall Loss 0.324156 Objective Loss 0.324156 LR 0.001000 Time 0.021759 +2023-10-05 21:21:12,980 - Epoch: [83][ 530/ 1236] Overall Loss 0.324167 Objective Loss 0.324167 LR 0.001000 Time 0.021729 +2023-10-05 21:21:13,184 - Epoch: [83][ 540/ 1236] Overall Loss 0.323787 Objective Loss 0.323787 LR 0.001000 Time 0.021703 +2023-10-05 21:21:13,386 - Epoch: [83][ 550/ 1236] Overall Loss 0.324069 Objective Loss 0.324069 LR 0.001000 Time 0.021676 +2023-10-05 21:21:13,589 - Epoch: [83][ 560/ 1236] Overall Loss 0.323938 Objective Loss 0.323938 LR 0.001000 Time 0.021650 +2023-10-05 21:21:13,792 - Epoch: [83][ 570/ 1236] Overall Loss 0.324219 Objective Loss 0.324219 LR 0.001000 Time 0.021626 +2023-10-05 21:21:13,995 - Epoch: [83][ 580/ 1236] Overall Loss 0.324509 Objective Loss 0.324509 LR 0.001000 Time 0.021602 +2023-10-05 21:21:14,197 - Epoch: [83][ 590/ 1236] Overall Loss 0.324642 Objective Loss 0.324642 LR 0.001000 Time 0.021579 +2023-10-05 21:21:14,400 - Epoch: [83][ 600/ 1236] Overall Loss 0.324644 Objective Loss 0.324644 LR 0.001000 Time 0.021557 +2023-10-05 21:21:14,603 - Epoch: [83][ 610/ 1236] Overall Loss 0.324346 Objective Loss 0.324346 LR 0.001000 Time 0.021535 +2023-10-05 21:21:14,806 - Epoch: [83][ 620/ 1236] Overall Loss 0.324827 Objective Loss 0.324827 LR 0.001000 Time 0.021515 +2023-10-05 21:21:15,009 - Epoch: [83][ 630/ 1236] Overall Loss 0.325058 Objective Loss 0.325058 LR 0.001000 Time 0.021495 +2023-10-05 21:21:15,212 - Epoch: [83][ 640/ 1236] Overall Loss 0.324516 Objective Loss 0.324516 LR 0.001000 Time 0.021476 +2023-10-05 21:21:15,416 - Epoch: [83][ 650/ 1236] Overall Loss 0.324070 Objective Loss 0.324070 LR 0.001000 Time 0.021458 +2023-10-05 21:21:15,621 - Epoch: [83][ 660/ 1236] Overall Loss 0.324071 Objective Loss 0.324071 LR 0.001000 Time 0.021443 +2023-10-05 21:21:15,824 - Epoch: [83][ 670/ 1236] Overall Loss 0.324285 Objective Loss 0.324285 LR 0.001000 Time 0.021426 +2023-10-05 21:21:16,028 - Epoch: [83][ 680/ 1236] Overall Loss 0.324519 Objective Loss 0.324519 LR 0.001000 Time 0.021410 +2023-10-05 21:21:16,231 - Epoch: [83][ 690/ 1236] Overall Loss 0.324448 Objective Loss 0.324448 LR 0.001000 Time 0.021394 +2023-10-05 21:21:16,442 - Epoch: [83][ 700/ 1236] Overall Loss 0.324865 Objective Loss 0.324865 LR 0.001000 Time 0.021389 +2023-10-05 21:21:16,651 - Epoch: [83][ 710/ 1236] Overall Loss 0.324537 Objective Loss 0.324537 LR 0.001000 Time 0.021382 +2023-10-05 21:21:16,864 - Epoch: [83][ 720/ 1236] Overall Loss 0.324459 Objective Loss 0.324459 LR 0.001000 Time 0.021380 +2023-10-05 21:21:17,074 - Epoch: [83][ 730/ 1236] Overall Loss 0.324430 Objective Loss 0.324430 LR 0.001000 Time 0.021375 +2023-10-05 21:21:17,287 - Epoch: [83][ 740/ 1236] Overall Loss 0.324458 Objective Loss 0.324458 LR 0.001000 Time 0.021373 +2023-10-05 21:21:17,498 - Epoch: [83][ 750/ 1236] Overall Loss 0.324214 Objective Loss 0.324214 LR 0.001000 Time 0.021369 +2023-10-05 21:21:17,711 - Epoch: [83][ 760/ 1236] Overall Loss 0.323907 Objective Loss 0.323907 LR 0.001000 Time 0.021367 +2023-10-05 21:21:17,922 - Epoch: [83][ 770/ 1236] Overall Loss 0.323693 Objective Loss 0.323693 LR 0.001000 Time 0.021363 +2023-10-05 21:21:18,135 - Epoch: [83][ 780/ 1236] Overall Loss 0.323486 Objective Loss 0.323486 LR 0.001000 Time 0.021363 +2023-10-05 21:21:18,345 - Epoch: [83][ 790/ 1236] Overall Loss 0.323461 Objective Loss 0.323461 LR 0.001000 Time 0.021357 +2023-10-05 21:21:18,554 - Epoch: [83][ 800/ 1236] Overall Loss 0.323325 Objective Loss 0.323325 LR 0.001000 Time 0.021352 +2023-10-05 21:21:18,763 - Epoch: [83][ 810/ 1236] Overall Loss 0.323104 Objective Loss 0.323104 LR 0.001000 Time 0.021346 +2023-10-05 21:21:18,977 - Epoch: [83][ 820/ 1236] Overall Loss 0.323125 Objective Loss 0.323125 LR 0.001000 Time 0.021345 +2023-10-05 21:21:19,192 - Epoch: [83][ 830/ 1236] Overall Loss 0.322808 Objective Loss 0.322808 LR 0.001000 Time 0.021347 +2023-10-05 21:21:19,404 - Epoch: [83][ 840/ 1236] Overall Loss 0.322564 Objective Loss 0.322564 LR 0.001000 Time 0.021345 +2023-10-05 21:21:19,614 - Epoch: [83][ 850/ 1236] Overall Loss 0.322378 Objective Loss 0.322378 LR 0.001000 Time 0.021340 +2023-10-05 21:21:19,826 - Epoch: [83][ 860/ 1236] Overall Loss 0.322667 Objective Loss 0.322667 LR 0.001000 Time 0.021339 +2023-10-05 21:21:20,037 - Epoch: [83][ 870/ 1236] Overall Loss 0.322554 Objective Loss 0.322554 LR 0.001000 Time 0.021336 +2023-10-05 21:21:20,250 - Epoch: [83][ 880/ 1236] Overall Loss 0.322659 Objective Loss 0.322659 LR 0.001000 Time 0.021335 +2023-10-05 21:21:20,460 - Epoch: [83][ 890/ 1236] Overall Loss 0.322459 Objective Loss 0.322459 LR 0.001000 Time 0.021330 +2023-10-05 21:21:20,672 - Epoch: [83][ 900/ 1236] Overall Loss 0.322540 Objective Loss 0.322540 LR 0.001000 Time 0.021329 +2023-10-05 21:21:20,880 - Epoch: [83][ 910/ 1236] Overall Loss 0.322815 Objective Loss 0.322815 LR 0.001000 Time 0.021323 +2023-10-05 21:21:21,093 - Epoch: [83][ 920/ 1236] Overall Loss 0.322860 Objective Loss 0.322860 LR 0.001000 Time 0.021322 +2023-10-05 21:21:21,301 - Epoch: [83][ 930/ 1236] Overall Loss 0.322952 Objective Loss 0.322952 LR 0.001000 Time 0.021316 +2023-10-05 21:21:21,513 - Epoch: [83][ 940/ 1236] Overall Loss 0.323165 Objective Loss 0.323165 LR 0.001000 Time 0.021314 +2023-10-05 21:21:21,721 - Epoch: [83][ 950/ 1236] Overall Loss 0.323194 Objective Loss 0.323194 LR 0.001000 Time 0.021309 +2023-10-05 21:21:21,933 - Epoch: [83][ 960/ 1236] Overall Loss 0.323386 Objective Loss 0.323386 LR 0.001000 Time 0.021307 +2023-10-05 21:21:22,141 - Epoch: [83][ 970/ 1236] Overall Loss 0.323845 Objective Loss 0.323845 LR 0.001000 Time 0.021302 +2023-10-05 21:21:22,353 - Epoch: [83][ 980/ 1236] Overall Loss 0.323598 Objective Loss 0.323598 LR 0.001000 Time 0.021300 +2023-10-05 21:21:22,561 - Epoch: [83][ 990/ 1236] Overall Loss 0.323551 Objective Loss 0.323551 LR 0.001000 Time 0.021295 +2023-10-05 21:21:22,773 - Epoch: [83][ 1000/ 1236] Overall Loss 0.323593 Objective Loss 0.323593 LR 0.001000 Time 0.021294 +2023-10-05 21:21:22,982 - Epoch: [83][ 1010/ 1236] Overall Loss 0.323613 Objective Loss 0.323613 LR 0.001000 Time 0.021290 +2023-10-05 21:21:23,195 - Epoch: [83][ 1020/ 1236] Overall Loss 0.323836 Objective Loss 0.323836 LR 0.001000 Time 0.021289 +2023-10-05 21:21:23,403 - Epoch: [83][ 1030/ 1236] Overall Loss 0.324007 Objective Loss 0.324007 LR 0.001000 Time 0.021284 +2023-10-05 21:21:23,615 - Epoch: [83][ 1040/ 1236] Overall Loss 0.323951 Objective Loss 0.323951 LR 0.001000 Time 0.021283 +2023-10-05 21:21:23,822 - Epoch: [83][ 1050/ 1236] Overall Loss 0.324357 Objective Loss 0.324357 LR 0.001000 Time 0.021278 +2023-10-05 21:21:24,035 - Epoch: [83][ 1060/ 1236] Overall Loss 0.323786 Objective Loss 0.323786 LR 0.001000 Time 0.021277 +2023-10-05 21:21:24,243 - Epoch: [83][ 1070/ 1236] Overall Loss 0.323822 Objective Loss 0.323822 LR 0.001000 Time 0.021272 +2023-10-05 21:21:24,455 - Epoch: [83][ 1080/ 1236] Overall Loss 0.323903 Objective Loss 0.323903 LR 0.001000 Time 0.021271 +2023-10-05 21:21:24,669 - Epoch: [83][ 1090/ 1236] Overall Loss 0.323510 Objective Loss 0.323510 LR 0.001000 Time 0.021273 +2023-10-05 21:21:24,881 - Epoch: [83][ 1100/ 1236] Overall Loss 0.323378 Objective Loss 0.323378 LR 0.001000 Time 0.021272 +2023-10-05 21:21:25,090 - Epoch: [83][ 1110/ 1236] Overall Loss 0.323915 Objective Loss 0.323915 LR 0.001000 Time 0.021268 +2023-10-05 21:21:25,302 - Epoch: [83][ 1120/ 1236] Overall Loss 0.323853 Objective Loss 0.323853 LR 0.001000 Time 0.021267 +2023-10-05 21:21:25,510 - Epoch: [83][ 1130/ 1236] Overall Loss 0.323647 Objective Loss 0.323647 LR 0.001000 Time 0.021263 +2023-10-05 21:21:25,722 - Epoch: [83][ 1140/ 1236] Overall Loss 0.323524 Objective Loss 0.323524 LR 0.001000 Time 0.021262 +2023-10-05 21:21:25,930 - Epoch: [83][ 1150/ 1236] Overall Loss 0.323527 Objective Loss 0.323527 LR 0.001000 Time 0.021257 +2023-10-05 21:21:26,142 - Epoch: [83][ 1160/ 1236] Overall Loss 0.323759 Objective Loss 0.323759 LR 0.001000 Time 0.021257 +2023-10-05 21:21:26,345 - Epoch: [83][ 1170/ 1236] Overall Loss 0.323767 Objective Loss 0.323767 LR 0.001000 Time 0.021249 +2023-10-05 21:21:26,548 - Epoch: [83][ 1180/ 1236] Overall Loss 0.323527 Objective Loss 0.323527 LR 0.001000 Time 0.021241 +2023-10-05 21:21:26,750 - Epoch: [83][ 1190/ 1236] Overall Loss 0.323488 Objective Loss 0.323488 LR 0.001000 Time 0.021231 +2023-10-05 21:21:26,959 - Epoch: [83][ 1200/ 1236] Overall Loss 0.323531 Objective Loss 0.323531 LR 0.001000 Time 0.021228 +2023-10-05 21:21:27,161 - Epoch: [83][ 1210/ 1236] Overall Loss 0.323583 Objective Loss 0.323583 LR 0.001000 Time 0.021220 +2023-10-05 21:21:27,364 - Epoch: [83][ 1220/ 1236] Overall Loss 0.323671 Objective Loss 0.323671 LR 0.001000 Time 0.021212 +2023-10-05 21:21:27,619 - Epoch: [83][ 1230/ 1236] Overall Loss 0.323821 Objective Loss 0.323821 LR 0.001000 Time 0.021247 +2023-10-05 21:21:27,736 - Epoch: [83][ 1236/ 1236] Overall Loss 0.324013 Objective Loss 0.324013 Top1 81.873727 Top5 96.741344 LR 0.001000 Time 0.021238 +2023-10-05 21:21:27,859 - --- validate (epoch=83)----------- +2023-10-05 21:21:27,859 - 29943 samples (256 per mini-batch) +2023-10-05 21:21:28,318 - Epoch: [83][ 10/ 117] Loss 0.353445 Top1 81.992188 Top5 97.578125 +2023-10-05 21:21:28,464 - Epoch: [83][ 20/ 117] Loss 0.366173 Top1 81.464844 Top5 97.441406 +2023-10-05 21:21:28,607 - Epoch: [83][ 30/ 117] Loss 0.363608 Top1 81.289062 Top5 97.486979 +2023-10-05 21:21:28,752 - Epoch: [83][ 40/ 117] Loss 0.360059 Top1 81.552734 Top5 97.451172 +2023-10-05 21:21:28,898 - Epoch: [83][ 50/ 117] Loss 0.367705 Top1 81.304688 Top5 97.437500 +2023-10-05 21:21:29,044 - Epoch: [83][ 60/ 117] Loss 0.361802 Top1 81.653646 Top5 97.506510 +2023-10-05 21:21:29,188 - Epoch: [83][ 70/ 117] Loss 0.364381 Top1 81.579241 Top5 97.500000 +2023-10-05 21:21:29,340 - Epoch: [83][ 80/ 117] Loss 0.363867 Top1 81.455078 Top5 97.480469 +2023-10-05 21:21:29,491 - Epoch: [83][ 90/ 117] Loss 0.364591 Top1 81.397569 Top5 97.434896 +2023-10-05 21:21:29,635 - Epoch: [83][ 100/ 117] Loss 0.366890 Top1 81.324219 Top5 97.355469 +2023-10-05 21:21:29,787 - Epoch: [83][ 110/ 117] Loss 0.366461 Top1 81.352983 Top5 97.393466 +2023-10-05 21:21:29,872 - Epoch: [83][ 117/ 117] Loss 0.363745 Top1 81.361253 Top5 97.391711 +2023-10-05 21:21:29,975 - ==> Top1: 81.361 Top5: 97.392 Loss: 0.364 + +2023-10-05 21:21:29,976 - ==> Confusion: +[[ 923 2 3 2 11 6 0 2 5 70 1 0 0 5 6 1 1 1 1 0 10] + [ 3 1001 0 0 10 41 3 30 2 0 4 1 0 0 3 4 4 0 17 1 7] + [ 7 1 925 15 1 1 43 14 0 1 4 1 8 1 1 7 1 3 8 5 9] + [ 3 1 15 930 2 6 5 0 2 0 9 0 5 3 50 5 2 8 30 2 11] + [ 33 2 3 0 952 10 1 0 1 6 2 1 1 2 12 5 14 1 0 2 2] + [ 5 23 1 2 3 977 1 27 2 1 3 11 2 15 8 2 3 0 4 12 14] + [ 0 2 16 0 0 0 1141 9 0 0 1 2 0 1 1 8 1 0 3 1 5] + [ 3 10 22 1 1 39 14 1046 3 3 1 8 4 0 0 4 0 1 46 7 5] + [ 15 2 1 1 1 7 0 1 938 49 19 2 1 10 25 4 2 1 8 0 2] + [ 135 1 1 0 5 4 3 0 26 881 0 0 0 34 10 5 1 1 1 4 7] + [ 3 4 9 7 0 0 8 3 15 3 957 2 2 16 3 1 1 2 8 0 9] + [ 1 0 0 0 1 7 2 1 0 0 1 935 36 8 0 4 1 19 0 15 4] + [ 0 1 6 7 0 3 0 3 1 0 0 56 939 3 2 6 0 24 1 4 12] + [ 1 0 2 1 2 12 1 0 11 12 2 4 3 1047 5 2 0 1 0 6 7] + [ 17 2 3 9 7 0 1 0 19 4 2 1 2 3 1004 0 0 4 16 0 7] + [ 0 3 3 1 1 1 4 0 0 0 0 5 12 1 0 1060 14 15 0 8 6] + [ 0 6 3 0 5 7 0 3 1 0 0 4 4 1 2 16 1085 1 3 9 11] + [ 1 0 1 1 0 0 1 0 0 0 0 9 20 1 4 5 0 989 1 1 4] + [ 1 6 6 20 2 0 1 41 3 0 3 1 2 3 10 0 1 0 960 0 8] + [ 0 2 2 1 1 8 15 12 0 0 0 13 4 6 0 9 6 0 3 1064 6] + [ 148 136 175 85 110 195 84 149 106 74 216 154 345 311 200 65 154 113 223 254 4608]] + +2023-10-05 21:21:29,977 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:21:29,977 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:21:29,983 - + +2023-10-05 21:21:29,983 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:21:31,079 - Epoch: [84][ 10/ 1236] Overall Loss 0.320084 Objective Loss 0.320084 LR 0.001000 Time 0.109531 +2023-10-05 21:21:31,281 - Epoch: [84][ 20/ 1236] Overall Loss 0.338065 Objective Loss 0.338065 LR 0.001000 Time 0.064873 +2023-10-05 21:21:31,482 - Epoch: [84][ 30/ 1236] Overall Loss 0.332298 Objective Loss 0.332298 LR 0.001000 Time 0.049916 +2023-10-05 21:21:31,684 - Epoch: [84][ 40/ 1236] Overall Loss 0.317247 Objective Loss 0.317247 LR 0.001000 Time 0.042492 +2023-10-05 21:21:31,887 - Epoch: [84][ 50/ 1236] Overall Loss 0.313385 Objective Loss 0.313385 LR 0.001000 Time 0.038038 +2023-10-05 21:21:32,094 - Epoch: [84][ 60/ 1236] Overall Loss 0.318080 Objective Loss 0.318080 LR 0.001000 Time 0.035147 +2023-10-05 21:21:32,295 - Epoch: [84][ 70/ 1236] Overall Loss 0.322251 Objective Loss 0.322251 LR 0.001000 Time 0.032998 +2023-10-05 21:21:32,498 - Epoch: [84][ 80/ 1236] Overall Loss 0.321657 Objective Loss 0.321657 LR 0.001000 Time 0.031406 +2023-10-05 21:21:32,700 - Epoch: [84][ 90/ 1236] Overall Loss 0.322808 Objective Loss 0.322808 LR 0.001000 Time 0.030153 +2023-10-05 21:21:32,902 - Epoch: [84][ 100/ 1236] Overall Loss 0.324275 Objective Loss 0.324275 LR 0.001000 Time 0.029155 +2023-10-05 21:21:33,103 - Epoch: [84][ 110/ 1236] Overall Loss 0.326639 Objective Loss 0.326639 LR 0.001000 Time 0.028325 +2023-10-05 21:21:33,306 - Epoch: [84][ 120/ 1236] Overall Loss 0.328246 Objective Loss 0.328246 LR 0.001000 Time 0.027653 +2023-10-05 21:21:33,506 - Epoch: [84][ 130/ 1236] Overall Loss 0.326639 Objective Loss 0.326639 LR 0.001000 Time 0.027067 +2023-10-05 21:21:33,708 - Epoch: [84][ 140/ 1236] Overall Loss 0.326256 Objective Loss 0.326256 LR 0.001000 Time 0.026574 +2023-10-05 21:21:33,909 - Epoch: [84][ 150/ 1236] Overall Loss 0.329012 Objective Loss 0.329012 LR 0.001000 Time 0.026134 +2023-10-05 21:21:34,112 - Epoch: [84][ 160/ 1236] Overall Loss 0.328985 Objective Loss 0.328985 LR 0.001000 Time 0.025773 +2023-10-05 21:21:34,316 - Epoch: [84][ 170/ 1236] Overall Loss 0.328901 Objective Loss 0.328901 LR 0.001000 Time 0.025450 +2023-10-05 21:21:34,521 - Epoch: [84][ 180/ 1236] Overall Loss 0.329775 Objective Loss 0.329775 LR 0.001000 Time 0.025173 +2023-10-05 21:21:34,724 - Epoch: [84][ 190/ 1236] Overall Loss 0.329214 Objective Loss 0.329214 LR 0.001000 Time 0.024919 +2023-10-05 21:21:34,928 - Epoch: [84][ 200/ 1236] Overall Loss 0.328693 Objective Loss 0.328693 LR 0.001000 Time 0.024690 +2023-10-05 21:21:35,132 - Epoch: [84][ 210/ 1236] Overall Loss 0.327485 Objective Loss 0.327485 LR 0.001000 Time 0.024476 +2023-10-05 21:21:35,338 - Epoch: [84][ 220/ 1236] Overall Loss 0.327101 Objective Loss 0.327101 LR 0.001000 Time 0.024301 +2023-10-05 21:21:35,541 - Epoch: [84][ 230/ 1236] Overall Loss 0.325481 Objective Loss 0.325481 LR 0.001000 Time 0.024121 +2023-10-05 21:21:35,746 - Epoch: [84][ 240/ 1236] Overall Loss 0.325560 Objective Loss 0.325560 LR 0.001000 Time 0.023968 +2023-10-05 21:21:35,951 - Epoch: [84][ 250/ 1236] Overall Loss 0.325646 Objective Loss 0.325646 LR 0.001000 Time 0.023828 +2023-10-05 21:21:36,156 - Epoch: [84][ 260/ 1236] Overall Loss 0.325412 Objective Loss 0.325412 LR 0.001000 Time 0.023698 +2023-10-05 21:21:36,360 - Epoch: [84][ 270/ 1236] Overall Loss 0.325152 Objective Loss 0.325152 LR 0.001000 Time 0.023569 +2023-10-05 21:21:36,565 - Epoch: [84][ 280/ 1236] Overall Loss 0.325002 Objective Loss 0.325002 LR 0.001000 Time 0.023457 +2023-10-05 21:21:36,768 - Epoch: [84][ 290/ 1236] Overall Loss 0.324674 Objective Loss 0.324674 LR 0.001000 Time 0.023350 +2023-10-05 21:21:36,973 - Epoch: [84][ 300/ 1236] Overall Loss 0.324738 Objective Loss 0.324738 LR 0.001000 Time 0.023253 +2023-10-05 21:21:37,177 - Epoch: [84][ 310/ 1236] Overall Loss 0.325118 Objective Loss 0.325118 LR 0.001000 Time 0.023158 +2023-10-05 21:21:37,381 - Epoch: [84][ 320/ 1236] Overall Loss 0.324421 Objective Loss 0.324421 LR 0.001000 Time 0.023073 +2023-10-05 21:21:37,584 - Epoch: [84][ 330/ 1236] Overall Loss 0.324046 Objective Loss 0.324046 LR 0.001000 Time 0.022989 +2023-10-05 21:21:37,790 - Epoch: [84][ 340/ 1236] Overall Loss 0.324035 Objective Loss 0.324035 LR 0.001000 Time 0.022914 +2023-10-05 21:21:37,993 - Epoch: [84][ 350/ 1236] Overall Loss 0.324465 Objective Loss 0.324465 LR 0.001000 Time 0.022839 +2023-10-05 21:21:38,198 - Epoch: [84][ 360/ 1236] Overall Loss 0.324589 Objective Loss 0.324589 LR 0.001000 Time 0.022775 +2023-10-05 21:21:38,401 - Epoch: [84][ 370/ 1236] Overall Loss 0.324523 Objective Loss 0.324523 LR 0.001000 Time 0.022705 +2023-10-05 21:21:38,605 - Epoch: [84][ 380/ 1236] Overall Loss 0.324600 Objective Loss 0.324600 LR 0.001000 Time 0.022646 +2023-10-05 21:21:38,807 - Epoch: [84][ 390/ 1236] Overall Loss 0.323770 Objective Loss 0.323770 LR 0.001000 Time 0.022582 +2023-10-05 21:21:39,010 - Epoch: [84][ 400/ 1236] Overall Loss 0.324186 Objective Loss 0.324186 LR 0.001000 Time 0.022524 +2023-10-05 21:21:39,213 - Epoch: [84][ 410/ 1236] Overall Loss 0.323643 Objective Loss 0.323643 LR 0.001000 Time 0.022468 +2023-10-05 21:21:39,416 - Epoch: [84][ 420/ 1236] Overall Loss 0.323308 Objective Loss 0.323308 LR 0.001000 Time 0.022416 +2023-10-05 21:21:39,618 - Epoch: [84][ 430/ 1236] Overall Loss 0.323854 Objective Loss 0.323854 LR 0.001000 Time 0.022363 +2023-10-05 21:21:39,821 - Epoch: [84][ 440/ 1236] Overall Loss 0.323417 Objective Loss 0.323417 LR 0.001000 Time 0.022315 +2023-10-05 21:21:40,023 - Epoch: [84][ 450/ 1236] Overall Loss 0.323891 Objective Loss 0.323891 LR 0.001000 Time 0.022267 +2023-10-05 21:21:40,226 - Epoch: [84][ 460/ 1236] Overall Loss 0.324016 Objective Loss 0.324016 LR 0.001000 Time 0.022223 +2023-10-05 21:21:40,428 - Epoch: [84][ 470/ 1236] Overall Loss 0.324726 Objective Loss 0.324726 LR 0.001000 Time 0.022180 +2023-10-05 21:21:40,631 - Epoch: [84][ 480/ 1236] Overall Loss 0.324801 Objective Loss 0.324801 LR 0.001000 Time 0.022140 +2023-10-05 21:21:40,833 - Epoch: [84][ 490/ 1236] Overall Loss 0.325004 Objective Loss 0.325004 LR 0.001000 Time 0.022100 +2023-10-05 21:21:41,036 - Epoch: [84][ 500/ 1236] Overall Loss 0.325326 Objective Loss 0.325326 LR 0.001000 Time 0.022062 +2023-10-05 21:21:41,238 - Epoch: [84][ 510/ 1236] Overall Loss 0.325180 Objective Loss 0.325180 LR 0.001000 Time 0.022025 +2023-10-05 21:21:41,440 - Epoch: [84][ 520/ 1236] Overall Loss 0.325009 Objective Loss 0.325009 LR 0.001000 Time 0.021989 +2023-10-05 21:21:41,641 - Epoch: [84][ 530/ 1236] Overall Loss 0.324958 Objective Loss 0.324958 LR 0.001000 Time 0.021954 +2023-10-05 21:21:41,843 - Epoch: [84][ 540/ 1236] Overall Loss 0.324727 Objective Loss 0.324727 LR 0.001000 Time 0.021921 +2023-10-05 21:21:42,045 - Epoch: [84][ 550/ 1236] Overall Loss 0.324364 Objective Loss 0.324364 LR 0.001000 Time 0.021888 +2023-10-05 21:21:42,247 - Epoch: [84][ 560/ 1236] Overall Loss 0.324720 Objective Loss 0.324720 LR 0.001000 Time 0.021857 +2023-10-05 21:21:42,449 - Epoch: [84][ 570/ 1236] Overall Loss 0.325124 Objective Loss 0.325124 LR 0.001000 Time 0.021827 +2023-10-05 21:21:42,650 - Epoch: [84][ 580/ 1236] Overall Loss 0.325501 Objective Loss 0.325501 LR 0.001000 Time 0.021798 +2023-10-05 21:21:42,852 - Epoch: [84][ 590/ 1236] Overall Loss 0.325805 Objective Loss 0.325805 LR 0.001000 Time 0.021769 +2023-10-05 21:21:43,054 - Epoch: [84][ 600/ 1236] Overall Loss 0.325638 Objective Loss 0.325638 LR 0.001000 Time 0.021742 +2023-10-05 21:21:43,256 - Epoch: [84][ 610/ 1236] Overall Loss 0.325376 Objective Loss 0.325376 LR 0.001000 Time 0.021717 +2023-10-05 21:21:43,458 - Epoch: [84][ 620/ 1236] Overall Loss 0.325383 Objective Loss 0.325383 LR 0.001000 Time 0.021692 +2023-10-05 21:21:43,676 - Epoch: [84][ 630/ 1236] Overall Loss 0.325214 Objective Loss 0.325214 LR 0.001000 Time 0.021693 +2023-10-05 21:21:43,879 - Epoch: [84][ 640/ 1236] Overall Loss 0.325270 Objective Loss 0.325270 LR 0.001000 Time 0.021671 +2023-10-05 21:21:44,085 - Epoch: [84][ 650/ 1236] Overall Loss 0.324871 Objective Loss 0.324871 LR 0.001000 Time 0.021654 +2023-10-05 21:21:44,288 - Epoch: [84][ 660/ 1236] Overall Loss 0.324663 Objective Loss 0.324663 LR 0.001000 Time 0.021633 +2023-10-05 21:21:44,491 - Epoch: [84][ 670/ 1236] Overall Loss 0.324542 Objective Loss 0.324542 LR 0.001000 Time 0.021612 +2023-10-05 21:21:44,697 - Epoch: [84][ 680/ 1236] Overall Loss 0.324441 Objective Loss 0.324441 LR 0.001000 Time 0.021596 +2023-10-05 21:21:44,899 - Epoch: [84][ 690/ 1236] Overall Loss 0.324356 Objective Loss 0.324356 LR 0.001000 Time 0.021576 +2023-10-05 21:21:45,102 - Epoch: [84][ 700/ 1236] Overall Loss 0.324630 Objective Loss 0.324630 LR 0.001000 Time 0.021558 +2023-10-05 21:21:45,309 - Epoch: [84][ 710/ 1236] Overall Loss 0.324664 Objective Loss 0.324664 LR 0.001000 Time 0.021544 +2023-10-05 21:21:45,522 - Epoch: [84][ 720/ 1236] Overall Loss 0.324798 Objective Loss 0.324798 LR 0.001000 Time 0.021541 +2023-10-05 21:21:45,730 - Epoch: [84][ 730/ 1236] Overall Loss 0.324305 Objective Loss 0.324305 LR 0.001000 Time 0.021531 +2023-10-05 21:21:45,943 - Epoch: [84][ 740/ 1236] Overall Loss 0.324133 Objective Loss 0.324133 LR 0.001000 Time 0.021526 +2023-10-05 21:21:46,151 - Epoch: [84][ 750/ 1236] Overall Loss 0.323743 Objective Loss 0.323743 LR 0.001000 Time 0.021517 +2023-10-05 21:21:46,364 - Epoch: [84][ 760/ 1236] Overall Loss 0.323615 Objective Loss 0.323615 LR 0.001000 Time 0.021513 +2023-10-05 21:21:46,572 - Epoch: [84][ 770/ 1236] Overall Loss 0.323510 Objective Loss 0.323510 LR 0.001000 Time 0.021504 +2023-10-05 21:21:46,784 - Epoch: [84][ 780/ 1236] Overall Loss 0.323428 Objective Loss 0.323428 LR 0.001000 Time 0.021500 +2023-10-05 21:21:46,993 - Epoch: [84][ 790/ 1236] Overall Loss 0.323427 Objective Loss 0.323427 LR 0.001000 Time 0.021491 +2023-10-05 21:21:47,205 - Epoch: [84][ 800/ 1236] Overall Loss 0.323623 Objective Loss 0.323623 LR 0.001000 Time 0.021488 +2023-10-05 21:21:47,414 - Epoch: [84][ 810/ 1236] Overall Loss 0.323786 Objective Loss 0.323786 LR 0.001000 Time 0.021479 +2023-10-05 21:21:47,626 - Epoch: [84][ 820/ 1236] Overall Loss 0.323741 Objective Loss 0.323741 LR 0.001000 Time 0.021476 +2023-10-05 21:21:47,834 - Epoch: [84][ 830/ 1236] Overall Loss 0.323465 Objective Loss 0.323465 LR 0.001000 Time 0.021468 +2023-10-05 21:21:48,047 - Epoch: [84][ 840/ 1236] Overall Loss 0.323485 Objective Loss 0.323485 LR 0.001000 Time 0.021465 +2023-10-05 21:21:48,256 - Epoch: [84][ 850/ 1236] Overall Loss 0.323191 Objective Loss 0.323191 LR 0.001000 Time 0.021458 +2023-10-05 21:21:48,468 - Epoch: [84][ 860/ 1236] Overall Loss 0.323338 Objective Loss 0.323338 LR 0.001000 Time 0.021455 +2023-10-05 21:21:48,676 - Epoch: [84][ 870/ 1236] Overall Loss 0.322949 Objective Loss 0.322949 LR 0.001000 Time 0.021447 +2023-10-05 21:21:48,889 - Epoch: [84][ 880/ 1236] Overall Loss 0.322726 Objective Loss 0.322726 LR 0.001000 Time 0.021445 +2023-10-05 21:21:49,097 - Epoch: [84][ 890/ 1236] Overall Loss 0.322399 Objective Loss 0.322399 LR 0.001000 Time 0.021438 +2023-10-05 21:21:49,310 - Epoch: [84][ 900/ 1236] Overall Loss 0.322763 Objective Loss 0.322763 LR 0.001000 Time 0.021435 +2023-10-05 21:21:49,518 - Epoch: [84][ 910/ 1236] Overall Loss 0.322830 Objective Loss 0.322830 LR 0.001000 Time 0.021428 +2023-10-05 21:21:49,727 - Epoch: [84][ 920/ 1236] Overall Loss 0.323015 Objective Loss 0.323015 LR 0.001000 Time 0.021422 +2023-10-05 21:21:49,929 - Epoch: [84][ 930/ 1236] Overall Loss 0.323091 Objective Loss 0.323091 LR 0.001000 Time 0.021408 +2023-10-05 21:21:50,138 - Epoch: [84][ 940/ 1236] Overall Loss 0.322726 Objective Loss 0.322726 LR 0.001000 Time 0.021403 +2023-10-05 21:21:50,346 - Epoch: [84][ 950/ 1236] Overall Loss 0.322683 Objective Loss 0.322683 LR 0.001000 Time 0.021396 +2023-10-05 21:21:50,559 - Epoch: [84][ 960/ 1236] Overall Loss 0.322233 Objective Loss 0.322233 LR 0.001000 Time 0.021395 +2023-10-05 21:21:50,767 - Epoch: [84][ 970/ 1236] Overall Loss 0.322018 Objective Loss 0.322018 LR 0.001000 Time 0.021388 +2023-10-05 21:21:50,980 - Epoch: [84][ 980/ 1236] Overall Loss 0.321774 Objective Loss 0.321774 LR 0.001000 Time 0.021387 +2023-10-05 21:21:51,188 - Epoch: [84][ 990/ 1236] Overall Loss 0.321795 Objective Loss 0.321795 LR 0.001000 Time 0.021381 +2023-10-05 21:21:51,401 - Epoch: [84][ 1000/ 1236] Overall Loss 0.322009 Objective Loss 0.322009 LR 0.001000 Time 0.021380 +2023-10-05 21:21:51,609 - Epoch: [84][ 1010/ 1236] Overall Loss 0.322127 Objective Loss 0.322127 LR 0.001000 Time 0.021374 +2023-10-05 21:21:51,825 - Epoch: [84][ 1020/ 1236] Overall Loss 0.322348 Objective Loss 0.322348 LR 0.001000 Time 0.021375 +2023-10-05 21:21:52,033 - Epoch: [84][ 1030/ 1236] Overall Loss 0.322443 Objective Loss 0.322443 LR 0.001000 Time 0.021370 +2023-10-05 21:21:52,246 - Epoch: [84][ 1040/ 1236] Overall Loss 0.322548 Objective Loss 0.322548 LR 0.001000 Time 0.021368 +2023-10-05 21:21:52,457 - Epoch: [84][ 1050/ 1236] Overall Loss 0.322749 Objective Loss 0.322749 LR 0.001000 Time 0.021366 +2023-10-05 21:21:52,670 - Epoch: [84][ 1060/ 1236] Overall Loss 0.322974 Objective Loss 0.322974 LR 0.001000 Time 0.021364 +2023-10-05 21:21:52,878 - Epoch: [84][ 1070/ 1236] Overall Loss 0.323029 Objective Loss 0.323029 LR 0.001000 Time 0.021359 +2023-10-05 21:21:53,091 - Epoch: [84][ 1080/ 1236] Overall Loss 0.323158 Objective Loss 0.323158 LR 0.001000 Time 0.021358 +2023-10-05 21:21:53,299 - Epoch: [84][ 1090/ 1236] Overall Loss 0.323317 Objective Loss 0.323317 LR 0.001000 Time 0.021353 +2023-10-05 21:21:53,513 - Epoch: [84][ 1100/ 1236] Overall Loss 0.323163 Objective Loss 0.323163 LR 0.001000 Time 0.021353 +2023-10-05 21:21:53,721 - Epoch: [84][ 1110/ 1236] Overall Loss 0.323386 Objective Loss 0.323386 LR 0.001000 Time 0.021348 +2023-10-05 21:21:53,934 - Epoch: [84][ 1120/ 1236] Overall Loss 0.323300 Objective Loss 0.323300 LR 0.001000 Time 0.021347 +2023-10-05 21:21:54,142 - Epoch: [84][ 1130/ 1236] Overall Loss 0.323299 Objective Loss 0.323299 LR 0.001000 Time 0.021342 +2023-10-05 21:21:54,355 - Epoch: [84][ 1140/ 1236] Overall Loss 0.323262 Objective Loss 0.323262 LR 0.001000 Time 0.021342 +2023-10-05 21:21:54,564 - Epoch: [84][ 1150/ 1236] Overall Loss 0.323368 Objective Loss 0.323368 LR 0.001000 Time 0.021337 +2023-10-05 21:21:54,777 - Epoch: [84][ 1160/ 1236] Overall Loss 0.323459 Objective Loss 0.323459 LR 0.001000 Time 0.021336 +2023-10-05 21:21:54,985 - Epoch: [84][ 1170/ 1236] Overall Loss 0.323261 Objective Loss 0.323261 LR 0.001000 Time 0.021332 +2023-10-05 21:21:55,198 - Epoch: [84][ 1180/ 1236] Overall Loss 0.323544 Objective Loss 0.323544 LR 0.001000 Time 0.021332 +2023-10-05 21:21:55,407 - Epoch: [84][ 1190/ 1236] Overall Loss 0.323446 Objective Loss 0.323446 LR 0.001000 Time 0.021327 +2023-10-05 21:21:55,620 - Epoch: [84][ 1200/ 1236] Overall Loss 0.323285 Objective Loss 0.323285 LR 0.001000 Time 0.021327 +2023-10-05 21:21:55,828 - Epoch: [84][ 1210/ 1236] Overall Loss 0.323289 Objective Loss 0.323289 LR 0.001000 Time 0.021322 +2023-10-05 21:21:56,041 - Epoch: [84][ 1220/ 1236] Overall Loss 0.323230 Objective Loss 0.323230 LR 0.001000 Time 0.021322 +2023-10-05 21:21:56,303 - Epoch: [84][ 1230/ 1236] Overall Loss 0.323179 Objective Loss 0.323179 LR 0.001000 Time 0.021361 +2023-10-05 21:21:56,420 - Epoch: [84][ 1236/ 1236] Overall Loss 0.323030 Objective Loss 0.323030 Top1 87.983707 Top5 98.167006 LR 0.001000 Time 0.021352 +2023-10-05 21:21:56,545 - --- validate (epoch=84)----------- +2023-10-05 21:21:56,546 - 29943 samples (256 per mini-batch) +2023-10-05 21:21:56,996 - Epoch: [84][ 10/ 117] Loss 0.366517 Top1 81.679688 Top5 96.835938 +2023-10-05 21:21:57,146 - Epoch: [84][ 20/ 117] Loss 0.375626 Top1 80.839844 Top5 96.835938 +2023-10-05 21:21:57,296 - Epoch: [84][ 30/ 117] Loss 0.365518 Top1 81.302083 Top5 97.096354 +2023-10-05 21:21:57,448 - Epoch: [84][ 40/ 117] Loss 0.366207 Top1 81.289062 Top5 97.207031 +2023-10-05 21:21:57,598 - Epoch: [84][ 50/ 117] Loss 0.361526 Top1 81.546875 Top5 97.375000 +2023-10-05 21:21:57,749 - Epoch: [84][ 60/ 117] Loss 0.359676 Top1 81.725260 Top5 97.298177 +2023-10-05 21:21:57,904 - Epoch: [84][ 70/ 117] Loss 0.354535 Top1 81.780134 Top5 97.315848 +2023-10-05 21:21:58,060 - Epoch: [84][ 80/ 117] Loss 0.359172 Top1 81.767578 Top5 97.275391 +2023-10-05 21:21:58,216 - Epoch: [84][ 90/ 117] Loss 0.362988 Top1 81.783854 Top5 97.282986 +2023-10-05 21:21:58,372 - Epoch: [84][ 100/ 117] Loss 0.361647 Top1 81.679688 Top5 97.304688 +2023-10-05 21:21:58,536 - Epoch: [84][ 110/ 117] Loss 0.361690 Top1 81.615767 Top5 97.329545 +2023-10-05 21:21:58,621 - Epoch: [84][ 117/ 117] Loss 0.362677 Top1 81.661824 Top5 97.318238 +2023-10-05 21:21:58,756 - ==> Top1: 81.662 Top5: 97.318 Loss: 0.363 + +2023-10-05 21:21:58,757 - ==> Confusion: +[[ 893 6 3 2 12 0 0 1 11 84 2 0 3 3 6 2 3 1 2 0 16] + [ 1 1042 1 0 5 20 2 26 1 0 1 2 0 0 0 4 5 0 15 0 6] + [ 5 2 964 20 1 0 8 12 0 0 8 4 8 2 2 4 1 0 5 4 6] + [ 5 1 24 950 0 3 1 1 1 0 2 1 5 5 42 4 1 6 26 0 11] + [ 21 8 1 0 964 7 0 1 1 7 1 1 1 3 7 4 13 3 0 0 7] + [ 5 52 0 0 3 942 3 29 5 0 3 11 2 18 6 1 5 1 5 17 8] + [ 1 10 36 1 0 0 1096 15 0 0 4 1 3 0 1 9 1 0 2 2 9] + [ 4 19 27 0 1 35 5 1040 1 1 6 6 3 2 1 1 2 0 42 15 7] + [ 18 3 1 0 1 6 0 0 969 43 9 0 3 6 18 0 2 2 6 1 1] + [ 93 0 1 0 4 7 0 0 39 919 0 1 0 23 12 3 2 2 0 4 9] + [ 4 2 10 15 2 1 7 7 15 2 957 3 0 9 2 0 2 0 5 1 9] + [ 1 0 0 0 1 12 0 3 0 0 0 951 27 2 0 2 2 15 0 16 3] + [ 1 0 5 4 2 5 0 2 1 0 2 38 961 1 2 6 4 16 3 3 12] + [ 3 0 4 2 2 20 1 1 24 7 13 5 3 1008 7 1 2 2 0 5 9] + [ 17 2 2 13 6 1 0 0 30 5 1 0 4 3 987 0 1 5 15 0 9] + [ 0 3 3 2 3 1 1 1 0 0 0 6 8 0 0 1059 16 10 1 10 10] + [ 0 16 1 1 10 7 2 0 3 0 0 4 0 0 3 8 1093 0 0 6 7] + [ 0 1 0 1 0 1 3 0 1 0 0 6 18 3 1 5 0 992 1 1 4] + [ 3 5 6 19 0 0 0 31 1 0 1 0 3 0 7 0 2 0 986 0 4] + [ 0 5 4 4 1 4 6 18 0 0 1 15 4 1 0 4 13 1 2 1065 4] + [ 137 251 174 110 86 155 51 127 133 91 189 165 387 252 173 63 195 96 214 242 4614]] + +2023-10-05 21:21:58,758 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:21:58,758 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:21:58,763 - + +2023-10-05 21:21:58,764 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:21:59,750 - Epoch: [85][ 10/ 1236] Overall Loss 0.339826 Objective Loss 0.339826 LR 0.001000 Time 0.098601 +2023-10-05 21:21:59,953 - Epoch: [85][ 20/ 1236] Overall Loss 0.336533 Objective Loss 0.336533 LR 0.001000 Time 0.059402 +2023-10-05 21:22:00,153 - Epoch: [85][ 30/ 1236] Overall Loss 0.321421 Objective Loss 0.321421 LR 0.001000 Time 0.046285 +2023-10-05 21:22:00,355 - Epoch: [85][ 40/ 1236] Overall Loss 0.318063 Objective Loss 0.318063 LR 0.001000 Time 0.039747 +2023-10-05 21:22:00,555 - Epoch: [85][ 50/ 1236] Overall Loss 0.309960 Objective Loss 0.309960 LR 0.001000 Time 0.035799 +2023-10-05 21:22:00,757 - Epoch: [85][ 60/ 1236] Overall Loss 0.313664 Objective Loss 0.313664 LR 0.001000 Time 0.033192 +2023-10-05 21:22:00,958 - Epoch: [85][ 70/ 1236] Overall Loss 0.314146 Objective Loss 0.314146 LR 0.001000 Time 0.031311 +2023-10-05 21:22:01,160 - Epoch: [85][ 80/ 1236] Overall Loss 0.313707 Objective Loss 0.313707 LR 0.001000 Time 0.029918 +2023-10-05 21:22:01,361 - Epoch: [85][ 90/ 1236] Overall Loss 0.313622 Objective Loss 0.313622 LR 0.001000 Time 0.028817 +2023-10-05 21:22:01,561 - Epoch: [85][ 100/ 1236] Overall Loss 0.318752 Objective Loss 0.318752 LR 0.001000 Time 0.027939 +2023-10-05 21:22:01,761 - Epoch: [85][ 110/ 1236] Overall Loss 0.319539 Objective Loss 0.319539 LR 0.001000 Time 0.027216 +2023-10-05 21:22:01,964 - Epoch: [85][ 120/ 1236] Overall Loss 0.318501 Objective Loss 0.318501 LR 0.001000 Time 0.026632 +2023-10-05 21:22:02,167 - Epoch: [85][ 130/ 1236] Overall Loss 0.318021 Objective Loss 0.318021 LR 0.001000 Time 0.026142 +2023-10-05 21:22:02,371 - Epoch: [85][ 140/ 1236] Overall Loss 0.316166 Objective Loss 0.316166 LR 0.001000 Time 0.025729 +2023-10-05 21:22:02,574 - Epoch: [85][ 150/ 1236] Overall Loss 0.315856 Objective Loss 0.315856 LR 0.001000 Time 0.025365 +2023-10-05 21:22:02,778 - Epoch: [85][ 160/ 1236] Overall Loss 0.314396 Objective Loss 0.314396 LR 0.001000 Time 0.025052 +2023-10-05 21:22:02,981 - Epoch: [85][ 170/ 1236] Overall Loss 0.315566 Objective Loss 0.315566 LR 0.001000 Time 0.024771 +2023-10-05 21:22:03,185 - Epoch: [85][ 180/ 1236] Overall Loss 0.315025 Objective Loss 0.315025 LR 0.001000 Time 0.024527 +2023-10-05 21:22:03,388 - Epoch: [85][ 190/ 1236] Overall Loss 0.315675 Objective Loss 0.315675 LR 0.001000 Time 0.024303 +2023-10-05 21:22:03,592 - Epoch: [85][ 200/ 1236] Overall Loss 0.317470 Objective Loss 0.317470 LR 0.001000 Time 0.024108 +2023-10-05 21:22:03,795 - Epoch: [85][ 210/ 1236] Overall Loss 0.317463 Objective Loss 0.317463 LR 0.001000 Time 0.023923 +2023-10-05 21:22:03,999 - Epoch: [85][ 220/ 1236] Overall Loss 0.317367 Objective Loss 0.317367 LR 0.001000 Time 0.023761 +2023-10-05 21:22:04,202 - Epoch: [85][ 230/ 1236] Overall Loss 0.316974 Objective Loss 0.316974 LR 0.001000 Time 0.023611 +2023-10-05 21:22:04,406 - Epoch: [85][ 240/ 1236] Overall Loss 0.317594 Objective Loss 0.317594 LR 0.001000 Time 0.023476 +2023-10-05 21:22:04,610 - Epoch: [85][ 250/ 1236] Overall Loss 0.317220 Objective Loss 0.317220 LR 0.001000 Time 0.023349 +2023-10-05 21:22:04,814 - Epoch: [85][ 260/ 1236] Overall Loss 0.318955 Objective Loss 0.318955 LR 0.001000 Time 0.023235 +2023-10-05 21:22:05,017 - Epoch: [85][ 270/ 1236] Overall Loss 0.318365 Objective Loss 0.318365 LR 0.001000 Time 0.023127 +2023-10-05 21:22:05,221 - Epoch: [85][ 280/ 1236] Overall Loss 0.318790 Objective Loss 0.318790 LR 0.001000 Time 0.023027 +2023-10-05 21:22:05,424 - Epoch: [85][ 290/ 1236] Overall Loss 0.320202 Objective Loss 0.320202 LR 0.001000 Time 0.022932 +2023-10-05 21:22:05,628 - Epoch: [85][ 300/ 1236] Overall Loss 0.318771 Objective Loss 0.318771 LR 0.001000 Time 0.022845 +2023-10-05 21:22:05,831 - Epoch: [85][ 310/ 1236] Overall Loss 0.317895 Objective Loss 0.317895 LR 0.001000 Time 0.022763 +2023-10-05 21:22:06,035 - Epoch: [85][ 320/ 1236] Overall Loss 0.318076 Objective Loss 0.318076 LR 0.001000 Time 0.022688 +2023-10-05 21:22:06,239 - Epoch: [85][ 330/ 1236] Overall Loss 0.318503 Objective Loss 0.318503 LR 0.001000 Time 0.022616 +2023-10-05 21:22:06,443 - Epoch: [85][ 340/ 1236] Overall Loss 0.318542 Objective Loss 0.318542 LR 0.001000 Time 0.022551 +2023-10-05 21:22:06,646 - Epoch: [85][ 350/ 1236] Overall Loss 0.317959 Objective Loss 0.317959 LR 0.001000 Time 0.022486 +2023-10-05 21:22:06,850 - Epoch: [85][ 360/ 1236] Overall Loss 0.318485 Objective Loss 0.318485 LR 0.001000 Time 0.022428 +2023-10-05 21:22:07,053 - Epoch: [85][ 370/ 1236] Overall Loss 0.318713 Objective Loss 0.318713 LR 0.001000 Time 0.022369 +2023-10-05 21:22:07,257 - Epoch: [85][ 380/ 1236] Overall Loss 0.318289 Objective Loss 0.318289 LR 0.001000 Time 0.022317 +2023-10-05 21:22:07,459 - Epoch: [85][ 390/ 1236] Overall Loss 0.318956 Objective Loss 0.318956 LR 0.001000 Time 0.022262 +2023-10-05 21:22:07,661 - Epoch: [85][ 400/ 1236] Overall Loss 0.318809 Objective Loss 0.318809 LR 0.001000 Time 0.022209 +2023-10-05 21:22:07,864 - Epoch: [85][ 410/ 1236] Overall Loss 0.318597 Objective Loss 0.318597 LR 0.001000 Time 0.022160 +2023-10-05 21:22:08,066 - Epoch: [85][ 420/ 1236] Overall Loss 0.318733 Objective Loss 0.318733 LR 0.001000 Time 0.022112 +2023-10-05 21:22:08,266 - Epoch: [85][ 430/ 1236] Overall Loss 0.319189 Objective Loss 0.319189 LR 0.001000 Time 0.022064 +2023-10-05 21:22:08,469 - Epoch: [85][ 440/ 1236] Overall Loss 0.319000 Objective Loss 0.319000 LR 0.001000 Time 0.022023 +2023-10-05 21:22:08,670 - Epoch: [85][ 450/ 1236] Overall Loss 0.318540 Objective Loss 0.318540 LR 0.001000 Time 0.021978 +2023-10-05 21:22:08,872 - Epoch: [85][ 460/ 1236] Overall Loss 0.319237 Objective Loss 0.319237 LR 0.001000 Time 0.021940 +2023-10-05 21:22:09,074 - Epoch: [85][ 470/ 1236] Overall Loss 0.319375 Objective Loss 0.319375 LR 0.001000 Time 0.021901 +2023-10-05 21:22:09,277 - Epoch: [85][ 480/ 1236] Overall Loss 0.319673 Objective Loss 0.319673 LR 0.001000 Time 0.021869 +2023-10-05 21:22:09,478 - Epoch: [85][ 490/ 1236] Overall Loss 0.319467 Objective Loss 0.319467 LR 0.001000 Time 0.021832 +2023-10-05 21:22:09,681 - Epoch: [85][ 500/ 1236] Overall Loss 0.319571 Objective Loss 0.319571 LR 0.001000 Time 0.021799 +2023-10-05 21:22:09,881 - Epoch: [85][ 510/ 1236] Overall Loss 0.320119 Objective Loss 0.320119 LR 0.001000 Time 0.021764 +2023-10-05 21:22:10,084 - Epoch: [85][ 520/ 1236] Overall Loss 0.320162 Objective Loss 0.320162 LR 0.001000 Time 0.021735 +2023-10-05 21:22:10,295 - Epoch: [85][ 530/ 1236] Overall Loss 0.320668 Objective Loss 0.320668 LR 0.001000 Time 0.021723 +2023-10-05 21:22:10,498 - Epoch: [85][ 540/ 1236] Overall Loss 0.320816 Objective Loss 0.320816 LR 0.001000 Time 0.021696 +2023-10-05 21:22:10,702 - Epoch: [85][ 550/ 1236] Overall Loss 0.320957 Objective Loss 0.320957 LR 0.001000 Time 0.021671 +2023-10-05 21:22:10,904 - Epoch: [85][ 560/ 1236] Overall Loss 0.320986 Objective Loss 0.320986 LR 0.001000 Time 0.021645 +2023-10-05 21:22:11,107 - Epoch: [85][ 570/ 1236] Overall Loss 0.321129 Objective Loss 0.321129 LR 0.001000 Time 0.021619 +2023-10-05 21:22:11,310 - Epoch: [85][ 580/ 1236] Overall Loss 0.321513 Objective Loss 0.321513 LR 0.001000 Time 0.021597 +2023-10-05 21:22:11,513 - Epoch: [85][ 590/ 1236] Overall Loss 0.321665 Objective Loss 0.321665 LR 0.001000 Time 0.021573 +2023-10-05 21:22:11,716 - Epoch: [85][ 600/ 1236] Overall Loss 0.321352 Objective Loss 0.321352 LR 0.001000 Time 0.021551 +2023-10-05 21:22:11,918 - Epoch: [85][ 610/ 1236] Overall Loss 0.321107 Objective Loss 0.321107 LR 0.001000 Time 0.021529 +2023-10-05 21:22:12,122 - Epoch: [85][ 620/ 1236] Overall Loss 0.320824 Objective Loss 0.320824 LR 0.001000 Time 0.021510 +2023-10-05 21:22:12,325 - Epoch: [85][ 630/ 1236] Overall Loss 0.321365 Objective Loss 0.321365 LR 0.001000 Time 0.021490 +2023-10-05 21:22:12,528 - Epoch: [85][ 640/ 1236] Overall Loss 0.321143 Objective Loss 0.321143 LR 0.001000 Time 0.021472 +2023-10-05 21:22:12,731 - Epoch: [85][ 650/ 1236] Overall Loss 0.320727 Objective Loss 0.320727 LR 0.001000 Time 0.021453 +2023-10-05 21:22:12,940 - Epoch: [85][ 660/ 1236] Overall Loss 0.320446 Objective Loss 0.320446 LR 0.001000 Time 0.021445 +2023-10-05 21:22:13,142 - Epoch: [85][ 670/ 1236] Overall Loss 0.320235 Objective Loss 0.320235 LR 0.001000 Time 0.021426 +2023-10-05 21:22:13,346 - Epoch: [85][ 680/ 1236] Overall Loss 0.320347 Objective Loss 0.320347 LR 0.001000 Time 0.021409 +2023-10-05 21:22:13,548 - Epoch: [85][ 690/ 1236] Overall Loss 0.320379 Objective Loss 0.320379 LR 0.001000 Time 0.021391 +2023-10-05 21:22:13,751 - Epoch: [85][ 700/ 1236] Overall Loss 0.320901 Objective Loss 0.320901 LR 0.001000 Time 0.021375 +2023-10-05 21:22:13,952 - Epoch: [85][ 710/ 1236] Overall Loss 0.321150 Objective Loss 0.321150 LR 0.001000 Time 0.021357 +2023-10-05 21:22:14,155 - Epoch: [85][ 720/ 1236] Overall Loss 0.321049 Objective Loss 0.321049 LR 0.001000 Time 0.021342 +2023-10-05 21:22:14,357 - Epoch: [85][ 730/ 1236] Overall Loss 0.321052 Objective Loss 0.321052 LR 0.001000 Time 0.021325 +2023-10-05 21:22:14,560 - Epoch: [85][ 740/ 1236] Overall Loss 0.321325 Objective Loss 0.321325 LR 0.001000 Time 0.021311 +2023-10-05 21:22:14,762 - Epoch: [85][ 750/ 1236] Overall Loss 0.321079 Objective Loss 0.321079 LR 0.001000 Time 0.021296 +2023-10-05 21:22:14,965 - Epoch: [85][ 760/ 1236] Overall Loss 0.321155 Objective Loss 0.321155 LR 0.001000 Time 0.021283 +2023-10-05 21:22:15,167 - Epoch: [85][ 770/ 1236] Overall Loss 0.321198 Objective Loss 0.321198 LR 0.001000 Time 0.021268 +2023-10-05 21:22:15,371 - Epoch: [85][ 780/ 1236] Overall Loss 0.321362 Objective Loss 0.321362 LR 0.001000 Time 0.021255 +2023-10-05 21:22:15,572 - Epoch: [85][ 790/ 1236] Overall Loss 0.321256 Objective Loss 0.321256 LR 0.001000 Time 0.021241 +2023-10-05 21:22:15,775 - Epoch: [85][ 800/ 1236] Overall Loss 0.320812 Objective Loss 0.320812 LR 0.001000 Time 0.021229 +2023-10-05 21:22:15,977 - Epoch: [85][ 810/ 1236] Overall Loss 0.320800 Objective Loss 0.320800 LR 0.001000 Time 0.021216 +2023-10-05 21:22:16,181 - Epoch: [85][ 820/ 1236] Overall Loss 0.320950 Objective Loss 0.320950 LR 0.001000 Time 0.021205 +2023-10-05 21:22:16,382 - Epoch: [85][ 830/ 1236] Overall Loss 0.320707 Objective Loss 0.320707 LR 0.001000 Time 0.021192 +2023-10-05 21:22:16,586 - Epoch: [85][ 840/ 1236] Overall Loss 0.320918 Objective Loss 0.320918 LR 0.001000 Time 0.021181 +2023-10-05 21:22:16,787 - Epoch: [85][ 850/ 1236] Overall Loss 0.320995 Objective Loss 0.320995 LR 0.001000 Time 0.021169 +2023-10-05 21:22:16,990 - Epoch: [85][ 860/ 1236] Overall Loss 0.321377 Objective Loss 0.321377 LR 0.001000 Time 0.021158 +2023-10-05 21:22:17,192 - Epoch: [85][ 870/ 1236] Overall Loss 0.320949 Objective Loss 0.320949 LR 0.001000 Time 0.021147 +2023-10-05 21:22:17,396 - Epoch: [85][ 880/ 1236] Overall Loss 0.321245 Objective Loss 0.321245 LR 0.001000 Time 0.021137 +2023-10-05 21:22:17,597 - Epoch: [85][ 890/ 1236] Overall Loss 0.321222 Objective Loss 0.321222 LR 0.001000 Time 0.021126 +2023-10-05 21:22:17,800 - Epoch: [85][ 900/ 1236] Overall Loss 0.321015 Objective Loss 0.321015 LR 0.001000 Time 0.021117 +2023-10-05 21:22:18,002 - Epoch: [85][ 910/ 1236] Overall Loss 0.321127 Objective Loss 0.321127 LR 0.001000 Time 0.021106 +2023-10-05 21:22:18,206 - Epoch: [85][ 920/ 1236] Overall Loss 0.321484 Objective Loss 0.321484 LR 0.001000 Time 0.021098 +2023-10-05 21:22:18,408 - Epoch: [85][ 930/ 1236] Overall Loss 0.321359 Objective Loss 0.321359 LR 0.001000 Time 0.021088 +2023-10-05 21:22:18,612 - Epoch: [85][ 940/ 1236] Overall Loss 0.321691 Objective Loss 0.321691 LR 0.001000 Time 0.021079 +2023-10-05 21:22:18,813 - Epoch: [85][ 950/ 1236] Overall Loss 0.321816 Objective Loss 0.321816 LR 0.001000 Time 0.021069 +2023-10-05 21:22:19,016 - Epoch: [85][ 960/ 1236] Overall Loss 0.322107 Objective Loss 0.322107 LR 0.001000 Time 0.021061 +2023-10-05 21:22:19,218 - Epoch: [85][ 970/ 1236] Overall Loss 0.322435 Objective Loss 0.322435 LR 0.001000 Time 0.021052 +2023-10-05 21:22:19,421 - Epoch: [85][ 980/ 1236] Overall Loss 0.322228 Objective Loss 0.322228 LR 0.001000 Time 0.021044 +2023-10-05 21:22:19,623 - Epoch: [85][ 990/ 1236] Overall Loss 0.322462 Objective Loss 0.322462 LR 0.001000 Time 0.021035 +2023-10-05 21:22:19,826 - Epoch: [85][ 1000/ 1236] Overall Loss 0.321975 Objective Loss 0.321975 LR 0.001000 Time 0.021027 +2023-10-05 21:22:20,028 - Epoch: [85][ 1010/ 1236] Overall Loss 0.322036 Objective Loss 0.322036 LR 0.001000 Time 0.021018 +2023-10-05 21:22:20,231 - Epoch: [85][ 1020/ 1236] Overall Loss 0.321912 Objective Loss 0.321912 LR 0.001000 Time 0.021011 +2023-10-05 21:22:20,433 - Epoch: [85][ 1030/ 1236] Overall Loss 0.322000 Objective Loss 0.322000 LR 0.001000 Time 0.021003 +2023-10-05 21:22:20,636 - Epoch: [85][ 1040/ 1236] Overall Loss 0.322135 Objective Loss 0.322135 LR 0.001000 Time 0.020996 +2023-10-05 21:22:20,838 - Epoch: [85][ 1050/ 1236] Overall Loss 0.322082 Objective Loss 0.322082 LR 0.001000 Time 0.020988 +2023-10-05 21:22:21,042 - Epoch: [85][ 1060/ 1236] Overall Loss 0.321931 Objective Loss 0.321931 LR 0.001000 Time 0.020981 +2023-10-05 21:22:21,243 - Epoch: [85][ 1070/ 1236] Overall Loss 0.321955 Objective Loss 0.321955 LR 0.001000 Time 0.020973 +2023-10-05 21:22:21,446 - Epoch: [85][ 1080/ 1236] Overall Loss 0.321786 Objective Loss 0.321786 LR 0.001000 Time 0.020967 +2023-10-05 21:22:21,648 - Epoch: [85][ 1090/ 1236] Overall Loss 0.321695 Objective Loss 0.321695 LR 0.001000 Time 0.020959 +2023-10-05 21:22:21,851 - Epoch: [85][ 1100/ 1236] Overall Loss 0.321542 Objective Loss 0.321542 LR 0.001000 Time 0.020953 +2023-10-05 21:22:22,053 - Epoch: [85][ 1110/ 1236] Overall Loss 0.321516 Objective Loss 0.321516 LR 0.001000 Time 0.020946 +2023-10-05 21:22:22,257 - Epoch: [85][ 1120/ 1236] Overall Loss 0.321338 Objective Loss 0.321338 LR 0.001000 Time 0.020940 +2023-10-05 21:22:22,458 - Epoch: [85][ 1130/ 1236] Overall Loss 0.321262 Objective Loss 0.321262 LR 0.001000 Time 0.020933 +2023-10-05 21:22:22,661 - Epoch: [85][ 1140/ 1236] Overall Loss 0.321172 Objective Loss 0.321172 LR 0.001000 Time 0.020927 +2023-10-05 21:22:22,863 - Epoch: [85][ 1150/ 1236] Overall Loss 0.321447 Objective Loss 0.321447 LR 0.001000 Time 0.020920 +2023-10-05 21:22:23,066 - Epoch: [85][ 1160/ 1236] Overall Loss 0.321760 Objective Loss 0.321760 LR 0.001000 Time 0.020915 +2023-10-05 21:22:23,268 - Epoch: [85][ 1170/ 1236] Overall Loss 0.321783 Objective Loss 0.321783 LR 0.001000 Time 0.020908 +2023-10-05 21:22:23,472 - Epoch: [85][ 1180/ 1236] Overall Loss 0.321803 Objective Loss 0.321803 LR 0.001000 Time 0.020903 +2023-10-05 21:22:23,673 - Epoch: [85][ 1190/ 1236] Overall Loss 0.321755 Objective Loss 0.321755 LR 0.001000 Time 0.020897 +2023-10-05 21:22:23,877 - Epoch: [85][ 1200/ 1236] Overall Loss 0.321859 Objective Loss 0.321859 LR 0.001000 Time 0.020892 +2023-10-05 21:22:24,078 - Epoch: [85][ 1210/ 1236] Overall Loss 0.321862 Objective Loss 0.321862 LR 0.001000 Time 0.020885 +2023-10-05 21:22:24,282 - Epoch: [85][ 1220/ 1236] Overall Loss 0.321897 Objective Loss 0.321897 LR 0.001000 Time 0.020880 +2023-10-05 21:22:24,535 - Epoch: [85][ 1230/ 1236] Overall Loss 0.322245 Objective Loss 0.322245 LR 0.001000 Time 0.020917 +2023-10-05 21:22:24,653 - Epoch: [85][ 1236/ 1236] Overall Loss 0.322301 Objective Loss 0.322301 Top1 82.484725 Top5 98.167006 LR 0.001000 Time 0.020910 +2023-10-05 21:22:24,772 - --- validate (epoch=85)----------- +2023-10-05 21:22:24,772 - 29943 samples (256 per mini-batch) +2023-10-05 21:22:25,224 - Epoch: [85][ 10/ 117] Loss 0.369354 Top1 80.703125 Top5 97.187500 +2023-10-05 21:22:25,373 - Epoch: [85][ 20/ 117] Loss 0.366339 Top1 80.917969 Top5 97.167969 +2023-10-05 21:22:25,520 - Epoch: [85][ 30/ 117] Loss 0.359773 Top1 80.625000 Top5 97.122396 +2023-10-05 21:22:25,668 - Epoch: [85][ 40/ 117] Loss 0.369017 Top1 80.498047 Top5 97.226562 +2023-10-05 21:22:25,814 - Epoch: [85][ 50/ 117] Loss 0.367023 Top1 80.664062 Top5 97.187500 +2023-10-05 21:22:25,962 - Epoch: [85][ 60/ 117] Loss 0.361678 Top1 80.729167 Top5 97.180990 +2023-10-05 21:22:26,107 - Epoch: [85][ 70/ 117] Loss 0.363681 Top1 80.569196 Top5 97.098214 +2023-10-05 21:22:26,256 - Epoch: [85][ 80/ 117] Loss 0.366950 Top1 80.371094 Top5 97.109375 +2023-10-05 21:22:26,401 - Epoch: [85][ 90/ 117] Loss 0.366570 Top1 80.360243 Top5 97.092014 +2023-10-05 21:22:26,551 - Epoch: [85][ 100/ 117] Loss 0.366326 Top1 80.390625 Top5 97.156250 +2023-10-05 21:22:26,704 - Epoch: [85][ 110/ 117] Loss 0.366693 Top1 80.340909 Top5 97.169744 +2023-10-05 21:22:26,789 - Epoch: [85][ 117/ 117] Loss 0.365511 Top1 80.389407 Top5 97.167952 +2023-10-05 21:22:26,908 - ==> Top1: 80.389 Top5: 97.168 Loss: 0.366 + +2023-10-05 21:22:26,909 - ==> Confusion: +[[ 909 3 11 2 3 3 1 1 4 82 1 1 0 1 4 6 5 0 1 0 12] + [ 2 1050 5 0 9 22 3 14 2 0 3 1 0 0 0 4 5 0 6 0 5] + [ 6 3 930 22 4 0 41 8 0 0 9 1 7 0 0 7 2 0 6 4 6] + [ 4 1 16 963 0 5 0 1 1 0 9 0 2 2 31 5 3 7 27 2 10] + [ 40 5 2 0 961 3 0 0 1 9 2 1 0 2 4 5 10 1 0 0 4] + [ 6 60 2 1 3 966 0 20 7 4 6 9 4 9 3 2 1 0 3 1 9] + [ 0 7 28 0 0 0 1125 8 0 0 3 4 0 0 1 4 1 1 1 5 3] + [ 3 40 23 0 3 39 4 1004 3 6 6 8 3 1 0 3 2 1 50 9 10] + [ 17 2 2 0 0 6 0 0 950 55 14 5 1 5 15 6 2 2 6 1 0] + [ 109 0 3 0 3 4 2 0 18 941 0 2 0 20 1 7 2 1 0 0 6] + [ 1 6 18 8 2 2 9 4 11 3 951 4 0 11 3 1 2 1 11 0 5] + [ 2 2 3 0 1 18 2 1 0 1 0 925 35 7 0 3 3 13 0 14 5] + [ 5 3 7 9 2 5 3 1 1 1 0 31 946 1 3 11 5 21 3 3 7] + [ 3 0 3 0 2 14 2 1 18 17 8 10 3 1013 4 1 9 1 0 4 6] + [ 22 6 5 16 6 0 0 0 33 3 0 1 1 0 978 0 1 1 17 0 11] + [ 1 4 2 1 4 2 0 0 0 0 0 10 3 0 0 1070 14 11 1 8 3] + [ 2 15 2 0 9 7 1 0 0 1 0 6 0 0 0 10 1096 0 1 4 7] + [ 1 0 1 3 0 1 1 0 0 1 1 7 18 0 3 12 2 982 3 0 2] + [ 4 6 14 18 0 1 1 29 1 0 1 1 0 0 5 0 3 0 978 0 6] + [ 0 5 0 3 1 9 14 11 0 0 2 12 8 2 0 7 15 1 3 1053 6] + [ 198 294 249 85 130 202 77 100 153 132 203 154 335 345 139 75 245 88 156 265 4280]] + +2023-10-05 21:22:26,910 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:22:26,910 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:22:26,916 - + +2023-10-05 21:22:26,916 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:22:27,917 - Epoch: [86][ 10/ 1236] Overall Loss 0.301753 Objective Loss 0.301753 LR 0.001000 Time 0.100080 +2023-10-05 21:22:28,119 - Epoch: [86][ 20/ 1236] Overall Loss 0.305928 Objective Loss 0.305928 LR 0.001000 Time 0.060126 +2023-10-05 21:22:28,321 - Epoch: [86][ 30/ 1236] Overall Loss 0.307746 Objective Loss 0.307746 LR 0.001000 Time 0.046787 +2023-10-05 21:22:28,523 - Epoch: [86][ 40/ 1236] Overall Loss 0.309197 Objective Loss 0.309197 LR 0.001000 Time 0.040135 +2023-10-05 21:22:28,725 - Epoch: [86][ 50/ 1236] Overall Loss 0.311566 Objective Loss 0.311566 LR 0.001000 Time 0.036133 +2023-10-05 21:22:28,926 - Epoch: [86][ 60/ 1236] Overall Loss 0.309359 Objective Loss 0.309359 LR 0.001000 Time 0.033464 +2023-10-05 21:22:29,125 - Epoch: [86][ 70/ 1236] Overall Loss 0.306101 Objective Loss 0.306101 LR 0.001000 Time 0.031518 +2023-10-05 21:22:29,325 - Epoch: [86][ 80/ 1236] Overall Loss 0.304963 Objective Loss 0.304963 LR 0.001000 Time 0.030080 +2023-10-05 21:22:29,524 - Epoch: [86][ 90/ 1236] Overall Loss 0.303609 Objective Loss 0.303609 LR 0.001000 Time 0.028941 +2023-10-05 21:22:29,724 - Epoch: [86][ 100/ 1236] Overall Loss 0.302797 Objective Loss 0.302797 LR 0.001000 Time 0.028048 +2023-10-05 21:22:29,923 - Epoch: [86][ 110/ 1236] Overall Loss 0.304415 Objective Loss 0.304415 LR 0.001000 Time 0.027300 +2023-10-05 21:22:30,123 - Epoch: [86][ 120/ 1236] Overall Loss 0.307481 Objective Loss 0.307481 LR 0.001000 Time 0.026691 +2023-10-05 21:22:30,321 - Epoch: [86][ 130/ 1236] Overall Loss 0.309197 Objective Loss 0.309197 LR 0.001000 Time 0.026159 +2023-10-05 21:22:30,521 - Epoch: [86][ 140/ 1236] Overall Loss 0.310011 Objective Loss 0.310011 LR 0.001000 Time 0.025716 +2023-10-05 21:22:30,720 - Epoch: [86][ 150/ 1236] Overall Loss 0.310228 Objective Loss 0.310228 LR 0.001000 Time 0.025323 +2023-10-05 21:22:30,920 - Epoch: [86][ 160/ 1236] Overall Loss 0.310255 Objective Loss 0.310255 LR 0.001000 Time 0.024990 +2023-10-05 21:22:31,118 - Epoch: [86][ 170/ 1236] Overall Loss 0.309904 Objective Loss 0.309904 LR 0.001000 Time 0.024684 +2023-10-05 21:22:31,318 - Epoch: [86][ 180/ 1236] Overall Loss 0.308411 Objective Loss 0.308411 LR 0.001000 Time 0.024421 +2023-10-05 21:22:31,519 - Epoch: [86][ 190/ 1236] Overall Loss 0.308901 Objective Loss 0.308901 LR 0.001000 Time 0.024191 +2023-10-05 21:22:31,719 - Epoch: [86][ 200/ 1236] Overall Loss 0.309349 Objective Loss 0.309349 LR 0.001000 Time 0.023978 +2023-10-05 21:22:31,918 - Epoch: [86][ 210/ 1236] Overall Loss 0.309774 Objective Loss 0.309774 LR 0.001000 Time 0.023782 +2023-10-05 21:22:32,119 - Epoch: [86][ 220/ 1236] Overall Loss 0.310297 Objective Loss 0.310297 LR 0.001000 Time 0.023615 +2023-10-05 21:22:32,317 - Epoch: [86][ 230/ 1236] Overall Loss 0.311163 Objective Loss 0.311163 LR 0.001000 Time 0.023449 +2023-10-05 21:22:32,517 - Epoch: [86][ 240/ 1236] Overall Loss 0.312097 Objective Loss 0.312097 LR 0.001000 Time 0.023303 +2023-10-05 21:22:32,716 - Epoch: [86][ 250/ 1236] Overall Loss 0.314626 Objective Loss 0.314626 LR 0.001000 Time 0.023167 +2023-10-05 21:22:32,917 - Epoch: [86][ 260/ 1236] Overall Loss 0.314911 Objective Loss 0.314911 LR 0.001000 Time 0.023045 +2023-10-05 21:22:33,116 - Epoch: [86][ 270/ 1236] Overall Loss 0.314934 Objective Loss 0.314934 LR 0.001000 Time 0.022929 +2023-10-05 21:22:33,317 - Epoch: [86][ 280/ 1236] Overall Loss 0.315426 Objective Loss 0.315426 LR 0.001000 Time 0.022827 +2023-10-05 21:22:33,518 - Epoch: [86][ 290/ 1236] Overall Loss 0.316093 Objective Loss 0.316093 LR 0.001000 Time 0.022731 +2023-10-05 21:22:33,719 - Epoch: [86][ 300/ 1236] Overall Loss 0.317167 Objective Loss 0.317167 LR 0.001000 Time 0.022642 +2023-10-05 21:22:33,923 - Epoch: [86][ 310/ 1236] Overall Loss 0.317956 Objective Loss 0.317956 LR 0.001000 Time 0.022568 +2023-10-05 21:22:34,123 - Epoch: [86][ 320/ 1236] Overall Loss 0.317421 Objective Loss 0.317421 LR 0.001000 Time 0.022488 +2023-10-05 21:22:34,327 - Epoch: [86][ 330/ 1236] Overall Loss 0.318102 Objective Loss 0.318102 LR 0.001000 Time 0.022422 +2023-10-05 21:22:34,530 - Epoch: [86][ 340/ 1236] Overall Loss 0.318691 Objective Loss 0.318691 LR 0.001000 Time 0.022359 +2023-10-05 21:22:34,737 - Epoch: [86][ 350/ 1236] Overall Loss 0.318906 Objective Loss 0.318906 LR 0.001000 Time 0.022310 +2023-10-05 21:22:34,943 - Epoch: [86][ 360/ 1236] Overall Loss 0.319801 Objective Loss 0.319801 LR 0.001000 Time 0.022262 +2023-10-05 21:22:35,149 - Epoch: [86][ 370/ 1236] Overall Loss 0.319933 Objective Loss 0.319933 LR 0.001000 Time 0.022214 +2023-10-05 21:22:35,353 - Epoch: [86][ 380/ 1236] Overall Loss 0.320132 Objective Loss 0.320132 LR 0.001000 Time 0.022166 +2023-10-05 21:22:35,560 - Epoch: [86][ 390/ 1236] Overall Loss 0.319749 Objective Loss 0.319749 LR 0.001000 Time 0.022127 +2023-10-05 21:22:35,763 - Epoch: [86][ 400/ 1236] Overall Loss 0.319704 Objective Loss 0.319704 LR 0.001000 Time 0.022081 +2023-10-05 21:22:35,970 - Epoch: [86][ 410/ 1236] Overall Loss 0.319262 Objective Loss 0.319262 LR 0.001000 Time 0.022046 +2023-10-05 21:22:36,175 - Epoch: [86][ 420/ 1236] Overall Loss 0.319557 Objective Loss 0.319557 LR 0.001000 Time 0.022008 +2023-10-05 21:22:36,384 - Epoch: [86][ 430/ 1236] Overall Loss 0.319285 Objective Loss 0.319285 LR 0.001000 Time 0.021982 +2023-10-05 21:22:36,593 - Epoch: [86][ 440/ 1236] Overall Loss 0.319114 Objective Loss 0.319114 LR 0.001000 Time 0.021956 +2023-10-05 21:22:36,802 - Epoch: [86][ 450/ 1236] Overall Loss 0.318405 Objective Loss 0.318405 LR 0.001000 Time 0.021931 +2023-10-05 21:22:37,007 - Epoch: [86][ 460/ 1236] Overall Loss 0.318181 Objective Loss 0.318181 LR 0.001000 Time 0.021899 +2023-10-05 21:22:37,215 - Epoch: [86][ 470/ 1236] Overall Loss 0.318136 Objective Loss 0.318136 LR 0.001000 Time 0.021875 +2023-10-05 21:22:37,420 - Epoch: [86][ 480/ 1236] Overall Loss 0.317969 Objective Loss 0.317969 LR 0.001000 Time 0.021846 +2023-10-05 21:22:37,628 - Epoch: [86][ 490/ 1236] Overall Loss 0.317871 Objective Loss 0.317871 LR 0.001000 Time 0.021825 +2023-10-05 21:22:37,833 - Epoch: [86][ 500/ 1236] Overall Loss 0.318251 Objective Loss 0.318251 LR 0.001000 Time 0.021797 +2023-10-05 21:22:38,042 - Epoch: [86][ 510/ 1236] Overall Loss 0.317859 Objective Loss 0.317859 LR 0.001000 Time 0.021777 +2023-10-05 21:22:38,246 - Epoch: [86][ 520/ 1236] Overall Loss 0.318234 Objective Loss 0.318234 LR 0.001000 Time 0.021750 +2023-10-05 21:22:38,451 - Epoch: [86][ 530/ 1236] Overall Loss 0.318155 Objective Loss 0.318155 LR 0.001000 Time 0.021726 +2023-10-05 21:22:38,653 - Epoch: [86][ 540/ 1236] Overall Loss 0.317434 Objective Loss 0.317434 LR 0.001000 Time 0.021697 +2023-10-05 21:22:38,859 - Epoch: [86][ 550/ 1236] Overall Loss 0.317415 Objective Loss 0.317415 LR 0.001000 Time 0.021676 +2023-10-05 21:22:39,068 - Epoch: [86][ 560/ 1236] Overall Loss 0.317061 Objective Loss 0.317061 LR 0.001000 Time 0.021662 +2023-10-05 21:22:39,281 - Epoch: [86][ 570/ 1236] Overall Loss 0.317169 Objective Loss 0.317169 LR 0.001000 Time 0.021656 +2023-10-05 21:22:39,482 - Epoch: [86][ 580/ 1236] Overall Loss 0.317135 Objective Loss 0.317135 LR 0.001000 Time 0.021627 +2023-10-05 21:22:39,683 - Epoch: [86][ 590/ 1236] Overall Loss 0.317518 Objective Loss 0.317518 LR 0.001000 Time 0.021602 +2023-10-05 21:22:39,884 - Epoch: [86][ 600/ 1236] Overall Loss 0.318053 Objective Loss 0.318053 LR 0.001000 Time 0.021575 +2023-10-05 21:22:40,085 - Epoch: [86][ 610/ 1236] Overall Loss 0.317852 Objective Loss 0.317852 LR 0.001000 Time 0.021551 +2023-10-05 21:22:40,286 - Epoch: [86][ 620/ 1236] Overall Loss 0.318167 Objective Loss 0.318167 LR 0.001000 Time 0.021526 +2023-10-05 21:22:40,488 - Epoch: [86][ 630/ 1236] Overall Loss 0.318254 Objective Loss 0.318254 LR 0.001000 Time 0.021504 +2023-10-05 21:22:40,688 - Epoch: [86][ 640/ 1236] Overall Loss 0.318656 Objective Loss 0.318656 LR 0.001000 Time 0.021481 +2023-10-05 21:22:40,890 - Epoch: [86][ 650/ 1236] Overall Loss 0.318540 Objective Loss 0.318540 LR 0.001000 Time 0.021460 +2023-10-05 21:22:41,089 - Epoch: [86][ 660/ 1236] Overall Loss 0.318950 Objective Loss 0.318950 LR 0.001000 Time 0.021436 +2023-10-05 21:22:41,291 - Epoch: [86][ 670/ 1236] Overall Loss 0.319025 Objective Loss 0.319025 LR 0.001000 Time 0.021417 +2023-10-05 21:22:41,492 - Epoch: [86][ 680/ 1236] Overall Loss 0.318806 Objective Loss 0.318806 LR 0.001000 Time 0.021397 +2023-10-05 21:22:41,704 - Epoch: [86][ 690/ 1236] Overall Loss 0.318725 Objective Loss 0.318725 LR 0.001000 Time 0.021393 +2023-10-05 21:22:41,910 - Epoch: [86][ 700/ 1236] Overall Loss 0.319147 Objective Loss 0.319147 LR 0.001000 Time 0.021382 +2023-10-05 21:22:42,112 - Epoch: [86][ 710/ 1236] Overall Loss 0.319107 Objective Loss 0.319107 LR 0.001000 Time 0.021365 +2023-10-05 21:22:42,319 - Epoch: [86][ 720/ 1236] Overall Loss 0.319176 Objective Loss 0.319176 LR 0.001000 Time 0.021355 +2023-10-05 21:22:42,533 - Epoch: [86][ 730/ 1236] Overall Loss 0.319488 Objective Loss 0.319488 LR 0.001000 Time 0.021355 +2023-10-05 21:22:42,742 - Epoch: [86][ 740/ 1236] Overall Loss 0.319783 Objective Loss 0.319783 LR 0.001000 Time 0.021349 +2023-10-05 21:22:42,956 - Epoch: [86][ 750/ 1236] Overall Loss 0.319834 Objective Loss 0.319834 LR 0.001000 Time 0.021349 +2023-10-05 21:22:43,166 - Epoch: [86][ 760/ 1236] Overall Loss 0.320065 Objective Loss 0.320065 LR 0.001000 Time 0.021343 +2023-10-05 21:22:43,380 - Epoch: [86][ 770/ 1236] Overall Loss 0.319993 Objective Loss 0.319993 LR 0.001000 Time 0.021344 +2023-10-05 21:22:43,591 - Epoch: [86][ 780/ 1236] Overall Loss 0.319784 Objective Loss 0.319784 LR 0.001000 Time 0.021340 +2023-10-05 21:22:43,804 - Epoch: [86][ 790/ 1236] Overall Loss 0.319644 Objective Loss 0.319644 LR 0.001000 Time 0.021340 +2023-10-05 21:22:44,014 - Epoch: [86][ 800/ 1236] Overall Loss 0.319455 Objective Loss 0.319455 LR 0.001000 Time 0.021335 +2023-10-05 21:22:44,228 - Epoch: [86][ 810/ 1236] Overall Loss 0.319349 Objective Loss 0.319349 LR 0.001000 Time 0.021335 +2023-10-05 21:22:44,437 - Epoch: [86][ 820/ 1236] Overall Loss 0.319468 Objective Loss 0.319468 LR 0.001000 Time 0.021329 +2023-10-05 21:22:44,651 - Epoch: [86][ 830/ 1236] Overall Loss 0.319436 Objective Loss 0.319436 LR 0.001000 Time 0.021330 +2023-10-05 21:22:44,860 - Epoch: [86][ 840/ 1236] Overall Loss 0.319609 Objective Loss 0.319609 LR 0.001000 Time 0.021324 +2023-10-05 21:22:45,073 - Epoch: [86][ 850/ 1236] Overall Loss 0.319385 Objective Loss 0.319385 LR 0.001000 Time 0.021324 +2023-10-05 21:22:45,282 - Epoch: [86][ 860/ 1236] Overall Loss 0.319319 Objective Loss 0.319319 LR 0.001000 Time 0.021319 +2023-10-05 21:22:45,498 - Epoch: [86][ 870/ 1236] Overall Loss 0.319453 Objective Loss 0.319453 LR 0.001000 Time 0.021321 +2023-10-05 21:22:45,714 - Epoch: [86][ 880/ 1236] Overall Loss 0.319256 Objective Loss 0.319256 LR 0.001000 Time 0.021324 +2023-10-05 21:22:45,930 - Epoch: [86][ 890/ 1236] Overall Loss 0.319532 Objective Loss 0.319532 LR 0.001000 Time 0.021327 +2023-10-05 21:22:46,140 - Epoch: [86][ 900/ 1236] Overall Loss 0.319475 Objective Loss 0.319475 LR 0.001000 Time 0.021323 +2023-10-05 21:22:46,348 - Epoch: [86][ 910/ 1236] Overall Loss 0.319306 Objective Loss 0.319306 LR 0.001000 Time 0.021317 +2023-10-05 21:22:46,556 - Epoch: [86][ 920/ 1236] Overall Loss 0.319089 Objective Loss 0.319089 LR 0.001000 Time 0.021311 +2023-10-05 21:22:46,764 - Epoch: [86][ 930/ 1236] Overall Loss 0.319031 Objective Loss 0.319031 LR 0.001000 Time 0.021305 +2023-10-05 21:22:46,971 - Epoch: [86][ 940/ 1236] Overall Loss 0.318677 Objective Loss 0.318677 LR 0.001000 Time 0.021297 +2023-10-05 21:22:47,178 - Epoch: [86][ 950/ 1236] Overall Loss 0.318881 Objective Loss 0.318881 LR 0.001000 Time 0.021291 +2023-10-05 21:22:47,387 - Epoch: [86][ 960/ 1236] Overall Loss 0.318784 Objective Loss 0.318784 LR 0.001000 Time 0.021286 +2023-10-05 21:22:47,599 - Epoch: [86][ 970/ 1236] Overall Loss 0.318693 Objective Loss 0.318693 LR 0.001000 Time 0.021284 +2023-10-05 21:22:47,808 - Epoch: [86][ 980/ 1236] Overall Loss 0.318977 Objective Loss 0.318977 LR 0.001000 Time 0.021279 +2023-10-05 21:22:48,018 - Epoch: [86][ 990/ 1236] Overall Loss 0.319084 Objective Loss 0.319084 LR 0.001000 Time 0.021276 +2023-10-05 21:22:48,227 - Epoch: [86][ 1000/ 1236] Overall Loss 0.319114 Objective Loss 0.319114 LR 0.001000 Time 0.021271 +2023-10-05 21:22:48,435 - Epoch: [86][ 1010/ 1236] Overall Loss 0.319336 Objective Loss 0.319336 LR 0.001000 Time 0.021266 +2023-10-05 21:22:48,643 - Epoch: [86][ 1020/ 1236] Overall Loss 0.319380 Objective Loss 0.319380 LR 0.001000 Time 0.021261 +2023-10-05 21:22:48,853 - Epoch: [86][ 1030/ 1236] Overall Loss 0.319564 Objective Loss 0.319564 LR 0.001000 Time 0.021258 +2023-10-05 21:22:49,061 - Epoch: [86][ 1040/ 1236] Overall Loss 0.319917 Objective Loss 0.319917 LR 0.001000 Time 0.021253 +2023-10-05 21:22:49,270 - Epoch: [86][ 1050/ 1236] Overall Loss 0.319976 Objective Loss 0.319976 LR 0.001000 Time 0.021250 +2023-10-05 21:22:49,481 - Epoch: [86][ 1060/ 1236] Overall Loss 0.320223 Objective Loss 0.320223 LR 0.001000 Time 0.021248 +2023-10-05 21:22:49,691 - Epoch: [86][ 1070/ 1236] Overall Loss 0.320408 Objective Loss 0.320408 LR 0.001000 Time 0.021244 +2023-10-05 21:22:49,900 - Epoch: [86][ 1080/ 1236] Overall Loss 0.320587 Objective Loss 0.320587 LR 0.001000 Time 0.021240 +2023-10-05 21:22:50,109 - Epoch: [86][ 1090/ 1236] Overall Loss 0.320459 Objective Loss 0.320459 LR 0.001000 Time 0.021237 +2023-10-05 21:22:50,318 - Epoch: [86][ 1100/ 1236] Overall Loss 0.320433 Objective Loss 0.320433 LR 0.001000 Time 0.021233 +2023-10-05 21:22:50,528 - Epoch: [86][ 1110/ 1236] Overall Loss 0.320454 Objective Loss 0.320454 LR 0.001000 Time 0.021230 +2023-10-05 21:22:50,736 - Epoch: [86][ 1120/ 1236] Overall Loss 0.320331 Objective Loss 0.320331 LR 0.001000 Time 0.021227 +2023-10-05 21:22:50,945 - Epoch: [86][ 1130/ 1236] Overall Loss 0.320529 Objective Loss 0.320529 LR 0.001000 Time 0.021223 +2023-10-05 21:22:51,154 - Epoch: [86][ 1140/ 1236] Overall Loss 0.320704 Objective Loss 0.320704 LR 0.001000 Time 0.021219 +2023-10-05 21:22:51,363 - Epoch: [86][ 1150/ 1236] Overall Loss 0.320697 Objective Loss 0.320697 LR 0.001000 Time 0.021217 +2023-10-05 21:22:51,579 - Epoch: [86][ 1160/ 1236] Overall Loss 0.320823 Objective Loss 0.320823 LR 0.001000 Time 0.021219 +2023-10-05 21:22:51,791 - Epoch: [86][ 1170/ 1236] Overall Loss 0.321201 Objective Loss 0.321201 LR 0.001000 Time 0.021219 +2023-10-05 21:22:52,000 - Epoch: [86][ 1180/ 1236] Overall Loss 0.320960 Objective Loss 0.320960 LR 0.001000 Time 0.021215 +2023-10-05 21:22:52,210 - Epoch: [86][ 1190/ 1236] Overall Loss 0.321062 Objective Loss 0.321062 LR 0.001000 Time 0.021213 +2023-10-05 21:22:52,420 - Epoch: [86][ 1200/ 1236] Overall Loss 0.321322 Objective Loss 0.321322 LR 0.001000 Time 0.021212 +2023-10-05 21:22:52,639 - Epoch: [86][ 1210/ 1236] Overall Loss 0.321219 Objective Loss 0.321219 LR 0.001000 Time 0.021217 +2023-10-05 21:22:52,849 - Epoch: [86][ 1220/ 1236] Overall Loss 0.321281 Objective Loss 0.321281 LR 0.001000 Time 0.021215 +2023-10-05 21:22:53,112 - Epoch: [86][ 1230/ 1236] Overall Loss 0.321612 Objective Loss 0.321612 LR 0.001000 Time 0.021255 +2023-10-05 21:22:53,230 - Epoch: [86][ 1236/ 1236] Overall Loss 0.321851 Objective Loss 0.321851 Top1 82.281059 Top5 97.352342 LR 0.001000 Time 0.021248 +2023-10-05 21:22:53,353 - --- validate (epoch=86)----------- +2023-10-05 21:22:53,353 - 29943 samples (256 per mini-batch) +2023-10-05 21:22:53,826 - Epoch: [86][ 10/ 117] Loss 0.385422 Top1 82.109375 Top5 97.070312 +2023-10-05 21:22:53,974 - Epoch: [86][ 20/ 117] Loss 0.375305 Top1 82.128906 Top5 97.265625 +2023-10-05 21:22:54,121 - Epoch: [86][ 30/ 117] Loss 0.380635 Top1 81.901042 Top5 97.135417 +2023-10-05 21:22:54,267 - Epoch: [86][ 40/ 117] Loss 0.372613 Top1 81.933594 Top5 97.148438 +2023-10-05 21:22:54,411 - Epoch: [86][ 50/ 117] Loss 0.365324 Top1 81.898438 Top5 97.242188 +2023-10-05 21:22:54,557 - Epoch: [86][ 60/ 117] Loss 0.361198 Top1 81.751302 Top5 97.220052 +2023-10-05 21:22:54,700 - Epoch: [86][ 70/ 117] Loss 0.355354 Top1 81.718750 Top5 97.254464 +2023-10-05 21:22:54,842 - Epoch: [86][ 80/ 117] Loss 0.355982 Top1 81.606445 Top5 97.285156 +2023-10-05 21:22:54,985 - Epoch: [86][ 90/ 117] Loss 0.360082 Top1 81.419271 Top5 97.230903 +2023-10-05 21:22:55,127 - Epoch: [86][ 100/ 117] Loss 0.363969 Top1 81.195312 Top5 97.277344 +2023-10-05 21:22:55,277 - Epoch: [86][ 110/ 117] Loss 0.360405 Top1 81.232244 Top5 97.286932 +2023-10-05 21:22:55,364 - Epoch: [86][ 117/ 117] Loss 0.358467 Top1 81.204288 Top5 97.338276 +2023-10-05 21:22:55,497 - ==> Top1: 81.204 Top5: 97.338 Loss: 0.358 + +2023-10-05 21:22:55,498 - ==> Confusion: +[[ 926 0 5 5 7 2 0 0 5 61 2 2 0 2 9 2 6 4 1 2 9] + [ 1 1033 4 1 12 18 0 21 2 0 3 6 0 0 0 2 6 1 14 1 6] + [ 4 0 932 35 2 0 35 9 0 0 6 2 8 0 1 3 3 2 4 1 9] + [ 3 1 13 958 0 4 2 1 4 0 7 0 9 5 45 1 1 9 17 0 9] + [ 23 3 2 0 966 6 0 1 0 6 1 6 2 4 13 3 11 2 0 0 1] + [ 5 53 0 3 3 954 2 25 5 2 1 10 2 18 10 1 5 0 2 4 11] + [ 0 3 25 1 0 1 1128 8 0 0 2 3 2 0 1 8 0 0 1 6 2] + [ 4 25 21 0 2 26 8 1048 3 0 6 8 6 2 0 0 0 1 46 8 4] + [ 18 3 0 0 0 2 0 2 946 48 14 5 2 11 30 1 1 2 2 0 2] + [ 116 0 3 0 8 6 1 0 27 917 0 3 2 20 4 2 1 4 0 0 5] + [ 1 6 13 16 0 3 7 2 19 3 948 3 0 8 5 0 1 2 9 1 6] + [ 2 0 1 1 0 10 1 6 0 0 0 959 20 5 0 1 5 15 0 8 1] + [ 0 5 4 9 3 0 1 0 2 0 0 52 951 4 7 3 3 13 2 4 5] + [ 2 1 2 2 1 8 0 1 13 15 6 5 3 1023 8 3 6 2 0 4 14] + [ 14 4 2 13 5 0 0 0 21 5 0 1 1 1 1014 0 2 1 8 0 9] + [ 2 4 5 1 4 1 0 0 0 0 0 11 8 0 1 1060 15 11 1 4 6] + [ 4 11 3 2 4 4 1 0 4 0 0 7 0 0 4 9 1097 0 1 3 7] + [ 0 1 0 1 0 0 1 0 1 1 0 4 23 1 3 5 1 994 0 0 2] + [ 3 9 9 19 0 0 2 19 2 0 0 0 5 0 31 0 1 0 960 1 7] + [ 0 4 5 3 1 6 9 19 0 0 0 19 7 2 1 8 9 0 2 1046 11] + [ 158 224 164 121 138 171 44 112 132 88 202 147 397 227 269 58 256 99 220 223 4455]] + +2023-10-05 21:22:55,499 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:22:55,499 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:22:55,505 - + +2023-10-05 21:22:55,505 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:22:56,514 - Epoch: [87][ 10/ 1236] Overall Loss 0.327024 Objective Loss 0.327024 LR 0.001000 Time 0.100850 +2023-10-05 21:22:56,719 - Epoch: [87][ 20/ 1236] Overall Loss 0.326651 Objective Loss 0.326651 LR 0.001000 Time 0.060635 +2023-10-05 21:22:56,920 - Epoch: [87][ 30/ 1236] Overall Loss 0.332835 Objective Loss 0.332835 LR 0.001000 Time 0.047132 +2023-10-05 21:22:57,125 - Epoch: [87][ 40/ 1236] Overall Loss 0.337295 Objective Loss 0.337295 LR 0.001000 Time 0.040451 +2023-10-05 21:22:57,332 - Epoch: [87][ 50/ 1236] Overall Loss 0.331192 Objective Loss 0.331192 LR 0.001000 Time 0.036501 +2023-10-05 21:22:57,537 - Epoch: [87][ 60/ 1236] Overall Loss 0.329319 Objective Loss 0.329319 LR 0.001000 Time 0.033824 +2023-10-05 21:22:57,740 - Epoch: [87][ 70/ 1236] Overall Loss 0.329664 Objective Loss 0.329664 LR 0.001000 Time 0.031885 +2023-10-05 21:22:57,943 - Epoch: [87][ 80/ 1236] Overall Loss 0.331052 Objective Loss 0.331052 LR 0.001000 Time 0.030442 +2023-10-05 21:22:58,146 - Epoch: [87][ 90/ 1236] Overall Loss 0.328404 Objective Loss 0.328404 LR 0.001000 Time 0.029307 +2023-10-05 21:22:58,352 - Epoch: [87][ 100/ 1236] Overall Loss 0.327256 Objective Loss 0.327256 LR 0.001000 Time 0.028432 +2023-10-05 21:22:58,557 - Epoch: [87][ 110/ 1236] Overall Loss 0.327270 Objective Loss 0.327270 LR 0.001000 Time 0.027704 +2023-10-05 21:22:58,765 - Epoch: [87][ 120/ 1236] Overall Loss 0.328280 Objective Loss 0.328280 LR 0.001000 Time 0.027131 +2023-10-05 21:22:58,974 - Epoch: [87][ 130/ 1236] Overall Loss 0.330856 Objective Loss 0.330856 LR 0.001000 Time 0.026644 +2023-10-05 21:22:59,196 - Epoch: [87][ 140/ 1236] Overall Loss 0.329481 Objective Loss 0.329481 LR 0.001000 Time 0.026327 +2023-10-05 21:22:59,410 - Epoch: [87][ 150/ 1236] Overall Loss 0.327527 Objective Loss 0.327527 LR 0.001000 Time 0.025998 +2023-10-05 21:22:59,630 - Epoch: [87][ 160/ 1236] Overall Loss 0.328971 Objective Loss 0.328971 LR 0.001000 Time 0.025745 +2023-10-05 21:22:59,844 - Epoch: [87][ 170/ 1236] Overall Loss 0.331152 Objective Loss 0.331152 LR 0.001000 Time 0.025490 +2023-10-05 21:23:00,064 - Epoch: [87][ 180/ 1236] Overall Loss 0.333520 Objective Loss 0.333520 LR 0.001000 Time 0.025291 +2023-10-05 21:23:00,278 - Epoch: [87][ 190/ 1236] Overall Loss 0.335265 Objective Loss 0.335265 LR 0.001000 Time 0.025086 +2023-10-05 21:23:00,500 - Epoch: [87][ 200/ 1236] Overall Loss 0.334562 Objective Loss 0.334562 LR 0.001000 Time 0.024939 +2023-10-05 21:23:00,706 - Epoch: [87][ 210/ 1236] Overall Loss 0.335664 Objective Loss 0.335664 LR 0.001000 Time 0.024732 +2023-10-05 21:23:00,915 - Epoch: [87][ 220/ 1236] Overall Loss 0.335714 Objective Loss 0.335714 LR 0.001000 Time 0.024556 +2023-10-05 21:23:01,121 - Epoch: [87][ 230/ 1236] Overall Loss 0.334109 Objective Loss 0.334109 LR 0.001000 Time 0.024383 +2023-10-05 21:23:01,331 - Epoch: [87][ 240/ 1236] Overall Loss 0.332449 Objective Loss 0.332449 LR 0.001000 Time 0.024239 +2023-10-05 21:23:01,537 - Epoch: [87][ 250/ 1236] Overall Loss 0.331631 Objective Loss 0.331631 LR 0.001000 Time 0.024095 +2023-10-05 21:23:01,748 - Epoch: [87][ 260/ 1236] Overall Loss 0.328922 Objective Loss 0.328922 LR 0.001000 Time 0.023975 +2023-10-05 21:23:01,954 - Epoch: [87][ 270/ 1236] Overall Loss 0.328410 Objective Loss 0.328410 LR 0.001000 Time 0.023851 +2023-10-05 21:23:02,164 - Epoch: [87][ 280/ 1236] Overall Loss 0.328479 Objective Loss 0.328479 LR 0.001000 Time 0.023745 +2023-10-05 21:23:02,371 - Epoch: [87][ 290/ 1236] Overall Loss 0.328889 Objective Loss 0.328889 LR 0.001000 Time 0.023641 +2023-10-05 21:23:02,583 - Epoch: [87][ 300/ 1236] Overall Loss 0.329487 Objective Loss 0.329487 LR 0.001000 Time 0.023559 +2023-10-05 21:23:02,790 - Epoch: [87][ 310/ 1236] Overall Loss 0.329375 Objective Loss 0.329375 LR 0.001000 Time 0.023465 +2023-10-05 21:23:03,000 - Epoch: [87][ 320/ 1236] Overall Loss 0.329922 Objective Loss 0.329922 LR 0.001000 Time 0.023387 +2023-10-05 21:23:03,207 - Epoch: [87][ 330/ 1236] Overall Loss 0.329564 Objective Loss 0.329564 LR 0.001000 Time 0.023304 +2023-10-05 21:23:03,416 - Epoch: [87][ 340/ 1236] Overall Loss 0.330154 Objective Loss 0.330154 LR 0.001000 Time 0.023234 +2023-10-05 21:23:03,624 - Epoch: [87][ 350/ 1236] Overall Loss 0.330396 Objective Loss 0.330396 LR 0.001000 Time 0.023162 +2023-10-05 21:23:03,833 - Epoch: [87][ 360/ 1236] Overall Loss 0.330773 Objective Loss 0.330773 LR 0.001000 Time 0.023099 +2023-10-05 21:23:04,038 - Epoch: [87][ 370/ 1236] Overall Loss 0.330773 Objective Loss 0.330773 LR 0.001000 Time 0.023028 +2023-10-05 21:23:04,253 - Epoch: [87][ 380/ 1236] Overall Loss 0.331170 Objective Loss 0.331170 LR 0.001000 Time 0.022986 +2023-10-05 21:23:04,462 - Epoch: [87][ 390/ 1236] Overall Loss 0.330900 Objective Loss 0.330900 LR 0.001000 Time 0.022930 +2023-10-05 21:23:04,669 - Epoch: [87][ 400/ 1236] Overall Loss 0.330286 Objective Loss 0.330286 LR 0.001000 Time 0.022875 +2023-10-05 21:23:04,873 - Epoch: [87][ 410/ 1236] Overall Loss 0.330312 Objective Loss 0.330312 LR 0.001000 Time 0.022814 +2023-10-05 21:23:05,082 - Epoch: [87][ 420/ 1236] Overall Loss 0.329537 Objective Loss 0.329537 LR 0.001000 Time 0.022766 +2023-10-05 21:23:05,286 - Epoch: [87][ 430/ 1236] Overall Loss 0.329249 Objective Loss 0.329249 LR 0.001000 Time 0.022712 +2023-10-05 21:23:05,492 - Epoch: [87][ 440/ 1236] Overall Loss 0.329276 Objective Loss 0.329276 LR 0.001000 Time 0.022663 +2023-10-05 21:23:05,696 - Epoch: [87][ 450/ 1236] Overall Loss 0.329867 Objective Loss 0.329867 LR 0.001000 Time 0.022612 +2023-10-05 21:23:05,903 - Epoch: [87][ 460/ 1236] Overall Loss 0.330300 Objective Loss 0.330300 LR 0.001000 Time 0.022570 +2023-10-05 21:23:06,109 - Epoch: [87][ 470/ 1236] Overall Loss 0.330214 Objective Loss 0.330214 LR 0.001000 Time 0.022526 +2023-10-05 21:23:06,321 - Epoch: [87][ 480/ 1236] Overall Loss 0.330900 Objective Loss 0.330900 LR 0.001000 Time 0.022497 +2023-10-05 21:23:06,525 - Epoch: [87][ 490/ 1236] Overall Loss 0.331932 Objective Loss 0.331932 LR 0.001000 Time 0.022455 +2023-10-05 21:23:06,734 - Epoch: [87][ 500/ 1236] Overall Loss 0.331436 Objective Loss 0.331436 LR 0.001000 Time 0.022423 +2023-10-05 21:23:06,938 - Epoch: [87][ 510/ 1236] Overall Loss 0.331979 Objective Loss 0.331979 LR 0.001000 Time 0.022383 +2023-10-05 21:23:07,146 - Epoch: [87][ 520/ 1236] Overall Loss 0.332549 Objective Loss 0.332549 LR 0.001000 Time 0.022350 +2023-10-05 21:23:07,349 - Epoch: [87][ 530/ 1236] Overall Loss 0.332560 Objective Loss 0.332560 LR 0.001000 Time 0.022312 +2023-10-05 21:23:07,557 - Epoch: [87][ 540/ 1236] Overall Loss 0.332184 Objective Loss 0.332184 LR 0.001000 Time 0.022283 +2023-10-05 21:23:07,761 - Epoch: [87][ 550/ 1236] Overall Loss 0.331803 Objective Loss 0.331803 LR 0.001000 Time 0.022249 +2023-10-05 21:23:07,971 - Epoch: [87][ 560/ 1236] Overall Loss 0.332303 Objective Loss 0.332303 LR 0.001000 Time 0.022226 +2023-10-05 21:23:08,173 - Epoch: [87][ 570/ 1236] Overall Loss 0.332215 Objective Loss 0.332215 LR 0.001000 Time 0.022190 +2023-10-05 21:23:08,377 - Epoch: [87][ 580/ 1236] Overall Loss 0.331959 Objective Loss 0.331959 LR 0.001000 Time 0.022157 +2023-10-05 21:23:08,579 - Epoch: [87][ 590/ 1236] Overall Loss 0.332535 Objective Loss 0.332535 LR 0.001000 Time 0.022124 +2023-10-05 21:23:08,779 - Epoch: [87][ 600/ 1236] Overall Loss 0.332787 Objective Loss 0.332787 LR 0.001000 Time 0.022087 +2023-10-05 21:23:08,979 - Epoch: [87][ 610/ 1236] Overall Loss 0.332250 Objective Loss 0.332250 LR 0.001000 Time 0.022053 +2023-10-05 21:23:09,179 - Epoch: [87][ 620/ 1236] Overall Loss 0.332323 Objective Loss 0.332323 LR 0.001000 Time 0.022019 +2023-10-05 21:23:09,379 - Epoch: [87][ 630/ 1236] Overall Loss 0.331682 Objective Loss 0.331682 LR 0.001000 Time 0.021987 +2023-10-05 21:23:09,579 - Epoch: [87][ 640/ 1236] Overall Loss 0.331575 Objective Loss 0.331575 LR 0.001000 Time 0.021955 +2023-10-05 21:23:09,779 - Epoch: [87][ 650/ 1236] Overall Loss 0.331477 Objective Loss 0.331477 LR 0.001000 Time 0.021925 +2023-10-05 21:23:09,979 - Epoch: [87][ 660/ 1236] Overall Loss 0.331632 Objective Loss 0.331632 LR 0.001000 Time 0.021894 +2023-10-05 21:23:10,179 - Epoch: [87][ 670/ 1236] Overall Loss 0.331917 Objective Loss 0.331917 LR 0.001000 Time 0.021867 +2023-10-05 21:23:10,379 - Epoch: [87][ 680/ 1236] Overall Loss 0.332374 Objective Loss 0.332374 LR 0.001000 Time 0.021838 +2023-10-05 21:23:10,586 - Epoch: [87][ 690/ 1236] Overall Loss 0.331954 Objective Loss 0.331954 LR 0.001000 Time 0.021820 +2023-10-05 21:23:10,787 - Epoch: [87][ 700/ 1236] Overall Loss 0.331745 Objective Loss 0.331745 LR 0.001000 Time 0.021796 +2023-10-05 21:23:10,990 - Epoch: [87][ 710/ 1236] Overall Loss 0.331350 Objective Loss 0.331350 LR 0.001000 Time 0.021774 +2023-10-05 21:23:11,190 - Epoch: [87][ 720/ 1236] Overall Loss 0.330939 Objective Loss 0.330939 LR 0.001000 Time 0.021750 +2023-10-05 21:23:11,393 - Epoch: [87][ 730/ 1236] Overall Loss 0.330467 Objective Loss 0.330467 LR 0.001000 Time 0.021729 +2023-10-05 21:23:11,593 - Epoch: [87][ 740/ 1236] Overall Loss 0.329841 Objective Loss 0.329841 LR 0.001000 Time 0.021706 +2023-10-05 21:23:11,797 - Epoch: [87][ 750/ 1236] Overall Loss 0.329345 Objective Loss 0.329345 LR 0.001000 Time 0.021688 +2023-10-05 21:23:11,998 - Epoch: [87][ 760/ 1236] Overall Loss 0.329315 Objective Loss 0.329315 LR 0.001000 Time 0.021666 +2023-10-05 21:23:12,201 - Epoch: [87][ 770/ 1236] Overall Loss 0.329079 Objective Loss 0.329079 LR 0.001000 Time 0.021647 +2023-10-05 21:23:12,401 - Epoch: [87][ 780/ 1236] Overall Loss 0.328871 Objective Loss 0.328871 LR 0.001000 Time 0.021627 +2023-10-05 21:23:12,605 - Epoch: [87][ 790/ 1236] Overall Loss 0.328764 Objective Loss 0.328764 LR 0.001000 Time 0.021610 +2023-10-05 21:23:12,805 - Epoch: [87][ 800/ 1236] Overall Loss 0.328775 Objective Loss 0.328775 LR 0.001000 Time 0.021589 +2023-10-05 21:23:13,008 - Epoch: [87][ 810/ 1236] Overall Loss 0.328978 Objective Loss 0.328978 LR 0.001000 Time 0.021573 +2023-10-05 21:23:13,217 - Epoch: [87][ 820/ 1236] Overall Loss 0.329450 Objective Loss 0.329450 LR 0.001000 Time 0.021564 +2023-10-05 21:23:13,417 - Epoch: [87][ 830/ 1236] Overall Loss 0.329754 Objective Loss 0.329754 LR 0.001000 Time 0.021545 +2023-10-05 21:23:13,616 - Epoch: [87][ 840/ 1236] Overall Loss 0.330000 Objective Loss 0.330000 LR 0.001000 Time 0.021525 +2023-10-05 21:23:13,815 - Epoch: [87][ 850/ 1236] Overall Loss 0.329520 Objective Loss 0.329520 LR 0.001000 Time 0.021506 +2023-10-05 21:23:14,013 - Epoch: [87][ 860/ 1236] Overall Loss 0.329144 Objective Loss 0.329144 LR 0.001000 Time 0.021486 +2023-10-05 21:23:14,213 - Epoch: [87][ 870/ 1236] Overall Loss 0.329113 Objective Loss 0.329113 LR 0.001000 Time 0.021468 +2023-10-05 21:23:14,411 - Epoch: [87][ 880/ 1236] Overall Loss 0.329140 Objective Loss 0.329140 LR 0.001000 Time 0.021448 +2023-10-05 21:23:14,610 - Epoch: [87][ 890/ 1236] Overall Loss 0.328683 Objective Loss 0.328683 LR 0.001000 Time 0.021431 +2023-10-05 21:23:14,808 - Epoch: [87][ 900/ 1236] Overall Loss 0.328480 Objective Loss 0.328480 LR 0.001000 Time 0.021412 +2023-10-05 21:23:15,008 - Epoch: [87][ 910/ 1236] Overall Loss 0.328174 Objective Loss 0.328174 LR 0.001000 Time 0.021396 +2023-10-05 21:23:15,206 - Epoch: [87][ 920/ 1236] Overall Loss 0.327950 Objective Loss 0.327950 LR 0.001000 Time 0.021378 +2023-10-05 21:23:15,405 - Epoch: [87][ 930/ 1236] Overall Loss 0.327504 Objective Loss 0.327504 LR 0.001000 Time 0.021362 +2023-10-05 21:23:15,605 - Epoch: [87][ 940/ 1236] Overall Loss 0.327574 Objective Loss 0.327574 LR 0.001000 Time 0.021347 +2023-10-05 21:23:15,806 - Epoch: [87][ 950/ 1236] Overall Loss 0.327888 Objective Loss 0.327888 LR 0.001000 Time 0.021334 +2023-10-05 21:23:16,005 - Epoch: [87][ 960/ 1236] Overall Loss 0.328117 Objective Loss 0.328117 LR 0.001000 Time 0.021319 +2023-10-05 21:23:16,205 - Epoch: [87][ 970/ 1236] Overall Loss 0.328075 Objective Loss 0.328075 LR 0.001000 Time 0.021305 +2023-10-05 21:23:16,403 - Epoch: [87][ 980/ 1236] Overall Loss 0.328224 Objective Loss 0.328224 LR 0.001000 Time 0.021288 +2023-10-05 21:23:16,602 - Epoch: [87][ 990/ 1236] Overall Loss 0.328338 Objective Loss 0.328338 LR 0.001000 Time 0.021274 +2023-10-05 21:23:16,800 - Epoch: [87][ 1000/ 1236] Overall Loss 0.328147 Objective Loss 0.328147 LR 0.001000 Time 0.021258 +2023-10-05 21:23:17,000 - Epoch: [87][ 1010/ 1236] Overall Loss 0.328043 Objective Loss 0.328043 LR 0.001000 Time 0.021246 +2023-10-05 21:23:17,198 - Epoch: [87][ 1020/ 1236] Overall Loss 0.328177 Objective Loss 0.328177 LR 0.001000 Time 0.021231 +2023-10-05 21:23:17,397 - Epoch: [87][ 1030/ 1236] Overall Loss 0.327925 Objective Loss 0.327925 LR 0.001000 Time 0.021218 +2023-10-05 21:23:17,595 - Epoch: [87][ 1040/ 1236] Overall Loss 0.327926 Objective Loss 0.327926 LR 0.001000 Time 0.021204 +2023-10-05 21:23:17,794 - Epoch: [87][ 1050/ 1236] Overall Loss 0.327947 Objective Loss 0.327947 LR 0.001000 Time 0.021191 +2023-10-05 21:23:17,992 - Epoch: [87][ 1060/ 1236] Overall Loss 0.327736 Objective Loss 0.327736 LR 0.001000 Time 0.021178 +2023-10-05 21:23:18,194 - Epoch: [87][ 1070/ 1236] Overall Loss 0.327542 Objective Loss 0.327542 LR 0.001000 Time 0.021168 +2023-10-05 21:23:18,398 - Epoch: [87][ 1080/ 1236] Overall Loss 0.327448 Objective Loss 0.327448 LR 0.001000 Time 0.021161 +2023-10-05 21:23:18,600 - Epoch: [87][ 1090/ 1236] Overall Loss 0.327540 Objective Loss 0.327540 LR 0.001000 Time 0.021151 +2023-10-05 21:23:18,799 - Epoch: [87][ 1100/ 1236] Overall Loss 0.327422 Objective Loss 0.327422 LR 0.001000 Time 0.021140 +2023-10-05 21:23:19,000 - Epoch: [87][ 1110/ 1236] Overall Loss 0.327478 Objective Loss 0.327478 LR 0.001000 Time 0.021130 +2023-10-05 21:23:19,199 - Epoch: [87][ 1120/ 1236] Overall Loss 0.327394 Objective Loss 0.327394 LR 0.001000 Time 0.021119 +2023-10-05 21:23:19,400 - Epoch: [87][ 1130/ 1236] Overall Loss 0.327077 Objective Loss 0.327077 LR 0.001000 Time 0.021109 +2023-10-05 21:23:19,600 - Epoch: [87][ 1140/ 1236] Overall Loss 0.327023 Objective Loss 0.327023 LR 0.001000 Time 0.021099 +2023-10-05 21:23:19,802 - Epoch: [87][ 1150/ 1236] Overall Loss 0.326869 Objective Loss 0.326869 LR 0.001000 Time 0.021092 +2023-10-05 21:23:20,005 - Epoch: [87][ 1160/ 1236] Overall Loss 0.326718 Objective Loss 0.326718 LR 0.001000 Time 0.021084 +2023-10-05 21:23:20,209 - Epoch: [87][ 1170/ 1236] Overall Loss 0.326760 Objective Loss 0.326760 LR 0.001000 Time 0.021078 +2023-10-05 21:23:20,410 - Epoch: [87][ 1180/ 1236] Overall Loss 0.326957 Objective Loss 0.326957 LR 0.001000 Time 0.021070 +2023-10-05 21:23:20,611 - Epoch: [87][ 1190/ 1236] Overall Loss 0.326690 Objective Loss 0.326690 LR 0.001000 Time 0.021061 +2023-10-05 21:23:20,810 - Epoch: [87][ 1200/ 1236] Overall Loss 0.326637 Objective Loss 0.326637 LR 0.001000 Time 0.021051 +2023-10-05 21:23:21,011 - Epoch: [87][ 1210/ 1236] Overall Loss 0.326616 Objective Loss 0.326616 LR 0.001000 Time 0.021043 +2023-10-05 21:23:21,210 - Epoch: [87][ 1220/ 1236] Overall Loss 0.326469 Objective Loss 0.326469 LR 0.001000 Time 0.021033 +2023-10-05 21:23:21,462 - Epoch: [87][ 1230/ 1236] Overall Loss 0.326408 Objective Loss 0.326408 LR 0.001000 Time 0.021067 +2023-10-05 21:23:21,581 - Epoch: [87][ 1236/ 1236] Overall Loss 0.326258 Objective Loss 0.326258 Top1 87.372709 Top5 97.759674 LR 0.001000 Time 0.021061 +2023-10-05 21:23:21,711 - --- validate (epoch=87)----------- +2023-10-05 21:23:21,711 - 29943 samples (256 per mini-batch) +2023-10-05 21:23:22,177 - Epoch: [87][ 10/ 117] Loss 0.342346 Top1 81.640625 Top5 97.421875 +2023-10-05 21:23:22,333 - Epoch: [87][ 20/ 117] Loss 0.351193 Top1 81.796875 Top5 97.792969 +2023-10-05 21:23:22,477 - Epoch: [87][ 30/ 117] Loss 0.353061 Top1 81.692708 Top5 97.760417 +2023-10-05 21:23:22,623 - Epoch: [87][ 40/ 117] Loss 0.345936 Top1 81.904297 Top5 97.792969 +2023-10-05 21:23:22,767 - Epoch: [87][ 50/ 117] Loss 0.340278 Top1 82.132812 Top5 97.710938 +2023-10-05 21:23:22,912 - Epoch: [87][ 60/ 117] Loss 0.339701 Top1 82.128906 Top5 97.623698 +2023-10-05 21:23:23,054 - Epoch: [87][ 70/ 117] Loss 0.337568 Top1 82.237723 Top5 97.578125 +2023-10-05 21:23:23,198 - Epoch: [87][ 80/ 117] Loss 0.344369 Top1 82.026367 Top5 97.563477 +2023-10-05 21:23:23,339 - Epoch: [87][ 90/ 117] Loss 0.353795 Top1 81.940104 Top5 97.552083 +2023-10-05 21:23:23,481 - Epoch: [87][ 100/ 117] Loss 0.351493 Top1 82.007812 Top5 97.554688 +2023-10-05 21:23:23,631 - Epoch: [87][ 110/ 117] Loss 0.350879 Top1 82.105824 Top5 97.542614 +2023-10-05 21:23:23,717 - Epoch: [87][ 117/ 117] Loss 0.350406 Top1 82.095982 Top5 97.541996 +2023-10-05 21:23:23,855 - ==> Top1: 82.096 Top5: 97.542 Loss: 0.350 + +2023-10-05 21:23:23,856 - ==> Confusion: +[[ 913 2 8 2 8 3 0 3 10 66 0 2 3 5 5 3 3 2 0 1 11] + [ 1 1037 2 0 5 32 1 23 3 0 2 0 0 1 1 2 1 2 11 0 7] + [ 7 3 935 16 1 2 33 13 0 1 6 2 9 1 0 5 1 1 6 9 5] + [ 2 1 23 942 0 1 0 1 5 0 11 0 8 4 26 6 0 9 30 3 17] + [ 23 9 0 0 969 8 0 0 1 6 3 1 2 1 8 5 9 2 0 1 2] + [ 4 24 0 1 2 990 3 24 1 3 5 9 2 20 6 2 2 1 2 2 13] + [ 0 7 32 0 0 4 1099 13 0 0 4 1 2 1 1 7 0 4 1 9 6] + [ 4 19 11 1 1 46 6 1033 3 2 3 16 3 1 0 2 0 0 51 7 9] + [ 16 3 0 0 0 10 0 2 965 41 11 2 3 11 13 4 0 0 5 0 3] + [ 105 2 2 1 5 5 1 1 35 911 0 1 1 28 3 7 0 4 0 0 7] + [ 1 5 14 6 0 2 1 6 13 1 970 3 0 10 2 1 1 0 6 0 11] + [ 1 0 1 0 0 16 1 3 0 0 0 966 13 3 0 1 1 14 0 12 3] + [ 4 1 7 6 1 1 1 3 2 0 0 55 949 2 6 3 1 17 0 5 4] + [ 1 1 2 0 2 8 0 4 13 8 7 5 3 1047 3 2 0 2 0 3 8] + [ 16 7 4 16 12 3 1 0 28 7 3 1 2 3 974 0 0 1 12 0 11] + [ 0 1 5 1 2 1 2 0 0 0 0 8 12 1 0 1051 12 19 3 8 8] + [ 0 18 1 1 10 9 1 1 2 0 0 9 5 2 3 8 1075 0 1 6 9] + [ 0 0 0 2 0 1 2 0 0 0 1 4 27 0 2 4 0 992 1 1 1] + [ 1 11 4 17 0 1 0 23 4 0 1 4 3 0 7 0 0 0 985 0 7] + [ 0 2 6 0 1 7 4 13 0 0 1 21 7 7 0 5 6 0 6 1056 10] + [ 155 207 146 76 100 208 55 122 126 73 183 145 361 337 118 54 177 91 191 257 4723]] + +2023-10-05 21:23:23,857 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:23:23,857 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:23:23,863 - + +2023-10-05 21:23:23,863 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:23:24,974 - Epoch: [88][ 10/ 1236] Overall Loss 0.299502 Objective Loss 0.299502 LR 0.001000 Time 0.111023 +2023-10-05 21:23:25,174 - Epoch: [88][ 20/ 1236] Overall Loss 0.311916 Objective Loss 0.311916 LR 0.001000 Time 0.065508 +2023-10-05 21:23:25,373 - Epoch: [88][ 30/ 1236] Overall Loss 0.312990 Objective Loss 0.312990 LR 0.001000 Time 0.050305 +2023-10-05 21:23:25,577 - Epoch: [88][ 40/ 1236] Overall Loss 0.318857 Objective Loss 0.318857 LR 0.001000 Time 0.042813 +2023-10-05 21:23:25,777 - Epoch: [88][ 50/ 1236] Overall Loss 0.321019 Objective Loss 0.321019 LR 0.001000 Time 0.038232 +2023-10-05 21:23:25,977 - Epoch: [88][ 60/ 1236] Overall Loss 0.321807 Objective Loss 0.321807 LR 0.001000 Time 0.035200 +2023-10-05 21:23:26,176 - Epoch: [88][ 70/ 1236] Overall Loss 0.320354 Objective Loss 0.320354 LR 0.001000 Time 0.033008 +2023-10-05 21:23:26,377 - Epoch: [88][ 80/ 1236] Overall Loss 0.320360 Objective Loss 0.320360 LR 0.001000 Time 0.031381 +2023-10-05 21:23:26,576 - Epoch: [88][ 90/ 1236] Overall Loss 0.319594 Objective Loss 0.319594 LR 0.001000 Time 0.030101 +2023-10-05 21:23:26,776 - Epoch: [88][ 100/ 1236] Overall Loss 0.316574 Objective Loss 0.316574 LR 0.001000 Time 0.029094 +2023-10-05 21:23:26,983 - Epoch: [88][ 110/ 1236] Overall Loss 0.314020 Objective Loss 0.314020 LR 0.001000 Time 0.028324 +2023-10-05 21:23:27,189 - Epoch: [88][ 120/ 1236] Overall Loss 0.314738 Objective Loss 0.314738 LR 0.001000 Time 0.027678 +2023-10-05 21:23:27,394 - Epoch: [88][ 130/ 1236] Overall Loss 0.316391 Objective Loss 0.316391 LR 0.001000 Time 0.027125 +2023-10-05 21:23:27,599 - Epoch: [88][ 140/ 1236] Overall Loss 0.316134 Objective Loss 0.316134 LR 0.001000 Time 0.026645 +2023-10-05 21:23:27,805 - Epoch: [88][ 150/ 1236] Overall Loss 0.318568 Objective Loss 0.318568 LR 0.001000 Time 0.026238 +2023-10-05 21:23:28,010 - Epoch: [88][ 160/ 1236] Overall Loss 0.320133 Objective Loss 0.320133 LR 0.001000 Time 0.025875 +2023-10-05 21:23:28,214 - Epoch: [88][ 170/ 1236] Overall Loss 0.320413 Objective Loss 0.320413 LR 0.001000 Time 0.025552 +2023-10-05 21:23:28,417 - Epoch: [88][ 180/ 1236] Overall Loss 0.320055 Objective Loss 0.320055 LR 0.001000 Time 0.025259 +2023-10-05 21:23:28,621 - Epoch: [88][ 190/ 1236] Overall Loss 0.317769 Objective Loss 0.317769 LR 0.001000 Time 0.025002 +2023-10-05 21:23:28,827 - Epoch: [88][ 200/ 1236] Overall Loss 0.317183 Objective Loss 0.317183 LR 0.001000 Time 0.024780 +2023-10-05 21:23:29,031 - Epoch: [88][ 210/ 1236] Overall Loss 0.316409 Objective Loss 0.316409 LR 0.001000 Time 0.024568 +2023-10-05 21:23:29,235 - Epoch: [88][ 220/ 1236] Overall Loss 0.314984 Objective Loss 0.314984 LR 0.001000 Time 0.024377 +2023-10-05 21:23:29,440 - Epoch: [88][ 230/ 1236] Overall Loss 0.315263 Objective Loss 0.315263 LR 0.001000 Time 0.024205 +2023-10-05 21:23:29,644 - Epoch: [88][ 240/ 1236] Overall Loss 0.314419 Objective Loss 0.314419 LR 0.001000 Time 0.024046 +2023-10-05 21:23:29,848 - Epoch: [88][ 250/ 1236] Overall Loss 0.314024 Objective Loss 0.314024 LR 0.001000 Time 0.023899 +2023-10-05 21:23:30,053 - Epoch: [88][ 260/ 1236] Overall Loss 0.313842 Objective Loss 0.313842 LR 0.001000 Time 0.023766 +2023-10-05 21:23:30,257 - Epoch: [88][ 270/ 1236] Overall Loss 0.313470 Objective Loss 0.313470 LR 0.001000 Time 0.023642 +2023-10-05 21:23:30,467 - Epoch: [88][ 280/ 1236] Overall Loss 0.313698 Objective Loss 0.313698 LR 0.001000 Time 0.023546 +2023-10-05 21:23:30,677 - Epoch: [88][ 290/ 1236] Overall Loss 0.313919 Objective Loss 0.313919 LR 0.001000 Time 0.023454 +2023-10-05 21:23:30,888 - Epoch: [88][ 300/ 1236] Overall Loss 0.313119 Objective Loss 0.313119 LR 0.001000 Time 0.023377 +2023-10-05 21:23:31,098 - Epoch: [88][ 310/ 1236] Overall Loss 0.311635 Objective Loss 0.311635 LR 0.001000 Time 0.023296 +2023-10-05 21:23:31,307 - Epoch: [88][ 320/ 1236] Overall Loss 0.311388 Objective Loss 0.311388 LR 0.001000 Time 0.023222 +2023-10-05 21:23:31,516 - Epoch: [88][ 330/ 1236] Overall Loss 0.312034 Objective Loss 0.312034 LR 0.001000 Time 0.023151 +2023-10-05 21:23:31,726 - Epoch: [88][ 340/ 1236] Overall Loss 0.313123 Objective Loss 0.313123 LR 0.001000 Time 0.023086 +2023-10-05 21:23:31,935 - Epoch: [88][ 350/ 1236] Overall Loss 0.314163 Objective Loss 0.314163 LR 0.001000 Time 0.023022 +2023-10-05 21:23:32,146 - Epoch: [88][ 360/ 1236] Overall Loss 0.314116 Objective Loss 0.314116 LR 0.001000 Time 0.022966 +2023-10-05 21:23:32,353 - Epoch: [88][ 370/ 1236] Overall Loss 0.314129 Objective Loss 0.314129 LR 0.001000 Time 0.022905 +2023-10-05 21:23:32,562 - Epoch: [88][ 380/ 1236] Overall Loss 0.314985 Objective Loss 0.314985 LR 0.001000 Time 0.022851 +2023-10-05 21:23:32,772 - Epoch: [88][ 390/ 1236] Overall Loss 0.315319 Objective Loss 0.315319 LR 0.001000 Time 0.022801 +2023-10-05 21:23:32,982 - Epoch: [88][ 400/ 1236] Overall Loss 0.315273 Objective Loss 0.315273 LR 0.001000 Time 0.022755 +2023-10-05 21:23:33,190 - Epoch: [88][ 410/ 1236] Overall Loss 0.314793 Objective Loss 0.314793 LR 0.001000 Time 0.022707 +2023-10-05 21:23:33,396 - Epoch: [88][ 420/ 1236] Overall Loss 0.315200 Objective Loss 0.315200 LR 0.001000 Time 0.022656 +2023-10-05 21:23:33,603 - Epoch: [88][ 430/ 1236] Overall Loss 0.314380 Objective Loss 0.314380 LR 0.001000 Time 0.022610 +2023-10-05 21:23:33,809 - Epoch: [88][ 440/ 1236] Overall Loss 0.314368 Objective Loss 0.314368 LR 0.001000 Time 0.022562 +2023-10-05 21:23:34,017 - Epoch: [88][ 450/ 1236] Overall Loss 0.315159 Objective Loss 0.315159 LR 0.001000 Time 0.022521 +2023-10-05 21:23:34,223 - Epoch: [88][ 460/ 1236] Overall Loss 0.316358 Objective Loss 0.316358 LR 0.001000 Time 0.022478 +2023-10-05 21:23:34,430 - Epoch: [88][ 470/ 1236] Overall Loss 0.317001 Objective Loss 0.317001 LR 0.001000 Time 0.022439 +2023-10-05 21:23:34,636 - Epoch: [88][ 480/ 1236] Overall Loss 0.316901 Objective Loss 0.316901 LR 0.001000 Time 0.022401 +2023-10-05 21:23:34,843 - Epoch: [88][ 490/ 1236] Overall Loss 0.317226 Objective Loss 0.317226 LR 0.001000 Time 0.022364 +2023-10-05 21:23:35,049 - Epoch: [88][ 500/ 1236] Overall Loss 0.317048 Objective Loss 0.317048 LR 0.001000 Time 0.022329 +2023-10-05 21:23:35,256 - Epoch: [88][ 510/ 1236] Overall Loss 0.316714 Objective Loss 0.316714 LR 0.001000 Time 0.022295 +2023-10-05 21:23:35,462 - Epoch: [88][ 520/ 1236] Overall Loss 0.316837 Objective Loss 0.316837 LR 0.001000 Time 0.022262 +2023-10-05 21:23:35,668 - Epoch: [88][ 530/ 1236] Overall Loss 0.316942 Objective Loss 0.316942 LR 0.001000 Time 0.022230 +2023-10-05 21:23:35,874 - Epoch: [88][ 540/ 1236] Overall Loss 0.317282 Objective Loss 0.317282 LR 0.001000 Time 0.022198 +2023-10-05 21:23:36,081 - Epoch: [88][ 550/ 1236] Overall Loss 0.318128 Objective Loss 0.318128 LR 0.001000 Time 0.022170 +2023-10-05 21:23:36,287 - Epoch: [88][ 560/ 1236] Overall Loss 0.317754 Objective Loss 0.317754 LR 0.001000 Time 0.022141 +2023-10-05 21:23:36,493 - Epoch: [88][ 570/ 1236] Overall Loss 0.317996 Objective Loss 0.317996 LR 0.001000 Time 0.022114 +2023-10-05 21:23:36,698 - Epoch: [88][ 580/ 1236] Overall Loss 0.318006 Objective Loss 0.318006 LR 0.001000 Time 0.022086 +2023-10-05 21:23:36,906 - Epoch: [88][ 590/ 1236] Overall Loss 0.318161 Objective Loss 0.318161 LR 0.001000 Time 0.022062 +2023-10-05 21:23:37,112 - Epoch: [88][ 600/ 1236] Overall Loss 0.317999 Objective Loss 0.317999 LR 0.001000 Time 0.022037 +2023-10-05 21:23:37,320 - Epoch: [88][ 610/ 1236] Overall Loss 0.317687 Objective Loss 0.317687 LR 0.001000 Time 0.022017 +2023-10-05 21:23:37,527 - Epoch: [88][ 620/ 1236] Overall Loss 0.317572 Objective Loss 0.317572 LR 0.001000 Time 0.021994 +2023-10-05 21:23:37,734 - Epoch: [88][ 630/ 1236] Overall Loss 0.317787 Objective Loss 0.317787 LR 0.001000 Time 0.021972 +2023-10-05 21:23:37,939 - Epoch: [88][ 640/ 1236] Overall Loss 0.318091 Objective Loss 0.318091 LR 0.001000 Time 0.021949 +2023-10-05 21:23:38,146 - Epoch: [88][ 650/ 1236] Overall Loss 0.319085 Objective Loss 0.319085 LR 0.001000 Time 0.021930 +2023-10-05 21:23:38,352 - Epoch: [88][ 660/ 1236] Overall Loss 0.319049 Objective Loss 0.319049 LR 0.001000 Time 0.021908 +2023-10-05 21:23:38,558 - Epoch: [88][ 670/ 1236] Overall Loss 0.319875 Objective Loss 0.319875 LR 0.001000 Time 0.021888 +2023-10-05 21:23:38,765 - Epoch: [88][ 680/ 1236] Overall Loss 0.319523 Objective Loss 0.319523 LR 0.001000 Time 0.021869 +2023-10-05 21:23:38,972 - Epoch: [88][ 690/ 1236] Overall Loss 0.319611 Objective Loss 0.319611 LR 0.001000 Time 0.021852 +2023-10-05 21:23:39,178 - Epoch: [88][ 700/ 1236] Overall Loss 0.319990 Objective Loss 0.319990 LR 0.001000 Time 0.021834 +2023-10-05 21:23:39,385 - Epoch: [88][ 710/ 1236] Overall Loss 0.319926 Objective Loss 0.319926 LR 0.001000 Time 0.021817 +2023-10-05 21:23:39,591 - Epoch: [88][ 720/ 1236] Overall Loss 0.319968 Objective Loss 0.319968 LR 0.001000 Time 0.021799 +2023-10-05 21:23:39,798 - Epoch: [88][ 730/ 1236] Overall Loss 0.319879 Objective Loss 0.319879 LR 0.001000 Time 0.021783 +2023-10-05 21:23:40,004 - Epoch: [88][ 740/ 1236] Overall Loss 0.319452 Objective Loss 0.319452 LR 0.001000 Time 0.021767 +2023-10-05 21:23:40,210 - Epoch: [88][ 750/ 1236] Overall Loss 0.319507 Objective Loss 0.319507 LR 0.001000 Time 0.021751 +2023-10-05 21:23:40,416 - Epoch: [88][ 760/ 1236] Overall Loss 0.319316 Objective Loss 0.319316 LR 0.001000 Time 0.021735 +2023-10-05 21:23:40,623 - Epoch: [88][ 770/ 1236] Overall Loss 0.319343 Objective Loss 0.319343 LR 0.001000 Time 0.021721 +2023-10-05 21:23:40,829 - Epoch: [88][ 780/ 1236] Overall Loss 0.319301 Objective Loss 0.319301 LR 0.001000 Time 0.021706 +2023-10-05 21:23:41,036 - Epoch: [88][ 790/ 1236] Overall Loss 0.318970 Objective Loss 0.318970 LR 0.001000 Time 0.021693 +2023-10-05 21:23:41,246 - Epoch: [88][ 800/ 1236] Overall Loss 0.319025 Objective Loss 0.319025 LR 0.001000 Time 0.021683 +2023-10-05 21:23:41,463 - Epoch: [88][ 810/ 1236] Overall Loss 0.319001 Objective Loss 0.319001 LR 0.001000 Time 0.021683 +2023-10-05 21:23:41,676 - Epoch: [88][ 820/ 1236] Overall Loss 0.318965 Objective Loss 0.318965 LR 0.001000 Time 0.021678 +2023-10-05 21:23:41,893 - Epoch: [88][ 830/ 1236] Overall Loss 0.318644 Objective Loss 0.318644 LR 0.001000 Time 0.021678 +2023-10-05 21:23:42,107 - Epoch: [88][ 840/ 1236] Overall Loss 0.319067 Objective Loss 0.319067 LR 0.001000 Time 0.021674 +2023-10-05 21:23:42,324 - Epoch: [88][ 850/ 1236] Overall Loss 0.319547 Objective Loss 0.319547 LR 0.001000 Time 0.021673 +2023-10-05 21:23:42,537 - Epoch: [88][ 860/ 1236] Overall Loss 0.319395 Objective Loss 0.319395 LR 0.001000 Time 0.021669 +2023-10-05 21:23:42,754 - Epoch: [88][ 870/ 1236] Overall Loss 0.319286 Objective Loss 0.319286 LR 0.001000 Time 0.021669 +2023-10-05 21:23:42,967 - Epoch: [88][ 880/ 1236] Overall Loss 0.319353 Objective Loss 0.319353 LR 0.001000 Time 0.021664 +2023-10-05 21:23:43,181 - Epoch: [88][ 890/ 1236] Overall Loss 0.319210 Objective Loss 0.319210 LR 0.001000 Time 0.021661 +2023-10-05 21:23:43,380 - Epoch: [88][ 900/ 1236] Overall Loss 0.319566 Objective Loss 0.319566 LR 0.001000 Time 0.021640 +2023-10-05 21:23:43,582 - Epoch: [88][ 910/ 1236] Overall Loss 0.319907 Objective Loss 0.319907 LR 0.001000 Time 0.021624 +2023-10-05 21:23:43,780 - Epoch: [88][ 920/ 1236] Overall Loss 0.319898 Objective Loss 0.319898 LR 0.001000 Time 0.021604 +2023-10-05 21:23:43,979 - Epoch: [88][ 930/ 1236] Overall Loss 0.320233 Objective Loss 0.320233 LR 0.001000 Time 0.021586 +2023-10-05 21:23:44,178 - Epoch: [88][ 940/ 1236] Overall Loss 0.320209 Objective Loss 0.320209 LR 0.001000 Time 0.021568 +2023-10-05 21:23:44,378 - Epoch: [88][ 950/ 1236] Overall Loss 0.320478 Objective Loss 0.320478 LR 0.001000 Time 0.021550 +2023-10-05 21:23:44,576 - Epoch: [88][ 960/ 1236] Overall Loss 0.320538 Objective Loss 0.320538 LR 0.001000 Time 0.021533 +2023-10-05 21:23:44,776 - Epoch: [88][ 970/ 1236] Overall Loss 0.320450 Objective Loss 0.320450 LR 0.001000 Time 0.021516 +2023-10-05 21:23:44,975 - Epoch: [88][ 980/ 1236] Overall Loss 0.320378 Objective Loss 0.320378 LR 0.001000 Time 0.021499 +2023-10-05 21:23:45,175 - Epoch: [88][ 990/ 1236] Overall Loss 0.320655 Objective Loss 0.320655 LR 0.001000 Time 0.021483 +2023-10-05 21:23:45,373 - Epoch: [88][ 1000/ 1236] Overall Loss 0.320789 Objective Loss 0.320789 LR 0.001000 Time 0.021467 +2023-10-05 21:23:45,573 - Epoch: [88][ 1010/ 1236] Overall Loss 0.320651 Objective Loss 0.320651 LR 0.001000 Time 0.021452 +2023-10-05 21:23:45,772 - Epoch: [88][ 1020/ 1236] Overall Loss 0.321077 Objective Loss 0.321077 LR 0.001000 Time 0.021436 +2023-10-05 21:23:45,971 - Epoch: [88][ 1030/ 1236] Overall Loss 0.321035 Objective Loss 0.321035 LR 0.001000 Time 0.021422 +2023-10-05 21:23:46,170 - Epoch: [88][ 1040/ 1236] Overall Loss 0.321445 Objective Loss 0.321445 LR 0.001000 Time 0.021407 +2023-10-05 21:23:46,370 - Epoch: [88][ 1050/ 1236] Overall Loss 0.321422 Objective Loss 0.321422 LR 0.001000 Time 0.021393 +2023-10-05 21:23:46,569 - Epoch: [88][ 1060/ 1236] Overall Loss 0.321849 Objective Loss 0.321849 LR 0.001000 Time 0.021378 +2023-10-05 21:23:46,769 - Epoch: [88][ 1070/ 1236] Overall Loss 0.321772 Objective Loss 0.321772 LR 0.001000 Time 0.021365 +2023-10-05 21:23:46,967 - Epoch: [88][ 1080/ 1236] Overall Loss 0.321725 Objective Loss 0.321725 LR 0.001000 Time 0.021351 +2023-10-05 21:23:47,167 - Epoch: [88][ 1090/ 1236] Overall Loss 0.321502 Objective Loss 0.321502 LR 0.001000 Time 0.021338 +2023-10-05 21:23:47,366 - Epoch: [88][ 1100/ 1236] Overall Loss 0.321429 Objective Loss 0.321429 LR 0.001000 Time 0.021324 +2023-10-05 21:23:47,565 - Epoch: [88][ 1110/ 1236] Overall Loss 0.321532 Objective Loss 0.321532 LR 0.001000 Time 0.021312 +2023-10-05 21:23:47,764 - Epoch: [88][ 1120/ 1236] Overall Loss 0.321651 Objective Loss 0.321651 LR 0.001000 Time 0.021298 +2023-10-05 21:23:47,964 - Epoch: [88][ 1130/ 1236] Overall Loss 0.322052 Objective Loss 0.322052 LR 0.001000 Time 0.021287 +2023-10-05 21:23:48,163 - Epoch: [88][ 1140/ 1236] Overall Loss 0.322324 Objective Loss 0.322324 LR 0.001000 Time 0.021274 +2023-10-05 21:23:48,363 - Epoch: [88][ 1150/ 1236] Overall Loss 0.322333 Objective Loss 0.322333 LR 0.001000 Time 0.021263 +2023-10-05 21:23:48,562 - Epoch: [88][ 1160/ 1236] Overall Loss 0.322409 Objective Loss 0.322409 LR 0.001000 Time 0.021251 +2023-10-05 21:23:48,762 - Epoch: [88][ 1170/ 1236] Overall Loss 0.322334 Objective Loss 0.322334 LR 0.001000 Time 0.021240 +2023-10-05 21:23:48,960 - Epoch: [88][ 1180/ 1236] Overall Loss 0.322437 Objective Loss 0.322437 LR 0.001000 Time 0.021228 +2023-10-05 21:23:49,163 - Epoch: [88][ 1190/ 1236] Overall Loss 0.322453 Objective Loss 0.322453 LR 0.001000 Time 0.021220 +2023-10-05 21:23:49,363 - Epoch: [88][ 1200/ 1236] Overall Loss 0.322560 Objective Loss 0.322560 LR 0.001000 Time 0.021209 +2023-10-05 21:23:49,562 - Epoch: [88][ 1210/ 1236] Overall Loss 0.322592 Objective Loss 0.322592 LR 0.001000 Time 0.021198 +2023-10-05 21:23:49,765 - Epoch: [88][ 1220/ 1236] Overall Loss 0.322571 Objective Loss 0.322571 LR 0.001000 Time 0.021190 +2023-10-05 21:23:50,018 - Epoch: [88][ 1230/ 1236] Overall Loss 0.322925 Objective Loss 0.322925 LR 0.001000 Time 0.021224 +2023-10-05 21:23:50,137 - Epoch: [88][ 1236/ 1236] Overall Loss 0.322840 Objective Loss 0.322840 Top1 84.521385 Top5 97.963340 LR 0.001000 Time 0.021217 +2023-10-05 21:23:50,257 - --- validate (epoch=88)----------- +2023-10-05 21:23:50,258 - 29943 samples (256 per mini-batch) +2023-10-05 21:23:50,724 - Epoch: [88][ 10/ 117] Loss 0.349835 Top1 81.132812 Top5 97.070312 +2023-10-05 21:23:50,877 - Epoch: [88][ 20/ 117] Loss 0.361928 Top1 80.410156 Top5 97.050781 +2023-10-05 21:23:51,032 - Epoch: [88][ 30/ 117] Loss 0.362439 Top1 80.572917 Top5 97.161458 +2023-10-05 21:23:51,186 - Epoch: [88][ 40/ 117] Loss 0.362443 Top1 80.566406 Top5 97.128906 +2023-10-05 21:23:51,341 - Epoch: [88][ 50/ 117] Loss 0.373683 Top1 80.382812 Top5 97.132812 +2023-10-05 21:23:51,493 - Epoch: [88][ 60/ 117] Loss 0.372614 Top1 80.501302 Top5 97.148438 +2023-10-05 21:23:51,645 - Epoch: [88][ 70/ 117] Loss 0.369911 Top1 80.535714 Top5 97.159598 +2023-10-05 21:23:51,795 - Epoch: [88][ 80/ 117] Loss 0.369348 Top1 80.537109 Top5 97.172852 +2023-10-05 21:23:51,947 - Epoch: [88][ 90/ 117] Loss 0.366404 Top1 80.603299 Top5 97.161458 +2023-10-05 21:23:52,102 - Epoch: [88][ 100/ 117] Loss 0.366340 Top1 80.574219 Top5 97.183594 +2023-10-05 21:23:52,267 - Epoch: [88][ 110/ 117] Loss 0.367349 Top1 80.376420 Top5 97.201705 +2023-10-05 21:23:52,354 - Epoch: [88][ 117/ 117] Loss 0.365872 Top1 80.419464 Top5 97.208029 +2023-10-05 21:23:52,487 - ==> Top1: 80.419 Top5: 97.208 Loss: 0.366 + +2023-10-05 21:23:52,488 - ==> Confusion: +[[ 898 3 4 2 14 3 0 1 9 85 1 3 1 3 4 2 4 3 3 0 7] + [ 3 1022 2 1 13 21 1 36 1 0 4 0 0 0 1 3 3 0 12 0 8] + [ 6 1 926 19 1 1 48 14 0 2 5 3 8 0 0 7 1 1 1 7 5] + [ 4 3 20 950 0 4 2 1 4 2 9 0 3 1 21 6 2 10 29 4 14] + [ 24 9 2 2 966 3 1 1 1 9 0 0 0 4 9 6 10 1 0 0 2] + [ 5 60 0 2 2 940 2 29 5 1 6 14 1 16 5 2 2 0 4 6 14] + [ 0 2 21 1 1 0 1124 8 0 0 2 2 1 0 1 12 1 2 1 6 6] + [ 2 21 22 1 2 27 10 1036 1 7 2 4 6 3 1 2 0 1 53 9 8] + [ 19 3 0 0 1 1 0 1 965 54 9 0 4 11 10 3 0 0 7 0 1] + [ 98 1 1 0 8 2 2 0 35 937 0 3 0 14 3 7 1 0 0 1 6] + [ 1 5 12 10 3 0 6 5 27 0 943 4 0 13 4 0 2 1 7 2 8] + [ 2 0 2 0 2 22 1 2 0 0 0 939 30 3 1 2 0 12 0 11 6] + [ 1 2 1 8 0 0 2 1 3 2 0 40 948 2 3 8 3 28 5 6 5] + [ 2 0 1 2 3 16 0 0 20 15 6 4 1 1030 3 3 1 2 1 2 7] + [ 13 5 4 22 9 0 0 0 38 6 0 2 3 2 961 1 2 2 23 0 8] + [ 1 1 1 2 5 1 2 0 0 0 0 9 5 2 0 1062 20 13 2 5 3] + [ 1 17 1 2 10 4 0 1 3 0 0 3 0 1 2 7 1093 0 0 5 11] + [ 0 0 0 0 1 0 1 0 0 0 1 3 21 0 2 5 1 997 3 0 3] + [ 1 7 9 13 2 0 1 20 3 2 5 0 3 0 7 0 1 0 982 2 10] + [ 0 3 7 4 2 6 9 12 0 0 1 19 2 0 0 4 13 1 2 1060 7] + [ 174 266 188 88 131 153 61 140 159 143 211 145 369 312 144 67 206 132 250 265 4301]] + +2023-10-05 21:23:52,489 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:23:52,489 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:23:52,495 - + +2023-10-05 21:23:52,495 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:23:53,503 - Epoch: [89][ 10/ 1236] Overall Loss 0.311566 Objective Loss 0.311566 LR 0.001000 Time 0.100710 +2023-10-05 21:23:53,708 - Epoch: [89][ 20/ 1236] Overall Loss 0.309984 Objective Loss 0.309984 LR 0.001000 Time 0.060579 +2023-10-05 21:23:53,913 - Epoch: [89][ 30/ 1236] Overall Loss 0.310198 Objective Loss 0.310198 LR 0.001000 Time 0.047215 +2023-10-05 21:23:54,119 - Epoch: [89][ 40/ 1236] Overall Loss 0.313521 Objective Loss 0.313521 LR 0.001000 Time 0.040546 +2023-10-05 21:23:54,322 - Epoch: [89][ 50/ 1236] Overall Loss 0.316105 Objective Loss 0.316105 LR 0.001000 Time 0.036480 +2023-10-05 21:23:54,527 - Epoch: [89][ 60/ 1236] Overall Loss 0.314186 Objective Loss 0.314186 LR 0.001000 Time 0.033824 +2023-10-05 21:23:54,730 - Epoch: [89][ 70/ 1236] Overall Loss 0.314465 Objective Loss 0.314465 LR 0.001000 Time 0.031877 +2023-10-05 21:23:54,936 - Epoch: [89][ 80/ 1236] Overall Loss 0.316142 Objective Loss 0.316142 LR 0.001000 Time 0.030458 +2023-10-05 21:23:55,138 - Epoch: [89][ 90/ 1236] Overall Loss 0.315386 Objective Loss 0.315386 LR 0.001000 Time 0.029317 +2023-10-05 21:23:55,344 - Epoch: [89][ 100/ 1236] Overall Loss 0.316591 Objective Loss 0.316591 LR 0.001000 Time 0.028438 +2023-10-05 21:23:55,550 - Epoch: [89][ 110/ 1236] Overall Loss 0.316437 Objective Loss 0.316437 LR 0.001000 Time 0.027729 +2023-10-05 21:23:55,758 - Epoch: [89][ 120/ 1236] Overall Loss 0.314939 Objective Loss 0.314939 LR 0.001000 Time 0.027147 +2023-10-05 21:23:55,964 - Epoch: [89][ 130/ 1236] Overall Loss 0.315882 Objective Loss 0.315882 LR 0.001000 Time 0.026636 +2023-10-05 21:23:56,171 - Epoch: [89][ 140/ 1236] Overall Loss 0.317444 Objective Loss 0.317444 LR 0.001000 Time 0.026214 +2023-10-05 21:23:56,377 - Epoch: [89][ 150/ 1236] Overall Loss 0.319358 Objective Loss 0.319358 LR 0.001000 Time 0.025835 +2023-10-05 21:23:56,585 - Epoch: [89][ 160/ 1236] Overall Loss 0.318678 Objective Loss 0.318678 LR 0.001000 Time 0.025518 +2023-10-05 21:23:56,791 - Epoch: [89][ 170/ 1236] Overall Loss 0.318087 Objective Loss 0.318087 LR 0.001000 Time 0.025226 +2023-10-05 21:23:57,002 - Epoch: [89][ 180/ 1236] Overall Loss 0.317650 Objective Loss 0.317650 LR 0.001000 Time 0.024996 +2023-10-05 21:23:57,209 - Epoch: [89][ 190/ 1236] Overall Loss 0.317452 Objective Loss 0.317452 LR 0.001000 Time 0.024765 +2023-10-05 21:23:57,424 - Epoch: [89][ 200/ 1236] Overall Loss 0.317504 Objective Loss 0.317504 LR 0.001000 Time 0.024600 +2023-10-05 21:23:57,637 - Epoch: [89][ 210/ 1236] Overall Loss 0.317738 Objective Loss 0.317738 LR 0.001000 Time 0.024433 +2023-10-05 21:23:57,852 - Epoch: [89][ 220/ 1236] Overall Loss 0.316110 Objective Loss 0.316110 LR 0.001000 Time 0.024300 +2023-10-05 21:23:58,061 - Epoch: [89][ 230/ 1236] Overall Loss 0.316456 Objective Loss 0.316456 LR 0.001000 Time 0.024148 +2023-10-05 21:23:58,274 - Epoch: [89][ 240/ 1236] Overall Loss 0.315877 Objective Loss 0.315877 LR 0.001000 Time 0.024030 +2023-10-05 21:23:58,483 - Epoch: [89][ 250/ 1236] Overall Loss 0.318087 Objective Loss 0.318087 LR 0.001000 Time 0.023902 +2023-10-05 21:23:58,698 - Epoch: [89][ 260/ 1236] Overall Loss 0.318513 Objective Loss 0.318513 LR 0.001000 Time 0.023809 +2023-10-05 21:23:58,907 - Epoch: [89][ 270/ 1236] Overall Loss 0.320337 Objective Loss 0.320337 LR 0.001000 Time 0.023698 +2023-10-05 21:23:59,120 - Epoch: [89][ 280/ 1236] Overall Loss 0.320698 Objective Loss 0.320698 LR 0.001000 Time 0.023613 +2023-10-05 21:23:59,329 - Epoch: [89][ 290/ 1236] Overall Loss 0.319000 Objective Loss 0.319000 LR 0.001000 Time 0.023518 +2023-10-05 21:23:59,544 - Epoch: [89][ 300/ 1236] Overall Loss 0.318323 Objective Loss 0.318323 LR 0.001000 Time 0.023450 +2023-10-05 21:23:59,754 - Epoch: [89][ 310/ 1236] Overall Loss 0.317842 Objective Loss 0.317842 LR 0.001000 Time 0.023367 +2023-10-05 21:23:59,968 - Epoch: [89][ 320/ 1236] Overall Loss 0.317949 Objective Loss 0.317949 LR 0.001000 Time 0.023306 +2023-10-05 21:24:00,177 - Epoch: [89][ 330/ 1236] Overall Loss 0.317797 Objective Loss 0.317797 LR 0.001000 Time 0.023232 +2023-10-05 21:24:00,392 - Epoch: [89][ 340/ 1236] Overall Loss 0.318109 Objective Loss 0.318109 LR 0.001000 Time 0.023180 +2023-10-05 21:24:00,602 - Epoch: [89][ 350/ 1236] Overall Loss 0.317926 Objective Loss 0.317926 LR 0.001000 Time 0.023115 +2023-10-05 21:24:00,811 - Epoch: [89][ 360/ 1236] Overall Loss 0.318768 Objective Loss 0.318768 LR 0.001000 Time 0.023052 +2023-10-05 21:24:01,017 - Epoch: [89][ 370/ 1236] Overall Loss 0.319238 Objective Loss 0.319238 LR 0.001000 Time 0.022985 +2023-10-05 21:24:01,233 - Epoch: [89][ 380/ 1236] Overall Loss 0.319514 Objective Loss 0.319514 LR 0.001000 Time 0.022947 +2023-10-05 21:24:01,442 - Epoch: [89][ 390/ 1236] Overall Loss 0.320022 Objective Loss 0.320022 LR 0.001000 Time 0.022894 +2023-10-05 21:24:01,659 - Epoch: [89][ 400/ 1236] Overall Loss 0.320257 Objective Loss 0.320257 LR 0.001000 Time 0.022863 +2023-10-05 21:24:01,869 - Epoch: [89][ 410/ 1236] Overall Loss 0.319941 Objective Loss 0.319941 LR 0.001000 Time 0.022816 +2023-10-05 21:24:02,085 - Epoch: [89][ 420/ 1236] Overall Loss 0.320265 Objective Loss 0.320265 LR 0.001000 Time 0.022787 +2023-10-05 21:24:02,295 - Epoch: [89][ 430/ 1236] Overall Loss 0.319941 Objective Loss 0.319941 LR 0.001000 Time 0.022744 +2023-10-05 21:24:02,511 - Epoch: [89][ 440/ 1236] Overall Loss 0.319601 Objective Loss 0.319601 LR 0.001000 Time 0.022718 +2023-10-05 21:24:02,721 - Epoch: [89][ 450/ 1236] Overall Loss 0.319910 Objective Loss 0.319910 LR 0.001000 Time 0.022679 +2023-10-05 21:24:02,937 - Epoch: [89][ 460/ 1236] Overall Loss 0.319392 Objective Loss 0.319392 LR 0.001000 Time 0.022655 +2023-10-05 21:24:03,147 - Epoch: [89][ 470/ 1236] Overall Loss 0.319156 Objective Loss 0.319156 LR 0.001000 Time 0.022618 +2023-10-05 21:24:03,363 - Epoch: [89][ 480/ 1236] Overall Loss 0.319117 Objective Loss 0.319117 LR 0.001000 Time 0.022597 +2023-10-05 21:24:03,573 - Epoch: [89][ 490/ 1236] Overall Loss 0.319084 Objective Loss 0.319084 LR 0.001000 Time 0.022563 +2023-10-05 21:24:03,793 - Epoch: [89][ 500/ 1236] Overall Loss 0.319661 Objective Loss 0.319661 LR 0.001000 Time 0.022552 +2023-10-05 21:24:04,004 - Epoch: [89][ 510/ 1236] Overall Loss 0.319938 Objective Loss 0.319938 LR 0.001000 Time 0.022522 +2023-10-05 21:24:04,221 - Epoch: [89][ 520/ 1236] Overall Loss 0.320139 Objective Loss 0.320139 LR 0.001000 Time 0.022505 +2023-10-05 21:24:04,431 - Epoch: [89][ 530/ 1236] Overall Loss 0.320342 Objective Loss 0.320342 LR 0.001000 Time 0.022476 +2023-10-05 21:24:04,648 - Epoch: [89][ 540/ 1236] Overall Loss 0.320259 Objective Loss 0.320259 LR 0.001000 Time 0.022461 +2023-10-05 21:24:04,858 - Epoch: [89][ 550/ 1236] Overall Loss 0.320119 Objective Loss 0.320119 LR 0.001000 Time 0.022434 +2023-10-05 21:24:05,075 - Epoch: [89][ 560/ 1236] Overall Loss 0.319872 Objective Loss 0.319872 LR 0.001000 Time 0.022421 +2023-10-05 21:24:05,282 - Epoch: [89][ 570/ 1236] Overall Loss 0.319953 Objective Loss 0.319953 LR 0.001000 Time 0.022389 +2023-10-05 21:24:05,488 - Epoch: [89][ 580/ 1236] Overall Loss 0.320456 Objective Loss 0.320456 LR 0.001000 Time 0.022358 +2023-10-05 21:24:05,691 - Epoch: [89][ 590/ 1236] Overall Loss 0.321235 Objective Loss 0.321235 LR 0.001000 Time 0.022323 +2023-10-05 21:24:05,900 - Epoch: [89][ 600/ 1236] Overall Loss 0.321883 Objective Loss 0.321883 LR 0.001000 Time 0.022298 +2023-10-05 21:24:06,108 - Epoch: [89][ 610/ 1236] Overall Loss 0.321730 Objective Loss 0.321730 LR 0.001000 Time 0.022273 +2023-10-05 21:24:06,316 - Epoch: [89][ 620/ 1236] Overall Loss 0.321828 Objective Loss 0.321828 LR 0.001000 Time 0.022247 +2023-10-05 21:24:06,519 - Epoch: [89][ 630/ 1236] Overall Loss 0.321488 Objective Loss 0.321488 LR 0.001000 Time 0.022217 +2023-10-05 21:24:06,726 - Epoch: [89][ 640/ 1236] Overall Loss 0.321847 Objective Loss 0.321847 LR 0.001000 Time 0.022192 +2023-10-05 21:24:06,929 - Epoch: [89][ 650/ 1236] Overall Loss 0.321906 Objective Loss 0.321906 LR 0.001000 Time 0.022162 +2023-10-05 21:24:07,135 - Epoch: [89][ 660/ 1236] Overall Loss 0.321931 Objective Loss 0.321931 LR 0.001000 Time 0.022138 +2023-10-05 21:24:07,338 - Epoch: [89][ 670/ 1236] Overall Loss 0.321835 Objective Loss 0.321835 LR 0.001000 Time 0.022111 +2023-10-05 21:24:07,545 - Epoch: [89][ 680/ 1236] Overall Loss 0.321883 Objective Loss 0.321883 LR 0.001000 Time 0.022088 +2023-10-05 21:24:07,748 - Epoch: [89][ 690/ 1236] Overall Loss 0.321933 Objective Loss 0.321933 LR 0.001000 Time 0.022062 +2023-10-05 21:24:07,955 - Epoch: [89][ 700/ 1236] Overall Loss 0.321430 Objective Loss 0.321430 LR 0.001000 Time 0.022041 +2023-10-05 21:24:08,158 - Epoch: [89][ 710/ 1236] Overall Loss 0.321043 Objective Loss 0.321043 LR 0.001000 Time 0.022017 +2023-10-05 21:24:08,364 - Epoch: [89][ 720/ 1236] Overall Loss 0.321146 Objective Loss 0.321146 LR 0.001000 Time 0.021997 +2023-10-05 21:24:08,568 - Epoch: [89][ 730/ 1236] Overall Loss 0.321458 Objective Loss 0.321458 LR 0.001000 Time 0.021974 +2023-10-05 21:24:08,774 - Epoch: [89][ 740/ 1236] Overall Loss 0.321764 Objective Loss 0.321764 LR 0.001000 Time 0.021955 +2023-10-05 21:24:08,977 - Epoch: [89][ 750/ 1236] Overall Loss 0.321955 Objective Loss 0.321955 LR 0.001000 Time 0.021932 +2023-10-05 21:24:09,184 - Epoch: [89][ 760/ 1236] Overall Loss 0.321792 Objective Loss 0.321792 LR 0.001000 Time 0.021915 +2023-10-05 21:24:09,387 - Epoch: [89][ 770/ 1236] Overall Loss 0.321552 Objective Loss 0.321552 LR 0.001000 Time 0.021894 +2023-10-05 21:24:09,593 - Epoch: [89][ 780/ 1236] Overall Loss 0.321550 Objective Loss 0.321550 LR 0.001000 Time 0.021877 +2023-10-05 21:24:09,797 - Epoch: [89][ 790/ 1236] Overall Loss 0.321557 Objective Loss 0.321557 LR 0.001000 Time 0.021857 +2023-10-05 21:24:10,003 - Epoch: [89][ 800/ 1236] Overall Loss 0.321888 Objective Loss 0.321888 LR 0.001000 Time 0.021842 +2023-10-05 21:24:10,207 - Epoch: [89][ 810/ 1236] Overall Loss 0.321681 Objective Loss 0.321681 LR 0.001000 Time 0.021823 +2023-10-05 21:24:10,413 - Epoch: [89][ 820/ 1236] Overall Loss 0.321516 Objective Loss 0.321516 LR 0.001000 Time 0.021808 +2023-10-05 21:24:10,617 - Epoch: [89][ 830/ 1236] Overall Loss 0.321313 Objective Loss 0.321313 LR 0.001000 Time 0.021790 +2023-10-05 21:24:10,823 - Epoch: [89][ 840/ 1236] Overall Loss 0.321306 Objective Loss 0.321306 LR 0.001000 Time 0.021776 +2023-10-05 21:24:11,027 - Epoch: [89][ 850/ 1236] Overall Loss 0.321058 Objective Loss 0.321058 LR 0.001000 Time 0.021758 +2023-10-05 21:24:11,233 - Epoch: [89][ 860/ 1236] Overall Loss 0.320968 Objective Loss 0.320968 LR 0.001000 Time 0.021745 +2023-10-05 21:24:11,436 - Epoch: [89][ 870/ 1236] Overall Loss 0.321086 Objective Loss 0.321086 LR 0.001000 Time 0.021728 +2023-10-05 21:24:11,642 - Epoch: [89][ 880/ 1236] Overall Loss 0.321233 Objective Loss 0.321233 LR 0.001000 Time 0.021714 +2023-10-05 21:24:11,845 - Epoch: [89][ 890/ 1236] Overall Loss 0.321615 Objective Loss 0.321615 LR 0.001000 Time 0.021698 +2023-10-05 21:24:12,052 - Epoch: [89][ 900/ 1236] Overall Loss 0.321517 Objective Loss 0.321517 LR 0.001000 Time 0.021686 +2023-10-05 21:24:12,255 - Epoch: [89][ 910/ 1236] Overall Loss 0.321606 Objective Loss 0.321606 LR 0.001000 Time 0.021671 +2023-10-05 21:24:12,462 - Epoch: [89][ 920/ 1236] Overall Loss 0.321764 Objective Loss 0.321764 LR 0.001000 Time 0.021659 +2023-10-05 21:24:12,664 - Epoch: [89][ 930/ 1236] Overall Loss 0.322064 Objective Loss 0.322064 LR 0.001000 Time 0.021644 +2023-10-05 21:24:12,871 - Epoch: [89][ 940/ 1236] Overall Loss 0.322116 Objective Loss 0.322116 LR 0.001000 Time 0.021633 +2023-10-05 21:24:13,074 - Epoch: [89][ 950/ 1236] Overall Loss 0.322498 Objective Loss 0.322498 LR 0.001000 Time 0.021619 +2023-10-05 21:24:13,281 - Epoch: [89][ 960/ 1236] Overall Loss 0.322599 Objective Loss 0.322599 LR 0.001000 Time 0.021608 +2023-10-05 21:24:13,484 - Epoch: [89][ 970/ 1236] Overall Loss 0.322709 Objective Loss 0.322709 LR 0.001000 Time 0.021595 +2023-10-05 21:24:13,690 - Epoch: [89][ 980/ 1236] Overall Loss 0.322776 Objective Loss 0.322776 LR 0.001000 Time 0.021584 +2023-10-05 21:24:13,894 - Epoch: [89][ 990/ 1236] Overall Loss 0.322941 Objective Loss 0.322941 LR 0.001000 Time 0.021571 +2023-10-05 21:24:14,100 - Epoch: [89][ 1000/ 1236] Overall Loss 0.323177 Objective Loss 0.323177 LR 0.001000 Time 0.021562 +2023-10-05 21:24:14,305 - Epoch: [89][ 1010/ 1236] Overall Loss 0.322919 Objective Loss 0.322919 LR 0.001000 Time 0.021550 +2023-10-05 21:24:14,512 - Epoch: [89][ 1020/ 1236] Overall Loss 0.323024 Objective Loss 0.323024 LR 0.001000 Time 0.021541 +2023-10-05 21:24:14,715 - Epoch: [89][ 1030/ 1236] Overall Loss 0.322859 Objective Loss 0.322859 LR 0.001000 Time 0.021529 +2023-10-05 21:24:14,922 - Epoch: [89][ 1040/ 1236] Overall Loss 0.322766 Objective Loss 0.322766 LR 0.001000 Time 0.021521 +2023-10-05 21:24:15,127 - Epoch: [89][ 1050/ 1236] Overall Loss 0.322621 Objective Loss 0.322621 LR 0.001000 Time 0.021511 +2023-10-05 21:24:15,328 - Epoch: [89][ 1060/ 1236] Overall Loss 0.322559 Objective Loss 0.322559 LR 0.001000 Time 0.021497 +2023-10-05 21:24:15,530 - Epoch: [89][ 1070/ 1236] Overall Loss 0.322747 Objective Loss 0.322747 LR 0.001000 Time 0.021484 +2023-10-05 21:24:15,730 - Epoch: [89][ 1080/ 1236] Overall Loss 0.322941 Objective Loss 0.322941 LR 0.001000 Time 0.021471 +2023-10-05 21:24:15,933 - Epoch: [89][ 1090/ 1236] Overall Loss 0.323261 Objective Loss 0.323261 LR 0.001000 Time 0.021460 +2023-10-05 21:24:16,134 - Epoch: [89][ 1100/ 1236] Overall Loss 0.323284 Objective Loss 0.323284 LR 0.001000 Time 0.021447 +2023-10-05 21:24:16,336 - Epoch: [89][ 1110/ 1236] Overall Loss 0.323124 Objective Loss 0.323124 LR 0.001000 Time 0.021436 +2023-10-05 21:24:16,537 - Epoch: [89][ 1120/ 1236] Overall Loss 0.323180 Objective Loss 0.323180 LR 0.001000 Time 0.021423 +2023-10-05 21:24:16,739 - Epoch: [89][ 1130/ 1236] Overall Loss 0.322870 Objective Loss 0.322870 LR 0.001000 Time 0.021412 +2023-10-05 21:24:16,940 - Epoch: [89][ 1140/ 1236] Overall Loss 0.322932 Objective Loss 0.322932 LR 0.001000 Time 0.021400 +2023-10-05 21:24:17,142 - Epoch: [89][ 1150/ 1236] Overall Loss 0.322876 Objective Loss 0.322876 LR 0.001000 Time 0.021390 +2023-10-05 21:24:17,343 - Epoch: [89][ 1160/ 1236] Overall Loss 0.322798 Objective Loss 0.322798 LR 0.001000 Time 0.021378 +2023-10-05 21:24:17,546 - Epoch: [89][ 1170/ 1236] Overall Loss 0.322883 Objective Loss 0.322883 LR 0.001000 Time 0.021368 +2023-10-05 21:24:17,746 - Epoch: [89][ 1180/ 1236] Overall Loss 0.322929 Objective Loss 0.322929 LR 0.001000 Time 0.021357 +2023-10-05 21:24:17,949 - Epoch: [89][ 1190/ 1236] Overall Loss 0.323098 Objective Loss 0.323098 LR 0.001000 Time 0.021348 +2023-10-05 21:24:18,149 - Epoch: [89][ 1200/ 1236] Overall Loss 0.323241 Objective Loss 0.323241 LR 0.001000 Time 0.021337 +2023-10-05 21:24:18,351 - Epoch: [89][ 1210/ 1236] Overall Loss 0.323348 Objective Loss 0.323348 LR 0.001000 Time 0.021327 +2023-10-05 21:24:18,552 - Epoch: [89][ 1220/ 1236] Overall Loss 0.323490 Objective Loss 0.323490 LR 0.001000 Time 0.021316 +2023-10-05 21:24:18,804 - Epoch: [89][ 1230/ 1236] Overall Loss 0.323627 Objective Loss 0.323627 LR 0.001000 Time 0.021348 +2023-10-05 21:24:18,921 - Epoch: [89][ 1236/ 1236] Overall Loss 0.323488 Objective Loss 0.323488 Top1 83.299389 Top5 97.556008 LR 0.001000 Time 0.021339 +2023-10-05 21:24:19,053 - --- validate (epoch=89)----------- +2023-10-05 21:24:19,054 - 29943 samples (256 per mini-batch) +2023-10-05 21:24:19,503 - Epoch: [89][ 10/ 117] Loss 0.359766 Top1 80.859375 Top5 97.539062 +2023-10-05 21:24:19,652 - Epoch: [89][ 20/ 117] Loss 0.357131 Top1 81.464844 Top5 97.656250 +2023-10-05 21:24:19,801 - Epoch: [89][ 30/ 117] Loss 0.359040 Top1 81.510417 Top5 97.604167 +2023-10-05 21:24:19,949 - Epoch: [89][ 40/ 117] Loss 0.373268 Top1 80.732422 Top5 97.343750 +2023-10-05 21:24:20,097 - Epoch: [89][ 50/ 117] Loss 0.369057 Top1 80.960938 Top5 97.312500 +2023-10-05 21:24:20,245 - Epoch: [89][ 60/ 117] Loss 0.367274 Top1 80.826823 Top5 97.291667 +2023-10-05 21:24:20,394 - Epoch: [89][ 70/ 117] Loss 0.369755 Top1 80.825893 Top5 97.310268 +2023-10-05 21:24:20,544 - Epoch: [89][ 80/ 117] Loss 0.370323 Top1 80.903320 Top5 97.319336 +2023-10-05 21:24:20,695 - Epoch: [89][ 90/ 117] Loss 0.366427 Top1 81.054688 Top5 97.330729 +2023-10-05 21:24:20,844 - Epoch: [89][ 100/ 117] Loss 0.366667 Top1 81.066406 Top5 97.328125 +2023-10-05 21:24:21,001 - Epoch: [89][ 110/ 117] Loss 0.367088 Top1 81.072443 Top5 97.357955 +2023-10-05 21:24:21,085 - Epoch: [89][ 117/ 117] Loss 0.364176 Top1 81.100758 Top5 97.364993 +2023-10-05 21:24:21,228 - ==> Top1: 81.101 Top5: 97.365 Loss: 0.364 + +2023-10-05 21:24:21,229 - ==> Confusion: +[[ 943 2 1 1 9 3 2 2 6 56 1 2 2 4 3 0 3 1 1 0 8] + [ 3 999 0 2 15 47 4 24 4 0 1 2 0 1 0 3 2 1 6 4 13] + [ 12 2 937 16 2 2 34 10 0 1 11 4 4 1 2 1 0 0 5 6 6] + [ 3 1 18 948 1 6 1 1 2 0 2 1 11 6 36 6 0 5 23 2 16] + [ 34 6 0 0 962 5 0 4 2 6 2 4 1 3 7 3 3 4 1 1 2] + [ 4 19 0 2 3 994 2 19 0 4 4 16 1 16 6 1 1 3 4 3 14] + [ 1 5 24 0 1 1 1115 12 0 0 2 6 1 0 1 9 0 2 3 5 3] + [ 10 15 8 0 2 40 8 1059 1 1 7 6 2 2 0 2 0 3 38 8 6] + [ 20 0 0 2 2 1 3 0 960 52 8 1 1 18 12 1 0 0 3 3 2] + [ 128 0 4 0 6 1 1 1 23 906 0 3 0 24 4 6 0 0 0 2 10] + [ 3 5 12 8 1 1 1 7 26 2 940 4 0 17 7 3 0 1 3 3 9] + [ 2 0 2 0 2 10 1 1 0 0 0 968 22 5 0 1 0 10 0 9 2] + [ 3 1 5 5 0 1 2 1 2 1 3 57 944 0 1 11 3 19 2 2 5] + [ 3 0 3 0 1 8 0 2 10 17 4 3 4 1045 3 3 0 3 0 2 8] + [ 21 2 3 9 6 0 0 0 25 6 2 1 0 3 1000 0 0 1 14 0 8] + [ 1 3 1 1 5 1 7 0 0 0 1 10 7 0 0 1060 10 17 0 3 7] + [ 4 22 1 1 12 7 1 3 2 0 0 12 2 2 4 10 1061 0 0 4 13] + [ 0 0 1 1 0 0 2 0 0 0 0 13 20 1 2 8 0 986 1 1 2] + [ 4 10 8 10 2 2 3 22 6 1 4 0 5 1 15 0 0 0 962 3 10] + [ 1 5 4 1 2 11 7 14 0 0 3 19 5 3 0 8 12 2 1 1049 5] + [ 223 144 167 68 123 226 72 144 113 116 188 205 363 416 169 78 99 105 183 257 4446]] + +2023-10-05 21:24:21,230 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:24:21,230 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:24:21,236 - + +2023-10-05 21:24:21,236 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:24:22,224 - Epoch: [90][ 10/ 1236] Overall Loss 0.293303 Objective Loss 0.293303 LR 0.001000 Time 0.098696 +2023-10-05 21:24:22,426 - Epoch: [90][ 20/ 1236] Overall Loss 0.290394 Objective Loss 0.290394 LR 0.001000 Time 0.059465 +2023-10-05 21:24:22,628 - Epoch: [90][ 30/ 1236] Overall Loss 0.299549 Objective Loss 0.299549 LR 0.001000 Time 0.046367 +2023-10-05 21:24:22,831 - Epoch: [90][ 40/ 1236] Overall Loss 0.297351 Objective Loss 0.297351 LR 0.001000 Time 0.039837 +2023-10-05 21:24:23,033 - Epoch: [90][ 50/ 1236] Overall Loss 0.296202 Objective Loss 0.296202 LR 0.001000 Time 0.035907 +2023-10-05 21:24:23,236 - Epoch: [90][ 60/ 1236] Overall Loss 0.299793 Objective Loss 0.299793 LR 0.001000 Time 0.033295 +2023-10-05 21:24:23,438 - Epoch: [90][ 70/ 1236] Overall Loss 0.301281 Objective Loss 0.301281 LR 0.001000 Time 0.031423 +2023-10-05 21:24:23,641 - Epoch: [90][ 80/ 1236] Overall Loss 0.300620 Objective Loss 0.300620 LR 0.001000 Time 0.030024 +2023-10-05 21:24:23,844 - Epoch: [90][ 90/ 1236] Overall Loss 0.302238 Objective Loss 0.302238 LR 0.001000 Time 0.028932 +2023-10-05 21:24:24,046 - Epoch: [90][ 100/ 1236] Overall Loss 0.303440 Objective Loss 0.303440 LR 0.001000 Time 0.028057 +2023-10-05 21:24:24,245 - Epoch: [90][ 110/ 1236] Overall Loss 0.303273 Objective Loss 0.303273 LR 0.001000 Time 0.027317 +2023-10-05 21:24:24,446 - Epoch: [90][ 120/ 1236] Overall Loss 0.302164 Objective Loss 0.302164 LR 0.001000 Time 0.026712 +2023-10-05 21:24:24,646 - Epoch: [90][ 130/ 1236] Overall Loss 0.305157 Objective Loss 0.305157 LR 0.001000 Time 0.026195 +2023-10-05 21:24:24,847 - Epoch: [90][ 140/ 1236] Overall Loss 0.303623 Objective Loss 0.303623 LR 0.001000 Time 0.025755 +2023-10-05 21:24:25,047 - Epoch: [90][ 150/ 1236] Overall Loss 0.301130 Objective Loss 0.301130 LR 0.001000 Time 0.025366 +2023-10-05 21:24:25,248 - Epoch: [90][ 160/ 1236] Overall Loss 0.301946 Objective Loss 0.301946 LR 0.001000 Time 0.025036 +2023-10-05 21:24:25,448 - Epoch: [90][ 170/ 1236] Overall Loss 0.302426 Objective Loss 0.302426 LR 0.001000 Time 0.024738 +2023-10-05 21:24:25,649 - Epoch: [90][ 180/ 1236] Overall Loss 0.302802 Objective Loss 0.302802 LR 0.001000 Time 0.024480 +2023-10-05 21:24:25,849 - Epoch: [90][ 190/ 1236] Overall Loss 0.304326 Objective Loss 0.304326 LR 0.001000 Time 0.024241 +2023-10-05 21:24:26,050 - Epoch: [90][ 200/ 1236] Overall Loss 0.303965 Objective Loss 0.303965 LR 0.001000 Time 0.024034 +2023-10-05 21:24:26,250 - Epoch: [90][ 210/ 1236] Overall Loss 0.303955 Objective Loss 0.303955 LR 0.001000 Time 0.023842 +2023-10-05 21:24:26,451 - Epoch: [90][ 220/ 1236] Overall Loss 0.305361 Objective Loss 0.305361 LR 0.001000 Time 0.023669 +2023-10-05 21:24:26,652 - Epoch: [90][ 230/ 1236] Overall Loss 0.306229 Objective Loss 0.306229 LR 0.001000 Time 0.023510 +2023-10-05 21:24:26,852 - Epoch: [90][ 240/ 1236] Overall Loss 0.307318 Objective Loss 0.307318 LR 0.001000 Time 0.023362 +2023-10-05 21:24:27,052 - Epoch: [90][ 250/ 1236] Overall Loss 0.307754 Objective Loss 0.307754 LR 0.001000 Time 0.023229 +2023-10-05 21:24:27,253 - Epoch: [90][ 260/ 1236] Overall Loss 0.309881 Objective Loss 0.309881 LR 0.001000 Time 0.023107 +2023-10-05 21:24:27,454 - Epoch: [90][ 270/ 1236] Overall Loss 0.309857 Objective Loss 0.309857 LR 0.001000 Time 0.022993 +2023-10-05 21:24:27,655 - Epoch: [90][ 280/ 1236] Overall Loss 0.309980 Objective Loss 0.309980 LR 0.001000 Time 0.022889 +2023-10-05 21:24:27,856 - Epoch: [90][ 290/ 1236] Overall Loss 0.311075 Objective Loss 0.311075 LR 0.001000 Time 0.022792 +2023-10-05 21:24:28,060 - Epoch: [90][ 300/ 1236] Overall Loss 0.310896 Objective Loss 0.310896 LR 0.001000 Time 0.022709 +2023-10-05 21:24:28,259 - Epoch: [90][ 310/ 1236] Overall Loss 0.311706 Objective Loss 0.311706 LR 0.001000 Time 0.022619 +2023-10-05 21:24:28,461 - Epoch: [90][ 320/ 1236] Overall Loss 0.313037 Objective Loss 0.313037 LR 0.001000 Time 0.022542 +2023-10-05 21:24:28,662 - Epoch: [90][ 330/ 1236] Overall Loss 0.312581 Objective Loss 0.312581 LR 0.001000 Time 0.022466 +2023-10-05 21:24:28,863 - Epoch: [90][ 340/ 1236] Overall Loss 0.312615 Objective Loss 0.312615 LR 0.001000 Time 0.022397 +2023-10-05 21:24:29,064 - Epoch: [90][ 350/ 1236] Overall Loss 0.312917 Objective Loss 0.312917 LR 0.001000 Time 0.022329 +2023-10-05 21:24:29,265 - Epoch: [90][ 360/ 1236] Overall Loss 0.313468 Objective Loss 0.313468 LR 0.001000 Time 0.022267 +2023-10-05 21:24:29,466 - Epoch: [90][ 370/ 1236] Overall Loss 0.314369 Objective Loss 0.314369 LR 0.001000 Time 0.022208 +2023-10-05 21:24:29,668 - Epoch: [90][ 380/ 1236] Overall Loss 0.314831 Objective Loss 0.314831 LR 0.001000 Time 0.022153 +2023-10-05 21:24:29,871 - Epoch: [90][ 390/ 1236] Overall Loss 0.315587 Objective Loss 0.315587 LR 0.001000 Time 0.022104 +2023-10-05 21:24:30,074 - Epoch: [90][ 400/ 1236] Overall Loss 0.316273 Objective Loss 0.316273 LR 0.001000 Time 0.022058 +2023-10-05 21:24:30,277 - Epoch: [90][ 410/ 1236] Overall Loss 0.316684 Objective Loss 0.316684 LR 0.001000 Time 0.022015 +2023-10-05 21:24:30,480 - Epoch: [90][ 420/ 1236] Overall Loss 0.317881 Objective Loss 0.317881 LR 0.001000 Time 0.021972 +2023-10-05 21:24:30,682 - Epoch: [90][ 430/ 1236] Overall Loss 0.317698 Objective Loss 0.317698 LR 0.001000 Time 0.021932 +2023-10-05 21:24:30,885 - Epoch: [90][ 440/ 1236] Overall Loss 0.317523 Objective Loss 0.317523 LR 0.001000 Time 0.021893 +2023-10-05 21:24:31,087 - Epoch: [90][ 450/ 1236] Overall Loss 0.317640 Objective Loss 0.317640 LR 0.001000 Time 0.021855 +2023-10-05 21:24:31,290 - Epoch: [90][ 460/ 1236] Overall Loss 0.318238 Objective Loss 0.318238 LR 0.001000 Time 0.021821 +2023-10-05 21:24:31,493 - Epoch: [90][ 470/ 1236] Overall Loss 0.318038 Objective Loss 0.318038 LR 0.001000 Time 0.021788 +2023-10-05 21:24:31,696 - Epoch: [90][ 480/ 1236] Overall Loss 0.317659 Objective Loss 0.317659 LR 0.001000 Time 0.021755 +2023-10-05 21:24:31,898 - Epoch: [90][ 490/ 1236] Overall Loss 0.317958 Objective Loss 0.317958 LR 0.001000 Time 0.021723 +2023-10-05 21:24:32,101 - Epoch: [90][ 500/ 1236] Overall Loss 0.317972 Objective Loss 0.317972 LR 0.001000 Time 0.021694 +2023-10-05 21:24:32,303 - Epoch: [90][ 510/ 1236] Overall Loss 0.318189 Objective Loss 0.318189 LR 0.001000 Time 0.021664 +2023-10-05 21:24:32,506 - Epoch: [90][ 520/ 1236] Overall Loss 0.318738 Objective Loss 0.318738 LR 0.001000 Time 0.021637 +2023-10-05 21:24:32,709 - Epoch: [90][ 530/ 1236] Overall Loss 0.318852 Objective Loss 0.318852 LR 0.001000 Time 0.021611 +2023-10-05 21:24:32,912 - Epoch: [90][ 540/ 1236] Overall Loss 0.319765 Objective Loss 0.319765 LR 0.001000 Time 0.021586 +2023-10-05 21:24:33,115 - Epoch: [90][ 550/ 1236] Overall Loss 0.319323 Objective Loss 0.319323 LR 0.001000 Time 0.021562 +2023-10-05 21:24:33,318 - Epoch: [90][ 560/ 1236] Overall Loss 0.319782 Objective Loss 0.319782 LR 0.001000 Time 0.021539 +2023-10-05 21:24:33,521 - Epoch: [90][ 570/ 1236] Overall Loss 0.319829 Objective Loss 0.319829 LR 0.001000 Time 0.021516 +2023-10-05 21:24:33,724 - Epoch: [90][ 580/ 1236] Overall Loss 0.320091 Objective Loss 0.320091 LR 0.001000 Time 0.021494 +2023-10-05 21:24:33,927 - Epoch: [90][ 590/ 1236] Overall Loss 0.320291 Objective Loss 0.320291 LR 0.001000 Time 0.021474 +2023-10-05 21:24:34,130 - Epoch: [90][ 600/ 1236] Overall Loss 0.320748 Objective Loss 0.320748 LR 0.001000 Time 0.021454 +2023-10-05 21:24:34,333 - Epoch: [90][ 610/ 1236] Overall Loss 0.320891 Objective Loss 0.320891 LR 0.001000 Time 0.021434 +2023-10-05 21:24:34,536 - Epoch: [90][ 620/ 1236] Overall Loss 0.320897 Objective Loss 0.320897 LR 0.001000 Time 0.021415 +2023-10-05 21:24:34,739 - Epoch: [90][ 630/ 1236] Overall Loss 0.321323 Objective Loss 0.321323 LR 0.001000 Time 0.021397 +2023-10-05 21:24:34,942 - Epoch: [90][ 640/ 1236] Overall Loss 0.321107 Objective Loss 0.321107 LR 0.001000 Time 0.021379 +2023-10-05 21:24:35,145 - Epoch: [90][ 650/ 1236] Overall Loss 0.321085 Objective Loss 0.321085 LR 0.001000 Time 0.021362 +2023-10-05 21:24:35,348 - Epoch: [90][ 660/ 1236] Overall Loss 0.321195 Objective Loss 0.321195 LR 0.001000 Time 0.021346 +2023-10-05 21:24:35,551 - Epoch: [90][ 670/ 1236] Overall Loss 0.321379 Objective Loss 0.321379 LR 0.001000 Time 0.021330 +2023-10-05 21:24:35,754 - Epoch: [90][ 680/ 1236] Overall Loss 0.321649 Objective Loss 0.321649 LR 0.001000 Time 0.021314 +2023-10-05 21:24:35,957 - Epoch: [90][ 690/ 1236] Overall Loss 0.321358 Objective Loss 0.321358 LR 0.001000 Time 0.021299 +2023-10-05 21:24:36,160 - Epoch: [90][ 700/ 1236] Overall Loss 0.320984 Objective Loss 0.320984 LR 0.001000 Time 0.021284 +2023-10-05 21:24:36,362 - Epoch: [90][ 710/ 1236] Overall Loss 0.321396 Objective Loss 0.321396 LR 0.001000 Time 0.021269 +2023-10-05 21:24:36,565 - Epoch: [90][ 720/ 1236] Overall Loss 0.321853 Objective Loss 0.321853 LR 0.001000 Time 0.021255 +2023-10-05 21:24:36,768 - Epoch: [90][ 730/ 1236] Overall Loss 0.321404 Objective Loss 0.321404 LR 0.001000 Time 0.021241 +2023-10-05 21:24:36,970 - Epoch: [90][ 740/ 1236] Overall Loss 0.321162 Objective Loss 0.321162 LR 0.001000 Time 0.021227 +2023-10-05 21:24:37,173 - Epoch: [90][ 750/ 1236] Overall Loss 0.320861 Objective Loss 0.320861 LR 0.001000 Time 0.021214 +2023-10-05 21:24:37,376 - Epoch: [90][ 760/ 1236] Overall Loss 0.320828 Objective Loss 0.320828 LR 0.001000 Time 0.021201 +2023-10-05 21:24:37,578 - Epoch: [90][ 770/ 1236] Overall Loss 0.320925 Objective Loss 0.320925 LR 0.001000 Time 0.021188 +2023-10-05 21:24:37,781 - Epoch: [90][ 780/ 1236] Overall Loss 0.320982 Objective Loss 0.320982 LR 0.001000 Time 0.021176 +2023-10-05 21:24:37,983 - Epoch: [90][ 790/ 1236] Overall Loss 0.320693 Objective Loss 0.320693 LR 0.001000 Time 0.021164 +2023-10-05 21:24:38,187 - Epoch: [90][ 800/ 1236] Overall Loss 0.320183 Objective Loss 0.320183 LR 0.001000 Time 0.021153 +2023-10-05 21:24:38,390 - Epoch: [90][ 810/ 1236] Overall Loss 0.319962 Objective Loss 0.319962 LR 0.001000 Time 0.021142 +2023-10-05 21:24:38,593 - Epoch: [90][ 820/ 1236] Overall Loss 0.320248 Objective Loss 0.320248 LR 0.001000 Time 0.021131 +2023-10-05 21:24:38,795 - Epoch: [90][ 830/ 1236] Overall Loss 0.320271 Objective Loss 0.320271 LR 0.001000 Time 0.021120 +2023-10-05 21:24:38,998 - Epoch: [90][ 840/ 1236] Overall Loss 0.320651 Objective Loss 0.320651 LR 0.001000 Time 0.021110 +2023-10-05 21:24:39,201 - Epoch: [90][ 850/ 1236] Overall Loss 0.321194 Objective Loss 0.321194 LR 0.001000 Time 0.021100 +2023-10-05 21:24:39,404 - Epoch: [90][ 860/ 1236] Overall Loss 0.321619 Objective Loss 0.321619 LR 0.001000 Time 0.021090 +2023-10-05 21:24:39,607 - Epoch: [90][ 870/ 1236] Overall Loss 0.321944 Objective Loss 0.321944 LR 0.001000 Time 0.021081 +2023-10-05 21:24:39,810 - Epoch: [90][ 880/ 1236] Overall Loss 0.322032 Objective Loss 0.322032 LR 0.001000 Time 0.021071 +2023-10-05 21:24:40,013 - Epoch: [90][ 890/ 1236] Overall Loss 0.322113 Objective Loss 0.322113 LR 0.001000 Time 0.021063 +2023-10-05 21:24:40,216 - Epoch: [90][ 900/ 1236] Overall Loss 0.322167 Objective Loss 0.322167 LR 0.001000 Time 0.021053 +2023-10-05 21:24:40,418 - Epoch: [90][ 910/ 1236] Overall Loss 0.322386 Objective Loss 0.322386 LR 0.001000 Time 0.021044 +2023-10-05 21:24:40,621 - Epoch: [90][ 920/ 1236] Overall Loss 0.322528 Objective Loss 0.322528 LR 0.001000 Time 0.021035 +2023-10-05 21:24:40,824 - Epoch: [90][ 930/ 1236] Overall Loss 0.322586 Objective Loss 0.322586 LR 0.001000 Time 0.021027 +2023-10-05 21:24:41,028 - Epoch: [90][ 940/ 1236] Overall Loss 0.323016 Objective Loss 0.323016 LR 0.001000 Time 0.021020 +2023-10-05 21:24:41,231 - Epoch: [90][ 950/ 1236] Overall Loss 0.322759 Objective Loss 0.322759 LR 0.001000 Time 0.021012 +2023-10-05 21:24:41,434 - Epoch: [90][ 960/ 1236] Overall Loss 0.322649 Objective Loss 0.322649 LR 0.001000 Time 0.021004 +2023-10-05 21:24:41,637 - Epoch: [90][ 970/ 1236] Overall Loss 0.323043 Objective Loss 0.323043 LR 0.001000 Time 0.020997 +2023-10-05 21:24:41,840 - Epoch: [90][ 980/ 1236] Overall Loss 0.323337 Objective Loss 0.323337 LR 0.001000 Time 0.020989 +2023-10-05 21:24:42,043 - Epoch: [90][ 990/ 1236] Overall Loss 0.323661 Objective Loss 0.323661 LR 0.001000 Time 0.020982 +2023-10-05 21:24:42,246 - Epoch: [90][ 1000/ 1236] Overall Loss 0.323908 Objective Loss 0.323908 LR 0.001000 Time 0.020975 +2023-10-05 21:24:42,448 - Epoch: [90][ 1010/ 1236] Overall Loss 0.323962 Objective Loss 0.323962 LR 0.001000 Time 0.020967 +2023-10-05 21:24:42,651 - Epoch: [90][ 1020/ 1236] Overall Loss 0.324094 Objective Loss 0.324094 LR 0.001000 Time 0.020960 +2023-10-05 21:24:42,853 - Epoch: [90][ 1030/ 1236] Overall Loss 0.324395 Objective Loss 0.324395 LR 0.001000 Time 0.020953 +2023-10-05 21:24:43,057 - Epoch: [90][ 1040/ 1236] Overall Loss 0.324618 Objective Loss 0.324618 LR 0.001000 Time 0.020946 +2023-10-05 21:24:43,259 - Epoch: [90][ 1050/ 1236] Overall Loss 0.324584 Objective Loss 0.324584 LR 0.001000 Time 0.020939 +2023-10-05 21:24:43,462 - Epoch: [90][ 1060/ 1236] Overall Loss 0.324868 Objective Loss 0.324868 LR 0.001000 Time 0.020933 +2023-10-05 21:24:43,665 - Epoch: [90][ 1070/ 1236] Overall Loss 0.325083 Objective Loss 0.325083 LR 0.001000 Time 0.020927 +2023-10-05 21:24:43,868 - Epoch: [90][ 1080/ 1236] Overall Loss 0.325004 Objective Loss 0.325004 LR 0.001000 Time 0.020921 +2023-10-05 21:24:44,071 - Epoch: [90][ 1090/ 1236] Overall Loss 0.325173 Objective Loss 0.325173 LR 0.001000 Time 0.020914 +2023-10-05 21:24:44,274 - Epoch: [90][ 1100/ 1236] Overall Loss 0.325345 Objective Loss 0.325345 LR 0.001000 Time 0.020909 +2023-10-05 21:24:44,477 - Epoch: [90][ 1110/ 1236] Overall Loss 0.325003 Objective Loss 0.325003 LR 0.001000 Time 0.020903 +2023-10-05 21:24:44,680 - Epoch: [90][ 1120/ 1236] Overall Loss 0.325402 Objective Loss 0.325402 LR 0.001000 Time 0.020897 +2023-10-05 21:24:44,883 - Epoch: [90][ 1130/ 1236] Overall Loss 0.325298 Objective Loss 0.325298 LR 0.001000 Time 0.020891 +2023-10-05 21:24:45,086 - Epoch: [90][ 1140/ 1236] Overall Loss 0.325320 Objective Loss 0.325320 LR 0.001000 Time 0.020886 +2023-10-05 21:24:45,289 - Epoch: [90][ 1150/ 1236] Overall Loss 0.325612 Objective Loss 0.325612 LR 0.001000 Time 0.020881 +2023-10-05 21:24:45,492 - Epoch: [90][ 1160/ 1236] Overall Loss 0.325539 Objective Loss 0.325539 LR 0.001000 Time 0.020875 +2023-10-05 21:24:45,695 - Epoch: [90][ 1170/ 1236] Overall Loss 0.325343 Objective Loss 0.325343 LR 0.001000 Time 0.020870 +2023-10-05 21:24:45,899 - Epoch: [90][ 1180/ 1236] Overall Loss 0.325509 Objective Loss 0.325509 LR 0.001000 Time 0.020865 +2023-10-05 21:24:46,101 - Epoch: [90][ 1190/ 1236] Overall Loss 0.325513 Objective Loss 0.325513 LR 0.001000 Time 0.020860 +2023-10-05 21:24:46,304 - Epoch: [90][ 1200/ 1236] Overall Loss 0.325536 Objective Loss 0.325536 LR 0.001000 Time 0.020855 +2023-10-05 21:24:46,507 - Epoch: [90][ 1210/ 1236] Overall Loss 0.325762 Objective Loss 0.325762 LR 0.001000 Time 0.020850 +2023-10-05 21:24:46,710 - Epoch: [90][ 1220/ 1236] Overall Loss 0.325908 Objective Loss 0.325908 LR 0.001000 Time 0.020845 +2023-10-05 21:24:46,965 - Epoch: [90][ 1230/ 1236] Overall Loss 0.325986 Objective Loss 0.325986 LR 0.001000 Time 0.020882 +2023-10-05 21:24:47,082 - Epoch: [90][ 1236/ 1236] Overall Loss 0.326083 Objective Loss 0.326083 Top1 81.059063 Top5 97.759674 LR 0.001000 Time 0.020876 +2023-10-05 21:24:47,215 - --- validate (epoch=90)----------- +2023-10-05 21:24:47,215 - 29943 samples (256 per mini-batch) +2023-10-05 21:24:47,675 - Epoch: [90][ 10/ 117] Loss 0.377461 Top1 82.070312 Top5 97.421875 +2023-10-05 21:24:47,834 - Epoch: [90][ 20/ 117] Loss 0.372312 Top1 82.070312 Top5 97.265625 +2023-10-05 21:24:47,994 - Epoch: [90][ 30/ 117] Loss 0.387636 Top1 81.757812 Top5 97.031250 +2023-10-05 21:24:48,153 - Epoch: [90][ 40/ 117] Loss 0.378077 Top1 81.757812 Top5 97.138672 +2023-10-05 21:24:48,312 - Epoch: [90][ 50/ 117] Loss 0.376332 Top1 81.765625 Top5 97.273438 +2023-10-05 21:24:48,468 - Epoch: [90][ 60/ 117] Loss 0.373227 Top1 81.647135 Top5 97.272135 +2023-10-05 21:24:48,625 - Epoch: [90][ 70/ 117] Loss 0.371087 Top1 81.696429 Top5 97.321429 +2023-10-05 21:24:48,781 - Epoch: [90][ 80/ 117] Loss 0.370612 Top1 81.733398 Top5 97.348633 +2023-10-05 21:24:48,939 - Epoch: [90][ 90/ 117] Loss 0.371346 Top1 81.597222 Top5 97.339410 +2023-10-05 21:24:49,094 - Epoch: [90][ 100/ 117] Loss 0.371500 Top1 81.640625 Top5 97.316406 +2023-10-05 21:24:49,255 - Epoch: [90][ 110/ 117] Loss 0.368911 Top1 81.644176 Top5 97.304688 +2023-10-05 21:24:49,340 - Epoch: [90][ 117/ 117] Loss 0.369620 Top1 81.568313 Top5 97.338276 +2023-10-05 21:24:49,477 - ==> Top1: 81.568 Top5: 97.338 Loss: 0.370 + +2023-10-05 21:24:49,478 - ==> Confusion: +[[ 892 4 3 1 12 3 0 0 5 93 1 0 1 3 2 5 5 5 2 0 13] + [ 1 1048 1 0 14 14 1 21 3 2 3 0 0 0 1 5 4 0 5 2 6] + [ 4 2 944 11 2 0 31 11 0 1 12 1 8 2 1 5 3 4 3 2 9] + [ 5 1 25 937 2 5 1 0 5 0 7 0 8 2 29 7 1 8 32 3 11] + [ 21 7 0 0 981 3 0 0 0 11 0 2 2 1 7 2 8 3 0 1 1] + [ 1 59 0 3 6 947 3 20 6 2 6 16 1 18 5 1 5 4 0 3 10] + [ 1 5 23 0 0 1 1117 8 0 0 6 3 0 0 1 11 1 0 0 8 6] + [ 4 36 26 0 2 47 8 1017 2 0 8 7 6 0 1 1 2 1 37 7 6] + [ 17 3 1 1 0 4 0 0 955 51 11 0 2 15 16 4 2 1 3 1 2] + [ 91 1 2 0 8 2 0 0 40 922 1 3 0 27 5 2 2 3 1 1 8] + [ 3 4 12 6 3 3 9 2 22 1 950 2 0 16 4 3 1 0 4 1 7] + [ 1 1 0 0 0 9 1 2 0 1 1 949 24 9 0 2 6 13 2 9 5] + [ 0 2 4 4 0 2 1 0 2 0 1 44 946 3 3 10 4 26 3 4 9] + [ 2 1 1 0 2 14 1 1 19 14 5 8 7 1015 1 6 0 3 0 7 12] + [ 14 6 3 5 5 0 0 0 34 6 2 2 1 2 994 0 4 1 11 0 11] + [ 0 4 1 0 3 1 3 0 0 0 0 5 7 2 1 1066 19 13 0 4 5] + [ 0 18 1 0 6 5 0 2 2 1 0 6 2 1 3 10 1090 0 0 5 9] + [ 0 0 0 3 0 1 0 0 1 0 0 6 13 1 3 7 0 998 0 0 5] + [ 6 15 12 15 2 0 0 29 6 0 10 0 4 0 15 0 1 0 943 4 6] + [ 0 5 9 1 2 6 11 12 0 0 2 14 5 0 0 8 12 2 2 1054 7] + [ 128 292 178 76 148 163 57 88 139 118 200 136 286 330 151 85 203 84 177 207 4659]] + +2023-10-05 21:24:49,480 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:24:49,480 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:24:49,485 - + +2023-10-05 21:24:49,485 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:24:50,474 - Epoch: [91][ 10/ 1236] Overall Loss 0.332125 Objective Loss 0.332125 LR 0.001000 Time 0.098765 +2023-10-05 21:24:50,676 - Epoch: [91][ 20/ 1236] Overall Loss 0.325490 Objective Loss 0.325490 LR 0.001000 Time 0.059486 +2023-10-05 21:24:50,876 - Epoch: [91][ 30/ 1236] Overall Loss 0.325754 Objective Loss 0.325754 LR 0.001000 Time 0.046319 +2023-10-05 21:24:51,079 - Epoch: [91][ 40/ 1236] Overall Loss 0.327835 Objective Loss 0.327835 LR 0.001000 Time 0.039792 +2023-10-05 21:24:51,279 - Epoch: [91][ 50/ 1236] Overall Loss 0.332859 Objective Loss 0.332859 LR 0.001000 Time 0.035838 +2023-10-05 21:24:51,482 - Epoch: [91][ 60/ 1236] Overall Loss 0.331385 Objective Loss 0.331385 LR 0.001000 Time 0.033233 +2023-10-05 21:24:51,683 - Epoch: [91][ 70/ 1236] Overall Loss 0.331092 Objective Loss 0.331092 LR 0.001000 Time 0.031352 +2023-10-05 21:24:51,885 - Epoch: [91][ 80/ 1236] Overall Loss 0.333742 Objective Loss 0.333742 LR 0.001000 Time 0.029956 +2023-10-05 21:24:52,088 - Epoch: [91][ 90/ 1236] Overall Loss 0.332398 Objective Loss 0.332398 LR 0.001000 Time 0.028881 +2023-10-05 21:24:52,292 - Epoch: [91][ 100/ 1236] Overall Loss 0.328111 Objective Loss 0.328111 LR 0.001000 Time 0.028028 +2023-10-05 21:24:52,498 - Epoch: [91][ 110/ 1236] Overall Loss 0.322757 Objective Loss 0.322757 LR 0.001000 Time 0.027348 +2023-10-05 21:24:52,703 - Epoch: [91][ 120/ 1236] Overall Loss 0.320419 Objective Loss 0.320419 LR 0.001000 Time 0.026775 +2023-10-05 21:24:52,907 - Epoch: [91][ 130/ 1236] Overall Loss 0.317892 Objective Loss 0.317892 LR 0.001000 Time 0.026285 +2023-10-05 21:24:53,109 - Epoch: [91][ 140/ 1236] Overall Loss 0.318740 Objective Loss 0.318740 LR 0.001000 Time 0.025847 +2023-10-05 21:24:53,313 - Epoch: [91][ 150/ 1236] Overall Loss 0.318293 Objective Loss 0.318293 LR 0.001000 Time 0.025482 +2023-10-05 21:24:53,515 - Epoch: [91][ 160/ 1236] Overall Loss 0.318641 Objective Loss 0.318641 LR 0.001000 Time 0.025149 +2023-10-05 21:24:53,719 - Epoch: [91][ 170/ 1236] Overall Loss 0.318075 Objective Loss 0.318075 LR 0.001000 Time 0.024868 +2023-10-05 21:24:53,921 - Epoch: [91][ 180/ 1236] Overall Loss 0.319874 Objective Loss 0.319874 LR 0.001000 Time 0.024607 +2023-10-05 21:24:54,125 - Epoch: [91][ 190/ 1236] Overall Loss 0.320839 Objective Loss 0.320839 LR 0.001000 Time 0.024383 +2023-10-05 21:24:54,327 - Epoch: [91][ 200/ 1236] Overall Loss 0.320594 Objective Loss 0.320594 LR 0.001000 Time 0.024173 +2023-10-05 21:24:54,531 - Epoch: [91][ 210/ 1236] Overall Loss 0.320909 Objective Loss 0.320909 LR 0.001000 Time 0.023991 +2023-10-05 21:24:54,733 - Epoch: [91][ 220/ 1236] Overall Loss 0.319703 Objective Loss 0.319703 LR 0.001000 Time 0.023817 +2023-10-05 21:24:54,937 - Epoch: [91][ 230/ 1236] Overall Loss 0.320162 Objective Loss 0.320162 LR 0.001000 Time 0.023667 +2023-10-05 21:24:55,140 - Epoch: [91][ 240/ 1236] Overall Loss 0.319983 Objective Loss 0.319983 LR 0.001000 Time 0.023522 +2023-10-05 21:24:55,344 - Epoch: [91][ 250/ 1236] Overall Loss 0.319805 Objective Loss 0.319805 LR 0.001000 Time 0.023396 +2023-10-05 21:24:55,546 - Epoch: [91][ 260/ 1236] Overall Loss 0.319378 Objective Loss 0.319378 LR 0.001000 Time 0.023273 +2023-10-05 21:24:55,750 - Epoch: [91][ 270/ 1236] Overall Loss 0.320509 Objective Loss 0.320509 LR 0.001000 Time 0.023165 +2023-10-05 21:24:55,952 - Epoch: [91][ 280/ 1236] Overall Loss 0.320213 Objective Loss 0.320213 LR 0.001000 Time 0.023059 +2023-10-05 21:24:56,156 - Epoch: [91][ 290/ 1236] Overall Loss 0.320311 Objective Loss 0.320311 LR 0.001000 Time 0.022965 +2023-10-05 21:24:56,358 - Epoch: [91][ 300/ 1236] Overall Loss 0.321461 Objective Loss 0.321461 LR 0.001000 Time 0.022873 +2023-10-05 21:24:56,563 - Epoch: [91][ 310/ 1236] Overall Loss 0.320624 Objective Loss 0.320624 LR 0.001000 Time 0.022793 +2023-10-05 21:24:56,765 - Epoch: [91][ 320/ 1236] Overall Loss 0.320377 Objective Loss 0.320377 LR 0.001000 Time 0.022713 +2023-10-05 21:24:56,970 - Epoch: [91][ 330/ 1236] Overall Loss 0.320900 Objective Loss 0.320900 LR 0.001000 Time 0.022643 +2023-10-05 21:24:57,172 - Epoch: [91][ 340/ 1236] Overall Loss 0.321388 Objective Loss 0.321388 LR 0.001000 Time 0.022572 +2023-10-05 21:24:57,375 - Epoch: [91][ 350/ 1236] Overall Loss 0.321556 Objective Loss 0.321556 LR 0.001000 Time 0.022507 +2023-10-05 21:24:57,578 - Epoch: [91][ 360/ 1236] Overall Loss 0.322256 Objective Loss 0.322256 LR 0.001000 Time 0.022444 +2023-10-05 21:24:57,782 - Epoch: [91][ 370/ 1236] Overall Loss 0.321861 Objective Loss 0.321861 LR 0.001000 Time 0.022388 +2023-10-05 21:24:57,986 - Epoch: [91][ 380/ 1236] Overall Loss 0.322607 Objective Loss 0.322607 LR 0.001000 Time 0.022334 +2023-10-05 21:24:58,191 - Epoch: [91][ 390/ 1236] Overall Loss 0.322943 Objective Loss 0.322943 LR 0.001000 Time 0.022287 +2023-10-05 21:24:58,395 - Epoch: [91][ 400/ 1236] Overall Loss 0.323421 Objective Loss 0.323421 LR 0.001000 Time 0.022237 +2023-10-05 21:24:58,599 - Epoch: [91][ 410/ 1236] Overall Loss 0.323887 Objective Loss 0.323887 LR 0.001000 Time 0.022193 +2023-10-05 21:24:58,802 - Epoch: [91][ 420/ 1236] Overall Loss 0.324228 Objective Loss 0.324228 LR 0.001000 Time 0.022148 +2023-10-05 21:24:59,007 - Epoch: [91][ 430/ 1236] Overall Loss 0.324662 Objective Loss 0.324662 LR 0.001000 Time 0.022108 +2023-10-05 21:24:59,211 - Epoch: [91][ 440/ 1236] Overall Loss 0.325030 Objective Loss 0.325030 LR 0.001000 Time 0.022068 +2023-10-05 21:24:59,415 - Epoch: [91][ 450/ 1236] Overall Loss 0.325163 Objective Loss 0.325163 LR 0.001000 Time 0.022031 +2023-10-05 21:24:59,619 - Epoch: [91][ 460/ 1236] Overall Loss 0.325411 Objective Loss 0.325411 LR 0.001000 Time 0.021994 +2023-10-05 21:24:59,823 - Epoch: [91][ 470/ 1236] Overall Loss 0.325441 Objective Loss 0.325441 LR 0.001000 Time 0.021959 +2023-10-05 21:25:00,026 - Epoch: [91][ 480/ 1236] Overall Loss 0.325627 Objective Loss 0.325627 LR 0.001000 Time 0.021925 +2023-10-05 21:25:00,230 - Epoch: [91][ 490/ 1236] Overall Loss 0.325099 Objective Loss 0.325099 LR 0.001000 Time 0.021893 +2023-10-05 21:25:00,434 - Epoch: [91][ 500/ 1236] Overall Loss 0.325382 Objective Loss 0.325382 LR 0.001000 Time 0.021861 +2023-10-05 21:25:00,638 - Epoch: [91][ 510/ 1236] Overall Loss 0.325343 Objective Loss 0.325343 LR 0.001000 Time 0.021832 +2023-10-05 21:25:00,841 - Epoch: [91][ 520/ 1236] Overall Loss 0.325313 Objective Loss 0.325313 LR 0.001000 Time 0.021803 +2023-10-05 21:25:01,045 - Epoch: [91][ 530/ 1236] Overall Loss 0.325940 Objective Loss 0.325940 LR 0.001000 Time 0.021776 +2023-10-05 21:25:01,249 - Epoch: [91][ 540/ 1236] Overall Loss 0.325928 Objective Loss 0.325928 LR 0.001000 Time 0.021748 +2023-10-05 21:25:01,453 - Epoch: [91][ 550/ 1236] Overall Loss 0.326523 Objective Loss 0.326523 LR 0.001000 Time 0.021723 +2023-10-05 21:25:01,656 - Epoch: [91][ 560/ 1236] Overall Loss 0.326946 Objective Loss 0.326946 LR 0.001000 Time 0.021698 +2023-10-05 21:25:01,860 - Epoch: [91][ 570/ 1236] Overall Loss 0.327139 Objective Loss 0.327139 LR 0.001000 Time 0.021675 +2023-10-05 21:25:02,064 - Epoch: [91][ 580/ 1236] Overall Loss 0.327207 Objective Loss 0.327207 LR 0.001000 Time 0.021651 +2023-10-05 21:25:02,268 - Epoch: [91][ 590/ 1236] Overall Loss 0.327873 Objective Loss 0.327873 LR 0.001000 Time 0.021630 +2023-10-05 21:25:02,472 - Epoch: [91][ 600/ 1236] Overall Loss 0.327681 Objective Loss 0.327681 LR 0.001000 Time 0.021608 +2023-10-05 21:25:02,676 - Epoch: [91][ 610/ 1236] Overall Loss 0.327906 Objective Loss 0.327906 LR 0.001000 Time 0.021589 +2023-10-05 21:25:02,880 - Epoch: [91][ 620/ 1236] Overall Loss 0.328085 Objective Loss 0.328085 LR 0.001000 Time 0.021568 +2023-10-05 21:25:03,084 - Epoch: [91][ 630/ 1236] Overall Loss 0.329027 Objective Loss 0.329027 LR 0.001000 Time 0.021549 +2023-10-05 21:25:03,287 - Epoch: [91][ 640/ 1236] Overall Loss 0.329484 Objective Loss 0.329484 LR 0.001000 Time 0.021530 +2023-10-05 21:25:03,490 - Epoch: [91][ 650/ 1236] Overall Loss 0.329450 Objective Loss 0.329450 LR 0.001000 Time 0.021509 +2023-10-05 21:25:03,691 - Epoch: [91][ 660/ 1236] Overall Loss 0.328393 Objective Loss 0.328393 LR 0.001000 Time 0.021488 +2023-10-05 21:25:03,893 - Epoch: [91][ 670/ 1236] Overall Loss 0.328220 Objective Loss 0.328220 LR 0.001000 Time 0.021468 +2023-10-05 21:25:04,094 - Epoch: [91][ 680/ 1236] Overall Loss 0.328151 Objective Loss 0.328151 LR 0.001000 Time 0.021448 +2023-10-05 21:25:04,296 - Epoch: [91][ 690/ 1236] Overall Loss 0.327953 Objective Loss 0.327953 LR 0.001000 Time 0.021428 +2023-10-05 21:25:04,497 - Epoch: [91][ 700/ 1236] Overall Loss 0.327721 Objective Loss 0.327721 LR 0.001000 Time 0.021409 +2023-10-05 21:25:04,698 - Epoch: [91][ 710/ 1236] Overall Loss 0.327496 Objective Loss 0.327496 LR 0.001000 Time 0.021391 +2023-10-05 21:25:04,899 - Epoch: [91][ 720/ 1236] Overall Loss 0.327725 Objective Loss 0.327725 LR 0.001000 Time 0.021372 +2023-10-05 21:25:05,101 - Epoch: [91][ 730/ 1236] Overall Loss 0.328479 Objective Loss 0.328479 LR 0.001000 Time 0.021355 +2023-10-05 21:25:05,302 - Epoch: [91][ 740/ 1236] Overall Loss 0.328106 Objective Loss 0.328106 LR 0.001000 Time 0.021338 +2023-10-05 21:25:05,503 - Epoch: [91][ 750/ 1236] Overall Loss 0.327979 Objective Loss 0.327979 LR 0.001000 Time 0.021322 +2023-10-05 21:25:05,704 - Epoch: [91][ 760/ 1236] Overall Loss 0.328506 Objective Loss 0.328506 LR 0.001000 Time 0.021305 +2023-10-05 21:25:05,906 - Epoch: [91][ 770/ 1236] Overall Loss 0.328546 Objective Loss 0.328546 LR 0.001000 Time 0.021290 +2023-10-05 21:25:06,107 - Epoch: [91][ 780/ 1236] Overall Loss 0.328290 Objective Loss 0.328290 LR 0.001000 Time 0.021274 +2023-10-05 21:25:06,308 - Epoch: [91][ 790/ 1236] Overall Loss 0.328400 Objective Loss 0.328400 LR 0.001000 Time 0.021260 +2023-10-05 21:25:06,509 - Epoch: [91][ 800/ 1236] Overall Loss 0.328486 Objective Loss 0.328486 LR 0.001000 Time 0.021245 +2023-10-05 21:25:06,711 - Epoch: [91][ 810/ 1236] Overall Loss 0.328538 Objective Loss 0.328538 LR 0.001000 Time 0.021230 +2023-10-05 21:25:06,912 - Epoch: [91][ 820/ 1236] Overall Loss 0.328278 Objective Loss 0.328278 LR 0.001000 Time 0.021217 +2023-10-05 21:25:07,114 - Epoch: [91][ 830/ 1236] Overall Loss 0.328344 Objective Loss 0.328344 LR 0.001000 Time 0.021204 +2023-10-05 21:25:07,315 - Epoch: [91][ 840/ 1236] Overall Loss 0.328445 Objective Loss 0.328445 LR 0.001000 Time 0.021191 +2023-10-05 21:25:07,517 - Epoch: [91][ 850/ 1236] Overall Loss 0.328374 Objective Loss 0.328374 LR 0.001000 Time 0.021178 +2023-10-05 21:25:07,718 - Epoch: [91][ 860/ 1236] Overall Loss 0.327972 Objective Loss 0.327972 LR 0.001000 Time 0.021165 +2023-10-05 21:25:07,920 - Epoch: [91][ 870/ 1236] Overall Loss 0.327838 Objective Loss 0.327838 LR 0.001000 Time 0.021153 +2023-10-05 21:25:08,121 - Epoch: [91][ 880/ 1236] Overall Loss 0.327760 Objective Loss 0.327760 LR 0.001000 Time 0.021141 +2023-10-05 21:25:08,322 - Epoch: [91][ 890/ 1236] Overall Loss 0.327383 Objective Loss 0.327383 LR 0.001000 Time 0.021130 +2023-10-05 21:25:08,524 - Epoch: [91][ 900/ 1236] Overall Loss 0.327831 Objective Loss 0.327831 LR 0.001000 Time 0.021118 +2023-10-05 21:25:08,725 - Epoch: [91][ 910/ 1236] Overall Loss 0.327549 Objective Loss 0.327549 LR 0.001000 Time 0.021108 +2023-10-05 21:25:08,927 - Epoch: [91][ 920/ 1236] Overall Loss 0.327454 Objective Loss 0.327454 LR 0.001000 Time 0.021096 +2023-10-05 21:25:09,128 - Epoch: [91][ 930/ 1236] Overall Loss 0.327426 Objective Loss 0.327426 LR 0.001000 Time 0.021086 +2023-10-05 21:25:09,329 - Epoch: [91][ 940/ 1236] Overall Loss 0.327163 Objective Loss 0.327163 LR 0.001000 Time 0.021075 +2023-10-05 21:25:09,531 - Epoch: [91][ 950/ 1236] Overall Loss 0.326810 Objective Loss 0.326810 LR 0.001000 Time 0.021066 +2023-10-05 21:25:09,733 - Epoch: [91][ 960/ 1236] Overall Loss 0.326394 Objective Loss 0.326394 LR 0.001000 Time 0.021056 +2023-10-05 21:25:09,934 - Epoch: [91][ 970/ 1236] Overall Loss 0.326532 Objective Loss 0.326532 LR 0.001000 Time 0.021046 +2023-10-05 21:25:10,135 - Epoch: [91][ 980/ 1236] Overall Loss 0.326454 Objective Loss 0.326454 LR 0.001000 Time 0.021036 +2023-10-05 21:25:10,337 - Epoch: [91][ 990/ 1236] Overall Loss 0.326796 Objective Loss 0.326796 LR 0.001000 Time 0.021027 +2023-10-05 21:25:10,538 - Epoch: [91][ 1000/ 1236] Overall Loss 0.326740 Objective Loss 0.326740 LR 0.001000 Time 0.021017 +2023-10-05 21:25:10,739 - Epoch: [91][ 1010/ 1236] Overall Loss 0.326959 Objective Loss 0.326959 LR 0.001000 Time 0.021008 +2023-10-05 21:25:10,940 - Epoch: [91][ 1020/ 1236] Overall Loss 0.326861 Objective Loss 0.326861 LR 0.001000 Time 0.020999 +2023-10-05 21:25:11,142 - Epoch: [91][ 1030/ 1236] Overall Loss 0.326652 Objective Loss 0.326652 LR 0.001000 Time 0.020991 +2023-10-05 21:25:11,343 - Epoch: [91][ 1040/ 1236] Overall Loss 0.326589 Objective Loss 0.326589 LR 0.001000 Time 0.020982 +2023-10-05 21:25:11,545 - Epoch: [91][ 1050/ 1236] Overall Loss 0.326495 Objective Loss 0.326495 LR 0.001000 Time 0.020974 +2023-10-05 21:25:11,746 - Epoch: [91][ 1060/ 1236] Overall Loss 0.326515 Objective Loss 0.326515 LR 0.001000 Time 0.020965 +2023-10-05 21:25:11,948 - Epoch: [91][ 1070/ 1236] Overall Loss 0.326211 Objective Loss 0.326211 LR 0.001000 Time 0.020958 +2023-10-05 21:25:12,149 - Epoch: [91][ 1080/ 1236] Overall Loss 0.326296 Objective Loss 0.326296 LR 0.001000 Time 0.020949 +2023-10-05 21:25:12,350 - Epoch: [91][ 1090/ 1236] Overall Loss 0.326452 Objective Loss 0.326452 LR 0.001000 Time 0.020942 +2023-10-05 21:25:12,552 - Epoch: [91][ 1100/ 1236] Overall Loss 0.326238 Objective Loss 0.326238 LR 0.001000 Time 0.020934 +2023-10-05 21:25:12,753 - Epoch: [91][ 1110/ 1236] Overall Loss 0.325909 Objective Loss 0.325909 LR 0.001000 Time 0.020927 +2023-10-05 21:25:12,955 - Epoch: [91][ 1120/ 1236] Overall Loss 0.326224 Objective Loss 0.326224 LR 0.001000 Time 0.020920 +2023-10-05 21:25:13,156 - Epoch: [91][ 1130/ 1236] Overall Loss 0.326267 Objective Loss 0.326267 LR 0.001000 Time 0.020913 +2023-10-05 21:25:13,357 - Epoch: [91][ 1140/ 1236] Overall Loss 0.326313 Objective Loss 0.326313 LR 0.001000 Time 0.020905 +2023-10-05 21:25:13,559 - Epoch: [91][ 1150/ 1236] Overall Loss 0.326281 Objective Loss 0.326281 LR 0.001000 Time 0.020899 +2023-10-05 21:25:13,761 - Epoch: [91][ 1160/ 1236] Overall Loss 0.326503 Objective Loss 0.326503 LR 0.001000 Time 0.020892 +2023-10-05 21:25:13,962 - Epoch: [91][ 1170/ 1236] Overall Loss 0.326638 Objective Loss 0.326638 LR 0.001000 Time 0.020885 +2023-10-05 21:25:14,164 - Epoch: [91][ 1180/ 1236] Overall Loss 0.326568 Objective Loss 0.326568 LR 0.001000 Time 0.020879 +2023-10-05 21:25:14,366 - Epoch: [91][ 1190/ 1236] Overall Loss 0.326765 Objective Loss 0.326765 LR 0.001000 Time 0.020873 +2023-10-05 21:25:14,567 - Epoch: [91][ 1200/ 1236] Overall Loss 0.327117 Objective Loss 0.327117 LR 0.001000 Time 0.020866 +2023-10-05 21:25:14,769 - Epoch: [91][ 1210/ 1236] Overall Loss 0.327082 Objective Loss 0.327082 LR 0.001000 Time 0.020860 +2023-10-05 21:25:14,970 - Epoch: [91][ 1220/ 1236] Overall Loss 0.327276 Objective Loss 0.327276 LR 0.001000 Time 0.020854 +2023-10-05 21:25:15,222 - Epoch: [91][ 1230/ 1236] Overall Loss 0.327475 Objective Loss 0.327475 LR 0.001000 Time 0.020889 +2023-10-05 21:25:15,339 - Epoch: [91][ 1236/ 1236] Overall Loss 0.327867 Objective Loss 0.327867 Top1 81.059063 Top5 96.945010 LR 0.001000 Time 0.020882 +2023-10-05 21:25:15,465 - --- validate (epoch=91)----------- +2023-10-05 21:25:15,466 - 29943 samples (256 per mini-batch) +2023-10-05 21:25:15,928 - Epoch: [91][ 10/ 117] Loss 0.351362 Top1 80.898438 Top5 97.539062 +2023-10-05 21:25:16,074 - Epoch: [91][ 20/ 117] Loss 0.375125 Top1 80.566406 Top5 97.382812 +2023-10-05 21:25:16,220 - Epoch: [91][ 30/ 117] Loss 0.365971 Top1 81.184896 Top5 97.486979 +2023-10-05 21:25:16,368 - Epoch: [91][ 40/ 117] Loss 0.373982 Top1 81.054688 Top5 97.285156 +2023-10-05 21:25:16,518 - Epoch: [91][ 50/ 117] Loss 0.371218 Top1 81.234375 Top5 97.460938 +2023-10-05 21:25:16,668 - Epoch: [91][ 60/ 117] Loss 0.371757 Top1 81.223958 Top5 97.408854 +2023-10-05 21:25:16,817 - Epoch: [91][ 70/ 117] Loss 0.366251 Top1 81.523438 Top5 97.438616 +2023-10-05 21:25:16,974 - Epoch: [91][ 80/ 117] Loss 0.366108 Top1 81.435547 Top5 97.436523 +2023-10-05 21:25:17,129 - Epoch: [91][ 90/ 117] Loss 0.366448 Top1 81.427951 Top5 97.430556 +2023-10-05 21:25:17,285 - Epoch: [91][ 100/ 117] Loss 0.365460 Top1 81.433594 Top5 97.425781 +2023-10-05 21:25:17,448 - Epoch: [91][ 110/ 117] Loss 0.366278 Top1 81.409801 Top5 97.379261 +2023-10-05 21:25:17,533 - Epoch: [91][ 117/ 117] Loss 0.365776 Top1 81.458104 Top5 97.348295 +2023-10-05 21:25:17,665 - ==> Top1: 81.458 Top5: 97.348 Loss: 0.366 + +2023-10-05 21:25:17,666 - ==> Confusion: +[[ 925 1 8 0 11 3 0 0 7 52 0 1 2 7 6 4 6 4 2 1 10] + [ 2 1014 2 1 12 36 5 19 2 0 3 1 1 2 1 6 8 0 10 1 5] + [ 5 0 940 19 1 3 43 4 0 0 3 3 8 2 2 6 1 2 0 8 6] + [ 1 1 21 964 1 7 0 0 1 0 8 0 7 2 25 6 2 5 22 4 12] + [ 28 7 2 0 964 3 0 0 0 9 0 5 0 1 12 3 11 2 0 0 3] + [ 7 22 0 1 3 992 2 14 2 2 6 9 1 17 11 1 5 1 4 5 11] + [ 0 6 24 0 0 0 1120 5 0 0 0 2 2 0 1 16 0 2 2 7 4] + [ 7 23 29 0 3 66 10 988 1 2 6 9 2 1 0 0 1 0 53 10 7] + [ 21 3 0 0 1 6 0 0 938 53 14 5 2 15 16 2 4 1 7 0 1] + [ 135 0 3 0 5 4 1 0 20 906 0 2 2 18 7 2 3 1 1 1 8] + [ 3 5 21 7 3 5 9 0 11 1 938 5 0 16 5 1 1 0 11 1 10] + [ 1 1 2 0 2 10 0 1 0 0 0 941 30 7 0 4 3 11 0 19 3] + [ 1 0 6 5 1 3 1 0 2 1 0 52 964 4 1 5 0 5 1 4 12] + [ 1 0 6 1 5 13 1 0 7 15 8 4 3 1034 3 3 0 1 0 6 8] + [ 15 0 4 13 14 1 0 0 32 2 1 2 5 1 975 0 0 1 17 0 18] + [ 1 0 2 0 4 3 2 0 0 0 0 9 8 1 0 1075 15 5 1 6 2] + [ 1 12 3 1 9 7 0 0 0 0 0 6 1 1 2 14 1087 0 1 7 9] + [ 1 0 1 0 0 0 0 0 0 0 0 10 31 0 3 13 2 969 1 1 6] + [ 2 6 11 20 1 1 0 20 5 0 3 4 4 6 6 0 2 0 964 4 9] + [ 0 1 6 2 2 8 14 10 0 0 0 19 5 2 0 9 13 2 3 1053 3] + [ 165 182 215 89 130 195 61 70 77 98 165 148 366 336 176 110 230 63 161 228 4640]] + +2023-10-05 21:25:17,667 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:25:17,667 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:25:17,673 - + +2023-10-05 21:25:17,673 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:25:18,781 - Epoch: [92][ 10/ 1236] Overall Loss 0.324476 Objective Loss 0.324476 LR 0.001000 Time 0.110761 +2023-10-05 21:25:18,985 - Epoch: [92][ 20/ 1236] Overall Loss 0.337120 Objective Loss 0.337120 LR 0.001000 Time 0.065534 +2023-10-05 21:25:19,193 - Epoch: [92][ 30/ 1236] Overall Loss 0.338379 Objective Loss 0.338379 LR 0.001000 Time 0.050621 +2023-10-05 21:25:19,406 - Epoch: [92][ 40/ 1236] Overall Loss 0.327684 Objective Loss 0.327684 LR 0.001000 Time 0.043279 +2023-10-05 21:25:19,614 - Epoch: [92][ 50/ 1236] Overall Loss 0.325128 Objective Loss 0.325128 LR 0.001000 Time 0.038782 +2023-10-05 21:25:19,826 - Epoch: [92][ 60/ 1236] Overall Loss 0.322892 Objective Loss 0.322892 LR 0.001000 Time 0.035855 +2023-10-05 21:25:20,034 - Epoch: [92][ 70/ 1236] Overall Loss 0.322503 Objective Loss 0.322503 LR 0.001000 Time 0.033697 +2023-10-05 21:25:20,247 - Epoch: [92][ 80/ 1236] Overall Loss 0.322248 Objective Loss 0.322248 LR 0.001000 Time 0.032139 +2023-10-05 21:25:20,455 - Epoch: [92][ 90/ 1236] Overall Loss 0.320163 Objective Loss 0.320163 LR 0.001000 Time 0.030875 +2023-10-05 21:25:20,667 - Epoch: [92][ 100/ 1236] Overall Loss 0.321450 Objective Loss 0.321450 LR 0.001000 Time 0.029903 +2023-10-05 21:25:20,873 - Epoch: [92][ 110/ 1236] Overall Loss 0.319804 Objective Loss 0.319804 LR 0.001000 Time 0.029057 +2023-10-05 21:25:21,085 - Epoch: [92][ 120/ 1236] Overall Loss 0.318621 Objective Loss 0.318621 LR 0.001000 Time 0.028403 +2023-10-05 21:25:21,295 - Epoch: [92][ 130/ 1236] Overall Loss 0.317439 Objective Loss 0.317439 LR 0.001000 Time 0.027831 +2023-10-05 21:25:21,509 - Epoch: [92][ 140/ 1236] Overall Loss 0.315232 Objective Loss 0.315232 LR 0.001000 Time 0.027365 +2023-10-05 21:25:21,719 - Epoch: [92][ 150/ 1236] Overall Loss 0.316561 Objective Loss 0.316561 LR 0.001000 Time 0.026938 +2023-10-05 21:25:21,932 - Epoch: [92][ 160/ 1236] Overall Loss 0.317440 Objective Loss 0.317440 LR 0.001000 Time 0.026586 +2023-10-05 21:25:22,142 - Epoch: [92][ 170/ 1236] Overall Loss 0.317990 Objective Loss 0.317990 LR 0.001000 Time 0.026254 +2023-10-05 21:25:22,356 - Epoch: [92][ 180/ 1236] Overall Loss 0.318948 Objective Loss 0.318948 LR 0.001000 Time 0.025982 +2023-10-05 21:25:22,566 - Epoch: [92][ 190/ 1236] Overall Loss 0.318502 Objective Loss 0.318502 LR 0.001000 Time 0.025718 +2023-10-05 21:25:22,779 - Epoch: [92][ 200/ 1236] Overall Loss 0.318338 Objective Loss 0.318338 LR 0.001000 Time 0.025497 +2023-10-05 21:25:22,989 - Epoch: [92][ 210/ 1236] Overall Loss 0.318285 Objective Loss 0.318285 LR 0.001000 Time 0.025282 +2023-10-05 21:25:23,203 - Epoch: [92][ 220/ 1236] Overall Loss 0.316238 Objective Loss 0.316238 LR 0.001000 Time 0.025102 +2023-10-05 21:25:23,413 - Epoch: [92][ 230/ 1236] Overall Loss 0.315575 Objective Loss 0.315575 LR 0.001000 Time 0.024922 +2023-10-05 21:25:23,626 - Epoch: [92][ 240/ 1236] Overall Loss 0.316106 Objective Loss 0.316106 LR 0.001000 Time 0.024772 +2023-10-05 21:25:23,836 - Epoch: [92][ 250/ 1236] Overall Loss 0.315270 Objective Loss 0.315270 LR 0.001000 Time 0.024619 +2023-10-05 21:25:24,039 - Epoch: [92][ 260/ 1236] Overall Loss 0.315651 Objective Loss 0.315651 LR 0.001000 Time 0.024453 +2023-10-05 21:25:24,243 - Epoch: [92][ 270/ 1236] Overall Loss 0.315858 Objective Loss 0.315858 LR 0.001000 Time 0.024299 +2023-10-05 21:25:24,447 - Epoch: [92][ 280/ 1236] Overall Loss 0.315820 Objective Loss 0.315820 LR 0.001000 Time 0.024159 +2023-10-05 21:25:24,650 - Epoch: [92][ 290/ 1236] Overall Loss 0.316630 Objective Loss 0.316630 LR 0.001000 Time 0.024025 +2023-10-05 21:25:24,854 - Epoch: [92][ 300/ 1236] Overall Loss 0.316816 Objective Loss 0.316816 LR 0.001000 Time 0.023904 +2023-10-05 21:25:25,057 - Epoch: [92][ 310/ 1236] Overall Loss 0.317006 Objective Loss 0.317006 LR 0.001000 Time 0.023785 +2023-10-05 21:25:25,264 - Epoch: [92][ 320/ 1236] Overall Loss 0.317853 Objective Loss 0.317853 LR 0.001000 Time 0.023688 +2023-10-05 21:25:25,469 - Epoch: [92][ 330/ 1236] Overall Loss 0.317537 Objective Loss 0.317537 LR 0.001000 Time 0.023591 +2023-10-05 21:25:25,675 - Epoch: [92][ 340/ 1236] Overall Loss 0.316821 Objective Loss 0.316821 LR 0.001000 Time 0.023501 +2023-10-05 21:25:25,879 - Epoch: [92][ 350/ 1236] Overall Loss 0.317495 Objective Loss 0.317495 LR 0.001000 Time 0.023410 +2023-10-05 21:25:26,082 - Epoch: [92][ 360/ 1236] Overall Loss 0.317549 Objective Loss 0.317549 LR 0.001000 Time 0.023322 +2023-10-05 21:25:26,284 - Epoch: [92][ 370/ 1236] Overall Loss 0.317541 Objective Loss 0.317541 LR 0.001000 Time 0.023238 +2023-10-05 21:25:26,487 - Epoch: [92][ 380/ 1236] Overall Loss 0.317513 Objective Loss 0.317513 LR 0.001000 Time 0.023160 +2023-10-05 21:25:26,689 - Epoch: [92][ 390/ 1236] Overall Loss 0.316944 Objective Loss 0.316944 LR 0.001000 Time 0.023084 +2023-10-05 21:25:26,893 - Epoch: [92][ 400/ 1236] Overall Loss 0.317288 Objective Loss 0.317288 LR 0.001000 Time 0.023014 +2023-10-05 21:25:27,095 - Epoch: [92][ 410/ 1236] Overall Loss 0.318010 Objective Loss 0.318010 LR 0.001000 Time 0.022942 +2023-10-05 21:25:27,298 - Epoch: [92][ 420/ 1236] Overall Loss 0.318257 Objective Loss 0.318257 LR 0.001000 Time 0.022877 +2023-10-05 21:25:27,500 - Epoch: [92][ 430/ 1236] Overall Loss 0.318100 Objective Loss 0.318100 LR 0.001000 Time 0.022814 +2023-10-05 21:25:27,702 - Epoch: [92][ 440/ 1236] Overall Loss 0.317941 Objective Loss 0.317941 LR 0.001000 Time 0.022756 +2023-10-05 21:25:27,905 - Epoch: [92][ 450/ 1236] Overall Loss 0.318346 Objective Loss 0.318346 LR 0.001000 Time 0.022699 +2023-10-05 21:25:28,107 - Epoch: [92][ 460/ 1236] Overall Loss 0.317777 Objective Loss 0.317777 LR 0.001000 Time 0.022645 +2023-10-05 21:25:28,309 - Epoch: [92][ 470/ 1236] Overall Loss 0.317466 Objective Loss 0.317466 LR 0.001000 Time 0.022593 +2023-10-05 21:25:28,512 - Epoch: [92][ 480/ 1236] Overall Loss 0.317971 Objective Loss 0.317971 LR 0.001000 Time 0.022544 +2023-10-05 21:25:28,715 - Epoch: [92][ 490/ 1236] Overall Loss 0.318732 Objective Loss 0.318732 LR 0.001000 Time 0.022494 +2023-10-05 21:25:28,917 - Epoch: [92][ 500/ 1236] Overall Loss 0.318856 Objective Loss 0.318856 LR 0.001000 Time 0.022448 +2023-10-05 21:25:29,119 - Epoch: [92][ 510/ 1236] Overall Loss 0.319523 Objective Loss 0.319523 LR 0.001000 Time 0.022403 +2023-10-05 21:25:29,322 - Epoch: [92][ 520/ 1236] Overall Loss 0.320057 Objective Loss 0.320057 LR 0.001000 Time 0.022362 +2023-10-05 21:25:29,524 - Epoch: [92][ 530/ 1236] Overall Loss 0.320360 Objective Loss 0.320360 LR 0.001000 Time 0.022320 +2023-10-05 21:25:29,727 - Epoch: [92][ 540/ 1236] Overall Loss 0.321202 Objective Loss 0.321202 LR 0.001000 Time 0.022282 +2023-10-05 21:25:29,929 - Epoch: [92][ 550/ 1236] Overall Loss 0.321772 Objective Loss 0.321772 LR 0.001000 Time 0.022244 +2023-10-05 21:25:30,132 - Epoch: [92][ 560/ 1236] Overall Loss 0.321491 Objective Loss 0.321491 LR 0.001000 Time 0.022208 +2023-10-05 21:25:30,334 - Epoch: [92][ 570/ 1236] Overall Loss 0.321996 Objective Loss 0.321996 LR 0.001000 Time 0.022172 +2023-10-05 21:25:30,537 - Epoch: [92][ 580/ 1236] Overall Loss 0.323144 Objective Loss 0.323144 LR 0.001000 Time 0.022140 +2023-10-05 21:25:30,739 - Epoch: [92][ 590/ 1236] Overall Loss 0.323409 Objective Loss 0.323409 LR 0.001000 Time 0.022106 +2023-10-05 21:25:30,942 - Epoch: [92][ 600/ 1236] Overall Loss 0.323010 Objective Loss 0.323010 LR 0.001000 Time 0.022076 +2023-10-05 21:25:31,146 - Epoch: [92][ 610/ 1236] Overall Loss 0.323304 Objective Loss 0.323304 LR 0.001000 Time 0.022046 +2023-10-05 21:25:31,349 - Epoch: [92][ 620/ 1236] Overall Loss 0.322948 Objective Loss 0.322948 LR 0.001000 Time 0.022017 +2023-10-05 21:25:31,551 - Epoch: [92][ 630/ 1236] Overall Loss 0.323438 Objective Loss 0.323438 LR 0.001000 Time 0.021986 +2023-10-05 21:25:31,754 - Epoch: [92][ 640/ 1236] Overall Loss 0.323685 Objective Loss 0.323685 LR 0.001000 Time 0.021960 +2023-10-05 21:25:31,957 - Epoch: [92][ 650/ 1236] Overall Loss 0.323844 Objective Loss 0.323844 LR 0.001000 Time 0.021933 +2023-10-05 21:25:32,160 - Epoch: [92][ 660/ 1236] Overall Loss 0.324086 Objective Loss 0.324086 LR 0.001000 Time 0.021908 +2023-10-05 21:25:32,362 - Epoch: [92][ 670/ 1236] Overall Loss 0.323911 Objective Loss 0.323911 LR 0.001000 Time 0.021880 +2023-10-05 21:25:32,565 - Epoch: [92][ 680/ 1236] Overall Loss 0.323557 Objective Loss 0.323557 LR 0.001000 Time 0.021856 +2023-10-05 21:25:32,767 - Epoch: [92][ 690/ 1236] Overall Loss 0.323727 Objective Loss 0.323727 LR 0.001000 Time 0.021831 +2023-10-05 21:25:32,970 - Epoch: [92][ 700/ 1236] Overall Loss 0.323389 Objective Loss 0.323389 LR 0.001000 Time 0.021808 +2023-10-05 21:25:33,172 - Epoch: [92][ 710/ 1236] Overall Loss 0.323879 Objective Loss 0.323879 LR 0.001000 Time 0.021785 +2023-10-05 21:25:33,375 - Epoch: [92][ 720/ 1236] Overall Loss 0.323914 Objective Loss 0.323914 LR 0.001000 Time 0.021764 +2023-10-05 21:25:33,578 - Epoch: [92][ 730/ 1236] Overall Loss 0.323879 Objective Loss 0.323879 LR 0.001000 Time 0.021741 +2023-10-05 21:25:33,781 - Epoch: [92][ 740/ 1236] Overall Loss 0.323978 Objective Loss 0.323978 LR 0.001000 Time 0.021721 +2023-10-05 21:25:33,983 - Epoch: [92][ 750/ 1236] Overall Loss 0.323450 Objective Loss 0.323450 LR 0.001000 Time 0.021701 +2023-10-05 21:25:34,186 - Epoch: [92][ 760/ 1236] Overall Loss 0.322976 Objective Loss 0.322976 LR 0.001000 Time 0.021682 +2023-10-05 21:25:34,389 - Epoch: [92][ 770/ 1236] Overall Loss 0.323277 Objective Loss 0.323277 LR 0.001000 Time 0.021663 +2023-10-05 21:25:34,592 - Epoch: [92][ 780/ 1236] Overall Loss 0.323496 Objective Loss 0.323496 LR 0.001000 Time 0.021645 +2023-10-05 21:25:34,795 - Epoch: [92][ 790/ 1236] Overall Loss 0.323404 Objective Loss 0.323404 LR 0.001000 Time 0.021626 +2023-10-05 21:25:34,998 - Epoch: [92][ 800/ 1236] Overall Loss 0.323300 Objective Loss 0.323300 LR 0.001000 Time 0.021609 +2023-10-05 21:25:35,201 - Epoch: [92][ 810/ 1236] Overall Loss 0.323580 Objective Loss 0.323580 LR 0.001000 Time 0.021590 +2023-10-05 21:25:35,404 - Epoch: [92][ 820/ 1236] Overall Loss 0.323901 Objective Loss 0.323901 LR 0.001000 Time 0.021575 +2023-10-05 21:25:35,606 - Epoch: [92][ 830/ 1236] Overall Loss 0.323760 Objective Loss 0.323760 LR 0.001000 Time 0.021556 +2023-10-05 21:25:35,809 - Epoch: [92][ 840/ 1236] Overall Loss 0.323549 Objective Loss 0.323549 LR 0.001000 Time 0.021541 +2023-10-05 21:25:36,012 - Epoch: [92][ 850/ 1236] Overall Loss 0.323659 Objective Loss 0.323659 LR 0.001000 Time 0.021524 +2023-10-05 21:25:36,215 - Epoch: [92][ 860/ 1236] Overall Loss 0.323573 Objective Loss 0.323573 LR 0.001000 Time 0.021509 +2023-10-05 21:25:36,417 - Epoch: [92][ 870/ 1236] Overall Loss 0.323612 Objective Loss 0.323612 LR 0.001000 Time 0.021495 +2023-10-05 21:25:36,620 - Epoch: [92][ 880/ 1236] Overall Loss 0.323647 Objective Loss 0.323647 LR 0.001000 Time 0.021481 +2023-10-05 21:25:36,822 - Epoch: [92][ 890/ 1236] Overall Loss 0.323492 Objective Loss 0.323492 LR 0.001000 Time 0.021465 +2023-10-05 21:25:37,026 - Epoch: [92][ 900/ 1236] Overall Loss 0.323736 Objective Loss 0.323736 LR 0.001000 Time 0.021451 +2023-10-05 21:25:37,228 - Epoch: [92][ 910/ 1236] Overall Loss 0.323954 Objective Loss 0.323954 LR 0.001000 Time 0.021438 +2023-10-05 21:25:37,431 - Epoch: [92][ 920/ 1236] Overall Loss 0.323888 Objective Loss 0.323888 LR 0.001000 Time 0.021425 +2023-10-05 21:25:37,634 - Epoch: [92][ 930/ 1236] Overall Loss 0.323677 Objective Loss 0.323677 LR 0.001000 Time 0.021412 +2023-10-05 21:25:37,837 - Epoch: [92][ 940/ 1236] Overall Loss 0.323807 Objective Loss 0.323807 LR 0.001000 Time 0.021400 +2023-10-05 21:25:38,039 - Epoch: [92][ 950/ 1236] Overall Loss 0.324199 Objective Loss 0.324199 LR 0.001000 Time 0.021388 +2023-10-05 21:25:38,244 - Epoch: [92][ 960/ 1236] Overall Loss 0.324339 Objective Loss 0.324339 LR 0.001000 Time 0.021378 +2023-10-05 21:25:38,449 - Epoch: [92][ 970/ 1236] Overall Loss 0.324036 Objective Loss 0.324036 LR 0.001000 Time 0.021368 +2023-10-05 21:25:38,652 - Epoch: [92][ 980/ 1236] Overall Loss 0.324107 Objective Loss 0.324107 LR 0.001000 Time 0.021357 +2023-10-05 21:25:38,854 - Epoch: [92][ 990/ 1236] Overall Loss 0.323931 Objective Loss 0.323931 LR 0.001000 Time 0.021345 +2023-10-05 21:25:39,057 - Epoch: [92][ 1000/ 1236] Overall Loss 0.323869 Objective Loss 0.323869 LR 0.001000 Time 0.021334 +2023-10-05 21:25:39,260 - Epoch: [92][ 1010/ 1236] Overall Loss 0.324024 Objective Loss 0.324024 LR 0.001000 Time 0.021323 +2023-10-05 21:25:39,463 - Epoch: [92][ 1020/ 1236] Overall Loss 0.323926 Objective Loss 0.323926 LR 0.001000 Time 0.021313 +2023-10-05 21:25:39,665 - Epoch: [92][ 1030/ 1236] Overall Loss 0.323667 Objective Loss 0.323667 LR 0.001000 Time 0.021302 +2023-10-05 21:25:39,868 - Epoch: [92][ 1040/ 1236] Overall Loss 0.323462 Objective Loss 0.323462 LR 0.001000 Time 0.021292 +2023-10-05 21:25:40,071 - Epoch: [92][ 1050/ 1236] Overall Loss 0.323330 Objective Loss 0.323330 LR 0.001000 Time 0.021282 +2023-10-05 21:25:40,274 - Epoch: [92][ 1060/ 1236] Overall Loss 0.323319 Objective Loss 0.323319 LR 0.001000 Time 0.021272 +2023-10-05 21:25:40,477 - Epoch: [92][ 1070/ 1236] Overall Loss 0.323510 Objective Loss 0.323510 LR 0.001000 Time 0.021263 +2023-10-05 21:25:40,680 - Epoch: [92][ 1080/ 1236] Overall Loss 0.323691 Objective Loss 0.323691 LR 0.001000 Time 0.021253 +2023-10-05 21:25:40,882 - Epoch: [92][ 1090/ 1236] Overall Loss 0.323555 Objective Loss 0.323555 LR 0.001000 Time 0.021244 +2023-10-05 21:25:41,085 - Epoch: [92][ 1100/ 1236] Overall Loss 0.323784 Objective Loss 0.323784 LR 0.001000 Time 0.021235 +2023-10-05 21:25:41,288 - Epoch: [92][ 1110/ 1236] Overall Loss 0.324113 Objective Loss 0.324113 LR 0.001000 Time 0.021226 +2023-10-05 21:25:41,490 - Epoch: [92][ 1120/ 1236] Overall Loss 0.324240 Objective Loss 0.324240 LR 0.001000 Time 0.021217 +2023-10-05 21:25:41,693 - Epoch: [92][ 1130/ 1236] Overall Loss 0.324532 Objective Loss 0.324532 LR 0.001000 Time 0.021208 +2023-10-05 21:25:41,896 - Epoch: [92][ 1140/ 1236] Overall Loss 0.324574 Objective Loss 0.324574 LR 0.001000 Time 0.021200 +2023-10-05 21:25:42,098 - Epoch: [92][ 1150/ 1236] Overall Loss 0.324610 Objective Loss 0.324610 LR 0.001000 Time 0.021191 +2023-10-05 21:25:42,302 - Epoch: [92][ 1160/ 1236] Overall Loss 0.324286 Objective Loss 0.324286 LR 0.001000 Time 0.021184 +2023-10-05 21:25:42,506 - Epoch: [92][ 1170/ 1236] Overall Loss 0.324272 Objective Loss 0.324272 LR 0.001000 Time 0.021177 +2023-10-05 21:25:42,710 - Epoch: [92][ 1180/ 1236] Overall Loss 0.324277 Objective Loss 0.324277 LR 0.001000 Time 0.021170 +2023-10-05 21:25:42,916 - Epoch: [92][ 1190/ 1236] Overall Loss 0.324323 Objective Loss 0.324323 LR 0.001000 Time 0.021165 +2023-10-05 21:25:43,120 - Epoch: [92][ 1200/ 1236] Overall Loss 0.324303 Objective Loss 0.324303 LR 0.001000 Time 0.021159 +2023-10-05 21:25:43,326 - Epoch: [92][ 1210/ 1236] Overall Loss 0.324297 Objective Loss 0.324297 LR 0.001000 Time 0.021154 +2023-10-05 21:25:43,531 - Epoch: [92][ 1220/ 1236] Overall Loss 0.324270 Objective Loss 0.324270 LR 0.001000 Time 0.021147 +2023-10-05 21:25:43,787 - Epoch: [92][ 1230/ 1236] Overall Loss 0.324426 Objective Loss 0.324426 LR 0.001000 Time 0.021183 +2023-10-05 21:25:43,904 - Epoch: [92][ 1236/ 1236] Overall Loss 0.324357 Objective Loss 0.324357 Top1 88.594705 Top5 98.370672 LR 0.001000 Time 0.021175 +2023-10-05 21:25:44,031 - --- validate (epoch=92)----------- +2023-10-05 21:25:44,031 - 29943 samples (256 per mini-batch) +2023-10-05 21:25:44,478 - Epoch: [92][ 10/ 117] Loss 0.383058 Top1 82.070312 Top5 97.226562 +2023-10-05 21:25:44,628 - Epoch: [92][ 20/ 117] Loss 0.389259 Top1 81.523438 Top5 97.011719 +2023-10-05 21:25:44,776 - Epoch: [92][ 30/ 117] Loss 0.374039 Top1 81.822917 Top5 97.213542 +2023-10-05 21:25:44,927 - Epoch: [92][ 40/ 117] Loss 0.366020 Top1 81.914062 Top5 97.373047 +2023-10-05 21:25:45,076 - Epoch: [92][ 50/ 117] Loss 0.362771 Top1 82.117188 Top5 97.390625 +2023-10-05 21:25:45,226 - Epoch: [92][ 60/ 117] Loss 0.366951 Top1 81.914062 Top5 97.408854 +2023-10-05 21:25:45,374 - Epoch: [92][ 70/ 117] Loss 0.371571 Top1 81.780134 Top5 97.393973 +2023-10-05 21:25:45,522 - Epoch: [92][ 80/ 117] Loss 0.372030 Top1 81.713867 Top5 97.402344 +2023-10-05 21:25:45,669 - Epoch: [92][ 90/ 117] Loss 0.370674 Top1 81.710069 Top5 97.456597 +2023-10-05 21:25:45,819 - Epoch: [92][ 100/ 117] Loss 0.369271 Top1 81.738281 Top5 97.480469 +2023-10-05 21:25:45,976 - Epoch: [92][ 110/ 117] Loss 0.369217 Top1 81.729403 Top5 97.471591 +2023-10-05 21:25:46,061 - Epoch: [92][ 117/ 117] Loss 0.366856 Top1 81.808770 Top5 97.485222 +2023-10-05 21:25:46,206 - ==> Top1: 81.809 Top5: 97.485 Loss: 0.367 + +2023-10-05 21:25:46,206 - ==> Confusion: +[[ 924 3 6 1 6 4 0 0 5 67 0 2 1 3 6 2 5 1 1 1 12] + [ 2 1021 3 1 13 30 1 14 2 0 2 2 0 1 2 4 13 0 8 1 11] + [ 4 0 931 31 5 1 22 5 0 1 4 3 10 2 2 6 3 0 8 9 9] + [ 2 0 14 978 1 6 2 0 2 1 5 0 2 4 23 7 1 6 24 2 9] + [ 22 4 0 0 973 4 1 0 3 13 0 2 1 1 7 3 10 1 0 1 4] + [ 1 29 0 1 9 987 0 14 5 3 3 10 4 12 8 2 8 0 1 6 13] + [ 0 5 33 1 1 0 1098 8 0 0 2 2 2 0 1 19 0 1 4 6 8] + [ 9 23 18 1 4 76 7 970 5 8 2 9 3 1 3 2 1 0 51 15 10] + [ 20 2 2 0 0 4 0 0 928 60 12 6 3 13 24 3 0 1 7 0 4] + [ 119 1 3 0 5 4 0 0 24 909 0 3 1 23 11 1 2 2 1 2 8] + [ 1 3 14 8 3 3 7 2 12 5 943 5 1 15 5 0 4 1 6 3 12] + [ 1 0 2 0 2 12 0 2 0 0 2 938 36 7 0 3 3 14 0 10 3] + [ 0 1 5 4 1 2 0 0 3 0 1 26 967 5 5 8 4 15 5 8 8] + [ 1 0 2 1 3 10 0 0 15 21 4 5 2 1038 2 1 2 1 0 1 10] + [ 15 0 5 14 12 0 0 0 17 5 1 2 1 2 1001 0 0 2 9 0 15] + [ 0 0 2 3 5 2 0 0 0 0 0 12 7 3 0 1053 20 11 2 7 7] + [ 1 15 2 1 5 4 0 0 3 0 0 6 0 1 5 12 1094 0 0 4 8] + [ 0 0 0 4 0 0 1 0 0 0 1 7 29 1 7 9 3 968 0 0 8] + [ 2 10 9 24 1 1 3 22 5 2 2 0 2 0 22 0 2 1 945 4 11] + [ 0 2 2 2 1 5 5 8 0 0 0 13 7 1 1 6 11 1 2 1076 9] + [ 135 186 154 90 121 216 37 67 88 94 185 121 358 322 182 87 277 72 150 209 4754]] + +2023-10-05 21:25:46,208 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:25:46,208 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:25:46,214 - + +2023-10-05 21:25:46,214 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:25:47,203 - Epoch: [93][ 10/ 1236] Overall Loss 0.329961 Objective Loss 0.329961 LR 0.001000 Time 0.098884 +2023-10-05 21:25:47,405 - Epoch: [93][ 20/ 1236] Overall Loss 0.315997 Objective Loss 0.315997 LR 0.001000 Time 0.059526 +2023-10-05 21:25:47,605 - Epoch: [93][ 30/ 1236] Overall Loss 0.315362 Objective Loss 0.315362 LR 0.001000 Time 0.046349 +2023-10-05 21:25:47,807 - Epoch: [93][ 40/ 1236] Overall Loss 0.312440 Objective Loss 0.312440 LR 0.001000 Time 0.039803 +2023-10-05 21:25:48,008 - Epoch: [93][ 50/ 1236] Overall Loss 0.310830 Objective Loss 0.310830 LR 0.001000 Time 0.035845 +2023-10-05 21:25:48,210 - Epoch: [93][ 60/ 1236] Overall Loss 0.314657 Objective Loss 0.314657 LR 0.001000 Time 0.033231 +2023-10-05 21:25:48,410 - Epoch: [93][ 70/ 1236] Overall Loss 0.311492 Objective Loss 0.311492 LR 0.001000 Time 0.031342 +2023-10-05 21:25:48,613 - Epoch: [93][ 80/ 1236] Overall Loss 0.309121 Objective Loss 0.309121 LR 0.001000 Time 0.029948 +2023-10-05 21:25:48,813 - Epoch: [93][ 90/ 1236] Overall Loss 0.309902 Objective Loss 0.309902 LR 0.001000 Time 0.028842 +2023-10-05 21:25:49,015 - Epoch: [93][ 100/ 1236] Overall Loss 0.313500 Objective Loss 0.313500 LR 0.001000 Time 0.027976 +2023-10-05 21:25:49,217 - Epoch: [93][ 110/ 1236] Overall Loss 0.315512 Objective Loss 0.315512 LR 0.001000 Time 0.027262 +2023-10-05 21:25:49,420 - Epoch: [93][ 120/ 1236] Overall Loss 0.315086 Objective Loss 0.315086 LR 0.001000 Time 0.026680 +2023-10-05 21:25:49,623 - Epoch: [93][ 130/ 1236] Overall Loss 0.312611 Objective Loss 0.312611 LR 0.001000 Time 0.026188 +2023-10-05 21:25:49,826 - Epoch: [93][ 140/ 1236] Overall Loss 0.312641 Objective Loss 0.312641 LR 0.001000 Time 0.025766 +2023-10-05 21:25:50,029 - Epoch: [93][ 150/ 1236] Overall Loss 0.314821 Objective Loss 0.314821 LR 0.001000 Time 0.025401 +2023-10-05 21:25:50,233 - Epoch: [93][ 160/ 1236] Overall Loss 0.316747 Objective Loss 0.316747 LR 0.001000 Time 0.025085 +2023-10-05 21:25:50,436 - Epoch: [93][ 170/ 1236] Overall Loss 0.315417 Objective Loss 0.315417 LR 0.001000 Time 0.024803 +2023-10-05 21:25:50,640 - Epoch: [93][ 180/ 1236] Overall Loss 0.316420 Objective Loss 0.316420 LR 0.001000 Time 0.024555 +2023-10-05 21:25:50,844 - Epoch: [93][ 190/ 1236] Overall Loss 0.316605 Objective Loss 0.316605 LR 0.001000 Time 0.024332 +2023-10-05 21:25:51,047 - Epoch: [93][ 200/ 1236] Overall Loss 0.314207 Objective Loss 0.314207 LR 0.001000 Time 0.024130 +2023-10-05 21:25:51,251 - Epoch: [93][ 210/ 1236] Overall Loss 0.315357 Objective Loss 0.315357 LR 0.001000 Time 0.023951 +2023-10-05 21:25:51,455 - Epoch: [93][ 220/ 1236] Overall Loss 0.314934 Objective Loss 0.314934 LR 0.001000 Time 0.023787 +2023-10-05 21:25:51,658 - Epoch: [93][ 230/ 1236] Overall Loss 0.314783 Objective Loss 0.314783 LR 0.001000 Time 0.023635 +2023-10-05 21:25:51,862 - Epoch: [93][ 240/ 1236] Overall Loss 0.314744 Objective Loss 0.314744 LR 0.001000 Time 0.023498 +2023-10-05 21:25:52,064 - Epoch: [93][ 250/ 1236] Overall Loss 0.315514 Objective Loss 0.315514 LR 0.001000 Time 0.023368 +2023-10-05 21:25:52,268 - Epoch: [93][ 260/ 1236] Overall Loss 0.315875 Objective Loss 0.315875 LR 0.001000 Time 0.023251 +2023-10-05 21:25:52,471 - Epoch: [93][ 270/ 1236] Overall Loss 0.315830 Objective Loss 0.315830 LR 0.001000 Time 0.023140 +2023-10-05 21:25:52,675 - Epoch: [93][ 280/ 1236] Overall Loss 0.316837 Objective Loss 0.316837 LR 0.001000 Time 0.023041 +2023-10-05 21:25:52,878 - Epoch: [93][ 290/ 1236] Overall Loss 0.318079 Objective Loss 0.318079 LR 0.001000 Time 0.022947 +2023-10-05 21:25:53,082 - Epoch: [93][ 300/ 1236] Overall Loss 0.317828 Objective Loss 0.317828 LR 0.001000 Time 0.022860 +2023-10-05 21:25:53,286 - Epoch: [93][ 310/ 1236] Overall Loss 0.317249 Objective Loss 0.317249 LR 0.001000 Time 0.022779 +2023-10-05 21:25:53,489 - Epoch: [93][ 320/ 1236] Overall Loss 0.317213 Objective Loss 0.317213 LR 0.001000 Time 0.022702 +2023-10-05 21:25:53,693 - Epoch: [93][ 330/ 1236] Overall Loss 0.317959 Objective Loss 0.317959 LR 0.001000 Time 0.022630 +2023-10-05 21:25:53,897 - Epoch: [93][ 340/ 1236] Overall Loss 0.317765 Objective Loss 0.317765 LR 0.001000 Time 0.022563 +2023-10-05 21:25:54,100 - Epoch: [93][ 350/ 1236] Overall Loss 0.317806 Objective Loss 0.317806 LR 0.001000 Time 0.022499 +2023-10-05 21:25:54,302 - Epoch: [93][ 360/ 1236] Overall Loss 0.318160 Objective Loss 0.318160 LR 0.001000 Time 0.022433 +2023-10-05 21:25:54,502 - Epoch: [93][ 370/ 1236] Overall Loss 0.318324 Objective Loss 0.318324 LR 0.001000 Time 0.022367 +2023-10-05 21:25:54,706 - Epoch: [93][ 380/ 1236] Overall Loss 0.319298 Objective Loss 0.319298 LR 0.001000 Time 0.022313 +2023-10-05 21:25:54,907 - Epoch: [93][ 390/ 1236] Overall Loss 0.319372 Objective Loss 0.319372 LR 0.001000 Time 0.022256 +2023-10-05 21:25:55,111 - Epoch: [93][ 400/ 1236] Overall Loss 0.319779 Objective Loss 0.319779 LR 0.001000 Time 0.022208 +2023-10-05 21:25:55,313 - Epoch: [93][ 410/ 1236] Overall Loss 0.319489 Objective Loss 0.319489 LR 0.001000 Time 0.022159 +2023-10-05 21:25:55,515 - Epoch: [93][ 420/ 1236] Overall Loss 0.319685 Objective Loss 0.319685 LR 0.001000 Time 0.022112 +2023-10-05 21:25:55,717 - Epoch: [93][ 430/ 1236] Overall Loss 0.320421 Objective Loss 0.320421 LR 0.001000 Time 0.022065 +2023-10-05 21:25:55,920 - Epoch: [93][ 440/ 1236] Overall Loss 0.320256 Objective Loss 0.320256 LR 0.001000 Time 0.022025 +2023-10-05 21:25:56,121 - Epoch: [93][ 450/ 1236] Overall Loss 0.320369 Objective Loss 0.320369 LR 0.001000 Time 0.021981 +2023-10-05 21:25:56,324 - Epoch: [93][ 460/ 1236] Overall Loss 0.319597 Objective Loss 0.319597 LR 0.001000 Time 0.021944 +2023-10-05 21:25:56,525 - Epoch: [93][ 470/ 1236] Overall Loss 0.319112 Objective Loss 0.319112 LR 0.001000 Time 0.021904 +2023-10-05 21:25:56,729 - Epoch: [93][ 480/ 1236] Overall Loss 0.318640 Objective Loss 0.318640 LR 0.001000 Time 0.021871 +2023-10-05 21:25:56,931 - Epoch: [93][ 490/ 1236] Overall Loss 0.319218 Objective Loss 0.319218 LR 0.001000 Time 0.021833 +2023-10-05 21:25:57,134 - Epoch: [93][ 500/ 1236] Overall Loss 0.318665 Objective Loss 0.318665 LR 0.001000 Time 0.021802 +2023-10-05 21:25:57,335 - Epoch: [93][ 510/ 1236] Overall Loss 0.318762 Objective Loss 0.318762 LR 0.001000 Time 0.021768 +2023-10-05 21:25:57,538 - Epoch: [93][ 520/ 1236] Overall Loss 0.317450 Objective Loss 0.317450 LR 0.001000 Time 0.021740 +2023-10-05 21:25:57,740 - Epoch: [93][ 530/ 1236] Overall Loss 0.316812 Objective Loss 0.316812 LR 0.001000 Time 0.021710 +2023-10-05 21:25:57,944 - Epoch: [93][ 540/ 1236] Overall Loss 0.317961 Objective Loss 0.317961 LR 0.001000 Time 0.021684 +2023-10-05 21:25:58,146 - Epoch: [93][ 550/ 1236] Overall Loss 0.318172 Objective Loss 0.318172 LR 0.001000 Time 0.021656 +2023-10-05 21:25:58,349 - Epoch: [93][ 560/ 1236] Overall Loss 0.318321 Objective Loss 0.318321 LR 0.001000 Time 0.021632 +2023-10-05 21:25:58,550 - Epoch: [93][ 570/ 1236] Overall Loss 0.318745 Objective Loss 0.318745 LR 0.001000 Time 0.021604 +2023-10-05 21:25:58,754 - Epoch: [93][ 580/ 1236] Overall Loss 0.318995 Objective Loss 0.318995 LR 0.001000 Time 0.021582 +2023-10-05 21:25:58,955 - Epoch: [93][ 590/ 1236] Overall Loss 0.319163 Objective Loss 0.319163 LR 0.001000 Time 0.021557 +2023-10-05 21:25:59,158 - Epoch: [93][ 600/ 1236] Overall Loss 0.319932 Objective Loss 0.319932 LR 0.001000 Time 0.021536 +2023-10-05 21:25:59,360 - Epoch: [93][ 610/ 1236] Overall Loss 0.320645 Objective Loss 0.320645 LR 0.001000 Time 0.021513 +2023-10-05 21:25:59,563 - Epoch: [93][ 620/ 1236] Overall Loss 0.321097 Objective Loss 0.321097 LR 0.001000 Time 0.021493 +2023-10-05 21:25:59,765 - Epoch: [93][ 630/ 1236] Overall Loss 0.321979 Objective Loss 0.321979 LR 0.001000 Time 0.021472 +2023-10-05 21:25:59,969 - Epoch: [93][ 640/ 1236] Overall Loss 0.322513 Objective Loss 0.322513 LR 0.001000 Time 0.021453 +2023-10-05 21:26:00,170 - Epoch: [93][ 650/ 1236] Overall Loss 0.322327 Objective Loss 0.322327 LR 0.001000 Time 0.021433 +2023-10-05 21:26:00,374 - Epoch: [93][ 660/ 1236] Overall Loss 0.322852 Objective Loss 0.322852 LR 0.001000 Time 0.021417 +2023-10-05 21:26:00,575 - Epoch: [93][ 670/ 1236] Overall Loss 0.323665 Objective Loss 0.323665 LR 0.001000 Time 0.021397 +2023-10-05 21:26:00,779 - Epoch: [93][ 680/ 1236] Overall Loss 0.323570 Objective Loss 0.323570 LR 0.001000 Time 0.021380 +2023-10-05 21:26:00,980 - Epoch: [93][ 690/ 1236] Overall Loss 0.323327 Objective Loss 0.323327 LR 0.001000 Time 0.021362 +2023-10-05 21:26:01,183 - Epoch: [93][ 700/ 1236] Overall Loss 0.323126 Objective Loss 0.323126 LR 0.001000 Time 0.021347 +2023-10-05 21:26:01,385 - Epoch: [93][ 710/ 1236] Overall Loss 0.323697 Objective Loss 0.323697 LR 0.001000 Time 0.021329 +2023-10-05 21:26:01,588 - Epoch: [93][ 720/ 1236] Overall Loss 0.323772 Objective Loss 0.323772 LR 0.001000 Time 0.021315 +2023-10-05 21:26:01,790 - Epoch: [93][ 730/ 1236] Overall Loss 0.323944 Objective Loss 0.323944 LR 0.001000 Time 0.021299 +2023-10-05 21:26:01,993 - Epoch: [93][ 740/ 1236] Overall Loss 0.324209 Objective Loss 0.324209 LR 0.001000 Time 0.021285 +2023-10-05 21:26:02,194 - Epoch: [93][ 750/ 1236] Overall Loss 0.324277 Objective Loss 0.324277 LR 0.001000 Time 0.021269 +2023-10-05 21:26:02,397 - Epoch: [93][ 760/ 1236] Overall Loss 0.324653 Objective Loss 0.324653 LR 0.001000 Time 0.021256 +2023-10-05 21:26:02,599 - Epoch: [93][ 770/ 1236] Overall Loss 0.324855 Objective Loss 0.324855 LR 0.001000 Time 0.021241 +2023-10-05 21:26:02,802 - Epoch: [93][ 780/ 1236] Overall Loss 0.324774 Objective Loss 0.324774 LR 0.001000 Time 0.021228 +2023-10-05 21:26:03,003 - Epoch: [93][ 790/ 1236] Overall Loss 0.325121 Objective Loss 0.325121 LR 0.001000 Time 0.021214 +2023-10-05 21:26:03,206 - Epoch: [93][ 800/ 1236] Overall Loss 0.325076 Objective Loss 0.325076 LR 0.001000 Time 0.021202 +2023-10-05 21:26:03,407 - Epoch: [93][ 810/ 1236] Overall Loss 0.325073 Objective Loss 0.325073 LR 0.001000 Time 0.021188 +2023-10-05 21:26:03,610 - Epoch: [93][ 820/ 1236] Overall Loss 0.325589 Objective Loss 0.325589 LR 0.001000 Time 0.021177 +2023-10-05 21:26:03,811 - Epoch: [93][ 830/ 1236] Overall Loss 0.325671 Objective Loss 0.325671 LR 0.001000 Time 0.021163 +2023-10-05 21:26:04,015 - Epoch: [93][ 840/ 1236] Overall Loss 0.326002 Objective Loss 0.326002 LR 0.001000 Time 0.021153 +2023-10-05 21:26:04,216 - Epoch: [93][ 850/ 1236] Overall Loss 0.325900 Objective Loss 0.325900 LR 0.001000 Time 0.021140 +2023-10-05 21:26:04,419 - Epoch: [93][ 860/ 1236] Overall Loss 0.325686 Objective Loss 0.325686 LR 0.001000 Time 0.021130 +2023-10-05 21:26:04,620 - Epoch: [93][ 870/ 1236] Overall Loss 0.325907 Objective Loss 0.325907 LR 0.001000 Time 0.021118 +2023-10-05 21:26:04,823 - Epoch: [93][ 880/ 1236] Overall Loss 0.326243 Objective Loss 0.326243 LR 0.001000 Time 0.021108 +2023-10-05 21:26:05,024 - Epoch: [93][ 890/ 1236] Overall Loss 0.326098 Objective Loss 0.326098 LR 0.001000 Time 0.021097 +2023-10-05 21:26:05,228 - Epoch: [93][ 900/ 1236] Overall Loss 0.326170 Objective Loss 0.326170 LR 0.001000 Time 0.021088 +2023-10-05 21:26:05,429 - Epoch: [93][ 910/ 1236] Overall Loss 0.326236 Objective Loss 0.326236 LR 0.001000 Time 0.021077 +2023-10-05 21:26:05,633 - Epoch: [93][ 920/ 1236] Overall Loss 0.326192 Objective Loss 0.326192 LR 0.001000 Time 0.021069 +2023-10-05 21:26:05,833 - Epoch: [93][ 930/ 1236] Overall Loss 0.325994 Objective Loss 0.325994 LR 0.001000 Time 0.021058 +2023-10-05 21:26:06,037 - Epoch: [93][ 940/ 1236] Overall Loss 0.326587 Objective Loss 0.326587 LR 0.001000 Time 0.021050 +2023-10-05 21:26:06,238 - Epoch: [93][ 950/ 1236] Overall Loss 0.326887 Objective Loss 0.326887 LR 0.001000 Time 0.021040 +2023-10-05 21:26:06,442 - Epoch: [93][ 960/ 1236] Overall Loss 0.327321 Objective Loss 0.327321 LR 0.001000 Time 0.021032 +2023-10-05 21:26:06,642 - Epoch: [93][ 970/ 1236] Overall Loss 0.327417 Objective Loss 0.327417 LR 0.001000 Time 0.021022 +2023-10-05 21:26:06,846 - Epoch: [93][ 980/ 1236] Overall Loss 0.327616 Objective Loss 0.327616 LR 0.001000 Time 0.021015 +2023-10-05 21:26:07,047 - Epoch: [93][ 990/ 1236] Overall Loss 0.327670 Objective Loss 0.327670 LR 0.001000 Time 0.021005 +2023-10-05 21:26:07,250 - Epoch: [93][ 1000/ 1236] Overall Loss 0.327650 Objective Loss 0.327650 LR 0.001000 Time 0.020998 +2023-10-05 21:26:07,451 - Epoch: [93][ 1010/ 1236] Overall Loss 0.327771 Objective Loss 0.327771 LR 0.001000 Time 0.020989 +2023-10-05 21:26:07,655 - Epoch: [93][ 1020/ 1236] Overall Loss 0.327897 Objective Loss 0.327897 LR 0.001000 Time 0.020982 +2023-10-05 21:26:07,855 - Epoch: [93][ 1030/ 1236] Overall Loss 0.327900 Objective Loss 0.327900 LR 0.001000 Time 0.020973 +2023-10-05 21:26:08,059 - Epoch: [93][ 1040/ 1236] Overall Loss 0.327528 Objective Loss 0.327528 LR 0.001000 Time 0.020966 +2023-10-05 21:26:08,260 - Epoch: [93][ 1050/ 1236] Overall Loss 0.327523 Objective Loss 0.327523 LR 0.001000 Time 0.020957 +2023-10-05 21:26:08,463 - Epoch: [93][ 1060/ 1236] Overall Loss 0.327418 Objective Loss 0.327418 LR 0.001000 Time 0.020951 +2023-10-05 21:26:08,665 - Epoch: [93][ 1070/ 1236] Overall Loss 0.328044 Objective Loss 0.328044 LR 0.001000 Time 0.020944 +2023-10-05 21:26:08,868 - Epoch: [93][ 1080/ 1236] Overall Loss 0.328103 Objective Loss 0.328103 LR 0.001000 Time 0.020938 +2023-10-05 21:26:09,070 - Epoch: [93][ 1090/ 1236] Overall Loss 0.328180 Objective Loss 0.328180 LR 0.001000 Time 0.020930 +2023-10-05 21:26:09,273 - Epoch: [93][ 1100/ 1236] Overall Loss 0.328116 Objective Loss 0.328116 LR 0.001000 Time 0.020925 +2023-10-05 21:26:09,476 - Epoch: [93][ 1110/ 1236] Overall Loss 0.327973 Objective Loss 0.327973 LR 0.001000 Time 0.020918 +2023-10-05 21:26:09,682 - Epoch: [93][ 1120/ 1236] Overall Loss 0.328283 Objective Loss 0.328283 LR 0.001000 Time 0.020915 +2023-10-05 21:26:09,885 - Epoch: [93][ 1130/ 1236] Overall Loss 0.328395 Objective Loss 0.328395 LR 0.001000 Time 0.020910 +2023-10-05 21:26:10,091 - Epoch: [93][ 1140/ 1236] Overall Loss 0.328653 Objective Loss 0.328653 LR 0.001000 Time 0.020907 +2023-10-05 21:26:10,294 - Epoch: [93][ 1150/ 1236] Overall Loss 0.328570 Objective Loss 0.328570 LR 0.001000 Time 0.020901 +2023-10-05 21:26:10,501 - Epoch: [93][ 1160/ 1236] Overall Loss 0.328964 Objective Loss 0.328964 LR 0.001000 Time 0.020899 +2023-10-05 21:26:10,703 - Epoch: [93][ 1170/ 1236] Overall Loss 0.329076 Objective Loss 0.329076 LR 0.001000 Time 0.020892 +2023-10-05 21:26:10,909 - Epoch: [93][ 1180/ 1236] Overall Loss 0.328976 Objective Loss 0.328976 LR 0.001000 Time 0.020889 +2023-10-05 21:26:11,111 - Epoch: [93][ 1190/ 1236] Overall Loss 0.329312 Objective Loss 0.329312 LR 0.001000 Time 0.020884 +2023-10-05 21:26:11,317 - Epoch: [93][ 1200/ 1236] Overall Loss 0.329397 Objective Loss 0.329397 LR 0.001000 Time 0.020881 +2023-10-05 21:26:11,519 - Epoch: [93][ 1210/ 1236] Overall Loss 0.329434 Objective Loss 0.329434 LR 0.001000 Time 0.020875 +2023-10-05 21:26:11,725 - Epoch: [93][ 1220/ 1236] Overall Loss 0.329351 Objective Loss 0.329351 LR 0.001000 Time 0.020873 +2023-10-05 21:26:11,982 - Epoch: [93][ 1230/ 1236] Overall Loss 0.329509 Objective Loss 0.329509 LR 0.001000 Time 0.020911 +2023-10-05 21:26:12,099 - Epoch: [93][ 1236/ 1236] Overall Loss 0.329542 Objective Loss 0.329542 Top1 81.873727 Top5 96.537678 LR 0.001000 Time 0.020904 +2023-10-05 21:26:12,231 - --- validate (epoch=93)----------- +2023-10-05 21:26:12,231 - 29943 samples (256 per mini-batch) +2023-10-05 21:26:12,685 - Epoch: [93][ 10/ 117] Loss 0.382383 Top1 81.289062 Top5 97.343750 +2023-10-05 21:26:12,836 - Epoch: [93][ 20/ 117] Loss 0.376845 Top1 80.898438 Top5 97.304688 +2023-10-05 21:26:12,987 - Epoch: [93][ 30/ 117] Loss 0.376503 Top1 80.846354 Top5 97.109375 +2023-10-05 21:26:13,137 - Epoch: [93][ 40/ 117] Loss 0.376853 Top1 80.781250 Top5 97.011719 +2023-10-05 21:26:13,283 - Epoch: [93][ 50/ 117] Loss 0.375131 Top1 80.718750 Top5 97.054688 +2023-10-05 21:26:13,433 - Epoch: [93][ 60/ 117] Loss 0.369336 Top1 80.820312 Top5 97.096354 +2023-10-05 21:26:13,581 - Epoch: [93][ 70/ 117] Loss 0.374465 Top1 80.591518 Top5 97.008929 +2023-10-05 21:26:13,730 - Epoch: [93][ 80/ 117] Loss 0.372545 Top1 80.664062 Top5 97.041016 +2023-10-05 21:26:13,878 - Epoch: [93][ 90/ 117] Loss 0.373813 Top1 80.425347 Top5 97.039931 +2023-10-05 21:26:14,029 - Epoch: [93][ 100/ 117] Loss 0.376595 Top1 80.402344 Top5 97.039062 +2023-10-05 21:26:14,185 - Epoch: [93][ 110/ 117] Loss 0.372568 Top1 80.621449 Top5 97.059659 +2023-10-05 21:26:14,272 - Epoch: [93][ 117/ 117] Loss 0.373503 Top1 80.656581 Top5 97.047724 +2023-10-05 21:26:14,395 - ==> Top1: 80.657 Top5: 97.048 Loss: 0.374 + +2023-10-05 21:26:14,396 - ==> Confusion: +[[ 932 3 1 1 11 4 0 1 4 55 1 0 1 3 8 3 11 1 0 0 10] + [ 3 1052 1 0 22 15 4 11 3 0 1 1 0 0 0 1 2 2 5 1 7] + [ 11 2 932 10 1 4 44 6 0 2 3 4 7 1 0 4 2 1 11 3 8] + [ 4 4 17 952 1 3 1 0 3 1 7 0 1 3 39 6 0 6 29 1 11] + [ 36 2 0 0 966 7 0 0 0 9 0 1 1 0 6 2 8 3 1 2 6] + [ 10 49 1 2 10 971 3 19 2 0 6 5 3 6 6 1 5 0 3 6 8] + [ 1 7 23 0 0 0 1131 4 0 0 4 4 1 0 1 6 0 2 0 4 3] + [ 7 35 13 1 8 45 10 1011 4 1 5 9 3 0 0 2 2 2 46 8 6] + [ 35 0 0 0 1 6 1 0 938 60 16 2 3 7 11 1 0 4 1 2 1] + [ 153 0 1 1 9 3 1 0 15 891 0 2 0 18 8 5 0 3 0 1 8] + [ 1 5 12 7 3 6 7 1 17 1 954 2 0 8 9 2 1 3 2 3 9] + [ 1 0 1 0 4 13 0 2 0 1 0 918 49 4 1 2 2 16 0 15 6] + [ 2 1 5 8 1 1 1 1 6 0 0 43 947 0 3 6 4 24 6 5 4] + [ 0 1 0 1 7 29 1 1 23 26 6 5 6 989 5 0 1 3 0 5 10] + [ 20 4 3 14 6 0 0 0 27 8 0 1 3 0 994 0 0 3 11 0 7] + [ 0 3 0 1 8 2 4 0 0 0 0 11 9 1 1 1046 19 13 2 7 7] + [ 2 22 2 2 7 4 1 1 2 0 1 5 1 1 4 7 1082 0 0 8 9] + [ 3 0 0 1 0 0 0 0 0 0 0 7 20 0 0 9 1 994 1 0 2] + [ 3 8 8 14 2 1 3 21 3 0 2 1 3 0 14 0 1 0 973 3 8] + [ 0 7 1 1 3 10 14 9 1 0 3 13 7 2 0 4 14 2 2 1051 8] + [ 191 248 135 75 172 196 93 72 126 106 179 146 370 232 155 64 319 102 253 243 4428]] + +2023-10-05 21:26:14,397 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:26:14,397 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:26:14,404 - + +2023-10-05 21:26:14,404 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:26:15,408 - Epoch: [94][ 10/ 1236] Overall Loss 0.300253 Objective Loss 0.300253 LR 0.001000 Time 0.100337 +2023-10-05 21:26:15,611 - Epoch: [94][ 20/ 1236] Overall Loss 0.325243 Objective Loss 0.325243 LR 0.001000 Time 0.060338 +2023-10-05 21:26:15,812 - Epoch: [94][ 30/ 1236] Overall Loss 0.324811 Objective Loss 0.324811 LR 0.001000 Time 0.046910 +2023-10-05 21:26:16,016 - Epoch: [94][ 40/ 1236] Overall Loss 0.328620 Objective Loss 0.328620 LR 0.001000 Time 0.040264 +2023-10-05 21:26:16,218 - Epoch: [94][ 50/ 1236] Overall Loss 0.327795 Objective Loss 0.327795 LR 0.001000 Time 0.036239 +2023-10-05 21:26:16,421 - Epoch: [94][ 60/ 1236] Overall Loss 0.329387 Objective Loss 0.329387 LR 0.001000 Time 0.033584 +2023-10-05 21:26:16,623 - Epoch: [94][ 70/ 1236] Overall Loss 0.327782 Objective Loss 0.327782 LR 0.001000 Time 0.031660 +2023-10-05 21:26:16,827 - Epoch: [94][ 80/ 1236] Overall Loss 0.325142 Objective Loss 0.325142 LR 0.001000 Time 0.030246 +2023-10-05 21:26:17,029 - Epoch: [94][ 90/ 1236] Overall Loss 0.322576 Objective Loss 0.322576 LR 0.001000 Time 0.029128 +2023-10-05 21:26:17,233 - Epoch: [94][ 100/ 1236] Overall Loss 0.324417 Objective Loss 0.324417 LR 0.001000 Time 0.028253 +2023-10-05 21:26:17,436 - Epoch: [94][ 110/ 1236] Overall Loss 0.325461 Objective Loss 0.325461 LR 0.001000 Time 0.027526 +2023-10-05 21:26:17,640 - Epoch: [94][ 120/ 1236] Overall Loss 0.322818 Objective Loss 0.322818 LR 0.001000 Time 0.026932 +2023-10-05 21:26:17,841 - Epoch: [94][ 130/ 1236] Overall Loss 0.324168 Objective Loss 0.324168 LR 0.001000 Time 0.026405 +2023-10-05 21:26:18,046 - Epoch: [94][ 140/ 1236] Overall Loss 0.325788 Objective Loss 0.325788 LR 0.001000 Time 0.025978 +2023-10-05 21:26:18,248 - Epoch: [94][ 150/ 1236] Overall Loss 0.324845 Objective Loss 0.324845 LR 0.001000 Time 0.025588 +2023-10-05 21:26:18,451 - Epoch: [94][ 160/ 1236] Overall Loss 0.323396 Objective Loss 0.323396 LR 0.001000 Time 0.025258 +2023-10-05 21:26:18,653 - Epoch: [94][ 170/ 1236] Overall Loss 0.322138 Objective Loss 0.322138 LR 0.001000 Time 0.024960 +2023-10-05 21:26:18,858 - Epoch: [94][ 180/ 1236] Overall Loss 0.324008 Objective Loss 0.324008 LR 0.001000 Time 0.024707 +2023-10-05 21:26:19,063 - Epoch: [94][ 190/ 1236] Overall Loss 0.323501 Objective Loss 0.323501 LR 0.001000 Time 0.024485 +2023-10-05 21:26:19,269 - Epoch: [94][ 200/ 1236] Overall Loss 0.323203 Objective Loss 0.323203 LR 0.001000 Time 0.024291 +2023-10-05 21:26:19,473 - Epoch: [94][ 210/ 1236] Overall Loss 0.322730 Objective Loss 0.322730 LR 0.001000 Time 0.024103 +2023-10-05 21:26:19,679 - Epoch: [94][ 220/ 1236] Overall Loss 0.323162 Objective Loss 0.323162 LR 0.001000 Time 0.023942 +2023-10-05 21:26:19,884 - Epoch: [94][ 230/ 1236] Overall Loss 0.323668 Objective Loss 0.323668 LR 0.001000 Time 0.023792 +2023-10-05 21:26:20,091 - Epoch: [94][ 240/ 1236] Overall Loss 0.323497 Objective Loss 0.323497 LR 0.001000 Time 0.023659 +2023-10-05 21:26:20,296 - Epoch: [94][ 250/ 1236] Overall Loss 0.322768 Objective Loss 0.322768 LR 0.001000 Time 0.023531 +2023-10-05 21:26:20,502 - Epoch: [94][ 260/ 1236] Overall Loss 0.322030 Objective Loss 0.322030 LR 0.001000 Time 0.023419 +2023-10-05 21:26:20,707 - Epoch: [94][ 270/ 1236] Overall Loss 0.321080 Objective Loss 0.321080 LR 0.001000 Time 0.023310 +2023-10-05 21:26:20,914 - Epoch: [94][ 280/ 1236] Overall Loss 0.322528 Objective Loss 0.322528 LR 0.001000 Time 0.023214 +2023-10-05 21:26:21,119 - Epoch: [94][ 290/ 1236] Overall Loss 0.321499 Objective Loss 0.321499 LR 0.001000 Time 0.023118 +2023-10-05 21:26:21,325 - Epoch: [94][ 300/ 1236] Overall Loss 0.322274 Objective Loss 0.322274 LR 0.001000 Time 0.023034 +2023-10-05 21:26:21,530 - Epoch: [94][ 310/ 1236] Overall Loss 0.322102 Objective Loss 0.322102 LR 0.001000 Time 0.022952 +2023-10-05 21:26:21,737 - Epoch: [94][ 320/ 1236] Overall Loss 0.323750 Objective Loss 0.323750 LR 0.001000 Time 0.022879 +2023-10-05 21:26:21,941 - Epoch: [94][ 330/ 1236] Overall Loss 0.324758 Objective Loss 0.324758 LR 0.001000 Time 0.022804 +2023-10-05 21:26:22,147 - Epoch: [94][ 340/ 1236] Overall Loss 0.324082 Objective Loss 0.324082 LR 0.001000 Time 0.022737 +2023-10-05 21:26:22,352 - Epoch: [94][ 350/ 1236] Overall Loss 0.324192 Objective Loss 0.324192 LR 0.001000 Time 0.022673 +2023-10-05 21:26:22,558 - Epoch: [94][ 360/ 1236] Overall Loss 0.324766 Objective Loss 0.324766 LR 0.001000 Time 0.022614 +2023-10-05 21:26:22,762 - Epoch: [94][ 370/ 1236] Overall Loss 0.324653 Objective Loss 0.324653 LR 0.001000 Time 0.022554 +2023-10-05 21:26:22,966 - Epoch: [94][ 380/ 1236] Overall Loss 0.324189 Objective Loss 0.324189 LR 0.001000 Time 0.022495 +2023-10-05 21:26:23,169 - Epoch: [94][ 390/ 1236] Overall Loss 0.324687 Objective Loss 0.324687 LR 0.001000 Time 0.022439 +2023-10-05 21:26:23,373 - Epoch: [94][ 400/ 1236] Overall Loss 0.324432 Objective Loss 0.324432 LR 0.001000 Time 0.022386 +2023-10-05 21:26:23,577 - Epoch: [94][ 410/ 1236] Overall Loss 0.325293 Objective Loss 0.325293 LR 0.001000 Time 0.022338 +2023-10-05 21:26:23,780 - Epoch: [94][ 420/ 1236] Overall Loss 0.326302 Objective Loss 0.326302 LR 0.001000 Time 0.022289 +2023-10-05 21:26:23,984 - Epoch: [94][ 430/ 1236] Overall Loss 0.326398 Objective Loss 0.326398 LR 0.001000 Time 0.022244 +2023-10-05 21:26:24,188 - Epoch: [94][ 440/ 1236] Overall Loss 0.327071 Objective Loss 0.327071 LR 0.001000 Time 0.022201 +2023-10-05 21:26:24,392 - Epoch: [94][ 450/ 1236] Overall Loss 0.327786 Objective Loss 0.327786 LR 0.001000 Time 0.022159 +2023-10-05 21:26:24,595 - Epoch: [94][ 460/ 1236] Overall Loss 0.327746 Objective Loss 0.327746 LR 0.001000 Time 0.022119 +2023-10-05 21:26:24,799 - Epoch: [94][ 470/ 1236] Overall Loss 0.328253 Objective Loss 0.328253 LR 0.001000 Time 0.022081 +2023-10-05 21:26:25,002 - Epoch: [94][ 480/ 1236] Overall Loss 0.328048 Objective Loss 0.328048 LR 0.001000 Time 0.022045 +2023-10-05 21:26:25,206 - Epoch: [94][ 490/ 1236] Overall Loss 0.328743 Objective Loss 0.328743 LR 0.001000 Time 0.022010 +2023-10-05 21:26:25,410 - Epoch: [94][ 500/ 1236] Overall Loss 0.329502 Objective Loss 0.329502 LR 0.001000 Time 0.021977 +2023-10-05 21:26:25,614 - Epoch: [94][ 510/ 1236] Overall Loss 0.330034 Objective Loss 0.330034 LR 0.001000 Time 0.021944 +2023-10-05 21:26:25,818 - Epoch: [94][ 520/ 1236] Overall Loss 0.330630 Objective Loss 0.330630 LR 0.001000 Time 0.021914 +2023-10-05 21:26:26,022 - Epoch: [94][ 530/ 1236] Overall Loss 0.330335 Objective Loss 0.330335 LR 0.001000 Time 0.021885 +2023-10-05 21:26:26,225 - Epoch: [94][ 540/ 1236] Overall Loss 0.330244 Objective Loss 0.330244 LR 0.001000 Time 0.021856 +2023-10-05 21:26:26,430 - Epoch: [94][ 550/ 1236] Overall Loss 0.329858 Objective Loss 0.329858 LR 0.001000 Time 0.021829 +2023-10-05 21:26:26,633 - Epoch: [94][ 560/ 1236] Overall Loss 0.330398 Objective Loss 0.330398 LR 0.001000 Time 0.021803 +2023-10-05 21:26:26,837 - Epoch: [94][ 570/ 1236] Overall Loss 0.330644 Objective Loss 0.330644 LR 0.001000 Time 0.021777 +2023-10-05 21:26:27,041 - Epoch: [94][ 580/ 1236] Overall Loss 0.330406 Objective Loss 0.330406 LR 0.001000 Time 0.021752 +2023-10-05 21:26:27,244 - Epoch: [94][ 590/ 1236] Overall Loss 0.330954 Objective Loss 0.330954 LR 0.001000 Time 0.021728 +2023-10-05 21:26:27,448 - Epoch: [94][ 600/ 1236] Overall Loss 0.330804 Objective Loss 0.330804 LR 0.001000 Time 0.021705 +2023-10-05 21:26:27,652 - Epoch: [94][ 610/ 1236] Overall Loss 0.330929 Objective Loss 0.330929 LR 0.001000 Time 0.021683 +2023-10-05 21:26:27,856 - Epoch: [94][ 620/ 1236] Overall Loss 0.331197 Objective Loss 0.331197 LR 0.001000 Time 0.021662 +2023-10-05 21:26:28,060 - Epoch: [94][ 630/ 1236] Overall Loss 0.331152 Objective Loss 0.331152 LR 0.001000 Time 0.021641 +2023-10-05 21:26:28,264 - Epoch: [94][ 640/ 1236] Overall Loss 0.330991 Objective Loss 0.330991 LR 0.001000 Time 0.021621 +2023-10-05 21:26:28,467 - Epoch: [94][ 650/ 1236] Overall Loss 0.330903 Objective Loss 0.330903 LR 0.001000 Time 0.021600 +2023-10-05 21:26:28,671 - Epoch: [94][ 660/ 1236] Overall Loss 0.330689 Objective Loss 0.330689 LR 0.001000 Time 0.021581 +2023-10-05 21:26:28,875 - Epoch: [94][ 670/ 1236] Overall Loss 0.330526 Objective Loss 0.330526 LR 0.001000 Time 0.021563 +2023-10-05 21:26:29,079 - Epoch: [94][ 680/ 1236] Overall Loss 0.329839 Objective Loss 0.329839 LR 0.001000 Time 0.021546 +2023-10-05 21:26:29,284 - Epoch: [94][ 690/ 1236] Overall Loss 0.330066 Objective Loss 0.330066 LR 0.001000 Time 0.021530 +2023-10-05 21:26:29,488 - Epoch: [94][ 700/ 1236] Overall Loss 0.329749 Objective Loss 0.329749 LR 0.001000 Time 0.021514 +2023-10-05 21:26:29,692 - Epoch: [94][ 710/ 1236] Overall Loss 0.329765 Objective Loss 0.329765 LR 0.001000 Time 0.021497 +2023-10-05 21:26:29,896 - Epoch: [94][ 720/ 1236] Overall Loss 0.329977 Objective Loss 0.329977 LR 0.001000 Time 0.021481 +2023-10-05 21:26:30,100 - Epoch: [94][ 730/ 1236] Overall Loss 0.330301 Objective Loss 0.330301 LR 0.001000 Time 0.021466 +2023-10-05 21:26:30,303 - Epoch: [94][ 740/ 1236] Overall Loss 0.330397 Objective Loss 0.330397 LR 0.001000 Time 0.021450 +2023-10-05 21:26:30,507 - Epoch: [94][ 750/ 1236] Overall Loss 0.330450 Objective Loss 0.330450 LR 0.001000 Time 0.021436 +2023-10-05 21:26:30,711 - Epoch: [94][ 760/ 1236] Overall Loss 0.330043 Objective Loss 0.330043 LR 0.001000 Time 0.021421 +2023-10-05 21:26:30,914 - Epoch: [94][ 770/ 1236] Overall Loss 0.329844 Objective Loss 0.329844 LR 0.001000 Time 0.021407 +2023-10-05 21:26:31,119 - Epoch: [94][ 780/ 1236] Overall Loss 0.330228 Objective Loss 0.330228 LR 0.001000 Time 0.021394 +2023-10-05 21:26:31,323 - Epoch: [94][ 790/ 1236] Overall Loss 0.330311 Objective Loss 0.330311 LR 0.001000 Time 0.021381 +2023-10-05 21:26:31,526 - Epoch: [94][ 800/ 1236] Overall Loss 0.330587 Objective Loss 0.330587 LR 0.001000 Time 0.021368 +2023-10-05 21:26:31,731 - Epoch: [94][ 810/ 1236] Overall Loss 0.330812 Objective Loss 0.330812 LR 0.001000 Time 0.021356 +2023-10-05 21:26:31,935 - Epoch: [94][ 820/ 1236] Overall Loss 0.330659 Objective Loss 0.330659 LR 0.001000 Time 0.021344 +2023-10-05 21:26:32,138 - Epoch: [94][ 830/ 1236] Overall Loss 0.330292 Objective Loss 0.330292 LR 0.001000 Time 0.021331 +2023-10-05 21:26:32,342 - Epoch: [94][ 840/ 1236] Overall Loss 0.330365 Objective Loss 0.330365 LR 0.001000 Time 0.021320 +2023-10-05 21:26:32,546 - Epoch: [94][ 850/ 1236] Overall Loss 0.330314 Objective Loss 0.330314 LR 0.001000 Time 0.021309 +2023-10-05 21:26:32,751 - Epoch: [94][ 860/ 1236] Overall Loss 0.330817 Objective Loss 0.330817 LR 0.001000 Time 0.021299 +2023-10-05 21:26:32,954 - Epoch: [94][ 870/ 1236] Overall Loss 0.331226 Objective Loss 0.331226 LR 0.001000 Time 0.021287 +2023-10-05 21:26:33,159 - Epoch: [94][ 880/ 1236] Overall Loss 0.331333 Objective Loss 0.331333 LR 0.001000 Time 0.021277 +2023-10-05 21:26:33,363 - Epoch: [94][ 890/ 1236] Overall Loss 0.331201 Objective Loss 0.331201 LR 0.001000 Time 0.021267 +2023-10-05 21:26:33,567 - Epoch: [94][ 900/ 1236] Overall Loss 0.331169 Objective Loss 0.331169 LR 0.001000 Time 0.021257 +2023-10-05 21:26:33,771 - Epoch: [94][ 910/ 1236] Overall Loss 0.331741 Objective Loss 0.331741 LR 0.001000 Time 0.021247 +2023-10-05 21:26:33,975 - Epoch: [94][ 920/ 1236] Overall Loss 0.331574 Objective Loss 0.331574 LR 0.001000 Time 0.021238 +2023-10-05 21:26:34,179 - Epoch: [94][ 930/ 1236] Overall Loss 0.331530 Objective Loss 0.331530 LR 0.001000 Time 0.021229 +2023-10-05 21:26:34,383 - Epoch: [94][ 940/ 1236] Overall Loss 0.331515 Objective Loss 0.331515 LR 0.001000 Time 0.021220 +2023-10-05 21:26:34,588 - Epoch: [94][ 950/ 1236] Overall Loss 0.331599 Objective Loss 0.331599 LR 0.001000 Time 0.021211 +2023-10-05 21:26:34,792 - Epoch: [94][ 960/ 1236] Overall Loss 0.331785 Objective Loss 0.331785 LR 0.001000 Time 0.021202 +2023-10-05 21:26:34,995 - Epoch: [94][ 970/ 1236] Overall Loss 0.331971 Objective Loss 0.331971 LR 0.001000 Time 0.021193 +2023-10-05 21:26:35,199 - Epoch: [94][ 980/ 1236] Overall Loss 0.332109 Objective Loss 0.332109 LR 0.001000 Time 0.021184 +2023-10-05 21:26:35,402 - Epoch: [94][ 990/ 1236] Overall Loss 0.331994 Objective Loss 0.331994 LR 0.001000 Time 0.021175 +2023-10-05 21:26:35,606 - Epoch: [94][ 1000/ 1236] Overall Loss 0.331924 Objective Loss 0.331924 LR 0.001000 Time 0.021168 +2023-10-05 21:26:35,809 - Epoch: [94][ 1010/ 1236] Overall Loss 0.332075 Objective Loss 0.332075 LR 0.001000 Time 0.021158 +2023-10-05 21:26:36,013 - Epoch: [94][ 1020/ 1236] Overall Loss 0.332020 Objective Loss 0.332020 LR 0.001000 Time 0.021151 +2023-10-05 21:26:36,217 - Epoch: [94][ 1030/ 1236] Overall Loss 0.332188 Objective Loss 0.332188 LR 0.001000 Time 0.021143 +2023-10-05 21:26:36,421 - Epoch: [94][ 1040/ 1236] Overall Loss 0.332171 Objective Loss 0.332171 LR 0.001000 Time 0.021135 +2023-10-05 21:26:36,623 - Epoch: [94][ 1050/ 1236] Overall Loss 0.331879 Objective Loss 0.331879 LR 0.001000 Time 0.021126 +2023-10-05 21:26:36,827 - Epoch: [94][ 1060/ 1236] Overall Loss 0.331760 Objective Loss 0.331760 LR 0.001000 Time 0.021119 +2023-10-05 21:26:37,031 - Epoch: [94][ 1070/ 1236] Overall Loss 0.331808 Objective Loss 0.331808 LR 0.001000 Time 0.021111 +2023-10-05 21:26:37,235 - Epoch: [94][ 1080/ 1236] Overall Loss 0.331817 Objective Loss 0.331817 LR 0.001000 Time 0.021105 +2023-10-05 21:26:37,438 - Epoch: [94][ 1090/ 1236] Overall Loss 0.331831 Objective Loss 0.331831 LR 0.001000 Time 0.021098 +2023-10-05 21:26:37,642 - Epoch: [94][ 1100/ 1236] Overall Loss 0.331915 Objective Loss 0.331915 LR 0.001000 Time 0.021091 +2023-10-05 21:26:37,845 - Epoch: [94][ 1110/ 1236] Overall Loss 0.331637 Objective Loss 0.331637 LR 0.001000 Time 0.021083 +2023-10-05 21:26:38,049 - Epoch: [94][ 1120/ 1236] Overall Loss 0.331566 Objective Loss 0.331566 LR 0.001000 Time 0.021077 +2023-10-05 21:26:38,253 - Epoch: [94][ 1130/ 1236] Overall Loss 0.331808 Objective Loss 0.331808 LR 0.001000 Time 0.021070 +2023-10-05 21:26:38,457 - Epoch: [94][ 1140/ 1236] Overall Loss 0.331732 Objective Loss 0.331732 LR 0.001000 Time 0.021064 +2023-10-05 21:26:38,660 - Epoch: [94][ 1150/ 1236] Overall Loss 0.331483 Objective Loss 0.331483 LR 0.001000 Time 0.021057 +2023-10-05 21:26:38,864 - Epoch: [94][ 1160/ 1236] Overall Loss 0.331461 Objective Loss 0.331461 LR 0.001000 Time 0.021051 +2023-10-05 21:26:39,068 - Epoch: [94][ 1170/ 1236] Overall Loss 0.331332 Objective Loss 0.331332 LR 0.001000 Time 0.021045 +2023-10-05 21:26:39,272 - Epoch: [94][ 1180/ 1236] Overall Loss 0.331037 Objective Loss 0.331037 LR 0.001000 Time 0.021039 +2023-10-05 21:26:39,476 - Epoch: [94][ 1190/ 1236] Overall Loss 0.330906 Objective Loss 0.330906 LR 0.001000 Time 0.021033 +2023-10-05 21:26:39,680 - Epoch: [94][ 1200/ 1236] Overall Loss 0.330488 Objective Loss 0.330488 LR 0.001000 Time 0.021028 +2023-10-05 21:26:39,885 - Epoch: [94][ 1210/ 1236] Overall Loss 0.330365 Objective Loss 0.330365 LR 0.001000 Time 0.021024 +2023-10-05 21:26:40,089 - Epoch: [94][ 1220/ 1236] Overall Loss 0.330485 Objective Loss 0.330485 LR 0.001000 Time 0.021019 +2023-10-05 21:26:40,346 - Epoch: [94][ 1230/ 1236] Overall Loss 0.330281 Objective Loss 0.330281 LR 0.001000 Time 0.021056 +2023-10-05 21:26:40,465 - Epoch: [94][ 1236/ 1236] Overall Loss 0.330264 Objective Loss 0.330264 Top1 82.484725 Top5 97.963340 LR 0.001000 Time 0.021050 +2023-10-05 21:26:40,620 - --- validate (epoch=94)----------- +2023-10-05 21:26:40,620 - 29943 samples (256 per mini-batch) +2023-10-05 21:26:41,081 - Epoch: [94][ 10/ 117] Loss 0.371556 Top1 80.898438 Top5 97.851562 +2023-10-05 21:26:41,233 - Epoch: [94][ 20/ 117] Loss 0.366821 Top1 81.406250 Top5 97.558594 +2023-10-05 21:26:41,382 - Epoch: [94][ 30/ 117] Loss 0.369182 Top1 81.250000 Top5 97.434896 +2023-10-05 21:26:41,535 - Epoch: [94][ 40/ 117] Loss 0.368037 Top1 81.396484 Top5 97.460938 +2023-10-05 21:26:41,685 - Epoch: [94][ 50/ 117] Loss 0.372217 Top1 81.414062 Top5 97.476562 +2023-10-05 21:26:41,846 - Epoch: [94][ 60/ 117] Loss 0.368702 Top1 81.490885 Top5 97.558594 +2023-10-05 21:26:41,997 - Epoch: [94][ 70/ 117] Loss 0.368828 Top1 81.450893 Top5 97.427455 +2023-10-05 21:26:42,155 - Epoch: [94][ 80/ 117] Loss 0.367415 Top1 81.386719 Top5 97.402344 +2023-10-05 21:26:42,312 - Epoch: [94][ 90/ 117] Loss 0.369377 Top1 81.471354 Top5 97.404514 +2023-10-05 21:26:42,471 - Epoch: [94][ 100/ 117] Loss 0.367484 Top1 81.476562 Top5 97.390625 +2023-10-05 21:26:42,637 - Epoch: [94][ 110/ 117] Loss 0.365845 Top1 81.516335 Top5 97.354403 +2023-10-05 21:26:42,723 - Epoch: [94][ 117/ 117] Loss 0.364389 Top1 81.524897 Top5 97.395051 +2023-10-05 21:26:42,855 - ==> Top1: 81.525 Top5: 97.395 Loss: 0.364 + +2023-10-05 21:26:42,856 - ==> Confusion: +[[ 926 3 2 0 13 3 0 0 7 58 2 0 0 2 10 1 12 1 0 0 10] + [ 1 1071 0 0 9 12 2 15 2 0 1 2 0 0 0 1 4 0 6 0 5] + [ 5 1 941 19 3 0 24 10 0 2 5 0 8 2 2 3 3 0 9 7 12] + [ 3 2 18 956 1 4 0 2 1 3 15 0 3 1 24 3 1 3 35 2 12] + [ 29 9 0 1 960 2 0 0 0 5 2 3 0 1 8 7 9 1 0 3 10] + [ 5 74 0 0 7 933 1 30 4 1 2 8 3 15 6 2 3 0 4 6 12] + [ 0 7 38 0 0 2 1098 14 0 0 5 1 0 0 1 8 1 1 4 7 4] + [ 1 29 14 0 2 30 7 1062 1 1 3 8 4 0 1 1 3 0 35 9 7] + [ 24 1 0 0 0 0 0 1 958 37 18 3 1 11 18 6 5 1 2 0 3] + [ 114 0 1 0 13 1 1 0 45 884 0 1 0 29 11 8 0 0 1 1 9] + [ 1 3 10 5 2 4 4 7 18 0 959 2 0 7 3 2 1 0 9 5 11] + [ 1 0 2 0 3 19 0 5 0 2 0 910 34 5 0 2 5 15 0 27 5] + [ 2 2 4 8 1 5 2 2 0 0 0 33 952 8 3 5 12 15 2 5 7] + [ 1 2 2 0 5 10 0 2 16 9 8 3 0 1037 6 2 5 1 0 3 7] + [ 19 0 3 15 5 0 0 0 27 5 7 1 1 1 981 1 0 4 19 0 12] + [ 0 0 6 2 6 2 3 0 0 0 0 7 6 1 0 1055 20 10 3 7 6] + [ 1 18 1 0 6 4 1 1 2 0 0 2 0 1 4 7 1102 1 0 5 5] + [ 2 1 1 2 0 0 0 0 0 2 0 2 24 1 7 9 2 977 1 2 5] + [ 4 13 8 12 1 0 1 31 2 0 2 0 1 0 10 0 3 0 970 0 10] + [ 0 2 2 1 3 9 12 17 1 0 4 9 5 3 0 2 16 1 2 1057 6] + [ 160 307 167 77 112 124 51 127 89 69 179 117 385 331 148 57 264 69 198 252 4622]] + +2023-10-05 21:26:42,858 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:26:42,858 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:26:42,864 - + +2023-10-05 21:26:42,864 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:26:43,870 - Epoch: [95][ 10/ 1236] Overall Loss 0.286045 Objective Loss 0.286045 LR 0.001000 Time 0.100579 +2023-10-05 21:26:44,074 - Epoch: [95][ 20/ 1236] Overall Loss 0.298377 Objective Loss 0.298377 LR 0.001000 Time 0.060492 +2023-10-05 21:26:44,278 - Epoch: [95][ 30/ 1236] Overall Loss 0.306935 Objective Loss 0.306935 LR 0.001000 Time 0.047117 +2023-10-05 21:26:44,483 - Epoch: [95][ 40/ 1236] Overall Loss 0.312601 Objective Loss 0.312601 LR 0.001000 Time 0.040435 +2023-10-05 21:26:44,686 - Epoch: [95][ 50/ 1236] Overall Loss 0.315964 Objective Loss 0.315964 LR 0.001000 Time 0.036405 +2023-10-05 21:26:44,889 - Epoch: [95][ 60/ 1236] Overall Loss 0.310252 Objective Loss 0.310252 LR 0.001000 Time 0.033721 +2023-10-05 21:26:45,093 - Epoch: [95][ 70/ 1236] Overall Loss 0.312339 Objective Loss 0.312339 LR 0.001000 Time 0.031813 +2023-10-05 21:26:45,296 - Epoch: [95][ 80/ 1236] Overall Loss 0.310777 Objective Loss 0.310777 LR 0.001000 Time 0.030370 +2023-10-05 21:26:45,500 - Epoch: [95][ 90/ 1236] Overall Loss 0.309225 Objective Loss 0.309225 LR 0.001000 Time 0.029261 +2023-10-05 21:26:45,706 - Epoch: [95][ 100/ 1236] Overall Loss 0.310990 Objective Loss 0.310990 LR 0.001000 Time 0.028383 +2023-10-05 21:26:45,906 - Epoch: [95][ 110/ 1236] Overall Loss 0.313769 Objective Loss 0.313769 LR 0.001000 Time 0.027619 +2023-10-05 21:26:46,108 - Epoch: [95][ 120/ 1236] Overall Loss 0.314833 Objective Loss 0.314833 LR 0.001000 Time 0.027004 +2023-10-05 21:26:46,311 - Epoch: [95][ 130/ 1236] Overall Loss 0.316264 Objective Loss 0.316264 LR 0.001000 Time 0.026479 +2023-10-05 21:26:46,525 - Epoch: [95][ 140/ 1236] Overall Loss 0.316506 Objective Loss 0.316506 LR 0.001000 Time 0.026119 +2023-10-05 21:26:46,729 - Epoch: [95][ 150/ 1236] Overall Loss 0.316553 Objective Loss 0.316553 LR 0.001000 Time 0.025735 +2023-10-05 21:26:46,935 - Epoch: [95][ 160/ 1236] Overall Loss 0.315977 Objective Loss 0.315977 LR 0.001000 Time 0.025408 +2023-10-05 21:26:47,139 - Epoch: [95][ 170/ 1236] Overall Loss 0.315984 Objective Loss 0.315984 LR 0.001000 Time 0.025113 +2023-10-05 21:26:47,345 - Epoch: [95][ 180/ 1236] Overall Loss 0.314659 Objective Loss 0.314659 LR 0.001000 Time 0.024859 +2023-10-05 21:26:47,549 - Epoch: [95][ 190/ 1236] Overall Loss 0.314995 Objective Loss 0.314995 LR 0.001000 Time 0.024625 +2023-10-05 21:26:47,755 - Epoch: [95][ 200/ 1236] Overall Loss 0.314203 Objective Loss 0.314203 LR 0.001000 Time 0.024419 +2023-10-05 21:26:47,959 - Epoch: [95][ 210/ 1236] Overall Loss 0.314931 Objective Loss 0.314931 LR 0.001000 Time 0.024228 +2023-10-05 21:26:48,167 - Epoch: [95][ 220/ 1236] Overall Loss 0.315415 Objective Loss 0.315415 LR 0.001000 Time 0.024067 +2023-10-05 21:26:48,373 - Epoch: [95][ 230/ 1236] Overall Loss 0.314396 Objective Loss 0.314396 LR 0.001000 Time 0.023917 +2023-10-05 21:26:48,582 - Epoch: [95][ 240/ 1236] Overall Loss 0.315553 Objective Loss 0.315553 LR 0.001000 Time 0.023788 +2023-10-05 21:26:48,788 - Epoch: [95][ 250/ 1236] Overall Loss 0.316277 Objective Loss 0.316277 LR 0.001000 Time 0.023661 +2023-10-05 21:26:48,997 - Epoch: [95][ 260/ 1236] Overall Loss 0.316717 Objective Loss 0.316717 LR 0.001000 Time 0.023551 +2023-10-05 21:26:49,203 - Epoch: [95][ 270/ 1236] Overall Loss 0.317530 Objective Loss 0.317530 LR 0.001000 Time 0.023443 +2023-10-05 21:26:49,411 - Epoch: [95][ 280/ 1236] Overall Loss 0.318397 Objective Loss 0.318397 LR 0.001000 Time 0.023345 +2023-10-05 21:26:49,617 - Epoch: [95][ 290/ 1236] Overall Loss 0.319299 Objective Loss 0.319299 LR 0.001000 Time 0.023250 +2023-10-05 21:26:49,826 - Epoch: [95][ 300/ 1236] Overall Loss 0.319491 Objective Loss 0.319491 LR 0.001000 Time 0.023170 +2023-10-05 21:26:50,032 - Epoch: [95][ 310/ 1236] Overall Loss 0.321303 Objective Loss 0.321303 LR 0.001000 Time 0.023087 +2023-10-05 21:26:50,241 - Epoch: [95][ 320/ 1236] Overall Loss 0.321633 Objective Loss 0.321633 LR 0.001000 Time 0.023017 +2023-10-05 21:26:50,447 - Epoch: [95][ 330/ 1236] Overall Loss 0.321889 Objective Loss 0.321889 LR 0.001000 Time 0.022943 +2023-10-05 21:26:50,663 - Epoch: [95][ 340/ 1236] Overall Loss 0.321825 Objective Loss 0.321825 LR 0.001000 Time 0.022900 +2023-10-05 21:26:50,870 - Epoch: [95][ 350/ 1236] Overall Loss 0.322871 Objective Loss 0.322871 LR 0.001000 Time 0.022836 +2023-10-05 21:26:51,078 - Epoch: [95][ 360/ 1236] Overall Loss 0.323627 Objective Loss 0.323627 LR 0.001000 Time 0.022777 +2023-10-05 21:26:51,284 - Epoch: [95][ 370/ 1236] Overall Loss 0.322478 Objective Loss 0.322478 LR 0.001000 Time 0.022720 +2023-10-05 21:26:51,493 - Epoch: [95][ 380/ 1236] Overall Loss 0.322075 Objective Loss 0.322075 LR 0.001000 Time 0.022669 +2023-10-05 21:26:51,700 - Epoch: [95][ 390/ 1236] Overall Loss 0.321830 Objective Loss 0.321830 LR 0.001000 Time 0.022619 +2023-10-05 21:26:51,909 - Epoch: [95][ 400/ 1236] Overall Loss 0.322385 Objective Loss 0.322385 LR 0.001000 Time 0.022574 +2023-10-05 21:26:52,115 - Epoch: [95][ 410/ 1236] Overall Loss 0.322219 Objective Loss 0.322219 LR 0.001000 Time 0.022525 +2023-10-05 21:26:52,321 - Epoch: [95][ 420/ 1236] Overall Loss 0.322302 Objective Loss 0.322302 LR 0.001000 Time 0.022479 +2023-10-05 21:26:52,526 - Epoch: [95][ 430/ 1236] Overall Loss 0.322257 Objective Loss 0.322257 LR 0.001000 Time 0.022432 +2023-10-05 21:26:52,733 - Epoch: [95][ 440/ 1236] Overall Loss 0.322515 Objective Loss 0.322515 LR 0.001000 Time 0.022390 +2023-10-05 21:26:52,937 - Epoch: [95][ 450/ 1236] Overall Loss 0.322358 Objective Loss 0.322358 LR 0.001000 Time 0.022347 +2023-10-05 21:26:53,144 - Epoch: [95][ 460/ 1236] Overall Loss 0.323283 Objective Loss 0.323283 LR 0.001000 Time 0.022309 +2023-10-05 21:26:53,349 - Epoch: [95][ 470/ 1236] Overall Loss 0.323315 Objective Loss 0.323315 LR 0.001000 Time 0.022269 +2023-10-05 21:26:53,555 - Epoch: [95][ 480/ 1236] Overall Loss 0.324109 Objective Loss 0.324109 LR 0.001000 Time 0.022234 +2023-10-05 21:26:53,760 - Epoch: [95][ 490/ 1236] Overall Loss 0.323460 Objective Loss 0.323460 LR 0.001000 Time 0.022198 +2023-10-05 21:26:53,967 - Epoch: [95][ 500/ 1236] Overall Loss 0.322674 Objective Loss 0.322674 LR 0.001000 Time 0.022166 +2023-10-05 21:26:54,172 - Epoch: [95][ 510/ 1236] Overall Loss 0.322507 Objective Loss 0.322507 LR 0.001000 Time 0.022133 +2023-10-05 21:26:54,378 - Epoch: [95][ 520/ 1236] Overall Loss 0.322811 Objective Loss 0.322811 LR 0.001000 Time 0.022104 +2023-10-05 21:26:54,584 - Epoch: [95][ 530/ 1236] Overall Loss 0.322667 Objective Loss 0.322667 LR 0.001000 Time 0.022074 +2023-10-05 21:26:54,791 - Epoch: [95][ 540/ 1236] Overall Loss 0.323285 Objective Loss 0.323285 LR 0.001000 Time 0.022047 +2023-10-05 21:26:54,996 - Epoch: [95][ 550/ 1236] Overall Loss 0.323871 Objective Loss 0.323871 LR 0.001000 Time 0.022019 +2023-10-05 21:26:55,203 - Epoch: [95][ 560/ 1236] Overall Loss 0.324247 Objective Loss 0.324247 LR 0.001000 Time 0.021994 +2023-10-05 21:26:55,408 - Epoch: [95][ 570/ 1236] Overall Loss 0.324342 Objective Loss 0.324342 LR 0.001000 Time 0.021968 +2023-10-05 21:26:55,615 - Epoch: [95][ 580/ 1236] Overall Loss 0.324703 Objective Loss 0.324703 LR 0.001000 Time 0.021945 +2023-10-05 21:26:55,822 - Epoch: [95][ 590/ 1236] Overall Loss 0.325101 Objective Loss 0.325101 LR 0.001000 Time 0.021924 +2023-10-05 21:26:56,027 - Epoch: [95][ 600/ 1236] Overall Loss 0.325286 Objective Loss 0.325286 LR 0.001000 Time 0.021900 +2023-10-05 21:26:56,231 - Epoch: [95][ 610/ 1236] Overall Loss 0.324846 Objective Loss 0.324846 LR 0.001000 Time 0.021874 +2023-10-05 21:26:56,437 - Epoch: [95][ 620/ 1236] Overall Loss 0.325322 Objective Loss 0.325322 LR 0.001000 Time 0.021853 +2023-10-05 21:26:56,641 - Epoch: [95][ 630/ 1236] Overall Loss 0.325531 Objective Loss 0.325531 LR 0.001000 Time 0.021829 +2023-10-05 21:26:56,847 - Epoch: [95][ 640/ 1236] Overall Loss 0.325381 Objective Loss 0.325381 LR 0.001000 Time 0.021810 +2023-10-05 21:26:57,051 - Epoch: [95][ 650/ 1236] Overall Loss 0.325366 Objective Loss 0.325366 LR 0.001000 Time 0.021787 +2023-10-05 21:26:57,258 - Epoch: [95][ 660/ 1236] Overall Loss 0.325184 Objective Loss 0.325184 LR 0.001000 Time 0.021769 +2023-10-05 21:26:57,461 - Epoch: [95][ 670/ 1236] Overall Loss 0.325003 Objective Loss 0.325003 LR 0.001000 Time 0.021748 +2023-10-05 21:26:57,668 - Epoch: [95][ 680/ 1236] Overall Loss 0.325127 Objective Loss 0.325127 LR 0.001000 Time 0.021731 +2023-10-05 21:26:57,871 - Epoch: [95][ 690/ 1236] Overall Loss 0.325246 Objective Loss 0.325246 LR 0.001000 Time 0.021711 +2023-10-05 21:26:58,078 - Epoch: [95][ 700/ 1236] Overall Loss 0.325387 Objective Loss 0.325387 LR 0.001000 Time 0.021695 +2023-10-05 21:26:58,281 - Epoch: [95][ 710/ 1236] Overall Loss 0.325823 Objective Loss 0.325823 LR 0.001000 Time 0.021675 +2023-10-05 21:26:58,488 - Epoch: [95][ 720/ 1236] Overall Loss 0.325758 Objective Loss 0.325758 LR 0.001000 Time 0.021660 +2023-10-05 21:26:58,691 - Epoch: [95][ 730/ 1236] Overall Loss 0.325524 Objective Loss 0.325524 LR 0.001000 Time 0.021642 +2023-10-05 21:26:58,897 - Epoch: [95][ 740/ 1236] Overall Loss 0.326083 Objective Loss 0.326083 LR 0.001000 Time 0.021628 +2023-10-05 21:26:59,101 - Epoch: [95][ 750/ 1236] Overall Loss 0.326347 Objective Loss 0.326347 LR 0.001000 Time 0.021610 +2023-10-05 21:26:59,307 - Epoch: [95][ 760/ 1236] Overall Loss 0.326345 Objective Loss 0.326345 LR 0.001000 Time 0.021597 +2023-10-05 21:26:59,511 - Epoch: [95][ 770/ 1236] Overall Loss 0.326478 Objective Loss 0.326478 LR 0.001000 Time 0.021580 +2023-10-05 21:26:59,717 - Epoch: [95][ 780/ 1236] Overall Loss 0.326588 Objective Loss 0.326588 LR 0.001000 Time 0.021568 +2023-10-05 21:26:59,921 - Epoch: [95][ 790/ 1236] Overall Loss 0.326621 Objective Loss 0.326621 LR 0.001000 Time 0.021552 +2023-10-05 21:27:00,127 - Epoch: [95][ 800/ 1236] Overall Loss 0.326539 Objective Loss 0.326539 LR 0.001000 Time 0.021540 +2023-10-05 21:27:00,331 - Epoch: [95][ 810/ 1236] Overall Loss 0.326431 Objective Loss 0.326431 LR 0.001000 Time 0.021525 +2023-10-05 21:27:00,537 - Epoch: [95][ 820/ 1236] Overall Loss 0.326424 Objective Loss 0.326424 LR 0.001000 Time 0.021514 +2023-10-05 21:27:00,741 - Epoch: [95][ 830/ 1236] Overall Loss 0.326301 Objective Loss 0.326301 LR 0.001000 Time 0.021499 +2023-10-05 21:27:00,947 - Epoch: [95][ 840/ 1236] Overall Loss 0.326454 Objective Loss 0.326454 LR 0.001000 Time 0.021489 +2023-10-05 21:27:01,151 - Epoch: [95][ 850/ 1236] Overall Loss 0.326786 Objective Loss 0.326786 LR 0.001000 Time 0.021475 +2023-10-05 21:27:01,357 - Epoch: [95][ 860/ 1236] Overall Loss 0.327047 Objective Loss 0.327047 LR 0.001000 Time 0.021465 +2023-10-05 21:27:01,561 - Epoch: [95][ 870/ 1236] Overall Loss 0.326830 Objective Loss 0.326830 LR 0.001000 Time 0.021452 +2023-10-05 21:27:01,767 - Epoch: [95][ 880/ 1236] Overall Loss 0.326937 Objective Loss 0.326937 LR 0.001000 Time 0.021442 +2023-10-05 21:27:01,971 - Epoch: [95][ 890/ 1236] Overall Loss 0.326849 Objective Loss 0.326849 LR 0.001000 Time 0.021429 +2023-10-05 21:27:02,177 - Epoch: [95][ 900/ 1236] Overall Loss 0.326725 Objective Loss 0.326725 LR 0.001000 Time 0.021420 +2023-10-05 21:27:02,381 - Epoch: [95][ 910/ 1236] Overall Loss 0.326897 Objective Loss 0.326897 LR 0.001000 Time 0.021408 +2023-10-05 21:27:02,587 - Epoch: [95][ 920/ 1236] Overall Loss 0.326850 Objective Loss 0.326850 LR 0.001000 Time 0.021399 +2023-10-05 21:27:02,791 - Epoch: [95][ 930/ 1236] Overall Loss 0.326445 Objective Loss 0.326445 LR 0.001000 Time 0.021388 +2023-10-05 21:27:02,997 - Epoch: [95][ 940/ 1236] Overall Loss 0.326715 Objective Loss 0.326715 LR 0.001000 Time 0.021379 +2023-10-05 21:27:03,201 - Epoch: [95][ 950/ 1236] Overall Loss 0.326683 Objective Loss 0.326683 LR 0.001000 Time 0.021368 +2023-10-05 21:27:03,407 - Epoch: [95][ 960/ 1236] Overall Loss 0.326895 Objective Loss 0.326895 LR 0.001000 Time 0.021360 +2023-10-05 21:27:03,611 - Epoch: [95][ 970/ 1236] Overall Loss 0.326904 Objective Loss 0.326904 LR 0.001000 Time 0.021350 +2023-10-05 21:27:03,817 - Epoch: [95][ 980/ 1236] Overall Loss 0.326915 Objective Loss 0.326915 LR 0.001000 Time 0.021342 +2023-10-05 21:27:04,021 - Epoch: [95][ 990/ 1236] Overall Loss 0.327115 Objective Loss 0.327115 LR 0.001000 Time 0.021332 +2023-10-05 21:27:04,227 - Epoch: [95][ 1000/ 1236] Overall Loss 0.327246 Objective Loss 0.327246 LR 0.001000 Time 0.021324 +2023-10-05 21:27:04,431 - Epoch: [95][ 1010/ 1236] Overall Loss 0.327295 Objective Loss 0.327295 LR 0.001000 Time 0.021314 +2023-10-05 21:27:04,637 - Epoch: [95][ 1020/ 1236] Overall Loss 0.327068 Objective Loss 0.327068 LR 0.001000 Time 0.021307 +2023-10-05 21:27:04,841 - Epoch: [95][ 1030/ 1236] Overall Loss 0.326877 Objective Loss 0.326877 LR 0.001000 Time 0.021298 +2023-10-05 21:27:05,047 - Epoch: [95][ 1040/ 1236] Overall Loss 0.326407 Objective Loss 0.326407 LR 0.001000 Time 0.021291 +2023-10-05 21:27:05,251 - Epoch: [95][ 1050/ 1236] Overall Loss 0.326282 Objective Loss 0.326282 LR 0.001000 Time 0.021282 +2023-10-05 21:27:05,457 - Epoch: [95][ 1060/ 1236] Overall Loss 0.326116 Objective Loss 0.326116 LR 0.001000 Time 0.021275 +2023-10-05 21:27:05,661 - Epoch: [95][ 1070/ 1236] Overall Loss 0.326055 Objective Loss 0.326055 LR 0.001000 Time 0.021266 +2023-10-05 21:27:05,867 - Epoch: [95][ 1080/ 1236] Overall Loss 0.326095 Objective Loss 0.326095 LR 0.001000 Time 0.021260 +2023-10-05 21:27:06,070 - Epoch: [95][ 1090/ 1236] Overall Loss 0.326187 Objective Loss 0.326187 LR 0.001000 Time 0.021251 +2023-10-05 21:27:06,277 - Epoch: [95][ 1100/ 1236] Overall Loss 0.326060 Objective Loss 0.326060 LR 0.001000 Time 0.021245 +2023-10-05 21:27:06,480 - Epoch: [95][ 1110/ 1236] Overall Loss 0.325924 Objective Loss 0.325924 LR 0.001000 Time 0.021237 +2023-10-05 21:27:06,687 - Epoch: [95][ 1120/ 1236] Overall Loss 0.325882 Objective Loss 0.325882 LR 0.001000 Time 0.021231 +2023-10-05 21:27:06,890 - Epoch: [95][ 1130/ 1236] Overall Loss 0.325948 Objective Loss 0.325948 LR 0.001000 Time 0.021223 +2023-10-05 21:27:07,097 - Epoch: [95][ 1140/ 1236] Overall Loss 0.325801 Objective Loss 0.325801 LR 0.001000 Time 0.021218 +2023-10-05 21:27:07,300 - Epoch: [95][ 1150/ 1236] Overall Loss 0.325822 Objective Loss 0.325822 LR 0.001000 Time 0.021210 +2023-10-05 21:27:07,506 - Epoch: [95][ 1160/ 1236] Overall Loss 0.325837 Objective Loss 0.325837 LR 0.001000 Time 0.021205 +2023-10-05 21:27:07,710 - Epoch: [95][ 1170/ 1236] Overall Loss 0.325836 Objective Loss 0.325836 LR 0.001000 Time 0.021197 +2023-10-05 21:27:07,916 - Epoch: [95][ 1180/ 1236] Overall Loss 0.325697 Objective Loss 0.325697 LR 0.001000 Time 0.021192 +2023-10-05 21:27:08,120 - Epoch: [95][ 1190/ 1236] Overall Loss 0.325671 Objective Loss 0.325671 LR 0.001000 Time 0.021185 +2023-10-05 21:27:08,326 - Epoch: [95][ 1200/ 1236] Overall Loss 0.325815 Objective Loss 0.325815 LR 0.001000 Time 0.021180 +2023-10-05 21:27:08,530 - Epoch: [95][ 1210/ 1236] Overall Loss 0.325843 Objective Loss 0.325843 LR 0.001000 Time 0.021173 +2023-10-05 21:27:08,736 - Epoch: [95][ 1220/ 1236] Overall Loss 0.326296 Objective Loss 0.326296 LR 0.001000 Time 0.021168 +2023-10-05 21:27:08,996 - Epoch: [95][ 1230/ 1236] Overall Loss 0.326429 Objective Loss 0.326429 LR 0.001000 Time 0.021207 +2023-10-05 21:27:09,115 - Epoch: [95][ 1236/ 1236] Overall Loss 0.326355 Objective Loss 0.326355 Top1 84.521385 Top5 98.778004 LR 0.001000 Time 0.021200 +2023-10-05 21:27:09,239 - --- validate (epoch=95)----------- +2023-10-05 21:27:09,240 - 29943 samples (256 per mini-batch) +2023-10-05 21:27:09,698 - Epoch: [95][ 10/ 117] Loss 0.344137 Top1 82.343750 Top5 97.851562 +2023-10-05 21:27:09,848 - Epoch: [95][ 20/ 117] Loss 0.367565 Top1 82.167969 Top5 97.539062 +2023-10-05 21:27:10,000 - Epoch: [95][ 30/ 117] Loss 0.357228 Top1 82.213542 Top5 97.500000 +2023-10-05 21:27:10,151 - Epoch: [95][ 40/ 117] Loss 0.351696 Top1 82.343750 Top5 97.548828 +2023-10-05 21:27:10,303 - Epoch: [95][ 50/ 117] Loss 0.358786 Top1 82.085938 Top5 97.601562 +2023-10-05 21:27:10,461 - Epoch: [95][ 60/ 117] Loss 0.358059 Top1 82.167969 Top5 97.558594 +2023-10-05 21:27:10,620 - Epoch: [95][ 70/ 117] Loss 0.355230 Top1 82.237723 Top5 97.566964 +2023-10-05 21:27:10,776 - Epoch: [95][ 80/ 117] Loss 0.357094 Top1 82.299805 Top5 97.539062 +2023-10-05 21:27:10,935 - Epoch: [95][ 90/ 117] Loss 0.358430 Top1 82.356771 Top5 97.534722 +2023-10-05 21:27:11,093 - Epoch: [95][ 100/ 117] Loss 0.360318 Top1 82.367188 Top5 97.539062 +2023-10-05 21:27:11,252 - Epoch: [95][ 110/ 117] Loss 0.361918 Top1 82.379261 Top5 97.514205 +2023-10-05 21:27:11,339 - Epoch: [95][ 117/ 117] Loss 0.362019 Top1 82.389874 Top5 97.535317 +2023-10-05 21:27:11,445 - ==> Top1: 82.390 Top5: 97.535 Loss: 0.362 + +2023-10-05 21:27:11,446 - ==> Confusion: +[[ 928 1 5 2 9 2 0 0 6 66 0 1 0 5 6 5 6 0 0 0 8] + [ 1 1031 4 2 5 21 0 23 4 0 5 1 0 1 0 5 10 0 9 1 8] + [ 6 0 944 15 0 1 25 10 0 5 8 0 8 4 1 4 3 2 6 4 10] + [ 4 1 17 965 1 5 0 1 2 1 9 0 2 4 34 5 1 3 13 3 18] + [ 28 5 1 0 957 2 0 0 0 7 2 2 3 2 12 4 12 2 0 1 10] + [ 5 38 0 3 3 950 0 34 1 2 5 9 1 23 8 1 1 1 1 8 22] + [ 1 3 43 0 0 2 1092 8 0 0 11 3 0 1 0 8 1 4 2 5 7] + [ 2 17 21 0 1 38 4 1052 0 2 4 12 6 0 0 1 0 0 38 10 10] + [ 26 0 0 1 0 7 0 0 952 37 14 0 3 14 19 3 4 1 5 0 3] + [ 111 0 4 1 7 2 0 0 26 909 0 0 0 34 9 5 1 0 0 4 6] + [ 2 1 12 7 2 2 5 6 17 2 946 3 0 20 3 1 2 1 5 1 15] + [ 1 2 1 0 0 11 0 8 0 0 0 947 24 10 0 5 1 14 0 5 6] + [ 1 1 6 10 2 2 1 5 1 0 1 40 942 6 4 9 3 18 1 3 12] + [ 2 0 0 0 7 11 0 1 15 11 7 3 1 1046 3 0 1 2 0 0 9] + [ 21 2 3 8 5 0 0 0 18 5 1 1 3 1 1006 1 2 1 12 0 11] + [ 0 2 3 1 5 0 2 0 0 0 1 7 9 3 0 1068 15 10 2 1 5] + [ 2 19 1 0 6 7 1 2 0 0 0 3 0 1 1 10 1093 0 1 6 8] + [ 1 0 1 3 0 0 1 0 0 0 0 6 26 0 2 7 0 980 1 3 7] + [ 1 6 10 16 3 2 2 41 2 0 3 1 7 0 11 0 3 0 951 0 9] + [ 0 4 3 1 5 8 6 14 0 0 3 19 5 2 0 8 15 1 1 1038 19] + [ 169 195 184 64 95 133 50 125 86 100 191 127 362 363 161 91 163 72 119 182 4873]] + +2023-10-05 21:27:11,448 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:27:11,448 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:27:11,454 - + +2023-10-05 21:27:11,454 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:27:12,580 - Epoch: [96][ 10/ 1236] Overall Loss 0.298937 Objective Loss 0.298937 LR 0.001000 Time 0.112547 +2023-10-05 21:27:12,784 - Epoch: [96][ 20/ 1236] Overall Loss 0.313181 Objective Loss 0.313181 LR 0.001000 Time 0.066476 +2023-10-05 21:27:12,985 - Epoch: [96][ 30/ 1236] Overall Loss 0.309857 Objective Loss 0.309857 LR 0.001000 Time 0.050997 +2023-10-05 21:27:13,187 - Epoch: [96][ 40/ 1236] Overall Loss 0.318263 Objective Loss 0.318263 LR 0.001000 Time 0.043287 +2023-10-05 21:27:13,387 - Epoch: [96][ 50/ 1236] Overall Loss 0.322523 Objective Loss 0.322523 LR 0.001000 Time 0.038621 +2023-10-05 21:27:13,588 - Epoch: [96][ 60/ 1236] Overall Loss 0.320727 Objective Loss 0.320727 LR 0.001000 Time 0.035538 +2023-10-05 21:27:13,788 - Epoch: [96][ 70/ 1236] Overall Loss 0.325816 Objective Loss 0.325816 LR 0.001000 Time 0.033315 +2023-10-05 21:27:13,991 - Epoch: [96][ 80/ 1236] Overall Loss 0.324959 Objective Loss 0.324959 LR 0.001000 Time 0.031677 +2023-10-05 21:27:14,190 - Epoch: [96][ 90/ 1236] Overall Loss 0.327773 Objective Loss 0.327773 LR 0.001000 Time 0.030370 +2023-10-05 21:27:14,392 - Epoch: [96][ 100/ 1236] Overall Loss 0.327473 Objective Loss 0.327473 LR 0.001000 Time 0.029350 +2023-10-05 21:27:14,592 - Epoch: [96][ 110/ 1236] Overall Loss 0.326705 Objective Loss 0.326705 LR 0.001000 Time 0.028497 +2023-10-05 21:27:14,794 - Epoch: [96][ 120/ 1236] Overall Loss 0.327726 Objective Loss 0.327726 LR 0.001000 Time 0.027798 +2023-10-05 21:27:14,993 - Epoch: [96][ 130/ 1236] Overall Loss 0.326705 Objective Loss 0.326705 LR 0.001000 Time 0.027192 +2023-10-05 21:27:15,195 - Epoch: [96][ 140/ 1236] Overall Loss 0.323535 Objective Loss 0.323535 LR 0.001000 Time 0.026688 +2023-10-05 21:27:15,395 - Epoch: [96][ 150/ 1236] Overall Loss 0.324499 Objective Loss 0.324499 LR 0.001000 Time 0.026238 +2023-10-05 21:27:15,597 - Epoch: [96][ 160/ 1236] Overall Loss 0.325204 Objective Loss 0.325204 LR 0.001000 Time 0.025856 +2023-10-05 21:27:15,796 - Epoch: [96][ 170/ 1236] Overall Loss 0.325491 Objective Loss 0.325491 LR 0.001000 Time 0.025509 +2023-10-05 21:27:15,998 - Epoch: [96][ 180/ 1236] Overall Loss 0.324763 Objective Loss 0.324763 LR 0.001000 Time 0.025211 +2023-10-05 21:27:16,198 - Epoch: [96][ 190/ 1236] Overall Loss 0.324495 Objective Loss 0.324495 LR 0.001000 Time 0.024933 +2023-10-05 21:27:16,400 - Epoch: [96][ 200/ 1236] Overall Loss 0.322685 Objective Loss 0.322685 LR 0.001000 Time 0.024695 +2023-10-05 21:27:16,600 - Epoch: [96][ 210/ 1236] Overall Loss 0.323371 Objective Loss 0.323371 LR 0.001000 Time 0.024470 +2023-10-05 21:27:16,802 - Epoch: [96][ 220/ 1236] Overall Loss 0.322640 Objective Loss 0.322640 LR 0.001000 Time 0.024274 +2023-10-05 21:27:17,002 - Epoch: [96][ 230/ 1236] Overall Loss 0.322879 Objective Loss 0.322879 LR 0.001000 Time 0.024087 +2023-10-05 21:27:17,202 - Epoch: [96][ 240/ 1236] Overall Loss 0.321838 Objective Loss 0.321838 LR 0.001000 Time 0.023915 +2023-10-05 21:27:17,402 - Epoch: [96][ 250/ 1236] Overall Loss 0.323035 Objective Loss 0.323035 LR 0.001000 Time 0.023758 +2023-10-05 21:27:17,602 - Epoch: [96][ 260/ 1236] Overall Loss 0.322566 Objective Loss 0.322566 LR 0.001000 Time 0.023613 +2023-10-05 21:27:17,804 - Epoch: [96][ 270/ 1236] Overall Loss 0.323906 Objective Loss 0.323906 LR 0.001000 Time 0.023483 +2023-10-05 21:27:18,004 - Epoch: [96][ 280/ 1236] Overall Loss 0.324548 Objective Loss 0.324548 LR 0.001000 Time 0.023357 +2023-10-05 21:27:18,206 - Epoch: [96][ 290/ 1236] Overall Loss 0.324530 Objective Loss 0.324530 LR 0.001000 Time 0.023246 +2023-10-05 21:27:18,406 - Epoch: [96][ 300/ 1236] Overall Loss 0.324496 Objective Loss 0.324496 LR 0.001000 Time 0.023138 +2023-10-05 21:27:18,608 - Epoch: [96][ 310/ 1236] Overall Loss 0.323848 Objective Loss 0.323848 LR 0.001000 Time 0.023042 +2023-10-05 21:27:18,808 - Epoch: [96][ 320/ 1236] Overall Loss 0.322876 Objective Loss 0.322876 LR 0.001000 Time 0.022947 +2023-10-05 21:27:19,009 - Epoch: [96][ 330/ 1236] Overall Loss 0.323133 Objective Loss 0.323133 LR 0.001000 Time 0.022859 +2023-10-05 21:27:19,208 - Epoch: [96][ 340/ 1236] Overall Loss 0.323300 Objective Loss 0.323300 LR 0.001000 Time 0.022772 +2023-10-05 21:27:19,409 - Epoch: [96][ 350/ 1236] Overall Loss 0.324020 Objective Loss 0.324020 LR 0.001000 Time 0.022695 +2023-10-05 21:27:19,609 - Epoch: [96][ 360/ 1236] Overall Loss 0.323584 Objective Loss 0.323584 LR 0.001000 Time 0.022617 +2023-10-05 21:27:19,810 - Epoch: [96][ 370/ 1236] Overall Loss 0.323915 Objective Loss 0.323915 LR 0.001000 Time 0.022549 +2023-10-05 21:27:20,012 - Epoch: [96][ 380/ 1236] Overall Loss 0.324985 Objective Loss 0.324985 LR 0.001000 Time 0.022485 +2023-10-05 21:27:20,214 - Epoch: [96][ 390/ 1236] Overall Loss 0.325248 Objective Loss 0.325248 LR 0.001000 Time 0.022426 +2023-10-05 21:27:20,414 - Epoch: [96][ 400/ 1236] Overall Loss 0.324991 Objective Loss 0.324991 LR 0.001000 Time 0.022365 +2023-10-05 21:27:20,616 - Epoch: [96][ 410/ 1236] Overall Loss 0.325177 Objective Loss 0.325177 LR 0.001000 Time 0.022312 +2023-10-05 21:27:20,816 - Epoch: [96][ 420/ 1236] Overall Loss 0.325610 Objective Loss 0.325610 LR 0.001000 Time 0.022255 +2023-10-05 21:27:21,018 - Epoch: [96][ 430/ 1236] Overall Loss 0.326344 Objective Loss 0.326344 LR 0.001000 Time 0.022207 +2023-10-05 21:27:21,218 - Epoch: [96][ 440/ 1236] Overall Loss 0.326678 Objective Loss 0.326678 LR 0.001000 Time 0.022156 +2023-10-05 21:27:21,420 - Epoch: [96][ 450/ 1236] Overall Loss 0.327094 Objective Loss 0.327094 LR 0.001000 Time 0.022112 +2023-10-05 21:27:21,620 - Epoch: [96][ 460/ 1236] Overall Loss 0.326719 Objective Loss 0.326719 LR 0.001000 Time 0.022066 +2023-10-05 21:27:21,822 - Epoch: [96][ 470/ 1236] Overall Loss 0.326074 Objective Loss 0.326074 LR 0.001000 Time 0.022026 +2023-10-05 21:27:22,022 - Epoch: [96][ 480/ 1236] Overall Loss 0.325633 Objective Loss 0.325633 LR 0.001000 Time 0.021982 +2023-10-05 21:27:22,224 - Epoch: [96][ 490/ 1236] Overall Loss 0.326301 Objective Loss 0.326301 LR 0.001000 Time 0.021945 +2023-10-05 21:27:22,424 - Epoch: [96][ 500/ 1236] Overall Loss 0.326677 Objective Loss 0.326677 LR 0.001000 Time 0.021905 +2023-10-05 21:27:22,626 - Epoch: [96][ 510/ 1236] Overall Loss 0.326767 Objective Loss 0.326767 LR 0.001000 Time 0.021871 +2023-10-05 21:27:22,826 - Epoch: [96][ 520/ 1236] Overall Loss 0.325843 Objective Loss 0.325843 LR 0.001000 Time 0.021835 +2023-10-05 21:27:23,029 - Epoch: [96][ 530/ 1236] Overall Loss 0.324859 Objective Loss 0.324859 LR 0.001000 Time 0.021804 +2023-10-05 21:27:23,229 - Epoch: [96][ 540/ 1236] Overall Loss 0.325355 Objective Loss 0.325355 LR 0.001000 Time 0.021770 +2023-10-05 21:27:23,431 - Epoch: [96][ 550/ 1236] Overall Loss 0.324890 Objective Loss 0.324890 LR 0.001000 Time 0.021742 +2023-10-05 21:27:23,631 - Epoch: [96][ 560/ 1236] Overall Loss 0.324665 Objective Loss 0.324665 LR 0.001000 Time 0.021710 +2023-10-05 21:27:23,833 - Epoch: [96][ 570/ 1236] Overall Loss 0.324700 Objective Loss 0.324700 LR 0.001000 Time 0.021683 +2023-10-05 21:27:24,033 - Epoch: [96][ 580/ 1236] Overall Loss 0.324082 Objective Loss 0.324082 LR 0.001000 Time 0.021653 +2023-10-05 21:27:24,235 - Epoch: [96][ 590/ 1236] Overall Loss 0.324137 Objective Loss 0.324137 LR 0.001000 Time 0.021628 +2023-10-05 21:27:24,436 - Epoch: [96][ 600/ 1236] Overall Loss 0.324372 Objective Loss 0.324372 LR 0.001000 Time 0.021602 +2023-10-05 21:27:24,638 - Epoch: [96][ 610/ 1236] Overall Loss 0.324823 Objective Loss 0.324823 LR 0.001000 Time 0.021579 +2023-10-05 21:27:24,838 - Epoch: [96][ 620/ 1236] Overall Loss 0.325166 Objective Loss 0.325166 LR 0.001000 Time 0.021553 +2023-10-05 21:27:25,040 - Epoch: [96][ 630/ 1236] Overall Loss 0.325793 Objective Loss 0.325793 LR 0.001000 Time 0.021531 +2023-10-05 21:27:25,240 - Epoch: [96][ 640/ 1236] Overall Loss 0.326249 Objective Loss 0.326249 LR 0.001000 Time 0.021506 +2023-10-05 21:27:25,442 - Epoch: [96][ 650/ 1236] Overall Loss 0.325800 Objective Loss 0.325800 LR 0.001000 Time 0.021486 +2023-10-05 21:27:25,643 - Epoch: [96][ 660/ 1236] Overall Loss 0.325707 Objective Loss 0.325707 LR 0.001000 Time 0.021464 +2023-10-05 21:27:25,845 - Epoch: [96][ 670/ 1236] Overall Loss 0.325420 Objective Loss 0.325420 LR 0.001000 Time 0.021445 +2023-10-05 21:27:26,045 - Epoch: [96][ 680/ 1236] Overall Loss 0.325272 Objective Loss 0.325272 LR 0.001000 Time 0.021423 +2023-10-05 21:27:26,247 - Epoch: [96][ 690/ 1236] Overall Loss 0.324609 Objective Loss 0.324609 LR 0.001000 Time 0.021405 +2023-10-05 21:27:26,447 - Epoch: [96][ 700/ 1236] Overall Loss 0.324654 Objective Loss 0.324654 LR 0.001000 Time 0.021384 +2023-10-05 21:27:26,649 - Epoch: [96][ 710/ 1236] Overall Loss 0.324989 Objective Loss 0.324989 LR 0.001000 Time 0.021367 +2023-10-05 21:27:26,849 - Epoch: [96][ 720/ 1236] Overall Loss 0.324969 Objective Loss 0.324969 LR 0.001000 Time 0.021348 +2023-10-05 21:27:27,062 - Epoch: [96][ 730/ 1236] Overall Loss 0.324538 Objective Loss 0.324538 LR 0.001000 Time 0.021346 +2023-10-05 21:27:27,269 - Epoch: [96][ 740/ 1236] Overall Loss 0.324273 Objective Loss 0.324273 LR 0.001000 Time 0.021336 +2023-10-05 21:27:27,481 - Epoch: [96][ 750/ 1236] Overall Loss 0.324032 Objective Loss 0.324032 LR 0.001000 Time 0.021335 +2023-10-05 21:27:27,685 - Epoch: [96][ 760/ 1236] Overall Loss 0.324191 Objective Loss 0.324191 LR 0.001000 Time 0.021321 +2023-10-05 21:27:27,887 - Epoch: [96][ 770/ 1236] Overall Loss 0.324231 Objective Loss 0.324231 LR 0.001000 Time 0.021307 +2023-10-05 21:27:28,087 - Epoch: [96][ 780/ 1236] Overall Loss 0.324266 Objective Loss 0.324266 LR 0.001000 Time 0.021290 +2023-10-05 21:27:28,289 - Epoch: [96][ 790/ 1236] Overall Loss 0.323885 Objective Loss 0.323885 LR 0.001000 Time 0.021276 +2023-10-05 21:27:28,489 - Epoch: [96][ 800/ 1236] Overall Loss 0.324005 Objective Loss 0.324005 LR 0.001000 Time 0.021259 +2023-10-05 21:27:28,691 - Epoch: [96][ 810/ 1236] Overall Loss 0.324102 Objective Loss 0.324102 LR 0.001000 Time 0.021246 +2023-10-05 21:27:28,891 - Epoch: [96][ 820/ 1236] Overall Loss 0.323898 Objective Loss 0.323898 LR 0.001000 Time 0.021230 +2023-10-05 21:27:29,093 - Epoch: [96][ 830/ 1236] Overall Loss 0.323850 Objective Loss 0.323850 LR 0.001000 Time 0.021217 +2023-10-05 21:27:29,293 - Epoch: [96][ 840/ 1236] Overall Loss 0.323911 Objective Loss 0.323911 LR 0.001000 Time 0.021202 +2023-10-05 21:27:29,495 - Epoch: [96][ 850/ 1236] Overall Loss 0.323684 Objective Loss 0.323684 LR 0.001000 Time 0.021190 +2023-10-05 21:27:29,695 - Epoch: [96][ 860/ 1236] Overall Loss 0.323818 Objective Loss 0.323818 LR 0.001000 Time 0.021176 +2023-10-05 21:27:29,897 - Epoch: [96][ 870/ 1236] Overall Loss 0.324182 Objective Loss 0.324182 LR 0.001000 Time 0.021165 +2023-10-05 21:27:30,097 - Epoch: [96][ 880/ 1236] Overall Loss 0.324338 Objective Loss 0.324338 LR 0.001000 Time 0.021151 +2023-10-05 21:27:30,300 - Epoch: [96][ 890/ 1236] Overall Loss 0.324348 Objective Loss 0.324348 LR 0.001000 Time 0.021140 +2023-10-05 21:27:30,499 - Epoch: [96][ 900/ 1236] Overall Loss 0.323702 Objective Loss 0.323702 LR 0.001000 Time 0.021127 +2023-10-05 21:27:30,701 - Epoch: [96][ 910/ 1236] Overall Loss 0.323673 Objective Loss 0.323673 LR 0.001000 Time 0.021116 +2023-10-05 21:27:30,901 - Epoch: [96][ 920/ 1236] Overall Loss 0.323684 Objective Loss 0.323684 LR 0.001000 Time 0.021104 +2023-10-05 21:27:31,104 - Epoch: [96][ 930/ 1236] Overall Loss 0.323897 Objective Loss 0.323897 LR 0.001000 Time 0.021094 +2023-10-05 21:27:31,304 - Epoch: [96][ 940/ 1236] Overall Loss 0.324253 Objective Loss 0.324253 LR 0.001000 Time 0.021082 +2023-10-05 21:27:31,506 - Epoch: [96][ 950/ 1236] Overall Loss 0.324003 Objective Loss 0.324003 LR 0.001000 Time 0.021073 +2023-10-05 21:27:31,706 - Epoch: [96][ 960/ 1236] Overall Loss 0.323828 Objective Loss 0.323828 LR 0.001000 Time 0.021061 +2023-10-05 21:27:31,908 - Epoch: [96][ 970/ 1236] Overall Loss 0.323617 Objective Loss 0.323617 LR 0.001000 Time 0.021052 +2023-10-05 21:27:32,108 - Epoch: [96][ 980/ 1236] Overall Loss 0.323729 Objective Loss 0.323729 LR 0.001000 Time 0.021041 +2023-10-05 21:27:32,311 - Epoch: [96][ 990/ 1236] Overall Loss 0.323708 Objective Loss 0.323708 LR 0.001000 Time 0.021033 +2023-10-05 21:27:32,510 - Epoch: [96][ 1000/ 1236] Overall Loss 0.323956 Objective Loss 0.323956 LR 0.001000 Time 0.021022 +2023-10-05 21:27:32,712 - Epoch: [96][ 1010/ 1236] Overall Loss 0.324099 Objective Loss 0.324099 LR 0.001000 Time 0.021013 +2023-10-05 21:27:32,912 - Epoch: [96][ 1020/ 1236] Overall Loss 0.324262 Objective Loss 0.324262 LR 0.001000 Time 0.021003 +2023-10-05 21:27:33,114 - Epoch: [96][ 1030/ 1236] Overall Loss 0.324347 Objective Loss 0.324347 LR 0.001000 Time 0.020995 +2023-10-05 21:27:33,314 - Epoch: [96][ 1040/ 1236] Overall Loss 0.324525 Objective Loss 0.324525 LR 0.001000 Time 0.020985 +2023-10-05 21:27:33,517 - Epoch: [96][ 1050/ 1236] Overall Loss 0.324295 Objective Loss 0.324295 LR 0.001000 Time 0.020977 +2023-10-05 21:27:33,717 - Epoch: [96][ 1060/ 1236] Overall Loss 0.324417 Objective Loss 0.324417 LR 0.001000 Time 0.020968 +2023-10-05 21:27:33,919 - Epoch: [96][ 1070/ 1236] Overall Loss 0.324402 Objective Loss 0.324402 LR 0.001000 Time 0.020960 +2023-10-05 21:27:34,118 - Epoch: [96][ 1080/ 1236] Overall Loss 0.324298 Objective Loss 0.324298 LR 0.001000 Time 0.020951 +2023-10-05 21:27:34,321 - Epoch: [96][ 1090/ 1236] Overall Loss 0.324219 Objective Loss 0.324219 LR 0.001000 Time 0.020944 +2023-10-05 21:27:34,520 - Epoch: [96][ 1100/ 1236] Overall Loss 0.324040 Objective Loss 0.324040 LR 0.001000 Time 0.020935 +2023-10-05 21:27:34,722 - Epoch: [96][ 1110/ 1236] Overall Loss 0.323984 Objective Loss 0.323984 LR 0.001000 Time 0.020928 +2023-10-05 21:27:34,922 - Epoch: [96][ 1120/ 1236] Overall Loss 0.323686 Objective Loss 0.323686 LR 0.001000 Time 0.020919 +2023-10-05 21:27:35,124 - Epoch: [96][ 1130/ 1236] Overall Loss 0.323500 Objective Loss 0.323500 LR 0.001000 Time 0.020912 +2023-10-05 21:27:35,324 - Epoch: [96][ 1140/ 1236] Overall Loss 0.323554 Objective Loss 0.323554 LR 0.001000 Time 0.020904 +2023-10-05 21:27:35,527 - Epoch: [96][ 1150/ 1236] Overall Loss 0.323272 Objective Loss 0.323272 LR 0.001000 Time 0.020898 +2023-10-05 21:27:35,727 - Epoch: [96][ 1160/ 1236] Overall Loss 0.323344 Objective Loss 0.323344 LR 0.001000 Time 0.020890 +2023-10-05 21:27:35,930 - Epoch: [96][ 1170/ 1236] Overall Loss 0.323376 Objective Loss 0.323376 LR 0.001000 Time 0.020885 +2023-10-05 21:27:36,130 - Epoch: [96][ 1180/ 1236] Overall Loss 0.323322 Objective Loss 0.323322 LR 0.001000 Time 0.020877 +2023-10-05 21:27:36,333 - Epoch: [96][ 1190/ 1236] Overall Loss 0.323373 Objective Loss 0.323373 LR 0.001000 Time 0.020872 +2023-10-05 21:27:36,533 - Epoch: [96][ 1200/ 1236] Overall Loss 0.323356 Objective Loss 0.323356 LR 0.001000 Time 0.020865 +2023-10-05 21:27:36,736 - Epoch: [96][ 1210/ 1236] Overall Loss 0.323239 Objective Loss 0.323239 LR 0.001000 Time 0.020859 +2023-10-05 21:27:36,936 - Epoch: [96][ 1220/ 1236] Overall Loss 0.323207 Objective Loss 0.323207 LR 0.001000 Time 0.020852 +2023-10-05 21:27:37,187 - Epoch: [96][ 1230/ 1236] Overall Loss 0.323537 Objective Loss 0.323537 LR 0.001000 Time 0.020887 +2023-10-05 21:27:37,305 - Epoch: [96][ 1236/ 1236] Overall Loss 0.323448 Objective Loss 0.323448 Top1 82.077393 Top5 96.945010 LR 0.001000 Time 0.020880 +2023-10-05 21:27:37,436 - --- validate (epoch=96)----------- +2023-10-05 21:27:37,436 - 29943 samples (256 per mini-batch) +2023-10-05 21:27:37,891 - Epoch: [96][ 10/ 117] Loss 0.337671 Top1 82.343750 Top5 97.460938 +2023-10-05 21:27:38,040 - Epoch: [96][ 20/ 117] Loss 0.354258 Top1 81.855469 Top5 97.421875 +2023-10-05 21:27:38,189 - Epoch: [96][ 30/ 117] Loss 0.360939 Top1 81.250000 Top5 97.513021 +2023-10-05 21:27:38,336 - Epoch: [96][ 40/ 117] Loss 0.364481 Top1 81.279297 Top5 97.548828 +2023-10-05 21:27:38,485 - Epoch: [96][ 50/ 117] Loss 0.364807 Top1 81.281250 Top5 97.585938 +2023-10-05 21:27:38,633 - Epoch: [96][ 60/ 117] Loss 0.369850 Top1 81.074219 Top5 97.454427 +2023-10-05 21:27:38,781 - Epoch: [96][ 70/ 117] Loss 0.369703 Top1 80.959821 Top5 97.444196 +2023-10-05 21:27:38,929 - Epoch: [96][ 80/ 117] Loss 0.369126 Top1 80.957031 Top5 97.460938 +2023-10-05 21:27:39,077 - Epoch: [96][ 90/ 117] Loss 0.369325 Top1 80.972222 Top5 97.404514 +2023-10-05 21:27:39,226 - Epoch: [96][ 100/ 117] Loss 0.370488 Top1 81.066406 Top5 97.359375 +2023-10-05 21:27:39,379 - Epoch: [96][ 110/ 117] Loss 0.370793 Top1 81.193182 Top5 97.382812 +2023-10-05 21:27:39,464 - Epoch: [96][ 117/ 117] Loss 0.369489 Top1 81.204288 Top5 97.395051 +2023-10-05 21:27:39,615 - ==> Top1: 81.204 Top5: 97.395 Loss: 0.369 + +2023-10-05 21:27:39,615 - ==> Confusion: +[[ 896 4 3 3 9 4 0 3 10 77 2 2 2 4 5 6 2 4 1 1 12] + [ 4 1003 1 1 7 29 4 28 10 0 3 3 0 0 0 3 6 0 15 3 11] + [ 7 2 919 17 4 2 32 15 0 1 9 3 6 1 0 7 0 1 8 5 17] + [ 1 1 8 952 0 4 2 2 4 1 11 1 9 5 31 4 1 9 26 2 15] + [ 27 4 1 0 970 7 0 2 0 8 0 1 0 2 5 7 10 2 0 2 2] + [ 3 30 0 1 3 969 0 27 7 1 3 11 6 18 6 1 3 1 5 6 15] + [ 0 6 33 0 0 1 1099 12 0 0 3 1 1 1 0 13 0 2 4 10 5] + [ 3 15 19 0 1 31 2 1046 5 4 3 11 4 2 0 0 0 1 60 4 7] + [ 18 3 0 0 0 2 0 0 971 44 8 1 5 10 15 2 0 1 4 4 1] + [ 87 0 2 0 5 6 0 1 30 939 0 0 0 24 7 9 0 1 1 1 6] + [ 5 6 11 5 4 2 6 5 31 3 937 4 0 12 6 2 1 1 6 0 6] + [ 0 0 1 0 1 11 0 6 2 0 0 929 44 5 0 3 0 16 0 9 8] + [ 3 3 4 1 1 4 2 4 1 0 0 46 957 2 2 4 2 14 4 4 10] + [ 2 1 0 0 5 19 1 0 22 12 9 5 3 1019 5 2 2 1 0 3 8] + [ 19 3 2 16 10 0 0 0 23 3 3 0 3 1 990 0 1 1 14 0 12] + [ 3 2 3 0 4 1 2 0 0 0 0 8 7 1 0 1072 12 12 2 4 1] + [ 2 8 2 1 6 7 0 0 5 0 1 6 1 0 5 14 1084 0 1 7 11] + [ 0 1 0 1 0 0 3 0 3 1 0 4 37 0 0 9 1 970 1 0 7] + [ 2 5 3 20 0 0 0 28 3 1 4 0 6 0 14 0 1 1 968 4 8] + [ 0 3 3 0 2 5 9 14 1 0 3 22 4 3 0 8 10 1 1 1054 9] + [ 132 198 131 82 125 205 37 128 141 117 179 150 368 349 213 68 156 71 221 263 4571]] + +2023-10-05 21:27:39,617 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:27:39,617 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:27:39,623 - + +2023-10-05 21:27:39,623 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:27:40,619 - Epoch: [97][ 10/ 1236] Overall Loss 0.302903 Objective Loss 0.302903 LR 0.001000 Time 0.099550 +2023-10-05 21:27:40,820 - Epoch: [97][ 20/ 1236] Overall Loss 0.323127 Objective Loss 0.323127 LR 0.001000 Time 0.059826 +2023-10-05 21:27:41,020 - Epoch: [97][ 30/ 1236] Overall Loss 0.320144 Objective Loss 0.320144 LR 0.001000 Time 0.046534 +2023-10-05 21:27:41,222 - Epoch: [97][ 40/ 1236] Overall Loss 0.315836 Objective Loss 0.315836 LR 0.001000 Time 0.039931 +2023-10-05 21:27:41,421 - Epoch: [97][ 50/ 1236] Overall Loss 0.317800 Objective Loss 0.317800 LR 0.001000 Time 0.035921 +2023-10-05 21:27:41,624 - Epoch: [97][ 60/ 1236] Overall Loss 0.318490 Objective Loss 0.318490 LR 0.001000 Time 0.033307 +2023-10-05 21:27:41,824 - Epoch: [97][ 70/ 1236] Overall Loss 0.319385 Objective Loss 0.319385 LR 0.001000 Time 0.031405 +2023-10-05 21:27:42,026 - Epoch: [97][ 80/ 1236] Overall Loss 0.318591 Objective Loss 0.318591 LR 0.001000 Time 0.029997 +2023-10-05 21:27:42,225 - Epoch: [97][ 90/ 1236] Overall Loss 0.323253 Objective Loss 0.323253 LR 0.001000 Time 0.028873 +2023-10-05 21:27:42,427 - Epoch: [97][ 100/ 1236] Overall Loss 0.325691 Objective Loss 0.325691 LR 0.001000 Time 0.028003 +2023-10-05 21:27:42,625 - Epoch: [97][ 110/ 1236] Overall Loss 0.326209 Objective Loss 0.326209 LR 0.001000 Time 0.027259 +2023-10-05 21:27:42,826 - Epoch: [97][ 120/ 1236] Overall Loss 0.324508 Objective Loss 0.324508 LR 0.001000 Time 0.026657 +2023-10-05 21:27:43,024 - Epoch: [97][ 130/ 1236] Overall Loss 0.321665 Objective Loss 0.321665 LR 0.001000 Time 0.026128 +2023-10-05 21:27:43,225 - Epoch: [97][ 140/ 1236] Overall Loss 0.322596 Objective Loss 0.322596 LR 0.001000 Time 0.025692 +2023-10-05 21:27:43,423 - Epoch: [97][ 150/ 1236] Overall Loss 0.321212 Objective Loss 0.321212 LR 0.001000 Time 0.025298 +2023-10-05 21:27:43,623 - Epoch: [97][ 160/ 1236] Overall Loss 0.322623 Objective Loss 0.322623 LR 0.001000 Time 0.024967 +2023-10-05 21:27:43,822 - Epoch: [97][ 170/ 1236] Overall Loss 0.323061 Objective Loss 0.323061 LR 0.001000 Time 0.024664 +2023-10-05 21:27:44,022 - Epoch: [97][ 180/ 1236] Overall Loss 0.324520 Objective Loss 0.324520 LR 0.001000 Time 0.024405 +2023-10-05 21:27:44,221 - Epoch: [97][ 190/ 1236] Overall Loss 0.324805 Objective Loss 0.324805 LR 0.001000 Time 0.024163 +2023-10-05 21:27:44,421 - Epoch: [97][ 200/ 1236] Overall Loss 0.325484 Objective Loss 0.325484 LR 0.001000 Time 0.023955 +2023-10-05 21:27:44,620 - Epoch: [97][ 210/ 1236] Overall Loss 0.326492 Objective Loss 0.326492 LR 0.001000 Time 0.023757 +2023-10-05 21:27:44,820 - Epoch: [97][ 220/ 1236] Overall Loss 0.326473 Objective Loss 0.326473 LR 0.001000 Time 0.023587 +2023-10-05 21:27:45,019 - Epoch: [97][ 230/ 1236] Overall Loss 0.324972 Objective Loss 0.324972 LR 0.001000 Time 0.023423 +2023-10-05 21:27:45,219 - Epoch: [97][ 240/ 1236] Overall Loss 0.324163 Objective Loss 0.324163 LR 0.001000 Time 0.023281 +2023-10-05 21:27:45,417 - Epoch: [97][ 250/ 1236] Overall Loss 0.323507 Objective Loss 0.323507 LR 0.001000 Time 0.023140 +2023-10-05 21:27:45,618 - Epoch: [97][ 260/ 1236] Overall Loss 0.324651 Objective Loss 0.324651 LR 0.001000 Time 0.023022 +2023-10-05 21:27:45,816 - Epoch: [97][ 270/ 1236] Overall Loss 0.324680 Objective Loss 0.324680 LR 0.001000 Time 0.022902 +2023-10-05 21:27:46,017 - Epoch: [97][ 280/ 1236] Overall Loss 0.325434 Objective Loss 0.325434 LR 0.001000 Time 0.022799 +2023-10-05 21:27:46,216 - Epoch: [97][ 290/ 1236] Overall Loss 0.325387 Objective Loss 0.325387 LR 0.001000 Time 0.022696 +2023-10-05 21:27:46,416 - Epoch: [97][ 300/ 1236] Overall Loss 0.325635 Objective Loss 0.325635 LR 0.001000 Time 0.022607 +2023-10-05 21:27:46,615 - Epoch: [97][ 310/ 1236] Overall Loss 0.325262 Objective Loss 0.325262 LR 0.001000 Time 0.022517 +2023-10-05 21:27:46,815 - Epoch: [97][ 320/ 1236] Overall Loss 0.324981 Objective Loss 0.324981 LR 0.001000 Time 0.022438 +2023-10-05 21:27:47,014 - Epoch: [97][ 330/ 1236] Overall Loss 0.325005 Objective Loss 0.325005 LR 0.001000 Time 0.022359 +2023-10-05 21:27:47,214 - Epoch: [97][ 340/ 1236] Overall Loss 0.325055 Objective Loss 0.325055 LR 0.001000 Time 0.022290 +2023-10-05 21:27:47,413 - Epoch: [97][ 350/ 1236] Overall Loss 0.325516 Objective Loss 0.325516 LR 0.001000 Time 0.022220 +2023-10-05 21:27:47,613 - Epoch: [97][ 360/ 1236] Overall Loss 0.325422 Objective Loss 0.325422 LR 0.001000 Time 0.022158 +2023-10-05 21:27:47,812 - Epoch: [97][ 370/ 1236] Overall Loss 0.325786 Objective Loss 0.325786 LR 0.001000 Time 0.022096 +2023-10-05 21:27:48,013 - Epoch: [97][ 380/ 1236] Overall Loss 0.325240 Objective Loss 0.325240 LR 0.001000 Time 0.022043 +2023-10-05 21:27:48,214 - Epoch: [97][ 390/ 1236] Overall Loss 0.325883 Objective Loss 0.325883 LR 0.001000 Time 0.021990 +2023-10-05 21:27:48,416 - Epoch: [97][ 400/ 1236] Overall Loss 0.325426 Objective Loss 0.325426 LR 0.001000 Time 0.021945 +2023-10-05 21:27:48,615 - Epoch: [97][ 410/ 1236] Overall Loss 0.325402 Objective Loss 0.325402 LR 0.001000 Time 0.021895 +2023-10-05 21:27:48,817 - Epoch: [97][ 420/ 1236] Overall Loss 0.325198 Objective Loss 0.325198 LR 0.001000 Time 0.021853 +2023-10-05 21:27:49,017 - Epoch: [97][ 430/ 1236] Overall Loss 0.324890 Objective Loss 0.324890 LR 0.001000 Time 0.021809 +2023-10-05 21:27:49,219 - Epoch: [97][ 440/ 1236] Overall Loss 0.325269 Objective Loss 0.325269 LR 0.001000 Time 0.021771 +2023-10-05 21:27:49,418 - Epoch: [97][ 450/ 1236] Overall Loss 0.325217 Objective Loss 0.325217 LR 0.001000 Time 0.021730 +2023-10-05 21:27:49,620 - Epoch: [97][ 460/ 1236] Overall Loss 0.324995 Objective Loss 0.324995 LR 0.001000 Time 0.021695 +2023-10-05 21:27:49,819 - Epoch: [97][ 470/ 1236] Overall Loss 0.325227 Objective Loss 0.325227 LR 0.001000 Time 0.021657 +2023-10-05 21:27:50,021 - Epoch: [97][ 480/ 1236] Overall Loss 0.325374 Objective Loss 0.325374 LR 0.001000 Time 0.021626 +2023-10-05 21:27:50,221 - Epoch: [97][ 490/ 1236] Overall Loss 0.325098 Objective Loss 0.325098 LR 0.001000 Time 0.021591 +2023-10-05 21:27:50,423 - Epoch: [97][ 500/ 1236] Overall Loss 0.324742 Objective Loss 0.324742 LR 0.001000 Time 0.021563 +2023-10-05 21:27:50,622 - Epoch: [97][ 510/ 1236] Overall Loss 0.323848 Objective Loss 0.323848 LR 0.001000 Time 0.021530 +2023-10-05 21:27:50,825 - Epoch: [97][ 520/ 1236] Overall Loss 0.323421 Objective Loss 0.323421 LR 0.001000 Time 0.021505 +2023-10-05 21:27:51,025 - Epoch: [97][ 530/ 1236] Overall Loss 0.323719 Objective Loss 0.323719 LR 0.001000 Time 0.021476 +2023-10-05 21:27:51,228 - Epoch: [97][ 540/ 1236] Overall Loss 0.324013 Objective Loss 0.324013 LR 0.001000 Time 0.021453 +2023-10-05 21:27:51,427 - Epoch: [97][ 550/ 1236] Overall Loss 0.324366 Objective Loss 0.324366 LR 0.001000 Time 0.021425 +2023-10-05 21:27:51,629 - Epoch: [97][ 560/ 1236] Overall Loss 0.324256 Objective Loss 0.324256 LR 0.001000 Time 0.021403 +2023-10-05 21:27:51,829 - Epoch: [97][ 570/ 1236] Overall Loss 0.324324 Objective Loss 0.324324 LR 0.001000 Time 0.021377 +2023-10-05 21:27:52,031 - Epoch: [97][ 580/ 1236] Overall Loss 0.323688 Objective Loss 0.323688 LR 0.001000 Time 0.021356 +2023-10-05 21:27:52,231 - Epoch: [97][ 590/ 1236] Overall Loss 0.323866 Objective Loss 0.323866 LR 0.001000 Time 0.021332 +2023-10-05 21:27:52,433 - Epoch: [97][ 600/ 1236] Overall Loss 0.324460 Objective Loss 0.324460 LR 0.001000 Time 0.021313 +2023-10-05 21:27:52,633 - Epoch: [97][ 610/ 1236] Overall Loss 0.324937 Objective Loss 0.324937 LR 0.001000 Time 0.021291 +2023-10-05 21:27:52,835 - Epoch: [97][ 620/ 1236] Overall Loss 0.325293 Objective Loss 0.325293 LR 0.001000 Time 0.021272 +2023-10-05 21:27:53,035 - Epoch: [97][ 630/ 1236] Overall Loss 0.326334 Objective Loss 0.326334 LR 0.001000 Time 0.021251 +2023-10-05 21:27:53,237 - Epoch: [97][ 640/ 1236] Overall Loss 0.326605 Objective Loss 0.326605 LR 0.001000 Time 0.021235 +2023-10-05 21:27:53,438 - Epoch: [97][ 650/ 1236] Overall Loss 0.326197 Objective Loss 0.326197 LR 0.001000 Time 0.021216 +2023-10-05 21:27:53,640 - Epoch: [97][ 660/ 1236] Overall Loss 0.327172 Objective Loss 0.327172 LR 0.001000 Time 0.021200 +2023-10-05 21:27:53,840 - Epoch: [97][ 670/ 1236] Overall Loss 0.326968 Objective Loss 0.326968 LR 0.001000 Time 0.021182 +2023-10-05 21:27:54,042 - Epoch: [97][ 680/ 1236] Overall Loss 0.326776 Objective Loss 0.326776 LR 0.001000 Time 0.021168 +2023-10-05 21:27:54,242 - Epoch: [97][ 690/ 1236] Overall Loss 0.326401 Objective Loss 0.326401 LR 0.001000 Time 0.021150 +2023-10-05 21:27:54,444 - Epoch: [97][ 700/ 1236] Overall Loss 0.326291 Objective Loss 0.326291 LR 0.001000 Time 0.021136 +2023-10-05 21:27:54,644 - Epoch: [97][ 710/ 1236] Overall Loss 0.325823 Objective Loss 0.325823 LR 0.001000 Time 0.021120 +2023-10-05 21:27:54,847 - Epoch: [97][ 720/ 1236] Overall Loss 0.325834 Objective Loss 0.325834 LR 0.001000 Time 0.021107 +2023-10-05 21:27:55,047 - Epoch: [97][ 730/ 1236] Overall Loss 0.325734 Objective Loss 0.325734 LR 0.001000 Time 0.021092 +2023-10-05 21:27:55,249 - Epoch: [97][ 740/ 1236] Overall Loss 0.325331 Objective Loss 0.325331 LR 0.001000 Time 0.021079 +2023-10-05 21:27:55,449 - Epoch: [97][ 750/ 1236] Overall Loss 0.325287 Objective Loss 0.325287 LR 0.001000 Time 0.021065 +2023-10-05 21:27:55,652 - Epoch: [97][ 760/ 1236] Overall Loss 0.325776 Objective Loss 0.325776 LR 0.001000 Time 0.021053 +2023-10-05 21:27:55,852 - Epoch: [97][ 770/ 1236] Overall Loss 0.325779 Objective Loss 0.325779 LR 0.001000 Time 0.021040 +2023-10-05 21:27:56,054 - Epoch: [97][ 780/ 1236] Overall Loss 0.325730 Objective Loss 0.325730 LR 0.001000 Time 0.021028 +2023-10-05 21:27:56,254 - Epoch: [97][ 790/ 1236] Overall Loss 0.325648 Objective Loss 0.325648 LR 0.001000 Time 0.021015 +2023-10-05 21:27:56,456 - Epoch: [97][ 800/ 1236] Overall Loss 0.325637 Objective Loss 0.325637 LR 0.001000 Time 0.021004 +2023-10-05 21:27:56,657 - Epoch: [97][ 810/ 1236] Overall Loss 0.325893 Objective Loss 0.325893 LR 0.001000 Time 0.020992 +2023-10-05 21:27:56,859 - Epoch: [97][ 820/ 1236] Overall Loss 0.326097 Objective Loss 0.326097 LR 0.001000 Time 0.020982 +2023-10-05 21:27:57,059 - Epoch: [97][ 830/ 1236] Overall Loss 0.326189 Objective Loss 0.326189 LR 0.001000 Time 0.020970 +2023-10-05 21:27:57,261 - Epoch: [97][ 840/ 1236] Overall Loss 0.326340 Objective Loss 0.326340 LR 0.001000 Time 0.020961 +2023-10-05 21:27:57,461 - Epoch: [97][ 850/ 1236] Overall Loss 0.326351 Objective Loss 0.326351 LR 0.001000 Time 0.020949 +2023-10-05 21:27:57,664 - Epoch: [97][ 860/ 1236] Overall Loss 0.327008 Objective Loss 0.327008 LR 0.001000 Time 0.020940 +2023-10-05 21:27:57,864 - Epoch: [97][ 870/ 1236] Overall Loss 0.327109 Objective Loss 0.327109 LR 0.001000 Time 0.020929 +2023-10-05 21:27:58,066 - Epoch: [97][ 880/ 1236] Overall Loss 0.327226 Objective Loss 0.327226 LR 0.001000 Time 0.020921 +2023-10-05 21:27:58,266 - Epoch: [97][ 890/ 1236] Overall Loss 0.327059 Objective Loss 0.327059 LR 0.001000 Time 0.020911 +2023-10-05 21:27:58,468 - Epoch: [97][ 900/ 1236] Overall Loss 0.327279 Objective Loss 0.327279 LR 0.001000 Time 0.020902 +2023-10-05 21:27:58,669 - Epoch: [97][ 910/ 1236] Overall Loss 0.327605 Objective Loss 0.327605 LR 0.001000 Time 0.020893 +2023-10-05 21:27:58,871 - Epoch: [97][ 920/ 1236] Overall Loss 0.328060 Objective Loss 0.328060 LR 0.001000 Time 0.020884 +2023-10-05 21:27:59,071 - Epoch: [97][ 930/ 1236] Overall Loss 0.328266 Objective Loss 0.328266 LR 0.001000 Time 0.020875 +2023-10-05 21:27:59,273 - Epoch: [97][ 940/ 1236] Overall Loss 0.328293 Objective Loss 0.328293 LR 0.001000 Time 0.020867 +2023-10-05 21:27:59,473 - Epoch: [97][ 950/ 1236] Overall Loss 0.328361 Objective Loss 0.328361 LR 0.001000 Time 0.020858 +2023-10-05 21:27:59,675 - Epoch: [97][ 960/ 1236] Overall Loss 0.328434 Objective Loss 0.328434 LR 0.001000 Time 0.020851 +2023-10-05 21:27:59,876 - Epoch: [97][ 970/ 1236] Overall Loss 0.328666 Objective Loss 0.328666 LR 0.001000 Time 0.020842 +2023-10-05 21:28:00,078 - Epoch: [97][ 980/ 1236] Overall Loss 0.328820 Objective Loss 0.328820 LR 0.001000 Time 0.020835 +2023-10-05 21:28:00,278 - Epoch: [97][ 990/ 1236] Overall Loss 0.328659 Objective Loss 0.328659 LR 0.001000 Time 0.020827 +2023-10-05 21:28:00,480 - Epoch: [97][ 1000/ 1236] Overall Loss 0.328820 Objective Loss 0.328820 LR 0.001000 Time 0.020820 +2023-10-05 21:28:00,680 - Epoch: [97][ 1010/ 1236] Overall Loss 0.328904 Objective Loss 0.328904 LR 0.001000 Time 0.020812 +2023-10-05 21:28:00,882 - Epoch: [97][ 1020/ 1236] Overall Loss 0.329127 Objective Loss 0.329127 LR 0.001000 Time 0.020805 +2023-10-05 21:28:01,082 - Epoch: [97][ 1030/ 1236] Overall Loss 0.328845 Objective Loss 0.328845 LR 0.001000 Time 0.020797 +2023-10-05 21:28:01,284 - Epoch: [97][ 1040/ 1236] Overall Loss 0.328662 Objective Loss 0.328662 LR 0.001000 Time 0.020791 +2023-10-05 21:28:01,485 - Epoch: [97][ 1050/ 1236] Overall Loss 0.328720 Objective Loss 0.328720 LR 0.001000 Time 0.020784 +2023-10-05 21:28:01,687 - Epoch: [97][ 1060/ 1236] Overall Loss 0.328833 Objective Loss 0.328833 LR 0.001000 Time 0.020778 +2023-10-05 21:28:01,887 - Epoch: [97][ 1070/ 1236] Overall Loss 0.328911 Objective Loss 0.328911 LR 0.001000 Time 0.020771 +2023-10-05 21:28:02,089 - Epoch: [97][ 1080/ 1236] Overall Loss 0.329286 Objective Loss 0.329286 LR 0.001000 Time 0.020765 +2023-10-05 21:28:02,290 - Epoch: [97][ 1090/ 1236] Overall Loss 0.329273 Objective Loss 0.329273 LR 0.001000 Time 0.020758 +2023-10-05 21:28:02,492 - Epoch: [97][ 1100/ 1236] Overall Loss 0.329256 Objective Loss 0.329256 LR 0.001000 Time 0.020752 +2023-10-05 21:28:02,692 - Epoch: [97][ 1110/ 1236] Overall Loss 0.329384 Objective Loss 0.329384 LR 0.001000 Time 0.020746 +2023-10-05 21:28:02,895 - Epoch: [97][ 1120/ 1236] Overall Loss 0.329421 Objective Loss 0.329421 LR 0.001000 Time 0.020741 +2023-10-05 21:28:03,095 - Epoch: [97][ 1130/ 1236] Overall Loss 0.329546 Objective Loss 0.329546 LR 0.001000 Time 0.020735 +2023-10-05 21:28:03,297 - Epoch: [97][ 1140/ 1236] Overall Loss 0.329499 Objective Loss 0.329499 LR 0.001000 Time 0.020730 +2023-10-05 21:28:03,497 - Epoch: [97][ 1150/ 1236] Overall Loss 0.329753 Objective Loss 0.329753 LR 0.001000 Time 0.020723 +2023-10-05 21:28:03,700 - Epoch: [97][ 1160/ 1236] Overall Loss 0.329742 Objective Loss 0.329742 LR 0.001000 Time 0.020718 +2023-10-05 21:28:03,900 - Epoch: [97][ 1170/ 1236] Overall Loss 0.329709 Objective Loss 0.329709 LR 0.001000 Time 0.020712 +2023-10-05 21:28:04,102 - Epoch: [97][ 1180/ 1236] Overall Loss 0.329521 Objective Loss 0.329521 LR 0.001000 Time 0.020707 +2023-10-05 21:28:04,302 - Epoch: [97][ 1190/ 1236] Overall Loss 0.329557 Objective Loss 0.329557 LR 0.001000 Time 0.020701 +2023-10-05 21:28:04,504 - Epoch: [97][ 1200/ 1236] Overall Loss 0.329515 Objective Loss 0.329515 LR 0.001000 Time 0.020697 +2023-10-05 21:28:04,704 - Epoch: [97][ 1210/ 1236] Overall Loss 0.329496 Objective Loss 0.329496 LR 0.001000 Time 0.020691 +2023-10-05 21:28:04,906 - Epoch: [97][ 1220/ 1236] Overall Loss 0.329552 Objective Loss 0.329552 LR 0.001000 Time 0.020686 +2023-10-05 21:28:05,160 - Epoch: [97][ 1230/ 1236] Overall Loss 0.329776 Objective Loss 0.329776 LR 0.001000 Time 0.020724 +2023-10-05 21:28:05,278 - Epoch: [97][ 1236/ 1236] Overall Loss 0.329701 Objective Loss 0.329701 Top1 83.910387 Top5 96.130346 LR 0.001000 Time 0.020719 +2023-10-05 21:28:05,415 - --- validate (epoch=97)----------- +2023-10-05 21:28:05,415 - 29943 samples (256 per mini-batch) +2023-10-05 21:28:05,873 - Epoch: [97][ 10/ 117] Loss 0.381971 Top1 80.664062 Top5 97.031250 +2023-10-05 21:28:06,022 - Epoch: [97][ 20/ 117] Loss 0.391813 Top1 80.703125 Top5 97.167969 +2023-10-05 21:28:06,172 - Epoch: [97][ 30/ 117] Loss 0.386918 Top1 80.716146 Top5 97.135417 +2023-10-05 21:28:06,321 - Epoch: [97][ 40/ 117] Loss 0.376504 Top1 80.859375 Top5 97.197266 +2023-10-05 21:28:06,470 - Epoch: [97][ 50/ 117] Loss 0.373787 Top1 81.031250 Top5 97.226562 +2023-10-05 21:28:06,619 - Epoch: [97][ 60/ 117] Loss 0.373467 Top1 81.022135 Top5 97.239583 +2023-10-05 21:28:06,767 - Epoch: [97][ 70/ 117] Loss 0.377494 Top1 80.864955 Top5 97.159598 +2023-10-05 21:28:06,914 - Epoch: [97][ 80/ 117] Loss 0.374054 Top1 80.869141 Top5 97.163086 +2023-10-05 21:28:07,062 - Epoch: [97][ 90/ 117] Loss 0.372776 Top1 80.898438 Top5 97.222222 +2023-10-05 21:28:07,210 - Epoch: [97][ 100/ 117] Loss 0.369276 Top1 80.937500 Top5 97.285156 +2023-10-05 21:28:07,364 - Epoch: [97][ 110/ 117] Loss 0.368927 Top1 81.044034 Top5 97.286932 +2023-10-05 21:28:07,449 - Epoch: [97][ 117/ 117] Loss 0.368014 Top1 81.074041 Top5 97.318238 +2023-10-05 21:28:07,592 - ==> Top1: 81.074 Top5: 97.318 Loss: 0.368 + +2023-10-05 21:28:07,592 - ==> Confusion: +[[ 915 5 5 0 14 3 0 2 2 66 1 3 3 4 6 2 6 1 0 0 12] + [ 0 1023 4 1 10 31 1 28 3 0 3 2 0 0 0 2 2 1 8 6 6] + [ 8 2 916 27 1 0 42 14 0 2 5 2 11 1 0 4 1 2 6 4 8] + [ 4 2 11 966 3 4 1 1 3 0 8 1 4 4 26 0 0 5 29 5 12] + [ 20 6 0 2 959 9 1 1 0 12 1 1 2 1 9 5 10 3 0 1 7] + [ 1 34 1 0 5 998 2 21 1 2 4 8 3 11 3 1 0 0 4 5 12] + [ 0 7 20 0 0 2 1123 5 0 0 1 1 2 2 1 9 0 2 3 7 6] + [ 4 21 20 0 1 39 7 1054 0 2 2 10 3 0 0 3 0 1 35 9 7] + [ 22 5 1 1 1 4 1 0 951 50 8 2 2 15 10 5 1 2 4 2 2] + [ 132 3 1 0 6 7 1 0 21 900 0 2 0 29 3 4 1 0 1 3 5] + [ 3 11 8 8 0 10 8 4 21 0 934 4 0 16 4 2 2 1 11 0 6] + [ 3 3 0 0 0 22 1 0 0 1 1 927 37 6 0 2 1 17 0 7 7] + [ 3 3 2 4 2 3 1 1 4 0 1 34 960 2 3 7 3 19 3 6 7] + [ 0 0 2 2 8 18 0 1 6 12 7 3 6 1040 3 2 1 1 0 4 3] + [ 26 4 3 21 9 1 0 0 30 5 4 1 5 1 947 1 1 8 21 0 13] + [ 0 1 6 1 4 1 2 0 0 0 0 6 11 2 1 1046 18 19 1 7 8] + [ 2 20 1 1 7 9 0 1 3 0 0 6 10 3 2 13 1066 0 0 5 12] + [ 1 0 0 0 0 2 1 0 0 0 0 5 18 0 1 3 0 1002 2 2 1] + [ 4 15 4 14 0 0 2 34 7 0 0 0 5 0 10 0 2 0 964 0 7] + [ 0 2 2 1 1 12 18 14 1 0 1 8 4 1 0 8 8 2 1 1059 9] + [ 154 223 142 129 123 233 63 145 126 98 158 162 410 331 130 64 167 93 182 246 4526]] + +2023-10-05 21:28:07,594 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:28:07,594 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:28:07,600 - + +2023-10-05 21:28:07,600 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:28:08,593 - Epoch: [98][ 10/ 1236] Overall Loss 0.333352 Objective Loss 0.333352 LR 0.001000 Time 0.099303 +2023-10-05 21:28:08,796 - Epoch: [98][ 20/ 1236] Overall Loss 0.327262 Objective Loss 0.327262 LR 0.001000 Time 0.059770 +2023-10-05 21:28:08,994 - Epoch: [98][ 30/ 1236] Overall Loss 0.317103 Objective Loss 0.317103 LR 0.001000 Time 0.046451 +2023-10-05 21:28:09,198 - Epoch: [98][ 40/ 1236] Overall Loss 0.312298 Objective Loss 0.312298 LR 0.001000 Time 0.039916 +2023-10-05 21:28:09,397 - Epoch: [98][ 50/ 1236] Overall Loss 0.308895 Objective Loss 0.308895 LR 0.001000 Time 0.035918 +2023-10-05 21:28:09,599 - Epoch: [98][ 60/ 1236] Overall Loss 0.312730 Objective Loss 0.312730 LR 0.001000 Time 0.033281 +2023-10-05 21:28:09,797 - Epoch: [98][ 70/ 1236] Overall Loss 0.314329 Objective Loss 0.314329 LR 0.001000 Time 0.031359 +2023-10-05 21:28:10,000 - Epoch: [98][ 80/ 1236] Overall Loss 0.314833 Objective Loss 0.314833 LR 0.001000 Time 0.029965 +2023-10-05 21:28:10,200 - Epoch: [98][ 90/ 1236] Overall Loss 0.314497 Objective Loss 0.314497 LR 0.001000 Time 0.028855 +2023-10-05 21:28:10,403 - Epoch: [98][ 100/ 1236] Overall Loss 0.312851 Objective Loss 0.312851 LR 0.001000 Time 0.027998 +2023-10-05 21:28:10,601 - Epoch: [98][ 110/ 1236] Overall Loss 0.314090 Objective Loss 0.314090 LR 0.001000 Time 0.027252 +2023-10-05 21:28:10,804 - Epoch: [98][ 120/ 1236] Overall Loss 0.314664 Objective Loss 0.314664 LR 0.001000 Time 0.026666 +2023-10-05 21:28:11,004 - Epoch: [98][ 130/ 1236] Overall Loss 0.314288 Objective Loss 0.314288 LR 0.001000 Time 0.026151 +2023-10-05 21:28:11,206 - Epoch: [98][ 140/ 1236] Overall Loss 0.313837 Objective Loss 0.313837 LR 0.001000 Time 0.025727 +2023-10-05 21:28:11,407 - Epoch: [98][ 150/ 1236] Overall Loss 0.315552 Objective Loss 0.315552 LR 0.001000 Time 0.025345 +2023-10-05 21:28:11,609 - Epoch: [98][ 160/ 1236] Overall Loss 0.315932 Objective Loss 0.315932 LR 0.001000 Time 0.025023 +2023-10-05 21:28:11,809 - Epoch: [98][ 170/ 1236] Overall Loss 0.316236 Objective Loss 0.316236 LR 0.001000 Time 0.024726 +2023-10-05 21:28:12,012 - Epoch: [98][ 180/ 1236] Overall Loss 0.316286 Objective Loss 0.316286 LR 0.001000 Time 0.024476 +2023-10-05 21:28:12,212 - Epoch: [98][ 190/ 1236] Overall Loss 0.315797 Objective Loss 0.315797 LR 0.001000 Time 0.024239 +2023-10-05 21:28:12,415 - Epoch: [98][ 200/ 1236] Overall Loss 0.315122 Objective Loss 0.315122 LR 0.001000 Time 0.024041 +2023-10-05 21:28:12,615 - Epoch: [98][ 210/ 1236] Overall Loss 0.315010 Objective Loss 0.315010 LR 0.001000 Time 0.023846 +2023-10-05 21:28:12,822 - Epoch: [98][ 220/ 1236] Overall Loss 0.315289 Objective Loss 0.315289 LR 0.001000 Time 0.023703 +2023-10-05 21:28:13,022 - Epoch: [98][ 230/ 1236] Overall Loss 0.316299 Objective Loss 0.316299 LR 0.001000 Time 0.023542 +2023-10-05 21:28:13,225 - Epoch: [98][ 240/ 1236] Overall Loss 0.317016 Objective Loss 0.317016 LR 0.001000 Time 0.023404 +2023-10-05 21:28:13,425 - Epoch: [98][ 250/ 1236] Overall Loss 0.317428 Objective Loss 0.317428 LR 0.001000 Time 0.023265 +2023-10-05 21:28:13,628 - Epoch: [98][ 260/ 1236] Overall Loss 0.317913 Objective Loss 0.317913 LR 0.001000 Time 0.023151 +2023-10-05 21:28:13,827 - Epoch: [98][ 270/ 1236] Overall Loss 0.319563 Objective Loss 0.319563 LR 0.001000 Time 0.023031 +2023-10-05 21:28:14,030 - Epoch: [98][ 280/ 1236] Overall Loss 0.318566 Objective Loss 0.318566 LR 0.001000 Time 0.022931 +2023-10-05 21:28:14,230 - Epoch: [98][ 290/ 1236] Overall Loss 0.318618 Objective Loss 0.318618 LR 0.001000 Time 0.022827 +2023-10-05 21:28:14,433 - Epoch: [98][ 300/ 1236] Overall Loss 0.319702 Objective Loss 0.319702 LR 0.001000 Time 0.022742 +2023-10-05 21:28:14,633 - Epoch: [98][ 310/ 1236] Overall Loss 0.319738 Objective Loss 0.319738 LR 0.001000 Time 0.022654 +2023-10-05 21:28:14,836 - Epoch: [98][ 320/ 1236] Overall Loss 0.319460 Objective Loss 0.319460 LR 0.001000 Time 0.022578 +2023-10-05 21:28:15,036 - Epoch: [98][ 330/ 1236] Overall Loss 0.318740 Objective Loss 0.318740 LR 0.001000 Time 0.022500 +2023-10-05 21:28:15,239 - Epoch: [98][ 340/ 1236] Overall Loss 0.318692 Objective Loss 0.318692 LR 0.001000 Time 0.022433 +2023-10-05 21:28:15,439 - Epoch: [98][ 350/ 1236] Overall Loss 0.318330 Objective Loss 0.318330 LR 0.001000 Time 0.022362 +2023-10-05 21:28:15,649 - Epoch: [98][ 360/ 1236] Overall Loss 0.318891 Objective Loss 0.318891 LR 0.001000 Time 0.022324 +2023-10-05 21:28:15,848 - Epoch: [98][ 370/ 1236] Overall Loss 0.318733 Objective Loss 0.318733 LR 0.001000 Time 0.022256 +2023-10-05 21:28:16,048 - Epoch: [98][ 380/ 1236] Overall Loss 0.319663 Objective Loss 0.319663 LR 0.001000 Time 0.022198 +2023-10-05 21:28:16,247 - Epoch: [98][ 390/ 1236] Overall Loss 0.319677 Objective Loss 0.319677 LR 0.001000 Time 0.022139 +2023-10-05 21:28:16,449 - Epoch: [98][ 400/ 1236] Overall Loss 0.319150 Objective Loss 0.319150 LR 0.001000 Time 0.022087 +2023-10-05 21:28:16,649 - Epoch: [98][ 410/ 1236] Overall Loss 0.318799 Objective Loss 0.318799 LR 0.001000 Time 0.022036 +2023-10-05 21:28:16,850 - Epoch: [98][ 420/ 1236] Overall Loss 0.319340 Objective Loss 0.319340 LR 0.001000 Time 0.021991 +2023-10-05 21:28:17,051 - Epoch: [98][ 430/ 1236] Overall Loss 0.319245 Objective Loss 0.319245 LR 0.001000 Time 0.021945 +2023-10-05 21:28:17,262 - Epoch: [98][ 440/ 1236] Overall Loss 0.319165 Objective Loss 0.319165 LR 0.001000 Time 0.021924 +2023-10-05 21:28:17,469 - Epoch: [98][ 450/ 1236] Overall Loss 0.319334 Objective Loss 0.319334 LR 0.001000 Time 0.021896 +2023-10-05 21:28:17,679 - Epoch: [98][ 460/ 1236] Overall Loss 0.319100 Objective Loss 0.319100 LR 0.001000 Time 0.021877 +2023-10-05 21:28:17,886 - Epoch: [98][ 470/ 1236] Overall Loss 0.319576 Objective Loss 0.319576 LR 0.001000 Time 0.021851 +2023-10-05 21:28:18,097 - Epoch: [98][ 480/ 1236] Overall Loss 0.320519 Objective Loss 0.320519 LR 0.001000 Time 0.021834 +2023-10-05 21:28:18,306 - Epoch: [98][ 490/ 1236] Overall Loss 0.320620 Objective Loss 0.320620 LR 0.001000 Time 0.021814 +2023-10-05 21:28:18,516 - Epoch: [98][ 500/ 1236] Overall Loss 0.320579 Objective Loss 0.320579 LR 0.001000 Time 0.021798 +2023-10-05 21:28:18,723 - Epoch: [98][ 510/ 1236] Overall Loss 0.320290 Objective Loss 0.320290 LR 0.001000 Time 0.021776 +2023-10-05 21:28:18,935 - Epoch: [98][ 520/ 1236] Overall Loss 0.319961 Objective Loss 0.319961 LR 0.001000 Time 0.021764 +2023-10-05 21:28:19,142 - Epoch: [98][ 530/ 1236] Overall Loss 0.319136 Objective Loss 0.319136 LR 0.001000 Time 0.021742 +2023-10-05 21:28:19,353 - Epoch: [98][ 540/ 1236] Overall Loss 0.319417 Objective Loss 0.319417 LR 0.001000 Time 0.021729 +2023-10-05 21:28:19,560 - Epoch: [98][ 550/ 1236] Overall Loss 0.319468 Objective Loss 0.319468 LR 0.001000 Time 0.021710 +2023-10-05 21:28:19,770 - Epoch: [98][ 560/ 1236] Overall Loss 0.319390 Objective Loss 0.319390 LR 0.001000 Time 0.021698 +2023-10-05 21:28:19,978 - Epoch: [98][ 570/ 1236] Overall Loss 0.318683 Objective Loss 0.318683 LR 0.001000 Time 0.021681 +2023-10-05 21:28:20,188 - Epoch: [98][ 580/ 1236] Overall Loss 0.319089 Objective Loss 0.319089 LR 0.001000 Time 0.021669 +2023-10-05 21:28:20,395 - Epoch: [98][ 590/ 1236] Overall Loss 0.319058 Objective Loss 0.319058 LR 0.001000 Time 0.021653 +2023-10-05 21:28:20,606 - Epoch: [98][ 600/ 1236] Overall Loss 0.319274 Objective Loss 0.319274 LR 0.001000 Time 0.021642 +2023-10-05 21:28:20,814 - Epoch: [98][ 610/ 1236] Overall Loss 0.319685 Objective Loss 0.319685 LR 0.001000 Time 0.021628 +2023-10-05 21:28:21,025 - Epoch: [98][ 620/ 1236] Overall Loss 0.319351 Objective Loss 0.319351 LR 0.001000 Time 0.021619 +2023-10-05 21:28:21,232 - Epoch: [98][ 630/ 1236] Overall Loss 0.318972 Objective Loss 0.318972 LR 0.001000 Time 0.021604 +2023-10-05 21:28:21,443 - Epoch: [98][ 640/ 1236] Overall Loss 0.319196 Objective Loss 0.319196 LR 0.001000 Time 0.021595 +2023-10-05 21:28:21,650 - Epoch: [98][ 650/ 1236] Overall Loss 0.319636 Objective Loss 0.319636 LR 0.001000 Time 0.021582 +2023-10-05 21:28:21,861 - Epoch: [98][ 660/ 1236] Overall Loss 0.319823 Objective Loss 0.319823 LR 0.001000 Time 0.021573 +2023-10-05 21:28:22,069 - Epoch: [98][ 670/ 1236] Overall Loss 0.319387 Objective Loss 0.319387 LR 0.001000 Time 0.021561 +2023-10-05 21:28:22,280 - Epoch: [98][ 680/ 1236] Overall Loss 0.319158 Objective Loss 0.319158 LR 0.001000 Time 0.021553 +2023-10-05 21:28:22,487 - Epoch: [98][ 690/ 1236] Overall Loss 0.318912 Objective Loss 0.318912 LR 0.001000 Time 0.021541 +2023-10-05 21:28:22,698 - Epoch: [98][ 700/ 1236] Overall Loss 0.318966 Objective Loss 0.318966 LR 0.001000 Time 0.021535 +2023-10-05 21:28:22,905 - Epoch: [98][ 710/ 1236] Overall Loss 0.318771 Objective Loss 0.318771 LR 0.001000 Time 0.021522 +2023-10-05 21:28:23,116 - Epoch: [98][ 720/ 1236] Overall Loss 0.319184 Objective Loss 0.319184 LR 0.001000 Time 0.021516 +2023-10-05 21:28:23,323 - Epoch: [98][ 730/ 1236] Overall Loss 0.319151 Objective Loss 0.319151 LR 0.001000 Time 0.021505 +2023-10-05 21:28:23,534 - Epoch: [98][ 740/ 1236] Overall Loss 0.319418 Objective Loss 0.319418 LR 0.001000 Time 0.021498 +2023-10-05 21:28:23,742 - Epoch: [98][ 750/ 1236] Overall Loss 0.319591 Objective Loss 0.319591 LR 0.001000 Time 0.021488 +2023-10-05 21:28:23,952 - Epoch: [98][ 760/ 1236] Overall Loss 0.319641 Objective Loss 0.319641 LR 0.001000 Time 0.021482 +2023-10-05 21:28:24,160 - Epoch: [98][ 770/ 1236] Overall Loss 0.319518 Objective Loss 0.319518 LR 0.001000 Time 0.021472 +2023-10-05 21:28:24,371 - Epoch: [98][ 780/ 1236] Overall Loss 0.319471 Objective Loss 0.319471 LR 0.001000 Time 0.021467 +2023-10-05 21:28:24,577 - Epoch: [98][ 790/ 1236] Overall Loss 0.319492 Objective Loss 0.319492 LR 0.001000 Time 0.021456 +2023-10-05 21:28:24,788 - Epoch: [98][ 800/ 1236] Overall Loss 0.319955 Objective Loss 0.319955 LR 0.001000 Time 0.021451 +2023-10-05 21:28:24,996 - Epoch: [98][ 810/ 1236] Overall Loss 0.320054 Objective Loss 0.320054 LR 0.001000 Time 0.021442 +2023-10-05 21:28:25,206 - Epoch: [98][ 820/ 1236] Overall Loss 0.320104 Objective Loss 0.320104 LR 0.001000 Time 0.021437 +2023-10-05 21:28:25,414 - Epoch: [98][ 830/ 1236] Overall Loss 0.319951 Objective Loss 0.319951 LR 0.001000 Time 0.021428 +2023-10-05 21:28:25,624 - Epoch: [98][ 840/ 1236] Overall Loss 0.320474 Objective Loss 0.320474 LR 0.001000 Time 0.021423 +2023-10-05 21:28:25,831 - Epoch: [98][ 850/ 1236] Overall Loss 0.320912 Objective Loss 0.320912 LR 0.001000 Time 0.021415 +2023-10-05 21:28:26,042 - Epoch: [98][ 860/ 1236] Overall Loss 0.320987 Objective Loss 0.320987 LR 0.001000 Time 0.021411 +2023-10-05 21:28:26,249 - Epoch: [98][ 870/ 1236] Overall Loss 0.320504 Objective Loss 0.320504 LR 0.001000 Time 0.021402 +2023-10-05 21:28:26,460 - Epoch: [98][ 880/ 1236] Overall Loss 0.320360 Objective Loss 0.320360 LR 0.001000 Time 0.021398 +2023-10-05 21:28:26,667 - Epoch: [98][ 890/ 1236] Overall Loss 0.320943 Objective Loss 0.320943 LR 0.001000 Time 0.021390 +2023-10-05 21:28:26,878 - Epoch: [98][ 900/ 1236] Overall Loss 0.321242 Objective Loss 0.321242 LR 0.001000 Time 0.021386 +2023-10-05 21:28:27,085 - Epoch: [98][ 910/ 1236] Overall Loss 0.321098 Objective Loss 0.321098 LR 0.001000 Time 0.021379 +2023-10-05 21:28:27,296 - Epoch: [98][ 920/ 1236] Overall Loss 0.321071 Objective Loss 0.321071 LR 0.001000 Time 0.021375 +2023-10-05 21:28:27,503 - Epoch: [98][ 930/ 1236] Overall Loss 0.321093 Objective Loss 0.321093 LR 0.001000 Time 0.021368 +2023-10-05 21:28:27,715 - Epoch: [98][ 940/ 1236] Overall Loss 0.321602 Objective Loss 0.321602 LR 0.001000 Time 0.021365 +2023-10-05 21:28:27,922 - Epoch: [98][ 950/ 1236] Overall Loss 0.321737 Objective Loss 0.321737 LR 0.001000 Time 0.021358 +2023-10-05 21:28:28,133 - Epoch: [98][ 960/ 1236] Overall Loss 0.322058 Objective Loss 0.322058 LR 0.001000 Time 0.021355 +2023-10-05 21:28:28,340 - Epoch: [98][ 970/ 1236] Overall Loss 0.321580 Objective Loss 0.321580 LR 0.001000 Time 0.021348 +2023-10-05 21:28:28,552 - Epoch: [98][ 980/ 1236] Overall Loss 0.321754 Objective Loss 0.321754 LR 0.001000 Time 0.021346 +2023-10-05 21:28:28,759 - Epoch: [98][ 990/ 1236] Overall Loss 0.321858 Objective Loss 0.321858 LR 0.001000 Time 0.021339 +2023-10-05 21:28:28,970 - Epoch: [98][ 1000/ 1236] Overall Loss 0.321593 Objective Loss 0.321593 LR 0.001000 Time 0.021336 +2023-10-05 21:28:29,177 - Epoch: [98][ 1010/ 1236] Overall Loss 0.321520 Objective Loss 0.321520 LR 0.001000 Time 0.021329 +2023-10-05 21:28:29,388 - Epoch: [98][ 1020/ 1236] Overall Loss 0.321258 Objective Loss 0.321258 LR 0.001000 Time 0.021326 +2023-10-05 21:28:29,595 - Epoch: [98][ 1030/ 1236] Overall Loss 0.321325 Objective Loss 0.321325 LR 0.001000 Time 0.021320 +2023-10-05 21:28:29,806 - Epoch: [98][ 1040/ 1236] Overall Loss 0.321025 Objective Loss 0.321025 LR 0.001000 Time 0.021318 +2023-10-05 21:28:30,013 - Epoch: [98][ 1050/ 1236] Overall Loss 0.320862 Objective Loss 0.320862 LR 0.001000 Time 0.021312 +2023-10-05 21:28:30,223 - Epoch: [98][ 1060/ 1236] Overall Loss 0.321016 Objective Loss 0.321016 LR 0.001000 Time 0.021309 +2023-10-05 21:28:30,431 - Epoch: [98][ 1070/ 1236] Overall Loss 0.321447 Objective Loss 0.321447 LR 0.001000 Time 0.021303 +2023-10-05 21:28:30,642 - Epoch: [98][ 1080/ 1236] Overall Loss 0.321308 Objective Loss 0.321308 LR 0.001000 Time 0.021301 +2023-10-05 21:28:30,849 - Epoch: [98][ 1090/ 1236] Overall Loss 0.321237 Objective Loss 0.321237 LR 0.001000 Time 0.021295 +2023-10-05 21:28:31,060 - Epoch: [98][ 1100/ 1236] Overall Loss 0.321418 Objective Loss 0.321418 LR 0.001000 Time 0.021293 +2023-10-05 21:28:31,267 - Epoch: [98][ 1110/ 1236] Overall Loss 0.321134 Objective Loss 0.321134 LR 0.001000 Time 0.021287 +2023-10-05 21:28:31,478 - Epoch: [98][ 1120/ 1236] Overall Loss 0.321136 Objective Loss 0.321136 LR 0.001000 Time 0.021286 +2023-10-05 21:28:31,686 - Epoch: [98][ 1130/ 1236] Overall Loss 0.321210 Objective Loss 0.321210 LR 0.001000 Time 0.021281 +2023-10-05 21:28:31,897 - Epoch: [98][ 1140/ 1236] Overall Loss 0.321313 Objective Loss 0.321313 LR 0.001000 Time 0.021279 +2023-10-05 21:28:32,104 - Epoch: [98][ 1150/ 1236] Overall Loss 0.321896 Objective Loss 0.321896 LR 0.001000 Time 0.021274 +2023-10-05 21:28:32,315 - Epoch: [98][ 1160/ 1236] Overall Loss 0.321875 Objective Loss 0.321875 LR 0.001000 Time 0.021272 +2023-10-05 21:28:32,523 - Epoch: [98][ 1170/ 1236] Overall Loss 0.322050 Objective Loss 0.322050 LR 0.001000 Time 0.021268 +2023-10-05 21:28:32,734 - Epoch: [98][ 1180/ 1236] Overall Loss 0.322129 Objective Loss 0.322129 LR 0.001000 Time 0.021266 +2023-10-05 21:28:32,941 - Epoch: [98][ 1190/ 1236] Overall Loss 0.322230 Objective Loss 0.322230 LR 0.001000 Time 0.021261 +2023-10-05 21:28:33,152 - Epoch: [98][ 1200/ 1236] Overall Loss 0.322443 Objective Loss 0.322443 LR 0.001000 Time 0.021259 +2023-10-05 21:28:33,360 - Epoch: [98][ 1210/ 1236] Overall Loss 0.322604 Objective Loss 0.322604 LR 0.001000 Time 0.021255 +2023-10-05 21:28:33,572 - Epoch: [98][ 1220/ 1236] Overall Loss 0.322562 Objective Loss 0.322562 LR 0.001000 Time 0.021255 +2023-10-05 21:28:33,836 - Epoch: [98][ 1230/ 1236] Overall Loss 0.323019 Objective Loss 0.323019 LR 0.001000 Time 0.021296 +2023-10-05 21:28:33,954 - Epoch: [98][ 1236/ 1236] Overall Loss 0.323052 Objective Loss 0.323052 Top1 85.336049 Top5 97.963340 LR 0.001000 Time 0.021288 +2023-10-05 21:28:34,095 - --- validate (epoch=98)----------- +2023-10-05 21:28:34,096 - 29943 samples (256 per mini-batch) +2023-10-05 21:28:34,551 - Epoch: [98][ 10/ 117] Loss 0.369710 Top1 82.226562 Top5 97.343750 +2023-10-05 21:28:34,701 - Epoch: [98][ 20/ 117] Loss 0.376663 Top1 81.699219 Top5 97.343750 +2023-10-05 21:28:34,850 - Epoch: [98][ 30/ 117] Loss 0.376658 Top1 81.796875 Top5 97.382812 +2023-10-05 21:28:35,001 - Epoch: [98][ 40/ 117] Loss 0.383356 Top1 81.826172 Top5 97.373047 +2023-10-05 21:28:35,151 - Epoch: [98][ 50/ 117] Loss 0.380421 Top1 81.765625 Top5 97.367188 +2023-10-05 21:28:35,304 - Epoch: [98][ 60/ 117] Loss 0.373156 Top1 81.979167 Top5 97.363281 +2023-10-05 21:28:35,460 - Epoch: [98][ 70/ 117] Loss 0.370329 Top1 81.947545 Top5 97.388393 +2023-10-05 21:28:35,618 - Epoch: [98][ 80/ 117] Loss 0.368341 Top1 81.982422 Top5 97.358398 +2023-10-05 21:28:35,774 - Epoch: [98][ 90/ 117] Loss 0.366404 Top1 82.087674 Top5 97.391493 +2023-10-05 21:28:35,931 - Epoch: [98][ 100/ 117] Loss 0.363269 Top1 82.203125 Top5 97.386719 +2023-10-05 21:28:36,089 - Epoch: [98][ 110/ 117] Loss 0.367635 Top1 82.084517 Top5 97.379261 +2023-10-05 21:28:36,173 - Epoch: [98][ 117/ 117] Loss 0.369577 Top1 82.002471 Top5 97.354974 +2023-10-05 21:28:36,312 - ==> Top1: 82.002 Top5: 97.355 Loss: 0.370 + +2023-10-05 21:28:36,312 - ==> Confusion: +[[ 902 1 7 3 12 3 0 1 4 78 0 1 1 2 7 3 6 1 1 0 17] + [ 1 1018 2 0 9 27 2 25 2 2 4 1 0 0 1 0 9 0 15 5 8] + [ 6 0 942 10 2 0 27 16 0 3 8 5 8 1 0 3 3 0 9 4 9] + [ 5 0 18 925 1 4 2 2 4 4 10 0 5 2 40 5 3 9 31 1 18] + [ 20 7 1 0 964 6 0 2 1 14 1 5 1 1 8 4 7 1 1 0 6] + [ 5 44 0 1 3 974 1 28 5 0 2 9 5 15 6 2 1 0 1 3 11] + [ 0 6 33 0 0 1 1104 13 0 0 6 2 2 0 1 8 0 2 3 5 5] + [ 1 11 12 0 0 27 4 1082 1 4 4 5 7 2 2 2 0 1 41 4 8] + [ 14 2 0 0 2 3 1 1 950 48 19 1 3 13 18 4 0 2 6 1 1] + [ 82 0 4 0 5 5 0 0 21 953 0 2 0 23 7 8 0 2 0 0 7] + [ 0 5 10 9 4 3 3 5 16 3 944 3 0 16 7 1 1 2 8 1 12] + [ 1 1 0 0 1 20 0 3 0 0 1 924 31 6 0 3 3 20 0 12 9] + [ 1 1 2 5 1 5 2 0 6 0 2 33 947 4 5 9 9 17 4 1 14] + [ 3 3 2 1 6 17 0 0 16 22 5 5 1 1015 6 0 5 0 0 4 8] + [ 13 5 4 7 6 2 0 0 23 6 0 1 2 3 993 0 1 2 13 0 20] + [ 1 3 5 0 2 0 3 0 0 0 0 9 7 1 1 1057 18 12 1 7 7] + [ 1 15 1 0 8 8 0 1 1 0 0 4 0 1 4 7 1091 0 2 3 14] + [ 0 0 1 1 0 2 1 0 0 0 0 1 23 2 2 7 4 987 1 1 5] + [ 1 12 4 13 0 1 0 42 4 0 3 1 2 0 7 0 1 0 961 0 16] + [ 0 1 1 0 2 11 8 20 0 0 4 13 8 2 0 7 8 1 2 1053 11] + [ 129 230 139 48 97 193 63 155 136 113 165 146 331 313 159 55 214 90 182 179 4768]] + +2023-10-05 21:28:36,314 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:28:36,314 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:28:36,320 - + +2023-10-05 21:28:36,320 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:28:37,302 - Epoch: [99][ 10/ 1236] Overall Loss 0.334281 Objective Loss 0.334281 LR 0.001000 Time 0.098169 +2023-10-05 21:28:37,513 - Epoch: [99][ 20/ 1236] Overall Loss 0.342404 Objective Loss 0.342404 LR 0.001000 Time 0.059624 +2023-10-05 21:28:37,720 - Epoch: [99][ 30/ 1236] Overall Loss 0.328947 Objective Loss 0.328947 LR 0.001000 Time 0.046641 +2023-10-05 21:28:37,932 - Epoch: [99][ 40/ 1236] Overall Loss 0.329812 Objective Loss 0.329812 LR 0.001000 Time 0.040263 +2023-10-05 21:28:38,139 - Epoch: [99][ 50/ 1236] Overall Loss 0.325766 Objective Loss 0.325766 LR 0.001000 Time 0.036345 +2023-10-05 21:28:38,349 - Epoch: [99][ 60/ 1236] Overall Loss 0.330758 Objective Loss 0.330758 LR 0.001000 Time 0.033794 +2023-10-05 21:28:38,557 - Epoch: [99][ 70/ 1236] Overall Loss 0.333797 Objective Loss 0.333797 LR 0.001000 Time 0.031923 +2023-10-05 21:28:38,768 - Epoch: [99][ 80/ 1236] Overall Loss 0.333808 Objective Loss 0.333808 LR 0.001000 Time 0.030566 +2023-10-05 21:28:38,976 - Epoch: [99][ 90/ 1236] Overall Loss 0.330300 Objective Loss 0.330300 LR 0.001000 Time 0.029485 +2023-10-05 21:28:39,192 - Epoch: [99][ 100/ 1236] Overall Loss 0.329859 Objective Loss 0.329859 LR 0.001000 Time 0.028688 +2023-10-05 21:28:39,398 - Epoch: [99][ 110/ 1236] Overall Loss 0.327629 Objective Loss 0.327629 LR 0.001000 Time 0.027950 +2023-10-05 21:28:39,608 - Epoch: [99][ 120/ 1236] Overall Loss 0.327413 Objective Loss 0.327413 LR 0.001000 Time 0.027368 +2023-10-05 21:28:39,814 - Epoch: [99][ 130/ 1236] Overall Loss 0.327201 Objective Loss 0.327201 LR 0.001000 Time 0.026841 +2023-10-05 21:28:40,024 - Epoch: [99][ 140/ 1236] Overall Loss 0.324531 Objective Loss 0.324531 LR 0.001000 Time 0.026426 +2023-10-05 21:28:40,230 - Epoch: [99][ 150/ 1236] Overall Loss 0.324402 Objective Loss 0.324402 LR 0.001000 Time 0.026031 +2023-10-05 21:28:40,439 - Epoch: [99][ 160/ 1236] Overall Loss 0.321851 Objective Loss 0.321851 LR 0.001000 Time 0.025714 +2023-10-05 21:28:40,645 - Epoch: [99][ 170/ 1236] Overall Loss 0.320737 Objective Loss 0.320737 LR 0.001000 Time 0.025407 +2023-10-05 21:28:40,855 - Epoch: [99][ 180/ 1236] Overall Loss 0.320809 Objective Loss 0.320809 LR 0.001000 Time 0.025161 +2023-10-05 21:28:41,061 - Epoch: [99][ 190/ 1236] Overall Loss 0.320058 Objective Loss 0.320058 LR 0.001000 Time 0.024918 +2023-10-05 21:28:41,268 - Epoch: [99][ 200/ 1236] Overall Loss 0.319574 Objective Loss 0.319574 LR 0.001000 Time 0.024706 +2023-10-05 21:28:41,467 - Epoch: [99][ 210/ 1236] Overall Loss 0.319998 Objective Loss 0.319998 LR 0.001000 Time 0.024476 +2023-10-05 21:28:41,668 - Epoch: [99][ 220/ 1236] Overall Loss 0.320387 Objective Loss 0.320387 LR 0.001000 Time 0.024275 +2023-10-05 21:28:41,867 - Epoch: [99][ 230/ 1236] Overall Loss 0.319326 Objective Loss 0.319326 LR 0.001000 Time 0.024083 +2023-10-05 21:28:42,067 - Epoch: [99][ 240/ 1236] Overall Loss 0.318955 Objective Loss 0.318955 LR 0.001000 Time 0.023913 +2023-10-05 21:28:42,266 - Epoch: [99][ 250/ 1236] Overall Loss 0.320260 Objective Loss 0.320260 LR 0.001000 Time 0.023750 +2023-10-05 21:28:42,466 - Epoch: [99][ 260/ 1236] Overall Loss 0.320496 Objective Loss 0.320496 LR 0.001000 Time 0.023606 +2023-10-05 21:28:42,665 - Epoch: [99][ 270/ 1236] Overall Loss 0.321689 Objective Loss 0.321689 LR 0.001000 Time 0.023466 +2023-10-05 21:28:42,865 - Epoch: [99][ 280/ 1236] Overall Loss 0.322656 Objective Loss 0.322656 LR 0.001000 Time 0.023342 +2023-10-05 21:28:43,064 - Epoch: [99][ 290/ 1236] Overall Loss 0.321163 Objective Loss 0.321163 LR 0.001000 Time 0.023222 +2023-10-05 21:28:43,265 - Epoch: [99][ 300/ 1236] Overall Loss 0.321260 Objective Loss 0.321260 LR 0.001000 Time 0.023116 +2023-10-05 21:28:43,464 - Epoch: [99][ 310/ 1236] Overall Loss 0.321247 Objective Loss 0.321247 LR 0.001000 Time 0.023012 +2023-10-05 21:28:43,665 - Epoch: [99][ 320/ 1236] Overall Loss 0.320546 Objective Loss 0.320546 LR 0.001000 Time 0.022920 +2023-10-05 21:28:43,864 - Epoch: [99][ 330/ 1236] Overall Loss 0.320129 Objective Loss 0.320129 LR 0.001000 Time 0.022828 +2023-10-05 21:28:44,065 - Epoch: [99][ 340/ 1236] Overall Loss 0.320448 Objective Loss 0.320448 LR 0.001000 Time 0.022746 +2023-10-05 21:28:44,264 - Epoch: [99][ 350/ 1236] Overall Loss 0.320075 Objective Loss 0.320075 LR 0.001000 Time 0.022664 +2023-10-05 21:28:44,465 - Epoch: [99][ 360/ 1236] Overall Loss 0.320976 Objective Loss 0.320976 LR 0.001000 Time 0.022591 +2023-10-05 21:28:44,664 - Epoch: [99][ 370/ 1236] Overall Loss 0.321007 Objective Loss 0.321007 LR 0.001000 Time 0.022519 +2023-10-05 21:28:44,866 - Epoch: [99][ 380/ 1236] Overall Loss 0.321171 Objective Loss 0.321171 LR 0.001000 Time 0.022457 +2023-10-05 21:28:45,067 - Epoch: [99][ 390/ 1236] Overall Loss 0.321038 Objective Loss 0.321038 LR 0.001000 Time 0.022394 +2023-10-05 21:28:45,268 - Epoch: [99][ 400/ 1236] Overall Loss 0.321122 Objective Loss 0.321122 LR 0.001000 Time 0.022338 +2023-10-05 21:28:45,468 - Epoch: [99][ 410/ 1236] Overall Loss 0.321672 Objective Loss 0.321672 LR 0.001000 Time 0.022279 +2023-10-05 21:28:45,670 - Epoch: [99][ 420/ 1236] Overall Loss 0.321661 Objective Loss 0.321661 LR 0.001000 Time 0.022229 +2023-10-05 21:28:45,871 - Epoch: [99][ 430/ 1236] Overall Loss 0.321805 Objective Loss 0.321805 LR 0.001000 Time 0.022178 +2023-10-05 21:28:46,073 - Epoch: [99][ 440/ 1236] Overall Loss 0.321535 Objective Loss 0.321535 LR 0.001000 Time 0.022133 +2023-10-05 21:28:46,273 - Epoch: [99][ 450/ 1236] Overall Loss 0.322046 Objective Loss 0.322046 LR 0.001000 Time 0.022085 +2023-10-05 21:28:46,475 - Epoch: [99][ 460/ 1236] Overall Loss 0.322299 Objective Loss 0.322299 LR 0.001000 Time 0.022043 +2023-10-05 21:28:46,675 - Epoch: [99][ 470/ 1236] Overall Loss 0.322120 Objective Loss 0.322120 LR 0.001000 Time 0.021998 +2023-10-05 21:28:46,877 - Epoch: [99][ 480/ 1236] Overall Loss 0.323112 Objective Loss 0.323112 LR 0.001000 Time 0.021960 +2023-10-05 21:28:47,077 - Epoch: [99][ 490/ 1236] Overall Loss 0.322645 Objective Loss 0.322645 LR 0.001000 Time 0.021919 +2023-10-05 21:28:47,279 - Epoch: [99][ 500/ 1236] Overall Loss 0.322200 Objective Loss 0.322200 LR 0.001000 Time 0.021885 +2023-10-05 21:28:47,479 - Epoch: [99][ 510/ 1236] Overall Loss 0.323197 Objective Loss 0.323197 LR 0.001000 Time 0.021846 +2023-10-05 21:28:47,681 - Epoch: [99][ 520/ 1236] Overall Loss 0.322692 Objective Loss 0.322692 LR 0.001000 Time 0.021815 +2023-10-05 21:28:47,881 - Epoch: [99][ 530/ 1236] Overall Loss 0.322681 Objective Loss 0.322681 LR 0.001000 Time 0.021779 +2023-10-05 21:28:48,083 - Epoch: [99][ 540/ 1236] Overall Loss 0.322843 Objective Loss 0.322843 LR 0.001000 Time 0.021749 +2023-10-05 21:28:48,283 - Epoch: [99][ 550/ 1236] Overall Loss 0.323109 Objective Loss 0.323109 LR 0.001000 Time 0.021717 +2023-10-05 21:28:48,485 - Epoch: [99][ 560/ 1236] Overall Loss 0.323000 Objective Loss 0.323000 LR 0.001000 Time 0.021690 +2023-10-05 21:28:48,685 - Epoch: [99][ 570/ 1236] Overall Loss 0.323395 Objective Loss 0.323395 LR 0.001000 Time 0.021659 +2023-10-05 21:28:48,887 - Epoch: [99][ 580/ 1236] Overall Loss 0.323761 Objective Loss 0.323761 LR 0.001000 Time 0.021633 +2023-10-05 21:28:49,088 - Epoch: [99][ 590/ 1236] Overall Loss 0.323878 Objective Loss 0.323878 LR 0.001000 Time 0.021606 +2023-10-05 21:28:49,290 - Epoch: [99][ 600/ 1236] Overall Loss 0.324184 Objective Loss 0.324184 LR 0.001000 Time 0.021582 +2023-10-05 21:28:49,490 - Epoch: [99][ 610/ 1236] Overall Loss 0.324808 Objective Loss 0.324808 LR 0.001000 Time 0.021555 +2023-10-05 21:28:49,692 - Epoch: [99][ 620/ 1236] Overall Loss 0.325167 Objective Loss 0.325167 LR 0.001000 Time 0.021533 +2023-10-05 21:28:49,892 - Epoch: [99][ 630/ 1236] Overall Loss 0.325107 Objective Loss 0.325107 LR 0.001000 Time 0.021508 +2023-10-05 21:28:50,094 - Epoch: [99][ 640/ 1236] Overall Loss 0.324438 Objective Loss 0.324438 LR 0.001000 Time 0.021487 +2023-10-05 21:28:50,294 - Epoch: [99][ 650/ 1236] Overall Loss 0.324082 Objective Loss 0.324082 LR 0.001000 Time 0.021464 +2023-10-05 21:28:50,496 - Epoch: [99][ 660/ 1236] Overall Loss 0.323286 Objective Loss 0.323286 LR 0.001000 Time 0.021444 +2023-10-05 21:28:50,696 - Epoch: [99][ 670/ 1236] Overall Loss 0.323182 Objective Loss 0.323182 LR 0.001000 Time 0.021422 +2023-10-05 21:28:50,898 - Epoch: [99][ 680/ 1236] Overall Loss 0.322659 Objective Loss 0.322659 LR 0.001000 Time 0.021404 +2023-10-05 21:28:51,098 - Epoch: [99][ 690/ 1236] Overall Loss 0.322787 Objective Loss 0.322787 LR 0.001000 Time 0.021383 +2023-10-05 21:28:51,300 - Epoch: [99][ 700/ 1236] Overall Loss 0.323203 Objective Loss 0.323203 LR 0.001000 Time 0.021366 +2023-10-05 21:28:51,501 - Epoch: [99][ 710/ 1236] Overall Loss 0.323416 Objective Loss 0.323416 LR 0.001000 Time 0.021347 +2023-10-05 21:28:51,703 - Epoch: [99][ 720/ 1236] Overall Loss 0.323399 Objective Loss 0.323399 LR 0.001000 Time 0.021330 +2023-10-05 21:28:51,903 - Epoch: [99][ 730/ 1236] Overall Loss 0.323623 Objective Loss 0.323623 LR 0.001000 Time 0.021312 +2023-10-05 21:28:52,105 - Epoch: [99][ 740/ 1236] Overall Loss 0.324319 Objective Loss 0.324319 LR 0.001000 Time 0.021297 +2023-10-05 21:28:52,305 - Epoch: [99][ 750/ 1236] Overall Loss 0.324063 Objective Loss 0.324063 LR 0.001000 Time 0.021279 +2023-10-05 21:28:52,507 - Epoch: [99][ 760/ 1236] Overall Loss 0.323845 Objective Loss 0.323845 LR 0.001000 Time 0.021264 +2023-10-05 21:28:52,707 - Epoch: [99][ 770/ 1236] Overall Loss 0.323743 Objective Loss 0.323743 LR 0.001000 Time 0.021248 +2023-10-05 21:28:52,910 - Epoch: [99][ 780/ 1236] Overall Loss 0.324063 Objective Loss 0.324063 LR 0.001000 Time 0.021234 +2023-10-05 21:28:53,110 - Epoch: [99][ 790/ 1236] Overall Loss 0.324231 Objective Loss 0.324231 LR 0.001000 Time 0.021218 +2023-10-05 21:28:53,312 - Epoch: [99][ 800/ 1236] Overall Loss 0.324539 Objective Loss 0.324539 LR 0.001000 Time 0.021206 +2023-10-05 21:28:53,512 - Epoch: [99][ 810/ 1236] Overall Loss 0.324805 Objective Loss 0.324805 LR 0.001000 Time 0.021190 +2023-10-05 21:28:53,714 - Epoch: [99][ 820/ 1236] Overall Loss 0.324929 Objective Loss 0.324929 LR 0.001000 Time 0.021178 +2023-10-05 21:28:53,914 - Epoch: [99][ 830/ 1236] Overall Loss 0.325518 Objective Loss 0.325518 LR 0.001000 Time 0.021163 +2023-10-05 21:28:54,116 - Epoch: [99][ 840/ 1236] Overall Loss 0.325556 Objective Loss 0.325556 LR 0.001000 Time 0.021151 +2023-10-05 21:28:54,316 - Epoch: [99][ 850/ 1236] Overall Loss 0.325830 Objective Loss 0.325830 LR 0.001000 Time 0.021137 +2023-10-05 21:28:54,518 - Epoch: [99][ 860/ 1236] Overall Loss 0.325638 Objective Loss 0.325638 LR 0.001000 Time 0.021126 +2023-10-05 21:28:54,718 - Epoch: [99][ 870/ 1236] Overall Loss 0.325747 Objective Loss 0.325747 LR 0.001000 Time 0.021113 +2023-10-05 21:28:54,921 - Epoch: [99][ 880/ 1236] Overall Loss 0.326042 Objective Loss 0.326042 LR 0.001000 Time 0.021102 +2023-10-05 21:28:55,121 - Epoch: [99][ 890/ 1236] Overall Loss 0.325876 Objective Loss 0.325876 LR 0.001000 Time 0.021090 +2023-10-05 21:28:55,323 - Epoch: [99][ 900/ 1236] Overall Loss 0.326178 Objective Loss 0.326178 LR 0.001000 Time 0.021080 +2023-10-05 21:28:55,523 - Epoch: [99][ 910/ 1236] Overall Loss 0.326253 Objective Loss 0.326253 LR 0.001000 Time 0.021067 +2023-10-05 21:28:55,725 - Epoch: [99][ 920/ 1236] Overall Loss 0.326433 Objective Loss 0.326433 LR 0.001000 Time 0.021058 +2023-10-05 21:28:55,926 - Epoch: [99][ 930/ 1236] Overall Loss 0.326298 Objective Loss 0.326298 LR 0.001000 Time 0.021046 +2023-10-05 21:28:56,128 - Epoch: [99][ 940/ 1236] Overall Loss 0.326320 Objective Loss 0.326320 LR 0.001000 Time 0.021037 +2023-10-05 21:28:56,328 - Epoch: [99][ 950/ 1236] Overall Loss 0.325926 Objective Loss 0.325926 LR 0.001000 Time 0.021026 +2023-10-05 21:28:56,530 - Epoch: [99][ 960/ 1236] Overall Loss 0.325678 Objective Loss 0.325678 LR 0.001000 Time 0.021017 +2023-10-05 21:28:56,730 - Epoch: [99][ 970/ 1236] Overall Loss 0.325668 Objective Loss 0.325668 LR 0.001000 Time 0.021006 +2023-10-05 21:28:56,932 - Epoch: [99][ 980/ 1236] Overall Loss 0.325531 Objective Loss 0.325531 LR 0.001000 Time 0.020998 +2023-10-05 21:28:57,133 - Epoch: [99][ 990/ 1236] Overall Loss 0.325810 Objective Loss 0.325810 LR 0.001000 Time 0.020988 +2023-10-05 21:28:57,335 - Epoch: [99][ 1000/ 1236] Overall Loss 0.325853 Objective Loss 0.325853 LR 0.001000 Time 0.020979 +2023-10-05 21:28:57,535 - Epoch: [99][ 1010/ 1236] Overall Loss 0.325900 Objective Loss 0.325900 LR 0.001000 Time 0.020970 +2023-10-05 21:28:57,737 - Epoch: [99][ 1020/ 1236] Overall Loss 0.325751 Objective Loss 0.325751 LR 0.001000 Time 0.020961 +2023-10-05 21:28:57,937 - Epoch: [99][ 1030/ 1236] Overall Loss 0.325891 Objective Loss 0.325891 LR 0.001000 Time 0.020952 +2023-10-05 21:28:58,139 - Epoch: [99][ 1040/ 1236] Overall Loss 0.325806 Objective Loss 0.325806 LR 0.001000 Time 0.020945 +2023-10-05 21:28:58,339 - Epoch: [99][ 1050/ 1236] Overall Loss 0.325854 Objective Loss 0.325854 LR 0.001000 Time 0.020936 +2023-10-05 21:28:58,541 - Epoch: [99][ 1060/ 1236] Overall Loss 0.325746 Objective Loss 0.325746 LR 0.001000 Time 0.020928 +2023-10-05 21:28:58,741 - Epoch: [99][ 1070/ 1236] Overall Loss 0.325762 Objective Loss 0.325762 LR 0.001000 Time 0.020919 +2023-10-05 21:28:58,943 - Epoch: [99][ 1080/ 1236] Overall Loss 0.325689 Objective Loss 0.325689 LR 0.001000 Time 0.020912 +2023-10-05 21:28:59,144 - Epoch: [99][ 1090/ 1236] Overall Loss 0.325663 Objective Loss 0.325663 LR 0.001000 Time 0.020904 +2023-10-05 21:28:59,346 - Epoch: [99][ 1100/ 1236] Overall Loss 0.325793 Objective Loss 0.325793 LR 0.001000 Time 0.020897 +2023-10-05 21:28:59,546 - Epoch: [99][ 1110/ 1236] Overall Loss 0.325752 Objective Loss 0.325752 LR 0.001000 Time 0.020889 +2023-10-05 21:28:59,748 - Epoch: [99][ 1120/ 1236] Overall Loss 0.325655 Objective Loss 0.325655 LR 0.001000 Time 0.020883 +2023-10-05 21:28:59,948 - Epoch: [99][ 1130/ 1236] Overall Loss 0.325293 Objective Loss 0.325293 LR 0.001000 Time 0.020875 +2023-10-05 21:29:00,150 - Epoch: [99][ 1140/ 1236] Overall Loss 0.325309 Objective Loss 0.325309 LR 0.001000 Time 0.020868 +2023-10-05 21:29:00,350 - Epoch: [99][ 1150/ 1236] Overall Loss 0.325257 Objective Loss 0.325257 LR 0.001000 Time 0.020861 +2023-10-05 21:29:00,553 - Epoch: [99][ 1160/ 1236] Overall Loss 0.325469 Objective Loss 0.325469 LR 0.001000 Time 0.020855 +2023-10-05 21:29:00,754 - Epoch: [99][ 1170/ 1236] Overall Loss 0.325388 Objective Loss 0.325388 LR 0.001000 Time 0.020848 +2023-10-05 21:29:00,956 - Epoch: [99][ 1180/ 1236] Overall Loss 0.325513 Objective Loss 0.325513 LR 0.001000 Time 0.020843 +2023-10-05 21:29:01,156 - Epoch: [99][ 1190/ 1236] Overall Loss 0.325628 Objective Loss 0.325628 LR 0.001000 Time 0.020835 +2023-10-05 21:29:01,358 - Epoch: [99][ 1200/ 1236] Overall Loss 0.325602 Objective Loss 0.325602 LR 0.001000 Time 0.020830 +2023-10-05 21:29:01,558 - Epoch: [99][ 1210/ 1236] Overall Loss 0.325921 Objective Loss 0.325921 LR 0.001000 Time 0.020823 +2023-10-05 21:29:01,760 - Epoch: [99][ 1220/ 1236] Overall Loss 0.325996 Objective Loss 0.325996 LR 0.001000 Time 0.020817 +2023-10-05 21:29:02,015 - Epoch: [99][ 1230/ 1236] Overall Loss 0.325805 Objective Loss 0.325805 LR 0.001000 Time 0.020855 +2023-10-05 21:29:02,132 - Epoch: [99][ 1236/ 1236] Overall Loss 0.325601 Objective Loss 0.325601 Top1 84.114053 Top5 97.963340 LR 0.001000 Time 0.020848 +2023-10-05 21:29:02,269 - --- validate (epoch=99)----------- +2023-10-05 21:29:02,269 - 29943 samples (256 per mini-batch) +2023-10-05 21:29:02,720 - Epoch: [99][ 10/ 117] Loss 0.398643 Top1 81.445312 Top5 97.304688 +2023-10-05 21:29:02,870 - Epoch: [99][ 20/ 117] Loss 0.389879 Top1 81.269531 Top5 97.265625 +2023-10-05 21:29:03,022 - Epoch: [99][ 30/ 117] Loss 0.382430 Top1 81.119792 Top5 97.226562 +2023-10-05 21:29:03,182 - Epoch: [99][ 40/ 117] Loss 0.368194 Top1 81.582031 Top5 97.470703 +2023-10-05 21:29:03,338 - Epoch: [99][ 50/ 117] Loss 0.367101 Top1 81.554688 Top5 97.437500 +2023-10-05 21:29:03,499 - Epoch: [99][ 60/ 117] Loss 0.363617 Top1 81.627604 Top5 97.402344 +2023-10-05 21:29:03,659 - Epoch: [99][ 70/ 117] Loss 0.367735 Top1 81.428571 Top5 97.343750 +2023-10-05 21:29:03,819 - Epoch: [99][ 80/ 117] Loss 0.365131 Top1 81.298828 Top5 97.290039 +2023-10-05 21:29:03,976 - Epoch: [99][ 90/ 117] Loss 0.364671 Top1 81.354167 Top5 97.326389 +2023-10-05 21:29:04,137 - Epoch: [99][ 100/ 117] Loss 0.362818 Top1 81.320312 Top5 97.328125 +2023-10-05 21:29:04,301 - Epoch: [99][ 110/ 117] Loss 0.363353 Top1 81.313920 Top5 97.336648 +2023-10-05 21:29:04,386 - Epoch: [99][ 117/ 117] Loss 0.361476 Top1 81.404669 Top5 97.358314 +2023-10-05 21:29:04,525 - ==> Top1: 81.405 Top5: 97.358 Loss: 0.361 + +2023-10-05 21:29:04,526 - ==> Confusion: +[[ 927 1 4 2 9 4 0 1 4 70 0 0 0 2 7 2 6 1 0 0 10] + [ 1 1035 2 1 13 24 2 26 0 0 1 0 0 0 2 5 5 0 7 1 6] + [ 8 2 951 11 1 2 24 8 0 1 5 3 10 1 3 5 0 1 6 5 9] + [ 4 0 24 940 2 4 0 0 6 2 3 0 4 4 32 5 1 7 34 3 14] + [ 31 8 0 0 967 2 0 0 0 9 0 2 0 2 10 5 5 1 2 2 4] + [ 8 43 0 1 3 981 0 20 2 1 2 13 2 11 4 1 5 0 5 4 10] + [ 0 6 34 0 0 2 1102 13 1 0 5 4 1 0 0 7 1 0 4 4 7] + [ 6 24 13 0 2 31 7 1037 0 3 3 10 1 1 1 6 0 1 55 7 10] + [ 19 2 1 0 1 2 1 0 924 70 12 4 1 13 24 6 2 2 2 1 2] + [ 109 0 2 0 4 4 2 0 25 923 0 0 0 22 8 7 1 1 0 4 7] + [ 3 6 16 10 2 1 7 3 20 1 927 4 1 20 8 3 0 0 10 1 10] + [ 1 0 0 0 1 14 0 1 2 1 0 961 24 4 0 2 0 13 0 8 3] + [ 1 2 2 7 0 3 1 0 2 1 0 38 964 0 2 5 1 23 3 2 11] + [ 1 0 0 1 1 21 0 3 13 16 9 4 4 1022 6 2 0 1 0 6 9] + [ 18 2 3 15 6 0 0 0 22 6 1 0 1 1 1000 0 2 2 11 0 11] + [ 1 0 5 1 5 0 2 0 0 0 0 15 8 1 2 1037 14 23 2 7 11] + [ 2 15 4 0 9 4 0 2 1 0 1 9 2 2 4 6 1085 0 2 5 8] + [ 0 1 2 3 0 2 1 0 0 0 0 6 19 0 1 6 0 990 1 2 4] + [ 2 13 8 12 2 0 0 27 2 0 2 2 6 2 14 0 3 1 964 1 7] + [ 0 3 7 2 2 7 12 12 0 0 0 22 4 2 0 6 5 1 0 1057 10] + [ 164 215 179 90 112 174 49 118 93 127 145 171 405 315 210 52 165 93 211 236 4581]] + +2023-10-05 21:29:04,527 - ==> Best [Top1: 82.714 Top5: 97.702 Sparsity:0.00 Params: 148928 on epoch: 72] +2023-10-05 21:29:04,527 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:29:04,534 - + +2023-10-05 21:29:04,534 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:29:05,509 - Epoch: [100][ 10/ 1236] Overall Loss 0.325309 Objective Loss 0.325309 LR 0.000500 Time 0.097508 +2023-10-05 21:29:05,712 - Epoch: [100][ 20/ 1236] Overall Loss 0.312768 Objective Loss 0.312768 LR 0.000500 Time 0.058893 +2023-10-05 21:29:05,913 - Epoch: [100][ 30/ 1236] Overall Loss 0.306899 Objective Loss 0.306899 LR 0.000500 Time 0.045931 +2023-10-05 21:29:06,116 - Epoch: [100][ 40/ 1236] Overall Loss 0.306467 Objective Loss 0.306467 LR 0.000500 Time 0.039523 +2023-10-05 21:29:06,317 - Epoch: [100][ 50/ 1236] Overall Loss 0.309214 Objective Loss 0.309214 LR 0.000500 Time 0.035625 +2023-10-05 21:29:06,520 - Epoch: [100][ 60/ 1236] Overall Loss 0.302902 Objective Loss 0.302902 LR 0.000500 Time 0.033072 +2023-10-05 21:29:06,721 - Epoch: [100][ 70/ 1236] Overall Loss 0.305524 Objective Loss 0.305524 LR 0.000500 Time 0.031209 +2023-10-05 21:29:06,924 - Epoch: [100][ 80/ 1236] Overall Loss 0.304324 Objective Loss 0.304324 LR 0.000500 Time 0.029848 +2023-10-05 21:29:07,125 - Epoch: [100][ 90/ 1236] Overall Loss 0.304878 Objective Loss 0.304878 LR 0.000500 Time 0.028755 +2023-10-05 21:29:07,328 - Epoch: [100][ 100/ 1236] Overall Loss 0.304162 Objective Loss 0.304162 LR 0.000500 Time 0.027912 +2023-10-05 21:29:07,529 - Epoch: [100][ 110/ 1236] Overall Loss 0.306886 Objective Loss 0.306886 LR 0.000500 Time 0.027193 +2023-10-05 21:29:07,733 - Epoch: [100][ 120/ 1236] Overall Loss 0.308692 Objective Loss 0.308692 LR 0.000500 Time 0.026627 +2023-10-05 21:29:07,940 - Epoch: [100][ 130/ 1236] Overall Loss 0.306746 Objective Loss 0.306746 LR 0.000500 Time 0.026161 +2023-10-05 21:29:08,146 - Epoch: [100][ 140/ 1236] Overall Loss 0.308027 Objective Loss 0.308027 LR 0.000500 Time 0.025761 +2023-10-05 21:29:08,351 - Epoch: [100][ 150/ 1236] Overall Loss 0.308879 Objective Loss 0.308879 LR 0.000500 Time 0.025411 +2023-10-05 21:29:08,557 - Epoch: [100][ 160/ 1236] Overall Loss 0.306927 Objective Loss 0.306927 LR 0.000500 Time 0.025107 +2023-10-05 21:29:08,763 - Epoch: [100][ 170/ 1236] Overall Loss 0.304846 Objective Loss 0.304846 LR 0.000500 Time 0.024837 +2023-10-05 21:29:08,969 - Epoch: [100][ 180/ 1236] Overall Loss 0.302694 Objective Loss 0.302694 LR 0.000500 Time 0.024598 +2023-10-05 21:29:09,175 - Epoch: [100][ 190/ 1236] Overall Loss 0.300177 Objective Loss 0.300177 LR 0.000500 Time 0.024383 +2023-10-05 21:29:09,381 - Epoch: [100][ 200/ 1236] Overall Loss 0.300625 Objective Loss 0.300625 LR 0.000500 Time 0.024190 +2023-10-05 21:29:09,586 - Epoch: [100][ 210/ 1236] Overall Loss 0.299869 Objective Loss 0.299869 LR 0.000500 Time 0.024014 +2023-10-05 21:29:09,792 - Epoch: [100][ 220/ 1236] Overall Loss 0.299556 Objective Loss 0.299556 LR 0.000500 Time 0.023857 +2023-10-05 21:29:09,998 - Epoch: [100][ 230/ 1236] Overall Loss 0.299241 Objective Loss 0.299241 LR 0.000500 Time 0.023711 +2023-10-05 21:29:10,204 - Epoch: [100][ 240/ 1236] Overall Loss 0.299154 Objective Loss 0.299154 LR 0.000500 Time 0.023578 +2023-10-05 21:29:10,408 - Epoch: [100][ 250/ 1236] Overall Loss 0.298851 Objective Loss 0.298851 LR 0.000500 Time 0.023452 +2023-10-05 21:29:10,615 - Epoch: [100][ 260/ 1236] Overall Loss 0.299169 Objective Loss 0.299169 LR 0.000500 Time 0.023343 +2023-10-05 21:29:10,821 - Epoch: [100][ 270/ 1236] Overall Loss 0.298556 Objective Loss 0.298556 LR 0.000500 Time 0.023238 +2023-10-05 21:29:11,026 - Epoch: [100][ 280/ 1236] Overall Loss 0.298433 Objective Loss 0.298433 LR 0.000500 Time 0.023141 +2023-10-05 21:29:11,232 - Epoch: [100][ 290/ 1236] Overall Loss 0.297759 Objective Loss 0.297759 LR 0.000500 Time 0.023050 +2023-10-05 21:29:11,438 - Epoch: [100][ 300/ 1236] Overall Loss 0.297246 Objective Loss 0.297246 LR 0.000500 Time 0.022967 +2023-10-05 21:29:11,644 - Epoch: [100][ 310/ 1236] Overall Loss 0.297784 Objective Loss 0.297784 LR 0.000500 Time 0.022888 +2023-10-05 21:29:11,850 - Epoch: [100][ 320/ 1236] Overall Loss 0.296615 Objective Loss 0.296615 LR 0.000500 Time 0.022815 +2023-10-05 21:29:12,057 - Epoch: [100][ 330/ 1236] Overall Loss 0.297107 Objective Loss 0.297107 LR 0.000500 Time 0.022747 +2023-10-05 21:29:12,263 - Epoch: [100][ 340/ 1236] Overall Loss 0.297171 Objective Loss 0.297171 LR 0.000500 Time 0.022683 +2023-10-05 21:29:12,469 - Epoch: [100][ 350/ 1236] Overall Loss 0.296877 Objective Loss 0.296877 LR 0.000500 Time 0.022621 +2023-10-05 21:29:12,675 - Epoch: [100][ 360/ 1236] Overall Loss 0.296900 Objective Loss 0.296900 LR 0.000500 Time 0.022563 +2023-10-05 21:29:12,881 - Epoch: [100][ 370/ 1236] Overall Loss 0.296632 Objective Loss 0.296632 LR 0.000500 Time 0.022508 +2023-10-05 21:29:13,086 - Epoch: [100][ 380/ 1236] Overall Loss 0.296802 Objective Loss 0.296802 LR 0.000500 Time 0.022456 +2023-10-05 21:29:13,293 - Epoch: [100][ 390/ 1236] Overall Loss 0.296778 Objective Loss 0.296778 LR 0.000500 Time 0.022409 +2023-10-05 21:29:13,506 - Epoch: [100][ 400/ 1236] Overall Loss 0.297307 Objective Loss 0.297307 LR 0.000500 Time 0.022380 +2023-10-05 21:29:13,713 - Epoch: [100][ 410/ 1236] Overall Loss 0.297592 Objective Loss 0.297592 LR 0.000500 Time 0.022338 +2023-10-05 21:29:13,923 - Epoch: [100][ 420/ 1236] Overall Loss 0.297386 Objective Loss 0.297386 LR 0.000500 Time 0.022307 +2023-10-05 21:29:14,128 - Epoch: [100][ 430/ 1236] Overall Loss 0.297324 Objective Loss 0.297324 LR 0.000500 Time 0.022264 +2023-10-05 21:29:14,329 - Epoch: [100][ 440/ 1236] Overall Loss 0.297202 Objective Loss 0.297202 LR 0.000500 Time 0.022212 +2023-10-05 21:29:14,527 - Epoch: [100][ 450/ 1236] Overall Loss 0.296903 Objective Loss 0.296903 LR 0.000500 Time 0.022159 +2023-10-05 21:29:14,727 - Epoch: [100][ 460/ 1236] Overall Loss 0.296680 Objective Loss 0.296680 LR 0.000500 Time 0.022111 +2023-10-05 21:29:14,925 - Epoch: [100][ 470/ 1236] Overall Loss 0.296836 Objective Loss 0.296836 LR 0.000500 Time 0.022062 +2023-10-05 21:29:15,126 - Epoch: [100][ 480/ 1236] Overall Loss 0.296747 Objective Loss 0.296747 LR 0.000500 Time 0.022020 +2023-10-05 21:29:15,324 - Epoch: [100][ 490/ 1236] Overall Loss 0.297010 Objective Loss 0.297010 LR 0.000500 Time 0.021974 +2023-10-05 21:29:15,525 - Epoch: [100][ 500/ 1236] Overall Loss 0.297524 Objective Loss 0.297524 LR 0.000500 Time 0.021935 +2023-10-05 21:29:15,723 - Epoch: [100][ 510/ 1236] Overall Loss 0.296935 Objective Loss 0.296935 LR 0.000500 Time 0.021893 +2023-10-05 21:29:15,923 - Epoch: [100][ 520/ 1236] Overall Loss 0.297567 Objective Loss 0.297567 LR 0.000500 Time 0.021856 +2023-10-05 21:29:16,122 - Epoch: [100][ 530/ 1236] Overall Loss 0.297271 Objective Loss 0.297271 LR 0.000500 Time 0.021817 +2023-10-05 21:29:16,322 - Epoch: [100][ 540/ 1236] Overall Loss 0.297126 Objective Loss 0.297126 LR 0.000500 Time 0.021783 +2023-10-05 21:29:16,520 - Epoch: [100][ 550/ 1236] Overall Loss 0.296327 Objective Loss 0.296327 LR 0.000500 Time 0.021747 +2023-10-05 21:29:16,721 - Epoch: [100][ 560/ 1236] Overall Loss 0.295455 Objective Loss 0.295455 LR 0.000500 Time 0.021717 +2023-10-05 21:29:16,919 - Epoch: [100][ 570/ 1236] Overall Loss 0.295790 Objective Loss 0.295790 LR 0.000500 Time 0.021683 +2023-10-05 21:29:17,120 - Epoch: [100][ 580/ 1236] Overall Loss 0.295552 Objective Loss 0.295552 LR 0.000500 Time 0.021655 +2023-10-05 21:29:17,318 - Epoch: [100][ 590/ 1236] Overall Loss 0.295106 Objective Loss 0.295106 LR 0.000500 Time 0.021623 +2023-10-05 21:29:17,519 - Epoch: [100][ 600/ 1236] Overall Loss 0.295342 Objective Loss 0.295342 LR 0.000500 Time 0.021596 +2023-10-05 21:29:17,717 - Epoch: [100][ 610/ 1236] Overall Loss 0.294593 Objective Loss 0.294593 LR 0.000500 Time 0.021567 +2023-10-05 21:29:17,918 - Epoch: [100][ 620/ 1236] Overall Loss 0.294742 Objective Loss 0.294742 LR 0.000500 Time 0.021542 +2023-10-05 21:29:18,116 - Epoch: [100][ 630/ 1236] Overall Loss 0.294714 Objective Loss 0.294714 LR 0.000500 Time 0.021515 +2023-10-05 21:29:18,317 - Epoch: [100][ 640/ 1236] Overall Loss 0.294345 Objective Loss 0.294345 LR 0.000500 Time 0.021492 +2023-10-05 21:29:18,515 - Epoch: [100][ 650/ 1236] Overall Loss 0.293822 Objective Loss 0.293822 LR 0.000500 Time 0.021466 +2023-10-05 21:29:18,716 - Epoch: [100][ 660/ 1236] Overall Loss 0.293432 Objective Loss 0.293432 LR 0.000500 Time 0.021444 +2023-10-05 21:29:18,913 - Epoch: [100][ 670/ 1236] Overall Loss 0.293399 Objective Loss 0.293399 LR 0.000500 Time 0.021418 +2023-10-05 21:29:19,114 - Epoch: [100][ 680/ 1236] Overall Loss 0.293028 Objective Loss 0.293028 LR 0.000500 Time 0.021397 +2023-10-05 21:29:19,312 - Epoch: [100][ 690/ 1236] Overall Loss 0.292954 Objective Loss 0.292954 LR 0.000500 Time 0.021374 +2023-10-05 21:29:19,513 - Epoch: [100][ 700/ 1236] Overall Loss 0.292965 Objective Loss 0.292965 LR 0.000500 Time 0.021355 +2023-10-05 21:29:19,711 - Epoch: [100][ 710/ 1236] Overall Loss 0.292813 Objective Loss 0.292813 LR 0.000500 Time 0.021333 +2023-10-05 21:29:19,912 - Epoch: [100][ 720/ 1236] Overall Loss 0.292752 Objective Loss 0.292752 LR 0.000500 Time 0.021315 +2023-10-05 21:29:20,110 - Epoch: [100][ 730/ 1236] Overall Loss 0.292461 Objective Loss 0.292461 LR 0.000500 Time 0.021294 +2023-10-05 21:29:20,311 - Epoch: [100][ 740/ 1236] Overall Loss 0.292518 Objective Loss 0.292518 LR 0.000500 Time 0.021277 +2023-10-05 21:29:20,509 - Epoch: [100][ 750/ 1236] Overall Loss 0.292331 Objective Loss 0.292331 LR 0.000500 Time 0.021257 +2023-10-05 21:29:20,710 - Epoch: [100][ 760/ 1236] Overall Loss 0.292472 Objective Loss 0.292472 LR 0.000500 Time 0.021241 +2023-10-05 21:29:20,908 - Epoch: [100][ 770/ 1236] Overall Loss 0.292157 Objective Loss 0.292157 LR 0.000500 Time 0.021222 +2023-10-05 21:29:21,108 - Epoch: [100][ 780/ 1236] Overall Loss 0.291921 Objective Loss 0.291921 LR 0.000500 Time 0.021207 +2023-10-05 21:29:21,307 - Epoch: [100][ 790/ 1236] Overall Loss 0.291962 Objective Loss 0.291962 LR 0.000500 Time 0.021189 +2023-10-05 21:29:21,508 - Epoch: [100][ 800/ 1236] Overall Loss 0.291570 Objective Loss 0.291570 LR 0.000500 Time 0.021175 +2023-10-05 21:29:21,706 - Epoch: [100][ 810/ 1236] Overall Loss 0.291676 Objective Loss 0.291676 LR 0.000500 Time 0.021158 +2023-10-05 21:29:21,907 - Epoch: [100][ 820/ 1236] Overall Loss 0.292266 Objective Loss 0.292266 LR 0.000500 Time 0.021144 +2023-10-05 21:29:22,106 - Epoch: [100][ 830/ 1236] Overall Loss 0.292196 Objective Loss 0.292196 LR 0.000500 Time 0.021128 +2023-10-05 21:29:22,306 - Epoch: [100][ 840/ 1236] Overall Loss 0.292310 Objective Loss 0.292310 LR 0.000500 Time 0.021115 +2023-10-05 21:29:22,504 - Epoch: [100][ 850/ 1236] Overall Loss 0.291883 Objective Loss 0.291883 LR 0.000500 Time 0.021100 +2023-10-05 21:29:22,705 - Epoch: [100][ 860/ 1236] Overall Loss 0.291401 Objective Loss 0.291401 LR 0.000500 Time 0.021087 +2023-10-05 21:29:22,903 - Epoch: [100][ 870/ 1236] Overall Loss 0.291616 Objective Loss 0.291616 LR 0.000500 Time 0.021072 +2023-10-05 21:29:23,104 - Epoch: [100][ 880/ 1236] Overall Loss 0.291764 Objective Loss 0.291764 LR 0.000500 Time 0.021060 +2023-10-05 21:29:23,302 - Epoch: [100][ 890/ 1236] Overall Loss 0.292156 Objective Loss 0.292156 LR 0.000500 Time 0.021046 +2023-10-05 21:29:23,502 - Epoch: [100][ 900/ 1236] Overall Loss 0.291776 Objective Loss 0.291776 LR 0.000500 Time 0.021034 +2023-10-05 21:29:23,700 - Epoch: [100][ 910/ 1236] Overall Loss 0.291595 Objective Loss 0.291595 LR 0.000500 Time 0.021020 +2023-10-05 21:29:23,901 - Epoch: [100][ 920/ 1236] Overall Loss 0.291799 Objective Loss 0.291799 LR 0.000500 Time 0.021009 +2023-10-05 21:29:24,102 - Epoch: [100][ 930/ 1236] Overall Loss 0.291055 Objective Loss 0.291055 LR 0.000500 Time 0.020999 +2023-10-05 21:29:24,303 - Epoch: [100][ 940/ 1236] Overall Loss 0.291012 Objective Loss 0.291012 LR 0.000500 Time 0.020990 +2023-10-05 21:29:24,504 - Epoch: [100][ 950/ 1236] Overall Loss 0.290692 Objective Loss 0.290692 LR 0.000500 Time 0.020980 +2023-10-05 21:29:24,706 - Epoch: [100][ 960/ 1236] Overall Loss 0.290913 Objective Loss 0.290913 LR 0.000500 Time 0.020971 +2023-10-05 21:29:24,907 - Epoch: [100][ 970/ 1236] Overall Loss 0.290991 Objective Loss 0.290991 LR 0.000500 Time 0.020962 +2023-10-05 21:29:25,110 - Epoch: [100][ 980/ 1236] Overall Loss 0.291132 Objective Loss 0.291132 LR 0.000500 Time 0.020955 +2023-10-05 21:29:25,313 - Epoch: [100][ 990/ 1236] Overall Loss 0.291126 Objective Loss 0.291126 LR 0.000500 Time 0.020948 +2023-10-05 21:29:25,517 - Epoch: [100][ 1000/ 1236] Overall Loss 0.291331 Objective Loss 0.291331 LR 0.000500 Time 0.020942 +2023-10-05 21:29:25,720 - Epoch: [100][ 1010/ 1236] Overall Loss 0.290932 Objective Loss 0.290932 LR 0.000500 Time 0.020935 +2023-10-05 21:29:25,923 - Epoch: [100][ 1020/ 1236] Overall Loss 0.290614 Objective Loss 0.290614 LR 0.000500 Time 0.020929 +2023-10-05 21:29:26,126 - Epoch: [100][ 1030/ 1236] Overall Loss 0.290207 Objective Loss 0.290207 LR 0.000500 Time 0.020923 +2023-10-05 21:29:26,329 - Epoch: [100][ 1040/ 1236] Overall Loss 0.289818 Objective Loss 0.289818 LR 0.000500 Time 0.020916 +2023-10-05 21:29:26,533 - Epoch: [100][ 1050/ 1236] Overall Loss 0.289626 Objective Loss 0.289626 LR 0.000500 Time 0.020911 +2023-10-05 21:29:26,737 - Epoch: [100][ 1060/ 1236] Overall Loss 0.289527 Objective Loss 0.289527 LR 0.000500 Time 0.020905 +2023-10-05 21:29:26,939 - Epoch: [100][ 1070/ 1236] Overall Loss 0.289508 Objective Loss 0.289508 LR 0.000500 Time 0.020899 +2023-10-05 21:29:27,143 - Epoch: [100][ 1080/ 1236] Overall Loss 0.289386 Objective Loss 0.289386 LR 0.000500 Time 0.020894 +2023-10-05 21:29:27,346 - Epoch: [100][ 1090/ 1236] Overall Loss 0.289496 Objective Loss 0.289496 LR 0.000500 Time 0.020888 +2023-10-05 21:29:27,549 - Epoch: [100][ 1100/ 1236] Overall Loss 0.289330 Objective Loss 0.289330 LR 0.000500 Time 0.020883 +2023-10-05 21:29:27,753 - Epoch: [100][ 1110/ 1236] Overall Loss 0.289263 Objective Loss 0.289263 LR 0.000500 Time 0.020878 +2023-10-05 21:29:27,956 - Epoch: [100][ 1120/ 1236] Overall Loss 0.288968 Objective Loss 0.288968 LR 0.000500 Time 0.020872 +2023-10-05 21:29:28,159 - Epoch: [100][ 1130/ 1236] Overall Loss 0.288959 Objective Loss 0.288959 LR 0.000500 Time 0.020867 +2023-10-05 21:29:28,362 - Epoch: [100][ 1140/ 1236] Overall Loss 0.288852 Objective Loss 0.288852 LR 0.000500 Time 0.020862 +2023-10-05 21:29:28,566 - Epoch: [100][ 1150/ 1236] Overall Loss 0.288841 Objective Loss 0.288841 LR 0.000500 Time 0.020857 +2023-10-05 21:29:28,769 - Epoch: [100][ 1160/ 1236] Overall Loss 0.288808 Objective Loss 0.288808 LR 0.000500 Time 0.020853 +2023-10-05 21:29:28,972 - Epoch: [100][ 1170/ 1236] Overall Loss 0.288804 Objective Loss 0.288804 LR 0.000500 Time 0.020848 +2023-10-05 21:29:29,175 - Epoch: [100][ 1180/ 1236] Overall Loss 0.288515 Objective Loss 0.288515 LR 0.000500 Time 0.020843 +2023-10-05 21:29:29,379 - Epoch: [100][ 1190/ 1236] Overall Loss 0.288531 Objective Loss 0.288531 LR 0.000500 Time 0.020838 +2023-10-05 21:29:29,582 - Epoch: [100][ 1200/ 1236] Overall Loss 0.288594 Objective Loss 0.288594 LR 0.000500 Time 0.020834 +2023-10-05 21:29:29,785 - Epoch: [100][ 1210/ 1236] Overall Loss 0.288657 Objective Loss 0.288657 LR 0.000500 Time 0.020829 +2023-10-05 21:29:29,989 - Epoch: [100][ 1220/ 1236] Overall Loss 0.288585 Objective Loss 0.288585 LR 0.000500 Time 0.020825 +2023-10-05 21:29:30,242 - Epoch: [100][ 1230/ 1236] Overall Loss 0.288461 Objective Loss 0.288461 LR 0.000500 Time 0.020862 +2023-10-05 21:29:30,359 - Epoch: [100][ 1236/ 1236] Overall Loss 0.288585 Objective Loss 0.288585 Top1 85.947047 Top5 97.352342 LR 0.000500 Time 0.020854 +2023-10-05 21:29:30,497 - --- validate (epoch=100)----------- +2023-10-05 21:29:30,497 - 29943 samples (256 per mini-batch) +2023-10-05 21:29:30,948 - Epoch: [100][ 10/ 117] Loss 0.336529 Top1 82.539062 Top5 97.812500 +2023-10-05 21:29:31,095 - Epoch: [100][ 20/ 117] Loss 0.346753 Top1 82.519531 Top5 97.500000 +2023-10-05 21:29:31,243 - Epoch: [100][ 30/ 117] Loss 0.347247 Top1 82.708333 Top5 97.552083 +2023-10-05 21:29:31,389 - Epoch: [100][ 40/ 117] Loss 0.342054 Top1 83.017578 Top5 97.626953 +2023-10-05 21:29:31,537 - Epoch: [100][ 50/ 117] Loss 0.342308 Top1 83.031250 Top5 97.640625 +2023-10-05 21:29:31,685 - Epoch: [100][ 60/ 117] Loss 0.338963 Top1 82.942708 Top5 97.656250 +2023-10-05 21:29:31,833 - Epoch: [100][ 70/ 117] Loss 0.341498 Top1 82.779018 Top5 97.645089 +2023-10-05 21:29:31,981 - Epoch: [100][ 80/ 117] Loss 0.339322 Top1 82.832031 Top5 97.651367 +2023-10-05 21:29:32,129 - Epoch: [100][ 90/ 117] Loss 0.340647 Top1 82.855903 Top5 97.651910 +2023-10-05 21:29:32,276 - Epoch: [100][ 100/ 117] Loss 0.338594 Top1 82.941406 Top5 97.656250 +2023-10-05 21:29:32,429 - Epoch: [100][ 110/ 117] Loss 0.339056 Top1 82.798295 Top5 97.645597 +2023-10-05 21:29:32,513 - Epoch: [100][ 117/ 117] Loss 0.335605 Top1 82.914204 Top5 97.655546 +2023-10-05 21:29:32,642 - ==> Top1: 82.914 Top5: 97.656 Loss: 0.336 + +2023-10-05 21:29:32,643 - ==> Confusion: +[[ 917 3 5 1 7 2 0 1 3 81 1 0 2 3 3 3 4 1 0 0 13] + [ 1 1061 1 0 6 23 3 12 0 0 2 2 0 0 1 4 2 1 5 2 5] + [ 5 1 943 16 2 0 31 6 0 2 8 2 9 3 0 2 2 1 6 5 12] + [ 2 0 10 960 1 3 1 0 5 0 11 0 8 6 29 4 2 7 20 2 18] + [ 21 3 0 0 973 6 0 1 0 10 1 1 1 3 4 4 13 1 3 0 5] + [ 5 43 0 0 3 986 0 17 4 2 5 13 3 9 6 1 5 0 1 4 9] + [ 0 4 22 0 0 0 1124 8 0 0 3 1 3 1 1 6 0 2 1 8 7] + [ 1 16 18 1 3 34 7 1050 2 5 4 7 3 2 0 3 0 2 37 12 11] + [ 19 2 0 1 0 8 0 0 960 36 10 1 5 15 20 1 0 0 8 0 3] + [ 97 1 1 0 8 3 1 0 25 936 0 1 0 25 4 8 0 0 2 2 5] + [ 4 4 8 4 2 2 6 2 20 0 957 3 0 14 8 1 1 0 7 2 8] + [ 1 0 0 0 1 14 0 2 1 0 0 944 36 4 0 4 2 14 0 9 3] + [ 0 1 2 9 1 3 1 3 2 1 0 35 968 4 3 5 2 16 1 2 9] + [ 0 1 2 0 2 16 0 0 7 11 3 4 2 1055 2 1 1 1 0 2 9] + [ 14 1 3 11 8 0 1 0 26 3 1 0 3 2 1001 0 0 2 13 0 12] + [ 1 4 4 1 2 1 1 0 0 0 0 6 4 1 1 1067 14 12 0 10 5] + [ 0 15 2 0 4 5 0 1 2 0 0 2 0 1 4 11 1100 0 0 6 8] + [ 0 0 0 2 0 0 2 0 1 0 0 1 19 1 1 6 1 997 1 4 2] + [ 2 10 10 12 1 2 0 23 4 1 1 0 6 1 12 0 0 0 973 3 7] + [ 0 5 1 1 2 3 9 7 1 0 1 15 5 3 0 5 5 2 0 1079 8] + [ 141 217 131 76 114 174 55 110 112 83 171 136 365 307 180 71 183 85 176 242 4776]] + +2023-10-05 21:29:32,644 - ==> Best [Top1: 82.914 Top5: 97.656 Sparsity:0.00 Params: 148928 on epoch: 100] +2023-10-05 21:29:32,645 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:29:32,658 - + +2023-10-05 21:29:32,658 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:29:33,759 - Epoch: [101][ 10/ 1236] Overall Loss 0.287724 Objective Loss 0.287724 LR 0.000500 Time 0.110095 +2023-10-05 21:29:33,963 - Epoch: [101][ 20/ 1236] Overall Loss 0.275712 Objective Loss 0.275712 LR 0.000500 Time 0.065182 +2023-10-05 21:29:34,164 - Epoch: [101][ 30/ 1236] Overall Loss 0.264832 Objective Loss 0.264832 LR 0.000500 Time 0.050155 +2023-10-05 21:29:34,366 - Epoch: [101][ 40/ 1236] Overall Loss 0.267327 Objective Loss 0.267327 LR 0.000500 Time 0.042657 +2023-10-05 21:29:34,566 - Epoch: [101][ 50/ 1236] Overall Loss 0.266954 Objective Loss 0.266954 LR 0.000500 Time 0.038129 +2023-10-05 21:29:34,768 - Epoch: [101][ 60/ 1236] Overall Loss 0.264841 Objective Loss 0.264841 LR 0.000500 Time 0.035136 +2023-10-05 21:29:34,969 - Epoch: [101][ 70/ 1236] Overall Loss 0.262626 Objective Loss 0.262626 LR 0.000500 Time 0.032973 +2023-10-05 21:29:35,171 - Epoch: [101][ 80/ 1236] Overall Loss 0.263315 Objective Loss 0.263315 LR 0.000500 Time 0.031375 +2023-10-05 21:29:35,371 - Epoch: [101][ 90/ 1236] Overall Loss 0.261871 Objective Loss 0.261871 LR 0.000500 Time 0.030114 +2023-10-05 21:29:35,574 - Epoch: [101][ 100/ 1236] Overall Loss 0.260445 Objective Loss 0.260445 LR 0.000500 Time 0.029120 +2023-10-05 21:29:35,773 - Epoch: [101][ 110/ 1236] Overall Loss 0.263953 Objective Loss 0.263953 LR 0.000500 Time 0.028281 +2023-10-05 21:29:35,974 - Epoch: [101][ 120/ 1236] Overall Loss 0.265317 Objective Loss 0.265317 LR 0.000500 Time 0.027597 +2023-10-05 21:29:36,174 - Epoch: [101][ 130/ 1236] Overall Loss 0.267148 Objective Loss 0.267148 LR 0.000500 Time 0.027009 +2023-10-05 21:29:36,374 - Epoch: [101][ 140/ 1236] Overall Loss 0.268494 Objective Loss 0.268494 LR 0.000500 Time 0.026509 +2023-10-05 21:29:36,578 - Epoch: [101][ 150/ 1236] Overall Loss 0.270822 Objective Loss 0.270822 LR 0.000500 Time 0.026098 +2023-10-05 21:29:36,778 - Epoch: [101][ 160/ 1236] Overall Loss 0.270025 Objective Loss 0.270025 LR 0.000500 Time 0.025717 +2023-10-05 21:29:36,978 - Epoch: [101][ 170/ 1236] Overall Loss 0.270972 Objective Loss 0.270972 LR 0.000500 Time 0.025379 +2023-10-05 21:29:37,179 - Epoch: [101][ 180/ 1236] Overall Loss 0.270359 Objective Loss 0.270359 LR 0.000500 Time 0.025081 +2023-10-05 21:29:37,379 - Epoch: [101][ 190/ 1236] Overall Loss 0.269187 Objective Loss 0.269187 LR 0.000500 Time 0.024812 +2023-10-05 21:29:37,580 - Epoch: [101][ 200/ 1236] Overall Loss 0.270300 Objective Loss 0.270300 LR 0.000500 Time 0.024573 +2023-10-05 21:29:37,780 - Epoch: [101][ 210/ 1236] Overall Loss 0.271615 Objective Loss 0.271615 LR 0.000500 Time 0.024355 +2023-10-05 21:29:37,981 - Epoch: [101][ 220/ 1236] Overall Loss 0.270247 Objective Loss 0.270247 LR 0.000500 Time 0.024159 +2023-10-05 21:29:38,181 - Epoch: [101][ 230/ 1236] Overall Loss 0.270089 Objective Loss 0.270089 LR 0.000500 Time 0.023977 +2023-10-05 21:29:38,382 - Epoch: [101][ 240/ 1236] Overall Loss 0.271003 Objective Loss 0.271003 LR 0.000500 Time 0.023813 +2023-10-05 21:29:38,582 - Epoch: [101][ 250/ 1236] Overall Loss 0.271817 Objective Loss 0.271817 LR 0.000500 Time 0.023661 +2023-10-05 21:29:38,781 - Epoch: [101][ 260/ 1236] Overall Loss 0.271438 Objective Loss 0.271438 LR 0.000500 Time 0.023514 +2023-10-05 21:29:38,981 - Epoch: [101][ 270/ 1236] Overall Loss 0.271277 Objective Loss 0.271277 LR 0.000500 Time 0.023383 +2023-10-05 21:29:39,181 - Epoch: [101][ 280/ 1236] Overall Loss 0.271244 Objective Loss 0.271244 LR 0.000500 Time 0.023264 +2023-10-05 21:29:39,382 - Epoch: [101][ 290/ 1236] Overall Loss 0.270567 Objective Loss 0.270567 LR 0.000500 Time 0.023151 +2023-10-05 21:29:39,582 - Epoch: [101][ 300/ 1236] Overall Loss 0.271619 Objective Loss 0.271619 LR 0.000500 Time 0.023044 +2023-10-05 21:29:39,782 - Epoch: [101][ 310/ 1236] Overall Loss 0.273129 Objective Loss 0.273129 LR 0.000500 Time 0.022947 +2023-10-05 21:29:39,981 - Epoch: [101][ 320/ 1236] Overall Loss 0.273491 Objective Loss 0.273491 LR 0.000500 Time 0.022850 +2023-10-05 21:29:40,182 - Epoch: [101][ 330/ 1236] Overall Loss 0.274304 Objective Loss 0.274304 LR 0.000500 Time 0.022766 +2023-10-05 21:29:40,386 - Epoch: [101][ 340/ 1236] Overall Loss 0.274436 Objective Loss 0.274436 LR 0.000500 Time 0.022696 +2023-10-05 21:29:40,589 - Epoch: [101][ 350/ 1236] Overall Loss 0.274974 Objective Loss 0.274974 LR 0.000500 Time 0.022626 +2023-10-05 21:29:40,793 - Epoch: [101][ 360/ 1236] Overall Loss 0.274480 Objective Loss 0.274480 LR 0.000500 Time 0.022564 +2023-10-05 21:29:40,994 - Epoch: [101][ 370/ 1236] Overall Loss 0.274440 Objective Loss 0.274440 LR 0.000500 Time 0.022497 +2023-10-05 21:29:41,197 - Epoch: [101][ 380/ 1236] Overall Loss 0.274172 Objective Loss 0.274172 LR 0.000500 Time 0.022436 +2023-10-05 21:29:41,398 - Epoch: [101][ 390/ 1236] Overall Loss 0.273394 Objective Loss 0.273394 LR 0.000500 Time 0.022377 +2023-10-05 21:29:41,601 - Epoch: [101][ 400/ 1236] Overall Loss 0.272923 Objective Loss 0.272923 LR 0.000500 Time 0.022323 +2023-10-05 21:29:41,802 - Epoch: [101][ 410/ 1236] Overall Loss 0.272466 Objective Loss 0.272466 LR 0.000500 Time 0.022270 +2023-10-05 21:29:42,004 - Epoch: [101][ 420/ 1236] Overall Loss 0.271751 Objective Loss 0.271751 LR 0.000500 Time 0.022218 +2023-10-05 21:29:42,206 - Epoch: [101][ 430/ 1236] Overall Loss 0.271134 Objective Loss 0.271134 LR 0.000500 Time 0.022170 +2023-10-05 21:29:42,408 - Epoch: [101][ 440/ 1236] Overall Loss 0.271097 Objective Loss 0.271097 LR 0.000500 Time 0.022124 +2023-10-05 21:29:42,609 - Epoch: [101][ 450/ 1236] Overall Loss 0.271296 Objective Loss 0.271296 LR 0.000500 Time 0.022080 +2023-10-05 21:29:42,810 - Epoch: [101][ 460/ 1236] Overall Loss 0.271524 Objective Loss 0.271524 LR 0.000500 Time 0.022036 +2023-10-05 21:29:43,010 - Epoch: [101][ 470/ 1236] Overall Loss 0.271118 Objective Loss 0.271118 LR 0.000500 Time 0.021993 +2023-10-05 21:29:43,212 - Epoch: [101][ 480/ 1236] Overall Loss 0.270504 Objective Loss 0.270504 LR 0.000500 Time 0.021953 +2023-10-05 21:29:43,414 - Epoch: [101][ 490/ 1236] Overall Loss 0.270971 Objective Loss 0.270971 LR 0.000500 Time 0.021917 +2023-10-05 21:29:43,615 - Epoch: [101][ 500/ 1236] Overall Loss 0.271134 Objective Loss 0.271134 LR 0.000500 Time 0.021881 +2023-10-05 21:29:43,817 - Epoch: [101][ 510/ 1236] Overall Loss 0.271044 Objective Loss 0.271044 LR 0.000500 Time 0.021846 +2023-10-05 21:29:44,019 - Epoch: [101][ 520/ 1236] Overall Loss 0.271371 Objective Loss 0.271371 LR 0.000500 Time 0.021814 +2023-10-05 21:29:44,220 - Epoch: [101][ 530/ 1236] Overall Loss 0.270844 Objective Loss 0.270844 LR 0.000500 Time 0.021782 +2023-10-05 21:29:44,422 - Epoch: [101][ 540/ 1236] Overall Loss 0.270977 Objective Loss 0.270977 LR 0.000500 Time 0.021752 +2023-10-05 21:29:44,624 - Epoch: [101][ 550/ 1236] Overall Loss 0.271365 Objective Loss 0.271365 LR 0.000500 Time 0.021722 +2023-10-05 21:29:44,826 - Epoch: [101][ 560/ 1236] Overall Loss 0.271425 Objective Loss 0.271425 LR 0.000500 Time 0.021694 +2023-10-05 21:29:45,028 - Epoch: [101][ 570/ 1236] Overall Loss 0.271008 Objective Loss 0.271008 LR 0.000500 Time 0.021667 +2023-10-05 21:29:45,229 - Epoch: [101][ 580/ 1236] Overall Loss 0.270690 Objective Loss 0.270690 LR 0.000500 Time 0.021640 +2023-10-05 21:29:45,431 - Epoch: [101][ 590/ 1236] Overall Loss 0.270663 Objective Loss 0.270663 LR 0.000500 Time 0.021615 +2023-10-05 21:29:45,633 - Epoch: [101][ 600/ 1236] Overall Loss 0.270841 Objective Loss 0.270841 LR 0.000500 Time 0.021591 +2023-10-05 21:29:45,835 - Epoch: [101][ 610/ 1236] Overall Loss 0.270341 Objective Loss 0.270341 LR 0.000500 Time 0.021567 +2023-10-05 21:29:46,037 - Epoch: [101][ 620/ 1236] Overall Loss 0.270306 Objective Loss 0.270306 LR 0.000500 Time 0.021545 +2023-10-05 21:29:46,238 - Epoch: [101][ 630/ 1236] Overall Loss 0.269860 Objective Loss 0.269860 LR 0.000500 Time 0.021522 +2023-10-05 21:29:46,440 - Epoch: [101][ 640/ 1236] Overall Loss 0.269642 Objective Loss 0.269642 LR 0.000500 Time 0.021501 +2023-10-05 21:29:46,642 - Epoch: [101][ 650/ 1236] Overall Loss 0.270576 Objective Loss 0.270576 LR 0.000500 Time 0.021481 +2023-10-05 21:29:46,845 - Epoch: [101][ 660/ 1236] Overall Loss 0.270338 Objective Loss 0.270338 LR 0.000500 Time 0.021462 +2023-10-05 21:29:47,047 - Epoch: [101][ 670/ 1236] Overall Loss 0.269904 Objective Loss 0.269904 LR 0.000500 Time 0.021442 +2023-10-05 21:29:47,249 - Epoch: [101][ 680/ 1236] Overall Loss 0.269525 Objective Loss 0.269525 LR 0.000500 Time 0.021424 +2023-10-05 21:29:47,450 - Epoch: [101][ 690/ 1236] Overall Loss 0.269859 Objective Loss 0.269859 LR 0.000500 Time 0.021405 +2023-10-05 21:29:47,658 - Epoch: [101][ 700/ 1236] Overall Loss 0.270010 Objective Loss 0.270010 LR 0.000500 Time 0.021395 +2023-10-05 21:29:47,860 - Epoch: [101][ 710/ 1236] Overall Loss 0.270094 Objective Loss 0.270094 LR 0.000500 Time 0.021378 +2023-10-05 21:29:48,063 - Epoch: [101][ 720/ 1236] Overall Loss 0.270218 Objective Loss 0.270218 LR 0.000500 Time 0.021362 +2023-10-05 21:29:48,265 - Epoch: [101][ 730/ 1236] Overall Loss 0.270531 Objective Loss 0.270531 LR 0.000500 Time 0.021346 +2023-10-05 21:29:48,468 - Epoch: [101][ 740/ 1236] Overall Loss 0.270863 Objective Loss 0.270863 LR 0.000500 Time 0.021332 +2023-10-05 21:29:48,670 - Epoch: [101][ 750/ 1236] Overall Loss 0.271088 Objective Loss 0.271088 LR 0.000500 Time 0.021316 +2023-10-05 21:29:48,873 - Epoch: [101][ 760/ 1236] Overall Loss 0.271552 Objective Loss 0.271552 LR 0.000500 Time 0.021302 +2023-10-05 21:29:49,076 - Epoch: [101][ 770/ 1236] Overall Loss 0.271874 Objective Loss 0.271874 LR 0.000500 Time 0.021289 +2023-10-05 21:29:49,280 - Epoch: [101][ 780/ 1236] Overall Loss 0.271668 Objective Loss 0.271668 LR 0.000500 Time 0.021277 +2023-10-05 21:29:49,482 - Epoch: [101][ 790/ 1236] Overall Loss 0.271910 Objective Loss 0.271910 LR 0.000500 Time 0.021263 +2023-10-05 21:29:49,686 - Epoch: [101][ 800/ 1236] Overall Loss 0.272009 Objective Loss 0.272009 LR 0.000500 Time 0.021252 +2023-10-05 21:29:49,888 - Epoch: [101][ 810/ 1236] Overall Loss 0.271866 Objective Loss 0.271866 LR 0.000500 Time 0.021238 +2023-10-05 21:29:50,090 - Epoch: [101][ 820/ 1236] Overall Loss 0.271945 Objective Loss 0.271945 LR 0.000500 Time 0.021226 +2023-10-05 21:29:50,292 - Epoch: [101][ 830/ 1236] Overall Loss 0.272342 Objective Loss 0.272342 LR 0.000500 Time 0.021213 +2023-10-05 21:29:50,495 - Epoch: [101][ 840/ 1236] Overall Loss 0.272045 Objective Loss 0.272045 LR 0.000500 Time 0.021201 +2023-10-05 21:29:50,696 - Epoch: [101][ 850/ 1236] Overall Loss 0.271991 Objective Loss 0.271991 LR 0.000500 Time 0.021188 +2023-10-05 21:29:50,899 - Epoch: [101][ 860/ 1236] Overall Loss 0.271836 Objective Loss 0.271836 LR 0.000500 Time 0.021177 +2023-10-05 21:29:51,100 - Epoch: [101][ 870/ 1236] Overall Loss 0.271854 Objective Loss 0.271854 LR 0.000500 Time 0.021165 +2023-10-05 21:29:51,303 - Epoch: [101][ 880/ 1236] Overall Loss 0.271534 Objective Loss 0.271534 LR 0.000500 Time 0.021154 +2023-10-05 21:29:51,504 - Epoch: [101][ 890/ 1236] Overall Loss 0.271621 Objective Loss 0.271621 LR 0.000500 Time 0.021142 +2023-10-05 21:29:51,705 - Epoch: [101][ 900/ 1236] Overall Loss 0.271622 Objective Loss 0.271622 LR 0.000500 Time 0.021130 +2023-10-05 21:29:51,906 - Epoch: [101][ 910/ 1236] Overall Loss 0.271640 Objective Loss 0.271640 LR 0.000500 Time 0.021119 +2023-10-05 21:29:52,109 - Epoch: [101][ 920/ 1236] Overall Loss 0.271693 Objective Loss 0.271693 LR 0.000500 Time 0.021109 +2023-10-05 21:29:52,310 - Epoch: [101][ 930/ 1236] Overall Loss 0.271790 Objective Loss 0.271790 LR 0.000500 Time 0.021098 +2023-10-05 21:29:52,521 - Epoch: [101][ 940/ 1236] Overall Loss 0.271857 Objective Loss 0.271857 LR 0.000500 Time 0.021098 +2023-10-05 21:29:52,728 - Epoch: [101][ 950/ 1236] Overall Loss 0.271693 Objective Loss 0.271693 LR 0.000500 Time 0.021094 +2023-10-05 21:29:52,939 - Epoch: [101][ 960/ 1236] Overall Loss 0.271581 Objective Loss 0.271581 LR 0.000500 Time 0.021094 +2023-10-05 21:29:53,147 - Epoch: [101][ 970/ 1236] Overall Loss 0.271571 Objective Loss 0.271571 LR 0.000500 Time 0.021089 +2023-10-05 21:29:53,358 - Epoch: [101][ 980/ 1236] Overall Loss 0.271785 Objective Loss 0.271785 LR 0.000500 Time 0.021090 +2023-10-05 21:29:53,565 - Epoch: [101][ 990/ 1236] Overall Loss 0.272022 Objective Loss 0.272022 LR 0.000500 Time 0.021086 +2023-10-05 21:29:53,777 - Epoch: [101][ 1000/ 1236] Overall Loss 0.272185 Objective Loss 0.272185 LR 0.000500 Time 0.021086 +2023-10-05 21:29:53,984 - Epoch: [101][ 1010/ 1236] Overall Loss 0.272330 Objective Loss 0.272330 LR 0.000500 Time 0.021082 +2023-10-05 21:29:54,195 - Epoch: [101][ 1020/ 1236] Overall Loss 0.272429 Objective Loss 0.272429 LR 0.000500 Time 0.021082 +2023-10-05 21:29:54,402 - Epoch: [101][ 1030/ 1236] Overall Loss 0.272418 Objective Loss 0.272418 LR 0.000500 Time 0.021078 +2023-10-05 21:29:54,614 - Epoch: [101][ 1040/ 1236] Overall Loss 0.272210 Objective Loss 0.272210 LR 0.000500 Time 0.021078 +2023-10-05 21:29:54,821 - Epoch: [101][ 1050/ 1236] Overall Loss 0.272155 Objective Loss 0.272155 LR 0.000500 Time 0.021075 +2023-10-05 21:29:55,032 - Epoch: [101][ 1060/ 1236] Overall Loss 0.272222 Objective Loss 0.272222 LR 0.000500 Time 0.021075 +2023-10-05 21:29:55,239 - Epoch: [101][ 1070/ 1236] Overall Loss 0.272182 Objective Loss 0.272182 LR 0.000500 Time 0.021071 +2023-10-05 21:29:55,451 - Epoch: [101][ 1080/ 1236] Overall Loss 0.272269 Objective Loss 0.272269 LR 0.000500 Time 0.021072 +2023-10-05 21:29:55,659 - Epoch: [101][ 1090/ 1236] Overall Loss 0.272226 Objective Loss 0.272226 LR 0.000500 Time 0.021069 +2023-10-05 21:29:55,870 - Epoch: [101][ 1100/ 1236] Overall Loss 0.272248 Objective Loss 0.272248 LR 0.000500 Time 0.021069 +2023-10-05 21:29:56,078 - Epoch: [101][ 1110/ 1236] Overall Loss 0.272368 Objective Loss 0.272368 LR 0.000500 Time 0.021066 +2023-10-05 21:29:56,289 - Epoch: [101][ 1120/ 1236] Overall Loss 0.272303 Objective Loss 0.272303 LR 0.000500 Time 0.021067 +2023-10-05 21:29:56,497 - Epoch: [101][ 1130/ 1236] Overall Loss 0.272383 Objective Loss 0.272383 LR 0.000500 Time 0.021064 +2023-10-05 21:29:56,709 - Epoch: [101][ 1140/ 1236] Overall Loss 0.272654 Objective Loss 0.272654 LR 0.000500 Time 0.021065 +2023-10-05 21:29:56,917 - Epoch: [101][ 1150/ 1236] Overall Loss 0.272749 Objective Loss 0.272749 LR 0.000500 Time 0.021062 +2023-10-05 21:29:57,128 - Epoch: [101][ 1160/ 1236] Overall Loss 0.272891 Objective Loss 0.272891 LR 0.000500 Time 0.021062 +2023-10-05 21:29:57,336 - Epoch: [101][ 1170/ 1236] Overall Loss 0.272925 Objective Loss 0.272925 LR 0.000500 Time 0.021060 +2023-10-05 21:29:57,547 - Epoch: [101][ 1180/ 1236] Overall Loss 0.272770 Objective Loss 0.272770 LR 0.000500 Time 0.021060 +2023-10-05 21:29:57,755 - Epoch: [101][ 1190/ 1236] Overall Loss 0.272680 Objective Loss 0.272680 LR 0.000500 Time 0.021057 +2023-10-05 21:29:57,966 - Epoch: [101][ 1200/ 1236] Overall Loss 0.272800 Objective Loss 0.272800 LR 0.000500 Time 0.021057 +2023-10-05 21:29:58,174 - Epoch: [101][ 1210/ 1236] Overall Loss 0.272866 Objective Loss 0.272866 LR 0.000500 Time 0.021055 +2023-10-05 21:29:58,385 - Epoch: [101][ 1220/ 1236] Overall Loss 0.272781 Objective Loss 0.272781 LR 0.000500 Time 0.021055 +2023-10-05 21:29:58,642 - Epoch: [101][ 1230/ 1236] Overall Loss 0.273039 Objective Loss 0.273039 LR 0.000500 Time 0.021093 +2023-10-05 21:29:58,760 - Epoch: [101][ 1236/ 1236] Overall Loss 0.272966 Objective Loss 0.272966 Top1 86.354379 Top5 98.167006 LR 0.000500 Time 0.021085 +2023-10-05 21:29:58,883 - --- validate (epoch=101)----------- +2023-10-05 21:29:58,884 - 29943 samples (256 per mini-batch) +2023-10-05 21:29:59,345 - Epoch: [101][ 10/ 117] Loss 0.309620 Top1 83.281250 Top5 97.617188 +2023-10-05 21:29:59,498 - Epoch: [101][ 20/ 117] Loss 0.319005 Top1 83.085938 Top5 97.656250 +2023-10-05 21:29:59,651 - Epoch: [101][ 30/ 117] Loss 0.323866 Top1 82.981771 Top5 97.682292 +2023-10-05 21:29:59,803 - Epoch: [101][ 40/ 117] Loss 0.329291 Top1 82.861328 Top5 97.626953 +2023-10-05 21:29:59,956 - Epoch: [101][ 50/ 117] Loss 0.330815 Top1 82.898438 Top5 97.804688 +2023-10-05 21:30:00,110 - Epoch: [101][ 60/ 117] Loss 0.335782 Top1 82.786458 Top5 97.766927 +2023-10-05 21:30:00,261 - Epoch: [101][ 70/ 117] Loss 0.334069 Top1 83.035714 Top5 97.845982 +2023-10-05 21:30:00,412 - Epoch: [101][ 80/ 117] Loss 0.333345 Top1 83.095703 Top5 97.817383 +2023-10-05 21:30:00,564 - Epoch: [101][ 90/ 117] Loss 0.333276 Top1 82.977431 Top5 97.860243 +2023-10-05 21:30:00,716 - Epoch: [101][ 100/ 117] Loss 0.335244 Top1 82.914062 Top5 97.851562 +2023-10-05 21:30:00,873 - Epoch: [101][ 110/ 117] Loss 0.339815 Top1 82.801847 Top5 97.844460 +2023-10-05 21:30:00,958 - Epoch: [101][ 117/ 117] Loss 0.340121 Top1 82.820693 Top5 97.859266 +2023-10-05 21:30:01,096 - ==> Top1: 82.821 Top5: 97.859 Loss: 0.340 + +2023-10-05 21:30:01,097 - ==> Confusion: +[[ 900 6 1 0 8 1 0 0 4 90 1 2 2 4 6 8 5 1 0 0 11] + [ 1 1054 1 0 7 22 2 10 1 1 0 1 1 0 2 5 5 1 12 1 4] + [ 6 1 938 23 2 2 26 6 0 1 6 2 9 2 1 5 2 1 5 6 12] + [ 4 2 10 960 1 2 0 0 4 0 9 0 15 2 32 7 2 7 16 1 15] + [ 20 4 1 0 965 7 0 0 0 9 1 2 2 1 10 5 12 2 1 2 6] + [ 3 52 0 0 6 973 1 16 1 1 5 11 5 12 4 2 6 0 4 4 10] + [ 0 3 25 0 1 1 1115 5 0 0 5 4 3 1 1 13 1 1 2 6 4] + [ 1 38 19 1 6 30 9 1017 1 4 5 7 6 3 1 2 0 1 45 8 14] + [ 19 1 0 0 1 6 0 0 960 49 8 2 5 10 15 4 3 3 2 1 0] + [ 86 0 1 0 5 1 2 0 25 938 0 1 0 31 7 9 0 2 0 4 7] + [ 3 6 11 5 2 1 4 4 16 2 954 2 0 20 4 1 0 0 2 5 11] + [ 0 0 0 0 1 10 0 0 0 1 0 941 39 6 0 3 3 16 0 9 6] + [ 2 0 3 6 1 2 0 1 2 0 0 23 982 7 3 9 3 13 4 2 5] + [ 0 0 1 0 3 13 0 0 11 8 5 2 1 1055 2 2 2 1 0 1 12] + [ 16 1 3 12 6 0 0 0 23 7 1 1 3 4 998 0 1 1 11 0 13] + [ 1 2 1 2 4 1 1 0 0 0 0 5 6 2 1 1073 13 8 1 9 4] + [ 0 11 0 1 5 6 0 0 0 0 1 1 4 1 4 11 1101 0 1 6 8] + [ 0 0 0 2 0 0 0 0 0 0 0 2 18 0 0 7 1 1003 0 3 2] + [ 1 9 8 17 1 2 1 20 3 0 2 0 8 2 10 0 1 0 972 2 9] + [ 0 5 1 1 2 4 10 6 0 0 0 10 8 2 0 10 14 2 2 1070 5] + [ 111 227 129 78 105 149 48 62 110 95 175 103 420 302 167 104 250 71 158 211 4830]] + +2023-10-05 21:30:01,098 - ==> Best [Top1: 82.914 Top5: 97.656 Sparsity:0.00 Params: 148928 on epoch: 100] +2023-10-05 21:30:01,098 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:30:01,104 - + +2023-10-05 21:30:01,104 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:30:02,088 - Epoch: [102][ 10/ 1236] Overall Loss 0.278084 Objective Loss 0.278084 LR 0.000500 Time 0.098286 +2023-10-05 21:30:02,289 - Epoch: [102][ 20/ 1236] Overall Loss 0.268366 Objective Loss 0.268366 LR 0.000500 Time 0.059216 +2023-10-05 21:30:02,490 - Epoch: [102][ 30/ 1236] Overall Loss 0.266101 Objective Loss 0.266101 LR 0.000500 Time 0.046166 +2023-10-05 21:30:02,692 - Epoch: [102][ 40/ 1236] Overall Loss 0.273077 Objective Loss 0.273077 LR 0.000500 Time 0.039659 +2023-10-05 21:30:02,894 - Epoch: [102][ 50/ 1236] Overall Loss 0.272494 Objective Loss 0.272494 LR 0.000500 Time 0.035754 +2023-10-05 21:30:03,095 - Epoch: [102][ 60/ 1236] Overall Loss 0.271852 Objective Loss 0.271852 LR 0.000500 Time 0.033143 +2023-10-05 21:30:03,296 - Epoch: [102][ 70/ 1236] Overall Loss 0.274851 Objective Loss 0.274851 LR 0.000500 Time 0.031278 +2023-10-05 21:30:03,498 - Epoch: [102][ 80/ 1236] Overall Loss 0.275801 Objective Loss 0.275801 LR 0.000500 Time 0.029886 +2023-10-05 21:30:03,699 - Epoch: [102][ 90/ 1236] Overall Loss 0.273217 Objective Loss 0.273217 LR 0.000500 Time 0.028799 +2023-10-05 21:30:03,900 - Epoch: [102][ 100/ 1236] Overall Loss 0.270627 Objective Loss 0.270627 LR 0.000500 Time 0.027924 +2023-10-05 21:30:04,100 - Epoch: [102][ 110/ 1236] Overall Loss 0.270492 Objective Loss 0.270492 LR 0.000500 Time 0.027198 +2023-10-05 21:30:04,301 - Epoch: [102][ 120/ 1236] Overall Loss 0.270937 Objective Loss 0.270937 LR 0.000500 Time 0.026603 +2023-10-05 21:30:04,500 - Epoch: [102][ 130/ 1236] Overall Loss 0.268707 Objective Loss 0.268707 LR 0.000500 Time 0.026086 +2023-10-05 21:30:04,703 - Epoch: [102][ 140/ 1236] Overall Loss 0.266986 Objective Loss 0.266986 LR 0.000500 Time 0.025673 +2023-10-05 21:30:04,904 - Epoch: [102][ 150/ 1236] Overall Loss 0.267701 Objective Loss 0.267701 LR 0.000500 Time 0.025294 +2023-10-05 21:30:05,104 - Epoch: [102][ 160/ 1236] Overall Loss 0.268221 Objective Loss 0.268221 LR 0.000500 Time 0.024964 +2023-10-05 21:30:05,304 - Epoch: [102][ 170/ 1236] Overall Loss 0.269741 Objective Loss 0.269741 LR 0.000500 Time 0.024669 +2023-10-05 21:30:05,504 - Epoch: [102][ 180/ 1236] Overall Loss 0.268914 Objective Loss 0.268914 LR 0.000500 Time 0.024409 +2023-10-05 21:30:05,704 - Epoch: [102][ 190/ 1236] Overall Loss 0.269011 Objective Loss 0.269011 LR 0.000500 Time 0.024177 +2023-10-05 21:30:05,907 - Epoch: [102][ 200/ 1236] Overall Loss 0.268104 Objective Loss 0.268104 LR 0.000500 Time 0.023980 +2023-10-05 21:30:06,110 - Epoch: [102][ 210/ 1236] Overall Loss 0.268503 Objective Loss 0.268503 LR 0.000500 Time 0.023802 +2023-10-05 21:30:06,314 - Epoch: [102][ 220/ 1236] Overall Loss 0.268849 Objective Loss 0.268849 LR 0.000500 Time 0.023647 +2023-10-05 21:30:06,517 - Epoch: [102][ 230/ 1236] Overall Loss 0.269155 Objective Loss 0.269155 LR 0.000500 Time 0.023500 +2023-10-05 21:30:06,721 - Epoch: [102][ 240/ 1236] Overall Loss 0.269672 Objective Loss 0.269672 LR 0.000500 Time 0.023369 +2023-10-05 21:30:06,924 - Epoch: [102][ 250/ 1236] Overall Loss 0.268583 Objective Loss 0.268583 LR 0.000500 Time 0.023244 +2023-10-05 21:30:07,128 - Epoch: [102][ 260/ 1236] Overall Loss 0.267427 Objective Loss 0.267427 LR 0.000500 Time 0.023134 +2023-10-05 21:30:07,331 - Epoch: [102][ 270/ 1236] Overall Loss 0.269265 Objective Loss 0.269265 LR 0.000500 Time 0.023028 +2023-10-05 21:30:07,535 - Epoch: [102][ 280/ 1236] Overall Loss 0.269488 Objective Loss 0.269488 LR 0.000500 Time 0.022934 +2023-10-05 21:30:07,738 - Epoch: [102][ 290/ 1236] Overall Loss 0.268480 Objective Loss 0.268480 LR 0.000500 Time 0.022841 +2023-10-05 21:30:07,942 - Epoch: [102][ 300/ 1236] Overall Loss 0.268646 Objective Loss 0.268646 LR 0.000500 Time 0.022759 +2023-10-05 21:30:08,145 - Epoch: [102][ 310/ 1236] Overall Loss 0.268530 Objective Loss 0.268530 LR 0.000500 Time 0.022678 +2023-10-05 21:30:08,349 - Epoch: [102][ 320/ 1236] Overall Loss 0.268632 Objective Loss 0.268632 LR 0.000500 Time 0.022607 +2023-10-05 21:30:08,553 - Epoch: [102][ 330/ 1236] Overall Loss 0.268894 Objective Loss 0.268894 LR 0.000500 Time 0.022536 +2023-10-05 21:30:08,757 - Epoch: [102][ 340/ 1236] Overall Loss 0.270101 Objective Loss 0.270101 LR 0.000500 Time 0.022474 +2023-10-05 21:30:08,960 - Epoch: [102][ 350/ 1236] Overall Loss 0.270921 Objective Loss 0.270921 LR 0.000500 Time 0.022411 +2023-10-05 21:30:09,164 - Epoch: [102][ 360/ 1236] Overall Loss 0.270113 Objective Loss 0.270113 LR 0.000500 Time 0.022355 +2023-10-05 21:30:09,368 - Epoch: [102][ 370/ 1236] Overall Loss 0.269494 Objective Loss 0.269494 LR 0.000500 Time 0.022300 +2023-10-05 21:30:09,571 - Epoch: [102][ 380/ 1236] Overall Loss 0.268756 Objective Loss 0.268756 LR 0.000500 Time 0.022248 +2023-10-05 21:30:09,776 - Epoch: [102][ 390/ 1236] Overall Loss 0.268554 Objective Loss 0.268554 LR 0.000500 Time 0.022201 +2023-10-05 21:30:09,982 - Epoch: [102][ 400/ 1236] Overall Loss 0.268824 Objective Loss 0.268824 LR 0.000500 Time 0.022160 +2023-10-05 21:30:10,183 - Epoch: [102][ 410/ 1236] Overall Loss 0.268799 Objective Loss 0.268799 LR 0.000500 Time 0.022108 +2023-10-05 21:30:10,385 - Epoch: [102][ 420/ 1236] Overall Loss 0.268388 Objective Loss 0.268388 LR 0.000500 Time 0.022063 +2023-10-05 21:30:10,586 - Epoch: [102][ 430/ 1236] Overall Loss 0.268551 Objective Loss 0.268551 LR 0.000500 Time 0.022016 +2023-10-05 21:30:10,788 - Epoch: [102][ 440/ 1236] Overall Loss 0.268500 Objective Loss 0.268500 LR 0.000500 Time 0.021974 +2023-10-05 21:30:10,989 - Epoch: [102][ 450/ 1236] Overall Loss 0.269050 Objective Loss 0.269050 LR 0.000500 Time 0.021931 +2023-10-05 21:30:11,192 - Epoch: [102][ 460/ 1236] Overall Loss 0.268599 Objective Loss 0.268599 LR 0.000500 Time 0.021894 +2023-10-05 21:30:11,393 - Epoch: [102][ 470/ 1236] Overall Loss 0.268144 Objective Loss 0.268144 LR 0.000500 Time 0.021856 +2023-10-05 21:30:11,595 - Epoch: [102][ 480/ 1236] Overall Loss 0.267492 Objective Loss 0.267492 LR 0.000500 Time 0.021822 +2023-10-05 21:30:11,796 - Epoch: [102][ 490/ 1236] Overall Loss 0.266944 Objective Loss 0.266944 LR 0.000500 Time 0.021786 +2023-10-05 21:30:11,999 - Epoch: [102][ 500/ 1236] Overall Loss 0.266706 Objective Loss 0.266706 LR 0.000500 Time 0.021755 +2023-10-05 21:30:12,200 - Epoch: [102][ 510/ 1236] Overall Loss 0.266542 Objective Loss 0.266542 LR 0.000500 Time 0.021721 +2023-10-05 21:30:12,403 - Epoch: [102][ 520/ 1236] Overall Loss 0.266059 Objective Loss 0.266059 LR 0.000500 Time 0.021692 +2023-10-05 21:30:12,603 - Epoch: [102][ 530/ 1236] Overall Loss 0.265800 Objective Loss 0.265800 LR 0.000500 Time 0.021660 +2023-10-05 21:30:12,806 - Epoch: [102][ 540/ 1236] Overall Loss 0.265836 Objective Loss 0.265836 LR 0.000500 Time 0.021634 +2023-10-05 21:30:13,006 - Epoch: [102][ 550/ 1236] Overall Loss 0.265599 Objective Loss 0.265599 LR 0.000500 Time 0.021605 +2023-10-05 21:30:13,209 - Epoch: [102][ 560/ 1236] Overall Loss 0.265465 Objective Loss 0.265465 LR 0.000500 Time 0.021581 +2023-10-05 21:30:13,410 - Epoch: [102][ 570/ 1236] Overall Loss 0.265615 Objective Loss 0.265615 LR 0.000500 Time 0.021553 +2023-10-05 21:30:13,613 - Epoch: [102][ 580/ 1236] Overall Loss 0.265721 Objective Loss 0.265721 LR 0.000500 Time 0.021531 +2023-10-05 21:30:13,813 - Epoch: [102][ 590/ 1236] Overall Loss 0.265829 Objective Loss 0.265829 LR 0.000500 Time 0.021505 +2023-10-05 21:30:14,016 - Epoch: [102][ 600/ 1236] Overall Loss 0.265823 Objective Loss 0.265823 LR 0.000500 Time 0.021484 +2023-10-05 21:30:14,217 - Epoch: [102][ 610/ 1236] Overall Loss 0.265559 Objective Loss 0.265559 LR 0.000500 Time 0.021461 +2023-10-05 21:30:14,420 - Epoch: [102][ 620/ 1236] Overall Loss 0.265096 Objective Loss 0.265096 LR 0.000500 Time 0.021441 +2023-10-05 21:30:14,621 - Epoch: [102][ 630/ 1236] Overall Loss 0.265007 Objective Loss 0.265007 LR 0.000500 Time 0.021420 +2023-10-05 21:30:14,824 - Epoch: [102][ 640/ 1236] Overall Loss 0.264974 Objective Loss 0.264974 LR 0.000500 Time 0.021402 +2023-10-05 21:30:15,025 - Epoch: [102][ 650/ 1236] Overall Loss 0.264796 Objective Loss 0.264796 LR 0.000500 Time 0.021381 +2023-10-05 21:30:15,228 - Epoch: [102][ 660/ 1236] Overall Loss 0.265262 Objective Loss 0.265262 LR 0.000500 Time 0.021364 +2023-10-05 21:30:15,428 - Epoch: [102][ 670/ 1236] Overall Loss 0.265551 Objective Loss 0.265551 LR 0.000500 Time 0.021344 +2023-10-05 21:30:15,631 - Epoch: [102][ 680/ 1236] Overall Loss 0.265478 Objective Loss 0.265478 LR 0.000500 Time 0.021328 +2023-10-05 21:30:15,833 - Epoch: [102][ 690/ 1236] Overall Loss 0.265075 Objective Loss 0.265075 LR 0.000500 Time 0.021310 +2023-10-05 21:30:16,036 - Epoch: [102][ 700/ 1236] Overall Loss 0.265073 Objective Loss 0.265073 LR 0.000500 Time 0.021295 +2023-10-05 21:30:16,236 - Epoch: [102][ 710/ 1236] Overall Loss 0.264975 Objective Loss 0.264975 LR 0.000500 Time 0.021277 +2023-10-05 21:30:16,440 - Epoch: [102][ 720/ 1236] Overall Loss 0.265658 Objective Loss 0.265658 LR 0.000500 Time 0.021263 +2023-10-05 21:30:16,640 - Epoch: [102][ 730/ 1236] Overall Loss 0.265883 Objective Loss 0.265883 LR 0.000500 Time 0.021247 +2023-10-05 21:30:16,843 - Epoch: [102][ 740/ 1236] Overall Loss 0.265964 Objective Loss 0.265964 LR 0.000500 Time 0.021234 +2023-10-05 21:30:17,044 - Epoch: [102][ 750/ 1236] Overall Loss 0.266201 Objective Loss 0.266201 LR 0.000500 Time 0.021217 +2023-10-05 21:30:17,249 - Epoch: [102][ 760/ 1236] Overall Loss 0.266532 Objective Loss 0.266532 LR 0.000500 Time 0.021208 +2023-10-05 21:30:17,451 - Epoch: [102][ 770/ 1236] Overall Loss 0.266188 Objective Loss 0.266188 LR 0.000500 Time 0.021194 +2023-10-05 21:30:17,655 - Epoch: [102][ 780/ 1236] Overall Loss 0.266356 Objective Loss 0.266356 LR 0.000500 Time 0.021183 +2023-10-05 21:30:17,855 - Epoch: [102][ 790/ 1236] Overall Loss 0.267232 Objective Loss 0.267232 LR 0.000500 Time 0.021168 +2023-10-05 21:30:18,058 - Epoch: [102][ 800/ 1236] Overall Loss 0.266823 Objective Loss 0.266823 LR 0.000500 Time 0.021157 +2023-10-05 21:30:18,259 - Epoch: [102][ 810/ 1236] Overall Loss 0.266890 Objective Loss 0.266890 LR 0.000500 Time 0.021143 +2023-10-05 21:30:18,463 - Epoch: [102][ 820/ 1236] Overall Loss 0.266954 Objective Loss 0.266954 LR 0.000500 Time 0.021134 +2023-10-05 21:30:18,663 - Epoch: [102][ 830/ 1236] Overall Loss 0.266967 Objective Loss 0.266967 LR 0.000500 Time 0.021120 +2023-10-05 21:30:18,866 - Epoch: [102][ 840/ 1236] Overall Loss 0.266703 Objective Loss 0.266703 LR 0.000500 Time 0.021110 +2023-10-05 21:30:19,067 - Epoch: [102][ 850/ 1236] Overall Loss 0.266750 Objective Loss 0.266750 LR 0.000500 Time 0.021097 +2023-10-05 21:30:19,270 - Epoch: [102][ 860/ 1236] Overall Loss 0.266570 Objective Loss 0.266570 LR 0.000500 Time 0.021088 +2023-10-05 21:30:19,471 - Epoch: [102][ 870/ 1236] Overall Loss 0.266940 Objective Loss 0.266940 LR 0.000500 Time 0.021076 +2023-10-05 21:30:19,675 - Epoch: [102][ 880/ 1236] Overall Loss 0.266774 Objective Loss 0.266774 LR 0.000500 Time 0.021067 +2023-10-05 21:30:19,875 - Epoch: [102][ 890/ 1236] Overall Loss 0.267064 Objective Loss 0.267064 LR 0.000500 Time 0.021056 +2023-10-05 21:30:20,079 - Epoch: [102][ 900/ 1236] Overall Loss 0.266921 Objective Loss 0.266921 LR 0.000500 Time 0.021047 +2023-10-05 21:30:20,279 - Epoch: [102][ 910/ 1236] Overall Loss 0.267243 Objective Loss 0.267243 LR 0.000500 Time 0.021036 +2023-10-05 21:30:20,483 - Epoch: [102][ 920/ 1236] Overall Loss 0.267578 Objective Loss 0.267578 LR 0.000500 Time 0.021028 +2023-10-05 21:30:20,683 - Epoch: [102][ 930/ 1236] Overall Loss 0.267547 Objective Loss 0.267547 LR 0.000500 Time 0.021017 +2023-10-05 21:30:20,887 - Epoch: [102][ 940/ 1236] Overall Loss 0.267809 Objective Loss 0.267809 LR 0.000500 Time 0.021010 +2023-10-05 21:30:21,087 - Epoch: [102][ 950/ 1236] Overall Loss 0.267576 Objective Loss 0.267576 LR 0.000500 Time 0.021000 +2023-10-05 21:30:21,291 - Epoch: [102][ 960/ 1236] Overall Loss 0.267817 Objective Loss 0.267817 LR 0.000500 Time 0.020992 +2023-10-05 21:30:21,492 - Epoch: [102][ 970/ 1236] Overall Loss 0.268062 Objective Loss 0.268062 LR 0.000500 Time 0.020983 +2023-10-05 21:30:21,696 - Epoch: [102][ 980/ 1236] Overall Loss 0.268129 Objective Loss 0.268129 LR 0.000500 Time 0.020977 +2023-10-05 21:30:21,898 - Epoch: [102][ 990/ 1236] Overall Loss 0.268211 Objective Loss 0.268211 LR 0.000500 Time 0.020969 +2023-10-05 21:30:22,102 - Epoch: [102][ 1000/ 1236] Overall Loss 0.267962 Objective Loss 0.267962 LR 0.000500 Time 0.020963 +2023-10-05 21:30:22,304 - Epoch: [102][ 1010/ 1236] Overall Loss 0.267805 Objective Loss 0.267805 LR 0.000500 Time 0.020955 +2023-10-05 21:30:22,508 - Epoch: [102][ 1020/ 1236] Overall Loss 0.267740 Objective Loss 0.267740 LR 0.000500 Time 0.020949 +2023-10-05 21:30:22,710 - Epoch: [102][ 1030/ 1236] Overall Loss 0.268003 Objective Loss 0.268003 LR 0.000500 Time 0.020941 +2023-10-05 21:30:22,914 - Epoch: [102][ 1040/ 1236] Overall Loss 0.267858 Objective Loss 0.267858 LR 0.000500 Time 0.020936 +2023-10-05 21:30:23,116 - Epoch: [102][ 1050/ 1236] Overall Loss 0.267676 Objective Loss 0.267676 LR 0.000500 Time 0.020928 +2023-10-05 21:30:23,320 - Epoch: [102][ 1060/ 1236] Overall Loss 0.267953 Objective Loss 0.267953 LR 0.000500 Time 0.020923 +2023-10-05 21:30:23,522 - Epoch: [102][ 1070/ 1236] Overall Loss 0.268201 Objective Loss 0.268201 LR 0.000500 Time 0.020916 +2023-10-05 21:30:23,726 - Epoch: [102][ 1080/ 1236] Overall Loss 0.268504 Objective Loss 0.268504 LR 0.000500 Time 0.020911 +2023-10-05 21:30:23,928 - Epoch: [102][ 1090/ 1236] Overall Loss 0.268520 Objective Loss 0.268520 LR 0.000500 Time 0.020904 +2023-10-05 21:30:24,132 - Epoch: [102][ 1100/ 1236] Overall Loss 0.268543 Objective Loss 0.268543 LR 0.000500 Time 0.020899 +2023-10-05 21:30:24,334 - Epoch: [102][ 1110/ 1236] Overall Loss 0.268455 Objective Loss 0.268455 LR 0.000500 Time 0.020893 +2023-10-05 21:30:24,538 - Epoch: [102][ 1120/ 1236] Overall Loss 0.268595 Objective Loss 0.268595 LR 0.000500 Time 0.020888 +2023-10-05 21:30:24,740 - Epoch: [102][ 1130/ 1236] Overall Loss 0.268223 Objective Loss 0.268223 LR 0.000500 Time 0.020882 +2023-10-05 21:30:24,945 - Epoch: [102][ 1140/ 1236] Overall Loss 0.268370 Objective Loss 0.268370 LR 0.000500 Time 0.020878 +2023-10-05 21:30:25,147 - Epoch: [102][ 1150/ 1236] Overall Loss 0.268237 Objective Loss 0.268237 LR 0.000500 Time 0.020872 +2023-10-05 21:30:25,351 - Epoch: [102][ 1160/ 1236] Overall Loss 0.268223 Objective Loss 0.268223 LR 0.000500 Time 0.020867 +2023-10-05 21:30:25,552 - Epoch: [102][ 1170/ 1236] Overall Loss 0.268251 Objective Loss 0.268251 LR 0.000500 Time 0.020860 +2023-10-05 21:30:25,756 - Epoch: [102][ 1180/ 1236] Overall Loss 0.268200 Objective Loss 0.268200 LR 0.000500 Time 0.020856 +2023-10-05 21:30:25,958 - Epoch: [102][ 1190/ 1236] Overall Loss 0.268296 Objective Loss 0.268296 LR 0.000500 Time 0.020850 +2023-10-05 21:30:26,162 - Epoch: [102][ 1200/ 1236] Overall Loss 0.268439 Objective Loss 0.268439 LR 0.000500 Time 0.020846 +2023-10-05 21:30:26,364 - Epoch: [102][ 1210/ 1236] Overall Loss 0.268536 Objective Loss 0.268536 LR 0.000500 Time 0.020841 +2023-10-05 21:30:26,568 - Epoch: [102][ 1220/ 1236] Overall Loss 0.268537 Objective Loss 0.268537 LR 0.000500 Time 0.020837 +2023-10-05 21:30:26,822 - Epoch: [102][ 1230/ 1236] Overall Loss 0.268613 Objective Loss 0.268613 LR 0.000500 Time 0.020874 +2023-10-05 21:30:26,940 - Epoch: [102][ 1236/ 1236] Overall Loss 0.268705 Objective Loss 0.268705 Top1 84.928717 Top5 97.556008 LR 0.000500 Time 0.020868 +2023-10-05 21:30:27,061 - --- validate (epoch=102)----------- +2023-10-05 21:30:27,061 - 29943 samples (256 per mini-batch) +2023-10-05 21:30:27,515 - Epoch: [102][ 10/ 117] Loss 0.320671 Top1 84.062500 Top5 97.851562 +2023-10-05 21:30:27,666 - Epoch: [102][ 20/ 117] Loss 0.317870 Top1 83.906250 Top5 97.792969 +2023-10-05 21:30:27,816 - Epoch: [102][ 30/ 117] Loss 0.324756 Top1 83.554688 Top5 97.708333 +2023-10-05 21:30:27,967 - Epoch: [102][ 40/ 117] Loss 0.330072 Top1 83.066406 Top5 97.675781 +2023-10-05 21:30:28,117 - Epoch: [102][ 50/ 117] Loss 0.329905 Top1 83.187500 Top5 97.812500 +2023-10-05 21:30:28,266 - Epoch: [102][ 60/ 117] Loss 0.329630 Top1 83.092448 Top5 97.819010 +2023-10-05 21:30:28,414 - Epoch: [102][ 70/ 117] Loss 0.330147 Top1 83.069196 Top5 97.818080 +2023-10-05 21:30:28,561 - Epoch: [102][ 80/ 117] Loss 0.329648 Top1 82.929688 Top5 97.783203 +2023-10-05 21:30:28,709 - Epoch: [102][ 90/ 117] Loss 0.330322 Top1 82.864583 Top5 97.769097 +2023-10-05 21:30:28,857 - Epoch: [102][ 100/ 117] Loss 0.331131 Top1 82.812500 Top5 97.769531 +2023-10-05 21:30:29,012 - Epoch: [102][ 110/ 117] Loss 0.330502 Top1 82.933239 Top5 97.776989 +2023-10-05 21:30:29,096 - Epoch: [102][ 117/ 117] Loss 0.331283 Top1 82.824032 Top5 97.765755 +2023-10-05 21:30:29,223 - ==> Top1: 82.824 Top5: 97.766 Loss: 0.331 + +2023-10-05 21:30:29,223 - ==> Confusion: +[[ 900 1 4 2 10 3 0 0 8 87 1 2 1 3 4 3 5 3 1 0 12] + [ 0 1054 3 0 5 21 1 14 0 0 3 3 0 0 0 3 3 2 9 2 8] + [ 3 0 926 18 4 1 42 5 0 3 6 3 9 1 0 5 1 2 7 6 14] + [ 2 0 17 960 0 2 2 1 5 0 10 1 7 3 26 3 5 5 17 3 20] + [ 18 2 0 0 976 8 1 1 0 9 2 3 2 3 8 3 7 3 0 1 3] + [ 3 54 0 0 3 962 1 16 2 0 6 11 3 12 7 2 4 1 3 9 17] + [ 0 4 22 0 0 0 1125 8 0 0 4 3 2 0 0 9 0 0 4 7 3] + [ 2 29 18 0 3 24 4 1048 0 3 5 8 8 1 0 1 1 0 40 11 12] + [ 17 0 0 1 1 2 0 0 978 40 13 2 3 7 14 5 0 0 3 2 1] + [ 85 0 1 0 8 4 1 0 43 925 0 1 0 28 5 7 0 3 0 1 7] + [ 1 3 9 2 0 0 6 3 20 1 968 6 0 13 2 3 1 1 4 2 8] + [ 2 1 1 0 0 9 0 3 0 0 0 947 27 9 0 2 2 18 0 11 3] + [ 0 0 2 6 2 3 0 1 4 0 2 29 976 8 1 4 2 16 2 3 7] + [ 0 0 1 0 3 8 0 0 24 9 9 4 2 1044 2 2 0 1 0 2 8] + [ 10 2 3 12 6 0 0 0 31 5 2 1 3 2 994 0 0 2 15 0 13] + [ 0 0 3 1 4 1 1 0 0 0 0 10 4 3 0 1068 15 8 2 8 6] + [ 0 14 1 0 5 1 0 2 1 0 1 4 0 2 2 12 1095 0 0 9 12] + [ 0 1 0 0 0 0 1 0 0 0 0 8 24 1 1 4 0 992 1 3 2] + [ 1 14 9 15 1 0 1 20 5 0 6 0 6 0 7 1 0 0 973 1 8] + [ 0 3 3 1 2 3 8 7 0 0 1 11 5 1 0 7 9 1 0 1079 11] + [ 133 231 124 68 103 111 52 101 124 101 204 120 379 331 138 61 230 78 178 228 4810]] + +2023-10-05 21:30:29,225 - ==> Best [Top1: 82.914 Top5: 97.656 Sparsity:0.00 Params: 148928 on epoch: 100] +2023-10-05 21:30:29,225 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:30:29,231 - + +2023-10-05 21:30:29,231 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:30:30,210 - Epoch: [103][ 10/ 1236] Overall Loss 0.267644 Objective Loss 0.267644 LR 0.000500 Time 0.097855 +2023-10-05 21:30:30,415 - Epoch: [103][ 20/ 1236] Overall Loss 0.274737 Objective Loss 0.274737 LR 0.000500 Time 0.059158 +2023-10-05 21:30:30,617 - Epoch: [103][ 30/ 1236] Overall Loss 0.270628 Objective Loss 0.270628 LR 0.000500 Time 0.046176 +2023-10-05 21:30:30,821 - Epoch: [103][ 40/ 1236] Overall Loss 0.273261 Objective Loss 0.273261 LR 0.000500 Time 0.039714 +2023-10-05 21:30:31,022 - Epoch: [103][ 50/ 1236] Overall Loss 0.268596 Objective Loss 0.268596 LR 0.000500 Time 0.035778 +2023-10-05 21:30:31,223 - Epoch: [103][ 60/ 1236] Overall Loss 0.266165 Objective Loss 0.266165 LR 0.000500 Time 0.033168 +2023-10-05 21:30:31,424 - Epoch: [103][ 70/ 1236] Overall Loss 0.262644 Objective Loss 0.262644 LR 0.000500 Time 0.031298 +2023-10-05 21:30:31,626 - Epoch: [103][ 80/ 1236] Overall Loss 0.264765 Objective Loss 0.264765 LR 0.000500 Time 0.029902 +2023-10-05 21:30:31,827 - Epoch: [103][ 90/ 1236] Overall Loss 0.262909 Objective Loss 0.262909 LR 0.000500 Time 0.028806 +2023-10-05 21:30:32,029 - Epoch: [103][ 100/ 1236] Overall Loss 0.267211 Objective Loss 0.267211 LR 0.000500 Time 0.027942 +2023-10-05 21:30:32,229 - Epoch: [103][ 110/ 1236] Overall Loss 0.267211 Objective Loss 0.267211 LR 0.000500 Time 0.027213 +2023-10-05 21:30:32,430 - Epoch: [103][ 120/ 1236] Overall Loss 0.265968 Objective Loss 0.265968 LR 0.000500 Time 0.026624 +2023-10-05 21:30:32,630 - Epoch: [103][ 130/ 1236] Overall Loss 0.265164 Objective Loss 0.265164 LR 0.000500 Time 0.026110 +2023-10-05 21:30:32,832 - Epoch: [103][ 140/ 1236] Overall Loss 0.264220 Objective Loss 0.264220 LR 0.000500 Time 0.025682 +2023-10-05 21:30:33,032 - Epoch: [103][ 150/ 1236] Overall Loss 0.265225 Objective Loss 0.265225 LR 0.000500 Time 0.025301 +2023-10-05 21:30:33,233 - Epoch: [103][ 160/ 1236] Overall Loss 0.263263 Objective Loss 0.263263 LR 0.000500 Time 0.024977 +2023-10-05 21:30:33,434 - Epoch: [103][ 170/ 1236] Overall Loss 0.264664 Objective Loss 0.264664 LR 0.000500 Time 0.024685 +2023-10-05 21:30:33,636 - Epoch: [103][ 180/ 1236] Overall Loss 0.263782 Objective Loss 0.263782 LR 0.000500 Time 0.024437 +2023-10-05 21:30:33,836 - Epoch: [103][ 190/ 1236] Overall Loss 0.262690 Objective Loss 0.262690 LR 0.000500 Time 0.024202 +2023-10-05 21:30:34,038 - Epoch: [103][ 200/ 1236] Overall Loss 0.264064 Objective Loss 0.264064 LR 0.000500 Time 0.023998 +2023-10-05 21:30:34,238 - Epoch: [103][ 210/ 1236] Overall Loss 0.263948 Objective Loss 0.263948 LR 0.000500 Time 0.023808 +2023-10-05 21:30:34,440 - Epoch: [103][ 220/ 1236] Overall Loss 0.263964 Objective Loss 0.263964 LR 0.000500 Time 0.023642 +2023-10-05 21:30:34,640 - Epoch: [103][ 230/ 1236] Overall Loss 0.263440 Objective Loss 0.263440 LR 0.000500 Time 0.023480 +2023-10-05 21:30:34,841 - Epoch: [103][ 240/ 1236] Overall Loss 0.263169 Objective Loss 0.263169 LR 0.000500 Time 0.023339 +2023-10-05 21:30:35,041 - Epoch: [103][ 250/ 1236] Overall Loss 0.261982 Objective Loss 0.261982 LR 0.000500 Time 0.023205 +2023-10-05 21:30:35,243 - Epoch: [103][ 260/ 1236] Overall Loss 0.262291 Objective Loss 0.262291 LR 0.000500 Time 0.023086 +2023-10-05 21:30:35,444 - Epoch: [103][ 270/ 1236] Overall Loss 0.261310 Objective Loss 0.261310 LR 0.000500 Time 0.022973 +2023-10-05 21:30:35,645 - Epoch: [103][ 280/ 1236] Overall Loss 0.260955 Objective Loss 0.260955 LR 0.000500 Time 0.022871 +2023-10-05 21:30:35,845 - Epoch: [103][ 290/ 1236] Overall Loss 0.261683 Objective Loss 0.261683 LR 0.000500 Time 0.022770 +2023-10-05 21:30:36,046 - Epoch: [103][ 300/ 1236] Overall Loss 0.262563 Objective Loss 0.262563 LR 0.000500 Time 0.022682 +2023-10-05 21:30:36,247 - Epoch: [103][ 310/ 1236] Overall Loss 0.263380 Objective Loss 0.263380 LR 0.000500 Time 0.022597 +2023-10-05 21:30:36,449 - Epoch: [103][ 320/ 1236] Overall Loss 0.263199 Objective Loss 0.263199 LR 0.000500 Time 0.022520 +2023-10-05 21:30:36,649 - Epoch: [103][ 330/ 1236] Overall Loss 0.262498 Objective Loss 0.262498 LR 0.000500 Time 0.022444 +2023-10-05 21:30:36,853 - Epoch: [103][ 340/ 1236] Overall Loss 0.262776 Objective Loss 0.262776 LR 0.000500 Time 0.022381 +2023-10-05 21:30:37,058 - Epoch: [103][ 350/ 1236] Overall Loss 0.262024 Objective Loss 0.262024 LR 0.000500 Time 0.022327 +2023-10-05 21:30:37,264 - Epoch: [103][ 360/ 1236] Overall Loss 0.262151 Objective Loss 0.262151 LR 0.000500 Time 0.022278 +2023-10-05 21:30:37,465 - Epoch: [103][ 370/ 1236] Overall Loss 0.262961 Objective Loss 0.262961 LR 0.000500 Time 0.022217 +2023-10-05 21:30:37,667 - Epoch: [103][ 380/ 1236] Overall Loss 0.262734 Objective Loss 0.262734 LR 0.000500 Time 0.022164 +2023-10-05 21:30:37,868 - Epoch: [103][ 390/ 1236] Overall Loss 0.262487 Objective Loss 0.262487 LR 0.000500 Time 0.022109 +2023-10-05 21:30:38,071 - Epoch: [103][ 400/ 1236] Overall Loss 0.262717 Objective Loss 0.262717 LR 0.000500 Time 0.022062 +2023-10-05 21:30:38,271 - Epoch: [103][ 410/ 1236] Overall Loss 0.262887 Objective Loss 0.262887 LR 0.000500 Time 0.022012 +2023-10-05 21:30:38,473 - Epoch: [103][ 420/ 1236] Overall Loss 0.263188 Objective Loss 0.263188 LR 0.000500 Time 0.021969 +2023-10-05 21:30:38,676 - Epoch: [103][ 430/ 1236] Overall Loss 0.263917 Objective Loss 0.263917 LR 0.000500 Time 0.021928 +2023-10-05 21:30:38,881 - Epoch: [103][ 440/ 1236] Overall Loss 0.264548 Objective Loss 0.264548 LR 0.000500 Time 0.021895 +2023-10-05 21:30:39,084 - Epoch: [103][ 450/ 1236] Overall Loss 0.264761 Objective Loss 0.264761 LR 0.000500 Time 0.021857 +2023-10-05 21:30:39,289 - Epoch: [103][ 460/ 1236] Overall Loss 0.264640 Objective Loss 0.264640 LR 0.000500 Time 0.021827 +2023-10-05 21:30:39,491 - Epoch: [103][ 470/ 1236] Overall Loss 0.264610 Objective Loss 0.264610 LR 0.000500 Time 0.021793 +2023-10-05 21:30:39,696 - Epoch: [103][ 480/ 1236] Overall Loss 0.264757 Objective Loss 0.264757 LR 0.000500 Time 0.021765 +2023-10-05 21:30:39,899 - Epoch: [103][ 490/ 1236] Overall Loss 0.264689 Objective Loss 0.264689 LR 0.000500 Time 0.021733 +2023-10-05 21:30:40,104 - Epoch: [103][ 500/ 1236] Overall Loss 0.264097 Objective Loss 0.264097 LR 0.000500 Time 0.021708 +2023-10-05 21:30:40,306 - Epoch: [103][ 510/ 1236] Overall Loss 0.263850 Objective Loss 0.263850 LR 0.000500 Time 0.021679 +2023-10-05 21:30:40,511 - Epoch: [103][ 520/ 1236] Overall Loss 0.263824 Objective Loss 0.263824 LR 0.000500 Time 0.021656 +2023-10-05 21:30:40,714 - Epoch: [103][ 530/ 1236] Overall Loss 0.263357 Objective Loss 0.263357 LR 0.000500 Time 0.021628 +2023-10-05 21:30:40,919 - Epoch: [103][ 540/ 1236] Overall Loss 0.263231 Objective Loss 0.263231 LR 0.000500 Time 0.021607 +2023-10-05 21:30:41,121 - Epoch: [103][ 550/ 1236] Overall Loss 0.262789 Objective Loss 0.262789 LR 0.000500 Time 0.021581 +2023-10-05 21:30:41,327 - Epoch: [103][ 560/ 1236] Overall Loss 0.262904 Objective Loss 0.262904 LR 0.000500 Time 0.021562 +2023-10-05 21:30:41,529 - Epoch: [103][ 570/ 1236] Overall Loss 0.262951 Objective Loss 0.262951 LR 0.000500 Time 0.021538 +2023-10-05 21:30:41,734 - Epoch: [103][ 580/ 1236] Overall Loss 0.263073 Objective Loss 0.263073 LR 0.000500 Time 0.021520 +2023-10-05 21:30:41,937 - Epoch: [103][ 590/ 1236] Overall Loss 0.262866 Objective Loss 0.262866 LR 0.000500 Time 0.021497 +2023-10-05 21:30:42,142 - Epoch: [103][ 600/ 1236] Overall Loss 0.262825 Objective Loss 0.262825 LR 0.000500 Time 0.021481 +2023-10-05 21:30:42,344 - Epoch: [103][ 610/ 1236] Overall Loss 0.262709 Objective Loss 0.262709 LR 0.000500 Time 0.021459 +2023-10-05 21:30:42,549 - Epoch: [103][ 620/ 1236] Overall Loss 0.262710 Objective Loss 0.262710 LR 0.000500 Time 0.021444 +2023-10-05 21:30:42,752 - Epoch: [103][ 630/ 1236] Overall Loss 0.262717 Objective Loss 0.262717 LR 0.000500 Time 0.021424 +2023-10-05 21:30:42,957 - Epoch: [103][ 640/ 1236] Overall Loss 0.262459 Objective Loss 0.262459 LR 0.000500 Time 0.021409 +2023-10-05 21:30:43,159 - Epoch: [103][ 650/ 1236] Overall Loss 0.262589 Objective Loss 0.262589 LR 0.000500 Time 0.021391 +2023-10-05 21:30:43,366 - Epoch: [103][ 660/ 1236] Overall Loss 0.263004 Objective Loss 0.263004 LR 0.000500 Time 0.021379 +2023-10-05 21:30:43,568 - Epoch: [103][ 670/ 1236] Overall Loss 0.262485 Objective Loss 0.262485 LR 0.000500 Time 0.021361 +2023-10-05 21:30:43,772 - Epoch: [103][ 680/ 1236] Overall Loss 0.262905 Objective Loss 0.262905 LR 0.000500 Time 0.021347 +2023-10-05 21:30:43,973 - Epoch: [103][ 690/ 1236] Overall Loss 0.262478 Objective Loss 0.262478 LR 0.000500 Time 0.021328 +2023-10-05 21:30:44,177 - Epoch: [103][ 700/ 1236] Overall Loss 0.262522 Objective Loss 0.262522 LR 0.000500 Time 0.021314 +2023-10-05 21:30:44,378 - Epoch: [103][ 710/ 1236] Overall Loss 0.262663 Objective Loss 0.262663 LR 0.000500 Time 0.021297 +2023-10-05 21:30:44,578 - Epoch: [103][ 720/ 1236] Overall Loss 0.262654 Objective Loss 0.262654 LR 0.000500 Time 0.021278 +2023-10-05 21:30:44,776 - Epoch: [103][ 730/ 1236] Overall Loss 0.262494 Objective Loss 0.262494 LR 0.000500 Time 0.021257 +2023-10-05 21:30:44,976 - Epoch: [103][ 740/ 1236] Overall Loss 0.262376 Objective Loss 0.262376 LR 0.000500 Time 0.021240 +2023-10-05 21:30:45,173 - Epoch: [103][ 750/ 1236] Overall Loss 0.262284 Objective Loss 0.262284 LR 0.000500 Time 0.021220 +2023-10-05 21:30:45,374 - Epoch: [103][ 760/ 1236] Overall Loss 0.262327 Objective Loss 0.262327 LR 0.000500 Time 0.021204 +2023-10-05 21:30:45,572 - Epoch: [103][ 770/ 1236] Overall Loss 0.262500 Objective Loss 0.262500 LR 0.000500 Time 0.021185 +2023-10-05 21:30:45,771 - Epoch: [103][ 780/ 1236] Overall Loss 0.262285 Objective Loss 0.262285 LR 0.000500 Time 0.021170 +2023-10-05 21:30:45,969 - Epoch: [103][ 790/ 1236] Overall Loss 0.262174 Objective Loss 0.262174 LR 0.000500 Time 0.021152 +2023-10-05 21:30:46,169 - Epoch: [103][ 800/ 1236] Overall Loss 0.262036 Objective Loss 0.262036 LR 0.000500 Time 0.021136 +2023-10-05 21:30:46,367 - Epoch: [103][ 810/ 1236] Overall Loss 0.261928 Objective Loss 0.261928 LR 0.000500 Time 0.021120 +2023-10-05 21:30:46,567 - Epoch: [103][ 820/ 1236] Overall Loss 0.262164 Objective Loss 0.262164 LR 0.000500 Time 0.021106 +2023-10-05 21:30:46,766 - Epoch: [103][ 830/ 1236] Overall Loss 0.262378 Objective Loss 0.262378 LR 0.000500 Time 0.021091 +2023-10-05 21:30:46,967 - Epoch: [103][ 840/ 1236] Overall Loss 0.261997 Objective Loss 0.261997 LR 0.000500 Time 0.021079 +2023-10-05 21:30:47,165 - Epoch: [103][ 850/ 1236] Overall Loss 0.261854 Objective Loss 0.261854 LR 0.000500 Time 0.021064 +2023-10-05 21:30:47,365 - Epoch: [103][ 860/ 1236] Overall Loss 0.262058 Objective Loss 0.262058 LR 0.000500 Time 0.021051 +2023-10-05 21:30:47,564 - Epoch: [103][ 870/ 1236] Overall Loss 0.262409 Objective Loss 0.262409 LR 0.000500 Time 0.021037 +2023-10-05 21:30:47,765 - Epoch: [103][ 880/ 1236] Overall Loss 0.262543 Objective Loss 0.262543 LR 0.000500 Time 0.021026 +2023-10-05 21:30:47,963 - Epoch: [103][ 890/ 1236] Overall Loss 0.262737 Objective Loss 0.262737 LR 0.000500 Time 0.021013 +2023-10-05 21:30:48,163 - Epoch: [103][ 900/ 1236] Overall Loss 0.262559 Objective Loss 0.262559 LR 0.000500 Time 0.021001 +2023-10-05 21:30:48,362 - Epoch: [103][ 910/ 1236] Overall Loss 0.262570 Objective Loss 0.262570 LR 0.000500 Time 0.020988 +2023-10-05 21:30:48,562 - Epoch: [103][ 920/ 1236] Overall Loss 0.262795 Objective Loss 0.262795 LR 0.000500 Time 0.020978 +2023-10-05 21:30:48,761 - Epoch: [103][ 930/ 1236] Overall Loss 0.263088 Objective Loss 0.263088 LR 0.000500 Time 0.020965 +2023-10-05 21:30:48,961 - Epoch: [103][ 940/ 1236] Overall Loss 0.262993 Objective Loss 0.262993 LR 0.000500 Time 0.020955 +2023-10-05 21:30:49,160 - Epoch: [103][ 950/ 1236] Overall Loss 0.263363 Objective Loss 0.263363 LR 0.000500 Time 0.020943 +2023-10-05 21:30:49,361 - Epoch: [103][ 960/ 1236] Overall Loss 0.263430 Objective Loss 0.263430 LR 0.000500 Time 0.020934 +2023-10-05 21:30:49,559 - Epoch: [103][ 970/ 1236] Overall Loss 0.263455 Objective Loss 0.263455 LR 0.000500 Time 0.020922 +2023-10-05 21:30:49,759 - Epoch: [103][ 980/ 1236] Overall Loss 0.263419 Objective Loss 0.263419 LR 0.000500 Time 0.020913 +2023-10-05 21:30:49,958 - Epoch: [103][ 990/ 1236] Overall Loss 0.263734 Objective Loss 0.263734 LR 0.000500 Time 0.020902 +2023-10-05 21:30:50,158 - Epoch: [103][ 1000/ 1236] Overall Loss 0.263572 Objective Loss 0.263572 LR 0.000500 Time 0.020893 +2023-10-05 21:30:50,357 - Epoch: [103][ 1010/ 1236] Overall Loss 0.263819 Objective Loss 0.263819 LR 0.000500 Time 0.020883 +2023-10-05 21:30:50,557 - Epoch: [103][ 1020/ 1236] Overall Loss 0.263823 Objective Loss 0.263823 LR 0.000500 Time 0.020874 +2023-10-05 21:30:50,756 - Epoch: [103][ 1030/ 1236] Overall Loss 0.264200 Objective Loss 0.264200 LR 0.000500 Time 0.020864 +2023-10-05 21:30:50,961 - Epoch: [103][ 1040/ 1236] Overall Loss 0.264213 Objective Loss 0.264213 LR 0.000500 Time 0.020860 +2023-10-05 21:30:51,159 - Epoch: [103][ 1050/ 1236] Overall Loss 0.264334 Objective Loss 0.264334 LR 0.000500 Time 0.020850 +2023-10-05 21:30:51,360 - Epoch: [103][ 1060/ 1236] Overall Loss 0.264238 Objective Loss 0.264238 LR 0.000500 Time 0.020842 +2023-10-05 21:30:51,558 - Epoch: [103][ 1070/ 1236] Overall Loss 0.264514 Objective Loss 0.264514 LR 0.000500 Time 0.020833 +2023-10-05 21:30:51,758 - Epoch: [103][ 1080/ 1236] Overall Loss 0.264660 Objective Loss 0.264660 LR 0.000500 Time 0.020825 +2023-10-05 21:30:51,957 - Epoch: [103][ 1090/ 1236] Overall Loss 0.264892 Objective Loss 0.264892 LR 0.000500 Time 0.020816 +2023-10-05 21:30:52,157 - Epoch: [103][ 1100/ 1236] Overall Loss 0.264674 Objective Loss 0.264674 LR 0.000500 Time 0.020808 +2023-10-05 21:30:52,357 - Epoch: [103][ 1110/ 1236] Overall Loss 0.264662 Objective Loss 0.264662 LR 0.000500 Time 0.020800 +2023-10-05 21:30:52,556 - Epoch: [103][ 1120/ 1236] Overall Loss 0.264669 Objective Loss 0.264669 LR 0.000500 Time 0.020792 +2023-10-05 21:30:52,755 - Epoch: [103][ 1130/ 1236] Overall Loss 0.264771 Objective Loss 0.264771 LR 0.000500 Time 0.020784 +2023-10-05 21:30:52,955 - Epoch: [103][ 1140/ 1236] Overall Loss 0.264784 Objective Loss 0.264784 LR 0.000500 Time 0.020777 +2023-10-05 21:30:53,154 - Epoch: [103][ 1150/ 1236] Overall Loss 0.264888 Objective Loss 0.264888 LR 0.000500 Time 0.020769 +2023-10-05 21:30:53,354 - Epoch: [103][ 1160/ 1236] Overall Loss 0.265021 Objective Loss 0.265021 LR 0.000500 Time 0.020762 +2023-10-05 21:30:53,553 - Epoch: [103][ 1170/ 1236] Overall Loss 0.264868 Objective Loss 0.264868 LR 0.000500 Time 0.020755 +2023-10-05 21:30:53,753 - Epoch: [103][ 1180/ 1236] Overall Loss 0.264793 Objective Loss 0.264793 LR 0.000500 Time 0.020748 +2023-10-05 21:30:53,952 - Epoch: [103][ 1190/ 1236] Overall Loss 0.265083 Objective Loss 0.265083 LR 0.000500 Time 0.020740 +2023-10-05 21:30:54,152 - Epoch: [103][ 1200/ 1236] Overall Loss 0.265418 Objective Loss 0.265418 LR 0.000500 Time 0.020734 +2023-10-05 21:30:54,351 - Epoch: [103][ 1210/ 1236] Overall Loss 0.265439 Objective Loss 0.265439 LR 0.000500 Time 0.020727 +2023-10-05 21:30:54,551 - Epoch: [103][ 1220/ 1236] Overall Loss 0.265346 Objective Loss 0.265346 LR 0.000500 Time 0.020721 +2023-10-05 21:30:54,805 - Epoch: [103][ 1230/ 1236] Overall Loss 0.265153 Objective Loss 0.265153 LR 0.000500 Time 0.020759 +2023-10-05 21:30:54,924 - Epoch: [103][ 1236/ 1236] Overall Loss 0.265510 Objective Loss 0.265510 Top1 84.114053 Top5 96.334012 LR 0.000500 Time 0.020754 +2023-10-05 21:30:55,059 - --- validate (epoch=103)----------- +2023-10-05 21:30:55,059 - 29943 samples (256 per mini-batch) +2023-10-05 21:30:55,517 - Epoch: [103][ 10/ 117] Loss 0.328938 Top1 81.484375 Top5 97.929688 +2023-10-05 21:30:55,666 - Epoch: [103][ 20/ 117] Loss 0.323267 Top1 82.832031 Top5 97.871094 +2023-10-05 21:30:55,816 - Epoch: [103][ 30/ 117] Loss 0.329835 Top1 83.072917 Top5 97.734375 +2023-10-05 21:30:55,964 - Epoch: [103][ 40/ 117] Loss 0.330146 Top1 82.910156 Top5 97.792969 +2023-10-05 21:30:56,113 - Epoch: [103][ 50/ 117] Loss 0.323776 Top1 83.156250 Top5 97.914062 +2023-10-05 21:30:56,262 - Epoch: [103][ 60/ 117] Loss 0.325467 Top1 83.131510 Top5 97.858073 +2023-10-05 21:30:56,411 - Epoch: [103][ 70/ 117] Loss 0.327916 Top1 83.169643 Top5 97.812500 +2023-10-05 21:30:56,559 - Epoch: [103][ 80/ 117] Loss 0.329535 Top1 83.149414 Top5 97.836914 +2023-10-05 21:30:56,709 - Epoch: [103][ 90/ 117] Loss 0.331142 Top1 83.237847 Top5 97.816840 +2023-10-05 21:30:56,856 - Epoch: [103][ 100/ 117] Loss 0.330755 Top1 83.281250 Top5 97.824219 +2023-10-05 21:30:57,012 - Epoch: [103][ 110/ 117] Loss 0.333159 Top1 83.306108 Top5 97.798295 +2023-10-05 21:30:57,098 - Epoch: [103][ 117/ 117] Loss 0.332574 Top1 83.278229 Top5 97.809171 +2023-10-05 21:30:57,217 - ==> Top1: 83.278 Top5: 97.809 Loss: 0.333 + +2023-10-05 21:30:57,217 - ==> Confusion: +[[ 909 5 5 0 14 1 0 0 7 78 1 0 1 3 6 1 3 2 1 0 13] + [ 0 1043 1 0 5 22 2 21 0 0 3 4 0 0 1 4 7 1 12 1 4] + [ 1 0 941 9 2 0 45 8 0 0 6 2 7 2 2 4 1 3 8 4 11] + [ 2 1 27 953 0 4 0 1 1 0 14 1 4 2 23 5 2 4 23 2 20] + [ 25 5 4 1 958 7 0 1 0 9 1 2 1 1 13 3 10 2 0 1 6] + [ 4 43 1 2 2 971 2 26 2 1 4 10 3 12 7 2 2 0 3 3 16] + [ 0 2 24 0 0 0 1127 6 0 0 5 1 1 1 1 8 1 1 3 2 8] + [ 0 24 21 1 1 26 5 1060 1 4 5 9 4 0 0 3 1 0 37 5 11] + [ 17 2 0 1 1 3 2 0 965 36 10 3 4 6 18 5 2 1 9 1 3] + [ 89 1 0 0 5 2 1 0 32 939 1 1 0 27 5 5 0 0 0 2 9] + [ 0 5 9 3 4 0 9 3 9 2 968 3 1 9 5 2 1 0 8 1 11] + [ 2 0 1 0 0 12 0 6 0 1 1 950 27 6 0 3 2 15 0 6 3] + [ 0 0 5 3 2 3 2 1 1 0 1 33 976 3 2 4 1 21 1 3 6] + [ 2 0 1 0 2 10 1 1 12 8 8 8 2 1040 2 5 0 3 0 4 10] + [ 11 3 3 13 3 0 0 0 27 2 2 2 2 2 1001 0 2 2 14 0 12] + [ 1 2 2 1 3 1 1 0 0 0 0 9 6 1 0 1067 15 9 0 9 7] + [ 0 13 1 0 2 3 1 2 1 0 0 6 2 1 4 11 1099 0 0 6 9] + [ 0 2 0 0 0 1 3 0 0 0 0 1 20 0 1 5 1 997 1 2 4] + [ 3 10 9 18 0 2 2 19 4 0 4 0 6 0 8 0 0 0 970 1 12] + [ 1 5 1 1 1 2 18 8 0 0 2 16 5 1 0 6 9 1 1 1066 8] + [ 124 223 150 63 82 128 72 100 105 87 193 156 361 295 137 73 189 73 153 205 4936]] + +2023-10-05 21:30:57,218 - ==> Best [Top1: 83.278 Top5: 97.809 Sparsity:0.00 Params: 148928 on epoch: 103] +2023-10-05 21:30:57,218 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:30:57,225 - + +2023-10-05 21:30:57,226 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:30:58,341 - Epoch: [104][ 10/ 1236] Overall Loss 0.246642 Objective Loss 0.246642 LR 0.000500 Time 0.111491 +2023-10-05 21:30:58,541 - Epoch: [104][ 20/ 1236] Overall Loss 0.253632 Objective Loss 0.253632 LR 0.000500 Time 0.065745 +2023-10-05 21:30:58,740 - Epoch: [104][ 30/ 1236] Overall Loss 0.257188 Objective Loss 0.257188 LR 0.000500 Time 0.050424 +2023-10-05 21:30:58,940 - Epoch: [104][ 40/ 1236] Overall Loss 0.260901 Objective Loss 0.260901 LR 0.000500 Time 0.042814 +2023-10-05 21:30:59,138 - Epoch: [104][ 50/ 1236] Overall Loss 0.256858 Objective Loss 0.256858 LR 0.000500 Time 0.038206 +2023-10-05 21:30:59,338 - Epoch: [104][ 60/ 1236] Overall Loss 0.256253 Objective Loss 0.256253 LR 0.000500 Time 0.035170 +2023-10-05 21:30:59,537 - Epoch: [104][ 70/ 1236] Overall Loss 0.257974 Objective Loss 0.257974 LR 0.000500 Time 0.032982 +2023-10-05 21:30:59,737 - Epoch: [104][ 80/ 1236] Overall Loss 0.257940 Objective Loss 0.257940 LR 0.000500 Time 0.031361 +2023-10-05 21:30:59,936 - Epoch: [104][ 90/ 1236] Overall Loss 0.260227 Objective Loss 0.260227 LR 0.000500 Time 0.030076 +2023-10-05 21:31:00,137 - Epoch: [104][ 100/ 1236] Overall Loss 0.260504 Objective Loss 0.260504 LR 0.000500 Time 0.029074 +2023-10-05 21:31:00,335 - Epoch: [104][ 110/ 1236] Overall Loss 0.258196 Objective Loss 0.258196 LR 0.000500 Time 0.028229 +2023-10-05 21:31:00,535 - Epoch: [104][ 120/ 1236] Overall Loss 0.259021 Objective Loss 0.259021 LR 0.000500 Time 0.027542 +2023-10-05 21:31:00,733 - Epoch: [104][ 130/ 1236] Overall Loss 0.260496 Objective Loss 0.260496 LR 0.000500 Time 0.026945 +2023-10-05 21:31:00,933 - Epoch: [104][ 140/ 1236] Overall Loss 0.258514 Objective Loss 0.258514 LR 0.000500 Time 0.026450 +2023-10-05 21:31:01,131 - Epoch: [104][ 150/ 1236] Overall Loss 0.257636 Objective Loss 0.257636 LR 0.000500 Time 0.026003 +2023-10-05 21:31:01,332 - Epoch: [104][ 160/ 1236] Overall Loss 0.259683 Objective Loss 0.259683 LR 0.000500 Time 0.025628 +2023-10-05 21:31:01,530 - Epoch: [104][ 170/ 1236] Overall Loss 0.260323 Objective Loss 0.260323 LR 0.000500 Time 0.025284 +2023-10-05 21:31:01,730 - Epoch: [104][ 180/ 1236] Overall Loss 0.260552 Objective Loss 0.260552 LR 0.000500 Time 0.024989 +2023-10-05 21:31:01,928 - Epoch: [104][ 190/ 1236] Overall Loss 0.260723 Objective Loss 0.260723 LR 0.000500 Time 0.024716 +2023-10-05 21:31:02,129 - Epoch: [104][ 200/ 1236] Overall Loss 0.260373 Objective Loss 0.260373 LR 0.000500 Time 0.024484 +2023-10-05 21:31:02,329 - Epoch: [104][ 210/ 1236] Overall Loss 0.261368 Objective Loss 0.261368 LR 0.000500 Time 0.024265 +2023-10-05 21:31:02,530 - Epoch: [104][ 220/ 1236] Overall Loss 0.261515 Objective Loss 0.261515 LR 0.000500 Time 0.024076 +2023-10-05 21:31:02,729 - Epoch: [104][ 230/ 1236] Overall Loss 0.261915 Objective Loss 0.261915 LR 0.000500 Time 0.023893 +2023-10-05 21:31:02,930 - Epoch: [104][ 240/ 1236] Overall Loss 0.262146 Objective Loss 0.262146 LR 0.000500 Time 0.023734 +2023-10-05 21:31:03,129 - Epoch: [104][ 250/ 1236] Overall Loss 0.262509 Objective Loss 0.262509 LR 0.000500 Time 0.023579 +2023-10-05 21:31:03,330 - Epoch: [104][ 260/ 1236] Overall Loss 0.261329 Objective Loss 0.261329 LR 0.000500 Time 0.023445 +2023-10-05 21:31:03,530 - Epoch: [104][ 270/ 1236] Overall Loss 0.260712 Objective Loss 0.260712 LR 0.000500 Time 0.023313 +2023-10-05 21:31:03,731 - Epoch: [104][ 280/ 1236] Overall Loss 0.259867 Objective Loss 0.259867 LR 0.000500 Time 0.023198 +2023-10-05 21:31:03,930 - Epoch: [104][ 290/ 1236] Overall Loss 0.260315 Objective Loss 0.260315 LR 0.000500 Time 0.023083 +2023-10-05 21:31:04,131 - Epoch: [104][ 300/ 1236] Overall Loss 0.259866 Objective Loss 0.259866 LR 0.000500 Time 0.022984 +2023-10-05 21:31:04,331 - Epoch: [104][ 310/ 1236] Overall Loss 0.259749 Objective Loss 0.259749 LR 0.000500 Time 0.022884 +2023-10-05 21:31:04,532 - Epoch: [104][ 320/ 1236] Overall Loss 0.260804 Objective Loss 0.260804 LR 0.000500 Time 0.022798 +2023-10-05 21:31:04,732 - Epoch: [104][ 330/ 1236] Overall Loss 0.259876 Objective Loss 0.259876 LR 0.000500 Time 0.022711 +2023-10-05 21:31:04,934 - Epoch: [104][ 340/ 1236] Overall Loss 0.259188 Objective Loss 0.259188 LR 0.000500 Time 0.022636 +2023-10-05 21:31:05,133 - Epoch: [104][ 350/ 1236] Overall Loss 0.258713 Objective Loss 0.258713 LR 0.000500 Time 0.022557 +2023-10-05 21:31:05,335 - Epoch: [104][ 360/ 1236] Overall Loss 0.258237 Objective Loss 0.258237 LR 0.000500 Time 0.022491 +2023-10-05 21:31:05,539 - Epoch: [104][ 370/ 1236] Overall Loss 0.258019 Objective Loss 0.258019 LR 0.000500 Time 0.022434 +2023-10-05 21:31:05,753 - Epoch: [104][ 380/ 1236] Overall Loss 0.259061 Objective Loss 0.259061 LR 0.000500 Time 0.022404 +2023-10-05 21:31:05,961 - Epoch: [104][ 390/ 1236] Overall Loss 0.258814 Objective Loss 0.258814 LR 0.000500 Time 0.022364 +2023-10-05 21:31:06,174 - Epoch: [104][ 400/ 1236] Overall Loss 0.259195 Objective Loss 0.259195 LR 0.000500 Time 0.022336 +2023-10-05 21:31:06,383 - Epoch: [104][ 410/ 1236] Overall Loss 0.259838 Objective Loss 0.259838 LR 0.000500 Time 0.022300 +2023-10-05 21:31:06,596 - Epoch: [104][ 420/ 1236] Overall Loss 0.260408 Objective Loss 0.260408 LR 0.000500 Time 0.022275 +2023-10-05 21:31:06,804 - Epoch: [104][ 430/ 1236] Overall Loss 0.260658 Objective Loss 0.260658 LR 0.000500 Time 0.022241 +2023-10-05 21:31:07,017 - Epoch: [104][ 440/ 1236] Overall Loss 0.260327 Objective Loss 0.260327 LR 0.000500 Time 0.022219 +2023-10-05 21:31:07,226 - Epoch: [104][ 450/ 1236] Overall Loss 0.260421 Objective Loss 0.260421 LR 0.000500 Time 0.022187 +2023-10-05 21:31:07,438 - Epoch: [104][ 460/ 1236] Overall Loss 0.260410 Objective Loss 0.260410 LR 0.000500 Time 0.022166 +2023-10-05 21:31:07,647 - Epoch: [104][ 470/ 1236] Overall Loss 0.260794 Objective Loss 0.260794 LR 0.000500 Time 0.022137 +2023-10-05 21:31:07,860 - Epoch: [104][ 480/ 1236] Overall Loss 0.260582 Objective Loss 0.260582 LR 0.000500 Time 0.022119 +2023-10-05 21:31:08,068 - Epoch: [104][ 490/ 1236] Overall Loss 0.260905 Objective Loss 0.260905 LR 0.000500 Time 0.022093 +2023-10-05 21:31:08,281 - Epoch: [104][ 500/ 1236] Overall Loss 0.260807 Objective Loss 0.260807 LR 0.000500 Time 0.022076 +2023-10-05 21:31:08,489 - Epoch: [104][ 510/ 1236] Overall Loss 0.261456 Objective Loss 0.261456 LR 0.000500 Time 0.022051 +2023-10-05 21:31:08,702 - Epoch: [104][ 520/ 1236] Overall Loss 0.261961 Objective Loss 0.261961 LR 0.000500 Time 0.022035 +2023-10-05 21:31:08,910 - Epoch: [104][ 530/ 1236] Overall Loss 0.261482 Objective Loss 0.261482 LR 0.000500 Time 0.022012 +2023-10-05 21:31:09,123 - Epoch: [104][ 540/ 1236] Overall Loss 0.261522 Objective Loss 0.261522 LR 0.000500 Time 0.021998 +2023-10-05 21:31:09,332 - Epoch: [104][ 550/ 1236] Overall Loss 0.260924 Objective Loss 0.260924 LR 0.000500 Time 0.021977 +2023-10-05 21:31:09,545 - Epoch: [104][ 560/ 1236] Overall Loss 0.261409 Objective Loss 0.261409 LR 0.000500 Time 0.021965 +2023-10-05 21:31:09,753 - Epoch: [104][ 570/ 1236] Overall Loss 0.262013 Objective Loss 0.262013 LR 0.000500 Time 0.021944 +2023-10-05 21:31:09,967 - Epoch: [104][ 580/ 1236] Overall Loss 0.261843 Objective Loss 0.261843 LR 0.000500 Time 0.021933 +2023-10-05 21:31:10,175 - Epoch: [104][ 590/ 1236] Overall Loss 0.261770 Objective Loss 0.261770 LR 0.000500 Time 0.021914 +2023-10-05 21:31:10,388 - Epoch: [104][ 600/ 1236] Overall Loss 0.261690 Objective Loss 0.261690 LR 0.000500 Time 0.021903 +2023-10-05 21:31:10,598 - Epoch: [104][ 610/ 1236] Overall Loss 0.261458 Objective Loss 0.261458 LR 0.000500 Time 0.021887 +2023-10-05 21:31:10,812 - Epoch: [104][ 620/ 1236] Overall Loss 0.261926 Objective Loss 0.261926 LR 0.000500 Time 0.021879 +2023-10-05 21:31:11,022 - Epoch: [104][ 630/ 1236] Overall Loss 0.262064 Objective Loss 0.262064 LR 0.000500 Time 0.021864 +2023-10-05 21:31:11,236 - Epoch: [104][ 640/ 1236] Overall Loss 0.262490 Objective Loss 0.262490 LR 0.000500 Time 0.021857 +2023-10-05 21:31:11,446 - Epoch: [104][ 650/ 1236] Overall Loss 0.261903 Objective Loss 0.261903 LR 0.000500 Time 0.021843 +2023-10-05 21:31:11,660 - Epoch: [104][ 660/ 1236] Overall Loss 0.261280 Objective Loss 0.261280 LR 0.000500 Time 0.021836 +2023-10-05 21:31:11,870 - Epoch: [104][ 670/ 1236] Overall Loss 0.261949 Objective Loss 0.261949 LR 0.000500 Time 0.021823 +2023-10-05 21:31:12,084 - Epoch: [104][ 680/ 1236] Overall Loss 0.261942 Objective Loss 0.261942 LR 0.000500 Time 0.021817 +2023-10-05 21:31:12,294 - Epoch: [104][ 690/ 1236] Overall Loss 0.261503 Objective Loss 0.261503 LR 0.000500 Time 0.021804 +2023-10-05 21:31:12,508 - Epoch: [104][ 700/ 1236] Overall Loss 0.261671 Objective Loss 0.261671 LR 0.000500 Time 0.021798 +2023-10-05 21:31:12,718 - Epoch: [104][ 710/ 1236] Overall Loss 0.261682 Objective Loss 0.261682 LR 0.000500 Time 0.021786 +2023-10-05 21:31:12,932 - Epoch: [104][ 720/ 1236] Overall Loss 0.261215 Objective Loss 0.261215 LR 0.000500 Time 0.021780 +2023-10-05 21:31:13,142 - Epoch: [104][ 730/ 1236] Overall Loss 0.261283 Objective Loss 0.261283 LR 0.000500 Time 0.021768 +2023-10-05 21:31:13,356 - Epoch: [104][ 740/ 1236] Overall Loss 0.260959 Objective Loss 0.260959 LR 0.000500 Time 0.021764 +2023-10-05 21:31:13,566 - Epoch: [104][ 750/ 1236] Overall Loss 0.260920 Objective Loss 0.260920 LR 0.000500 Time 0.021753 +2023-10-05 21:31:13,780 - Epoch: [104][ 760/ 1236] Overall Loss 0.260819 Objective Loss 0.260819 LR 0.000500 Time 0.021748 +2023-10-05 21:31:13,990 - Epoch: [104][ 770/ 1236] Overall Loss 0.261058 Objective Loss 0.261058 LR 0.000500 Time 0.021737 +2023-10-05 21:31:14,205 - Epoch: [104][ 780/ 1236] Overall Loss 0.261066 Objective Loss 0.261066 LR 0.000500 Time 0.021733 +2023-10-05 21:31:14,414 - Epoch: [104][ 790/ 1236] Overall Loss 0.261120 Objective Loss 0.261120 LR 0.000500 Time 0.021723 +2023-10-05 21:31:14,629 - Epoch: [104][ 800/ 1236] Overall Loss 0.261081 Objective Loss 0.261081 LR 0.000500 Time 0.021719 +2023-10-05 21:31:14,838 - Epoch: [104][ 810/ 1236] Overall Loss 0.261021 Objective Loss 0.261021 LR 0.000500 Time 0.021710 +2023-10-05 21:31:15,053 - Epoch: [104][ 820/ 1236] Overall Loss 0.261026 Objective Loss 0.261026 LR 0.000500 Time 0.021706 +2023-10-05 21:31:15,263 - Epoch: [104][ 830/ 1236] Overall Loss 0.261018 Objective Loss 0.261018 LR 0.000500 Time 0.021697 +2023-10-05 21:31:15,477 - Epoch: [104][ 840/ 1236] Overall Loss 0.260956 Objective Loss 0.260956 LR 0.000500 Time 0.021693 +2023-10-05 21:31:15,687 - Epoch: [104][ 850/ 1236] Overall Loss 0.260923 Objective Loss 0.260923 LR 0.000500 Time 0.021685 +2023-10-05 21:31:15,901 - Epoch: [104][ 860/ 1236] Overall Loss 0.261091 Objective Loss 0.261091 LR 0.000500 Time 0.021681 +2023-10-05 21:31:16,111 - Epoch: [104][ 870/ 1236] Overall Loss 0.261032 Objective Loss 0.261032 LR 0.000500 Time 0.021673 +2023-10-05 21:31:16,325 - Epoch: [104][ 880/ 1236] Overall Loss 0.261194 Objective Loss 0.261194 LR 0.000500 Time 0.021670 +2023-10-05 21:31:16,535 - Epoch: [104][ 890/ 1236] Overall Loss 0.261395 Objective Loss 0.261395 LR 0.000500 Time 0.021662 +2023-10-05 21:31:16,749 - Epoch: [104][ 900/ 1236] Overall Loss 0.261380 Objective Loss 0.261380 LR 0.000500 Time 0.021659 +2023-10-05 21:31:16,959 - Epoch: [104][ 910/ 1236] Overall Loss 0.261275 Objective Loss 0.261275 LR 0.000500 Time 0.021651 +2023-10-05 21:31:17,174 - Epoch: [104][ 920/ 1236] Overall Loss 0.260995 Objective Loss 0.260995 LR 0.000500 Time 0.021648 +2023-10-05 21:31:17,383 - Epoch: [104][ 930/ 1236] Overall Loss 0.260957 Objective Loss 0.260957 LR 0.000500 Time 0.021641 +2023-10-05 21:31:17,598 - Epoch: [104][ 940/ 1236] Overall Loss 0.261002 Objective Loss 0.261002 LR 0.000500 Time 0.021638 +2023-10-05 21:31:17,808 - Epoch: [104][ 950/ 1236] Overall Loss 0.260836 Objective Loss 0.260836 LR 0.000500 Time 0.021631 +2023-10-05 21:31:18,022 - Epoch: [104][ 960/ 1236] Overall Loss 0.260921 Objective Loss 0.260921 LR 0.000500 Time 0.021629 +2023-10-05 21:31:18,232 - Epoch: [104][ 970/ 1236] Overall Loss 0.260816 Objective Loss 0.260816 LR 0.000500 Time 0.021622 +2023-10-05 21:31:18,447 - Epoch: [104][ 980/ 1236] Overall Loss 0.260817 Objective Loss 0.260817 LR 0.000500 Time 0.021620 +2023-10-05 21:31:18,656 - Epoch: [104][ 990/ 1236] Overall Loss 0.260653 Objective Loss 0.260653 LR 0.000500 Time 0.021613 +2023-10-05 21:31:18,871 - Epoch: [104][ 1000/ 1236] Overall Loss 0.260845 Objective Loss 0.260845 LR 0.000500 Time 0.021611 +2023-10-05 21:31:19,081 - Epoch: [104][ 1010/ 1236] Overall Loss 0.260761 Objective Loss 0.260761 LR 0.000500 Time 0.021604 +2023-10-05 21:31:19,295 - Epoch: [104][ 1020/ 1236] Overall Loss 0.260695 Objective Loss 0.260695 LR 0.000500 Time 0.021603 +2023-10-05 21:31:19,505 - Epoch: [104][ 1030/ 1236] Overall Loss 0.260389 Objective Loss 0.260389 LR 0.000500 Time 0.021596 +2023-10-05 21:31:19,719 - Epoch: [104][ 1040/ 1236] Overall Loss 0.260305 Objective Loss 0.260305 LR 0.000500 Time 0.021594 +2023-10-05 21:31:19,929 - Epoch: [104][ 1050/ 1236] Overall Loss 0.260434 Objective Loss 0.260434 LR 0.000500 Time 0.021588 +2023-10-05 21:31:20,136 - Epoch: [104][ 1060/ 1236] Overall Loss 0.260887 Objective Loss 0.260887 LR 0.000500 Time 0.021580 +2023-10-05 21:31:20,340 - Epoch: [104][ 1070/ 1236] Overall Loss 0.260824 Objective Loss 0.260824 LR 0.000500 Time 0.021568 +2023-10-05 21:31:20,544 - Epoch: [104][ 1080/ 1236] Overall Loss 0.260842 Objective Loss 0.260842 LR 0.000500 Time 0.021557 +2023-10-05 21:31:20,747 - Epoch: [104][ 1090/ 1236] Overall Loss 0.261079 Objective Loss 0.261079 LR 0.000500 Time 0.021545 +2023-10-05 21:31:20,952 - Epoch: [104][ 1100/ 1236] Overall Loss 0.260972 Objective Loss 0.260972 LR 0.000500 Time 0.021535 +2023-10-05 21:31:21,155 - Epoch: [104][ 1110/ 1236] Overall Loss 0.261125 Objective Loss 0.261125 LR 0.000500 Time 0.021524 +2023-10-05 21:31:21,360 - Epoch: [104][ 1120/ 1236] Overall Loss 0.261329 Objective Loss 0.261329 LR 0.000500 Time 0.021514 +2023-10-05 21:31:21,563 - Epoch: [104][ 1130/ 1236] Overall Loss 0.261319 Objective Loss 0.261319 LR 0.000500 Time 0.021502 +2023-10-05 21:31:21,767 - Epoch: [104][ 1140/ 1236] Overall Loss 0.261344 Objective Loss 0.261344 LR 0.000500 Time 0.021493 +2023-10-05 21:31:21,970 - Epoch: [104][ 1150/ 1236] Overall Loss 0.261234 Objective Loss 0.261234 LR 0.000500 Time 0.021482 +2023-10-05 21:31:22,174 - Epoch: [104][ 1160/ 1236] Overall Loss 0.261391 Objective Loss 0.261391 LR 0.000500 Time 0.021473 +2023-10-05 21:31:22,378 - Epoch: [104][ 1170/ 1236] Overall Loss 0.261391 Objective Loss 0.261391 LR 0.000500 Time 0.021463 +2023-10-05 21:31:22,582 - Epoch: [104][ 1180/ 1236] Overall Loss 0.261643 Objective Loss 0.261643 LR 0.000500 Time 0.021454 +2023-10-05 21:31:22,786 - Epoch: [104][ 1190/ 1236] Overall Loss 0.261591 Objective Loss 0.261591 LR 0.000500 Time 0.021444 +2023-10-05 21:31:22,990 - Epoch: [104][ 1200/ 1236] Overall Loss 0.261446 Objective Loss 0.261446 LR 0.000500 Time 0.021436 +2023-10-05 21:31:23,194 - Epoch: [104][ 1210/ 1236] Overall Loss 0.261226 Objective Loss 0.261226 LR 0.000500 Time 0.021426 +2023-10-05 21:31:23,398 - Epoch: [104][ 1220/ 1236] Overall Loss 0.261134 Objective Loss 0.261134 LR 0.000500 Time 0.021418 +2023-10-05 21:31:23,656 - Epoch: [104][ 1230/ 1236] Overall Loss 0.261246 Objective Loss 0.261246 LR 0.000500 Time 0.021454 +2023-10-05 21:31:23,775 - Epoch: [104][ 1236/ 1236] Overall Loss 0.261330 Objective Loss 0.261330 Top1 84.521385 Top5 97.963340 LR 0.000500 Time 0.021445 +2023-10-05 21:31:23,905 - --- validate (epoch=104)----------- +2023-10-05 21:31:23,905 - 29943 samples (256 per mini-batch) +2023-10-05 21:31:24,372 - Epoch: [104][ 10/ 117] Loss 0.337998 Top1 83.007812 Top5 97.695312 +2023-10-05 21:31:24,524 - Epoch: [104][ 20/ 117] Loss 0.325910 Top1 83.164062 Top5 97.695312 +2023-10-05 21:31:24,673 - Epoch: [104][ 30/ 117] Loss 0.333944 Top1 82.630208 Top5 97.643229 +2023-10-05 21:31:24,823 - Epoch: [104][ 40/ 117] Loss 0.336025 Top1 82.568359 Top5 97.617188 +2023-10-05 21:31:24,972 - Epoch: [104][ 50/ 117] Loss 0.343055 Top1 82.648438 Top5 97.640625 +2023-10-05 21:31:25,123 - Epoch: [104][ 60/ 117] Loss 0.342577 Top1 82.688802 Top5 97.617188 +2023-10-05 21:31:25,271 - Epoch: [104][ 70/ 117] Loss 0.346384 Top1 82.606027 Top5 97.516741 +2023-10-05 21:31:25,429 - Epoch: [104][ 80/ 117] Loss 0.343708 Top1 82.612305 Top5 97.587891 +2023-10-05 21:31:25,583 - Epoch: [104][ 90/ 117] Loss 0.342417 Top1 82.604167 Top5 97.634549 +2023-10-05 21:31:25,740 - Epoch: [104][ 100/ 117] Loss 0.340934 Top1 82.605469 Top5 97.648438 +2023-10-05 21:31:25,905 - Epoch: [104][ 110/ 117] Loss 0.338074 Top1 82.780540 Top5 97.659801 +2023-10-05 21:31:25,991 - Epoch: [104][ 117/ 117] Loss 0.340963 Top1 82.773937 Top5 97.622149 +2023-10-05 21:31:26,131 - ==> Top1: 82.774 Top5: 97.622 Loss: 0.341 + +2023-10-05 21:31:26,132 - ==> Confusion: +[[ 922 3 3 3 14 3 0 1 7 69 1 1 0 4 4 1 3 0 1 0 10] + [ 2 1058 2 0 11 16 1 15 0 0 0 0 0 0 0 5 7 2 4 1 7] + [ 4 2 949 16 3 0 26 9 0 1 6 2 8 4 1 2 3 1 6 3 10] + [ 3 1 22 963 1 4 2 0 3 1 3 0 3 6 34 3 1 3 17 2 17] + [ 22 4 0 0 984 1 1 0 0 5 1 1 0 0 12 5 7 1 0 2 4] + [ 6 68 0 0 4 945 3 21 6 2 5 13 2 13 9 2 2 0 1 1 13] + [ 0 3 25 0 1 0 1126 9 0 0 4 3 1 1 1 4 0 0 1 7 5] + [ 4 31 17 0 4 25 3 1050 1 2 4 11 4 2 0 4 0 0 39 8 9] + [ 18 3 0 1 1 1 1 0 958 48 11 1 1 8 19 6 2 0 7 1 2] + [ 98 0 1 0 7 2 0 0 27 935 0 2 0 22 8 6 0 1 0 1 9] + [ 1 3 9 5 3 2 8 4 14 4 954 3 0 16 7 3 1 3 5 1 7] + [ 1 0 3 0 0 11 0 3 0 1 0 946 26 11 0 2 2 15 0 7 7] + [ 2 0 5 7 1 4 0 1 3 1 1 42 950 6 4 7 5 15 2 1 11] + [ 1 1 1 1 4 2 1 0 18 14 3 5 2 1053 3 2 0 0 0 2 6] + [ 13 1 3 10 8 0 0 0 21 3 0 1 1 1 1014 0 2 2 13 0 8] + [ 0 2 3 0 3 0 1 0 0 0 0 10 6 2 1 1071 16 7 1 7 4] + [ 0 15 3 0 9 5 0 0 1 0 1 3 0 0 2 11 1103 0 0 2 6] + [ 1 0 0 1 0 0 2 0 0 0 1 5 20 0 2 11 0 988 1 2 4] + [ 3 17 11 17 3 2 0 21 2 0 5 0 3 0 12 0 1 0 960 2 9] + [ 0 4 1 1 1 6 12 10 1 0 2 14 4 5 1 7 18 1 3 1053 8] + [ 148 245 159 72 141 119 42 102 124 125 190 165 295 316 184 66 217 61 159 172 4803]] + +2023-10-05 21:31:26,133 - ==> Best [Top1: 83.278 Top5: 97.809 Sparsity:0.00 Params: 148928 on epoch: 103] +2023-10-05 21:31:26,133 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:31:26,139 - + +2023-10-05 21:31:26,139 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:31:27,154 - Epoch: [105][ 10/ 1236] Overall Loss 0.237933 Objective Loss 0.237933 LR 0.000500 Time 0.101463 +2023-10-05 21:31:27,369 - Epoch: [105][ 20/ 1236] Overall Loss 0.247317 Objective Loss 0.247317 LR 0.000500 Time 0.061445 +2023-10-05 21:31:27,579 - Epoch: [105][ 30/ 1236] Overall Loss 0.242636 Objective Loss 0.242636 LR 0.000500 Time 0.047960 +2023-10-05 21:31:27,794 - Epoch: [105][ 40/ 1236] Overall Loss 0.244652 Objective Loss 0.244652 LR 0.000500 Time 0.041331 +2023-10-05 21:31:28,003 - Epoch: [105][ 50/ 1236] Overall Loss 0.246627 Objective Loss 0.246627 LR 0.000500 Time 0.037249 +2023-10-05 21:31:28,218 - Epoch: [105][ 60/ 1236] Overall Loss 0.245441 Objective Loss 0.245441 LR 0.000500 Time 0.034610 +2023-10-05 21:31:28,428 - Epoch: [105][ 70/ 1236] Overall Loss 0.245714 Objective Loss 0.245714 LR 0.000500 Time 0.032661 +2023-10-05 21:31:28,642 - Epoch: [105][ 80/ 1236] Overall Loss 0.245616 Objective Loss 0.245616 LR 0.000500 Time 0.031253 +2023-10-05 21:31:28,851 - Epoch: [105][ 90/ 1236] Overall Loss 0.244364 Objective Loss 0.244364 LR 0.000500 Time 0.030095 +2023-10-05 21:31:29,064 - Epoch: [105][ 100/ 1236] Overall Loss 0.248037 Objective Loss 0.248037 LR 0.000500 Time 0.029211 +2023-10-05 21:31:29,275 - Epoch: [105][ 110/ 1236] Overall Loss 0.246238 Objective Loss 0.246238 LR 0.000500 Time 0.028472 +2023-10-05 21:31:29,487 - Epoch: [105][ 120/ 1236] Overall Loss 0.245279 Objective Loss 0.245279 LR 0.000500 Time 0.027864 +2023-10-05 21:31:29,703 - Epoch: [105][ 130/ 1236] Overall Loss 0.247735 Objective Loss 0.247735 LR 0.000500 Time 0.027382 +2023-10-05 21:31:29,914 - Epoch: [105][ 140/ 1236] Overall Loss 0.248858 Objective Loss 0.248858 LR 0.000500 Time 0.026929 +2023-10-05 21:31:30,130 - Epoch: [105][ 150/ 1236] Overall Loss 0.247811 Objective Loss 0.247811 LR 0.000500 Time 0.026572 +2023-10-05 21:31:30,341 - Epoch: [105][ 160/ 1236] Overall Loss 0.247499 Objective Loss 0.247499 LR 0.000500 Time 0.026228 +2023-10-05 21:31:30,557 - Epoch: [105][ 170/ 1236] Overall Loss 0.249182 Objective Loss 0.249182 LR 0.000500 Time 0.025955 +2023-10-05 21:31:30,768 - Epoch: [105][ 180/ 1236] Overall Loss 0.249963 Objective Loss 0.249963 LR 0.000500 Time 0.025681 +2023-10-05 21:31:30,984 - Epoch: [105][ 190/ 1236] Overall Loss 0.251516 Objective Loss 0.251516 LR 0.000500 Time 0.025468 +2023-10-05 21:31:31,195 - Epoch: [105][ 200/ 1236] Overall Loss 0.250549 Objective Loss 0.250549 LR 0.000500 Time 0.025249 +2023-10-05 21:31:31,411 - Epoch: [105][ 210/ 1236] Overall Loss 0.250885 Objective Loss 0.250885 LR 0.000500 Time 0.025074 +2023-10-05 21:31:31,622 - Epoch: [105][ 220/ 1236] Overall Loss 0.250327 Objective Loss 0.250327 LR 0.000500 Time 0.024892 +2023-10-05 21:31:31,838 - Epoch: [105][ 230/ 1236] Overall Loss 0.250652 Objective Loss 0.250652 LR 0.000500 Time 0.024747 +2023-10-05 21:31:32,050 - Epoch: [105][ 240/ 1236] Overall Loss 0.251339 Objective Loss 0.251339 LR 0.000500 Time 0.024595 +2023-10-05 21:31:32,265 - Epoch: [105][ 250/ 1236] Overall Loss 0.251814 Objective Loss 0.251814 LR 0.000500 Time 0.024473 +2023-10-05 21:31:32,477 - Epoch: [105][ 260/ 1236] Overall Loss 0.251765 Objective Loss 0.251765 LR 0.000500 Time 0.024343 +2023-10-05 21:31:32,693 - Epoch: [105][ 270/ 1236] Overall Loss 0.252234 Objective Loss 0.252234 LR 0.000500 Time 0.024241 +2023-10-05 21:31:32,904 - Epoch: [105][ 280/ 1236] Overall Loss 0.252107 Objective Loss 0.252107 LR 0.000500 Time 0.024127 +2023-10-05 21:31:33,120 - Epoch: [105][ 290/ 1236] Overall Loss 0.253047 Objective Loss 0.253047 LR 0.000500 Time 0.024041 +2023-10-05 21:31:33,331 - Epoch: [105][ 300/ 1236] Overall Loss 0.254103 Objective Loss 0.254103 LR 0.000500 Time 0.023941 +2023-10-05 21:31:33,547 - Epoch: [105][ 310/ 1236] Overall Loss 0.255467 Objective Loss 0.255467 LR 0.000500 Time 0.023866 +2023-10-05 21:31:33,759 - Epoch: [105][ 320/ 1236] Overall Loss 0.256069 Objective Loss 0.256069 LR 0.000500 Time 0.023781 +2023-10-05 21:31:33,975 - Epoch: [105][ 330/ 1236] Overall Loss 0.256128 Objective Loss 0.256128 LR 0.000500 Time 0.023715 +2023-10-05 21:31:34,187 - Epoch: [105][ 340/ 1236] Overall Loss 0.256504 Objective Loss 0.256504 LR 0.000500 Time 0.023638 +2023-10-05 21:31:34,395 - Epoch: [105][ 350/ 1236] Overall Loss 0.256190 Objective Loss 0.256190 LR 0.000500 Time 0.023557 +2023-10-05 21:31:34,599 - Epoch: [105][ 360/ 1236] Overall Loss 0.256075 Objective Loss 0.256075 LR 0.000500 Time 0.023468 +2023-10-05 21:31:34,804 - Epoch: [105][ 370/ 1236] Overall Loss 0.255948 Objective Loss 0.255948 LR 0.000500 Time 0.023386 +2023-10-05 21:31:35,006 - Epoch: [105][ 380/ 1236] Overall Loss 0.256120 Objective Loss 0.256120 LR 0.000500 Time 0.023303 +2023-10-05 21:31:35,215 - Epoch: [105][ 390/ 1236] Overall Loss 0.256133 Objective Loss 0.256133 LR 0.000500 Time 0.023239 +2023-10-05 21:31:35,425 - Epoch: [105][ 400/ 1236] Overall Loss 0.256624 Objective Loss 0.256624 LR 0.000500 Time 0.023181 +2023-10-05 21:31:35,640 - Epoch: [105][ 410/ 1236] Overall Loss 0.256570 Objective Loss 0.256570 LR 0.000500 Time 0.023140 +2023-10-05 21:31:35,849 - Epoch: [105][ 420/ 1236] Overall Loss 0.256886 Objective Loss 0.256886 LR 0.000500 Time 0.023087 +2023-10-05 21:31:36,064 - Epoch: [105][ 430/ 1236] Overall Loss 0.256703 Objective Loss 0.256703 LR 0.000500 Time 0.023050 +2023-10-05 21:31:36,274 - Epoch: [105][ 440/ 1236] Overall Loss 0.257763 Objective Loss 0.257763 LR 0.000500 Time 0.023001 +2023-10-05 21:31:36,489 - Epoch: [105][ 450/ 1236] Overall Loss 0.258022 Objective Loss 0.258022 LR 0.000500 Time 0.022967 +2023-10-05 21:31:36,698 - Epoch: [105][ 460/ 1236] Overall Loss 0.256912 Objective Loss 0.256912 LR 0.000500 Time 0.022922 +2023-10-05 21:31:36,903 - Epoch: [105][ 470/ 1236] Overall Loss 0.257068 Objective Loss 0.257068 LR 0.000500 Time 0.022869 +2023-10-05 21:31:37,106 - Epoch: [105][ 480/ 1236] Overall Loss 0.257329 Objective Loss 0.257329 LR 0.000500 Time 0.022815 +2023-10-05 21:31:37,310 - Epoch: [105][ 490/ 1236] Overall Loss 0.257558 Objective Loss 0.257558 LR 0.000500 Time 0.022766 +2023-10-05 21:31:37,513 - Epoch: [105][ 500/ 1236] Overall Loss 0.257240 Objective Loss 0.257240 LR 0.000500 Time 0.022715 +2023-10-05 21:31:37,716 - Epoch: [105][ 510/ 1236] Overall Loss 0.257004 Objective Loss 0.257004 LR 0.000500 Time 0.022668 +2023-10-05 21:31:37,919 - Epoch: [105][ 520/ 1236] Overall Loss 0.257158 Objective Loss 0.257158 LR 0.000500 Time 0.022621 +2023-10-05 21:31:38,123 - Epoch: [105][ 530/ 1236] Overall Loss 0.256667 Objective Loss 0.256667 LR 0.000500 Time 0.022579 +2023-10-05 21:31:38,326 - Epoch: [105][ 540/ 1236] Overall Loss 0.256296 Objective Loss 0.256296 LR 0.000500 Time 0.022535 +2023-10-05 21:31:38,530 - Epoch: [105][ 550/ 1236] Overall Loss 0.255997 Objective Loss 0.255997 LR 0.000500 Time 0.022496 +2023-10-05 21:31:38,733 - Epoch: [105][ 560/ 1236] Overall Loss 0.256445 Objective Loss 0.256445 LR 0.000500 Time 0.022456 +2023-10-05 21:31:38,938 - Epoch: [105][ 570/ 1236] Overall Loss 0.256499 Objective Loss 0.256499 LR 0.000500 Time 0.022421 +2023-10-05 21:31:39,141 - Epoch: [105][ 580/ 1236] Overall Loss 0.256714 Objective Loss 0.256714 LR 0.000500 Time 0.022384 +2023-10-05 21:31:39,349 - Epoch: [105][ 590/ 1236] Overall Loss 0.256716 Objective Loss 0.256716 LR 0.000500 Time 0.022357 +2023-10-05 21:31:39,559 - Epoch: [105][ 600/ 1236] Overall Loss 0.256978 Objective Loss 0.256978 LR 0.000500 Time 0.022333 +2023-10-05 21:31:39,773 - Epoch: [105][ 610/ 1236] Overall Loss 0.256646 Objective Loss 0.256646 LR 0.000500 Time 0.022319 +2023-10-05 21:31:39,983 - Epoch: [105][ 620/ 1236] Overall Loss 0.256782 Objective Loss 0.256782 LR 0.000500 Time 0.022295 +2023-10-05 21:31:40,197 - Epoch: [105][ 630/ 1236] Overall Loss 0.257243 Objective Loss 0.257243 LR 0.000500 Time 0.022282 +2023-10-05 21:31:40,406 - Epoch: [105][ 640/ 1236] Overall Loss 0.257461 Objective Loss 0.257461 LR 0.000500 Time 0.022260 +2023-10-05 21:31:40,621 - Epoch: [105][ 650/ 1236] Overall Loss 0.257321 Objective Loss 0.257321 LR 0.000500 Time 0.022248 +2023-10-05 21:31:40,824 - Epoch: [105][ 660/ 1236] Overall Loss 0.257207 Objective Loss 0.257207 LR 0.000500 Time 0.022218 +2023-10-05 21:31:41,030 - Epoch: [105][ 670/ 1236] Overall Loss 0.257480 Objective Loss 0.257480 LR 0.000500 Time 0.022192 +2023-10-05 21:31:41,233 - Epoch: [105][ 680/ 1236] Overall Loss 0.257501 Objective Loss 0.257501 LR 0.000500 Time 0.022164 +2023-10-05 21:31:41,437 - Epoch: [105][ 690/ 1236] Overall Loss 0.257473 Objective Loss 0.257473 LR 0.000500 Time 0.022138 +2023-10-05 21:31:41,641 - Epoch: [105][ 700/ 1236] Overall Loss 0.257203 Objective Loss 0.257203 LR 0.000500 Time 0.022112 +2023-10-05 21:31:41,845 - Epoch: [105][ 710/ 1236] Overall Loss 0.256936 Objective Loss 0.256936 LR 0.000500 Time 0.022088 +2023-10-05 21:31:42,048 - Epoch: [105][ 720/ 1236] Overall Loss 0.256753 Objective Loss 0.256753 LR 0.000500 Time 0.022063 +2023-10-05 21:31:42,253 - Epoch: [105][ 730/ 1236] Overall Loss 0.256702 Objective Loss 0.256702 LR 0.000500 Time 0.022040 +2023-10-05 21:31:42,456 - Epoch: [105][ 740/ 1236] Overall Loss 0.256499 Objective Loss 0.256499 LR 0.000500 Time 0.022016 +2023-10-05 21:31:42,661 - Epoch: [105][ 750/ 1236] Overall Loss 0.256736 Objective Loss 0.256736 LR 0.000500 Time 0.021996 +2023-10-05 21:31:42,864 - Epoch: [105][ 760/ 1236] Overall Loss 0.256806 Objective Loss 0.256806 LR 0.000500 Time 0.021973 +2023-10-05 21:31:43,070 - Epoch: [105][ 770/ 1236] Overall Loss 0.256865 Objective Loss 0.256865 LR 0.000500 Time 0.021955 +2023-10-05 21:31:43,271 - Epoch: [105][ 780/ 1236] Overall Loss 0.257116 Objective Loss 0.257116 LR 0.000500 Time 0.021931 +2023-10-05 21:31:43,476 - Epoch: [105][ 790/ 1236] Overall Loss 0.257237 Objective Loss 0.257237 LR 0.000500 Time 0.021912 +2023-10-05 21:31:43,677 - Epoch: [105][ 800/ 1236] Overall Loss 0.256882 Objective Loss 0.256882 LR 0.000500 Time 0.021889 +2023-10-05 21:31:43,882 - Epoch: [105][ 810/ 1236] Overall Loss 0.257221 Objective Loss 0.257221 LR 0.000500 Time 0.021871 +2023-10-05 21:31:44,083 - Epoch: [105][ 820/ 1236] Overall Loss 0.257102 Objective Loss 0.257102 LR 0.000500 Time 0.021849 +2023-10-05 21:31:44,287 - Epoch: [105][ 830/ 1236] Overall Loss 0.257035 Objective Loss 0.257035 LR 0.000500 Time 0.021832 +2023-10-05 21:31:44,489 - Epoch: [105][ 840/ 1236] Overall Loss 0.256787 Objective Loss 0.256787 LR 0.000500 Time 0.021812 +2023-10-05 21:31:44,694 - Epoch: [105][ 850/ 1236] Overall Loss 0.256503 Objective Loss 0.256503 LR 0.000500 Time 0.021796 +2023-10-05 21:31:44,895 - Epoch: [105][ 860/ 1236] Overall Loss 0.256813 Objective Loss 0.256813 LR 0.000500 Time 0.021776 +2023-10-05 21:31:45,100 - Epoch: [105][ 870/ 1236] Overall Loss 0.256707 Objective Loss 0.256707 LR 0.000500 Time 0.021760 +2023-10-05 21:31:45,301 - Epoch: [105][ 880/ 1236] Overall Loss 0.256716 Objective Loss 0.256716 LR 0.000500 Time 0.021742 +2023-10-05 21:31:45,506 - Epoch: [105][ 890/ 1236] Overall Loss 0.256945 Objective Loss 0.256945 LR 0.000500 Time 0.021727 +2023-10-05 21:31:45,707 - Epoch: [105][ 900/ 1236] Overall Loss 0.257139 Objective Loss 0.257139 LR 0.000500 Time 0.021709 +2023-10-05 21:31:45,912 - Epoch: [105][ 910/ 1236] Overall Loss 0.256965 Objective Loss 0.256965 LR 0.000500 Time 0.021695 +2023-10-05 21:31:46,115 - Epoch: [105][ 920/ 1236] Overall Loss 0.256941 Objective Loss 0.256941 LR 0.000500 Time 0.021679 +2023-10-05 21:31:46,319 - Epoch: [105][ 930/ 1236] Overall Loss 0.257019 Objective Loss 0.257019 LR 0.000500 Time 0.021665 +2023-10-05 21:31:46,521 - Epoch: [105][ 940/ 1236] Overall Loss 0.257307 Objective Loss 0.257307 LR 0.000500 Time 0.021649 +2023-10-05 21:31:46,726 - Epoch: [105][ 950/ 1236] Overall Loss 0.257706 Objective Loss 0.257706 LR 0.000500 Time 0.021636 +2023-10-05 21:31:46,927 - Epoch: [105][ 960/ 1236] Overall Loss 0.257758 Objective Loss 0.257758 LR 0.000500 Time 0.021620 +2023-10-05 21:31:47,131 - Epoch: [105][ 970/ 1236] Overall Loss 0.257756 Objective Loss 0.257756 LR 0.000500 Time 0.021607 +2023-10-05 21:31:47,333 - Epoch: [105][ 980/ 1236] Overall Loss 0.258021 Objective Loss 0.258021 LR 0.000500 Time 0.021593 +2023-10-05 21:31:47,537 - Epoch: [105][ 990/ 1236] Overall Loss 0.258320 Objective Loss 0.258320 LR 0.000500 Time 0.021581 +2023-10-05 21:31:47,739 - Epoch: [105][ 1000/ 1236] Overall Loss 0.258546 Objective Loss 0.258546 LR 0.000500 Time 0.021566 +2023-10-05 21:31:47,944 - Epoch: [105][ 1010/ 1236] Overall Loss 0.258221 Objective Loss 0.258221 LR 0.000500 Time 0.021555 +2023-10-05 21:31:48,145 - Epoch: [105][ 1020/ 1236] Overall Loss 0.258141 Objective Loss 0.258141 LR 0.000500 Time 0.021541 +2023-10-05 21:31:48,350 - Epoch: [105][ 1030/ 1236] Overall Loss 0.258075 Objective Loss 0.258075 LR 0.000500 Time 0.021530 +2023-10-05 21:31:48,551 - Epoch: [105][ 1040/ 1236] Overall Loss 0.258153 Objective Loss 0.258153 LR 0.000500 Time 0.021516 +2023-10-05 21:31:48,756 - Epoch: [105][ 1050/ 1236] Overall Loss 0.257960 Objective Loss 0.257960 LR 0.000500 Time 0.021506 +2023-10-05 21:31:48,957 - Epoch: [105][ 1060/ 1236] Overall Loss 0.258044 Objective Loss 0.258044 LR 0.000500 Time 0.021493 +2023-10-05 21:31:49,162 - Epoch: [105][ 1070/ 1236] Overall Loss 0.258140 Objective Loss 0.258140 LR 0.000500 Time 0.021483 +2023-10-05 21:31:49,363 - Epoch: [105][ 1080/ 1236] Overall Loss 0.258122 Objective Loss 0.258122 LR 0.000500 Time 0.021470 +2023-10-05 21:31:49,567 - Epoch: [105][ 1090/ 1236] Overall Loss 0.258172 Objective Loss 0.258172 LR 0.000500 Time 0.021460 +2023-10-05 21:31:49,769 - Epoch: [105][ 1100/ 1236] Overall Loss 0.258281 Objective Loss 0.258281 LR 0.000500 Time 0.021448 +2023-10-05 21:31:49,973 - Epoch: [105][ 1110/ 1236] Overall Loss 0.258343 Objective Loss 0.258343 LR 0.000500 Time 0.021438 +2023-10-05 21:31:50,175 - Epoch: [105][ 1120/ 1236] Overall Loss 0.258491 Objective Loss 0.258491 LR 0.000500 Time 0.021426 +2023-10-05 21:31:50,379 - Epoch: [105][ 1130/ 1236] Overall Loss 0.258707 Objective Loss 0.258707 LR 0.000500 Time 0.021417 +2023-10-05 21:31:50,581 - Epoch: [105][ 1140/ 1236] Overall Loss 0.258756 Objective Loss 0.258756 LR 0.000500 Time 0.021406 +2023-10-05 21:31:50,785 - Epoch: [105][ 1150/ 1236] Overall Loss 0.258791 Objective Loss 0.258791 LR 0.000500 Time 0.021397 +2023-10-05 21:31:50,987 - Epoch: [105][ 1160/ 1236] Overall Loss 0.258776 Objective Loss 0.258776 LR 0.000500 Time 0.021386 +2023-10-05 21:31:51,192 - Epoch: [105][ 1170/ 1236] Overall Loss 0.258997 Objective Loss 0.258997 LR 0.000500 Time 0.021379 +2023-10-05 21:31:51,394 - Epoch: [105][ 1180/ 1236] Overall Loss 0.259224 Objective Loss 0.259224 LR 0.000500 Time 0.021368 +2023-10-05 21:31:51,598 - Epoch: [105][ 1190/ 1236] Overall Loss 0.259378 Objective Loss 0.259378 LR 0.000500 Time 0.021360 +2023-10-05 21:31:51,800 - Epoch: [105][ 1200/ 1236] Overall Loss 0.259582 Objective Loss 0.259582 LR 0.000500 Time 0.021350 +2023-10-05 21:31:52,005 - Epoch: [105][ 1210/ 1236] Overall Loss 0.259708 Objective Loss 0.259708 LR 0.000500 Time 0.021343 +2023-10-05 21:31:52,207 - Epoch: [105][ 1220/ 1236] Overall Loss 0.259803 Objective Loss 0.259803 LR 0.000500 Time 0.021333 +2023-10-05 21:31:52,462 - Epoch: [105][ 1230/ 1236] Overall Loss 0.259933 Objective Loss 0.259933 LR 0.000500 Time 0.021366 +2023-10-05 21:31:52,581 - Epoch: [105][ 1236/ 1236] Overall Loss 0.259996 Objective Loss 0.259996 Top1 84.114053 Top5 98.370672 LR 0.000500 Time 0.021359 +2023-10-05 21:31:52,725 - --- validate (epoch=105)----------- +2023-10-05 21:31:52,725 - 29943 samples (256 per mini-batch) +2023-10-05 21:31:53,200 - Epoch: [105][ 10/ 117] Loss 0.356182 Top1 83.164062 Top5 97.695312 +2023-10-05 21:31:53,348 - Epoch: [105][ 20/ 117] Loss 0.333650 Top1 83.320312 Top5 97.714844 +2023-10-05 21:31:53,494 - Epoch: [105][ 30/ 117] Loss 0.318494 Top1 83.776042 Top5 97.968750 +2023-10-05 21:31:53,642 - Epoch: [105][ 40/ 117] Loss 0.326194 Top1 83.818359 Top5 97.880859 +2023-10-05 21:31:53,789 - Epoch: [105][ 50/ 117] Loss 0.321877 Top1 83.750000 Top5 97.796875 +2023-10-05 21:31:53,934 - Epoch: [105][ 60/ 117] Loss 0.329920 Top1 83.502604 Top5 97.805990 +2023-10-05 21:31:54,078 - Epoch: [105][ 70/ 117] Loss 0.329890 Top1 83.498884 Top5 97.728795 +2023-10-05 21:31:54,222 - Epoch: [105][ 80/ 117] Loss 0.329355 Top1 83.422852 Top5 97.724609 +2023-10-05 21:31:54,366 - Epoch: [105][ 90/ 117] Loss 0.328645 Top1 83.454861 Top5 97.760417 +2023-10-05 21:31:54,511 - Epoch: [105][ 100/ 117] Loss 0.336724 Top1 83.261719 Top5 97.757812 +2023-10-05 21:31:54,663 - Epoch: [105][ 110/ 117] Loss 0.335922 Top1 83.270597 Top5 97.773438 +2023-10-05 21:31:54,750 - Epoch: [105][ 117/ 117] Loss 0.334248 Top1 83.231473 Top5 97.759076 +2023-10-05 21:31:54,879 - ==> Top1: 83.231 Top5: 97.759 Loss: 0.334 + +2023-10-05 21:31:54,880 - ==> Confusion: +[[ 925 1 8 1 5 1 0 2 2 73 1 0 1 4 4 5 4 2 0 0 11] + [ 1 1038 1 0 6 29 0 24 2 0 2 4 0 0 2 3 3 0 8 1 7] + [ 3 1 960 12 1 0 24 9 0 3 6 1 8 0 0 6 2 1 0 7 12] + [ 4 1 20 973 0 4 1 0 2 0 9 1 5 5 24 3 2 5 15 3 12] + [ 29 4 1 2 962 4 0 1 0 10 2 1 1 2 8 7 6 3 1 1 5] + [ 3 39 1 3 3 975 0 23 3 2 4 13 2 14 6 2 2 0 3 6 12] + [ 0 3 29 0 1 1 1107 11 0 0 3 1 1 0 1 12 0 3 0 8 10] + [ 3 14 21 0 1 30 6 1069 0 5 4 10 3 2 0 2 0 0 31 11 6] + [ 23 0 1 0 1 6 1 0 962 48 10 1 3 9 9 4 1 1 6 1 2] + [ 99 0 1 0 4 4 1 0 26 934 0 2 0 21 4 10 1 0 0 0 12] + [ 2 5 12 4 2 1 5 5 8 1 955 3 1 19 6 1 3 1 8 1 10] + [ 2 0 1 0 1 6 0 1 0 1 0 974 17 5 0 3 1 13 0 6 4] + [ 2 0 2 6 1 2 0 4 2 0 2 40 969 1 4 5 3 12 1 1 11] + [ 2 0 1 2 5 11 0 0 13 11 4 5 1 1050 3 1 0 0 0 3 7] + [ 17 2 3 13 12 0 0 0 32 6 2 0 2 1 984 0 0 1 13 1 12] + [ 2 3 3 1 4 0 1 0 0 0 0 8 8 2 1 1070 11 9 0 8 3] + [ 0 11 3 1 4 9 0 0 2 0 0 3 2 2 2 14 1089 0 1 4 14] + [ 1 1 1 2 0 0 1 0 0 0 0 4 23 1 1 6 1 990 1 3 2] + [ 2 5 12 20 1 0 0 26 2 0 1 0 3 1 8 0 2 0 976 1 8] + [ 0 3 5 1 2 4 10 13 0 0 0 12 6 3 0 6 8 1 2 1064 12] + [ 159 201 153 89 98 141 51 114 109 104 211 136 317 338 130 82 174 64 150 188 4896]] + +2023-10-05 21:31:54,881 - ==> Best [Top1: 83.278 Top5: 97.809 Sparsity:0.00 Params: 148928 on epoch: 103] +2023-10-05 21:31:54,881 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:31:54,887 - + +2023-10-05 21:31:54,887 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:31:56,023 - Epoch: [106][ 10/ 1236] Overall Loss 0.232911 Objective Loss 0.232911 LR 0.000500 Time 0.113523 +2023-10-05 21:31:56,227 - Epoch: [106][ 20/ 1236] Overall Loss 0.242251 Objective Loss 0.242251 LR 0.000500 Time 0.066954 +2023-10-05 21:31:56,428 - Epoch: [106][ 30/ 1236] Overall Loss 0.246990 Objective Loss 0.246990 LR 0.000500 Time 0.051311 +2023-10-05 21:31:56,630 - Epoch: [106][ 40/ 1236] Overall Loss 0.255846 Objective Loss 0.255846 LR 0.000500 Time 0.043523 +2023-10-05 21:31:56,831 - Epoch: [106][ 50/ 1236] Overall Loss 0.255610 Objective Loss 0.255610 LR 0.000500 Time 0.038834 +2023-10-05 21:31:57,034 - Epoch: [106][ 60/ 1236] Overall Loss 0.255463 Objective Loss 0.255463 LR 0.000500 Time 0.035741 +2023-10-05 21:31:57,235 - Epoch: [106][ 70/ 1236] Overall Loss 0.254547 Objective Loss 0.254547 LR 0.000500 Time 0.033501 +2023-10-05 21:31:57,437 - Epoch: [106][ 80/ 1236] Overall Loss 0.252716 Objective Loss 0.252716 LR 0.000500 Time 0.031839 +2023-10-05 21:31:57,638 - Epoch: [106][ 90/ 1236] Overall Loss 0.254955 Objective Loss 0.254955 LR 0.000500 Time 0.030528 +2023-10-05 21:31:57,841 - Epoch: [106][ 100/ 1236] Overall Loss 0.254889 Objective Loss 0.254889 LR 0.000500 Time 0.029498 +2023-10-05 21:31:58,041 - Epoch: [106][ 110/ 1236] Overall Loss 0.253814 Objective Loss 0.253814 LR 0.000500 Time 0.028635 +2023-10-05 21:31:58,244 - Epoch: [106][ 120/ 1236] Overall Loss 0.255521 Objective Loss 0.255521 LR 0.000500 Time 0.027934 +2023-10-05 21:31:58,444 - Epoch: [106][ 130/ 1236] Overall Loss 0.255436 Objective Loss 0.255436 LR 0.000500 Time 0.027327 +2023-10-05 21:31:58,647 - Epoch: [106][ 140/ 1236] Overall Loss 0.255718 Objective Loss 0.255718 LR 0.000500 Time 0.026820 +2023-10-05 21:31:58,848 - Epoch: [106][ 150/ 1236] Overall Loss 0.257715 Objective Loss 0.257715 LR 0.000500 Time 0.026369 +2023-10-05 21:31:59,052 - Epoch: [106][ 160/ 1236] Overall Loss 0.257683 Objective Loss 0.257683 LR 0.000500 Time 0.025993 +2023-10-05 21:31:59,253 - Epoch: [106][ 170/ 1236] Overall Loss 0.258590 Objective Loss 0.258590 LR 0.000500 Time 0.025646 +2023-10-05 21:31:59,461 - Epoch: [106][ 180/ 1236] Overall Loss 0.258711 Objective Loss 0.258711 LR 0.000500 Time 0.025374 +2023-10-05 21:31:59,674 - Epoch: [106][ 190/ 1236] Overall Loss 0.257994 Objective Loss 0.257994 LR 0.000500 Time 0.025155 +2023-10-05 21:31:59,875 - Epoch: [106][ 200/ 1236] Overall Loss 0.258019 Objective Loss 0.258019 LR 0.000500 Time 0.024902 +2023-10-05 21:32:00,074 - Epoch: [106][ 210/ 1236] Overall Loss 0.257847 Objective Loss 0.257847 LR 0.000500 Time 0.024663 +2023-10-05 21:32:00,275 - Epoch: [106][ 220/ 1236] Overall Loss 0.256886 Objective Loss 0.256886 LR 0.000500 Time 0.024454 +2023-10-05 21:32:00,474 - Epoch: [106][ 230/ 1236] Overall Loss 0.256485 Objective Loss 0.256485 LR 0.000500 Time 0.024253 +2023-10-05 21:32:00,675 - Epoch: [106][ 240/ 1236] Overall Loss 0.258059 Objective Loss 0.258059 LR 0.000500 Time 0.024079 +2023-10-05 21:32:00,874 - Epoch: [106][ 250/ 1236] Overall Loss 0.257458 Objective Loss 0.257458 LR 0.000500 Time 0.023910 +2023-10-05 21:32:01,075 - Epoch: [106][ 260/ 1236] Overall Loss 0.256677 Objective Loss 0.256677 LR 0.000500 Time 0.023761 +2023-10-05 21:32:01,273 - Epoch: [106][ 270/ 1236] Overall Loss 0.256554 Objective Loss 0.256554 LR 0.000500 Time 0.023616 +2023-10-05 21:32:01,474 - Epoch: [106][ 280/ 1236] Overall Loss 0.256215 Objective Loss 0.256215 LR 0.000500 Time 0.023487 +2023-10-05 21:32:01,673 - Epoch: [106][ 290/ 1236] Overall Loss 0.256155 Objective Loss 0.256155 LR 0.000500 Time 0.023362 +2023-10-05 21:32:01,874 - Epoch: [106][ 300/ 1236] Overall Loss 0.256308 Objective Loss 0.256308 LR 0.000500 Time 0.023251 +2023-10-05 21:32:02,072 - Epoch: [106][ 310/ 1236] Overall Loss 0.255974 Objective Loss 0.255974 LR 0.000500 Time 0.023141 +2023-10-05 21:32:02,273 - Epoch: [106][ 320/ 1236] Overall Loss 0.255902 Objective Loss 0.255902 LR 0.000500 Time 0.023044 +2023-10-05 21:32:02,472 - Epoch: [106][ 330/ 1236] Overall Loss 0.255427 Objective Loss 0.255427 LR 0.000500 Time 0.022948 +2023-10-05 21:32:02,674 - Epoch: [106][ 340/ 1236] Overall Loss 0.255091 Objective Loss 0.255091 LR 0.000500 Time 0.022866 +2023-10-05 21:32:02,873 - Epoch: [106][ 350/ 1236] Overall Loss 0.256096 Objective Loss 0.256096 LR 0.000500 Time 0.022781 +2023-10-05 21:32:03,074 - Epoch: [106][ 360/ 1236] Overall Loss 0.255414 Objective Loss 0.255414 LR 0.000500 Time 0.022705 +2023-10-05 21:32:03,273 - Epoch: [106][ 370/ 1236] Overall Loss 0.254908 Objective Loss 0.254908 LR 0.000500 Time 0.022628 +2023-10-05 21:32:03,474 - Epoch: [106][ 380/ 1236] Overall Loss 0.254415 Objective Loss 0.254415 LR 0.000500 Time 0.022560 +2023-10-05 21:32:03,673 - Epoch: [106][ 390/ 1236] Overall Loss 0.253976 Objective Loss 0.253976 LR 0.000500 Time 0.022491 +2023-10-05 21:32:03,874 - Epoch: [106][ 400/ 1236] Overall Loss 0.253703 Objective Loss 0.253703 LR 0.000500 Time 0.022431 +2023-10-05 21:32:04,073 - Epoch: [106][ 410/ 1236] Overall Loss 0.253593 Objective Loss 0.253593 LR 0.000500 Time 0.022368 +2023-10-05 21:32:04,274 - Epoch: [106][ 420/ 1236] Overall Loss 0.253468 Objective Loss 0.253468 LR 0.000500 Time 0.022312 +2023-10-05 21:32:04,473 - Epoch: [106][ 430/ 1236] Overall Loss 0.253382 Objective Loss 0.253382 LR 0.000500 Time 0.022255 +2023-10-05 21:32:04,673 - Epoch: [106][ 440/ 1236] Overall Loss 0.253762 Objective Loss 0.253762 LR 0.000500 Time 0.022205 +2023-10-05 21:32:04,872 - Epoch: [106][ 450/ 1236] Overall Loss 0.253210 Objective Loss 0.253210 LR 0.000500 Time 0.022152 +2023-10-05 21:32:05,073 - Epoch: [106][ 460/ 1236] Overall Loss 0.253018 Objective Loss 0.253018 LR 0.000500 Time 0.022107 +2023-10-05 21:32:05,272 - Epoch: [106][ 470/ 1236] Overall Loss 0.253100 Objective Loss 0.253100 LR 0.000500 Time 0.022059 +2023-10-05 21:32:05,473 - Epoch: [106][ 480/ 1236] Overall Loss 0.253347 Objective Loss 0.253347 LR 0.000500 Time 0.022017 +2023-10-05 21:32:05,672 - Epoch: [106][ 490/ 1236] Overall Loss 0.253549 Objective Loss 0.253549 LR 0.000500 Time 0.021973 +2023-10-05 21:32:05,873 - Epoch: [106][ 500/ 1236] Overall Loss 0.253814 Objective Loss 0.253814 LR 0.000500 Time 0.021935 +2023-10-05 21:32:06,072 - Epoch: [106][ 510/ 1236] Overall Loss 0.253793 Objective Loss 0.253793 LR 0.000500 Time 0.021894 +2023-10-05 21:32:06,273 - Epoch: [106][ 520/ 1236] Overall Loss 0.253960 Objective Loss 0.253960 LR 0.000500 Time 0.021858 +2023-10-05 21:32:06,472 - Epoch: [106][ 530/ 1236] Overall Loss 0.253820 Objective Loss 0.253820 LR 0.000500 Time 0.021822 +2023-10-05 21:32:06,673 - Epoch: [106][ 540/ 1236] Overall Loss 0.253901 Objective Loss 0.253901 LR 0.000500 Time 0.021790 +2023-10-05 21:32:06,872 - Epoch: [106][ 550/ 1236] Overall Loss 0.253994 Objective Loss 0.253994 LR 0.000500 Time 0.021755 +2023-10-05 21:32:07,072 - Epoch: [106][ 560/ 1236] Overall Loss 0.253900 Objective Loss 0.253900 LR 0.000500 Time 0.021723 +2023-10-05 21:32:07,271 - Epoch: [106][ 570/ 1236] Overall Loss 0.253935 Objective Loss 0.253935 LR 0.000500 Time 0.021690 +2023-10-05 21:32:07,472 - Epoch: [106][ 580/ 1236] Overall Loss 0.254034 Objective Loss 0.254034 LR 0.000500 Time 0.021662 +2023-10-05 21:32:07,672 - Epoch: [106][ 590/ 1236] Overall Loss 0.254298 Objective Loss 0.254298 LR 0.000500 Time 0.021632 +2023-10-05 21:32:07,873 - Epoch: [106][ 600/ 1236] Overall Loss 0.254587 Objective Loss 0.254587 LR 0.000500 Time 0.021606 +2023-10-05 21:32:08,072 - Epoch: [106][ 610/ 1236] Overall Loss 0.254346 Objective Loss 0.254346 LR 0.000500 Time 0.021577 +2023-10-05 21:32:08,273 - Epoch: [106][ 620/ 1236] Overall Loss 0.254008 Objective Loss 0.254008 LR 0.000500 Time 0.021553 +2023-10-05 21:32:08,472 - Epoch: [106][ 630/ 1236] Overall Loss 0.254259 Objective Loss 0.254259 LR 0.000500 Time 0.021526 +2023-10-05 21:32:08,673 - Epoch: [106][ 640/ 1236] Overall Loss 0.254564 Objective Loss 0.254564 LR 0.000500 Time 0.021504 +2023-10-05 21:32:08,872 - Epoch: [106][ 650/ 1236] Overall Loss 0.254472 Objective Loss 0.254472 LR 0.000500 Time 0.021478 +2023-10-05 21:32:09,073 - Epoch: [106][ 660/ 1236] Overall Loss 0.254188 Objective Loss 0.254188 LR 0.000500 Time 0.021457 +2023-10-05 21:32:09,272 - Epoch: [106][ 670/ 1236] Overall Loss 0.254323 Objective Loss 0.254323 LR 0.000500 Time 0.021434 +2023-10-05 21:32:09,473 - Epoch: [106][ 680/ 1236] Overall Loss 0.254324 Objective Loss 0.254324 LR 0.000500 Time 0.021414 +2023-10-05 21:32:09,672 - Epoch: [106][ 690/ 1236] Overall Loss 0.254497 Objective Loss 0.254497 LR 0.000500 Time 0.021391 +2023-10-05 21:32:09,874 - Epoch: [106][ 700/ 1236] Overall Loss 0.254814 Objective Loss 0.254814 LR 0.000500 Time 0.021373 +2023-10-05 21:32:10,073 - Epoch: [106][ 710/ 1236] Overall Loss 0.254519 Objective Loss 0.254519 LR 0.000500 Time 0.021352 +2023-10-05 21:32:10,274 - Epoch: [106][ 720/ 1236] Overall Loss 0.254587 Objective Loss 0.254587 LR 0.000500 Time 0.021335 +2023-10-05 21:32:10,474 - Epoch: [106][ 730/ 1236] Overall Loss 0.254315 Objective Loss 0.254315 LR 0.000500 Time 0.021316 +2023-10-05 21:32:10,676 - Epoch: [106][ 740/ 1236] Overall Loss 0.254123 Objective Loss 0.254123 LR 0.000500 Time 0.021299 +2023-10-05 21:32:10,874 - Epoch: [106][ 750/ 1236] Overall Loss 0.254176 Objective Loss 0.254176 LR 0.000500 Time 0.021280 +2023-10-05 21:32:11,076 - Epoch: [106][ 760/ 1236] Overall Loss 0.254268 Objective Loss 0.254268 LR 0.000500 Time 0.021264 +2023-10-05 21:32:11,273 - Epoch: [106][ 770/ 1236] Overall Loss 0.254241 Objective Loss 0.254241 LR 0.000500 Time 0.021244 +2023-10-05 21:32:11,474 - Epoch: [106][ 780/ 1236] Overall Loss 0.254299 Objective Loss 0.254299 LR 0.000500 Time 0.021229 +2023-10-05 21:32:11,678 - Epoch: [106][ 790/ 1236] Overall Loss 0.254540 Objective Loss 0.254540 LR 0.000500 Time 0.021218 +2023-10-05 21:32:11,879 - Epoch: [106][ 800/ 1236] Overall Loss 0.254388 Objective Loss 0.254388 LR 0.000500 Time 0.021204 +2023-10-05 21:32:12,078 - Epoch: [106][ 810/ 1236] Overall Loss 0.254553 Objective Loss 0.254553 LR 0.000500 Time 0.021186 +2023-10-05 21:32:12,279 - Epoch: [106][ 820/ 1236] Overall Loss 0.254555 Objective Loss 0.254555 LR 0.000500 Time 0.021173 +2023-10-05 21:32:12,477 - Epoch: [106][ 830/ 1236] Overall Loss 0.255004 Objective Loss 0.255004 LR 0.000500 Time 0.021156 +2023-10-05 21:32:12,679 - Epoch: [106][ 840/ 1236] Overall Loss 0.254809 Objective Loss 0.254809 LR 0.000500 Time 0.021144 +2023-10-05 21:32:12,882 - Epoch: [106][ 850/ 1236] Overall Loss 0.254968 Objective Loss 0.254968 LR 0.000500 Time 0.021134 +2023-10-05 21:32:13,083 - Epoch: [106][ 860/ 1236] Overall Loss 0.255119 Objective Loss 0.255119 LR 0.000500 Time 0.021122 +2023-10-05 21:32:13,283 - Epoch: [106][ 870/ 1236] Overall Loss 0.255290 Objective Loss 0.255290 LR 0.000500 Time 0.021108 +2023-10-05 21:32:13,483 - Epoch: [106][ 880/ 1236] Overall Loss 0.255406 Objective Loss 0.255406 LR 0.000500 Time 0.021096 +2023-10-05 21:32:13,685 - Epoch: [106][ 890/ 1236] Overall Loss 0.255372 Objective Loss 0.255372 LR 0.000500 Time 0.021085 +2023-10-05 21:32:13,890 - Epoch: [106][ 900/ 1236] Overall Loss 0.255560 Objective Loss 0.255560 LR 0.000500 Time 0.021078 +2023-10-05 21:32:14,098 - Epoch: [106][ 910/ 1236] Overall Loss 0.255444 Objective Loss 0.255444 LR 0.000500 Time 0.021074 +2023-10-05 21:32:14,301 - Epoch: [106][ 920/ 1236] Overall Loss 0.255519 Objective Loss 0.255519 LR 0.000500 Time 0.021066 +2023-10-05 21:32:14,504 - Epoch: [106][ 930/ 1236] Overall Loss 0.255454 Objective Loss 0.255454 LR 0.000500 Time 0.021057 +2023-10-05 21:32:14,707 - Epoch: [106][ 940/ 1236] Overall Loss 0.255408 Objective Loss 0.255408 LR 0.000500 Time 0.021048 +2023-10-05 21:32:14,908 - Epoch: [106][ 950/ 1236] Overall Loss 0.255251 Objective Loss 0.255251 LR 0.000500 Time 0.021039 +2023-10-05 21:32:15,112 - Epoch: [106][ 960/ 1236] Overall Loss 0.254983 Objective Loss 0.254983 LR 0.000500 Time 0.021031 +2023-10-05 21:32:15,315 - Epoch: [106][ 970/ 1236] Overall Loss 0.254949 Objective Loss 0.254949 LR 0.000500 Time 0.021023 +2023-10-05 21:32:15,518 - Epoch: [106][ 980/ 1236] Overall Loss 0.254905 Objective Loss 0.254905 LR 0.000500 Time 0.021015 +2023-10-05 21:32:15,719 - Epoch: [106][ 990/ 1236] Overall Loss 0.255199 Objective Loss 0.255199 LR 0.000500 Time 0.021006 +2023-10-05 21:32:15,922 - Epoch: [106][ 1000/ 1236] Overall Loss 0.255040 Objective Loss 0.255040 LR 0.000500 Time 0.020998 +2023-10-05 21:32:16,124 - Epoch: [106][ 1010/ 1236] Overall Loss 0.254954 Objective Loss 0.254954 LR 0.000500 Time 0.020989 +2023-10-05 21:32:16,327 - Epoch: [106][ 1020/ 1236] Overall Loss 0.255063 Objective Loss 0.255063 LR 0.000500 Time 0.020983 +2023-10-05 21:32:16,528 - Epoch: [106][ 1030/ 1236] Overall Loss 0.255346 Objective Loss 0.255346 LR 0.000500 Time 0.020974 +2023-10-05 21:32:16,731 - Epoch: [106][ 1040/ 1236] Overall Loss 0.255329 Objective Loss 0.255329 LR 0.000500 Time 0.020967 +2023-10-05 21:32:16,934 - Epoch: [106][ 1050/ 1236] Overall Loss 0.255245 Objective Loss 0.255245 LR 0.000500 Time 0.020960 +2023-10-05 21:32:17,137 - Epoch: [106][ 1060/ 1236] Overall Loss 0.255224 Objective Loss 0.255224 LR 0.000500 Time 0.020953 +2023-10-05 21:32:17,337 - Epoch: [106][ 1070/ 1236] Overall Loss 0.255215 Objective Loss 0.255215 LR 0.000500 Time 0.020944 +2023-10-05 21:32:17,541 - Epoch: [106][ 1080/ 1236] Overall Loss 0.255920 Objective Loss 0.255920 LR 0.000500 Time 0.020939 +2023-10-05 21:32:17,743 - Epoch: [106][ 1090/ 1236] Overall Loss 0.255656 Objective Loss 0.255656 LR 0.000500 Time 0.020932 +2023-10-05 21:32:17,946 - Epoch: [106][ 1100/ 1236] Overall Loss 0.255533 Objective Loss 0.255533 LR 0.000500 Time 0.020926 +2023-10-05 21:32:18,148 - Epoch: [106][ 1110/ 1236] Overall Loss 0.255658 Objective Loss 0.255658 LR 0.000500 Time 0.020919 +2023-10-05 21:32:18,352 - Epoch: [106][ 1120/ 1236] Overall Loss 0.255627 Objective Loss 0.255627 LR 0.000500 Time 0.020914 +2023-10-05 21:32:18,554 - Epoch: [106][ 1130/ 1236] Overall Loss 0.255507 Objective Loss 0.255507 LR 0.000500 Time 0.020907 +2023-10-05 21:32:18,758 - Epoch: [106][ 1140/ 1236] Overall Loss 0.255547 Objective Loss 0.255547 LR 0.000500 Time 0.020902 +2023-10-05 21:32:18,960 - Epoch: [106][ 1150/ 1236] Overall Loss 0.255610 Objective Loss 0.255610 LR 0.000500 Time 0.020896 +2023-10-05 21:32:19,164 - Epoch: [106][ 1160/ 1236] Overall Loss 0.255423 Objective Loss 0.255423 LR 0.000500 Time 0.020891 +2023-10-05 21:32:19,366 - Epoch: [106][ 1170/ 1236] Overall Loss 0.255238 Objective Loss 0.255238 LR 0.000500 Time 0.020885 +2023-10-05 21:32:19,570 - Epoch: [106][ 1180/ 1236] Overall Loss 0.255192 Objective Loss 0.255192 LR 0.000500 Time 0.020881 +2023-10-05 21:32:19,772 - Epoch: [106][ 1190/ 1236] Overall Loss 0.255124 Objective Loss 0.255124 LR 0.000500 Time 0.020875 +2023-10-05 21:32:19,976 - Epoch: [106][ 1200/ 1236] Overall Loss 0.254996 Objective Loss 0.254996 LR 0.000500 Time 0.020870 +2023-10-05 21:32:20,177 - Epoch: [106][ 1210/ 1236] Overall Loss 0.255011 Objective Loss 0.255011 LR 0.000500 Time 0.020863 +2023-10-05 21:32:20,380 - Epoch: [106][ 1220/ 1236] Overall Loss 0.255157 Objective Loss 0.255157 LR 0.000500 Time 0.020859 +2023-10-05 21:32:20,634 - Epoch: [106][ 1230/ 1236] Overall Loss 0.255116 Objective Loss 0.255116 LR 0.000500 Time 0.020895 +2023-10-05 21:32:20,752 - Epoch: [106][ 1236/ 1236] Overall Loss 0.255083 Objective Loss 0.255083 Top1 83.095723 Top5 98.167006 LR 0.000500 Time 0.020889 +2023-10-05 21:32:20,866 - --- validate (epoch=106)----------- +2023-10-05 21:32:20,867 - 29943 samples (256 per mini-batch) +2023-10-05 21:32:21,323 - Epoch: [106][ 10/ 117] Loss 0.320827 Top1 83.945312 Top5 98.554688 +2023-10-05 21:32:21,474 - Epoch: [106][ 20/ 117] Loss 0.351144 Top1 82.519531 Top5 98.066406 +2023-10-05 21:32:21,623 - Epoch: [106][ 30/ 117] Loss 0.344327 Top1 82.981771 Top5 97.916667 +2023-10-05 21:32:21,775 - Epoch: [106][ 40/ 117] Loss 0.340166 Top1 83.115234 Top5 97.861328 +2023-10-05 21:32:21,924 - Epoch: [106][ 50/ 117] Loss 0.338154 Top1 83.343750 Top5 97.906250 +2023-10-05 21:32:22,075 - Epoch: [106][ 60/ 117] Loss 0.342939 Top1 83.268229 Top5 97.903646 +2023-10-05 21:32:22,230 - Epoch: [106][ 70/ 117] Loss 0.342232 Top1 83.320312 Top5 97.829241 +2023-10-05 21:32:22,387 - Epoch: [106][ 80/ 117] Loss 0.338803 Top1 83.374023 Top5 97.851562 +2023-10-05 21:32:22,543 - Epoch: [106][ 90/ 117] Loss 0.339306 Top1 83.381076 Top5 97.834201 +2023-10-05 21:32:22,699 - Epoch: [106][ 100/ 117] Loss 0.335669 Top1 83.496094 Top5 97.847656 +2023-10-05 21:32:22,853 - Epoch: [106][ 110/ 117] Loss 0.337381 Top1 83.458807 Top5 97.848011 +2023-10-05 21:32:22,939 - Epoch: [106][ 117/ 117] Loss 0.333234 Top1 83.542063 Top5 97.872625 +2023-10-05 21:32:23,085 - ==> Top1: 83.542 Top5: 97.873 Loss: 0.333 + +2023-10-05 21:32:23,086 - ==> Confusion: +[[ 940 4 6 0 6 2 0 0 7 53 1 1 0 3 4 3 6 1 0 0 13] + [ 3 1031 3 0 12 30 1 17 1 0 1 3 0 0 3 3 6 0 12 0 5] + [ 7 0 963 11 2 1 23 8 0 1 5 3 6 3 0 2 1 1 4 6 9] + [ 4 1 16 968 0 4 1 0 6 1 8 0 4 5 22 4 0 5 27 0 13] + [ 32 6 0 0 950 6 0 2 0 10 2 1 0 3 14 4 10 2 0 1 7] + [ 6 29 0 0 5 984 1 24 3 1 8 6 1 12 8 1 3 0 5 5 14] + [ 0 3 27 0 2 2 1114 12 0 0 3 5 0 0 1 8 0 1 1 5 7] + [ 2 16 21 0 4 37 4 1060 2 6 5 11 1 1 0 2 0 0 28 7 11] + [ 17 3 0 0 1 5 1 0 975 42 8 1 2 8 14 5 0 0 4 2 1] + [ 115 0 0 0 5 2 0 0 35 918 0 4 0 18 3 9 0 1 0 0 9] + [ 5 3 12 3 1 2 3 4 11 1 969 2 0 18 4 2 1 0 2 2 8] + [ 1 0 3 0 0 14 0 2 0 1 0 962 11 8 0 2 1 15 0 11 4] + [ 1 1 3 5 1 5 0 0 2 0 3 43 954 5 1 10 4 13 2 1 14] + [ 1 0 1 1 5 16 0 0 12 13 7 3 1 1045 2 1 0 0 0 2 9] + [ 16 1 3 12 6 0 0 0 31 2 3 0 1 1 997 0 2 2 12 1 11] + [ 1 1 3 0 3 1 0 0 0 0 0 8 3 4 1 1064 20 9 2 8 6] + [ 1 12 1 0 5 7 1 1 1 0 0 3 0 0 3 13 1101 0 1 3 8] + [ 0 1 1 1 0 0 2 0 1 1 0 2 20 2 1 8 0 991 1 2 4] + [ 2 8 7 16 1 2 0 28 2 0 5 1 7 0 12 0 1 0 965 0 11] + [ 0 4 4 1 1 6 9 11 1 0 0 21 2 2 0 7 11 1 1 1058 12] + [ 151 182 141 74 99 169 45 98 110 73 186 155 289 291 163 69 238 65 137 164 5006]] + +2023-10-05 21:32:23,087 - ==> Best [Top1: 83.542 Top5: 97.873 Sparsity:0.00 Params: 148928 on epoch: 106] +2023-10-05 21:32:23,087 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:32:23,101 - + +2023-10-05 21:32:23,101 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:32:24,117 - Epoch: [107][ 10/ 1236] Overall Loss 0.234973 Objective Loss 0.234973 LR 0.000500 Time 0.101569 +2023-10-05 21:32:24,319 - Epoch: [107][ 20/ 1236] Overall Loss 0.243941 Objective Loss 0.243941 LR 0.000500 Time 0.060846 +2023-10-05 21:32:24,519 - Epoch: [107][ 30/ 1236] Overall Loss 0.245493 Objective Loss 0.245493 LR 0.000500 Time 0.047230 +2023-10-05 21:32:24,721 - Epoch: [107][ 40/ 1236] Overall Loss 0.243814 Objective Loss 0.243814 LR 0.000500 Time 0.040464 +2023-10-05 21:32:24,921 - Epoch: [107][ 50/ 1236] Overall Loss 0.241072 Objective Loss 0.241072 LR 0.000500 Time 0.036368 +2023-10-05 21:32:25,123 - Epoch: [107][ 60/ 1236] Overall Loss 0.246564 Objective Loss 0.246564 LR 0.000500 Time 0.033661 +2023-10-05 21:32:25,323 - Epoch: [107][ 70/ 1236] Overall Loss 0.249673 Objective Loss 0.249673 LR 0.000500 Time 0.031702 +2023-10-05 21:32:25,524 - Epoch: [107][ 80/ 1236] Overall Loss 0.251731 Objective Loss 0.251731 LR 0.000500 Time 0.030255 +2023-10-05 21:32:25,724 - Epoch: [107][ 90/ 1236] Overall Loss 0.253599 Objective Loss 0.253599 LR 0.000500 Time 0.029111 +2023-10-05 21:32:25,926 - Epoch: [107][ 100/ 1236] Overall Loss 0.256310 Objective Loss 0.256310 LR 0.000500 Time 0.028213 +2023-10-05 21:32:26,133 - Epoch: [107][ 110/ 1236] Overall Loss 0.255209 Objective Loss 0.255209 LR 0.000500 Time 0.027528 +2023-10-05 21:32:26,351 - Epoch: [107][ 120/ 1236] Overall Loss 0.254584 Objective Loss 0.254584 LR 0.000500 Time 0.027045 +2023-10-05 21:32:26,564 - Epoch: [107][ 130/ 1236] Overall Loss 0.254290 Objective Loss 0.254290 LR 0.000500 Time 0.026604 +2023-10-05 21:32:26,782 - Epoch: [107][ 140/ 1236] Overall Loss 0.252146 Objective Loss 0.252146 LR 0.000500 Time 0.026256 +2023-10-05 21:32:26,996 - Epoch: [107][ 150/ 1236] Overall Loss 0.252003 Objective Loss 0.252003 LR 0.000500 Time 0.025926 +2023-10-05 21:32:27,214 - Epoch: [107][ 160/ 1236] Overall Loss 0.253477 Objective Loss 0.253477 LR 0.000500 Time 0.025663 +2023-10-05 21:32:27,428 - Epoch: [107][ 170/ 1236] Overall Loss 0.253225 Objective Loss 0.253225 LR 0.000500 Time 0.025409 +2023-10-05 21:32:27,646 - Epoch: [107][ 180/ 1236] Overall Loss 0.253058 Objective Loss 0.253058 LR 0.000500 Time 0.025206 +2023-10-05 21:32:27,860 - Epoch: [107][ 190/ 1236] Overall Loss 0.252868 Objective Loss 0.252868 LR 0.000500 Time 0.025003 +2023-10-05 21:32:28,077 - Epoch: [107][ 200/ 1236] Overall Loss 0.252404 Objective Loss 0.252404 LR 0.000500 Time 0.024839 +2023-10-05 21:32:28,288 - Epoch: [107][ 210/ 1236] Overall Loss 0.251660 Objective Loss 0.251660 LR 0.000500 Time 0.024659 +2023-10-05 21:32:28,496 - Epoch: [107][ 220/ 1236] Overall Loss 0.251411 Objective Loss 0.251411 LR 0.000500 Time 0.024481 +2023-10-05 21:32:28,704 - Epoch: [107][ 230/ 1236] Overall Loss 0.251399 Objective Loss 0.251399 LR 0.000500 Time 0.024315 +2023-10-05 21:32:28,911 - Epoch: [107][ 240/ 1236] Overall Loss 0.252100 Objective Loss 0.252100 LR 0.000500 Time 0.024166 +2023-10-05 21:32:29,118 - Epoch: [107][ 250/ 1236] Overall Loss 0.252208 Objective Loss 0.252208 LR 0.000500 Time 0.024025 +2023-10-05 21:32:29,326 - Epoch: [107][ 260/ 1236] Overall Loss 0.252372 Objective Loss 0.252372 LR 0.000500 Time 0.023899 +2023-10-05 21:32:29,533 - Epoch: [107][ 270/ 1236] Overall Loss 0.252886 Objective Loss 0.252886 LR 0.000500 Time 0.023778 +2023-10-05 21:32:29,742 - Epoch: [107][ 280/ 1236] Overall Loss 0.252507 Objective Loss 0.252507 LR 0.000500 Time 0.023670 +2023-10-05 21:32:29,948 - Epoch: [107][ 290/ 1236] Overall Loss 0.251905 Objective Loss 0.251905 LR 0.000500 Time 0.023565 +2023-10-05 21:32:30,156 - Epoch: [107][ 300/ 1236] Overall Loss 0.251621 Objective Loss 0.251621 LR 0.000500 Time 0.023471 +2023-10-05 21:32:30,363 - Epoch: [107][ 310/ 1236] Overall Loss 0.250532 Objective Loss 0.250532 LR 0.000500 Time 0.023380 +2023-10-05 21:32:30,572 - Epoch: [107][ 320/ 1236] Overall Loss 0.250512 Objective Loss 0.250512 LR 0.000500 Time 0.023299 +2023-10-05 21:32:30,779 - Epoch: [107][ 330/ 1236] Overall Loss 0.250823 Objective Loss 0.250823 LR 0.000500 Time 0.023220 +2023-10-05 21:32:30,988 - Epoch: [107][ 340/ 1236] Overall Loss 0.250459 Objective Loss 0.250459 LR 0.000500 Time 0.023148 +2023-10-05 21:32:31,194 - Epoch: [107][ 350/ 1236] Overall Loss 0.251449 Objective Loss 0.251449 LR 0.000500 Time 0.023075 +2023-10-05 21:32:31,408 - Epoch: [107][ 360/ 1236] Overall Loss 0.250921 Objective Loss 0.250921 LR 0.000500 Time 0.023027 +2023-10-05 21:32:31,617 - Epoch: [107][ 370/ 1236] Overall Loss 0.250442 Objective Loss 0.250442 LR 0.000500 Time 0.022968 +2023-10-05 21:32:31,830 - Epoch: [107][ 380/ 1236] Overall Loss 0.250063 Objective Loss 0.250063 LR 0.000500 Time 0.022924 +2023-10-05 21:32:32,039 - Epoch: [107][ 390/ 1236] Overall Loss 0.251130 Objective Loss 0.251130 LR 0.000500 Time 0.022872 +2023-10-05 21:32:32,252 - Epoch: [107][ 400/ 1236] Overall Loss 0.250843 Objective Loss 0.250843 LR 0.000500 Time 0.022832 +2023-10-05 21:32:32,460 - Epoch: [107][ 410/ 1236] Overall Loss 0.250552 Objective Loss 0.250552 LR 0.000500 Time 0.022782 +2023-10-05 21:32:32,674 - Epoch: [107][ 420/ 1236] Overall Loss 0.249525 Objective Loss 0.249525 LR 0.000500 Time 0.022746 +2023-10-05 21:32:32,882 - Epoch: [107][ 430/ 1236] Overall Loss 0.250322 Objective Loss 0.250322 LR 0.000500 Time 0.022702 +2023-10-05 21:32:33,096 - Epoch: [107][ 440/ 1236] Overall Loss 0.251183 Objective Loss 0.251183 LR 0.000500 Time 0.022670 +2023-10-05 21:32:33,304 - Epoch: [107][ 450/ 1236] Overall Loss 0.251032 Objective Loss 0.251032 LR 0.000500 Time 0.022629 +2023-10-05 21:32:33,518 - Epoch: [107][ 460/ 1236] Overall Loss 0.251386 Objective Loss 0.251386 LR 0.000500 Time 0.022600 +2023-10-05 21:32:33,726 - Epoch: [107][ 470/ 1236] Overall Loss 0.251490 Objective Loss 0.251490 LR 0.000500 Time 0.022563 +2023-10-05 21:32:33,933 - Epoch: [107][ 480/ 1236] Overall Loss 0.251405 Objective Loss 0.251405 LR 0.000500 Time 0.022522 +2023-10-05 21:32:34,134 - Epoch: [107][ 490/ 1236] Overall Loss 0.251517 Objective Loss 0.251517 LR 0.000500 Time 0.022473 +2023-10-05 21:32:34,337 - Epoch: [107][ 500/ 1236] Overall Loss 0.251881 Objective Loss 0.251881 LR 0.000500 Time 0.022428 +2023-10-05 21:32:34,538 - Epoch: [107][ 510/ 1236] Overall Loss 0.252326 Objective Loss 0.252326 LR 0.000500 Time 0.022381 +2023-10-05 21:32:34,741 - Epoch: [107][ 520/ 1236] Overall Loss 0.252645 Objective Loss 0.252645 LR 0.000500 Time 0.022341 +2023-10-05 21:32:34,942 - Epoch: [107][ 530/ 1236] Overall Loss 0.252910 Objective Loss 0.252910 LR 0.000500 Time 0.022298 +2023-10-05 21:32:35,145 - Epoch: [107][ 540/ 1236] Overall Loss 0.252523 Objective Loss 0.252523 LR 0.000500 Time 0.022259 +2023-10-05 21:32:35,345 - Epoch: [107][ 550/ 1236] Overall Loss 0.252696 Objective Loss 0.252696 LR 0.000500 Time 0.022219 +2023-10-05 21:32:35,548 - Epoch: [107][ 560/ 1236] Overall Loss 0.252783 Objective Loss 0.252783 LR 0.000500 Time 0.022184 +2023-10-05 21:32:35,749 - Epoch: [107][ 570/ 1236] Overall Loss 0.252828 Objective Loss 0.252828 LR 0.000500 Time 0.022147 +2023-10-05 21:32:35,952 - Epoch: [107][ 580/ 1236] Overall Loss 0.252748 Objective Loss 0.252748 LR 0.000500 Time 0.022114 +2023-10-05 21:32:36,153 - Epoch: [107][ 590/ 1236] Overall Loss 0.252552 Objective Loss 0.252552 LR 0.000500 Time 0.022080 +2023-10-05 21:32:36,356 - Epoch: [107][ 600/ 1236] Overall Loss 0.252828 Objective Loss 0.252828 LR 0.000500 Time 0.022048 +2023-10-05 21:32:36,557 - Epoch: [107][ 610/ 1236] Overall Loss 0.252870 Objective Loss 0.252870 LR 0.000500 Time 0.022016 +2023-10-05 21:32:36,760 - Epoch: [107][ 620/ 1236] Overall Loss 0.253152 Objective Loss 0.253152 LR 0.000500 Time 0.021987 +2023-10-05 21:32:36,961 - Epoch: [107][ 630/ 1236] Overall Loss 0.253193 Objective Loss 0.253193 LR 0.000500 Time 0.021958 +2023-10-05 21:32:37,164 - Epoch: [107][ 640/ 1236] Overall Loss 0.253042 Objective Loss 0.253042 LR 0.000500 Time 0.021930 +2023-10-05 21:32:37,365 - Epoch: [107][ 650/ 1236] Overall Loss 0.253023 Objective Loss 0.253023 LR 0.000500 Time 0.021902 +2023-10-05 21:32:37,568 - Epoch: [107][ 660/ 1236] Overall Loss 0.253033 Objective Loss 0.253033 LR 0.000500 Time 0.021877 +2023-10-05 21:32:37,769 - Epoch: [107][ 670/ 1236] Overall Loss 0.253152 Objective Loss 0.253152 LR 0.000500 Time 0.021850 +2023-10-05 21:32:37,972 - Epoch: [107][ 680/ 1236] Overall Loss 0.253655 Objective Loss 0.253655 LR 0.000500 Time 0.021826 +2023-10-05 21:32:38,173 - Epoch: [107][ 690/ 1236] Overall Loss 0.253539 Objective Loss 0.253539 LR 0.000500 Time 0.021801 +2023-10-05 21:32:38,376 - Epoch: [107][ 700/ 1236] Overall Loss 0.253211 Objective Loss 0.253211 LR 0.000500 Time 0.021779 +2023-10-05 21:32:38,577 - Epoch: [107][ 710/ 1236] Overall Loss 0.252980 Objective Loss 0.252980 LR 0.000500 Time 0.021755 +2023-10-05 21:32:38,780 - Epoch: [107][ 720/ 1236] Overall Loss 0.253221 Objective Loss 0.253221 LR 0.000500 Time 0.021734 +2023-10-05 21:32:38,981 - Epoch: [107][ 730/ 1236] Overall Loss 0.252725 Objective Loss 0.252725 LR 0.000500 Time 0.021711 +2023-10-05 21:32:39,184 - Epoch: [107][ 740/ 1236] Overall Loss 0.252691 Objective Loss 0.252691 LR 0.000500 Time 0.021692 +2023-10-05 21:32:39,386 - Epoch: [107][ 750/ 1236] Overall Loss 0.252655 Objective Loss 0.252655 LR 0.000500 Time 0.021671 +2023-10-05 21:32:39,590 - Epoch: [107][ 760/ 1236] Overall Loss 0.252616 Objective Loss 0.252616 LR 0.000500 Time 0.021654 +2023-10-05 21:32:39,792 - Epoch: [107][ 770/ 1236] Overall Loss 0.252449 Objective Loss 0.252449 LR 0.000500 Time 0.021634 +2023-10-05 21:32:40,005 - Epoch: [107][ 780/ 1236] Overall Loss 0.252521 Objective Loss 0.252521 LR 0.000500 Time 0.021629 +2023-10-05 21:32:40,214 - Epoch: [107][ 790/ 1236] Overall Loss 0.252211 Objective Loss 0.252211 LR 0.000500 Time 0.021620 +2023-10-05 21:32:40,421 - Epoch: [107][ 800/ 1236] Overall Loss 0.251951 Objective Loss 0.251951 LR 0.000500 Time 0.021608 +2023-10-05 21:32:40,622 - Epoch: [107][ 810/ 1236] Overall Loss 0.251525 Objective Loss 0.251525 LR 0.000500 Time 0.021590 +2023-10-05 21:32:40,826 - Epoch: [107][ 820/ 1236] Overall Loss 0.251265 Objective Loss 0.251265 LR 0.000500 Time 0.021574 +2023-10-05 21:32:41,027 - Epoch: [107][ 830/ 1236] Overall Loss 0.251272 Objective Loss 0.251272 LR 0.000500 Time 0.021556 +2023-10-05 21:32:41,231 - Epoch: [107][ 840/ 1236] Overall Loss 0.251195 Objective Loss 0.251195 LR 0.000500 Time 0.021542 +2023-10-05 21:32:41,433 - Epoch: [107][ 850/ 1236] Overall Loss 0.251150 Objective Loss 0.251150 LR 0.000500 Time 0.021525 +2023-10-05 21:32:41,636 - Epoch: [107][ 860/ 1236] Overall Loss 0.251232 Objective Loss 0.251232 LR 0.000500 Time 0.021511 +2023-10-05 21:32:41,838 - Epoch: [107][ 870/ 1236] Overall Loss 0.250822 Objective Loss 0.250822 LR 0.000500 Time 0.021495 +2023-10-05 21:32:42,041 - Epoch: [107][ 880/ 1236] Overall Loss 0.250946 Objective Loss 0.250946 LR 0.000500 Time 0.021482 +2023-10-05 21:32:42,243 - Epoch: [107][ 890/ 1236] Overall Loss 0.251004 Objective Loss 0.251004 LR 0.000500 Time 0.021466 +2023-10-05 21:32:42,446 - Epoch: [107][ 900/ 1236] Overall Loss 0.251341 Objective Loss 0.251341 LR 0.000500 Time 0.021454 +2023-10-05 21:32:42,648 - Epoch: [107][ 910/ 1236] Overall Loss 0.251418 Objective Loss 0.251418 LR 0.000500 Time 0.021439 +2023-10-05 21:32:42,851 - Epoch: [107][ 920/ 1236] Overall Loss 0.251240 Objective Loss 0.251240 LR 0.000500 Time 0.021426 +2023-10-05 21:32:43,052 - Epoch: [107][ 930/ 1236] Overall Loss 0.251085 Objective Loss 0.251085 LR 0.000500 Time 0.021412 +2023-10-05 21:32:43,257 - Epoch: [107][ 940/ 1236] Overall Loss 0.250770 Objective Loss 0.250770 LR 0.000500 Time 0.021401 +2023-10-05 21:32:43,459 - Epoch: [107][ 950/ 1236] Overall Loss 0.250957 Objective Loss 0.250957 LR 0.000500 Time 0.021388 +2023-10-05 21:32:43,662 - Epoch: [107][ 960/ 1236] Overall Loss 0.251232 Objective Loss 0.251232 LR 0.000500 Time 0.021377 +2023-10-05 21:32:43,864 - Epoch: [107][ 970/ 1236] Overall Loss 0.251382 Objective Loss 0.251382 LR 0.000500 Time 0.021364 +2023-10-05 21:32:44,067 - Epoch: [107][ 980/ 1236] Overall Loss 0.251143 Objective Loss 0.251143 LR 0.000500 Time 0.021353 +2023-10-05 21:32:44,269 - Epoch: [107][ 990/ 1236] Overall Loss 0.251098 Objective Loss 0.251098 LR 0.000500 Time 0.021341 +2023-10-05 21:32:44,472 - Epoch: [107][ 1000/ 1236] Overall Loss 0.251155 Objective Loss 0.251155 LR 0.000500 Time 0.021330 +2023-10-05 21:32:44,673 - Epoch: [107][ 1010/ 1236] Overall Loss 0.251065 Objective Loss 0.251065 LR 0.000500 Time 0.021319 +2023-10-05 21:32:44,877 - Epoch: [107][ 1020/ 1236] Overall Loss 0.250841 Objective Loss 0.250841 LR 0.000500 Time 0.021309 +2023-10-05 21:32:45,079 - Epoch: [107][ 1030/ 1236] Overall Loss 0.250768 Objective Loss 0.250768 LR 0.000500 Time 0.021297 +2023-10-05 21:32:45,282 - Epoch: [107][ 1040/ 1236] Overall Loss 0.250972 Objective Loss 0.250972 LR 0.000500 Time 0.021287 +2023-10-05 21:32:45,484 - Epoch: [107][ 1050/ 1236] Overall Loss 0.251277 Objective Loss 0.251277 LR 0.000500 Time 0.021277 +2023-10-05 21:32:45,687 - Epoch: [107][ 1060/ 1236] Overall Loss 0.251622 Objective Loss 0.251622 LR 0.000500 Time 0.021268 +2023-10-05 21:32:45,889 - Epoch: [107][ 1070/ 1236] Overall Loss 0.251489 Objective Loss 0.251489 LR 0.000500 Time 0.021257 +2023-10-05 21:32:46,092 - Epoch: [107][ 1080/ 1236] Overall Loss 0.251783 Objective Loss 0.251783 LR 0.000500 Time 0.021248 +2023-10-05 21:32:46,294 - Epoch: [107][ 1090/ 1236] Overall Loss 0.251700 Objective Loss 0.251700 LR 0.000500 Time 0.021238 +2023-10-05 21:32:46,497 - Epoch: [107][ 1100/ 1236] Overall Loss 0.251681 Objective Loss 0.251681 LR 0.000500 Time 0.021229 +2023-10-05 21:32:46,699 - Epoch: [107][ 1110/ 1236] Overall Loss 0.251890 Objective Loss 0.251890 LR 0.000500 Time 0.021220 +2023-10-05 21:32:46,903 - Epoch: [107][ 1120/ 1236] Overall Loss 0.251917 Objective Loss 0.251917 LR 0.000500 Time 0.021212 +2023-10-05 21:32:47,106 - Epoch: [107][ 1130/ 1236] Overall Loss 0.251827 Objective Loss 0.251827 LR 0.000500 Time 0.021203 +2023-10-05 21:32:47,309 - Epoch: [107][ 1140/ 1236] Overall Loss 0.251930 Objective Loss 0.251930 LR 0.000500 Time 0.021195 +2023-10-05 21:32:47,520 - Epoch: [107][ 1150/ 1236] Overall Loss 0.251771 Objective Loss 0.251771 LR 0.000500 Time 0.021194 +2023-10-05 21:32:47,722 - Epoch: [107][ 1160/ 1236] Overall Loss 0.252017 Objective Loss 0.252017 LR 0.000500 Time 0.021185 +2023-10-05 21:32:47,923 - Epoch: [107][ 1170/ 1236] Overall Loss 0.251991 Objective Loss 0.251991 LR 0.000500 Time 0.021176 +2023-10-05 21:32:48,126 - Epoch: [107][ 1180/ 1236] Overall Loss 0.252264 Objective Loss 0.252264 LR 0.000500 Time 0.021167 +2023-10-05 21:32:48,326 - Epoch: [107][ 1190/ 1236] Overall Loss 0.252764 Objective Loss 0.252764 LR 0.000500 Time 0.021158 +2023-10-05 21:32:48,529 - Epoch: [107][ 1200/ 1236] Overall Loss 0.252894 Objective Loss 0.252894 LR 0.000500 Time 0.021150 +2023-10-05 21:32:48,729 - Epoch: [107][ 1210/ 1236] Overall Loss 0.253072 Objective Loss 0.253072 LR 0.000500 Time 0.021141 +2023-10-05 21:32:48,932 - Epoch: [107][ 1220/ 1236] Overall Loss 0.253080 Objective Loss 0.253080 LR 0.000500 Time 0.021133 +2023-10-05 21:32:49,188 - Epoch: [107][ 1230/ 1236] Overall Loss 0.253005 Objective Loss 0.253005 LR 0.000500 Time 0.021169 +2023-10-05 21:32:49,305 - Epoch: [107][ 1236/ 1236] Overall Loss 0.252840 Objective Loss 0.252840 Top1 83.706721 Top5 97.963340 LR 0.000500 Time 0.021161 +2023-10-05 21:32:49,424 - --- validate (epoch=107)----------- +2023-10-05 21:32:49,424 - 29943 samples (256 per mini-batch) +2023-10-05 21:32:49,902 - Epoch: [107][ 10/ 117] Loss 0.313467 Top1 84.414062 Top5 98.046875 +2023-10-05 21:32:50,051 - Epoch: [107][ 20/ 117] Loss 0.304586 Top1 84.179688 Top5 97.812500 +2023-10-05 21:32:50,200 - Epoch: [107][ 30/ 117] Loss 0.300593 Top1 84.388021 Top5 97.942708 +2023-10-05 21:32:50,348 - Epoch: [107][ 40/ 117] Loss 0.309674 Top1 84.160156 Top5 97.880859 +2023-10-05 21:32:50,498 - Epoch: [107][ 50/ 117] Loss 0.312649 Top1 84.195312 Top5 97.867188 +2023-10-05 21:32:50,647 - Epoch: [107][ 60/ 117] Loss 0.319399 Top1 84.069010 Top5 97.916667 +2023-10-05 21:32:50,796 - Epoch: [107][ 70/ 117] Loss 0.319394 Top1 84.006696 Top5 97.879464 +2023-10-05 21:32:50,945 - Epoch: [107][ 80/ 117] Loss 0.320392 Top1 84.042969 Top5 97.880859 +2023-10-05 21:32:51,095 - Epoch: [107][ 90/ 117] Loss 0.319421 Top1 84.040799 Top5 97.873264 +2023-10-05 21:32:51,245 - Epoch: [107][ 100/ 117] Loss 0.320681 Top1 83.996094 Top5 97.871094 +2023-10-05 21:32:51,403 - Epoch: [107][ 110/ 117] Loss 0.324348 Top1 83.927557 Top5 97.848011 +2023-10-05 21:32:51,489 - Epoch: [107][ 117/ 117] Loss 0.324917 Top1 83.886050 Top5 97.879304 +2023-10-05 21:32:51,634 - ==> Top1: 83.886 Top5: 97.879 Loss: 0.325 + +2023-10-05 21:32:51,635 - ==> Confusion: +[[ 919 2 5 0 11 5 0 0 5 75 2 0 0 1 5 2 6 1 0 0 11] + [ 1 1050 2 0 5 31 1 19 0 0 0 1 1 0 1 3 1 1 6 0 8] + [ 7 2 958 14 0 0 24 11 0 0 4 1 9 2 0 2 1 1 6 6 8] + [ 3 1 14 982 2 4 0 2 2 0 4 0 4 5 20 3 0 5 20 1 17] + [ 14 9 1 0 977 4 0 2 0 9 1 2 0 1 13 2 7 2 0 3 3] + [ 4 39 1 2 2 983 2 20 2 2 4 9 3 10 7 3 3 1 2 2 15] + [ 0 5 23 1 0 3 1118 10 0 0 1 3 2 0 1 8 0 0 0 7 9] + [ 2 23 16 2 3 22 3 1076 0 4 3 9 3 3 1 1 0 1 33 5 8] + [ 14 2 0 0 1 5 0 1 968 38 11 2 4 10 22 2 0 3 3 2 1] + [ 104 1 2 0 9 2 0 0 36 908 0 1 0 28 6 7 1 1 0 5 8] + [ 1 5 12 7 2 0 6 4 12 1 958 5 2 9 6 1 1 1 6 3 11] + [ 1 1 2 0 0 13 0 2 0 0 0 952 25 4 0 4 1 17 1 6 6] + [ 0 1 2 6 0 4 1 1 0 1 0 33 973 2 2 4 2 22 5 1 8] + [ 2 0 0 1 5 7 0 0 8 8 7 6 2 1054 4 1 2 0 0 1 11] + [ 9 4 3 17 7 2 0 0 22 0 1 2 1 2 999 0 1 0 18 0 13] + [ 1 4 4 0 3 0 0 0 0 0 0 7 8 2 0 1068 11 12 1 8 5] + [ 0 10 1 0 5 6 1 1 0 0 0 4 2 1 3 11 1101 0 1 3 11] + [ 0 0 0 2 0 0 1 0 2 0 0 2 19 0 2 7 0 998 1 2 2] + [ 2 8 5 14 2 1 0 21 2 0 2 2 5 0 8 0 0 1 988 0 7] + [ 1 3 1 0 2 8 11 9 1 0 0 17 4 1 0 7 11 2 1 1063 10] + [ 110 194 118 91 95 175 49 108 98 72 173 132 357 266 161 70 158 86 177 190 5025]] + +2023-10-05 21:32:51,636 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:32:51,636 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:32:51,643 - + +2023-10-05 21:32:51,643 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:32:52,648 - Epoch: [108][ 10/ 1236] Overall Loss 0.247021 Objective Loss 0.247021 LR 0.000500 Time 0.100383 +2023-10-05 21:32:52,849 - Epoch: [108][ 20/ 1236] Overall Loss 0.236222 Objective Loss 0.236222 LR 0.000500 Time 0.060234 +2023-10-05 21:32:53,048 - Epoch: [108][ 30/ 1236] Overall Loss 0.236388 Objective Loss 0.236388 LR 0.000500 Time 0.046787 +2023-10-05 21:32:53,250 - Epoch: [108][ 40/ 1236] Overall Loss 0.228928 Objective Loss 0.228928 LR 0.000500 Time 0.040115 +2023-10-05 21:32:53,448 - Epoch: [108][ 50/ 1236] Overall Loss 0.233766 Objective Loss 0.233766 LR 0.000500 Time 0.036063 +2023-10-05 21:32:53,650 - Epoch: [108][ 60/ 1236] Overall Loss 0.238482 Objective Loss 0.238482 LR 0.000500 Time 0.033403 +2023-10-05 21:32:53,848 - Epoch: [108][ 70/ 1236] Overall Loss 0.237884 Objective Loss 0.237884 LR 0.000500 Time 0.031463 +2023-10-05 21:32:54,050 - Epoch: [108][ 80/ 1236] Overall Loss 0.239834 Objective Loss 0.239834 LR 0.000500 Time 0.030044 +2023-10-05 21:32:54,248 - Epoch: [108][ 90/ 1236] Overall Loss 0.241606 Objective Loss 0.241606 LR 0.000500 Time 0.028907 +2023-10-05 21:32:54,450 - Epoch: [108][ 100/ 1236] Overall Loss 0.242382 Objective Loss 0.242382 LR 0.000500 Time 0.028026 +2023-10-05 21:32:54,648 - Epoch: [108][ 110/ 1236] Overall Loss 0.241027 Objective Loss 0.241027 LR 0.000500 Time 0.027281 +2023-10-05 21:32:54,849 - Epoch: [108][ 120/ 1236] Overall Loss 0.243754 Objective Loss 0.243754 LR 0.000500 Time 0.026676 +2023-10-05 21:32:55,048 - Epoch: [108][ 130/ 1236] Overall Loss 0.244336 Objective Loss 0.244336 LR 0.000500 Time 0.026151 +2023-10-05 21:32:55,249 - Epoch: [108][ 140/ 1236] Overall Loss 0.243166 Objective Loss 0.243166 LR 0.000500 Time 0.025716 +2023-10-05 21:32:55,447 - Epoch: [108][ 150/ 1236] Overall Loss 0.243066 Objective Loss 0.243066 LR 0.000500 Time 0.025325 +2023-10-05 21:32:55,649 - Epoch: [108][ 160/ 1236] Overall Loss 0.242870 Objective Loss 0.242870 LR 0.000500 Time 0.024998 +2023-10-05 21:32:55,847 - Epoch: [108][ 170/ 1236] Overall Loss 0.242322 Objective Loss 0.242322 LR 0.000500 Time 0.024695 +2023-10-05 21:32:56,049 - Epoch: [108][ 180/ 1236] Overall Loss 0.241936 Objective Loss 0.241936 LR 0.000500 Time 0.024439 +2023-10-05 21:32:56,248 - Epoch: [108][ 190/ 1236] Overall Loss 0.241217 Objective Loss 0.241217 LR 0.000500 Time 0.024198 +2023-10-05 21:32:56,449 - Epoch: [108][ 200/ 1236] Overall Loss 0.241928 Objective Loss 0.241928 LR 0.000500 Time 0.023993 +2023-10-05 21:32:56,650 - Epoch: [108][ 210/ 1236] Overall Loss 0.241256 Objective Loss 0.241256 LR 0.000500 Time 0.023807 +2023-10-05 21:32:56,850 - Epoch: [108][ 220/ 1236] Overall Loss 0.242077 Objective Loss 0.242077 LR 0.000500 Time 0.023633 +2023-10-05 21:32:57,052 - Epoch: [108][ 230/ 1236] Overall Loss 0.242433 Objective Loss 0.242433 LR 0.000500 Time 0.023482 +2023-10-05 21:32:57,256 - Epoch: [108][ 240/ 1236] Overall Loss 0.241695 Objective Loss 0.241695 LR 0.000500 Time 0.023351 +2023-10-05 21:32:57,465 - Epoch: [108][ 250/ 1236] Overall Loss 0.242677 Objective Loss 0.242677 LR 0.000500 Time 0.023253 +2023-10-05 21:32:57,670 - Epoch: [108][ 260/ 1236] Overall Loss 0.242424 Objective Loss 0.242424 LR 0.000500 Time 0.023146 +2023-10-05 21:32:57,873 - Epoch: [108][ 270/ 1236] Overall Loss 0.243510 Objective Loss 0.243510 LR 0.000500 Time 0.023038 +2023-10-05 21:32:58,072 - Epoch: [108][ 280/ 1236] Overall Loss 0.244262 Objective Loss 0.244262 LR 0.000500 Time 0.022923 +2023-10-05 21:32:58,272 - Epoch: [108][ 290/ 1236] Overall Loss 0.244986 Objective Loss 0.244986 LR 0.000500 Time 0.022823 +2023-10-05 21:32:58,471 - Epoch: [108][ 300/ 1236] Overall Loss 0.244966 Objective Loss 0.244966 LR 0.000500 Time 0.022724 +2023-10-05 21:32:58,671 - Epoch: [108][ 310/ 1236] Overall Loss 0.244632 Objective Loss 0.244632 LR 0.000500 Time 0.022636 +2023-10-05 21:32:58,870 - Epoch: [108][ 320/ 1236] Overall Loss 0.244803 Objective Loss 0.244803 LR 0.000500 Time 0.022548 +2023-10-05 21:32:59,068 - Epoch: [108][ 330/ 1236] Overall Loss 0.244070 Objective Loss 0.244070 LR 0.000500 Time 0.022464 +2023-10-05 21:32:59,267 - Epoch: [108][ 340/ 1236] Overall Loss 0.243520 Objective Loss 0.243520 LR 0.000500 Time 0.022388 +2023-10-05 21:32:59,467 - Epoch: [108][ 350/ 1236] Overall Loss 0.243428 Objective Loss 0.243428 LR 0.000500 Time 0.022319 +2023-10-05 21:32:59,666 - Epoch: [108][ 360/ 1236] Overall Loss 0.243748 Objective Loss 0.243748 LR 0.000500 Time 0.022250 +2023-10-05 21:32:59,868 - Epoch: [108][ 370/ 1236] Overall Loss 0.242706 Objective Loss 0.242706 LR 0.000500 Time 0.022193 +2023-10-05 21:33:00,073 - Epoch: [108][ 380/ 1236] Overall Loss 0.242586 Objective Loss 0.242586 LR 0.000500 Time 0.022149 +2023-10-05 21:33:00,283 - Epoch: [108][ 390/ 1236] Overall Loss 0.242670 Objective Loss 0.242670 LR 0.000500 Time 0.022117 +2023-10-05 21:33:00,488 - Epoch: [108][ 400/ 1236] Overall Loss 0.243330 Objective Loss 0.243330 LR 0.000500 Time 0.022077 +2023-10-05 21:33:00,697 - Epoch: [108][ 410/ 1236] Overall Loss 0.243350 Objective Loss 0.243350 LR 0.000500 Time 0.022047 +2023-10-05 21:33:00,903 - Epoch: [108][ 420/ 1236] Overall Loss 0.243378 Objective Loss 0.243378 LR 0.000500 Time 0.022012 +2023-10-05 21:33:01,111 - Epoch: [108][ 430/ 1236] Overall Loss 0.243484 Objective Loss 0.243484 LR 0.000500 Time 0.021984 +2023-10-05 21:33:01,311 - Epoch: [108][ 440/ 1236] Overall Loss 0.243617 Objective Loss 0.243617 LR 0.000500 Time 0.021938 +2023-10-05 21:33:01,514 - Epoch: [108][ 450/ 1236] Overall Loss 0.243712 Objective Loss 0.243712 LR 0.000500 Time 0.021901 +2023-10-05 21:33:01,715 - Epoch: [108][ 460/ 1236] Overall Loss 0.243502 Objective Loss 0.243502 LR 0.000500 Time 0.021861 +2023-10-05 21:33:01,918 - Epoch: [108][ 470/ 1236] Overall Loss 0.244170 Objective Loss 0.244170 LR 0.000500 Time 0.021826 +2023-10-05 21:33:02,119 - Epoch: [108][ 480/ 1236] Overall Loss 0.244385 Objective Loss 0.244385 LR 0.000500 Time 0.021789 +2023-10-05 21:33:02,322 - Epoch: [108][ 490/ 1236] Overall Loss 0.244803 Objective Loss 0.244803 LR 0.000500 Time 0.021759 +2023-10-05 21:33:02,525 - Epoch: [108][ 500/ 1236] Overall Loss 0.245053 Objective Loss 0.245053 LR 0.000500 Time 0.021730 +2023-10-05 21:33:02,729 - Epoch: [108][ 510/ 1236] Overall Loss 0.244976 Objective Loss 0.244976 LR 0.000500 Time 0.021703 +2023-10-05 21:33:02,933 - Epoch: [108][ 520/ 1236] Overall Loss 0.245612 Objective Loss 0.245612 LR 0.000500 Time 0.021676 +2023-10-05 21:33:03,140 - Epoch: [108][ 530/ 1236] Overall Loss 0.245988 Objective Loss 0.245988 LR 0.000500 Time 0.021659 +2023-10-05 21:33:03,347 - Epoch: [108][ 540/ 1236] Overall Loss 0.245990 Objective Loss 0.245990 LR 0.000500 Time 0.021639 +2023-10-05 21:33:03,559 - Epoch: [108][ 550/ 1236] Overall Loss 0.246585 Objective Loss 0.246585 LR 0.000500 Time 0.021631 +2023-10-05 21:33:03,765 - Epoch: [108][ 560/ 1236] Overall Loss 0.246742 Objective Loss 0.246742 LR 0.000500 Time 0.021611 +2023-10-05 21:33:03,974 - Epoch: [108][ 570/ 1236] Overall Loss 0.247147 Objective Loss 0.247147 LR 0.000500 Time 0.021598 +2023-10-05 21:33:04,175 - Epoch: [108][ 580/ 1236] Overall Loss 0.247885 Objective Loss 0.247885 LR 0.000500 Time 0.021572 +2023-10-05 21:33:04,378 - Epoch: [108][ 590/ 1236] Overall Loss 0.248147 Objective Loss 0.248147 LR 0.000500 Time 0.021551 +2023-10-05 21:33:04,578 - Epoch: [108][ 600/ 1236] Overall Loss 0.248192 Objective Loss 0.248192 LR 0.000500 Time 0.021524 +2023-10-05 21:33:04,780 - Epoch: [108][ 610/ 1236] Overall Loss 0.248834 Objective Loss 0.248834 LR 0.000500 Time 0.021501 +2023-10-05 21:33:04,980 - Epoch: [108][ 620/ 1236] Overall Loss 0.248550 Objective Loss 0.248550 LR 0.000500 Time 0.021476 +2023-10-05 21:33:05,182 - Epoch: [108][ 630/ 1236] Overall Loss 0.248473 Objective Loss 0.248473 LR 0.000500 Time 0.021455 +2023-10-05 21:33:05,381 - Epoch: [108][ 640/ 1236] Overall Loss 0.248769 Objective Loss 0.248769 LR 0.000500 Time 0.021432 +2023-10-05 21:33:05,583 - Epoch: [108][ 650/ 1236] Overall Loss 0.249006 Objective Loss 0.249006 LR 0.000500 Time 0.021412 +2023-10-05 21:33:05,783 - Epoch: [108][ 660/ 1236] Overall Loss 0.249081 Objective Loss 0.249081 LR 0.000500 Time 0.021390 +2023-10-05 21:33:05,986 - Epoch: [108][ 670/ 1236] Overall Loss 0.249053 Objective Loss 0.249053 LR 0.000500 Time 0.021372 +2023-10-05 21:33:06,186 - Epoch: [108][ 680/ 1236] Overall Loss 0.248985 Objective Loss 0.248985 LR 0.000500 Time 0.021352 +2023-10-05 21:33:06,388 - Epoch: [108][ 690/ 1236] Overall Loss 0.248797 Objective Loss 0.248797 LR 0.000500 Time 0.021334 +2023-10-05 21:33:06,587 - Epoch: [108][ 700/ 1236] Overall Loss 0.248643 Objective Loss 0.248643 LR 0.000500 Time 0.021314 +2023-10-05 21:33:06,789 - Epoch: [108][ 710/ 1236] Overall Loss 0.248547 Objective Loss 0.248547 LR 0.000500 Time 0.021298 +2023-10-05 21:33:06,989 - Epoch: [108][ 720/ 1236] Overall Loss 0.248476 Objective Loss 0.248476 LR 0.000500 Time 0.021280 +2023-10-05 21:33:07,191 - Epoch: [108][ 730/ 1236] Overall Loss 0.248285 Objective Loss 0.248285 LR 0.000500 Time 0.021264 +2023-10-05 21:33:07,391 - Epoch: [108][ 740/ 1236] Overall Loss 0.248394 Objective Loss 0.248394 LR 0.000500 Time 0.021246 +2023-10-05 21:33:07,593 - Epoch: [108][ 750/ 1236] Overall Loss 0.248559 Objective Loss 0.248559 LR 0.000500 Time 0.021232 +2023-10-05 21:33:07,793 - Epoch: [108][ 760/ 1236] Overall Loss 0.248954 Objective Loss 0.248954 LR 0.000500 Time 0.021215 +2023-10-05 21:33:07,995 - Epoch: [108][ 770/ 1236] Overall Loss 0.249097 Objective Loss 0.249097 LR 0.000500 Time 0.021201 +2023-10-05 21:33:08,194 - Epoch: [108][ 780/ 1236] Overall Loss 0.249088 Objective Loss 0.249088 LR 0.000500 Time 0.021185 +2023-10-05 21:33:08,396 - Epoch: [108][ 790/ 1236] Overall Loss 0.248840 Objective Loss 0.248840 LR 0.000500 Time 0.021172 +2023-10-05 21:33:08,596 - Epoch: [108][ 800/ 1236] Overall Loss 0.249047 Objective Loss 0.249047 LR 0.000500 Time 0.021156 +2023-10-05 21:33:08,798 - Epoch: [108][ 810/ 1236] Overall Loss 0.248740 Objective Loss 0.248740 LR 0.000500 Time 0.021144 +2023-10-05 21:33:08,998 - Epoch: [108][ 820/ 1236] Overall Loss 0.248889 Objective Loss 0.248889 LR 0.000500 Time 0.021129 +2023-10-05 21:33:09,200 - Epoch: [108][ 830/ 1236] Overall Loss 0.249200 Objective Loss 0.249200 LR 0.000500 Time 0.021118 +2023-10-05 21:33:09,400 - Epoch: [108][ 840/ 1236] Overall Loss 0.249077 Objective Loss 0.249077 LR 0.000500 Time 0.021104 +2023-10-05 21:33:09,602 - Epoch: [108][ 850/ 1236] Overall Loss 0.248865 Objective Loss 0.248865 LR 0.000500 Time 0.021093 +2023-10-05 21:33:09,801 - Epoch: [108][ 860/ 1236] Overall Loss 0.248674 Objective Loss 0.248674 LR 0.000500 Time 0.021080 +2023-10-05 21:33:10,003 - Epoch: [108][ 870/ 1236] Overall Loss 0.248674 Objective Loss 0.248674 LR 0.000500 Time 0.021069 +2023-10-05 21:33:10,203 - Epoch: [108][ 880/ 1236] Overall Loss 0.248903 Objective Loss 0.248903 LR 0.000500 Time 0.021056 +2023-10-05 21:33:10,405 - Epoch: [108][ 890/ 1236] Overall Loss 0.248976 Objective Loss 0.248976 LR 0.000500 Time 0.021046 +2023-10-05 21:33:10,605 - Epoch: [108][ 900/ 1236] Overall Loss 0.249427 Objective Loss 0.249427 LR 0.000500 Time 0.021034 +2023-10-05 21:33:10,806 - Epoch: [108][ 910/ 1236] Overall Loss 0.249572 Objective Loss 0.249572 LR 0.000500 Time 0.021024 +2023-10-05 21:33:11,006 - Epoch: [108][ 920/ 1236] Overall Loss 0.249608 Objective Loss 0.249608 LR 0.000500 Time 0.021012 +2023-10-05 21:33:11,208 - Epoch: [108][ 930/ 1236] Overall Loss 0.249791 Objective Loss 0.249791 LR 0.000500 Time 0.021003 +2023-10-05 21:33:11,408 - Epoch: [108][ 940/ 1236] Overall Loss 0.249872 Objective Loss 0.249872 LR 0.000500 Time 0.020992 +2023-10-05 21:33:11,610 - Epoch: [108][ 950/ 1236] Overall Loss 0.249903 Objective Loss 0.249903 LR 0.000500 Time 0.020983 +2023-10-05 21:33:11,810 - Epoch: [108][ 960/ 1236] Overall Loss 0.249786 Objective Loss 0.249786 LR 0.000500 Time 0.020972 +2023-10-05 21:33:12,012 - Epoch: [108][ 970/ 1236] Overall Loss 0.249637 Objective Loss 0.249637 LR 0.000500 Time 0.020964 +2023-10-05 21:33:12,212 - Epoch: [108][ 980/ 1236] Overall Loss 0.249729 Objective Loss 0.249729 LR 0.000500 Time 0.020954 +2023-10-05 21:33:12,414 - Epoch: [108][ 990/ 1236] Overall Loss 0.249631 Objective Loss 0.249631 LR 0.000500 Time 0.020946 +2023-10-05 21:33:12,614 - Epoch: [108][ 1000/ 1236] Overall Loss 0.249632 Objective Loss 0.249632 LR 0.000500 Time 0.020936 +2023-10-05 21:33:12,817 - Epoch: [108][ 1010/ 1236] Overall Loss 0.249497 Objective Loss 0.249497 LR 0.000500 Time 0.020929 +2023-10-05 21:33:13,017 - Epoch: [108][ 1020/ 1236] Overall Loss 0.249381 Objective Loss 0.249381 LR 0.000500 Time 0.020919 +2023-10-05 21:33:13,219 - Epoch: [108][ 1030/ 1236] Overall Loss 0.249337 Objective Loss 0.249337 LR 0.000500 Time 0.020912 +2023-10-05 21:33:13,419 - Epoch: [108][ 1040/ 1236] Overall Loss 0.249700 Objective Loss 0.249700 LR 0.000500 Time 0.020903 +2023-10-05 21:33:13,621 - Epoch: [108][ 1050/ 1236] Overall Loss 0.249554 Objective Loss 0.249554 LR 0.000500 Time 0.020896 +2023-10-05 21:33:13,821 - Epoch: [108][ 1060/ 1236] Overall Loss 0.249571 Objective Loss 0.249571 LR 0.000500 Time 0.020887 +2023-10-05 21:33:14,023 - Epoch: [108][ 1070/ 1236] Overall Loss 0.249533 Objective Loss 0.249533 LR 0.000500 Time 0.020881 +2023-10-05 21:33:14,223 - Epoch: [108][ 1080/ 1236] Overall Loss 0.249408 Objective Loss 0.249408 LR 0.000500 Time 0.020872 +2023-10-05 21:33:14,426 - Epoch: [108][ 1090/ 1236] Overall Loss 0.249281 Objective Loss 0.249281 LR 0.000500 Time 0.020866 +2023-10-05 21:33:14,625 - Epoch: [108][ 1100/ 1236] Overall Loss 0.249411 Objective Loss 0.249411 LR 0.000500 Time 0.020858 +2023-10-05 21:33:14,828 - Epoch: [108][ 1110/ 1236] Overall Loss 0.249240 Objective Loss 0.249240 LR 0.000500 Time 0.020852 +2023-10-05 21:33:15,028 - Epoch: [108][ 1120/ 1236] Overall Loss 0.249521 Objective Loss 0.249521 LR 0.000500 Time 0.020844 +2023-10-05 21:33:15,230 - Epoch: [108][ 1130/ 1236] Overall Loss 0.249723 Objective Loss 0.249723 LR 0.000500 Time 0.020838 +2023-10-05 21:33:15,430 - Epoch: [108][ 1140/ 1236] Overall Loss 0.249803 Objective Loss 0.249803 LR 0.000500 Time 0.020830 +2023-10-05 21:33:15,632 - Epoch: [108][ 1150/ 1236] Overall Loss 0.249915 Objective Loss 0.249915 LR 0.000500 Time 0.020825 +2023-10-05 21:33:15,832 - Epoch: [108][ 1160/ 1236] Overall Loss 0.249702 Objective Loss 0.249702 LR 0.000500 Time 0.020817 +2023-10-05 21:33:16,034 - Epoch: [108][ 1170/ 1236] Overall Loss 0.249872 Objective Loss 0.249872 LR 0.000500 Time 0.020812 +2023-10-05 21:33:16,234 - Epoch: [108][ 1180/ 1236] Overall Loss 0.249879 Objective Loss 0.249879 LR 0.000500 Time 0.020805 +2023-10-05 21:33:16,436 - Epoch: [108][ 1190/ 1236] Overall Loss 0.250087 Objective Loss 0.250087 LR 0.000500 Time 0.020800 +2023-10-05 21:33:16,636 - Epoch: [108][ 1200/ 1236] Overall Loss 0.250196 Objective Loss 0.250196 LR 0.000500 Time 0.020792 +2023-10-05 21:33:16,839 - Epoch: [108][ 1210/ 1236] Overall Loss 0.250237 Objective Loss 0.250237 LR 0.000500 Time 0.020788 +2023-10-05 21:33:17,039 - Epoch: [108][ 1220/ 1236] Overall Loss 0.250418 Objective Loss 0.250418 LR 0.000500 Time 0.020781 +2023-10-05 21:33:17,291 - Epoch: [108][ 1230/ 1236] Overall Loss 0.250506 Objective Loss 0.250506 LR 0.000500 Time 0.020817 +2023-10-05 21:33:17,409 - Epoch: [108][ 1236/ 1236] Overall Loss 0.250422 Objective Loss 0.250422 Top1 86.354379 Top5 98.574338 LR 0.000500 Time 0.020811 +2023-10-05 21:33:17,556 - --- validate (epoch=108)----------- +2023-10-05 21:33:17,556 - 29943 samples (256 per mini-batch) +2023-10-05 21:33:18,007 - Epoch: [108][ 10/ 117] Loss 0.297213 Top1 85.039062 Top5 97.695312 +2023-10-05 21:33:18,152 - Epoch: [108][ 20/ 117] Loss 0.304639 Top1 84.511719 Top5 97.871094 +2023-10-05 21:33:18,296 - Epoch: [108][ 30/ 117] Loss 0.318051 Top1 83.893229 Top5 97.747396 +2023-10-05 21:33:18,439 - Epoch: [108][ 40/ 117] Loss 0.317322 Top1 83.701172 Top5 97.900391 +2023-10-05 21:33:18,583 - Epoch: [108][ 50/ 117] Loss 0.316781 Top1 83.632812 Top5 97.976562 +2023-10-05 21:33:18,728 - Epoch: [108][ 60/ 117] Loss 0.324097 Top1 83.404948 Top5 97.929688 +2023-10-05 21:33:18,872 - Epoch: [108][ 70/ 117] Loss 0.319074 Top1 83.476562 Top5 97.957589 +2023-10-05 21:33:19,018 - Epoch: [108][ 80/ 117] Loss 0.326967 Top1 83.266602 Top5 97.919922 +2023-10-05 21:33:19,163 - Epoch: [108][ 90/ 117] Loss 0.322807 Top1 83.337674 Top5 97.925347 +2023-10-05 21:33:19,309 - Epoch: [108][ 100/ 117] Loss 0.325293 Top1 83.386719 Top5 97.917969 +2023-10-05 21:33:19,461 - Epoch: [108][ 110/ 117] Loss 0.328279 Top1 83.270597 Top5 97.915483 +2023-10-05 21:33:19,546 - Epoch: [108][ 117/ 117] Loss 0.326861 Top1 83.318305 Top5 97.912701 +2023-10-05 21:33:19,666 - ==> Top1: 83.318 Top5: 97.913 Loss: 0.327 + +2023-10-05 21:33:19,666 - ==> Confusion: +[[ 939 1 5 1 5 2 0 0 5 59 2 2 1 3 6 0 5 2 2 2 8] + [ 1 1043 2 0 11 22 1 16 4 0 2 1 0 0 1 3 6 1 8 1 8] + [ 5 0 954 6 4 0 30 5 0 0 11 2 9 0 1 4 4 0 5 6 10] + [ 3 2 15 956 1 2 2 2 6 3 10 0 8 4 28 4 0 7 19 3 14] + [ 23 5 0 0 969 7 0 3 1 12 1 1 0 2 8 3 9 2 0 2 2] + [ 6 52 0 0 5 970 3 17 0 2 5 8 2 11 6 1 1 0 4 8 15] + [ 1 3 23 0 0 1 1125 6 0 0 3 3 1 0 1 9 0 1 0 8 6] + [ 6 25 25 0 2 32 5 1046 1 3 4 8 1 4 0 0 0 1 37 8 10] + [ 17 2 0 0 1 2 0 0 977 37 12 1 4 8 19 2 1 0 2 3 1] + [ 104 0 4 0 4 5 2 0 26 923 1 4 0 22 6 0 3 0 2 5 8] + [ 2 5 13 4 1 1 4 0 10 1 966 2 0 16 6 1 1 0 6 1 13] + [ 1 0 2 0 2 11 0 3 0 1 0 948 24 5 0 2 3 14 0 17 2] + [ 2 1 6 5 1 3 0 0 1 0 1 42 962 9 2 6 1 12 3 4 7] + [ 2 1 2 1 5 13 1 2 8 13 4 3 1 1049 3 2 1 0 0 2 6] + [ 10 0 3 5 10 0 0 0 23 3 1 1 4 1 1009 0 0 2 14 0 15] + [ 2 2 1 0 5 1 1 0 0 0 0 8 7 4 0 1067 17 9 0 7 3] + [ 1 12 3 0 2 7 1 3 0 0 0 4 1 0 4 14 1095 0 1 5 8] + [ 0 2 0 0 0 0 2 0 1 0 0 4 25 0 2 5 1 987 1 4 4] + [ 2 7 9 11 0 1 1 20 3 0 5 1 2 0 15 0 1 0 984 1 5] + [ 0 3 1 1 1 3 9 11 1 0 0 11 2 2 0 8 10 0 2 1075 12] + [ 153 217 147 74 94 137 59 97 104 75 190 113 344 277 203 60 189 74 162 232 4904]] + +2023-10-05 21:33:19,668 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:33:19,668 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:33:19,674 - + +2023-10-05 21:33:19,674 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:33:20,666 - Epoch: [109][ 10/ 1236] Overall Loss 0.217797 Objective Loss 0.217797 LR 0.000500 Time 0.099158 +2023-10-05 21:33:20,877 - Epoch: [109][ 20/ 1236] Overall Loss 0.241288 Objective Loss 0.241288 LR 0.000500 Time 0.060131 +2023-10-05 21:33:21,085 - Epoch: [109][ 30/ 1236] Overall Loss 0.244346 Objective Loss 0.244346 LR 0.000500 Time 0.046982 +2023-10-05 21:33:21,296 - Epoch: [109][ 40/ 1236] Overall Loss 0.243728 Objective Loss 0.243728 LR 0.000500 Time 0.040516 +2023-10-05 21:33:21,503 - Epoch: [109][ 50/ 1236] Overall Loss 0.243350 Objective Loss 0.243350 LR 0.000500 Time 0.036545 +2023-10-05 21:33:21,714 - Epoch: [109][ 60/ 1236] Overall Loss 0.247191 Objective Loss 0.247191 LR 0.000500 Time 0.033973 +2023-10-05 21:33:21,921 - Epoch: [109][ 70/ 1236] Overall Loss 0.247517 Objective Loss 0.247517 LR 0.000500 Time 0.032073 +2023-10-05 21:33:22,132 - Epoch: [109][ 80/ 1236] Overall Loss 0.247783 Objective Loss 0.247783 LR 0.000500 Time 0.030698 +2023-10-05 21:33:22,340 - Epoch: [109][ 90/ 1236] Overall Loss 0.251745 Objective Loss 0.251745 LR 0.000500 Time 0.029586 +2023-10-05 21:33:22,551 - Epoch: [109][ 100/ 1236] Overall Loss 0.250818 Objective Loss 0.250818 LR 0.000500 Time 0.028737 +2023-10-05 21:33:22,757 - Epoch: [109][ 110/ 1236] Overall Loss 0.249982 Objective Loss 0.249982 LR 0.000500 Time 0.027992 +2023-10-05 21:33:22,967 - Epoch: [109][ 120/ 1236] Overall Loss 0.248338 Objective Loss 0.248338 LR 0.000500 Time 0.027412 +2023-10-05 21:33:23,173 - Epoch: [109][ 130/ 1236] Overall Loss 0.247950 Objective Loss 0.247950 LR 0.000500 Time 0.026885 +2023-10-05 21:33:23,383 - Epoch: [109][ 140/ 1236] Overall Loss 0.248837 Objective Loss 0.248837 LR 0.000500 Time 0.026464 +2023-10-05 21:33:23,589 - Epoch: [109][ 150/ 1236] Overall Loss 0.249528 Objective Loss 0.249528 LR 0.000500 Time 0.026070 +2023-10-05 21:33:23,799 - Epoch: [109][ 160/ 1236] Overall Loss 0.249519 Objective Loss 0.249519 LR 0.000500 Time 0.025746 +2023-10-05 21:33:24,007 - Epoch: [109][ 170/ 1236] Overall Loss 0.247482 Objective Loss 0.247482 LR 0.000500 Time 0.025456 +2023-10-05 21:33:24,220 - Epoch: [109][ 180/ 1236] Overall Loss 0.248966 Objective Loss 0.248966 LR 0.000500 Time 0.025224 +2023-10-05 21:33:24,429 - Epoch: [109][ 190/ 1236] Overall Loss 0.247170 Objective Loss 0.247170 LR 0.000500 Time 0.024992 +2023-10-05 21:33:24,634 - Epoch: [109][ 200/ 1236] Overall Loss 0.246169 Objective Loss 0.246169 LR 0.000500 Time 0.024770 +2023-10-05 21:33:24,836 - Epoch: [109][ 210/ 1236] Overall Loss 0.246163 Objective Loss 0.246163 LR 0.000500 Time 0.024548 +2023-10-05 21:33:25,039 - Epoch: [109][ 220/ 1236] Overall Loss 0.246035 Objective Loss 0.246035 LR 0.000500 Time 0.024353 +2023-10-05 21:33:25,240 - Epoch: [109][ 230/ 1236] Overall Loss 0.245022 Objective Loss 0.245022 LR 0.000500 Time 0.024169 +2023-10-05 21:33:25,443 - Epoch: [109][ 240/ 1236] Overall Loss 0.244292 Objective Loss 0.244292 LR 0.000500 Time 0.024005 +2023-10-05 21:33:25,645 - Epoch: [109][ 250/ 1236] Overall Loss 0.244126 Objective Loss 0.244126 LR 0.000500 Time 0.023853 +2023-10-05 21:33:25,848 - Epoch: [109][ 260/ 1236] Overall Loss 0.244326 Objective Loss 0.244326 LR 0.000500 Time 0.023714 +2023-10-05 21:33:26,050 - Epoch: [109][ 270/ 1236] Overall Loss 0.244471 Objective Loss 0.244471 LR 0.000500 Time 0.023582 +2023-10-05 21:33:26,253 - Epoch: [109][ 280/ 1236] Overall Loss 0.244053 Objective Loss 0.244053 LR 0.000500 Time 0.023462 +2023-10-05 21:33:26,456 - Epoch: [109][ 290/ 1236] Overall Loss 0.244276 Objective Loss 0.244276 LR 0.000500 Time 0.023353 +2023-10-05 21:33:26,662 - Epoch: [109][ 300/ 1236] Overall Loss 0.244731 Objective Loss 0.244731 LR 0.000500 Time 0.023259 +2023-10-05 21:33:26,866 - Epoch: [109][ 310/ 1236] Overall Loss 0.244922 Objective Loss 0.244922 LR 0.000500 Time 0.023167 +2023-10-05 21:33:27,072 - Epoch: [109][ 320/ 1236] Overall Loss 0.245300 Objective Loss 0.245300 LR 0.000500 Time 0.023085 +2023-10-05 21:33:27,276 - Epoch: [109][ 330/ 1236] Overall Loss 0.245076 Objective Loss 0.245076 LR 0.000500 Time 0.023004 +2023-10-05 21:33:27,482 - Epoch: [109][ 340/ 1236] Overall Loss 0.245274 Objective Loss 0.245274 LR 0.000500 Time 0.022931 +2023-10-05 21:33:27,686 - Epoch: [109][ 350/ 1236] Overall Loss 0.244409 Objective Loss 0.244409 LR 0.000500 Time 0.022858 +2023-10-05 21:33:27,893 - Epoch: [109][ 360/ 1236] Overall Loss 0.244498 Objective Loss 0.244498 LR 0.000500 Time 0.022796 +2023-10-05 21:33:28,097 - Epoch: [109][ 370/ 1236] Overall Loss 0.244335 Objective Loss 0.244335 LR 0.000500 Time 0.022731 +2023-10-05 21:33:28,301 - Epoch: [109][ 380/ 1236] Overall Loss 0.243577 Objective Loss 0.243577 LR 0.000500 Time 0.022668 +2023-10-05 21:33:28,504 - Epoch: [109][ 390/ 1236] Overall Loss 0.243414 Objective Loss 0.243414 LR 0.000500 Time 0.022607 +2023-10-05 21:33:28,707 - Epoch: [109][ 400/ 1236] Overall Loss 0.243874 Objective Loss 0.243874 LR 0.000500 Time 0.022549 +2023-10-05 21:33:28,910 - Epoch: [109][ 410/ 1236] Overall Loss 0.244122 Objective Loss 0.244122 LR 0.000500 Time 0.022493 +2023-10-05 21:33:29,114 - Epoch: [109][ 420/ 1236] Overall Loss 0.245084 Objective Loss 0.245084 LR 0.000500 Time 0.022441 +2023-10-05 21:33:29,317 - Epoch: [109][ 430/ 1236] Overall Loss 0.245353 Objective Loss 0.245353 LR 0.000500 Time 0.022392 +2023-10-05 21:33:29,529 - Epoch: [109][ 440/ 1236] Overall Loss 0.245195 Objective Loss 0.245195 LR 0.000500 Time 0.022365 +2023-10-05 21:33:29,733 - Epoch: [109][ 450/ 1236] Overall Loss 0.244736 Objective Loss 0.244736 LR 0.000500 Time 0.022319 +2023-10-05 21:33:29,936 - Epoch: [109][ 460/ 1236] Overall Loss 0.243628 Objective Loss 0.243628 LR 0.000500 Time 0.022275 +2023-10-05 21:33:30,139 - Epoch: [109][ 470/ 1236] Overall Loss 0.243090 Objective Loss 0.243090 LR 0.000500 Time 0.022232 +2023-10-05 21:33:30,342 - Epoch: [109][ 480/ 1236] Overall Loss 0.242647 Objective Loss 0.242647 LR 0.000500 Time 0.022192 +2023-10-05 21:33:30,546 - Epoch: [109][ 490/ 1236] Overall Loss 0.242586 Objective Loss 0.242586 LR 0.000500 Time 0.022153 +2023-10-05 21:33:30,750 - Epoch: [109][ 500/ 1236] Overall Loss 0.242803 Objective Loss 0.242803 LR 0.000500 Time 0.022119 +2023-10-05 21:33:30,955 - Epoch: [109][ 510/ 1236] Overall Loss 0.242675 Objective Loss 0.242675 LR 0.000500 Time 0.022086 +2023-10-05 21:33:31,159 - Epoch: [109][ 520/ 1236] Overall Loss 0.242558 Objective Loss 0.242558 LR 0.000500 Time 0.022052 +2023-10-05 21:33:31,362 - Epoch: [109][ 530/ 1236] Overall Loss 0.242838 Objective Loss 0.242838 LR 0.000500 Time 0.022018 +2023-10-05 21:33:31,565 - Epoch: [109][ 540/ 1236] Overall Loss 0.242951 Objective Loss 0.242951 LR 0.000500 Time 0.021986 +2023-10-05 21:33:31,767 - Epoch: [109][ 550/ 1236] Overall Loss 0.243068 Objective Loss 0.243068 LR 0.000500 Time 0.021954 +2023-10-05 21:33:31,970 - Epoch: [109][ 560/ 1236] Overall Loss 0.244021 Objective Loss 0.244021 LR 0.000500 Time 0.021924 +2023-10-05 21:33:32,174 - Epoch: [109][ 570/ 1236] Overall Loss 0.244492 Objective Loss 0.244492 LR 0.000500 Time 0.021896 +2023-10-05 21:33:32,377 - Epoch: [109][ 580/ 1236] Overall Loss 0.244798 Objective Loss 0.244798 LR 0.000500 Time 0.021868 +2023-10-05 21:33:32,580 - Epoch: [109][ 590/ 1236] Overall Loss 0.245054 Objective Loss 0.245054 LR 0.000500 Time 0.021841 +2023-10-05 21:33:32,784 - Epoch: [109][ 600/ 1236] Overall Loss 0.244809 Objective Loss 0.244809 LR 0.000500 Time 0.021816 +2023-10-05 21:33:32,986 - Epoch: [109][ 610/ 1236] Overall Loss 0.244805 Objective Loss 0.244805 LR 0.000500 Time 0.021789 +2023-10-05 21:33:33,191 - Epoch: [109][ 620/ 1236] Overall Loss 0.244699 Objective Loss 0.244699 LR 0.000500 Time 0.021768 +2023-10-05 21:33:33,394 - Epoch: [109][ 630/ 1236] Overall Loss 0.244953 Objective Loss 0.244953 LR 0.000500 Time 0.021743 +2023-10-05 21:33:33,598 - Epoch: [109][ 640/ 1236] Overall Loss 0.245415 Objective Loss 0.245415 LR 0.000500 Time 0.021722 +2023-10-05 21:33:33,801 - Epoch: [109][ 650/ 1236] Overall Loss 0.245303 Objective Loss 0.245303 LR 0.000500 Time 0.021700 +2023-10-05 21:33:34,006 - Epoch: [109][ 660/ 1236] Overall Loss 0.245124 Objective Loss 0.245124 LR 0.000500 Time 0.021682 +2023-10-05 21:33:34,223 - Epoch: [109][ 670/ 1236] Overall Loss 0.245290 Objective Loss 0.245290 LR 0.000500 Time 0.021681 +2023-10-05 21:33:34,428 - Epoch: [109][ 680/ 1236] Overall Loss 0.245385 Objective Loss 0.245385 LR 0.000500 Time 0.021664 +2023-10-05 21:33:34,632 - Epoch: [109][ 690/ 1236] Overall Loss 0.245604 Objective Loss 0.245604 LR 0.000500 Time 0.021644 +2023-10-05 21:33:34,836 - Epoch: [109][ 700/ 1236] Overall Loss 0.245331 Objective Loss 0.245331 LR 0.000500 Time 0.021627 +2023-10-05 21:33:35,039 - Epoch: [109][ 710/ 1236] Overall Loss 0.244846 Objective Loss 0.244846 LR 0.000500 Time 0.021607 +2023-10-05 21:33:35,244 - Epoch: [109][ 720/ 1236] Overall Loss 0.245051 Objective Loss 0.245051 LR 0.000500 Time 0.021591 +2023-10-05 21:33:35,447 - Epoch: [109][ 730/ 1236] Overall Loss 0.245205 Objective Loss 0.245205 LR 0.000500 Time 0.021573 +2023-10-05 21:33:35,651 - Epoch: [109][ 740/ 1236] Overall Loss 0.245114 Objective Loss 0.245114 LR 0.000500 Time 0.021557 +2023-10-05 21:33:35,854 - Epoch: [109][ 750/ 1236] Overall Loss 0.245028 Objective Loss 0.245028 LR 0.000500 Time 0.021539 +2023-10-05 21:33:36,058 - Epoch: [109][ 760/ 1236] Overall Loss 0.245257 Objective Loss 0.245257 LR 0.000500 Time 0.021524 +2023-10-05 21:33:36,261 - Epoch: [109][ 770/ 1236] Overall Loss 0.245419 Objective Loss 0.245419 LR 0.000500 Time 0.021508 +2023-10-05 21:33:36,465 - Epoch: [109][ 780/ 1236] Overall Loss 0.244955 Objective Loss 0.244955 LR 0.000500 Time 0.021493 +2023-10-05 21:33:36,668 - Epoch: [109][ 790/ 1236] Overall Loss 0.244850 Objective Loss 0.244850 LR 0.000500 Time 0.021478 +2023-10-05 21:33:36,872 - Epoch: [109][ 800/ 1236] Overall Loss 0.244795 Objective Loss 0.244795 LR 0.000500 Time 0.021464 +2023-10-05 21:33:37,075 - Epoch: [109][ 810/ 1236] Overall Loss 0.244627 Objective Loss 0.244627 LR 0.000500 Time 0.021449 +2023-10-05 21:33:37,279 - Epoch: [109][ 820/ 1236] Overall Loss 0.244524 Objective Loss 0.244524 LR 0.000500 Time 0.021436 +2023-10-05 21:33:37,483 - Epoch: [109][ 830/ 1236] Overall Loss 0.244482 Objective Loss 0.244482 LR 0.000500 Time 0.021423 +2023-10-05 21:33:37,687 - Epoch: [109][ 840/ 1236] Overall Loss 0.244806 Objective Loss 0.244806 LR 0.000500 Time 0.021410 +2023-10-05 21:33:37,890 - Epoch: [109][ 850/ 1236] Overall Loss 0.244781 Objective Loss 0.244781 LR 0.000500 Time 0.021397 +2023-10-05 21:33:38,094 - Epoch: [109][ 860/ 1236] Overall Loss 0.244825 Objective Loss 0.244825 LR 0.000500 Time 0.021385 +2023-10-05 21:33:38,297 - Epoch: [109][ 870/ 1236] Overall Loss 0.244762 Objective Loss 0.244762 LR 0.000500 Time 0.021372 +2023-10-05 21:33:38,501 - Epoch: [109][ 880/ 1236] Overall Loss 0.244962 Objective Loss 0.244962 LR 0.000500 Time 0.021361 +2023-10-05 21:33:38,704 - Epoch: [109][ 890/ 1236] Overall Loss 0.245022 Objective Loss 0.245022 LR 0.000500 Time 0.021349 +2023-10-05 21:33:38,908 - Epoch: [109][ 900/ 1236] Overall Loss 0.245219 Objective Loss 0.245219 LR 0.000500 Time 0.021338 +2023-10-05 21:33:39,111 - Epoch: [109][ 910/ 1236] Overall Loss 0.245220 Objective Loss 0.245220 LR 0.000500 Time 0.021326 +2023-10-05 21:33:39,316 - Epoch: [109][ 920/ 1236] Overall Loss 0.245352 Objective Loss 0.245352 LR 0.000500 Time 0.021316 +2023-10-05 21:33:39,520 - Epoch: [109][ 930/ 1236] Overall Loss 0.245388 Objective Loss 0.245388 LR 0.000500 Time 0.021306 +2023-10-05 21:33:39,724 - Epoch: [109][ 940/ 1236] Overall Loss 0.245483 Objective Loss 0.245483 LR 0.000500 Time 0.021296 +2023-10-05 21:33:39,926 - Epoch: [109][ 950/ 1236] Overall Loss 0.245553 Objective Loss 0.245553 LR 0.000500 Time 0.021285 +2023-10-05 21:33:40,131 - Epoch: [109][ 960/ 1236] Overall Loss 0.245504 Objective Loss 0.245504 LR 0.000500 Time 0.021275 +2023-10-05 21:33:40,333 - Epoch: [109][ 970/ 1236] Overall Loss 0.245522 Objective Loss 0.245522 LR 0.000500 Time 0.021265 +2023-10-05 21:33:40,537 - Epoch: [109][ 980/ 1236] Overall Loss 0.245536 Objective Loss 0.245536 LR 0.000500 Time 0.021256 +2023-10-05 21:33:40,740 - Epoch: [109][ 990/ 1236] Overall Loss 0.245287 Objective Loss 0.245287 LR 0.000500 Time 0.021245 +2023-10-05 21:33:40,944 - Epoch: [109][ 1000/ 1236] Overall Loss 0.245209 Objective Loss 0.245209 LR 0.000500 Time 0.021236 +2023-10-05 21:33:41,146 - Epoch: [109][ 1010/ 1236] Overall Loss 0.245494 Objective Loss 0.245494 LR 0.000500 Time 0.021226 +2023-10-05 21:33:41,350 - Epoch: [109][ 1020/ 1236] Overall Loss 0.245747 Objective Loss 0.245747 LR 0.000500 Time 0.021218 +2023-10-05 21:33:41,553 - Epoch: [109][ 1030/ 1236] Overall Loss 0.245911 Objective Loss 0.245911 LR 0.000500 Time 0.021208 +2023-10-05 21:33:41,757 - Epoch: [109][ 1040/ 1236] Overall Loss 0.246278 Objective Loss 0.246278 LR 0.000500 Time 0.021200 +2023-10-05 21:33:41,959 - Epoch: [109][ 1050/ 1236] Overall Loss 0.246357 Objective Loss 0.246357 LR 0.000500 Time 0.021191 +2023-10-05 21:33:42,163 - Epoch: [109][ 1060/ 1236] Overall Loss 0.246265 Objective Loss 0.246265 LR 0.000500 Time 0.021183 +2023-10-05 21:33:42,374 - Epoch: [109][ 1070/ 1236] Overall Loss 0.246409 Objective Loss 0.246409 LR 0.000500 Time 0.021181 +2023-10-05 21:33:42,581 - Epoch: [109][ 1080/ 1236] Overall Loss 0.246313 Objective Loss 0.246313 LR 0.000500 Time 0.021177 +2023-10-05 21:33:42,789 - Epoch: [109][ 1090/ 1236] Overall Loss 0.246817 Objective Loss 0.246817 LR 0.000500 Time 0.021173 +2023-10-05 21:33:42,998 - Epoch: [109][ 1100/ 1236] Overall Loss 0.247171 Objective Loss 0.247171 LR 0.000500 Time 0.021170 +2023-10-05 21:33:43,209 - Epoch: [109][ 1110/ 1236] Overall Loss 0.247447 Objective Loss 0.247447 LR 0.000500 Time 0.021169 +2023-10-05 21:33:43,416 - Epoch: [109][ 1120/ 1236] Overall Loss 0.247562 Objective Loss 0.247562 LR 0.000500 Time 0.021165 +2023-10-05 21:33:43,624 - Epoch: [109][ 1130/ 1236] Overall Loss 0.247749 Objective Loss 0.247749 LR 0.000500 Time 0.021161 +2023-10-05 21:33:43,836 - Epoch: [109][ 1140/ 1236] Overall Loss 0.247879 Objective Loss 0.247879 LR 0.000500 Time 0.021161 +2023-10-05 21:33:44,044 - Epoch: [109][ 1150/ 1236] Overall Loss 0.248126 Objective Loss 0.248126 LR 0.000500 Time 0.021157 +2023-10-05 21:33:44,251 - Epoch: [109][ 1160/ 1236] Overall Loss 0.248078 Objective Loss 0.248078 LR 0.000500 Time 0.021153 +2023-10-05 21:33:44,459 - Epoch: [109][ 1170/ 1236] Overall Loss 0.248024 Objective Loss 0.248024 LR 0.000500 Time 0.021150 +2023-10-05 21:33:44,666 - Epoch: [109][ 1180/ 1236] Overall Loss 0.248105 Objective Loss 0.248105 LR 0.000500 Time 0.021146 +2023-10-05 21:33:44,874 - Epoch: [109][ 1190/ 1236] Overall Loss 0.248286 Objective Loss 0.248286 LR 0.000500 Time 0.021143 +2023-10-05 21:33:45,081 - Epoch: [109][ 1200/ 1236] Overall Loss 0.248208 Objective Loss 0.248208 LR 0.000500 Time 0.021139 +2023-10-05 21:33:45,289 - Epoch: [109][ 1210/ 1236] Overall Loss 0.248134 Objective Loss 0.248134 LR 0.000500 Time 0.021135 +2023-10-05 21:33:45,496 - Epoch: [109][ 1220/ 1236] Overall Loss 0.248407 Objective Loss 0.248407 LR 0.000500 Time 0.021132 +2023-10-05 21:33:45,758 - Epoch: [109][ 1230/ 1236] Overall Loss 0.248497 Objective Loss 0.248497 LR 0.000500 Time 0.021173 +2023-10-05 21:33:45,877 - Epoch: [109][ 1236/ 1236] Overall Loss 0.248892 Objective Loss 0.248892 Top1 86.150713 Top5 98.778004 LR 0.000500 Time 0.021166 +2023-10-05 21:33:46,012 - --- validate (epoch=109)----------- +2023-10-05 21:33:46,012 - 29943 samples (256 per mini-batch) +2023-10-05 21:33:46,479 - Epoch: [109][ 10/ 117] Loss 0.325312 Top1 84.414062 Top5 97.968750 +2023-10-05 21:33:46,640 - Epoch: [109][ 20/ 117] Loss 0.314988 Top1 84.511719 Top5 97.773438 +2023-10-05 21:33:46,801 - Epoch: [109][ 30/ 117] Loss 0.314891 Top1 84.544271 Top5 97.929688 +2023-10-05 21:33:46,961 - Epoch: [109][ 40/ 117] Loss 0.309571 Top1 84.648438 Top5 97.939453 +2023-10-05 21:33:47,122 - Epoch: [109][ 50/ 117] Loss 0.307315 Top1 84.515625 Top5 97.968750 +2023-10-05 21:33:47,288 - Epoch: [109][ 60/ 117] Loss 0.312189 Top1 84.322917 Top5 97.910156 +2023-10-05 21:33:47,445 - Epoch: [109][ 70/ 117] Loss 0.316653 Top1 84.174107 Top5 97.907366 +2023-10-05 21:33:47,603 - Epoch: [109][ 80/ 117] Loss 0.319798 Top1 84.096680 Top5 97.910156 +2023-10-05 21:33:47,761 - Epoch: [109][ 90/ 117] Loss 0.321170 Top1 83.993056 Top5 97.881944 +2023-10-05 21:33:47,916 - Epoch: [109][ 100/ 117] Loss 0.323480 Top1 83.812500 Top5 97.855469 +2023-10-05 21:33:48,081 - Epoch: [109][ 110/ 117] Loss 0.324604 Top1 83.806818 Top5 97.851562 +2023-10-05 21:33:48,167 - Epoch: [109][ 117/ 117] Loss 0.323470 Top1 83.799219 Top5 97.852587 +2023-10-05 21:33:48,312 - ==> Top1: 83.799 Top5: 97.853 Loss: 0.323 + +2023-10-05 21:33:48,313 - ==> Confusion: +[[ 934 1 3 1 7 1 0 2 7 59 3 1 0 1 6 2 4 1 2 0 15] + [ 0 1064 1 0 9 21 1 14 1 0 0 0 0 0 2 5 2 0 7 1 3] + [ 5 3 957 12 1 1 24 9 0 1 6 1 6 2 2 5 5 0 4 3 9] + [ 2 4 17 955 0 1 1 1 4 0 9 0 5 2 29 2 1 5 33 2 16] + [ 30 3 0 0 979 2 2 1 0 6 1 2 1 0 7 5 3 1 0 1 6] + [ 5 51 1 1 3 955 3 21 3 1 3 11 4 16 6 2 5 0 6 3 16] + [ 0 3 22 0 0 1 1132 11 0 0 3 2 2 0 1 4 0 0 0 5 5] + [ 5 24 12 0 3 26 5 1059 1 3 3 8 3 1 2 0 1 0 43 10 9] + [ 17 2 0 2 1 1 1 0 977 42 13 2 3 6 15 2 0 1 2 1 1] + [ 114 0 1 0 7 2 0 0 30 920 0 1 0 24 4 5 0 0 1 2 8] + [ 5 7 12 4 0 0 5 1 9 3 962 3 2 10 4 1 1 1 11 1 11] + [ 1 0 2 0 4 12 0 3 0 1 0 941 25 7 0 2 2 20 0 12 3] + [ 1 2 3 4 1 2 0 1 1 1 2 36 971 5 1 5 2 13 4 4 9] + [ 1 0 0 2 7 3 0 1 9 12 6 3 4 1052 3 3 2 0 0 3 8] + [ 10 2 3 13 8 0 0 0 23 2 2 1 1 2 1009 0 0 1 14 0 10] + [ 0 4 1 0 4 0 0 0 0 0 0 5 6 2 1 1079 9 8 0 9 6] + [ 0 25 2 0 8 1 0 2 0 0 0 3 0 0 4 8 1091 0 0 5 12] + [ 0 0 0 3 0 0 0 0 0 0 0 1 18 0 0 6 0 1002 1 3 4] + [ 2 8 9 14 0 1 0 24 2 0 2 1 2 0 11 0 1 0 982 0 9] + [ 0 5 4 2 2 3 11 12 1 0 2 11 0 2 0 8 9 3 4 1066 7] + [ 134 267 150 75 103 137 49 102 90 68 173 131 328 269 165 55 187 58 182 177 5005]] + +2023-10-05 21:33:48,314 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:33:48,314 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:33:48,320 - + +2023-10-05 21:33:48,320 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:33:49,449 - Epoch: [110][ 10/ 1236] Overall Loss 0.217849 Objective Loss 0.217849 LR 0.000500 Time 0.112828 +2023-10-05 21:33:49,654 - Epoch: [110][ 20/ 1236] Overall Loss 0.238420 Objective Loss 0.238420 LR 0.000500 Time 0.066640 +2023-10-05 21:33:49,857 - Epoch: [110][ 30/ 1236] Overall Loss 0.238459 Objective Loss 0.238459 LR 0.000500 Time 0.051167 +2023-10-05 21:33:50,061 - Epoch: [110][ 40/ 1236] Overall Loss 0.244913 Objective Loss 0.244913 LR 0.000500 Time 0.043471 +2023-10-05 21:33:50,263 - Epoch: [110][ 50/ 1236] Overall Loss 0.244312 Objective Loss 0.244312 LR 0.000500 Time 0.038805 +2023-10-05 21:33:50,466 - Epoch: [110][ 60/ 1236] Overall Loss 0.240146 Objective Loss 0.240146 LR 0.000500 Time 0.035720 +2023-10-05 21:33:50,668 - Epoch: [110][ 70/ 1236] Overall Loss 0.237727 Objective Loss 0.237727 LR 0.000500 Time 0.033494 +2023-10-05 21:33:50,872 - Epoch: [110][ 80/ 1236] Overall Loss 0.239190 Objective Loss 0.239190 LR 0.000500 Time 0.031861 +2023-10-05 21:33:51,076 - Epoch: [110][ 90/ 1236] Overall Loss 0.237442 Objective Loss 0.237442 LR 0.000500 Time 0.030575 +2023-10-05 21:33:51,281 - Epoch: [110][ 100/ 1236] Overall Loss 0.239366 Objective Loss 0.239366 LR 0.000500 Time 0.029570 +2023-10-05 21:33:51,486 - Epoch: [110][ 110/ 1236] Overall Loss 0.242634 Objective Loss 0.242634 LR 0.000500 Time 0.028740 +2023-10-05 21:33:51,691 - Epoch: [110][ 120/ 1236] Overall Loss 0.243130 Objective Loss 0.243130 LR 0.000500 Time 0.028053 +2023-10-05 21:33:51,896 - Epoch: [110][ 130/ 1236] Overall Loss 0.244739 Objective Loss 0.244739 LR 0.000500 Time 0.027467 +2023-10-05 21:33:52,101 - Epoch: [110][ 140/ 1236] Overall Loss 0.245090 Objective Loss 0.245090 LR 0.000500 Time 0.026967 +2023-10-05 21:33:52,307 - Epoch: [110][ 150/ 1236] Overall Loss 0.244312 Objective Loss 0.244312 LR 0.000500 Time 0.026538 +2023-10-05 21:33:52,512 - Epoch: [110][ 160/ 1236] Overall Loss 0.244817 Objective Loss 0.244817 LR 0.000500 Time 0.026161 +2023-10-05 21:33:52,717 - Epoch: [110][ 170/ 1236] Overall Loss 0.245370 Objective Loss 0.245370 LR 0.000500 Time 0.025826 +2023-10-05 21:33:52,919 - Epoch: [110][ 180/ 1236] Overall Loss 0.244543 Objective Loss 0.244543 LR 0.000500 Time 0.025513 +2023-10-05 21:33:53,121 - Epoch: [110][ 190/ 1236] Overall Loss 0.245810 Objective Loss 0.245810 LR 0.000500 Time 0.025229 +2023-10-05 21:33:53,324 - Epoch: [110][ 200/ 1236] Overall Loss 0.246265 Objective Loss 0.246265 LR 0.000500 Time 0.024979 +2023-10-05 21:33:53,526 - Epoch: [110][ 210/ 1236] Overall Loss 0.246744 Objective Loss 0.246744 LR 0.000500 Time 0.024751 +2023-10-05 21:33:53,728 - Epoch: [110][ 220/ 1236] Overall Loss 0.247318 Objective Loss 0.247318 LR 0.000500 Time 0.024545 +2023-10-05 21:33:53,930 - Epoch: [110][ 230/ 1236] Overall Loss 0.247340 Objective Loss 0.247340 LR 0.000500 Time 0.024354 +2023-10-05 21:33:54,133 - Epoch: [110][ 240/ 1236] Overall Loss 0.247846 Objective Loss 0.247846 LR 0.000500 Time 0.024181 +2023-10-05 21:33:54,334 - Epoch: [110][ 250/ 1236] Overall Loss 0.248715 Objective Loss 0.248715 LR 0.000500 Time 0.024020 +2023-10-05 21:33:54,537 - Epoch: [110][ 260/ 1236] Overall Loss 0.247088 Objective Loss 0.247088 LR 0.000500 Time 0.023872 +2023-10-05 21:33:54,739 - Epoch: [110][ 270/ 1236] Overall Loss 0.248103 Objective Loss 0.248103 LR 0.000500 Time 0.023735 +2023-10-05 21:33:54,941 - Epoch: [110][ 280/ 1236] Overall Loss 0.247782 Objective Loss 0.247782 LR 0.000500 Time 0.023608 +2023-10-05 21:33:55,143 - Epoch: [110][ 290/ 1236] Overall Loss 0.247635 Objective Loss 0.247635 LR 0.000500 Time 0.023489 +2023-10-05 21:33:55,344 - Epoch: [110][ 300/ 1236] Overall Loss 0.247365 Objective Loss 0.247365 LR 0.000500 Time 0.023375 +2023-10-05 21:33:55,546 - Epoch: [110][ 310/ 1236] Overall Loss 0.246947 Objective Loss 0.246947 LR 0.000500 Time 0.023272 +2023-10-05 21:33:55,749 - Epoch: [110][ 320/ 1236] Overall Loss 0.246778 Objective Loss 0.246778 LR 0.000500 Time 0.023179 +2023-10-05 21:33:55,951 - Epoch: [110][ 330/ 1236] Overall Loss 0.245802 Objective Loss 0.245802 LR 0.000500 Time 0.023087 +2023-10-05 21:33:56,154 - Epoch: [110][ 340/ 1236] Overall Loss 0.246266 Objective Loss 0.246266 LR 0.000500 Time 0.023004 +2023-10-05 21:33:56,357 - Epoch: [110][ 350/ 1236] Overall Loss 0.246043 Objective Loss 0.246043 LR 0.000500 Time 0.022927 +2023-10-05 21:33:56,566 - Epoch: [110][ 360/ 1236] Overall Loss 0.245677 Objective Loss 0.245677 LR 0.000500 Time 0.022869 +2023-10-05 21:33:56,773 - Epoch: [110][ 370/ 1236] Overall Loss 0.246136 Objective Loss 0.246136 LR 0.000500 Time 0.022810 +2023-10-05 21:33:56,979 - Epoch: [110][ 380/ 1236] Overall Loss 0.245887 Objective Loss 0.245887 LR 0.000500 Time 0.022748 +2023-10-05 21:33:57,183 - Epoch: [110][ 390/ 1236] Overall Loss 0.246128 Objective Loss 0.246128 LR 0.000500 Time 0.022688 +2023-10-05 21:33:57,388 - Epoch: [110][ 400/ 1236] Overall Loss 0.245783 Objective Loss 0.245783 LR 0.000500 Time 0.022633 +2023-10-05 21:33:57,592 - Epoch: [110][ 410/ 1236] Overall Loss 0.245556 Objective Loss 0.245556 LR 0.000500 Time 0.022577 +2023-10-05 21:33:57,797 - Epoch: [110][ 420/ 1236] Overall Loss 0.245158 Objective Loss 0.245158 LR 0.000500 Time 0.022526 +2023-10-05 21:33:58,000 - Epoch: [110][ 430/ 1236] Overall Loss 0.244916 Objective Loss 0.244916 LR 0.000500 Time 0.022475 +2023-10-05 21:33:58,205 - Epoch: [110][ 440/ 1236] Overall Loss 0.245180 Objective Loss 0.245180 LR 0.000500 Time 0.022429 +2023-10-05 21:33:58,409 - Epoch: [110][ 450/ 1236] Overall Loss 0.245568 Objective Loss 0.245568 LR 0.000500 Time 0.022383 +2023-10-05 21:33:58,614 - Epoch: [110][ 460/ 1236] Overall Loss 0.245563 Objective Loss 0.245563 LR 0.000500 Time 0.022341 +2023-10-05 21:33:58,818 - Epoch: [110][ 470/ 1236] Overall Loss 0.245119 Objective Loss 0.245119 LR 0.000500 Time 0.022298 +2023-10-05 21:33:59,023 - Epoch: [110][ 480/ 1236] Overall Loss 0.245463 Objective Loss 0.245463 LR 0.000500 Time 0.022260 +2023-10-05 21:33:59,226 - Epoch: [110][ 490/ 1236] Overall Loss 0.245165 Objective Loss 0.245165 LR 0.000500 Time 0.022220 +2023-10-05 21:33:59,431 - Epoch: [110][ 500/ 1236] Overall Loss 0.245667 Objective Loss 0.245667 LR 0.000500 Time 0.022185 +2023-10-05 21:33:59,635 - Epoch: [110][ 510/ 1236] Overall Loss 0.245485 Objective Loss 0.245485 LR 0.000500 Time 0.022149 +2023-10-05 21:33:59,841 - Epoch: [110][ 520/ 1236] Overall Loss 0.245805 Objective Loss 0.245805 LR 0.000500 Time 0.022117 +2023-10-05 21:34:00,045 - Epoch: [110][ 530/ 1236] Overall Loss 0.246186 Objective Loss 0.246186 LR 0.000500 Time 0.022085 +2023-10-05 21:34:00,265 - Epoch: [110][ 540/ 1236] Overall Loss 0.246155 Objective Loss 0.246155 LR 0.000500 Time 0.022082 +2023-10-05 21:34:00,480 - Epoch: [110][ 550/ 1236] Overall Loss 0.246642 Objective Loss 0.246642 LR 0.000500 Time 0.022072 +2023-10-05 21:34:00,697 - Epoch: [110][ 560/ 1236] Overall Loss 0.246508 Objective Loss 0.246508 LR 0.000500 Time 0.022065 +2023-10-05 21:34:00,905 - Epoch: [110][ 570/ 1236] Overall Loss 0.246823 Objective Loss 0.246823 LR 0.000500 Time 0.022040 +2023-10-05 21:34:01,111 - Epoch: [110][ 580/ 1236] Overall Loss 0.246764 Objective Loss 0.246764 LR 0.000500 Time 0.022015 +2023-10-05 21:34:01,317 - Epoch: [110][ 590/ 1236] Overall Loss 0.246771 Objective Loss 0.246771 LR 0.000500 Time 0.021990 +2023-10-05 21:34:01,523 - Epoch: [110][ 600/ 1236] Overall Loss 0.247052 Objective Loss 0.247052 LR 0.000500 Time 0.021967 +2023-10-05 21:34:01,730 - Epoch: [110][ 610/ 1236] Overall Loss 0.246934 Objective Loss 0.246934 LR 0.000500 Time 0.021945 +2023-10-05 21:34:01,934 - Epoch: [110][ 620/ 1236] Overall Loss 0.246964 Objective Loss 0.246964 LR 0.000500 Time 0.021920 +2023-10-05 21:34:02,148 - Epoch: [110][ 630/ 1236] Overall Loss 0.246974 Objective Loss 0.246974 LR 0.000500 Time 0.021912 +2023-10-05 21:34:02,354 - Epoch: [110][ 640/ 1236] Overall Loss 0.247189 Objective Loss 0.247189 LR 0.000500 Time 0.021890 +2023-10-05 21:34:02,566 - Epoch: [110][ 650/ 1236] Overall Loss 0.247323 Objective Loss 0.247323 LR 0.000500 Time 0.021879 +2023-10-05 21:34:02,771 - Epoch: [110][ 660/ 1236] Overall Loss 0.247028 Objective Loss 0.247028 LR 0.000500 Time 0.021857 +2023-10-05 21:34:02,977 - Epoch: [110][ 670/ 1236] Overall Loss 0.246988 Objective Loss 0.246988 LR 0.000500 Time 0.021837 +2023-10-05 21:34:03,182 - Epoch: [110][ 680/ 1236] Overall Loss 0.247177 Objective Loss 0.247177 LR 0.000500 Time 0.021819 +2023-10-05 21:34:03,388 - Epoch: [110][ 690/ 1236] Overall Loss 0.246711 Objective Loss 0.246711 LR 0.000500 Time 0.021800 +2023-10-05 21:34:03,595 - Epoch: [110][ 700/ 1236] Overall Loss 0.246571 Objective Loss 0.246571 LR 0.000500 Time 0.021783 +2023-10-05 21:34:03,800 - Epoch: [110][ 710/ 1236] Overall Loss 0.246658 Objective Loss 0.246658 LR 0.000500 Time 0.021765 +2023-10-05 21:34:04,004 - Epoch: [110][ 720/ 1236] Overall Loss 0.246638 Objective Loss 0.246638 LR 0.000500 Time 0.021746 +2023-10-05 21:34:04,210 - Epoch: [110][ 730/ 1236] Overall Loss 0.246464 Objective Loss 0.246464 LR 0.000500 Time 0.021729 +2023-10-05 21:34:04,416 - Epoch: [110][ 740/ 1236] Overall Loss 0.246647 Objective Loss 0.246647 LR 0.000500 Time 0.021713 +2023-10-05 21:34:04,620 - Epoch: [110][ 750/ 1236] Overall Loss 0.246670 Objective Loss 0.246670 LR 0.000500 Time 0.021695 +2023-10-05 21:34:04,828 - Epoch: [110][ 760/ 1236] Overall Loss 0.246352 Objective Loss 0.246352 LR 0.000500 Time 0.021683 +2023-10-05 21:34:05,034 - Epoch: [110][ 770/ 1236] Overall Loss 0.246427 Objective Loss 0.246427 LR 0.000500 Time 0.021668 +2023-10-05 21:34:05,242 - Epoch: [110][ 780/ 1236] Overall Loss 0.246490 Objective Loss 0.246490 LR 0.000500 Time 0.021657 +2023-10-05 21:34:05,448 - Epoch: [110][ 790/ 1236] Overall Loss 0.246803 Objective Loss 0.246803 LR 0.000500 Time 0.021643 +2023-10-05 21:34:05,656 - Epoch: [110][ 800/ 1236] Overall Loss 0.246535 Objective Loss 0.246535 LR 0.000500 Time 0.021633 +2023-10-05 21:34:05,863 - Epoch: [110][ 810/ 1236] Overall Loss 0.246305 Objective Loss 0.246305 LR 0.000500 Time 0.021620 +2023-10-05 21:34:06,072 - Epoch: [110][ 820/ 1236] Overall Loss 0.246226 Objective Loss 0.246226 LR 0.000500 Time 0.021611 +2023-10-05 21:34:06,278 - Epoch: [110][ 830/ 1236] Overall Loss 0.246255 Objective Loss 0.246255 LR 0.000500 Time 0.021598 +2023-10-05 21:34:06,487 - Epoch: [110][ 840/ 1236] Overall Loss 0.246032 Objective Loss 0.246032 LR 0.000500 Time 0.021589 +2023-10-05 21:34:06,693 - Epoch: [110][ 850/ 1236] Overall Loss 0.245856 Objective Loss 0.245856 LR 0.000500 Time 0.021577 +2023-10-05 21:34:06,902 - Epoch: [110][ 860/ 1236] Overall Loss 0.246058 Objective Loss 0.246058 LR 0.000500 Time 0.021569 +2023-10-05 21:34:07,108 - Epoch: [110][ 870/ 1236] Overall Loss 0.245867 Objective Loss 0.245867 LR 0.000500 Time 0.021557 +2023-10-05 21:34:07,317 - Epoch: [110][ 880/ 1236] Overall Loss 0.245656 Objective Loss 0.245656 LR 0.000500 Time 0.021549 +2023-10-05 21:34:07,523 - Epoch: [110][ 890/ 1236] Overall Loss 0.245411 Objective Loss 0.245411 LR 0.000500 Time 0.021538 +2023-10-05 21:34:07,732 - Epoch: [110][ 900/ 1236] Overall Loss 0.245174 Objective Loss 0.245174 LR 0.000500 Time 0.021531 +2023-10-05 21:34:07,938 - Epoch: [110][ 910/ 1236] Overall Loss 0.245139 Objective Loss 0.245139 LR 0.000500 Time 0.021521 +2023-10-05 21:34:08,148 - Epoch: [110][ 920/ 1236] Overall Loss 0.245315 Objective Loss 0.245315 LR 0.000500 Time 0.021514 +2023-10-05 21:34:08,354 - Epoch: [110][ 930/ 1236] Overall Loss 0.245416 Objective Loss 0.245416 LR 0.000500 Time 0.021503 +2023-10-05 21:34:08,563 - Epoch: [110][ 940/ 1236] Overall Loss 0.245444 Objective Loss 0.245444 LR 0.000500 Time 0.021497 +2023-10-05 21:34:08,768 - Epoch: [110][ 950/ 1236] Overall Loss 0.245359 Objective Loss 0.245359 LR 0.000500 Time 0.021486 +2023-10-05 21:34:08,977 - Epoch: [110][ 960/ 1236] Overall Loss 0.245380 Objective Loss 0.245380 LR 0.000500 Time 0.021480 +2023-10-05 21:34:09,184 - Epoch: [110][ 970/ 1236] Overall Loss 0.245692 Objective Loss 0.245692 LR 0.000500 Time 0.021471 +2023-10-05 21:34:09,393 - Epoch: [110][ 980/ 1236] Overall Loss 0.245862 Objective Loss 0.245862 LR 0.000500 Time 0.021465 +2023-10-05 21:34:09,599 - Epoch: [110][ 990/ 1236] Overall Loss 0.245854 Objective Loss 0.245854 LR 0.000500 Time 0.021455 +2023-10-05 21:34:09,808 - Epoch: [110][ 1000/ 1236] Overall Loss 0.246158 Objective Loss 0.246158 LR 0.000500 Time 0.021449 +2023-10-05 21:34:10,014 - Epoch: [110][ 1010/ 1236] Overall Loss 0.246218 Objective Loss 0.246218 LR 0.000500 Time 0.021441 +2023-10-05 21:34:10,223 - Epoch: [110][ 1020/ 1236] Overall Loss 0.246239 Objective Loss 0.246239 LR 0.000500 Time 0.021435 +2023-10-05 21:34:10,429 - Epoch: [110][ 1030/ 1236] Overall Loss 0.246340 Objective Loss 0.246340 LR 0.000500 Time 0.021427 +2023-10-05 21:34:10,637 - Epoch: [110][ 1040/ 1236] Overall Loss 0.246530 Objective Loss 0.246530 LR 0.000500 Time 0.021421 +2023-10-05 21:34:10,844 - Epoch: [110][ 1050/ 1236] Overall Loss 0.246765 Objective Loss 0.246765 LR 0.000500 Time 0.021413 +2023-10-05 21:34:11,052 - Epoch: [110][ 1060/ 1236] Overall Loss 0.246837 Objective Loss 0.246837 LR 0.000500 Time 0.021407 +2023-10-05 21:34:11,258 - Epoch: [110][ 1070/ 1236] Overall Loss 0.246852 Objective Loss 0.246852 LR 0.000500 Time 0.021399 +2023-10-05 21:34:11,467 - Epoch: [110][ 1080/ 1236] Overall Loss 0.246834 Objective Loss 0.246834 LR 0.000500 Time 0.021394 +2023-10-05 21:34:11,673 - Epoch: [110][ 1090/ 1236] Overall Loss 0.246644 Objective Loss 0.246644 LR 0.000500 Time 0.021387 +2023-10-05 21:34:11,882 - Epoch: [110][ 1100/ 1236] Overall Loss 0.246745 Objective Loss 0.246745 LR 0.000500 Time 0.021381 +2023-10-05 21:34:12,088 - Epoch: [110][ 1110/ 1236] Overall Loss 0.246973 Objective Loss 0.246973 LR 0.000500 Time 0.021374 +2023-10-05 21:34:12,296 - Epoch: [110][ 1120/ 1236] Overall Loss 0.247101 Objective Loss 0.247101 LR 0.000500 Time 0.021369 +2023-10-05 21:34:12,503 - Epoch: [110][ 1130/ 1236] Overall Loss 0.247391 Objective Loss 0.247391 LR 0.000500 Time 0.021362 +2023-10-05 21:34:12,711 - Epoch: [110][ 1140/ 1236] Overall Loss 0.247413 Objective Loss 0.247413 LR 0.000500 Time 0.021357 +2023-10-05 21:34:12,917 - Epoch: [110][ 1150/ 1236] Overall Loss 0.247376 Objective Loss 0.247376 LR 0.000500 Time 0.021350 +2023-10-05 21:34:13,126 - Epoch: [110][ 1160/ 1236] Overall Loss 0.247271 Objective Loss 0.247271 LR 0.000500 Time 0.021346 +2023-10-05 21:34:13,335 - Epoch: [110][ 1170/ 1236] Overall Loss 0.247323 Objective Loss 0.247323 LR 0.000500 Time 0.021342 +2023-10-05 21:34:13,554 - Epoch: [110][ 1180/ 1236] Overall Loss 0.247366 Objective Loss 0.247366 LR 0.000500 Time 0.021346 +2023-10-05 21:34:13,767 - Epoch: [110][ 1190/ 1236] Overall Loss 0.247250 Objective Loss 0.247250 LR 0.000500 Time 0.021345 +2023-10-05 21:34:13,987 - Epoch: [110][ 1200/ 1236] Overall Loss 0.247343 Objective Loss 0.247343 LR 0.000500 Time 0.021350 +2023-10-05 21:34:14,207 - Epoch: [110][ 1210/ 1236] Overall Loss 0.247464 Objective Loss 0.247464 LR 0.000500 Time 0.021356 +2023-10-05 21:34:14,431 - Epoch: [110][ 1220/ 1236] Overall Loss 0.247560 Objective Loss 0.247560 LR 0.000500 Time 0.021364 +2023-10-05 21:34:14,697 - Epoch: [110][ 1230/ 1236] Overall Loss 0.247615 Objective Loss 0.247615 LR 0.000500 Time 0.021406 +2023-10-05 21:34:14,816 - Epoch: [110][ 1236/ 1236] Overall Loss 0.247752 Objective Loss 0.247752 Top1 85.743381 Top5 97.352342 LR 0.000500 Time 0.021398 +2023-10-05 21:34:14,961 - --- validate (epoch=110)----------- +2023-10-05 21:34:14,961 - 29943 samples (256 per mini-batch) +2023-10-05 21:34:15,433 - Epoch: [110][ 10/ 117] Loss 0.311446 Top1 84.375000 Top5 97.890625 +2023-10-05 21:34:15,597 - Epoch: [110][ 20/ 117] Loss 0.323011 Top1 83.867188 Top5 97.851562 +2023-10-05 21:34:15,760 - Epoch: [110][ 30/ 117] Loss 0.322224 Top1 83.919271 Top5 97.851562 +2023-10-05 21:34:15,923 - Epoch: [110][ 40/ 117] Loss 0.328285 Top1 83.671875 Top5 97.617188 +2023-10-05 21:34:16,084 - Epoch: [110][ 50/ 117] Loss 0.327967 Top1 83.789062 Top5 97.812500 +2023-10-05 21:34:16,248 - Epoch: [110][ 60/ 117] Loss 0.320983 Top1 83.854167 Top5 97.812500 +2023-10-05 21:34:16,404 - Epoch: [110][ 70/ 117] Loss 0.320589 Top1 83.772321 Top5 97.879464 +2023-10-05 21:34:16,568 - Epoch: [110][ 80/ 117] Loss 0.319259 Top1 83.862305 Top5 97.875977 +2023-10-05 21:34:16,723 - Epoch: [110][ 90/ 117] Loss 0.320097 Top1 83.802083 Top5 97.855903 +2023-10-05 21:34:16,887 - Epoch: [110][ 100/ 117] Loss 0.321777 Top1 83.687500 Top5 97.832031 +2023-10-05 21:34:17,051 - Epoch: [110][ 110/ 117] Loss 0.320584 Top1 83.721591 Top5 97.858665 +2023-10-05 21:34:17,137 - Epoch: [110][ 117/ 117] Loss 0.321984 Top1 83.749123 Top5 97.875964 +2023-10-05 21:34:17,274 - ==> Top1: 83.749 Top5: 97.876 Loss: 0.322 + +2023-10-05 21:34:17,274 - ==> Confusion: +[[ 935 1 3 1 9 1 0 1 6 58 1 2 1 4 4 2 4 1 1 1 14] + [ 0 1045 3 0 9 21 1 21 2 0 4 3 0 0 1 3 1 0 11 1 5] + [ 2 1 950 21 1 0 33 10 0 2 4 2 8 2 0 3 3 1 3 1 9] + [ 3 1 12 984 0 6 1 0 4 1 8 0 5 1 14 4 1 6 24 1 13] + [ 27 3 3 0 976 4 1 2 0 9 1 1 0 0 8 2 7 2 0 0 4] + [ 3 32 1 1 3 997 1 20 0 2 5 10 4 8 3 1 5 0 4 3 13] + [ 0 3 20 0 0 1 1132 7 0 0 5 2 1 0 1 8 0 0 2 4 5] + [ 3 16 23 1 1 31 5 1063 0 4 4 7 2 1 0 3 0 0 42 5 7] + [ 17 2 0 0 1 4 2 0 962 48 14 3 4 9 14 2 1 0 5 0 1] + [ 108 0 1 1 5 8 1 0 21 941 0 0 1 19 2 2 2 1 0 3 3] + [ 1 4 11 7 2 3 5 2 8 1 973 5 0 8 5 2 0 0 4 0 12] + [ 1 1 2 0 0 18 0 2 0 0 0 943 28 4 1 2 0 18 1 9 5] + [ 0 1 3 7 0 3 1 2 1 1 2 27 973 2 1 5 2 20 2 4 11] + [ 1 0 2 2 5 13 1 0 13 15 7 4 2 1031 3 1 0 0 0 1 18] + [ 14 2 3 20 8 1 0 0 27 3 3 1 1 2 990 0 0 2 11 0 13] + [ 0 1 1 0 5 1 2 0 0 0 0 13 3 1 0 1067 12 12 0 9 7] + [ 1 19 0 0 6 6 0 2 1 0 0 6 1 0 2 7 1095 0 2 5 8] + [ 0 0 0 1 0 0 1 0 1 1 0 0 17 0 0 7 0 1005 1 2 2] + [ 2 6 11 21 1 1 0 24 3 0 2 1 1 1 6 0 1 0 976 0 11] + [ 1 4 3 2 2 6 14 11 1 1 2 14 3 1 0 4 9 1 4 1060 9] + [ 137 188 145 83 97 182 73 128 79 86 182 143 315 276 122 70 166 68 186 200 4979]] + +2023-10-05 21:34:17,276 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:34:17,276 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:34:17,282 - + +2023-10-05 21:34:17,282 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:34:18,293 - Epoch: [111][ 10/ 1236] Overall Loss 0.253960 Objective Loss 0.253960 LR 0.000500 Time 0.101005 +2023-10-05 21:34:18,496 - Epoch: [111][ 20/ 1236] Overall Loss 0.251436 Objective Loss 0.251436 LR 0.000500 Time 0.060663 +2023-10-05 21:34:18,700 - Epoch: [111][ 30/ 1236] Overall Loss 0.248429 Objective Loss 0.248429 LR 0.000500 Time 0.047227 +2023-10-05 21:34:18,905 - Epoch: [111][ 40/ 1236] Overall Loss 0.249926 Objective Loss 0.249926 LR 0.000500 Time 0.040532 +2023-10-05 21:34:19,108 - Epoch: [111][ 50/ 1236] Overall Loss 0.243354 Objective Loss 0.243354 LR 0.000500 Time 0.036478 +2023-10-05 21:34:19,312 - Epoch: [111][ 60/ 1236] Overall Loss 0.242294 Objective Loss 0.242294 LR 0.000500 Time 0.033792 +2023-10-05 21:34:19,514 - Epoch: [111][ 70/ 1236] Overall Loss 0.240979 Objective Loss 0.240979 LR 0.000500 Time 0.031849 +2023-10-05 21:34:19,718 - Epoch: [111][ 80/ 1236] Overall Loss 0.246805 Objective Loss 0.246805 LR 0.000500 Time 0.030412 +2023-10-05 21:34:19,922 - Epoch: [111][ 90/ 1236] Overall Loss 0.245690 Objective Loss 0.245690 LR 0.000500 Time 0.029295 +2023-10-05 21:34:20,127 - Epoch: [111][ 100/ 1236] Overall Loss 0.245543 Objective Loss 0.245543 LR 0.000500 Time 0.028414 +2023-10-05 21:34:20,332 - Epoch: [111][ 110/ 1236] Overall Loss 0.246462 Objective Loss 0.246462 LR 0.000500 Time 0.027688 +2023-10-05 21:34:20,537 - Epoch: [111][ 120/ 1236] Overall Loss 0.245847 Objective Loss 0.245847 LR 0.000500 Time 0.027085 +2023-10-05 21:34:20,741 - Epoch: [111][ 130/ 1236] Overall Loss 0.244415 Objective Loss 0.244415 LR 0.000500 Time 0.026570 +2023-10-05 21:34:20,946 - Epoch: [111][ 140/ 1236] Overall Loss 0.242534 Objective Loss 0.242534 LR 0.000500 Time 0.026133 +2023-10-05 21:34:21,150 - Epoch: [111][ 150/ 1236] Overall Loss 0.241296 Objective Loss 0.241296 LR 0.000500 Time 0.025749 +2023-10-05 21:34:21,355 - Epoch: [111][ 160/ 1236] Overall Loss 0.240308 Objective Loss 0.240308 LR 0.000500 Time 0.025420 +2023-10-05 21:34:21,559 - Epoch: [111][ 170/ 1236] Overall Loss 0.240565 Objective Loss 0.240565 LR 0.000500 Time 0.025123 +2023-10-05 21:34:21,764 - Epoch: [111][ 180/ 1236] Overall Loss 0.240147 Objective Loss 0.240147 LR 0.000500 Time 0.024865 +2023-10-05 21:34:21,969 - Epoch: [111][ 190/ 1236] Overall Loss 0.240240 Objective Loss 0.240240 LR 0.000500 Time 0.024629 +2023-10-05 21:34:22,174 - Epoch: [111][ 200/ 1236] Overall Loss 0.240634 Objective Loss 0.240634 LR 0.000500 Time 0.024421 +2023-10-05 21:34:22,376 - Epoch: [111][ 210/ 1236] Overall Loss 0.240385 Objective Loss 0.240385 LR 0.000500 Time 0.024222 +2023-10-05 21:34:22,582 - Epoch: [111][ 220/ 1236] Overall Loss 0.239330 Objective Loss 0.239330 LR 0.000500 Time 0.024053 +2023-10-05 21:34:22,786 - Epoch: [111][ 230/ 1236] Overall Loss 0.239386 Objective Loss 0.239386 LR 0.000500 Time 0.023894 +2023-10-05 21:34:22,991 - Epoch: [111][ 240/ 1236] Overall Loss 0.239029 Objective Loss 0.239029 LR 0.000500 Time 0.023752 +2023-10-05 21:34:23,196 - Epoch: [111][ 250/ 1236] Overall Loss 0.239431 Objective Loss 0.239431 LR 0.000500 Time 0.023620 +2023-10-05 21:34:23,402 - Epoch: [111][ 260/ 1236] Overall Loss 0.240026 Objective Loss 0.240026 LR 0.000500 Time 0.023501 +2023-10-05 21:34:23,606 - Epoch: [111][ 270/ 1236] Overall Loss 0.240028 Objective Loss 0.240028 LR 0.000500 Time 0.023386 +2023-10-05 21:34:23,811 - Epoch: [111][ 280/ 1236] Overall Loss 0.240893 Objective Loss 0.240893 LR 0.000500 Time 0.023283 +2023-10-05 21:34:24,016 - Epoch: [111][ 290/ 1236] Overall Loss 0.241318 Objective Loss 0.241318 LR 0.000500 Time 0.023185 +2023-10-05 21:34:24,222 - Epoch: [111][ 300/ 1236] Overall Loss 0.242606 Objective Loss 0.242606 LR 0.000500 Time 0.023096 +2023-10-05 21:34:24,428 - Epoch: [111][ 310/ 1236] Overall Loss 0.242225 Objective Loss 0.242225 LR 0.000500 Time 0.023014 +2023-10-05 21:34:24,634 - Epoch: [111][ 320/ 1236] Overall Loss 0.242887 Objective Loss 0.242887 LR 0.000500 Time 0.022938 +2023-10-05 21:34:24,840 - Epoch: [111][ 330/ 1236] Overall Loss 0.241692 Objective Loss 0.241692 LR 0.000500 Time 0.022868 +2023-10-05 21:34:25,046 - Epoch: [111][ 340/ 1236] Overall Loss 0.242048 Objective Loss 0.242048 LR 0.000500 Time 0.022799 +2023-10-05 21:34:25,251 - Epoch: [111][ 350/ 1236] Overall Loss 0.242479 Objective Loss 0.242479 LR 0.000500 Time 0.022732 +2023-10-05 21:34:25,457 - Epoch: [111][ 360/ 1236] Overall Loss 0.243060 Objective Loss 0.243060 LR 0.000500 Time 0.022671 +2023-10-05 21:34:25,668 - Epoch: [111][ 370/ 1236] Overall Loss 0.243850 Objective Loss 0.243850 LR 0.000500 Time 0.022628 +2023-10-05 21:34:25,884 - Epoch: [111][ 380/ 1236] Overall Loss 0.244044 Objective Loss 0.244044 LR 0.000500 Time 0.022600 +2023-10-05 21:34:26,094 - Epoch: [111][ 390/ 1236] Overall Loss 0.243826 Objective Loss 0.243826 LR 0.000500 Time 0.022559 +2023-10-05 21:34:26,308 - Epoch: [111][ 400/ 1236] Overall Loss 0.244315 Objective Loss 0.244315 LR 0.000500 Time 0.022529 +2023-10-05 21:34:26,520 - Epoch: [111][ 410/ 1236] Overall Loss 0.244215 Objective Loss 0.244215 LR 0.000500 Time 0.022496 +2023-10-05 21:34:26,735 - Epoch: [111][ 420/ 1236] Overall Loss 0.243679 Objective Loss 0.243679 LR 0.000500 Time 0.022471 +2023-10-05 21:34:26,945 - Epoch: [111][ 430/ 1236] Overall Loss 0.243518 Objective Loss 0.243518 LR 0.000500 Time 0.022436 +2023-10-05 21:34:27,149 - Epoch: [111][ 440/ 1236] Overall Loss 0.244004 Objective Loss 0.244004 LR 0.000500 Time 0.022390 +2023-10-05 21:34:27,354 - Epoch: [111][ 450/ 1236] Overall Loss 0.244094 Objective Loss 0.244094 LR 0.000500 Time 0.022347 +2023-10-05 21:34:27,559 - Epoch: [111][ 460/ 1236] Overall Loss 0.244445 Objective Loss 0.244445 LR 0.000500 Time 0.022305 +2023-10-05 21:34:27,761 - Epoch: [111][ 470/ 1236] Overall Loss 0.244551 Objective Loss 0.244551 LR 0.000500 Time 0.022260 +2023-10-05 21:34:27,965 - Epoch: [111][ 480/ 1236] Overall Loss 0.244567 Objective Loss 0.244567 LR 0.000500 Time 0.022222 +2023-10-05 21:34:28,169 - Epoch: [111][ 490/ 1236] Overall Loss 0.244567 Objective Loss 0.244567 LR 0.000500 Time 0.022183 +2023-10-05 21:34:28,373 - Epoch: [111][ 500/ 1236] Overall Loss 0.244465 Objective Loss 0.244465 LR 0.000500 Time 0.022147 +2023-10-05 21:34:28,576 - Epoch: [111][ 510/ 1236] Overall Loss 0.245228 Objective Loss 0.245228 LR 0.000500 Time 0.022109 +2023-10-05 21:34:28,781 - Epoch: [111][ 520/ 1236] Overall Loss 0.245238 Objective Loss 0.245238 LR 0.000500 Time 0.022078 +2023-10-05 21:34:28,984 - Epoch: [111][ 530/ 1236] Overall Loss 0.244748 Objective Loss 0.244748 LR 0.000500 Time 0.022044 +2023-10-05 21:34:29,199 - Epoch: [111][ 540/ 1236] Overall Loss 0.244481 Objective Loss 0.244481 LR 0.000500 Time 0.022034 +2023-10-05 21:34:29,409 - Epoch: [111][ 550/ 1236] Overall Loss 0.244671 Objective Loss 0.244671 LR 0.000500 Time 0.022015 +2023-10-05 21:34:29,624 - Epoch: [111][ 560/ 1236] Overall Loss 0.244740 Objective Loss 0.244740 LR 0.000500 Time 0.022004 +2023-10-05 21:34:29,833 - Epoch: [111][ 570/ 1236] Overall Loss 0.244634 Objective Loss 0.244634 LR 0.000500 Time 0.021985 +2023-10-05 21:34:30,047 - Epoch: [111][ 580/ 1236] Overall Loss 0.244533 Objective Loss 0.244533 LR 0.000500 Time 0.021975 +2023-10-05 21:34:30,257 - Epoch: [111][ 590/ 1236] Overall Loss 0.244491 Objective Loss 0.244491 LR 0.000500 Time 0.021958 +2023-10-05 21:34:30,472 - Epoch: [111][ 600/ 1236] Overall Loss 0.244783 Objective Loss 0.244783 LR 0.000500 Time 0.021950 +2023-10-05 21:34:30,682 - Epoch: [111][ 610/ 1236] Overall Loss 0.245139 Objective Loss 0.245139 LR 0.000500 Time 0.021933 +2023-10-05 21:34:30,890 - Epoch: [111][ 620/ 1236] Overall Loss 0.245121 Objective Loss 0.245121 LR 0.000500 Time 0.021914 +2023-10-05 21:34:31,092 - Epoch: [111][ 630/ 1236] Overall Loss 0.245268 Objective Loss 0.245268 LR 0.000500 Time 0.021887 +2023-10-05 21:34:31,297 - Epoch: [111][ 640/ 1236] Overall Loss 0.245132 Objective Loss 0.245132 LR 0.000500 Time 0.021864 +2023-10-05 21:34:31,506 - Epoch: [111][ 650/ 1236] Overall Loss 0.244985 Objective Loss 0.244985 LR 0.000500 Time 0.021848 +2023-10-05 21:34:31,720 - Epoch: [111][ 660/ 1236] Overall Loss 0.244846 Objective Loss 0.244846 LR 0.000500 Time 0.021842 +2023-10-05 21:34:31,930 - Epoch: [111][ 670/ 1236] Overall Loss 0.244962 Objective Loss 0.244962 LR 0.000500 Time 0.021829 +2023-10-05 21:34:32,145 - Epoch: [111][ 680/ 1236] Overall Loss 0.244634 Objective Loss 0.244634 LR 0.000500 Time 0.021823 +2023-10-05 21:34:32,355 - Epoch: [111][ 690/ 1236] Overall Loss 0.244566 Objective Loss 0.244566 LR 0.000500 Time 0.021812 +2023-10-05 21:34:32,570 - Epoch: [111][ 700/ 1236] Overall Loss 0.244543 Objective Loss 0.244543 LR 0.000500 Time 0.021805 +2023-10-05 21:34:32,779 - Epoch: [111][ 710/ 1236] Overall Loss 0.244355 Objective Loss 0.244355 LR 0.000500 Time 0.021793 +2023-10-05 21:34:32,994 - Epoch: [111][ 720/ 1236] Overall Loss 0.244779 Objective Loss 0.244779 LR 0.000500 Time 0.021789 +2023-10-05 21:34:33,204 - Epoch: [111][ 730/ 1236] Overall Loss 0.244375 Objective Loss 0.244375 LR 0.000500 Time 0.021777 +2023-10-05 21:34:33,418 - Epoch: [111][ 740/ 1236] Overall Loss 0.244766 Objective Loss 0.244766 LR 0.000500 Time 0.021772 +2023-10-05 21:34:33,628 - Epoch: [111][ 750/ 1236] Overall Loss 0.244382 Objective Loss 0.244382 LR 0.000500 Time 0.021761 +2023-10-05 21:34:33,843 - Epoch: [111][ 760/ 1236] Overall Loss 0.244536 Objective Loss 0.244536 LR 0.000500 Time 0.021757 +2023-10-05 21:34:34,054 - Epoch: [111][ 770/ 1236] Overall Loss 0.244612 Objective Loss 0.244612 LR 0.000500 Time 0.021748 +2023-10-05 21:34:34,268 - Epoch: [111][ 780/ 1236] Overall Loss 0.244512 Objective Loss 0.244512 LR 0.000500 Time 0.021743 +2023-10-05 21:34:34,478 - Epoch: [111][ 790/ 1236] Overall Loss 0.244434 Objective Loss 0.244434 LR 0.000500 Time 0.021734 +2023-10-05 21:34:34,693 - Epoch: [111][ 800/ 1236] Overall Loss 0.244642 Objective Loss 0.244642 LR 0.000500 Time 0.021730 +2023-10-05 21:34:34,903 - Epoch: [111][ 810/ 1236] Overall Loss 0.244880 Objective Loss 0.244880 LR 0.000500 Time 0.021721 +2023-10-05 21:34:35,118 - Epoch: [111][ 820/ 1236] Overall Loss 0.244971 Objective Loss 0.244971 LR 0.000500 Time 0.021718 +2023-10-05 21:34:35,327 - Epoch: [111][ 830/ 1236] Overall Loss 0.244839 Objective Loss 0.244839 LR 0.000500 Time 0.021708 +2023-10-05 21:34:35,542 - Epoch: [111][ 840/ 1236] Overall Loss 0.244723 Objective Loss 0.244723 LR 0.000500 Time 0.021705 +2023-10-05 21:34:35,748 - Epoch: [111][ 850/ 1236] Overall Loss 0.244721 Objective Loss 0.244721 LR 0.000500 Time 0.021692 +2023-10-05 21:34:35,953 - Epoch: [111][ 860/ 1236] Overall Loss 0.244703 Objective Loss 0.244703 LR 0.000500 Time 0.021677 +2023-10-05 21:34:36,156 - Epoch: [111][ 870/ 1236] Overall Loss 0.244551 Objective Loss 0.244551 LR 0.000500 Time 0.021660 +2023-10-05 21:34:36,361 - Epoch: [111][ 880/ 1236] Overall Loss 0.244412 Objective Loss 0.244412 LR 0.000500 Time 0.021647 +2023-10-05 21:34:36,563 - Epoch: [111][ 890/ 1236] Overall Loss 0.244496 Objective Loss 0.244496 LR 0.000500 Time 0.021631 +2023-10-05 21:34:36,768 - Epoch: [111][ 900/ 1236] Overall Loss 0.244583 Objective Loss 0.244583 LR 0.000500 Time 0.021617 +2023-10-05 21:34:36,971 - Epoch: [111][ 910/ 1236] Overall Loss 0.244515 Objective Loss 0.244515 LR 0.000500 Time 0.021603 +2023-10-05 21:34:37,175 - Epoch: [111][ 920/ 1236] Overall Loss 0.244683 Objective Loss 0.244683 LR 0.000500 Time 0.021589 +2023-10-05 21:34:37,378 - Epoch: [111][ 930/ 1236] Overall Loss 0.244644 Objective Loss 0.244644 LR 0.000500 Time 0.021575 +2023-10-05 21:34:37,582 - Epoch: [111][ 940/ 1236] Overall Loss 0.244921 Objective Loss 0.244921 LR 0.000500 Time 0.021563 +2023-10-05 21:34:37,785 - Epoch: [111][ 950/ 1236] Overall Loss 0.245126 Objective Loss 0.245126 LR 0.000500 Time 0.021549 +2023-10-05 21:34:37,989 - Epoch: [111][ 960/ 1236] Overall Loss 0.244932 Objective Loss 0.244932 LR 0.000500 Time 0.021537 +2023-10-05 21:34:38,193 - Epoch: [111][ 970/ 1236] Overall Loss 0.245034 Objective Loss 0.245034 LR 0.000500 Time 0.021524 +2023-10-05 21:34:38,397 - Epoch: [111][ 980/ 1236] Overall Loss 0.244866 Objective Loss 0.244866 LR 0.000500 Time 0.021513 +2023-10-05 21:34:38,599 - Epoch: [111][ 990/ 1236] Overall Loss 0.244866 Objective Loss 0.244866 LR 0.000500 Time 0.021500 +2023-10-05 21:34:38,804 - Epoch: [111][ 1000/ 1236] Overall Loss 0.244613 Objective Loss 0.244613 LR 0.000500 Time 0.021489 +2023-10-05 21:34:39,006 - Epoch: [111][ 1010/ 1236] Overall Loss 0.244743 Objective Loss 0.244743 LR 0.000500 Time 0.021476 +2023-10-05 21:34:39,211 - Epoch: [111][ 1020/ 1236] Overall Loss 0.244489 Objective Loss 0.244489 LR 0.000500 Time 0.021466 +2023-10-05 21:34:39,413 - Epoch: [111][ 1030/ 1236] Overall Loss 0.244660 Objective Loss 0.244660 LR 0.000500 Time 0.021454 +2023-10-05 21:34:39,618 - Epoch: [111][ 1040/ 1236] Overall Loss 0.244878 Objective Loss 0.244878 LR 0.000500 Time 0.021444 +2023-10-05 21:34:39,823 - Epoch: [111][ 1050/ 1236] Overall Loss 0.244810 Objective Loss 0.244810 LR 0.000500 Time 0.021435 +2023-10-05 21:34:40,027 - Epoch: [111][ 1060/ 1236] Overall Loss 0.244754 Objective Loss 0.244754 LR 0.000500 Time 0.021425 +2023-10-05 21:34:40,231 - Epoch: [111][ 1070/ 1236] Overall Loss 0.244746 Objective Loss 0.244746 LR 0.000500 Time 0.021414 +2023-10-05 21:34:40,437 - Epoch: [111][ 1080/ 1236] Overall Loss 0.245133 Objective Loss 0.245133 LR 0.000500 Time 0.021406 +2023-10-05 21:34:40,640 - Epoch: [111][ 1090/ 1236] Overall Loss 0.245204 Objective Loss 0.245204 LR 0.000500 Time 0.021396 +2023-10-05 21:34:40,844 - Epoch: [111][ 1100/ 1236] Overall Loss 0.245185 Objective Loss 0.245185 LR 0.000500 Time 0.021387 +2023-10-05 21:34:41,047 - Epoch: [111][ 1110/ 1236] Overall Loss 0.245052 Objective Loss 0.245052 LR 0.000500 Time 0.021377 +2023-10-05 21:34:41,251 - Epoch: [111][ 1120/ 1236] Overall Loss 0.245301 Objective Loss 0.245301 LR 0.000500 Time 0.021368 +2023-10-05 21:34:41,454 - Epoch: [111][ 1130/ 1236] Overall Loss 0.245156 Objective Loss 0.245156 LR 0.000500 Time 0.021358 +2023-10-05 21:34:41,658 - Epoch: [111][ 1140/ 1236] Overall Loss 0.245110 Objective Loss 0.245110 LR 0.000500 Time 0.021349 +2023-10-05 21:34:41,864 - Epoch: [111][ 1150/ 1236] Overall Loss 0.245321 Objective Loss 0.245321 LR 0.000500 Time 0.021343 +2023-10-05 21:34:42,079 - Epoch: [111][ 1160/ 1236] Overall Loss 0.245499 Objective Loss 0.245499 LR 0.000500 Time 0.021344 +2023-10-05 21:34:42,289 - Epoch: [111][ 1170/ 1236] Overall Loss 0.245351 Objective Loss 0.245351 LR 0.000500 Time 0.021341 +2023-10-05 21:34:42,504 - Epoch: [111][ 1180/ 1236] Overall Loss 0.245278 Objective Loss 0.245278 LR 0.000500 Time 0.021342 +2023-10-05 21:34:42,714 - Epoch: [111][ 1190/ 1236] Overall Loss 0.245337 Objective Loss 0.245337 LR 0.000500 Time 0.021338 +2023-10-05 21:34:42,929 - Epoch: [111][ 1200/ 1236] Overall Loss 0.245252 Objective Loss 0.245252 LR 0.000500 Time 0.021339 +2023-10-05 21:34:43,132 - Epoch: [111][ 1210/ 1236] Overall Loss 0.245307 Objective Loss 0.245307 LR 0.000500 Time 0.021330 +2023-10-05 21:34:43,336 - Epoch: [111][ 1220/ 1236] Overall Loss 0.245436 Objective Loss 0.245436 LR 0.000500 Time 0.021323 +2023-10-05 21:34:43,593 - Epoch: [111][ 1230/ 1236] Overall Loss 0.245435 Objective Loss 0.245435 LR 0.000500 Time 0.021358 +2023-10-05 21:34:43,711 - Epoch: [111][ 1236/ 1236] Overall Loss 0.245550 Objective Loss 0.245550 Top1 85.539715 Top5 97.759674 LR 0.000500 Time 0.021350 +2023-10-05 21:34:43,830 - --- validate (epoch=111)----------- +2023-10-05 21:34:43,830 - 29943 samples (256 per mini-batch) +2023-10-05 21:34:44,288 - Epoch: [111][ 10/ 117] Loss 0.325269 Top1 82.968750 Top5 98.164062 +2023-10-05 21:34:44,437 - Epoch: [111][ 20/ 117] Loss 0.322113 Top1 83.476562 Top5 97.988281 +2023-10-05 21:34:44,584 - Epoch: [111][ 30/ 117] Loss 0.315099 Top1 83.359375 Top5 97.955729 +2023-10-05 21:34:44,732 - Epoch: [111][ 40/ 117] Loss 0.333828 Top1 83.105469 Top5 97.880859 +2023-10-05 21:34:44,879 - Epoch: [111][ 50/ 117] Loss 0.340172 Top1 82.750000 Top5 97.789062 +2023-10-05 21:34:45,026 - Epoch: [111][ 60/ 117] Loss 0.340417 Top1 82.799479 Top5 97.825521 +2023-10-05 21:34:45,177 - Epoch: [111][ 70/ 117] Loss 0.342560 Top1 82.773438 Top5 97.862723 +2023-10-05 21:34:45,328 - Epoch: [111][ 80/ 117] Loss 0.337065 Top1 82.929688 Top5 97.866211 +2023-10-05 21:34:45,480 - Epoch: [111][ 90/ 117] Loss 0.338311 Top1 82.834201 Top5 97.829861 +2023-10-05 21:34:45,631 - Epoch: [111][ 100/ 117] Loss 0.336663 Top1 82.921875 Top5 97.816406 +2023-10-05 21:34:45,787 - Epoch: [111][ 110/ 117] Loss 0.333771 Top1 83.082386 Top5 97.848011 +2023-10-05 21:34:45,873 - Epoch: [111][ 117/ 117] Loss 0.333407 Top1 83.067829 Top5 97.862606 +2023-10-05 21:34:46,009 - ==> Top1: 83.068 Top5: 97.863 Loss: 0.333 + +2023-10-05 21:34:46,009 - ==> Confusion: +[[ 930 2 2 0 11 3 0 2 4 63 0 1 1 4 8 3 6 1 0 0 9] + [ 2 1056 0 1 5 22 2 19 1 0 1 0 0 0 0 4 5 0 5 1 7] + [ 4 2 953 4 1 0 29 11 0 1 3 3 8 2 0 6 3 2 4 6 14] + [ 3 0 24 934 2 8 1 0 2 0 7 0 4 7 32 4 3 9 29 4 16] + [ 23 5 1 0 974 4 0 1 0 7 1 2 0 3 8 6 5 2 3 1 4] + [ 3 45 1 1 5 970 1 29 0 1 2 12 2 16 5 1 4 1 0 6 11] + [ 1 7 25 0 0 0 1115 11 0 0 0 2 1 0 1 10 0 2 2 9 5] + [ 4 25 21 0 5 39 3 1041 0 4 3 7 2 1 0 2 0 1 38 12 10] + [ 19 2 0 2 2 6 0 0 954 50 10 3 0 7 23 3 0 1 1 4 2] + [ 116 0 3 0 9 5 0 0 29 908 0 0 0 25 6 6 1 0 1 3 7] + [ 4 4 11 4 0 0 6 4 6 3 967 4 0 14 4 1 2 1 3 3 12] + [ 1 0 1 0 2 9 0 1 0 2 0 976 13 6 0 1 1 13 0 6 3] + [ 0 1 2 2 0 3 1 2 0 0 1 46 973 2 1 5 3 9 2 5 10] + [ 1 0 1 1 4 6 2 0 7 9 5 4 3 1058 4 2 1 0 0 2 9] + [ 13 3 3 11 8 0 0 0 20 6 2 0 3 1 1007 0 1 3 8 0 12] + [ 0 1 1 0 5 0 0 1 0 0 0 14 9 1 1 1061 16 11 1 8 4] + [ 0 10 0 0 11 5 0 2 1 0 0 5 2 0 2 7 1100 1 0 6 9] + [ 1 0 0 1 0 0 0 0 1 0 0 13 14 0 1 5 2 994 1 1 4] + [ 1 10 13 13 2 0 0 30 1 0 4 0 7 0 15 0 1 0 960 1 10] + [ 0 3 4 2 2 7 9 8 0 0 1 19 4 1 0 8 7 1 2 1063 11] + [ 134 242 138 53 119 200 41 104 88 75 171 148 343 317 152 73 249 74 123 182 4879]] + +2023-10-05 21:34:46,011 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:34:46,011 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:34:46,017 - + +2023-10-05 21:34:46,017 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:34:47,024 - Epoch: [112][ 10/ 1236] Overall Loss 0.252062 Objective Loss 0.252062 LR 0.000500 Time 0.100642 +2023-10-05 21:34:47,231 - Epoch: [112][ 20/ 1236] Overall Loss 0.236471 Objective Loss 0.236471 LR 0.000500 Time 0.060657 +2023-10-05 21:34:47,432 - Epoch: [112][ 30/ 1236] Overall Loss 0.232363 Objective Loss 0.232363 LR 0.000500 Time 0.047126 +2023-10-05 21:34:47,641 - Epoch: [112][ 40/ 1236] Overall Loss 0.245904 Objective Loss 0.245904 LR 0.000500 Time 0.040573 +2023-10-05 21:34:47,854 - Epoch: [112][ 50/ 1236] Overall Loss 0.248401 Objective Loss 0.248401 LR 0.000500 Time 0.036707 +2023-10-05 21:34:48,067 - Epoch: [112][ 60/ 1236] Overall Loss 0.247405 Objective Loss 0.247405 LR 0.000500 Time 0.034124 +2023-10-05 21:34:48,275 - Epoch: [112][ 70/ 1236] Overall Loss 0.247798 Objective Loss 0.247798 LR 0.000500 Time 0.032218 +2023-10-05 21:34:48,487 - Epoch: [112][ 80/ 1236] Overall Loss 0.245376 Objective Loss 0.245376 LR 0.000500 Time 0.030843 +2023-10-05 21:34:48,695 - Epoch: [112][ 90/ 1236] Overall Loss 0.250810 Objective Loss 0.250810 LR 0.000500 Time 0.029726 +2023-10-05 21:34:48,903 - Epoch: [112][ 100/ 1236] Overall Loss 0.254851 Objective Loss 0.254851 LR 0.000500 Time 0.028823 +2023-10-05 21:34:49,105 - Epoch: [112][ 110/ 1236] Overall Loss 0.252629 Objective Loss 0.252629 LR 0.000500 Time 0.028036 +2023-10-05 21:34:49,307 - Epoch: [112][ 120/ 1236] Overall Loss 0.252572 Objective Loss 0.252572 LR 0.000500 Time 0.027381 +2023-10-05 21:34:49,507 - Epoch: [112][ 130/ 1236] Overall Loss 0.252046 Objective Loss 0.252046 LR 0.000500 Time 0.026814 +2023-10-05 21:34:49,710 - Epoch: [112][ 140/ 1236] Overall Loss 0.252793 Objective Loss 0.252793 LR 0.000500 Time 0.026341 +2023-10-05 21:34:49,910 - Epoch: [112][ 150/ 1236] Overall Loss 0.253318 Objective Loss 0.253318 LR 0.000500 Time 0.025922 +2023-10-05 21:34:50,113 - Epoch: [112][ 160/ 1236] Overall Loss 0.252890 Objective Loss 0.252890 LR 0.000500 Time 0.025566 +2023-10-05 21:34:50,314 - Epoch: [112][ 170/ 1236] Overall Loss 0.249949 Objective Loss 0.249949 LR 0.000500 Time 0.025241 +2023-10-05 21:34:50,516 - Epoch: [112][ 180/ 1236] Overall Loss 0.250272 Objective Loss 0.250272 LR 0.000500 Time 0.024961 +2023-10-05 21:34:50,717 - Epoch: [112][ 190/ 1236] Overall Loss 0.250805 Objective Loss 0.250805 LR 0.000500 Time 0.024702 +2023-10-05 21:34:50,919 - Epoch: [112][ 200/ 1236] Overall Loss 0.251583 Objective Loss 0.251583 LR 0.000500 Time 0.024477 +2023-10-05 21:34:51,120 - Epoch: [112][ 210/ 1236] Overall Loss 0.252097 Objective Loss 0.252097 LR 0.000500 Time 0.024265 +2023-10-05 21:34:51,323 - Epoch: [112][ 220/ 1236] Overall Loss 0.252125 Objective Loss 0.252125 LR 0.000500 Time 0.024081 +2023-10-05 21:34:51,523 - Epoch: [112][ 230/ 1236] Overall Loss 0.252839 Objective Loss 0.252839 LR 0.000500 Time 0.023906 +2023-10-05 21:34:51,726 - Epoch: [112][ 240/ 1236] Overall Loss 0.252401 Objective Loss 0.252401 LR 0.000500 Time 0.023751 +2023-10-05 21:34:51,926 - Epoch: [112][ 250/ 1236] Overall Loss 0.252490 Objective Loss 0.252490 LR 0.000500 Time 0.023602 +2023-10-05 21:34:52,129 - Epoch: [112][ 260/ 1236] Overall Loss 0.251988 Objective Loss 0.251988 LR 0.000500 Time 0.023471 +2023-10-05 21:34:52,330 - Epoch: [112][ 270/ 1236] Overall Loss 0.251915 Objective Loss 0.251915 LR 0.000500 Time 0.023345 +2023-10-05 21:34:52,530 - Epoch: [112][ 280/ 1236] Overall Loss 0.250964 Objective Loss 0.250964 LR 0.000500 Time 0.023227 +2023-10-05 21:34:52,729 - Epoch: [112][ 290/ 1236] Overall Loss 0.250457 Objective Loss 0.250457 LR 0.000500 Time 0.023110 +2023-10-05 21:34:52,932 - Epoch: [112][ 300/ 1236] Overall Loss 0.250288 Objective Loss 0.250288 LR 0.000500 Time 0.023014 +2023-10-05 21:34:53,133 - Epoch: [112][ 310/ 1236] Overall Loss 0.250271 Objective Loss 0.250271 LR 0.000500 Time 0.022918 +2023-10-05 21:34:53,333 - Epoch: [112][ 320/ 1236] Overall Loss 0.249420 Objective Loss 0.249420 LR 0.000500 Time 0.022828 +2023-10-05 21:34:53,535 - Epoch: [112][ 330/ 1236] Overall Loss 0.248638 Objective Loss 0.248638 LR 0.000500 Time 0.022746 +2023-10-05 21:34:53,737 - Epoch: [112][ 340/ 1236] Overall Loss 0.247735 Objective Loss 0.247735 LR 0.000500 Time 0.022672 +2023-10-05 21:34:53,939 - Epoch: [112][ 350/ 1236] Overall Loss 0.247982 Objective Loss 0.247982 LR 0.000500 Time 0.022599 +2023-10-05 21:34:54,141 - Epoch: [112][ 360/ 1236] Overall Loss 0.248023 Objective Loss 0.248023 LR 0.000500 Time 0.022532 +2023-10-05 21:34:54,344 - Epoch: [112][ 370/ 1236] Overall Loss 0.248320 Objective Loss 0.248320 LR 0.000500 Time 0.022469 +2023-10-05 21:34:54,556 - Epoch: [112][ 380/ 1236] Overall Loss 0.247971 Objective Loss 0.247971 LR 0.000500 Time 0.022435 +2023-10-05 21:34:54,764 - Epoch: [112][ 390/ 1236] Overall Loss 0.247365 Objective Loss 0.247365 LR 0.000500 Time 0.022394 +2023-10-05 21:34:54,977 - Epoch: [112][ 400/ 1236] Overall Loss 0.247250 Objective Loss 0.247250 LR 0.000500 Time 0.022364 +2023-10-05 21:34:55,186 - Epoch: [112][ 410/ 1236] Overall Loss 0.246444 Objective Loss 0.246444 LR 0.000500 Time 0.022327 +2023-10-05 21:34:55,399 - Epoch: [112][ 420/ 1236] Overall Loss 0.245545 Objective Loss 0.245545 LR 0.000500 Time 0.022302 +2023-10-05 21:34:55,608 - Epoch: [112][ 430/ 1236] Overall Loss 0.245209 Objective Loss 0.245209 LR 0.000500 Time 0.022269 +2023-10-05 21:34:55,821 - Epoch: [112][ 440/ 1236] Overall Loss 0.245004 Objective Loss 0.245004 LR 0.000500 Time 0.022247 +2023-10-05 21:34:56,030 - Epoch: [112][ 450/ 1236] Overall Loss 0.244464 Objective Loss 0.244464 LR 0.000500 Time 0.022216 +2023-10-05 21:34:56,243 - Epoch: [112][ 460/ 1236] Overall Loss 0.244128 Objective Loss 0.244128 LR 0.000500 Time 0.022195 +2023-10-05 21:34:56,452 - Epoch: [112][ 470/ 1236] Overall Loss 0.244181 Objective Loss 0.244181 LR 0.000500 Time 0.022168 +2023-10-05 21:34:56,666 - Epoch: [112][ 480/ 1236] Overall Loss 0.244413 Objective Loss 0.244413 LR 0.000500 Time 0.022150 +2023-10-05 21:34:56,874 - Epoch: [112][ 490/ 1236] Overall Loss 0.244155 Objective Loss 0.244155 LR 0.000500 Time 0.022123 +2023-10-05 21:34:57,088 - Epoch: [112][ 500/ 1236] Overall Loss 0.244208 Objective Loss 0.244208 LR 0.000500 Time 0.022107 +2023-10-05 21:34:57,297 - Epoch: [112][ 510/ 1236] Overall Loss 0.243730 Objective Loss 0.243730 LR 0.000500 Time 0.022082 +2023-10-05 21:34:57,510 - Epoch: [112][ 520/ 1236] Overall Loss 0.244268 Objective Loss 0.244268 LR 0.000500 Time 0.022068 +2023-10-05 21:34:57,720 - Epoch: [112][ 530/ 1236] Overall Loss 0.244499 Objective Loss 0.244499 LR 0.000500 Time 0.022046 +2023-10-05 21:34:57,933 - Epoch: [112][ 540/ 1236] Overall Loss 0.244689 Objective Loss 0.244689 LR 0.000500 Time 0.022032 +2023-10-05 21:34:58,142 - Epoch: [112][ 550/ 1236] Overall Loss 0.244787 Objective Loss 0.244787 LR 0.000500 Time 0.022011 +2023-10-05 21:34:58,356 - Epoch: [112][ 560/ 1236] Overall Loss 0.244923 Objective Loss 0.244923 LR 0.000500 Time 0.021999 +2023-10-05 21:34:58,565 - Epoch: [112][ 570/ 1236] Overall Loss 0.245258 Objective Loss 0.245258 LR 0.000500 Time 0.021979 +2023-10-05 21:34:58,778 - Epoch: [112][ 580/ 1236] Overall Loss 0.245320 Objective Loss 0.245320 LR 0.000500 Time 0.021967 +2023-10-05 21:34:58,987 - Epoch: [112][ 590/ 1236] Overall Loss 0.245943 Objective Loss 0.245943 LR 0.000500 Time 0.021948 +2023-10-05 21:34:59,200 - Epoch: [112][ 600/ 1236] Overall Loss 0.246020 Objective Loss 0.246020 LR 0.000500 Time 0.021937 +2023-10-05 21:34:59,409 - Epoch: [112][ 610/ 1236] Overall Loss 0.246472 Objective Loss 0.246472 LR 0.000500 Time 0.021921 +2023-10-05 21:34:59,623 - Epoch: [112][ 620/ 1236] Overall Loss 0.246491 Objective Loss 0.246491 LR 0.000500 Time 0.021910 +2023-10-05 21:34:59,832 - Epoch: [112][ 630/ 1236] Overall Loss 0.246413 Objective Loss 0.246413 LR 0.000500 Time 0.021894 +2023-10-05 21:35:00,046 - Epoch: [112][ 640/ 1236] Overall Loss 0.246089 Objective Loss 0.246089 LR 0.000500 Time 0.021885 +2023-10-05 21:35:00,255 - Epoch: [112][ 650/ 1236] Overall Loss 0.245942 Objective Loss 0.245942 LR 0.000500 Time 0.021871 +2023-10-05 21:35:00,469 - Epoch: [112][ 660/ 1236] Overall Loss 0.246140 Objective Loss 0.246140 LR 0.000500 Time 0.021862 +2023-10-05 21:35:00,678 - Epoch: [112][ 670/ 1236] Overall Loss 0.246036 Objective Loss 0.246036 LR 0.000500 Time 0.021848 +2023-10-05 21:35:00,891 - Epoch: [112][ 680/ 1236] Overall Loss 0.246376 Objective Loss 0.246376 LR 0.000500 Time 0.021840 +2023-10-05 21:35:01,100 - Epoch: [112][ 690/ 1236] Overall Loss 0.246155 Objective Loss 0.246155 LR 0.000500 Time 0.021826 +2023-10-05 21:35:01,314 - Epoch: [112][ 700/ 1236] Overall Loss 0.246135 Objective Loss 0.246135 LR 0.000500 Time 0.021818 +2023-10-05 21:35:01,523 - Epoch: [112][ 710/ 1236] Overall Loss 0.246073 Objective Loss 0.246073 LR 0.000500 Time 0.021805 +2023-10-05 21:35:01,736 - Epoch: [112][ 720/ 1236] Overall Loss 0.245915 Objective Loss 0.245915 LR 0.000500 Time 0.021798 +2023-10-05 21:35:01,945 - Epoch: [112][ 730/ 1236] Overall Loss 0.245729 Objective Loss 0.245729 LR 0.000500 Time 0.021785 +2023-10-05 21:35:02,159 - Epoch: [112][ 740/ 1236] Overall Loss 0.245631 Objective Loss 0.245631 LR 0.000500 Time 0.021779 +2023-10-05 21:35:02,368 - Epoch: [112][ 750/ 1236] Overall Loss 0.245766 Objective Loss 0.245766 LR 0.000500 Time 0.021767 +2023-10-05 21:35:02,581 - Epoch: [112][ 760/ 1236] Overall Loss 0.245881 Objective Loss 0.245881 LR 0.000500 Time 0.021761 +2023-10-05 21:35:02,783 - Epoch: [112][ 770/ 1236] Overall Loss 0.246201 Objective Loss 0.246201 LR 0.000500 Time 0.021741 +2023-10-05 21:35:02,997 - Epoch: [112][ 780/ 1236] Overall Loss 0.246296 Objective Loss 0.246296 LR 0.000500 Time 0.021735 +2023-10-05 21:35:03,206 - Epoch: [112][ 790/ 1236] Overall Loss 0.246068 Objective Loss 0.246068 LR 0.000500 Time 0.021724 +2023-10-05 21:35:03,420 - Epoch: [112][ 800/ 1236] Overall Loss 0.246060 Objective Loss 0.246060 LR 0.000500 Time 0.021719 +2023-10-05 21:35:03,629 - Epoch: [112][ 810/ 1236] Overall Loss 0.246049 Objective Loss 0.246049 LR 0.000500 Time 0.021709 +2023-10-05 21:35:03,842 - Epoch: [112][ 820/ 1236] Overall Loss 0.246076 Objective Loss 0.246076 LR 0.000500 Time 0.021704 +2023-10-05 21:35:04,051 - Epoch: [112][ 830/ 1236] Overall Loss 0.245712 Objective Loss 0.245712 LR 0.000500 Time 0.021694 +2023-10-05 21:35:04,265 - Epoch: [112][ 840/ 1236] Overall Loss 0.245560 Objective Loss 0.245560 LR 0.000500 Time 0.021689 +2023-10-05 21:35:04,474 - Epoch: [112][ 850/ 1236] Overall Loss 0.245348 Objective Loss 0.245348 LR 0.000500 Time 0.021680 +2023-10-05 21:35:04,684 - Epoch: [112][ 860/ 1236] Overall Loss 0.245343 Objective Loss 0.245343 LR 0.000500 Time 0.021672 +2023-10-05 21:35:04,889 - Epoch: [112][ 870/ 1236] Overall Loss 0.245289 Objective Loss 0.245289 LR 0.000500 Time 0.021658 +2023-10-05 21:35:05,102 - Epoch: [112][ 880/ 1236] Overall Loss 0.245457 Objective Loss 0.245457 LR 0.000500 Time 0.021653 +2023-10-05 21:35:05,311 - Epoch: [112][ 890/ 1236] Overall Loss 0.245285 Objective Loss 0.245285 LR 0.000500 Time 0.021644 +2023-10-05 21:35:05,520 - Epoch: [112][ 900/ 1236] Overall Loss 0.245022 Objective Loss 0.245022 LR 0.000500 Time 0.021636 +2023-10-05 21:35:05,728 - Epoch: [112][ 910/ 1236] Overall Loss 0.245014 Objective Loss 0.245014 LR 0.000500 Time 0.021627 +2023-10-05 21:35:05,941 - Epoch: [112][ 920/ 1236] Overall Loss 0.244908 Objective Loss 0.244908 LR 0.000500 Time 0.021623 +2023-10-05 21:35:06,144 - Epoch: [112][ 930/ 1236] Overall Loss 0.244604 Objective Loss 0.244604 LR 0.000500 Time 0.021608 +2023-10-05 21:35:06,347 - Epoch: [112][ 940/ 1236] Overall Loss 0.244687 Objective Loss 0.244687 LR 0.000500 Time 0.021594 +2023-10-05 21:35:06,550 - Epoch: [112][ 950/ 1236] Overall Loss 0.244790 Objective Loss 0.244790 LR 0.000500 Time 0.021579 +2023-10-05 21:35:06,753 - Epoch: [112][ 960/ 1236] Overall Loss 0.244819 Objective Loss 0.244819 LR 0.000500 Time 0.021566 +2023-10-05 21:35:06,956 - Epoch: [112][ 970/ 1236] Overall Loss 0.244510 Objective Loss 0.244510 LR 0.000500 Time 0.021552 +2023-10-05 21:35:07,159 - Epoch: [112][ 980/ 1236] Overall Loss 0.244720 Objective Loss 0.244720 LR 0.000500 Time 0.021539 +2023-10-05 21:35:07,362 - Epoch: [112][ 990/ 1236] Overall Loss 0.245074 Objective Loss 0.245074 LR 0.000500 Time 0.021526 +2023-10-05 21:35:07,565 - Epoch: [112][ 1000/ 1236] Overall Loss 0.244860 Objective Loss 0.244860 LR 0.000500 Time 0.021514 +2023-10-05 21:35:07,767 - Epoch: [112][ 1010/ 1236] Overall Loss 0.244803 Objective Loss 0.244803 LR 0.000500 Time 0.021501 +2023-10-05 21:35:07,971 - Epoch: [112][ 1020/ 1236] Overall Loss 0.244990 Objective Loss 0.244990 LR 0.000500 Time 0.021489 +2023-10-05 21:35:08,173 - Epoch: [112][ 1030/ 1236] Overall Loss 0.245146 Objective Loss 0.245146 LR 0.000500 Time 0.021476 +2023-10-05 21:35:08,376 - Epoch: [112][ 1040/ 1236] Overall Loss 0.245401 Objective Loss 0.245401 LR 0.000500 Time 0.021465 +2023-10-05 21:35:08,579 - Epoch: [112][ 1050/ 1236] Overall Loss 0.245244 Objective Loss 0.245244 LR 0.000500 Time 0.021453 +2023-10-05 21:35:08,782 - Epoch: [112][ 1060/ 1236] Overall Loss 0.245533 Objective Loss 0.245533 LR 0.000500 Time 0.021442 +2023-10-05 21:35:08,984 - Epoch: [112][ 1070/ 1236] Overall Loss 0.245621 Objective Loss 0.245621 LR 0.000500 Time 0.021431 +2023-10-05 21:35:09,188 - Epoch: [112][ 1080/ 1236] Overall Loss 0.245536 Objective Loss 0.245536 LR 0.000500 Time 0.021420 +2023-10-05 21:35:09,391 - Epoch: [112][ 1090/ 1236] Overall Loss 0.245478 Objective Loss 0.245478 LR 0.000500 Time 0.021409 +2023-10-05 21:35:09,594 - Epoch: [112][ 1100/ 1236] Overall Loss 0.245387 Objective Loss 0.245387 LR 0.000500 Time 0.021399 +2023-10-05 21:35:09,796 - Epoch: [112][ 1110/ 1236] Overall Loss 0.245279 Objective Loss 0.245279 LR 0.000500 Time 0.021388 +2023-10-05 21:35:10,003 - Epoch: [112][ 1120/ 1236] Overall Loss 0.245278 Objective Loss 0.245278 LR 0.000500 Time 0.021382 +2023-10-05 21:35:10,219 - Epoch: [112][ 1130/ 1236] Overall Loss 0.245294 Objective Loss 0.245294 LR 0.000500 Time 0.021383 +2023-10-05 21:35:10,430 - Epoch: [112][ 1140/ 1236] Overall Loss 0.245207 Objective Loss 0.245207 LR 0.000500 Time 0.021380 +2023-10-05 21:35:10,636 - Epoch: [112][ 1150/ 1236] Overall Loss 0.245126 Objective Loss 0.245126 LR 0.000500 Time 0.021373 +2023-10-05 21:35:10,840 - Epoch: [112][ 1160/ 1236] Overall Loss 0.244897 Objective Loss 0.244897 LR 0.000500 Time 0.021365 +2023-10-05 21:35:11,046 - Epoch: [112][ 1170/ 1236] Overall Loss 0.244748 Objective Loss 0.244748 LR 0.000500 Time 0.021358 +2023-10-05 21:35:11,250 - Epoch: [112][ 1180/ 1236] Overall Loss 0.244459 Objective Loss 0.244459 LR 0.000500 Time 0.021349 +2023-10-05 21:35:11,456 - Epoch: [112][ 1190/ 1236] Overall Loss 0.244535 Objective Loss 0.244535 LR 0.000500 Time 0.021343 +2023-10-05 21:35:11,661 - Epoch: [112][ 1200/ 1236] Overall Loss 0.244431 Objective Loss 0.244431 LR 0.000500 Time 0.021335 +2023-10-05 21:35:11,867 - Epoch: [112][ 1210/ 1236] Overall Loss 0.244514 Objective Loss 0.244514 LR 0.000500 Time 0.021329 +2023-10-05 21:35:12,072 - Epoch: [112][ 1220/ 1236] Overall Loss 0.244357 Objective Loss 0.244357 LR 0.000500 Time 0.021322 +2023-10-05 21:35:12,332 - Epoch: [112][ 1230/ 1236] Overall Loss 0.244320 Objective Loss 0.244320 LR 0.000500 Time 0.021359 +2023-10-05 21:35:12,451 - Epoch: [112][ 1236/ 1236] Overall Loss 0.244512 Objective Loss 0.244512 Top1 85.336049 Top5 98.167006 LR 0.000500 Time 0.021352 +2023-10-05 21:35:12,583 - --- validate (epoch=112)----------- +2023-10-05 21:35:12,583 - 29943 samples (256 per mini-batch) +2023-10-05 21:35:13,053 - Epoch: [112][ 10/ 117] Loss 0.324621 Top1 83.710938 Top5 97.539062 +2023-10-05 21:35:13,215 - Epoch: [112][ 20/ 117] Loss 0.340501 Top1 83.496094 Top5 97.480469 +2023-10-05 21:35:13,374 - Epoch: [112][ 30/ 117] Loss 0.332574 Top1 83.919271 Top5 97.630208 +2023-10-05 21:35:13,535 - Epoch: [112][ 40/ 117] Loss 0.339102 Top1 83.613281 Top5 97.646484 +2023-10-05 21:35:13,693 - Epoch: [112][ 50/ 117] Loss 0.336325 Top1 83.515625 Top5 97.656250 +2023-10-05 21:35:13,840 - Epoch: [112][ 60/ 117] Loss 0.339683 Top1 83.411458 Top5 97.636719 +2023-10-05 21:35:13,983 - Epoch: [112][ 70/ 117] Loss 0.334204 Top1 83.582589 Top5 97.706473 +2023-10-05 21:35:14,134 - Epoch: [112][ 80/ 117] Loss 0.337730 Top1 83.398438 Top5 97.734375 +2023-10-05 21:35:14,287 - Epoch: [112][ 90/ 117] Loss 0.337526 Top1 83.454861 Top5 97.712674 +2023-10-05 21:35:14,440 - Epoch: [112][ 100/ 117] Loss 0.339396 Top1 83.363281 Top5 97.750000 +2023-10-05 21:35:14,601 - Epoch: [112][ 110/ 117] Loss 0.339337 Top1 83.423295 Top5 97.805398 +2023-10-05 21:35:14,687 - Epoch: [112][ 117/ 117] Loss 0.337624 Top1 83.475270 Top5 97.822529 +2023-10-05 21:35:14,833 - ==> Top1: 83.475 Top5: 97.823 Loss: 0.338 + +2023-10-05 21:35:14,833 - ==> Confusion: +[[ 922 4 4 2 7 4 0 1 5 70 0 2 0 3 6 3 5 0 0 0 12] + [ 1 1059 1 1 5 25 1 16 1 0 1 0 0 1 0 4 2 0 6 0 7] + [ 5 2 935 13 3 0 42 12 0 0 5 1 7 0 1 2 3 1 6 10 8] + [ 1 1 19 956 1 8 3 2 4 1 10 0 5 4 21 7 4 7 19 1 15] + [ 27 7 1 0 954 6 0 0 0 14 1 2 2 1 9 5 9 1 0 3 8] + [ 3 29 2 2 4 991 4 21 5 3 2 1 4 15 8 2 2 0 1 4 13] + [ 0 5 12 0 0 2 1143 6 0 0 3 2 2 0 1 6 0 1 2 4 2] + [ 1 23 14 0 2 41 10 1056 2 2 3 8 2 2 0 4 1 0 28 9 10] + [ 15 1 0 0 4 1 1 0 979 40 12 2 5 6 12 1 2 0 3 0 5] + [ 120 1 1 0 7 2 0 0 31 906 0 0 0 27 3 7 1 1 0 3 9] + [ 1 8 13 4 0 0 7 3 10 1 965 3 0 13 3 2 2 0 4 5 9] + [ 1 1 2 0 3 15 1 3 0 1 0 936 26 6 0 3 1 18 0 14 4] + [ 0 2 6 6 0 1 3 1 2 0 2 39 954 3 0 9 4 19 4 5 8] + [ 0 0 2 1 3 8 1 0 10 12 8 1 1 1058 2 0 0 1 0 1 10] + [ 11 6 3 14 8 1 1 0 28 4 3 0 1 1 991 0 1 2 7 1 18] + [ 0 3 1 0 5 1 1 0 0 0 0 6 4 2 1 1072 11 10 0 13 4] + [ 1 17 2 0 5 5 0 2 0 0 0 4 2 1 3 10 1096 0 0 7 6] + [ 0 1 0 1 0 0 2 0 1 1 0 4 16 1 0 5 1 999 0 2 4] + [ 1 10 8 18 1 0 1 40 3 0 4 0 4 0 11 0 1 0 956 1 9] + [ 0 5 3 3 2 7 17 13 1 0 2 18 1 2 0 5 9 1 2 1050 11] + [ 110 221 149 76 78 200 77 106 98 80 163 108 340 320 103 66 201 71 142 179 5017]] + +2023-10-05 21:35:14,835 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:35:14,835 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:35:14,841 - + +2023-10-05 21:35:14,841 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:35:15,848 - Epoch: [113][ 10/ 1236] Overall Loss 0.207270 Objective Loss 0.207270 LR 0.000500 Time 0.100639 +2023-10-05 21:35:16,053 - Epoch: [113][ 20/ 1236] Overall Loss 0.216928 Objective Loss 0.216928 LR 0.000500 Time 0.060564 +2023-10-05 21:35:16,258 - Epoch: [113][ 30/ 1236] Overall Loss 0.228134 Objective Loss 0.228134 LR 0.000500 Time 0.047189 +2023-10-05 21:35:16,463 - Epoch: [113][ 40/ 1236] Overall Loss 0.233092 Objective Loss 0.233092 LR 0.000500 Time 0.040517 +2023-10-05 21:35:16,668 - Epoch: [113][ 50/ 1236] Overall Loss 0.239338 Objective Loss 0.239338 LR 0.000500 Time 0.036502 +2023-10-05 21:35:16,873 - Epoch: [113][ 60/ 1236] Overall Loss 0.239967 Objective Loss 0.239967 LR 0.000500 Time 0.033831 +2023-10-05 21:35:17,078 - Epoch: [113][ 70/ 1236] Overall Loss 0.244892 Objective Loss 0.244892 LR 0.000500 Time 0.031920 +2023-10-05 21:35:17,283 - Epoch: [113][ 80/ 1236] Overall Loss 0.246003 Objective Loss 0.246003 LR 0.000500 Time 0.030493 +2023-10-05 21:35:17,488 - Epoch: [113][ 90/ 1236] Overall Loss 0.245615 Objective Loss 0.245615 LR 0.000500 Time 0.029376 +2023-10-05 21:35:17,693 - Epoch: [113][ 100/ 1236] Overall Loss 0.246806 Objective Loss 0.246806 LR 0.000500 Time 0.028487 +2023-10-05 21:35:17,897 - Epoch: [113][ 110/ 1236] Overall Loss 0.246910 Objective Loss 0.246910 LR 0.000500 Time 0.027747 +2023-10-05 21:35:18,102 - Epoch: [113][ 120/ 1236] Overall Loss 0.245896 Objective Loss 0.245896 LR 0.000500 Time 0.027139 +2023-10-05 21:35:18,306 - Epoch: [113][ 130/ 1236] Overall Loss 0.245940 Objective Loss 0.245940 LR 0.000500 Time 0.026623 +2023-10-05 21:35:18,511 - Epoch: [113][ 140/ 1236] Overall Loss 0.245420 Objective Loss 0.245420 LR 0.000500 Time 0.026181 +2023-10-05 21:35:18,715 - Epoch: [113][ 150/ 1236] Overall Loss 0.245089 Objective Loss 0.245089 LR 0.000500 Time 0.025793 +2023-10-05 21:35:18,920 - Epoch: [113][ 160/ 1236] Overall Loss 0.244653 Objective Loss 0.244653 LR 0.000500 Time 0.025458 +2023-10-05 21:35:19,124 - Epoch: [113][ 170/ 1236] Overall Loss 0.244812 Objective Loss 0.244812 LR 0.000500 Time 0.025160 +2023-10-05 21:35:19,329 - Epoch: [113][ 180/ 1236] Overall Loss 0.244569 Objective Loss 0.244569 LR 0.000500 Time 0.024897 +2023-10-05 21:35:19,532 - Epoch: [113][ 190/ 1236] Overall Loss 0.244852 Objective Loss 0.244852 LR 0.000500 Time 0.024658 +2023-10-05 21:35:19,737 - Epoch: [113][ 200/ 1236] Overall Loss 0.243792 Objective Loss 0.243792 LR 0.000500 Time 0.024446 +2023-10-05 21:35:19,941 - Epoch: [113][ 210/ 1236] Overall Loss 0.244097 Objective Loss 0.244097 LR 0.000500 Time 0.024253 +2023-10-05 21:35:20,146 - Epoch: [113][ 220/ 1236] Overall Loss 0.243325 Objective Loss 0.243325 LR 0.000500 Time 0.024079 +2023-10-05 21:35:20,350 - Epoch: [113][ 230/ 1236] Overall Loss 0.243665 Objective Loss 0.243665 LR 0.000500 Time 0.023918 +2023-10-05 21:35:20,555 - Epoch: [113][ 240/ 1236] Overall Loss 0.243343 Objective Loss 0.243343 LR 0.000500 Time 0.023773 +2023-10-05 21:35:20,759 - Epoch: [113][ 250/ 1236] Overall Loss 0.243508 Objective Loss 0.243508 LR 0.000500 Time 0.023637 +2023-10-05 21:35:20,963 - Epoch: [113][ 260/ 1236] Overall Loss 0.244572 Objective Loss 0.244572 LR 0.000500 Time 0.023514 +2023-10-05 21:35:21,165 - Epoch: [113][ 270/ 1236] Overall Loss 0.245915 Objective Loss 0.245915 LR 0.000500 Time 0.023390 +2023-10-05 21:35:21,369 - Epoch: [113][ 280/ 1236] Overall Loss 0.245320 Objective Loss 0.245320 LR 0.000500 Time 0.023280 +2023-10-05 21:35:21,570 - Epoch: [113][ 290/ 1236] Overall Loss 0.244701 Objective Loss 0.244701 LR 0.000500 Time 0.023170 +2023-10-05 21:35:21,773 - Epoch: [113][ 300/ 1236] Overall Loss 0.243976 Objective Loss 0.243976 LR 0.000500 Time 0.023074 +2023-10-05 21:35:21,976 - Epoch: [113][ 310/ 1236] Overall Loss 0.244145 Objective Loss 0.244145 LR 0.000500 Time 0.022981 +2023-10-05 21:35:22,177 - Epoch: [113][ 320/ 1236] Overall Loss 0.243908 Objective Loss 0.243908 LR 0.000500 Time 0.022891 +2023-10-05 21:35:22,378 - Epoch: [113][ 330/ 1236] Overall Loss 0.243816 Objective Loss 0.243816 LR 0.000500 Time 0.022805 +2023-10-05 21:35:22,581 - Epoch: [113][ 340/ 1236] Overall Loss 0.242730 Objective Loss 0.242730 LR 0.000500 Time 0.022730 +2023-10-05 21:35:22,782 - Epoch: [113][ 350/ 1236] Overall Loss 0.241975 Objective Loss 0.241975 LR 0.000500 Time 0.022655 +2023-10-05 21:35:22,984 - Epoch: [113][ 360/ 1236] Overall Loss 0.241733 Objective Loss 0.241733 LR 0.000500 Time 0.022585 +2023-10-05 21:35:23,186 - Epoch: [113][ 370/ 1236] Overall Loss 0.241787 Objective Loss 0.241787 LR 0.000500 Time 0.022520 +2023-10-05 21:35:23,390 - Epoch: [113][ 380/ 1236] Overall Loss 0.241769 Objective Loss 0.241769 LR 0.000500 Time 0.022463 +2023-10-05 21:35:23,593 - Epoch: [113][ 390/ 1236] Overall Loss 0.241629 Objective Loss 0.241629 LR 0.000500 Time 0.022407 +2023-10-05 21:35:23,797 - Epoch: [113][ 400/ 1236] Overall Loss 0.241616 Objective Loss 0.241616 LR 0.000500 Time 0.022356 +2023-10-05 21:35:24,000 - Epoch: [113][ 410/ 1236] Overall Loss 0.241737 Objective Loss 0.241737 LR 0.000500 Time 0.022304 +2023-10-05 21:35:24,204 - Epoch: [113][ 420/ 1236] Overall Loss 0.241248 Objective Loss 0.241248 LR 0.000500 Time 0.022257 +2023-10-05 21:35:24,406 - Epoch: [113][ 430/ 1236] Overall Loss 0.241570 Objective Loss 0.241570 LR 0.000500 Time 0.022210 +2023-10-05 21:35:24,610 - Epoch: [113][ 440/ 1236] Overall Loss 0.240962 Objective Loss 0.240962 LR 0.000500 Time 0.022168 +2023-10-05 21:35:24,813 - Epoch: [113][ 450/ 1236] Overall Loss 0.241898 Objective Loss 0.241898 LR 0.000500 Time 0.022125 +2023-10-05 21:35:25,017 - Epoch: [113][ 460/ 1236] Overall Loss 0.242115 Objective Loss 0.242115 LR 0.000500 Time 0.022086 +2023-10-05 21:35:25,219 - Epoch: [113][ 470/ 1236] Overall Loss 0.242317 Objective Loss 0.242317 LR 0.000500 Time 0.022046 +2023-10-05 21:35:25,423 - Epoch: [113][ 480/ 1236] Overall Loss 0.241920 Objective Loss 0.241920 LR 0.000500 Time 0.022011 +2023-10-05 21:35:25,626 - Epoch: [113][ 490/ 1236] Overall Loss 0.242140 Objective Loss 0.242140 LR 0.000500 Time 0.021975 +2023-10-05 21:35:25,830 - Epoch: [113][ 500/ 1236] Overall Loss 0.241927 Objective Loss 0.241927 LR 0.000500 Time 0.021942 +2023-10-05 21:35:26,033 - Epoch: [113][ 510/ 1236] Overall Loss 0.242755 Objective Loss 0.242755 LR 0.000500 Time 0.021910 +2023-10-05 21:35:26,237 - Epoch: [113][ 520/ 1236] Overall Loss 0.243037 Objective Loss 0.243037 LR 0.000500 Time 0.021880 +2023-10-05 21:35:26,439 - Epoch: [113][ 530/ 1236] Overall Loss 0.242840 Objective Loss 0.242840 LR 0.000500 Time 0.021848 +2023-10-05 21:35:26,643 - Epoch: [113][ 540/ 1236] Overall Loss 0.242667 Objective Loss 0.242667 LR 0.000500 Time 0.021821 +2023-10-05 21:35:26,846 - Epoch: [113][ 550/ 1236] Overall Loss 0.243464 Objective Loss 0.243464 LR 0.000500 Time 0.021792 +2023-10-05 21:35:27,050 - Epoch: [113][ 560/ 1236] Overall Loss 0.243118 Objective Loss 0.243118 LR 0.000500 Time 0.021766 +2023-10-05 21:35:27,253 - Epoch: [113][ 570/ 1236] Overall Loss 0.242890 Objective Loss 0.242890 LR 0.000500 Time 0.021740 +2023-10-05 21:35:27,457 - Epoch: [113][ 580/ 1236] Overall Loss 0.242589 Objective Loss 0.242589 LR 0.000500 Time 0.021716 +2023-10-05 21:35:27,659 - Epoch: [113][ 590/ 1236] Overall Loss 0.242121 Objective Loss 0.242121 LR 0.000500 Time 0.021690 +2023-10-05 21:35:27,863 - Epoch: [113][ 600/ 1236] Overall Loss 0.241698 Objective Loss 0.241698 LR 0.000500 Time 0.021668 +2023-10-05 21:35:28,066 - Epoch: [113][ 610/ 1236] Overall Loss 0.241855 Objective Loss 0.241855 LR 0.000500 Time 0.021645 +2023-10-05 21:35:28,270 - Epoch: [113][ 620/ 1236] Overall Loss 0.241612 Objective Loss 0.241612 LR 0.000500 Time 0.021624 +2023-10-05 21:35:28,473 - Epoch: [113][ 630/ 1236] Overall Loss 0.241336 Objective Loss 0.241336 LR 0.000500 Time 0.021603 +2023-10-05 21:35:28,677 - Epoch: [113][ 640/ 1236] Overall Loss 0.241471 Objective Loss 0.241471 LR 0.000500 Time 0.021583 +2023-10-05 21:35:28,880 - Epoch: [113][ 650/ 1236] Overall Loss 0.241597 Objective Loss 0.241597 LR 0.000500 Time 0.021563 +2023-10-05 21:35:29,084 - Epoch: [113][ 660/ 1236] Overall Loss 0.241191 Objective Loss 0.241191 LR 0.000500 Time 0.021545 +2023-10-05 21:35:29,287 - Epoch: [113][ 670/ 1236] Overall Loss 0.240914 Objective Loss 0.240914 LR 0.000500 Time 0.021526 +2023-10-05 21:35:29,491 - Epoch: [113][ 680/ 1236] Overall Loss 0.240843 Objective Loss 0.240843 LR 0.000500 Time 0.021509 +2023-10-05 21:35:29,694 - Epoch: [113][ 690/ 1236] Overall Loss 0.240641 Objective Loss 0.240641 LR 0.000500 Time 0.021490 +2023-10-05 21:35:29,898 - Epoch: [113][ 700/ 1236] Overall Loss 0.240478 Objective Loss 0.240478 LR 0.000500 Time 0.021474 +2023-10-05 21:35:30,101 - Epoch: [113][ 710/ 1236] Overall Loss 0.240572 Objective Loss 0.240572 LR 0.000500 Time 0.021457 +2023-10-05 21:35:30,305 - Epoch: [113][ 720/ 1236] Overall Loss 0.240716 Objective Loss 0.240716 LR 0.000500 Time 0.021442 +2023-10-05 21:35:30,508 - Epoch: [113][ 730/ 1236] Overall Loss 0.240809 Objective Loss 0.240809 LR 0.000500 Time 0.021425 +2023-10-05 21:35:30,711 - Epoch: [113][ 740/ 1236] Overall Loss 0.240939 Objective Loss 0.240939 LR 0.000500 Time 0.021410 +2023-10-05 21:35:30,915 - Epoch: [113][ 750/ 1236] Overall Loss 0.241182 Objective Loss 0.241182 LR 0.000500 Time 0.021396 +2023-10-05 21:35:31,120 - Epoch: [113][ 760/ 1236] Overall Loss 0.241334 Objective Loss 0.241334 LR 0.000500 Time 0.021384 +2023-10-05 21:35:31,324 - Epoch: [113][ 770/ 1236] Overall Loss 0.241602 Objective Loss 0.241602 LR 0.000500 Time 0.021370 +2023-10-05 21:35:31,529 - Epoch: [113][ 780/ 1236] Overall Loss 0.241625 Objective Loss 0.241625 LR 0.000500 Time 0.021359 +2023-10-05 21:35:31,732 - Epoch: [113][ 790/ 1236] Overall Loss 0.241594 Objective Loss 0.241594 LR 0.000500 Time 0.021345 +2023-10-05 21:35:31,937 - Epoch: [113][ 800/ 1236] Overall Loss 0.241519 Objective Loss 0.241519 LR 0.000500 Time 0.021334 +2023-10-05 21:35:32,141 - Epoch: [113][ 810/ 1236] Overall Loss 0.241643 Objective Loss 0.241643 LR 0.000500 Time 0.021322 +2023-10-05 21:35:32,346 - Epoch: [113][ 820/ 1236] Overall Loss 0.241483 Objective Loss 0.241483 LR 0.000500 Time 0.021312 +2023-10-05 21:35:32,550 - Epoch: [113][ 830/ 1236] Overall Loss 0.241532 Objective Loss 0.241532 LR 0.000500 Time 0.021300 +2023-10-05 21:35:32,754 - Epoch: [113][ 840/ 1236] Overall Loss 0.241609 Objective Loss 0.241609 LR 0.000500 Time 0.021289 +2023-10-05 21:35:32,956 - Epoch: [113][ 850/ 1236] Overall Loss 0.241986 Objective Loss 0.241986 LR 0.000500 Time 0.021276 +2023-10-05 21:35:33,159 - Epoch: [113][ 860/ 1236] Overall Loss 0.242471 Objective Loss 0.242471 LR 0.000500 Time 0.021264 +2023-10-05 21:35:33,362 - Epoch: [113][ 870/ 1236] Overall Loss 0.242670 Objective Loss 0.242670 LR 0.000500 Time 0.021252 +2023-10-05 21:35:33,565 - Epoch: [113][ 880/ 1236] Overall Loss 0.242691 Objective Loss 0.242691 LR 0.000500 Time 0.021241 +2023-10-05 21:35:33,768 - Epoch: [113][ 890/ 1236] Overall Loss 0.242338 Objective Loss 0.242338 LR 0.000500 Time 0.021230 +2023-10-05 21:35:33,971 - Epoch: [113][ 900/ 1236] Overall Loss 0.242103 Objective Loss 0.242103 LR 0.000500 Time 0.021220 +2023-10-05 21:35:34,175 - Epoch: [113][ 910/ 1236] Overall Loss 0.242111 Objective Loss 0.242111 LR 0.000500 Time 0.021210 +2023-10-05 21:35:34,379 - Epoch: [113][ 920/ 1236] Overall Loss 0.242167 Objective Loss 0.242167 LR 0.000500 Time 0.021201 +2023-10-05 21:35:34,582 - Epoch: [113][ 930/ 1236] Overall Loss 0.242293 Objective Loss 0.242293 LR 0.000500 Time 0.021191 +2023-10-05 21:35:34,785 - Epoch: [113][ 940/ 1236] Overall Loss 0.242297 Objective Loss 0.242297 LR 0.000500 Time 0.021181 +2023-10-05 21:35:34,987 - Epoch: [113][ 950/ 1236] Overall Loss 0.242329 Objective Loss 0.242329 LR 0.000500 Time 0.021171 +2023-10-05 21:35:35,190 - Epoch: [113][ 960/ 1236] Overall Loss 0.242310 Objective Loss 0.242310 LR 0.000500 Time 0.021161 +2023-10-05 21:35:35,392 - Epoch: [113][ 970/ 1236] Overall Loss 0.242247 Objective Loss 0.242247 LR 0.000500 Time 0.021151 +2023-10-05 21:35:35,596 - Epoch: [113][ 980/ 1236] Overall Loss 0.242117 Objective Loss 0.242117 LR 0.000500 Time 0.021143 +2023-10-05 21:35:35,798 - Epoch: [113][ 990/ 1236] Overall Loss 0.241842 Objective Loss 0.241842 LR 0.000500 Time 0.021133 +2023-10-05 21:35:36,001 - Epoch: [113][ 1000/ 1236] Overall Loss 0.241624 Objective Loss 0.241624 LR 0.000500 Time 0.021124 +2023-10-05 21:35:36,204 - Epoch: [113][ 1010/ 1236] Overall Loss 0.241789 Objective Loss 0.241789 LR 0.000500 Time 0.021115 +2023-10-05 21:35:36,407 - Epoch: [113][ 1020/ 1236] Overall Loss 0.241723 Objective Loss 0.241723 LR 0.000500 Time 0.021107 +2023-10-05 21:35:36,609 - Epoch: [113][ 1030/ 1236] Overall Loss 0.241882 Objective Loss 0.241882 LR 0.000500 Time 0.021098 +2023-10-05 21:35:36,812 - Epoch: [113][ 1040/ 1236] Overall Loss 0.241908 Objective Loss 0.241908 LR 0.000500 Time 0.021090 +2023-10-05 21:35:37,015 - Epoch: [113][ 1050/ 1236] Overall Loss 0.241926 Objective Loss 0.241926 LR 0.000500 Time 0.021082 +2023-10-05 21:35:37,218 - Epoch: [113][ 1060/ 1236] Overall Loss 0.241936 Objective Loss 0.241936 LR 0.000500 Time 0.021074 +2023-10-05 21:35:37,420 - Epoch: [113][ 1070/ 1236] Overall Loss 0.241764 Objective Loss 0.241764 LR 0.000500 Time 0.021066 +2023-10-05 21:35:37,624 - Epoch: [113][ 1080/ 1236] Overall Loss 0.241642 Objective Loss 0.241642 LR 0.000500 Time 0.021059 +2023-10-05 21:35:37,826 - Epoch: [113][ 1090/ 1236] Overall Loss 0.241439 Objective Loss 0.241439 LR 0.000500 Time 0.021051 +2023-10-05 21:35:38,029 - Epoch: [113][ 1100/ 1236] Overall Loss 0.241508 Objective Loss 0.241508 LR 0.000500 Time 0.021044 +2023-10-05 21:35:38,232 - Epoch: [113][ 1110/ 1236] Overall Loss 0.241112 Objective Loss 0.241112 LR 0.000500 Time 0.021037 +2023-10-05 21:35:38,435 - Epoch: [113][ 1120/ 1236] Overall Loss 0.241177 Objective Loss 0.241177 LR 0.000500 Time 0.021030 +2023-10-05 21:35:38,637 - Epoch: [113][ 1130/ 1236] Overall Loss 0.241211 Objective Loss 0.241211 LR 0.000500 Time 0.021022 +2023-10-05 21:35:38,840 - Epoch: [113][ 1140/ 1236] Overall Loss 0.241115 Objective Loss 0.241115 LR 0.000500 Time 0.021016 +2023-10-05 21:35:39,043 - Epoch: [113][ 1150/ 1236] Overall Loss 0.241095 Objective Loss 0.241095 LR 0.000500 Time 0.021009 +2023-10-05 21:35:39,246 - Epoch: [113][ 1160/ 1236] Overall Loss 0.241320 Objective Loss 0.241320 LR 0.000500 Time 0.021003 +2023-10-05 21:35:39,448 - Epoch: [113][ 1170/ 1236] Overall Loss 0.241395 Objective Loss 0.241395 LR 0.000500 Time 0.020996 +2023-10-05 21:35:39,652 - Epoch: [113][ 1180/ 1236] Overall Loss 0.241638 Objective Loss 0.241638 LR 0.000500 Time 0.020990 +2023-10-05 21:35:39,854 - Epoch: [113][ 1190/ 1236] Overall Loss 0.241793 Objective Loss 0.241793 LR 0.000500 Time 0.020983 +2023-10-05 21:35:40,058 - Epoch: [113][ 1200/ 1236] Overall Loss 0.241886 Objective Loss 0.241886 LR 0.000500 Time 0.020978 +2023-10-05 21:35:40,260 - Epoch: [113][ 1210/ 1236] Overall Loss 0.241911 Objective Loss 0.241911 LR 0.000500 Time 0.020971 +2023-10-05 21:35:40,463 - Epoch: [113][ 1220/ 1236] Overall Loss 0.241779 Objective Loss 0.241779 LR 0.000500 Time 0.020966 +2023-10-05 21:35:40,719 - Epoch: [113][ 1230/ 1236] Overall Loss 0.242002 Objective Loss 0.242002 LR 0.000500 Time 0.021003 +2023-10-05 21:35:40,838 - Epoch: [113][ 1236/ 1236] Overall Loss 0.241905 Objective Loss 0.241905 Top1 87.169043 Top5 97.963340 LR 0.000500 Time 0.020997 +2023-10-05 21:35:40,972 - --- validate (epoch=113)----------- +2023-10-05 21:35:40,973 - 29943 samples (256 per mini-batch) +2023-10-05 21:35:41,438 - Epoch: [113][ 10/ 117] Loss 0.352805 Top1 83.593750 Top5 97.968750 +2023-10-05 21:35:41,601 - Epoch: [113][ 20/ 117] Loss 0.342414 Top1 84.160156 Top5 97.851562 +2023-10-05 21:35:41,757 - Epoch: [113][ 30/ 117] Loss 0.331966 Top1 84.531250 Top5 97.864583 +2023-10-05 21:35:41,920 - Epoch: [113][ 40/ 117] Loss 0.337310 Top1 84.169922 Top5 97.812500 +2023-10-05 21:35:42,076 - Epoch: [113][ 50/ 117] Loss 0.333434 Top1 84.187500 Top5 97.914062 +2023-10-05 21:35:42,236 - Epoch: [113][ 60/ 117] Loss 0.326956 Top1 84.277344 Top5 97.968750 +2023-10-05 21:35:42,392 - Epoch: [113][ 70/ 117] Loss 0.324260 Top1 84.185268 Top5 98.013393 +2023-10-05 21:35:42,552 - Epoch: [113][ 80/ 117] Loss 0.327531 Top1 84.008789 Top5 98.017578 +2023-10-05 21:35:42,707 - Epoch: [113][ 90/ 117] Loss 0.327686 Top1 83.914931 Top5 97.951389 +2023-10-05 21:35:42,868 - Epoch: [113][ 100/ 117] Loss 0.329928 Top1 83.960938 Top5 97.941406 +2023-10-05 21:35:43,029 - Epoch: [113][ 110/ 117] Loss 0.331100 Top1 83.849432 Top5 97.890625 +2023-10-05 21:35:43,115 - Epoch: [113][ 117/ 117] Loss 0.332557 Top1 83.819257 Top5 97.899342 +2023-10-05 21:35:43,254 - ==> Top1: 83.819 Top5: 97.899 Loss: 0.333 + +2023-10-05 21:35:43,255 - ==> Confusion: +[[ 907 1 3 1 12 1 0 2 10 78 1 2 2 4 6 4 4 0 2 0 10] + [ 1 1053 3 0 10 13 1 21 0 0 4 1 0 0 1 4 2 0 12 1 4] + [ 3 2 951 12 3 0 16 14 0 2 7 2 5 0 3 3 4 4 11 7 7] + [ 3 1 16 940 2 3 1 0 3 0 9 0 3 5 38 4 3 7 33 3 15] + [ 19 3 0 0 978 3 0 1 0 9 4 0 1 1 10 3 6 1 3 2 6] + [ 4 38 0 0 4 967 1 28 2 1 5 5 2 20 8 1 3 0 6 5 16] + [ 0 4 33 0 0 2 1106 11 0 0 6 3 1 0 1 8 0 2 4 5 5] + [ 4 16 15 0 2 23 5 1072 2 5 3 6 0 2 2 2 0 0 45 4 10] + [ 19 1 0 1 0 3 2 1 955 48 18 2 2 13 12 2 0 2 4 2 2] + [ 96 0 3 1 8 3 0 0 21 938 0 1 0 28 5 1 0 1 1 4 8] + [ 4 5 13 7 2 0 3 2 11 2 963 1 0 9 3 2 2 0 13 1 10] + [ 1 1 2 0 1 11 0 2 1 2 0 937 31 8 0 2 1 19 0 10 6] + [ 2 1 3 2 1 3 0 1 2 0 2 30 978 3 1 4 2 18 1 2 12] + [ 2 0 1 1 5 2 0 1 6 11 7 2 3 1065 2 2 1 0 0 1 7] + [ 11 2 3 9 11 0 0 0 26 4 2 1 0 1 1000 0 0 1 17 0 13] + [ 1 1 2 3 5 1 3 0 0 0 1 8 6 1 2 1061 12 12 0 10 5] + [ 0 11 2 0 6 4 0 2 3 0 0 2 3 1 2 10 1102 0 1 3 9] + [ 0 0 0 0 0 0 0 0 0 1 0 0 19 0 2 6 1 1001 1 4 3] + [ 1 4 9 9 2 0 0 23 3 0 1 0 5 0 6 0 1 0 998 0 6] + [ 1 5 4 1 2 7 8 9 1 0 3 17 3 1 0 6 12 1 4 1056 11] + [ 150 179 157 59 110 123 40 97 92 89 186 118 319 316 128 56 165 76 215 160 5070]] + +2023-10-05 21:35:43,256 - ==> Best [Top1: 83.886 Top5: 97.879 Sparsity:0.00 Params: 148928 on epoch: 107] +2023-10-05 21:35:43,256 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:35:43,262 - + +2023-10-05 21:35:43,262 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:35:44,389 - Epoch: [114][ 10/ 1236] Overall Loss 0.224354 Objective Loss 0.224354 LR 0.000500 Time 0.112648 +2023-10-05 21:35:44,592 - Epoch: [114][ 20/ 1236] Overall Loss 0.233019 Objective Loss 0.233019 LR 0.000500 Time 0.066456 +2023-10-05 21:35:44,795 - Epoch: [114][ 30/ 1236] Overall Loss 0.232900 Objective Loss 0.232900 LR 0.000500 Time 0.051052 +2023-10-05 21:35:44,999 - Epoch: [114][ 40/ 1236] Overall Loss 0.239380 Objective Loss 0.239380 LR 0.000500 Time 0.043379 +2023-10-05 21:35:45,203 - Epoch: [114][ 50/ 1236] Overall Loss 0.239504 Objective Loss 0.239504 LR 0.000500 Time 0.038777 +2023-10-05 21:35:45,408 - Epoch: [114][ 60/ 1236] Overall Loss 0.240486 Objective Loss 0.240486 LR 0.000500 Time 0.035726 +2023-10-05 21:35:45,612 - Epoch: [114][ 70/ 1236] Overall Loss 0.241875 Objective Loss 0.241875 LR 0.000500 Time 0.033537 +2023-10-05 21:35:45,816 - Epoch: [114][ 80/ 1236] Overall Loss 0.242017 Objective Loss 0.242017 LR 0.000500 Time 0.031887 +2023-10-05 21:35:46,020 - Epoch: [114][ 90/ 1236] Overall Loss 0.241923 Objective Loss 0.241923 LR 0.000500 Time 0.030605 +2023-10-05 21:35:46,223 - Epoch: [114][ 100/ 1236] Overall Loss 0.241134 Objective Loss 0.241134 LR 0.000500 Time 0.029567 +2023-10-05 21:35:46,425 - Epoch: [114][ 110/ 1236] Overall Loss 0.238506 Objective Loss 0.238506 LR 0.000500 Time 0.028716 +2023-10-05 21:35:46,631 - Epoch: [114][ 120/ 1236] Overall Loss 0.239599 Objective Loss 0.239599 LR 0.000500 Time 0.028031 +2023-10-05 21:35:46,836 - Epoch: [114][ 130/ 1236] Overall Loss 0.240049 Objective Loss 0.240049 LR 0.000500 Time 0.027453 +2023-10-05 21:35:47,043 - Epoch: [114][ 140/ 1236] Overall Loss 0.238173 Objective Loss 0.238173 LR 0.000500 Time 0.026963 +2023-10-05 21:35:47,249 - Epoch: [114][ 150/ 1236] Overall Loss 0.237575 Objective Loss 0.237575 LR 0.000500 Time 0.026535 +2023-10-05 21:35:47,456 - Epoch: [114][ 160/ 1236] Overall Loss 0.236299 Objective Loss 0.236299 LR 0.000500 Time 0.026168 +2023-10-05 21:35:47,663 - Epoch: [114][ 170/ 1236] Overall Loss 0.236987 Objective Loss 0.236987 LR 0.000500 Time 0.025842 +2023-10-05 21:35:47,872 - Epoch: [114][ 180/ 1236] Overall Loss 0.237776 Objective Loss 0.237776 LR 0.000500 Time 0.025569 +2023-10-05 21:35:48,079 - Epoch: [114][ 190/ 1236] Overall Loss 0.238515 Objective Loss 0.238515 LR 0.000500 Time 0.025307 +2023-10-05 21:35:48,286 - Epoch: [114][ 200/ 1236] Overall Loss 0.239699 Objective Loss 0.239699 LR 0.000500 Time 0.025073 +2023-10-05 21:35:48,492 - Epoch: [114][ 210/ 1236] Overall Loss 0.240560 Objective Loss 0.240560 LR 0.000500 Time 0.024860 +2023-10-05 21:35:48,698 - Epoch: [114][ 220/ 1236] Overall Loss 0.240820 Objective Loss 0.240820 LR 0.000500 Time 0.024667 +2023-10-05 21:35:48,904 - Epoch: [114][ 230/ 1236] Overall Loss 0.240801 Objective Loss 0.240801 LR 0.000500 Time 0.024488 +2023-10-05 21:35:49,111 - Epoch: [114][ 240/ 1236] Overall Loss 0.241131 Objective Loss 0.241131 LR 0.000500 Time 0.024327 +2023-10-05 21:35:49,318 - Epoch: [114][ 250/ 1236] Overall Loss 0.240243 Objective Loss 0.240243 LR 0.000500 Time 0.024179 +2023-10-05 21:35:49,524 - Epoch: [114][ 260/ 1236] Overall Loss 0.239149 Objective Loss 0.239149 LR 0.000500 Time 0.024042 +2023-10-05 21:35:49,731 - Epoch: [114][ 270/ 1236] Overall Loss 0.239305 Objective Loss 0.239305 LR 0.000500 Time 0.023916 +2023-10-05 21:35:49,938 - Epoch: [114][ 280/ 1236] Overall Loss 0.239419 Objective Loss 0.239419 LR 0.000500 Time 0.023798 +2023-10-05 21:35:50,144 - Epoch: [114][ 290/ 1236] Overall Loss 0.239312 Objective Loss 0.239312 LR 0.000500 Time 0.023688 +2023-10-05 21:35:50,351 - Epoch: [114][ 300/ 1236] Overall Loss 0.239605 Objective Loss 0.239605 LR 0.000500 Time 0.023585 +2023-10-05 21:35:50,557 - Epoch: [114][ 310/ 1236] Overall Loss 0.239162 Objective Loss 0.239162 LR 0.000500 Time 0.023489 +2023-10-05 21:35:50,764 - Epoch: [114][ 320/ 1236] Overall Loss 0.238935 Objective Loss 0.238935 LR 0.000500 Time 0.023401 +2023-10-05 21:35:50,969 - Epoch: [114][ 330/ 1236] Overall Loss 0.238759 Objective Loss 0.238759 LR 0.000500 Time 0.023312 +2023-10-05 21:35:51,175 - Epoch: [114][ 340/ 1236] Overall Loss 0.239025 Objective Loss 0.239025 LR 0.000500 Time 0.023231 +2023-10-05 21:35:51,381 - Epoch: [114][ 350/ 1236] Overall Loss 0.239122 Objective Loss 0.239122 LR 0.000500 Time 0.023155 +2023-10-05 21:35:51,588 - Epoch: [114][ 360/ 1236] Overall Loss 0.239134 Objective Loss 0.239134 LR 0.000500 Time 0.023084 +2023-10-05 21:35:51,793 - Epoch: [114][ 370/ 1236] Overall Loss 0.239236 Objective Loss 0.239236 LR 0.000500 Time 0.023013 +2023-10-05 21:35:51,998 - Epoch: [114][ 380/ 1236] Overall Loss 0.239215 Objective Loss 0.239215 LR 0.000500 Time 0.022946 +2023-10-05 21:35:52,203 - Epoch: [114][ 390/ 1236] Overall Loss 0.239073 Objective Loss 0.239073 LR 0.000500 Time 0.022884 +2023-10-05 21:35:52,408 - Epoch: [114][ 400/ 1236] Overall Loss 0.239338 Objective Loss 0.239338 LR 0.000500 Time 0.022823 +2023-10-05 21:35:52,613 - Epoch: [114][ 410/ 1236] Overall Loss 0.238832 Objective Loss 0.238832 LR 0.000500 Time 0.022766 +2023-10-05 21:35:52,818 - Epoch: [114][ 420/ 1236] Overall Loss 0.238707 Objective Loss 0.238707 LR 0.000500 Time 0.022711 +2023-10-05 21:35:53,024 - Epoch: [114][ 430/ 1236] Overall Loss 0.238170 Objective Loss 0.238170 LR 0.000500 Time 0.022659 +2023-10-05 21:35:53,228 - Epoch: [114][ 440/ 1236] Overall Loss 0.237377 Objective Loss 0.237377 LR 0.000500 Time 0.022608 +2023-10-05 21:35:53,434 - Epoch: [114][ 450/ 1236] Overall Loss 0.237263 Objective Loss 0.237263 LR 0.000500 Time 0.022561 +2023-10-05 21:35:53,638 - Epoch: [114][ 460/ 1236] Overall Loss 0.236898 Objective Loss 0.236898 LR 0.000500 Time 0.022515 +2023-10-05 21:35:53,843 - Epoch: [114][ 470/ 1236] Overall Loss 0.237359 Objective Loss 0.237359 LR 0.000500 Time 0.022471 +2023-10-05 21:35:54,048 - Epoch: [114][ 480/ 1236] Overall Loss 0.236834 Objective Loss 0.236834 LR 0.000500 Time 0.022428 +2023-10-05 21:35:54,253 - Epoch: [114][ 490/ 1236] Overall Loss 0.236983 Objective Loss 0.236983 LR 0.000500 Time 0.022389 +2023-10-05 21:35:54,458 - Epoch: [114][ 500/ 1236] Overall Loss 0.237162 Objective Loss 0.237162 LR 0.000500 Time 0.022350 +2023-10-05 21:35:54,663 - Epoch: [114][ 510/ 1236] Overall Loss 0.237210 Objective Loss 0.237210 LR 0.000500 Time 0.022313 +2023-10-05 21:35:54,868 - Epoch: [114][ 520/ 1236] Overall Loss 0.237385 Objective Loss 0.237385 LR 0.000500 Time 0.022278 +2023-10-05 21:35:55,074 - Epoch: [114][ 530/ 1236] Overall Loss 0.237712 Objective Loss 0.237712 LR 0.000500 Time 0.022244 +2023-10-05 21:35:55,278 - Epoch: [114][ 540/ 1236] Overall Loss 0.237542 Objective Loss 0.237542 LR 0.000500 Time 0.022210 +2023-10-05 21:35:55,484 - Epoch: [114][ 550/ 1236] Overall Loss 0.237433 Objective Loss 0.237433 LR 0.000500 Time 0.022179 +2023-10-05 21:35:55,688 - Epoch: [114][ 560/ 1236] Overall Loss 0.237416 Objective Loss 0.237416 LR 0.000500 Time 0.022148 +2023-10-05 21:35:55,893 - Epoch: [114][ 570/ 1236] Overall Loss 0.237532 Objective Loss 0.237532 LR 0.000500 Time 0.022118 +2023-10-05 21:35:56,097 - Epoch: [114][ 580/ 1236] Overall Loss 0.238002 Objective Loss 0.238002 LR 0.000500 Time 0.022087 +2023-10-05 21:35:56,302 - Epoch: [114][ 590/ 1236] Overall Loss 0.237843 Objective Loss 0.237843 LR 0.000500 Time 0.022060 +2023-10-05 21:35:56,506 - Epoch: [114][ 600/ 1236] Overall Loss 0.237740 Objective Loss 0.237740 LR 0.000500 Time 0.022032 +2023-10-05 21:35:56,711 - Epoch: [114][ 610/ 1236] Overall Loss 0.237753 Objective Loss 0.237753 LR 0.000500 Time 0.022007 +2023-10-05 21:35:56,916 - Epoch: [114][ 620/ 1236] Overall Loss 0.237713 Objective Loss 0.237713 LR 0.000500 Time 0.021981 +2023-10-05 21:35:57,122 - Epoch: [114][ 630/ 1236] Overall Loss 0.237680 Objective Loss 0.237680 LR 0.000500 Time 0.021958 +2023-10-05 21:35:57,327 - Epoch: [114][ 640/ 1236] Overall Loss 0.237162 Objective Loss 0.237162 LR 0.000500 Time 0.021934 +2023-10-05 21:35:57,532 - Epoch: [114][ 650/ 1236] Overall Loss 0.236951 Objective Loss 0.236951 LR 0.000500 Time 0.021912 +2023-10-05 21:35:57,736 - Epoch: [114][ 660/ 1236] Overall Loss 0.237410 Objective Loss 0.237410 LR 0.000500 Time 0.021889 +2023-10-05 21:35:57,942 - Epoch: [114][ 670/ 1236] Overall Loss 0.237901 Objective Loss 0.237901 LR 0.000500 Time 0.021868 +2023-10-05 21:35:58,146 - Epoch: [114][ 680/ 1236] Overall Loss 0.237496 Objective Loss 0.237496 LR 0.000500 Time 0.021847 +2023-10-05 21:35:58,352 - Epoch: [114][ 690/ 1236] Overall Loss 0.237515 Objective Loss 0.237515 LR 0.000500 Time 0.021828 +2023-10-05 21:35:58,556 - Epoch: [114][ 700/ 1236] Overall Loss 0.237618 Objective Loss 0.237618 LR 0.000500 Time 0.021808 +2023-10-05 21:35:58,762 - Epoch: [114][ 710/ 1236] Overall Loss 0.238290 Objective Loss 0.238290 LR 0.000500 Time 0.021789 +2023-10-05 21:35:58,966 - Epoch: [114][ 720/ 1236] Overall Loss 0.238310 Objective Loss 0.238310 LR 0.000500 Time 0.021770 +2023-10-05 21:35:59,172 - Epoch: [114][ 730/ 1236] Overall Loss 0.238421 Objective Loss 0.238421 LR 0.000500 Time 0.021753 +2023-10-05 21:35:59,376 - Epoch: [114][ 740/ 1236] Overall Loss 0.238302 Objective Loss 0.238302 LR 0.000500 Time 0.021735 +2023-10-05 21:35:59,582 - Epoch: [114][ 750/ 1236] Overall Loss 0.238162 Objective Loss 0.238162 LR 0.000500 Time 0.021718 +2023-10-05 21:35:59,786 - Epoch: [114][ 760/ 1236] Overall Loss 0.238604 Objective Loss 0.238604 LR 0.000500 Time 0.021701 +2023-10-05 21:35:59,991 - Epoch: [114][ 770/ 1236] Overall Loss 0.238761 Objective Loss 0.238761 LR 0.000500 Time 0.021685 +2023-10-05 21:36:00,196 - Epoch: [114][ 780/ 1236] Overall Loss 0.239641 Objective Loss 0.239641 LR 0.000500 Time 0.021669 +2023-10-05 21:36:00,402 - Epoch: [114][ 790/ 1236] Overall Loss 0.239818 Objective Loss 0.239818 LR 0.000500 Time 0.021654 +2023-10-05 21:36:00,606 - Epoch: [114][ 800/ 1236] Overall Loss 0.240201 Objective Loss 0.240201 LR 0.000500 Time 0.021639 +2023-10-05 21:36:00,812 - Epoch: [114][ 810/ 1236] Overall Loss 0.240343 Objective Loss 0.240343 LR 0.000500 Time 0.021625 +2023-10-05 21:36:01,016 - Epoch: [114][ 820/ 1236] Overall Loss 0.240214 Objective Loss 0.240214 LR 0.000500 Time 0.021610 +2023-10-05 21:36:01,222 - Epoch: [114][ 830/ 1236] Overall Loss 0.240291 Objective Loss 0.240291 LR 0.000500 Time 0.021597 +2023-10-05 21:36:01,426 - Epoch: [114][ 840/ 1236] Overall Loss 0.240710 Objective Loss 0.240710 LR 0.000500 Time 0.021583 +2023-10-05 21:36:01,631 - Epoch: [114][ 850/ 1236] Overall Loss 0.240817 Objective Loss 0.240817 LR 0.000500 Time 0.021570 +2023-10-05 21:36:01,836 - Epoch: [114][ 860/ 1236] Overall Loss 0.240801 Objective Loss 0.240801 LR 0.000500 Time 0.021557 +2023-10-05 21:36:02,042 - Epoch: [114][ 870/ 1236] Overall Loss 0.240647 Objective Loss 0.240647 LR 0.000500 Time 0.021545 +2023-10-05 21:36:02,247 - Epoch: [114][ 880/ 1236] Overall Loss 0.240595 Objective Loss 0.240595 LR 0.000500 Time 0.021531 +2023-10-05 21:36:02,453 - Epoch: [114][ 890/ 1236] Overall Loss 0.240877 Objective Loss 0.240877 LR 0.000500 Time 0.021519 +2023-10-05 21:36:02,657 - Epoch: [114][ 900/ 1236] Overall Loss 0.240666 Objective Loss 0.240666 LR 0.000500 Time 0.021506 +2023-10-05 21:36:02,862 - Epoch: [114][ 910/ 1236] Overall Loss 0.240473 Objective Loss 0.240473 LR 0.000500 Time 0.021495 +2023-10-05 21:36:03,068 - Epoch: [114][ 920/ 1236] Overall Loss 0.240714 Objective Loss 0.240714 LR 0.000500 Time 0.021484 +2023-10-05 21:36:03,273 - Epoch: [114][ 930/ 1236] Overall Loss 0.240491 Objective Loss 0.240491 LR 0.000500 Time 0.021473 +2023-10-05 21:36:03,477 - Epoch: [114][ 940/ 1236] Overall Loss 0.240670 Objective Loss 0.240670 LR 0.000500 Time 0.021462 +2023-10-05 21:36:03,683 - Epoch: [114][ 950/ 1236] Overall Loss 0.240716 Objective Loss 0.240716 LR 0.000500 Time 0.021452 +2023-10-05 21:36:03,886 - Epoch: [114][ 960/ 1236] Overall Loss 0.240756 Objective Loss 0.240756 LR 0.000500 Time 0.021440 +2023-10-05 21:36:04,092 - Epoch: [114][ 970/ 1236] Overall Loss 0.241035 Objective Loss 0.241035 LR 0.000500 Time 0.021431 +2023-10-05 21:36:04,296 - Epoch: [114][ 980/ 1236] Overall Loss 0.241081 Objective Loss 0.241081 LR 0.000500 Time 0.021420 +2023-10-05 21:36:04,502 - Epoch: [114][ 990/ 1236] Overall Loss 0.241237 Objective Loss 0.241237 LR 0.000500 Time 0.021411 +2023-10-05 21:36:04,706 - Epoch: [114][ 1000/ 1236] Overall Loss 0.241148 Objective Loss 0.241148 LR 0.000500 Time 0.021401 +2023-10-05 21:36:04,912 - Epoch: [114][ 1010/ 1236] Overall Loss 0.241208 Objective Loss 0.241208 LR 0.000500 Time 0.021392 +2023-10-05 21:36:05,116 - Epoch: [114][ 1020/ 1236] Overall Loss 0.241331 Objective Loss 0.241331 LR 0.000500 Time 0.021383 +2023-10-05 21:36:05,320 - Epoch: [114][ 1030/ 1236] Overall Loss 0.240956 Objective Loss 0.240956 LR 0.000500 Time 0.021373 +2023-10-05 21:36:05,525 - Epoch: [114][ 1040/ 1236] Overall Loss 0.241105 Objective Loss 0.241105 LR 0.000500 Time 0.021364 +2023-10-05 21:36:05,732 - Epoch: [114][ 1050/ 1236] Overall Loss 0.241290 Objective Loss 0.241290 LR 0.000500 Time 0.021357 +2023-10-05 21:36:05,937 - Epoch: [114][ 1060/ 1236] Overall Loss 0.241559 Objective Loss 0.241559 LR 0.000500 Time 0.021349 +2023-10-05 21:36:06,142 - Epoch: [114][ 1070/ 1236] Overall Loss 0.241448 Objective Loss 0.241448 LR 0.000500 Time 0.021340 +2023-10-05 21:36:06,347 - Epoch: [114][ 1080/ 1236] Overall Loss 0.241239 Objective Loss 0.241239 LR 0.000500 Time 0.021332 +2023-10-05 21:36:06,552 - Epoch: [114][ 1090/ 1236] Overall Loss 0.241378 Objective Loss 0.241378 LR 0.000500 Time 0.021324 +2023-10-05 21:36:06,757 - Epoch: [114][ 1100/ 1236] Overall Loss 0.241180 Objective Loss 0.241180 LR 0.000500 Time 0.021316 +2023-10-05 21:36:06,962 - Epoch: [114][ 1110/ 1236] Overall Loss 0.241205 Objective Loss 0.241205 LR 0.000500 Time 0.021309 +2023-10-05 21:36:07,167 - Epoch: [114][ 1120/ 1236] Overall Loss 0.241350 Objective Loss 0.241350 LR 0.000500 Time 0.021301 +2023-10-05 21:36:07,372 - Epoch: [114][ 1130/ 1236] Overall Loss 0.241404 Objective Loss 0.241404 LR 0.000500 Time 0.021293 +2023-10-05 21:36:07,576 - Epoch: [114][ 1140/ 1236] Overall Loss 0.241384 Objective Loss 0.241384 LR 0.000500 Time 0.021285 +2023-10-05 21:36:07,781 - Epoch: [114][ 1150/ 1236] Overall Loss 0.241605 Objective Loss 0.241605 LR 0.000500 Time 0.021278 +2023-10-05 21:36:07,986 - Epoch: [114][ 1160/ 1236] Overall Loss 0.241856 Objective Loss 0.241856 LR 0.000500 Time 0.021271 +2023-10-05 21:36:08,191 - Epoch: [114][ 1170/ 1236] Overall Loss 0.241784 Objective Loss 0.241784 LR 0.000500 Time 0.021264 +2023-10-05 21:36:08,398 - Epoch: [114][ 1180/ 1236] Overall Loss 0.241801 Objective Loss 0.241801 LR 0.000500 Time 0.021258 +2023-10-05 21:36:08,604 - Epoch: [114][ 1190/ 1236] Overall Loss 0.241880 Objective Loss 0.241880 LR 0.000500 Time 0.021253 +2023-10-05 21:36:08,810 - Epoch: [114][ 1200/ 1236] Overall Loss 0.241915 Objective Loss 0.241915 LR 0.000500 Time 0.021247 +2023-10-05 21:36:09,015 - Epoch: [114][ 1210/ 1236] Overall Loss 0.241973 Objective Loss 0.241973 LR 0.000500 Time 0.021240 +2023-10-05 21:36:09,219 - Epoch: [114][ 1220/ 1236] Overall Loss 0.242106 Objective Loss 0.242106 LR 0.000500 Time 0.021234 +2023-10-05 21:36:09,477 - Epoch: [114][ 1230/ 1236] Overall Loss 0.242122 Objective Loss 0.242122 LR 0.000500 Time 0.021270 +2023-10-05 21:36:09,596 - Epoch: [114][ 1236/ 1236] Overall Loss 0.242382 Objective Loss 0.242382 Top1 85.336049 Top5 98.167006 LR 0.000500 Time 0.021263 +2023-10-05 21:36:09,717 - --- validate (epoch=114)----------- +2023-10-05 21:36:09,717 - 29943 samples (256 per mini-batch) +2023-10-05 21:36:10,166 - Epoch: [114][ 10/ 117] Loss 0.323019 Top1 84.218750 Top5 98.046875 +2023-10-05 21:36:10,311 - Epoch: [114][ 20/ 117] Loss 0.327140 Top1 84.296875 Top5 98.105469 +2023-10-05 21:36:10,457 - Epoch: [114][ 30/ 117] Loss 0.333849 Top1 83.984375 Top5 98.125000 +2023-10-05 21:36:10,600 - Epoch: [114][ 40/ 117] Loss 0.326159 Top1 84.130859 Top5 98.095703 +2023-10-05 21:36:10,741 - Epoch: [114][ 50/ 117] Loss 0.318439 Top1 84.296875 Top5 98.117188 +2023-10-05 21:36:10,883 - Epoch: [114][ 60/ 117] Loss 0.316319 Top1 84.394531 Top5 98.085938 +2023-10-05 21:36:11,027 - Epoch: [114][ 70/ 117] Loss 0.319869 Top1 84.341518 Top5 98.074777 +2023-10-05 21:36:11,173 - Epoch: [114][ 80/ 117] Loss 0.323998 Top1 84.252930 Top5 98.037109 +2023-10-05 21:36:11,318 - Epoch: [114][ 90/ 117] Loss 0.323857 Top1 84.244792 Top5 98.046875 +2023-10-05 21:36:11,462 - Epoch: [114][ 100/ 117] Loss 0.327579 Top1 84.089844 Top5 97.980469 +2023-10-05 21:36:11,612 - Epoch: [114][ 110/ 117] Loss 0.329115 Top1 84.105114 Top5 97.933239 +2023-10-05 21:36:11,698 - Epoch: [114][ 117/ 117] Loss 0.328458 Top1 84.116488 Top5 97.959456 +2023-10-05 21:36:11,850 - ==> Top1: 84.116 Top5: 97.959 Loss: 0.328 + +2023-10-05 21:36:11,851 - ==> Confusion: +[[ 927 3 2 1 9 2 1 1 3 61 1 1 1 1 9 4 3 1 0 0 19] + [ 1 1058 2 0 8 18 1 14 2 0 3 0 0 1 1 4 3 0 6 1 8] + [ 3 4 956 12 0 0 22 12 0 0 4 0 10 2 0 5 2 1 9 6 8] + [ 1 1 17 971 2 2 2 1 2 1 7 0 2 3 21 4 1 7 26 3 15] + [ 20 4 0 0 980 1 0 0 0 8 0 1 1 1 9 6 8 1 2 1 7] + [ 2 47 0 1 5 968 2 26 3 0 6 8 3 12 5 1 2 1 2 11 11] + [ 0 4 16 0 0 1 1129 14 0 0 5 2 1 0 1 5 0 0 1 7 5] + [ 4 24 7 0 3 32 3 1067 0 4 4 6 2 2 0 5 0 0 33 9 13] + [ 21 2 0 2 1 3 1 0 961 38 15 2 2 9 19 3 0 1 6 1 2] + [ 104 0 3 2 10 3 1 0 31 917 0 0 0 19 8 4 1 0 1 3 12] + [ 2 3 13 6 2 0 2 3 8 1 976 3 0 12 3 1 1 0 5 3 9] + [ 1 0 3 0 0 14 0 3 0 2 0 932 34 4 0 3 2 17 0 16 4] + [ 1 1 3 8 1 2 1 2 0 0 1 34 962 4 0 10 2 18 1 4 13] + [ 2 0 1 2 3 6 0 0 10 12 2 7 3 1048 4 2 2 0 0 2 13] + [ 12 4 3 19 6 1 1 0 18 2 1 1 1 2 1000 0 0 0 11 0 19] + [ 0 1 1 0 5 1 2 0 0 0 0 8 3 4 1 1067 17 10 1 8 5] + [ 1 12 1 0 5 1 1 3 3 0 0 1 0 0 1 14 1100 1 0 5 12] + [ 0 0 0 3 0 0 0 0 1 0 0 1 17 1 2 8 0 998 1 1 5] + [ 2 9 11 18 0 0 1 25 3 0 1 0 4 1 6 0 2 0 977 0 8] + [ 0 3 2 2 1 6 9 11 1 0 1 10 4 1 0 6 9 1 1 1077 7] + [ 121 186 129 94 105 128 52 102 91 73 175 97 336 277 135 86 165 70 140 227 5116]] + +2023-10-05 21:36:11,852 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:36:11,852 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:36:11,866 - + +2023-10-05 21:36:11,866 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:36:12,885 - Epoch: [115][ 10/ 1236] Overall Loss 0.228005 Objective Loss 0.228005 LR 0.000500 Time 0.101881 +2023-10-05 21:36:13,093 - Epoch: [115][ 20/ 1236] Overall Loss 0.235768 Objective Loss 0.235768 LR 0.000500 Time 0.061298 +2023-10-05 21:36:13,301 - Epoch: [115][ 30/ 1236] Overall Loss 0.232181 Objective Loss 0.232181 LR 0.000500 Time 0.047783 +2023-10-05 21:36:13,509 - Epoch: [115][ 40/ 1236] Overall Loss 0.225744 Objective Loss 0.225744 LR 0.000500 Time 0.041043 +2023-10-05 21:36:13,721 - Epoch: [115][ 50/ 1236] Overall Loss 0.228008 Objective Loss 0.228008 LR 0.000500 Time 0.037058 +2023-10-05 21:36:13,933 - Epoch: [115][ 60/ 1236] Overall Loss 0.228541 Objective Loss 0.228541 LR 0.000500 Time 0.034384 +2023-10-05 21:36:14,141 - Epoch: [115][ 70/ 1236] Overall Loss 0.230950 Objective Loss 0.230950 LR 0.000500 Time 0.032447 +2023-10-05 21:36:14,354 - Epoch: [115][ 80/ 1236] Overall Loss 0.231069 Objective Loss 0.231069 LR 0.000500 Time 0.031045 +2023-10-05 21:36:14,574 - Epoch: [115][ 90/ 1236] Overall Loss 0.231088 Objective Loss 0.231088 LR 0.000500 Time 0.030031 +2023-10-05 21:36:14,797 - Epoch: [115][ 100/ 1236] Overall Loss 0.232615 Objective Loss 0.232615 LR 0.000500 Time 0.029258 +2023-10-05 21:36:15,015 - Epoch: [115][ 110/ 1236] Overall Loss 0.234401 Objective Loss 0.234401 LR 0.000500 Time 0.028574 +2023-10-05 21:36:15,233 - Epoch: [115][ 120/ 1236] Overall Loss 0.235450 Objective Loss 0.235450 LR 0.000500 Time 0.028005 +2023-10-05 21:36:15,442 - Epoch: [115][ 130/ 1236] Overall Loss 0.235522 Objective Loss 0.235522 LR 0.000500 Time 0.027454 +2023-10-05 21:36:15,659 - Epoch: [115][ 140/ 1236] Overall Loss 0.234906 Objective Loss 0.234906 LR 0.000500 Time 0.027044 +2023-10-05 21:36:15,871 - Epoch: [115][ 150/ 1236] Overall Loss 0.234557 Objective Loss 0.234557 LR 0.000500 Time 0.026651 +2023-10-05 21:36:16,082 - Epoch: [115][ 160/ 1236] Overall Loss 0.235187 Objective Loss 0.235187 LR 0.000500 Time 0.026300 +2023-10-05 21:36:16,290 - Epoch: [115][ 170/ 1236] Overall Loss 0.236594 Objective Loss 0.236594 LR 0.000500 Time 0.025977 +2023-10-05 21:36:16,503 - Epoch: [115][ 180/ 1236] Overall Loss 0.236370 Objective Loss 0.236370 LR 0.000500 Time 0.025712 +2023-10-05 21:36:16,713 - Epoch: [115][ 190/ 1236] Overall Loss 0.236754 Objective Loss 0.236754 LR 0.000500 Time 0.025461 +2023-10-05 21:36:16,935 - Epoch: [115][ 200/ 1236] Overall Loss 0.237174 Objective Loss 0.237174 LR 0.000500 Time 0.025296 +2023-10-05 21:36:17,160 - Epoch: [115][ 210/ 1236] Overall Loss 0.236989 Objective Loss 0.236989 LR 0.000500 Time 0.025159 +2023-10-05 21:36:17,377 - Epoch: [115][ 220/ 1236] Overall Loss 0.235727 Objective Loss 0.235727 LR 0.000500 Time 0.025002 +2023-10-05 21:36:17,596 - Epoch: [115][ 230/ 1236] Overall Loss 0.235965 Objective Loss 0.235965 LR 0.000500 Time 0.024865 +2023-10-05 21:36:17,812 - Epoch: [115][ 240/ 1236] Overall Loss 0.235934 Objective Loss 0.235934 LR 0.000500 Time 0.024727 +2023-10-05 21:36:18,019 - Epoch: [115][ 250/ 1236] Overall Loss 0.235677 Objective Loss 0.235677 LR 0.000500 Time 0.024566 +2023-10-05 21:36:18,229 - Epoch: [115][ 260/ 1236] Overall Loss 0.235452 Objective Loss 0.235452 LR 0.000500 Time 0.024427 +2023-10-05 21:36:18,440 - Epoch: [115][ 270/ 1236] Overall Loss 0.235468 Objective Loss 0.235468 LR 0.000500 Time 0.024301 +2023-10-05 21:36:18,652 - Epoch: [115][ 280/ 1236] Overall Loss 0.237015 Objective Loss 0.237015 LR 0.000500 Time 0.024188 +2023-10-05 21:36:18,857 - Epoch: [115][ 290/ 1236] Overall Loss 0.236665 Objective Loss 0.236665 LR 0.000500 Time 0.024059 +2023-10-05 21:36:19,059 - Epoch: [115][ 300/ 1236] Overall Loss 0.237430 Objective Loss 0.237430 LR 0.000500 Time 0.023930 +2023-10-05 21:36:19,260 - Epoch: [115][ 310/ 1236] Overall Loss 0.236707 Objective Loss 0.236707 LR 0.000500 Time 0.023808 +2023-10-05 21:36:19,464 - Epoch: [115][ 320/ 1236] Overall Loss 0.237595 Objective Loss 0.237595 LR 0.000500 Time 0.023698 +2023-10-05 21:36:19,665 - Epoch: [115][ 330/ 1236] Overall Loss 0.236872 Objective Loss 0.236872 LR 0.000500 Time 0.023588 +2023-10-05 21:36:19,865 - Epoch: [115][ 340/ 1236] Overall Loss 0.236941 Objective Loss 0.236941 LR 0.000500 Time 0.023483 +2023-10-05 21:36:20,066 - Epoch: [115][ 350/ 1236] Overall Loss 0.236849 Objective Loss 0.236849 LR 0.000500 Time 0.023385 +2023-10-05 21:36:20,267 - Epoch: [115][ 360/ 1236] Overall Loss 0.237573 Objective Loss 0.237573 LR 0.000500 Time 0.023295 +2023-10-05 21:36:20,470 - Epoch: [115][ 370/ 1236] Overall Loss 0.238443 Objective Loss 0.238443 LR 0.000500 Time 0.023210 +2023-10-05 21:36:20,671 - Epoch: [115][ 380/ 1236] Overall Loss 0.238440 Objective Loss 0.238440 LR 0.000500 Time 0.023128 +2023-10-05 21:36:20,870 - Epoch: [115][ 390/ 1236] Overall Loss 0.238395 Objective Loss 0.238395 LR 0.000500 Time 0.023045 +2023-10-05 21:36:21,068 - Epoch: [115][ 400/ 1236] Overall Loss 0.238870 Objective Loss 0.238870 LR 0.000500 Time 0.022963 +2023-10-05 21:36:21,267 - Epoch: [115][ 410/ 1236] Overall Loss 0.238434 Objective Loss 0.238434 LR 0.000500 Time 0.022888 +2023-10-05 21:36:21,465 - Epoch: [115][ 420/ 1236] Overall Loss 0.239454 Objective Loss 0.239454 LR 0.000500 Time 0.022813 +2023-10-05 21:36:21,665 - Epoch: [115][ 430/ 1236] Overall Loss 0.239457 Objective Loss 0.239457 LR 0.000500 Time 0.022747 +2023-10-05 21:36:21,862 - Epoch: [115][ 440/ 1236] Overall Loss 0.239631 Objective Loss 0.239631 LR 0.000500 Time 0.022678 +2023-10-05 21:36:22,062 - Epoch: [115][ 450/ 1236] Overall Loss 0.239510 Objective Loss 0.239510 LR 0.000500 Time 0.022617 +2023-10-05 21:36:22,259 - Epoch: [115][ 460/ 1236] Overall Loss 0.239443 Objective Loss 0.239443 LR 0.000500 Time 0.022553 +2023-10-05 21:36:22,458 - Epoch: [115][ 470/ 1236] Overall Loss 0.239902 Objective Loss 0.239902 LR 0.000500 Time 0.022497 +2023-10-05 21:36:22,656 - Epoch: [115][ 480/ 1236] Overall Loss 0.239786 Objective Loss 0.239786 LR 0.000500 Time 0.022440 +2023-10-05 21:36:22,855 - Epoch: [115][ 490/ 1236] Overall Loss 0.240035 Objective Loss 0.240035 LR 0.000500 Time 0.022388 +2023-10-05 21:36:23,053 - Epoch: [115][ 500/ 1236] Overall Loss 0.239811 Objective Loss 0.239811 LR 0.000500 Time 0.022335 +2023-10-05 21:36:23,252 - Epoch: [115][ 510/ 1236] Overall Loss 0.240512 Objective Loss 0.240512 LR 0.000500 Time 0.022288 +2023-10-05 21:36:23,450 - Epoch: [115][ 520/ 1236] Overall Loss 0.239877 Objective Loss 0.239877 LR 0.000500 Time 0.022239 +2023-10-05 21:36:23,650 - Epoch: [115][ 530/ 1236] Overall Loss 0.239652 Objective Loss 0.239652 LR 0.000500 Time 0.022195 +2023-10-05 21:36:23,847 - Epoch: [115][ 540/ 1236] Overall Loss 0.239512 Objective Loss 0.239512 LR 0.000500 Time 0.022149 +2023-10-05 21:36:24,046 - Epoch: [115][ 550/ 1236] Overall Loss 0.239753 Objective Loss 0.239753 LR 0.000500 Time 0.022108 +2023-10-05 21:36:24,244 - Epoch: [115][ 560/ 1236] Overall Loss 0.239610 Objective Loss 0.239610 LR 0.000500 Time 0.022065 +2023-10-05 21:36:24,443 - Epoch: [115][ 570/ 1236] Overall Loss 0.239385 Objective Loss 0.239385 LR 0.000500 Time 0.022027 +2023-10-05 21:36:24,641 - Epoch: [115][ 580/ 1236] Overall Loss 0.239263 Objective Loss 0.239263 LR 0.000500 Time 0.021988 +2023-10-05 21:36:24,840 - Epoch: [115][ 590/ 1236] Overall Loss 0.239812 Objective Loss 0.239812 LR 0.000500 Time 0.021953 +2023-10-05 21:36:25,041 - Epoch: [115][ 600/ 1236] Overall Loss 0.240142 Objective Loss 0.240142 LR 0.000500 Time 0.021922 +2023-10-05 21:36:25,241 - Epoch: [115][ 610/ 1236] Overall Loss 0.240275 Objective Loss 0.240275 LR 0.000500 Time 0.021888 +2023-10-05 21:36:25,437 - Epoch: [115][ 620/ 1236] Overall Loss 0.240465 Objective Loss 0.240465 LR 0.000500 Time 0.021851 +2023-10-05 21:36:25,632 - Epoch: [115][ 630/ 1236] Overall Loss 0.239960 Objective Loss 0.239960 LR 0.000500 Time 0.021815 +2023-10-05 21:36:25,835 - Epoch: [115][ 640/ 1236] Overall Loss 0.239848 Objective Loss 0.239848 LR 0.000500 Time 0.021791 +2023-10-05 21:36:26,032 - Epoch: [115][ 650/ 1236] Overall Loss 0.240016 Objective Loss 0.240016 LR 0.000500 Time 0.021757 +2023-10-05 21:36:26,228 - Epoch: [115][ 660/ 1236] Overall Loss 0.240227 Objective Loss 0.240227 LR 0.000500 Time 0.021724 +2023-10-05 21:36:26,423 - Epoch: [115][ 670/ 1236] Overall Loss 0.240179 Objective Loss 0.240179 LR 0.000500 Time 0.021691 +2023-10-05 21:36:26,619 - Epoch: [115][ 680/ 1236] Overall Loss 0.240242 Objective Loss 0.240242 LR 0.000500 Time 0.021659 +2023-10-05 21:36:26,814 - Epoch: [115][ 690/ 1236] Overall Loss 0.239906 Objective Loss 0.239906 LR 0.000500 Time 0.021629 +2023-10-05 21:36:27,010 - Epoch: [115][ 700/ 1236] Overall Loss 0.239748 Objective Loss 0.239748 LR 0.000500 Time 0.021598 +2023-10-05 21:36:27,205 - Epoch: [115][ 710/ 1236] Overall Loss 0.239595 Objective Loss 0.239595 LR 0.000500 Time 0.021569 +2023-10-05 21:36:27,400 - Epoch: [115][ 720/ 1236] Overall Loss 0.240138 Objective Loss 0.240138 LR 0.000500 Time 0.021540 +2023-10-05 21:36:27,596 - Epoch: [115][ 730/ 1236] Overall Loss 0.239816 Objective Loss 0.239816 LR 0.000500 Time 0.021513 +2023-10-05 21:36:27,792 - Epoch: [115][ 740/ 1236] Overall Loss 0.239984 Objective Loss 0.239984 LR 0.000500 Time 0.021486 +2023-10-05 21:36:27,987 - Epoch: [115][ 750/ 1236] Overall Loss 0.240055 Objective Loss 0.240055 LR 0.000500 Time 0.021460 +2023-10-05 21:36:28,183 - Epoch: [115][ 760/ 1236] Overall Loss 0.239936 Objective Loss 0.239936 LR 0.000500 Time 0.021435 +2023-10-05 21:36:28,379 - Epoch: [115][ 770/ 1236] Overall Loss 0.239729 Objective Loss 0.239729 LR 0.000500 Time 0.021411 +2023-10-05 21:36:28,574 - Epoch: [115][ 780/ 1236] Overall Loss 0.239588 Objective Loss 0.239588 LR 0.000500 Time 0.021387 +2023-10-05 21:36:28,770 - Epoch: [115][ 790/ 1236] Overall Loss 0.239493 Objective Loss 0.239493 LR 0.000500 Time 0.021364 +2023-10-05 21:36:28,966 - Epoch: [115][ 800/ 1236] Overall Loss 0.239551 Objective Loss 0.239551 LR 0.000500 Time 0.021341 +2023-10-05 21:36:29,161 - Epoch: [115][ 810/ 1236] Overall Loss 0.239697 Objective Loss 0.239697 LR 0.000500 Time 0.021319 +2023-10-05 21:36:29,357 - Epoch: [115][ 820/ 1236] Overall Loss 0.239611 Objective Loss 0.239611 LR 0.000500 Time 0.021297 +2023-10-05 21:36:29,552 - Epoch: [115][ 830/ 1236] Overall Loss 0.239439 Objective Loss 0.239439 LR 0.000500 Time 0.021275 +2023-10-05 21:36:29,748 - Epoch: [115][ 840/ 1236] Overall Loss 0.239305 Objective Loss 0.239305 LR 0.000500 Time 0.021255 +2023-10-05 21:36:29,944 - Epoch: [115][ 850/ 1236] Overall Loss 0.239058 Objective Loss 0.239058 LR 0.000500 Time 0.021235 +2023-10-05 21:36:30,139 - Epoch: [115][ 860/ 1236] Overall Loss 0.238870 Objective Loss 0.238870 LR 0.000500 Time 0.021215 +2023-10-05 21:36:30,335 - Epoch: [115][ 870/ 1236] Overall Loss 0.239006 Objective Loss 0.239006 LR 0.000500 Time 0.021196 +2023-10-05 21:36:30,530 - Epoch: [115][ 880/ 1236] Overall Loss 0.239089 Objective Loss 0.239089 LR 0.000500 Time 0.021177 +2023-10-05 21:36:30,726 - Epoch: [115][ 890/ 1236] Overall Loss 0.238881 Objective Loss 0.238881 LR 0.000500 Time 0.021158 +2023-10-05 21:36:30,922 - Epoch: [115][ 900/ 1236] Overall Loss 0.238563 Objective Loss 0.238563 LR 0.000500 Time 0.021141 +2023-10-05 21:36:31,118 - Epoch: [115][ 910/ 1236] Overall Loss 0.238818 Objective Loss 0.238818 LR 0.000500 Time 0.021123 +2023-10-05 21:36:31,314 - Epoch: [115][ 920/ 1236] Overall Loss 0.238749 Objective Loss 0.238749 LR 0.000500 Time 0.021107 +2023-10-05 21:36:31,510 - Epoch: [115][ 930/ 1236] Overall Loss 0.238791 Objective Loss 0.238791 LR 0.000500 Time 0.021090 +2023-10-05 21:36:31,706 - Epoch: [115][ 940/ 1236] Overall Loss 0.238503 Objective Loss 0.238503 LR 0.000500 Time 0.021074 +2023-10-05 21:36:31,902 - Epoch: [115][ 950/ 1236] Overall Loss 0.238722 Objective Loss 0.238722 LR 0.000500 Time 0.021058 +2023-10-05 21:36:32,098 - Epoch: [115][ 960/ 1236] Overall Loss 0.238861 Objective Loss 0.238861 LR 0.000500 Time 0.021043 +2023-10-05 21:36:32,294 - Epoch: [115][ 970/ 1236] Overall Loss 0.238852 Objective Loss 0.238852 LR 0.000500 Time 0.021027 +2023-10-05 21:36:32,489 - Epoch: [115][ 980/ 1236] Overall Loss 0.239013 Objective Loss 0.239013 LR 0.000500 Time 0.021012 +2023-10-05 21:36:32,685 - Epoch: [115][ 990/ 1236] Overall Loss 0.239211 Objective Loss 0.239211 LR 0.000500 Time 0.020998 +2023-10-05 21:36:32,881 - Epoch: [115][ 1000/ 1236] Overall Loss 0.239308 Objective Loss 0.239308 LR 0.000500 Time 0.020984 +2023-10-05 21:36:33,077 - Epoch: [115][ 1010/ 1236] Overall Loss 0.239338 Objective Loss 0.239338 LR 0.000500 Time 0.020970 +2023-10-05 21:36:33,273 - Epoch: [115][ 1020/ 1236] Overall Loss 0.239589 Objective Loss 0.239589 LR 0.000500 Time 0.020956 +2023-10-05 21:36:33,469 - Epoch: [115][ 1030/ 1236] Overall Loss 0.239973 Objective Loss 0.239973 LR 0.000500 Time 0.020942 +2023-10-05 21:36:33,664 - Epoch: [115][ 1040/ 1236] Overall Loss 0.239855 Objective Loss 0.239855 LR 0.000500 Time 0.020929 +2023-10-05 21:36:33,860 - Epoch: [115][ 1050/ 1236] Overall Loss 0.239914 Objective Loss 0.239914 LR 0.000500 Time 0.020916 +2023-10-05 21:36:34,056 - Epoch: [115][ 1060/ 1236] Overall Loss 0.239909 Objective Loss 0.239909 LR 0.000500 Time 0.020903 +2023-10-05 21:36:34,252 - Epoch: [115][ 1070/ 1236] Overall Loss 0.239807 Objective Loss 0.239807 LR 0.000500 Time 0.020890 +2023-10-05 21:36:34,447 - Epoch: [115][ 1080/ 1236] Overall Loss 0.239739 Objective Loss 0.239739 LR 0.000500 Time 0.020878 +2023-10-05 21:36:34,643 - Epoch: [115][ 1090/ 1236] Overall Loss 0.239855 Objective Loss 0.239855 LR 0.000500 Time 0.020866 +2023-10-05 21:36:34,839 - Epoch: [115][ 1100/ 1236] Overall Loss 0.239563 Objective Loss 0.239563 LR 0.000500 Time 0.020854 +2023-10-05 21:36:35,034 - Epoch: [115][ 1110/ 1236] Overall Loss 0.239729 Objective Loss 0.239729 LR 0.000500 Time 0.020842 +2023-10-05 21:36:35,230 - Epoch: [115][ 1120/ 1236] Overall Loss 0.239688 Objective Loss 0.239688 LR 0.000500 Time 0.020830 +2023-10-05 21:36:35,426 - Epoch: [115][ 1130/ 1236] Overall Loss 0.239525 Objective Loss 0.239525 LR 0.000500 Time 0.020819 +2023-10-05 21:36:35,622 - Epoch: [115][ 1140/ 1236] Overall Loss 0.239541 Objective Loss 0.239541 LR 0.000500 Time 0.020808 +2023-10-05 21:36:35,817 - Epoch: [115][ 1150/ 1236] Overall Loss 0.239606 Objective Loss 0.239606 LR 0.000500 Time 0.020797 +2023-10-05 21:36:36,013 - Epoch: [115][ 1160/ 1236] Overall Loss 0.239326 Objective Loss 0.239326 LR 0.000500 Time 0.020786 +2023-10-05 21:36:36,209 - Epoch: [115][ 1170/ 1236] Overall Loss 0.239225 Objective Loss 0.239225 LR 0.000500 Time 0.020776 +2023-10-05 21:36:36,405 - Epoch: [115][ 1180/ 1236] Overall Loss 0.239243 Objective Loss 0.239243 LR 0.000500 Time 0.020765 +2023-10-05 21:36:36,601 - Epoch: [115][ 1190/ 1236] Overall Loss 0.239306 Objective Loss 0.239306 LR 0.000500 Time 0.020755 +2023-10-05 21:36:36,796 - Epoch: [115][ 1200/ 1236] Overall Loss 0.239319 Objective Loss 0.239319 LR 0.000500 Time 0.020745 +2023-10-05 21:36:36,992 - Epoch: [115][ 1210/ 1236] Overall Loss 0.239397 Objective Loss 0.239397 LR 0.000500 Time 0.020735 +2023-10-05 21:36:37,188 - Epoch: [115][ 1220/ 1236] Overall Loss 0.239427 Objective Loss 0.239427 LR 0.000500 Time 0.020726 +2023-10-05 21:36:37,438 - Epoch: [115][ 1230/ 1236] Overall Loss 0.239519 Objective Loss 0.239519 LR 0.000500 Time 0.020760 +2023-10-05 21:36:37,555 - Epoch: [115][ 1236/ 1236] Overall Loss 0.239664 Objective Loss 0.239664 Top1 84.317719 Top5 98.778004 LR 0.000500 Time 0.020754 +2023-10-05 21:36:37,689 - --- validate (epoch=115)----------- +2023-10-05 21:36:37,689 - 29943 samples (256 per mini-batch) +2023-10-05 21:36:38,154 - Epoch: [115][ 10/ 117] Loss 0.343265 Top1 83.789062 Top5 97.812500 +2023-10-05 21:36:38,303 - Epoch: [115][ 20/ 117] Loss 0.330065 Top1 83.730469 Top5 97.968750 +2023-10-05 21:36:38,451 - Epoch: [115][ 30/ 117] Loss 0.329581 Top1 83.750000 Top5 97.929688 +2023-10-05 21:36:38,599 - Epoch: [115][ 40/ 117] Loss 0.332257 Top1 83.925781 Top5 97.910156 +2023-10-05 21:36:38,746 - Epoch: [115][ 50/ 117] Loss 0.329482 Top1 84.015625 Top5 97.890625 +2023-10-05 21:36:38,895 - Epoch: [115][ 60/ 117] Loss 0.327904 Top1 83.977865 Top5 97.936198 +2023-10-05 21:36:39,042 - Epoch: [115][ 70/ 117] Loss 0.325207 Top1 84.040179 Top5 98.035714 +2023-10-05 21:36:39,193 - Epoch: [115][ 80/ 117] Loss 0.327688 Top1 83.881836 Top5 98.007812 +2023-10-05 21:36:39,345 - Epoch: [115][ 90/ 117] Loss 0.325766 Top1 83.953993 Top5 97.981771 +2023-10-05 21:36:39,495 - Epoch: [115][ 100/ 117] Loss 0.326463 Top1 84.019531 Top5 97.988281 +2023-10-05 21:36:39,652 - Epoch: [115][ 110/ 117] Loss 0.325755 Top1 83.991477 Top5 97.954545 +2023-10-05 21:36:39,737 - Epoch: [115][ 117/ 117] Loss 0.329029 Top1 83.922787 Top5 97.912701 +2023-10-05 21:36:39,853 - ==> Top1: 83.923 Top5: 97.913 Loss: 0.329 + +2023-10-05 21:36:39,853 - ==> Confusion: +[[ 919 2 7 1 12 0 0 1 6 67 3 2 2 3 7 2 4 0 0 1 11] + [ 1 1053 1 0 10 21 1 15 1 0 1 1 0 1 0 4 5 0 9 3 4] + [ 5 2 949 10 1 1 28 12 0 0 8 2 7 2 2 1 2 2 6 9 7] + [ 3 0 14 941 1 4 0 2 5 0 14 0 9 5 35 4 2 7 26 1 16] + [ 18 5 2 0 979 3 0 1 0 11 0 2 0 1 8 3 10 1 0 1 5] + [ 3 43 0 1 2 984 0 14 2 0 4 8 0 14 7 2 4 0 2 11 15] + [ 0 7 17 0 0 1 1125 10 0 0 1 1 1 1 1 11 0 0 2 9 4] + [ 3 28 19 0 1 36 8 1048 1 2 4 6 3 2 0 3 0 0 37 8 9] + [ 21 3 0 1 1 1 1 0 970 45 10 3 2 6 9 2 2 0 3 6 3] + [ 99 1 4 0 8 4 1 0 31 921 0 2 0 23 6 6 0 1 0 2 10] + [ 3 4 10 10 1 0 7 3 10 4 962 4 0 15 3 2 2 0 3 1 9] + [ 1 0 2 0 0 7 0 2 0 3 0 951 29 3 0 2 0 17 0 14 4] + [ 1 1 1 3 1 3 1 1 1 0 2 39 970 4 2 6 0 16 0 4 12] + [ 0 0 1 1 3 8 0 0 8 8 6 1 3 1064 4 2 0 1 0 1 8] + [ 12 2 3 7 7 0 0 0 29 3 1 2 1 2 1008 0 0 2 8 0 14] + [ 0 3 2 0 5 1 3 0 0 0 0 7 7 1 1 1068 14 8 0 9 5] + [ 1 14 2 0 4 3 0 2 0 0 0 5 1 1 2 9 1101 0 0 5 11] + [ 0 0 0 1 0 1 2 0 0 1 0 2 20 1 2 6 0 995 1 3 3] + [ 1 10 8 13 2 0 1 20 6 0 1 1 2 0 11 1 3 0 982 0 6] + [ 0 4 5 0 1 6 10 10 0 0 1 14 5 1 0 4 10 1 2 1069 9] + [ 144 185 125 59 122 152 58 100 87 64 169 97 311 308 146 81 170 75 163 219 5070]] + +2023-10-05 21:36:39,855 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:36:39,855 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:36:39,861 - + +2023-10-05 21:36:39,861 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:36:40,852 - Epoch: [116][ 10/ 1236] Overall Loss 0.224219 Objective Loss 0.224219 LR 0.000500 Time 0.099057 +2023-10-05 21:36:41,051 - Epoch: [116][ 20/ 1236] Overall Loss 0.223786 Objective Loss 0.223786 LR 0.000500 Time 0.059461 +2023-10-05 21:36:41,248 - Epoch: [116][ 30/ 1236] Overall Loss 0.228436 Objective Loss 0.228436 LR 0.000500 Time 0.046208 +2023-10-05 21:36:41,448 - Epoch: [116][ 40/ 1236] Overall Loss 0.230930 Objective Loss 0.230930 LR 0.000500 Time 0.039633 +2023-10-05 21:36:41,645 - Epoch: [116][ 50/ 1236] Overall Loss 0.231743 Objective Loss 0.231743 LR 0.000500 Time 0.035645 +2023-10-05 21:36:41,844 - Epoch: [116][ 60/ 1236] Overall Loss 0.234343 Objective Loss 0.234343 LR 0.000500 Time 0.033023 +2023-10-05 21:36:42,042 - Epoch: [116][ 70/ 1236] Overall Loss 0.235517 Objective Loss 0.235517 LR 0.000500 Time 0.031121 +2023-10-05 21:36:42,241 - Epoch: [116][ 80/ 1236] Overall Loss 0.230330 Objective Loss 0.230330 LR 0.000500 Time 0.029717 +2023-10-05 21:36:42,439 - Epoch: [116][ 90/ 1236] Overall Loss 0.230121 Objective Loss 0.230121 LR 0.000500 Time 0.028607 +2023-10-05 21:36:42,638 - Epoch: [116][ 100/ 1236] Overall Loss 0.230757 Objective Loss 0.230757 LR 0.000500 Time 0.027736 +2023-10-05 21:36:42,835 - Epoch: [116][ 110/ 1236] Overall Loss 0.232158 Objective Loss 0.232158 LR 0.000500 Time 0.027006 +2023-10-05 21:36:43,035 - Epoch: [116][ 120/ 1236] Overall Loss 0.232878 Objective Loss 0.232878 LR 0.000500 Time 0.026413 +2023-10-05 21:36:43,232 - Epoch: [116][ 130/ 1236] Overall Loss 0.231556 Objective Loss 0.231556 LR 0.000500 Time 0.025897 +2023-10-05 21:36:43,431 - Epoch: [116][ 140/ 1236] Overall Loss 0.233568 Objective Loss 0.233568 LR 0.000500 Time 0.025467 +2023-10-05 21:36:43,629 - Epoch: [116][ 150/ 1236] Overall Loss 0.233717 Objective Loss 0.233717 LR 0.000500 Time 0.025084 +2023-10-05 21:36:43,828 - Epoch: [116][ 160/ 1236] Overall Loss 0.234035 Objective Loss 0.234035 LR 0.000500 Time 0.024759 +2023-10-05 21:36:44,026 - Epoch: [116][ 170/ 1236] Overall Loss 0.234506 Objective Loss 0.234506 LR 0.000500 Time 0.024467 +2023-10-05 21:36:44,226 - Epoch: [116][ 180/ 1236] Overall Loss 0.235308 Objective Loss 0.235308 LR 0.000500 Time 0.024215 +2023-10-05 21:36:44,424 - Epoch: [116][ 190/ 1236] Overall Loss 0.235193 Objective Loss 0.235193 LR 0.000500 Time 0.023982 +2023-10-05 21:36:44,624 - Epoch: [116][ 200/ 1236] Overall Loss 0.236783 Objective Loss 0.236783 LR 0.000500 Time 0.023781 +2023-10-05 21:36:44,822 - Epoch: [116][ 210/ 1236] Overall Loss 0.235126 Objective Loss 0.235126 LR 0.000500 Time 0.023591 +2023-10-05 21:36:45,022 - Epoch: [116][ 220/ 1236] Overall Loss 0.236889 Objective Loss 0.236889 LR 0.000500 Time 0.023425 +2023-10-05 21:36:45,220 - Epoch: [116][ 230/ 1236] Overall Loss 0.238597 Objective Loss 0.238597 LR 0.000500 Time 0.023268 +2023-10-05 21:36:45,420 - Epoch: [116][ 240/ 1236] Overall Loss 0.239108 Objective Loss 0.239108 LR 0.000500 Time 0.023130 +2023-10-05 21:36:45,619 - Epoch: [116][ 250/ 1236] Overall Loss 0.239573 Objective Loss 0.239573 LR 0.000500 Time 0.022998 +2023-10-05 21:36:45,819 - Epoch: [116][ 260/ 1236] Overall Loss 0.238857 Objective Loss 0.238857 LR 0.000500 Time 0.022882 +2023-10-05 21:36:46,017 - Epoch: [116][ 270/ 1236] Overall Loss 0.238031 Objective Loss 0.238031 LR 0.000500 Time 0.022768 +2023-10-05 21:36:46,217 - Epoch: [116][ 280/ 1236] Overall Loss 0.239063 Objective Loss 0.239063 LR 0.000500 Time 0.022667 +2023-10-05 21:36:46,416 - Epoch: [116][ 290/ 1236] Overall Loss 0.239684 Objective Loss 0.239684 LR 0.000500 Time 0.022568 +2023-10-05 21:36:46,616 - Epoch: [116][ 300/ 1236] Overall Loss 0.239376 Objective Loss 0.239376 LR 0.000500 Time 0.022483 +2023-10-05 21:36:46,815 - Epoch: [116][ 310/ 1236] Overall Loss 0.239683 Objective Loss 0.239683 LR 0.000500 Time 0.022398 +2023-10-05 21:36:47,015 - Epoch: [116][ 320/ 1236] Overall Loss 0.240110 Objective Loss 0.240110 LR 0.000500 Time 0.022322 +2023-10-05 21:36:47,214 - Epoch: [116][ 330/ 1236] Overall Loss 0.239702 Objective Loss 0.239702 LR 0.000500 Time 0.022248 +2023-10-05 21:36:47,421 - Epoch: [116][ 340/ 1236] Overall Loss 0.239553 Objective Loss 0.239553 LR 0.000500 Time 0.022201 +2023-10-05 21:36:47,627 - Epoch: [116][ 350/ 1236] Overall Loss 0.240353 Objective Loss 0.240353 LR 0.000500 Time 0.022153 +2023-10-05 21:36:47,832 - Epoch: [116][ 360/ 1236] Overall Loss 0.239886 Objective Loss 0.239886 LR 0.000500 Time 0.022108 +2023-10-05 21:36:48,035 - Epoch: [116][ 370/ 1236] Overall Loss 0.240555 Objective Loss 0.240555 LR 0.000500 Time 0.022057 +2023-10-05 21:36:48,239 - Epoch: [116][ 380/ 1236] Overall Loss 0.240545 Objective Loss 0.240545 LR 0.000500 Time 0.022012 +2023-10-05 21:36:48,443 - Epoch: [116][ 390/ 1236] Overall Loss 0.241408 Objective Loss 0.241408 LR 0.000500 Time 0.021970 +2023-10-05 21:36:48,647 - Epoch: [116][ 400/ 1236] Overall Loss 0.241758 Objective Loss 0.241758 LR 0.000500 Time 0.021929 +2023-10-05 21:36:48,851 - Epoch: [116][ 410/ 1236] Overall Loss 0.241142 Objective Loss 0.241142 LR 0.000500 Time 0.021888 +2023-10-05 21:36:49,053 - Epoch: [116][ 420/ 1236] Overall Loss 0.240398 Objective Loss 0.240398 LR 0.000500 Time 0.021845 +2023-10-05 21:36:49,252 - Epoch: [116][ 430/ 1236] Overall Loss 0.240139 Objective Loss 0.240139 LR 0.000500 Time 0.021801 +2023-10-05 21:36:49,453 - Epoch: [116][ 440/ 1236] Overall Loss 0.239791 Objective Loss 0.239791 LR 0.000500 Time 0.021761 +2023-10-05 21:36:49,653 - Epoch: [116][ 450/ 1236] Overall Loss 0.239297 Objective Loss 0.239297 LR 0.000500 Time 0.021721 +2023-10-05 21:36:49,854 - Epoch: [116][ 460/ 1236] Overall Loss 0.239824 Objective Loss 0.239824 LR 0.000500 Time 0.021685 +2023-10-05 21:36:50,054 - Epoch: [116][ 470/ 1236] Overall Loss 0.239639 Objective Loss 0.239639 LR 0.000500 Time 0.021648 +2023-10-05 21:36:50,255 - Epoch: [116][ 480/ 1236] Overall Loss 0.239582 Objective Loss 0.239582 LR 0.000500 Time 0.021615 +2023-10-05 21:36:50,455 - Epoch: [116][ 490/ 1236] Overall Loss 0.239698 Objective Loss 0.239698 LR 0.000500 Time 0.021580 +2023-10-05 21:36:50,656 - Epoch: [116][ 500/ 1236] Overall Loss 0.240005 Objective Loss 0.240005 LR 0.000500 Time 0.021551 +2023-10-05 21:36:50,856 - Epoch: [116][ 510/ 1236] Overall Loss 0.240536 Objective Loss 0.240536 LR 0.000500 Time 0.021520 +2023-10-05 21:36:51,057 - Epoch: [116][ 520/ 1236] Overall Loss 0.241021 Objective Loss 0.241021 LR 0.000500 Time 0.021492 +2023-10-05 21:36:51,256 - Epoch: [116][ 530/ 1236] Overall Loss 0.241056 Objective Loss 0.241056 LR 0.000500 Time 0.021462 +2023-10-05 21:36:51,457 - Epoch: [116][ 540/ 1236] Overall Loss 0.241387 Objective Loss 0.241387 LR 0.000500 Time 0.021436 +2023-10-05 21:36:51,657 - Epoch: [116][ 550/ 1236] Overall Loss 0.241011 Objective Loss 0.241011 LR 0.000500 Time 0.021409 +2023-10-05 21:36:51,860 - Epoch: [116][ 560/ 1236] Overall Loss 0.241110 Objective Loss 0.241110 LR 0.000500 Time 0.021388 +2023-10-05 21:36:52,062 - Epoch: [116][ 570/ 1236] Overall Loss 0.241261 Objective Loss 0.241261 LR 0.000500 Time 0.021366 +2023-10-05 21:36:52,266 - Epoch: [116][ 580/ 1236] Overall Loss 0.241230 Objective Loss 0.241230 LR 0.000500 Time 0.021349 +2023-10-05 21:36:52,467 - Epoch: [116][ 590/ 1236] Overall Loss 0.241114 Objective Loss 0.241114 LR 0.000500 Time 0.021328 +2023-10-05 21:36:52,671 - Epoch: [116][ 600/ 1236] Overall Loss 0.241432 Objective Loss 0.241432 LR 0.000500 Time 0.021311 +2023-10-05 21:36:52,872 - Epoch: [116][ 610/ 1236] Overall Loss 0.241367 Objective Loss 0.241367 LR 0.000500 Time 0.021291 +2023-10-05 21:36:53,076 - Epoch: [116][ 620/ 1236] Overall Loss 0.241390 Objective Loss 0.241390 LR 0.000500 Time 0.021276 +2023-10-05 21:36:53,277 - Epoch: [116][ 630/ 1236] Overall Loss 0.241103 Objective Loss 0.241103 LR 0.000500 Time 0.021257 +2023-10-05 21:36:53,481 - Epoch: [116][ 640/ 1236] Overall Loss 0.240691 Objective Loss 0.240691 LR 0.000500 Time 0.021243 +2023-10-05 21:36:53,682 - Epoch: [116][ 650/ 1236] Overall Loss 0.240204 Objective Loss 0.240204 LR 0.000500 Time 0.021225 +2023-10-05 21:36:53,886 - Epoch: [116][ 660/ 1236] Overall Loss 0.240095 Objective Loss 0.240095 LR 0.000500 Time 0.021213 +2023-10-05 21:36:54,088 - Epoch: [116][ 670/ 1236] Overall Loss 0.240031 Objective Loss 0.240031 LR 0.000500 Time 0.021196 +2023-10-05 21:36:54,292 - Epoch: [116][ 680/ 1236] Overall Loss 0.239934 Objective Loss 0.239934 LR 0.000500 Time 0.021184 +2023-10-05 21:36:54,493 - Epoch: [116][ 690/ 1236] Overall Loss 0.239813 Objective Loss 0.239813 LR 0.000500 Time 0.021168 +2023-10-05 21:36:54,697 - Epoch: [116][ 700/ 1236] Overall Loss 0.239937 Objective Loss 0.239937 LR 0.000500 Time 0.021157 +2023-10-05 21:36:54,899 - Epoch: [116][ 710/ 1236] Overall Loss 0.239668 Objective Loss 0.239668 LR 0.000500 Time 0.021142 +2023-10-05 21:36:55,103 - Epoch: [116][ 720/ 1236] Overall Loss 0.239322 Objective Loss 0.239322 LR 0.000500 Time 0.021132 +2023-10-05 21:36:55,304 - Epoch: [116][ 730/ 1236] Overall Loss 0.239481 Objective Loss 0.239481 LR 0.000500 Time 0.021118 +2023-10-05 21:36:55,508 - Epoch: [116][ 740/ 1236] Overall Loss 0.239641 Objective Loss 0.239641 LR 0.000500 Time 0.021107 +2023-10-05 21:36:55,709 - Epoch: [116][ 750/ 1236] Overall Loss 0.239741 Objective Loss 0.239741 LR 0.000500 Time 0.021094 +2023-10-05 21:36:55,913 - Epoch: [116][ 760/ 1236] Overall Loss 0.239623 Objective Loss 0.239623 LR 0.000500 Time 0.021084 +2023-10-05 21:36:56,114 - Epoch: [116][ 770/ 1236] Overall Loss 0.239440 Objective Loss 0.239440 LR 0.000500 Time 0.021071 +2023-10-05 21:36:56,318 - Epoch: [116][ 780/ 1236] Overall Loss 0.239194 Objective Loss 0.239194 LR 0.000500 Time 0.021062 +2023-10-05 21:36:56,520 - Epoch: [116][ 790/ 1236] Overall Loss 0.239565 Objective Loss 0.239565 LR 0.000500 Time 0.021050 +2023-10-05 21:36:56,724 - Epoch: [116][ 800/ 1236] Overall Loss 0.239798 Objective Loss 0.239798 LR 0.000500 Time 0.021041 +2023-10-05 21:36:56,925 - Epoch: [116][ 810/ 1236] Overall Loss 0.239706 Objective Loss 0.239706 LR 0.000500 Time 0.021030 +2023-10-05 21:36:57,129 - Epoch: [116][ 820/ 1236] Overall Loss 0.239670 Objective Loss 0.239670 LR 0.000500 Time 0.021022 +2023-10-05 21:36:57,331 - Epoch: [116][ 830/ 1236] Overall Loss 0.239274 Objective Loss 0.239274 LR 0.000500 Time 0.021011 +2023-10-05 21:36:57,535 - Epoch: [116][ 840/ 1236] Overall Loss 0.239178 Objective Loss 0.239178 LR 0.000500 Time 0.021004 +2023-10-05 21:36:57,736 - Epoch: [116][ 850/ 1236] Overall Loss 0.238991 Objective Loss 0.238991 LR 0.000500 Time 0.020993 +2023-10-05 21:36:57,940 - Epoch: [116][ 860/ 1236] Overall Loss 0.239079 Objective Loss 0.239079 LR 0.000500 Time 0.020985 +2023-10-05 21:36:58,141 - Epoch: [116][ 870/ 1236] Overall Loss 0.239276 Objective Loss 0.239276 LR 0.000500 Time 0.020975 +2023-10-05 21:36:58,345 - Epoch: [116][ 880/ 1236] Overall Loss 0.239257 Objective Loss 0.239257 LR 0.000500 Time 0.020968 +2023-10-05 21:36:58,546 - Epoch: [116][ 890/ 1236] Overall Loss 0.239223 Objective Loss 0.239223 LR 0.000500 Time 0.020958 +2023-10-05 21:36:58,750 - Epoch: [116][ 900/ 1236] Overall Loss 0.239239 Objective Loss 0.239239 LR 0.000500 Time 0.020951 +2023-10-05 21:36:58,951 - Epoch: [116][ 910/ 1236] Overall Loss 0.239185 Objective Loss 0.239185 LR 0.000500 Time 0.020942 +2023-10-05 21:36:59,155 - Epoch: [116][ 920/ 1236] Overall Loss 0.239092 Objective Loss 0.239092 LR 0.000500 Time 0.020935 +2023-10-05 21:36:59,356 - Epoch: [116][ 930/ 1236] Overall Loss 0.239044 Objective Loss 0.239044 LR 0.000500 Time 0.020926 +2023-10-05 21:36:59,560 - Epoch: [116][ 940/ 1236] Overall Loss 0.239350 Objective Loss 0.239350 LR 0.000500 Time 0.020921 +2023-10-05 21:36:59,762 - Epoch: [116][ 950/ 1236] Overall Loss 0.239342 Objective Loss 0.239342 LR 0.000500 Time 0.020912 +2023-10-05 21:36:59,966 - Epoch: [116][ 960/ 1236] Overall Loss 0.239121 Objective Loss 0.239121 LR 0.000500 Time 0.020906 +2023-10-05 21:37:00,167 - Epoch: [116][ 970/ 1236] Overall Loss 0.239158 Objective Loss 0.239158 LR 0.000500 Time 0.020898 +2023-10-05 21:37:00,371 - Epoch: [116][ 980/ 1236] Overall Loss 0.239209 Objective Loss 0.239209 LR 0.000500 Time 0.020892 +2023-10-05 21:37:00,572 - Epoch: [116][ 990/ 1236] Overall Loss 0.238897 Objective Loss 0.238897 LR 0.000500 Time 0.020884 +2023-10-05 21:37:00,777 - Epoch: [116][ 1000/ 1236] Overall Loss 0.238725 Objective Loss 0.238725 LR 0.000500 Time 0.020880 +2023-10-05 21:37:00,977 - Epoch: [116][ 1010/ 1236] Overall Loss 0.238737 Objective Loss 0.238737 LR 0.000500 Time 0.020871 +2023-10-05 21:37:01,178 - Epoch: [116][ 1020/ 1236] Overall Loss 0.238783 Objective Loss 0.238783 LR 0.000500 Time 0.020863 +2023-10-05 21:37:01,378 - Epoch: [116][ 1030/ 1236] Overall Loss 0.238830 Objective Loss 0.238830 LR 0.000500 Time 0.020854 +2023-10-05 21:37:01,579 - Epoch: [116][ 1040/ 1236] Overall Loss 0.238708 Objective Loss 0.238708 LR 0.000500 Time 0.020847 +2023-10-05 21:37:01,780 - Epoch: [116][ 1050/ 1236] Overall Loss 0.238317 Objective Loss 0.238317 LR 0.000500 Time 0.020839 +2023-10-05 21:37:01,981 - Epoch: [116][ 1060/ 1236] Overall Loss 0.238446 Objective Loss 0.238446 LR 0.000500 Time 0.020832 +2023-10-05 21:37:02,181 - Epoch: [116][ 1070/ 1236] Overall Loss 0.238324 Objective Loss 0.238324 LR 0.000500 Time 0.020824 +2023-10-05 21:37:02,383 - Epoch: [116][ 1080/ 1236] Overall Loss 0.238377 Objective Loss 0.238377 LR 0.000500 Time 0.020817 +2023-10-05 21:37:02,583 - Epoch: [116][ 1090/ 1236] Overall Loss 0.238543 Objective Loss 0.238543 LR 0.000500 Time 0.020810 +2023-10-05 21:37:02,785 - Epoch: [116][ 1100/ 1236] Overall Loss 0.238687 Objective Loss 0.238687 LR 0.000500 Time 0.020804 +2023-10-05 21:37:02,985 - Epoch: [116][ 1110/ 1236] Overall Loss 0.238917 Objective Loss 0.238917 LR 0.000500 Time 0.020796 +2023-10-05 21:37:03,186 - Epoch: [116][ 1120/ 1236] Overall Loss 0.238993 Objective Loss 0.238993 LR 0.000500 Time 0.020790 +2023-10-05 21:37:03,386 - Epoch: [116][ 1130/ 1236] Overall Loss 0.238858 Objective Loss 0.238858 LR 0.000500 Time 0.020783 +2023-10-05 21:37:03,588 - Epoch: [116][ 1140/ 1236] Overall Loss 0.239083 Objective Loss 0.239083 LR 0.000500 Time 0.020777 +2023-10-05 21:37:03,788 - Epoch: [116][ 1150/ 1236] Overall Loss 0.239249 Objective Loss 0.239249 LR 0.000500 Time 0.020770 +2023-10-05 21:37:03,990 - Epoch: [116][ 1160/ 1236] Overall Loss 0.239801 Objective Loss 0.239801 LR 0.000500 Time 0.020765 +2023-10-05 21:37:04,190 - Epoch: [116][ 1170/ 1236] Overall Loss 0.239817 Objective Loss 0.239817 LR 0.000500 Time 0.020758 +2023-10-05 21:37:04,392 - Epoch: [116][ 1180/ 1236] Overall Loss 0.239763 Objective Loss 0.239763 LR 0.000500 Time 0.020753 +2023-10-05 21:37:04,592 - Epoch: [116][ 1190/ 1236] Overall Loss 0.240062 Objective Loss 0.240062 LR 0.000500 Time 0.020747 +2023-10-05 21:37:04,794 - Epoch: [116][ 1200/ 1236] Overall Loss 0.240178 Objective Loss 0.240178 LR 0.000500 Time 0.020741 +2023-10-05 21:37:04,994 - Epoch: [116][ 1210/ 1236] Overall Loss 0.240261 Objective Loss 0.240261 LR 0.000500 Time 0.020735 +2023-10-05 21:37:05,196 - Epoch: [116][ 1220/ 1236] Overall Loss 0.240415 Objective Loss 0.240415 LR 0.000500 Time 0.020730 +2023-10-05 21:37:05,448 - Epoch: [116][ 1230/ 1236] Overall Loss 0.240267 Objective Loss 0.240267 LR 0.000500 Time 0.020767 +2023-10-05 21:37:05,565 - Epoch: [116][ 1236/ 1236] Overall Loss 0.240273 Objective Loss 0.240273 Top1 86.965377 Top5 99.185336 LR 0.000500 Time 0.020760 +2023-10-05 21:37:05,704 - --- validate (epoch=116)----------- +2023-10-05 21:37:05,704 - 29943 samples (256 per mini-batch) +2023-10-05 21:37:06,167 - Epoch: [116][ 10/ 117] Loss 0.374215 Top1 82.617188 Top5 97.187500 +2023-10-05 21:37:06,326 - Epoch: [116][ 20/ 117] Loss 0.370578 Top1 82.949219 Top5 97.187500 +2023-10-05 21:37:06,479 - Epoch: [116][ 30/ 117] Loss 0.349245 Top1 83.424479 Top5 97.500000 +2023-10-05 21:37:06,637 - Epoch: [116][ 40/ 117] Loss 0.345995 Top1 83.652344 Top5 97.548828 +2023-10-05 21:37:06,789 - Epoch: [116][ 50/ 117] Loss 0.339887 Top1 83.820312 Top5 97.585938 +2023-10-05 21:37:06,947 - Epoch: [116][ 60/ 117] Loss 0.338154 Top1 83.691406 Top5 97.565104 +2023-10-05 21:37:07,100 - Epoch: [116][ 70/ 117] Loss 0.341316 Top1 83.643973 Top5 97.661830 +2023-10-05 21:37:07,258 - Epoch: [116][ 80/ 117] Loss 0.338858 Top1 83.647461 Top5 97.631836 +2023-10-05 21:37:07,409 - Epoch: [116][ 90/ 117] Loss 0.338017 Top1 83.745660 Top5 97.664931 +2023-10-05 21:37:07,559 - Epoch: [116][ 100/ 117] Loss 0.334952 Top1 83.882812 Top5 97.750000 +2023-10-05 21:37:07,715 - Epoch: [116][ 110/ 117] Loss 0.332821 Top1 83.966619 Top5 97.755682 +2023-10-05 21:37:07,800 - Epoch: [116][ 117/ 117] Loss 0.330997 Top1 84.043015 Top5 97.779114 +2023-10-05 21:37:07,896 - ==> Top1: 84.043 Top5: 97.779 Loss: 0.331 + +2023-10-05 21:37:07,897 - ==> Confusion: +[[ 899 1 6 0 17 2 1 0 4 85 3 2 0 3 8 3 6 0 1 0 9] + [ 1 1042 3 0 11 22 1 20 1 0 1 0 0 0 3 4 5 0 5 4 8] + [ 4 3 955 16 1 2 21 7 0 1 5 2 9 1 2 2 1 2 5 3 14] + [ 1 0 19 944 2 3 2 3 4 2 10 0 8 2 39 3 1 5 27 1 13] + [ 16 3 2 1 974 3 1 2 0 10 1 1 3 1 18 4 4 1 0 1 4] + [ 6 30 1 0 4 985 1 23 4 1 4 3 3 12 9 1 4 0 4 8 13] + [ 0 4 32 0 0 2 1111 8 0 0 1 3 2 0 1 8 0 2 2 10 5] + [ 3 15 22 0 4 30 5 1064 2 3 1 7 5 4 0 2 0 0 34 10 7] + [ 15 1 1 0 3 2 0 0 962 49 13 2 3 10 16 3 2 1 4 2 0] + [ 85 1 2 0 11 6 2 0 16 956 0 2 3 17 9 0 2 0 0 1 6] + [ 1 8 14 6 6 0 4 2 9 4 965 5 0 11 4 1 0 2 2 2 7] + [ 2 1 0 0 3 10 0 1 0 2 0 959 29 2 0 2 0 15 0 8 1] + [ 1 2 2 6 1 5 1 1 1 0 1 34 977 5 3 9 2 8 0 1 8] + [ 3 0 0 1 5 8 0 0 10 19 4 3 4 1044 5 3 1 1 0 1 7] + [ 11 3 4 6 3 0 0 0 20 2 2 1 4 1 1023 0 1 1 8 0 11] + [ 0 3 2 1 6 1 2 0 0 0 0 11 7 1 0 1073 10 6 0 6 5] + [ 0 15 4 0 9 3 0 0 3 0 0 2 3 1 2 11 1091 0 1 3 13] + [ 0 2 0 3 0 0 0 0 1 0 0 2 25 1 3 11 2 983 1 0 4] + [ 3 13 8 12 1 0 0 17 2 1 6 1 4 0 19 0 1 0 971 0 9] + [ 1 4 5 2 1 1 6 12 0 0 2 18 5 1 0 6 10 2 1 1069 6] + [ 123 159 138 65 122 129 42 119 111 100 181 127 328 290 177 77 136 63 126 174 5118]] + +2023-10-05 21:37:07,898 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:37:07,898 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:37:07,904 - + +2023-10-05 21:37:07,904 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:37:08,890 - Epoch: [117][ 10/ 1236] Overall Loss 0.212538 Objective Loss 0.212538 LR 0.000500 Time 0.098558 +2023-10-05 21:37:09,089 - Epoch: [117][ 20/ 1236] Overall Loss 0.227324 Objective Loss 0.227324 LR 0.000500 Time 0.059226 +2023-10-05 21:37:09,288 - Epoch: [117][ 30/ 1236] Overall Loss 0.225763 Objective Loss 0.225763 LR 0.000500 Time 0.046091 +2023-10-05 21:37:09,488 - Epoch: [117][ 40/ 1236] Overall Loss 0.227380 Objective Loss 0.227380 LR 0.000500 Time 0.039554 +2023-10-05 21:37:09,686 - Epoch: [117][ 50/ 1236] Overall Loss 0.225570 Objective Loss 0.225570 LR 0.000500 Time 0.035607 +2023-10-05 21:37:09,886 - Epoch: [117][ 60/ 1236] Overall Loss 0.226490 Objective Loss 0.226490 LR 0.000500 Time 0.033001 +2023-10-05 21:37:10,085 - Epoch: [117][ 70/ 1236] Overall Loss 0.230468 Objective Loss 0.230468 LR 0.000500 Time 0.031114 +2023-10-05 21:37:10,284 - Epoch: [117][ 80/ 1236] Overall Loss 0.230114 Objective Loss 0.230114 LR 0.000500 Time 0.029717 +2023-10-05 21:37:10,483 - Epoch: [117][ 90/ 1236] Overall Loss 0.232442 Objective Loss 0.232442 LR 0.000500 Time 0.028623 +2023-10-05 21:37:10,683 - Epoch: [117][ 100/ 1236] Overall Loss 0.233686 Objective Loss 0.233686 LR 0.000500 Time 0.027758 +2023-10-05 21:37:10,882 - Epoch: [117][ 110/ 1236] Overall Loss 0.235199 Objective Loss 0.235199 LR 0.000500 Time 0.027037 +2023-10-05 21:37:11,082 - Epoch: [117][ 120/ 1236] Overall Loss 0.236744 Objective Loss 0.236744 LR 0.000500 Time 0.026444 +2023-10-05 21:37:11,280 - Epoch: [117][ 130/ 1236] Overall Loss 0.238055 Objective Loss 0.238055 LR 0.000500 Time 0.025934 +2023-10-05 21:37:11,480 - Epoch: [117][ 140/ 1236] Overall Loss 0.236713 Objective Loss 0.236713 LR 0.000500 Time 0.025505 +2023-10-05 21:37:11,677 - Epoch: [117][ 150/ 1236] Overall Loss 0.236311 Objective Loss 0.236311 LR 0.000500 Time 0.025119 +2023-10-05 21:37:11,877 - Epoch: [117][ 160/ 1236] Overall Loss 0.235755 Objective Loss 0.235755 LR 0.000500 Time 0.024794 +2023-10-05 21:37:12,075 - Epoch: [117][ 170/ 1236] Overall Loss 0.236237 Objective Loss 0.236237 LR 0.000500 Time 0.024501 +2023-10-05 21:37:12,275 - Epoch: [117][ 180/ 1236] Overall Loss 0.234581 Objective Loss 0.234581 LR 0.000500 Time 0.024246 +2023-10-05 21:37:12,473 - Epoch: [117][ 190/ 1236] Overall Loss 0.233846 Objective Loss 0.233846 LR 0.000500 Time 0.024010 +2023-10-05 21:37:12,675 - Epoch: [117][ 200/ 1236] Overall Loss 0.234405 Objective Loss 0.234405 LR 0.000500 Time 0.023821 +2023-10-05 21:37:12,878 - Epoch: [117][ 210/ 1236] Overall Loss 0.234319 Objective Loss 0.234319 LR 0.000500 Time 0.023651 +2023-10-05 21:37:13,080 - Epoch: [117][ 220/ 1236] Overall Loss 0.235330 Objective Loss 0.235330 LR 0.000500 Time 0.023491 +2023-10-05 21:37:13,283 - Epoch: [117][ 230/ 1236] Overall Loss 0.235410 Objective Loss 0.235410 LR 0.000500 Time 0.023351 +2023-10-05 21:37:13,485 - Epoch: [117][ 240/ 1236] Overall Loss 0.235413 Objective Loss 0.235413 LR 0.000500 Time 0.023219 +2023-10-05 21:37:13,688 - Epoch: [117][ 250/ 1236] Overall Loss 0.235851 Objective Loss 0.235851 LR 0.000500 Time 0.023100 +2023-10-05 21:37:13,890 - Epoch: [117][ 260/ 1236] Overall Loss 0.236070 Objective Loss 0.236070 LR 0.000500 Time 0.022988 +2023-10-05 21:37:14,093 - Epoch: [117][ 270/ 1236] Overall Loss 0.235641 Objective Loss 0.235641 LR 0.000500 Time 0.022887 +2023-10-05 21:37:14,295 - Epoch: [117][ 280/ 1236] Overall Loss 0.235595 Objective Loss 0.235595 LR 0.000500 Time 0.022789 +2023-10-05 21:37:14,498 - Epoch: [117][ 290/ 1236] Overall Loss 0.235602 Objective Loss 0.235602 LR 0.000500 Time 0.022702 +2023-10-05 21:37:14,700 - Epoch: [117][ 300/ 1236] Overall Loss 0.235376 Objective Loss 0.235376 LR 0.000500 Time 0.022618 +2023-10-05 21:37:14,903 - Epoch: [117][ 310/ 1236] Overall Loss 0.234563 Objective Loss 0.234563 LR 0.000500 Time 0.022543 +2023-10-05 21:37:15,105 - Epoch: [117][ 320/ 1236] Overall Loss 0.235408 Objective Loss 0.235408 LR 0.000500 Time 0.022468 +2023-10-05 21:37:15,308 - Epoch: [117][ 330/ 1236] Overall Loss 0.235651 Objective Loss 0.235651 LR 0.000500 Time 0.022401 +2023-10-05 21:37:15,510 - Epoch: [117][ 340/ 1236] Overall Loss 0.236052 Objective Loss 0.236052 LR 0.000500 Time 0.022336 +2023-10-05 21:37:15,714 - Epoch: [117][ 350/ 1236] Overall Loss 0.234537 Objective Loss 0.234537 LR 0.000500 Time 0.022279 +2023-10-05 21:37:15,916 - Epoch: [117][ 360/ 1236] Overall Loss 0.234598 Objective Loss 0.234598 LR 0.000500 Time 0.022219 +2023-10-05 21:37:16,120 - Epoch: [117][ 370/ 1236] Overall Loss 0.234299 Objective Loss 0.234299 LR 0.000500 Time 0.022170 +2023-10-05 21:37:16,322 - Epoch: [117][ 380/ 1236] Overall Loss 0.234666 Objective Loss 0.234666 LR 0.000500 Time 0.022118 +2023-10-05 21:37:16,526 - Epoch: [117][ 390/ 1236] Overall Loss 0.234821 Objective Loss 0.234821 LR 0.000500 Time 0.022073 +2023-10-05 21:37:16,728 - Epoch: [117][ 400/ 1236] Overall Loss 0.234756 Objective Loss 0.234756 LR 0.000500 Time 0.022026 +2023-10-05 21:37:16,933 - Epoch: [117][ 410/ 1236] Overall Loss 0.235148 Objective Loss 0.235148 LR 0.000500 Time 0.021986 +2023-10-05 21:37:17,135 - Epoch: [117][ 420/ 1236] Overall Loss 0.234727 Objective Loss 0.234727 LR 0.000500 Time 0.021944 +2023-10-05 21:37:17,340 - Epoch: [117][ 430/ 1236] Overall Loss 0.235316 Objective Loss 0.235316 LR 0.000500 Time 0.021908 +2023-10-05 21:37:17,542 - Epoch: [117][ 440/ 1236] Overall Loss 0.235279 Objective Loss 0.235279 LR 0.000500 Time 0.021869 +2023-10-05 21:37:17,746 - Epoch: [117][ 450/ 1236] Overall Loss 0.235687 Objective Loss 0.235687 LR 0.000500 Time 0.021836 +2023-10-05 21:37:17,949 - Epoch: [117][ 460/ 1236] Overall Loss 0.236417 Objective Loss 0.236417 LR 0.000500 Time 0.021801 +2023-10-05 21:37:18,154 - Epoch: [117][ 470/ 1236] Overall Loss 0.236491 Objective Loss 0.236491 LR 0.000500 Time 0.021772 +2023-10-05 21:37:18,356 - Epoch: [117][ 480/ 1236] Overall Loss 0.236221 Objective Loss 0.236221 LR 0.000500 Time 0.021740 +2023-10-05 21:37:18,561 - Epoch: [117][ 490/ 1236] Overall Loss 0.235909 Objective Loss 0.235909 LR 0.000500 Time 0.021713 +2023-10-05 21:37:18,763 - Epoch: [117][ 500/ 1236] Overall Loss 0.235992 Objective Loss 0.235992 LR 0.000500 Time 0.021682 +2023-10-05 21:37:18,968 - Epoch: [117][ 510/ 1236] Overall Loss 0.235928 Objective Loss 0.235928 LR 0.000500 Time 0.021658 +2023-10-05 21:37:19,170 - Epoch: [117][ 520/ 1236] Overall Loss 0.235912 Objective Loss 0.235912 LR 0.000500 Time 0.021630 +2023-10-05 21:37:19,375 - Epoch: [117][ 530/ 1236] Overall Loss 0.235883 Objective Loss 0.235883 LR 0.000500 Time 0.021607 +2023-10-05 21:37:19,577 - Epoch: [117][ 540/ 1236] Overall Loss 0.235886 Objective Loss 0.235886 LR 0.000500 Time 0.021580 +2023-10-05 21:37:19,781 - Epoch: [117][ 550/ 1236] Overall Loss 0.235934 Objective Loss 0.235934 LR 0.000500 Time 0.021559 +2023-10-05 21:37:19,984 - Epoch: [117][ 560/ 1236] Overall Loss 0.236041 Objective Loss 0.236041 LR 0.000500 Time 0.021535 +2023-10-05 21:37:20,188 - Epoch: [117][ 570/ 1236] Overall Loss 0.235937 Objective Loss 0.235937 LR 0.000500 Time 0.021515 +2023-10-05 21:37:20,391 - Epoch: [117][ 580/ 1236] Overall Loss 0.236067 Objective Loss 0.236067 LR 0.000500 Time 0.021493 +2023-10-05 21:37:20,595 - Epoch: [117][ 590/ 1236] Overall Loss 0.236808 Objective Loss 0.236808 LR 0.000500 Time 0.021475 +2023-10-05 21:37:20,798 - Epoch: [117][ 600/ 1236] Overall Loss 0.237269 Objective Loss 0.237269 LR 0.000500 Time 0.021454 +2023-10-05 21:37:21,002 - Epoch: [117][ 610/ 1236] Overall Loss 0.237527 Objective Loss 0.237527 LR 0.000500 Time 0.021437 +2023-10-05 21:37:21,205 - Epoch: [117][ 620/ 1236] Overall Loss 0.237550 Objective Loss 0.237550 LR 0.000500 Time 0.021418 +2023-10-05 21:37:21,410 - Epoch: [117][ 630/ 1236] Overall Loss 0.237454 Objective Loss 0.237454 LR 0.000500 Time 0.021402 +2023-10-05 21:37:21,612 - Epoch: [117][ 640/ 1236] Overall Loss 0.237444 Objective Loss 0.237444 LR 0.000500 Time 0.021383 +2023-10-05 21:37:21,817 - Epoch: [117][ 650/ 1236] Overall Loss 0.237700 Objective Loss 0.237700 LR 0.000500 Time 0.021369 +2023-10-05 21:37:22,021 - Epoch: [117][ 660/ 1236] Overall Loss 0.237333 Objective Loss 0.237333 LR 0.000500 Time 0.021353 +2023-10-05 21:37:22,226 - Epoch: [117][ 670/ 1236] Overall Loss 0.237180 Objective Loss 0.237180 LR 0.000500 Time 0.021340 +2023-10-05 21:37:22,429 - Epoch: [117][ 680/ 1236] Overall Loss 0.237359 Objective Loss 0.237359 LR 0.000500 Time 0.021324 +2023-10-05 21:37:22,634 - Epoch: [117][ 690/ 1236] Overall Loss 0.237967 Objective Loss 0.237967 LR 0.000500 Time 0.021312 +2023-10-05 21:37:22,837 - Epoch: [117][ 700/ 1236] Overall Loss 0.238103 Objective Loss 0.238103 LR 0.000500 Time 0.021296 +2023-10-05 21:37:23,042 - Epoch: [117][ 710/ 1236] Overall Loss 0.238184 Objective Loss 0.238184 LR 0.000500 Time 0.021286 +2023-10-05 21:37:23,245 - Epoch: [117][ 720/ 1236] Overall Loss 0.237902 Objective Loss 0.237902 LR 0.000500 Time 0.021271 +2023-10-05 21:37:23,450 - Epoch: [117][ 730/ 1236] Overall Loss 0.238419 Objective Loss 0.238419 LR 0.000500 Time 0.021260 +2023-10-05 21:37:23,654 - Epoch: [117][ 740/ 1236] Overall Loss 0.237894 Objective Loss 0.237894 LR 0.000500 Time 0.021247 +2023-10-05 21:37:23,859 - Epoch: [117][ 750/ 1236] Overall Loss 0.237665 Objective Loss 0.237665 LR 0.000500 Time 0.021237 +2023-10-05 21:37:24,062 - Epoch: [117][ 760/ 1236] Overall Loss 0.237472 Objective Loss 0.237472 LR 0.000500 Time 0.021225 +2023-10-05 21:37:24,268 - Epoch: [117][ 770/ 1236] Overall Loss 0.237112 Objective Loss 0.237112 LR 0.000500 Time 0.021215 +2023-10-05 21:37:24,471 - Epoch: [117][ 780/ 1236] Overall Loss 0.237236 Objective Loss 0.237236 LR 0.000500 Time 0.021203 +2023-10-05 21:37:24,676 - Epoch: [117][ 790/ 1236] Overall Loss 0.237025 Objective Loss 0.237025 LR 0.000500 Time 0.021194 +2023-10-05 21:37:24,879 - Epoch: [117][ 800/ 1236] Overall Loss 0.237068 Objective Loss 0.237068 LR 0.000500 Time 0.021183 +2023-10-05 21:37:25,084 - Epoch: [117][ 810/ 1236] Overall Loss 0.237302 Objective Loss 0.237302 LR 0.000500 Time 0.021174 +2023-10-05 21:37:25,287 - Epoch: [117][ 820/ 1236] Overall Loss 0.237339 Objective Loss 0.237339 LR 0.000500 Time 0.021162 +2023-10-05 21:37:25,492 - Epoch: [117][ 830/ 1236] Overall Loss 0.237428 Objective Loss 0.237428 LR 0.000500 Time 0.021154 +2023-10-05 21:37:25,695 - Epoch: [117][ 840/ 1236] Overall Loss 0.237380 Objective Loss 0.237380 LR 0.000500 Time 0.021143 +2023-10-05 21:37:25,900 - Epoch: [117][ 850/ 1236] Overall Loss 0.237756 Objective Loss 0.237756 LR 0.000500 Time 0.021136 +2023-10-05 21:37:26,103 - Epoch: [117][ 860/ 1236] Overall Loss 0.237901 Objective Loss 0.237901 LR 0.000500 Time 0.021126 +2023-10-05 21:37:26,309 - Epoch: [117][ 870/ 1236] Overall Loss 0.237591 Objective Loss 0.237591 LR 0.000500 Time 0.021119 +2023-10-05 21:37:26,512 - Epoch: [117][ 880/ 1236] Overall Loss 0.237690 Objective Loss 0.237690 LR 0.000500 Time 0.021109 +2023-10-05 21:37:26,717 - Epoch: [117][ 890/ 1236] Overall Loss 0.237838 Objective Loss 0.237838 LR 0.000500 Time 0.021102 +2023-10-05 21:37:26,920 - Epoch: [117][ 900/ 1236] Overall Loss 0.237854 Objective Loss 0.237854 LR 0.000500 Time 0.021093 +2023-10-05 21:37:27,126 - Epoch: [117][ 910/ 1236] Overall Loss 0.237821 Objective Loss 0.237821 LR 0.000500 Time 0.021087 +2023-10-05 21:37:27,328 - Epoch: [117][ 920/ 1236] Overall Loss 0.237817 Objective Loss 0.237817 LR 0.000500 Time 0.021077 +2023-10-05 21:37:27,534 - Epoch: [117][ 930/ 1236] Overall Loss 0.238054 Objective Loss 0.238054 LR 0.000500 Time 0.021071 +2023-10-05 21:37:27,737 - Epoch: [117][ 940/ 1236] Overall Loss 0.238167 Objective Loss 0.238167 LR 0.000500 Time 0.021063 +2023-10-05 21:37:27,942 - Epoch: [117][ 950/ 1236] Overall Loss 0.237952 Objective Loss 0.237952 LR 0.000500 Time 0.021057 +2023-10-05 21:37:28,149 - Epoch: [117][ 960/ 1236] Overall Loss 0.238023 Objective Loss 0.238023 LR 0.000500 Time 0.021052 +2023-10-05 21:37:28,364 - Epoch: [117][ 970/ 1236] Overall Loss 0.238051 Objective Loss 0.238051 LR 0.000500 Time 0.021057 +2023-10-05 21:37:28,574 - Epoch: [117][ 980/ 1236] Overall Loss 0.237901 Objective Loss 0.237901 LR 0.000500 Time 0.021055 +2023-10-05 21:37:28,790 - Epoch: [117][ 990/ 1236] Overall Loss 0.238331 Objective Loss 0.238331 LR 0.000500 Time 0.021061 +2023-10-05 21:37:28,992 - Epoch: [117][ 1000/ 1236] Overall Loss 0.238644 Objective Loss 0.238644 LR 0.000500 Time 0.021053 +2023-10-05 21:37:29,198 - Epoch: [117][ 1010/ 1236] Overall Loss 0.238526 Objective Loss 0.238526 LR 0.000500 Time 0.021047 +2023-10-05 21:37:29,401 - Epoch: [117][ 1020/ 1236] Overall Loss 0.238447 Objective Loss 0.238447 LR 0.000500 Time 0.021039 +2023-10-05 21:37:29,606 - Epoch: [117][ 1030/ 1236] Overall Loss 0.238465 Objective Loss 0.238465 LR 0.000500 Time 0.021034 +2023-10-05 21:37:29,808 - Epoch: [117][ 1040/ 1236] Overall Loss 0.238254 Objective Loss 0.238254 LR 0.000500 Time 0.021026 +2023-10-05 21:37:30,013 - Epoch: [117][ 1050/ 1236] Overall Loss 0.238210 Objective Loss 0.238210 LR 0.000500 Time 0.021020 +2023-10-05 21:37:30,216 - Epoch: [117][ 1060/ 1236] Overall Loss 0.238277 Objective Loss 0.238277 LR 0.000500 Time 0.021013 +2023-10-05 21:37:30,421 - Epoch: [117][ 1070/ 1236] Overall Loss 0.238250 Objective Loss 0.238250 LR 0.000500 Time 0.021008 +2023-10-05 21:37:30,624 - Epoch: [117][ 1080/ 1236] Overall Loss 0.238377 Objective Loss 0.238377 LR 0.000500 Time 0.021001 +2023-10-05 21:37:30,828 - Epoch: [117][ 1090/ 1236] Overall Loss 0.238111 Objective Loss 0.238111 LR 0.000500 Time 0.020995 +2023-10-05 21:37:31,031 - Epoch: [117][ 1100/ 1236] Overall Loss 0.238290 Objective Loss 0.238290 LR 0.000500 Time 0.020988 +2023-10-05 21:37:31,236 - Epoch: [117][ 1110/ 1236] Overall Loss 0.238276 Objective Loss 0.238276 LR 0.000500 Time 0.020983 +2023-10-05 21:37:31,439 - Epoch: [117][ 1120/ 1236] Overall Loss 0.238195 Objective Loss 0.238195 LR 0.000500 Time 0.020977 +2023-10-05 21:37:31,643 - Epoch: [117][ 1130/ 1236] Overall Loss 0.238096 Objective Loss 0.238096 LR 0.000500 Time 0.020972 +2023-10-05 21:37:31,846 - Epoch: [117][ 1140/ 1236] Overall Loss 0.237922 Objective Loss 0.237922 LR 0.000500 Time 0.020966 +2023-10-05 21:37:32,051 - Epoch: [117][ 1150/ 1236] Overall Loss 0.237821 Objective Loss 0.237821 LR 0.000500 Time 0.020962 +2023-10-05 21:37:32,254 - Epoch: [117][ 1160/ 1236] Overall Loss 0.237657 Objective Loss 0.237657 LR 0.000500 Time 0.020955 +2023-10-05 21:37:32,459 - Epoch: [117][ 1170/ 1236] Overall Loss 0.237746 Objective Loss 0.237746 LR 0.000500 Time 0.020951 +2023-10-05 21:37:32,662 - Epoch: [117][ 1180/ 1236] Overall Loss 0.237630 Objective Loss 0.237630 LR 0.000500 Time 0.020945 +2023-10-05 21:37:32,867 - Epoch: [117][ 1190/ 1236] Overall Loss 0.237628 Objective Loss 0.237628 LR 0.000500 Time 0.020941 +2023-10-05 21:37:33,070 - Epoch: [117][ 1200/ 1236] Overall Loss 0.237527 Objective Loss 0.237527 LR 0.000500 Time 0.020935 +2023-10-05 21:37:33,275 - Epoch: [117][ 1210/ 1236] Overall Loss 0.237728 Objective Loss 0.237728 LR 0.000500 Time 0.020931 +2023-10-05 21:37:33,477 - Epoch: [117][ 1220/ 1236] Overall Loss 0.237546 Objective Loss 0.237546 LR 0.000500 Time 0.020926 +2023-10-05 21:37:33,736 - Epoch: [117][ 1230/ 1236] Overall Loss 0.237412 Objective Loss 0.237412 LR 0.000500 Time 0.020965 +2023-10-05 21:37:33,854 - Epoch: [117][ 1236/ 1236] Overall Loss 0.237368 Objective Loss 0.237368 Top1 85.743381 Top5 97.963340 LR 0.000500 Time 0.020959 +2023-10-05 21:37:33,991 - --- validate (epoch=117)----------- +2023-10-05 21:37:33,991 - 29943 samples (256 per mini-batch) +2023-10-05 21:37:34,455 - Epoch: [117][ 10/ 117] Loss 0.312312 Top1 85.195312 Top5 98.203125 +2023-10-05 21:37:34,609 - Epoch: [117][ 20/ 117] Loss 0.309244 Top1 85.136719 Top5 98.007812 +2023-10-05 21:37:34,761 - Epoch: [117][ 30/ 117] Loss 0.323932 Top1 84.648438 Top5 97.734375 +2023-10-05 21:37:34,915 - Epoch: [117][ 40/ 117] Loss 0.322159 Top1 84.277344 Top5 97.714844 +2023-10-05 21:37:35,067 - Epoch: [117][ 50/ 117] Loss 0.314231 Top1 84.421875 Top5 97.789062 +2023-10-05 21:37:35,219 - Epoch: [117][ 60/ 117] Loss 0.317490 Top1 84.114583 Top5 97.812500 +2023-10-05 21:37:35,368 - Epoch: [117][ 70/ 117] Loss 0.321839 Top1 84.068080 Top5 97.823661 +2023-10-05 21:37:35,521 - Epoch: [117][ 80/ 117] Loss 0.323705 Top1 83.979492 Top5 97.822266 +2023-10-05 21:37:35,675 - Epoch: [117][ 90/ 117] Loss 0.325018 Top1 83.810764 Top5 97.777778 +2023-10-05 21:37:35,829 - Epoch: [117][ 100/ 117] Loss 0.326987 Top1 83.730469 Top5 97.812500 +2023-10-05 21:37:35,983 - Epoch: [117][ 110/ 117] Loss 0.326872 Top1 83.767756 Top5 97.840909 +2023-10-05 21:37:36,069 - Epoch: [117][ 117/ 117] Loss 0.329243 Top1 83.762482 Top5 97.839228 +2023-10-05 21:37:36,211 - ==> Top1: 83.762 Top5: 97.839 Loss: 0.329 + +2023-10-05 21:37:36,212 - ==> Confusion: +[[ 929 3 9 2 7 3 0 0 5 61 0 1 0 2 8 2 4 2 2 0 10] + [ 2 1053 4 0 8 20 1 15 3 0 1 1 0 0 2 2 2 0 7 2 8] + [ 2 2 951 14 1 0 34 8 0 0 5 2 9 0 2 2 2 0 7 7 8] + [ 1 0 16 968 0 3 4 1 1 1 6 0 8 3 32 3 2 8 18 2 12] + [ 30 9 1 0 951 4 0 0 0 9 1 3 0 2 14 6 11 1 0 1 7] + [ 4 46 0 3 3 952 2 28 4 0 5 9 2 18 5 1 5 0 3 9 17] + [ 0 5 23 1 0 0 1131 6 0 0 4 2 1 0 1 5 0 1 1 5 5] + [ 4 29 21 0 2 30 5 1037 2 4 5 6 2 0 0 2 0 0 45 13 11] + [ 18 1 0 0 1 1 1 0 970 39 12 2 1 13 14 5 1 1 7 0 2] + [ 117 0 2 0 5 5 1 0 37 907 1 1 0 23 4 4 1 0 1 4 6] + [ 5 5 10 6 4 1 5 4 8 1 962 0 2 11 5 1 1 2 5 1 14] + [ 2 1 3 0 1 10 0 3 0 0 0 941 34 5 0 3 0 19 0 9 4] + [ 0 0 3 5 1 2 2 3 0 1 1 32 978 2 3 3 4 16 3 2 7] + [ 2 1 1 1 2 2 2 0 12 12 9 3 5 1047 3 3 2 1 0 2 9] + [ 16 3 3 8 4 0 0 0 23 5 0 2 1 1 1013 0 1 2 10 0 9] + [ 1 1 3 1 3 0 0 0 0 0 1 9 8 0 0 1069 13 10 2 8 5] + [ 0 17 2 0 5 3 0 1 1 0 0 3 2 0 2 9 1102 0 2 3 9] + [ 0 0 0 0 0 0 1 0 0 0 0 3 17 1 1 6 0 1004 1 2 2] + [ 2 10 7 16 0 0 0 15 2 0 1 1 1 0 7 0 1 1 995 2 7] + [ 0 3 3 0 1 3 12 7 1 0 1 16 7 3 0 6 10 1 2 1067 9] + [ 132 202 139 88 72 120 56 87 102 78 180 95 358 282 170 73 179 91 144 203 5054]] + +2023-10-05 21:37:36,213 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:37:36,213 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:37:36,219 - + +2023-10-05 21:37:36,219 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:37:37,352 - Epoch: [118][ 10/ 1236] Overall Loss 0.258902 Objective Loss 0.258902 LR 0.000500 Time 0.113260 +2023-10-05 21:37:37,557 - Epoch: [118][ 20/ 1236] Overall Loss 0.245284 Objective Loss 0.245284 LR 0.000500 Time 0.066842 +2023-10-05 21:37:37,759 - Epoch: [118][ 30/ 1236] Overall Loss 0.236414 Objective Loss 0.236414 LR 0.000500 Time 0.051287 +2023-10-05 21:37:37,964 - Epoch: [118][ 40/ 1236] Overall Loss 0.242331 Objective Loss 0.242331 LR 0.000500 Time 0.043572 +2023-10-05 21:37:38,166 - Epoch: [118][ 50/ 1236] Overall Loss 0.234929 Objective Loss 0.234929 LR 0.000500 Time 0.038896 +2023-10-05 21:37:38,371 - Epoch: [118][ 60/ 1236] Overall Loss 0.234288 Objective Loss 0.234288 LR 0.000500 Time 0.035819 +2023-10-05 21:37:38,573 - Epoch: [118][ 70/ 1236] Overall Loss 0.230310 Objective Loss 0.230310 LR 0.000500 Time 0.033591 +2023-10-05 21:37:38,778 - Epoch: [118][ 80/ 1236] Overall Loss 0.229690 Objective Loss 0.229690 LR 0.000500 Time 0.031945 +2023-10-05 21:37:38,981 - Epoch: [118][ 90/ 1236] Overall Loss 0.228950 Objective Loss 0.228950 LR 0.000500 Time 0.030642 +2023-10-05 21:37:39,186 - Epoch: [118][ 100/ 1236] Overall Loss 0.230147 Objective Loss 0.230147 LR 0.000500 Time 0.029626 +2023-10-05 21:37:39,389 - Epoch: [118][ 110/ 1236] Overall Loss 0.232280 Objective Loss 0.232280 LR 0.000500 Time 0.028778 +2023-10-05 21:37:39,596 - Epoch: [118][ 120/ 1236] Overall Loss 0.233449 Objective Loss 0.233449 LR 0.000500 Time 0.028098 +2023-10-05 21:37:39,801 - Epoch: [118][ 130/ 1236] Overall Loss 0.233566 Objective Loss 0.233566 LR 0.000500 Time 0.027512 +2023-10-05 21:37:40,007 - Epoch: [118][ 140/ 1236] Overall Loss 0.235009 Objective Loss 0.235009 LR 0.000500 Time 0.027016 +2023-10-05 21:37:40,211 - Epoch: [118][ 150/ 1236] Overall Loss 0.237060 Objective Loss 0.237060 LR 0.000500 Time 0.026574 +2023-10-05 21:37:40,418 - Epoch: [118][ 160/ 1236] Overall Loss 0.237114 Objective Loss 0.237114 LR 0.000500 Time 0.026204 +2023-10-05 21:37:40,623 - Epoch: [118][ 170/ 1236] Overall Loss 0.237392 Objective Loss 0.237392 LR 0.000500 Time 0.025867 +2023-10-05 21:37:40,830 - Epoch: [118][ 180/ 1236] Overall Loss 0.237496 Objective Loss 0.237496 LR 0.000500 Time 0.025576 +2023-10-05 21:37:41,035 - Epoch: [118][ 190/ 1236] Overall Loss 0.237203 Objective Loss 0.237203 LR 0.000500 Time 0.025308 +2023-10-05 21:37:41,242 - Epoch: [118][ 200/ 1236] Overall Loss 0.237424 Objective Loss 0.237424 LR 0.000500 Time 0.025075 +2023-10-05 21:37:41,447 - Epoch: [118][ 210/ 1236] Overall Loss 0.237611 Objective Loss 0.237611 LR 0.000500 Time 0.024856 +2023-10-05 21:37:41,653 - Epoch: [118][ 220/ 1236] Overall Loss 0.237206 Objective Loss 0.237206 LR 0.000500 Time 0.024664 +2023-10-05 21:37:41,858 - Epoch: [118][ 230/ 1236] Overall Loss 0.237055 Objective Loss 0.237055 LR 0.000500 Time 0.024480 +2023-10-05 21:37:42,065 - Epoch: [118][ 240/ 1236] Overall Loss 0.236919 Objective Loss 0.236919 LR 0.000500 Time 0.024319 +2023-10-05 21:37:42,270 - Epoch: [118][ 250/ 1236] Overall Loss 0.237029 Objective Loss 0.237029 LR 0.000500 Time 0.024166 +2023-10-05 21:37:42,477 - Epoch: [118][ 260/ 1236] Overall Loss 0.236910 Objective Loss 0.236910 LR 0.000500 Time 0.024030 +2023-10-05 21:37:42,681 - Epoch: [118][ 270/ 1236] Overall Loss 0.237852 Objective Loss 0.237852 LR 0.000500 Time 0.023896 +2023-10-05 21:37:42,888 - Epoch: [118][ 280/ 1236] Overall Loss 0.238184 Objective Loss 0.238184 LR 0.000500 Time 0.023779 +2023-10-05 21:37:43,093 - Epoch: [118][ 290/ 1236] Overall Loss 0.238653 Objective Loss 0.238653 LR 0.000500 Time 0.023666 +2023-10-05 21:37:43,299 - Epoch: [118][ 300/ 1236] Overall Loss 0.238238 Objective Loss 0.238238 LR 0.000500 Time 0.023561 +2023-10-05 21:37:43,501 - Epoch: [118][ 310/ 1236] Overall Loss 0.238152 Objective Loss 0.238152 LR 0.000500 Time 0.023453 +2023-10-05 21:37:43,704 - Epoch: [118][ 320/ 1236] Overall Loss 0.238826 Objective Loss 0.238826 LR 0.000500 Time 0.023352 +2023-10-05 21:37:43,906 - Epoch: [118][ 330/ 1236] Overall Loss 0.239065 Objective Loss 0.239065 LR 0.000500 Time 0.023257 +2023-10-05 21:37:44,110 - Epoch: [118][ 340/ 1236] Overall Loss 0.239778 Objective Loss 0.239778 LR 0.000500 Time 0.023170 +2023-10-05 21:37:44,312 - Epoch: [118][ 350/ 1236] Overall Loss 0.238886 Objective Loss 0.238886 LR 0.000500 Time 0.023086 +2023-10-05 21:37:44,516 - Epoch: [118][ 360/ 1236] Overall Loss 0.238622 Objective Loss 0.238622 LR 0.000500 Time 0.023008 +2023-10-05 21:37:44,719 - Epoch: [118][ 370/ 1236] Overall Loss 0.239568 Objective Loss 0.239568 LR 0.000500 Time 0.022935 +2023-10-05 21:37:44,923 - Epoch: [118][ 380/ 1236] Overall Loss 0.239536 Objective Loss 0.239536 LR 0.000500 Time 0.022868 +2023-10-05 21:37:45,126 - Epoch: [118][ 390/ 1236] Overall Loss 0.239519 Objective Loss 0.239519 LR 0.000500 Time 0.022801 +2023-10-05 21:37:45,330 - Epoch: [118][ 400/ 1236] Overall Loss 0.239097 Objective Loss 0.239097 LR 0.000500 Time 0.022741 +2023-10-05 21:37:45,533 - Epoch: [118][ 410/ 1236] Overall Loss 0.239042 Objective Loss 0.239042 LR 0.000500 Time 0.022681 +2023-10-05 21:37:45,737 - Epoch: [118][ 420/ 1236] Overall Loss 0.238757 Objective Loss 0.238757 LR 0.000500 Time 0.022626 +2023-10-05 21:37:45,940 - Epoch: [118][ 430/ 1236] Overall Loss 0.238278 Objective Loss 0.238278 LR 0.000500 Time 0.022571 +2023-10-05 21:37:46,145 - Epoch: [118][ 440/ 1236] Overall Loss 0.237455 Objective Loss 0.237455 LR 0.000500 Time 0.022522 +2023-10-05 21:37:46,348 - Epoch: [118][ 450/ 1236] Overall Loss 0.237348 Objective Loss 0.237348 LR 0.000500 Time 0.022472 +2023-10-05 21:37:46,552 - Epoch: [118][ 460/ 1236] Overall Loss 0.237023 Objective Loss 0.237023 LR 0.000500 Time 0.022426 +2023-10-05 21:37:46,755 - Epoch: [118][ 470/ 1236] Overall Loss 0.237376 Objective Loss 0.237376 LR 0.000500 Time 0.022381 +2023-10-05 21:37:46,960 - Epoch: [118][ 480/ 1236] Overall Loss 0.237764 Objective Loss 0.237764 LR 0.000500 Time 0.022339 +2023-10-05 21:37:47,163 - Epoch: [118][ 490/ 1236] Overall Loss 0.237785 Objective Loss 0.237785 LR 0.000500 Time 0.022297 +2023-10-05 21:37:47,367 - Epoch: [118][ 500/ 1236] Overall Loss 0.237500 Objective Loss 0.237500 LR 0.000500 Time 0.022259 +2023-10-05 21:37:47,570 - Epoch: [118][ 510/ 1236] Overall Loss 0.237273 Objective Loss 0.237273 LR 0.000500 Time 0.022219 +2023-10-05 21:37:47,774 - Epoch: [118][ 520/ 1236] Overall Loss 0.237326 Objective Loss 0.237326 LR 0.000500 Time 0.022184 +2023-10-05 21:37:47,976 - Epoch: [118][ 530/ 1236] Overall Loss 0.237466 Objective Loss 0.237466 LR 0.000500 Time 0.022147 +2023-10-05 21:37:48,180 - Epoch: [118][ 540/ 1236] Overall Loss 0.237344 Objective Loss 0.237344 LR 0.000500 Time 0.022114 +2023-10-05 21:37:48,383 - Epoch: [118][ 550/ 1236] Overall Loss 0.237225 Objective Loss 0.237225 LR 0.000500 Time 0.022080 +2023-10-05 21:37:48,587 - Epoch: [118][ 560/ 1236] Overall Loss 0.236660 Objective Loss 0.236660 LR 0.000500 Time 0.022050 +2023-10-05 21:37:48,790 - Epoch: [118][ 570/ 1236] Overall Loss 0.236885 Objective Loss 0.236885 LR 0.000500 Time 0.022018 +2023-10-05 21:37:48,994 - Epoch: [118][ 580/ 1236] Overall Loss 0.236776 Objective Loss 0.236776 LR 0.000500 Time 0.021990 +2023-10-05 21:37:49,197 - Epoch: [118][ 590/ 1236] Overall Loss 0.236950 Objective Loss 0.236950 LR 0.000500 Time 0.021960 +2023-10-05 21:37:49,401 - Epoch: [118][ 600/ 1236] Overall Loss 0.236923 Objective Loss 0.236923 LR 0.000500 Time 0.021934 +2023-10-05 21:37:49,604 - Epoch: [118][ 610/ 1236] Overall Loss 0.236945 Objective Loss 0.236945 LR 0.000500 Time 0.021905 +2023-10-05 21:37:49,808 - Epoch: [118][ 620/ 1236] Overall Loss 0.236674 Objective Loss 0.236674 LR 0.000500 Time 0.021881 +2023-10-05 21:37:50,011 - Epoch: [118][ 630/ 1236] Overall Loss 0.236290 Objective Loss 0.236290 LR 0.000500 Time 0.021855 +2023-10-05 21:37:50,215 - Epoch: [118][ 640/ 1236] Overall Loss 0.236580 Objective Loss 0.236580 LR 0.000500 Time 0.021832 +2023-10-05 21:37:50,418 - Epoch: [118][ 650/ 1236] Overall Loss 0.236607 Objective Loss 0.236607 LR 0.000500 Time 0.021808 +2023-10-05 21:37:50,622 - Epoch: [118][ 660/ 1236] Overall Loss 0.236602 Objective Loss 0.236602 LR 0.000500 Time 0.021786 +2023-10-05 21:37:50,825 - Epoch: [118][ 670/ 1236] Overall Loss 0.236602 Objective Loss 0.236602 LR 0.000500 Time 0.021763 +2023-10-05 21:37:51,029 - Epoch: [118][ 680/ 1236] Overall Loss 0.236498 Objective Loss 0.236498 LR 0.000500 Time 0.021743 +2023-10-05 21:37:51,232 - Epoch: [118][ 690/ 1236] Overall Loss 0.236640 Objective Loss 0.236640 LR 0.000500 Time 0.021721 +2023-10-05 21:37:51,435 - Epoch: [118][ 700/ 1236] Overall Loss 0.236294 Objective Loss 0.236294 LR 0.000500 Time 0.021701 +2023-10-05 21:37:51,637 - Epoch: [118][ 710/ 1236] Overall Loss 0.236436 Objective Loss 0.236436 LR 0.000500 Time 0.021680 +2023-10-05 21:37:51,841 - Epoch: [118][ 720/ 1236] Overall Loss 0.236917 Objective Loss 0.236917 LR 0.000500 Time 0.021661 +2023-10-05 21:37:52,044 - Epoch: [118][ 730/ 1236] Overall Loss 0.237198 Objective Loss 0.237198 LR 0.000500 Time 0.021642 +2023-10-05 21:37:52,249 - Epoch: [118][ 740/ 1236] Overall Loss 0.236820 Objective Loss 0.236820 LR 0.000500 Time 0.021626 +2023-10-05 21:37:52,454 - Epoch: [118][ 750/ 1236] Overall Loss 0.236754 Objective Loss 0.236754 LR 0.000500 Time 0.021611 +2023-10-05 21:37:52,659 - Epoch: [118][ 760/ 1236] Overall Loss 0.236885 Objective Loss 0.236885 LR 0.000500 Time 0.021595 +2023-10-05 21:37:52,865 - Epoch: [118][ 770/ 1236] Overall Loss 0.237152 Objective Loss 0.237152 LR 0.000500 Time 0.021581 +2023-10-05 21:37:53,069 - Epoch: [118][ 780/ 1236] Overall Loss 0.237174 Objective Loss 0.237174 LR 0.000500 Time 0.021567 +2023-10-05 21:37:53,275 - Epoch: [118][ 790/ 1236] Overall Loss 0.237210 Objective Loss 0.237210 LR 0.000500 Time 0.021553 +2023-10-05 21:37:53,479 - Epoch: [118][ 800/ 1236] Overall Loss 0.238009 Objective Loss 0.238009 LR 0.000500 Time 0.021539 +2023-10-05 21:37:53,685 - Epoch: [118][ 810/ 1236] Overall Loss 0.238152 Objective Loss 0.238152 LR 0.000500 Time 0.021527 +2023-10-05 21:37:53,889 - Epoch: [118][ 820/ 1236] Overall Loss 0.238364 Objective Loss 0.238364 LR 0.000500 Time 0.021513 +2023-10-05 21:37:54,095 - Epoch: [118][ 830/ 1236] Overall Loss 0.238339 Objective Loss 0.238339 LR 0.000500 Time 0.021501 +2023-10-05 21:37:54,299 - Epoch: [118][ 840/ 1236] Overall Loss 0.238551 Objective Loss 0.238551 LR 0.000500 Time 0.021488 +2023-10-05 21:37:54,505 - Epoch: [118][ 850/ 1236] Overall Loss 0.238291 Objective Loss 0.238291 LR 0.000500 Time 0.021477 +2023-10-05 21:37:54,710 - Epoch: [118][ 860/ 1236] Overall Loss 0.238179 Objective Loss 0.238179 LR 0.000500 Time 0.021465 +2023-10-05 21:37:54,915 - Epoch: [118][ 870/ 1236] Overall Loss 0.238144 Objective Loss 0.238144 LR 0.000500 Time 0.021454 +2023-10-05 21:37:55,120 - Epoch: [118][ 880/ 1236] Overall Loss 0.238267 Objective Loss 0.238267 LR 0.000500 Time 0.021442 +2023-10-05 21:37:55,326 - Epoch: [118][ 890/ 1236] Overall Loss 0.238251 Objective Loss 0.238251 LR 0.000500 Time 0.021432 +2023-10-05 21:37:55,530 - Epoch: [118][ 900/ 1236] Overall Loss 0.238512 Objective Loss 0.238512 LR 0.000500 Time 0.021421 +2023-10-05 21:37:55,737 - Epoch: [118][ 910/ 1236] Overall Loss 0.238351 Objective Loss 0.238351 LR 0.000500 Time 0.021412 +2023-10-05 21:37:55,941 - Epoch: [118][ 920/ 1236] Overall Loss 0.238355 Objective Loss 0.238355 LR 0.000500 Time 0.021401 +2023-10-05 21:37:56,148 - Epoch: [118][ 930/ 1236] Overall Loss 0.238153 Objective Loss 0.238153 LR 0.000500 Time 0.021393 +2023-10-05 21:37:56,353 - Epoch: [118][ 940/ 1236] Overall Loss 0.238319 Objective Loss 0.238319 LR 0.000500 Time 0.021383 +2023-10-05 21:37:56,559 - Epoch: [118][ 950/ 1236] Overall Loss 0.238457 Objective Loss 0.238457 LR 0.000500 Time 0.021374 +2023-10-05 21:37:56,763 - Epoch: [118][ 960/ 1236] Overall Loss 0.238550 Objective Loss 0.238550 LR 0.000500 Time 0.021364 +2023-10-05 21:37:56,969 - Epoch: [118][ 970/ 1236] Overall Loss 0.238480 Objective Loss 0.238480 LR 0.000500 Time 0.021356 +2023-10-05 21:37:57,174 - Epoch: [118][ 980/ 1236] Overall Loss 0.238690 Objective Loss 0.238690 LR 0.000500 Time 0.021346 +2023-10-05 21:37:57,379 - Epoch: [118][ 990/ 1236] Overall Loss 0.238214 Objective Loss 0.238214 LR 0.000500 Time 0.021338 +2023-10-05 21:37:57,584 - Epoch: [118][ 1000/ 1236] Overall Loss 0.238033 Objective Loss 0.238033 LR 0.000500 Time 0.021329 +2023-10-05 21:37:57,789 - Epoch: [118][ 1010/ 1236] Overall Loss 0.237979 Objective Loss 0.237979 LR 0.000500 Time 0.021321 +2023-10-05 21:37:57,994 - Epoch: [118][ 1020/ 1236] Overall Loss 0.238135 Objective Loss 0.238135 LR 0.000500 Time 0.021312 +2023-10-05 21:37:58,200 - Epoch: [118][ 1030/ 1236] Overall Loss 0.238064 Objective Loss 0.238064 LR 0.000500 Time 0.021305 +2023-10-05 21:37:58,404 - Epoch: [118][ 1040/ 1236] Overall Loss 0.238184 Objective Loss 0.238184 LR 0.000500 Time 0.021296 +2023-10-05 21:37:58,610 - Epoch: [118][ 1050/ 1236] Overall Loss 0.238216 Objective Loss 0.238216 LR 0.000500 Time 0.021289 +2023-10-05 21:37:58,815 - Epoch: [118][ 1060/ 1236] Overall Loss 0.238256 Objective Loss 0.238256 LR 0.000500 Time 0.021281 +2023-10-05 21:37:59,020 - Epoch: [118][ 1070/ 1236] Overall Loss 0.238101 Objective Loss 0.238101 LR 0.000500 Time 0.021274 +2023-10-05 21:37:59,225 - Epoch: [118][ 1080/ 1236] Overall Loss 0.237824 Objective Loss 0.237824 LR 0.000500 Time 0.021266 +2023-10-05 21:37:59,431 - Epoch: [118][ 1090/ 1236] Overall Loss 0.237602 Objective Loss 0.237602 LR 0.000500 Time 0.021259 +2023-10-05 21:37:59,636 - Epoch: [118][ 1100/ 1236] Overall Loss 0.237651 Objective Loss 0.237651 LR 0.000500 Time 0.021252 +2023-10-05 21:37:59,841 - Epoch: [118][ 1110/ 1236] Overall Loss 0.237647 Objective Loss 0.237647 LR 0.000500 Time 0.021246 +2023-10-05 21:38:00,046 - Epoch: [118][ 1120/ 1236] Overall Loss 0.237667 Objective Loss 0.237667 LR 0.000500 Time 0.021238 +2023-10-05 21:38:00,252 - Epoch: [118][ 1130/ 1236] Overall Loss 0.237547 Objective Loss 0.237547 LR 0.000500 Time 0.021233 +2023-10-05 21:38:00,457 - Epoch: [118][ 1140/ 1236] Overall Loss 0.237397 Objective Loss 0.237397 LR 0.000500 Time 0.021225 +2023-10-05 21:38:00,663 - Epoch: [118][ 1150/ 1236] Overall Loss 0.237176 Objective Loss 0.237176 LR 0.000500 Time 0.021220 +2023-10-05 21:38:00,868 - Epoch: [118][ 1160/ 1236] Overall Loss 0.237219 Objective Loss 0.237219 LR 0.000500 Time 0.021213 +2023-10-05 21:38:01,074 - Epoch: [118][ 1170/ 1236] Overall Loss 0.237442 Objective Loss 0.237442 LR 0.000500 Time 0.021208 +2023-10-05 21:38:01,279 - Epoch: [118][ 1180/ 1236] Overall Loss 0.237229 Objective Loss 0.237229 LR 0.000500 Time 0.021201 +2023-10-05 21:38:01,484 - Epoch: [118][ 1190/ 1236] Overall Loss 0.237051 Objective Loss 0.237051 LR 0.000500 Time 0.021196 +2023-10-05 21:38:01,689 - Epoch: [118][ 1200/ 1236] Overall Loss 0.236923 Objective Loss 0.236923 LR 0.000500 Time 0.021190 +2023-10-05 21:38:01,895 - Epoch: [118][ 1210/ 1236] Overall Loss 0.237205 Objective Loss 0.237205 LR 0.000500 Time 0.021184 +2023-10-05 21:38:02,100 - Epoch: [118][ 1220/ 1236] Overall Loss 0.237188 Objective Loss 0.237188 LR 0.000500 Time 0.021179 +2023-10-05 21:38:02,358 - Epoch: [118][ 1230/ 1236] Overall Loss 0.237252 Objective Loss 0.237252 LR 0.000500 Time 0.021216 +2023-10-05 21:38:02,477 - Epoch: [118][ 1236/ 1236] Overall Loss 0.237193 Objective Loss 0.237193 Top1 88.187373 Top5 98.574338 LR 0.000500 Time 0.021209 +2023-10-05 21:38:02,606 - --- validate (epoch=118)----------- +2023-10-05 21:38:02,606 - 29943 samples (256 per mini-batch) +2023-10-05 21:38:03,072 - Epoch: [118][ 10/ 117] Loss 0.347017 Top1 83.085938 Top5 97.890625 +2023-10-05 21:38:03,234 - Epoch: [118][ 20/ 117] Loss 0.327152 Top1 83.437500 Top5 97.910156 +2023-10-05 21:38:03,389 - Epoch: [118][ 30/ 117] Loss 0.322527 Top1 83.372396 Top5 97.955729 +2023-10-05 21:38:03,550 - Epoch: [118][ 40/ 117] Loss 0.328000 Top1 83.242188 Top5 97.900391 +2023-10-05 21:38:03,705 - Epoch: [118][ 50/ 117] Loss 0.321822 Top1 83.515625 Top5 97.804688 +2023-10-05 21:38:03,866 - Epoch: [118][ 60/ 117] Loss 0.325507 Top1 83.515625 Top5 97.812500 +2023-10-05 21:38:04,022 - Epoch: [118][ 70/ 117] Loss 0.322867 Top1 83.710938 Top5 97.857143 +2023-10-05 21:38:04,183 - Epoch: [118][ 80/ 117] Loss 0.323758 Top1 83.710938 Top5 97.856445 +2023-10-05 21:38:04,338 - Epoch: [118][ 90/ 117] Loss 0.320857 Top1 83.736979 Top5 97.877604 +2023-10-05 21:38:04,499 - Epoch: [118][ 100/ 117] Loss 0.319539 Top1 83.664062 Top5 97.859375 +2023-10-05 21:38:04,662 - Epoch: [118][ 110/ 117] Loss 0.324482 Top1 83.536932 Top5 97.844460 +2023-10-05 21:38:04,748 - Epoch: [118][ 117/ 117] Loss 0.322644 Top1 83.572120 Top5 97.842568 +2023-10-05 21:38:04,857 - ==> Top1: 83.572 Top5: 97.843 Loss: 0.323 + +2023-10-05 21:38:04,858 - ==> Confusion: +[[ 930 1 1 0 6 3 0 1 9 70 1 1 1 2 4 2 4 1 2 1 10] + [ 2 1061 2 0 9 20 1 14 3 0 2 0 0 0 0 4 3 0 6 0 4] + [ 3 1 941 22 4 1 28 10 0 2 6 2 8 0 2 4 1 2 5 4 10] + [ 2 0 10 985 2 6 1 1 5 0 8 0 4 4 23 3 1 6 13 2 13] + [ 20 7 0 0 977 5 0 0 0 7 1 1 2 5 7 4 6 1 1 1 5] + [ 3 46 0 1 3 979 2 18 2 2 1 9 1 15 8 1 4 0 2 5 14] + [ 0 6 20 0 1 2 1120 10 0 0 5 4 3 0 1 7 0 1 2 4 5] + [ 3 26 13 0 3 41 4 1049 2 4 5 7 1 2 1 5 0 0 35 8 9] + [ 16 2 0 0 2 3 1 0 983 41 10 2 1 10 10 2 0 0 2 2 2] + [ 106 0 2 0 8 4 0 0 34 920 1 1 0 22 4 8 1 0 0 3 5] + [ 3 4 8 10 2 1 1 3 12 4 969 3 0 13 3 1 3 1 1 2 9] + [ 1 2 1 0 2 11 1 2 0 1 0 966 18 4 0 3 1 16 0 4 2] + [ 2 4 3 7 0 2 0 1 2 1 1 34 974 5 0 5 2 19 0 1 5] + [ 5 0 2 0 3 5 0 0 12 15 4 3 3 1058 4 0 0 0 0 1 4] + [ 15 4 3 14 4 0 0 0 36 3 1 2 1 2 995 0 1 2 9 0 9] + [ 0 4 3 1 6 2 1 0 0 0 0 7 3 3 1 1065 17 11 0 7 3] + [ 3 13 1 0 8 5 0 2 3 0 0 2 1 1 4 13 1096 0 0 3 6] + [ 0 0 0 1 0 0 0 0 0 0 0 1 19 3 0 6 0 1002 1 2 3] + [ 2 11 6 21 1 0 0 21 3 0 5 0 0 1 9 0 2 0 975 0 11] + [ 1 4 4 2 3 5 11 12 1 0 1 13 2 3 0 4 8 2 2 1064 10] + [ 144 214 107 73 126 158 49 88 128 103 195 149 334 303 152 65 224 76 130 173 4914]] + +2023-10-05 21:38:04,859 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:38:04,859 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:38:04,865 - + +2023-10-05 21:38:04,865 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:38:05,873 - Epoch: [119][ 10/ 1236] Overall Loss 0.201843 Objective Loss 0.201843 LR 0.000500 Time 0.100743 +2023-10-05 21:38:06,077 - Epoch: [119][ 20/ 1236] Overall Loss 0.208940 Objective Loss 0.208940 LR 0.000500 Time 0.060537 +2023-10-05 21:38:06,279 - Epoch: [119][ 30/ 1236] Overall Loss 0.219051 Objective Loss 0.219051 LR 0.000500 Time 0.047094 +2023-10-05 21:38:06,482 - Epoch: [119][ 40/ 1236] Overall Loss 0.221166 Objective Loss 0.221166 LR 0.000500 Time 0.040385 +2023-10-05 21:38:06,685 - Epoch: [119][ 50/ 1236] Overall Loss 0.223517 Objective Loss 0.223517 LR 0.000500 Time 0.036351 +2023-10-05 21:38:06,887 - Epoch: [119][ 60/ 1236] Overall Loss 0.219163 Objective Loss 0.219163 LR 0.000500 Time 0.033662 +2023-10-05 21:38:07,090 - Epoch: [119][ 70/ 1236] Overall Loss 0.219279 Objective Loss 0.219279 LR 0.000500 Time 0.031736 +2023-10-05 21:38:07,293 - Epoch: [119][ 80/ 1236] Overall Loss 0.222315 Objective Loss 0.222315 LR 0.000500 Time 0.030310 +2023-10-05 21:38:07,498 - Epoch: [119][ 90/ 1236] Overall Loss 0.220759 Objective Loss 0.220759 LR 0.000500 Time 0.029213 +2023-10-05 21:38:07,703 - Epoch: [119][ 100/ 1236] Overall Loss 0.224734 Objective Loss 0.224734 LR 0.000500 Time 0.028342 +2023-10-05 21:38:07,908 - Epoch: [119][ 110/ 1236] Overall Loss 0.223919 Objective Loss 0.223919 LR 0.000500 Time 0.027622 +2023-10-05 21:38:08,115 - Epoch: [119][ 120/ 1236] Overall Loss 0.221566 Objective Loss 0.221566 LR 0.000500 Time 0.027039 +2023-10-05 21:38:08,322 - Epoch: [119][ 130/ 1236] Overall Loss 0.222063 Objective Loss 0.222063 LR 0.000500 Time 0.026546 +2023-10-05 21:38:08,529 - Epoch: [119][ 140/ 1236] Overall Loss 0.222286 Objective Loss 0.222286 LR 0.000500 Time 0.026127 +2023-10-05 21:38:08,735 - Epoch: [119][ 150/ 1236] Overall Loss 0.220819 Objective Loss 0.220819 LR 0.000500 Time 0.025757 +2023-10-05 21:38:08,942 - Epoch: [119][ 160/ 1236] Overall Loss 0.221599 Objective Loss 0.221599 LR 0.000500 Time 0.025438 +2023-10-05 21:38:09,149 - Epoch: [119][ 170/ 1236] Overall Loss 0.222884 Objective Loss 0.222884 LR 0.000500 Time 0.025155 +2023-10-05 21:38:09,356 - Epoch: [119][ 180/ 1236] Overall Loss 0.223037 Objective Loss 0.223037 LR 0.000500 Time 0.024906 +2023-10-05 21:38:09,563 - Epoch: [119][ 190/ 1236] Overall Loss 0.223686 Objective Loss 0.223686 LR 0.000500 Time 0.024680 +2023-10-05 21:38:09,770 - Epoch: [119][ 200/ 1236] Overall Loss 0.223822 Objective Loss 0.223822 LR 0.000500 Time 0.024479 +2023-10-05 21:38:09,977 - Epoch: [119][ 210/ 1236] Overall Loss 0.224453 Objective Loss 0.224453 LR 0.000500 Time 0.024295 +2023-10-05 21:38:10,184 - Epoch: [119][ 220/ 1236] Overall Loss 0.224227 Objective Loss 0.224227 LR 0.000500 Time 0.024132 +2023-10-05 21:38:10,390 - Epoch: [119][ 230/ 1236] Overall Loss 0.224917 Objective Loss 0.224917 LR 0.000500 Time 0.023978 +2023-10-05 21:38:10,598 - Epoch: [119][ 240/ 1236] Overall Loss 0.224952 Objective Loss 0.224952 LR 0.000500 Time 0.023841 +2023-10-05 21:38:10,804 - Epoch: [119][ 250/ 1236] Overall Loss 0.223990 Objective Loss 0.223990 LR 0.000500 Time 0.023711 +2023-10-05 21:38:11,011 - Epoch: [119][ 260/ 1236] Overall Loss 0.225157 Objective Loss 0.225157 LR 0.000500 Time 0.023594 +2023-10-05 21:38:11,218 - Epoch: [119][ 270/ 1236] Overall Loss 0.226715 Objective Loss 0.226715 LR 0.000500 Time 0.023483 +2023-10-05 21:38:11,425 - Epoch: [119][ 280/ 1236] Overall Loss 0.226719 Objective Loss 0.226719 LR 0.000500 Time 0.023382 +2023-10-05 21:38:11,631 - Epoch: [119][ 290/ 1236] Overall Loss 0.226567 Objective Loss 0.226567 LR 0.000500 Time 0.023285 +2023-10-05 21:38:11,839 - Epoch: [119][ 300/ 1236] Overall Loss 0.227662 Objective Loss 0.227662 LR 0.000500 Time 0.023199 +2023-10-05 21:38:12,045 - Epoch: [119][ 310/ 1236] Overall Loss 0.227825 Objective Loss 0.227825 LR 0.000500 Time 0.023116 +2023-10-05 21:38:12,253 - Epoch: [119][ 320/ 1236] Overall Loss 0.228159 Objective Loss 0.228159 LR 0.000500 Time 0.023040 +2023-10-05 21:38:12,460 - Epoch: [119][ 330/ 1236] Overall Loss 0.227007 Objective Loss 0.227007 LR 0.000500 Time 0.022967 +2023-10-05 21:38:12,667 - Epoch: [119][ 340/ 1236] Overall Loss 0.227383 Objective Loss 0.227383 LR 0.000500 Time 0.022900 +2023-10-05 21:38:12,874 - Epoch: [119][ 350/ 1236] Overall Loss 0.227853 Objective Loss 0.227853 LR 0.000500 Time 0.022835 +2023-10-05 21:38:13,081 - Epoch: [119][ 360/ 1236] Overall Loss 0.227893 Objective Loss 0.227893 LR 0.000500 Time 0.022775 +2023-10-05 21:38:13,288 - Epoch: [119][ 370/ 1236] Overall Loss 0.228169 Objective Loss 0.228169 LR 0.000500 Time 0.022718 +2023-10-05 21:38:13,492 - Epoch: [119][ 380/ 1236] Overall Loss 0.228866 Objective Loss 0.228866 LR 0.000500 Time 0.022655 +2023-10-05 21:38:13,694 - Epoch: [119][ 390/ 1236] Overall Loss 0.228712 Objective Loss 0.228712 LR 0.000500 Time 0.022592 +2023-10-05 21:38:13,897 - Epoch: [119][ 400/ 1236] Overall Loss 0.229549 Objective Loss 0.229549 LR 0.000500 Time 0.022535 +2023-10-05 21:38:14,100 - Epoch: [119][ 410/ 1236] Overall Loss 0.229450 Objective Loss 0.229450 LR 0.000500 Time 0.022479 +2023-10-05 21:38:14,303 - Epoch: [119][ 420/ 1236] Overall Loss 0.230137 Objective Loss 0.230137 LR 0.000500 Time 0.022427 +2023-10-05 21:38:14,506 - Epoch: [119][ 430/ 1236] Overall Loss 0.230038 Objective Loss 0.230038 LR 0.000500 Time 0.022376 +2023-10-05 21:38:14,710 - Epoch: [119][ 440/ 1236] Overall Loss 0.230295 Objective Loss 0.230295 LR 0.000500 Time 0.022329 +2023-10-05 21:38:14,912 - Epoch: [119][ 450/ 1236] Overall Loss 0.230346 Objective Loss 0.230346 LR 0.000500 Time 0.022282 +2023-10-05 21:38:15,115 - Epoch: [119][ 460/ 1236] Overall Loss 0.230159 Objective Loss 0.230159 LR 0.000500 Time 0.022239 +2023-10-05 21:38:15,318 - Epoch: [119][ 470/ 1236] Overall Loss 0.229984 Objective Loss 0.229984 LR 0.000500 Time 0.022196 +2023-10-05 21:38:15,521 - Epoch: [119][ 480/ 1236] Overall Loss 0.230355 Objective Loss 0.230355 LR 0.000500 Time 0.022156 +2023-10-05 21:38:15,724 - Epoch: [119][ 490/ 1236] Overall Loss 0.230461 Objective Loss 0.230461 LR 0.000500 Time 0.022117 +2023-10-05 21:38:15,927 - Epoch: [119][ 500/ 1236] Overall Loss 0.230547 Objective Loss 0.230547 LR 0.000500 Time 0.022080 +2023-10-05 21:38:16,129 - Epoch: [119][ 510/ 1236] Overall Loss 0.230640 Objective Loss 0.230640 LR 0.000500 Time 0.022044 +2023-10-05 21:38:16,333 - Epoch: [119][ 520/ 1236] Overall Loss 0.230981 Objective Loss 0.230981 LR 0.000500 Time 0.022011 +2023-10-05 21:38:16,535 - Epoch: [119][ 530/ 1236] Overall Loss 0.230813 Objective Loss 0.230813 LR 0.000500 Time 0.021977 +2023-10-05 21:38:16,739 - Epoch: [119][ 540/ 1236] Overall Loss 0.230988 Objective Loss 0.230988 LR 0.000500 Time 0.021946 +2023-10-05 21:38:16,942 - Epoch: [119][ 550/ 1236] Overall Loss 0.230935 Objective Loss 0.230935 LR 0.000500 Time 0.021915 +2023-10-05 21:38:17,145 - Epoch: [119][ 560/ 1236] Overall Loss 0.231144 Objective Loss 0.231144 LR 0.000500 Time 0.021886 +2023-10-05 21:38:17,348 - Epoch: [119][ 570/ 1236] Overall Loss 0.231276 Objective Loss 0.231276 LR 0.000500 Time 0.021857 +2023-10-05 21:38:17,553 - Epoch: [119][ 580/ 1236] Overall Loss 0.231100 Objective Loss 0.231100 LR 0.000500 Time 0.021834 +2023-10-05 21:38:17,758 - Epoch: [119][ 590/ 1236] Overall Loss 0.230920 Objective Loss 0.230920 LR 0.000500 Time 0.021810 +2023-10-05 21:38:17,965 - Epoch: [119][ 600/ 1236] Overall Loss 0.230985 Objective Loss 0.230985 LR 0.000500 Time 0.021791 +2023-10-05 21:38:18,169 - Epoch: [119][ 610/ 1236] Overall Loss 0.231633 Objective Loss 0.231633 LR 0.000500 Time 0.021769 +2023-10-05 21:38:18,376 - Epoch: [119][ 620/ 1236] Overall Loss 0.231639 Objective Loss 0.231639 LR 0.000500 Time 0.021751 +2023-10-05 21:38:18,581 - Epoch: [119][ 630/ 1236] Overall Loss 0.231656 Objective Loss 0.231656 LR 0.000500 Time 0.021730 +2023-10-05 21:38:18,788 - Epoch: [119][ 640/ 1236] Overall Loss 0.231904 Objective Loss 0.231904 LR 0.000500 Time 0.021712 +2023-10-05 21:38:18,993 - Epoch: [119][ 650/ 1236] Overall Loss 0.232597 Objective Loss 0.232597 LR 0.000500 Time 0.021693 +2023-10-05 21:38:19,200 - Epoch: [119][ 660/ 1236] Overall Loss 0.232951 Objective Loss 0.232951 LR 0.000500 Time 0.021677 +2023-10-05 21:38:19,404 - Epoch: [119][ 670/ 1236] Overall Loss 0.233131 Objective Loss 0.233131 LR 0.000500 Time 0.021659 +2023-10-05 21:38:19,611 - Epoch: [119][ 680/ 1236] Overall Loss 0.233004 Objective Loss 0.233004 LR 0.000500 Time 0.021643 +2023-10-05 21:38:19,815 - Epoch: [119][ 690/ 1236] Overall Loss 0.233384 Objective Loss 0.233384 LR 0.000500 Time 0.021625 +2023-10-05 21:38:20,022 - Epoch: [119][ 700/ 1236] Overall Loss 0.233365 Objective Loss 0.233365 LR 0.000500 Time 0.021611 +2023-10-05 21:38:20,226 - Epoch: [119][ 710/ 1236] Overall Loss 0.233593 Objective Loss 0.233593 LR 0.000500 Time 0.021594 +2023-10-05 21:38:20,434 - Epoch: [119][ 720/ 1236] Overall Loss 0.233243 Objective Loss 0.233243 LR 0.000500 Time 0.021583 +2023-10-05 21:38:20,638 - Epoch: [119][ 730/ 1236] Overall Loss 0.232904 Objective Loss 0.232904 LR 0.000500 Time 0.021566 +2023-10-05 21:38:20,845 - Epoch: [119][ 740/ 1236] Overall Loss 0.232991 Objective Loss 0.232991 LR 0.000500 Time 0.021553 +2023-10-05 21:38:21,049 - Epoch: [119][ 750/ 1236] Overall Loss 0.232913 Objective Loss 0.232913 LR 0.000500 Time 0.021538 +2023-10-05 21:38:21,256 - Epoch: [119][ 760/ 1236] Overall Loss 0.232829 Objective Loss 0.232829 LR 0.000500 Time 0.021526 +2023-10-05 21:38:21,460 - Epoch: [119][ 770/ 1236] Overall Loss 0.232985 Objective Loss 0.232985 LR 0.000500 Time 0.021511 +2023-10-05 21:38:21,667 - Epoch: [119][ 780/ 1236] Overall Loss 0.233069 Objective Loss 0.233069 LR 0.000500 Time 0.021500 +2023-10-05 21:38:21,872 - Epoch: [119][ 790/ 1236] Overall Loss 0.233090 Objective Loss 0.233090 LR 0.000500 Time 0.021487 +2023-10-05 21:38:22,080 - Epoch: [119][ 800/ 1236] Overall Loss 0.233398 Objective Loss 0.233398 LR 0.000500 Time 0.021478 +2023-10-05 21:38:22,286 - Epoch: [119][ 810/ 1236] Overall Loss 0.233692 Objective Loss 0.233692 LR 0.000500 Time 0.021466 +2023-10-05 21:38:22,493 - Epoch: [119][ 820/ 1236] Overall Loss 0.233490 Objective Loss 0.233490 LR 0.000500 Time 0.021457 +2023-10-05 21:38:22,698 - Epoch: [119][ 830/ 1236] Overall Loss 0.233603 Objective Loss 0.233603 LR 0.000500 Time 0.021445 +2023-10-05 21:38:22,905 - Epoch: [119][ 840/ 1236] Overall Loss 0.233907 Objective Loss 0.233907 LR 0.000500 Time 0.021436 +2023-10-05 21:38:23,110 - Epoch: [119][ 850/ 1236] Overall Loss 0.234178 Objective Loss 0.234178 LR 0.000500 Time 0.021424 +2023-10-05 21:38:23,318 - Epoch: [119][ 860/ 1236] Overall Loss 0.234417 Objective Loss 0.234417 LR 0.000500 Time 0.021416 +2023-10-05 21:38:23,522 - Epoch: [119][ 870/ 1236] Overall Loss 0.234146 Objective Loss 0.234146 LR 0.000500 Time 0.021404 +2023-10-05 21:38:23,730 - Epoch: [119][ 880/ 1236] Overall Loss 0.234147 Objective Loss 0.234147 LR 0.000500 Time 0.021396 +2023-10-05 21:38:23,934 - Epoch: [119][ 890/ 1236] Overall Loss 0.234356 Objective Loss 0.234356 LR 0.000500 Time 0.021386 +2023-10-05 21:38:24,143 - Epoch: [119][ 900/ 1236] Overall Loss 0.234910 Objective Loss 0.234910 LR 0.000500 Time 0.021379 +2023-10-05 21:38:24,347 - Epoch: [119][ 910/ 1236] Overall Loss 0.235338 Objective Loss 0.235338 LR 0.000500 Time 0.021368 +2023-10-05 21:38:24,556 - Epoch: [119][ 920/ 1236] Overall Loss 0.235025 Objective Loss 0.235025 LR 0.000500 Time 0.021363 +2023-10-05 21:38:24,761 - Epoch: [119][ 930/ 1236] Overall Loss 0.235165 Objective Loss 0.235165 LR 0.000500 Time 0.021353 +2023-10-05 21:38:24,966 - Epoch: [119][ 940/ 1236] Overall Loss 0.235175 Objective Loss 0.235175 LR 0.000500 Time 0.021343 +2023-10-05 21:38:25,169 - Epoch: [119][ 950/ 1236] Overall Loss 0.235472 Objective Loss 0.235472 LR 0.000500 Time 0.021333 +2023-10-05 21:38:25,374 - Epoch: [119][ 960/ 1236] Overall Loss 0.235161 Objective Loss 0.235161 LR 0.000500 Time 0.021324 +2023-10-05 21:38:25,578 - Epoch: [119][ 970/ 1236] Overall Loss 0.235101 Objective Loss 0.235101 LR 0.000500 Time 0.021313 +2023-10-05 21:38:25,782 - Epoch: [119][ 980/ 1236] Overall Loss 0.235189 Objective Loss 0.235189 LR 0.000500 Time 0.021304 +2023-10-05 21:38:25,991 - Epoch: [119][ 990/ 1236] Overall Loss 0.235202 Objective Loss 0.235202 LR 0.000500 Time 0.021299 +2023-10-05 21:38:26,203 - Epoch: [119][ 1000/ 1236] Overall Loss 0.235324 Objective Loss 0.235324 LR 0.000500 Time 0.021298 +2023-10-05 21:38:26,413 - Epoch: [119][ 1010/ 1236] Overall Loss 0.235520 Objective Loss 0.235520 LR 0.000500 Time 0.021295 +2023-10-05 21:38:26,628 - Epoch: [119][ 1020/ 1236] Overall Loss 0.235674 Objective Loss 0.235674 LR 0.000500 Time 0.021296 +2023-10-05 21:38:26,838 - Epoch: [119][ 1030/ 1236] Overall Loss 0.235641 Objective Loss 0.235641 LR 0.000500 Time 0.021293 +2023-10-05 21:38:27,043 - Epoch: [119][ 1040/ 1236] Overall Loss 0.235566 Objective Loss 0.235566 LR 0.000500 Time 0.021285 +2023-10-05 21:38:27,248 - Epoch: [119][ 1050/ 1236] Overall Loss 0.235344 Objective Loss 0.235344 LR 0.000500 Time 0.021277 +2023-10-05 21:38:27,451 - Epoch: [119][ 1060/ 1236] Overall Loss 0.235255 Objective Loss 0.235255 LR 0.000500 Time 0.021268 +2023-10-05 21:38:27,655 - Epoch: [119][ 1070/ 1236] Overall Loss 0.235185 Objective Loss 0.235185 LR 0.000500 Time 0.021259 +2023-10-05 21:38:27,859 - Epoch: [119][ 1080/ 1236] Overall Loss 0.235112 Objective Loss 0.235112 LR 0.000500 Time 0.021251 +2023-10-05 21:38:28,063 - Epoch: [119][ 1090/ 1236] Overall Loss 0.235198 Objective Loss 0.235198 LR 0.000500 Time 0.021243 +2023-10-05 21:38:28,268 - Epoch: [119][ 1100/ 1236] Overall Loss 0.235299 Objective Loss 0.235299 LR 0.000500 Time 0.021235 +2023-10-05 21:38:28,471 - Epoch: [119][ 1110/ 1236] Overall Loss 0.235227 Objective Loss 0.235227 LR 0.000500 Time 0.021227 +2023-10-05 21:38:28,675 - Epoch: [119][ 1120/ 1236] Overall Loss 0.235200 Objective Loss 0.235200 LR 0.000500 Time 0.021219 +2023-10-05 21:38:28,879 - Epoch: [119][ 1130/ 1236] Overall Loss 0.235383 Objective Loss 0.235383 LR 0.000500 Time 0.021212 +2023-10-05 21:38:29,083 - Epoch: [119][ 1140/ 1236] Overall Loss 0.235579 Objective Loss 0.235579 LR 0.000500 Time 0.021204 +2023-10-05 21:38:29,287 - Epoch: [119][ 1150/ 1236] Overall Loss 0.235382 Objective Loss 0.235382 LR 0.000500 Time 0.021197 +2023-10-05 21:38:29,493 - Epoch: [119][ 1160/ 1236] Overall Loss 0.235500 Objective Loss 0.235500 LR 0.000500 Time 0.021192 +2023-10-05 21:38:29,698 - Epoch: [119][ 1170/ 1236] Overall Loss 0.235573 Objective Loss 0.235573 LR 0.000500 Time 0.021185 +2023-10-05 21:38:29,899 - Epoch: [119][ 1180/ 1236] Overall Loss 0.235387 Objective Loss 0.235387 LR 0.000500 Time 0.021176 +2023-10-05 21:38:30,103 - Epoch: [119][ 1190/ 1236] Overall Loss 0.235304 Objective Loss 0.235304 LR 0.000500 Time 0.021169 +2023-10-05 21:38:30,305 - Epoch: [119][ 1200/ 1236] Overall Loss 0.235309 Objective Loss 0.235309 LR 0.000500 Time 0.021160 +2023-10-05 21:38:30,509 - Epoch: [119][ 1210/ 1236] Overall Loss 0.235146 Objective Loss 0.235146 LR 0.000500 Time 0.021154 +2023-10-05 21:38:30,710 - Epoch: [119][ 1220/ 1236] Overall Loss 0.235214 Objective Loss 0.235214 LR 0.000500 Time 0.021145 +2023-10-05 21:38:30,963 - Epoch: [119][ 1230/ 1236] Overall Loss 0.235199 Objective Loss 0.235199 LR 0.000500 Time 0.021179 +2023-10-05 21:38:31,081 - Epoch: [119][ 1236/ 1236] Overall Loss 0.235175 Objective Loss 0.235175 Top1 86.965377 Top5 97.556008 LR 0.000500 Time 0.021171 +2023-10-05 21:38:31,209 - --- validate (epoch=119)----------- +2023-10-05 21:38:31,210 - 29943 samples (256 per mini-batch) +2023-10-05 21:38:31,666 - Epoch: [119][ 10/ 117] Loss 0.294956 Top1 85.195312 Top5 98.398438 +2023-10-05 21:38:31,818 - Epoch: [119][ 20/ 117] Loss 0.299068 Top1 84.511719 Top5 98.046875 +2023-10-05 21:38:31,968 - Epoch: [119][ 30/ 117] Loss 0.306695 Top1 84.322917 Top5 97.981771 +2023-10-05 21:38:32,120 - Epoch: [119][ 40/ 117] Loss 0.312412 Top1 84.316406 Top5 98.007812 +2023-10-05 21:38:32,280 - Epoch: [119][ 50/ 117] Loss 0.324403 Top1 83.835938 Top5 97.976562 +2023-10-05 21:38:32,438 - Epoch: [119][ 60/ 117] Loss 0.327913 Top1 83.828125 Top5 97.884115 +2023-10-05 21:38:32,599 - Epoch: [119][ 70/ 117] Loss 0.327399 Top1 83.956473 Top5 97.868304 +2023-10-05 21:38:32,755 - Epoch: [119][ 80/ 117] Loss 0.328766 Top1 83.852539 Top5 97.846680 +2023-10-05 21:38:32,912 - Epoch: [119][ 90/ 117] Loss 0.329939 Top1 83.871528 Top5 97.825521 +2023-10-05 21:38:33,067 - Epoch: [119][ 100/ 117] Loss 0.331864 Top1 83.921875 Top5 97.820312 +2023-10-05 21:38:33,227 - Epoch: [119][ 110/ 117] Loss 0.330280 Top1 83.966619 Top5 97.826705 +2023-10-05 21:38:33,311 - Epoch: [119][ 117/ 117] Loss 0.328886 Top1 83.989580 Top5 97.855926 +2023-10-05 21:38:33,421 - ==> Top1: 83.990 Top5: 97.856 Loss: 0.329 + +2023-10-05 21:38:33,422 - ==> Confusion: +[[ 934 1 5 2 8 2 0 1 10 60 1 2 2 3 4 0 4 0 0 1 10] + [ 1 1056 2 0 8 21 1 18 2 0 3 0 0 0 0 4 1 0 10 2 2] + [ 4 1 941 17 3 0 31 8 0 3 2 3 11 1 2 4 2 1 10 6 6] + [ 2 0 13 960 1 5 1 0 2 1 11 0 9 5 30 3 3 5 23 3 12] + [ 23 9 1 0 977 2 0 0 0 7 3 2 1 1 10 1 7 3 1 0 2] + [ 4 49 1 1 8 965 2 21 1 1 3 9 4 14 6 1 3 0 2 6 15] + [ 0 5 17 0 2 1 1125 11 0 0 3 2 2 0 0 8 1 2 2 3 7] + [ 4 21 14 1 2 32 8 1061 4 1 4 5 5 2 0 2 1 0 35 8 8] + [ 23 3 0 1 1 1 0 0 971 36 13 1 4 10 15 2 1 0 6 0 1] + [ 117 0 2 0 10 6 0 0 28 917 0 1 2 17 6 3 1 0 2 1 6] + [ 4 6 11 3 2 1 5 6 15 2 951 3 0 16 4 1 2 1 3 4 13] + [ 1 0 0 0 0 12 1 1 0 2 0 948 35 3 0 4 2 14 1 5 6] + [ 0 0 2 2 1 3 0 3 1 0 2 28 987 7 1 4 1 13 1 0 12] + [ 2 0 2 2 6 6 0 0 13 9 2 3 3 1051 4 1 4 0 0 4 7] + [ 14 4 3 8 5 0 0 0 25 3 4 2 1 0 1011 0 1 0 12 0 8] + [ 1 3 2 0 4 1 3 0 0 0 0 9 9 1 1 1062 17 9 0 7 5] + [ 1 14 1 0 8 3 0 1 1 0 0 4 1 1 4 7 1101 0 1 4 9] + [ 0 0 0 1 0 1 0 0 0 0 0 5 27 1 2 7 0 989 0 1 4] + [ 3 7 7 10 0 0 1 21 2 0 2 1 4 0 14 0 1 0 983 1 11] + [ 0 5 1 0 1 6 10 6 1 0 3 12 6 1 0 3 8 2 2 1075 10] + [ 132 192 140 62 140 130 65 80 121 86 169 117 343 281 139 47 172 64 166 175 5084]] + +2023-10-05 21:38:33,423 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:38:33,423 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:38:33,429 - + +2023-10-05 21:38:33,429 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:38:34,439 - Epoch: [120][ 10/ 1236] Overall Loss 0.241425 Objective Loss 0.241425 LR 0.000500 Time 0.100895 +2023-10-05 21:38:34,642 - Epoch: [120][ 20/ 1236] Overall Loss 0.229749 Objective Loss 0.229749 LR 0.000500 Time 0.060576 +2023-10-05 21:38:34,842 - Epoch: [120][ 30/ 1236] Overall Loss 0.227126 Objective Loss 0.227126 LR 0.000500 Time 0.047068 +2023-10-05 21:38:35,045 - Epoch: [120][ 40/ 1236] Overall Loss 0.227239 Objective Loss 0.227239 LR 0.000500 Time 0.040358 +2023-10-05 21:38:35,246 - Epoch: [120][ 50/ 1236] Overall Loss 0.230327 Objective Loss 0.230327 LR 0.000500 Time 0.036297 +2023-10-05 21:38:35,449 - Epoch: [120][ 60/ 1236] Overall Loss 0.232047 Objective Loss 0.232047 LR 0.000500 Time 0.033619 +2023-10-05 21:38:35,649 - Epoch: [120][ 70/ 1236] Overall Loss 0.231142 Objective Loss 0.231142 LR 0.000500 Time 0.031678 +2023-10-05 21:38:35,852 - Epoch: [120][ 80/ 1236] Overall Loss 0.233214 Objective Loss 0.233214 LR 0.000500 Time 0.030248 +2023-10-05 21:38:36,060 - Epoch: [120][ 90/ 1236] Overall Loss 0.231598 Objective Loss 0.231598 LR 0.000500 Time 0.029190 +2023-10-05 21:38:36,263 - Epoch: [120][ 100/ 1236] Overall Loss 0.231514 Objective Loss 0.231514 LR 0.000500 Time 0.028304 +2023-10-05 21:38:36,464 - Epoch: [120][ 110/ 1236] Overall Loss 0.231751 Objective Loss 0.231751 LR 0.000500 Time 0.027548 +2023-10-05 21:38:36,665 - Epoch: [120][ 120/ 1236] Overall Loss 0.230595 Objective Loss 0.230595 LR 0.000500 Time 0.026931 +2023-10-05 21:38:36,865 - Epoch: [120][ 130/ 1236] Overall Loss 0.231134 Objective Loss 0.231134 LR 0.000500 Time 0.026389 +2023-10-05 21:38:37,066 - Epoch: [120][ 140/ 1236] Overall Loss 0.230173 Objective Loss 0.230173 LR 0.000500 Time 0.025939 +2023-10-05 21:38:37,265 - Epoch: [120][ 150/ 1236] Overall Loss 0.231695 Objective Loss 0.231695 LR 0.000500 Time 0.025538 +2023-10-05 21:38:37,467 - Epoch: [120][ 160/ 1236] Overall Loss 0.231114 Objective Loss 0.231114 LR 0.000500 Time 0.025198 +2023-10-05 21:38:37,666 - Epoch: [120][ 170/ 1236] Overall Loss 0.230516 Objective Loss 0.230516 LR 0.000500 Time 0.024886 +2023-10-05 21:38:37,867 - Epoch: [120][ 180/ 1236] Overall Loss 0.231268 Objective Loss 0.231268 LR 0.000500 Time 0.024617 +2023-10-05 21:38:38,066 - Epoch: [120][ 190/ 1236] Overall Loss 0.231314 Objective Loss 0.231314 LR 0.000500 Time 0.024369 +2023-10-05 21:38:38,268 - Epoch: [120][ 200/ 1236] Overall Loss 0.230924 Objective Loss 0.230924 LR 0.000500 Time 0.024156 +2023-10-05 21:38:38,468 - Epoch: [120][ 210/ 1236] Overall Loss 0.230513 Objective Loss 0.230513 LR 0.000500 Time 0.023955 +2023-10-05 21:38:38,669 - Epoch: [120][ 220/ 1236] Overall Loss 0.229951 Objective Loss 0.229951 LR 0.000500 Time 0.023780 +2023-10-05 21:38:38,868 - Epoch: [120][ 230/ 1236] Overall Loss 0.230534 Objective Loss 0.230534 LR 0.000500 Time 0.023611 +2023-10-05 21:38:39,070 - Epoch: [120][ 240/ 1236] Overall Loss 0.232017 Objective Loss 0.232017 LR 0.000500 Time 0.023465 +2023-10-05 21:38:39,269 - Epoch: [120][ 250/ 1236] Overall Loss 0.230914 Objective Loss 0.230914 LR 0.000500 Time 0.023323 +2023-10-05 21:38:39,470 - Epoch: [120][ 260/ 1236] Overall Loss 0.230240 Objective Loss 0.230240 LR 0.000500 Time 0.023198 +2023-10-05 21:38:39,670 - Epoch: [120][ 270/ 1236] Overall Loss 0.231490 Objective Loss 0.231490 LR 0.000500 Time 0.023077 +2023-10-05 21:38:39,871 - Epoch: [120][ 280/ 1236] Overall Loss 0.231769 Objective Loss 0.231769 LR 0.000500 Time 0.022969 +2023-10-05 21:38:40,071 - Epoch: [120][ 290/ 1236] Overall Loss 0.230930 Objective Loss 0.230930 LR 0.000500 Time 0.022864 +2023-10-05 21:38:40,272 - Epoch: [120][ 300/ 1236] Overall Loss 0.230984 Objective Loss 0.230984 LR 0.000500 Time 0.022772 +2023-10-05 21:38:40,472 - Epoch: [120][ 310/ 1236] Overall Loss 0.233549 Objective Loss 0.233549 LR 0.000500 Time 0.022682 +2023-10-05 21:38:40,672 - Epoch: [120][ 320/ 1236] Overall Loss 0.233470 Objective Loss 0.233470 LR 0.000500 Time 0.022598 +2023-10-05 21:38:40,872 - Epoch: [120][ 330/ 1236] Overall Loss 0.233164 Objective Loss 0.233164 LR 0.000500 Time 0.022518 +2023-10-05 21:38:41,074 - Epoch: [120][ 340/ 1236] Overall Loss 0.232965 Objective Loss 0.232965 LR 0.000500 Time 0.022448 +2023-10-05 21:38:41,274 - Epoch: [120][ 350/ 1236] Overall Loss 0.232693 Objective Loss 0.232693 LR 0.000500 Time 0.022377 +2023-10-05 21:38:41,474 - Epoch: [120][ 360/ 1236] Overall Loss 0.232359 Objective Loss 0.232359 LR 0.000500 Time 0.022310 +2023-10-05 21:38:41,674 - Epoch: [120][ 370/ 1236] Overall Loss 0.231952 Objective Loss 0.231952 LR 0.000500 Time 0.022247 +2023-10-05 21:38:41,877 - Epoch: [120][ 380/ 1236] Overall Loss 0.232144 Objective Loss 0.232144 LR 0.000500 Time 0.022195 +2023-10-05 21:38:42,078 - Epoch: [120][ 390/ 1236] Overall Loss 0.231682 Objective Loss 0.231682 LR 0.000500 Time 0.022140 +2023-10-05 21:38:42,281 - Epoch: [120][ 400/ 1236] Overall Loss 0.230899 Objective Loss 0.230899 LR 0.000500 Time 0.022093 +2023-10-05 21:38:42,482 - Epoch: [120][ 410/ 1236] Overall Loss 0.230958 Objective Loss 0.230958 LR 0.000500 Time 0.022044 +2023-10-05 21:38:42,685 - Epoch: [120][ 420/ 1236] Overall Loss 0.230910 Objective Loss 0.230910 LR 0.000500 Time 0.022001 +2023-10-05 21:38:42,886 - Epoch: [120][ 430/ 1236] Overall Loss 0.231087 Objective Loss 0.231087 LR 0.000500 Time 0.021955 +2023-10-05 21:38:43,089 - Epoch: [120][ 440/ 1236] Overall Loss 0.230870 Objective Loss 0.230870 LR 0.000500 Time 0.021916 +2023-10-05 21:38:43,290 - Epoch: [120][ 450/ 1236] Overall Loss 0.230696 Objective Loss 0.230696 LR 0.000500 Time 0.021875 +2023-10-05 21:38:43,493 - Epoch: [120][ 460/ 1236] Overall Loss 0.231212 Objective Loss 0.231212 LR 0.000500 Time 0.021841 +2023-10-05 21:38:43,694 - Epoch: [120][ 470/ 1236] Overall Loss 0.232147 Objective Loss 0.232147 LR 0.000500 Time 0.021803 +2023-10-05 21:38:43,897 - Epoch: [120][ 480/ 1236] Overall Loss 0.231814 Objective Loss 0.231814 LR 0.000500 Time 0.021770 +2023-10-05 21:38:44,098 - Epoch: [120][ 490/ 1236] Overall Loss 0.231794 Objective Loss 0.231794 LR 0.000500 Time 0.021735 +2023-10-05 21:38:44,305 - Epoch: [120][ 500/ 1236] Overall Loss 0.232052 Objective Loss 0.232052 LR 0.000500 Time 0.021715 +2023-10-05 21:38:44,507 - Epoch: [120][ 510/ 1236] Overall Loss 0.232171 Objective Loss 0.232171 LR 0.000500 Time 0.021684 +2023-10-05 21:38:44,710 - Epoch: [120][ 520/ 1236] Overall Loss 0.232350 Objective Loss 0.232350 LR 0.000500 Time 0.021656 +2023-10-05 21:38:44,914 - Epoch: [120][ 530/ 1236] Overall Loss 0.232247 Objective Loss 0.232247 LR 0.000500 Time 0.021633 +2023-10-05 21:38:45,118 - Epoch: [120][ 540/ 1236] Overall Loss 0.232191 Objective Loss 0.232191 LR 0.000500 Time 0.021608 +2023-10-05 21:38:45,319 - Epoch: [120][ 550/ 1236] Overall Loss 0.232163 Objective Loss 0.232163 LR 0.000500 Time 0.021581 +2023-10-05 21:38:45,525 - Epoch: [120][ 560/ 1236] Overall Loss 0.232115 Objective Loss 0.232115 LR 0.000500 Time 0.021562 +2023-10-05 21:38:45,726 - Epoch: [120][ 570/ 1236] Overall Loss 0.232129 Objective Loss 0.232129 LR 0.000500 Time 0.021537 +2023-10-05 21:38:45,929 - Epoch: [120][ 580/ 1236] Overall Loss 0.232218 Objective Loss 0.232218 LR 0.000500 Time 0.021515 +2023-10-05 21:38:46,130 - Epoch: [120][ 590/ 1236] Overall Loss 0.232288 Objective Loss 0.232288 LR 0.000500 Time 0.021490 +2023-10-05 21:38:46,333 - Epoch: [120][ 600/ 1236] Overall Loss 0.231847 Objective Loss 0.231847 LR 0.000500 Time 0.021470 +2023-10-05 21:38:46,534 - Epoch: [120][ 610/ 1236] Overall Loss 0.232394 Objective Loss 0.232394 LR 0.000500 Time 0.021447 +2023-10-05 21:38:46,737 - Epoch: [120][ 620/ 1236] Overall Loss 0.231976 Objective Loss 0.231976 LR 0.000500 Time 0.021428 +2023-10-05 21:38:46,938 - Epoch: [120][ 630/ 1236] Overall Loss 0.232136 Objective Loss 0.232136 LR 0.000500 Time 0.021406 +2023-10-05 21:38:47,141 - Epoch: [120][ 640/ 1236] Overall Loss 0.232220 Objective Loss 0.232220 LR 0.000500 Time 0.021389 +2023-10-05 21:38:47,342 - Epoch: [120][ 650/ 1236] Overall Loss 0.232542 Objective Loss 0.232542 LR 0.000500 Time 0.021368 +2023-10-05 21:38:47,545 - Epoch: [120][ 660/ 1236] Overall Loss 0.232642 Objective Loss 0.232642 LR 0.000500 Time 0.021352 +2023-10-05 21:38:47,747 - Epoch: [120][ 670/ 1236] Overall Loss 0.232807 Objective Loss 0.232807 LR 0.000500 Time 0.021333 +2023-10-05 21:38:47,949 - Epoch: [120][ 680/ 1236] Overall Loss 0.232853 Objective Loss 0.232853 LR 0.000500 Time 0.021317 +2023-10-05 21:38:48,151 - Epoch: [120][ 690/ 1236] Overall Loss 0.233339 Objective Loss 0.233339 LR 0.000500 Time 0.021299 +2023-10-05 21:38:48,353 - Epoch: [120][ 700/ 1236] Overall Loss 0.233300 Objective Loss 0.233300 LR 0.000500 Time 0.021284 +2023-10-05 21:38:48,555 - Epoch: [120][ 710/ 1236] Overall Loss 0.233024 Objective Loss 0.233024 LR 0.000500 Time 0.021267 +2023-10-05 21:38:48,758 - Epoch: [120][ 720/ 1236] Overall Loss 0.233154 Objective Loss 0.233154 LR 0.000500 Time 0.021253 +2023-10-05 21:38:48,959 - Epoch: [120][ 730/ 1236] Overall Loss 0.233042 Objective Loss 0.233042 LR 0.000500 Time 0.021238 +2023-10-05 21:38:49,162 - Epoch: [120][ 740/ 1236] Overall Loss 0.233069 Objective Loss 0.233069 LR 0.000500 Time 0.021224 +2023-10-05 21:38:49,363 - Epoch: [120][ 750/ 1236] Overall Loss 0.233251 Objective Loss 0.233251 LR 0.000500 Time 0.021209 +2023-10-05 21:38:49,566 - Epoch: [120][ 760/ 1236] Overall Loss 0.233008 Objective Loss 0.233008 LR 0.000500 Time 0.021196 +2023-10-05 21:38:49,767 - Epoch: [120][ 770/ 1236] Overall Loss 0.233359 Objective Loss 0.233359 LR 0.000500 Time 0.021182 +2023-10-05 21:38:49,970 - Epoch: [120][ 780/ 1236] Overall Loss 0.233298 Objective Loss 0.233298 LR 0.000500 Time 0.021170 +2023-10-05 21:38:50,171 - Epoch: [120][ 790/ 1236] Overall Loss 0.233900 Objective Loss 0.233900 LR 0.000500 Time 0.021156 +2023-10-05 21:38:50,373 - Epoch: [120][ 800/ 1236] Overall Loss 0.233726 Objective Loss 0.233726 LR 0.000500 Time 0.021144 +2023-10-05 21:38:50,575 - Epoch: [120][ 810/ 1236] Overall Loss 0.233972 Objective Loss 0.233972 LR 0.000500 Time 0.021131 +2023-10-05 21:38:50,778 - Epoch: [120][ 820/ 1236] Overall Loss 0.234132 Objective Loss 0.234132 LR 0.000500 Time 0.021121 +2023-10-05 21:38:50,979 - Epoch: [120][ 830/ 1236] Overall Loss 0.234276 Objective Loss 0.234276 LR 0.000500 Time 0.021108 +2023-10-05 21:38:51,182 - Epoch: [120][ 840/ 1236] Overall Loss 0.234336 Objective Loss 0.234336 LR 0.000500 Time 0.021098 +2023-10-05 21:38:51,383 - Epoch: [120][ 850/ 1236] Overall Loss 0.234496 Objective Loss 0.234496 LR 0.000500 Time 0.021086 +2023-10-05 21:38:51,585 - Epoch: [120][ 860/ 1236] Overall Loss 0.234395 Objective Loss 0.234395 LR 0.000500 Time 0.021076 +2023-10-05 21:38:51,786 - Epoch: [120][ 870/ 1236] Overall Loss 0.234241 Objective Loss 0.234241 LR 0.000500 Time 0.021064 +2023-10-05 21:38:51,989 - Epoch: [120][ 880/ 1236] Overall Loss 0.234340 Objective Loss 0.234340 LR 0.000500 Time 0.021055 +2023-10-05 21:38:52,190 - Epoch: [120][ 890/ 1236] Overall Loss 0.234501 Objective Loss 0.234501 LR 0.000500 Time 0.021044 +2023-10-05 21:38:52,393 - Epoch: [120][ 900/ 1236] Overall Loss 0.234527 Objective Loss 0.234527 LR 0.000500 Time 0.021036 +2023-10-05 21:38:52,595 - Epoch: [120][ 910/ 1236] Overall Loss 0.234648 Objective Loss 0.234648 LR 0.000500 Time 0.021025 +2023-10-05 21:38:52,797 - Epoch: [120][ 920/ 1236] Overall Loss 0.234656 Objective Loss 0.234656 LR 0.000500 Time 0.021017 +2023-10-05 21:38:52,998 - Epoch: [120][ 930/ 1236] Overall Loss 0.234888 Objective Loss 0.234888 LR 0.000500 Time 0.021006 +2023-10-05 21:38:53,202 - Epoch: [120][ 940/ 1236] Overall Loss 0.234669 Objective Loss 0.234669 LR 0.000500 Time 0.020999 +2023-10-05 21:38:53,403 - Epoch: [120][ 950/ 1236] Overall Loss 0.234691 Objective Loss 0.234691 LR 0.000500 Time 0.020989 +2023-10-05 21:38:53,606 - Epoch: [120][ 960/ 1236] Overall Loss 0.234612 Objective Loss 0.234612 LR 0.000500 Time 0.020981 +2023-10-05 21:38:53,807 - Epoch: [120][ 970/ 1236] Overall Loss 0.234730 Objective Loss 0.234730 LR 0.000500 Time 0.020973 +2023-10-05 21:38:54,010 - Epoch: [120][ 980/ 1236] Overall Loss 0.234938 Objective Loss 0.234938 LR 0.000500 Time 0.020965 +2023-10-05 21:38:54,211 - Epoch: [120][ 990/ 1236] Overall Loss 0.235189 Objective Loss 0.235189 LR 0.000500 Time 0.020956 +2023-10-05 21:38:54,414 - Epoch: [120][ 1000/ 1236] Overall Loss 0.234836 Objective Loss 0.234836 LR 0.000500 Time 0.020949 +2023-10-05 21:38:54,615 - Epoch: [120][ 1010/ 1236] Overall Loss 0.234747 Objective Loss 0.234747 LR 0.000500 Time 0.020940 +2023-10-05 21:38:54,818 - Epoch: [120][ 1020/ 1236] Overall Loss 0.234449 Objective Loss 0.234449 LR 0.000500 Time 0.020934 +2023-10-05 21:38:55,019 - Epoch: [120][ 1030/ 1236] Overall Loss 0.234484 Objective Loss 0.234484 LR 0.000500 Time 0.020925 +2023-10-05 21:38:55,222 - Epoch: [120][ 1040/ 1236] Overall Loss 0.234263 Objective Loss 0.234263 LR 0.000500 Time 0.020919 +2023-10-05 21:38:55,423 - Epoch: [120][ 1050/ 1236] Overall Loss 0.234266 Objective Loss 0.234266 LR 0.000500 Time 0.020911 +2023-10-05 21:38:55,626 - Epoch: [120][ 1060/ 1236] Overall Loss 0.234529 Objective Loss 0.234529 LR 0.000500 Time 0.020904 +2023-10-05 21:38:55,827 - Epoch: [120][ 1070/ 1236] Overall Loss 0.234666 Objective Loss 0.234666 LR 0.000500 Time 0.020897 +2023-10-05 21:38:56,030 - Epoch: [120][ 1080/ 1236] Overall Loss 0.234697 Objective Loss 0.234697 LR 0.000500 Time 0.020891 +2023-10-05 21:38:56,231 - Epoch: [120][ 1090/ 1236] Overall Loss 0.234695 Objective Loss 0.234695 LR 0.000500 Time 0.020884 +2023-10-05 21:38:56,434 - Epoch: [120][ 1100/ 1236] Overall Loss 0.234419 Objective Loss 0.234419 LR 0.000500 Time 0.020878 +2023-10-05 21:38:56,635 - Epoch: [120][ 1110/ 1236] Overall Loss 0.234697 Objective Loss 0.234697 LR 0.000500 Time 0.020871 +2023-10-05 21:38:56,838 - Epoch: [120][ 1120/ 1236] Overall Loss 0.234400 Objective Loss 0.234400 LR 0.000500 Time 0.020865 +2023-10-05 21:38:57,039 - Epoch: [120][ 1130/ 1236] Overall Loss 0.234277 Objective Loss 0.234277 LR 0.000500 Time 0.020858 +2023-10-05 21:38:57,242 - Epoch: [120][ 1140/ 1236] Overall Loss 0.234062 Objective Loss 0.234062 LR 0.000500 Time 0.020853 +2023-10-05 21:38:57,443 - Epoch: [120][ 1150/ 1236] Overall Loss 0.234201 Objective Loss 0.234201 LR 0.000500 Time 0.020846 +2023-10-05 21:38:57,646 - Epoch: [120][ 1160/ 1236] Overall Loss 0.234204 Objective Loss 0.234204 LR 0.000500 Time 0.020841 +2023-10-05 21:38:57,847 - Epoch: [120][ 1170/ 1236] Overall Loss 0.234152 Objective Loss 0.234152 LR 0.000500 Time 0.020834 +2023-10-05 21:38:58,049 - Epoch: [120][ 1180/ 1236] Overall Loss 0.233874 Objective Loss 0.233874 LR 0.000500 Time 0.020829 +2023-10-05 21:38:58,251 - Epoch: [120][ 1190/ 1236] Overall Loss 0.233643 Objective Loss 0.233643 LR 0.000500 Time 0.020823 +2023-10-05 21:38:58,454 - Epoch: [120][ 1200/ 1236] Overall Loss 0.233642 Objective Loss 0.233642 LR 0.000500 Time 0.020818 +2023-10-05 21:38:58,655 - Epoch: [120][ 1210/ 1236] Overall Loss 0.233582 Objective Loss 0.233582 LR 0.000500 Time 0.020812 +2023-10-05 21:38:58,858 - Epoch: [120][ 1220/ 1236] Overall Loss 0.233732 Objective Loss 0.233732 LR 0.000500 Time 0.020808 +2023-10-05 21:38:59,112 - Epoch: [120][ 1230/ 1236] Overall Loss 0.233696 Objective Loss 0.233696 LR 0.000500 Time 0.020845 +2023-10-05 21:38:59,230 - Epoch: [120][ 1236/ 1236] Overall Loss 0.233525 Objective Loss 0.233525 Top1 88.187373 Top5 99.185336 LR 0.000500 Time 0.020839 +2023-10-05 21:38:59,357 - --- validate (epoch=120)----------- +2023-10-05 21:38:59,357 - 29943 samples (256 per mini-batch) +2023-10-05 21:38:59,814 - Epoch: [120][ 10/ 117] Loss 0.337261 Top1 83.710938 Top5 97.656250 +2023-10-05 21:38:59,958 - Epoch: [120][ 20/ 117] Loss 0.323736 Top1 83.671875 Top5 97.929688 +2023-10-05 21:39:00,101 - Epoch: [120][ 30/ 117] Loss 0.326620 Top1 83.684896 Top5 98.046875 +2023-10-05 21:39:00,244 - Epoch: [120][ 40/ 117] Loss 0.332534 Top1 83.378906 Top5 98.027344 +2023-10-05 21:39:00,387 - Epoch: [120][ 50/ 117] Loss 0.333320 Top1 83.492188 Top5 97.992188 +2023-10-05 21:39:00,529 - Epoch: [120][ 60/ 117] Loss 0.338004 Top1 83.385417 Top5 97.903646 +2023-10-05 21:39:00,674 - Epoch: [120][ 70/ 117] Loss 0.333680 Top1 83.487723 Top5 97.963170 +2023-10-05 21:39:00,818 - Epoch: [120][ 80/ 117] Loss 0.334575 Top1 83.505859 Top5 97.924805 +2023-10-05 21:39:00,963 - Epoch: [120][ 90/ 117] Loss 0.332839 Top1 83.550347 Top5 97.955729 +2023-10-05 21:39:01,107 - Epoch: [120][ 100/ 117] Loss 0.331079 Top1 83.488281 Top5 97.925781 +2023-10-05 21:39:01,258 - Epoch: [120][ 110/ 117] Loss 0.332661 Top1 83.412642 Top5 97.904830 +2023-10-05 21:39:01,344 - Epoch: [120][ 117/ 117] Loss 0.333326 Top1 83.411816 Top5 97.932739 +2023-10-05 21:39:01,472 - ==> Top1: 83.412 Top5: 97.933 Loss: 0.333 + +2023-10-05 21:39:01,473 - ==> Confusion: +[[ 929 1 5 1 1 0 0 0 7 75 1 5 1 2 5 1 2 1 0 0 13] + [ 2 1064 3 0 1 19 1 15 5 0 1 0 0 0 1 4 3 0 6 0 6] + [ 5 1 946 16 0 0 34 7 0 4 0 1 12 1 1 2 3 1 10 5 7] + [ 3 1 22 958 0 4 0 2 3 0 5 0 5 6 29 7 2 6 21 1 14] + [ 38 7 0 0 960 5 0 1 0 4 1 2 2 3 8 3 8 2 1 3 2] + [ 6 40 1 2 3 989 1 13 3 2 4 8 1 13 6 1 4 0 3 3 13] + [ 0 2 21 1 0 0 1127 5 0 0 3 2 2 1 1 11 0 1 2 5 7] + [ 6 17 11 1 0 35 7 1059 2 5 4 10 3 1 0 2 1 2 39 4 9] + [ 18 0 1 0 1 3 0 0 979 40 9 2 4 11 12 3 1 0 2 1 2] + [ 111 0 3 0 2 5 0 0 33 914 0 0 0 28 2 7 0 0 0 5 9] + [ 3 4 10 5 0 3 4 5 19 2 957 4 0 16 5 3 1 0 3 1 8] + [ 1 0 0 1 1 16 1 0 0 0 0 957 19 6 0 2 4 17 0 8 2] + [ 1 1 4 2 0 2 1 0 2 0 1 41 969 3 2 8 2 19 2 2 6] + [ 3 0 3 0 2 8 0 0 9 8 2 5 2 1061 2 3 1 2 0 2 6] + [ 19 1 4 8 2 0 0 0 23 2 0 2 2 3 1015 0 0 1 9 0 10] + [ 0 1 4 0 3 1 1 0 0 0 0 13 7 2 1 1071 15 9 0 4 2] + [ 1 19 1 0 5 2 1 0 2 0 0 1 0 0 4 9 1104 0 1 2 9] + [ 0 0 0 3 0 1 0 0 2 0 0 0 17 1 1 4 0 1007 0 0 2] + [ 2 15 10 18 1 0 2 33 4 1 1 0 1 0 17 0 1 0 952 1 9] + [ 0 7 3 1 1 8 13 6 1 0 0 16 5 2 0 8 13 3 1 1061 3] + [ 146 205 152 69 98 158 44 103 123 99 176 101 357 361 152 73 173 91 117 210 4897]] + +2023-10-05 21:39:01,474 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:39:01,474 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:39:01,480 - + +2023-10-05 21:39:01,480 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:39:02,486 - Epoch: [121][ 10/ 1236] Overall Loss 0.209741 Objective Loss 0.209741 LR 0.000500 Time 0.100524 +2023-10-05 21:39:02,686 - Epoch: [121][ 20/ 1236] Overall Loss 0.218744 Objective Loss 0.218744 LR 0.000500 Time 0.060254 +2023-10-05 21:39:02,884 - Epoch: [121][ 30/ 1236] Overall Loss 0.232480 Objective Loss 0.232480 LR 0.000500 Time 0.046764 +2023-10-05 21:39:03,084 - Epoch: [121][ 40/ 1236] Overall Loss 0.231897 Objective Loss 0.231897 LR 0.000500 Time 0.040074 +2023-10-05 21:39:03,283 - Epoch: [121][ 50/ 1236] Overall Loss 0.229696 Objective Loss 0.229696 LR 0.000500 Time 0.036020 +2023-10-05 21:39:03,483 - Epoch: [121][ 60/ 1236] Overall Loss 0.229972 Objective Loss 0.229972 LR 0.000500 Time 0.033350 +2023-10-05 21:39:03,681 - Epoch: [121][ 70/ 1236] Overall Loss 0.234473 Objective Loss 0.234473 LR 0.000500 Time 0.031410 +2023-10-05 21:39:03,881 - Epoch: [121][ 80/ 1236] Overall Loss 0.233437 Objective Loss 0.233437 LR 0.000500 Time 0.029985 +2023-10-05 21:39:04,080 - Epoch: [121][ 90/ 1236] Overall Loss 0.235966 Objective Loss 0.235966 LR 0.000500 Time 0.028853 +2023-10-05 21:39:04,280 - Epoch: [121][ 100/ 1236] Overall Loss 0.234844 Objective Loss 0.234844 LR 0.000500 Time 0.027968 +2023-10-05 21:39:04,478 - Epoch: [121][ 110/ 1236] Overall Loss 0.233811 Objective Loss 0.233811 LR 0.000500 Time 0.027224 +2023-10-05 21:39:04,679 - Epoch: [121][ 120/ 1236] Overall Loss 0.229481 Objective Loss 0.229481 LR 0.000500 Time 0.026624 +2023-10-05 21:39:04,877 - Epoch: [121][ 130/ 1236] Overall Loss 0.230654 Objective Loss 0.230654 LR 0.000500 Time 0.026096 +2023-10-05 21:39:05,077 - Epoch: [121][ 140/ 1236] Overall Loss 0.230537 Objective Loss 0.230537 LR 0.000500 Time 0.025661 +2023-10-05 21:39:05,276 - Epoch: [121][ 150/ 1236] Overall Loss 0.230299 Objective Loss 0.230299 LR 0.000500 Time 0.025270 +2023-10-05 21:39:05,476 - Epoch: [121][ 160/ 1236] Overall Loss 0.231379 Objective Loss 0.231379 LR 0.000500 Time 0.024941 +2023-10-05 21:39:05,674 - Epoch: [121][ 170/ 1236] Overall Loss 0.230137 Objective Loss 0.230137 LR 0.000500 Time 0.024637 +2023-10-05 21:39:05,874 - Epoch: [121][ 180/ 1236] Overall Loss 0.229898 Objective Loss 0.229898 LR 0.000500 Time 0.024379 +2023-10-05 21:39:06,072 - Epoch: [121][ 190/ 1236] Overall Loss 0.231215 Objective Loss 0.231215 LR 0.000500 Time 0.024137 +2023-10-05 21:39:06,273 - Epoch: [121][ 200/ 1236] Overall Loss 0.230670 Objective Loss 0.230670 LR 0.000500 Time 0.023931 +2023-10-05 21:39:06,477 - Epoch: [121][ 210/ 1236] Overall Loss 0.230968 Objective Loss 0.230968 LR 0.000500 Time 0.023759 +2023-10-05 21:39:06,679 - Epoch: [121][ 220/ 1236] Overall Loss 0.230744 Objective Loss 0.230744 LR 0.000500 Time 0.023598 +2023-10-05 21:39:06,883 - Epoch: [121][ 230/ 1236] Overall Loss 0.230402 Objective Loss 0.230402 LR 0.000500 Time 0.023455 +2023-10-05 21:39:07,086 - Epoch: [121][ 240/ 1236] Overall Loss 0.231244 Objective Loss 0.231244 LR 0.000500 Time 0.023320 +2023-10-05 21:39:07,289 - Epoch: [121][ 250/ 1236] Overall Loss 0.230361 Objective Loss 0.230361 LR 0.000500 Time 0.023201 +2023-10-05 21:39:07,493 - Epoch: [121][ 260/ 1236] Overall Loss 0.229176 Objective Loss 0.229176 LR 0.000500 Time 0.023092 +2023-10-05 21:39:07,697 - Epoch: [121][ 270/ 1236] Overall Loss 0.229277 Objective Loss 0.229277 LR 0.000500 Time 0.022990 +2023-10-05 21:39:07,901 - Epoch: [121][ 280/ 1236] Overall Loss 0.229333 Objective Loss 0.229333 LR 0.000500 Time 0.022895 +2023-10-05 21:39:08,105 - Epoch: [121][ 290/ 1236] Overall Loss 0.230243 Objective Loss 0.230243 LR 0.000500 Time 0.022806 +2023-10-05 21:39:08,309 - Epoch: [121][ 300/ 1236] Overall Loss 0.229669 Objective Loss 0.229669 LR 0.000500 Time 0.022725 +2023-10-05 21:39:08,513 - Epoch: [121][ 310/ 1236] Overall Loss 0.229208 Objective Loss 0.229208 LR 0.000500 Time 0.022649 +2023-10-05 21:39:08,717 - Epoch: [121][ 320/ 1236] Overall Loss 0.229574 Objective Loss 0.229574 LR 0.000500 Time 0.022579 +2023-10-05 21:39:08,921 - Epoch: [121][ 330/ 1236] Overall Loss 0.230038 Objective Loss 0.230038 LR 0.000500 Time 0.022511 +2023-10-05 21:39:09,125 - Epoch: [121][ 340/ 1236] Overall Loss 0.230042 Objective Loss 0.230042 LR 0.000500 Time 0.022446 +2023-10-05 21:39:09,329 - Epoch: [121][ 350/ 1236] Overall Loss 0.230680 Objective Loss 0.230680 LR 0.000500 Time 0.022387 +2023-10-05 21:39:09,533 - Epoch: [121][ 360/ 1236] Overall Loss 0.230497 Objective Loss 0.230497 LR 0.000500 Time 0.022331 +2023-10-05 21:39:09,737 - Epoch: [121][ 370/ 1236] Overall Loss 0.230209 Objective Loss 0.230209 LR 0.000500 Time 0.022277 +2023-10-05 21:39:09,941 - Epoch: [121][ 380/ 1236] Overall Loss 0.230162 Objective Loss 0.230162 LR 0.000500 Time 0.022227 +2023-10-05 21:39:10,146 - Epoch: [121][ 390/ 1236] Overall Loss 0.230089 Objective Loss 0.230089 LR 0.000500 Time 0.022180 +2023-10-05 21:39:10,349 - Epoch: [121][ 400/ 1236] Overall Loss 0.230051 Objective Loss 0.230051 LR 0.000500 Time 0.022134 +2023-10-05 21:39:10,554 - Epoch: [121][ 410/ 1236] Overall Loss 0.229988 Objective Loss 0.229988 LR 0.000500 Time 0.022092 +2023-10-05 21:39:10,758 - Epoch: [121][ 420/ 1236] Overall Loss 0.229549 Objective Loss 0.229549 LR 0.000500 Time 0.022050 +2023-10-05 21:39:10,962 - Epoch: [121][ 430/ 1236] Overall Loss 0.229322 Objective Loss 0.229322 LR 0.000500 Time 0.022012 +2023-10-05 21:39:11,166 - Epoch: [121][ 440/ 1236] Overall Loss 0.229910 Objective Loss 0.229910 LR 0.000500 Time 0.021973 +2023-10-05 21:39:11,370 - Epoch: [121][ 450/ 1236] Overall Loss 0.230111 Objective Loss 0.230111 LR 0.000500 Time 0.021938 +2023-10-05 21:39:11,574 - Epoch: [121][ 460/ 1236] Overall Loss 0.230501 Objective Loss 0.230501 LR 0.000500 Time 0.021903 +2023-10-05 21:39:11,778 - Epoch: [121][ 470/ 1236] Overall Loss 0.230348 Objective Loss 0.230348 LR 0.000500 Time 0.021871 +2023-10-05 21:39:11,982 - Epoch: [121][ 480/ 1236] Overall Loss 0.230101 Objective Loss 0.230101 LR 0.000500 Time 0.021839 +2023-10-05 21:39:12,186 - Epoch: [121][ 490/ 1236] Overall Loss 0.230129 Objective Loss 0.230129 LR 0.000500 Time 0.021809 +2023-10-05 21:39:12,390 - Epoch: [121][ 500/ 1236] Overall Loss 0.230048 Objective Loss 0.230048 LR 0.000500 Time 0.021780 +2023-10-05 21:39:12,595 - Epoch: [121][ 510/ 1236] Overall Loss 0.229980 Objective Loss 0.229980 LR 0.000500 Time 0.021753 +2023-10-05 21:39:12,798 - Epoch: [121][ 520/ 1236] Overall Loss 0.229808 Objective Loss 0.229808 LR 0.000500 Time 0.021726 +2023-10-05 21:39:13,003 - Epoch: [121][ 530/ 1236] Overall Loss 0.230136 Objective Loss 0.230136 LR 0.000500 Time 0.021700 +2023-10-05 21:39:13,207 - Epoch: [121][ 540/ 1236] Overall Loss 0.230100 Objective Loss 0.230100 LR 0.000500 Time 0.021675 +2023-10-05 21:39:13,411 - Epoch: [121][ 550/ 1236] Overall Loss 0.230183 Objective Loss 0.230183 LR 0.000500 Time 0.021652 +2023-10-05 21:39:13,615 - Epoch: [121][ 560/ 1236] Overall Loss 0.230137 Objective Loss 0.230137 LR 0.000500 Time 0.021628 +2023-10-05 21:39:13,819 - Epoch: [121][ 570/ 1236] Overall Loss 0.230580 Objective Loss 0.230580 LR 0.000500 Time 0.021607 +2023-10-05 21:39:14,023 - Epoch: [121][ 580/ 1236] Overall Loss 0.230742 Objective Loss 0.230742 LR 0.000500 Time 0.021584 +2023-10-05 21:39:14,227 - Epoch: [121][ 590/ 1236] Overall Loss 0.230875 Objective Loss 0.230875 LR 0.000500 Time 0.021564 +2023-10-05 21:39:14,432 - Epoch: [121][ 600/ 1236] Overall Loss 0.230796 Objective Loss 0.230796 LR 0.000500 Time 0.021546 +2023-10-05 21:39:14,637 - Epoch: [121][ 610/ 1236] Overall Loss 0.231077 Objective Loss 0.231077 LR 0.000500 Time 0.021528 +2023-10-05 21:39:14,843 - Epoch: [121][ 620/ 1236] Overall Loss 0.230780 Objective Loss 0.230780 LR 0.000500 Time 0.021512 +2023-10-05 21:39:15,048 - Epoch: [121][ 630/ 1236] Overall Loss 0.230787 Objective Loss 0.230787 LR 0.000500 Time 0.021496 +2023-10-05 21:39:15,254 - Epoch: [121][ 640/ 1236] Overall Loss 0.230835 Objective Loss 0.230835 LR 0.000500 Time 0.021481 +2023-10-05 21:39:15,459 - Epoch: [121][ 650/ 1236] Overall Loss 0.230839 Objective Loss 0.230839 LR 0.000500 Time 0.021465 +2023-10-05 21:39:15,665 - Epoch: [121][ 660/ 1236] Overall Loss 0.230475 Objective Loss 0.230475 LR 0.000500 Time 0.021451 +2023-10-05 21:39:15,870 - Epoch: [121][ 670/ 1236] Overall Loss 0.230609 Objective Loss 0.230609 LR 0.000500 Time 0.021436 +2023-10-05 21:39:16,076 - Epoch: [121][ 680/ 1236] Overall Loss 0.230941 Objective Loss 0.230941 LR 0.000500 Time 0.021424 +2023-10-05 21:39:16,281 - Epoch: [121][ 690/ 1236] Overall Loss 0.230691 Objective Loss 0.230691 LR 0.000500 Time 0.021410 +2023-10-05 21:39:16,487 - Epoch: [121][ 700/ 1236] Overall Loss 0.230305 Objective Loss 0.230305 LR 0.000500 Time 0.021397 +2023-10-05 21:39:16,692 - Epoch: [121][ 710/ 1236] Overall Loss 0.230527 Objective Loss 0.230527 LR 0.000500 Time 0.021384 +2023-10-05 21:39:16,898 - Epoch: [121][ 720/ 1236] Overall Loss 0.230585 Objective Loss 0.230585 LR 0.000500 Time 0.021372 +2023-10-05 21:39:17,103 - Epoch: [121][ 730/ 1236] Overall Loss 0.230607 Objective Loss 0.230607 LR 0.000500 Time 0.021360 +2023-10-05 21:39:17,309 - Epoch: [121][ 740/ 1236] Overall Loss 0.230512 Objective Loss 0.230512 LR 0.000500 Time 0.021348 +2023-10-05 21:39:17,514 - Epoch: [121][ 750/ 1236] Overall Loss 0.230828 Objective Loss 0.230828 LR 0.000500 Time 0.021337 +2023-10-05 21:39:17,720 - Epoch: [121][ 760/ 1236] Overall Loss 0.230911 Objective Loss 0.230911 LR 0.000500 Time 0.021326 +2023-10-05 21:39:17,925 - Epoch: [121][ 770/ 1236] Overall Loss 0.230925 Objective Loss 0.230925 LR 0.000500 Time 0.021315 +2023-10-05 21:39:18,130 - Epoch: [121][ 780/ 1236] Overall Loss 0.231025 Objective Loss 0.231025 LR 0.000500 Time 0.021305 +2023-10-05 21:39:18,335 - Epoch: [121][ 790/ 1236] Overall Loss 0.231063 Objective Loss 0.231063 LR 0.000500 Time 0.021294 +2023-10-05 21:39:18,541 - Epoch: [121][ 800/ 1236] Overall Loss 0.231411 Objective Loss 0.231411 LR 0.000500 Time 0.021284 +2023-10-05 21:39:18,746 - Epoch: [121][ 810/ 1236] Overall Loss 0.231269 Objective Loss 0.231269 LR 0.000500 Time 0.021274 +2023-10-05 21:39:18,952 - Epoch: [121][ 820/ 1236] Overall Loss 0.231082 Objective Loss 0.231082 LR 0.000500 Time 0.021265 +2023-10-05 21:39:19,157 - Epoch: [121][ 830/ 1236] Overall Loss 0.231016 Objective Loss 0.231016 LR 0.000500 Time 0.021256 +2023-10-05 21:39:19,363 - Epoch: [121][ 840/ 1236] Overall Loss 0.231452 Objective Loss 0.231452 LR 0.000500 Time 0.021247 +2023-10-05 21:39:19,568 - Epoch: [121][ 850/ 1236] Overall Loss 0.231431 Objective Loss 0.231431 LR 0.000500 Time 0.021238 +2023-10-05 21:39:19,774 - Epoch: [121][ 860/ 1236] Overall Loss 0.231544 Objective Loss 0.231544 LR 0.000500 Time 0.021229 +2023-10-05 21:39:19,979 - Epoch: [121][ 870/ 1236] Overall Loss 0.231529 Objective Loss 0.231529 LR 0.000500 Time 0.021220 +2023-10-05 21:39:20,185 - Epoch: [121][ 880/ 1236] Overall Loss 0.231642 Objective Loss 0.231642 LR 0.000500 Time 0.021213 +2023-10-05 21:39:20,390 - Epoch: [121][ 890/ 1236] Overall Loss 0.231599 Objective Loss 0.231599 LR 0.000500 Time 0.021204 +2023-10-05 21:39:20,595 - Epoch: [121][ 900/ 1236] Overall Loss 0.231772 Objective Loss 0.231772 LR 0.000500 Time 0.021197 +2023-10-05 21:39:20,800 - Epoch: [121][ 910/ 1236] Overall Loss 0.231901 Objective Loss 0.231901 LR 0.000500 Time 0.021189 +2023-10-05 21:39:21,006 - Epoch: [121][ 920/ 1236] Overall Loss 0.232042 Objective Loss 0.232042 LR 0.000500 Time 0.021182 +2023-10-05 21:39:21,211 - Epoch: [121][ 930/ 1236] Overall Loss 0.232153 Objective Loss 0.232153 LR 0.000500 Time 0.021174 +2023-10-05 21:39:21,417 - Epoch: [121][ 940/ 1236] Overall Loss 0.232232 Objective Loss 0.232232 LR 0.000500 Time 0.021167 +2023-10-05 21:39:21,622 - Epoch: [121][ 950/ 1236] Overall Loss 0.232157 Objective Loss 0.232157 LR 0.000500 Time 0.021159 +2023-10-05 21:39:21,827 - Epoch: [121][ 960/ 1236] Overall Loss 0.232029 Objective Loss 0.232029 LR 0.000500 Time 0.021153 +2023-10-05 21:39:22,032 - Epoch: [121][ 970/ 1236] Overall Loss 0.231985 Objective Loss 0.231985 LR 0.000500 Time 0.021145 +2023-10-05 21:39:22,238 - Epoch: [121][ 980/ 1236] Overall Loss 0.232007 Objective Loss 0.232007 LR 0.000500 Time 0.021139 +2023-10-05 21:39:22,443 - Epoch: [121][ 990/ 1236] Overall Loss 0.232282 Objective Loss 0.232282 LR 0.000500 Time 0.021132 +2023-10-05 21:39:22,649 - Epoch: [121][ 1000/ 1236] Overall Loss 0.232239 Objective Loss 0.232239 LR 0.000500 Time 0.021126 +2023-10-05 21:39:22,854 - Epoch: [121][ 1010/ 1236] Overall Loss 0.232140 Objective Loss 0.232140 LR 0.000500 Time 0.021120 +2023-10-05 21:39:23,060 - Epoch: [121][ 1020/ 1236] Overall Loss 0.232186 Objective Loss 0.232186 LR 0.000500 Time 0.021114 +2023-10-05 21:39:23,265 - Epoch: [121][ 1030/ 1236] Overall Loss 0.232204 Objective Loss 0.232204 LR 0.000500 Time 0.021108 +2023-10-05 21:39:23,471 - Epoch: [121][ 1040/ 1236] Overall Loss 0.232202 Objective Loss 0.232202 LR 0.000500 Time 0.021102 +2023-10-05 21:39:23,676 - Epoch: [121][ 1050/ 1236] Overall Loss 0.232125 Objective Loss 0.232125 LR 0.000500 Time 0.021096 +2023-10-05 21:39:23,882 - Epoch: [121][ 1060/ 1236] Overall Loss 0.232341 Objective Loss 0.232341 LR 0.000500 Time 0.021091 +2023-10-05 21:39:24,087 - Epoch: [121][ 1070/ 1236] Overall Loss 0.232320 Objective Loss 0.232320 LR 0.000500 Time 0.021085 +2023-10-05 21:39:24,292 - Epoch: [121][ 1080/ 1236] Overall Loss 0.232410 Objective Loss 0.232410 LR 0.000500 Time 0.021080 +2023-10-05 21:39:24,497 - Epoch: [121][ 1090/ 1236] Overall Loss 0.232618 Objective Loss 0.232618 LR 0.000500 Time 0.021074 +2023-10-05 21:39:24,703 - Epoch: [121][ 1100/ 1236] Overall Loss 0.232673 Objective Loss 0.232673 LR 0.000500 Time 0.021069 +2023-10-05 21:39:24,908 - Epoch: [121][ 1110/ 1236] Overall Loss 0.232600 Objective Loss 0.232600 LR 0.000500 Time 0.021063 +2023-10-05 21:39:25,114 - Epoch: [121][ 1120/ 1236] Overall Loss 0.232534 Objective Loss 0.232534 LR 0.000500 Time 0.021058 +2023-10-05 21:39:25,318 - Epoch: [121][ 1130/ 1236] Overall Loss 0.232490 Objective Loss 0.232490 LR 0.000500 Time 0.021053 +2023-10-05 21:39:25,524 - Epoch: [121][ 1140/ 1236] Overall Loss 0.232695 Objective Loss 0.232695 LR 0.000500 Time 0.021048 +2023-10-05 21:39:25,729 - Epoch: [121][ 1150/ 1236] Overall Loss 0.232957 Objective Loss 0.232957 LR 0.000500 Time 0.021043 +2023-10-05 21:39:25,935 - Epoch: [121][ 1160/ 1236] Overall Loss 0.233018 Objective Loss 0.233018 LR 0.000500 Time 0.021039 +2023-10-05 21:39:26,140 - Epoch: [121][ 1170/ 1236] Overall Loss 0.232959 Objective Loss 0.232959 LR 0.000500 Time 0.021034 +2023-10-05 21:39:26,346 - Epoch: [121][ 1180/ 1236] Overall Loss 0.232842 Objective Loss 0.232842 LR 0.000500 Time 0.021030 +2023-10-05 21:39:26,551 - Epoch: [121][ 1190/ 1236] Overall Loss 0.232791 Objective Loss 0.232791 LR 0.000500 Time 0.021025 +2023-10-05 21:39:26,757 - Epoch: [121][ 1200/ 1236] Overall Loss 0.232799 Objective Loss 0.232799 LR 0.000500 Time 0.021021 +2023-10-05 21:39:26,962 - Epoch: [121][ 1210/ 1236] Overall Loss 0.232797 Objective Loss 0.232797 LR 0.000500 Time 0.021016 +2023-10-05 21:39:27,168 - Epoch: [121][ 1220/ 1236] Overall Loss 0.232911 Objective Loss 0.232911 LR 0.000500 Time 0.021012 +2023-10-05 21:39:27,424 - Epoch: [121][ 1230/ 1236] Overall Loss 0.232900 Objective Loss 0.232900 LR 0.000500 Time 0.021050 +2023-10-05 21:39:27,543 - Epoch: [121][ 1236/ 1236] Overall Loss 0.232842 Objective Loss 0.232842 Top1 87.169043 Top5 98.370672 LR 0.000500 Time 0.021043 +2023-10-05 21:39:27,663 - --- validate (epoch=121)----------- +2023-10-05 21:39:27,663 - 29943 samples (256 per mini-batch) +2023-10-05 21:39:28,120 - Epoch: [121][ 10/ 117] Loss 0.354123 Top1 83.671875 Top5 97.929688 +2023-10-05 21:39:28,267 - Epoch: [121][ 20/ 117] Loss 0.346412 Top1 83.984375 Top5 98.066406 +2023-10-05 21:39:28,412 - Epoch: [121][ 30/ 117] Loss 0.341233 Top1 83.893229 Top5 98.046875 +2023-10-05 21:39:28,558 - Epoch: [121][ 40/ 117] Loss 0.339353 Top1 84.023438 Top5 97.949219 +2023-10-05 21:39:28,702 - Epoch: [121][ 50/ 117] Loss 0.337799 Top1 83.945312 Top5 97.914062 +2023-10-05 21:39:28,851 - Epoch: [121][ 60/ 117] Loss 0.328857 Top1 84.075521 Top5 97.949219 +2023-10-05 21:39:28,995 - Epoch: [121][ 70/ 117] Loss 0.324555 Top1 84.162946 Top5 97.929688 +2023-10-05 21:39:29,139 - Epoch: [121][ 80/ 117] Loss 0.332458 Top1 83.994141 Top5 97.880859 +2023-10-05 21:39:29,283 - Epoch: [121][ 90/ 117] Loss 0.330178 Top1 84.079861 Top5 97.929688 +2023-10-05 21:39:29,429 - Epoch: [121][ 100/ 117] Loss 0.328757 Top1 84.140625 Top5 97.937500 +2023-10-05 21:39:29,580 - Epoch: [121][ 110/ 117] Loss 0.330260 Top1 84.112216 Top5 97.908381 +2023-10-05 21:39:29,666 - Epoch: [121][ 117/ 117] Loss 0.332558 Top1 84.086431 Top5 97.885983 +2023-10-05 21:39:29,788 - ==> Top1: 84.086 Top5: 97.886 Loss: 0.333 + +2023-10-05 21:39:29,789 - ==> Confusion: +[[ 904 1 2 3 12 1 0 0 5 93 1 1 1 2 3 0 7 1 1 0 12] + [ 1 1045 2 1 8 18 1 18 5 0 1 3 0 0 0 2 7 0 8 3 8] + [ 3 1 948 7 1 0 23 16 0 0 6 3 8 4 1 5 1 3 7 5 14] + [ 2 1 20 925 3 4 2 3 5 1 13 0 5 4 34 4 2 6 27 2 26] + [ 24 7 0 0 976 4 0 0 0 6 2 3 1 1 5 2 9 1 0 3 6] + [ 3 36 1 1 2 974 0 16 1 6 8 13 0 20 6 1 5 0 2 2 19] + [ 0 6 28 1 0 1 1104 9 0 0 4 3 1 0 1 11 0 0 3 5 14] + [ 3 15 18 0 0 24 5 1061 2 5 4 14 1 2 0 3 1 0 41 7 12] + [ 15 4 1 0 0 3 0 0 965 47 13 3 2 15 12 4 0 0 4 0 1] + [ 81 0 2 0 6 3 1 0 23 966 0 5 0 18 2 2 1 0 0 3 6] + [ 2 4 9 6 1 0 1 7 9 2 966 5 0 17 4 1 2 1 4 3 9] + [ 1 0 0 0 0 11 0 1 0 1 0 969 17 7 0 3 2 11 0 5 7] + [ 0 2 3 2 0 3 0 3 3 0 2 41 960 6 3 7 3 13 4 2 11] + [ 2 0 0 0 4 6 0 0 4 17 2 6 1 1061 2 2 1 1 0 1 9] + [ 13 6 3 4 5 0 0 0 26 4 3 2 2 1 1003 0 0 1 16 0 12] + [ 0 2 2 0 3 1 1 0 0 0 0 9 3 3 1 1072 16 9 0 8 4] + [ 1 4 0 0 7 5 0 2 2 0 0 4 0 1 2 10 1109 0 1 2 11] + [ 0 1 0 1 0 0 1 0 0 1 0 3 23 1 0 7 2 991 1 1 5] + [ 1 12 10 11 0 0 1 22 1 2 3 2 1 1 9 0 3 0 976 2 11] + [ 0 2 6 1 1 4 9 9 1 0 1 19 3 5 0 9 10 2 5 1054 11] + [ 129 169 137 39 111 123 43 110 116 110 176 144 289 319 106 55 185 62 133 200 5149]] + +2023-10-05 21:39:29,790 - ==> Best [Top1: 84.116 Top5: 97.959 Sparsity:0.00 Params: 148928 on epoch: 114] +2023-10-05 21:39:29,790 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:39:29,796 - + +2023-10-05 21:39:29,796 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:39:30,792 - Epoch: [122][ 10/ 1236] Overall Loss 0.240705 Objective Loss 0.240705 LR 0.000500 Time 0.099519 +2023-10-05 21:39:30,993 - Epoch: [122][ 20/ 1236] Overall Loss 0.232597 Objective Loss 0.232597 LR 0.000500 Time 0.059807 +2023-10-05 21:39:31,194 - Epoch: [122][ 30/ 1236] Overall Loss 0.226905 Objective Loss 0.226905 LR 0.000500 Time 0.046566 +2023-10-05 21:39:31,398 - Epoch: [122][ 40/ 1236] Overall Loss 0.231580 Objective Loss 0.231580 LR 0.000500 Time 0.040014 +2023-10-05 21:39:31,599 - Epoch: [122][ 50/ 1236] Overall Loss 0.228578 Objective Loss 0.228578 LR 0.000500 Time 0.036026 +2023-10-05 21:39:31,803 - Epoch: [122][ 60/ 1236] Overall Loss 0.225059 Objective Loss 0.225059 LR 0.000500 Time 0.033423 +2023-10-05 21:39:32,004 - Epoch: [122][ 70/ 1236] Overall Loss 0.229030 Objective Loss 0.229030 LR 0.000500 Time 0.031510 +2023-10-05 21:39:32,206 - Epoch: [122][ 80/ 1236] Overall Loss 0.225221 Objective Loss 0.225221 LR 0.000500 Time 0.030091 +2023-10-05 21:39:32,407 - Epoch: [122][ 90/ 1236] Overall Loss 0.224638 Objective Loss 0.224638 LR 0.000500 Time 0.028973 +2023-10-05 21:39:32,610 - Epoch: [122][ 100/ 1236] Overall Loss 0.226096 Objective Loss 0.226096 LR 0.000500 Time 0.028106 +2023-10-05 21:39:32,809 - Epoch: [122][ 110/ 1236] Overall Loss 0.225359 Objective Loss 0.225359 LR 0.000500 Time 0.027358 +2023-10-05 21:39:33,014 - Epoch: [122][ 120/ 1236] Overall Loss 0.227101 Objective Loss 0.227101 LR 0.000500 Time 0.026778 +2023-10-05 21:39:33,214 - Epoch: [122][ 130/ 1236] Overall Loss 0.226107 Objective Loss 0.226107 LR 0.000500 Time 0.026257 +2023-10-05 21:39:33,418 - Epoch: [122][ 140/ 1236] Overall Loss 0.226560 Objective Loss 0.226560 LR 0.000500 Time 0.025839 +2023-10-05 21:39:33,619 - Epoch: [122][ 150/ 1236] Overall Loss 0.225747 Objective Loss 0.225747 LR 0.000500 Time 0.025452 +2023-10-05 21:39:33,823 - Epoch: [122][ 160/ 1236] Overall Loss 0.226001 Objective Loss 0.226001 LR 0.000500 Time 0.025133 +2023-10-05 21:39:34,023 - Epoch: [122][ 170/ 1236] Overall Loss 0.226699 Objective Loss 0.226699 LR 0.000500 Time 0.024831 +2023-10-05 21:39:34,228 - Epoch: [122][ 180/ 1236] Overall Loss 0.227065 Objective Loss 0.227065 LR 0.000500 Time 0.024585 +2023-10-05 21:39:34,429 - Epoch: [122][ 190/ 1236] Overall Loss 0.227802 Objective Loss 0.227802 LR 0.000500 Time 0.024347 +2023-10-05 21:39:34,633 - Epoch: [122][ 200/ 1236] Overall Loss 0.229254 Objective Loss 0.229254 LR 0.000500 Time 0.024148 +2023-10-05 21:39:34,834 - Epoch: [122][ 210/ 1236] Overall Loss 0.228052 Objective Loss 0.228052 LR 0.000500 Time 0.023953 +2023-10-05 21:39:35,038 - Epoch: [122][ 220/ 1236] Overall Loss 0.228822 Objective Loss 0.228822 LR 0.000500 Time 0.023790 +2023-10-05 21:39:35,238 - Epoch: [122][ 230/ 1236] Overall Loss 0.228967 Objective Loss 0.228967 LR 0.000500 Time 0.023626 +2023-10-05 21:39:35,441 - Epoch: [122][ 240/ 1236] Overall Loss 0.228248 Objective Loss 0.228248 LR 0.000500 Time 0.023484 +2023-10-05 21:39:35,641 - Epoch: [122][ 250/ 1236] Overall Loss 0.226116 Objective Loss 0.226116 LR 0.000500 Time 0.023346 +2023-10-05 21:39:35,846 - Epoch: [122][ 260/ 1236] Overall Loss 0.226924 Objective Loss 0.226924 LR 0.000500 Time 0.023232 +2023-10-05 21:39:36,046 - Epoch: [122][ 270/ 1236] Overall Loss 0.226422 Objective Loss 0.226422 LR 0.000500 Time 0.023113 +2023-10-05 21:39:36,250 - Epoch: [122][ 280/ 1236] Overall Loss 0.226670 Objective Loss 0.226670 LR 0.000500 Time 0.023016 +2023-10-05 21:39:36,451 - Epoch: [122][ 290/ 1236] Overall Loss 0.226784 Objective Loss 0.226784 LR 0.000500 Time 0.022914 +2023-10-05 21:39:36,656 - Epoch: [122][ 300/ 1236] Overall Loss 0.226504 Objective Loss 0.226504 LR 0.000500 Time 0.022830 +2023-10-05 21:39:36,856 - Epoch: [122][ 310/ 1236] Overall Loss 0.226247 Objective Loss 0.226247 LR 0.000500 Time 0.022739 +2023-10-05 21:39:37,061 - Epoch: [122][ 320/ 1236] Overall Loss 0.226120 Objective Loss 0.226120 LR 0.000500 Time 0.022666 +2023-10-05 21:39:37,259 - Epoch: [122][ 330/ 1236] Overall Loss 0.226413 Objective Loss 0.226413 LR 0.000500 Time 0.022580 +2023-10-05 21:39:37,464 - Epoch: [122][ 340/ 1236] Overall Loss 0.226488 Objective Loss 0.226488 LR 0.000500 Time 0.022516 +2023-10-05 21:39:37,665 - Epoch: [122][ 350/ 1236] Overall Loss 0.226348 Objective Loss 0.226348 LR 0.000500 Time 0.022447 +2023-10-05 21:39:37,869 - Epoch: [122][ 360/ 1236] Overall Loss 0.226465 Objective Loss 0.226465 LR 0.000500 Time 0.022388 +2023-10-05 21:39:38,070 - Epoch: [122][ 370/ 1236] Overall Loss 0.227076 Objective Loss 0.227076 LR 0.000500 Time 0.022327 +2023-10-05 21:39:38,274 - Epoch: [122][ 380/ 1236] Overall Loss 0.227104 Objective Loss 0.227104 LR 0.000500 Time 0.022275 +2023-10-05 21:39:38,473 - Epoch: [122][ 390/ 1236] Overall Loss 0.227145 Objective Loss 0.227145 LR 0.000500 Time 0.022213 +2023-10-05 21:39:38,678 - Epoch: [122][ 400/ 1236] Overall Loss 0.227494 Objective Loss 0.227494 LR 0.000500 Time 0.022168 +2023-10-05 21:39:38,879 - Epoch: [122][ 410/ 1236] Overall Loss 0.227204 Objective Loss 0.227204 LR 0.000500 Time 0.022117 +2023-10-05 21:39:39,082 - Epoch: [122][ 420/ 1236] Overall Loss 0.227454 Objective Loss 0.227454 LR 0.000500 Time 0.022075 +2023-10-05 21:39:39,281 - Epoch: [122][ 430/ 1236] Overall Loss 0.228372 Objective Loss 0.228372 LR 0.000500 Time 0.022023 +2023-10-05 21:39:39,487 - Epoch: [122][ 440/ 1236] Overall Loss 0.227753 Objective Loss 0.227753 LR 0.000500 Time 0.021989 +2023-10-05 21:39:39,688 - Epoch: [122][ 450/ 1236] Overall Loss 0.227758 Objective Loss 0.227758 LR 0.000500 Time 0.021947 +2023-10-05 21:39:39,890 - Epoch: [122][ 460/ 1236] Overall Loss 0.228088 Objective Loss 0.228088 LR 0.000500 Time 0.021907 +2023-10-05 21:39:40,090 - Epoch: [122][ 470/ 1236] Overall Loss 0.227778 Objective Loss 0.227778 LR 0.000500 Time 0.021867 +2023-10-05 21:39:40,294 - Epoch: [122][ 480/ 1236] Overall Loss 0.228065 Objective Loss 0.228065 LR 0.000500 Time 0.021835 +2023-10-05 21:39:40,496 - Epoch: [122][ 490/ 1236] Overall Loss 0.228418 Objective Loss 0.228418 LR 0.000500 Time 0.021800 +2023-10-05 21:39:40,697 - Epoch: [122][ 500/ 1236] Overall Loss 0.228082 Objective Loss 0.228082 LR 0.000500 Time 0.021766 +2023-10-05 21:39:40,898 - Epoch: [122][ 510/ 1236] Overall Loss 0.228737 Objective Loss 0.228737 LR 0.000500 Time 0.021734 +2023-10-05 21:39:41,103 - Epoch: [122][ 520/ 1236] Overall Loss 0.228820 Objective Loss 0.228820 LR 0.000500 Time 0.021709 +2023-10-05 21:39:41,304 - Epoch: [122][ 530/ 1236] Overall Loss 0.228798 Objective Loss 0.228798 LR 0.000500 Time 0.021677 +2023-10-05 21:39:41,506 - Epoch: [122][ 540/ 1236] Overall Loss 0.228765 Objective Loss 0.228765 LR 0.000500 Time 0.021650 +2023-10-05 21:39:41,707 - Epoch: [122][ 550/ 1236] Overall Loss 0.228625 Objective Loss 0.228625 LR 0.000500 Time 0.021621 +2023-10-05 21:39:41,911 - Epoch: [122][ 560/ 1236] Overall Loss 0.228789 Objective Loss 0.228789 LR 0.000500 Time 0.021599 +2023-10-05 21:39:42,112 - Epoch: [122][ 570/ 1236] Overall Loss 0.228407 Objective Loss 0.228407 LR 0.000500 Time 0.021572 +2023-10-05 21:39:42,316 - Epoch: [122][ 580/ 1236] Overall Loss 0.228571 Objective Loss 0.228571 LR 0.000500 Time 0.021551 +2023-10-05 21:39:42,518 - Epoch: [122][ 590/ 1236] Overall Loss 0.228463 Objective Loss 0.228463 LR 0.000500 Time 0.021527 +2023-10-05 21:39:42,722 - Epoch: [122][ 600/ 1236] Overall Loss 0.228624 Objective Loss 0.228624 LR 0.000500 Time 0.021507 +2023-10-05 21:39:42,920 - Epoch: [122][ 610/ 1236] Overall Loss 0.228538 Objective Loss 0.228538 LR 0.000500 Time 0.021480 +2023-10-05 21:39:43,120 - Epoch: [122][ 620/ 1236] Overall Loss 0.228807 Objective Loss 0.228807 LR 0.000500 Time 0.021455 +2023-10-05 21:39:43,320 - Epoch: [122][ 630/ 1236] Overall Loss 0.228788 Objective Loss 0.228788 LR 0.000500 Time 0.021431 +2023-10-05 21:39:43,525 - Epoch: [122][ 640/ 1236] Overall Loss 0.228535 Objective Loss 0.228535 LR 0.000500 Time 0.021415 +2023-10-05 21:39:43,725 - Epoch: [122][ 650/ 1236] Overall Loss 0.228283 Objective Loss 0.228283 LR 0.000500 Time 0.021394 +2023-10-05 21:39:43,930 - Epoch: [122][ 660/ 1236] Overall Loss 0.228659 Objective Loss 0.228659 LR 0.000500 Time 0.021379 +2023-10-05 21:39:44,131 - Epoch: [122][ 670/ 1236] Overall Loss 0.228706 Objective Loss 0.228706 LR 0.000500 Time 0.021360 +2023-10-05 21:39:44,335 - Epoch: [122][ 680/ 1236] Overall Loss 0.228947 Objective Loss 0.228947 LR 0.000500 Time 0.021346 +2023-10-05 21:39:44,534 - Epoch: [122][ 690/ 1236] Overall Loss 0.229390 Objective Loss 0.229390 LR 0.000500 Time 0.021324 +2023-10-05 21:39:44,739 - Epoch: [122][ 700/ 1236] Overall Loss 0.229713 Objective Loss 0.229713 LR 0.000500 Time 0.021311 +2023-10-05 21:39:44,939 - Epoch: [122][ 710/ 1236] Overall Loss 0.229318 Objective Loss 0.229318 LR 0.000500 Time 0.021293 +2023-10-05 21:39:45,143 - Epoch: [122][ 720/ 1236] Overall Loss 0.228965 Objective Loss 0.228965 LR 0.000500 Time 0.021280 +2023-10-05 21:39:45,344 - Epoch: [122][ 730/ 1236] Overall Loss 0.229161 Objective Loss 0.229161 LR 0.000500 Time 0.021264 +2023-10-05 21:39:45,549 - Epoch: [122][ 740/ 1236] Overall Loss 0.229132 Objective Loss 0.229132 LR 0.000500 Time 0.021252 +2023-10-05 21:39:45,750 - Epoch: [122][ 750/ 1236] Overall Loss 0.229402 Objective Loss 0.229402 LR 0.000500 Time 0.021236 +2023-10-05 21:39:45,954 - Epoch: [122][ 760/ 1236] Overall Loss 0.229054 Objective Loss 0.229054 LR 0.000500 Time 0.021224 +2023-10-05 21:39:46,155 - Epoch: [122][ 770/ 1236] Overall Loss 0.228710 Objective Loss 0.228710 LR 0.000500 Time 0.021209 +2023-10-05 21:39:46,359 - Epoch: [122][ 780/ 1236] Overall Loss 0.229004 Objective Loss 0.229004 LR 0.000500 Time 0.021199 +2023-10-05 21:39:46,560 - Epoch: [122][ 790/ 1236] Overall Loss 0.229091 Objective Loss 0.229091 LR 0.000500 Time 0.021184 +2023-10-05 21:39:46,764 - Epoch: [122][ 800/ 1236] Overall Loss 0.229244 Objective Loss 0.229244 LR 0.000500 Time 0.021174 +2023-10-05 21:39:46,965 - Epoch: [122][ 810/ 1236] Overall Loss 0.229481 Objective Loss 0.229481 LR 0.000500 Time 0.021160 +2023-10-05 21:39:47,169 - Epoch: [122][ 820/ 1236] Overall Loss 0.229492 Objective Loss 0.229492 LR 0.000500 Time 0.021151 +2023-10-05 21:39:47,370 - Epoch: [122][ 830/ 1236] Overall Loss 0.229723 Objective Loss 0.229723 LR 0.000500 Time 0.021138 +2023-10-05 21:39:47,575 - Epoch: [122][ 840/ 1236] Overall Loss 0.230109 Objective Loss 0.230109 LR 0.000500 Time 0.021129 +2023-10-05 21:39:47,776 - Epoch: [122][ 850/ 1236] Overall Loss 0.230310 Objective Loss 0.230310 LR 0.000500 Time 0.021117 +2023-10-05 21:39:47,980 - Epoch: [122][ 860/ 1236] Overall Loss 0.230628 Objective Loss 0.230628 LR 0.000500 Time 0.021108 +2023-10-05 21:39:48,181 - Epoch: [122][ 870/ 1236] Overall Loss 0.230498 Objective Loss 0.230498 LR 0.000500 Time 0.021096 +2023-10-05 21:39:48,385 - Epoch: [122][ 880/ 1236] Overall Loss 0.230629 Objective Loss 0.230629 LR 0.000500 Time 0.021088 +2023-10-05 21:39:48,586 - Epoch: [122][ 890/ 1236] Overall Loss 0.231138 Objective Loss 0.231138 LR 0.000500 Time 0.021076 +2023-10-05 21:39:48,791 - Epoch: [122][ 900/ 1236] Overall Loss 0.231326 Objective Loss 0.231326 LR 0.000500 Time 0.021069 +2023-10-05 21:39:48,989 - Epoch: [122][ 910/ 1236] Overall Loss 0.231295 Objective Loss 0.231295 LR 0.000500 Time 0.021055 +2023-10-05 21:39:49,191 - Epoch: [122][ 920/ 1236] Overall Loss 0.231493 Objective Loss 0.231493 LR 0.000500 Time 0.021046 +2023-10-05 21:39:49,391 - Epoch: [122][ 930/ 1236] Overall Loss 0.231604 Objective Loss 0.231604 LR 0.000500 Time 0.021034 +2023-10-05 21:39:49,592 - Epoch: [122][ 940/ 1236] Overall Loss 0.231797 Objective Loss 0.231797 LR 0.000500 Time 0.021024 +2023-10-05 21:39:49,791 - Epoch: [122][ 950/ 1236] Overall Loss 0.232173 Objective Loss 0.232173 LR 0.000500 Time 0.021012 +2023-10-05 21:39:49,992 - Epoch: [122][ 960/ 1236] Overall Loss 0.232174 Objective Loss 0.232174 LR 0.000500 Time 0.021002 +2023-10-05 21:39:50,191 - Epoch: [122][ 970/ 1236] Overall Loss 0.232510 Objective Loss 0.232510 LR 0.000500 Time 0.020990 +2023-10-05 21:39:50,392 - Epoch: [122][ 980/ 1236] Overall Loss 0.232569 Objective Loss 0.232569 LR 0.000500 Time 0.020981 +2023-10-05 21:39:50,590 - Epoch: [122][ 990/ 1236] Overall Loss 0.232823 Objective Loss 0.232823 LR 0.000500 Time 0.020969 +2023-10-05 21:39:50,792 - Epoch: [122][ 1000/ 1236] Overall Loss 0.232872 Objective Loss 0.232872 LR 0.000500 Time 0.020961 +2023-10-05 21:39:50,991 - Epoch: [122][ 1010/ 1236] Overall Loss 0.232721 Objective Loss 0.232721 LR 0.000500 Time 0.020950 +2023-10-05 21:39:51,192 - Epoch: [122][ 1020/ 1236] Overall Loss 0.232534 Objective Loss 0.232534 LR 0.000500 Time 0.020942 +2023-10-05 21:39:51,392 - Epoch: [122][ 1030/ 1236] Overall Loss 0.232744 Objective Loss 0.232744 LR 0.000500 Time 0.020932 +2023-10-05 21:39:51,594 - Epoch: [122][ 1040/ 1236] Overall Loss 0.232673 Objective Loss 0.232673 LR 0.000500 Time 0.020925 +2023-10-05 21:39:51,794 - Epoch: [122][ 1050/ 1236] Overall Loss 0.232810 Objective Loss 0.232810 LR 0.000500 Time 0.020916 +2023-10-05 21:39:51,996 - Epoch: [122][ 1060/ 1236] Overall Loss 0.233021 Objective Loss 0.233021 LR 0.000500 Time 0.020908 +2023-10-05 21:39:52,196 - Epoch: [122][ 1070/ 1236] Overall Loss 0.233331 Objective Loss 0.233331 LR 0.000500 Time 0.020899 +2023-10-05 21:39:52,398 - Epoch: [122][ 1080/ 1236] Overall Loss 0.233687 Objective Loss 0.233687 LR 0.000500 Time 0.020892 +2023-10-05 21:39:52,606 - Epoch: [122][ 1090/ 1236] Overall Loss 0.233622 Objective Loss 0.233622 LR 0.000500 Time 0.020891 +2023-10-05 21:39:52,815 - Epoch: [122][ 1100/ 1236] Overall Loss 0.233841 Objective Loss 0.233841 LR 0.000500 Time 0.020892 +2023-10-05 21:39:53,031 - Epoch: [122][ 1110/ 1236] Overall Loss 0.233875 Objective Loss 0.233875 LR 0.000500 Time 0.020897 +2023-10-05 21:39:53,240 - Epoch: [122][ 1120/ 1236] Overall Loss 0.233770 Objective Loss 0.233770 LR 0.000500 Time 0.020897 +2023-10-05 21:39:53,455 - Epoch: [122][ 1130/ 1236] Overall Loss 0.233857 Objective Loss 0.233857 LR 0.000500 Time 0.020902 +2023-10-05 21:39:53,657 - Epoch: [122][ 1140/ 1236] Overall Loss 0.233721 Objective Loss 0.233721 LR 0.000500 Time 0.020896 +2023-10-05 21:39:53,862 - Epoch: [122][ 1150/ 1236] Overall Loss 0.233691 Objective Loss 0.233691 LR 0.000500 Time 0.020892 +2023-10-05 21:39:54,065 - Epoch: [122][ 1160/ 1236] Overall Loss 0.233552 Objective Loss 0.233552 LR 0.000500 Time 0.020886 +2023-10-05 21:39:54,269 - Epoch: [122][ 1170/ 1236] Overall Loss 0.233662 Objective Loss 0.233662 LR 0.000500 Time 0.020882 +2023-10-05 21:39:54,472 - Epoch: [122][ 1180/ 1236] Overall Loss 0.233667 Objective Loss 0.233667 LR 0.000500 Time 0.020877 +2023-10-05 21:39:54,676 - Epoch: [122][ 1190/ 1236] Overall Loss 0.233886 Objective Loss 0.233886 LR 0.000500 Time 0.020873 +2023-10-05 21:39:54,879 - Epoch: [122][ 1200/ 1236] Overall Loss 0.233841 Objective Loss 0.233841 LR 0.000500 Time 0.020867 +2023-10-05 21:39:55,084 - Epoch: [122][ 1210/ 1236] Overall Loss 0.233801 Objective Loss 0.233801 LR 0.000500 Time 0.020864 +2023-10-05 21:39:55,286 - Epoch: [122][ 1220/ 1236] Overall Loss 0.233716 Objective Loss 0.233716 LR 0.000500 Time 0.020859 +2023-10-05 21:39:55,543 - Epoch: [122][ 1230/ 1236] Overall Loss 0.233731 Objective Loss 0.233731 LR 0.000500 Time 0.020898 +2023-10-05 21:39:55,662 - Epoch: [122][ 1236/ 1236] Overall Loss 0.233748 Objective Loss 0.233748 Top1 83.706721 Top5 98.167006 LR 0.000500 Time 0.020892 +2023-10-05 21:39:55,787 - --- validate (epoch=122)----------- +2023-10-05 21:39:55,787 - 29943 samples (256 per mini-batch) +2023-10-05 21:39:56,246 - Epoch: [122][ 10/ 117] Loss 0.325915 Top1 85.390625 Top5 98.085938 +2023-10-05 21:39:56,391 - Epoch: [122][ 20/ 117] Loss 0.316608 Top1 85.136719 Top5 98.320312 +2023-10-05 21:39:56,535 - Epoch: [122][ 30/ 117] Loss 0.313920 Top1 84.791667 Top5 98.320312 +2023-10-05 21:39:56,679 - Epoch: [122][ 40/ 117] Loss 0.320697 Top1 84.531250 Top5 98.193359 +2023-10-05 21:39:56,823 - Epoch: [122][ 50/ 117] Loss 0.322476 Top1 84.640625 Top5 98.109375 +2023-10-05 21:39:56,966 - Epoch: [122][ 60/ 117] Loss 0.318178 Top1 84.596354 Top5 98.118490 +2023-10-05 21:39:57,110 - Epoch: [122][ 70/ 117] Loss 0.318485 Top1 84.575893 Top5 98.052455 +2023-10-05 21:39:57,254 - Epoch: [122][ 80/ 117] Loss 0.318041 Top1 84.501953 Top5 98.012695 +2023-10-05 21:39:57,398 - Epoch: [122][ 90/ 117] Loss 0.319549 Top1 84.440104 Top5 97.977431 +2023-10-05 21:39:57,543 - Epoch: [122][ 100/ 117] Loss 0.321556 Top1 84.390625 Top5 97.964844 +2023-10-05 21:39:57,695 - Epoch: [122][ 110/ 117] Loss 0.324578 Top1 84.414062 Top5 97.982955 +2023-10-05 21:39:57,781 - Epoch: [122][ 117/ 117] Loss 0.323296 Top1 84.470494 Top5 97.999532 +2023-10-05 21:39:57,916 - ==> Top1: 84.470 Top5: 98.000 Loss: 0.323 + +2023-10-05 21:39:57,917 - ==> Confusion: +[[ 919 1 4 1 6 3 1 0 4 73 1 4 1 2 8 3 6 0 0 0 13] + [ 1 1055 1 0 8 22 1 18 2 0 1 1 0 0 2 4 3 0 6 1 5] + [ 3 1 972 8 3 0 13 5 0 1 8 2 9 2 1 4 3 1 6 3 11] + [ 1 0 19 950 2 5 2 1 1 1 6 0 7 2 37 8 1 10 18 1 17] + [ 21 7 1 1 968 2 0 0 0 7 2 3 1 3 10 3 9 1 0 2 9] + [ 4 36 2 1 4 987 1 20 1 2 4 8 0 17 6 1 5 0 2 2 13] + [ 0 4 32 0 0 0 1104 16 0 0 4 1 2 0 1 11 0 1 1 6 8] + [ 3 16 18 0 2 32 3 1074 0 4 2 7 2 1 1 2 1 0 34 6 10] + [ 19 5 1 0 1 3 0 0 955 45 12 2 1 9 21 5 1 0 4 2 3] + [ 108 0 2 0 3 4 0 1 17 924 1 3 0 29 11 5 1 0 0 2 8] + [ 3 9 11 7 1 0 2 5 8 0 966 4 0 14 5 2 1 0 7 2 6] + [ 1 1 1 0 1 13 0 1 1 0 0 966 14 6 0 2 2 14 0 9 3] + [ 0 0 4 2 1 1 1 1 1 0 1 40 975 3 2 4 4 14 2 3 9] + [ 3 0 3 0 3 6 0 0 5 14 6 5 4 1051 3 2 1 1 0 2 10] + [ 12 2 3 8 6 0 0 0 14 3 2 1 3 1 1019 0 0 2 11 0 14] + [ 0 4 2 0 5 1 3 0 0 0 0 10 5 1 0 1069 13 7 0 8 6] + [ 0 10 1 0 4 7 0 3 1 0 0 2 1 0 4 7 1107 0 0 5 9] + [ 0 0 1 2 1 0 2 0 1 0 0 3 17 0 2 4 0 1001 2 2 0] + [ 1 10 10 13 1 0 2 25 2 0 1 0 3 1 14 0 1 0 974 0 10] + [ 0 4 1 1 1 7 7 11 0 0 2 12 7 3 1 7 11 4 2 1060 11] + [ 112 190 134 45 94 153 31 111 79 80 164 133 317 280 164 65 154 70 132 200 5197]] + +2023-10-05 21:39:57,918 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:39:57,918 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:39:57,931 - + +2023-10-05 21:39:57,931 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:39:59,055 - Epoch: [123][ 10/ 1236] Overall Loss 0.212146 Objective Loss 0.212146 LR 0.000500 Time 0.112306 +2023-10-05 21:39:59,259 - Epoch: [123][ 20/ 1236] Overall Loss 0.222643 Objective Loss 0.222643 LR 0.000500 Time 0.066344 +2023-10-05 21:39:59,461 - Epoch: [123][ 30/ 1236] Overall Loss 0.222119 Objective Loss 0.222119 LR 0.000500 Time 0.050935 +2023-10-05 21:39:59,665 - Epoch: [123][ 40/ 1236] Overall Loss 0.219959 Objective Loss 0.219959 LR 0.000500 Time 0.043301 +2023-10-05 21:39:59,867 - Epoch: [123][ 50/ 1236] Overall Loss 0.219537 Objective Loss 0.219537 LR 0.000500 Time 0.038669 +2023-10-05 21:40:00,071 - Epoch: [123][ 60/ 1236] Overall Loss 0.224840 Objective Loss 0.224840 LR 0.000500 Time 0.035617 +2023-10-05 21:40:00,273 - Epoch: [123][ 70/ 1236] Overall Loss 0.225894 Objective Loss 0.225894 LR 0.000500 Time 0.033411 +2023-10-05 21:40:00,477 - Epoch: [123][ 80/ 1236] Overall Loss 0.227915 Objective Loss 0.227915 LR 0.000500 Time 0.031780 +2023-10-05 21:40:00,679 - Epoch: [123][ 90/ 1236] Overall Loss 0.227373 Objective Loss 0.227373 LR 0.000500 Time 0.030492 +2023-10-05 21:40:00,883 - Epoch: [123][ 100/ 1236] Overall Loss 0.225750 Objective Loss 0.225750 LR 0.000500 Time 0.029481 +2023-10-05 21:40:01,085 - Epoch: [123][ 110/ 1236] Overall Loss 0.224823 Objective Loss 0.224823 LR 0.000500 Time 0.028638 +2023-10-05 21:40:01,289 - Epoch: [123][ 120/ 1236] Overall Loss 0.224131 Objective Loss 0.224131 LR 0.000500 Time 0.027945 +2023-10-05 21:40:01,493 - Epoch: [123][ 130/ 1236] Overall Loss 0.225045 Objective Loss 0.225045 LR 0.000500 Time 0.027358 +2023-10-05 21:40:01,696 - Epoch: [123][ 140/ 1236] Overall Loss 0.225617 Objective Loss 0.225617 LR 0.000500 Time 0.026853 +2023-10-05 21:40:01,898 - Epoch: [123][ 150/ 1236] Overall Loss 0.224898 Objective Loss 0.224898 LR 0.000500 Time 0.026409 +2023-10-05 21:40:02,102 - Epoch: [123][ 160/ 1236] Overall Loss 0.224289 Objective Loss 0.224289 LR 0.000500 Time 0.026030 +2023-10-05 21:40:02,304 - Epoch: [123][ 170/ 1236] Overall Loss 0.223610 Objective Loss 0.223610 LR 0.000500 Time 0.025684 +2023-10-05 21:40:02,508 - Epoch: [123][ 180/ 1236] Overall Loss 0.222593 Objective Loss 0.222593 LR 0.000500 Time 0.025388 +2023-10-05 21:40:02,709 - Epoch: [123][ 190/ 1236] Overall Loss 0.222271 Objective Loss 0.222271 LR 0.000500 Time 0.025110 +2023-10-05 21:40:02,913 - Epoch: [123][ 200/ 1236] Overall Loss 0.222496 Objective Loss 0.222496 LR 0.000500 Time 0.024872 +2023-10-05 21:40:03,117 - Epoch: [123][ 210/ 1236] Overall Loss 0.222164 Objective Loss 0.222164 LR 0.000500 Time 0.024659 +2023-10-05 21:40:03,323 - Epoch: [123][ 220/ 1236] Overall Loss 0.224393 Objective Loss 0.224393 LR 0.000500 Time 0.024472 +2023-10-05 21:40:03,528 - Epoch: [123][ 230/ 1236] Overall Loss 0.224620 Objective Loss 0.224620 LR 0.000500 Time 0.024297 +2023-10-05 21:40:03,734 - Epoch: [123][ 240/ 1236] Overall Loss 0.225056 Objective Loss 0.225056 LR 0.000500 Time 0.024141 +2023-10-05 21:40:03,938 - Epoch: [123][ 250/ 1236] Overall Loss 0.224808 Objective Loss 0.224808 LR 0.000500 Time 0.023993 +2023-10-05 21:40:04,144 - Epoch: [123][ 260/ 1236] Overall Loss 0.224897 Objective Loss 0.224897 LR 0.000500 Time 0.023859 +2023-10-05 21:40:04,348 - Epoch: [123][ 270/ 1236] Overall Loss 0.224647 Objective Loss 0.224647 LR 0.000500 Time 0.023731 +2023-10-05 21:40:04,554 - Epoch: [123][ 280/ 1236] Overall Loss 0.225480 Objective Loss 0.225480 LR 0.000500 Time 0.023616 +2023-10-05 21:40:04,758 - Epoch: [123][ 290/ 1236] Overall Loss 0.225316 Objective Loss 0.225316 LR 0.000500 Time 0.023506 +2023-10-05 21:40:04,964 - Epoch: [123][ 300/ 1236] Overall Loss 0.225149 Objective Loss 0.225149 LR 0.000500 Time 0.023407 +2023-10-05 21:40:05,169 - Epoch: [123][ 310/ 1236] Overall Loss 0.224524 Objective Loss 0.224524 LR 0.000500 Time 0.023312 +2023-10-05 21:40:05,375 - Epoch: [123][ 320/ 1236] Overall Loss 0.225081 Objective Loss 0.225081 LR 0.000500 Time 0.023225 +2023-10-05 21:40:05,580 - Epoch: [123][ 330/ 1236] Overall Loss 0.224913 Objective Loss 0.224913 LR 0.000500 Time 0.023142 +2023-10-05 21:40:05,786 - Epoch: [123][ 340/ 1236] Overall Loss 0.225339 Objective Loss 0.225339 LR 0.000500 Time 0.023066 +2023-10-05 21:40:05,991 - Epoch: [123][ 350/ 1236] Overall Loss 0.224919 Objective Loss 0.224919 LR 0.000500 Time 0.022992 +2023-10-05 21:40:06,197 - Epoch: [123][ 360/ 1236] Overall Loss 0.225234 Objective Loss 0.225234 LR 0.000500 Time 0.022925 +2023-10-05 21:40:06,400 - Epoch: [123][ 370/ 1236] Overall Loss 0.225355 Objective Loss 0.225355 LR 0.000500 Time 0.022854 +2023-10-05 21:40:06,605 - Epoch: [123][ 380/ 1236] Overall Loss 0.225866 Objective Loss 0.225866 LR 0.000500 Time 0.022791 +2023-10-05 21:40:06,809 - Epoch: [123][ 390/ 1236] Overall Loss 0.225529 Objective Loss 0.225529 LR 0.000500 Time 0.022727 +2023-10-05 21:40:07,013 - Epoch: [123][ 400/ 1236] Overall Loss 0.225504 Objective Loss 0.225504 LR 0.000500 Time 0.022670 +2023-10-05 21:40:07,217 - Epoch: [123][ 410/ 1236] Overall Loss 0.225405 Objective Loss 0.225405 LR 0.000500 Time 0.022611 +2023-10-05 21:40:07,421 - Epoch: [123][ 420/ 1236] Overall Loss 0.225838 Objective Loss 0.225838 LR 0.000500 Time 0.022559 +2023-10-05 21:40:07,624 - Epoch: [123][ 430/ 1236] Overall Loss 0.225967 Objective Loss 0.225967 LR 0.000500 Time 0.022506 +2023-10-05 21:40:07,829 - Epoch: [123][ 440/ 1236] Overall Loss 0.225768 Objective Loss 0.225768 LR 0.000500 Time 0.022460 +2023-10-05 21:40:08,032 - Epoch: [123][ 450/ 1236] Overall Loss 0.226487 Objective Loss 0.226487 LR 0.000500 Time 0.022411 +2023-10-05 21:40:08,237 - Epoch: [123][ 460/ 1236] Overall Loss 0.226667 Objective Loss 0.226667 LR 0.000500 Time 0.022368 +2023-10-05 21:40:08,440 - Epoch: [123][ 470/ 1236] Overall Loss 0.226827 Objective Loss 0.226827 LR 0.000500 Time 0.022323 +2023-10-05 21:40:08,645 - Epoch: [123][ 480/ 1236] Overall Loss 0.227225 Objective Loss 0.227225 LR 0.000500 Time 0.022285 +2023-10-05 21:40:08,848 - Epoch: [123][ 490/ 1236] Overall Loss 0.227293 Objective Loss 0.227293 LR 0.000500 Time 0.022244 +2023-10-05 21:40:09,053 - Epoch: [123][ 500/ 1236] Overall Loss 0.227445 Objective Loss 0.227445 LR 0.000500 Time 0.022208 +2023-10-05 21:40:09,257 - Epoch: [123][ 510/ 1236] Overall Loss 0.227427 Objective Loss 0.227427 LR 0.000500 Time 0.022170 +2023-10-05 21:40:09,462 - Epoch: [123][ 520/ 1236] Overall Loss 0.228107 Objective Loss 0.228107 LR 0.000500 Time 0.022137 +2023-10-05 21:40:09,665 - Epoch: [123][ 530/ 1236] Overall Loss 0.228428 Objective Loss 0.228428 LR 0.000500 Time 0.022102 +2023-10-05 21:40:09,870 - Epoch: [123][ 540/ 1236] Overall Loss 0.228325 Objective Loss 0.228325 LR 0.000500 Time 0.022071 +2023-10-05 21:40:10,073 - Epoch: [123][ 550/ 1236] Overall Loss 0.228177 Objective Loss 0.228177 LR 0.000500 Time 0.022039 +2023-10-05 21:40:10,278 - Epoch: [123][ 560/ 1236] Overall Loss 0.228570 Objective Loss 0.228570 LR 0.000500 Time 0.022012 +2023-10-05 21:40:10,483 - Epoch: [123][ 570/ 1236] Overall Loss 0.228468 Objective Loss 0.228468 LR 0.000500 Time 0.021985 +2023-10-05 21:40:10,688 - Epoch: [123][ 580/ 1236] Overall Loss 0.228344 Objective Loss 0.228344 LR 0.000500 Time 0.021958 +2023-10-05 21:40:10,891 - Epoch: [123][ 590/ 1236] Overall Loss 0.228705 Objective Loss 0.228705 LR 0.000500 Time 0.021929 +2023-10-05 21:40:11,096 - Epoch: [123][ 600/ 1236] Overall Loss 0.229026 Objective Loss 0.229026 LR 0.000500 Time 0.021905 +2023-10-05 21:40:11,299 - Epoch: [123][ 610/ 1236] Overall Loss 0.229263 Objective Loss 0.229263 LR 0.000500 Time 0.021878 +2023-10-05 21:40:11,504 - Epoch: [123][ 620/ 1236] Overall Loss 0.229535 Objective Loss 0.229535 LR 0.000500 Time 0.021855 +2023-10-05 21:40:11,708 - Epoch: [123][ 630/ 1236] Overall Loss 0.229413 Objective Loss 0.229413 LR 0.000500 Time 0.021830 +2023-10-05 21:40:11,913 - Epoch: [123][ 640/ 1236] Overall Loss 0.229614 Objective Loss 0.229614 LR 0.000500 Time 0.021809 +2023-10-05 21:40:12,116 - Epoch: [123][ 650/ 1236] Overall Loss 0.229390 Objective Loss 0.229390 LR 0.000500 Time 0.021786 +2023-10-05 21:40:12,321 - Epoch: [123][ 660/ 1236] Overall Loss 0.229758 Objective Loss 0.229758 LR 0.000500 Time 0.021766 +2023-10-05 21:40:12,524 - Epoch: [123][ 670/ 1236] Overall Loss 0.229613 Objective Loss 0.229613 LR 0.000500 Time 0.021744 +2023-10-05 21:40:12,729 - Epoch: [123][ 680/ 1236] Overall Loss 0.229871 Objective Loss 0.229871 LR 0.000500 Time 0.021725 +2023-10-05 21:40:12,932 - Epoch: [123][ 690/ 1236] Overall Loss 0.229998 Objective Loss 0.229998 LR 0.000500 Time 0.021704 +2023-10-05 21:40:13,137 - Epoch: [123][ 700/ 1236] Overall Loss 0.230027 Objective Loss 0.230027 LR 0.000500 Time 0.021686 +2023-10-05 21:40:13,340 - Epoch: [123][ 710/ 1236] Overall Loss 0.230511 Objective Loss 0.230511 LR 0.000500 Time 0.021666 +2023-10-05 21:40:13,545 - Epoch: [123][ 720/ 1236] Overall Loss 0.230466 Objective Loss 0.230466 LR 0.000500 Time 0.021650 +2023-10-05 21:40:13,749 - Epoch: [123][ 730/ 1236] Overall Loss 0.230622 Objective Loss 0.230622 LR 0.000500 Time 0.021631 +2023-10-05 21:40:13,954 - Epoch: [123][ 740/ 1236] Overall Loss 0.230856 Objective Loss 0.230856 LR 0.000500 Time 0.021615 +2023-10-05 21:40:14,157 - Epoch: [123][ 750/ 1236] Overall Loss 0.231321 Objective Loss 0.231321 LR 0.000500 Time 0.021597 +2023-10-05 21:40:14,362 - Epoch: [123][ 760/ 1236] Overall Loss 0.231613 Objective Loss 0.231613 LR 0.000500 Time 0.021582 +2023-10-05 21:40:14,565 - Epoch: [123][ 770/ 1236] Overall Loss 0.231442 Objective Loss 0.231442 LR 0.000500 Time 0.021566 +2023-10-05 21:40:14,770 - Epoch: [123][ 780/ 1236] Overall Loss 0.231565 Objective Loss 0.231565 LR 0.000500 Time 0.021551 +2023-10-05 21:40:14,973 - Epoch: [123][ 790/ 1236] Overall Loss 0.231698 Objective Loss 0.231698 LR 0.000500 Time 0.021535 +2023-10-05 21:40:15,178 - Epoch: [123][ 800/ 1236] Overall Loss 0.231910 Objective Loss 0.231910 LR 0.000500 Time 0.021522 +2023-10-05 21:40:15,381 - Epoch: [123][ 810/ 1236] Overall Loss 0.231772 Objective Loss 0.231772 LR 0.000500 Time 0.021507 +2023-10-05 21:40:15,587 - Epoch: [123][ 820/ 1236] Overall Loss 0.231646 Objective Loss 0.231646 LR 0.000500 Time 0.021494 +2023-10-05 21:40:15,790 - Epoch: [123][ 830/ 1236] Overall Loss 0.231912 Objective Loss 0.231912 LR 0.000500 Time 0.021480 +2023-10-05 21:40:15,995 - Epoch: [123][ 840/ 1236] Overall Loss 0.232431 Objective Loss 0.232431 LR 0.000500 Time 0.021467 +2023-10-05 21:40:16,198 - Epoch: [123][ 850/ 1236] Overall Loss 0.232656 Objective Loss 0.232656 LR 0.000500 Time 0.021454 +2023-10-05 21:40:16,403 - Epoch: [123][ 860/ 1236] Overall Loss 0.233274 Objective Loss 0.233274 LR 0.000500 Time 0.021442 +2023-10-05 21:40:16,607 - Epoch: [123][ 870/ 1236] Overall Loss 0.233750 Objective Loss 0.233750 LR 0.000500 Time 0.021429 +2023-10-05 21:40:16,812 - Epoch: [123][ 880/ 1236] Overall Loss 0.233919 Objective Loss 0.233919 LR 0.000500 Time 0.021418 +2023-10-05 21:40:17,015 - Epoch: [123][ 890/ 1236] Overall Loss 0.233692 Objective Loss 0.233692 LR 0.000500 Time 0.021405 +2023-10-05 21:40:17,220 - Epoch: [123][ 900/ 1236] Overall Loss 0.233506 Objective Loss 0.233506 LR 0.000500 Time 0.021395 +2023-10-05 21:40:17,423 - Epoch: [123][ 910/ 1236] Overall Loss 0.233566 Objective Loss 0.233566 LR 0.000500 Time 0.021383 +2023-10-05 21:40:17,629 - Epoch: [123][ 920/ 1236] Overall Loss 0.233701 Objective Loss 0.233701 LR 0.000500 Time 0.021373 +2023-10-05 21:40:17,832 - Epoch: [123][ 930/ 1236] Overall Loss 0.233895 Objective Loss 0.233895 LR 0.000500 Time 0.021362 +2023-10-05 21:40:18,037 - Epoch: [123][ 940/ 1236] Overall Loss 0.234452 Objective Loss 0.234452 LR 0.000500 Time 0.021352 +2023-10-05 21:40:18,240 - Epoch: [123][ 950/ 1236] Overall Loss 0.234503 Objective Loss 0.234503 LR 0.000500 Time 0.021341 +2023-10-05 21:40:18,445 - Epoch: [123][ 960/ 1236] Overall Loss 0.234700 Objective Loss 0.234700 LR 0.000500 Time 0.021332 +2023-10-05 21:40:18,648 - Epoch: [123][ 970/ 1236] Overall Loss 0.234383 Objective Loss 0.234383 LR 0.000500 Time 0.021321 +2023-10-05 21:40:18,853 - Epoch: [123][ 980/ 1236] Overall Loss 0.234361 Objective Loss 0.234361 LR 0.000500 Time 0.021312 +2023-10-05 21:40:19,057 - Epoch: [123][ 990/ 1236] Overall Loss 0.234514 Objective Loss 0.234514 LR 0.000500 Time 0.021302 +2023-10-05 21:40:19,262 - Epoch: [123][ 1000/ 1236] Overall Loss 0.234360 Objective Loss 0.234360 LR 0.000500 Time 0.021294 +2023-10-05 21:40:19,465 - Epoch: [123][ 1010/ 1236] Overall Loss 0.234344 Objective Loss 0.234344 LR 0.000500 Time 0.021284 +2023-10-05 21:40:19,671 - Epoch: [123][ 1020/ 1236] Overall Loss 0.234464 Objective Loss 0.234464 LR 0.000500 Time 0.021276 +2023-10-05 21:40:19,874 - Epoch: [123][ 1030/ 1236] Overall Loss 0.234592 Objective Loss 0.234592 LR 0.000500 Time 0.021267 +2023-10-05 21:40:20,079 - Epoch: [123][ 1040/ 1236] Overall Loss 0.234731 Objective Loss 0.234731 LR 0.000500 Time 0.021259 +2023-10-05 21:40:20,282 - Epoch: [123][ 1050/ 1236] Overall Loss 0.234758 Objective Loss 0.234758 LR 0.000500 Time 0.021250 +2023-10-05 21:40:20,487 - Epoch: [123][ 1060/ 1236] Overall Loss 0.234596 Objective Loss 0.234596 LR 0.000500 Time 0.021242 +2023-10-05 21:40:20,690 - Epoch: [123][ 1070/ 1236] Overall Loss 0.234442 Objective Loss 0.234442 LR 0.000500 Time 0.021233 +2023-10-05 21:40:20,895 - Epoch: [123][ 1080/ 1236] Overall Loss 0.234415 Objective Loss 0.234415 LR 0.000500 Time 0.021225 +2023-10-05 21:40:21,097 - Epoch: [123][ 1090/ 1236] Overall Loss 0.234422 Objective Loss 0.234422 LR 0.000500 Time 0.021216 +2023-10-05 21:40:21,302 - Epoch: [123][ 1100/ 1236] Overall Loss 0.234215 Objective Loss 0.234215 LR 0.000500 Time 0.021209 +2023-10-05 21:40:21,505 - Epoch: [123][ 1110/ 1236] Overall Loss 0.234056 Objective Loss 0.234056 LR 0.000500 Time 0.021201 +2023-10-05 21:40:21,710 - Epoch: [123][ 1120/ 1236] Overall Loss 0.234024 Objective Loss 0.234024 LR 0.000500 Time 0.021194 +2023-10-05 21:40:21,913 - Epoch: [123][ 1130/ 1236] Overall Loss 0.234033 Objective Loss 0.234033 LR 0.000500 Time 0.021186 +2023-10-05 21:40:22,118 - Epoch: [123][ 1140/ 1236] Overall Loss 0.234084 Objective Loss 0.234084 LR 0.000500 Time 0.021179 +2023-10-05 21:40:22,321 - Epoch: [123][ 1150/ 1236] Overall Loss 0.233962 Objective Loss 0.233962 LR 0.000500 Time 0.021172 +2023-10-05 21:40:22,526 - Epoch: [123][ 1160/ 1236] Overall Loss 0.234053 Objective Loss 0.234053 LR 0.000500 Time 0.021165 +2023-10-05 21:40:22,729 - Epoch: [123][ 1170/ 1236] Overall Loss 0.234137 Objective Loss 0.234137 LR 0.000500 Time 0.021158 +2023-10-05 21:40:22,934 - Epoch: [123][ 1180/ 1236] Overall Loss 0.234324 Objective Loss 0.234324 LR 0.000500 Time 0.021152 +2023-10-05 21:40:23,138 - Epoch: [123][ 1190/ 1236] Overall Loss 0.234336 Objective Loss 0.234336 LR 0.000500 Time 0.021145 +2023-10-05 21:40:23,343 - Epoch: [123][ 1200/ 1236] Overall Loss 0.234112 Objective Loss 0.234112 LR 0.000500 Time 0.021139 +2023-10-05 21:40:23,547 - Epoch: [123][ 1210/ 1236] Overall Loss 0.233928 Objective Loss 0.233928 LR 0.000500 Time 0.021133 +2023-10-05 21:40:23,752 - Epoch: [123][ 1220/ 1236] Overall Loss 0.234187 Objective Loss 0.234187 LR 0.000500 Time 0.021127 +2023-10-05 21:40:24,010 - Epoch: [123][ 1230/ 1236] Overall Loss 0.234080 Objective Loss 0.234080 LR 0.000500 Time 0.021165 +2023-10-05 21:40:24,129 - Epoch: [123][ 1236/ 1236] Overall Loss 0.234125 Objective Loss 0.234125 Top1 87.169043 Top5 98.167006 LR 0.000500 Time 0.021159 +2023-10-05 21:40:24,251 - --- validate (epoch=123)----------- +2023-10-05 21:40:24,251 - 29943 samples (256 per mini-batch) +2023-10-05 21:40:24,716 - Epoch: [123][ 10/ 117] Loss 0.299520 Top1 83.632812 Top5 98.007812 +2023-10-05 21:40:24,874 - Epoch: [123][ 20/ 117] Loss 0.304658 Top1 83.984375 Top5 98.222656 +2023-10-05 21:40:25,029 - Epoch: [123][ 30/ 117] Loss 0.314946 Top1 84.036458 Top5 98.203125 +2023-10-05 21:40:25,190 - Epoch: [123][ 40/ 117] Loss 0.319109 Top1 83.974609 Top5 98.115234 +2023-10-05 21:40:25,345 - Epoch: [123][ 50/ 117] Loss 0.334625 Top1 83.851562 Top5 97.976562 +2023-10-05 21:40:25,505 - Epoch: [123][ 60/ 117] Loss 0.328860 Top1 84.101562 Top5 97.923177 +2023-10-05 21:40:25,661 - Epoch: [123][ 70/ 117] Loss 0.328983 Top1 84.062500 Top5 97.946429 +2023-10-05 21:40:25,822 - Epoch: [123][ 80/ 117] Loss 0.329740 Top1 83.945312 Top5 97.866211 +2023-10-05 21:40:25,977 - Epoch: [123][ 90/ 117] Loss 0.329429 Top1 83.914931 Top5 97.860243 +2023-10-05 21:40:26,136 - Epoch: [123][ 100/ 117] Loss 0.326431 Top1 83.902344 Top5 97.867188 +2023-10-05 21:40:26,298 - Epoch: [123][ 110/ 117] Loss 0.321925 Top1 83.987926 Top5 97.894176 +2023-10-05 21:40:26,383 - Epoch: [123][ 117/ 117] Loss 0.323070 Top1 83.926126 Top5 97.872625 +2023-10-05 21:40:26,499 - ==> Top1: 83.926 Top5: 97.873 Loss: 0.323 + +2023-10-05 21:40:26,499 - ==> Confusion: +[[ 929 0 5 3 9 0 1 0 5 67 1 0 4 2 6 3 2 1 1 0 11] + [ 1 1055 3 0 3 23 2 18 4 0 1 1 0 0 1 1 3 0 9 0 6] + [ 4 1 957 14 3 0 25 9 0 0 10 1 8 2 2 4 0 1 6 3 6] + [ 4 0 13 963 3 2 1 2 2 1 8 0 7 2 26 4 0 1 35 1 14] + [ 25 6 2 0 965 4 0 2 0 11 1 1 1 2 8 1 9 1 1 2 8] + [ 4 46 1 2 2 961 3 25 2 0 4 9 1 19 9 2 2 0 5 6 13] + [ 0 4 27 0 0 1 1117 10 0 0 5 2 2 0 1 6 0 1 2 4 9] + [ 8 14 9 0 4 23 5 1070 0 4 4 8 4 1 0 1 0 3 43 9 8] + [ 19 1 0 0 2 2 1 0 960 51 10 3 4 11 18 1 1 0 4 0 1] + [ 95 0 1 0 9 4 1 0 20 942 1 1 0 22 9 2 4 0 0 2 6] + [ 2 5 8 8 2 1 5 6 10 3 964 4 0 13 6 2 1 0 2 0 11] + [ 2 0 2 0 0 8 0 5 0 0 0 961 21 4 0 4 2 12 0 11 3] + [ 0 0 4 5 0 1 1 0 0 0 3 36 983 3 1 3 2 9 3 4 10] + [ 4 0 2 0 3 9 0 0 10 15 8 2 2 1048 4 1 1 1 0 2 7] + [ 15 2 4 9 6 0 0 0 19 5 2 2 2 3 1004 0 0 0 16 0 12] + [ 0 1 2 0 6 0 4 0 0 0 0 9 8 1 0 1065 14 9 1 7 7] + [ 2 14 2 0 8 3 1 2 1 0 0 2 1 1 2 9 1101 0 0 2 10] + [ 1 0 0 0 0 1 1 0 1 0 1 4 23 0 1 4 1 993 2 0 5] + [ 1 7 4 14 2 0 1 20 3 0 2 1 1 0 9 0 0 0 994 0 9] + [ 0 3 2 2 2 5 8 7 0 1 2 15 5 2 0 6 9 1 2 1072 8] + [ 128 176 140 68 83 137 41 92 83 89 161 124 364 296 195 59 173 64 200 206 5026]] + +2023-10-05 21:40:26,501 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:40:26,501 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:40:26,507 - + +2023-10-05 21:40:26,507 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:40:27,513 - Epoch: [124][ 10/ 1236] Overall Loss 0.194711 Objective Loss 0.194711 LR 0.000500 Time 0.100554 +2023-10-05 21:40:27,727 - Epoch: [124][ 20/ 1236] Overall Loss 0.209047 Objective Loss 0.209047 LR 0.000500 Time 0.060962 +2023-10-05 21:40:27,935 - Epoch: [124][ 30/ 1236] Overall Loss 0.208429 Objective Loss 0.208429 LR 0.000500 Time 0.047583 +2023-10-05 21:40:28,149 - Epoch: [124][ 40/ 1236] Overall Loss 0.209971 Objective Loss 0.209971 LR 0.000500 Time 0.041021 +2023-10-05 21:40:28,358 - Epoch: [124][ 50/ 1236] Overall Loss 0.216667 Objective Loss 0.216667 LR 0.000500 Time 0.036995 +2023-10-05 21:40:28,571 - Epoch: [124][ 60/ 1236] Overall Loss 0.213326 Objective Loss 0.213326 LR 0.000500 Time 0.034379 +2023-10-05 21:40:28,781 - Epoch: [124][ 70/ 1236] Overall Loss 0.216143 Objective Loss 0.216143 LR 0.000500 Time 0.032454 +2023-10-05 21:40:28,979 - Epoch: [124][ 80/ 1236] Overall Loss 0.214880 Objective Loss 0.214880 LR 0.000500 Time 0.030874 +2023-10-05 21:40:29,177 - Epoch: [124][ 90/ 1236] Overall Loss 0.217582 Objective Loss 0.217582 LR 0.000500 Time 0.029633 +2023-10-05 21:40:29,376 - Epoch: [124][ 100/ 1236] Overall Loss 0.214395 Objective Loss 0.214395 LR 0.000500 Time 0.028658 +2023-10-05 21:40:29,573 - Epoch: [124][ 110/ 1236] Overall Loss 0.212351 Objective Loss 0.212351 LR 0.000500 Time 0.027842 +2023-10-05 21:40:29,772 - Epoch: [124][ 120/ 1236] Overall Loss 0.212023 Objective Loss 0.212023 LR 0.000500 Time 0.027178 +2023-10-05 21:40:29,969 - Epoch: [124][ 130/ 1236] Overall Loss 0.216170 Objective Loss 0.216170 LR 0.000500 Time 0.026602 +2023-10-05 21:40:30,169 - Epoch: [124][ 140/ 1236] Overall Loss 0.217002 Objective Loss 0.217002 LR 0.000500 Time 0.026122 +2023-10-05 21:40:30,366 - Epoch: [124][ 150/ 1236] Overall Loss 0.216062 Objective Loss 0.216062 LR 0.000500 Time 0.025693 +2023-10-05 21:40:30,565 - Epoch: [124][ 160/ 1236] Overall Loss 0.215093 Objective Loss 0.215093 LR 0.000500 Time 0.025330 +2023-10-05 21:40:30,762 - Epoch: [124][ 170/ 1236] Overall Loss 0.215607 Objective Loss 0.215607 LR 0.000500 Time 0.024996 +2023-10-05 21:40:30,961 - Epoch: [124][ 180/ 1236] Overall Loss 0.216960 Objective Loss 0.216960 LR 0.000500 Time 0.024711 +2023-10-05 21:40:31,158 - Epoch: [124][ 190/ 1236] Overall Loss 0.217419 Objective Loss 0.217419 LR 0.000500 Time 0.024448 +2023-10-05 21:40:31,358 - Epoch: [124][ 200/ 1236] Overall Loss 0.218962 Objective Loss 0.218962 LR 0.000500 Time 0.024220 +2023-10-05 21:40:31,555 - Epoch: [124][ 210/ 1236] Overall Loss 0.220095 Objective Loss 0.220095 LR 0.000500 Time 0.024004 +2023-10-05 21:40:31,754 - Epoch: [124][ 220/ 1236] Overall Loss 0.221019 Objective Loss 0.221019 LR 0.000500 Time 0.023816 +2023-10-05 21:40:31,952 - Epoch: [124][ 230/ 1236] Overall Loss 0.220625 Objective Loss 0.220625 LR 0.000500 Time 0.023638 +2023-10-05 21:40:32,151 - Epoch: [124][ 240/ 1236] Overall Loss 0.221408 Objective Loss 0.221408 LR 0.000500 Time 0.023484 +2023-10-05 21:40:32,348 - Epoch: [124][ 250/ 1236] Overall Loss 0.222188 Objective Loss 0.222188 LR 0.000500 Time 0.023331 +2023-10-05 21:40:32,547 - Epoch: [124][ 260/ 1236] Overall Loss 0.222706 Objective Loss 0.222706 LR 0.000500 Time 0.023198 +2023-10-05 21:40:32,745 - Epoch: [124][ 270/ 1236] Overall Loss 0.222913 Objective Loss 0.222913 LR 0.000500 Time 0.023068 +2023-10-05 21:40:32,944 - Epoch: [124][ 280/ 1236] Overall Loss 0.223071 Objective Loss 0.223071 LR 0.000500 Time 0.022956 +2023-10-05 21:40:33,141 - Epoch: [124][ 290/ 1236] Overall Loss 0.223389 Objective Loss 0.223389 LR 0.000500 Time 0.022842 +2023-10-05 21:40:33,341 - Epoch: [124][ 300/ 1236] Overall Loss 0.223964 Objective Loss 0.223964 LR 0.000500 Time 0.022746 +2023-10-05 21:40:33,538 - Epoch: [124][ 310/ 1236] Overall Loss 0.223799 Objective Loss 0.223799 LR 0.000500 Time 0.022646 +2023-10-05 21:40:33,738 - Epoch: [124][ 320/ 1236] Overall Loss 0.223355 Objective Loss 0.223355 LR 0.000500 Time 0.022562 +2023-10-05 21:40:33,935 - Epoch: [124][ 330/ 1236] Overall Loss 0.223492 Objective Loss 0.223492 LR 0.000500 Time 0.022475 +2023-10-05 21:40:34,135 - Epoch: [124][ 340/ 1236] Overall Loss 0.223436 Objective Loss 0.223436 LR 0.000500 Time 0.022399 +2023-10-05 21:40:34,332 - Epoch: [124][ 350/ 1236] Overall Loss 0.223276 Objective Loss 0.223276 LR 0.000500 Time 0.022323 +2023-10-05 21:40:34,532 - Epoch: [124][ 360/ 1236] Overall Loss 0.223325 Objective Loss 0.223325 LR 0.000500 Time 0.022258 +2023-10-05 21:40:34,730 - Epoch: [124][ 370/ 1236] Overall Loss 0.223379 Objective Loss 0.223379 LR 0.000500 Time 0.022191 +2023-10-05 21:40:34,931 - Epoch: [124][ 380/ 1236] Overall Loss 0.224293 Objective Loss 0.224293 LR 0.000500 Time 0.022135 +2023-10-05 21:40:35,130 - Epoch: [124][ 390/ 1236] Overall Loss 0.223991 Objective Loss 0.223991 LR 0.000500 Time 0.022076 +2023-10-05 21:40:35,331 - Epoch: [124][ 400/ 1236] Overall Loss 0.224767 Objective Loss 0.224767 LR 0.000500 Time 0.022026 +2023-10-05 21:40:35,529 - Epoch: [124][ 410/ 1236] Overall Loss 0.225073 Objective Loss 0.225073 LR 0.000500 Time 0.021970 +2023-10-05 21:40:35,729 - Epoch: [124][ 420/ 1236] Overall Loss 0.225472 Objective Loss 0.225472 LR 0.000500 Time 0.021924 +2023-10-05 21:40:35,927 - Epoch: [124][ 430/ 1236] Overall Loss 0.225161 Objective Loss 0.225161 LR 0.000500 Time 0.021873 +2023-10-05 21:40:36,128 - Epoch: [124][ 440/ 1236] Overall Loss 0.225565 Objective Loss 0.225565 LR 0.000500 Time 0.021832 +2023-10-05 21:40:36,326 - Epoch: [124][ 450/ 1236] Overall Loss 0.225775 Objective Loss 0.225775 LR 0.000500 Time 0.021785 +2023-10-05 21:40:36,525 - Epoch: [124][ 460/ 1236] Overall Loss 0.225918 Objective Loss 0.225918 LR 0.000500 Time 0.021745 +2023-10-05 21:40:36,723 - Epoch: [124][ 470/ 1236] Overall Loss 0.225945 Objective Loss 0.225945 LR 0.000500 Time 0.021703 +2023-10-05 21:40:36,924 - Epoch: [124][ 480/ 1236] Overall Loss 0.225698 Objective Loss 0.225698 LR 0.000500 Time 0.021668 +2023-10-05 21:40:37,122 - Epoch: [124][ 490/ 1236] Overall Loss 0.225712 Objective Loss 0.225712 LR 0.000500 Time 0.021629 +2023-10-05 21:40:37,322 - Epoch: [124][ 500/ 1236] Overall Loss 0.225572 Objective Loss 0.225572 LR 0.000500 Time 0.021597 +2023-10-05 21:40:37,520 - Epoch: [124][ 510/ 1236] Overall Loss 0.225950 Objective Loss 0.225950 LR 0.000500 Time 0.021561 +2023-10-05 21:40:37,721 - Epoch: [124][ 520/ 1236] Overall Loss 0.225857 Objective Loss 0.225857 LR 0.000500 Time 0.021532 +2023-10-05 21:40:37,919 - Epoch: [124][ 530/ 1236] Overall Loss 0.225733 Objective Loss 0.225733 LR 0.000500 Time 0.021498 +2023-10-05 21:40:38,120 - Epoch: [124][ 540/ 1236] Overall Loss 0.225521 Objective Loss 0.225521 LR 0.000500 Time 0.021472 +2023-10-05 21:40:38,318 - Epoch: [124][ 550/ 1236] Overall Loss 0.225504 Objective Loss 0.225504 LR 0.000500 Time 0.021440 +2023-10-05 21:40:38,518 - Epoch: [124][ 560/ 1236] Overall Loss 0.225396 Objective Loss 0.225396 LR 0.000500 Time 0.021416 +2023-10-05 21:40:38,717 - Epoch: [124][ 570/ 1236] Overall Loss 0.225477 Objective Loss 0.225477 LR 0.000500 Time 0.021387 +2023-10-05 21:40:38,918 - Epoch: [124][ 580/ 1236] Overall Loss 0.225775 Objective Loss 0.225775 LR 0.000500 Time 0.021364 +2023-10-05 21:40:39,115 - Epoch: [124][ 590/ 1236] Overall Loss 0.226094 Objective Loss 0.226094 LR 0.000500 Time 0.021337 +2023-10-05 21:40:39,316 - Epoch: [124][ 600/ 1236] Overall Loss 0.226313 Objective Loss 0.226313 LR 0.000500 Time 0.021315 +2023-10-05 21:40:39,514 - Epoch: [124][ 610/ 1236] Overall Loss 0.226428 Objective Loss 0.226428 LR 0.000500 Time 0.021290 +2023-10-05 21:40:39,715 - Epoch: [124][ 620/ 1236] Overall Loss 0.226532 Objective Loss 0.226532 LR 0.000500 Time 0.021270 +2023-10-05 21:40:39,913 - Epoch: [124][ 630/ 1236] Overall Loss 0.226646 Objective Loss 0.226646 LR 0.000500 Time 0.021247 +2023-10-05 21:40:40,114 - Epoch: [124][ 640/ 1236] Overall Loss 0.226723 Objective Loss 0.226723 LR 0.000500 Time 0.021228 +2023-10-05 21:40:40,312 - Epoch: [124][ 650/ 1236] Overall Loss 0.226780 Objective Loss 0.226780 LR 0.000500 Time 0.021205 +2023-10-05 21:40:40,513 - Epoch: [124][ 660/ 1236] Overall Loss 0.227199 Objective Loss 0.227199 LR 0.000500 Time 0.021188 +2023-10-05 21:40:40,711 - Epoch: [124][ 670/ 1236] Overall Loss 0.226794 Objective Loss 0.226794 LR 0.000500 Time 0.021166 +2023-10-05 21:40:40,911 - Epoch: [124][ 680/ 1236] Overall Loss 0.227134 Objective Loss 0.227134 LR 0.000500 Time 0.021150 +2023-10-05 21:40:41,110 - Epoch: [124][ 690/ 1236] Overall Loss 0.226806 Objective Loss 0.226806 LR 0.000500 Time 0.021130 +2023-10-05 21:40:41,310 - Epoch: [124][ 700/ 1236] Overall Loss 0.227227 Objective Loss 0.227227 LR 0.000500 Time 0.021114 +2023-10-05 21:40:41,508 - Epoch: [124][ 710/ 1236] Overall Loss 0.227528 Objective Loss 0.227528 LR 0.000500 Time 0.021096 +2023-10-05 21:40:41,709 - Epoch: [124][ 720/ 1236] Overall Loss 0.227517 Objective Loss 0.227517 LR 0.000500 Time 0.021081 +2023-10-05 21:40:41,908 - Epoch: [124][ 730/ 1236] Overall Loss 0.227632 Objective Loss 0.227632 LR 0.000500 Time 0.021064 +2023-10-05 21:40:42,109 - Epoch: [124][ 740/ 1236] Overall Loss 0.227593 Objective Loss 0.227593 LR 0.000500 Time 0.021050 +2023-10-05 21:40:42,307 - Epoch: [124][ 750/ 1236] Overall Loss 0.227589 Objective Loss 0.227589 LR 0.000500 Time 0.021033 +2023-10-05 21:40:42,507 - Epoch: [124][ 760/ 1236] Overall Loss 0.227518 Objective Loss 0.227518 LR 0.000500 Time 0.021020 +2023-10-05 21:40:42,705 - Epoch: [124][ 770/ 1236] Overall Loss 0.227544 Objective Loss 0.227544 LR 0.000500 Time 0.021003 +2023-10-05 21:40:42,906 - Epoch: [124][ 780/ 1236] Overall Loss 0.227411 Objective Loss 0.227411 LR 0.000500 Time 0.020991 +2023-10-05 21:40:43,104 - Epoch: [124][ 790/ 1236] Overall Loss 0.227308 Objective Loss 0.227308 LR 0.000500 Time 0.020976 +2023-10-05 21:40:43,305 - Epoch: [124][ 800/ 1236] Overall Loss 0.227203 Objective Loss 0.227203 LR 0.000500 Time 0.020964 +2023-10-05 21:40:43,503 - Epoch: [124][ 810/ 1236] Overall Loss 0.227655 Objective Loss 0.227655 LR 0.000500 Time 0.020949 +2023-10-05 21:40:43,703 - Epoch: [124][ 820/ 1236] Overall Loss 0.227655 Objective Loss 0.227655 LR 0.000500 Time 0.020938 +2023-10-05 21:40:43,901 - Epoch: [124][ 830/ 1236] Overall Loss 0.228073 Objective Loss 0.228073 LR 0.000500 Time 0.020924 +2023-10-05 21:40:44,102 - Epoch: [124][ 840/ 1236] Overall Loss 0.228048 Objective Loss 0.228048 LR 0.000500 Time 0.020914 +2023-10-05 21:40:44,300 - Epoch: [124][ 850/ 1236] Overall Loss 0.228094 Objective Loss 0.228094 LR 0.000500 Time 0.020901 +2023-10-05 21:40:44,501 - Epoch: [124][ 860/ 1236] Overall Loss 0.228069 Objective Loss 0.228069 LR 0.000500 Time 0.020891 +2023-10-05 21:40:44,699 - Epoch: [124][ 870/ 1236] Overall Loss 0.228218 Objective Loss 0.228218 LR 0.000500 Time 0.020878 +2023-10-05 21:40:44,900 - Epoch: [124][ 880/ 1236] Overall Loss 0.228741 Objective Loss 0.228741 LR 0.000500 Time 0.020869 +2023-10-05 21:40:45,099 - Epoch: [124][ 890/ 1236] Overall Loss 0.228723 Objective Loss 0.228723 LR 0.000500 Time 0.020857 +2023-10-05 21:40:45,299 - Epoch: [124][ 900/ 1236] Overall Loss 0.228664 Objective Loss 0.228664 LR 0.000500 Time 0.020848 +2023-10-05 21:40:45,497 - Epoch: [124][ 910/ 1236] Overall Loss 0.228924 Objective Loss 0.228924 LR 0.000500 Time 0.020836 +2023-10-05 21:40:45,698 - Epoch: [124][ 920/ 1236] Overall Loss 0.229186 Objective Loss 0.229186 LR 0.000500 Time 0.020827 +2023-10-05 21:40:45,896 - Epoch: [124][ 930/ 1236] Overall Loss 0.229066 Objective Loss 0.229066 LR 0.000500 Time 0.020816 +2023-10-05 21:40:46,097 - Epoch: [124][ 940/ 1236] Overall Loss 0.229038 Objective Loss 0.229038 LR 0.000500 Time 0.020808 +2023-10-05 21:40:46,295 - Epoch: [124][ 950/ 1236] Overall Loss 0.229019 Objective Loss 0.229019 LR 0.000500 Time 0.020797 +2023-10-05 21:40:46,496 - Epoch: [124][ 960/ 1236] Overall Loss 0.229438 Objective Loss 0.229438 LR 0.000500 Time 0.020789 +2023-10-05 21:40:46,694 - Epoch: [124][ 970/ 1236] Overall Loss 0.229206 Objective Loss 0.229206 LR 0.000500 Time 0.020778 +2023-10-05 21:40:46,894 - Epoch: [124][ 980/ 1236] Overall Loss 0.229025 Objective Loss 0.229025 LR 0.000500 Time 0.020771 +2023-10-05 21:40:47,092 - Epoch: [124][ 990/ 1236] Overall Loss 0.229333 Objective Loss 0.229333 LR 0.000500 Time 0.020760 +2023-10-05 21:40:47,293 - Epoch: [124][ 1000/ 1236] Overall Loss 0.229141 Objective Loss 0.229141 LR 0.000500 Time 0.020753 +2023-10-05 21:40:47,491 - Epoch: [124][ 1010/ 1236] Overall Loss 0.229003 Objective Loss 0.229003 LR 0.000500 Time 0.020744 +2023-10-05 21:40:47,692 - Epoch: [124][ 1020/ 1236] Overall Loss 0.229189 Objective Loss 0.229189 LR 0.000500 Time 0.020737 +2023-10-05 21:40:47,890 - Epoch: [124][ 1030/ 1236] Overall Loss 0.229176 Objective Loss 0.229176 LR 0.000500 Time 0.020727 +2023-10-05 21:40:48,090 - Epoch: [124][ 1040/ 1236] Overall Loss 0.229079 Objective Loss 0.229079 LR 0.000500 Time 0.020721 +2023-10-05 21:40:48,288 - Epoch: [124][ 1050/ 1236] Overall Loss 0.229432 Objective Loss 0.229432 LR 0.000500 Time 0.020712 +2023-10-05 21:40:48,489 - Epoch: [124][ 1060/ 1236] Overall Loss 0.229451 Objective Loss 0.229451 LR 0.000500 Time 0.020705 +2023-10-05 21:40:48,687 - Epoch: [124][ 1070/ 1236] Overall Loss 0.229674 Objective Loss 0.229674 LR 0.000500 Time 0.020697 +2023-10-05 21:40:48,888 - Epoch: [124][ 1080/ 1236] Overall Loss 0.229647 Objective Loss 0.229647 LR 0.000500 Time 0.020691 +2023-10-05 21:40:49,086 - Epoch: [124][ 1090/ 1236] Overall Loss 0.229891 Objective Loss 0.229891 LR 0.000500 Time 0.020682 +2023-10-05 21:40:49,287 - Epoch: [124][ 1100/ 1236] Overall Loss 0.230082 Objective Loss 0.230082 LR 0.000500 Time 0.020677 +2023-10-05 21:40:49,485 - Epoch: [124][ 1110/ 1236] Overall Loss 0.230106 Objective Loss 0.230106 LR 0.000500 Time 0.020668 +2023-10-05 21:40:49,685 - Epoch: [124][ 1120/ 1236] Overall Loss 0.230265 Objective Loss 0.230265 LR 0.000500 Time 0.020663 +2023-10-05 21:40:49,884 - Epoch: [124][ 1130/ 1236] Overall Loss 0.230245 Objective Loss 0.230245 LR 0.000500 Time 0.020655 +2023-10-05 21:40:50,084 - Epoch: [124][ 1140/ 1236] Overall Loss 0.230504 Objective Loss 0.230504 LR 0.000500 Time 0.020649 +2023-10-05 21:40:50,283 - Epoch: [124][ 1150/ 1236] Overall Loss 0.230380 Objective Loss 0.230380 LR 0.000500 Time 0.020642 +2023-10-05 21:40:50,483 - Epoch: [124][ 1160/ 1236] Overall Loss 0.230550 Objective Loss 0.230550 LR 0.000500 Time 0.020637 +2023-10-05 21:40:50,681 - Epoch: [124][ 1170/ 1236] Overall Loss 0.230591 Objective Loss 0.230591 LR 0.000500 Time 0.020630 +2023-10-05 21:40:50,882 - Epoch: [124][ 1180/ 1236] Overall Loss 0.230413 Objective Loss 0.230413 LR 0.000500 Time 0.020625 +2023-10-05 21:40:51,080 - Epoch: [124][ 1190/ 1236] Overall Loss 0.230763 Objective Loss 0.230763 LR 0.000500 Time 0.020617 +2023-10-05 21:40:51,281 - Epoch: [124][ 1200/ 1236] Overall Loss 0.230672 Objective Loss 0.230672 LR 0.000500 Time 0.020613 +2023-10-05 21:40:51,479 - Epoch: [124][ 1210/ 1236] Overall Loss 0.230653 Objective Loss 0.230653 LR 0.000500 Time 0.020605 +2023-10-05 21:40:51,680 - Epoch: [124][ 1220/ 1236] Overall Loss 0.230536 Objective Loss 0.230536 LR 0.000500 Time 0.020601 +2023-10-05 21:40:51,929 - Epoch: [124][ 1230/ 1236] Overall Loss 0.230262 Objective Loss 0.230262 LR 0.000500 Time 0.020636 +2023-10-05 21:40:52,046 - Epoch: [124][ 1236/ 1236] Overall Loss 0.230377 Objective Loss 0.230377 Top1 87.576375 Top5 98.574338 LR 0.000500 Time 0.020630 +2023-10-05 21:40:52,166 - --- validate (epoch=124)----------- +2023-10-05 21:40:52,167 - 29943 samples (256 per mini-batch) +2023-10-05 21:40:52,623 - Epoch: [124][ 10/ 117] Loss 0.333285 Top1 83.828125 Top5 97.617188 +2023-10-05 21:40:52,770 - Epoch: [124][ 20/ 117] Loss 0.320344 Top1 84.707031 Top5 97.851562 +2023-10-05 21:40:52,915 - Epoch: [124][ 30/ 117] Loss 0.316869 Top1 84.700521 Top5 97.890625 +2023-10-05 21:40:53,061 - Epoch: [124][ 40/ 117] Loss 0.313777 Top1 84.882812 Top5 97.998047 +2023-10-05 21:40:53,207 - Epoch: [124][ 50/ 117] Loss 0.321623 Top1 84.601562 Top5 98.046875 +2023-10-05 21:40:53,351 - Epoch: [124][ 60/ 117] Loss 0.322046 Top1 84.589844 Top5 98.072917 +2023-10-05 21:40:53,496 - Epoch: [124][ 70/ 117] Loss 0.322918 Top1 84.525670 Top5 98.058036 +2023-10-05 21:40:53,640 - Epoch: [124][ 80/ 117] Loss 0.319219 Top1 84.624023 Top5 98.061523 +2023-10-05 21:40:53,787 - Epoch: [124][ 90/ 117] Loss 0.315804 Top1 84.639757 Top5 98.064236 +2023-10-05 21:40:53,932 - Epoch: [124][ 100/ 117] Loss 0.318422 Top1 84.558594 Top5 98.050781 +2023-10-05 21:40:54,086 - Epoch: [124][ 110/ 117] Loss 0.320108 Top1 84.456676 Top5 98.064631 +2023-10-05 21:40:54,171 - Epoch: [124][ 117/ 117] Loss 0.318489 Top1 84.437097 Top5 98.073005 +2023-10-05 21:40:54,289 - ==> Top1: 84.437 Top5: 98.073 Loss: 0.318 + +2023-10-05 21:40:54,289 - ==> Confusion: +[[ 920 1 7 1 9 2 1 0 6 76 1 0 2 2 5 0 4 1 1 1 10] + [ 2 1060 2 0 2 19 1 17 2 0 2 0 0 0 1 3 1 0 13 1 5] + [ 3 2 961 15 1 1 27 8 1 1 4 3 8 0 1 1 0 0 6 2 11] + [ 2 0 12 978 0 3 1 5 4 1 9 0 5 1 26 3 1 5 17 0 16] + [ 27 4 2 0 966 4 0 1 0 8 1 3 2 1 9 4 9 1 0 1 7] + [ 4 51 0 1 1 980 2 23 2 1 4 4 2 13 4 1 2 0 2 4 15] + [ 0 6 23 0 1 1 1128 6 0 0 3 2 3 0 1 4 0 0 2 4 7] + [ 2 18 13 1 1 33 6 1062 0 2 6 6 5 0 0 2 0 0 42 7 12] + [ 20 4 0 0 0 1 0 0 977 42 10 2 3 7 11 1 1 0 7 0 3] + [ 94 1 3 0 6 4 1 0 37 934 1 2 2 17 4 1 1 2 0 1 8] + [ 6 5 9 6 1 3 5 5 11 1 961 4 0 11 3 1 1 0 4 3 13] + [ 2 0 1 0 0 13 0 6 0 0 1 962 22 3 0 2 0 13 0 8 2] + [ 0 0 6 3 0 1 0 3 1 1 2 35 973 3 3 3 3 16 2 4 9] + [ 2 1 2 0 0 16 0 2 18 11 15 3 3 1030 3 1 3 0 0 1 8] + [ 10 3 3 10 7 0 0 0 26 6 2 1 4 2 992 0 0 1 17 0 17] + [ 2 4 2 0 3 1 2 0 0 1 0 9 4 2 1 1064 14 11 0 9 5] + [ 0 13 2 0 5 5 0 2 1 0 0 2 0 0 1 12 1100 0 2 5 11] + [ 0 0 0 0 1 0 3 0 0 1 0 5 22 0 0 5 0 994 1 0 6] + [ 1 6 4 16 1 1 1 31 1 0 2 0 1 0 9 1 2 0 983 0 8] + [ 0 3 4 0 1 4 12 13 1 0 2 14 1 0 0 9 7 2 2 1067 10] + [ 129 193 132 65 82 147 54 104 108 81 164 116 341 230 127 59 174 72 159 177 5191]] + +2023-10-05 21:40:54,291 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:40:54,291 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:40:54,297 - + +2023-10-05 21:40:54,297 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:40:55,283 - Epoch: [125][ 10/ 1236] Overall Loss 0.236667 Objective Loss 0.236667 LR 0.000500 Time 0.098572 +2023-10-05 21:40:55,484 - Epoch: [125][ 20/ 1236] Overall Loss 0.223850 Objective Loss 0.223850 LR 0.000500 Time 0.059323 +2023-10-05 21:40:55,685 - Epoch: [125][ 30/ 1236] Overall Loss 0.229434 Objective Loss 0.229434 LR 0.000500 Time 0.046219 +2023-10-05 21:40:55,887 - Epoch: [125][ 40/ 1236] Overall Loss 0.231307 Objective Loss 0.231307 LR 0.000500 Time 0.039712 +2023-10-05 21:40:56,087 - Epoch: [125][ 50/ 1236] Overall Loss 0.227503 Objective Loss 0.227503 LR 0.000500 Time 0.035759 +2023-10-05 21:40:56,288 - Epoch: [125][ 60/ 1236] Overall Loss 0.228839 Objective Loss 0.228839 LR 0.000500 Time 0.033143 +2023-10-05 21:40:56,487 - Epoch: [125][ 70/ 1236] Overall Loss 0.232954 Objective Loss 0.232954 LR 0.000500 Time 0.031257 +2023-10-05 21:40:56,689 - Epoch: [125][ 80/ 1236] Overall Loss 0.232076 Objective Loss 0.232076 LR 0.000500 Time 0.029861 +2023-10-05 21:40:56,890 - Epoch: [125][ 90/ 1236] Overall Loss 0.230009 Objective Loss 0.230009 LR 0.000500 Time 0.028773 +2023-10-05 21:40:57,092 - Epoch: [125][ 100/ 1236] Overall Loss 0.227416 Objective Loss 0.227416 LR 0.000500 Time 0.027914 +2023-10-05 21:40:57,293 - Epoch: [125][ 110/ 1236] Overall Loss 0.227396 Objective Loss 0.227396 LR 0.000500 Time 0.027197 +2023-10-05 21:40:57,494 - Epoch: [125][ 120/ 1236] Overall Loss 0.227706 Objective Loss 0.227706 LR 0.000500 Time 0.026601 +2023-10-05 21:40:57,697 - Epoch: [125][ 130/ 1236] Overall Loss 0.226401 Objective Loss 0.226401 LR 0.000500 Time 0.026119 +2023-10-05 21:40:57,902 - Epoch: [125][ 140/ 1236] Overall Loss 0.225959 Objective Loss 0.225959 LR 0.000500 Time 0.025712 +2023-10-05 21:40:58,102 - Epoch: [125][ 150/ 1236] Overall Loss 0.225354 Objective Loss 0.225354 LR 0.000500 Time 0.025325 +2023-10-05 21:40:58,308 - Epoch: [125][ 160/ 1236] Overall Loss 0.225494 Objective Loss 0.225494 LR 0.000500 Time 0.025028 +2023-10-05 21:40:58,514 - Epoch: [125][ 170/ 1236] Overall Loss 0.226240 Objective Loss 0.226240 LR 0.000500 Time 0.024765 +2023-10-05 21:40:58,720 - Epoch: [125][ 180/ 1236] Overall Loss 0.226358 Objective Loss 0.226358 LR 0.000500 Time 0.024532 +2023-10-05 21:40:58,926 - Epoch: [125][ 190/ 1236] Overall Loss 0.226382 Objective Loss 0.226382 LR 0.000500 Time 0.024324 +2023-10-05 21:40:59,133 - Epoch: [125][ 200/ 1236] Overall Loss 0.225380 Objective Loss 0.225380 LR 0.000500 Time 0.024135 +2023-10-05 21:40:59,338 - Epoch: [125][ 210/ 1236] Overall Loss 0.225478 Objective Loss 0.225478 LR 0.000500 Time 0.023964 +2023-10-05 21:40:59,544 - Epoch: [125][ 220/ 1236] Overall Loss 0.225741 Objective Loss 0.225741 LR 0.000500 Time 0.023806 +2023-10-05 21:40:59,750 - Epoch: [125][ 230/ 1236] Overall Loss 0.225975 Objective Loss 0.225975 LR 0.000500 Time 0.023664 +2023-10-05 21:40:59,957 - Epoch: [125][ 240/ 1236] Overall Loss 0.225989 Objective Loss 0.225989 LR 0.000500 Time 0.023537 +2023-10-05 21:41:00,163 - Epoch: [125][ 250/ 1236] Overall Loss 0.225708 Objective Loss 0.225708 LR 0.000500 Time 0.023419 +2023-10-05 21:41:00,369 - Epoch: [125][ 260/ 1236] Overall Loss 0.225078 Objective Loss 0.225078 LR 0.000500 Time 0.023309 +2023-10-05 21:41:00,576 - Epoch: [125][ 270/ 1236] Overall Loss 0.224910 Objective Loss 0.224910 LR 0.000500 Time 0.023209 +2023-10-05 21:41:00,782 - Epoch: [125][ 280/ 1236] Overall Loss 0.225534 Objective Loss 0.225534 LR 0.000500 Time 0.023114 +2023-10-05 21:41:00,989 - Epoch: [125][ 290/ 1236] Overall Loss 0.225809 Objective Loss 0.225809 LR 0.000500 Time 0.023028 +2023-10-05 21:41:01,195 - Epoch: [125][ 300/ 1236] Overall Loss 0.226048 Objective Loss 0.226048 LR 0.000500 Time 0.022946 +2023-10-05 21:41:01,401 - Epoch: [125][ 310/ 1236] Overall Loss 0.225034 Objective Loss 0.225034 LR 0.000500 Time 0.022871 +2023-10-05 21:41:01,608 - Epoch: [125][ 320/ 1236] Overall Loss 0.224248 Objective Loss 0.224248 LR 0.000500 Time 0.022801 +2023-10-05 21:41:01,815 - Epoch: [125][ 330/ 1236] Overall Loss 0.224212 Objective Loss 0.224212 LR 0.000500 Time 0.022734 +2023-10-05 21:41:02,022 - Epoch: [125][ 340/ 1236] Overall Loss 0.223647 Objective Loss 0.223647 LR 0.000500 Time 0.022672 +2023-10-05 21:41:02,230 - Epoch: [125][ 350/ 1236] Overall Loss 0.223779 Objective Loss 0.223779 LR 0.000500 Time 0.022617 +2023-10-05 21:41:02,436 - Epoch: [125][ 360/ 1236] Overall Loss 0.223856 Objective Loss 0.223856 LR 0.000500 Time 0.022561 +2023-10-05 21:41:02,643 - Epoch: [125][ 370/ 1236] Overall Loss 0.223773 Objective Loss 0.223773 LR 0.000500 Time 0.022509 +2023-10-05 21:41:02,846 - Epoch: [125][ 380/ 1236] Overall Loss 0.223225 Objective Loss 0.223225 LR 0.000500 Time 0.022449 +2023-10-05 21:41:03,048 - Epoch: [125][ 390/ 1236] Overall Loss 0.223036 Objective Loss 0.223036 LR 0.000500 Time 0.022387 +2023-10-05 21:41:03,250 - Epoch: [125][ 400/ 1236] Overall Loss 0.222505 Objective Loss 0.222505 LR 0.000500 Time 0.022333 +2023-10-05 21:41:03,453 - Epoch: [125][ 410/ 1236] Overall Loss 0.222506 Objective Loss 0.222506 LR 0.000500 Time 0.022280 +2023-10-05 21:41:03,656 - Epoch: [125][ 420/ 1236] Overall Loss 0.222386 Objective Loss 0.222386 LR 0.000500 Time 0.022232 +2023-10-05 21:41:03,857 - Epoch: [125][ 430/ 1236] Overall Loss 0.221851 Objective Loss 0.221851 LR 0.000500 Time 0.022179 +2023-10-05 21:41:04,058 - Epoch: [125][ 440/ 1236] Overall Loss 0.222305 Objective Loss 0.222305 LR 0.000500 Time 0.022132 +2023-10-05 21:41:04,259 - Epoch: [125][ 450/ 1236] Overall Loss 0.222654 Objective Loss 0.222654 LR 0.000500 Time 0.022085 +2023-10-05 21:41:04,462 - Epoch: [125][ 460/ 1236] Overall Loss 0.223041 Objective Loss 0.223041 LR 0.000500 Time 0.022045 +2023-10-05 21:41:04,664 - Epoch: [125][ 470/ 1236] Overall Loss 0.223266 Objective Loss 0.223266 LR 0.000500 Time 0.022003 +2023-10-05 21:41:04,867 - Epoch: [125][ 480/ 1236] Overall Loss 0.223293 Objective Loss 0.223293 LR 0.000500 Time 0.021966 +2023-10-05 21:41:05,067 - Epoch: [125][ 490/ 1236] Overall Loss 0.223041 Objective Loss 0.223041 LR 0.000500 Time 0.021925 +2023-10-05 21:41:05,269 - Epoch: [125][ 500/ 1236] Overall Loss 0.222957 Objective Loss 0.222957 LR 0.000500 Time 0.021890 +2023-10-05 21:41:05,469 - Epoch: [125][ 510/ 1236] Overall Loss 0.223203 Objective Loss 0.223203 LR 0.000500 Time 0.021853 +2023-10-05 21:41:05,671 - Epoch: [125][ 520/ 1236] Overall Loss 0.223250 Objective Loss 0.223250 LR 0.000500 Time 0.021820 +2023-10-05 21:41:05,871 - Epoch: [125][ 530/ 1236] Overall Loss 0.223561 Objective Loss 0.223561 LR 0.000500 Time 0.021785 +2023-10-05 21:41:06,073 - Epoch: [125][ 540/ 1236] Overall Loss 0.224012 Objective Loss 0.224012 LR 0.000500 Time 0.021755 +2023-10-05 21:41:06,273 - Epoch: [125][ 550/ 1236] Overall Loss 0.224635 Objective Loss 0.224635 LR 0.000500 Time 0.021722 +2023-10-05 21:41:06,475 - Epoch: [125][ 560/ 1236] Overall Loss 0.224686 Objective Loss 0.224686 LR 0.000500 Time 0.021694 +2023-10-05 21:41:06,675 - Epoch: [125][ 570/ 1236] Overall Loss 0.224960 Objective Loss 0.224960 LR 0.000500 Time 0.021664 +2023-10-05 21:41:06,877 - Epoch: [125][ 580/ 1236] Overall Loss 0.224857 Objective Loss 0.224857 LR 0.000500 Time 0.021638 +2023-10-05 21:41:07,077 - Epoch: [125][ 590/ 1236] Overall Loss 0.225074 Objective Loss 0.225074 LR 0.000500 Time 0.021610 +2023-10-05 21:41:07,279 - Epoch: [125][ 600/ 1236] Overall Loss 0.224845 Objective Loss 0.224845 LR 0.000500 Time 0.021586 +2023-10-05 21:41:07,479 - Epoch: [125][ 610/ 1236] Overall Loss 0.224924 Objective Loss 0.224924 LR 0.000500 Time 0.021560 +2023-10-05 21:41:07,681 - Epoch: [125][ 620/ 1236] Overall Loss 0.225313 Objective Loss 0.225313 LR 0.000500 Time 0.021537 +2023-10-05 21:41:07,882 - Epoch: [125][ 630/ 1236] Overall Loss 0.225732 Objective Loss 0.225732 LR 0.000500 Time 0.021513 +2023-10-05 21:41:08,084 - Epoch: [125][ 640/ 1236] Overall Loss 0.225872 Objective Loss 0.225872 LR 0.000500 Time 0.021492 +2023-10-05 21:41:08,284 - Epoch: [125][ 650/ 1236] Overall Loss 0.226078 Objective Loss 0.226078 LR 0.000500 Time 0.021469 +2023-10-05 21:41:08,486 - Epoch: [125][ 660/ 1236] Overall Loss 0.226460 Objective Loss 0.226460 LR 0.000500 Time 0.021449 +2023-10-05 21:41:08,686 - Epoch: [125][ 670/ 1236] Overall Loss 0.226709 Objective Loss 0.226709 LR 0.000500 Time 0.021427 +2023-10-05 21:41:08,888 - Epoch: [125][ 680/ 1236] Overall Loss 0.227165 Objective Loss 0.227165 LR 0.000500 Time 0.021408 +2023-10-05 21:41:09,088 - Epoch: [125][ 690/ 1236] Overall Loss 0.227274 Objective Loss 0.227274 LR 0.000500 Time 0.021388 +2023-10-05 21:41:09,290 - Epoch: [125][ 700/ 1236] Overall Loss 0.227218 Objective Loss 0.227218 LR 0.000500 Time 0.021370 +2023-10-05 21:41:09,490 - Epoch: [125][ 710/ 1236] Overall Loss 0.227148 Objective Loss 0.227148 LR 0.000500 Time 0.021350 +2023-10-05 21:41:09,692 - Epoch: [125][ 720/ 1236] Overall Loss 0.227102 Objective Loss 0.227102 LR 0.000500 Time 0.021334 +2023-10-05 21:41:09,893 - Epoch: [125][ 730/ 1236] Overall Loss 0.227268 Objective Loss 0.227268 LR 0.000500 Time 0.021316 +2023-10-05 21:41:10,096 - Epoch: [125][ 740/ 1236] Overall Loss 0.227482 Objective Loss 0.227482 LR 0.000500 Time 0.021302 +2023-10-05 21:41:10,297 - Epoch: [125][ 750/ 1236] Overall Loss 0.227454 Objective Loss 0.227454 LR 0.000500 Time 0.021283 +2023-10-05 21:41:10,500 - Epoch: [125][ 760/ 1236] Overall Loss 0.227452 Objective Loss 0.227452 LR 0.000500 Time 0.021270 +2023-10-05 21:41:10,700 - Epoch: [125][ 770/ 1236] Overall Loss 0.227508 Objective Loss 0.227508 LR 0.000500 Time 0.021253 +2023-10-05 21:41:10,902 - Epoch: [125][ 780/ 1236] Overall Loss 0.227609 Objective Loss 0.227609 LR 0.000500 Time 0.021239 +2023-10-05 21:41:11,102 - Epoch: [125][ 790/ 1236] Overall Loss 0.227710 Objective Loss 0.227710 LR 0.000500 Time 0.021223 +2023-10-05 21:41:11,304 - Epoch: [125][ 800/ 1236] Overall Loss 0.227402 Objective Loss 0.227402 LR 0.000500 Time 0.021210 +2023-10-05 21:41:11,504 - Epoch: [125][ 810/ 1236] Overall Loss 0.227541 Objective Loss 0.227541 LR 0.000500 Time 0.021195 +2023-10-05 21:41:11,706 - Epoch: [125][ 820/ 1236] Overall Loss 0.227814 Objective Loss 0.227814 LR 0.000500 Time 0.021182 +2023-10-05 21:41:11,907 - Epoch: [125][ 830/ 1236] Overall Loss 0.228151 Objective Loss 0.228151 LR 0.000500 Time 0.021168 +2023-10-05 21:41:12,109 - Epoch: [125][ 840/ 1236] Overall Loss 0.228027 Objective Loss 0.228027 LR 0.000500 Time 0.021156 +2023-10-05 21:41:12,309 - Epoch: [125][ 850/ 1236] Overall Loss 0.227751 Objective Loss 0.227751 LR 0.000500 Time 0.021142 +2023-10-05 21:41:12,511 - Epoch: [125][ 860/ 1236] Overall Loss 0.227927 Objective Loss 0.227927 LR 0.000500 Time 0.021131 +2023-10-05 21:41:12,711 - Epoch: [125][ 870/ 1236] Overall Loss 0.228258 Objective Loss 0.228258 LR 0.000500 Time 0.021118 +2023-10-05 21:41:12,913 - Epoch: [125][ 880/ 1236] Overall Loss 0.228419 Objective Loss 0.228419 LR 0.000500 Time 0.021107 +2023-10-05 21:41:13,114 - Epoch: [125][ 890/ 1236] Overall Loss 0.228296 Objective Loss 0.228296 LR 0.000500 Time 0.021095 +2023-10-05 21:41:13,317 - Epoch: [125][ 900/ 1236] Overall Loss 0.228524 Objective Loss 0.228524 LR 0.000500 Time 0.021086 +2023-10-05 21:41:13,519 - Epoch: [125][ 910/ 1236] Overall Loss 0.228590 Objective Loss 0.228590 LR 0.000500 Time 0.021074 +2023-10-05 21:41:13,722 - Epoch: [125][ 920/ 1236] Overall Loss 0.228639 Objective Loss 0.228639 LR 0.000500 Time 0.021065 +2023-10-05 21:41:13,924 - Epoch: [125][ 930/ 1236] Overall Loss 0.228482 Objective Loss 0.228482 LR 0.000500 Time 0.021054 +2023-10-05 21:41:14,127 - Epoch: [125][ 940/ 1236] Overall Loss 0.228423 Objective Loss 0.228423 LR 0.000500 Time 0.021045 +2023-10-05 21:41:14,328 - Epoch: [125][ 950/ 1236] Overall Loss 0.228345 Objective Loss 0.228345 LR 0.000500 Time 0.021035 +2023-10-05 21:41:14,531 - Epoch: [125][ 960/ 1236] Overall Loss 0.228251 Objective Loss 0.228251 LR 0.000500 Time 0.021027 +2023-10-05 21:41:14,732 - Epoch: [125][ 970/ 1236] Overall Loss 0.228460 Objective Loss 0.228460 LR 0.000500 Time 0.021017 +2023-10-05 21:41:14,934 - Epoch: [125][ 980/ 1236] Overall Loss 0.228710 Objective Loss 0.228710 LR 0.000500 Time 0.021008 +2023-10-05 21:41:15,134 - Epoch: [125][ 990/ 1236] Overall Loss 0.228733 Objective Loss 0.228733 LR 0.000500 Time 0.020998 +2023-10-05 21:41:15,336 - Epoch: [125][ 1000/ 1236] Overall Loss 0.228769 Objective Loss 0.228769 LR 0.000500 Time 0.020989 +2023-10-05 21:41:15,536 - Epoch: [125][ 1010/ 1236] Overall Loss 0.228627 Objective Loss 0.228627 LR 0.000500 Time 0.020979 +2023-10-05 21:41:15,738 - Epoch: [125][ 1020/ 1236] Overall Loss 0.228951 Objective Loss 0.228951 LR 0.000500 Time 0.020971 +2023-10-05 21:41:15,938 - Epoch: [125][ 1030/ 1236] Overall Loss 0.229081 Objective Loss 0.229081 LR 0.000500 Time 0.020962 +2023-10-05 21:41:16,140 - Epoch: [125][ 1040/ 1236] Overall Loss 0.229119 Objective Loss 0.229119 LR 0.000500 Time 0.020954 +2023-10-05 21:41:16,341 - Epoch: [125][ 1050/ 1236] Overall Loss 0.229219 Objective Loss 0.229219 LR 0.000500 Time 0.020945 +2023-10-05 21:41:16,544 - Epoch: [125][ 1060/ 1236] Overall Loss 0.229266 Objective Loss 0.229266 LR 0.000500 Time 0.020939 +2023-10-05 21:41:16,746 - Epoch: [125][ 1070/ 1236] Overall Loss 0.229023 Objective Loss 0.229023 LR 0.000500 Time 0.020930 +2023-10-05 21:41:16,949 - Epoch: [125][ 1080/ 1236] Overall Loss 0.229115 Objective Loss 0.229115 LR 0.000500 Time 0.020924 +2023-10-05 21:41:17,151 - Epoch: [125][ 1090/ 1236] Overall Loss 0.229470 Objective Loss 0.229470 LR 0.000500 Time 0.020916 +2023-10-05 21:41:17,354 - Epoch: [125][ 1100/ 1236] Overall Loss 0.229653 Objective Loss 0.229653 LR 0.000500 Time 0.020910 +2023-10-05 21:41:17,556 - Epoch: [125][ 1110/ 1236] Overall Loss 0.229288 Objective Loss 0.229288 LR 0.000500 Time 0.020903 +2023-10-05 21:41:17,759 - Epoch: [125][ 1120/ 1236] Overall Loss 0.229285 Objective Loss 0.229285 LR 0.000500 Time 0.020898 +2023-10-05 21:41:17,960 - Epoch: [125][ 1130/ 1236] Overall Loss 0.229194 Objective Loss 0.229194 LR 0.000500 Time 0.020889 +2023-10-05 21:41:18,162 - Epoch: [125][ 1140/ 1236] Overall Loss 0.229659 Objective Loss 0.229659 LR 0.000500 Time 0.020882 +2023-10-05 21:41:18,362 - Epoch: [125][ 1150/ 1236] Overall Loss 0.229555 Objective Loss 0.229555 LR 0.000500 Time 0.020875 +2023-10-05 21:41:18,564 - Epoch: [125][ 1160/ 1236] Overall Loss 0.229509 Objective Loss 0.229509 LR 0.000500 Time 0.020869 +2023-10-05 21:41:18,765 - Epoch: [125][ 1170/ 1236] Overall Loss 0.229641 Objective Loss 0.229641 LR 0.000500 Time 0.020861 +2023-10-05 21:41:18,967 - Epoch: [125][ 1180/ 1236] Overall Loss 0.229814 Objective Loss 0.229814 LR 0.000500 Time 0.020856 +2023-10-05 21:41:19,167 - Epoch: [125][ 1190/ 1236] Overall Loss 0.229812 Objective Loss 0.229812 LR 0.000500 Time 0.020848 +2023-10-05 21:41:19,369 - Epoch: [125][ 1200/ 1236] Overall Loss 0.229898 Objective Loss 0.229898 LR 0.000500 Time 0.020843 +2023-10-05 21:41:19,570 - Epoch: [125][ 1210/ 1236] Overall Loss 0.229771 Objective Loss 0.229771 LR 0.000500 Time 0.020836 +2023-10-05 21:41:19,772 - Epoch: [125][ 1220/ 1236] Overall Loss 0.229952 Objective Loss 0.229952 LR 0.000500 Time 0.020830 +2023-10-05 21:41:20,025 - Epoch: [125][ 1230/ 1236] Overall Loss 0.230251 Objective Loss 0.230251 LR 0.000500 Time 0.020867 +2023-10-05 21:41:20,143 - Epoch: [125][ 1236/ 1236] Overall Loss 0.230172 Objective Loss 0.230172 Top1 87.169043 Top5 97.963340 LR 0.000500 Time 0.020861 +2023-10-05 21:41:20,281 - --- validate (epoch=125)----------- +2023-10-05 21:41:20,282 - 29943 samples (256 per mini-batch) +2023-10-05 21:41:20,739 - Epoch: [125][ 10/ 117] Loss 0.359020 Top1 83.710938 Top5 97.617188 +2023-10-05 21:41:20,889 - Epoch: [125][ 20/ 117] Loss 0.336772 Top1 84.316406 Top5 97.832031 +2023-10-05 21:41:21,035 - Epoch: [125][ 30/ 117] Loss 0.330456 Top1 84.388021 Top5 97.773438 +2023-10-05 21:41:21,184 - Epoch: [125][ 40/ 117] Loss 0.327072 Top1 84.228516 Top5 97.666016 +2023-10-05 21:41:21,330 - Epoch: [125][ 50/ 117] Loss 0.325884 Top1 84.171875 Top5 97.718750 +2023-10-05 21:41:21,478 - Epoch: [125][ 60/ 117] Loss 0.332087 Top1 84.218750 Top5 97.799479 +2023-10-05 21:41:21,627 - Epoch: [125][ 70/ 117] Loss 0.334060 Top1 84.174107 Top5 97.812500 +2023-10-05 21:41:21,777 - Epoch: [125][ 80/ 117] Loss 0.332832 Top1 84.106445 Top5 97.851562 +2023-10-05 21:41:21,924 - Epoch: [125][ 90/ 117] Loss 0.330151 Top1 84.210069 Top5 97.799479 +2023-10-05 21:41:22,074 - Epoch: [125][ 100/ 117] Loss 0.328838 Top1 84.226562 Top5 97.816406 +2023-10-05 21:41:22,228 - Epoch: [125][ 110/ 117] Loss 0.329081 Top1 84.232955 Top5 97.848011 +2023-10-05 21:41:22,313 - Epoch: [125][ 117/ 117] Loss 0.327439 Top1 84.223358 Top5 97.849247 +2023-10-05 21:41:22,464 - ==> Top1: 84.223 Top5: 97.849 Loss: 0.327 + +2023-10-05 21:41:22,464 - ==> Confusion: +[[ 943 0 7 2 1 1 0 1 7 62 1 1 1 3 4 0 2 1 0 1 12] + [ 0 1063 2 0 8 14 1 12 2 1 2 1 0 1 2 2 3 1 8 1 7] + [ 2 2 956 20 0 0 20 9 0 2 5 3 5 2 2 4 3 0 9 4 8] + [ 4 0 11 977 1 2 0 3 5 1 12 0 2 3 26 3 1 4 18 1 15] + [ 25 7 1 0 964 2 1 0 0 10 0 2 0 3 11 3 11 2 1 2 5] + [ 5 61 0 1 2 967 2 18 1 1 2 9 1 16 6 1 2 0 5 3 13] + [ 0 5 31 0 1 0 1120 6 0 0 5 2 1 2 1 5 0 3 1 3 5] + [ 4 27 15 0 3 31 4 1049 0 1 4 10 2 1 0 0 1 0 48 6 12] + [ 19 2 1 1 0 2 1 0 967 38 11 2 1 14 15 4 1 0 7 2 1] + [ 108 0 5 0 4 6 2 0 34 913 1 1 0 23 4 8 1 1 0 1 7] + [ 4 5 12 5 2 1 5 5 11 2 960 1 0 12 5 0 1 1 7 2 12] + [ 2 1 1 0 0 8 0 5 0 1 0 967 19 6 0 2 2 13 0 5 3] + [ 0 1 3 7 0 2 0 1 0 1 1 44 966 2 1 7 3 15 4 2 8] + [ 4 0 1 2 2 5 1 0 12 9 6 3 2 1051 6 5 1 0 0 0 9] + [ 13 3 4 10 7 1 0 0 32 6 0 0 2 1 1002 0 1 1 12 0 6] + [ 0 2 2 1 3 1 1 0 0 0 0 12 5 1 0 1074 11 9 0 7 5] + [ 2 14 1 1 5 3 0 0 1 0 0 2 1 2 3 11 1098 0 2 4 11] + [ 0 1 0 6 0 0 1 0 2 0 0 4 20 0 2 5 0 990 1 3 3] + [ 1 10 5 14 1 0 0 19 2 3 5 1 0 0 10 0 1 0 989 0 7] + [ 0 3 4 3 1 7 13 12 0 0 2 13 3 1 0 8 12 1 2 1055 12] + [ 149 205 150 78 80 108 42 87 134 81 144 126 316 283 151 56 129 75 169 194 5148]] + +2023-10-05 21:41:22,466 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:41:22,466 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:41:22,471 - + +2023-10-05 21:41:22,471 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:41:23,559 - Epoch: [126][ 10/ 1236] Overall Loss 0.244389 Objective Loss 0.244389 LR 0.000500 Time 0.108695 +2023-10-05 21:41:23,758 - Epoch: [126][ 20/ 1236] Overall Loss 0.236660 Objective Loss 0.236660 LR 0.000500 Time 0.064299 +2023-10-05 21:41:23,956 - Epoch: [126][ 30/ 1236] Overall Loss 0.232557 Objective Loss 0.232557 LR 0.000500 Time 0.049456 +2023-10-05 21:41:24,156 - Epoch: [126][ 40/ 1236] Overall Loss 0.230744 Objective Loss 0.230744 LR 0.000500 Time 0.042073 +2023-10-05 21:41:24,354 - Epoch: [126][ 50/ 1236] Overall Loss 0.235352 Objective Loss 0.235352 LR 0.000500 Time 0.037612 +2023-10-05 21:41:24,553 - Epoch: [126][ 60/ 1236] Overall Loss 0.230853 Objective Loss 0.230853 LR 0.000500 Time 0.034662 +2023-10-05 21:41:24,752 - Epoch: [126][ 70/ 1236] Overall Loss 0.228525 Objective Loss 0.228525 LR 0.000500 Time 0.032541 +2023-10-05 21:41:24,951 - Epoch: [126][ 80/ 1236] Overall Loss 0.228003 Objective Loss 0.228003 LR 0.000500 Time 0.030963 +2023-10-05 21:41:25,150 - Epoch: [126][ 90/ 1236] Overall Loss 0.226609 Objective Loss 0.226609 LR 0.000500 Time 0.029724 +2023-10-05 21:41:25,349 - Epoch: [126][ 100/ 1236] Overall Loss 0.225585 Objective Loss 0.225585 LR 0.000500 Time 0.028743 +2023-10-05 21:41:25,547 - Epoch: [126][ 110/ 1236] Overall Loss 0.225046 Objective Loss 0.225046 LR 0.000500 Time 0.027927 +2023-10-05 21:41:25,747 - Epoch: [126][ 120/ 1236] Overall Loss 0.225828 Objective Loss 0.225828 LR 0.000500 Time 0.027255 +2023-10-05 21:41:25,945 - Epoch: [126][ 130/ 1236] Overall Loss 0.227175 Objective Loss 0.227175 LR 0.000500 Time 0.026683 +2023-10-05 21:41:26,144 - Epoch: [126][ 140/ 1236] Overall Loss 0.225691 Objective Loss 0.225691 LR 0.000500 Time 0.026197 +2023-10-05 21:41:26,342 - Epoch: [126][ 150/ 1236] Overall Loss 0.226315 Objective Loss 0.226315 LR 0.000500 Time 0.025770 +2023-10-05 21:41:26,542 - Epoch: [126][ 160/ 1236] Overall Loss 0.225754 Objective Loss 0.225754 LR 0.000500 Time 0.025405 +2023-10-05 21:41:26,740 - Epoch: [126][ 170/ 1236] Overall Loss 0.226689 Objective Loss 0.226689 LR 0.000500 Time 0.025071 +2023-10-05 21:41:26,939 - Epoch: [126][ 180/ 1236] Overall Loss 0.227774 Objective Loss 0.227774 LR 0.000500 Time 0.024784 +2023-10-05 21:41:27,137 - Epoch: [126][ 190/ 1236] Overall Loss 0.228298 Objective Loss 0.228298 LR 0.000500 Time 0.024519 +2023-10-05 21:41:27,336 - Epoch: [126][ 200/ 1236] Overall Loss 0.228187 Objective Loss 0.228187 LR 0.000500 Time 0.024285 +2023-10-05 21:41:27,534 - Epoch: [126][ 210/ 1236] Overall Loss 0.228401 Objective Loss 0.228401 LR 0.000500 Time 0.024073 +2023-10-05 21:41:27,734 - Epoch: [126][ 220/ 1236] Overall Loss 0.228047 Objective Loss 0.228047 LR 0.000500 Time 0.023886 +2023-10-05 21:41:27,933 - Epoch: [126][ 230/ 1236] Overall Loss 0.227113 Objective Loss 0.227113 LR 0.000500 Time 0.023708 +2023-10-05 21:41:28,132 - Epoch: [126][ 240/ 1236] Overall Loss 0.225690 Objective Loss 0.225690 LR 0.000500 Time 0.023549 +2023-10-05 21:41:28,330 - Epoch: [126][ 250/ 1236] Overall Loss 0.224952 Objective Loss 0.224952 LR 0.000500 Time 0.023400 +2023-10-05 21:41:28,530 - Epoch: [126][ 260/ 1236] Overall Loss 0.224063 Objective Loss 0.224063 LR 0.000500 Time 0.023266 +2023-10-05 21:41:28,728 - Epoch: [126][ 270/ 1236] Overall Loss 0.224579 Objective Loss 0.224579 LR 0.000500 Time 0.023138 +2023-10-05 21:41:28,928 - Epoch: [126][ 280/ 1236] Overall Loss 0.224019 Objective Loss 0.224019 LR 0.000500 Time 0.023024 +2023-10-05 21:41:29,127 - Epoch: [126][ 290/ 1236] Overall Loss 0.223574 Objective Loss 0.223574 LR 0.000500 Time 0.022913 +2023-10-05 21:41:29,326 - Epoch: [126][ 300/ 1236] Overall Loss 0.223941 Objective Loss 0.223941 LR 0.000500 Time 0.022813 +2023-10-05 21:41:29,525 - Epoch: [126][ 310/ 1236] Overall Loss 0.223190 Objective Loss 0.223190 LR 0.000500 Time 0.022717 +2023-10-05 21:41:29,725 - Epoch: [126][ 320/ 1236] Overall Loss 0.224243 Objective Loss 0.224243 LR 0.000500 Time 0.022631 +2023-10-05 21:41:29,924 - Epoch: [126][ 330/ 1236] Overall Loss 0.223896 Objective Loss 0.223896 LR 0.000500 Time 0.022545 +2023-10-05 21:41:30,123 - Epoch: [126][ 340/ 1236] Overall Loss 0.224104 Objective Loss 0.224104 LR 0.000500 Time 0.022469 +2023-10-05 21:41:30,322 - Epoch: [126][ 350/ 1236] Overall Loss 0.224155 Objective Loss 0.224155 LR 0.000500 Time 0.022394 +2023-10-05 21:41:30,521 - Epoch: [126][ 360/ 1236] Overall Loss 0.224876 Objective Loss 0.224876 LR 0.000500 Time 0.022323 +2023-10-05 21:41:30,719 - Epoch: [126][ 370/ 1236] Overall Loss 0.224104 Objective Loss 0.224104 LR 0.000500 Time 0.022254 +2023-10-05 21:41:30,918 - Epoch: [126][ 380/ 1236] Overall Loss 0.224615 Objective Loss 0.224615 LR 0.000500 Time 0.022192 +2023-10-05 21:41:31,117 - Epoch: [126][ 390/ 1236] Overall Loss 0.224907 Objective Loss 0.224907 LR 0.000500 Time 0.022132 +2023-10-05 21:41:31,317 - Epoch: [126][ 400/ 1236] Overall Loss 0.224582 Objective Loss 0.224582 LR 0.000500 Time 0.022078 +2023-10-05 21:41:31,515 - Epoch: [126][ 410/ 1236] Overall Loss 0.224389 Objective Loss 0.224389 LR 0.000500 Time 0.022022 +2023-10-05 21:41:31,715 - Epoch: [126][ 420/ 1236] Overall Loss 0.224622 Objective Loss 0.224622 LR 0.000500 Time 0.021973 +2023-10-05 21:41:31,914 - Epoch: [126][ 430/ 1236] Overall Loss 0.224492 Objective Loss 0.224492 LR 0.000500 Time 0.021923 +2023-10-05 21:41:32,113 - Epoch: [126][ 440/ 1236] Overall Loss 0.224060 Objective Loss 0.224060 LR 0.000500 Time 0.021877 +2023-10-05 21:41:32,312 - Epoch: [126][ 450/ 1236] Overall Loss 0.224115 Objective Loss 0.224115 LR 0.000500 Time 0.021831 +2023-10-05 21:41:32,511 - Epoch: [126][ 460/ 1236] Overall Loss 0.224242 Objective Loss 0.224242 LR 0.000500 Time 0.021790 +2023-10-05 21:41:32,710 - Epoch: [126][ 470/ 1236] Overall Loss 0.224226 Objective Loss 0.224226 LR 0.000500 Time 0.021748 +2023-10-05 21:41:32,910 - Epoch: [126][ 480/ 1236] Overall Loss 0.224859 Objective Loss 0.224859 LR 0.000500 Time 0.021710 +2023-10-05 21:41:33,108 - Epoch: [126][ 490/ 1236] Overall Loss 0.225265 Objective Loss 0.225265 LR 0.000500 Time 0.021671 +2023-10-05 21:41:33,308 - Epoch: [126][ 500/ 1236] Overall Loss 0.225340 Objective Loss 0.225340 LR 0.000500 Time 0.021638 +2023-10-05 21:41:33,506 - Epoch: [126][ 510/ 1236] Overall Loss 0.225671 Objective Loss 0.225671 LR 0.000500 Time 0.021601 +2023-10-05 21:41:33,706 - Epoch: [126][ 520/ 1236] Overall Loss 0.225761 Objective Loss 0.225761 LR 0.000500 Time 0.021569 +2023-10-05 21:41:33,905 - Epoch: [126][ 530/ 1236] Overall Loss 0.226697 Objective Loss 0.226697 LR 0.000500 Time 0.021537 +2023-10-05 21:41:34,105 - Epoch: [126][ 540/ 1236] Overall Loss 0.226637 Objective Loss 0.226637 LR 0.000500 Time 0.021507 +2023-10-05 21:41:34,303 - Epoch: [126][ 550/ 1236] Overall Loss 0.226662 Objective Loss 0.226662 LR 0.000500 Time 0.021476 +2023-10-05 21:41:34,503 - Epoch: [126][ 560/ 1236] Overall Loss 0.225828 Objective Loss 0.225828 LR 0.000500 Time 0.021450 +2023-10-05 21:41:34,701 - Epoch: [126][ 570/ 1236] Overall Loss 0.226200 Objective Loss 0.226200 LR 0.000500 Time 0.021420 +2023-10-05 21:41:34,901 - Epoch: [126][ 580/ 1236] Overall Loss 0.226452 Objective Loss 0.226452 LR 0.000500 Time 0.021395 +2023-10-05 21:41:35,100 - Epoch: [126][ 590/ 1236] Overall Loss 0.226424 Objective Loss 0.226424 LR 0.000500 Time 0.021368 +2023-10-05 21:41:35,300 - Epoch: [126][ 600/ 1236] Overall Loss 0.226504 Objective Loss 0.226504 LR 0.000500 Time 0.021345 +2023-10-05 21:41:35,498 - Epoch: [126][ 610/ 1236] Overall Loss 0.226229 Objective Loss 0.226229 LR 0.000500 Time 0.021319 +2023-10-05 21:41:35,698 - Epoch: [126][ 620/ 1236] Overall Loss 0.225896 Objective Loss 0.225896 LR 0.000500 Time 0.021297 +2023-10-05 21:41:35,897 - Epoch: [126][ 630/ 1236] Overall Loss 0.226091 Objective Loss 0.226091 LR 0.000500 Time 0.021274 +2023-10-05 21:41:36,097 - Epoch: [126][ 640/ 1236] Overall Loss 0.225938 Objective Loss 0.225938 LR 0.000500 Time 0.021254 +2023-10-05 21:41:36,295 - Epoch: [126][ 650/ 1236] Overall Loss 0.225595 Objective Loss 0.225595 LR 0.000500 Time 0.021232 +2023-10-05 21:41:36,495 - Epoch: [126][ 660/ 1236] Overall Loss 0.225614 Objective Loss 0.225614 LR 0.000500 Time 0.021213 +2023-10-05 21:41:36,694 - Epoch: [126][ 670/ 1236] Overall Loss 0.225490 Objective Loss 0.225490 LR 0.000500 Time 0.021191 +2023-10-05 21:41:36,894 - Epoch: [126][ 680/ 1236] Overall Loss 0.225701 Objective Loss 0.225701 LR 0.000500 Time 0.021173 +2023-10-05 21:41:37,092 - Epoch: [126][ 690/ 1236] Overall Loss 0.225741 Objective Loss 0.225741 LR 0.000500 Time 0.021154 +2023-10-05 21:41:37,292 - Epoch: [126][ 700/ 1236] Overall Loss 0.225980 Objective Loss 0.225980 LR 0.000500 Time 0.021137 +2023-10-05 21:41:37,491 - Epoch: [126][ 710/ 1236] Overall Loss 0.226095 Objective Loss 0.226095 LR 0.000500 Time 0.021118 +2023-10-05 21:41:37,690 - Epoch: [126][ 720/ 1236] Overall Loss 0.225997 Objective Loss 0.225997 LR 0.000500 Time 0.021102 +2023-10-05 21:41:37,889 - Epoch: [126][ 730/ 1236] Overall Loss 0.225548 Objective Loss 0.225548 LR 0.000500 Time 0.021085 +2023-10-05 21:41:38,089 - Epoch: [126][ 740/ 1236] Overall Loss 0.225687 Objective Loss 0.225687 LR 0.000500 Time 0.021069 +2023-10-05 21:41:38,288 - Epoch: [126][ 750/ 1236] Overall Loss 0.225686 Objective Loss 0.225686 LR 0.000500 Time 0.021053 +2023-10-05 21:41:38,487 - Epoch: [126][ 760/ 1236] Overall Loss 0.226175 Objective Loss 0.226175 LR 0.000500 Time 0.021038 +2023-10-05 21:41:38,686 - Epoch: [126][ 770/ 1236] Overall Loss 0.225676 Objective Loss 0.225676 LR 0.000500 Time 0.021022 +2023-10-05 21:41:38,886 - Epoch: [126][ 780/ 1236] Overall Loss 0.225825 Objective Loss 0.225825 LR 0.000500 Time 0.021008 +2023-10-05 21:41:39,084 - Epoch: [126][ 790/ 1236] Overall Loss 0.225534 Objective Loss 0.225534 LR 0.000500 Time 0.020993 +2023-10-05 21:41:39,284 - Epoch: [126][ 800/ 1236] Overall Loss 0.225432 Objective Loss 0.225432 LR 0.000500 Time 0.020981 +2023-10-05 21:41:39,483 - Epoch: [126][ 810/ 1236] Overall Loss 0.225380 Objective Loss 0.225380 LR 0.000500 Time 0.020966 +2023-10-05 21:41:39,683 - Epoch: [126][ 820/ 1236] Overall Loss 0.225664 Objective Loss 0.225664 LR 0.000500 Time 0.020954 +2023-10-05 21:41:39,881 - Epoch: [126][ 830/ 1236] Overall Loss 0.225579 Objective Loss 0.225579 LR 0.000500 Time 0.020940 +2023-10-05 21:41:40,081 - Epoch: [126][ 840/ 1236] Overall Loss 0.225831 Objective Loss 0.225831 LR 0.000500 Time 0.020929 +2023-10-05 21:41:40,280 - Epoch: [126][ 850/ 1236] Overall Loss 0.226035 Objective Loss 0.226035 LR 0.000500 Time 0.020915 +2023-10-05 21:41:40,479 - Epoch: [126][ 860/ 1236] Overall Loss 0.226175 Objective Loss 0.226175 LR 0.000500 Time 0.020904 +2023-10-05 21:41:40,678 - Epoch: [126][ 870/ 1236] Overall Loss 0.225996 Objective Loss 0.225996 LR 0.000500 Time 0.020892 +2023-10-05 21:41:40,878 - Epoch: [126][ 880/ 1236] Overall Loss 0.225747 Objective Loss 0.225747 LR 0.000500 Time 0.020882 +2023-10-05 21:41:41,077 - Epoch: [126][ 890/ 1236] Overall Loss 0.225574 Objective Loss 0.225574 LR 0.000500 Time 0.020869 +2023-10-05 21:41:41,277 - Epoch: [126][ 900/ 1236] Overall Loss 0.225286 Objective Loss 0.225286 LR 0.000500 Time 0.020859 +2023-10-05 21:41:41,475 - Epoch: [126][ 910/ 1236] Overall Loss 0.225448 Objective Loss 0.225448 LR 0.000500 Time 0.020848 +2023-10-05 21:41:41,675 - Epoch: [126][ 920/ 1236] Overall Loss 0.225672 Objective Loss 0.225672 LR 0.000500 Time 0.020838 +2023-10-05 21:41:41,873 - Epoch: [126][ 930/ 1236] Overall Loss 0.225776 Objective Loss 0.225776 LR 0.000500 Time 0.020827 +2023-10-05 21:41:42,073 - Epoch: [126][ 940/ 1236] Overall Loss 0.225787 Objective Loss 0.225787 LR 0.000500 Time 0.020818 +2023-10-05 21:41:42,272 - Epoch: [126][ 950/ 1236] Overall Loss 0.225812 Objective Loss 0.225812 LR 0.000500 Time 0.020807 +2023-10-05 21:41:42,472 - Epoch: [126][ 960/ 1236] Overall Loss 0.225867 Objective Loss 0.225867 LR 0.000500 Time 0.020798 +2023-10-05 21:41:42,671 - Epoch: [126][ 970/ 1236] Overall Loss 0.225962 Objective Loss 0.225962 LR 0.000500 Time 0.020789 +2023-10-05 21:41:42,870 - Epoch: [126][ 980/ 1236] Overall Loss 0.225974 Objective Loss 0.225974 LR 0.000500 Time 0.020780 +2023-10-05 21:41:43,069 - Epoch: [126][ 990/ 1236] Overall Loss 0.225786 Objective Loss 0.225786 LR 0.000500 Time 0.020770 +2023-10-05 21:41:43,269 - Epoch: [126][ 1000/ 1236] Overall Loss 0.225826 Objective Loss 0.225826 LR 0.000500 Time 0.020762 +2023-10-05 21:41:43,468 - Epoch: [126][ 1010/ 1236] Overall Loss 0.225858 Objective Loss 0.225858 LR 0.000500 Time 0.020753 +2023-10-05 21:41:43,668 - Epoch: [126][ 1020/ 1236] Overall Loss 0.225834 Objective Loss 0.225834 LR 0.000500 Time 0.020745 +2023-10-05 21:41:43,866 - Epoch: [126][ 1030/ 1236] Overall Loss 0.225563 Objective Loss 0.225563 LR 0.000500 Time 0.020736 +2023-10-05 21:41:44,066 - Epoch: [126][ 1040/ 1236] Overall Loss 0.225706 Objective Loss 0.225706 LR 0.000500 Time 0.020729 +2023-10-05 21:41:44,265 - Epoch: [126][ 1050/ 1236] Overall Loss 0.225420 Objective Loss 0.225420 LR 0.000500 Time 0.020721 +2023-10-05 21:41:44,465 - Epoch: [126][ 1060/ 1236] Overall Loss 0.225520 Objective Loss 0.225520 LR 0.000500 Time 0.020713 +2023-10-05 21:41:44,663 - Epoch: [126][ 1070/ 1236] Overall Loss 0.225707 Objective Loss 0.225707 LR 0.000500 Time 0.020705 +2023-10-05 21:41:44,863 - Epoch: [126][ 1080/ 1236] Overall Loss 0.225746 Objective Loss 0.225746 LR 0.000500 Time 0.020698 +2023-10-05 21:41:45,062 - Epoch: [126][ 1090/ 1236] Overall Loss 0.225638 Objective Loss 0.225638 LR 0.000500 Time 0.020690 +2023-10-05 21:41:45,262 - Epoch: [126][ 1100/ 1236] Overall Loss 0.225566 Objective Loss 0.225566 LR 0.000500 Time 0.020683 +2023-10-05 21:41:45,460 - Epoch: [126][ 1110/ 1236] Overall Loss 0.225842 Objective Loss 0.225842 LR 0.000500 Time 0.020676 +2023-10-05 21:41:45,660 - Epoch: [126][ 1120/ 1236] Overall Loss 0.226108 Objective Loss 0.226108 LR 0.000500 Time 0.020669 +2023-10-05 21:41:45,859 - Epoch: [126][ 1130/ 1236] Overall Loss 0.226093 Objective Loss 0.226093 LR 0.000500 Time 0.020662 +2023-10-05 21:41:46,059 - Epoch: [126][ 1140/ 1236] Overall Loss 0.226254 Objective Loss 0.226254 LR 0.000500 Time 0.020656 +2023-10-05 21:41:46,258 - Epoch: [126][ 1150/ 1236] Overall Loss 0.226252 Objective Loss 0.226252 LR 0.000500 Time 0.020649 +2023-10-05 21:41:46,458 - Epoch: [126][ 1160/ 1236] Overall Loss 0.226417 Objective Loss 0.226417 LR 0.000500 Time 0.020643 +2023-10-05 21:41:46,656 - Epoch: [126][ 1170/ 1236] Overall Loss 0.226561 Objective Loss 0.226561 LR 0.000500 Time 0.020636 +2023-10-05 21:41:46,856 - Epoch: [126][ 1180/ 1236] Overall Loss 0.226509 Objective Loss 0.226509 LR 0.000500 Time 0.020630 +2023-10-05 21:41:47,055 - Epoch: [126][ 1190/ 1236] Overall Loss 0.226533 Objective Loss 0.226533 LR 0.000500 Time 0.020624 +2023-10-05 21:41:47,256 - Epoch: [126][ 1200/ 1236] Overall Loss 0.226601 Objective Loss 0.226601 LR 0.000500 Time 0.020619 +2023-10-05 21:41:47,455 - Epoch: [126][ 1210/ 1236] Overall Loss 0.226593 Objective Loss 0.226593 LR 0.000500 Time 0.020613 +2023-10-05 21:41:47,656 - Epoch: [126][ 1220/ 1236] Overall Loss 0.226850 Objective Loss 0.226850 LR 0.000500 Time 0.020608 +2023-10-05 21:41:47,906 - Epoch: [126][ 1230/ 1236] Overall Loss 0.226995 Objective Loss 0.226995 LR 0.000500 Time 0.020644 +2023-10-05 21:41:48,023 - Epoch: [126][ 1236/ 1236] Overall Loss 0.226882 Objective Loss 0.226882 Top1 87.983707 Top5 99.592668 LR 0.000500 Time 0.020638 +2023-10-05 21:41:48,157 - --- validate (epoch=126)----------- +2023-10-05 21:41:48,157 - 29943 samples (256 per mini-batch) +2023-10-05 21:41:48,612 - Epoch: [126][ 10/ 117] Loss 0.319251 Top1 84.453125 Top5 98.203125 +2023-10-05 21:41:48,761 - Epoch: [126][ 20/ 117] Loss 0.310702 Top1 84.746094 Top5 98.007812 +2023-10-05 21:41:48,907 - Epoch: [126][ 30/ 117] Loss 0.318628 Top1 84.023438 Top5 97.864583 +2023-10-05 21:41:49,055 - Epoch: [126][ 40/ 117] Loss 0.326716 Top1 83.925781 Top5 97.890625 +2023-10-05 21:41:49,201 - Epoch: [126][ 50/ 117] Loss 0.319539 Top1 84.000000 Top5 98.039062 +2023-10-05 21:41:49,347 - Epoch: [126][ 60/ 117] Loss 0.318585 Top1 84.231771 Top5 97.988281 +2023-10-05 21:41:49,494 - Epoch: [126][ 70/ 117] Loss 0.324062 Top1 84.068080 Top5 97.957589 +2023-10-05 21:41:49,642 - Epoch: [126][ 80/ 117] Loss 0.326531 Top1 83.994141 Top5 97.963867 +2023-10-05 21:41:49,789 - Epoch: [126][ 90/ 117] Loss 0.326660 Top1 84.062500 Top5 97.968750 +2023-10-05 21:41:49,936 - Epoch: [126][ 100/ 117] Loss 0.325009 Top1 84.046875 Top5 97.960938 +2023-10-05 21:41:50,089 - Epoch: [126][ 110/ 117] Loss 0.321439 Top1 84.140625 Top5 98.007812 +2023-10-05 21:41:50,174 - Epoch: [126][ 117/ 117] Loss 0.320762 Top1 84.166583 Top5 97.986174 +2023-10-05 21:41:50,310 - ==> Top1: 84.167 Top5: 97.986 Loss: 0.321 + +2023-10-05 21:41:50,311 - ==> Confusion: +[[ 918 2 6 1 12 2 0 0 9 69 2 0 3 4 4 1 4 0 2 0 11] + [ 2 1048 1 0 4 19 1 19 0 0 3 1 0 0 1 3 3 1 17 3 5] + [ 1 1 962 11 0 0 27 8 0 0 5 1 10 3 0 1 3 1 11 1 10] + [ 2 1 14 975 0 4 2 1 3 0 7 0 6 0 18 4 0 2 32 2 16] + [ 18 6 1 0 970 4 0 1 0 8 1 2 4 1 12 2 10 1 1 2 6] + [ 5 44 1 3 1 989 2 22 0 0 2 6 2 9 6 1 3 1 4 6 9] + [ 0 5 27 0 0 2 1110 10 0 0 7 1 3 0 1 5 1 2 2 7 8] + [ 2 23 18 1 1 25 5 1056 0 0 5 8 4 2 2 2 1 0 47 6 10] + [ 16 2 1 0 0 1 0 0 964 38 18 2 2 12 14 2 2 1 6 3 5] + [ 98 1 5 0 9 4 0 0 33 919 2 1 1 24 8 4 2 0 0 2 6] + [ 2 4 13 8 1 1 1 6 11 0 962 3 0 13 4 1 0 0 8 3 12] + [ 0 1 3 0 0 12 0 3 0 1 0 945 28 6 0 1 3 16 0 11 5] + [ 0 1 4 6 0 1 0 3 1 0 2 28 979 4 0 6 2 12 5 3 11] + [ 2 1 2 0 1 16 1 1 13 7 7 4 2 1044 4 0 2 1 1 1 9] + [ 11 1 2 16 3 1 0 0 22 3 2 1 3 3 1004 0 1 1 16 0 11] + [ 0 2 4 1 6 1 1 0 0 0 0 9 4 2 0 1067 12 5 1 11 8] + [ 0 12 3 1 8 4 0 2 1 0 0 3 0 1 4 12 1097 0 2 5 6] + [ 0 0 0 3 0 1 2 0 2 0 0 3 19 0 0 5 0 997 2 2 2] + [ 1 3 7 10 1 0 0 11 2 1 1 0 2 0 8 0 0 0 1015 1 5] + [ 0 2 5 1 1 4 8 8 1 0 2 10 6 0 0 4 8 2 2 1083 5] + [ 110 206 156 93 81 146 34 99 99 65 169 107 331 263 146 48 164 69 212 209 5098]] + +2023-10-05 21:41:50,312 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:41:50,312 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:41:50,318 - + +2023-10-05 21:41:50,318 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:41:51,294 - Epoch: [127][ 10/ 1236] Overall Loss 0.203181 Objective Loss 0.203181 LR 0.000500 Time 0.097534 +2023-10-05 21:41:51,492 - Epoch: [127][ 20/ 1236] Overall Loss 0.208359 Objective Loss 0.208359 LR 0.000500 Time 0.058674 +2023-10-05 21:41:51,689 - Epoch: [127][ 30/ 1236] Overall Loss 0.213725 Objective Loss 0.213725 LR 0.000500 Time 0.045658 +2023-10-05 21:41:51,887 - Epoch: [127][ 40/ 1236] Overall Loss 0.218386 Objective Loss 0.218386 LR 0.000500 Time 0.039190 +2023-10-05 21:41:52,084 - Epoch: [127][ 50/ 1236] Overall Loss 0.223892 Objective Loss 0.223892 LR 0.000500 Time 0.035293 +2023-10-05 21:41:52,283 - Epoch: [127][ 60/ 1236] Overall Loss 0.221422 Objective Loss 0.221422 LR 0.000500 Time 0.032716 +2023-10-05 21:41:52,480 - Epoch: [127][ 70/ 1236] Overall Loss 0.225363 Objective Loss 0.225363 LR 0.000500 Time 0.030852 +2023-10-05 21:41:52,679 - Epoch: [127][ 80/ 1236] Overall Loss 0.225789 Objective Loss 0.225789 LR 0.000500 Time 0.029474 +2023-10-05 21:41:52,876 - Epoch: [127][ 90/ 1236] Overall Loss 0.226562 Objective Loss 0.226562 LR 0.000500 Time 0.028393 +2023-10-05 21:41:53,075 - Epoch: [127][ 100/ 1236] Overall Loss 0.226030 Objective Loss 0.226030 LR 0.000500 Time 0.027539 +2023-10-05 21:41:53,273 - Epoch: [127][ 110/ 1236] Overall Loss 0.227518 Objective Loss 0.227518 LR 0.000500 Time 0.026826 +2023-10-05 21:41:53,471 - Epoch: [127][ 120/ 1236] Overall Loss 0.226960 Objective Loss 0.226960 LR 0.000500 Time 0.026244 +2023-10-05 21:41:53,668 - Epoch: [127][ 130/ 1236] Overall Loss 0.226689 Objective Loss 0.226689 LR 0.000500 Time 0.025739 +2023-10-05 21:41:53,867 - Epoch: [127][ 140/ 1236] Overall Loss 0.226399 Objective Loss 0.226399 LR 0.000500 Time 0.025320 +2023-10-05 21:41:54,065 - Epoch: [127][ 150/ 1236] Overall Loss 0.226406 Objective Loss 0.226406 LR 0.000500 Time 0.024945 +2023-10-05 21:41:54,264 - Epoch: [127][ 160/ 1236] Overall Loss 0.227631 Objective Loss 0.227631 LR 0.000500 Time 0.024626 +2023-10-05 21:41:54,461 - Epoch: [127][ 170/ 1236] Overall Loss 0.228834 Objective Loss 0.228834 LR 0.000500 Time 0.024337 +2023-10-05 21:41:54,660 - Epoch: [127][ 180/ 1236] Overall Loss 0.228571 Objective Loss 0.228571 LR 0.000500 Time 0.024087 +2023-10-05 21:41:54,857 - Epoch: [127][ 190/ 1236] Overall Loss 0.229180 Objective Loss 0.229180 LR 0.000500 Time 0.023857 +2023-10-05 21:41:55,056 - Epoch: [127][ 200/ 1236] Overall Loss 0.228916 Objective Loss 0.228916 LR 0.000500 Time 0.023657 +2023-10-05 21:41:55,254 - Epoch: [127][ 210/ 1236] Overall Loss 0.229630 Objective Loss 0.229630 LR 0.000500 Time 0.023469 +2023-10-05 21:41:55,452 - Epoch: [127][ 220/ 1236] Overall Loss 0.229906 Objective Loss 0.229906 LR 0.000500 Time 0.023303 +2023-10-05 21:41:55,649 - Epoch: [127][ 230/ 1236] Overall Loss 0.229101 Objective Loss 0.229101 LR 0.000500 Time 0.023146 +2023-10-05 21:41:55,848 - Epoch: [127][ 240/ 1236] Overall Loss 0.228664 Objective Loss 0.228664 LR 0.000500 Time 0.023008 +2023-10-05 21:41:56,045 - Epoch: [127][ 250/ 1236] Overall Loss 0.227918 Objective Loss 0.227918 LR 0.000500 Time 0.022876 +2023-10-05 21:41:56,244 - Epoch: [127][ 260/ 1236] Overall Loss 0.227093 Objective Loss 0.227093 LR 0.000500 Time 0.022759 +2023-10-05 21:41:56,441 - Epoch: [127][ 270/ 1236] Overall Loss 0.226674 Objective Loss 0.226674 LR 0.000500 Time 0.022645 +2023-10-05 21:41:56,640 - Epoch: [127][ 280/ 1236] Overall Loss 0.225394 Objective Loss 0.225394 LR 0.000500 Time 0.022545 +2023-10-05 21:41:56,837 - Epoch: [127][ 290/ 1236] Overall Loss 0.225623 Objective Loss 0.225623 LR 0.000500 Time 0.022447 +2023-10-05 21:41:57,036 - Epoch: [127][ 300/ 1236] Overall Loss 0.225720 Objective Loss 0.225720 LR 0.000500 Time 0.022360 +2023-10-05 21:41:57,233 - Epoch: [127][ 310/ 1236] Overall Loss 0.225419 Objective Loss 0.225419 LR 0.000500 Time 0.022274 +2023-10-05 21:41:57,432 - Epoch: [127][ 320/ 1236] Overall Loss 0.225500 Objective Loss 0.225500 LR 0.000500 Time 0.022198 +2023-10-05 21:41:57,630 - Epoch: [127][ 330/ 1236] Overall Loss 0.225496 Objective Loss 0.225496 LR 0.000500 Time 0.022122 +2023-10-05 21:41:57,828 - Epoch: [127][ 340/ 1236] Overall Loss 0.226015 Objective Loss 0.226015 LR 0.000500 Time 0.022056 +2023-10-05 21:41:58,026 - Epoch: [127][ 350/ 1236] Overall Loss 0.225825 Objective Loss 0.225825 LR 0.000500 Time 0.021988 +2023-10-05 21:41:58,225 - Epoch: [127][ 360/ 1236] Overall Loss 0.226079 Objective Loss 0.226079 LR 0.000500 Time 0.021929 +2023-10-05 21:41:58,422 - Epoch: [127][ 370/ 1236] Overall Loss 0.225788 Objective Loss 0.225788 LR 0.000500 Time 0.021869 +2023-10-05 21:41:58,621 - Epoch: [127][ 380/ 1236] Overall Loss 0.226066 Objective Loss 0.226066 LR 0.000500 Time 0.021816 +2023-10-05 21:41:58,819 - Epoch: [127][ 390/ 1236] Overall Loss 0.225100 Objective Loss 0.225100 LR 0.000500 Time 0.021763 +2023-10-05 21:41:59,018 - Epoch: [127][ 400/ 1236] Overall Loss 0.225146 Objective Loss 0.225146 LR 0.000500 Time 0.021715 +2023-10-05 21:41:59,215 - Epoch: [127][ 410/ 1236] Overall Loss 0.225123 Objective Loss 0.225123 LR 0.000500 Time 0.021666 +2023-10-05 21:41:59,414 - Epoch: [127][ 420/ 1236] Overall Loss 0.224683 Objective Loss 0.224683 LR 0.000500 Time 0.021623 +2023-10-05 21:41:59,611 - Epoch: [127][ 430/ 1236] Overall Loss 0.224280 Objective Loss 0.224280 LR 0.000500 Time 0.021579 +2023-10-05 21:41:59,810 - Epoch: [127][ 440/ 1236] Overall Loss 0.224789 Objective Loss 0.224789 LR 0.000500 Time 0.021539 +2023-10-05 21:42:00,008 - Epoch: [127][ 450/ 1236] Overall Loss 0.224482 Objective Loss 0.224482 LR 0.000500 Time 0.021500 +2023-10-05 21:42:00,207 - Epoch: [127][ 460/ 1236] Overall Loss 0.224078 Objective Loss 0.224078 LR 0.000500 Time 0.021464 +2023-10-05 21:42:00,404 - Epoch: [127][ 470/ 1236] Overall Loss 0.223908 Objective Loss 0.223908 LR 0.000500 Time 0.021426 +2023-10-05 21:42:00,603 - Epoch: [127][ 480/ 1236] Overall Loss 0.223687 Objective Loss 0.223687 LR 0.000500 Time 0.021394 +2023-10-05 21:42:00,801 - Epoch: [127][ 490/ 1236] Overall Loss 0.223796 Objective Loss 0.223796 LR 0.000500 Time 0.021359 +2023-10-05 21:42:01,000 - Epoch: [127][ 500/ 1236] Overall Loss 0.223687 Objective Loss 0.223687 LR 0.000500 Time 0.021329 +2023-10-05 21:42:01,197 - Epoch: [127][ 510/ 1236] Overall Loss 0.223998 Objective Loss 0.223998 LR 0.000500 Time 0.021297 +2023-10-05 21:42:01,396 - Epoch: [127][ 520/ 1236] Overall Loss 0.224265 Objective Loss 0.224265 LR 0.000500 Time 0.021270 +2023-10-05 21:42:01,593 - Epoch: [127][ 530/ 1236] Overall Loss 0.224396 Objective Loss 0.224396 LR 0.000500 Time 0.021240 +2023-10-05 21:42:01,792 - Epoch: [127][ 540/ 1236] Overall Loss 0.224206 Objective Loss 0.224206 LR 0.000500 Time 0.021215 +2023-10-05 21:42:01,989 - Epoch: [127][ 550/ 1236] Overall Loss 0.224090 Objective Loss 0.224090 LR 0.000500 Time 0.021186 +2023-10-05 21:42:02,188 - Epoch: [127][ 560/ 1236] Overall Loss 0.224235 Objective Loss 0.224235 LR 0.000500 Time 0.021163 +2023-10-05 21:42:02,385 - Epoch: [127][ 570/ 1236] Overall Loss 0.224318 Objective Loss 0.224318 LR 0.000500 Time 0.021137 +2023-10-05 21:42:02,584 - Epoch: [127][ 580/ 1236] Overall Loss 0.224242 Objective Loss 0.224242 LR 0.000500 Time 0.021115 +2023-10-05 21:42:02,782 - Epoch: [127][ 590/ 1236] Overall Loss 0.223728 Objective Loss 0.223728 LR 0.000500 Time 0.021091 +2023-10-05 21:42:02,980 - Epoch: [127][ 600/ 1236] Overall Loss 0.223203 Objective Loss 0.223203 LR 0.000500 Time 0.021069 +2023-10-05 21:42:03,177 - Epoch: [127][ 610/ 1236] Overall Loss 0.223143 Objective Loss 0.223143 LR 0.000500 Time 0.021047 +2023-10-05 21:42:03,376 - Epoch: [127][ 620/ 1236] Overall Loss 0.222853 Objective Loss 0.222853 LR 0.000500 Time 0.021028 +2023-10-05 21:42:03,574 - Epoch: [127][ 630/ 1236] Overall Loss 0.222721 Objective Loss 0.222721 LR 0.000500 Time 0.021007 +2023-10-05 21:42:03,773 - Epoch: [127][ 640/ 1236] Overall Loss 0.222817 Objective Loss 0.222817 LR 0.000500 Time 0.020989 +2023-10-05 21:42:03,970 - Epoch: [127][ 650/ 1236] Overall Loss 0.222696 Objective Loss 0.222696 LR 0.000500 Time 0.020970 +2023-10-05 21:42:04,169 - Epoch: [127][ 660/ 1236] Overall Loss 0.223277 Objective Loss 0.223277 LR 0.000500 Time 0.020953 +2023-10-05 21:42:04,367 - Epoch: [127][ 670/ 1236] Overall Loss 0.223692 Objective Loss 0.223692 LR 0.000500 Time 0.020934 +2023-10-05 21:42:04,566 - Epoch: [127][ 680/ 1236] Overall Loss 0.223761 Objective Loss 0.223761 LR 0.000500 Time 0.020919 +2023-10-05 21:42:04,763 - Epoch: [127][ 690/ 1236] Overall Loss 0.224007 Objective Loss 0.224007 LR 0.000500 Time 0.020901 +2023-10-05 21:42:04,962 - Epoch: [127][ 700/ 1236] Overall Loss 0.224187 Objective Loss 0.224187 LR 0.000500 Time 0.020886 +2023-10-05 21:42:05,160 - Epoch: [127][ 710/ 1236] Overall Loss 0.224011 Objective Loss 0.224011 LR 0.000500 Time 0.020870 +2023-10-05 21:42:05,359 - Epoch: [127][ 720/ 1236] Overall Loss 0.224105 Objective Loss 0.224105 LR 0.000500 Time 0.020856 +2023-10-05 21:42:05,556 - Epoch: [127][ 730/ 1236] Overall Loss 0.224405 Objective Loss 0.224405 LR 0.000500 Time 0.020840 +2023-10-05 21:42:05,755 - Epoch: [127][ 740/ 1236] Overall Loss 0.224415 Objective Loss 0.224415 LR 0.000500 Time 0.020827 +2023-10-05 21:42:05,952 - Epoch: [127][ 750/ 1236] Overall Loss 0.224374 Objective Loss 0.224374 LR 0.000500 Time 0.020812 +2023-10-05 21:42:06,152 - Epoch: [127][ 760/ 1236] Overall Loss 0.224860 Objective Loss 0.224860 LR 0.000500 Time 0.020800 +2023-10-05 21:42:06,349 - Epoch: [127][ 770/ 1236] Overall Loss 0.225054 Objective Loss 0.225054 LR 0.000500 Time 0.020785 +2023-10-05 21:42:06,548 - Epoch: [127][ 780/ 1236] Overall Loss 0.224967 Objective Loss 0.224967 LR 0.000500 Time 0.020774 +2023-10-05 21:42:06,745 - Epoch: [127][ 790/ 1236] Overall Loss 0.224989 Objective Loss 0.224989 LR 0.000500 Time 0.020760 +2023-10-05 21:42:06,945 - Epoch: [127][ 800/ 1236] Overall Loss 0.225108 Objective Loss 0.225108 LR 0.000500 Time 0.020749 +2023-10-05 21:42:07,142 - Epoch: [127][ 810/ 1236] Overall Loss 0.225286 Objective Loss 0.225286 LR 0.000500 Time 0.020737 +2023-10-05 21:42:07,345 - Epoch: [127][ 820/ 1236] Overall Loss 0.225177 Objective Loss 0.225177 LR 0.000500 Time 0.020724 +2023-10-05 21:42:07,543 - Epoch: [127][ 830/ 1236] Overall Loss 0.224779 Objective Loss 0.224779 LR 0.000500 Time 0.020712 +2023-10-05 21:42:07,742 - Epoch: [127][ 840/ 1236] Overall Loss 0.224746 Objective Loss 0.224746 LR 0.000500 Time 0.020702 +2023-10-05 21:42:07,939 - Epoch: [127][ 850/ 1236] Overall Loss 0.225335 Objective Loss 0.225335 LR 0.000500 Time 0.020690 +2023-10-05 21:42:08,138 - Epoch: [127][ 860/ 1236] Overall Loss 0.225362 Objective Loss 0.225362 LR 0.000500 Time 0.020681 +2023-10-05 21:42:08,336 - Epoch: [127][ 870/ 1236] Overall Loss 0.225256 Objective Loss 0.225256 LR 0.000500 Time 0.020669 +2023-10-05 21:42:08,535 - Epoch: [127][ 880/ 1236] Overall Loss 0.225310 Objective Loss 0.225310 LR 0.000500 Time 0.020660 +2023-10-05 21:42:08,732 - Epoch: [127][ 890/ 1236] Overall Loss 0.225271 Objective Loss 0.225271 LR 0.000500 Time 0.020649 +2023-10-05 21:42:08,931 - Epoch: [127][ 900/ 1236] Overall Loss 0.225500 Objective Loss 0.225500 LR 0.000500 Time 0.020641 +2023-10-05 21:42:09,128 - Epoch: [127][ 910/ 1236] Overall Loss 0.225526 Objective Loss 0.225526 LR 0.000500 Time 0.020631 +2023-10-05 21:42:09,327 - Epoch: [127][ 920/ 1236] Overall Loss 0.225390 Objective Loss 0.225390 LR 0.000500 Time 0.020622 +2023-10-05 21:42:09,525 - Epoch: [127][ 930/ 1236] Overall Loss 0.225222 Objective Loss 0.225222 LR 0.000500 Time 0.020613 +2023-10-05 21:42:09,724 - Epoch: [127][ 940/ 1236] Overall Loss 0.225304 Objective Loss 0.225304 LR 0.000500 Time 0.020605 +2023-10-05 21:42:09,922 - Epoch: [127][ 950/ 1236] Overall Loss 0.225258 Objective Loss 0.225258 LR 0.000500 Time 0.020595 +2023-10-05 21:42:10,121 - Epoch: [127][ 960/ 1236] Overall Loss 0.225588 Objective Loss 0.225588 LR 0.000500 Time 0.020588 +2023-10-05 21:42:10,318 - Epoch: [127][ 970/ 1236] Overall Loss 0.225870 Objective Loss 0.225870 LR 0.000500 Time 0.020579 +2023-10-05 21:42:10,521 - Epoch: [127][ 980/ 1236] Overall Loss 0.226271 Objective Loss 0.226271 LR 0.000500 Time 0.020575 +2023-10-05 21:42:10,720 - Epoch: [127][ 990/ 1236] Overall Loss 0.226445 Objective Loss 0.226445 LR 0.000500 Time 0.020568 +2023-10-05 21:42:10,919 - Epoch: [127][ 1000/ 1236] Overall Loss 0.226668 Objective Loss 0.226668 LR 0.000500 Time 0.020562 +2023-10-05 21:42:11,118 - Epoch: [127][ 1010/ 1236] Overall Loss 0.226939 Objective Loss 0.226939 LR 0.000500 Time 0.020555 +2023-10-05 21:42:11,318 - Epoch: [127][ 1020/ 1236] Overall Loss 0.227156 Objective Loss 0.227156 LR 0.000500 Time 0.020549 +2023-10-05 21:42:11,517 - Epoch: [127][ 1030/ 1236] Overall Loss 0.227267 Objective Loss 0.227267 LR 0.000500 Time 0.020542 +2023-10-05 21:42:11,717 - Epoch: [127][ 1040/ 1236] Overall Loss 0.227387 Objective Loss 0.227387 LR 0.000500 Time 0.020537 +2023-10-05 21:42:11,916 - Epoch: [127][ 1050/ 1236] Overall Loss 0.227438 Objective Loss 0.227438 LR 0.000500 Time 0.020530 +2023-10-05 21:42:12,116 - Epoch: [127][ 1060/ 1236] Overall Loss 0.227772 Objective Loss 0.227772 LR 0.000500 Time 0.020525 +2023-10-05 21:42:12,315 - Epoch: [127][ 1070/ 1236] Overall Loss 0.227815 Objective Loss 0.227815 LR 0.000500 Time 0.020519 +2023-10-05 21:42:12,515 - Epoch: [127][ 1080/ 1236] Overall Loss 0.228015 Objective Loss 0.228015 LR 0.000500 Time 0.020513 +2023-10-05 21:42:12,713 - Epoch: [127][ 1090/ 1236] Overall Loss 0.228229 Objective Loss 0.228229 LR 0.000500 Time 0.020507 +2023-10-05 21:42:12,913 - Epoch: [127][ 1100/ 1236] Overall Loss 0.228341 Objective Loss 0.228341 LR 0.000500 Time 0.020502 +2023-10-05 21:42:13,112 - Epoch: [127][ 1110/ 1236] Overall Loss 0.228338 Objective Loss 0.228338 LR 0.000500 Time 0.020496 +2023-10-05 21:42:13,312 - Epoch: [127][ 1120/ 1236] Overall Loss 0.228519 Objective Loss 0.228519 LR 0.000500 Time 0.020491 +2023-10-05 21:42:13,511 - Epoch: [127][ 1130/ 1236] Overall Loss 0.228284 Objective Loss 0.228284 LR 0.000500 Time 0.020486 +2023-10-05 21:42:13,711 - Epoch: [127][ 1140/ 1236] Overall Loss 0.228105 Objective Loss 0.228105 LR 0.000500 Time 0.020481 +2023-10-05 21:42:13,909 - Epoch: [127][ 1150/ 1236] Overall Loss 0.228130 Objective Loss 0.228130 LR 0.000500 Time 0.020475 +2023-10-05 21:42:14,109 - Epoch: [127][ 1160/ 1236] Overall Loss 0.228383 Objective Loss 0.228383 LR 0.000500 Time 0.020471 +2023-10-05 21:42:14,308 - Epoch: [127][ 1170/ 1236] Overall Loss 0.228526 Objective Loss 0.228526 LR 0.000500 Time 0.020465 +2023-10-05 21:42:14,508 - Epoch: [127][ 1180/ 1236] Overall Loss 0.228519 Objective Loss 0.228519 LR 0.000500 Time 0.020461 +2023-10-05 21:42:14,707 - Epoch: [127][ 1190/ 1236] Overall Loss 0.228463 Objective Loss 0.228463 LR 0.000500 Time 0.020456 +2023-10-05 21:42:14,907 - Epoch: [127][ 1200/ 1236] Overall Loss 0.228418 Objective Loss 0.228418 LR 0.000500 Time 0.020452 +2023-10-05 21:42:15,106 - Epoch: [127][ 1210/ 1236] Overall Loss 0.228482 Objective Loss 0.228482 LR 0.000500 Time 0.020447 +2023-10-05 21:42:15,306 - Epoch: [127][ 1220/ 1236] Overall Loss 0.228522 Objective Loss 0.228522 LR 0.000500 Time 0.020443 +2023-10-05 21:42:15,556 - Epoch: [127][ 1230/ 1236] Overall Loss 0.228823 Objective Loss 0.228823 LR 0.000500 Time 0.020480 +2023-10-05 21:42:15,673 - Epoch: [127][ 1236/ 1236] Overall Loss 0.228730 Objective Loss 0.228730 Top1 86.965377 Top5 97.963340 LR 0.000500 Time 0.020476 +2023-10-05 21:42:15,806 - --- validate (epoch=127)----------- +2023-10-05 21:42:15,807 - 29943 samples (256 per mini-batch) +2023-10-05 21:42:16,257 - Epoch: [127][ 10/ 117] Loss 0.292172 Top1 83.828125 Top5 98.164062 +2023-10-05 21:42:16,406 - Epoch: [127][ 20/ 117] Loss 0.332840 Top1 83.750000 Top5 98.007812 +2023-10-05 21:42:16,554 - Epoch: [127][ 30/ 117] Loss 0.338402 Top1 83.268229 Top5 97.903646 +2023-10-05 21:42:16,702 - Epoch: [127][ 40/ 117] Loss 0.327991 Top1 83.593750 Top5 97.958984 +2023-10-05 21:42:16,850 - Epoch: [127][ 50/ 117] Loss 0.325598 Top1 83.695312 Top5 98.007812 +2023-10-05 21:42:16,997 - Epoch: [127][ 60/ 117] Loss 0.329248 Top1 83.704427 Top5 97.949219 +2023-10-05 21:42:17,144 - Epoch: [127][ 70/ 117] Loss 0.327992 Top1 83.755580 Top5 97.907366 +2023-10-05 21:42:17,296 - Epoch: [127][ 80/ 117] Loss 0.329413 Top1 83.632812 Top5 97.939453 +2023-10-05 21:42:17,443 - Epoch: [127][ 90/ 117] Loss 0.331332 Top1 83.541667 Top5 97.938368 +2023-10-05 21:42:17,590 - Epoch: [127][ 100/ 117] Loss 0.334287 Top1 83.453125 Top5 97.929688 +2023-10-05 21:42:17,744 - Epoch: [127][ 110/ 117] Loss 0.332673 Top1 83.444602 Top5 97.894176 +2023-10-05 21:42:17,829 - Epoch: [127][ 117/ 117] Loss 0.332385 Top1 83.491968 Top5 97.892663 +2023-10-05 21:42:17,947 - ==> Top1: 83.492 Top5: 97.893 Loss: 0.332 + +2023-10-05 21:42:17,948 - ==> Confusion: +[[ 941 2 1 1 7 3 0 0 4 63 1 2 3 2 6 1 1 3 1 0 8] + [ 4 1039 1 0 7 33 1 17 3 0 1 0 0 0 2 3 2 2 6 5 5] + [ 8 1 955 16 1 1 21 11 0 0 3 2 9 4 0 1 1 3 9 3 7] + [ 5 1 17 941 1 5 2 0 5 0 9 2 12 5 30 6 3 6 21 0 18] + [ 34 6 1 0 968 6 0 0 0 10 0 1 0 0 8 5 7 1 0 0 3] + [ 5 26 1 0 5 1001 1 16 2 1 3 15 1 12 7 0 1 0 2 3 14] + [ 0 8 25 2 1 2 1115 6 0 0 1 2 2 0 1 7 1 2 3 8 5] + [ 7 19 12 0 0 41 2 1043 0 3 2 12 3 3 0 2 0 1 48 6 14] + [ 25 3 0 0 0 2 0 0 960 48 9 2 5 11 16 2 1 1 2 0 2] + [ 114 0 0 0 7 3 0 0 24 936 0 2 1 14 6 4 2 0 0 1 5] + [ 3 6 12 6 2 1 6 5 11 3 943 3 1 25 3 1 3 1 7 2 9] + [ 0 0 0 0 0 9 0 1 0 1 0 967 19 6 0 4 1 17 0 8 2] + [ 0 1 1 4 0 2 1 1 1 2 0 45 966 6 3 4 1 23 0 2 5] + [ 5 0 1 0 0 10 0 1 11 14 5 4 3 1049 2 3 1 1 0 0 9] + [ 15 3 4 6 4 0 0 0 20 7 3 2 2 1 1022 0 1 0 9 0 2] + [ 1 3 1 0 2 1 1 0 0 0 0 8 4 2 0 1074 15 13 0 4 5] + [ 2 13 1 0 7 7 0 2 1 0 0 4 1 2 4 10 1094 0 1 4 8] + [ 0 0 0 4 0 0 0 0 0 1 0 5 19 0 1 5 0 1000 0 2 1] + [ 2 7 7 12 1 0 0 32 2 2 1 0 3 1 17 0 1 1 969 1 9] + [ 0 2 1 0 1 8 5 8 0 0 2 25 6 2 0 7 11 1 3 1058 12] + [ 185 149 149 58 125 177 37 97 108 94 136 152 344 326 152 62 175 88 160 172 4959]] + +2023-10-05 21:42:17,949 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:42:17,949 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:42:17,955 - + +2023-10-05 21:42:17,955 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:42:19,060 - Epoch: [128][ 10/ 1236] Overall Loss 0.209377 Objective Loss 0.209377 LR 0.000500 Time 0.110378 +2023-10-05 21:42:19,259 - Epoch: [128][ 20/ 1236] Overall Loss 0.215400 Objective Loss 0.215400 LR 0.000500 Time 0.065135 +2023-10-05 21:42:19,457 - Epoch: [128][ 30/ 1236] Overall Loss 0.212829 Objective Loss 0.212829 LR 0.000500 Time 0.050030 +2023-10-05 21:42:19,657 - Epoch: [128][ 40/ 1236] Overall Loss 0.223890 Objective Loss 0.223890 LR 0.000500 Time 0.042499 +2023-10-05 21:42:19,855 - Epoch: [128][ 50/ 1236] Overall Loss 0.224089 Objective Loss 0.224089 LR 0.000500 Time 0.037966 +2023-10-05 21:42:20,055 - Epoch: [128][ 60/ 1236] Overall Loss 0.225285 Objective Loss 0.225285 LR 0.000500 Time 0.034965 +2023-10-05 21:42:20,254 - Epoch: [128][ 70/ 1236] Overall Loss 0.223354 Objective Loss 0.223354 LR 0.000500 Time 0.032802 +2023-10-05 21:42:20,454 - Epoch: [128][ 80/ 1236] Overall Loss 0.223792 Objective Loss 0.223792 LR 0.000500 Time 0.031192 +2023-10-05 21:42:20,652 - Epoch: [128][ 90/ 1236] Overall Loss 0.227406 Objective Loss 0.227406 LR 0.000500 Time 0.029931 +2023-10-05 21:42:20,852 - Epoch: [128][ 100/ 1236] Overall Loss 0.227119 Objective Loss 0.227119 LR 0.000500 Time 0.028931 +2023-10-05 21:42:21,051 - Epoch: [128][ 110/ 1236] Overall Loss 0.225074 Objective Loss 0.225074 LR 0.000500 Time 0.028102 +2023-10-05 21:42:21,250 - Epoch: [128][ 120/ 1236] Overall Loss 0.223765 Objective Loss 0.223765 LR 0.000500 Time 0.027421 +2023-10-05 21:42:21,448 - Epoch: [128][ 130/ 1236] Overall Loss 0.223035 Objective Loss 0.223035 LR 0.000500 Time 0.026830 +2023-10-05 21:42:21,647 - Epoch: [128][ 140/ 1236] Overall Loss 0.222382 Objective Loss 0.222382 LR 0.000500 Time 0.026335 +2023-10-05 21:42:21,846 - Epoch: [128][ 150/ 1236] Overall Loss 0.220943 Objective Loss 0.220943 LR 0.000500 Time 0.025898 +2023-10-05 21:42:22,044 - Epoch: [128][ 160/ 1236] Overall Loss 0.221080 Objective Loss 0.221080 LR 0.000500 Time 0.025518 +2023-10-05 21:42:22,243 - Epoch: [128][ 170/ 1236] Overall Loss 0.222299 Objective Loss 0.222299 LR 0.000500 Time 0.025183 +2023-10-05 21:42:22,441 - Epoch: [128][ 180/ 1236] Overall Loss 0.223363 Objective Loss 0.223363 LR 0.000500 Time 0.024885 +2023-10-05 21:42:22,639 - Epoch: [128][ 190/ 1236] Overall Loss 0.222855 Objective Loss 0.222855 LR 0.000500 Time 0.024613 +2023-10-05 21:42:22,838 - Epoch: [128][ 200/ 1236] Overall Loss 0.223401 Objective Loss 0.223401 LR 0.000500 Time 0.024379 +2023-10-05 21:42:23,037 - Epoch: [128][ 210/ 1236] Overall Loss 0.223158 Objective Loss 0.223158 LR 0.000500 Time 0.024162 +2023-10-05 21:42:23,237 - Epoch: [128][ 220/ 1236] Overall Loss 0.222859 Objective Loss 0.222859 LR 0.000500 Time 0.023970 +2023-10-05 21:42:23,435 - Epoch: [128][ 230/ 1236] Overall Loss 0.222954 Objective Loss 0.222954 LR 0.000500 Time 0.023789 +2023-10-05 21:42:23,633 - Epoch: [128][ 240/ 1236] Overall Loss 0.223741 Objective Loss 0.223741 LR 0.000500 Time 0.023622 +2023-10-05 21:42:23,832 - Epoch: [128][ 250/ 1236] Overall Loss 0.224608 Objective Loss 0.224608 LR 0.000500 Time 0.023470 +2023-10-05 21:42:24,031 - Epoch: [128][ 260/ 1236] Overall Loss 0.223723 Objective Loss 0.223723 LR 0.000500 Time 0.023333 +2023-10-05 21:42:24,230 - Epoch: [128][ 270/ 1236] Overall Loss 0.223643 Objective Loss 0.223643 LR 0.000500 Time 0.023204 +2023-10-05 21:42:24,430 - Epoch: [128][ 280/ 1236] Overall Loss 0.223862 Objective Loss 0.223862 LR 0.000500 Time 0.023087 +2023-10-05 21:42:24,628 - Epoch: [128][ 290/ 1236] Overall Loss 0.223751 Objective Loss 0.223751 LR 0.000500 Time 0.022974 +2023-10-05 21:42:24,828 - Epoch: [128][ 300/ 1236] Overall Loss 0.224213 Objective Loss 0.224213 LR 0.000500 Time 0.022873 +2023-10-05 21:42:25,027 - Epoch: [128][ 310/ 1236] Overall Loss 0.224214 Objective Loss 0.224214 LR 0.000500 Time 0.022775 +2023-10-05 21:42:25,227 - Epoch: [128][ 320/ 1236] Overall Loss 0.223825 Objective Loss 0.223825 LR 0.000500 Time 0.022687 +2023-10-05 21:42:25,425 - Epoch: [128][ 330/ 1236] Overall Loss 0.223827 Objective Loss 0.223827 LR 0.000500 Time 0.022600 +2023-10-05 21:42:25,625 - Epoch: [128][ 340/ 1236] Overall Loss 0.223213 Objective Loss 0.223213 LR 0.000500 Time 0.022521 +2023-10-05 21:42:25,824 - Epoch: [128][ 350/ 1236] Overall Loss 0.223894 Objective Loss 0.223894 LR 0.000500 Time 0.022445 +2023-10-05 21:42:26,023 - Epoch: [128][ 360/ 1236] Overall Loss 0.224451 Objective Loss 0.224451 LR 0.000500 Time 0.022375 +2023-10-05 21:42:26,222 - Epoch: [128][ 370/ 1236] Overall Loss 0.224146 Objective Loss 0.224146 LR 0.000500 Time 0.022306 +2023-10-05 21:42:26,422 - Epoch: [128][ 380/ 1236] Overall Loss 0.223710 Objective Loss 0.223710 LR 0.000500 Time 0.022245 +2023-10-05 21:42:26,621 - Epoch: [128][ 390/ 1236] Overall Loss 0.223722 Objective Loss 0.223722 LR 0.000500 Time 0.022183 +2023-10-05 21:42:26,821 - Epoch: [128][ 400/ 1236] Overall Loss 0.223876 Objective Loss 0.223876 LR 0.000500 Time 0.022127 +2023-10-05 21:42:27,019 - Epoch: [128][ 410/ 1236] Overall Loss 0.223584 Objective Loss 0.223584 LR 0.000500 Time 0.022072 +2023-10-05 21:42:27,220 - Epoch: [128][ 420/ 1236] Overall Loss 0.223858 Objective Loss 0.223858 LR 0.000500 Time 0.022022 +2023-10-05 21:42:27,419 - Epoch: [128][ 430/ 1236] Overall Loss 0.223975 Objective Loss 0.223975 LR 0.000500 Time 0.021972 +2023-10-05 21:42:27,618 - Epoch: [128][ 440/ 1236] Overall Loss 0.224422 Objective Loss 0.224422 LR 0.000500 Time 0.021924 +2023-10-05 21:42:27,815 - Epoch: [128][ 450/ 1236] Overall Loss 0.224246 Objective Loss 0.224246 LR 0.000500 Time 0.021876 +2023-10-05 21:42:28,015 - Epoch: [128][ 460/ 1236] Overall Loss 0.224567 Objective Loss 0.224567 LR 0.000500 Time 0.021833 +2023-10-05 21:42:28,213 - Epoch: [128][ 470/ 1236] Overall Loss 0.224802 Objective Loss 0.224802 LR 0.000500 Time 0.021789 +2023-10-05 21:42:28,412 - Epoch: [128][ 480/ 1236] Overall Loss 0.224390 Objective Loss 0.224390 LR 0.000500 Time 0.021749 +2023-10-05 21:42:28,610 - Epoch: [128][ 490/ 1236] Overall Loss 0.224226 Objective Loss 0.224226 LR 0.000500 Time 0.021708 +2023-10-05 21:42:28,810 - Epoch: [128][ 500/ 1236] Overall Loss 0.224243 Objective Loss 0.224243 LR 0.000500 Time 0.021674 +2023-10-05 21:42:29,010 - Epoch: [128][ 510/ 1236] Overall Loss 0.224205 Objective Loss 0.224205 LR 0.000500 Time 0.021640 +2023-10-05 21:42:29,210 - Epoch: [128][ 520/ 1236] Overall Loss 0.224624 Objective Loss 0.224624 LR 0.000500 Time 0.021608 +2023-10-05 21:42:29,410 - Epoch: [128][ 530/ 1236] Overall Loss 0.224731 Objective Loss 0.224731 LR 0.000500 Time 0.021576 +2023-10-05 21:42:29,610 - Epoch: [128][ 540/ 1236] Overall Loss 0.224626 Objective Loss 0.224626 LR 0.000500 Time 0.021547 +2023-10-05 21:42:29,809 - Epoch: [128][ 550/ 1236] Overall Loss 0.224216 Objective Loss 0.224216 LR 0.000500 Time 0.021517 +2023-10-05 21:42:30,009 - Epoch: [128][ 560/ 1236] Overall Loss 0.224491 Objective Loss 0.224491 LR 0.000500 Time 0.021490 +2023-10-05 21:42:30,209 - Epoch: [128][ 570/ 1236] Overall Loss 0.224677 Objective Loss 0.224677 LR 0.000500 Time 0.021461 +2023-10-05 21:42:30,409 - Epoch: [128][ 580/ 1236] Overall Loss 0.224426 Objective Loss 0.224426 LR 0.000500 Time 0.021436 +2023-10-05 21:42:30,609 - Epoch: [128][ 590/ 1236] Overall Loss 0.224461 Objective Loss 0.224461 LR 0.000500 Time 0.021411 +2023-10-05 21:42:30,807 - Epoch: [128][ 600/ 1236] Overall Loss 0.224648 Objective Loss 0.224648 LR 0.000500 Time 0.021385 +2023-10-05 21:42:31,005 - Epoch: [128][ 610/ 1236] Overall Loss 0.224641 Objective Loss 0.224641 LR 0.000500 Time 0.021358 +2023-10-05 21:42:31,204 - Epoch: [128][ 620/ 1236] Overall Loss 0.224734 Objective Loss 0.224734 LR 0.000500 Time 0.021334 +2023-10-05 21:42:31,400 - Epoch: [128][ 630/ 1236] Overall Loss 0.224643 Objective Loss 0.224643 LR 0.000500 Time 0.021306 +2023-10-05 21:42:31,599 - Epoch: [128][ 640/ 1236] Overall Loss 0.224717 Objective Loss 0.224717 LR 0.000500 Time 0.021284 +2023-10-05 21:42:31,797 - Epoch: [128][ 650/ 1236] Overall Loss 0.224626 Objective Loss 0.224626 LR 0.000500 Time 0.021259 +2023-10-05 21:42:31,996 - Epoch: [128][ 660/ 1236] Overall Loss 0.224811 Objective Loss 0.224811 LR 0.000500 Time 0.021239 +2023-10-05 21:42:32,194 - Epoch: [128][ 670/ 1236] Overall Loss 0.224583 Objective Loss 0.224583 LR 0.000500 Time 0.021217 +2023-10-05 21:42:32,393 - Epoch: [128][ 680/ 1236] Overall Loss 0.224497 Objective Loss 0.224497 LR 0.000500 Time 0.021197 +2023-10-05 21:42:32,591 - Epoch: [128][ 690/ 1236] Overall Loss 0.224540 Objective Loss 0.224540 LR 0.000500 Time 0.021176 +2023-10-05 21:42:32,790 - Epoch: [128][ 700/ 1236] Overall Loss 0.224193 Objective Loss 0.224193 LR 0.000500 Time 0.021157 +2023-10-05 21:42:32,987 - Epoch: [128][ 710/ 1236] Overall Loss 0.224190 Objective Loss 0.224190 LR 0.000500 Time 0.021136 +2023-10-05 21:42:33,186 - Epoch: [128][ 720/ 1236] Overall Loss 0.223771 Objective Loss 0.223771 LR 0.000500 Time 0.021119 +2023-10-05 21:42:33,384 - Epoch: [128][ 730/ 1236] Overall Loss 0.224156 Objective Loss 0.224156 LR 0.000500 Time 0.021100 +2023-10-05 21:42:33,583 - Epoch: [128][ 740/ 1236] Overall Loss 0.223970 Objective Loss 0.223970 LR 0.000500 Time 0.021083 +2023-10-05 21:42:33,779 - Epoch: [128][ 750/ 1236] Overall Loss 0.224044 Objective Loss 0.224044 LR 0.000500 Time 0.021064 +2023-10-05 21:42:33,978 - Epoch: [128][ 760/ 1236] Overall Loss 0.224016 Objective Loss 0.224016 LR 0.000500 Time 0.021048 +2023-10-05 21:42:34,176 - Epoch: [128][ 770/ 1236] Overall Loss 0.224018 Objective Loss 0.224018 LR 0.000500 Time 0.021031 +2023-10-05 21:42:34,375 - Epoch: [128][ 780/ 1236] Overall Loss 0.224030 Objective Loss 0.224030 LR 0.000500 Time 0.021016 +2023-10-05 21:42:34,573 - Epoch: [128][ 790/ 1236] Overall Loss 0.223736 Objective Loss 0.223736 LR 0.000500 Time 0.021000 +2023-10-05 21:42:34,772 - Epoch: [128][ 800/ 1236] Overall Loss 0.223927 Objective Loss 0.223927 LR 0.000500 Time 0.020986 +2023-10-05 21:42:34,969 - Epoch: [128][ 810/ 1236] Overall Loss 0.223645 Objective Loss 0.223645 LR 0.000500 Time 0.020970 +2023-10-05 21:42:35,168 - Epoch: [128][ 820/ 1236] Overall Loss 0.224116 Objective Loss 0.224116 LR 0.000500 Time 0.020957 +2023-10-05 21:42:35,366 - Epoch: [128][ 830/ 1236] Overall Loss 0.224319 Objective Loss 0.224319 LR 0.000500 Time 0.020942 +2023-10-05 21:42:35,565 - Epoch: [128][ 840/ 1236] Overall Loss 0.224182 Objective Loss 0.224182 LR 0.000500 Time 0.020929 +2023-10-05 21:42:35,763 - Epoch: [128][ 850/ 1236] Overall Loss 0.224305 Objective Loss 0.224305 LR 0.000500 Time 0.020915 +2023-10-05 21:42:35,962 - Epoch: [128][ 860/ 1236] Overall Loss 0.224346 Objective Loss 0.224346 LR 0.000500 Time 0.020903 +2023-10-05 21:42:36,160 - Epoch: [128][ 870/ 1236] Overall Loss 0.224406 Objective Loss 0.224406 LR 0.000500 Time 0.020890 +2023-10-05 21:42:36,359 - Epoch: [128][ 880/ 1236] Overall Loss 0.224322 Objective Loss 0.224322 LR 0.000500 Time 0.020879 +2023-10-05 21:42:36,556 - Epoch: [128][ 890/ 1236] Overall Loss 0.224789 Objective Loss 0.224789 LR 0.000500 Time 0.020865 +2023-10-05 21:42:36,755 - Epoch: [128][ 900/ 1236] Overall Loss 0.224982 Objective Loss 0.224982 LR 0.000500 Time 0.020854 +2023-10-05 21:42:36,953 - Epoch: [128][ 910/ 1236] Overall Loss 0.225135 Objective Loss 0.225135 LR 0.000500 Time 0.020842 +2023-10-05 21:42:37,151 - Epoch: [128][ 920/ 1236] Overall Loss 0.225383 Objective Loss 0.225383 LR 0.000500 Time 0.020830 +2023-10-05 21:42:37,348 - Epoch: [128][ 930/ 1236] Overall Loss 0.225468 Objective Loss 0.225468 LR 0.000500 Time 0.020818 +2023-10-05 21:42:37,546 - Epoch: [128][ 940/ 1236] Overall Loss 0.225540 Objective Loss 0.225540 LR 0.000500 Time 0.020807 +2023-10-05 21:42:37,743 - Epoch: [128][ 950/ 1236] Overall Loss 0.225384 Objective Loss 0.225384 LR 0.000500 Time 0.020794 +2023-10-05 21:42:37,942 - Epoch: [128][ 960/ 1236] Overall Loss 0.225410 Objective Loss 0.225410 LR 0.000500 Time 0.020785 +2023-10-05 21:42:38,139 - Epoch: [128][ 970/ 1236] Overall Loss 0.225534 Objective Loss 0.225534 LR 0.000500 Time 0.020774 +2023-10-05 21:42:38,338 - Epoch: [128][ 980/ 1236] Overall Loss 0.225665 Objective Loss 0.225665 LR 0.000500 Time 0.020765 +2023-10-05 21:42:38,536 - Epoch: [128][ 990/ 1236] Overall Loss 0.226064 Objective Loss 0.226064 LR 0.000500 Time 0.020754 +2023-10-05 21:42:38,735 - Epoch: [128][ 1000/ 1236] Overall Loss 0.226736 Objective Loss 0.226736 LR 0.000500 Time 0.020745 +2023-10-05 21:42:38,931 - Epoch: [128][ 1010/ 1236] Overall Loss 0.226427 Objective Loss 0.226427 LR 0.000500 Time 0.020734 +2023-10-05 21:42:39,131 - Epoch: [128][ 1020/ 1236] Overall Loss 0.226297 Objective Loss 0.226297 LR 0.000500 Time 0.020726 +2023-10-05 21:42:39,328 - Epoch: [128][ 1030/ 1236] Overall Loss 0.226495 Objective Loss 0.226495 LR 0.000500 Time 0.020716 +2023-10-05 21:42:39,528 - Epoch: [128][ 1040/ 1236] Overall Loss 0.226451 Objective Loss 0.226451 LR 0.000500 Time 0.020708 +2023-10-05 21:42:39,726 - Epoch: [128][ 1050/ 1236] Overall Loss 0.226433 Objective Loss 0.226433 LR 0.000500 Time 0.020699 +2023-10-05 21:42:39,925 - Epoch: [128][ 1060/ 1236] Overall Loss 0.226455 Objective Loss 0.226455 LR 0.000500 Time 0.020691 +2023-10-05 21:42:40,122 - Epoch: [128][ 1070/ 1236] Overall Loss 0.226431 Objective Loss 0.226431 LR 0.000500 Time 0.020682 +2023-10-05 21:42:40,321 - Epoch: [128][ 1080/ 1236] Overall Loss 0.226428 Objective Loss 0.226428 LR 0.000500 Time 0.020675 +2023-10-05 21:42:40,519 - Epoch: [128][ 1090/ 1236] Overall Loss 0.226748 Objective Loss 0.226748 LR 0.000500 Time 0.020666 +2023-10-05 21:42:40,718 - Epoch: [128][ 1100/ 1236] Overall Loss 0.226679 Objective Loss 0.226679 LR 0.000500 Time 0.020659 +2023-10-05 21:42:40,916 - Epoch: [128][ 1110/ 1236] Overall Loss 0.226837 Objective Loss 0.226837 LR 0.000500 Time 0.020651 +2023-10-05 21:42:41,115 - Epoch: [128][ 1120/ 1236] Overall Loss 0.226997 Objective Loss 0.226997 LR 0.000500 Time 0.020644 +2023-10-05 21:42:41,313 - Epoch: [128][ 1130/ 1236] Overall Loss 0.227137 Objective Loss 0.227137 LR 0.000500 Time 0.020636 +2023-10-05 21:42:41,512 - Epoch: [128][ 1140/ 1236] Overall Loss 0.226911 Objective Loss 0.226911 LR 0.000500 Time 0.020630 +2023-10-05 21:42:41,710 - Epoch: [128][ 1150/ 1236] Overall Loss 0.227112 Objective Loss 0.227112 LR 0.000500 Time 0.020622 +2023-10-05 21:42:41,909 - Epoch: [128][ 1160/ 1236] Overall Loss 0.227375 Objective Loss 0.227375 LR 0.000500 Time 0.020616 +2023-10-05 21:42:42,107 - Epoch: [128][ 1170/ 1236] Overall Loss 0.227633 Objective Loss 0.227633 LR 0.000500 Time 0.020608 +2023-10-05 21:42:42,306 - Epoch: [128][ 1180/ 1236] Overall Loss 0.227722 Objective Loss 0.227722 LR 0.000500 Time 0.020602 +2023-10-05 21:42:42,504 - Epoch: [128][ 1190/ 1236] Overall Loss 0.227842 Objective Loss 0.227842 LR 0.000500 Time 0.020595 +2023-10-05 21:42:42,703 - Epoch: [128][ 1200/ 1236] Overall Loss 0.227912 Objective Loss 0.227912 LR 0.000500 Time 0.020589 +2023-10-05 21:42:42,901 - Epoch: [128][ 1210/ 1236] Overall Loss 0.228012 Objective Loss 0.228012 LR 0.000500 Time 0.020582 +2023-10-05 21:42:43,100 - Epoch: [128][ 1220/ 1236] Overall Loss 0.228070 Objective Loss 0.228070 LR 0.000500 Time 0.020576 +2023-10-05 21:42:43,348 - Epoch: [128][ 1230/ 1236] Overall Loss 0.228099 Objective Loss 0.228099 LR 0.000500 Time 0.020611 +2023-10-05 21:42:43,465 - Epoch: [128][ 1236/ 1236] Overall Loss 0.228158 Objective Loss 0.228158 Top1 86.558045 Top5 98.370672 LR 0.000500 Time 0.020604 +2023-10-05 21:42:43,590 - --- validate (epoch=128)----------- +2023-10-05 21:42:43,591 - 29943 samples (256 per mini-batch) +2023-10-05 21:42:44,057 - Epoch: [128][ 10/ 117] Loss 0.344925 Top1 83.632812 Top5 97.656250 +2023-10-05 21:42:44,207 - Epoch: [128][ 20/ 117] Loss 0.339040 Top1 83.476562 Top5 97.773438 +2023-10-05 21:42:44,353 - Epoch: [128][ 30/ 117] Loss 0.349243 Top1 82.981771 Top5 97.825521 +2023-10-05 21:42:44,501 - Epoch: [128][ 40/ 117] Loss 0.332132 Top1 83.427734 Top5 97.841797 +2023-10-05 21:42:44,649 - Epoch: [128][ 50/ 117] Loss 0.327260 Top1 83.632812 Top5 97.812500 +2023-10-05 21:42:44,800 - Epoch: [128][ 60/ 117] Loss 0.328196 Top1 83.509115 Top5 97.792969 +2023-10-05 21:42:44,949 - Epoch: [128][ 70/ 117] Loss 0.329881 Top1 83.448661 Top5 97.823661 +2023-10-05 21:42:45,099 - Epoch: [128][ 80/ 117] Loss 0.328880 Top1 83.544922 Top5 97.817383 +2023-10-05 21:42:45,247 - Epoch: [128][ 90/ 117] Loss 0.326511 Top1 83.619792 Top5 97.834201 +2023-10-05 21:42:45,396 - Epoch: [128][ 100/ 117] Loss 0.323552 Top1 83.792969 Top5 97.894531 +2023-10-05 21:42:45,550 - Epoch: [128][ 110/ 117] Loss 0.322967 Top1 83.686080 Top5 97.940341 +2023-10-05 21:42:45,635 - Epoch: [128][ 117/ 117] Loss 0.321099 Top1 83.742444 Top5 97.966136 +2023-10-05 21:42:45,755 - ==> Top1: 83.742 Top5: 97.966 Loss: 0.321 + +2023-10-05 21:42:45,756 - ==> Confusion: +[[ 925 2 7 2 11 1 0 0 9 61 1 2 0 2 7 3 3 1 1 0 12] + [ 3 1057 2 0 8 16 2 16 3 0 2 0 0 1 0 4 0 0 8 3 6] + [ 1 1 954 16 1 1 27 12 0 0 6 3 9 3 0 1 2 2 9 2 6] + [ 3 1 20 970 0 4 0 0 6 0 10 0 4 3 22 5 1 3 23 1 13] + [ 22 8 0 0 973 4 0 1 0 7 1 1 1 1 5 3 12 2 2 3 4] + [ 5 34 1 0 2 995 2 12 3 1 5 13 1 11 9 0 1 0 4 8 9] + [ 0 6 27 0 1 2 1121 6 0 0 2 4 0 1 1 4 0 1 3 8 4] + [ 3 25 10 0 4 37 5 1040 2 2 3 10 5 2 0 0 0 0 52 5 13] + [ 16 5 0 0 0 4 1 0 972 44 13 1 4 6 11 2 2 0 5 1 2] + [ 89 1 3 0 4 4 3 0 37 940 0 0 0 15 10 1 4 0 0 3 5] + [ 1 5 9 5 2 1 1 5 10 1 972 4 1 11 6 2 0 1 8 0 8] + [ 1 0 1 0 0 10 0 2 0 0 0 955 26 5 0 2 1 18 0 13 1] + [ 0 2 1 9 0 3 0 1 1 0 1 45 970 2 0 5 0 18 2 1 7] + [ 3 0 3 0 1 15 0 1 8 12 4 7 4 1045 3 1 2 1 0 1 8] + [ 11 5 3 16 4 0 0 0 17 2 2 2 2 1 1012 0 0 0 12 0 12] + [ 2 3 2 0 3 1 1 0 0 0 0 9 5 2 0 1066 14 10 1 8 7] + [ 0 18 1 0 5 6 1 1 3 0 0 4 1 3 3 8 1097 0 0 6 4] + [ 0 0 0 4 0 0 1 0 0 0 0 7 17 0 1 2 0 1000 3 3 0] + [ 1 6 4 19 1 0 1 12 2 1 3 0 2 0 14 1 2 0 990 1 8] + [ 0 4 1 3 1 9 5 11 1 0 2 15 4 2 0 5 12 2 3 1067 5] + [ 120 200 152 87 105 147 57 79 118 72 192 115 347 253 155 54 149 68 211 270 4954]] + +2023-10-05 21:42:45,757 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:42:45,757 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:42:45,763 - + +2023-10-05 21:42:45,763 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:42:46,756 - Epoch: [129][ 10/ 1236] Overall Loss 0.217087 Objective Loss 0.217087 LR 0.000500 Time 0.099271 +2023-10-05 21:42:46,959 - Epoch: [129][ 20/ 1236] Overall Loss 0.228170 Objective Loss 0.228170 LR 0.000500 Time 0.059734 +2023-10-05 21:42:47,158 - Epoch: [129][ 30/ 1236] Overall Loss 0.242452 Objective Loss 0.242452 LR 0.000500 Time 0.046454 +2023-10-05 21:42:47,361 - Epoch: [129][ 40/ 1236] Overall Loss 0.242258 Objective Loss 0.242258 LR 0.000500 Time 0.039899 +2023-10-05 21:42:47,561 - Epoch: [129][ 50/ 1236] Overall Loss 0.244395 Objective Loss 0.244395 LR 0.000500 Time 0.035921 +2023-10-05 21:42:47,763 - Epoch: [129][ 60/ 1236] Overall Loss 0.236751 Objective Loss 0.236751 LR 0.000500 Time 0.033294 +2023-10-05 21:42:47,963 - Epoch: [129][ 70/ 1236] Overall Loss 0.237099 Objective Loss 0.237099 LR 0.000500 Time 0.031392 +2023-10-05 21:42:48,166 - Epoch: [129][ 80/ 1236] Overall Loss 0.234235 Objective Loss 0.234235 LR 0.000500 Time 0.030001 +2023-10-05 21:42:48,366 - Epoch: [129][ 90/ 1236] Overall Loss 0.234822 Objective Loss 0.234822 LR 0.000500 Time 0.028884 +2023-10-05 21:42:48,568 - Epoch: [129][ 100/ 1236] Overall Loss 0.234429 Objective Loss 0.234429 LR 0.000500 Time 0.028017 +2023-10-05 21:42:48,768 - Epoch: [129][ 110/ 1236] Overall Loss 0.232121 Objective Loss 0.232121 LR 0.000500 Time 0.027282 +2023-10-05 21:42:48,970 - Epoch: [129][ 120/ 1236] Overall Loss 0.234152 Objective Loss 0.234152 LR 0.000500 Time 0.026688 +2023-10-05 21:42:49,170 - Epoch: [129][ 130/ 1236] Overall Loss 0.235015 Objective Loss 0.235015 LR 0.000500 Time 0.026171 +2023-10-05 21:42:49,373 - Epoch: [129][ 140/ 1236] Overall Loss 0.232731 Objective Loss 0.232731 LR 0.000500 Time 0.025748 +2023-10-05 21:42:49,572 - Epoch: [129][ 150/ 1236] Overall Loss 0.234116 Objective Loss 0.234116 LR 0.000500 Time 0.025360 +2023-10-05 21:42:49,775 - Epoch: [129][ 160/ 1236] Overall Loss 0.233579 Objective Loss 0.233579 LR 0.000500 Time 0.025036 +2023-10-05 21:42:49,974 - Epoch: [129][ 170/ 1236] Overall Loss 0.233308 Objective Loss 0.233308 LR 0.000500 Time 0.024737 +2023-10-05 21:42:50,175 - Epoch: [129][ 180/ 1236] Overall Loss 0.232922 Objective Loss 0.232922 LR 0.000500 Time 0.024476 +2023-10-05 21:42:50,375 - Epoch: [129][ 190/ 1236] Overall Loss 0.233671 Objective Loss 0.233671 LR 0.000500 Time 0.024238 +2023-10-05 21:42:50,577 - Epoch: [129][ 200/ 1236] Overall Loss 0.233520 Objective Loss 0.233520 LR 0.000500 Time 0.024035 +2023-10-05 21:42:50,777 - Epoch: [129][ 210/ 1236] Overall Loss 0.234895 Objective Loss 0.234895 LR 0.000500 Time 0.023843 +2023-10-05 21:42:50,980 - Epoch: [129][ 220/ 1236] Overall Loss 0.234329 Objective Loss 0.234329 LR 0.000500 Time 0.023680 +2023-10-05 21:42:51,180 - Epoch: [129][ 230/ 1236] Overall Loss 0.234911 Objective Loss 0.234911 LR 0.000500 Time 0.023517 +2023-10-05 21:42:51,382 - Epoch: [129][ 240/ 1236] Overall Loss 0.234988 Objective Loss 0.234988 LR 0.000500 Time 0.023378 +2023-10-05 21:42:51,582 - Epoch: [129][ 250/ 1236] Overall Loss 0.234491 Objective Loss 0.234491 LR 0.000500 Time 0.023242 +2023-10-05 21:42:51,785 - Epoch: [129][ 260/ 1236] Overall Loss 0.234286 Objective Loss 0.234286 LR 0.000500 Time 0.023127 +2023-10-05 21:42:51,985 - Epoch: [129][ 270/ 1236] Overall Loss 0.234257 Objective Loss 0.234257 LR 0.000500 Time 0.023011 +2023-10-05 21:42:52,187 - Epoch: [129][ 280/ 1236] Overall Loss 0.233194 Objective Loss 0.233194 LR 0.000500 Time 0.022908 +2023-10-05 21:42:52,387 - Epoch: [129][ 290/ 1236] Overall Loss 0.233013 Objective Loss 0.233013 LR 0.000500 Time 0.022807 +2023-10-05 21:42:52,590 - Epoch: [129][ 300/ 1236] Overall Loss 0.232673 Objective Loss 0.232673 LR 0.000500 Time 0.022722 +2023-10-05 21:42:52,790 - Epoch: [129][ 310/ 1236] Overall Loss 0.232448 Objective Loss 0.232448 LR 0.000500 Time 0.022632 +2023-10-05 21:42:52,992 - Epoch: [129][ 320/ 1236] Overall Loss 0.232716 Objective Loss 0.232716 LR 0.000500 Time 0.022556 +2023-10-05 21:42:53,193 - Epoch: [129][ 330/ 1236] Overall Loss 0.232385 Objective Loss 0.232385 LR 0.000500 Time 0.022479 +2023-10-05 21:42:53,396 - Epoch: [129][ 340/ 1236] Overall Loss 0.233059 Objective Loss 0.233059 LR 0.000500 Time 0.022414 +2023-10-05 21:42:53,595 - Epoch: [129][ 350/ 1236] Overall Loss 0.232974 Objective Loss 0.232974 LR 0.000500 Time 0.022343 +2023-10-05 21:42:53,798 - Epoch: [129][ 360/ 1236] Overall Loss 0.232509 Objective Loss 0.232509 LR 0.000500 Time 0.022285 +2023-10-05 21:42:53,998 - Epoch: [129][ 370/ 1236] Overall Loss 0.231671 Objective Loss 0.231671 LR 0.000500 Time 0.022222 +2023-10-05 21:42:54,201 - Epoch: [129][ 380/ 1236] Overall Loss 0.231617 Objective Loss 0.231617 LR 0.000500 Time 0.022170 +2023-10-05 21:42:54,401 - Epoch: [129][ 390/ 1236] Overall Loss 0.232305 Objective Loss 0.232305 LR 0.000500 Time 0.022114 +2023-10-05 21:42:54,604 - Epoch: [129][ 400/ 1236] Overall Loss 0.232336 Objective Loss 0.232336 LR 0.000500 Time 0.022067 +2023-10-05 21:42:54,804 - Epoch: [129][ 410/ 1236] Overall Loss 0.232647 Objective Loss 0.232647 LR 0.000500 Time 0.022016 +2023-10-05 21:42:55,007 - Epoch: [129][ 420/ 1236] Overall Loss 0.232136 Objective Loss 0.232136 LR 0.000500 Time 0.021974 +2023-10-05 21:42:55,206 - Epoch: [129][ 430/ 1236] Overall Loss 0.232565 Objective Loss 0.232565 LR 0.000500 Time 0.021927 +2023-10-05 21:42:55,409 - Epoch: [129][ 440/ 1236] Overall Loss 0.233177 Objective Loss 0.233177 LR 0.000500 Time 0.021889 +2023-10-05 21:42:55,609 - Epoch: [129][ 450/ 1236] Overall Loss 0.232560 Objective Loss 0.232560 LR 0.000500 Time 0.021846 +2023-10-05 21:42:55,812 - Epoch: [129][ 460/ 1236] Overall Loss 0.232748 Objective Loss 0.232748 LR 0.000500 Time 0.021810 +2023-10-05 21:42:56,011 - Epoch: [129][ 470/ 1236] Overall Loss 0.232326 Objective Loss 0.232326 LR 0.000500 Time 0.021771 +2023-10-05 21:42:56,214 - Epoch: [129][ 480/ 1236] Overall Loss 0.232496 Objective Loss 0.232496 LR 0.000500 Time 0.021739 +2023-10-05 21:42:56,414 - Epoch: [129][ 490/ 1236] Overall Loss 0.232621 Objective Loss 0.232621 LR 0.000500 Time 0.021702 +2023-10-05 21:42:56,617 - Epoch: [129][ 500/ 1236] Overall Loss 0.232564 Objective Loss 0.232564 LR 0.000500 Time 0.021673 +2023-10-05 21:42:56,816 - Epoch: [129][ 510/ 1236] Overall Loss 0.232663 Objective Loss 0.232663 LR 0.000500 Time 0.021639 +2023-10-05 21:42:57,019 - Epoch: [129][ 520/ 1236] Overall Loss 0.232744 Objective Loss 0.232744 LR 0.000500 Time 0.021613 +2023-10-05 21:42:57,219 - Epoch: [129][ 530/ 1236] Overall Loss 0.232417 Objective Loss 0.232417 LR 0.000500 Time 0.021582 +2023-10-05 21:42:57,422 - Epoch: [129][ 540/ 1236] Overall Loss 0.232650 Objective Loss 0.232650 LR 0.000500 Time 0.021556 +2023-10-05 21:42:57,622 - Epoch: [129][ 550/ 1236] Overall Loss 0.232376 Objective Loss 0.232376 LR 0.000500 Time 0.021527 +2023-10-05 21:42:57,824 - Epoch: [129][ 560/ 1236] Overall Loss 0.232510 Objective Loss 0.232510 LR 0.000500 Time 0.021503 +2023-10-05 21:42:58,024 - Epoch: [129][ 570/ 1236] Overall Loss 0.232368 Objective Loss 0.232368 LR 0.000500 Time 0.021477 +2023-10-05 21:42:58,227 - Epoch: [129][ 580/ 1236] Overall Loss 0.232738 Objective Loss 0.232738 LR 0.000500 Time 0.021456 +2023-10-05 21:42:58,427 - Epoch: [129][ 590/ 1236] Overall Loss 0.232785 Objective Loss 0.232785 LR 0.000500 Time 0.021430 +2023-10-05 21:42:58,630 - Epoch: [129][ 600/ 1236] Overall Loss 0.233417 Objective Loss 0.233417 LR 0.000500 Time 0.021410 +2023-10-05 21:42:58,830 - Epoch: [129][ 610/ 1236] Overall Loss 0.233777 Objective Loss 0.233777 LR 0.000500 Time 0.021387 +2023-10-05 21:42:59,033 - Epoch: [129][ 620/ 1236] Overall Loss 0.233926 Objective Loss 0.233926 LR 0.000500 Time 0.021369 +2023-10-05 21:42:59,233 - Epoch: [129][ 630/ 1236] Overall Loss 0.233993 Objective Loss 0.233993 LR 0.000500 Time 0.021347 +2023-10-05 21:42:59,435 - Epoch: [129][ 640/ 1236] Overall Loss 0.234454 Objective Loss 0.234454 LR 0.000500 Time 0.021329 +2023-10-05 21:42:59,636 - Epoch: [129][ 650/ 1236] Overall Loss 0.234917 Objective Loss 0.234917 LR 0.000500 Time 0.021308 +2023-10-05 21:42:59,838 - Epoch: [129][ 660/ 1236] Overall Loss 0.234892 Objective Loss 0.234892 LR 0.000500 Time 0.021292 +2023-10-05 21:43:00,038 - Epoch: [129][ 670/ 1236] Overall Loss 0.235005 Objective Loss 0.235005 LR 0.000500 Time 0.021272 +2023-10-05 21:43:00,241 - Epoch: [129][ 680/ 1236] Overall Loss 0.235121 Objective Loss 0.235121 LR 0.000500 Time 0.021257 +2023-10-05 21:43:00,441 - Epoch: [129][ 690/ 1236] Overall Loss 0.234728 Objective Loss 0.234728 LR 0.000500 Time 0.021239 +2023-10-05 21:43:00,645 - Epoch: [129][ 700/ 1236] Overall Loss 0.234673 Objective Loss 0.234673 LR 0.000500 Time 0.021225 +2023-10-05 21:43:00,844 - Epoch: [129][ 710/ 1236] Overall Loss 0.234796 Objective Loss 0.234796 LR 0.000500 Time 0.021207 +2023-10-05 21:43:01,047 - Epoch: [129][ 720/ 1236] Overall Loss 0.234667 Objective Loss 0.234667 LR 0.000500 Time 0.021194 +2023-10-05 21:43:01,247 - Epoch: [129][ 730/ 1236] Overall Loss 0.234035 Objective Loss 0.234035 LR 0.000500 Time 0.021177 +2023-10-05 21:43:01,449 - Epoch: [129][ 740/ 1236] Overall Loss 0.234141 Objective Loss 0.234141 LR 0.000500 Time 0.021163 +2023-10-05 21:43:01,650 - Epoch: [129][ 750/ 1236] Overall Loss 0.233952 Objective Loss 0.233952 LR 0.000500 Time 0.021148 +2023-10-05 21:43:01,852 - Epoch: [129][ 760/ 1236] Overall Loss 0.233914 Objective Loss 0.233914 LR 0.000500 Time 0.021135 +2023-10-05 21:43:02,052 - Epoch: [129][ 770/ 1236] Overall Loss 0.233581 Objective Loss 0.233581 LR 0.000500 Time 0.021120 +2023-10-05 21:43:02,255 - Epoch: [129][ 780/ 1236] Overall Loss 0.233735 Objective Loss 0.233735 LR 0.000500 Time 0.021109 +2023-10-05 21:43:02,455 - Epoch: [129][ 790/ 1236] Overall Loss 0.234160 Objective Loss 0.234160 LR 0.000500 Time 0.021094 +2023-10-05 21:43:02,658 - Epoch: [129][ 800/ 1236] Overall Loss 0.234439 Objective Loss 0.234439 LR 0.000500 Time 0.021084 +2023-10-05 21:43:02,857 - Epoch: [129][ 810/ 1236] Overall Loss 0.234376 Objective Loss 0.234376 LR 0.000500 Time 0.021070 +2023-10-05 21:43:03,061 - Epoch: [129][ 820/ 1236] Overall Loss 0.234300 Objective Loss 0.234300 LR 0.000500 Time 0.021060 +2023-10-05 21:43:03,260 - Epoch: [129][ 830/ 1236] Overall Loss 0.234404 Objective Loss 0.234404 LR 0.000500 Time 0.021047 +2023-10-05 21:43:03,462 - Epoch: [129][ 840/ 1236] Overall Loss 0.234510 Objective Loss 0.234510 LR 0.000500 Time 0.021036 +2023-10-05 21:43:03,662 - Epoch: [129][ 850/ 1236] Overall Loss 0.234648 Objective Loss 0.234648 LR 0.000500 Time 0.021024 +2023-10-05 21:43:03,866 - Epoch: [129][ 860/ 1236] Overall Loss 0.234708 Objective Loss 0.234708 LR 0.000500 Time 0.021015 +2023-10-05 21:43:04,065 - Epoch: [129][ 870/ 1236] Overall Loss 0.234666 Objective Loss 0.234666 LR 0.000500 Time 0.021003 +2023-10-05 21:43:04,268 - Epoch: [129][ 880/ 1236] Overall Loss 0.234830 Objective Loss 0.234830 LR 0.000500 Time 0.020994 +2023-10-05 21:43:04,468 - Epoch: [129][ 890/ 1236] Overall Loss 0.234667 Objective Loss 0.234667 LR 0.000500 Time 0.020983 +2023-10-05 21:43:04,671 - Epoch: [129][ 900/ 1236] Overall Loss 0.234594 Objective Loss 0.234594 LR 0.000500 Time 0.020975 +2023-10-05 21:43:04,871 - Epoch: [129][ 910/ 1236] Overall Loss 0.234244 Objective Loss 0.234244 LR 0.000500 Time 0.020964 +2023-10-05 21:43:05,074 - Epoch: [129][ 920/ 1236] Overall Loss 0.234254 Objective Loss 0.234254 LR 0.000500 Time 0.020956 +2023-10-05 21:43:05,274 - Epoch: [129][ 930/ 1236] Overall Loss 0.234515 Objective Loss 0.234515 LR 0.000500 Time 0.020945 +2023-10-05 21:43:05,477 - Epoch: [129][ 940/ 1236] Overall Loss 0.234196 Objective Loss 0.234196 LR 0.000500 Time 0.020938 +2023-10-05 21:43:05,677 - Epoch: [129][ 950/ 1236] Overall Loss 0.233847 Objective Loss 0.233847 LR 0.000500 Time 0.020927 +2023-10-05 21:43:05,880 - Epoch: [129][ 960/ 1236] Overall Loss 0.234001 Objective Loss 0.234001 LR 0.000500 Time 0.020920 +2023-10-05 21:43:06,079 - Epoch: [129][ 970/ 1236] Overall Loss 0.233937 Objective Loss 0.233937 LR 0.000500 Time 0.020910 +2023-10-05 21:43:06,282 - Epoch: [129][ 980/ 1236] Overall Loss 0.233987 Objective Loss 0.233987 LR 0.000500 Time 0.020903 +2023-10-05 21:43:06,482 - Epoch: [129][ 990/ 1236] Overall Loss 0.233948 Objective Loss 0.233948 LR 0.000500 Time 0.020894 +2023-10-05 21:43:06,686 - Epoch: [129][ 1000/ 1236] Overall Loss 0.234207 Objective Loss 0.234207 LR 0.000500 Time 0.020888 +2023-10-05 21:43:06,885 - Epoch: [129][ 1010/ 1236] Overall Loss 0.234517 Objective Loss 0.234517 LR 0.000500 Time 0.020879 +2023-10-05 21:43:07,088 - Epoch: [129][ 1020/ 1236] Overall Loss 0.234482 Objective Loss 0.234482 LR 0.000500 Time 0.020873 +2023-10-05 21:43:07,288 - Epoch: [129][ 1030/ 1236] Overall Loss 0.234584 Objective Loss 0.234584 LR 0.000500 Time 0.020864 +2023-10-05 21:43:07,491 - Epoch: [129][ 1040/ 1236] Overall Loss 0.234372 Objective Loss 0.234372 LR 0.000500 Time 0.020858 +2023-10-05 21:43:07,691 - Epoch: [129][ 1050/ 1236] Overall Loss 0.234322 Objective Loss 0.234322 LR 0.000500 Time 0.020849 +2023-10-05 21:43:07,894 - Epoch: [129][ 1060/ 1236] Overall Loss 0.234152 Objective Loss 0.234152 LR 0.000500 Time 0.020844 +2023-10-05 21:43:08,094 - Epoch: [129][ 1070/ 1236] Overall Loss 0.234260 Objective Loss 0.234260 LR 0.000500 Time 0.020835 +2023-10-05 21:43:08,296 - Epoch: [129][ 1080/ 1236] Overall Loss 0.234156 Objective Loss 0.234156 LR 0.000500 Time 0.020830 +2023-10-05 21:43:08,497 - Epoch: [129][ 1090/ 1236] Overall Loss 0.233883 Objective Loss 0.233883 LR 0.000500 Time 0.020822 +2023-10-05 21:43:08,699 - Epoch: [129][ 1100/ 1236] Overall Loss 0.233755 Objective Loss 0.233755 LR 0.000500 Time 0.020816 +2023-10-05 21:43:08,899 - Epoch: [129][ 1110/ 1236] Overall Loss 0.233624 Objective Loss 0.233624 LR 0.000500 Time 0.020809 +2023-10-05 21:43:09,102 - Epoch: [129][ 1120/ 1236] Overall Loss 0.233640 Objective Loss 0.233640 LR 0.000500 Time 0.020804 +2023-10-05 21:43:09,302 - Epoch: [129][ 1130/ 1236] Overall Loss 0.233516 Objective Loss 0.233516 LR 0.000500 Time 0.020797 +2023-10-05 21:43:09,506 - Epoch: [129][ 1140/ 1236] Overall Loss 0.233633 Objective Loss 0.233633 LR 0.000500 Time 0.020793 +2023-10-05 21:43:09,722 - Epoch: [129][ 1150/ 1236] Overall Loss 0.233627 Objective Loss 0.233627 LR 0.000500 Time 0.020799 +2023-10-05 21:43:09,936 - Epoch: [129][ 1160/ 1236] Overall Loss 0.233259 Objective Loss 0.233259 LR 0.000500 Time 0.020804 +2023-10-05 21:43:10,152 - Epoch: [129][ 1170/ 1236] Overall Loss 0.232910 Objective Loss 0.232910 LR 0.000500 Time 0.020811 +2023-10-05 21:43:10,366 - Epoch: [129][ 1180/ 1236] Overall Loss 0.233017 Objective Loss 0.233017 LR 0.000500 Time 0.020815 +2023-10-05 21:43:10,582 - Epoch: [129][ 1190/ 1236] Overall Loss 0.233108 Objective Loss 0.233108 LR 0.000500 Time 0.020822 +2023-10-05 21:43:10,797 - Epoch: [129][ 1200/ 1236] Overall Loss 0.233223 Objective Loss 0.233223 LR 0.000500 Time 0.020827 +2023-10-05 21:43:11,013 - Epoch: [129][ 1210/ 1236] Overall Loss 0.233135 Objective Loss 0.233135 LR 0.000500 Time 0.020833 +2023-10-05 21:43:11,227 - Epoch: [129][ 1220/ 1236] Overall Loss 0.233241 Objective Loss 0.233241 LR 0.000500 Time 0.020838 +2023-10-05 21:43:11,490 - Epoch: [129][ 1230/ 1236] Overall Loss 0.233270 Objective Loss 0.233270 LR 0.000500 Time 0.020882 +2023-10-05 21:43:11,609 - Epoch: [129][ 1236/ 1236] Overall Loss 0.233366 Objective Loss 0.233366 Top1 84.928717 Top5 98.778004 LR 0.000500 Time 0.020877 +2023-10-05 21:43:11,739 - --- validate (epoch=129)----------- +2023-10-05 21:43:11,739 - 29943 samples (256 per mini-batch) +2023-10-05 21:43:12,203 - Epoch: [129][ 10/ 117] Loss 0.323500 Top1 83.632812 Top5 97.890625 +2023-10-05 21:43:12,362 - Epoch: [129][ 20/ 117] Loss 0.314849 Top1 83.828125 Top5 97.968750 +2023-10-05 21:43:12,518 - Epoch: [129][ 30/ 117] Loss 0.310832 Top1 84.114583 Top5 97.929688 +2023-10-05 21:43:12,676 - Epoch: [129][ 40/ 117] Loss 0.312605 Top1 83.925781 Top5 98.027344 +2023-10-05 21:43:12,832 - Epoch: [129][ 50/ 117] Loss 0.314456 Top1 83.976562 Top5 98.007812 +2023-10-05 21:43:12,991 - Epoch: [129][ 60/ 117] Loss 0.322253 Top1 84.003906 Top5 97.981771 +2023-10-05 21:43:13,146 - Epoch: [129][ 70/ 117] Loss 0.326242 Top1 83.895089 Top5 98.007812 +2023-10-05 21:43:13,303 - Epoch: [129][ 80/ 117] Loss 0.326653 Top1 83.964844 Top5 97.988281 +2023-10-05 21:43:13,458 - Epoch: [129][ 90/ 117] Loss 0.328212 Top1 83.836806 Top5 97.960069 +2023-10-05 21:43:13,615 - Epoch: [129][ 100/ 117] Loss 0.328452 Top1 83.839844 Top5 97.925781 +2023-10-05 21:43:13,779 - Epoch: [129][ 110/ 117] Loss 0.327826 Top1 83.959517 Top5 97.926136 +2023-10-05 21:43:13,865 - Epoch: [129][ 117/ 117] Loss 0.328154 Top1 83.926126 Top5 97.929399 +2023-10-05 21:43:14,017 - ==> Top1: 83.926 Top5: 97.929 Loss: 0.328 + +2023-10-05 21:43:14,018 - ==> Confusion: +[[ 895 0 5 3 8 2 1 1 5 94 1 0 0 2 8 6 3 2 0 1 13] + [ 2 1034 3 1 12 18 2 19 1 0 3 1 0 0 1 3 4 2 13 3 9] + [ 2 1 953 16 4 0 29 10 0 1 4 1 6 1 2 2 1 3 7 3 10] + [ 2 0 13 969 1 3 0 1 3 1 7 0 5 4 31 3 0 5 22 3 16] + [ 20 4 1 0 971 0 0 0 0 13 2 0 0 3 17 2 8 2 1 1 5] + [ 4 30 0 1 4 988 1 25 3 2 3 3 1 10 10 1 0 0 2 12 16] + [ 0 3 22 0 1 1 1131 8 0 0 6 1 1 1 1 3 0 2 0 8 2] + [ 4 12 14 1 1 39 6 1058 0 5 2 7 2 0 0 1 0 1 45 4 16] + [ 11 4 0 0 1 3 0 0 955 61 8 0 3 13 12 5 0 0 11 1 1] + [ 60 0 3 0 2 3 0 0 23 991 1 0 0 16 6 5 2 0 0 1 6] + [ 2 6 7 11 2 1 9 2 7 2 961 4 0 9 5 2 1 1 7 4 10] + [ 3 1 2 0 1 19 0 2 1 1 0 920 32 4 0 6 1 17 1 20 4] + [ 0 2 3 5 0 5 2 0 3 0 2 33 978 1 2 4 2 13 2 4 7] + [ 3 0 1 0 0 8 1 0 13 16 5 2 4 1036 5 2 6 0 1 5 11] + [ 11 1 2 13 5 0 0 0 16 6 1 0 2 1 1019 0 0 1 12 0 11] + [ 2 2 3 1 5 1 0 0 0 0 0 5 6 2 0 1067 13 10 0 10 7] + [ 0 11 1 1 8 4 2 1 0 0 0 2 0 3 4 12 1091 0 0 6 15] + [ 0 1 0 4 0 1 3 0 1 0 0 0 23 1 1 7 0 992 1 1 2] + [ 3 4 3 16 0 0 1 15 3 0 2 0 1 0 12 0 1 0 996 2 9] + [ 0 3 8 5 1 8 12 10 0 0 1 14 4 1 0 4 7 1 2 1065 6] + [ 105 145 139 74 117 133 57 84 92 113 163 98 353 273 191 62 125 76 187 258 5060]] + +2023-10-05 21:43:14,019 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:43:14,019 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:43:14,025 - + +2023-10-05 21:43:14,025 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:43:15,030 - Epoch: [130][ 10/ 1236] Overall Loss 0.240551 Objective Loss 0.240551 LR 0.000500 Time 0.100376 +2023-10-05 21:43:15,232 - Epoch: [130][ 20/ 1236] Overall Loss 0.237047 Objective Loss 0.237047 LR 0.000500 Time 0.060290 +2023-10-05 21:43:15,433 - Epoch: [130][ 30/ 1236] Overall Loss 0.226590 Objective Loss 0.226590 LR 0.000500 Time 0.046885 +2023-10-05 21:43:15,635 - Epoch: [130][ 40/ 1236] Overall Loss 0.223300 Objective Loss 0.223300 LR 0.000500 Time 0.040197 +2023-10-05 21:43:15,836 - Epoch: [130][ 50/ 1236] Overall Loss 0.222493 Objective Loss 0.222493 LR 0.000500 Time 0.036180 +2023-10-05 21:43:16,038 - Epoch: [130][ 60/ 1236] Overall Loss 0.224158 Objective Loss 0.224158 LR 0.000500 Time 0.033510 +2023-10-05 21:43:16,239 - Epoch: [130][ 70/ 1236] Overall Loss 0.222198 Objective Loss 0.222198 LR 0.000500 Time 0.031592 +2023-10-05 21:43:16,442 - Epoch: [130][ 80/ 1236] Overall Loss 0.222571 Objective Loss 0.222571 LR 0.000500 Time 0.030169 +2023-10-05 21:43:16,643 - Epoch: [130][ 90/ 1236] Overall Loss 0.222905 Objective Loss 0.222905 LR 0.000500 Time 0.029050 +2023-10-05 21:43:16,845 - Epoch: [130][ 100/ 1236] Overall Loss 0.221585 Objective Loss 0.221585 LR 0.000500 Time 0.028163 +2023-10-05 21:43:17,046 - Epoch: [130][ 110/ 1236] Overall Loss 0.220653 Objective Loss 0.220653 LR 0.000500 Time 0.027428 +2023-10-05 21:43:17,249 - Epoch: [130][ 120/ 1236] Overall Loss 0.221675 Objective Loss 0.221675 LR 0.000500 Time 0.026824 +2023-10-05 21:43:17,450 - Epoch: [130][ 130/ 1236] Overall Loss 0.222294 Objective Loss 0.222294 LR 0.000500 Time 0.026309 +2023-10-05 21:43:17,652 - Epoch: [130][ 140/ 1236] Overall Loss 0.221630 Objective Loss 0.221630 LR 0.000500 Time 0.025870 +2023-10-05 21:43:17,854 - Epoch: [130][ 150/ 1236] Overall Loss 0.220814 Objective Loss 0.220814 LR 0.000500 Time 0.025489 +2023-10-05 21:43:18,057 - Epoch: [130][ 160/ 1236] Overall Loss 0.219537 Objective Loss 0.219537 LR 0.000500 Time 0.025162 +2023-10-05 21:43:18,259 - Epoch: [130][ 170/ 1236] Overall Loss 0.220800 Objective Loss 0.220800 LR 0.000500 Time 0.024868 +2023-10-05 21:43:18,462 - Epoch: [130][ 180/ 1236] Overall Loss 0.221273 Objective Loss 0.221273 LR 0.000500 Time 0.024612 +2023-10-05 21:43:18,664 - Epoch: [130][ 190/ 1236] Overall Loss 0.221661 Objective Loss 0.221661 LR 0.000500 Time 0.024379 +2023-10-05 21:43:18,867 - Epoch: [130][ 200/ 1236] Overall Loss 0.221047 Objective Loss 0.221047 LR 0.000500 Time 0.024172 +2023-10-05 21:43:19,069 - Epoch: [130][ 210/ 1236] Overall Loss 0.221617 Objective Loss 0.221617 LR 0.000500 Time 0.023982 +2023-10-05 21:43:19,272 - Epoch: [130][ 220/ 1236] Overall Loss 0.221506 Objective Loss 0.221506 LR 0.000500 Time 0.023812 +2023-10-05 21:43:19,474 - Epoch: [130][ 230/ 1236] Overall Loss 0.221036 Objective Loss 0.221036 LR 0.000500 Time 0.023655 +2023-10-05 21:43:19,677 - Epoch: [130][ 240/ 1236] Overall Loss 0.221554 Objective Loss 0.221554 LR 0.000500 Time 0.023513 +2023-10-05 21:43:19,879 - Epoch: [130][ 250/ 1236] Overall Loss 0.221571 Objective Loss 0.221571 LR 0.000500 Time 0.023379 +2023-10-05 21:43:20,082 - Epoch: [130][ 260/ 1236] Overall Loss 0.220872 Objective Loss 0.220872 LR 0.000500 Time 0.023259 +2023-10-05 21:43:20,284 - Epoch: [130][ 270/ 1236] Overall Loss 0.221747 Objective Loss 0.221747 LR 0.000500 Time 0.023145 +2023-10-05 21:43:20,488 - Epoch: [130][ 280/ 1236] Overall Loss 0.222253 Objective Loss 0.222253 LR 0.000500 Time 0.023044 +2023-10-05 21:43:20,689 - Epoch: [130][ 290/ 1236] Overall Loss 0.222668 Objective Loss 0.222668 LR 0.000500 Time 0.022943 +2023-10-05 21:43:20,893 - Epoch: [130][ 300/ 1236] Overall Loss 0.222552 Objective Loss 0.222552 LR 0.000500 Time 0.022856 +2023-10-05 21:43:21,094 - Epoch: [130][ 310/ 1236] Overall Loss 0.223018 Objective Loss 0.223018 LR 0.000500 Time 0.022767 +2023-10-05 21:43:21,299 - Epoch: [130][ 320/ 1236] Overall Loss 0.223851 Objective Loss 0.223851 LR 0.000500 Time 0.022693 +2023-10-05 21:43:21,500 - Epoch: [130][ 330/ 1236] Overall Loss 0.224592 Objective Loss 0.224592 LR 0.000500 Time 0.022614 +2023-10-05 21:43:21,704 - Epoch: [130][ 340/ 1236] Overall Loss 0.224728 Objective Loss 0.224728 LR 0.000500 Time 0.022548 +2023-10-05 21:43:21,905 - Epoch: [130][ 350/ 1236] Overall Loss 0.225254 Objective Loss 0.225254 LR 0.000500 Time 0.022478 +2023-10-05 21:43:22,109 - Epoch: [130][ 360/ 1236] Overall Loss 0.224751 Objective Loss 0.224751 LR 0.000500 Time 0.022419 +2023-10-05 21:43:22,311 - Epoch: [130][ 370/ 1236] Overall Loss 0.225047 Objective Loss 0.225047 LR 0.000500 Time 0.022358 +2023-10-05 21:43:22,517 - Epoch: [130][ 380/ 1236] Overall Loss 0.225248 Objective Loss 0.225248 LR 0.000500 Time 0.022309 +2023-10-05 21:43:22,719 - Epoch: [130][ 390/ 1236] Overall Loss 0.225381 Objective Loss 0.225381 LR 0.000500 Time 0.022255 +2023-10-05 21:43:22,924 - Epoch: [130][ 400/ 1236] Overall Loss 0.225162 Objective Loss 0.225162 LR 0.000500 Time 0.022209 +2023-10-05 21:43:23,126 - Epoch: [130][ 410/ 1236] Overall Loss 0.225393 Objective Loss 0.225393 LR 0.000500 Time 0.022161 +2023-10-05 21:43:23,332 - Epoch: [130][ 420/ 1236] Overall Loss 0.225539 Objective Loss 0.225539 LR 0.000500 Time 0.022121 +2023-10-05 21:43:23,534 - Epoch: [130][ 430/ 1236] Overall Loss 0.225522 Objective Loss 0.225522 LR 0.000500 Time 0.022077 +2023-10-05 21:43:23,739 - Epoch: [130][ 440/ 1236] Overall Loss 0.225307 Objective Loss 0.225307 LR 0.000500 Time 0.022039 +2023-10-05 21:43:23,941 - Epoch: [130][ 450/ 1236] Overall Loss 0.225501 Objective Loss 0.225501 LR 0.000500 Time 0.021999 +2023-10-05 21:43:24,146 - Epoch: [130][ 460/ 1236] Overall Loss 0.225422 Objective Loss 0.225422 LR 0.000500 Time 0.021965 +2023-10-05 21:43:24,349 - Epoch: [130][ 470/ 1236] Overall Loss 0.226064 Objective Loss 0.226064 LR 0.000500 Time 0.021927 +2023-10-05 21:43:24,554 - Epoch: [130][ 480/ 1236] Overall Loss 0.226297 Objective Loss 0.226297 LR 0.000500 Time 0.021897 +2023-10-05 21:43:24,756 - Epoch: [130][ 490/ 1236] Overall Loss 0.226478 Objective Loss 0.226478 LR 0.000500 Time 0.021863 +2023-10-05 21:43:24,960 - Epoch: [130][ 500/ 1236] Overall Loss 0.226555 Objective Loss 0.226555 LR 0.000500 Time 0.021833 +2023-10-05 21:43:25,162 - Epoch: [130][ 510/ 1236] Overall Loss 0.226827 Objective Loss 0.226827 LR 0.000500 Time 0.021801 +2023-10-05 21:43:25,368 - Epoch: [130][ 520/ 1236] Overall Loss 0.226754 Objective Loss 0.226754 LR 0.000500 Time 0.021775 +2023-10-05 21:43:25,570 - Epoch: [130][ 530/ 1236] Overall Loss 0.226913 Objective Loss 0.226913 LR 0.000500 Time 0.021746 +2023-10-05 21:43:25,775 - Epoch: [130][ 540/ 1236] Overall Loss 0.226910 Objective Loss 0.226910 LR 0.000500 Time 0.021722 +2023-10-05 21:43:25,976 - Epoch: [130][ 550/ 1236] Overall Loss 0.226979 Objective Loss 0.226979 LR 0.000500 Time 0.021692 +2023-10-05 21:43:26,182 - Epoch: [130][ 560/ 1236] Overall Loss 0.227165 Objective Loss 0.227165 LR 0.000500 Time 0.021671 +2023-10-05 21:43:26,384 - Epoch: [130][ 570/ 1236] Overall Loss 0.226520 Objective Loss 0.226520 LR 0.000500 Time 0.021646 +2023-10-05 21:43:26,589 - Epoch: [130][ 580/ 1236] Overall Loss 0.226106 Objective Loss 0.226106 LR 0.000500 Time 0.021624 +2023-10-05 21:43:26,791 - Epoch: [130][ 590/ 1236] Overall Loss 0.225822 Objective Loss 0.225822 LR 0.000500 Time 0.021599 +2023-10-05 21:43:26,995 - Epoch: [130][ 600/ 1236] Overall Loss 0.225676 Objective Loss 0.225676 LR 0.000500 Time 0.021578 +2023-10-05 21:43:27,197 - Epoch: [130][ 610/ 1236] Overall Loss 0.225513 Objective Loss 0.225513 LR 0.000500 Time 0.021556 +2023-10-05 21:43:27,402 - Epoch: [130][ 620/ 1236] Overall Loss 0.225603 Objective Loss 0.225603 LR 0.000500 Time 0.021539 +2023-10-05 21:43:27,604 - Epoch: [130][ 630/ 1236] Overall Loss 0.225710 Objective Loss 0.225710 LR 0.000500 Time 0.021517 +2023-10-05 21:43:27,810 - Epoch: [130][ 640/ 1236] Overall Loss 0.225596 Objective Loss 0.225596 LR 0.000500 Time 0.021501 +2023-10-05 21:43:28,012 - Epoch: [130][ 650/ 1236] Overall Loss 0.225913 Objective Loss 0.225913 LR 0.000500 Time 0.021482 +2023-10-05 21:43:28,217 - Epoch: [130][ 660/ 1236] Overall Loss 0.225867 Objective Loss 0.225867 LR 0.000500 Time 0.021466 +2023-10-05 21:43:28,420 - Epoch: [130][ 670/ 1236] Overall Loss 0.226121 Objective Loss 0.226121 LR 0.000500 Time 0.021447 +2023-10-05 21:43:28,625 - Epoch: [130][ 680/ 1236] Overall Loss 0.226417 Objective Loss 0.226417 LR 0.000500 Time 0.021433 +2023-10-05 21:43:28,827 - Epoch: [130][ 690/ 1236] Overall Loss 0.226655 Objective Loss 0.226655 LR 0.000500 Time 0.021415 +2023-10-05 21:43:29,032 - Epoch: [130][ 700/ 1236] Overall Loss 0.226556 Objective Loss 0.226556 LR 0.000500 Time 0.021401 +2023-10-05 21:43:29,235 - Epoch: [130][ 710/ 1236] Overall Loss 0.226499 Objective Loss 0.226499 LR 0.000500 Time 0.021385 +2023-10-05 21:43:29,440 - Epoch: [130][ 720/ 1236] Overall Loss 0.226353 Objective Loss 0.226353 LR 0.000500 Time 0.021372 +2023-10-05 21:43:29,642 - Epoch: [130][ 730/ 1236] Overall Loss 0.226522 Objective Loss 0.226522 LR 0.000500 Time 0.021356 +2023-10-05 21:43:29,848 - Epoch: [130][ 740/ 1236] Overall Loss 0.227012 Objective Loss 0.227012 LR 0.000500 Time 0.021345 +2023-10-05 21:43:30,049 - Epoch: [130][ 750/ 1236] Overall Loss 0.227255 Objective Loss 0.227255 LR 0.000500 Time 0.021328 +2023-10-05 21:43:30,254 - Epoch: [130][ 760/ 1236] Overall Loss 0.227088 Objective Loss 0.227088 LR 0.000500 Time 0.021317 +2023-10-05 21:43:30,457 - Epoch: [130][ 770/ 1236] Overall Loss 0.226914 Objective Loss 0.226914 LR 0.000500 Time 0.021303 +2023-10-05 21:43:30,661 - Epoch: [130][ 780/ 1236] Overall Loss 0.227008 Objective Loss 0.227008 LR 0.000500 Time 0.021291 +2023-10-05 21:43:30,863 - Epoch: [130][ 790/ 1236] Overall Loss 0.227355 Objective Loss 0.227355 LR 0.000500 Time 0.021277 +2023-10-05 21:43:31,068 - Epoch: [130][ 800/ 1236] Overall Loss 0.227182 Objective Loss 0.227182 LR 0.000500 Time 0.021267 +2023-10-05 21:43:31,270 - Epoch: [130][ 810/ 1236] Overall Loss 0.227599 Objective Loss 0.227599 LR 0.000500 Time 0.021253 +2023-10-05 21:43:31,476 - Epoch: [130][ 820/ 1236] Overall Loss 0.227341 Objective Loss 0.227341 LR 0.000500 Time 0.021244 +2023-10-05 21:43:31,678 - Epoch: [130][ 830/ 1236] Overall Loss 0.227382 Objective Loss 0.227382 LR 0.000500 Time 0.021232 +2023-10-05 21:43:31,883 - Epoch: [130][ 840/ 1236] Overall Loss 0.227235 Objective Loss 0.227235 LR 0.000500 Time 0.021223 +2023-10-05 21:43:32,085 - Epoch: [130][ 850/ 1236] Overall Loss 0.226970 Objective Loss 0.226970 LR 0.000500 Time 0.021210 +2023-10-05 21:43:32,291 - Epoch: [130][ 860/ 1236] Overall Loss 0.226991 Objective Loss 0.226991 LR 0.000500 Time 0.021202 +2023-10-05 21:43:32,493 - Epoch: [130][ 870/ 1236] Overall Loss 0.226946 Objective Loss 0.226946 LR 0.000500 Time 0.021190 +2023-10-05 21:43:32,698 - Epoch: [130][ 880/ 1236] Overall Loss 0.227099 Objective Loss 0.227099 LR 0.000500 Time 0.021182 +2023-10-05 21:43:32,900 - Epoch: [130][ 890/ 1236] Overall Loss 0.227076 Objective Loss 0.227076 LR 0.000500 Time 0.021171 +2023-10-05 21:43:33,104 - Epoch: [130][ 900/ 1236] Overall Loss 0.227153 Objective Loss 0.227153 LR 0.000500 Time 0.021162 +2023-10-05 21:43:33,307 - Epoch: [130][ 910/ 1236] Overall Loss 0.227259 Objective Loss 0.227259 LR 0.000500 Time 0.021152 +2023-10-05 21:43:33,512 - Epoch: [130][ 920/ 1236] Overall Loss 0.227384 Objective Loss 0.227384 LR 0.000500 Time 0.021144 +2023-10-05 21:43:33,715 - Epoch: [130][ 930/ 1236] Overall Loss 0.227531 Objective Loss 0.227531 LR 0.000500 Time 0.021134 +2023-10-05 21:43:33,920 - Epoch: [130][ 940/ 1236] Overall Loss 0.227637 Objective Loss 0.227637 LR 0.000500 Time 0.021127 +2023-10-05 21:43:34,122 - Epoch: [130][ 950/ 1236] Overall Loss 0.227391 Objective Loss 0.227391 LR 0.000500 Time 0.021118 +2023-10-05 21:43:34,327 - Epoch: [130][ 960/ 1236] Overall Loss 0.227175 Objective Loss 0.227175 LR 0.000500 Time 0.021111 +2023-10-05 21:43:34,530 - Epoch: [130][ 970/ 1236] Overall Loss 0.227210 Objective Loss 0.227210 LR 0.000500 Time 0.021102 +2023-10-05 21:43:34,735 - Epoch: [130][ 980/ 1236] Overall Loss 0.227113 Objective Loss 0.227113 LR 0.000500 Time 0.021095 +2023-10-05 21:43:34,937 - Epoch: [130][ 990/ 1236] Overall Loss 0.227218 Objective Loss 0.227218 LR 0.000500 Time 0.021086 +2023-10-05 21:43:35,143 - Epoch: [130][ 1000/ 1236] Overall Loss 0.227333 Objective Loss 0.227333 LR 0.000500 Time 0.021080 +2023-10-05 21:43:35,345 - Epoch: [130][ 1010/ 1236] Overall Loss 0.227135 Objective Loss 0.227135 LR 0.000500 Time 0.021072 +2023-10-05 21:43:35,550 - Epoch: [130][ 1020/ 1236] Overall Loss 0.227339 Objective Loss 0.227339 LR 0.000500 Time 0.021066 +2023-10-05 21:43:35,753 - Epoch: [130][ 1030/ 1236] Overall Loss 0.227247 Objective Loss 0.227247 LR 0.000500 Time 0.021057 +2023-10-05 21:43:35,957 - Epoch: [130][ 1040/ 1236] Overall Loss 0.227345 Objective Loss 0.227345 LR 0.000500 Time 0.021051 +2023-10-05 21:43:36,158 - Epoch: [130][ 1050/ 1236] Overall Loss 0.227302 Objective Loss 0.227302 LR 0.000500 Time 0.021042 +2023-10-05 21:43:36,364 - Epoch: [130][ 1060/ 1236] Overall Loss 0.227506 Objective Loss 0.227506 LR 0.000500 Time 0.021037 +2023-10-05 21:43:36,566 - Epoch: [130][ 1070/ 1236] Overall Loss 0.227330 Objective Loss 0.227330 LR 0.000500 Time 0.021030 +2023-10-05 21:43:36,770 - Epoch: [130][ 1080/ 1236] Overall Loss 0.227629 Objective Loss 0.227629 LR 0.000500 Time 0.021023 +2023-10-05 21:43:36,972 - Epoch: [130][ 1090/ 1236] Overall Loss 0.227600 Objective Loss 0.227600 LR 0.000500 Time 0.021015 +2023-10-05 21:43:37,177 - Epoch: [130][ 1100/ 1236] Overall Loss 0.227658 Objective Loss 0.227658 LR 0.000500 Time 0.021010 +2023-10-05 21:43:37,379 - Epoch: [130][ 1110/ 1236] Overall Loss 0.227726 Objective Loss 0.227726 LR 0.000500 Time 0.021003 +2023-10-05 21:43:37,584 - Epoch: [130][ 1120/ 1236] Overall Loss 0.227876 Objective Loss 0.227876 LR 0.000500 Time 0.020998 +2023-10-05 21:43:37,787 - Epoch: [130][ 1130/ 1236] Overall Loss 0.227848 Objective Loss 0.227848 LR 0.000500 Time 0.020991 +2023-10-05 21:43:37,992 - Epoch: [130][ 1140/ 1236] Overall Loss 0.227727 Objective Loss 0.227727 LR 0.000500 Time 0.020986 +2023-10-05 21:43:38,194 - Epoch: [130][ 1150/ 1236] Overall Loss 0.227731 Objective Loss 0.227731 LR 0.000500 Time 0.020980 +2023-10-05 21:43:38,399 - Epoch: [130][ 1160/ 1236] Overall Loss 0.228019 Objective Loss 0.228019 LR 0.000500 Time 0.020975 +2023-10-05 21:43:38,601 - Epoch: [130][ 1170/ 1236] Overall Loss 0.228065 Objective Loss 0.228065 LR 0.000500 Time 0.020968 +2023-10-05 21:43:38,807 - Epoch: [130][ 1180/ 1236] Overall Loss 0.227987 Objective Loss 0.227987 LR 0.000500 Time 0.020964 +2023-10-05 21:43:39,009 - Epoch: [130][ 1190/ 1236] Overall Loss 0.228091 Objective Loss 0.228091 LR 0.000500 Time 0.020958 +2023-10-05 21:43:39,214 - Epoch: [130][ 1200/ 1236] Overall Loss 0.228070 Objective Loss 0.228070 LR 0.000500 Time 0.020953 +2023-10-05 21:43:39,417 - Epoch: [130][ 1210/ 1236] Overall Loss 0.228023 Objective Loss 0.228023 LR 0.000500 Time 0.020948 +2023-10-05 21:43:39,622 - Epoch: [130][ 1220/ 1236] Overall Loss 0.228210 Objective Loss 0.228210 LR 0.000500 Time 0.020944 +2023-10-05 21:43:39,881 - Epoch: [130][ 1230/ 1236] Overall Loss 0.227951 Objective Loss 0.227951 LR 0.000500 Time 0.020984 +2023-10-05 21:43:40,000 - Epoch: [130][ 1236/ 1236] Overall Loss 0.228060 Objective Loss 0.228060 Top1 87.983707 Top5 99.185336 LR 0.000500 Time 0.020978 +2023-10-05 21:43:40,121 - --- validate (epoch=130)----------- +2023-10-05 21:43:40,122 - 29943 samples (256 per mini-batch) +2023-10-05 21:43:40,599 - Epoch: [130][ 10/ 117] Loss 0.358705 Top1 83.281250 Top5 97.773438 +2023-10-05 21:43:40,761 - Epoch: [130][ 20/ 117] Loss 0.346075 Top1 84.238281 Top5 97.890625 +2023-10-05 21:43:40,916 - Epoch: [130][ 30/ 117] Loss 0.345907 Top1 83.945312 Top5 97.786458 +2023-10-05 21:43:41,073 - Epoch: [130][ 40/ 117] Loss 0.335251 Top1 84.150391 Top5 97.998047 +2023-10-05 21:43:41,228 - Epoch: [130][ 50/ 117] Loss 0.336374 Top1 84.179688 Top5 98.007812 +2023-10-05 21:43:41,384 - Epoch: [130][ 60/ 117] Loss 0.328871 Top1 84.225260 Top5 98.014323 +2023-10-05 21:43:41,535 - Epoch: [130][ 70/ 117] Loss 0.330239 Top1 84.241071 Top5 97.985491 +2023-10-05 21:43:41,688 - Epoch: [130][ 80/ 117] Loss 0.324814 Top1 84.389648 Top5 97.998047 +2023-10-05 21:43:41,840 - Epoch: [130][ 90/ 117] Loss 0.325946 Top1 84.292535 Top5 97.951389 +2023-10-05 21:43:41,993 - Epoch: [130][ 100/ 117] Loss 0.325401 Top1 84.183594 Top5 97.968750 +2023-10-05 21:43:42,150 - Epoch: [130][ 110/ 117] Loss 0.327705 Top1 84.183239 Top5 97.926136 +2023-10-05 21:43:42,236 - Epoch: [130][ 117/ 117] Loss 0.325590 Top1 84.256755 Top5 97.916040 +2023-10-05 21:43:42,335 - ==> Top1: 84.257 Top5: 97.916 Loss: 0.326 + +2023-10-05 21:43:42,335 - ==> Confusion: +[[ 927 0 3 1 10 3 0 1 5 74 1 0 0 1 4 1 6 0 0 0 13] + [ 0 1033 2 0 8 31 1 25 4 0 4 2 0 0 0 3 2 0 6 3 7] + [ 4 1 954 8 1 0 23 13 0 1 7 3 8 2 4 3 1 0 8 5 10] + [ 3 2 20 945 0 2 0 0 8 1 11 0 6 3 25 4 1 4 31 2 21] + [ 21 4 1 0 981 4 0 2 1 10 1 0 0 1 7 4 7 2 0 1 3] + [ 4 25 0 0 6 996 1 19 1 0 3 16 0 14 6 1 6 1 3 2 12] + [ 0 5 28 0 0 2 1114 11 0 0 5 3 2 1 1 7 0 1 0 5 6] + [ 3 9 10 0 2 36 5 1078 4 2 2 8 2 2 0 5 0 1 33 7 9] + [ 22 2 0 0 1 2 0 0 968 45 12 2 2 9 12 5 2 0 1 2 2] + [ 93 0 2 0 4 4 0 1 21 952 2 1 0 18 5 5 0 0 0 3 8] + [ 3 3 10 1 2 2 6 8 13 2 968 3 0 11 2 1 3 1 3 2 9] + [ 1 0 1 0 0 11 1 1 0 1 0 969 21 3 0 1 2 16 0 5 2] + [ 1 2 3 3 0 2 2 0 2 0 2 37 976 2 1 4 3 11 2 2 13] + [ 5 0 2 0 1 9 0 0 7 11 4 5 1 1057 4 4 0 0 0 0 9] + [ 12 2 3 7 7 2 0 0 37 3 2 1 1 2 995 1 1 3 11 0 11] + [ 2 2 3 0 5 1 0 0 0 0 0 8 4 1 0 1065 18 7 1 12 5] + [ 0 8 1 1 8 4 1 2 2 0 0 5 1 3 5 5 1104 0 1 3 7] + [ 0 0 0 1 0 1 2 0 1 0 0 2 22 1 0 5 1 992 1 2 7] + [ 2 9 4 10 0 2 1 24 2 1 3 2 1 0 8 0 0 0 985 1 13] + [ 0 4 1 0 2 3 8 10 1 0 2 19 3 4 0 5 10 1 1 1072 6] + [ 175 142 127 46 94 144 39 110 120 81 177 132 336 307 132 63 197 62 137 186 5098]] + +2023-10-05 21:43:42,337 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:43:42,337 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:43:42,343 - + +2023-10-05 21:43:42,343 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:43:43,358 - Epoch: [131][ 10/ 1236] Overall Loss 0.218474 Objective Loss 0.218474 LR 0.000500 Time 0.101470 +2023-10-05 21:43:43,562 - Epoch: [131][ 20/ 1236] Overall Loss 0.207401 Objective Loss 0.207401 LR 0.000500 Time 0.060927 +2023-10-05 21:43:43,764 - Epoch: [131][ 30/ 1236] Overall Loss 0.217853 Objective Loss 0.217853 LR 0.000500 Time 0.047336 +2023-10-05 21:43:43,968 - Epoch: [131][ 40/ 1236] Overall Loss 0.212978 Objective Loss 0.212978 LR 0.000500 Time 0.040596 +2023-10-05 21:43:44,170 - Epoch: [131][ 50/ 1236] Overall Loss 0.216607 Objective Loss 0.216607 LR 0.000500 Time 0.036510 +2023-10-05 21:43:44,374 - Epoch: [131][ 60/ 1236] Overall Loss 0.217173 Objective Loss 0.217173 LR 0.000500 Time 0.033821 +2023-10-05 21:43:44,577 - Epoch: [131][ 70/ 1236] Overall Loss 0.211592 Objective Loss 0.211592 LR 0.000500 Time 0.031872 +2023-10-05 21:43:44,781 - Epoch: [131][ 80/ 1236] Overall Loss 0.214217 Objective Loss 0.214217 LR 0.000500 Time 0.030437 +2023-10-05 21:43:44,983 - Epoch: [131][ 90/ 1236] Overall Loss 0.213137 Objective Loss 0.213137 LR 0.000500 Time 0.029293 +2023-10-05 21:43:45,187 - Epoch: [131][ 100/ 1236] Overall Loss 0.213634 Objective Loss 0.213634 LR 0.000500 Time 0.028403 +2023-10-05 21:43:45,393 - Epoch: [131][ 110/ 1236] Overall Loss 0.214434 Objective Loss 0.214434 LR 0.000500 Time 0.027693 +2023-10-05 21:43:45,601 - Epoch: [131][ 120/ 1236] Overall Loss 0.214217 Objective Loss 0.214217 LR 0.000500 Time 0.027115 +2023-10-05 21:43:45,803 - Epoch: [131][ 130/ 1236] Overall Loss 0.217194 Objective Loss 0.217194 LR 0.000500 Time 0.026578 +2023-10-05 21:43:46,007 - Epoch: [131][ 140/ 1236] Overall Loss 0.217023 Objective Loss 0.217023 LR 0.000500 Time 0.026132 +2023-10-05 21:43:46,208 - Epoch: [131][ 150/ 1236] Overall Loss 0.217658 Objective Loss 0.217658 LR 0.000500 Time 0.025730 +2023-10-05 21:43:46,412 - Epoch: [131][ 160/ 1236] Overall Loss 0.217722 Objective Loss 0.217722 LR 0.000500 Time 0.025393 +2023-10-05 21:43:46,613 - Epoch: [131][ 170/ 1236] Overall Loss 0.217854 Objective Loss 0.217854 LR 0.000500 Time 0.025081 +2023-10-05 21:43:46,817 - Epoch: [131][ 180/ 1236] Overall Loss 0.218495 Objective Loss 0.218495 LR 0.000500 Time 0.024818 +2023-10-05 21:43:47,018 - Epoch: [131][ 190/ 1236] Overall Loss 0.217521 Objective Loss 0.217521 LR 0.000500 Time 0.024571 +2023-10-05 21:43:47,223 - Epoch: [131][ 200/ 1236] Overall Loss 0.219636 Objective Loss 0.219636 LR 0.000500 Time 0.024361 +2023-10-05 21:43:47,424 - Epoch: [131][ 210/ 1236] Overall Loss 0.218981 Objective Loss 0.218981 LR 0.000500 Time 0.024157 +2023-10-05 21:43:47,628 - Epoch: [131][ 220/ 1236] Overall Loss 0.218979 Objective Loss 0.218979 LR 0.000500 Time 0.023984 +2023-10-05 21:43:47,829 - Epoch: [131][ 230/ 1236] Overall Loss 0.219722 Objective Loss 0.219722 LR 0.000500 Time 0.023814 +2023-10-05 21:43:48,033 - Epoch: [131][ 240/ 1236] Overall Loss 0.220502 Objective Loss 0.220502 LR 0.000500 Time 0.023671 +2023-10-05 21:43:48,234 - Epoch: [131][ 250/ 1236] Overall Loss 0.220258 Objective Loss 0.220258 LR 0.000500 Time 0.023526 +2023-10-05 21:43:48,437 - Epoch: [131][ 260/ 1236] Overall Loss 0.221066 Objective Loss 0.221066 LR 0.000500 Time 0.023404 +2023-10-05 21:43:48,639 - Epoch: [131][ 270/ 1236] Overall Loss 0.221618 Objective Loss 0.221618 LR 0.000500 Time 0.023281 +2023-10-05 21:43:48,843 - Epoch: [131][ 280/ 1236] Overall Loss 0.222411 Objective Loss 0.222411 LR 0.000500 Time 0.023176 +2023-10-05 21:43:49,044 - Epoch: [131][ 290/ 1236] Overall Loss 0.222315 Objective Loss 0.222315 LR 0.000500 Time 0.023070 +2023-10-05 21:43:49,248 - Epoch: [131][ 300/ 1236] Overall Loss 0.222730 Objective Loss 0.222730 LR 0.000500 Time 0.022980 +2023-10-05 21:43:49,451 - Epoch: [131][ 310/ 1236] Overall Loss 0.222214 Objective Loss 0.222214 LR 0.000500 Time 0.022892 +2023-10-05 21:43:49,654 - Epoch: [131][ 320/ 1236] Overall Loss 0.221683 Objective Loss 0.221683 LR 0.000500 Time 0.022811 +2023-10-05 21:43:49,855 - Epoch: [131][ 330/ 1236] Overall Loss 0.221838 Objective Loss 0.221838 LR 0.000500 Time 0.022728 +2023-10-05 21:43:50,058 - Epoch: [131][ 340/ 1236] Overall Loss 0.225803 Objective Loss 0.225803 LR 0.000500 Time 0.022655 +2023-10-05 21:43:50,261 - Epoch: [131][ 350/ 1236] Overall Loss 0.229923 Objective Loss 0.229923 LR 0.000500 Time 0.022587 +2023-10-05 21:43:50,466 - Epoch: [131][ 360/ 1236] Overall Loss 0.234401 Objective Loss 0.234401 LR 0.000500 Time 0.022526 +2023-10-05 21:43:50,677 - Epoch: [131][ 370/ 1236] Overall Loss 0.237168 Objective Loss 0.237168 LR 0.000500 Time 0.022487 +2023-10-05 21:43:50,892 - Epoch: [131][ 380/ 1236] Overall Loss 0.240659 Objective Loss 0.240659 LR 0.000500 Time 0.022460 +2023-10-05 21:43:51,101 - Epoch: [131][ 390/ 1236] Overall Loss 0.243290 Objective Loss 0.243290 LR 0.000500 Time 0.022420 +2023-10-05 21:43:51,316 - Epoch: [131][ 400/ 1236] Overall Loss 0.246234 Objective Loss 0.246234 LR 0.000500 Time 0.022396 +2023-10-05 21:43:51,526 - Epoch: [131][ 410/ 1236] Overall Loss 0.247802 Objective Loss 0.247802 LR 0.000500 Time 0.022360 +2023-10-05 21:43:51,740 - Epoch: [131][ 420/ 1236] Overall Loss 0.249773 Objective Loss 0.249773 LR 0.000500 Time 0.022338 +2023-10-05 21:43:51,950 - Epoch: [131][ 430/ 1236] Overall Loss 0.251567 Objective Loss 0.251567 LR 0.000500 Time 0.022305 +2023-10-05 21:43:52,164 - Epoch: [131][ 440/ 1236] Overall Loss 0.253856 Objective Loss 0.253856 LR 0.000500 Time 0.022285 +2023-10-05 21:43:52,374 - Epoch: [131][ 450/ 1236] Overall Loss 0.255769 Objective Loss 0.255769 LR 0.000500 Time 0.022255 +2023-10-05 21:43:52,589 - Epoch: [131][ 460/ 1236] Overall Loss 0.257404 Objective Loss 0.257404 LR 0.000500 Time 0.022237 +2023-10-05 21:43:52,798 - Epoch: [131][ 470/ 1236] Overall Loss 0.258882 Objective Loss 0.258882 LR 0.000500 Time 0.022209 +2023-10-05 21:43:53,013 - Epoch: [131][ 480/ 1236] Overall Loss 0.260461 Objective Loss 0.260461 LR 0.000500 Time 0.022194 +2023-10-05 21:43:53,215 - Epoch: [131][ 490/ 1236] Overall Loss 0.261912 Objective Loss 0.261912 LR 0.000500 Time 0.022152 +2023-10-05 21:43:53,419 - Epoch: [131][ 500/ 1236] Overall Loss 0.263531 Objective Loss 0.263531 LR 0.000500 Time 0.022116 +2023-10-05 21:43:53,621 - Epoch: [131][ 510/ 1236] Overall Loss 0.265171 Objective Loss 0.265171 LR 0.000500 Time 0.022078 +2023-10-05 21:43:53,825 - Epoch: [131][ 520/ 1236] Overall Loss 0.266445 Objective Loss 0.266445 LR 0.000500 Time 0.022045 +2023-10-05 21:43:54,026 - Epoch: [131][ 530/ 1236] Overall Loss 0.267144 Objective Loss 0.267144 LR 0.000500 Time 0.022009 +2023-10-05 21:43:54,231 - Epoch: [131][ 540/ 1236] Overall Loss 0.268274 Objective Loss 0.268274 LR 0.000500 Time 0.021979 +2023-10-05 21:43:54,432 - Epoch: [131][ 550/ 1236] Overall Loss 0.269254 Objective Loss 0.269254 LR 0.000500 Time 0.021945 +2023-10-05 21:43:54,636 - Epoch: [131][ 560/ 1236] Overall Loss 0.270090 Objective Loss 0.270090 LR 0.000500 Time 0.021917 +2023-10-05 21:43:54,838 - Epoch: [131][ 570/ 1236] Overall Loss 0.271314 Objective Loss 0.271314 LR 0.000500 Time 0.021886 +2023-10-05 21:43:55,042 - Epoch: [131][ 580/ 1236] Overall Loss 0.272240 Objective Loss 0.272240 LR 0.000500 Time 0.021859 +2023-10-05 21:43:55,244 - Epoch: [131][ 590/ 1236] Overall Loss 0.273266 Objective Loss 0.273266 LR 0.000500 Time 0.021830 +2023-10-05 21:43:55,448 - Epoch: [131][ 600/ 1236] Overall Loss 0.274063 Objective Loss 0.274063 LR 0.000500 Time 0.021806 +2023-10-05 21:43:55,650 - Epoch: [131][ 610/ 1236] Overall Loss 0.274814 Objective Loss 0.274814 LR 0.000500 Time 0.021779 +2023-10-05 21:43:55,854 - Epoch: [131][ 620/ 1236] Overall Loss 0.275799 Objective Loss 0.275799 LR 0.000500 Time 0.021756 +2023-10-05 21:43:56,055 - Epoch: [131][ 630/ 1236] Overall Loss 0.276236 Objective Loss 0.276236 LR 0.000500 Time 0.021730 +2023-10-05 21:43:56,260 - Epoch: [131][ 640/ 1236] Overall Loss 0.276810 Objective Loss 0.276810 LR 0.000500 Time 0.021709 +2023-10-05 21:43:56,461 - Epoch: [131][ 650/ 1236] Overall Loss 0.277229 Objective Loss 0.277229 LR 0.000500 Time 0.021685 +2023-10-05 21:43:56,666 - Epoch: [131][ 660/ 1236] Overall Loss 0.277730 Objective Loss 0.277730 LR 0.000500 Time 0.021666 +2023-10-05 21:43:56,867 - Epoch: [131][ 670/ 1236] Overall Loss 0.278570 Objective Loss 0.278570 LR 0.000500 Time 0.021643 +2023-10-05 21:43:57,072 - Epoch: [131][ 680/ 1236] Overall Loss 0.279221 Objective Loss 0.279221 LR 0.000500 Time 0.021624 +2023-10-05 21:43:57,273 - Epoch: [131][ 690/ 1236] Overall Loss 0.279852 Objective Loss 0.279852 LR 0.000500 Time 0.021603 +2023-10-05 21:43:57,477 - Epoch: [131][ 700/ 1236] Overall Loss 0.280513 Objective Loss 0.280513 LR 0.000500 Time 0.021585 +2023-10-05 21:43:57,679 - Epoch: [131][ 710/ 1236] Overall Loss 0.280812 Objective Loss 0.280812 LR 0.000500 Time 0.021564 +2023-10-05 21:43:57,883 - Epoch: [131][ 720/ 1236] Overall Loss 0.281301 Objective Loss 0.281301 LR 0.000500 Time 0.021547 +2023-10-05 21:43:58,085 - Epoch: [131][ 730/ 1236] Overall Loss 0.281768 Objective Loss 0.281768 LR 0.000500 Time 0.021528 +2023-10-05 21:43:58,289 - Epoch: [131][ 740/ 1236] Overall Loss 0.282165 Objective Loss 0.282165 LR 0.000500 Time 0.021513 +2023-10-05 21:43:58,491 - Epoch: [131][ 750/ 1236] Overall Loss 0.282396 Objective Loss 0.282396 LR 0.000500 Time 0.021495 +2023-10-05 21:43:58,695 - Epoch: [131][ 760/ 1236] Overall Loss 0.282547 Objective Loss 0.282547 LR 0.000500 Time 0.021480 +2023-10-05 21:43:58,897 - Epoch: [131][ 770/ 1236] Overall Loss 0.282790 Objective Loss 0.282790 LR 0.000500 Time 0.021463 +2023-10-05 21:43:59,101 - Epoch: [131][ 780/ 1236] Overall Loss 0.282802 Objective Loss 0.282802 LR 0.000500 Time 0.021449 +2023-10-05 21:43:59,303 - Epoch: [131][ 790/ 1236] Overall Loss 0.282796 Objective Loss 0.282796 LR 0.000500 Time 0.021433 +2023-10-05 21:43:59,507 - Epoch: [131][ 800/ 1236] Overall Loss 0.283023 Objective Loss 0.283023 LR 0.000500 Time 0.021420 +2023-10-05 21:43:59,711 - Epoch: [131][ 810/ 1236] Overall Loss 0.283202 Objective Loss 0.283202 LR 0.000500 Time 0.021407 +2023-10-05 21:43:59,916 - Epoch: [131][ 820/ 1236] Overall Loss 0.283544 Objective Loss 0.283544 LR 0.000500 Time 0.021395 +2023-10-05 21:44:00,120 - Epoch: [131][ 830/ 1236] Overall Loss 0.283970 Objective Loss 0.283970 LR 0.000500 Time 0.021383 +2023-10-05 21:44:00,325 - Epoch: [131][ 840/ 1236] Overall Loss 0.284091 Objective Loss 0.284091 LR 0.000500 Time 0.021372 +2023-10-05 21:44:00,529 - Epoch: [131][ 850/ 1236] Overall Loss 0.284324 Objective Loss 0.284324 LR 0.000500 Time 0.021360 +2023-10-05 21:44:00,734 - Epoch: [131][ 860/ 1236] Overall Loss 0.284097 Objective Loss 0.284097 LR 0.000500 Time 0.021349 +2023-10-05 21:44:00,938 - Epoch: [131][ 870/ 1236] Overall Loss 0.284347 Objective Loss 0.284347 LR 0.000500 Time 0.021339 +2023-10-05 21:44:01,144 - Epoch: [131][ 880/ 1236] Overall Loss 0.284715 Objective Loss 0.284715 LR 0.000500 Time 0.021329 +2023-10-05 21:44:01,348 - Epoch: [131][ 890/ 1236] Overall Loss 0.284735 Objective Loss 0.284735 LR 0.000500 Time 0.021319 +2023-10-05 21:44:01,553 - Epoch: [131][ 900/ 1236] Overall Loss 0.284733 Objective Loss 0.284733 LR 0.000500 Time 0.021308 +2023-10-05 21:44:01,757 - Epoch: [131][ 910/ 1236] Overall Loss 0.284973 Objective Loss 0.284973 LR 0.000500 Time 0.021298 +2023-10-05 21:44:01,962 - Epoch: [131][ 920/ 1236] Overall Loss 0.285321 Objective Loss 0.285321 LR 0.000500 Time 0.021289 +2023-10-05 21:44:02,166 - Epoch: [131][ 930/ 1236] Overall Loss 0.285335 Objective Loss 0.285335 LR 0.000500 Time 0.021280 +2023-10-05 21:44:02,371 - Epoch: [131][ 940/ 1236] Overall Loss 0.285607 Objective Loss 0.285607 LR 0.000500 Time 0.021270 +2023-10-05 21:44:02,575 - Epoch: [131][ 950/ 1236] Overall Loss 0.285730 Objective Loss 0.285730 LR 0.000500 Time 0.021262 +2023-10-05 21:44:02,780 - Epoch: [131][ 960/ 1236] Overall Loss 0.286013 Objective Loss 0.286013 LR 0.000500 Time 0.021253 +2023-10-05 21:44:02,984 - Epoch: [131][ 970/ 1236] Overall Loss 0.286224 Objective Loss 0.286224 LR 0.000500 Time 0.021244 +2023-10-05 21:44:03,189 - Epoch: [131][ 980/ 1236] Overall Loss 0.286506 Objective Loss 0.286506 LR 0.000500 Time 0.021236 +2023-10-05 21:44:03,394 - Epoch: [131][ 990/ 1236] Overall Loss 0.286538 Objective Loss 0.286538 LR 0.000500 Time 0.021228 +2023-10-05 21:44:03,598 - Epoch: [131][ 1000/ 1236] Overall Loss 0.286542 Objective Loss 0.286542 LR 0.000500 Time 0.021220 +2023-10-05 21:44:03,803 - Epoch: [131][ 1010/ 1236] Overall Loss 0.286796 Objective Loss 0.286796 LR 0.000500 Time 0.021212 +2023-10-05 21:44:04,008 - Epoch: [131][ 1020/ 1236] Overall Loss 0.286879 Objective Loss 0.286879 LR 0.000500 Time 0.021204 +2023-10-05 21:44:04,213 - Epoch: [131][ 1030/ 1236] Overall Loss 0.287070 Objective Loss 0.287070 LR 0.000500 Time 0.021197 +2023-10-05 21:44:04,418 - Epoch: [131][ 1040/ 1236] Overall Loss 0.286912 Objective Loss 0.286912 LR 0.000500 Time 0.021190 +2023-10-05 21:44:04,622 - Epoch: [131][ 1050/ 1236] Overall Loss 0.287024 Objective Loss 0.287024 LR 0.000500 Time 0.021183 +2023-10-05 21:44:04,827 - Epoch: [131][ 1060/ 1236] Overall Loss 0.286974 Objective Loss 0.286974 LR 0.000500 Time 0.021176 +2023-10-05 21:44:05,031 - Epoch: [131][ 1070/ 1236] Overall Loss 0.287233 Objective Loss 0.287233 LR 0.000500 Time 0.021169 +2023-10-05 21:44:05,236 - Epoch: [131][ 1080/ 1236] Overall Loss 0.287452 Objective Loss 0.287452 LR 0.000500 Time 0.021162 +2023-10-05 21:44:05,441 - Epoch: [131][ 1090/ 1236] Overall Loss 0.287597 Objective Loss 0.287597 LR 0.000500 Time 0.021155 +2023-10-05 21:44:05,646 - Epoch: [131][ 1100/ 1236] Overall Loss 0.287562 Objective Loss 0.287562 LR 0.000500 Time 0.021149 +2023-10-05 21:44:05,850 - Epoch: [131][ 1110/ 1236] Overall Loss 0.287702 Objective Loss 0.287702 LR 0.000500 Time 0.021142 +2023-10-05 21:44:06,055 - Epoch: [131][ 1120/ 1236] Overall Loss 0.287802 Objective Loss 0.287802 LR 0.000500 Time 0.021136 +2023-10-05 21:44:06,260 - Epoch: [131][ 1130/ 1236] Overall Loss 0.288030 Objective Loss 0.288030 LR 0.000500 Time 0.021130 +2023-10-05 21:44:06,465 - Epoch: [131][ 1140/ 1236] Overall Loss 0.287909 Objective Loss 0.287909 LR 0.000500 Time 0.021124 +2023-10-05 21:44:06,670 - Epoch: [131][ 1150/ 1236] Overall Loss 0.287969 Objective Loss 0.287969 LR 0.000500 Time 0.021118 +2023-10-05 21:44:06,875 - Epoch: [131][ 1160/ 1236] Overall Loss 0.288083 Objective Loss 0.288083 LR 0.000500 Time 0.021113 +2023-10-05 21:44:07,079 - Epoch: [131][ 1170/ 1236] Overall Loss 0.288165 Objective Loss 0.288165 LR 0.000500 Time 0.021107 +2023-10-05 21:44:07,284 - Epoch: [131][ 1180/ 1236] Overall Loss 0.288186 Objective Loss 0.288186 LR 0.000500 Time 0.021101 +2023-10-05 21:44:07,489 - Epoch: [131][ 1190/ 1236] Overall Loss 0.288239 Objective Loss 0.288239 LR 0.000500 Time 0.021095 +2023-10-05 21:44:07,694 - Epoch: [131][ 1200/ 1236] Overall Loss 0.288438 Objective Loss 0.288438 LR 0.000500 Time 0.021090 +2023-10-05 21:44:07,899 - Epoch: [131][ 1210/ 1236] Overall Loss 0.288562 Objective Loss 0.288562 LR 0.000500 Time 0.021085 +2023-10-05 21:44:08,104 - Epoch: [131][ 1220/ 1236] Overall Loss 0.288770 Objective Loss 0.288770 LR 0.000500 Time 0.021080 +2023-10-05 21:44:08,364 - Epoch: [131][ 1230/ 1236] Overall Loss 0.288908 Objective Loss 0.288908 LR 0.000500 Time 0.021120 +2023-10-05 21:44:08,482 - Epoch: [131][ 1236/ 1236] Overall Loss 0.288912 Objective Loss 0.288912 Top1 83.503055 Top5 97.352342 LR 0.000500 Time 0.021113 +2023-10-05 21:44:08,620 - --- validate (epoch=131)----------- +2023-10-05 21:44:08,620 - 29943 samples (256 per mini-batch) +2023-10-05 21:44:09,082 - Epoch: [131][ 10/ 117] Loss 0.351537 Top1 83.320312 Top5 97.265625 +2023-10-05 21:44:09,235 - Epoch: [131][ 20/ 117] Loss 0.357457 Top1 82.441406 Top5 97.421875 +2023-10-05 21:44:09,386 - Epoch: [131][ 30/ 117] Loss 0.356645 Top1 82.330729 Top5 97.500000 +2023-10-05 21:44:09,539 - Epoch: [131][ 40/ 117] Loss 0.348900 Top1 82.470703 Top5 97.666016 +2023-10-05 21:44:09,689 - Epoch: [131][ 50/ 117] Loss 0.351376 Top1 82.281250 Top5 97.687500 +2023-10-05 21:44:09,840 - Epoch: [131][ 60/ 117] Loss 0.346881 Top1 82.343750 Top5 97.721354 +2023-10-05 21:44:09,990 - Epoch: [131][ 70/ 117] Loss 0.349778 Top1 82.271205 Top5 97.617188 +2023-10-05 21:44:10,139 - Epoch: [131][ 80/ 117] Loss 0.351876 Top1 82.153320 Top5 97.612305 +2023-10-05 21:44:10,288 - Epoch: [131][ 90/ 117] Loss 0.347983 Top1 82.339410 Top5 97.682292 +2023-10-05 21:44:10,438 - Epoch: [131][ 100/ 117] Loss 0.348953 Top1 82.207031 Top5 97.683594 +2023-10-05 21:44:10,595 - Epoch: [131][ 110/ 117] Loss 0.349193 Top1 82.233665 Top5 97.709517 +2023-10-05 21:44:10,681 - Epoch: [131][ 117/ 117] Loss 0.349873 Top1 82.272985 Top5 97.698961 +2023-10-05 21:44:10,820 - ==> Top1: 82.273 Top5: 97.699 Loss: 0.350 + +2023-10-05 21:44:10,820 - ==> Confusion: +[[ 927 1 4 2 4 0 1 0 6 77 1 1 2 2 3 4 1 0 1 1 12] + [ 3 1051 4 0 8 21 2 13 3 0 2 2 0 0 1 4 5 0 10 0 2] + [ 9 1 956 12 2 0 22 9 0 0 6 2 9 1 3 3 0 2 8 3 8] + [ 4 1 10 968 0 5 1 2 4 0 8 0 6 3 29 4 2 6 23 1 12] + [ 26 5 1 0 971 7 0 3 1 6 0 2 1 0 10 1 9 2 0 1 4] + [ 8 34 1 1 2 990 0 13 2 2 4 14 0 12 6 1 6 0 3 4 13] + [ 0 6 17 0 2 2 1123 8 0 0 3 2 3 0 1 10 0 2 1 7 4] + [ 4 23 11 1 1 42 3 1062 2 1 4 8 3 2 0 3 0 1 36 5 6] + [ 17 2 2 0 1 1 1 1 982 36 10 1 3 9 12 3 1 0 5 0 2] + [ 112 1 3 0 2 4 0 0 38 915 0 2 0 21 2 6 1 0 0 4 8] + [ 2 5 9 7 1 1 6 5 14 2 968 3 0 11 4 1 1 0 4 2 7] + [ 3 2 1 0 0 14 0 1 0 1 0 951 26 4 0 4 4 15 1 7 1] + [ 0 1 3 5 0 1 1 3 0 0 1 33 987 2 1 4 2 14 1 2 7] + [ 1 0 0 1 2 14 0 0 15 16 5 3 2 1043 2 3 3 0 0 3 6] + [ 14 1 3 5 7 1 1 0 42 2 1 1 3 2 989 0 0 0 16 0 13] + [ 0 4 1 1 3 1 2 0 0 1 0 10 8 1 0 1068 14 9 1 6 4] + [ 0 14 2 0 4 3 0 2 1 0 0 5 0 1 3 9 1106 0 0 4 7] + [ 0 0 0 1 0 0 1 0 1 0 0 0 22 0 1 5 0 1001 3 2 1] + [ 3 7 6 16 0 0 0 26 1 0 1 1 0 0 11 0 2 0 986 1 7] + [ 0 3 2 2 2 8 8 10 1 0 0 17 3 3 0 6 9 1 3 1069 5] + [ 174 255 152 76 124 188 47 110 132 83 202 124 469 317 156 69 202 89 171 243 4522]] + +2023-10-05 21:44:10,822 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:44:10,822 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:44:10,828 - + +2023-10-05 21:44:10,828 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:44:11,950 - Epoch: [132][ 10/ 1236] Overall Loss 0.299343 Objective Loss 0.299343 LR 0.000500 Time 0.112223 +2023-10-05 21:44:12,156 - Epoch: [132][ 20/ 1236] Overall Loss 0.294398 Objective Loss 0.294398 LR 0.000500 Time 0.066351 +2023-10-05 21:44:12,361 - Epoch: [132][ 30/ 1236] Overall Loss 0.288768 Objective Loss 0.288768 LR 0.000500 Time 0.051056 +2023-10-05 21:44:12,565 - Epoch: [132][ 40/ 1236] Overall Loss 0.289274 Objective Loss 0.289274 LR 0.000500 Time 0.043391 +2023-10-05 21:44:12,769 - Epoch: [132][ 50/ 1236] Overall Loss 0.292534 Objective Loss 0.292534 LR 0.000500 Time 0.038793 +2023-10-05 21:44:12,974 - Epoch: [132][ 60/ 1236] Overall Loss 0.290189 Objective Loss 0.290189 LR 0.000500 Time 0.035730 +2023-10-05 21:44:13,179 - Epoch: [132][ 70/ 1236] Overall Loss 0.287481 Objective Loss 0.287481 LR 0.000500 Time 0.033544 +2023-10-05 21:44:13,383 - Epoch: [132][ 80/ 1236] Overall Loss 0.287780 Objective Loss 0.287780 LR 0.000500 Time 0.031901 +2023-10-05 21:44:13,588 - Epoch: [132][ 90/ 1236] Overall Loss 0.287190 Objective Loss 0.287190 LR 0.000500 Time 0.030631 +2023-10-05 21:44:13,793 - Epoch: [132][ 100/ 1236] Overall Loss 0.289104 Objective Loss 0.289104 LR 0.000500 Time 0.029610 +2023-10-05 21:44:13,997 - Epoch: [132][ 110/ 1236] Overall Loss 0.289137 Objective Loss 0.289137 LR 0.000500 Time 0.028772 +2023-10-05 21:44:14,202 - Epoch: [132][ 120/ 1236] Overall Loss 0.289275 Objective Loss 0.289275 LR 0.000500 Time 0.028082 +2023-10-05 21:44:14,406 - Epoch: [132][ 130/ 1236] Overall Loss 0.292323 Objective Loss 0.292323 LR 0.000500 Time 0.027491 +2023-10-05 21:44:14,611 - Epoch: [132][ 140/ 1236] Overall Loss 0.290438 Objective Loss 0.290438 LR 0.000500 Time 0.026987 +2023-10-05 21:44:14,816 - Epoch: [132][ 150/ 1236] Overall Loss 0.289546 Objective Loss 0.289546 LR 0.000500 Time 0.026549 +2023-10-05 21:44:15,021 - Epoch: [132][ 160/ 1236] Overall Loss 0.289695 Objective Loss 0.289695 LR 0.000500 Time 0.026172 +2023-10-05 21:44:15,226 - Epoch: [132][ 170/ 1236] Overall Loss 0.288410 Objective Loss 0.288410 LR 0.000500 Time 0.025834 +2023-10-05 21:44:15,431 - Epoch: [132][ 180/ 1236] Overall Loss 0.290254 Objective Loss 0.290254 LR 0.000500 Time 0.025538 +2023-10-05 21:44:15,636 - Epoch: [132][ 190/ 1236] Overall Loss 0.289062 Objective Loss 0.289062 LR 0.000500 Time 0.025269 +2023-10-05 21:44:15,841 - Epoch: [132][ 200/ 1236] Overall Loss 0.288386 Objective Loss 0.288386 LR 0.000500 Time 0.025031 +2023-10-05 21:44:16,046 - Epoch: [132][ 210/ 1236] Overall Loss 0.288774 Objective Loss 0.288774 LR 0.000500 Time 0.024811 +2023-10-05 21:44:16,251 - Epoch: [132][ 220/ 1236] Overall Loss 0.288471 Objective Loss 0.288471 LR 0.000500 Time 0.024616 +2023-10-05 21:44:16,456 - Epoch: [132][ 230/ 1236] Overall Loss 0.288588 Objective Loss 0.288588 LR 0.000500 Time 0.024434 +2023-10-05 21:44:16,662 - Epoch: [132][ 240/ 1236] Overall Loss 0.288580 Objective Loss 0.288580 LR 0.000500 Time 0.024272 +2023-10-05 21:44:16,866 - Epoch: [132][ 250/ 1236] Overall Loss 0.288221 Objective Loss 0.288221 LR 0.000500 Time 0.024116 +2023-10-05 21:44:17,071 - Epoch: [132][ 260/ 1236] Overall Loss 0.287614 Objective Loss 0.287614 LR 0.000500 Time 0.023978 +2023-10-05 21:44:17,276 - Epoch: [132][ 270/ 1236] Overall Loss 0.287453 Objective Loss 0.287453 LR 0.000500 Time 0.023846 +2023-10-05 21:44:17,481 - Epoch: [132][ 280/ 1236] Overall Loss 0.286847 Objective Loss 0.286847 LR 0.000500 Time 0.023727 +2023-10-05 21:44:17,686 - Epoch: [132][ 290/ 1236] Overall Loss 0.285899 Objective Loss 0.285899 LR 0.000500 Time 0.023613 +2023-10-05 21:44:17,892 - Epoch: [132][ 300/ 1236] Overall Loss 0.285981 Objective Loss 0.285981 LR 0.000500 Time 0.023512 +2023-10-05 21:44:18,097 - Epoch: [132][ 310/ 1236] Overall Loss 0.285808 Objective Loss 0.285808 LR 0.000500 Time 0.023412 +2023-10-05 21:44:18,303 - Epoch: [132][ 320/ 1236] Overall Loss 0.285948 Objective Loss 0.285948 LR 0.000500 Time 0.023325 +2023-10-05 21:44:18,509 - Epoch: [132][ 330/ 1236] Overall Loss 0.285443 Objective Loss 0.285443 LR 0.000500 Time 0.023241 +2023-10-05 21:44:18,715 - Epoch: [132][ 340/ 1236] Overall Loss 0.286278 Objective Loss 0.286278 LR 0.000500 Time 0.023162 +2023-10-05 21:44:18,920 - Epoch: [132][ 350/ 1236] Overall Loss 0.285977 Objective Loss 0.285977 LR 0.000500 Time 0.023084 +2023-10-05 21:44:19,126 - Epoch: [132][ 360/ 1236] Overall Loss 0.286424 Objective Loss 0.286424 LR 0.000500 Time 0.023014 +2023-10-05 21:44:19,330 - Epoch: [132][ 370/ 1236] Overall Loss 0.286551 Objective Loss 0.286551 LR 0.000500 Time 0.022945 +2023-10-05 21:44:19,535 - Epoch: [132][ 380/ 1236] Overall Loss 0.286809 Objective Loss 0.286809 LR 0.000500 Time 0.022877 +2023-10-05 21:44:19,738 - Epoch: [132][ 390/ 1236] Overall Loss 0.287266 Objective Loss 0.287266 LR 0.000500 Time 0.022812 +2023-10-05 21:44:19,942 - Epoch: [132][ 400/ 1236] Overall Loss 0.287118 Objective Loss 0.287118 LR 0.000500 Time 0.022750 +2023-10-05 21:44:20,145 - Epoch: [132][ 410/ 1236] Overall Loss 0.286907 Objective Loss 0.286907 LR 0.000500 Time 0.022689 +2023-10-05 21:44:20,349 - Epoch: [132][ 420/ 1236] Overall Loss 0.287129 Objective Loss 0.287129 LR 0.000500 Time 0.022634 +2023-10-05 21:44:20,554 - Epoch: [132][ 430/ 1236] Overall Loss 0.287324 Objective Loss 0.287324 LR 0.000500 Time 0.022583 +2023-10-05 21:44:20,758 - Epoch: [132][ 440/ 1236] Overall Loss 0.286909 Objective Loss 0.286909 LR 0.000500 Time 0.022534 +2023-10-05 21:44:20,963 - Epoch: [132][ 450/ 1236] Overall Loss 0.286822 Objective Loss 0.286822 LR 0.000500 Time 0.022488 +2023-10-05 21:44:21,168 - Epoch: [132][ 460/ 1236] Overall Loss 0.287223 Objective Loss 0.287223 LR 0.000500 Time 0.022443 +2023-10-05 21:44:21,372 - Epoch: [132][ 470/ 1236] Overall Loss 0.287143 Objective Loss 0.287143 LR 0.000500 Time 0.022399 +2023-10-05 21:44:21,576 - Epoch: [132][ 480/ 1236] Overall Loss 0.286474 Objective Loss 0.286474 LR 0.000500 Time 0.022357 +2023-10-05 21:44:21,780 - Epoch: [132][ 490/ 1236] Overall Loss 0.286308 Objective Loss 0.286308 LR 0.000500 Time 0.022317 +2023-10-05 21:44:21,985 - Epoch: [132][ 500/ 1236] Overall Loss 0.286106 Objective Loss 0.286106 LR 0.000500 Time 0.022278 +2023-10-05 21:44:22,189 - Epoch: [132][ 510/ 1236] Overall Loss 0.286215 Objective Loss 0.286215 LR 0.000500 Time 0.022241 +2023-10-05 21:44:22,393 - Epoch: [132][ 520/ 1236] Overall Loss 0.285972 Objective Loss 0.285972 LR 0.000500 Time 0.022205 +2023-10-05 21:44:22,597 - Epoch: [132][ 530/ 1236] Overall Loss 0.286128 Objective Loss 0.286128 LR 0.000500 Time 0.022171 +2023-10-05 21:44:22,802 - Epoch: [132][ 540/ 1236] Overall Loss 0.286392 Objective Loss 0.286392 LR 0.000500 Time 0.022139 +2023-10-05 21:44:23,006 - Epoch: [132][ 550/ 1236] Overall Loss 0.286207 Objective Loss 0.286207 LR 0.000500 Time 0.022106 +2023-10-05 21:44:23,210 - Epoch: [132][ 560/ 1236] Overall Loss 0.286720 Objective Loss 0.286720 LR 0.000500 Time 0.022076 +2023-10-05 21:44:23,414 - Epoch: [132][ 570/ 1236] Overall Loss 0.286723 Objective Loss 0.286723 LR 0.000500 Time 0.022046 +2023-10-05 21:44:23,619 - Epoch: [132][ 580/ 1236] Overall Loss 0.286892 Objective Loss 0.286892 LR 0.000500 Time 0.022018 +2023-10-05 21:44:23,823 - Epoch: [132][ 590/ 1236] Overall Loss 0.287403 Objective Loss 0.287403 LR 0.000500 Time 0.021990 +2023-10-05 21:44:24,028 - Epoch: [132][ 600/ 1236] Overall Loss 0.287288 Objective Loss 0.287288 LR 0.000500 Time 0.021964 +2023-10-05 21:44:24,232 - Epoch: [132][ 610/ 1236] Overall Loss 0.287013 Objective Loss 0.287013 LR 0.000500 Time 0.021938 +2023-10-05 21:44:24,436 - Epoch: [132][ 620/ 1236] Overall Loss 0.287266 Objective Loss 0.287266 LR 0.000500 Time 0.021913 +2023-10-05 21:44:24,640 - Epoch: [132][ 630/ 1236] Overall Loss 0.287441 Objective Loss 0.287441 LR 0.000500 Time 0.021889 +2023-10-05 21:44:24,845 - Epoch: [132][ 640/ 1236] Overall Loss 0.287413 Objective Loss 0.287413 LR 0.000500 Time 0.021866 +2023-10-05 21:44:25,049 - Epoch: [132][ 650/ 1236] Overall Loss 0.287064 Objective Loss 0.287064 LR 0.000500 Time 0.021843 +2023-10-05 21:44:25,253 - Epoch: [132][ 660/ 1236] Overall Loss 0.287267 Objective Loss 0.287267 LR 0.000500 Time 0.021820 +2023-10-05 21:44:25,456 - Epoch: [132][ 670/ 1236] Overall Loss 0.287196 Objective Loss 0.287196 LR 0.000500 Time 0.021798 +2023-10-05 21:44:25,662 - Epoch: [132][ 680/ 1236] Overall Loss 0.287346 Objective Loss 0.287346 LR 0.000500 Time 0.021779 +2023-10-05 21:44:25,864 - Epoch: [132][ 690/ 1236] Overall Loss 0.287333 Objective Loss 0.287333 LR 0.000500 Time 0.021756 +2023-10-05 21:44:26,069 - Epoch: [132][ 700/ 1236] Overall Loss 0.287229 Objective Loss 0.287229 LR 0.000500 Time 0.021737 +2023-10-05 21:44:26,272 - Epoch: [132][ 710/ 1236] Overall Loss 0.287163 Objective Loss 0.287163 LR 0.000500 Time 0.021716 +2023-10-05 21:44:26,476 - Epoch: [132][ 720/ 1236] Overall Loss 0.287469 Objective Loss 0.287469 LR 0.000500 Time 0.021698 +2023-10-05 21:44:26,679 - Epoch: [132][ 730/ 1236] Overall Loss 0.287493 Objective Loss 0.287493 LR 0.000500 Time 0.021678 +2023-10-05 21:44:26,883 - Epoch: [132][ 740/ 1236] Overall Loss 0.287763 Objective Loss 0.287763 LR 0.000500 Time 0.021661 +2023-10-05 21:44:27,086 - Epoch: [132][ 750/ 1236] Overall Loss 0.287644 Objective Loss 0.287644 LR 0.000500 Time 0.021642 +2023-10-05 21:44:27,290 - Epoch: [132][ 760/ 1236] Overall Loss 0.287498 Objective Loss 0.287498 LR 0.000500 Time 0.021626 +2023-10-05 21:44:27,493 - Epoch: [132][ 770/ 1236] Overall Loss 0.287897 Objective Loss 0.287897 LR 0.000500 Time 0.021608 +2023-10-05 21:44:27,698 - Epoch: [132][ 780/ 1236] Overall Loss 0.287763 Objective Loss 0.287763 LR 0.000500 Time 0.021592 +2023-10-05 21:44:27,900 - Epoch: [132][ 790/ 1236] Overall Loss 0.287401 Objective Loss 0.287401 LR 0.000500 Time 0.021575 +2023-10-05 21:44:28,105 - Epoch: [132][ 800/ 1236] Overall Loss 0.287124 Objective Loss 0.287124 LR 0.000500 Time 0.021561 +2023-10-05 21:44:28,308 - Epoch: [132][ 810/ 1236] Overall Loss 0.287466 Objective Loss 0.287466 LR 0.000500 Time 0.021545 +2023-10-05 21:44:28,513 - Epoch: [132][ 820/ 1236] Overall Loss 0.287481 Objective Loss 0.287481 LR 0.000500 Time 0.021531 +2023-10-05 21:44:28,715 - Epoch: [132][ 830/ 1236] Overall Loss 0.287500 Objective Loss 0.287500 LR 0.000500 Time 0.021515 +2023-10-05 21:44:28,920 - Epoch: [132][ 840/ 1236] Overall Loss 0.287376 Objective Loss 0.287376 LR 0.000500 Time 0.021502 +2023-10-05 21:44:29,122 - Epoch: [132][ 850/ 1236] Overall Loss 0.287167 Objective Loss 0.287167 LR 0.000500 Time 0.021487 +2023-10-05 21:44:29,327 - Epoch: [132][ 860/ 1236] Overall Loss 0.286801 Objective Loss 0.286801 LR 0.000500 Time 0.021475 +2023-10-05 21:44:29,529 - Epoch: [132][ 870/ 1236] Overall Loss 0.286958 Objective Loss 0.286958 LR 0.000500 Time 0.021460 +2023-10-05 21:44:29,734 - Epoch: [132][ 880/ 1236] Overall Loss 0.287001 Objective Loss 0.287001 LR 0.000500 Time 0.021448 +2023-10-05 21:44:29,936 - Epoch: [132][ 890/ 1236] Overall Loss 0.287040 Objective Loss 0.287040 LR 0.000500 Time 0.021434 +2023-10-05 21:44:30,141 - Epoch: [132][ 900/ 1236] Overall Loss 0.286959 Objective Loss 0.286959 LR 0.000500 Time 0.021423 +2023-10-05 21:44:30,343 - Epoch: [132][ 910/ 1236] Overall Loss 0.286644 Objective Loss 0.286644 LR 0.000500 Time 0.021410 +2023-10-05 21:44:30,548 - Epoch: [132][ 920/ 1236] Overall Loss 0.286689 Objective Loss 0.286689 LR 0.000500 Time 0.021399 +2023-10-05 21:44:30,750 - Epoch: [132][ 930/ 1236] Overall Loss 0.286537 Objective Loss 0.286537 LR 0.000500 Time 0.021386 +2023-10-05 21:44:30,955 - Epoch: [132][ 940/ 1236] Overall Loss 0.286324 Objective Loss 0.286324 LR 0.000500 Time 0.021376 +2023-10-05 21:44:31,157 - Epoch: [132][ 950/ 1236] Overall Loss 0.286541 Objective Loss 0.286541 LR 0.000500 Time 0.021364 +2023-10-05 21:44:31,362 - Epoch: [132][ 960/ 1236] Overall Loss 0.286398 Objective Loss 0.286398 LR 0.000500 Time 0.021354 +2023-10-05 21:44:31,565 - Epoch: [132][ 970/ 1236] Overall Loss 0.286360 Objective Loss 0.286360 LR 0.000500 Time 0.021342 +2023-10-05 21:44:31,769 - Epoch: [132][ 980/ 1236] Overall Loss 0.286509 Objective Loss 0.286509 LR 0.000500 Time 0.021333 +2023-10-05 21:44:31,972 - Epoch: [132][ 990/ 1236] Overall Loss 0.286434 Objective Loss 0.286434 LR 0.000500 Time 0.021322 +2023-10-05 21:44:32,176 - Epoch: [132][ 1000/ 1236] Overall Loss 0.286365 Objective Loss 0.286365 LR 0.000500 Time 0.021313 +2023-10-05 21:44:32,379 - Epoch: [132][ 1010/ 1236] Overall Loss 0.286410 Objective Loss 0.286410 LR 0.000500 Time 0.021302 +2023-10-05 21:44:32,584 - Epoch: [132][ 1020/ 1236] Overall Loss 0.286454 Objective Loss 0.286454 LR 0.000500 Time 0.021294 +2023-10-05 21:44:32,786 - Epoch: [132][ 1030/ 1236] Overall Loss 0.286633 Objective Loss 0.286633 LR 0.000500 Time 0.021283 +2023-10-05 21:44:32,991 - Epoch: [132][ 1040/ 1236] Overall Loss 0.286479 Objective Loss 0.286479 LR 0.000500 Time 0.021275 +2023-10-05 21:44:33,193 - Epoch: [132][ 1050/ 1236] Overall Loss 0.286454 Objective Loss 0.286454 LR 0.000500 Time 0.021265 +2023-10-05 21:44:33,398 - Epoch: [132][ 1060/ 1236] Overall Loss 0.286410 Objective Loss 0.286410 LR 0.000500 Time 0.021257 +2023-10-05 21:44:33,600 - Epoch: [132][ 1070/ 1236] Overall Loss 0.286610 Objective Loss 0.286610 LR 0.000500 Time 0.021247 +2023-10-05 21:44:33,804 - Epoch: [132][ 1080/ 1236] Overall Loss 0.286784 Objective Loss 0.286784 LR 0.000500 Time 0.021239 +2023-10-05 21:44:34,007 - Epoch: [132][ 1090/ 1236] Overall Loss 0.286720 Objective Loss 0.286720 LR 0.000500 Time 0.021230 +2023-10-05 21:44:34,212 - Epoch: [132][ 1100/ 1236] Overall Loss 0.286667 Objective Loss 0.286667 LR 0.000500 Time 0.021223 +2023-10-05 21:44:34,415 - Epoch: [132][ 1110/ 1236] Overall Loss 0.286672 Objective Loss 0.286672 LR 0.000500 Time 0.021214 +2023-10-05 21:44:34,619 - Epoch: [132][ 1120/ 1236] Overall Loss 0.286705 Objective Loss 0.286705 LR 0.000500 Time 0.021206 +2023-10-05 21:44:34,822 - Epoch: [132][ 1130/ 1236] Overall Loss 0.286749 Objective Loss 0.286749 LR 0.000500 Time 0.021198 +2023-10-05 21:44:35,026 - Epoch: [132][ 1140/ 1236] Overall Loss 0.286634 Objective Loss 0.286634 LR 0.000500 Time 0.021191 +2023-10-05 21:44:35,229 - Epoch: [132][ 1150/ 1236] Overall Loss 0.286550 Objective Loss 0.286550 LR 0.000500 Time 0.021183 +2023-10-05 21:44:35,433 - Epoch: [132][ 1160/ 1236] Overall Loss 0.286583 Objective Loss 0.286583 LR 0.000500 Time 0.021176 +2023-10-05 21:44:35,636 - Epoch: [132][ 1170/ 1236] Overall Loss 0.286788 Objective Loss 0.286788 LR 0.000500 Time 0.021168 +2023-10-05 21:44:35,841 - Epoch: [132][ 1180/ 1236] Overall Loss 0.286735 Objective Loss 0.286735 LR 0.000500 Time 0.021162 +2023-10-05 21:44:36,044 - Epoch: [132][ 1190/ 1236] Overall Loss 0.286644 Objective Loss 0.286644 LR 0.000500 Time 0.021154 +2023-10-05 21:44:36,248 - Epoch: [132][ 1200/ 1236] Overall Loss 0.286665 Objective Loss 0.286665 LR 0.000500 Time 0.021148 +2023-10-05 21:44:36,451 - Epoch: [132][ 1210/ 1236] Overall Loss 0.286581 Objective Loss 0.286581 LR 0.000500 Time 0.021141 +2023-10-05 21:44:36,656 - Epoch: [132][ 1220/ 1236] Overall Loss 0.286427 Objective Loss 0.286427 LR 0.000500 Time 0.021135 +2023-10-05 21:44:36,915 - Epoch: [132][ 1230/ 1236] Overall Loss 0.286294 Objective Loss 0.286294 LR 0.000500 Time 0.021173 +2023-10-05 21:44:37,033 - Epoch: [132][ 1236/ 1236] Overall Loss 0.286353 Objective Loss 0.286353 Top1 83.910387 Top5 97.352342 LR 0.000500 Time 0.021166 +2023-10-05 21:44:37,161 - --- validate (epoch=132)----------- +2023-10-05 21:44:37,162 - 29943 samples (256 per mini-batch) +2023-10-05 21:44:37,629 - Epoch: [132][ 10/ 117] Loss 0.370378 Top1 82.343750 Top5 97.148438 +2023-10-05 21:44:37,788 - Epoch: [132][ 20/ 117] Loss 0.361909 Top1 82.617188 Top5 97.285156 +2023-10-05 21:44:37,944 - Epoch: [132][ 30/ 117] Loss 0.347665 Top1 83.111979 Top5 97.578125 +2023-10-05 21:44:38,104 - Epoch: [132][ 40/ 117] Loss 0.349440 Top1 83.076172 Top5 97.548828 +2023-10-05 21:44:38,260 - Epoch: [132][ 50/ 117] Loss 0.340964 Top1 83.164062 Top5 97.632812 +2023-10-05 21:44:38,417 - Epoch: [132][ 60/ 117] Loss 0.338423 Top1 83.131510 Top5 97.792969 +2023-10-05 21:44:38,574 - Epoch: [132][ 70/ 117] Loss 0.335577 Top1 83.253348 Top5 97.806920 +2023-10-05 21:44:38,734 - Epoch: [132][ 80/ 117] Loss 0.334232 Top1 83.295898 Top5 97.812500 +2023-10-05 21:44:38,891 - Epoch: [132][ 90/ 117] Loss 0.334553 Top1 83.294271 Top5 97.834201 +2023-10-05 21:44:39,050 - Epoch: [132][ 100/ 117] Loss 0.334961 Top1 83.292969 Top5 97.796875 +2023-10-05 21:44:39,217 - Epoch: [132][ 110/ 117] Loss 0.337795 Top1 83.235085 Top5 97.837358 +2023-10-05 21:44:39,303 - Epoch: [132][ 117/ 117] Loss 0.339003 Top1 83.234813 Top5 97.802491 +2023-10-05 21:44:39,447 - ==> Top1: 83.235 Top5: 97.802 Loss: 0.339 + +2023-10-05 21:44:39,448 - ==> Confusion: +[[ 943 1 3 0 8 1 0 1 4 52 1 1 2 4 8 4 2 0 1 0 14] + [ 2 1037 5 0 10 24 1 23 2 0 1 1 0 0 0 5 2 0 15 1 2] + [ 2 0 954 14 2 0 29 8 0 1 6 3 10 1 1 4 1 2 4 5 9] + [ 0 0 13 964 0 5 2 1 2 0 8 0 8 4 23 5 1 10 25 3 15] + [ 28 5 1 0 974 4 0 2 0 2 1 1 0 0 9 6 10 1 0 1 5] + [ 5 31 0 1 3 988 1 25 1 0 4 11 2 13 7 1 3 0 4 5 11] + [ 0 5 22 0 0 1 1128 6 0 0 1 2 3 0 1 10 0 0 1 7 4] + [ 5 8 14 1 1 31 5 1069 1 1 3 11 1 3 0 3 0 0 44 9 8] + [ 22 2 0 0 0 2 0 1 966 36 12 1 0 10 17 3 5 1 8 1 2] + [ 118 0 3 0 10 5 1 0 30 899 0 0 0 21 5 6 3 2 0 6 10] + [ 2 4 12 7 1 1 2 6 11 1 964 4 0 12 3 1 2 0 4 4 12] + [ 2 1 1 0 0 15 0 1 0 0 0 953 28 4 0 2 2 15 0 5 6] + [ 0 0 3 3 1 1 0 1 0 0 1 30 989 2 2 6 2 13 2 2 10] + [ 2 0 1 0 2 12 0 0 9 7 5 4 2 1056 3 2 2 1 0 3 8] + [ 16 1 3 13 6 0 0 0 19 0 0 2 2 3 1006 0 2 1 16 0 11] + [ 1 3 2 1 3 0 1 0 0 1 0 8 8 1 0 1073 13 7 1 8 3] + [ 2 12 1 0 5 5 0 1 1 0 0 3 1 1 2 9 1103 0 0 5 10] + [ 0 0 0 2 0 0 1 0 0 0 0 1 22 0 2 5 0 999 1 3 2] + [ 1 2 5 19 1 0 0 23 2 0 2 0 4 0 6 0 1 0 992 1 9] + [ 0 4 3 1 2 5 8 8 0 0 3 15 2 2 0 8 11 1 4 1066 9] + [ 136 204 189 74 106 149 58 98 85 60 191 134 420 317 175 73 162 73 194 207 4800]] + +2023-10-05 21:44:39,449 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:44:39,449 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:44:39,455 - + +2023-10-05 21:44:39,456 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:44:40,474 - Epoch: [133][ 10/ 1236] Overall Loss 0.290605 Objective Loss 0.290605 LR 0.000500 Time 0.101775 +2023-10-05 21:44:40,678 - Epoch: [133][ 20/ 1236] Overall Loss 0.289522 Objective Loss 0.289522 LR 0.000500 Time 0.061092 +2023-10-05 21:44:40,880 - Epoch: [133][ 30/ 1236] Overall Loss 0.275499 Objective Loss 0.275499 LR 0.000500 Time 0.047448 +2023-10-05 21:44:41,084 - Epoch: [133][ 40/ 1236] Overall Loss 0.274162 Objective Loss 0.274162 LR 0.000500 Time 0.040677 +2023-10-05 21:44:41,286 - Epoch: [133][ 50/ 1236] Overall Loss 0.277514 Objective Loss 0.277514 LR 0.000500 Time 0.036579 +2023-10-05 21:44:41,490 - Epoch: [133][ 60/ 1236] Overall Loss 0.282032 Objective Loss 0.282032 LR 0.000500 Time 0.033875 +2023-10-05 21:44:41,693 - Epoch: [133][ 70/ 1236] Overall Loss 0.279824 Objective Loss 0.279824 LR 0.000500 Time 0.031932 +2023-10-05 21:44:41,897 - Epoch: [133][ 80/ 1236] Overall Loss 0.279187 Objective Loss 0.279187 LR 0.000500 Time 0.030483 +2023-10-05 21:44:42,101 - Epoch: [133][ 90/ 1236] Overall Loss 0.282242 Objective Loss 0.282242 LR 0.000500 Time 0.029360 +2023-10-05 21:44:42,306 - Epoch: [133][ 100/ 1236] Overall Loss 0.279356 Objective Loss 0.279356 LR 0.000500 Time 0.028472 +2023-10-05 21:44:42,511 - Epoch: [133][ 110/ 1236] Overall Loss 0.278609 Objective Loss 0.278609 LR 0.000500 Time 0.027743 +2023-10-05 21:44:42,714 - Epoch: [133][ 120/ 1236] Overall Loss 0.277846 Objective Loss 0.277846 LR 0.000500 Time 0.027116 +2023-10-05 21:44:42,916 - Epoch: [133][ 130/ 1236] Overall Loss 0.276351 Objective Loss 0.276351 LR 0.000500 Time 0.026584 +2023-10-05 21:44:43,120 - Epoch: [133][ 140/ 1236] Overall Loss 0.275651 Objective Loss 0.275651 LR 0.000500 Time 0.026138 +2023-10-05 21:44:43,323 - Epoch: [133][ 150/ 1236] Overall Loss 0.276077 Objective Loss 0.276077 LR 0.000500 Time 0.025750 +2023-10-05 21:44:43,528 - Epoch: [133][ 160/ 1236] Overall Loss 0.275610 Objective Loss 0.275610 LR 0.000500 Time 0.025417 +2023-10-05 21:44:43,732 - Epoch: [133][ 170/ 1236] Overall Loss 0.275099 Objective Loss 0.275099 LR 0.000500 Time 0.025122 +2023-10-05 21:44:43,936 - Epoch: [133][ 180/ 1236] Overall Loss 0.273422 Objective Loss 0.273422 LR 0.000500 Time 0.024860 +2023-10-05 21:44:44,141 - Epoch: [133][ 190/ 1236] Overall Loss 0.272471 Objective Loss 0.272471 LR 0.000500 Time 0.024623 +2023-10-05 21:44:44,345 - Epoch: [133][ 200/ 1236] Overall Loss 0.272208 Objective Loss 0.272208 LR 0.000500 Time 0.024413 +2023-10-05 21:44:44,549 - Epoch: [133][ 210/ 1236] Overall Loss 0.273310 Objective Loss 0.273310 LR 0.000500 Time 0.024221 +2023-10-05 21:44:44,754 - Epoch: [133][ 220/ 1236] Overall Loss 0.272648 Objective Loss 0.272648 LR 0.000500 Time 0.024050 +2023-10-05 21:44:44,956 - Epoch: [133][ 230/ 1236] Overall Loss 0.273462 Objective Loss 0.273462 LR 0.000500 Time 0.023881 +2023-10-05 21:44:45,160 - Epoch: [133][ 240/ 1236] Overall Loss 0.272902 Objective Loss 0.272902 LR 0.000500 Time 0.023732 +2023-10-05 21:44:45,361 - Epoch: [133][ 250/ 1236] Overall Loss 0.273949 Objective Loss 0.273949 LR 0.000500 Time 0.023589 +2023-10-05 21:44:45,566 - Epoch: [133][ 260/ 1236] Overall Loss 0.273562 Objective Loss 0.273562 LR 0.000500 Time 0.023466 +2023-10-05 21:44:45,766 - Epoch: [133][ 270/ 1236] Overall Loss 0.274712 Objective Loss 0.274712 LR 0.000500 Time 0.023339 +2023-10-05 21:44:45,969 - Epoch: [133][ 280/ 1236] Overall Loss 0.274248 Objective Loss 0.274248 LR 0.000500 Time 0.023229 +2023-10-05 21:44:46,172 - Epoch: [133][ 290/ 1236] Overall Loss 0.275466 Objective Loss 0.275466 LR 0.000500 Time 0.023127 +2023-10-05 21:44:46,378 - Epoch: [133][ 300/ 1236] Overall Loss 0.276005 Objective Loss 0.276005 LR 0.000500 Time 0.023041 +2023-10-05 21:44:46,583 - Epoch: [133][ 310/ 1236] Overall Loss 0.275909 Objective Loss 0.275909 LR 0.000500 Time 0.022959 +2023-10-05 21:44:46,790 - Epoch: [133][ 320/ 1236] Overall Loss 0.276337 Objective Loss 0.276337 LR 0.000500 Time 0.022885 +2023-10-05 21:44:46,994 - Epoch: [133][ 330/ 1236] Overall Loss 0.275834 Objective Loss 0.275834 LR 0.000500 Time 0.022811 +2023-10-05 21:44:47,201 - Epoch: [133][ 340/ 1236] Overall Loss 0.275404 Objective Loss 0.275404 LR 0.000500 Time 0.022746 +2023-10-05 21:44:47,405 - Epoch: [133][ 350/ 1236] Overall Loss 0.275191 Objective Loss 0.275191 LR 0.000500 Time 0.022680 +2023-10-05 21:44:47,612 - Epoch: [133][ 360/ 1236] Overall Loss 0.275384 Objective Loss 0.275384 LR 0.000500 Time 0.022622 +2023-10-05 21:44:47,817 - Epoch: [133][ 370/ 1236] Overall Loss 0.275979 Objective Loss 0.275979 LR 0.000500 Time 0.022565 +2023-10-05 21:44:48,023 - Epoch: [133][ 380/ 1236] Overall Loss 0.275803 Objective Loss 0.275803 LR 0.000500 Time 0.022511 +2023-10-05 21:44:48,226 - Epoch: [133][ 390/ 1236] Overall Loss 0.275625 Objective Loss 0.275625 LR 0.000500 Time 0.022454 +2023-10-05 21:44:48,431 - Epoch: [133][ 400/ 1236] Overall Loss 0.275973 Objective Loss 0.275973 LR 0.000500 Time 0.022404 +2023-10-05 21:44:48,634 - Epoch: [133][ 410/ 1236] Overall Loss 0.276732 Objective Loss 0.276732 LR 0.000500 Time 0.022351 +2023-10-05 21:44:48,839 - Epoch: [133][ 420/ 1236] Overall Loss 0.277125 Objective Loss 0.277125 LR 0.000500 Time 0.022306 +2023-10-05 21:44:49,041 - Epoch: [133][ 430/ 1236] Overall Loss 0.277302 Objective Loss 0.277302 LR 0.000500 Time 0.022257 +2023-10-05 21:44:49,245 - Epoch: [133][ 440/ 1236] Overall Loss 0.277415 Objective Loss 0.277415 LR 0.000500 Time 0.022214 +2023-10-05 21:44:49,447 - Epoch: [133][ 450/ 1236] Overall Loss 0.277827 Objective Loss 0.277827 LR 0.000500 Time 0.022168 +2023-10-05 21:44:49,651 - Epoch: [133][ 460/ 1236] Overall Loss 0.277611 Objective Loss 0.277611 LR 0.000500 Time 0.022130 +2023-10-05 21:44:49,853 - Epoch: [133][ 470/ 1236] Overall Loss 0.277621 Objective Loss 0.277621 LR 0.000500 Time 0.022088 +2023-10-05 21:44:50,057 - Epoch: [133][ 480/ 1236] Overall Loss 0.277126 Objective Loss 0.277126 LR 0.000500 Time 0.022053 +2023-10-05 21:44:50,259 - Epoch: [133][ 490/ 1236] Overall Loss 0.276908 Objective Loss 0.276908 LR 0.000500 Time 0.022014 +2023-10-05 21:44:50,463 - Epoch: [133][ 500/ 1236] Overall Loss 0.276912 Objective Loss 0.276912 LR 0.000500 Time 0.021981 +2023-10-05 21:44:50,665 - Epoch: [133][ 510/ 1236] Overall Loss 0.276841 Objective Loss 0.276841 LR 0.000500 Time 0.021944 +2023-10-05 21:44:50,869 - Epoch: [133][ 520/ 1236] Overall Loss 0.276393 Objective Loss 0.276393 LR 0.000500 Time 0.021914 +2023-10-05 21:44:51,070 - Epoch: [133][ 530/ 1236] Overall Loss 0.276847 Objective Loss 0.276847 LR 0.000500 Time 0.021880 +2023-10-05 21:44:51,275 - Epoch: [133][ 540/ 1236] Overall Loss 0.276878 Objective Loss 0.276878 LR 0.000500 Time 0.021853 +2023-10-05 21:44:51,476 - Epoch: [133][ 550/ 1236] Overall Loss 0.276844 Objective Loss 0.276844 LR 0.000500 Time 0.021821 +2023-10-05 21:44:51,680 - Epoch: [133][ 560/ 1236] Overall Loss 0.276480 Objective Loss 0.276480 LR 0.000500 Time 0.021796 +2023-10-05 21:44:51,882 - Epoch: [133][ 570/ 1236] Overall Loss 0.276466 Objective Loss 0.276466 LR 0.000500 Time 0.021767 +2023-10-05 21:44:52,087 - Epoch: [133][ 580/ 1236] Overall Loss 0.276618 Objective Loss 0.276618 LR 0.000500 Time 0.021743 +2023-10-05 21:44:52,288 - Epoch: [133][ 590/ 1236] Overall Loss 0.276970 Objective Loss 0.276970 LR 0.000500 Time 0.021716 +2023-10-05 21:44:52,492 - Epoch: [133][ 600/ 1236] Overall Loss 0.276653 Objective Loss 0.276653 LR 0.000500 Time 0.021694 +2023-10-05 21:44:52,694 - Epoch: [133][ 610/ 1236] Overall Loss 0.276889 Objective Loss 0.276889 LR 0.000500 Time 0.021668 +2023-10-05 21:44:52,898 - Epoch: [133][ 620/ 1236] Overall Loss 0.277223 Objective Loss 0.277223 LR 0.000500 Time 0.021647 +2023-10-05 21:44:53,100 - Epoch: [133][ 630/ 1236] Overall Loss 0.277281 Objective Loss 0.277281 LR 0.000500 Time 0.021624 +2023-10-05 21:44:53,304 - Epoch: [133][ 640/ 1236] Overall Loss 0.277243 Objective Loss 0.277243 LR 0.000500 Time 0.021604 +2023-10-05 21:44:53,506 - Epoch: [133][ 650/ 1236] Overall Loss 0.277594 Objective Loss 0.277594 LR 0.000500 Time 0.021582 +2023-10-05 21:44:53,711 - Epoch: [133][ 660/ 1236] Overall Loss 0.277975 Objective Loss 0.277975 LR 0.000500 Time 0.021564 +2023-10-05 21:44:53,912 - Epoch: [133][ 670/ 1236] Overall Loss 0.277819 Objective Loss 0.277819 LR 0.000500 Time 0.021543 +2023-10-05 21:44:54,116 - Epoch: [133][ 680/ 1236] Overall Loss 0.277505 Objective Loss 0.277505 LR 0.000500 Time 0.021526 +2023-10-05 21:44:54,318 - Epoch: [133][ 690/ 1236] Overall Loss 0.277485 Objective Loss 0.277485 LR 0.000500 Time 0.021505 +2023-10-05 21:44:54,522 - Epoch: [133][ 700/ 1236] Overall Loss 0.277432 Objective Loss 0.277432 LR 0.000500 Time 0.021489 +2023-10-05 21:44:54,724 - Epoch: [133][ 710/ 1236] Overall Loss 0.277183 Objective Loss 0.277183 LR 0.000500 Time 0.021470 +2023-10-05 21:44:54,928 - Epoch: [133][ 720/ 1236] Overall Loss 0.277611 Objective Loss 0.277611 LR 0.000500 Time 0.021455 +2023-10-05 21:44:55,130 - Epoch: [133][ 730/ 1236] Overall Loss 0.277582 Objective Loss 0.277582 LR 0.000500 Time 0.021437 +2023-10-05 21:44:55,334 - Epoch: [133][ 740/ 1236] Overall Loss 0.277442 Objective Loss 0.277442 LR 0.000500 Time 0.021423 +2023-10-05 21:44:55,536 - Epoch: [133][ 750/ 1236] Overall Loss 0.277344 Objective Loss 0.277344 LR 0.000500 Time 0.021405 +2023-10-05 21:44:55,740 - Epoch: [133][ 760/ 1236] Overall Loss 0.277508 Objective Loss 0.277508 LR 0.000500 Time 0.021392 +2023-10-05 21:44:55,942 - Epoch: [133][ 770/ 1236] Overall Loss 0.277618 Objective Loss 0.277618 LR 0.000500 Time 0.021376 +2023-10-05 21:44:56,146 - Epoch: [133][ 780/ 1236] Overall Loss 0.277701 Objective Loss 0.277701 LR 0.000500 Time 0.021363 +2023-10-05 21:44:56,348 - Epoch: [133][ 790/ 1236] Overall Loss 0.277730 Objective Loss 0.277730 LR 0.000500 Time 0.021348 +2023-10-05 21:44:56,552 - Epoch: [133][ 800/ 1236] Overall Loss 0.278018 Objective Loss 0.278018 LR 0.000500 Time 0.021336 +2023-10-05 21:44:56,753 - Epoch: [133][ 810/ 1236] Overall Loss 0.277996 Objective Loss 0.277996 LR 0.000500 Time 0.021320 +2023-10-05 21:44:56,956 - Epoch: [133][ 820/ 1236] Overall Loss 0.277859 Objective Loss 0.277859 LR 0.000500 Time 0.021308 +2023-10-05 21:44:57,158 - Epoch: [133][ 830/ 1236] Overall Loss 0.277760 Objective Loss 0.277760 LR 0.000500 Time 0.021294 +2023-10-05 21:44:57,362 - Epoch: [133][ 840/ 1236] Overall Loss 0.277939 Objective Loss 0.277939 LR 0.000500 Time 0.021283 +2023-10-05 21:44:57,564 - Epoch: [133][ 850/ 1236] Overall Loss 0.277847 Objective Loss 0.277847 LR 0.000500 Time 0.021270 +2023-10-05 21:44:57,769 - Epoch: [133][ 860/ 1236] Overall Loss 0.278099 Objective Loss 0.278099 LR 0.000500 Time 0.021260 +2023-10-05 21:44:57,971 - Epoch: [133][ 870/ 1236] Overall Loss 0.277836 Objective Loss 0.277836 LR 0.000500 Time 0.021247 +2023-10-05 21:44:58,175 - Epoch: [133][ 880/ 1236] Overall Loss 0.278014 Objective Loss 0.278014 LR 0.000500 Time 0.021237 +2023-10-05 21:44:58,377 - Epoch: [133][ 890/ 1236] Overall Loss 0.277680 Objective Loss 0.277680 LR 0.000500 Time 0.021225 +2023-10-05 21:44:58,581 - Epoch: [133][ 900/ 1236] Overall Loss 0.277572 Objective Loss 0.277572 LR 0.000500 Time 0.021215 +2023-10-05 21:44:58,783 - Epoch: [133][ 910/ 1236] Overall Loss 0.277865 Objective Loss 0.277865 LR 0.000500 Time 0.021204 +2023-10-05 21:44:58,987 - Epoch: [133][ 920/ 1236] Overall Loss 0.277756 Objective Loss 0.277756 LR 0.000500 Time 0.021195 +2023-10-05 21:44:59,189 - Epoch: [133][ 930/ 1236] Overall Loss 0.277728 Objective Loss 0.277728 LR 0.000500 Time 0.021184 +2023-10-05 21:44:59,393 - Epoch: [133][ 940/ 1236] Overall Loss 0.277554 Objective Loss 0.277554 LR 0.000500 Time 0.021175 +2023-10-05 21:44:59,595 - Epoch: [133][ 950/ 1236] Overall Loss 0.277511 Objective Loss 0.277511 LR 0.000500 Time 0.021164 +2023-10-05 21:44:59,799 - Epoch: [133][ 960/ 1236] Overall Loss 0.277886 Objective Loss 0.277886 LR 0.000500 Time 0.021156 +2023-10-05 21:45:00,001 - Epoch: [133][ 970/ 1236] Overall Loss 0.278009 Objective Loss 0.278009 LR 0.000500 Time 0.021146 +2023-10-05 21:45:00,205 - Epoch: [133][ 980/ 1236] Overall Loss 0.278032 Objective Loss 0.278032 LR 0.000500 Time 0.021138 +2023-10-05 21:45:00,407 - Epoch: [133][ 990/ 1236] Overall Loss 0.277804 Objective Loss 0.277804 LR 0.000500 Time 0.021128 +2023-10-05 21:45:00,611 - Epoch: [133][ 1000/ 1236] Overall Loss 0.277539 Objective Loss 0.277539 LR 0.000500 Time 0.021121 +2023-10-05 21:45:00,822 - Epoch: [133][ 1010/ 1236] Overall Loss 0.277292 Objective Loss 0.277292 LR 0.000500 Time 0.021120 +2023-10-05 21:45:01,037 - Epoch: [133][ 1020/ 1236] Overall Loss 0.277214 Objective Loss 0.277214 LR 0.000500 Time 0.021124 +2023-10-05 21:45:01,247 - Epoch: [133][ 1030/ 1236] Overall Loss 0.277429 Objective Loss 0.277429 LR 0.000500 Time 0.021121 +2023-10-05 21:45:01,461 - Epoch: [133][ 1040/ 1236] Overall Loss 0.277322 Objective Loss 0.277322 LR 0.000500 Time 0.021124 +2023-10-05 21:45:01,671 - Epoch: [133][ 1050/ 1236] Overall Loss 0.277337 Objective Loss 0.277337 LR 0.000500 Time 0.021122 +2023-10-05 21:45:01,875 - Epoch: [133][ 1060/ 1236] Overall Loss 0.277123 Objective Loss 0.277123 LR 0.000500 Time 0.021116 +2023-10-05 21:45:02,076 - Epoch: [133][ 1070/ 1236] Overall Loss 0.277113 Objective Loss 0.277113 LR 0.000500 Time 0.021106 +2023-10-05 21:45:02,280 - Epoch: [133][ 1080/ 1236] Overall Loss 0.277100 Objective Loss 0.277100 LR 0.000500 Time 0.021099 +2023-10-05 21:45:02,482 - Epoch: [133][ 1090/ 1236] Overall Loss 0.276829 Objective Loss 0.276829 LR 0.000500 Time 0.021090 +2023-10-05 21:45:02,686 - Epoch: [133][ 1100/ 1236] Overall Loss 0.276933 Objective Loss 0.276933 LR 0.000500 Time 0.021083 +2023-10-05 21:45:02,888 - Epoch: [133][ 1110/ 1236] Overall Loss 0.276943 Objective Loss 0.276943 LR 0.000500 Time 0.021075 +2023-10-05 21:45:03,092 - Epoch: [133][ 1120/ 1236] Overall Loss 0.276821 Objective Loss 0.276821 LR 0.000500 Time 0.021069 +2023-10-05 21:45:03,293 - Epoch: [133][ 1130/ 1236] Overall Loss 0.276874 Objective Loss 0.276874 LR 0.000500 Time 0.021060 +2023-10-05 21:45:03,497 - Epoch: [133][ 1140/ 1236] Overall Loss 0.276929 Objective Loss 0.276929 LR 0.000500 Time 0.021054 +2023-10-05 21:45:03,699 - Epoch: [133][ 1150/ 1236] Overall Loss 0.276691 Objective Loss 0.276691 LR 0.000500 Time 0.021046 +2023-10-05 21:45:03,903 - Epoch: [133][ 1160/ 1236] Overall Loss 0.276706 Objective Loss 0.276706 LR 0.000500 Time 0.021041 +2023-10-05 21:45:04,105 - Epoch: [133][ 1170/ 1236] Overall Loss 0.276695 Objective Loss 0.276695 LR 0.000500 Time 0.021033 +2023-10-05 21:45:04,309 - Epoch: [133][ 1180/ 1236] Overall Loss 0.276427 Objective Loss 0.276427 LR 0.000500 Time 0.021027 +2023-10-05 21:45:04,510 - Epoch: [133][ 1190/ 1236] Overall Loss 0.276203 Objective Loss 0.276203 LR 0.000500 Time 0.021019 +2023-10-05 21:45:04,714 - Epoch: [133][ 1200/ 1236] Overall Loss 0.275976 Objective Loss 0.275976 LR 0.000500 Time 0.021014 +2023-10-05 21:45:04,916 - Epoch: [133][ 1210/ 1236] Overall Loss 0.276216 Objective Loss 0.276216 LR 0.000500 Time 0.021007 +2023-10-05 21:45:05,120 - Epoch: [133][ 1220/ 1236] Overall Loss 0.276250 Objective Loss 0.276250 LR 0.000500 Time 0.021002 +2023-10-05 21:45:05,379 - Epoch: [133][ 1230/ 1236] Overall Loss 0.276391 Objective Loss 0.276391 LR 0.000500 Time 0.021041 +2023-10-05 21:45:05,498 - Epoch: [133][ 1236/ 1236] Overall Loss 0.276356 Objective Loss 0.276356 Top1 85.743381 Top5 98.574338 LR 0.000500 Time 0.021035 +2023-10-05 21:45:05,602 - --- validate (epoch=133)----------- +2023-10-05 21:45:05,602 - 29943 samples (256 per mini-batch) +2023-10-05 21:45:06,070 - Epoch: [133][ 10/ 117] Loss 0.328525 Top1 84.648438 Top5 97.617188 +2023-10-05 21:45:06,226 - Epoch: [133][ 20/ 117] Loss 0.338594 Top1 83.496094 Top5 97.753906 +2023-10-05 21:45:06,382 - Epoch: [133][ 30/ 117] Loss 0.342520 Top1 83.385417 Top5 97.747396 +2023-10-05 21:45:06,537 - Epoch: [133][ 40/ 117] Loss 0.338052 Top1 83.212891 Top5 97.880859 +2023-10-05 21:45:06,693 - Epoch: [133][ 50/ 117] Loss 0.340879 Top1 83.195312 Top5 97.695312 +2023-10-05 21:45:06,847 - Epoch: [133][ 60/ 117] Loss 0.337162 Top1 83.274740 Top5 97.747396 +2023-10-05 21:45:06,998 - Epoch: [133][ 70/ 117] Loss 0.335944 Top1 83.275670 Top5 97.762277 +2023-10-05 21:45:07,149 - Epoch: [133][ 80/ 117] Loss 0.337923 Top1 83.129883 Top5 97.734375 +2023-10-05 21:45:07,299 - Epoch: [133][ 90/ 117] Loss 0.337118 Top1 83.151042 Top5 97.712674 +2023-10-05 21:45:07,451 - Epoch: [133][ 100/ 117] Loss 0.334040 Top1 83.292969 Top5 97.757812 +2023-10-05 21:45:07,608 - Epoch: [133][ 110/ 117] Loss 0.333551 Top1 83.327415 Top5 97.748580 +2023-10-05 21:45:07,694 - Epoch: [133][ 117/ 117] Loss 0.332036 Top1 83.294927 Top5 97.725679 +2023-10-05 21:45:07,837 - ==> Top1: 83.295 Top5: 97.726 Loss: 0.332 + +2023-10-05 21:45:07,838 - ==> Confusion: +[[ 931 3 2 0 3 2 1 0 5 75 1 0 2 1 7 2 3 1 0 1 10] + [ 2 1058 3 0 9 20 1 17 1 0 2 3 0 0 1 3 1 0 5 1 4] + [ 4 2 960 17 3 0 23 7 0 2 2 3 9 3 2 2 0 2 3 4 8] + [ 4 1 10 962 0 3 2 2 4 1 8 0 7 2 30 5 0 6 25 2 15] + [ 34 4 1 0 966 4 0 0 0 9 0 1 0 0 11 4 7 4 0 1 4] + [ 5 37 0 1 3 985 1 22 0 1 2 12 1 14 7 1 4 0 2 6 12] + [ 0 8 25 0 0 1 1114 14 0 0 2 2 1 0 1 6 0 0 0 11 6] + [ 3 17 11 0 3 28 4 1076 1 3 3 7 2 2 0 5 0 1 38 7 7] + [ 15 3 3 0 1 0 0 0 973 54 9 2 1 9 7 3 0 0 6 1 2] + [ 95 0 3 1 2 3 1 0 26 952 1 2 1 14 2 3 2 2 0 2 7] + [ 4 9 7 7 1 3 2 6 18 2 954 4 0 13 4 3 2 0 5 0 9] + [ 1 0 0 0 0 14 0 2 0 1 0 965 16 3 0 1 1 15 0 12 4] + [ 2 0 4 7 1 1 0 0 4 1 0 38 978 0 2 7 2 10 0 4 7] + [ 2 0 1 1 2 13 0 0 14 19 4 3 3 1031 4 3 2 0 1 6 10] + [ 19 2 5 9 3 0 0 0 30 1 0 1 1 3 1004 0 1 1 11 0 10] + [ 0 1 3 0 3 1 2 0 0 0 1 11 6 2 1 1068 12 10 0 9 4] + [ 3 13 1 0 6 4 0 1 0 0 0 5 0 0 4 10 1101 0 0 7 6] + [ 0 0 0 3 1 0 2 0 1 1 0 3 17 1 0 4 0 1001 1 0 3] + [ 1 4 10 17 1 0 1 26 2 1 3 0 1 0 10 0 1 0 980 2 8] + [ 0 2 1 1 1 4 7 10 1 0 1 14 2 2 0 5 11 1 3 1080 6] + [ 163 234 172 65 103 134 50 110 117 95 174 134 361 317 180 64 132 66 167 265 4802]] + +2023-10-05 21:45:07,839 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:45:07,839 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:45:07,845 - + +2023-10-05 21:45:07,845 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:45:08,867 - Epoch: [134][ 10/ 1236] Overall Loss 0.267450 Objective Loss 0.267450 LR 0.000500 Time 0.102189 +2023-10-05 21:45:09,072 - Epoch: [134][ 20/ 1236] Overall Loss 0.267108 Objective Loss 0.267108 LR 0.000500 Time 0.061326 +2023-10-05 21:45:09,275 - Epoch: [134][ 30/ 1236] Overall Loss 0.270234 Objective Loss 0.270234 LR 0.000500 Time 0.047625 +2023-10-05 21:45:09,480 - Epoch: [134][ 40/ 1236] Overall Loss 0.267412 Objective Loss 0.267412 LR 0.000500 Time 0.040836 +2023-10-05 21:45:09,682 - Epoch: [134][ 50/ 1236] Overall Loss 0.269438 Objective Loss 0.269438 LR 0.000500 Time 0.036714 +2023-10-05 21:45:09,888 - Epoch: [134][ 60/ 1236] Overall Loss 0.266913 Objective Loss 0.266913 LR 0.000500 Time 0.034009 +2023-10-05 21:45:10,090 - Epoch: [134][ 70/ 1236] Overall Loss 0.274196 Objective Loss 0.274196 LR 0.000500 Time 0.032043 +2023-10-05 21:45:10,296 - Epoch: [134][ 80/ 1236] Overall Loss 0.268681 Objective Loss 0.268681 LR 0.000500 Time 0.030599 +2023-10-05 21:45:10,499 - Epoch: [134][ 90/ 1236] Overall Loss 0.269497 Objective Loss 0.269497 LR 0.000500 Time 0.029449 +2023-10-05 21:45:10,703 - Epoch: [134][ 100/ 1236] Overall Loss 0.269441 Objective Loss 0.269441 LR 0.000500 Time 0.028540 +2023-10-05 21:45:10,903 - Epoch: [134][ 110/ 1236] Overall Loss 0.269142 Objective Loss 0.269142 LR 0.000500 Time 0.027769 +2023-10-05 21:45:11,112 - Epoch: [134][ 120/ 1236] Overall Loss 0.269068 Objective Loss 0.269068 LR 0.000500 Time 0.027186 +2023-10-05 21:45:11,318 - Epoch: [134][ 130/ 1236] Overall Loss 0.269508 Objective Loss 0.269508 LR 0.000500 Time 0.026681 +2023-10-05 21:45:11,526 - Epoch: [134][ 140/ 1236] Overall Loss 0.268678 Objective Loss 0.268678 LR 0.000500 Time 0.026259 +2023-10-05 21:45:11,728 - Epoch: [134][ 150/ 1236] Overall Loss 0.268582 Objective Loss 0.268582 LR 0.000500 Time 0.025848 +2023-10-05 21:45:11,932 - Epoch: [134][ 160/ 1236] Overall Loss 0.268709 Objective Loss 0.268709 LR 0.000500 Time 0.025504 +2023-10-05 21:45:12,133 - Epoch: [134][ 170/ 1236] Overall Loss 0.267622 Objective Loss 0.267622 LR 0.000500 Time 0.025185 +2023-10-05 21:45:12,337 - Epoch: [134][ 180/ 1236] Overall Loss 0.267864 Objective Loss 0.267864 LR 0.000500 Time 0.024917 +2023-10-05 21:45:12,538 - Epoch: [134][ 190/ 1236] Overall Loss 0.267696 Objective Loss 0.267696 LR 0.000500 Time 0.024662 +2023-10-05 21:45:12,742 - Epoch: [134][ 200/ 1236] Overall Loss 0.267370 Objective Loss 0.267370 LR 0.000500 Time 0.024448 +2023-10-05 21:45:12,943 - Epoch: [134][ 210/ 1236] Overall Loss 0.267936 Objective Loss 0.267936 LR 0.000500 Time 0.024242 +2023-10-05 21:45:13,148 - Epoch: [134][ 220/ 1236] Overall Loss 0.268812 Objective Loss 0.268812 LR 0.000500 Time 0.024066 +2023-10-05 21:45:13,349 - Epoch: [134][ 230/ 1236] Overall Loss 0.268661 Objective Loss 0.268661 LR 0.000500 Time 0.023894 +2023-10-05 21:45:13,553 - Epoch: [134][ 240/ 1236] Overall Loss 0.268212 Objective Loss 0.268212 LR 0.000500 Time 0.023748 +2023-10-05 21:45:13,755 - Epoch: [134][ 250/ 1236] Overall Loss 0.268332 Objective Loss 0.268332 LR 0.000500 Time 0.023602 +2023-10-05 21:45:13,959 - Epoch: [134][ 260/ 1236] Overall Loss 0.270057 Objective Loss 0.270057 LR 0.000500 Time 0.023477 +2023-10-05 21:45:14,160 - Epoch: [134][ 270/ 1236] Overall Loss 0.270238 Objective Loss 0.270238 LR 0.000500 Time 0.023351 +2023-10-05 21:45:14,363 - Epoch: [134][ 280/ 1236] Overall Loss 0.269830 Objective Loss 0.269830 LR 0.000500 Time 0.023243 +2023-10-05 21:45:14,565 - Epoch: [134][ 290/ 1236] Overall Loss 0.270421 Objective Loss 0.270421 LR 0.000500 Time 0.023136 +2023-10-05 21:45:14,770 - Epoch: [134][ 300/ 1236] Overall Loss 0.269914 Objective Loss 0.269914 LR 0.000500 Time 0.023047 +2023-10-05 21:45:14,972 - Epoch: [134][ 310/ 1236] Overall Loss 0.269338 Objective Loss 0.269338 LR 0.000500 Time 0.022955 +2023-10-05 21:45:15,176 - Epoch: [134][ 320/ 1236] Overall Loss 0.268756 Objective Loss 0.268756 LR 0.000500 Time 0.022874 +2023-10-05 21:45:15,379 - Epoch: [134][ 330/ 1236] Overall Loss 0.268200 Objective Loss 0.268200 LR 0.000500 Time 0.022792 +2023-10-05 21:45:15,583 - Epoch: [134][ 340/ 1236] Overall Loss 0.267935 Objective Loss 0.267935 LR 0.000500 Time 0.022721 +2023-10-05 21:45:15,784 - Epoch: [134][ 350/ 1236] Overall Loss 0.267672 Objective Loss 0.267672 LR 0.000500 Time 0.022647 +2023-10-05 21:45:15,989 - Epoch: [134][ 360/ 1236] Overall Loss 0.267229 Objective Loss 0.267229 LR 0.000500 Time 0.022585 +2023-10-05 21:45:16,190 - Epoch: [134][ 370/ 1236] Overall Loss 0.267513 Objective Loss 0.267513 LR 0.000500 Time 0.022519 +2023-10-05 21:45:16,397 - Epoch: [134][ 380/ 1236] Overall Loss 0.267128 Objective Loss 0.267128 LR 0.000500 Time 0.022467 +2023-10-05 21:45:16,599 - Epoch: [134][ 390/ 1236] Overall Loss 0.266944 Objective Loss 0.266944 LR 0.000500 Time 0.022411 +2023-10-05 21:45:16,805 - Epoch: [134][ 400/ 1236] Overall Loss 0.267335 Objective Loss 0.267335 LR 0.000500 Time 0.022364 +2023-10-05 21:45:17,008 - Epoch: [134][ 410/ 1236] Overall Loss 0.266850 Objective Loss 0.266850 LR 0.000500 Time 0.022312 +2023-10-05 21:45:17,214 - Epoch: [134][ 420/ 1236] Overall Loss 0.266938 Objective Loss 0.266938 LR 0.000500 Time 0.022270 +2023-10-05 21:45:17,416 - Epoch: [134][ 430/ 1236] Overall Loss 0.267179 Objective Loss 0.267179 LR 0.000500 Time 0.022223 +2023-10-05 21:45:17,622 - Epoch: [134][ 440/ 1236] Overall Loss 0.267276 Objective Loss 0.267276 LR 0.000500 Time 0.022184 +2023-10-05 21:45:17,825 - Epoch: [134][ 450/ 1236] Overall Loss 0.266840 Objective Loss 0.266840 LR 0.000500 Time 0.022142 +2023-10-05 21:45:18,031 - Epoch: [134][ 460/ 1236] Overall Loss 0.267005 Objective Loss 0.267005 LR 0.000500 Time 0.022107 +2023-10-05 21:45:18,234 - Epoch: [134][ 470/ 1236] Overall Loss 0.266905 Objective Loss 0.266905 LR 0.000500 Time 0.022068 +2023-10-05 21:45:18,440 - Epoch: [134][ 480/ 1236] Overall Loss 0.266881 Objective Loss 0.266881 LR 0.000500 Time 0.022037 +2023-10-05 21:45:18,644 - Epoch: [134][ 490/ 1236] Overall Loss 0.266997 Objective Loss 0.266997 LR 0.000500 Time 0.022002 +2023-10-05 21:45:18,850 - Epoch: [134][ 500/ 1236] Overall Loss 0.267109 Objective Loss 0.267109 LR 0.000500 Time 0.021973 +2023-10-05 21:45:19,052 - Epoch: [134][ 510/ 1236] Overall Loss 0.266888 Objective Loss 0.266888 LR 0.000500 Time 0.021940 +2023-10-05 21:45:19,258 - Epoch: [134][ 520/ 1236] Overall Loss 0.267062 Objective Loss 0.267062 LR 0.000500 Time 0.021912 +2023-10-05 21:45:19,461 - Epoch: [134][ 530/ 1236] Overall Loss 0.267073 Objective Loss 0.267073 LR 0.000500 Time 0.021882 +2023-10-05 21:45:19,667 - Epoch: [134][ 540/ 1236] Overall Loss 0.267374 Objective Loss 0.267374 LR 0.000500 Time 0.021856 +2023-10-05 21:45:19,870 - Epoch: [134][ 550/ 1236] Overall Loss 0.267635 Objective Loss 0.267635 LR 0.000500 Time 0.021827 +2023-10-05 21:45:20,075 - Epoch: [134][ 560/ 1236] Overall Loss 0.267631 Objective Loss 0.267631 LR 0.000500 Time 0.021804 +2023-10-05 21:45:20,278 - Epoch: [134][ 570/ 1236] Overall Loss 0.267434 Objective Loss 0.267434 LR 0.000500 Time 0.021777 +2023-10-05 21:45:20,483 - Epoch: [134][ 580/ 1236] Overall Loss 0.267124 Objective Loss 0.267124 LR 0.000500 Time 0.021755 +2023-10-05 21:45:20,686 - Epoch: [134][ 590/ 1236] Overall Loss 0.267109 Objective Loss 0.267109 LR 0.000500 Time 0.021729 +2023-10-05 21:45:20,891 - Epoch: [134][ 600/ 1236] Overall Loss 0.266835 Objective Loss 0.266835 LR 0.000500 Time 0.021709 +2023-10-05 21:45:21,095 - Epoch: [134][ 610/ 1236] Overall Loss 0.266751 Objective Loss 0.266751 LR 0.000500 Time 0.021685 +2023-10-05 21:45:21,300 - Epoch: [134][ 620/ 1236] Overall Loss 0.266785 Objective Loss 0.266785 LR 0.000500 Time 0.021666 +2023-10-05 21:45:21,503 - Epoch: [134][ 630/ 1236] Overall Loss 0.267077 Objective Loss 0.267077 LR 0.000500 Time 0.021644 +2023-10-05 21:45:21,708 - Epoch: [134][ 640/ 1236] Overall Loss 0.267356 Objective Loss 0.267356 LR 0.000500 Time 0.021626 +2023-10-05 21:45:21,912 - Epoch: [134][ 650/ 1236] Overall Loss 0.267634 Objective Loss 0.267634 LR 0.000500 Time 0.021605 +2023-10-05 21:45:22,117 - Epoch: [134][ 660/ 1236] Overall Loss 0.267444 Objective Loss 0.267444 LR 0.000500 Time 0.021588 +2023-10-05 21:45:22,320 - Epoch: [134][ 670/ 1236] Overall Loss 0.267174 Objective Loss 0.267174 LR 0.000500 Time 0.021569 +2023-10-05 21:45:22,525 - Epoch: [134][ 680/ 1236] Overall Loss 0.267464 Objective Loss 0.267464 LR 0.000500 Time 0.021553 +2023-10-05 21:45:22,728 - Epoch: [134][ 690/ 1236] Overall Loss 0.267188 Objective Loss 0.267188 LR 0.000500 Time 0.021534 +2023-10-05 21:45:22,934 - Epoch: [134][ 700/ 1236] Overall Loss 0.266681 Objective Loss 0.266681 LR 0.000500 Time 0.021520 +2023-10-05 21:45:23,137 - Epoch: [134][ 710/ 1236] Overall Loss 0.266518 Objective Loss 0.266518 LR 0.000500 Time 0.021502 +2023-10-05 21:45:23,342 - Epoch: [134][ 720/ 1236] Overall Loss 0.267012 Objective Loss 0.267012 LR 0.000500 Time 0.021489 +2023-10-05 21:45:23,545 - Epoch: [134][ 730/ 1236] Overall Loss 0.266955 Objective Loss 0.266955 LR 0.000500 Time 0.021472 +2023-10-05 21:45:23,751 - Epoch: [134][ 740/ 1236] Overall Loss 0.266732 Objective Loss 0.266732 LR 0.000500 Time 0.021459 +2023-10-05 21:45:23,954 - Epoch: [134][ 750/ 1236] Overall Loss 0.266691 Objective Loss 0.266691 LR 0.000500 Time 0.021443 +2023-10-05 21:45:24,159 - Epoch: [134][ 760/ 1236] Overall Loss 0.266896 Objective Loss 0.266896 LR 0.000500 Time 0.021431 +2023-10-05 21:45:24,362 - Epoch: [134][ 770/ 1236] Overall Loss 0.267037 Objective Loss 0.267037 LR 0.000500 Time 0.021416 +2023-10-05 21:45:24,568 - Epoch: [134][ 780/ 1236] Overall Loss 0.266858 Objective Loss 0.266858 LR 0.000500 Time 0.021404 +2023-10-05 21:45:24,771 - Epoch: [134][ 790/ 1236] Overall Loss 0.266793 Objective Loss 0.266793 LR 0.000500 Time 0.021390 +2023-10-05 21:45:24,977 - Epoch: [134][ 800/ 1236] Overall Loss 0.266664 Objective Loss 0.266664 LR 0.000500 Time 0.021380 +2023-10-05 21:45:25,180 - Epoch: [134][ 810/ 1236] Overall Loss 0.266822 Objective Loss 0.266822 LR 0.000500 Time 0.021367 +2023-10-05 21:45:25,386 - Epoch: [134][ 820/ 1236] Overall Loss 0.266762 Objective Loss 0.266762 LR 0.000500 Time 0.021356 +2023-10-05 21:45:25,589 - Epoch: [134][ 830/ 1236] Overall Loss 0.266758 Objective Loss 0.266758 LR 0.000500 Time 0.021343 +2023-10-05 21:45:25,795 - Epoch: [134][ 840/ 1236] Overall Loss 0.266989 Objective Loss 0.266989 LR 0.000500 Time 0.021334 +2023-10-05 21:45:25,998 - Epoch: [134][ 850/ 1236] Overall Loss 0.266949 Objective Loss 0.266949 LR 0.000500 Time 0.021321 +2023-10-05 21:45:26,203 - Epoch: [134][ 860/ 1236] Overall Loss 0.267093 Objective Loss 0.267093 LR 0.000500 Time 0.021312 +2023-10-05 21:45:26,407 - Epoch: [134][ 870/ 1236] Overall Loss 0.267167 Objective Loss 0.267167 LR 0.000500 Time 0.021300 +2023-10-05 21:45:26,612 - Epoch: [134][ 880/ 1236] Overall Loss 0.267395 Objective Loss 0.267395 LR 0.000500 Time 0.021291 +2023-10-05 21:45:26,815 - Epoch: [134][ 890/ 1236] Overall Loss 0.267627 Objective Loss 0.267627 LR 0.000500 Time 0.021280 +2023-10-05 21:45:27,021 - Epoch: [134][ 900/ 1236] Overall Loss 0.267464 Objective Loss 0.267464 LR 0.000500 Time 0.021271 +2023-10-05 21:45:27,224 - Epoch: [134][ 910/ 1236] Overall Loss 0.267791 Objective Loss 0.267791 LR 0.000500 Time 0.021260 +2023-10-05 21:45:27,429 - Epoch: [134][ 920/ 1236] Overall Loss 0.267563 Objective Loss 0.267563 LR 0.000500 Time 0.021252 +2023-10-05 21:45:27,632 - Epoch: [134][ 930/ 1236] Overall Loss 0.267653 Objective Loss 0.267653 LR 0.000500 Time 0.021242 +2023-10-05 21:45:27,838 - Epoch: [134][ 940/ 1236] Overall Loss 0.267672 Objective Loss 0.267672 LR 0.000500 Time 0.021234 +2023-10-05 21:45:28,041 - Epoch: [134][ 950/ 1236] Overall Loss 0.267718 Objective Loss 0.267718 LR 0.000500 Time 0.021225 +2023-10-05 21:45:28,247 - Epoch: [134][ 960/ 1236] Overall Loss 0.267658 Objective Loss 0.267658 LR 0.000500 Time 0.021217 +2023-10-05 21:45:28,450 - Epoch: [134][ 970/ 1236] Overall Loss 0.267623 Objective Loss 0.267623 LR 0.000500 Time 0.021208 +2023-10-05 21:45:28,656 - Epoch: [134][ 980/ 1236] Overall Loss 0.267578 Objective Loss 0.267578 LR 0.000500 Time 0.021201 +2023-10-05 21:45:28,859 - Epoch: [134][ 990/ 1236] Overall Loss 0.267605 Objective Loss 0.267605 LR 0.000500 Time 0.021192 +2023-10-05 21:45:29,065 - Epoch: [134][ 1000/ 1236] Overall Loss 0.267525 Objective Loss 0.267525 LR 0.000500 Time 0.021185 +2023-10-05 21:45:29,269 - Epoch: [134][ 1010/ 1236] Overall Loss 0.267438 Objective Loss 0.267438 LR 0.000500 Time 0.021177 +2023-10-05 21:45:29,474 - Epoch: [134][ 1020/ 1236] Overall Loss 0.267415 Objective Loss 0.267415 LR 0.000500 Time 0.021170 +2023-10-05 21:45:29,677 - Epoch: [134][ 1030/ 1236] Overall Loss 0.267488 Objective Loss 0.267488 LR 0.000500 Time 0.021162 +2023-10-05 21:45:29,883 - Epoch: [134][ 1040/ 1236] Overall Loss 0.267567 Objective Loss 0.267567 LR 0.000500 Time 0.021155 +2023-10-05 21:45:30,086 - Epoch: [134][ 1050/ 1236] Overall Loss 0.267640 Objective Loss 0.267640 LR 0.000500 Time 0.021147 +2023-10-05 21:45:30,291 - Epoch: [134][ 1060/ 1236] Overall Loss 0.267590 Objective Loss 0.267590 LR 0.000500 Time 0.021141 +2023-10-05 21:45:30,495 - Epoch: [134][ 1070/ 1236] Overall Loss 0.267664 Objective Loss 0.267664 LR 0.000500 Time 0.021133 +2023-10-05 21:45:30,700 - Epoch: [134][ 1080/ 1236] Overall Loss 0.267556 Objective Loss 0.267556 LR 0.000500 Time 0.021127 +2023-10-05 21:45:30,903 - Epoch: [134][ 1090/ 1236] Overall Loss 0.267429 Objective Loss 0.267429 LR 0.000500 Time 0.021119 +2023-10-05 21:45:31,108 - Epoch: [134][ 1100/ 1236] Overall Loss 0.267541 Objective Loss 0.267541 LR 0.000500 Time 0.021114 +2023-10-05 21:45:31,312 - Epoch: [134][ 1110/ 1236] Overall Loss 0.267597 Objective Loss 0.267597 LR 0.000500 Time 0.021106 +2023-10-05 21:45:31,517 - Epoch: [134][ 1120/ 1236] Overall Loss 0.267525 Objective Loss 0.267525 LR 0.000500 Time 0.021101 +2023-10-05 21:45:31,721 - Epoch: [134][ 1130/ 1236] Overall Loss 0.267600 Objective Loss 0.267600 LR 0.000500 Time 0.021094 +2023-10-05 21:45:31,926 - Epoch: [134][ 1140/ 1236] Overall Loss 0.267679 Objective Loss 0.267679 LR 0.000500 Time 0.021089 +2023-10-05 21:45:32,129 - Epoch: [134][ 1150/ 1236] Overall Loss 0.267717 Objective Loss 0.267717 LR 0.000500 Time 0.021082 +2023-10-05 21:45:32,335 - Epoch: [134][ 1160/ 1236] Overall Loss 0.267860 Objective Loss 0.267860 LR 0.000500 Time 0.021077 +2023-10-05 21:45:32,538 - Epoch: [134][ 1170/ 1236] Overall Loss 0.267844 Objective Loss 0.267844 LR 0.000500 Time 0.021070 +2023-10-05 21:45:32,743 - Epoch: [134][ 1180/ 1236] Overall Loss 0.267854 Objective Loss 0.267854 LR 0.000500 Time 0.021065 +2023-10-05 21:45:32,947 - Epoch: [134][ 1190/ 1236] Overall Loss 0.267916 Objective Loss 0.267916 LR 0.000500 Time 0.021059 +2023-10-05 21:45:33,152 - Epoch: [134][ 1200/ 1236] Overall Loss 0.267808 Objective Loss 0.267808 LR 0.000500 Time 0.021054 +2023-10-05 21:45:33,355 - Epoch: [134][ 1210/ 1236] Overall Loss 0.267790 Objective Loss 0.267790 LR 0.000500 Time 0.021048 +2023-10-05 21:45:33,560 - Epoch: [134][ 1220/ 1236] Overall Loss 0.267670 Objective Loss 0.267670 LR 0.000500 Time 0.021043 +2023-10-05 21:45:33,818 - Epoch: [134][ 1230/ 1236] Overall Loss 0.267709 Objective Loss 0.267709 LR 0.000500 Time 0.021082 +2023-10-05 21:45:33,938 - Epoch: [134][ 1236/ 1236] Overall Loss 0.267705 Objective Loss 0.267705 Top1 85.743381 Top5 98.778004 LR 0.000500 Time 0.021076 +2023-10-05 21:45:34,081 - --- validate (epoch=134)----------- +2023-10-05 21:45:34,081 - 29943 samples (256 per mini-batch) +2023-10-05 21:45:34,543 - Epoch: [134][ 10/ 117] Loss 0.301160 Top1 83.085938 Top5 98.046875 +2023-10-05 21:45:34,697 - Epoch: [134][ 20/ 117] Loss 0.313784 Top1 82.968750 Top5 97.910156 +2023-10-05 21:45:34,848 - Epoch: [134][ 30/ 117] Loss 0.315378 Top1 83.059896 Top5 97.786458 +2023-10-05 21:45:35,001 - Epoch: [134][ 40/ 117] Loss 0.319273 Top1 82.968750 Top5 97.851562 +2023-10-05 21:45:35,153 - Epoch: [134][ 50/ 117] Loss 0.320193 Top1 83.000000 Top5 97.796875 +2023-10-05 21:45:35,305 - Epoch: [134][ 60/ 117] Loss 0.324509 Top1 82.923177 Top5 97.714844 +2023-10-05 21:45:35,456 - Epoch: [134][ 70/ 117] Loss 0.326925 Top1 82.946429 Top5 97.672991 +2023-10-05 21:45:35,605 - Epoch: [134][ 80/ 117] Loss 0.328288 Top1 82.900391 Top5 97.729492 +2023-10-05 21:45:35,752 - Epoch: [134][ 90/ 117] Loss 0.327158 Top1 82.999132 Top5 97.760417 +2023-10-05 21:45:35,903 - Epoch: [134][ 100/ 117] Loss 0.324827 Top1 83.105469 Top5 97.757812 +2023-10-05 21:45:36,060 - Epoch: [134][ 110/ 117] Loss 0.325873 Top1 83.142756 Top5 97.776989 +2023-10-05 21:45:36,146 - Epoch: [134][ 117/ 117] Loss 0.326176 Top1 83.124603 Top5 97.755736 +2023-10-05 21:45:36,294 - ==> Top1: 83.125 Top5: 97.756 Loss: 0.326 + +2023-10-05 21:45:36,294 - ==> Confusion: +[[ 928 4 4 0 8 0 0 0 5 66 2 1 1 2 8 0 3 2 3 0 13] + [ 3 1061 2 0 5 17 1 15 2 1 1 1 0 0 0 4 2 0 11 1 4] + [ 5 3 961 11 1 0 27 11 0 0 2 2 8 0 1 4 1 2 5 3 9] + [ 3 2 13 980 2 5 0 1 3 0 7 0 5 1 19 2 1 3 25 4 13] + [ 19 12 2 0 968 4 0 1 0 9 1 1 0 1 8 3 13 2 0 3 3] + [ 5 58 0 1 4 948 2 33 1 2 1 11 0 18 3 2 3 1 2 5 16] + [ 0 5 24 0 0 0 1121 9 0 0 2 2 2 1 2 11 0 0 1 8 3] + [ 2 22 10 0 3 21 3 1071 1 4 3 7 6 2 0 2 0 0 44 10 7] + [ 20 6 0 0 1 3 0 0 978 43 9 3 1 6 9 2 2 0 3 1 2] + [ 95 1 4 0 5 2 0 0 24 949 0 0 0 12 4 7 2 2 0 3 9] + [ 4 7 10 5 1 0 3 3 17 2 958 5 0 13 4 0 4 0 4 1 12] + [ 1 0 0 0 0 10 1 2 0 1 0 958 15 10 0 2 1 19 1 10 4] + [ 1 1 5 6 0 0 0 2 0 0 0 37 985 1 1 8 1 11 1 2 6] + [ 5 0 1 0 2 3 0 1 8 18 5 5 1 1050 4 2 2 1 0 3 8] + [ 12 3 4 16 6 0 0 0 26 0 3 1 4 2 1001 0 0 1 13 0 9] + [ 0 3 2 1 4 0 1 0 0 0 1 6 5 2 0 1070 12 12 0 12 3] + [ 0 18 1 0 3 2 0 1 3 0 0 8 1 0 3 10 1097 0 0 5 9] + [ 0 1 0 2 0 0 3 0 1 1 0 0 16 1 2 6 0 1002 1 0 2] + [ 2 4 5 16 0 0 0 27 2 0 2 0 3 0 7 0 2 0 991 2 5] + [ 0 2 3 4 2 1 9 11 1 0 2 15 3 2 0 2 13 0 2 1075 5] + [ 111 284 171 71 110 117 40 99 97 72 203 119 447 306 153 76 154 74 190 273 4738]] + +2023-10-05 21:45:36,296 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:45:36,296 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:45:36,302 - + +2023-10-05 21:45:36,302 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:45:37,317 - Epoch: [135][ 10/ 1236] Overall Loss 0.259658 Objective Loss 0.259658 LR 0.000500 Time 0.101486 +2023-10-05 21:45:37,521 - Epoch: [135][ 20/ 1236] Overall Loss 0.262106 Objective Loss 0.262106 LR 0.000500 Time 0.060914 +2023-10-05 21:45:37,724 - Epoch: [135][ 30/ 1236] Overall Loss 0.269697 Objective Loss 0.269697 LR 0.000500 Time 0.047350 +2023-10-05 21:45:37,928 - Epoch: [135][ 40/ 1236] Overall Loss 0.264956 Objective Loss 0.264956 LR 0.000500 Time 0.040600 +2023-10-05 21:45:38,130 - Epoch: [135][ 50/ 1236] Overall Loss 0.261099 Objective Loss 0.261099 LR 0.000500 Time 0.036519 +2023-10-05 21:45:38,333 - Epoch: [135][ 60/ 1236] Overall Loss 0.263058 Objective Loss 0.263058 LR 0.000500 Time 0.033819 +2023-10-05 21:45:38,536 - Epoch: [135][ 70/ 1236] Overall Loss 0.261287 Objective Loss 0.261287 LR 0.000500 Time 0.031877 +2023-10-05 21:45:38,740 - Epoch: [135][ 80/ 1236] Overall Loss 0.264047 Objective Loss 0.264047 LR 0.000500 Time 0.030436 +2023-10-05 21:45:38,943 - Epoch: [135][ 90/ 1236] Overall Loss 0.265558 Objective Loss 0.265558 LR 0.000500 Time 0.029307 +2023-10-05 21:45:39,148 - Epoch: [135][ 100/ 1236] Overall Loss 0.266264 Objective Loss 0.266264 LR 0.000500 Time 0.028426 +2023-10-05 21:45:39,351 - Epoch: [135][ 110/ 1236] Overall Loss 0.266211 Objective Loss 0.266211 LR 0.000500 Time 0.027681 +2023-10-05 21:45:39,554 - Epoch: [135][ 120/ 1236] Overall Loss 0.267531 Objective Loss 0.267531 LR 0.000500 Time 0.027061 +2023-10-05 21:45:39,754 - Epoch: [135][ 130/ 1236] Overall Loss 0.266581 Objective Loss 0.266581 LR 0.000500 Time 0.026517 +2023-10-05 21:45:39,957 - Epoch: [135][ 140/ 1236] Overall Loss 0.266941 Objective Loss 0.266941 LR 0.000500 Time 0.026069 +2023-10-05 21:45:40,159 - Epoch: [135][ 150/ 1236] Overall Loss 0.266977 Objective Loss 0.266977 LR 0.000500 Time 0.025675 +2023-10-05 21:45:40,361 - Epoch: [135][ 160/ 1236] Overall Loss 0.266203 Objective Loss 0.266203 LR 0.000500 Time 0.025335 +2023-10-05 21:45:40,564 - Epoch: [135][ 170/ 1236] Overall Loss 0.267315 Objective Loss 0.267315 LR 0.000500 Time 0.025033 +2023-10-05 21:45:40,765 - Epoch: [135][ 180/ 1236] Overall Loss 0.266508 Objective Loss 0.266508 LR 0.000500 Time 0.024761 +2023-10-05 21:45:40,968 - Epoch: [135][ 190/ 1236] Overall Loss 0.265335 Objective Loss 0.265335 LR 0.000500 Time 0.024521 +2023-10-05 21:45:41,169 - Epoch: [135][ 200/ 1236] Overall Loss 0.265754 Objective Loss 0.265754 LR 0.000500 Time 0.024303 +2023-10-05 21:45:41,371 - Epoch: [135][ 210/ 1236] Overall Loss 0.265187 Objective Loss 0.265187 LR 0.000500 Time 0.024105 +2023-10-05 21:45:41,574 - Epoch: [135][ 220/ 1236] Overall Loss 0.264595 Objective Loss 0.264595 LR 0.000500 Time 0.023931 +2023-10-05 21:45:41,777 - Epoch: [135][ 230/ 1236] Overall Loss 0.264416 Objective Loss 0.264416 LR 0.000500 Time 0.023767 +2023-10-05 21:45:41,980 - Epoch: [135][ 240/ 1236] Overall Loss 0.263520 Objective Loss 0.263520 LR 0.000500 Time 0.023622 +2023-10-05 21:45:42,181 - Epoch: [135][ 250/ 1236] Overall Loss 0.263814 Objective Loss 0.263814 LR 0.000500 Time 0.023479 +2023-10-05 21:45:42,384 - Epoch: [135][ 260/ 1236] Overall Loss 0.263407 Objective Loss 0.263407 LR 0.000500 Time 0.023356 +2023-10-05 21:45:42,586 - Epoch: [135][ 270/ 1236] Overall Loss 0.262925 Objective Loss 0.262925 LR 0.000500 Time 0.023237 +2023-10-05 21:45:42,789 - Epoch: [135][ 280/ 1236] Overall Loss 0.262997 Objective Loss 0.262997 LR 0.000500 Time 0.023131 +2023-10-05 21:45:42,990 - Epoch: [135][ 290/ 1236] Overall Loss 0.263530 Objective Loss 0.263530 LR 0.000500 Time 0.023029 +2023-10-05 21:45:43,194 - Epoch: [135][ 300/ 1236] Overall Loss 0.263275 Objective Loss 0.263275 LR 0.000500 Time 0.022937 +2023-10-05 21:45:43,396 - Epoch: [135][ 310/ 1236] Overall Loss 0.263200 Objective Loss 0.263200 LR 0.000500 Time 0.022848 +2023-10-05 21:45:43,598 - Epoch: [135][ 320/ 1236] Overall Loss 0.263871 Objective Loss 0.263871 LR 0.000500 Time 0.022764 +2023-10-05 21:45:43,800 - Epoch: [135][ 330/ 1236] Overall Loss 0.264021 Objective Loss 0.264021 LR 0.000500 Time 0.022686 +2023-10-05 21:45:44,003 - Epoch: [135][ 340/ 1236] Overall Loss 0.263705 Objective Loss 0.263705 LR 0.000500 Time 0.022615 +2023-10-05 21:45:44,205 - Epoch: [135][ 350/ 1236] Overall Loss 0.263735 Objective Loss 0.263735 LR 0.000500 Time 0.022544 +2023-10-05 21:45:44,408 - Epoch: [135][ 360/ 1236] Overall Loss 0.263500 Objective Loss 0.263500 LR 0.000500 Time 0.022481 +2023-10-05 21:45:44,610 - Epoch: [135][ 370/ 1236] Overall Loss 0.263323 Objective Loss 0.263323 LR 0.000500 Time 0.022420 +2023-10-05 21:45:44,815 - Epoch: [135][ 380/ 1236] Overall Loss 0.263305 Objective Loss 0.263305 LR 0.000500 Time 0.022366 +2023-10-05 21:45:45,017 - Epoch: [135][ 390/ 1236] Overall Loss 0.263105 Objective Loss 0.263105 LR 0.000500 Time 0.022311 +2023-10-05 21:45:45,221 - Epoch: [135][ 400/ 1236] Overall Loss 0.262979 Objective Loss 0.262979 LR 0.000500 Time 0.022262 +2023-10-05 21:45:45,424 - Epoch: [135][ 410/ 1236] Overall Loss 0.262858 Objective Loss 0.262858 LR 0.000500 Time 0.022212 +2023-10-05 21:45:45,628 - Epoch: [135][ 420/ 1236] Overall Loss 0.263739 Objective Loss 0.263739 LR 0.000500 Time 0.022169 +2023-10-05 21:45:45,830 - Epoch: [135][ 430/ 1236] Overall Loss 0.263974 Objective Loss 0.263974 LR 0.000500 Time 0.022123 +2023-10-05 21:45:46,034 - Epoch: [135][ 440/ 1236] Overall Loss 0.264346 Objective Loss 0.264346 LR 0.000500 Time 0.022084 +2023-10-05 21:45:46,237 - Epoch: [135][ 450/ 1236] Overall Loss 0.263222 Objective Loss 0.263222 LR 0.000500 Time 0.022042 +2023-10-05 21:45:46,441 - Epoch: [135][ 460/ 1236] Overall Loss 0.263536 Objective Loss 0.263536 LR 0.000500 Time 0.022006 +2023-10-05 21:45:46,644 - Epoch: [135][ 470/ 1236] Overall Loss 0.263964 Objective Loss 0.263964 LR 0.000500 Time 0.021968 +2023-10-05 21:45:46,848 - Epoch: [135][ 480/ 1236] Overall Loss 0.263829 Objective Loss 0.263829 LR 0.000500 Time 0.021936 +2023-10-05 21:45:47,050 - Epoch: [135][ 490/ 1236] Overall Loss 0.263809 Objective Loss 0.263809 LR 0.000500 Time 0.021900 +2023-10-05 21:45:47,254 - Epoch: [135][ 500/ 1236] Overall Loss 0.263461 Objective Loss 0.263461 LR 0.000500 Time 0.021870 +2023-10-05 21:45:47,457 - Epoch: [135][ 510/ 1236] Overall Loss 0.263073 Objective Loss 0.263073 LR 0.000500 Time 0.021837 +2023-10-05 21:45:47,661 - Epoch: [135][ 520/ 1236] Overall Loss 0.263106 Objective Loss 0.263106 LR 0.000500 Time 0.021809 +2023-10-05 21:45:47,864 - Epoch: [135][ 530/ 1236] Overall Loss 0.262772 Objective Loss 0.262772 LR 0.000500 Time 0.021779 +2023-10-05 21:45:48,068 - Epoch: [135][ 540/ 1236] Overall Loss 0.262469 Objective Loss 0.262469 LR 0.000500 Time 0.021754 +2023-10-05 21:45:48,271 - Epoch: [135][ 550/ 1236] Overall Loss 0.261994 Objective Loss 0.261994 LR 0.000500 Time 0.021726 +2023-10-05 21:45:48,475 - Epoch: [135][ 560/ 1236] Overall Loss 0.261879 Objective Loss 0.261879 LR 0.000500 Time 0.021702 +2023-10-05 21:45:48,677 - Epoch: [135][ 570/ 1236] Overall Loss 0.262207 Objective Loss 0.262207 LR 0.000500 Time 0.021676 +2023-10-05 21:45:48,881 - Epoch: [135][ 580/ 1236] Overall Loss 0.261823 Objective Loss 0.261823 LR 0.000500 Time 0.021653 +2023-10-05 21:45:49,084 - Epoch: [135][ 590/ 1236] Overall Loss 0.261867 Objective Loss 0.261867 LR 0.000500 Time 0.021629 +2023-10-05 21:45:49,288 - Epoch: [135][ 600/ 1236] Overall Loss 0.261640 Objective Loss 0.261640 LR 0.000500 Time 0.021608 +2023-10-05 21:45:49,490 - Epoch: [135][ 610/ 1236] Overall Loss 0.261923 Objective Loss 0.261923 LR 0.000500 Time 0.021585 +2023-10-05 21:45:49,695 - Epoch: [135][ 620/ 1236] Overall Loss 0.262597 Objective Loss 0.262597 LR 0.000500 Time 0.021566 +2023-10-05 21:45:49,898 - Epoch: [135][ 630/ 1236] Overall Loss 0.262745 Objective Loss 0.262745 LR 0.000500 Time 0.021545 +2023-10-05 21:45:50,101 - Epoch: [135][ 640/ 1236] Overall Loss 0.262346 Objective Loss 0.262346 LR 0.000500 Time 0.021527 +2023-10-05 21:45:50,304 - Epoch: [135][ 650/ 1236] Overall Loss 0.261921 Objective Loss 0.261921 LR 0.000500 Time 0.021506 +2023-10-05 21:45:50,508 - Epoch: [135][ 660/ 1236] Overall Loss 0.261931 Objective Loss 0.261931 LR 0.000500 Time 0.021489 +2023-10-05 21:45:50,711 - Epoch: [135][ 670/ 1236] Overall Loss 0.261791 Objective Loss 0.261791 LR 0.000500 Time 0.021471 +2023-10-05 21:45:50,915 - Epoch: [135][ 680/ 1236] Overall Loss 0.262173 Objective Loss 0.262173 LR 0.000500 Time 0.021454 +2023-10-05 21:45:51,118 - Epoch: [135][ 690/ 1236] Overall Loss 0.262357 Objective Loss 0.262357 LR 0.000500 Time 0.021437 +2023-10-05 21:45:51,322 - Epoch: [135][ 700/ 1236] Overall Loss 0.262384 Objective Loss 0.262384 LR 0.000500 Time 0.021422 +2023-10-05 21:45:51,525 - Epoch: [135][ 710/ 1236] Overall Loss 0.262287 Objective Loss 0.262287 LR 0.000500 Time 0.021406 +2023-10-05 21:45:51,729 - Epoch: [135][ 720/ 1236] Overall Loss 0.262363 Objective Loss 0.262363 LR 0.000500 Time 0.021392 +2023-10-05 21:45:51,931 - Epoch: [135][ 730/ 1236] Overall Loss 0.262738 Objective Loss 0.262738 LR 0.000500 Time 0.021375 +2023-10-05 21:45:52,136 - Epoch: [135][ 740/ 1236] Overall Loss 0.262551 Objective Loss 0.262551 LR 0.000500 Time 0.021362 +2023-10-05 21:45:52,338 - Epoch: [135][ 750/ 1236] Overall Loss 0.262352 Objective Loss 0.262352 LR 0.000500 Time 0.021346 +2023-10-05 21:45:52,542 - Epoch: [135][ 760/ 1236] Overall Loss 0.262482 Objective Loss 0.262482 LR 0.000500 Time 0.021334 +2023-10-05 21:45:52,745 - Epoch: [135][ 770/ 1236] Overall Loss 0.262610 Objective Loss 0.262610 LR 0.000500 Time 0.021319 +2023-10-05 21:45:52,949 - Epoch: [135][ 780/ 1236] Overall Loss 0.262302 Objective Loss 0.262302 LR 0.000500 Time 0.021308 +2023-10-05 21:45:53,152 - Epoch: [135][ 790/ 1236] Overall Loss 0.262260 Objective Loss 0.262260 LR 0.000500 Time 0.021295 +2023-10-05 21:45:53,356 - Epoch: [135][ 800/ 1236] Overall Loss 0.262329 Objective Loss 0.262329 LR 0.000500 Time 0.021283 +2023-10-05 21:45:53,559 - Epoch: [135][ 810/ 1236] Overall Loss 0.262620 Objective Loss 0.262620 LR 0.000500 Time 0.021270 +2023-10-05 21:45:53,763 - Epoch: [135][ 820/ 1236] Overall Loss 0.262686 Objective Loss 0.262686 LR 0.000500 Time 0.021259 +2023-10-05 21:45:53,966 - Epoch: [135][ 830/ 1236] Overall Loss 0.262662 Objective Loss 0.262662 LR 0.000500 Time 0.021247 +2023-10-05 21:45:54,170 - Epoch: [135][ 840/ 1236] Overall Loss 0.262612 Objective Loss 0.262612 LR 0.000500 Time 0.021237 +2023-10-05 21:45:54,373 - Epoch: [135][ 850/ 1236] Overall Loss 0.262542 Objective Loss 0.262542 LR 0.000500 Time 0.021225 +2023-10-05 21:45:54,577 - Epoch: [135][ 860/ 1236] Overall Loss 0.262567 Objective Loss 0.262567 LR 0.000500 Time 0.021215 +2023-10-05 21:45:54,780 - Epoch: [135][ 870/ 1236] Overall Loss 0.262736 Objective Loss 0.262736 LR 0.000500 Time 0.021204 +2023-10-05 21:45:54,984 - Epoch: [135][ 880/ 1236] Overall Loss 0.262627 Objective Loss 0.262627 LR 0.000500 Time 0.021195 +2023-10-05 21:45:55,186 - Epoch: [135][ 890/ 1236] Overall Loss 0.262530 Objective Loss 0.262530 LR 0.000500 Time 0.021184 +2023-10-05 21:45:55,391 - Epoch: [135][ 900/ 1236] Overall Loss 0.262554 Objective Loss 0.262554 LR 0.000500 Time 0.021175 +2023-10-05 21:45:55,593 - Epoch: [135][ 910/ 1236] Overall Loss 0.262530 Objective Loss 0.262530 LR 0.000500 Time 0.021164 +2023-10-05 21:45:55,798 - Epoch: [135][ 920/ 1236] Overall Loss 0.262810 Objective Loss 0.262810 LR 0.000500 Time 0.021156 +2023-10-05 21:45:56,000 - Epoch: [135][ 930/ 1236] Overall Loss 0.263143 Objective Loss 0.263143 LR 0.000500 Time 0.021146 +2023-10-05 21:45:56,205 - Epoch: [135][ 940/ 1236] Overall Loss 0.263148 Objective Loss 0.263148 LR 0.000500 Time 0.021139 +2023-10-05 21:45:56,407 - Epoch: [135][ 950/ 1236] Overall Loss 0.263214 Objective Loss 0.263214 LR 0.000500 Time 0.021129 +2023-10-05 21:45:56,611 - Epoch: [135][ 960/ 1236] Overall Loss 0.263185 Objective Loss 0.263185 LR 0.000500 Time 0.021120 +2023-10-05 21:45:56,813 - Epoch: [135][ 970/ 1236] Overall Loss 0.263226 Objective Loss 0.263226 LR 0.000500 Time 0.021111 +2023-10-05 21:45:57,017 - Epoch: [135][ 980/ 1236] Overall Loss 0.263390 Objective Loss 0.263390 LR 0.000500 Time 0.021104 +2023-10-05 21:45:57,220 - Epoch: [135][ 990/ 1236] Overall Loss 0.263398 Objective Loss 0.263398 LR 0.000500 Time 0.021095 +2023-10-05 21:45:57,424 - Epoch: [135][ 1000/ 1236] Overall Loss 0.263489 Objective Loss 0.263489 LR 0.000500 Time 0.021088 +2023-10-05 21:45:57,627 - Epoch: [135][ 1010/ 1236] Overall Loss 0.263617 Objective Loss 0.263617 LR 0.000500 Time 0.021080 +2023-10-05 21:45:57,831 - Epoch: [135][ 1020/ 1236] Overall Loss 0.263605 Objective Loss 0.263605 LR 0.000500 Time 0.021073 +2023-10-05 21:45:58,034 - Epoch: [135][ 1030/ 1236] Overall Loss 0.263539 Objective Loss 0.263539 LR 0.000500 Time 0.021064 +2023-10-05 21:45:58,238 - Epoch: [135][ 1040/ 1236] Overall Loss 0.263670 Objective Loss 0.263670 LR 0.000500 Time 0.021058 +2023-10-05 21:45:58,440 - Epoch: [135][ 1050/ 1236] Overall Loss 0.263669 Objective Loss 0.263669 LR 0.000500 Time 0.021050 +2023-10-05 21:45:58,645 - Epoch: [135][ 1060/ 1236] Overall Loss 0.263757 Objective Loss 0.263757 LR 0.000500 Time 0.021044 +2023-10-05 21:45:58,847 - Epoch: [135][ 1070/ 1236] Overall Loss 0.263702 Objective Loss 0.263702 LR 0.000500 Time 0.021036 +2023-10-05 21:45:59,052 - Epoch: [135][ 1080/ 1236] Overall Loss 0.263851 Objective Loss 0.263851 LR 0.000500 Time 0.021030 +2023-10-05 21:45:59,254 - Epoch: [135][ 1090/ 1236] Overall Loss 0.263853 Objective Loss 0.263853 LR 0.000500 Time 0.021023 +2023-10-05 21:45:59,459 - Epoch: [135][ 1100/ 1236] Overall Loss 0.263829 Objective Loss 0.263829 LR 0.000500 Time 0.021017 +2023-10-05 21:45:59,661 - Epoch: [135][ 1110/ 1236] Overall Loss 0.264032 Objective Loss 0.264032 LR 0.000500 Time 0.021010 +2023-10-05 21:45:59,866 - Epoch: [135][ 1120/ 1236] Overall Loss 0.263988 Objective Loss 0.263988 LR 0.000500 Time 0.021005 +2023-10-05 21:46:00,069 - Epoch: [135][ 1130/ 1236] Overall Loss 0.264040 Objective Loss 0.264040 LR 0.000500 Time 0.020998 +2023-10-05 21:46:00,273 - Epoch: [135][ 1140/ 1236] Overall Loss 0.264047 Objective Loss 0.264047 LR 0.000500 Time 0.020992 +2023-10-05 21:46:00,475 - Epoch: [135][ 1150/ 1236] Overall Loss 0.264067 Objective Loss 0.264067 LR 0.000500 Time 0.020986 +2023-10-05 21:46:00,680 - Epoch: [135][ 1160/ 1236] Overall Loss 0.263928 Objective Loss 0.263928 LR 0.000500 Time 0.020981 +2023-10-05 21:46:00,882 - Epoch: [135][ 1170/ 1236] Overall Loss 0.263941 Objective Loss 0.263941 LR 0.000500 Time 0.020974 +2023-10-05 21:46:01,086 - Epoch: [135][ 1180/ 1236] Overall Loss 0.264302 Objective Loss 0.264302 LR 0.000500 Time 0.020969 +2023-10-05 21:46:01,288 - Epoch: [135][ 1190/ 1236] Overall Loss 0.264354 Objective Loss 0.264354 LR 0.000500 Time 0.020962 +2023-10-05 21:46:01,492 - Epoch: [135][ 1200/ 1236] Overall Loss 0.264361 Objective Loss 0.264361 LR 0.000500 Time 0.020957 +2023-10-05 21:46:01,695 - Epoch: [135][ 1210/ 1236] Overall Loss 0.264448 Objective Loss 0.264448 LR 0.000500 Time 0.020951 +2023-10-05 21:46:01,899 - Epoch: [135][ 1220/ 1236] Overall Loss 0.264478 Objective Loss 0.264478 LR 0.000500 Time 0.020947 +2023-10-05 21:46:02,156 - Epoch: [135][ 1230/ 1236] Overall Loss 0.264527 Objective Loss 0.264527 LR 0.000500 Time 0.020985 +2023-10-05 21:46:02,276 - Epoch: [135][ 1236/ 1236] Overall Loss 0.264338 Objective Loss 0.264338 Top1 86.354379 Top5 98.574338 LR 0.000500 Time 0.020980 +2023-10-05 21:46:02,420 - --- validate (epoch=135)----------- +2023-10-05 21:46:02,421 - 29943 samples (256 per mini-batch) +2023-10-05 21:46:02,884 - Epoch: [135][ 10/ 117] Loss 0.324493 Top1 82.968750 Top5 97.773438 +2023-10-05 21:46:03,037 - Epoch: [135][ 20/ 117] Loss 0.339156 Top1 83.183594 Top5 97.890625 +2023-10-05 21:46:03,188 - Epoch: [135][ 30/ 117] Loss 0.328381 Top1 83.984375 Top5 97.838542 +2023-10-05 21:46:03,342 - Epoch: [135][ 40/ 117] Loss 0.334375 Top1 83.867188 Top5 97.744141 +2023-10-05 21:46:03,494 - Epoch: [135][ 50/ 117] Loss 0.337582 Top1 83.828125 Top5 97.781250 +2023-10-05 21:46:03,647 - Epoch: [135][ 60/ 117] Loss 0.332838 Top1 83.828125 Top5 97.773438 +2023-10-05 21:46:03,799 - Epoch: [135][ 70/ 117] Loss 0.331459 Top1 83.850446 Top5 97.851562 +2023-10-05 21:46:03,949 - Epoch: [135][ 80/ 117] Loss 0.332737 Top1 83.759766 Top5 97.846680 +2023-10-05 21:46:04,099 - Epoch: [135][ 90/ 117] Loss 0.334265 Top1 83.676215 Top5 97.795139 +2023-10-05 21:46:04,250 - Epoch: [135][ 100/ 117] Loss 0.332070 Top1 83.722656 Top5 97.812500 +2023-10-05 21:46:04,407 - Epoch: [135][ 110/ 117] Loss 0.329186 Top1 83.824574 Top5 97.816051 +2023-10-05 21:46:04,493 - Epoch: [135][ 117/ 117] Loss 0.329519 Top1 83.872691 Top5 97.842568 +2023-10-05 21:46:04,628 - ==> Top1: 83.873 Top5: 97.843 Loss: 0.330 + +2023-10-05 21:46:04,628 - ==> Confusion: +[[ 929 1 4 0 9 1 0 0 7 59 2 0 1 2 12 3 2 0 3 0 15] + [ 0 1064 4 0 3 17 1 18 0 0 2 0 0 1 1 6 2 1 7 0 4] + [ 4 2 955 13 3 1 31 5 0 0 4 2 12 2 1 2 1 1 8 5 4] + [ 1 0 12 960 1 4 2 1 2 0 10 0 6 5 35 4 1 8 20 0 17] + [ 21 7 2 1 965 3 1 0 0 8 0 1 0 1 13 4 17 1 0 1 4] + [ 3 36 1 0 7 968 1 21 0 0 5 14 2 19 6 1 4 1 2 9 16] + [ 0 6 15 0 1 0 1132 2 0 0 3 4 2 0 1 9 0 1 1 6 8] + [ 4 20 9 0 5 35 5 1056 1 4 4 9 2 2 1 1 0 0 41 10 9] + [ 18 3 0 0 0 2 0 0 967 34 16 1 2 13 20 4 0 0 4 4 1] + [ 109 0 2 0 2 5 1 0 35 913 1 1 0 24 8 9 1 1 0 0 7] + [ 4 8 9 2 0 0 4 1 11 1 974 2 2 14 2 3 0 0 4 1 11] + [ 2 2 0 0 0 8 0 0 0 1 0 964 20 4 0 4 1 14 0 10 5] + [ 1 0 3 4 0 2 0 0 1 0 0 34 973 5 3 11 2 15 1 2 11] + [ 3 0 0 1 2 7 1 0 10 8 4 2 4 1058 3 3 2 1 0 4 6] + [ 11 3 3 10 7 0 0 0 16 1 3 1 1 4 1022 0 0 1 8 0 10] + [ 0 4 2 1 3 0 2 0 0 0 0 4 5 1 1 1077 14 7 0 10 3] + [ 1 10 2 0 5 4 0 0 1 0 0 1 0 0 5 9 1106 0 0 3 14] + [ 0 1 1 3 0 0 2 0 0 1 0 3 21 0 2 6 2 992 1 0 3] + [ 3 5 9 17 0 1 0 25 1 1 3 0 0 0 8 0 1 0 982 2 10] + [ 0 4 1 0 1 3 12 8 1 0 0 10 8 0 0 5 14 1 3 1074 7] + [ 118 224 145 69 107 132 57 72 100 67 220 107 376 313 158 71 161 57 141 227 4983]] + +2023-10-05 21:46:04,630 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:46:04,630 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:46:04,636 - + +2023-10-05 21:46:04,636 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:46:05,764 - Epoch: [136][ 10/ 1236] Overall Loss 0.247841 Objective Loss 0.247841 LR 0.000500 Time 0.112728 +2023-10-05 21:46:05,967 - Epoch: [136][ 20/ 1236] Overall Loss 0.242943 Objective Loss 0.242943 LR 0.000500 Time 0.066533 +2023-10-05 21:46:06,170 - Epoch: [136][ 30/ 1236] Overall Loss 0.252944 Objective Loss 0.252944 LR 0.000500 Time 0.051091 +2023-10-05 21:46:06,373 - Epoch: [136][ 40/ 1236] Overall Loss 0.257665 Objective Loss 0.257665 LR 0.000500 Time 0.043393 +2023-10-05 21:46:06,575 - Epoch: [136][ 50/ 1236] Overall Loss 0.254831 Objective Loss 0.254831 LR 0.000500 Time 0.038739 +2023-10-05 21:46:06,779 - Epoch: [136][ 60/ 1236] Overall Loss 0.254942 Objective Loss 0.254942 LR 0.000500 Time 0.035676 +2023-10-05 21:46:06,981 - Epoch: [136][ 70/ 1236] Overall Loss 0.253443 Objective Loss 0.253443 LR 0.000500 Time 0.033460 +2023-10-05 21:46:07,184 - Epoch: [136][ 80/ 1236] Overall Loss 0.252327 Objective Loss 0.252327 LR 0.000500 Time 0.031810 +2023-10-05 21:46:07,386 - Epoch: [136][ 90/ 1236] Overall Loss 0.254672 Objective Loss 0.254672 LR 0.000500 Time 0.030515 +2023-10-05 21:46:07,589 - Epoch: [136][ 100/ 1236] Overall Loss 0.255744 Objective Loss 0.255744 LR 0.000500 Time 0.029490 +2023-10-05 21:46:07,790 - Epoch: [136][ 110/ 1236] Overall Loss 0.254511 Objective Loss 0.254511 LR 0.000500 Time 0.028637 +2023-10-05 21:46:07,993 - Epoch: [136][ 120/ 1236] Overall Loss 0.254842 Objective Loss 0.254842 LR 0.000500 Time 0.027938 +2023-10-05 21:46:08,195 - Epoch: [136][ 130/ 1236] Overall Loss 0.256472 Objective Loss 0.256472 LR 0.000500 Time 0.027338 +2023-10-05 21:46:08,398 - Epoch: [136][ 140/ 1236] Overall Loss 0.255698 Objective Loss 0.255698 LR 0.000500 Time 0.026837 +2023-10-05 21:46:08,601 - Epoch: [136][ 150/ 1236] Overall Loss 0.253436 Objective Loss 0.253436 LR 0.000500 Time 0.026396 +2023-10-05 21:46:08,806 - Epoch: [136][ 160/ 1236] Overall Loss 0.253710 Objective Loss 0.253710 LR 0.000500 Time 0.026026 +2023-10-05 21:46:09,007 - Epoch: [136][ 170/ 1236] Overall Loss 0.255070 Objective Loss 0.255070 LR 0.000500 Time 0.025677 +2023-10-05 21:46:09,211 - Epoch: [136][ 180/ 1236] Overall Loss 0.256093 Objective Loss 0.256093 LR 0.000500 Time 0.025379 +2023-10-05 21:46:09,412 - Epoch: [136][ 190/ 1236] Overall Loss 0.257589 Objective Loss 0.257589 LR 0.000500 Time 0.025101 +2023-10-05 21:46:09,616 - Epoch: [136][ 200/ 1236] Overall Loss 0.257158 Objective Loss 0.257158 LR 0.000500 Time 0.024862 +2023-10-05 21:46:09,817 - Epoch: [136][ 210/ 1236] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.000500 Time 0.024635 +2023-10-05 21:46:10,021 - Epoch: [136][ 220/ 1236] Overall Loss 0.258425 Objective Loss 0.258425 LR 0.000500 Time 0.024441 +2023-10-05 21:46:10,223 - Epoch: [136][ 230/ 1236] Overall Loss 0.259444 Objective Loss 0.259444 LR 0.000500 Time 0.024253 +2023-10-05 21:46:10,427 - Epoch: [136][ 240/ 1236] Overall Loss 0.259692 Objective Loss 0.259692 LR 0.000500 Time 0.024091 +2023-10-05 21:46:10,628 - Epoch: [136][ 250/ 1236] Overall Loss 0.258583 Objective Loss 0.258583 LR 0.000500 Time 0.023932 +2023-10-05 21:46:10,832 - Epoch: [136][ 260/ 1236] Overall Loss 0.259136 Objective Loss 0.259136 LR 0.000500 Time 0.023794 +2023-10-05 21:46:11,033 - Epoch: [136][ 270/ 1236] Overall Loss 0.258291 Objective Loss 0.258291 LR 0.000500 Time 0.023658 +2023-10-05 21:46:11,238 - Epoch: [136][ 280/ 1236] Overall Loss 0.258270 Objective Loss 0.258270 LR 0.000500 Time 0.023541 +2023-10-05 21:46:11,439 - Epoch: [136][ 290/ 1236] Overall Loss 0.258226 Objective Loss 0.258226 LR 0.000500 Time 0.023422 +2023-10-05 21:46:11,643 - Epoch: [136][ 300/ 1236] Overall Loss 0.258393 Objective Loss 0.258393 LR 0.000500 Time 0.023319 +2023-10-05 21:46:11,845 - Epoch: [136][ 310/ 1236] Overall Loss 0.258461 Objective Loss 0.258461 LR 0.000500 Time 0.023216 +2023-10-05 21:46:12,048 - Epoch: [136][ 320/ 1236] Overall Loss 0.259828 Objective Loss 0.259828 LR 0.000500 Time 0.023126 +2023-10-05 21:46:12,250 - Epoch: [136][ 330/ 1236] Overall Loss 0.259700 Objective Loss 0.259700 LR 0.000500 Time 0.023034 +2023-10-05 21:46:12,453 - Epoch: [136][ 340/ 1236] Overall Loss 0.259873 Objective Loss 0.259873 LR 0.000500 Time 0.022954 +2023-10-05 21:46:12,655 - Epoch: [136][ 350/ 1236] Overall Loss 0.259635 Objective Loss 0.259635 LR 0.000500 Time 0.022873 +2023-10-05 21:46:12,859 - Epoch: [136][ 360/ 1236] Overall Loss 0.258757 Objective Loss 0.258757 LR 0.000500 Time 0.022803 +2023-10-05 21:46:13,061 - Epoch: [136][ 370/ 1236] Overall Loss 0.258884 Objective Loss 0.258884 LR 0.000500 Time 0.022733 +2023-10-05 21:46:13,266 - Epoch: [136][ 380/ 1236] Overall Loss 0.258958 Objective Loss 0.258958 LR 0.000500 Time 0.022673 +2023-10-05 21:46:13,469 - Epoch: [136][ 390/ 1236] Overall Loss 0.259025 Objective Loss 0.259025 LR 0.000500 Time 0.022613 +2023-10-05 21:46:13,676 - Epoch: [136][ 400/ 1236] Overall Loss 0.259417 Objective Loss 0.259417 LR 0.000500 Time 0.022563 +2023-10-05 21:46:13,880 - Epoch: [136][ 410/ 1236] Overall Loss 0.259103 Objective Loss 0.259103 LR 0.000500 Time 0.022508 +2023-10-05 21:46:14,085 - Epoch: [136][ 420/ 1236] Overall Loss 0.259400 Objective Loss 0.259400 LR 0.000500 Time 0.022460 +2023-10-05 21:46:14,288 - Epoch: [136][ 430/ 1236] Overall Loss 0.259384 Objective Loss 0.259384 LR 0.000500 Time 0.022410 +2023-10-05 21:46:14,493 - Epoch: [136][ 440/ 1236] Overall Loss 0.259394 Objective Loss 0.259394 LR 0.000500 Time 0.022366 +2023-10-05 21:46:14,697 - Epoch: [136][ 450/ 1236] Overall Loss 0.259391 Objective Loss 0.259391 LR 0.000500 Time 0.022320 +2023-10-05 21:46:14,904 - Epoch: [136][ 460/ 1236] Overall Loss 0.259344 Objective Loss 0.259344 LR 0.000500 Time 0.022285 +2023-10-05 21:46:15,107 - Epoch: [136][ 470/ 1236] Overall Loss 0.259374 Objective Loss 0.259374 LR 0.000500 Time 0.022241 +2023-10-05 21:46:15,312 - Epoch: [136][ 480/ 1236] Overall Loss 0.258620 Objective Loss 0.258620 LR 0.000500 Time 0.022204 +2023-10-05 21:46:15,514 - Epoch: [136][ 490/ 1236] Overall Loss 0.258881 Objective Loss 0.258881 LR 0.000500 Time 0.022164 +2023-10-05 21:46:15,719 - Epoch: [136][ 500/ 1236] Overall Loss 0.258496 Objective Loss 0.258496 LR 0.000500 Time 0.022129 +2023-10-05 21:46:15,922 - Epoch: [136][ 510/ 1236] Overall Loss 0.258576 Objective Loss 0.258576 LR 0.000500 Time 0.022093 +2023-10-05 21:46:16,127 - Epoch: [136][ 520/ 1236] Overall Loss 0.259038 Objective Loss 0.259038 LR 0.000500 Time 0.022061 +2023-10-05 21:46:16,330 - Epoch: [136][ 530/ 1236] Overall Loss 0.259236 Objective Loss 0.259236 LR 0.000500 Time 0.022027 +2023-10-05 21:46:16,535 - Epoch: [136][ 540/ 1236] Overall Loss 0.259153 Objective Loss 0.259153 LR 0.000500 Time 0.021998 +2023-10-05 21:46:16,737 - Epoch: [136][ 550/ 1236] Overall Loss 0.259409 Objective Loss 0.259409 LR 0.000500 Time 0.021965 +2023-10-05 21:46:16,942 - Epoch: [136][ 560/ 1236] Overall Loss 0.258849 Objective Loss 0.258849 LR 0.000500 Time 0.021939 +2023-10-05 21:46:17,145 - Epoch: [136][ 570/ 1236] Overall Loss 0.258664 Objective Loss 0.258664 LR 0.000500 Time 0.021908 +2023-10-05 21:46:17,350 - Epoch: [136][ 580/ 1236] Overall Loss 0.258736 Objective Loss 0.258736 LR 0.000500 Time 0.021884 +2023-10-05 21:46:17,552 - Epoch: [136][ 590/ 1236] Overall Loss 0.258907 Objective Loss 0.258907 LR 0.000500 Time 0.021855 +2023-10-05 21:46:17,757 - Epoch: [136][ 600/ 1236] Overall Loss 0.258493 Objective Loss 0.258493 LR 0.000500 Time 0.021832 +2023-10-05 21:46:17,959 - Epoch: [136][ 610/ 1236] Overall Loss 0.257996 Objective Loss 0.257996 LR 0.000500 Time 0.021805 +2023-10-05 21:46:18,164 - Epoch: [136][ 620/ 1236] Overall Loss 0.258065 Objective Loss 0.258065 LR 0.000500 Time 0.021783 +2023-10-05 21:46:18,367 - Epoch: [136][ 630/ 1236] Overall Loss 0.258323 Objective Loss 0.258323 LR 0.000500 Time 0.021759 +2023-10-05 21:46:18,572 - Epoch: [136][ 640/ 1236] Overall Loss 0.258398 Objective Loss 0.258398 LR 0.000500 Time 0.021738 +2023-10-05 21:46:18,774 - Epoch: [136][ 650/ 1236] Overall Loss 0.258389 Objective Loss 0.258389 LR 0.000500 Time 0.021714 +2023-10-05 21:46:18,984 - Epoch: [136][ 660/ 1236] Overall Loss 0.258246 Objective Loss 0.258246 LR 0.000500 Time 0.021703 +2023-10-05 21:46:19,203 - Epoch: [136][ 670/ 1236] Overall Loss 0.258216 Objective Loss 0.258216 LR 0.000500 Time 0.021706 +2023-10-05 21:46:19,418 - Epoch: [136][ 680/ 1236] Overall Loss 0.258188 Objective Loss 0.258188 LR 0.000500 Time 0.021702 +2023-10-05 21:46:19,634 - Epoch: [136][ 690/ 1236] Overall Loss 0.258586 Objective Loss 0.258586 LR 0.000500 Time 0.021700 +2023-10-05 21:46:19,847 - Epoch: [136][ 700/ 1236] Overall Loss 0.258982 Objective Loss 0.258982 LR 0.000500 Time 0.021693 +2023-10-05 21:46:20,068 - Epoch: [136][ 710/ 1236] Overall Loss 0.258660 Objective Loss 0.258660 LR 0.000500 Time 0.021699 +2023-10-05 21:46:20,282 - Epoch: [136][ 720/ 1236] Overall Loss 0.258505 Objective Loss 0.258505 LR 0.000500 Time 0.021694 +2023-10-05 21:46:20,499 - Epoch: [136][ 730/ 1236] Overall Loss 0.258631 Objective Loss 0.258631 LR 0.000500 Time 0.021694 +2023-10-05 21:46:20,704 - Epoch: [136][ 740/ 1236] Overall Loss 0.258636 Objective Loss 0.258636 LR 0.000500 Time 0.021678 +2023-10-05 21:46:20,906 - Epoch: [136][ 750/ 1236] Overall Loss 0.258802 Objective Loss 0.258802 LR 0.000500 Time 0.021658 +2023-10-05 21:46:21,109 - Epoch: [136][ 760/ 1236] Overall Loss 0.258677 Objective Loss 0.258677 LR 0.000500 Time 0.021639 +2023-10-05 21:46:21,311 - Epoch: [136][ 770/ 1236] Overall Loss 0.258557 Objective Loss 0.258557 LR 0.000500 Time 0.021619 +2023-10-05 21:46:21,514 - Epoch: [136][ 780/ 1236] Overall Loss 0.258393 Objective Loss 0.258393 LR 0.000500 Time 0.021602 +2023-10-05 21:46:21,716 - Epoch: [136][ 790/ 1236] Overall Loss 0.258151 Objective Loss 0.258151 LR 0.000500 Time 0.021584 +2023-10-05 21:46:21,919 - Epoch: [136][ 800/ 1236] Overall Loss 0.257901 Objective Loss 0.257901 LR 0.000500 Time 0.021567 +2023-10-05 21:46:22,121 - Epoch: [136][ 810/ 1236] Overall Loss 0.258014 Objective Loss 0.258014 LR 0.000500 Time 0.021550 +2023-10-05 21:46:22,324 - Epoch: [136][ 820/ 1236] Overall Loss 0.258037 Objective Loss 0.258037 LR 0.000500 Time 0.021534 +2023-10-05 21:46:22,526 - Epoch: [136][ 830/ 1236] Overall Loss 0.257769 Objective Loss 0.257769 LR 0.000500 Time 0.021517 +2023-10-05 21:46:22,728 - Epoch: [136][ 840/ 1236] Overall Loss 0.257818 Objective Loss 0.257818 LR 0.000500 Time 0.021502 +2023-10-05 21:46:22,930 - Epoch: [136][ 850/ 1236] Overall Loss 0.257683 Objective Loss 0.257683 LR 0.000500 Time 0.021486 +2023-10-05 21:46:23,133 - Epoch: [136][ 860/ 1236] Overall Loss 0.257684 Objective Loss 0.257684 LR 0.000500 Time 0.021472 +2023-10-05 21:46:23,336 - Epoch: [136][ 870/ 1236] Overall Loss 0.258034 Objective Loss 0.258034 LR 0.000500 Time 0.021457 +2023-10-05 21:46:23,540 - Epoch: [136][ 880/ 1236] Overall Loss 0.258157 Objective Loss 0.258157 LR 0.000500 Time 0.021445 +2023-10-05 21:46:23,742 - Epoch: [136][ 890/ 1236] Overall Loss 0.258201 Objective Loss 0.258201 LR 0.000500 Time 0.021431 +2023-10-05 21:46:23,945 - Epoch: [136][ 900/ 1236] Overall Loss 0.258419 Objective Loss 0.258419 LR 0.000500 Time 0.021418 +2023-10-05 21:46:24,148 - Epoch: [136][ 910/ 1236] Overall Loss 0.258966 Objective Loss 0.258966 LR 0.000500 Time 0.021405 +2023-10-05 21:46:24,351 - Epoch: [136][ 920/ 1236] Overall Loss 0.258835 Objective Loss 0.258835 LR 0.000500 Time 0.021393 +2023-10-05 21:46:24,554 - Epoch: [136][ 930/ 1236] Overall Loss 0.258735 Objective Loss 0.258735 LR 0.000500 Time 0.021380 +2023-10-05 21:46:24,757 - Epoch: [136][ 940/ 1236] Overall Loss 0.258791 Objective Loss 0.258791 LR 0.000500 Time 0.021369 +2023-10-05 21:46:24,962 - Epoch: [136][ 950/ 1236] Overall Loss 0.258718 Objective Loss 0.258718 LR 0.000500 Time 0.021358 +2023-10-05 21:46:25,165 - Epoch: [136][ 960/ 1236] Overall Loss 0.258592 Objective Loss 0.258592 LR 0.000500 Time 0.021347 +2023-10-05 21:46:25,368 - Epoch: [136][ 970/ 1236] Overall Loss 0.258733 Objective Loss 0.258733 LR 0.000500 Time 0.021335 +2023-10-05 21:46:25,571 - Epoch: [136][ 980/ 1236] Overall Loss 0.258683 Objective Loss 0.258683 LR 0.000500 Time 0.021325 +2023-10-05 21:46:25,775 - Epoch: [136][ 990/ 1236] Overall Loss 0.258715 Objective Loss 0.258715 LR 0.000500 Time 0.021315 +2023-10-05 21:46:25,981 - Epoch: [136][ 1000/ 1236] Overall Loss 0.258630 Objective Loss 0.258630 LR 0.000500 Time 0.021308 +2023-10-05 21:46:26,184 - Epoch: [136][ 1010/ 1236] Overall Loss 0.258775 Objective Loss 0.258775 LR 0.000500 Time 0.021298 +2023-10-05 21:46:26,392 - Epoch: [136][ 1020/ 1236] Overall Loss 0.258823 Objective Loss 0.258823 LR 0.000500 Time 0.021292 +2023-10-05 21:46:26,598 - Epoch: [136][ 1030/ 1236] Overall Loss 0.258567 Objective Loss 0.258567 LR 0.000500 Time 0.021284 +2023-10-05 21:46:26,803 - Epoch: [136][ 1040/ 1236] Overall Loss 0.258574 Objective Loss 0.258574 LR 0.000500 Time 0.021277 +2023-10-05 21:46:27,010 - Epoch: [136][ 1050/ 1236] Overall Loss 0.258174 Objective Loss 0.258174 LR 0.000500 Time 0.021270 +2023-10-05 21:46:27,216 - Epoch: [136][ 1060/ 1236] Overall Loss 0.258210 Objective Loss 0.258210 LR 0.000500 Time 0.021264 +2023-10-05 21:46:27,422 - Epoch: [136][ 1070/ 1236] Overall Loss 0.258340 Objective Loss 0.258340 LR 0.000500 Time 0.021258 +2023-10-05 21:46:27,628 - Epoch: [136][ 1080/ 1236] Overall Loss 0.258452 Objective Loss 0.258452 LR 0.000500 Time 0.021251 +2023-10-05 21:46:27,833 - Epoch: [136][ 1090/ 1236] Overall Loss 0.258290 Objective Loss 0.258290 LR 0.000500 Time 0.021244 +2023-10-05 21:46:28,039 - Epoch: [136][ 1100/ 1236] Overall Loss 0.258243 Objective Loss 0.258243 LR 0.000500 Time 0.021238 +2023-10-05 21:46:28,246 - Epoch: [136][ 1110/ 1236] Overall Loss 0.258305 Objective Loss 0.258305 LR 0.000500 Time 0.021232 +2023-10-05 21:46:28,452 - Epoch: [136][ 1120/ 1236] Overall Loss 0.258369 Objective Loss 0.258369 LR 0.000500 Time 0.021226 +2023-10-05 21:46:28,658 - Epoch: [136][ 1130/ 1236] Overall Loss 0.258379 Objective Loss 0.258379 LR 0.000500 Time 0.021220 +2023-10-05 21:46:28,870 - Epoch: [136][ 1140/ 1236] Overall Loss 0.258477 Objective Loss 0.258477 LR 0.000500 Time 0.021219 +2023-10-05 21:46:29,077 - Epoch: [136][ 1150/ 1236] Overall Loss 0.258454 Objective Loss 0.258454 LR 0.000500 Time 0.021215 +2023-10-05 21:46:29,283 - Epoch: [136][ 1160/ 1236] Overall Loss 0.258437 Objective Loss 0.258437 LR 0.000500 Time 0.021209 +2023-10-05 21:46:29,488 - Epoch: [136][ 1170/ 1236] Overall Loss 0.258434 Objective Loss 0.258434 LR 0.000500 Time 0.021203 +2023-10-05 21:46:29,697 - Epoch: [136][ 1180/ 1236] Overall Loss 0.258612 Objective Loss 0.258612 LR 0.000500 Time 0.021200 +2023-10-05 21:46:29,904 - Epoch: [136][ 1190/ 1236] Overall Loss 0.258569 Objective Loss 0.258569 LR 0.000500 Time 0.021195 +2023-10-05 21:46:30,110 - Epoch: [136][ 1200/ 1236] Overall Loss 0.258503 Objective Loss 0.258503 LR 0.000500 Time 0.021190 +2023-10-05 21:46:30,315 - Epoch: [136][ 1210/ 1236] Overall Loss 0.258446 Objective Loss 0.258446 LR 0.000500 Time 0.021184 +2023-10-05 21:46:30,521 - Epoch: [136][ 1220/ 1236] Overall Loss 0.258363 Objective Loss 0.258363 LR 0.000500 Time 0.021179 +2023-10-05 21:46:30,782 - Epoch: [136][ 1230/ 1236] Overall Loss 0.258376 Objective Loss 0.258376 LR 0.000500 Time 0.021219 +2023-10-05 21:46:30,902 - Epoch: [136][ 1236/ 1236] Overall Loss 0.258339 Objective Loss 0.258339 Top1 87.983707 Top5 97.963340 LR 0.000500 Time 0.021212 +2023-10-05 21:46:31,032 - --- validate (epoch=136)----------- +2023-10-05 21:46:31,032 - 29943 samples (256 per mini-batch) +2023-10-05 21:46:31,520 - Epoch: [136][ 10/ 117] Loss 0.353422 Top1 82.421875 Top5 97.890625 +2023-10-05 21:46:31,680 - Epoch: [136][ 20/ 117] Loss 0.345076 Top1 83.046875 Top5 97.929688 +2023-10-05 21:46:31,839 - Epoch: [136][ 30/ 117] Loss 0.344578 Top1 83.046875 Top5 97.838542 +2023-10-05 21:46:32,003 - Epoch: [136][ 40/ 117] Loss 0.346651 Top1 83.056641 Top5 97.861328 +2023-10-05 21:46:32,158 - Epoch: [136][ 50/ 117] Loss 0.340819 Top1 83.156250 Top5 97.867188 +2023-10-05 21:46:32,316 - Epoch: [136][ 60/ 117] Loss 0.335944 Top1 83.268229 Top5 97.871094 +2023-10-05 21:46:32,477 - Epoch: [136][ 70/ 117] Loss 0.334202 Top1 83.325893 Top5 97.907366 +2023-10-05 21:46:32,641 - Epoch: [136][ 80/ 117] Loss 0.338057 Top1 83.383789 Top5 97.846680 +2023-10-05 21:46:32,802 - Epoch: [136][ 90/ 117] Loss 0.334873 Top1 83.467882 Top5 97.877604 +2023-10-05 21:46:32,966 - Epoch: [136][ 100/ 117] Loss 0.334702 Top1 83.496094 Top5 97.878906 +2023-10-05 21:46:33,133 - Epoch: [136][ 110/ 117] Loss 0.333642 Top1 83.458807 Top5 97.901278 +2023-10-05 21:46:33,221 - Epoch: [136][ 117/ 117] Loss 0.332704 Top1 83.522025 Top5 97.906021 +2023-10-05 21:46:33,339 - ==> Top1: 83.522 Top5: 97.906 Loss: 0.333 + +2023-10-05 21:46:33,340 - ==> Confusion: +[[ 931 4 4 1 7 0 0 0 7 69 2 0 2 2 5 3 1 0 0 0 12] + [ 2 1046 3 0 3 19 1 23 1 1 3 0 0 0 0 4 5 0 10 1 9] + [ 5 2 969 8 2 1 17 9 0 1 4 5 8 2 1 3 0 1 7 1 10] + [ 1 0 19 940 0 3 0 1 2 1 14 1 8 1 27 7 0 7 37 0 20] + [ 25 14 1 0 959 1 0 1 1 8 1 1 2 3 4 6 16 1 0 0 6] + [ 4 35 0 1 3 983 1 23 1 1 4 8 2 15 5 2 5 0 2 4 17] + [ 0 5 26 0 0 0 1119 8 0 0 1 4 2 0 0 9 0 0 4 6 7] + [ 3 15 7 0 3 27 3 1077 1 3 3 11 5 1 0 2 0 0 43 9 5] + [ 18 1 0 0 0 1 0 0 978 43 13 1 3 10 8 6 2 0 3 1 1] + [ 115 0 1 0 3 4 1 0 40 906 2 3 0 23 3 7 0 3 0 0 8] + [ 4 7 11 2 0 1 5 1 8 3 969 2 3 12 3 2 1 0 6 2 11] + [ 3 0 2 0 0 15 0 5 0 1 0 946 22 7 0 4 0 15 0 11 4] + [ 1 2 4 3 0 1 0 1 1 0 1 31 982 4 0 8 2 11 4 3 9] + [ 3 0 1 1 1 5 0 0 12 12 9 4 1 1054 3 2 3 0 0 2 6] + [ 15 4 3 7 2 1 1 0 38 4 4 0 4 1 987 0 1 1 15 0 13] + [ 1 2 1 0 2 1 0 0 0 0 0 9 4 1 1 1079 14 7 1 7 4] + [ 1 16 1 0 3 4 1 2 1 0 0 2 0 0 2 9 1102 1 0 4 12] + [ 0 0 0 1 0 1 1 0 0 1 0 7 17 0 1 10 0 988 3 2 6] + [ 1 5 9 13 1 1 1 21 2 0 2 0 0 0 4 0 1 0 997 2 8] + [ 0 3 2 1 2 5 12 7 2 0 1 18 4 0 0 3 10 1 2 1073 6] + [ 143 216 148 41 98 150 56 102 113 72 221 135 362 324 126 75 133 64 168 234 4924]] + +2023-10-05 21:46:33,342 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:46:33,342 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:46:33,348 - + +2023-10-05 21:46:33,348 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:46:34,385 - Epoch: [137][ 10/ 1236] Overall Loss 0.244231 Objective Loss 0.244231 LR 0.000500 Time 0.103724 +2023-10-05 21:46:34,590 - Epoch: [137][ 20/ 1236] Overall Loss 0.245515 Objective Loss 0.245515 LR 0.000500 Time 0.062063 +2023-10-05 21:46:34,794 - Epoch: [137][ 30/ 1236] Overall Loss 0.252230 Objective Loss 0.252230 LR 0.000500 Time 0.048155 +2023-10-05 21:46:34,997 - Epoch: [137][ 40/ 1236] Overall Loss 0.254420 Objective Loss 0.254420 LR 0.000500 Time 0.041195 +2023-10-05 21:46:35,201 - Epoch: [137][ 50/ 1236] Overall Loss 0.256974 Objective Loss 0.256974 LR 0.000500 Time 0.037036 +2023-10-05 21:46:35,407 - Epoch: [137][ 60/ 1236] Overall Loss 0.261966 Objective Loss 0.261966 LR 0.000500 Time 0.034285 +2023-10-05 21:46:35,612 - Epoch: [137][ 70/ 1236] Overall Loss 0.258180 Objective Loss 0.258180 LR 0.000500 Time 0.032312 +2023-10-05 21:46:35,818 - Epoch: [137][ 80/ 1236] Overall Loss 0.257164 Objective Loss 0.257164 LR 0.000500 Time 0.030843 +2023-10-05 21:46:36,023 - Epoch: [137][ 90/ 1236] Overall Loss 0.255577 Objective Loss 0.255577 LR 0.000500 Time 0.029688 +2023-10-05 21:46:36,229 - Epoch: [137][ 100/ 1236] Overall Loss 0.255820 Objective Loss 0.255820 LR 0.000500 Time 0.028775 +2023-10-05 21:46:36,434 - Epoch: [137][ 110/ 1236] Overall Loss 0.254221 Objective Loss 0.254221 LR 0.000500 Time 0.028020 +2023-10-05 21:46:36,640 - Epoch: [137][ 120/ 1236] Overall Loss 0.256348 Objective Loss 0.256348 LR 0.000500 Time 0.027398 +2023-10-05 21:46:36,849 - Epoch: [137][ 130/ 1236] Overall Loss 0.255913 Objective Loss 0.255913 LR 0.000500 Time 0.026890 +2023-10-05 21:46:37,064 - Epoch: [137][ 140/ 1236] Overall Loss 0.257424 Objective Loss 0.257424 LR 0.000500 Time 0.026503 +2023-10-05 21:46:37,282 - Epoch: [137][ 150/ 1236] Overall Loss 0.255633 Objective Loss 0.255633 LR 0.000500 Time 0.026190 +2023-10-05 21:46:37,501 - Epoch: [137][ 160/ 1236] Overall Loss 0.255802 Objective Loss 0.255802 LR 0.000500 Time 0.025919 +2023-10-05 21:46:37,713 - Epoch: [137][ 170/ 1236] Overall Loss 0.254771 Objective Loss 0.254771 LR 0.000500 Time 0.025637 +2023-10-05 21:46:37,921 - Epoch: [137][ 180/ 1236] Overall Loss 0.254498 Objective Loss 0.254498 LR 0.000500 Time 0.025365 +2023-10-05 21:46:38,128 - Epoch: [137][ 190/ 1236] Overall Loss 0.255143 Objective Loss 0.255143 LR 0.000500 Time 0.025120 +2023-10-05 21:46:38,336 - Epoch: [137][ 200/ 1236] Overall Loss 0.256064 Objective Loss 0.256064 LR 0.000500 Time 0.024900 +2023-10-05 21:46:38,543 - Epoch: [137][ 210/ 1236] Overall Loss 0.256332 Objective Loss 0.256332 LR 0.000500 Time 0.024698 +2023-10-05 21:46:38,751 - Epoch: [137][ 220/ 1236] Overall Loss 0.257289 Objective Loss 0.257289 LR 0.000500 Time 0.024516 +2023-10-05 21:46:38,959 - Epoch: [137][ 230/ 1236] Overall Loss 0.257323 Objective Loss 0.257323 LR 0.000500 Time 0.024355 +2023-10-05 21:46:39,170 - Epoch: [137][ 240/ 1236] Overall Loss 0.257041 Objective Loss 0.257041 LR 0.000500 Time 0.024217 +2023-10-05 21:46:39,379 - Epoch: [137][ 250/ 1236] Overall Loss 0.257348 Objective Loss 0.257348 LR 0.000500 Time 0.024083 +2023-10-05 21:46:39,591 - Epoch: [137][ 260/ 1236] Overall Loss 0.257186 Objective Loss 0.257186 LR 0.000500 Time 0.023969 +2023-10-05 21:46:39,800 - Epoch: [137][ 270/ 1236] Overall Loss 0.257787 Objective Loss 0.257787 LR 0.000500 Time 0.023855 +2023-10-05 21:46:40,016 - Epoch: [137][ 280/ 1236] Overall Loss 0.258559 Objective Loss 0.258559 LR 0.000500 Time 0.023773 +2023-10-05 21:46:40,230 - Epoch: [137][ 290/ 1236] Overall Loss 0.258447 Objective Loss 0.258447 LR 0.000500 Time 0.023688 +2023-10-05 21:46:40,459 - Epoch: [137][ 300/ 1236] Overall Loss 0.259244 Objective Loss 0.259244 LR 0.000500 Time 0.023661 +2023-10-05 21:46:40,686 - Epoch: [137][ 310/ 1236] Overall Loss 0.258167 Objective Loss 0.258167 LR 0.000500 Time 0.023628 +2023-10-05 21:46:40,911 - Epoch: [137][ 320/ 1236] Overall Loss 0.257234 Objective Loss 0.257234 LR 0.000500 Time 0.023593 +2023-10-05 21:46:41,127 - Epoch: [137][ 330/ 1236] Overall Loss 0.257537 Objective Loss 0.257537 LR 0.000500 Time 0.023529 +2023-10-05 21:46:41,345 - Epoch: [137][ 340/ 1236] Overall Loss 0.257948 Objective Loss 0.257948 LR 0.000500 Time 0.023477 +2023-10-05 21:46:41,550 - Epoch: [137][ 350/ 1236] Overall Loss 0.257700 Objective Loss 0.257700 LR 0.000500 Time 0.023390 +2023-10-05 21:46:41,755 - Epoch: [137][ 360/ 1236] Overall Loss 0.257218 Objective Loss 0.257218 LR 0.000500 Time 0.023310 +2023-10-05 21:46:41,962 - Epoch: [137][ 370/ 1236] Overall Loss 0.256988 Objective Loss 0.256988 LR 0.000500 Time 0.023238 +2023-10-05 21:46:42,172 - Epoch: [137][ 380/ 1236] Overall Loss 0.256567 Objective Loss 0.256567 LR 0.000500 Time 0.023177 +2023-10-05 21:46:42,376 - Epoch: [137][ 390/ 1236] Overall Loss 0.256682 Objective Loss 0.256682 LR 0.000500 Time 0.023105 +2023-10-05 21:46:42,588 - Epoch: [137][ 400/ 1236] Overall Loss 0.256019 Objective Loss 0.256019 LR 0.000500 Time 0.023057 +2023-10-05 21:46:42,800 - Epoch: [137][ 410/ 1236] Overall Loss 0.255620 Objective Loss 0.255620 LR 0.000500 Time 0.023011 +2023-10-05 21:46:43,019 - Epoch: [137][ 420/ 1236] Overall Loss 0.255016 Objective Loss 0.255016 LR 0.000500 Time 0.022982 +2023-10-05 21:46:43,229 - Epoch: [137][ 430/ 1236] Overall Loss 0.254393 Objective Loss 0.254393 LR 0.000500 Time 0.022934 +2023-10-05 21:46:43,441 - Epoch: [137][ 440/ 1236] Overall Loss 0.254287 Objective Loss 0.254287 LR 0.000500 Time 0.022896 +2023-10-05 21:46:43,654 - Epoch: [137][ 450/ 1236] Overall Loss 0.253781 Objective Loss 0.253781 LR 0.000500 Time 0.022858 +2023-10-05 21:46:43,870 - Epoch: [137][ 460/ 1236] Overall Loss 0.254069 Objective Loss 0.254069 LR 0.000500 Time 0.022830 +2023-10-05 21:46:44,100 - Epoch: [137][ 470/ 1236] Overall Loss 0.253891 Objective Loss 0.253891 LR 0.000500 Time 0.022832 +2023-10-05 21:46:44,317 - Epoch: [137][ 480/ 1236] Overall Loss 0.254081 Objective Loss 0.254081 LR 0.000500 Time 0.022808 +2023-10-05 21:46:44,528 - Epoch: [137][ 490/ 1236] Overall Loss 0.253904 Objective Loss 0.253904 LR 0.000500 Time 0.022773 +2023-10-05 21:46:44,743 - Epoch: [137][ 500/ 1236] Overall Loss 0.253998 Objective Loss 0.253998 LR 0.000500 Time 0.022746 +2023-10-05 21:46:44,954 - Epoch: [137][ 510/ 1236] Overall Loss 0.253632 Objective Loss 0.253632 LR 0.000500 Time 0.022713 +2023-10-05 21:46:45,169 - Epoch: [137][ 520/ 1236] Overall Loss 0.253676 Objective Loss 0.253676 LR 0.000500 Time 0.022689 +2023-10-05 21:46:45,381 - Epoch: [137][ 530/ 1236] Overall Loss 0.253460 Objective Loss 0.253460 LR 0.000500 Time 0.022659 +2023-10-05 21:46:45,595 - Epoch: [137][ 540/ 1236] Overall Loss 0.252933 Objective Loss 0.252933 LR 0.000500 Time 0.022635 +2023-10-05 21:46:45,806 - Epoch: [137][ 550/ 1236] Overall Loss 0.253329 Objective Loss 0.253329 LR 0.000500 Time 0.022607 +2023-10-05 21:46:46,020 - Epoch: [137][ 560/ 1236] Overall Loss 0.253551 Objective Loss 0.253551 LR 0.000500 Time 0.022585 +2023-10-05 21:46:46,233 - Epoch: [137][ 570/ 1236] Overall Loss 0.253733 Objective Loss 0.253733 LR 0.000500 Time 0.022562 +2023-10-05 21:46:46,447 - Epoch: [137][ 580/ 1236] Overall Loss 0.253474 Objective Loss 0.253474 LR 0.000500 Time 0.022541 +2023-10-05 21:46:46,656 - Epoch: [137][ 590/ 1236] Overall Loss 0.253324 Objective Loss 0.253324 LR 0.000500 Time 0.022512 +2023-10-05 21:46:46,870 - Epoch: [137][ 600/ 1236] Overall Loss 0.253225 Objective Loss 0.253225 LR 0.000500 Time 0.022493 +2023-10-05 21:46:47,083 - Epoch: [137][ 610/ 1236] Overall Loss 0.253174 Objective Loss 0.253174 LR 0.000500 Time 0.022472 +2023-10-05 21:46:47,298 - Epoch: [137][ 620/ 1236] Overall Loss 0.253507 Objective Loss 0.253507 LR 0.000500 Time 0.022456 +2023-10-05 21:46:47,506 - Epoch: [137][ 630/ 1236] Overall Loss 0.253112 Objective Loss 0.253112 LR 0.000500 Time 0.022430 +2023-10-05 21:46:47,721 - Epoch: [137][ 640/ 1236] Overall Loss 0.253006 Objective Loss 0.253006 LR 0.000500 Time 0.022415 +2023-10-05 21:46:47,933 - Epoch: [137][ 650/ 1236] Overall Loss 0.253132 Objective Loss 0.253132 LR 0.000500 Time 0.022395 +2023-10-05 21:46:48,149 - Epoch: [137][ 660/ 1236] Overall Loss 0.252781 Objective Loss 0.252781 LR 0.000500 Time 0.022382 +2023-10-05 21:46:48,361 - Epoch: [137][ 670/ 1236] Overall Loss 0.252745 Objective Loss 0.252745 LR 0.000500 Time 0.022364 +2023-10-05 21:46:48,576 - Epoch: [137][ 680/ 1236] Overall Loss 0.253051 Objective Loss 0.253051 LR 0.000500 Time 0.022351 +2023-10-05 21:46:48,788 - Epoch: [137][ 690/ 1236] Overall Loss 0.252956 Objective Loss 0.252956 LR 0.000500 Time 0.022333 +2023-10-05 21:46:49,003 - Epoch: [137][ 700/ 1236] Overall Loss 0.252973 Objective Loss 0.252973 LR 0.000500 Time 0.022321 +2023-10-05 21:46:49,215 - Epoch: [137][ 710/ 1236] Overall Loss 0.252805 Objective Loss 0.252805 LR 0.000500 Time 0.022304 +2023-10-05 21:46:49,430 - Epoch: [137][ 720/ 1236] Overall Loss 0.252635 Objective Loss 0.252635 LR 0.000500 Time 0.022293 +2023-10-05 21:46:49,642 - Epoch: [137][ 730/ 1236] Overall Loss 0.252931 Objective Loss 0.252931 LR 0.000500 Time 0.022276 +2023-10-05 21:46:49,857 - Epoch: [137][ 740/ 1236] Overall Loss 0.253534 Objective Loss 0.253534 LR 0.000500 Time 0.022265 +2023-10-05 21:46:50,068 - Epoch: [137][ 750/ 1236] Overall Loss 0.253691 Objective Loss 0.253691 LR 0.000500 Time 0.022250 +2023-10-05 21:46:50,283 - Epoch: [137][ 760/ 1236] Overall Loss 0.253758 Objective Loss 0.253758 LR 0.000500 Time 0.022240 +2023-10-05 21:46:50,495 - Epoch: [137][ 770/ 1236] Overall Loss 0.254156 Objective Loss 0.254156 LR 0.000500 Time 0.022225 +2023-10-05 21:46:50,710 - Epoch: [137][ 780/ 1236] Overall Loss 0.254243 Objective Loss 0.254243 LR 0.000500 Time 0.022216 +2023-10-05 21:46:50,922 - Epoch: [137][ 790/ 1236] Overall Loss 0.254054 Objective Loss 0.254054 LR 0.000500 Time 0.022201 +2023-10-05 21:46:51,137 - Epoch: [137][ 800/ 1236] Overall Loss 0.253944 Objective Loss 0.253944 LR 0.000500 Time 0.022193 +2023-10-05 21:46:51,348 - Epoch: [137][ 810/ 1236] Overall Loss 0.253795 Objective Loss 0.253795 LR 0.000500 Time 0.022179 +2023-10-05 21:46:51,563 - Epoch: [137][ 820/ 1236] Overall Loss 0.254082 Objective Loss 0.254082 LR 0.000500 Time 0.022170 +2023-10-05 21:46:51,775 - Epoch: [137][ 830/ 1236] Overall Loss 0.254362 Objective Loss 0.254362 LR 0.000500 Time 0.022157 +2023-10-05 21:46:51,990 - Epoch: [137][ 840/ 1236] Overall Loss 0.254241 Objective Loss 0.254241 LR 0.000500 Time 0.022149 +2023-10-05 21:46:52,201 - Epoch: [137][ 850/ 1236] Overall Loss 0.254164 Objective Loss 0.254164 LR 0.000500 Time 0.022137 +2023-10-05 21:46:52,416 - Epoch: [137][ 860/ 1236] Overall Loss 0.254121 Objective Loss 0.254121 LR 0.000500 Time 0.022129 +2023-10-05 21:46:52,628 - Epoch: [137][ 870/ 1236] Overall Loss 0.254106 Objective Loss 0.254106 LR 0.000500 Time 0.022117 +2023-10-05 21:46:52,842 - Epoch: [137][ 880/ 1236] Overall Loss 0.253919 Objective Loss 0.253919 LR 0.000500 Time 0.022109 +2023-10-05 21:46:53,054 - Epoch: [137][ 890/ 1236] Overall Loss 0.253904 Objective Loss 0.253904 LR 0.000500 Time 0.022098 +2023-10-05 21:46:53,269 - Epoch: [137][ 900/ 1236] Overall Loss 0.253796 Objective Loss 0.253796 LR 0.000500 Time 0.022091 +2023-10-05 21:46:53,480 - Epoch: [137][ 910/ 1236] Overall Loss 0.253663 Objective Loss 0.253663 LR 0.000500 Time 0.022080 +2023-10-05 21:46:53,695 - Epoch: [137][ 920/ 1236] Overall Loss 0.254204 Objective Loss 0.254204 LR 0.000500 Time 0.022073 +2023-10-05 21:46:53,907 - Epoch: [137][ 930/ 1236] Overall Loss 0.254302 Objective Loss 0.254302 LR 0.000500 Time 0.022063 +2023-10-05 21:46:54,122 - Epoch: [137][ 940/ 1236] Overall Loss 0.254364 Objective Loss 0.254364 LR 0.000500 Time 0.022056 +2023-10-05 21:46:54,333 - Epoch: [137][ 950/ 1236] Overall Loss 0.254176 Objective Loss 0.254176 LR 0.000500 Time 0.022046 +2023-10-05 21:46:54,548 - Epoch: [137][ 960/ 1236] Overall Loss 0.254305 Objective Loss 0.254305 LR 0.000500 Time 0.022040 +2023-10-05 21:46:54,760 - Epoch: [137][ 970/ 1236] Overall Loss 0.254308 Objective Loss 0.254308 LR 0.000500 Time 0.022030 +2023-10-05 21:46:54,975 - Epoch: [137][ 980/ 1236] Overall Loss 0.254253 Objective Loss 0.254253 LR 0.000500 Time 0.022025 +2023-10-05 21:46:55,183 - Epoch: [137][ 990/ 1236] Overall Loss 0.254279 Objective Loss 0.254279 LR 0.000500 Time 0.022012 +2023-10-05 21:46:55,397 - Epoch: [137][ 1000/ 1236] Overall Loss 0.254131 Objective Loss 0.254131 LR 0.000500 Time 0.022006 +2023-10-05 21:46:55,608 - Epoch: [137][ 1010/ 1236] Overall Loss 0.254045 Objective Loss 0.254045 LR 0.000500 Time 0.021996 +2023-10-05 21:46:55,822 - Epoch: [137][ 1020/ 1236] Overall Loss 0.253947 Objective Loss 0.253947 LR 0.000500 Time 0.021990 +2023-10-05 21:46:56,033 - Epoch: [137][ 1030/ 1236] Overall Loss 0.253925 Objective Loss 0.253925 LR 0.000500 Time 0.021981 +2023-10-05 21:46:56,247 - Epoch: [137][ 1040/ 1236] Overall Loss 0.253829 Objective Loss 0.253829 LR 0.000500 Time 0.021975 +2023-10-05 21:46:56,459 - Epoch: [137][ 1050/ 1236] Overall Loss 0.253827 Objective Loss 0.253827 LR 0.000500 Time 0.021967 +2023-10-05 21:46:56,673 - Epoch: [137][ 1060/ 1236] Overall Loss 0.253788 Objective Loss 0.253788 LR 0.000500 Time 0.021961 +2023-10-05 21:46:56,884 - Epoch: [137][ 1070/ 1236] Overall Loss 0.254009 Objective Loss 0.254009 LR 0.000500 Time 0.021952 +2023-10-05 21:46:57,098 - Epoch: [137][ 1080/ 1236] Overall Loss 0.254029 Objective Loss 0.254029 LR 0.000500 Time 0.021947 +2023-10-05 21:46:57,309 - Epoch: [137][ 1090/ 1236] Overall Loss 0.254148 Objective Loss 0.254148 LR 0.000500 Time 0.021939 +2023-10-05 21:46:57,523 - Epoch: [137][ 1100/ 1236] Overall Loss 0.254122 Objective Loss 0.254122 LR 0.000500 Time 0.021934 +2023-10-05 21:46:57,733 - Epoch: [137][ 1110/ 1236] Overall Loss 0.254232 Objective Loss 0.254232 LR 0.000500 Time 0.021925 +2023-10-05 21:46:57,948 - Epoch: [137][ 1120/ 1236] Overall Loss 0.254260 Objective Loss 0.254260 LR 0.000500 Time 0.021921 +2023-10-05 21:46:58,162 - Epoch: [137][ 1130/ 1236] Overall Loss 0.254271 Objective Loss 0.254271 LR 0.000500 Time 0.021916 +2023-10-05 21:46:58,374 - Epoch: [137][ 1140/ 1236] Overall Loss 0.254361 Objective Loss 0.254361 LR 0.000500 Time 0.021909 +2023-10-05 21:46:58,589 - Epoch: [137][ 1150/ 1236] Overall Loss 0.254444 Objective Loss 0.254444 LR 0.000500 Time 0.021905 +2023-10-05 21:46:58,800 - Epoch: [137][ 1160/ 1236] Overall Loss 0.254453 Objective Loss 0.254453 LR 0.000500 Time 0.021898 +2023-10-05 21:46:59,014 - Epoch: [137][ 1170/ 1236] Overall Loss 0.254469 Objective Loss 0.254469 LR 0.000500 Time 0.021893 +2023-10-05 21:46:59,225 - Epoch: [137][ 1180/ 1236] Overall Loss 0.254511 Objective Loss 0.254511 LR 0.000500 Time 0.021886 +2023-10-05 21:46:59,436 - Epoch: [137][ 1190/ 1236] Overall Loss 0.254354 Objective Loss 0.254354 LR 0.000500 Time 0.021879 +2023-10-05 21:46:59,648 - Epoch: [137][ 1200/ 1236] Overall Loss 0.254494 Objective Loss 0.254494 LR 0.000500 Time 0.021873 +2023-10-05 21:46:59,862 - Epoch: [137][ 1210/ 1236] Overall Loss 0.254349 Objective Loss 0.254349 LR 0.000500 Time 0.021868 +2023-10-05 21:47:00,073 - Epoch: [137][ 1220/ 1236] Overall Loss 0.254486 Objective Loss 0.254486 LR 0.000500 Time 0.021862 +2023-10-05 21:47:00,334 - Epoch: [137][ 1230/ 1236] Overall Loss 0.254446 Objective Loss 0.254446 LR 0.000500 Time 0.021896 +2023-10-05 21:47:00,451 - Epoch: [137][ 1236/ 1236] Overall Loss 0.254562 Objective Loss 0.254562 Top1 86.150713 Top5 98.574338 LR 0.000500 Time 0.021885 +2023-10-05 21:47:00,576 - --- validate (epoch=137)----------- +2023-10-05 21:47:00,576 - 29943 samples (256 per mini-batch) +2023-10-05 21:47:01,036 - Epoch: [137][ 10/ 117] Loss 0.373230 Top1 82.382812 Top5 97.578125 +2023-10-05 21:47:01,194 - Epoch: [137][ 20/ 117] Loss 0.357585 Top1 82.656250 Top5 97.519531 +2023-10-05 21:47:01,354 - Epoch: [137][ 30/ 117] Loss 0.351286 Top1 82.864583 Top5 97.669271 +2023-10-05 21:47:01,517 - Epoch: [137][ 40/ 117] Loss 0.346083 Top1 82.822266 Top5 97.587891 +2023-10-05 21:47:01,676 - Epoch: [137][ 50/ 117] Loss 0.341160 Top1 82.750000 Top5 97.593750 +2023-10-05 21:47:01,838 - Epoch: [137][ 60/ 117] Loss 0.332500 Top1 82.845052 Top5 97.695312 +2023-10-05 21:47:01,988 - Epoch: [137][ 70/ 117] Loss 0.329703 Top1 82.739955 Top5 97.678571 +2023-10-05 21:47:02,135 - Epoch: [137][ 80/ 117] Loss 0.331478 Top1 82.646484 Top5 97.666016 +2023-10-05 21:47:02,291 - Epoch: [137][ 90/ 117] Loss 0.327749 Top1 82.764757 Top5 97.656250 +2023-10-05 21:47:02,447 - Epoch: [137][ 100/ 117] Loss 0.330742 Top1 82.726562 Top5 97.683594 +2023-10-05 21:47:02,612 - Epoch: [137][ 110/ 117] Loss 0.330756 Top1 82.766335 Top5 97.674006 +2023-10-05 21:47:02,698 - Epoch: [137][ 117/ 117] Loss 0.333354 Top1 82.687106 Top5 97.645527 +2023-10-05 21:47:02,842 - ==> Top1: 82.687 Top5: 97.646 Loss: 0.333 + +2023-10-05 21:47:02,843 - ==> Confusion: +[[ 907 3 4 1 14 2 0 0 4 82 0 0 2 3 7 0 4 2 3 0 12] + [ 2 1043 2 0 13 21 1 21 1 0 1 0 0 1 0 3 3 0 11 2 6] + [ 5 0 963 6 4 1 28 4 0 2 6 1 8 3 2 4 1 1 7 6 4] + [ 3 1 23 936 2 3 0 2 3 1 13 1 15 2 30 3 1 11 22 3 14] + [ 18 7 0 0 970 4 0 0 2 9 0 4 1 2 9 4 11 2 0 2 5] + [ 5 41 1 0 2 981 1 27 2 3 4 12 0 11 3 1 4 0 2 6 10] + [ 0 9 20 0 1 0 1123 6 0 1 0 3 2 0 0 5 1 3 1 12 4] + [ 2 19 18 0 5 28 4 1055 1 5 2 16 2 1 1 2 0 0 42 11 4] + [ 14 3 1 0 0 3 0 0 967 54 11 0 3 11 13 5 1 0 1 0 2] + [ 91 0 2 1 5 2 0 0 20 951 0 2 0 24 7 7 0 0 0 2 5] + [ 2 4 10 2 1 0 4 4 10 2 966 3 1 20 2 1 2 1 6 2 10] + [ 3 1 0 0 0 10 0 2 0 0 0 981 13 6 0 1 1 13 0 4 0] + [ 0 1 1 2 0 2 0 1 0 0 1 53 980 3 0 5 2 11 1 2 3] + [ 2 1 0 0 3 4 0 0 12 16 3 7 2 1054 2 1 0 2 0 2 8] + [ 10 3 4 8 6 1 0 0 32 2 4 1 5 1 1002 0 0 2 10 0 10] + [ 0 3 3 0 5 0 1 0 0 2 0 11 10 0 1 1065 14 6 0 10 3] + [ 1 13 1 0 5 5 0 0 1 0 0 8 2 2 3 9 1101 0 0 4 6] + [ 0 0 0 1 0 3 1 0 0 0 0 5 24 0 0 5 0 993 1 2 3] + [ 0 7 10 14 2 0 0 25 2 0 4 5 5 1 10 0 2 0 968 2 11] + [ 1 3 1 0 1 3 8 8 1 0 2 17 3 4 0 4 9 1 3 1080 3] + [ 140 203 162 48 116 149 53 88 124 105 206 181 418 348 170 58 172 76 159 256 4673]] + +2023-10-05 21:47:02,844 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:47:02,844 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:47:02,850 - + +2023-10-05 21:47:02,851 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:47:03,859 - Epoch: [138][ 10/ 1236] Overall Loss 0.232987 Objective Loss 0.232987 LR 0.000500 Time 0.100746 +2023-10-05 21:47:04,063 - Epoch: [138][ 20/ 1236] Overall Loss 0.252095 Objective Loss 0.252095 LR 0.000500 Time 0.060569 +2023-10-05 21:47:04,266 - Epoch: [138][ 30/ 1236] Overall Loss 0.252109 Objective Loss 0.252109 LR 0.000500 Time 0.047125 +2023-10-05 21:47:04,469 - Epoch: [138][ 40/ 1236] Overall Loss 0.252661 Objective Loss 0.252661 LR 0.000500 Time 0.040413 +2023-10-05 21:47:04,672 - Epoch: [138][ 50/ 1236] Overall Loss 0.258815 Objective Loss 0.258815 LR 0.000500 Time 0.036390 +2023-10-05 21:47:04,876 - Epoch: [138][ 60/ 1236] Overall Loss 0.256211 Objective Loss 0.256211 LR 0.000500 Time 0.033714 +2023-10-05 21:47:05,079 - Epoch: [138][ 70/ 1236] Overall Loss 0.261734 Objective Loss 0.261734 LR 0.000500 Time 0.031790 +2023-10-05 21:47:05,283 - Epoch: [138][ 80/ 1236] Overall Loss 0.259422 Objective Loss 0.259422 LR 0.000500 Time 0.030361 +2023-10-05 21:47:05,485 - Epoch: [138][ 90/ 1236] Overall Loss 0.256356 Objective Loss 0.256356 LR 0.000500 Time 0.029233 +2023-10-05 21:47:05,689 - Epoch: [138][ 100/ 1236] Overall Loss 0.254943 Objective Loss 0.254943 LR 0.000500 Time 0.028346 +2023-10-05 21:47:05,891 - Epoch: [138][ 110/ 1236] Overall Loss 0.251586 Objective Loss 0.251586 LR 0.000500 Time 0.027600 +2023-10-05 21:47:06,093 - Epoch: [138][ 120/ 1236] Overall Loss 0.251646 Objective Loss 0.251646 LR 0.000500 Time 0.026986 +2023-10-05 21:47:06,295 - Epoch: [138][ 130/ 1236] Overall Loss 0.251557 Objective Loss 0.251557 LR 0.000500 Time 0.026455 +2023-10-05 21:47:06,498 - Epoch: [138][ 140/ 1236] Overall Loss 0.250493 Objective Loss 0.250493 LR 0.000500 Time 0.026019 +2023-10-05 21:47:06,700 - Epoch: [138][ 150/ 1236] Overall Loss 0.250534 Objective Loss 0.250534 LR 0.000500 Time 0.025625 +2023-10-05 21:47:06,902 - Epoch: [138][ 160/ 1236] Overall Loss 0.250889 Objective Loss 0.250889 LR 0.000500 Time 0.025287 +2023-10-05 21:47:07,103 - Epoch: [138][ 170/ 1236] Overall Loss 0.251747 Objective Loss 0.251747 LR 0.000500 Time 0.024980 +2023-10-05 21:47:07,307 - Epoch: [138][ 180/ 1236] Overall Loss 0.252445 Objective Loss 0.252445 LR 0.000500 Time 0.024719 +2023-10-05 21:47:07,508 - Epoch: [138][ 190/ 1236] Overall Loss 0.252193 Objective Loss 0.252193 LR 0.000500 Time 0.024474 +2023-10-05 21:47:07,711 - Epoch: [138][ 200/ 1236] Overall Loss 0.251303 Objective Loss 0.251303 LR 0.000500 Time 0.024267 +2023-10-05 21:47:07,912 - Epoch: [138][ 210/ 1236] Overall Loss 0.252877 Objective Loss 0.252877 LR 0.000500 Time 0.024065 +2023-10-05 21:47:08,115 - Epoch: [138][ 220/ 1236] Overall Loss 0.252133 Objective Loss 0.252133 LR 0.000500 Time 0.023893 +2023-10-05 21:47:08,316 - Epoch: [138][ 230/ 1236] Overall Loss 0.252384 Objective Loss 0.252384 LR 0.000500 Time 0.023725 +2023-10-05 21:47:08,519 - Epoch: [138][ 240/ 1236] Overall Loss 0.251549 Objective Loss 0.251549 LR 0.000500 Time 0.023583 +2023-10-05 21:47:08,720 - Epoch: [138][ 250/ 1236] Overall Loss 0.252171 Objective Loss 0.252171 LR 0.000500 Time 0.023441 +2023-10-05 21:47:08,924 - Epoch: [138][ 260/ 1236] Overall Loss 0.253212 Objective Loss 0.253212 LR 0.000500 Time 0.023321 +2023-10-05 21:47:09,125 - Epoch: [138][ 270/ 1236] Overall Loss 0.253048 Objective Loss 0.253048 LR 0.000500 Time 0.023201 +2023-10-05 21:47:09,328 - Epoch: [138][ 280/ 1236] Overall Loss 0.253041 Objective Loss 0.253041 LR 0.000500 Time 0.023097 +2023-10-05 21:47:09,529 - Epoch: [138][ 290/ 1236] Overall Loss 0.253097 Objective Loss 0.253097 LR 0.000500 Time 0.022992 +2023-10-05 21:47:09,733 - Epoch: [138][ 300/ 1236] Overall Loss 0.253874 Objective Loss 0.253874 LR 0.000500 Time 0.022904 +2023-10-05 21:47:09,934 - Epoch: [138][ 310/ 1236] Overall Loss 0.253303 Objective Loss 0.253303 LR 0.000500 Time 0.022813 +2023-10-05 21:47:10,138 - Epoch: [138][ 320/ 1236] Overall Loss 0.252788 Objective Loss 0.252788 LR 0.000500 Time 0.022736 +2023-10-05 21:47:10,338 - Epoch: [138][ 330/ 1236] Overall Loss 0.253066 Objective Loss 0.253066 LR 0.000500 Time 0.022654 +2023-10-05 21:47:10,542 - Epoch: [138][ 340/ 1236] Overall Loss 0.252710 Objective Loss 0.252710 LR 0.000500 Time 0.022586 +2023-10-05 21:47:10,745 - Epoch: [138][ 350/ 1236] Overall Loss 0.252811 Objective Loss 0.252811 LR 0.000500 Time 0.022519 +2023-10-05 21:47:10,949 - Epoch: [138][ 360/ 1236] Overall Loss 0.252664 Objective Loss 0.252664 LR 0.000500 Time 0.022459 +2023-10-05 21:47:11,151 - Epoch: [138][ 370/ 1236] Overall Loss 0.252677 Objective Loss 0.252677 LR 0.000500 Time 0.022398 +2023-10-05 21:47:11,355 - Epoch: [138][ 380/ 1236] Overall Loss 0.252727 Objective Loss 0.252727 LR 0.000500 Time 0.022345 +2023-10-05 21:47:11,558 - Epoch: [138][ 390/ 1236] Overall Loss 0.252633 Objective Loss 0.252633 LR 0.000500 Time 0.022292 +2023-10-05 21:47:11,763 - Epoch: [138][ 400/ 1236] Overall Loss 0.252071 Objective Loss 0.252071 LR 0.000500 Time 0.022245 +2023-10-05 21:47:11,966 - Epoch: [138][ 410/ 1236] Overall Loss 0.251922 Objective Loss 0.251922 LR 0.000500 Time 0.022196 +2023-10-05 21:47:12,170 - Epoch: [138][ 420/ 1236] Overall Loss 0.252493 Objective Loss 0.252493 LR 0.000500 Time 0.022153 +2023-10-05 21:47:12,373 - Epoch: [138][ 430/ 1236] Overall Loss 0.252625 Objective Loss 0.252625 LR 0.000500 Time 0.022109 +2023-10-05 21:47:12,577 - Epoch: [138][ 440/ 1236] Overall Loss 0.252087 Objective Loss 0.252087 LR 0.000500 Time 0.022068 +2023-10-05 21:47:12,780 - Epoch: [138][ 450/ 1236] Overall Loss 0.252797 Objective Loss 0.252797 LR 0.000500 Time 0.022029 +2023-10-05 21:47:12,984 - Epoch: [138][ 460/ 1236] Overall Loss 0.253286 Objective Loss 0.253286 LR 0.000500 Time 0.021993 +2023-10-05 21:47:13,187 - Epoch: [138][ 470/ 1236] Overall Loss 0.253211 Objective Loss 0.253211 LR 0.000500 Time 0.021957 +2023-10-05 21:47:13,391 - Epoch: [138][ 480/ 1236] Overall Loss 0.252692 Objective Loss 0.252692 LR 0.000500 Time 0.021924 +2023-10-05 21:47:13,594 - Epoch: [138][ 490/ 1236] Overall Loss 0.252002 Objective Loss 0.252002 LR 0.000500 Time 0.021890 +2023-10-05 21:47:13,798 - Epoch: [138][ 500/ 1236] Overall Loss 0.252615 Objective Loss 0.252615 LR 0.000500 Time 0.021860 +2023-10-05 21:47:14,001 - Epoch: [138][ 510/ 1236] Overall Loss 0.252913 Objective Loss 0.252913 LR 0.000500 Time 0.021829 +2023-10-05 21:47:14,206 - Epoch: [138][ 520/ 1236] Overall Loss 0.252909 Objective Loss 0.252909 LR 0.000500 Time 0.021801 +2023-10-05 21:47:14,409 - Epoch: [138][ 530/ 1236] Overall Loss 0.252640 Objective Loss 0.252640 LR 0.000500 Time 0.021772 +2023-10-05 21:47:14,613 - Epoch: [138][ 540/ 1236] Overall Loss 0.252704 Objective Loss 0.252704 LR 0.000500 Time 0.021747 +2023-10-05 21:47:14,816 - Epoch: [138][ 550/ 1236] Overall Loss 0.253148 Objective Loss 0.253148 LR 0.000500 Time 0.021720 +2023-10-05 21:47:15,021 - Epoch: [138][ 560/ 1236] Overall Loss 0.252990 Objective Loss 0.252990 LR 0.000500 Time 0.021697 +2023-10-05 21:47:15,224 - Epoch: [138][ 570/ 1236] Overall Loss 0.252999 Objective Loss 0.252999 LR 0.000500 Time 0.021672 +2023-10-05 21:47:15,428 - Epoch: [138][ 580/ 1236] Overall Loss 0.252931 Objective Loss 0.252931 LR 0.000500 Time 0.021650 +2023-10-05 21:47:15,631 - Epoch: [138][ 590/ 1236] Overall Loss 0.253310 Objective Loss 0.253310 LR 0.000500 Time 0.021627 +2023-10-05 21:47:15,835 - Epoch: [138][ 600/ 1236] Overall Loss 0.253446 Objective Loss 0.253446 LR 0.000500 Time 0.021606 +2023-10-05 21:47:16,039 - Epoch: [138][ 610/ 1236] Overall Loss 0.253234 Objective Loss 0.253234 LR 0.000500 Time 0.021585 +2023-10-05 21:47:16,244 - Epoch: [138][ 620/ 1236] Overall Loss 0.253568 Objective Loss 0.253568 LR 0.000500 Time 0.021566 +2023-10-05 21:47:16,447 - Epoch: [138][ 630/ 1236] Overall Loss 0.253404 Objective Loss 0.253404 LR 0.000500 Time 0.021546 +2023-10-05 21:47:16,651 - Epoch: [138][ 640/ 1236] Overall Loss 0.253206 Objective Loss 0.253206 LR 0.000500 Time 0.021528 +2023-10-05 21:47:16,854 - Epoch: [138][ 650/ 1236] Overall Loss 0.253265 Objective Loss 0.253265 LR 0.000500 Time 0.021509 +2023-10-05 21:47:17,059 - Epoch: [138][ 660/ 1236] Overall Loss 0.253717 Objective Loss 0.253717 LR 0.000500 Time 0.021492 +2023-10-05 21:47:17,262 - Epoch: [138][ 670/ 1236] Overall Loss 0.253693 Objective Loss 0.253693 LR 0.000500 Time 0.021474 +2023-10-05 21:47:17,466 - Epoch: [138][ 680/ 1236] Overall Loss 0.253643 Objective Loss 0.253643 LR 0.000500 Time 0.021458 +2023-10-05 21:47:17,670 - Epoch: [138][ 690/ 1236] Overall Loss 0.253470 Objective Loss 0.253470 LR 0.000500 Time 0.021442 +2023-10-05 21:47:17,874 - Epoch: [138][ 700/ 1236] Overall Loss 0.253668 Objective Loss 0.253668 LR 0.000500 Time 0.021427 +2023-10-05 21:47:18,077 - Epoch: [138][ 710/ 1236] Overall Loss 0.253852 Objective Loss 0.253852 LR 0.000500 Time 0.021411 +2023-10-05 21:47:18,282 - Epoch: [138][ 720/ 1236] Overall Loss 0.254066 Objective Loss 0.254066 LR 0.000500 Time 0.021397 +2023-10-05 21:47:18,485 - Epoch: [138][ 730/ 1236] Overall Loss 0.254269 Objective Loss 0.254269 LR 0.000500 Time 0.021382 +2023-10-05 21:47:18,690 - Epoch: [138][ 740/ 1236] Overall Loss 0.254228 Objective Loss 0.254228 LR 0.000500 Time 0.021369 +2023-10-05 21:47:18,892 - Epoch: [138][ 750/ 1236] Overall Loss 0.254330 Objective Loss 0.254330 LR 0.000500 Time 0.021353 +2023-10-05 21:47:19,096 - Epoch: [138][ 760/ 1236] Overall Loss 0.254627 Objective Loss 0.254627 LR 0.000500 Time 0.021340 +2023-10-05 21:47:19,299 - Epoch: [138][ 770/ 1236] Overall Loss 0.254204 Objective Loss 0.254204 LR 0.000500 Time 0.021326 +2023-10-05 21:47:19,503 - Epoch: [138][ 780/ 1236] Overall Loss 0.254484 Objective Loss 0.254484 LR 0.000500 Time 0.021314 +2023-10-05 21:47:19,706 - Epoch: [138][ 790/ 1236] Overall Loss 0.254406 Objective Loss 0.254406 LR 0.000500 Time 0.021300 +2023-10-05 21:47:19,910 - Epoch: [138][ 800/ 1236] Overall Loss 0.254078 Objective Loss 0.254078 LR 0.000500 Time 0.021289 +2023-10-05 21:47:20,113 - Epoch: [138][ 810/ 1236] Overall Loss 0.253854 Objective Loss 0.253854 LR 0.000500 Time 0.021276 +2023-10-05 21:47:20,317 - Epoch: [138][ 820/ 1236] Overall Loss 0.254072 Objective Loss 0.254072 LR 0.000500 Time 0.021265 +2023-10-05 21:47:20,520 - Epoch: [138][ 830/ 1236] Overall Loss 0.254024 Objective Loss 0.254024 LR 0.000500 Time 0.021253 +2023-10-05 21:47:20,724 - Epoch: [138][ 840/ 1236] Overall Loss 0.254062 Objective Loss 0.254062 LR 0.000500 Time 0.021242 +2023-10-05 21:47:20,927 - Epoch: [138][ 850/ 1236] Overall Loss 0.254223 Objective Loss 0.254223 LR 0.000500 Time 0.021231 +2023-10-05 21:47:21,130 - Epoch: [138][ 860/ 1236] Overall Loss 0.254422 Objective Loss 0.254422 LR 0.000500 Time 0.021220 +2023-10-05 21:47:21,334 - Epoch: [138][ 870/ 1236] Overall Loss 0.254400 Objective Loss 0.254400 LR 0.000500 Time 0.021209 +2023-10-05 21:47:21,537 - Epoch: [138][ 880/ 1236] Overall Loss 0.254559 Objective Loss 0.254559 LR 0.000500 Time 0.021199 +2023-10-05 21:47:21,740 - Epoch: [138][ 890/ 1236] Overall Loss 0.254758 Objective Loss 0.254758 LR 0.000500 Time 0.021189 +2023-10-05 21:47:21,944 - Epoch: [138][ 900/ 1236] Overall Loss 0.254457 Objective Loss 0.254457 LR 0.000500 Time 0.021180 +2023-10-05 21:47:22,147 - Epoch: [138][ 910/ 1236] Overall Loss 0.254268 Objective Loss 0.254268 LR 0.000500 Time 0.021170 +2023-10-05 21:47:22,351 - Epoch: [138][ 920/ 1236] Overall Loss 0.254607 Objective Loss 0.254607 LR 0.000500 Time 0.021161 +2023-10-05 21:47:22,554 - Epoch: [138][ 930/ 1236] Overall Loss 0.254568 Objective Loss 0.254568 LR 0.000500 Time 0.021151 +2023-10-05 21:47:22,757 - Epoch: [138][ 940/ 1236] Overall Loss 0.254972 Objective Loss 0.254972 LR 0.000500 Time 0.021142 +2023-10-05 21:47:22,961 - Epoch: [138][ 950/ 1236] Overall Loss 0.254709 Objective Loss 0.254709 LR 0.000500 Time 0.021133 +2023-10-05 21:47:23,164 - Epoch: [138][ 960/ 1236] Overall Loss 0.254447 Objective Loss 0.254447 LR 0.000500 Time 0.021125 +2023-10-05 21:47:23,367 - Epoch: [138][ 970/ 1236] Overall Loss 0.254299 Objective Loss 0.254299 LR 0.000500 Time 0.021116 +2023-10-05 21:47:23,571 - Epoch: [138][ 980/ 1236] Overall Loss 0.254072 Objective Loss 0.254072 LR 0.000500 Time 0.021108 +2023-10-05 21:47:23,775 - Epoch: [138][ 990/ 1236] Overall Loss 0.254324 Objective Loss 0.254324 LR 0.000500 Time 0.021100 +2023-10-05 21:47:23,978 - Epoch: [138][ 1000/ 1236] Overall Loss 0.254387 Objective Loss 0.254387 LR 0.000500 Time 0.021093 +2023-10-05 21:47:24,182 - Epoch: [138][ 1010/ 1236] Overall Loss 0.254389 Objective Loss 0.254389 LR 0.000500 Time 0.021085 +2023-10-05 21:47:24,389 - Epoch: [138][ 1020/ 1236] Overall Loss 0.254646 Objective Loss 0.254646 LR 0.000500 Time 0.021077 +2023-10-05 21:47:24,592 - Epoch: [138][ 1030/ 1236] Overall Loss 0.254835 Objective Loss 0.254835 LR 0.000500 Time 0.021069 +2023-10-05 21:47:24,796 - Epoch: [138][ 1040/ 1236] Overall Loss 0.254854 Objective Loss 0.254854 LR 0.000500 Time 0.021062 +2023-10-05 21:47:24,999 - Epoch: [138][ 1050/ 1236] Overall Loss 0.254692 Objective Loss 0.254692 LR 0.000500 Time 0.021054 +2023-10-05 21:47:25,203 - Epoch: [138][ 1060/ 1236] Overall Loss 0.254948 Objective Loss 0.254948 LR 0.000500 Time 0.021048 +2023-10-05 21:47:25,406 - Epoch: [138][ 1070/ 1236] Overall Loss 0.254871 Objective Loss 0.254871 LR 0.000500 Time 0.021041 +2023-10-05 21:47:25,610 - Epoch: [138][ 1080/ 1236] Overall Loss 0.254973 Objective Loss 0.254973 LR 0.000500 Time 0.021035 +2023-10-05 21:47:25,813 - Epoch: [138][ 1090/ 1236] Overall Loss 0.254810 Objective Loss 0.254810 LR 0.000500 Time 0.021027 +2023-10-05 21:47:26,017 - Epoch: [138][ 1100/ 1236] Overall Loss 0.254725 Objective Loss 0.254725 LR 0.000500 Time 0.021021 +2023-10-05 21:47:26,220 - Epoch: [138][ 1110/ 1236] Overall Loss 0.254655 Objective Loss 0.254655 LR 0.000500 Time 0.021015 +2023-10-05 21:47:26,424 - Epoch: [138][ 1120/ 1236] Overall Loss 0.254626 Objective Loss 0.254626 LR 0.000500 Time 0.021009 +2023-10-05 21:47:26,627 - Epoch: [138][ 1130/ 1236] Overall Loss 0.254464 Objective Loss 0.254464 LR 0.000500 Time 0.021002 +2023-10-05 21:47:26,831 - Epoch: [138][ 1140/ 1236] Overall Loss 0.254297 Objective Loss 0.254297 LR 0.000500 Time 0.020997 +2023-10-05 21:47:27,034 - Epoch: [138][ 1150/ 1236] Overall Loss 0.254252 Objective Loss 0.254252 LR 0.000500 Time 0.020990 +2023-10-05 21:47:27,238 - Epoch: [138][ 1160/ 1236] Overall Loss 0.254375 Objective Loss 0.254375 LR 0.000500 Time 0.020985 +2023-10-05 21:47:27,442 - Epoch: [138][ 1170/ 1236] Overall Loss 0.254404 Objective Loss 0.254404 LR 0.000500 Time 0.020979 +2023-10-05 21:47:27,646 - Epoch: [138][ 1180/ 1236] Overall Loss 0.254229 Objective Loss 0.254229 LR 0.000500 Time 0.020975 +2023-10-05 21:47:27,850 - Epoch: [138][ 1190/ 1236] Overall Loss 0.254051 Objective Loss 0.254051 LR 0.000500 Time 0.020969 +2023-10-05 21:47:28,054 - Epoch: [138][ 1200/ 1236] Overall Loss 0.254233 Objective Loss 0.254233 LR 0.000500 Time 0.020964 +2023-10-05 21:47:28,258 - Epoch: [138][ 1210/ 1236] Overall Loss 0.254238 Objective Loss 0.254238 LR 0.000500 Time 0.020959 +2023-10-05 21:47:28,462 - Epoch: [138][ 1220/ 1236] Overall Loss 0.254054 Objective Loss 0.254054 LR 0.000500 Time 0.020954 +2023-10-05 21:47:28,718 - Epoch: [138][ 1230/ 1236] Overall Loss 0.254246 Objective Loss 0.254246 LR 0.000500 Time 0.020992 +2023-10-05 21:47:28,837 - Epoch: [138][ 1236/ 1236] Overall Loss 0.254198 Objective Loss 0.254198 Top1 86.150713 Top5 97.759674 LR 0.000500 Time 0.020986 +2023-10-05 21:47:28,972 - --- validate (epoch=138)----------- +2023-10-05 21:47:28,972 - 29943 samples (256 per mini-batch) +2023-10-05 21:47:29,430 - Epoch: [138][ 10/ 117] Loss 0.320416 Top1 83.437500 Top5 98.125000 +2023-10-05 21:47:29,581 - Epoch: [138][ 20/ 117] Loss 0.328188 Top1 83.574219 Top5 97.734375 +2023-10-05 21:47:29,730 - Epoch: [138][ 30/ 117] Loss 0.319208 Top1 83.684896 Top5 97.760417 +2023-10-05 21:47:29,880 - Epoch: [138][ 40/ 117] Loss 0.327792 Top1 83.720703 Top5 97.744141 +2023-10-05 21:47:30,029 - Epoch: [138][ 50/ 117] Loss 0.321025 Top1 83.656250 Top5 97.796875 +2023-10-05 21:47:30,182 - Epoch: [138][ 60/ 117] Loss 0.322804 Top1 83.515625 Top5 97.779948 +2023-10-05 21:47:30,336 - Epoch: [138][ 70/ 117] Loss 0.322085 Top1 83.560268 Top5 97.862723 +2023-10-05 21:47:30,486 - Epoch: [138][ 80/ 117] Loss 0.319891 Top1 83.510742 Top5 97.895508 +2023-10-05 21:47:30,634 - Epoch: [138][ 90/ 117] Loss 0.321118 Top1 83.528646 Top5 97.903646 +2023-10-05 21:47:30,783 - Epoch: [138][ 100/ 117] Loss 0.321602 Top1 83.589844 Top5 97.902344 +2023-10-05 21:47:30,938 - Epoch: [138][ 110/ 117] Loss 0.319255 Top1 83.678977 Top5 97.936790 +2023-10-05 21:47:31,024 - Epoch: [138][ 117/ 117] Loss 0.321725 Top1 83.675650 Top5 97.936079 +2023-10-05 21:47:31,171 - ==> Top1: 83.676 Top5: 97.936 Loss: 0.322 + +2023-10-05 21:47:31,172 - ==> Confusion: +[[ 936 1 7 0 5 4 1 0 5 60 0 0 0 4 6 3 2 1 2 0 13] + [ 4 1043 3 0 10 24 1 26 0 0 1 0 0 0 1 3 4 0 6 2 3] + [ 5 1 966 12 1 2 27 7 0 0 3 2 8 1 1 4 1 1 3 3 8] + [ 3 1 14 959 3 6 0 2 1 1 3 1 7 4 34 3 0 7 25 2 13] + [ 24 4 2 0 967 7 1 1 1 10 1 2 0 2 9 2 8 2 1 0 6] + [ 4 24 0 2 3 1002 1 21 1 1 4 5 1 13 9 2 4 1 0 3 15] + [ 0 5 23 0 0 1 1131 8 0 0 0 2 1 0 2 6 0 0 1 7 4] + [ 4 15 22 0 4 33 4 1063 0 4 3 13 1 0 1 3 0 0 35 6 7] + [ 21 2 0 0 2 5 1 2 953 44 12 1 0 14 18 2 1 2 5 2 2] + [ 111 0 1 0 2 3 0 0 22 927 0 2 0 24 7 9 1 2 2 0 6] + [ 2 3 12 7 0 1 5 3 9 1 965 4 2 12 6 1 2 0 7 0 11] + [ 2 1 0 0 0 13 0 0 0 1 0 965 15 6 0 2 1 13 0 13 3] + [ 2 1 5 7 0 1 0 0 0 0 0 39 984 0 3 4 2 8 2 1 9] + [ 1 1 1 0 4 13 0 0 9 12 6 1 3 1052 3 0 1 1 0 1 10] + [ 12 1 3 3 5 0 0 0 16 2 2 1 5 2 1027 0 1 1 9 0 11] + [ 1 5 2 1 5 0 3 0 0 0 0 7 6 2 1 1061 14 11 0 10 5] + [ 1 12 2 0 6 3 0 1 2 0 0 4 2 1 3 8 1101 0 0 5 10] + [ 0 0 1 1 0 2 2 0 0 1 0 1 21 3 1 7 1 993 1 0 3] + [ 1 0 6 21 1 1 1 25 1 0 1 1 2 0 12 0 1 0 985 2 7] + [ 0 1 3 2 1 4 11 6 0 0 2 10 6 2 0 1 6 2 3 1084 8] + [ 131 162 181 63 107 186 62 90 79 80 182 100 361 315 174 71 175 72 172 251 4891]] + +2023-10-05 21:47:31,173 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:47:31,173 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:47:31,179 - + +2023-10-05 21:47:31,179 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:47:32,182 - Epoch: [139][ 10/ 1236] Overall Loss 0.232029 Objective Loss 0.232029 LR 0.000500 Time 0.100190 +2023-10-05 21:47:32,387 - Epoch: [139][ 20/ 1236] Overall Loss 0.235554 Objective Loss 0.235554 LR 0.000500 Time 0.060326 +2023-10-05 21:47:32,589 - Epoch: [139][ 30/ 1236] Overall Loss 0.235826 Objective Loss 0.235826 LR 0.000500 Time 0.046948 +2023-10-05 21:47:32,796 - Epoch: [139][ 40/ 1236] Overall Loss 0.235623 Objective Loss 0.235623 LR 0.000500 Time 0.040390 +2023-10-05 21:47:33,000 - Epoch: [139][ 50/ 1236] Overall Loss 0.244431 Objective Loss 0.244431 LR 0.000500 Time 0.036385 +2023-10-05 21:47:33,208 - Epoch: [139][ 60/ 1236] Overall Loss 0.244289 Objective Loss 0.244289 LR 0.000500 Time 0.033776 +2023-10-05 21:47:33,412 - Epoch: [139][ 70/ 1236] Overall Loss 0.240028 Objective Loss 0.240028 LR 0.000500 Time 0.031866 +2023-10-05 21:47:33,628 - Epoch: [139][ 80/ 1236] Overall Loss 0.243348 Objective Loss 0.243348 LR 0.000500 Time 0.030571 +2023-10-05 21:47:33,843 - Epoch: [139][ 90/ 1236] Overall Loss 0.238872 Objective Loss 0.238872 LR 0.000500 Time 0.029562 +2023-10-05 21:47:34,055 - Epoch: [139][ 100/ 1236] Overall Loss 0.240415 Objective Loss 0.240415 LR 0.000500 Time 0.028715 +2023-10-05 21:47:34,263 - Epoch: [139][ 110/ 1236] Overall Loss 0.241137 Objective Loss 0.241137 LR 0.000500 Time 0.027995 +2023-10-05 21:47:34,470 - Epoch: [139][ 120/ 1236] Overall Loss 0.243408 Objective Loss 0.243408 LR 0.000500 Time 0.027388 +2023-10-05 21:47:34,675 - Epoch: [139][ 130/ 1236] Overall Loss 0.241043 Objective Loss 0.241043 LR 0.000500 Time 0.026851 +2023-10-05 21:47:34,880 - Epoch: [139][ 140/ 1236] Overall Loss 0.242211 Objective Loss 0.242211 LR 0.000500 Time 0.026397 +2023-10-05 21:47:35,085 - Epoch: [139][ 150/ 1236] Overall Loss 0.242741 Objective Loss 0.242741 LR 0.000500 Time 0.025997 +2023-10-05 21:47:35,290 - Epoch: [139][ 160/ 1236] Overall Loss 0.242106 Objective Loss 0.242106 LR 0.000500 Time 0.025655 +2023-10-05 21:47:35,495 - Epoch: [139][ 170/ 1236] Overall Loss 0.241390 Objective Loss 0.241390 LR 0.000500 Time 0.025346 +2023-10-05 21:47:35,700 - Epoch: [139][ 180/ 1236] Overall Loss 0.241518 Objective Loss 0.241518 LR 0.000500 Time 0.025078 +2023-10-05 21:47:35,905 - Epoch: [139][ 190/ 1236] Overall Loss 0.241667 Objective Loss 0.241667 LR 0.000500 Time 0.024832 +2023-10-05 21:47:36,110 - Epoch: [139][ 200/ 1236] Overall Loss 0.242725 Objective Loss 0.242725 LR 0.000500 Time 0.024616 +2023-10-05 21:47:36,315 - Epoch: [139][ 210/ 1236] Overall Loss 0.242960 Objective Loss 0.242960 LR 0.000500 Time 0.024415 +2023-10-05 21:47:36,520 - Epoch: [139][ 220/ 1236] Overall Loss 0.242782 Objective Loss 0.242782 LR 0.000500 Time 0.024238 +2023-10-05 21:47:36,725 - Epoch: [139][ 230/ 1236] Overall Loss 0.242813 Objective Loss 0.242813 LR 0.000500 Time 0.024072 +2023-10-05 21:47:36,930 - Epoch: [139][ 240/ 1236] Overall Loss 0.242062 Objective Loss 0.242062 LR 0.000500 Time 0.023923 +2023-10-05 21:47:37,135 - Epoch: [139][ 250/ 1236] Overall Loss 0.242153 Objective Loss 0.242153 LR 0.000500 Time 0.023783 +2023-10-05 21:47:37,341 - Epoch: [139][ 260/ 1236] Overall Loss 0.242302 Objective Loss 0.242302 LR 0.000500 Time 0.023659 +2023-10-05 21:47:37,545 - Epoch: [139][ 270/ 1236] Overall Loss 0.243055 Objective Loss 0.243055 LR 0.000500 Time 0.023536 +2023-10-05 21:47:37,750 - Epoch: [139][ 280/ 1236] Overall Loss 0.243370 Objective Loss 0.243370 LR 0.000500 Time 0.023429 +2023-10-05 21:47:37,955 - Epoch: [139][ 290/ 1236] Overall Loss 0.244710 Objective Loss 0.244710 LR 0.000500 Time 0.023324 +2023-10-05 21:47:38,161 - Epoch: [139][ 300/ 1236] Overall Loss 0.244715 Objective Loss 0.244715 LR 0.000500 Time 0.023231 +2023-10-05 21:47:38,366 - Epoch: [139][ 310/ 1236] Overall Loss 0.244536 Objective Loss 0.244536 LR 0.000500 Time 0.023142 +2023-10-05 21:47:38,572 - Epoch: [139][ 320/ 1236] Overall Loss 0.244859 Objective Loss 0.244859 LR 0.000500 Time 0.023061 +2023-10-05 21:47:38,779 - Epoch: [139][ 330/ 1236] Overall Loss 0.244814 Objective Loss 0.244814 LR 0.000500 Time 0.022988 +2023-10-05 21:47:38,984 - Epoch: [139][ 340/ 1236] Overall Loss 0.245102 Objective Loss 0.245102 LR 0.000500 Time 0.022916 +2023-10-05 21:47:39,189 - Epoch: [139][ 350/ 1236] Overall Loss 0.244748 Objective Loss 0.244748 LR 0.000500 Time 0.022845 +2023-10-05 21:47:39,395 - Epoch: [139][ 360/ 1236] Overall Loss 0.244248 Objective Loss 0.244248 LR 0.000500 Time 0.022782 +2023-10-05 21:47:39,599 - Epoch: [139][ 370/ 1236] Overall Loss 0.243832 Objective Loss 0.243832 LR 0.000500 Time 0.022717 +2023-10-05 21:47:39,809 - Epoch: [139][ 380/ 1236] Overall Loss 0.243674 Objective Loss 0.243674 LR 0.000500 Time 0.022670 +2023-10-05 21:47:40,013 - Epoch: [139][ 390/ 1236] Overall Loss 0.243933 Objective Loss 0.243933 LR 0.000500 Time 0.022611 +2023-10-05 21:47:40,218 - Epoch: [139][ 400/ 1236] Overall Loss 0.244327 Objective Loss 0.244327 LR 0.000500 Time 0.022557 +2023-10-05 21:47:40,422 - Epoch: [139][ 410/ 1236] Overall Loss 0.243559 Objective Loss 0.243559 LR 0.000500 Time 0.022503 +2023-10-05 21:47:40,635 - Epoch: [139][ 420/ 1236] Overall Loss 0.243594 Objective Loss 0.243594 LR 0.000500 Time 0.022473 +2023-10-05 21:47:40,846 - Epoch: [139][ 430/ 1236] Overall Loss 0.244121 Objective Loss 0.244121 LR 0.000500 Time 0.022441 +2023-10-05 21:47:41,057 - Epoch: [139][ 440/ 1236] Overall Loss 0.244476 Objective Loss 0.244476 LR 0.000500 Time 0.022408 +2023-10-05 21:47:41,272 - Epoch: [139][ 450/ 1236] Overall Loss 0.245153 Objective Loss 0.245153 LR 0.000500 Time 0.022388 +2023-10-05 21:47:41,484 - Epoch: [139][ 460/ 1236] Overall Loss 0.245262 Objective Loss 0.245262 LR 0.000500 Time 0.022360 +2023-10-05 21:47:41,698 - Epoch: [139][ 470/ 1236] Overall Loss 0.245457 Objective Loss 0.245457 LR 0.000500 Time 0.022339 +2023-10-05 21:47:41,907 - Epoch: [139][ 480/ 1236] Overall Loss 0.245740 Objective Loss 0.245740 LR 0.000500 Time 0.022309 +2023-10-05 21:47:42,121 - Epoch: [139][ 490/ 1236] Overall Loss 0.246487 Objective Loss 0.246487 LR 0.000500 Time 0.022289 +2023-10-05 21:47:42,333 - Epoch: [139][ 500/ 1236] Overall Loss 0.246572 Objective Loss 0.246572 LR 0.000500 Time 0.022265 +2023-10-05 21:47:42,546 - Epoch: [139][ 510/ 1236] Overall Loss 0.246562 Objective Loss 0.246562 LR 0.000500 Time 0.022245 +2023-10-05 21:47:42,756 - Epoch: [139][ 520/ 1236] Overall Loss 0.247135 Objective Loss 0.247135 LR 0.000500 Time 0.022221 +2023-10-05 21:47:42,970 - Epoch: [139][ 530/ 1236] Overall Loss 0.246933 Objective Loss 0.246933 LR 0.000500 Time 0.022205 +2023-10-05 21:47:43,180 - Epoch: [139][ 540/ 1236] Overall Loss 0.246872 Objective Loss 0.246872 LR 0.000500 Time 0.022182 +2023-10-05 21:47:43,394 - Epoch: [139][ 550/ 1236] Overall Loss 0.246677 Objective Loss 0.246677 LR 0.000500 Time 0.022167 +2023-10-05 21:47:43,606 - Epoch: [139][ 560/ 1236] Overall Loss 0.246493 Objective Loss 0.246493 LR 0.000500 Time 0.022147 +2023-10-05 21:47:43,820 - Epoch: [139][ 570/ 1236] Overall Loss 0.246515 Objective Loss 0.246515 LR 0.000500 Time 0.022133 +2023-10-05 21:47:44,031 - Epoch: [139][ 580/ 1236] Overall Loss 0.246139 Objective Loss 0.246139 LR 0.000500 Time 0.022114 +2023-10-05 21:47:44,245 - Epoch: [139][ 590/ 1236] Overall Loss 0.246244 Objective Loss 0.246244 LR 0.000500 Time 0.022102 +2023-10-05 21:47:44,456 - Epoch: [139][ 600/ 1236] Overall Loss 0.246264 Objective Loss 0.246264 LR 0.000500 Time 0.022085 +2023-10-05 21:47:44,670 - Epoch: [139][ 610/ 1236] Overall Loss 0.246179 Objective Loss 0.246179 LR 0.000500 Time 0.022073 +2023-10-05 21:47:44,881 - Epoch: [139][ 620/ 1236] Overall Loss 0.246176 Objective Loss 0.246176 LR 0.000500 Time 0.022056 +2023-10-05 21:47:45,096 - Epoch: [139][ 630/ 1236] Overall Loss 0.246086 Objective Loss 0.246086 LR 0.000500 Time 0.022047 +2023-10-05 21:47:45,309 - Epoch: [139][ 640/ 1236] Overall Loss 0.246429 Objective Loss 0.246429 LR 0.000500 Time 0.022034 +2023-10-05 21:47:45,529 - Epoch: [139][ 650/ 1236] Overall Loss 0.246795 Objective Loss 0.246795 LR 0.000500 Time 0.022033 +2023-10-05 21:47:45,746 - Epoch: [139][ 660/ 1236] Overall Loss 0.246586 Objective Loss 0.246586 LR 0.000500 Time 0.022027 +2023-10-05 21:47:45,962 - Epoch: [139][ 670/ 1236] Overall Loss 0.246657 Objective Loss 0.246657 LR 0.000500 Time 0.022019 +2023-10-05 21:47:46,168 - Epoch: [139][ 680/ 1236] Overall Loss 0.246486 Objective Loss 0.246486 LR 0.000500 Time 0.021999 +2023-10-05 21:47:46,374 - Epoch: [139][ 690/ 1236] Overall Loss 0.246641 Objective Loss 0.246641 LR 0.000500 Time 0.021977 +2023-10-05 21:47:46,583 - Epoch: [139][ 700/ 1236] Overall Loss 0.246178 Objective Loss 0.246178 LR 0.000500 Time 0.021961 +2023-10-05 21:47:46,792 - Epoch: [139][ 710/ 1236] Overall Loss 0.246386 Objective Loss 0.246386 LR 0.000500 Time 0.021945 +2023-10-05 21:47:47,010 - Epoch: [139][ 720/ 1236] Overall Loss 0.246203 Objective Loss 0.246203 LR 0.000500 Time 0.021943 +2023-10-05 21:47:47,224 - Epoch: [139][ 730/ 1236] Overall Loss 0.246411 Objective Loss 0.246411 LR 0.000500 Time 0.021934 +2023-10-05 21:47:47,435 - Epoch: [139][ 740/ 1236] Overall Loss 0.247055 Objective Loss 0.247055 LR 0.000500 Time 0.021923 +2023-10-05 21:47:47,642 - Epoch: [139][ 750/ 1236] Overall Loss 0.247008 Objective Loss 0.247008 LR 0.000500 Time 0.021905 +2023-10-05 21:47:47,850 - Epoch: [139][ 760/ 1236] Overall Loss 0.247230 Objective Loss 0.247230 LR 0.000500 Time 0.021890 +2023-10-05 21:47:48,058 - Epoch: [139][ 770/ 1236] Overall Loss 0.247165 Objective Loss 0.247165 LR 0.000500 Time 0.021874 +2023-10-05 21:47:48,265 - Epoch: [139][ 780/ 1236] Overall Loss 0.247427 Objective Loss 0.247427 LR 0.000500 Time 0.021859 +2023-10-05 21:47:48,472 - Epoch: [139][ 790/ 1236] Overall Loss 0.247640 Objective Loss 0.247640 LR 0.000500 Time 0.021844 +2023-10-05 21:47:48,679 - Epoch: [139][ 800/ 1236] Overall Loss 0.247878 Objective Loss 0.247878 LR 0.000500 Time 0.021829 +2023-10-05 21:47:48,892 - Epoch: [139][ 810/ 1236] Overall Loss 0.247809 Objective Loss 0.247809 LR 0.000500 Time 0.021822 +2023-10-05 21:47:49,093 - Epoch: [139][ 820/ 1236] Overall Loss 0.247936 Objective Loss 0.247936 LR 0.000500 Time 0.021801 +2023-10-05 21:47:49,292 - Epoch: [139][ 830/ 1236] Overall Loss 0.248075 Objective Loss 0.248075 LR 0.000500 Time 0.021777 +2023-10-05 21:47:49,491 - Epoch: [139][ 840/ 1236] Overall Loss 0.248199 Objective Loss 0.248199 LR 0.000500 Time 0.021755 +2023-10-05 21:47:49,691 - Epoch: [139][ 850/ 1236] Overall Loss 0.248428 Objective Loss 0.248428 LR 0.000500 Time 0.021734 +2023-10-05 21:47:49,891 - Epoch: [139][ 860/ 1236] Overall Loss 0.248344 Objective Loss 0.248344 LR 0.000500 Time 0.021714 +2023-10-05 21:47:50,091 - Epoch: [139][ 870/ 1236] Overall Loss 0.248265 Objective Loss 0.248265 LR 0.000500 Time 0.021693 +2023-10-05 21:47:50,291 - Epoch: [139][ 880/ 1236] Overall Loss 0.248294 Objective Loss 0.248294 LR 0.000500 Time 0.021674 +2023-10-05 21:47:50,491 - Epoch: [139][ 890/ 1236] Overall Loss 0.248535 Objective Loss 0.248535 LR 0.000500 Time 0.021654 +2023-10-05 21:47:50,692 - Epoch: [139][ 900/ 1236] Overall Loss 0.248477 Objective Loss 0.248477 LR 0.000500 Time 0.021637 +2023-10-05 21:47:50,892 - Epoch: [139][ 910/ 1236] Overall Loss 0.248383 Objective Loss 0.248383 LR 0.000500 Time 0.021619 +2023-10-05 21:47:51,092 - Epoch: [139][ 920/ 1236] Overall Loss 0.248455 Objective Loss 0.248455 LR 0.000500 Time 0.021601 +2023-10-05 21:47:51,292 - Epoch: [139][ 930/ 1236] Overall Loss 0.248353 Objective Loss 0.248353 LR 0.000500 Time 0.021584 +2023-10-05 21:47:51,499 - Epoch: [139][ 940/ 1236] Overall Loss 0.248209 Objective Loss 0.248209 LR 0.000500 Time 0.021573 +2023-10-05 21:47:51,706 - Epoch: [139][ 950/ 1236] Overall Loss 0.248310 Objective Loss 0.248310 LR 0.000500 Time 0.021564 +2023-10-05 21:47:51,915 - Epoch: [139][ 960/ 1236] Overall Loss 0.248474 Objective Loss 0.248474 LR 0.000500 Time 0.021557 +2023-10-05 21:47:52,125 - Epoch: [139][ 970/ 1236] Overall Loss 0.248566 Objective Loss 0.248566 LR 0.000500 Time 0.021551 +2023-10-05 21:47:52,335 - Epoch: [139][ 980/ 1236] Overall Loss 0.248306 Objective Loss 0.248306 LR 0.000500 Time 0.021544 +2023-10-05 21:47:52,545 - Epoch: [139][ 990/ 1236] Overall Loss 0.248294 Objective Loss 0.248294 LR 0.000500 Time 0.021538 +2023-10-05 21:47:52,755 - Epoch: [139][ 1000/ 1236] Overall Loss 0.248353 Objective Loss 0.248353 LR 0.000500 Time 0.021532 +2023-10-05 21:47:52,965 - Epoch: [139][ 1010/ 1236] Overall Loss 0.248261 Objective Loss 0.248261 LR 0.000500 Time 0.021527 +2023-10-05 21:47:53,175 - Epoch: [139][ 1020/ 1236] Overall Loss 0.248167 Objective Loss 0.248167 LR 0.000500 Time 0.021521 +2023-10-05 21:47:53,385 - Epoch: [139][ 1030/ 1236] Overall Loss 0.248342 Objective Loss 0.248342 LR 0.000500 Time 0.021516 +2023-10-05 21:47:53,595 - Epoch: [139][ 1040/ 1236] Overall Loss 0.248140 Objective Loss 0.248140 LR 0.000500 Time 0.021510 +2023-10-05 21:47:53,805 - Epoch: [139][ 1050/ 1236] Overall Loss 0.248085 Objective Loss 0.248085 LR 0.000500 Time 0.021505 +2023-10-05 21:47:54,014 - Epoch: [139][ 1060/ 1236] Overall Loss 0.248101 Objective Loss 0.248101 LR 0.000500 Time 0.021499 +2023-10-05 21:47:54,224 - Epoch: [139][ 1070/ 1236] Overall Loss 0.247982 Objective Loss 0.247982 LR 0.000500 Time 0.021493 +2023-10-05 21:47:54,434 - Epoch: [139][ 1080/ 1236] Overall Loss 0.248000 Objective Loss 0.248000 LR 0.000500 Time 0.021488 +2023-10-05 21:47:54,643 - Epoch: [139][ 1090/ 1236] Overall Loss 0.247952 Objective Loss 0.247952 LR 0.000500 Time 0.021482 +2023-10-05 21:47:54,852 - Epoch: [139][ 1100/ 1236] Overall Loss 0.247930 Objective Loss 0.247930 LR 0.000500 Time 0.021476 +2023-10-05 21:47:55,062 - Epoch: [139][ 1110/ 1236] Overall Loss 0.247756 Objective Loss 0.247756 LR 0.000500 Time 0.021472 +2023-10-05 21:47:55,271 - Epoch: [139][ 1120/ 1236] Overall Loss 0.247886 Objective Loss 0.247886 LR 0.000500 Time 0.021466 +2023-10-05 21:47:55,481 - Epoch: [139][ 1130/ 1236] Overall Loss 0.247892 Objective Loss 0.247892 LR 0.000500 Time 0.021461 +2023-10-05 21:47:55,690 - Epoch: [139][ 1140/ 1236] Overall Loss 0.247925 Objective Loss 0.247925 LR 0.000500 Time 0.021456 +2023-10-05 21:47:55,900 - Epoch: [139][ 1150/ 1236] Overall Loss 0.248109 Objective Loss 0.248109 LR 0.000500 Time 0.021452 +2023-10-05 21:47:56,110 - Epoch: [139][ 1160/ 1236] Overall Loss 0.248517 Objective Loss 0.248517 LR 0.000500 Time 0.021447 +2023-10-05 21:47:56,320 - Epoch: [139][ 1170/ 1236] Overall Loss 0.248780 Objective Loss 0.248780 LR 0.000500 Time 0.021444 +2023-10-05 21:47:56,530 - Epoch: [139][ 1180/ 1236] Overall Loss 0.248850 Objective Loss 0.248850 LR 0.000500 Time 0.021439 +2023-10-05 21:47:56,739 - Epoch: [139][ 1190/ 1236] Overall Loss 0.248951 Objective Loss 0.248951 LR 0.000500 Time 0.021434 +2023-10-05 21:47:56,947 - Epoch: [139][ 1200/ 1236] Overall Loss 0.249000 Objective Loss 0.249000 LR 0.000500 Time 0.021429 +2023-10-05 21:47:57,156 - Epoch: [139][ 1210/ 1236] Overall Loss 0.249130 Objective Loss 0.249130 LR 0.000500 Time 0.021424 +2023-10-05 21:47:57,365 - Epoch: [139][ 1220/ 1236] Overall Loss 0.249022 Objective Loss 0.249022 LR 0.000500 Time 0.021419 +2023-10-05 21:47:57,630 - Epoch: [139][ 1230/ 1236] Overall Loss 0.248643 Objective Loss 0.248643 LR 0.000500 Time 0.021460 +2023-10-05 21:47:57,748 - Epoch: [139][ 1236/ 1236] Overall Loss 0.248498 Objective Loss 0.248498 Top1 89.205703 Top5 98.370672 LR 0.000500 Time 0.021451 +2023-10-05 21:47:57,873 - --- validate (epoch=139)----------- +2023-10-05 21:47:57,873 - 29943 samples (256 per mini-batch) +2023-10-05 21:47:58,327 - Epoch: [139][ 10/ 117] Loss 0.286737 Top1 83.984375 Top5 97.500000 +2023-10-05 21:47:58,479 - Epoch: [139][ 20/ 117] Loss 0.289195 Top1 84.375000 Top5 97.910156 +2023-10-05 21:47:58,631 - Epoch: [139][ 30/ 117] Loss 0.301523 Top1 84.127604 Top5 97.747396 +2023-10-05 21:47:58,785 - Epoch: [139][ 40/ 117] Loss 0.309203 Top1 84.169922 Top5 97.666016 +2023-10-05 21:47:58,938 - Epoch: [139][ 50/ 117] Loss 0.309217 Top1 84.226562 Top5 97.710938 +2023-10-05 21:47:59,087 - Epoch: [139][ 60/ 117] Loss 0.312366 Top1 84.166667 Top5 97.832031 +2023-10-05 21:47:59,236 - Epoch: [139][ 70/ 117] Loss 0.315638 Top1 84.168527 Top5 97.795759 +2023-10-05 21:47:59,386 - Epoch: [139][ 80/ 117] Loss 0.317659 Top1 84.145508 Top5 97.827148 +2023-10-05 21:47:59,534 - Epoch: [139][ 90/ 117] Loss 0.316656 Top1 84.105903 Top5 97.825521 +2023-10-05 21:47:59,683 - Epoch: [139][ 100/ 117] Loss 0.317397 Top1 83.996094 Top5 97.847656 +2023-10-05 21:47:59,840 - Epoch: [139][ 110/ 117] Loss 0.320956 Top1 83.987926 Top5 97.837358 +2023-10-05 21:47:59,926 - Epoch: [139][ 117/ 117] Loss 0.317935 Top1 84.022977 Top5 97.859266 +2023-10-05 21:48:00,046 - ==> Top1: 84.023 Top5: 97.859 Loss: 0.318 + +2023-10-05 21:48:00,047 - ==> Confusion: +[[ 908 3 2 2 12 2 0 0 5 77 2 1 1 6 6 2 3 0 1 0 17] + [ 2 1054 2 0 9 21 2 16 1 0 0 1 0 0 1 3 4 0 5 1 9] + [ 3 1 960 10 6 1 28 7 0 1 5 1 9 1 0 5 1 1 6 2 8] + [ 5 1 13 961 1 3 2 2 2 1 10 0 1 3 28 5 0 9 24 2 16] + [ 20 5 1 0 982 4 0 1 1 8 1 1 2 1 4 6 4 2 1 2 4] + [ 3 45 2 0 4 984 0 16 1 0 4 7 2 14 7 2 3 1 2 8 11] + [ 0 8 24 0 1 0 1121 7 0 0 2 2 1 0 0 10 0 3 4 5 3] + [ 4 17 15 0 3 29 4 1055 0 2 3 11 2 2 1 3 0 1 44 10 12] + [ 19 3 0 0 1 1 0 0 967 40 10 3 2 12 20 4 0 1 3 0 3] + [ 87 0 5 0 6 5 0 0 21 955 0 2 0 21 5 4 0 1 0 1 6] + [ 4 5 8 5 1 1 3 2 11 1 972 3 0 12 4 2 2 1 4 2 10] + [ 1 1 0 0 1 12 0 0 0 1 0 961 23 8 0 2 0 17 0 5 3] + [ 1 3 3 1 0 0 0 1 1 0 0 38 977 1 3 6 3 13 2 5 10] + [ 0 0 3 0 2 6 0 0 6 11 7 5 2 1059 4 2 1 0 0 4 7] + [ 16 1 2 9 7 0 0 0 18 2 3 1 4 2 1009 0 2 2 9 0 14] + [ 1 5 0 0 3 0 1 0 0 0 0 9 8 2 0 1070 14 10 0 9 2] + [ 0 21 2 0 7 2 0 0 4 0 0 3 0 1 4 9 1096 0 0 3 9] + [ 1 0 0 0 1 0 1 0 0 0 0 3 18 0 0 5 0 1005 1 1 2] + [ 1 9 6 16 0 0 0 17 2 0 4 0 4 0 8 0 2 0 989 3 7] + [ 0 2 3 3 1 4 12 11 2 0 0 15 3 2 0 5 9 1 3 1069 7] + [ 119 218 165 54 102 139 46 95 85 75 167 124 353 324 141 62 149 84 169 229 5005]] + +2023-10-05 21:48:00,049 - ==> Best [Top1: 84.470 Top5: 98.000 Sparsity:0.00 Params: 148928 on epoch: 122] +2023-10-05 21:48:00,049 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:48:00,055 - + +2023-10-05 21:48:00,055 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:48:01,272 - Epoch: [140][ 10/ 1236] Overall Loss 0.242700 Objective Loss 0.242700 LR 0.000250 Time 0.121637 +2023-10-05 21:48:01,480 - Epoch: [140][ 20/ 1236] Overall Loss 0.235359 Objective Loss 0.235359 LR 0.000250 Time 0.071207 +2023-10-05 21:48:01,686 - Epoch: [140][ 30/ 1236] Overall Loss 0.231663 Objective Loss 0.231663 LR 0.000250 Time 0.054306 +2023-10-05 21:48:01,894 - Epoch: [140][ 40/ 1236] Overall Loss 0.230866 Objective Loss 0.230866 LR 0.000250 Time 0.045936 +2023-10-05 21:48:02,099 - Epoch: [140][ 50/ 1236] Overall Loss 0.231548 Objective Loss 0.231548 LR 0.000250 Time 0.040828 +2023-10-05 21:48:02,307 - Epoch: [140][ 60/ 1236] Overall Loss 0.231385 Objective Loss 0.231385 LR 0.000250 Time 0.037485 +2023-10-05 21:48:02,512 - Epoch: [140][ 70/ 1236] Overall Loss 0.229887 Objective Loss 0.229887 LR 0.000250 Time 0.035053 +2023-10-05 21:48:02,721 - Epoch: [140][ 80/ 1236] Overall Loss 0.232619 Objective Loss 0.232619 LR 0.000250 Time 0.033275 +2023-10-05 21:48:02,925 - Epoch: [140][ 90/ 1236] Overall Loss 0.231241 Objective Loss 0.231241 LR 0.000250 Time 0.031840 +2023-10-05 21:48:03,132 - Epoch: [140][ 100/ 1236] Overall Loss 0.230214 Objective Loss 0.230214 LR 0.000250 Time 0.030731 +2023-10-05 21:48:03,337 - Epoch: [140][ 110/ 1236] Overall Loss 0.230250 Objective Loss 0.230250 LR 0.000250 Time 0.029789 +2023-10-05 21:48:03,543 - Epoch: [140][ 120/ 1236] Overall Loss 0.231158 Objective Loss 0.231158 LR 0.000250 Time 0.029020 +2023-10-05 21:48:03,747 - Epoch: [140][ 130/ 1236] Overall Loss 0.230569 Objective Loss 0.230569 LR 0.000250 Time 0.028353 +2023-10-05 21:48:03,950 - Epoch: [140][ 140/ 1236] Overall Loss 0.229470 Objective Loss 0.229470 LR 0.000250 Time 0.027776 +2023-10-05 21:48:04,149 - Epoch: [140][ 150/ 1236] Overall Loss 0.229936 Objective Loss 0.229936 LR 0.000250 Time 0.027249 +2023-10-05 21:48:04,350 - Epoch: [140][ 160/ 1236] Overall Loss 0.228741 Objective Loss 0.228741 LR 0.000250 Time 0.026803 +2023-10-05 21:48:04,549 - Epoch: [140][ 170/ 1236] Overall Loss 0.228771 Objective Loss 0.228771 LR 0.000250 Time 0.026393 +2023-10-05 21:48:04,752 - Epoch: [140][ 180/ 1236] Overall Loss 0.229099 Objective Loss 0.229099 LR 0.000250 Time 0.026056 +2023-10-05 21:48:04,949 - Epoch: [140][ 190/ 1236] Overall Loss 0.229315 Objective Loss 0.229315 LR 0.000250 Time 0.025721 +2023-10-05 21:48:05,147 - Epoch: [140][ 200/ 1236] Overall Loss 0.229663 Objective Loss 0.229663 LR 0.000250 Time 0.025420 +2023-10-05 21:48:05,345 - Epoch: [140][ 210/ 1236] Overall Loss 0.231033 Objective Loss 0.231033 LR 0.000250 Time 0.025150 +2023-10-05 21:48:05,541 - Epoch: [140][ 220/ 1236] Overall Loss 0.231545 Objective Loss 0.231545 LR 0.000250 Time 0.024898 +2023-10-05 21:48:05,739 - Epoch: [140][ 230/ 1236] Overall Loss 0.230618 Objective Loss 0.230618 LR 0.000250 Time 0.024674 +2023-10-05 21:48:05,936 - Epoch: [140][ 240/ 1236] Overall Loss 0.230835 Objective Loss 0.230835 LR 0.000250 Time 0.024467 +2023-10-05 21:48:06,133 - Epoch: [140][ 250/ 1236] Overall Loss 0.231000 Objective Loss 0.231000 LR 0.000250 Time 0.024275 +2023-10-05 21:48:06,330 - Epoch: [140][ 260/ 1236] Overall Loss 0.231596 Objective Loss 0.231596 LR 0.000250 Time 0.024099 +2023-10-05 21:48:06,529 - Epoch: [140][ 270/ 1236] Overall Loss 0.231610 Objective Loss 0.231610 LR 0.000250 Time 0.023940 +2023-10-05 21:48:06,727 - Epoch: [140][ 280/ 1236] Overall Loss 0.232038 Objective Loss 0.232038 LR 0.000250 Time 0.023792 +2023-10-05 21:48:06,925 - Epoch: [140][ 290/ 1236] Overall Loss 0.231753 Objective Loss 0.231753 LR 0.000250 Time 0.023651 +2023-10-05 21:48:07,122 - Epoch: [140][ 300/ 1236] Overall Loss 0.231951 Objective Loss 0.231951 LR 0.000250 Time 0.023519 +2023-10-05 21:48:07,318 - Epoch: [140][ 310/ 1236] Overall Loss 0.232957 Objective Loss 0.232957 LR 0.000250 Time 0.023394 +2023-10-05 21:48:07,516 - Epoch: [140][ 320/ 1236] Overall Loss 0.231797 Objective Loss 0.231797 LR 0.000250 Time 0.023281 +2023-10-05 21:48:07,713 - Epoch: [140][ 330/ 1236] Overall Loss 0.230492 Objective Loss 0.230492 LR 0.000250 Time 0.023172 +2023-10-05 21:48:07,910 - Epoch: [140][ 340/ 1236] Overall Loss 0.230777 Objective Loss 0.230777 LR 0.000250 Time 0.023069 +2023-10-05 21:48:08,107 - Epoch: [140][ 350/ 1236] Overall Loss 0.230505 Objective Loss 0.230505 LR 0.000250 Time 0.022972 +2023-10-05 21:48:08,305 - Epoch: [140][ 360/ 1236] Overall Loss 0.230542 Objective Loss 0.230542 LR 0.000250 Time 0.022881 +2023-10-05 21:48:08,503 - Epoch: [140][ 370/ 1236] Overall Loss 0.231105 Objective Loss 0.231105 LR 0.000250 Time 0.022798 +2023-10-05 21:48:08,700 - Epoch: [140][ 380/ 1236] Overall Loss 0.230589 Objective Loss 0.230589 LR 0.000250 Time 0.022715 +2023-10-05 21:48:08,898 - Epoch: [140][ 390/ 1236] Overall Loss 0.230618 Objective Loss 0.230618 LR 0.000250 Time 0.022639 +2023-10-05 21:48:09,096 - Epoch: [140][ 400/ 1236] Overall Loss 0.231838 Objective Loss 0.231838 LR 0.000250 Time 0.022569 +2023-10-05 21:48:09,294 - Epoch: [140][ 410/ 1236] Overall Loss 0.232339 Objective Loss 0.232339 LR 0.000250 Time 0.022500 +2023-10-05 21:48:09,492 - Epoch: [140][ 420/ 1236] Overall Loss 0.232486 Objective Loss 0.232486 LR 0.000250 Time 0.022435 +2023-10-05 21:48:09,690 - Epoch: [140][ 430/ 1236] Overall Loss 0.232620 Objective Loss 0.232620 LR 0.000250 Time 0.022372 +2023-10-05 21:48:09,889 - Epoch: [140][ 440/ 1236] Overall Loss 0.233145 Objective Loss 0.233145 LR 0.000250 Time 0.022315 +2023-10-05 21:48:10,087 - Epoch: [140][ 450/ 1236] Overall Loss 0.232750 Objective Loss 0.232750 LR 0.000250 Time 0.022258 +2023-10-05 21:48:10,285 - Epoch: [140][ 460/ 1236] Overall Loss 0.232934 Objective Loss 0.232934 LR 0.000250 Time 0.022205 +2023-10-05 21:48:10,485 - Epoch: [140][ 470/ 1236] Overall Loss 0.233600 Objective Loss 0.233600 LR 0.000250 Time 0.022158 +2023-10-05 21:48:10,685 - Epoch: [140][ 480/ 1236] Overall Loss 0.233759 Objective Loss 0.233759 LR 0.000250 Time 0.022112 +2023-10-05 21:48:10,883 - Epoch: [140][ 490/ 1236] Overall Loss 0.233383 Objective Loss 0.233383 LR 0.000250 Time 0.022064 +2023-10-05 21:48:11,081 - Epoch: [140][ 500/ 1236] Overall Loss 0.233046 Objective Loss 0.233046 LR 0.000250 Time 0.022018 +2023-10-05 21:48:11,279 - Epoch: [140][ 510/ 1236] Overall Loss 0.232577 Objective Loss 0.232577 LR 0.000250 Time 0.021973 +2023-10-05 21:48:11,477 - Epoch: [140][ 520/ 1236] Overall Loss 0.232232 Objective Loss 0.232232 LR 0.000250 Time 0.021931 +2023-10-05 21:48:11,675 - Epoch: [140][ 530/ 1236] Overall Loss 0.232234 Objective Loss 0.232234 LR 0.000250 Time 0.021891 +2023-10-05 21:48:11,873 - Epoch: [140][ 540/ 1236] Overall Loss 0.232213 Objective Loss 0.232213 LR 0.000250 Time 0.021852 +2023-10-05 21:48:12,071 - Epoch: [140][ 550/ 1236] Overall Loss 0.232087 Objective Loss 0.232087 LR 0.000250 Time 0.021814 +2023-10-05 21:48:12,268 - Epoch: [140][ 560/ 1236] Overall Loss 0.231793 Objective Loss 0.231793 LR 0.000250 Time 0.021776 +2023-10-05 21:48:12,466 - Epoch: [140][ 570/ 1236] Overall Loss 0.231199 Objective Loss 0.231199 LR 0.000250 Time 0.021741 +2023-10-05 21:48:12,664 - Epoch: [140][ 580/ 1236] Overall Loss 0.231098 Objective Loss 0.231098 LR 0.000250 Time 0.021707 +2023-10-05 21:48:12,862 - Epoch: [140][ 590/ 1236] Overall Loss 0.231071 Objective Loss 0.231071 LR 0.000250 Time 0.021673 +2023-10-05 21:48:13,060 - Epoch: [140][ 600/ 1236] Overall Loss 0.230975 Objective Loss 0.230975 LR 0.000250 Time 0.021642 +2023-10-05 21:48:13,258 - Epoch: [140][ 610/ 1236] Overall Loss 0.230439 Objective Loss 0.230439 LR 0.000250 Time 0.021610 +2023-10-05 21:48:13,456 - Epoch: [140][ 620/ 1236] Overall Loss 0.230088 Objective Loss 0.230088 LR 0.000250 Time 0.021582 +2023-10-05 21:48:13,655 - Epoch: [140][ 630/ 1236] Overall Loss 0.230299 Objective Loss 0.230299 LR 0.000250 Time 0.021555 +2023-10-05 21:48:13,856 - Epoch: [140][ 640/ 1236] Overall Loss 0.230490 Objective Loss 0.230490 LR 0.000250 Time 0.021531 +2023-10-05 21:48:14,054 - Epoch: [140][ 650/ 1236] Overall Loss 0.229943 Objective Loss 0.229943 LR 0.000250 Time 0.021504 +2023-10-05 21:48:14,252 - Epoch: [140][ 660/ 1236] Overall Loss 0.230179 Objective Loss 0.230179 LR 0.000250 Time 0.021477 +2023-10-05 21:48:14,449 - Epoch: [140][ 670/ 1236] Overall Loss 0.229920 Objective Loss 0.229920 LR 0.000250 Time 0.021451 +2023-10-05 21:48:14,646 - Epoch: [140][ 680/ 1236] Overall Loss 0.229792 Objective Loss 0.229792 LR 0.000250 Time 0.021425 +2023-10-05 21:48:14,843 - Epoch: [140][ 690/ 1236] Overall Loss 0.229731 Objective Loss 0.229731 LR 0.000250 Time 0.021399 +2023-10-05 21:48:15,040 - Epoch: [140][ 700/ 1236] Overall Loss 0.230164 Objective Loss 0.230164 LR 0.000250 Time 0.021375 +2023-10-05 21:48:15,238 - Epoch: [140][ 710/ 1236] Overall Loss 0.230212 Objective Loss 0.230212 LR 0.000250 Time 0.021351 +2023-10-05 21:48:15,436 - Epoch: [140][ 720/ 1236] Overall Loss 0.230473 Objective Loss 0.230473 LR 0.000250 Time 0.021329 +2023-10-05 21:48:15,634 - Epoch: [140][ 730/ 1236] Overall Loss 0.230588 Objective Loss 0.230588 LR 0.000250 Time 0.021307 +2023-10-05 21:48:15,831 - Epoch: [140][ 740/ 1236] Overall Loss 0.230555 Objective Loss 0.230555 LR 0.000250 Time 0.021286 +2023-10-05 21:48:16,029 - Epoch: [140][ 750/ 1236] Overall Loss 0.230521 Objective Loss 0.230521 LR 0.000250 Time 0.021265 +2023-10-05 21:48:16,226 - Epoch: [140][ 760/ 1236] Overall Loss 0.230498 Objective Loss 0.230498 LR 0.000250 Time 0.021244 +2023-10-05 21:48:16,423 - Epoch: [140][ 770/ 1236] Overall Loss 0.230102 Objective Loss 0.230102 LR 0.000250 Time 0.021225 +2023-10-05 21:48:16,621 - Epoch: [140][ 780/ 1236] Overall Loss 0.230298 Objective Loss 0.230298 LR 0.000250 Time 0.021205 +2023-10-05 21:48:16,818 - Epoch: [140][ 790/ 1236] Overall Loss 0.230225 Objective Loss 0.230225 LR 0.000250 Time 0.021186 +2023-10-05 21:48:17,015 - Epoch: [140][ 800/ 1236] Overall Loss 0.230257 Objective Loss 0.230257 LR 0.000250 Time 0.021167 +2023-10-05 21:48:17,212 - Epoch: [140][ 810/ 1236] Overall Loss 0.230303 Objective Loss 0.230303 LR 0.000250 Time 0.021149 +2023-10-05 21:48:17,410 - Epoch: [140][ 820/ 1236] Overall Loss 0.230248 Objective Loss 0.230248 LR 0.000250 Time 0.021132 +2023-10-05 21:48:17,608 - Epoch: [140][ 830/ 1236] Overall Loss 0.230130 Objective Loss 0.230130 LR 0.000250 Time 0.021115 +2023-10-05 21:48:17,805 - Epoch: [140][ 840/ 1236] Overall Loss 0.230264 Objective Loss 0.230264 LR 0.000250 Time 0.021098 +2023-10-05 21:48:18,002 - Epoch: [140][ 850/ 1236] Overall Loss 0.230316 Objective Loss 0.230316 LR 0.000250 Time 0.021082 +2023-10-05 21:48:18,199 - Epoch: [140][ 860/ 1236] Overall Loss 0.230210 Objective Loss 0.230210 LR 0.000250 Time 0.021065 +2023-10-05 21:48:18,396 - Epoch: [140][ 870/ 1236] Overall Loss 0.230402 Objective Loss 0.230402 LR 0.000250 Time 0.021049 +2023-10-05 21:48:18,593 - Epoch: [140][ 880/ 1236] Overall Loss 0.230127 Objective Loss 0.230127 LR 0.000250 Time 0.021034 +2023-10-05 21:48:18,790 - Epoch: [140][ 890/ 1236] Overall Loss 0.230158 Objective Loss 0.230158 LR 0.000250 Time 0.021019 +2023-10-05 21:48:18,987 - Epoch: [140][ 900/ 1236] Overall Loss 0.229938 Objective Loss 0.229938 LR 0.000250 Time 0.021003 +2023-10-05 21:48:19,185 - Epoch: [140][ 910/ 1236] Overall Loss 0.230013 Objective Loss 0.230013 LR 0.000250 Time 0.020989 +2023-10-05 21:48:19,382 - Epoch: [140][ 920/ 1236] Overall Loss 0.229993 Objective Loss 0.229993 LR 0.000250 Time 0.020975 +2023-10-05 21:48:19,579 - Epoch: [140][ 930/ 1236] Overall Loss 0.230113 Objective Loss 0.230113 LR 0.000250 Time 0.020962 +2023-10-05 21:48:19,777 - Epoch: [140][ 940/ 1236] Overall Loss 0.230008 Objective Loss 0.230008 LR 0.000250 Time 0.020948 +2023-10-05 21:48:19,974 - Epoch: [140][ 950/ 1236] Overall Loss 0.229736 Objective Loss 0.229736 LR 0.000250 Time 0.020935 +2023-10-05 21:48:20,171 - Epoch: [140][ 960/ 1236] Overall Loss 0.230097 Objective Loss 0.230097 LR 0.000250 Time 0.020922 +2023-10-05 21:48:20,368 - Epoch: [140][ 970/ 1236] Overall Loss 0.229938 Objective Loss 0.229938 LR 0.000250 Time 0.020909 +2023-10-05 21:48:20,567 - Epoch: [140][ 980/ 1236] Overall Loss 0.229640 Objective Loss 0.229640 LR 0.000250 Time 0.020898 +2023-10-05 21:48:20,764 - Epoch: [140][ 990/ 1236] Overall Loss 0.229581 Objective Loss 0.229581 LR 0.000250 Time 0.020886 +2023-10-05 21:48:20,962 - Epoch: [140][ 1000/ 1236] Overall Loss 0.229518 Objective Loss 0.229518 LR 0.000250 Time 0.020875 +2023-10-05 21:48:21,161 - Epoch: [140][ 1010/ 1236] Overall Loss 0.229618 Objective Loss 0.229618 LR 0.000250 Time 0.020865 +2023-10-05 21:48:21,360 - Epoch: [140][ 1020/ 1236] Overall Loss 0.229563 Objective Loss 0.229563 LR 0.000250 Time 0.020855 +2023-10-05 21:48:21,558 - Epoch: [140][ 1030/ 1236] Overall Loss 0.229550 Objective Loss 0.229550 LR 0.000250 Time 0.020845 +2023-10-05 21:48:21,756 - Epoch: [140][ 1040/ 1236] Overall Loss 0.229518 Objective Loss 0.229518 LR 0.000250 Time 0.020834 +2023-10-05 21:48:21,956 - Epoch: [140][ 1050/ 1236] Overall Loss 0.229608 Objective Loss 0.229608 LR 0.000250 Time 0.020826 +2023-10-05 21:48:22,154 - Epoch: [140][ 1060/ 1236] Overall Loss 0.229615 Objective Loss 0.229615 LR 0.000250 Time 0.020816 +2023-10-05 21:48:22,352 - Epoch: [140][ 1070/ 1236] Overall Loss 0.229852 Objective Loss 0.229852 LR 0.000250 Time 0.020807 +2023-10-05 21:48:22,551 - Epoch: [140][ 1080/ 1236] Overall Loss 0.229854 Objective Loss 0.229854 LR 0.000250 Time 0.020797 +2023-10-05 21:48:22,749 - Epoch: [140][ 1090/ 1236] Overall Loss 0.229988 Objective Loss 0.229988 LR 0.000250 Time 0.020788 +2023-10-05 21:48:22,947 - Epoch: [140][ 1100/ 1236] Overall Loss 0.229923 Objective Loss 0.229923 LR 0.000250 Time 0.020779 +2023-10-05 21:48:23,145 - Epoch: [140][ 1110/ 1236] Overall Loss 0.229690 Objective Loss 0.229690 LR 0.000250 Time 0.020770 +2023-10-05 21:48:23,343 - Epoch: [140][ 1120/ 1236] Overall Loss 0.229674 Objective Loss 0.229674 LR 0.000250 Time 0.020761 +2023-10-05 21:48:23,541 - Epoch: [140][ 1130/ 1236] Overall Loss 0.229715 Objective Loss 0.229715 LR 0.000250 Time 0.020753 +2023-10-05 21:48:23,739 - Epoch: [140][ 1140/ 1236] Overall Loss 0.229781 Objective Loss 0.229781 LR 0.000250 Time 0.020744 +2023-10-05 21:48:23,939 - Epoch: [140][ 1150/ 1236] Overall Loss 0.229813 Objective Loss 0.229813 LR 0.000250 Time 0.020737 +2023-10-05 21:48:24,137 - Epoch: [140][ 1160/ 1236] Overall Loss 0.229945 Objective Loss 0.229945 LR 0.000250 Time 0.020729 +2023-10-05 21:48:24,335 - Epoch: [140][ 1170/ 1236] Overall Loss 0.230057 Objective Loss 0.230057 LR 0.000250 Time 0.020721 +2023-10-05 21:48:24,533 - Epoch: [140][ 1180/ 1236] Overall Loss 0.230039 Objective Loss 0.230039 LR 0.000250 Time 0.020712 +2023-10-05 21:48:24,731 - Epoch: [140][ 1190/ 1236] Overall Loss 0.230178 Objective Loss 0.230178 LR 0.000250 Time 0.020704 +2023-10-05 21:48:24,928 - Epoch: [140][ 1200/ 1236] Overall Loss 0.230302 Objective Loss 0.230302 LR 0.000250 Time 0.020696 +2023-10-05 21:48:25,126 - Epoch: [140][ 1210/ 1236] Overall Loss 0.230314 Objective Loss 0.230314 LR 0.000250 Time 0.020688 +2023-10-05 21:48:25,323 - Epoch: [140][ 1220/ 1236] Overall Loss 0.230149 Objective Loss 0.230149 LR 0.000250 Time 0.020680 +2023-10-05 21:48:25,572 - Epoch: [140][ 1230/ 1236] Overall Loss 0.230282 Objective Loss 0.230282 LR 0.000250 Time 0.020714 +2023-10-05 21:48:25,691 - Epoch: [140][ 1236/ 1236] Overall Loss 0.230487 Objective Loss 0.230487 Top1 84.725051 Top5 97.352342 LR 0.000250 Time 0.020710 +2023-10-05 21:48:25,818 - --- validate (epoch=140)----------- +2023-10-05 21:48:25,818 - 29943 samples (256 per mini-batch) +2023-10-05 21:48:26,311 - Epoch: [140][ 10/ 117] Loss 0.326884 Top1 85.390625 Top5 97.734375 +2023-10-05 21:48:26,464 - Epoch: [140][ 20/ 117] Loss 0.313785 Top1 84.921875 Top5 97.988281 +2023-10-05 21:48:26,618 - Epoch: [140][ 30/ 117] Loss 0.315760 Top1 84.986979 Top5 97.955729 +2023-10-05 21:48:26,766 - Epoch: [140][ 40/ 117] Loss 0.303973 Top1 85.058594 Top5 97.978516 +2023-10-05 21:48:26,918 - Epoch: [140][ 50/ 117] Loss 0.310450 Top1 84.890625 Top5 97.835938 +2023-10-05 21:48:27,065 - Epoch: [140][ 60/ 117] Loss 0.311857 Top1 85.032552 Top5 97.858073 +2023-10-05 21:48:27,222 - Epoch: [140][ 70/ 117] Loss 0.314049 Top1 84.988839 Top5 97.862723 +2023-10-05 21:48:27,373 - Epoch: [140][ 80/ 117] Loss 0.314798 Top1 84.946289 Top5 97.866211 +2023-10-05 21:48:27,527 - Epoch: [140][ 90/ 117] Loss 0.316450 Top1 84.930556 Top5 97.938368 +2023-10-05 21:48:27,681 - Epoch: [140][ 100/ 117] Loss 0.315052 Top1 84.820312 Top5 97.964844 +2023-10-05 21:48:27,844 - Epoch: [140][ 110/ 117] Loss 0.315952 Top1 84.733665 Top5 97.936790 +2023-10-05 21:48:27,932 - Epoch: [140][ 117/ 117] Loss 0.316327 Top1 84.757706 Top5 97.922720 +2023-10-05 21:48:28,076 - ==> Top1: 84.758 Top5: 97.923 Loss: 0.316 + +2023-10-05 21:48:28,077 - ==> Confusion: +[[ 928 3 4 2 9 2 1 0 5 73 1 0 0 1 4 0 3 1 0 0 13] + [ 1 1065 3 0 11 18 1 12 0 0 0 3 0 0 0 2 1 0 5 0 9] + [ 5 2 960 13 2 1 26 6 0 0 4 3 10 1 1 4 0 1 4 3 10] + [ 1 1 16 955 1 5 2 2 2 1 7 0 8 3 29 6 0 7 24 2 17] + [ 25 7 1 0 978 3 0 1 0 10 0 1 0 2 3 5 5 3 0 2 4] + [ 5 44 0 0 4 985 0 22 0 1 4 8 1 9 4 2 4 1 3 4 15] + [ 0 8 24 0 0 1 1120 6 0 0 2 4 2 0 0 6 0 2 2 7 7] + [ 3 17 18 0 3 22 5 1071 1 4 3 13 0 2 1 3 0 0 36 6 10] + [ 19 2 0 0 3 4 0 0 971 48 10 2 0 7 14 4 1 1 2 0 1] + [ 97 0 4 0 7 4 0 0 20 954 0 3 0 10 5 4 0 2 0 2 7] + [ 5 4 9 6 2 1 5 5 13 3 957 4 1 13 2 2 3 0 5 3 10] + [ 1 0 1 0 0 12 0 2 0 1 0 971 13 4 0 3 0 15 0 8 4] + [ 1 1 4 4 0 1 0 3 0 0 1 38 974 2 2 7 2 14 2 1 11] + [ 2 1 4 0 2 7 0 0 11 20 5 7 2 1042 2 2 2 0 0 1 9] + [ 13 2 2 13 6 0 1 0 23 3 1 1 3 1 1003 0 1 2 8 0 18] + [ 0 3 2 0 2 0 2 0 0 0 0 11 5 1 1 1068 14 12 0 10 3] + [ 2 12 1 0 6 3 0 0 1 0 0 6 1 0 3 10 1099 0 0 3 14] + [ 0 0 0 1 1 1 3 0 0 0 0 2 21 1 0 4 0 999 1 0 4] + [ 1 12 7 19 0 0 0 25 1 0 2 0 1 0 7 0 1 0 981 2 9] + [ 1 2 1 2 1 3 13 8 1 0 1 17 5 2 0 5 8 1 4 1062 15] + [ 139 189 158 53 107 121 49 90 105 105 172 116 313 253 138 63 128 70 131 169 5236]] + +2023-10-05 21:48:28,078 - ==> Best [Top1: 84.758 Top5: 97.923 Sparsity:0.00 Params: 148928 on epoch: 140] +2023-10-05 21:48:28,078 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:48:28,085 - + +2023-10-05 21:48:28,085 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:48:29,086 - Epoch: [141][ 10/ 1236] Overall Loss 0.239085 Objective Loss 0.239085 LR 0.000250 Time 0.100063 +2023-10-05 21:48:29,290 - Epoch: [141][ 20/ 1236] Overall Loss 0.230912 Objective Loss 0.230912 LR 0.000250 Time 0.060205 +2023-10-05 21:48:29,494 - Epoch: [141][ 30/ 1236] Overall Loss 0.223968 Objective Loss 0.223968 LR 0.000250 Time 0.046912 +2023-10-05 21:48:29,698 - Epoch: [141][ 40/ 1236] Overall Loss 0.221972 Objective Loss 0.221972 LR 0.000250 Time 0.040280 +2023-10-05 21:48:29,903 - Epoch: [141][ 50/ 1236] Overall Loss 0.223969 Objective Loss 0.223969 LR 0.000250 Time 0.036313 +2023-10-05 21:48:30,108 - Epoch: [141][ 60/ 1236] Overall Loss 0.225575 Objective Loss 0.225575 LR 0.000250 Time 0.033679 +2023-10-05 21:48:30,314 - Epoch: [141][ 70/ 1236] Overall Loss 0.227762 Objective Loss 0.227762 LR 0.000250 Time 0.031805 +2023-10-05 21:48:30,522 - Epoch: [141][ 80/ 1236] Overall Loss 0.229533 Objective Loss 0.229533 LR 0.000250 Time 0.030427 +2023-10-05 21:48:30,729 - Epoch: [141][ 90/ 1236] Overall Loss 0.230891 Objective Loss 0.230891 LR 0.000250 Time 0.029344 +2023-10-05 21:48:30,927 - Epoch: [141][ 100/ 1236] Overall Loss 0.229817 Objective Loss 0.229817 LR 0.000250 Time 0.028385 +2023-10-05 21:48:31,124 - Epoch: [141][ 110/ 1236] Overall Loss 0.230118 Objective Loss 0.230118 LR 0.000250 Time 0.027595 +2023-10-05 21:48:31,322 - Epoch: [141][ 120/ 1236] Overall Loss 0.229340 Objective Loss 0.229340 LR 0.000250 Time 0.026938 +2023-10-05 21:48:31,519 - Epoch: [141][ 130/ 1236] Overall Loss 0.233136 Objective Loss 0.233136 LR 0.000250 Time 0.026380 +2023-10-05 21:48:31,716 - Epoch: [141][ 140/ 1236] Overall Loss 0.232238 Objective Loss 0.232238 LR 0.000250 Time 0.025902 +2023-10-05 21:48:31,913 - Epoch: [141][ 150/ 1236] Overall Loss 0.232285 Objective Loss 0.232285 LR 0.000250 Time 0.025487 +2023-10-05 21:48:32,110 - Epoch: [141][ 160/ 1236] Overall Loss 0.231894 Objective Loss 0.231894 LR 0.000250 Time 0.025120 +2023-10-05 21:48:32,308 - Epoch: [141][ 170/ 1236] Overall Loss 0.230771 Objective Loss 0.230771 LR 0.000250 Time 0.024808 +2023-10-05 21:48:32,505 - Epoch: [141][ 180/ 1236] Overall Loss 0.228544 Objective Loss 0.228544 LR 0.000250 Time 0.024521 +2023-10-05 21:48:32,715 - Epoch: [141][ 190/ 1236] Overall Loss 0.227381 Objective Loss 0.227381 LR 0.000250 Time 0.024335 +2023-10-05 21:48:32,924 - Epoch: [141][ 200/ 1236] Overall Loss 0.228471 Objective Loss 0.228471 LR 0.000250 Time 0.024160 +2023-10-05 21:48:33,135 - Epoch: [141][ 210/ 1236] Overall Loss 0.227627 Objective Loss 0.227627 LR 0.000250 Time 0.024016 +2023-10-05 21:48:33,347 - Epoch: [141][ 220/ 1236] Overall Loss 0.228970 Objective Loss 0.228970 LR 0.000250 Time 0.023884 +2023-10-05 21:48:33,559 - Epoch: [141][ 230/ 1236] Overall Loss 0.229770 Objective Loss 0.229770 LR 0.000250 Time 0.023765 +2023-10-05 21:48:33,771 - Epoch: [141][ 240/ 1236] Overall Loss 0.230170 Objective Loss 0.230170 LR 0.000250 Time 0.023657 +2023-10-05 21:48:33,983 - Epoch: [141][ 250/ 1236] Overall Loss 0.229899 Objective Loss 0.229899 LR 0.000250 Time 0.023556 +2023-10-05 21:48:34,195 - Epoch: [141][ 260/ 1236] Overall Loss 0.229552 Objective Loss 0.229552 LR 0.000250 Time 0.023463 +2023-10-05 21:48:34,407 - Epoch: [141][ 270/ 1236] Overall Loss 0.229262 Objective Loss 0.229262 LR 0.000250 Time 0.023377 +2023-10-05 21:48:34,619 - Epoch: [141][ 280/ 1236] Overall Loss 0.229641 Objective Loss 0.229641 LR 0.000250 Time 0.023297 +2023-10-05 21:48:34,831 - Epoch: [141][ 290/ 1236] Overall Loss 0.229932 Objective Loss 0.229932 LR 0.000250 Time 0.023223 +2023-10-05 21:48:35,043 - Epoch: [141][ 300/ 1236] Overall Loss 0.229914 Objective Loss 0.229914 LR 0.000250 Time 0.023154 +2023-10-05 21:48:35,255 - Epoch: [141][ 310/ 1236] Overall Loss 0.229562 Objective Loss 0.229562 LR 0.000250 Time 0.023089 +2023-10-05 21:48:35,466 - Epoch: [141][ 320/ 1236] Overall Loss 0.229193 Objective Loss 0.229193 LR 0.000250 Time 0.023026 +2023-10-05 21:48:35,677 - Epoch: [141][ 330/ 1236] Overall Loss 0.228777 Objective Loss 0.228777 LR 0.000250 Time 0.022966 +2023-10-05 21:48:35,888 - Epoch: [141][ 340/ 1236] Overall Loss 0.228710 Objective Loss 0.228710 LR 0.000250 Time 0.022912 +2023-10-05 21:48:36,100 - Epoch: [141][ 350/ 1236] Overall Loss 0.228763 Objective Loss 0.228763 LR 0.000250 Time 0.022861 +2023-10-05 21:48:36,312 - Epoch: [141][ 360/ 1236] Overall Loss 0.228989 Objective Loss 0.228989 LR 0.000250 Time 0.022813 +2023-10-05 21:48:36,524 - Epoch: [141][ 370/ 1236] Overall Loss 0.228649 Objective Loss 0.228649 LR 0.000250 Time 0.022768 +2023-10-05 21:48:36,736 - Epoch: [141][ 380/ 1236] Overall Loss 0.228828 Objective Loss 0.228828 LR 0.000250 Time 0.022726 +2023-10-05 21:48:36,948 - Epoch: [141][ 390/ 1236] Overall Loss 0.229072 Objective Loss 0.229072 LR 0.000250 Time 0.022685 +2023-10-05 21:48:37,160 - Epoch: [141][ 400/ 1236] Overall Loss 0.229053 Objective Loss 0.229053 LR 0.000250 Time 0.022646 +2023-10-05 21:48:37,372 - Epoch: [141][ 410/ 1236] Overall Loss 0.229100 Objective Loss 0.229100 LR 0.000250 Time 0.022610 +2023-10-05 21:48:37,583 - Epoch: [141][ 420/ 1236] Overall Loss 0.228721 Objective Loss 0.228721 LR 0.000250 Time 0.022574 +2023-10-05 21:48:37,795 - Epoch: [141][ 430/ 1236] Overall Loss 0.228356 Objective Loss 0.228356 LR 0.000250 Time 0.022541 +2023-10-05 21:48:38,007 - Epoch: [141][ 440/ 1236] Overall Loss 0.227816 Objective Loss 0.227816 LR 0.000250 Time 0.022509 +2023-10-05 21:48:38,219 - Epoch: [141][ 450/ 1236] Overall Loss 0.226716 Objective Loss 0.226716 LR 0.000250 Time 0.022478 +2023-10-05 21:48:38,431 - Epoch: [141][ 460/ 1236] Overall Loss 0.226380 Objective Loss 0.226380 LR 0.000250 Time 0.022449 +2023-10-05 21:48:38,642 - Epoch: [141][ 470/ 1236] Overall Loss 0.225809 Objective Loss 0.225809 LR 0.000250 Time 0.022419 +2023-10-05 21:48:38,851 - Epoch: [141][ 480/ 1236] Overall Loss 0.225384 Objective Loss 0.225384 LR 0.000250 Time 0.022387 +2023-10-05 21:48:39,059 - Epoch: [141][ 490/ 1236] Overall Loss 0.225308 Objective Loss 0.225308 LR 0.000250 Time 0.022354 +2023-10-05 21:48:39,269 - Epoch: [141][ 500/ 1236] Overall Loss 0.225312 Objective Loss 0.225312 LR 0.000250 Time 0.022328 +2023-10-05 21:48:39,481 - Epoch: [141][ 510/ 1236] Overall Loss 0.225785 Objective Loss 0.225785 LR 0.000250 Time 0.022304 +2023-10-05 21:48:39,693 - Epoch: [141][ 520/ 1236] Overall Loss 0.225896 Objective Loss 0.225896 LR 0.000250 Time 0.022282 +2023-10-05 21:48:39,905 - Epoch: [141][ 530/ 1236] Overall Loss 0.225519 Objective Loss 0.225519 LR 0.000250 Time 0.022261 +2023-10-05 21:48:40,117 - Epoch: [141][ 540/ 1236] Overall Loss 0.225460 Objective Loss 0.225460 LR 0.000250 Time 0.022240 +2023-10-05 21:48:40,330 - Epoch: [141][ 550/ 1236] Overall Loss 0.225791 Objective Loss 0.225791 LR 0.000250 Time 0.022221 +2023-10-05 21:48:40,542 - Epoch: [141][ 560/ 1236] Overall Loss 0.225796 Objective Loss 0.225796 LR 0.000250 Time 0.022202 +2023-10-05 21:48:40,754 - Epoch: [141][ 570/ 1236] Overall Loss 0.225374 Objective Loss 0.225374 LR 0.000250 Time 0.022184 +2023-10-05 21:48:40,966 - Epoch: [141][ 580/ 1236] Overall Loss 0.224928 Objective Loss 0.224928 LR 0.000250 Time 0.022166 +2023-10-05 21:48:41,178 - Epoch: [141][ 590/ 1236] Overall Loss 0.224955 Objective Loss 0.224955 LR 0.000250 Time 0.022149 +2023-10-05 21:48:41,390 - Epoch: [141][ 600/ 1236] Overall Loss 0.225084 Objective Loss 0.225084 LR 0.000250 Time 0.022132 +2023-10-05 21:48:41,602 - Epoch: [141][ 610/ 1236] Overall Loss 0.225001 Objective Loss 0.225001 LR 0.000250 Time 0.022116 +2023-10-05 21:48:41,814 - Epoch: [141][ 620/ 1236] Overall Loss 0.225203 Objective Loss 0.225203 LR 0.000250 Time 0.022101 +2023-10-05 21:48:42,025 - Epoch: [141][ 630/ 1236] Overall Loss 0.225363 Objective Loss 0.225363 LR 0.000250 Time 0.022084 +2023-10-05 21:48:42,236 - Epoch: [141][ 640/ 1236] Overall Loss 0.225232 Objective Loss 0.225232 LR 0.000250 Time 0.022069 +2023-10-05 21:48:42,448 - Epoch: [141][ 650/ 1236] Overall Loss 0.225294 Objective Loss 0.225294 LR 0.000250 Time 0.022054 +2023-10-05 21:48:42,660 - Epoch: [141][ 660/ 1236] Overall Loss 0.225267 Objective Loss 0.225267 LR 0.000250 Time 0.022040 +2023-10-05 21:48:42,871 - Epoch: [141][ 670/ 1236] Overall Loss 0.224985 Objective Loss 0.224985 LR 0.000250 Time 0.022026 +2023-10-05 21:48:43,083 - Epoch: [141][ 680/ 1236] Overall Loss 0.225242 Objective Loss 0.225242 LR 0.000250 Time 0.022013 +2023-10-05 21:48:43,295 - Epoch: [141][ 690/ 1236] Overall Loss 0.225251 Objective Loss 0.225251 LR 0.000250 Time 0.022000 +2023-10-05 21:48:43,506 - Epoch: [141][ 700/ 1236] Overall Loss 0.224988 Objective Loss 0.224988 LR 0.000250 Time 0.021987 +2023-10-05 21:48:43,717 - Epoch: [141][ 710/ 1236] Overall Loss 0.225127 Objective Loss 0.225127 LR 0.000250 Time 0.021974 +2023-10-05 21:48:43,928 - Epoch: [141][ 720/ 1236] Overall Loss 0.225239 Objective Loss 0.225239 LR 0.000250 Time 0.021961 +2023-10-05 21:48:44,140 - Epoch: [141][ 730/ 1236] Overall Loss 0.224845 Objective Loss 0.224845 LR 0.000250 Time 0.021950 +2023-10-05 21:48:44,348 - Epoch: [141][ 740/ 1236] Overall Loss 0.224549 Objective Loss 0.224549 LR 0.000250 Time 0.021934 +2023-10-05 21:48:44,556 - Epoch: [141][ 750/ 1236] Overall Loss 0.224677 Objective Loss 0.224677 LR 0.000250 Time 0.021918 +2023-10-05 21:48:44,765 - Epoch: [141][ 760/ 1236] Overall Loss 0.224627 Objective Loss 0.224627 LR 0.000250 Time 0.021904 +2023-10-05 21:48:44,973 - Epoch: [141][ 770/ 1236] Overall Loss 0.224409 Objective Loss 0.224409 LR 0.000250 Time 0.021889 +2023-10-05 21:48:45,183 - Epoch: [141][ 780/ 1236] Overall Loss 0.224695 Objective Loss 0.224695 LR 0.000250 Time 0.021877 +2023-10-05 21:48:45,393 - Epoch: [141][ 790/ 1236] Overall Loss 0.224598 Objective Loss 0.224598 LR 0.000250 Time 0.021866 +2023-10-05 21:48:45,603 - Epoch: [141][ 800/ 1236] Overall Loss 0.224429 Objective Loss 0.224429 LR 0.000250 Time 0.021855 +2023-10-05 21:48:45,814 - Epoch: [141][ 810/ 1236] Overall Loss 0.224520 Objective Loss 0.224520 LR 0.000250 Time 0.021844 +2023-10-05 21:48:46,023 - Epoch: [141][ 820/ 1236] Overall Loss 0.224629 Objective Loss 0.224629 LR 0.000250 Time 0.021833 +2023-10-05 21:48:46,234 - Epoch: [141][ 830/ 1236] Overall Loss 0.224618 Objective Loss 0.224618 LR 0.000250 Time 0.021823 +2023-10-05 21:48:46,444 - Epoch: [141][ 840/ 1236] Overall Loss 0.224617 Objective Loss 0.224617 LR 0.000250 Time 0.021813 +2023-10-05 21:48:46,654 - Epoch: [141][ 850/ 1236] Overall Loss 0.224719 Objective Loss 0.224719 LR 0.000250 Time 0.021803 +2023-10-05 21:48:46,864 - Epoch: [141][ 860/ 1236] Overall Loss 0.224725 Objective Loss 0.224725 LR 0.000250 Time 0.021793 +2023-10-05 21:48:47,073 - Epoch: [141][ 870/ 1236] Overall Loss 0.224859 Objective Loss 0.224859 LR 0.000250 Time 0.021783 +2023-10-05 21:48:47,283 - Epoch: [141][ 880/ 1236] Overall Loss 0.224998 Objective Loss 0.224998 LR 0.000250 Time 0.021773 +2023-10-05 21:48:47,493 - Epoch: [141][ 890/ 1236] Overall Loss 0.225083 Objective Loss 0.225083 LR 0.000250 Time 0.021764 +2023-10-05 21:48:47,703 - Epoch: [141][ 900/ 1236] Overall Loss 0.225078 Objective Loss 0.225078 LR 0.000250 Time 0.021755 +2023-10-05 21:48:47,913 - Epoch: [141][ 910/ 1236] Overall Loss 0.225174 Objective Loss 0.225174 LR 0.000250 Time 0.021746 +2023-10-05 21:48:48,123 - Epoch: [141][ 920/ 1236] Overall Loss 0.225355 Objective Loss 0.225355 LR 0.000250 Time 0.021737 +2023-10-05 21:48:48,333 - Epoch: [141][ 930/ 1236] Overall Loss 0.225358 Objective Loss 0.225358 LR 0.000250 Time 0.021729 +2023-10-05 21:48:48,543 - Epoch: [141][ 940/ 1236] Overall Loss 0.225347 Objective Loss 0.225347 LR 0.000250 Time 0.021721 +2023-10-05 21:48:48,753 - Epoch: [141][ 950/ 1236] Overall Loss 0.225449 Objective Loss 0.225449 LR 0.000250 Time 0.021713 +2023-10-05 21:48:48,963 - Epoch: [141][ 960/ 1236] Overall Loss 0.225283 Objective Loss 0.225283 LR 0.000250 Time 0.021705 +2023-10-05 21:48:49,172 - Epoch: [141][ 970/ 1236] Overall Loss 0.225405 Objective Loss 0.225405 LR 0.000250 Time 0.021697 +2023-10-05 21:48:49,380 - Epoch: [141][ 980/ 1236] Overall Loss 0.225709 Objective Loss 0.225709 LR 0.000250 Time 0.021686 +2023-10-05 21:48:49,591 - Epoch: [141][ 990/ 1236] Overall Loss 0.225855 Objective Loss 0.225855 LR 0.000250 Time 0.021680 +2023-10-05 21:48:49,797 - Epoch: [141][ 1000/ 1236] Overall Loss 0.225847 Objective Loss 0.225847 LR 0.000250 Time 0.021669 +2023-10-05 21:48:50,003 - Epoch: [141][ 1010/ 1236] Overall Loss 0.225824 Objective Loss 0.225824 LR 0.000250 Time 0.021658 +2023-10-05 21:48:50,211 - Epoch: [141][ 1020/ 1236] Overall Loss 0.225714 Objective Loss 0.225714 LR 0.000250 Time 0.021649 +2023-10-05 21:48:50,418 - Epoch: [141][ 1030/ 1236] Overall Loss 0.225734 Objective Loss 0.225734 LR 0.000250 Time 0.021640 +2023-10-05 21:48:50,626 - Epoch: [141][ 1040/ 1236] Overall Loss 0.225497 Objective Loss 0.225497 LR 0.000250 Time 0.021631 +2023-10-05 21:48:50,834 - Epoch: [141][ 1050/ 1236] Overall Loss 0.225479 Objective Loss 0.225479 LR 0.000250 Time 0.021622 +2023-10-05 21:48:51,042 - Epoch: [141][ 1060/ 1236] Overall Loss 0.225536 Objective Loss 0.225536 LR 0.000250 Time 0.021614 +2023-10-05 21:48:51,249 - Epoch: [141][ 1070/ 1236] Overall Loss 0.225497 Objective Loss 0.225497 LR 0.000250 Time 0.021605 +2023-10-05 21:48:51,456 - Epoch: [141][ 1080/ 1236] Overall Loss 0.225535 Objective Loss 0.225535 LR 0.000250 Time 0.021597 +2023-10-05 21:48:51,664 - Epoch: [141][ 1090/ 1236] Overall Loss 0.225492 Objective Loss 0.225492 LR 0.000250 Time 0.021589 +2023-10-05 21:48:51,871 - Epoch: [141][ 1100/ 1236] Overall Loss 0.225653 Objective Loss 0.225653 LR 0.000250 Time 0.021581 +2023-10-05 21:48:52,079 - Epoch: [141][ 1110/ 1236] Overall Loss 0.225209 Objective Loss 0.225209 LR 0.000250 Time 0.021573 +2023-10-05 21:48:52,286 - Epoch: [141][ 1120/ 1236] Overall Loss 0.225118 Objective Loss 0.225118 LR 0.000250 Time 0.021566 +2023-10-05 21:48:52,494 - Epoch: [141][ 1130/ 1236] Overall Loss 0.225073 Objective Loss 0.225073 LR 0.000250 Time 0.021558 +2023-10-05 21:48:52,702 - Epoch: [141][ 1140/ 1236] Overall Loss 0.224897 Objective Loss 0.224897 LR 0.000250 Time 0.021551 +2023-10-05 21:48:52,909 - Epoch: [141][ 1150/ 1236] Overall Loss 0.224902 Objective Loss 0.224902 LR 0.000250 Time 0.021543 +2023-10-05 21:48:53,117 - Epoch: [141][ 1160/ 1236] Overall Loss 0.225091 Objective Loss 0.225091 LR 0.000250 Time 0.021536 +2023-10-05 21:48:53,325 - Epoch: [141][ 1170/ 1236] Overall Loss 0.224951 Objective Loss 0.224951 LR 0.000250 Time 0.021530 +2023-10-05 21:48:53,534 - Epoch: [141][ 1180/ 1236] Overall Loss 0.224933 Objective Loss 0.224933 LR 0.000250 Time 0.021524 +2023-10-05 21:48:53,744 - Epoch: [141][ 1190/ 1236] Overall Loss 0.224721 Objective Loss 0.224721 LR 0.000250 Time 0.021520 +2023-10-05 21:48:53,950 - Epoch: [141][ 1200/ 1236] Overall Loss 0.224827 Objective Loss 0.224827 LR 0.000250 Time 0.021512 +2023-10-05 21:48:54,155 - Epoch: [141][ 1210/ 1236] Overall Loss 0.224965 Objective Loss 0.224965 LR 0.000250 Time 0.021503 +2023-10-05 21:48:54,361 - Epoch: [141][ 1220/ 1236] Overall Loss 0.224869 Objective Loss 0.224869 LR 0.000250 Time 0.021495 +2023-10-05 21:48:54,617 - Epoch: [141][ 1230/ 1236] Overall Loss 0.224994 Objective Loss 0.224994 LR 0.000250 Time 0.021528 +2023-10-05 21:48:54,735 - Epoch: [141][ 1236/ 1236] Overall Loss 0.224980 Objective Loss 0.224980 Top1 86.354379 Top5 97.556008 LR 0.000250 Time 0.021519 +2023-10-05 21:48:54,852 - --- validate (epoch=141)----------- +2023-10-05 21:48:54,852 - 29943 samples (256 per mini-batch) +2023-10-05 21:48:55,309 - Epoch: [141][ 10/ 117] Loss 0.288503 Top1 85.351562 Top5 97.773438 +2023-10-05 21:48:55,460 - Epoch: [141][ 20/ 117] Loss 0.304387 Top1 84.960938 Top5 97.851562 +2023-10-05 21:48:55,611 - Epoch: [141][ 30/ 117] Loss 0.304038 Top1 84.687500 Top5 97.942708 +2023-10-05 21:48:55,762 - Epoch: [141][ 40/ 117] Loss 0.305711 Top1 84.892578 Top5 97.998047 +2023-10-05 21:48:55,913 - Epoch: [141][ 50/ 117] Loss 0.311120 Top1 84.914062 Top5 98.007812 +2023-10-05 21:48:56,064 - Epoch: [141][ 60/ 117] Loss 0.313947 Top1 84.973958 Top5 98.027344 +2023-10-05 21:48:56,219 - Epoch: [141][ 70/ 117] Loss 0.312606 Top1 85.000000 Top5 98.035714 +2023-10-05 21:48:56,371 - Epoch: [141][ 80/ 117] Loss 0.316721 Top1 84.877930 Top5 98.017578 +2023-10-05 21:48:56,522 - Epoch: [141][ 90/ 117] Loss 0.318222 Top1 84.856771 Top5 97.960069 +2023-10-05 21:48:56,674 - Epoch: [141][ 100/ 117] Loss 0.314812 Top1 84.898438 Top5 97.996094 +2023-10-05 21:48:56,830 - Epoch: [141][ 110/ 117] Loss 0.315151 Top1 84.836648 Top5 97.965199 +2023-10-05 21:48:56,915 - Epoch: [141][ 117/ 117] Loss 0.318118 Top1 84.764386 Top5 97.952777 +2023-10-05 21:48:57,041 - ==> Top1: 84.764 Top5: 97.953 Loss: 0.318 + +2023-10-05 21:48:57,042 - ==> Confusion: +[[ 932 4 6 1 10 2 0 0 6 66 2 0 1 1 5 1 3 0 0 0 10] + [ 0 1040 3 0 9 24 1 24 1 1 3 1 0 0 0 4 4 0 8 2 6] + [ 2 1 965 15 3 0 21 8 0 1 1 0 10 1 0 5 1 1 6 6 9] + [ 2 1 13 968 1 4 0 1 2 1 9 1 6 3 20 4 0 7 28 1 17] + [ 22 8 1 1 977 3 0 2 0 6 0 2 0 2 5 6 10 1 0 1 3] + [ 3 33 0 0 4 988 1 26 1 1 4 8 0 13 3 2 4 1 5 6 13] + [ 0 6 27 0 0 0 1125 9 0 0 1 2 1 0 1 8 1 0 1 5 4] + [ 1 10 14 0 3 26 6 1091 0 1 3 10 0 2 0 3 1 1 30 9 7] + [ 18 2 1 0 3 3 0 1 970 43 12 0 1 16 8 3 1 0 3 1 3] + [ 118 0 4 0 11 3 0 0 26 915 0 0 1 23 3 5 3 0 0 0 7] + [ 3 6 11 5 1 2 4 2 9 3 966 3 0 13 2 0 3 1 6 2 11] + [ 0 1 1 0 1 11 0 2 0 1 0 958 19 10 0 5 1 14 0 5 6] + [ 1 1 3 5 0 1 1 4 0 0 1 36 977 1 2 6 2 13 2 3 9] + [ 2 0 1 1 2 5 1 1 7 10 5 1 3 1065 3 2 1 0 0 1 8] + [ 14 0 4 10 11 0 0 0 28 3 1 0 3 3 992 0 3 1 13 0 15] + [ 1 4 2 0 3 0 2 0 0 0 0 6 5 1 0 1080 12 7 1 8 2] + [ 2 11 1 0 3 2 0 1 0 0 0 3 1 1 4 12 1107 0 0 3 10] + [ 0 0 0 2 0 0 2 0 0 1 0 1 20 2 0 6 0 1000 1 0 3] + [ 1 7 8 21 0 0 0 23 1 0 3 0 1 0 5 0 0 0 991 2 5] + [ 0 1 3 2 1 5 13 8 2 0 1 11 5 1 0 7 9 0 3 1072 8] + [ 116 179 150 63 108 133 53 95 78 60 173 111 313 296 109 73 181 67 137 208 5202]] + +2023-10-05 21:48:57,043 - ==> Best [Top1: 84.764 Top5: 97.953 Sparsity:0.00 Params: 148928 on epoch: 141] +2023-10-05 21:48:57,043 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:48:57,050 - + +2023-10-05 21:48:57,050 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:48:58,045 - Epoch: [142][ 10/ 1236] Overall Loss 0.251141 Objective Loss 0.251141 LR 0.000250 Time 0.099483 +2023-10-05 21:48:58,247 - Epoch: [142][ 20/ 1236] Overall Loss 0.226737 Objective Loss 0.226737 LR 0.000250 Time 0.059831 +2023-10-05 21:48:58,447 - Epoch: [142][ 30/ 1236] Overall Loss 0.228387 Objective Loss 0.228387 LR 0.000250 Time 0.046535 +2023-10-05 21:48:58,648 - Epoch: [142][ 40/ 1236] Overall Loss 0.223417 Objective Loss 0.223417 LR 0.000250 Time 0.039915 +2023-10-05 21:48:58,847 - Epoch: [142][ 50/ 1236] Overall Loss 0.225780 Objective Loss 0.225780 LR 0.000250 Time 0.035896 +2023-10-05 21:48:59,046 - Epoch: [142][ 60/ 1236] Overall Loss 0.229425 Objective Loss 0.229425 LR 0.000250 Time 0.033234 +2023-10-05 21:48:59,245 - Epoch: [142][ 70/ 1236] Overall Loss 0.229157 Objective Loss 0.229157 LR 0.000250 Time 0.031320 +2023-10-05 21:48:59,445 - Epoch: [142][ 80/ 1236] Overall Loss 0.227696 Objective Loss 0.227696 LR 0.000250 Time 0.029899 +2023-10-05 21:48:59,643 - Epoch: [142][ 90/ 1236] Overall Loss 0.226785 Objective Loss 0.226785 LR 0.000250 Time 0.028780 +2023-10-05 21:48:59,843 - Epoch: [142][ 100/ 1236] Overall Loss 0.226526 Objective Loss 0.226526 LR 0.000250 Time 0.027897 +2023-10-05 21:49:00,042 - Epoch: [142][ 110/ 1236] Overall Loss 0.226300 Objective Loss 0.226300 LR 0.000250 Time 0.027162 +2023-10-05 21:49:00,241 - Epoch: [142][ 120/ 1236] Overall Loss 0.224339 Objective Loss 0.224339 LR 0.000250 Time 0.026560 +2023-10-05 21:49:00,440 - Epoch: [142][ 130/ 1236] Overall Loss 0.222263 Objective Loss 0.222263 LR 0.000250 Time 0.026042 +2023-10-05 21:49:00,640 - Epoch: [142][ 140/ 1236] Overall Loss 0.222240 Objective Loss 0.222240 LR 0.000250 Time 0.025607 +2023-10-05 21:49:00,838 - Epoch: [142][ 150/ 1236] Overall Loss 0.222363 Objective Loss 0.222363 LR 0.000250 Time 0.025221 +2023-10-05 21:49:01,038 - Epoch: [142][ 160/ 1236] Overall Loss 0.222460 Objective Loss 0.222460 LR 0.000250 Time 0.024891 +2023-10-05 21:49:01,237 - Epoch: [142][ 170/ 1236] Overall Loss 0.221859 Objective Loss 0.221859 LR 0.000250 Time 0.024594 +2023-10-05 21:49:01,437 - Epoch: [142][ 180/ 1236] Overall Loss 0.221053 Objective Loss 0.221053 LR 0.000250 Time 0.024337 +2023-10-05 21:49:01,635 - Epoch: [142][ 190/ 1236] Overall Loss 0.220853 Objective Loss 0.220853 LR 0.000250 Time 0.024099 +2023-10-05 21:49:01,835 - Epoch: [142][ 200/ 1236] Overall Loss 0.219504 Objective Loss 0.219504 LR 0.000250 Time 0.023892 +2023-10-05 21:49:02,034 - Epoch: [142][ 210/ 1236] Overall Loss 0.219222 Objective Loss 0.219222 LR 0.000250 Time 0.023698 +2023-10-05 21:49:02,234 - Epoch: [142][ 220/ 1236] Overall Loss 0.220158 Objective Loss 0.220158 LR 0.000250 Time 0.023528 +2023-10-05 21:49:02,432 - Epoch: [142][ 230/ 1236] Overall Loss 0.220194 Objective Loss 0.220194 LR 0.000250 Time 0.023367 +2023-10-05 21:49:02,632 - Epoch: [142][ 240/ 1236] Overall Loss 0.220404 Objective Loss 0.220404 LR 0.000250 Time 0.023225 +2023-10-05 21:49:02,831 - Epoch: [142][ 250/ 1236] Overall Loss 0.220013 Objective Loss 0.220013 LR 0.000250 Time 0.023089 +2023-10-05 21:49:03,031 - Epoch: [142][ 260/ 1236] Overall Loss 0.218607 Objective Loss 0.218607 LR 0.000250 Time 0.022969 +2023-10-05 21:49:03,229 - Epoch: [142][ 270/ 1236] Overall Loss 0.218513 Objective Loss 0.218513 LR 0.000250 Time 0.022852 +2023-10-05 21:49:03,429 - Epoch: [142][ 280/ 1236] Overall Loss 0.218161 Objective Loss 0.218161 LR 0.000250 Time 0.022748 +2023-10-05 21:49:03,628 - Epoch: [142][ 290/ 1236] Overall Loss 0.218483 Objective Loss 0.218483 LR 0.000250 Time 0.022648 +2023-10-05 21:49:03,828 - Epoch: [142][ 300/ 1236] Overall Loss 0.218663 Objective Loss 0.218663 LR 0.000250 Time 0.022559 +2023-10-05 21:49:04,027 - Epoch: [142][ 310/ 1236] Overall Loss 0.218661 Objective Loss 0.218661 LR 0.000250 Time 0.022471 +2023-10-05 21:49:04,227 - Epoch: [142][ 320/ 1236] Overall Loss 0.218905 Objective Loss 0.218905 LR 0.000250 Time 0.022393 +2023-10-05 21:49:04,426 - Epoch: [142][ 330/ 1236] Overall Loss 0.218740 Objective Loss 0.218740 LR 0.000250 Time 0.022316 +2023-10-05 21:49:04,625 - Epoch: [142][ 340/ 1236] Overall Loss 0.218952 Objective Loss 0.218952 LR 0.000250 Time 0.022246 +2023-10-05 21:49:04,824 - Epoch: [142][ 350/ 1236] Overall Loss 0.219360 Objective Loss 0.219360 LR 0.000250 Time 0.022177 +2023-10-05 21:49:05,024 - Epoch: [142][ 360/ 1236] Overall Loss 0.219560 Objective Loss 0.219560 LR 0.000250 Time 0.022116 +2023-10-05 21:49:05,226 - Epoch: [142][ 370/ 1236] Overall Loss 0.219608 Objective Loss 0.219608 LR 0.000250 Time 0.022062 +2023-10-05 21:49:05,431 - Epoch: [142][ 380/ 1236] Overall Loss 0.219635 Objective Loss 0.219635 LR 0.000250 Time 0.022020 +2023-10-05 21:49:05,636 - Epoch: [142][ 390/ 1236] Overall Loss 0.220110 Objective Loss 0.220110 LR 0.000250 Time 0.021980 +2023-10-05 21:49:05,841 - Epoch: [142][ 400/ 1236] Overall Loss 0.220360 Objective Loss 0.220360 LR 0.000250 Time 0.021942 +2023-10-05 21:49:06,046 - Epoch: [142][ 410/ 1236] Overall Loss 0.220190 Objective Loss 0.220190 LR 0.000250 Time 0.021906 +2023-10-05 21:49:06,251 - Epoch: [142][ 420/ 1236] Overall Loss 0.220506 Objective Loss 0.220506 LR 0.000250 Time 0.021871 +2023-10-05 21:49:06,456 - Epoch: [142][ 430/ 1236] Overall Loss 0.220642 Objective Loss 0.220642 LR 0.000250 Time 0.021839 +2023-10-05 21:49:06,660 - Epoch: [142][ 440/ 1236] Overall Loss 0.220658 Objective Loss 0.220658 LR 0.000250 Time 0.021808 +2023-10-05 21:49:06,866 - Epoch: [142][ 450/ 1236] Overall Loss 0.220301 Objective Loss 0.220301 LR 0.000250 Time 0.021778 +2023-10-05 21:49:07,071 - Epoch: [142][ 460/ 1236] Overall Loss 0.220311 Objective Loss 0.220311 LR 0.000250 Time 0.021749 +2023-10-05 21:49:07,275 - Epoch: [142][ 470/ 1236] Overall Loss 0.219829 Objective Loss 0.219829 LR 0.000250 Time 0.021722 +2023-10-05 21:49:07,480 - Epoch: [142][ 480/ 1236] Overall Loss 0.219755 Objective Loss 0.219755 LR 0.000250 Time 0.021696 +2023-10-05 21:49:07,686 - Epoch: [142][ 490/ 1236] Overall Loss 0.219661 Objective Loss 0.219661 LR 0.000250 Time 0.021671 +2023-10-05 21:49:07,891 - Epoch: [142][ 500/ 1236] Overall Loss 0.219522 Objective Loss 0.219522 LR 0.000250 Time 0.021647 +2023-10-05 21:49:08,096 - Epoch: [142][ 510/ 1236] Overall Loss 0.219102 Objective Loss 0.219102 LR 0.000250 Time 0.021624 +2023-10-05 21:49:08,301 - Epoch: [142][ 520/ 1236] Overall Loss 0.219160 Objective Loss 0.219160 LR 0.000250 Time 0.021602 +2023-10-05 21:49:08,506 - Epoch: [142][ 530/ 1236] Overall Loss 0.219392 Objective Loss 0.219392 LR 0.000250 Time 0.021581 +2023-10-05 21:49:08,711 - Epoch: [142][ 540/ 1236] Overall Loss 0.219424 Objective Loss 0.219424 LR 0.000250 Time 0.021560 +2023-10-05 21:49:08,916 - Epoch: [142][ 550/ 1236] Overall Loss 0.219860 Objective Loss 0.219860 LR 0.000250 Time 0.021540 +2023-10-05 21:49:09,121 - Epoch: [142][ 560/ 1236] Overall Loss 0.220240 Objective Loss 0.220240 LR 0.000250 Time 0.021521 +2023-10-05 21:49:09,326 - Epoch: [142][ 570/ 1236] Overall Loss 0.220255 Objective Loss 0.220255 LR 0.000250 Time 0.021503 +2023-10-05 21:49:09,531 - Epoch: [142][ 580/ 1236] Overall Loss 0.220471 Objective Loss 0.220471 LR 0.000250 Time 0.021485 +2023-10-05 21:49:09,736 - Epoch: [142][ 590/ 1236] Overall Loss 0.221233 Objective Loss 0.221233 LR 0.000250 Time 0.021468 +2023-10-05 21:49:09,940 - Epoch: [142][ 600/ 1236] Overall Loss 0.221904 Objective Loss 0.221904 LR 0.000250 Time 0.021449 +2023-10-05 21:49:10,142 - Epoch: [142][ 610/ 1236] Overall Loss 0.221768 Objective Loss 0.221768 LR 0.000250 Time 0.021429 +2023-10-05 21:49:10,345 - Epoch: [142][ 620/ 1236] Overall Loss 0.221833 Objective Loss 0.221833 LR 0.000250 Time 0.021410 +2023-10-05 21:49:10,547 - Epoch: [142][ 630/ 1236] Overall Loss 0.221838 Objective Loss 0.221838 LR 0.000250 Time 0.021390 +2023-10-05 21:49:10,749 - Epoch: [142][ 640/ 1236] Overall Loss 0.221973 Objective Loss 0.221973 LR 0.000250 Time 0.021372 +2023-10-05 21:49:10,950 - Epoch: [142][ 650/ 1236] Overall Loss 0.221967 Objective Loss 0.221967 LR 0.000250 Time 0.021351 +2023-10-05 21:49:11,153 - Epoch: [142][ 660/ 1236] Overall Loss 0.222181 Objective Loss 0.222181 LR 0.000250 Time 0.021334 +2023-10-05 21:49:11,354 - Epoch: [142][ 670/ 1236] Overall Loss 0.222077 Objective Loss 0.222077 LR 0.000250 Time 0.021316 +2023-10-05 21:49:11,557 - Epoch: [142][ 680/ 1236] Overall Loss 0.222123 Objective Loss 0.222123 LR 0.000250 Time 0.021300 +2023-10-05 21:49:11,758 - Epoch: [142][ 690/ 1236] Overall Loss 0.222133 Objective Loss 0.222133 LR 0.000250 Time 0.021282 +2023-10-05 21:49:11,961 - Epoch: [142][ 700/ 1236] Overall Loss 0.222172 Objective Loss 0.222172 LR 0.000250 Time 0.021267 +2023-10-05 21:49:12,162 - Epoch: [142][ 710/ 1236] Overall Loss 0.222145 Objective Loss 0.222145 LR 0.000250 Time 0.021250 +2023-10-05 21:49:12,364 - Epoch: [142][ 720/ 1236] Overall Loss 0.221726 Objective Loss 0.221726 LR 0.000250 Time 0.021236 +2023-10-05 21:49:12,565 - Epoch: [142][ 730/ 1236] Overall Loss 0.221599 Objective Loss 0.221599 LR 0.000250 Time 0.021220 +2023-10-05 21:49:12,767 - Epoch: [142][ 740/ 1236] Overall Loss 0.221502 Objective Loss 0.221502 LR 0.000250 Time 0.021206 +2023-10-05 21:49:12,969 - Epoch: [142][ 750/ 1236] Overall Loss 0.221560 Objective Loss 0.221560 LR 0.000250 Time 0.021191 +2023-10-05 21:49:13,174 - Epoch: [142][ 760/ 1236] Overall Loss 0.221568 Objective Loss 0.221568 LR 0.000250 Time 0.021182 +2023-10-05 21:49:13,379 - Epoch: [142][ 770/ 1236] Overall Loss 0.221628 Objective Loss 0.221628 LR 0.000250 Time 0.021173 +2023-10-05 21:49:13,584 - Epoch: [142][ 780/ 1236] Overall Loss 0.221491 Objective Loss 0.221491 LR 0.000250 Time 0.021164 +2023-10-05 21:49:13,789 - Epoch: [142][ 790/ 1236] Overall Loss 0.221210 Objective Loss 0.221210 LR 0.000250 Time 0.021155 +2023-10-05 21:49:13,994 - Epoch: [142][ 800/ 1236] Overall Loss 0.221168 Objective Loss 0.221168 LR 0.000250 Time 0.021147 +2023-10-05 21:49:14,199 - Epoch: [142][ 810/ 1236] Overall Loss 0.221245 Objective Loss 0.221245 LR 0.000250 Time 0.021138 +2023-10-05 21:49:14,407 - Epoch: [142][ 820/ 1236] Overall Loss 0.221175 Objective Loss 0.221175 LR 0.000250 Time 0.021128 +2023-10-05 21:49:14,612 - Epoch: [142][ 830/ 1236] Overall Loss 0.220904 Objective Loss 0.220904 LR 0.000250 Time 0.021120 +2023-10-05 21:49:14,817 - Epoch: [142][ 840/ 1236] Overall Loss 0.220844 Objective Loss 0.220844 LR 0.000250 Time 0.021112 +2023-10-05 21:49:15,022 - Epoch: [142][ 850/ 1236] Overall Loss 0.221077 Objective Loss 0.221077 LR 0.000250 Time 0.021105 +2023-10-05 21:49:15,227 - Epoch: [142][ 860/ 1236] Overall Loss 0.221173 Objective Loss 0.221173 LR 0.000250 Time 0.021097 +2023-10-05 21:49:15,432 - Epoch: [142][ 870/ 1236] Overall Loss 0.221302 Objective Loss 0.221302 LR 0.000250 Time 0.021090 +2023-10-05 21:49:15,637 - Epoch: [142][ 880/ 1236] Overall Loss 0.221291 Objective Loss 0.221291 LR 0.000250 Time 0.021083 +2023-10-05 21:49:15,842 - Epoch: [142][ 890/ 1236] Overall Loss 0.221390 Objective Loss 0.221390 LR 0.000250 Time 0.021077 +2023-10-05 21:49:16,047 - Epoch: [142][ 900/ 1236] Overall Loss 0.221785 Objective Loss 0.221785 LR 0.000250 Time 0.021070 +2023-10-05 21:49:16,253 - Epoch: [142][ 910/ 1236] Overall Loss 0.221822 Objective Loss 0.221822 LR 0.000250 Time 0.021064 +2023-10-05 21:49:16,458 - Epoch: [142][ 920/ 1236] Overall Loss 0.221559 Objective Loss 0.221559 LR 0.000250 Time 0.021057 +2023-10-05 21:49:16,663 - Epoch: [142][ 930/ 1236] Overall Loss 0.221546 Objective Loss 0.221546 LR 0.000250 Time 0.021051 +2023-10-05 21:49:16,868 - Epoch: [142][ 940/ 1236] Overall Loss 0.221599 Objective Loss 0.221599 LR 0.000250 Time 0.021045 +2023-10-05 21:49:17,073 - Epoch: [142][ 950/ 1236] Overall Loss 0.221711 Objective Loss 0.221711 LR 0.000250 Time 0.021039 +2023-10-05 21:49:17,278 - Epoch: [142][ 960/ 1236] Overall Loss 0.221833 Objective Loss 0.221833 LR 0.000250 Time 0.021033 +2023-10-05 21:49:17,483 - Epoch: [142][ 970/ 1236] Overall Loss 0.221953 Objective Loss 0.221953 LR 0.000250 Time 0.021027 +2023-10-05 21:49:17,688 - Epoch: [142][ 980/ 1236] Overall Loss 0.222170 Objective Loss 0.222170 LR 0.000250 Time 0.021021 +2023-10-05 21:49:17,893 - Epoch: [142][ 990/ 1236] Overall Loss 0.222411 Objective Loss 0.222411 LR 0.000250 Time 0.021016 +2023-10-05 21:49:18,098 - Epoch: [142][ 1000/ 1236] Overall Loss 0.222611 Objective Loss 0.222611 LR 0.000250 Time 0.021010 +2023-10-05 21:49:18,303 - Epoch: [142][ 1010/ 1236] Overall Loss 0.222653 Objective Loss 0.222653 LR 0.000250 Time 0.021005 +2023-10-05 21:49:18,508 - Epoch: [142][ 1020/ 1236] Overall Loss 0.222663 Objective Loss 0.222663 LR 0.000250 Time 0.020999 +2023-10-05 21:49:18,713 - Epoch: [142][ 1030/ 1236] Overall Loss 0.222827 Objective Loss 0.222827 LR 0.000250 Time 0.020995 +2023-10-05 21:49:18,918 - Epoch: [142][ 1040/ 1236] Overall Loss 0.222771 Objective Loss 0.222771 LR 0.000250 Time 0.020989 +2023-10-05 21:49:19,123 - Epoch: [142][ 1050/ 1236] Overall Loss 0.222783 Objective Loss 0.222783 LR 0.000250 Time 0.020984 +2023-10-05 21:49:19,328 - Epoch: [142][ 1060/ 1236] Overall Loss 0.222671 Objective Loss 0.222671 LR 0.000250 Time 0.020979 +2023-10-05 21:49:19,533 - Epoch: [142][ 1070/ 1236] Overall Loss 0.222536 Objective Loss 0.222536 LR 0.000250 Time 0.020975 +2023-10-05 21:49:19,738 - Epoch: [142][ 1080/ 1236] Overall Loss 0.222939 Objective Loss 0.222939 LR 0.000250 Time 0.020970 +2023-10-05 21:49:19,943 - Epoch: [142][ 1090/ 1236] Overall Loss 0.222931 Objective Loss 0.222931 LR 0.000250 Time 0.020966 +2023-10-05 21:49:20,148 - Epoch: [142][ 1100/ 1236] Overall Loss 0.222841 Objective Loss 0.222841 LR 0.000250 Time 0.020961 +2023-10-05 21:49:20,353 - Epoch: [142][ 1110/ 1236] Overall Loss 0.222834 Objective Loss 0.222834 LR 0.000250 Time 0.020957 +2023-10-05 21:49:20,558 - Epoch: [142][ 1120/ 1236] Overall Loss 0.222734 Objective Loss 0.222734 LR 0.000250 Time 0.020952 +2023-10-05 21:49:20,763 - Epoch: [142][ 1130/ 1236] Overall Loss 0.222629 Objective Loss 0.222629 LR 0.000250 Time 0.020948 +2023-10-05 21:49:20,968 - Epoch: [142][ 1140/ 1236] Overall Loss 0.222629 Objective Loss 0.222629 LR 0.000250 Time 0.020944 +2023-10-05 21:49:21,173 - Epoch: [142][ 1150/ 1236] Overall Loss 0.222974 Objective Loss 0.222974 LR 0.000250 Time 0.020940 +2023-10-05 21:49:21,378 - Epoch: [142][ 1160/ 1236] Overall Loss 0.222982 Objective Loss 0.222982 LR 0.000250 Time 0.020935 +2023-10-05 21:49:21,583 - Epoch: [142][ 1170/ 1236] Overall Loss 0.222830 Objective Loss 0.222830 LR 0.000250 Time 0.020932 +2023-10-05 21:49:21,788 - Epoch: [142][ 1180/ 1236] Overall Loss 0.222963 Objective Loss 0.222963 LR 0.000250 Time 0.020927 +2023-10-05 21:49:21,993 - Epoch: [142][ 1190/ 1236] Overall Loss 0.223234 Objective Loss 0.223234 LR 0.000250 Time 0.020924 +2023-10-05 21:49:22,198 - Epoch: [142][ 1200/ 1236] Overall Loss 0.223049 Objective Loss 0.223049 LR 0.000250 Time 0.020920 +2023-10-05 21:49:22,403 - Epoch: [142][ 1210/ 1236] Overall Loss 0.223068 Objective Loss 0.223068 LR 0.000250 Time 0.020916 +2023-10-05 21:49:22,608 - Epoch: [142][ 1220/ 1236] Overall Loss 0.222950 Objective Loss 0.222950 LR 0.000250 Time 0.020912 +2023-10-05 21:49:22,865 - Epoch: [142][ 1230/ 1236] Overall Loss 0.222873 Objective Loss 0.222873 LR 0.000250 Time 0.020951 +2023-10-05 21:49:22,982 - Epoch: [142][ 1236/ 1236] Overall Loss 0.222943 Objective Loss 0.222943 Top1 87.576375 Top5 97.963340 LR 0.000250 Time 0.020944 +2023-10-05 21:49:23,114 - --- validate (epoch=142)----------- +2023-10-05 21:49:23,115 - 29943 samples (256 per mini-batch) +2023-10-05 21:49:23,571 - Epoch: [142][ 10/ 117] Loss 0.313621 Top1 84.648438 Top5 97.773438 +2023-10-05 21:49:23,723 - Epoch: [142][ 20/ 117] Loss 0.301484 Top1 85.039062 Top5 97.636719 +2023-10-05 21:49:23,874 - Epoch: [142][ 30/ 117] Loss 0.312258 Top1 85.000000 Top5 97.799479 +2023-10-05 21:49:24,025 - Epoch: [142][ 40/ 117] Loss 0.311922 Top1 84.843750 Top5 97.880859 +2023-10-05 21:49:24,174 - Epoch: [142][ 50/ 117] Loss 0.312807 Top1 84.625000 Top5 97.929688 +2023-10-05 21:49:24,323 - Epoch: [142][ 60/ 117] Loss 0.309835 Top1 84.654948 Top5 97.897135 +2023-10-05 21:49:24,471 - Epoch: [142][ 70/ 117] Loss 0.310035 Top1 84.614955 Top5 97.968750 +2023-10-05 21:49:24,620 - Epoch: [142][ 80/ 117] Loss 0.311300 Top1 84.599609 Top5 97.958984 +2023-10-05 21:49:24,768 - Epoch: [142][ 90/ 117] Loss 0.312316 Top1 84.552951 Top5 97.903646 +2023-10-05 21:49:24,915 - Epoch: [142][ 100/ 117] Loss 0.313626 Top1 84.523438 Top5 97.937500 +2023-10-05 21:49:25,069 - Epoch: [142][ 110/ 117] Loss 0.314173 Top1 84.570312 Top5 97.936790 +2023-10-05 21:49:25,154 - Epoch: [142][ 117/ 117] Loss 0.313046 Top1 84.624119 Top5 97.926060 +2023-10-05 21:49:25,301 - ==> Top1: 84.624 Top5: 97.926 Loss: 0.313 + +2023-10-05 21:49:25,302 - ==> Confusion: +[[ 943 1 4 1 8 4 0 1 4 62 1 1 0 2 5 1 2 0 0 0 10] + [ 1 1055 3 0 7 20 1 26 1 0 1 1 0 0 0 4 2 0 4 1 4] + [ 4 1 972 12 1 1 24 9 0 0 3 3 10 1 0 3 0 1 3 3 5] + [ 6 1 19 967 0 3 0 1 2 0 8 1 5 1 22 3 1 9 22 3 15] + [ 23 8 0 0 972 5 0 1 0 12 1 1 0 1 8 6 6 1 0 1 4] + [ 3 37 0 1 4 990 1 24 1 3 3 6 1 10 3 2 4 0 2 5 16] + [ 0 7 21 0 0 0 1134 7 0 0 2 2 1 0 1 5 0 0 1 7 3] + [ 3 12 15 0 3 29 3 1085 0 3 4 10 2 1 0 3 0 0 29 8 8] + [ 18 2 0 1 2 4 0 2 970 45 11 0 2 9 9 3 0 2 4 1 4] + [ 108 0 4 0 9 2 0 0 25 936 0 3 1 17 2 4 0 0 0 0 8] + [ 4 5 11 3 1 1 5 5 7 2 974 1 2 8 4 2 2 0 4 2 10] + [ 1 0 1 0 0 12 0 2 0 1 0 957 21 10 0 7 0 17 0 5 1] + [ 0 1 5 6 0 0 0 3 0 0 0 40 975 1 2 8 3 15 1 3 5] + [ 2 0 0 0 2 4 1 0 10 16 8 4 1 1054 4 2 0 1 0 0 10] + [ 14 3 4 7 8 1 0 0 28 1 1 2 5 3 1000 0 1 0 8 0 15] + [ 0 4 2 0 2 0 0 0 0 0 0 6 5 2 1 1075 12 12 0 10 3] + [ 1 14 3 0 8 3 0 2 2 0 0 1 0 0 3 12 1099 0 0 3 10] + [ 0 0 0 1 0 0 3 0 0 1 0 1 16 0 0 5 0 1007 1 1 2] + [ 1 6 6 22 2 1 0 29 1 0 3 0 1 0 7 0 1 0 979 1 8] + [ 0 1 4 3 3 5 14 6 1 0 0 16 2 0 0 3 10 1 2 1073 8] + [ 154 201 156 65 100 145 50 113 104 81 179 113 333 273 125 56 129 80 138 188 5122]] + +2023-10-05 21:49:25,303 - ==> Best [Top1: 84.764 Top5: 97.953 Sparsity:0.00 Params: 148928 on epoch: 141] +2023-10-05 21:49:25,303 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:49:25,309 - + +2023-10-05 21:49:25,309 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:49:26,304 - Epoch: [143][ 10/ 1236] Overall Loss 0.204317 Objective Loss 0.204317 LR 0.000250 Time 0.099461 +2023-10-05 21:49:26,508 - Epoch: [143][ 20/ 1236] Overall Loss 0.194012 Objective Loss 0.194012 LR 0.000250 Time 0.059867 +2023-10-05 21:49:26,711 - Epoch: [143][ 30/ 1236] Overall Loss 0.197948 Objective Loss 0.197948 LR 0.000250 Time 0.046688 +2023-10-05 21:49:26,914 - Epoch: [143][ 40/ 1236] Overall Loss 0.203076 Objective Loss 0.203076 LR 0.000250 Time 0.040085 +2023-10-05 21:49:27,118 - Epoch: [143][ 50/ 1236] Overall Loss 0.204369 Objective Loss 0.204369 LR 0.000250 Time 0.036128 +2023-10-05 21:49:27,321 - Epoch: [143][ 60/ 1236] Overall Loss 0.204526 Objective Loss 0.204526 LR 0.000250 Time 0.033490 +2023-10-05 21:49:27,525 - Epoch: [143][ 70/ 1236] Overall Loss 0.207153 Objective Loss 0.207153 LR 0.000250 Time 0.031617 +2023-10-05 21:49:27,728 - Epoch: [143][ 80/ 1236] Overall Loss 0.206063 Objective Loss 0.206063 LR 0.000250 Time 0.030201 +2023-10-05 21:49:27,933 - Epoch: [143][ 90/ 1236] Overall Loss 0.207256 Objective Loss 0.207256 LR 0.000250 Time 0.029111 +2023-10-05 21:49:28,136 - Epoch: [143][ 100/ 1236] Overall Loss 0.209111 Objective Loss 0.209111 LR 0.000250 Time 0.028229 +2023-10-05 21:49:28,340 - Epoch: [143][ 110/ 1236] Overall Loss 0.211148 Objective Loss 0.211148 LR 0.000250 Time 0.027513 +2023-10-05 21:49:28,541 - Epoch: [143][ 120/ 1236] Overall Loss 0.211038 Objective Loss 0.211038 LR 0.000250 Time 0.026899 +2023-10-05 21:49:28,746 - Epoch: [143][ 130/ 1236] Overall Loss 0.212202 Objective Loss 0.212202 LR 0.000250 Time 0.026397 +2023-10-05 21:49:28,949 - Epoch: [143][ 140/ 1236] Overall Loss 0.214021 Objective Loss 0.214021 LR 0.000250 Time 0.025961 +2023-10-05 21:49:29,153 - Epoch: [143][ 150/ 1236] Overall Loss 0.214014 Objective Loss 0.214014 LR 0.000250 Time 0.025588 +2023-10-05 21:49:29,358 - Epoch: [143][ 160/ 1236] Overall Loss 0.213426 Objective Loss 0.213426 LR 0.000250 Time 0.025268 +2023-10-05 21:49:29,564 - Epoch: [143][ 170/ 1236] Overall Loss 0.214682 Objective Loss 0.214682 LR 0.000250 Time 0.024990 +2023-10-05 21:49:29,769 - Epoch: [143][ 180/ 1236] Overall Loss 0.214628 Objective Loss 0.214628 LR 0.000250 Time 0.024742 +2023-10-05 21:49:29,976 - Epoch: [143][ 190/ 1236] Overall Loss 0.214351 Objective Loss 0.214351 LR 0.000250 Time 0.024523 +2023-10-05 21:49:30,181 - Epoch: [143][ 200/ 1236] Overall Loss 0.213940 Objective Loss 0.213940 LR 0.000250 Time 0.024323 +2023-10-05 21:49:30,387 - Epoch: [143][ 210/ 1236] Overall Loss 0.213946 Objective Loss 0.213946 LR 0.000250 Time 0.024146 +2023-10-05 21:49:30,593 - Epoch: [143][ 220/ 1236] Overall Loss 0.214323 Objective Loss 0.214323 LR 0.000250 Time 0.023981 +2023-10-05 21:49:30,799 - Epoch: [143][ 230/ 1236] Overall Loss 0.214662 Objective Loss 0.214662 LR 0.000250 Time 0.023831 +2023-10-05 21:49:31,004 - Epoch: [143][ 240/ 1236] Overall Loss 0.214631 Objective Loss 0.214631 LR 0.000250 Time 0.023693 +2023-10-05 21:49:31,210 - Epoch: [143][ 250/ 1236] Overall Loss 0.214916 Objective Loss 0.214916 LR 0.000250 Time 0.023567 +2023-10-05 21:49:31,416 - Epoch: [143][ 260/ 1236] Overall Loss 0.214658 Objective Loss 0.214658 LR 0.000250 Time 0.023451 +2023-10-05 21:49:31,622 - Epoch: [143][ 270/ 1236] Overall Loss 0.215228 Objective Loss 0.215228 LR 0.000250 Time 0.023344 +2023-10-05 21:49:31,827 - Epoch: [143][ 280/ 1236] Overall Loss 0.214959 Objective Loss 0.214959 LR 0.000250 Time 0.023242 +2023-10-05 21:49:32,034 - Epoch: [143][ 290/ 1236] Overall Loss 0.213680 Objective Loss 0.213680 LR 0.000250 Time 0.023152 +2023-10-05 21:49:32,239 - Epoch: [143][ 300/ 1236] Overall Loss 0.213905 Objective Loss 0.213905 LR 0.000250 Time 0.023063 +2023-10-05 21:49:32,445 - Epoch: [143][ 310/ 1236] Overall Loss 0.213954 Objective Loss 0.213954 LR 0.000250 Time 0.022983 +2023-10-05 21:49:32,651 - Epoch: [143][ 320/ 1236] Overall Loss 0.214632 Objective Loss 0.214632 LR 0.000250 Time 0.022906 +2023-10-05 21:49:32,857 - Epoch: [143][ 330/ 1236] Overall Loss 0.214909 Objective Loss 0.214909 LR 0.000250 Time 0.022837 +2023-10-05 21:49:33,063 - Epoch: [143][ 340/ 1236] Overall Loss 0.214416 Objective Loss 0.214416 LR 0.000250 Time 0.022769 +2023-10-05 21:49:33,269 - Epoch: [143][ 350/ 1236] Overall Loss 0.214865 Objective Loss 0.214865 LR 0.000250 Time 0.022707 +2023-10-05 21:49:33,475 - Epoch: [143][ 360/ 1236] Overall Loss 0.214470 Objective Loss 0.214470 LR 0.000250 Time 0.022646 +2023-10-05 21:49:33,680 - Epoch: [143][ 370/ 1236] Overall Loss 0.214431 Objective Loss 0.214431 LR 0.000250 Time 0.022587 +2023-10-05 21:49:33,884 - Epoch: [143][ 380/ 1236] Overall Loss 0.214942 Objective Loss 0.214942 LR 0.000250 Time 0.022530 +2023-10-05 21:49:34,088 - Epoch: [143][ 390/ 1236] Overall Loss 0.215088 Objective Loss 0.215088 LR 0.000250 Time 0.022473 +2023-10-05 21:49:34,291 - Epoch: [143][ 400/ 1236] Overall Loss 0.215627 Objective Loss 0.215627 LR 0.000250 Time 0.022418 +2023-10-05 21:49:34,494 - Epoch: [143][ 410/ 1236] Overall Loss 0.215655 Objective Loss 0.215655 LR 0.000250 Time 0.022365 +2023-10-05 21:49:34,697 - Epoch: [143][ 420/ 1236] Overall Loss 0.215744 Objective Loss 0.215744 LR 0.000250 Time 0.022315 +2023-10-05 21:49:34,898 - Epoch: [143][ 430/ 1236] Overall Loss 0.215527 Objective Loss 0.215527 LR 0.000250 Time 0.022264 +2023-10-05 21:49:35,101 - Epoch: [143][ 440/ 1236] Overall Loss 0.215247 Objective Loss 0.215247 LR 0.000250 Time 0.022217 +2023-10-05 21:49:35,302 - Epoch: [143][ 450/ 1236] Overall Loss 0.215608 Objective Loss 0.215608 LR 0.000250 Time 0.022171 +2023-10-05 21:49:35,505 - Epoch: [143][ 460/ 1236] Overall Loss 0.215916 Objective Loss 0.215916 LR 0.000250 Time 0.022129 +2023-10-05 21:49:35,707 - Epoch: [143][ 470/ 1236] Overall Loss 0.216752 Objective Loss 0.216752 LR 0.000250 Time 0.022084 +2023-10-05 21:49:35,910 - Epoch: [143][ 480/ 1236] Overall Loss 0.217131 Objective Loss 0.217131 LR 0.000250 Time 0.022045 +2023-10-05 21:49:36,111 - Epoch: [143][ 490/ 1236] Overall Loss 0.217186 Objective Loss 0.217186 LR 0.000250 Time 0.022006 +2023-10-05 21:49:36,314 - Epoch: [143][ 500/ 1236] Overall Loss 0.217620 Objective Loss 0.217620 LR 0.000250 Time 0.021970 +2023-10-05 21:49:36,515 - Epoch: [143][ 510/ 1236] Overall Loss 0.217842 Objective Loss 0.217842 LR 0.000250 Time 0.021934 +2023-10-05 21:49:36,718 - Epoch: [143][ 520/ 1236] Overall Loss 0.218186 Objective Loss 0.218186 LR 0.000250 Time 0.021901 +2023-10-05 21:49:36,919 - Epoch: [143][ 530/ 1236] Overall Loss 0.218321 Objective Loss 0.218321 LR 0.000250 Time 0.021867 +2023-10-05 21:49:37,121 - Epoch: [143][ 540/ 1236] Overall Loss 0.218720 Objective Loss 0.218720 LR 0.000250 Time 0.021836 +2023-10-05 21:49:37,323 - Epoch: [143][ 550/ 1236] Overall Loss 0.218526 Objective Loss 0.218526 LR 0.000250 Time 0.021805 +2023-10-05 21:49:37,526 - Epoch: [143][ 560/ 1236] Overall Loss 0.218638 Objective Loss 0.218638 LR 0.000250 Time 0.021777 +2023-10-05 21:49:37,727 - Epoch: [143][ 570/ 1236] Overall Loss 0.218255 Objective Loss 0.218255 LR 0.000250 Time 0.021747 +2023-10-05 21:49:37,930 - Epoch: [143][ 580/ 1236] Overall Loss 0.218201 Objective Loss 0.218201 LR 0.000250 Time 0.021722 +2023-10-05 21:49:38,132 - Epoch: [143][ 590/ 1236] Overall Loss 0.218177 Objective Loss 0.218177 LR 0.000250 Time 0.021695 +2023-10-05 21:49:38,334 - Epoch: [143][ 600/ 1236] Overall Loss 0.218200 Objective Loss 0.218200 LR 0.000250 Time 0.021670 +2023-10-05 21:49:38,537 - Epoch: [143][ 610/ 1236] Overall Loss 0.218493 Objective Loss 0.218493 LR 0.000250 Time 0.021647 +2023-10-05 21:49:38,739 - Epoch: [143][ 620/ 1236] Overall Loss 0.218334 Objective Loss 0.218334 LR 0.000250 Time 0.021623 +2023-10-05 21:49:38,941 - Epoch: [143][ 630/ 1236] Overall Loss 0.218112 Objective Loss 0.218112 LR 0.000250 Time 0.021600 +2023-10-05 21:49:39,144 - Epoch: [143][ 640/ 1236] Overall Loss 0.218105 Objective Loss 0.218105 LR 0.000250 Time 0.021579 +2023-10-05 21:49:39,346 - Epoch: [143][ 650/ 1236] Overall Loss 0.218172 Objective Loss 0.218172 LR 0.000250 Time 0.021556 +2023-10-05 21:49:39,549 - Epoch: [143][ 660/ 1236] Overall Loss 0.218225 Objective Loss 0.218225 LR 0.000250 Time 0.021536 +2023-10-05 21:49:39,751 - Epoch: [143][ 670/ 1236] Overall Loss 0.218458 Objective Loss 0.218458 LR 0.000250 Time 0.021513 +2023-10-05 21:49:39,954 - Epoch: [143][ 680/ 1236] Overall Loss 0.218530 Objective Loss 0.218530 LR 0.000250 Time 0.021495 +2023-10-05 21:49:40,156 - Epoch: [143][ 690/ 1236] Overall Loss 0.218179 Objective Loss 0.218179 LR 0.000250 Time 0.021476 +2023-10-05 21:49:40,359 - Epoch: [143][ 700/ 1236] Overall Loss 0.218192 Objective Loss 0.218192 LR 0.000250 Time 0.021458 +2023-10-05 21:49:40,561 - Epoch: [143][ 710/ 1236] Overall Loss 0.218287 Objective Loss 0.218287 LR 0.000250 Time 0.021439 +2023-10-05 21:49:40,763 - Epoch: [143][ 720/ 1236] Overall Loss 0.218284 Objective Loss 0.218284 LR 0.000250 Time 0.021422 +2023-10-05 21:49:40,965 - Epoch: [143][ 730/ 1236] Overall Loss 0.218290 Objective Loss 0.218290 LR 0.000250 Time 0.021405 +2023-10-05 21:49:41,168 - Epoch: [143][ 740/ 1236] Overall Loss 0.218389 Objective Loss 0.218389 LR 0.000250 Time 0.021390 +2023-10-05 21:49:41,370 - Epoch: [143][ 750/ 1236] Overall Loss 0.218612 Objective Loss 0.218612 LR 0.000250 Time 0.021374 +2023-10-05 21:49:41,573 - Epoch: [143][ 760/ 1236] Overall Loss 0.218672 Objective Loss 0.218672 LR 0.000250 Time 0.021359 +2023-10-05 21:49:41,775 - Epoch: [143][ 770/ 1236] Overall Loss 0.218491 Objective Loss 0.218491 LR 0.000250 Time 0.021343 +2023-10-05 21:49:41,978 - Epoch: [143][ 780/ 1236] Overall Loss 0.218568 Objective Loss 0.218568 LR 0.000250 Time 0.021329 +2023-10-05 21:49:42,180 - Epoch: [143][ 790/ 1236] Overall Loss 0.218651 Objective Loss 0.218651 LR 0.000250 Time 0.021315 +2023-10-05 21:49:42,382 - Epoch: [143][ 800/ 1236] Overall Loss 0.218669 Objective Loss 0.218669 LR 0.000250 Time 0.021301 +2023-10-05 21:49:42,585 - Epoch: [143][ 810/ 1236] Overall Loss 0.218602 Objective Loss 0.218602 LR 0.000250 Time 0.021288 +2023-10-05 21:49:42,787 - Epoch: [143][ 820/ 1236] Overall Loss 0.218668 Objective Loss 0.218668 LR 0.000250 Time 0.021274 +2023-10-05 21:49:42,989 - Epoch: [143][ 830/ 1236] Overall Loss 0.219053 Objective Loss 0.219053 LR 0.000250 Time 0.021261 +2023-10-05 21:49:43,192 - Epoch: [143][ 840/ 1236] Overall Loss 0.219049 Objective Loss 0.219049 LR 0.000250 Time 0.021249 +2023-10-05 21:49:43,394 - Epoch: [143][ 850/ 1236] Overall Loss 0.219186 Objective Loss 0.219186 LR 0.000250 Time 0.021235 +2023-10-05 21:49:43,597 - Epoch: [143][ 860/ 1236] Overall Loss 0.219290 Objective Loss 0.219290 LR 0.000250 Time 0.021223 +2023-10-05 21:49:43,799 - Epoch: [143][ 870/ 1236] Overall Loss 0.219337 Objective Loss 0.219337 LR 0.000250 Time 0.021212 +2023-10-05 21:49:44,002 - Epoch: [143][ 880/ 1236] Overall Loss 0.219157 Objective Loss 0.219157 LR 0.000250 Time 0.021200 +2023-10-05 21:49:44,204 - Epoch: [143][ 890/ 1236] Overall Loss 0.219176 Objective Loss 0.219176 LR 0.000250 Time 0.021187 +2023-10-05 21:49:44,407 - Epoch: [143][ 900/ 1236] Overall Loss 0.218957 Objective Loss 0.218957 LR 0.000250 Time 0.021177 +2023-10-05 21:49:44,609 - Epoch: [143][ 910/ 1236] Overall Loss 0.219045 Objective Loss 0.219045 LR 0.000250 Time 0.021166 +2023-10-05 21:49:44,812 - Epoch: [143][ 920/ 1236] Overall Loss 0.219296 Objective Loss 0.219296 LR 0.000250 Time 0.021156 +2023-10-05 21:49:45,014 - Epoch: [143][ 930/ 1236] Overall Loss 0.219193 Objective Loss 0.219193 LR 0.000250 Time 0.021146 +2023-10-05 21:49:45,217 - Epoch: [143][ 940/ 1236] Overall Loss 0.219295 Objective Loss 0.219295 LR 0.000250 Time 0.021136 +2023-10-05 21:49:45,419 - Epoch: [143][ 950/ 1236] Overall Loss 0.219420 Objective Loss 0.219420 LR 0.000250 Time 0.021126 +2023-10-05 21:49:45,622 - Epoch: [143][ 960/ 1236] Overall Loss 0.219374 Objective Loss 0.219374 LR 0.000250 Time 0.021117 +2023-10-05 21:49:45,824 - Epoch: [143][ 970/ 1236] Overall Loss 0.219400 Objective Loss 0.219400 LR 0.000250 Time 0.021107 +2023-10-05 21:49:46,027 - Epoch: [143][ 980/ 1236] Overall Loss 0.219854 Objective Loss 0.219854 LR 0.000250 Time 0.021099 +2023-10-05 21:49:46,229 - Epoch: [143][ 990/ 1236] Overall Loss 0.219496 Objective Loss 0.219496 LR 0.000250 Time 0.021088 +2023-10-05 21:49:46,432 - Epoch: [143][ 1000/ 1236] Overall Loss 0.219597 Objective Loss 0.219597 LR 0.000250 Time 0.021079 +2023-10-05 21:49:46,634 - Epoch: [143][ 1010/ 1236] Overall Loss 0.219542 Objective Loss 0.219542 LR 0.000250 Time 0.021070 +2023-10-05 21:49:46,836 - Epoch: [143][ 1020/ 1236] Overall Loss 0.219541 Objective Loss 0.219541 LR 0.000250 Time 0.021062 +2023-10-05 21:49:47,039 - Epoch: [143][ 1030/ 1236] Overall Loss 0.219498 Objective Loss 0.219498 LR 0.000250 Time 0.021054 +2023-10-05 21:49:47,241 - Epoch: [143][ 1040/ 1236] Overall Loss 0.219401 Objective Loss 0.219401 LR 0.000250 Time 0.021046 +2023-10-05 21:49:47,444 - Epoch: [143][ 1050/ 1236] Overall Loss 0.219359 Objective Loss 0.219359 LR 0.000250 Time 0.021038 +2023-10-05 21:49:47,647 - Epoch: [143][ 1060/ 1236] Overall Loss 0.219530 Objective Loss 0.219530 LR 0.000250 Time 0.021030 +2023-10-05 21:49:47,849 - Epoch: [143][ 1070/ 1236] Overall Loss 0.219593 Objective Loss 0.219593 LR 0.000250 Time 0.021023 +2023-10-05 21:49:48,051 - Epoch: [143][ 1080/ 1236] Overall Loss 0.219778 Objective Loss 0.219778 LR 0.000250 Time 0.021015 +2023-10-05 21:49:48,253 - Epoch: [143][ 1090/ 1236] Overall Loss 0.219988 Objective Loss 0.219988 LR 0.000250 Time 0.021007 +2023-10-05 21:49:48,456 - Epoch: [143][ 1100/ 1236] Overall Loss 0.219982 Objective Loss 0.219982 LR 0.000250 Time 0.021000 +2023-10-05 21:49:48,658 - Epoch: [143][ 1110/ 1236] Overall Loss 0.220204 Objective Loss 0.220204 LR 0.000250 Time 0.020992 +2023-10-05 21:49:48,861 - Epoch: [143][ 1120/ 1236] Overall Loss 0.220307 Objective Loss 0.220307 LR 0.000250 Time 0.020986 +2023-10-05 21:49:49,062 - Epoch: [143][ 1130/ 1236] Overall Loss 0.220272 Objective Loss 0.220272 LR 0.000250 Time 0.020978 +2023-10-05 21:49:49,265 - Epoch: [143][ 1140/ 1236] Overall Loss 0.220334 Objective Loss 0.220334 LR 0.000250 Time 0.020972 +2023-10-05 21:49:49,467 - Epoch: [143][ 1150/ 1236] Overall Loss 0.220422 Objective Loss 0.220422 LR 0.000250 Time 0.020965 +2023-10-05 21:49:49,670 - Epoch: [143][ 1160/ 1236] Overall Loss 0.220696 Objective Loss 0.220696 LR 0.000250 Time 0.020958 +2023-10-05 21:49:49,873 - Epoch: [143][ 1170/ 1236] Overall Loss 0.220977 Objective Loss 0.220977 LR 0.000250 Time 0.020953 +2023-10-05 21:49:50,074 - Epoch: [143][ 1180/ 1236] Overall Loss 0.220950 Objective Loss 0.220950 LR 0.000250 Time 0.020945 +2023-10-05 21:49:50,276 - Epoch: [143][ 1190/ 1236] Overall Loss 0.221173 Objective Loss 0.221173 LR 0.000250 Time 0.020939 +2023-10-05 21:49:50,479 - Epoch: [143][ 1200/ 1236] Overall Loss 0.221216 Objective Loss 0.221216 LR 0.000250 Time 0.020933 +2023-10-05 21:49:50,682 - Epoch: [143][ 1210/ 1236] Overall Loss 0.221427 Objective Loss 0.221427 LR 0.000250 Time 0.020928 +2023-10-05 21:49:50,884 - Epoch: [143][ 1220/ 1236] Overall Loss 0.221350 Objective Loss 0.221350 LR 0.000250 Time 0.020922 +2023-10-05 21:49:51,138 - Epoch: [143][ 1230/ 1236] Overall Loss 0.221514 Objective Loss 0.221514 LR 0.000250 Time 0.020958 +2023-10-05 21:49:51,256 - Epoch: [143][ 1236/ 1236] Overall Loss 0.221503 Objective Loss 0.221503 Top1 89.205703 Top5 98.370672 LR 0.000250 Time 0.020951 +2023-10-05 21:49:51,396 - --- validate (epoch=143)----------- +2023-10-05 21:49:51,396 - 29943 samples (256 per mini-batch) +2023-10-05 21:49:51,850 - Epoch: [143][ 10/ 117] Loss 0.321424 Top1 84.023438 Top5 98.398438 +2023-10-05 21:49:52,001 - Epoch: [143][ 20/ 117] Loss 0.321621 Top1 84.140625 Top5 98.125000 +2023-10-05 21:49:52,151 - Epoch: [143][ 30/ 117] Loss 0.309235 Top1 84.505208 Top5 98.111979 +2023-10-05 21:49:52,300 - Epoch: [143][ 40/ 117] Loss 0.307189 Top1 84.941406 Top5 98.125000 +2023-10-05 21:49:52,448 - Epoch: [143][ 50/ 117] Loss 0.301891 Top1 84.914062 Top5 98.062500 +2023-10-05 21:49:52,596 - Epoch: [143][ 60/ 117] Loss 0.302999 Top1 84.980469 Top5 98.079427 +2023-10-05 21:49:52,744 - Epoch: [143][ 70/ 117] Loss 0.309040 Top1 84.771205 Top5 98.030134 +2023-10-05 21:49:52,898 - Epoch: [143][ 80/ 117] Loss 0.309080 Top1 84.824219 Top5 97.998047 +2023-10-05 21:49:53,053 - Epoch: [143][ 90/ 117] Loss 0.307471 Top1 84.913194 Top5 97.990451 +2023-10-05 21:49:53,207 - Epoch: [143][ 100/ 117] Loss 0.311371 Top1 84.664062 Top5 97.980469 +2023-10-05 21:49:53,372 - Epoch: [143][ 110/ 117] Loss 0.310266 Top1 84.715909 Top5 97.979403 +2023-10-05 21:49:53,458 - Epoch: [143][ 117/ 117] Loss 0.309844 Top1 84.714290 Top5 97.986174 +2023-10-05 21:49:53,584 - ==> Top1: 84.714 Top5: 97.986 Loss: 0.310 + +2023-10-05 21:49:53,585 - ==> Confusion: +[[ 915 3 4 2 10 1 0 0 6 83 1 1 1 2 3 0 2 1 0 0 15] + [ 0 1061 3 1 8 20 1 18 0 0 0 2 0 0 1 3 1 0 4 2 6] + [ 4 3 968 16 1 0 20 6 0 1 2 1 9 3 0 3 0 1 5 3 10] + [ 1 1 11 991 0 5 0 0 2 2 6 1 9 2 21 5 0 5 12 1 14] + [ 22 6 1 0 976 3 0 0 0 9 2 2 0 1 8 5 6 2 0 2 5] + [ 4 35 1 0 3 988 0 21 3 4 5 3 1 16 6 2 2 0 2 5 15] + [ 0 6 33 0 0 0 1116 5 0 0 1 3 2 1 1 9 0 1 1 6 6] + [ 2 20 16 0 4 26 5 1067 2 3 4 11 2 3 0 3 0 1 35 6 8] + [ 17 3 0 0 2 3 0 0 978 40 10 0 3 12 15 3 0 0 1 0 2] + [ 87 1 3 0 5 3 0 0 26 954 0 1 2 21 2 6 0 0 0 0 8] + [ 2 5 9 5 1 1 2 3 7 2 975 1 1 13 6 2 1 1 6 2 8] + [ 1 1 1 0 0 14 0 0 0 1 1 949 26 7 0 4 0 16 1 7 6] + [ 1 0 5 7 0 2 0 1 0 0 1 36 986 0 1 7 2 11 1 2 5] + [ 1 0 0 0 4 3 0 0 10 16 5 5 1 1063 2 1 0 1 0 0 7] + [ 13 2 3 13 7 0 0 0 29 0 2 2 2 4 1002 0 1 1 8 0 12] + [ 0 4 1 1 4 0 1 0 0 0 0 4 7 3 1 1068 10 12 1 11 6] + [ 1 17 1 0 3 3 0 0 1 0 0 3 0 2 3 10 1098 0 0 4 15] + [ 0 0 0 1 0 0 1 0 0 1 0 0 14 1 0 6 0 1011 0 1 2] + [ 1 5 8 20 1 0 0 16 1 0 1 0 2 2 10 0 0 0 994 1 6] + [ 0 2 2 3 1 4 13 4 2 1 0 17 4 1 1 6 9 1 2 1072 7] + [ 114 171 151 85 104 143 30 83 110 76 188 93 372 299 162 57 100 76 158 199 5134]] + +2023-10-05 21:49:53,586 - ==> Best [Top1: 84.764 Top5: 97.953 Sparsity:0.00 Params: 148928 on epoch: 141] +2023-10-05 21:49:53,586 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:49:53,592 - + +2023-10-05 21:49:53,592 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:49:54,569 - Epoch: [144][ 10/ 1236] Overall Loss 0.230428 Objective Loss 0.230428 LR 0.000250 Time 0.097625 +2023-10-05 21:49:54,769 - Epoch: [144][ 20/ 1236] Overall Loss 0.227505 Objective Loss 0.227505 LR 0.000250 Time 0.058830 +2023-10-05 21:49:54,970 - Epoch: [144][ 30/ 1236] Overall Loss 0.232028 Objective Loss 0.232028 LR 0.000250 Time 0.045894 +2023-10-05 21:49:55,169 - Epoch: [144][ 40/ 1236] Overall Loss 0.227018 Objective Loss 0.227018 LR 0.000250 Time 0.039397 +2023-10-05 21:49:55,370 - Epoch: [144][ 50/ 1236] Overall Loss 0.228346 Objective Loss 0.228346 LR 0.000250 Time 0.035514 +2023-10-05 21:49:55,569 - Epoch: [144][ 60/ 1236] Overall Loss 0.224244 Objective Loss 0.224244 LR 0.000250 Time 0.032920 +2023-10-05 21:49:55,770 - Epoch: [144][ 70/ 1236] Overall Loss 0.219803 Objective Loss 0.219803 LR 0.000250 Time 0.031073 +2023-10-05 21:49:55,971 - Epoch: [144][ 80/ 1236] Overall Loss 0.219034 Objective Loss 0.219034 LR 0.000250 Time 0.029699 +2023-10-05 21:49:56,170 - Epoch: [144][ 90/ 1236] Overall Loss 0.218516 Objective Loss 0.218516 LR 0.000250 Time 0.028609 +2023-10-05 21:49:56,369 - Epoch: [144][ 100/ 1236] Overall Loss 0.216694 Objective Loss 0.216694 LR 0.000250 Time 0.027735 +2023-10-05 21:49:56,569 - Epoch: [144][ 110/ 1236] Overall Loss 0.216157 Objective Loss 0.216157 LR 0.000250 Time 0.027032 +2023-10-05 21:49:56,770 - Epoch: [144][ 120/ 1236] Overall Loss 0.215887 Objective Loss 0.215887 LR 0.000250 Time 0.026448 +2023-10-05 21:49:56,971 - Epoch: [144][ 130/ 1236] Overall Loss 0.216680 Objective Loss 0.216680 LR 0.000250 Time 0.025953 +2023-10-05 21:49:57,171 - Epoch: [144][ 140/ 1236] Overall Loss 0.217961 Objective Loss 0.217961 LR 0.000250 Time 0.025530 +2023-10-05 21:49:57,372 - Epoch: [144][ 150/ 1236] Overall Loss 0.217188 Objective Loss 0.217188 LR 0.000250 Time 0.025163 +2023-10-05 21:49:57,572 - Epoch: [144][ 160/ 1236] Overall Loss 0.218109 Objective Loss 0.218109 LR 0.000250 Time 0.024839 +2023-10-05 21:49:57,772 - Epoch: [144][ 170/ 1236] Overall Loss 0.218370 Objective Loss 0.218370 LR 0.000250 Time 0.024555 +2023-10-05 21:49:57,974 - Epoch: [144][ 180/ 1236] Overall Loss 0.217849 Objective Loss 0.217849 LR 0.000250 Time 0.024307 +2023-10-05 21:49:58,174 - Epoch: [144][ 190/ 1236] Overall Loss 0.218198 Objective Loss 0.218198 LR 0.000250 Time 0.024082 +2023-10-05 21:49:58,375 - Epoch: [144][ 200/ 1236] Overall Loss 0.217824 Objective Loss 0.217824 LR 0.000250 Time 0.023879 +2023-10-05 21:49:58,575 - Epoch: [144][ 210/ 1236] Overall Loss 0.217168 Objective Loss 0.217168 LR 0.000250 Time 0.023695 +2023-10-05 21:49:58,777 - Epoch: [144][ 220/ 1236] Overall Loss 0.217355 Objective Loss 0.217355 LR 0.000250 Time 0.023530 +2023-10-05 21:49:58,977 - Epoch: [144][ 230/ 1236] Overall Loss 0.217988 Objective Loss 0.217988 LR 0.000250 Time 0.023379 +2023-10-05 21:49:59,178 - Epoch: [144][ 240/ 1236] Overall Loss 0.217956 Objective Loss 0.217956 LR 0.000250 Time 0.023239 +2023-10-05 21:49:59,379 - Epoch: [144][ 250/ 1236] Overall Loss 0.218073 Objective Loss 0.218073 LR 0.000250 Time 0.023111 +2023-10-05 21:49:59,579 - Epoch: [144][ 260/ 1236] Overall Loss 0.217130 Objective Loss 0.217130 LR 0.000250 Time 0.022991 +2023-10-05 21:49:59,779 - Epoch: [144][ 270/ 1236] Overall Loss 0.216017 Objective Loss 0.216017 LR 0.000250 Time 0.022880 +2023-10-05 21:49:59,980 - Epoch: [144][ 280/ 1236] Overall Loss 0.215588 Objective Loss 0.215588 LR 0.000250 Time 0.022780 +2023-10-05 21:50:00,181 - Epoch: [144][ 290/ 1236] Overall Loss 0.215785 Objective Loss 0.215785 LR 0.000250 Time 0.022686 +2023-10-05 21:50:00,382 - Epoch: [144][ 300/ 1236] Overall Loss 0.215731 Objective Loss 0.215731 LR 0.000250 Time 0.022596 +2023-10-05 21:50:00,582 - Epoch: [144][ 310/ 1236] Overall Loss 0.216454 Objective Loss 0.216454 LR 0.000250 Time 0.022514 +2023-10-05 21:50:00,783 - Epoch: [144][ 320/ 1236] Overall Loss 0.216351 Objective Loss 0.216351 LR 0.000250 Time 0.022436 +2023-10-05 21:50:00,984 - Epoch: [144][ 330/ 1236] Overall Loss 0.216884 Objective Loss 0.216884 LR 0.000250 Time 0.022364 +2023-10-05 21:50:01,184 - Epoch: [144][ 340/ 1236] Overall Loss 0.216699 Objective Loss 0.216699 LR 0.000250 Time 0.022295 +2023-10-05 21:50:01,385 - Epoch: [144][ 350/ 1236] Overall Loss 0.216437 Objective Loss 0.216437 LR 0.000250 Time 0.022229 +2023-10-05 21:50:01,585 - Epoch: [144][ 360/ 1236] Overall Loss 0.216647 Objective Loss 0.216647 LR 0.000250 Time 0.022168 +2023-10-05 21:50:01,786 - Epoch: [144][ 370/ 1236] Overall Loss 0.216612 Objective Loss 0.216612 LR 0.000250 Time 0.022111 +2023-10-05 21:50:01,986 - Epoch: [144][ 380/ 1236] Overall Loss 0.216413 Objective Loss 0.216413 LR 0.000250 Time 0.022055 +2023-10-05 21:50:02,185 - Epoch: [144][ 390/ 1236] Overall Loss 0.217064 Objective Loss 0.217064 LR 0.000250 Time 0.021998 +2023-10-05 21:50:02,386 - Epoch: [144][ 400/ 1236] Overall Loss 0.216900 Objective Loss 0.216900 LR 0.000250 Time 0.021949 +2023-10-05 21:50:02,588 - Epoch: [144][ 410/ 1236] Overall Loss 0.216729 Objective Loss 0.216729 LR 0.000250 Time 0.021905 +2023-10-05 21:50:02,789 - Epoch: [144][ 420/ 1236] Overall Loss 0.217010 Objective Loss 0.217010 LR 0.000250 Time 0.021861 +2023-10-05 21:50:02,990 - Epoch: [144][ 430/ 1236] Overall Loss 0.217062 Objective Loss 0.217062 LR 0.000250 Time 0.021821 +2023-10-05 21:50:03,191 - Epoch: [144][ 440/ 1236] Overall Loss 0.218369 Objective Loss 0.218369 LR 0.000250 Time 0.021782 +2023-10-05 21:50:03,394 - Epoch: [144][ 450/ 1236] Overall Loss 0.218396 Objective Loss 0.218396 LR 0.000250 Time 0.021746 +2023-10-05 21:50:03,593 - Epoch: [144][ 460/ 1236] Overall Loss 0.217941 Objective Loss 0.217941 LR 0.000250 Time 0.021707 +2023-10-05 21:50:03,794 - Epoch: [144][ 470/ 1236] Overall Loss 0.218162 Objective Loss 0.218162 LR 0.000250 Time 0.021672 +2023-10-05 21:50:03,996 - Epoch: [144][ 480/ 1236] Overall Loss 0.217869 Objective Loss 0.217869 LR 0.000250 Time 0.021639 +2023-10-05 21:50:04,198 - Epoch: [144][ 490/ 1236] Overall Loss 0.217508 Objective Loss 0.217508 LR 0.000250 Time 0.021610 +2023-10-05 21:50:04,400 - Epoch: [144][ 500/ 1236] Overall Loss 0.218115 Objective Loss 0.218115 LR 0.000250 Time 0.021580 +2023-10-05 21:50:04,603 - Epoch: [144][ 510/ 1236] Overall Loss 0.218436 Objective Loss 0.218436 LR 0.000250 Time 0.021556 +2023-10-05 21:50:04,806 - Epoch: [144][ 520/ 1236] Overall Loss 0.218274 Objective Loss 0.218274 LR 0.000250 Time 0.021530 +2023-10-05 21:50:05,009 - Epoch: [144][ 530/ 1236] Overall Loss 0.218266 Objective Loss 0.218266 LR 0.000250 Time 0.021506 +2023-10-05 21:50:05,211 - Epoch: [144][ 540/ 1236] Overall Loss 0.218266 Objective Loss 0.218266 LR 0.000250 Time 0.021481 +2023-10-05 21:50:05,414 - Epoch: [144][ 550/ 1236] Overall Loss 0.218052 Objective Loss 0.218052 LR 0.000250 Time 0.021460 +2023-10-05 21:50:05,616 - Epoch: [144][ 560/ 1236] Overall Loss 0.217822 Objective Loss 0.217822 LR 0.000250 Time 0.021435 +2023-10-05 21:50:05,819 - Epoch: [144][ 570/ 1236] Overall Loss 0.217579 Objective Loss 0.217579 LR 0.000250 Time 0.021415 +2023-10-05 21:50:06,022 - Epoch: [144][ 580/ 1236] Overall Loss 0.217752 Objective Loss 0.217752 LR 0.000250 Time 0.021396 +2023-10-05 21:50:06,224 - Epoch: [144][ 590/ 1236] Overall Loss 0.217731 Objective Loss 0.217731 LR 0.000250 Time 0.021375 +2023-10-05 21:50:06,427 - Epoch: [144][ 600/ 1236] Overall Loss 0.217854 Objective Loss 0.217854 LR 0.000250 Time 0.021356 +2023-10-05 21:50:06,630 - Epoch: [144][ 610/ 1236] Overall Loss 0.218113 Objective Loss 0.218113 LR 0.000250 Time 0.021338 +2023-10-05 21:50:06,835 - Epoch: [144][ 620/ 1236] Overall Loss 0.217936 Objective Loss 0.217936 LR 0.000250 Time 0.021324 +2023-10-05 21:50:07,039 - Epoch: [144][ 630/ 1236] Overall Loss 0.217881 Objective Loss 0.217881 LR 0.000250 Time 0.021308 +2023-10-05 21:50:07,244 - Epoch: [144][ 640/ 1236] Overall Loss 0.218182 Objective Loss 0.218182 LR 0.000250 Time 0.021296 +2023-10-05 21:50:07,448 - Epoch: [144][ 650/ 1236] Overall Loss 0.217936 Objective Loss 0.217936 LR 0.000250 Time 0.021282 +2023-10-05 21:50:07,653 - Epoch: [144][ 660/ 1236] Overall Loss 0.218165 Objective Loss 0.218165 LR 0.000250 Time 0.021269 +2023-10-05 21:50:07,857 - Epoch: [144][ 670/ 1236] Overall Loss 0.217828 Objective Loss 0.217828 LR 0.000250 Time 0.021254 +2023-10-05 21:50:08,061 - Epoch: [144][ 680/ 1236] Overall Loss 0.217972 Objective Loss 0.217972 LR 0.000250 Time 0.021240 +2023-10-05 21:50:08,263 - Epoch: [144][ 690/ 1236] Overall Loss 0.217857 Objective Loss 0.217857 LR 0.000250 Time 0.021225 +2023-10-05 21:50:08,466 - Epoch: [144][ 700/ 1236] Overall Loss 0.217461 Objective Loss 0.217461 LR 0.000250 Time 0.021211 +2023-10-05 21:50:08,668 - Epoch: [144][ 710/ 1236] Overall Loss 0.217150 Objective Loss 0.217150 LR 0.000250 Time 0.021197 +2023-10-05 21:50:08,872 - Epoch: [144][ 720/ 1236] Overall Loss 0.217529 Objective Loss 0.217529 LR 0.000250 Time 0.021185 +2023-10-05 21:50:09,074 - Epoch: [144][ 730/ 1236] Overall Loss 0.217615 Objective Loss 0.217615 LR 0.000250 Time 0.021171 +2023-10-05 21:50:09,278 - Epoch: [144][ 740/ 1236] Overall Loss 0.217526 Objective Loss 0.217526 LR 0.000250 Time 0.021160 +2023-10-05 21:50:09,480 - Epoch: [144][ 750/ 1236] Overall Loss 0.217583 Objective Loss 0.217583 LR 0.000250 Time 0.021146 +2023-10-05 21:50:09,683 - Epoch: [144][ 760/ 1236] Overall Loss 0.217374 Objective Loss 0.217374 LR 0.000250 Time 0.021135 +2023-10-05 21:50:09,886 - Epoch: [144][ 770/ 1236] Overall Loss 0.217152 Objective Loss 0.217152 LR 0.000250 Time 0.021124 +2023-10-05 21:50:10,089 - Epoch: [144][ 780/ 1236] Overall Loss 0.217328 Objective Loss 0.217328 LR 0.000250 Time 0.021113 +2023-10-05 21:50:10,292 - Epoch: [144][ 790/ 1236] Overall Loss 0.217498 Objective Loss 0.217498 LR 0.000250 Time 0.021101 +2023-10-05 21:50:10,495 - Epoch: [144][ 800/ 1236] Overall Loss 0.217372 Objective Loss 0.217372 LR 0.000250 Time 0.021091 +2023-10-05 21:50:10,698 - Epoch: [144][ 810/ 1236] Overall Loss 0.217246 Objective Loss 0.217246 LR 0.000250 Time 0.021081 +2023-10-05 21:50:10,901 - Epoch: [144][ 820/ 1236] Overall Loss 0.217014 Objective Loss 0.217014 LR 0.000250 Time 0.021071 +2023-10-05 21:50:11,104 - Epoch: [144][ 830/ 1236] Overall Loss 0.217065 Objective Loss 0.217065 LR 0.000250 Time 0.021061 +2023-10-05 21:50:11,307 - Epoch: [144][ 840/ 1236] Overall Loss 0.217183 Objective Loss 0.217183 LR 0.000250 Time 0.021052 +2023-10-05 21:50:11,510 - Epoch: [144][ 850/ 1236] Overall Loss 0.217099 Objective Loss 0.217099 LR 0.000250 Time 0.021042 +2023-10-05 21:50:11,713 - Epoch: [144][ 860/ 1236] Overall Loss 0.217272 Objective Loss 0.217272 LR 0.000250 Time 0.021033 +2023-10-05 21:50:11,915 - Epoch: [144][ 870/ 1236] Overall Loss 0.217601 Objective Loss 0.217601 LR 0.000250 Time 0.021024 +2023-10-05 21:50:12,119 - Epoch: [144][ 880/ 1236] Overall Loss 0.217940 Objective Loss 0.217940 LR 0.000250 Time 0.021016 +2023-10-05 21:50:12,325 - Epoch: [144][ 890/ 1236] Overall Loss 0.218071 Objective Loss 0.218071 LR 0.000250 Time 0.021011 +2023-10-05 21:50:12,528 - Epoch: [144][ 900/ 1236] Overall Loss 0.218258 Objective Loss 0.218258 LR 0.000250 Time 0.021003 +2023-10-05 21:50:12,731 - Epoch: [144][ 910/ 1236] Overall Loss 0.218638 Objective Loss 0.218638 LR 0.000250 Time 0.020995 +2023-10-05 21:50:12,934 - Epoch: [144][ 920/ 1236] Overall Loss 0.218335 Objective Loss 0.218335 LR 0.000250 Time 0.020987 +2023-10-05 21:50:13,137 - Epoch: [144][ 930/ 1236] Overall Loss 0.218562 Objective Loss 0.218562 LR 0.000250 Time 0.020979 +2023-10-05 21:50:13,341 - Epoch: [144][ 940/ 1236] Overall Loss 0.218469 Objective Loss 0.218469 LR 0.000250 Time 0.020972 +2023-10-05 21:50:13,543 - Epoch: [144][ 950/ 1236] Overall Loss 0.218467 Objective Loss 0.218467 LR 0.000250 Time 0.020964 +2023-10-05 21:50:13,746 - Epoch: [144][ 960/ 1236] Overall Loss 0.218547 Objective Loss 0.218547 LR 0.000250 Time 0.020957 +2023-10-05 21:50:13,949 - Epoch: [144][ 970/ 1236] Overall Loss 0.218549 Objective Loss 0.218549 LR 0.000250 Time 0.020950 +2023-10-05 21:50:14,152 - Epoch: [144][ 980/ 1236] Overall Loss 0.218640 Objective Loss 0.218640 LR 0.000250 Time 0.020943 +2023-10-05 21:50:14,355 - Epoch: [144][ 990/ 1236] Overall Loss 0.218615 Objective Loss 0.218615 LR 0.000250 Time 0.020936 +2023-10-05 21:50:14,558 - Epoch: [144][ 1000/ 1236] Overall Loss 0.218490 Objective Loss 0.218490 LR 0.000250 Time 0.020929 +2023-10-05 21:50:14,761 - Epoch: [144][ 1010/ 1236] Overall Loss 0.218529 Objective Loss 0.218529 LR 0.000250 Time 0.020922 +2023-10-05 21:50:14,965 - Epoch: [144][ 1020/ 1236] Overall Loss 0.218585 Objective Loss 0.218585 LR 0.000250 Time 0.020916 +2023-10-05 21:50:15,167 - Epoch: [144][ 1030/ 1236] Overall Loss 0.218490 Objective Loss 0.218490 LR 0.000250 Time 0.020909 +2023-10-05 21:50:15,371 - Epoch: [144][ 1040/ 1236] Overall Loss 0.218521 Objective Loss 0.218521 LR 0.000250 Time 0.020904 +2023-10-05 21:50:15,574 - Epoch: [144][ 1050/ 1236] Overall Loss 0.218329 Objective Loss 0.218329 LR 0.000250 Time 0.020898 +2023-10-05 21:50:15,777 - Epoch: [144][ 1060/ 1236] Overall Loss 0.218398 Objective Loss 0.218398 LR 0.000250 Time 0.020892 +2023-10-05 21:50:15,980 - Epoch: [144][ 1070/ 1236] Overall Loss 0.218279 Objective Loss 0.218279 LR 0.000250 Time 0.020886 +2023-10-05 21:50:16,183 - Epoch: [144][ 1080/ 1236] Overall Loss 0.218152 Objective Loss 0.218152 LR 0.000250 Time 0.020881 +2023-10-05 21:50:16,387 - Epoch: [144][ 1090/ 1236] Overall Loss 0.218328 Objective Loss 0.218328 LR 0.000250 Time 0.020876 +2023-10-05 21:50:16,590 - Epoch: [144][ 1100/ 1236] Overall Loss 0.218281 Objective Loss 0.218281 LR 0.000250 Time 0.020870 +2023-10-05 21:50:16,795 - Epoch: [144][ 1110/ 1236] Overall Loss 0.218315 Objective Loss 0.218315 LR 0.000250 Time 0.020866 +2023-10-05 21:50:16,998 - Epoch: [144][ 1120/ 1236] Overall Loss 0.218190 Objective Loss 0.218190 LR 0.000250 Time 0.020861 +2023-10-05 21:50:17,202 - Epoch: [144][ 1130/ 1236] Overall Loss 0.217902 Objective Loss 0.217902 LR 0.000250 Time 0.020856 +2023-10-05 21:50:17,411 - Epoch: [144][ 1140/ 1236] Overall Loss 0.217931 Objective Loss 0.217931 LR 0.000250 Time 0.020856 +2023-10-05 21:50:17,625 - Epoch: [144][ 1150/ 1236] Overall Loss 0.217857 Objective Loss 0.217857 LR 0.000250 Time 0.020861 +2023-10-05 21:50:17,834 - Epoch: [144][ 1160/ 1236] Overall Loss 0.218077 Objective Loss 0.218077 LR 0.000250 Time 0.020862 +2023-10-05 21:50:18,049 - Epoch: [144][ 1170/ 1236] Overall Loss 0.218175 Objective Loss 0.218175 LR 0.000250 Time 0.020867 +2023-10-05 21:50:18,258 - Epoch: [144][ 1180/ 1236] Overall Loss 0.218224 Objective Loss 0.218224 LR 0.000250 Time 0.020867 +2023-10-05 21:50:18,473 - Epoch: [144][ 1190/ 1236] Overall Loss 0.218178 Objective Loss 0.218178 LR 0.000250 Time 0.020872 +2023-10-05 21:50:18,682 - Epoch: [144][ 1200/ 1236] Overall Loss 0.218379 Objective Loss 0.218379 LR 0.000250 Time 0.020871 +2023-10-05 21:50:18,897 - Epoch: [144][ 1210/ 1236] Overall Loss 0.218394 Objective Loss 0.218394 LR 0.000250 Time 0.020876 +2023-10-05 21:50:19,106 - Epoch: [144][ 1220/ 1236] Overall Loss 0.218467 Objective Loss 0.218467 LR 0.000250 Time 0.020876 +2023-10-05 21:50:19,367 - Epoch: [144][ 1230/ 1236] Overall Loss 0.218601 Objective Loss 0.218601 LR 0.000250 Time 0.020919 +2023-10-05 21:50:19,486 - Epoch: [144][ 1236/ 1236] Overall Loss 0.218550 Objective Loss 0.218550 Top1 89.409369 Top5 98.574338 LR 0.000250 Time 0.020913 +2023-10-05 21:50:19,619 - --- validate (epoch=144)----------- +2023-10-05 21:50:19,620 - 29943 samples (256 per mini-batch) +2023-10-05 21:50:20,078 - Epoch: [144][ 10/ 117] Loss 0.288394 Top1 85.117188 Top5 98.007812 +2023-10-05 21:50:20,226 - Epoch: [144][ 20/ 117] Loss 0.293427 Top1 84.238281 Top5 97.968750 +2023-10-05 21:50:20,373 - Epoch: [144][ 30/ 117] Loss 0.301633 Top1 84.414062 Top5 98.007812 +2023-10-05 21:50:20,519 - Epoch: [144][ 40/ 117] Loss 0.301283 Top1 84.335938 Top5 97.978516 +2023-10-05 21:50:20,665 - Epoch: [144][ 50/ 117] Loss 0.305069 Top1 84.289062 Top5 97.960938 +2023-10-05 21:50:20,809 - Epoch: [144][ 60/ 117] Loss 0.307211 Top1 84.329427 Top5 97.968750 +2023-10-05 21:50:20,953 - Epoch: [144][ 70/ 117] Loss 0.306029 Top1 84.308036 Top5 97.968750 +2023-10-05 21:50:21,101 - Epoch: [144][ 80/ 117] Loss 0.301565 Top1 84.472656 Top5 97.993164 +2023-10-05 21:50:21,255 - Epoch: [144][ 90/ 117] Loss 0.304073 Top1 84.561632 Top5 97.951389 +2023-10-05 21:50:21,408 - Epoch: [144][ 100/ 117] Loss 0.305594 Top1 84.460938 Top5 97.953125 +2023-10-05 21:50:21,569 - Epoch: [144][ 110/ 117] Loss 0.304541 Top1 84.609375 Top5 97.975852 +2023-10-05 21:50:21,655 - Epoch: [144][ 117/ 117] Loss 0.305254 Top1 84.654176 Top5 97.969475 +2023-10-05 21:50:21,792 - ==> Top1: 84.654 Top5: 97.969 Loss: 0.305 + +2023-10-05 21:50:21,792 - ==> Confusion: +[[ 942 4 3 0 6 3 0 0 6 63 1 0 2 2 4 1 2 1 0 0 10] + [ 0 1061 3 0 8 24 1 15 0 0 0 3 0 0 0 2 1 0 6 3 4] + [ 7 1 970 16 1 0 20 6 0 0 3 3 10 1 1 4 1 1 2 5 4] + [ 4 0 13 990 1 4 1 1 0 1 6 0 3 1 27 4 0 7 14 2 10] + [ 22 7 1 0 975 7 0 0 0 8 2 0 0 2 9 5 7 1 0 1 3] + [ 4 31 0 1 1 1002 1 20 2 2 3 8 1 12 6 2 3 0 1 3 13] + [ 0 7 27 0 0 1 1122 7 0 0 3 1 1 0 1 7 0 1 0 11 2] + [ 2 12 13 1 2 36 4 1082 0 1 4 11 2 1 0 2 0 1 26 8 10] + [ 22 6 0 1 0 4 0 1 966 46 10 1 1 8 12 5 0 0 3 0 3] + [ 92 0 4 0 3 4 1 1 22 958 1 2 1 16 3 5 0 2 0 0 4] + [ 2 5 14 5 2 4 3 2 11 3 968 2 2 10 5 1 1 1 2 1 9] + [ 2 0 2 0 0 15 0 1 0 1 0 967 15 5 0 4 0 17 0 5 1] + [ 0 1 5 6 0 3 3 0 0 0 1 39 976 1 0 6 2 18 0 2 5] + [ 2 0 0 0 3 8 0 0 6 10 4 5 2 1065 3 3 1 1 0 0 6] + [ 14 3 3 13 8 1 0 0 22 1 2 2 1 2 1007 0 1 1 8 0 12] + [ 0 4 2 0 2 0 0 0 0 1 1 8 6 2 1 1069 12 13 1 10 2] + [ 2 14 2 1 4 2 0 1 1 0 0 4 1 1 2 13 1097 0 0 5 11] + [ 0 0 0 1 1 0 4 0 0 0 0 2 11 1 1 5 0 1008 1 0 3] + [ 1 6 6 23 1 1 0 27 1 0 2 0 3 0 9 0 1 1 980 0 6] + [ 0 1 4 1 1 5 11 5 1 1 0 12 2 1 0 8 9 2 3 1081 4] + [ 146 176 178 70 90 170 45 82 90 66 173 107 373 294 155 66 121 83 128 230 5062]] + +2023-10-05 21:50:21,794 - ==> Best [Top1: 84.764 Top5: 97.953 Sparsity:0.00 Params: 148928 on epoch: 141] +2023-10-05 21:50:21,794 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:50:21,800 - + +2023-10-05 21:50:21,800 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:50:22,923 - Epoch: [145][ 10/ 1236] Overall Loss 0.212878 Objective Loss 0.212878 LR 0.000250 Time 0.112303 +2023-10-05 21:50:23,128 - Epoch: [145][ 20/ 1236] Overall Loss 0.206708 Objective Loss 0.206708 LR 0.000250 Time 0.066391 +2023-10-05 21:50:23,331 - Epoch: [145][ 30/ 1236] Overall Loss 0.217130 Objective Loss 0.217130 LR 0.000250 Time 0.050991 +2023-10-05 21:50:23,535 - Epoch: [145][ 40/ 1236] Overall Loss 0.215091 Objective Loss 0.215091 LR 0.000250 Time 0.043353 +2023-10-05 21:50:23,738 - Epoch: [145][ 50/ 1236] Overall Loss 0.217106 Objective Loss 0.217106 LR 0.000250 Time 0.038726 +2023-10-05 21:50:23,943 - Epoch: [145][ 60/ 1236] Overall Loss 0.217342 Objective Loss 0.217342 LR 0.000250 Time 0.035680 +2023-10-05 21:50:24,146 - Epoch: [145][ 70/ 1236] Overall Loss 0.214926 Objective Loss 0.214926 LR 0.000250 Time 0.033477 +2023-10-05 21:50:24,350 - Epoch: [145][ 80/ 1236] Overall Loss 0.212764 Objective Loss 0.212764 LR 0.000250 Time 0.031846 +2023-10-05 21:50:24,553 - Epoch: [145][ 90/ 1236] Overall Loss 0.214115 Objective Loss 0.214115 LR 0.000250 Time 0.030558 +2023-10-05 21:50:24,758 - Epoch: [145][ 100/ 1236] Overall Loss 0.214795 Objective Loss 0.214795 LR 0.000250 Time 0.029545 +2023-10-05 21:50:24,962 - Epoch: [145][ 110/ 1236] Overall Loss 0.213136 Objective Loss 0.213136 LR 0.000250 Time 0.028713 +2023-10-05 21:50:25,166 - Epoch: [145][ 120/ 1236] Overall Loss 0.214405 Objective Loss 0.214405 LR 0.000250 Time 0.028017 +2023-10-05 21:50:25,368 - Epoch: [145][ 130/ 1236] Overall Loss 0.214210 Objective Loss 0.214210 LR 0.000250 Time 0.027413 +2023-10-05 21:50:25,570 - Epoch: [145][ 140/ 1236] Overall Loss 0.214103 Objective Loss 0.214103 LR 0.000250 Time 0.026897 +2023-10-05 21:50:25,771 - Epoch: [145][ 150/ 1236] Overall Loss 0.215815 Objective Loss 0.215815 LR 0.000250 Time 0.026440 +2023-10-05 21:50:25,973 - Epoch: [145][ 160/ 1236] Overall Loss 0.215330 Objective Loss 0.215330 LR 0.000250 Time 0.026050 +2023-10-05 21:50:26,175 - Epoch: [145][ 170/ 1236] Overall Loss 0.215379 Objective Loss 0.215379 LR 0.000250 Time 0.025704 +2023-10-05 21:50:26,377 - Epoch: [145][ 180/ 1236] Overall Loss 0.215555 Objective Loss 0.215555 LR 0.000250 Time 0.025396 +2023-10-05 21:50:26,579 - Epoch: [145][ 190/ 1236] Overall Loss 0.214945 Objective Loss 0.214945 LR 0.000250 Time 0.025121 +2023-10-05 21:50:26,782 - Epoch: [145][ 200/ 1236] Overall Loss 0.214267 Objective Loss 0.214267 LR 0.000250 Time 0.024877 +2023-10-05 21:50:26,985 - Epoch: [145][ 210/ 1236] Overall Loss 0.215331 Objective Loss 0.215331 LR 0.000250 Time 0.024655 +2023-10-05 21:50:27,187 - Epoch: [145][ 220/ 1236] Overall Loss 0.215891 Objective Loss 0.215891 LR 0.000250 Time 0.024453 +2023-10-05 21:50:27,389 - Epoch: [145][ 230/ 1236] Overall Loss 0.215730 Objective Loss 0.215730 LR 0.000250 Time 0.024267 +2023-10-05 21:50:27,592 - Epoch: [145][ 240/ 1236] Overall Loss 0.215718 Objective Loss 0.215718 LR 0.000250 Time 0.024100 +2023-10-05 21:50:27,794 - Epoch: [145][ 250/ 1236] Overall Loss 0.214516 Objective Loss 0.214516 LR 0.000250 Time 0.023943 +2023-10-05 21:50:27,997 - Epoch: [145][ 260/ 1236] Overall Loss 0.214045 Objective Loss 0.214045 LR 0.000250 Time 0.023798 +2023-10-05 21:50:28,198 - Epoch: [145][ 270/ 1236] Overall Loss 0.214579 Objective Loss 0.214579 LR 0.000250 Time 0.023663 +2023-10-05 21:50:28,401 - Epoch: [145][ 280/ 1236] Overall Loss 0.213375 Objective Loss 0.213375 LR 0.000250 Time 0.023541 +2023-10-05 21:50:28,603 - Epoch: [145][ 290/ 1236] Overall Loss 0.213490 Objective Loss 0.213490 LR 0.000250 Time 0.023425 +2023-10-05 21:50:28,806 - Epoch: [145][ 300/ 1236] Overall Loss 0.212887 Objective Loss 0.212887 LR 0.000250 Time 0.023319 +2023-10-05 21:50:29,009 - Epoch: [145][ 310/ 1236] Overall Loss 0.213414 Objective Loss 0.213414 LR 0.000250 Time 0.023219 +2023-10-05 21:50:29,211 - Epoch: [145][ 320/ 1236] Overall Loss 0.214064 Objective Loss 0.214064 LR 0.000250 Time 0.023126 +2023-10-05 21:50:29,413 - Epoch: [145][ 330/ 1236] Overall Loss 0.214326 Objective Loss 0.214326 LR 0.000250 Time 0.023036 +2023-10-05 21:50:29,616 - Epoch: [145][ 340/ 1236] Overall Loss 0.215041 Objective Loss 0.215041 LR 0.000250 Time 0.022954 +2023-10-05 21:50:29,818 - Epoch: [145][ 350/ 1236] Overall Loss 0.215829 Objective Loss 0.215829 LR 0.000250 Time 0.022875 +2023-10-05 21:50:30,021 - Epoch: [145][ 360/ 1236] Overall Loss 0.216379 Objective Loss 0.216379 LR 0.000250 Time 0.022803 +2023-10-05 21:50:30,224 - Epoch: [145][ 370/ 1236] Overall Loss 0.215990 Objective Loss 0.215990 LR 0.000250 Time 0.022733 +2023-10-05 21:50:30,429 - Epoch: [145][ 380/ 1236] Overall Loss 0.216010 Objective Loss 0.216010 LR 0.000250 Time 0.022672 +2023-10-05 21:50:30,631 - Epoch: [145][ 390/ 1236] Overall Loss 0.215798 Objective Loss 0.215798 LR 0.000250 Time 0.022610 +2023-10-05 21:50:30,836 - Epoch: [145][ 400/ 1236] Overall Loss 0.215829 Objective Loss 0.215829 LR 0.000250 Time 0.022556 +2023-10-05 21:50:31,040 - Epoch: [145][ 410/ 1236] Overall Loss 0.215947 Objective Loss 0.215947 LR 0.000250 Time 0.022501 +2023-10-05 21:50:31,244 - Epoch: [145][ 420/ 1236] Overall Loss 0.216277 Objective Loss 0.216277 LR 0.000250 Time 0.022452 +2023-10-05 21:50:31,448 - Epoch: [145][ 430/ 1236] Overall Loss 0.215959 Objective Loss 0.215959 LR 0.000250 Time 0.022402 +2023-10-05 21:50:31,652 - Epoch: [145][ 440/ 1236] Overall Loss 0.215893 Objective Loss 0.215893 LR 0.000250 Time 0.022357 +2023-10-05 21:50:31,855 - Epoch: [145][ 450/ 1236] Overall Loss 0.215876 Objective Loss 0.215876 LR 0.000250 Time 0.022311 +2023-10-05 21:50:32,061 - Epoch: [145][ 460/ 1236] Overall Loss 0.215784 Objective Loss 0.215784 LR 0.000250 Time 0.022271 +2023-10-05 21:50:32,264 - Epoch: [145][ 470/ 1236] Overall Loss 0.215420 Objective Loss 0.215420 LR 0.000250 Time 0.022230 +2023-10-05 21:50:32,469 - Epoch: [145][ 480/ 1236] Overall Loss 0.215460 Objective Loss 0.215460 LR 0.000250 Time 0.022192 +2023-10-05 21:50:32,672 - Epoch: [145][ 490/ 1236] Overall Loss 0.215357 Objective Loss 0.215357 LR 0.000250 Time 0.022153 +2023-10-05 21:50:32,877 - Epoch: [145][ 500/ 1236] Overall Loss 0.215265 Objective Loss 0.215265 LR 0.000250 Time 0.022119 +2023-10-05 21:50:33,080 - Epoch: [145][ 510/ 1236] Overall Loss 0.215351 Objective Loss 0.215351 LR 0.000250 Time 0.022084 +2023-10-05 21:50:33,285 - Epoch: [145][ 520/ 1236] Overall Loss 0.215179 Objective Loss 0.215179 LR 0.000250 Time 0.022052 +2023-10-05 21:50:33,488 - Epoch: [145][ 530/ 1236] Overall Loss 0.215054 Objective Loss 0.215054 LR 0.000250 Time 0.022019 +2023-10-05 21:50:33,693 - Epoch: [145][ 540/ 1236] Overall Loss 0.215540 Objective Loss 0.215540 LR 0.000250 Time 0.021989 +2023-10-05 21:50:33,896 - Epoch: [145][ 550/ 1236] Overall Loss 0.215263 Objective Loss 0.215263 LR 0.000250 Time 0.021959 +2023-10-05 21:50:34,100 - Epoch: [145][ 560/ 1236] Overall Loss 0.215257 Objective Loss 0.215257 LR 0.000250 Time 0.021929 +2023-10-05 21:50:34,303 - Epoch: [145][ 570/ 1236] Overall Loss 0.215403 Objective Loss 0.215403 LR 0.000250 Time 0.021900 +2023-10-05 21:50:34,508 - Epoch: [145][ 580/ 1236] Overall Loss 0.215487 Objective Loss 0.215487 LR 0.000250 Time 0.021875 +2023-10-05 21:50:34,711 - Epoch: [145][ 590/ 1236] Overall Loss 0.215728 Objective Loss 0.215728 LR 0.000250 Time 0.021848 +2023-10-05 21:50:34,916 - Epoch: [145][ 600/ 1236] Overall Loss 0.215703 Objective Loss 0.215703 LR 0.000250 Time 0.021825 +2023-10-05 21:50:35,119 - Epoch: [145][ 610/ 1236] Overall Loss 0.215576 Objective Loss 0.215576 LR 0.000250 Time 0.021800 +2023-10-05 21:50:35,324 - Epoch: [145][ 620/ 1236] Overall Loss 0.215188 Objective Loss 0.215188 LR 0.000250 Time 0.021778 +2023-10-05 21:50:35,527 - Epoch: [145][ 630/ 1236] Overall Loss 0.215150 Objective Loss 0.215150 LR 0.000250 Time 0.021754 +2023-10-05 21:50:35,732 - Epoch: [145][ 640/ 1236] Overall Loss 0.215824 Objective Loss 0.215824 LR 0.000250 Time 0.021734 +2023-10-05 21:50:35,935 - Epoch: [145][ 650/ 1236] Overall Loss 0.215725 Objective Loss 0.215725 LR 0.000250 Time 0.021711 +2023-10-05 21:50:36,140 - Epoch: [145][ 660/ 1236] Overall Loss 0.215612 Objective Loss 0.215612 LR 0.000250 Time 0.021693 +2023-10-05 21:50:36,343 - Epoch: [145][ 670/ 1236] Overall Loss 0.215724 Objective Loss 0.215724 LR 0.000250 Time 0.021672 +2023-10-05 21:50:36,548 - Epoch: [145][ 680/ 1236] Overall Loss 0.215442 Objective Loss 0.215442 LR 0.000250 Time 0.021653 +2023-10-05 21:50:36,751 - Epoch: [145][ 690/ 1236] Overall Loss 0.215439 Objective Loss 0.215439 LR 0.000250 Time 0.021634 +2023-10-05 21:50:36,956 - Epoch: [145][ 700/ 1236] Overall Loss 0.215677 Objective Loss 0.215677 LR 0.000250 Time 0.021617 +2023-10-05 21:50:37,159 - Epoch: [145][ 710/ 1236] Overall Loss 0.215834 Objective Loss 0.215834 LR 0.000250 Time 0.021598 +2023-10-05 21:50:37,364 - Epoch: [145][ 720/ 1236] Overall Loss 0.215996 Objective Loss 0.215996 LR 0.000250 Time 0.021582 +2023-10-05 21:50:37,567 - Epoch: [145][ 730/ 1236] Overall Loss 0.216244 Objective Loss 0.216244 LR 0.000250 Time 0.021564 +2023-10-05 21:50:37,772 - Epoch: [145][ 740/ 1236] Overall Loss 0.216391 Objective Loss 0.216391 LR 0.000250 Time 0.021549 +2023-10-05 21:50:37,975 - Epoch: [145][ 750/ 1236] Overall Loss 0.216495 Objective Loss 0.216495 LR 0.000250 Time 0.021532 +2023-10-05 21:50:38,180 - Epoch: [145][ 760/ 1236] Overall Loss 0.216472 Objective Loss 0.216472 LR 0.000250 Time 0.021518 +2023-10-05 21:50:38,382 - Epoch: [145][ 770/ 1236] Overall Loss 0.216224 Objective Loss 0.216224 LR 0.000250 Time 0.021500 +2023-10-05 21:50:38,586 - Epoch: [145][ 780/ 1236] Overall Loss 0.216034 Objective Loss 0.216034 LR 0.000250 Time 0.021486 +2023-10-05 21:50:38,788 - Epoch: [145][ 790/ 1236] Overall Loss 0.216252 Objective Loss 0.216252 LR 0.000250 Time 0.021469 +2023-10-05 21:50:38,992 - Epoch: [145][ 800/ 1236] Overall Loss 0.216501 Objective Loss 0.216501 LR 0.000250 Time 0.021455 +2023-10-05 21:50:39,194 - Epoch: [145][ 810/ 1236] Overall Loss 0.216459 Objective Loss 0.216459 LR 0.000250 Time 0.021440 +2023-10-05 21:50:39,398 - Epoch: [145][ 820/ 1236] Overall Loss 0.216517 Objective Loss 0.216517 LR 0.000250 Time 0.021426 +2023-10-05 21:50:39,601 - Epoch: [145][ 830/ 1236] Overall Loss 0.216320 Objective Loss 0.216320 LR 0.000250 Time 0.021412 +2023-10-05 21:50:39,804 - Epoch: [145][ 840/ 1236] Overall Loss 0.216355 Objective Loss 0.216355 LR 0.000250 Time 0.021399 +2023-10-05 21:50:40,007 - Epoch: [145][ 850/ 1236] Overall Loss 0.216218 Objective Loss 0.216218 LR 0.000250 Time 0.021385 +2023-10-05 21:50:40,210 - Epoch: [145][ 860/ 1236] Overall Loss 0.216400 Objective Loss 0.216400 LR 0.000250 Time 0.021373 +2023-10-05 21:50:40,413 - Epoch: [145][ 870/ 1236] Overall Loss 0.216307 Objective Loss 0.216307 LR 0.000250 Time 0.021359 +2023-10-05 21:50:40,617 - Epoch: [145][ 880/ 1236] Overall Loss 0.216408 Objective Loss 0.216408 LR 0.000250 Time 0.021348 +2023-10-05 21:50:40,819 - Epoch: [145][ 890/ 1236] Overall Loss 0.216448 Objective Loss 0.216448 LR 0.000250 Time 0.021335 +2023-10-05 21:50:41,023 - Epoch: [145][ 900/ 1236] Overall Loss 0.216753 Objective Loss 0.216753 LR 0.000250 Time 0.021324 +2023-10-05 21:50:41,225 - Epoch: [145][ 910/ 1236] Overall Loss 0.216686 Objective Loss 0.216686 LR 0.000250 Time 0.021311 +2023-10-05 21:50:41,429 - Epoch: [145][ 920/ 1236] Overall Loss 0.216740 Objective Loss 0.216740 LR 0.000250 Time 0.021301 +2023-10-05 21:50:41,632 - Epoch: [145][ 930/ 1236] Overall Loss 0.216555 Objective Loss 0.216555 LR 0.000250 Time 0.021290 +2023-10-05 21:50:41,836 - Epoch: [145][ 940/ 1236] Overall Loss 0.216460 Objective Loss 0.216460 LR 0.000250 Time 0.021280 +2023-10-05 21:50:42,038 - Epoch: [145][ 950/ 1236] Overall Loss 0.216320 Objective Loss 0.216320 LR 0.000250 Time 0.021268 +2023-10-05 21:50:42,242 - Epoch: [145][ 960/ 1236] Overall Loss 0.216338 Objective Loss 0.216338 LR 0.000250 Time 0.021259 +2023-10-05 21:50:42,444 - Epoch: [145][ 970/ 1236] Overall Loss 0.216372 Objective Loss 0.216372 LR 0.000250 Time 0.021248 +2023-10-05 21:50:42,648 - Epoch: [145][ 980/ 1236] Overall Loss 0.216179 Objective Loss 0.216179 LR 0.000250 Time 0.021239 +2023-10-05 21:50:42,850 - Epoch: [145][ 990/ 1236] Overall Loss 0.216143 Objective Loss 0.216143 LR 0.000250 Time 0.021228 +2023-10-05 21:50:43,054 - Epoch: [145][ 1000/ 1236] Overall Loss 0.216042 Objective Loss 0.216042 LR 0.000250 Time 0.021219 +2023-10-05 21:50:43,256 - Epoch: [145][ 1010/ 1236] Overall Loss 0.216056 Objective Loss 0.216056 LR 0.000250 Time 0.021209 +2023-10-05 21:50:43,460 - Epoch: [145][ 1020/ 1236] Overall Loss 0.216145 Objective Loss 0.216145 LR 0.000250 Time 0.021200 +2023-10-05 21:50:43,663 - Epoch: [145][ 1030/ 1236] Overall Loss 0.216096 Objective Loss 0.216096 LR 0.000250 Time 0.021191 +2023-10-05 21:50:43,867 - Epoch: [145][ 1040/ 1236] Overall Loss 0.216217 Objective Loss 0.216217 LR 0.000250 Time 0.021183 +2023-10-05 21:50:44,069 - Epoch: [145][ 1050/ 1236] Overall Loss 0.216559 Objective Loss 0.216559 LR 0.000250 Time 0.021174 +2023-10-05 21:50:44,273 - Epoch: [145][ 1060/ 1236] Overall Loss 0.216598 Objective Loss 0.216598 LR 0.000250 Time 0.021166 +2023-10-05 21:50:44,476 - Epoch: [145][ 1070/ 1236] Overall Loss 0.216784 Objective Loss 0.216784 LR 0.000250 Time 0.021157 +2023-10-05 21:50:44,679 - Epoch: [145][ 1080/ 1236] Overall Loss 0.216864 Objective Loss 0.216864 LR 0.000250 Time 0.021150 +2023-10-05 21:50:44,882 - Epoch: [145][ 1090/ 1236] Overall Loss 0.216771 Objective Loss 0.216771 LR 0.000250 Time 0.021141 +2023-10-05 21:50:45,086 - Epoch: [145][ 1100/ 1236] Overall Loss 0.216593 Objective Loss 0.216593 LR 0.000250 Time 0.021134 +2023-10-05 21:50:45,288 - Epoch: [145][ 1110/ 1236] Overall Loss 0.216550 Objective Loss 0.216550 LR 0.000250 Time 0.021125 +2023-10-05 21:50:45,492 - Epoch: [145][ 1120/ 1236] Overall Loss 0.216261 Objective Loss 0.216261 LR 0.000250 Time 0.021118 +2023-10-05 21:50:45,694 - Epoch: [145][ 1130/ 1236] Overall Loss 0.216299 Objective Loss 0.216299 LR 0.000250 Time 0.021110 +2023-10-05 21:50:45,898 - Epoch: [145][ 1140/ 1236] Overall Loss 0.216247 Objective Loss 0.216247 LR 0.000250 Time 0.021104 +2023-10-05 21:50:46,100 - Epoch: [145][ 1150/ 1236] Overall Loss 0.216342 Objective Loss 0.216342 LR 0.000250 Time 0.021096 +2023-10-05 21:50:46,304 - Epoch: [145][ 1160/ 1236] Overall Loss 0.216050 Objective Loss 0.216050 LR 0.000250 Time 0.021090 +2023-10-05 21:50:46,506 - Epoch: [145][ 1170/ 1236] Overall Loss 0.215902 Objective Loss 0.215902 LR 0.000250 Time 0.021082 +2023-10-05 21:50:46,711 - Epoch: [145][ 1180/ 1236] Overall Loss 0.215962 Objective Loss 0.215962 LR 0.000250 Time 0.021076 +2023-10-05 21:50:46,913 - Epoch: [145][ 1190/ 1236] Overall Loss 0.215945 Objective Loss 0.215945 LR 0.000250 Time 0.021068 +2023-10-05 21:50:47,116 - Epoch: [145][ 1200/ 1236] Overall Loss 0.215796 Objective Loss 0.215796 LR 0.000250 Time 0.021062 +2023-10-05 21:50:47,319 - Epoch: [145][ 1210/ 1236] Overall Loss 0.215786 Objective Loss 0.215786 LR 0.000250 Time 0.021055 +2023-10-05 21:50:47,523 - Epoch: [145][ 1220/ 1236] Overall Loss 0.215821 Objective Loss 0.215821 LR 0.000250 Time 0.021049 +2023-10-05 21:50:47,781 - Epoch: [145][ 1230/ 1236] Overall Loss 0.215750 Objective Loss 0.215750 LR 0.000250 Time 0.021088 +2023-10-05 21:50:47,900 - Epoch: [145][ 1236/ 1236] Overall Loss 0.215857 Objective Loss 0.215857 Top1 87.780041 Top5 98.574338 LR 0.000250 Time 0.021081 +2023-10-05 21:50:48,033 - --- validate (epoch=145)----------- +2023-10-05 21:50:48,033 - 29943 samples (256 per mini-batch) +2023-10-05 21:50:48,492 - Epoch: [145][ 10/ 117] Loss 0.329344 Top1 83.945312 Top5 97.226562 +2023-10-05 21:50:48,642 - Epoch: [145][ 20/ 117] Loss 0.323777 Top1 83.750000 Top5 97.636719 +2023-10-05 21:50:48,790 - Epoch: [145][ 30/ 117] Loss 0.321446 Top1 83.945312 Top5 97.812500 +2023-10-05 21:50:48,939 - Epoch: [145][ 40/ 117] Loss 0.315074 Top1 84.003906 Top5 97.851562 +2023-10-05 21:50:49,086 - Epoch: [145][ 50/ 117] Loss 0.309557 Top1 84.351562 Top5 97.859375 +2023-10-05 21:50:49,236 - Epoch: [145][ 60/ 117] Loss 0.307506 Top1 84.472656 Top5 97.845052 +2023-10-05 21:50:49,384 - Epoch: [145][ 70/ 117] Loss 0.311511 Top1 84.308036 Top5 97.795759 +2023-10-05 21:50:49,532 - Epoch: [145][ 80/ 117] Loss 0.308188 Top1 84.536133 Top5 97.841797 +2023-10-05 21:50:49,679 - Epoch: [145][ 90/ 117] Loss 0.308741 Top1 84.461806 Top5 97.855903 +2023-10-05 21:50:49,828 - Epoch: [145][ 100/ 117] Loss 0.305824 Top1 84.617188 Top5 97.886719 +2023-10-05 21:50:49,982 - Epoch: [145][ 110/ 117] Loss 0.308323 Top1 84.751420 Top5 97.897727 +2023-10-05 21:50:50,067 - Epoch: [145][ 117/ 117] Loss 0.309746 Top1 84.674214 Top5 97.899342 +2023-10-05 21:50:50,202 - ==> Top1: 84.674 Top5: 97.899 Loss: 0.310 + +2023-10-05 21:50:50,202 - ==> Confusion: +[[ 933 2 6 2 4 1 0 0 6 68 2 1 1 1 7 3 3 0 0 0 10] + [ 0 1063 2 0 8 18 1 17 0 0 1 3 0 0 0 2 4 0 8 1 3] + [ 5 2 972 12 2 0 19 8 0 0 3 5 7 1 0 2 0 3 5 5 5] + [ 6 1 13 976 0 1 1 2 1 0 3 1 5 1 31 3 0 9 21 2 12] + [ 20 8 3 0 972 6 0 0 0 7 0 2 0 2 11 4 8 2 0 2 3] + [ 4 34 0 1 2 993 1 22 1 2 5 8 0 15 5 2 2 0 3 2 14] + [ 0 9 29 0 0 0 1123 4 0 0 1 1 3 0 1 9 1 0 0 6 4] + [ 2 16 13 0 2 30 7 1073 2 3 3 14 0 2 0 5 0 0 34 3 9] + [ 17 5 1 0 0 3 0 0 968 41 10 1 1 9 20 6 0 1 4 0 2] + [ 88 0 2 0 5 3 0 1 26 949 1 1 0 15 12 10 0 0 1 0 5] + [ 2 7 10 4 1 0 3 2 5 2 981 2 0 9 5 1 1 1 6 1 10] + [ 1 0 1 0 0 14 0 3 0 1 0 965 13 4 0 4 2 19 0 6 2] + [ 1 2 4 5 0 1 0 1 1 0 1 39 983 3 1 5 2 13 1 4 1] + [ 2 0 1 0 2 7 0 0 12 10 7 4 3 1060 3 1 1 1 0 0 5] + [ 10 2 3 9 5 0 0 0 22 1 2 2 2 3 1015 0 0 2 8 0 15] + [ 0 3 1 0 4 0 0 0 0 1 0 7 6 3 1 1067 16 13 1 8 3] + [ 1 13 2 0 4 2 0 1 2 0 0 4 1 1 2 9 1103 0 1 3 12] + [ 0 0 0 2 1 0 1 0 0 0 0 2 12 0 0 3 0 1013 1 1 2] + [ 0 4 6 20 1 1 0 23 1 1 3 0 1 0 6 0 2 0 991 2 6] + [ 0 2 2 2 2 8 12 11 1 0 1 17 3 1 0 6 8 2 1 1070 3] + [ 120 195 157 67 93 150 44 94 111 61 181 128 343 282 175 70 120 83 167 180 5084]] + +2023-10-05 21:50:50,204 - ==> Best [Top1: 84.764 Top5: 97.953 Sparsity:0.00 Params: 148928 on epoch: 141] +2023-10-05 21:50:50,204 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:50:50,210 - + +2023-10-05 21:50:50,210 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:50:51,196 - Epoch: [146][ 10/ 1236] Overall Loss 0.232004 Objective Loss 0.232004 LR 0.000250 Time 0.098540 +2023-10-05 21:50:51,396 - Epoch: [146][ 20/ 1236] Overall Loss 0.222509 Objective Loss 0.222509 LR 0.000250 Time 0.059244 +2023-10-05 21:50:51,594 - Epoch: [146][ 30/ 1236] Overall Loss 0.226228 Objective Loss 0.226228 LR 0.000250 Time 0.046115 +2023-10-05 21:50:51,794 - Epoch: [146][ 40/ 1236] Overall Loss 0.225979 Objective Loss 0.225979 LR 0.000250 Time 0.039579 +2023-10-05 21:50:51,993 - Epoch: [146][ 50/ 1236] Overall Loss 0.221411 Objective Loss 0.221411 LR 0.000250 Time 0.035625 +2023-10-05 21:50:52,193 - Epoch: [146][ 60/ 1236] Overall Loss 0.221732 Objective Loss 0.221732 LR 0.000250 Time 0.033020 +2023-10-05 21:50:52,392 - Epoch: [146][ 70/ 1236] Overall Loss 0.219848 Objective Loss 0.219848 LR 0.000250 Time 0.031137 +2023-10-05 21:50:52,592 - Epoch: [146][ 80/ 1236] Overall Loss 0.217988 Objective Loss 0.217988 LR 0.000250 Time 0.029744 +2023-10-05 21:50:52,791 - Epoch: [146][ 90/ 1236] Overall Loss 0.217158 Objective Loss 0.217158 LR 0.000250 Time 0.028642 +2023-10-05 21:50:52,991 - Epoch: [146][ 100/ 1236] Overall Loss 0.215884 Objective Loss 0.215884 LR 0.000250 Time 0.027779 +2023-10-05 21:50:53,190 - Epoch: [146][ 110/ 1236] Overall Loss 0.219338 Objective Loss 0.219338 LR 0.000250 Time 0.027057 +2023-10-05 21:50:53,388 - Epoch: [146][ 120/ 1236] Overall Loss 0.218004 Objective Loss 0.218004 LR 0.000250 Time 0.026450 +2023-10-05 21:50:53,587 - Epoch: [146][ 130/ 1236] Overall Loss 0.216246 Objective Loss 0.216246 LR 0.000250 Time 0.025943 +2023-10-05 21:50:53,787 - Epoch: [146][ 140/ 1236] Overall Loss 0.215660 Objective Loss 0.215660 LR 0.000250 Time 0.025517 +2023-10-05 21:50:53,986 - Epoch: [146][ 150/ 1236] Overall Loss 0.215452 Objective Loss 0.215452 LR 0.000250 Time 0.025137 +2023-10-05 21:50:54,186 - Epoch: [146][ 160/ 1236] Overall Loss 0.215436 Objective Loss 0.215436 LR 0.000250 Time 0.024816 +2023-10-05 21:50:54,384 - Epoch: [146][ 170/ 1236] Overall Loss 0.215318 Objective Loss 0.215318 LR 0.000250 Time 0.024521 +2023-10-05 21:50:54,585 - Epoch: [146][ 180/ 1236] Overall Loss 0.215377 Objective Loss 0.215377 LR 0.000250 Time 0.024269 +2023-10-05 21:50:54,783 - Epoch: [146][ 190/ 1236] Overall Loss 0.215927 Objective Loss 0.215927 LR 0.000250 Time 0.024036 +2023-10-05 21:50:54,984 - Epoch: [146][ 200/ 1236] Overall Loss 0.215831 Objective Loss 0.215831 LR 0.000250 Time 0.023833 +2023-10-05 21:50:55,183 - Epoch: [146][ 210/ 1236] Overall Loss 0.216699 Objective Loss 0.216699 LR 0.000250 Time 0.023645 +2023-10-05 21:50:55,383 - Epoch: [146][ 220/ 1236] Overall Loss 0.217766 Objective Loss 0.217766 LR 0.000250 Time 0.023477 +2023-10-05 21:50:55,582 - Epoch: [146][ 230/ 1236] Overall Loss 0.216667 Objective Loss 0.216667 LR 0.000250 Time 0.023319 +2023-10-05 21:50:55,782 - Epoch: [146][ 240/ 1236] Overall Loss 0.216878 Objective Loss 0.216878 LR 0.000250 Time 0.023180 +2023-10-05 21:50:55,981 - Epoch: [146][ 250/ 1236] Overall Loss 0.216491 Objective Loss 0.216491 LR 0.000250 Time 0.023047 +2023-10-05 21:50:56,181 - Epoch: [146][ 260/ 1236] Overall Loss 0.216383 Objective Loss 0.216383 LR 0.000250 Time 0.022930 +2023-10-05 21:50:56,380 - Epoch: [146][ 270/ 1236] Overall Loss 0.216006 Objective Loss 0.216006 LR 0.000250 Time 0.022816 +2023-10-05 21:50:56,575 - Epoch: [146][ 280/ 1236] Overall Loss 0.215642 Objective Loss 0.215642 LR 0.000250 Time 0.022698 +2023-10-05 21:50:56,771 - Epoch: [146][ 290/ 1236] Overall Loss 0.215229 Objective Loss 0.215229 LR 0.000250 Time 0.022588 +2023-10-05 21:50:56,966 - Epoch: [146][ 300/ 1236] Overall Loss 0.215366 Objective Loss 0.215366 LR 0.000250 Time 0.022486 +2023-10-05 21:50:57,162 - Epoch: [146][ 310/ 1236] Overall Loss 0.214936 Objective Loss 0.214936 LR 0.000250 Time 0.022390 +2023-10-05 21:50:57,357 - Epoch: [146][ 320/ 1236] Overall Loss 0.214866 Objective Loss 0.214866 LR 0.000250 Time 0.022300 +2023-10-05 21:50:57,552 - Epoch: [146][ 330/ 1236] Overall Loss 0.214480 Objective Loss 0.214480 LR 0.000250 Time 0.022215 +2023-10-05 21:50:57,748 - Epoch: [146][ 340/ 1236] Overall Loss 0.214823 Objective Loss 0.214823 LR 0.000250 Time 0.022136 +2023-10-05 21:50:57,943 - Epoch: [146][ 350/ 1236] Overall Loss 0.215070 Objective Loss 0.215070 LR 0.000250 Time 0.022060 +2023-10-05 21:50:58,138 - Epoch: [146][ 360/ 1236] Overall Loss 0.214484 Objective Loss 0.214484 LR 0.000250 Time 0.021989 +2023-10-05 21:50:58,334 - Epoch: [146][ 370/ 1236] Overall Loss 0.214790 Objective Loss 0.214790 LR 0.000250 Time 0.021923 +2023-10-05 21:50:58,529 - Epoch: [146][ 380/ 1236] Overall Loss 0.214742 Objective Loss 0.214742 LR 0.000250 Time 0.021859 +2023-10-05 21:50:58,725 - Epoch: [146][ 390/ 1236] Overall Loss 0.214398 Objective Loss 0.214398 LR 0.000250 Time 0.021799 +2023-10-05 21:50:58,920 - Epoch: [146][ 400/ 1236] Overall Loss 0.214419 Objective Loss 0.214419 LR 0.000250 Time 0.021742 +2023-10-05 21:50:59,116 - Epoch: [146][ 410/ 1236] Overall Loss 0.214354 Objective Loss 0.214354 LR 0.000250 Time 0.021688 +2023-10-05 21:50:59,311 - Epoch: [146][ 420/ 1236] Overall Loss 0.214604 Objective Loss 0.214604 LR 0.000250 Time 0.021637 +2023-10-05 21:50:59,506 - Epoch: [146][ 430/ 1236] Overall Loss 0.214430 Objective Loss 0.214430 LR 0.000250 Time 0.021587 +2023-10-05 21:50:59,702 - Epoch: [146][ 440/ 1236] Overall Loss 0.214685 Objective Loss 0.214685 LR 0.000250 Time 0.021540 +2023-10-05 21:50:59,897 - Epoch: [146][ 450/ 1236] Overall Loss 0.214701 Objective Loss 0.214701 LR 0.000250 Time 0.021495 +2023-10-05 21:51:00,092 - Epoch: [146][ 460/ 1236] Overall Loss 0.215144 Objective Loss 0.215144 LR 0.000250 Time 0.021452 +2023-10-05 21:51:00,288 - Epoch: [146][ 470/ 1236] Overall Loss 0.215248 Objective Loss 0.215248 LR 0.000250 Time 0.021411 +2023-10-05 21:51:00,484 - Epoch: [146][ 480/ 1236] Overall Loss 0.215873 Objective Loss 0.215873 LR 0.000250 Time 0.021372 +2023-10-05 21:51:00,679 - Epoch: [146][ 490/ 1236] Overall Loss 0.215833 Objective Loss 0.215833 LR 0.000250 Time 0.021334 +2023-10-05 21:51:00,874 - Epoch: [146][ 500/ 1236] Overall Loss 0.215832 Objective Loss 0.215832 LR 0.000250 Time 0.021297 +2023-10-05 21:51:01,070 - Epoch: [146][ 510/ 1236] Overall Loss 0.216024 Objective Loss 0.216024 LR 0.000250 Time 0.021262 +2023-10-05 21:51:01,265 - Epoch: [146][ 520/ 1236] Overall Loss 0.215775 Objective Loss 0.215775 LR 0.000250 Time 0.021229 +2023-10-05 21:51:01,461 - Epoch: [146][ 530/ 1236] Overall Loss 0.216061 Objective Loss 0.216061 LR 0.000250 Time 0.021197 +2023-10-05 21:51:01,656 - Epoch: [146][ 540/ 1236] Overall Loss 0.216113 Objective Loss 0.216113 LR 0.000250 Time 0.021166 +2023-10-05 21:51:01,851 - Epoch: [146][ 550/ 1236] Overall Loss 0.216345 Objective Loss 0.216345 LR 0.000250 Time 0.021135 +2023-10-05 21:51:02,047 - Epoch: [146][ 560/ 1236] Overall Loss 0.216037 Objective Loss 0.216037 LR 0.000250 Time 0.021107 +2023-10-05 21:51:02,243 - Epoch: [146][ 570/ 1236] Overall Loss 0.216381 Objective Loss 0.216381 LR 0.000250 Time 0.021079 +2023-10-05 21:51:02,438 - Epoch: [146][ 580/ 1236] Overall Loss 0.216443 Objective Loss 0.216443 LR 0.000250 Time 0.021053 +2023-10-05 21:51:02,634 - Epoch: [146][ 590/ 1236] Overall Loss 0.216766 Objective Loss 0.216766 LR 0.000250 Time 0.021028 +2023-10-05 21:51:02,830 - Epoch: [146][ 600/ 1236] Overall Loss 0.217008 Objective Loss 0.217008 LR 0.000250 Time 0.021003 +2023-10-05 21:51:03,026 - Epoch: [146][ 610/ 1236] Overall Loss 0.217001 Objective Loss 0.217001 LR 0.000250 Time 0.020979 +2023-10-05 21:51:03,221 - Epoch: [146][ 620/ 1236] Overall Loss 0.217117 Objective Loss 0.217117 LR 0.000250 Time 0.020955 +2023-10-05 21:51:03,417 - Epoch: [146][ 630/ 1236] Overall Loss 0.217316 Objective Loss 0.217316 LR 0.000250 Time 0.020930 +2023-10-05 21:51:03,612 - Epoch: [146][ 640/ 1236] Overall Loss 0.217247 Objective Loss 0.217247 LR 0.000250 Time 0.020908 +2023-10-05 21:51:03,808 - Epoch: [146][ 650/ 1236] Overall Loss 0.217536 Objective Loss 0.217536 LR 0.000250 Time 0.020887 +2023-10-05 21:51:04,003 - Epoch: [146][ 660/ 1236] Overall Loss 0.217670 Objective Loss 0.217670 LR 0.000250 Time 0.020866 +2023-10-05 21:51:04,199 - Epoch: [146][ 670/ 1236] Overall Loss 0.217539 Objective Loss 0.217539 LR 0.000250 Time 0.020846 +2023-10-05 21:51:04,394 - Epoch: [146][ 680/ 1236] Overall Loss 0.217715 Objective Loss 0.217715 LR 0.000250 Time 0.020827 +2023-10-05 21:51:04,590 - Epoch: [146][ 690/ 1236] Overall Loss 0.217409 Objective Loss 0.217409 LR 0.000250 Time 0.020808 +2023-10-05 21:51:04,785 - Epoch: [146][ 700/ 1236] Overall Loss 0.217158 Objective Loss 0.217158 LR 0.000250 Time 0.020789 +2023-10-05 21:51:04,981 - Epoch: [146][ 710/ 1236] Overall Loss 0.217068 Objective Loss 0.217068 LR 0.000250 Time 0.020771 +2023-10-05 21:51:05,177 - Epoch: [146][ 720/ 1236] Overall Loss 0.216974 Objective Loss 0.216974 LR 0.000250 Time 0.020754 +2023-10-05 21:51:05,372 - Epoch: [146][ 730/ 1236] Overall Loss 0.216976 Objective Loss 0.216976 LR 0.000250 Time 0.020737 +2023-10-05 21:51:05,567 - Epoch: [146][ 740/ 1236] Overall Loss 0.217055 Objective Loss 0.217055 LR 0.000250 Time 0.020720 +2023-10-05 21:51:05,762 - Epoch: [146][ 750/ 1236] Overall Loss 0.217114 Objective Loss 0.217114 LR 0.000250 Time 0.020704 +2023-10-05 21:51:05,957 - Epoch: [146][ 760/ 1236] Overall Loss 0.217021 Objective Loss 0.217021 LR 0.000250 Time 0.020687 +2023-10-05 21:51:06,152 - Epoch: [146][ 770/ 1236] Overall Loss 0.216884 Objective Loss 0.216884 LR 0.000250 Time 0.020672 +2023-10-05 21:51:06,347 - Epoch: [146][ 780/ 1236] Overall Loss 0.216675 Objective Loss 0.216675 LR 0.000250 Time 0.020656 +2023-10-05 21:51:06,542 - Epoch: [146][ 790/ 1236] Overall Loss 0.216480 Objective Loss 0.216480 LR 0.000250 Time 0.020641 +2023-10-05 21:51:06,737 - Epoch: [146][ 800/ 1236] Overall Loss 0.216143 Objective Loss 0.216143 LR 0.000250 Time 0.020627 +2023-10-05 21:51:06,933 - Epoch: [146][ 810/ 1236] Overall Loss 0.216293 Objective Loss 0.216293 LR 0.000250 Time 0.020613 +2023-10-05 21:51:07,128 - Epoch: [146][ 820/ 1236] Overall Loss 0.216167 Objective Loss 0.216167 LR 0.000250 Time 0.020599 +2023-10-05 21:51:07,323 - Epoch: [146][ 830/ 1236] Overall Loss 0.216175 Objective Loss 0.216175 LR 0.000250 Time 0.020586 +2023-10-05 21:51:07,518 - Epoch: [146][ 840/ 1236] Overall Loss 0.215966 Objective Loss 0.215966 LR 0.000250 Time 0.020573 +2023-10-05 21:51:07,714 - Epoch: [146][ 850/ 1236] Overall Loss 0.216333 Objective Loss 0.216333 LR 0.000250 Time 0.020560 +2023-10-05 21:51:07,909 - Epoch: [146][ 860/ 1236] Overall Loss 0.215987 Objective Loss 0.215987 LR 0.000250 Time 0.020548 +2023-10-05 21:51:08,104 - Epoch: [146][ 870/ 1236] Overall Loss 0.216321 Objective Loss 0.216321 LR 0.000250 Time 0.020536 +2023-10-05 21:51:08,299 - Epoch: [146][ 880/ 1236] Overall Loss 0.216273 Objective Loss 0.216273 LR 0.000250 Time 0.020524 +2023-10-05 21:51:08,495 - Epoch: [146][ 890/ 1236] Overall Loss 0.216323 Objective Loss 0.216323 LR 0.000250 Time 0.020513 +2023-10-05 21:51:08,691 - Epoch: [146][ 900/ 1236] Overall Loss 0.216186 Objective Loss 0.216186 LR 0.000250 Time 0.020502 +2023-10-05 21:51:08,885 - Epoch: [146][ 910/ 1236] Overall Loss 0.216146 Objective Loss 0.216146 LR 0.000250 Time 0.020491 +2023-10-05 21:51:09,080 - Epoch: [146][ 920/ 1236] Overall Loss 0.216116 Objective Loss 0.216116 LR 0.000250 Time 0.020479 +2023-10-05 21:51:09,275 - Epoch: [146][ 930/ 1236] Overall Loss 0.216371 Objective Loss 0.216371 LR 0.000250 Time 0.020469 +2023-10-05 21:51:09,470 - Epoch: [146][ 940/ 1236] Overall Loss 0.216428 Objective Loss 0.216428 LR 0.000250 Time 0.020458 +2023-10-05 21:51:09,665 - Epoch: [146][ 950/ 1236] Overall Loss 0.216399 Objective Loss 0.216399 LR 0.000250 Time 0.020448 +2023-10-05 21:51:09,860 - Epoch: [146][ 960/ 1236] Overall Loss 0.216328 Objective Loss 0.216328 LR 0.000250 Time 0.020438 +2023-10-05 21:51:10,056 - Epoch: [146][ 970/ 1236] Overall Loss 0.216361 Objective Loss 0.216361 LR 0.000250 Time 0.020429 +2023-10-05 21:51:10,251 - Epoch: [146][ 980/ 1236] Overall Loss 0.216512 Objective Loss 0.216512 LR 0.000250 Time 0.020419 +2023-10-05 21:51:10,446 - Epoch: [146][ 990/ 1236] Overall Loss 0.216637 Objective Loss 0.216637 LR 0.000250 Time 0.020409 +2023-10-05 21:51:10,641 - Epoch: [146][ 1000/ 1236] Overall Loss 0.216735 Objective Loss 0.216735 LR 0.000250 Time 0.020400 +2023-10-05 21:51:10,836 - Epoch: [146][ 1010/ 1236] Overall Loss 0.216681 Objective Loss 0.216681 LR 0.000250 Time 0.020391 +2023-10-05 21:51:11,032 - Epoch: [146][ 1020/ 1236] Overall Loss 0.216574 Objective Loss 0.216574 LR 0.000250 Time 0.020382 +2023-10-05 21:51:11,227 - Epoch: [146][ 1030/ 1236] Overall Loss 0.216578 Objective Loss 0.216578 LR 0.000250 Time 0.020374 +2023-10-05 21:51:11,422 - Epoch: [146][ 1040/ 1236] Overall Loss 0.216530 Objective Loss 0.216530 LR 0.000250 Time 0.020365 +2023-10-05 21:51:11,618 - Epoch: [146][ 1050/ 1236] Overall Loss 0.216260 Objective Loss 0.216260 LR 0.000250 Time 0.020357 +2023-10-05 21:51:11,813 - Epoch: [146][ 1060/ 1236] Overall Loss 0.216222 Objective Loss 0.216222 LR 0.000250 Time 0.020349 +2023-10-05 21:51:12,008 - Epoch: [146][ 1070/ 1236] Overall Loss 0.216395 Objective Loss 0.216395 LR 0.000250 Time 0.020341 +2023-10-05 21:51:12,203 - Epoch: [146][ 1080/ 1236] Overall Loss 0.216231 Objective Loss 0.216231 LR 0.000250 Time 0.020333 +2023-10-05 21:51:12,399 - Epoch: [146][ 1090/ 1236] Overall Loss 0.216282 Objective Loss 0.216282 LR 0.000250 Time 0.020326 +2023-10-05 21:51:12,594 - Epoch: [146][ 1100/ 1236] Overall Loss 0.216221 Objective Loss 0.216221 LR 0.000250 Time 0.020318 +2023-10-05 21:51:12,789 - Epoch: [146][ 1110/ 1236] Overall Loss 0.216154 Objective Loss 0.216154 LR 0.000250 Time 0.020311 +2023-10-05 21:51:12,984 - Epoch: [146][ 1120/ 1236] Overall Loss 0.216156 Objective Loss 0.216156 LR 0.000250 Time 0.020303 +2023-10-05 21:51:13,180 - Epoch: [146][ 1130/ 1236] Overall Loss 0.216316 Objective Loss 0.216316 LR 0.000250 Time 0.020297 +2023-10-05 21:51:13,375 - Epoch: [146][ 1140/ 1236] Overall Loss 0.216149 Objective Loss 0.216149 LR 0.000250 Time 0.020290 +2023-10-05 21:51:13,571 - Epoch: [146][ 1150/ 1236] Overall Loss 0.216181 Objective Loss 0.216181 LR 0.000250 Time 0.020283 +2023-10-05 21:51:13,766 - Epoch: [146][ 1160/ 1236] Overall Loss 0.216075 Objective Loss 0.216075 LR 0.000250 Time 0.020276 +2023-10-05 21:51:13,961 - Epoch: [146][ 1170/ 1236] Overall Loss 0.216142 Objective Loss 0.216142 LR 0.000250 Time 0.020270 +2023-10-05 21:51:14,157 - Epoch: [146][ 1180/ 1236] Overall Loss 0.216187 Objective Loss 0.216187 LR 0.000250 Time 0.020263 +2023-10-05 21:51:14,352 - Epoch: [146][ 1190/ 1236] Overall Loss 0.216048 Objective Loss 0.216048 LR 0.000250 Time 0.020257 +2023-10-05 21:51:14,547 - Epoch: [146][ 1200/ 1236] Overall Loss 0.215966 Objective Loss 0.215966 LR 0.000250 Time 0.020251 +2023-10-05 21:51:14,742 - Epoch: [146][ 1210/ 1236] Overall Loss 0.215778 Objective Loss 0.215778 LR 0.000250 Time 0.020244 +2023-10-05 21:51:14,937 - Epoch: [146][ 1220/ 1236] Overall Loss 0.215722 Objective Loss 0.215722 LR 0.000250 Time 0.020238 +2023-10-05 21:51:15,185 - Epoch: [146][ 1230/ 1236] Overall Loss 0.215673 Objective Loss 0.215673 LR 0.000250 Time 0.020275 +2023-10-05 21:51:15,302 - Epoch: [146][ 1236/ 1236] Overall Loss 0.215832 Objective Loss 0.215832 Top1 85.539715 Top5 98.778004 LR 0.000250 Time 0.020271 +2023-10-05 21:51:15,428 - --- validate (epoch=146)----------- +2023-10-05 21:51:15,428 - 29943 samples (256 per mini-batch) +2023-10-05 21:51:15,878 - Epoch: [146][ 10/ 117] Loss 0.299125 Top1 84.882812 Top5 98.242188 +2023-10-05 21:51:16,026 - Epoch: [146][ 20/ 117] Loss 0.316557 Top1 84.726562 Top5 97.812500 +2023-10-05 21:51:16,173 - Epoch: [146][ 30/ 117] Loss 0.311628 Top1 84.934896 Top5 97.929688 +2023-10-05 21:51:16,322 - Epoch: [146][ 40/ 117] Loss 0.313132 Top1 84.980469 Top5 97.900391 +2023-10-05 21:51:16,472 - Epoch: [146][ 50/ 117] Loss 0.313547 Top1 84.937500 Top5 97.921875 +2023-10-05 21:51:16,620 - Epoch: [146][ 60/ 117] Loss 0.311681 Top1 85.013021 Top5 97.910156 +2023-10-05 21:51:16,767 - Epoch: [146][ 70/ 117] Loss 0.310932 Top1 85.044643 Top5 97.940848 +2023-10-05 21:51:16,915 - Epoch: [146][ 80/ 117] Loss 0.307727 Top1 85.029297 Top5 97.939453 +2023-10-05 21:51:17,063 - Epoch: [146][ 90/ 117] Loss 0.305295 Top1 85.065104 Top5 97.925347 +2023-10-05 21:51:17,212 - Epoch: [146][ 100/ 117] Loss 0.307512 Top1 84.976562 Top5 97.910156 +2023-10-05 21:51:17,365 - Epoch: [146][ 110/ 117] Loss 0.308146 Top1 84.960938 Top5 97.894176 +2023-10-05 21:51:17,450 - Epoch: [146][ 117/ 117] Loss 0.307744 Top1 84.994823 Top5 97.885983 +2023-10-05 21:51:17,597 - ==> Top1: 84.995 Top5: 97.886 Loss: 0.308 + +2023-10-05 21:51:17,597 - ==> Confusion: +[[ 937 3 2 0 5 2 0 1 7 64 4 0 1 0 8 2 3 0 0 0 11] + [ 0 1067 2 0 8 15 1 22 2 0 1 0 0 0 0 5 0 0 5 0 3] + [ 3 2 958 14 2 0 23 11 0 2 7 4 7 1 1 3 0 1 5 2 10] + [ 6 0 9 967 0 3 1 1 2 0 10 1 5 1 24 4 0 6 33 1 15] + [ 22 5 2 0 979 5 0 1 0 9 1 1 1 0 6 5 7 1 0 3 2] + [ 4 37 0 0 2 992 1 21 0 2 6 8 1 9 6 1 3 1 4 3 15] + [ 0 7 19 0 0 0 1130 4 0 0 1 2 1 0 1 8 0 0 4 6 8] + [ 1 14 10 0 2 24 4 1086 1 3 5 14 2 1 0 4 0 0 35 5 7] + [ 19 3 1 0 0 2 0 0 971 47 11 5 2 7 11 3 1 0 5 0 1] + [ 87 1 2 0 4 2 0 1 22 964 1 3 0 12 3 5 0 1 0 1 10] + [ 3 5 9 3 1 1 3 4 11 1 980 3 0 11 2 1 2 0 4 1 8] + [ 1 0 0 0 0 12 0 2 0 1 0 974 15 5 0 2 1 15 0 3 4] + [ 3 1 0 5 0 2 0 0 1 0 2 41 978 1 1 4 2 13 1 1 12] + [ 3 1 0 0 4 9 0 1 15 15 11 6 1 1040 3 1 0 1 0 0 8] + [ 11 1 3 8 5 0 0 0 24 3 1 0 1 3 1009 0 0 3 14 0 15] + [ 0 5 2 0 2 0 2 0 0 1 0 6 7 2 1 1069 13 12 0 8 4] + [ 1 12 2 0 7 2 0 1 1 0 0 6 0 1 2 9 1101 0 1 3 12] + [ 0 0 0 3 1 1 2 0 0 0 0 3 14 1 1 3 0 1004 1 0 4] + [ 2 5 5 15 1 0 0 20 1 0 5 0 0 0 8 0 1 0 997 1 7] + [ 0 3 2 1 3 5 13 9 0 0 1 19 4 3 0 4 9 2 3 1062 9] + [ 127 190 149 56 122 127 35 100 122 91 198 105 307 245 161 55 124 80 164 162 5185]] + +2023-10-05 21:51:17,599 - ==> Best [Top1: 84.995 Top5: 97.886 Sparsity:0.00 Params: 148928 on epoch: 146] +2023-10-05 21:51:17,599 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:51:17,612 - + +2023-10-05 21:51:17,612 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:51:18,591 - Epoch: [147][ 10/ 1236] Overall Loss 0.215402 Objective Loss 0.215402 LR 0.000250 Time 0.097876 +2023-10-05 21:51:18,789 - Epoch: [147][ 20/ 1236] Overall Loss 0.214231 Objective Loss 0.214231 LR 0.000250 Time 0.058821 +2023-10-05 21:51:18,988 - Epoch: [147][ 30/ 1236] Overall Loss 0.213950 Objective Loss 0.213950 LR 0.000250 Time 0.045815 +2023-10-05 21:51:19,188 - Epoch: [147][ 40/ 1236] Overall Loss 0.216303 Objective Loss 0.216303 LR 0.000250 Time 0.039351 +2023-10-05 21:51:19,386 - Epoch: [147][ 50/ 1236] Overall Loss 0.215289 Objective Loss 0.215289 LR 0.000250 Time 0.035436 +2023-10-05 21:51:19,586 - Epoch: [147][ 60/ 1236] Overall Loss 0.214710 Objective Loss 0.214710 LR 0.000250 Time 0.032857 +2023-10-05 21:51:19,784 - Epoch: [147][ 70/ 1236] Overall Loss 0.211427 Objective Loss 0.211427 LR 0.000250 Time 0.030988 +2023-10-05 21:51:19,984 - Epoch: [147][ 80/ 1236] Overall Loss 0.211284 Objective Loss 0.211284 LR 0.000250 Time 0.029609 +2023-10-05 21:51:20,182 - Epoch: [147][ 90/ 1236] Overall Loss 0.213588 Objective Loss 0.213588 LR 0.000250 Time 0.028516 +2023-10-05 21:51:20,382 - Epoch: [147][ 100/ 1236] Overall Loss 0.211102 Objective Loss 0.211102 LR 0.000250 Time 0.027660 +2023-10-05 21:51:20,580 - Epoch: [147][ 110/ 1236] Overall Loss 0.207856 Objective Loss 0.207856 LR 0.000250 Time 0.026944 +2023-10-05 21:51:20,780 - Epoch: [147][ 120/ 1236] Overall Loss 0.208797 Objective Loss 0.208797 LR 0.000250 Time 0.026363 +2023-10-05 21:51:20,978 - Epoch: [147][ 130/ 1236] Overall Loss 0.207888 Objective Loss 0.207888 LR 0.000250 Time 0.025859 +2023-10-05 21:51:21,179 - Epoch: [147][ 140/ 1236] Overall Loss 0.208344 Objective Loss 0.208344 LR 0.000250 Time 0.025439 +2023-10-05 21:51:21,377 - Epoch: [147][ 150/ 1236] Overall Loss 0.208323 Objective Loss 0.208323 LR 0.000250 Time 0.025062 +2023-10-05 21:51:21,577 - Epoch: [147][ 160/ 1236] Overall Loss 0.208263 Objective Loss 0.208263 LR 0.000250 Time 0.024744 +2023-10-05 21:51:21,777 - Epoch: [147][ 170/ 1236] Overall Loss 0.207660 Objective Loss 0.207660 LR 0.000250 Time 0.024466 +2023-10-05 21:51:21,977 - Epoch: [147][ 180/ 1236] Overall Loss 0.209735 Objective Loss 0.209735 LR 0.000250 Time 0.024212 +2023-10-05 21:51:22,179 - Epoch: [147][ 190/ 1236] Overall Loss 0.208795 Objective Loss 0.208795 LR 0.000250 Time 0.024000 +2023-10-05 21:51:22,380 - Epoch: [147][ 200/ 1236] Overall Loss 0.208528 Objective Loss 0.208528 LR 0.000250 Time 0.023802 +2023-10-05 21:51:22,582 - Epoch: [147][ 210/ 1236] Overall Loss 0.208686 Objective Loss 0.208686 LR 0.000250 Time 0.023630 +2023-10-05 21:51:22,782 - Epoch: [147][ 220/ 1236] Overall Loss 0.209414 Objective Loss 0.209414 LR 0.000250 Time 0.023463 +2023-10-05 21:51:22,984 - Epoch: [147][ 230/ 1236] Overall Loss 0.209994 Objective Loss 0.209994 LR 0.000250 Time 0.023322 +2023-10-05 21:51:23,184 - Epoch: [147][ 240/ 1236] Overall Loss 0.208546 Objective Loss 0.208546 LR 0.000250 Time 0.023181 +2023-10-05 21:51:23,388 - Epoch: [147][ 250/ 1236] Overall Loss 0.209084 Objective Loss 0.209084 LR 0.000250 Time 0.023066 +2023-10-05 21:51:23,587 - Epoch: [147][ 260/ 1236] Overall Loss 0.209562 Objective Loss 0.209562 LR 0.000250 Time 0.022946 +2023-10-05 21:51:23,790 - Epoch: [147][ 270/ 1236] Overall Loss 0.209723 Objective Loss 0.209723 LR 0.000250 Time 0.022844 +2023-10-05 21:51:23,990 - Epoch: [147][ 280/ 1236] Overall Loss 0.209680 Objective Loss 0.209680 LR 0.000250 Time 0.022740 +2023-10-05 21:51:24,192 - Epoch: [147][ 290/ 1236] Overall Loss 0.209070 Objective Loss 0.209070 LR 0.000250 Time 0.022654 +2023-10-05 21:51:24,393 - Epoch: [147][ 300/ 1236] Overall Loss 0.208669 Objective Loss 0.208669 LR 0.000250 Time 0.022565 +2023-10-05 21:51:24,596 - Epoch: [147][ 310/ 1236] Overall Loss 0.208576 Objective Loss 0.208576 LR 0.000250 Time 0.022491 +2023-10-05 21:51:24,795 - Epoch: [147][ 320/ 1236] Overall Loss 0.208564 Objective Loss 0.208564 LR 0.000250 Time 0.022411 +2023-10-05 21:51:24,999 - Epoch: [147][ 330/ 1236] Overall Loss 0.208651 Objective Loss 0.208651 LR 0.000250 Time 0.022347 +2023-10-05 21:51:25,198 - Epoch: [147][ 340/ 1236] Overall Loss 0.209063 Objective Loss 0.209063 LR 0.000250 Time 0.022275 +2023-10-05 21:51:25,401 - Epoch: [147][ 350/ 1236] Overall Loss 0.209132 Objective Loss 0.209132 LR 0.000250 Time 0.022217 +2023-10-05 21:51:25,601 - Epoch: [147][ 360/ 1236] Overall Loss 0.210059 Objective Loss 0.210059 LR 0.000250 Time 0.022154 +2023-10-05 21:51:25,805 - Epoch: [147][ 370/ 1236] Overall Loss 0.209729 Objective Loss 0.209729 LR 0.000250 Time 0.022107 +2023-10-05 21:51:26,006 - Epoch: [147][ 380/ 1236] Overall Loss 0.209593 Objective Loss 0.209593 LR 0.000250 Time 0.022053 +2023-10-05 21:51:26,210 - Epoch: [147][ 390/ 1236] Overall Loss 0.209590 Objective Loss 0.209590 LR 0.000250 Time 0.022009 +2023-10-05 21:51:26,411 - Epoch: [147][ 400/ 1236] Overall Loss 0.209619 Objective Loss 0.209619 LR 0.000250 Time 0.021961 +2023-10-05 21:51:26,615 - Epoch: [147][ 410/ 1236] Overall Loss 0.209451 Objective Loss 0.209451 LR 0.000250 Time 0.021922 +2023-10-05 21:51:26,816 - Epoch: [147][ 420/ 1236] Overall Loss 0.209569 Objective Loss 0.209569 LR 0.000250 Time 0.021878 +2023-10-05 21:51:27,020 - Epoch: [147][ 430/ 1236] Overall Loss 0.209334 Objective Loss 0.209334 LR 0.000250 Time 0.021843 +2023-10-05 21:51:27,221 - Epoch: [147][ 440/ 1236] Overall Loss 0.209220 Objective Loss 0.209220 LR 0.000250 Time 0.021803 +2023-10-05 21:51:27,425 - Epoch: [147][ 450/ 1236] Overall Loss 0.208977 Objective Loss 0.208977 LR 0.000250 Time 0.021771 +2023-10-05 21:51:27,626 - Epoch: [147][ 460/ 1236] Overall Loss 0.209031 Objective Loss 0.209031 LR 0.000250 Time 0.021734 +2023-10-05 21:51:27,830 - Epoch: [147][ 470/ 1236] Overall Loss 0.209324 Objective Loss 0.209324 LR 0.000250 Time 0.021704 +2023-10-05 21:51:28,031 - Epoch: [147][ 480/ 1236] Overall Loss 0.209288 Objective Loss 0.209288 LR 0.000250 Time 0.021670 +2023-10-05 21:51:28,235 - Epoch: [147][ 490/ 1236] Overall Loss 0.209919 Objective Loss 0.209919 LR 0.000250 Time 0.021644 +2023-10-05 21:51:28,436 - Epoch: [147][ 500/ 1236] Overall Loss 0.210198 Objective Loss 0.210198 LR 0.000250 Time 0.021612 +2023-10-05 21:51:28,640 - Epoch: [147][ 510/ 1236] Overall Loss 0.210379 Objective Loss 0.210379 LR 0.000250 Time 0.021588 +2023-10-05 21:51:28,840 - Epoch: [147][ 520/ 1236] Overall Loss 0.209868 Objective Loss 0.209868 LR 0.000250 Time 0.021556 +2023-10-05 21:51:29,042 - Epoch: [147][ 530/ 1236] Overall Loss 0.209996 Objective Loss 0.209996 LR 0.000250 Time 0.021529 +2023-10-05 21:51:29,242 - Epoch: [147][ 540/ 1236] Overall Loss 0.209750 Objective Loss 0.209750 LR 0.000250 Time 0.021500 +2023-10-05 21:51:29,443 - Epoch: [147][ 550/ 1236] Overall Loss 0.209990 Objective Loss 0.209990 LR 0.000250 Time 0.021475 +2023-10-05 21:51:29,642 - Epoch: [147][ 560/ 1236] Overall Loss 0.209946 Objective Loss 0.209946 LR 0.000250 Time 0.021446 +2023-10-05 21:51:29,844 - Epoch: [147][ 570/ 1236] Overall Loss 0.210026 Objective Loss 0.210026 LR 0.000250 Time 0.021423 +2023-10-05 21:51:30,044 - Epoch: [147][ 580/ 1236] Overall Loss 0.210159 Objective Loss 0.210159 LR 0.000250 Time 0.021397 +2023-10-05 21:51:30,246 - Epoch: [147][ 590/ 1236] Overall Loss 0.210456 Objective Loss 0.210456 LR 0.000250 Time 0.021377 +2023-10-05 21:51:30,445 - Epoch: [147][ 600/ 1236] Overall Loss 0.210400 Objective Loss 0.210400 LR 0.000250 Time 0.021352 +2023-10-05 21:51:30,647 - Epoch: [147][ 610/ 1236] Overall Loss 0.210750 Objective Loss 0.210750 LR 0.000250 Time 0.021332 +2023-10-05 21:51:30,846 - Epoch: [147][ 620/ 1236] Overall Loss 0.211061 Objective Loss 0.211061 LR 0.000250 Time 0.021310 +2023-10-05 21:51:31,048 - Epoch: [147][ 630/ 1236] Overall Loss 0.211470 Objective Loss 0.211470 LR 0.000250 Time 0.021291 +2023-10-05 21:51:31,248 - Epoch: [147][ 640/ 1236] Overall Loss 0.211437 Objective Loss 0.211437 LR 0.000250 Time 0.021270 +2023-10-05 21:51:31,449 - Epoch: [147][ 650/ 1236] Overall Loss 0.211624 Objective Loss 0.211624 LR 0.000250 Time 0.021252 +2023-10-05 21:51:31,649 - Epoch: [147][ 660/ 1236] Overall Loss 0.211900 Objective Loss 0.211900 LR 0.000250 Time 0.021232 +2023-10-05 21:51:31,851 - Epoch: [147][ 670/ 1236] Overall Loss 0.211708 Objective Loss 0.211708 LR 0.000250 Time 0.021216 +2023-10-05 21:51:32,050 - Epoch: [147][ 680/ 1236] Overall Loss 0.211716 Objective Loss 0.211716 LR 0.000250 Time 0.021197 +2023-10-05 21:51:32,256 - Epoch: [147][ 690/ 1236] Overall Loss 0.212014 Objective Loss 0.212014 LR 0.000250 Time 0.021187 +2023-10-05 21:51:32,464 - Epoch: [147][ 700/ 1236] Overall Loss 0.212170 Objective Loss 0.212170 LR 0.000250 Time 0.021181 +2023-10-05 21:51:32,667 - Epoch: [147][ 710/ 1236] Overall Loss 0.212746 Objective Loss 0.212746 LR 0.000250 Time 0.021169 +2023-10-05 21:51:32,872 - Epoch: [147][ 720/ 1236] Overall Loss 0.212826 Objective Loss 0.212826 LR 0.000250 Time 0.021158 +2023-10-05 21:51:33,075 - Epoch: [147][ 730/ 1236] Overall Loss 0.213059 Objective Loss 0.213059 LR 0.000250 Time 0.021146 +2023-10-05 21:51:33,280 - Epoch: [147][ 740/ 1236] Overall Loss 0.212952 Objective Loss 0.212952 LR 0.000250 Time 0.021136 +2023-10-05 21:51:33,483 - Epoch: [147][ 750/ 1236] Overall Loss 0.212883 Objective Loss 0.212883 LR 0.000250 Time 0.021126 +2023-10-05 21:51:33,688 - Epoch: [147][ 760/ 1236] Overall Loss 0.213485 Objective Loss 0.213485 LR 0.000250 Time 0.021116 +2023-10-05 21:51:33,891 - Epoch: [147][ 770/ 1236] Overall Loss 0.213593 Objective Loss 0.213593 LR 0.000250 Time 0.021106 +2023-10-05 21:51:34,096 - Epoch: [147][ 780/ 1236] Overall Loss 0.213848 Objective Loss 0.213848 LR 0.000250 Time 0.021097 +2023-10-05 21:51:34,300 - Epoch: [147][ 790/ 1236] Overall Loss 0.213917 Objective Loss 0.213917 LR 0.000250 Time 0.021088 +2023-10-05 21:51:34,504 - Epoch: [147][ 800/ 1236] Overall Loss 0.213835 Objective Loss 0.213835 LR 0.000250 Time 0.021079 +2023-10-05 21:51:34,708 - Epoch: [147][ 810/ 1236] Overall Loss 0.214116 Objective Loss 0.214116 LR 0.000250 Time 0.021070 +2023-10-05 21:51:34,912 - Epoch: [147][ 820/ 1236] Overall Loss 0.214275 Objective Loss 0.214275 LR 0.000250 Time 0.021062 +2023-10-05 21:51:35,116 - Epoch: [147][ 830/ 1236] Overall Loss 0.214183 Objective Loss 0.214183 LR 0.000250 Time 0.021053 +2023-10-05 21:51:35,322 - Epoch: [147][ 840/ 1236] Overall Loss 0.214107 Objective Loss 0.214107 LR 0.000250 Time 0.021047 +2023-10-05 21:51:35,527 - Epoch: [147][ 850/ 1236] Overall Loss 0.214132 Objective Loss 0.214132 LR 0.000250 Time 0.021041 +2023-10-05 21:51:35,733 - Epoch: [147][ 860/ 1236] Overall Loss 0.214062 Objective Loss 0.214062 LR 0.000250 Time 0.021036 +2023-10-05 21:51:35,939 - Epoch: [147][ 870/ 1236] Overall Loss 0.214259 Objective Loss 0.214259 LR 0.000250 Time 0.021030 +2023-10-05 21:51:36,145 - Epoch: [147][ 880/ 1236] Overall Loss 0.214121 Objective Loss 0.214121 LR 0.000250 Time 0.021025 +2023-10-05 21:51:36,351 - Epoch: [147][ 890/ 1236] Overall Loss 0.214154 Objective Loss 0.214154 LR 0.000250 Time 0.021019 +2023-10-05 21:51:36,557 - Epoch: [147][ 900/ 1236] Overall Loss 0.214316 Objective Loss 0.214316 LR 0.000250 Time 0.021014 +2023-10-05 21:51:36,763 - Epoch: [147][ 910/ 1236] Overall Loss 0.214084 Objective Loss 0.214084 LR 0.000250 Time 0.021009 +2023-10-05 21:51:36,969 - Epoch: [147][ 920/ 1236] Overall Loss 0.213953 Objective Loss 0.213953 LR 0.000250 Time 0.021005 +2023-10-05 21:51:37,174 - Epoch: [147][ 930/ 1236] Overall Loss 0.213973 Objective Loss 0.213973 LR 0.000250 Time 0.021000 +2023-10-05 21:51:37,381 - Epoch: [147][ 940/ 1236] Overall Loss 0.213871 Objective Loss 0.213871 LR 0.000250 Time 0.020996 +2023-10-05 21:51:37,587 - Epoch: [147][ 950/ 1236] Overall Loss 0.213853 Objective Loss 0.213853 LR 0.000250 Time 0.020991 +2023-10-05 21:51:37,793 - Epoch: [147][ 960/ 1236] Overall Loss 0.213875 Objective Loss 0.213875 LR 0.000250 Time 0.020987 +2023-10-05 21:51:37,999 - Epoch: [147][ 970/ 1236] Overall Loss 0.213760 Objective Loss 0.213760 LR 0.000250 Time 0.020982 +2023-10-05 21:51:38,205 - Epoch: [147][ 980/ 1236] Overall Loss 0.213989 Objective Loss 0.213989 LR 0.000250 Time 0.020978 +2023-10-05 21:51:38,411 - Epoch: [147][ 990/ 1236] Overall Loss 0.213996 Objective Loss 0.213996 LR 0.000250 Time 0.020974 +2023-10-05 21:51:38,617 - Epoch: [147][ 1000/ 1236] Overall Loss 0.213831 Objective Loss 0.213831 LR 0.000250 Time 0.020970 +2023-10-05 21:51:38,823 - Epoch: [147][ 1010/ 1236] Overall Loss 0.213881 Objective Loss 0.213881 LR 0.000250 Time 0.020966 +2023-10-05 21:51:39,029 - Epoch: [147][ 1020/ 1236] Overall Loss 0.214075 Objective Loss 0.214075 LR 0.000250 Time 0.020962 +2023-10-05 21:51:39,235 - Epoch: [147][ 1030/ 1236] Overall Loss 0.214186 Objective Loss 0.214186 LR 0.000250 Time 0.020958 +2023-10-05 21:51:39,442 - Epoch: [147][ 1040/ 1236] Overall Loss 0.214040 Objective Loss 0.214040 LR 0.000250 Time 0.020955 +2023-10-05 21:51:39,647 - Epoch: [147][ 1050/ 1236] Overall Loss 0.213886 Objective Loss 0.213886 LR 0.000250 Time 0.020951 +2023-10-05 21:51:39,854 - Epoch: [147][ 1060/ 1236] Overall Loss 0.213693 Objective Loss 0.213693 LR 0.000250 Time 0.020948 +2023-10-05 21:51:40,059 - Epoch: [147][ 1070/ 1236] Overall Loss 0.213748 Objective Loss 0.213748 LR 0.000250 Time 0.020944 +2023-10-05 21:51:40,266 - Epoch: [147][ 1080/ 1236] Overall Loss 0.213835 Objective Loss 0.213835 LR 0.000250 Time 0.020941 +2023-10-05 21:51:40,471 - Epoch: [147][ 1090/ 1236] Overall Loss 0.213902 Objective Loss 0.213902 LR 0.000250 Time 0.020937 +2023-10-05 21:51:40,677 - Epoch: [147][ 1100/ 1236] Overall Loss 0.213959 Objective Loss 0.213959 LR 0.000250 Time 0.020934 +2023-10-05 21:51:40,883 - Epoch: [147][ 1110/ 1236] Overall Loss 0.214028 Objective Loss 0.214028 LR 0.000250 Time 0.020930 +2023-10-05 21:51:41,089 - Epoch: [147][ 1120/ 1236] Overall Loss 0.213909 Objective Loss 0.213909 LR 0.000250 Time 0.020927 +2023-10-05 21:51:41,295 - Epoch: [147][ 1130/ 1236] Overall Loss 0.213925 Objective Loss 0.213925 LR 0.000250 Time 0.020924 +2023-10-05 21:51:41,502 - Epoch: [147][ 1140/ 1236] Overall Loss 0.213896 Objective Loss 0.213896 LR 0.000250 Time 0.020921 +2023-10-05 21:51:41,707 - Epoch: [147][ 1150/ 1236] Overall Loss 0.213684 Objective Loss 0.213684 LR 0.000250 Time 0.020918 +2023-10-05 21:51:41,913 - Epoch: [147][ 1160/ 1236] Overall Loss 0.213625 Objective Loss 0.213625 LR 0.000250 Time 0.020915 +2023-10-05 21:51:42,119 - Epoch: [147][ 1170/ 1236] Overall Loss 0.213502 Objective Loss 0.213502 LR 0.000250 Time 0.020912 +2023-10-05 21:51:42,325 - Epoch: [147][ 1180/ 1236] Overall Loss 0.213354 Objective Loss 0.213354 LR 0.000250 Time 0.020909 +2023-10-05 21:51:42,531 - Epoch: [147][ 1190/ 1236] Overall Loss 0.213297 Objective Loss 0.213297 LR 0.000250 Time 0.020906 +2023-10-05 21:51:42,738 - Epoch: [147][ 1200/ 1236] Overall Loss 0.213326 Objective Loss 0.213326 LR 0.000250 Time 0.020904 +2023-10-05 21:51:42,944 - Epoch: [147][ 1210/ 1236] Overall Loss 0.213415 Objective Loss 0.213415 LR 0.000250 Time 0.020901 +2023-10-05 21:51:43,150 - Epoch: [147][ 1220/ 1236] Overall Loss 0.213647 Objective Loss 0.213647 LR 0.000250 Time 0.020898 +2023-10-05 21:51:43,409 - Epoch: [147][ 1230/ 1236] Overall Loss 0.213708 Objective Loss 0.213708 LR 0.000250 Time 0.020939 +2023-10-05 21:51:43,528 - Epoch: [147][ 1236/ 1236] Overall Loss 0.213788 Objective Loss 0.213788 Top1 89.613035 Top5 97.963340 LR 0.000250 Time 0.020933 +2023-10-05 21:51:43,662 - --- validate (epoch=147)----------- +2023-10-05 21:51:43,662 - 29943 samples (256 per mini-batch) +2023-10-05 21:51:44,123 - Epoch: [147][ 10/ 117] Loss 0.345762 Top1 84.101562 Top5 97.812500 +2023-10-05 21:51:44,272 - Epoch: [147][ 20/ 117] Loss 0.327291 Top1 84.296875 Top5 97.910156 +2023-10-05 21:51:44,418 - Epoch: [147][ 30/ 117] Loss 0.321985 Top1 84.440104 Top5 97.864583 +2023-10-05 21:51:44,565 - Epoch: [147][ 40/ 117] Loss 0.319816 Top1 84.501953 Top5 97.929688 +2023-10-05 21:51:44,711 - Epoch: [147][ 50/ 117] Loss 0.315086 Top1 84.828125 Top5 97.984375 +2023-10-05 21:51:44,858 - Epoch: [147][ 60/ 117] Loss 0.311649 Top1 84.850260 Top5 97.916667 +2023-10-05 21:51:45,003 - Epoch: [147][ 70/ 117] Loss 0.313982 Top1 84.737723 Top5 97.929688 +2023-10-05 21:51:45,150 - Epoch: [147][ 80/ 117] Loss 0.311168 Top1 84.746094 Top5 97.978516 +2023-10-05 21:51:45,297 - Epoch: [147][ 90/ 117] Loss 0.310044 Top1 84.782986 Top5 98.003472 +2023-10-05 21:51:45,443 - Epoch: [147][ 100/ 117] Loss 0.310264 Top1 84.746094 Top5 98.003906 +2023-10-05 21:51:45,598 - Epoch: [147][ 110/ 117] Loss 0.312723 Top1 84.680398 Top5 97.993608 +2023-10-05 21:51:45,684 - Epoch: [147][ 117/ 117] Loss 0.311783 Top1 84.684233 Top5 98.006212 +2023-10-05 21:51:45,801 - ==> Top1: 84.684 Top5: 98.006 Loss: 0.312 + +2023-10-05 21:51:45,802 - ==> Confusion: +[[ 926 4 2 4 8 1 0 0 4 73 2 1 0 1 4 1 2 2 0 0 15] + [ 0 1062 3 1 11 16 1 14 0 0 1 2 0 0 0 4 2 0 7 1 6] + [ 3 2 972 10 2 0 22 6 0 3 2 2 8 1 1 3 1 1 4 4 9] + [ 2 1 14 980 3 2 1 2 1 1 3 1 5 3 20 2 0 7 26 1 14] + [ 19 7 3 0 978 2 0 0 0 8 2 3 0 2 3 4 11 1 0 0 7] + [ 4 28 1 1 4 992 1 20 1 2 3 6 0 16 4 2 3 0 3 7 18] + [ 1 7 20 0 0 0 1131 4 0 0 1 3 3 0 1 7 0 0 1 5 7] + [ 3 17 11 0 2 28 7 1074 1 6 2 11 1 1 0 2 1 0 35 8 8] + [ 20 4 0 0 0 1 0 1 951 50 9 2 2 17 18 3 1 1 4 3 2] + [ 81 0 4 0 2 1 0 0 17 969 1 3 1 21 5 5 0 1 0 0 8] + [ 2 7 13 6 1 1 4 5 5 3 973 1 1 12 2 2 1 1 5 0 8] + [ 1 1 2 0 0 13 0 3 0 1 0 948 22 7 0 5 1 15 0 11 5] + [ 1 3 3 2 0 1 1 0 1 0 0 37 983 1 1 4 3 11 2 4 10] + [ 2 0 1 0 4 5 0 0 7 10 5 3 2 1065 2 3 0 1 0 2 7] + [ 12 3 4 20 5 0 1 0 17 1 2 1 1 2 1004 0 1 2 9 0 16] + [ 2 4 0 0 3 0 2 0 0 0 0 5 7 2 1 1076 13 6 0 10 3] + [ 0 17 1 0 4 2 0 1 1 0 0 3 0 1 2 10 1102 0 1 3 13] + [ 0 0 0 1 0 0 1 0 1 0 0 1 26 1 0 6 1 994 1 1 4] + [ 1 8 10 22 1 0 0 24 2 0 1 0 0 0 8 0 0 0 979 2 10] + [ 0 2 3 2 2 4 14 8 0 0 1 10 6 1 0 5 8 1 4 1078 3] + [ 119 185 157 65 101 126 51 87 86 77 154 88 338 317 160 68 156 54 157 239 5120]] + +2023-10-05 21:51:45,803 - ==> Best [Top1: 84.995 Top5: 97.886 Sparsity:0.00 Params: 148928 on epoch: 146] +2023-10-05 21:51:45,803 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:51:45,809 - + +2023-10-05 21:51:45,809 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:51:46,930 - Epoch: [148][ 10/ 1236] Overall Loss 0.218023 Objective Loss 0.218023 LR 0.000250 Time 0.112046 +2023-10-05 21:51:47,135 - Epoch: [148][ 20/ 1236] Overall Loss 0.207637 Objective Loss 0.207637 LR 0.000250 Time 0.066226 +2023-10-05 21:51:47,338 - Epoch: [148][ 30/ 1236] Overall Loss 0.212218 Objective Loss 0.212218 LR 0.000250 Time 0.050906 +2023-10-05 21:51:47,543 - Epoch: [148][ 40/ 1236] Overall Loss 0.220274 Objective Loss 0.220274 LR 0.000250 Time 0.043286 +2023-10-05 21:51:47,746 - Epoch: [148][ 50/ 1236] Overall Loss 0.215071 Objective Loss 0.215071 LR 0.000250 Time 0.038691 +2023-10-05 21:51:47,951 - Epoch: [148][ 60/ 1236] Overall Loss 0.212812 Objective Loss 0.212812 LR 0.000250 Time 0.035652 +2023-10-05 21:51:48,155 - Epoch: [148][ 70/ 1236] Overall Loss 0.213069 Objective Loss 0.213069 LR 0.000250 Time 0.033465 +2023-10-05 21:51:48,360 - Epoch: [148][ 80/ 1236] Overall Loss 0.214043 Objective Loss 0.214043 LR 0.000250 Time 0.031837 +2023-10-05 21:51:48,564 - Epoch: [148][ 90/ 1236] Overall Loss 0.213185 Objective Loss 0.213185 LR 0.000250 Time 0.030566 +2023-10-05 21:51:48,769 - Epoch: [148][ 100/ 1236] Overall Loss 0.210767 Objective Loss 0.210767 LR 0.000250 Time 0.029556 +2023-10-05 21:51:48,974 - Epoch: [148][ 110/ 1236] Overall Loss 0.210639 Objective Loss 0.210639 LR 0.000250 Time 0.028726 +2023-10-05 21:51:49,179 - Epoch: [148][ 120/ 1236] Overall Loss 0.209340 Objective Loss 0.209340 LR 0.000250 Time 0.028040 +2023-10-05 21:51:49,381 - Epoch: [148][ 130/ 1236] Overall Loss 0.209637 Objective Loss 0.209637 LR 0.000250 Time 0.027437 +2023-10-05 21:51:49,584 - Epoch: [148][ 140/ 1236] Overall Loss 0.210985 Objective Loss 0.210985 LR 0.000250 Time 0.026925 +2023-10-05 21:51:49,785 - Epoch: [148][ 150/ 1236] Overall Loss 0.212271 Objective Loss 0.212271 LR 0.000250 Time 0.026469 +2023-10-05 21:51:49,990 - Epoch: [148][ 160/ 1236] Overall Loss 0.210397 Objective Loss 0.210397 LR 0.000250 Time 0.026090 +2023-10-05 21:51:50,190 - Epoch: [148][ 170/ 1236] Overall Loss 0.209611 Objective Loss 0.209611 LR 0.000250 Time 0.025733 +2023-10-05 21:51:50,394 - Epoch: [148][ 180/ 1236] Overall Loss 0.210194 Objective Loss 0.210194 LR 0.000250 Time 0.025433 +2023-10-05 21:51:50,595 - Epoch: [148][ 190/ 1236] Overall Loss 0.210038 Objective Loss 0.210038 LR 0.000250 Time 0.025153 +2023-10-05 21:51:50,799 - Epoch: [148][ 200/ 1236] Overall Loss 0.209517 Objective Loss 0.209517 LR 0.000250 Time 0.024911 +2023-10-05 21:51:51,000 - Epoch: [148][ 210/ 1236] Overall Loss 0.209044 Objective Loss 0.209044 LR 0.000250 Time 0.024681 +2023-10-05 21:51:51,203 - Epoch: [148][ 220/ 1236] Overall Loss 0.208569 Objective Loss 0.208569 LR 0.000250 Time 0.024479 +2023-10-05 21:51:51,404 - Epoch: [148][ 230/ 1236] Overall Loss 0.208419 Objective Loss 0.208419 LR 0.000250 Time 0.024286 +2023-10-05 21:51:51,607 - Epoch: [148][ 240/ 1236] Overall Loss 0.208951 Objective Loss 0.208951 LR 0.000250 Time 0.024120 +2023-10-05 21:51:51,808 - Epoch: [148][ 250/ 1236] Overall Loss 0.209209 Objective Loss 0.209209 LR 0.000250 Time 0.023957 +2023-10-05 21:51:52,011 - Epoch: [148][ 260/ 1236] Overall Loss 0.210202 Objective Loss 0.210202 LR 0.000250 Time 0.023817 +2023-10-05 21:51:52,212 - Epoch: [148][ 270/ 1236] Overall Loss 0.210286 Objective Loss 0.210286 LR 0.000250 Time 0.023677 +2023-10-05 21:51:52,415 - Epoch: [148][ 280/ 1236] Overall Loss 0.210045 Objective Loss 0.210045 LR 0.000250 Time 0.023556 +2023-10-05 21:51:52,616 - Epoch: [148][ 290/ 1236] Overall Loss 0.209980 Objective Loss 0.209980 LR 0.000250 Time 0.023434 +2023-10-05 21:51:52,819 - Epoch: [148][ 300/ 1236] Overall Loss 0.210256 Objective Loss 0.210256 LR 0.000250 Time 0.023329 +2023-10-05 21:51:53,020 - Epoch: [148][ 310/ 1236] Overall Loss 0.210008 Objective Loss 0.210008 LR 0.000250 Time 0.023223 +2023-10-05 21:51:53,224 - Epoch: [148][ 320/ 1236] Overall Loss 0.209844 Objective Loss 0.209844 LR 0.000250 Time 0.023132 +2023-10-05 21:51:53,425 - Epoch: [148][ 330/ 1236] Overall Loss 0.210354 Objective Loss 0.210354 LR 0.000250 Time 0.023041 +2023-10-05 21:51:53,629 - Epoch: [148][ 340/ 1236] Overall Loss 0.210036 Objective Loss 0.210036 LR 0.000250 Time 0.022961 +2023-10-05 21:51:53,830 - Epoch: [148][ 350/ 1236] Overall Loss 0.209703 Objective Loss 0.209703 LR 0.000250 Time 0.022879 +2023-10-05 21:51:54,034 - Epoch: [148][ 360/ 1236] Overall Loss 0.209580 Objective Loss 0.209580 LR 0.000250 Time 0.022808 +2023-10-05 21:51:54,235 - Epoch: [148][ 370/ 1236] Overall Loss 0.210359 Objective Loss 0.210359 LR 0.000250 Time 0.022735 +2023-10-05 21:51:54,439 - Epoch: [148][ 380/ 1236] Overall Loss 0.210588 Objective Loss 0.210588 LR 0.000250 Time 0.022672 +2023-10-05 21:51:54,640 - Epoch: [148][ 390/ 1236] Overall Loss 0.210715 Objective Loss 0.210715 LR 0.000250 Time 0.022606 +2023-10-05 21:51:54,844 - Epoch: [148][ 400/ 1236] Overall Loss 0.211222 Objective Loss 0.211222 LR 0.000250 Time 0.022549 +2023-10-05 21:51:55,045 - Epoch: [148][ 410/ 1236] Overall Loss 0.211221 Objective Loss 0.211221 LR 0.000250 Time 0.022489 +2023-10-05 21:51:55,249 - Epoch: [148][ 420/ 1236] Overall Loss 0.211375 Objective Loss 0.211375 LR 0.000250 Time 0.022438 +2023-10-05 21:51:55,450 - Epoch: [148][ 430/ 1236] Overall Loss 0.211396 Objective Loss 0.211396 LR 0.000250 Time 0.022383 +2023-10-05 21:51:55,653 - Epoch: [148][ 440/ 1236] Overall Loss 0.211568 Objective Loss 0.211568 LR 0.000250 Time 0.022335 +2023-10-05 21:51:55,854 - Epoch: [148][ 450/ 1236] Overall Loss 0.211303 Objective Loss 0.211303 LR 0.000250 Time 0.022285 +2023-10-05 21:51:56,058 - Epoch: [148][ 460/ 1236] Overall Loss 0.211075 Objective Loss 0.211075 LR 0.000250 Time 0.022243 +2023-10-05 21:51:56,259 - Epoch: [148][ 470/ 1236] Overall Loss 0.210584 Objective Loss 0.210584 LR 0.000250 Time 0.022196 +2023-10-05 21:51:56,463 - Epoch: [148][ 480/ 1236] Overall Loss 0.210403 Objective Loss 0.210403 LR 0.000250 Time 0.022158 +2023-10-05 21:51:56,664 - Epoch: [148][ 490/ 1236] Overall Loss 0.210657 Objective Loss 0.210657 LR 0.000250 Time 0.022116 +2023-10-05 21:51:56,868 - Epoch: [148][ 500/ 1236] Overall Loss 0.210651 Objective Loss 0.210651 LR 0.000250 Time 0.022080 +2023-10-05 21:51:57,069 - Epoch: [148][ 510/ 1236] Overall Loss 0.210727 Objective Loss 0.210727 LR 0.000250 Time 0.022042 +2023-10-05 21:51:57,273 - Epoch: [148][ 520/ 1236] Overall Loss 0.210573 Objective Loss 0.210573 LR 0.000250 Time 0.022009 +2023-10-05 21:51:57,474 - Epoch: [148][ 530/ 1236] Overall Loss 0.210305 Objective Loss 0.210305 LR 0.000250 Time 0.021972 +2023-10-05 21:51:57,678 - Epoch: [148][ 540/ 1236] Overall Loss 0.210107 Objective Loss 0.210107 LR 0.000250 Time 0.021942 +2023-10-05 21:51:57,879 - Epoch: [148][ 550/ 1236] Overall Loss 0.210220 Objective Loss 0.210220 LR 0.000250 Time 0.021909 +2023-10-05 21:51:58,083 - Epoch: [148][ 560/ 1236] Overall Loss 0.210268 Objective Loss 0.210268 LR 0.000250 Time 0.021881 +2023-10-05 21:51:58,284 - Epoch: [148][ 570/ 1236] Overall Loss 0.210227 Objective Loss 0.210227 LR 0.000250 Time 0.021850 +2023-10-05 21:51:58,488 - Epoch: [148][ 580/ 1236] Overall Loss 0.209865 Objective Loss 0.209865 LR 0.000250 Time 0.021823 +2023-10-05 21:51:58,689 - Epoch: [148][ 590/ 1236] Overall Loss 0.209824 Objective Loss 0.209824 LR 0.000250 Time 0.021794 +2023-10-05 21:51:58,893 - Epoch: [148][ 600/ 1236] Overall Loss 0.209658 Objective Loss 0.209658 LR 0.000250 Time 0.021770 +2023-10-05 21:51:59,094 - Epoch: [148][ 610/ 1236] Overall Loss 0.209930 Objective Loss 0.209930 LR 0.000250 Time 0.021742 +2023-10-05 21:51:59,298 - Epoch: [148][ 620/ 1236] Overall Loss 0.209885 Objective Loss 0.209885 LR 0.000250 Time 0.021719 +2023-10-05 21:51:59,499 - Epoch: [148][ 630/ 1236] Overall Loss 0.209732 Objective Loss 0.209732 LR 0.000250 Time 0.021694 +2023-10-05 21:51:59,704 - Epoch: [148][ 640/ 1236] Overall Loss 0.209718 Objective Loss 0.209718 LR 0.000250 Time 0.021674 +2023-10-05 21:51:59,904 - Epoch: [148][ 650/ 1236] Overall Loss 0.209938 Objective Loss 0.209938 LR 0.000250 Time 0.021648 +2023-10-05 21:52:00,108 - Epoch: [148][ 660/ 1236] Overall Loss 0.210025 Objective Loss 0.210025 LR 0.000250 Time 0.021628 +2023-10-05 21:52:00,309 - Epoch: [148][ 670/ 1236] Overall Loss 0.209864 Objective Loss 0.209864 LR 0.000250 Time 0.021605 +2023-10-05 21:52:00,513 - Epoch: [148][ 680/ 1236] Overall Loss 0.209659 Objective Loss 0.209659 LR 0.000250 Time 0.021587 +2023-10-05 21:52:00,714 - Epoch: [148][ 690/ 1236] Overall Loss 0.209996 Objective Loss 0.209996 LR 0.000250 Time 0.021565 +2023-10-05 21:52:00,918 - Epoch: [148][ 700/ 1236] Overall Loss 0.209996 Objective Loss 0.209996 LR 0.000250 Time 0.021547 +2023-10-05 21:52:01,119 - Epoch: [148][ 710/ 1236] Overall Loss 0.210001 Objective Loss 0.210001 LR 0.000250 Time 0.021527 +2023-10-05 21:52:01,323 - Epoch: [148][ 720/ 1236] Overall Loss 0.209857 Objective Loss 0.209857 LR 0.000250 Time 0.021510 +2023-10-05 21:52:01,524 - Epoch: [148][ 730/ 1236] Overall Loss 0.210202 Objective Loss 0.210202 LR 0.000250 Time 0.021491 +2023-10-05 21:52:01,727 - Epoch: [148][ 740/ 1236] Overall Loss 0.210515 Objective Loss 0.210515 LR 0.000250 Time 0.021475 +2023-10-05 21:52:01,929 - Epoch: [148][ 750/ 1236] Overall Loss 0.210578 Objective Loss 0.210578 LR 0.000250 Time 0.021457 +2023-10-05 21:52:02,132 - Epoch: [148][ 760/ 1236] Overall Loss 0.210421 Objective Loss 0.210421 LR 0.000250 Time 0.021442 +2023-10-05 21:52:02,334 - Epoch: [148][ 770/ 1236] Overall Loss 0.210553 Objective Loss 0.210553 LR 0.000250 Time 0.021425 +2023-10-05 21:52:02,538 - Epoch: [148][ 780/ 1236] Overall Loss 0.211088 Objective Loss 0.211088 LR 0.000250 Time 0.021411 +2023-10-05 21:52:02,739 - Epoch: [148][ 790/ 1236] Overall Loss 0.211031 Objective Loss 0.211031 LR 0.000250 Time 0.021394 +2023-10-05 21:52:02,943 - Epoch: [148][ 800/ 1236] Overall Loss 0.211062 Objective Loss 0.211062 LR 0.000250 Time 0.021381 +2023-10-05 21:52:03,144 - Epoch: [148][ 810/ 1236] Overall Loss 0.211092 Objective Loss 0.211092 LR 0.000250 Time 0.021365 +2023-10-05 21:52:03,348 - Epoch: [148][ 820/ 1236] Overall Loss 0.210913 Objective Loss 0.210913 LR 0.000250 Time 0.021353 +2023-10-05 21:52:03,549 - Epoch: [148][ 830/ 1236] Overall Loss 0.211008 Objective Loss 0.211008 LR 0.000250 Time 0.021338 +2023-10-05 21:52:03,753 - Epoch: [148][ 840/ 1236] Overall Loss 0.210792 Objective Loss 0.210792 LR 0.000250 Time 0.021326 +2023-10-05 21:52:03,954 - Epoch: [148][ 850/ 1236] Overall Loss 0.210751 Objective Loss 0.210751 LR 0.000250 Time 0.021311 +2023-10-05 21:52:04,158 - Epoch: [148][ 860/ 1236] Overall Loss 0.211004 Objective Loss 0.211004 LR 0.000250 Time 0.021300 +2023-10-05 21:52:04,359 - Epoch: [148][ 870/ 1236] Overall Loss 0.210936 Objective Loss 0.210936 LR 0.000250 Time 0.021286 +2023-10-05 21:52:04,563 - Epoch: [148][ 880/ 1236] Overall Loss 0.210754 Objective Loss 0.210754 LR 0.000250 Time 0.021275 +2023-10-05 21:52:04,764 - Epoch: [148][ 890/ 1236] Overall Loss 0.210837 Objective Loss 0.210837 LR 0.000250 Time 0.021262 +2023-10-05 21:52:04,968 - Epoch: [148][ 900/ 1236] Overall Loss 0.210799 Objective Loss 0.210799 LR 0.000250 Time 0.021252 +2023-10-05 21:52:05,169 - Epoch: [148][ 910/ 1236] Overall Loss 0.210568 Objective Loss 0.210568 LR 0.000250 Time 0.021239 +2023-10-05 21:52:05,373 - Epoch: [148][ 920/ 1236] Overall Loss 0.210531 Objective Loss 0.210531 LR 0.000250 Time 0.021229 +2023-10-05 21:52:05,574 - Epoch: [148][ 930/ 1236] Overall Loss 0.210608 Objective Loss 0.210608 LR 0.000250 Time 0.021217 +2023-10-05 21:52:05,778 - Epoch: [148][ 940/ 1236] Overall Loss 0.210789 Objective Loss 0.210789 LR 0.000250 Time 0.021208 +2023-10-05 21:52:05,979 - Epoch: [148][ 950/ 1236] Overall Loss 0.210784 Objective Loss 0.210784 LR 0.000250 Time 0.021196 +2023-10-05 21:52:06,183 - Epoch: [148][ 960/ 1236] Overall Loss 0.210879 Objective Loss 0.210879 LR 0.000250 Time 0.021187 +2023-10-05 21:52:06,384 - Epoch: [148][ 970/ 1236] Overall Loss 0.211170 Objective Loss 0.211170 LR 0.000250 Time 0.021176 +2023-10-05 21:52:06,588 - Epoch: [148][ 980/ 1236] Overall Loss 0.211162 Objective Loss 0.211162 LR 0.000250 Time 0.021167 +2023-10-05 21:52:06,789 - Epoch: [148][ 990/ 1236] Overall Loss 0.211277 Objective Loss 0.211277 LR 0.000250 Time 0.021157 +2023-10-05 21:52:06,993 - Epoch: [148][ 1000/ 1236] Overall Loss 0.211554 Objective Loss 0.211554 LR 0.000250 Time 0.021149 +2023-10-05 21:52:07,194 - Epoch: [148][ 1010/ 1236] Overall Loss 0.211381 Objective Loss 0.211381 LR 0.000250 Time 0.021138 +2023-10-05 21:52:07,398 - Epoch: [148][ 1020/ 1236] Overall Loss 0.211648 Objective Loss 0.211648 LR 0.000250 Time 0.021130 +2023-10-05 21:52:07,599 - Epoch: [148][ 1030/ 1236] Overall Loss 0.211682 Objective Loss 0.211682 LR 0.000250 Time 0.021120 +2023-10-05 21:52:07,803 - Epoch: [148][ 1040/ 1236] Overall Loss 0.211898 Objective Loss 0.211898 LR 0.000250 Time 0.021113 +2023-10-05 21:52:08,004 - Epoch: [148][ 1050/ 1236] Overall Loss 0.212051 Objective Loss 0.212051 LR 0.000250 Time 0.021103 +2023-10-05 21:52:08,208 - Epoch: [148][ 1060/ 1236] Overall Loss 0.212137 Objective Loss 0.212137 LR 0.000250 Time 0.021096 +2023-10-05 21:52:08,409 - Epoch: [148][ 1070/ 1236] Overall Loss 0.212170 Objective Loss 0.212170 LR 0.000250 Time 0.021086 +2023-10-05 21:52:08,613 - Epoch: [148][ 1080/ 1236] Overall Loss 0.211939 Objective Loss 0.211939 LR 0.000250 Time 0.021079 +2023-10-05 21:52:08,814 - Epoch: [148][ 1090/ 1236] Overall Loss 0.212102 Objective Loss 0.212102 LR 0.000250 Time 0.021071 +2023-10-05 21:52:09,018 - Epoch: [148][ 1100/ 1236] Overall Loss 0.211826 Objective Loss 0.211826 LR 0.000250 Time 0.021064 +2023-10-05 21:52:09,219 - Epoch: [148][ 1110/ 1236] Overall Loss 0.212027 Objective Loss 0.212027 LR 0.000250 Time 0.021055 +2023-10-05 21:52:09,423 - Epoch: [148][ 1120/ 1236] Overall Loss 0.212259 Objective Loss 0.212259 LR 0.000250 Time 0.021049 +2023-10-05 21:52:09,624 - Epoch: [148][ 1130/ 1236] Overall Loss 0.212131 Objective Loss 0.212131 LR 0.000250 Time 0.021040 +2023-10-05 21:52:09,828 - Epoch: [148][ 1140/ 1236] Overall Loss 0.212158 Objective Loss 0.212158 LR 0.000250 Time 0.021034 +2023-10-05 21:52:10,029 - Epoch: [148][ 1150/ 1236] Overall Loss 0.212064 Objective Loss 0.212064 LR 0.000250 Time 0.021026 +2023-10-05 21:52:10,233 - Epoch: [148][ 1160/ 1236] Overall Loss 0.212102 Objective Loss 0.212102 LR 0.000250 Time 0.021020 +2023-10-05 21:52:10,434 - Epoch: [148][ 1170/ 1236] Overall Loss 0.212198 Objective Loss 0.212198 LR 0.000250 Time 0.021013 +2023-10-05 21:52:10,638 - Epoch: [148][ 1180/ 1236] Overall Loss 0.212067 Objective Loss 0.212067 LR 0.000250 Time 0.021007 +2023-10-05 21:52:10,839 - Epoch: [148][ 1190/ 1236] Overall Loss 0.212020 Objective Loss 0.212020 LR 0.000250 Time 0.020999 +2023-10-05 21:52:11,043 - Epoch: [148][ 1200/ 1236] Overall Loss 0.212068 Objective Loss 0.212068 LR 0.000250 Time 0.020993 +2023-10-05 21:52:11,244 - Epoch: [148][ 1210/ 1236] Overall Loss 0.212125 Objective Loss 0.212125 LR 0.000250 Time 0.020986 +2023-10-05 21:52:11,448 - Epoch: [148][ 1220/ 1236] Overall Loss 0.212073 Objective Loss 0.212073 LR 0.000250 Time 0.020981 +2023-10-05 21:52:11,706 - Epoch: [148][ 1230/ 1236] Overall Loss 0.212105 Objective Loss 0.212105 LR 0.000250 Time 0.021020 +2023-10-05 21:52:11,824 - Epoch: [148][ 1236/ 1236] Overall Loss 0.212168 Objective Loss 0.212168 Top1 86.354379 Top5 97.963340 LR 0.000250 Time 0.021013 +2023-10-05 21:52:11,956 - --- validate (epoch=148)----------- +2023-10-05 21:52:11,956 - 29943 samples (256 per mini-batch) +2023-10-05 21:52:12,423 - Epoch: [148][ 10/ 117] Loss 0.308678 Top1 84.765625 Top5 98.085938 +2023-10-05 21:52:12,588 - Epoch: [148][ 20/ 117] Loss 0.305674 Top1 84.726562 Top5 98.066406 +2023-10-05 21:52:12,748 - Epoch: [148][ 30/ 117] Loss 0.297789 Top1 85.013021 Top5 98.177083 +2023-10-05 21:52:12,912 - Epoch: [148][ 40/ 117] Loss 0.307299 Top1 85.107422 Top5 98.027344 +2023-10-05 21:52:13,072 - Epoch: [148][ 50/ 117] Loss 0.309685 Top1 85.054688 Top5 98.015625 +2023-10-05 21:52:13,240 - Epoch: [148][ 60/ 117] Loss 0.309300 Top1 85.110677 Top5 97.988281 +2023-10-05 21:52:13,400 - Epoch: [148][ 70/ 117] Loss 0.305525 Top1 85.228795 Top5 97.979911 +2023-10-05 21:52:13,563 - Epoch: [148][ 80/ 117] Loss 0.304256 Top1 85.263672 Top5 98.032227 +2023-10-05 21:52:13,720 - Epoch: [148][ 90/ 117] Loss 0.307938 Top1 85.073785 Top5 98.003472 +2023-10-05 21:52:13,882 - Epoch: [148][ 100/ 117] Loss 0.308336 Top1 85.046875 Top5 98.003906 +2023-10-05 21:52:14,045 - Epoch: [148][ 110/ 117] Loss 0.307949 Top1 85.060369 Top5 98.004261 +2023-10-05 21:52:14,131 - Epoch: [148][ 117/ 117] Loss 0.308405 Top1 85.068296 Top5 98.029590 +2023-10-05 21:52:14,279 - ==> Top1: 85.068 Top5: 98.030 Loss: 0.308 + +2023-10-05 21:52:14,279 - ==> Confusion: +[[ 944 2 2 3 6 3 0 0 4 59 1 0 1 2 4 1 4 1 0 0 13] + [ 1 1047 5 0 7 22 1 22 0 0 1 1 0 0 0 3 4 0 7 1 9] + [ 5 1 968 16 1 2 19 8 0 2 4 0 10 2 0 3 1 1 4 3 6] + [ 4 0 14 990 0 3 0 2 1 1 5 1 3 1 24 3 0 7 14 1 15] + [ 25 8 2 0 971 7 1 0 0 8 2 1 1 1 5 4 7 2 0 1 4] + [ 4 23 1 1 3 1019 0 16 1 0 4 5 2 9 4 1 6 0 1 5 11] + [ 0 5 27 0 0 1 1117 6 0 1 2 4 1 0 1 7 0 0 1 10 8] + [ 3 13 15 0 2 34 4 1077 1 3 4 14 1 2 1 3 0 0 26 6 9] + [ 19 2 0 0 2 3 0 2 973 44 10 0 2 13 10 3 0 0 3 1 2] + [ 109 1 1 0 4 4 1 0 21 944 0 1 0 16 3 7 0 0 0 1 6] + [ 2 2 12 4 1 0 2 4 8 2 972 3 4 14 4 2 1 0 5 1 10] + [ 2 1 0 0 0 15 0 2 0 1 1 956 22 6 0 6 1 14 0 4 4] + [ 0 1 2 5 0 4 0 1 0 0 1 38 979 1 3 6 1 16 1 1 8] + [ 2 0 0 0 2 12 0 1 16 13 4 5 1 1047 4 1 1 1 0 1 8] + [ 15 1 3 17 5 0 0 0 22 3 3 2 3 1 1000 0 1 0 7 0 18] + [ 0 3 1 0 3 0 1 0 0 0 0 7 5 2 1 1075 12 13 1 8 2] + [ 1 9 2 1 6 3 0 2 2 0 0 4 1 0 2 11 1098 0 0 3 16] + [ 0 0 0 2 1 0 3 0 0 1 0 2 11 1 1 4 0 1009 1 0 2] + [ 3 5 9 30 1 0 0 27 3 0 2 1 3 1 7 0 0 0 965 2 9] + [ 0 2 5 2 2 6 9 6 0 0 2 14 3 1 0 5 9 2 2 1072 10] + [ 141 161 165 81 90 159 26 92 98 71 179 124 328 236 123 64 128 71 130 189 5249]] + +2023-10-05 21:52:14,281 - ==> Best [Top1: 85.068 Top5: 98.030 Sparsity:0.00 Params: 148928 on epoch: 148] +2023-10-05 21:52:14,281 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:52:14,294 - + +2023-10-05 21:52:14,294 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:52:15,295 - Epoch: [149][ 10/ 1236] Overall Loss 0.194041 Objective Loss 0.194041 LR 0.000250 Time 0.100033 +2023-10-05 21:52:15,498 - Epoch: [149][ 20/ 1236] Overall Loss 0.196554 Objective Loss 0.196554 LR 0.000250 Time 0.060136 +2023-10-05 21:52:15,698 - Epoch: [149][ 30/ 1236] Overall Loss 0.200218 Objective Loss 0.200218 LR 0.000250 Time 0.046759 +2023-10-05 21:52:15,901 - Epoch: [149][ 40/ 1236] Overall Loss 0.200110 Objective Loss 0.200110 LR 0.000250 Time 0.040132 +2023-10-05 21:52:16,101 - Epoch: [149][ 50/ 1236] Overall Loss 0.200729 Objective Loss 0.200729 LR 0.000250 Time 0.036112 +2023-10-05 21:52:16,304 - Epoch: [149][ 60/ 1236] Overall Loss 0.204005 Objective Loss 0.204005 LR 0.000250 Time 0.033462 +2023-10-05 21:52:16,505 - Epoch: [149][ 70/ 1236] Overall Loss 0.205044 Objective Loss 0.205044 LR 0.000250 Time 0.031549 +2023-10-05 21:52:16,708 - Epoch: [149][ 80/ 1236] Overall Loss 0.206043 Objective Loss 0.206043 LR 0.000250 Time 0.030141 +2023-10-05 21:52:16,909 - Epoch: [149][ 90/ 1236] Overall Loss 0.207235 Objective Loss 0.207235 LR 0.000250 Time 0.029019 +2023-10-05 21:52:17,112 - Epoch: [149][ 100/ 1236] Overall Loss 0.207409 Objective Loss 0.207409 LR 0.000250 Time 0.028143 +2023-10-05 21:52:17,312 - Epoch: [149][ 110/ 1236] Overall Loss 0.206046 Objective Loss 0.206046 LR 0.000250 Time 0.027405 +2023-10-05 21:52:17,516 - Epoch: [149][ 120/ 1236] Overall Loss 0.207646 Objective Loss 0.207646 LR 0.000250 Time 0.026818 +2023-10-05 21:52:17,721 - Epoch: [149][ 130/ 1236] Overall Loss 0.208610 Objective Loss 0.208610 LR 0.000250 Time 0.026325 +2023-10-05 21:52:17,926 - Epoch: [149][ 140/ 1236] Overall Loss 0.208680 Objective Loss 0.208680 LR 0.000250 Time 0.025907 +2023-10-05 21:52:18,130 - Epoch: [149][ 150/ 1236] Overall Loss 0.208390 Objective Loss 0.208390 LR 0.000250 Time 0.025539 +2023-10-05 21:52:18,335 - Epoch: [149][ 160/ 1236] Overall Loss 0.208489 Objective Loss 0.208489 LR 0.000250 Time 0.025222 +2023-10-05 21:52:18,540 - Epoch: [149][ 170/ 1236] Overall Loss 0.210028 Objective Loss 0.210028 LR 0.000250 Time 0.024940 +2023-10-05 21:52:18,743 - Epoch: [149][ 180/ 1236] Overall Loss 0.210348 Objective Loss 0.210348 LR 0.000250 Time 0.024680 +2023-10-05 21:52:18,943 - Epoch: [149][ 190/ 1236] Overall Loss 0.209897 Objective Loss 0.209897 LR 0.000250 Time 0.024434 +2023-10-05 21:52:19,146 - Epoch: [149][ 200/ 1236] Overall Loss 0.209765 Objective Loss 0.209765 LR 0.000250 Time 0.024227 +2023-10-05 21:52:19,347 - Epoch: [149][ 210/ 1236] Overall Loss 0.209305 Objective Loss 0.209305 LR 0.000250 Time 0.024027 +2023-10-05 21:52:19,551 - Epoch: [149][ 220/ 1236] Overall Loss 0.209532 Objective Loss 0.209532 LR 0.000250 Time 0.023858 +2023-10-05 21:52:19,751 - Epoch: [149][ 230/ 1236] Overall Loss 0.211294 Objective Loss 0.211294 LR 0.000250 Time 0.023691 +2023-10-05 21:52:19,954 - Epoch: [149][ 240/ 1236] Overall Loss 0.211446 Objective Loss 0.211446 LR 0.000250 Time 0.023549 +2023-10-05 21:52:20,155 - Epoch: [149][ 250/ 1236] Overall Loss 0.211710 Objective Loss 0.211710 LR 0.000250 Time 0.023409 +2023-10-05 21:52:20,358 - Epoch: [149][ 260/ 1236] Overall Loss 0.210783 Objective Loss 0.210783 LR 0.000250 Time 0.023289 +2023-10-05 21:52:20,559 - Epoch: [149][ 270/ 1236] Overall Loss 0.210589 Objective Loss 0.210589 LR 0.000250 Time 0.023168 +2023-10-05 21:52:20,762 - Epoch: [149][ 280/ 1236] Overall Loss 0.210599 Objective Loss 0.210599 LR 0.000250 Time 0.023064 +2023-10-05 21:52:20,964 - Epoch: [149][ 290/ 1236] Overall Loss 0.210103 Objective Loss 0.210103 LR 0.000250 Time 0.022964 +2023-10-05 21:52:21,166 - Epoch: [149][ 300/ 1236] Overall Loss 0.209839 Objective Loss 0.209839 LR 0.000250 Time 0.022871 +2023-10-05 21:52:21,366 - Epoch: [149][ 310/ 1236] Overall Loss 0.210450 Objective Loss 0.210450 LR 0.000250 Time 0.022778 +2023-10-05 21:52:21,569 - Epoch: [149][ 320/ 1236] Overall Loss 0.211027 Objective Loss 0.211027 LR 0.000250 Time 0.022697 +2023-10-05 21:52:21,770 - Epoch: [149][ 330/ 1236] Overall Loss 0.210875 Objective Loss 0.210875 LR 0.000250 Time 0.022620 +2023-10-05 21:52:21,973 - Epoch: [149][ 340/ 1236] Overall Loss 0.210983 Objective Loss 0.210983 LR 0.000250 Time 0.022551 +2023-10-05 21:52:22,175 - Epoch: [149][ 350/ 1236] Overall Loss 0.210898 Objective Loss 0.210898 LR 0.000250 Time 0.022481 +2023-10-05 21:52:22,379 - Epoch: [149][ 360/ 1236] Overall Loss 0.211187 Objective Loss 0.211187 LR 0.000250 Time 0.022422 +2023-10-05 21:52:22,582 - Epoch: [149][ 370/ 1236] Overall Loss 0.211568 Objective Loss 0.211568 LR 0.000250 Time 0.022364 +2023-10-05 21:52:22,785 - Epoch: [149][ 380/ 1236] Overall Loss 0.211456 Objective Loss 0.211456 LR 0.000250 Time 0.022310 +2023-10-05 21:52:22,988 - Epoch: [149][ 390/ 1236] Overall Loss 0.211299 Objective Loss 0.211299 LR 0.000250 Time 0.022258 +2023-10-05 21:52:23,192 - Epoch: [149][ 400/ 1236] Overall Loss 0.210752 Objective Loss 0.210752 LR 0.000250 Time 0.022209 +2023-10-05 21:52:23,394 - Epoch: [149][ 410/ 1236] Overall Loss 0.210829 Objective Loss 0.210829 LR 0.000250 Time 0.022161 +2023-10-05 21:52:23,597 - Epoch: [149][ 420/ 1236] Overall Loss 0.209967 Objective Loss 0.209967 LR 0.000250 Time 0.022116 +2023-10-05 21:52:23,800 - Epoch: [149][ 430/ 1236] Overall Loss 0.209653 Objective Loss 0.209653 LR 0.000250 Time 0.022072 +2023-10-05 21:52:24,003 - Epoch: [149][ 440/ 1236] Overall Loss 0.209266 Objective Loss 0.209266 LR 0.000250 Time 0.022032 +2023-10-05 21:52:24,206 - Epoch: [149][ 450/ 1236] Overall Loss 0.209097 Objective Loss 0.209097 LR 0.000250 Time 0.021992 +2023-10-05 21:52:24,410 - Epoch: [149][ 460/ 1236] Overall Loss 0.209529 Objective Loss 0.209529 LR 0.000250 Time 0.021956 +2023-10-05 21:52:24,613 - Epoch: [149][ 470/ 1236] Overall Loss 0.209304 Objective Loss 0.209304 LR 0.000250 Time 0.021921 +2023-10-05 21:52:24,817 - Epoch: [149][ 480/ 1236] Overall Loss 0.209324 Objective Loss 0.209324 LR 0.000250 Time 0.021888 +2023-10-05 21:52:25,020 - Epoch: [149][ 490/ 1236] Overall Loss 0.209143 Objective Loss 0.209143 LR 0.000250 Time 0.021854 +2023-10-05 21:52:25,223 - Epoch: [149][ 500/ 1236] Overall Loss 0.209093 Objective Loss 0.209093 LR 0.000250 Time 0.021823 +2023-10-05 21:52:25,426 - Epoch: [149][ 510/ 1236] Overall Loss 0.209122 Objective Loss 0.209122 LR 0.000250 Time 0.021792 +2023-10-05 21:52:25,631 - Epoch: [149][ 520/ 1236] Overall Loss 0.209152 Objective Loss 0.209152 LR 0.000250 Time 0.021767 +2023-10-05 21:52:25,836 - Epoch: [149][ 530/ 1236] Overall Loss 0.209670 Objective Loss 0.209670 LR 0.000250 Time 0.021742 +2023-10-05 21:52:26,041 - Epoch: [149][ 540/ 1236] Overall Loss 0.209492 Objective Loss 0.209492 LR 0.000250 Time 0.021719 +2023-10-05 21:52:26,245 - Epoch: [149][ 550/ 1236] Overall Loss 0.210069 Objective Loss 0.210069 LR 0.000250 Time 0.021694 +2023-10-05 21:52:26,451 - Epoch: [149][ 560/ 1236] Overall Loss 0.209930 Objective Loss 0.209930 LR 0.000250 Time 0.021674 +2023-10-05 21:52:26,656 - Epoch: [149][ 570/ 1236] Overall Loss 0.210021 Objective Loss 0.210021 LR 0.000250 Time 0.021653 +2023-10-05 21:52:26,861 - Epoch: [149][ 580/ 1236] Overall Loss 0.209810 Objective Loss 0.209810 LR 0.000250 Time 0.021633 +2023-10-05 21:52:27,064 - Epoch: [149][ 590/ 1236] Overall Loss 0.209532 Objective Loss 0.209532 LR 0.000250 Time 0.021609 +2023-10-05 21:52:27,267 - Epoch: [149][ 600/ 1236] Overall Loss 0.209701 Objective Loss 0.209701 LR 0.000250 Time 0.021588 +2023-10-05 21:52:27,470 - Epoch: [149][ 610/ 1236] Overall Loss 0.209504 Objective Loss 0.209504 LR 0.000250 Time 0.021566 +2023-10-05 21:52:27,674 - Epoch: [149][ 620/ 1236] Overall Loss 0.209307 Objective Loss 0.209307 LR 0.000250 Time 0.021546 +2023-10-05 21:52:27,876 - Epoch: [149][ 630/ 1236] Overall Loss 0.209246 Objective Loss 0.209246 LR 0.000250 Time 0.021525 +2023-10-05 21:52:28,079 - Epoch: [149][ 640/ 1236] Overall Loss 0.209200 Objective Loss 0.209200 LR 0.000250 Time 0.021505 +2023-10-05 21:52:28,282 - Epoch: [149][ 650/ 1236] Overall Loss 0.209652 Objective Loss 0.209652 LR 0.000250 Time 0.021486 +2023-10-05 21:52:28,486 - Epoch: [149][ 660/ 1236] Overall Loss 0.209768 Objective Loss 0.209768 LR 0.000250 Time 0.021469 +2023-10-05 21:52:28,688 - Epoch: [149][ 670/ 1236] Overall Loss 0.210065 Objective Loss 0.210065 LR 0.000250 Time 0.021450 +2023-10-05 21:52:28,892 - Epoch: [149][ 680/ 1236] Overall Loss 0.210484 Objective Loss 0.210484 LR 0.000250 Time 0.021433 +2023-10-05 21:52:29,095 - Epoch: [149][ 690/ 1236] Overall Loss 0.210715 Objective Loss 0.210715 LR 0.000250 Time 0.021416 +2023-10-05 21:52:29,299 - Epoch: [149][ 700/ 1236] Overall Loss 0.210720 Objective Loss 0.210720 LR 0.000250 Time 0.021401 +2023-10-05 21:52:29,502 - Epoch: [149][ 710/ 1236] Overall Loss 0.210898 Objective Loss 0.210898 LR 0.000250 Time 0.021385 +2023-10-05 21:52:29,705 - Epoch: [149][ 720/ 1236] Overall Loss 0.210934 Objective Loss 0.210934 LR 0.000250 Time 0.021370 +2023-10-05 21:52:29,908 - Epoch: [149][ 730/ 1236] Overall Loss 0.211139 Objective Loss 0.211139 LR 0.000250 Time 0.021354 +2023-10-05 21:52:30,112 - Epoch: [149][ 740/ 1236] Overall Loss 0.211217 Objective Loss 0.211217 LR 0.000250 Time 0.021341 +2023-10-05 21:52:30,314 - Epoch: [149][ 750/ 1236] Overall Loss 0.210847 Objective Loss 0.210847 LR 0.000250 Time 0.021326 +2023-10-05 21:52:30,518 - Epoch: [149][ 760/ 1236] Overall Loss 0.210795 Objective Loss 0.210795 LR 0.000250 Time 0.021313 +2023-10-05 21:52:30,721 - Epoch: [149][ 770/ 1236] Overall Loss 0.210824 Objective Loss 0.210824 LR 0.000250 Time 0.021299 +2023-10-05 21:52:30,924 - Epoch: [149][ 780/ 1236] Overall Loss 0.210893 Objective Loss 0.210893 LR 0.000250 Time 0.021286 +2023-10-05 21:52:31,127 - Epoch: [149][ 790/ 1236] Overall Loss 0.211062 Objective Loss 0.211062 LR 0.000250 Time 0.021273 +2023-10-05 21:52:31,330 - Epoch: [149][ 800/ 1236] Overall Loss 0.211308 Objective Loss 0.211308 LR 0.000250 Time 0.021261 +2023-10-05 21:52:31,535 - Epoch: [149][ 810/ 1236] Overall Loss 0.211234 Objective Loss 0.211234 LR 0.000250 Time 0.021250 +2023-10-05 21:52:31,738 - Epoch: [149][ 820/ 1236] Overall Loss 0.211521 Objective Loss 0.211521 LR 0.000250 Time 0.021239 +2023-10-05 21:52:31,942 - Epoch: [149][ 830/ 1236] Overall Loss 0.211249 Objective Loss 0.211249 LR 0.000250 Time 0.021228 +2023-10-05 21:52:32,145 - Epoch: [149][ 840/ 1236] Overall Loss 0.211110 Objective Loss 0.211110 LR 0.000250 Time 0.021217 +2023-10-05 21:52:32,349 - Epoch: [149][ 850/ 1236] Overall Loss 0.211353 Objective Loss 0.211353 LR 0.000250 Time 0.021206 +2023-10-05 21:52:32,552 - Epoch: [149][ 860/ 1236] Overall Loss 0.211325 Objective Loss 0.211325 LR 0.000250 Time 0.021195 +2023-10-05 21:52:32,755 - Epoch: [149][ 870/ 1236] Overall Loss 0.211327 Objective Loss 0.211327 LR 0.000250 Time 0.021185 +2023-10-05 21:52:32,957 - Epoch: [149][ 880/ 1236] Overall Loss 0.211384 Objective Loss 0.211384 LR 0.000250 Time 0.021174 +2023-10-05 21:52:33,161 - Epoch: [149][ 890/ 1236] Overall Loss 0.211325 Objective Loss 0.211325 LR 0.000250 Time 0.021164 +2023-10-05 21:52:33,364 - Epoch: [149][ 900/ 1236] Overall Loss 0.211283 Objective Loss 0.211283 LR 0.000250 Time 0.021154 +2023-10-05 21:52:33,567 - Epoch: [149][ 910/ 1236] Overall Loss 0.211282 Objective Loss 0.211282 LR 0.000250 Time 0.021144 +2023-10-05 21:52:33,770 - Epoch: [149][ 920/ 1236] Overall Loss 0.211223 Objective Loss 0.211223 LR 0.000250 Time 0.021135 +2023-10-05 21:52:33,973 - Epoch: [149][ 930/ 1236] Overall Loss 0.211022 Objective Loss 0.211022 LR 0.000250 Time 0.021126 +2023-10-05 21:52:34,176 - Epoch: [149][ 940/ 1236] Overall Loss 0.211055 Objective Loss 0.211055 LR 0.000250 Time 0.021117 +2023-10-05 21:52:34,380 - Epoch: [149][ 950/ 1236] Overall Loss 0.211111 Objective Loss 0.211111 LR 0.000250 Time 0.021108 +2023-10-05 21:52:34,583 - Epoch: [149][ 960/ 1236] Overall Loss 0.211204 Objective Loss 0.211204 LR 0.000250 Time 0.021100 +2023-10-05 21:52:34,786 - Epoch: [149][ 970/ 1236] Overall Loss 0.211242 Objective Loss 0.211242 LR 0.000250 Time 0.021091 +2023-10-05 21:52:34,989 - Epoch: [149][ 980/ 1236] Overall Loss 0.211115 Objective Loss 0.211115 LR 0.000250 Time 0.021083 +2023-10-05 21:52:35,195 - Epoch: [149][ 990/ 1236] Overall Loss 0.211104 Objective Loss 0.211104 LR 0.000250 Time 0.021078 +2023-10-05 21:52:35,400 - Epoch: [149][ 1000/ 1236] Overall Loss 0.211402 Objective Loss 0.211402 LR 0.000250 Time 0.021072 +2023-10-05 21:52:35,604 - Epoch: [149][ 1010/ 1236] Overall Loss 0.211647 Objective Loss 0.211647 LR 0.000250 Time 0.021065 +2023-10-05 21:52:35,807 - Epoch: [149][ 1020/ 1236] Overall Loss 0.211478 Objective Loss 0.211478 LR 0.000250 Time 0.021057 +2023-10-05 21:52:36,010 - Epoch: [149][ 1030/ 1236] Overall Loss 0.211615 Objective Loss 0.211615 LR 0.000250 Time 0.021049 +2023-10-05 21:52:36,213 - Epoch: [149][ 1040/ 1236] Overall Loss 0.211687 Objective Loss 0.211687 LR 0.000250 Time 0.021042 +2023-10-05 21:52:36,417 - Epoch: [149][ 1050/ 1236] Overall Loss 0.211706 Objective Loss 0.211706 LR 0.000250 Time 0.021035 +2023-10-05 21:52:36,620 - Epoch: [149][ 1060/ 1236] Overall Loss 0.211492 Objective Loss 0.211492 LR 0.000250 Time 0.021028 +2023-10-05 21:52:36,823 - Epoch: [149][ 1070/ 1236] Overall Loss 0.211477 Objective Loss 0.211477 LR 0.000250 Time 0.021021 +2023-10-05 21:52:37,026 - Epoch: [149][ 1080/ 1236] Overall Loss 0.211384 Objective Loss 0.211384 LR 0.000250 Time 0.021014 +2023-10-05 21:52:37,229 - Epoch: [149][ 1090/ 1236] Overall Loss 0.211634 Objective Loss 0.211634 LR 0.000250 Time 0.021008 +2023-10-05 21:52:37,432 - Epoch: [149][ 1100/ 1236] Overall Loss 0.211472 Objective Loss 0.211472 LR 0.000250 Time 0.021001 +2023-10-05 21:52:37,635 - Epoch: [149][ 1110/ 1236] Overall Loss 0.211828 Objective Loss 0.211828 LR 0.000250 Time 0.020994 +2023-10-05 21:52:37,838 - Epoch: [149][ 1120/ 1236] Overall Loss 0.211780 Objective Loss 0.211780 LR 0.000250 Time 0.020988 +2023-10-05 21:52:38,041 - Epoch: [149][ 1130/ 1236] Overall Loss 0.211771 Objective Loss 0.211771 LR 0.000250 Time 0.020981 +2023-10-05 21:52:38,244 - Epoch: [149][ 1140/ 1236] Overall Loss 0.211976 Objective Loss 0.211976 LR 0.000250 Time 0.020974 +2023-10-05 21:52:38,446 - Epoch: [149][ 1150/ 1236] Overall Loss 0.211757 Objective Loss 0.211757 LR 0.000250 Time 0.020968 +2023-10-05 21:52:38,650 - Epoch: [149][ 1160/ 1236] Overall Loss 0.211834 Objective Loss 0.211834 LR 0.000250 Time 0.020962 +2023-10-05 21:52:38,853 - Epoch: [149][ 1170/ 1236] Overall Loss 0.211811 Objective Loss 0.211811 LR 0.000250 Time 0.020956 +2023-10-05 21:52:39,056 - Epoch: [149][ 1180/ 1236] Overall Loss 0.211682 Objective Loss 0.211682 LR 0.000250 Time 0.020951 +2023-10-05 21:52:39,259 - Epoch: [149][ 1190/ 1236] Overall Loss 0.211453 Objective Loss 0.211453 LR 0.000250 Time 0.020945 +2023-10-05 21:52:39,462 - Epoch: [149][ 1200/ 1236] Overall Loss 0.211494 Objective Loss 0.211494 LR 0.000250 Time 0.020939 +2023-10-05 21:52:39,665 - Epoch: [149][ 1210/ 1236] Overall Loss 0.211583 Objective Loss 0.211583 LR 0.000250 Time 0.020933 +2023-10-05 21:52:39,867 - Epoch: [149][ 1220/ 1236] Overall Loss 0.211461 Objective Loss 0.211461 LR 0.000250 Time 0.020928 +2023-10-05 21:52:40,124 - Epoch: [149][ 1230/ 1236] Overall Loss 0.211505 Objective Loss 0.211505 LR 0.000250 Time 0.020966 +2023-10-05 21:52:40,242 - Epoch: [149][ 1236/ 1236] Overall Loss 0.211512 Objective Loss 0.211512 Top1 90.835031 Top5 98.167006 LR 0.000250 Time 0.020960 +2023-10-05 21:52:40,385 - --- validate (epoch=149)----------- +2023-10-05 21:52:40,385 - 29943 samples (256 per mini-batch) +2023-10-05 21:52:40,845 - Epoch: [149][ 10/ 117] Loss 0.315745 Top1 84.335938 Top5 98.007812 +2023-10-05 21:52:40,999 - Epoch: [149][ 20/ 117] Loss 0.325720 Top1 84.257812 Top5 97.949219 +2023-10-05 21:52:41,151 - Epoch: [149][ 30/ 117] Loss 0.316450 Top1 84.518229 Top5 98.125000 +2023-10-05 21:52:41,305 - Epoch: [149][ 40/ 117] Loss 0.321565 Top1 84.287109 Top5 98.115234 +2023-10-05 21:52:41,455 - Epoch: [149][ 50/ 117] Loss 0.317677 Top1 84.406250 Top5 98.085938 +2023-10-05 21:52:41,606 - Epoch: [149][ 60/ 117] Loss 0.319710 Top1 84.453125 Top5 98.027344 +2023-10-05 21:52:41,756 - Epoch: [149][ 70/ 117] Loss 0.322883 Top1 84.375000 Top5 97.929688 +2023-10-05 21:52:41,907 - Epoch: [149][ 80/ 117] Loss 0.319271 Top1 84.501953 Top5 97.949219 +2023-10-05 21:52:42,055 - Epoch: [149][ 90/ 117] Loss 0.318232 Top1 84.518229 Top5 97.973090 +2023-10-05 21:52:42,205 - Epoch: [149][ 100/ 117] Loss 0.314224 Top1 84.609375 Top5 98.000000 +2023-10-05 21:52:42,361 - Epoch: [149][ 110/ 117] Loss 0.314182 Top1 84.598722 Top5 98.000710 +2023-10-05 21:52:42,447 - Epoch: [149][ 117/ 117] Loss 0.313237 Top1 84.637478 Top5 98.006212 +2023-10-05 21:52:42,581 - ==> Top1: 84.637 Top5: 98.006 Loss: 0.313 + +2023-10-05 21:52:42,582 - ==> Confusion: +[[ 941 3 7 1 4 3 0 0 4 57 1 0 1 2 6 3 4 1 0 0 12] + [ 0 1045 3 0 6 26 2 24 1 0 3 2 0 1 0 3 3 0 7 2 3] + [ 4 1 952 19 2 0 25 6 0 0 6 1 9 2 1 3 1 1 5 6 12] + [ 2 0 9 976 1 4 1 2 1 1 7 1 2 1 30 4 0 5 24 3 15] + [ 30 6 1 0 960 6 0 0 0 9 2 1 0 2 8 3 12 1 0 2 7] + [ 3 27 1 0 1 1001 0 17 0 2 7 6 1 14 7 2 5 1 3 7 11] + [ 0 5 19 0 0 1 1129 6 0 1 0 2 2 0 1 6 0 0 4 9 6] + [ 2 13 13 0 3 32 5 1070 0 5 5 12 1 1 0 4 0 0 34 11 7] + [ 17 1 0 0 0 2 0 0 975 47 13 3 3 8 12 5 0 0 2 0 1] + [ 110 0 2 0 4 3 1 0 15 946 0 1 0 16 5 8 0 1 0 1 6] + [ 3 4 10 4 0 0 2 2 10 2 973 3 3 11 6 1 2 0 6 1 10] + [ 1 1 1 0 0 14 0 2 0 1 0 950 19 8 0 5 3 15 0 10 5] + [ 1 2 2 8 0 2 0 0 0 0 0 36 985 6 3 3 3 9 0 3 5] + [ 3 0 2 0 1 4 0 0 7 17 4 3 2 1062 3 2 1 0 0 1 7] + [ 13 0 3 9 4 1 0 0 30 0 2 2 2 2 1008 0 1 1 9 0 14] + [ 0 3 1 0 2 0 1 1 0 0 1 6 7 1 0 1072 16 10 0 11 2] + [ 0 12 1 0 4 3 0 1 2 0 0 4 0 0 2 9 1108 0 0 4 11] + [ 0 0 0 3 1 0 3 0 1 1 0 2 25 2 0 4 2 989 2 1 2] + [ 2 7 6 20 1 0 1 24 2 0 7 0 1 0 8 0 0 0 980 2 7] + [ 0 2 4 0 1 5 8 6 1 0 3 15 3 1 0 4 10 1 2 1082 4] + [ 129 164 139 64 88 152 41 103 107 84 175 102 381 282 136 58 153 56 121 231 5139]] + +2023-10-05 21:52:42,584 - ==> Best [Top1: 85.068 Top5: 98.030 Sparsity:0.00 Params: 148928 on epoch: 148] +2023-10-05 21:52:42,584 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:52:42,590 - + +2023-10-05 21:52:42,590 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:52:43,719 - Epoch: [150][ 10/ 1236] Overall Loss 0.214582 Objective Loss 0.214582 LR 0.000250 Time 0.112857 +2023-10-05 21:52:43,923 - Epoch: [150][ 20/ 1236] Overall Loss 0.209998 Objective Loss 0.209998 LR 0.000250 Time 0.066623 +2023-10-05 21:52:44,126 - Epoch: [150][ 30/ 1236] Overall Loss 0.211335 Objective Loss 0.211335 LR 0.000250 Time 0.051160 +2023-10-05 21:52:44,330 - Epoch: [150][ 40/ 1236] Overall Loss 0.217928 Objective Loss 0.217928 LR 0.000250 Time 0.043465 +2023-10-05 21:52:44,534 - Epoch: [150][ 50/ 1236] Overall Loss 0.214976 Objective Loss 0.214976 LR 0.000250 Time 0.038836 +2023-10-05 21:52:44,738 - Epoch: [150][ 60/ 1236] Overall Loss 0.215879 Objective Loss 0.215879 LR 0.000250 Time 0.035768 +2023-10-05 21:52:44,942 - Epoch: [150][ 70/ 1236] Overall Loss 0.212690 Objective Loss 0.212690 LR 0.000250 Time 0.033559 +2023-10-05 21:52:45,144 - Epoch: [150][ 80/ 1236] Overall Loss 0.209869 Objective Loss 0.209869 LR 0.000250 Time 0.031890 +2023-10-05 21:52:45,347 - Epoch: [150][ 90/ 1236] Overall Loss 0.209306 Objective Loss 0.209306 LR 0.000250 Time 0.030599 +2023-10-05 21:52:45,548 - Epoch: [150][ 100/ 1236] Overall Loss 0.210955 Objective Loss 0.210955 LR 0.000250 Time 0.029546 +2023-10-05 21:52:45,750 - Epoch: [150][ 110/ 1236] Overall Loss 0.210385 Objective Loss 0.210385 LR 0.000250 Time 0.028696 +2023-10-05 21:52:45,957 - Epoch: [150][ 120/ 1236] Overall Loss 0.209285 Objective Loss 0.209285 LR 0.000250 Time 0.028027 +2023-10-05 21:52:46,165 - Epoch: [150][ 130/ 1236] Overall Loss 0.208749 Objective Loss 0.208749 LR 0.000250 Time 0.027467 +2023-10-05 21:52:46,372 - Epoch: [150][ 140/ 1236] Overall Loss 0.209425 Objective Loss 0.209425 LR 0.000250 Time 0.026981 +2023-10-05 21:52:46,580 - Epoch: [150][ 150/ 1236] Overall Loss 0.207980 Objective Loss 0.207980 LR 0.000250 Time 0.026566 +2023-10-05 21:52:46,787 - Epoch: [150][ 160/ 1236] Overall Loss 0.208177 Objective Loss 0.208177 LR 0.000250 Time 0.026197 +2023-10-05 21:52:46,995 - Epoch: [150][ 170/ 1236] Overall Loss 0.209356 Objective Loss 0.209356 LR 0.000250 Time 0.025876 +2023-10-05 21:52:47,202 - Epoch: [150][ 180/ 1236] Overall Loss 0.209801 Objective Loss 0.209801 LR 0.000250 Time 0.025585 +2023-10-05 21:52:47,410 - Epoch: [150][ 190/ 1236] Overall Loss 0.209269 Objective Loss 0.209269 LR 0.000250 Time 0.025331 +2023-10-05 21:52:47,615 - Epoch: [150][ 200/ 1236] Overall Loss 0.208238 Objective Loss 0.208238 LR 0.000250 Time 0.025091 +2023-10-05 21:52:47,823 - Epoch: [150][ 210/ 1236] Overall Loss 0.207888 Objective Loss 0.207888 LR 0.000250 Time 0.024884 +2023-10-05 21:52:48,030 - Epoch: [150][ 220/ 1236] Overall Loss 0.208198 Objective Loss 0.208198 LR 0.000250 Time 0.024690 +2023-10-05 21:52:48,238 - Epoch: [150][ 230/ 1236] Overall Loss 0.207582 Objective Loss 0.207582 LR 0.000250 Time 0.024520 +2023-10-05 21:52:48,445 - Epoch: [150][ 240/ 1236] Overall Loss 0.207277 Objective Loss 0.207277 LR 0.000250 Time 0.024360 +2023-10-05 21:52:48,653 - Epoch: [150][ 250/ 1236] Overall Loss 0.207141 Objective Loss 0.207141 LR 0.000250 Time 0.024216 +2023-10-05 21:52:48,860 - Epoch: [150][ 260/ 1236] Overall Loss 0.207326 Objective Loss 0.207326 LR 0.000250 Time 0.024079 +2023-10-05 21:52:49,068 - Epoch: [150][ 270/ 1236] Overall Loss 0.207453 Objective Loss 0.207453 LR 0.000250 Time 0.023956 +2023-10-05 21:52:49,274 - Epoch: [150][ 280/ 1236] Overall Loss 0.207786 Objective Loss 0.207786 LR 0.000250 Time 0.023838 +2023-10-05 21:52:49,482 - Epoch: [150][ 290/ 1236] Overall Loss 0.207487 Objective Loss 0.207487 LR 0.000250 Time 0.023732 +2023-10-05 21:52:49,690 - Epoch: [150][ 300/ 1236] Overall Loss 0.207877 Objective Loss 0.207877 LR 0.000250 Time 0.023630 +2023-10-05 21:52:49,898 - Epoch: [150][ 310/ 1236] Overall Loss 0.208057 Objective Loss 0.208057 LR 0.000250 Time 0.023538 +2023-10-05 21:52:50,105 - Epoch: [150][ 320/ 1236] Overall Loss 0.207898 Objective Loss 0.207898 LR 0.000250 Time 0.023448 +2023-10-05 21:52:50,311 - Epoch: [150][ 330/ 1236] Overall Loss 0.208831 Objective Loss 0.208831 LR 0.000250 Time 0.023362 +2023-10-05 21:52:50,516 - Epoch: [150][ 340/ 1236] Overall Loss 0.208549 Objective Loss 0.208549 LR 0.000250 Time 0.023278 +2023-10-05 21:52:50,723 - Epoch: [150][ 350/ 1236] Overall Loss 0.208052 Objective Loss 0.208052 LR 0.000250 Time 0.023202 +2023-10-05 21:52:50,928 - Epoch: [150][ 360/ 1236] Overall Loss 0.208088 Objective Loss 0.208088 LR 0.000250 Time 0.023127 +2023-10-05 21:52:51,134 - Epoch: [150][ 370/ 1236] Overall Loss 0.208382 Objective Loss 0.208382 LR 0.000250 Time 0.023058 +2023-10-05 21:52:51,340 - Epoch: [150][ 380/ 1236] Overall Loss 0.208532 Objective Loss 0.208532 LR 0.000250 Time 0.022992 +2023-10-05 21:52:51,546 - Epoch: [150][ 390/ 1236] Overall Loss 0.208699 Objective Loss 0.208699 LR 0.000250 Time 0.022930 +2023-10-05 21:52:51,752 - Epoch: [150][ 400/ 1236] Overall Loss 0.208771 Objective Loss 0.208771 LR 0.000250 Time 0.022871 +2023-10-05 21:52:51,958 - Epoch: [150][ 410/ 1236] Overall Loss 0.208381 Objective Loss 0.208381 LR 0.000250 Time 0.022814 +2023-10-05 21:52:52,164 - Epoch: [150][ 420/ 1236] Overall Loss 0.208913 Objective Loss 0.208913 LR 0.000250 Time 0.022759 +2023-10-05 21:52:52,370 - Epoch: [150][ 430/ 1236] Overall Loss 0.209911 Objective Loss 0.209911 LR 0.000250 Time 0.022709 +2023-10-05 21:52:52,575 - Epoch: [150][ 440/ 1236] Overall Loss 0.209496 Objective Loss 0.209496 LR 0.000250 Time 0.022659 +2023-10-05 21:52:52,782 - Epoch: [150][ 450/ 1236] Overall Loss 0.209406 Objective Loss 0.209406 LR 0.000250 Time 0.022613 +2023-10-05 21:52:52,987 - Epoch: [150][ 460/ 1236] Overall Loss 0.210171 Objective Loss 0.210171 LR 0.000250 Time 0.022567 +2023-10-05 21:52:53,194 - Epoch: [150][ 470/ 1236] Overall Loss 0.210192 Objective Loss 0.210192 LR 0.000250 Time 0.022526 +2023-10-05 21:52:53,399 - Epoch: [150][ 480/ 1236] Overall Loss 0.209932 Objective Loss 0.209932 LR 0.000250 Time 0.022484 +2023-10-05 21:52:53,606 - Epoch: [150][ 490/ 1236] Overall Loss 0.209879 Objective Loss 0.209879 LR 0.000250 Time 0.022446 +2023-10-05 21:52:53,811 - Epoch: [150][ 500/ 1236] Overall Loss 0.209956 Objective Loss 0.209956 LR 0.000250 Time 0.022407 +2023-10-05 21:52:54,018 - Epoch: [150][ 510/ 1236] Overall Loss 0.209646 Objective Loss 0.209646 LR 0.000250 Time 0.022371 +2023-10-05 21:52:54,223 - Epoch: [150][ 520/ 1236] Overall Loss 0.209284 Objective Loss 0.209284 LR 0.000250 Time 0.022336 +2023-10-05 21:52:54,431 - Epoch: [150][ 530/ 1236] Overall Loss 0.209562 Objective Loss 0.209562 LR 0.000250 Time 0.022305 +2023-10-05 21:52:54,643 - Epoch: [150][ 540/ 1236] Overall Loss 0.209713 Objective Loss 0.209713 LR 0.000250 Time 0.022285 +2023-10-05 21:52:54,852 - Epoch: [150][ 550/ 1236] Overall Loss 0.209877 Objective Loss 0.209877 LR 0.000250 Time 0.022260 +2023-10-05 21:52:55,065 - Epoch: [150][ 560/ 1236] Overall Loss 0.209911 Objective Loss 0.209911 LR 0.000250 Time 0.022242 +2023-10-05 21:52:55,274 - Epoch: [150][ 570/ 1236] Overall Loss 0.209698 Objective Loss 0.209698 LR 0.000250 Time 0.022217 +2023-10-05 21:52:55,487 - Epoch: [150][ 580/ 1236] Overall Loss 0.209592 Objective Loss 0.209592 LR 0.000250 Time 0.022201 +2023-10-05 21:52:55,696 - Epoch: [150][ 590/ 1236] Overall Loss 0.209464 Objective Loss 0.209464 LR 0.000250 Time 0.022178 +2023-10-05 21:52:55,910 - Epoch: [150][ 600/ 1236] Overall Loss 0.209129 Objective Loss 0.209129 LR 0.000250 Time 0.022163 +2023-10-05 21:52:56,118 - Epoch: [150][ 610/ 1236] Overall Loss 0.208753 Objective Loss 0.208753 LR 0.000250 Time 0.022142 +2023-10-05 21:52:56,332 - Epoch: [150][ 620/ 1236] Overall Loss 0.209142 Objective Loss 0.209142 LR 0.000250 Time 0.022129 +2023-10-05 21:52:56,541 - Epoch: [150][ 630/ 1236] Overall Loss 0.209097 Objective Loss 0.209097 LR 0.000250 Time 0.022108 +2023-10-05 21:52:56,754 - Epoch: [150][ 640/ 1236] Overall Loss 0.209531 Objective Loss 0.209531 LR 0.000250 Time 0.022095 +2023-10-05 21:52:56,963 - Epoch: [150][ 650/ 1236] Overall Loss 0.209155 Objective Loss 0.209155 LR 0.000250 Time 0.022076 +2023-10-05 21:52:57,176 - Epoch: [150][ 660/ 1236] Overall Loss 0.209505 Objective Loss 0.209505 LR 0.000250 Time 0.022065 +2023-10-05 21:52:57,385 - Epoch: [150][ 670/ 1236] Overall Loss 0.209702 Objective Loss 0.209702 LR 0.000250 Time 0.022047 +2023-10-05 21:52:57,598 - Epoch: [150][ 680/ 1236] Overall Loss 0.210042 Objective Loss 0.210042 LR 0.000250 Time 0.022036 +2023-10-05 21:52:57,800 - Epoch: [150][ 690/ 1236] Overall Loss 0.209851 Objective Loss 0.209851 LR 0.000250 Time 0.022009 +2023-10-05 21:52:58,004 - Epoch: [150][ 700/ 1236] Overall Loss 0.209877 Objective Loss 0.209877 LR 0.000250 Time 0.021985 +2023-10-05 21:52:58,205 - Epoch: [150][ 710/ 1236] Overall Loss 0.210128 Objective Loss 0.210128 LR 0.000250 Time 0.021958 +2023-10-05 21:52:58,409 - Epoch: [150][ 720/ 1236] Overall Loss 0.210143 Objective Loss 0.210143 LR 0.000250 Time 0.021936 +2023-10-05 21:52:58,611 - Epoch: [150][ 730/ 1236] Overall Loss 0.210121 Objective Loss 0.210121 LR 0.000250 Time 0.021911 +2023-10-05 21:52:58,814 - Epoch: [150][ 740/ 1236] Overall Loss 0.210212 Objective Loss 0.210212 LR 0.000250 Time 0.021889 +2023-10-05 21:52:59,016 - Epoch: [150][ 750/ 1236] Overall Loss 0.210395 Objective Loss 0.210395 LR 0.000250 Time 0.021865 +2023-10-05 21:52:59,219 - Epoch: [150][ 760/ 1236] Overall Loss 0.210221 Objective Loss 0.210221 LR 0.000250 Time 0.021844 +2023-10-05 21:52:59,421 - Epoch: [150][ 770/ 1236] Overall Loss 0.210275 Objective Loss 0.210275 LR 0.000250 Time 0.021822 +2023-10-05 21:52:59,625 - Epoch: [150][ 780/ 1236] Overall Loss 0.210300 Objective Loss 0.210300 LR 0.000250 Time 0.021803 +2023-10-05 21:52:59,826 - Epoch: [150][ 790/ 1236] Overall Loss 0.210702 Objective Loss 0.210702 LR 0.000250 Time 0.021782 +2023-10-05 21:53:00,030 - Epoch: [150][ 800/ 1236] Overall Loss 0.210790 Objective Loss 0.210790 LR 0.000250 Time 0.021764 +2023-10-05 21:53:00,231 - Epoch: [150][ 810/ 1236] Overall Loss 0.210964 Objective Loss 0.210964 LR 0.000250 Time 0.021744 +2023-10-05 21:53:00,435 - Epoch: [150][ 820/ 1236] Overall Loss 0.210740 Objective Loss 0.210740 LR 0.000250 Time 0.021726 +2023-10-05 21:53:00,636 - Epoch: [150][ 830/ 1236] Overall Loss 0.211019 Objective Loss 0.211019 LR 0.000250 Time 0.021707 +2023-10-05 21:53:00,840 - Epoch: [150][ 840/ 1236] Overall Loss 0.210702 Objective Loss 0.210702 LR 0.000250 Time 0.021690 +2023-10-05 21:53:01,042 - Epoch: [150][ 850/ 1236] Overall Loss 0.210687 Objective Loss 0.210687 LR 0.000250 Time 0.021672 +2023-10-05 21:53:01,245 - Epoch: [150][ 860/ 1236] Overall Loss 0.210644 Objective Loss 0.210644 LR 0.000250 Time 0.021656 +2023-10-05 21:53:01,447 - Epoch: [150][ 870/ 1236] Overall Loss 0.210259 Objective Loss 0.210259 LR 0.000250 Time 0.021639 +2023-10-05 21:53:01,650 - Epoch: [150][ 880/ 1236] Overall Loss 0.210007 Objective Loss 0.210007 LR 0.000250 Time 0.021624 +2023-10-05 21:53:01,852 - Epoch: [150][ 890/ 1236] Overall Loss 0.209925 Objective Loss 0.209925 LR 0.000250 Time 0.021607 +2023-10-05 21:53:02,055 - Epoch: [150][ 900/ 1236] Overall Loss 0.210205 Objective Loss 0.210205 LR 0.000250 Time 0.021592 +2023-10-05 21:53:02,257 - Epoch: [150][ 910/ 1236] Overall Loss 0.210296 Objective Loss 0.210296 LR 0.000250 Time 0.021576 +2023-10-05 21:53:02,460 - Epoch: [150][ 920/ 1236] Overall Loss 0.210550 Objective Loss 0.210550 LR 0.000250 Time 0.021562 +2023-10-05 21:53:02,662 - Epoch: [150][ 930/ 1236] Overall Loss 0.210721 Objective Loss 0.210721 LR 0.000250 Time 0.021547 +2023-10-05 21:53:02,866 - Epoch: [150][ 940/ 1236] Overall Loss 0.210685 Objective Loss 0.210685 LR 0.000250 Time 0.021534 +2023-10-05 21:53:03,067 - Epoch: [150][ 950/ 1236] Overall Loss 0.210510 Objective Loss 0.210510 LR 0.000250 Time 0.021519 +2023-10-05 21:53:03,271 - Epoch: [150][ 960/ 1236] Overall Loss 0.210407 Objective Loss 0.210407 LR 0.000250 Time 0.021507 +2023-10-05 21:53:03,472 - Epoch: [150][ 970/ 1236] Overall Loss 0.210355 Objective Loss 0.210355 LR 0.000250 Time 0.021492 +2023-10-05 21:53:03,675 - Epoch: [150][ 980/ 1236] Overall Loss 0.210377 Objective Loss 0.210377 LR 0.000250 Time 0.021480 +2023-10-05 21:53:03,877 - Epoch: [150][ 990/ 1236] Overall Loss 0.210514 Objective Loss 0.210514 LR 0.000250 Time 0.021466 +2023-10-05 21:53:04,080 - Epoch: [150][ 1000/ 1236] Overall Loss 0.210451 Objective Loss 0.210451 LR 0.000250 Time 0.021455 +2023-10-05 21:53:04,282 - Epoch: [150][ 1010/ 1236] Overall Loss 0.210238 Objective Loss 0.210238 LR 0.000250 Time 0.021442 +2023-10-05 21:53:04,485 - Epoch: [150][ 1020/ 1236] Overall Loss 0.210184 Objective Loss 0.210184 LR 0.000250 Time 0.021430 +2023-10-05 21:53:04,687 - Epoch: [150][ 1030/ 1236] Overall Loss 0.210122 Objective Loss 0.210122 LR 0.000250 Time 0.021418 +2023-10-05 21:53:04,891 - Epoch: [150][ 1040/ 1236] Overall Loss 0.210027 Objective Loss 0.210027 LR 0.000250 Time 0.021408 +2023-10-05 21:53:05,092 - Epoch: [150][ 1050/ 1236] Overall Loss 0.209904 Objective Loss 0.209904 LR 0.000250 Time 0.021395 +2023-10-05 21:53:05,295 - Epoch: [150][ 1060/ 1236] Overall Loss 0.209745 Objective Loss 0.209745 LR 0.000250 Time 0.021385 +2023-10-05 21:53:05,497 - Epoch: [150][ 1070/ 1236] Overall Loss 0.209813 Objective Loss 0.209813 LR 0.000250 Time 0.021373 +2023-10-05 21:53:05,700 - Epoch: [150][ 1080/ 1236] Overall Loss 0.209702 Objective Loss 0.209702 LR 0.000250 Time 0.021363 +2023-10-05 21:53:05,902 - Epoch: [150][ 1090/ 1236] Overall Loss 0.209654 Objective Loss 0.209654 LR 0.000250 Time 0.021351 +2023-10-05 21:53:06,105 - Epoch: [150][ 1100/ 1236] Overall Loss 0.209567 Objective Loss 0.209567 LR 0.000250 Time 0.021342 +2023-10-05 21:53:06,307 - Epoch: [150][ 1110/ 1236] Overall Loss 0.209584 Objective Loss 0.209584 LR 0.000250 Time 0.021331 +2023-10-05 21:53:06,514 - Epoch: [150][ 1120/ 1236] Overall Loss 0.209432 Objective Loss 0.209432 LR 0.000250 Time 0.021325 +2023-10-05 21:53:06,724 - Epoch: [150][ 1130/ 1236] Overall Loss 0.209291 Objective Loss 0.209291 LR 0.000250 Time 0.021322 +2023-10-05 21:53:06,938 - Epoch: [150][ 1140/ 1236] Overall Loss 0.209430 Objective Loss 0.209430 LR 0.000250 Time 0.021323 +2023-10-05 21:53:07,148 - Epoch: [150][ 1150/ 1236] Overall Loss 0.209462 Objective Loss 0.209462 LR 0.000250 Time 0.021319 +2023-10-05 21:53:07,362 - Epoch: [150][ 1160/ 1236] Overall Loss 0.209329 Objective Loss 0.209329 LR 0.000250 Time 0.021320 +2023-10-05 21:53:07,572 - Epoch: [150][ 1170/ 1236] Overall Loss 0.209372 Objective Loss 0.209372 LR 0.000250 Time 0.021317 +2023-10-05 21:53:07,787 - Epoch: [150][ 1180/ 1236] Overall Loss 0.209239 Objective Loss 0.209239 LR 0.000250 Time 0.021318 +2023-10-05 21:53:07,996 - Epoch: [150][ 1190/ 1236] Overall Loss 0.209383 Objective Loss 0.209383 LR 0.000250 Time 0.021314 +2023-10-05 21:53:08,210 - Epoch: [150][ 1200/ 1236] Overall Loss 0.209431 Objective Loss 0.209431 LR 0.000250 Time 0.021315 +2023-10-05 21:53:08,420 - Epoch: [150][ 1210/ 1236] Overall Loss 0.209596 Objective Loss 0.209596 LR 0.000250 Time 0.021312 +2023-10-05 21:53:08,626 - Epoch: [150][ 1220/ 1236] Overall Loss 0.209531 Objective Loss 0.209531 LR 0.000250 Time 0.021306 +2023-10-05 21:53:08,882 - Epoch: [150][ 1230/ 1236] Overall Loss 0.209510 Objective Loss 0.209510 LR 0.000250 Time 0.021341 +2023-10-05 21:53:09,001 - Epoch: [150][ 1236/ 1236] Overall Loss 0.209499 Objective Loss 0.209499 Top1 86.558045 Top5 97.963340 LR 0.000250 Time 0.021333 +2023-10-05 21:53:09,136 - --- validate (epoch=150)----------- +2023-10-05 21:53:09,137 - 29943 samples (256 per mini-batch) +2023-10-05 21:53:09,600 - Epoch: [150][ 10/ 117] Loss 0.323151 Top1 84.921875 Top5 97.812500 +2023-10-05 21:53:09,759 - Epoch: [150][ 20/ 117] Loss 0.334935 Top1 84.511719 Top5 97.773438 +2023-10-05 21:53:09,915 - Epoch: [150][ 30/ 117] Loss 0.335031 Top1 84.114583 Top5 97.968750 +2023-10-05 21:53:10,073 - Epoch: [150][ 40/ 117] Loss 0.325969 Top1 84.326172 Top5 97.822266 +2023-10-05 21:53:10,228 - Epoch: [150][ 50/ 117] Loss 0.320464 Top1 84.531250 Top5 97.835938 +2023-10-05 21:53:10,385 - Epoch: [150][ 60/ 117] Loss 0.317290 Top1 84.674479 Top5 97.877604 +2023-10-05 21:53:10,540 - Epoch: [150][ 70/ 117] Loss 0.309225 Top1 84.748884 Top5 97.912946 +2023-10-05 21:53:10,697 - Epoch: [150][ 80/ 117] Loss 0.309767 Top1 84.775391 Top5 97.915039 +2023-10-05 21:53:10,852 - Epoch: [150][ 90/ 117] Loss 0.309564 Top1 84.843750 Top5 97.981771 +2023-10-05 21:53:11,005 - Epoch: [150][ 100/ 117] Loss 0.305029 Top1 84.824219 Top5 98.000000 +2023-10-05 21:53:11,167 - Epoch: [150][ 110/ 117] Loss 0.303724 Top1 84.939631 Top5 97.993608 +2023-10-05 21:53:11,253 - Epoch: [150][ 117/ 117] Loss 0.304337 Top1 84.921351 Top5 98.002872 +2023-10-05 21:53:11,354 - ==> Top1: 84.921 Top5: 98.003 Loss: 0.304 + +2023-10-05 21:53:11,355 - ==> Confusion: +[[ 933 4 3 1 8 3 0 0 6 65 1 0 1 2 5 2 2 2 1 0 11] + [ 0 1050 2 0 10 20 1 24 1 0 1 1 0 0 0 4 1 0 8 2 6] + [ 6 0 966 10 3 0 21 7 0 2 2 3 6 1 1 4 1 3 6 5 9] + [ 5 1 20 965 0 3 0 2 0 1 6 1 6 3 26 3 0 5 31 1 10] + [ 19 6 2 1 977 4 0 1 0 9 0 1 0 3 8 5 8 1 0 2 3] + [ 3 32 0 0 2 984 0 23 2 1 6 11 0 18 6 1 6 0 4 5 12] + [ 0 5 24 0 0 1 1128 7 0 1 2 2 3 1 1 7 0 0 2 3 4] + [ 3 16 13 0 2 24 3 1093 0 5 3 10 0 2 0 2 0 1 28 6 7] + [ 21 2 0 0 0 2 0 0 967 44 12 1 2 12 11 6 1 0 5 0 3] + [ 96 1 3 1 4 3 0 0 19 954 0 1 0 18 4 7 0 2 0 1 5] + [ 2 4 6 5 0 0 1 4 8 2 982 2 0 13 3 3 2 0 5 2 9] + [ 1 0 1 0 0 11 0 2 0 1 1 962 18 8 0 5 1 16 0 5 3] + [ 0 1 5 4 0 1 1 1 1 1 1 31 991 0 1 7 2 12 3 2 3] + [ 2 0 2 0 1 4 0 0 5 17 8 5 2 1065 2 1 0 0 0 1 4] + [ 12 0 4 10 6 0 0 0 22 1 4 1 3 2 1011 0 2 1 10 0 12] + [ 1 4 1 1 3 0 2 0 0 0 0 7 6 1 0 1076 11 11 0 9 1] + [ 0 13 1 0 6 2 0 2 3 0 0 4 0 2 2 10 1102 0 1 3 10] + [ 0 0 0 4 1 0 2 0 0 0 0 2 17 2 0 4 0 1004 1 1 0] + [ 1 6 7 17 1 0 0 20 2 1 3 1 0 0 7 0 0 0 993 2 7] + [ 0 4 3 2 1 6 8 7 0 0 1 15 4 4 0 6 8 1 4 1075 3] + [ 142 163 174 52 109 108 39 93 95 85 184 108 351 313 124 68 150 63 147 187 5150]] + +2023-10-05 21:53:11,356 - ==> Best [Top1: 85.068 Top5: 98.030 Sparsity:0.00 Params: 148928 on epoch: 148] +2023-10-05 21:53:11,357 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:53:11,362 - + +2023-10-05 21:53:11,363 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:53:12,372 - Epoch: [151][ 10/ 1236] Overall Loss 0.206151 Objective Loss 0.206151 LR 0.000250 Time 0.100935 +2023-10-05 21:53:12,577 - Epoch: [151][ 20/ 1236] Overall Loss 0.210452 Objective Loss 0.210452 LR 0.000250 Time 0.060673 +2023-10-05 21:53:12,779 - Epoch: [151][ 30/ 1236] Overall Loss 0.208493 Objective Loss 0.208493 LR 0.000250 Time 0.047157 +2023-10-05 21:53:12,981 - Epoch: [151][ 40/ 1236] Overall Loss 0.204355 Objective Loss 0.204355 LR 0.000250 Time 0.040422 +2023-10-05 21:53:13,183 - Epoch: [151][ 50/ 1236] Overall Loss 0.206551 Objective Loss 0.206551 LR 0.000250 Time 0.036361 +2023-10-05 21:53:13,385 - Epoch: [151][ 60/ 1236] Overall Loss 0.205818 Objective Loss 0.205818 LR 0.000250 Time 0.033676 +2023-10-05 21:53:13,587 - Epoch: [151][ 70/ 1236] Overall Loss 0.202471 Objective Loss 0.202471 LR 0.000250 Time 0.031745 +2023-10-05 21:53:13,791 - Epoch: [151][ 80/ 1236] Overall Loss 0.203361 Objective Loss 0.203361 LR 0.000250 Time 0.030313 +2023-10-05 21:53:13,992 - Epoch: [151][ 90/ 1236] Overall Loss 0.204985 Objective Loss 0.204985 LR 0.000250 Time 0.029182 +2023-10-05 21:53:14,194 - Epoch: [151][ 100/ 1236] Overall Loss 0.205596 Objective Loss 0.205596 LR 0.000250 Time 0.028278 +2023-10-05 21:53:14,395 - Epoch: [151][ 110/ 1236] Overall Loss 0.205560 Objective Loss 0.205560 LR 0.000250 Time 0.027535 +2023-10-05 21:53:14,596 - Epoch: [151][ 120/ 1236] Overall Loss 0.207930 Objective Loss 0.207930 LR 0.000250 Time 0.026905 +2023-10-05 21:53:14,797 - Epoch: [151][ 130/ 1236] Overall Loss 0.208061 Objective Loss 0.208061 LR 0.000250 Time 0.026381 +2023-10-05 21:53:14,998 - Epoch: [151][ 140/ 1236] Overall Loss 0.208905 Objective Loss 0.208905 LR 0.000250 Time 0.025934 +2023-10-05 21:53:15,200 - Epoch: [151][ 150/ 1236] Overall Loss 0.208398 Objective Loss 0.208398 LR 0.000250 Time 0.025544 +2023-10-05 21:53:15,402 - Epoch: [151][ 160/ 1236] Overall Loss 0.208242 Objective Loss 0.208242 LR 0.000250 Time 0.025213 +2023-10-05 21:53:15,603 - Epoch: [151][ 170/ 1236] Overall Loss 0.207984 Objective Loss 0.207984 LR 0.000250 Time 0.024908 +2023-10-05 21:53:15,806 - Epoch: [151][ 180/ 1236] Overall Loss 0.208145 Objective Loss 0.208145 LR 0.000250 Time 0.024649 +2023-10-05 21:53:16,006 - Epoch: [151][ 190/ 1236] Overall Loss 0.208333 Objective Loss 0.208333 LR 0.000250 Time 0.024404 +2023-10-05 21:53:16,209 - Epoch: [151][ 200/ 1236] Overall Loss 0.207844 Objective Loss 0.207844 LR 0.000250 Time 0.024195 +2023-10-05 21:53:16,410 - Epoch: [151][ 210/ 1236] Overall Loss 0.207758 Objective Loss 0.207758 LR 0.000250 Time 0.023997 +2023-10-05 21:53:16,612 - Epoch: [151][ 220/ 1236] Overall Loss 0.207147 Objective Loss 0.207147 LR 0.000250 Time 0.023826 +2023-10-05 21:53:16,813 - Epoch: [151][ 230/ 1236] Overall Loss 0.206961 Objective Loss 0.206961 LR 0.000250 Time 0.023661 +2023-10-05 21:53:17,015 - Epoch: [151][ 240/ 1236] Overall Loss 0.206699 Objective Loss 0.206699 LR 0.000250 Time 0.023515 +2023-10-05 21:53:17,215 - Epoch: [151][ 250/ 1236] Overall Loss 0.206233 Objective Loss 0.206233 LR 0.000250 Time 0.023375 +2023-10-05 21:53:17,418 - Epoch: [151][ 260/ 1236] Overall Loss 0.206019 Objective Loss 0.206019 LR 0.000250 Time 0.023255 +2023-10-05 21:53:17,619 - Epoch: [151][ 270/ 1236] Overall Loss 0.205862 Objective Loss 0.205862 LR 0.000250 Time 0.023136 +2023-10-05 21:53:17,822 - Epoch: [151][ 280/ 1236] Overall Loss 0.206429 Objective Loss 0.206429 LR 0.000250 Time 0.023032 +2023-10-05 21:53:18,022 - Epoch: [151][ 290/ 1236] Overall Loss 0.206093 Objective Loss 0.206093 LR 0.000250 Time 0.022927 +2023-10-05 21:53:18,224 - Epoch: [151][ 300/ 1236] Overall Loss 0.205845 Objective Loss 0.205845 LR 0.000250 Time 0.022836 +2023-10-05 21:53:18,425 - Epoch: [151][ 310/ 1236] Overall Loss 0.205780 Objective Loss 0.205780 LR 0.000250 Time 0.022744 +2023-10-05 21:53:18,627 - Epoch: [151][ 320/ 1236] Overall Loss 0.205986 Objective Loss 0.205986 LR 0.000250 Time 0.022666 +2023-10-05 21:53:18,828 - Epoch: [151][ 330/ 1236] Overall Loss 0.205804 Objective Loss 0.205804 LR 0.000250 Time 0.022587 +2023-10-05 21:53:19,031 - Epoch: [151][ 340/ 1236] Overall Loss 0.206543 Objective Loss 0.206543 LR 0.000250 Time 0.022519 +2023-10-05 21:53:19,232 - Epoch: [151][ 350/ 1236] Overall Loss 0.206113 Objective Loss 0.206113 LR 0.000250 Time 0.022448 +2023-10-05 21:53:19,435 - Epoch: [151][ 360/ 1236] Overall Loss 0.206505 Objective Loss 0.206505 LR 0.000250 Time 0.022387 +2023-10-05 21:53:19,636 - Epoch: [151][ 370/ 1236] Overall Loss 0.206532 Objective Loss 0.206532 LR 0.000250 Time 0.022325 +2023-10-05 21:53:19,841 - Epoch: [151][ 380/ 1236] Overall Loss 0.206091 Objective Loss 0.206091 LR 0.000250 Time 0.022275 +2023-10-05 21:53:20,044 - Epoch: [151][ 390/ 1236] Overall Loss 0.206054 Objective Loss 0.206054 LR 0.000250 Time 0.022223 +2023-10-05 21:53:20,249 - Epoch: [151][ 400/ 1236] Overall Loss 0.205729 Objective Loss 0.205729 LR 0.000250 Time 0.022179 +2023-10-05 21:53:20,452 - Epoch: [151][ 410/ 1236] Overall Loss 0.205793 Objective Loss 0.205793 LR 0.000250 Time 0.022132 +2023-10-05 21:53:20,656 - Epoch: [151][ 420/ 1236] Overall Loss 0.205631 Objective Loss 0.205631 LR 0.000250 Time 0.022090 +2023-10-05 21:53:20,858 - Epoch: [151][ 430/ 1236] Overall Loss 0.205231 Objective Loss 0.205231 LR 0.000250 Time 0.022045 +2023-10-05 21:53:21,061 - Epoch: [151][ 440/ 1236] Overall Loss 0.205920 Objective Loss 0.205920 LR 0.000250 Time 0.022006 +2023-10-05 21:53:21,263 - Epoch: [151][ 450/ 1236] Overall Loss 0.206185 Objective Loss 0.206185 LR 0.000250 Time 0.021963 +2023-10-05 21:53:21,467 - Epoch: [151][ 460/ 1236] Overall Loss 0.206013 Objective Loss 0.206013 LR 0.000250 Time 0.021929 +2023-10-05 21:53:21,668 - Epoch: [151][ 470/ 1236] Overall Loss 0.206512 Objective Loss 0.206512 LR 0.000250 Time 0.021890 +2023-10-05 21:53:21,872 - Epoch: [151][ 480/ 1236] Overall Loss 0.206517 Objective Loss 0.206517 LR 0.000250 Time 0.021858 +2023-10-05 21:53:22,074 - Epoch: [151][ 490/ 1236] Overall Loss 0.206625 Objective Loss 0.206625 LR 0.000250 Time 0.021822 +2023-10-05 21:53:22,277 - Epoch: [151][ 500/ 1236] Overall Loss 0.206641 Objective Loss 0.206641 LR 0.000250 Time 0.021793 +2023-10-05 21:53:22,479 - Epoch: [151][ 510/ 1236] Overall Loss 0.206558 Objective Loss 0.206558 LR 0.000250 Time 0.021759 +2023-10-05 21:53:22,683 - Epoch: [151][ 520/ 1236] Overall Loss 0.206854 Objective Loss 0.206854 LR 0.000250 Time 0.021733 +2023-10-05 21:53:22,884 - Epoch: [151][ 530/ 1236] Overall Loss 0.207542 Objective Loss 0.207542 LR 0.000250 Time 0.021702 +2023-10-05 21:53:23,088 - Epoch: [151][ 540/ 1236] Overall Loss 0.207870 Objective Loss 0.207870 LR 0.000250 Time 0.021677 +2023-10-05 21:53:23,289 - Epoch: [151][ 550/ 1236] Overall Loss 0.207857 Objective Loss 0.207857 LR 0.000250 Time 0.021648 +2023-10-05 21:53:23,493 - Epoch: [151][ 560/ 1236] Overall Loss 0.207918 Objective Loss 0.207918 LR 0.000250 Time 0.021625 +2023-10-05 21:53:23,695 - Epoch: [151][ 570/ 1236] Overall Loss 0.208120 Objective Loss 0.208120 LR 0.000250 Time 0.021599 +2023-10-05 21:53:23,899 - Epoch: [151][ 580/ 1236] Overall Loss 0.208522 Objective Loss 0.208522 LR 0.000250 Time 0.021577 +2023-10-05 21:53:24,100 - Epoch: [151][ 590/ 1236] Overall Loss 0.208069 Objective Loss 0.208069 LR 0.000250 Time 0.021552 +2023-10-05 21:53:24,304 - Epoch: [151][ 600/ 1236] Overall Loss 0.208126 Objective Loss 0.208126 LR 0.000250 Time 0.021532 +2023-10-05 21:53:24,505 - Epoch: [151][ 610/ 1236] Overall Loss 0.208239 Objective Loss 0.208239 LR 0.000250 Time 0.021509 +2023-10-05 21:53:24,709 - Epoch: [151][ 620/ 1236] Overall Loss 0.208211 Objective Loss 0.208211 LR 0.000250 Time 0.021490 +2023-10-05 21:53:24,911 - Epoch: [151][ 630/ 1236] Overall Loss 0.208390 Objective Loss 0.208390 LR 0.000250 Time 0.021468 +2023-10-05 21:53:25,115 - Epoch: [151][ 640/ 1236] Overall Loss 0.208531 Objective Loss 0.208531 LR 0.000250 Time 0.021451 +2023-10-05 21:53:25,316 - Epoch: [151][ 650/ 1236] Overall Loss 0.208539 Objective Loss 0.208539 LR 0.000250 Time 0.021430 +2023-10-05 21:53:25,519 - Epoch: [151][ 660/ 1236] Overall Loss 0.208481 Objective Loss 0.208481 LR 0.000250 Time 0.021413 +2023-10-05 21:53:25,721 - Epoch: [151][ 670/ 1236] Overall Loss 0.208808 Objective Loss 0.208808 LR 0.000250 Time 0.021393 +2023-10-05 21:53:25,924 - Epoch: [151][ 680/ 1236] Overall Loss 0.208864 Objective Loss 0.208864 LR 0.000250 Time 0.021378 +2023-10-05 21:53:26,126 - Epoch: [151][ 690/ 1236] Overall Loss 0.208662 Objective Loss 0.208662 LR 0.000250 Time 0.021359 +2023-10-05 21:53:26,330 - Epoch: [151][ 700/ 1236] Overall Loss 0.208771 Objective Loss 0.208771 LR 0.000250 Time 0.021346 +2023-10-05 21:53:26,532 - Epoch: [151][ 710/ 1236] Overall Loss 0.208969 Objective Loss 0.208969 LR 0.000250 Time 0.021328 +2023-10-05 21:53:26,736 - Epoch: [151][ 720/ 1236] Overall Loss 0.208894 Objective Loss 0.208894 LR 0.000250 Time 0.021315 +2023-10-05 21:53:26,937 - Epoch: [151][ 730/ 1236] Overall Loss 0.208752 Objective Loss 0.208752 LR 0.000250 Time 0.021298 +2023-10-05 21:53:27,141 - Epoch: [151][ 740/ 1236] Overall Loss 0.208670 Objective Loss 0.208670 LR 0.000250 Time 0.021286 +2023-10-05 21:53:27,343 - Epoch: [151][ 750/ 1236] Overall Loss 0.208604 Objective Loss 0.208604 LR 0.000250 Time 0.021270 +2023-10-05 21:53:27,546 - Epoch: [151][ 760/ 1236] Overall Loss 0.208486 Objective Loss 0.208486 LR 0.000250 Time 0.021258 +2023-10-05 21:53:27,748 - Epoch: [151][ 770/ 1236] Overall Loss 0.208601 Objective Loss 0.208601 LR 0.000250 Time 0.021243 +2023-10-05 21:53:27,952 - Epoch: [151][ 780/ 1236] Overall Loss 0.208446 Objective Loss 0.208446 LR 0.000250 Time 0.021232 +2023-10-05 21:53:28,153 - Epoch: [151][ 790/ 1236] Overall Loss 0.208363 Objective Loss 0.208363 LR 0.000250 Time 0.021218 +2023-10-05 21:53:28,357 - Epoch: [151][ 800/ 1236] Overall Loss 0.208353 Objective Loss 0.208353 LR 0.000250 Time 0.021207 +2023-10-05 21:53:28,559 - Epoch: [151][ 810/ 1236] Overall Loss 0.208326 Objective Loss 0.208326 LR 0.000250 Time 0.021193 +2023-10-05 21:53:28,763 - Epoch: [151][ 820/ 1236] Overall Loss 0.208454 Objective Loss 0.208454 LR 0.000250 Time 0.021183 +2023-10-05 21:53:28,964 - Epoch: [151][ 830/ 1236] Overall Loss 0.208417 Objective Loss 0.208417 LR 0.000250 Time 0.021171 +2023-10-05 21:53:29,168 - Epoch: [151][ 840/ 1236] Overall Loss 0.208592 Objective Loss 0.208592 LR 0.000250 Time 0.021161 +2023-10-05 21:53:29,370 - Epoch: [151][ 850/ 1236] Overall Loss 0.208705 Objective Loss 0.208705 LR 0.000250 Time 0.021149 +2023-10-05 21:53:29,574 - Epoch: [151][ 860/ 1236] Overall Loss 0.208662 Objective Loss 0.208662 LR 0.000250 Time 0.021140 +2023-10-05 21:53:29,775 - Epoch: [151][ 870/ 1236] Overall Loss 0.208586 Objective Loss 0.208586 LR 0.000250 Time 0.021128 +2023-10-05 21:53:29,979 - Epoch: [151][ 880/ 1236] Overall Loss 0.208210 Objective Loss 0.208210 LR 0.000250 Time 0.021119 +2023-10-05 21:53:30,181 - Epoch: [151][ 890/ 1236] Overall Loss 0.208216 Objective Loss 0.208216 LR 0.000250 Time 0.021108 +2023-10-05 21:53:30,384 - Epoch: [151][ 900/ 1236] Overall Loss 0.208285 Objective Loss 0.208285 LR 0.000250 Time 0.021099 +2023-10-05 21:53:30,586 - Epoch: [151][ 910/ 1236] Overall Loss 0.208213 Objective Loss 0.208213 LR 0.000250 Time 0.021088 +2023-10-05 21:53:30,789 - Epoch: [151][ 920/ 1236] Overall Loss 0.208134 Objective Loss 0.208134 LR 0.000250 Time 0.021080 +2023-10-05 21:53:30,991 - Epoch: [151][ 930/ 1236] Overall Loss 0.207959 Objective Loss 0.207959 LR 0.000250 Time 0.021069 +2023-10-05 21:53:31,195 - Epoch: [151][ 940/ 1236] Overall Loss 0.207765 Objective Loss 0.207765 LR 0.000250 Time 0.021062 +2023-10-05 21:53:31,396 - Epoch: [151][ 950/ 1236] Overall Loss 0.207420 Objective Loss 0.207420 LR 0.000250 Time 0.021052 +2023-10-05 21:53:31,601 - Epoch: [151][ 960/ 1236] Overall Loss 0.207451 Objective Loss 0.207451 LR 0.000250 Time 0.021045 +2023-10-05 21:53:31,802 - Epoch: [151][ 970/ 1236] Overall Loss 0.207539 Objective Loss 0.207539 LR 0.000250 Time 0.021035 +2023-10-05 21:53:32,006 - Epoch: [151][ 980/ 1236] Overall Loss 0.207283 Objective Loss 0.207283 LR 0.000250 Time 0.021028 +2023-10-05 21:53:32,208 - Epoch: [151][ 990/ 1236] Overall Loss 0.207308 Objective Loss 0.207308 LR 0.000250 Time 0.021019 +2023-10-05 21:53:32,412 - Epoch: [151][ 1000/ 1236] Overall Loss 0.207133 Objective Loss 0.207133 LR 0.000250 Time 0.021013 +2023-10-05 21:53:32,613 - Epoch: [151][ 1010/ 1236] Overall Loss 0.207222 Objective Loss 0.207222 LR 0.000250 Time 0.021004 +2023-10-05 21:53:32,820 - Epoch: [151][ 1020/ 1236] Overall Loss 0.207318 Objective Loss 0.207318 LR 0.000250 Time 0.021000 +2023-10-05 21:53:33,022 - Epoch: [151][ 1030/ 1236] Overall Loss 0.207158 Objective Loss 0.207158 LR 0.000250 Time 0.020992 +2023-10-05 21:53:33,226 - Epoch: [151][ 1040/ 1236] Overall Loss 0.207381 Objective Loss 0.207381 LR 0.000250 Time 0.020987 +2023-10-05 21:53:33,428 - Epoch: [151][ 1050/ 1236] Overall Loss 0.207582 Objective Loss 0.207582 LR 0.000250 Time 0.020979 +2023-10-05 21:53:33,632 - Epoch: [151][ 1060/ 1236] Overall Loss 0.207527 Objective Loss 0.207527 LR 0.000250 Time 0.020973 +2023-10-05 21:53:33,834 - Epoch: [151][ 1070/ 1236] Overall Loss 0.207349 Objective Loss 0.207349 LR 0.000250 Time 0.020965 +2023-10-05 21:53:34,039 - Epoch: [151][ 1080/ 1236] Overall Loss 0.207349 Objective Loss 0.207349 LR 0.000250 Time 0.020960 +2023-10-05 21:53:34,241 - Epoch: [151][ 1090/ 1236] Overall Loss 0.207457 Objective Loss 0.207457 LR 0.000250 Time 0.020953 +2023-10-05 21:53:34,445 - Epoch: [151][ 1100/ 1236] Overall Loss 0.207579 Objective Loss 0.207579 LR 0.000250 Time 0.020948 +2023-10-05 21:53:34,646 - Epoch: [151][ 1110/ 1236] Overall Loss 0.207613 Objective Loss 0.207613 LR 0.000250 Time 0.020940 +2023-10-05 21:53:34,851 - Epoch: [151][ 1120/ 1236] Overall Loss 0.207615 Objective Loss 0.207615 LR 0.000250 Time 0.020936 +2023-10-05 21:53:35,052 - Epoch: [151][ 1130/ 1236] Overall Loss 0.207625 Objective Loss 0.207625 LR 0.000250 Time 0.020928 +2023-10-05 21:53:35,257 - Epoch: [151][ 1140/ 1236] Overall Loss 0.207882 Objective Loss 0.207882 LR 0.000250 Time 0.020924 +2023-10-05 21:53:35,458 - Epoch: [151][ 1150/ 1236] Overall Loss 0.207766 Objective Loss 0.207766 LR 0.000250 Time 0.020917 +2023-10-05 21:53:35,663 - Epoch: [151][ 1160/ 1236] Overall Loss 0.207737 Objective Loss 0.207737 LR 0.000250 Time 0.020912 +2023-10-05 21:53:35,864 - Epoch: [151][ 1170/ 1236] Overall Loss 0.207855 Objective Loss 0.207855 LR 0.000250 Time 0.020906 +2023-10-05 21:53:36,069 - Epoch: [151][ 1180/ 1236] Overall Loss 0.208012 Objective Loss 0.208012 LR 0.000250 Time 0.020901 +2023-10-05 21:53:36,270 - Epoch: [151][ 1190/ 1236] Overall Loss 0.207957 Objective Loss 0.207957 LR 0.000250 Time 0.020895 +2023-10-05 21:53:36,475 - Epoch: [151][ 1200/ 1236] Overall Loss 0.207922 Objective Loss 0.207922 LR 0.000250 Time 0.020891 +2023-10-05 21:53:36,676 - Epoch: [151][ 1210/ 1236] Overall Loss 0.208011 Objective Loss 0.208011 LR 0.000250 Time 0.020884 +2023-10-05 21:53:36,880 - Epoch: [151][ 1220/ 1236] Overall Loss 0.207898 Objective Loss 0.207898 LR 0.000250 Time 0.020880 +2023-10-05 21:53:37,137 - Epoch: [151][ 1230/ 1236] Overall Loss 0.208082 Objective Loss 0.208082 LR 0.000250 Time 0.020919 +2023-10-05 21:53:37,256 - Epoch: [151][ 1236/ 1236] Overall Loss 0.208156 Objective Loss 0.208156 Top1 89.409369 Top5 97.759674 LR 0.000250 Time 0.020914 +2023-10-05 21:53:37,384 - --- validate (epoch=151)----------- +2023-10-05 21:53:37,385 - 29943 samples (256 per mini-batch) +2023-10-05 21:53:37,854 - Epoch: [151][ 10/ 117] Loss 0.302234 Top1 85.234375 Top5 98.085938 +2023-10-05 21:53:38,003 - Epoch: [151][ 20/ 117] Loss 0.293046 Top1 85.546875 Top5 98.222656 +2023-10-05 21:53:38,152 - Epoch: [151][ 30/ 117] Loss 0.308402 Top1 84.973958 Top5 98.177083 +2023-10-05 21:53:38,300 - Epoch: [151][ 40/ 117] Loss 0.315192 Top1 84.736328 Top5 97.900391 +2023-10-05 21:53:38,449 - Epoch: [151][ 50/ 117] Loss 0.316874 Top1 84.750000 Top5 97.890625 +2023-10-05 21:53:38,597 - Epoch: [151][ 60/ 117] Loss 0.313994 Top1 84.993490 Top5 97.955729 +2023-10-05 21:53:38,745 - Epoch: [151][ 70/ 117] Loss 0.315601 Top1 84.955357 Top5 97.979911 +2023-10-05 21:53:38,893 - Epoch: [151][ 80/ 117] Loss 0.319160 Top1 84.794922 Top5 97.905273 +2023-10-05 21:53:39,041 - Epoch: [151][ 90/ 117] Loss 0.315099 Top1 84.917535 Top5 97.947049 +2023-10-05 21:53:39,190 - Epoch: [151][ 100/ 117] Loss 0.315217 Top1 84.917969 Top5 97.945312 +2023-10-05 21:53:39,344 - Epoch: [151][ 110/ 117] Loss 0.313398 Top1 84.825994 Top5 97.926136 +2023-10-05 21:53:39,430 - Epoch: [151][ 117/ 117] Loss 0.313740 Top1 84.817821 Top5 97.946098 +2023-10-05 21:53:39,566 - ==> Top1: 84.818 Top5: 97.946 Loss: 0.314 + +2023-10-05 21:53:39,567 - ==> Confusion: +[[ 933 2 6 1 5 3 0 0 4 73 1 0 2 2 1 1 2 1 1 0 12] + [ 2 1067 3 0 9 20 1 9 2 0 0 1 0 0 0 4 1 0 8 0 4] + [ 3 0 973 11 4 1 18 5 0 2 5 2 7 3 0 4 1 2 6 3 6] + [ 2 1 10 972 1 4 0 1 1 1 8 1 4 2 25 3 0 4 30 2 17] + [ 22 8 2 0 980 4 0 1 1 10 0 0 0 0 4 3 7 2 0 2 4] + [ 5 48 1 1 4 980 1 21 0 3 5 5 0 11 4 1 5 1 4 5 11] + [ 0 10 25 0 0 0 1122 7 0 0 0 3 1 0 1 9 0 1 1 5 6] + [ 5 23 18 0 3 18 5 1068 1 3 5 12 1 1 1 5 0 0 38 4 7] + [ 18 2 0 0 0 0 1 0 985 50 8 3 1 3 8 5 2 1 2 0 0] + [ 88 0 4 0 7 4 0 0 19 963 1 1 0 13 3 8 1 0 0 1 6] + [ 2 3 11 3 3 0 2 3 13 5 969 3 0 13 2 1 3 0 4 2 11] + [ 2 1 2 0 0 15 0 2 0 1 0 951 22 6 0 4 3 17 0 3 6] + [ 2 2 2 6 1 1 0 1 3 0 2 29 987 0 1 2 2 11 2 2 12] + [ 2 0 1 0 2 7 0 1 20 22 8 5 2 1031 3 3 2 0 0 0 10] + [ 13 3 3 11 9 1 0 0 32 4 3 0 2 2 991 0 1 2 13 0 11] + [ 1 4 1 0 4 0 1 0 0 0 0 7 6 2 0 1073 11 11 0 8 5] + [ 0 13 1 0 9 4 0 0 4 0 1 3 0 0 3 8 1105 0 1 2 7] + [ 0 0 1 2 0 1 2 0 0 1 0 3 15 1 0 5 0 1004 1 1 1] + [ 2 6 6 17 1 0 0 19 4 1 4 0 1 0 6 0 0 0 996 0 5] + [ 0 3 4 3 3 10 10 13 1 0 1 20 3 0 0 4 11 1 5 1053 7] + [ 139 188 161 57 114 134 36 93 141 94 167 108 324 248 126 56 128 60 181 156 5194]] + +2023-10-05 21:53:39,569 - ==> Best [Top1: 85.068 Top5: 98.030 Sparsity:0.00 Params: 148928 on epoch: 148] +2023-10-05 21:53:39,569 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:53:39,575 - + +2023-10-05 21:53:39,575 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:53:40,587 - Epoch: [152][ 10/ 1236] Overall Loss 0.217109 Objective Loss 0.217109 LR 0.000250 Time 0.101210 +2023-10-05 21:53:40,792 - Epoch: [152][ 20/ 1236] Overall Loss 0.208430 Objective Loss 0.208430 LR 0.000250 Time 0.060808 +2023-10-05 21:53:40,994 - Epoch: [152][ 30/ 1236] Overall Loss 0.211361 Objective Loss 0.211361 LR 0.000250 Time 0.047257 +2023-10-05 21:53:41,198 - Epoch: [152][ 40/ 1236] Overall Loss 0.203443 Objective Loss 0.203443 LR 0.000250 Time 0.040537 +2023-10-05 21:53:41,400 - Epoch: [152][ 50/ 1236] Overall Loss 0.204434 Objective Loss 0.204434 LR 0.000250 Time 0.036459 +2023-10-05 21:53:41,604 - Epoch: [152][ 60/ 1236] Overall Loss 0.200982 Objective Loss 0.200982 LR 0.000250 Time 0.033781 +2023-10-05 21:53:41,806 - Epoch: [152][ 70/ 1236] Overall Loss 0.202823 Objective Loss 0.202823 LR 0.000250 Time 0.031835 +2023-10-05 21:53:42,010 - Epoch: [152][ 80/ 1236] Overall Loss 0.204639 Objective Loss 0.204639 LR 0.000250 Time 0.030406 +2023-10-05 21:53:42,214 - Epoch: [152][ 90/ 1236] Overall Loss 0.206969 Objective Loss 0.206969 LR 0.000250 Time 0.029286 +2023-10-05 21:53:42,419 - Epoch: [152][ 100/ 1236] Overall Loss 0.204538 Objective Loss 0.204538 LR 0.000250 Time 0.028409 +2023-10-05 21:53:42,622 - Epoch: [152][ 110/ 1236] Overall Loss 0.205613 Objective Loss 0.205613 LR 0.000250 Time 0.027662 +2023-10-05 21:53:42,825 - Epoch: [152][ 120/ 1236] Overall Loss 0.206297 Objective Loss 0.206297 LR 0.000250 Time 0.027049 +2023-10-05 21:53:43,026 - Epoch: [152][ 130/ 1236] Overall Loss 0.206448 Objective Loss 0.206448 LR 0.000250 Time 0.026511 +2023-10-05 21:53:43,230 - Epoch: [152][ 140/ 1236] Overall Loss 0.207583 Objective Loss 0.207583 LR 0.000250 Time 0.026070 +2023-10-05 21:53:43,431 - Epoch: [152][ 150/ 1236] Overall Loss 0.206552 Objective Loss 0.206552 LR 0.000250 Time 0.025672 +2023-10-05 21:53:43,635 - Epoch: [152][ 160/ 1236] Overall Loss 0.207082 Objective Loss 0.207082 LR 0.000250 Time 0.025338 +2023-10-05 21:53:43,836 - Epoch: [152][ 170/ 1236] Overall Loss 0.206705 Objective Loss 0.206705 LR 0.000250 Time 0.025030 +2023-10-05 21:53:44,040 - Epoch: [152][ 180/ 1236] Overall Loss 0.206885 Objective Loss 0.206885 LR 0.000250 Time 0.024768 +2023-10-05 21:53:44,241 - Epoch: [152][ 190/ 1236] Overall Loss 0.206669 Objective Loss 0.206669 LR 0.000250 Time 0.024523 +2023-10-05 21:53:44,445 - Epoch: [152][ 200/ 1236] Overall Loss 0.206635 Objective Loss 0.206635 LR 0.000250 Time 0.024313 +2023-10-05 21:53:44,646 - Epoch: [152][ 210/ 1236] Overall Loss 0.206365 Objective Loss 0.206365 LR 0.000250 Time 0.024113 +2023-10-05 21:53:44,850 - Epoch: [152][ 220/ 1236] Overall Loss 0.206378 Objective Loss 0.206378 LR 0.000250 Time 0.023940 +2023-10-05 21:53:45,051 - Epoch: [152][ 230/ 1236] Overall Loss 0.206337 Objective Loss 0.206337 LR 0.000250 Time 0.023773 +2023-10-05 21:53:45,254 - Epoch: [152][ 240/ 1236] Overall Loss 0.205958 Objective Loss 0.205958 LR 0.000250 Time 0.023627 +2023-10-05 21:53:45,456 - Epoch: [152][ 250/ 1236] Overall Loss 0.206341 Objective Loss 0.206341 LR 0.000250 Time 0.023486 +2023-10-05 21:53:45,660 - Epoch: [152][ 260/ 1236] Overall Loss 0.205745 Objective Loss 0.205745 LR 0.000250 Time 0.023369 +2023-10-05 21:53:45,865 - Epoch: [152][ 270/ 1236] Overall Loss 0.205147 Objective Loss 0.205147 LR 0.000250 Time 0.023260 +2023-10-05 21:53:46,070 - Epoch: [152][ 280/ 1236] Overall Loss 0.204162 Objective Loss 0.204162 LR 0.000250 Time 0.023160 +2023-10-05 21:53:46,274 - Epoch: [152][ 290/ 1236] Overall Loss 0.204545 Objective Loss 0.204545 LR 0.000250 Time 0.023064 +2023-10-05 21:53:46,479 - Epoch: [152][ 300/ 1236] Overall Loss 0.205655 Objective Loss 0.205655 LR 0.000250 Time 0.022978 +2023-10-05 21:53:46,684 - Epoch: [152][ 310/ 1236] Overall Loss 0.205408 Objective Loss 0.205408 LR 0.000250 Time 0.022896 +2023-10-05 21:53:46,889 - Epoch: [152][ 320/ 1236] Overall Loss 0.204931 Objective Loss 0.204931 LR 0.000250 Time 0.022821 +2023-10-05 21:53:47,094 - Epoch: [152][ 330/ 1236] Overall Loss 0.205082 Objective Loss 0.205082 LR 0.000250 Time 0.022749 +2023-10-05 21:53:47,299 - Epoch: [152][ 340/ 1236] Overall Loss 0.204800 Objective Loss 0.204800 LR 0.000250 Time 0.022683 +2023-10-05 21:53:47,504 - Epoch: [152][ 350/ 1236] Overall Loss 0.204814 Objective Loss 0.204814 LR 0.000250 Time 0.022618 +2023-10-05 21:53:47,709 - Epoch: [152][ 360/ 1236] Overall Loss 0.204344 Objective Loss 0.204344 LR 0.000250 Time 0.022559 +2023-10-05 21:53:47,914 - Epoch: [152][ 370/ 1236] Overall Loss 0.205231 Objective Loss 0.205231 LR 0.000250 Time 0.022503 +2023-10-05 21:53:48,119 - Epoch: [152][ 380/ 1236] Overall Loss 0.205278 Objective Loss 0.205278 LR 0.000250 Time 0.022448 +2023-10-05 21:53:48,323 - Epoch: [152][ 390/ 1236] Overall Loss 0.205177 Objective Loss 0.205177 LR 0.000250 Time 0.022394 +2023-10-05 21:53:48,527 - Epoch: [152][ 400/ 1236] Overall Loss 0.205172 Objective Loss 0.205172 LR 0.000250 Time 0.022344 +2023-10-05 21:53:48,730 - Epoch: [152][ 410/ 1236] Overall Loss 0.205372 Objective Loss 0.205372 LR 0.000250 Time 0.022295 +2023-10-05 21:53:48,935 - Epoch: [152][ 420/ 1236] Overall Loss 0.205622 Objective Loss 0.205622 LR 0.000250 Time 0.022250 +2023-10-05 21:53:49,139 - Epoch: [152][ 430/ 1236] Overall Loss 0.205096 Objective Loss 0.205096 LR 0.000250 Time 0.022206 +2023-10-05 21:53:49,343 - Epoch: [152][ 440/ 1236] Overall Loss 0.205328 Objective Loss 0.205328 LR 0.000250 Time 0.022165 +2023-10-05 21:53:49,547 - Epoch: [152][ 450/ 1236] Overall Loss 0.205547 Objective Loss 0.205547 LR 0.000250 Time 0.022125 +2023-10-05 21:53:49,751 - Epoch: [152][ 460/ 1236] Overall Loss 0.205753 Objective Loss 0.205753 LR 0.000250 Time 0.022087 +2023-10-05 21:53:49,956 - Epoch: [152][ 470/ 1236] Overall Loss 0.206175 Objective Loss 0.206175 LR 0.000250 Time 0.022051 +2023-10-05 21:53:50,160 - Epoch: [152][ 480/ 1236] Overall Loss 0.206166 Objective Loss 0.206166 LR 0.000250 Time 0.022016 +2023-10-05 21:53:50,364 - Epoch: [152][ 490/ 1236] Overall Loss 0.205992 Objective Loss 0.205992 LR 0.000250 Time 0.021982 +2023-10-05 21:53:50,568 - Epoch: [152][ 500/ 1236] Overall Loss 0.206364 Objective Loss 0.206364 LR 0.000250 Time 0.021950 +2023-10-05 21:53:50,772 - Epoch: [152][ 510/ 1236] Overall Loss 0.206162 Objective Loss 0.206162 LR 0.000250 Time 0.021919 +2023-10-05 21:53:50,976 - Epoch: [152][ 520/ 1236] Overall Loss 0.206671 Objective Loss 0.206671 LR 0.000250 Time 0.021890 +2023-10-05 21:53:51,180 - Epoch: [152][ 530/ 1236] Overall Loss 0.206538 Objective Loss 0.206538 LR 0.000250 Time 0.021862 +2023-10-05 21:53:51,384 - Epoch: [152][ 540/ 1236] Overall Loss 0.206672 Objective Loss 0.206672 LR 0.000250 Time 0.021834 +2023-10-05 21:53:51,588 - Epoch: [152][ 550/ 1236] Overall Loss 0.206466 Objective Loss 0.206466 LR 0.000250 Time 0.021807 +2023-10-05 21:53:51,792 - Epoch: [152][ 560/ 1236] Overall Loss 0.206827 Objective Loss 0.206827 LR 0.000250 Time 0.021782 +2023-10-05 21:53:51,996 - Epoch: [152][ 570/ 1236] Overall Loss 0.206878 Objective Loss 0.206878 LR 0.000250 Time 0.021757 +2023-10-05 21:53:52,201 - Epoch: [152][ 580/ 1236] Overall Loss 0.206618 Objective Loss 0.206618 LR 0.000250 Time 0.021734 +2023-10-05 21:53:52,405 - Epoch: [152][ 590/ 1236] Overall Loss 0.206447 Objective Loss 0.206447 LR 0.000250 Time 0.021710 +2023-10-05 21:53:52,609 - Epoch: [152][ 600/ 1236] Overall Loss 0.206552 Objective Loss 0.206552 LR 0.000250 Time 0.021688 +2023-10-05 21:53:52,812 - Epoch: [152][ 610/ 1236] Overall Loss 0.206190 Objective Loss 0.206190 LR 0.000250 Time 0.021665 +2023-10-05 21:53:53,017 - Epoch: [152][ 620/ 1236] Overall Loss 0.206027 Objective Loss 0.206027 LR 0.000250 Time 0.021646 +2023-10-05 21:53:53,222 - Epoch: [152][ 630/ 1236] Overall Loss 0.206031 Objective Loss 0.206031 LR 0.000250 Time 0.021627 +2023-10-05 21:53:53,426 - Epoch: [152][ 640/ 1236] Overall Loss 0.206218 Objective Loss 0.206218 LR 0.000250 Time 0.021608 +2023-10-05 21:53:53,630 - Epoch: [152][ 650/ 1236] Overall Loss 0.206398 Objective Loss 0.206398 LR 0.000250 Time 0.021589 +2023-10-05 21:53:53,834 - Epoch: [152][ 660/ 1236] Overall Loss 0.205995 Objective Loss 0.205995 LR 0.000250 Time 0.021570 +2023-10-05 21:53:54,037 - Epoch: [152][ 670/ 1236] Overall Loss 0.205804 Objective Loss 0.205804 LR 0.000250 Time 0.021551 +2023-10-05 21:53:54,242 - Epoch: [152][ 680/ 1236] Overall Loss 0.206107 Objective Loss 0.206107 LR 0.000250 Time 0.021535 +2023-10-05 21:53:54,446 - Epoch: [152][ 690/ 1236] Overall Loss 0.206213 Objective Loss 0.206213 LR 0.000250 Time 0.021517 +2023-10-05 21:53:54,650 - Epoch: [152][ 700/ 1236] Overall Loss 0.206255 Objective Loss 0.206255 LR 0.000250 Time 0.021501 +2023-10-05 21:53:54,853 - Epoch: [152][ 710/ 1236] Overall Loss 0.205925 Objective Loss 0.205925 LR 0.000250 Time 0.021484 +2023-10-05 21:53:55,058 - Epoch: [152][ 720/ 1236] Overall Loss 0.205911 Objective Loss 0.205911 LR 0.000250 Time 0.021469 +2023-10-05 21:53:55,261 - Epoch: [152][ 730/ 1236] Overall Loss 0.205577 Objective Loss 0.205577 LR 0.000250 Time 0.021453 +2023-10-05 21:53:55,466 - Epoch: [152][ 740/ 1236] Overall Loss 0.205314 Objective Loss 0.205314 LR 0.000250 Time 0.021439 +2023-10-05 21:53:55,669 - Epoch: [152][ 750/ 1236] Overall Loss 0.205308 Objective Loss 0.205308 LR 0.000250 Time 0.021424 +2023-10-05 21:53:55,874 - Epoch: [152][ 760/ 1236] Overall Loss 0.205381 Objective Loss 0.205381 LR 0.000250 Time 0.021412 +2023-10-05 21:53:56,078 - Epoch: [152][ 770/ 1236] Overall Loss 0.205233 Objective Loss 0.205233 LR 0.000250 Time 0.021398 +2023-10-05 21:53:56,283 - Epoch: [152][ 780/ 1236] Overall Loss 0.205378 Objective Loss 0.205378 LR 0.000250 Time 0.021386 +2023-10-05 21:53:56,487 - Epoch: [152][ 790/ 1236] Overall Loss 0.205452 Objective Loss 0.205452 LR 0.000250 Time 0.021373 +2023-10-05 21:53:56,692 - Epoch: [152][ 800/ 1236] Overall Loss 0.205712 Objective Loss 0.205712 LR 0.000250 Time 0.021362 +2023-10-05 21:53:56,897 - Epoch: [152][ 810/ 1236] Overall Loss 0.205672 Objective Loss 0.205672 LR 0.000250 Time 0.021350 +2023-10-05 21:53:57,101 - Epoch: [152][ 820/ 1236] Overall Loss 0.205792 Objective Loss 0.205792 LR 0.000250 Time 0.021339 +2023-10-05 21:53:57,305 - Epoch: [152][ 830/ 1236] Overall Loss 0.206046 Objective Loss 0.206046 LR 0.000250 Time 0.021327 +2023-10-05 21:53:57,510 - Epoch: [152][ 840/ 1236] Overall Loss 0.206293 Objective Loss 0.206293 LR 0.000250 Time 0.021317 +2023-10-05 21:53:57,715 - Epoch: [152][ 850/ 1236] Overall Loss 0.206350 Objective Loss 0.206350 LR 0.000250 Time 0.021306 +2023-10-05 21:53:57,919 - Epoch: [152][ 860/ 1236] Overall Loss 0.206264 Objective Loss 0.206264 LR 0.000250 Time 0.021295 +2023-10-05 21:53:58,123 - Epoch: [152][ 870/ 1236] Overall Loss 0.206362 Objective Loss 0.206362 LR 0.000250 Time 0.021285 +2023-10-05 21:53:58,328 - Epoch: [152][ 880/ 1236] Overall Loss 0.206345 Objective Loss 0.206345 LR 0.000250 Time 0.021275 +2023-10-05 21:53:58,532 - Epoch: [152][ 890/ 1236] Overall Loss 0.206223 Objective Loss 0.206223 LR 0.000250 Time 0.021265 +2023-10-05 21:53:58,737 - Epoch: [152][ 900/ 1236] Overall Loss 0.206209 Objective Loss 0.206209 LR 0.000250 Time 0.021257 +2023-10-05 21:53:58,941 - Epoch: [152][ 910/ 1236] Overall Loss 0.206533 Objective Loss 0.206533 LR 0.000250 Time 0.021247 +2023-10-05 21:53:59,146 - Epoch: [152][ 920/ 1236] Overall Loss 0.206411 Objective Loss 0.206411 LR 0.000250 Time 0.021239 +2023-10-05 21:53:59,350 - Epoch: [152][ 930/ 1236] Overall Loss 0.206410 Objective Loss 0.206410 LR 0.000250 Time 0.021229 +2023-10-05 21:53:59,555 - Epoch: [152][ 940/ 1236] Overall Loss 0.206529 Objective Loss 0.206529 LR 0.000250 Time 0.021221 +2023-10-05 21:53:59,759 - Epoch: [152][ 950/ 1236] Overall Loss 0.206608 Objective Loss 0.206608 LR 0.000250 Time 0.021212 +2023-10-05 21:53:59,963 - Epoch: [152][ 960/ 1236] Overall Loss 0.206698 Objective Loss 0.206698 LR 0.000250 Time 0.021203 +2023-10-05 21:54:00,167 - Epoch: [152][ 970/ 1236] Overall Loss 0.206504 Objective Loss 0.206504 LR 0.000250 Time 0.021194 +2023-10-05 21:54:00,373 - Epoch: [152][ 980/ 1236] Overall Loss 0.206574 Objective Loss 0.206574 LR 0.000250 Time 0.021187 +2023-10-05 21:54:00,576 - Epoch: [152][ 990/ 1236] Overall Loss 0.206518 Objective Loss 0.206518 LR 0.000250 Time 0.021179 +2023-10-05 21:54:00,781 - Epoch: [152][ 1000/ 1236] Overall Loss 0.206258 Objective Loss 0.206258 LR 0.000250 Time 0.021172 +2023-10-05 21:54:00,985 - Epoch: [152][ 1010/ 1236] Overall Loss 0.206311 Objective Loss 0.206311 LR 0.000250 Time 0.021163 +2023-10-05 21:54:01,190 - Epoch: [152][ 1020/ 1236] Overall Loss 0.206573 Objective Loss 0.206573 LR 0.000250 Time 0.021157 +2023-10-05 21:54:01,394 - Epoch: [152][ 1030/ 1236] Overall Loss 0.206649 Objective Loss 0.206649 LR 0.000250 Time 0.021149 +2023-10-05 21:54:01,600 - Epoch: [152][ 1040/ 1236] Overall Loss 0.206439 Objective Loss 0.206439 LR 0.000250 Time 0.021143 +2023-10-05 21:54:01,804 - Epoch: [152][ 1050/ 1236] Overall Loss 0.206570 Objective Loss 0.206570 LR 0.000250 Time 0.021136 +2023-10-05 21:54:02,009 - Epoch: [152][ 1060/ 1236] Overall Loss 0.206458 Objective Loss 0.206458 LR 0.000250 Time 0.021129 +2023-10-05 21:54:02,212 - Epoch: [152][ 1070/ 1236] Overall Loss 0.206198 Objective Loss 0.206198 LR 0.000250 Time 0.021122 +2023-10-05 21:54:02,417 - Epoch: [152][ 1080/ 1236] Overall Loss 0.206204 Objective Loss 0.206204 LR 0.000250 Time 0.021116 +2023-10-05 21:54:02,621 - Epoch: [152][ 1090/ 1236] Overall Loss 0.206310 Objective Loss 0.206310 LR 0.000250 Time 0.021108 +2023-10-05 21:54:02,827 - Epoch: [152][ 1100/ 1236] Overall Loss 0.206322 Objective Loss 0.206322 LR 0.000250 Time 0.021103 +2023-10-05 21:54:03,031 - Epoch: [152][ 1110/ 1236] Overall Loss 0.206287 Objective Loss 0.206287 LR 0.000250 Time 0.021097 +2023-10-05 21:54:03,236 - Epoch: [152][ 1120/ 1236] Overall Loss 0.206372 Objective Loss 0.206372 LR 0.000250 Time 0.021091 +2023-10-05 21:54:03,440 - Epoch: [152][ 1130/ 1236] Overall Loss 0.206467 Objective Loss 0.206467 LR 0.000250 Time 0.021085 +2023-10-05 21:54:03,645 - Epoch: [152][ 1140/ 1236] Overall Loss 0.206696 Objective Loss 0.206696 LR 0.000250 Time 0.021079 +2023-10-05 21:54:03,849 - Epoch: [152][ 1150/ 1236] Overall Loss 0.206713 Objective Loss 0.206713 LR 0.000250 Time 0.021073 +2023-10-05 21:54:04,054 - Epoch: [152][ 1160/ 1236] Overall Loss 0.206711 Objective Loss 0.206711 LR 0.000250 Time 0.021067 +2023-10-05 21:54:04,257 - Epoch: [152][ 1170/ 1236] Overall Loss 0.206609 Objective Loss 0.206609 LR 0.000250 Time 0.021061 +2023-10-05 21:54:04,472 - Epoch: [152][ 1180/ 1236] Overall Loss 0.206655 Objective Loss 0.206655 LR 0.000250 Time 0.021064 +2023-10-05 21:54:04,682 - Epoch: [152][ 1190/ 1236] Overall Loss 0.206544 Objective Loss 0.206544 LR 0.000250 Time 0.021063 +2023-10-05 21:54:04,896 - Epoch: [152][ 1200/ 1236] Overall Loss 0.206500 Objective Loss 0.206500 LR 0.000250 Time 0.021066 +2023-10-05 21:54:05,106 - Epoch: [152][ 1210/ 1236] Overall Loss 0.206679 Objective Loss 0.206679 LR 0.000250 Time 0.021065 +2023-10-05 21:54:05,321 - Epoch: [152][ 1220/ 1236] Overall Loss 0.206600 Objective Loss 0.206600 LR 0.000250 Time 0.021069 +2023-10-05 21:54:05,583 - Epoch: [152][ 1230/ 1236] Overall Loss 0.206736 Objective Loss 0.206736 LR 0.000250 Time 0.021110 +2023-10-05 21:54:05,701 - Epoch: [152][ 1236/ 1236] Overall Loss 0.206791 Objective Loss 0.206791 Top1 88.798371 Top5 98.574338 LR 0.000250 Time 0.021103 +2023-10-05 21:54:05,827 - --- validate (epoch=152)----------- +2023-10-05 21:54:05,827 - 29943 samples (256 per mini-batch) +2023-10-05 21:54:06,286 - Epoch: [152][ 10/ 117] Loss 0.274234 Top1 85.820312 Top5 98.203125 +2023-10-05 21:54:06,440 - Epoch: [152][ 20/ 117] Loss 0.295769 Top1 85.214844 Top5 98.027344 +2023-10-05 21:54:06,592 - Epoch: [152][ 30/ 117] Loss 0.308622 Top1 84.921875 Top5 97.994792 +2023-10-05 21:54:06,755 - Epoch: [152][ 40/ 117] Loss 0.301666 Top1 85.097656 Top5 98.085938 +2023-10-05 21:54:06,913 - Epoch: [152][ 50/ 117] Loss 0.310908 Top1 84.828125 Top5 98.023438 +2023-10-05 21:54:07,076 - Epoch: [152][ 60/ 117] Loss 0.309050 Top1 84.889323 Top5 98.098958 +2023-10-05 21:54:07,232 - Epoch: [152][ 70/ 117] Loss 0.308684 Top1 85.072545 Top5 98.102679 +2023-10-05 21:54:07,394 - Epoch: [152][ 80/ 117] Loss 0.310305 Top1 85.092773 Top5 98.125000 +2023-10-05 21:54:07,550 - Epoch: [152][ 90/ 117] Loss 0.310599 Top1 85.130208 Top5 98.129340 +2023-10-05 21:54:07,712 - Epoch: [152][ 100/ 117] Loss 0.311778 Top1 85.101562 Top5 98.082031 +2023-10-05 21:54:07,874 - Epoch: [152][ 110/ 117] Loss 0.310924 Top1 85.209517 Top5 98.110795 +2023-10-05 21:54:07,960 - Epoch: [152][ 117/ 117] Loss 0.310123 Top1 85.198544 Top5 98.089704 +2023-10-05 21:54:08,100 - ==> Top1: 85.199 Top5: 98.090 Loss: 0.310 + +2023-10-05 21:54:08,101 - ==> Confusion: +[[ 934 3 3 0 5 4 0 0 4 70 1 0 0 1 6 3 3 1 0 0 12] + [ 1 1071 2 0 5 15 1 17 1 0 2 1 0 0 0 4 1 0 4 0 6] + [ 5 1 949 16 3 0 34 8 0 2 6 1 7 4 0 4 1 1 2 1 11] + [ 1 1 15 963 1 5 1 1 2 1 9 1 5 1 30 5 1 9 18 0 19] + [ 22 8 1 0 971 3 0 1 0 12 0 0 0 2 7 3 11 2 0 2 5] + [ 4 46 0 1 4 985 2 26 0 1 3 7 0 10 6 1 6 0 2 3 9] + [ 1 7 16 0 1 1 1130 9 0 0 2 1 1 0 1 8 0 0 1 5 7] + [ 3 20 8 0 1 24 6 1087 0 5 4 9 1 1 1 2 0 1 33 3 9] + [ 16 6 1 0 0 1 1 1 987 41 8 1 0 6 11 5 1 0 1 0 2] + [ 100 0 3 0 5 2 0 1 21 945 1 1 1 18 3 7 0 1 0 1 9] + [ 2 10 6 4 1 1 3 5 10 3 976 2 0 12 4 0 2 1 0 1 10] + [ 2 2 0 0 0 16 0 1 0 1 0 952 18 4 0 4 3 20 0 9 3] + [ 0 3 1 5 0 2 1 1 1 0 1 33 984 1 3 7 3 10 1 4 7] + [ 2 0 2 0 3 5 1 1 16 10 8 6 2 1043 2 2 1 0 0 1 14] + [ 15 3 5 7 4 0 0 0 27 1 3 0 2 2 1008 0 1 2 9 0 12] + [ 1 4 1 0 6 0 1 0 0 0 0 8 5 1 0 1073 13 8 0 9 4] + [ 0 12 1 0 5 3 0 0 3 0 0 2 1 0 3 9 1104 0 0 2 16] + [ 0 0 0 0 1 1 2 0 2 0 0 2 18 0 0 4 0 1002 1 0 5] + [ 1 9 7 21 1 0 1 23 1 1 4 0 0 1 7 0 1 0 979 1 10] + [ 0 3 1 2 2 9 10 9 1 0 2 14 4 1 0 4 11 1 3 1067 8] + [ 116 207 123 47 93 125 40 116 131 84 191 97 315 260 131 62 138 52 118 158 5301]] + +2023-10-05 21:54:08,102 - ==> Best [Top1: 85.199 Top5: 98.090 Sparsity:0.00 Params: 148928 on epoch: 152] +2023-10-05 21:54:08,102 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:54:08,116 - + +2023-10-05 21:54:08,116 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:54:09,118 - Epoch: [153][ 10/ 1236] Overall Loss 0.186810 Objective Loss 0.186810 LR 0.000250 Time 0.100194 +2023-10-05 21:54:09,326 - Epoch: [153][ 20/ 1236] Overall Loss 0.192218 Objective Loss 0.192218 LR 0.000250 Time 0.060467 +2023-10-05 21:54:09,530 - Epoch: [153][ 30/ 1236] Overall Loss 0.193747 Objective Loss 0.193747 LR 0.000250 Time 0.047111 +2023-10-05 21:54:09,738 - Epoch: [153][ 40/ 1236] Overall Loss 0.196847 Objective Loss 0.196847 LR 0.000250 Time 0.040509 +2023-10-05 21:54:09,942 - Epoch: [153][ 50/ 1236] Overall Loss 0.193114 Objective Loss 0.193114 LR 0.000250 Time 0.036486 +2023-10-05 21:54:10,150 - Epoch: [153][ 60/ 1236] Overall Loss 0.195090 Objective Loss 0.195090 LR 0.000250 Time 0.033860 +2023-10-05 21:54:10,353 - Epoch: [153][ 70/ 1236] Overall Loss 0.197288 Objective Loss 0.197288 LR 0.000250 Time 0.031931 +2023-10-05 21:54:10,562 - Epoch: [153][ 80/ 1236] Overall Loss 0.199986 Objective Loss 0.199986 LR 0.000250 Time 0.030542 +2023-10-05 21:54:10,766 - Epoch: [153][ 90/ 1236] Overall Loss 0.199155 Objective Loss 0.199155 LR 0.000250 Time 0.029412 +2023-10-05 21:54:10,974 - Epoch: [153][ 100/ 1236] Overall Loss 0.198274 Objective Loss 0.198274 LR 0.000250 Time 0.028545 +2023-10-05 21:54:11,177 - Epoch: [153][ 110/ 1236] Overall Loss 0.198176 Objective Loss 0.198176 LR 0.000250 Time 0.027798 +2023-10-05 21:54:11,383 - Epoch: [153][ 120/ 1236] Overall Loss 0.199302 Objective Loss 0.199302 LR 0.000250 Time 0.027196 +2023-10-05 21:54:11,583 - Epoch: [153][ 130/ 1236] Overall Loss 0.199265 Objective Loss 0.199265 LR 0.000250 Time 0.026636 +2023-10-05 21:54:11,784 - Epoch: [153][ 140/ 1236] Overall Loss 0.199685 Objective Loss 0.199685 LR 0.000250 Time 0.026169 +2023-10-05 21:54:11,988 - Epoch: [153][ 150/ 1236] Overall Loss 0.199354 Objective Loss 0.199354 LR 0.000250 Time 0.025777 +2023-10-05 21:54:12,204 - Epoch: [153][ 160/ 1236] Overall Loss 0.199715 Objective Loss 0.199715 LR 0.000250 Time 0.025516 +2023-10-05 21:54:12,410 - Epoch: [153][ 170/ 1236] Overall Loss 0.199717 Objective Loss 0.199717 LR 0.000250 Time 0.025226 +2023-10-05 21:54:12,616 - Epoch: [153][ 180/ 1236] Overall Loss 0.200373 Objective Loss 0.200373 LR 0.000250 Time 0.024966 +2023-10-05 21:54:12,820 - Epoch: [153][ 190/ 1236] Overall Loss 0.200259 Objective Loss 0.200259 LR 0.000250 Time 0.024724 +2023-10-05 21:54:13,026 - Epoch: [153][ 200/ 1236] Overall Loss 0.199882 Objective Loss 0.199882 LR 0.000250 Time 0.024517 +2023-10-05 21:54:13,230 - Epoch: [153][ 210/ 1236] Overall Loss 0.200850 Objective Loss 0.200850 LR 0.000250 Time 0.024319 +2023-10-05 21:54:13,436 - Epoch: [153][ 220/ 1236] Overall Loss 0.201110 Objective Loss 0.201110 LR 0.000250 Time 0.024148 +2023-10-05 21:54:13,640 - Epoch: [153][ 230/ 1236] Overall Loss 0.201450 Objective Loss 0.201450 LR 0.000250 Time 0.023984 +2023-10-05 21:54:13,846 - Epoch: [153][ 240/ 1236] Overall Loss 0.202643 Objective Loss 0.202643 LR 0.000250 Time 0.023841 +2023-10-05 21:54:14,050 - Epoch: [153][ 250/ 1236] Overall Loss 0.202972 Objective Loss 0.202972 LR 0.000250 Time 0.023702 +2023-10-05 21:54:14,256 - Epoch: [153][ 260/ 1236] Overall Loss 0.202469 Objective Loss 0.202469 LR 0.000250 Time 0.023582 +2023-10-05 21:54:14,460 - Epoch: [153][ 270/ 1236] Overall Loss 0.202751 Objective Loss 0.202751 LR 0.000250 Time 0.023462 +2023-10-05 21:54:14,666 - Epoch: [153][ 280/ 1236] Overall Loss 0.202564 Objective Loss 0.202564 LR 0.000250 Time 0.023359 +2023-10-05 21:54:14,870 - Epoch: [153][ 290/ 1236] Overall Loss 0.202329 Objective Loss 0.202329 LR 0.000250 Time 0.023257 +2023-10-05 21:54:15,076 - Epoch: [153][ 300/ 1236] Overall Loss 0.202034 Objective Loss 0.202034 LR 0.000250 Time 0.023166 +2023-10-05 21:54:15,280 - Epoch: [153][ 310/ 1236] Overall Loss 0.201409 Objective Loss 0.201409 LR 0.000250 Time 0.023077 +2023-10-05 21:54:15,486 - Epoch: [153][ 320/ 1236] Overall Loss 0.200775 Objective Loss 0.200775 LR 0.000250 Time 0.022998 +2023-10-05 21:54:15,690 - Epoch: [153][ 330/ 1236] Overall Loss 0.201961 Objective Loss 0.201961 LR 0.000250 Time 0.022918 +2023-10-05 21:54:15,896 - Epoch: [153][ 340/ 1236] Overall Loss 0.201873 Objective Loss 0.201873 LR 0.000250 Time 0.022850 +2023-10-05 21:54:16,100 - Epoch: [153][ 350/ 1236] Overall Loss 0.202348 Objective Loss 0.202348 LR 0.000250 Time 0.022779 +2023-10-05 21:54:16,306 - Epoch: [153][ 360/ 1236] Overall Loss 0.202461 Objective Loss 0.202461 LR 0.000250 Time 0.022717 +2023-10-05 21:54:16,508 - Epoch: [153][ 370/ 1236] Overall Loss 0.202102 Objective Loss 0.202102 LR 0.000250 Time 0.022647 +2023-10-05 21:54:16,712 - Epoch: [153][ 380/ 1236] Overall Loss 0.202311 Objective Loss 0.202311 LR 0.000250 Time 0.022588 +2023-10-05 21:54:16,915 - Epoch: [153][ 390/ 1236] Overall Loss 0.202732 Objective Loss 0.202732 LR 0.000250 Time 0.022529 +2023-10-05 21:54:17,118 - Epoch: [153][ 400/ 1236] Overall Loss 0.202888 Objective Loss 0.202888 LR 0.000250 Time 0.022472 +2023-10-05 21:54:17,321 - Epoch: [153][ 410/ 1236] Overall Loss 0.203022 Objective Loss 0.203022 LR 0.000250 Time 0.022417 +2023-10-05 21:54:17,523 - Epoch: [153][ 420/ 1236] Overall Loss 0.203192 Objective Loss 0.203192 LR 0.000250 Time 0.022365 +2023-10-05 21:54:17,725 - Epoch: [153][ 430/ 1236] Overall Loss 0.203170 Objective Loss 0.203170 LR 0.000250 Time 0.022314 +2023-10-05 21:54:17,928 - Epoch: [153][ 440/ 1236] Overall Loss 0.203389 Objective Loss 0.203389 LR 0.000250 Time 0.022266 +2023-10-05 21:54:18,129 - Epoch: [153][ 450/ 1236] Overall Loss 0.203360 Objective Loss 0.203360 LR 0.000250 Time 0.022218 +2023-10-05 21:54:18,332 - Epoch: [153][ 460/ 1236] Overall Loss 0.203439 Objective Loss 0.203439 LR 0.000250 Time 0.022176 +2023-10-05 21:54:18,533 - Epoch: [153][ 470/ 1236] Overall Loss 0.203194 Objective Loss 0.203194 LR 0.000250 Time 0.022131 +2023-10-05 21:54:18,736 - Epoch: [153][ 480/ 1236] Overall Loss 0.203473 Objective Loss 0.203473 LR 0.000250 Time 0.022092 +2023-10-05 21:54:18,938 - Epoch: [153][ 490/ 1236] Overall Loss 0.204077 Objective Loss 0.204077 LR 0.000250 Time 0.022052 +2023-10-05 21:54:19,141 - Epoch: [153][ 500/ 1236] Overall Loss 0.203785 Objective Loss 0.203785 LR 0.000250 Time 0.022016 +2023-10-05 21:54:19,342 - Epoch: [153][ 510/ 1236] Overall Loss 0.203843 Objective Loss 0.203843 LR 0.000250 Time 0.021978 +2023-10-05 21:54:19,545 - Epoch: [153][ 520/ 1236] Overall Loss 0.204197 Objective Loss 0.204197 LR 0.000250 Time 0.021945 +2023-10-05 21:54:19,746 - Epoch: [153][ 530/ 1236] Overall Loss 0.204228 Objective Loss 0.204228 LR 0.000250 Time 0.021910 +2023-10-05 21:54:19,949 - Epoch: [153][ 540/ 1236] Overall Loss 0.204295 Objective Loss 0.204295 LR 0.000250 Time 0.021879 +2023-10-05 21:54:20,151 - Epoch: [153][ 550/ 1236] Overall Loss 0.204153 Objective Loss 0.204153 LR 0.000250 Time 0.021848 +2023-10-05 21:54:20,353 - Epoch: [153][ 560/ 1236] Overall Loss 0.203919 Objective Loss 0.203919 LR 0.000250 Time 0.021818 +2023-10-05 21:54:20,554 - Epoch: [153][ 570/ 1236] Overall Loss 0.203867 Objective Loss 0.203867 LR 0.000250 Time 0.021788 +2023-10-05 21:54:20,757 - Epoch: [153][ 580/ 1236] Overall Loss 0.204081 Objective Loss 0.204081 LR 0.000250 Time 0.021760 +2023-10-05 21:54:20,959 - Epoch: [153][ 590/ 1236] Overall Loss 0.204116 Objective Loss 0.204116 LR 0.000250 Time 0.021734 +2023-10-05 21:54:21,161 - Epoch: [153][ 600/ 1236] Overall Loss 0.204023 Objective Loss 0.204023 LR 0.000250 Time 0.021708 +2023-10-05 21:54:21,363 - Epoch: [153][ 610/ 1236] Overall Loss 0.204652 Objective Loss 0.204652 LR 0.000250 Time 0.021682 +2023-10-05 21:54:21,565 - Epoch: [153][ 620/ 1236] Overall Loss 0.205182 Objective Loss 0.205182 LR 0.000250 Time 0.021659 +2023-10-05 21:54:21,767 - Epoch: [153][ 630/ 1236] Overall Loss 0.205327 Objective Loss 0.205327 LR 0.000250 Time 0.021634 +2023-10-05 21:54:21,970 - Epoch: [153][ 640/ 1236] Overall Loss 0.205449 Objective Loss 0.205449 LR 0.000250 Time 0.021613 +2023-10-05 21:54:22,171 - Epoch: [153][ 650/ 1236] Overall Loss 0.205370 Objective Loss 0.205370 LR 0.000250 Time 0.021590 +2023-10-05 21:54:22,374 - Epoch: [153][ 660/ 1236] Overall Loss 0.205743 Objective Loss 0.205743 LR 0.000250 Time 0.021570 +2023-10-05 21:54:22,576 - Epoch: [153][ 670/ 1236] Overall Loss 0.205873 Objective Loss 0.205873 LR 0.000250 Time 0.021548 +2023-10-05 21:54:22,779 - Epoch: [153][ 680/ 1236] Overall Loss 0.205801 Objective Loss 0.205801 LR 0.000250 Time 0.021529 +2023-10-05 21:54:22,980 - Epoch: [153][ 690/ 1236] Overall Loss 0.205738 Objective Loss 0.205738 LR 0.000250 Time 0.021509 +2023-10-05 21:54:23,183 - Epoch: [153][ 700/ 1236] Overall Loss 0.205758 Objective Loss 0.205758 LR 0.000250 Time 0.021490 +2023-10-05 21:54:23,385 - Epoch: [153][ 710/ 1236] Overall Loss 0.205469 Objective Loss 0.205469 LR 0.000250 Time 0.021471 +2023-10-05 21:54:23,588 - Epoch: [153][ 720/ 1236] Overall Loss 0.205509 Objective Loss 0.205509 LR 0.000250 Time 0.021455 +2023-10-05 21:54:23,789 - Epoch: [153][ 730/ 1236] Overall Loss 0.205444 Objective Loss 0.205444 LR 0.000250 Time 0.021437 +2023-10-05 21:54:23,992 - Epoch: [153][ 740/ 1236] Overall Loss 0.205655 Objective Loss 0.205655 LR 0.000250 Time 0.021420 +2023-10-05 21:54:24,193 - Epoch: [153][ 750/ 1236] Overall Loss 0.205391 Objective Loss 0.205391 LR 0.000250 Time 0.021403 +2023-10-05 21:54:24,396 - Epoch: [153][ 760/ 1236] Overall Loss 0.205685 Objective Loss 0.205685 LR 0.000250 Time 0.021387 +2023-10-05 21:54:24,598 - Epoch: [153][ 770/ 1236] Overall Loss 0.205605 Objective Loss 0.205605 LR 0.000250 Time 0.021371 +2023-10-05 21:54:24,801 - Epoch: [153][ 780/ 1236] Overall Loss 0.205705 Objective Loss 0.205705 LR 0.000250 Time 0.021357 +2023-10-05 21:54:25,002 - Epoch: [153][ 790/ 1236] Overall Loss 0.205405 Objective Loss 0.205405 LR 0.000250 Time 0.021341 +2023-10-05 21:54:25,205 - Epoch: [153][ 800/ 1236] Overall Loss 0.205170 Objective Loss 0.205170 LR 0.000250 Time 0.021327 +2023-10-05 21:54:25,407 - Epoch: [153][ 810/ 1236] Overall Loss 0.205247 Objective Loss 0.205247 LR 0.000250 Time 0.021312 +2023-10-05 21:54:25,609 - Epoch: [153][ 820/ 1236] Overall Loss 0.205282 Objective Loss 0.205282 LR 0.000250 Time 0.021298 +2023-10-05 21:54:25,811 - Epoch: [153][ 830/ 1236] Overall Loss 0.205001 Objective Loss 0.205001 LR 0.000250 Time 0.021285 +2023-10-05 21:54:26,014 - Epoch: [153][ 840/ 1236] Overall Loss 0.204999 Objective Loss 0.204999 LR 0.000250 Time 0.021273 +2023-10-05 21:54:26,215 - Epoch: [153][ 850/ 1236] Overall Loss 0.204786 Objective Loss 0.204786 LR 0.000250 Time 0.021259 +2023-10-05 21:54:26,418 - Epoch: [153][ 860/ 1236] Overall Loss 0.204776 Objective Loss 0.204776 LR 0.000250 Time 0.021247 +2023-10-05 21:54:26,620 - Epoch: [153][ 870/ 1236] Overall Loss 0.204650 Objective Loss 0.204650 LR 0.000250 Time 0.021234 +2023-10-05 21:54:26,823 - Epoch: [153][ 880/ 1236] Overall Loss 0.204450 Objective Loss 0.204450 LR 0.000250 Time 0.021224 +2023-10-05 21:54:27,024 - Epoch: [153][ 890/ 1236] Overall Loss 0.204495 Objective Loss 0.204495 LR 0.000250 Time 0.021211 +2023-10-05 21:54:27,227 - Epoch: [153][ 900/ 1236] Overall Loss 0.204090 Objective Loss 0.204090 LR 0.000250 Time 0.021201 +2023-10-05 21:54:27,429 - Epoch: [153][ 910/ 1236] Overall Loss 0.204187 Objective Loss 0.204187 LR 0.000250 Time 0.021189 +2023-10-05 21:54:27,631 - Epoch: [153][ 920/ 1236] Overall Loss 0.204332 Objective Loss 0.204332 LR 0.000250 Time 0.021178 +2023-10-05 21:54:27,833 - Epoch: [153][ 930/ 1236] Overall Loss 0.204236 Objective Loss 0.204236 LR 0.000250 Time 0.021167 +2023-10-05 21:54:28,036 - Epoch: [153][ 940/ 1236] Overall Loss 0.204373 Objective Loss 0.204373 LR 0.000250 Time 0.021157 +2023-10-05 21:54:28,238 - Epoch: [153][ 950/ 1236] Overall Loss 0.204501 Objective Loss 0.204501 LR 0.000250 Time 0.021147 +2023-10-05 21:54:28,441 - Epoch: [153][ 960/ 1236] Overall Loss 0.204682 Objective Loss 0.204682 LR 0.000250 Time 0.021138 +2023-10-05 21:54:28,642 - Epoch: [153][ 970/ 1236] Overall Loss 0.204648 Objective Loss 0.204648 LR 0.000250 Time 0.021127 +2023-10-05 21:54:28,846 - Epoch: [153][ 980/ 1236] Overall Loss 0.204751 Objective Loss 0.204751 LR 0.000250 Time 0.021119 +2023-10-05 21:54:29,047 - Epoch: [153][ 990/ 1236] Overall Loss 0.204657 Objective Loss 0.204657 LR 0.000250 Time 0.021109 +2023-10-05 21:54:29,250 - Epoch: [153][ 1000/ 1236] Overall Loss 0.204987 Objective Loss 0.204987 LR 0.000250 Time 0.021100 +2023-10-05 21:54:29,452 - Epoch: [153][ 1010/ 1236] Overall Loss 0.205224 Objective Loss 0.205224 LR 0.000250 Time 0.021091 +2023-10-05 21:54:29,655 - Epoch: [153][ 1020/ 1236] Overall Loss 0.205288 Objective Loss 0.205288 LR 0.000250 Time 0.021082 +2023-10-05 21:54:29,857 - Epoch: [153][ 1030/ 1236] Overall Loss 0.205368 Objective Loss 0.205368 LR 0.000250 Time 0.021073 +2023-10-05 21:54:30,059 - Epoch: [153][ 1040/ 1236] Overall Loss 0.205581 Objective Loss 0.205581 LR 0.000250 Time 0.021065 +2023-10-05 21:54:30,261 - Epoch: [153][ 1050/ 1236] Overall Loss 0.205411 Objective Loss 0.205411 LR 0.000250 Time 0.021056 +2023-10-05 21:54:30,464 - Epoch: [153][ 1060/ 1236] Overall Loss 0.205160 Objective Loss 0.205160 LR 0.000250 Time 0.021049 +2023-10-05 21:54:30,665 - Epoch: [153][ 1070/ 1236] Overall Loss 0.205442 Objective Loss 0.205442 LR 0.000250 Time 0.021040 +2023-10-05 21:54:30,868 - Epoch: [153][ 1080/ 1236] Overall Loss 0.205481 Objective Loss 0.205481 LR 0.000250 Time 0.021032 +2023-10-05 21:54:31,070 - Epoch: [153][ 1090/ 1236] Overall Loss 0.205476 Objective Loss 0.205476 LR 0.000250 Time 0.021024 +2023-10-05 21:54:31,272 - Epoch: [153][ 1100/ 1236] Overall Loss 0.205507 Objective Loss 0.205507 LR 0.000250 Time 0.021017 +2023-10-05 21:54:31,474 - Epoch: [153][ 1110/ 1236] Overall Loss 0.205668 Objective Loss 0.205668 LR 0.000250 Time 0.021009 +2023-10-05 21:54:31,678 - Epoch: [153][ 1120/ 1236] Overall Loss 0.205742 Objective Loss 0.205742 LR 0.000250 Time 0.021003 +2023-10-05 21:54:31,879 - Epoch: [153][ 1130/ 1236] Overall Loss 0.205865 Objective Loss 0.205865 LR 0.000250 Time 0.020995 +2023-10-05 21:54:32,082 - Epoch: [153][ 1140/ 1236] Overall Loss 0.205730 Objective Loss 0.205730 LR 0.000250 Time 0.020988 +2023-10-05 21:54:32,283 - Epoch: [153][ 1150/ 1236] Overall Loss 0.206023 Objective Loss 0.206023 LR 0.000250 Time 0.020981 +2023-10-05 21:54:32,490 - Epoch: [153][ 1160/ 1236] Overall Loss 0.206025 Objective Loss 0.206025 LR 0.000250 Time 0.020978 +2023-10-05 21:54:32,695 - Epoch: [153][ 1170/ 1236] Overall Loss 0.205932 Objective Loss 0.205932 LR 0.000250 Time 0.020973 +2023-10-05 21:54:32,901 - Epoch: [153][ 1180/ 1236] Overall Loss 0.205906 Objective Loss 0.205906 LR 0.000250 Time 0.020970 +2023-10-05 21:54:33,105 - Epoch: [153][ 1190/ 1236] Overall Loss 0.205951 Objective Loss 0.205951 LR 0.000250 Time 0.020965 +2023-10-05 21:54:33,311 - Epoch: [153][ 1200/ 1236] Overall Loss 0.206177 Objective Loss 0.206177 LR 0.000250 Time 0.020961 +2023-10-05 21:54:33,515 - Epoch: [153][ 1210/ 1236] Overall Loss 0.205926 Objective Loss 0.205926 LR 0.000250 Time 0.020956 +2023-10-05 21:54:33,721 - Epoch: [153][ 1220/ 1236] Overall Loss 0.205881 Objective Loss 0.205881 LR 0.000250 Time 0.020953 +2023-10-05 21:54:33,981 - Epoch: [153][ 1230/ 1236] Overall Loss 0.205897 Objective Loss 0.205897 LR 0.000250 Time 0.020994 +2023-10-05 21:54:34,100 - Epoch: [153][ 1236/ 1236] Overall Loss 0.205833 Objective Loss 0.205833 Top1 90.020367 Top5 98.167006 LR 0.000250 Time 0.020988 +2023-10-05 21:54:34,226 - --- validate (epoch=153)----------- +2023-10-05 21:54:34,227 - 29943 samples (256 per mini-batch) +2023-10-05 21:54:34,685 - Epoch: [153][ 10/ 117] Loss 0.273539 Top1 85.625000 Top5 98.437500 +2023-10-05 21:54:34,833 - Epoch: [153][ 20/ 117] Loss 0.284055 Top1 85.859375 Top5 98.378906 +2023-10-05 21:54:34,981 - Epoch: [153][ 30/ 117] Loss 0.282825 Top1 85.872396 Top5 98.346354 +2023-10-05 21:54:35,129 - Epoch: [153][ 40/ 117] Loss 0.288650 Top1 85.673828 Top5 98.154297 +2023-10-05 21:54:35,277 - Epoch: [153][ 50/ 117] Loss 0.298041 Top1 85.578125 Top5 98.164062 +2023-10-05 21:54:35,425 - Epoch: [153][ 60/ 117] Loss 0.303658 Top1 85.390625 Top5 98.072917 +2023-10-05 21:54:35,573 - Epoch: [153][ 70/ 117] Loss 0.300874 Top1 85.546875 Top5 98.091518 +2023-10-05 21:54:35,722 - Epoch: [153][ 80/ 117] Loss 0.300835 Top1 85.590820 Top5 98.105469 +2023-10-05 21:54:35,871 - Epoch: [153][ 90/ 117] Loss 0.300565 Top1 85.499132 Top5 98.090278 +2023-10-05 21:54:36,020 - Epoch: [153][ 100/ 117] Loss 0.303970 Top1 85.433594 Top5 98.089844 +2023-10-05 21:54:36,174 - Epoch: [153][ 110/ 117] Loss 0.308388 Top1 85.344460 Top5 98.064631 +2023-10-05 21:54:36,260 - Epoch: [153][ 117/ 117] Loss 0.308219 Top1 85.302074 Top5 98.066326 +2023-10-05 21:54:36,354 - ==> Top1: 85.302 Top5: 98.066 Loss: 0.308 + +2023-10-05 21:54:36,354 - ==> Confusion: +[[ 918 2 3 0 4 3 0 0 5 86 2 0 2 2 5 1 2 1 0 0 14] + [ 2 1057 3 0 9 20 1 19 1 0 1 1 0 0 1 3 0 0 7 0 6] + [ 3 2 964 16 3 1 22 6 0 1 3 0 7 3 0 2 0 3 7 2 11] + [ 2 1 12 986 2 3 0 0 1 1 3 1 8 1 19 4 0 5 25 1 14] + [ 19 5 1 1 976 5 0 0 0 11 1 1 0 2 11 2 4 2 0 1 8] + [ 4 26 2 1 3 995 1 20 0 1 5 9 0 13 4 2 4 0 5 4 17] + [ 0 5 27 0 0 1 1134 5 0 0 0 3 1 0 1 4 0 0 1 3 6] + [ 4 19 15 0 4 23 5 1064 1 2 3 12 0 2 1 4 0 0 44 5 10] + [ 19 2 1 0 1 0 0 0 977 47 12 4 2 6 11 2 2 0 1 0 2] + [ 84 0 2 1 5 2 2 0 20 967 1 2 0 19 2 4 1 0 0 0 7] + [ 2 3 8 8 0 2 3 4 12 2 970 6 0 12 3 1 3 0 3 1 10] + [ 1 0 1 0 1 13 0 2 0 1 0 959 20 4 0 6 2 16 0 6 3] + [ 1 2 4 3 0 3 0 0 0 0 1 36 985 2 3 4 2 11 2 1 8] + [ 3 0 2 0 3 3 0 1 8 14 4 4 2 1058 4 2 1 0 0 0 10] + [ 13 1 3 9 3 0 0 0 28 2 1 1 1 3 1013 0 1 1 8 0 13] + [ 0 3 3 1 3 0 1 0 0 0 0 7 7 2 1 1068 12 14 0 10 2] + [ 0 6 2 0 8 7 0 0 3 0 0 3 0 0 2 11 1095 0 0 3 21] + [ 0 0 0 0 1 0 2 0 0 1 0 3 19 1 0 5 0 1002 1 0 3] + [ 1 5 7 15 1 0 0 22 1 1 2 0 1 1 8 0 0 0 996 1 6] + [ 1 2 1 3 1 6 8 7 1 0 3 16 4 1 0 6 7 1 2 1072 10] + [ 132 143 172 68 84 134 34 90 124 80 185 101 326 287 132 53 93 78 144 159 5286]] + +2023-10-05 21:54:36,356 - ==> Best [Top1: 85.302 Top5: 98.066 Sparsity:0.00 Params: 148928 on epoch: 153] +2023-10-05 21:54:36,356 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:54:36,369 - + +2023-10-05 21:54:36,369 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:54:37,494 - Epoch: [154][ 10/ 1236] Overall Loss 0.188345 Objective Loss 0.188345 LR 0.000250 Time 0.112448 +2023-10-05 21:54:37,700 - Epoch: [154][ 20/ 1236] Overall Loss 0.177240 Objective Loss 0.177240 LR 0.000250 Time 0.066506 +2023-10-05 21:54:37,904 - Epoch: [154][ 30/ 1236] Overall Loss 0.181561 Objective Loss 0.181561 LR 0.000250 Time 0.051115 +2023-10-05 21:54:38,111 - Epoch: [154][ 40/ 1236] Overall Loss 0.185800 Objective Loss 0.185800 LR 0.000250 Time 0.043494 +2023-10-05 21:54:38,314 - Epoch: [154][ 50/ 1236] Overall Loss 0.188685 Objective Loss 0.188685 LR 0.000250 Time 0.038865 +2023-10-05 21:54:38,520 - Epoch: [154][ 60/ 1236] Overall Loss 0.186929 Objective Loss 0.186929 LR 0.000250 Time 0.035818 +2023-10-05 21:54:38,724 - Epoch: [154][ 70/ 1236] Overall Loss 0.189490 Objective Loss 0.189490 LR 0.000250 Time 0.033598 +2023-10-05 21:54:38,929 - Epoch: [154][ 80/ 1236] Overall Loss 0.188334 Objective Loss 0.188334 LR 0.000250 Time 0.031967 +2023-10-05 21:54:39,133 - Epoch: [154][ 90/ 1236] Overall Loss 0.188953 Objective Loss 0.188953 LR 0.000250 Time 0.030673 +2023-10-05 21:54:39,338 - Epoch: [154][ 100/ 1236] Overall Loss 0.190884 Objective Loss 0.190884 LR 0.000250 Time 0.029656 +2023-10-05 21:54:39,543 - Epoch: [154][ 110/ 1236] Overall Loss 0.192915 Objective Loss 0.192915 LR 0.000250 Time 0.028821 +2023-10-05 21:54:39,749 - Epoch: [154][ 120/ 1236] Overall Loss 0.194590 Objective Loss 0.194590 LR 0.000250 Time 0.028130 +2023-10-05 21:54:39,954 - Epoch: [154][ 130/ 1236] Overall Loss 0.195529 Objective Loss 0.195529 LR 0.000250 Time 0.027537 +2023-10-05 21:54:40,159 - Epoch: [154][ 140/ 1236] Overall Loss 0.193411 Objective Loss 0.193411 LR 0.000250 Time 0.027035 +2023-10-05 21:54:40,364 - Epoch: [154][ 150/ 1236] Overall Loss 0.194986 Objective Loss 0.194986 LR 0.000250 Time 0.026595 +2023-10-05 21:54:40,569 - Epoch: [154][ 160/ 1236] Overall Loss 0.195354 Objective Loss 0.195354 LR 0.000250 Time 0.026215 +2023-10-05 21:54:40,773 - Epoch: [154][ 170/ 1236] Overall Loss 0.195118 Objective Loss 0.195118 LR 0.000250 Time 0.025873 +2023-10-05 21:54:40,980 - Epoch: [154][ 180/ 1236] Overall Loss 0.196695 Objective Loss 0.196695 LR 0.000250 Time 0.025578 +2023-10-05 21:54:41,184 - Epoch: [154][ 190/ 1236] Overall Loss 0.196240 Objective Loss 0.196240 LR 0.000250 Time 0.025308 +2023-10-05 21:54:41,390 - Epoch: [154][ 200/ 1236] Overall Loss 0.196752 Objective Loss 0.196752 LR 0.000250 Time 0.025070 +2023-10-05 21:54:41,595 - Epoch: [154][ 210/ 1236] Overall Loss 0.195932 Objective Loss 0.195932 LR 0.000250 Time 0.024848 +2023-10-05 21:54:41,800 - Epoch: [154][ 220/ 1236] Overall Loss 0.196532 Objective Loss 0.196532 LR 0.000250 Time 0.024652 +2023-10-05 21:54:42,005 - Epoch: [154][ 230/ 1236] Overall Loss 0.197376 Objective Loss 0.197376 LR 0.000250 Time 0.024468 +2023-10-05 21:54:42,211 - Epoch: [154][ 240/ 1236] Overall Loss 0.197658 Objective Loss 0.197658 LR 0.000250 Time 0.024305 +2023-10-05 21:54:42,416 - Epoch: [154][ 250/ 1236] Overall Loss 0.196973 Objective Loss 0.196973 LR 0.000250 Time 0.024151 +2023-10-05 21:54:42,621 - Epoch: [154][ 260/ 1236] Overall Loss 0.196846 Objective Loss 0.196846 LR 0.000250 Time 0.024010 +2023-10-05 21:54:42,825 - Epoch: [154][ 270/ 1236] Overall Loss 0.197012 Objective Loss 0.197012 LR 0.000250 Time 0.023877 +2023-10-05 21:54:43,031 - Epoch: [154][ 280/ 1236] Overall Loss 0.196454 Objective Loss 0.196454 LR 0.000250 Time 0.023757 +2023-10-05 21:54:43,235 - Epoch: [154][ 290/ 1236] Overall Loss 0.196689 Objective Loss 0.196689 LR 0.000250 Time 0.023641 +2023-10-05 21:54:43,441 - Epoch: [154][ 300/ 1236] Overall Loss 0.197537 Objective Loss 0.197537 LR 0.000250 Time 0.023537 +2023-10-05 21:54:43,645 - Epoch: [154][ 310/ 1236] Overall Loss 0.197361 Objective Loss 0.197361 LR 0.000250 Time 0.023436 +2023-10-05 21:54:43,851 - Epoch: [154][ 320/ 1236] Overall Loss 0.197267 Objective Loss 0.197267 LR 0.000250 Time 0.023345 +2023-10-05 21:54:44,055 - Epoch: [154][ 330/ 1236] Overall Loss 0.197246 Objective Loss 0.197246 LR 0.000250 Time 0.023256 +2023-10-05 21:54:44,261 - Epoch: [154][ 340/ 1236] Overall Loss 0.197866 Objective Loss 0.197866 LR 0.000250 Time 0.023176 +2023-10-05 21:54:44,465 - Epoch: [154][ 350/ 1236] Overall Loss 0.197361 Objective Loss 0.197361 LR 0.000250 Time 0.023097 +2023-10-05 21:54:44,671 - Epoch: [154][ 360/ 1236] Overall Loss 0.197388 Objective Loss 0.197388 LR 0.000250 Time 0.023025 +2023-10-05 21:54:44,875 - Epoch: [154][ 370/ 1236] Overall Loss 0.197922 Objective Loss 0.197922 LR 0.000250 Time 0.022955 +2023-10-05 21:54:45,080 - Epoch: [154][ 380/ 1236] Overall Loss 0.197963 Objective Loss 0.197963 LR 0.000250 Time 0.022887 +2023-10-05 21:54:45,282 - Epoch: [154][ 390/ 1236] Overall Loss 0.198052 Objective Loss 0.198052 LR 0.000250 Time 0.022819 +2023-10-05 21:54:45,486 - Epoch: [154][ 400/ 1236] Overall Loss 0.197938 Objective Loss 0.197938 LR 0.000250 Time 0.022756 +2023-10-05 21:54:45,687 - Epoch: [154][ 410/ 1236] Overall Loss 0.198069 Objective Loss 0.198069 LR 0.000250 Time 0.022691 +2023-10-05 21:54:45,890 - Epoch: [154][ 420/ 1236] Overall Loss 0.198086 Objective Loss 0.198086 LR 0.000250 Time 0.022634 +2023-10-05 21:54:46,091 - Epoch: [154][ 430/ 1236] Overall Loss 0.198538 Objective Loss 0.198538 LR 0.000250 Time 0.022573 +2023-10-05 21:54:46,296 - Epoch: [154][ 440/ 1236] Overall Loss 0.198620 Objective Loss 0.198620 LR 0.000250 Time 0.022526 +2023-10-05 21:54:46,498 - Epoch: [154][ 450/ 1236] Overall Loss 0.198472 Objective Loss 0.198472 LR 0.000250 Time 0.022473 +2023-10-05 21:54:46,702 - Epoch: [154][ 460/ 1236] Overall Loss 0.199167 Objective Loss 0.199167 LR 0.000250 Time 0.022427 +2023-10-05 21:54:46,903 - Epoch: [154][ 470/ 1236] Overall Loss 0.199391 Objective Loss 0.199391 LR 0.000250 Time 0.022377 +2023-10-05 21:54:47,106 - Epoch: [154][ 480/ 1236] Overall Loss 0.199648 Objective Loss 0.199648 LR 0.000250 Time 0.022334 +2023-10-05 21:54:47,308 - Epoch: [154][ 490/ 1236] Overall Loss 0.199787 Objective Loss 0.199787 LR 0.000250 Time 0.022289 +2023-10-05 21:54:47,511 - Epoch: [154][ 500/ 1236] Overall Loss 0.199642 Objective Loss 0.199642 LR 0.000250 Time 0.022250 +2023-10-05 21:54:47,713 - Epoch: [154][ 510/ 1236] Overall Loss 0.199710 Objective Loss 0.199710 LR 0.000250 Time 0.022207 +2023-10-05 21:54:47,916 - Epoch: [154][ 520/ 1236] Overall Loss 0.199917 Objective Loss 0.199917 LR 0.000250 Time 0.022171 +2023-10-05 21:54:48,117 - Epoch: [154][ 530/ 1236] Overall Loss 0.200093 Objective Loss 0.200093 LR 0.000250 Time 0.022131 +2023-10-05 21:54:48,320 - Epoch: [154][ 540/ 1236] Overall Loss 0.200406 Objective Loss 0.200406 LR 0.000250 Time 0.022097 +2023-10-05 21:54:48,522 - Epoch: [154][ 550/ 1236] Overall Loss 0.200464 Objective Loss 0.200464 LR 0.000250 Time 0.022060 +2023-10-05 21:54:48,725 - Epoch: [154][ 560/ 1236] Overall Loss 0.200408 Objective Loss 0.200408 LR 0.000250 Time 0.022029 +2023-10-05 21:54:48,926 - Epoch: [154][ 570/ 1236] Overall Loss 0.200379 Objective Loss 0.200379 LR 0.000250 Time 0.021994 +2023-10-05 21:54:49,129 - Epoch: [154][ 580/ 1236] Overall Loss 0.200144 Objective Loss 0.200144 LR 0.000250 Time 0.021965 +2023-10-05 21:54:49,331 - Epoch: [154][ 590/ 1236] Overall Loss 0.200127 Objective Loss 0.200127 LR 0.000250 Time 0.021933 +2023-10-05 21:54:49,534 - Epoch: [154][ 600/ 1236] Overall Loss 0.200431 Objective Loss 0.200431 LR 0.000250 Time 0.021906 +2023-10-05 21:54:49,735 - Epoch: [154][ 610/ 1236] Overall Loss 0.200253 Objective Loss 0.200253 LR 0.000250 Time 0.021876 +2023-10-05 21:54:49,939 - Epoch: [154][ 620/ 1236] Overall Loss 0.200251 Objective Loss 0.200251 LR 0.000250 Time 0.021851 +2023-10-05 21:54:50,140 - Epoch: [154][ 630/ 1236] Overall Loss 0.200080 Objective Loss 0.200080 LR 0.000250 Time 0.021823 +2023-10-05 21:54:50,343 - Epoch: [154][ 640/ 1236] Overall Loss 0.200141 Objective Loss 0.200141 LR 0.000250 Time 0.021799 +2023-10-05 21:54:50,544 - Epoch: [154][ 650/ 1236] Overall Loss 0.200025 Objective Loss 0.200025 LR 0.000250 Time 0.021772 +2023-10-05 21:54:50,748 - Epoch: [154][ 660/ 1236] Overall Loss 0.199801 Objective Loss 0.199801 LR 0.000250 Time 0.021750 +2023-10-05 21:54:50,949 - Epoch: [154][ 670/ 1236] Overall Loss 0.200025 Objective Loss 0.200025 LR 0.000250 Time 0.021725 +2023-10-05 21:54:51,153 - Epoch: [154][ 680/ 1236] Overall Loss 0.200221 Objective Loss 0.200221 LR 0.000250 Time 0.021705 +2023-10-05 21:54:51,354 - Epoch: [154][ 690/ 1236] Overall Loss 0.200590 Objective Loss 0.200590 LR 0.000250 Time 0.021682 +2023-10-05 21:54:51,557 - Epoch: [154][ 700/ 1236] Overall Loss 0.200477 Objective Loss 0.200477 LR 0.000250 Time 0.021662 +2023-10-05 21:54:51,759 - Epoch: [154][ 710/ 1236] Overall Loss 0.200784 Objective Loss 0.200784 LR 0.000250 Time 0.021640 +2023-10-05 21:54:51,960 - Epoch: [154][ 720/ 1236] Overall Loss 0.200664 Objective Loss 0.200664 LR 0.000250 Time 0.021619 +2023-10-05 21:54:52,162 - Epoch: [154][ 730/ 1236] Overall Loss 0.200757 Objective Loss 0.200757 LR 0.000250 Time 0.021598 +2023-10-05 21:54:52,366 - Epoch: [154][ 740/ 1236] Overall Loss 0.200939 Objective Loss 0.200939 LR 0.000250 Time 0.021581 +2023-10-05 21:54:52,567 - Epoch: [154][ 750/ 1236] Overall Loss 0.201100 Objective Loss 0.201100 LR 0.000250 Time 0.021561 +2023-10-05 21:54:52,771 - Epoch: [154][ 760/ 1236] Overall Loss 0.201029 Objective Loss 0.201029 LR 0.000250 Time 0.021546 +2023-10-05 21:54:52,972 - Epoch: [154][ 770/ 1236] Overall Loss 0.201155 Objective Loss 0.201155 LR 0.000250 Time 0.021527 +2023-10-05 21:54:53,176 - Epoch: [154][ 780/ 1236] Overall Loss 0.201613 Objective Loss 0.201613 LR 0.000250 Time 0.021512 +2023-10-05 21:54:53,377 - Epoch: [154][ 790/ 1236] Overall Loss 0.201426 Objective Loss 0.201426 LR 0.000250 Time 0.021493 +2023-10-05 21:54:53,581 - Epoch: [154][ 800/ 1236] Overall Loss 0.201406 Objective Loss 0.201406 LR 0.000250 Time 0.021479 +2023-10-05 21:54:53,782 - Epoch: [154][ 810/ 1236] Overall Loss 0.201888 Objective Loss 0.201888 LR 0.000250 Time 0.021462 +2023-10-05 21:54:53,986 - Epoch: [154][ 820/ 1236] Overall Loss 0.201819 Objective Loss 0.201819 LR 0.000250 Time 0.021448 +2023-10-05 21:54:54,187 - Epoch: [154][ 830/ 1236] Overall Loss 0.201929 Objective Loss 0.201929 LR 0.000250 Time 0.021431 +2023-10-05 21:54:54,391 - Epoch: [154][ 840/ 1236] Overall Loss 0.202039 Objective Loss 0.202039 LR 0.000250 Time 0.021419 +2023-10-05 21:54:54,592 - Epoch: [154][ 850/ 1236] Overall Loss 0.202152 Objective Loss 0.202152 LR 0.000250 Time 0.021403 +2023-10-05 21:54:54,796 - Epoch: [154][ 860/ 1236] Overall Loss 0.202338 Objective Loss 0.202338 LR 0.000250 Time 0.021391 +2023-10-05 21:54:54,997 - Epoch: [154][ 870/ 1236] Overall Loss 0.202295 Objective Loss 0.202295 LR 0.000250 Time 0.021375 +2023-10-05 21:54:55,201 - Epoch: [154][ 880/ 1236] Overall Loss 0.202519 Objective Loss 0.202519 LR 0.000250 Time 0.021364 +2023-10-05 21:54:55,402 - Epoch: [154][ 890/ 1236] Overall Loss 0.202684 Objective Loss 0.202684 LR 0.000250 Time 0.021349 +2023-10-05 21:54:55,606 - Epoch: [154][ 900/ 1236] Overall Loss 0.202448 Objective Loss 0.202448 LR 0.000250 Time 0.021338 +2023-10-05 21:54:55,807 - Epoch: [154][ 910/ 1236] Overall Loss 0.202490 Objective Loss 0.202490 LR 0.000250 Time 0.021324 +2023-10-05 21:54:56,011 - Epoch: [154][ 920/ 1236] Overall Loss 0.202629 Objective Loss 0.202629 LR 0.000250 Time 0.021314 +2023-10-05 21:54:56,212 - Epoch: [154][ 930/ 1236] Overall Loss 0.202483 Objective Loss 0.202483 LR 0.000250 Time 0.021301 +2023-10-05 21:54:56,416 - Epoch: [154][ 940/ 1236] Overall Loss 0.202710 Objective Loss 0.202710 LR 0.000250 Time 0.021290 +2023-10-05 21:54:56,617 - Epoch: [154][ 950/ 1236] Overall Loss 0.202893 Objective Loss 0.202893 LR 0.000250 Time 0.021278 +2023-10-05 21:54:56,821 - Epoch: [154][ 960/ 1236] Overall Loss 0.203020 Objective Loss 0.203020 LR 0.000250 Time 0.021268 +2023-10-05 21:54:57,022 - Epoch: [154][ 970/ 1236] Overall Loss 0.203327 Objective Loss 0.203327 LR 0.000250 Time 0.021256 +2023-10-05 21:54:57,226 - Epoch: [154][ 980/ 1236] Overall Loss 0.203443 Objective Loss 0.203443 LR 0.000250 Time 0.021247 +2023-10-05 21:54:57,427 - Epoch: [154][ 990/ 1236] Overall Loss 0.203515 Objective Loss 0.203515 LR 0.000250 Time 0.021235 +2023-10-05 21:54:57,631 - Epoch: [154][ 1000/ 1236] Overall Loss 0.203675 Objective Loss 0.203675 LR 0.000250 Time 0.021226 +2023-10-05 21:54:57,832 - Epoch: [154][ 1010/ 1236] Overall Loss 0.203479 Objective Loss 0.203479 LR 0.000250 Time 0.021215 +2023-10-05 21:54:58,036 - Epoch: [154][ 1020/ 1236] Overall Loss 0.203781 Objective Loss 0.203781 LR 0.000250 Time 0.021206 +2023-10-05 21:54:58,237 - Epoch: [154][ 1030/ 1236] Overall Loss 0.203837 Objective Loss 0.203837 LR 0.000250 Time 0.021195 +2023-10-05 21:54:58,441 - Epoch: [154][ 1040/ 1236] Overall Loss 0.203621 Objective Loss 0.203621 LR 0.000250 Time 0.021187 +2023-10-05 21:54:58,642 - Epoch: [154][ 1050/ 1236] Overall Loss 0.203803 Objective Loss 0.203803 LR 0.000250 Time 0.021176 +2023-10-05 21:54:58,846 - Epoch: [154][ 1060/ 1236] Overall Loss 0.204224 Objective Loss 0.204224 LR 0.000250 Time 0.021168 +2023-10-05 21:54:59,047 - Epoch: [154][ 1070/ 1236] Overall Loss 0.204382 Objective Loss 0.204382 LR 0.000250 Time 0.021158 +2023-10-05 21:54:59,251 - Epoch: [154][ 1080/ 1236] Overall Loss 0.204517 Objective Loss 0.204517 LR 0.000250 Time 0.021151 +2023-10-05 21:54:59,452 - Epoch: [154][ 1090/ 1236] Overall Loss 0.204379 Objective Loss 0.204379 LR 0.000250 Time 0.021141 +2023-10-05 21:54:59,656 - Epoch: [154][ 1100/ 1236] Overall Loss 0.204242 Objective Loss 0.204242 LR 0.000250 Time 0.021134 +2023-10-05 21:54:59,857 - Epoch: [154][ 1110/ 1236] Overall Loss 0.204337 Objective Loss 0.204337 LR 0.000250 Time 0.021124 +2023-10-05 21:55:00,061 - Epoch: [154][ 1120/ 1236] Overall Loss 0.204473 Objective Loss 0.204473 LR 0.000250 Time 0.021117 +2023-10-05 21:55:00,262 - Epoch: [154][ 1130/ 1236] Overall Loss 0.204454 Objective Loss 0.204454 LR 0.000250 Time 0.021108 +2023-10-05 21:55:00,466 - Epoch: [154][ 1140/ 1236] Overall Loss 0.204340 Objective Loss 0.204340 LR 0.000250 Time 0.021101 +2023-10-05 21:55:00,667 - Epoch: [154][ 1150/ 1236] Overall Loss 0.204248 Objective Loss 0.204248 LR 0.000250 Time 0.021093 +2023-10-05 21:55:00,871 - Epoch: [154][ 1160/ 1236] Overall Loss 0.204206 Objective Loss 0.204206 LR 0.000250 Time 0.021086 +2023-10-05 21:55:01,072 - Epoch: [154][ 1170/ 1236] Overall Loss 0.204099 Objective Loss 0.204099 LR 0.000250 Time 0.021078 +2023-10-05 21:55:01,276 - Epoch: [154][ 1180/ 1236] Overall Loss 0.204206 Objective Loss 0.204206 LR 0.000250 Time 0.021071 +2023-10-05 21:55:01,477 - Epoch: [154][ 1190/ 1236] Overall Loss 0.204253 Objective Loss 0.204253 LR 0.000250 Time 0.021063 +2023-10-05 21:55:01,681 - Epoch: [154][ 1200/ 1236] Overall Loss 0.204392 Objective Loss 0.204392 LR 0.000250 Time 0.021057 +2023-10-05 21:55:01,882 - Epoch: [154][ 1210/ 1236] Overall Loss 0.204309 Objective Loss 0.204309 LR 0.000250 Time 0.021049 +2023-10-05 21:55:02,086 - Epoch: [154][ 1220/ 1236] Overall Loss 0.204274 Objective Loss 0.204274 LR 0.000250 Time 0.021043 +2023-10-05 21:55:02,341 - Epoch: [154][ 1230/ 1236] Overall Loss 0.204099 Objective Loss 0.204099 LR 0.000250 Time 0.021079 +2023-10-05 21:55:02,460 - Epoch: [154][ 1236/ 1236] Overall Loss 0.204198 Objective Loss 0.204198 Top1 85.132383 Top5 97.148676 LR 0.000250 Time 0.021073 +2023-10-05 21:55:02,577 - --- validate (epoch=154)----------- +2023-10-05 21:55:02,578 - 29943 samples (256 per mini-batch) +2023-10-05 21:55:03,039 - Epoch: [154][ 10/ 117] Loss 0.310364 Top1 85.234375 Top5 98.242188 +2023-10-05 21:55:03,189 - Epoch: [154][ 20/ 117] Loss 0.316297 Top1 85.117188 Top5 98.085938 +2023-10-05 21:55:03,338 - Epoch: [154][ 30/ 117] Loss 0.307535 Top1 85.234375 Top5 98.164062 +2023-10-05 21:55:03,486 - Epoch: [154][ 40/ 117] Loss 0.302650 Top1 85.556641 Top5 98.232422 +2023-10-05 21:55:03,634 - Epoch: [154][ 50/ 117] Loss 0.300565 Top1 85.640625 Top5 98.265625 +2023-10-05 21:55:03,782 - Epoch: [154][ 60/ 117] Loss 0.305391 Top1 85.677083 Top5 98.216146 +2023-10-05 21:55:03,930 - Epoch: [154][ 70/ 117] Loss 0.307299 Top1 85.675223 Top5 98.158482 +2023-10-05 21:55:04,077 - Epoch: [154][ 80/ 117] Loss 0.305645 Top1 85.620117 Top5 98.134766 +2023-10-05 21:55:04,226 - Epoch: [154][ 90/ 117] Loss 0.309934 Top1 85.525174 Top5 98.133681 +2023-10-05 21:55:04,375 - Epoch: [154][ 100/ 117] Loss 0.307657 Top1 85.476562 Top5 98.125000 +2023-10-05 21:55:04,530 - Epoch: [154][ 110/ 117] Loss 0.308001 Top1 85.557528 Top5 98.121449 +2023-10-05 21:55:04,616 - Epoch: [154][ 117/ 117] Loss 0.310053 Top1 85.452359 Top5 98.109742 +2023-10-05 21:55:04,727 - ==> Top1: 85.452 Top5: 98.110 Loss: 0.310 + +2023-10-05 21:55:04,728 - ==> Confusion: +[[ 924 0 3 1 6 3 0 0 5 79 2 0 1 1 4 1 3 1 0 0 16] + [ 2 1057 2 0 8 15 2 18 0 0 3 2 0 0 2 2 4 0 10 1 3] + [ 2 4 958 15 1 0 32 7 0 3 4 0 5 3 2 3 0 3 5 1 8] + [ 5 1 11 981 1 3 2 1 1 1 6 1 8 1 17 3 0 5 23 1 17] + [ 23 9 2 0 969 3 0 0 1 10 0 2 0 1 8 2 10 2 0 2 6] + [ 3 36 0 0 3 984 0 23 0 0 6 6 1 12 8 2 4 0 3 5 20] + [ 0 6 24 0 0 0 1130 5 0 0 0 3 1 0 1 7 0 0 3 3 8] + [ 6 14 16 0 3 29 5 1055 1 3 3 10 3 3 1 1 2 0 45 6 12] + [ 17 2 1 0 2 3 1 0 976 41 7 2 1 10 11 4 1 1 6 0 3] + [ 88 0 1 1 6 4 0 0 17 956 1 1 0 22 6 7 0 1 0 1 7] + [ 3 3 7 6 0 1 3 3 10 2 975 5 0 10 4 1 3 1 6 1 9] + [ 2 0 2 0 0 12 0 1 0 0 0 962 19 7 0 4 1 16 0 5 4] + [ 0 1 5 5 0 1 0 0 1 0 1 40 972 3 1 6 2 13 5 3 9] + [ 3 0 1 0 1 8 0 1 8 13 8 4 2 1054 3 3 0 0 0 0 10] + [ 12 1 3 8 4 0 1 0 23 1 3 0 1 3 1012 0 1 2 12 0 14] + [ 2 3 2 1 3 0 2 0 0 0 1 8 6 2 1 1070 10 12 0 8 3] + [ 0 14 2 1 5 3 0 0 1 0 0 4 0 0 2 9 1103 0 1 3 13] + [ 0 0 1 2 1 0 2 0 0 0 0 1 13 2 1 3 0 1009 1 0 2] + [ 1 6 7 19 1 0 0 16 1 0 2 0 0 1 10 0 0 0 993 2 9] + [ 0 2 2 3 1 8 12 9 1 0 0 15 2 2 0 6 7 2 3 1064 13] + [ 127 162 123 55 88 122 42 82 96 68 171 100 301 269 136 50 134 68 154 174 5383]] + +2023-10-05 21:55:04,729 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:55:04,729 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:55:04,742 - + +2023-10-05 21:55:04,742 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:55:05,755 - Epoch: [155][ 10/ 1236] Overall Loss 0.203574 Objective Loss 0.203574 LR 0.000250 Time 0.101235 +2023-10-05 21:55:05,960 - Epoch: [155][ 20/ 1236] Overall Loss 0.193421 Objective Loss 0.193421 LR 0.000250 Time 0.060860 +2023-10-05 21:55:06,163 - Epoch: [155][ 30/ 1236] Overall Loss 0.193273 Objective Loss 0.193273 LR 0.000250 Time 0.047319 +2023-10-05 21:55:06,368 - Epoch: [155][ 40/ 1236] Overall Loss 0.197374 Objective Loss 0.197374 LR 0.000250 Time 0.040609 +2023-10-05 21:55:06,571 - Epoch: [155][ 50/ 1236] Overall Loss 0.200302 Objective Loss 0.200302 LR 0.000250 Time 0.036538 +2023-10-05 21:55:06,776 - Epoch: [155][ 60/ 1236] Overall Loss 0.200584 Objective Loss 0.200584 LR 0.000250 Time 0.033861 +2023-10-05 21:55:06,979 - Epoch: [155][ 70/ 1236] Overall Loss 0.202784 Objective Loss 0.202784 LR 0.000250 Time 0.031920 +2023-10-05 21:55:07,185 - Epoch: [155][ 80/ 1236] Overall Loss 0.202370 Objective Loss 0.202370 LR 0.000250 Time 0.030494 +2023-10-05 21:55:07,388 - Epoch: [155][ 90/ 1236] Overall Loss 0.201034 Objective Loss 0.201034 LR 0.000250 Time 0.029359 +2023-10-05 21:55:07,593 - Epoch: [155][ 100/ 1236] Overall Loss 0.198772 Objective Loss 0.198772 LR 0.000250 Time 0.028475 +2023-10-05 21:55:07,796 - Epoch: [155][ 110/ 1236] Overall Loss 0.198599 Objective Loss 0.198599 LR 0.000250 Time 0.027726 +2023-10-05 21:55:07,999 - Epoch: [155][ 120/ 1236] Overall Loss 0.197397 Objective Loss 0.197397 LR 0.000250 Time 0.027107 +2023-10-05 21:55:08,201 - Epoch: [155][ 130/ 1236] Overall Loss 0.197730 Objective Loss 0.197730 LR 0.000250 Time 0.026571 +2023-10-05 21:55:08,403 - Epoch: [155][ 140/ 1236] Overall Loss 0.199233 Objective Loss 0.199233 LR 0.000250 Time 0.026112 +2023-10-05 21:55:08,605 - Epoch: [155][ 150/ 1236] Overall Loss 0.198344 Objective Loss 0.198344 LR 0.000250 Time 0.025716 +2023-10-05 21:55:08,808 - Epoch: [155][ 160/ 1236] Overall Loss 0.197209 Objective Loss 0.197209 LR 0.000250 Time 0.025378 +2023-10-05 21:55:09,010 - Epoch: [155][ 170/ 1236] Overall Loss 0.198168 Objective Loss 0.198168 LR 0.000250 Time 0.025071 +2023-10-05 21:55:09,213 - Epoch: [155][ 180/ 1236] Overall Loss 0.197574 Objective Loss 0.197574 LR 0.000250 Time 0.024805 +2023-10-05 21:55:09,416 - Epoch: [155][ 190/ 1236] Overall Loss 0.198066 Objective Loss 0.198066 LR 0.000250 Time 0.024562 +2023-10-05 21:55:09,617 - Epoch: [155][ 200/ 1236] Overall Loss 0.199045 Objective Loss 0.199045 LR 0.000250 Time 0.024339 +2023-10-05 21:55:09,819 - Epoch: [155][ 210/ 1236] Overall Loss 0.198711 Objective Loss 0.198711 LR 0.000250 Time 0.024141 +2023-10-05 21:55:10,023 - Epoch: [155][ 220/ 1236] Overall Loss 0.199053 Objective Loss 0.199053 LR 0.000250 Time 0.023967 +2023-10-05 21:55:10,227 - Epoch: [155][ 230/ 1236] Overall Loss 0.199302 Objective Loss 0.199302 LR 0.000250 Time 0.023812 +2023-10-05 21:55:10,440 - Epoch: [155][ 240/ 1236] Overall Loss 0.198787 Objective Loss 0.198787 LR 0.000250 Time 0.023706 +2023-10-05 21:55:10,647 - Epoch: [155][ 250/ 1236] Overall Loss 0.198503 Objective Loss 0.198503 LR 0.000250 Time 0.023585 +2023-10-05 21:55:10,859 - Epoch: [155][ 260/ 1236] Overall Loss 0.198548 Objective Loss 0.198548 LR 0.000250 Time 0.023491 +2023-10-05 21:55:11,065 - Epoch: [155][ 270/ 1236] Overall Loss 0.198587 Objective Loss 0.198587 LR 0.000250 Time 0.023385 +2023-10-05 21:55:11,277 - Epoch: [155][ 280/ 1236] Overall Loss 0.198841 Objective Loss 0.198841 LR 0.000250 Time 0.023306 +2023-10-05 21:55:11,484 - Epoch: [155][ 290/ 1236] Overall Loss 0.198410 Objective Loss 0.198410 LR 0.000250 Time 0.023214 +2023-10-05 21:55:11,696 - Epoch: [155][ 300/ 1236] Overall Loss 0.198176 Objective Loss 0.198176 LR 0.000250 Time 0.023147 +2023-10-05 21:55:11,904 - Epoch: [155][ 310/ 1236] Overall Loss 0.197897 Objective Loss 0.197897 LR 0.000250 Time 0.023068 +2023-10-05 21:55:12,116 - Epoch: [155][ 320/ 1236] Overall Loss 0.197611 Objective Loss 0.197611 LR 0.000250 Time 0.023008 +2023-10-05 21:55:12,318 - Epoch: [155][ 330/ 1236] Overall Loss 0.198736 Objective Loss 0.198736 LR 0.000250 Time 0.022924 +2023-10-05 21:55:12,520 - Epoch: [155][ 340/ 1236] Overall Loss 0.198330 Objective Loss 0.198330 LR 0.000250 Time 0.022842 +2023-10-05 21:55:12,721 - Epoch: [155][ 350/ 1236] Overall Loss 0.199107 Objective Loss 0.199107 LR 0.000250 Time 0.022764 +2023-10-05 21:55:12,925 - Epoch: [155][ 360/ 1236] Overall Loss 0.200203 Objective Loss 0.200203 LR 0.000250 Time 0.022698 +2023-10-05 21:55:13,128 - Epoch: [155][ 370/ 1236] Overall Loss 0.199950 Objective Loss 0.199950 LR 0.000250 Time 0.022630 +2023-10-05 21:55:13,333 - Epoch: [155][ 380/ 1236] Overall Loss 0.200274 Objective Loss 0.200274 LR 0.000250 Time 0.022573 +2023-10-05 21:55:13,535 - Epoch: [155][ 390/ 1236] Overall Loss 0.200718 Objective Loss 0.200718 LR 0.000250 Time 0.022511 +2023-10-05 21:55:13,740 - Epoch: [155][ 400/ 1236] Overall Loss 0.201182 Objective Loss 0.201182 LR 0.000250 Time 0.022461 +2023-10-05 21:55:13,941 - Epoch: [155][ 410/ 1236] Overall Loss 0.201108 Objective Loss 0.201108 LR 0.000250 Time 0.022403 +2023-10-05 21:55:14,146 - Epoch: [155][ 420/ 1236] Overall Loss 0.201136 Objective Loss 0.201136 LR 0.000250 Time 0.022355 +2023-10-05 21:55:14,348 - Epoch: [155][ 430/ 1236] Overall Loss 0.201735 Objective Loss 0.201735 LR 0.000250 Time 0.022305 +2023-10-05 21:55:14,553 - Epoch: [155][ 440/ 1236] Overall Loss 0.201936 Objective Loss 0.201936 LR 0.000250 Time 0.022264 +2023-10-05 21:55:14,756 - Epoch: [155][ 450/ 1236] Overall Loss 0.202226 Objective Loss 0.202226 LR 0.000250 Time 0.022218 +2023-10-05 21:55:14,961 - Epoch: [155][ 460/ 1236] Overall Loss 0.201782 Objective Loss 0.201782 LR 0.000250 Time 0.022181 +2023-10-05 21:55:15,163 - Epoch: [155][ 470/ 1236] Overall Loss 0.202459 Objective Loss 0.202459 LR 0.000250 Time 0.022138 +2023-10-05 21:55:15,368 - Epoch: [155][ 480/ 1236] Overall Loss 0.202652 Objective Loss 0.202652 LR 0.000250 Time 0.022104 +2023-10-05 21:55:15,570 - Epoch: [155][ 490/ 1236] Overall Loss 0.202435 Objective Loss 0.202435 LR 0.000250 Time 0.022064 +2023-10-05 21:55:15,776 - Epoch: [155][ 500/ 1236] Overall Loss 0.202610 Objective Loss 0.202610 LR 0.000250 Time 0.022034 +2023-10-05 21:55:15,978 - Epoch: [155][ 510/ 1236] Overall Loss 0.202482 Objective Loss 0.202482 LR 0.000250 Time 0.021997 +2023-10-05 21:55:16,183 - Epoch: [155][ 520/ 1236] Overall Loss 0.202634 Objective Loss 0.202634 LR 0.000250 Time 0.021968 +2023-10-05 21:55:16,386 - Epoch: [155][ 530/ 1236] Overall Loss 0.202505 Objective Loss 0.202505 LR 0.000250 Time 0.021934 +2023-10-05 21:55:16,591 - Epoch: [155][ 540/ 1236] Overall Loss 0.202657 Objective Loss 0.202657 LR 0.000250 Time 0.021908 +2023-10-05 21:55:16,793 - Epoch: [155][ 550/ 1236] Overall Loss 0.202638 Objective Loss 0.202638 LR 0.000250 Time 0.021875 +2023-10-05 21:55:16,998 - Epoch: [155][ 560/ 1236] Overall Loss 0.202416 Objective Loss 0.202416 LR 0.000250 Time 0.021851 +2023-10-05 21:55:17,200 - Epoch: [155][ 570/ 1236] Overall Loss 0.202610 Objective Loss 0.202610 LR 0.000250 Time 0.021821 +2023-10-05 21:55:17,405 - Epoch: [155][ 580/ 1236] Overall Loss 0.202625 Objective Loss 0.202625 LR 0.000250 Time 0.021798 +2023-10-05 21:55:17,607 - Epoch: [155][ 590/ 1236] Overall Loss 0.202945 Objective Loss 0.202945 LR 0.000250 Time 0.021770 +2023-10-05 21:55:17,812 - Epoch: [155][ 600/ 1236] Overall Loss 0.202804 Objective Loss 0.202804 LR 0.000250 Time 0.021749 +2023-10-05 21:55:18,015 - Epoch: [155][ 610/ 1236] Overall Loss 0.202544 Objective Loss 0.202544 LR 0.000250 Time 0.021723 +2023-10-05 21:55:18,220 - Epoch: [155][ 620/ 1236] Overall Loss 0.202829 Objective Loss 0.202829 LR 0.000250 Time 0.021704 +2023-10-05 21:55:18,422 - Epoch: [155][ 630/ 1236] Overall Loss 0.202542 Objective Loss 0.202542 LR 0.000250 Time 0.021679 +2023-10-05 21:55:18,628 - Epoch: [155][ 640/ 1236] Overall Loss 0.202876 Objective Loss 0.202876 LR 0.000250 Time 0.021661 +2023-10-05 21:55:18,830 - Epoch: [155][ 650/ 1236] Overall Loss 0.202835 Objective Loss 0.202835 LR 0.000250 Time 0.021638 +2023-10-05 21:55:19,036 - Epoch: [155][ 660/ 1236] Overall Loss 0.202946 Objective Loss 0.202946 LR 0.000250 Time 0.021622 +2023-10-05 21:55:19,237 - Epoch: [155][ 670/ 1236] Overall Loss 0.202744 Objective Loss 0.202744 LR 0.000250 Time 0.021599 +2023-10-05 21:55:19,443 - Epoch: [155][ 680/ 1236] Overall Loss 0.202709 Objective Loss 0.202709 LR 0.000250 Time 0.021584 +2023-10-05 21:55:19,645 - Epoch: [155][ 690/ 1236] Overall Loss 0.202927 Objective Loss 0.202927 LR 0.000250 Time 0.021563 +2023-10-05 21:55:19,850 - Epoch: [155][ 700/ 1236] Overall Loss 0.202893 Objective Loss 0.202893 LR 0.000250 Time 0.021548 +2023-10-05 21:55:20,052 - Epoch: [155][ 710/ 1236] Overall Loss 0.202684 Objective Loss 0.202684 LR 0.000250 Time 0.021529 +2023-10-05 21:55:20,258 - Epoch: [155][ 720/ 1236] Overall Loss 0.202598 Objective Loss 0.202598 LR 0.000250 Time 0.021514 +2023-10-05 21:55:20,460 - Epoch: [155][ 730/ 1236] Overall Loss 0.202815 Objective Loss 0.202815 LR 0.000250 Time 0.021496 +2023-10-05 21:55:20,665 - Epoch: [155][ 740/ 1236] Overall Loss 0.203004 Objective Loss 0.203004 LR 0.000250 Time 0.021482 +2023-10-05 21:55:20,867 - Epoch: [155][ 750/ 1236] Overall Loss 0.203127 Objective Loss 0.203127 LR 0.000250 Time 0.021465 +2023-10-05 21:55:21,072 - Epoch: [155][ 760/ 1236] Overall Loss 0.203332 Objective Loss 0.203332 LR 0.000250 Time 0.021452 +2023-10-05 21:55:21,274 - Epoch: [155][ 770/ 1236] Overall Loss 0.203314 Objective Loss 0.203314 LR 0.000250 Time 0.021435 +2023-10-05 21:55:21,480 - Epoch: [155][ 780/ 1236] Overall Loss 0.203529 Objective Loss 0.203529 LR 0.000250 Time 0.021424 +2023-10-05 21:55:21,682 - Epoch: [155][ 790/ 1236] Overall Loss 0.203375 Objective Loss 0.203375 LR 0.000250 Time 0.021408 +2023-10-05 21:55:21,888 - Epoch: [155][ 800/ 1236] Overall Loss 0.203440 Objective Loss 0.203440 LR 0.000250 Time 0.021397 +2023-10-05 21:55:22,090 - Epoch: [155][ 810/ 1236] Overall Loss 0.203207 Objective Loss 0.203207 LR 0.000250 Time 0.021382 +2023-10-05 21:55:22,292 - Epoch: [155][ 820/ 1236] Overall Loss 0.203380 Objective Loss 0.203380 LR 0.000250 Time 0.021368 +2023-10-05 21:55:22,495 - Epoch: [155][ 830/ 1236] Overall Loss 0.203484 Objective Loss 0.203484 LR 0.000250 Time 0.021354 +2023-10-05 21:55:22,700 - Epoch: [155][ 840/ 1236] Overall Loss 0.203704 Objective Loss 0.203704 LR 0.000250 Time 0.021343 +2023-10-05 21:55:22,902 - Epoch: [155][ 850/ 1236] Overall Loss 0.203696 Objective Loss 0.203696 LR 0.000250 Time 0.021329 +2023-10-05 21:55:23,107 - Epoch: [155][ 860/ 1236] Overall Loss 0.203682 Objective Loss 0.203682 LR 0.000250 Time 0.021320 +2023-10-05 21:55:23,309 - Epoch: [155][ 870/ 1236] Overall Loss 0.203962 Objective Loss 0.203962 LR 0.000250 Time 0.021307 +2023-10-05 21:55:23,512 - Epoch: [155][ 880/ 1236] Overall Loss 0.204065 Objective Loss 0.204065 LR 0.000250 Time 0.021295 +2023-10-05 21:55:23,714 - Epoch: [155][ 890/ 1236] Overall Loss 0.204095 Objective Loss 0.204095 LR 0.000250 Time 0.021282 +2023-10-05 21:55:23,919 - Epoch: [155][ 900/ 1236] Overall Loss 0.204340 Objective Loss 0.204340 LR 0.000250 Time 0.021273 +2023-10-05 21:55:24,121 - Epoch: [155][ 910/ 1236] Overall Loss 0.204242 Objective Loss 0.204242 LR 0.000250 Time 0.021260 +2023-10-05 21:55:24,327 - Epoch: [155][ 920/ 1236] Overall Loss 0.204272 Objective Loss 0.204272 LR 0.000250 Time 0.021252 +2023-10-05 21:55:24,529 - Epoch: [155][ 930/ 1236] Overall Loss 0.204189 Objective Loss 0.204189 LR 0.000250 Time 0.021241 +2023-10-05 21:55:24,735 - Epoch: [155][ 940/ 1236] Overall Loss 0.203936 Objective Loss 0.203936 LR 0.000250 Time 0.021233 +2023-10-05 21:55:24,936 - Epoch: [155][ 950/ 1236] Overall Loss 0.203965 Objective Loss 0.203965 LR 0.000250 Time 0.021222 +2023-10-05 21:55:25,139 - Epoch: [155][ 960/ 1236] Overall Loss 0.204072 Objective Loss 0.204072 LR 0.000250 Time 0.021211 +2023-10-05 21:55:25,341 - Epoch: [155][ 970/ 1236] Overall Loss 0.203949 Objective Loss 0.203949 LR 0.000250 Time 0.021201 +2023-10-05 21:55:25,545 - Epoch: [155][ 980/ 1236] Overall Loss 0.204140 Objective Loss 0.204140 LR 0.000250 Time 0.021192 +2023-10-05 21:55:25,747 - Epoch: [155][ 990/ 1236] Overall Loss 0.203869 Objective Loss 0.203869 LR 0.000250 Time 0.021182 +2023-10-05 21:55:25,950 - Epoch: [155][ 1000/ 1236] Overall Loss 0.203737 Objective Loss 0.203737 LR 0.000250 Time 0.021172 +2023-10-05 21:55:26,152 - Epoch: [155][ 1010/ 1236] Overall Loss 0.203861 Objective Loss 0.203861 LR 0.000250 Time 0.021163 +2023-10-05 21:55:26,358 - Epoch: [155][ 1020/ 1236] Overall Loss 0.203999 Objective Loss 0.203999 LR 0.000250 Time 0.021156 +2023-10-05 21:55:26,560 - Epoch: [155][ 1030/ 1236] Overall Loss 0.204219 Objective Loss 0.204219 LR 0.000250 Time 0.021147 +2023-10-05 21:55:26,766 - Epoch: [155][ 1040/ 1236] Overall Loss 0.204485 Objective Loss 0.204485 LR 0.000250 Time 0.021141 +2023-10-05 21:55:26,965 - Epoch: [155][ 1050/ 1236] Overall Loss 0.204460 Objective Loss 0.204460 LR 0.000250 Time 0.021130 +2023-10-05 21:55:27,171 - Epoch: [155][ 1060/ 1236] Overall Loss 0.204441 Objective Loss 0.204441 LR 0.000250 Time 0.021124 +2023-10-05 21:55:27,373 - Epoch: [155][ 1070/ 1236] Overall Loss 0.204153 Objective Loss 0.204153 LR 0.000250 Time 0.021115 +2023-10-05 21:55:27,575 - Epoch: [155][ 1080/ 1236] Overall Loss 0.204001 Objective Loss 0.204001 LR 0.000250 Time 0.021106 +2023-10-05 21:55:27,774 - Epoch: [155][ 1090/ 1236] Overall Loss 0.204095 Objective Loss 0.204095 LR 0.000250 Time 0.021095 +2023-10-05 21:55:27,979 - Epoch: [155][ 1100/ 1236] Overall Loss 0.204099 Objective Loss 0.204099 LR 0.000250 Time 0.021090 +2023-10-05 21:55:28,181 - Epoch: [155][ 1110/ 1236] Overall Loss 0.204171 Objective Loss 0.204171 LR 0.000250 Time 0.021081 +2023-10-05 21:55:28,387 - Epoch: [155][ 1120/ 1236] Overall Loss 0.204227 Objective Loss 0.204227 LR 0.000250 Time 0.021077 +2023-10-05 21:55:28,589 - Epoch: [155][ 1130/ 1236] Overall Loss 0.204215 Objective Loss 0.204215 LR 0.000250 Time 0.021068 +2023-10-05 21:55:28,795 - Epoch: [155][ 1140/ 1236] Overall Loss 0.204109 Objective Loss 0.204109 LR 0.000250 Time 0.021064 +2023-10-05 21:55:28,997 - Epoch: [155][ 1150/ 1236] Overall Loss 0.204234 Objective Loss 0.204234 LR 0.000250 Time 0.021056 +2023-10-05 21:55:29,203 - Epoch: [155][ 1160/ 1236] Overall Loss 0.204346 Objective Loss 0.204346 LR 0.000250 Time 0.021051 +2023-10-05 21:55:29,405 - Epoch: [155][ 1170/ 1236] Overall Loss 0.204476 Objective Loss 0.204476 LR 0.000250 Time 0.021044 +2023-10-05 21:55:29,610 - Epoch: [155][ 1180/ 1236] Overall Loss 0.204332 Objective Loss 0.204332 LR 0.000250 Time 0.021040 +2023-10-05 21:55:29,812 - Epoch: [155][ 1190/ 1236] Overall Loss 0.204466 Objective Loss 0.204466 LR 0.000250 Time 0.021032 +2023-10-05 21:55:30,015 - Epoch: [155][ 1200/ 1236] Overall Loss 0.204586 Objective Loss 0.204586 LR 0.000250 Time 0.021025 +2023-10-05 21:55:30,229 - Epoch: [155][ 1210/ 1236] Overall Loss 0.204526 Objective Loss 0.204526 LR 0.000250 Time 0.021029 +2023-10-05 21:55:30,441 - Epoch: [155][ 1220/ 1236] Overall Loss 0.204619 Objective Loss 0.204619 LR 0.000250 Time 0.021030 +2023-10-05 21:55:30,703 - Epoch: [155][ 1230/ 1236] Overall Loss 0.204629 Objective Loss 0.204629 LR 0.000250 Time 0.021071 +2023-10-05 21:55:30,821 - Epoch: [155][ 1236/ 1236] Overall Loss 0.204495 Objective Loss 0.204495 Top1 89.816701 Top5 98.574338 LR 0.000250 Time 0.021064 +2023-10-05 21:55:30,952 - --- validate (epoch=155)----------- +2023-10-05 21:55:30,952 - 29943 samples (256 per mini-batch) +2023-10-05 21:55:31,404 - Epoch: [155][ 10/ 117] Loss 0.303250 Top1 84.375000 Top5 97.812500 +2023-10-05 21:55:31,556 - Epoch: [155][ 20/ 117] Loss 0.287325 Top1 84.980469 Top5 97.949219 +2023-10-05 21:55:31,704 - Epoch: [155][ 30/ 117] Loss 0.297042 Top1 84.622396 Top5 97.825521 +2023-10-05 21:55:31,854 - Epoch: [155][ 40/ 117] Loss 0.297873 Top1 84.648438 Top5 97.939453 +2023-10-05 21:55:32,002 - Epoch: [155][ 50/ 117] Loss 0.298949 Top1 84.617188 Top5 97.945312 +2023-10-05 21:55:32,152 - Epoch: [155][ 60/ 117] Loss 0.299104 Top1 84.869792 Top5 97.994792 +2023-10-05 21:55:32,299 - Epoch: [155][ 70/ 117] Loss 0.297443 Top1 85.078125 Top5 98.013393 +2023-10-05 21:55:32,450 - Epoch: [155][ 80/ 117] Loss 0.301884 Top1 85.048828 Top5 98.002930 +2023-10-05 21:55:32,597 - Epoch: [155][ 90/ 117] Loss 0.303172 Top1 85.039062 Top5 98.051215 +2023-10-05 21:55:32,747 - Epoch: [155][ 100/ 117] Loss 0.303997 Top1 85.054688 Top5 98.042969 +2023-10-05 21:55:32,901 - Epoch: [155][ 110/ 117] Loss 0.306968 Top1 84.989347 Top5 98.039773 +2023-10-05 21:55:32,986 - Epoch: [155][ 117/ 117] Loss 0.306667 Top1 85.024881 Top5 98.026250 +2023-10-05 21:55:33,102 - ==> Top1: 85.025 Top5: 98.026 Loss: 0.307 + +2023-10-05 21:55:33,103 - ==> Confusion: +[[ 930 2 3 2 6 2 0 0 3 79 1 1 1 2 3 1 3 1 0 0 10] + [ 1 1057 2 1 7 19 2 19 0 0 1 1 0 0 1 4 1 0 8 1 6] + [ 4 2 963 19 1 0 18 8 0 4 4 4 9 0 1 3 1 1 4 4 6] + [ 1 1 13 977 1 4 1 1 2 1 4 1 3 1 27 3 0 6 27 2 13] + [ 22 6 1 0 969 4 0 1 0 12 2 0 2 0 7 5 11 1 0 2 5] + [ 4 40 1 0 4 997 1 14 1 1 4 8 0 11 7 0 4 0 4 4 11] + [ 0 8 21 0 0 0 1119 8 0 1 4 5 1 0 1 11 1 1 2 5 3] + [ 2 18 9 0 2 27 3 1079 2 4 4 12 2 3 0 3 0 0 32 8 8] + [ 17 1 0 1 1 1 0 1 979 45 9 2 3 5 12 4 1 0 3 2 2] + [ 94 0 3 1 6 2 0 1 18 957 1 1 1 15 5 6 0 1 0 0 7] + [ 6 4 8 3 1 2 3 2 11 1 975 4 0 8 4 1 3 0 5 1 11] + [ 1 1 1 0 0 14 0 2 0 1 0 959 18 7 0 5 0 18 0 6 2] + [ 1 1 1 5 0 4 0 1 2 0 2 28 979 1 2 5 2 18 1 4 11] + [ 3 0 1 0 2 6 0 0 13 13 10 6 3 1043 4 3 1 1 0 0 10] + [ 15 3 2 11 4 0 0 0 25 2 1 1 1 3 1013 0 1 1 11 0 7] + [ 1 3 1 0 3 0 3 0 0 0 0 6 6 2 1 1073 10 13 0 8 4] + [ 1 14 1 0 5 4 0 1 1 0 0 4 0 0 2 9 1107 0 0 3 9] + [ 0 0 0 2 0 0 2 0 0 0 0 0 15 1 0 4 1 1008 1 0 4] + [ 1 7 7 18 1 0 0 17 2 1 3 0 1 0 7 0 1 0 995 0 7] + [ 0 2 2 2 1 7 6 11 1 0 1 14 4 1 0 7 10 1 3 1073 6] + [ 143 180 131 67 85 125 31 98 128 84 174 109 311 263 153 63 149 80 158 166 5207]] + +2023-10-05 21:55:33,104 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:55:33,104 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:55:33,110 - + +2023-10-05 21:55:33,110 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:55:34,097 - Epoch: [156][ 10/ 1236] Overall Loss 0.180086 Objective Loss 0.180086 LR 0.000250 Time 0.098589 +2023-10-05 21:55:34,299 - Epoch: [156][ 20/ 1236] Overall Loss 0.194470 Objective Loss 0.194470 LR 0.000250 Time 0.059398 +2023-10-05 21:55:34,501 - Epoch: [156][ 30/ 1236] Overall Loss 0.196507 Objective Loss 0.196507 LR 0.000250 Time 0.046310 +2023-10-05 21:55:34,703 - Epoch: [156][ 40/ 1236] Overall Loss 0.192730 Objective Loss 0.192730 LR 0.000250 Time 0.039782 +2023-10-05 21:55:34,905 - Epoch: [156][ 50/ 1236] Overall Loss 0.197141 Objective Loss 0.197141 LR 0.000250 Time 0.035846 +2023-10-05 21:55:35,107 - Epoch: [156][ 60/ 1236] Overall Loss 0.202058 Objective Loss 0.202058 LR 0.000250 Time 0.033236 +2023-10-05 21:55:35,309 - Epoch: [156][ 70/ 1236] Overall Loss 0.204079 Objective Loss 0.204079 LR 0.000250 Time 0.031365 +2023-10-05 21:55:35,511 - Epoch: [156][ 80/ 1236] Overall Loss 0.204634 Objective Loss 0.204634 LR 0.000250 Time 0.029970 +2023-10-05 21:55:35,714 - Epoch: [156][ 90/ 1236] Overall Loss 0.204254 Objective Loss 0.204254 LR 0.000250 Time 0.028895 +2023-10-05 21:55:35,918 - Epoch: [156][ 100/ 1236] Overall Loss 0.204028 Objective Loss 0.204028 LR 0.000250 Time 0.028041 +2023-10-05 21:55:36,122 - Epoch: [156][ 110/ 1236] Overall Loss 0.201249 Objective Loss 0.201249 LR 0.000250 Time 0.027344 +2023-10-05 21:55:36,327 - Epoch: [156][ 120/ 1236] Overall Loss 0.202699 Objective Loss 0.202699 LR 0.000250 Time 0.026770 +2023-10-05 21:55:36,532 - Epoch: [156][ 130/ 1236] Overall Loss 0.204472 Objective Loss 0.204472 LR 0.000250 Time 0.026281 +2023-10-05 21:55:36,737 - Epoch: [156][ 140/ 1236] Overall Loss 0.204517 Objective Loss 0.204517 LR 0.000250 Time 0.025864 +2023-10-05 21:55:36,942 - Epoch: [156][ 150/ 1236] Overall Loss 0.204651 Objective Loss 0.204651 LR 0.000250 Time 0.025503 +2023-10-05 21:55:37,146 - Epoch: [156][ 160/ 1236] Overall Loss 0.203130 Objective Loss 0.203130 LR 0.000250 Time 0.025179 +2023-10-05 21:55:37,346 - Epoch: [156][ 170/ 1236] Overall Loss 0.204469 Objective Loss 0.204469 LR 0.000250 Time 0.024873 +2023-10-05 21:55:37,547 - Epoch: [156][ 180/ 1236] Overall Loss 0.205872 Objective Loss 0.205872 LR 0.000250 Time 0.024608 +2023-10-05 21:55:37,747 - Epoch: [156][ 190/ 1236] Overall Loss 0.206433 Objective Loss 0.206433 LR 0.000250 Time 0.024364 +2023-10-05 21:55:37,948 - Epoch: [156][ 200/ 1236] Overall Loss 0.205950 Objective Loss 0.205950 LR 0.000250 Time 0.024150 +2023-10-05 21:55:38,149 - Epoch: [156][ 210/ 1236] Overall Loss 0.205337 Objective Loss 0.205337 LR 0.000250 Time 0.023952 +2023-10-05 21:55:38,350 - Epoch: [156][ 220/ 1236] Overall Loss 0.205475 Objective Loss 0.205475 LR 0.000250 Time 0.023777 +2023-10-05 21:55:38,551 - Epoch: [156][ 230/ 1236] Overall Loss 0.205498 Objective Loss 0.205498 LR 0.000250 Time 0.023615 +2023-10-05 21:55:38,751 - Epoch: [156][ 240/ 1236] Overall Loss 0.205282 Objective Loss 0.205282 LR 0.000250 Time 0.023465 +2023-10-05 21:55:38,952 - Epoch: [156][ 250/ 1236] Overall Loss 0.204720 Objective Loss 0.204720 LR 0.000250 Time 0.023329 +2023-10-05 21:55:39,153 - Epoch: [156][ 260/ 1236] Overall Loss 0.203770 Objective Loss 0.203770 LR 0.000250 Time 0.023202 +2023-10-05 21:55:39,354 - Epoch: [156][ 270/ 1236] Overall Loss 0.203909 Objective Loss 0.203909 LR 0.000250 Time 0.023087 +2023-10-05 21:55:39,555 - Epoch: [156][ 280/ 1236] Overall Loss 0.204490 Objective Loss 0.204490 LR 0.000250 Time 0.022977 +2023-10-05 21:55:39,756 - Epoch: [156][ 290/ 1236] Overall Loss 0.203920 Objective Loss 0.203920 LR 0.000250 Time 0.022876 +2023-10-05 21:55:39,957 - Epoch: [156][ 300/ 1236] Overall Loss 0.204600 Objective Loss 0.204600 LR 0.000250 Time 0.022782 +2023-10-05 21:55:40,157 - Epoch: [156][ 310/ 1236] Overall Loss 0.203601 Objective Loss 0.203601 LR 0.000250 Time 0.022693 +2023-10-05 21:55:40,359 - Epoch: [156][ 320/ 1236] Overall Loss 0.203498 Objective Loss 0.203498 LR 0.000250 Time 0.022613 +2023-10-05 21:55:40,560 - Epoch: [156][ 330/ 1236] Overall Loss 0.203502 Objective Loss 0.203502 LR 0.000250 Time 0.022536 +2023-10-05 21:55:40,761 - Epoch: [156][ 340/ 1236] Overall Loss 0.203602 Objective Loss 0.203602 LR 0.000250 Time 0.022464 +2023-10-05 21:55:40,962 - Epoch: [156][ 350/ 1236] Overall Loss 0.203772 Objective Loss 0.203772 LR 0.000250 Time 0.022395 +2023-10-05 21:55:41,163 - Epoch: [156][ 360/ 1236] Overall Loss 0.203775 Objective Loss 0.203775 LR 0.000250 Time 0.022330 +2023-10-05 21:55:41,365 - Epoch: [156][ 370/ 1236] Overall Loss 0.203462 Objective Loss 0.203462 LR 0.000250 Time 0.022271 +2023-10-05 21:55:41,568 - Epoch: [156][ 380/ 1236] Overall Loss 0.203414 Objective Loss 0.203414 LR 0.000250 Time 0.022218 +2023-10-05 21:55:41,770 - Epoch: [156][ 390/ 1236] Overall Loss 0.202635 Objective Loss 0.202635 LR 0.000250 Time 0.022165 +2023-10-05 21:55:41,972 - Epoch: [156][ 400/ 1236] Overall Loss 0.202343 Objective Loss 0.202343 LR 0.000250 Time 0.022117 +2023-10-05 21:55:42,175 - Epoch: [156][ 410/ 1236] Overall Loss 0.203004 Objective Loss 0.203004 LR 0.000250 Time 0.022070 +2023-10-05 21:55:42,378 - Epoch: [156][ 420/ 1236] Overall Loss 0.202921 Objective Loss 0.202921 LR 0.000250 Time 0.022026 +2023-10-05 21:55:42,580 - Epoch: [156][ 430/ 1236] Overall Loss 0.203211 Objective Loss 0.203211 LR 0.000250 Time 0.021985 +2023-10-05 21:55:42,782 - Epoch: [156][ 440/ 1236] Overall Loss 0.203276 Objective Loss 0.203276 LR 0.000250 Time 0.021944 +2023-10-05 21:55:42,985 - Epoch: [156][ 450/ 1236] Overall Loss 0.202880 Objective Loss 0.202880 LR 0.000250 Time 0.021904 +2023-10-05 21:55:43,188 - Epoch: [156][ 460/ 1236] Overall Loss 0.202891 Objective Loss 0.202891 LR 0.000250 Time 0.021868 +2023-10-05 21:55:43,390 - Epoch: [156][ 470/ 1236] Overall Loss 0.202880 Objective Loss 0.202880 LR 0.000250 Time 0.021833 +2023-10-05 21:55:43,593 - Epoch: [156][ 480/ 1236] Overall Loss 0.203110 Objective Loss 0.203110 LR 0.000250 Time 0.021800 +2023-10-05 21:55:43,795 - Epoch: [156][ 490/ 1236] Overall Loss 0.202910 Objective Loss 0.202910 LR 0.000250 Time 0.021766 +2023-10-05 21:55:43,998 - Epoch: [156][ 500/ 1236] Overall Loss 0.202850 Objective Loss 0.202850 LR 0.000250 Time 0.021736 +2023-10-05 21:55:44,199 - Epoch: [156][ 510/ 1236] Overall Loss 0.203132 Objective Loss 0.203132 LR 0.000250 Time 0.021705 +2023-10-05 21:55:44,403 - Epoch: [156][ 520/ 1236] Overall Loss 0.203347 Objective Loss 0.203347 LR 0.000250 Time 0.021677 +2023-10-05 21:55:44,605 - Epoch: [156][ 530/ 1236] Overall Loss 0.203005 Objective Loss 0.203005 LR 0.000250 Time 0.021649 +2023-10-05 21:55:44,808 - Epoch: [156][ 540/ 1236] Overall Loss 0.202703 Objective Loss 0.202703 LR 0.000250 Time 0.021623 +2023-10-05 21:55:45,010 - Epoch: [156][ 550/ 1236] Overall Loss 0.202377 Objective Loss 0.202377 LR 0.000250 Time 0.021597 +2023-10-05 21:55:45,213 - Epoch: [156][ 560/ 1236] Overall Loss 0.202329 Objective Loss 0.202329 LR 0.000250 Time 0.021572 +2023-10-05 21:55:45,415 - Epoch: [156][ 570/ 1236] Overall Loss 0.202095 Objective Loss 0.202095 LR 0.000250 Time 0.021548 +2023-10-05 21:55:45,617 - Epoch: [156][ 580/ 1236] Overall Loss 0.202544 Objective Loss 0.202544 LR 0.000250 Time 0.021525 +2023-10-05 21:55:45,820 - Epoch: [156][ 590/ 1236] Overall Loss 0.202571 Objective Loss 0.202571 LR 0.000250 Time 0.021504 +2023-10-05 21:55:46,023 - Epoch: [156][ 600/ 1236] Overall Loss 0.202518 Objective Loss 0.202518 LR 0.000250 Time 0.021482 +2023-10-05 21:55:46,225 - Epoch: [156][ 610/ 1236] Overall Loss 0.202653 Objective Loss 0.202653 LR 0.000250 Time 0.021461 +2023-10-05 21:55:46,428 - Epoch: [156][ 620/ 1236] Overall Loss 0.202630 Objective Loss 0.202630 LR 0.000250 Time 0.021441 +2023-10-05 21:55:46,630 - Epoch: [156][ 630/ 1236] Overall Loss 0.202603 Objective Loss 0.202603 LR 0.000250 Time 0.021421 +2023-10-05 21:55:46,833 - Epoch: [156][ 640/ 1236] Overall Loss 0.202996 Objective Loss 0.202996 LR 0.000250 Time 0.021402 +2023-10-05 21:55:47,035 - Epoch: [156][ 650/ 1236] Overall Loss 0.203216 Objective Loss 0.203216 LR 0.000250 Time 0.021384 +2023-10-05 21:55:47,238 - Epoch: [156][ 660/ 1236] Overall Loss 0.202988 Objective Loss 0.202988 LR 0.000250 Time 0.021367 +2023-10-05 21:55:47,440 - Epoch: [156][ 670/ 1236] Overall Loss 0.203139 Objective Loss 0.203139 LR 0.000250 Time 0.021349 +2023-10-05 21:55:47,643 - Epoch: [156][ 680/ 1236] Overall Loss 0.203387 Objective Loss 0.203387 LR 0.000250 Time 0.021333 +2023-10-05 21:55:47,845 - Epoch: [156][ 690/ 1236] Overall Loss 0.203203 Objective Loss 0.203203 LR 0.000250 Time 0.021316 +2023-10-05 21:55:48,048 - Epoch: [156][ 700/ 1236] Overall Loss 0.202729 Objective Loss 0.202729 LR 0.000250 Time 0.021301 +2023-10-05 21:55:48,250 - Epoch: [156][ 710/ 1236] Overall Loss 0.202781 Objective Loss 0.202781 LR 0.000250 Time 0.021285 +2023-10-05 21:55:48,453 - Epoch: [156][ 720/ 1236] Overall Loss 0.202871 Objective Loss 0.202871 LR 0.000250 Time 0.021270 +2023-10-05 21:55:48,655 - Epoch: [156][ 730/ 1236] Overall Loss 0.203003 Objective Loss 0.203003 LR 0.000250 Time 0.021256 +2023-10-05 21:55:48,858 - Epoch: [156][ 740/ 1236] Overall Loss 0.203156 Objective Loss 0.203156 LR 0.000250 Time 0.021242 +2023-10-05 21:55:49,060 - Epoch: [156][ 750/ 1236] Overall Loss 0.203475 Objective Loss 0.203475 LR 0.000250 Time 0.021227 +2023-10-05 21:55:49,263 - Epoch: [156][ 760/ 1236] Overall Loss 0.203337 Objective Loss 0.203337 LR 0.000250 Time 0.021214 +2023-10-05 21:55:49,465 - Epoch: [156][ 770/ 1236] Overall Loss 0.203183 Objective Loss 0.203183 LR 0.000250 Time 0.021201 +2023-10-05 21:55:49,668 - Epoch: [156][ 780/ 1236] Overall Loss 0.203205 Objective Loss 0.203205 LR 0.000250 Time 0.021188 +2023-10-05 21:55:49,870 - Epoch: [156][ 790/ 1236] Overall Loss 0.203026 Objective Loss 0.203026 LR 0.000250 Time 0.021176 +2023-10-05 21:55:50,073 - Epoch: [156][ 800/ 1236] Overall Loss 0.202949 Objective Loss 0.202949 LR 0.000250 Time 0.021164 +2023-10-05 21:55:50,275 - Epoch: [156][ 810/ 1236] Overall Loss 0.202536 Objective Loss 0.202536 LR 0.000250 Time 0.021152 +2023-10-05 21:55:50,478 - Epoch: [156][ 820/ 1236] Overall Loss 0.202654 Objective Loss 0.202654 LR 0.000250 Time 0.021141 +2023-10-05 21:55:50,680 - Epoch: [156][ 830/ 1236] Overall Loss 0.202777 Objective Loss 0.202777 LR 0.000250 Time 0.021129 +2023-10-05 21:55:50,883 - Epoch: [156][ 840/ 1236] Overall Loss 0.202891 Objective Loss 0.202891 LR 0.000250 Time 0.021119 +2023-10-05 21:55:51,085 - Epoch: [156][ 850/ 1236] Overall Loss 0.202933 Objective Loss 0.202933 LR 0.000250 Time 0.021108 +2023-10-05 21:55:51,288 - Epoch: [156][ 860/ 1236] Overall Loss 0.202901 Objective Loss 0.202901 LR 0.000250 Time 0.021098 +2023-10-05 21:55:51,490 - Epoch: [156][ 870/ 1236] Overall Loss 0.202954 Objective Loss 0.202954 LR 0.000250 Time 0.021087 +2023-10-05 21:55:51,693 - Epoch: [156][ 880/ 1236] Overall Loss 0.202874 Objective Loss 0.202874 LR 0.000250 Time 0.021077 +2023-10-05 21:55:51,895 - Epoch: [156][ 890/ 1236] Overall Loss 0.202801 Objective Loss 0.202801 LR 0.000250 Time 0.021067 +2023-10-05 21:55:52,108 - Epoch: [156][ 900/ 1236] Overall Loss 0.202688 Objective Loss 0.202688 LR 0.000250 Time 0.021070 +2023-10-05 21:55:52,318 - Epoch: [156][ 910/ 1236] Overall Loss 0.202317 Objective Loss 0.202317 LR 0.000250 Time 0.021068 +2023-10-05 21:55:52,521 - Epoch: [156][ 920/ 1236] Overall Loss 0.202285 Objective Loss 0.202285 LR 0.000250 Time 0.021060 +2023-10-05 21:55:52,723 - Epoch: [156][ 930/ 1236] Overall Loss 0.201982 Objective Loss 0.201982 LR 0.000250 Time 0.021050 +2023-10-05 21:55:52,926 - Epoch: [156][ 940/ 1236] Overall Loss 0.202204 Objective Loss 0.202204 LR 0.000250 Time 0.021042 +2023-10-05 21:55:53,128 - Epoch: [156][ 950/ 1236] Overall Loss 0.202240 Objective Loss 0.202240 LR 0.000250 Time 0.021033 +2023-10-05 21:55:53,331 - Epoch: [156][ 960/ 1236] Overall Loss 0.202024 Objective Loss 0.202024 LR 0.000250 Time 0.021025 +2023-10-05 21:55:53,533 - Epoch: [156][ 970/ 1236] Overall Loss 0.202058 Objective Loss 0.202058 LR 0.000250 Time 0.021016 +2023-10-05 21:55:53,736 - Epoch: [156][ 980/ 1236] Overall Loss 0.202120 Objective Loss 0.202120 LR 0.000250 Time 0.021008 +2023-10-05 21:55:53,938 - Epoch: [156][ 990/ 1236] Overall Loss 0.202082 Objective Loss 0.202082 LR 0.000250 Time 0.020999 +2023-10-05 21:55:54,141 - Epoch: [156][ 1000/ 1236] Overall Loss 0.202063 Objective Loss 0.202063 LR 0.000250 Time 0.020992 +2023-10-05 21:55:54,343 - Epoch: [156][ 1010/ 1236] Overall Loss 0.201869 Objective Loss 0.201869 LR 0.000250 Time 0.020984 +2023-10-05 21:55:54,546 - Epoch: [156][ 1020/ 1236] Overall Loss 0.201772 Objective Loss 0.201772 LR 0.000250 Time 0.020977 +2023-10-05 21:55:54,748 - Epoch: [156][ 1030/ 1236] Overall Loss 0.201524 Objective Loss 0.201524 LR 0.000250 Time 0.020969 +2023-10-05 21:55:54,951 - Epoch: [156][ 1040/ 1236] Overall Loss 0.201432 Objective Loss 0.201432 LR 0.000250 Time 0.020962 +2023-10-05 21:55:55,153 - Epoch: [156][ 1050/ 1236] Overall Loss 0.201224 Objective Loss 0.201224 LR 0.000250 Time 0.020954 +2023-10-05 21:55:55,356 - Epoch: [156][ 1060/ 1236] Overall Loss 0.201442 Objective Loss 0.201442 LR 0.000250 Time 0.020948 +2023-10-05 21:55:55,558 - Epoch: [156][ 1070/ 1236] Overall Loss 0.201450 Objective Loss 0.201450 LR 0.000250 Time 0.020941 +2023-10-05 21:55:55,761 - Epoch: [156][ 1080/ 1236] Overall Loss 0.201464 Objective Loss 0.201464 LR 0.000250 Time 0.020934 +2023-10-05 21:55:55,963 - Epoch: [156][ 1090/ 1236] Overall Loss 0.201617 Objective Loss 0.201617 LR 0.000250 Time 0.020927 +2023-10-05 21:55:56,166 - Epoch: [156][ 1100/ 1236] Overall Loss 0.201776 Objective Loss 0.201776 LR 0.000250 Time 0.020921 +2023-10-05 21:55:56,368 - Epoch: [156][ 1110/ 1236] Overall Loss 0.201690 Objective Loss 0.201690 LR 0.000250 Time 0.020914 +2023-10-05 21:55:56,572 - Epoch: [156][ 1120/ 1236] Overall Loss 0.201621 Objective Loss 0.201621 LR 0.000250 Time 0.020909 +2023-10-05 21:55:56,774 - Epoch: [156][ 1130/ 1236] Overall Loss 0.201480 Objective Loss 0.201480 LR 0.000250 Time 0.020902 +2023-10-05 21:55:56,977 - Epoch: [156][ 1140/ 1236] Overall Loss 0.201352 Objective Loss 0.201352 LR 0.000250 Time 0.020897 +2023-10-05 21:55:57,179 - Epoch: [156][ 1150/ 1236] Overall Loss 0.201366 Objective Loss 0.201366 LR 0.000250 Time 0.020890 +2023-10-05 21:55:57,382 - Epoch: [156][ 1160/ 1236] Overall Loss 0.201415 Objective Loss 0.201415 LR 0.000250 Time 0.020885 +2023-10-05 21:55:57,584 - Epoch: [156][ 1170/ 1236] Overall Loss 0.201416 Objective Loss 0.201416 LR 0.000250 Time 0.020879 +2023-10-05 21:55:57,787 - Epoch: [156][ 1180/ 1236] Overall Loss 0.201415 Objective Loss 0.201415 LR 0.000250 Time 0.020873 +2023-10-05 21:55:57,989 - Epoch: [156][ 1190/ 1236] Overall Loss 0.201363 Objective Loss 0.201363 LR 0.000250 Time 0.020868 +2023-10-05 21:55:58,192 - Epoch: [156][ 1200/ 1236] Overall Loss 0.201476 Objective Loss 0.201476 LR 0.000250 Time 0.020863 +2023-10-05 21:55:58,394 - Epoch: [156][ 1210/ 1236] Overall Loss 0.201355 Objective Loss 0.201355 LR 0.000250 Time 0.020857 +2023-10-05 21:55:58,597 - Epoch: [156][ 1220/ 1236] Overall Loss 0.201225 Objective Loss 0.201225 LR 0.000250 Time 0.020852 +2023-10-05 21:55:58,855 - Epoch: [156][ 1230/ 1236] Overall Loss 0.201243 Objective Loss 0.201243 LR 0.000250 Time 0.020891 +2023-10-05 21:55:58,972 - Epoch: [156][ 1236/ 1236] Overall Loss 0.201315 Objective Loss 0.201315 Top1 86.150713 Top5 98.778004 LR 0.000250 Time 0.020885 +2023-10-05 21:55:59,102 - --- validate (epoch=156)----------- +2023-10-05 21:55:59,102 - 29943 samples (256 per mini-batch) +2023-10-05 21:55:59,563 - Epoch: [156][ 10/ 117] Loss 0.292153 Top1 86.484375 Top5 98.515625 +2023-10-05 21:55:59,712 - Epoch: [156][ 20/ 117] Loss 0.300793 Top1 85.625000 Top5 98.300781 +2023-10-05 21:55:59,862 - Epoch: [156][ 30/ 117] Loss 0.303448 Top1 85.442708 Top5 98.281250 +2023-10-05 21:56:00,010 - Epoch: [156][ 40/ 117] Loss 0.301799 Top1 85.351562 Top5 98.095703 +2023-10-05 21:56:00,160 - Epoch: [156][ 50/ 117] Loss 0.305509 Top1 85.289062 Top5 98.117188 +2023-10-05 21:56:00,309 - Epoch: [156][ 60/ 117] Loss 0.303392 Top1 85.462240 Top5 98.020833 +2023-10-05 21:56:00,461 - Epoch: [156][ 70/ 117] Loss 0.305035 Top1 85.401786 Top5 98.058036 +2023-10-05 21:56:00,617 - Epoch: [156][ 80/ 117] Loss 0.305134 Top1 85.458984 Top5 98.041992 +2023-10-05 21:56:00,769 - Epoch: [156][ 90/ 117] Loss 0.307900 Top1 85.347222 Top5 98.007812 +2023-10-05 21:56:00,919 - Epoch: [156][ 100/ 117] Loss 0.309661 Top1 85.246094 Top5 98.003906 +2023-10-05 21:56:01,075 - Epoch: [156][ 110/ 117] Loss 0.308681 Top1 85.202415 Top5 98.039773 +2023-10-05 21:56:01,160 - Epoch: [156][ 117/ 117] Loss 0.307922 Top1 85.215242 Top5 98.049628 +2023-10-05 21:56:01,296 - ==> Top1: 85.215 Top5: 98.050 Loss: 0.308 + +2023-10-05 21:56:01,296 - ==> Confusion: +[[ 929 4 1 1 8 4 0 0 4 75 2 2 0 1 5 0 3 1 0 0 10] + [ 0 1065 3 0 5 16 1 19 0 0 1 3 0 0 1 2 1 0 7 0 7] + [ 4 1 962 17 2 0 18 11 0 3 5 2 8 1 1 3 0 1 6 4 7] + [ 2 1 8 977 1 7 2 1 2 1 8 1 4 1 23 3 0 5 26 0 16] + [ 20 7 0 0 974 4 0 1 0 10 1 1 0 1 11 2 8 2 0 3 5] + [ 5 39 0 1 5 999 1 19 0 0 6 5 1 9 5 1 2 0 3 3 12] + [ 0 6 28 0 0 0 1111 11 0 1 2 3 1 1 1 10 0 0 4 7 5] + [ 5 20 9 0 3 28 2 1072 3 4 1 12 0 3 1 1 0 0 38 8 8] + [ 16 3 0 0 2 1 0 0 976 49 11 1 2 9 11 3 0 1 2 0 2] + [ 88 1 1 0 6 1 1 1 20 967 1 3 1 13 5 1 0 1 0 0 8] + [ 2 10 9 4 1 1 2 2 11 4 974 3 2 10 4 0 1 0 2 1 10] + [ 2 0 0 0 0 16 0 3 0 0 0 955 19 6 0 1 1 15 0 14 3] + [ 2 2 2 4 0 3 1 1 1 0 0 34 990 0 1 3 2 14 1 2 5] + [ 1 0 1 0 3 15 0 0 10 20 6 2 2 1043 4 2 1 0 0 1 8] + [ 13 2 3 6 4 0 0 0 30 3 3 1 3 1 1008 0 1 1 12 0 10] + [ 0 2 2 0 3 0 0 0 1 0 1 6 6 3 1 1070 14 13 0 9 3] + [ 1 20 1 0 6 3 0 1 3 0 0 2 0 0 3 11 1097 0 0 3 10] + [ 0 0 0 0 1 0 3 0 0 0 0 3 17 1 1 3 0 1004 2 1 2] + [ 1 6 5 17 1 0 1 19 1 0 1 0 1 1 12 0 0 0 992 1 9] + [ 0 3 3 2 2 6 9 9 1 1 3 11 4 0 0 4 6 1 2 1078 7] + [ 126 205 130 55 92 160 39 86 112 103 164 120 327 267 127 50 100 76 136 157 5273]] + +2023-10-05 21:56:01,298 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:56:01,298 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:56:01,304 - + +2023-10-05 21:56:01,304 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:56:02,290 - Epoch: [157][ 10/ 1236] Overall Loss 0.198860 Objective Loss 0.198860 LR 0.000250 Time 0.098589 +2023-10-05 21:56:02,493 - Epoch: [157][ 20/ 1236] Overall Loss 0.193275 Objective Loss 0.193275 LR 0.000250 Time 0.059415 +2023-10-05 21:56:02,695 - Epoch: [157][ 30/ 1236] Overall Loss 0.199768 Objective Loss 0.199768 LR 0.000250 Time 0.046327 +2023-10-05 21:56:02,898 - Epoch: [157][ 40/ 1236] Overall Loss 0.196574 Objective Loss 0.196574 LR 0.000250 Time 0.039817 +2023-10-05 21:56:03,100 - Epoch: [157][ 50/ 1236] Overall Loss 0.198834 Objective Loss 0.198834 LR 0.000250 Time 0.035882 +2023-10-05 21:56:03,303 - Epoch: [157][ 60/ 1236] Overall Loss 0.195674 Objective Loss 0.195674 LR 0.000250 Time 0.033282 +2023-10-05 21:56:03,505 - Epoch: [157][ 70/ 1236] Overall Loss 0.194867 Objective Loss 0.194867 LR 0.000250 Time 0.031408 +2023-10-05 21:56:03,707 - Epoch: [157][ 80/ 1236] Overall Loss 0.198190 Objective Loss 0.198190 LR 0.000250 Time 0.030003 +2023-10-05 21:56:03,908 - Epoch: [157][ 90/ 1236] Overall Loss 0.195593 Objective Loss 0.195593 LR 0.000250 Time 0.028897 +2023-10-05 21:56:04,111 - Epoch: [157][ 100/ 1236] Overall Loss 0.196031 Objective Loss 0.196031 LR 0.000250 Time 0.028029 +2023-10-05 21:56:04,311 - Epoch: [157][ 110/ 1236] Overall Loss 0.194739 Objective Loss 0.194739 LR 0.000250 Time 0.027301 +2023-10-05 21:56:04,513 - Epoch: [157][ 120/ 1236] Overall Loss 0.194424 Objective Loss 0.194424 LR 0.000250 Time 0.026701 +2023-10-05 21:56:04,714 - Epoch: [157][ 130/ 1236] Overall Loss 0.193461 Objective Loss 0.193461 LR 0.000250 Time 0.026197 +2023-10-05 21:56:04,916 - Epoch: [157][ 140/ 1236] Overall Loss 0.194286 Objective Loss 0.194286 LR 0.000250 Time 0.025760 +2023-10-05 21:56:05,116 - Epoch: [157][ 150/ 1236] Overall Loss 0.194265 Objective Loss 0.194265 LR 0.000250 Time 0.025376 +2023-10-05 21:56:05,318 - Epoch: [157][ 160/ 1236] Overall Loss 0.195411 Objective Loss 0.195411 LR 0.000250 Time 0.025051 +2023-10-05 21:56:05,518 - Epoch: [157][ 170/ 1236] Overall Loss 0.195301 Objective Loss 0.195301 LR 0.000250 Time 0.024754 +2023-10-05 21:56:05,722 - Epoch: [157][ 180/ 1236] Overall Loss 0.196006 Objective Loss 0.196006 LR 0.000250 Time 0.024511 +2023-10-05 21:56:05,923 - Epoch: [157][ 190/ 1236] Overall Loss 0.195695 Objective Loss 0.195695 LR 0.000250 Time 0.024277 +2023-10-05 21:56:06,128 - Epoch: [157][ 200/ 1236] Overall Loss 0.195863 Objective Loss 0.195863 LR 0.000250 Time 0.024082 +2023-10-05 21:56:06,331 - Epoch: [157][ 210/ 1236] Overall Loss 0.195912 Objective Loss 0.195912 LR 0.000250 Time 0.023903 +2023-10-05 21:56:06,536 - Epoch: [157][ 220/ 1236] Overall Loss 0.195626 Objective Loss 0.195626 LR 0.000250 Time 0.023745 +2023-10-05 21:56:06,744 - Epoch: [157][ 230/ 1236] Overall Loss 0.196298 Objective Loss 0.196298 LR 0.000250 Time 0.023615 +2023-10-05 21:56:06,951 - Epoch: [157][ 240/ 1236] Overall Loss 0.196051 Objective Loss 0.196051 LR 0.000250 Time 0.023494 +2023-10-05 21:56:07,156 - Epoch: [157][ 250/ 1236] Overall Loss 0.195943 Objective Loss 0.195943 LR 0.000250 Time 0.023370 +2023-10-05 21:56:07,361 - Epoch: [157][ 260/ 1236] Overall Loss 0.195403 Objective Loss 0.195403 LR 0.000250 Time 0.023258 +2023-10-05 21:56:07,567 - Epoch: [157][ 270/ 1236] Overall Loss 0.195364 Objective Loss 0.195364 LR 0.000250 Time 0.023161 +2023-10-05 21:56:07,770 - Epoch: [157][ 280/ 1236] Overall Loss 0.196053 Objective Loss 0.196053 LR 0.000250 Time 0.023055 +2023-10-05 21:56:07,977 - Epoch: [157][ 290/ 1236] Overall Loss 0.196643 Objective Loss 0.196643 LR 0.000250 Time 0.022974 +2023-10-05 21:56:08,178 - Epoch: [157][ 300/ 1236] Overall Loss 0.196692 Objective Loss 0.196692 LR 0.000250 Time 0.022876 +2023-10-05 21:56:08,383 - Epoch: [157][ 310/ 1236] Overall Loss 0.197729 Objective Loss 0.197729 LR 0.000250 Time 0.022798 +2023-10-05 21:56:08,591 - Epoch: [157][ 320/ 1236] Overall Loss 0.197927 Objective Loss 0.197927 LR 0.000250 Time 0.022734 +2023-10-05 21:56:08,795 - Epoch: [157][ 330/ 1236] Overall Loss 0.197216 Objective Loss 0.197216 LR 0.000250 Time 0.022664 +2023-10-05 21:56:09,011 - Epoch: [157][ 340/ 1236] Overall Loss 0.196908 Objective Loss 0.196908 LR 0.000250 Time 0.022630 +2023-10-05 21:56:09,222 - Epoch: [157][ 350/ 1236] Overall Loss 0.196672 Objective Loss 0.196672 LR 0.000250 Time 0.022587 +2023-10-05 21:56:09,438 - Epoch: [157][ 360/ 1236] Overall Loss 0.196639 Objective Loss 0.196639 LR 0.000250 Time 0.022558 +2023-10-05 21:56:09,650 - Epoch: [157][ 370/ 1236] Overall Loss 0.197033 Objective Loss 0.197033 LR 0.000250 Time 0.022519 +2023-10-05 21:56:09,862 - Epoch: [157][ 380/ 1236] Overall Loss 0.197066 Objective Loss 0.197066 LR 0.000250 Time 0.022485 +2023-10-05 21:56:10,070 - Epoch: [157][ 390/ 1236] Overall Loss 0.197212 Objective Loss 0.197212 LR 0.000250 Time 0.022440 +2023-10-05 21:56:10,283 - Epoch: [157][ 400/ 1236] Overall Loss 0.197184 Objective Loss 0.197184 LR 0.000250 Time 0.022410 +2023-10-05 21:56:10,490 - Epoch: [157][ 410/ 1236] Overall Loss 0.197339 Objective Loss 0.197339 LR 0.000250 Time 0.022370 +2023-10-05 21:56:10,703 - Epoch: [157][ 420/ 1236] Overall Loss 0.198120 Objective Loss 0.198120 LR 0.000250 Time 0.022342 +2023-10-05 21:56:10,910 - Epoch: [157][ 430/ 1236] Overall Loss 0.197907 Objective Loss 0.197907 LR 0.000250 Time 0.022304 +2023-10-05 21:56:11,123 - Epoch: [157][ 440/ 1236] Overall Loss 0.197911 Objective Loss 0.197911 LR 0.000250 Time 0.022280 +2023-10-05 21:56:11,330 - Epoch: [157][ 450/ 1236] Overall Loss 0.197960 Objective Loss 0.197960 LR 0.000250 Time 0.022245 +2023-10-05 21:56:11,543 - Epoch: [157][ 460/ 1236] Overall Loss 0.197463 Objective Loss 0.197463 LR 0.000250 Time 0.022223 +2023-10-05 21:56:11,750 - Epoch: [157][ 470/ 1236] Overall Loss 0.197570 Objective Loss 0.197570 LR 0.000250 Time 0.022190 +2023-10-05 21:56:11,962 - Epoch: [157][ 480/ 1236] Overall Loss 0.197598 Objective Loss 0.197598 LR 0.000250 Time 0.022169 +2023-10-05 21:56:12,169 - Epoch: [157][ 490/ 1236] Overall Loss 0.197730 Objective Loss 0.197730 LR 0.000250 Time 0.022139 +2023-10-05 21:56:12,382 - Epoch: [157][ 500/ 1236] Overall Loss 0.197870 Objective Loss 0.197870 LR 0.000250 Time 0.022120 +2023-10-05 21:56:12,583 - Epoch: [157][ 510/ 1236] Overall Loss 0.197836 Objective Loss 0.197836 LR 0.000250 Time 0.022081 +2023-10-05 21:56:12,787 - Epoch: [157][ 520/ 1236] Overall Loss 0.198156 Objective Loss 0.198156 LR 0.000250 Time 0.022047 +2023-10-05 21:56:12,988 - Epoch: [157][ 530/ 1236] Overall Loss 0.198099 Objective Loss 0.198099 LR 0.000250 Time 0.022009 +2023-10-05 21:56:13,191 - Epoch: [157][ 540/ 1236] Overall Loss 0.197743 Objective Loss 0.197743 LR 0.000250 Time 0.021978 +2023-10-05 21:56:13,392 - Epoch: [157][ 550/ 1236] Overall Loss 0.198015 Objective Loss 0.198015 LR 0.000250 Time 0.021944 +2023-10-05 21:56:13,596 - Epoch: [157][ 560/ 1236] Overall Loss 0.197946 Objective Loss 0.197946 LR 0.000250 Time 0.021915 +2023-10-05 21:56:13,797 - Epoch: [157][ 570/ 1236] Overall Loss 0.197558 Objective Loss 0.197558 LR 0.000250 Time 0.021882 +2023-10-05 21:56:14,000 - Epoch: [157][ 580/ 1236] Overall Loss 0.197675 Objective Loss 0.197675 LR 0.000250 Time 0.021855 +2023-10-05 21:56:14,201 - Epoch: [157][ 590/ 1236] Overall Loss 0.197806 Objective Loss 0.197806 LR 0.000250 Time 0.021825 +2023-10-05 21:56:14,405 - Epoch: [157][ 600/ 1236] Overall Loss 0.197630 Objective Loss 0.197630 LR 0.000250 Time 0.021800 +2023-10-05 21:56:14,606 - Epoch: [157][ 610/ 1236] Overall Loss 0.197708 Objective Loss 0.197708 LR 0.000250 Time 0.021772 +2023-10-05 21:56:14,810 - Epoch: [157][ 620/ 1236] Overall Loss 0.197750 Objective Loss 0.197750 LR 0.000250 Time 0.021749 +2023-10-05 21:56:15,011 - Epoch: [157][ 630/ 1236] Overall Loss 0.197869 Objective Loss 0.197869 LR 0.000250 Time 0.021723 +2023-10-05 21:56:15,215 - Epoch: [157][ 640/ 1236] Overall Loss 0.198088 Objective Loss 0.198088 LR 0.000250 Time 0.021701 +2023-10-05 21:56:15,416 - Epoch: [157][ 650/ 1236] Overall Loss 0.197853 Objective Loss 0.197853 LR 0.000250 Time 0.021676 +2023-10-05 21:56:15,619 - Epoch: [157][ 660/ 1236] Overall Loss 0.198032 Objective Loss 0.198032 LR 0.000250 Time 0.021655 +2023-10-05 21:56:15,820 - Epoch: [157][ 670/ 1236] Overall Loss 0.198451 Objective Loss 0.198451 LR 0.000250 Time 0.021632 +2023-10-05 21:56:16,024 - Epoch: [157][ 680/ 1236] Overall Loss 0.198503 Objective Loss 0.198503 LR 0.000250 Time 0.021612 +2023-10-05 21:56:16,225 - Epoch: [157][ 690/ 1236] Overall Loss 0.198628 Objective Loss 0.198628 LR 0.000250 Time 0.021590 +2023-10-05 21:56:16,429 - Epoch: [157][ 700/ 1236] Overall Loss 0.198818 Objective Loss 0.198818 LR 0.000250 Time 0.021573 +2023-10-05 21:56:16,630 - Epoch: [157][ 710/ 1236] Overall Loss 0.199113 Objective Loss 0.199113 LR 0.000250 Time 0.021552 +2023-10-05 21:56:16,834 - Epoch: [157][ 720/ 1236] Overall Loss 0.199054 Objective Loss 0.199054 LR 0.000250 Time 0.021535 +2023-10-05 21:56:17,035 - Epoch: [157][ 730/ 1236] Overall Loss 0.199391 Objective Loss 0.199391 LR 0.000250 Time 0.021515 +2023-10-05 21:56:17,239 - Epoch: [157][ 740/ 1236] Overall Loss 0.199359 Objective Loss 0.199359 LR 0.000250 Time 0.021499 +2023-10-05 21:56:17,440 - Epoch: [157][ 750/ 1236] Overall Loss 0.199226 Objective Loss 0.199226 LR 0.000250 Time 0.021481 +2023-10-05 21:56:17,644 - Epoch: [157][ 760/ 1236] Overall Loss 0.199279 Objective Loss 0.199279 LR 0.000250 Time 0.021466 +2023-10-05 21:56:17,845 - Epoch: [157][ 770/ 1236] Overall Loss 0.199325 Objective Loss 0.199325 LR 0.000250 Time 0.021448 +2023-10-05 21:56:18,049 - Epoch: [157][ 780/ 1236] Overall Loss 0.199518 Objective Loss 0.199518 LR 0.000250 Time 0.021434 +2023-10-05 21:56:18,250 - Epoch: [157][ 790/ 1236] Overall Loss 0.199531 Objective Loss 0.199531 LR 0.000250 Time 0.021417 +2023-10-05 21:56:18,454 - Epoch: [157][ 800/ 1236] Overall Loss 0.199998 Objective Loss 0.199998 LR 0.000250 Time 0.021403 +2023-10-05 21:56:18,655 - Epoch: [157][ 810/ 1236] Overall Loss 0.200283 Objective Loss 0.200283 LR 0.000250 Time 0.021387 +2023-10-05 21:56:18,859 - Epoch: [157][ 820/ 1236] Overall Loss 0.200562 Objective Loss 0.200562 LR 0.000250 Time 0.021374 +2023-10-05 21:56:19,060 - Epoch: [157][ 830/ 1236] Overall Loss 0.200541 Objective Loss 0.200541 LR 0.000250 Time 0.021359 +2023-10-05 21:56:19,264 - Epoch: [157][ 840/ 1236] Overall Loss 0.200348 Objective Loss 0.200348 LR 0.000250 Time 0.021347 +2023-10-05 21:56:19,465 - Epoch: [157][ 850/ 1236] Overall Loss 0.200373 Objective Loss 0.200373 LR 0.000250 Time 0.021332 +2023-10-05 21:56:19,669 - Epoch: [157][ 860/ 1236] Overall Loss 0.200490 Objective Loss 0.200490 LR 0.000250 Time 0.021320 +2023-10-05 21:56:19,870 - Epoch: [157][ 870/ 1236] Overall Loss 0.200233 Objective Loss 0.200233 LR 0.000250 Time 0.021306 +2023-10-05 21:56:20,074 - Epoch: [157][ 880/ 1236] Overall Loss 0.200056 Objective Loss 0.200056 LR 0.000250 Time 0.021295 +2023-10-05 21:56:20,275 - Epoch: [157][ 890/ 1236] Overall Loss 0.199753 Objective Loss 0.199753 LR 0.000250 Time 0.021281 +2023-10-05 21:56:20,478 - Epoch: [157][ 900/ 1236] Overall Loss 0.199590 Objective Loss 0.199590 LR 0.000250 Time 0.021270 +2023-10-05 21:56:20,680 - Epoch: [157][ 910/ 1236] Overall Loss 0.199310 Objective Loss 0.199310 LR 0.000250 Time 0.021257 +2023-10-05 21:56:20,883 - Epoch: [157][ 920/ 1236] Overall Loss 0.199200 Objective Loss 0.199200 LR 0.000250 Time 0.021247 +2023-10-05 21:56:21,084 - Epoch: [157][ 930/ 1236] Overall Loss 0.199209 Objective Loss 0.199209 LR 0.000250 Time 0.021235 +2023-10-05 21:56:21,288 - Epoch: [157][ 940/ 1236] Overall Loss 0.199125 Objective Loss 0.199125 LR 0.000250 Time 0.021225 +2023-10-05 21:56:21,493 - Epoch: [157][ 950/ 1236] Overall Loss 0.199184 Objective Loss 0.199184 LR 0.000250 Time 0.021217 +2023-10-05 21:56:21,696 - Epoch: [157][ 960/ 1236] Overall Loss 0.198877 Objective Loss 0.198877 LR 0.000250 Time 0.021207 +2023-10-05 21:56:21,901 - Epoch: [157][ 970/ 1236] Overall Loss 0.198905 Objective Loss 0.198905 LR 0.000250 Time 0.021200 +2023-10-05 21:56:22,107 - Epoch: [157][ 980/ 1236] Overall Loss 0.198871 Objective Loss 0.198871 LR 0.000250 Time 0.021193 +2023-10-05 21:56:22,311 - Epoch: [157][ 990/ 1236] Overall Loss 0.199148 Objective Loss 0.199148 LR 0.000250 Time 0.021185 +2023-10-05 21:56:22,517 - Epoch: [157][ 1000/ 1236] Overall Loss 0.199375 Objective Loss 0.199375 LR 0.000250 Time 0.021179 +2023-10-05 21:56:22,722 - Epoch: [157][ 1010/ 1236] Overall Loss 0.199326 Objective Loss 0.199326 LR 0.000250 Time 0.021172 +2023-10-05 21:56:22,931 - Epoch: [157][ 1020/ 1236] Overall Loss 0.199399 Objective Loss 0.199399 LR 0.000250 Time 0.021169 +2023-10-05 21:56:23,137 - Epoch: [157][ 1030/ 1236] Overall Loss 0.199351 Objective Loss 0.199351 LR 0.000250 Time 0.021163 +2023-10-05 21:56:23,346 - Epoch: [157][ 1040/ 1236] Overall Loss 0.199493 Objective Loss 0.199493 LR 0.000250 Time 0.021159 +2023-10-05 21:56:23,554 - Epoch: [157][ 1050/ 1236] Overall Loss 0.199609 Objective Loss 0.199609 LR 0.000250 Time 0.021156 +2023-10-05 21:56:23,774 - Epoch: [157][ 1060/ 1236] Overall Loss 0.199560 Objective Loss 0.199560 LR 0.000250 Time 0.021164 +2023-10-05 21:56:24,006 - Epoch: [157][ 1070/ 1236] Overall Loss 0.199710 Objective Loss 0.199710 LR 0.000250 Time 0.021182 +2023-10-05 21:56:24,243 - Epoch: [157][ 1080/ 1236] Overall Loss 0.199653 Objective Loss 0.199653 LR 0.000250 Time 0.021205 +2023-10-05 21:56:24,460 - Epoch: [157][ 1090/ 1236] Overall Loss 0.199460 Objective Loss 0.199460 LR 0.000250 Time 0.021209 +2023-10-05 21:56:24,676 - Epoch: [157][ 1100/ 1236] Overall Loss 0.199480 Objective Loss 0.199480 LR 0.000250 Time 0.021213 +2023-10-05 21:56:24,888 - Epoch: [157][ 1110/ 1236] Overall Loss 0.199510 Objective Loss 0.199510 LR 0.000250 Time 0.021212 +2023-10-05 21:56:25,108 - Epoch: [157][ 1120/ 1236] Overall Loss 0.199689 Objective Loss 0.199689 LR 0.000250 Time 0.021219 +2023-10-05 21:56:25,318 - Epoch: [157][ 1130/ 1236] Overall Loss 0.199678 Objective Loss 0.199678 LR 0.000250 Time 0.021216 +2023-10-05 21:56:25,531 - Epoch: [157][ 1140/ 1236] Overall Loss 0.199899 Objective Loss 0.199899 LR 0.000250 Time 0.021216 +2023-10-05 21:56:25,738 - Epoch: [157][ 1150/ 1236] Overall Loss 0.200059 Objective Loss 0.200059 LR 0.000250 Time 0.021212 +2023-10-05 21:56:25,943 - Epoch: [157][ 1160/ 1236] Overall Loss 0.200219 Objective Loss 0.200219 LR 0.000250 Time 0.021205 +2023-10-05 21:56:26,150 - Epoch: [157][ 1170/ 1236] Overall Loss 0.200084 Objective Loss 0.200084 LR 0.000250 Time 0.021201 +2023-10-05 21:56:26,365 - Epoch: [157][ 1180/ 1236] Overall Loss 0.200001 Objective Loss 0.200001 LR 0.000250 Time 0.021203 +2023-10-05 21:56:26,569 - Epoch: [157][ 1190/ 1236] Overall Loss 0.199998 Objective Loss 0.199998 LR 0.000250 Time 0.021196 +2023-10-05 21:56:26,784 - Epoch: [157][ 1200/ 1236] Overall Loss 0.200126 Objective Loss 0.200126 LR 0.000250 Time 0.021198 +2023-10-05 21:56:26,994 - Epoch: [157][ 1210/ 1236] Overall Loss 0.200305 Objective Loss 0.200305 LR 0.000250 Time 0.021196 +2023-10-05 21:56:27,209 - Epoch: [157][ 1220/ 1236] Overall Loss 0.200220 Objective Loss 0.200220 LR 0.000250 Time 0.021198 +2023-10-05 21:56:27,474 - Epoch: [157][ 1230/ 1236] Overall Loss 0.200076 Objective Loss 0.200076 LR 0.000250 Time 0.021241 +2023-10-05 21:56:27,594 - Epoch: [157][ 1236/ 1236] Overall Loss 0.200065 Objective Loss 0.200065 Top1 87.780041 Top5 98.167006 LR 0.000250 Time 0.021235 +2023-10-05 21:56:27,710 - --- validate (epoch=157)----------- +2023-10-05 21:56:27,710 - 29943 samples (256 per mini-batch) +2023-10-05 21:56:28,182 - Epoch: [157][ 10/ 117] Loss 0.311415 Top1 84.726562 Top5 97.695312 +2023-10-05 21:56:28,332 - Epoch: [157][ 20/ 117] Loss 0.301485 Top1 85.000000 Top5 98.027344 +2023-10-05 21:56:28,483 - Epoch: [157][ 30/ 117] Loss 0.307910 Top1 85.182292 Top5 98.111979 +2023-10-05 21:56:28,631 - Epoch: [157][ 40/ 117] Loss 0.307874 Top1 85.019531 Top5 98.125000 +2023-10-05 21:56:28,781 - Epoch: [157][ 50/ 117] Loss 0.307670 Top1 84.882812 Top5 98.046875 +2023-10-05 21:56:28,931 - Epoch: [157][ 60/ 117] Loss 0.309542 Top1 84.824219 Top5 98.040365 +2023-10-05 21:56:29,080 - Epoch: [157][ 70/ 117] Loss 0.307265 Top1 85.005580 Top5 98.085938 +2023-10-05 21:56:29,229 - Epoch: [157][ 80/ 117] Loss 0.305449 Top1 85.019531 Top5 98.066406 +2023-10-05 21:56:29,378 - Epoch: [157][ 90/ 117] Loss 0.304842 Top1 84.982639 Top5 98.029514 +2023-10-05 21:56:29,527 - Epoch: [157][ 100/ 117] Loss 0.307855 Top1 84.886719 Top5 98.007812 +2023-10-05 21:56:29,683 - Epoch: [157][ 110/ 117] Loss 0.310016 Top1 84.889915 Top5 98.000710 +2023-10-05 21:56:29,769 - Epoch: [157][ 117/ 117] Loss 0.311066 Top1 84.807801 Top5 97.992853 +2023-10-05 21:56:29,889 - ==> Top1: 84.808 Top5: 97.993 Loss: 0.311 + +2023-10-05 21:56:29,890 - ==> Confusion: +[[ 907 2 5 0 4 2 0 0 8 90 1 0 1 2 4 2 2 1 0 0 19] + [ 1 1067 3 0 5 18 2 13 0 0 1 0 0 1 1 4 1 0 7 1 6] + [ 1 2 971 12 1 0 21 6 0 2 6 4 8 4 1 2 1 2 4 3 5] + [ 4 1 16 960 1 4 0 0 2 1 9 1 10 1 31 3 0 5 24 2 14] + [ 23 7 1 1 970 4 0 2 0 11 1 1 2 3 7 1 6 2 0 2 6] + [ 4 32 0 0 5 999 1 15 1 2 4 9 0 12 3 0 3 0 6 5 15] + [ 0 7 23 0 0 1 1122 5 0 1 2 3 1 0 1 12 0 0 1 6 6] + [ 3 25 19 0 2 29 2 1065 1 1 4 10 4 2 0 4 0 0 31 7 9] + [ 15 2 0 0 0 0 0 0 969 50 10 1 3 13 14 6 0 1 4 0 1] + [ 80 0 2 1 5 1 0 0 19 969 1 1 1 19 4 8 0 0 0 1 7] + [ 3 4 12 5 1 1 5 2 9 3 966 3 1 15 4 1 1 0 5 2 10] + [ 1 0 1 0 0 13 0 1 0 1 0 966 17 5 1 3 1 16 0 5 4] + [ 0 1 3 6 0 3 0 0 0 0 1 31 992 2 2 4 2 10 1 4 6] + [ 4 0 1 0 2 4 0 0 11 19 4 6 1 1052 4 2 0 0 0 0 9] + [ 13 1 2 7 3 0 0 0 22 3 3 1 2 3 1015 0 0 1 9 0 16] + [ 0 4 2 1 3 0 1 0 0 0 0 9 8 1 1 1070 12 12 0 8 2] + [ 1 13 1 0 9 5 0 1 2 0 0 3 0 2 3 7 1104 0 0 2 8] + [ 0 0 0 1 1 0 3 0 1 0 0 1 15 3 0 5 0 1002 1 0 5] + [ 2 10 10 15 1 1 0 23 1 0 1 1 2 0 11 0 0 0 981 1 8] + [ 0 3 5 0 1 10 6 6 1 0 1 15 3 2 0 6 11 2 1 1073 6] + [ 108 196 172 56 88 138 38 80 115 102 174 124 354 303 147 68 124 60 110 174 5174]] + +2023-10-05 21:56:29,891 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:56:29,891 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:56:29,897 - + +2023-10-05 21:56:29,898 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:56:31,024 - Epoch: [158][ 10/ 1236] Overall Loss 0.177652 Objective Loss 0.177652 LR 0.000250 Time 0.112585 +2023-10-05 21:56:31,226 - Epoch: [158][ 20/ 1236] Overall Loss 0.181043 Objective Loss 0.181043 LR 0.000250 Time 0.066370 +2023-10-05 21:56:31,427 - Epoch: [158][ 30/ 1236] Overall Loss 0.187919 Objective Loss 0.187919 LR 0.000250 Time 0.050926 +2023-10-05 21:56:31,629 - Epoch: [158][ 40/ 1236] Overall Loss 0.182934 Objective Loss 0.182934 LR 0.000250 Time 0.043240 +2023-10-05 21:56:31,830 - Epoch: [158][ 50/ 1236] Overall Loss 0.186261 Objective Loss 0.186261 LR 0.000250 Time 0.038604 +2023-10-05 21:56:32,032 - Epoch: [158][ 60/ 1236] Overall Loss 0.183977 Objective Loss 0.183977 LR 0.000250 Time 0.035537 +2023-10-05 21:56:32,233 - Epoch: [158][ 70/ 1236] Overall Loss 0.184342 Objective Loss 0.184342 LR 0.000250 Time 0.033321 +2023-10-05 21:56:32,435 - Epoch: [158][ 80/ 1236] Overall Loss 0.186728 Objective Loss 0.186728 LR 0.000250 Time 0.031678 +2023-10-05 21:56:32,635 - Epoch: [158][ 90/ 1236] Overall Loss 0.187161 Objective Loss 0.187161 LR 0.000250 Time 0.030381 +2023-10-05 21:56:32,836 - Epoch: [158][ 100/ 1236] Overall Loss 0.188381 Objective Loss 0.188381 LR 0.000250 Time 0.029350 +2023-10-05 21:56:33,042 - Epoch: [158][ 110/ 1236] Overall Loss 0.189155 Objective Loss 0.189155 LR 0.000250 Time 0.028547 +2023-10-05 21:56:33,249 - Epoch: [158][ 120/ 1236] Overall Loss 0.190978 Objective Loss 0.190978 LR 0.000250 Time 0.027887 +2023-10-05 21:56:33,454 - Epoch: [158][ 130/ 1236] Overall Loss 0.191748 Objective Loss 0.191748 LR 0.000250 Time 0.027320 +2023-10-05 21:56:33,661 - Epoch: [158][ 140/ 1236] Overall Loss 0.191591 Objective Loss 0.191591 LR 0.000250 Time 0.026841 +2023-10-05 21:56:33,865 - Epoch: [158][ 150/ 1236] Overall Loss 0.193413 Objective Loss 0.193413 LR 0.000250 Time 0.026409 +2023-10-05 21:56:34,071 - Epoch: [158][ 160/ 1236] Overall Loss 0.194087 Objective Loss 0.194087 LR 0.000250 Time 0.026043 +2023-10-05 21:56:34,277 - Epoch: [158][ 170/ 1236] Overall Loss 0.194201 Objective Loss 0.194201 LR 0.000250 Time 0.025718 +2023-10-05 21:56:34,484 - Epoch: [158][ 180/ 1236] Overall Loss 0.194580 Objective Loss 0.194580 LR 0.000250 Time 0.025434 +2023-10-05 21:56:34,690 - Epoch: [158][ 190/ 1236] Overall Loss 0.195350 Objective Loss 0.195350 LR 0.000250 Time 0.025179 +2023-10-05 21:56:34,897 - Epoch: [158][ 200/ 1236] Overall Loss 0.195254 Objective Loss 0.195254 LR 0.000250 Time 0.024951 +2023-10-05 21:56:35,103 - Epoch: [158][ 210/ 1236] Overall Loss 0.194504 Objective Loss 0.194504 LR 0.000250 Time 0.024740 +2023-10-05 21:56:35,312 - Epoch: [158][ 220/ 1236] Overall Loss 0.195037 Objective Loss 0.195037 LR 0.000250 Time 0.024565 +2023-10-05 21:56:35,521 - Epoch: [158][ 230/ 1236] Overall Loss 0.195258 Objective Loss 0.195258 LR 0.000250 Time 0.024403 +2023-10-05 21:56:35,731 - Epoch: [158][ 240/ 1236] Overall Loss 0.196371 Objective Loss 0.196371 LR 0.000250 Time 0.024259 +2023-10-05 21:56:35,941 - Epoch: [158][ 250/ 1236] Overall Loss 0.196864 Objective Loss 0.196864 LR 0.000250 Time 0.024121 +2023-10-05 21:56:36,150 - Epoch: [158][ 260/ 1236] Overall Loss 0.197042 Objective Loss 0.197042 LR 0.000250 Time 0.023991 +2023-10-05 21:56:36,359 - Epoch: [158][ 270/ 1236] Overall Loss 0.196693 Objective Loss 0.196693 LR 0.000250 Time 0.023874 +2023-10-05 21:56:36,568 - Epoch: [158][ 280/ 1236] Overall Loss 0.196685 Objective Loss 0.196685 LR 0.000250 Time 0.023767 +2023-10-05 21:56:36,776 - Epoch: [158][ 290/ 1236] Overall Loss 0.196736 Objective Loss 0.196736 LR 0.000250 Time 0.023662 +2023-10-05 21:56:36,985 - Epoch: [158][ 300/ 1236] Overall Loss 0.197578 Objective Loss 0.197578 LR 0.000250 Time 0.023567 +2023-10-05 21:56:37,193 - Epoch: [158][ 310/ 1236] Overall Loss 0.197660 Objective Loss 0.197660 LR 0.000250 Time 0.023474 +2023-10-05 21:56:37,404 - Epoch: [158][ 320/ 1236] Overall Loss 0.197486 Objective Loss 0.197486 LR 0.000250 Time 0.023397 +2023-10-05 21:56:37,611 - Epoch: [158][ 330/ 1236] Overall Loss 0.196989 Objective Loss 0.196989 LR 0.000250 Time 0.023316 +2023-10-05 21:56:37,822 - Epoch: [158][ 340/ 1236] Overall Loss 0.197299 Objective Loss 0.197299 LR 0.000250 Time 0.023249 +2023-10-05 21:56:38,031 - Epoch: [158][ 350/ 1236] Overall Loss 0.197123 Objective Loss 0.197123 LR 0.000250 Time 0.023177 +2023-10-05 21:56:38,235 - Epoch: [158][ 360/ 1236] Overall Loss 0.198020 Objective Loss 0.198020 LR 0.000250 Time 0.023098 +2023-10-05 21:56:38,437 - Epoch: [158][ 370/ 1236] Overall Loss 0.198171 Objective Loss 0.198171 LR 0.000250 Time 0.023019 +2023-10-05 21:56:38,641 - Epoch: [158][ 380/ 1236] Overall Loss 0.199012 Objective Loss 0.199012 LR 0.000250 Time 0.022948 +2023-10-05 21:56:38,845 - Epoch: [158][ 390/ 1236] Overall Loss 0.198448 Objective Loss 0.198448 LR 0.000250 Time 0.022883 +2023-10-05 21:56:39,048 - Epoch: [158][ 400/ 1236] Overall Loss 0.197954 Objective Loss 0.197954 LR 0.000250 Time 0.022817 +2023-10-05 21:56:39,250 - Epoch: [158][ 410/ 1236] Overall Loss 0.197954 Objective Loss 0.197954 LR 0.000250 Time 0.022753 +2023-10-05 21:56:39,453 - Epoch: [158][ 420/ 1236] Overall Loss 0.198731 Objective Loss 0.198731 LR 0.000250 Time 0.022694 +2023-10-05 21:56:39,656 - Epoch: [158][ 430/ 1236] Overall Loss 0.198707 Objective Loss 0.198707 LR 0.000250 Time 0.022635 +2023-10-05 21:56:39,858 - Epoch: [158][ 440/ 1236] Overall Loss 0.198894 Objective Loss 0.198894 LR 0.000250 Time 0.022580 +2023-10-05 21:56:40,060 - Epoch: [158][ 450/ 1236] Overall Loss 0.198971 Objective Loss 0.198971 LR 0.000250 Time 0.022527 +2023-10-05 21:56:40,263 - Epoch: [158][ 460/ 1236] Overall Loss 0.199217 Objective Loss 0.199217 LR 0.000250 Time 0.022477 +2023-10-05 21:56:40,465 - Epoch: [158][ 470/ 1236] Overall Loss 0.198788 Objective Loss 0.198788 LR 0.000250 Time 0.022429 +2023-10-05 21:56:40,668 - Epoch: [158][ 480/ 1236] Overall Loss 0.198533 Objective Loss 0.198533 LR 0.000250 Time 0.022383 +2023-10-05 21:56:40,870 - Epoch: [158][ 490/ 1236] Overall Loss 0.198604 Objective Loss 0.198604 LR 0.000250 Time 0.022338 +2023-10-05 21:56:41,073 - Epoch: [158][ 500/ 1236] Overall Loss 0.198682 Objective Loss 0.198682 LR 0.000250 Time 0.022296 +2023-10-05 21:56:41,275 - Epoch: [158][ 510/ 1236] Overall Loss 0.198658 Objective Loss 0.198658 LR 0.000250 Time 0.022255 +2023-10-05 21:56:41,486 - Epoch: [158][ 520/ 1236] Overall Loss 0.198814 Objective Loss 0.198814 LR 0.000250 Time 0.022231 +2023-10-05 21:56:41,693 - Epoch: [158][ 530/ 1236] Overall Loss 0.198619 Objective Loss 0.198619 LR 0.000250 Time 0.022201 +2023-10-05 21:56:41,896 - Epoch: [158][ 540/ 1236] Overall Loss 0.198373 Objective Loss 0.198373 LR 0.000250 Time 0.022165 +2023-10-05 21:56:42,096 - Epoch: [158][ 550/ 1236] Overall Loss 0.198421 Objective Loss 0.198421 LR 0.000250 Time 0.022126 +2023-10-05 21:56:42,298 - Epoch: [158][ 560/ 1236] Overall Loss 0.198394 Objective Loss 0.198394 LR 0.000250 Time 0.022092 +2023-10-05 21:56:42,499 - Epoch: [158][ 570/ 1236] Overall Loss 0.198161 Objective Loss 0.198161 LR 0.000250 Time 0.022055 +2023-10-05 21:56:42,701 - Epoch: [158][ 580/ 1236] Overall Loss 0.198094 Objective Loss 0.198094 LR 0.000250 Time 0.022023 +2023-10-05 21:56:42,902 - Epoch: [158][ 590/ 1236] Overall Loss 0.197771 Objective Loss 0.197771 LR 0.000250 Time 0.021989 +2023-10-05 21:56:43,104 - Epoch: [158][ 600/ 1236] Overall Loss 0.197771 Objective Loss 0.197771 LR 0.000250 Time 0.021959 +2023-10-05 21:56:43,304 - Epoch: [158][ 610/ 1236] Overall Loss 0.198168 Objective Loss 0.198168 LR 0.000250 Time 0.021927 +2023-10-05 21:56:43,506 - Epoch: [158][ 620/ 1236] Overall Loss 0.198321 Objective Loss 0.198321 LR 0.000250 Time 0.021899 +2023-10-05 21:56:43,707 - Epoch: [158][ 630/ 1236] Overall Loss 0.198509 Objective Loss 0.198509 LR 0.000250 Time 0.021869 +2023-10-05 21:56:43,909 - Epoch: [158][ 640/ 1236] Overall Loss 0.198496 Objective Loss 0.198496 LR 0.000250 Time 0.021843 +2023-10-05 21:56:44,110 - Epoch: [158][ 650/ 1236] Overall Loss 0.198528 Objective Loss 0.198528 LR 0.000250 Time 0.021815 +2023-10-05 21:56:44,312 - Epoch: [158][ 660/ 1236] Overall Loss 0.199044 Objective Loss 0.199044 LR 0.000250 Time 0.021790 +2023-10-05 21:56:44,512 - Epoch: [158][ 670/ 1236] Overall Loss 0.198934 Objective Loss 0.198934 LR 0.000250 Time 0.021763 +2023-10-05 21:56:44,714 - Epoch: [158][ 680/ 1236] Overall Loss 0.199166 Objective Loss 0.199166 LR 0.000250 Time 0.021740 +2023-10-05 21:56:44,915 - Epoch: [158][ 690/ 1236] Overall Loss 0.199487 Objective Loss 0.199487 LR 0.000250 Time 0.021715 +2023-10-05 21:56:45,117 - Epoch: [158][ 700/ 1236] Overall Loss 0.199116 Objective Loss 0.199116 LR 0.000250 Time 0.021694 +2023-10-05 21:56:45,318 - Epoch: [158][ 710/ 1236] Overall Loss 0.199274 Objective Loss 0.199274 LR 0.000250 Time 0.021670 +2023-10-05 21:56:45,520 - Epoch: [158][ 720/ 1236] Overall Loss 0.199347 Objective Loss 0.199347 LR 0.000250 Time 0.021649 +2023-10-05 21:56:45,720 - Epoch: [158][ 730/ 1236] Overall Loss 0.199454 Objective Loss 0.199454 LR 0.000250 Time 0.021627 +2023-10-05 21:56:45,922 - Epoch: [158][ 740/ 1236] Overall Loss 0.199483 Objective Loss 0.199483 LR 0.000250 Time 0.021607 +2023-10-05 21:56:46,123 - Epoch: [158][ 750/ 1236] Overall Loss 0.199453 Objective Loss 0.199453 LR 0.000250 Time 0.021586 +2023-10-05 21:56:46,325 - Epoch: [158][ 760/ 1236] Overall Loss 0.199805 Objective Loss 0.199805 LR 0.000250 Time 0.021568 +2023-10-05 21:56:46,525 - Epoch: [158][ 770/ 1236] Overall Loss 0.199763 Objective Loss 0.199763 LR 0.000250 Time 0.021547 +2023-10-05 21:56:46,727 - Epoch: [158][ 780/ 1236] Overall Loss 0.199767 Objective Loss 0.199767 LR 0.000250 Time 0.021530 +2023-10-05 21:56:46,928 - Epoch: [158][ 790/ 1236] Overall Loss 0.200213 Objective Loss 0.200213 LR 0.000250 Time 0.021511 +2023-10-05 21:56:47,130 - Epoch: [158][ 800/ 1236] Overall Loss 0.200179 Objective Loss 0.200179 LR 0.000250 Time 0.021494 +2023-10-05 21:56:47,331 - Epoch: [158][ 810/ 1236] Overall Loss 0.200040 Objective Loss 0.200040 LR 0.000250 Time 0.021476 +2023-10-05 21:56:47,533 - Epoch: [158][ 820/ 1236] Overall Loss 0.199973 Objective Loss 0.199973 LR 0.000250 Time 0.021460 +2023-10-05 21:56:47,733 - Epoch: [158][ 830/ 1236] Overall Loss 0.199969 Objective Loss 0.199969 LR 0.000250 Time 0.021443 +2023-10-05 21:56:47,935 - Epoch: [158][ 840/ 1236] Overall Loss 0.200171 Objective Loss 0.200171 LR 0.000250 Time 0.021428 +2023-10-05 21:56:48,136 - Epoch: [158][ 850/ 1236] Overall Loss 0.199982 Objective Loss 0.199982 LR 0.000250 Time 0.021411 +2023-10-05 21:56:48,338 - Epoch: [158][ 860/ 1236] Overall Loss 0.200159 Objective Loss 0.200159 LR 0.000250 Time 0.021397 +2023-10-05 21:56:48,538 - Epoch: [158][ 870/ 1236] Overall Loss 0.200205 Objective Loss 0.200205 LR 0.000250 Time 0.021381 +2023-10-05 21:56:48,741 - Epoch: [158][ 880/ 1236] Overall Loss 0.199913 Objective Loss 0.199913 LR 0.000250 Time 0.021367 +2023-10-05 21:56:48,941 - Epoch: [158][ 890/ 1236] Overall Loss 0.200007 Objective Loss 0.200007 LR 0.000250 Time 0.021352 +2023-10-05 21:56:49,143 - Epoch: [158][ 900/ 1236] Overall Loss 0.200229 Objective Loss 0.200229 LR 0.000250 Time 0.021339 +2023-10-05 21:56:49,344 - Epoch: [158][ 910/ 1236] Overall Loss 0.200176 Objective Loss 0.200176 LR 0.000250 Time 0.021324 +2023-10-05 21:56:49,546 - Epoch: [158][ 920/ 1236] Overall Loss 0.200262 Objective Loss 0.200262 LR 0.000250 Time 0.021312 +2023-10-05 21:56:49,750 - Epoch: [158][ 930/ 1236] Overall Loss 0.200211 Objective Loss 0.200211 LR 0.000250 Time 0.021302 +2023-10-05 21:56:49,953 - Epoch: [158][ 940/ 1236] Overall Loss 0.199845 Objective Loss 0.199845 LR 0.000250 Time 0.021291 +2023-10-05 21:56:50,153 - Epoch: [158][ 950/ 1236] Overall Loss 0.199986 Objective Loss 0.199986 LR 0.000250 Time 0.021277 +2023-10-05 21:56:50,357 - Epoch: [158][ 960/ 1236] Overall Loss 0.199981 Objective Loss 0.199981 LR 0.000250 Time 0.021268 +2023-10-05 21:56:50,558 - Epoch: [158][ 970/ 1236] Overall Loss 0.199889 Objective Loss 0.199889 LR 0.000250 Time 0.021255 +2023-10-05 21:56:50,760 - Epoch: [158][ 980/ 1236] Overall Loss 0.199738 Objective Loss 0.199738 LR 0.000250 Time 0.021244 +2023-10-05 21:56:50,961 - Epoch: [158][ 990/ 1236] Overall Loss 0.199700 Objective Loss 0.199700 LR 0.000250 Time 0.021232 +2023-10-05 21:56:51,163 - Epoch: [158][ 1000/ 1236] Overall Loss 0.199660 Objective Loss 0.199660 LR 0.000250 Time 0.021222 +2023-10-05 21:56:51,364 - Epoch: [158][ 1010/ 1236] Overall Loss 0.199478 Objective Loss 0.199478 LR 0.000250 Time 0.021210 +2023-10-05 21:56:51,567 - Epoch: [158][ 1020/ 1236] Overall Loss 0.199537 Objective Loss 0.199537 LR 0.000250 Time 0.021200 +2023-10-05 21:56:51,767 - Epoch: [158][ 1030/ 1236] Overall Loss 0.199567 Objective Loss 0.199567 LR 0.000250 Time 0.021189 +2023-10-05 21:56:51,970 - Epoch: [158][ 1040/ 1236] Overall Loss 0.199671 Objective Loss 0.199671 LR 0.000250 Time 0.021180 +2023-10-05 21:56:52,170 - Epoch: [158][ 1050/ 1236] Overall Loss 0.199876 Objective Loss 0.199876 LR 0.000250 Time 0.021169 +2023-10-05 21:56:52,373 - Epoch: [158][ 1060/ 1236] Overall Loss 0.199949 Objective Loss 0.199949 LR 0.000250 Time 0.021160 +2023-10-05 21:56:52,573 - Epoch: [158][ 1070/ 1236] Overall Loss 0.200158 Objective Loss 0.200158 LR 0.000250 Time 0.021149 +2023-10-05 21:56:52,776 - Epoch: [158][ 1080/ 1236] Overall Loss 0.200289 Objective Loss 0.200289 LR 0.000250 Time 0.021140 +2023-10-05 21:56:52,976 - Epoch: [158][ 1090/ 1236] Overall Loss 0.200238 Objective Loss 0.200238 LR 0.000250 Time 0.021130 +2023-10-05 21:56:53,179 - Epoch: [158][ 1100/ 1236] Overall Loss 0.200210 Objective Loss 0.200210 LR 0.000250 Time 0.021121 +2023-10-05 21:56:53,379 - Epoch: [158][ 1110/ 1236] Overall Loss 0.200101 Objective Loss 0.200101 LR 0.000250 Time 0.021111 +2023-10-05 21:56:53,581 - Epoch: [158][ 1120/ 1236] Overall Loss 0.199940 Objective Loss 0.199940 LR 0.000250 Time 0.021103 +2023-10-05 21:56:53,782 - Epoch: [158][ 1130/ 1236] Overall Loss 0.199850 Objective Loss 0.199850 LR 0.000250 Time 0.021094 +2023-10-05 21:56:53,984 - Epoch: [158][ 1140/ 1236] Overall Loss 0.199942 Objective Loss 0.199942 LR 0.000250 Time 0.021086 +2023-10-05 21:56:54,185 - Epoch: [158][ 1150/ 1236] Overall Loss 0.199997 Objective Loss 0.199997 LR 0.000250 Time 0.021077 +2023-10-05 21:56:54,387 - Epoch: [158][ 1160/ 1236] Overall Loss 0.199988 Objective Loss 0.199988 LR 0.000250 Time 0.021069 +2023-10-05 21:56:54,588 - Epoch: [158][ 1170/ 1236] Overall Loss 0.200063 Objective Loss 0.200063 LR 0.000250 Time 0.021060 +2023-10-05 21:56:54,790 - Epoch: [158][ 1180/ 1236] Overall Loss 0.200093 Objective Loss 0.200093 LR 0.000250 Time 0.021053 +2023-10-05 21:56:54,991 - Epoch: [158][ 1190/ 1236] Overall Loss 0.200217 Objective Loss 0.200217 LR 0.000250 Time 0.021045 +2023-10-05 21:56:55,193 - Epoch: [158][ 1200/ 1236] Overall Loss 0.200173 Objective Loss 0.200173 LR 0.000250 Time 0.021038 +2023-10-05 21:56:55,394 - Epoch: [158][ 1210/ 1236] Overall Loss 0.200174 Objective Loss 0.200174 LR 0.000250 Time 0.021029 +2023-10-05 21:56:55,596 - Epoch: [158][ 1220/ 1236] Overall Loss 0.200192 Objective Loss 0.200192 LR 0.000250 Time 0.021023 +2023-10-05 21:56:55,849 - Epoch: [158][ 1230/ 1236] Overall Loss 0.200252 Objective Loss 0.200252 LR 0.000250 Time 0.021057 +2023-10-05 21:56:55,968 - Epoch: [158][ 1236/ 1236] Overall Loss 0.200102 Objective Loss 0.200102 Top1 89.409369 Top5 99.185336 LR 0.000250 Time 0.021050 +2023-10-05 21:56:56,089 - --- validate (epoch=158)----------- +2023-10-05 21:56:56,089 - 29943 samples (256 per mini-batch) +2023-10-05 21:56:56,557 - Epoch: [158][ 10/ 117] Loss 0.300585 Top1 84.296875 Top5 97.929688 +2023-10-05 21:56:56,707 - Epoch: [158][ 20/ 117] Loss 0.323280 Top1 84.121094 Top5 98.046875 +2023-10-05 21:56:56,854 - Epoch: [158][ 30/ 117] Loss 0.334161 Top1 84.101562 Top5 97.929688 +2023-10-05 21:56:57,003 - Epoch: [158][ 40/ 117] Loss 0.327019 Top1 84.130859 Top5 97.998047 +2023-10-05 21:56:57,151 - Epoch: [158][ 50/ 117] Loss 0.315538 Top1 84.468750 Top5 98.109375 +2023-10-05 21:56:57,302 - Epoch: [158][ 60/ 117] Loss 0.316988 Top1 84.505208 Top5 98.085938 +2023-10-05 21:56:57,450 - Epoch: [158][ 70/ 117] Loss 0.312788 Top1 84.570312 Top5 98.085938 +2023-10-05 21:56:57,601 - Epoch: [158][ 80/ 117] Loss 0.308572 Top1 84.746094 Top5 98.100586 +2023-10-05 21:56:57,751 - Epoch: [158][ 90/ 117] Loss 0.310056 Top1 84.804688 Top5 98.098958 +2023-10-05 21:56:57,903 - Epoch: [158][ 100/ 117] Loss 0.308238 Top1 84.914062 Top5 98.121094 +2023-10-05 21:56:58,061 - Epoch: [158][ 110/ 117] Loss 0.309297 Top1 84.982244 Top5 98.103693 +2023-10-05 21:56:58,147 - Epoch: [158][ 117/ 117] Loss 0.310862 Top1 84.914671 Top5 98.093043 +2023-10-05 21:56:58,280 - ==> Top1: 84.915 Top5: 98.093 Loss: 0.311 + +2023-10-05 21:56:58,281 - ==> Confusion: +[[ 924 2 3 0 6 3 0 0 7 71 2 0 2 1 7 1 2 2 0 0 17] + [ 1 1063 2 0 6 19 1 14 0 0 1 1 0 1 1 3 3 0 7 0 8] + [ 4 1 970 14 3 0 19 6 0 1 3 1 7 1 2 3 1 1 11 4 4] + [ 2 1 14 962 1 4 1 0 0 2 5 1 7 1 27 2 1 8 33 2 15] + [ 21 8 2 0 970 5 0 1 0 7 0 2 1 2 11 4 8 2 0 1 5] + [ 4 38 0 1 7 981 2 18 0 2 5 7 1 16 6 2 3 1 3 5 14] + [ 0 7 24 0 0 0 1125 7 0 0 1 3 0 0 1 6 0 0 3 7 7] + [ 5 24 16 0 4 28 2 1048 0 3 4 10 0 3 1 1 0 0 51 10 8] + [ 16 2 0 2 2 2 0 0 966 42 10 1 3 15 17 3 0 0 6 0 2] + [ 90 0 3 1 3 3 0 0 20 953 1 2 2 16 9 5 0 1 1 0 9] + [ 3 6 9 5 1 1 6 3 10 2 971 3 0 9 4 1 2 0 5 2 10] + [ 0 0 1 0 1 15 0 1 0 1 1 960 19 7 0 3 0 15 0 7 4] + [ 1 2 1 5 0 2 0 1 0 0 1 36 985 1 1 3 2 13 0 5 9] + [ 1 0 1 0 1 6 0 0 6 11 6 4 3 1059 5 2 0 0 0 3 11] + [ 11 1 4 8 3 0 0 0 22 1 3 1 1 3 1020 0 0 1 11 0 11] + [ 0 3 2 0 3 0 0 1 0 0 0 8 7 3 1 1073 9 9 0 12 3] + [ 1 14 1 0 7 4 0 0 1 0 0 4 0 1 3 12 1098 0 0 4 11] + [ 0 0 1 2 1 0 3 0 1 0 0 3 18 1 0 5 0 999 1 0 3] + [ 1 8 6 16 1 0 0 20 1 0 1 0 1 2 9 0 1 0 994 0 7] + [ 0 1 5 3 2 5 9 7 1 0 1 13 3 2 0 6 8 0 2 1076 8] + [ 103 191 149 55 99 106 41 74 89 84 182 117 341 282 172 55 107 77 175 177 5229]] + +2023-10-05 21:56:58,283 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:56:58,283 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:56:58,289 - + +2023-10-05 21:56:58,289 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:56:59,309 - Epoch: [159][ 10/ 1236] Overall Loss 0.179544 Objective Loss 0.179544 LR 0.000250 Time 0.102001 +2023-10-05 21:56:59,511 - Epoch: [159][ 20/ 1236] Overall Loss 0.169422 Objective Loss 0.169422 LR 0.000250 Time 0.061045 +2023-10-05 21:56:59,711 - Epoch: [159][ 30/ 1236] Overall Loss 0.178727 Objective Loss 0.178727 LR 0.000250 Time 0.047343 +2023-10-05 21:56:59,912 - Epoch: [159][ 40/ 1236] Overall Loss 0.185486 Objective Loss 0.185486 LR 0.000250 Time 0.040537 +2023-10-05 21:57:00,111 - Epoch: [159][ 50/ 1236] Overall Loss 0.188473 Objective Loss 0.188473 LR 0.000250 Time 0.036408 +2023-10-05 21:57:00,312 - Epoch: [159][ 60/ 1236] Overall Loss 0.187700 Objective Loss 0.187700 LR 0.000250 Time 0.033680 +2023-10-05 21:57:00,511 - Epoch: [159][ 70/ 1236] Overall Loss 0.188516 Objective Loss 0.188516 LR 0.000250 Time 0.031703 +2023-10-05 21:57:00,713 - Epoch: [159][ 80/ 1236] Overall Loss 0.190965 Objective Loss 0.190965 LR 0.000250 Time 0.030259 +2023-10-05 21:57:00,912 - Epoch: [159][ 90/ 1236] Overall Loss 0.192810 Objective Loss 0.192810 LR 0.000250 Time 0.029110 +2023-10-05 21:57:01,114 - Epoch: [159][ 100/ 1236] Overall Loss 0.193557 Objective Loss 0.193557 LR 0.000250 Time 0.028215 +2023-10-05 21:57:01,314 - Epoch: [159][ 110/ 1236] Overall Loss 0.196564 Objective Loss 0.196564 LR 0.000250 Time 0.027466 +2023-10-05 21:57:01,516 - Epoch: [159][ 120/ 1236] Overall Loss 0.195734 Objective Loss 0.195734 LR 0.000250 Time 0.026853 +2023-10-05 21:57:01,716 - Epoch: [159][ 130/ 1236] Overall Loss 0.194590 Objective Loss 0.194590 LR 0.000250 Time 0.026325 +2023-10-05 21:57:01,918 - Epoch: [159][ 140/ 1236] Overall Loss 0.194269 Objective Loss 0.194269 LR 0.000250 Time 0.025885 +2023-10-05 21:57:02,120 - Epoch: [159][ 150/ 1236] Overall Loss 0.194234 Objective Loss 0.194234 LR 0.000250 Time 0.025507 +2023-10-05 21:57:02,321 - Epoch: [159][ 160/ 1236] Overall Loss 0.196746 Objective Loss 0.196746 LR 0.000250 Time 0.025162 +2023-10-05 21:57:02,523 - Epoch: [159][ 170/ 1236] Overall Loss 0.195988 Objective Loss 0.195988 LR 0.000250 Time 0.024872 +2023-10-05 21:57:02,727 - Epoch: [159][ 180/ 1236] Overall Loss 0.198052 Objective Loss 0.198052 LR 0.000250 Time 0.024618 +2023-10-05 21:57:02,928 - Epoch: [159][ 190/ 1236] Overall Loss 0.197148 Objective Loss 0.197148 LR 0.000250 Time 0.024382 +2023-10-05 21:57:03,132 - Epoch: [159][ 200/ 1236] Overall Loss 0.196679 Objective Loss 0.196679 LR 0.000250 Time 0.024180 +2023-10-05 21:57:03,333 - Epoch: [159][ 210/ 1236] Overall Loss 0.197010 Objective Loss 0.197010 LR 0.000250 Time 0.023986 +2023-10-05 21:57:03,537 - Epoch: [159][ 220/ 1236] Overall Loss 0.196897 Objective Loss 0.196897 LR 0.000250 Time 0.023818 +2023-10-05 21:57:03,738 - Epoch: [159][ 230/ 1236] Overall Loss 0.196712 Objective Loss 0.196712 LR 0.000250 Time 0.023658 +2023-10-05 21:57:03,942 - Epoch: [159][ 240/ 1236] Overall Loss 0.197311 Objective Loss 0.197311 LR 0.000250 Time 0.023517 +2023-10-05 21:57:04,144 - Epoch: [159][ 250/ 1236] Overall Loss 0.197228 Objective Loss 0.197228 LR 0.000250 Time 0.023384 +2023-10-05 21:57:04,347 - Epoch: [159][ 260/ 1236] Overall Loss 0.197765 Objective Loss 0.197765 LR 0.000250 Time 0.023266 +2023-10-05 21:57:04,550 - Epoch: [159][ 270/ 1236] Overall Loss 0.198858 Objective Loss 0.198858 LR 0.000250 Time 0.023154 +2023-10-05 21:57:04,750 - Epoch: [159][ 280/ 1236] Overall Loss 0.199749 Objective Loss 0.199749 LR 0.000250 Time 0.023042 +2023-10-05 21:57:04,952 - Epoch: [159][ 290/ 1236] Overall Loss 0.200006 Objective Loss 0.200006 LR 0.000250 Time 0.022942 +2023-10-05 21:57:05,153 - Epoch: [159][ 300/ 1236] Overall Loss 0.200580 Objective Loss 0.200580 LR 0.000250 Time 0.022844 +2023-10-05 21:57:05,353 - Epoch: [159][ 310/ 1236] Overall Loss 0.200730 Objective Loss 0.200730 LR 0.000250 Time 0.022752 +2023-10-05 21:57:05,553 - Epoch: [159][ 320/ 1236] Overall Loss 0.200551 Objective Loss 0.200551 LR 0.000250 Time 0.022664 +2023-10-05 21:57:05,755 - Epoch: [159][ 330/ 1236] Overall Loss 0.200762 Objective Loss 0.200762 LR 0.000250 Time 0.022588 +2023-10-05 21:57:05,955 - Epoch: [159][ 340/ 1236] Overall Loss 0.200702 Objective Loss 0.200702 LR 0.000250 Time 0.022513 +2023-10-05 21:57:06,155 - Epoch: [159][ 350/ 1236] Overall Loss 0.200501 Objective Loss 0.200501 LR 0.000250 Time 0.022439 +2023-10-05 21:57:06,352 - Epoch: [159][ 360/ 1236] Overall Loss 0.200616 Objective Loss 0.200616 LR 0.000250 Time 0.022363 +2023-10-05 21:57:06,552 - Epoch: [159][ 370/ 1236] Overall Loss 0.200207 Objective Loss 0.200207 LR 0.000250 Time 0.022299 +2023-10-05 21:57:06,749 - Epoch: [159][ 380/ 1236] Overall Loss 0.200008 Objective Loss 0.200008 LR 0.000250 Time 0.022229 +2023-10-05 21:57:06,949 - Epoch: [159][ 390/ 1236] Overall Loss 0.200260 Objective Loss 0.200260 LR 0.000250 Time 0.022170 +2023-10-05 21:57:07,146 - Epoch: [159][ 400/ 1236] Overall Loss 0.200028 Objective Loss 0.200028 LR 0.000250 Time 0.022108 +2023-10-05 21:57:07,345 - Epoch: [159][ 410/ 1236] Overall Loss 0.199615 Objective Loss 0.199615 LR 0.000250 Time 0.022054 +2023-10-05 21:57:07,543 - Epoch: [159][ 420/ 1236] Overall Loss 0.199823 Objective Loss 0.199823 LR 0.000250 Time 0.021998 +2023-10-05 21:57:07,742 - Epoch: [159][ 430/ 1236] Overall Loss 0.199717 Objective Loss 0.199717 LR 0.000250 Time 0.021950 +2023-10-05 21:57:07,940 - Epoch: [159][ 440/ 1236] Overall Loss 0.199608 Objective Loss 0.199608 LR 0.000250 Time 0.021898 +2023-10-05 21:57:08,139 - Epoch: [159][ 450/ 1236] Overall Loss 0.199471 Objective Loss 0.199471 LR 0.000250 Time 0.021853 +2023-10-05 21:57:08,336 - Epoch: [159][ 460/ 1236] Overall Loss 0.199640 Objective Loss 0.199640 LR 0.000250 Time 0.021807 +2023-10-05 21:57:08,535 - Epoch: [159][ 470/ 1236] Overall Loss 0.199931 Objective Loss 0.199931 LR 0.000250 Time 0.021765 +2023-10-05 21:57:08,733 - Epoch: [159][ 480/ 1236] Overall Loss 0.199968 Objective Loss 0.199968 LR 0.000250 Time 0.021723 +2023-10-05 21:57:08,932 - Epoch: [159][ 490/ 1236] Overall Loss 0.199828 Objective Loss 0.199828 LR 0.000250 Time 0.021685 +2023-10-05 21:57:09,130 - Epoch: [159][ 500/ 1236] Overall Loss 0.199734 Objective Loss 0.199734 LR 0.000250 Time 0.021646 +2023-10-05 21:57:09,329 - Epoch: [159][ 510/ 1236] Overall Loss 0.199662 Objective Loss 0.199662 LR 0.000250 Time 0.021611 +2023-10-05 21:57:09,526 - Epoch: [159][ 520/ 1236] Overall Loss 0.199910 Objective Loss 0.199910 LR 0.000250 Time 0.021575 +2023-10-05 21:57:09,725 - Epoch: [159][ 530/ 1236] Overall Loss 0.200146 Objective Loss 0.200146 LR 0.000250 Time 0.021543 +2023-10-05 21:57:09,923 - Epoch: [159][ 540/ 1236] Overall Loss 0.199954 Objective Loss 0.199954 LR 0.000250 Time 0.021509 +2023-10-05 21:57:10,122 - Epoch: [159][ 550/ 1236] Overall Loss 0.200308 Objective Loss 0.200308 LR 0.000250 Time 0.021479 +2023-10-05 21:57:10,319 - Epoch: [159][ 560/ 1236] Overall Loss 0.200160 Objective Loss 0.200160 LR 0.000250 Time 0.021448 +2023-10-05 21:57:10,518 - Epoch: [159][ 570/ 1236] Overall Loss 0.199799 Objective Loss 0.199799 LR 0.000250 Time 0.021420 +2023-10-05 21:57:10,716 - Epoch: [159][ 580/ 1236] Overall Loss 0.199964 Objective Loss 0.199964 LR 0.000250 Time 0.021391 +2023-10-05 21:57:10,915 - Epoch: [159][ 590/ 1236] Overall Loss 0.199592 Objective Loss 0.199592 LR 0.000250 Time 0.021365 +2023-10-05 21:57:11,113 - Epoch: [159][ 600/ 1236] Overall Loss 0.199816 Objective Loss 0.199816 LR 0.000250 Time 0.021338 +2023-10-05 21:57:11,312 - Epoch: [159][ 610/ 1236] Overall Loss 0.199737 Objective Loss 0.199737 LR 0.000250 Time 0.021314 +2023-10-05 21:57:11,509 - Epoch: [159][ 620/ 1236] Overall Loss 0.199590 Objective Loss 0.199590 LR 0.000250 Time 0.021288 +2023-10-05 21:57:11,708 - Epoch: [159][ 630/ 1236] Overall Loss 0.199681 Objective Loss 0.199681 LR 0.000250 Time 0.021265 +2023-10-05 21:57:11,906 - Epoch: [159][ 640/ 1236] Overall Loss 0.199652 Objective Loss 0.199652 LR 0.000250 Time 0.021241 +2023-10-05 21:57:12,105 - Epoch: [159][ 650/ 1236] Overall Loss 0.199434 Objective Loss 0.199434 LR 0.000250 Time 0.021220 +2023-10-05 21:57:12,302 - Epoch: [159][ 660/ 1236] Overall Loss 0.199453 Objective Loss 0.199453 LR 0.000250 Time 0.021197 +2023-10-05 21:57:12,501 - Epoch: [159][ 670/ 1236] Overall Loss 0.199228 Objective Loss 0.199228 LR 0.000250 Time 0.021177 +2023-10-05 21:57:12,699 - Epoch: [159][ 680/ 1236] Overall Loss 0.199128 Objective Loss 0.199128 LR 0.000250 Time 0.021156 +2023-10-05 21:57:12,898 - Epoch: [159][ 690/ 1236] Overall Loss 0.199015 Objective Loss 0.199015 LR 0.000250 Time 0.021137 +2023-10-05 21:57:13,095 - Epoch: [159][ 700/ 1236] Overall Loss 0.199085 Objective Loss 0.199085 LR 0.000250 Time 0.021117 +2023-10-05 21:57:13,294 - Epoch: [159][ 710/ 1236] Overall Loss 0.199431 Objective Loss 0.199431 LR 0.000250 Time 0.021100 +2023-10-05 21:57:13,492 - Epoch: [159][ 720/ 1236] Overall Loss 0.199466 Objective Loss 0.199466 LR 0.000250 Time 0.021081 +2023-10-05 21:57:13,691 - Epoch: [159][ 730/ 1236] Overall Loss 0.199536 Objective Loss 0.199536 LR 0.000250 Time 0.021064 +2023-10-05 21:57:13,889 - Epoch: [159][ 740/ 1236] Overall Loss 0.199808 Objective Loss 0.199808 LR 0.000250 Time 0.021046 +2023-10-05 21:57:14,088 - Epoch: [159][ 750/ 1236] Overall Loss 0.199378 Objective Loss 0.199378 LR 0.000250 Time 0.021030 +2023-10-05 21:57:14,285 - Epoch: [159][ 760/ 1236] Overall Loss 0.199431 Objective Loss 0.199431 LR 0.000250 Time 0.021013 +2023-10-05 21:57:14,484 - Epoch: [159][ 770/ 1236] Overall Loss 0.199164 Objective Loss 0.199164 LR 0.000250 Time 0.020998 +2023-10-05 21:57:14,682 - Epoch: [159][ 780/ 1236] Overall Loss 0.199343 Objective Loss 0.199343 LR 0.000250 Time 0.020982 +2023-10-05 21:57:14,881 - Epoch: [159][ 790/ 1236] Overall Loss 0.199553 Objective Loss 0.199553 LR 0.000250 Time 0.020968 +2023-10-05 21:57:15,079 - Epoch: [159][ 800/ 1236] Overall Loss 0.199699 Objective Loss 0.199699 LR 0.000250 Time 0.020953 +2023-10-05 21:57:15,278 - Epoch: [159][ 810/ 1236] Overall Loss 0.199628 Objective Loss 0.199628 LR 0.000250 Time 0.020939 +2023-10-05 21:57:15,475 - Epoch: [159][ 820/ 1236] Overall Loss 0.199693 Objective Loss 0.199693 LR 0.000250 Time 0.020925 +2023-10-05 21:57:15,675 - Epoch: [159][ 830/ 1236] Overall Loss 0.199419 Objective Loss 0.199419 LR 0.000250 Time 0.020912 +2023-10-05 21:57:15,872 - Epoch: [159][ 840/ 1236] Overall Loss 0.199340 Objective Loss 0.199340 LR 0.000250 Time 0.020898 +2023-10-05 21:57:16,071 - Epoch: [159][ 850/ 1236] Overall Loss 0.199500 Objective Loss 0.199500 LR 0.000250 Time 0.020886 +2023-10-05 21:57:16,269 - Epoch: [159][ 860/ 1236] Overall Loss 0.199872 Objective Loss 0.199872 LR 0.000250 Time 0.020873 +2023-10-05 21:57:16,468 - Epoch: [159][ 870/ 1236] Overall Loss 0.199774 Objective Loss 0.199774 LR 0.000250 Time 0.020861 +2023-10-05 21:57:16,666 - Epoch: [159][ 880/ 1236] Overall Loss 0.199609 Objective Loss 0.199609 LR 0.000250 Time 0.020848 +2023-10-05 21:57:16,865 - Epoch: [159][ 890/ 1236] Overall Loss 0.199517 Objective Loss 0.199517 LR 0.000250 Time 0.020837 +2023-10-05 21:57:17,062 - Epoch: [159][ 900/ 1236] Overall Loss 0.199783 Objective Loss 0.199783 LR 0.000250 Time 0.020825 +2023-10-05 21:57:17,261 - Epoch: [159][ 910/ 1236] Overall Loss 0.199494 Objective Loss 0.199494 LR 0.000250 Time 0.020814 +2023-10-05 21:57:17,459 - Epoch: [159][ 920/ 1236] Overall Loss 0.199542 Objective Loss 0.199542 LR 0.000250 Time 0.020802 +2023-10-05 21:57:17,658 - Epoch: [159][ 930/ 1236] Overall Loss 0.199477 Objective Loss 0.199477 LR 0.000250 Time 0.020792 +2023-10-05 21:57:17,856 - Epoch: [159][ 940/ 1236] Overall Loss 0.199570 Objective Loss 0.199570 LR 0.000250 Time 0.020781 +2023-10-05 21:57:18,055 - Epoch: [159][ 950/ 1236] Overall Loss 0.199522 Objective Loss 0.199522 LR 0.000250 Time 0.020772 +2023-10-05 21:57:18,253 - Epoch: [159][ 960/ 1236] Overall Loss 0.199603 Objective Loss 0.199603 LR 0.000250 Time 0.020761 +2023-10-05 21:57:18,452 - Epoch: [159][ 970/ 1236] Overall Loss 0.199442 Objective Loss 0.199442 LR 0.000250 Time 0.020751 +2023-10-05 21:57:18,650 - Epoch: [159][ 980/ 1236] Overall Loss 0.199461 Objective Loss 0.199461 LR 0.000250 Time 0.020741 +2023-10-05 21:57:18,849 - Epoch: [159][ 990/ 1236] Overall Loss 0.199326 Objective Loss 0.199326 LR 0.000250 Time 0.020732 +2023-10-05 21:57:19,046 - Epoch: [159][ 1000/ 1236] Overall Loss 0.199334 Objective Loss 0.199334 LR 0.000250 Time 0.020722 +2023-10-05 21:57:19,245 - Epoch: [159][ 1010/ 1236] Overall Loss 0.199303 Objective Loss 0.199303 LR 0.000250 Time 0.020714 +2023-10-05 21:57:19,443 - Epoch: [159][ 1020/ 1236] Overall Loss 0.199213 Objective Loss 0.199213 LR 0.000250 Time 0.020704 +2023-10-05 21:57:19,642 - Epoch: [159][ 1030/ 1236] Overall Loss 0.199284 Objective Loss 0.199284 LR 0.000250 Time 0.020696 +2023-10-05 21:57:19,839 - Epoch: [159][ 1040/ 1236] Overall Loss 0.199325 Objective Loss 0.199325 LR 0.000250 Time 0.020687 +2023-10-05 21:57:20,038 - Epoch: [159][ 1050/ 1236] Overall Loss 0.199357 Objective Loss 0.199357 LR 0.000250 Time 0.020679 +2023-10-05 21:57:20,236 - Epoch: [159][ 1060/ 1236] Overall Loss 0.199206 Objective Loss 0.199206 LR 0.000250 Time 0.020670 +2023-10-05 21:57:20,435 - Epoch: [159][ 1070/ 1236] Overall Loss 0.199236 Objective Loss 0.199236 LR 0.000250 Time 0.020663 +2023-10-05 21:57:20,633 - Epoch: [159][ 1080/ 1236] Overall Loss 0.199048 Objective Loss 0.199048 LR 0.000250 Time 0.020654 +2023-10-05 21:57:20,832 - Epoch: [159][ 1090/ 1236] Overall Loss 0.198785 Objective Loss 0.198785 LR 0.000250 Time 0.020647 +2023-10-05 21:57:21,029 - Epoch: [159][ 1100/ 1236] Overall Loss 0.198854 Objective Loss 0.198854 LR 0.000250 Time 0.020638 +2023-10-05 21:57:21,228 - Epoch: [159][ 1110/ 1236] Overall Loss 0.198741 Objective Loss 0.198741 LR 0.000250 Time 0.020631 +2023-10-05 21:57:21,426 - Epoch: [159][ 1120/ 1236] Overall Loss 0.198681 Objective Loss 0.198681 LR 0.000250 Time 0.020623 +2023-10-05 21:57:21,625 - Epoch: [159][ 1130/ 1236] Overall Loss 0.198698 Objective Loss 0.198698 LR 0.000250 Time 0.020616 +2023-10-05 21:57:21,823 - Epoch: [159][ 1140/ 1236] Overall Loss 0.198687 Objective Loss 0.198687 LR 0.000250 Time 0.020609 +2023-10-05 21:57:22,022 - Epoch: [159][ 1150/ 1236] Overall Loss 0.198744 Objective Loss 0.198744 LR 0.000250 Time 0.020602 +2023-10-05 21:57:22,219 - Epoch: [159][ 1160/ 1236] Overall Loss 0.198619 Objective Loss 0.198619 LR 0.000250 Time 0.020595 +2023-10-05 21:57:22,418 - Epoch: [159][ 1170/ 1236] Overall Loss 0.198544 Objective Loss 0.198544 LR 0.000250 Time 0.020589 +2023-10-05 21:57:22,616 - Epoch: [159][ 1180/ 1236] Overall Loss 0.198457 Objective Loss 0.198457 LR 0.000250 Time 0.020581 +2023-10-05 21:57:22,815 - Epoch: [159][ 1190/ 1236] Overall Loss 0.198495 Objective Loss 0.198495 LR 0.000250 Time 0.020575 +2023-10-05 21:57:23,013 - Epoch: [159][ 1200/ 1236] Overall Loss 0.198507 Objective Loss 0.198507 LR 0.000250 Time 0.020568 +2023-10-05 21:57:23,212 - Epoch: [159][ 1210/ 1236] Overall Loss 0.198594 Objective Loss 0.198594 LR 0.000250 Time 0.020563 +2023-10-05 21:57:23,410 - Epoch: [159][ 1220/ 1236] Overall Loss 0.198680 Objective Loss 0.198680 LR 0.000250 Time 0.020556 +2023-10-05 21:57:23,657 - Epoch: [159][ 1230/ 1236] Overall Loss 0.198638 Objective Loss 0.198638 LR 0.000250 Time 0.020590 +2023-10-05 21:57:23,774 - Epoch: [159][ 1236/ 1236] Overall Loss 0.198749 Objective Loss 0.198749 Top1 87.780041 Top5 97.352342 LR 0.000250 Time 0.020584 +2023-10-05 21:57:23,907 - --- validate (epoch=159)----------- +2023-10-05 21:57:23,907 - 29943 samples (256 per mini-batch) +2023-10-05 21:57:24,360 - Epoch: [159][ 10/ 117] Loss 0.291308 Top1 85.507812 Top5 98.085938 +2023-10-05 21:57:24,511 - Epoch: [159][ 20/ 117] Loss 0.306237 Top1 85.253906 Top5 98.125000 +2023-10-05 21:57:24,662 - Epoch: [159][ 30/ 117] Loss 0.324151 Top1 84.700521 Top5 98.072917 +2023-10-05 21:57:24,813 - Epoch: [159][ 40/ 117] Loss 0.322114 Top1 84.716797 Top5 98.046875 +2023-10-05 21:57:24,966 - Epoch: [159][ 50/ 117] Loss 0.318183 Top1 84.882812 Top5 97.968750 +2023-10-05 21:57:25,116 - Epoch: [159][ 60/ 117] Loss 0.314777 Top1 84.947917 Top5 97.910156 +2023-10-05 21:57:25,266 - Epoch: [159][ 70/ 117] Loss 0.314527 Top1 84.960938 Top5 97.907366 +2023-10-05 21:57:25,417 - Epoch: [159][ 80/ 117] Loss 0.309776 Top1 85.087891 Top5 97.988281 +2023-10-05 21:57:25,566 - Epoch: [159][ 90/ 117] Loss 0.306338 Top1 85.095486 Top5 98.042535 +2023-10-05 21:57:25,713 - Epoch: [159][ 100/ 117] Loss 0.306902 Top1 85.062500 Top5 98.070312 +2023-10-05 21:57:25,874 - Epoch: [159][ 110/ 117] Loss 0.306800 Top1 85.056818 Top5 98.025568 +2023-10-05 21:57:25,960 - Epoch: [159][ 117/ 117] Loss 0.307758 Top1 85.008182 Top5 98.026250 +2023-10-05 21:57:26,077 - ==> Top1: 85.008 Top5: 98.026 Loss: 0.308 + +2023-10-05 21:57:26,078 - ==> Confusion: +[[ 935 1 6 1 7 3 0 0 3 67 1 0 1 2 6 2 4 0 0 0 11] + [ 0 1056 2 0 10 19 1 15 1 0 1 2 0 0 4 3 1 0 6 3 7] + [ 4 1 968 14 4 0 22 5 0 0 4 2 6 1 0 4 0 1 7 4 9] + [ 3 1 15 970 0 4 1 1 1 1 6 1 6 1 30 4 1 7 20 2 14] + [ 21 5 2 0 975 4 0 1 0 10 1 1 0 3 10 2 7 3 1 2 2] + [ 6 26 1 1 2 1000 1 21 1 1 3 4 0 14 9 0 4 0 4 6 12] + [ 0 5 29 0 1 0 1114 8 0 0 1 2 1 0 1 9 0 1 1 12 6] + [ 4 20 15 0 3 25 3 1071 1 2 3 10 1 3 1 1 1 0 38 7 9] + [ 18 1 0 0 1 0 0 1 971 47 9 0 2 14 16 2 0 0 1 4 2] + [ 86 0 3 1 3 3 0 0 16 970 1 1 1 19 2 5 0 0 0 1 7] + [ 2 5 11 5 0 3 5 4 10 3 971 3 0 11 2 1 3 1 1 1 11] + [ 1 0 1 0 0 15 0 0 0 1 0 961 19 6 0 3 1 16 0 8 3] + [ 1 2 3 2 0 3 0 2 1 0 1 41 975 2 1 4 3 11 2 4 10] + [ 1 0 1 0 2 6 0 0 9 11 5 6 2 1059 3 3 1 0 0 0 10] + [ 12 1 3 3 4 1 0 0 25 3 4 1 4 2 1013 0 1 1 13 0 10] + [ 0 2 3 0 2 0 0 1 0 0 0 6 9 2 1 1070 15 11 1 9 2] + [ 1 16 1 0 4 3 0 0 1 0 0 5 0 1 3 12 1099 0 0 4 11] + [ 1 0 1 3 1 0 2 0 1 0 0 2 22 2 0 6 0 991 1 1 4] + [ 1 6 6 20 1 1 0 18 1 0 1 0 2 0 8 0 1 0 993 1 8] + [ 0 0 5 3 3 7 10 10 1 0 0 15 3 4 0 6 8 2 1 1070 4] + [ 132 149 152 60 94 128 39 90 100 82 186 120 320 289 154 67 138 63 133 187 5222]] + +2023-10-05 21:57:26,079 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:57:26,079 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:57:26,085 - + +2023-10-05 21:57:26,085 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:57:27,070 - Epoch: [160][ 10/ 1236] Overall Loss 0.195423 Objective Loss 0.195423 LR 0.000250 Time 0.098469 +2023-10-05 21:57:27,270 - Epoch: [160][ 20/ 1236] Overall Loss 0.190302 Objective Loss 0.190302 LR 0.000250 Time 0.059205 +2023-10-05 21:57:27,469 - Epoch: [160][ 30/ 1236] Overall Loss 0.190832 Objective Loss 0.190832 LR 0.000250 Time 0.046084 +2023-10-05 21:57:27,669 - Epoch: [160][ 40/ 1236] Overall Loss 0.195110 Objective Loss 0.195110 LR 0.000250 Time 0.039552 +2023-10-05 21:57:27,867 - Epoch: [160][ 50/ 1236] Overall Loss 0.196420 Objective Loss 0.196420 LR 0.000250 Time 0.035610 +2023-10-05 21:57:28,067 - Epoch: [160][ 60/ 1236] Overall Loss 0.196586 Objective Loss 0.196586 LR 0.000250 Time 0.032997 +2023-10-05 21:57:28,266 - Epoch: [160][ 70/ 1236] Overall Loss 0.195165 Objective Loss 0.195165 LR 0.000250 Time 0.031126 +2023-10-05 21:57:28,466 - Epoch: [160][ 80/ 1236] Overall Loss 0.194663 Objective Loss 0.194663 LR 0.000250 Time 0.029728 +2023-10-05 21:57:28,665 - Epoch: [160][ 90/ 1236] Overall Loss 0.194828 Objective Loss 0.194828 LR 0.000250 Time 0.028628 +2023-10-05 21:57:28,865 - Epoch: [160][ 100/ 1236] Overall Loss 0.195270 Objective Loss 0.195270 LR 0.000250 Time 0.027760 +2023-10-05 21:57:29,063 - Epoch: [160][ 110/ 1236] Overall Loss 0.192779 Objective Loss 0.192779 LR 0.000250 Time 0.027039 +2023-10-05 21:57:29,263 - Epoch: [160][ 120/ 1236] Overall Loss 0.191495 Objective Loss 0.191495 LR 0.000250 Time 0.026445 +2023-10-05 21:57:29,461 - Epoch: [160][ 130/ 1236] Overall Loss 0.191463 Objective Loss 0.191463 LR 0.000250 Time 0.025936 +2023-10-05 21:57:29,661 - Epoch: [160][ 140/ 1236] Overall Loss 0.193975 Objective Loss 0.193975 LR 0.000250 Time 0.025506 +2023-10-05 21:57:29,859 - Epoch: [160][ 150/ 1236] Overall Loss 0.194990 Objective Loss 0.194990 LR 0.000250 Time 0.025125 +2023-10-05 21:57:30,059 - Epoch: [160][ 160/ 1236] Overall Loss 0.194805 Objective Loss 0.194805 LR 0.000250 Time 0.024804 +2023-10-05 21:57:30,258 - Epoch: [160][ 170/ 1236] Overall Loss 0.194471 Objective Loss 0.194471 LR 0.000250 Time 0.024510 +2023-10-05 21:57:30,456 - Epoch: [160][ 180/ 1236] Overall Loss 0.195660 Objective Loss 0.195660 LR 0.000250 Time 0.024251 +2023-10-05 21:57:30,656 - Epoch: [160][ 190/ 1236] Overall Loss 0.194400 Objective Loss 0.194400 LR 0.000250 Time 0.024022 +2023-10-05 21:57:30,855 - Epoch: [160][ 200/ 1236] Overall Loss 0.195006 Objective Loss 0.195006 LR 0.000250 Time 0.023818 +2023-10-05 21:57:31,054 - Epoch: [160][ 210/ 1236] Overall Loss 0.195947 Objective Loss 0.195947 LR 0.000250 Time 0.023629 +2023-10-05 21:57:31,254 - Epoch: [160][ 220/ 1236] Overall Loss 0.194824 Objective Loss 0.194824 LR 0.000250 Time 0.023462 +2023-10-05 21:57:31,452 - Epoch: [160][ 230/ 1236] Overall Loss 0.195299 Objective Loss 0.195299 LR 0.000250 Time 0.023300 +2023-10-05 21:57:31,652 - Epoch: [160][ 240/ 1236] Overall Loss 0.195089 Objective Loss 0.195089 LR 0.000250 Time 0.023162 +2023-10-05 21:57:31,851 - Epoch: [160][ 250/ 1236] Overall Loss 0.195101 Objective Loss 0.195101 LR 0.000250 Time 0.023030 +2023-10-05 21:57:32,051 - Epoch: [160][ 260/ 1236] Overall Loss 0.194333 Objective Loss 0.194333 LR 0.000250 Time 0.022912 +2023-10-05 21:57:32,250 - Epoch: [160][ 270/ 1236] Overall Loss 0.194179 Objective Loss 0.194179 LR 0.000250 Time 0.022800 +2023-10-05 21:57:32,450 - Epoch: [160][ 280/ 1236] Overall Loss 0.194278 Objective Loss 0.194278 LR 0.000250 Time 0.022696 +2023-10-05 21:57:32,648 - Epoch: [160][ 290/ 1236] Overall Loss 0.194140 Objective Loss 0.194140 LR 0.000250 Time 0.022597 +2023-10-05 21:57:32,848 - Epoch: [160][ 300/ 1236] Overall Loss 0.194608 Objective Loss 0.194608 LR 0.000250 Time 0.022510 +2023-10-05 21:57:33,047 - Epoch: [160][ 310/ 1236] Overall Loss 0.194127 Objective Loss 0.194127 LR 0.000250 Time 0.022423 +2023-10-05 21:57:33,247 - Epoch: [160][ 320/ 1236] Overall Loss 0.194423 Objective Loss 0.194423 LR 0.000250 Time 0.022346 +2023-10-05 21:57:33,446 - Epoch: [160][ 330/ 1236] Overall Loss 0.195247 Objective Loss 0.195247 LR 0.000250 Time 0.022272 +2023-10-05 21:57:33,647 - Epoch: [160][ 340/ 1236] Overall Loss 0.194853 Objective Loss 0.194853 LR 0.000250 Time 0.022205 +2023-10-05 21:57:33,845 - Epoch: [160][ 350/ 1236] Overall Loss 0.194648 Objective Loss 0.194648 LR 0.000250 Time 0.022137 +2023-10-05 21:57:34,045 - Epoch: [160][ 360/ 1236] Overall Loss 0.194451 Objective Loss 0.194451 LR 0.000250 Time 0.022077 +2023-10-05 21:57:34,244 - Epoch: [160][ 370/ 1236] Overall Loss 0.194753 Objective Loss 0.194753 LR 0.000250 Time 0.022017 +2023-10-05 21:57:34,445 - Epoch: [160][ 380/ 1236] Overall Loss 0.194353 Objective Loss 0.194353 LR 0.000250 Time 0.021964 +2023-10-05 21:57:34,644 - Epoch: [160][ 390/ 1236] Overall Loss 0.194088 Objective Loss 0.194088 LR 0.000250 Time 0.021911 +2023-10-05 21:57:34,845 - Epoch: [160][ 400/ 1236] Overall Loss 0.193992 Objective Loss 0.193992 LR 0.000250 Time 0.021864 +2023-10-05 21:57:35,044 - Epoch: [160][ 410/ 1236] Overall Loss 0.193714 Objective Loss 0.193714 LR 0.000250 Time 0.021817 +2023-10-05 21:57:35,245 - Epoch: [160][ 420/ 1236] Overall Loss 0.193823 Objective Loss 0.193823 LR 0.000250 Time 0.021774 +2023-10-05 21:57:35,444 - Epoch: [160][ 430/ 1236] Overall Loss 0.193808 Objective Loss 0.193808 LR 0.000250 Time 0.021731 +2023-10-05 21:57:35,645 - Epoch: [160][ 440/ 1236] Overall Loss 0.194198 Objective Loss 0.194198 LR 0.000250 Time 0.021691 +2023-10-05 21:57:35,844 - Epoch: [160][ 450/ 1236] Overall Loss 0.194057 Objective Loss 0.194057 LR 0.000250 Time 0.021651 +2023-10-05 21:57:36,044 - Epoch: [160][ 460/ 1236] Overall Loss 0.194322 Objective Loss 0.194322 LR 0.000250 Time 0.021615 +2023-10-05 21:57:36,243 - Epoch: [160][ 470/ 1236] Overall Loss 0.194503 Objective Loss 0.194503 LR 0.000250 Time 0.021578 +2023-10-05 21:57:36,444 - Epoch: [160][ 480/ 1236] Overall Loss 0.194441 Objective Loss 0.194441 LR 0.000250 Time 0.021546 +2023-10-05 21:57:36,643 - Epoch: [160][ 490/ 1236] Overall Loss 0.194489 Objective Loss 0.194489 LR 0.000250 Time 0.021513 +2023-10-05 21:57:36,844 - Epoch: [160][ 500/ 1236] Overall Loss 0.194288 Objective Loss 0.194288 LR 0.000250 Time 0.021483 +2023-10-05 21:57:37,043 - Epoch: [160][ 510/ 1236] Overall Loss 0.194218 Objective Loss 0.194218 LR 0.000250 Time 0.021452 +2023-10-05 21:57:37,244 - Epoch: [160][ 520/ 1236] Overall Loss 0.194007 Objective Loss 0.194007 LR 0.000250 Time 0.021424 +2023-10-05 21:57:37,443 - Epoch: [160][ 530/ 1236] Overall Loss 0.194445 Objective Loss 0.194445 LR 0.000250 Time 0.021396 +2023-10-05 21:57:37,644 - Epoch: [160][ 540/ 1236] Overall Loss 0.194209 Objective Loss 0.194209 LR 0.000250 Time 0.021370 +2023-10-05 21:57:37,843 - Epoch: [160][ 550/ 1236] Overall Loss 0.194254 Objective Loss 0.194254 LR 0.000250 Time 0.021344 +2023-10-05 21:57:38,044 - Epoch: [160][ 560/ 1236] Overall Loss 0.194472 Objective Loss 0.194472 LR 0.000250 Time 0.021320 +2023-10-05 21:57:38,243 - Epoch: [160][ 570/ 1236] Overall Loss 0.195253 Objective Loss 0.195253 LR 0.000250 Time 0.021296 +2023-10-05 21:57:38,444 - Epoch: [160][ 580/ 1236] Overall Loss 0.195330 Objective Loss 0.195330 LR 0.000250 Time 0.021274 +2023-10-05 21:57:38,643 - Epoch: [160][ 590/ 1236] Overall Loss 0.195300 Objective Loss 0.195300 LR 0.000250 Time 0.021251 +2023-10-05 21:57:38,844 - Epoch: [160][ 600/ 1236] Overall Loss 0.195202 Objective Loss 0.195202 LR 0.000250 Time 0.021230 +2023-10-05 21:57:39,043 - Epoch: [160][ 610/ 1236] Overall Loss 0.195418 Objective Loss 0.195418 LR 0.000250 Time 0.021209 +2023-10-05 21:57:39,244 - Epoch: [160][ 620/ 1236] Overall Loss 0.195494 Objective Loss 0.195494 LR 0.000250 Time 0.021189 +2023-10-05 21:57:39,443 - Epoch: [160][ 630/ 1236] Overall Loss 0.195691 Objective Loss 0.195691 LR 0.000250 Time 0.021169 +2023-10-05 21:57:39,644 - Epoch: [160][ 640/ 1236] Overall Loss 0.195768 Objective Loss 0.195768 LR 0.000250 Time 0.021151 +2023-10-05 21:57:39,843 - Epoch: [160][ 650/ 1236] Overall Loss 0.195595 Objective Loss 0.195595 LR 0.000250 Time 0.021131 +2023-10-05 21:57:40,043 - Epoch: [160][ 660/ 1236] Overall Loss 0.195543 Objective Loss 0.195543 LR 0.000250 Time 0.021115 +2023-10-05 21:57:40,243 - Epoch: [160][ 670/ 1236] Overall Loss 0.196267 Objective Loss 0.196267 LR 0.000250 Time 0.021096 +2023-10-05 21:57:40,443 - Epoch: [160][ 680/ 1236] Overall Loss 0.196076 Objective Loss 0.196076 LR 0.000250 Time 0.021080 +2023-10-05 21:57:40,642 - Epoch: [160][ 690/ 1236] Overall Loss 0.195947 Objective Loss 0.195947 LR 0.000250 Time 0.021063 +2023-10-05 21:57:40,843 - Epoch: [160][ 700/ 1236] Overall Loss 0.195933 Objective Loss 0.195933 LR 0.000250 Time 0.021048 +2023-10-05 21:57:41,042 - Epoch: [160][ 710/ 1236] Overall Loss 0.196104 Objective Loss 0.196104 LR 0.000250 Time 0.021032 +2023-10-05 21:57:41,243 - Epoch: [160][ 720/ 1236] Overall Loss 0.196212 Objective Loss 0.196212 LR 0.000250 Time 0.021019 +2023-10-05 21:57:41,442 - Epoch: [160][ 730/ 1236] Overall Loss 0.196302 Objective Loss 0.196302 LR 0.000250 Time 0.021003 +2023-10-05 21:57:41,643 - Epoch: [160][ 740/ 1236] Overall Loss 0.196377 Objective Loss 0.196377 LR 0.000250 Time 0.020990 +2023-10-05 21:57:41,844 - Epoch: [160][ 750/ 1236] Overall Loss 0.196070 Objective Loss 0.196070 LR 0.000250 Time 0.020978 +2023-10-05 21:57:42,045 - Epoch: [160][ 760/ 1236] Overall Loss 0.196164 Objective Loss 0.196164 LR 0.000250 Time 0.020966 +2023-10-05 21:57:42,244 - Epoch: [160][ 770/ 1236] Overall Loss 0.196160 Objective Loss 0.196160 LR 0.000250 Time 0.020952 +2023-10-05 21:57:42,445 - Epoch: [160][ 780/ 1236] Overall Loss 0.196283 Objective Loss 0.196283 LR 0.000250 Time 0.020940 +2023-10-05 21:57:42,644 - Epoch: [160][ 790/ 1236] Overall Loss 0.196389 Objective Loss 0.196389 LR 0.000250 Time 0.020927 +2023-10-05 21:57:42,845 - Epoch: [160][ 800/ 1236] Overall Loss 0.196697 Objective Loss 0.196697 LR 0.000250 Time 0.020916 +2023-10-05 21:57:43,044 - Epoch: [160][ 810/ 1236] Overall Loss 0.197079 Objective Loss 0.197079 LR 0.000250 Time 0.020903 +2023-10-05 21:57:43,245 - Epoch: [160][ 820/ 1236] Overall Loss 0.196971 Objective Loss 0.196971 LR 0.000250 Time 0.020892 +2023-10-05 21:57:43,444 - Epoch: [160][ 830/ 1236] Overall Loss 0.197120 Objective Loss 0.197120 LR 0.000250 Time 0.020881 +2023-10-05 21:57:43,645 - Epoch: [160][ 840/ 1236] Overall Loss 0.197280 Objective Loss 0.197280 LR 0.000250 Time 0.020871 +2023-10-05 21:57:43,844 - Epoch: [160][ 850/ 1236] Overall Loss 0.197343 Objective Loss 0.197343 LR 0.000250 Time 0.020859 +2023-10-05 21:57:44,045 - Epoch: [160][ 860/ 1236] Overall Loss 0.197416 Objective Loss 0.197416 LR 0.000250 Time 0.020849 +2023-10-05 21:57:44,244 - Epoch: [160][ 870/ 1236] Overall Loss 0.197424 Objective Loss 0.197424 LR 0.000250 Time 0.020839 +2023-10-05 21:57:44,445 - Epoch: [160][ 880/ 1236] Overall Loss 0.197756 Objective Loss 0.197756 LR 0.000250 Time 0.020830 +2023-10-05 21:57:44,644 - Epoch: [160][ 890/ 1236] Overall Loss 0.198101 Objective Loss 0.198101 LR 0.000250 Time 0.020819 +2023-10-05 21:57:44,845 - Epoch: [160][ 900/ 1236] Overall Loss 0.198138 Objective Loss 0.198138 LR 0.000250 Time 0.020810 +2023-10-05 21:57:45,044 - Epoch: [160][ 910/ 1236] Overall Loss 0.197988 Objective Loss 0.197988 LR 0.000250 Time 0.020800 +2023-10-05 21:57:45,245 - Epoch: [160][ 920/ 1236] Overall Loss 0.198009 Objective Loss 0.198009 LR 0.000250 Time 0.020792 +2023-10-05 21:57:45,444 - Epoch: [160][ 930/ 1236] Overall Loss 0.198075 Objective Loss 0.198075 LR 0.000250 Time 0.020782 +2023-10-05 21:57:45,645 - Epoch: [160][ 940/ 1236] Overall Loss 0.198323 Objective Loss 0.198323 LR 0.000250 Time 0.020774 +2023-10-05 21:57:45,844 - Epoch: [160][ 950/ 1236] Overall Loss 0.198308 Objective Loss 0.198308 LR 0.000250 Time 0.020765 +2023-10-05 21:57:46,045 - Epoch: [160][ 960/ 1236] Overall Loss 0.198207 Objective Loss 0.198207 LR 0.000250 Time 0.020757 +2023-10-05 21:57:46,244 - Epoch: [160][ 970/ 1236] Overall Loss 0.198211 Objective Loss 0.198211 LR 0.000250 Time 0.020748 +2023-10-05 21:57:46,445 - Epoch: [160][ 980/ 1236] Overall Loss 0.198115 Objective Loss 0.198115 LR 0.000250 Time 0.020741 +2023-10-05 21:57:46,644 - Epoch: [160][ 990/ 1236] Overall Loss 0.197860 Objective Loss 0.197860 LR 0.000250 Time 0.020732 +2023-10-05 21:57:46,845 - Epoch: [160][ 1000/ 1236] Overall Loss 0.197972 Objective Loss 0.197972 LR 0.000250 Time 0.020725 +2023-10-05 21:57:47,044 - Epoch: [160][ 1010/ 1236] Overall Loss 0.197946 Objective Loss 0.197946 LR 0.000250 Time 0.020717 +2023-10-05 21:57:47,244 - Epoch: [160][ 1020/ 1236] Overall Loss 0.197968 Objective Loss 0.197968 LR 0.000250 Time 0.020710 +2023-10-05 21:57:47,444 - Epoch: [160][ 1030/ 1236] Overall Loss 0.197963 Objective Loss 0.197963 LR 0.000250 Time 0.020702 +2023-10-05 21:57:47,644 - Epoch: [160][ 1040/ 1236] Overall Loss 0.197920 Objective Loss 0.197920 LR 0.000250 Time 0.020696 +2023-10-05 21:57:47,843 - Epoch: [160][ 1050/ 1236] Overall Loss 0.198015 Objective Loss 0.198015 LR 0.000250 Time 0.020688 +2023-10-05 21:57:48,044 - Epoch: [160][ 1060/ 1236] Overall Loss 0.198302 Objective Loss 0.198302 LR 0.000250 Time 0.020682 +2023-10-05 21:57:48,243 - Epoch: [160][ 1070/ 1236] Overall Loss 0.198316 Objective Loss 0.198316 LR 0.000250 Time 0.020675 +2023-10-05 21:57:48,444 - Epoch: [160][ 1080/ 1236] Overall Loss 0.198454 Objective Loss 0.198454 LR 0.000250 Time 0.020669 +2023-10-05 21:57:48,643 - Epoch: [160][ 1090/ 1236] Overall Loss 0.198507 Objective Loss 0.198507 LR 0.000250 Time 0.020662 +2023-10-05 21:57:48,844 - Epoch: [160][ 1100/ 1236] Overall Loss 0.198383 Objective Loss 0.198383 LR 0.000250 Time 0.020656 +2023-10-05 21:57:49,043 - Epoch: [160][ 1110/ 1236] Overall Loss 0.198502 Objective Loss 0.198502 LR 0.000250 Time 0.020649 +2023-10-05 21:57:49,244 - Epoch: [160][ 1120/ 1236] Overall Loss 0.198501 Objective Loss 0.198501 LR 0.000250 Time 0.020644 +2023-10-05 21:57:49,443 - Epoch: [160][ 1130/ 1236] Overall Loss 0.198505 Objective Loss 0.198505 LR 0.000250 Time 0.020637 +2023-10-05 21:57:49,644 - Epoch: [160][ 1140/ 1236] Overall Loss 0.198715 Objective Loss 0.198715 LR 0.000250 Time 0.020632 +2023-10-05 21:57:49,843 - Epoch: [160][ 1150/ 1236] Overall Loss 0.198819 Objective Loss 0.198819 LR 0.000250 Time 0.020626 +2023-10-05 21:57:50,044 - Epoch: [160][ 1160/ 1236] Overall Loss 0.198787 Objective Loss 0.198787 LR 0.000250 Time 0.020620 +2023-10-05 21:57:50,244 - Epoch: [160][ 1170/ 1236] Overall Loss 0.198810 Objective Loss 0.198810 LR 0.000250 Time 0.020614 +2023-10-05 21:57:50,444 - Epoch: [160][ 1180/ 1236] Overall Loss 0.198797 Objective Loss 0.198797 LR 0.000250 Time 0.020609 +2023-10-05 21:57:50,643 - Epoch: [160][ 1190/ 1236] Overall Loss 0.198901 Objective Loss 0.198901 LR 0.000250 Time 0.020604 +2023-10-05 21:57:50,842 - Epoch: [160][ 1200/ 1236] Overall Loss 0.199072 Objective Loss 0.199072 LR 0.000250 Time 0.020597 +2023-10-05 21:57:51,042 - Epoch: [160][ 1210/ 1236] Overall Loss 0.199067 Objective Loss 0.199067 LR 0.000250 Time 0.020592 +2023-10-05 21:57:51,243 - Epoch: [160][ 1220/ 1236] Overall Loss 0.199148 Objective Loss 0.199148 LR 0.000250 Time 0.020588 +2023-10-05 21:57:51,493 - Epoch: [160][ 1230/ 1236] Overall Loss 0.199165 Objective Loss 0.199165 LR 0.000250 Time 0.020623 +2023-10-05 21:57:51,610 - Epoch: [160][ 1236/ 1236] Overall Loss 0.199104 Objective Loss 0.199104 Top1 87.983707 Top5 99.389002 LR 0.000250 Time 0.020618 +2023-10-05 21:57:51,739 - --- validate (epoch=160)----------- +2023-10-05 21:57:51,740 - 29943 samples (256 per mini-batch) +2023-10-05 21:57:52,193 - Epoch: [160][ 10/ 117] Loss 0.324295 Top1 85.546875 Top5 97.734375 +2023-10-05 21:57:52,338 - Epoch: [160][ 20/ 117] Loss 0.309114 Top1 85.800781 Top5 97.929688 +2023-10-05 21:57:52,481 - Epoch: [160][ 30/ 117] Loss 0.317047 Top1 85.690104 Top5 97.994792 +2023-10-05 21:57:52,624 - Epoch: [160][ 40/ 117] Loss 0.318286 Top1 85.380859 Top5 97.880859 +2023-10-05 21:57:52,769 - Epoch: [160][ 50/ 117] Loss 0.315260 Top1 85.250000 Top5 97.921875 +2023-10-05 21:57:52,913 - Epoch: [160][ 60/ 117] Loss 0.314711 Top1 85.032552 Top5 97.949219 +2023-10-05 21:57:53,057 - Epoch: [160][ 70/ 117] Loss 0.314949 Top1 85.005580 Top5 97.952009 +2023-10-05 21:57:53,201 - Epoch: [160][ 80/ 117] Loss 0.314471 Top1 85.004883 Top5 97.968750 +2023-10-05 21:57:53,345 - Epoch: [160][ 90/ 117] Loss 0.312294 Top1 84.960938 Top5 97.899306 +2023-10-05 21:57:53,490 - Epoch: [160][ 100/ 117] Loss 0.313828 Top1 84.871094 Top5 97.914062 +2023-10-05 21:57:53,640 - Epoch: [160][ 110/ 117] Loss 0.314105 Top1 84.904119 Top5 97.926136 +2023-10-05 21:57:53,725 - Epoch: [160][ 117/ 117] Loss 0.312530 Top1 84.911332 Top5 97.969475 +2023-10-05 21:57:53,847 - ==> Top1: 84.911 Top5: 97.969 Loss: 0.313 + +2023-10-05 21:57:53,848 - ==> Confusion: +[[ 932 1 2 2 3 2 0 1 7 74 1 1 0 2 4 2 3 3 0 0 10] + [ 0 1073 1 0 5 18 1 10 3 0 1 3 0 0 1 2 2 0 6 0 5] + [ 5 1 970 10 1 1 23 7 0 1 5 2 8 3 0 3 1 0 6 4 5] + [ 3 1 11 960 0 4 1 3 1 1 7 1 8 1 29 3 0 7 31 1 16] + [ 31 6 0 0 966 2 0 1 0 10 1 1 0 3 8 4 6 4 0 1 6] + [ 3 37 1 1 2 988 0 28 3 1 5 8 0 14 2 2 2 0 5 2 12] + [ 0 7 23 0 0 0 1116 13 0 0 2 2 1 0 1 9 0 3 1 10 3] + [ 2 18 17 0 2 31 0 1080 1 5 3 6 1 2 0 4 1 0 28 11 6] + [ 15 3 0 0 0 1 0 2 987 43 8 4 1 5 10 6 0 0 3 0 1] + [ 101 0 2 1 3 3 0 0 24 954 1 2 1 12 2 4 0 1 0 1 7] + [ 3 4 9 5 0 2 3 4 16 2 969 2 1 13 3 1 0 1 4 1 10] + [ 1 0 1 0 0 12 0 4 0 1 0 970 15 6 0 2 0 16 0 3 4] + [ 1 4 1 2 0 3 0 2 0 0 2 43 974 2 1 5 2 14 2 4 6] + [ 3 0 1 0 2 4 0 0 17 15 6 7 2 1049 3 2 1 0 0 0 7] + [ 13 1 2 9 4 1 0 0 35 4 1 1 1 4 993 0 1 2 16 0 13] + [ 0 5 1 0 4 0 1 0 0 0 0 9 10 2 0 1062 12 13 0 11 4] + [ 0 13 1 0 5 4 0 2 2 0 0 6 0 1 2 10 1101 0 0 4 10] + [ 0 0 0 2 1 0 4 0 1 0 0 2 15 0 0 3 0 1007 2 0 1] + [ 1 9 4 12 1 0 0 24 2 0 1 0 0 1 10 1 0 0 992 0 10] + [ 0 3 3 2 1 9 9 10 2 0 0 17 3 4 0 6 7 2 3 1064 7] + [ 138 189 148 46 73 150 32 111 136 80 163 143 325 253 137 41 106 85 156 175 5218]] + +2023-10-05 21:57:53,849 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:57:53,849 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:57:53,855 - + +2023-10-05 21:57:53,855 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:57:54,834 - Epoch: [161][ 10/ 1236] Overall Loss 0.228993 Objective Loss 0.228993 LR 0.000250 Time 0.097789 +2023-10-05 21:57:55,034 - Epoch: [161][ 20/ 1236] Overall Loss 0.210378 Objective Loss 0.210378 LR 0.000250 Time 0.058896 +2023-10-05 21:57:55,232 - Epoch: [161][ 30/ 1236] Overall Loss 0.210075 Objective Loss 0.210075 LR 0.000250 Time 0.045861 +2023-10-05 21:57:55,434 - Epoch: [161][ 40/ 1236] Overall Loss 0.201573 Objective Loss 0.201573 LR 0.000250 Time 0.039423 +2023-10-05 21:57:55,632 - Epoch: [161][ 50/ 1236] Overall Loss 0.209037 Objective Loss 0.209037 LR 0.000250 Time 0.035502 +2023-10-05 21:57:55,833 - Epoch: [161][ 60/ 1236] Overall Loss 0.204558 Objective Loss 0.204558 LR 0.000250 Time 0.032930 +2023-10-05 21:57:56,032 - Epoch: [161][ 70/ 1236] Overall Loss 0.203319 Objective Loss 0.203319 LR 0.000250 Time 0.031062 +2023-10-05 21:57:56,233 - Epoch: [161][ 80/ 1236] Overall Loss 0.201610 Objective Loss 0.201610 LR 0.000250 Time 0.029688 +2023-10-05 21:57:56,432 - Epoch: [161][ 90/ 1236] Overall Loss 0.202526 Objective Loss 0.202526 LR 0.000250 Time 0.028591 +2023-10-05 21:57:56,632 - Epoch: [161][ 100/ 1236] Overall Loss 0.201873 Objective Loss 0.201873 LR 0.000250 Time 0.027736 +2023-10-05 21:57:56,830 - Epoch: [161][ 110/ 1236] Overall Loss 0.200908 Objective Loss 0.200908 LR 0.000250 Time 0.027011 +2023-10-05 21:57:57,031 - Epoch: [161][ 120/ 1236] Overall Loss 0.200942 Objective Loss 0.200942 LR 0.000250 Time 0.026430 +2023-10-05 21:57:57,230 - Epoch: [161][ 130/ 1236] Overall Loss 0.199776 Objective Loss 0.199776 LR 0.000250 Time 0.025921 +2023-10-05 21:57:57,430 - Epoch: [161][ 140/ 1236] Overall Loss 0.200705 Objective Loss 0.200705 LR 0.000250 Time 0.025502 +2023-10-05 21:57:57,629 - Epoch: [161][ 150/ 1236] Overall Loss 0.200513 Objective Loss 0.200513 LR 0.000250 Time 0.025122 +2023-10-05 21:57:57,830 - Epoch: [161][ 160/ 1236] Overall Loss 0.199283 Objective Loss 0.199283 LR 0.000250 Time 0.024808 +2023-10-05 21:57:58,029 - Epoch: [161][ 170/ 1236] Overall Loss 0.198861 Objective Loss 0.198861 LR 0.000250 Time 0.024514 +2023-10-05 21:57:58,229 - Epoch: [161][ 180/ 1236] Overall Loss 0.199260 Objective Loss 0.199260 LR 0.000250 Time 0.024266 +2023-10-05 21:57:58,428 - Epoch: [161][ 190/ 1236] Overall Loss 0.199024 Objective Loss 0.199024 LR 0.000250 Time 0.024032 +2023-10-05 21:57:58,629 - Epoch: [161][ 200/ 1236] Overall Loss 0.198395 Objective Loss 0.198395 LR 0.000250 Time 0.023834 +2023-10-05 21:57:58,828 - Epoch: [161][ 210/ 1236] Overall Loss 0.197775 Objective Loss 0.197775 LR 0.000250 Time 0.023643 +2023-10-05 21:57:59,028 - Epoch: [161][ 220/ 1236] Overall Loss 0.196719 Objective Loss 0.196719 LR 0.000250 Time 0.023479 +2023-10-05 21:57:59,227 - Epoch: [161][ 230/ 1236] Overall Loss 0.196418 Objective Loss 0.196418 LR 0.000250 Time 0.023320 +2023-10-05 21:57:59,428 - Epoch: [161][ 240/ 1236] Overall Loss 0.195886 Objective Loss 0.195886 LR 0.000250 Time 0.023185 +2023-10-05 21:57:59,627 - Epoch: [161][ 250/ 1236] Overall Loss 0.195527 Objective Loss 0.195527 LR 0.000250 Time 0.023051 +2023-10-05 21:57:59,827 - Epoch: [161][ 260/ 1236] Overall Loss 0.196085 Objective Loss 0.196085 LR 0.000250 Time 0.022935 +2023-10-05 21:58:00,026 - Epoch: [161][ 270/ 1236] Overall Loss 0.195930 Objective Loss 0.195930 LR 0.000250 Time 0.022820 +2023-10-05 21:58:00,227 - Epoch: [161][ 280/ 1236] Overall Loss 0.195986 Objective Loss 0.195986 LR 0.000250 Time 0.022722 +2023-10-05 21:58:00,426 - Epoch: [161][ 290/ 1236] Overall Loss 0.195766 Objective Loss 0.195766 LR 0.000250 Time 0.022622 +2023-10-05 21:58:00,627 - Epoch: [161][ 300/ 1236] Overall Loss 0.196080 Objective Loss 0.196080 LR 0.000250 Time 0.022537 +2023-10-05 21:58:00,825 - Epoch: [161][ 310/ 1236] Overall Loss 0.196003 Objective Loss 0.196003 LR 0.000250 Time 0.022448 +2023-10-05 21:58:01,026 - Epoch: [161][ 320/ 1236] Overall Loss 0.196143 Objective Loss 0.196143 LR 0.000250 Time 0.022374 +2023-10-05 21:58:01,225 - Epoch: [161][ 330/ 1236] Overall Loss 0.196016 Objective Loss 0.196016 LR 0.000250 Time 0.022296 +2023-10-05 21:58:01,426 - Epoch: [161][ 340/ 1236] Overall Loss 0.195772 Objective Loss 0.195772 LR 0.000250 Time 0.022230 +2023-10-05 21:58:01,624 - Epoch: [161][ 350/ 1236] Overall Loss 0.195628 Objective Loss 0.195628 LR 0.000250 Time 0.022162 +2023-10-05 21:58:01,825 - Epoch: [161][ 360/ 1236] Overall Loss 0.195793 Objective Loss 0.195793 LR 0.000250 Time 0.022103 +2023-10-05 21:58:02,024 - Epoch: [161][ 370/ 1236] Overall Loss 0.196012 Objective Loss 0.196012 LR 0.000250 Time 0.022042 +2023-10-05 21:58:02,225 - Epoch: [161][ 380/ 1236] Overall Loss 0.196102 Objective Loss 0.196102 LR 0.000250 Time 0.021990 +2023-10-05 21:58:02,424 - Epoch: [161][ 390/ 1236] Overall Loss 0.195533 Objective Loss 0.195533 LR 0.000250 Time 0.021935 +2023-10-05 21:58:02,625 - Epoch: [161][ 400/ 1236] Overall Loss 0.195830 Objective Loss 0.195830 LR 0.000250 Time 0.021888 +2023-10-05 21:58:02,823 - Epoch: [161][ 410/ 1236] Overall Loss 0.195680 Objective Loss 0.195680 LR 0.000250 Time 0.021838 +2023-10-05 21:58:03,024 - Epoch: [161][ 420/ 1236] Overall Loss 0.196065 Objective Loss 0.196065 LR 0.000250 Time 0.021796 +2023-10-05 21:58:03,223 - Epoch: [161][ 430/ 1236] Overall Loss 0.195838 Objective Loss 0.195838 LR 0.000250 Time 0.021751 +2023-10-05 21:58:03,424 - Epoch: [161][ 440/ 1236] Overall Loss 0.195656 Objective Loss 0.195656 LR 0.000250 Time 0.021713 +2023-10-05 21:58:03,623 - Epoch: [161][ 450/ 1236] Overall Loss 0.195749 Objective Loss 0.195749 LR 0.000250 Time 0.021672 +2023-10-05 21:58:03,825 - Epoch: [161][ 460/ 1236] Overall Loss 0.195330 Objective Loss 0.195330 LR 0.000250 Time 0.021637 +2023-10-05 21:58:04,023 - Epoch: [161][ 470/ 1236] Overall Loss 0.195550 Objective Loss 0.195550 LR 0.000250 Time 0.021599 +2023-10-05 21:58:04,224 - Epoch: [161][ 480/ 1236] Overall Loss 0.195583 Objective Loss 0.195583 LR 0.000250 Time 0.021567 +2023-10-05 21:58:04,423 - Epoch: [161][ 490/ 1236] Overall Loss 0.195690 Objective Loss 0.195690 LR 0.000250 Time 0.021532 +2023-10-05 21:58:04,624 - Epoch: [161][ 500/ 1236] Overall Loss 0.195888 Objective Loss 0.195888 LR 0.000250 Time 0.021503 +2023-10-05 21:58:04,824 - Epoch: [161][ 510/ 1236] Overall Loss 0.196113 Objective Loss 0.196113 LR 0.000250 Time 0.021471 +2023-10-05 21:58:05,025 - Epoch: [161][ 520/ 1236] Overall Loss 0.196251 Objective Loss 0.196251 LR 0.000250 Time 0.021444 +2023-10-05 21:58:05,223 - Epoch: [161][ 530/ 1236] Overall Loss 0.196173 Objective Loss 0.196173 LR 0.000250 Time 0.021413 +2023-10-05 21:58:05,424 - Epoch: [161][ 540/ 1236] Overall Loss 0.195835 Objective Loss 0.195835 LR 0.000250 Time 0.021388 +2023-10-05 21:58:05,622 - Epoch: [161][ 550/ 1236] Overall Loss 0.195614 Objective Loss 0.195614 LR 0.000250 Time 0.021360 +2023-10-05 21:58:05,824 - Epoch: [161][ 560/ 1236] Overall Loss 0.195464 Objective Loss 0.195464 LR 0.000250 Time 0.021337 +2023-10-05 21:58:06,022 - Epoch: [161][ 570/ 1236] Overall Loss 0.195477 Objective Loss 0.195477 LR 0.000250 Time 0.021311 +2023-10-05 21:58:06,223 - Epoch: [161][ 580/ 1236] Overall Loss 0.195887 Objective Loss 0.195887 LR 0.000250 Time 0.021289 +2023-10-05 21:58:06,422 - Epoch: [161][ 590/ 1236] Overall Loss 0.196067 Objective Loss 0.196067 LR 0.000250 Time 0.021264 +2023-10-05 21:58:06,623 - Epoch: [161][ 600/ 1236] Overall Loss 0.195961 Objective Loss 0.195961 LR 0.000250 Time 0.021244 +2023-10-05 21:58:06,822 - Epoch: [161][ 610/ 1236] Overall Loss 0.196354 Objective Loss 0.196354 LR 0.000250 Time 0.021221 +2023-10-05 21:58:07,022 - Epoch: [161][ 620/ 1236] Overall Loss 0.196700 Objective Loss 0.196700 LR 0.000250 Time 0.021202 +2023-10-05 21:58:07,221 - Epoch: [161][ 630/ 1236] Overall Loss 0.196577 Objective Loss 0.196577 LR 0.000250 Time 0.021180 +2023-10-05 21:58:07,422 - Epoch: [161][ 640/ 1236] Overall Loss 0.196521 Objective Loss 0.196521 LR 0.000250 Time 0.021163 +2023-10-05 21:58:07,621 - Epoch: [161][ 650/ 1236] Overall Loss 0.196573 Objective Loss 0.196573 LR 0.000250 Time 0.021143 +2023-10-05 21:58:07,822 - Epoch: [161][ 660/ 1236] Overall Loss 0.196319 Objective Loss 0.196319 LR 0.000250 Time 0.021127 +2023-10-05 21:58:08,021 - Epoch: [161][ 670/ 1236] Overall Loss 0.196498 Objective Loss 0.196498 LR 0.000250 Time 0.021108 +2023-10-05 21:58:08,222 - Epoch: [161][ 680/ 1236] Overall Loss 0.196272 Objective Loss 0.196272 LR 0.000250 Time 0.021093 +2023-10-05 21:58:08,422 - Epoch: [161][ 690/ 1236] Overall Loss 0.196427 Objective Loss 0.196427 LR 0.000250 Time 0.021076 +2023-10-05 21:58:08,623 - Epoch: [161][ 700/ 1236] Overall Loss 0.196854 Objective Loss 0.196854 LR 0.000250 Time 0.021061 +2023-10-05 21:58:08,822 - Epoch: [161][ 710/ 1236] Overall Loss 0.196795 Objective Loss 0.196795 LR 0.000250 Time 0.021044 +2023-10-05 21:58:09,023 - Epoch: [161][ 720/ 1236] Overall Loss 0.196502 Objective Loss 0.196502 LR 0.000250 Time 0.021031 +2023-10-05 21:58:09,222 - Epoch: [161][ 730/ 1236] Overall Loss 0.196260 Objective Loss 0.196260 LR 0.000250 Time 0.021015 +2023-10-05 21:58:09,423 - Epoch: [161][ 740/ 1236] Overall Loss 0.196164 Objective Loss 0.196164 LR 0.000250 Time 0.021002 +2023-10-05 21:58:09,622 - Epoch: [161][ 750/ 1236] Overall Loss 0.196345 Objective Loss 0.196345 LR 0.000250 Time 0.020987 +2023-10-05 21:58:09,823 - Epoch: [161][ 760/ 1236] Overall Loss 0.196630 Objective Loss 0.196630 LR 0.000250 Time 0.020975 +2023-10-05 21:58:10,021 - Epoch: [161][ 770/ 1236] Overall Loss 0.196930 Objective Loss 0.196930 LR 0.000250 Time 0.020960 +2023-10-05 21:58:10,223 - Epoch: [161][ 780/ 1236] Overall Loss 0.196956 Objective Loss 0.196956 LR 0.000250 Time 0.020949 +2023-10-05 21:58:10,421 - Epoch: [161][ 790/ 1236] Overall Loss 0.197071 Objective Loss 0.197071 LR 0.000250 Time 0.020934 +2023-10-05 21:58:10,622 - Epoch: [161][ 800/ 1236] Overall Loss 0.197056 Objective Loss 0.197056 LR 0.000250 Time 0.020923 +2023-10-05 21:58:10,821 - Epoch: [161][ 810/ 1236] Overall Loss 0.197127 Objective Loss 0.197127 LR 0.000250 Time 0.020911 +2023-10-05 21:58:11,022 - Epoch: [161][ 820/ 1236] Overall Loss 0.197355 Objective Loss 0.197355 LR 0.000250 Time 0.020900 +2023-10-05 21:58:11,221 - Epoch: [161][ 830/ 1236] Overall Loss 0.197333 Objective Loss 0.197333 LR 0.000250 Time 0.020887 +2023-10-05 21:58:11,422 - Epoch: [161][ 840/ 1236] Overall Loss 0.197306 Objective Loss 0.197306 LR 0.000250 Time 0.020877 +2023-10-05 21:58:11,621 - Epoch: [161][ 850/ 1236] Overall Loss 0.197347 Objective Loss 0.197347 LR 0.000250 Time 0.020866 +2023-10-05 21:58:11,822 - Epoch: [161][ 860/ 1236] Overall Loss 0.197530 Objective Loss 0.197530 LR 0.000250 Time 0.020857 +2023-10-05 21:58:12,022 - Epoch: [161][ 870/ 1236] Overall Loss 0.197238 Objective Loss 0.197238 LR 0.000250 Time 0.020846 +2023-10-05 21:58:12,223 - Epoch: [161][ 880/ 1236] Overall Loss 0.197038 Objective Loss 0.197038 LR 0.000250 Time 0.020837 +2023-10-05 21:58:12,422 - Epoch: [161][ 890/ 1236] Overall Loss 0.196810 Objective Loss 0.196810 LR 0.000250 Time 0.020826 +2023-10-05 21:58:12,623 - Epoch: [161][ 900/ 1236] Overall Loss 0.196851 Objective Loss 0.196851 LR 0.000250 Time 0.020817 +2023-10-05 21:58:12,822 - Epoch: [161][ 910/ 1236] Overall Loss 0.196966 Objective Loss 0.196966 LR 0.000250 Time 0.020807 +2023-10-05 21:58:13,023 - Epoch: [161][ 920/ 1236] Overall Loss 0.196904 Objective Loss 0.196904 LR 0.000250 Time 0.020799 +2023-10-05 21:58:13,221 - Epoch: [161][ 930/ 1236] Overall Loss 0.196878 Objective Loss 0.196878 LR 0.000250 Time 0.020788 +2023-10-05 21:58:13,422 - Epoch: [161][ 940/ 1236] Overall Loss 0.196535 Objective Loss 0.196535 LR 0.000250 Time 0.020780 +2023-10-05 21:58:13,619 - Epoch: [161][ 950/ 1236] Overall Loss 0.196661 Objective Loss 0.196661 LR 0.000250 Time 0.020769 +2023-10-05 21:58:13,820 - Epoch: [161][ 960/ 1236] Overall Loss 0.196665 Objective Loss 0.196665 LR 0.000250 Time 0.020762 +2023-10-05 21:58:14,019 - Epoch: [161][ 970/ 1236] Overall Loss 0.196618 Objective Loss 0.196618 LR 0.000250 Time 0.020752 +2023-10-05 21:58:14,220 - Epoch: [161][ 980/ 1236] Overall Loss 0.196519 Objective Loss 0.196519 LR 0.000250 Time 0.020745 +2023-10-05 21:58:14,418 - Epoch: [161][ 990/ 1236] Overall Loss 0.196504 Objective Loss 0.196504 LR 0.000250 Time 0.020736 +2023-10-05 21:58:14,619 - Epoch: [161][ 1000/ 1236] Overall Loss 0.196525 Objective Loss 0.196525 LR 0.000250 Time 0.020729 +2023-10-05 21:58:14,817 - Epoch: [161][ 1010/ 1236] Overall Loss 0.196558 Objective Loss 0.196558 LR 0.000250 Time 0.020720 +2023-10-05 21:58:15,019 - Epoch: [161][ 1020/ 1236] Overall Loss 0.196520 Objective Loss 0.196520 LR 0.000250 Time 0.020713 +2023-10-05 21:58:15,217 - Epoch: [161][ 1030/ 1236] Overall Loss 0.196277 Objective Loss 0.196277 LR 0.000250 Time 0.020705 +2023-10-05 21:58:15,418 - Epoch: [161][ 1040/ 1236] Overall Loss 0.196307 Objective Loss 0.196307 LR 0.000250 Time 0.020699 +2023-10-05 21:58:15,615 - Epoch: [161][ 1050/ 1236] Overall Loss 0.196581 Objective Loss 0.196581 LR 0.000250 Time 0.020689 +2023-10-05 21:58:15,815 - Epoch: [161][ 1060/ 1236] Overall Loss 0.196689 Objective Loss 0.196689 LR 0.000250 Time 0.020682 +2023-10-05 21:58:16,012 - Epoch: [161][ 1070/ 1236] Overall Loss 0.196710 Objective Loss 0.196710 LR 0.000250 Time 0.020672 +2023-10-05 21:58:16,212 - Epoch: [161][ 1080/ 1236] Overall Loss 0.196695 Objective Loss 0.196695 LR 0.000250 Time 0.020666 +2023-10-05 21:58:16,409 - Epoch: [161][ 1090/ 1236] Overall Loss 0.196848 Objective Loss 0.196848 LR 0.000250 Time 0.020656 +2023-10-05 21:58:16,608 - Epoch: [161][ 1100/ 1236] Overall Loss 0.196906 Objective Loss 0.196906 LR 0.000250 Time 0.020650 +2023-10-05 21:58:16,805 - Epoch: [161][ 1110/ 1236] Overall Loss 0.196963 Objective Loss 0.196963 LR 0.000250 Time 0.020641 +2023-10-05 21:58:17,005 - Epoch: [161][ 1120/ 1236] Overall Loss 0.196995 Objective Loss 0.196995 LR 0.000250 Time 0.020635 +2023-10-05 21:58:17,202 - Epoch: [161][ 1130/ 1236] Overall Loss 0.197157 Objective Loss 0.197157 LR 0.000250 Time 0.020626 +2023-10-05 21:58:17,402 - Epoch: [161][ 1140/ 1236] Overall Loss 0.197248 Objective Loss 0.197248 LR 0.000250 Time 0.020620 +2023-10-05 21:58:17,599 - Epoch: [161][ 1150/ 1236] Overall Loss 0.197396 Objective Loss 0.197396 LR 0.000250 Time 0.020612 +2023-10-05 21:58:17,798 - Epoch: [161][ 1160/ 1236] Overall Loss 0.197468 Objective Loss 0.197468 LR 0.000250 Time 0.020606 +2023-10-05 21:58:17,996 - Epoch: [161][ 1170/ 1236] Overall Loss 0.197399 Objective Loss 0.197399 LR 0.000250 Time 0.020598 +2023-10-05 21:58:18,195 - Epoch: [161][ 1180/ 1236] Overall Loss 0.197597 Objective Loss 0.197597 LR 0.000250 Time 0.020593 +2023-10-05 21:58:18,392 - Epoch: [161][ 1190/ 1236] Overall Loss 0.197960 Objective Loss 0.197960 LR 0.000250 Time 0.020585 +2023-10-05 21:58:18,592 - Epoch: [161][ 1200/ 1236] Overall Loss 0.197960 Objective Loss 0.197960 LR 0.000250 Time 0.020579 +2023-10-05 21:58:18,789 - Epoch: [161][ 1210/ 1236] Overall Loss 0.198006 Objective Loss 0.198006 LR 0.000250 Time 0.020572 +2023-10-05 21:58:18,989 - Epoch: [161][ 1220/ 1236] Overall Loss 0.197894 Objective Loss 0.197894 LR 0.000250 Time 0.020567 +2023-10-05 21:58:19,237 - Epoch: [161][ 1230/ 1236] Overall Loss 0.198013 Objective Loss 0.198013 LR 0.000250 Time 0.020601 +2023-10-05 21:58:19,354 - Epoch: [161][ 1236/ 1236] Overall Loss 0.198055 Objective Loss 0.198055 Top1 88.391039 Top5 98.981670 LR 0.000250 Time 0.020595 +2023-10-05 21:58:19,480 - --- validate (epoch=161)----------- +2023-10-05 21:58:19,480 - 29943 samples (256 per mini-batch) +2023-10-05 21:58:19,933 - Epoch: [161][ 10/ 117] Loss 0.327845 Top1 85.351562 Top5 97.734375 +2023-10-05 21:58:20,086 - Epoch: [161][ 20/ 117] Loss 0.311972 Top1 85.371094 Top5 97.871094 +2023-10-05 21:58:20,235 - Epoch: [161][ 30/ 117] Loss 0.312548 Top1 85.403646 Top5 97.968750 +2023-10-05 21:58:20,387 - Epoch: [161][ 40/ 117] Loss 0.306623 Top1 85.585938 Top5 98.007812 +2023-10-05 21:58:20,541 - Epoch: [161][ 50/ 117] Loss 0.312388 Top1 85.320312 Top5 98.000000 +2023-10-05 21:58:20,691 - Epoch: [161][ 60/ 117] Loss 0.313200 Top1 85.253906 Top5 98.027344 +2023-10-05 21:58:20,841 - Epoch: [161][ 70/ 117] Loss 0.310324 Top1 85.217634 Top5 98.080357 +2023-10-05 21:58:20,989 - Epoch: [161][ 80/ 117] Loss 0.312451 Top1 85.224609 Top5 98.095703 +2023-10-05 21:58:21,139 - Epoch: [161][ 90/ 117] Loss 0.311974 Top1 85.230035 Top5 98.081597 +2023-10-05 21:58:21,286 - Epoch: [161][ 100/ 117] Loss 0.311461 Top1 85.183594 Top5 98.062500 +2023-10-05 21:58:21,443 - Epoch: [161][ 110/ 117] Loss 0.310452 Top1 85.248580 Top5 98.100142 +2023-10-05 21:58:21,529 - Epoch: [161][ 117/ 117] Loss 0.309942 Top1 85.275357 Top5 98.089704 +2023-10-05 21:58:21,645 - ==> Top1: 85.275 Top5: 98.090 Loss: 0.310 + +2023-10-05 21:58:21,646 - ==> Confusion: +[[ 926 2 3 3 8 3 0 0 5 73 2 0 0 1 6 2 3 0 0 0 13] + [ 0 1077 2 0 7 12 1 11 1 0 2 1 0 0 1 4 2 0 3 0 7] + [ 1 1 968 13 3 0 23 4 0 0 3 0 7 2 0 5 1 1 4 9 11] + [ 2 1 9 953 0 4 1 2 2 1 8 1 9 1 27 3 0 7 35 1 22] + [ 22 10 1 0 975 2 0 1 0 8 0 1 0 1 5 6 8 2 0 1 7] + [ 4 37 0 0 1 990 2 22 0 2 3 3 0 16 5 2 4 0 3 5 17] + [ 0 9 21 0 0 1 1119 9 0 1 0 2 1 0 1 10 0 1 2 7 7] + [ 4 22 12 0 3 31 3 1069 1 3 2 9 0 3 1 4 0 0 36 6 9] + [ 16 4 0 0 0 1 1 0 972 46 8 3 1 12 10 6 1 0 2 1 5] + [ 88 1 3 0 3 2 0 0 19 959 1 1 0 17 8 8 1 1 0 0 7] + [ 3 8 9 3 1 0 6 5 17 0 957 2 0 15 6 2 3 0 7 2 7] + [ 1 0 1 0 1 15 0 3 0 1 0 950 19 4 0 3 1 16 0 15 5] + [ 0 2 3 4 0 3 0 1 0 0 1 33 981 2 0 7 3 13 4 4 7] + [ 5 0 1 0 1 2 0 0 7 15 4 6 4 1059 3 3 0 1 0 0 8] + [ 13 5 3 11 6 0 0 0 22 3 3 1 2 1 1006 0 1 1 13 0 10] + [ 0 4 1 0 3 0 0 0 0 0 0 5 8 2 1 1070 17 12 0 9 2] + [ 0 20 1 0 6 1 0 0 2 0 0 1 0 1 4 8 1104 0 0 4 9] + [ 0 1 0 3 1 0 2 0 0 1 0 2 14 0 1 7 0 1000 1 0 5] + [ 1 5 7 15 1 0 0 23 1 1 5 0 0 0 7 0 0 0 992 0 10] + [ 0 3 3 2 2 3 7 11 1 0 1 13 3 1 0 4 7 1 0 1082 8] + [ 107 217 129 44 87 116 37 83 98 80 142 90 333 283 126 61 144 57 144 202 5325]] + +2023-10-05 21:58:21,647 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:58:21,647 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:58:21,653 - + +2023-10-05 21:58:21,653 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:58:22,781 - Epoch: [162][ 10/ 1236] Overall Loss 0.185570 Objective Loss 0.185570 LR 0.000250 Time 0.112769 +2023-10-05 21:58:22,985 - Epoch: [162][ 20/ 1236] Overall Loss 0.198746 Objective Loss 0.198746 LR 0.000250 Time 0.066531 +2023-10-05 21:58:23,186 - Epoch: [162][ 30/ 1236] Overall Loss 0.194712 Objective Loss 0.194712 LR 0.000250 Time 0.051060 +2023-10-05 21:58:23,389 - Epoch: [162][ 40/ 1236] Overall Loss 0.195716 Objective Loss 0.195716 LR 0.000250 Time 0.043364 +2023-10-05 21:58:23,591 - Epoch: [162][ 50/ 1236] Overall Loss 0.192819 Objective Loss 0.192819 LR 0.000250 Time 0.038718 +2023-10-05 21:58:23,794 - Epoch: [162][ 60/ 1236] Overall Loss 0.192702 Objective Loss 0.192702 LR 0.000250 Time 0.035646 +2023-10-05 21:58:23,996 - Epoch: [162][ 70/ 1236] Overall Loss 0.191180 Objective Loss 0.191180 LR 0.000250 Time 0.033432 +2023-10-05 21:58:24,199 - Epoch: [162][ 80/ 1236] Overall Loss 0.191159 Objective Loss 0.191159 LR 0.000250 Time 0.031789 +2023-10-05 21:58:24,401 - Epoch: [162][ 90/ 1236] Overall Loss 0.190476 Objective Loss 0.190476 LR 0.000250 Time 0.030495 +2023-10-05 21:58:24,604 - Epoch: [162][ 100/ 1236] Overall Loss 0.190911 Objective Loss 0.190911 LR 0.000250 Time 0.029475 +2023-10-05 21:58:24,806 - Epoch: [162][ 110/ 1236] Overall Loss 0.191136 Objective Loss 0.191136 LR 0.000250 Time 0.028622 +2023-10-05 21:58:25,010 - Epoch: [162][ 120/ 1236] Overall Loss 0.192958 Objective Loss 0.192958 LR 0.000250 Time 0.027931 +2023-10-05 21:58:25,215 - Epoch: [162][ 130/ 1236] Overall Loss 0.193317 Objective Loss 0.193317 LR 0.000250 Time 0.027362 +2023-10-05 21:58:25,421 - Epoch: [162][ 140/ 1236] Overall Loss 0.193919 Objective Loss 0.193919 LR 0.000250 Time 0.026878 +2023-10-05 21:58:25,628 - Epoch: [162][ 150/ 1236] Overall Loss 0.194618 Objective Loss 0.194618 LR 0.000250 Time 0.026463 +2023-10-05 21:58:25,834 - Epoch: [162][ 160/ 1236] Overall Loss 0.195029 Objective Loss 0.195029 LR 0.000250 Time 0.026096 +2023-10-05 21:58:26,040 - Epoch: [162][ 170/ 1236] Overall Loss 0.194467 Objective Loss 0.194467 LR 0.000250 Time 0.025771 +2023-10-05 21:58:26,246 - Epoch: [162][ 180/ 1236] Overall Loss 0.196033 Objective Loss 0.196033 LR 0.000250 Time 0.025481 +2023-10-05 21:58:26,454 - Epoch: [162][ 190/ 1236] Overall Loss 0.196077 Objective Loss 0.196077 LR 0.000250 Time 0.025229 +2023-10-05 21:58:26,660 - Epoch: [162][ 200/ 1236] Overall Loss 0.196469 Objective Loss 0.196469 LR 0.000250 Time 0.024996 +2023-10-05 21:58:26,867 - Epoch: [162][ 210/ 1236] Overall Loss 0.196080 Objective Loss 0.196080 LR 0.000250 Time 0.024790 +2023-10-05 21:58:27,073 - Epoch: [162][ 220/ 1236] Overall Loss 0.195748 Objective Loss 0.195748 LR 0.000250 Time 0.024599 +2023-10-05 21:58:27,280 - Epoch: [162][ 230/ 1236] Overall Loss 0.194776 Objective Loss 0.194776 LR 0.000250 Time 0.024427 +2023-10-05 21:58:27,486 - Epoch: [162][ 240/ 1236] Overall Loss 0.194312 Objective Loss 0.194312 LR 0.000250 Time 0.024266 +2023-10-05 21:58:27,693 - Epoch: [162][ 250/ 1236] Overall Loss 0.194699 Objective Loss 0.194699 LR 0.000250 Time 0.024123 +2023-10-05 21:58:27,902 - Epoch: [162][ 260/ 1236] Overall Loss 0.194212 Objective Loss 0.194212 LR 0.000250 Time 0.024000 +2023-10-05 21:58:28,116 - Epoch: [162][ 270/ 1236] Overall Loss 0.194515 Objective Loss 0.194515 LR 0.000250 Time 0.023901 +2023-10-05 21:58:28,332 - Epoch: [162][ 280/ 1236] Overall Loss 0.194680 Objective Loss 0.194680 LR 0.000250 Time 0.023818 +2023-10-05 21:58:28,546 - Epoch: [162][ 290/ 1236] Overall Loss 0.195908 Objective Loss 0.195908 LR 0.000250 Time 0.023734 +2023-10-05 21:58:28,762 - Epoch: [162][ 300/ 1236] Overall Loss 0.195873 Objective Loss 0.195873 LR 0.000250 Time 0.023661 +2023-10-05 21:58:28,976 - Epoch: [162][ 310/ 1236] Overall Loss 0.195649 Objective Loss 0.195649 LR 0.000250 Time 0.023588 +2023-10-05 21:58:29,192 - Epoch: [162][ 320/ 1236] Overall Loss 0.196306 Objective Loss 0.196306 LR 0.000250 Time 0.023525 +2023-10-05 21:58:29,407 - Epoch: [162][ 330/ 1236] Overall Loss 0.196741 Objective Loss 0.196741 LR 0.000250 Time 0.023460 +2023-10-05 21:58:29,623 - Epoch: [162][ 340/ 1236] Overall Loss 0.195983 Objective Loss 0.195983 LR 0.000250 Time 0.023405 +2023-10-05 21:58:29,837 - Epoch: [162][ 350/ 1236] Overall Loss 0.195565 Objective Loss 0.195565 LR 0.000250 Time 0.023346 +2023-10-05 21:58:30,052 - Epoch: [162][ 360/ 1236] Overall Loss 0.195379 Objective Loss 0.195379 LR 0.000250 Time 0.023296 +2023-10-05 21:58:30,266 - Epoch: [162][ 370/ 1236] Overall Loss 0.195134 Objective Loss 0.195134 LR 0.000250 Time 0.023242 +2023-10-05 21:58:30,481 - Epoch: [162][ 380/ 1236] Overall Loss 0.195579 Objective Loss 0.195579 LR 0.000250 Time 0.023197 +2023-10-05 21:58:30,694 - Epoch: [162][ 390/ 1236] Overall Loss 0.196011 Objective Loss 0.196011 LR 0.000250 Time 0.023147 +2023-10-05 21:58:30,910 - Epoch: [162][ 400/ 1236] Overall Loss 0.196529 Objective Loss 0.196529 LR 0.000250 Time 0.023107 +2023-10-05 21:58:31,122 - Epoch: [162][ 410/ 1236] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000250 Time 0.023060 +2023-10-05 21:58:31,337 - Epoch: [162][ 420/ 1236] Overall Loss 0.196218 Objective Loss 0.196218 LR 0.000250 Time 0.023021 +2023-10-05 21:58:31,549 - Epoch: [162][ 430/ 1236] Overall Loss 0.196348 Objective Loss 0.196348 LR 0.000250 Time 0.022978 +2023-10-05 21:58:31,763 - Epoch: [162][ 440/ 1236] Overall Loss 0.196530 Objective Loss 0.196530 LR 0.000250 Time 0.022942 +2023-10-05 21:58:31,975 - Epoch: [162][ 450/ 1236] Overall Loss 0.196964 Objective Loss 0.196964 LR 0.000250 Time 0.022902 +2023-10-05 21:58:32,189 - Epoch: [162][ 460/ 1236] Overall Loss 0.197141 Objective Loss 0.197141 LR 0.000250 Time 0.022870 +2023-10-05 21:58:32,401 - Epoch: [162][ 470/ 1236] Overall Loss 0.197370 Objective Loss 0.197370 LR 0.000250 Time 0.022833 +2023-10-05 21:58:32,616 - Epoch: [162][ 480/ 1236] Overall Loss 0.197110 Objective Loss 0.197110 LR 0.000250 Time 0.022804 +2023-10-05 21:58:32,828 - Epoch: [162][ 490/ 1236] Overall Loss 0.196347 Objective Loss 0.196347 LR 0.000250 Time 0.022771 +2023-10-05 21:58:33,042 - Epoch: [162][ 500/ 1236] Overall Loss 0.196076 Objective Loss 0.196076 LR 0.000250 Time 0.022743 +2023-10-05 21:58:33,254 - Epoch: [162][ 510/ 1236] Overall Loss 0.196175 Objective Loss 0.196175 LR 0.000250 Time 0.022712 +2023-10-05 21:58:33,469 - Epoch: [162][ 520/ 1236] Overall Loss 0.195884 Objective Loss 0.195884 LR 0.000250 Time 0.022687 +2023-10-05 21:58:33,681 - Epoch: [162][ 530/ 1236] Overall Loss 0.195791 Objective Loss 0.195791 LR 0.000250 Time 0.022659 +2023-10-05 21:58:33,893 - Epoch: [162][ 540/ 1236] Overall Loss 0.196563 Objective Loss 0.196563 LR 0.000250 Time 0.022632 +2023-10-05 21:58:34,101 - Epoch: [162][ 550/ 1236] Overall Loss 0.196589 Objective Loss 0.196589 LR 0.000250 Time 0.022598 +2023-10-05 21:58:34,314 - Epoch: [162][ 560/ 1236] Overall Loss 0.196762 Objective Loss 0.196762 LR 0.000250 Time 0.022573 +2023-10-05 21:58:34,522 - Epoch: [162][ 570/ 1236] Overall Loss 0.196576 Objective Loss 0.196576 LR 0.000250 Time 0.022542 +2023-10-05 21:58:34,734 - Epoch: [162][ 580/ 1236] Overall Loss 0.196428 Objective Loss 0.196428 LR 0.000250 Time 0.022518 +2023-10-05 21:58:34,942 - Epoch: [162][ 590/ 1236] Overall Loss 0.196061 Objective Loss 0.196061 LR 0.000250 Time 0.022489 +2023-10-05 21:58:35,155 - Epoch: [162][ 600/ 1236] Overall Loss 0.196085 Objective Loss 0.196085 LR 0.000250 Time 0.022468 +2023-10-05 21:58:35,363 - Epoch: [162][ 610/ 1236] Overall Loss 0.196352 Objective Loss 0.196352 LR 0.000250 Time 0.022440 +2023-10-05 21:58:35,571 - Epoch: [162][ 620/ 1236] Overall Loss 0.196659 Objective Loss 0.196659 LR 0.000250 Time 0.022414 +2023-10-05 21:58:35,775 - Epoch: [162][ 630/ 1236] Overall Loss 0.196567 Objective Loss 0.196567 LR 0.000250 Time 0.022381 +2023-10-05 21:58:35,980 - Epoch: [162][ 640/ 1236] Overall Loss 0.196570 Objective Loss 0.196570 LR 0.000250 Time 0.022351 +2023-10-05 21:58:36,183 - Epoch: [162][ 650/ 1236] Overall Loss 0.196817 Objective Loss 0.196817 LR 0.000250 Time 0.022319 +2023-10-05 21:58:36,388 - Epoch: [162][ 660/ 1236] Overall Loss 0.196703 Objective Loss 0.196703 LR 0.000250 Time 0.022291 +2023-10-05 21:58:36,592 - Epoch: [162][ 670/ 1236] Overall Loss 0.196799 Objective Loss 0.196799 LR 0.000250 Time 0.022262 +2023-10-05 21:58:36,796 - Epoch: [162][ 680/ 1236] Overall Loss 0.196691 Objective Loss 0.196691 LR 0.000250 Time 0.022235 +2023-10-05 21:58:36,999 - Epoch: [162][ 690/ 1236] Overall Loss 0.196665 Objective Loss 0.196665 LR 0.000250 Time 0.022206 +2023-10-05 21:58:37,203 - Epoch: [162][ 700/ 1236] Overall Loss 0.196866 Objective Loss 0.196866 LR 0.000250 Time 0.022179 +2023-10-05 21:58:37,404 - Epoch: [162][ 710/ 1236] Overall Loss 0.196978 Objective Loss 0.196978 LR 0.000250 Time 0.022150 +2023-10-05 21:58:37,607 - Epoch: [162][ 720/ 1236] Overall Loss 0.196862 Objective Loss 0.196862 LR 0.000250 Time 0.022124 +2023-10-05 21:58:37,809 - Epoch: [162][ 730/ 1236] Overall Loss 0.196996 Objective Loss 0.196996 LR 0.000250 Time 0.022097 +2023-10-05 21:58:38,012 - Epoch: [162][ 740/ 1236] Overall Loss 0.197069 Objective Loss 0.197069 LR 0.000250 Time 0.022072 +2023-10-05 21:58:38,214 - Epoch: [162][ 750/ 1236] Overall Loss 0.197109 Objective Loss 0.197109 LR 0.000250 Time 0.022047 +2023-10-05 21:58:38,417 - Epoch: [162][ 760/ 1236] Overall Loss 0.196994 Objective Loss 0.196994 LR 0.000250 Time 0.022024 +2023-10-05 21:58:38,619 - Epoch: [162][ 770/ 1236] Overall Loss 0.196733 Objective Loss 0.196733 LR 0.000250 Time 0.021999 +2023-10-05 21:58:38,822 - Epoch: [162][ 780/ 1236] Overall Loss 0.196829 Objective Loss 0.196829 LR 0.000250 Time 0.021977 +2023-10-05 21:58:39,023 - Epoch: [162][ 790/ 1236] Overall Loss 0.196568 Objective Loss 0.196568 LR 0.000250 Time 0.021953 +2023-10-05 21:58:39,226 - Epoch: [162][ 800/ 1236] Overall Loss 0.196586 Objective Loss 0.196586 LR 0.000250 Time 0.021932 +2023-10-05 21:58:39,428 - Epoch: [162][ 810/ 1236] Overall Loss 0.196290 Objective Loss 0.196290 LR 0.000250 Time 0.021910 +2023-10-05 21:58:39,631 - Epoch: [162][ 820/ 1236] Overall Loss 0.196266 Objective Loss 0.196266 LR 0.000250 Time 0.021890 +2023-10-05 21:58:39,833 - Epoch: [162][ 830/ 1236] Overall Loss 0.196363 Objective Loss 0.196363 LR 0.000250 Time 0.021868 +2023-10-05 21:58:40,036 - Epoch: [162][ 840/ 1236] Overall Loss 0.196122 Objective Loss 0.196122 LR 0.000250 Time 0.021849 +2023-10-05 21:58:40,237 - Epoch: [162][ 850/ 1236] Overall Loss 0.195832 Objective Loss 0.195832 LR 0.000250 Time 0.021829 +2023-10-05 21:58:40,441 - Epoch: [162][ 860/ 1236] Overall Loss 0.195682 Objective Loss 0.195682 LR 0.000250 Time 0.021811 +2023-10-05 21:58:40,642 - Epoch: [162][ 870/ 1236] Overall Loss 0.195546 Objective Loss 0.195546 LR 0.000250 Time 0.021792 +2023-10-05 21:58:40,845 - Epoch: [162][ 880/ 1236] Overall Loss 0.195654 Objective Loss 0.195654 LR 0.000250 Time 0.021775 +2023-10-05 21:58:41,047 - Epoch: [162][ 890/ 1236] Overall Loss 0.195512 Objective Loss 0.195512 LR 0.000250 Time 0.021756 +2023-10-05 21:58:41,249 - Epoch: [162][ 900/ 1236] Overall Loss 0.195487 Objective Loss 0.195487 LR 0.000250 Time 0.021739 +2023-10-05 21:58:41,451 - Epoch: [162][ 910/ 1236] Overall Loss 0.195745 Objective Loss 0.195745 LR 0.000250 Time 0.021721 +2023-10-05 21:58:41,654 - Epoch: [162][ 920/ 1236] Overall Loss 0.195809 Objective Loss 0.195809 LR 0.000250 Time 0.021706 +2023-10-05 21:58:41,856 - Epoch: [162][ 930/ 1236] Overall Loss 0.196046 Objective Loss 0.196046 LR 0.000250 Time 0.021689 +2023-10-05 21:58:42,059 - Epoch: [162][ 940/ 1236] Overall Loss 0.196061 Objective Loss 0.196061 LR 0.000250 Time 0.021673 +2023-10-05 21:58:42,260 - Epoch: [162][ 950/ 1236] Overall Loss 0.195994 Objective Loss 0.195994 LR 0.000250 Time 0.021657 +2023-10-05 21:58:42,463 - Epoch: [162][ 960/ 1236] Overall Loss 0.196005 Objective Loss 0.196005 LR 0.000250 Time 0.021642 +2023-10-05 21:58:42,665 - Epoch: [162][ 970/ 1236] Overall Loss 0.195568 Objective Loss 0.195568 LR 0.000250 Time 0.021627 +2023-10-05 21:58:42,868 - Epoch: [162][ 980/ 1236] Overall Loss 0.195731 Objective Loss 0.195731 LR 0.000250 Time 0.021613 +2023-10-05 21:58:43,069 - Epoch: [162][ 990/ 1236] Overall Loss 0.195976 Objective Loss 0.195976 LR 0.000250 Time 0.021598 +2023-10-05 21:58:43,273 - Epoch: [162][ 1000/ 1236] Overall Loss 0.195774 Objective Loss 0.195774 LR 0.000250 Time 0.021585 +2023-10-05 21:58:43,474 - Epoch: [162][ 1010/ 1236] Overall Loss 0.195739 Objective Loss 0.195739 LR 0.000250 Time 0.021570 +2023-10-05 21:58:43,677 - Epoch: [162][ 1020/ 1236] Overall Loss 0.195662 Objective Loss 0.195662 LR 0.000250 Time 0.021557 +2023-10-05 21:58:43,878 - Epoch: [162][ 1030/ 1236] Overall Loss 0.195739 Objective Loss 0.195739 LR 0.000250 Time 0.021543 +2023-10-05 21:58:44,085 - Epoch: [162][ 1040/ 1236] Overall Loss 0.195506 Objective Loss 0.195506 LR 0.000250 Time 0.021534 +2023-10-05 21:58:44,291 - Epoch: [162][ 1050/ 1236] Overall Loss 0.195619 Objective Loss 0.195619 LR 0.000250 Time 0.021525 +2023-10-05 21:58:44,497 - Epoch: [162][ 1060/ 1236] Overall Loss 0.195584 Objective Loss 0.195584 LR 0.000250 Time 0.021516 +2023-10-05 21:58:44,704 - Epoch: [162][ 1070/ 1236] Overall Loss 0.195887 Objective Loss 0.195887 LR 0.000250 Time 0.021507 +2023-10-05 21:58:44,909 - Epoch: [162][ 1080/ 1236] Overall Loss 0.195916 Objective Loss 0.195916 LR 0.000250 Time 0.021498 +2023-10-05 21:58:45,115 - Epoch: [162][ 1090/ 1236] Overall Loss 0.195858 Objective Loss 0.195858 LR 0.000250 Time 0.021490 +2023-10-05 21:58:45,322 - Epoch: [162][ 1100/ 1236] Overall Loss 0.196087 Objective Loss 0.196087 LR 0.000250 Time 0.021482 +2023-10-05 21:58:45,528 - Epoch: [162][ 1110/ 1236] Overall Loss 0.196116 Objective Loss 0.196116 LR 0.000250 Time 0.021473 +2023-10-05 21:58:45,734 - Epoch: [162][ 1120/ 1236] Overall Loss 0.196079 Objective Loss 0.196079 LR 0.000250 Time 0.021466 +2023-10-05 21:58:45,940 - Epoch: [162][ 1130/ 1236] Overall Loss 0.196188 Objective Loss 0.196188 LR 0.000250 Time 0.021458 +2023-10-05 21:58:46,147 - Epoch: [162][ 1140/ 1236] Overall Loss 0.196243 Objective Loss 0.196243 LR 0.000250 Time 0.021450 +2023-10-05 21:58:46,353 - Epoch: [162][ 1150/ 1236] Overall Loss 0.196479 Objective Loss 0.196479 LR 0.000250 Time 0.021442 +2023-10-05 21:58:46,559 - Epoch: [162][ 1160/ 1236] Overall Loss 0.196546 Objective Loss 0.196546 LR 0.000250 Time 0.021434 +2023-10-05 21:58:46,764 - Epoch: [162][ 1170/ 1236] Overall Loss 0.196533 Objective Loss 0.196533 LR 0.000250 Time 0.021427 +2023-10-05 21:58:46,971 - Epoch: [162][ 1180/ 1236] Overall Loss 0.196652 Objective Loss 0.196652 LR 0.000250 Time 0.021420 +2023-10-05 21:58:47,176 - Epoch: [162][ 1190/ 1236] Overall Loss 0.196675 Objective Loss 0.196675 LR 0.000250 Time 0.021412 +2023-10-05 21:58:47,383 - Epoch: [162][ 1200/ 1236] Overall Loss 0.196524 Objective Loss 0.196524 LR 0.000250 Time 0.021405 +2023-10-05 21:58:47,589 - Epoch: [162][ 1210/ 1236] Overall Loss 0.196710 Objective Loss 0.196710 LR 0.000250 Time 0.021398 +2023-10-05 21:58:47,795 - Epoch: [162][ 1220/ 1236] Overall Loss 0.196740 Objective Loss 0.196740 LR 0.000250 Time 0.021392 +2023-10-05 21:58:48,055 - Epoch: [162][ 1230/ 1236] Overall Loss 0.196613 Objective Loss 0.196613 LR 0.000250 Time 0.021429 +2023-10-05 21:58:48,174 - Epoch: [162][ 1236/ 1236] Overall Loss 0.196558 Objective Loss 0.196558 Top1 90.020367 Top5 98.778004 LR 0.000250 Time 0.021421 +2023-10-05 21:58:48,298 - --- validate (epoch=162)----------- +2023-10-05 21:58:48,299 - 29943 samples (256 per mini-batch) +2023-10-05 21:58:48,772 - Epoch: [162][ 10/ 117] Loss 0.286269 Top1 85.664062 Top5 98.554688 +2023-10-05 21:58:48,935 - Epoch: [162][ 20/ 117] Loss 0.295794 Top1 85.390625 Top5 98.457031 +2023-10-05 21:58:49,098 - Epoch: [162][ 30/ 117] Loss 0.287073 Top1 85.234375 Top5 98.320312 +2023-10-05 21:58:49,260 - Epoch: [162][ 40/ 117] Loss 0.294546 Top1 84.941406 Top5 98.212891 +2023-10-05 21:58:49,422 - Epoch: [162][ 50/ 117] Loss 0.299086 Top1 84.953125 Top5 98.234375 +2023-10-05 21:58:49,584 - Epoch: [162][ 60/ 117] Loss 0.299650 Top1 85.006510 Top5 98.170573 +2023-10-05 21:58:49,742 - Epoch: [162][ 70/ 117] Loss 0.302009 Top1 85.022321 Top5 98.186384 +2023-10-05 21:58:49,902 - Epoch: [162][ 80/ 117] Loss 0.304586 Top1 85.014648 Top5 98.178711 +2023-10-05 21:58:50,062 - Epoch: [162][ 90/ 117] Loss 0.303234 Top1 85.056424 Top5 98.168403 +2023-10-05 21:58:50,225 - Epoch: [162][ 100/ 117] Loss 0.303149 Top1 85.082031 Top5 98.171875 +2023-10-05 21:58:50,396 - Epoch: [162][ 110/ 117] Loss 0.304241 Top1 85.003551 Top5 98.192472 +2023-10-05 21:58:50,482 - Epoch: [162][ 117/ 117] Loss 0.307449 Top1 84.954747 Top5 98.189894 +2023-10-05 21:58:50,619 - ==> Top1: 84.955 Top5: 98.190 Loss: 0.307 + +2023-10-05 21:58:50,620 - ==> Confusion: +[[ 948 2 6 1 7 4 0 0 5 45 2 0 1 1 5 0 4 1 2 0 16] + [ 2 1073 2 0 6 18 1 11 0 0 0 1 0 0 1 4 2 0 5 0 5] + [ 8 1 968 17 3 0 15 8 0 1 3 1 9 4 0 4 1 1 3 3 6] + [ 3 1 9 976 1 2 1 1 0 1 5 0 5 0 30 3 0 8 22 1 20] + [ 22 8 1 0 980 3 1 1 0 6 1 1 0 3 8 1 7 1 0 2 4] + [ 5 33 0 0 5 1001 1 19 1 0 4 4 0 11 5 2 3 0 8 3 11] + [ 0 7 26 0 1 0 1115 9 0 0 3 2 1 0 1 8 0 1 3 7 7] + [ 4 25 14 0 4 25 4 1066 3 1 2 5 2 2 0 3 0 0 42 7 9] + [ 20 3 0 1 0 4 0 0 968 41 12 2 2 10 14 5 1 0 5 0 1] + [ 116 0 3 0 5 4 0 0 25 919 0 0 0 19 11 7 1 1 0 1 7] + [ 3 2 8 6 2 1 3 4 8 1 970 1 1 13 3 3 3 0 9 1 11] + [ 1 0 0 0 0 17 0 3 0 1 0 954 26 4 0 5 2 14 0 4 4] + [ 2 1 1 5 1 4 1 2 0 0 1 34 983 0 0 6 3 15 2 2 5] + [ 1 0 2 0 2 7 0 0 8 8 3 6 1 1068 5 1 1 0 0 0 6] + [ 13 1 3 7 3 0 0 0 22 1 2 1 2 1 1022 0 1 1 9 0 12] + [ 0 4 1 0 3 0 1 0 0 0 0 6 7 1 1 1066 15 16 0 8 5] + [ 0 16 1 0 5 3 0 0 2 0 0 4 0 1 3 9 1105 0 0 2 10] + [ 0 0 0 2 0 0 2 0 0 0 0 2 18 2 0 7 0 999 1 0 5] + [ 2 9 5 16 1 0 0 18 1 1 2 0 1 1 12 0 0 0 990 1 8] + [ 0 2 3 4 2 5 5 11 1 0 2 17 5 2 0 7 10 1 2 1060 13] + [ 136 192 132 65 105 151 25 81 94 58 155 104 326 287 175 58 154 67 166 167 5207]] + +2023-10-05 21:58:50,621 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:58:50,621 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:58:50,627 - + +2023-10-05 21:58:50,627 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:58:51,652 - Epoch: [163][ 10/ 1236] Overall Loss 0.189423 Objective Loss 0.189423 LR 0.000250 Time 0.102366 +2023-10-05 21:58:51,855 - Epoch: [163][ 20/ 1236] Overall Loss 0.193218 Objective Loss 0.193218 LR 0.000250 Time 0.061328 +2023-10-05 21:58:52,060 - Epoch: [163][ 30/ 1236] Overall Loss 0.187283 Objective Loss 0.187283 LR 0.000250 Time 0.047714 +2023-10-05 21:58:52,265 - Epoch: [163][ 40/ 1236] Overall Loss 0.181127 Objective Loss 0.181127 LR 0.000250 Time 0.040902 +2023-10-05 21:58:52,470 - Epoch: [163][ 50/ 1236] Overall Loss 0.178986 Objective Loss 0.178986 LR 0.000250 Time 0.036817 +2023-10-05 21:58:52,675 - Epoch: [163][ 60/ 1236] Overall Loss 0.180055 Objective Loss 0.180055 LR 0.000250 Time 0.034085 +2023-10-05 21:58:52,880 - Epoch: [163][ 70/ 1236] Overall Loss 0.183809 Objective Loss 0.183809 LR 0.000250 Time 0.032144 +2023-10-05 21:58:53,085 - Epoch: [163][ 80/ 1236] Overall Loss 0.186426 Objective Loss 0.186426 LR 0.000250 Time 0.030682 +2023-10-05 21:58:53,291 - Epoch: [163][ 90/ 1236] Overall Loss 0.188123 Objective Loss 0.188123 LR 0.000250 Time 0.029558 +2023-10-05 21:58:53,496 - Epoch: [163][ 100/ 1236] Overall Loss 0.187619 Objective Loss 0.187619 LR 0.000250 Time 0.028647 +2023-10-05 21:58:53,701 - Epoch: [163][ 110/ 1236] Overall Loss 0.187149 Objective Loss 0.187149 LR 0.000250 Time 0.027907 +2023-10-05 21:58:53,905 - Epoch: [163][ 120/ 1236] Overall Loss 0.187435 Objective Loss 0.187435 LR 0.000250 Time 0.027274 +2023-10-05 21:58:54,110 - Epoch: [163][ 130/ 1236] Overall Loss 0.188130 Objective Loss 0.188130 LR 0.000250 Time 0.026752 +2023-10-05 21:58:54,312 - Epoch: [163][ 140/ 1236] Overall Loss 0.189171 Objective Loss 0.189171 LR 0.000250 Time 0.026281 +2023-10-05 21:58:54,517 - Epoch: [163][ 150/ 1236] Overall Loss 0.190022 Objective Loss 0.190022 LR 0.000250 Time 0.025892 +2023-10-05 21:58:54,724 - Epoch: [163][ 160/ 1236] Overall Loss 0.190583 Objective Loss 0.190583 LR 0.000250 Time 0.025564 +2023-10-05 21:58:54,930 - Epoch: [163][ 170/ 1236] Overall Loss 0.189929 Objective Loss 0.189929 LR 0.000250 Time 0.025275 +2023-10-05 21:58:55,137 - Epoch: [163][ 180/ 1236] Overall Loss 0.191977 Objective Loss 0.191977 LR 0.000250 Time 0.025019 +2023-10-05 21:58:55,344 - Epoch: [163][ 190/ 1236] Overall Loss 0.193393 Objective Loss 0.193393 LR 0.000250 Time 0.024787 +2023-10-05 21:58:55,551 - Epoch: [163][ 200/ 1236] Overall Loss 0.192167 Objective Loss 0.192167 LR 0.000250 Time 0.024580 +2023-10-05 21:58:55,758 - Epoch: [163][ 210/ 1236] Overall Loss 0.190998 Objective Loss 0.190998 LR 0.000250 Time 0.024397 +2023-10-05 21:58:55,965 - Epoch: [163][ 220/ 1236] Overall Loss 0.191074 Objective Loss 0.191074 LR 0.000250 Time 0.024228 +2023-10-05 21:58:56,173 - Epoch: [163][ 230/ 1236] Overall Loss 0.190129 Objective Loss 0.190129 LR 0.000250 Time 0.024077 +2023-10-05 21:58:56,380 - Epoch: [163][ 240/ 1236] Overall Loss 0.190001 Objective Loss 0.190001 LR 0.000250 Time 0.023935 +2023-10-05 21:58:56,587 - Epoch: [163][ 250/ 1236] Overall Loss 0.189382 Objective Loss 0.189382 LR 0.000250 Time 0.023802 +2023-10-05 21:58:56,793 - Epoch: [163][ 260/ 1236] Overall Loss 0.190452 Objective Loss 0.190452 LR 0.000250 Time 0.023680 +2023-10-05 21:58:57,001 - Epoch: [163][ 270/ 1236] Overall Loss 0.190986 Objective Loss 0.190986 LR 0.000250 Time 0.023572 +2023-10-05 21:58:57,208 - Epoch: [163][ 280/ 1236] Overall Loss 0.190961 Objective Loss 0.190961 LR 0.000250 Time 0.023468 +2023-10-05 21:58:57,415 - Epoch: [163][ 290/ 1236] Overall Loss 0.190862 Objective Loss 0.190862 LR 0.000250 Time 0.023369 +2023-10-05 21:58:57,622 - Epoch: [163][ 300/ 1236] Overall Loss 0.190786 Objective Loss 0.190786 LR 0.000250 Time 0.023278 +2023-10-05 21:58:57,829 - Epoch: [163][ 310/ 1236] Overall Loss 0.190680 Objective Loss 0.190680 LR 0.000250 Time 0.023196 +2023-10-05 21:58:58,037 - Epoch: [163][ 320/ 1236] Overall Loss 0.190249 Objective Loss 0.190249 LR 0.000250 Time 0.023119 +2023-10-05 21:58:58,244 - Epoch: [163][ 330/ 1236] Overall Loss 0.190482 Objective Loss 0.190482 LR 0.000250 Time 0.023047 +2023-10-05 21:58:58,452 - Epoch: [163][ 340/ 1236] Overall Loss 0.190476 Objective Loss 0.190476 LR 0.000250 Time 0.022977 +2023-10-05 21:58:58,660 - Epoch: [163][ 350/ 1236] Overall Loss 0.190533 Objective Loss 0.190533 LR 0.000250 Time 0.022914 +2023-10-05 21:58:58,866 - Epoch: [163][ 360/ 1236] Overall Loss 0.190960 Objective Loss 0.190960 LR 0.000250 Time 0.022849 +2023-10-05 21:58:59,070 - Epoch: [163][ 370/ 1236] Overall Loss 0.191008 Objective Loss 0.191008 LR 0.000250 Time 0.022784 +2023-10-05 21:58:59,275 - Epoch: [163][ 380/ 1236] Overall Loss 0.190893 Objective Loss 0.190893 LR 0.000250 Time 0.022723 +2023-10-05 21:58:59,480 - Epoch: [163][ 390/ 1236] Overall Loss 0.190409 Objective Loss 0.190409 LR 0.000250 Time 0.022663 +2023-10-05 21:58:59,685 - Epoch: [163][ 400/ 1236] Overall Loss 0.190885 Objective Loss 0.190885 LR 0.000250 Time 0.022609 +2023-10-05 21:58:59,889 - Epoch: [163][ 410/ 1236] Overall Loss 0.190356 Objective Loss 0.190356 LR 0.000250 Time 0.022555 +2023-10-05 21:59:00,095 - Epoch: [163][ 420/ 1236] Overall Loss 0.190050 Objective Loss 0.190050 LR 0.000250 Time 0.022506 +2023-10-05 21:59:00,299 - Epoch: [163][ 430/ 1236] Overall Loss 0.190403 Objective Loss 0.190403 LR 0.000250 Time 0.022457 +2023-10-05 21:59:00,504 - Epoch: [163][ 440/ 1236] Overall Loss 0.190234 Objective Loss 0.190234 LR 0.000250 Time 0.022411 +2023-10-05 21:59:00,708 - Epoch: [163][ 450/ 1236] Overall Loss 0.190539 Objective Loss 0.190539 LR 0.000250 Time 0.022367 +2023-10-05 21:59:00,913 - Epoch: [163][ 460/ 1236] Overall Loss 0.190461 Objective Loss 0.190461 LR 0.000250 Time 0.022325 +2023-10-05 21:59:01,118 - Epoch: [163][ 470/ 1236] Overall Loss 0.191141 Objective Loss 0.191141 LR 0.000250 Time 0.022285 +2023-10-05 21:59:01,323 - Epoch: [163][ 480/ 1236] Overall Loss 0.191698 Objective Loss 0.191698 LR 0.000250 Time 0.022247 +2023-10-05 21:59:01,527 - Epoch: [163][ 490/ 1236] Overall Loss 0.191701 Objective Loss 0.191701 LR 0.000250 Time 0.022210 +2023-10-05 21:59:01,732 - Epoch: [163][ 500/ 1236] Overall Loss 0.191576 Objective Loss 0.191576 LR 0.000250 Time 0.022174 +2023-10-05 21:59:01,937 - Epoch: [163][ 510/ 1236] Overall Loss 0.192001 Objective Loss 0.192001 LR 0.000250 Time 0.022140 +2023-10-05 21:59:02,142 - Epoch: [163][ 520/ 1236] Overall Loss 0.192028 Objective Loss 0.192028 LR 0.000250 Time 0.022109 +2023-10-05 21:59:02,345 - Epoch: [163][ 530/ 1236] Overall Loss 0.192169 Objective Loss 0.192169 LR 0.000250 Time 0.022074 +2023-10-05 21:59:02,549 - Epoch: [163][ 540/ 1236] Overall Loss 0.192272 Objective Loss 0.192272 LR 0.000250 Time 0.022042 +2023-10-05 21:59:02,749 - Epoch: [163][ 550/ 1236] Overall Loss 0.192755 Objective Loss 0.192755 LR 0.000250 Time 0.022005 +2023-10-05 21:59:02,953 - Epoch: [163][ 560/ 1236] Overall Loss 0.193102 Objective Loss 0.193102 LR 0.000250 Time 0.021974 +2023-10-05 21:59:03,153 - Epoch: [163][ 570/ 1236] Overall Loss 0.193218 Objective Loss 0.193218 LR 0.000250 Time 0.021939 +2023-10-05 21:59:03,356 - Epoch: [163][ 580/ 1236] Overall Loss 0.193600 Objective Loss 0.193600 LR 0.000250 Time 0.021910 +2023-10-05 21:59:03,556 - Epoch: [163][ 590/ 1236] Overall Loss 0.193453 Objective Loss 0.193453 LR 0.000250 Time 0.021878 +2023-10-05 21:59:03,759 - Epoch: [163][ 600/ 1236] Overall Loss 0.193755 Objective Loss 0.193755 LR 0.000250 Time 0.021851 +2023-10-05 21:59:03,960 - Epoch: [163][ 610/ 1236] Overall Loss 0.193792 Objective Loss 0.193792 LR 0.000250 Time 0.021821 +2023-10-05 21:59:04,163 - Epoch: [163][ 620/ 1236] Overall Loss 0.193831 Objective Loss 0.193831 LR 0.000250 Time 0.021797 +2023-10-05 21:59:04,363 - Epoch: [163][ 630/ 1236] Overall Loss 0.194291 Objective Loss 0.194291 LR 0.000250 Time 0.021768 +2023-10-05 21:59:04,567 - Epoch: [163][ 640/ 1236] Overall Loss 0.194018 Objective Loss 0.194018 LR 0.000250 Time 0.021745 +2023-10-05 21:59:04,767 - Epoch: [163][ 650/ 1236] Overall Loss 0.193769 Objective Loss 0.193769 LR 0.000250 Time 0.021718 +2023-10-05 21:59:04,970 - Epoch: [163][ 660/ 1236] Overall Loss 0.193810 Objective Loss 0.193810 LR 0.000250 Time 0.021696 +2023-10-05 21:59:05,171 - Epoch: [163][ 670/ 1236] Overall Loss 0.193571 Objective Loss 0.193571 LR 0.000250 Time 0.021671 +2023-10-05 21:59:05,374 - Epoch: [163][ 680/ 1236] Overall Loss 0.193453 Objective Loss 0.193453 LR 0.000250 Time 0.021650 +2023-10-05 21:59:05,575 - Epoch: [163][ 690/ 1236] Overall Loss 0.193610 Objective Loss 0.193610 LR 0.000250 Time 0.021627 +2023-10-05 21:59:05,778 - Epoch: [163][ 700/ 1236] Overall Loss 0.194089 Objective Loss 0.194089 LR 0.000250 Time 0.021608 +2023-10-05 21:59:05,978 - Epoch: [163][ 710/ 1236] Overall Loss 0.194262 Objective Loss 0.194262 LR 0.000250 Time 0.021585 +2023-10-05 21:59:06,182 - Epoch: [163][ 720/ 1236] Overall Loss 0.194331 Objective Loss 0.194331 LR 0.000250 Time 0.021568 +2023-10-05 21:59:06,382 - Epoch: [163][ 730/ 1236] Overall Loss 0.194503 Objective Loss 0.194503 LR 0.000250 Time 0.021546 +2023-10-05 21:59:06,585 - Epoch: [163][ 740/ 1236] Overall Loss 0.194617 Objective Loss 0.194617 LR 0.000250 Time 0.021529 +2023-10-05 21:59:06,785 - Epoch: [163][ 750/ 1236] Overall Loss 0.194592 Objective Loss 0.194592 LR 0.000250 Time 0.021509 +2023-10-05 21:59:06,989 - Epoch: [163][ 760/ 1236] Overall Loss 0.194667 Objective Loss 0.194667 LR 0.000250 Time 0.021493 +2023-10-05 21:59:07,189 - Epoch: [163][ 770/ 1236] Overall Loss 0.194740 Objective Loss 0.194740 LR 0.000250 Time 0.021474 +2023-10-05 21:59:07,393 - Epoch: [163][ 780/ 1236] Overall Loss 0.194811 Objective Loss 0.194811 LR 0.000250 Time 0.021459 +2023-10-05 21:59:07,593 - Epoch: [163][ 790/ 1236] Overall Loss 0.194800 Objective Loss 0.194800 LR 0.000250 Time 0.021440 +2023-10-05 21:59:07,796 - Epoch: [163][ 800/ 1236] Overall Loss 0.194794 Objective Loss 0.194794 LR 0.000250 Time 0.021426 +2023-10-05 21:59:07,996 - Epoch: [163][ 810/ 1236] Overall Loss 0.194976 Objective Loss 0.194976 LR 0.000250 Time 0.021408 +2023-10-05 21:59:08,200 - Epoch: [163][ 820/ 1236] Overall Loss 0.194675 Objective Loss 0.194675 LR 0.000250 Time 0.021395 +2023-10-05 21:59:08,401 - Epoch: [163][ 830/ 1236] Overall Loss 0.194405 Objective Loss 0.194405 LR 0.000250 Time 0.021378 +2023-10-05 21:59:08,604 - Epoch: [163][ 840/ 1236] Overall Loss 0.194457 Objective Loss 0.194457 LR 0.000250 Time 0.021365 +2023-10-05 21:59:08,804 - Epoch: [163][ 850/ 1236] Overall Loss 0.194421 Objective Loss 0.194421 LR 0.000250 Time 0.021349 +2023-10-05 21:59:09,008 - Epoch: [163][ 860/ 1236] Overall Loss 0.194362 Objective Loss 0.194362 LR 0.000250 Time 0.021337 +2023-10-05 21:59:09,208 - Epoch: [163][ 870/ 1236] Overall Loss 0.194305 Objective Loss 0.194305 LR 0.000250 Time 0.021322 +2023-10-05 21:59:09,412 - Epoch: [163][ 880/ 1236] Overall Loss 0.194324 Objective Loss 0.194324 LR 0.000250 Time 0.021311 +2023-10-05 21:59:09,612 - Epoch: [163][ 890/ 1236] Overall Loss 0.194260 Objective Loss 0.194260 LR 0.000250 Time 0.021296 +2023-10-05 21:59:09,816 - Epoch: [163][ 900/ 1236] Overall Loss 0.194023 Objective Loss 0.194023 LR 0.000250 Time 0.021285 +2023-10-05 21:59:10,018 - Epoch: [163][ 910/ 1236] Overall Loss 0.194099 Objective Loss 0.194099 LR 0.000250 Time 0.021273 +2023-10-05 21:59:10,221 - Epoch: [163][ 920/ 1236] Overall Loss 0.194176 Objective Loss 0.194176 LR 0.000250 Time 0.021262 +2023-10-05 21:59:10,421 - Epoch: [163][ 930/ 1236] Overall Loss 0.194159 Objective Loss 0.194159 LR 0.000250 Time 0.021248 +2023-10-05 21:59:10,625 - Epoch: [163][ 940/ 1236] Overall Loss 0.194078 Objective Loss 0.194078 LR 0.000250 Time 0.021238 +2023-10-05 21:59:10,825 - Epoch: [163][ 950/ 1236] Overall Loss 0.194168 Objective Loss 0.194168 LR 0.000250 Time 0.021225 +2023-10-05 21:59:11,028 - Epoch: [163][ 960/ 1236] Overall Loss 0.194318 Objective Loss 0.194318 LR 0.000250 Time 0.021215 +2023-10-05 21:59:11,229 - Epoch: [163][ 970/ 1236] Overall Loss 0.194152 Objective Loss 0.194152 LR 0.000250 Time 0.021203 +2023-10-05 21:59:11,432 - Epoch: [163][ 980/ 1236] Overall Loss 0.194199 Objective Loss 0.194199 LR 0.000250 Time 0.021194 +2023-10-05 21:59:11,633 - Epoch: [163][ 990/ 1236] Overall Loss 0.194321 Objective Loss 0.194321 LR 0.000250 Time 0.021182 +2023-10-05 21:59:11,836 - Epoch: [163][ 1000/ 1236] Overall Loss 0.194080 Objective Loss 0.194080 LR 0.000250 Time 0.021173 +2023-10-05 21:59:12,037 - Epoch: [163][ 1010/ 1236] Overall Loss 0.194137 Objective Loss 0.194137 LR 0.000250 Time 0.021162 +2023-10-05 21:59:12,240 - Epoch: [163][ 1020/ 1236] Overall Loss 0.194261 Objective Loss 0.194261 LR 0.000250 Time 0.021153 +2023-10-05 21:59:12,441 - Epoch: [163][ 1030/ 1236] Overall Loss 0.194123 Objective Loss 0.194123 LR 0.000250 Time 0.021143 +2023-10-05 21:59:12,644 - Epoch: [163][ 1040/ 1236] Overall Loss 0.194406 Objective Loss 0.194406 LR 0.000250 Time 0.021135 +2023-10-05 21:59:12,845 - Epoch: [163][ 1050/ 1236] Overall Loss 0.194333 Objective Loss 0.194333 LR 0.000250 Time 0.021124 +2023-10-05 21:59:13,048 - Epoch: [163][ 1060/ 1236] Overall Loss 0.194518 Objective Loss 0.194518 LR 0.000250 Time 0.021116 +2023-10-05 21:59:13,248 - Epoch: [163][ 1070/ 1236] Overall Loss 0.194392 Objective Loss 0.194392 LR 0.000250 Time 0.021106 +2023-10-05 21:59:13,452 - Epoch: [163][ 1080/ 1236] Overall Loss 0.194243 Objective Loss 0.194243 LR 0.000250 Time 0.021098 +2023-10-05 21:59:13,653 - Epoch: [163][ 1090/ 1236] Overall Loss 0.194246 Objective Loss 0.194246 LR 0.000250 Time 0.021089 +2023-10-05 21:59:13,856 - Epoch: [163][ 1100/ 1236] Overall Loss 0.194151 Objective Loss 0.194151 LR 0.000250 Time 0.021081 +2023-10-05 21:59:14,056 - Epoch: [163][ 1110/ 1236] Overall Loss 0.194109 Objective Loss 0.194109 LR 0.000250 Time 0.021071 +2023-10-05 21:59:14,259 - Epoch: [163][ 1120/ 1236] Overall Loss 0.194335 Objective Loss 0.194335 LR 0.000250 Time 0.021064 +2023-10-05 21:59:14,460 - Epoch: [163][ 1130/ 1236] Overall Loss 0.194265 Objective Loss 0.194265 LR 0.000250 Time 0.021055 +2023-10-05 21:59:14,663 - Epoch: [163][ 1140/ 1236] Overall Loss 0.194279 Objective Loss 0.194279 LR 0.000250 Time 0.021048 +2023-10-05 21:59:14,864 - Epoch: [163][ 1150/ 1236] Overall Loss 0.194236 Objective Loss 0.194236 LR 0.000250 Time 0.021040 +2023-10-05 21:59:15,066 - Epoch: [163][ 1160/ 1236] Overall Loss 0.194123 Objective Loss 0.194123 LR 0.000250 Time 0.021033 +2023-10-05 21:59:15,267 - Epoch: [163][ 1170/ 1236] Overall Loss 0.193892 Objective Loss 0.193892 LR 0.000250 Time 0.021024 +2023-10-05 21:59:15,470 - Epoch: [163][ 1180/ 1236] Overall Loss 0.194238 Objective Loss 0.194238 LR 0.000250 Time 0.021018 +2023-10-05 21:59:15,671 - Epoch: [163][ 1190/ 1236] Overall Loss 0.194452 Objective Loss 0.194452 LR 0.000250 Time 0.021010 +2023-10-05 21:59:15,874 - Epoch: [163][ 1200/ 1236] Overall Loss 0.194577 Objective Loss 0.194577 LR 0.000250 Time 0.021003 +2023-10-05 21:59:16,075 - Epoch: [163][ 1210/ 1236] Overall Loss 0.194568 Objective Loss 0.194568 LR 0.000250 Time 0.020996 +2023-10-05 21:59:16,278 - Epoch: [163][ 1220/ 1236] Overall Loss 0.194781 Objective Loss 0.194781 LR 0.000250 Time 0.020990 +2023-10-05 21:59:16,534 - Epoch: [163][ 1230/ 1236] Overall Loss 0.194864 Objective Loss 0.194864 LR 0.000250 Time 0.021027 +2023-10-05 21:59:16,653 - Epoch: [163][ 1236/ 1236] Overall Loss 0.194870 Objective Loss 0.194870 Top1 86.761711 Top5 98.574338 LR 0.000250 Time 0.021021 +2023-10-05 21:59:16,783 - --- validate (epoch=163)----------- +2023-10-05 21:59:16,783 - 29943 samples (256 per mini-batch) +2023-10-05 21:59:17,242 - Epoch: [163][ 10/ 117] Loss 0.286196 Top1 86.406250 Top5 98.203125 +2023-10-05 21:59:17,393 - Epoch: [163][ 20/ 117] Loss 0.310691 Top1 85.175781 Top5 98.222656 +2023-10-05 21:59:17,542 - Epoch: [163][ 30/ 117] Loss 0.309052 Top1 85.273438 Top5 98.255208 +2023-10-05 21:59:17,692 - Epoch: [163][ 40/ 117] Loss 0.306997 Top1 85.322266 Top5 98.134766 +2023-10-05 21:59:17,839 - Epoch: [163][ 50/ 117] Loss 0.309949 Top1 85.218750 Top5 98.148438 +2023-10-05 21:59:17,988 - Epoch: [163][ 60/ 117] Loss 0.309993 Top1 85.149740 Top5 98.059896 +2023-10-05 21:59:18,136 - Epoch: [163][ 70/ 117] Loss 0.312799 Top1 85.161830 Top5 98.074777 +2023-10-05 21:59:18,286 - Epoch: [163][ 80/ 117] Loss 0.306346 Top1 85.307617 Top5 98.100586 +2023-10-05 21:59:18,434 - Epoch: [163][ 90/ 117] Loss 0.309102 Top1 85.269097 Top5 98.138021 +2023-10-05 21:59:18,583 - Epoch: [163][ 100/ 117] Loss 0.311946 Top1 85.210938 Top5 98.148438 +2023-10-05 21:59:18,738 - Epoch: [163][ 110/ 117] Loss 0.306312 Top1 85.348011 Top5 98.164062 +2023-10-05 21:59:18,824 - Epoch: [163][ 117/ 117] Loss 0.307797 Top1 85.365528 Top5 98.126440 +2023-10-05 21:59:18,955 - ==> Top1: 85.366 Top5: 98.126 Loss: 0.308 + +2023-10-05 21:59:18,956 - ==> Confusion: +[[ 927 2 3 2 11 1 0 0 8 63 1 0 1 2 6 2 5 0 1 0 15] + [ 0 1060 2 0 9 22 1 11 1 0 1 2 0 0 1 3 3 0 5 2 8] + [ 6 1 956 15 3 0 26 6 0 2 2 2 11 2 0 5 1 1 5 1 11] + [ 3 1 10 981 2 4 1 2 0 1 8 0 4 0 25 3 1 8 16 0 19] + [ 23 7 0 0 974 3 0 1 1 8 0 2 1 1 6 2 9 2 0 2 8] + [ 4 30 0 0 6 996 0 16 3 1 5 6 0 15 6 1 6 0 4 4 13] + [ 0 6 22 0 0 0 1125 7 0 0 1 3 0 0 1 10 0 1 1 8 6] + [ 3 22 14 0 2 33 3 1066 1 3 1 11 0 2 0 2 2 0 37 5 11] + [ 17 2 0 0 0 2 0 1 982 36 11 2 1 9 14 6 0 0 5 0 1] + [ 108 0 1 1 4 2 0 0 29 926 0 0 0 19 6 10 1 1 1 0 10] + [ 2 6 9 3 2 0 2 6 13 1 975 1 1 12 3 0 2 0 5 2 8] + [ 0 0 2 0 1 13 0 0 0 1 0 967 14 6 0 5 2 18 0 3 3] + [ 1 1 1 4 0 2 0 2 1 0 1 36 977 1 1 9 2 16 1 3 9] + [ 1 0 1 0 2 5 0 0 9 12 7 6 3 1059 4 1 1 0 0 0 8] + [ 15 0 4 5 5 0 0 0 21 2 2 0 1 2 1024 0 1 2 7 0 10] + [ 0 4 0 0 7 0 1 0 0 0 0 6 6 1 1 1080 12 8 0 6 2] + [ 0 13 1 0 5 4 0 1 2 0 0 5 1 0 2 9 1107 0 0 3 8] + [ 0 0 0 0 1 0 2 0 0 0 0 2 15 2 0 11 0 998 1 0 6] + [ 1 6 6 18 1 1 0 17 2 0 4 1 1 0 12 0 1 0 990 1 6] + [ 1 3 2 2 2 5 9 5 0 0 2 18 2 1 0 7 10 2 2 1070 9] + [ 112 184 115 63 97 125 34 83 94 76 161 114 321 271 148 63 160 57 139 167 5321]] + +2023-10-05 21:59:18,957 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:59:18,957 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:59:18,963 - + +2023-10-05 21:59:18,963 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:59:19,981 - Epoch: [164][ 10/ 1236] Overall Loss 0.189367 Objective Loss 0.189367 LR 0.000250 Time 0.101701 +2023-10-05 21:59:20,182 - Epoch: [164][ 20/ 1236] Overall Loss 0.185740 Objective Loss 0.185740 LR 0.000250 Time 0.060870 +2023-10-05 21:59:20,381 - Epoch: [164][ 30/ 1236] Overall Loss 0.184464 Objective Loss 0.184464 LR 0.000250 Time 0.047208 +2023-10-05 21:59:20,582 - Epoch: [164][ 40/ 1236] Overall Loss 0.187930 Objective Loss 0.187930 LR 0.000250 Time 0.040426 +2023-10-05 21:59:20,781 - Epoch: [164][ 50/ 1236] Overall Loss 0.185734 Objective Loss 0.185734 LR 0.000250 Time 0.036322 +2023-10-05 21:59:20,983 - Epoch: [164][ 60/ 1236] Overall Loss 0.189678 Objective Loss 0.189678 LR 0.000250 Time 0.033615 +2023-10-05 21:59:21,183 - Epoch: [164][ 70/ 1236] Overall Loss 0.187613 Objective Loss 0.187613 LR 0.000250 Time 0.031665 +2023-10-05 21:59:21,384 - Epoch: [164][ 80/ 1236] Overall Loss 0.189013 Objective Loss 0.189013 LR 0.000250 Time 0.030226 +2023-10-05 21:59:21,584 - Epoch: [164][ 90/ 1236] Overall Loss 0.190372 Objective Loss 0.190372 LR 0.000250 Time 0.029076 +2023-10-05 21:59:21,785 - Epoch: [164][ 100/ 1236] Overall Loss 0.193030 Objective Loss 0.193030 LR 0.000250 Time 0.028178 +2023-10-05 21:59:21,984 - Epoch: [164][ 110/ 1236] Overall Loss 0.193623 Objective Loss 0.193623 LR 0.000250 Time 0.027427 +2023-10-05 21:59:22,186 - Epoch: [164][ 120/ 1236] Overall Loss 0.192040 Objective Loss 0.192040 LR 0.000250 Time 0.026815 +2023-10-05 21:59:22,385 - Epoch: [164][ 130/ 1236] Overall Loss 0.192736 Objective Loss 0.192736 LR 0.000250 Time 0.026283 +2023-10-05 21:59:22,586 - Epoch: [164][ 140/ 1236] Overall Loss 0.192208 Objective Loss 0.192208 LR 0.000250 Time 0.025839 +2023-10-05 21:59:22,785 - Epoch: [164][ 150/ 1236] Overall Loss 0.192494 Objective Loss 0.192494 LR 0.000250 Time 0.025444 +2023-10-05 21:59:22,986 - Epoch: [164][ 160/ 1236] Overall Loss 0.193477 Objective Loss 0.193477 LR 0.000250 Time 0.025107 +2023-10-05 21:59:23,186 - Epoch: [164][ 170/ 1236] Overall Loss 0.192621 Objective Loss 0.192621 LR 0.000250 Time 0.024803 +2023-10-05 21:59:23,387 - Epoch: [164][ 180/ 1236] Overall Loss 0.192110 Objective Loss 0.192110 LR 0.000250 Time 0.024540 +2023-10-05 21:59:23,587 - Epoch: [164][ 190/ 1236] Overall Loss 0.190985 Objective Loss 0.190985 LR 0.000250 Time 0.024296 +2023-10-05 21:59:23,788 - Epoch: [164][ 200/ 1236] Overall Loss 0.190814 Objective Loss 0.190814 LR 0.000250 Time 0.024087 +2023-10-05 21:59:23,988 - Epoch: [164][ 210/ 1236] Overall Loss 0.190839 Objective Loss 0.190839 LR 0.000250 Time 0.023888 +2023-10-05 21:59:24,189 - Epoch: [164][ 220/ 1236] Overall Loss 0.191485 Objective Loss 0.191485 LR 0.000250 Time 0.023718 +2023-10-05 21:59:24,389 - Epoch: [164][ 230/ 1236] Overall Loss 0.191248 Objective Loss 0.191248 LR 0.000250 Time 0.023552 +2023-10-05 21:59:24,590 - Epoch: [164][ 240/ 1236] Overall Loss 0.191629 Objective Loss 0.191629 LR 0.000250 Time 0.023408 +2023-10-05 21:59:24,790 - Epoch: [164][ 250/ 1236] Overall Loss 0.192006 Objective Loss 0.192006 LR 0.000250 Time 0.023268 +2023-10-05 21:59:24,991 - Epoch: [164][ 260/ 1236] Overall Loss 0.191667 Objective Loss 0.191667 LR 0.000250 Time 0.023146 +2023-10-05 21:59:25,191 - Epoch: [164][ 270/ 1236] Overall Loss 0.191973 Objective Loss 0.191973 LR 0.000250 Time 0.023028 +2023-10-05 21:59:25,391 - Epoch: [164][ 280/ 1236] Overall Loss 0.191922 Objective Loss 0.191922 LR 0.000250 Time 0.022921 +2023-10-05 21:59:25,591 - Epoch: [164][ 290/ 1236] Overall Loss 0.191982 Objective Loss 0.191982 LR 0.000250 Time 0.022818 +2023-10-05 21:59:25,792 - Epoch: [164][ 300/ 1236] Overall Loss 0.192170 Objective Loss 0.192170 LR 0.000250 Time 0.022727 +2023-10-05 21:59:25,992 - Epoch: [164][ 310/ 1236] Overall Loss 0.191167 Objective Loss 0.191167 LR 0.000250 Time 0.022635 +2023-10-05 21:59:26,193 - Epoch: [164][ 320/ 1236] Overall Loss 0.190980 Objective Loss 0.190980 LR 0.000250 Time 0.022555 +2023-10-05 21:59:26,392 - Epoch: [164][ 330/ 1236] Overall Loss 0.190939 Objective Loss 0.190939 LR 0.000250 Time 0.022475 +2023-10-05 21:59:26,594 - Epoch: [164][ 340/ 1236] Overall Loss 0.190743 Objective Loss 0.190743 LR 0.000250 Time 0.022406 +2023-10-05 21:59:26,793 - Epoch: [164][ 350/ 1236] Overall Loss 0.191019 Objective Loss 0.191019 LR 0.000250 Time 0.022335 +2023-10-05 21:59:26,995 - Epoch: [164][ 360/ 1236] Overall Loss 0.190970 Objective Loss 0.190970 LR 0.000250 Time 0.022274 +2023-10-05 21:59:27,195 - Epoch: [164][ 370/ 1236] Overall Loss 0.191452 Objective Loss 0.191452 LR 0.000250 Time 0.022212 +2023-10-05 21:59:27,398 - Epoch: [164][ 380/ 1236] Overall Loss 0.190768 Objective Loss 0.190768 LR 0.000250 Time 0.022160 +2023-10-05 21:59:27,599 - Epoch: [164][ 390/ 1236] Overall Loss 0.190727 Objective Loss 0.190727 LR 0.000250 Time 0.022107 +2023-10-05 21:59:27,802 - Epoch: [164][ 400/ 1236] Overall Loss 0.190980 Objective Loss 0.190980 LR 0.000250 Time 0.022060 +2023-10-05 21:59:28,003 - Epoch: [164][ 410/ 1236] Overall Loss 0.191338 Objective Loss 0.191338 LR 0.000250 Time 0.022013 +2023-10-05 21:59:28,206 - Epoch: [164][ 420/ 1236] Overall Loss 0.191257 Objective Loss 0.191257 LR 0.000250 Time 0.021970 +2023-10-05 21:59:28,407 - Epoch: [164][ 430/ 1236] Overall Loss 0.191543 Objective Loss 0.191543 LR 0.000250 Time 0.021927 +2023-10-05 21:59:28,610 - Epoch: [164][ 440/ 1236] Overall Loss 0.192049 Objective Loss 0.192049 LR 0.000250 Time 0.021888 +2023-10-05 21:59:28,811 - Epoch: [164][ 450/ 1236] Overall Loss 0.192379 Objective Loss 0.192379 LR 0.000250 Time 0.021847 +2023-10-05 21:59:29,014 - Epoch: [164][ 460/ 1236] Overall Loss 0.192450 Objective Loss 0.192450 LR 0.000250 Time 0.021812 +2023-10-05 21:59:29,215 - Epoch: [164][ 470/ 1236] Overall Loss 0.192667 Objective Loss 0.192667 LR 0.000250 Time 0.021776 +2023-10-05 21:59:29,418 - Epoch: [164][ 480/ 1236] Overall Loss 0.192316 Objective Loss 0.192316 LR 0.000250 Time 0.021744 +2023-10-05 21:59:29,619 - Epoch: [164][ 490/ 1236] Overall Loss 0.192085 Objective Loss 0.192085 LR 0.000250 Time 0.021710 +2023-10-05 21:59:29,822 - Epoch: [164][ 500/ 1236] Overall Loss 0.193135 Objective Loss 0.193135 LR 0.000250 Time 0.021681 +2023-10-05 21:59:30,023 - Epoch: [164][ 510/ 1236] Overall Loss 0.193206 Objective Loss 0.193206 LR 0.000250 Time 0.021649 +2023-10-05 21:59:30,226 - Epoch: [164][ 520/ 1236] Overall Loss 0.193049 Objective Loss 0.193049 LR 0.000250 Time 0.021622 +2023-10-05 21:59:30,427 - Epoch: [164][ 530/ 1236] Overall Loss 0.193063 Objective Loss 0.193063 LR 0.000250 Time 0.021593 +2023-10-05 21:59:30,630 - Epoch: [164][ 540/ 1236] Overall Loss 0.193134 Objective Loss 0.193134 LR 0.000250 Time 0.021568 +2023-10-05 21:59:30,831 - Epoch: [164][ 550/ 1236] Overall Loss 0.193452 Objective Loss 0.193452 LR 0.000250 Time 0.021541 +2023-10-05 21:59:31,034 - Epoch: [164][ 560/ 1236] Overall Loss 0.192986 Objective Loss 0.192986 LR 0.000250 Time 0.021518 +2023-10-05 21:59:31,235 - Epoch: [164][ 570/ 1236] Overall Loss 0.192928 Objective Loss 0.192928 LR 0.000250 Time 0.021494 +2023-10-05 21:59:31,438 - Epoch: [164][ 580/ 1236] Overall Loss 0.192711 Objective Loss 0.192711 LR 0.000250 Time 0.021472 +2023-10-05 21:59:31,639 - Epoch: [164][ 590/ 1236] Overall Loss 0.192511 Objective Loss 0.192511 LR 0.000250 Time 0.021449 +2023-10-05 21:59:31,842 - Epoch: [164][ 600/ 1236] Overall Loss 0.192751 Objective Loss 0.192751 LR 0.000250 Time 0.021428 +2023-10-05 21:59:32,043 - Epoch: [164][ 610/ 1236] Overall Loss 0.192919 Objective Loss 0.192919 LR 0.000250 Time 0.021406 +2023-10-05 21:59:32,246 - Epoch: [164][ 620/ 1236] Overall Loss 0.192919 Objective Loss 0.192919 LR 0.000250 Time 0.021388 +2023-10-05 21:59:32,447 - Epoch: [164][ 630/ 1236] Overall Loss 0.192808 Objective Loss 0.192808 LR 0.000250 Time 0.021367 +2023-10-05 21:59:32,650 - Epoch: [164][ 640/ 1236] Overall Loss 0.192807 Objective Loss 0.192807 LR 0.000250 Time 0.021350 +2023-10-05 21:59:32,851 - Epoch: [164][ 650/ 1236] Overall Loss 0.192857 Objective Loss 0.192857 LR 0.000250 Time 0.021330 +2023-10-05 21:59:33,054 - Epoch: [164][ 660/ 1236] Overall Loss 0.193358 Objective Loss 0.193358 LR 0.000250 Time 0.021314 +2023-10-05 21:59:33,255 - Epoch: [164][ 670/ 1236] Overall Loss 0.193404 Objective Loss 0.193404 LR 0.000250 Time 0.021296 +2023-10-05 21:59:33,458 - Epoch: [164][ 680/ 1236] Overall Loss 0.193892 Objective Loss 0.193892 LR 0.000250 Time 0.021280 +2023-10-05 21:59:33,659 - Epoch: [164][ 690/ 1236] Overall Loss 0.193870 Objective Loss 0.193870 LR 0.000250 Time 0.021263 +2023-10-05 21:59:33,862 - Epoch: [164][ 700/ 1236] Overall Loss 0.194195 Objective Loss 0.194195 LR 0.000250 Time 0.021249 +2023-10-05 21:59:34,063 - Epoch: [164][ 710/ 1236] Overall Loss 0.194084 Objective Loss 0.194084 LR 0.000250 Time 0.021233 +2023-10-05 21:59:34,266 - Epoch: [164][ 720/ 1236] Overall Loss 0.194265 Objective Loss 0.194265 LR 0.000250 Time 0.021219 +2023-10-05 21:59:34,467 - Epoch: [164][ 730/ 1236] Overall Loss 0.194311 Objective Loss 0.194311 LR 0.000250 Time 0.021203 +2023-10-05 21:59:34,670 - Epoch: [164][ 740/ 1236] Overall Loss 0.194544 Objective Loss 0.194544 LR 0.000250 Time 0.021190 +2023-10-05 21:59:34,871 - Epoch: [164][ 750/ 1236] Overall Loss 0.194572 Objective Loss 0.194572 LR 0.000250 Time 0.021176 +2023-10-05 21:59:35,074 - Epoch: [164][ 760/ 1236] Overall Loss 0.194646 Objective Loss 0.194646 LR 0.000250 Time 0.021163 +2023-10-05 21:59:35,275 - Epoch: [164][ 770/ 1236] Overall Loss 0.194902 Objective Loss 0.194902 LR 0.000250 Time 0.021149 +2023-10-05 21:59:35,478 - Epoch: [164][ 780/ 1236] Overall Loss 0.195215 Objective Loss 0.195215 LR 0.000250 Time 0.021138 +2023-10-05 21:59:35,679 - Epoch: [164][ 790/ 1236] Overall Loss 0.195266 Objective Loss 0.195266 LR 0.000250 Time 0.021124 +2023-10-05 21:59:35,882 - Epoch: [164][ 800/ 1236] Overall Loss 0.195032 Objective Loss 0.195032 LR 0.000250 Time 0.021113 +2023-10-05 21:59:36,083 - Epoch: [164][ 810/ 1236] Overall Loss 0.194843 Objective Loss 0.194843 LR 0.000250 Time 0.021101 +2023-10-05 21:59:36,286 - Epoch: [164][ 820/ 1236] Overall Loss 0.194813 Objective Loss 0.194813 LR 0.000250 Time 0.021090 +2023-10-05 21:59:36,487 - Epoch: [164][ 830/ 1236] Overall Loss 0.194831 Objective Loss 0.194831 LR 0.000250 Time 0.021078 +2023-10-05 21:59:36,690 - Epoch: [164][ 840/ 1236] Overall Loss 0.195154 Objective Loss 0.195154 LR 0.000250 Time 0.021068 +2023-10-05 21:59:36,891 - Epoch: [164][ 850/ 1236] Overall Loss 0.195192 Objective Loss 0.195192 LR 0.000250 Time 0.021057 +2023-10-05 21:59:37,094 - Epoch: [164][ 860/ 1236] Overall Loss 0.195331 Objective Loss 0.195331 LR 0.000250 Time 0.021047 +2023-10-05 21:59:37,295 - Epoch: [164][ 870/ 1236] Overall Loss 0.195243 Objective Loss 0.195243 LR 0.000250 Time 0.021036 +2023-10-05 21:59:37,498 - Epoch: [164][ 880/ 1236] Overall Loss 0.195222 Objective Loss 0.195222 LR 0.000250 Time 0.021027 +2023-10-05 21:59:37,699 - Epoch: [164][ 890/ 1236] Overall Loss 0.195317 Objective Loss 0.195317 LR 0.000250 Time 0.021017 +2023-10-05 21:59:37,902 - Epoch: [164][ 900/ 1236] Overall Loss 0.195447 Objective Loss 0.195447 LR 0.000250 Time 0.021008 +2023-10-05 21:59:38,103 - Epoch: [164][ 910/ 1236] Overall Loss 0.195405 Objective Loss 0.195405 LR 0.000250 Time 0.020998 +2023-10-05 21:59:38,306 - Epoch: [164][ 920/ 1236] Overall Loss 0.195349 Objective Loss 0.195349 LR 0.000250 Time 0.020990 +2023-10-05 21:59:38,508 - Epoch: [164][ 930/ 1236] Overall Loss 0.195227 Objective Loss 0.195227 LR 0.000250 Time 0.020980 +2023-10-05 21:59:38,711 - Epoch: [164][ 940/ 1236] Overall Loss 0.195104 Objective Loss 0.195104 LR 0.000250 Time 0.020973 +2023-10-05 21:59:38,912 - Epoch: [164][ 950/ 1236] Overall Loss 0.195021 Objective Loss 0.195021 LR 0.000250 Time 0.020963 +2023-10-05 21:59:39,115 - Epoch: [164][ 960/ 1236] Overall Loss 0.195033 Objective Loss 0.195033 LR 0.000250 Time 0.020956 +2023-10-05 21:59:39,316 - Epoch: [164][ 970/ 1236] Overall Loss 0.195139 Objective Loss 0.195139 LR 0.000250 Time 0.020947 +2023-10-05 21:59:39,519 - Epoch: [164][ 980/ 1236] Overall Loss 0.195173 Objective Loss 0.195173 LR 0.000250 Time 0.020940 +2023-10-05 21:59:39,720 - Epoch: [164][ 990/ 1236] Overall Loss 0.195384 Objective Loss 0.195384 LR 0.000250 Time 0.020932 +2023-10-05 21:59:39,923 - Epoch: [164][ 1000/ 1236] Overall Loss 0.195358 Objective Loss 0.195358 LR 0.000250 Time 0.020925 +2023-10-05 21:59:40,125 - Epoch: [164][ 1010/ 1236] Overall Loss 0.195283 Objective Loss 0.195283 LR 0.000250 Time 0.020917 +2023-10-05 21:59:40,328 - Epoch: [164][ 1020/ 1236] Overall Loss 0.195140 Objective Loss 0.195140 LR 0.000250 Time 0.020910 +2023-10-05 21:59:40,529 - Epoch: [164][ 1030/ 1236] Overall Loss 0.195034 Objective Loss 0.195034 LR 0.000250 Time 0.020902 +2023-10-05 21:59:40,732 - Epoch: [164][ 1040/ 1236] Overall Loss 0.194914 Objective Loss 0.194914 LR 0.000250 Time 0.020896 +2023-10-05 21:59:40,933 - Epoch: [164][ 1050/ 1236] Overall Loss 0.195002 Objective Loss 0.195002 LR 0.000250 Time 0.020888 +2023-10-05 21:59:41,135 - Epoch: [164][ 1060/ 1236] Overall Loss 0.194961 Objective Loss 0.194961 LR 0.000250 Time 0.020882 +2023-10-05 21:59:41,337 - Epoch: [164][ 1070/ 1236] Overall Loss 0.194878 Objective Loss 0.194878 LR 0.000250 Time 0.020875 +2023-10-05 21:59:41,540 - Epoch: [164][ 1080/ 1236] Overall Loss 0.194837 Objective Loss 0.194837 LR 0.000250 Time 0.020869 +2023-10-05 21:59:41,741 - Epoch: [164][ 1090/ 1236] Overall Loss 0.194840 Objective Loss 0.194840 LR 0.000250 Time 0.020862 +2023-10-05 21:59:41,943 - Epoch: [164][ 1100/ 1236] Overall Loss 0.194789 Objective Loss 0.194789 LR 0.000250 Time 0.020856 +2023-10-05 21:59:42,145 - Epoch: [164][ 1110/ 1236] Overall Loss 0.194759 Objective Loss 0.194759 LR 0.000250 Time 0.020849 +2023-10-05 21:59:42,348 - Epoch: [164][ 1120/ 1236] Overall Loss 0.194878 Objective Loss 0.194878 LR 0.000250 Time 0.020844 +2023-10-05 21:59:42,550 - Epoch: [164][ 1130/ 1236] Overall Loss 0.194857 Objective Loss 0.194857 LR 0.000250 Time 0.020838 +2023-10-05 21:59:42,753 - Epoch: [164][ 1140/ 1236] Overall Loss 0.194979 Objective Loss 0.194979 LR 0.000250 Time 0.020833 +2023-10-05 21:59:42,954 - Epoch: [164][ 1150/ 1236] Overall Loss 0.194922 Objective Loss 0.194922 LR 0.000250 Time 0.020826 +2023-10-05 21:59:43,157 - Epoch: [164][ 1160/ 1236] Overall Loss 0.194826 Objective Loss 0.194826 LR 0.000250 Time 0.020821 +2023-10-05 21:59:43,358 - Epoch: [164][ 1170/ 1236] Overall Loss 0.194853 Objective Loss 0.194853 LR 0.000250 Time 0.020815 +2023-10-05 21:59:43,561 - Epoch: [164][ 1180/ 1236] Overall Loss 0.194834 Objective Loss 0.194834 LR 0.000250 Time 0.020811 +2023-10-05 21:59:43,763 - Epoch: [164][ 1190/ 1236] Overall Loss 0.194726 Objective Loss 0.194726 LR 0.000250 Time 0.020805 +2023-10-05 21:59:43,965 - Epoch: [164][ 1200/ 1236] Overall Loss 0.194642 Objective Loss 0.194642 LR 0.000250 Time 0.020800 +2023-10-05 21:59:44,167 - Epoch: [164][ 1210/ 1236] Overall Loss 0.194334 Objective Loss 0.194334 LR 0.000250 Time 0.020794 +2023-10-05 21:59:44,370 - Epoch: [164][ 1220/ 1236] Overall Loss 0.194531 Objective Loss 0.194531 LR 0.000250 Time 0.020790 +2023-10-05 21:59:44,624 - Epoch: [164][ 1230/ 1236] Overall Loss 0.194574 Objective Loss 0.194574 LR 0.000250 Time 0.020828 +2023-10-05 21:59:44,743 - Epoch: [164][ 1236/ 1236] Overall Loss 0.194531 Objective Loss 0.194531 Top1 89.002037 Top5 97.759674 LR 0.000250 Time 0.020822 +2023-10-05 21:59:44,868 - --- validate (epoch=164)----------- +2023-10-05 21:59:44,868 - 29943 samples (256 per mini-batch) +2023-10-05 21:59:45,327 - Epoch: [164][ 10/ 117] Loss 0.312087 Top1 85.312500 Top5 98.320312 +2023-10-05 21:59:45,479 - Epoch: [164][ 20/ 117] Loss 0.292794 Top1 86.054688 Top5 98.417969 +2023-10-05 21:59:45,627 - Epoch: [164][ 30/ 117] Loss 0.298636 Top1 85.703125 Top5 98.346354 +2023-10-05 21:59:45,777 - Epoch: [164][ 40/ 117] Loss 0.299297 Top1 85.556641 Top5 98.261719 +2023-10-05 21:59:45,925 - Epoch: [164][ 50/ 117] Loss 0.304739 Top1 85.351562 Top5 98.195312 +2023-10-05 21:59:46,075 - Epoch: [164][ 60/ 117] Loss 0.310979 Top1 85.279948 Top5 98.138021 +2023-10-05 21:59:46,223 - Epoch: [164][ 70/ 117] Loss 0.311621 Top1 85.239955 Top5 98.102679 +2023-10-05 21:59:46,373 - Epoch: [164][ 80/ 117] Loss 0.310325 Top1 85.278320 Top5 98.110352 +2023-10-05 21:59:46,522 - Epoch: [164][ 90/ 117] Loss 0.310058 Top1 85.208333 Top5 98.042535 +2023-10-05 21:59:46,671 - Epoch: [164][ 100/ 117] Loss 0.309005 Top1 85.320312 Top5 98.035156 +2023-10-05 21:59:46,824 - Epoch: [164][ 110/ 117] Loss 0.310404 Top1 85.269886 Top5 98.050426 +2023-10-05 21:59:46,910 - Epoch: [164][ 117/ 117] Loss 0.308416 Top1 85.228601 Top5 98.096383 +2023-10-05 21:59:47,028 - ==> Top1: 85.229 Top5: 98.096 Loss: 0.308 + +2023-10-05 21:59:47,029 - ==> Confusion: +[[ 910 3 5 2 9 2 2 0 3 83 1 0 1 1 6 2 2 2 0 0 16] + [ 0 1066 1 0 10 21 2 14 1 0 0 2 0 0 1 2 1 0 3 0 7] + [ 3 2 964 14 2 0 25 7 0 2 5 0 7 1 1 3 0 2 6 5 7] + [ 2 1 11 973 0 2 2 2 1 1 5 1 5 2 28 4 0 7 24 1 17] + [ 16 9 2 0 977 1 0 1 0 12 0 2 1 0 9 2 8 3 0 2 5] + [ 4 29 0 1 5 1001 2 20 1 1 3 9 0 10 6 1 3 0 4 3 13] + [ 0 5 16 0 1 0 1133 8 0 0 1 2 1 0 1 7 0 0 2 7 7] + [ 4 22 12 0 2 30 5 1072 0 3 2 8 0 1 1 3 0 0 36 6 11] + [ 18 2 0 0 2 2 0 2 971 42 11 0 0 10 17 4 0 3 3 0 2] + [ 89 0 1 0 4 3 0 0 22 959 0 2 1 18 6 6 0 2 0 0 6] + [ 3 6 6 5 2 1 8 8 10 2 970 2 1 8 3 1 1 0 3 2 11] + [ 1 0 1 0 0 11 0 1 0 1 0 963 16 7 0 5 1 17 0 6 5] + [ 1 2 2 5 1 1 1 2 1 0 1 34 979 1 1 3 2 19 2 3 7] + [ 2 1 2 1 3 6 0 2 10 14 7 6 1 1042 3 2 0 1 0 1 15] + [ 10 2 3 9 8 0 0 0 24 0 2 0 2 3 1013 0 1 2 11 0 11] + [ 1 3 1 0 1 0 1 0 0 0 0 6 5 1 1 1072 15 15 1 8 3] + [ 0 17 1 0 6 3 0 0 1 0 0 4 0 1 3 10 1098 0 1 4 12] + [ 0 0 0 2 1 0 2 0 1 0 0 3 13 1 1 3 0 1007 0 0 4] + [ 1 9 5 15 1 0 0 19 1 1 2 0 0 0 11 0 0 0 997 1 5] + [ 1 2 3 2 2 6 13 8 1 0 0 12 4 0 0 5 7 2 4 1072 8] + [ 106 196 129 53 95 132 49 89 105 75 161 110 342 237 162 55 113 79 158 178 5281]] + +2023-10-05 21:59:47,030 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 21:59:47,030 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 21:59:47,036 - + +2023-10-05 21:59:47,036 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 21:59:48,042 - Epoch: [165][ 10/ 1236] Overall Loss 0.191674 Objective Loss 0.191674 LR 0.000250 Time 0.100541 +2023-10-05 21:59:48,244 - Epoch: [165][ 20/ 1236] Overall Loss 0.183729 Objective Loss 0.183729 LR 0.000250 Time 0.060336 +2023-10-05 21:59:48,444 - Epoch: [165][ 30/ 1236] Overall Loss 0.188996 Objective Loss 0.188996 LR 0.000250 Time 0.046880 +2023-10-05 21:59:48,646 - Epoch: [165][ 40/ 1236] Overall Loss 0.190591 Objective Loss 0.190591 LR 0.000250 Time 0.040196 +2023-10-05 21:59:48,845 - Epoch: [165][ 50/ 1236] Overall Loss 0.193117 Objective Loss 0.193117 LR 0.000250 Time 0.036144 +2023-10-05 21:59:49,047 - Epoch: [165][ 60/ 1236] Overall Loss 0.191000 Objective Loss 0.191000 LR 0.000250 Time 0.033477 +2023-10-05 21:59:49,247 - Epoch: [165][ 70/ 1236] Overall Loss 0.188405 Objective Loss 0.188405 LR 0.000250 Time 0.031549 +2023-10-05 21:59:49,449 - Epoch: [165][ 80/ 1236] Overall Loss 0.189772 Objective Loss 0.189772 LR 0.000250 Time 0.030126 +2023-10-05 21:59:49,649 - Epoch: [165][ 90/ 1236] Overall Loss 0.189637 Objective Loss 0.189637 LR 0.000250 Time 0.028998 +2023-10-05 21:59:49,849 - Epoch: [165][ 100/ 1236] Overall Loss 0.188785 Objective Loss 0.188785 LR 0.000250 Time 0.028093 +2023-10-05 21:59:50,048 - Epoch: [165][ 110/ 1236] Overall Loss 0.190106 Objective Loss 0.190106 LR 0.000250 Time 0.027345 +2023-10-05 21:59:50,249 - Epoch: [165][ 120/ 1236] Overall Loss 0.189016 Objective Loss 0.189016 LR 0.000250 Time 0.026736 +2023-10-05 21:59:50,447 - Epoch: [165][ 130/ 1236] Overall Loss 0.191218 Objective Loss 0.191218 LR 0.000250 Time 0.026205 +2023-10-05 21:59:50,648 - Epoch: [165][ 140/ 1236] Overall Loss 0.192227 Objective Loss 0.192227 LR 0.000250 Time 0.025764 +2023-10-05 21:59:50,847 - Epoch: [165][ 150/ 1236] Overall Loss 0.191498 Objective Loss 0.191498 LR 0.000250 Time 0.025370 +2023-10-05 21:59:51,046 - Epoch: [165][ 160/ 1236] Overall Loss 0.190637 Objective Loss 0.190637 LR 0.000250 Time 0.025028 +2023-10-05 21:59:51,244 - Epoch: [165][ 170/ 1236] Overall Loss 0.190826 Objective Loss 0.190826 LR 0.000250 Time 0.024715 +2023-10-05 21:59:51,445 - Epoch: [165][ 180/ 1236] Overall Loss 0.190149 Objective Loss 0.190149 LR 0.000250 Time 0.024456 +2023-10-05 21:59:51,644 - Epoch: [165][ 190/ 1236] Overall Loss 0.189478 Objective Loss 0.189478 LR 0.000250 Time 0.024215 +2023-10-05 21:59:51,844 - Epoch: [165][ 200/ 1236] Overall Loss 0.189283 Objective Loss 0.189283 LR 0.000250 Time 0.024004 +2023-10-05 21:59:52,043 - Epoch: [165][ 210/ 1236] Overall Loss 0.190062 Objective Loss 0.190062 LR 0.000250 Time 0.023806 +2023-10-05 21:59:52,243 - Epoch: [165][ 220/ 1236] Overall Loss 0.189280 Objective Loss 0.189280 LR 0.000250 Time 0.023632 +2023-10-05 21:59:52,441 - Epoch: [165][ 230/ 1236] Overall Loss 0.189048 Objective Loss 0.189048 LR 0.000250 Time 0.023466 +2023-10-05 21:59:52,642 - Epoch: [165][ 240/ 1236] Overall Loss 0.189982 Objective Loss 0.189982 LR 0.000250 Time 0.023324 +2023-10-05 21:59:52,841 - Epoch: [165][ 250/ 1236] Overall Loss 0.191491 Objective Loss 0.191491 LR 0.000250 Time 0.023185 +2023-10-05 21:59:53,042 - Epoch: [165][ 260/ 1236] Overall Loss 0.191499 Objective Loss 0.191499 LR 0.000250 Time 0.023064 +2023-10-05 21:59:53,241 - Epoch: [165][ 270/ 1236] Overall Loss 0.191799 Objective Loss 0.191799 LR 0.000250 Time 0.022946 +2023-10-05 21:59:53,441 - Epoch: [165][ 280/ 1236] Overall Loss 0.192932 Objective Loss 0.192932 LR 0.000250 Time 0.022841 +2023-10-05 21:59:53,640 - Epoch: [165][ 290/ 1236] Overall Loss 0.192821 Objective Loss 0.192821 LR 0.000250 Time 0.022737 +2023-10-05 21:59:53,841 - Epoch: [165][ 300/ 1236] Overall Loss 0.193562 Objective Loss 0.193562 LR 0.000250 Time 0.022648 +2023-10-05 21:59:54,040 - Epoch: [165][ 310/ 1236] Overall Loss 0.193703 Objective Loss 0.193703 LR 0.000250 Time 0.022558 +2023-10-05 21:59:54,241 - Epoch: [165][ 320/ 1236] Overall Loss 0.193051 Objective Loss 0.193051 LR 0.000250 Time 0.022481 +2023-10-05 21:59:54,440 - Epoch: [165][ 330/ 1236] Overall Loss 0.193687 Objective Loss 0.193687 LR 0.000250 Time 0.022401 +2023-10-05 21:59:54,646 - Epoch: [165][ 340/ 1236] Overall Loss 0.193879 Objective Loss 0.193879 LR 0.000250 Time 0.022347 +2023-10-05 21:59:54,852 - Epoch: [165][ 350/ 1236] Overall Loss 0.193652 Objective Loss 0.193652 LR 0.000250 Time 0.022296 +2023-10-05 21:59:55,059 - Epoch: [165][ 360/ 1236] Overall Loss 0.193670 Objective Loss 0.193670 LR 0.000250 Time 0.022250 +2023-10-05 21:59:55,259 - Epoch: [165][ 370/ 1236] Overall Loss 0.193853 Objective Loss 0.193853 LR 0.000250 Time 0.022189 +2023-10-05 21:59:55,461 - Epoch: [165][ 380/ 1236] Overall Loss 0.193931 Objective Loss 0.193931 LR 0.000250 Time 0.022136 +2023-10-05 21:59:55,662 - Epoch: [165][ 390/ 1236] Overall Loss 0.194792 Objective Loss 0.194792 LR 0.000250 Time 0.022081 +2023-10-05 21:59:55,864 - Epoch: [165][ 400/ 1236] Overall Loss 0.194609 Objective Loss 0.194609 LR 0.000250 Time 0.022033 +2023-10-05 21:59:56,064 - Epoch: [165][ 410/ 1236] Overall Loss 0.194909 Objective Loss 0.194909 LR 0.000250 Time 0.021984 +2023-10-05 21:59:56,267 - Epoch: [165][ 420/ 1236] Overall Loss 0.194591 Objective Loss 0.194591 LR 0.000250 Time 0.021942 +2023-10-05 21:59:56,468 - Epoch: [165][ 430/ 1236] Overall Loss 0.194714 Objective Loss 0.194714 LR 0.000250 Time 0.021898 +2023-10-05 21:59:56,670 - Epoch: [165][ 440/ 1236] Overall Loss 0.194301 Objective Loss 0.194301 LR 0.000250 Time 0.021860 +2023-10-05 21:59:56,871 - Epoch: [165][ 450/ 1236] Overall Loss 0.193954 Objective Loss 0.193954 LR 0.000250 Time 0.021819 +2023-10-05 21:59:57,074 - Epoch: [165][ 460/ 1236] Overall Loss 0.193585 Objective Loss 0.193585 LR 0.000250 Time 0.021785 +2023-10-05 21:59:57,275 - Epoch: [165][ 470/ 1236] Overall Loss 0.194015 Objective Loss 0.194015 LR 0.000250 Time 0.021749 +2023-10-05 21:59:57,477 - Epoch: [165][ 480/ 1236] Overall Loss 0.194204 Objective Loss 0.194204 LR 0.000250 Time 0.021716 +2023-10-05 21:59:57,678 - Epoch: [165][ 490/ 1236] Overall Loss 0.194038 Objective Loss 0.194038 LR 0.000250 Time 0.021682 +2023-10-05 21:59:57,880 - Epoch: [165][ 500/ 1236] Overall Loss 0.193857 Objective Loss 0.193857 LR 0.000250 Time 0.021653 +2023-10-05 21:59:58,081 - Epoch: [165][ 510/ 1236] Overall Loss 0.193424 Objective Loss 0.193424 LR 0.000250 Time 0.021622 +2023-10-05 21:59:58,284 - Epoch: [165][ 520/ 1236] Overall Loss 0.193334 Objective Loss 0.193334 LR 0.000250 Time 0.021595 +2023-10-05 21:59:58,485 - Epoch: [165][ 530/ 1236] Overall Loss 0.193430 Objective Loss 0.193430 LR 0.000250 Time 0.021566 +2023-10-05 21:59:58,688 - Epoch: [165][ 540/ 1236] Overall Loss 0.193130 Objective Loss 0.193130 LR 0.000250 Time 0.021541 +2023-10-05 21:59:58,889 - Epoch: [165][ 550/ 1236] Overall Loss 0.193319 Objective Loss 0.193319 LR 0.000250 Time 0.021515 +2023-10-05 21:59:59,091 - Epoch: [165][ 560/ 1236] Overall Loss 0.193467 Objective Loss 0.193467 LR 0.000250 Time 0.021491 +2023-10-05 21:59:59,292 - Epoch: [165][ 570/ 1236] Overall Loss 0.193411 Objective Loss 0.193411 LR 0.000250 Time 0.021466 +2023-10-05 21:59:59,495 - Epoch: [165][ 580/ 1236] Overall Loss 0.193195 Objective Loss 0.193195 LR 0.000250 Time 0.021444 +2023-10-05 21:59:59,696 - Epoch: [165][ 590/ 1236] Overall Loss 0.192785 Objective Loss 0.192785 LR 0.000250 Time 0.021421 +2023-10-05 21:59:59,898 - Epoch: [165][ 600/ 1236] Overall Loss 0.193017 Objective Loss 0.193017 LR 0.000250 Time 0.021401 +2023-10-05 22:00:00,099 - Epoch: [165][ 610/ 1236] Overall Loss 0.192935 Objective Loss 0.192935 LR 0.000250 Time 0.021379 +2023-10-05 22:00:00,302 - Epoch: [165][ 620/ 1236] Overall Loss 0.192531 Objective Loss 0.192531 LR 0.000250 Time 0.021360 +2023-10-05 22:00:00,503 - Epoch: [165][ 630/ 1236] Overall Loss 0.192458 Objective Loss 0.192458 LR 0.000250 Time 0.021340 +2023-10-05 22:00:00,706 - Epoch: [165][ 640/ 1236] Overall Loss 0.192429 Objective Loss 0.192429 LR 0.000250 Time 0.021323 +2023-10-05 22:00:00,907 - Epoch: [165][ 650/ 1236] Overall Loss 0.192638 Objective Loss 0.192638 LR 0.000250 Time 0.021304 +2023-10-05 22:00:01,110 - Epoch: [165][ 660/ 1236] Overall Loss 0.192806 Objective Loss 0.192806 LR 0.000250 Time 0.021288 +2023-10-05 22:00:01,311 - Epoch: [165][ 670/ 1236] Overall Loss 0.193057 Objective Loss 0.193057 LR 0.000250 Time 0.021270 +2023-10-05 22:00:01,514 - Epoch: [165][ 680/ 1236] Overall Loss 0.193105 Objective Loss 0.193105 LR 0.000250 Time 0.021255 +2023-10-05 22:00:01,715 - Epoch: [165][ 690/ 1236] Overall Loss 0.192981 Objective Loss 0.192981 LR 0.000250 Time 0.021238 +2023-10-05 22:00:01,918 - Epoch: [165][ 700/ 1236] Overall Loss 0.193156 Objective Loss 0.193156 LR 0.000250 Time 0.021224 +2023-10-05 22:00:02,119 - Epoch: [165][ 710/ 1236] Overall Loss 0.193444 Objective Loss 0.193444 LR 0.000250 Time 0.021208 +2023-10-05 22:00:02,321 - Epoch: [165][ 720/ 1236] Overall Loss 0.193467 Objective Loss 0.193467 LR 0.000250 Time 0.021194 +2023-10-05 22:00:02,522 - Epoch: [165][ 730/ 1236] Overall Loss 0.193427 Objective Loss 0.193427 LR 0.000250 Time 0.021179 +2023-10-05 22:00:02,725 - Epoch: [165][ 740/ 1236] Overall Loss 0.193881 Objective Loss 0.193881 LR 0.000250 Time 0.021166 +2023-10-05 22:00:02,926 - Epoch: [165][ 750/ 1236] Overall Loss 0.193647 Objective Loss 0.193647 LR 0.000250 Time 0.021151 +2023-10-05 22:00:03,129 - Epoch: [165][ 760/ 1236] Overall Loss 0.193679 Objective Loss 0.193679 LR 0.000250 Time 0.021139 +2023-10-05 22:00:03,330 - Epoch: [165][ 770/ 1236] Overall Loss 0.193869 Objective Loss 0.193869 LR 0.000250 Time 0.021125 +2023-10-05 22:00:03,533 - Epoch: [165][ 780/ 1236] Overall Loss 0.193936 Objective Loss 0.193936 LR 0.000250 Time 0.021114 +2023-10-05 22:00:03,734 - Epoch: [165][ 790/ 1236] Overall Loss 0.193804 Objective Loss 0.193804 LR 0.000250 Time 0.021101 +2023-10-05 22:00:03,936 - Epoch: [165][ 800/ 1236] Overall Loss 0.193636 Objective Loss 0.193636 LR 0.000250 Time 0.021090 +2023-10-05 22:00:04,137 - Epoch: [165][ 810/ 1236] Overall Loss 0.193448 Objective Loss 0.193448 LR 0.000250 Time 0.021077 +2023-10-05 22:00:04,340 - Epoch: [165][ 820/ 1236] Overall Loss 0.193583 Objective Loss 0.193583 LR 0.000250 Time 0.021067 +2023-10-05 22:00:04,541 - Epoch: [165][ 830/ 1236] Overall Loss 0.193525 Objective Loss 0.193525 LR 0.000250 Time 0.021055 +2023-10-05 22:00:04,743 - Epoch: [165][ 840/ 1236] Overall Loss 0.193303 Objective Loss 0.193303 LR 0.000250 Time 0.021044 +2023-10-05 22:00:04,944 - Epoch: [165][ 850/ 1236] Overall Loss 0.193181 Objective Loss 0.193181 LR 0.000250 Time 0.021033 +2023-10-05 22:00:05,147 - Epoch: [165][ 860/ 1236] Overall Loss 0.193302 Objective Loss 0.193302 LR 0.000250 Time 0.021023 +2023-10-05 22:00:05,347 - Epoch: [165][ 870/ 1236] Overall Loss 0.193319 Objective Loss 0.193319 LR 0.000250 Time 0.021012 +2023-10-05 22:00:05,550 - Epoch: [165][ 880/ 1236] Overall Loss 0.193326 Objective Loss 0.193326 LR 0.000250 Time 0.021003 +2023-10-05 22:00:05,751 - Epoch: [165][ 890/ 1236] Overall Loss 0.193429 Objective Loss 0.193429 LR 0.000250 Time 0.020992 +2023-10-05 22:00:05,953 - Epoch: [165][ 900/ 1236] Overall Loss 0.193636 Objective Loss 0.193636 LR 0.000250 Time 0.020984 +2023-10-05 22:00:06,154 - Epoch: [165][ 910/ 1236] Overall Loss 0.193647 Objective Loss 0.193647 LR 0.000250 Time 0.020974 +2023-10-05 22:00:06,357 - Epoch: [165][ 920/ 1236] Overall Loss 0.193637 Objective Loss 0.193637 LR 0.000250 Time 0.020966 +2023-10-05 22:00:06,558 - Epoch: [165][ 930/ 1236] Overall Loss 0.193700 Objective Loss 0.193700 LR 0.000250 Time 0.020956 +2023-10-05 22:00:06,761 - Epoch: [165][ 940/ 1236] Overall Loss 0.193795 Objective Loss 0.193795 LR 0.000250 Time 0.020948 +2023-10-05 22:00:06,962 - Epoch: [165][ 950/ 1236] Overall Loss 0.193981 Objective Loss 0.193981 LR 0.000250 Time 0.020939 +2023-10-05 22:00:07,164 - Epoch: [165][ 960/ 1236] Overall Loss 0.193645 Objective Loss 0.193645 LR 0.000250 Time 0.020931 +2023-10-05 22:00:07,365 - Epoch: [165][ 970/ 1236] Overall Loss 0.193554 Objective Loss 0.193554 LR 0.000250 Time 0.020922 +2023-10-05 22:00:07,568 - Epoch: [165][ 980/ 1236] Overall Loss 0.193489 Objective Loss 0.193489 LR 0.000250 Time 0.020915 +2023-10-05 22:00:07,769 - Epoch: [165][ 990/ 1236] Overall Loss 0.193451 Objective Loss 0.193451 LR 0.000250 Time 0.020907 +2023-10-05 22:00:07,971 - Epoch: [165][ 1000/ 1236] Overall Loss 0.193622 Objective Loss 0.193622 LR 0.000250 Time 0.020899 +2023-10-05 22:00:08,172 - Epoch: [165][ 1010/ 1236] Overall Loss 0.193625 Objective Loss 0.193625 LR 0.000250 Time 0.020891 +2023-10-05 22:00:08,374 - Epoch: [165][ 1020/ 1236] Overall Loss 0.193621 Objective Loss 0.193621 LR 0.000250 Time 0.020885 +2023-10-05 22:00:08,575 - Epoch: [165][ 1030/ 1236] Overall Loss 0.193544 Objective Loss 0.193544 LR 0.000250 Time 0.020877 +2023-10-05 22:00:08,778 - Epoch: [165][ 1040/ 1236] Overall Loss 0.193614 Objective Loss 0.193614 LR 0.000250 Time 0.020870 +2023-10-05 22:00:08,979 - Epoch: [165][ 1050/ 1236] Overall Loss 0.193423 Objective Loss 0.193423 LR 0.000250 Time 0.020863 +2023-10-05 22:00:09,182 - Epoch: [165][ 1060/ 1236] Overall Loss 0.193299 Objective Loss 0.193299 LR 0.000250 Time 0.020857 +2023-10-05 22:00:09,382 - Epoch: [165][ 1070/ 1236] Overall Loss 0.193292 Objective Loss 0.193292 LR 0.000250 Time 0.020849 +2023-10-05 22:00:09,585 - Epoch: [165][ 1080/ 1236] Overall Loss 0.193289 Objective Loss 0.193289 LR 0.000250 Time 0.020843 +2023-10-05 22:00:09,786 - Epoch: [165][ 1090/ 1236] Overall Loss 0.193331 Objective Loss 0.193331 LR 0.000250 Time 0.020836 +2023-10-05 22:00:09,989 - Epoch: [165][ 1100/ 1236] Overall Loss 0.193227 Objective Loss 0.193227 LR 0.000250 Time 0.020831 +2023-10-05 22:00:10,190 - Epoch: [165][ 1110/ 1236] Overall Loss 0.193255 Objective Loss 0.193255 LR 0.000250 Time 0.020824 +2023-10-05 22:00:10,393 - Epoch: [165][ 1120/ 1236] Overall Loss 0.193405 Objective Loss 0.193405 LR 0.000250 Time 0.020819 +2023-10-05 22:00:10,601 - Epoch: [165][ 1130/ 1236] Overall Loss 0.193648 Objective Loss 0.193648 LR 0.000250 Time 0.020818 +2023-10-05 22:00:10,813 - Epoch: [165][ 1140/ 1236] Overall Loss 0.193807 Objective Loss 0.193807 LR 0.000250 Time 0.020822 +2023-10-05 22:00:11,022 - Epoch: [165][ 1150/ 1236] Overall Loss 0.193895 Objective Loss 0.193895 LR 0.000250 Time 0.020822 +2023-10-05 22:00:11,234 - Epoch: [165][ 1160/ 1236] Overall Loss 0.193785 Objective Loss 0.193785 LR 0.000250 Time 0.020825 +2023-10-05 22:00:11,442 - Epoch: [165][ 1170/ 1236] Overall Loss 0.193723 Objective Loss 0.193723 LR 0.000250 Time 0.020825 +2023-10-05 22:00:11,655 - Epoch: [165][ 1180/ 1236] Overall Loss 0.193587 Objective Loss 0.193587 LR 0.000250 Time 0.020829 +2023-10-05 22:00:11,864 - Epoch: [165][ 1190/ 1236] Overall Loss 0.193669 Objective Loss 0.193669 LR 0.000250 Time 0.020829 +2023-10-05 22:00:12,077 - Epoch: [165][ 1200/ 1236] Overall Loss 0.193736 Objective Loss 0.193736 LR 0.000250 Time 0.020832 +2023-10-05 22:00:12,285 - Epoch: [165][ 1210/ 1236] Overall Loss 0.193670 Objective Loss 0.193670 LR 0.000250 Time 0.020832 +2023-10-05 22:00:12,498 - Epoch: [165][ 1220/ 1236] Overall Loss 0.193745 Objective Loss 0.193745 LR 0.000250 Time 0.020836 +2023-10-05 22:00:12,759 - Epoch: [165][ 1230/ 1236] Overall Loss 0.193941 Objective Loss 0.193941 LR 0.000250 Time 0.020878 +2023-10-05 22:00:12,878 - Epoch: [165][ 1236/ 1236] Overall Loss 0.194053 Objective Loss 0.194053 Top1 87.983707 Top5 98.574338 LR 0.000250 Time 0.020873 +2023-10-05 22:00:12,994 - --- validate (epoch=165)----------- +2023-10-05 22:00:12,995 - 29943 samples (256 per mini-batch) +2023-10-05 22:00:13,453 - Epoch: [165][ 10/ 117] Loss 0.301707 Top1 85.156250 Top5 98.007812 +2023-10-05 22:00:13,599 - Epoch: [165][ 20/ 117] Loss 0.311151 Top1 85.683594 Top5 98.125000 +2023-10-05 22:00:13,746 - Epoch: [165][ 30/ 117] Loss 0.306430 Top1 85.403646 Top5 98.085938 +2023-10-05 22:00:13,893 - Epoch: [165][ 40/ 117] Loss 0.308757 Top1 85.341797 Top5 98.027344 +2023-10-05 22:00:14,039 - Epoch: [165][ 50/ 117] Loss 0.313470 Top1 85.281250 Top5 97.984375 +2023-10-05 22:00:14,189 - Epoch: [165][ 60/ 117] Loss 0.312768 Top1 85.195312 Top5 97.988281 +2023-10-05 22:00:14,338 - Epoch: [165][ 70/ 117] Loss 0.318285 Top1 85.089286 Top5 98.007812 +2023-10-05 22:00:14,486 - Epoch: [165][ 80/ 117] Loss 0.313826 Top1 85.102539 Top5 98.061523 +2023-10-05 22:00:14,633 - Epoch: [165][ 90/ 117] Loss 0.313723 Top1 85.047743 Top5 98.081597 +2023-10-05 22:00:14,782 - Epoch: [165][ 100/ 117] Loss 0.311040 Top1 85.183594 Top5 98.070312 +2023-10-05 22:00:14,937 - Epoch: [165][ 110/ 117] Loss 0.309774 Top1 85.145597 Top5 98.096591 +2023-10-05 22:00:15,022 - Epoch: [165][ 117/ 117] Loss 0.311300 Top1 85.064957 Top5 98.093043 +2023-10-05 22:00:15,155 - ==> Top1: 85.065 Top5: 98.093 Loss: 0.311 + +2023-10-05 22:00:15,156 - ==> Confusion: +[[ 908 2 6 3 6 2 0 0 7 80 1 1 2 2 8 5 2 2 1 0 12] + [ 0 1065 2 0 6 15 1 16 3 0 1 1 0 1 1 5 1 0 5 1 7] + [ 1 2 955 19 0 0 28 5 0 0 6 3 9 1 2 3 1 1 6 2 12] + [ 1 1 9 974 0 5 1 0 1 1 4 0 7 2 24 6 1 7 25 3 17] + [ 17 8 2 0 968 1 0 1 0 14 0 3 2 5 6 4 10 3 0 1 5] + [ 2 33 1 1 1 992 1 19 2 0 3 9 1 14 6 2 3 1 5 2 18] + [ 0 3 16 0 0 0 1134 6 0 0 1 4 2 0 1 8 0 1 1 5 9] + [ 3 25 12 0 2 26 4 1060 2 3 2 12 2 4 0 2 0 1 39 9 10] + [ 19 1 0 0 0 5 0 0 976 39 11 2 2 9 14 3 2 2 3 0 1] + [ 73 0 2 0 2 3 0 0 26 962 0 1 1 22 5 8 0 2 0 0 12] + [ 2 3 8 8 1 0 5 4 7 1 976 3 1 14 4 2 1 0 3 2 8] + [ 1 0 1 0 0 12 0 2 0 1 0 960 20 6 0 5 3 15 0 6 3] + [ 2 1 3 4 0 1 0 0 0 0 0 31 992 0 1 6 3 12 1 2 9] + [ 2 0 1 0 0 5 0 0 11 10 7 3 3 1062 3 1 0 0 0 3 8] + [ 10 3 3 8 4 1 0 0 26 2 5 1 2 4 1005 0 1 2 12 0 12] + [ 0 3 1 0 1 0 0 0 0 0 0 7 6 2 0 1080 11 14 0 5 4] + [ 0 12 1 0 4 2 0 0 1 0 0 6 0 2 2 14 1100 0 1 3 13] + [ 0 0 0 0 0 0 2 0 0 0 0 2 28 2 0 5 0 994 1 0 4] + [ 1 3 3 20 1 0 0 18 1 0 3 0 2 0 10 0 1 0 992 2 11] + [ 0 4 2 3 1 5 11 6 0 0 1 15 6 4 0 9 9 1 2 1065 8] + [ 93 181 105 71 73 121 42 81 92 65 191 94 371 302 155 68 137 74 159 179 5251]] + +2023-10-05 22:00:15,157 - ==> Best [Top1: 85.452 Top5: 98.110 Sparsity:0.00 Params: 148928 on epoch: 154] +2023-10-05 22:00:15,157 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:00:15,163 - + +2023-10-05 22:00:15,163 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:00:16,170 - Epoch: [166][ 10/ 1236] Overall Loss 0.186713 Objective Loss 0.186713 LR 0.000250 Time 0.100635 +2023-10-05 22:00:16,371 - Epoch: [166][ 20/ 1236] Overall Loss 0.184902 Objective Loss 0.184902 LR 0.000250 Time 0.060338 +2023-10-05 22:00:16,570 - Epoch: [166][ 30/ 1236] Overall Loss 0.188881 Objective Loss 0.188881 LR 0.000250 Time 0.046855 +2023-10-05 22:00:16,770 - Epoch: [166][ 40/ 1236] Overall Loss 0.188082 Objective Loss 0.188082 LR 0.000250 Time 0.040119 +2023-10-05 22:00:16,968 - Epoch: [166][ 50/ 1236] Overall Loss 0.188312 Objective Loss 0.188312 LR 0.000250 Time 0.036068 +2023-10-05 22:00:17,170 - Epoch: [166][ 60/ 1236] Overall Loss 0.193945 Objective Loss 0.193945 LR 0.000250 Time 0.033402 +2023-10-05 22:00:17,369 - Epoch: [166][ 70/ 1236] Overall Loss 0.191504 Objective Loss 0.191504 LR 0.000250 Time 0.031475 +2023-10-05 22:00:17,568 - Epoch: [166][ 80/ 1236] Overall Loss 0.192745 Objective Loss 0.192745 LR 0.000250 Time 0.030029 +2023-10-05 22:00:17,768 - Epoch: [166][ 90/ 1236] Overall Loss 0.192026 Objective Loss 0.192026 LR 0.000250 Time 0.028906 +2023-10-05 22:00:17,968 - Epoch: [166][ 100/ 1236] Overall Loss 0.191157 Objective Loss 0.191157 LR 0.000250 Time 0.028009 +2023-10-05 22:00:18,167 - Epoch: [166][ 110/ 1236] Overall Loss 0.193175 Objective Loss 0.193175 LR 0.000250 Time 0.027272 +2023-10-05 22:00:18,367 - Epoch: [166][ 120/ 1236] Overall Loss 0.192370 Objective Loss 0.192370 LR 0.000250 Time 0.026665 +2023-10-05 22:00:18,566 - Epoch: [166][ 130/ 1236] Overall Loss 0.193180 Objective Loss 0.193180 LR 0.000250 Time 0.026137 +2023-10-05 22:00:18,766 - Epoch: [166][ 140/ 1236] Overall Loss 0.194098 Objective Loss 0.194098 LR 0.000250 Time 0.025698 +2023-10-05 22:00:18,965 - Epoch: [166][ 150/ 1236] Overall Loss 0.193187 Objective Loss 0.193187 LR 0.000250 Time 0.025311 +2023-10-05 22:00:19,166 - Epoch: [166][ 160/ 1236] Overall Loss 0.194735 Objective Loss 0.194735 LR 0.000250 Time 0.024985 +2023-10-05 22:00:19,367 - Epoch: [166][ 170/ 1236] Overall Loss 0.194985 Objective Loss 0.194985 LR 0.000250 Time 0.024694 +2023-10-05 22:00:19,567 - Epoch: [166][ 180/ 1236] Overall Loss 0.196412 Objective Loss 0.196412 LR 0.000250 Time 0.024431 +2023-10-05 22:00:19,768 - Epoch: [166][ 190/ 1236] Overall Loss 0.195781 Objective Loss 0.195781 LR 0.000250 Time 0.024199 +2023-10-05 22:00:19,968 - Epoch: [166][ 200/ 1236] Overall Loss 0.195165 Objective Loss 0.195165 LR 0.000250 Time 0.023989 +2023-10-05 22:00:20,168 - Epoch: [166][ 210/ 1236] Overall Loss 0.195224 Objective Loss 0.195224 LR 0.000250 Time 0.023799 +2023-10-05 22:00:20,369 - Epoch: [166][ 220/ 1236] Overall Loss 0.195329 Objective Loss 0.195329 LR 0.000250 Time 0.023625 +2023-10-05 22:00:20,569 - Epoch: [166][ 230/ 1236] Overall Loss 0.195017 Objective Loss 0.195017 LR 0.000250 Time 0.023470 +2023-10-05 22:00:20,769 - Epoch: [166][ 240/ 1236] Overall Loss 0.194882 Objective Loss 0.194882 LR 0.000250 Time 0.023324 +2023-10-05 22:00:20,970 - Epoch: [166][ 250/ 1236] Overall Loss 0.193830 Objective Loss 0.193830 LR 0.000250 Time 0.023192 +2023-10-05 22:00:21,171 - Epoch: [166][ 260/ 1236] Overall Loss 0.193360 Objective Loss 0.193360 LR 0.000250 Time 0.023072 +2023-10-05 22:00:21,373 - Epoch: [166][ 270/ 1236] Overall Loss 0.194401 Objective Loss 0.194401 LR 0.000250 Time 0.022962 +2023-10-05 22:00:21,573 - Epoch: [166][ 280/ 1236] Overall Loss 0.194487 Objective Loss 0.194487 LR 0.000250 Time 0.022857 +2023-10-05 22:00:21,774 - Epoch: [166][ 290/ 1236] Overall Loss 0.194597 Objective Loss 0.194597 LR 0.000250 Time 0.022760 +2023-10-05 22:00:21,974 - Epoch: [166][ 300/ 1236] Overall Loss 0.194778 Objective Loss 0.194778 LR 0.000250 Time 0.022668 +2023-10-05 22:00:22,175 - Epoch: [166][ 310/ 1236] Overall Loss 0.194666 Objective Loss 0.194666 LR 0.000250 Time 0.022583 +2023-10-05 22:00:22,375 - Epoch: [166][ 320/ 1236] Overall Loss 0.194869 Objective Loss 0.194869 LR 0.000250 Time 0.022503 +2023-10-05 22:00:22,576 - Epoch: [166][ 330/ 1236] Overall Loss 0.194868 Objective Loss 0.194868 LR 0.000250 Time 0.022427 +2023-10-05 22:00:22,776 - Epoch: [166][ 340/ 1236] Overall Loss 0.195378 Objective Loss 0.195378 LR 0.000250 Time 0.022356 +2023-10-05 22:00:22,977 - Epoch: [166][ 350/ 1236] Overall Loss 0.195018 Objective Loss 0.195018 LR 0.000250 Time 0.022290 +2023-10-05 22:00:23,177 - Epoch: [166][ 360/ 1236] Overall Loss 0.194526 Objective Loss 0.194526 LR 0.000250 Time 0.022227 +2023-10-05 22:00:23,379 - Epoch: [166][ 370/ 1236] Overall Loss 0.195007 Objective Loss 0.195007 LR 0.000250 Time 0.022169 +2023-10-05 22:00:23,580 - Epoch: [166][ 380/ 1236] Overall Loss 0.195135 Objective Loss 0.195135 LR 0.000250 Time 0.022115 +2023-10-05 22:00:23,782 - Epoch: [166][ 390/ 1236] Overall Loss 0.194951 Objective Loss 0.194951 LR 0.000250 Time 0.022065 +2023-10-05 22:00:23,983 - Epoch: [166][ 400/ 1236] Overall Loss 0.194847 Objective Loss 0.194847 LR 0.000250 Time 0.022013 +2023-10-05 22:00:24,185 - Epoch: [166][ 410/ 1236] Overall Loss 0.194952 Objective Loss 0.194952 LR 0.000250 Time 0.021969 +2023-10-05 22:00:24,384 - Epoch: [166][ 420/ 1236] Overall Loss 0.194499 Objective Loss 0.194499 LR 0.000250 Time 0.021920 +2023-10-05 22:00:24,586 - Epoch: [166][ 430/ 1236] Overall Loss 0.194540 Objective Loss 0.194540 LR 0.000250 Time 0.021878 +2023-10-05 22:00:24,786 - Epoch: [166][ 440/ 1236] Overall Loss 0.194432 Objective Loss 0.194432 LR 0.000250 Time 0.021832 +2023-10-05 22:00:24,988 - Epoch: [166][ 450/ 1236] Overall Loss 0.194088 Objective Loss 0.194088 LR 0.000250 Time 0.021794 +2023-10-05 22:00:25,188 - Epoch: [166][ 460/ 1236] Overall Loss 0.194124 Objective Loss 0.194124 LR 0.000250 Time 0.021755 +2023-10-05 22:00:25,390 - Epoch: [166][ 470/ 1236] Overall Loss 0.194079 Objective Loss 0.194079 LR 0.000250 Time 0.021720 +2023-10-05 22:00:25,589 - Epoch: [166][ 480/ 1236] Overall Loss 0.193913 Objective Loss 0.193913 LR 0.000250 Time 0.021682 +2023-10-05 22:00:25,791 - Epoch: [166][ 490/ 1236] Overall Loss 0.193974 Objective Loss 0.193974 LR 0.000250 Time 0.021650 +2023-10-05 22:00:25,990 - Epoch: [166][ 500/ 1236] Overall Loss 0.193671 Objective Loss 0.193671 LR 0.000250 Time 0.021615 +2023-10-05 22:00:26,192 - Epoch: [166][ 510/ 1236] Overall Loss 0.193538 Objective Loss 0.193538 LR 0.000250 Time 0.021586 +2023-10-05 22:00:26,392 - Epoch: [166][ 520/ 1236] Overall Loss 0.193242 Objective Loss 0.193242 LR 0.000250 Time 0.021556 +2023-10-05 22:00:26,594 - Epoch: [166][ 530/ 1236] Overall Loss 0.192755 Objective Loss 0.192755 LR 0.000250 Time 0.021528 +2023-10-05 22:00:26,794 - Epoch: [166][ 540/ 1236] Overall Loss 0.193084 Objective Loss 0.193084 LR 0.000250 Time 0.021500 +2023-10-05 22:00:26,996 - Epoch: [166][ 550/ 1236] Overall Loss 0.193361 Objective Loss 0.193361 LR 0.000250 Time 0.021475 +2023-10-05 22:00:27,196 - Epoch: [166][ 560/ 1236] Overall Loss 0.193112 Objective Loss 0.193112 LR 0.000250 Time 0.021449 +2023-10-05 22:00:27,398 - Epoch: [166][ 570/ 1236] Overall Loss 0.193323 Objective Loss 0.193323 LR 0.000250 Time 0.021426 +2023-10-05 22:00:27,598 - Epoch: [166][ 580/ 1236] Overall Loss 0.193714 Objective Loss 0.193714 LR 0.000250 Time 0.021401 +2023-10-05 22:00:27,801 - Epoch: [166][ 590/ 1236] Overall Loss 0.193734 Objective Loss 0.193734 LR 0.000250 Time 0.021380 +2023-10-05 22:00:28,001 - Epoch: [166][ 600/ 1236] Overall Loss 0.194154 Objective Loss 0.194154 LR 0.000250 Time 0.021357 +2023-10-05 22:00:28,203 - Epoch: [166][ 610/ 1236] Overall Loss 0.194184 Objective Loss 0.194184 LR 0.000250 Time 0.021338 +2023-10-05 22:00:28,404 - Epoch: [166][ 620/ 1236] Overall Loss 0.193717 Objective Loss 0.193717 LR 0.000250 Time 0.021317 +2023-10-05 22:00:28,605 - Epoch: [166][ 630/ 1236] Overall Loss 0.193679 Objective Loss 0.193679 LR 0.000250 Time 0.021298 +2023-10-05 22:00:28,806 - Epoch: [166][ 640/ 1236] Overall Loss 0.193544 Objective Loss 0.193544 LR 0.000250 Time 0.021277 +2023-10-05 22:00:29,007 - Epoch: [166][ 650/ 1236] Overall Loss 0.193100 Objective Loss 0.193100 LR 0.000250 Time 0.021259 +2023-10-05 22:00:29,208 - Epoch: [166][ 660/ 1236] Overall Loss 0.192915 Objective Loss 0.192915 LR 0.000250 Time 0.021240 +2023-10-05 22:00:29,410 - Epoch: [166][ 670/ 1236] Overall Loss 0.192888 Objective Loss 0.192888 LR 0.000250 Time 0.021224 +2023-10-05 22:00:29,610 - Epoch: [166][ 680/ 1236] Overall Loss 0.192773 Objective Loss 0.192773 LR 0.000250 Time 0.021206 +2023-10-05 22:00:29,812 - Epoch: [166][ 690/ 1236] Overall Loss 0.192291 Objective Loss 0.192291 LR 0.000250 Time 0.021190 +2023-10-05 22:00:30,011 - Epoch: [166][ 700/ 1236] Overall Loss 0.192259 Objective Loss 0.192259 LR 0.000250 Time 0.021172 +2023-10-05 22:00:30,213 - Epoch: [166][ 710/ 1236] Overall Loss 0.191861 Objective Loss 0.191861 LR 0.000250 Time 0.021157 +2023-10-05 22:00:30,413 - Epoch: [166][ 720/ 1236] Overall Loss 0.191672 Objective Loss 0.191672 LR 0.000250 Time 0.021141 +2023-10-05 22:00:30,615 - Epoch: [166][ 730/ 1236] Overall Loss 0.191519 Objective Loss 0.191519 LR 0.000250 Time 0.021127 +2023-10-05 22:00:30,814 - Epoch: [166][ 740/ 1236] Overall Loss 0.191703 Objective Loss 0.191703 LR 0.000250 Time 0.021110 +2023-10-05 22:00:31,016 - Epoch: [166][ 750/ 1236] Overall Loss 0.191818 Objective Loss 0.191818 LR 0.000250 Time 0.021097 +2023-10-05 22:00:31,215 - Epoch: [166][ 760/ 1236] Overall Loss 0.191803 Objective Loss 0.191803 LR 0.000250 Time 0.021081 +2023-10-05 22:00:31,417 - Epoch: [166][ 770/ 1236] Overall Loss 0.191523 Objective Loss 0.191523 LR 0.000250 Time 0.021070 +2023-10-05 22:00:31,617 - Epoch: [166][ 780/ 1236] Overall Loss 0.191780 Objective Loss 0.191780 LR 0.000250 Time 0.021055 +2023-10-05 22:00:31,818 - Epoch: [166][ 790/ 1236] Overall Loss 0.191682 Objective Loss 0.191682 LR 0.000250 Time 0.021043 +2023-10-05 22:00:32,019 - Epoch: [166][ 800/ 1236] Overall Loss 0.192013 Objective Loss 0.192013 LR 0.000250 Time 0.021030 +2023-10-05 22:00:32,220 - Epoch: [166][ 810/ 1236] Overall Loss 0.192066 Objective Loss 0.192066 LR 0.000250 Time 0.021019 +2023-10-05 22:00:32,421 - Epoch: [166][ 820/ 1236] Overall Loss 0.191886 Objective Loss 0.191886 LR 0.000250 Time 0.021006 +2023-10-05 22:00:32,622 - Epoch: [166][ 830/ 1236] Overall Loss 0.191986 Objective Loss 0.191986 LR 0.000250 Time 0.020995 +2023-10-05 22:00:32,823 - Epoch: [166][ 840/ 1236] Overall Loss 0.192188 Objective Loss 0.192188 LR 0.000250 Time 0.020984 +2023-10-05 22:00:33,025 - Epoch: [166][ 850/ 1236] Overall Loss 0.192252 Objective Loss 0.192252 LR 0.000250 Time 0.020975 +2023-10-05 22:00:33,226 - Epoch: [166][ 860/ 1236] Overall Loss 0.192356 Objective Loss 0.192356 LR 0.000250 Time 0.020964 +2023-10-05 22:00:33,428 - Epoch: [166][ 870/ 1236] Overall Loss 0.192251 Objective Loss 0.192251 LR 0.000250 Time 0.020954 +2023-10-05 22:00:33,628 - Epoch: [166][ 880/ 1236] Overall Loss 0.192433 Objective Loss 0.192433 LR 0.000250 Time 0.020943 +2023-10-05 22:00:33,830 - Epoch: [166][ 890/ 1236] Overall Loss 0.192566 Objective Loss 0.192566 LR 0.000250 Time 0.020934 +2023-10-05 22:00:34,031 - Epoch: [166][ 900/ 1236] Overall Loss 0.192397 Objective Loss 0.192397 LR 0.000250 Time 0.020925 +2023-10-05 22:00:34,233 - Epoch: [166][ 910/ 1236] Overall Loss 0.192437 Objective Loss 0.192437 LR 0.000250 Time 0.020916 +2023-10-05 22:00:34,433 - Epoch: [166][ 920/ 1236] Overall Loss 0.192673 Objective Loss 0.192673 LR 0.000250 Time 0.020906 +2023-10-05 22:00:34,635 - Epoch: [166][ 930/ 1236] Overall Loss 0.192951 Objective Loss 0.192951 LR 0.000250 Time 0.020897 +2023-10-05 22:00:34,835 - Epoch: [166][ 940/ 1236] Overall Loss 0.192853 Objective Loss 0.192853 LR 0.000250 Time 0.020888 +2023-10-05 22:00:35,038 - Epoch: [166][ 950/ 1236] Overall Loss 0.192960 Objective Loss 0.192960 LR 0.000250 Time 0.020881 +2023-10-05 22:00:35,238 - Epoch: [166][ 960/ 1236] Overall Loss 0.192984 Objective Loss 0.192984 LR 0.000250 Time 0.020872 +2023-10-05 22:00:35,440 - Epoch: [166][ 970/ 1236] Overall Loss 0.192846 Objective Loss 0.192846 LR 0.000250 Time 0.020864 +2023-10-05 22:00:35,640 - Epoch: [166][ 980/ 1236] Overall Loss 0.192887 Objective Loss 0.192887 LR 0.000250 Time 0.020855 +2023-10-05 22:00:35,842 - Epoch: [166][ 990/ 1236] Overall Loss 0.192709 Objective Loss 0.192709 LR 0.000250 Time 0.020848 +2023-10-05 22:00:36,041 - Epoch: [166][ 1000/ 1236] Overall Loss 0.193033 Objective Loss 0.193033 LR 0.000250 Time 0.020838 +2023-10-05 22:00:36,243 - Epoch: [166][ 1010/ 1236] Overall Loss 0.192845 Objective Loss 0.192845 LR 0.000250 Time 0.020831 +2023-10-05 22:00:36,443 - Epoch: [166][ 1020/ 1236] Overall Loss 0.192820 Objective Loss 0.192820 LR 0.000250 Time 0.020822 +2023-10-05 22:00:36,644 - Epoch: [166][ 1030/ 1236] Overall Loss 0.192644 Objective Loss 0.192644 LR 0.000250 Time 0.020815 +2023-10-05 22:00:36,844 - Epoch: [166][ 1040/ 1236] Overall Loss 0.192566 Objective Loss 0.192566 LR 0.000250 Time 0.020807 +2023-10-05 22:00:37,046 - Epoch: [166][ 1050/ 1236] Overall Loss 0.192508 Objective Loss 0.192508 LR 0.000250 Time 0.020801 +2023-10-05 22:00:37,246 - Epoch: [166][ 1060/ 1236] Overall Loss 0.192654 Objective Loss 0.192654 LR 0.000250 Time 0.020793 +2023-10-05 22:00:37,449 - Epoch: [166][ 1070/ 1236] Overall Loss 0.192508 Objective Loss 0.192508 LR 0.000250 Time 0.020788 +2023-10-05 22:00:37,649 - Epoch: [166][ 1080/ 1236] Overall Loss 0.192429 Objective Loss 0.192429 LR 0.000250 Time 0.020780 +2023-10-05 22:00:37,851 - Epoch: [166][ 1090/ 1236] Overall Loss 0.192452 Objective Loss 0.192452 LR 0.000250 Time 0.020775 +2023-10-05 22:00:38,051 - Epoch: [166][ 1100/ 1236] Overall Loss 0.192232 Objective Loss 0.192232 LR 0.000250 Time 0.020767 +2023-10-05 22:00:38,254 - Epoch: [166][ 1110/ 1236] Overall Loss 0.192237 Objective Loss 0.192237 LR 0.000250 Time 0.020762 +2023-10-05 22:00:38,454 - Epoch: [166][ 1120/ 1236] Overall Loss 0.192252 Objective Loss 0.192252 LR 0.000250 Time 0.020756 +2023-10-05 22:00:38,657 - Epoch: [166][ 1130/ 1236] Overall Loss 0.192112 Objective Loss 0.192112 LR 0.000250 Time 0.020751 +2023-10-05 22:00:38,857 - Epoch: [166][ 1140/ 1236] Overall Loss 0.191961 Objective Loss 0.191961 LR 0.000250 Time 0.020744 +2023-10-05 22:00:39,059 - Epoch: [166][ 1150/ 1236] Overall Loss 0.192090 Objective Loss 0.192090 LR 0.000250 Time 0.020739 +2023-10-05 22:00:39,259 - Epoch: [166][ 1160/ 1236] Overall Loss 0.192293 Objective Loss 0.192293 LR 0.000250 Time 0.020732 +2023-10-05 22:00:39,461 - Epoch: [166][ 1170/ 1236] Overall Loss 0.192311 Objective Loss 0.192311 LR 0.000250 Time 0.020728 +2023-10-05 22:00:39,662 - Epoch: [166][ 1180/ 1236] Overall Loss 0.192444 Objective Loss 0.192444 LR 0.000250 Time 0.020720 +2023-10-05 22:00:39,864 - Epoch: [166][ 1190/ 1236] Overall Loss 0.192537 Objective Loss 0.192537 LR 0.000250 Time 0.020716 +2023-10-05 22:00:40,066 - Epoch: [166][ 1200/ 1236] Overall Loss 0.192663 Objective Loss 0.192663 LR 0.000250 Time 0.020711 +2023-10-05 22:00:40,268 - Epoch: [166][ 1210/ 1236] Overall Loss 0.192802 Objective Loss 0.192802 LR 0.000250 Time 0.020706 +2023-10-05 22:00:40,468 - Epoch: [166][ 1220/ 1236] Overall Loss 0.192928 Objective Loss 0.192928 LR 0.000250 Time 0.020701 +2023-10-05 22:00:40,719 - Epoch: [166][ 1230/ 1236] Overall Loss 0.192625 Objective Loss 0.192625 LR 0.000250 Time 0.020736 +2023-10-05 22:00:40,837 - Epoch: [166][ 1236/ 1236] Overall Loss 0.192732 Objective Loss 0.192732 Top1 87.983707 Top5 98.778004 LR 0.000250 Time 0.020730 +2023-10-05 22:00:40,968 - --- validate (epoch=166)----------- +2023-10-05 22:00:40,968 - 29943 samples (256 per mini-batch) +2023-10-05 22:00:41,422 - Epoch: [166][ 10/ 117] Loss 0.332875 Top1 86.015625 Top5 97.968750 +2023-10-05 22:00:41,572 - Epoch: [166][ 20/ 117] Loss 0.315573 Top1 85.722656 Top5 98.105469 +2023-10-05 22:00:41,719 - Epoch: [166][ 30/ 117] Loss 0.306963 Top1 85.572917 Top5 98.151042 +2023-10-05 22:00:41,867 - Epoch: [166][ 40/ 117] Loss 0.305408 Top1 85.664062 Top5 98.173828 +2023-10-05 22:00:42,014 - Epoch: [166][ 50/ 117] Loss 0.304913 Top1 85.570312 Top5 98.171875 +2023-10-05 22:00:42,162 - Epoch: [166][ 60/ 117] Loss 0.307875 Top1 85.397135 Top5 98.183594 +2023-10-05 22:00:42,310 - Epoch: [166][ 70/ 117] Loss 0.307711 Top1 85.407366 Top5 98.147321 +2023-10-05 22:00:42,458 - Epoch: [166][ 80/ 117] Loss 0.308456 Top1 85.458984 Top5 98.144531 +2023-10-05 22:00:42,604 - Epoch: [166][ 90/ 117] Loss 0.306873 Top1 85.473090 Top5 98.185764 +2023-10-05 22:00:42,760 - Epoch: [166][ 100/ 117] Loss 0.309119 Top1 85.476562 Top5 98.179688 +2023-10-05 22:00:42,919 - Epoch: [166][ 110/ 117] Loss 0.308992 Top1 85.529119 Top5 98.188920 +2023-10-05 22:00:43,004 - Epoch: [166][ 117/ 117] Loss 0.310108 Top1 85.492436 Top5 98.169856 +2023-10-05 22:00:43,151 - ==> Top1: 85.492 Top5: 98.170 Loss: 0.310 + +2023-10-05 22:00:43,151 - ==> Confusion: +[[ 935 0 8 1 7 4 0 0 6 64 2 0 0 1 6 2 2 1 0 0 11] + [ 0 1038 2 0 6 26 1 23 3 0 0 3 0 0 2 3 3 0 8 2 11] + [ 5 1 977 12 4 0 22 7 0 0 6 0 8 1 1 2 0 1 2 1 6] + [ 0 0 11 986 1 5 2 1 1 1 4 0 2 1 26 3 0 7 19 2 17] + [ 20 4 3 0 970 4 0 2 0 9 0 2 0 5 11 1 10 2 1 0 6] + [ 4 27 1 2 3 998 0 20 2 1 8 5 1 11 6 2 4 0 3 5 13] + [ 0 4 34 0 0 0 1117 6 0 0 2 2 2 1 1 6 0 1 4 5 6] + [ 5 9 14 0 3 23 6 1084 3 2 3 13 2 4 0 0 1 0 29 6 11] + [ 18 1 0 0 0 2 1 0 978 43 10 1 3 7 12 4 2 0 3 1 3] + [ 100 0 5 1 3 2 0 0 26 943 0 2 2 16 5 3 0 3 0 0 8] + [ 1 3 17 4 0 0 3 4 11 0 973 3 2 8 5 0 2 0 6 1 10] + [ 0 0 3 0 0 15 0 3 0 1 0 941 31 6 0 5 1 17 0 8 4] + [ 2 1 5 5 1 2 0 2 1 0 1 28 982 1 0 3 3 16 2 2 11] + [ 2 0 0 1 1 7 0 1 11 16 7 5 1 1048 4 3 1 1 0 1 9] + [ 15 1 4 8 2 0 0 0 25 2 1 1 3 1 1012 0 1 1 11 0 13] + [ 0 3 2 0 2 1 2 0 0 0 0 7 6 2 1 1067 14 13 0 10 4] + [ 1 9 2 0 3 4 0 1 1 0 0 4 0 1 3 11 1101 0 0 5 15] + [ 0 0 0 2 1 0 3 0 0 0 0 1 21 2 0 7 0 998 1 0 2] + [ 0 1 10 23 1 0 1 15 1 0 3 0 1 0 6 0 1 0 994 2 9] + [ 0 2 4 2 2 4 6 10 0 0 1 12 4 2 0 5 9 1 4 1073 11] + [ 132 137 172 57 80 124 34 86 117 62 164 98 306 263 154 54 124 57 138 162 5384]] + +2023-10-05 22:00:43,153 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:00:43,153 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:00:43,166 - + +2023-10-05 22:00:43,166 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:00:44,270 - Epoch: [167][ 10/ 1236] Overall Loss 0.195800 Objective Loss 0.195800 LR 0.000250 Time 0.110320 +2023-10-05 22:00:44,472 - Epoch: [167][ 20/ 1236] Overall Loss 0.188134 Objective Loss 0.188134 LR 0.000250 Time 0.065265 +2023-10-05 22:00:44,672 - Epoch: [167][ 30/ 1236] Overall Loss 0.187754 Objective Loss 0.187754 LR 0.000250 Time 0.050163 +2023-10-05 22:00:44,875 - Epoch: [167][ 40/ 1236] Overall Loss 0.190680 Objective Loss 0.190680 LR 0.000250 Time 0.042671 +2023-10-05 22:00:45,075 - Epoch: [167][ 50/ 1236] Overall Loss 0.190044 Objective Loss 0.190044 LR 0.000250 Time 0.038132 +2023-10-05 22:00:45,277 - Epoch: [167][ 60/ 1236] Overall Loss 0.194104 Objective Loss 0.194104 LR 0.000250 Time 0.035139 +2023-10-05 22:00:45,477 - Epoch: [167][ 70/ 1236] Overall Loss 0.194005 Objective Loss 0.194005 LR 0.000250 Time 0.032975 +2023-10-05 22:00:45,679 - Epoch: [167][ 80/ 1236] Overall Loss 0.193989 Objective Loss 0.193989 LR 0.000250 Time 0.031378 +2023-10-05 22:00:45,879 - Epoch: [167][ 90/ 1236] Overall Loss 0.194941 Objective Loss 0.194941 LR 0.000250 Time 0.030111 +2023-10-05 22:00:46,082 - Epoch: [167][ 100/ 1236] Overall Loss 0.194701 Objective Loss 0.194701 LR 0.000250 Time 0.029119 +2023-10-05 22:00:46,281 - Epoch: [167][ 110/ 1236] Overall Loss 0.193665 Objective Loss 0.193665 LR 0.000250 Time 0.028286 +2023-10-05 22:00:46,483 - Epoch: [167][ 120/ 1236] Overall Loss 0.194334 Objective Loss 0.194334 LR 0.000250 Time 0.027605 +2023-10-05 22:00:46,682 - Epoch: [167][ 130/ 1236] Overall Loss 0.196105 Objective Loss 0.196105 LR 0.000250 Time 0.027014 +2023-10-05 22:00:46,884 - Epoch: [167][ 140/ 1236] Overall Loss 0.196177 Objective Loss 0.196177 LR 0.000250 Time 0.026519 +2023-10-05 22:00:47,083 - Epoch: [167][ 150/ 1236] Overall Loss 0.195318 Objective Loss 0.195318 LR 0.000250 Time 0.026080 +2023-10-05 22:00:47,285 - Epoch: [167][ 160/ 1236] Overall Loss 0.195168 Objective Loss 0.195168 LR 0.000250 Time 0.025706 +2023-10-05 22:00:47,484 - Epoch: [167][ 170/ 1236] Overall Loss 0.195585 Objective Loss 0.195585 LR 0.000250 Time 0.025367 +2023-10-05 22:00:47,686 - Epoch: [167][ 180/ 1236] Overall Loss 0.195965 Objective Loss 0.195965 LR 0.000250 Time 0.025078 +2023-10-05 22:00:47,886 - Epoch: [167][ 190/ 1236] Overall Loss 0.194799 Objective Loss 0.194799 LR 0.000250 Time 0.024808 +2023-10-05 22:00:48,088 - Epoch: [167][ 200/ 1236] Overall Loss 0.195565 Objective Loss 0.195565 LR 0.000250 Time 0.024573 +2023-10-05 22:00:48,288 - Epoch: [167][ 210/ 1236] Overall Loss 0.194896 Objective Loss 0.194896 LR 0.000250 Time 0.024354 +2023-10-05 22:00:48,490 - Epoch: [167][ 220/ 1236] Overall Loss 0.194174 Objective Loss 0.194174 LR 0.000250 Time 0.024163 +2023-10-05 22:00:48,690 - Epoch: [167][ 230/ 1236] Overall Loss 0.193426 Objective Loss 0.193426 LR 0.000250 Time 0.023982 +2023-10-05 22:00:48,892 - Epoch: [167][ 240/ 1236] Overall Loss 0.192868 Objective Loss 0.192868 LR 0.000250 Time 0.023824 +2023-10-05 22:00:49,092 - Epoch: [167][ 250/ 1236] Overall Loss 0.192261 Objective Loss 0.192261 LR 0.000250 Time 0.023670 +2023-10-05 22:00:49,294 - Epoch: [167][ 260/ 1236] Overall Loss 0.192043 Objective Loss 0.192043 LR 0.000250 Time 0.023535 +2023-10-05 22:00:49,494 - Epoch: [167][ 270/ 1236] Overall Loss 0.192355 Objective Loss 0.192355 LR 0.000250 Time 0.023401 +2023-10-05 22:00:49,696 - Epoch: [167][ 280/ 1236] Overall Loss 0.191917 Objective Loss 0.191917 LR 0.000250 Time 0.023284 +2023-10-05 22:00:49,897 - Epoch: [167][ 290/ 1236] Overall Loss 0.192161 Objective Loss 0.192161 LR 0.000250 Time 0.023173 +2023-10-05 22:00:50,098 - Epoch: [167][ 300/ 1236] Overall Loss 0.191965 Objective Loss 0.191965 LR 0.000250 Time 0.023069 +2023-10-05 22:00:50,298 - Epoch: [167][ 310/ 1236] Overall Loss 0.192330 Objective Loss 0.192330 LR 0.000250 Time 0.022972 +2023-10-05 22:00:50,500 - Epoch: [167][ 320/ 1236] Overall Loss 0.191434 Objective Loss 0.191434 LR 0.000250 Time 0.022883 +2023-10-05 22:00:50,700 - Epoch: [167][ 330/ 1236] Overall Loss 0.191792 Objective Loss 0.191792 LR 0.000250 Time 0.022795 +2023-10-05 22:00:50,902 - Epoch: [167][ 340/ 1236] Overall Loss 0.192133 Objective Loss 0.192133 LR 0.000250 Time 0.022718 +2023-10-05 22:00:51,102 - Epoch: [167][ 350/ 1236] Overall Loss 0.191897 Objective Loss 0.191897 LR 0.000250 Time 0.022639 +2023-10-05 22:00:51,304 - Epoch: [167][ 360/ 1236] Overall Loss 0.191632 Objective Loss 0.191632 LR 0.000250 Time 0.022569 +2023-10-05 22:00:51,505 - Epoch: [167][ 370/ 1236] Overall Loss 0.191526 Objective Loss 0.191526 LR 0.000250 Time 0.022501 +2023-10-05 22:00:51,707 - Epoch: [167][ 380/ 1236] Overall Loss 0.191611 Objective Loss 0.191611 LR 0.000250 Time 0.022439 +2023-10-05 22:00:51,907 - Epoch: [167][ 390/ 1236] Overall Loss 0.191661 Objective Loss 0.191661 LR 0.000250 Time 0.022376 +2023-10-05 22:00:52,108 - Epoch: [167][ 400/ 1236] Overall Loss 0.191643 Objective Loss 0.191643 LR 0.000250 Time 0.022320 +2023-10-05 22:00:52,309 - Epoch: [167][ 410/ 1236] Overall Loss 0.191693 Objective Loss 0.191693 LR 0.000250 Time 0.022264 +2023-10-05 22:00:52,511 - Epoch: [167][ 420/ 1236] Overall Loss 0.191876 Objective Loss 0.191876 LR 0.000250 Time 0.022215 +2023-10-05 22:00:52,712 - Epoch: [167][ 430/ 1236] Overall Loss 0.191421 Objective Loss 0.191421 LR 0.000250 Time 0.022165 +2023-10-05 22:00:52,914 - Epoch: [167][ 440/ 1236] Overall Loss 0.191276 Objective Loss 0.191276 LR 0.000250 Time 0.022120 +2023-10-05 22:00:53,115 - Epoch: [167][ 450/ 1236] Overall Loss 0.191887 Objective Loss 0.191887 LR 0.000250 Time 0.022074 +2023-10-05 22:00:53,318 - Epoch: [167][ 460/ 1236] Overall Loss 0.191655 Objective Loss 0.191655 LR 0.000250 Time 0.022034 +2023-10-05 22:00:53,519 - Epoch: [167][ 470/ 1236] Overall Loss 0.191443 Objective Loss 0.191443 LR 0.000250 Time 0.021992 +2023-10-05 22:00:53,722 - Epoch: [167][ 480/ 1236] Overall Loss 0.191430 Objective Loss 0.191430 LR 0.000250 Time 0.021956 +2023-10-05 22:00:53,923 - Epoch: [167][ 490/ 1236] Overall Loss 0.191614 Objective Loss 0.191614 LR 0.000250 Time 0.021917 +2023-10-05 22:00:54,125 - Epoch: [167][ 500/ 1236] Overall Loss 0.191789 Objective Loss 0.191789 LR 0.000250 Time 0.021883 +2023-10-05 22:00:54,324 - Epoch: [167][ 510/ 1236] Overall Loss 0.191578 Objective Loss 0.191578 LR 0.000250 Time 0.021844 +2023-10-05 22:00:54,525 - Epoch: [167][ 520/ 1236] Overall Loss 0.191691 Objective Loss 0.191691 LR 0.000250 Time 0.021809 +2023-10-05 22:00:54,723 - Epoch: [167][ 530/ 1236] Overall Loss 0.191452 Objective Loss 0.191452 LR 0.000250 Time 0.021771 +2023-10-05 22:00:54,923 - Epoch: [167][ 540/ 1236] Overall Loss 0.191434 Objective Loss 0.191434 LR 0.000250 Time 0.021738 +2023-10-05 22:00:55,122 - Epoch: [167][ 550/ 1236] Overall Loss 0.191356 Objective Loss 0.191356 LR 0.000250 Time 0.021704 +2023-10-05 22:00:55,322 - Epoch: [167][ 560/ 1236] Overall Loss 0.191603 Objective Loss 0.191603 LR 0.000250 Time 0.021673 +2023-10-05 22:00:55,521 - Epoch: [167][ 570/ 1236] Overall Loss 0.191814 Objective Loss 0.191814 LR 0.000250 Time 0.021641 +2023-10-05 22:00:55,721 - Epoch: [167][ 580/ 1236] Overall Loss 0.191757 Objective Loss 0.191757 LR 0.000250 Time 0.021612 +2023-10-05 22:00:55,920 - Epoch: [167][ 590/ 1236] Overall Loss 0.191471 Objective Loss 0.191471 LR 0.000250 Time 0.021583 +2023-10-05 22:00:56,121 - Epoch: [167][ 600/ 1236] Overall Loss 0.191554 Objective Loss 0.191554 LR 0.000250 Time 0.021556 +2023-10-05 22:00:56,319 - Epoch: [167][ 610/ 1236] Overall Loss 0.192009 Objective Loss 0.192009 LR 0.000250 Time 0.021528 +2023-10-05 22:00:56,520 - Epoch: [167][ 620/ 1236] Overall Loss 0.191835 Objective Loss 0.191835 LR 0.000250 Time 0.021504 +2023-10-05 22:00:56,718 - Epoch: [167][ 630/ 1236] Overall Loss 0.191643 Objective Loss 0.191643 LR 0.000250 Time 0.021477 +2023-10-05 22:00:56,919 - Epoch: [167][ 640/ 1236] Overall Loss 0.191318 Objective Loss 0.191318 LR 0.000250 Time 0.021454 +2023-10-05 22:00:57,117 - Epoch: [167][ 650/ 1236] Overall Loss 0.191108 Objective Loss 0.191108 LR 0.000250 Time 0.021429 +2023-10-05 22:00:57,317 - Epoch: [167][ 660/ 1236] Overall Loss 0.190796 Objective Loss 0.190796 LR 0.000250 Time 0.021407 +2023-10-05 22:00:57,516 - Epoch: [167][ 670/ 1236] Overall Loss 0.190736 Objective Loss 0.190736 LR 0.000250 Time 0.021383 +2023-10-05 22:00:57,716 - Epoch: [167][ 680/ 1236] Overall Loss 0.190682 Objective Loss 0.190682 LR 0.000250 Time 0.021362 +2023-10-05 22:00:57,915 - Epoch: [167][ 690/ 1236] Overall Loss 0.190598 Objective Loss 0.190598 LR 0.000250 Time 0.021341 +2023-10-05 22:00:58,115 - Epoch: [167][ 700/ 1236] Overall Loss 0.190661 Objective Loss 0.190661 LR 0.000250 Time 0.021322 +2023-10-05 22:00:58,314 - Epoch: [167][ 710/ 1236] Overall Loss 0.190777 Objective Loss 0.190777 LR 0.000250 Time 0.021301 +2023-10-05 22:00:58,515 - Epoch: [167][ 720/ 1236] Overall Loss 0.191187 Objective Loss 0.191187 LR 0.000250 Time 0.021284 +2023-10-05 22:00:58,714 - Epoch: [167][ 730/ 1236] Overall Loss 0.191412 Objective Loss 0.191412 LR 0.000250 Time 0.021265 +2023-10-05 22:00:58,915 - Epoch: [167][ 740/ 1236] Overall Loss 0.191408 Objective Loss 0.191408 LR 0.000250 Time 0.021248 +2023-10-05 22:00:59,113 - Epoch: [167][ 750/ 1236] Overall Loss 0.191545 Objective Loss 0.191545 LR 0.000250 Time 0.021229 +2023-10-05 22:00:59,313 - Epoch: [167][ 760/ 1236] Overall Loss 0.191639 Objective Loss 0.191639 LR 0.000250 Time 0.021212 +2023-10-05 22:00:59,512 - Epoch: [167][ 770/ 1236] Overall Loss 0.191300 Objective Loss 0.191300 LR 0.000250 Time 0.021195 +2023-10-05 22:00:59,713 - Epoch: [167][ 780/ 1236] Overall Loss 0.191431 Objective Loss 0.191431 LR 0.000250 Time 0.021180 +2023-10-05 22:00:59,911 - Epoch: [167][ 790/ 1236] Overall Loss 0.191018 Objective Loss 0.191018 LR 0.000250 Time 0.021162 +2023-10-05 22:01:00,112 - Epoch: [167][ 800/ 1236] Overall Loss 0.191135 Objective Loss 0.191135 LR 0.000250 Time 0.021148 +2023-10-05 22:01:00,310 - Epoch: [167][ 810/ 1236] Overall Loss 0.191103 Objective Loss 0.191103 LR 0.000250 Time 0.021132 +2023-10-05 22:01:00,511 - Epoch: [167][ 820/ 1236] Overall Loss 0.191216 Objective Loss 0.191216 LR 0.000250 Time 0.021118 +2023-10-05 22:01:00,709 - Epoch: [167][ 830/ 1236] Overall Loss 0.191108 Objective Loss 0.191108 LR 0.000250 Time 0.021102 +2023-10-05 22:01:00,909 - Epoch: [167][ 840/ 1236] Overall Loss 0.191158 Objective Loss 0.191158 LR 0.000250 Time 0.021089 +2023-10-05 22:01:01,108 - Epoch: [167][ 850/ 1236] Overall Loss 0.190989 Objective Loss 0.190989 LR 0.000250 Time 0.021074 +2023-10-05 22:01:01,308 - Epoch: [167][ 860/ 1236] Overall Loss 0.191096 Objective Loss 0.191096 LR 0.000250 Time 0.021061 +2023-10-05 22:01:01,507 - Epoch: [167][ 870/ 1236] Overall Loss 0.190949 Objective Loss 0.190949 LR 0.000250 Time 0.021048 +2023-10-05 22:01:01,707 - Epoch: [167][ 880/ 1236] Overall Loss 0.190779 Objective Loss 0.190779 LR 0.000250 Time 0.021035 +2023-10-05 22:01:01,906 - Epoch: [167][ 890/ 1236] Overall Loss 0.190676 Objective Loss 0.190676 LR 0.000250 Time 0.021022 +2023-10-05 22:01:02,106 - Epoch: [167][ 900/ 1236] Overall Loss 0.190894 Objective Loss 0.190894 LR 0.000250 Time 0.021010 +2023-10-05 22:01:02,304 - Epoch: [167][ 910/ 1236] Overall Loss 0.191065 Objective Loss 0.191065 LR 0.000250 Time 0.020997 +2023-10-05 22:01:02,505 - Epoch: [167][ 920/ 1236] Overall Loss 0.191004 Objective Loss 0.191004 LR 0.000250 Time 0.020986 +2023-10-05 22:01:02,703 - Epoch: [167][ 930/ 1236] Overall Loss 0.191260 Objective Loss 0.191260 LR 0.000250 Time 0.020974 +2023-10-05 22:01:02,904 - Epoch: [167][ 940/ 1236] Overall Loss 0.191602 Objective Loss 0.191602 LR 0.000250 Time 0.020964 +2023-10-05 22:01:03,102 - Epoch: [167][ 950/ 1236] Overall Loss 0.191503 Objective Loss 0.191503 LR 0.000250 Time 0.020952 +2023-10-05 22:01:03,303 - Epoch: [167][ 960/ 1236] Overall Loss 0.191287 Objective Loss 0.191287 LR 0.000250 Time 0.020942 +2023-10-05 22:01:03,501 - Epoch: [167][ 970/ 1236] Overall Loss 0.191322 Objective Loss 0.191322 LR 0.000250 Time 0.020930 +2023-10-05 22:01:03,702 - Epoch: [167][ 980/ 1236] Overall Loss 0.191355 Objective Loss 0.191355 LR 0.000250 Time 0.020921 +2023-10-05 22:01:03,901 - Epoch: [167][ 990/ 1236] Overall Loss 0.191379 Objective Loss 0.191379 LR 0.000250 Time 0.020910 +2023-10-05 22:01:04,101 - Epoch: [167][ 1000/ 1236] Overall Loss 0.191301 Objective Loss 0.191301 LR 0.000250 Time 0.020901 +2023-10-05 22:01:04,300 - Epoch: [167][ 1010/ 1236] Overall Loss 0.191483 Objective Loss 0.191483 LR 0.000250 Time 0.020891 +2023-10-05 22:01:04,500 - Epoch: [167][ 1020/ 1236] Overall Loss 0.191548 Objective Loss 0.191548 LR 0.000250 Time 0.020882 +2023-10-05 22:01:04,699 - Epoch: [167][ 1030/ 1236] Overall Loss 0.191906 Objective Loss 0.191906 LR 0.000250 Time 0.020871 +2023-10-05 22:01:04,899 - Epoch: [167][ 1040/ 1236] Overall Loss 0.191806 Objective Loss 0.191806 LR 0.000250 Time 0.020863 +2023-10-05 22:01:05,098 - Epoch: [167][ 1050/ 1236] Overall Loss 0.191670 Objective Loss 0.191670 LR 0.000250 Time 0.020853 +2023-10-05 22:01:05,298 - Epoch: [167][ 1060/ 1236] Overall Loss 0.191640 Objective Loss 0.191640 LR 0.000250 Time 0.020845 +2023-10-05 22:01:05,496 - Epoch: [167][ 1070/ 1236] Overall Loss 0.191928 Objective Loss 0.191928 LR 0.000250 Time 0.020836 +2023-10-05 22:01:05,697 - Epoch: [167][ 1080/ 1236] Overall Loss 0.191949 Objective Loss 0.191949 LR 0.000250 Time 0.020828 +2023-10-05 22:01:05,896 - Epoch: [167][ 1090/ 1236] Overall Loss 0.191823 Objective Loss 0.191823 LR 0.000250 Time 0.020819 +2023-10-05 22:01:06,096 - Epoch: [167][ 1100/ 1236] Overall Loss 0.191904 Objective Loss 0.191904 LR 0.000250 Time 0.020811 +2023-10-05 22:01:06,295 - Epoch: [167][ 1110/ 1236] Overall Loss 0.192000 Objective Loss 0.192000 LR 0.000250 Time 0.020803 +2023-10-05 22:01:06,495 - Epoch: [167][ 1120/ 1236] Overall Loss 0.192180 Objective Loss 0.192180 LR 0.000250 Time 0.020795 +2023-10-05 22:01:06,694 - Epoch: [167][ 1130/ 1236] Overall Loss 0.192191 Objective Loss 0.192191 LR 0.000250 Time 0.020787 +2023-10-05 22:01:06,894 - Epoch: [167][ 1140/ 1236] Overall Loss 0.192323 Objective Loss 0.192323 LR 0.000250 Time 0.020780 +2023-10-05 22:01:07,093 - Epoch: [167][ 1150/ 1236] Overall Loss 0.192321 Objective Loss 0.192321 LR 0.000250 Time 0.020772 +2023-10-05 22:01:07,293 - Epoch: [167][ 1160/ 1236] Overall Loss 0.192418 Objective Loss 0.192418 LR 0.000250 Time 0.020765 +2023-10-05 22:01:07,492 - Epoch: [167][ 1170/ 1236] Overall Loss 0.192407 Objective Loss 0.192407 LR 0.000250 Time 0.020757 +2023-10-05 22:01:07,692 - Epoch: [167][ 1180/ 1236] Overall Loss 0.192261 Objective Loss 0.192261 LR 0.000250 Time 0.020751 +2023-10-05 22:01:07,891 - Epoch: [167][ 1190/ 1236] Overall Loss 0.192302 Objective Loss 0.192302 LR 0.000250 Time 0.020743 +2023-10-05 22:01:08,091 - Epoch: [167][ 1200/ 1236] Overall Loss 0.192366 Objective Loss 0.192366 LR 0.000250 Time 0.020737 +2023-10-05 22:01:08,290 - Epoch: [167][ 1210/ 1236] Overall Loss 0.192359 Objective Loss 0.192359 LR 0.000250 Time 0.020729 +2023-10-05 22:01:08,490 - Epoch: [167][ 1220/ 1236] Overall Loss 0.192340 Objective Loss 0.192340 LR 0.000250 Time 0.020723 +2023-10-05 22:01:08,742 - Epoch: [167][ 1230/ 1236] Overall Loss 0.192257 Objective Loss 0.192257 LR 0.000250 Time 0.020760 +2023-10-05 22:01:08,860 - Epoch: [167][ 1236/ 1236] Overall Loss 0.192173 Objective Loss 0.192173 Top1 88.798371 Top5 97.352342 LR 0.000250 Time 0.020754 +2023-10-05 22:01:08,998 - --- validate (epoch=167)----------- +2023-10-05 22:01:08,999 - 29943 samples (256 per mini-batch) +2023-10-05 22:01:09,454 - Epoch: [167][ 10/ 117] Loss 0.316005 Top1 84.765625 Top5 97.929688 +2023-10-05 22:01:09,605 - Epoch: [167][ 20/ 117] Loss 0.305413 Top1 85.214844 Top5 98.027344 +2023-10-05 22:01:09,753 - Epoch: [167][ 30/ 117] Loss 0.306480 Top1 85.273438 Top5 98.059896 +2023-10-05 22:01:09,901 - Epoch: [167][ 40/ 117] Loss 0.308811 Top1 85.292969 Top5 98.046875 +2023-10-05 22:01:10,048 - Epoch: [167][ 50/ 117] Loss 0.310419 Top1 85.382812 Top5 98.023438 +2023-10-05 22:01:10,198 - Epoch: [167][ 60/ 117] Loss 0.309290 Top1 85.266927 Top5 98.001302 +2023-10-05 22:01:10,346 - Epoch: [167][ 70/ 117] Loss 0.311431 Top1 85.189732 Top5 98.007812 +2023-10-05 22:01:10,494 - Epoch: [167][ 80/ 117] Loss 0.309187 Top1 85.209961 Top5 98.002930 +2023-10-05 22:01:10,641 - Epoch: [167][ 90/ 117] Loss 0.307341 Top1 85.221354 Top5 98.051215 +2023-10-05 22:01:10,791 - Epoch: [167][ 100/ 117] Loss 0.307447 Top1 85.164062 Top5 98.078125 +2023-10-05 22:01:10,946 - Epoch: [167][ 110/ 117] Loss 0.308793 Top1 85.205966 Top5 98.128551 +2023-10-05 22:01:11,030 - Epoch: [167][ 117/ 117] Loss 0.308183 Top1 85.185185 Top5 98.119761 +2023-10-05 22:01:11,163 - ==> Top1: 85.185 Top5: 98.120 Loss: 0.308 + +2023-10-05 22:01:11,164 - ==> Confusion: +[[ 933 2 6 1 6 2 0 0 5 65 2 1 1 1 6 1 3 1 1 0 13] + [ 0 1060 2 0 8 17 1 17 0 0 2 3 0 0 0 2 2 0 9 0 8] + [ 9 2 963 11 3 1 24 10 0 1 3 1 5 1 0 3 0 2 7 3 7] + [ 2 1 9 973 1 5 0 2 1 1 6 0 3 2 26 4 0 5 28 2 18] + [ 25 7 1 0 969 2 0 1 0 11 0 3 0 3 6 2 11 2 0 1 6] + [ 4 31 2 0 2 994 2 23 0 0 3 7 0 14 5 2 5 0 7 2 13] + [ 0 6 19 1 0 0 1130 11 0 0 0 1 0 0 1 3 1 1 3 8 6] + [ 5 14 11 0 3 24 2 1085 0 2 0 9 1 3 1 2 0 0 43 3 10] + [ 18 4 1 0 0 2 0 2 975 35 9 1 2 11 15 5 2 0 3 1 3] + [ 105 0 3 0 5 2 0 0 17 950 1 2 0 18 3 6 0 2 0 0 5] + [ 3 6 10 5 2 1 2 3 13 0 965 2 0 13 3 2 3 0 7 1 12] + [ 1 0 1 1 1 13 0 4 0 1 0 948 26 4 0 4 3 18 1 6 3] + [ 2 1 1 5 0 0 0 1 0 0 1 38 984 1 0 4 3 11 4 3 9] + [ 3 0 1 0 1 6 0 1 9 12 7 5 0 1058 4 1 0 0 0 0 11] + [ 13 1 3 10 6 0 1 0 26 1 1 1 3 3 1007 0 1 2 14 0 8] + [ 1 4 1 0 2 0 1 0 0 0 0 9 6 1 1 1068 19 10 0 9 2] + [ 1 9 2 0 7 4 0 0 0 0 0 2 0 0 3 9 1108 0 1 3 12] + [ 0 0 0 2 1 0 2 0 0 0 0 1 18 0 1 6 0 1002 1 0 4] + [ 1 6 6 15 1 0 0 20 2 0 2 0 3 0 4 0 1 0 998 1 8] + [ 0 2 3 1 2 6 10 10 0 1 1 16 3 0 0 6 8 2 3 1068 10] + [ 133 178 130 55 95 123 27 107 101 78 157 102 319 284 155 53 167 63 141 168 5269]] + +2023-10-05 22:01:11,165 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:01:11,165 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:01:11,171 - + +2023-10-05 22:01:11,171 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:01:12,157 - Epoch: [168][ 10/ 1236] Overall Loss 0.204747 Objective Loss 0.204747 LR 0.000250 Time 0.098473 +2023-10-05 22:01:12,357 - Epoch: [168][ 20/ 1236] Overall Loss 0.198266 Objective Loss 0.198266 LR 0.000250 Time 0.059222 +2023-10-05 22:01:12,555 - Epoch: [168][ 30/ 1236] Overall Loss 0.195666 Objective Loss 0.195666 LR 0.000250 Time 0.046091 +2023-10-05 22:01:12,754 - Epoch: [168][ 40/ 1236] Overall Loss 0.198210 Objective Loss 0.198210 LR 0.000250 Time 0.039540 +2023-10-05 22:01:12,953 - Epoch: [168][ 50/ 1236] Overall Loss 0.196635 Objective Loss 0.196635 LR 0.000250 Time 0.035597 +2023-10-05 22:01:13,152 - Epoch: [168][ 60/ 1236] Overall Loss 0.194278 Objective Loss 0.194278 LR 0.000250 Time 0.032982 +2023-10-05 22:01:13,351 - Epoch: [168][ 70/ 1236] Overall Loss 0.191554 Objective Loss 0.191554 LR 0.000250 Time 0.031106 +2023-10-05 22:01:13,551 - Epoch: [168][ 80/ 1236] Overall Loss 0.190580 Objective Loss 0.190580 LR 0.000250 Time 0.029714 +2023-10-05 22:01:13,750 - Epoch: [168][ 90/ 1236] Overall Loss 0.190314 Objective Loss 0.190314 LR 0.000250 Time 0.028614 +2023-10-05 22:01:13,949 - Epoch: [168][ 100/ 1236] Overall Loss 0.191736 Objective Loss 0.191736 LR 0.000250 Time 0.027746 +2023-10-05 22:01:14,148 - Epoch: [168][ 110/ 1236] Overall Loss 0.191389 Objective Loss 0.191389 LR 0.000250 Time 0.027026 +2023-10-05 22:01:14,348 - Epoch: [168][ 120/ 1236] Overall Loss 0.192191 Objective Loss 0.192191 LR 0.000250 Time 0.026437 +2023-10-05 22:01:14,547 - Epoch: [168][ 130/ 1236] Overall Loss 0.194058 Objective Loss 0.194058 LR 0.000250 Time 0.025929 +2023-10-05 22:01:14,746 - Epoch: [168][ 140/ 1236] Overall Loss 0.193674 Objective Loss 0.193674 LR 0.000250 Time 0.025500 +2023-10-05 22:01:14,945 - Epoch: [168][ 150/ 1236] Overall Loss 0.192510 Objective Loss 0.192510 LR 0.000250 Time 0.025122 +2023-10-05 22:01:15,145 - Epoch: [168][ 160/ 1236] Overall Loss 0.193496 Objective Loss 0.193496 LR 0.000250 Time 0.024798 +2023-10-05 22:01:15,343 - Epoch: [168][ 170/ 1236] Overall Loss 0.193147 Objective Loss 0.193147 LR 0.000250 Time 0.024505 +2023-10-05 22:01:15,543 - Epoch: [168][ 180/ 1236] Overall Loss 0.193925 Objective Loss 0.193925 LR 0.000250 Time 0.024251 +2023-10-05 22:01:15,741 - Epoch: [168][ 190/ 1236] Overall Loss 0.193364 Objective Loss 0.193364 LR 0.000250 Time 0.024018 +2023-10-05 22:01:15,941 - Epoch: [168][ 200/ 1236] Overall Loss 0.192921 Objective Loss 0.192921 LR 0.000250 Time 0.023814 +2023-10-05 22:01:16,140 - Epoch: [168][ 210/ 1236] Overall Loss 0.193690 Objective Loss 0.193690 LR 0.000250 Time 0.023624 +2023-10-05 22:01:16,340 - Epoch: [168][ 220/ 1236] Overall Loss 0.192579 Objective Loss 0.192579 LR 0.000250 Time 0.023457 +2023-10-05 22:01:16,538 - Epoch: [168][ 230/ 1236] Overall Loss 0.192036 Objective Loss 0.192036 LR 0.000250 Time 0.023298 +2023-10-05 22:01:16,738 - Epoch: [168][ 240/ 1236] Overall Loss 0.190858 Objective Loss 0.190858 LR 0.000250 Time 0.023159 +2023-10-05 22:01:16,936 - Epoch: [168][ 250/ 1236] Overall Loss 0.191634 Objective Loss 0.191634 LR 0.000250 Time 0.023025 +2023-10-05 22:01:17,136 - Epoch: [168][ 260/ 1236] Overall Loss 0.191928 Objective Loss 0.191928 LR 0.000250 Time 0.022905 +2023-10-05 22:01:17,334 - Epoch: [168][ 270/ 1236] Overall Loss 0.192721 Objective Loss 0.192721 LR 0.000250 Time 0.022791 +2023-10-05 22:01:17,534 - Epoch: [168][ 280/ 1236] Overall Loss 0.192501 Objective Loss 0.192501 LR 0.000250 Time 0.022689 +2023-10-05 22:01:17,733 - Epoch: [168][ 290/ 1236] Overall Loss 0.192394 Objective Loss 0.192394 LR 0.000250 Time 0.022590 +2023-10-05 22:01:17,933 - Epoch: [168][ 300/ 1236] Overall Loss 0.193158 Objective Loss 0.193158 LR 0.000250 Time 0.022502 +2023-10-05 22:01:18,132 - Epoch: [168][ 310/ 1236] Overall Loss 0.193608 Objective Loss 0.193608 LR 0.000250 Time 0.022417 +2023-10-05 22:01:18,331 - Epoch: [168][ 320/ 1236] Overall Loss 0.193257 Objective Loss 0.193257 LR 0.000250 Time 0.022339 +2023-10-05 22:01:18,530 - Epoch: [168][ 330/ 1236] Overall Loss 0.193171 Objective Loss 0.193171 LR 0.000250 Time 0.022263 +2023-10-05 22:01:18,730 - Epoch: [168][ 340/ 1236] Overall Loss 0.193418 Objective Loss 0.193418 LR 0.000250 Time 0.022195 +2023-10-05 22:01:18,929 - Epoch: [168][ 350/ 1236] Overall Loss 0.193650 Objective Loss 0.193650 LR 0.000250 Time 0.022128 +2023-10-05 22:01:19,128 - Epoch: [168][ 360/ 1236] Overall Loss 0.193712 Objective Loss 0.193712 LR 0.000250 Time 0.022067 +2023-10-05 22:01:19,327 - Epoch: [168][ 370/ 1236] Overall Loss 0.193415 Objective Loss 0.193415 LR 0.000250 Time 0.022008 +2023-10-05 22:01:19,527 - Epoch: [168][ 380/ 1236] Overall Loss 0.192988 Objective Loss 0.192988 LR 0.000250 Time 0.021954 +2023-10-05 22:01:19,727 - Epoch: [168][ 390/ 1236] Overall Loss 0.192857 Objective Loss 0.192857 LR 0.000250 Time 0.021901 +2023-10-05 22:01:19,927 - Epoch: [168][ 400/ 1236] Overall Loss 0.192506 Objective Loss 0.192506 LR 0.000250 Time 0.021853 +2023-10-05 22:01:20,126 - Epoch: [168][ 410/ 1236] Overall Loss 0.193516 Objective Loss 0.193516 LR 0.000250 Time 0.021805 +2023-10-05 22:01:20,326 - Epoch: [168][ 420/ 1236] Overall Loss 0.193434 Objective Loss 0.193434 LR 0.000250 Time 0.021761 +2023-10-05 22:01:20,524 - Epoch: [168][ 430/ 1236] Overall Loss 0.193723 Objective Loss 0.193723 LR 0.000250 Time 0.021716 +2023-10-05 22:01:20,724 - Epoch: [168][ 440/ 1236] Overall Loss 0.193603 Objective Loss 0.193603 LR 0.000250 Time 0.021676 +2023-10-05 22:01:20,923 - Epoch: [168][ 450/ 1236] Overall Loss 0.194109 Objective Loss 0.194109 LR 0.000250 Time 0.021635 +2023-10-05 22:01:21,123 - Epoch: [168][ 460/ 1236] Overall Loss 0.194223 Objective Loss 0.194223 LR 0.000250 Time 0.021598 +2023-10-05 22:01:21,322 - Epoch: [168][ 470/ 1236] Overall Loss 0.194330 Objective Loss 0.194330 LR 0.000250 Time 0.021561 +2023-10-05 22:01:21,521 - Epoch: [168][ 480/ 1236] Overall Loss 0.194478 Objective Loss 0.194478 LR 0.000250 Time 0.021528 +2023-10-05 22:01:21,721 - Epoch: [168][ 490/ 1236] Overall Loss 0.194365 Objective Loss 0.194365 LR 0.000250 Time 0.021494 +2023-10-05 22:01:21,921 - Epoch: [168][ 500/ 1236] Overall Loss 0.194317 Objective Loss 0.194317 LR 0.000250 Time 0.021464 +2023-10-05 22:01:22,120 - Epoch: [168][ 510/ 1236] Overall Loss 0.194155 Objective Loss 0.194155 LR 0.000250 Time 0.021432 +2023-10-05 22:01:22,319 - Epoch: [168][ 520/ 1236] Overall Loss 0.194339 Objective Loss 0.194339 LR 0.000250 Time 0.021404 +2023-10-05 22:01:22,518 - Epoch: [168][ 530/ 1236] Overall Loss 0.194239 Objective Loss 0.194239 LR 0.000250 Time 0.021375 +2023-10-05 22:01:22,718 - Epoch: [168][ 540/ 1236] Overall Loss 0.193943 Objective Loss 0.193943 LR 0.000250 Time 0.021349 +2023-10-05 22:01:22,917 - Epoch: [168][ 550/ 1236] Overall Loss 0.193688 Objective Loss 0.193688 LR 0.000250 Time 0.021322 +2023-10-05 22:01:23,118 - Epoch: [168][ 560/ 1236] Overall Loss 0.193450 Objective Loss 0.193450 LR 0.000250 Time 0.021298 +2023-10-05 22:01:23,316 - Epoch: [168][ 570/ 1236] Overall Loss 0.193269 Objective Loss 0.193269 LR 0.000250 Time 0.021273 +2023-10-05 22:01:23,516 - Epoch: [168][ 580/ 1236] Overall Loss 0.193312 Objective Loss 0.193312 LR 0.000250 Time 0.021250 +2023-10-05 22:01:23,715 - Epoch: [168][ 590/ 1236] Overall Loss 0.193088 Objective Loss 0.193088 LR 0.000250 Time 0.021227 +2023-10-05 22:01:23,915 - Epoch: [168][ 600/ 1236] Overall Loss 0.193048 Objective Loss 0.193048 LR 0.000250 Time 0.021205 +2023-10-05 22:01:24,114 - Epoch: [168][ 610/ 1236] Overall Loss 0.193148 Objective Loss 0.193148 LR 0.000250 Time 0.021184 +2023-10-05 22:01:24,314 - Epoch: [168][ 620/ 1236] Overall Loss 0.192825 Objective Loss 0.192825 LR 0.000250 Time 0.021163 +2023-10-05 22:01:24,513 - Epoch: [168][ 630/ 1236] Overall Loss 0.192282 Objective Loss 0.192282 LR 0.000250 Time 0.021142 +2023-10-05 22:01:24,712 - Epoch: [168][ 640/ 1236] Overall Loss 0.192133 Objective Loss 0.192133 LR 0.000250 Time 0.021124 +2023-10-05 22:01:24,911 - Epoch: [168][ 650/ 1236] Overall Loss 0.191845 Objective Loss 0.191845 LR 0.000250 Time 0.021104 +2023-10-05 22:01:25,111 - Epoch: [168][ 660/ 1236] Overall Loss 0.192318 Objective Loss 0.192318 LR 0.000250 Time 0.021087 +2023-10-05 22:01:25,310 - Epoch: [168][ 670/ 1236] Overall Loss 0.192466 Objective Loss 0.192466 LR 0.000250 Time 0.021069 +2023-10-05 22:01:25,510 - Epoch: [168][ 680/ 1236] Overall Loss 0.192611 Objective Loss 0.192611 LR 0.000250 Time 0.021053 +2023-10-05 22:01:25,709 - Epoch: [168][ 690/ 1236] Overall Loss 0.192871 Objective Loss 0.192871 LR 0.000250 Time 0.021035 +2023-10-05 22:01:25,909 - Epoch: [168][ 700/ 1236] Overall Loss 0.192833 Objective Loss 0.192833 LR 0.000250 Time 0.021020 +2023-10-05 22:01:26,108 - Epoch: [168][ 710/ 1236] Overall Loss 0.193052 Objective Loss 0.193052 LR 0.000250 Time 0.021003 +2023-10-05 22:01:26,308 - Epoch: [168][ 720/ 1236] Overall Loss 0.193419 Objective Loss 0.193419 LR 0.000250 Time 0.020989 +2023-10-05 22:01:26,507 - Epoch: [168][ 730/ 1236] Overall Loss 0.193192 Objective Loss 0.193192 LR 0.000250 Time 0.020973 +2023-10-05 22:01:26,707 - Epoch: [168][ 740/ 1236] Overall Loss 0.193211 Objective Loss 0.193211 LR 0.000250 Time 0.020960 +2023-10-05 22:01:26,906 - Epoch: [168][ 750/ 1236] Overall Loss 0.193315 Objective Loss 0.193315 LR 0.000250 Time 0.020945 +2023-10-05 22:01:27,105 - Epoch: [168][ 760/ 1236] Overall Loss 0.193188 Objective Loss 0.193188 LR 0.000250 Time 0.020932 +2023-10-05 22:01:27,304 - Epoch: [168][ 770/ 1236] Overall Loss 0.193100 Objective Loss 0.193100 LR 0.000250 Time 0.020918 +2023-10-05 22:01:27,504 - Epoch: [168][ 780/ 1236] Overall Loss 0.192963 Objective Loss 0.192963 LR 0.000250 Time 0.020906 +2023-10-05 22:01:27,704 - Epoch: [168][ 790/ 1236] Overall Loss 0.193064 Objective Loss 0.193064 LR 0.000250 Time 0.020893 +2023-10-05 22:01:27,904 - Epoch: [168][ 800/ 1236] Overall Loss 0.192914 Objective Loss 0.192914 LR 0.000250 Time 0.020882 +2023-10-05 22:01:28,103 - Epoch: [168][ 810/ 1236] Overall Loss 0.193038 Objective Loss 0.193038 LR 0.000250 Time 0.020869 +2023-10-05 22:01:28,303 - Epoch: [168][ 820/ 1236] Overall Loss 0.192862 Objective Loss 0.192862 LR 0.000250 Time 0.020858 +2023-10-05 22:01:28,502 - Epoch: [168][ 830/ 1236] Overall Loss 0.192959 Objective Loss 0.192959 LR 0.000250 Time 0.020846 +2023-10-05 22:01:28,701 - Epoch: [168][ 840/ 1236] Overall Loss 0.192939 Objective Loss 0.192939 LR 0.000250 Time 0.020835 +2023-10-05 22:01:28,901 - Epoch: [168][ 850/ 1236] Overall Loss 0.192844 Objective Loss 0.192844 LR 0.000250 Time 0.020824 +2023-10-05 22:01:29,101 - Epoch: [168][ 860/ 1236] Overall Loss 0.192964 Objective Loss 0.192964 LR 0.000250 Time 0.020814 +2023-10-05 22:01:29,300 - Epoch: [168][ 870/ 1236] Overall Loss 0.192836 Objective Loss 0.192836 LR 0.000250 Time 0.020803 +2023-10-05 22:01:29,500 - Epoch: [168][ 880/ 1236] Overall Loss 0.192859 Objective Loss 0.192859 LR 0.000250 Time 0.020794 +2023-10-05 22:01:29,699 - Epoch: [168][ 890/ 1236] Overall Loss 0.192742 Objective Loss 0.192742 LR 0.000250 Time 0.020784 +2023-10-05 22:01:29,899 - Epoch: [168][ 900/ 1236] Overall Loss 0.192601 Objective Loss 0.192601 LR 0.000250 Time 0.020775 +2023-10-05 22:01:30,098 - Epoch: [168][ 910/ 1236] Overall Loss 0.192514 Objective Loss 0.192514 LR 0.000250 Time 0.020764 +2023-10-05 22:01:30,298 - Epoch: [168][ 920/ 1236] Overall Loss 0.192345 Objective Loss 0.192345 LR 0.000250 Time 0.020756 +2023-10-05 22:01:30,497 - Epoch: [168][ 930/ 1236] Overall Loss 0.192197 Objective Loss 0.192197 LR 0.000250 Time 0.020746 +2023-10-05 22:01:30,696 - Epoch: [168][ 940/ 1236] Overall Loss 0.192216 Objective Loss 0.192216 LR 0.000250 Time 0.020738 +2023-10-05 22:01:30,895 - Epoch: [168][ 950/ 1236] Overall Loss 0.192162 Objective Loss 0.192162 LR 0.000250 Time 0.020728 +2023-10-05 22:01:31,095 - Epoch: [168][ 960/ 1236] Overall Loss 0.192091 Objective Loss 0.192091 LR 0.000250 Time 0.020720 +2023-10-05 22:01:31,293 - Epoch: [168][ 970/ 1236] Overall Loss 0.192020 Objective Loss 0.192020 LR 0.000250 Time 0.020710 +2023-10-05 22:01:31,492 - Epoch: [168][ 980/ 1236] Overall Loss 0.191926 Objective Loss 0.191926 LR 0.000250 Time 0.020702 +2023-10-05 22:01:31,691 - Epoch: [168][ 990/ 1236] Overall Loss 0.191694 Objective Loss 0.191694 LR 0.000250 Time 0.020694 +2023-10-05 22:01:31,891 - Epoch: [168][ 1000/ 1236] Overall Loss 0.191867 Objective Loss 0.191867 LR 0.000250 Time 0.020686 +2023-10-05 22:01:32,090 - Epoch: [168][ 1010/ 1236] Overall Loss 0.191887 Objective Loss 0.191887 LR 0.000250 Time 0.020678 +2023-10-05 22:01:32,290 - Epoch: [168][ 1020/ 1236] Overall Loss 0.191640 Objective Loss 0.191640 LR 0.000250 Time 0.020671 +2023-10-05 22:01:32,488 - Epoch: [168][ 1030/ 1236] Overall Loss 0.191300 Objective Loss 0.191300 LR 0.000250 Time 0.020662 +2023-10-05 22:01:32,688 - Epoch: [168][ 1040/ 1236] Overall Loss 0.191308 Objective Loss 0.191308 LR 0.000250 Time 0.020656 +2023-10-05 22:01:32,887 - Epoch: [168][ 1050/ 1236] Overall Loss 0.191454 Objective Loss 0.191454 LR 0.000250 Time 0.020648 +2023-10-05 22:01:33,087 - Epoch: [168][ 1060/ 1236] Overall Loss 0.191598 Objective Loss 0.191598 LR 0.000250 Time 0.020642 +2023-10-05 22:01:33,286 - Epoch: [168][ 1070/ 1236] Overall Loss 0.191470 Objective Loss 0.191470 LR 0.000250 Time 0.020635 +2023-10-05 22:01:33,487 - Epoch: [168][ 1080/ 1236] Overall Loss 0.191527 Objective Loss 0.191527 LR 0.000250 Time 0.020629 +2023-10-05 22:01:33,686 - Epoch: [168][ 1090/ 1236] Overall Loss 0.191531 Objective Loss 0.191531 LR 0.000250 Time 0.020622 +2023-10-05 22:01:33,887 - Epoch: [168][ 1100/ 1236] Overall Loss 0.191717 Objective Loss 0.191717 LR 0.000250 Time 0.020617 +2023-10-05 22:01:34,085 - Epoch: [168][ 1110/ 1236] Overall Loss 0.191791 Objective Loss 0.191791 LR 0.000250 Time 0.020610 +2023-10-05 22:01:34,286 - Epoch: [168][ 1120/ 1236] Overall Loss 0.191868 Objective Loss 0.191868 LR 0.000250 Time 0.020604 +2023-10-05 22:01:34,485 - Epoch: [168][ 1130/ 1236] Overall Loss 0.191731 Objective Loss 0.191731 LR 0.000250 Time 0.020598 +2023-10-05 22:01:34,685 - Epoch: [168][ 1140/ 1236] Overall Loss 0.191641 Objective Loss 0.191641 LR 0.000250 Time 0.020592 +2023-10-05 22:01:34,884 - Epoch: [168][ 1150/ 1236] Overall Loss 0.191695 Objective Loss 0.191695 LR 0.000250 Time 0.020586 +2023-10-05 22:01:35,084 - Epoch: [168][ 1160/ 1236] Overall Loss 0.191755 Objective Loss 0.191755 LR 0.000250 Time 0.020581 +2023-10-05 22:01:35,284 - Epoch: [168][ 1170/ 1236] Overall Loss 0.191665 Objective Loss 0.191665 LR 0.000250 Time 0.020575 +2023-10-05 22:01:35,484 - Epoch: [168][ 1180/ 1236] Overall Loss 0.191623 Objective Loss 0.191623 LR 0.000250 Time 0.020571 +2023-10-05 22:01:35,683 - Epoch: [168][ 1190/ 1236] Overall Loss 0.191778 Objective Loss 0.191778 LR 0.000250 Time 0.020565 +2023-10-05 22:01:35,884 - Epoch: [168][ 1200/ 1236] Overall Loss 0.191592 Objective Loss 0.191592 LR 0.000250 Time 0.020560 +2023-10-05 22:01:36,082 - Epoch: [168][ 1210/ 1236] Overall Loss 0.191597 Objective Loss 0.191597 LR 0.000250 Time 0.020554 +2023-10-05 22:01:36,282 - Epoch: [168][ 1220/ 1236] Overall Loss 0.191688 Objective Loss 0.191688 LR 0.000250 Time 0.020549 +2023-10-05 22:01:36,536 - Epoch: [168][ 1230/ 1236] Overall Loss 0.191652 Objective Loss 0.191652 LR 0.000250 Time 0.020588 +2023-10-05 22:01:36,653 - Epoch: [168][ 1236/ 1236] Overall Loss 0.191773 Objective Loss 0.191773 Top1 85.947047 Top5 98.167006 LR 0.000250 Time 0.020583 +2023-10-05 22:01:36,780 - --- validate (epoch=168)----------- +2023-10-05 22:01:36,780 - 29943 samples (256 per mini-batch) +2023-10-05 22:01:37,232 - Epoch: [168][ 10/ 117] Loss 0.304712 Top1 85.898438 Top5 98.554688 +2023-10-05 22:01:37,381 - Epoch: [168][ 20/ 117] Loss 0.317341 Top1 85.468750 Top5 98.339844 +2023-10-05 22:01:37,529 - Epoch: [168][ 30/ 117] Loss 0.329507 Top1 85.195312 Top5 98.281250 +2023-10-05 22:01:37,677 - Epoch: [168][ 40/ 117] Loss 0.320345 Top1 85.292969 Top5 98.281250 +2023-10-05 22:01:37,826 - Epoch: [168][ 50/ 117] Loss 0.316567 Top1 85.203125 Top5 98.210938 +2023-10-05 22:01:37,974 - Epoch: [168][ 60/ 117] Loss 0.319120 Top1 85.175781 Top5 98.144531 +2023-10-05 22:01:38,122 - Epoch: [168][ 70/ 117] Loss 0.312894 Top1 85.251116 Top5 98.141741 +2023-10-05 22:01:38,269 - Epoch: [168][ 80/ 117] Loss 0.313545 Top1 85.214844 Top5 98.100586 +2023-10-05 22:01:38,417 - Epoch: [168][ 90/ 117] Loss 0.317834 Top1 85.134549 Top5 98.125000 +2023-10-05 22:01:38,564 - Epoch: [168][ 100/ 117] Loss 0.315710 Top1 85.195312 Top5 98.128906 +2023-10-05 22:01:38,718 - Epoch: [168][ 110/ 117] Loss 0.312826 Top1 85.248580 Top5 98.160511 +2023-10-05 22:01:38,803 - Epoch: [168][ 117/ 117] Loss 0.313192 Top1 85.261998 Top5 98.166516 +2023-10-05 22:01:38,940 - ==> Top1: 85.262 Top5: 98.167 Loss: 0.313 + +2023-10-05 22:01:38,940 - ==> Confusion: +[[ 930 4 2 2 6 2 0 0 6 63 1 0 0 2 3 5 5 1 1 0 17] + [ 0 1075 3 0 6 12 1 11 0 0 0 3 0 0 2 3 3 0 6 0 6] + [ 1 1 970 14 2 0 21 8 0 0 4 0 7 3 0 3 2 1 5 4 10] + [ 2 1 16 981 2 5 0 1 3 1 2 0 5 1 20 3 0 4 21 2 19] + [ 19 4 2 0 961 4 0 1 0 6 1 2 1 3 12 5 21 1 0 1 6] + [ 3 40 1 0 2 995 2 15 0 0 2 7 1 11 7 2 8 0 4 4 12] + [ 0 6 21 0 0 2 1128 4 0 0 3 2 1 0 1 9 0 0 1 7 6] + [ 4 21 13 0 2 28 5 1057 0 1 3 10 0 3 2 1 1 0 51 5 11] + [ 21 2 1 1 0 1 0 0 964 41 14 1 2 7 15 2 2 2 8 3 2] + [ 107 1 6 0 6 2 0 0 21 927 1 0 0 23 4 8 2 1 0 1 9] + [ 3 4 6 8 0 1 5 4 8 0 970 0 1 14 4 3 3 0 5 2 12] + [ 0 0 2 0 1 18 0 1 0 1 0 946 24 5 0 6 2 17 0 8 4] + [ 0 2 3 7 0 1 0 1 0 0 2 34 975 2 0 4 2 15 3 5 12] + [ 2 0 1 1 0 7 0 0 7 12 6 5 3 1055 4 2 2 0 0 1 11] + [ 13 3 2 11 4 1 1 0 21 1 2 2 2 2 1001 0 2 0 18 0 15] + [ 1 4 1 0 2 0 0 0 0 0 0 7 7 2 1 1077 13 6 0 9 4] + [ 0 10 1 1 2 4 0 0 0 0 0 2 1 0 2 9 1114 0 0 3 12] + [ 0 0 0 4 0 0 3 0 0 0 0 1 16 1 0 10 0 997 0 1 5] + [ 1 6 4 19 1 0 0 21 2 0 1 0 0 1 6 0 1 0 995 2 8] + [ 0 2 4 3 2 7 11 9 0 0 0 16 2 0 1 6 11 1 4 1065 8] + [ 100 199 135 62 86 116 35 83 86 52 167 97 321 252 121 64 214 55 139 174 5347]] + +2023-10-05 22:01:38,942 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:01:38,942 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:01:38,948 - + +2023-10-05 22:01:38,948 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:01:39,926 - Epoch: [169][ 10/ 1236] Overall Loss 0.179956 Objective Loss 0.179956 LR 0.000250 Time 0.097773 +2023-10-05 22:01:40,126 - Epoch: [169][ 20/ 1236] Overall Loss 0.185560 Objective Loss 0.185560 LR 0.000250 Time 0.058891 +2023-10-05 22:01:40,323 - Epoch: [169][ 30/ 1236] Overall Loss 0.186611 Objective Loss 0.186611 LR 0.000250 Time 0.045808 +2023-10-05 22:01:40,523 - Epoch: [169][ 40/ 1236] Overall Loss 0.185661 Objective Loss 0.185661 LR 0.000250 Time 0.039352 +2023-10-05 22:01:40,720 - Epoch: [169][ 50/ 1236] Overall Loss 0.187448 Objective Loss 0.187448 LR 0.000250 Time 0.035416 +2023-10-05 22:01:40,922 - Epoch: [169][ 60/ 1236] Overall Loss 0.185675 Objective Loss 0.185675 LR 0.000250 Time 0.032861 +2023-10-05 22:01:41,121 - Epoch: [169][ 70/ 1236] Overall Loss 0.186903 Objective Loss 0.186903 LR 0.000250 Time 0.031003 +2023-10-05 22:01:41,322 - Epoch: [169][ 80/ 1236] Overall Loss 0.188940 Objective Loss 0.188940 LR 0.000250 Time 0.029638 +2023-10-05 22:01:41,521 - Epoch: [169][ 90/ 1236] Overall Loss 0.189080 Objective Loss 0.189080 LR 0.000250 Time 0.028556 +2023-10-05 22:01:41,722 - Epoch: [169][ 100/ 1236] Overall Loss 0.189242 Objective Loss 0.189242 LR 0.000250 Time 0.027708 +2023-10-05 22:01:41,920 - Epoch: [169][ 110/ 1236] Overall Loss 0.189976 Objective Loss 0.189976 LR 0.000250 Time 0.026985 +2023-10-05 22:01:42,120 - Epoch: [169][ 120/ 1236] Overall Loss 0.189274 Objective Loss 0.189274 LR 0.000250 Time 0.026401 +2023-10-05 22:01:42,318 - Epoch: [169][ 130/ 1236] Overall Loss 0.188938 Objective Loss 0.188938 LR 0.000250 Time 0.025889 +2023-10-05 22:01:42,517 - Epoch: [169][ 140/ 1236] Overall Loss 0.187587 Objective Loss 0.187587 LR 0.000250 Time 0.025458 +2023-10-05 22:01:42,715 - Epoch: [169][ 150/ 1236] Overall Loss 0.187880 Objective Loss 0.187880 LR 0.000250 Time 0.025077 +2023-10-05 22:01:42,915 - Epoch: [169][ 160/ 1236] Overall Loss 0.188537 Objective Loss 0.188537 LR 0.000250 Time 0.024756 +2023-10-05 22:01:43,113 - Epoch: [169][ 170/ 1236] Overall Loss 0.191306 Objective Loss 0.191306 LR 0.000250 Time 0.024463 +2023-10-05 22:01:43,313 - Epoch: [169][ 180/ 1236] Overall Loss 0.191460 Objective Loss 0.191460 LR 0.000250 Time 0.024211 +2023-10-05 22:01:43,510 - Epoch: [169][ 190/ 1236] Overall Loss 0.191524 Objective Loss 0.191524 LR 0.000250 Time 0.023975 +2023-10-05 22:01:43,710 - Epoch: [169][ 200/ 1236] Overall Loss 0.190914 Objective Loss 0.190914 LR 0.000250 Time 0.023775 +2023-10-05 22:01:43,908 - Epoch: [169][ 210/ 1236] Overall Loss 0.190744 Objective Loss 0.190744 LR 0.000250 Time 0.023584 +2023-10-05 22:01:44,108 - Epoch: [169][ 220/ 1236] Overall Loss 0.190652 Objective Loss 0.190652 LR 0.000250 Time 0.023419 +2023-10-05 22:01:44,306 - Epoch: [169][ 230/ 1236] Overall Loss 0.191693 Objective Loss 0.191693 LR 0.000250 Time 0.023259 +2023-10-05 22:01:44,505 - Epoch: [169][ 240/ 1236] Overall Loss 0.191640 Objective Loss 0.191640 LR 0.000250 Time 0.023121 +2023-10-05 22:01:44,704 - Epoch: [169][ 250/ 1236] Overall Loss 0.191771 Objective Loss 0.191771 LR 0.000250 Time 0.022988 +2023-10-05 22:01:44,904 - Epoch: [169][ 260/ 1236] Overall Loss 0.191176 Objective Loss 0.191176 LR 0.000250 Time 0.022871 +2023-10-05 22:01:45,101 - Epoch: [169][ 270/ 1236] Overall Loss 0.190877 Objective Loss 0.190877 LR 0.000250 Time 0.022755 +2023-10-05 22:01:45,301 - Epoch: [169][ 280/ 1236] Overall Loss 0.190306 Objective Loss 0.190306 LR 0.000250 Time 0.022655 +2023-10-05 22:01:45,499 - Epoch: [169][ 290/ 1236] Overall Loss 0.189977 Objective Loss 0.189977 LR 0.000250 Time 0.022554 +2023-10-05 22:01:45,699 - Epoch: [169][ 300/ 1236] Overall Loss 0.189902 Objective Loss 0.189902 LR 0.000250 Time 0.022469 +2023-10-05 22:01:45,897 - Epoch: [169][ 310/ 1236] Overall Loss 0.190936 Objective Loss 0.190936 LR 0.000250 Time 0.022382 +2023-10-05 22:01:46,097 - Epoch: [169][ 320/ 1236] Overall Loss 0.191486 Objective Loss 0.191486 LR 0.000250 Time 0.022306 +2023-10-05 22:01:46,295 - Epoch: [169][ 330/ 1236] Overall Loss 0.192160 Objective Loss 0.192160 LR 0.000250 Time 0.022229 +2023-10-05 22:01:46,495 - Epoch: [169][ 340/ 1236] Overall Loss 0.191921 Objective Loss 0.191921 LR 0.000250 Time 0.022162 +2023-10-05 22:01:46,693 - Epoch: [169][ 350/ 1236] Overall Loss 0.191762 Objective Loss 0.191762 LR 0.000250 Time 0.022094 +2023-10-05 22:01:46,893 - Epoch: [169][ 360/ 1236] Overall Loss 0.191879 Objective Loss 0.191879 LR 0.000250 Time 0.022035 +2023-10-05 22:01:47,091 - Epoch: [169][ 370/ 1236] Overall Loss 0.192121 Objective Loss 0.192121 LR 0.000250 Time 0.021975 +2023-10-05 22:01:47,300 - Epoch: [169][ 380/ 1236] Overall Loss 0.192087 Objective Loss 0.192087 LR 0.000250 Time 0.021944 +2023-10-05 22:01:47,505 - Epoch: [169][ 390/ 1236] Overall Loss 0.192937 Objective Loss 0.192937 LR 0.000250 Time 0.021906 +2023-10-05 22:01:47,709 - Epoch: [169][ 400/ 1236] Overall Loss 0.192996 Objective Loss 0.192996 LR 0.000250 Time 0.021868 +2023-10-05 22:01:47,907 - Epoch: [169][ 410/ 1236] Overall Loss 0.192385 Objective Loss 0.192385 LR 0.000250 Time 0.021818 +2023-10-05 22:01:48,109 - Epoch: [169][ 420/ 1236] Overall Loss 0.192163 Objective Loss 0.192163 LR 0.000250 Time 0.021778 +2023-10-05 22:01:48,308 - Epoch: [169][ 430/ 1236] Overall Loss 0.191792 Objective Loss 0.191792 LR 0.000250 Time 0.021734 +2023-10-05 22:01:48,509 - Epoch: [169][ 440/ 1236] Overall Loss 0.191616 Objective Loss 0.191616 LR 0.000250 Time 0.021696 +2023-10-05 22:01:48,709 - Epoch: [169][ 450/ 1236] Overall Loss 0.192039 Objective Loss 0.192039 LR 0.000250 Time 0.021656 +2023-10-05 22:01:48,910 - Epoch: [169][ 460/ 1236] Overall Loss 0.191660 Objective Loss 0.191660 LR 0.000250 Time 0.021622 +2023-10-05 22:01:49,109 - Epoch: [169][ 470/ 1236] Overall Loss 0.191438 Objective Loss 0.191438 LR 0.000250 Time 0.021585 +2023-10-05 22:01:49,310 - Epoch: [169][ 480/ 1236] Overall Loss 0.191054 Objective Loss 0.191054 LR 0.000250 Time 0.021554 +2023-10-05 22:01:49,508 - Epoch: [169][ 490/ 1236] Overall Loss 0.190764 Objective Loss 0.190764 LR 0.000250 Time 0.021517 +2023-10-05 22:01:49,711 - Epoch: [169][ 500/ 1236] Overall Loss 0.190787 Objective Loss 0.190787 LR 0.000250 Time 0.021491 +2023-10-05 22:01:49,909 - Epoch: [169][ 510/ 1236] Overall Loss 0.190901 Objective Loss 0.190901 LR 0.000250 Time 0.021459 +2023-10-05 22:01:50,110 - Epoch: [169][ 520/ 1236] Overall Loss 0.191338 Objective Loss 0.191338 LR 0.000250 Time 0.021432 +2023-10-05 22:01:50,310 - Epoch: [169][ 530/ 1236] Overall Loss 0.191089 Objective Loss 0.191089 LR 0.000250 Time 0.021402 +2023-10-05 22:01:50,511 - Epoch: [169][ 540/ 1236] Overall Loss 0.191327 Objective Loss 0.191327 LR 0.000250 Time 0.021378 +2023-10-05 22:01:50,709 - Epoch: [169][ 550/ 1236] Overall Loss 0.191023 Objective Loss 0.191023 LR 0.000250 Time 0.021350 +2023-10-05 22:01:50,911 - Epoch: [169][ 560/ 1236] Overall Loss 0.190498 Objective Loss 0.190498 LR 0.000250 Time 0.021327 +2023-10-05 22:01:51,110 - Epoch: [169][ 570/ 1236] Overall Loss 0.190527 Objective Loss 0.190527 LR 0.000250 Time 0.021303 +2023-10-05 22:01:51,311 - Epoch: [169][ 580/ 1236] Overall Loss 0.190711 Objective Loss 0.190711 LR 0.000250 Time 0.021282 +2023-10-05 22:01:51,510 - Epoch: [169][ 590/ 1236] Overall Loss 0.190575 Objective Loss 0.190575 LR 0.000250 Time 0.021258 +2023-10-05 22:01:51,712 - Epoch: [169][ 600/ 1236] Overall Loss 0.190631 Objective Loss 0.190631 LR 0.000250 Time 0.021239 +2023-10-05 22:01:51,912 - Epoch: [169][ 610/ 1236] Overall Loss 0.191006 Objective Loss 0.191006 LR 0.000250 Time 0.021217 +2023-10-05 22:01:52,113 - Epoch: [169][ 620/ 1236] Overall Loss 0.190659 Objective Loss 0.190659 LR 0.000250 Time 0.021199 +2023-10-05 22:01:52,312 - Epoch: [169][ 630/ 1236] Overall Loss 0.190676 Objective Loss 0.190676 LR 0.000250 Time 0.021178 +2023-10-05 22:01:52,513 - Epoch: [169][ 640/ 1236] Overall Loss 0.190767 Objective Loss 0.190767 LR 0.000250 Time 0.021161 +2023-10-05 22:01:52,712 - Epoch: [169][ 650/ 1236] Overall Loss 0.190765 Objective Loss 0.190765 LR 0.000250 Time 0.021140 +2023-10-05 22:01:52,913 - Epoch: [169][ 660/ 1236] Overall Loss 0.190831 Objective Loss 0.190831 LR 0.000250 Time 0.021124 +2023-10-05 22:01:53,112 - Epoch: [169][ 670/ 1236] Overall Loss 0.190919 Objective Loss 0.190919 LR 0.000250 Time 0.021105 +2023-10-05 22:01:53,313 - Epoch: [169][ 680/ 1236] Overall Loss 0.190990 Objective Loss 0.190990 LR 0.000250 Time 0.021090 +2023-10-05 22:01:53,512 - Epoch: [169][ 690/ 1236] Overall Loss 0.190733 Objective Loss 0.190733 LR 0.000250 Time 0.021072 +2023-10-05 22:01:53,713 - Epoch: [169][ 700/ 1236] Overall Loss 0.190766 Objective Loss 0.190766 LR 0.000250 Time 0.021058 +2023-10-05 22:01:53,912 - Epoch: [169][ 710/ 1236] Overall Loss 0.190663 Objective Loss 0.190663 LR 0.000250 Time 0.021041 +2023-10-05 22:01:54,114 - Epoch: [169][ 720/ 1236] Overall Loss 0.190947 Objective Loss 0.190947 LR 0.000250 Time 0.021029 +2023-10-05 22:01:54,312 - Epoch: [169][ 730/ 1236] Overall Loss 0.190746 Objective Loss 0.190746 LR 0.000250 Time 0.021012 +2023-10-05 22:01:54,514 - Epoch: [169][ 740/ 1236] Overall Loss 0.190747 Objective Loss 0.190747 LR 0.000250 Time 0.021000 +2023-10-05 22:01:54,714 - Epoch: [169][ 750/ 1236] Overall Loss 0.190887 Objective Loss 0.190887 LR 0.000250 Time 0.020986 +2023-10-05 22:01:54,916 - Epoch: [169][ 760/ 1236] Overall Loss 0.190951 Objective Loss 0.190951 LR 0.000250 Time 0.020975 +2023-10-05 22:01:55,116 - Epoch: [169][ 770/ 1236] Overall Loss 0.191155 Objective Loss 0.191155 LR 0.000250 Time 0.020962 +2023-10-05 22:01:55,317 - Epoch: [169][ 780/ 1236] Overall Loss 0.190840 Objective Loss 0.190840 LR 0.000250 Time 0.020951 +2023-10-05 22:01:55,519 - Epoch: [169][ 790/ 1236] Overall Loss 0.190907 Objective Loss 0.190907 LR 0.000250 Time 0.020941 +2023-10-05 22:01:55,723 - Epoch: [169][ 800/ 1236] Overall Loss 0.190828 Objective Loss 0.190828 LR 0.000250 Time 0.020934 +2023-10-05 22:01:55,925 - Epoch: [169][ 810/ 1236] Overall Loss 0.190947 Objective Loss 0.190947 LR 0.000250 Time 0.020924 +2023-10-05 22:01:56,127 - Epoch: [169][ 820/ 1236] Overall Loss 0.190948 Objective Loss 0.190948 LR 0.000250 Time 0.020915 +2023-10-05 22:01:56,327 - Epoch: [169][ 830/ 1236] Overall Loss 0.191167 Objective Loss 0.191167 LR 0.000250 Time 0.020904 +2023-10-05 22:01:56,531 - Epoch: [169][ 840/ 1236] Overall Loss 0.191170 Objective Loss 0.191170 LR 0.000250 Time 0.020897 +2023-10-05 22:01:56,738 - Epoch: [169][ 850/ 1236] Overall Loss 0.191319 Objective Loss 0.191319 LR 0.000250 Time 0.020894 +2023-10-05 22:01:56,938 - Epoch: [169][ 860/ 1236] Overall Loss 0.191438 Objective Loss 0.191438 LR 0.000250 Time 0.020884 +2023-10-05 22:01:57,138 - Epoch: [169][ 870/ 1236] Overall Loss 0.191317 Objective Loss 0.191317 LR 0.000250 Time 0.020873 +2023-10-05 22:01:57,339 - Epoch: [169][ 880/ 1236] Overall Loss 0.191061 Objective Loss 0.191061 LR 0.000250 Time 0.020863 +2023-10-05 22:01:57,538 - Epoch: [169][ 890/ 1236] Overall Loss 0.191113 Objective Loss 0.191113 LR 0.000250 Time 0.020852 +2023-10-05 22:01:57,739 - Epoch: [169][ 900/ 1236] Overall Loss 0.191209 Objective Loss 0.191209 LR 0.000250 Time 0.020843 +2023-10-05 22:01:57,938 - Epoch: [169][ 910/ 1236] Overall Loss 0.191338 Objective Loss 0.191338 LR 0.000250 Time 0.020833 +2023-10-05 22:01:58,140 - Epoch: [169][ 920/ 1236] Overall Loss 0.191315 Objective Loss 0.191315 LR 0.000250 Time 0.020826 +2023-10-05 22:01:58,340 - Epoch: [169][ 930/ 1236] Overall Loss 0.191025 Objective Loss 0.191025 LR 0.000250 Time 0.020816 +2023-10-05 22:01:58,541 - Epoch: [169][ 940/ 1236] Overall Loss 0.190908 Objective Loss 0.190908 LR 0.000250 Time 0.020808 +2023-10-05 22:01:58,740 - Epoch: [169][ 950/ 1236] Overall Loss 0.190897 Objective Loss 0.190897 LR 0.000250 Time 0.020798 +2023-10-05 22:01:58,941 - Epoch: [169][ 960/ 1236] Overall Loss 0.190927 Objective Loss 0.190927 LR 0.000250 Time 0.020790 +2023-10-05 22:01:59,141 - Epoch: [169][ 970/ 1236] Overall Loss 0.190918 Objective Loss 0.190918 LR 0.000250 Time 0.020782 +2023-10-05 22:01:59,342 - Epoch: [169][ 980/ 1236] Overall Loss 0.190789 Objective Loss 0.190789 LR 0.000250 Time 0.020775 +2023-10-05 22:01:59,541 - Epoch: [169][ 990/ 1236] Overall Loss 0.190643 Objective Loss 0.190643 LR 0.000250 Time 0.020766 +2023-10-05 22:01:59,743 - Epoch: [169][ 1000/ 1236] Overall Loss 0.190544 Objective Loss 0.190544 LR 0.000250 Time 0.020759 +2023-10-05 22:01:59,941 - Epoch: [169][ 1010/ 1236] Overall Loss 0.190482 Objective Loss 0.190482 LR 0.000250 Time 0.020750 +2023-10-05 22:02:00,142 - Epoch: [169][ 1020/ 1236] Overall Loss 0.190641 Objective Loss 0.190641 LR 0.000250 Time 0.020743 +2023-10-05 22:02:00,340 - Epoch: [169][ 1030/ 1236] Overall Loss 0.190527 Objective Loss 0.190527 LR 0.000250 Time 0.020734 +2023-10-05 22:02:00,542 - Epoch: [169][ 1040/ 1236] Overall Loss 0.190465 Objective Loss 0.190465 LR 0.000250 Time 0.020728 +2023-10-05 22:02:00,741 - Epoch: [169][ 1050/ 1236] Overall Loss 0.190509 Objective Loss 0.190509 LR 0.000250 Time 0.020719 +2023-10-05 22:02:00,943 - Epoch: [169][ 1060/ 1236] Overall Loss 0.190460 Objective Loss 0.190460 LR 0.000250 Time 0.020714 +2023-10-05 22:02:01,141 - Epoch: [169][ 1070/ 1236] Overall Loss 0.190368 Objective Loss 0.190368 LR 0.000250 Time 0.020706 +2023-10-05 22:02:01,345 - Epoch: [169][ 1080/ 1236] Overall Loss 0.190505 Objective Loss 0.190505 LR 0.000250 Time 0.020703 +2023-10-05 22:02:01,550 - Epoch: [169][ 1090/ 1236] Overall Loss 0.190438 Objective Loss 0.190438 LR 0.000250 Time 0.020701 +2023-10-05 22:02:01,753 - Epoch: [169][ 1100/ 1236] Overall Loss 0.190428 Objective Loss 0.190428 LR 0.000250 Time 0.020696 +2023-10-05 22:02:01,952 - Epoch: [169][ 1110/ 1236] Overall Loss 0.190379 Objective Loss 0.190379 LR 0.000250 Time 0.020689 +2023-10-05 22:02:02,159 - Epoch: [169][ 1120/ 1236] Overall Loss 0.190382 Objective Loss 0.190382 LR 0.000250 Time 0.020689 +2023-10-05 22:02:02,365 - Epoch: [169][ 1130/ 1236] Overall Loss 0.190479 Objective Loss 0.190479 LR 0.000250 Time 0.020688 +2023-10-05 22:02:02,566 - Epoch: [169][ 1140/ 1236] Overall Loss 0.190455 Objective Loss 0.190455 LR 0.000250 Time 0.020682 +2023-10-05 22:02:02,765 - Epoch: [169][ 1150/ 1236] Overall Loss 0.190608 Objective Loss 0.190608 LR 0.000250 Time 0.020675 +2023-10-05 22:02:02,970 - Epoch: [169][ 1160/ 1236] Overall Loss 0.190501 Objective Loss 0.190501 LR 0.000250 Time 0.020673 +2023-10-05 22:02:03,177 - Epoch: [169][ 1170/ 1236] Overall Loss 0.190386 Objective Loss 0.190386 LR 0.000250 Time 0.020673 +2023-10-05 22:02:03,392 - Epoch: [169][ 1180/ 1236] Overall Loss 0.190269 Objective Loss 0.190269 LR 0.000250 Time 0.020680 +2023-10-05 22:02:03,601 - Epoch: [169][ 1190/ 1236] Overall Loss 0.190233 Objective Loss 0.190233 LR 0.000250 Time 0.020681 +2023-10-05 22:02:03,814 - Epoch: [169][ 1200/ 1236] Overall Loss 0.190324 Objective Loss 0.190324 LR 0.000250 Time 0.020686 +2023-10-05 22:02:04,022 - Epoch: [169][ 1210/ 1236] Overall Loss 0.190349 Objective Loss 0.190349 LR 0.000250 Time 0.020687 +2023-10-05 22:02:04,237 - Epoch: [169][ 1220/ 1236] Overall Loss 0.190445 Objective Loss 0.190445 LR 0.000250 Time 0.020693 +2023-10-05 22:02:04,500 - Epoch: [169][ 1230/ 1236] Overall Loss 0.190459 Objective Loss 0.190459 LR 0.000250 Time 0.020739 +2023-10-05 22:02:04,618 - Epoch: [169][ 1236/ 1236] Overall Loss 0.190491 Objective Loss 0.190491 Top1 86.965377 Top5 98.574338 LR 0.000250 Time 0.020734 +2023-10-05 22:02:04,747 - --- validate (epoch=169)----------- +2023-10-05 22:02:04,747 - 29943 samples (256 per mini-batch) +2023-10-05 22:02:05,233 - Epoch: [169][ 10/ 117] Loss 0.300720 Top1 85.742188 Top5 98.007812 +2023-10-05 22:02:05,384 - Epoch: [169][ 20/ 117] Loss 0.296357 Top1 85.468750 Top5 98.085938 +2023-10-05 22:02:05,536 - Epoch: [169][ 30/ 117] Loss 0.297113 Top1 85.533854 Top5 98.229167 +2023-10-05 22:02:05,693 - Epoch: [169][ 40/ 117] Loss 0.293211 Top1 85.576172 Top5 98.300781 +2023-10-05 22:02:05,846 - Epoch: [169][ 50/ 117] Loss 0.292768 Top1 85.570312 Top5 98.257812 +2023-10-05 22:02:06,001 - Epoch: [169][ 60/ 117] Loss 0.300325 Top1 85.338542 Top5 98.242188 +2023-10-05 22:02:06,163 - Epoch: [169][ 70/ 117] Loss 0.300319 Top1 85.279018 Top5 98.186384 +2023-10-05 22:02:06,322 - Epoch: [169][ 80/ 117] Loss 0.300325 Top1 85.380859 Top5 98.144531 +2023-10-05 22:02:06,477 - Epoch: [169][ 90/ 117] Loss 0.300350 Top1 85.329861 Top5 98.142361 +2023-10-05 22:02:06,629 - Epoch: [169][ 100/ 117] Loss 0.302345 Top1 85.324219 Top5 98.125000 +2023-10-05 22:02:06,789 - Epoch: [169][ 110/ 117] Loss 0.305748 Top1 85.305398 Top5 98.135653 +2023-10-05 22:02:06,876 - Epoch: [169][ 117/ 117] Loss 0.304284 Top1 85.402264 Top5 98.159837 +2023-10-05 22:02:06,992 - ==> Top1: 85.402 Top5: 98.160 Loss: 0.304 + +2023-10-05 22:02:06,993 - ==> Confusion: +[[ 931 4 3 0 7 3 0 0 2 71 3 0 0 0 6 3 3 0 0 0 14] + [ 0 1074 2 0 3 16 1 13 1 0 1 2 0 0 0 3 2 0 5 2 6] + [ 2 1 976 11 2 0 22 8 0 2 6 1 7 2 0 4 0 2 1 2 7] + [ 3 1 14 979 0 2 1 3 2 2 4 0 7 2 21 4 0 8 19 0 17] + [ 20 7 1 0 971 4 1 3 0 10 2 1 2 2 8 2 10 2 0 1 3] + [ 3 36 1 1 3 987 1 23 2 2 4 9 0 15 4 1 5 0 6 4 9] + [ 0 5 15 1 0 0 1138 11 0 0 0 1 2 0 0 5 1 0 0 8 4] + [ 3 17 12 0 1 28 6 1087 0 7 2 9 0 1 0 2 0 0 29 5 9] + [ 20 5 0 0 1 2 1 1 971 43 10 0 2 9 14 4 0 0 4 1 1] + [ 106 2 3 1 1 3 0 0 13 947 1 1 1 21 2 7 0 1 0 1 8] + [ 2 2 11 4 2 0 3 4 11 3 973 4 1 13 3 0 3 0 4 0 10] + [ 0 0 2 0 1 15 0 2 0 1 0 949 22 4 0 6 2 20 0 7 4] + [ 2 0 3 5 1 2 0 2 1 0 1 28 995 2 1 6 3 9 1 2 4] + [ 3 0 1 0 1 5 0 0 7 13 8 5 4 1058 3 2 0 0 0 0 9] + [ 11 2 3 13 3 1 0 0 23 2 2 1 1 3 1006 0 2 1 13 0 14] + [ 1 3 2 0 1 0 2 0 0 1 0 5 6 1 1 1075 13 14 0 6 3] + [ 0 12 1 0 4 5 0 1 1 0 0 5 0 0 2 13 1101 0 0 4 12] + [ 0 0 0 1 0 0 3 0 0 0 0 2 18 1 0 2 0 1006 1 1 3] + [ 0 5 7 16 1 1 2 24 1 0 1 0 2 1 9 0 1 0 989 2 6] + [ 0 2 5 1 1 9 8 8 0 0 2 13 2 0 0 3 7 3 2 1076 10] + [ 125 184 153 60 84 139 45 92 87 71 173 86 324 260 132 61 155 61 141 189 5283]] + +2023-10-05 22:02:06,994 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:02:06,994 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:02:07,000 - + +2023-10-05 22:02:07,000 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:02:08,140 - Epoch: [170][ 10/ 1236] Overall Loss 0.176720 Objective Loss 0.176720 LR 0.000250 Time 0.113934 +2023-10-05 22:02:08,349 - Epoch: [170][ 20/ 1236] Overall Loss 0.193011 Objective Loss 0.193011 LR 0.000250 Time 0.067365 +2023-10-05 22:02:08,554 - Epoch: [170][ 30/ 1236] Overall Loss 0.188623 Objective Loss 0.188623 LR 0.000250 Time 0.051743 +2023-10-05 22:02:08,763 - Epoch: [170][ 40/ 1236] Overall Loss 0.190821 Objective Loss 0.190821 LR 0.000250 Time 0.044010 +2023-10-05 22:02:08,968 - Epoch: [170][ 50/ 1236] Overall Loss 0.191661 Objective Loss 0.191661 LR 0.000250 Time 0.039299 +2023-10-05 22:02:09,176 - Epoch: [170][ 60/ 1236] Overall Loss 0.191407 Objective Loss 0.191407 LR 0.000250 Time 0.036217 +2023-10-05 22:02:09,381 - Epoch: [170][ 70/ 1236] Overall Loss 0.189607 Objective Loss 0.189607 LR 0.000250 Time 0.033965 +2023-10-05 22:02:09,591 - Epoch: [170][ 80/ 1236] Overall Loss 0.192130 Objective Loss 0.192130 LR 0.000250 Time 0.032331 +2023-10-05 22:02:09,796 - Epoch: [170][ 90/ 1236] Overall Loss 0.190643 Objective Loss 0.190643 LR 0.000250 Time 0.031018 +2023-10-05 22:02:10,004 - Epoch: [170][ 100/ 1236] Overall Loss 0.188072 Objective Loss 0.188072 LR 0.000250 Time 0.029987 +2023-10-05 22:02:10,208 - Epoch: [170][ 110/ 1236] Overall Loss 0.187397 Objective Loss 0.187397 LR 0.000250 Time 0.029115 +2023-10-05 22:02:10,415 - Epoch: [170][ 120/ 1236] Overall Loss 0.187833 Objective Loss 0.187833 LR 0.000250 Time 0.028406 +2023-10-05 22:02:10,618 - Epoch: [170][ 130/ 1236] Overall Loss 0.188246 Objective Loss 0.188246 LR 0.000250 Time 0.027783 +2023-10-05 22:02:10,824 - Epoch: [170][ 140/ 1236] Overall Loss 0.187593 Objective Loss 0.187593 LR 0.000250 Time 0.027268 +2023-10-05 22:02:11,027 - Epoch: [170][ 150/ 1236] Overall Loss 0.187329 Objective Loss 0.187329 LR 0.000250 Time 0.026801 +2023-10-05 22:02:11,233 - Epoch: [170][ 160/ 1236] Overall Loss 0.188560 Objective Loss 0.188560 LR 0.000250 Time 0.026412 +2023-10-05 22:02:11,437 - Epoch: [170][ 170/ 1236] Overall Loss 0.188277 Objective Loss 0.188277 LR 0.000250 Time 0.026055 +2023-10-05 22:02:11,644 - Epoch: [170][ 180/ 1236] Overall Loss 0.188176 Objective Loss 0.188176 LR 0.000250 Time 0.025752 +2023-10-05 22:02:11,847 - Epoch: [170][ 190/ 1236] Overall Loss 0.188911 Objective Loss 0.188911 LR 0.000250 Time 0.025467 +2023-10-05 22:02:12,054 - Epoch: [170][ 200/ 1236] Overall Loss 0.188845 Objective Loss 0.188845 LR 0.000250 Time 0.025224 +2023-10-05 22:02:12,257 - Epoch: [170][ 210/ 1236] Overall Loss 0.188170 Objective Loss 0.188170 LR 0.000250 Time 0.024990 +2023-10-05 22:02:12,464 - Epoch: [170][ 220/ 1236] Overall Loss 0.187793 Objective Loss 0.187793 LR 0.000250 Time 0.024790 +2023-10-05 22:02:12,667 - Epoch: [170][ 230/ 1236] Overall Loss 0.187984 Objective Loss 0.187984 LR 0.000250 Time 0.024596 +2023-10-05 22:02:12,874 - Epoch: [170][ 240/ 1236] Overall Loss 0.187741 Objective Loss 0.187741 LR 0.000250 Time 0.024430 +2023-10-05 22:02:13,077 - Epoch: [170][ 250/ 1236] Overall Loss 0.188015 Objective Loss 0.188015 LR 0.000250 Time 0.024265 +2023-10-05 22:02:13,284 - Epoch: [170][ 260/ 1236] Overall Loss 0.187949 Objective Loss 0.187949 LR 0.000250 Time 0.024125 +2023-10-05 22:02:13,487 - Epoch: [170][ 270/ 1236] Overall Loss 0.187156 Objective Loss 0.187156 LR 0.000250 Time 0.023983 +2023-10-05 22:02:13,694 - Epoch: [170][ 280/ 1236] Overall Loss 0.186636 Objective Loss 0.186636 LR 0.000250 Time 0.023862 +2023-10-05 22:02:13,897 - Epoch: [170][ 290/ 1236] Overall Loss 0.187184 Objective Loss 0.187184 LR 0.000250 Time 0.023740 +2023-10-05 22:02:14,102 - Epoch: [170][ 300/ 1236] Overall Loss 0.187566 Objective Loss 0.187566 LR 0.000250 Time 0.023631 +2023-10-05 22:02:14,304 - Epoch: [170][ 310/ 1236] Overall Loss 0.186627 Objective Loss 0.186627 LR 0.000250 Time 0.023519 +2023-10-05 22:02:14,509 - Epoch: [170][ 320/ 1236] Overall Loss 0.187147 Objective Loss 0.187147 LR 0.000250 Time 0.023423 +2023-10-05 22:02:14,715 - Epoch: [170][ 330/ 1236] Overall Loss 0.187230 Objective Loss 0.187230 LR 0.000250 Time 0.023337 +2023-10-05 22:02:14,926 - Epoch: [170][ 340/ 1236] Overall Loss 0.187087 Objective Loss 0.187087 LR 0.000250 Time 0.023268 +2023-10-05 22:02:15,134 - Epoch: [170][ 350/ 1236] Overall Loss 0.187475 Objective Loss 0.187475 LR 0.000250 Time 0.023199 +2023-10-05 22:02:15,346 - Epoch: [170][ 360/ 1236] Overall Loss 0.188186 Objective Loss 0.188186 LR 0.000250 Time 0.023140 +2023-10-05 22:02:15,554 - Epoch: [170][ 370/ 1236] Overall Loss 0.188589 Objective Loss 0.188589 LR 0.000250 Time 0.023076 +2023-10-05 22:02:15,763 - Epoch: [170][ 380/ 1236] Overall Loss 0.188559 Objective Loss 0.188559 LR 0.000250 Time 0.023018 +2023-10-05 22:02:15,969 - Epoch: [170][ 390/ 1236] Overall Loss 0.187985 Objective Loss 0.187985 LR 0.000250 Time 0.022954 +2023-10-05 22:02:16,178 - Epoch: [170][ 400/ 1236] Overall Loss 0.188045 Objective Loss 0.188045 LR 0.000250 Time 0.022903 +2023-10-05 22:02:16,384 - Epoch: [170][ 410/ 1236] Overall Loss 0.187799 Objective Loss 0.187799 LR 0.000250 Time 0.022844 +2023-10-05 22:02:16,593 - Epoch: [170][ 420/ 1236] Overall Loss 0.188040 Objective Loss 0.188040 LR 0.000250 Time 0.022798 +2023-10-05 22:02:16,799 - Epoch: [170][ 430/ 1236] Overall Loss 0.187963 Objective Loss 0.187963 LR 0.000250 Time 0.022744 +2023-10-05 22:02:17,008 - Epoch: [170][ 440/ 1236] Overall Loss 0.187810 Objective Loss 0.187810 LR 0.000250 Time 0.022702 +2023-10-05 22:02:17,213 - Epoch: [170][ 450/ 1236] Overall Loss 0.187832 Objective Loss 0.187832 LR 0.000250 Time 0.022653 +2023-10-05 22:02:17,423 - Epoch: [170][ 460/ 1236] Overall Loss 0.187768 Objective Loss 0.187768 LR 0.000250 Time 0.022615 +2023-10-05 22:02:17,628 - Epoch: [170][ 470/ 1236] Overall Loss 0.187485 Objective Loss 0.187485 LR 0.000250 Time 0.022570 +2023-10-05 22:02:17,836 - Epoch: [170][ 480/ 1236] Overall Loss 0.187634 Objective Loss 0.187634 LR 0.000250 Time 0.022532 +2023-10-05 22:02:18,041 - Epoch: [170][ 490/ 1236] Overall Loss 0.187601 Objective Loss 0.187601 LR 0.000250 Time 0.022490 +2023-10-05 22:02:18,251 - Epoch: [170][ 500/ 1236] Overall Loss 0.187582 Objective Loss 0.187582 LR 0.000250 Time 0.022458 +2023-10-05 22:02:18,455 - Epoch: [170][ 510/ 1236] Overall Loss 0.187453 Objective Loss 0.187453 LR 0.000250 Time 0.022418 +2023-10-05 22:02:18,665 - Epoch: [170][ 520/ 1236] Overall Loss 0.187220 Objective Loss 0.187220 LR 0.000250 Time 0.022389 +2023-10-05 22:02:18,871 - Epoch: [170][ 530/ 1236] Overall Loss 0.187318 Objective Loss 0.187318 LR 0.000250 Time 0.022354 +2023-10-05 22:02:19,079 - Epoch: [170][ 540/ 1236] Overall Loss 0.187539 Objective Loss 0.187539 LR 0.000250 Time 0.022324 +2023-10-05 22:02:19,284 - Epoch: [170][ 550/ 1236] Overall Loss 0.187323 Objective Loss 0.187323 LR 0.000250 Time 0.022290 +2023-10-05 22:02:19,492 - Epoch: [170][ 560/ 1236] Overall Loss 0.187509 Objective Loss 0.187509 LR 0.000250 Time 0.022264 +2023-10-05 22:02:19,697 - Epoch: [170][ 570/ 1236] Overall Loss 0.187610 Objective Loss 0.187610 LR 0.000250 Time 0.022232 +2023-10-05 22:02:19,904 - Epoch: [170][ 580/ 1236] Overall Loss 0.187621 Objective Loss 0.187621 LR 0.000250 Time 0.022205 +2023-10-05 22:02:20,106 - Epoch: [170][ 590/ 1236] Overall Loss 0.187409 Objective Loss 0.187409 LR 0.000250 Time 0.022170 +2023-10-05 22:02:20,312 - Epoch: [170][ 600/ 1236] Overall Loss 0.187497 Objective Loss 0.187497 LR 0.000250 Time 0.022144 +2023-10-05 22:02:20,514 - Epoch: [170][ 610/ 1236] Overall Loss 0.187535 Objective Loss 0.187535 LR 0.000250 Time 0.022111 +2023-10-05 22:02:20,720 - Epoch: [170][ 620/ 1236] Overall Loss 0.187654 Objective Loss 0.187654 LR 0.000250 Time 0.022086 +2023-10-05 22:02:20,925 - Epoch: [170][ 630/ 1236] Overall Loss 0.187529 Objective Loss 0.187529 LR 0.000250 Time 0.022060 +2023-10-05 22:02:21,132 - Epoch: [170][ 640/ 1236] Overall Loss 0.187515 Objective Loss 0.187515 LR 0.000250 Time 0.022038 +2023-10-05 22:02:21,335 - Epoch: [170][ 650/ 1236] Overall Loss 0.187785 Objective Loss 0.187785 LR 0.000250 Time 0.022012 +2023-10-05 22:02:21,543 - Epoch: [170][ 660/ 1236] Overall Loss 0.187908 Objective Loss 0.187908 LR 0.000250 Time 0.021992 +2023-10-05 22:02:21,747 - Epoch: [170][ 670/ 1236] Overall Loss 0.187621 Objective Loss 0.187621 LR 0.000250 Time 0.021967 +2023-10-05 22:02:21,954 - Epoch: [170][ 680/ 1236] Overall Loss 0.187974 Objective Loss 0.187974 LR 0.000250 Time 0.021949 +2023-10-05 22:02:22,158 - Epoch: [170][ 690/ 1236] Overall Loss 0.188235 Objective Loss 0.188235 LR 0.000250 Time 0.021925 +2023-10-05 22:02:22,365 - Epoch: [170][ 700/ 1236] Overall Loss 0.188418 Objective Loss 0.188418 LR 0.000250 Time 0.021908 +2023-10-05 22:02:22,569 - Epoch: [170][ 710/ 1236] Overall Loss 0.188688 Objective Loss 0.188688 LR 0.000250 Time 0.021885 +2023-10-05 22:02:22,776 - Epoch: [170][ 720/ 1236] Overall Loss 0.188818 Objective Loss 0.188818 LR 0.000250 Time 0.021868 +2023-10-05 22:02:22,979 - Epoch: [170][ 730/ 1236] Overall Loss 0.188878 Objective Loss 0.188878 LR 0.000250 Time 0.021847 +2023-10-05 22:02:23,186 - Epoch: [170][ 740/ 1236] Overall Loss 0.188768 Objective Loss 0.188768 LR 0.000250 Time 0.021831 +2023-10-05 22:02:23,391 - Epoch: [170][ 750/ 1236] Overall Loss 0.188672 Objective Loss 0.188672 LR 0.000250 Time 0.021812 +2023-10-05 22:02:23,597 - Epoch: [170][ 760/ 1236] Overall Loss 0.188763 Objective Loss 0.188763 LR 0.000250 Time 0.021797 +2023-10-05 22:02:23,802 - Epoch: [170][ 770/ 1236] Overall Loss 0.188934 Objective Loss 0.188934 LR 0.000250 Time 0.021778 +2023-10-05 22:02:24,009 - Epoch: [170][ 780/ 1236] Overall Loss 0.188796 Objective Loss 0.188796 LR 0.000250 Time 0.021764 +2023-10-05 22:02:24,212 - Epoch: [170][ 790/ 1236] Overall Loss 0.189023 Objective Loss 0.189023 LR 0.000250 Time 0.021746 +2023-10-05 22:02:24,419 - Epoch: [170][ 800/ 1236] Overall Loss 0.189057 Objective Loss 0.189057 LR 0.000250 Time 0.021732 +2023-10-05 22:02:24,623 - Epoch: [170][ 810/ 1236] Overall Loss 0.189186 Objective Loss 0.189186 LR 0.000250 Time 0.021715 +2023-10-05 22:02:24,829 - Epoch: [170][ 820/ 1236] Overall Loss 0.188906 Objective Loss 0.188906 LR 0.000250 Time 0.021701 +2023-10-05 22:02:25,033 - Epoch: [170][ 830/ 1236] Overall Loss 0.188867 Objective Loss 0.188867 LR 0.000250 Time 0.021685 +2023-10-05 22:02:25,241 - Epoch: [170][ 840/ 1236] Overall Loss 0.188923 Objective Loss 0.188923 LR 0.000250 Time 0.021673 +2023-10-05 22:02:25,443 - Epoch: [170][ 850/ 1236] Overall Loss 0.188902 Objective Loss 0.188902 LR 0.000250 Time 0.021656 +2023-10-05 22:02:25,647 - Epoch: [170][ 860/ 1236] Overall Loss 0.188867 Objective Loss 0.188867 LR 0.000250 Time 0.021641 +2023-10-05 22:02:25,849 - Epoch: [170][ 870/ 1236] Overall Loss 0.188934 Objective Loss 0.188934 LR 0.000250 Time 0.021623 +2023-10-05 22:02:26,054 - Epoch: [170][ 880/ 1236] Overall Loss 0.189109 Objective Loss 0.189109 LR 0.000250 Time 0.021610 +2023-10-05 22:02:26,256 - Epoch: [170][ 890/ 1236] Overall Loss 0.188996 Objective Loss 0.188996 LR 0.000250 Time 0.021594 +2023-10-05 22:02:26,461 - Epoch: [170][ 900/ 1236] Overall Loss 0.188682 Objective Loss 0.188682 LR 0.000250 Time 0.021581 +2023-10-05 22:02:26,662 - Epoch: [170][ 910/ 1236] Overall Loss 0.188778 Objective Loss 0.188778 LR 0.000250 Time 0.021565 +2023-10-05 22:02:26,867 - Epoch: [170][ 920/ 1236] Overall Loss 0.188779 Objective Loss 0.188779 LR 0.000250 Time 0.021553 +2023-10-05 22:02:27,069 - Epoch: [170][ 930/ 1236] Overall Loss 0.188817 Objective Loss 0.188817 LR 0.000250 Time 0.021538 +2023-10-05 22:02:27,274 - Epoch: [170][ 940/ 1236] Overall Loss 0.188700 Objective Loss 0.188700 LR 0.000250 Time 0.021526 +2023-10-05 22:02:27,475 - Epoch: [170][ 950/ 1236] Overall Loss 0.188453 Objective Loss 0.188453 LR 0.000250 Time 0.021511 +2023-10-05 22:02:27,679 - Epoch: [170][ 960/ 1236] Overall Loss 0.188784 Objective Loss 0.188784 LR 0.000250 Time 0.021500 +2023-10-05 22:02:27,881 - Epoch: [170][ 970/ 1236] Overall Loss 0.188733 Objective Loss 0.188733 LR 0.000250 Time 0.021485 +2023-10-05 22:02:28,085 - Epoch: [170][ 980/ 1236] Overall Loss 0.188616 Objective Loss 0.188616 LR 0.000250 Time 0.021474 +2023-10-05 22:02:28,287 - Epoch: [170][ 990/ 1236] Overall Loss 0.188780 Objective Loss 0.188780 LR 0.000250 Time 0.021460 +2023-10-05 22:02:28,491 - Epoch: [170][ 1000/ 1236] Overall Loss 0.188823 Objective Loss 0.188823 LR 0.000250 Time 0.021450 +2023-10-05 22:02:28,693 - Epoch: [170][ 1010/ 1236] Overall Loss 0.188681 Objective Loss 0.188681 LR 0.000250 Time 0.021437 +2023-10-05 22:02:28,899 - Epoch: [170][ 1020/ 1236] Overall Loss 0.188811 Objective Loss 0.188811 LR 0.000250 Time 0.021429 +2023-10-05 22:02:29,102 - Epoch: [170][ 1030/ 1236] Overall Loss 0.188879 Objective Loss 0.188879 LR 0.000250 Time 0.021417 +2023-10-05 22:02:29,306 - Epoch: [170][ 1040/ 1236] Overall Loss 0.188912 Objective Loss 0.188912 LR 0.000250 Time 0.021407 +2023-10-05 22:02:29,510 - Epoch: [170][ 1050/ 1236] Overall Loss 0.188871 Objective Loss 0.188871 LR 0.000250 Time 0.021397 +2023-10-05 22:02:29,715 - Epoch: [170][ 1060/ 1236] Overall Loss 0.188790 Objective Loss 0.188790 LR 0.000250 Time 0.021388 +2023-10-05 22:02:29,918 - Epoch: [170][ 1070/ 1236] Overall Loss 0.188859 Objective Loss 0.188859 LR 0.000250 Time 0.021378 +2023-10-05 22:02:30,122 - Epoch: [170][ 1080/ 1236] Overall Loss 0.188933 Objective Loss 0.188933 LR 0.000250 Time 0.021368 +2023-10-05 22:02:30,324 - Epoch: [170][ 1090/ 1236] Overall Loss 0.189075 Objective Loss 0.189075 LR 0.000250 Time 0.021357 +2023-10-05 22:02:30,527 - Epoch: [170][ 1100/ 1236] Overall Loss 0.189066 Objective Loss 0.189066 LR 0.000250 Time 0.021347 +2023-10-05 22:02:30,729 - Epoch: [170][ 1110/ 1236] Overall Loss 0.188995 Objective Loss 0.188995 LR 0.000250 Time 0.021336 +2023-10-05 22:02:30,933 - Epoch: [170][ 1120/ 1236] Overall Loss 0.188777 Objective Loss 0.188777 LR 0.000250 Time 0.021328 +2023-10-05 22:02:31,134 - Epoch: [170][ 1130/ 1236] Overall Loss 0.188852 Objective Loss 0.188852 LR 0.000250 Time 0.021317 +2023-10-05 22:02:31,338 - Epoch: [170][ 1140/ 1236] Overall Loss 0.188829 Objective Loss 0.188829 LR 0.000250 Time 0.021308 +2023-10-05 22:02:31,539 - Epoch: [170][ 1150/ 1236] Overall Loss 0.188946 Objective Loss 0.188946 LR 0.000250 Time 0.021298 +2023-10-05 22:02:31,743 - Epoch: [170][ 1160/ 1236] Overall Loss 0.188963 Objective Loss 0.188963 LR 0.000250 Time 0.021289 +2023-10-05 22:02:31,945 - Epoch: [170][ 1170/ 1236] Overall Loss 0.189121 Objective Loss 0.189121 LR 0.000250 Time 0.021280 +2023-10-05 22:02:32,149 - Epoch: [170][ 1180/ 1236] Overall Loss 0.189074 Objective Loss 0.189074 LR 0.000250 Time 0.021272 +2023-10-05 22:02:32,349 - Epoch: [170][ 1190/ 1236] Overall Loss 0.189208 Objective Loss 0.189208 LR 0.000250 Time 0.021261 +2023-10-05 22:02:32,552 - Epoch: [170][ 1200/ 1236] Overall Loss 0.189441 Objective Loss 0.189441 LR 0.000250 Time 0.021253 +2023-10-05 22:02:32,753 - Epoch: [170][ 1210/ 1236] Overall Loss 0.189530 Objective Loss 0.189530 LR 0.000250 Time 0.021243 +2023-10-05 22:02:32,956 - Epoch: [170][ 1220/ 1236] Overall Loss 0.189575 Objective Loss 0.189575 LR 0.000250 Time 0.021235 +2023-10-05 22:02:33,212 - Epoch: [170][ 1230/ 1236] Overall Loss 0.189624 Objective Loss 0.189624 LR 0.000250 Time 0.021270 +2023-10-05 22:02:33,331 - Epoch: [170][ 1236/ 1236] Overall Loss 0.189673 Objective Loss 0.189673 Top1 90.020367 Top5 98.778004 LR 0.000250 Time 0.021263 +2023-10-05 22:02:33,446 - --- validate (epoch=170)----------- +2023-10-05 22:02:33,447 - 29943 samples (256 per mini-batch) +2023-10-05 22:02:33,905 - Epoch: [170][ 10/ 117] Loss 0.325922 Top1 85.781250 Top5 98.437500 +2023-10-05 22:02:34,052 - Epoch: [170][ 20/ 117] Loss 0.322025 Top1 84.921875 Top5 98.320312 +2023-10-05 22:02:34,199 - Epoch: [170][ 30/ 117] Loss 0.316234 Top1 84.843750 Top5 98.216146 +2023-10-05 22:02:34,346 - Epoch: [170][ 40/ 117] Loss 0.313244 Top1 85.000000 Top5 98.173828 +2023-10-05 22:02:34,491 - Epoch: [170][ 50/ 117] Loss 0.310925 Top1 85.171875 Top5 98.125000 +2023-10-05 22:02:34,637 - Epoch: [170][ 60/ 117] Loss 0.303588 Top1 85.358073 Top5 98.177083 +2023-10-05 22:02:34,786 - Epoch: [170][ 70/ 117] Loss 0.305544 Top1 85.273438 Top5 98.175223 +2023-10-05 22:02:34,939 - Epoch: [170][ 80/ 117] Loss 0.304010 Top1 85.092773 Top5 98.144531 +2023-10-05 22:02:35,091 - Epoch: [170][ 90/ 117] Loss 0.308230 Top1 85.026042 Top5 98.094618 +2023-10-05 22:02:35,244 - Epoch: [170][ 100/ 117] Loss 0.309545 Top1 85.042969 Top5 98.121094 +2023-10-05 22:02:35,405 - Epoch: [170][ 110/ 117] Loss 0.311226 Top1 85.110085 Top5 98.149858 +2023-10-05 22:02:35,491 - Epoch: [170][ 117/ 117] Loss 0.308553 Top1 85.238620 Top5 98.176535 +2023-10-05 22:02:35,639 - ==> Top1: 85.239 Top5: 98.177 Loss: 0.309 + +2023-10-05 22:02:35,639 - ==> Confusion: +[[ 930 2 7 0 4 3 0 0 4 69 1 0 2 2 7 4 2 2 0 0 11] + [ 0 1066 3 0 10 11 1 17 2 0 1 1 0 0 1 4 2 1 6 1 4] + [ 4 1 971 10 3 0 23 7 0 1 4 1 9 2 0 3 1 1 6 4 5] + [ 4 1 14 960 0 2 0 1 1 1 7 1 7 3 28 3 0 8 27 4 17] + [ 19 5 2 0 978 0 0 1 0 7 0 1 1 2 10 4 12 2 0 2 4] + [ 4 37 0 1 7 974 0 15 0 2 6 14 1 18 4 2 4 0 5 6 16] + [ 0 7 22 0 0 0 1122 9 0 1 3 2 1 0 1 7 0 0 1 9 6] + [ 3 18 11 0 2 26 5 1066 1 4 6 9 1 1 0 2 1 0 48 5 9] + [ 20 3 0 0 0 0 1 1 961 41 11 2 3 13 19 5 0 0 6 1 2] + [ 105 0 1 0 3 3 1 0 14 945 1 3 0 21 4 8 1 0 0 1 8] + [ 3 8 10 1 1 1 1 4 5 1 985 4 1 11 3 1 2 0 4 0 7] + [ 1 0 2 0 1 9 0 1 0 1 0 962 19 4 0 6 1 15 0 8 5] + [ 0 1 5 4 0 1 0 0 1 0 0 34 987 3 0 3 3 16 2 2 6] + [ 2 0 1 0 2 4 0 0 5 14 6 4 2 1062 3 1 2 0 0 2 9] + [ 11 2 3 6 7 0 0 0 16 2 3 1 2 2 1021 0 2 2 11 0 10] + [ 0 4 2 0 2 0 0 0 0 1 0 7 7 1 1 1071 14 14 0 9 1] + [ 1 16 1 0 6 2 0 0 1 0 0 2 1 0 4 11 1101 0 0 3 12] + [ 0 0 0 1 0 0 2 0 0 0 0 2 16 1 1 6 0 1007 1 0 1] + [ 1 5 5 18 1 0 1 21 1 0 1 0 1 0 7 0 0 0 996 0 10] + [ 0 2 3 2 2 5 8 5 0 0 3 15 4 1 0 5 7 2 5 1076 7] + [ 120 162 168 53 125 113 31 78 79 72 182 96 318 273 147 66 148 72 153 167 5282]] + +2023-10-05 22:02:35,641 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:02:35,641 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:02:35,647 - + +2023-10-05 22:02:35,647 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:02:36,653 - Epoch: [171][ 10/ 1236] Overall Loss 0.175682 Objective Loss 0.175682 LR 0.000250 Time 0.100546 +2023-10-05 22:02:36,857 - Epoch: [171][ 20/ 1236] Overall Loss 0.189803 Objective Loss 0.189803 LR 0.000250 Time 0.060470 +2023-10-05 22:02:37,060 - Epoch: [171][ 30/ 1236] Overall Loss 0.190620 Objective Loss 0.190620 LR 0.000250 Time 0.047079 +2023-10-05 22:02:37,265 - Epoch: [171][ 40/ 1236] Overall Loss 0.192780 Objective Loss 0.192780 LR 0.000250 Time 0.040421 +2023-10-05 22:02:37,468 - Epoch: [171][ 50/ 1236] Overall Loss 0.192506 Objective Loss 0.192506 LR 0.000250 Time 0.036392 +2023-10-05 22:02:37,673 - Epoch: [171][ 60/ 1236] Overall Loss 0.193795 Objective Loss 0.193795 LR 0.000250 Time 0.033732 +2023-10-05 22:02:37,876 - Epoch: [171][ 70/ 1236] Overall Loss 0.194374 Objective Loss 0.194374 LR 0.000250 Time 0.031814 +2023-10-05 22:02:38,081 - Epoch: [171][ 80/ 1236] Overall Loss 0.193831 Objective Loss 0.193831 LR 0.000250 Time 0.030395 +2023-10-05 22:02:38,283 - Epoch: [171][ 90/ 1236] Overall Loss 0.191726 Objective Loss 0.191726 LR 0.000250 Time 0.029252 +2023-10-05 22:02:38,487 - Epoch: [171][ 100/ 1236] Overall Loss 0.191345 Objective Loss 0.191345 LR 0.000250 Time 0.028360 +2023-10-05 22:02:38,690 - Epoch: [171][ 110/ 1236] Overall Loss 0.191319 Objective Loss 0.191319 LR 0.000250 Time 0.027625 +2023-10-05 22:02:38,895 - Epoch: [171][ 120/ 1236] Overall Loss 0.190025 Objective Loss 0.190025 LR 0.000250 Time 0.027028 +2023-10-05 22:02:39,100 - Epoch: [171][ 130/ 1236] Overall Loss 0.189093 Objective Loss 0.189093 LR 0.000250 Time 0.026525 +2023-10-05 22:02:39,305 - Epoch: [171][ 140/ 1236] Overall Loss 0.189026 Objective Loss 0.189026 LR 0.000250 Time 0.026095 +2023-10-05 22:02:39,509 - Epoch: [171][ 150/ 1236] Overall Loss 0.188102 Objective Loss 0.188102 LR 0.000250 Time 0.025715 +2023-10-05 22:02:39,715 - Epoch: [171][ 160/ 1236] Overall Loss 0.188783 Objective Loss 0.188783 LR 0.000250 Time 0.025389 +2023-10-05 22:02:39,920 - Epoch: [171][ 170/ 1236] Overall Loss 0.187224 Objective Loss 0.187224 LR 0.000250 Time 0.025098 +2023-10-05 22:02:40,126 - Epoch: [171][ 180/ 1236] Overall Loss 0.185772 Objective Loss 0.185772 LR 0.000250 Time 0.024846 +2023-10-05 22:02:40,330 - Epoch: [171][ 190/ 1236] Overall Loss 0.185224 Objective Loss 0.185224 LR 0.000250 Time 0.024612 +2023-10-05 22:02:40,535 - Epoch: [171][ 200/ 1236] Overall Loss 0.185635 Objective Loss 0.185635 LR 0.000250 Time 0.024406 +2023-10-05 22:02:40,740 - Epoch: [171][ 210/ 1236] Overall Loss 0.185346 Objective Loss 0.185346 LR 0.000250 Time 0.024216 +2023-10-05 22:02:40,945 - Epoch: [171][ 220/ 1236] Overall Loss 0.185499 Objective Loss 0.185499 LR 0.000250 Time 0.024047 +2023-10-05 22:02:41,150 - Epoch: [171][ 230/ 1236] Overall Loss 0.184795 Objective Loss 0.184795 LR 0.000250 Time 0.023890 +2023-10-05 22:02:41,355 - Epoch: [171][ 240/ 1236] Overall Loss 0.185271 Objective Loss 0.185271 LR 0.000250 Time 0.023749 +2023-10-05 22:02:41,560 - Epoch: [171][ 250/ 1236] Overall Loss 0.184811 Objective Loss 0.184811 LR 0.000250 Time 0.023616 +2023-10-05 22:02:41,765 - Epoch: [171][ 260/ 1236] Overall Loss 0.184856 Objective Loss 0.184856 LR 0.000250 Time 0.023496 +2023-10-05 22:02:41,970 - Epoch: [171][ 270/ 1236] Overall Loss 0.184721 Objective Loss 0.184721 LR 0.000250 Time 0.023382 +2023-10-05 22:02:42,175 - Epoch: [171][ 280/ 1236] Overall Loss 0.184435 Objective Loss 0.184435 LR 0.000250 Time 0.023279 +2023-10-05 22:02:42,380 - Epoch: [171][ 290/ 1236] Overall Loss 0.184795 Objective Loss 0.184795 LR 0.000250 Time 0.023181 +2023-10-05 22:02:42,585 - Epoch: [171][ 300/ 1236] Overall Loss 0.185285 Objective Loss 0.185285 LR 0.000250 Time 0.023092 +2023-10-05 22:02:42,789 - Epoch: [171][ 310/ 1236] Overall Loss 0.185827 Objective Loss 0.185827 LR 0.000250 Time 0.023006 +2023-10-05 22:02:42,995 - Epoch: [171][ 320/ 1236] Overall Loss 0.186061 Objective Loss 0.186061 LR 0.000250 Time 0.022928 +2023-10-05 22:02:43,200 - Epoch: [171][ 330/ 1236] Overall Loss 0.186356 Objective Loss 0.186356 LR 0.000250 Time 0.022852 +2023-10-05 22:02:43,405 - Epoch: [171][ 340/ 1236] Overall Loss 0.186027 Objective Loss 0.186027 LR 0.000250 Time 0.022783 +2023-10-05 22:02:43,610 - Epoch: [171][ 350/ 1236] Overall Loss 0.185951 Objective Loss 0.185951 LR 0.000250 Time 0.022716 +2023-10-05 22:02:43,815 - Epoch: [171][ 360/ 1236] Overall Loss 0.185773 Objective Loss 0.185773 LR 0.000250 Time 0.022655 +2023-10-05 22:02:44,019 - Epoch: [171][ 370/ 1236] Overall Loss 0.186173 Objective Loss 0.186173 LR 0.000250 Time 0.022594 +2023-10-05 22:02:44,222 - Epoch: [171][ 380/ 1236] Overall Loss 0.186133 Objective Loss 0.186133 LR 0.000250 Time 0.022532 +2023-10-05 22:02:44,426 - Epoch: [171][ 390/ 1236] Overall Loss 0.186534 Objective Loss 0.186534 LR 0.000250 Time 0.022476 +2023-10-05 22:02:44,631 - Epoch: [171][ 400/ 1236] Overall Loss 0.186823 Objective Loss 0.186823 LR 0.000250 Time 0.022425 +2023-10-05 22:02:44,838 - Epoch: [171][ 410/ 1236] Overall Loss 0.186681 Objective Loss 0.186681 LR 0.000250 Time 0.022383 +2023-10-05 22:02:45,043 - Epoch: [171][ 420/ 1236] Overall Loss 0.187113 Objective Loss 0.187113 LR 0.000250 Time 0.022337 +2023-10-05 22:02:45,247 - Epoch: [171][ 430/ 1236] Overall Loss 0.187023 Objective Loss 0.187023 LR 0.000250 Time 0.022290 +2023-10-05 22:02:45,451 - Epoch: [171][ 440/ 1236] Overall Loss 0.186902 Objective Loss 0.186902 LR 0.000250 Time 0.022246 +2023-10-05 22:02:45,654 - Epoch: [171][ 450/ 1236] Overall Loss 0.187118 Objective Loss 0.187118 LR 0.000250 Time 0.022203 +2023-10-05 22:02:45,858 - Epoch: [171][ 460/ 1236] Overall Loss 0.186853 Objective Loss 0.186853 LR 0.000250 Time 0.022163 +2023-10-05 22:02:46,063 - Epoch: [171][ 470/ 1236] Overall Loss 0.186754 Objective Loss 0.186754 LR 0.000250 Time 0.022126 +2023-10-05 22:02:46,270 - Epoch: [171][ 480/ 1236] Overall Loss 0.186753 Objective Loss 0.186753 LR 0.000250 Time 0.022097 +2023-10-05 22:02:46,482 - Epoch: [171][ 490/ 1236] Overall Loss 0.186565 Objective Loss 0.186565 LR 0.000250 Time 0.022078 +2023-10-05 22:02:46,686 - Epoch: [171][ 500/ 1236] Overall Loss 0.186579 Objective Loss 0.186579 LR 0.000250 Time 0.022044 +2023-10-05 22:02:46,890 - Epoch: [171][ 510/ 1236] Overall Loss 0.187060 Objective Loss 0.187060 LR 0.000250 Time 0.022010 +2023-10-05 22:02:47,094 - Epoch: [171][ 520/ 1236] Overall Loss 0.187248 Objective Loss 0.187248 LR 0.000250 Time 0.021978 +2023-10-05 22:02:47,298 - Epoch: [171][ 530/ 1236] Overall Loss 0.187274 Objective Loss 0.187274 LR 0.000250 Time 0.021948 +2023-10-05 22:02:47,501 - Epoch: [171][ 540/ 1236] Overall Loss 0.187256 Objective Loss 0.187256 LR 0.000250 Time 0.021917 +2023-10-05 22:02:47,705 - Epoch: [171][ 550/ 1236] Overall Loss 0.187041 Objective Loss 0.187041 LR 0.000250 Time 0.021888 +2023-10-05 22:02:47,911 - Epoch: [171][ 560/ 1236] Overall Loss 0.187079 Objective Loss 0.187079 LR 0.000250 Time 0.021865 +2023-10-05 22:02:48,117 - Epoch: [171][ 570/ 1236] Overall Loss 0.187099 Objective Loss 0.187099 LR 0.000250 Time 0.021842 +2023-10-05 22:02:48,325 - Epoch: [171][ 580/ 1236] Overall Loss 0.187357 Objective Loss 0.187357 LR 0.000250 Time 0.021824 +2023-10-05 22:02:48,529 - Epoch: [171][ 590/ 1236] Overall Loss 0.187804 Objective Loss 0.187804 LR 0.000250 Time 0.021799 +2023-10-05 22:02:48,737 - Epoch: [171][ 600/ 1236] Overall Loss 0.187707 Objective Loss 0.187707 LR 0.000250 Time 0.021781 +2023-10-05 22:02:48,944 - Epoch: [171][ 610/ 1236] Overall Loss 0.187732 Objective Loss 0.187732 LR 0.000250 Time 0.021763 +2023-10-05 22:02:49,150 - Epoch: [171][ 620/ 1236] Overall Loss 0.187868 Objective Loss 0.187868 LR 0.000250 Time 0.021744 +2023-10-05 22:02:49,358 - Epoch: [171][ 630/ 1236] Overall Loss 0.187593 Objective Loss 0.187593 LR 0.000250 Time 0.021728 +2023-10-05 22:02:49,564 - Epoch: [171][ 640/ 1236] Overall Loss 0.187681 Objective Loss 0.187681 LR 0.000250 Time 0.021710 +2023-10-05 22:02:49,771 - Epoch: [171][ 650/ 1236] Overall Loss 0.187470 Objective Loss 0.187470 LR 0.000250 Time 0.021693 +2023-10-05 22:02:49,977 - Epoch: [171][ 660/ 1236] Overall Loss 0.187153 Objective Loss 0.187153 LR 0.000250 Time 0.021677 +2023-10-05 22:02:50,184 - Epoch: [171][ 670/ 1236] Overall Loss 0.187334 Objective Loss 0.187334 LR 0.000250 Time 0.021661 +2023-10-05 22:02:50,390 - Epoch: [171][ 680/ 1236] Overall Loss 0.187107 Objective Loss 0.187107 LR 0.000250 Time 0.021645 +2023-10-05 22:02:50,597 - Epoch: [171][ 690/ 1236] Overall Loss 0.186871 Objective Loss 0.186871 LR 0.000250 Time 0.021631 +2023-10-05 22:02:50,803 - Epoch: [171][ 700/ 1236] Overall Loss 0.186877 Objective Loss 0.186877 LR 0.000250 Time 0.021616 +2023-10-05 22:02:51,014 - Epoch: [171][ 710/ 1236] Overall Loss 0.186517 Objective Loss 0.186517 LR 0.000250 Time 0.021608 +2023-10-05 22:02:51,220 - Epoch: [171][ 720/ 1236] Overall Loss 0.186292 Objective Loss 0.186292 LR 0.000250 Time 0.021593 +2023-10-05 22:02:51,430 - Epoch: [171][ 730/ 1236] Overall Loss 0.186322 Objective Loss 0.186322 LR 0.000250 Time 0.021585 +2023-10-05 22:02:51,636 - Epoch: [171][ 740/ 1236] Overall Loss 0.186322 Objective Loss 0.186322 LR 0.000250 Time 0.021571 +2023-10-05 22:02:51,845 - Epoch: [171][ 750/ 1236] Overall Loss 0.186512 Objective Loss 0.186512 LR 0.000250 Time 0.021561 +2023-10-05 22:02:52,053 - Epoch: [171][ 760/ 1236] Overall Loss 0.186556 Objective Loss 0.186556 LR 0.000250 Time 0.021551 +2023-10-05 22:02:52,265 - Epoch: [171][ 770/ 1236] Overall Loss 0.186516 Objective Loss 0.186516 LR 0.000250 Time 0.021545 +2023-10-05 22:02:52,472 - Epoch: [171][ 780/ 1236] Overall Loss 0.186344 Objective Loss 0.186344 LR 0.000250 Time 0.021535 +2023-10-05 22:02:52,684 - Epoch: [171][ 790/ 1236] Overall Loss 0.186584 Objective Loss 0.186584 LR 0.000250 Time 0.021529 +2023-10-05 22:02:52,892 - Epoch: [171][ 800/ 1236] Overall Loss 0.186595 Objective Loss 0.186595 LR 0.000250 Time 0.021520 +2023-10-05 22:02:53,098 - Epoch: [171][ 810/ 1236] Overall Loss 0.186758 Objective Loss 0.186758 LR 0.000250 Time 0.021508 +2023-10-05 22:02:53,304 - Epoch: [171][ 820/ 1236] Overall Loss 0.186787 Objective Loss 0.186787 LR 0.000250 Time 0.021497 +2023-10-05 22:02:53,511 - Epoch: [171][ 830/ 1236] Overall Loss 0.186879 Objective Loss 0.186879 LR 0.000250 Time 0.021486 +2023-10-05 22:02:53,717 - Epoch: [171][ 840/ 1236] Overall Loss 0.186916 Objective Loss 0.186916 LR 0.000250 Time 0.021475 +2023-10-05 22:02:53,923 - Epoch: [171][ 850/ 1236] Overall Loss 0.186982 Objective Loss 0.186982 LR 0.000250 Time 0.021464 +2023-10-05 22:02:54,129 - Epoch: [171][ 860/ 1236] Overall Loss 0.187092 Objective Loss 0.187092 LR 0.000250 Time 0.021453 +2023-10-05 22:02:54,334 - Epoch: [171][ 870/ 1236] Overall Loss 0.186941 Objective Loss 0.186941 LR 0.000250 Time 0.021442 +2023-10-05 22:02:54,540 - Epoch: [171][ 880/ 1236] Overall Loss 0.186864 Objective Loss 0.186864 LR 0.000250 Time 0.021432 +2023-10-05 22:02:54,746 - Epoch: [171][ 890/ 1236] Overall Loss 0.187090 Objective Loss 0.187090 LR 0.000250 Time 0.021422 +2023-10-05 22:02:54,952 - Epoch: [171][ 900/ 1236] Overall Loss 0.187071 Objective Loss 0.187071 LR 0.000250 Time 0.021413 +2023-10-05 22:02:55,159 - Epoch: [171][ 910/ 1236] Overall Loss 0.187346 Objective Loss 0.187346 LR 0.000250 Time 0.021404 +2023-10-05 22:02:55,365 - Epoch: [171][ 920/ 1236] Overall Loss 0.187319 Objective Loss 0.187319 LR 0.000250 Time 0.021395 +2023-10-05 22:02:55,572 - Epoch: [171][ 930/ 1236] Overall Loss 0.187423 Objective Loss 0.187423 LR 0.000250 Time 0.021386 +2023-10-05 22:02:55,777 - Epoch: [171][ 940/ 1236] Overall Loss 0.187583 Objective Loss 0.187583 LR 0.000250 Time 0.021377 +2023-10-05 22:02:55,984 - Epoch: [171][ 950/ 1236] Overall Loss 0.187487 Objective Loss 0.187487 LR 0.000250 Time 0.021369 +2023-10-05 22:02:56,190 - Epoch: [171][ 960/ 1236] Overall Loss 0.187486 Objective Loss 0.187486 LR 0.000250 Time 0.021360 +2023-10-05 22:02:56,402 - Epoch: [171][ 970/ 1236] Overall Loss 0.187501 Objective Loss 0.187501 LR 0.000250 Time 0.021359 +2023-10-05 22:02:56,621 - Epoch: [171][ 980/ 1236] Overall Loss 0.187615 Objective Loss 0.187615 LR 0.000250 Time 0.021364 +2023-10-05 22:02:56,834 - Epoch: [171][ 990/ 1236] Overall Loss 0.187721 Objective Loss 0.187721 LR 0.000250 Time 0.021362 +2023-10-05 22:02:57,052 - Epoch: [171][ 1000/ 1236] Overall Loss 0.187702 Objective Loss 0.187702 LR 0.000250 Time 0.021366 +2023-10-05 22:02:57,264 - Epoch: [171][ 1010/ 1236] Overall Loss 0.187835 Objective Loss 0.187835 LR 0.000250 Time 0.021364 +2023-10-05 22:02:57,479 - Epoch: [171][ 1020/ 1236] Overall Loss 0.187991 Objective Loss 0.187991 LR 0.000250 Time 0.021365 +2023-10-05 22:02:57,688 - Epoch: [171][ 1030/ 1236] Overall Loss 0.188033 Objective Loss 0.188033 LR 0.000250 Time 0.021360 +2023-10-05 22:02:57,901 - Epoch: [171][ 1040/ 1236] Overall Loss 0.188003 Objective Loss 0.188003 LR 0.000250 Time 0.021359 +2023-10-05 22:02:58,111 - Epoch: [171][ 1050/ 1236] Overall Loss 0.188110 Objective Loss 0.188110 LR 0.000250 Time 0.021355 +2023-10-05 22:02:58,324 - Epoch: [171][ 1060/ 1236] Overall Loss 0.188300 Objective Loss 0.188300 LR 0.000250 Time 0.021355 +2023-10-05 22:02:58,534 - Epoch: [171][ 1070/ 1236] Overall Loss 0.188063 Objective Loss 0.188063 LR 0.000250 Time 0.021351 +2023-10-05 22:02:58,747 - Epoch: [171][ 1080/ 1236] Overall Loss 0.188560 Objective Loss 0.188560 LR 0.000250 Time 0.021350 +2023-10-05 22:02:58,956 - Epoch: [171][ 1090/ 1236] Overall Loss 0.188738 Objective Loss 0.188738 LR 0.000250 Time 0.021346 +2023-10-05 22:02:59,169 - Epoch: [171][ 1100/ 1236] Overall Loss 0.188861 Objective Loss 0.188861 LR 0.000250 Time 0.021345 +2023-10-05 22:02:59,379 - Epoch: [171][ 1110/ 1236] Overall Loss 0.189064 Objective Loss 0.189064 LR 0.000250 Time 0.021341 +2023-10-05 22:02:59,600 - Epoch: [171][ 1120/ 1236] Overall Loss 0.189107 Objective Loss 0.189107 LR 0.000250 Time 0.021347 +2023-10-05 22:02:59,819 - Epoch: [171][ 1130/ 1236] Overall Loss 0.189159 Objective Loss 0.189159 LR 0.000250 Time 0.021351 +2023-10-05 22:03:00,041 - Epoch: [171][ 1140/ 1236] Overall Loss 0.189313 Objective Loss 0.189313 LR 0.000250 Time 0.021359 +2023-10-05 22:03:00,260 - Epoch: [171][ 1150/ 1236] Overall Loss 0.189240 Objective Loss 0.189240 LR 0.000250 Time 0.021363 +2023-10-05 22:03:00,483 - Epoch: [171][ 1160/ 1236] Overall Loss 0.189118 Objective Loss 0.189118 LR 0.000250 Time 0.021370 +2023-10-05 22:03:00,702 - Epoch: [171][ 1170/ 1236] Overall Loss 0.189140 Objective Loss 0.189140 LR 0.000250 Time 0.021375 +2023-10-05 22:03:00,924 - Epoch: [171][ 1180/ 1236] Overall Loss 0.189141 Objective Loss 0.189141 LR 0.000250 Time 0.021381 +2023-10-05 22:03:01,142 - Epoch: [171][ 1190/ 1236] Overall Loss 0.189026 Objective Loss 0.189026 LR 0.000250 Time 0.021384 +2023-10-05 22:03:01,364 - Epoch: [171][ 1200/ 1236] Overall Loss 0.188983 Objective Loss 0.188983 LR 0.000250 Time 0.021391 +2023-10-05 22:03:01,584 - Epoch: [171][ 1210/ 1236] Overall Loss 0.188911 Objective Loss 0.188911 LR 0.000250 Time 0.021395 +2023-10-05 22:03:01,806 - Epoch: [171][ 1220/ 1236] Overall Loss 0.188998 Objective Loss 0.188998 LR 0.000250 Time 0.021401 +2023-10-05 22:03:02,083 - Epoch: [171][ 1230/ 1236] Overall Loss 0.189116 Objective Loss 0.189116 LR 0.000250 Time 0.021452 +2023-10-05 22:03:02,205 - Epoch: [171][ 1236/ 1236] Overall Loss 0.189148 Objective Loss 0.189148 Top1 89.816701 Top5 98.574338 LR 0.000250 Time 0.021446 +2023-10-05 22:03:02,342 - --- validate (epoch=171)----------- +2023-10-05 22:03:02,343 - 29943 samples (256 per mini-batch) +2023-10-05 22:03:02,846 - Epoch: [171][ 10/ 117] Loss 0.278518 Top1 86.328125 Top5 98.125000 +2023-10-05 22:03:02,996 - Epoch: [171][ 20/ 117] Loss 0.299653 Top1 85.468750 Top5 98.183594 +2023-10-05 22:03:03,147 - Epoch: [171][ 30/ 117] Loss 0.302712 Top1 85.429688 Top5 98.151042 +2023-10-05 22:03:03,295 - Epoch: [171][ 40/ 117] Loss 0.312390 Top1 85.322266 Top5 98.193359 +2023-10-05 22:03:03,446 - Epoch: [171][ 50/ 117] Loss 0.313144 Top1 85.132812 Top5 98.054688 +2023-10-05 22:03:03,595 - Epoch: [171][ 60/ 117] Loss 0.309417 Top1 85.123698 Top5 98.072917 +2023-10-05 22:03:03,752 - Epoch: [171][ 70/ 117] Loss 0.306076 Top1 85.117188 Top5 98.119420 +2023-10-05 22:03:03,911 - Epoch: [171][ 80/ 117] Loss 0.305832 Top1 85.190430 Top5 98.105469 +2023-10-05 22:03:04,069 - Epoch: [171][ 90/ 117] Loss 0.306189 Top1 85.164931 Top5 98.125000 +2023-10-05 22:03:04,227 - Epoch: [171][ 100/ 117] Loss 0.308650 Top1 85.207031 Top5 98.101562 +2023-10-05 22:03:04,393 - Epoch: [171][ 110/ 117] Loss 0.309337 Top1 85.156250 Top5 98.103693 +2023-10-05 22:03:04,479 - Epoch: [171][ 117/ 117] Loss 0.305681 Top1 85.218582 Top5 98.109742 +2023-10-05 22:03:04,620 - ==> Top1: 85.219 Top5: 98.110 Loss: 0.306 + +2023-10-05 22:03:04,620 - ==> Confusion: +[[ 929 4 2 1 12 4 0 0 4 72 1 0 1 2 6 2 1 0 0 0 9] + [ 1 1066 3 0 6 13 1 21 2 0 1 0 0 1 0 4 1 0 7 0 4] + [ 7 1 966 12 2 0 21 11 0 0 3 1 9 2 1 4 2 1 7 3 3] + [ 5 1 7 968 0 3 1 2 1 1 11 1 6 2 25 3 0 3 34 0 15] + [ 17 4 0 0 994 3 0 1 0 8 0 1 1 1 9 2 5 2 0 2 0] + [ 3 36 1 0 6 991 1 18 0 3 5 9 0 9 6 2 4 0 5 2 15] + [ 0 4 22 0 2 0 1128 11 0 0 2 2 1 0 1 7 0 1 0 7 3] + [ 2 15 11 0 1 28 6 1086 0 2 2 9 1 3 0 3 0 0 36 7 6] + [ 21 1 1 1 0 2 1 0 966 39 12 2 1 15 17 6 0 1 2 0 1] + [ 92 0 3 2 8 1 0 1 16 953 0 2 0 21 9 6 0 0 0 0 5] + [ 2 5 8 4 1 2 5 6 10 1 973 3 0 12 3 1 1 2 5 1 8] + [ 1 0 1 0 1 12 0 4 0 1 0 947 30 6 0 4 2 16 0 5 5] + [ 1 0 3 6 0 3 0 3 0 0 1 31 988 2 1 2 2 15 1 3 6] + [ 2 0 1 0 3 4 0 0 6 9 4 5 2 1071 2 1 1 0 0 0 8] + [ 11 1 3 9 8 0 0 0 20 1 6 1 1 2 1009 0 1 1 16 0 11] + [ 1 4 2 0 2 1 1 0 0 0 1 7 4 1 1 1073 13 12 0 7 4] + [ 1 14 1 1 12 2 0 1 1 0 0 4 1 1 4 12 1093 0 0 3 10] + [ 0 0 0 3 1 0 3 0 1 0 0 0 17 1 0 5 0 1003 2 0 2] + [ 1 7 5 15 1 0 2 27 1 0 4 1 2 1 8 0 0 0 984 0 9] + [ 0 3 2 2 3 8 10 9 0 0 2 14 3 3 0 6 11 3 2 1061 10] + [ 110 184 133 65 140 116 41 100 90 60 168 87 323 300 141 62 144 71 145 157 5268]] + +2023-10-05 22:03:04,622 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:03:04,622 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:03:04,631 - + +2023-10-05 22:03:04,631 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:03:05,752 - Epoch: [172][ 10/ 1236] Overall Loss 0.184954 Objective Loss 0.184954 LR 0.000250 Time 0.112005 +2023-10-05 22:03:05,955 - Epoch: [172][ 20/ 1236] Overall Loss 0.179951 Objective Loss 0.179951 LR 0.000250 Time 0.066146 +2023-10-05 22:03:06,156 - Epoch: [172][ 30/ 1236] Overall Loss 0.191195 Objective Loss 0.191195 LR 0.000250 Time 0.050778 +2023-10-05 22:03:06,359 - Epoch: [172][ 40/ 1236] Overall Loss 0.187407 Objective Loss 0.187407 LR 0.000250 Time 0.043148 +2023-10-05 22:03:06,560 - Epoch: [172][ 50/ 1236] Overall Loss 0.188260 Objective Loss 0.188260 LR 0.000250 Time 0.038536 +2023-10-05 22:03:06,763 - Epoch: [172][ 60/ 1236] Overall Loss 0.188637 Objective Loss 0.188637 LR 0.000250 Time 0.035495 +2023-10-05 22:03:06,964 - Epoch: [172][ 70/ 1236] Overall Loss 0.188888 Objective Loss 0.188888 LR 0.000250 Time 0.033291 +2023-10-05 22:03:07,167 - Epoch: [172][ 80/ 1236] Overall Loss 0.188567 Objective Loss 0.188567 LR 0.000250 Time 0.031657 +2023-10-05 22:03:07,366 - Epoch: [172][ 90/ 1236] Overall Loss 0.188294 Objective Loss 0.188294 LR 0.000250 Time 0.030347 +2023-10-05 22:03:07,567 - Epoch: [172][ 100/ 1236] Overall Loss 0.188033 Objective Loss 0.188033 LR 0.000250 Time 0.029324 +2023-10-05 22:03:07,769 - Epoch: [172][ 110/ 1236] Overall Loss 0.188914 Objective Loss 0.188914 LR 0.000250 Time 0.028484 +2023-10-05 22:03:07,978 - Epoch: [172][ 120/ 1236] Overall Loss 0.188752 Objective Loss 0.188752 LR 0.000250 Time 0.027850 +2023-10-05 22:03:08,190 - Epoch: [172][ 130/ 1236] Overall Loss 0.188710 Objective Loss 0.188710 LR 0.000250 Time 0.027331 +2023-10-05 22:03:08,405 - Epoch: [172][ 140/ 1236] Overall Loss 0.189437 Objective Loss 0.189437 LR 0.000250 Time 0.026912 +2023-10-05 22:03:08,616 - Epoch: [172][ 150/ 1236] Overall Loss 0.188375 Objective Loss 0.188375 LR 0.000250 Time 0.026524 +2023-10-05 22:03:08,833 - Epoch: [172][ 160/ 1236] Overall Loss 0.187745 Objective Loss 0.187745 LR 0.000250 Time 0.026217 +2023-10-05 22:03:09,037 - Epoch: [172][ 170/ 1236] Overall Loss 0.186559 Objective Loss 0.186559 LR 0.000250 Time 0.025873 +2023-10-05 22:03:09,243 - Epoch: [172][ 180/ 1236] Overall Loss 0.184976 Objective Loss 0.184976 LR 0.000250 Time 0.025581 +2023-10-05 22:03:09,448 - Epoch: [172][ 190/ 1236] Overall Loss 0.184647 Objective Loss 0.184647 LR 0.000250 Time 0.025308 +2023-10-05 22:03:09,655 - Epoch: [172][ 200/ 1236] Overall Loss 0.185130 Objective Loss 0.185130 LR 0.000250 Time 0.025079 +2023-10-05 22:03:09,862 - Epoch: [172][ 210/ 1236] Overall Loss 0.185505 Objective Loss 0.185505 LR 0.000250 Time 0.024868 +2023-10-05 22:03:10,069 - Epoch: [172][ 220/ 1236] Overall Loss 0.185907 Objective Loss 0.185907 LR 0.000250 Time 0.024674 +2023-10-05 22:03:10,275 - Epoch: [172][ 230/ 1236] Overall Loss 0.186187 Objective Loss 0.186187 LR 0.000250 Time 0.024496 +2023-10-05 22:03:10,482 - Epoch: [172][ 240/ 1236] Overall Loss 0.186311 Objective Loss 0.186311 LR 0.000250 Time 0.024334 +2023-10-05 22:03:10,688 - Epoch: [172][ 250/ 1236] Overall Loss 0.185440 Objective Loss 0.185440 LR 0.000250 Time 0.024185 +2023-10-05 22:03:10,895 - Epoch: [172][ 260/ 1236] Overall Loss 0.185520 Objective Loss 0.185520 LR 0.000250 Time 0.024047 +2023-10-05 22:03:11,101 - Epoch: [172][ 270/ 1236] Overall Loss 0.186499 Objective Loss 0.186499 LR 0.000250 Time 0.023920 +2023-10-05 22:03:11,308 - Epoch: [172][ 280/ 1236] Overall Loss 0.186186 Objective Loss 0.186186 LR 0.000250 Time 0.023801 +2023-10-05 22:03:11,515 - Epoch: [172][ 290/ 1236] Overall Loss 0.186157 Objective Loss 0.186157 LR 0.000250 Time 0.023692 +2023-10-05 22:03:11,721 - Epoch: [172][ 300/ 1236] Overall Loss 0.185638 Objective Loss 0.185638 LR 0.000250 Time 0.023589 +2023-10-05 22:03:11,929 - Epoch: [172][ 310/ 1236] Overall Loss 0.185778 Objective Loss 0.185778 LR 0.000250 Time 0.023496 +2023-10-05 22:03:12,134 - Epoch: [172][ 320/ 1236] Overall Loss 0.185126 Objective Loss 0.185126 LR 0.000250 Time 0.023402 +2023-10-05 22:03:12,339 - Epoch: [172][ 330/ 1236] Overall Loss 0.184949 Objective Loss 0.184949 LR 0.000250 Time 0.023312 +2023-10-05 22:03:12,544 - Epoch: [172][ 340/ 1236] Overall Loss 0.185339 Objective Loss 0.185339 LR 0.000250 Time 0.023229 +2023-10-05 22:03:12,749 - Epoch: [172][ 350/ 1236] Overall Loss 0.185778 Objective Loss 0.185778 LR 0.000250 Time 0.023149 +2023-10-05 22:03:12,953 - Epoch: [172][ 360/ 1236] Overall Loss 0.185635 Objective Loss 0.185635 LR 0.000250 Time 0.023073 +2023-10-05 22:03:13,156 - Epoch: [172][ 370/ 1236] Overall Loss 0.185504 Objective Loss 0.185504 LR 0.000250 Time 0.022998 +2023-10-05 22:03:13,361 - Epoch: [172][ 380/ 1236] Overall Loss 0.185356 Objective Loss 0.185356 LR 0.000250 Time 0.022929 +2023-10-05 22:03:13,566 - Epoch: [172][ 390/ 1236] Overall Loss 0.185326 Objective Loss 0.185326 LR 0.000250 Time 0.022866 +2023-10-05 22:03:13,772 - Epoch: [172][ 400/ 1236] Overall Loss 0.185168 Objective Loss 0.185168 LR 0.000250 Time 0.022810 +2023-10-05 22:03:13,979 - Epoch: [172][ 410/ 1236] Overall Loss 0.185518 Objective Loss 0.185518 LR 0.000250 Time 0.022758 +2023-10-05 22:03:14,188 - Epoch: [172][ 420/ 1236] Overall Loss 0.184902 Objective Loss 0.184902 LR 0.000250 Time 0.022712 +2023-10-05 22:03:14,395 - Epoch: [172][ 430/ 1236] Overall Loss 0.185198 Objective Loss 0.185198 LR 0.000250 Time 0.022664 +2023-10-05 22:03:14,602 - Epoch: [172][ 440/ 1236] Overall Loss 0.185224 Objective Loss 0.185224 LR 0.000250 Time 0.022617 +2023-10-05 22:03:14,805 - Epoch: [172][ 450/ 1236] Overall Loss 0.185659 Objective Loss 0.185659 LR 0.000250 Time 0.022565 +2023-10-05 22:03:15,009 - Epoch: [172][ 460/ 1236] Overall Loss 0.185730 Objective Loss 0.185730 LR 0.000250 Time 0.022518 +2023-10-05 22:03:15,212 - Epoch: [172][ 470/ 1236] Overall Loss 0.185613 Objective Loss 0.185613 LR 0.000250 Time 0.022471 +2023-10-05 22:03:15,417 - Epoch: [172][ 480/ 1236] Overall Loss 0.185962 Objective Loss 0.185962 LR 0.000250 Time 0.022428 +2023-10-05 22:03:15,620 - Epoch: [172][ 490/ 1236] Overall Loss 0.185778 Objective Loss 0.185778 LR 0.000250 Time 0.022384 +2023-10-05 22:03:15,824 - Epoch: [172][ 500/ 1236] Overall Loss 0.186020 Objective Loss 0.186020 LR 0.000250 Time 0.022343 +2023-10-05 22:03:16,027 - Epoch: [172][ 510/ 1236] Overall Loss 0.186545 Objective Loss 0.186545 LR 0.000250 Time 0.022302 +2023-10-05 22:03:16,231 - Epoch: [172][ 520/ 1236] Overall Loss 0.186362 Objective Loss 0.186362 LR 0.000250 Time 0.022266 +2023-10-05 22:03:16,435 - Epoch: [172][ 530/ 1236] Overall Loss 0.186458 Objective Loss 0.186458 LR 0.000250 Time 0.022229 +2023-10-05 22:03:16,639 - Epoch: [172][ 540/ 1236] Overall Loss 0.186212 Objective Loss 0.186212 LR 0.000250 Time 0.022194 +2023-10-05 22:03:16,842 - Epoch: [172][ 550/ 1236] Overall Loss 0.186227 Objective Loss 0.186227 LR 0.000250 Time 0.022159 +2023-10-05 22:03:17,046 - Epoch: [172][ 560/ 1236] Overall Loss 0.186783 Objective Loss 0.186783 LR 0.000250 Time 0.022127 +2023-10-05 22:03:17,249 - Epoch: [172][ 570/ 1236] Overall Loss 0.186877 Objective Loss 0.186877 LR 0.000250 Time 0.022095 +2023-10-05 22:03:17,453 - Epoch: [172][ 580/ 1236] Overall Loss 0.186773 Objective Loss 0.186773 LR 0.000250 Time 0.022066 +2023-10-05 22:03:17,657 - Epoch: [172][ 590/ 1236] Overall Loss 0.186648 Objective Loss 0.186648 LR 0.000250 Time 0.022035 +2023-10-05 22:03:17,861 - Epoch: [172][ 600/ 1236] Overall Loss 0.187034 Objective Loss 0.187034 LR 0.000250 Time 0.022008 +2023-10-05 22:03:18,064 - Epoch: [172][ 610/ 1236] Overall Loss 0.187002 Objective Loss 0.187002 LR 0.000250 Time 0.021980 +2023-10-05 22:03:18,269 - Epoch: [172][ 620/ 1236] Overall Loss 0.187465 Objective Loss 0.187465 LR 0.000250 Time 0.021955 +2023-10-05 22:03:18,473 - Epoch: [172][ 630/ 1236] Overall Loss 0.187840 Objective Loss 0.187840 LR 0.000250 Time 0.021929 +2023-10-05 22:03:18,677 - Epoch: [172][ 640/ 1236] Overall Loss 0.187784 Objective Loss 0.187784 LR 0.000250 Time 0.021905 +2023-10-05 22:03:18,880 - Epoch: [172][ 650/ 1236] Overall Loss 0.187995 Objective Loss 0.187995 LR 0.000250 Time 0.021880 +2023-10-05 22:03:19,084 - Epoch: [172][ 660/ 1236] Overall Loss 0.188147 Objective Loss 0.188147 LR 0.000250 Time 0.021857 +2023-10-05 22:03:19,287 - Epoch: [172][ 670/ 1236] Overall Loss 0.188074 Objective Loss 0.188074 LR 0.000250 Time 0.021833 +2023-10-05 22:03:19,491 - Epoch: [172][ 680/ 1236] Overall Loss 0.187612 Objective Loss 0.187612 LR 0.000250 Time 0.021811 +2023-10-05 22:03:19,694 - Epoch: [172][ 690/ 1236] Overall Loss 0.187312 Objective Loss 0.187312 LR 0.000250 Time 0.021789 +2023-10-05 22:03:19,897 - Epoch: [172][ 700/ 1236] Overall Loss 0.187211 Objective Loss 0.187211 LR 0.000250 Time 0.021768 +2023-10-05 22:03:20,101 - Epoch: [172][ 710/ 1236] Overall Loss 0.187684 Objective Loss 0.187684 LR 0.000250 Time 0.021747 +2023-10-05 22:03:20,305 - Epoch: [172][ 720/ 1236] Overall Loss 0.187656 Objective Loss 0.187656 LR 0.000250 Time 0.021728 +2023-10-05 22:03:20,508 - Epoch: [172][ 730/ 1236] Overall Loss 0.187800 Objective Loss 0.187800 LR 0.000250 Time 0.021708 +2023-10-05 22:03:20,712 - Epoch: [172][ 740/ 1236] Overall Loss 0.187816 Objective Loss 0.187816 LR 0.000250 Time 0.021690 +2023-10-05 22:03:20,915 - Epoch: [172][ 750/ 1236] Overall Loss 0.187795 Objective Loss 0.187795 LR 0.000250 Time 0.021671 +2023-10-05 22:03:21,119 - Epoch: [172][ 760/ 1236] Overall Loss 0.187903 Objective Loss 0.187903 LR 0.000250 Time 0.021654 +2023-10-05 22:03:21,322 - Epoch: [172][ 770/ 1236] Overall Loss 0.187724 Objective Loss 0.187724 LR 0.000250 Time 0.021636 +2023-10-05 22:03:21,526 - Epoch: [172][ 780/ 1236] Overall Loss 0.187717 Objective Loss 0.187717 LR 0.000250 Time 0.021620 +2023-10-05 22:03:21,729 - Epoch: [172][ 790/ 1236] Overall Loss 0.187650 Objective Loss 0.187650 LR 0.000250 Time 0.021603 +2023-10-05 22:03:21,934 - Epoch: [172][ 800/ 1236] Overall Loss 0.187797 Objective Loss 0.187797 LR 0.000250 Time 0.021588 +2023-10-05 22:03:22,137 - Epoch: [172][ 810/ 1236] Overall Loss 0.187811 Objective Loss 0.187811 LR 0.000250 Time 0.021571 +2023-10-05 22:03:22,341 - Epoch: [172][ 820/ 1236] Overall Loss 0.187680 Objective Loss 0.187680 LR 0.000250 Time 0.021557 +2023-10-05 22:03:22,544 - Epoch: [172][ 830/ 1236] Overall Loss 0.187742 Objective Loss 0.187742 LR 0.000250 Time 0.021541 +2023-10-05 22:03:22,748 - Epoch: [172][ 840/ 1236] Overall Loss 0.187713 Objective Loss 0.187713 LR 0.000250 Time 0.021527 +2023-10-05 22:03:22,951 - Epoch: [172][ 850/ 1236] Overall Loss 0.187642 Objective Loss 0.187642 LR 0.000250 Time 0.021513 +2023-10-05 22:03:23,155 - Epoch: [172][ 860/ 1236] Overall Loss 0.187550 Objective Loss 0.187550 LR 0.000250 Time 0.021499 +2023-10-05 22:03:23,358 - Epoch: [172][ 870/ 1236] Overall Loss 0.187696 Objective Loss 0.187696 LR 0.000250 Time 0.021485 +2023-10-05 22:03:23,562 - Epoch: [172][ 880/ 1236] Overall Loss 0.187752 Objective Loss 0.187752 LR 0.000250 Time 0.021473 +2023-10-05 22:03:23,765 - Epoch: [172][ 890/ 1236] Overall Loss 0.187618 Objective Loss 0.187618 LR 0.000250 Time 0.021459 +2023-10-05 22:03:23,969 - Epoch: [172][ 900/ 1236] Overall Loss 0.187577 Objective Loss 0.187577 LR 0.000250 Time 0.021447 +2023-10-05 22:03:24,172 - Epoch: [172][ 910/ 1236] Overall Loss 0.187501 Objective Loss 0.187501 LR 0.000250 Time 0.021434 +2023-10-05 22:03:24,377 - Epoch: [172][ 920/ 1236] Overall Loss 0.187477 Objective Loss 0.187477 LR 0.000250 Time 0.021423 +2023-10-05 22:03:24,580 - Epoch: [172][ 930/ 1236] Overall Loss 0.187538 Objective Loss 0.187538 LR 0.000250 Time 0.021411 +2023-10-05 22:03:24,784 - Epoch: [172][ 940/ 1236] Overall Loss 0.187889 Objective Loss 0.187889 LR 0.000250 Time 0.021400 +2023-10-05 22:03:24,987 - Epoch: [172][ 950/ 1236] Overall Loss 0.187925 Objective Loss 0.187925 LR 0.000250 Time 0.021388 +2023-10-05 22:03:25,192 - Epoch: [172][ 960/ 1236] Overall Loss 0.188083 Objective Loss 0.188083 LR 0.000250 Time 0.021377 +2023-10-05 22:03:25,395 - Epoch: [172][ 970/ 1236] Overall Loss 0.188057 Objective Loss 0.188057 LR 0.000250 Time 0.021366 +2023-10-05 22:03:25,599 - Epoch: [172][ 980/ 1236] Overall Loss 0.188303 Objective Loss 0.188303 LR 0.000250 Time 0.021356 +2023-10-05 22:03:25,802 - Epoch: [172][ 990/ 1236] Overall Loss 0.188326 Objective Loss 0.188326 LR 0.000250 Time 0.021345 +2023-10-05 22:03:26,006 - Epoch: [172][ 1000/ 1236] Overall Loss 0.188465 Objective Loss 0.188465 LR 0.000250 Time 0.021335 +2023-10-05 22:03:26,209 - Epoch: [172][ 1010/ 1236] Overall Loss 0.188564 Objective Loss 0.188564 LR 0.000250 Time 0.021325 +2023-10-05 22:03:26,413 - Epoch: [172][ 1020/ 1236] Overall Loss 0.188485 Objective Loss 0.188485 LR 0.000250 Time 0.021315 +2023-10-05 22:03:26,616 - Epoch: [172][ 1030/ 1236] Overall Loss 0.188401 Objective Loss 0.188401 LR 0.000250 Time 0.021305 +2023-10-05 22:03:26,820 - Epoch: [172][ 1040/ 1236] Overall Loss 0.188405 Objective Loss 0.188405 LR 0.000250 Time 0.021296 +2023-10-05 22:03:27,023 - Epoch: [172][ 1050/ 1236] Overall Loss 0.188398 Objective Loss 0.188398 LR 0.000250 Time 0.021286 +2023-10-05 22:03:27,227 - Epoch: [172][ 1060/ 1236] Overall Loss 0.188308 Objective Loss 0.188308 LR 0.000250 Time 0.021278 +2023-10-05 22:03:27,431 - Epoch: [172][ 1070/ 1236] Overall Loss 0.188276 Objective Loss 0.188276 LR 0.000250 Time 0.021268 +2023-10-05 22:03:27,635 - Epoch: [172][ 1080/ 1236] Overall Loss 0.188272 Objective Loss 0.188272 LR 0.000250 Time 0.021260 +2023-10-05 22:03:27,838 - Epoch: [172][ 1090/ 1236] Overall Loss 0.188312 Objective Loss 0.188312 LR 0.000250 Time 0.021251 +2023-10-05 22:03:28,042 - Epoch: [172][ 1100/ 1236] Overall Loss 0.188379 Objective Loss 0.188379 LR 0.000250 Time 0.021243 +2023-10-05 22:03:28,245 - Epoch: [172][ 1110/ 1236] Overall Loss 0.188445 Objective Loss 0.188445 LR 0.000250 Time 0.021234 +2023-10-05 22:03:28,449 - Epoch: [172][ 1120/ 1236] Overall Loss 0.188479 Objective Loss 0.188479 LR 0.000250 Time 0.021227 +2023-10-05 22:03:28,653 - Epoch: [172][ 1130/ 1236] Overall Loss 0.188482 Objective Loss 0.188482 LR 0.000250 Time 0.021218 +2023-10-05 22:03:28,857 - Epoch: [172][ 1140/ 1236] Overall Loss 0.188579 Objective Loss 0.188579 LR 0.000250 Time 0.021211 +2023-10-05 22:03:29,060 - Epoch: [172][ 1150/ 1236] Overall Loss 0.188813 Objective Loss 0.188813 LR 0.000250 Time 0.021203 +2023-10-05 22:03:29,264 - Epoch: [172][ 1160/ 1236] Overall Loss 0.188710 Objective Loss 0.188710 LR 0.000250 Time 0.021196 +2023-10-05 22:03:29,468 - Epoch: [172][ 1170/ 1236] Overall Loss 0.188892 Objective Loss 0.188892 LR 0.000250 Time 0.021188 +2023-10-05 22:03:29,672 - Epoch: [172][ 1180/ 1236] Overall Loss 0.188837 Objective Loss 0.188837 LR 0.000250 Time 0.021182 +2023-10-05 22:03:29,875 - Epoch: [172][ 1190/ 1236] Overall Loss 0.188762 Objective Loss 0.188762 LR 0.000250 Time 0.021174 +2023-10-05 22:03:30,079 - Epoch: [172][ 1200/ 1236] Overall Loss 0.188615 Objective Loss 0.188615 LR 0.000250 Time 0.021168 +2023-10-05 22:03:30,283 - Epoch: [172][ 1210/ 1236] Overall Loss 0.188691 Objective Loss 0.188691 LR 0.000250 Time 0.021160 +2023-10-05 22:03:30,487 - Epoch: [172][ 1220/ 1236] Overall Loss 0.188760 Objective Loss 0.188760 LR 0.000250 Time 0.021154 +2023-10-05 22:03:30,743 - Epoch: [172][ 1230/ 1236] Overall Loss 0.188921 Objective Loss 0.188921 LR 0.000250 Time 0.021190 +2023-10-05 22:03:30,861 - Epoch: [172][ 1236/ 1236] Overall Loss 0.188862 Objective Loss 0.188862 Top1 89.816701 Top5 99.185336 LR 0.000250 Time 0.021182 +2023-10-05 22:03:30,990 - --- validate (epoch=172)----------- +2023-10-05 22:03:30,990 - 29943 samples (256 per mini-batch) +2023-10-05 22:03:31,439 - Epoch: [172][ 10/ 117] Loss 0.292991 Top1 85.039062 Top5 98.242188 +2023-10-05 22:03:31,591 - Epoch: [172][ 20/ 117] Loss 0.305646 Top1 85.351562 Top5 98.222656 +2023-10-05 22:03:31,742 - Epoch: [172][ 30/ 117] Loss 0.310721 Top1 85.520833 Top5 98.216146 +2023-10-05 22:03:31,893 - Epoch: [172][ 40/ 117] Loss 0.317678 Top1 85.322266 Top5 98.154297 +2023-10-05 22:03:32,044 - Epoch: [172][ 50/ 117] Loss 0.311864 Top1 85.406250 Top5 98.195312 +2023-10-05 22:03:32,195 - Epoch: [172][ 60/ 117] Loss 0.316028 Top1 85.436198 Top5 98.209635 +2023-10-05 22:03:32,345 - Epoch: [172][ 70/ 117] Loss 0.316247 Top1 85.368304 Top5 98.147321 +2023-10-05 22:03:32,493 - Epoch: [172][ 80/ 117] Loss 0.317969 Top1 85.419922 Top5 98.125000 +2023-10-05 22:03:32,639 - Epoch: [172][ 90/ 117] Loss 0.316323 Top1 85.416667 Top5 98.133681 +2023-10-05 22:03:32,788 - Epoch: [172][ 100/ 117] Loss 0.317374 Top1 85.324219 Top5 98.132812 +2023-10-05 22:03:32,944 - Epoch: [172][ 110/ 117] Loss 0.318291 Top1 85.252131 Top5 98.160511 +2023-10-05 22:03:33,028 - Epoch: [172][ 117/ 117] Loss 0.315917 Top1 85.268677 Top5 98.189894 +2023-10-05 22:03:33,176 - ==> Top1: 85.269 Top5: 98.190 Loss: 0.316 + +2023-10-05 22:03:33,177 - ==> Confusion: +[[ 952 3 3 0 6 4 0 0 9 47 3 0 0 0 5 2 1 1 0 1 13] + [ 0 1066 3 0 10 12 1 14 1 0 0 1 0 1 0 4 3 0 10 1 4] + [ 6 2 974 7 2 1 19 6 0 0 3 2 7 2 0 5 3 0 7 4 6] + [ 2 1 15 969 1 3 0 1 2 1 8 1 4 1 24 4 0 6 27 0 19] + [ 24 4 2 1 980 3 1 1 0 8 0 2 0 1 9 2 6 3 0 1 2] + [ 3 46 0 0 7 969 1 23 0 2 6 10 0 13 8 2 4 0 5 0 17] + [ 0 4 33 0 0 1 1117 8 0 0 2 3 1 0 1 4 0 0 1 9 7] + [ 4 26 16 0 2 22 3 1062 3 2 2 12 2 3 1 4 0 0 38 5 11] + [ 23 1 2 0 0 0 1 0 978 35 11 2 2 9 14 3 1 1 4 0 2] + [ 114 0 4 0 6 2 0 0 19 931 1 0 2 17 6 6 1 2 0 0 8] + [ 3 7 12 2 0 1 1 2 15 2 972 2 0 10 4 0 2 0 9 2 7] + [ 0 0 1 0 0 16 0 1 0 0 0 960 17 7 0 2 3 16 0 8 4] + [ 2 2 1 6 0 1 0 2 1 0 2 35 982 3 1 2 2 13 1 3 9] + [ 2 0 3 0 3 4 0 0 9 11 7 5 4 1051 2 4 0 0 0 3 11] + [ 17 2 3 5 4 1 0 0 26 1 2 1 2 3 1009 0 0 1 13 0 11] + [ 1 3 1 0 2 1 0 0 0 0 0 8 6 1 2 1074 15 9 0 9 2] + [ 1 12 1 0 8 2 0 1 1 0 0 2 1 0 3 9 1102 0 1 3 14] + [ 0 0 0 1 1 0 3 0 1 0 0 0 18 3 2 4 0 1000 1 0 4] + [ 1 8 5 13 1 0 0 17 2 0 1 0 0 1 10 0 0 0 1000 0 9] + [ 0 4 3 2 2 3 8 9 0 1 4 19 3 0 0 5 7 2 2 1065 13] + [ 141 195 148 57 95 91 30 90 109 71 193 103 281 265 148 48 133 73 149 166 5319]] + +2023-10-05 22:03:33,178 - ==> Best [Top1: 85.492 Top5: 98.170 Sparsity:0.00 Params: 148928 on epoch: 166] +2023-10-05 22:03:33,178 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:03:33,184 - + +2023-10-05 22:03:33,184 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:03:34,166 - Epoch: [173][ 10/ 1236] Overall Loss 0.192966 Objective Loss 0.192966 LR 0.000250 Time 0.098147 +2023-10-05 22:03:34,369 - Epoch: [173][ 20/ 1236] Overall Loss 0.191010 Objective Loss 0.191010 LR 0.000250 Time 0.059186 +2023-10-05 22:03:34,571 - Epoch: [173][ 30/ 1236] Overall Loss 0.187733 Objective Loss 0.187733 LR 0.000250 Time 0.046179 +2023-10-05 22:03:34,774 - Epoch: [173][ 40/ 1236] Overall Loss 0.189353 Objective Loss 0.189353 LR 0.000250 Time 0.039699 +2023-10-05 22:03:34,976 - Epoch: [173][ 50/ 1236] Overall Loss 0.186168 Objective Loss 0.186168 LR 0.000250 Time 0.035801 +2023-10-05 22:03:35,179 - Epoch: [173][ 60/ 1236] Overall Loss 0.185362 Objective Loss 0.185362 LR 0.000250 Time 0.033205 +2023-10-05 22:03:35,381 - Epoch: [173][ 70/ 1236] Overall Loss 0.185049 Objective Loss 0.185049 LR 0.000250 Time 0.031345 +2023-10-05 22:03:35,583 - Epoch: [173][ 80/ 1236] Overall Loss 0.186229 Objective Loss 0.186229 LR 0.000250 Time 0.029943 +2023-10-05 22:03:35,783 - Epoch: [173][ 90/ 1236] Overall Loss 0.187125 Objective Loss 0.187125 LR 0.000250 Time 0.028838 +2023-10-05 22:03:35,984 - Epoch: [173][ 100/ 1236] Overall Loss 0.188539 Objective Loss 0.188539 LR 0.000250 Time 0.027959 +2023-10-05 22:03:36,187 - Epoch: [173][ 110/ 1236] Overall Loss 0.187454 Objective Loss 0.187454 LR 0.000250 Time 0.027264 +2023-10-05 22:03:36,391 - Epoch: [173][ 120/ 1236] Overall Loss 0.186877 Objective Loss 0.186877 LR 0.000250 Time 0.026689 +2023-10-05 22:03:36,593 - Epoch: [173][ 130/ 1236] Overall Loss 0.188424 Objective Loss 0.188424 LR 0.000250 Time 0.026188 +2023-10-05 22:03:36,797 - Epoch: [173][ 140/ 1236] Overall Loss 0.189910 Objective Loss 0.189910 LR 0.000250 Time 0.025772 +2023-10-05 22:03:37,001 - Epoch: [173][ 150/ 1236] Overall Loss 0.188719 Objective Loss 0.188719 LR 0.000250 Time 0.025406 +2023-10-05 22:03:37,205 - Epoch: [173][ 160/ 1236] Overall Loss 0.188501 Objective Loss 0.188501 LR 0.000250 Time 0.025093 +2023-10-05 22:03:37,408 - Epoch: [173][ 170/ 1236] Overall Loss 0.188010 Objective Loss 0.188010 LR 0.000250 Time 0.024811 +2023-10-05 22:03:37,613 - Epoch: [173][ 180/ 1236] Overall Loss 0.187481 Objective Loss 0.187481 LR 0.000250 Time 0.024566 +2023-10-05 22:03:37,816 - Epoch: [173][ 190/ 1236] Overall Loss 0.188107 Objective Loss 0.188107 LR 0.000250 Time 0.024341 +2023-10-05 22:03:38,020 - Epoch: [173][ 200/ 1236] Overall Loss 0.188589 Objective Loss 0.188589 LR 0.000250 Time 0.024141 +2023-10-05 22:03:38,223 - Epoch: [173][ 210/ 1236] Overall Loss 0.188991 Objective Loss 0.188991 LR 0.000250 Time 0.023960 +2023-10-05 22:03:38,428 - Epoch: [173][ 220/ 1236] Overall Loss 0.188999 Objective Loss 0.188999 LR 0.000250 Time 0.023798 +2023-10-05 22:03:38,631 - Epoch: [173][ 230/ 1236] Overall Loss 0.189487 Objective Loss 0.189487 LR 0.000250 Time 0.023645 +2023-10-05 22:03:38,835 - Epoch: [173][ 240/ 1236] Overall Loss 0.189212 Objective Loss 0.189212 LR 0.000250 Time 0.023509 +2023-10-05 22:03:39,038 - Epoch: [173][ 250/ 1236] Overall Loss 0.189038 Objective Loss 0.189038 LR 0.000250 Time 0.023381 +2023-10-05 22:03:39,243 - Epoch: [173][ 260/ 1236] Overall Loss 0.188243 Objective Loss 0.188243 LR 0.000250 Time 0.023266 +2023-10-05 22:03:39,446 - Epoch: [173][ 270/ 1236] Overall Loss 0.187806 Objective Loss 0.187806 LR 0.000250 Time 0.023156 +2023-10-05 22:03:39,650 - Epoch: [173][ 280/ 1236] Overall Loss 0.187474 Objective Loss 0.187474 LR 0.000250 Time 0.023057 +2023-10-05 22:03:39,853 - Epoch: [173][ 290/ 1236] Overall Loss 0.187410 Objective Loss 0.187410 LR 0.000250 Time 0.022962 +2023-10-05 22:03:40,058 - Epoch: [173][ 300/ 1236] Overall Loss 0.187646 Objective Loss 0.187646 LR 0.000250 Time 0.022878 +2023-10-05 22:03:40,263 - Epoch: [173][ 310/ 1236] Overall Loss 0.187909 Objective Loss 0.187909 LR 0.000250 Time 0.022799 +2023-10-05 22:03:40,467 - Epoch: [173][ 320/ 1236] Overall Loss 0.188440 Objective Loss 0.188440 LR 0.000250 Time 0.022723 +2023-10-05 22:03:40,670 - Epoch: [173][ 330/ 1236] Overall Loss 0.188796 Objective Loss 0.188796 LR 0.000250 Time 0.022650 +2023-10-05 22:03:40,875 - Epoch: [173][ 340/ 1236] Overall Loss 0.188460 Objective Loss 0.188460 LR 0.000250 Time 0.022584 +2023-10-05 22:03:41,079 - Epoch: [173][ 350/ 1236] Overall Loss 0.188494 Objective Loss 0.188494 LR 0.000250 Time 0.022521 +2023-10-05 22:03:41,284 - Epoch: [173][ 360/ 1236] Overall Loss 0.187627 Objective Loss 0.187627 LR 0.000250 Time 0.022464 +2023-10-05 22:03:41,487 - Epoch: [173][ 370/ 1236] Overall Loss 0.187464 Objective Loss 0.187464 LR 0.000250 Time 0.022405 +2023-10-05 22:03:41,690 - Epoch: [173][ 380/ 1236] Overall Loss 0.187805 Objective Loss 0.187805 LR 0.000250 Time 0.022349 +2023-10-05 22:03:41,892 - Epoch: [173][ 390/ 1236] Overall Loss 0.187984 Objective Loss 0.187984 LR 0.000250 Time 0.022293 +2023-10-05 22:03:42,095 - Epoch: [173][ 400/ 1236] Overall Loss 0.187883 Objective Loss 0.187883 LR 0.000250 Time 0.022242 +2023-10-05 22:03:42,297 - Epoch: [173][ 410/ 1236] Overall Loss 0.187802 Objective Loss 0.187802 LR 0.000250 Time 0.022192 +2023-10-05 22:03:42,500 - Epoch: [173][ 420/ 1236] Overall Loss 0.187919 Objective Loss 0.187919 LR 0.000250 Time 0.022146 +2023-10-05 22:03:42,702 - Epoch: [173][ 430/ 1236] Overall Loss 0.188236 Objective Loss 0.188236 LR 0.000250 Time 0.022100 +2023-10-05 22:03:42,905 - Epoch: [173][ 440/ 1236] Overall Loss 0.188119 Objective Loss 0.188119 LR 0.000250 Time 0.022058 +2023-10-05 22:03:43,107 - Epoch: [173][ 450/ 1236] Overall Loss 0.188231 Objective Loss 0.188231 LR 0.000250 Time 0.022016 +2023-10-05 22:03:43,310 - Epoch: [173][ 460/ 1236] Overall Loss 0.187990 Objective Loss 0.187990 LR 0.000250 Time 0.021978 +2023-10-05 22:03:43,512 - Epoch: [173][ 470/ 1236] Overall Loss 0.188295 Objective Loss 0.188295 LR 0.000250 Time 0.021940 +2023-10-05 22:03:43,715 - Epoch: [173][ 480/ 1236] Overall Loss 0.187975 Objective Loss 0.187975 LR 0.000250 Time 0.021905 +2023-10-05 22:03:43,918 - Epoch: [173][ 490/ 1236] Overall Loss 0.187916 Objective Loss 0.187916 LR 0.000250 Time 0.021870 +2023-10-05 22:03:44,120 - Epoch: [173][ 500/ 1236] Overall Loss 0.187878 Objective Loss 0.187878 LR 0.000250 Time 0.021837 +2023-10-05 22:03:44,322 - Epoch: [173][ 510/ 1236] Overall Loss 0.188510 Objective Loss 0.188510 LR 0.000250 Time 0.021805 +2023-10-05 22:03:44,525 - Epoch: [173][ 520/ 1236] Overall Loss 0.188898 Objective Loss 0.188898 LR 0.000250 Time 0.021775 +2023-10-05 22:03:44,727 - Epoch: [173][ 530/ 1236] Overall Loss 0.189080 Objective Loss 0.189080 LR 0.000250 Time 0.021745 +2023-10-05 22:03:44,930 - Epoch: [173][ 540/ 1236] Overall Loss 0.188992 Objective Loss 0.188992 LR 0.000250 Time 0.021718 +2023-10-05 22:03:45,132 - Epoch: [173][ 550/ 1236] Overall Loss 0.189280 Objective Loss 0.189280 LR 0.000250 Time 0.021689 +2023-10-05 22:03:45,335 - Epoch: [173][ 560/ 1236] Overall Loss 0.189071 Objective Loss 0.189071 LR 0.000250 Time 0.021664 +2023-10-05 22:03:45,537 - Epoch: [173][ 570/ 1236] Overall Loss 0.188954 Objective Loss 0.188954 LR 0.000250 Time 0.021637 +2023-10-05 22:03:45,740 - Epoch: [173][ 580/ 1236] Overall Loss 0.188929 Objective Loss 0.188929 LR 0.000250 Time 0.021614 +2023-10-05 22:03:45,942 - Epoch: [173][ 590/ 1236] Overall Loss 0.188517 Objective Loss 0.188517 LR 0.000250 Time 0.021589 +2023-10-05 22:03:46,145 - Epoch: [173][ 600/ 1236] Overall Loss 0.188543 Objective Loss 0.188543 LR 0.000250 Time 0.021567 +2023-10-05 22:03:46,347 - Epoch: [173][ 610/ 1236] Overall Loss 0.188523 Objective Loss 0.188523 LR 0.000250 Time 0.021544 +2023-10-05 22:03:46,550 - Epoch: [173][ 620/ 1236] Overall Loss 0.188273 Objective Loss 0.188273 LR 0.000250 Time 0.021523 +2023-10-05 22:03:46,752 - Epoch: [173][ 630/ 1236] Overall Loss 0.188286 Objective Loss 0.188286 LR 0.000250 Time 0.021502 +2023-10-05 22:03:46,955 - Epoch: [173][ 640/ 1236] Overall Loss 0.188539 Objective Loss 0.188539 LR 0.000250 Time 0.021482 +2023-10-05 22:03:47,158 - Epoch: [173][ 650/ 1236] Overall Loss 0.188689 Objective Loss 0.188689 LR 0.000250 Time 0.021462 +2023-10-05 22:03:47,360 - Epoch: [173][ 660/ 1236] Overall Loss 0.188938 Objective Loss 0.188938 LR 0.000250 Time 0.021444 +2023-10-05 22:03:47,562 - Epoch: [173][ 670/ 1236] Overall Loss 0.188860 Objective Loss 0.188860 LR 0.000250 Time 0.021425 +2023-10-05 22:03:47,765 - Epoch: [173][ 680/ 1236] Overall Loss 0.188842 Objective Loss 0.188842 LR 0.000250 Time 0.021408 +2023-10-05 22:03:47,967 - Epoch: [173][ 690/ 1236] Overall Loss 0.188767 Objective Loss 0.188767 LR 0.000250 Time 0.021390 +2023-10-05 22:03:48,170 - Epoch: [173][ 700/ 1236] Overall Loss 0.188553 Objective Loss 0.188553 LR 0.000250 Time 0.021374 +2023-10-05 22:03:48,372 - Epoch: [173][ 710/ 1236] Overall Loss 0.188603 Objective Loss 0.188603 LR 0.000250 Time 0.021357 +2023-10-05 22:03:48,576 - Epoch: [173][ 720/ 1236] Overall Loss 0.188499 Objective Loss 0.188499 LR 0.000250 Time 0.021342 +2023-10-05 22:03:48,777 - Epoch: [173][ 730/ 1236] Overall Loss 0.188618 Objective Loss 0.188618 LR 0.000250 Time 0.021326 +2023-10-05 22:03:48,980 - Epoch: [173][ 740/ 1236] Overall Loss 0.188271 Objective Loss 0.188271 LR 0.000250 Time 0.021311 +2023-10-05 22:03:49,182 - Epoch: [173][ 750/ 1236] Overall Loss 0.188541 Objective Loss 0.188541 LR 0.000250 Time 0.021296 +2023-10-05 22:03:49,385 - Epoch: [173][ 760/ 1236] Overall Loss 0.188501 Objective Loss 0.188501 LR 0.000250 Time 0.021282 +2023-10-05 22:03:49,587 - Epoch: [173][ 770/ 1236] Overall Loss 0.188330 Objective Loss 0.188330 LR 0.000250 Time 0.021268 +2023-10-05 22:03:49,791 - Epoch: [173][ 780/ 1236] Overall Loss 0.188093 Objective Loss 0.188093 LR 0.000250 Time 0.021255 +2023-10-05 22:03:49,992 - Epoch: [173][ 790/ 1236] Overall Loss 0.188109 Objective Loss 0.188109 LR 0.000250 Time 0.021241 +2023-10-05 22:03:50,196 - Epoch: [173][ 800/ 1236] Overall Loss 0.188271 Objective Loss 0.188271 LR 0.000250 Time 0.021229 +2023-10-05 22:03:50,397 - Epoch: [173][ 810/ 1236] Overall Loss 0.188280 Objective Loss 0.188280 LR 0.000250 Time 0.021216 +2023-10-05 22:03:50,601 - Epoch: [173][ 820/ 1236] Overall Loss 0.188047 Objective Loss 0.188047 LR 0.000250 Time 0.021205 +2023-10-05 22:03:50,802 - Epoch: [173][ 830/ 1236] Overall Loss 0.187805 Objective Loss 0.187805 LR 0.000250 Time 0.021192 +2023-10-05 22:03:51,005 - Epoch: [173][ 840/ 1236] Overall Loss 0.187551 Objective Loss 0.187551 LR 0.000250 Time 0.021181 +2023-10-05 22:03:51,207 - Epoch: [173][ 850/ 1236] Overall Loss 0.187601 Objective Loss 0.187601 LR 0.000250 Time 0.021169 +2023-10-05 22:03:51,411 - Epoch: [173][ 860/ 1236] Overall Loss 0.187710 Objective Loss 0.187710 LR 0.000250 Time 0.021159 +2023-10-05 22:03:51,613 - Epoch: [173][ 870/ 1236] Overall Loss 0.187860 Objective Loss 0.187860 LR 0.000250 Time 0.021148 +2023-10-05 22:03:51,816 - Epoch: [173][ 880/ 1236] Overall Loss 0.187951 Objective Loss 0.187951 LR 0.000250 Time 0.021138 +2023-10-05 22:03:52,018 - Epoch: [173][ 890/ 1236] Overall Loss 0.188095 Objective Loss 0.188095 LR 0.000250 Time 0.021127 +2023-10-05 22:03:52,222 - Epoch: [173][ 900/ 1236] Overall Loss 0.188121 Objective Loss 0.188121 LR 0.000250 Time 0.021118 +2023-10-05 22:03:52,424 - Epoch: [173][ 910/ 1236] Overall Loss 0.188054 Objective Loss 0.188054 LR 0.000250 Time 0.021108 +2023-10-05 22:03:52,627 - Epoch: [173][ 920/ 1236] Overall Loss 0.188069 Objective Loss 0.188069 LR 0.000250 Time 0.021099 +2023-10-05 22:03:52,830 - Epoch: [173][ 930/ 1236] Overall Loss 0.187919 Objective Loss 0.187919 LR 0.000250 Time 0.021089 +2023-10-05 22:03:53,033 - Epoch: [173][ 940/ 1236] Overall Loss 0.187916 Objective Loss 0.187916 LR 0.000250 Time 0.021081 +2023-10-05 22:03:53,234 - Epoch: [173][ 950/ 1236] Overall Loss 0.187726 Objective Loss 0.187726 LR 0.000250 Time 0.021071 +2023-10-05 22:03:53,438 - Epoch: [173][ 960/ 1236] Overall Loss 0.187977 Objective Loss 0.187977 LR 0.000250 Time 0.021063 +2023-10-05 22:03:53,639 - Epoch: [173][ 970/ 1236] Overall Loss 0.187922 Objective Loss 0.187922 LR 0.000250 Time 0.021053 +2023-10-05 22:03:53,843 - Epoch: [173][ 980/ 1236] Overall Loss 0.187747 Objective Loss 0.187747 LR 0.000250 Time 0.021045 +2023-10-05 22:03:54,045 - Epoch: [173][ 990/ 1236] Overall Loss 0.187825 Objective Loss 0.187825 LR 0.000250 Time 0.021037 +2023-10-05 22:03:54,249 - Epoch: [173][ 1000/ 1236] Overall Loss 0.188000 Objective Loss 0.188000 LR 0.000250 Time 0.021030 +2023-10-05 22:03:54,450 - Epoch: [173][ 1010/ 1236] Overall Loss 0.188002 Objective Loss 0.188002 LR 0.000250 Time 0.021021 +2023-10-05 22:03:54,654 - Epoch: [173][ 1020/ 1236] Overall Loss 0.187812 Objective Loss 0.187812 LR 0.000250 Time 0.021014 +2023-10-05 22:03:54,856 - Epoch: [173][ 1030/ 1236] Overall Loss 0.187960 Objective Loss 0.187960 LR 0.000250 Time 0.021005 +2023-10-05 22:03:55,059 - Epoch: [173][ 1040/ 1236] Overall Loss 0.187884 Objective Loss 0.187884 LR 0.000250 Time 0.020998 +2023-10-05 22:03:55,261 - Epoch: [173][ 1050/ 1236] Overall Loss 0.187827 Objective Loss 0.187827 LR 0.000250 Time 0.020990 +2023-10-05 22:03:55,463 - Epoch: [173][ 1060/ 1236] Overall Loss 0.188041 Objective Loss 0.188041 LR 0.000250 Time 0.020983 +2023-10-05 22:03:55,664 - Epoch: [173][ 1070/ 1236] Overall Loss 0.188133 Objective Loss 0.188133 LR 0.000250 Time 0.020975 +2023-10-05 22:03:55,868 - Epoch: [173][ 1080/ 1236] Overall Loss 0.188135 Objective Loss 0.188135 LR 0.000250 Time 0.020968 +2023-10-05 22:03:56,070 - Epoch: [173][ 1090/ 1236] Overall Loss 0.188192 Objective Loss 0.188192 LR 0.000250 Time 0.020961 +2023-10-05 22:03:56,273 - Epoch: [173][ 1100/ 1236] Overall Loss 0.187903 Objective Loss 0.187903 LR 0.000250 Time 0.020955 +2023-10-05 22:03:56,474 - Epoch: [173][ 1110/ 1236] Overall Loss 0.187967 Objective Loss 0.187967 LR 0.000250 Time 0.020947 +2023-10-05 22:03:56,678 - Epoch: [173][ 1120/ 1236] Overall Loss 0.187938 Objective Loss 0.187938 LR 0.000250 Time 0.020941 +2023-10-05 22:03:56,879 - Epoch: [173][ 1130/ 1236] Overall Loss 0.187971 Objective Loss 0.187971 LR 0.000250 Time 0.020934 +2023-10-05 22:03:57,083 - Epoch: [173][ 1140/ 1236] Overall Loss 0.187966 Objective Loss 0.187966 LR 0.000250 Time 0.020929 +2023-10-05 22:03:57,285 - Epoch: [173][ 1150/ 1236] Overall Loss 0.187973 Objective Loss 0.187973 LR 0.000250 Time 0.020922 +2023-10-05 22:03:57,488 - Epoch: [173][ 1160/ 1236] Overall Loss 0.187880 Objective Loss 0.187880 LR 0.000250 Time 0.020917 +2023-10-05 22:03:57,690 - Epoch: [173][ 1170/ 1236] Overall Loss 0.187806 Objective Loss 0.187806 LR 0.000250 Time 0.020911 +2023-10-05 22:03:57,894 - Epoch: [173][ 1180/ 1236] Overall Loss 0.187922 Objective Loss 0.187922 LR 0.000250 Time 0.020905 +2023-10-05 22:03:58,096 - Epoch: [173][ 1190/ 1236] Overall Loss 0.187852 Objective Loss 0.187852 LR 0.000250 Time 0.020899 +2023-10-05 22:03:58,299 - Epoch: [173][ 1200/ 1236] Overall Loss 0.187943 Objective Loss 0.187943 LR 0.000250 Time 0.020894 +2023-10-05 22:03:58,502 - Epoch: [173][ 1210/ 1236] Overall Loss 0.187987 Objective Loss 0.187987 LR 0.000250 Time 0.020888 +2023-10-05 22:03:58,705 - Epoch: [173][ 1220/ 1236] Overall Loss 0.188039 Objective Loss 0.188039 LR 0.000250 Time 0.020883 +2023-10-05 22:03:58,960 - Epoch: [173][ 1230/ 1236] Overall Loss 0.188206 Objective Loss 0.188206 LR 0.000250 Time 0.020921 +2023-10-05 22:03:59,078 - Epoch: [173][ 1236/ 1236] Overall Loss 0.188133 Objective Loss 0.188133 Top1 90.835031 Top5 99.185336 LR 0.000250 Time 0.020914 +2023-10-05 22:03:59,209 - --- validate (epoch=173)----------- +2023-10-05 22:03:59,209 - 29943 samples (256 per mini-batch) +2023-10-05 22:03:59,656 - Epoch: [173][ 10/ 117] Loss 0.297875 Top1 85.546875 Top5 98.320312 +2023-10-05 22:03:59,802 - Epoch: [173][ 20/ 117] Loss 0.300783 Top1 85.937500 Top5 98.281250 +2023-10-05 22:03:59,949 - Epoch: [173][ 30/ 117] Loss 0.292369 Top1 86.119792 Top5 98.203125 +2023-10-05 22:04:00,094 - Epoch: [173][ 40/ 117] Loss 0.294535 Top1 86.132812 Top5 98.222656 +2023-10-05 22:04:00,240 - Epoch: [173][ 50/ 117] Loss 0.296141 Top1 86.000000 Top5 98.242188 +2023-10-05 22:04:00,385 - Epoch: [173][ 60/ 117] Loss 0.295771 Top1 86.126302 Top5 98.300781 +2023-10-05 22:04:00,532 - Epoch: [173][ 70/ 117] Loss 0.299337 Top1 85.931920 Top5 98.225446 +2023-10-05 22:04:00,676 - Epoch: [173][ 80/ 117] Loss 0.297842 Top1 86.010742 Top5 98.251953 +2023-10-05 22:04:00,823 - Epoch: [173][ 90/ 117] Loss 0.299474 Top1 86.032986 Top5 98.237847 +2023-10-05 22:04:00,969 - Epoch: [173][ 100/ 117] Loss 0.303921 Top1 85.828125 Top5 98.253906 +2023-10-05 22:04:01,121 - Epoch: [173][ 110/ 117] Loss 0.307369 Top1 85.770597 Top5 98.252841 +2023-10-05 22:04:01,206 - Epoch: [173][ 117/ 117] Loss 0.307245 Top1 85.846442 Top5 98.233310 +2023-10-05 22:04:01,326 - ==> Top1: 85.846 Top5: 98.233 Loss: 0.307 + +2023-10-05 22:04:01,327 - ==> Confusion: +[[ 933 4 1 0 1 4 0 0 9 76 1 0 0 1 5 1 3 1 0 0 10] + [ 1 1061 1 0 7 15 2 17 4 0 0 0 0 0 0 4 2 0 9 3 5] + [ 8 1 972 8 0 0 25 5 0 2 4 0 9 1 0 4 1 2 4 2 8] + [ 3 1 12 971 1 1 1 0 0 1 11 1 2 2 29 3 1 5 21 1 22] + [ 28 1 1 0 962 3 1 1 1 15 0 2 0 3 10 3 11 1 0 2 5] + [ 3 40 1 1 3 993 1 17 1 2 3 6 1 12 4 2 6 0 3 3 14] + [ 0 4 24 0 0 0 1131 4 0 1 0 3 1 0 1 3 0 1 4 5 9] + [ 4 21 13 0 2 26 5 1065 2 4 3 7 2 3 0 2 1 0 44 4 10] + [ 18 2 1 0 1 2 0 1 977 44 8 2 2 11 12 3 1 0 3 0 1] + [ 85 0 1 0 2 3 0 0 22 964 0 1 2 15 6 6 0 2 0 0 10] + [ 2 8 10 4 1 2 4 2 14 3 970 4 1 7 3 0 1 0 4 2 11] + [ 2 0 2 0 0 18 0 2 0 1 0 954 20 7 0 3 3 15 0 4 4] + [ 1 3 2 3 0 2 0 2 0 0 0 38 973 2 1 5 4 17 3 4 8] + [ 2 0 1 1 1 8 0 0 12 14 8 3 3 1051 4 1 1 1 0 0 8] + [ 11 4 2 10 2 0 0 0 30 3 2 0 1 0 1008 0 1 2 11 0 14] + [ 1 3 1 0 1 0 2 0 0 1 0 7 6 1 1 1069 17 14 0 7 3] + [ 0 14 1 0 6 1 0 0 2 0 0 4 1 0 3 11 1101 0 0 2 15] + [ 0 0 0 1 1 0 3 0 2 0 0 0 13 0 1 4 0 1010 1 0 2] + [ 1 6 3 23 1 0 0 22 2 0 1 1 2 1 8 0 0 0 988 0 9] + [ 0 2 4 1 1 5 9 7 0 0 0 13 2 1 0 7 9 3 3 1076 9] + [ 121 173 152 58 66 102 32 79 114 77 160 100 301 267 150 41 105 68 119 144 5476]] + +2023-10-05 22:04:01,328 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:04:01,328 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:04:01,342 - + +2023-10-05 22:04:01,342 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:04:02,327 - Epoch: [174][ 10/ 1236] Overall Loss 0.181038 Objective Loss 0.181038 LR 0.000250 Time 0.098422 +2023-10-05 22:04:02,530 - Epoch: [174][ 20/ 1236] Overall Loss 0.166897 Objective Loss 0.166897 LR 0.000250 Time 0.059337 +2023-10-05 22:04:02,733 - Epoch: [174][ 30/ 1236] Overall Loss 0.165329 Objective Loss 0.165329 LR 0.000250 Time 0.046346 +2023-10-05 22:04:02,936 - Epoch: [174][ 40/ 1236] Overall Loss 0.164678 Objective Loss 0.164678 LR 0.000250 Time 0.039819 +2023-10-05 22:04:03,140 - Epoch: [174][ 50/ 1236] Overall Loss 0.167779 Objective Loss 0.167779 LR 0.000250 Time 0.035933 +2023-10-05 22:04:03,343 - Epoch: [174][ 60/ 1236] Overall Loss 0.171237 Objective Loss 0.171237 LR 0.000250 Time 0.033319 +2023-10-05 22:04:03,548 - Epoch: [174][ 70/ 1236] Overall Loss 0.177239 Objective Loss 0.177239 LR 0.000250 Time 0.031475 +2023-10-05 22:04:03,750 - Epoch: [174][ 80/ 1236] Overall Loss 0.181492 Objective Loss 0.181492 LR 0.000250 Time 0.030072 +2023-10-05 22:04:03,955 - Epoch: [174][ 90/ 1236] Overall Loss 0.184036 Objective Loss 0.184036 LR 0.000250 Time 0.028994 +2023-10-05 22:04:04,156 - Epoch: [174][ 100/ 1236] Overall Loss 0.184170 Objective Loss 0.184170 LR 0.000250 Time 0.028109 +2023-10-05 22:04:04,360 - Epoch: [174][ 110/ 1236] Overall Loss 0.184908 Objective Loss 0.184908 LR 0.000250 Time 0.027399 +2023-10-05 22:04:04,561 - Epoch: [174][ 120/ 1236] Overall Loss 0.186272 Objective Loss 0.186272 LR 0.000250 Time 0.026794 +2023-10-05 22:04:04,765 - Epoch: [174][ 130/ 1236] Overall Loss 0.185667 Objective Loss 0.185667 LR 0.000250 Time 0.026294 +2023-10-05 22:04:04,966 - Epoch: [174][ 140/ 1236] Overall Loss 0.186712 Objective Loss 0.186712 LR 0.000250 Time 0.025855 +2023-10-05 22:04:05,170 - Epoch: [174][ 150/ 1236] Overall Loss 0.186039 Objective Loss 0.186039 LR 0.000250 Time 0.025484 +2023-10-05 22:04:05,371 - Epoch: [174][ 160/ 1236] Overall Loss 0.185196 Objective Loss 0.185196 LR 0.000250 Time 0.025149 +2023-10-05 22:04:05,575 - Epoch: [174][ 170/ 1236] Overall Loss 0.184879 Objective Loss 0.184879 LR 0.000250 Time 0.024864 +2023-10-05 22:04:05,777 - Epoch: [174][ 180/ 1236] Overall Loss 0.183500 Objective Loss 0.183500 LR 0.000250 Time 0.024603 +2023-10-05 22:04:05,981 - Epoch: [174][ 190/ 1236] Overall Loss 0.183530 Objective Loss 0.183530 LR 0.000250 Time 0.024378 +2023-10-05 22:04:06,182 - Epoch: [174][ 200/ 1236] Overall Loss 0.184186 Objective Loss 0.184186 LR 0.000250 Time 0.024167 +2023-10-05 22:04:06,386 - Epoch: [174][ 210/ 1236] Overall Loss 0.183864 Objective Loss 0.183864 LR 0.000250 Time 0.023983 +2023-10-05 22:04:06,588 - Epoch: [174][ 220/ 1236] Overall Loss 0.184029 Objective Loss 0.184029 LR 0.000250 Time 0.023811 +2023-10-05 22:04:06,791 - Epoch: [174][ 230/ 1236] Overall Loss 0.183434 Objective Loss 0.183434 LR 0.000250 Time 0.023656 +2023-10-05 22:04:06,992 - Epoch: [174][ 240/ 1236] Overall Loss 0.183732 Objective Loss 0.183732 LR 0.000250 Time 0.023509 +2023-10-05 22:04:07,195 - Epoch: [174][ 250/ 1236] Overall Loss 0.184090 Objective Loss 0.184090 LR 0.000250 Time 0.023378 +2023-10-05 22:04:07,397 - Epoch: [174][ 260/ 1236] Overall Loss 0.185194 Objective Loss 0.185194 LR 0.000250 Time 0.023255 +2023-10-05 22:04:07,601 - Epoch: [174][ 270/ 1236] Overall Loss 0.183824 Objective Loss 0.183824 LR 0.000250 Time 0.023146 +2023-10-05 22:04:07,803 - Epoch: [174][ 280/ 1236] Overall Loss 0.183375 Objective Loss 0.183375 LR 0.000250 Time 0.023041 +2023-10-05 22:04:08,007 - Epoch: [174][ 290/ 1236] Overall Loss 0.183502 Objective Loss 0.183502 LR 0.000250 Time 0.022946 +2023-10-05 22:04:08,208 - Epoch: [174][ 300/ 1236] Overall Loss 0.183678 Objective Loss 0.183678 LR 0.000250 Time 0.022850 +2023-10-05 22:04:08,411 - Epoch: [174][ 310/ 1236] Overall Loss 0.183689 Objective Loss 0.183689 LR 0.000250 Time 0.022768 +2023-10-05 22:04:08,613 - Epoch: [174][ 320/ 1236] Overall Loss 0.183347 Objective Loss 0.183347 LR 0.000250 Time 0.022686 +2023-10-05 22:04:08,816 - Epoch: [174][ 330/ 1236] Overall Loss 0.183080 Objective Loss 0.183080 LR 0.000250 Time 0.022615 +2023-10-05 22:04:09,018 - Epoch: [174][ 340/ 1236] Overall Loss 0.183153 Objective Loss 0.183153 LR 0.000250 Time 0.022543 +2023-10-05 22:04:09,222 - Epoch: [174][ 350/ 1236] Overall Loss 0.183369 Objective Loss 0.183369 LR 0.000250 Time 0.022479 +2023-10-05 22:04:09,423 - Epoch: [174][ 360/ 1236] Overall Loss 0.182932 Objective Loss 0.182932 LR 0.000250 Time 0.022414 +2023-10-05 22:04:09,627 - Epoch: [174][ 370/ 1236] Overall Loss 0.183256 Objective Loss 0.183256 LR 0.000250 Time 0.022357 +2023-10-05 22:04:09,829 - Epoch: [174][ 380/ 1236] Overall Loss 0.183674 Objective Loss 0.183674 LR 0.000250 Time 0.022300 +2023-10-05 22:04:10,034 - Epoch: [174][ 390/ 1236] Overall Loss 0.184104 Objective Loss 0.184104 LR 0.000250 Time 0.022252 +2023-10-05 22:04:10,237 - Epoch: [174][ 400/ 1236] Overall Loss 0.183905 Objective Loss 0.183905 LR 0.000250 Time 0.022203 +2023-10-05 22:04:10,442 - Epoch: [174][ 410/ 1236] Overall Loss 0.183782 Objective Loss 0.183782 LR 0.000250 Time 0.022162 +2023-10-05 22:04:10,645 - Epoch: [174][ 420/ 1236] Overall Loss 0.183910 Objective Loss 0.183910 LR 0.000250 Time 0.022116 +2023-10-05 22:04:10,850 - Epoch: [174][ 430/ 1236] Overall Loss 0.183749 Objective Loss 0.183749 LR 0.000250 Time 0.022077 +2023-10-05 22:04:11,053 - Epoch: [174][ 440/ 1236] Overall Loss 0.184328 Objective Loss 0.184328 LR 0.000250 Time 0.022036 +2023-10-05 22:04:11,257 - Epoch: [174][ 450/ 1236] Overall Loss 0.184947 Objective Loss 0.184947 LR 0.000250 Time 0.022000 +2023-10-05 22:04:11,460 - Epoch: [174][ 460/ 1236] Overall Loss 0.184894 Objective Loss 0.184894 LR 0.000250 Time 0.021962 +2023-10-05 22:04:11,665 - Epoch: [174][ 470/ 1236] Overall Loss 0.184598 Objective Loss 0.184598 LR 0.000250 Time 0.021929 +2023-10-05 22:04:11,868 - Epoch: [174][ 480/ 1236] Overall Loss 0.184472 Objective Loss 0.184472 LR 0.000250 Time 0.021894 +2023-10-05 22:04:12,073 - Epoch: [174][ 490/ 1236] Overall Loss 0.184394 Objective Loss 0.184394 LR 0.000250 Time 0.021865 +2023-10-05 22:04:12,276 - Epoch: [174][ 500/ 1236] Overall Loss 0.184963 Objective Loss 0.184963 LR 0.000250 Time 0.021834 +2023-10-05 22:04:12,481 - Epoch: [174][ 510/ 1236] Overall Loss 0.185202 Objective Loss 0.185202 LR 0.000250 Time 0.021806 +2023-10-05 22:04:12,684 - Epoch: [174][ 520/ 1236] Overall Loss 0.185151 Objective Loss 0.185151 LR 0.000250 Time 0.021777 +2023-10-05 22:04:12,889 - Epoch: [174][ 530/ 1236] Overall Loss 0.185199 Objective Loss 0.185199 LR 0.000250 Time 0.021752 +2023-10-05 22:04:13,092 - Epoch: [174][ 540/ 1236] Overall Loss 0.185220 Objective Loss 0.185220 LR 0.000250 Time 0.021724 +2023-10-05 22:04:13,296 - Epoch: [174][ 550/ 1236] Overall Loss 0.185122 Objective Loss 0.185122 LR 0.000250 Time 0.021701 +2023-10-05 22:04:13,499 - Epoch: [174][ 560/ 1236] Overall Loss 0.185399 Objective Loss 0.185399 LR 0.000250 Time 0.021675 +2023-10-05 22:04:13,704 - Epoch: [174][ 570/ 1236] Overall Loss 0.185302 Objective Loss 0.185302 LR 0.000250 Time 0.021653 +2023-10-05 22:04:13,907 - Epoch: [174][ 580/ 1236] Overall Loss 0.185432 Objective Loss 0.185432 LR 0.000250 Time 0.021629 +2023-10-05 22:04:14,112 - Epoch: [174][ 590/ 1236] Overall Loss 0.185601 Objective Loss 0.185601 LR 0.000250 Time 0.021609 +2023-10-05 22:04:14,315 - Epoch: [174][ 600/ 1236] Overall Loss 0.185526 Objective Loss 0.185526 LR 0.000250 Time 0.021588 +2023-10-05 22:04:14,520 - Epoch: [174][ 610/ 1236] Overall Loss 0.185807 Objective Loss 0.185807 LR 0.000250 Time 0.021568 +2023-10-05 22:04:14,723 - Epoch: [174][ 620/ 1236] Overall Loss 0.186158 Objective Loss 0.186158 LR 0.000250 Time 0.021548 +2023-10-05 22:04:14,927 - Epoch: [174][ 630/ 1236] Overall Loss 0.185668 Objective Loss 0.185668 LR 0.000250 Time 0.021530 +2023-10-05 22:04:15,130 - Epoch: [174][ 640/ 1236] Overall Loss 0.185905 Objective Loss 0.185905 LR 0.000250 Time 0.021510 +2023-10-05 22:04:15,335 - Epoch: [174][ 650/ 1236] Overall Loss 0.186154 Objective Loss 0.186154 LR 0.000250 Time 0.021493 +2023-10-05 22:04:15,538 - Epoch: [174][ 660/ 1236] Overall Loss 0.186007 Objective Loss 0.186007 LR 0.000250 Time 0.021475 +2023-10-05 22:04:15,743 - Epoch: [174][ 670/ 1236] Overall Loss 0.186185 Objective Loss 0.186185 LR 0.000250 Time 0.021459 +2023-10-05 22:04:15,946 - Epoch: [174][ 680/ 1236] Overall Loss 0.186108 Objective Loss 0.186108 LR 0.000250 Time 0.021442 +2023-10-05 22:04:16,151 - Epoch: [174][ 690/ 1236] Overall Loss 0.186073 Objective Loss 0.186073 LR 0.000250 Time 0.021427 +2023-10-05 22:04:16,354 - Epoch: [174][ 700/ 1236] Overall Loss 0.186237 Objective Loss 0.186237 LR 0.000250 Time 0.021411 +2023-10-05 22:04:16,558 - Epoch: [174][ 710/ 1236] Overall Loss 0.186445 Objective Loss 0.186445 LR 0.000250 Time 0.021397 +2023-10-05 22:04:16,762 - Epoch: [174][ 720/ 1236] Overall Loss 0.186115 Objective Loss 0.186115 LR 0.000250 Time 0.021381 +2023-10-05 22:04:16,966 - Epoch: [174][ 730/ 1236] Overall Loss 0.186112 Objective Loss 0.186112 LR 0.000250 Time 0.021368 +2023-10-05 22:04:17,169 - Epoch: [174][ 740/ 1236] Overall Loss 0.186219 Objective Loss 0.186219 LR 0.000250 Time 0.021353 +2023-10-05 22:04:17,373 - Epoch: [174][ 750/ 1236] Overall Loss 0.186282 Objective Loss 0.186282 LR 0.000250 Time 0.021340 +2023-10-05 22:04:17,576 - Epoch: [174][ 760/ 1236] Overall Loss 0.186610 Objective Loss 0.186610 LR 0.000250 Time 0.021326 +2023-10-05 22:04:17,781 - Epoch: [174][ 770/ 1236] Overall Loss 0.186536 Objective Loss 0.186536 LR 0.000250 Time 0.021314 +2023-10-05 22:04:17,984 - Epoch: [174][ 780/ 1236] Overall Loss 0.186591 Objective Loss 0.186591 LR 0.000250 Time 0.021301 +2023-10-05 22:04:18,188 - Epoch: [174][ 790/ 1236] Overall Loss 0.186690 Objective Loss 0.186690 LR 0.000250 Time 0.021289 +2023-10-05 22:04:18,391 - Epoch: [174][ 800/ 1236] Overall Loss 0.187051 Objective Loss 0.187051 LR 0.000250 Time 0.021276 +2023-10-05 22:04:18,595 - Epoch: [174][ 810/ 1236] Overall Loss 0.186785 Objective Loss 0.186785 LR 0.000250 Time 0.021265 +2023-10-05 22:04:18,798 - Epoch: [174][ 820/ 1236] Overall Loss 0.186728 Objective Loss 0.186728 LR 0.000250 Time 0.021253 +2023-10-05 22:04:19,002 - Epoch: [174][ 830/ 1236] Overall Loss 0.186674 Objective Loss 0.186674 LR 0.000250 Time 0.021243 +2023-10-05 22:04:19,206 - Epoch: [174][ 840/ 1236] Overall Loss 0.186568 Objective Loss 0.186568 LR 0.000250 Time 0.021231 +2023-10-05 22:04:19,410 - Epoch: [174][ 850/ 1236] Overall Loss 0.186731 Objective Loss 0.186731 LR 0.000250 Time 0.021221 +2023-10-05 22:04:19,613 - Epoch: [174][ 860/ 1236] Overall Loss 0.186748 Objective Loss 0.186748 LR 0.000250 Time 0.021210 +2023-10-05 22:04:19,817 - Epoch: [174][ 870/ 1236] Overall Loss 0.186768 Objective Loss 0.186768 LR 0.000250 Time 0.021201 +2023-10-05 22:04:20,021 - Epoch: [174][ 880/ 1236] Overall Loss 0.186654 Objective Loss 0.186654 LR 0.000250 Time 0.021191 +2023-10-05 22:04:20,225 - Epoch: [174][ 890/ 1236] Overall Loss 0.186512 Objective Loss 0.186512 LR 0.000250 Time 0.021182 +2023-10-05 22:04:20,428 - Epoch: [174][ 900/ 1236] Overall Loss 0.186411 Objective Loss 0.186411 LR 0.000250 Time 0.021172 +2023-10-05 22:04:20,632 - Epoch: [174][ 910/ 1236] Overall Loss 0.186377 Objective Loss 0.186377 LR 0.000250 Time 0.021163 +2023-10-05 22:04:20,836 - Epoch: [174][ 920/ 1236] Overall Loss 0.186400 Objective Loss 0.186400 LR 0.000250 Time 0.021154 +2023-10-05 22:04:21,040 - Epoch: [174][ 930/ 1236] Overall Loss 0.186297 Objective Loss 0.186297 LR 0.000250 Time 0.021146 +2023-10-05 22:04:21,243 - Epoch: [174][ 940/ 1236] Overall Loss 0.186349 Objective Loss 0.186349 LR 0.000250 Time 0.021137 +2023-10-05 22:04:21,448 - Epoch: [174][ 950/ 1236] Overall Loss 0.186351 Objective Loss 0.186351 LR 0.000250 Time 0.021129 +2023-10-05 22:04:21,651 - Epoch: [174][ 960/ 1236] Overall Loss 0.186469 Objective Loss 0.186469 LR 0.000250 Time 0.021120 +2023-10-05 22:04:21,855 - Epoch: [174][ 970/ 1236] Overall Loss 0.186699 Objective Loss 0.186699 LR 0.000250 Time 0.021113 +2023-10-05 22:04:22,058 - Epoch: [174][ 980/ 1236] Overall Loss 0.186769 Objective Loss 0.186769 LR 0.000250 Time 0.021104 +2023-10-05 22:04:22,263 - Epoch: [174][ 990/ 1236] Overall Loss 0.186938 Objective Loss 0.186938 LR 0.000250 Time 0.021097 +2023-10-05 22:04:22,466 - Epoch: [174][ 1000/ 1236] Overall Loss 0.186914 Objective Loss 0.186914 LR 0.000250 Time 0.021089 +2023-10-05 22:04:22,670 - Epoch: [174][ 1010/ 1236] Overall Loss 0.186910 Objective Loss 0.186910 LR 0.000250 Time 0.021082 +2023-10-05 22:04:22,874 - Epoch: [174][ 1020/ 1236] Overall Loss 0.186787 Objective Loss 0.186787 LR 0.000250 Time 0.021075 +2023-10-05 22:04:23,078 - Epoch: [174][ 1030/ 1236] Overall Loss 0.186939 Objective Loss 0.186939 LR 0.000250 Time 0.021068 +2023-10-05 22:04:23,281 - Epoch: [174][ 1040/ 1236] Overall Loss 0.187232 Objective Loss 0.187232 LR 0.000250 Time 0.021060 +2023-10-05 22:04:23,486 - Epoch: [174][ 1050/ 1236] Overall Loss 0.187369 Objective Loss 0.187369 LR 0.000250 Time 0.021054 +2023-10-05 22:04:23,689 - Epoch: [174][ 1060/ 1236] Overall Loss 0.187401 Objective Loss 0.187401 LR 0.000250 Time 0.021047 +2023-10-05 22:04:23,893 - Epoch: [174][ 1070/ 1236] Overall Loss 0.187667 Objective Loss 0.187667 LR 0.000250 Time 0.021041 +2023-10-05 22:04:24,096 - Epoch: [174][ 1080/ 1236] Overall Loss 0.187591 Objective Loss 0.187591 LR 0.000250 Time 0.021034 +2023-10-05 22:04:24,301 - Epoch: [174][ 1090/ 1236] Overall Loss 0.187753 Objective Loss 0.187753 LR 0.000250 Time 0.021028 +2023-10-05 22:04:24,504 - Epoch: [174][ 1100/ 1236] Overall Loss 0.187871 Objective Loss 0.187871 LR 0.000250 Time 0.021022 +2023-10-05 22:04:24,708 - Epoch: [174][ 1110/ 1236] Overall Loss 0.187841 Objective Loss 0.187841 LR 0.000250 Time 0.021016 +2023-10-05 22:04:24,912 - Epoch: [174][ 1120/ 1236] Overall Loss 0.188038 Objective Loss 0.188038 LR 0.000250 Time 0.021009 +2023-10-05 22:04:25,116 - Epoch: [174][ 1130/ 1236] Overall Loss 0.187986 Objective Loss 0.187986 LR 0.000250 Time 0.021004 +2023-10-05 22:04:25,319 - Epoch: [174][ 1140/ 1236] Overall Loss 0.187892 Objective Loss 0.187892 LR 0.000250 Time 0.020998 +2023-10-05 22:04:25,525 - Epoch: [174][ 1150/ 1236] Overall Loss 0.187744 Objective Loss 0.187744 LR 0.000250 Time 0.020993 +2023-10-05 22:04:25,728 - Epoch: [174][ 1160/ 1236] Overall Loss 0.187566 Objective Loss 0.187566 LR 0.000250 Time 0.020988 +2023-10-05 22:04:25,934 - Epoch: [174][ 1170/ 1236] Overall Loss 0.187535 Objective Loss 0.187535 LR 0.000250 Time 0.020983 +2023-10-05 22:04:26,138 - Epoch: [174][ 1180/ 1236] Overall Loss 0.187573 Objective Loss 0.187573 LR 0.000250 Time 0.020978 +2023-10-05 22:04:26,343 - Epoch: [174][ 1190/ 1236] Overall Loss 0.187557 Objective Loss 0.187557 LR 0.000250 Time 0.020974 +2023-10-05 22:04:26,547 - Epoch: [174][ 1200/ 1236] Overall Loss 0.187429 Objective Loss 0.187429 LR 0.000250 Time 0.020969 +2023-10-05 22:04:26,752 - Epoch: [174][ 1210/ 1236] Overall Loss 0.187522 Objective Loss 0.187522 LR 0.000250 Time 0.020965 +2023-10-05 22:04:26,956 - Epoch: [174][ 1220/ 1236] Overall Loss 0.187650 Objective Loss 0.187650 LR 0.000250 Time 0.020960 +2023-10-05 22:04:27,212 - Epoch: [174][ 1230/ 1236] Overall Loss 0.187492 Objective Loss 0.187492 LR 0.000250 Time 0.020997 +2023-10-05 22:04:27,329 - Epoch: [174][ 1236/ 1236] Overall Loss 0.187512 Objective Loss 0.187512 Top1 89.613035 Top5 98.370672 LR 0.000250 Time 0.020990 +2023-10-05 22:04:27,445 - --- validate (epoch=174)----------- +2023-10-05 22:04:27,445 - 29943 samples (256 per mini-batch) +2023-10-05 22:04:27,896 - Epoch: [174][ 10/ 117] Loss 0.310235 Top1 85.546875 Top5 98.750000 +2023-10-05 22:04:28,044 - Epoch: [174][ 20/ 117] Loss 0.297460 Top1 85.527344 Top5 98.281250 +2023-10-05 22:04:28,194 - Epoch: [174][ 30/ 117] Loss 0.288244 Top1 86.028646 Top5 98.281250 +2023-10-05 22:04:28,342 - Epoch: [174][ 40/ 117] Loss 0.285139 Top1 86.054688 Top5 98.281250 +2023-10-05 22:04:28,491 - Epoch: [174][ 50/ 117] Loss 0.287663 Top1 86.023438 Top5 98.250000 +2023-10-05 22:04:28,641 - Epoch: [174][ 60/ 117] Loss 0.286298 Top1 85.983073 Top5 98.274740 +2023-10-05 22:04:28,788 - Epoch: [174][ 70/ 117] Loss 0.291759 Top1 85.820312 Top5 98.286830 +2023-10-05 22:04:28,937 - Epoch: [174][ 80/ 117] Loss 0.296893 Top1 85.639648 Top5 98.251953 +2023-10-05 22:04:29,084 - Epoch: [174][ 90/ 117] Loss 0.300344 Top1 85.607639 Top5 98.242188 +2023-10-05 22:04:29,233 - Epoch: [174][ 100/ 117] Loss 0.296991 Top1 85.714844 Top5 98.261719 +2023-10-05 22:04:29,388 - Epoch: [174][ 110/ 117] Loss 0.300955 Top1 85.632102 Top5 98.245739 +2023-10-05 22:04:29,473 - Epoch: [174][ 117/ 117] Loss 0.302393 Top1 85.555890 Top5 98.229970 +2023-10-05 22:04:29,614 - ==> Top1: 85.556 Top5: 98.230 Loss: 0.302 + +2023-10-05 22:04:29,615 - ==> Confusion: +[[ 937 3 3 1 3 3 0 0 4 67 1 1 0 2 6 2 1 2 0 0 14] + [ 1 1054 2 0 8 18 1 22 1 0 1 2 0 1 1 3 1 0 6 1 8] + [ 6 2 970 11 2 0 17 11 0 1 4 4 6 1 0 3 0 3 4 4 7] + [ 1 1 9 984 1 3 0 2 2 1 8 1 8 1 21 3 0 5 21 0 17] + [ 23 8 0 0 976 3 0 0 0 12 1 1 0 1 4 4 10 2 0 1 4] + [ 3 32 0 1 3 1007 2 15 0 1 3 9 0 11 5 2 3 0 3 3 13] + [ 0 5 24 0 0 0 1126 8 0 0 2 2 1 0 1 8 0 1 1 9 3] + [ 5 16 11 0 2 29 3 1085 2 4 3 10 2 1 0 1 0 0 30 8 6] + [ 20 4 1 0 1 3 0 1 985 35 6 3 2 7 13 2 1 0 3 0 2] + [ 107 0 5 0 4 3 0 0 19 945 1 1 0 17 3 5 0 1 0 2 6] + [ 2 6 7 8 0 1 7 5 16 3 960 2 1 13 3 1 1 0 3 4 10] + [ 1 0 0 0 0 9 0 1 0 1 0 973 16 7 0 3 1 15 0 5 3] + [ 1 2 2 2 0 1 0 1 0 0 1 34 993 2 0 4 3 11 3 3 5] + [ 2 0 1 0 1 7 0 0 8 16 8 4 4 1053 3 1 1 0 0 0 10] + [ 15 3 2 11 4 1 0 0 27 1 2 0 2 2 1003 0 1 3 11 0 13] + [ 1 3 1 0 2 0 0 0 0 0 0 8 6 2 1 1075 12 11 0 8 4] + [ 1 13 1 1 4 4 0 0 1 0 0 6 0 2 2 13 1096 0 0 5 12] + [ 0 0 0 0 1 0 2 0 0 1 0 3 20 1 1 6 0 998 2 0 3] + [ 1 7 6 20 1 0 0 22 1 0 1 0 1 0 6 0 0 0 993 0 9] + [ 0 3 3 1 2 6 5 6 0 0 1 17 2 0 0 8 8 3 4 1071 12] + [ 158 176 147 55 76 139 26 94 102 68 157 130 322 260 114 47 115 72 145 168 5334]] + +2023-10-05 22:04:29,616 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:04:29,617 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:04:29,622 - + +2023-10-05 22:04:29,623 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:04:30,596 - Epoch: [175][ 10/ 1236] Overall Loss 0.167335 Objective Loss 0.167335 LR 0.000250 Time 0.097259 +2023-10-05 22:04:30,797 - Epoch: [175][ 20/ 1236] Overall Loss 0.180884 Objective Loss 0.180884 LR 0.000250 Time 0.058670 +2023-10-05 22:04:30,996 - Epoch: [175][ 30/ 1236] Overall Loss 0.188781 Objective Loss 0.188781 LR 0.000250 Time 0.045737 +2023-10-05 22:04:31,197 - Epoch: [175][ 40/ 1236] Overall Loss 0.187894 Objective Loss 0.187894 LR 0.000250 Time 0.039321 +2023-10-05 22:04:31,396 - Epoch: [175][ 50/ 1236] Overall Loss 0.187404 Objective Loss 0.187404 LR 0.000250 Time 0.035439 +2023-10-05 22:04:31,597 - Epoch: [175][ 60/ 1236] Overall Loss 0.185774 Objective Loss 0.185774 LR 0.000250 Time 0.032868 +2023-10-05 22:04:31,796 - Epoch: [175][ 70/ 1236] Overall Loss 0.186780 Objective Loss 0.186780 LR 0.000250 Time 0.031012 +2023-10-05 22:04:31,997 - Epoch: [175][ 80/ 1236] Overall Loss 0.187345 Objective Loss 0.187345 LR 0.000250 Time 0.029645 +2023-10-05 22:04:32,198 - Epoch: [175][ 90/ 1236] Overall Loss 0.185865 Objective Loss 0.185865 LR 0.000250 Time 0.028583 +2023-10-05 22:04:32,401 - Epoch: [175][ 100/ 1236] Overall Loss 0.187462 Objective Loss 0.187462 LR 0.000250 Time 0.027754 +2023-10-05 22:04:32,602 - Epoch: [175][ 110/ 1236] Overall Loss 0.187020 Objective Loss 0.187020 LR 0.000250 Time 0.027058 +2023-10-05 22:04:32,805 - Epoch: [175][ 120/ 1236] Overall Loss 0.188210 Objective Loss 0.188210 LR 0.000250 Time 0.026490 +2023-10-05 22:04:33,007 - Epoch: [175][ 130/ 1236] Overall Loss 0.188355 Objective Loss 0.188355 LR 0.000250 Time 0.026002 +2023-10-05 22:04:33,211 - Epoch: [175][ 140/ 1236] Overall Loss 0.188312 Objective Loss 0.188312 LR 0.000250 Time 0.025601 +2023-10-05 22:04:33,413 - Epoch: [175][ 150/ 1236] Overall Loss 0.187038 Objective Loss 0.187038 LR 0.000250 Time 0.025239 +2023-10-05 22:04:33,616 - Epoch: [175][ 160/ 1236] Overall Loss 0.186744 Objective Loss 0.186744 LR 0.000250 Time 0.024933 +2023-10-05 22:04:33,818 - Epoch: [175][ 170/ 1236] Overall Loss 0.188168 Objective Loss 0.188168 LR 0.000250 Time 0.024650 +2023-10-05 22:04:34,021 - Epoch: [175][ 180/ 1236] Overall Loss 0.187882 Objective Loss 0.187882 LR 0.000250 Time 0.024409 +2023-10-05 22:04:34,223 - Epoch: [175][ 190/ 1236] Overall Loss 0.187086 Objective Loss 0.187086 LR 0.000250 Time 0.024184 +2023-10-05 22:04:34,426 - Epoch: [175][ 200/ 1236] Overall Loss 0.187001 Objective Loss 0.187001 LR 0.000250 Time 0.023991 +2023-10-05 22:04:34,628 - Epoch: [175][ 210/ 1236] Overall Loss 0.186553 Objective Loss 0.186553 LR 0.000250 Time 0.023807 +2023-10-05 22:04:34,831 - Epoch: [175][ 220/ 1236] Overall Loss 0.186099 Objective Loss 0.186099 LR 0.000250 Time 0.023647 +2023-10-05 22:04:35,033 - Epoch: [175][ 230/ 1236] Overall Loss 0.184857 Objective Loss 0.184857 LR 0.000250 Time 0.023495 +2023-10-05 22:04:35,236 - Epoch: [175][ 240/ 1236] Overall Loss 0.184509 Objective Loss 0.184509 LR 0.000250 Time 0.023362 +2023-10-05 22:04:35,438 - Epoch: [175][ 250/ 1236] Overall Loss 0.184818 Objective Loss 0.184818 LR 0.000250 Time 0.023233 +2023-10-05 22:04:35,641 - Epoch: [175][ 260/ 1236] Overall Loss 0.185023 Objective Loss 0.185023 LR 0.000250 Time 0.023121 +2023-10-05 22:04:35,843 - Epoch: [175][ 270/ 1236] Overall Loss 0.184793 Objective Loss 0.184793 LR 0.000250 Time 0.023010 +2023-10-05 22:04:36,046 - Epoch: [175][ 280/ 1236] Overall Loss 0.185124 Objective Loss 0.185124 LR 0.000250 Time 0.022914 +2023-10-05 22:04:36,248 - Epoch: [175][ 290/ 1236] Overall Loss 0.185413 Objective Loss 0.185413 LR 0.000250 Time 0.022818 +2023-10-05 22:04:36,451 - Epoch: [175][ 300/ 1236] Overall Loss 0.185386 Objective Loss 0.185386 LR 0.000250 Time 0.022735 +2023-10-05 22:04:36,653 - Epoch: [175][ 310/ 1236] Overall Loss 0.185438 Objective Loss 0.185438 LR 0.000250 Time 0.022651 +2023-10-05 22:04:36,856 - Epoch: [175][ 320/ 1236] Overall Loss 0.185357 Objective Loss 0.185357 LR 0.000250 Time 0.022578 +2023-10-05 22:04:37,058 - Epoch: [175][ 330/ 1236] Overall Loss 0.185397 Objective Loss 0.185397 LR 0.000250 Time 0.022504 +2023-10-05 22:04:37,261 - Epoch: [175][ 340/ 1236] Overall Loss 0.185444 Objective Loss 0.185444 LR 0.000250 Time 0.022439 +2023-10-05 22:04:37,463 - Epoch: [175][ 350/ 1236] Overall Loss 0.185294 Objective Loss 0.185294 LR 0.000250 Time 0.022374 +2023-10-05 22:04:37,666 - Epoch: [175][ 360/ 1236] Overall Loss 0.185812 Objective Loss 0.185812 LR 0.000250 Time 0.022317 +2023-10-05 22:04:37,868 - Epoch: [175][ 370/ 1236] Overall Loss 0.185581 Objective Loss 0.185581 LR 0.000250 Time 0.022258 +2023-10-05 22:04:38,071 - Epoch: [175][ 380/ 1236] Overall Loss 0.185765 Objective Loss 0.185765 LR 0.000250 Time 0.022207 +2023-10-05 22:04:38,270 - Epoch: [175][ 390/ 1236] Overall Loss 0.185602 Objective Loss 0.185602 LR 0.000250 Time 0.022147 +2023-10-05 22:04:38,472 - Epoch: [175][ 400/ 1236] Overall Loss 0.185411 Objective Loss 0.185411 LR 0.000250 Time 0.022097 +2023-10-05 22:04:38,671 - Epoch: [175][ 410/ 1236] Overall Loss 0.185446 Objective Loss 0.185446 LR 0.000250 Time 0.022043 +2023-10-05 22:04:38,873 - Epoch: [175][ 420/ 1236] Overall Loss 0.185544 Objective Loss 0.185544 LR 0.000250 Time 0.021998 +2023-10-05 22:04:39,072 - Epoch: [175][ 430/ 1236] Overall Loss 0.185641 Objective Loss 0.185641 LR 0.000250 Time 0.021949 +2023-10-05 22:04:39,273 - Epoch: [175][ 440/ 1236] Overall Loss 0.185612 Objective Loss 0.185612 LR 0.000250 Time 0.021907 +2023-10-05 22:04:39,473 - Epoch: [175][ 450/ 1236] Overall Loss 0.185138 Objective Loss 0.185138 LR 0.000250 Time 0.021863 +2023-10-05 22:04:39,674 - Epoch: [175][ 460/ 1236] Overall Loss 0.185344 Objective Loss 0.185344 LR 0.000250 Time 0.021825 +2023-10-05 22:04:39,874 - Epoch: [175][ 470/ 1236] Overall Loss 0.185042 Objective Loss 0.185042 LR 0.000250 Time 0.021784 +2023-10-05 22:04:40,075 - Epoch: [175][ 480/ 1236] Overall Loss 0.185375 Objective Loss 0.185375 LR 0.000250 Time 0.021749 +2023-10-05 22:04:40,274 - Epoch: [175][ 490/ 1236] Overall Loss 0.185137 Objective Loss 0.185137 LR 0.000250 Time 0.021710 +2023-10-05 22:04:40,475 - Epoch: [175][ 500/ 1236] Overall Loss 0.185847 Objective Loss 0.185847 LR 0.000250 Time 0.021679 +2023-10-05 22:04:40,674 - Epoch: [175][ 510/ 1236] Overall Loss 0.185930 Objective Loss 0.185930 LR 0.000250 Time 0.021644 +2023-10-05 22:04:40,876 - Epoch: [175][ 520/ 1236] Overall Loss 0.186143 Objective Loss 0.186143 LR 0.000250 Time 0.021614 +2023-10-05 22:04:41,074 - Epoch: [175][ 530/ 1236] Overall Loss 0.186303 Objective Loss 0.186303 LR 0.000250 Time 0.021581 +2023-10-05 22:04:41,276 - Epoch: [175][ 540/ 1236] Overall Loss 0.186729 Objective Loss 0.186729 LR 0.000250 Time 0.021554 +2023-10-05 22:04:41,475 - Epoch: [175][ 550/ 1236] Overall Loss 0.187213 Objective Loss 0.187213 LR 0.000250 Time 0.021523 +2023-10-05 22:04:41,676 - Epoch: [175][ 560/ 1236] Overall Loss 0.187396 Objective Loss 0.187396 LR 0.000250 Time 0.021498 +2023-10-05 22:04:41,875 - Epoch: [175][ 570/ 1236] Overall Loss 0.187351 Objective Loss 0.187351 LR 0.000250 Time 0.021470 +2023-10-05 22:04:42,077 - Epoch: [175][ 580/ 1236] Overall Loss 0.187369 Objective Loss 0.187369 LR 0.000250 Time 0.021446 +2023-10-05 22:04:42,276 - Epoch: [175][ 590/ 1236] Overall Loss 0.187344 Objective Loss 0.187344 LR 0.000250 Time 0.021420 +2023-10-05 22:04:42,478 - Epoch: [175][ 600/ 1236] Overall Loss 0.187009 Objective Loss 0.187009 LR 0.000250 Time 0.021399 +2023-10-05 22:04:42,677 - Epoch: [175][ 610/ 1236] Overall Loss 0.187077 Objective Loss 0.187077 LR 0.000250 Time 0.021374 +2023-10-05 22:04:42,878 - Epoch: [175][ 620/ 1236] Overall Loss 0.187155 Objective Loss 0.187155 LR 0.000250 Time 0.021354 +2023-10-05 22:04:43,078 - Epoch: [175][ 630/ 1236] Overall Loss 0.187394 Objective Loss 0.187394 LR 0.000250 Time 0.021331 +2023-10-05 22:04:43,279 - Epoch: [175][ 640/ 1236] Overall Loss 0.187410 Objective Loss 0.187410 LR 0.000250 Time 0.021312 +2023-10-05 22:04:43,479 - Epoch: [175][ 650/ 1236] Overall Loss 0.187431 Objective Loss 0.187431 LR 0.000250 Time 0.021290 +2023-10-05 22:04:43,680 - Epoch: [175][ 660/ 1236] Overall Loss 0.187369 Objective Loss 0.187369 LR 0.000250 Time 0.021273 +2023-10-05 22:04:43,879 - Epoch: [175][ 670/ 1236] Overall Loss 0.187608 Objective Loss 0.187608 LR 0.000250 Time 0.021252 +2023-10-05 22:04:44,081 - Epoch: [175][ 680/ 1236] Overall Loss 0.187542 Objective Loss 0.187542 LR 0.000250 Time 0.021236 +2023-10-05 22:04:44,280 - Epoch: [175][ 690/ 1236] Overall Loss 0.187687 Objective Loss 0.187687 LR 0.000250 Time 0.021216 +2023-10-05 22:04:44,481 - Epoch: [175][ 700/ 1236] Overall Loss 0.187702 Objective Loss 0.187702 LR 0.000250 Time 0.021201 +2023-10-05 22:04:44,681 - Epoch: [175][ 710/ 1236] Overall Loss 0.187600 Objective Loss 0.187600 LR 0.000250 Time 0.021183 +2023-10-05 22:04:44,882 - Epoch: [175][ 720/ 1236] Overall Loss 0.187401 Objective Loss 0.187401 LR 0.000250 Time 0.021168 +2023-10-05 22:04:45,082 - Epoch: [175][ 730/ 1236] Overall Loss 0.187565 Objective Loss 0.187565 LR 0.000250 Time 0.021150 +2023-10-05 22:04:45,283 - Epoch: [175][ 740/ 1236] Overall Loss 0.187712 Objective Loss 0.187712 LR 0.000250 Time 0.021136 +2023-10-05 22:04:45,480 - Epoch: [175][ 750/ 1236] Overall Loss 0.188012 Objective Loss 0.188012 LR 0.000250 Time 0.021117 +2023-10-05 22:04:45,680 - Epoch: [175][ 760/ 1236] Overall Loss 0.188144 Objective Loss 0.188144 LR 0.000250 Time 0.021101 +2023-10-05 22:04:45,877 - Epoch: [175][ 770/ 1236] Overall Loss 0.188052 Objective Loss 0.188052 LR 0.000250 Time 0.021083 +2023-10-05 22:04:46,076 - Epoch: [175][ 780/ 1236] Overall Loss 0.187849 Objective Loss 0.187849 LR 0.000250 Time 0.021068 +2023-10-05 22:04:46,274 - Epoch: [175][ 790/ 1236] Overall Loss 0.187642 Objective Loss 0.187642 LR 0.000250 Time 0.021051 +2023-10-05 22:04:46,473 - Epoch: [175][ 800/ 1236] Overall Loss 0.187692 Objective Loss 0.187692 LR 0.000250 Time 0.021037 +2023-10-05 22:04:46,670 - Epoch: [175][ 810/ 1236] Overall Loss 0.187658 Objective Loss 0.187658 LR 0.000250 Time 0.021020 +2023-10-05 22:04:46,869 - Epoch: [175][ 820/ 1236] Overall Loss 0.187557 Objective Loss 0.187557 LR 0.000250 Time 0.021006 +2023-10-05 22:04:47,067 - Epoch: [175][ 830/ 1236] Overall Loss 0.187330 Objective Loss 0.187330 LR 0.000250 Time 0.020990 +2023-10-05 22:04:47,266 - Epoch: [175][ 840/ 1236] Overall Loss 0.187229 Objective Loss 0.187229 LR 0.000250 Time 0.020977 +2023-10-05 22:04:47,463 - Epoch: [175][ 850/ 1236] Overall Loss 0.187269 Objective Loss 0.187269 LR 0.000250 Time 0.020962 +2023-10-05 22:04:47,662 - Epoch: [175][ 860/ 1236] Overall Loss 0.186850 Objective Loss 0.186850 LR 0.000250 Time 0.020950 +2023-10-05 22:04:47,860 - Epoch: [175][ 870/ 1236] Overall Loss 0.186537 Objective Loss 0.186537 LR 0.000250 Time 0.020936 +2023-10-05 22:04:48,059 - Epoch: [175][ 880/ 1236] Overall Loss 0.186751 Objective Loss 0.186751 LR 0.000250 Time 0.020923 +2023-10-05 22:04:48,256 - Epoch: [175][ 890/ 1236] Overall Loss 0.186849 Objective Loss 0.186849 LR 0.000250 Time 0.020910 +2023-10-05 22:04:48,455 - Epoch: [175][ 900/ 1236] Overall Loss 0.186670 Objective Loss 0.186670 LR 0.000250 Time 0.020898 +2023-10-05 22:04:48,652 - Epoch: [175][ 910/ 1236] Overall Loss 0.187031 Objective Loss 0.187031 LR 0.000250 Time 0.020885 +2023-10-05 22:04:48,852 - Epoch: [175][ 920/ 1236] Overall Loss 0.187483 Objective Loss 0.187483 LR 0.000250 Time 0.020874 +2023-10-05 22:04:49,049 - Epoch: [175][ 930/ 1236] Overall Loss 0.187577 Objective Loss 0.187577 LR 0.000250 Time 0.020862 +2023-10-05 22:04:49,248 - Epoch: [175][ 940/ 1236] Overall Loss 0.187860 Objective Loss 0.187860 LR 0.000250 Time 0.020851 +2023-10-05 22:04:49,445 - Epoch: [175][ 950/ 1236] Overall Loss 0.187928 Objective Loss 0.187928 LR 0.000250 Time 0.020839 +2023-10-05 22:04:49,644 - Epoch: [175][ 960/ 1236] Overall Loss 0.187962 Objective Loss 0.187962 LR 0.000250 Time 0.020829 +2023-10-05 22:04:49,845 - Epoch: [175][ 970/ 1236] Overall Loss 0.187761 Objective Loss 0.187761 LR 0.000250 Time 0.020821 +2023-10-05 22:04:50,048 - Epoch: [175][ 980/ 1236] Overall Loss 0.187678 Objective Loss 0.187678 LR 0.000250 Time 0.020815 +2023-10-05 22:04:50,249 - Epoch: [175][ 990/ 1236] Overall Loss 0.187659 Objective Loss 0.187659 LR 0.000250 Time 0.020808 +2023-10-05 22:04:50,451 - Epoch: [175][ 1000/ 1236] Overall Loss 0.187672 Objective Loss 0.187672 LR 0.000250 Time 0.020801 +2023-10-05 22:04:50,652 - Epoch: [175][ 1010/ 1236] Overall Loss 0.187625 Objective Loss 0.187625 LR 0.000250 Time 0.020794 +2023-10-05 22:04:50,854 - Epoch: [175][ 1020/ 1236] Overall Loss 0.187630 Objective Loss 0.187630 LR 0.000250 Time 0.020787 +2023-10-05 22:04:51,055 - Epoch: [175][ 1030/ 1236] Overall Loss 0.187837 Objective Loss 0.187837 LR 0.000250 Time 0.020781 +2023-10-05 22:04:51,257 - Epoch: [175][ 1040/ 1236] Overall Loss 0.187591 Objective Loss 0.187591 LR 0.000250 Time 0.020775 +2023-10-05 22:04:51,459 - Epoch: [175][ 1050/ 1236] Overall Loss 0.187564 Objective Loss 0.187564 LR 0.000250 Time 0.020769 +2023-10-05 22:04:51,660 - Epoch: [175][ 1060/ 1236] Overall Loss 0.187569 Objective Loss 0.187569 LR 0.000250 Time 0.020763 +2023-10-05 22:04:51,862 - Epoch: [175][ 1070/ 1236] Overall Loss 0.187691 Objective Loss 0.187691 LR 0.000250 Time 0.020757 +2023-10-05 22:04:52,064 - Epoch: [175][ 1080/ 1236] Overall Loss 0.187602 Objective Loss 0.187602 LR 0.000250 Time 0.020751 +2023-10-05 22:04:52,266 - Epoch: [175][ 1090/ 1236] Overall Loss 0.187565 Objective Loss 0.187565 LR 0.000250 Time 0.020745 +2023-10-05 22:04:52,467 - Epoch: [175][ 1100/ 1236] Overall Loss 0.187698 Objective Loss 0.187698 LR 0.000250 Time 0.020740 +2023-10-05 22:04:52,669 - Epoch: [175][ 1110/ 1236] Overall Loss 0.187709 Objective Loss 0.187709 LR 0.000250 Time 0.020735 +2023-10-05 22:04:52,871 - Epoch: [175][ 1120/ 1236] Overall Loss 0.187662 Objective Loss 0.187662 LR 0.000250 Time 0.020729 +2023-10-05 22:04:53,073 - Epoch: [175][ 1130/ 1236] Overall Loss 0.187621 Objective Loss 0.187621 LR 0.000250 Time 0.020724 +2023-10-05 22:04:53,274 - Epoch: [175][ 1140/ 1236] Overall Loss 0.187615 Objective Loss 0.187615 LR 0.000250 Time 0.020719 +2023-10-05 22:04:53,476 - Epoch: [175][ 1150/ 1236] Overall Loss 0.187639 Objective Loss 0.187639 LR 0.000250 Time 0.020714 +2023-10-05 22:04:53,678 - Epoch: [175][ 1160/ 1236] Overall Loss 0.187698 Objective Loss 0.187698 LR 0.000250 Time 0.020709 +2023-10-05 22:04:53,880 - Epoch: [175][ 1170/ 1236] Overall Loss 0.187807 Objective Loss 0.187807 LR 0.000250 Time 0.020704 +2023-10-05 22:04:54,082 - Epoch: [175][ 1180/ 1236] Overall Loss 0.187680 Objective Loss 0.187680 LR 0.000250 Time 0.020699 +2023-10-05 22:04:54,283 - Epoch: [175][ 1190/ 1236] Overall Loss 0.187594 Objective Loss 0.187594 LR 0.000250 Time 0.020695 +2023-10-05 22:04:54,485 - Epoch: [175][ 1200/ 1236] Overall Loss 0.187593 Objective Loss 0.187593 LR 0.000250 Time 0.020690 +2023-10-05 22:04:54,687 - Epoch: [175][ 1210/ 1236] Overall Loss 0.187432 Objective Loss 0.187432 LR 0.000250 Time 0.020685 +2023-10-05 22:04:54,889 - Epoch: [175][ 1220/ 1236] Overall Loss 0.187541 Objective Loss 0.187541 LR 0.000250 Time 0.020681 +2023-10-05 22:04:55,142 - Epoch: [175][ 1230/ 1236] Overall Loss 0.187679 Objective Loss 0.187679 LR 0.000250 Time 0.020719 +2023-10-05 22:04:55,260 - Epoch: [175][ 1236/ 1236] Overall Loss 0.187723 Objective Loss 0.187723 Top1 90.835031 Top5 99.185336 LR 0.000250 Time 0.020714 +2023-10-05 22:04:55,376 - --- validate (epoch=175)----------- +2023-10-05 22:04:55,376 - 29943 samples (256 per mini-batch) +2023-10-05 22:04:55,824 - Epoch: [175][ 10/ 117] Loss 0.301863 Top1 85.468750 Top5 98.281250 +2023-10-05 22:04:55,971 - Epoch: [175][ 20/ 117] Loss 0.299164 Top1 85.761719 Top5 98.144531 +2023-10-05 22:04:56,117 - Epoch: [175][ 30/ 117] Loss 0.295829 Top1 86.080729 Top5 98.059896 +2023-10-05 22:04:56,263 - Epoch: [175][ 40/ 117] Loss 0.296650 Top1 86.113281 Top5 98.017578 +2023-10-05 22:04:56,411 - Epoch: [175][ 50/ 117] Loss 0.295993 Top1 86.085938 Top5 98.054688 +2023-10-05 22:04:56,558 - Epoch: [175][ 60/ 117] Loss 0.298453 Top1 85.800781 Top5 98.085938 +2023-10-05 22:04:56,706 - Epoch: [175][ 70/ 117] Loss 0.297751 Top1 85.574777 Top5 98.108259 +2023-10-05 22:04:56,852 - Epoch: [175][ 80/ 117] Loss 0.302863 Top1 85.327148 Top5 98.081055 +2023-10-05 22:04:57,002 - Epoch: [175][ 90/ 117] Loss 0.302622 Top1 85.329861 Top5 98.103299 +2023-10-05 22:04:57,149 - Epoch: [175][ 100/ 117] Loss 0.302334 Top1 85.253906 Top5 98.140625 +2023-10-05 22:04:57,303 - Epoch: [175][ 110/ 117] Loss 0.302508 Top1 85.362216 Top5 98.196023 +2023-10-05 22:04:57,388 - Epoch: [175][ 117/ 117] Loss 0.305066 Top1 85.318772 Top5 98.196573 +2023-10-05 22:04:57,499 - ==> Top1: 85.319 Top5: 98.197 Loss: 0.305 + +2023-10-05 22:04:57,500 - ==> Confusion: +[[ 946 3 4 1 6 3 0 0 3 58 1 0 0 2 8 1 2 1 0 0 11] + [ 0 1053 2 0 9 16 1 23 3 0 1 2 0 0 1 4 3 0 7 1 5] + [ 8 1 977 11 2 0 16 4 0 1 8 2 7 1 0 3 1 3 3 2 6] + [ 3 1 15 986 0 2 0 2 3 1 3 1 3 1 18 3 0 7 21 2 17] + [ 24 6 2 0 978 1 1 0 0 7 0 1 0 3 4 1 12 2 0 1 7] + [ 4 39 3 0 4 998 0 15 0 0 4 7 0 11 4 2 4 0 4 4 13] + [ 0 4 25 0 0 1 1123 8 0 0 2 2 3 0 1 8 0 1 2 7 4] + [ 7 17 10 0 2 34 2 1068 2 2 3 8 1 2 0 3 0 0 46 4 7] + [ 20 1 3 0 0 1 1 0 975 47 9 4 2 4 16 2 0 0 2 0 2] + [ 124 0 5 0 4 2 0 0 11 934 1 0 0 18 5 5 0 0 0 0 10] + [ 3 6 9 6 0 2 6 4 11 5 959 1 1 10 5 1 3 0 7 2 12] + [ 1 0 3 0 2 15 0 3 0 1 0 955 19 4 0 3 1 16 0 10 2] + [ 4 1 4 5 0 1 0 1 0 0 1 32 987 0 2 4 2 11 4 3 6] + [ 3 0 2 0 2 6 0 0 11 12 4 3 2 1061 3 1 1 0 0 0 8] + [ 13 2 2 7 7 0 0 0 23 5 1 1 2 2 1010 0 2 2 11 0 11] + [ 1 3 4 0 3 0 0 0 0 0 0 6 6 1 1 1067 17 14 0 8 3] + [ 0 14 1 0 4 3 0 1 1 0 0 4 1 1 3 9 1102 0 0 3 14] + [ 0 0 1 1 1 0 2 0 0 0 0 2 11 2 1 4 0 1008 2 0 3] + [ 2 5 5 23 1 0 1 26 1 0 2 0 1 0 10 0 0 0 980 1 10] + [ 0 2 7 1 1 6 7 8 0 0 1 14 3 0 0 8 10 2 5 1071 6] + [ 137 171 171 58 90 130 32 80 80 75 143 104 310 270 141 47 152 72 146 187 5309]] + +2023-10-05 22:04:57,502 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:04:57,502 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:04:57,508 - + +2023-10-05 22:04:57,508 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:04:58,606 - Epoch: [176][ 10/ 1236] Overall Loss 0.184912 Objective Loss 0.184912 LR 0.000250 Time 0.109753 +2023-10-05 22:04:58,808 - Epoch: [176][ 20/ 1236] Overall Loss 0.176197 Objective Loss 0.176197 LR 0.000250 Time 0.064960 +2023-10-05 22:04:59,009 - Epoch: [176][ 30/ 1236] Overall Loss 0.177646 Objective Loss 0.177646 LR 0.000250 Time 0.050005 +2023-10-05 22:04:59,211 - Epoch: [176][ 40/ 1236] Overall Loss 0.175419 Objective Loss 0.175419 LR 0.000250 Time 0.042539 +2023-10-05 22:04:59,412 - Epoch: [176][ 50/ 1236] Overall Loss 0.174745 Objective Loss 0.174745 LR 0.000250 Time 0.038054 +2023-10-05 22:04:59,614 - Epoch: [176][ 60/ 1236] Overall Loss 0.172892 Objective Loss 0.172892 LR 0.000250 Time 0.035071 +2023-10-05 22:04:59,816 - Epoch: [176][ 70/ 1236] Overall Loss 0.172655 Objective Loss 0.172655 LR 0.000250 Time 0.032935 +2023-10-05 22:05:00,018 - Epoch: [176][ 80/ 1236] Overall Loss 0.171133 Objective Loss 0.171133 LR 0.000250 Time 0.031335 +2023-10-05 22:05:00,220 - Epoch: [176][ 90/ 1236] Overall Loss 0.172425 Objective Loss 0.172425 LR 0.000250 Time 0.030095 +2023-10-05 22:05:00,421 - Epoch: [176][ 100/ 1236] Overall Loss 0.172775 Objective Loss 0.172775 LR 0.000250 Time 0.029096 +2023-10-05 22:05:00,625 - Epoch: [176][ 110/ 1236] Overall Loss 0.176936 Objective Loss 0.176936 LR 0.000250 Time 0.028306 +2023-10-05 22:05:00,831 - Epoch: [176][ 120/ 1236] Overall Loss 0.177848 Objective Loss 0.177848 LR 0.000250 Time 0.027659 +2023-10-05 22:05:01,036 - Epoch: [176][ 130/ 1236] Overall Loss 0.176756 Objective Loss 0.176756 LR 0.000250 Time 0.027098 +2023-10-05 22:05:01,242 - Epoch: [176][ 140/ 1236] Overall Loss 0.176761 Objective Loss 0.176761 LR 0.000250 Time 0.026629 +2023-10-05 22:05:01,447 - Epoch: [176][ 150/ 1236] Overall Loss 0.177954 Objective Loss 0.177954 LR 0.000250 Time 0.026208 +2023-10-05 22:05:01,653 - Epoch: [176][ 160/ 1236] Overall Loss 0.179323 Objective Loss 0.179323 LR 0.000250 Time 0.025855 +2023-10-05 22:05:01,858 - Epoch: [176][ 170/ 1236] Overall Loss 0.181407 Objective Loss 0.181407 LR 0.000250 Time 0.025530 +2023-10-05 22:05:02,063 - Epoch: [176][ 180/ 1236] Overall Loss 0.181210 Objective Loss 0.181210 LR 0.000250 Time 0.025252 +2023-10-05 22:05:02,269 - Epoch: [176][ 190/ 1236] Overall Loss 0.182341 Objective Loss 0.182341 LR 0.000250 Time 0.024996 +2023-10-05 22:05:02,474 - Epoch: [176][ 200/ 1236] Overall Loss 0.182238 Objective Loss 0.182238 LR 0.000250 Time 0.024773 +2023-10-05 22:05:02,679 - Epoch: [176][ 210/ 1236] Overall Loss 0.183520 Objective Loss 0.183520 LR 0.000250 Time 0.024563 +2023-10-05 22:05:02,885 - Epoch: [176][ 220/ 1236] Overall Loss 0.184676 Objective Loss 0.184676 LR 0.000250 Time 0.024380 +2023-10-05 22:05:03,091 - Epoch: [176][ 230/ 1236] Overall Loss 0.184523 Objective Loss 0.184523 LR 0.000250 Time 0.024206 +2023-10-05 22:05:03,296 - Epoch: [176][ 240/ 1236] Overall Loss 0.183811 Objective Loss 0.183811 LR 0.000250 Time 0.024052 +2023-10-05 22:05:03,501 - Epoch: [176][ 250/ 1236] Overall Loss 0.183100 Objective Loss 0.183100 LR 0.000250 Time 0.023905 +2023-10-05 22:05:03,707 - Epoch: [176][ 260/ 1236] Overall Loss 0.183199 Objective Loss 0.183199 LR 0.000250 Time 0.023777 +2023-10-05 22:05:03,913 - Epoch: [176][ 270/ 1236] Overall Loss 0.183158 Objective Loss 0.183158 LR 0.000250 Time 0.023651 +2023-10-05 22:05:04,119 - Epoch: [176][ 280/ 1236] Overall Loss 0.182860 Objective Loss 0.182860 LR 0.000250 Time 0.023541 +2023-10-05 22:05:04,324 - Epoch: [176][ 290/ 1236] Overall Loss 0.182815 Objective Loss 0.182815 LR 0.000250 Time 0.023431 +2023-10-05 22:05:04,529 - Epoch: [176][ 300/ 1236] Overall Loss 0.183952 Objective Loss 0.183952 LR 0.000250 Time 0.023334 +2023-10-05 22:05:04,734 - Epoch: [176][ 310/ 1236] Overall Loss 0.183926 Objective Loss 0.183926 LR 0.000250 Time 0.023238 +2023-10-05 22:05:04,941 - Epoch: [176][ 320/ 1236] Overall Loss 0.183633 Objective Loss 0.183633 LR 0.000250 Time 0.023156 +2023-10-05 22:05:05,146 - Epoch: [176][ 330/ 1236] Overall Loss 0.183661 Objective Loss 0.183661 LR 0.000250 Time 0.023071 +2023-10-05 22:05:05,352 - Epoch: [176][ 340/ 1236] Overall Loss 0.183568 Objective Loss 0.183568 LR 0.000250 Time 0.022997 +2023-10-05 22:05:05,557 - Epoch: [176][ 350/ 1236] Overall Loss 0.183741 Objective Loss 0.183741 LR 0.000250 Time 0.022921 +2023-10-05 22:05:05,763 - Epoch: [176][ 360/ 1236] Overall Loss 0.183431 Objective Loss 0.183431 LR 0.000250 Time 0.022856 +2023-10-05 22:05:05,965 - Epoch: [176][ 370/ 1236] Overall Loss 0.182927 Objective Loss 0.182927 LR 0.000250 Time 0.022781 +2023-10-05 22:05:06,167 - Epoch: [176][ 380/ 1236] Overall Loss 0.182884 Objective Loss 0.182884 LR 0.000250 Time 0.022712 +2023-10-05 22:05:06,368 - Epoch: [176][ 390/ 1236] Overall Loss 0.182454 Objective Loss 0.182454 LR 0.000250 Time 0.022644 +2023-10-05 22:05:06,571 - Epoch: [176][ 400/ 1236] Overall Loss 0.182182 Objective Loss 0.182182 LR 0.000250 Time 0.022583 +2023-10-05 22:05:06,773 - Epoch: [176][ 410/ 1236] Overall Loss 0.182103 Objective Loss 0.182103 LR 0.000250 Time 0.022525 +2023-10-05 22:05:06,976 - Epoch: [176][ 420/ 1236] Overall Loss 0.182165 Objective Loss 0.182165 LR 0.000250 Time 0.022470 +2023-10-05 22:05:07,178 - Epoch: [176][ 430/ 1236] Overall Loss 0.182125 Objective Loss 0.182125 LR 0.000250 Time 0.022418 +2023-10-05 22:05:07,380 - Epoch: [176][ 440/ 1236] Overall Loss 0.182881 Objective Loss 0.182881 LR 0.000250 Time 0.022367 +2023-10-05 22:05:07,582 - Epoch: [176][ 450/ 1236] Overall Loss 0.183307 Objective Loss 0.183307 LR 0.000250 Time 0.022318 +2023-10-05 22:05:07,784 - Epoch: [176][ 460/ 1236] Overall Loss 0.183709 Objective Loss 0.183709 LR 0.000250 Time 0.022271 +2023-10-05 22:05:07,986 - Epoch: [176][ 470/ 1236] Overall Loss 0.183143 Objective Loss 0.183143 LR 0.000250 Time 0.022226 +2023-10-05 22:05:08,188 - Epoch: [176][ 480/ 1236] Overall Loss 0.182949 Objective Loss 0.182949 LR 0.000250 Time 0.022183 +2023-10-05 22:05:08,390 - Epoch: [176][ 490/ 1236] Overall Loss 0.182895 Objective Loss 0.182895 LR 0.000250 Time 0.022142 +2023-10-05 22:05:08,592 - Epoch: [176][ 500/ 1236] Overall Loss 0.183343 Objective Loss 0.183343 LR 0.000250 Time 0.022103 +2023-10-05 22:05:08,794 - Epoch: [176][ 510/ 1236] Overall Loss 0.183430 Objective Loss 0.183430 LR 0.000250 Time 0.022065 +2023-10-05 22:05:08,996 - Epoch: [176][ 520/ 1236] Overall Loss 0.183267 Objective Loss 0.183267 LR 0.000250 Time 0.022029 +2023-10-05 22:05:09,198 - Epoch: [176][ 530/ 1236] Overall Loss 0.183188 Objective Loss 0.183188 LR 0.000250 Time 0.021993 +2023-10-05 22:05:09,401 - Epoch: [176][ 540/ 1236] Overall Loss 0.183449 Objective Loss 0.183449 LR 0.000250 Time 0.021960 +2023-10-05 22:05:09,603 - Epoch: [176][ 550/ 1236] Overall Loss 0.183454 Objective Loss 0.183454 LR 0.000250 Time 0.021928 +2023-10-05 22:05:09,805 - Epoch: [176][ 560/ 1236] Overall Loss 0.183635 Objective Loss 0.183635 LR 0.000250 Time 0.021896 +2023-10-05 22:05:10,007 - Epoch: [176][ 570/ 1236] Overall Loss 0.183756 Objective Loss 0.183756 LR 0.000250 Time 0.021866 +2023-10-05 22:05:10,209 - Epoch: [176][ 580/ 1236] Overall Loss 0.184168 Objective Loss 0.184168 LR 0.000250 Time 0.021837 +2023-10-05 22:05:10,411 - Epoch: [176][ 590/ 1236] Overall Loss 0.184107 Objective Loss 0.184107 LR 0.000250 Time 0.021809 +2023-10-05 22:05:10,613 - Epoch: [176][ 600/ 1236] Overall Loss 0.184152 Objective Loss 0.184152 LR 0.000250 Time 0.021782 +2023-10-05 22:05:10,815 - Epoch: [176][ 610/ 1236] Overall Loss 0.184161 Objective Loss 0.184161 LR 0.000250 Time 0.021755 +2023-10-05 22:05:11,018 - Epoch: [176][ 620/ 1236] Overall Loss 0.184259 Objective Loss 0.184259 LR 0.000250 Time 0.021730 +2023-10-05 22:05:11,219 - Epoch: [176][ 630/ 1236] Overall Loss 0.184286 Objective Loss 0.184286 LR 0.000250 Time 0.021705 +2023-10-05 22:05:11,421 - Epoch: [176][ 640/ 1236] Overall Loss 0.184321 Objective Loss 0.184321 LR 0.000250 Time 0.021681 +2023-10-05 22:05:11,623 - Epoch: [176][ 650/ 1236] Overall Loss 0.184347 Objective Loss 0.184347 LR 0.000250 Time 0.021658 +2023-10-05 22:05:11,825 - Epoch: [176][ 660/ 1236] Overall Loss 0.184549 Objective Loss 0.184549 LR 0.000250 Time 0.021635 +2023-10-05 22:05:12,027 - Epoch: [176][ 670/ 1236] Overall Loss 0.184536 Objective Loss 0.184536 LR 0.000250 Time 0.021613 +2023-10-05 22:05:12,230 - Epoch: [176][ 680/ 1236] Overall Loss 0.184607 Objective Loss 0.184607 LR 0.000250 Time 0.021592 +2023-10-05 22:05:12,432 - Epoch: [176][ 690/ 1236] Overall Loss 0.184636 Objective Loss 0.184636 LR 0.000250 Time 0.021572 +2023-10-05 22:05:12,634 - Epoch: [176][ 700/ 1236] Overall Loss 0.184501 Objective Loss 0.184501 LR 0.000250 Time 0.021552 +2023-10-05 22:05:12,836 - Epoch: [176][ 710/ 1236] Overall Loss 0.184370 Objective Loss 0.184370 LR 0.000250 Time 0.021532 +2023-10-05 22:05:13,038 - Epoch: [176][ 720/ 1236] Overall Loss 0.184396 Objective Loss 0.184396 LR 0.000250 Time 0.021513 +2023-10-05 22:05:13,240 - Epoch: [176][ 730/ 1236] Overall Loss 0.184378 Objective Loss 0.184378 LR 0.000250 Time 0.021495 +2023-10-05 22:05:13,442 - Epoch: [176][ 740/ 1236] Overall Loss 0.184315 Objective Loss 0.184315 LR 0.000250 Time 0.021477 +2023-10-05 22:05:13,644 - Epoch: [176][ 750/ 1236] Overall Loss 0.184184 Objective Loss 0.184184 LR 0.000250 Time 0.021459 +2023-10-05 22:05:13,846 - Epoch: [176][ 760/ 1236] Overall Loss 0.184477 Objective Loss 0.184477 LR 0.000250 Time 0.021442 +2023-10-05 22:05:14,047 - Epoch: [176][ 770/ 1236] Overall Loss 0.184365 Objective Loss 0.184365 LR 0.000250 Time 0.021425 +2023-10-05 22:05:14,249 - Epoch: [176][ 780/ 1236] Overall Loss 0.184084 Objective Loss 0.184084 LR 0.000250 Time 0.021408 +2023-10-05 22:05:14,451 - Epoch: [176][ 790/ 1236] Overall Loss 0.184131 Objective Loss 0.184131 LR 0.000250 Time 0.021392 +2023-10-05 22:05:14,653 - Epoch: [176][ 800/ 1236] Overall Loss 0.183950 Objective Loss 0.183950 LR 0.000250 Time 0.021377 +2023-10-05 22:05:14,855 - Epoch: [176][ 810/ 1236] Overall Loss 0.183781 Objective Loss 0.183781 LR 0.000250 Time 0.021362 +2023-10-05 22:05:15,057 - Epoch: [176][ 820/ 1236] Overall Loss 0.183755 Objective Loss 0.183755 LR 0.000250 Time 0.021348 +2023-10-05 22:05:15,259 - Epoch: [176][ 830/ 1236] Overall Loss 0.183823 Objective Loss 0.183823 LR 0.000250 Time 0.021334 +2023-10-05 22:05:15,461 - Epoch: [176][ 840/ 1236] Overall Loss 0.183730 Objective Loss 0.183730 LR 0.000250 Time 0.021319 +2023-10-05 22:05:15,663 - Epoch: [176][ 850/ 1236] Overall Loss 0.184018 Objective Loss 0.184018 LR 0.000250 Time 0.021306 +2023-10-05 22:05:15,865 - Epoch: [176][ 860/ 1236] Overall Loss 0.183976 Objective Loss 0.183976 LR 0.000250 Time 0.021293 +2023-10-05 22:05:16,067 - Epoch: [176][ 870/ 1236] Overall Loss 0.183957 Objective Loss 0.183957 LR 0.000250 Time 0.021280 +2023-10-05 22:05:16,269 - Epoch: [176][ 880/ 1236] Overall Loss 0.184300 Objective Loss 0.184300 LR 0.000250 Time 0.021267 +2023-10-05 22:05:16,471 - Epoch: [176][ 890/ 1236] Overall Loss 0.184055 Objective Loss 0.184055 LR 0.000250 Time 0.021255 +2023-10-05 22:05:16,673 - Epoch: [176][ 900/ 1236] Overall Loss 0.184135 Objective Loss 0.184135 LR 0.000250 Time 0.021243 +2023-10-05 22:05:16,875 - Epoch: [176][ 910/ 1236] Overall Loss 0.184278 Objective Loss 0.184278 LR 0.000250 Time 0.021231 +2023-10-05 22:05:17,077 - Epoch: [176][ 920/ 1236] Overall Loss 0.184375 Objective Loss 0.184375 LR 0.000250 Time 0.021220 +2023-10-05 22:05:17,279 - Epoch: [176][ 930/ 1236] Overall Loss 0.184460 Objective Loss 0.184460 LR 0.000250 Time 0.021208 +2023-10-05 22:05:17,481 - Epoch: [176][ 940/ 1236] Overall Loss 0.184402 Objective Loss 0.184402 LR 0.000250 Time 0.021197 +2023-10-05 22:05:17,683 - Epoch: [176][ 950/ 1236] Overall Loss 0.184563 Objective Loss 0.184563 LR 0.000250 Time 0.021187 +2023-10-05 22:05:17,885 - Epoch: [176][ 960/ 1236] Overall Loss 0.184476 Objective Loss 0.184476 LR 0.000250 Time 0.021176 +2023-10-05 22:05:18,087 - Epoch: [176][ 970/ 1236] Overall Loss 0.184453 Objective Loss 0.184453 LR 0.000250 Time 0.021166 +2023-10-05 22:05:18,290 - Epoch: [176][ 980/ 1236] Overall Loss 0.184299 Objective Loss 0.184299 LR 0.000250 Time 0.021156 +2023-10-05 22:05:18,492 - Epoch: [176][ 990/ 1236] Overall Loss 0.184304 Objective Loss 0.184304 LR 0.000250 Time 0.021146 +2023-10-05 22:05:18,694 - Epoch: [176][ 1000/ 1236] Overall Loss 0.184218 Objective Loss 0.184218 LR 0.000250 Time 0.021136 +2023-10-05 22:05:18,896 - Epoch: [176][ 1010/ 1236] Overall Loss 0.184084 Objective Loss 0.184084 LR 0.000250 Time 0.021127 +2023-10-05 22:05:19,098 - Epoch: [176][ 1020/ 1236] Overall Loss 0.183903 Objective Loss 0.183903 LR 0.000250 Time 0.021117 +2023-10-05 22:05:19,300 - Epoch: [176][ 1030/ 1236] Overall Loss 0.184106 Objective Loss 0.184106 LR 0.000250 Time 0.021108 +2023-10-05 22:05:19,502 - Epoch: [176][ 1040/ 1236] Overall Loss 0.184096 Objective Loss 0.184096 LR 0.000250 Time 0.021099 +2023-10-05 22:05:19,704 - Epoch: [176][ 1050/ 1236] Overall Loss 0.184174 Objective Loss 0.184174 LR 0.000250 Time 0.021090 +2023-10-05 22:05:19,906 - Epoch: [176][ 1060/ 1236] Overall Loss 0.184121 Objective Loss 0.184121 LR 0.000250 Time 0.021082 +2023-10-05 22:05:20,108 - Epoch: [176][ 1070/ 1236] Overall Loss 0.184395 Objective Loss 0.184395 LR 0.000250 Time 0.021073 +2023-10-05 22:05:20,311 - Epoch: [176][ 1080/ 1236] Overall Loss 0.184480 Objective Loss 0.184480 LR 0.000250 Time 0.021065 +2023-10-05 22:05:20,512 - Epoch: [176][ 1090/ 1236] Overall Loss 0.184206 Objective Loss 0.184206 LR 0.000250 Time 0.021057 +2023-10-05 22:05:20,715 - Epoch: [176][ 1100/ 1236] Overall Loss 0.184151 Objective Loss 0.184151 LR 0.000250 Time 0.021049 +2023-10-05 22:05:20,917 - Epoch: [176][ 1110/ 1236] Overall Loss 0.184341 Objective Loss 0.184341 LR 0.000250 Time 0.021041 +2023-10-05 22:05:21,119 - Epoch: [176][ 1120/ 1236] Overall Loss 0.184523 Objective Loss 0.184523 LR 0.000250 Time 0.021033 +2023-10-05 22:05:21,321 - Epoch: [176][ 1130/ 1236] Overall Loss 0.184530 Objective Loss 0.184530 LR 0.000250 Time 0.021026 +2023-10-05 22:05:21,523 - Epoch: [176][ 1140/ 1236] Overall Loss 0.184763 Objective Loss 0.184763 LR 0.000250 Time 0.021018 +2023-10-05 22:05:21,725 - Epoch: [176][ 1150/ 1236] Overall Loss 0.184917 Objective Loss 0.184917 LR 0.000250 Time 0.021011 +2023-10-05 22:05:21,927 - Epoch: [176][ 1160/ 1236] Overall Loss 0.184963 Objective Loss 0.184963 LR 0.000250 Time 0.021003 +2023-10-05 22:05:22,129 - Epoch: [176][ 1170/ 1236] Overall Loss 0.185009 Objective Loss 0.185009 LR 0.000250 Time 0.020996 +2023-10-05 22:05:22,331 - Epoch: [176][ 1180/ 1236] Overall Loss 0.184970 Objective Loss 0.184970 LR 0.000250 Time 0.020989 +2023-10-05 22:05:22,533 - Epoch: [176][ 1190/ 1236] Overall Loss 0.185100 Objective Loss 0.185100 LR 0.000250 Time 0.020983 +2023-10-05 22:05:22,735 - Epoch: [176][ 1200/ 1236] Overall Loss 0.185136 Objective Loss 0.185136 LR 0.000250 Time 0.020976 +2023-10-05 22:05:22,937 - Epoch: [176][ 1210/ 1236] Overall Loss 0.185352 Objective Loss 0.185352 LR 0.000250 Time 0.020969 +2023-10-05 22:05:23,140 - Epoch: [176][ 1220/ 1236] Overall Loss 0.185277 Objective Loss 0.185277 LR 0.000250 Time 0.020963 +2023-10-05 22:05:23,393 - Epoch: [176][ 1230/ 1236] Overall Loss 0.185323 Objective Loss 0.185323 LR 0.000250 Time 0.020998 +2023-10-05 22:05:23,511 - Epoch: [176][ 1236/ 1236] Overall Loss 0.185494 Objective Loss 0.185494 Top1 86.761711 Top5 98.370672 LR 0.000250 Time 0.020991 +2023-10-05 22:05:23,636 - --- validate (epoch=176)----------- +2023-10-05 22:05:23,636 - 29943 samples (256 per mini-batch) +2023-10-05 22:05:24,092 - Epoch: [176][ 10/ 117] Loss 0.314442 Top1 84.218750 Top5 97.734375 +2023-10-05 22:05:24,242 - Epoch: [176][ 20/ 117] Loss 0.312572 Top1 84.960938 Top5 97.773438 +2023-10-05 22:05:24,389 - Epoch: [176][ 30/ 117] Loss 0.316492 Top1 85.013021 Top5 97.838542 +2023-10-05 22:05:24,538 - Epoch: [176][ 40/ 117] Loss 0.308979 Top1 85.322266 Top5 97.968750 +2023-10-05 22:05:24,685 - Epoch: [176][ 50/ 117] Loss 0.305970 Top1 85.335938 Top5 98.070312 +2023-10-05 22:05:24,832 - Epoch: [176][ 60/ 117] Loss 0.300269 Top1 85.546875 Top5 98.131510 +2023-10-05 22:05:24,984 - Epoch: [176][ 70/ 117] Loss 0.298986 Top1 85.619420 Top5 98.147321 +2023-10-05 22:05:25,138 - Epoch: [176][ 80/ 117] Loss 0.302025 Top1 85.590820 Top5 98.115234 +2023-10-05 22:05:25,292 - Epoch: [176][ 90/ 117] Loss 0.301617 Top1 85.546875 Top5 98.159722 +2023-10-05 22:05:25,448 - Epoch: [176][ 100/ 117] Loss 0.302798 Top1 85.531250 Top5 98.128906 +2023-10-05 22:05:25,609 - Epoch: [176][ 110/ 117] Loss 0.305442 Top1 85.493608 Top5 98.156960 +2023-10-05 22:05:25,694 - Epoch: [176][ 117/ 117] Loss 0.305142 Top1 85.502455 Top5 98.139799 +2023-10-05 22:05:25,807 - ==> Top1: 85.502 Top5: 98.140 Loss: 0.305 + +2023-10-05 22:05:25,808 - ==> Confusion: +[[ 931 5 1 2 9 3 0 0 6 67 1 1 0 2 4 2 1 1 1 0 13] + [ 1 1056 2 0 8 22 1 18 1 0 0 2 0 0 1 3 2 1 6 0 7] + [ 6 1 964 12 0 0 25 8 0 1 4 1 5 2 0 3 1 3 5 7 8] + [ 3 1 8 973 1 4 1 1 1 2 10 1 5 1 22 2 0 4 27 1 21] + [ 20 4 2 0 983 3 1 0 0 10 1 1 0 1 10 2 5 1 1 1 4] + [ 3 33 1 0 4 1001 0 18 1 0 8 5 0 10 5 2 3 0 5 3 14] + [ 0 5 20 0 0 2 1126 10 0 0 2 5 1 0 1 6 0 0 0 8 5] + [ 4 18 8 0 1 27 3 1083 1 3 5 11 2 1 0 5 1 0 32 5 8] + [ 20 2 2 0 0 2 0 1 981 36 11 1 1 4 17 5 1 0 3 1 1] + [ 91 0 3 1 6 2 0 0 17 955 0 2 1 25 4 1 0 0 0 3 8] + [ 3 6 7 3 0 2 4 5 8 1 975 4 0 14 5 1 1 0 3 3 8] + [ 1 0 1 1 1 12 0 2 0 0 0 961 14 8 0 0 1 17 0 10 6] + [ 1 1 3 4 0 1 0 1 0 0 1 38 985 1 2 1 2 11 3 3 10] + [ 1 0 2 1 2 12 0 1 9 9 5 5 2 1055 2 2 1 0 0 3 7] + [ 11 0 4 12 6 0 0 0 30 0 3 1 1 2 1005 0 2 2 8 0 14] + [ 0 3 2 0 2 0 0 1 0 1 0 12 5 2 1 1066 18 8 0 7 6] + [ 1 15 1 0 6 6 0 1 1 0 0 6 0 1 2 10 1097 0 0 4 10] + [ 0 0 0 2 0 0 3 0 2 0 0 3 20 0 0 5 1 998 1 1 2] + [ 1 9 6 22 1 0 0 21 2 0 5 0 0 0 8 0 0 0 984 2 7] + [ 0 2 2 1 2 6 8 8 0 0 2 17 4 1 0 5 6 1 4 1073 10] + [ 135 161 140 48 90 132 42 99 88 71 195 100 309 279 133 44 141 57 122 169 5350]] + +2023-10-05 22:05:25,810 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:05:25,810 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:05:25,816 - + +2023-10-05 22:05:25,816 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:05:26,801 - Epoch: [177][ 10/ 1236] Overall Loss 0.164191 Objective Loss 0.164191 LR 0.000250 Time 0.098450 +2023-10-05 22:05:27,004 - Epoch: [177][ 20/ 1236] Overall Loss 0.167518 Objective Loss 0.167518 LR 0.000250 Time 0.059356 +2023-10-05 22:05:27,206 - Epoch: [177][ 30/ 1236] Overall Loss 0.169865 Objective Loss 0.169865 LR 0.000250 Time 0.046284 +2023-10-05 22:05:27,409 - Epoch: [177][ 40/ 1236] Overall Loss 0.172476 Objective Loss 0.172476 LR 0.000250 Time 0.039778 +2023-10-05 22:05:27,610 - Epoch: [177][ 50/ 1236] Overall Loss 0.179204 Objective Loss 0.179204 LR 0.000250 Time 0.035838 +2023-10-05 22:05:27,813 - Epoch: [177][ 60/ 1236] Overall Loss 0.177356 Objective Loss 0.177356 LR 0.000250 Time 0.033241 +2023-10-05 22:05:28,015 - Epoch: [177][ 70/ 1236] Overall Loss 0.178086 Objective Loss 0.178086 LR 0.000250 Time 0.031379 +2023-10-05 22:05:28,218 - Epoch: [177][ 80/ 1236] Overall Loss 0.181477 Objective Loss 0.181477 LR 0.000250 Time 0.029985 +2023-10-05 22:05:28,420 - Epoch: [177][ 90/ 1236] Overall Loss 0.181973 Objective Loss 0.181973 LR 0.000250 Time 0.028902 +2023-10-05 22:05:28,623 - Epoch: [177][ 100/ 1236] Overall Loss 0.183350 Objective Loss 0.183350 LR 0.000250 Time 0.028036 +2023-10-05 22:05:28,824 - Epoch: [177][ 110/ 1236] Overall Loss 0.182785 Objective Loss 0.182785 LR 0.000250 Time 0.027311 +2023-10-05 22:05:29,024 - Epoch: [177][ 120/ 1236] Overall Loss 0.182719 Objective Loss 0.182719 LR 0.000250 Time 0.026700 +2023-10-05 22:05:29,224 - Epoch: [177][ 130/ 1236] Overall Loss 0.183576 Objective Loss 0.183576 LR 0.000250 Time 0.026182 +2023-10-05 22:05:29,425 - Epoch: [177][ 140/ 1236] Overall Loss 0.184316 Objective Loss 0.184316 LR 0.000250 Time 0.025745 +2023-10-05 22:05:29,625 - Epoch: [177][ 150/ 1236] Overall Loss 0.183398 Objective Loss 0.183398 LR 0.000250 Time 0.025358 +2023-10-05 22:05:29,826 - Epoch: [177][ 160/ 1236] Overall Loss 0.183020 Objective Loss 0.183020 LR 0.000250 Time 0.025028 +2023-10-05 22:05:30,028 - Epoch: [177][ 170/ 1236] Overall Loss 0.183150 Objective Loss 0.183150 LR 0.000250 Time 0.024743 +2023-10-05 22:05:30,229 - Epoch: [177][ 180/ 1236] Overall Loss 0.183437 Objective Loss 0.183437 LR 0.000250 Time 0.024482 +2023-10-05 22:05:30,429 - Epoch: [177][ 190/ 1236] Overall Loss 0.184571 Objective Loss 0.184571 LR 0.000250 Time 0.024243 +2023-10-05 22:05:30,630 - Epoch: [177][ 200/ 1236] Overall Loss 0.183815 Objective Loss 0.183815 LR 0.000250 Time 0.024035 +2023-10-05 22:05:30,829 - Epoch: [177][ 210/ 1236] Overall Loss 0.182520 Objective Loss 0.182520 LR 0.000250 Time 0.023835 +2023-10-05 22:05:31,028 - Epoch: [177][ 220/ 1236] Overall Loss 0.183293 Objective Loss 0.183293 LR 0.000250 Time 0.023655 +2023-10-05 22:05:31,226 - Epoch: [177][ 230/ 1236] Overall Loss 0.183343 Objective Loss 0.183343 LR 0.000250 Time 0.023489 +2023-10-05 22:05:31,426 - Epoch: [177][ 240/ 1236] Overall Loss 0.184173 Objective Loss 0.184173 LR 0.000250 Time 0.023341 +2023-10-05 22:05:31,625 - Epoch: [177][ 250/ 1236] Overall Loss 0.183904 Objective Loss 0.183904 LR 0.000250 Time 0.023201 +2023-10-05 22:05:31,825 - Epoch: [177][ 260/ 1236] Overall Loss 0.184605 Objective Loss 0.184605 LR 0.000250 Time 0.023075 +2023-10-05 22:05:32,023 - Epoch: [177][ 270/ 1236] Overall Loss 0.184265 Objective Loss 0.184265 LR 0.000250 Time 0.022954 +2023-10-05 22:05:32,223 - Epoch: [177][ 280/ 1236] Overall Loss 0.184771 Objective Loss 0.184771 LR 0.000250 Time 0.022847 +2023-10-05 22:05:32,422 - Epoch: [177][ 290/ 1236] Overall Loss 0.185126 Objective Loss 0.185126 LR 0.000250 Time 0.022744 +2023-10-05 22:05:32,621 - Epoch: [177][ 300/ 1236] Overall Loss 0.185073 Objective Loss 0.185073 LR 0.000250 Time 0.022649 +2023-10-05 22:05:32,822 - Epoch: [177][ 310/ 1236] Overall Loss 0.184552 Objective Loss 0.184552 LR 0.000250 Time 0.022564 +2023-10-05 22:05:33,023 - Epoch: [177][ 320/ 1236] Overall Loss 0.184612 Objective Loss 0.184612 LR 0.000250 Time 0.022487 +2023-10-05 22:05:33,221 - Epoch: [177][ 330/ 1236] Overall Loss 0.184254 Objective Loss 0.184254 LR 0.000250 Time 0.022405 +2023-10-05 22:05:33,422 - Epoch: [177][ 340/ 1236] Overall Loss 0.183546 Objective Loss 0.183546 LR 0.000250 Time 0.022335 +2023-10-05 22:05:33,622 - Epoch: [177][ 350/ 1236] Overall Loss 0.183893 Objective Loss 0.183893 LR 0.000250 Time 0.022268 +2023-10-05 22:05:33,824 - Epoch: [177][ 360/ 1236] Overall Loss 0.183429 Objective Loss 0.183429 LR 0.000250 Time 0.022211 +2023-10-05 22:05:34,025 - Epoch: [177][ 370/ 1236] Overall Loss 0.183541 Objective Loss 0.183541 LR 0.000250 Time 0.022152 +2023-10-05 22:05:34,228 - Epoch: [177][ 380/ 1236] Overall Loss 0.183877 Objective Loss 0.183877 LR 0.000250 Time 0.022103 +2023-10-05 22:05:34,428 - Epoch: [177][ 390/ 1236] Overall Loss 0.183528 Objective Loss 0.183528 LR 0.000250 Time 0.022049 +2023-10-05 22:05:34,632 - Epoch: [177][ 400/ 1236] Overall Loss 0.183646 Objective Loss 0.183646 LR 0.000250 Time 0.022004 +2023-10-05 22:05:34,832 - Epoch: [177][ 410/ 1236] Overall Loss 0.183332 Objective Loss 0.183332 LR 0.000250 Time 0.021956 +2023-10-05 22:05:35,035 - Epoch: [177][ 420/ 1236] Overall Loss 0.183444 Objective Loss 0.183444 LR 0.000250 Time 0.021915 +2023-10-05 22:05:35,235 - Epoch: [177][ 430/ 1236] Overall Loss 0.183331 Objective Loss 0.183331 LR 0.000250 Time 0.021870 +2023-10-05 22:05:35,438 - Epoch: [177][ 440/ 1236] Overall Loss 0.182746 Objective Loss 0.182746 LR 0.000250 Time 0.021834 +2023-10-05 22:05:35,639 - Epoch: [177][ 450/ 1236] Overall Loss 0.182576 Objective Loss 0.182576 LR 0.000250 Time 0.021794 +2023-10-05 22:05:35,842 - Epoch: [177][ 460/ 1236] Overall Loss 0.182627 Objective Loss 0.182627 LR 0.000250 Time 0.021760 +2023-10-05 22:05:36,042 - Epoch: [177][ 470/ 1236] Overall Loss 0.183097 Objective Loss 0.183097 LR 0.000250 Time 0.021723 +2023-10-05 22:05:36,245 - Epoch: [177][ 480/ 1236] Overall Loss 0.183101 Objective Loss 0.183101 LR 0.000250 Time 0.021692 +2023-10-05 22:05:36,446 - Epoch: [177][ 490/ 1236] Overall Loss 0.183251 Objective Loss 0.183251 LR 0.000250 Time 0.021658 +2023-10-05 22:05:36,648 - Epoch: [177][ 500/ 1236] Overall Loss 0.183671 Objective Loss 0.183671 LR 0.000250 Time 0.021629 +2023-10-05 22:05:36,848 - Epoch: [177][ 510/ 1236] Overall Loss 0.183545 Objective Loss 0.183545 LR 0.000250 Time 0.021597 +2023-10-05 22:05:37,052 - Epoch: [177][ 520/ 1236] Overall Loss 0.183416 Objective Loss 0.183416 LR 0.000250 Time 0.021572 +2023-10-05 22:05:37,252 - Epoch: [177][ 530/ 1236] Overall Loss 0.183260 Objective Loss 0.183260 LR 0.000250 Time 0.021543 +2023-10-05 22:05:37,455 - Epoch: [177][ 540/ 1236] Overall Loss 0.183298 Objective Loss 0.183298 LR 0.000250 Time 0.021518 +2023-10-05 22:05:37,655 - Epoch: [177][ 550/ 1236] Overall Loss 0.183248 Objective Loss 0.183248 LR 0.000250 Time 0.021491 +2023-10-05 22:05:37,858 - Epoch: [177][ 560/ 1236] Overall Loss 0.183191 Objective Loss 0.183191 LR 0.000250 Time 0.021469 +2023-10-05 22:05:38,059 - Epoch: [177][ 570/ 1236] Overall Loss 0.183570 Objective Loss 0.183570 LR 0.000250 Time 0.021443 +2023-10-05 22:05:38,262 - Epoch: [177][ 580/ 1236] Overall Loss 0.183383 Objective Loss 0.183383 LR 0.000250 Time 0.021423 +2023-10-05 22:05:38,462 - Epoch: [177][ 590/ 1236] Overall Loss 0.182993 Objective Loss 0.182993 LR 0.000250 Time 0.021399 +2023-10-05 22:05:38,665 - Epoch: [177][ 600/ 1236] Overall Loss 0.183020 Objective Loss 0.183020 LR 0.000250 Time 0.021380 +2023-10-05 22:05:38,865 - Epoch: [177][ 610/ 1236] Overall Loss 0.183092 Objective Loss 0.183092 LR 0.000250 Time 0.021357 +2023-10-05 22:05:39,068 - Epoch: [177][ 620/ 1236] Overall Loss 0.183077 Objective Loss 0.183077 LR 0.000250 Time 0.021339 +2023-10-05 22:05:39,268 - Epoch: [177][ 630/ 1236] Overall Loss 0.183184 Objective Loss 0.183184 LR 0.000250 Time 0.021318 +2023-10-05 22:05:39,471 - Epoch: [177][ 640/ 1236] Overall Loss 0.183366 Objective Loss 0.183366 LR 0.000250 Time 0.021301 +2023-10-05 22:05:39,672 - Epoch: [177][ 650/ 1236] Overall Loss 0.183593 Objective Loss 0.183593 LR 0.000250 Time 0.021282 +2023-10-05 22:05:39,874 - Epoch: [177][ 660/ 1236] Overall Loss 0.183850 Objective Loss 0.183850 LR 0.000250 Time 0.021265 +2023-10-05 22:05:40,075 - Epoch: [177][ 670/ 1236] Overall Loss 0.183871 Objective Loss 0.183871 LR 0.000250 Time 0.021247 +2023-10-05 22:05:40,277 - Epoch: [177][ 680/ 1236] Overall Loss 0.183869 Objective Loss 0.183869 LR 0.000250 Time 0.021232 +2023-10-05 22:05:40,478 - Epoch: [177][ 690/ 1236] Overall Loss 0.183962 Objective Loss 0.183962 LR 0.000250 Time 0.021214 +2023-10-05 22:05:40,681 - Epoch: [177][ 700/ 1236] Overall Loss 0.183986 Objective Loss 0.183986 LR 0.000250 Time 0.021200 +2023-10-05 22:05:40,881 - Epoch: [177][ 710/ 1236] Overall Loss 0.183855 Objective Loss 0.183855 LR 0.000250 Time 0.021183 +2023-10-05 22:05:41,083 - Epoch: [177][ 720/ 1236] Overall Loss 0.183862 Objective Loss 0.183862 LR 0.000250 Time 0.021170 +2023-10-05 22:05:41,283 - Epoch: [177][ 730/ 1236] Overall Loss 0.183934 Objective Loss 0.183934 LR 0.000250 Time 0.021153 +2023-10-05 22:05:41,486 - Epoch: [177][ 740/ 1236] Overall Loss 0.183900 Objective Loss 0.183900 LR 0.000250 Time 0.021141 +2023-10-05 22:05:41,687 - Epoch: [177][ 750/ 1236] Overall Loss 0.184109 Objective Loss 0.184109 LR 0.000250 Time 0.021126 +2023-10-05 22:05:41,890 - Epoch: [177][ 760/ 1236] Overall Loss 0.184128 Objective Loss 0.184128 LR 0.000250 Time 0.021114 +2023-10-05 22:05:42,090 - Epoch: [177][ 770/ 1236] Overall Loss 0.184529 Objective Loss 0.184529 LR 0.000250 Time 0.021100 +2023-10-05 22:05:42,293 - Epoch: [177][ 780/ 1236] Overall Loss 0.184508 Objective Loss 0.184508 LR 0.000250 Time 0.021089 +2023-10-05 22:05:42,493 - Epoch: [177][ 790/ 1236] Overall Loss 0.184538 Objective Loss 0.184538 LR 0.000250 Time 0.021075 +2023-10-05 22:05:42,696 - Epoch: [177][ 800/ 1236] Overall Loss 0.184722 Objective Loss 0.184722 LR 0.000250 Time 0.021065 +2023-10-05 22:05:42,897 - Epoch: [177][ 810/ 1236] Overall Loss 0.184855 Objective Loss 0.184855 LR 0.000250 Time 0.021052 +2023-10-05 22:05:43,099 - Epoch: [177][ 820/ 1236] Overall Loss 0.184834 Objective Loss 0.184834 LR 0.000250 Time 0.021042 +2023-10-05 22:05:43,300 - Epoch: [177][ 830/ 1236] Overall Loss 0.184826 Objective Loss 0.184826 LR 0.000250 Time 0.021030 +2023-10-05 22:05:43,502 - Epoch: [177][ 840/ 1236] Overall Loss 0.184942 Objective Loss 0.184942 LR 0.000250 Time 0.021019 +2023-10-05 22:05:43,703 - Epoch: [177][ 850/ 1236] Overall Loss 0.185208 Objective Loss 0.185208 LR 0.000250 Time 0.021008 +2023-10-05 22:05:43,905 - Epoch: [177][ 860/ 1236] Overall Loss 0.185014 Objective Loss 0.185014 LR 0.000250 Time 0.020999 +2023-10-05 22:05:44,106 - Epoch: [177][ 870/ 1236] Overall Loss 0.185104 Objective Loss 0.185104 LR 0.000250 Time 0.020988 +2023-10-05 22:05:44,308 - Epoch: [177][ 880/ 1236] Overall Loss 0.185281 Objective Loss 0.185281 LR 0.000250 Time 0.020979 +2023-10-05 22:05:44,508 - Epoch: [177][ 890/ 1236] Overall Loss 0.185338 Objective Loss 0.185338 LR 0.000250 Time 0.020967 +2023-10-05 22:05:44,711 - Epoch: [177][ 900/ 1236] Overall Loss 0.185437 Objective Loss 0.185437 LR 0.000250 Time 0.020960 +2023-10-05 22:05:44,911 - Epoch: [177][ 910/ 1236] Overall Loss 0.185284 Objective Loss 0.185284 LR 0.000250 Time 0.020949 +2023-10-05 22:05:45,114 - Epoch: [177][ 920/ 1236] Overall Loss 0.185181 Objective Loss 0.185181 LR 0.000250 Time 0.020941 +2023-10-05 22:05:45,315 - Epoch: [177][ 930/ 1236] Overall Loss 0.185280 Objective Loss 0.185280 LR 0.000250 Time 0.020931 +2023-10-05 22:05:45,517 - Epoch: [177][ 940/ 1236] Overall Loss 0.185138 Objective Loss 0.185138 LR 0.000250 Time 0.020924 +2023-10-05 22:05:45,718 - Epoch: [177][ 950/ 1236] Overall Loss 0.184962 Objective Loss 0.184962 LR 0.000250 Time 0.020914 +2023-10-05 22:05:45,920 - Epoch: [177][ 960/ 1236] Overall Loss 0.184938 Objective Loss 0.184938 LR 0.000250 Time 0.020907 +2023-10-05 22:05:46,120 - Epoch: [177][ 970/ 1236] Overall Loss 0.184760 Objective Loss 0.184760 LR 0.000250 Time 0.020897 +2023-10-05 22:05:46,323 - Epoch: [177][ 980/ 1236] Overall Loss 0.184696 Objective Loss 0.184696 LR 0.000250 Time 0.020890 +2023-10-05 22:05:46,523 - Epoch: [177][ 990/ 1236] Overall Loss 0.184822 Objective Loss 0.184822 LR 0.000250 Time 0.020881 +2023-10-05 22:05:46,726 - Epoch: [177][ 1000/ 1236] Overall Loss 0.184915 Objective Loss 0.184915 LR 0.000250 Time 0.020874 +2023-10-05 22:05:46,926 - Epoch: [177][ 1010/ 1236] Overall Loss 0.185045 Objective Loss 0.185045 LR 0.000250 Time 0.020866 +2023-10-05 22:05:47,129 - Epoch: [177][ 1020/ 1236] Overall Loss 0.185202 Objective Loss 0.185202 LR 0.000250 Time 0.020860 +2023-10-05 22:05:47,329 - Epoch: [177][ 1030/ 1236] Overall Loss 0.185194 Objective Loss 0.185194 LR 0.000250 Time 0.020851 +2023-10-05 22:05:47,532 - Epoch: [177][ 1040/ 1236] Overall Loss 0.185141 Objective Loss 0.185141 LR 0.000250 Time 0.020845 +2023-10-05 22:05:47,732 - Epoch: [177][ 1050/ 1236] Overall Loss 0.184982 Objective Loss 0.184982 LR 0.000250 Time 0.020837 +2023-10-05 22:05:47,935 - Epoch: [177][ 1060/ 1236] Overall Loss 0.184964 Objective Loss 0.184964 LR 0.000250 Time 0.020832 +2023-10-05 22:05:48,135 - Epoch: [177][ 1070/ 1236] Overall Loss 0.185148 Objective Loss 0.185148 LR 0.000250 Time 0.020823 +2023-10-05 22:05:48,338 - Epoch: [177][ 1080/ 1236] Overall Loss 0.185064 Objective Loss 0.185064 LR 0.000250 Time 0.020818 +2023-10-05 22:05:48,538 - Epoch: [177][ 1090/ 1236] Overall Loss 0.185093 Objective Loss 0.185093 LR 0.000250 Time 0.020810 +2023-10-05 22:05:48,741 - Epoch: [177][ 1100/ 1236] Overall Loss 0.185097 Objective Loss 0.185097 LR 0.000250 Time 0.020805 +2023-10-05 22:05:48,941 - Epoch: [177][ 1110/ 1236] Overall Loss 0.185074 Objective Loss 0.185074 LR 0.000250 Time 0.020798 +2023-10-05 22:05:49,144 - Epoch: [177][ 1120/ 1236] Overall Loss 0.184965 Objective Loss 0.184965 LR 0.000250 Time 0.020793 +2023-10-05 22:05:49,344 - Epoch: [177][ 1130/ 1236] Overall Loss 0.185052 Objective Loss 0.185052 LR 0.000250 Time 0.020786 +2023-10-05 22:05:49,546 - Epoch: [177][ 1140/ 1236] Overall Loss 0.184961 Objective Loss 0.184961 LR 0.000250 Time 0.020781 +2023-10-05 22:05:49,747 - Epoch: [177][ 1150/ 1236] Overall Loss 0.184958 Objective Loss 0.184958 LR 0.000250 Time 0.020774 +2023-10-05 22:05:49,950 - Epoch: [177][ 1160/ 1236] Overall Loss 0.185062 Objective Loss 0.185062 LR 0.000250 Time 0.020769 +2023-10-05 22:05:50,149 - Epoch: [177][ 1170/ 1236] Overall Loss 0.185082 Objective Loss 0.185082 LR 0.000250 Time 0.020762 +2023-10-05 22:05:50,353 - Epoch: [177][ 1180/ 1236] Overall Loss 0.185157 Objective Loss 0.185157 LR 0.000250 Time 0.020758 +2023-10-05 22:05:50,553 - Epoch: [177][ 1190/ 1236] Overall Loss 0.185306 Objective Loss 0.185306 LR 0.000250 Time 0.020751 +2023-10-05 22:05:50,756 - Epoch: [177][ 1200/ 1236] Overall Loss 0.185344 Objective Loss 0.185344 LR 0.000250 Time 0.020748 +2023-10-05 22:05:50,955 - Epoch: [177][ 1210/ 1236] Overall Loss 0.185392 Objective Loss 0.185392 LR 0.000250 Time 0.020740 +2023-10-05 22:05:51,155 - Epoch: [177][ 1220/ 1236] Overall Loss 0.185469 Objective Loss 0.185469 LR 0.000250 Time 0.020734 +2023-10-05 22:05:51,405 - Epoch: [177][ 1230/ 1236] Overall Loss 0.185464 Objective Loss 0.185464 LR 0.000250 Time 0.020769 +2023-10-05 22:05:51,522 - Epoch: [177][ 1236/ 1236] Overall Loss 0.185441 Objective Loss 0.185441 Top1 86.761711 Top5 98.167006 LR 0.000250 Time 0.020762 +2023-10-05 22:05:51,652 - --- validate (epoch=177)----------- +2023-10-05 22:05:51,653 - 29943 samples (256 per mini-batch) +2023-10-05 22:05:52,101 - Epoch: [177][ 10/ 117] Loss 0.319312 Top1 85.664062 Top5 98.125000 +2023-10-05 22:05:52,246 - Epoch: [177][ 20/ 117] Loss 0.295203 Top1 85.566406 Top5 98.007812 +2023-10-05 22:05:52,393 - Epoch: [177][ 30/ 117] Loss 0.305787 Top1 85.546875 Top5 98.007812 +2023-10-05 22:05:52,539 - Epoch: [177][ 40/ 117] Loss 0.308436 Top1 85.429688 Top5 98.056641 +2023-10-05 22:05:52,686 - Epoch: [177][ 50/ 117] Loss 0.306151 Top1 85.546875 Top5 98.093750 +2023-10-05 22:05:52,833 - Epoch: [177][ 60/ 117] Loss 0.311047 Top1 85.494792 Top5 98.066406 +2023-10-05 22:05:52,980 - Epoch: [177][ 70/ 117] Loss 0.311285 Top1 85.373884 Top5 98.041295 +2023-10-05 22:05:53,126 - Epoch: [177][ 80/ 117] Loss 0.310124 Top1 85.268555 Top5 98.085938 +2023-10-05 22:05:53,273 - Epoch: [177][ 90/ 117] Loss 0.312110 Top1 85.208333 Top5 98.059896 +2023-10-05 22:05:53,419 - Epoch: [177][ 100/ 117] Loss 0.307949 Top1 85.343750 Top5 98.085938 +2023-10-05 22:05:53,573 - Epoch: [177][ 110/ 117] Loss 0.310099 Top1 85.319602 Top5 98.096591 +2023-10-05 22:05:53,658 - Epoch: [177][ 117/ 117] Loss 0.310364 Top1 85.188525 Top5 98.083024 +2023-10-05 22:05:53,804 - ==> Top1: 85.189 Top5: 98.083 Loss: 0.310 + +2023-10-05 22:05:53,805 - ==> Confusion: +[[ 949 2 3 1 5 1 0 0 5 53 1 0 2 2 5 2 2 1 1 0 15] + [ 1 1059 2 0 6 14 1 22 3 0 2 2 0 0 1 3 2 0 6 2 5] + [ 5 2 957 10 2 1 26 10 0 1 8 3 6 1 0 2 2 3 7 4 6] + [ 1 1 5 968 1 1 1 1 2 2 5 0 5 3 22 3 0 5 38 2 23] + [ 25 7 0 0 985 2 0 1 0 7 0 2 1 1 6 1 7 2 0 2 1] + [ 4 44 1 0 2 980 0 26 1 1 8 9 1 11 4 2 2 0 3 5 12] + [ 0 5 16 0 0 1 1132 9 0 0 2 3 1 1 1 7 0 0 2 5 6] + [ 3 18 11 0 0 23 4 1079 1 2 3 11 1 1 1 1 0 1 49 4 5] + [ 18 2 0 0 3 2 1 0 992 32 8 2 3 5 10 2 1 0 6 0 2] + [ 112 0 3 0 8 3 0 0 23 930 1 2 0 17 2 8 0 1 0 0 9] + [ 2 9 8 2 1 2 4 6 10 3 968 0 2 9 5 1 2 0 9 2 8] + [ 0 0 1 0 2 12 0 2 0 1 0 962 19 2 0 3 0 16 0 10 5] + [ 1 1 0 3 0 3 0 1 0 0 2 38 976 2 4 3 2 14 3 6 9] + [ 1 0 1 0 1 6 0 1 15 15 7 6 3 1045 4 2 1 1 0 2 8] + [ 12 1 3 9 7 0 0 0 35 3 0 2 2 1 997 0 0 2 15 0 12] + [ 2 4 2 0 2 0 0 0 0 0 0 7 7 1 1 1074 14 8 0 9 3] + [ 0 13 1 0 6 1 0 2 2 0 0 5 0 0 3 12 1101 0 0 3 12] + [ 0 0 0 0 0 0 4 0 0 0 0 4 18 1 0 4 0 1000 2 0 5] + [ 3 9 3 15 1 0 1 22 2 0 1 1 1 1 6 0 0 0 990 1 11] + [ 0 3 6 3 2 7 9 12 2 0 2 13 3 2 0 6 12 2 5 1055 8] + [ 123 196 136 44 85 112 51 98 117 64 174 102 293 271 119 52 138 65 169 187 5309]] + +2023-10-05 22:05:53,807 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:05:53,807 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:05:53,812 - + +2023-10-05 22:05:53,812 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:05:54,788 - Epoch: [178][ 10/ 1236] Overall Loss 0.191201 Objective Loss 0.191201 LR 0.000250 Time 0.097474 +2023-10-05 22:05:54,990 - Epoch: [178][ 20/ 1236] Overall Loss 0.193059 Objective Loss 0.193059 LR 0.000250 Time 0.058847 +2023-10-05 22:05:55,190 - Epoch: [178][ 30/ 1236] Overall Loss 0.180292 Objective Loss 0.180292 LR 0.000250 Time 0.045871 +2023-10-05 22:05:55,391 - Epoch: [178][ 40/ 1236] Overall Loss 0.185010 Objective Loss 0.185010 LR 0.000250 Time 0.039421 +2023-10-05 22:05:55,590 - Epoch: [178][ 50/ 1236] Overall Loss 0.185318 Objective Loss 0.185318 LR 0.000250 Time 0.035517 +2023-10-05 22:05:55,793 - Epoch: [178][ 60/ 1236] Overall Loss 0.182622 Objective Loss 0.182622 LR 0.000250 Time 0.032965 +2023-10-05 22:05:55,993 - Epoch: [178][ 70/ 1236] Overall Loss 0.185791 Objective Loss 0.185791 LR 0.000250 Time 0.031112 +2023-10-05 22:05:56,195 - Epoch: [178][ 80/ 1236] Overall Loss 0.187299 Objective Loss 0.187299 LR 0.000250 Time 0.029743 +2023-10-05 22:05:56,395 - Epoch: [178][ 90/ 1236] Overall Loss 0.185995 Objective Loss 0.185995 LR 0.000250 Time 0.028658 +2023-10-05 22:05:56,597 - Epoch: [178][ 100/ 1236] Overall Loss 0.183383 Objective Loss 0.183383 LR 0.000250 Time 0.027809 +2023-10-05 22:05:56,797 - Epoch: [178][ 110/ 1236] Overall Loss 0.181740 Objective Loss 0.181740 LR 0.000250 Time 0.027097 +2023-10-05 22:05:56,999 - Epoch: [178][ 120/ 1236] Overall Loss 0.184016 Objective Loss 0.184016 LR 0.000250 Time 0.026518 +2023-10-05 22:05:57,199 - Epoch: [178][ 130/ 1236] Overall Loss 0.184393 Objective Loss 0.184393 LR 0.000250 Time 0.026016 +2023-10-05 22:05:57,401 - Epoch: [178][ 140/ 1236] Overall Loss 0.185299 Objective Loss 0.185299 LR 0.000250 Time 0.025597 +2023-10-05 22:05:57,601 - Epoch: [178][ 150/ 1236] Overall Loss 0.185453 Objective Loss 0.185453 LR 0.000250 Time 0.025222 +2023-10-05 22:05:57,803 - Epoch: [178][ 160/ 1236] Overall Loss 0.186333 Objective Loss 0.186333 LR 0.000250 Time 0.024905 +2023-10-05 22:05:58,003 - Epoch: [178][ 170/ 1236] Overall Loss 0.186168 Objective Loss 0.186168 LR 0.000250 Time 0.024616 +2023-10-05 22:05:58,206 - Epoch: [178][ 180/ 1236] Overall Loss 0.185969 Objective Loss 0.185969 LR 0.000250 Time 0.024370 +2023-10-05 22:05:58,406 - Epoch: [178][ 190/ 1236] Overall Loss 0.185571 Objective Loss 0.185571 LR 0.000250 Time 0.024139 +2023-10-05 22:05:58,608 - Epoch: [178][ 200/ 1236] Overall Loss 0.186600 Objective Loss 0.186600 LR 0.000250 Time 0.023942 +2023-10-05 22:05:58,808 - Epoch: [178][ 210/ 1236] Overall Loss 0.186820 Objective Loss 0.186820 LR 0.000250 Time 0.023753 +2023-10-05 22:05:59,011 - Epoch: [178][ 220/ 1236] Overall Loss 0.186876 Objective Loss 0.186876 LR 0.000250 Time 0.023592 +2023-10-05 22:05:59,211 - Epoch: [178][ 230/ 1236] Overall Loss 0.185932 Objective Loss 0.185932 LR 0.000250 Time 0.023435 +2023-10-05 22:05:59,413 - Epoch: [178][ 240/ 1236] Overall Loss 0.185692 Objective Loss 0.185692 LR 0.000250 Time 0.023300 +2023-10-05 22:05:59,613 - Epoch: [178][ 250/ 1236] Overall Loss 0.185847 Objective Loss 0.185847 LR 0.000250 Time 0.023167 +2023-10-05 22:05:59,815 - Epoch: [178][ 260/ 1236] Overall Loss 0.185833 Objective Loss 0.185833 LR 0.000250 Time 0.023051 +2023-10-05 22:06:00,015 - Epoch: [178][ 270/ 1236] Overall Loss 0.186236 Objective Loss 0.186236 LR 0.000250 Time 0.022936 +2023-10-05 22:06:00,218 - Epoch: [178][ 280/ 1236] Overall Loss 0.188031 Objective Loss 0.188031 LR 0.000250 Time 0.022839 +2023-10-05 22:06:00,418 - Epoch: [178][ 290/ 1236] Overall Loss 0.187561 Objective Loss 0.187561 LR 0.000250 Time 0.022741 +2023-10-05 22:06:00,620 - Epoch: [178][ 300/ 1236] Overall Loss 0.187123 Objective Loss 0.187123 LR 0.000250 Time 0.022656 +2023-10-05 22:06:00,821 - Epoch: [178][ 310/ 1236] Overall Loss 0.187365 Objective Loss 0.187365 LR 0.000250 Time 0.022571 +2023-10-05 22:06:01,023 - Epoch: [178][ 320/ 1236] Overall Loss 0.187233 Objective Loss 0.187233 LR 0.000250 Time 0.022497 +2023-10-05 22:06:01,223 - Epoch: [178][ 330/ 1236] Overall Loss 0.187152 Objective Loss 0.187152 LR 0.000250 Time 0.022421 +2023-10-05 22:06:01,426 - Epoch: [178][ 340/ 1236] Overall Loss 0.187226 Objective Loss 0.187226 LR 0.000250 Time 0.022356 +2023-10-05 22:06:01,626 - Epoch: [178][ 350/ 1236] Overall Loss 0.186937 Objective Loss 0.186937 LR 0.000250 Time 0.022287 +2023-10-05 22:06:01,828 - Epoch: [178][ 360/ 1236] Overall Loss 0.187102 Objective Loss 0.187102 LR 0.000250 Time 0.022230 +2023-10-05 22:06:02,029 - Epoch: [178][ 370/ 1236] Overall Loss 0.186534 Objective Loss 0.186534 LR 0.000250 Time 0.022170 +2023-10-05 22:06:02,231 - Epoch: [178][ 380/ 1236] Overall Loss 0.186661 Objective Loss 0.186661 LR 0.000250 Time 0.022118 +2023-10-05 22:06:02,432 - Epoch: [178][ 390/ 1236] Overall Loss 0.186384 Objective Loss 0.186384 LR 0.000250 Time 0.022064 +2023-10-05 22:06:02,634 - Epoch: [178][ 400/ 1236] Overall Loss 0.185963 Objective Loss 0.185963 LR 0.000250 Time 0.022017 +2023-10-05 22:06:02,834 - Epoch: [178][ 410/ 1236] Overall Loss 0.185489 Objective Loss 0.185489 LR 0.000250 Time 0.021968 +2023-10-05 22:06:03,036 - Epoch: [178][ 420/ 1236] Overall Loss 0.185817 Objective Loss 0.185817 LR 0.000250 Time 0.021925 +2023-10-05 22:06:03,236 - Epoch: [178][ 430/ 1236] Overall Loss 0.186257 Objective Loss 0.186257 LR 0.000250 Time 0.021879 +2023-10-05 22:06:03,439 - Epoch: [178][ 440/ 1236] Overall Loss 0.186453 Objective Loss 0.186453 LR 0.000250 Time 0.021841 +2023-10-05 22:06:03,639 - Epoch: [178][ 450/ 1236] Overall Loss 0.186650 Objective Loss 0.186650 LR 0.000250 Time 0.021801 +2023-10-05 22:06:03,841 - Epoch: [178][ 460/ 1236] Overall Loss 0.186375 Objective Loss 0.186375 LR 0.000250 Time 0.021766 +2023-10-05 22:06:04,040 - Epoch: [178][ 470/ 1236] Overall Loss 0.186337 Objective Loss 0.186337 LR 0.000250 Time 0.021725 +2023-10-05 22:06:04,239 - Epoch: [178][ 480/ 1236] Overall Loss 0.186652 Objective Loss 0.186652 LR 0.000250 Time 0.021686 +2023-10-05 22:06:04,436 - Epoch: [178][ 490/ 1236] Overall Loss 0.186526 Objective Loss 0.186526 LR 0.000250 Time 0.021645 +2023-10-05 22:06:04,635 - Epoch: [178][ 500/ 1236] Overall Loss 0.186760 Objective Loss 0.186760 LR 0.000250 Time 0.021609 +2023-10-05 22:06:04,833 - Epoch: [178][ 510/ 1236] Overall Loss 0.186502 Objective Loss 0.186502 LR 0.000250 Time 0.021572 +2023-10-05 22:06:05,032 - Epoch: [178][ 520/ 1236] Overall Loss 0.186540 Objective Loss 0.186540 LR 0.000250 Time 0.021540 +2023-10-05 22:06:05,229 - Epoch: [178][ 530/ 1236] Overall Loss 0.186861 Objective Loss 0.186861 LR 0.000250 Time 0.021505 +2023-10-05 22:06:05,428 - Epoch: [178][ 540/ 1236] Overall Loss 0.186288 Objective Loss 0.186288 LR 0.000250 Time 0.021474 +2023-10-05 22:06:05,625 - Epoch: [178][ 550/ 1236] Overall Loss 0.186174 Objective Loss 0.186174 LR 0.000250 Time 0.021442 +2023-10-05 22:06:05,825 - Epoch: [178][ 560/ 1236] Overall Loss 0.185886 Objective Loss 0.185886 LR 0.000250 Time 0.021414 +2023-10-05 22:06:06,022 - Epoch: [178][ 570/ 1236] Overall Loss 0.186213 Objective Loss 0.186213 LR 0.000250 Time 0.021385 +2023-10-05 22:06:06,221 - Epoch: [178][ 580/ 1236] Overall Loss 0.186603 Objective Loss 0.186603 LR 0.000250 Time 0.021358 +2023-10-05 22:06:06,419 - Epoch: [178][ 590/ 1236] Overall Loss 0.186526 Objective Loss 0.186526 LR 0.000250 Time 0.021330 +2023-10-05 22:06:06,618 - Epoch: [178][ 600/ 1236] Overall Loss 0.186303 Objective Loss 0.186303 LR 0.000250 Time 0.021307 +2023-10-05 22:06:06,816 - Epoch: [178][ 610/ 1236] Overall Loss 0.186493 Objective Loss 0.186493 LR 0.000250 Time 0.021281 +2023-10-05 22:06:07,018 - Epoch: [178][ 620/ 1236] Overall Loss 0.186574 Objective Loss 0.186574 LR 0.000250 Time 0.021263 +2023-10-05 22:06:07,218 - Epoch: [178][ 630/ 1236] Overall Loss 0.186347 Objective Loss 0.186347 LR 0.000250 Time 0.021242 +2023-10-05 22:06:07,421 - Epoch: [178][ 640/ 1236] Overall Loss 0.186720 Objective Loss 0.186720 LR 0.000250 Time 0.021227 +2023-10-05 22:06:07,620 - Epoch: [178][ 650/ 1236] Overall Loss 0.186764 Objective Loss 0.186764 LR 0.000250 Time 0.021207 +2023-10-05 22:06:07,823 - Epoch: [178][ 660/ 1236] Overall Loss 0.186629 Objective Loss 0.186629 LR 0.000250 Time 0.021192 +2023-10-05 22:06:08,023 - Epoch: [178][ 670/ 1236] Overall Loss 0.186758 Objective Loss 0.186758 LR 0.000250 Time 0.021173 +2023-10-05 22:06:08,226 - Epoch: [178][ 680/ 1236] Overall Loss 0.186890 Objective Loss 0.186890 LR 0.000250 Time 0.021159 +2023-10-05 22:06:08,426 - Epoch: [178][ 690/ 1236] Overall Loss 0.186923 Objective Loss 0.186923 LR 0.000250 Time 0.021143 +2023-10-05 22:06:08,628 - Epoch: [178][ 700/ 1236] Overall Loss 0.187090 Objective Loss 0.187090 LR 0.000250 Time 0.021129 +2023-10-05 22:06:08,828 - Epoch: [178][ 710/ 1236] Overall Loss 0.187061 Objective Loss 0.187061 LR 0.000250 Time 0.021113 +2023-10-05 22:06:09,031 - Epoch: [178][ 720/ 1236] Overall Loss 0.187027 Objective Loss 0.187027 LR 0.000250 Time 0.021101 +2023-10-05 22:06:09,231 - Epoch: [178][ 730/ 1236] Overall Loss 0.186750 Objective Loss 0.186750 LR 0.000250 Time 0.021085 +2023-10-05 22:06:09,434 - Epoch: [178][ 740/ 1236] Overall Loss 0.186599 Objective Loss 0.186599 LR 0.000250 Time 0.021074 +2023-10-05 22:06:09,634 - Epoch: [178][ 750/ 1236] Overall Loss 0.186457 Objective Loss 0.186457 LR 0.000250 Time 0.021059 +2023-10-05 22:06:09,837 - Epoch: [178][ 760/ 1236] Overall Loss 0.186640 Objective Loss 0.186640 LR 0.000250 Time 0.021048 +2023-10-05 22:06:10,037 - Epoch: [178][ 770/ 1236] Overall Loss 0.186654 Objective Loss 0.186654 LR 0.000250 Time 0.021034 +2023-10-05 22:06:10,239 - Epoch: [178][ 780/ 1236] Overall Loss 0.186536 Objective Loss 0.186536 LR 0.000250 Time 0.021024 +2023-10-05 22:06:10,439 - Epoch: [178][ 790/ 1236] Overall Loss 0.186785 Objective Loss 0.186785 LR 0.000250 Time 0.021010 +2023-10-05 22:06:10,642 - Epoch: [178][ 800/ 1236] Overall Loss 0.186647 Objective Loss 0.186647 LR 0.000250 Time 0.021001 +2023-10-05 22:06:10,842 - Epoch: [178][ 810/ 1236] Overall Loss 0.186695 Objective Loss 0.186695 LR 0.000250 Time 0.020988 +2023-10-05 22:06:11,045 - Epoch: [178][ 820/ 1236] Overall Loss 0.186819 Objective Loss 0.186819 LR 0.000250 Time 0.020979 +2023-10-05 22:06:11,245 - Epoch: [178][ 830/ 1236] Overall Loss 0.186672 Objective Loss 0.186672 LR 0.000250 Time 0.020967 +2023-10-05 22:06:11,446 - Epoch: [178][ 840/ 1236] Overall Loss 0.186883 Objective Loss 0.186883 LR 0.000250 Time 0.020956 +2023-10-05 22:06:11,647 - Epoch: [178][ 850/ 1236] Overall Loss 0.187090 Objective Loss 0.187090 LR 0.000250 Time 0.020945 +2023-10-05 22:06:11,850 - Epoch: [178][ 860/ 1236] Overall Loss 0.187126 Objective Loss 0.187126 LR 0.000250 Time 0.020937 +2023-10-05 22:06:12,049 - Epoch: [178][ 870/ 1236] Overall Loss 0.187078 Objective Loss 0.187078 LR 0.000250 Time 0.020925 +2023-10-05 22:06:12,252 - Epoch: [178][ 880/ 1236] Overall Loss 0.187038 Objective Loss 0.187038 LR 0.000250 Time 0.020918 +2023-10-05 22:06:12,452 - Epoch: [178][ 890/ 1236] Overall Loss 0.186976 Objective Loss 0.186976 LR 0.000250 Time 0.020907 +2023-10-05 22:06:12,655 - Epoch: [178][ 900/ 1236] Overall Loss 0.186902 Objective Loss 0.186902 LR 0.000250 Time 0.020900 +2023-10-05 22:06:12,855 - Epoch: [178][ 910/ 1236] Overall Loss 0.186735 Objective Loss 0.186735 LR 0.000250 Time 0.020890 +2023-10-05 22:06:13,058 - Epoch: [178][ 920/ 1236] Overall Loss 0.186679 Objective Loss 0.186679 LR 0.000250 Time 0.020883 +2023-10-05 22:06:13,258 - Epoch: [178][ 930/ 1236] Overall Loss 0.186862 Objective Loss 0.186862 LR 0.000250 Time 0.020873 +2023-10-05 22:06:13,461 - Epoch: [178][ 940/ 1236] Overall Loss 0.186707 Objective Loss 0.186707 LR 0.000250 Time 0.020867 +2023-10-05 22:06:13,661 - Epoch: [178][ 950/ 1236] Overall Loss 0.186735 Objective Loss 0.186735 LR 0.000250 Time 0.020857 +2023-10-05 22:06:13,864 - Epoch: [178][ 960/ 1236] Overall Loss 0.186757 Objective Loss 0.186757 LR 0.000250 Time 0.020851 +2023-10-05 22:06:14,064 - Epoch: [178][ 970/ 1236] Overall Loss 0.186889 Objective Loss 0.186889 LR 0.000250 Time 0.020842 +2023-10-05 22:06:14,267 - Epoch: [178][ 980/ 1236] Overall Loss 0.186894 Objective Loss 0.186894 LR 0.000250 Time 0.020836 +2023-10-05 22:06:14,467 - Epoch: [178][ 990/ 1236] Overall Loss 0.187047 Objective Loss 0.187047 LR 0.000250 Time 0.020827 +2023-10-05 22:06:14,669 - Epoch: [178][ 1000/ 1236] Overall Loss 0.186975 Objective Loss 0.186975 LR 0.000250 Time 0.020821 +2023-10-05 22:06:14,870 - Epoch: [178][ 1010/ 1236] Overall Loss 0.186899 Objective Loss 0.186899 LR 0.000250 Time 0.020813 +2023-10-05 22:06:15,073 - Epoch: [178][ 1020/ 1236] Overall Loss 0.186938 Objective Loss 0.186938 LR 0.000250 Time 0.020807 +2023-10-05 22:06:15,272 - Epoch: [178][ 1030/ 1236] Overall Loss 0.186786 Objective Loss 0.186786 LR 0.000250 Time 0.020799 +2023-10-05 22:06:15,475 - Epoch: [178][ 1040/ 1236] Overall Loss 0.186768 Objective Loss 0.186768 LR 0.000250 Time 0.020793 +2023-10-05 22:06:15,675 - Epoch: [178][ 1050/ 1236] Overall Loss 0.186756 Objective Loss 0.186756 LR 0.000250 Time 0.020785 +2023-10-05 22:06:15,879 - Epoch: [178][ 1060/ 1236] Overall Loss 0.186838 Objective Loss 0.186838 LR 0.000250 Time 0.020781 +2023-10-05 22:06:16,078 - Epoch: [178][ 1070/ 1236] Overall Loss 0.186987 Objective Loss 0.186987 LR 0.000250 Time 0.020773 +2023-10-05 22:06:16,281 - Epoch: [178][ 1080/ 1236] Overall Loss 0.186946 Objective Loss 0.186946 LR 0.000250 Time 0.020768 +2023-10-05 22:06:16,481 - Epoch: [178][ 1090/ 1236] Overall Loss 0.187006 Objective Loss 0.187006 LR 0.000250 Time 0.020761 +2023-10-05 22:06:16,684 - Epoch: [178][ 1100/ 1236] Overall Loss 0.187006 Objective Loss 0.187006 LR 0.000250 Time 0.020756 +2023-10-05 22:06:16,884 - Epoch: [178][ 1110/ 1236] Overall Loss 0.186982 Objective Loss 0.186982 LR 0.000250 Time 0.020749 +2023-10-05 22:06:17,087 - Epoch: [178][ 1120/ 1236] Overall Loss 0.186927 Objective Loss 0.186927 LR 0.000250 Time 0.020745 +2023-10-05 22:06:17,287 - Epoch: [178][ 1130/ 1236] Overall Loss 0.187018 Objective Loss 0.187018 LR 0.000250 Time 0.020737 +2023-10-05 22:06:17,490 - Epoch: [178][ 1140/ 1236] Overall Loss 0.187000 Objective Loss 0.187000 LR 0.000250 Time 0.020734 +2023-10-05 22:06:17,690 - Epoch: [178][ 1150/ 1236] Overall Loss 0.186886 Objective Loss 0.186886 LR 0.000250 Time 0.020727 +2023-10-05 22:06:17,894 - Epoch: [178][ 1160/ 1236] Overall Loss 0.186961 Objective Loss 0.186961 LR 0.000250 Time 0.020723 +2023-10-05 22:06:18,094 - Epoch: [178][ 1170/ 1236] Overall Loss 0.187133 Objective Loss 0.187133 LR 0.000250 Time 0.020717 +2023-10-05 22:06:18,297 - Epoch: [178][ 1180/ 1236] Overall Loss 0.187127 Objective Loss 0.187127 LR 0.000250 Time 0.020713 +2023-10-05 22:06:18,497 - Epoch: [178][ 1190/ 1236] Overall Loss 0.186851 Objective Loss 0.186851 LR 0.000250 Time 0.020707 +2023-10-05 22:06:18,700 - Epoch: [178][ 1200/ 1236] Overall Loss 0.186883 Objective Loss 0.186883 LR 0.000250 Time 0.020703 +2023-10-05 22:06:18,900 - Epoch: [178][ 1210/ 1236] Overall Loss 0.187012 Objective Loss 0.187012 LR 0.000250 Time 0.020697 +2023-10-05 22:06:19,103 - Epoch: [178][ 1220/ 1236] Overall Loss 0.187059 Objective Loss 0.187059 LR 0.000250 Time 0.020694 +2023-10-05 22:06:19,354 - Epoch: [178][ 1230/ 1236] Overall Loss 0.187193 Objective Loss 0.187193 LR 0.000250 Time 0.020729 +2023-10-05 22:06:19,471 - Epoch: [178][ 1236/ 1236] Overall Loss 0.187079 Objective Loss 0.187079 Top1 90.224033 Top5 98.778004 LR 0.000250 Time 0.020723 +2023-10-05 22:06:19,604 - --- validate (epoch=178)----------- +2023-10-05 22:06:19,605 - 29943 samples (256 per mini-batch) +2023-10-05 22:06:20,051 - Epoch: [178][ 10/ 117] Loss 0.309823 Top1 84.570312 Top5 98.437500 +2023-10-05 22:06:20,199 - Epoch: [178][ 20/ 117] Loss 0.308004 Top1 84.492188 Top5 98.281250 +2023-10-05 22:06:20,349 - Epoch: [178][ 30/ 117] Loss 0.316103 Top1 84.505208 Top5 98.085938 +2023-10-05 22:06:20,498 - Epoch: [178][ 40/ 117] Loss 0.320467 Top1 84.570312 Top5 98.007812 +2023-10-05 22:06:20,647 - Epoch: [178][ 50/ 117] Loss 0.315758 Top1 84.820312 Top5 98.054688 +2023-10-05 22:06:20,796 - Epoch: [178][ 60/ 117] Loss 0.319426 Top1 84.928385 Top5 98.111979 +2023-10-05 22:06:20,945 - Epoch: [178][ 70/ 117] Loss 0.318474 Top1 84.994420 Top5 98.141741 +2023-10-05 22:06:21,094 - Epoch: [178][ 80/ 117] Loss 0.319060 Top1 85.092773 Top5 98.168945 +2023-10-05 22:06:21,243 - Epoch: [178][ 90/ 117] Loss 0.317751 Top1 85.151910 Top5 98.133681 +2023-10-05 22:06:21,392 - Epoch: [178][ 100/ 117] Loss 0.313401 Top1 85.226562 Top5 98.148438 +2023-10-05 22:06:21,548 - Epoch: [178][ 110/ 117] Loss 0.314328 Top1 85.195312 Top5 98.117898 +2023-10-05 22:06:21,632 - Epoch: [178][ 117/ 117] Loss 0.312490 Top1 85.188525 Top5 98.116421 +2023-10-05 22:06:21,780 - ==> Top1: 85.189 Top5: 98.116 Loss: 0.312 + +2023-10-05 22:06:21,781 - ==> Confusion: +[[ 942 2 3 4 2 1 0 0 8 57 1 0 1 2 4 2 4 2 0 1 14] + [ 0 1072 2 0 3 17 1 14 2 0 0 1 0 0 1 4 3 0 6 0 5] + [ 3 1 959 13 3 0 31 10 0 0 6 2 9 1 1 3 1 1 7 1 4] + [ 0 1 9 974 1 1 1 3 2 1 3 0 4 2 26 4 0 2 36 3 16] + [ 28 9 1 0 965 2 2 1 0 10 1 1 0 1 9 2 12 1 0 1 4] + [ 3 46 1 3 3 982 0 21 4 0 7 4 0 8 7 2 5 0 3 3 14] + [ 0 7 12 0 0 1 1134 9 0 0 4 2 2 0 1 5 0 0 3 7 4] + [ 2 20 9 0 2 23 4 1076 3 5 4 10 0 2 0 3 0 0 41 7 7] + [ 17 2 0 0 3 0 1 0 995 31 10 3 3 5 8 0 0 0 4 2 5] + [ 111 1 3 1 4 2 0 0 27 933 0 2 0 15 4 4 0 1 0 1 10] + [ 1 8 7 4 1 2 3 4 13 2 979 2 0 8 4 2 2 0 4 2 5] + [ 0 0 1 0 1 18 0 2 0 1 0 959 17 3 0 4 1 15 0 5 8] + [ 1 1 4 8 0 2 0 1 0 0 1 30 975 2 1 7 2 15 4 4 10] + [ 3 0 1 0 3 6 0 0 19 12 13 6 3 1036 3 2 1 0 0 0 11] + [ 10 4 4 5 3 1 0 0 34 1 4 1 2 1 1008 0 0 2 12 0 9] + [ 1 4 2 0 2 0 1 0 0 0 0 4 6 1 1 1074 16 11 0 8 3] + [ 3 15 0 0 4 1 0 0 2 0 0 4 0 0 3 5 1107 0 1 3 13] + [ 0 0 0 1 0 0 3 0 1 0 0 1 11 3 1 7 1 1007 2 0 0] + [ 2 8 3 18 1 0 0 22 1 0 3 0 3 1 8 0 0 0 989 1 8] + [ 0 6 2 2 2 8 13 9 0 0 3 14 4 0 0 6 11 1 5 1060 6] + [ 130 209 129 58 70 121 47 70 123 63 204 104 299 241 147 55 143 68 155 187 5282]] + +2023-10-05 22:06:21,782 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:06:21,782 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:06:21,788 - + +2023-10-05 22:06:21,788 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:06:22,756 - Epoch: [179][ 10/ 1236] Overall Loss 0.180197 Objective Loss 0.180197 LR 0.000250 Time 0.096710 +2023-10-05 22:06:22,958 - Epoch: [179][ 20/ 1236] Overall Loss 0.179901 Objective Loss 0.179901 LR 0.000250 Time 0.058438 +2023-10-05 22:06:23,157 - Epoch: [179][ 30/ 1236] Overall Loss 0.184282 Objective Loss 0.184282 LR 0.000250 Time 0.045582 +2023-10-05 22:06:23,359 - Epoch: [179][ 40/ 1236] Overall Loss 0.181392 Objective Loss 0.181392 LR 0.000250 Time 0.039230 +2023-10-05 22:06:23,558 - Epoch: [179][ 50/ 1236] Overall Loss 0.185336 Objective Loss 0.185336 LR 0.000250 Time 0.035363 +2023-10-05 22:06:23,760 - Epoch: [179][ 60/ 1236] Overall Loss 0.185819 Objective Loss 0.185819 LR 0.000250 Time 0.032831 +2023-10-05 22:06:23,960 - Epoch: [179][ 70/ 1236] Overall Loss 0.187594 Objective Loss 0.187594 LR 0.000250 Time 0.030986 +2023-10-05 22:06:24,162 - Epoch: [179][ 80/ 1236] Overall Loss 0.183777 Objective Loss 0.183777 LR 0.000250 Time 0.029634 +2023-10-05 22:06:24,361 - Epoch: [179][ 90/ 1236] Overall Loss 0.182302 Objective Loss 0.182302 LR 0.000250 Time 0.028550 +2023-10-05 22:06:24,563 - Epoch: [179][ 100/ 1236] Overall Loss 0.182142 Objective Loss 0.182142 LR 0.000250 Time 0.027711 +2023-10-05 22:06:24,763 - Epoch: [179][ 110/ 1236] Overall Loss 0.181178 Objective Loss 0.181178 LR 0.000250 Time 0.027005 +2023-10-05 22:06:24,961 - Epoch: [179][ 120/ 1236] Overall Loss 0.181990 Objective Loss 0.181990 LR 0.000250 Time 0.026405 +2023-10-05 22:06:25,159 - Epoch: [179][ 130/ 1236] Overall Loss 0.181883 Objective Loss 0.181883 LR 0.000250 Time 0.025890 +2023-10-05 22:06:25,357 - Epoch: [179][ 140/ 1236] Overall Loss 0.180894 Objective Loss 0.180894 LR 0.000250 Time 0.025455 +2023-10-05 22:06:25,556 - Epoch: [179][ 150/ 1236] Overall Loss 0.180498 Objective Loss 0.180498 LR 0.000250 Time 0.025075 +2023-10-05 22:06:25,754 - Epoch: [179][ 160/ 1236] Overall Loss 0.182277 Objective Loss 0.182277 LR 0.000250 Time 0.024746 +2023-10-05 22:06:25,952 - Epoch: [179][ 170/ 1236] Overall Loss 0.182069 Objective Loss 0.182069 LR 0.000250 Time 0.024452 +2023-10-05 22:06:26,150 - Epoch: [179][ 180/ 1236] Overall Loss 0.183780 Objective Loss 0.183780 LR 0.000250 Time 0.024195 +2023-10-05 22:06:26,348 - Epoch: [179][ 190/ 1236] Overall Loss 0.183843 Objective Loss 0.183843 LR 0.000250 Time 0.023959 +2023-10-05 22:06:26,547 - Epoch: [179][ 200/ 1236] Overall Loss 0.184088 Objective Loss 0.184088 LR 0.000250 Time 0.023753 +2023-10-05 22:06:26,744 - Epoch: [179][ 210/ 1236] Overall Loss 0.183795 Objective Loss 0.183795 LR 0.000250 Time 0.023561 +2023-10-05 22:06:26,945 - Epoch: [179][ 220/ 1236] Overall Loss 0.184608 Objective Loss 0.184608 LR 0.000250 Time 0.023401 +2023-10-05 22:06:27,146 - Epoch: [179][ 230/ 1236] Overall Loss 0.183625 Objective Loss 0.183625 LR 0.000250 Time 0.023258 +2023-10-05 22:06:27,346 - Epoch: [179][ 240/ 1236] Overall Loss 0.183435 Objective Loss 0.183435 LR 0.000250 Time 0.023119 +2023-10-05 22:06:27,548 - Epoch: [179][ 250/ 1236] Overall Loss 0.183199 Objective Loss 0.183199 LR 0.000250 Time 0.023000 +2023-10-05 22:06:27,749 - Epoch: [179][ 260/ 1236] Overall Loss 0.182894 Objective Loss 0.182894 LR 0.000250 Time 0.022889 +2023-10-05 22:06:27,952 - Epoch: [179][ 270/ 1236] Overall Loss 0.183240 Objective Loss 0.183240 LR 0.000250 Time 0.022792 +2023-10-05 22:06:28,157 - Epoch: [179][ 280/ 1236] Overall Loss 0.183176 Objective Loss 0.183176 LR 0.000250 Time 0.022706 +2023-10-05 22:06:28,360 - Epoch: [179][ 290/ 1236] Overall Loss 0.184039 Objective Loss 0.184039 LR 0.000250 Time 0.022623 +2023-10-05 22:06:28,564 - Epoch: [179][ 300/ 1236] Overall Loss 0.183811 Objective Loss 0.183811 LR 0.000250 Time 0.022547 +2023-10-05 22:06:28,766 - Epoch: [179][ 310/ 1236] Overall Loss 0.183263 Objective Loss 0.183263 LR 0.000250 Time 0.022472 +2023-10-05 22:06:28,969 - Epoch: [179][ 320/ 1236] Overall Loss 0.183581 Objective Loss 0.183581 LR 0.000250 Time 0.022402 +2023-10-05 22:06:29,170 - Epoch: [179][ 330/ 1236] Overall Loss 0.183890 Objective Loss 0.183890 LR 0.000250 Time 0.022333 +2023-10-05 22:06:29,369 - Epoch: [179][ 340/ 1236] Overall Loss 0.184102 Objective Loss 0.184102 LR 0.000250 Time 0.022260 +2023-10-05 22:06:29,570 - Epoch: [179][ 350/ 1236] Overall Loss 0.183352 Objective Loss 0.183352 LR 0.000250 Time 0.022196 +2023-10-05 22:06:29,770 - Epoch: [179][ 360/ 1236] Overall Loss 0.183250 Objective Loss 0.183250 LR 0.000250 Time 0.022134 +2023-10-05 22:06:29,976 - Epoch: [179][ 370/ 1236] Overall Loss 0.183622 Objective Loss 0.183622 LR 0.000250 Time 0.022092 +2023-10-05 22:06:30,182 - Epoch: [179][ 380/ 1236] Overall Loss 0.184139 Objective Loss 0.184139 LR 0.000250 Time 0.022052 +2023-10-05 22:06:30,388 - Epoch: [179][ 390/ 1236] Overall Loss 0.184159 Objective Loss 0.184159 LR 0.000250 Time 0.022013 +2023-10-05 22:06:30,594 - Epoch: [179][ 400/ 1236] Overall Loss 0.184029 Objective Loss 0.184029 LR 0.000250 Time 0.021978 +2023-10-05 22:06:30,800 - Epoch: [179][ 410/ 1236] Overall Loss 0.184328 Objective Loss 0.184328 LR 0.000250 Time 0.021942 +2023-10-05 22:06:31,006 - Epoch: [179][ 420/ 1236] Overall Loss 0.184365 Objective Loss 0.184365 LR 0.000250 Time 0.021910 +2023-10-05 22:06:31,212 - Epoch: [179][ 430/ 1236] Overall Loss 0.184354 Objective Loss 0.184354 LR 0.000250 Time 0.021878 +2023-10-05 22:06:31,417 - Epoch: [179][ 440/ 1236] Overall Loss 0.184374 Objective Loss 0.184374 LR 0.000250 Time 0.021847 +2023-10-05 22:06:31,623 - Epoch: [179][ 450/ 1236] Overall Loss 0.184478 Objective Loss 0.184478 LR 0.000250 Time 0.021818 +2023-10-05 22:06:31,829 - Epoch: [179][ 460/ 1236] Overall Loss 0.184297 Objective Loss 0.184297 LR 0.000250 Time 0.021791 +2023-10-05 22:06:32,035 - Epoch: [179][ 470/ 1236] Overall Loss 0.184393 Objective Loss 0.184393 LR 0.000250 Time 0.021764 +2023-10-05 22:06:32,241 - Epoch: [179][ 480/ 1236] Overall Loss 0.184649 Objective Loss 0.184649 LR 0.000250 Time 0.021740 +2023-10-05 22:06:32,447 - Epoch: [179][ 490/ 1236] Overall Loss 0.184439 Objective Loss 0.184439 LR 0.000250 Time 0.021715 +2023-10-05 22:06:32,654 - Epoch: [179][ 500/ 1236] Overall Loss 0.184601 Objective Loss 0.184601 LR 0.000250 Time 0.021693 +2023-10-05 22:06:32,859 - Epoch: [179][ 510/ 1236] Overall Loss 0.184175 Objective Loss 0.184175 LR 0.000250 Time 0.021670 +2023-10-05 22:06:33,065 - Epoch: [179][ 520/ 1236] Overall Loss 0.184711 Objective Loss 0.184711 LR 0.000250 Time 0.021649 +2023-10-05 22:06:33,271 - Epoch: [179][ 530/ 1236] Overall Loss 0.184292 Objective Loss 0.184292 LR 0.000250 Time 0.021627 +2023-10-05 22:06:33,477 - Epoch: [179][ 540/ 1236] Overall Loss 0.184124 Objective Loss 0.184124 LR 0.000250 Time 0.021608 +2023-10-05 22:06:33,683 - Epoch: [179][ 550/ 1236] Overall Loss 0.184389 Objective Loss 0.184389 LR 0.000250 Time 0.021589 +2023-10-05 22:06:33,889 - Epoch: [179][ 560/ 1236] Overall Loss 0.184193 Objective Loss 0.184193 LR 0.000250 Time 0.021571 +2023-10-05 22:06:34,095 - Epoch: [179][ 570/ 1236] Overall Loss 0.183457 Objective Loss 0.183457 LR 0.000250 Time 0.021553 +2023-10-05 22:06:34,301 - Epoch: [179][ 580/ 1236] Overall Loss 0.183614 Objective Loss 0.183614 LR 0.000250 Time 0.021536 +2023-10-05 22:06:34,507 - Epoch: [179][ 590/ 1236] Overall Loss 0.183451 Objective Loss 0.183451 LR 0.000250 Time 0.021518 +2023-10-05 22:06:34,713 - Epoch: [179][ 600/ 1236] Overall Loss 0.183559 Objective Loss 0.183559 LR 0.000250 Time 0.021502 +2023-10-05 22:06:34,918 - Epoch: [179][ 610/ 1236] Overall Loss 0.183472 Objective Loss 0.183472 LR 0.000250 Time 0.021486 +2023-10-05 22:06:35,124 - Epoch: [179][ 620/ 1236] Overall Loss 0.183440 Objective Loss 0.183440 LR 0.000250 Time 0.021472 +2023-10-05 22:06:35,330 - Epoch: [179][ 630/ 1236] Overall Loss 0.183367 Objective Loss 0.183367 LR 0.000250 Time 0.021457 +2023-10-05 22:06:35,537 - Epoch: [179][ 640/ 1236] Overall Loss 0.183427 Objective Loss 0.183427 LR 0.000250 Time 0.021444 +2023-10-05 22:06:35,742 - Epoch: [179][ 650/ 1236] Overall Loss 0.183546 Objective Loss 0.183546 LR 0.000250 Time 0.021429 +2023-10-05 22:06:35,949 - Epoch: [179][ 660/ 1236] Overall Loss 0.183577 Objective Loss 0.183577 LR 0.000250 Time 0.021417 +2023-10-05 22:06:36,154 - Epoch: [179][ 670/ 1236] Overall Loss 0.183612 Objective Loss 0.183612 LR 0.000250 Time 0.021403 +2023-10-05 22:06:36,361 - Epoch: [179][ 680/ 1236] Overall Loss 0.183774 Objective Loss 0.183774 LR 0.000250 Time 0.021392 +2023-10-05 22:06:36,567 - Epoch: [179][ 690/ 1236] Overall Loss 0.183972 Objective Loss 0.183972 LR 0.000250 Time 0.021380 +2023-10-05 22:06:36,773 - Epoch: [179][ 700/ 1236] Overall Loss 0.183676 Objective Loss 0.183676 LR 0.000250 Time 0.021369 +2023-10-05 22:06:36,979 - Epoch: [179][ 710/ 1236] Overall Loss 0.184012 Objective Loss 0.184012 LR 0.000250 Time 0.021356 +2023-10-05 22:06:37,185 - Epoch: [179][ 720/ 1236] Overall Loss 0.184055 Objective Loss 0.184055 LR 0.000250 Time 0.021346 +2023-10-05 22:06:37,391 - Epoch: [179][ 730/ 1236] Overall Loss 0.183655 Objective Loss 0.183655 LR 0.000250 Time 0.021335 +2023-10-05 22:06:37,597 - Epoch: [179][ 740/ 1236] Overall Loss 0.183514 Objective Loss 0.183514 LR 0.000250 Time 0.021325 +2023-10-05 22:06:37,803 - Epoch: [179][ 750/ 1236] Overall Loss 0.183501 Objective Loss 0.183501 LR 0.000250 Time 0.021314 +2023-10-05 22:06:38,009 - Epoch: [179][ 760/ 1236] Overall Loss 0.183541 Objective Loss 0.183541 LR 0.000250 Time 0.021305 +2023-10-05 22:06:38,215 - Epoch: [179][ 770/ 1236] Overall Loss 0.183953 Objective Loss 0.183953 LR 0.000250 Time 0.021295 +2023-10-05 22:06:38,421 - Epoch: [179][ 780/ 1236] Overall Loss 0.183989 Objective Loss 0.183989 LR 0.000250 Time 0.021286 +2023-10-05 22:06:38,627 - Epoch: [179][ 790/ 1236] Overall Loss 0.183947 Objective Loss 0.183947 LR 0.000250 Time 0.021276 +2023-10-05 22:06:38,833 - Epoch: [179][ 800/ 1236] Overall Loss 0.183995 Objective Loss 0.183995 LR 0.000250 Time 0.021268 +2023-10-05 22:06:39,039 - Epoch: [179][ 810/ 1236] Overall Loss 0.184162 Objective Loss 0.184162 LR 0.000250 Time 0.021259 +2023-10-05 22:06:39,245 - Epoch: [179][ 820/ 1236] Overall Loss 0.184237 Objective Loss 0.184237 LR 0.000250 Time 0.021250 +2023-10-05 22:06:39,451 - Epoch: [179][ 830/ 1236] Overall Loss 0.184053 Objective Loss 0.184053 LR 0.000250 Time 0.021242 +2023-10-05 22:06:39,657 - Epoch: [179][ 840/ 1236] Overall Loss 0.183870 Objective Loss 0.183870 LR 0.000250 Time 0.021234 +2023-10-05 22:06:39,863 - Epoch: [179][ 850/ 1236] Overall Loss 0.183768 Objective Loss 0.183768 LR 0.000250 Time 0.021226 +2023-10-05 22:06:40,069 - Epoch: [179][ 860/ 1236] Overall Loss 0.183720 Objective Loss 0.183720 LR 0.000250 Time 0.021219 +2023-10-05 22:06:40,275 - Epoch: [179][ 870/ 1236] Overall Loss 0.183547 Objective Loss 0.183547 LR 0.000250 Time 0.021211 +2023-10-05 22:06:40,481 - Epoch: [179][ 880/ 1236] Overall Loss 0.183521 Objective Loss 0.183521 LR 0.000250 Time 0.021204 +2023-10-05 22:06:40,687 - Epoch: [179][ 890/ 1236] Overall Loss 0.183496 Objective Loss 0.183496 LR 0.000250 Time 0.021196 +2023-10-05 22:06:40,894 - Epoch: [179][ 900/ 1236] Overall Loss 0.183553 Objective Loss 0.183553 LR 0.000250 Time 0.021190 +2023-10-05 22:06:41,099 - Epoch: [179][ 910/ 1236] Overall Loss 0.183362 Objective Loss 0.183362 LR 0.000250 Time 0.021182 +2023-10-05 22:06:41,306 - Epoch: [179][ 920/ 1236] Overall Loss 0.183273 Objective Loss 0.183273 LR 0.000250 Time 0.021176 +2023-10-05 22:06:41,512 - Epoch: [179][ 930/ 1236] Overall Loss 0.183396 Objective Loss 0.183396 LR 0.000250 Time 0.021170 +2023-10-05 22:06:41,718 - Epoch: [179][ 940/ 1236] Overall Loss 0.183171 Objective Loss 0.183171 LR 0.000250 Time 0.021163 +2023-10-05 22:06:41,924 - Epoch: [179][ 950/ 1236] Overall Loss 0.183044 Objective Loss 0.183044 LR 0.000250 Time 0.021157 +2023-10-05 22:06:42,131 - Epoch: [179][ 960/ 1236] Overall Loss 0.182908 Objective Loss 0.182908 LR 0.000250 Time 0.021151 +2023-10-05 22:06:42,336 - Epoch: [179][ 970/ 1236] Overall Loss 0.182933 Objective Loss 0.182933 LR 0.000250 Time 0.021145 +2023-10-05 22:06:42,542 - Epoch: [179][ 980/ 1236] Overall Loss 0.182798 Objective Loss 0.182798 LR 0.000250 Time 0.021139 +2023-10-05 22:06:42,748 - Epoch: [179][ 990/ 1236] Overall Loss 0.182710 Objective Loss 0.182710 LR 0.000250 Time 0.021133 +2023-10-05 22:06:42,955 - Epoch: [179][ 1000/ 1236] Overall Loss 0.182742 Objective Loss 0.182742 LR 0.000250 Time 0.021128 +2023-10-05 22:06:43,160 - Epoch: [179][ 1010/ 1236] Overall Loss 0.182757 Objective Loss 0.182757 LR 0.000250 Time 0.021122 +2023-10-05 22:06:43,367 - Epoch: [179][ 1020/ 1236] Overall Loss 0.182781 Objective Loss 0.182781 LR 0.000250 Time 0.021117 +2023-10-05 22:06:43,573 - Epoch: [179][ 1030/ 1236] Overall Loss 0.182704 Objective Loss 0.182704 LR 0.000250 Time 0.021112 +2023-10-05 22:06:43,779 - Epoch: [179][ 1040/ 1236] Overall Loss 0.182568 Objective Loss 0.182568 LR 0.000250 Time 0.021107 +2023-10-05 22:06:43,984 - Epoch: [179][ 1050/ 1236] Overall Loss 0.182590 Objective Loss 0.182590 LR 0.000250 Time 0.021100 +2023-10-05 22:06:44,189 - Epoch: [179][ 1060/ 1236] Overall Loss 0.182431 Objective Loss 0.182431 LR 0.000250 Time 0.021094 +2023-10-05 22:06:44,394 - Epoch: [179][ 1070/ 1236] Overall Loss 0.182424 Objective Loss 0.182424 LR 0.000250 Time 0.021088 +2023-10-05 22:06:44,599 - Epoch: [179][ 1080/ 1236] Overall Loss 0.182506 Objective Loss 0.182506 LR 0.000250 Time 0.021083 +2023-10-05 22:06:44,803 - Epoch: [179][ 1090/ 1236] Overall Loss 0.182570 Objective Loss 0.182570 LR 0.000250 Time 0.021076 +2023-10-05 22:06:45,009 - Epoch: [179][ 1100/ 1236] Overall Loss 0.182524 Objective Loss 0.182524 LR 0.000250 Time 0.021071 +2023-10-05 22:06:45,213 - Epoch: [179][ 1110/ 1236] Overall Loss 0.182400 Objective Loss 0.182400 LR 0.000250 Time 0.021065 +2023-10-05 22:06:45,418 - Epoch: [179][ 1120/ 1236] Overall Loss 0.182394 Objective Loss 0.182394 LR 0.000250 Time 0.021060 +2023-10-05 22:06:45,623 - Epoch: [179][ 1130/ 1236] Overall Loss 0.182428 Objective Loss 0.182428 LR 0.000250 Time 0.021055 +2023-10-05 22:06:45,828 - Epoch: [179][ 1140/ 1236] Overall Loss 0.182491 Objective Loss 0.182491 LR 0.000250 Time 0.021050 +2023-10-05 22:06:46,041 - Epoch: [179][ 1150/ 1236] Overall Loss 0.182428 Objective Loss 0.182428 LR 0.000250 Time 0.021051 +2023-10-05 22:06:46,258 - Epoch: [179][ 1160/ 1236] Overall Loss 0.182349 Objective Loss 0.182349 LR 0.000250 Time 0.021056 +2023-10-05 22:06:46,471 - Epoch: [179][ 1170/ 1236] Overall Loss 0.182361 Objective Loss 0.182361 LR 0.000250 Time 0.021058 +2023-10-05 22:06:46,689 - Epoch: [179][ 1180/ 1236] Overall Loss 0.182400 Objective Loss 0.182400 LR 0.000250 Time 0.021064 +2023-10-05 22:06:46,903 - Epoch: [179][ 1190/ 1236] Overall Loss 0.182465 Objective Loss 0.182465 LR 0.000250 Time 0.021066 +2023-10-05 22:06:47,120 - Epoch: [179][ 1200/ 1236] Overall Loss 0.182736 Objective Loss 0.182736 LR 0.000250 Time 0.021072 +2023-10-05 22:06:47,330 - Epoch: [179][ 1210/ 1236] Overall Loss 0.182944 Objective Loss 0.182944 LR 0.000250 Time 0.021070 +2023-10-05 22:06:47,548 - Epoch: [179][ 1220/ 1236] Overall Loss 0.182796 Objective Loss 0.182796 LR 0.000250 Time 0.021076 +2023-10-05 22:06:47,809 - Epoch: [179][ 1230/ 1236] Overall Loss 0.182704 Objective Loss 0.182704 LR 0.000250 Time 0.021117 +2023-10-05 22:06:47,928 - Epoch: [179][ 1236/ 1236] Overall Loss 0.182669 Objective Loss 0.182669 Top1 86.761711 Top5 98.370672 LR 0.000250 Time 0.021110 +2023-10-05 22:06:48,048 - --- validate (epoch=179)----------- +2023-10-05 22:06:48,049 - 29943 samples (256 per mini-batch) +2023-10-05 22:06:48,499 - Epoch: [179][ 10/ 117] Loss 0.343850 Top1 85.468750 Top5 97.890625 +2023-10-05 22:06:48,648 - Epoch: [179][ 20/ 117] Loss 0.324599 Top1 85.273438 Top5 97.851562 +2023-10-05 22:06:48,795 - Epoch: [179][ 30/ 117] Loss 0.310377 Top1 85.716146 Top5 98.111979 +2023-10-05 22:06:48,953 - Epoch: [179][ 40/ 117] Loss 0.303517 Top1 85.810547 Top5 98.134766 +2023-10-05 22:06:49,105 - Epoch: [179][ 50/ 117] Loss 0.305640 Top1 85.531250 Top5 98.078125 +2023-10-05 22:06:49,263 - Epoch: [179][ 60/ 117] Loss 0.312492 Top1 85.332031 Top5 98.079427 +2023-10-05 22:06:49,415 - Epoch: [179][ 70/ 117] Loss 0.314886 Top1 85.256696 Top5 98.152902 +2023-10-05 22:06:49,572 - Epoch: [179][ 80/ 117] Loss 0.314070 Top1 85.288086 Top5 98.115234 +2023-10-05 22:06:49,724 - Epoch: [179][ 90/ 117] Loss 0.317050 Top1 85.156250 Top5 98.142361 +2023-10-05 22:06:49,882 - Epoch: [179][ 100/ 117] Loss 0.315921 Top1 85.152344 Top5 98.136719 +2023-10-05 22:06:50,041 - Epoch: [179][ 110/ 117] Loss 0.313641 Top1 85.195312 Top5 98.142756 +2023-10-05 22:06:50,127 - Epoch: [179][ 117/ 117] Loss 0.313375 Top1 85.151788 Top5 98.143139 +2023-10-05 22:06:50,260 - ==> Top1: 85.152 Top5: 98.143 Loss: 0.313 + +2023-10-05 22:06:50,261 - ==> Confusion: +[[ 953 1 1 0 6 3 0 0 4 59 2 0 1 1 6 1 1 0 0 0 11] + [ 1 1058 1 0 7 17 1 19 2 0 1 3 0 0 1 3 3 0 5 1 8] + [ 8 3 965 12 3 0 21 6 0 0 4 3 7 2 4 2 0 3 4 3 6] + [ 5 3 10 958 0 4 0 2 2 1 5 1 8 2 35 4 0 5 24 2 18] + [ 31 5 0 0 985 0 0 1 0 9 1 1 0 1 5 1 6 1 0 2 1] + [ 4 37 1 1 9 985 1 17 1 2 5 10 0 10 5 1 3 0 3 3 18] + [ 0 8 21 1 0 0 1121 11 0 0 1 4 1 1 1 5 0 0 1 5 10] + [ 7 21 9 0 1 22 3 1086 1 3 4 9 0 2 1 2 0 0 33 7 7] + [ 20 1 0 0 2 2 1 0 972 46 9 1 2 8 15 3 1 1 2 0 3] + [ 126 0 3 0 3 3 0 0 16 941 1 1 0 9 1 3 0 3 0 2 7] + [ 5 5 9 3 2 1 4 9 15 3 965 5 0 8 6 0 2 0 3 1 7] + [ 1 1 2 0 0 15 1 1 0 0 0 962 15 4 0 4 0 18 0 9 2] + [ 1 1 1 5 0 2 0 1 1 1 2 43 973 1 2 3 2 17 3 3 6] + [ 4 0 1 0 3 6 0 1 14 20 7 5 3 1039 4 2 1 0 0 1 8] + [ 16 3 4 4 5 0 0 0 28 5 1 1 1 1 1008 0 1 3 9 0 11] + [ 1 3 2 0 3 0 1 0 0 0 0 12 6 1 1 1072 13 10 0 7 2] + [ 2 13 1 0 11 4 0 2 1 0 0 3 2 0 2 9 1097 0 0 2 12] + [ 0 0 0 3 1 1 2 0 1 0 0 2 15 1 0 3 2 1004 0 1 2] + [ 1 9 6 17 2 0 0 28 1 0 2 1 2 0 10 0 1 0 978 2 8] + [ 1 2 3 3 3 6 9 13 1 0 2 15 5 0 0 6 6 2 3 1064 8] + [ 193 158 138 52 110 108 30 92 122 96 158 112 317 259 148 48 105 83 115 150 5311]] + +2023-10-05 22:06:50,262 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:06:50,262 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:06:50,268 - + +2023-10-05 22:06:50,268 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:06:51,382 - Epoch: [180][ 10/ 1236] Overall Loss 0.168000 Objective Loss 0.168000 LR 0.000125 Time 0.111310 +2023-10-05 22:06:51,586 - Epoch: [180][ 20/ 1236] Overall Loss 0.178484 Objective Loss 0.178484 LR 0.000125 Time 0.065834 +2023-10-05 22:06:51,789 - Epoch: [180][ 30/ 1236] Overall Loss 0.179515 Objective Loss 0.179515 LR 0.000125 Time 0.050652 +2023-10-05 22:06:51,993 - Epoch: [180][ 40/ 1236] Overall Loss 0.176590 Objective Loss 0.176590 LR 0.000125 Time 0.043063 +2023-10-05 22:06:52,196 - Epoch: [180][ 50/ 1236] Overall Loss 0.178476 Objective Loss 0.178476 LR 0.000125 Time 0.038508 +2023-10-05 22:06:52,399 - Epoch: [180][ 60/ 1236] Overall Loss 0.180149 Objective Loss 0.180149 LR 0.000125 Time 0.035472 +2023-10-05 22:06:52,603 - Epoch: [180][ 70/ 1236] Overall Loss 0.183366 Objective Loss 0.183366 LR 0.000125 Time 0.033304 +2023-10-05 22:06:52,806 - Epoch: [180][ 80/ 1236] Overall Loss 0.183639 Objective Loss 0.183639 LR 0.000125 Time 0.031678 +2023-10-05 22:06:53,009 - Epoch: [180][ 90/ 1236] Overall Loss 0.185396 Objective Loss 0.185396 LR 0.000125 Time 0.030413 +2023-10-05 22:06:53,213 - Epoch: [180][ 100/ 1236] Overall Loss 0.184575 Objective Loss 0.184575 LR 0.000125 Time 0.029404 +2023-10-05 22:06:53,416 - Epoch: [180][ 110/ 1236] Overall Loss 0.183771 Objective Loss 0.183771 LR 0.000125 Time 0.028571 +2023-10-05 22:06:53,620 - Epoch: [180][ 120/ 1236] Overall Loss 0.184140 Objective Loss 0.184140 LR 0.000125 Time 0.027894 +2023-10-05 22:06:53,825 - Epoch: [180][ 130/ 1236] Overall Loss 0.184284 Objective Loss 0.184284 LR 0.000125 Time 0.027315 +2023-10-05 22:06:54,029 - Epoch: [180][ 140/ 1236] Overall Loss 0.182984 Objective Loss 0.182984 LR 0.000125 Time 0.026820 +2023-10-05 22:06:54,233 - Epoch: [180][ 150/ 1236] Overall Loss 0.182314 Objective Loss 0.182314 LR 0.000125 Time 0.026392 +2023-10-05 22:06:54,438 - Epoch: [180][ 160/ 1236] Overall Loss 0.182696 Objective Loss 0.182696 LR 0.000125 Time 0.026021 +2023-10-05 22:06:54,644 - Epoch: [180][ 170/ 1236] Overall Loss 0.181623 Objective Loss 0.181623 LR 0.000125 Time 0.025696 +2023-10-05 22:06:54,850 - Epoch: [180][ 180/ 1236] Overall Loss 0.180868 Objective Loss 0.180868 LR 0.000125 Time 0.025411 +2023-10-05 22:06:55,056 - Epoch: [180][ 190/ 1236] Overall Loss 0.180380 Objective Loss 0.180380 LR 0.000125 Time 0.025153 +2023-10-05 22:06:55,261 - Epoch: [180][ 200/ 1236] Overall Loss 0.180369 Objective Loss 0.180369 LR 0.000125 Time 0.024921 +2023-10-05 22:06:55,467 - Epoch: [180][ 210/ 1236] Overall Loss 0.179174 Objective Loss 0.179174 LR 0.000125 Time 0.024710 +2023-10-05 22:06:55,673 - Epoch: [180][ 220/ 1236] Overall Loss 0.179490 Objective Loss 0.179490 LR 0.000125 Time 0.024522 +2023-10-05 22:06:55,878 - Epoch: [180][ 230/ 1236] Overall Loss 0.179531 Objective Loss 0.179531 LR 0.000125 Time 0.024346 +2023-10-05 22:06:56,085 - Epoch: [180][ 240/ 1236] Overall Loss 0.179152 Objective Loss 0.179152 LR 0.000125 Time 0.024188 +2023-10-05 22:06:56,290 - Epoch: [180][ 250/ 1236] Overall Loss 0.178959 Objective Loss 0.178959 LR 0.000125 Time 0.024040 +2023-10-05 22:06:56,496 - Epoch: [180][ 260/ 1236] Overall Loss 0.177792 Objective Loss 0.177792 LR 0.000125 Time 0.023905 +2023-10-05 22:06:56,702 - Epoch: [180][ 270/ 1236] Overall Loss 0.177606 Objective Loss 0.177606 LR 0.000125 Time 0.023779 +2023-10-05 22:06:56,908 - Epoch: [180][ 280/ 1236] Overall Loss 0.177112 Objective Loss 0.177112 LR 0.000125 Time 0.023664 +2023-10-05 22:06:57,113 - Epoch: [180][ 290/ 1236] Overall Loss 0.177294 Objective Loss 0.177294 LR 0.000125 Time 0.023554 +2023-10-05 22:06:57,319 - Epoch: [180][ 300/ 1236] Overall Loss 0.177403 Objective Loss 0.177403 LR 0.000125 Time 0.023454 +2023-10-05 22:06:57,525 - Epoch: [180][ 310/ 1236] Overall Loss 0.177032 Objective Loss 0.177032 LR 0.000125 Time 0.023358 +2023-10-05 22:06:57,731 - Epoch: [180][ 320/ 1236] Overall Loss 0.176837 Objective Loss 0.176837 LR 0.000125 Time 0.023271 +2023-10-05 22:06:57,937 - Epoch: [180][ 330/ 1236] Overall Loss 0.176231 Objective Loss 0.176231 LR 0.000125 Time 0.023187 +2023-10-05 22:06:58,143 - Epoch: [180][ 340/ 1236] Overall Loss 0.175599 Objective Loss 0.175599 LR 0.000125 Time 0.023110 +2023-10-05 22:06:58,349 - Epoch: [180][ 350/ 1236] Overall Loss 0.176019 Objective Loss 0.176019 LR 0.000125 Time 0.023036 +2023-10-05 22:06:58,555 - Epoch: [180][ 360/ 1236] Overall Loss 0.175986 Objective Loss 0.175986 LR 0.000125 Time 0.022967 +2023-10-05 22:06:58,760 - Epoch: [180][ 370/ 1236] Overall Loss 0.175774 Objective Loss 0.175774 LR 0.000125 Time 0.022898 +2023-10-05 22:06:58,964 - Epoch: [180][ 380/ 1236] Overall Loss 0.176013 Objective Loss 0.176013 LR 0.000125 Time 0.022833 +2023-10-05 22:06:59,168 - Epoch: [180][ 390/ 1236] Overall Loss 0.175779 Objective Loss 0.175779 LR 0.000125 Time 0.022770 +2023-10-05 22:06:59,373 - Epoch: [180][ 400/ 1236] Overall Loss 0.175548 Objective Loss 0.175548 LR 0.000125 Time 0.022711 +2023-10-05 22:06:59,577 - Epoch: [180][ 410/ 1236] Overall Loss 0.175282 Objective Loss 0.175282 LR 0.000125 Time 0.022653 +2023-10-05 22:06:59,781 - Epoch: [180][ 420/ 1236] Overall Loss 0.175599 Objective Loss 0.175599 LR 0.000125 Time 0.022599 +2023-10-05 22:06:59,984 - Epoch: [180][ 430/ 1236] Overall Loss 0.175220 Objective Loss 0.175220 LR 0.000125 Time 0.022546 +2023-10-05 22:07:00,189 - Epoch: [180][ 440/ 1236] Overall Loss 0.175582 Objective Loss 0.175582 LR 0.000125 Time 0.022497 +2023-10-05 22:07:00,392 - Epoch: [180][ 450/ 1236] Overall Loss 0.175570 Objective Loss 0.175570 LR 0.000125 Time 0.022449 +2023-10-05 22:07:00,596 - Epoch: [180][ 460/ 1236] Overall Loss 0.175511 Objective Loss 0.175511 LR 0.000125 Time 0.022404 +2023-10-05 22:07:00,800 - Epoch: [180][ 470/ 1236] Overall Loss 0.175270 Objective Loss 0.175270 LR 0.000125 Time 0.022360 +2023-10-05 22:07:01,004 - Epoch: [180][ 480/ 1236] Overall Loss 0.175153 Objective Loss 0.175153 LR 0.000125 Time 0.022319 +2023-10-05 22:07:01,208 - Epoch: [180][ 490/ 1236] Overall Loss 0.175021 Objective Loss 0.175021 LR 0.000125 Time 0.022279 +2023-10-05 22:07:01,413 - Epoch: [180][ 500/ 1236] Overall Loss 0.175141 Objective Loss 0.175141 LR 0.000125 Time 0.022241 +2023-10-05 22:07:01,617 - Epoch: [180][ 510/ 1236] Overall Loss 0.175478 Objective Loss 0.175478 LR 0.000125 Time 0.022204 +2023-10-05 22:07:01,821 - Epoch: [180][ 520/ 1236] Overall Loss 0.175412 Objective Loss 0.175412 LR 0.000125 Time 0.022169 +2023-10-05 22:07:02,025 - Epoch: [180][ 530/ 1236] Overall Loss 0.175853 Objective Loss 0.175853 LR 0.000125 Time 0.022135 +2023-10-05 22:07:02,229 - Epoch: [180][ 540/ 1236] Overall Loss 0.175558 Objective Loss 0.175558 LR 0.000125 Time 0.022103 +2023-10-05 22:07:02,433 - Epoch: [180][ 550/ 1236] Overall Loss 0.175660 Objective Loss 0.175660 LR 0.000125 Time 0.022071 +2023-10-05 22:07:02,637 - Epoch: [180][ 560/ 1236] Overall Loss 0.175383 Objective Loss 0.175383 LR 0.000125 Time 0.022041 +2023-10-05 22:07:02,841 - Epoch: [180][ 570/ 1236] Overall Loss 0.175699 Objective Loss 0.175699 LR 0.000125 Time 0.022011 +2023-10-05 22:07:03,045 - Epoch: [180][ 580/ 1236] Overall Loss 0.175612 Objective Loss 0.175612 LR 0.000125 Time 0.021983 +2023-10-05 22:07:03,249 - Epoch: [180][ 590/ 1236] Overall Loss 0.175762 Objective Loss 0.175762 LR 0.000125 Time 0.021956 +2023-10-05 22:07:03,453 - Epoch: [180][ 600/ 1236] Overall Loss 0.175585 Objective Loss 0.175585 LR 0.000125 Time 0.021929 +2023-10-05 22:07:03,657 - Epoch: [180][ 610/ 1236] Overall Loss 0.175632 Objective Loss 0.175632 LR 0.000125 Time 0.021903 +2023-10-05 22:07:03,862 - Epoch: [180][ 620/ 1236] Overall Loss 0.175946 Objective Loss 0.175946 LR 0.000125 Time 0.021879 +2023-10-05 22:07:04,066 - Epoch: [180][ 630/ 1236] Overall Loss 0.176054 Objective Loss 0.176054 LR 0.000125 Time 0.021855 +2023-10-05 22:07:04,270 - Epoch: [180][ 640/ 1236] Overall Loss 0.176047 Objective Loss 0.176047 LR 0.000125 Time 0.021833 +2023-10-05 22:07:04,474 - Epoch: [180][ 650/ 1236] Overall Loss 0.176057 Objective Loss 0.176057 LR 0.000125 Time 0.021810 +2023-10-05 22:07:04,678 - Epoch: [180][ 660/ 1236] Overall Loss 0.176118 Objective Loss 0.176118 LR 0.000125 Time 0.021788 +2023-10-05 22:07:04,882 - Epoch: [180][ 670/ 1236] Overall Loss 0.176160 Objective Loss 0.176160 LR 0.000125 Time 0.021767 +2023-10-05 22:07:05,086 - Epoch: [180][ 680/ 1236] Overall Loss 0.176053 Objective Loss 0.176053 LR 0.000125 Time 0.021746 +2023-10-05 22:07:05,291 - Epoch: [180][ 690/ 1236] Overall Loss 0.176014 Objective Loss 0.176014 LR 0.000125 Time 0.021726 +2023-10-05 22:07:05,495 - Epoch: [180][ 700/ 1236] Overall Loss 0.175743 Objective Loss 0.175743 LR 0.000125 Time 0.021707 +2023-10-05 22:07:05,699 - Epoch: [180][ 710/ 1236] Overall Loss 0.175542 Objective Loss 0.175542 LR 0.000125 Time 0.021689 +2023-10-05 22:07:05,903 - Epoch: [180][ 720/ 1236] Overall Loss 0.175504 Objective Loss 0.175504 LR 0.000125 Time 0.021671 +2023-10-05 22:07:06,107 - Epoch: [180][ 730/ 1236] Overall Loss 0.175660 Objective Loss 0.175660 LR 0.000125 Time 0.021653 +2023-10-05 22:07:06,311 - Epoch: [180][ 740/ 1236] Overall Loss 0.175518 Objective Loss 0.175518 LR 0.000125 Time 0.021636 +2023-10-05 22:07:06,515 - Epoch: [180][ 750/ 1236] Overall Loss 0.175402 Objective Loss 0.175402 LR 0.000125 Time 0.021619 +2023-10-05 22:07:06,720 - Epoch: [180][ 760/ 1236] Overall Loss 0.175223 Objective Loss 0.175223 LR 0.000125 Time 0.021602 +2023-10-05 22:07:06,924 - Epoch: [180][ 770/ 1236] Overall Loss 0.175551 Objective Loss 0.175551 LR 0.000125 Time 0.021586 +2023-10-05 22:07:07,128 - Epoch: [180][ 780/ 1236] Overall Loss 0.175596 Objective Loss 0.175596 LR 0.000125 Time 0.021571 +2023-10-05 22:07:07,332 - Epoch: [180][ 790/ 1236] Overall Loss 0.175671 Objective Loss 0.175671 LR 0.000125 Time 0.021556 +2023-10-05 22:07:07,536 - Epoch: [180][ 800/ 1236] Overall Loss 0.175652 Objective Loss 0.175652 LR 0.000125 Time 0.021541 +2023-10-05 22:07:07,741 - Epoch: [180][ 810/ 1236] Overall Loss 0.176109 Objective Loss 0.176109 LR 0.000125 Time 0.021527 +2023-10-05 22:07:07,945 - Epoch: [180][ 820/ 1236] Overall Loss 0.176393 Objective Loss 0.176393 LR 0.000125 Time 0.021513 +2023-10-05 22:07:08,149 - Epoch: [180][ 830/ 1236] Overall Loss 0.176492 Objective Loss 0.176492 LR 0.000125 Time 0.021500 +2023-10-05 22:07:08,353 - Epoch: [180][ 840/ 1236] Overall Loss 0.176695 Objective Loss 0.176695 LR 0.000125 Time 0.021486 +2023-10-05 22:07:08,557 - Epoch: [180][ 850/ 1236] Overall Loss 0.176756 Objective Loss 0.176756 LR 0.000125 Time 0.021473 +2023-10-05 22:07:08,762 - Epoch: [180][ 860/ 1236] Overall Loss 0.176782 Objective Loss 0.176782 LR 0.000125 Time 0.021460 +2023-10-05 22:07:08,966 - Epoch: [180][ 870/ 1236] Overall Loss 0.176782 Objective Loss 0.176782 LR 0.000125 Time 0.021448 +2023-10-05 22:07:09,170 - Epoch: [180][ 880/ 1236] Overall Loss 0.176715 Objective Loss 0.176715 LR 0.000125 Time 0.021436 +2023-10-05 22:07:09,374 - Epoch: [180][ 890/ 1236] Overall Loss 0.176818 Objective Loss 0.176818 LR 0.000125 Time 0.021424 +2023-10-05 22:07:09,578 - Epoch: [180][ 900/ 1236] Overall Loss 0.176889 Objective Loss 0.176889 LR 0.000125 Time 0.021412 +2023-10-05 22:07:09,782 - Epoch: [180][ 910/ 1236] Overall Loss 0.176969 Objective Loss 0.176969 LR 0.000125 Time 0.021401 +2023-10-05 22:07:09,987 - Epoch: [180][ 920/ 1236] Overall Loss 0.176847 Objective Loss 0.176847 LR 0.000125 Time 0.021391 +2023-10-05 22:07:10,191 - Epoch: [180][ 930/ 1236] Overall Loss 0.177109 Objective Loss 0.177109 LR 0.000125 Time 0.021380 +2023-10-05 22:07:10,396 - Epoch: [180][ 940/ 1236] Overall Loss 0.177129 Objective Loss 0.177129 LR 0.000125 Time 0.021369 +2023-10-05 22:07:10,600 - Epoch: [180][ 950/ 1236] Overall Loss 0.177045 Objective Loss 0.177045 LR 0.000125 Time 0.021359 +2023-10-05 22:07:10,804 - Epoch: [180][ 960/ 1236] Overall Loss 0.176828 Objective Loss 0.176828 LR 0.000125 Time 0.021349 +2023-10-05 22:07:11,008 - Epoch: [180][ 970/ 1236] Overall Loss 0.176749 Objective Loss 0.176749 LR 0.000125 Time 0.021339 +2023-10-05 22:07:11,212 - Epoch: [180][ 980/ 1236] Overall Loss 0.176760 Objective Loss 0.176760 LR 0.000125 Time 0.021329 +2023-10-05 22:07:11,417 - Epoch: [180][ 990/ 1236] Overall Loss 0.176680 Objective Loss 0.176680 LR 0.000125 Time 0.021319 +2023-10-05 22:07:11,620 - Epoch: [180][ 1000/ 1236] Overall Loss 0.176755 Objective Loss 0.176755 LR 0.000125 Time 0.021310 +2023-10-05 22:07:11,825 - Epoch: [180][ 1010/ 1236] Overall Loss 0.176779 Objective Loss 0.176779 LR 0.000125 Time 0.021300 +2023-10-05 22:07:12,029 - Epoch: [180][ 1020/ 1236] Overall Loss 0.176673 Objective Loss 0.176673 LR 0.000125 Time 0.021291 +2023-10-05 22:07:12,233 - Epoch: [180][ 1030/ 1236] Overall Loss 0.176697 Objective Loss 0.176697 LR 0.000125 Time 0.021282 +2023-10-05 22:07:12,437 - Epoch: [180][ 1040/ 1236] Overall Loss 0.176667 Objective Loss 0.176667 LR 0.000125 Time 0.021273 +2023-10-05 22:07:12,641 - Epoch: [180][ 1050/ 1236] Overall Loss 0.176701 Objective Loss 0.176701 LR 0.000125 Time 0.021265 +2023-10-05 22:07:12,845 - Epoch: [180][ 1060/ 1236] Overall Loss 0.176617 Objective Loss 0.176617 LR 0.000125 Time 0.021257 +2023-10-05 22:07:13,049 - Epoch: [180][ 1070/ 1236] Overall Loss 0.176406 Objective Loss 0.176406 LR 0.000125 Time 0.021248 +2023-10-05 22:07:13,253 - Epoch: [180][ 1080/ 1236] Overall Loss 0.176574 Objective Loss 0.176574 LR 0.000125 Time 0.021240 +2023-10-05 22:07:13,457 - Epoch: [180][ 1090/ 1236] Overall Loss 0.176666 Objective Loss 0.176666 LR 0.000125 Time 0.021232 +2023-10-05 22:07:13,662 - Epoch: [180][ 1100/ 1236] Overall Loss 0.176642 Objective Loss 0.176642 LR 0.000125 Time 0.021225 +2023-10-05 22:07:13,866 - Epoch: [180][ 1110/ 1236] Overall Loss 0.176533 Objective Loss 0.176533 LR 0.000125 Time 0.021217 +2023-10-05 22:07:14,070 - Epoch: [180][ 1120/ 1236] Overall Loss 0.176344 Objective Loss 0.176344 LR 0.000125 Time 0.021210 +2023-10-05 22:07:14,274 - Epoch: [180][ 1130/ 1236] Overall Loss 0.176397 Objective Loss 0.176397 LR 0.000125 Time 0.021202 +2023-10-05 22:07:14,478 - Epoch: [180][ 1140/ 1236] Overall Loss 0.176373 Objective Loss 0.176373 LR 0.000125 Time 0.021195 +2023-10-05 22:07:14,682 - Epoch: [180][ 1150/ 1236] Overall Loss 0.176321 Objective Loss 0.176321 LR 0.000125 Time 0.021188 +2023-10-05 22:07:14,887 - Epoch: [180][ 1160/ 1236] Overall Loss 0.176434 Objective Loss 0.176434 LR 0.000125 Time 0.021181 +2023-10-05 22:07:15,091 - Epoch: [180][ 1170/ 1236] Overall Loss 0.176778 Objective Loss 0.176778 LR 0.000125 Time 0.021174 +2023-10-05 22:07:15,295 - Epoch: [180][ 1180/ 1236] Overall Loss 0.176785 Objective Loss 0.176785 LR 0.000125 Time 0.021168 +2023-10-05 22:07:15,499 - Epoch: [180][ 1190/ 1236] Overall Loss 0.176816 Objective Loss 0.176816 LR 0.000125 Time 0.021161 +2023-10-05 22:07:15,703 - Epoch: [180][ 1200/ 1236] Overall Loss 0.176675 Objective Loss 0.176675 LR 0.000125 Time 0.021154 +2023-10-05 22:07:15,908 - Epoch: [180][ 1210/ 1236] Overall Loss 0.176820 Objective Loss 0.176820 LR 0.000125 Time 0.021148 +2023-10-05 22:07:16,112 - Epoch: [180][ 1220/ 1236] Overall Loss 0.176836 Objective Loss 0.176836 LR 0.000125 Time 0.021142 +2023-10-05 22:07:16,369 - Epoch: [180][ 1230/ 1236] Overall Loss 0.176936 Objective Loss 0.176936 LR 0.000125 Time 0.021179 +2023-10-05 22:07:16,488 - Epoch: [180][ 1236/ 1236] Overall Loss 0.177034 Objective Loss 0.177034 Top1 88.187373 Top5 98.370672 LR 0.000125 Time 0.021172 +2023-10-05 22:07:16,616 - --- validate (epoch=180)----------- +2023-10-05 22:07:16,616 - 29943 samples (256 per mini-batch) +2023-10-05 22:07:17,076 - Epoch: [180][ 10/ 117] Loss 0.328647 Top1 85.195312 Top5 98.281250 +2023-10-05 22:07:17,235 - Epoch: [180][ 20/ 117] Loss 0.330003 Top1 84.902344 Top5 98.300781 +2023-10-05 22:07:17,390 - Epoch: [180][ 30/ 117] Loss 0.315770 Top1 85.546875 Top5 98.151042 +2023-10-05 22:07:17,548 - Epoch: [180][ 40/ 117] Loss 0.314292 Top1 85.634766 Top5 98.154297 +2023-10-05 22:07:17,702 - Epoch: [180][ 50/ 117] Loss 0.301496 Top1 85.601562 Top5 98.171875 +2023-10-05 22:07:17,858 - Epoch: [180][ 60/ 117] Loss 0.301040 Top1 85.332031 Top5 98.079427 +2023-10-05 22:07:18,012 - Epoch: [180][ 70/ 117] Loss 0.304669 Top1 85.223214 Top5 98.108259 +2023-10-05 22:07:18,169 - Epoch: [180][ 80/ 117] Loss 0.308731 Top1 85.288086 Top5 98.115234 +2023-10-05 22:07:18,323 - Epoch: [180][ 90/ 117] Loss 0.306678 Top1 85.377604 Top5 98.151042 +2023-10-05 22:07:18,480 - Epoch: [180][ 100/ 117] Loss 0.308771 Top1 85.417969 Top5 98.132812 +2023-10-05 22:07:18,642 - Epoch: [180][ 110/ 117] Loss 0.307974 Top1 85.472301 Top5 98.178267 +2023-10-05 22:07:18,728 - Epoch: [180][ 117/ 117] Loss 0.306393 Top1 85.462379 Top5 98.206593 +2023-10-05 22:07:18,859 - ==> Top1: 85.462 Top5: 98.207 Loss: 0.306 + +2023-10-05 22:07:18,860 - ==> Confusion: +[[ 924 0 6 1 7 3 0 0 5 73 1 0 0 2 6 4 3 1 0 0 14] + [ 1 1057 1 0 9 17 1 17 0 0 0 1 0 0 1 4 3 0 7 2 10] + [ 3 2 973 14 2 0 21 7 0 0 4 2 7 2 2 3 1 1 6 3 3] + [ 1 1 8 972 1 4 0 0 2 1 5 1 7 3 31 6 0 5 22 2 17] + [ 19 7 0 0 972 3 1 1 0 9 1 2 0 1 8 5 10 2 0 3 6] + [ 3 25 2 1 4 990 0 18 2 0 5 9 0 18 8 2 5 0 4 2 18] + [ 0 5 14 0 0 0 1136 8 0 0 3 4 1 1 1 5 0 0 2 5 6] + [ 2 23 16 0 2 28 1 1073 1 4 3 10 1 3 0 2 0 0 33 9 7] + [ 18 2 1 0 1 0 1 0 981 40 9 2 1 8 13 2 2 0 5 2 1] + [ 80 0 3 1 2 2 0 0 20 969 0 2 0 16 5 5 0 2 0 2 10] + [ 3 6 9 3 1 1 6 6 11 2 968 1 2 12 4 1 4 0 3 2 8] + [ 1 0 3 0 0 13 0 3 0 0 0 956 22 3 0 5 2 15 0 6 6] + [ 1 2 2 3 0 1 0 1 0 0 1 33 990 2 1 5 2 17 1 3 3] + [ 2 0 1 0 2 5 0 0 10 13 8 3 2 1059 4 2 1 0 0 0 7] + [ 10 1 3 8 2 1 0 0 29 3 2 1 2 2 1010 0 0 2 10 0 15] + [ 1 2 2 0 3 0 1 1 0 1 0 6 6 2 1 1074 13 11 0 7 3] + [ 1 9 1 0 5 3 0 0 2 0 0 3 1 0 3 11 1107 0 0 2 13] + [ 0 0 0 3 0 1 3 0 0 0 0 3 20 3 1 5 0 995 0 0 4] + [ 0 7 6 16 1 0 0 19 1 0 2 0 4 1 12 0 0 0 988 1 10] + [ 0 2 3 3 2 6 8 8 0 0 1 17 4 4 0 4 9 0 3 1066 12] + [ 104 176 141 62 82 125 31 69 104 76 178 88 351 282 159 60 134 60 127 166 5330]] + +2023-10-05 22:07:18,861 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:07:18,861 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:07:18,867 - + +2023-10-05 22:07:18,867 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:07:19,872 - Epoch: [181][ 10/ 1236] Overall Loss 0.165334 Objective Loss 0.165334 LR 0.000125 Time 0.100437 +2023-10-05 22:07:20,076 - Epoch: [181][ 20/ 1236] Overall Loss 0.181739 Objective Loss 0.181739 LR 0.000125 Time 0.060373 +2023-10-05 22:07:20,278 - Epoch: [181][ 30/ 1236] Overall Loss 0.180947 Objective Loss 0.180947 LR 0.000125 Time 0.046966 +2023-10-05 22:07:20,481 - Epoch: [181][ 40/ 1236] Overall Loss 0.180630 Objective Loss 0.180630 LR 0.000125 Time 0.040292 +2023-10-05 22:07:20,683 - Epoch: [181][ 50/ 1236] Overall Loss 0.180106 Objective Loss 0.180106 LR 0.000125 Time 0.036269 +2023-10-05 22:07:20,886 - Epoch: [181][ 60/ 1236] Overall Loss 0.178986 Objective Loss 0.178986 LR 0.000125 Time 0.033600 +2023-10-05 22:07:21,088 - Epoch: [181][ 70/ 1236] Overall Loss 0.179107 Objective Loss 0.179107 LR 0.000125 Time 0.031683 +2023-10-05 22:07:21,290 - Epoch: [181][ 80/ 1236] Overall Loss 0.177930 Objective Loss 0.177930 LR 0.000125 Time 0.030253 +2023-10-05 22:07:21,492 - Epoch: [181][ 90/ 1236] Overall Loss 0.178555 Objective Loss 0.178555 LR 0.000125 Time 0.029125 +2023-10-05 22:07:21,694 - Epoch: [181][ 100/ 1236] Overall Loss 0.178592 Objective Loss 0.178592 LR 0.000125 Time 0.028227 +2023-10-05 22:07:21,895 - Epoch: [181][ 110/ 1236] Overall Loss 0.179623 Objective Loss 0.179623 LR 0.000125 Time 0.027492 +2023-10-05 22:07:22,097 - Epoch: [181][ 120/ 1236] Overall Loss 0.179729 Objective Loss 0.179729 LR 0.000125 Time 0.026878 +2023-10-05 22:07:22,299 - Epoch: [181][ 130/ 1236] Overall Loss 0.178297 Objective Loss 0.178297 LR 0.000125 Time 0.026358 +2023-10-05 22:07:22,500 - Epoch: [181][ 140/ 1236] Overall Loss 0.176687 Objective Loss 0.176687 LR 0.000125 Time 0.025913 +2023-10-05 22:07:22,702 - Epoch: [181][ 150/ 1236] Overall Loss 0.176402 Objective Loss 0.176402 LR 0.000125 Time 0.025527 +2023-10-05 22:07:22,905 - Epoch: [181][ 160/ 1236] Overall Loss 0.176415 Objective Loss 0.176415 LR 0.000125 Time 0.025198 +2023-10-05 22:07:23,107 - Epoch: [181][ 170/ 1236] Overall Loss 0.178985 Objective Loss 0.178985 LR 0.000125 Time 0.024905 +2023-10-05 22:07:23,309 - Epoch: [181][ 180/ 1236] Overall Loss 0.177816 Objective Loss 0.177816 LR 0.000125 Time 0.024641 +2023-10-05 22:07:23,509 - Epoch: [181][ 190/ 1236] Overall Loss 0.177912 Objective Loss 0.177912 LR 0.000125 Time 0.024396 +2023-10-05 22:07:23,710 - Epoch: [181][ 200/ 1236] Overall Loss 0.177382 Objective Loss 0.177382 LR 0.000125 Time 0.024176 +2023-10-05 22:07:23,911 - Epoch: [181][ 210/ 1236] Overall Loss 0.178259 Objective Loss 0.178259 LR 0.000125 Time 0.023982 +2023-10-05 22:07:24,111 - Epoch: [181][ 220/ 1236] Overall Loss 0.178386 Objective Loss 0.178386 LR 0.000125 Time 0.023800 +2023-10-05 22:07:24,313 - Epoch: [181][ 230/ 1236] Overall Loss 0.177983 Objective Loss 0.177983 LR 0.000125 Time 0.023643 +2023-10-05 22:07:24,514 - Epoch: [181][ 240/ 1236] Overall Loss 0.177823 Objective Loss 0.177823 LR 0.000125 Time 0.023492 +2023-10-05 22:07:24,714 - Epoch: [181][ 250/ 1236] Overall Loss 0.178238 Objective Loss 0.178238 LR 0.000125 Time 0.023351 +2023-10-05 22:07:24,914 - Epoch: [181][ 260/ 1236] Overall Loss 0.178636 Objective Loss 0.178636 LR 0.000125 Time 0.023222 +2023-10-05 22:07:25,115 - Epoch: [181][ 270/ 1236] Overall Loss 0.177822 Objective Loss 0.177822 LR 0.000125 Time 0.023103 +2023-10-05 22:07:25,316 - Epoch: [181][ 280/ 1236] Overall Loss 0.178051 Objective Loss 0.178051 LR 0.000125 Time 0.022997 +2023-10-05 22:07:25,518 - Epoch: [181][ 290/ 1236] Overall Loss 0.177529 Objective Loss 0.177529 LR 0.000125 Time 0.022899 +2023-10-05 22:07:25,722 - Epoch: [181][ 300/ 1236] Overall Loss 0.177321 Objective Loss 0.177321 LR 0.000125 Time 0.022813 +2023-10-05 22:07:25,926 - Epoch: [181][ 310/ 1236] Overall Loss 0.177906 Objective Loss 0.177906 LR 0.000125 Time 0.022734 +2023-10-05 22:07:26,128 - Epoch: [181][ 320/ 1236] Overall Loss 0.177434 Objective Loss 0.177434 LR 0.000125 Time 0.022654 +2023-10-05 22:07:26,330 - Epoch: [181][ 330/ 1236] Overall Loss 0.177396 Objective Loss 0.177396 LR 0.000125 Time 0.022579 +2023-10-05 22:07:26,530 - Epoch: [181][ 340/ 1236] Overall Loss 0.177529 Objective Loss 0.177529 LR 0.000125 Time 0.022502 +2023-10-05 22:07:26,731 - Epoch: [181][ 350/ 1236] Overall Loss 0.177196 Objective Loss 0.177196 LR 0.000125 Time 0.022433 +2023-10-05 22:07:26,933 - Epoch: [181][ 360/ 1236] Overall Loss 0.176652 Objective Loss 0.176652 LR 0.000125 Time 0.022370 +2023-10-05 22:07:27,139 - Epoch: [181][ 370/ 1236] Overall Loss 0.176568 Objective Loss 0.176568 LR 0.000125 Time 0.022320 +2023-10-05 22:07:27,351 - Epoch: [181][ 380/ 1236] Overall Loss 0.176098 Objective Loss 0.176098 LR 0.000125 Time 0.022291 +2023-10-05 22:07:27,560 - Epoch: [181][ 390/ 1236] Overall Loss 0.175746 Objective Loss 0.175746 LR 0.000125 Time 0.022255 +2023-10-05 22:07:27,773 - Epoch: [181][ 400/ 1236] Overall Loss 0.175799 Objective Loss 0.175799 LR 0.000125 Time 0.022229 +2023-10-05 22:07:27,982 - Epoch: [181][ 410/ 1236] Overall Loss 0.175899 Objective Loss 0.175899 LR 0.000125 Time 0.022196 +2023-10-05 22:07:28,195 - Epoch: [181][ 420/ 1236] Overall Loss 0.175797 Objective Loss 0.175797 LR 0.000125 Time 0.022173 +2023-10-05 22:07:28,404 - Epoch: [181][ 430/ 1236] Overall Loss 0.175309 Objective Loss 0.175309 LR 0.000125 Time 0.022143 +2023-10-05 22:07:28,617 - Epoch: [181][ 440/ 1236] Overall Loss 0.174774 Objective Loss 0.174774 LR 0.000125 Time 0.022123 +2023-10-05 22:07:28,828 - Epoch: [181][ 450/ 1236] Overall Loss 0.174945 Objective Loss 0.174945 LR 0.000125 Time 0.022099 +2023-10-05 22:07:29,043 - Epoch: [181][ 460/ 1236] Overall Loss 0.175263 Objective Loss 0.175263 LR 0.000125 Time 0.022086 +2023-10-05 22:07:29,254 - Epoch: [181][ 470/ 1236] Overall Loss 0.175123 Objective Loss 0.175123 LR 0.000125 Time 0.022065 +2023-10-05 22:07:29,469 - Epoch: [181][ 480/ 1236] Overall Loss 0.174993 Objective Loss 0.174993 LR 0.000125 Time 0.022052 +2023-10-05 22:07:29,681 - Epoch: [181][ 490/ 1236] Overall Loss 0.175047 Objective Loss 0.175047 LR 0.000125 Time 0.022034 +2023-10-05 22:07:29,896 - Epoch: [181][ 500/ 1236] Overall Loss 0.175046 Objective Loss 0.175046 LR 0.000125 Time 0.022023 +2023-10-05 22:07:30,108 - Epoch: [181][ 510/ 1236] Overall Loss 0.175050 Objective Loss 0.175050 LR 0.000125 Time 0.022006 +2023-10-05 22:07:30,323 - Epoch: [181][ 520/ 1236] Overall Loss 0.175207 Objective Loss 0.175207 LR 0.000125 Time 0.021995 +2023-10-05 22:07:30,535 - Epoch: [181][ 530/ 1236] Overall Loss 0.174894 Objective Loss 0.174894 LR 0.000125 Time 0.021979 +2023-10-05 22:07:30,749 - Epoch: [181][ 540/ 1236] Overall Loss 0.174932 Objective Loss 0.174932 LR 0.000125 Time 0.021969 +2023-10-05 22:07:30,961 - Epoch: [181][ 550/ 1236] Overall Loss 0.174911 Objective Loss 0.174911 LR 0.000125 Time 0.021954 +2023-10-05 22:07:31,172 - Epoch: [181][ 560/ 1236] Overall Loss 0.174690 Objective Loss 0.174690 LR 0.000125 Time 0.021938 +2023-10-05 22:07:31,378 - Epoch: [181][ 570/ 1236] Overall Loss 0.174651 Objective Loss 0.174651 LR 0.000125 Time 0.021914 +2023-10-05 22:07:31,582 - Epoch: [181][ 580/ 1236] Overall Loss 0.174329 Objective Loss 0.174329 LR 0.000125 Time 0.021887 +2023-10-05 22:07:31,788 - Epoch: [181][ 590/ 1236] Overall Loss 0.174160 Objective Loss 0.174160 LR 0.000125 Time 0.021865 +2023-10-05 22:07:31,990 - Epoch: [181][ 600/ 1236] Overall Loss 0.173740 Objective Loss 0.173740 LR 0.000125 Time 0.021837 +2023-10-05 22:07:32,193 - Epoch: [181][ 610/ 1236] Overall Loss 0.173392 Objective Loss 0.173392 LR 0.000125 Time 0.021810 +2023-10-05 22:07:32,396 - Epoch: [181][ 620/ 1236] Overall Loss 0.173423 Objective Loss 0.173423 LR 0.000125 Time 0.021785 +2023-10-05 22:07:32,599 - Epoch: [181][ 630/ 1236] Overall Loss 0.173482 Objective Loss 0.173482 LR 0.000125 Time 0.021761 +2023-10-05 22:07:32,802 - Epoch: [181][ 640/ 1236] Overall Loss 0.173544 Objective Loss 0.173544 LR 0.000125 Time 0.021738 +2023-10-05 22:07:33,004 - Epoch: [181][ 650/ 1236] Overall Loss 0.173624 Objective Loss 0.173624 LR 0.000125 Time 0.021714 +2023-10-05 22:07:33,207 - Epoch: [181][ 660/ 1236] Overall Loss 0.173587 Objective Loss 0.173587 LR 0.000125 Time 0.021692 +2023-10-05 22:07:33,409 - Epoch: [181][ 670/ 1236] Overall Loss 0.173741 Objective Loss 0.173741 LR 0.000125 Time 0.021669 +2023-10-05 22:07:33,611 - Epoch: [181][ 680/ 1236] Overall Loss 0.174152 Objective Loss 0.174152 LR 0.000125 Time 0.021647 +2023-10-05 22:07:33,814 - Epoch: [181][ 690/ 1236] Overall Loss 0.174460 Objective Loss 0.174460 LR 0.000125 Time 0.021627 +2023-10-05 22:07:34,016 - Epoch: [181][ 700/ 1236] Overall Loss 0.174205 Objective Loss 0.174205 LR 0.000125 Time 0.021606 +2023-10-05 22:07:34,219 - Epoch: [181][ 710/ 1236] Overall Loss 0.174388 Objective Loss 0.174388 LR 0.000125 Time 0.021587 +2023-10-05 22:07:34,421 - Epoch: [181][ 720/ 1236] Overall Loss 0.174033 Objective Loss 0.174033 LR 0.000125 Time 0.021568 +2023-10-05 22:07:34,624 - Epoch: [181][ 730/ 1236] Overall Loss 0.174373 Objective Loss 0.174373 LR 0.000125 Time 0.021550 +2023-10-05 22:07:34,826 - Epoch: [181][ 740/ 1236] Overall Loss 0.174645 Objective Loss 0.174645 LR 0.000125 Time 0.021531 +2023-10-05 22:07:35,029 - Epoch: [181][ 750/ 1236] Overall Loss 0.174793 Objective Loss 0.174793 LR 0.000125 Time 0.021514 +2023-10-05 22:07:35,231 - Epoch: [181][ 760/ 1236] Overall Loss 0.174743 Objective Loss 0.174743 LR 0.000125 Time 0.021497 +2023-10-05 22:07:35,434 - Epoch: [181][ 770/ 1236] Overall Loss 0.174476 Objective Loss 0.174476 LR 0.000125 Time 0.021480 +2023-10-05 22:07:35,636 - Epoch: [181][ 780/ 1236] Overall Loss 0.174516 Objective Loss 0.174516 LR 0.000125 Time 0.021464 +2023-10-05 22:07:35,839 - Epoch: [181][ 790/ 1236] Overall Loss 0.174667 Objective Loss 0.174667 LR 0.000125 Time 0.021448 +2023-10-05 22:07:36,041 - Epoch: [181][ 800/ 1236] Overall Loss 0.174666 Objective Loss 0.174666 LR 0.000125 Time 0.021433 +2023-10-05 22:07:36,245 - Epoch: [181][ 810/ 1236] Overall Loss 0.174596 Objective Loss 0.174596 LR 0.000125 Time 0.021419 +2023-10-05 22:07:36,447 - Epoch: [181][ 820/ 1236] Overall Loss 0.174823 Objective Loss 0.174823 LR 0.000125 Time 0.021404 +2023-10-05 22:07:36,650 - Epoch: [181][ 830/ 1236] Overall Loss 0.174763 Objective Loss 0.174763 LR 0.000125 Time 0.021390 +2023-10-05 22:07:36,852 - Epoch: [181][ 840/ 1236] Overall Loss 0.174641 Objective Loss 0.174641 LR 0.000125 Time 0.021376 +2023-10-05 22:07:37,055 - Epoch: [181][ 850/ 1236] Overall Loss 0.174723 Objective Loss 0.174723 LR 0.000125 Time 0.021363 +2023-10-05 22:07:37,257 - Epoch: [181][ 860/ 1236] Overall Loss 0.174459 Objective Loss 0.174459 LR 0.000125 Time 0.021349 +2023-10-05 22:07:37,460 - Epoch: [181][ 870/ 1236] Overall Loss 0.174594 Objective Loss 0.174594 LR 0.000125 Time 0.021336 +2023-10-05 22:07:37,663 - Epoch: [181][ 880/ 1236] Overall Loss 0.174670 Objective Loss 0.174670 LR 0.000125 Time 0.021324 +2023-10-05 22:07:37,865 - Epoch: [181][ 890/ 1236] Overall Loss 0.174536 Objective Loss 0.174536 LR 0.000125 Time 0.021311 +2023-10-05 22:07:38,067 - Epoch: [181][ 900/ 1236] Overall Loss 0.174472 Objective Loss 0.174472 LR 0.000125 Time 0.021299 +2023-10-05 22:07:38,270 - Epoch: [181][ 910/ 1236] Overall Loss 0.174447 Objective Loss 0.174447 LR 0.000125 Time 0.021287 +2023-10-05 22:07:38,472 - Epoch: [181][ 920/ 1236] Overall Loss 0.174180 Objective Loss 0.174180 LR 0.000125 Time 0.021275 +2023-10-05 22:07:38,675 - Epoch: [181][ 930/ 1236] Overall Loss 0.174440 Objective Loss 0.174440 LR 0.000125 Time 0.021264 +2023-10-05 22:07:38,877 - Epoch: [181][ 940/ 1236] Overall Loss 0.174569 Objective Loss 0.174569 LR 0.000125 Time 0.021253 +2023-10-05 22:07:39,080 - Epoch: [181][ 950/ 1236] Overall Loss 0.174779 Objective Loss 0.174779 LR 0.000125 Time 0.021242 +2023-10-05 22:07:39,285 - Epoch: [181][ 960/ 1236] Overall Loss 0.174841 Objective Loss 0.174841 LR 0.000125 Time 0.021233 +2023-10-05 22:07:39,489 - Epoch: [181][ 970/ 1236] Overall Loss 0.174978 Objective Loss 0.174978 LR 0.000125 Time 0.021225 +2023-10-05 22:07:39,695 - Epoch: [181][ 980/ 1236] Overall Loss 0.175160 Objective Loss 0.175160 LR 0.000125 Time 0.021217 +2023-10-05 22:07:39,899 - Epoch: [181][ 990/ 1236] Overall Loss 0.175049 Objective Loss 0.175049 LR 0.000125 Time 0.021210 +2023-10-05 22:07:40,105 - Epoch: [181][ 1000/ 1236] Overall Loss 0.175154 Objective Loss 0.175154 LR 0.000125 Time 0.021202 +2023-10-05 22:07:40,309 - Epoch: [181][ 1010/ 1236] Overall Loss 0.175262 Objective Loss 0.175262 LR 0.000125 Time 0.021195 +2023-10-05 22:07:40,515 - Epoch: [181][ 1020/ 1236] Overall Loss 0.175342 Objective Loss 0.175342 LR 0.000125 Time 0.021188 +2023-10-05 22:07:40,719 - Epoch: [181][ 1030/ 1236] Overall Loss 0.175340 Objective Loss 0.175340 LR 0.000125 Time 0.021180 +2023-10-05 22:07:40,924 - Epoch: [181][ 1040/ 1236] Overall Loss 0.175415 Objective Loss 0.175415 LR 0.000125 Time 0.021173 +2023-10-05 22:07:41,128 - Epoch: [181][ 1050/ 1236] Overall Loss 0.175333 Objective Loss 0.175333 LR 0.000125 Time 0.021166 +2023-10-05 22:07:41,334 - Epoch: [181][ 1060/ 1236] Overall Loss 0.175359 Objective Loss 0.175359 LR 0.000125 Time 0.021160 +2023-10-05 22:07:41,537 - Epoch: [181][ 1070/ 1236] Overall Loss 0.175241 Objective Loss 0.175241 LR 0.000125 Time 0.021151 +2023-10-05 22:07:41,742 - Epoch: [181][ 1080/ 1236] Overall Loss 0.175494 Objective Loss 0.175494 LR 0.000125 Time 0.021144 +2023-10-05 22:07:41,945 - Epoch: [181][ 1090/ 1236] Overall Loss 0.175352 Objective Loss 0.175352 LR 0.000125 Time 0.021136 +2023-10-05 22:07:42,150 - Epoch: [181][ 1100/ 1236] Overall Loss 0.175470 Objective Loss 0.175470 LR 0.000125 Time 0.021131 +2023-10-05 22:07:42,353 - Epoch: [181][ 1110/ 1236] Overall Loss 0.175489 Objective Loss 0.175489 LR 0.000125 Time 0.021123 +2023-10-05 22:07:42,558 - Epoch: [181][ 1120/ 1236] Overall Loss 0.175575 Objective Loss 0.175575 LR 0.000125 Time 0.021117 +2023-10-05 22:07:42,761 - Epoch: [181][ 1130/ 1236] Overall Loss 0.175451 Objective Loss 0.175451 LR 0.000125 Time 0.021109 +2023-10-05 22:07:42,966 - Epoch: [181][ 1140/ 1236] Overall Loss 0.175400 Objective Loss 0.175400 LR 0.000125 Time 0.021103 +2023-10-05 22:07:43,169 - Epoch: [181][ 1150/ 1236] Overall Loss 0.175277 Objective Loss 0.175277 LR 0.000125 Time 0.021096 +2023-10-05 22:07:43,374 - Epoch: [181][ 1160/ 1236] Overall Loss 0.175220 Objective Loss 0.175220 LR 0.000125 Time 0.021091 +2023-10-05 22:07:43,577 - Epoch: [181][ 1170/ 1236] Overall Loss 0.175276 Objective Loss 0.175276 LR 0.000125 Time 0.021084 +2023-10-05 22:07:43,782 - Epoch: [181][ 1180/ 1236] Overall Loss 0.175169 Objective Loss 0.175169 LR 0.000125 Time 0.021078 +2023-10-05 22:07:43,985 - Epoch: [181][ 1190/ 1236] Overall Loss 0.175212 Objective Loss 0.175212 LR 0.000125 Time 0.021071 +2023-10-05 22:07:44,190 - Epoch: [181][ 1200/ 1236] Overall Loss 0.175234 Objective Loss 0.175234 LR 0.000125 Time 0.021066 +2023-10-05 22:07:44,392 - Epoch: [181][ 1210/ 1236] Overall Loss 0.175155 Objective Loss 0.175155 LR 0.000125 Time 0.021059 +2023-10-05 22:07:44,598 - Epoch: [181][ 1220/ 1236] Overall Loss 0.174977 Objective Loss 0.174977 LR 0.000125 Time 0.021054 +2023-10-05 22:07:44,856 - Epoch: [181][ 1230/ 1236] Overall Loss 0.174900 Objective Loss 0.174900 LR 0.000125 Time 0.021093 +2023-10-05 22:07:44,975 - Epoch: [181][ 1236/ 1236] Overall Loss 0.174875 Objective Loss 0.174875 Top1 88.594705 Top5 98.167006 LR 0.000125 Time 0.021086 +2023-10-05 22:07:45,113 - --- validate (epoch=181)----------- +2023-10-05 22:07:45,114 - 29943 samples (256 per mini-batch) +2023-10-05 22:07:45,568 - Epoch: [181][ 10/ 117] Loss 0.310334 Top1 85.781250 Top5 98.203125 +2023-10-05 22:07:45,715 - Epoch: [181][ 20/ 117] Loss 0.311681 Top1 86.074219 Top5 98.222656 +2023-10-05 22:07:45,862 - Epoch: [181][ 30/ 117] Loss 0.312919 Top1 85.833333 Top5 98.229167 +2023-10-05 22:07:46,010 - Epoch: [181][ 40/ 117] Loss 0.304400 Top1 85.800781 Top5 98.242188 +2023-10-05 22:07:46,158 - Epoch: [181][ 50/ 117] Loss 0.303016 Top1 85.656250 Top5 98.179688 +2023-10-05 22:07:46,305 - Epoch: [181][ 60/ 117] Loss 0.300700 Top1 85.774740 Top5 98.157552 +2023-10-05 22:07:46,453 - Epoch: [181][ 70/ 117] Loss 0.300314 Top1 85.770089 Top5 98.197545 +2023-10-05 22:07:46,600 - Epoch: [181][ 80/ 117] Loss 0.301370 Top1 85.742188 Top5 98.217773 +2023-10-05 22:07:46,746 - Epoch: [181][ 90/ 117] Loss 0.300438 Top1 85.868056 Top5 98.194444 +2023-10-05 22:07:46,895 - Epoch: [181][ 100/ 117] Loss 0.299617 Top1 85.851562 Top5 98.234375 +2023-10-05 22:07:47,055 - Epoch: [181][ 110/ 117] Loss 0.302090 Top1 85.838068 Top5 98.210227 +2023-10-05 22:07:47,141 - Epoch: [181][ 117/ 117] Loss 0.304350 Top1 85.803026 Top5 98.206593 +2023-10-05 22:07:47,276 - ==> Top1: 85.803 Top5: 98.207 Loss: 0.304 + +2023-10-05 22:07:47,276 - ==> Confusion: +[[ 940 3 4 1 7 4 0 0 5 60 1 0 0 2 5 1 3 1 0 0 13] + [ 0 1062 3 0 9 14 1 19 0 0 1 1 0 0 0 4 2 0 8 0 7] + [ 4 1 975 11 2 0 20 7 0 0 3 2 7 1 0 4 1 2 6 3 7] + [ 2 1 10 976 1 4 1 2 0 1 5 1 6 1 24 4 0 5 29 3 13] + [ 27 8 1 1 967 3 1 1 0 8 1 2 0 2 5 1 10 2 0 1 9] + [ 3 32 2 1 2 983 0 29 1 0 5 7 0 12 7 2 3 0 5 6 16] + [ 0 4 17 0 0 0 1135 10 0 0 1 1 2 1 1 5 0 0 2 7 5] + [ 4 15 12 0 1 22 3 1094 1 4 3 6 2 3 1 2 0 0 34 5 6] + [ 19 2 0 0 3 1 1 0 987 37 9 1 2 8 9 2 1 0 5 0 2] + [ 97 0 3 2 7 2 0 0 17 952 1 1 0 21 2 3 0 2 0 2 7] + [ 4 5 7 4 0 1 1 6 12 1 972 4 0 11 4 1 3 0 8 0 9] + [ 1 0 3 0 1 13 0 1 0 1 0 952 24 4 0 5 2 16 0 9 3] + [ 1 0 2 3 0 1 0 2 1 0 2 34 987 2 0 3 2 15 3 2 8] + [ 2 0 1 0 1 4 0 1 12 16 8 5 1 1050 4 1 1 0 0 1 11] + [ 13 2 3 4 9 0 0 0 24 3 4 1 2 2 1003 0 0 2 13 1 15] + [ 1 3 1 0 3 0 2 0 0 0 0 4 6 4 1 1076 12 10 0 6 5] + [ 1 12 1 0 4 5 0 0 1 0 0 5 0 0 2 10 1100 0 0 4 16] + [ 0 0 0 2 0 0 2 0 0 1 0 1 16 2 0 7 0 1003 1 0 3] + [ 0 4 6 18 1 0 1 25 1 0 1 0 2 1 11 0 1 0 989 1 6] + [ 0 4 5 3 2 6 6 10 0 0 1 13 3 1 0 5 5 2 3 1069 14] + [ 123 169 151 49 91 125 40 96 82 63 164 95 296 270 135 57 122 62 144 151 5420]] + +2023-10-05 22:07:47,278 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:07:47,278 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:07:47,284 - + +2023-10-05 22:07:47,284 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:07:48,287 - Epoch: [182][ 10/ 1236] Overall Loss 0.174358 Objective Loss 0.174358 LR 0.000125 Time 0.100242 +2023-10-05 22:07:48,491 - Epoch: [182][ 20/ 1236] Overall Loss 0.173368 Objective Loss 0.173368 LR 0.000125 Time 0.060319 +2023-10-05 22:07:48,693 - Epoch: [182][ 30/ 1236] Overall Loss 0.167743 Objective Loss 0.167743 LR 0.000125 Time 0.046944 +2023-10-05 22:07:48,898 - Epoch: [182][ 40/ 1236] Overall Loss 0.162549 Objective Loss 0.162549 LR 0.000125 Time 0.040307 +2023-10-05 22:07:49,100 - Epoch: [182][ 50/ 1236] Overall Loss 0.165108 Objective Loss 0.165108 LR 0.000125 Time 0.036283 +2023-10-05 22:07:49,305 - Epoch: [182][ 60/ 1236] Overall Loss 0.166668 Objective Loss 0.166668 LR 0.000125 Time 0.033642 +2023-10-05 22:07:49,507 - Epoch: [182][ 70/ 1236] Overall Loss 0.168540 Objective Loss 0.168540 LR 0.000125 Time 0.031721 +2023-10-05 22:07:49,711 - Epoch: [182][ 80/ 1236] Overall Loss 0.169370 Objective Loss 0.169370 LR 0.000125 Time 0.030305 +2023-10-05 22:07:49,914 - Epoch: [182][ 90/ 1236] Overall Loss 0.168817 Objective Loss 0.168817 LR 0.000125 Time 0.029186 +2023-10-05 22:07:50,117 - Epoch: [182][ 100/ 1236] Overall Loss 0.168914 Objective Loss 0.168914 LR 0.000125 Time 0.028297 +2023-10-05 22:07:50,322 - Epoch: [182][ 110/ 1236] Overall Loss 0.168872 Objective Loss 0.168872 LR 0.000125 Time 0.027580 +2023-10-05 22:07:50,527 - Epoch: [182][ 120/ 1236] Overall Loss 0.169444 Objective Loss 0.169444 LR 0.000125 Time 0.026987 +2023-10-05 22:07:50,731 - Epoch: [182][ 130/ 1236] Overall Loss 0.169923 Objective Loss 0.169923 LR 0.000125 Time 0.026476 +2023-10-05 22:07:50,936 - Epoch: [182][ 140/ 1236] Overall Loss 0.169360 Objective Loss 0.169360 LR 0.000125 Time 0.026047 +2023-10-05 22:07:51,140 - Epoch: [182][ 150/ 1236] Overall Loss 0.168873 Objective Loss 0.168873 LR 0.000125 Time 0.025672 +2023-10-05 22:07:51,345 - Epoch: [182][ 160/ 1236] Overall Loss 0.169559 Objective Loss 0.169559 LR 0.000125 Time 0.025346 +2023-10-05 22:07:51,549 - Epoch: [182][ 170/ 1236] Overall Loss 0.171524 Objective Loss 0.171524 LR 0.000125 Time 0.025052 +2023-10-05 22:07:51,754 - Epoch: [182][ 180/ 1236] Overall Loss 0.171744 Objective Loss 0.171744 LR 0.000125 Time 0.024798 +2023-10-05 22:07:51,959 - Epoch: [182][ 190/ 1236] Overall Loss 0.171305 Objective Loss 0.171305 LR 0.000125 Time 0.024567 +2023-10-05 22:07:52,164 - Epoch: [182][ 200/ 1236] Overall Loss 0.171725 Objective Loss 0.171725 LR 0.000125 Time 0.024363 +2023-10-05 22:07:52,368 - Epoch: [182][ 210/ 1236] Overall Loss 0.170654 Objective Loss 0.170654 LR 0.000125 Time 0.024175 +2023-10-05 22:07:52,573 - Epoch: [182][ 220/ 1236] Overall Loss 0.170098 Objective Loss 0.170098 LR 0.000125 Time 0.024006 +2023-10-05 22:07:52,777 - Epoch: [182][ 230/ 1236] Overall Loss 0.170216 Objective Loss 0.170216 LR 0.000125 Time 0.023849 +2023-10-05 22:07:52,982 - Epoch: [182][ 240/ 1236] Overall Loss 0.170042 Objective Loss 0.170042 LR 0.000125 Time 0.023708 +2023-10-05 22:07:53,187 - Epoch: [182][ 250/ 1236] Overall Loss 0.170491 Objective Loss 0.170491 LR 0.000125 Time 0.023576 +2023-10-05 22:07:53,392 - Epoch: [182][ 260/ 1236] Overall Loss 0.171487 Objective Loss 0.171487 LR 0.000125 Time 0.023456 +2023-10-05 22:07:53,596 - Epoch: [182][ 270/ 1236] Overall Loss 0.171092 Objective Loss 0.171092 LR 0.000125 Time 0.023342 +2023-10-05 22:07:53,800 - Epoch: [182][ 280/ 1236] Overall Loss 0.171309 Objective Loss 0.171309 LR 0.000125 Time 0.023237 +2023-10-05 22:07:54,004 - Epoch: [182][ 290/ 1236] Overall Loss 0.171129 Objective Loss 0.171129 LR 0.000125 Time 0.023138 +2023-10-05 22:07:54,209 - Epoch: [182][ 300/ 1236] Overall Loss 0.170901 Objective Loss 0.170901 LR 0.000125 Time 0.023048 +2023-10-05 22:07:54,414 - Epoch: [182][ 310/ 1236] Overall Loss 0.170642 Objective Loss 0.170642 LR 0.000125 Time 0.022963 +2023-10-05 22:07:54,619 - Epoch: [182][ 320/ 1236] Overall Loss 0.170339 Objective Loss 0.170339 LR 0.000125 Time 0.022886 +2023-10-05 22:07:54,823 - Epoch: [182][ 330/ 1236] Overall Loss 0.170180 Objective Loss 0.170180 LR 0.000125 Time 0.022811 +2023-10-05 22:07:55,028 - Epoch: [182][ 340/ 1236] Overall Loss 0.170640 Objective Loss 0.170640 LR 0.000125 Time 0.022742 +2023-10-05 22:07:55,232 - Epoch: [182][ 350/ 1236] Overall Loss 0.171315 Objective Loss 0.171315 LR 0.000125 Time 0.022675 +2023-10-05 22:07:55,438 - Epoch: [182][ 360/ 1236] Overall Loss 0.171528 Objective Loss 0.171528 LR 0.000125 Time 0.022614 +2023-10-05 22:07:55,640 - Epoch: [182][ 370/ 1236] Overall Loss 0.171871 Objective Loss 0.171871 LR 0.000125 Time 0.022549 +2023-10-05 22:07:55,844 - Epoch: [182][ 380/ 1236] Overall Loss 0.172378 Objective Loss 0.172378 LR 0.000125 Time 0.022491 +2023-10-05 22:07:56,045 - Epoch: [182][ 390/ 1236] Overall Loss 0.172447 Objective Loss 0.172447 LR 0.000125 Time 0.022430 +2023-10-05 22:07:56,249 - Epoch: [182][ 400/ 1236] Overall Loss 0.172021 Objective Loss 0.172021 LR 0.000125 Time 0.022377 +2023-10-05 22:07:56,450 - Epoch: [182][ 410/ 1236] Overall Loss 0.172178 Objective Loss 0.172178 LR 0.000125 Time 0.022321 +2023-10-05 22:07:56,655 - Epoch: [182][ 420/ 1236] Overall Loss 0.172050 Objective Loss 0.172050 LR 0.000125 Time 0.022276 +2023-10-05 22:07:56,858 - Epoch: [182][ 430/ 1236] Overall Loss 0.172000 Objective Loss 0.172000 LR 0.000125 Time 0.022230 +2023-10-05 22:07:57,062 - Epoch: [182][ 440/ 1236] Overall Loss 0.172395 Objective Loss 0.172395 LR 0.000125 Time 0.022188 +2023-10-05 22:07:57,266 - Epoch: [182][ 450/ 1236] Overall Loss 0.172570 Objective Loss 0.172570 LR 0.000125 Time 0.022147 +2023-10-05 22:07:57,470 - Epoch: [182][ 460/ 1236] Overall Loss 0.172518 Objective Loss 0.172518 LR 0.000125 Time 0.022107 +2023-10-05 22:07:57,670 - Epoch: [182][ 470/ 1236] Overall Loss 0.172711 Objective Loss 0.172711 LR 0.000125 Time 0.022063 +2023-10-05 22:07:57,872 - Epoch: [182][ 480/ 1236] Overall Loss 0.172301 Objective Loss 0.172301 LR 0.000125 Time 0.022024 +2023-10-05 22:07:58,075 - Epoch: [182][ 490/ 1236] Overall Loss 0.172470 Objective Loss 0.172470 LR 0.000125 Time 0.021987 +2023-10-05 22:07:58,278 - Epoch: [182][ 500/ 1236] Overall Loss 0.172511 Objective Loss 0.172511 LR 0.000125 Time 0.021952 +2023-10-05 22:07:58,481 - Epoch: [182][ 510/ 1236] Overall Loss 0.172477 Objective Loss 0.172477 LR 0.000125 Time 0.021919 +2023-10-05 22:07:58,683 - Epoch: [182][ 520/ 1236] Overall Loss 0.173081 Objective Loss 0.173081 LR 0.000125 Time 0.021886 +2023-10-05 22:07:58,885 - Epoch: [182][ 530/ 1236] Overall Loss 0.172603 Objective Loss 0.172603 LR 0.000125 Time 0.021854 +2023-10-05 22:07:59,088 - Epoch: [182][ 540/ 1236] Overall Loss 0.172535 Objective Loss 0.172535 LR 0.000125 Time 0.021824 +2023-10-05 22:07:59,290 - Epoch: [182][ 550/ 1236] Overall Loss 0.172223 Objective Loss 0.172223 LR 0.000125 Time 0.021795 +2023-10-05 22:07:59,493 - Epoch: [182][ 560/ 1236] Overall Loss 0.172623 Objective Loss 0.172623 LR 0.000125 Time 0.021767 +2023-10-05 22:07:59,695 - Epoch: [182][ 570/ 1236] Overall Loss 0.173008 Objective Loss 0.173008 LR 0.000125 Time 0.021739 +2023-10-05 22:07:59,898 - Epoch: [182][ 580/ 1236] Overall Loss 0.172873 Objective Loss 0.172873 LR 0.000125 Time 0.021713 +2023-10-05 22:08:00,101 - Epoch: [182][ 590/ 1236] Overall Loss 0.172939 Objective Loss 0.172939 LR 0.000125 Time 0.021689 +2023-10-05 22:08:00,307 - Epoch: [182][ 600/ 1236] Overall Loss 0.172648 Objective Loss 0.172648 LR 0.000125 Time 0.021670 +2023-10-05 22:08:00,514 - Epoch: [182][ 610/ 1236] Overall Loss 0.172734 Objective Loss 0.172734 LR 0.000125 Time 0.021654 +2023-10-05 22:08:00,721 - Epoch: [182][ 620/ 1236] Overall Loss 0.172991 Objective Loss 0.172991 LR 0.000125 Time 0.021638 +2023-10-05 22:08:00,927 - Epoch: [182][ 630/ 1236] Overall Loss 0.172902 Objective Loss 0.172902 LR 0.000125 Time 0.021620 +2023-10-05 22:08:01,132 - Epoch: [182][ 640/ 1236] Overall Loss 0.172567 Objective Loss 0.172567 LR 0.000125 Time 0.021603 +2023-10-05 22:08:01,340 - Epoch: [182][ 650/ 1236] Overall Loss 0.172430 Objective Loss 0.172430 LR 0.000125 Time 0.021589 +2023-10-05 22:08:01,546 - Epoch: [182][ 660/ 1236] Overall Loss 0.172377 Objective Loss 0.172377 LR 0.000125 Time 0.021573 +2023-10-05 22:08:01,753 - Epoch: [182][ 670/ 1236] Overall Loss 0.172570 Objective Loss 0.172570 LR 0.000125 Time 0.021560 +2023-10-05 22:08:01,960 - Epoch: [182][ 680/ 1236] Overall Loss 0.172633 Objective Loss 0.172633 LR 0.000125 Time 0.021547 +2023-10-05 22:08:02,168 - Epoch: [182][ 690/ 1236] Overall Loss 0.172186 Objective Loss 0.172186 LR 0.000125 Time 0.021535 +2023-10-05 22:08:02,375 - Epoch: [182][ 700/ 1236] Overall Loss 0.172153 Objective Loss 0.172153 LR 0.000125 Time 0.021522 +2023-10-05 22:08:02,582 - Epoch: [182][ 710/ 1236] Overall Loss 0.172458 Objective Loss 0.172458 LR 0.000125 Time 0.021511 +2023-10-05 22:08:02,789 - Epoch: [182][ 720/ 1236] Overall Loss 0.172405 Objective Loss 0.172405 LR 0.000125 Time 0.021498 +2023-10-05 22:08:02,996 - Epoch: [182][ 730/ 1236] Overall Loss 0.172203 Objective Loss 0.172203 LR 0.000125 Time 0.021487 +2023-10-05 22:08:03,203 - Epoch: [182][ 740/ 1236] Overall Loss 0.172267 Objective Loss 0.172267 LR 0.000125 Time 0.021476 +2023-10-05 22:08:03,411 - Epoch: [182][ 750/ 1236] Overall Loss 0.172189 Objective Loss 0.172189 LR 0.000125 Time 0.021466 +2023-10-05 22:08:03,618 - Epoch: [182][ 760/ 1236] Overall Loss 0.172106 Objective Loss 0.172106 LR 0.000125 Time 0.021455 +2023-10-05 22:08:03,825 - Epoch: [182][ 770/ 1236] Overall Loss 0.172181 Objective Loss 0.172181 LR 0.000125 Time 0.021445 +2023-10-05 22:08:04,032 - Epoch: [182][ 780/ 1236] Overall Loss 0.172249 Objective Loss 0.172249 LR 0.000125 Time 0.021435 +2023-10-05 22:08:04,240 - Epoch: [182][ 790/ 1236] Overall Loss 0.172267 Objective Loss 0.172267 LR 0.000125 Time 0.021427 +2023-10-05 22:08:04,447 - Epoch: [182][ 800/ 1236] Overall Loss 0.172254 Objective Loss 0.172254 LR 0.000125 Time 0.021417 +2023-10-05 22:08:04,655 - Epoch: [182][ 810/ 1236] Overall Loss 0.172542 Objective Loss 0.172542 LR 0.000125 Time 0.021409 +2023-10-05 22:08:04,862 - Epoch: [182][ 820/ 1236] Overall Loss 0.172555 Objective Loss 0.172555 LR 0.000125 Time 0.021400 +2023-10-05 22:08:05,068 - Epoch: [182][ 830/ 1236] Overall Loss 0.172628 Objective Loss 0.172628 LR 0.000125 Time 0.021390 +2023-10-05 22:08:05,274 - Epoch: [182][ 840/ 1236] Overall Loss 0.172561 Objective Loss 0.172561 LR 0.000125 Time 0.021380 +2023-10-05 22:08:05,480 - Epoch: [182][ 850/ 1236] Overall Loss 0.172684 Objective Loss 0.172684 LR 0.000125 Time 0.021370 +2023-10-05 22:08:05,686 - Epoch: [182][ 860/ 1236] Overall Loss 0.172615 Objective Loss 0.172615 LR 0.000125 Time 0.021360 +2023-10-05 22:08:05,892 - Epoch: [182][ 870/ 1236] Overall Loss 0.172292 Objective Loss 0.172292 LR 0.000125 Time 0.021351 +2023-10-05 22:08:06,097 - Epoch: [182][ 880/ 1236] Overall Loss 0.172306 Objective Loss 0.172306 LR 0.000125 Time 0.021341 +2023-10-05 22:08:06,303 - Epoch: [182][ 890/ 1236] Overall Loss 0.172701 Objective Loss 0.172701 LR 0.000125 Time 0.021332 +2023-10-05 22:08:06,509 - Epoch: [182][ 900/ 1236] Overall Loss 0.172784 Objective Loss 0.172784 LR 0.000125 Time 0.021323 +2023-10-05 22:08:06,714 - Epoch: [182][ 910/ 1236] Overall Loss 0.172973 Objective Loss 0.172973 LR 0.000125 Time 0.021315 +2023-10-05 22:08:06,920 - Epoch: [182][ 920/ 1236] Overall Loss 0.172957 Objective Loss 0.172957 LR 0.000125 Time 0.021306 +2023-10-05 22:08:07,126 - Epoch: [182][ 930/ 1236] Overall Loss 0.173264 Objective Loss 0.173264 LR 0.000125 Time 0.021298 +2023-10-05 22:08:07,332 - Epoch: [182][ 940/ 1236] Overall Loss 0.173199 Objective Loss 0.173199 LR 0.000125 Time 0.021290 +2023-10-05 22:08:07,537 - Epoch: [182][ 950/ 1236] Overall Loss 0.173347 Objective Loss 0.173347 LR 0.000125 Time 0.021282 +2023-10-05 22:08:07,743 - Epoch: [182][ 960/ 1236] Overall Loss 0.173329 Objective Loss 0.173329 LR 0.000125 Time 0.021274 +2023-10-05 22:08:07,949 - Epoch: [182][ 970/ 1236] Overall Loss 0.173141 Objective Loss 0.173141 LR 0.000125 Time 0.021266 +2023-10-05 22:08:08,154 - Epoch: [182][ 980/ 1236] Overall Loss 0.173127 Objective Loss 0.173127 LR 0.000125 Time 0.021259 +2023-10-05 22:08:08,360 - Epoch: [182][ 990/ 1236] Overall Loss 0.173131 Objective Loss 0.173131 LR 0.000125 Time 0.021251 +2023-10-05 22:08:08,566 - Epoch: [182][ 1000/ 1236] Overall Loss 0.173180 Objective Loss 0.173180 LR 0.000125 Time 0.021244 +2023-10-05 22:08:08,772 - Epoch: [182][ 1010/ 1236] Overall Loss 0.173105 Objective Loss 0.173105 LR 0.000125 Time 0.021237 +2023-10-05 22:08:08,978 - Epoch: [182][ 1020/ 1236] Overall Loss 0.173240 Objective Loss 0.173240 LR 0.000125 Time 0.021230 +2023-10-05 22:08:09,183 - Epoch: [182][ 1030/ 1236] Overall Loss 0.173191 Objective Loss 0.173191 LR 0.000125 Time 0.021223 +2023-10-05 22:08:09,389 - Epoch: [182][ 1040/ 1236] Overall Loss 0.173275 Objective Loss 0.173275 LR 0.000125 Time 0.021217 +2023-10-05 22:08:09,595 - Epoch: [182][ 1050/ 1236] Overall Loss 0.173323 Objective Loss 0.173323 LR 0.000125 Time 0.021210 +2023-10-05 22:08:09,800 - Epoch: [182][ 1060/ 1236] Overall Loss 0.173421 Objective Loss 0.173421 LR 0.000125 Time 0.021204 +2023-10-05 22:08:10,006 - Epoch: [182][ 1070/ 1236] Overall Loss 0.173626 Objective Loss 0.173626 LR 0.000125 Time 0.021198 +2023-10-05 22:08:10,212 - Epoch: [182][ 1080/ 1236] Overall Loss 0.173702 Objective Loss 0.173702 LR 0.000125 Time 0.021192 +2023-10-05 22:08:10,418 - Epoch: [182][ 1090/ 1236] Overall Loss 0.173688 Objective Loss 0.173688 LR 0.000125 Time 0.021186 +2023-10-05 22:08:10,624 - Epoch: [182][ 1100/ 1236] Overall Loss 0.173648 Objective Loss 0.173648 LR 0.000125 Time 0.021180 +2023-10-05 22:08:10,830 - Epoch: [182][ 1110/ 1236] Overall Loss 0.173400 Objective Loss 0.173400 LR 0.000125 Time 0.021174 +2023-10-05 22:08:11,036 - Epoch: [182][ 1120/ 1236] Overall Loss 0.173367 Objective Loss 0.173367 LR 0.000125 Time 0.021169 +2023-10-05 22:08:11,241 - Epoch: [182][ 1130/ 1236] Overall Loss 0.173479 Objective Loss 0.173479 LR 0.000125 Time 0.021162 +2023-10-05 22:08:11,447 - Epoch: [182][ 1140/ 1236] Overall Loss 0.173554 Objective Loss 0.173554 LR 0.000125 Time 0.021157 +2023-10-05 22:08:11,652 - Epoch: [182][ 1150/ 1236] Overall Loss 0.173500 Objective Loss 0.173500 LR 0.000125 Time 0.021152 +2023-10-05 22:08:11,858 - Epoch: [182][ 1160/ 1236] Overall Loss 0.173422 Objective Loss 0.173422 LR 0.000125 Time 0.021146 +2023-10-05 22:08:12,064 - Epoch: [182][ 1170/ 1236] Overall Loss 0.173317 Objective Loss 0.173317 LR 0.000125 Time 0.021141 +2023-10-05 22:08:12,270 - Epoch: [182][ 1180/ 1236] Overall Loss 0.173314 Objective Loss 0.173314 LR 0.000125 Time 0.021136 +2023-10-05 22:08:12,476 - Epoch: [182][ 1190/ 1236] Overall Loss 0.173339 Objective Loss 0.173339 LR 0.000125 Time 0.021131 +2023-10-05 22:08:12,681 - Epoch: [182][ 1200/ 1236] Overall Loss 0.173256 Objective Loss 0.173256 LR 0.000125 Time 0.021126 +2023-10-05 22:08:12,887 - Epoch: [182][ 1210/ 1236] Overall Loss 0.173073 Objective Loss 0.173073 LR 0.000125 Time 0.021121 +2023-10-05 22:08:13,093 - Epoch: [182][ 1220/ 1236] Overall Loss 0.173356 Objective Loss 0.173356 LR 0.000125 Time 0.021116 +2023-10-05 22:08:13,353 - Epoch: [182][ 1230/ 1236] Overall Loss 0.173401 Objective Loss 0.173401 LR 0.000125 Time 0.021156 +2023-10-05 22:08:13,472 - Epoch: [182][ 1236/ 1236] Overall Loss 0.173401 Objective Loss 0.173401 Top1 90.224033 Top5 98.574338 LR 0.000125 Time 0.021150 +2023-10-05 22:08:13,592 - --- validate (epoch=182)----------- +2023-10-05 22:08:13,592 - 29943 samples (256 per mini-batch) +2023-10-05 22:08:14,041 - Epoch: [182][ 10/ 117] Loss 0.339590 Top1 85.117188 Top5 98.164062 +2023-10-05 22:08:14,189 - Epoch: [182][ 20/ 117] Loss 0.315761 Top1 85.722656 Top5 98.320312 +2023-10-05 22:08:14,338 - Epoch: [182][ 30/ 117] Loss 0.307425 Top1 85.572917 Top5 98.307292 +2023-10-05 22:08:14,486 - Epoch: [182][ 40/ 117] Loss 0.299717 Top1 85.859375 Top5 98.310547 +2023-10-05 22:08:14,635 - Epoch: [182][ 50/ 117] Loss 0.293463 Top1 85.992188 Top5 98.359375 +2023-10-05 22:08:14,783 - Epoch: [182][ 60/ 117] Loss 0.295063 Top1 85.983073 Top5 98.339844 +2023-10-05 22:08:14,932 - Epoch: [182][ 70/ 117] Loss 0.296593 Top1 85.970982 Top5 98.331473 +2023-10-05 22:08:15,076 - Epoch: [182][ 80/ 117] Loss 0.300080 Top1 85.747070 Top5 98.256836 +2023-10-05 22:08:15,222 - Epoch: [182][ 90/ 117] Loss 0.296501 Top1 85.876736 Top5 98.276910 +2023-10-05 22:08:15,370 - Epoch: [182][ 100/ 117] Loss 0.298377 Top1 85.781250 Top5 98.285156 +2023-10-05 22:08:15,524 - Epoch: [182][ 110/ 117] Loss 0.301429 Top1 85.742188 Top5 98.252841 +2023-10-05 22:08:15,609 - Epoch: [182][ 117/ 117] Loss 0.301174 Top1 85.829743 Top5 98.273386 +2023-10-05 22:08:15,725 - ==> Top1: 85.830 Top5: 98.273 Loss: 0.301 + +2023-10-05 22:08:15,726 - ==> Confusion: +[[ 934 2 2 0 7 4 0 0 7 66 1 0 0 2 4 2 5 0 0 0 14] + [ 0 1062 2 0 7 22 1 15 0 0 0 2 0 0 0 3 2 1 8 0 6] + [ 2 2 970 10 2 1 23 7 0 0 4 3 6 1 1 3 1 3 4 3 10] + [ 0 1 8 979 2 4 4 1 1 1 4 0 9 2 25 3 1 4 23 1 16] + [ 26 6 0 0 970 3 1 1 0 9 0 2 0 1 6 1 13 2 0 2 7] + [ 3 30 1 2 5 998 1 19 0 0 3 8 0 10 7 2 3 0 2 2 20] + [ 0 5 21 0 0 0 1135 8 0 0 1 1 1 0 1 7 0 0 1 6 4] + [ 4 18 12 0 3 28 3 1075 1 2 1 10 1 4 0 2 0 0 39 5 10] + [ 19 2 0 0 2 2 0 1 970 42 10 2 1 11 14 2 1 1 5 0 4] + [ 96 1 1 1 3 2 0 0 20 949 0 0 0 23 6 4 0 1 0 3 9] + [ 3 5 8 3 0 1 4 3 12 0 972 2 1 12 4 1 4 0 5 2 11] + [ 0 0 2 0 1 11 0 2 0 0 0 961 21 3 0 3 2 17 0 7 5] + [ 1 1 3 6 1 1 0 1 0 0 1 35 987 1 2 3 3 11 3 2 6] + [ 3 0 1 0 2 8 0 0 10 12 7 6 2 1053 3 1 1 0 0 1 9] + [ 13 2 2 3 4 1 0 0 29 3 4 1 2 1 1008 0 2 1 13 0 12] + [ 1 2 1 0 2 0 1 0 0 0 0 8 7 1 1 1076 15 9 0 7 3] + [ 1 12 1 0 5 5 0 0 1 0 0 4 1 0 2 10 1100 0 0 2 17] + [ 1 0 0 0 0 0 3 0 0 0 0 2 17 2 0 4 0 1001 1 0 7] + [ 1 4 4 18 1 0 0 22 1 0 1 0 2 1 9 0 1 0 994 1 8] + [ 1 3 3 3 2 7 8 8 0 0 1 14 3 1 0 5 10 1 4 1067 11] + [ 118 166 144 49 95 134 36 78 85 68 175 89 314 266 137 50 125 69 123 145 5439]] + +2023-10-05 22:08:15,727 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:08:15,727 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:08:15,733 - + +2023-10-05 22:08:15,733 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:08:16,725 - Epoch: [183][ 10/ 1236] Overall Loss 0.177074 Objective Loss 0.177074 LR 0.000125 Time 0.099142 +2023-10-05 22:08:16,930 - Epoch: [183][ 20/ 1236] Overall Loss 0.164468 Objective Loss 0.164468 LR 0.000125 Time 0.059788 +2023-10-05 22:08:17,132 - Epoch: [183][ 30/ 1236] Overall Loss 0.166482 Objective Loss 0.166482 LR 0.000125 Time 0.046596 +2023-10-05 22:08:17,337 - Epoch: [183][ 40/ 1236] Overall Loss 0.167103 Objective Loss 0.167103 LR 0.000125 Time 0.040067 +2023-10-05 22:08:17,540 - Epoch: [183][ 50/ 1236] Overall Loss 0.166183 Objective Loss 0.166183 LR 0.000125 Time 0.036107 +2023-10-05 22:08:17,745 - Epoch: [183][ 60/ 1236] Overall Loss 0.165419 Objective Loss 0.165419 LR 0.000125 Time 0.033486 +2023-10-05 22:08:17,949 - Epoch: [183][ 70/ 1236] Overall Loss 0.168442 Objective Loss 0.168442 LR 0.000125 Time 0.031621 +2023-10-05 22:08:18,154 - Epoch: [183][ 80/ 1236] Overall Loss 0.169517 Objective Loss 0.169517 LR 0.000125 Time 0.030217 +2023-10-05 22:08:18,358 - Epoch: [183][ 90/ 1236] Overall Loss 0.169285 Objective Loss 0.169285 LR 0.000125 Time 0.029124 +2023-10-05 22:08:18,562 - Epoch: [183][ 100/ 1236] Overall Loss 0.170873 Objective Loss 0.170873 LR 0.000125 Time 0.028249 +2023-10-05 22:08:18,766 - Epoch: [183][ 110/ 1236] Overall Loss 0.171396 Objective Loss 0.171396 LR 0.000125 Time 0.027534 +2023-10-05 22:08:18,971 - Epoch: [183][ 120/ 1236] Overall Loss 0.170742 Objective Loss 0.170742 LR 0.000125 Time 0.026947 +2023-10-05 22:08:19,175 - Epoch: [183][ 130/ 1236] Overall Loss 0.170595 Objective Loss 0.170595 LR 0.000125 Time 0.026431 +2023-10-05 22:08:19,379 - Epoch: [183][ 140/ 1236] Overall Loss 0.169893 Objective Loss 0.169893 LR 0.000125 Time 0.026000 +2023-10-05 22:08:19,583 - Epoch: [183][ 150/ 1236] Overall Loss 0.170003 Objective Loss 0.170003 LR 0.000125 Time 0.025623 +2023-10-05 22:08:19,784 - Epoch: [183][ 160/ 1236] Overall Loss 0.170164 Objective Loss 0.170164 LR 0.000125 Time 0.025276 +2023-10-05 22:08:19,985 - Epoch: [183][ 170/ 1236] Overall Loss 0.169472 Objective Loss 0.169472 LR 0.000125 Time 0.024969 +2023-10-05 22:08:20,187 - Epoch: [183][ 180/ 1236] Overall Loss 0.170106 Objective Loss 0.170106 LR 0.000125 Time 0.024704 +2023-10-05 22:08:20,388 - Epoch: [183][ 190/ 1236] Overall Loss 0.170538 Objective Loss 0.170538 LR 0.000125 Time 0.024460 +2023-10-05 22:08:20,591 - Epoch: [183][ 200/ 1236] Overall Loss 0.169946 Objective Loss 0.169946 LR 0.000125 Time 0.024246 +2023-10-05 22:08:20,792 - Epoch: [183][ 210/ 1236] Overall Loss 0.169755 Objective Loss 0.169755 LR 0.000125 Time 0.024047 +2023-10-05 22:08:20,993 - Epoch: [183][ 220/ 1236] Overall Loss 0.169626 Objective Loss 0.169626 LR 0.000125 Time 0.023870 +2023-10-05 22:08:21,195 - Epoch: [183][ 230/ 1236] Overall Loss 0.170107 Objective Loss 0.170107 LR 0.000125 Time 0.023705 +2023-10-05 22:08:21,397 - Epoch: [183][ 240/ 1236] Overall Loss 0.170151 Objective Loss 0.170151 LR 0.000125 Time 0.023557 +2023-10-05 22:08:21,598 - Epoch: [183][ 250/ 1236] Overall Loss 0.169167 Objective Loss 0.169167 LR 0.000125 Time 0.023419 +2023-10-05 22:08:21,800 - Epoch: [183][ 260/ 1236] Overall Loss 0.169636 Objective Loss 0.169636 LR 0.000125 Time 0.023294 +2023-10-05 22:08:22,002 - Epoch: [183][ 270/ 1236] Overall Loss 0.169827 Objective Loss 0.169827 LR 0.000125 Time 0.023176 +2023-10-05 22:08:22,204 - Epoch: [183][ 280/ 1236] Overall Loss 0.170090 Objective Loss 0.170090 LR 0.000125 Time 0.023071 +2023-10-05 22:08:22,406 - Epoch: [183][ 290/ 1236] Overall Loss 0.169962 Objective Loss 0.169962 LR 0.000125 Time 0.022969 +2023-10-05 22:08:22,608 - Epoch: [183][ 300/ 1236] Overall Loss 0.170792 Objective Loss 0.170792 LR 0.000125 Time 0.022878 +2023-10-05 22:08:22,809 - Epoch: [183][ 310/ 1236] Overall Loss 0.170349 Objective Loss 0.170349 LR 0.000125 Time 0.022787 +2023-10-05 22:08:23,011 - Epoch: [183][ 320/ 1236] Overall Loss 0.170384 Objective Loss 0.170384 LR 0.000125 Time 0.022706 +2023-10-05 22:08:23,212 - Epoch: [183][ 330/ 1236] Overall Loss 0.170381 Objective Loss 0.170381 LR 0.000125 Time 0.022626 +2023-10-05 22:08:23,414 - Epoch: [183][ 340/ 1236] Overall Loss 0.170340 Objective Loss 0.170340 LR 0.000125 Time 0.022553 +2023-10-05 22:08:23,615 - Epoch: [183][ 350/ 1236] Overall Loss 0.170883 Objective Loss 0.170883 LR 0.000125 Time 0.022482 +2023-10-05 22:08:23,818 - Epoch: [183][ 360/ 1236] Overall Loss 0.170603 Objective Loss 0.170603 LR 0.000125 Time 0.022418 +2023-10-05 22:08:24,019 - Epoch: [183][ 370/ 1236] Overall Loss 0.171107 Objective Loss 0.171107 LR 0.000125 Time 0.022355 +2023-10-05 22:08:24,221 - Epoch: [183][ 380/ 1236] Overall Loss 0.171456 Objective Loss 0.171456 LR 0.000125 Time 0.022298 +2023-10-05 22:08:24,422 - Epoch: [183][ 390/ 1236] Overall Loss 0.171605 Objective Loss 0.171605 LR 0.000125 Time 0.022242 +2023-10-05 22:08:24,625 - Epoch: [183][ 400/ 1236] Overall Loss 0.171795 Objective Loss 0.171795 LR 0.000125 Time 0.022191 +2023-10-05 22:08:24,826 - Epoch: [183][ 410/ 1236] Overall Loss 0.172164 Objective Loss 0.172164 LR 0.000125 Time 0.022139 +2023-10-05 22:08:25,029 - Epoch: [183][ 420/ 1236] Overall Loss 0.172451 Objective Loss 0.172451 LR 0.000125 Time 0.022094 +2023-10-05 22:08:25,231 - Epoch: [183][ 430/ 1236] Overall Loss 0.172362 Objective Loss 0.172362 LR 0.000125 Time 0.022050 +2023-10-05 22:08:25,434 - Epoch: [183][ 440/ 1236] Overall Loss 0.172880 Objective Loss 0.172880 LR 0.000125 Time 0.022009 +2023-10-05 22:08:25,636 - Epoch: [183][ 450/ 1236] Overall Loss 0.172750 Objective Loss 0.172750 LR 0.000125 Time 0.021968 +2023-10-05 22:08:25,839 - Epoch: [183][ 460/ 1236] Overall Loss 0.172533 Objective Loss 0.172533 LR 0.000125 Time 0.021931 +2023-10-05 22:08:26,041 - Epoch: [183][ 470/ 1236] Overall Loss 0.172419 Objective Loss 0.172419 LR 0.000125 Time 0.021894 +2023-10-05 22:08:26,244 - Epoch: [183][ 480/ 1236] Overall Loss 0.172286 Objective Loss 0.172286 LR 0.000125 Time 0.021859 +2023-10-05 22:08:26,446 - Epoch: [183][ 490/ 1236] Overall Loss 0.171852 Objective Loss 0.171852 LR 0.000125 Time 0.021825 +2023-10-05 22:08:26,649 - Epoch: [183][ 500/ 1236] Overall Loss 0.171824 Objective Loss 0.171824 LR 0.000125 Time 0.021794 +2023-10-05 22:08:26,851 - Epoch: [183][ 510/ 1236] Overall Loss 0.172223 Objective Loss 0.172223 LR 0.000125 Time 0.021762 +2023-10-05 22:08:27,054 - Epoch: [183][ 520/ 1236] Overall Loss 0.172202 Objective Loss 0.172202 LR 0.000125 Time 0.021733 +2023-10-05 22:08:27,256 - Epoch: [183][ 530/ 1236] Overall Loss 0.171837 Objective Loss 0.171837 LR 0.000125 Time 0.021704 +2023-10-05 22:08:27,459 - Epoch: [183][ 540/ 1236] Overall Loss 0.171766 Objective Loss 0.171766 LR 0.000125 Time 0.021677 +2023-10-05 22:08:27,661 - Epoch: [183][ 550/ 1236] Overall Loss 0.171398 Objective Loss 0.171398 LR 0.000125 Time 0.021649 +2023-10-05 22:08:27,864 - Epoch: [183][ 560/ 1236] Overall Loss 0.171418 Objective Loss 0.171418 LR 0.000125 Time 0.021625 +2023-10-05 22:08:28,066 - Epoch: [183][ 570/ 1236] Overall Loss 0.171275 Objective Loss 0.171275 LR 0.000125 Time 0.021599 +2023-10-05 22:08:28,269 - Epoch: [183][ 580/ 1236] Overall Loss 0.171463 Objective Loss 0.171463 LR 0.000125 Time 0.021576 +2023-10-05 22:08:28,471 - Epoch: [183][ 590/ 1236] Overall Loss 0.171542 Objective Loss 0.171542 LR 0.000125 Time 0.021552 +2023-10-05 22:08:28,674 - Epoch: [183][ 600/ 1236] Overall Loss 0.171421 Objective Loss 0.171421 LR 0.000125 Time 0.021531 +2023-10-05 22:08:28,876 - Epoch: [183][ 610/ 1236] Overall Loss 0.171269 Objective Loss 0.171269 LR 0.000125 Time 0.021509 +2023-10-05 22:08:29,079 - Epoch: [183][ 620/ 1236] Overall Loss 0.171305 Objective Loss 0.171305 LR 0.000125 Time 0.021488 +2023-10-05 22:08:29,281 - Epoch: [183][ 630/ 1236] Overall Loss 0.171112 Objective Loss 0.171112 LR 0.000125 Time 0.021468 +2023-10-05 22:08:29,484 - Epoch: [183][ 640/ 1236] Overall Loss 0.171066 Objective Loss 0.171066 LR 0.000125 Time 0.021449 +2023-10-05 22:08:29,686 - Epoch: [183][ 650/ 1236] Overall Loss 0.170883 Objective Loss 0.170883 LR 0.000125 Time 0.021429 +2023-10-05 22:08:29,889 - Epoch: [183][ 660/ 1236] Overall Loss 0.170792 Objective Loss 0.170792 LR 0.000125 Time 0.021411 +2023-10-05 22:08:30,091 - Epoch: [183][ 670/ 1236] Overall Loss 0.170479 Objective Loss 0.170479 LR 0.000125 Time 0.021393 +2023-10-05 22:08:30,294 - Epoch: [183][ 680/ 1236] Overall Loss 0.170672 Objective Loss 0.170672 LR 0.000125 Time 0.021376 +2023-10-05 22:08:30,496 - Epoch: [183][ 690/ 1236] Overall Loss 0.171258 Objective Loss 0.171258 LR 0.000125 Time 0.021359 +2023-10-05 22:08:30,699 - Epoch: [183][ 700/ 1236] Overall Loss 0.171003 Objective Loss 0.171003 LR 0.000125 Time 0.021343 +2023-10-05 22:08:30,901 - Epoch: [183][ 710/ 1236] Overall Loss 0.171078 Objective Loss 0.171078 LR 0.000125 Time 0.021327 +2023-10-05 22:08:31,104 - Epoch: [183][ 720/ 1236] Overall Loss 0.170969 Objective Loss 0.170969 LR 0.000125 Time 0.021312 +2023-10-05 22:08:31,306 - Epoch: [183][ 730/ 1236] Overall Loss 0.170923 Objective Loss 0.170923 LR 0.000125 Time 0.021296 +2023-10-05 22:08:31,509 - Epoch: [183][ 740/ 1236] Overall Loss 0.171114 Objective Loss 0.171114 LR 0.000125 Time 0.021282 +2023-10-05 22:08:31,711 - Epoch: [183][ 750/ 1236] Overall Loss 0.171327 Objective Loss 0.171327 LR 0.000125 Time 0.021268 +2023-10-05 22:08:31,914 - Epoch: [183][ 760/ 1236] Overall Loss 0.171459 Objective Loss 0.171459 LR 0.000125 Time 0.021254 +2023-10-05 22:08:32,116 - Epoch: [183][ 770/ 1236] Overall Loss 0.171368 Objective Loss 0.171368 LR 0.000125 Time 0.021240 +2023-10-05 22:08:32,319 - Epoch: [183][ 780/ 1236] Overall Loss 0.171358 Objective Loss 0.171358 LR 0.000125 Time 0.021227 +2023-10-05 22:08:32,521 - Epoch: [183][ 790/ 1236] Overall Loss 0.171264 Objective Loss 0.171264 LR 0.000125 Time 0.021214 +2023-10-05 22:08:32,724 - Epoch: [183][ 800/ 1236] Overall Loss 0.171546 Objective Loss 0.171546 LR 0.000125 Time 0.021202 +2023-10-05 22:08:32,926 - Epoch: [183][ 810/ 1236] Overall Loss 0.171551 Objective Loss 0.171551 LR 0.000125 Time 0.021189 +2023-10-05 22:08:33,129 - Epoch: [183][ 820/ 1236] Overall Loss 0.171694 Objective Loss 0.171694 LR 0.000125 Time 0.021178 +2023-10-05 22:08:33,331 - Epoch: [183][ 830/ 1236] Overall Loss 0.171709 Objective Loss 0.171709 LR 0.000125 Time 0.021166 +2023-10-05 22:08:33,534 - Epoch: [183][ 840/ 1236] Overall Loss 0.172136 Objective Loss 0.172136 LR 0.000125 Time 0.021155 +2023-10-05 22:08:33,736 - Epoch: [183][ 850/ 1236] Overall Loss 0.171946 Objective Loss 0.171946 LR 0.000125 Time 0.021144 +2023-10-05 22:08:33,939 - Epoch: [183][ 860/ 1236] Overall Loss 0.172302 Objective Loss 0.172302 LR 0.000125 Time 0.021133 +2023-10-05 22:08:34,141 - Epoch: [183][ 870/ 1236] Overall Loss 0.172286 Objective Loss 0.172286 LR 0.000125 Time 0.021122 +2023-10-05 22:08:34,344 - Epoch: [183][ 880/ 1236] Overall Loss 0.172235 Objective Loss 0.172235 LR 0.000125 Time 0.021113 +2023-10-05 22:08:34,546 - Epoch: [183][ 890/ 1236] Overall Loss 0.172403 Objective Loss 0.172403 LR 0.000125 Time 0.021102 +2023-10-05 22:08:34,749 - Epoch: [183][ 900/ 1236] Overall Loss 0.172363 Objective Loss 0.172363 LR 0.000125 Time 0.021092 +2023-10-05 22:08:34,951 - Epoch: [183][ 910/ 1236] Overall Loss 0.172286 Objective Loss 0.172286 LR 0.000125 Time 0.021082 +2023-10-05 22:08:35,154 - Epoch: [183][ 920/ 1236] Overall Loss 0.172074 Objective Loss 0.172074 LR 0.000125 Time 0.021073 +2023-10-05 22:08:35,356 - Epoch: [183][ 930/ 1236] Overall Loss 0.172047 Objective Loss 0.172047 LR 0.000125 Time 0.021064 +2023-10-05 22:08:35,559 - Epoch: [183][ 940/ 1236] Overall Loss 0.172019 Objective Loss 0.172019 LR 0.000125 Time 0.021055 +2023-10-05 22:08:35,761 - Epoch: [183][ 950/ 1236] Overall Loss 0.171944 Objective Loss 0.171944 LR 0.000125 Time 0.021046 +2023-10-05 22:08:35,964 - Epoch: [183][ 960/ 1236] Overall Loss 0.171866 Objective Loss 0.171866 LR 0.000125 Time 0.021037 +2023-10-05 22:08:36,166 - Epoch: [183][ 970/ 1236] Overall Loss 0.171735 Objective Loss 0.171735 LR 0.000125 Time 0.021029 +2023-10-05 22:08:36,369 - Epoch: [183][ 980/ 1236] Overall Loss 0.171768 Objective Loss 0.171768 LR 0.000125 Time 0.021021 +2023-10-05 22:08:36,571 - Epoch: [183][ 990/ 1236] Overall Loss 0.172019 Objective Loss 0.172019 LR 0.000125 Time 0.021012 +2023-10-05 22:08:36,774 - Epoch: [183][ 1000/ 1236] Overall Loss 0.171952 Objective Loss 0.171952 LR 0.000125 Time 0.021005 +2023-10-05 22:08:36,976 - Epoch: [183][ 1010/ 1236] Overall Loss 0.172000 Objective Loss 0.172000 LR 0.000125 Time 0.020997 +2023-10-05 22:08:37,179 - Epoch: [183][ 1020/ 1236] Overall Loss 0.171814 Objective Loss 0.171814 LR 0.000125 Time 0.020989 +2023-10-05 22:08:37,381 - Epoch: [183][ 1030/ 1236] Overall Loss 0.172171 Objective Loss 0.172171 LR 0.000125 Time 0.020981 +2023-10-05 22:08:37,584 - Epoch: [183][ 1040/ 1236] Overall Loss 0.172147 Objective Loss 0.172147 LR 0.000125 Time 0.020974 +2023-10-05 22:08:37,786 - Epoch: [183][ 1050/ 1236] Overall Loss 0.172043 Objective Loss 0.172043 LR 0.000125 Time 0.020967 +2023-10-05 22:08:37,989 - Epoch: [183][ 1060/ 1236] Overall Loss 0.172099 Objective Loss 0.172099 LR 0.000125 Time 0.020960 +2023-10-05 22:08:38,191 - Epoch: [183][ 1070/ 1236] Overall Loss 0.172148 Objective Loss 0.172148 LR 0.000125 Time 0.020953 +2023-10-05 22:08:38,394 - Epoch: [183][ 1080/ 1236] Overall Loss 0.172153 Objective Loss 0.172153 LR 0.000125 Time 0.020946 +2023-10-05 22:08:38,596 - Epoch: [183][ 1090/ 1236] Overall Loss 0.172196 Objective Loss 0.172196 LR 0.000125 Time 0.020939 +2023-10-05 22:08:38,799 - Epoch: [183][ 1100/ 1236] Overall Loss 0.172253 Objective Loss 0.172253 LR 0.000125 Time 0.020933 +2023-10-05 22:08:39,001 - Epoch: [183][ 1110/ 1236] Overall Loss 0.172292 Objective Loss 0.172292 LR 0.000125 Time 0.020926 +2023-10-05 22:08:39,204 - Epoch: [183][ 1120/ 1236] Overall Loss 0.172337 Objective Loss 0.172337 LR 0.000125 Time 0.020920 +2023-10-05 22:08:39,406 - Epoch: [183][ 1130/ 1236] Overall Loss 0.172370 Objective Loss 0.172370 LR 0.000125 Time 0.020914 +2023-10-05 22:08:39,609 - Epoch: [183][ 1140/ 1236] Overall Loss 0.172212 Objective Loss 0.172212 LR 0.000125 Time 0.020908 +2023-10-05 22:08:39,811 - Epoch: [183][ 1150/ 1236] Overall Loss 0.172006 Objective Loss 0.172006 LR 0.000125 Time 0.020901 +2023-10-05 22:08:40,014 - Epoch: [183][ 1160/ 1236] Overall Loss 0.171982 Objective Loss 0.171982 LR 0.000125 Time 0.020896 +2023-10-05 22:08:40,216 - Epoch: [183][ 1170/ 1236] Overall Loss 0.171977 Objective Loss 0.171977 LR 0.000125 Time 0.020890 +2023-10-05 22:08:40,419 - Epoch: [183][ 1180/ 1236] Overall Loss 0.171883 Objective Loss 0.171883 LR 0.000125 Time 0.020884 +2023-10-05 22:08:40,621 - Epoch: [183][ 1190/ 1236] Overall Loss 0.171858 Objective Loss 0.171858 LR 0.000125 Time 0.020879 +2023-10-05 22:08:40,824 - Epoch: [183][ 1200/ 1236] Overall Loss 0.171960 Objective Loss 0.171960 LR 0.000125 Time 0.020873 +2023-10-05 22:08:41,026 - Epoch: [183][ 1210/ 1236] Overall Loss 0.171990 Objective Loss 0.171990 LR 0.000125 Time 0.020868 +2023-10-05 22:08:41,229 - Epoch: [183][ 1220/ 1236] Overall Loss 0.171898 Objective Loss 0.171898 LR 0.000125 Time 0.020862 +2023-10-05 22:08:41,486 - Epoch: [183][ 1230/ 1236] Overall Loss 0.171904 Objective Loss 0.171904 LR 0.000125 Time 0.020901 +2023-10-05 22:08:41,605 - Epoch: [183][ 1236/ 1236] Overall Loss 0.171850 Objective Loss 0.171850 Top1 92.057026 Top5 99.389002 LR 0.000125 Time 0.020896 +2023-10-05 22:08:41,716 - --- validate (epoch=183)----------- +2023-10-05 22:08:41,717 - 29943 samples (256 per mini-batch) +2023-10-05 22:08:42,180 - Epoch: [183][ 10/ 117] Loss 0.286952 Top1 85.625000 Top5 98.242188 +2023-10-05 22:08:42,328 - Epoch: [183][ 20/ 117] Loss 0.295409 Top1 86.171875 Top5 98.203125 +2023-10-05 22:08:42,476 - Epoch: [183][ 30/ 117] Loss 0.298216 Top1 85.768229 Top5 98.307292 +2023-10-05 22:08:42,625 - Epoch: [183][ 40/ 117] Loss 0.302472 Top1 85.634766 Top5 98.203125 +2023-10-05 22:08:42,775 - Epoch: [183][ 50/ 117] Loss 0.304790 Top1 85.609375 Top5 98.210938 +2023-10-05 22:08:42,925 - Epoch: [183][ 60/ 117] Loss 0.298728 Top1 85.618490 Top5 98.229167 +2023-10-05 22:08:43,071 - Epoch: [183][ 70/ 117] Loss 0.300940 Top1 85.786830 Top5 98.264509 +2023-10-05 22:08:43,220 - Epoch: [183][ 80/ 117] Loss 0.305201 Top1 85.703125 Top5 98.256836 +2023-10-05 22:08:43,367 - Epoch: [183][ 90/ 117] Loss 0.306356 Top1 85.664062 Top5 98.246528 +2023-10-05 22:08:43,515 - Epoch: [183][ 100/ 117] Loss 0.303731 Top1 85.738281 Top5 98.238281 +2023-10-05 22:08:43,667 - Epoch: [183][ 110/ 117] Loss 0.302322 Top1 85.841619 Top5 98.231534 +2023-10-05 22:08:43,753 - Epoch: [183][ 117/ 117] Loss 0.300462 Top1 85.829743 Top5 98.233310 +2023-10-05 22:08:43,897 - ==> Top1: 85.830 Top5: 98.233 Loss: 0.300 + +2023-10-05 22:08:43,898 - ==> Confusion: +[[ 935 1 6 3 6 1 0 0 5 66 1 0 1 2 5 3 1 1 0 0 13] + [ 0 1055 1 0 12 20 1 19 2 0 0 2 0 0 1 3 2 1 7 1 4] + [ 3 2 979 11 2 0 17 6 0 1 4 2 8 2 2 3 0 1 5 3 5] + [ 4 1 11 971 0 1 0 3 1 2 4 0 7 2 31 2 0 8 20 3 18] + [ 25 7 1 0 976 0 1 2 0 11 0 1 1 2 8 1 7 2 0 2 3] + [ 5 38 3 1 5 982 0 21 1 1 5 8 0 13 7 1 1 0 2 5 17] + [ 0 7 24 1 0 0 1123 10 0 0 1 5 1 0 1 6 0 0 2 6 4] + [ 3 22 12 0 3 22 1 1081 1 3 3 10 1 2 1 2 0 1 36 7 7] + [ 18 3 0 0 0 0 0 1 983 39 7 2 3 9 15 2 1 1 2 2 1] + [ 87 0 1 2 6 2 0 0 20 957 0 0 0 23 4 1 0 1 0 5 10] + [ 2 4 10 3 0 2 3 4 13 2 969 2 0 12 5 2 4 0 6 2 8] + [ 1 0 1 1 0 10 0 3 0 1 0 972 14 3 0 4 1 13 0 6 5] + [ 1 1 3 4 0 1 0 2 1 0 0 34 990 3 1 2 2 12 1 2 8] + [ 2 0 1 0 0 4 0 0 8 15 4 2 3 1068 3 1 0 0 0 1 7] + [ 13 1 3 3 4 1 0 0 23 4 2 1 2 3 1017 0 0 2 10 0 12] + [ 0 2 2 0 3 0 0 1 0 1 0 5 7 3 1 1076 12 11 0 8 2] + [ 1 12 1 0 7 2 0 1 1 0 0 6 0 2 3 9 1095 0 0 5 16] + [ 0 0 1 0 1 0 3 0 1 0 0 3 16 1 0 5 0 1001 2 0 4] + [ 2 8 3 16 1 0 0 23 1 0 1 0 1 0 11 1 1 1 991 1 6] + [ 0 3 3 3 2 3 8 9 1 0 2 14 3 2 0 5 6 2 5 1069 12] + [ 132 154 143 51 101 107 36 84 109 76 164 104 294 286 156 55 105 59 133 146 5410]] + +2023-10-05 22:08:43,899 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:08:43,899 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:08:43,905 - + +2023-10-05 22:08:43,905 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:08:45,027 - Epoch: [184][ 10/ 1236] Overall Loss 0.165526 Objective Loss 0.165526 LR 0.000125 Time 0.112076 +2023-10-05 22:08:45,229 - Epoch: [184][ 20/ 1236] Overall Loss 0.166122 Objective Loss 0.166122 LR 0.000125 Time 0.066157 +2023-10-05 22:08:45,430 - Epoch: [184][ 30/ 1236] Overall Loss 0.164491 Objective Loss 0.164491 LR 0.000125 Time 0.050776 +2023-10-05 22:08:45,632 - Epoch: [184][ 40/ 1236] Overall Loss 0.167842 Objective Loss 0.167842 LR 0.000125 Time 0.043133 +2023-10-05 22:08:45,833 - Epoch: [184][ 50/ 1236] Overall Loss 0.170267 Objective Loss 0.170267 LR 0.000125 Time 0.038515 +2023-10-05 22:08:46,036 - Epoch: [184][ 60/ 1236] Overall Loss 0.171433 Objective Loss 0.171433 LR 0.000125 Time 0.035470 +2023-10-05 22:08:46,237 - Epoch: [184][ 70/ 1236] Overall Loss 0.173388 Objective Loss 0.173388 LR 0.000125 Time 0.033270 +2023-10-05 22:08:46,439 - Epoch: [184][ 80/ 1236] Overall Loss 0.171600 Objective Loss 0.171600 LR 0.000125 Time 0.031639 +2023-10-05 22:08:46,641 - Epoch: [184][ 90/ 1236] Overall Loss 0.172528 Objective Loss 0.172528 LR 0.000125 Time 0.030355 +2023-10-05 22:08:46,843 - Epoch: [184][ 100/ 1236] Overall Loss 0.169625 Objective Loss 0.169625 LR 0.000125 Time 0.029337 +2023-10-05 22:08:47,044 - Epoch: [184][ 110/ 1236] Overall Loss 0.168669 Objective Loss 0.168669 LR 0.000125 Time 0.028496 +2023-10-05 22:08:47,246 - Epoch: [184][ 120/ 1236] Overall Loss 0.168907 Objective Loss 0.168907 LR 0.000125 Time 0.027803 +2023-10-05 22:08:47,447 - Epoch: [184][ 130/ 1236] Overall Loss 0.168266 Objective Loss 0.168266 LR 0.000125 Time 0.027209 +2023-10-05 22:08:47,649 - Epoch: [184][ 140/ 1236] Overall Loss 0.167135 Objective Loss 0.167135 LR 0.000125 Time 0.026707 +2023-10-05 22:08:47,850 - Epoch: [184][ 150/ 1236] Overall Loss 0.166658 Objective Loss 0.166658 LR 0.000125 Time 0.026264 +2023-10-05 22:08:48,053 - Epoch: [184][ 160/ 1236] Overall Loss 0.166365 Objective Loss 0.166365 LR 0.000125 Time 0.025885 +2023-10-05 22:08:48,254 - Epoch: [184][ 170/ 1236] Overall Loss 0.167155 Objective Loss 0.167155 LR 0.000125 Time 0.025544 +2023-10-05 22:08:48,456 - Epoch: [184][ 180/ 1236] Overall Loss 0.167802 Objective Loss 0.167802 LR 0.000125 Time 0.025247 +2023-10-05 22:08:48,657 - Epoch: [184][ 190/ 1236] Overall Loss 0.167642 Objective Loss 0.167642 LR 0.000125 Time 0.024972 +2023-10-05 22:08:48,860 - Epoch: [184][ 200/ 1236] Overall Loss 0.167498 Objective Loss 0.167498 LR 0.000125 Time 0.024736 +2023-10-05 22:08:49,061 - Epoch: [184][ 210/ 1236] Overall Loss 0.167939 Objective Loss 0.167939 LR 0.000125 Time 0.024515 +2023-10-05 22:08:49,263 - Epoch: [184][ 220/ 1236] Overall Loss 0.168838 Objective Loss 0.168838 LR 0.000125 Time 0.024318 +2023-10-05 22:08:49,467 - Epoch: [184][ 230/ 1236] Overall Loss 0.169167 Objective Loss 0.169167 LR 0.000125 Time 0.024147 +2023-10-05 22:08:49,679 - Epoch: [184][ 240/ 1236] Overall Loss 0.169036 Objective Loss 0.169036 LR 0.000125 Time 0.024023 +2023-10-05 22:08:49,887 - Epoch: [184][ 250/ 1236] Overall Loss 0.169717 Objective Loss 0.169717 LR 0.000125 Time 0.023891 +2023-10-05 22:08:50,099 - Epoch: [184][ 260/ 1236] Overall Loss 0.169454 Objective Loss 0.169454 LR 0.000125 Time 0.023786 +2023-10-05 22:08:50,306 - Epoch: [184][ 270/ 1236] Overall Loss 0.169709 Objective Loss 0.169709 LR 0.000125 Time 0.023671 +2023-10-05 22:08:50,518 - Epoch: [184][ 280/ 1236] Overall Loss 0.169592 Objective Loss 0.169592 LR 0.000125 Time 0.023582 +2023-10-05 22:08:50,726 - Epoch: [184][ 290/ 1236] Overall Loss 0.169589 Objective Loss 0.169589 LR 0.000125 Time 0.023483 +2023-10-05 22:08:50,940 - Epoch: [184][ 300/ 1236] Overall Loss 0.169847 Objective Loss 0.169847 LR 0.000125 Time 0.023414 +2023-10-05 22:08:51,151 - Epoch: [184][ 310/ 1236] Overall Loss 0.170123 Objective Loss 0.170123 LR 0.000125 Time 0.023339 +2023-10-05 22:08:51,362 - Epoch: [184][ 320/ 1236] Overall Loss 0.170171 Objective Loss 0.170171 LR 0.000125 Time 0.023266 +2023-10-05 22:08:51,563 - Epoch: [184][ 330/ 1236] Overall Loss 0.170515 Objective Loss 0.170515 LR 0.000125 Time 0.023169 +2023-10-05 22:08:51,765 - Epoch: [184][ 340/ 1236] Overall Loss 0.170857 Objective Loss 0.170857 LR 0.000125 Time 0.023081 +2023-10-05 22:08:51,966 - Epoch: [184][ 350/ 1236] Overall Loss 0.171572 Objective Loss 0.171572 LR 0.000125 Time 0.022996 +2023-10-05 22:08:52,168 - Epoch: [184][ 360/ 1236] Overall Loss 0.171979 Objective Loss 0.171979 LR 0.000125 Time 0.022918 +2023-10-05 22:08:52,371 - Epoch: [184][ 370/ 1236] Overall Loss 0.171750 Objective Loss 0.171750 LR 0.000125 Time 0.022844 +2023-10-05 22:08:52,573 - Epoch: [184][ 380/ 1236] Overall Loss 0.171554 Objective Loss 0.171554 LR 0.000125 Time 0.022774 +2023-10-05 22:08:52,774 - Epoch: [184][ 390/ 1236] Overall Loss 0.172139 Objective Loss 0.172139 LR 0.000125 Time 0.022706 +2023-10-05 22:08:52,976 - Epoch: [184][ 400/ 1236] Overall Loss 0.171951 Objective Loss 0.171951 LR 0.000125 Time 0.022642 +2023-10-05 22:08:53,178 - Epoch: [184][ 410/ 1236] Overall Loss 0.171683 Objective Loss 0.171683 LR 0.000125 Time 0.022581 +2023-10-05 22:08:53,380 - Epoch: [184][ 420/ 1236] Overall Loss 0.171814 Objective Loss 0.171814 LR 0.000125 Time 0.022523 +2023-10-05 22:08:53,582 - Epoch: [184][ 430/ 1236] Overall Loss 0.171444 Objective Loss 0.171444 LR 0.000125 Time 0.022468 +2023-10-05 22:08:53,784 - Epoch: [184][ 440/ 1236] Overall Loss 0.171346 Objective Loss 0.171346 LR 0.000125 Time 0.022417 +2023-10-05 22:08:53,986 - Epoch: [184][ 450/ 1236] Overall Loss 0.171170 Objective Loss 0.171170 LR 0.000125 Time 0.022366 +2023-10-05 22:08:54,188 - Epoch: [184][ 460/ 1236] Overall Loss 0.171004 Objective Loss 0.171004 LR 0.000125 Time 0.022318 +2023-10-05 22:08:54,389 - Epoch: [184][ 470/ 1236] Overall Loss 0.171061 Objective Loss 0.171061 LR 0.000125 Time 0.022271 +2023-10-05 22:08:54,591 - Epoch: [184][ 480/ 1236] Overall Loss 0.170637 Objective Loss 0.170637 LR 0.000125 Time 0.022227 +2023-10-05 22:08:54,793 - Epoch: [184][ 490/ 1236] Overall Loss 0.170254 Objective Loss 0.170254 LR 0.000125 Time 0.022185 +2023-10-05 22:08:54,995 - Epoch: [184][ 500/ 1236] Overall Loss 0.169825 Objective Loss 0.169825 LR 0.000125 Time 0.022144 +2023-10-05 22:08:55,197 - Epoch: [184][ 510/ 1236] Overall Loss 0.169793 Objective Loss 0.169793 LR 0.000125 Time 0.022105 +2023-10-05 22:08:55,399 - Epoch: [184][ 520/ 1236] Overall Loss 0.169725 Objective Loss 0.169725 LR 0.000125 Time 0.022068 +2023-10-05 22:08:55,601 - Epoch: [184][ 530/ 1236] Overall Loss 0.169477 Objective Loss 0.169477 LR 0.000125 Time 0.022032 +2023-10-05 22:08:55,802 - Epoch: [184][ 540/ 1236] Overall Loss 0.169558 Objective Loss 0.169558 LR 0.000125 Time 0.021997 +2023-10-05 22:08:56,004 - Epoch: [184][ 550/ 1236] Overall Loss 0.169372 Objective Loss 0.169372 LR 0.000125 Time 0.021963 +2023-10-05 22:08:56,206 - Epoch: [184][ 560/ 1236] Overall Loss 0.169467 Objective Loss 0.169467 LR 0.000125 Time 0.021931 +2023-10-05 22:08:56,408 - Epoch: [184][ 570/ 1236] Overall Loss 0.169364 Objective Loss 0.169364 LR 0.000125 Time 0.021900 +2023-10-05 22:08:56,610 - Epoch: [184][ 580/ 1236] Overall Loss 0.169251 Objective Loss 0.169251 LR 0.000125 Time 0.021869 +2023-10-05 22:08:56,811 - Epoch: [184][ 590/ 1236] Overall Loss 0.169467 Objective Loss 0.169467 LR 0.000125 Time 0.021840 +2023-10-05 22:08:57,013 - Epoch: [184][ 600/ 1236] Overall Loss 0.169806 Objective Loss 0.169806 LR 0.000125 Time 0.021812 +2023-10-05 22:08:57,215 - Epoch: [184][ 610/ 1236] Overall Loss 0.169935 Objective Loss 0.169935 LR 0.000125 Time 0.021785 +2023-10-05 22:08:57,417 - Epoch: [184][ 620/ 1236] Overall Loss 0.170240 Objective Loss 0.170240 LR 0.000125 Time 0.021758 +2023-10-05 22:08:57,619 - Epoch: [184][ 630/ 1236] Overall Loss 0.170264 Objective Loss 0.170264 LR 0.000125 Time 0.021733 +2023-10-05 22:08:57,820 - Epoch: [184][ 640/ 1236] Overall Loss 0.170183 Objective Loss 0.170183 LR 0.000125 Time 0.021708 +2023-10-05 22:08:58,022 - Epoch: [184][ 650/ 1236] Overall Loss 0.170226 Objective Loss 0.170226 LR 0.000125 Time 0.021684 +2023-10-05 22:08:58,223 - Epoch: [184][ 660/ 1236] Overall Loss 0.170192 Objective Loss 0.170192 LR 0.000125 Time 0.021659 +2023-10-05 22:08:58,425 - Epoch: [184][ 670/ 1236] Overall Loss 0.170087 Objective Loss 0.170087 LR 0.000125 Time 0.021637 +2023-10-05 22:08:58,627 - Epoch: [184][ 680/ 1236] Overall Loss 0.170032 Objective Loss 0.170032 LR 0.000125 Time 0.021615 +2023-10-05 22:08:58,829 - Epoch: [184][ 690/ 1236] Overall Loss 0.170217 Objective Loss 0.170217 LR 0.000125 Time 0.021594 +2023-10-05 22:08:59,030 - Epoch: [184][ 700/ 1236] Overall Loss 0.170235 Objective Loss 0.170235 LR 0.000125 Time 0.021573 +2023-10-05 22:08:59,232 - Epoch: [184][ 710/ 1236] Overall Loss 0.170179 Objective Loss 0.170179 LR 0.000125 Time 0.021553 +2023-10-05 22:08:59,433 - Epoch: [184][ 720/ 1236] Overall Loss 0.170376 Objective Loss 0.170376 LR 0.000125 Time 0.021532 +2023-10-05 22:08:59,635 - Epoch: [184][ 730/ 1236] Overall Loss 0.170244 Objective Loss 0.170244 LR 0.000125 Time 0.021514 +2023-10-05 22:08:59,837 - Epoch: [184][ 740/ 1236] Overall Loss 0.170186 Objective Loss 0.170186 LR 0.000125 Time 0.021495 +2023-10-05 22:09:00,039 - Epoch: [184][ 750/ 1236] Overall Loss 0.170066 Objective Loss 0.170066 LR 0.000125 Time 0.021477 +2023-10-05 22:09:00,240 - Epoch: [184][ 760/ 1236] Overall Loss 0.170308 Objective Loss 0.170308 LR 0.000125 Time 0.021459 +2023-10-05 22:09:00,442 - Epoch: [184][ 770/ 1236] Overall Loss 0.170255 Objective Loss 0.170255 LR 0.000125 Time 0.021442 +2023-10-05 22:09:00,643 - Epoch: [184][ 780/ 1236] Overall Loss 0.170259 Objective Loss 0.170259 LR 0.000125 Time 0.021424 +2023-10-05 22:09:00,845 - Epoch: [184][ 790/ 1236] Overall Loss 0.170303 Objective Loss 0.170303 LR 0.000125 Time 0.021408 +2023-10-05 22:09:01,047 - Epoch: [184][ 800/ 1236] Overall Loss 0.170204 Objective Loss 0.170204 LR 0.000125 Time 0.021393 +2023-10-05 22:09:01,249 - Epoch: [184][ 810/ 1236] Overall Loss 0.170288 Objective Loss 0.170288 LR 0.000125 Time 0.021378 +2023-10-05 22:09:01,451 - Epoch: [184][ 820/ 1236] Overall Loss 0.170362 Objective Loss 0.170362 LR 0.000125 Time 0.021363 +2023-10-05 22:09:01,653 - Epoch: [184][ 830/ 1236] Overall Loss 0.170401 Objective Loss 0.170401 LR 0.000125 Time 0.021348 +2023-10-05 22:09:01,854 - Epoch: [184][ 840/ 1236] Overall Loss 0.170146 Objective Loss 0.170146 LR 0.000125 Time 0.021334 +2023-10-05 22:09:02,056 - Epoch: [184][ 850/ 1236] Overall Loss 0.170326 Objective Loss 0.170326 LR 0.000125 Time 0.021320 +2023-10-05 22:09:02,258 - Epoch: [184][ 860/ 1236] Overall Loss 0.170354 Objective Loss 0.170354 LR 0.000125 Time 0.021306 +2023-10-05 22:09:02,460 - Epoch: [184][ 870/ 1236] Overall Loss 0.170478 Objective Loss 0.170478 LR 0.000125 Time 0.021293 +2023-10-05 22:09:02,662 - Epoch: [184][ 880/ 1236] Overall Loss 0.170451 Objective Loss 0.170451 LR 0.000125 Time 0.021281 +2023-10-05 22:09:02,864 - Epoch: [184][ 890/ 1236] Overall Loss 0.170652 Objective Loss 0.170652 LR 0.000125 Time 0.021268 +2023-10-05 22:09:03,066 - Epoch: [184][ 900/ 1236] Overall Loss 0.170753 Objective Loss 0.170753 LR 0.000125 Time 0.021256 +2023-10-05 22:09:03,268 - Epoch: [184][ 910/ 1236] Overall Loss 0.170850 Objective Loss 0.170850 LR 0.000125 Time 0.021243 +2023-10-05 22:09:03,469 - Epoch: [184][ 920/ 1236] Overall Loss 0.170553 Objective Loss 0.170553 LR 0.000125 Time 0.021232 +2023-10-05 22:09:03,670 - Epoch: [184][ 930/ 1236] Overall Loss 0.170384 Objective Loss 0.170384 LR 0.000125 Time 0.021219 +2023-10-05 22:09:03,872 - Epoch: [184][ 940/ 1236] Overall Loss 0.170639 Objective Loss 0.170639 LR 0.000125 Time 0.021207 +2023-10-05 22:09:04,074 - Epoch: [184][ 950/ 1236] Overall Loss 0.170726 Objective Loss 0.170726 LR 0.000125 Time 0.021197 +2023-10-05 22:09:04,276 - Epoch: [184][ 960/ 1236] Overall Loss 0.170814 Objective Loss 0.170814 LR 0.000125 Time 0.021186 +2023-10-05 22:09:04,478 - Epoch: [184][ 970/ 1236] Overall Loss 0.170665 Objective Loss 0.170665 LR 0.000125 Time 0.021175 +2023-10-05 22:09:04,680 - Epoch: [184][ 980/ 1236] Overall Loss 0.170791 Objective Loss 0.170791 LR 0.000125 Time 0.021165 +2023-10-05 22:09:04,882 - Epoch: [184][ 990/ 1236] Overall Loss 0.170666 Objective Loss 0.170666 LR 0.000125 Time 0.021155 +2023-10-05 22:09:05,083 - Epoch: [184][ 1000/ 1236] Overall Loss 0.170674 Objective Loss 0.170674 LR 0.000125 Time 0.021144 +2023-10-05 22:09:05,285 - Epoch: [184][ 1010/ 1236] Overall Loss 0.170823 Objective Loss 0.170823 LR 0.000125 Time 0.021134 +2023-10-05 22:09:05,487 - Epoch: [184][ 1020/ 1236] Overall Loss 0.170884 Objective Loss 0.170884 LR 0.000125 Time 0.021124 +2023-10-05 22:09:05,688 - Epoch: [184][ 1030/ 1236] Overall Loss 0.170664 Objective Loss 0.170664 LR 0.000125 Time 0.021114 +2023-10-05 22:09:05,890 - Epoch: [184][ 1040/ 1236] Overall Loss 0.170875 Objective Loss 0.170875 LR 0.000125 Time 0.021105 +2023-10-05 22:09:06,090 - Epoch: [184][ 1050/ 1236] Overall Loss 0.170954 Objective Loss 0.170954 LR 0.000125 Time 0.021095 +2023-10-05 22:09:06,292 - Epoch: [184][ 1060/ 1236] Overall Loss 0.170910 Objective Loss 0.170910 LR 0.000125 Time 0.021086 +2023-10-05 22:09:06,495 - Epoch: [184][ 1070/ 1236] Overall Loss 0.170952 Objective Loss 0.170952 LR 0.000125 Time 0.021078 +2023-10-05 22:09:06,696 - Epoch: [184][ 1080/ 1236] Overall Loss 0.170936 Objective Loss 0.170936 LR 0.000125 Time 0.021069 +2023-10-05 22:09:06,898 - Epoch: [184][ 1090/ 1236] Overall Loss 0.170904 Objective Loss 0.170904 LR 0.000125 Time 0.021061 +2023-10-05 22:09:07,100 - Epoch: [184][ 1100/ 1236] Overall Loss 0.170875 Objective Loss 0.170875 LR 0.000125 Time 0.021053 +2023-10-05 22:09:07,302 - Epoch: [184][ 1110/ 1236] Overall Loss 0.170749 Objective Loss 0.170749 LR 0.000125 Time 0.021045 +2023-10-05 22:09:07,504 - Epoch: [184][ 1120/ 1236] Overall Loss 0.170959 Objective Loss 0.170959 LR 0.000125 Time 0.021037 +2023-10-05 22:09:07,705 - Epoch: [184][ 1130/ 1236] Overall Loss 0.170885 Objective Loss 0.170885 LR 0.000125 Time 0.021028 +2023-10-05 22:09:07,906 - Epoch: [184][ 1140/ 1236] Overall Loss 0.171177 Objective Loss 0.171177 LR 0.000125 Time 0.021020 +2023-10-05 22:09:08,108 - Epoch: [184][ 1150/ 1236] Overall Loss 0.171213 Objective Loss 0.171213 LR 0.000125 Time 0.021012 +2023-10-05 22:09:08,310 - Epoch: [184][ 1160/ 1236] Overall Loss 0.171296 Objective Loss 0.171296 LR 0.000125 Time 0.021005 +2023-10-05 22:09:08,512 - Epoch: [184][ 1170/ 1236] Overall Loss 0.171138 Objective Loss 0.171138 LR 0.000125 Time 0.020998 +2023-10-05 22:09:08,714 - Epoch: [184][ 1180/ 1236] Overall Loss 0.171077 Objective Loss 0.171077 LR 0.000125 Time 0.020991 +2023-10-05 22:09:08,916 - Epoch: [184][ 1190/ 1236] Overall Loss 0.171035 Objective Loss 0.171035 LR 0.000125 Time 0.020984 +2023-10-05 22:09:09,118 - Epoch: [184][ 1200/ 1236] Overall Loss 0.171098 Objective Loss 0.171098 LR 0.000125 Time 0.020977 +2023-10-05 22:09:09,320 - Epoch: [184][ 1210/ 1236] Overall Loss 0.171240 Objective Loss 0.171240 LR 0.000125 Time 0.020971 +2023-10-05 22:09:09,522 - Epoch: [184][ 1220/ 1236] Overall Loss 0.171243 Objective Loss 0.171243 LR 0.000125 Time 0.020964 +2023-10-05 22:09:09,779 - Epoch: [184][ 1230/ 1236] Overall Loss 0.171383 Objective Loss 0.171383 LR 0.000125 Time 0.021002 +2023-10-05 22:09:09,898 - Epoch: [184][ 1236/ 1236] Overall Loss 0.171294 Objective Loss 0.171294 Top1 89.409369 Top5 98.574338 LR 0.000125 Time 0.020996 +2023-10-05 22:09:10,016 - --- validate (epoch=184)----------- +2023-10-05 22:09:10,017 - 29943 samples (256 per mini-batch) +2023-10-05 22:09:10,466 - Epoch: [184][ 10/ 117] Loss 0.294198 Top1 85.976562 Top5 98.320312 +2023-10-05 22:09:10,611 - Epoch: [184][ 20/ 117] Loss 0.294249 Top1 85.820312 Top5 98.183594 +2023-10-05 22:09:10,756 - Epoch: [184][ 30/ 117] Loss 0.286417 Top1 86.289062 Top5 98.229167 +2023-10-05 22:09:10,903 - Epoch: [184][ 40/ 117] Loss 0.287023 Top1 86.240234 Top5 98.203125 +2023-10-05 22:09:11,051 - Epoch: [184][ 50/ 117] Loss 0.291613 Top1 86.140625 Top5 98.273438 +2023-10-05 22:09:11,197 - Epoch: [184][ 60/ 117] Loss 0.293573 Top1 86.087240 Top5 98.274740 +2023-10-05 22:09:11,344 - Epoch: [184][ 70/ 117] Loss 0.297504 Top1 86.077009 Top5 98.264509 +2023-10-05 22:09:11,490 - Epoch: [184][ 80/ 117] Loss 0.293645 Top1 86.152344 Top5 98.247070 +2023-10-05 22:09:11,638 - Epoch: [184][ 90/ 117] Loss 0.299315 Top1 85.946181 Top5 98.250868 +2023-10-05 22:09:11,785 - Epoch: [184][ 100/ 117] Loss 0.301741 Top1 85.738281 Top5 98.234375 +2023-10-05 22:09:11,939 - Epoch: [184][ 110/ 117] Loss 0.302953 Top1 85.614347 Top5 98.213778 +2023-10-05 22:09:12,024 - Epoch: [184][ 117/ 117] Loss 0.305937 Top1 85.545870 Top5 98.183215 +2023-10-05 22:09:12,153 - ==> Top1: 85.546 Top5: 98.183 Loss: 0.306 + +2023-10-05 22:09:12,153 - ==> Confusion: +[[ 950 2 6 3 7 1 0 0 2 52 1 0 1 2 6 2 3 2 0 0 10] + [ 0 1062 2 0 11 15 1 17 0 0 0 3 0 0 0 3 3 0 8 0 6] + [ 5 2 980 9 1 0 18 6 0 0 3 3 9 1 1 3 0 2 5 3 5] + [ 2 1 13 967 0 1 3 1 1 1 3 0 7 3 25 4 0 4 28 2 23] + [ 25 5 1 0 978 2 0 1 0 7 0 2 0 3 6 2 10 1 2 0 5] + [ 3 33 1 1 3 990 2 23 0 0 3 10 1 10 5 2 3 0 2 4 20] + [ 0 5 33 0 0 0 1122 6 0 0 1 2 2 0 1 6 0 0 2 6 5] + [ 5 17 13 0 3 26 3 1074 1 5 2 10 1 1 0 1 0 0 40 7 9] + [ 19 4 1 1 0 2 1 0 964 47 9 4 2 8 17 2 2 0 4 1 1] + [ 116 0 4 1 4 3 0 1 14 940 1 1 1 15 3 4 0 1 0 1 9] + [ 4 5 10 3 0 1 7 4 8 2 965 2 2 13 5 2 3 0 7 1 9] + [ 1 0 1 0 0 10 0 3 0 1 0 975 13 2 0 4 2 14 0 5 4] + [ 1 1 2 7 0 0 0 2 0 0 1 40 986 1 0 2 2 15 1 2 5] + [ 3 0 1 2 0 9 0 1 9 15 8 5 3 1046 4 3 1 0 0 0 9] + [ 15 1 4 5 4 0 0 0 21 4 2 1 2 2 1014 0 1 2 10 0 13] + [ 1 3 1 0 3 0 1 0 0 0 0 9 6 1 1 1074 14 11 0 6 3] + [ 1 13 1 0 4 3 0 0 1 0 0 4 0 0 2 9 1106 0 0 2 15] + [ 1 0 0 0 0 0 2 0 0 0 0 2 19 2 1 5 0 1001 1 1 3] + [ 1 4 6 22 1 0 0 20 1 0 2 0 3 0 7 0 1 0 990 2 8] + [ 0 3 4 3 2 7 8 9 0 0 1 18 6 0 0 6 10 1 4 1059 11] + [ 139 154 164 47 88 123 29 93 83 72 141 124 314 261 145 53 128 60 142 173 5372]] + +2023-10-05 22:09:12,155 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:09:12,155 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:09:12,161 - + +2023-10-05 22:09:12,161 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:09:13,166 - Epoch: [185][ 10/ 1236] Overall Loss 0.184009 Objective Loss 0.184009 LR 0.000125 Time 0.100458 +2023-10-05 22:09:13,368 - Epoch: [185][ 20/ 1236] Overall Loss 0.176866 Objective Loss 0.176866 LR 0.000125 Time 0.060331 +2023-10-05 22:09:13,571 - Epoch: [185][ 30/ 1236] Overall Loss 0.172094 Objective Loss 0.172094 LR 0.000125 Time 0.046968 +2023-10-05 22:09:13,773 - Epoch: [185][ 40/ 1236] Overall Loss 0.173941 Objective Loss 0.173941 LR 0.000125 Time 0.040267 +2023-10-05 22:09:13,976 - Epoch: [185][ 50/ 1236] Overall Loss 0.176741 Objective Loss 0.176741 LR 0.000125 Time 0.036259 +2023-10-05 22:09:14,177 - Epoch: [185][ 60/ 1236] Overall Loss 0.176864 Objective Loss 0.176864 LR 0.000125 Time 0.033574 +2023-10-05 22:09:14,380 - Epoch: [185][ 70/ 1236] Overall Loss 0.177481 Objective Loss 0.177481 LR 0.000125 Time 0.031665 +2023-10-05 22:09:14,582 - Epoch: [185][ 80/ 1236] Overall Loss 0.174874 Objective Loss 0.174874 LR 0.000125 Time 0.030231 +2023-10-05 22:09:14,785 - Epoch: [185][ 90/ 1236] Overall Loss 0.174755 Objective Loss 0.174755 LR 0.000125 Time 0.029119 +2023-10-05 22:09:14,988 - Epoch: [185][ 100/ 1236] Overall Loss 0.173964 Objective Loss 0.173964 LR 0.000125 Time 0.028233 +2023-10-05 22:09:15,191 - Epoch: [185][ 110/ 1236] Overall Loss 0.174208 Objective Loss 0.174208 LR 0.000125 Time 0.027514 +2023-10-05 22:09:15,395 - Epoch: [185][ 120/ 1236] Overall Loss 0.174564 Objective Loss 0.174564 LR 0.000125 Time 0.026916 +2023-10-05 22:09:15,599 - Epoch: [185][ 130/ 1236] Overall Loss 0.174351 Objective Loss 0.174351 LR 0.000125 Time 0.026414 +2023-10-05 22:09:15,804 - Epoch: [185][ 140/ 1236] Overall Loss 0.173010 Objective Loss 0.173010 LR 0.000125 Time 0.025986 +2023-10-05 22:09:16,007 - Epoch: [185][ 150/ 1236] Overall Loss 0.174403 Objective Loss 0.174403 LR 0.000125 Time 0.025608 +2023-10-05 22:09:16,212 - Epoch: [185][ 160/ 1236] Overall Loss 0.172349 Objective Loss 0.172349 LR 0.000125 Time 0.025283 +2023-10-05 22:09:16,416 - Epoch: [185][ 170/ 1236] Overall Loss 0.172174 Objective Loss 0.172174 LR 0.000125 Time 0.024993 +2023-10-05 22:09:16,620 - Epoch: [185][ 180/ 1236] Overall Loss 0.171928 Objective Loss 0.171928 LR 0.000125 Time 0.024739 +2023-10-05 22:09:16,824 - Epoch: [185][ 190/ 1236] Overall Loss 0.171383 Objective Loss 0.171383 LR 0.000125 Time 0.024508 +2023-10-05 22:09:17,028 - Epoch: [185][ 200/ 1236] Overall Loss 0.171907 Objective Loss 0.171907 LR 0.000125 Time 0.024303 +2023-10-05 22:09:17,232 - Epoch: [185][ 210/ 1236] Overall Loss 0.171907 Objective Loss 0.171907 LR 0.000125 Time 0.024115 +2023-10-05 22:09:17,437 - Epoch: [185][ 220/ 1236] Overall Loss 0.172324 Objective Loss 0.172324 LR 0.000125 Time 0.023947 +2023-10-05 22:09:17,641 - Epoch: [185][ 230/ 1236] Overall Loss 0.171697 Objective Loss 0.171697 LR 0.000125 Time 0.023790 +2023-10-05 22:09:17,845 - Epoch: [185][ 240/ 1236] Overall Loss 0.172184 Objective Loss 0.172184 LR 0.000125 Time 0.023651 +2023-10-05 22:09:18,049 - Epoch: [185][ 250/ 1236] Overall Loss 0.172434 Objective Loss 0.172434 LR 0.000125 Time 0.023519 +2023-10-05 22:09:18,254 - Epoch: [185][ 260/ 1236] Overall Loss 0.171933 Objective Loss 0.171933 LR 0.000125 Time 0.023400 +2023-10-05 22:09:18,458 - Epoch: [185][ 270/ 1236] Overall Loss 0.172765 Objective Loss 0.172765 LR 0.000125 Time 0.023287 +2023-10-05 22:09:18,662 - Epoch: [185][ 280/ 1236] Overall Loss 0.172644 Objective Loss 0.172644 LR 0.000125 Time 0.023185 +2023-10-05 22:09:18,866 - Epoch: [185][ 290/ 1236] Overall Loss 0.172033 Objective Loss 0.172033 LR 0.000125 Time 0.023088 +2023-10-05 22:09:19,071 - Epoch: [185][ 300/ 1236] Overall Loss 0.171807 Objective Loss 0.171807 LR 0.000125 Time 0.022999 +2023-10-05 22:09:19,275 - Epoch: [185][ 310/ 1236] Overall Loss 0.172266 Objective Loss 0.172266 LR 0.000125 Time 0.022913 +2023-10-05 22:09:19,480 - Epoch: [185][ 320/ 1236] Overall Loss 0.172490 Objective Loss 0.172490 LR 0.000125 Time 0.022837 +2023-10-05 22:09:19,684 - Epoch: [185][ 330/ 1236] Overall Loss 0.172086 Objective Loss 0.172086 LR 0.000125 Time 0.022762 +2023-10-05 22:09:19,888 - Epoch: [185][ 340/ 1236] Overall Loss 0.172041 Objective Loss 0.172041 LR 0.000125 Time 0.022693 +2023-10-05 22:09:20,092 - Epoch: [185][ 350/ 1236] Overall Loss 0.171984 Objective Loss 0.171984 LR 0.000125 Time 0.022627 +2023-10-05 22:09:20,297 - Epoch: [185][ 360/ 1236] Overall Loss 0.171638 Objective Loss 0.171638 LR 0.000125 Time 0.022566 +2023-10-05 22:09:20,501 - Epoch: [185][ 370/ 1236] Overall Loss 0.171864 Objective Loss 0.171864 LR 0.000125 Time 0.022507 +2023-10-05 22:09:20,705 - Epoch: [185][ 380/ 1236] Overall Loss 0.172250 Objective Loss 0.172250 LR 0.000125 Time 0.022451 +2023-10-05 22:09:20,908 - Epoch: [185][ 390/ 1236] Overall Loss 0.172270 Objective Loss 0.172270 LR 0.000125 Time 0.022394 +2023-10-05 22:09:21,110 - Epoch: [185][ 400/ 1236] Overall Loss 0.172214 Objective Loss 0.172214 LR 0.000125 Time 0.022339 +2023-10-05 22:09:21,313 - Epoch: [185][ 410/ 1236] Overall Loss 0.172254 Objective Loss 0.172254 LR 0.000125 Time 0.022288 +2023-10-05 22:09:21,515 - Epoch: [185][ 420/ 1236] Overall Loss 0.172280 Objective Loss 0.172280 LR 0.000125 Time 0.022238 +2023-10-05 22:09:21,718 - Epoch: [185][ 430/ 1236] Overall Loss 0.172137 Objective Loss 0.172137 LR 0.000125 Time 0.022192 +2023-10-05 22:09:21,921 - Epoch: [185][ 440/ 1236] Overall Loss 0.172088 Objective Loss 0.172088 LR 0.000125 Time 0.022147 +2023-10-05 22:09:22,123 - Epoch: [185][ 450/ 1236] Overall Loss 0.172162 Objective Loss 0.172162 LR 0.000125 Time 0.022104 +2023-10-05 22:09:22,325 - Epoch: [185][ 460/ 1236] Overall Loss 0.172370 Objective Loss 0.172370 LR 0.000125 Time 0.022062 +2023-10-05 22:09:22,528 - Epoch: [185][ 470/ 1236] Overall Loss 0.172224 Objective Loss 0.172224 LR 0.000125 Time 0.022023 +2023-10-05 22:09:22,730 - Epoch: [185][ 480/ 1236] Overall Loss 0.172328 Objective Loss 0.172328 LR 0.000125 Time 0.021986 +2023-10-05 22:09:22,933 - Epoch: [185][ 490/ 1236] Overall Loss 0.172481 Objective Loss 0.172481 LR 0.000125 Time 0.021950 +2023-10-05 22:09:23,136 - Epoch: [185][ 500/ 1236] Overall Loss 0.172279 Objective Loss 0.172279 LR 0.000125 Time 0.021915 +2023-10-05 22:09:23,338 - Epoch: [185][ 510/ 1236] Overall Loss 0.171970 Objective Loss 0.171970 LR 0.000125 Time 0.021882 +2023-10-05 22:09:23,540 - Epoch: [185][ 520/ 1236] Overall Loss 0.171894 Objective Loss 0.171894 LR 0.000125 Time 0.021849 +2023-10-05 22:09:23,743 - Epoch: [185][ 530/ 1236] Overall Loss 0.172020 Objective Loss 0.172020 LR 0.000125 Time 0.021819 +2023-10-05 22:09:23,945 - Epoch: [185][ 540/ 1236] Overall Loss 0.172002 Objective Loss 0.172002 LR 0.000125 Time 0.021789 +2023-10-05 22:09:24,148 - Epoch: [185][ 550/ 1236] Overall Loss 0.172179 Objective Loss 0.172179 LR 0.000125 Time 0.021760 +2023-10-05 22:09:24,350 - Epoch: [185][ 560/ 1236] Overall Loss 0.171658 Objective Loss 0.171658 LR 0.000125 Time 0.021733 +2023-10-05 22:09:24,553 - Epoch: [185][ 570/ 1236] Overall Loss 0.171664 Objective Loss 0.171664 LR 0.000125 Time 0.021706 +2023-10-05 22:09:24,756 - Epoch: [185][ 580/ 1236] Overall Loss 0.171618 Objective Loss 0.171618 LR 0.000125 Time 0.021680 +2023-10-05 22:09:24,958 - Epoch: [185][ 590/ 1236] Overall Loss 0.171938 Objective Loss 0.171938 LR 0.000125 Time 0.021656 +2023-10-05 22:09:25,161 - Epoch: [185][ 600/ 1236] Overall Loss 0.172289 Objective Loss 0.172289 LR 0.000125 Time 0.021632 +2023-10-05 22:09:25,363 - Epoch: [185][ 610/ 1236] Overall Loss 0.172375 Objective Loss 0.172375 LR 0.000125 Time 0.021609 +2023-10-05 22:09:25,565 - Epoch: [185][ 620/ 1236] Overall Loss 0.172283 Objective Loss 0.172283 LR 0.000125 Time 0.021586 +2023-10-05 22:09:25,768 - Epoch: [185][ 630/ 1236] Overall Loss 0.172459 Objective Loss 0.172459 LR 0.000125 Time 0.021564 +2023-10-05 22:09:25,970 - Epoch: [185][ 640/ 1236] Overall Loss 0.172472 Objective Loss 0.172472 LR 0.000125 Time 0.021543 +2023-10-05 22:09:26,173 - Epoch: [185][ 650/ 1236] Overall Loss 0.172115 Objective Loss 0.172115 LR 0.000125 Time 0.021523 +2023-10-05 22:09:26,375 - Epoch: [185][ 660/ 1236] Overall Loss 0.172317 Objective Loss 0.172317 LR 0.000125 Time 0.021503 +2023-10-05 22:09:26,578 - Epoch: [185][ 670/ 1236] Overall Loss 0.171988 Objective Loss 0.171988 LR 0.000125 Time 0.021484 +2023-10-05 22:09:26,780 - Epoch: [185][ 680/ 1236] Overall Loss 0.171983 Objective Loss 0.171983 LR 0.000125 Time 0.021465 +2023-10-05 22:09:26,983 - Epoch: [185][ 690/ 1236] Overall Loss 0.172021 Objective Loss 0.172021 LR 0.000125 Time 0.021447 +2023-10-05 22:09:27,185 - Epoch: [185][ 700/ 1236] Overall Loss 0.171993 Objective Loss 0.171993 LR 0.000125 Time 0.021429 +2023-10-05 22:09:27,388 - Epoch: [185][ 710/ 1236] Overall Loss 0.172075 Objective Loss 0.172075 LR 0.000125 Time 0.021412 +2023-10-05 22:09:27,590 - Epoch: [185][ 720/ 1236] Overall Loss 0.171996 Objective Loss 0.171996 LR 0.000125 Time 0.021395 +2023-10-05 22:09:27,793 - Epoch: [185][ 730/ 1236] Overall Loss 0.172027 Objective Loss 0.172027 LR 0.000125 Time 0.021380 +2023-10-05 22:09:27,995 - Epoch: [185][ 740/ 1236] Overall Loss 0.172326 Objective Loss 0.172326 LR 0.000125 Time 0.021364 +2023-10-05 22:09:28,198 - Epoch: [185][ 750/ 1236] Overall Loss 0.172362 Objective Loss 0.172362 LR 0.000125 Time 0.021349 +2023-10-05 22:09:28,400 - Epoch: [185][ 760/ 1236] Overall Loss 0.172254 Objective Loss 0.172254 LR 0.000125 Time 0.021333 +2023-10-05 22:09:28,603 - Epoch: [185][ 770/ 1236] Overall Loss 0.172172 Objective Loss 0.172172 LR 0.000125 Time 0.021319 +2023-10-05 22:09:28,806 - Epoch: [185][ 780/ 1236] Overall Loss 0.172111 Objective Loss 0.172111 LR 0.000125 Time 0.021305 +2023-10-05 22:09:29,008 - Epoch: [185][ 790/ 1236] Overall Loss 0.172377 Objective Loss 0.172377 LR 0.000125 Time 0.021291 +2023-10-05 22:09:29,210 - Epoch: [185][ 800/ 1236] Overall Loss 0.172431 Objective Loss 0.172431 LR 0.000125 Time 0.021277 +2023-10-05 22:09:29,413 - Epoch: [185][ 810/ 1236] Overall Loss 0.172079 Objective Loss 0.172079 LR 0.000125 Time 0.021265 +2023-10-05 22:09:29,616 - Epoch: [185][ 820/ 1236] Overall Loss 0.172031 Objective Loss 0.172031 LR 0.000125 Time 0.021252 +2023-10-05 22:09:29,818 - Epoch: [185][ 830/ 1236] Overall Loss 0.172077 Objective Loss 0.172077 LR 0.000125 Time 0.021240 +2023-10-05 22:09:30,020 - Epoch: [185][ 840/ 1236] Overall Loss 0.172095 Objective Loss 0.172095 LR 0.000125 Time 0.021227 +2023-10-05 22:09:30,223 - Epoch: [185][ 850/ 1236] Overall Loss 0.172131 Objective Loss 0.172131 LR 0.000125 Time 0.021215 +2023-10-05 22:09:30,425 - Epoch: [185][ 860/ 1236] Overall Loss 0.172006 Objective Loss 0.172006 LR 0.000125 Time 0.021203 +2023-10-05 22:09:30,628 - Epoch: [185][ 870/ 1236] Overall Loss 0.171821 Objective Loss 0.171821 LR 0.000125 Time 0.021193 +2023-10-05 22:09:30,831 - Epoch: [185][ 880/ 1236] Overall Loss 0.171712 Objective Loss 0.171712 LR 0.000125 Time 0.021181 +2023-10-05 22:09:31,039 - Epoch: [185][ 890/ 1236] Overall Loss 0.171520 Objective Loss 0.171520 LR 0.000125 Time 0.021178 +2023-10-05 22:09:31,252 - Epoch: [185][ 900/ 1236] Overall Loss 0.171511 Objective Loss 0.171511 LR 0.000125 Time 0.021178 +2023-10-05 22:09:31,453 - Epoch: [185][ 910/ 1236] Overall Loss 0.171428 Objective Loss 0.171428 LR 0.000125 Time 0.021166 +2023-10-05 22:09:31,658 - Epoch: [185][ 920/ 1236] Overall Loss 0.171414 Objective Loss 0.171414 LR 0.000125 Time 0.021158 +2023-10-05 22:09:31,858 - Epoch: [185][ 930/ 1236] Overall Loss 0.171499 Objective Loss 0.171499 LR 0.000125 Time 0.021146 +2023-10-05 22:09:32,063 - Epoch: [185][ 940/ 1236] Overall Loss 0.171494 Objective Loss 0.171494 LR 0.000125 Time 0.021138 +2023-10-05 22:09:32,264 - Epoch: [185][ 950/ 1236] Overall Loss 0.171324 Objective Loss 0.171324 LR 0.000125 Time 0.021127 +2023-10-05 22:09:32,468 - Epoch: [185][ 960/ 1236] Overall Loss 0.171389 Objective Loss 0.171389 LR 0.000125 Time 0.021119 +2023-10-05 22:09:32,670 - Epoch: [185][ 970/ 1236] Overall Loss 0.171362 Objective Loss 0.171362 LR 0.000125 Time 0.021108 +2023-10-05 22:09:32,874 - Epoch: [185][ 980/ 1236] Overall Loss 0.171245 Objective Loss 0.171245 LR 0.000125 Time 0.021101 +2023-10-05 22:09:33,075 - Epoch: [185][ 990/ 1236] Overall Loss 0.171141 Objective Loss 0.171141 LR 0.000125 Time 0.021090 +2023-10-05 22:09:33,279 - Epoch: [185][ 1000/ 1236] Overall Loss 0.171012 Objective Loss 0.171012 LR 0.000125 Time 0.021083 +2023-10-05 22:09:33,479 - Epoch: [185][ 1010/ 1236] Overall Loss 0.170822 Objective Loss 0.170822 LR 0.000125 Time 0.021073 +2023-10-05 22:09:33,684 - Epoch: [185][ 1020/ 1236] Overall Loss 0.170914 Objective Loss 0.170914 LR 0.000125 Time 0.021066 +2023-10-05 22:09:33,885 - Epoch: [185][ 1030/ 1236] Overall Loss 0.171089 Objective Loss 0.171089 LR 0.000125 Time 0.021056 +2023-10-05 22:09:34,089 - Epoch: [185][ 1040/ 1236] Overall Loss 0.171040 Objective Loss 0.171040 LR 0.000125 Time 0.021049 +2023-10-05 22:09:34,290 - Epoch: [185][ 1050/ 1236] Overall Loss 0.171223 Objective Loss 0.171223 LR 0.000125 Time 0.021040 +2023-10-05 22:09:34,494 - Epoch: [185][ 1060/ 1236] Overall Loss 0.171401 Objective Loss 0.171401 LR 0.000125 Time 0.021034 +2023-10-05 22:09:34,695 - Epoch: [185][ 1070/ 1236] Overall Loss 0.171222 Objective Loss 0.171222 LR 0.000125 Time 0.021025 +2023-10-05 22:09:34,899 - Epoch: [185][ 1080/ 1236] Overall Loss 0.171255 Objective Loss 0.171255 LR 0.000125 Time 0.021019 +2023-10-05 22:09:35,100 - Epoch: [185][ 1090/ 1236] Overall Loss 0.171203 Objective Loss 0.171203 LR 0.000125 Time 0.021010 +2023-10-05 22:09:35,304 - Epoch: [185][ 1100/ 1236] Overall Loss 0.171450 Objective Loss 0.171450 LR 0.000125 Time 0.021004 +2023-10-05 22:09:35,505 - Epoch: [185][ 1110/ 1236] Overall Loss 0.171449 Objective Loss 0.171449 LR 0.000125 Time 0.020995 +2023-10-05 22:09:35,709 - Epoch: [185][ 1120/ 1236] Overall Loss 0.171454 Objective Loss 0.171454 LR 0.000125 Time 0.020990 +2023-10-05 22:09:35,910 - Epoch: [185][ 1130/ 1236] Overall Loss 0.171423 Objective Loss 0.171423 LR 0.000125 Time 0.020981 +2023-10-05 22:09:36,111 - Epoch: [185][ 1140/ 1236] Overall Loss 0.171329 Objective Loss 0.171329 LR 0.000125 Time 0.020974 +2023-10-05 22:09:36,312 - Epoch: [185][ 1150/ 1236] Overall Loss 0.171276 Objective Loss 0.171276 LR 0.000125 Time 0.020966 +2023-10-05 22:09:36,515 - Epoch: [185][ 1160/ 1236] Overall Loss 0.171326 Objective Loss 0.171326 LR 0.000125 Time 0.020960 +2023-10-05 22:09:36,716 - Epoch: [185][ 1170/ 1236] Overall Loss 0.171340 Objective Loss 0.171340 LR 0.000125 Time 0.020952 +2023-10-05 22:09:36,920 - Epoch: [185][ 1180/ 1236] Overall Loss 0.171328 Objective Loss 0.171328 LR 0.000125 Time 0.020947 +2023-10-05 22:09:37,121 - Epoch: [185][ 1190/ 1236] Overall Loss 0.171388 Objective Loss 0.171388 LR 0.000125 Time 0.020940 +2023-10-05 22:09:37,325 - Epoch: [185][ 1200/ 1236] Overall Loss 0.171351 Objective Loss 0.171351 LR 0.000125 Time 0.020935 +2023-10-05 22:09:37,525 - Epoch: [185][ 1210/ 1236] Overall Loss 0.171448 Objective Loss 0.171448 LR 0.000125 Time 0.020927 +2023-10-05 22:09:37,730 - Epoch: [185][ 1220/ 1236] Overall Loss 0.171294 Objective Loss 0.171294 LR 0.000125 Time 0.020923 +2023-10-05 22:09:37,986 - Epoch: [185][ 1230/ 1236] Overall Loss 0.171414 Objective Loss 0.171414 LR 0.000125 Time 0.020961 +2023-10-05 22:09:38,104 - Epoch: [185][ 1236/ 1236] Overall Loss 0.171396 Objective Loss 0.171396 Top1 89.205703 Top5 97.556008 LR 0.000125 Time 0.020955 +2023-10-05 22:09:38,224 - --- validate (epoch=185)----------- +2023-10-05 22:09:38,224 - 29943 samples (256 per mini-batch) +2023-10-05 22:09:38,674 - Epoch: [185][ 10/ 117] Loss 0.303560 Top1 86.289062 Top5 98.242188 +2023-10-05 22:09:38,822 - Epoch: [185][ 20/ 117] Loss 0.305320 Top1 86.230469 Top5 98.105469 +2023-10-05 22:09:38,969 - Epoch: [185][ 30/ 117] Loss 0.317765 Top1 85.703125 Top5 98.007812 +2023-10-05 22:09:39,117 - Epoch: [185][ 40/ 117] Loss 0.313570 Top1 85.585938 Top5 98.056641 +2023-10-05 22:09:39,264 - Epoch: [185][ 50/ 117] Loss 0.311639 Top1 85.414062 Top5 98.117188 +2023-10-05 22:09:39,414 - Epoch: [185][ 60/ 117] Loss 0.308266 Top1 85.566406 Top5 98.125000 +2023-10-05 22:09:39,560 - Epoch: [185][ 70/ 117] Loss 0.305591 Top1 85.747768 Top5 98.152902 +2023-10-05 22:09:39,708 - Epoch: [185][ 80/ 117] Loss 0.306810 Top1 85.693359 Top5 98.090820 +2023-10-05 22:09:39,855 - Epoch: [185][ 90/ 117] Loss 0.304417 Top1 85.664062 Top5 98.098958 +2023-10-05 22:09:40,004 - Epoch: [185][ 100/ 117] Loss 0.304019 Top1 85.742188 Top5 98.121094 +2023-10-05 22:09:40,159 - Epoch: [185][ 110/ 117] Loss 0.303740 Top1 85.749290 Top5 98.096591 +2023-10-05 22:09:40,245 - Epoch: [185][ 117/ 117] Loss 0.303000 Top1 85.799686 Top5 98.136459 +2023-10-05 22:09:40,380 - ==> Top1: 85.800 Top5: 98.136 Loss: 0.303 + +2023-10-05 22:09:40,381 - ==> Confusion: +[[ 932 4 4 2 6 1 0 0 3 70 1 0 1 2 3 3 3 1 0 0 14] + [ 0 1066 1 0 10 15 1 15 0 0 0 1 0 0 2 4 3 0 7 1 5] + [ 5 3 967 10 2 0 20 9 0 0 5 2 7 2 1 4 1 2 5 4 7] + [ 2 2 10 962 1 5 1 2 3 1 5 1 6 2 32 3 0 5 26 2 18] + [ 22 6 0 0 979 3 1 1 0 9 0 1 0 2 5 2 11 3 0 2 3] + [ 5 41 0 1 4 989 0 22 1 2 6 8 0 9 3 2 2 0 1 4 16] + [ 0 4 16 0 0 0 1134 9 0 0 2 4 0 0 1 5 0 0 2 8 6] + [ 4 24 9 0 3 26 4 1066 1 3 5 9 2 1 1 2 0 0 45 5 8] + [ 20 2 0 0 0 3 1 0 976 47 8 0 1 8 12 2 1 0 4 2 2] + [ 88 0 4 2 7 3 0 0 20 965 1 2 1 14 2 2 0 1 0 1 6] + [ 4 7 7 3 0 2 2 6 11 1 971 0 1 11 6 2 1 1 6 1 10] + [ 0 1 2 0 0 9 0 2 0 1 0 968 16 4 0 5 2 13 0 7 5] + [ 2 1 1 4 1 1 0 2 0 0 2 33 990 0 0 2 3 12 4 4 6] + [ 2 0 1 0 3 4 0 0 10 14 9 2 3 1057 4 1 1 0 0 1 7] + [ 14 3 3 6 3 0 0 0 23 4 1 1 3 1 1010 0 1 2 8 0 18] + [ 1 2 1 0 2 0 1 0 0 1 0 8 6 2 1 1076 12 10 0 8 3] + [ 1 15 1 0 7 1 0 0 1 0 0 2 0 0 3 10 1107 0 0 3 10] + [ 0 0 1 1 0 0 3 0 0 0 0 2 15 1 0 5 0 1003 2 0 5] + [ 2 13 4 18 1 0 0 20 2 0 1 0 0 0 8 0 0 0 989 1 9] + [ 0 3 3 3 2 8 8 11 0 0 1 18 2 0 0 6 11 2 3 1063 8] + [ 122 180 126 55 102 119 36 87 90 81 166 98 287 265 139 57 147 55 131 141 5421]] + +2023-10-05 22:09:40,382 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:09:40,382 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:09:40,388 - + +2023-10-05 22:09:40,388 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:09:41,391 - Epoch: [186][ 10/ 1236] Overall Loss 0.176189 Objective Loss 0.176189 LR 0.000125 Time 0.100153 +2023-10-05 22:09:41,591 - Epoch: [186][ 20/ 1236] Overall Loss 0.175110 Objective Loss 0.175110 LR 0.000125 Time 0.060075 +2023-10-05 22:09:41,790 - Epoch: [186][ 30/ 1236] Overall Loss 0.172398 Objective Loss 0.172398 LR 0.000125 Time 0.046676 +2023-10-05 22:09:41,991 - Epoch: [186][ 40/ 1236] Overall Loss 0.170177 Objective Loss 0.170177 LR 0.000125 Time 0.040022 +2023-10-05 22:09:42,190 - Epoch: [186][ 50/ 1236] Overall Loss 0.164542 Objective Loss 0.164542 LR 0.000125 Time 0.035993 +2023-10-05 22:09:42,391 - Epoch: [186][ 60/ 1236] Overall Loss 0.169669 Objective Loss 0.169669 LR 0.000125 Time 0.033334 +2023-10-05 22:09:42,591 - Epoch: [186][ 70/ 1236] Overall Loss 0.170211 Objective Loss 0.170211 LR 0.000125 Time 0.031413 +2023-10-05 22:09:42,791 - Epoch: [186][ 80/ 1236] Overall Loss 0.172263 Objective Loss 0.172263 LR 0.000125 Time 0.029991 +2023-10-05 22:09:42,991 - Epoch: [186][ 90/ 1236] Overall Loss 0.170856 Objective Loss 0.170856 LR 0.000125 Time 0.028868 +2023-10-05 22:09:43,191 - Epoch: [186][ 100/ 1236] Overall Loss 0.170180 Objective Loss 0.170180 LR 0.000125 Time 0.027981 +2023-10-05 22:09:43,392 - Epoch: [186][ 110/ 1236] Overall Loss 0.168832 Objective Loss 0.168832 LR 0.000125 Time 0.027258 +2023-10-05 22:09:43,594 - Epoch: [186][ 120/ 1236] Overall Loss 0.169006 Objective Loss 0.169006 LR 0.000125 Time 0.026667 +2023-10-05 22:09:43,794 - Epoch: [186][ 130/ 1236] Overall Loss 0.168734 Objective Loss 0.168734 LR 0.000125 Time 0.026157 +2023-10-05 22:09:43,995 - Epoch: [186][ 140/ 1236] Overall Loss 0.168320 Objective Loss 0.168320 LR 0.000125 Time 0.025721 +2023-10-05 22:09:44,196 - Epoch: [186][ 150/ 1236] Overall Loss 0.168951 Objective Loss 0.168951 LR 0.000125 Time 0.025341 +2023-10-05 22:09:44,395 - Epoch: [186][ 160/ 1236] Overall Loss 0.168829 Objective Loss 0.168829 LR 0.000125 Time 0.025003 +2023-10-05 22:09:44,596 - Epoch: [186][ 170/ 1236] Overall Loss 0.168684 Objective Loss 0.168684 LR 0.000125 Time 0.024711 +2023-10-05 22:09:44,796 - Epoch: [186][ 180/ 1236] Overall Loss 0.168741 Objective Loss 0.168741 LR 0.000125 Time 0.024445 +2023-10-05 22:09:44,997 - Epoch: [186][ 190/ 1236] Overall Loss 0.168537 Objective Loss 0.168537 LR 0.000125 Time 0.024216 +2023-10-05 22:09:45,197 - Epoch: [186][ 200/ 1236] Overall Loss 0.168917 Objective Loss 0.168917 LR 0.000125 Time 0.024002 +2023-10-05 22:09:45,397 - Epoch: [186][ 210/ 1236] Overall Loss 0.169121 Objective Loss 0.169121 LR 0.000125 Time 0.023812 +2023-10-05 22:09:45,597 - Epoch: [186][ 220/ 1236] Overall Loss 0.169209 Objective Loss 0.169209 LR 0.000125 Time 0.023636 +2023-10-05 22:09:45,798 - Epoch: [186][ 230/ 1236] Overall Loss 0.170592 Objective Loss 0.170592 LR 0.000125 Time 0.023480 +2023-10-05 22:09:45,998 - Epoch: [186][ 240/ 1236] Overall Loss 0.169860 Objective Loss 0.169860 LR 0.000125 Time 0.023333 +2023-10-05 22:09:46,199 - Epoch: [186][ 250/ 1236] Overall Loss 0.169425 Objective Loss 0.169425 LR 0.000125 Time 0.023202 +2023-10-05 22:09:46,398 - Epoch: [186][ 260/ 1236] Overall Loss 0.169672 Objective Loss 0.169672 LR 0.000125 Time 0.023076 +2023-10-05 22:09:46,599 - Epoch: [186][ 270/ 1236] Overall Loss 0.170072 Objective Loss 0.170072 LR 0.000125 Time 0.022965 +2023-10-05 22:09:46,799 - Epoch: [186][ 280/ 1236] Overall Loss 0.169632 Objective Loss 0.169632 LR 0.000125 Time 0.022858 +2023-10-05 22:09:47,001 - Epoch: [186][ 290/ 1236] Overall Loss 0.169469 Objective Loss 0.169469 LR 0.000125 Time 0.022763 +2023-10-05 22:09:47,201 - Epoch: [186][ 300/ 1236] Overall Loss 0.169700 Objective Loss 0.169700 LR 0.000125 Time 0.022670 +2023-10-05 22:09:47,402 - Epoch: [186][ 310/ 1236] Overall Loss 0.169981 Objective Loss 0.169981 LR 0.000125 Time 0.022586 +2023-10-05 22:09:47,602 - Epoch: [186][ 320/ 1236] Overall Loss 0.169536 Objective Loss 0.169536 LR 0.000125 Time 0.022504 +2023-10-05 22:09:47,803 - Epoch: [186][ 330/ 1236] Overall Loss 0.169499 Objective Loss 0.169499 LR 0.000125 Time 0.022431 +2023-10-05 22:09:48,003 - Epoch: [186][ 340/ 1236] Overall Loss 0.169553 Objective Loss 0.169553 LR 0.000125 Time 0.022360 +2023-10-05 22:09:48,211 - Epoch: [186][ 350/ 1236] Overall Loss 0.169522 Objective Loss 0.169522 LR 0.000125 Time 0.022312 +2023-10-05 22:09:48,419 - Epoch: [186][ 360/ 1236] Overall Loss 0.169481 Objective Loss 0.169481 LR 0.000125 Time 0.022268 +2023-10-05 22:09:48,623 - Epoch: [186][ 370/ 1236] Overall Loss 0.169336 Objective Loss 0.169336 LR 0.000125 Time 0.022216 +2023-10-05 22:09:48,825 - Epoch: [186][ 380/ 1236] Overall Loss 0.169180 Objective Loss 0.169180 LR 0.000125 Time 0.022163 +2023-10-05 22:09:49,029 - Epoch: [186][ 390/ 1236] Overall Loss 0.169234 Objective Loss 0.169234 LR 0.000125 Time 0.022115 +2023-10-05 22:09:49,231 - Epoch: [186][ 400/ 1236] Overall Loss 0.169323 Objective Loss 0.169323 LR 0.000125 Time 0.022068 +2023-10-05 22:09:49,445 - Epoch: [186][ 410/ 1236] Overall Loss 0.169993 Objective Loss 0.169993 LR 0.000125 Time 0.022052 +2023-10-05 22:09:49,654 - Epoch: [186][ 420/ 1236] Overall Loss 0.169906 Objective Loss 0.169906 LR 0.000125 Time 0.022023 +2023-10-05 22:09:49,869 - Epoch: [186][ 430/ 1236] Overall Loss 0.169500 Objective Loss 0.169500 LR 0.000125 Time 0.022009 +2023-10-05 22:09:50,078 - Epoch: [186][ 440/ 1236] Overall Loss 0.169590 Objective Loss 0.169590 LR 0.000125 Time 0.021984 +2023-10-05 22:09:50,293 - Epoch: [186][ 450/ 1236] Overall Loss 0.169532 Objective Loss 0.169532 LR 0.000125 Time 0.021971 +2023-10-05 22:09:50,502 - Epoch: [186][ 460/ 1236] Overall Loss 0.169654 Objective Loss 0.169654 LR 0.000125 Time 0.021948 +2023-10-05 22:09:50,716 - Epoch: [186][ 470/ 1236] Overall Loss 0.169985 Objective Loss 0.169985 LR 0.000125 Time 0.021936 +2023-10-05 22:09:50,922 - Epoch: [186][ 480/ 1236] Overall Loss 0.169736 Objective Loss 0.169736 LR 0.000125 Time 0.021907 +2023-10-05 22:09:51,137 - Epoch: [186][ 490/ 1236] Overall Loss 0.169400 Objective Loss 0.169400 LR 0.000125 Time 0.021898 +2023-10-05 22:09:51,346 - Epoch: [186][ 500/ 1236] Overall Loss 0.169609 Objective Loss 0.169609 LR 0.000125 Time 0.021878 +2023-10-05 22:09:51,561 - Epoch: [186][ 510/ 1236] Overall Loss 0.169452 Objective Loss 0.169452 LR 0.000125 Time 0.021868 +2023-10-05 22:09:51,770 - Epoch: [186][ 520/ 1236] Overall Loss 0.169342 Objective Loss 0.169342 LR 0.000125 Time 0.021850 +2023-10-05 22:09:51,985 - Epoch: [186][ 530/ 1236] Overall Loss 0.169100 Objective Loss 0.169100 LR 0.000125 Time 0.021842 +2023-10-05 22:09:52,194 - Epoch: [186][ 540/ 1236] Overall Loss 0.168955 Objective Loss 0.168955 LR 0.000125 Time 0.021825 +2023-10-05 22:09:52,409 - Epoch: [186][ 550/ 1236] Overall Loss 0.169241 Objective Loss 0.169241 LR 0.000125 Time 0.021817 +2023-10-05 22:09:52,619 - Epoch: [186][ 560/ 1236] Overall Loss 0.169039 Objective Loss 0.169039 LR 0.000125 Time 0.021802 +2023-10-05 22:09:52,829 - Epoch: [186][ 570/ 1236] Overall Loss 0.169485 Objective Loss 0.169485 LR 0.000125 Time 0.021788 +2023-10-05 22:09:53,036 - Epoch: [186][ 580/ 1236] Overall Loss 0.169342 Objective Loss 0.169342 LR 0.000125 Time 0.021768 +2023-10-05 22:09:53,250 - Epoch: [186][ 590/ 1236] Overall Loss 0.169356 Objective Loss 0.169356 LR 0.000125 Time 0.021762 +2023-10-05 22:09:53,460 - Epoch: [186][ 600/ 1236] Overall Loss 0.169339 Objective Loss 0.169339 LR 0.000125 Time 0.021748 +2023-10-05 22:09:53,674 - Epoch: [186][ 610/ 1236] Overall Loss 0.169268 Objective Loss 0.169268 LR 0.000125 Time 0.021743 +2023-10-05 22:09:53,887 - Epoch: [186][ 620/ 1236] Overall Loss 0.169262 Objective Loss 0.169262 LR 0.000125 Time 0.021734 +2023-10-05 22:09:54,102 - Epoch: [186][ 630/ 1236] Overall Loss 0.169499 Objective Loss 0.169499 LR 0.000125 Time 0.021731 +2023-10-05 22:09:54,315 - Epoch: [186][ 640/ 1236] Overall Loss 0.169298 Objective Loss 0.169298 LR 0.000125 Time 0.021723 +2023-10-05 22:09:54,531 - Epoch: [186][ 650/ 1236] Overall Loss 0.169134 Objective Loss 0.169134 LR 0.000125 Time 0.021720 +2023-10-05 22:09:54,743 - Epoch: [186][ 660/ 1236] Overall Loss 0.169303 Objective Loss 0.169303 LR 0.000125 Time 0.021712 +2023-10-05 22:09:54,958 - Epoch: [186][ 670/ 1236] Overall Loss 0.169219 Objective Loss 0.169219 LR 0.000125 Time 0.021709 +2023-10-05 22:09:55,171 - Epoch: [186][ 680/ 1236] Overall Loss 0.169222 Objective Loss 0.169222 LR 0.000125 Time 0.021702 +2023-10-05 22:09:55,386 - Epoch: [186][ 690/ 1236] Overall Loss 0.169241 Objective Loss 0.169241 LR 0.000125 Time 0.021699 +2023-10-05 22:09:55,599 - Epoch: [186][ 700/ 1236] Overall Loss 0.169315 Objective Loss 0.169315 LR 0.000125 Time 0.021692 +2023-10-05 22:09:55,814 - Epoch: [186][ 710/ 1236] Overall Loss 0.169347 Objective Loss 0.169347 LR 0.000125 Time 0.021690 +2023-10-05 22:09:56,027 - Epoch: [186][ 720/ 1236] Overall Loss 0.169386 Objective Loss 0.169386 LR 0.000125 Time 0.021683 +2023-10-05 22:09:56,243 - Epoch: [186][ 730/ 1236] Overall Loss 0.169370 Objective Loss 0.169370 LR 0.000125 Time 0.021681 +2023-10-05 22:09:56,455 - Epoch: [186][ 740/ 1236] Overall Loss 0.169256 Objective Loss 0.169256 LR 0.000125 Time 0.021675 +2023-10-05 22:09:56,670 - Epoch: [186][ 750/ 1236] Overall Loss 0.169274 Objective Loss 0.169274 LR 0.000125 Time 0.021673 +2023-10-05 22:09:56,883 - Epoch: [186][ 760/ 1236] Overall Loss 0.169219 Objective Loss 0.169219 LR 0.000125 Time 0.021667 +2023-10-05 22:09:57,099 - Epoch: [186][ 770/ 1236] Overall Loss 0.169514 Objective Loss 0.169514 LR 0.000125 Time 0.021665 +2023-10-05 22:09:57,311 - Epoch: [186][ 780/ 1236] Overall Loss 0.169622 Objective Loss 0.169622 LR 0.000125 Time 0.021659 +2023-10-05 22:09:57,527 - Epoch: [186][ 790/ 1236] Overall Loss 0.169766 Objective Loss 0.169766 LR 0.000125 Time 0.021658 +2023-10-05 22:09:57,739 - Epoch: [186][ 800/ 1236] Overall Loss 0.169782 Objective Loss 0.169782 LR 0.000125 Time 0.021652 +2023-10-05 22:09:57,955 - Epoch: [186][ 810/ 1236] Overall Loss 0.170052 Objective Loss 0.170052 LR 0.000125 Time 0.021651 +2023-10-05 22:09:58,168 - Epoch: [186][ 820/ 1236] Overall Loss 0.170102 Objective Loss 0.170102 LR 0.000125 Time 0.021646 +2023-10-05 22:09:58,384 - Epoch: [186][ 830/ 1236] Overall Loss 0.170200 Objective Loss 0.170200 LR 0.000125 Time 0.021645 +2023-10-05 22:09:58,596 - Epoch: [186][ 840/ 1236] Overall Loss 0.170186 Objective Loss 0.170186 LR 0.000125 Time 0.021640 +2023-10-05 22:09:58,812 - Epoch: [186][ 850/ 1236] Overall Loss 0.169996 Objective Loss 0.169996 LR 0.000125 Time 0.021639 +2023-10-05 22:09:59,024 - Epoch: [186][ 860/ 1236] Overall Loss 0.169747 Objective Loss 0.169747 LR 0.000125 Time 0.021634 +2023-10-05 22:09:59,240 - Epoch: [186][ 870/ 1236] Overall Loss 0.169616 Objective Loss 0.169616 LR 0.000125 Time 0.021633 +2023-10-05 22:09:59,453 - Epoch: [186][ 880/ 1236] Overall Loss 0.169610 Objective Loss 0.169610 LR 0.000125 Time 0.021628 +2023-10-05 22:09:59,669 - Epoch: [186][ 890/ 1236] Overall Loss 0.169302 Objective Loss 0.169302 LR 0.000125 Time 0.021628 +2023-10-05 22:09:59,881 - Epoch: [186][ 900/ 1236] Overall Loss 0.169261 Objective Loss 0.169261 LR 0.000125 Time 0.021623 +2023-10-05 22:10:00,097 - Epoch: [186][ 910/ 1236] Overall Loss 0.169297 Objective Loss 0.169297 LR 0.000125 Time 0.021622 +2023-10-05 22:10:00,309 - Epoch: [186][ 920/ 1236] Overall Loss 0.169225 Objective Loss 0.169225 LR 0.000125 Time 0.021618 +2023-10-05 22:10:00,526 - Epoch: [186][ 930/ 1236] Overall Loss 0.169449 Objective Loss 0.169449 LR 0.000125 Time 0.021617 +2023-10-05 22:10:00,738 - Epoch: [186][ 940/ 1236] Overall Loss 0.169412 Objective Loss 0.169412 LR 0.000125 Time 0.021613 +2023-10-05 22:10:00,954 - Epoch: [186][ 950/ 1236] Overall Loss 0.169315 Objective Loss 0.169315 LR 0.000125 Time 0.021612 +2023-10-05 22:10:01,160 - Epoch: [186][ 960/ 1236] Overall Loss 0.169256 Objective Loss 0.169256 LR 0.000125 Time 0.021601 +2023-10-05 22:10:01,365 - Epoch: [186][ 970/ 1236] Overall Loss 0.169263 Objective Loss 0.169263 LR 0.000125 Time 0.021590 +2023-10-05 22:10:01,571 - Epoch: [186][ 980/ 1236] Overall Loss 0.169191 Objective Loss 0.169191 LR 0.000125 Time 0.021579 +2023-10-05 22:10:01,776 - Epoch: [186][ 990/ 1236] Overall Loss 0.169294 Objective Loss 0.169294 LR 0.000125 Time 0.021569 +2023-10-05 22:10:01,983 - Epoch: [186][ 1000/ 1236] Overall Loss 0.169390 Objective Loss 0.169390 LR 0.000125 Time 0.021559 +2023-10-05 22:10:02,188 - Epoch: [186][ 1010/ 1236] Overall Loss 0.169354 Objective Loss 0.169354 LR 0.000125 Time 0.021549 +2023-10-05 22:10:02,395 - Epoch: [186][ 1020/ 1236] Overall Loss 0.169190 Objective Loss 0.169190 LR 0.000125 Time 0.021539 +2023-10-05 22:10:02,600 - Epoch: [186][ 1030/ 1236] Overall Loss 0.169014 Objective Loss 0.169014 LR 0.000125 Time 0.021529 +2023-10-05 22:10:02,806 - Epoch: [186][ 1040/ 1236] Overall Loss 0.168937 Objective Loss 0.168937 LR 0.000125 Time 0.021520 +2023-10-05 22:10:03,012 - Epoch: [186][ 1050/ 1236] Overall Loss 0.169188 Objective Loss 0.169188 LR 0.000125 Time 0.021510 +2023-10-05 22:10:03,218 - Epoch: [186][ 1060/ 1236] Overall Loss 0.169186 Objective Loss 0.169186 LR 0.000125 Time 0.021502 +2023-10-05 22:10:03,423 - Epoch: [186][ 1070/ 1236] Overall Loss 0.169173 Objective Loss 0.169173 LR 0.000125 Time 0.021492 +2023-10-05 22:10:03,630 - Epoch: [186][ 1080/ 1236] Overall Loss 0.169185 Objective Loss 0.169185 LR 0.000125 Time 0.021484 +2023-10-05 22:10:03,836 - Epoch: [186][ 1090/ 1236] Overall Loss 0.169372 Objective Loss 0.169372 LR 0.000125 Time 0.021476 +2023-10-05 22:10:04,042 - Epoch: [186][ 1100/ 1236] Overall Loss 0.169603 Objective Loss 0.169603 LR 0.000125 Time 0.021468 +2023-10-05 22:10:04,247 - Epoch: [186][ 1110/ 1236] Overall Loss 0.169506 Objective Loss 0.169506 LR 0.000125 Time 0.021459 +2023-10-05 22:10:04,454 - Epoch: [186][ 1120/ 1236] Overall Loss 0.169343 Objective Loss 0.169343 LR 0.000125 Time 0.021451 +2023-10-05 22:10:04,660 - Epoch: [186][ 1130/ 1236] Overall Loss 0.169403 Objective Loss 0.169403 LR 0.000125 Time 0.021443 +2023-10-05 22:10:04,866 - Epoch: [186][ 1140/ 1236] Overall Loss 0.169317 Objective Loss 0.169317 LR 0.000125 Time 0.021436 +2023-10-05 22:10:05,071 - Epoch: [186][ 1150/ 1236] Overall Loss 0.169444 Objective Loss 0.169444 LR 0.000125 Time 0.021428 +2023-10-05 22:10:05,278 - Epoch: [186][ 1160/ 1236] Overall Loss 0.169382 Objective Loss 0.169382 LR 0.000125 Time 0.021421 +2023-10-05 22:10:05,483 - Epoch: [186][ 1170/ 1236] Overall Loss 0.169498 Objective Loss 0.169498 LR 0.000125 Time 0.021413 +2023-10-05 22:10:05,689 - Epoch: [186][ 1180/ 1236] Overall Loss 0.169389 Objective Loss 0.169389 LR 0.000125 Time 0.021406 +2023-10-05 22:10:05,895 - Epoch: [186][ 1190/ 1236] Overall Loss 0.169345 Objective Loss 0.169345 LR 0.000125 Time 0.021398 +2023-10-05 22:10:06,101 - Epoch: [186][ 1200/ 1236] Overall Loss 0.169527 Objective Loss 0.169527 LR 0.000125 Time 0.021392 +2023-10-05 22:10:06,306 - Epoch: [186][ 1210/ 1236] Overall Loss 0.169516 Objective Loss 0.169516 LR 0.000125 Time 0.021384 +2023-10-05 22:10:06,513 - Epoch: [186][ 1220/ 1236] Overall Loss 0.169574 Objective Loss 0.169574 LR 0.000125 Time 0.021378 +2023-10-05 22:10:06,770 - Epoch: [186][ 1230/ 1236] Overall Loss 0.169642 Objective Loss 0.169642 LR 0.000125 Time 0.021413 +2023-10-05 22:10:06,889 - Epoch: [186][ 1236/ 1236] Overall Loss 0.169642 Objective Loss 0.169642 Top1 89.816701 Top5 98.981670 LR 0.000125 Time 0.021406 +2023-10-05 22:10:07,027 - --- validate (epoch=186)----------- +2023-10-05 22:10:07,027 - 29943 samples (256 per mini-batch) +2023-10-05 22:10:07,484 - Epoch: [186][ 10/ 117] Loss 0.275018 Top1 86.015625 Top5 98.398438 +2023-10-05 22:10:07,636 - Epoch: [186][ 20/ 117] Loss 0.290873 Top1 85.605469 Top5 98.339844 +2023-10-05 22:10:07,786 - Epoch: [186][ 30/ 117] Loss 0.305439 Top1 85.468750 Top5 98.125000 +2023-10-05 22:10:07,939 - Epoch: [186][ 40/ 117] Loss 0.308913 Top1 85.341797 Top5 98.203125 +2023-10-05 22:10:08,090 - Epoch: [186][ 50/ 117] Loss 0.311864 Top1 85.195312 Top5 98.203125 +2023-10-05 22:10:08,242 - Epoch: [186][ 60/ 117] Loss 0.306447 Top1 85.670573 Top5 98.203125 +2023-10-05 22:10:08,394 - Epoch: [186][ 70/ 117] Loss 0.305797 Top1 85.630580 Top5 98.186384 +2023-10-05 22:10:08,543 - Epoch: [186][ 80/ 117] Loss 0.302285 Top1 85.605469 Top5 98.168945 +2023-10-05 22:10:08,692 - Epoch: [186][ 90/ 117] Loss 0.305168 Top1 85.607639 Top5 98.168403 +2023-10-05 22:10:08,843 - Epoch: [186][ 100/ 117] Loss 0.303852 Top1 85.515625 Top5 98.191406 +2023-10-05 22:10:09,000 - Epoch: [186][ 110/ 117] Loss 0.303209 Top1 85.543324 Top5 98.231534 +2023-10-05 22:10:09,087 - Epoch: [186][ 117/ 117] Loss 0.301523 Top1 85.595966 Top5 98.206593 +2023-10-05 22:10:09,243 - ==> Top1: 85.596 Top5: 98.207 Loss: 0.302 + +2023-10-05 22:10:09,244 - ==> Confusion: +[[ 939 2 2 0 8 2 0 0 6 64 1 0 1 2 5 2 2 1 1 0 12] + [ 0 1062 1 0 11 18 1 18 1 0 1 1 0 0 2 4 2 1 4 2 2] + [ 7 2 971 8 2 0 24 6 0 0 4 3 8 1 1 4 0 3 6 1 5] + [ 1 1 14 961 2 3 2 1 2 1 7 0 5 2 33 4 0 5 25 2 18] + [ 19 5 0 0 983 4 0 1 1 9 1 2 0 1 7 2 7 2 0 2 4] + [ 3 30 1 1 2 992 0 21 1 0 6 12 0 10 6 2 6 0 3 5 15] + [ 0 3 16 0 0 0 1139 9 0 0 2 3 1 0 1 5 0 0 2 7 3] + [ 3 17 14 0 2 26 3 1081 2 3 1 12 3 3 2 2 0 0 32 5 7] + [ 18 1 0 0 0 1 1 0 970 44 13 3 2 11 13 6 0 1 3 0 2] + [ 101 0 4 0 5 3 0 0 19 949 2 1 0 17 4 5 1 1 0 2 5] + [ 3 6 9 1 1 1 3 5 10 1 983 4 1 11 3 2 1 0 1 1 6] + [ 2 0 2 0 0 11 0 2 0 1 0 967 13 6 0 4 2 16 0 5 4] + [ 1 1 0 3 0 2 0 2 0 0 2 33 996 1 0 2 2 14 1 2 6] + [ 3 0 1 0 1 6 0 0 11 10 8 3 2 1061 3 1 1 0 0 0 8] + [ 13 4 3 8 8 1 0 0 18 3 2 1 1 3 1007 0 1 2 11 1 14] + [ 1 3 2 0 2 0 0 0 0 0 0 7 5 3 1 1072 16 12 0 9 1] + [ 0 13 1 0 7 4 0 0 2 0 0 4 2 0 3 12 1096 0 0 4 13] + [ 1 0 0 1 0 0 3 0 0 0 0 3 18 1 0 4 0 1002 2 0 3] + [ 2 10 7 17 2 0 0 23 1 0 1 0 1 0 11 0 0 0 987 1 5] + [ 0 2 2 2 2 6 7 7 0 0 3 15 1 1 0 6 6 3 3 1081 5] + [ 138 153 153 47 109 134 44 89 83 75 177 100 293 300 126 55 149 63 118 168 5331]] + +2023-10-05 22:10:09,245 - ==> Best [Top1: 85.846 Top5: 98.233 Sparsity:0.00 Params: 148928 on epoch: 173] +2023-10-05 22:10:09,245 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:10:09,251 - + +2023-10-05 22:10:09,251 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:10:10,246 - Epoch: [187][ 10/ 1236] Overall Loss 0.177755 Objective Loss 0.177755 LR 0.000125 Time 0.099472 +2023-10-05 22:10:10,452 - Epoch: [187][ 20/ 1236] Overall Loss 0.175832 Objective Loss 0.175832 LR 0.000125 Time 0.059981 +2023-10-05 22:10:10,653 - Epoch: [187][ 30/ 1236] Overall Loss 0.169243 Objective Loss 0.169243 LR 0.000125 Time 0.046690 +2023-10-05 22:10:10,857 - Epoch: [187][ 40/ 1236] Overall Loss 0.172396 Objective Loss 0.172396 LR 0.000125 Time 0.040113 +2023-10-05 22:10:11,058 - Epoch: [187][ 50/ 1236] Overall Loss 0.171652 Objective Loss 0.171652 LR 0.000125 Time 0.036109 +2023-10-05 22:10:11,262 - Epoch: [187][ 60/ 1236] Overall Loss 0.170665 Objective Loss 0.170665 LR 0.000125 Time 0.033482 +2023-10-05 22:10:11,464 - Epoch: [187][ 70/ 1236] Overall Loss 0.168963 Objective Loss 0.168963 LR 0.000125 Time 0.031580 +2023-10-05 22:10:11,669 - Epoch: [187][ 80/ 1236] Overall Loss 0.167895 Objective Loss 0.167895 LR 0.000125 Time 0.030181 +2023-10-05 22:10:11,871 - Epoch: [187][ 90/ 1236] Overall Loss 0.167839 Objective Loss 0.167839 LR 0.000125 Time 0.029071 +2023-10-05 22:10:12,076 - Epoch: [187][ 100/ 1236] Overall Loss 0.169232 Objective Loss 0.169232 LR 0.000125 Time 0.028210 +2023-10-05 22:10:12,277 - Epoch: [187][ 110/ 1236] Overall Loss 0.169975 Objective Loss 0.169975 LR 0.000125 Time 0.027472 +2023-10-05 22:10:12,480 - Epoch: [187][ 120/ 1236] Overall Loss 0.171138 Objective Loss 0.171138 LR 0.000125 Time 0.026869 +2023-10-05 22:10:12,683 - Epoch: [187][ 130/ 1236] Overall Loss 0.171453 Objective Loss 0.171453 LR 0.000125 Time 0.026364 +2023-10-05 22:10:12,887 - Epoch: [187][ 140/ 1236] Overall Loss 0.169585 Objective Loss 0.169585 LR 0.000125 Time 0.025934 +2023-10-05 22:10:13,091 - Epoch: [187][ 150/ 1236] Overall Loss 0.169036 Objective Loss 0.169036 LR 0.000125 Time 0.025564 +2023-10-05 22:10:13,295 - Epoch: [187][ 160/ 1236] Overall Loss 0.169356 Objective Loss 0.169356 LR 0.000125 Time 0.025239 +2023-10-05 22:10:13,500 - Epoch: [187][ 170/ 1236] Overall Loss 0.168690 Objective Loss 0.168690 LR 0.000125 Time 0.024954 +2023-10-05 22:10:13,702 - Epoch: [187][ 180/ 1236] Overall Loss 0.169399 Objective Loss 0.169399 LR 0.000125 Time 0.024690 +2023-10-05 22:10:13,904 - Epoch: [187][ 190/ 1236] Overall Loss 0.170561 Objective Loss 0.170561 LR 0.000125 Time 0.024452 +2023-10-05 22:10:14,107 - Epoch: [187][ 200/ 1236] Overall Loss 0.170168 Objective Loss 0.170168 LR 0.000125 Time 0.024241 +2023-10-05 22:10:14,309 - Epoch: [187][ 210/ 1236] Overall Loss 0.170564 Objective Loss 0.170564 LR 0.000125 Time 0.024048 +2023-10-05 22:10:14,511 - Epoch: [187][ 220/ 1236] Overall Loss 0.170260 Objective Loss 0.170260 LR 0.000125 Time 0.023872 +2023-10-05 22:10:14,713 - Epoch: [187][ 230/ 1236] Overall Loss 0.168895 Objective Loss 0.168895 LR 0.000125 Time 0.023711 +2023-10-05 22:10:14,916 - Epoch: [187][ 240/ 1236] Overall Loss 0.169088 Objective Loss 0.169088 LR 0.000125 Time 0.023566 +2023-10-05 22:10:15,116 - Epoch: [187][ 250/ 1236] Overall Loss 0.169527 Objective Loss 0.169527 LR 0.000125 Time 0.023424 +2023-10-05 22:10:15,319 - Epoch: [187][ 260/ 1236] Overall Loss 0.170533 Objective Loss 0.170533 LR 0.000125 Time 0.023300 +2023-10-05 22:10:15,521 - Epoch: [187][ 270/ 1236] Overall Loss 0.169976 Objective Loss 0.169976 LR 0.000125 Time 0.023185 +2023-10-05 22:10:15,723 - Epoch: [187][ 280/ 1236] Overall Loss 0.169442 Objective Loss 0.169442 LR 0.000125 Time 0.023079 +2023-10-05 22:10:15,925 - Epoch: [187][ 290/ 1236] Overall Loss 0.169222 Objective Loss 0.169222 LR 0.000125 Time 0.022978 +2023-10-05 22:10:16,128 - Epoch: [187][ 300/ 1236] Overall Loss 0.170351 Objective Loss 0.170351 LR 0.000125 Time 0.022886 +2023-10-05 22:10:16,330 - Epoch: [187][ 310/ 1236] Overall Loss 0.170035 Objective Loss 0.170035 LR 0.000125 Time 0.022798 +2023-10-05 22:10:16,532 - Epoch: [187][ 320/ 1236] Overall Loss 0.170055 Objective Loss 0.170055 LR 0.000125 Time 0.022716 +2023-10-05 22:10:16,733 - Epoch: [187][ 330/ 1236] Overall Loss 0.170194 Objective Loss 0.170194 LR 0.000125 Time 0.022638 +2023-10-05 22:10:16,936 - Epoch: [187][ 340/ 1236] Overall Loss 0.169889 Objective Loss 0.169889 LR 0.000125 Time 0.022567 +2023-10-05 22:10:17,138 - Epoch: [187][ 350/ 1236] Overall Loss 0.170574 Objective Loss 0.170574 LR 0.000125 Time 0.022498 +2023-10-05 22:10:17,341 - Epoch: [187][ 360/ 1236] Overall Loss 0.170766 Objective Loss 0.170766 LR 0.000125 Time 0.022435 +2023-10-05 22:10:17,545 - Epoch: [187][ 370/ 1236] Overall Loss 0.170612 Objective Loss 0.170612 LR 0.000125 Time 0.022380 +2023-10-05 22:10:17,751 - Epoch: [187][ 380/ 1236] Overall Loss 0.170418 Objective Loss 0.170418 LR 0.000125 Time 0.022334 +2023-10-05 22:10:17,954 - Epoch: [187][ 390/ 1236] Overall Loss 0.170093 Objective Loss 0.170093 LR 0.000125 Time 0.022279 +2023-10-05 22:10:18,161 - Epoch: [187][ 400/ 1236] Overall Loss 0.170049 Objective Loss 0.170049 LR 0.000125 Time 0.022239 +2023-10-05 22:10:18,364 - Epoch: [187][ 410/ 1236] Overall Loss 0.170046 Objective Loss 0.170046 LR 0.000125 Time 0.022192 +2023-10-05 22:10:18,571 - Epoch: [187][ 420/ 1236] Overall Loss 0.170183 Objective Loss 0.170183 LR 0.000125 Time 0.022154 +2023-10-05 22:10:18,773 - Epoch: [187][ 430/ 1236] Overall Loss 0.170252 Objective Loss 0.170252 LR 0.000125 Time 0.022109 +2023-10-05 22:10:18,979 - Epoch: [187][ 440/ 1236] Overall Loss 0.170633 Objective Loss 0.170633 LR 0.000125 Time 0.022073 +2023-10-05 22:10:19,182 - Epoch: [187][ 450/ 1236] Overall Loss 0.170443 Objective Loss 0.170443 LR 0.000125 Time 0.022033 +2023-10-05 22:10:19,386 - Epoch: [187][ 460/ 1236] Overall Loss 0.170841 Objective Loss 0.170841 LR 0.000125 Time 0.021996 +2023-10-05 22:10:19,588 - Epoch: [187][ 470/ 1236] Overall Loss 0.170874 Objective Loss 0.170874 LR 0.000125 Time 0.021958 +2023-10-05 22:10:19,792 - Epoch: [187][ 480/ 1236] Overall Loss 0.171336 Objective Loss 0.171336 LR 0.000125 Time 0.021926 +2023-10-05 22:10:19,995 - Epoch: [187][ 490/ 1236] Overall Loss 0.171560 Objective Loss 0.171560 LR 0.000125 Time 0.021892 +2023-10-05 22:10:20,199 - Epoch: [187][ 500/ 1236] Overall Loss 0.171478 Objective Loss 0.171478 LR 0.000125 Time 0.021861 +2023-10-05 22:10:20,400 - Epoch: [187][ 510/ 1236] Overall Loss 0.171356 Objective Loss 0.171356 LR 0.000125 Time 0.021825 +2023-10-05 22:10:20,605 - Epoch: [187][ 520/ 1236] Overall Loss 0.171083 Objective Loss 0.171083 LR 0.000125 Time 0.021799 +2023-10-05 22:10:20,808 - Epoch: [187][ 530/ 1236] Overall Loss 0.170967 Objective Loss 0.170967 LR 0.000125 Time 0.021770 +2023-10-05 22:10:21,012 - Epoch: [187][ 540/ 1236] Overall Loss 0.170894 Objective Loss 0.170894 LR 0.000125 Time 0.021745 +2023-10-05 22:10:21,215 - Epoch: [187][ 550/ 1236] Overall Loss 0.170489 Objective Loss 0.170489 LR 0.000125 Time 0.021718 +2023-10-05 22:10:21,419 - Epoch: [187][ 560/ 1236] Overall Loss 0.170404 Objective Loss 0.170404 LR 0.000125 Time 0.021694 +2023-10-05 22:10:21,622 - Epoch: [187][ 570/ 1236] Overall Loss 0.170107 Objective Loss 0.170107 LR 0.000125 Time 0.021669 +2023-10-05 22:10:21,826 - Epoch: [187][ 580/ 1236] Overall Loss 0.169920 Objective Loss 0.169920 LR 0.000125 Time 0.021647 +2023-10-05 22:10:22,030 - Epoch: [187][ 590/ 1236] Overall Loss 0.169957 Objective Loss 0.169957 LR 0.000125 Time 0.021624 +2023-10-05 22:10:22,234 - Epoch: [187][ 600/ 1236] Overall Loss 0.170267 Objective Loss 0.170267 LR 0.000125 Time 0.021603 +2023-10-05 22:10:22,436 - Epoch: [187][ 610/ 1236] Overall Loss 0.170026 Objective Loss 0.170026 LR 0.000125 Time 0.021579 +2023-10-05 22:10:22,641 - Epoch: [187][ 620/ 1236] Overall Loss 0.169879 Objective Loss 0.169879 LR 0.000125 Time 0.021562 +2023-10-05 22:10:22,844 - Epoch: [187][ 630/ 1236] Overall Loss 0.169680 Objective Loss 0.169680 LR 0.000125 Time 0.021541 +2023-10-05 22:10:23,048 - Epoch: [187][ 640/ 1236] Overall Loss 0.169358 Objective Loss 0.169358 LR 0.000125 Time 0.021523 +2023-10-05 22:10:23,251 - Epoch: [187][ 650/ 1236] Overall Loss 0.169363 Objective Loss 0.169363 LR 0.000125 Time 0.021504 +2023-10-05 22:10:23,456 - Epoch: [187][ 660/ 1236] Overall Loss 0.169632 Objective Loss 0.169632 LR 0.000125 Time 0.021487 +2023-10-05 22:10:23,659 - Epoch: [187][ 670/ 1236] Overall Loss 0.169506 Objective Loss 0.169506 LR 0.000125 Time 0.021469 +2023-10-05 22:10:23,863 - Epoch: [187][ 680/ 1236] Overall Loss 0.169707 Objective Loss 0.169707 LR 0.000125 Time 0.021454 +2023-10-05 22:10:24,067 - Epoch: [187][ 690/ 1236] Overall Loss 0.169784 Objective Loss 0.169784 LR 0.000125 Time 0.021437 +2023-10-05 22:10:24,270 - Epoch: [187][ 700/ 1236] Overall Loss 0.169523 Objective Loss 0.169523 LR 0.000125 Time 0.021421 +2023-10-05 22:10:24,473 - Epoch: [187][ 710/ 1236] Overall Loss 0.169542 Objective Loss 0.169542 LR 0.000125 Time 0.021404 +2023-10-05 22:10:24,677 - Epoch: [187][ 720/ 1236] Overall Loss 0.169274 Objective Loss 0.169274 LR 0.000125 Time 0.021390 +2023-10-05 22:10:24,880 - Epoch: [187][ 730/ 1236] Overall Loss 0.169258 Objective Loss 0.169258 LR 0.000125 Time 0.021375 +2023-10-05 22:10:25,085 - Epoch: [187][ 740/ 1236] Overall Loss 0.169160 Objective Loss 0.169160 LR 0.000125 Time 0.021362 +2023-10-05 22:10:25,288 - Epoch: [187][ 750/ 1236] Overall Loss 0.168963 Objective Loss 0.168963 LR 0.000125 Time 0.021347 +2023-10-05 22:10:25,493 - Epoch: [187][ 760/ 1236] Overall Loss 0.168928 Objective Loss 0.168928 LR 0.000125 Time 0.021336 +2023-10-05 22:10:25,696 - Epoch: [187][ 770/ 1236] Overall Loss 0.169228 Objective Loss 0.169228 LR 0.000125 Time 0.021322 +2023-10-05 22:10:25,900 - Epoch: [187][ 780/ 1236] Overall Loss 0.169199 Objective Loss 0.169199 LR 0.000125 Time 0.021310 +2023-10-05 22:10:26,103 - Epoch: [187][ 790/ 1236] Overall Loss 0.168981 Objective Loss 0.168981 LR 0.000125 Time 0.021297 +2023-10-05 22:10:26,308 - Epoch: [187][ 800/ 1236] Overall Loss 0.169235 Objective Loss 0.169235 LR 0.000125 Time 0.021286 +2023-10-05 22:10:26,511 - Epoch: [187][ 810/ 1236] Overall Loss 0.169204 Objective Loss 0.169204 LR 0.000125 Time 0.021273 +2023-10-05 22:10:26,715 - Epoch: [187][ 820/ 1236] Overall Loss 0.169112 Objective Loss 0.169112 LR 0.000125 Time 0.021262 +2023-10-05 22:10:26,918 - Epoch: [187][ 830/ 1236] Overall Loss 0.169012 Objective Loss 0.169012 LR 0.000125 Time 0.021250 +2023-10-05 22:10:27,122 - Epoch: [187][ 840/ 1236] Overall Loss 0.169045 Objective Loss 0.169045 LR 0.000125 Time 0.021240 +2023-10-05 22:10:27,326 - Epoch: [187][ 850/ 1236] Overall Loss 0.169195 Objective Loss 0.169195 LR 0.000125 Time 0.021229 +2023-10-05 22:10:27,531 - Epoch: [187][ 860/ 1236] Overall Loss 0.169329 Objective Loss 0.169329 LR 0.000125 Time 0.021220 +2023-10-05 22:10:27,734 - Epoch: [187][ 870/ 1236] Overall Loss 0.169450 Objective Loss 0.169450 LR 0.000125 Time 0.021209 +2023-10-05 22:10:27,938 - Epoch: [187][ 880/ 1236] Overall Loss 0.169587 Objective Loss 0.169587 LR 0.000125 Time 0.021200 +2023-10-05 22:10:28,142 - Epoch: [187][ 890/ 1236] Overall Loss 0.169597 Objective Loss 0.169597 LR 0.000125 Time 0.021190 +2023-10-05 22:10:28,347 - Epoch: [187][ 900/ 1236] Overall Loss 0.169343 Objective Loss 0.169343 LR 0.000125 Time 0.021182 +2023-10-05 22:10:28,550 - Epoch: [187][ 910/ 1236] Overall Loss 0.169273 Objective Loss 0.169273 LR 0.000125 Time 0.021172 +2023-10-05 22:10:28,754 - Epoch: [187][ 920/ 1236] Overall Loss 0.169238 Objective Loss 0.169238 LR 0.000125 Time 0.021163 +2023-10-05 22:10:28,955 - Epoch: [187][ 930/ 1236] Overall Loss 0.169252 Objective Loss 0.169252 LR 0.000125 Time 0.021152 +2023-10-05 22:10:29,159 - Epoch: [187][ 940/ 1236] Overall Loss 0.169213 Objective Loss 0.169213 LR 0.000125 Time 0.021144 +2023-10-05 22:10:29,360 - Epoch: [187][ 950/ 1236] Overall Loss 0.169226 Objective Loss 0.169226 LR 0.000125 Time 0.021132 +2023-10-05 22:10:29,564 - Epoch: [187][ 960/ 1236] Overall Loss 0.169475 Objective Loss 0.169475 LR 0.000125 Time 0.021124 +2023-10-05 22:10:29,766 - Epoch: [187][ 970/ 1236] Overall Loss 0.169595 Objective Loss 0.169595 LR 0.000125 Time 0.021114 +2023-10-05 22:10:29,971 - Epoch: [187][ 980/ 1236] Overall Loss 0.169822 Objective Loss 0.169822 LR 0.000125 Time 0.021107 +2023-10-05 22:10:30,173 - Epoch: [187][ 990/ 1236] Overall Loss 0.169636 Objective Loss 0.169636 LR 0.000125 Time 0.021098 +2023-10-05 22:10:30,378 - Epoch: [187][ 1000/ 1236] Overall Loss 0.169714 Objective Loss 0.169714 LR 0.000125 Time 0.021091 +2023-10-05 22:10:30,580 - Epoch: [187][ 1010/ 1236] Overall Loss 0.169902 Objective Loss 0.169902 LR 0.000125 Time 0.021083 +2023-10-05 22:10:30,785 - Epoch: [187][ 1020/ 1236] Overall Loss 0.169932 Objective Loss 0.169932 LR 0.000125 Time 0.021076 +2023-10-05 22:10:30,987 - Epoch: [187][ 1030/ 1236] Overall Loss 0.170045 Objective Loss 0.170045 LR 0.000125 Time 0.021068 +2023-10-05 22:10:31,192 - Epoch: [187][ 1040/ 1236] Overall Loss 0.169882 Objective Loss 0.169882 LR 0.000125 Time 0.021062 +2023-10-05 22:10:31,395 - Epoch: [187][ 1050/ 1236] Overall Loss 0.169845 Objective Loss 0.169845 LR 0.000125 Time 0.021054 +2023-10-05 22:10:31,600 - Epoch: [187][ 1060/ 1236] Overall Loss 0.169860 Objective Loss 0.169860 LR 0.000125 Time 0.021048 +2023-10-05 22:10:31,802 - Epoch: [187][ 1070/ 1236] Overall Loss 0.170061 Objective Loss 0.170061 LR 0.000125 Time 0.021041 +2023-10-05 22:10:32,007 - Epoch: [187][ 1080/ 1236] Overall Loss 0.170215 Objective Loss 0.170215 LR 0.000125 Time 0.021035 +2023-10-05 22:10:32,209 - Epoch: [187][ 1090/ 1236] Overall Loss 0.170229 Objective Loss 0.170229 LR 0.000125 Time 0.021027 +2023-10-05 22:10:32,415 - Epoch: [187][ 1100/ 1236] Overall Loss 0.170184 Objective Loss 0.170184 LR 0.000125 Time 0.021023 +2023-10-05 22:10:32,618 - Epoch: [187][ 1110/ 1236] Overall Loss 0.170268 Objective Loss 0.170268 LR 0.000125 Time 0.021016 +2023-10-05 22:10:32,823 - Epoch: [187][ 1120/ 1236] Overall Loss 0.170359 Objective Loss 0.170359 LR 0.000125 Time 0.021011 +2023-10-05 22:10:33,025 - Epoch: [187][ 1130/ 1236] Overall Loss 0.170189 Objective Loss 0.170189 LR 0.000125 Time 0.021004 +2023-10-05 22:10:33,230 - Epoch: [187][ 1140/ 1236] Overall Loss 0.170158 Objective Loss 0.170158 LR 0.000125 Time 0.020999 +2023-10-05 22:10:33,433 - Epoch: [187][ 1150/ 1236] Overall Loss 0.170156 Objective Loss 0.170156 LR 0.000125 Time 0.020992 +2023-10-05 22:10:33,638 - Epoch: [187][ 1160/ 1236] Overall Loss 0.170297 Objective Loss 0.170297 LR 0.000125 Time 0.020988 +2023-10-05 22:10:33,840 - Epoch: [187][ 1170/ 1236] Overall Loss 0.170279 Objective Loss 0.170279 LR 0.000125 Time 0.020981 +2023-10-05 22:10:34,045 - Epoch: [187][ 1180/ 1236] Overall Loss 0.170327 Objective Loss 0.170327 LR 0.000125 Time 0.020976 +2023-10-05 22:10:34,252 - Epoch: [187][ 1190/ 1236] Overall Loss 0.170330 Objective Loss 0.170330 LR 0.000125 Time 0.020974 +2023-10-05 22:10:34,466 - Epoch: [187][ 1200/ 1236] Overall Loss 0.170302 Objective Loss 0.170302 LR 0.000125 Time 0.020977 +2023-10-05 22:10:34,676 - Epoch: [187][ 1210/ 1236] Overall Loss 0.170378 Objective Loss 0.170378 LR 0.000125 Time 0.020977 +2023-10-05 22:10:34,891 - Epoch: [187][ 1220/ 1236] Overall Loss 0.170383 Objective Loss 0.170383 LR 0.000125 Time 0.020981 +2023-10-05 22:10:35,153 - Epoch: [187][ 1230/ 1236] Overall Loss 0.170386 Objective Loss 0.170386 LR 0.000125 Time 0.021023 +2023-10-05 22:10:35,273 - Epoch: [187][ 1236/ 1236] Overall Loss 0.170322 Objective Loss 0.170322 Top1 93.686354 Top5 98.981670 LR 0.000125 Time 0.021018 +2023-10-05 22:10:35,407 - --- validate (epoch=187)----------- +2023-10-05 22:10:35,407 - 29943 samples (256 per mini-batch) +2023-10-05 22:10:35,866 - Epoch: [187][ 10/ 117] Loss 0.275152 Top1 86.132812 Top5 98.125000 +2023-10-05 22:10:36,017 - Epoch: [187][ 20/ 117] Loss 0.282871 Top1 86.542969 Top5 98.164062 +2023-10-05 22:10:36,165 - Epoch: [187][ 30/ 117] Loss 0.294840 Top1 86.158854 Top5 98.242188 +2023-10-05 22:10:36,312 - Epoch: [187][ 40/ 117] Loss 0.288254 Top1 86.113281 Top5 98.339844 +2023-10-05 22:10:36,459 - Epoch: [187][ 50/ 117] Loss 0.296693 Top1 86.070312 Top5 98.335938 +2023-10-05 22:10:36,607 - Epoch: [187][ 60/ 117] Loss 0.298909 Top1 86.145833 Top5 98.300781 +2023-10-05 22:10:36,755 - Epoch: [187][ 70/ 117] Loss 0.299011 Top1 86.077009 Top5 98.286830 +2023-10-05 22:10:36,902 - Epoch: [187][ 80/ 117] Loss 0.295500 Top1 86.176758 Top5 98.349609 +2023-10-05 22:10:37,051 - Epoch: [187][ 90/ 117] Loss 0.296877 Top1 86.167535 Top5 98.333333 +2023-10-05 22:10:37,199 - Epoch: [187][ 100/ 117] Loss 0.297198 Top1 86.105469 Top5 98.316406 +2023-10-05 22:10:37,353 - Epoch: [187][ 110/ 117] Loss 0.295632 Top1 86.175426 Top5 98.327415 +2023-10-05 22:10:37,442 - Epoch: [187][ 117/ 117] Loss 0.298393 Top1 86.170390 Top5 98.326821 +2023-10-05 22:10:37,565 - ==> Top1: 86.170 Top5: 98.327 Loss: 0.298 + +2023-10-05 22:10:37,565 - ==> Confusion: +[[ 936 3 5 3 6 1 0 0 4 67 1 0 2 2 4 1 1 1 1 0 12] + [ 0 1064 2 0 8 18 1 15 0 0 2 1 0 0 1 3 2 1 7 2 4] + [ 4 3 976 12 1 0 15 5 0 1 3 3 6 1 2 3 0 3 6 3 9] + [ 1 1 12 975 1 3 2 2 2 3 5 1 9 2 27 3 0 5 17 1 17] + [ 25 7 0 0 970 5 1 1 0 12 1 2 0 1 6 1 7 1 0 2 8] + [ 4 30 1 1 2 992 1 22 1 1 5 11 0 11 6 2 2 0 1 5 18] + [ 0 3 23 0 0 0 1133 9 0 1 2 2 2 0 1 6 0 0 2 4 3] + [ 5 17 12 0 2 31 2 1076 1 5 5 10 1 3 1 0 0 0 31 5 11] + [ 21 1 2 0 0 1 0 1 981 42 7 3 2 7 10 3 1 0 3 1 3] + [ 91 1 4 0 5 3 0 0 15 969 1 1 0 13 2 3 0 3 0 0 8] + [ 2 7 9 3 1 0 3 2 10 1 980 4 0 11 3 2 1 0 3 1 10] + [ 0 0 2 0 0 9 0 3 0 1 0 967 18 7 0 4 1 15 0 4 4] + [ 0 0 1 3 0 2 0 2 0 1 2 28 998 2 1 2 2 12 1 2 9] + [ 3 0 1 0 1 6 0 1 11 19 8 2 3 1051 3 1 1 0 0 1 7] + [ 14 3 3 10 4 1 0 0 28 3 3 1 1 3 1000 0 1 2 9 0 15] + [ 0 2 3 0 4 0 1 0 0 1 1 7 6 3 0 1071 15 8 0 8 4] + [ 2 12 1 0 6 3 0 0 1 0 0 6 0 1 3 11 1096 0 1 5 13] + [ 1 0 1 3 1 0 3 0 1 1 0 2 16 2 0 3 0 1000 0 1 3] + [ 0 6 4 18 1 0 0 26 2 0 2 0 1 0 12 0 1 0 987 0 8] + [ 0 2 2 4 2 4 9 8 0 0 1 14 3 1 0 6 6 1 3 1081 5] + [ 128 163 144 56 66 116 30 85 94 85 164 99 311 264 113 47 96 59 119 166 5500]] + +2023-10-05 22:10:37,567 - ==> Best [Top1: 86.170 Top5: 98.327 Sparsity:0.00 Params: 148928 on epoch: 187] +2023-10-05 22:10:37,567 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:10:37,575 - + +2023-10-05 22:10:37,575 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:10:38,549 - Epoch: [188][ 10/ 1236] Overall Loss 0.153864 Objective Loss 0.153864 LR 0.000125 Time 0.097344 +2023-10-05 22:10:38,748 - Epoch: [188][ 20/ 1236] Overall Loss 0.160059 Objective Loss 0.160059 LR 0.000125 Time 0.058621 +2023-10-05 22:10:38,946 - Epoch: [188][ 30/ 1236] Overall Loss 0.162391 Objective Loss 0.162391 LR 0.000125 Time 0.045663 +2023-10-05 22:10:39,145 - Epoch: [188][ 40/ 1236] Overall Loss 0.164839 Objective Loss 0.164839 LR 0.000125 Time 0.039225 +2023-10-05 22:10:39,343 - Epoch: [188][ 50/ 1236] Overall Loss 0.168156 Objective Loss 0.168156 LR 0.000125 Time 0.035323 +2023-10-05 22:10:39,543 - Epoch: [188][ 60/ 1236] Overall Loss 0.164267 Objective Loss 0.164267 LR 0.000125 Time 0.032758 +2023-10-05 22:10:39,741 - Epoch: [188][ 70/ 1236] Overall Loss 0.167919 Objective Loss 0.167919 LR 0.000125 Time 0.030901 +2023-10-05 22:10:39,941 - Epoch: [188][ 80/ 1236] Overall Loss 0.169703 Objective Loss 0.169703 LR 0.000125 Time 0.029535 +2023-10-05 22:10:40,138 - Epoch: [188][ 90/ 1236] Overall Loss 0.168548 Objective Loss 0.168548 LR 0.000125 Time 0.028443 +2023-10-05 22:10:40,340 - Epoch: [188][ 100/ 1236] Overall Loss 0.169546 Objective Loss 0.169546 LR 0.000125 Time 0.027611 +2023-10-05 22:10:40,539 - Epoch: [188][ 110/ 1236] Overall Loss 0.168997 Objective Loss 0.168997 LR 0.000125 Time 0.026908 +2023-10-05 22:10:40,737 - Epoch: [188][ 120/ 1236] Overall Loss 0.169951 Objective Loss 0.169951 LR 0.000125 Time 0.026315 +2023-10-05 22:10:40,936 - Epoch: [188][ 130/ 1236] Overall Loss 0.170146 Objective Loss 0.170146 LR 0.000125 Time 0.025814 +2023-10-05 22:10:41,135 - Epoch: [188][ 140/ 1236] Overall Loss 0.169946 Objective Loss 0.169946 LR 0.000125 Time 0.025390 +2023-10-05 22:10:41,332 - Epoch: [188][ 150/ 1236] Overall Loss 0.170237 Objective Loss 0.170237 LR 0.000125 Time 0.025008 +2023-10-05 22:10:41,531 - Epoch: [188][ 160/ 1236] Overall Loss 0.169089 Objective Loss 0.169089 LR 0.000125 Time 0.024691 +2023-10-05 22:10:41,730 - Epoch: [188][ 170/ 1236] Overall Loss 0.170595 Objective Loss 0.170595 LR 0.000125 Time 0.024404 +2023-10-05 22:10:41,930 - Epoch: [188][ 180/ 1236] Overall Loss 0.170207 Objective Loss 0.170207 LR 0.000125 Time 0.024156 +2023-10-05 22:10:42,128 - Epoch: [188][ 190/ 1236] Overall Loss 0.170784 Objective Loss 0.170784 LR 0.000125 Time 0.023927 +2023-10-05 22:10:42,328 - Epoch: [188][ 200/ 1236] Overall Loss 0.170713 Objective Loss 0.170713 LR 0.000125 Time 0.023727 +2023-10-05 22:10:42,526 - Epoch: [188][ 210/ 1236] Overall Loss 0.170442 Objective Loss 0.170442 LR 0.000125 Time 0.023541 +2023-10-05 22:10:42,726 - Epoch: [188][ 220/ 1236] Overall Loss 0.169952 Objective Loss 0.169952 LR 0.000125 Time 0.023379 +2023-10-05 22:10:42,924 - Epoch: [188][ 230/ 1236] Overall Loss 0.169682 Objective Loss 0.169682 LR 0.000125 Time 0.023222 +2023-10-05 22:10:43,124 - Epoch: [188][ 240/ 1236] Overall Loss 0.168357 Objective Loss 0.168357 LR 0.000125 Time 0.023085 +2023-10-05 22:10:43,326 - Epoch: [188][ 250/ 1236] Overall Loss 0.167958 Objective Loss 0.167958 LR 0.000125 Time 0.022966 +2023-10-05 22:10:43,525 - Epoch: [188][ 260/ 1236] Overall Loss 0.168159 Objective Loss 0.168159 LR 0.000125 Time 0.022849 +2023-10-05 22:10:43,724 - Epoch: [188][ 270/ 1236] Overall Loss 0.168604 Objective Loss 0.168604 LR 0.000125 Time 0.022738 +2023-10-05 22:10:43,924 - Epoch: [188][ 280/ 1236] Overall Loss 0.168905 Objective Loss 0.168905 LR 0.000125 Time 0.022638 +2023-10-05 22:10:44,123 - Epoch: [188][ 290/ 1236] Overall Loss 0.169256 Objective Loss 0.169256 LR 0.000125 Time 0.022541 +2023-10-05 22:10:44,322 - Epoch: [188][ 300/ 1236] Overall Loss 0.168983 Objective Loss 0.168983 LR 0.000125 Time 0.022453 +2023-10-05 22:10:44,519 - Epoch: [188][ 310/ 1236] Overall Loss 0.168905 Objective Loss 0.168905 LR 0.000125 Time 0.022365 +2023-10-05 22:10:44,719 - Epoch: [188][ 320/ 1236] Overall Loss 0.168849 Objective Loss 0.168849 LR 0.000125 Time 0.022289 +2023-10-05 22:10:44,918 - Epoch: [188][ 330/ 1236] Overall Loss 0.168573 Objective Loss 0.168573 LR 0.000125 Time 0.022215 +2023-10-05 22:10:45,118 - Epoch: [188][ 340/ 1236] Overall Loss 0.167994 Objective Loss 0.167994 LR 0.000125 Time 0.022148 +2023-10-05 22:10:45,318 - Epoch: [188][ 350/ 1236] Overall Loss 0.168784 Objective Loss 0.168784 LR 0.000125 Time 0.022087 +2023-10-05 22:10:45,521 - Epoch: [188][ 360/ 1236] Overall Loss 0.168638 Objective Loss 0.168638 LR 0.000125 Time 0.022035 +2023-10-05 22:10:45,720 - Epoch: [188][ 370/ 1236] Overall Loss 0.168397 Objective Loss 0.168397 LR 0.000125 Time 0.021977 +2023-10-05 22:10:45,924 - Epoch: [188][ 380/ 1236] Overall Loss 0.167888 Objective Loss 0.167888 LR 0.000125 Time 0.021935 +2023-10-05 22:10:46,131 - Epoch: [188][ 390/ 1236] Overall Loss 0.167739 Objective Loss 0.167739 LR 0.000125 Time 0.021901 +2023-10-05 22:10:46,336 - Epoch: [188][ 400/ 1236] Overall Loss 0.167724 Objective Loss 0.167724 LR 0.000125 Time 0.021867 +2023-10-05 22:10:46,533 - Epoch: [188][ 410/ 1236] Overall Loss 0.167352 Objective Loss 0.167352 LR 0.000125 Time 0.021813 +2023-10-05 22:10:46,736 - Epoch: [188][ 420/ 1236] Overall Loss 0.167660 Objective Loss 0.167660 LR 0.000125 Time 0.021776 +2023-10-05 22:10:46,937 - Epoch: [188][ 430/ 1236] Overall Loss 0.167553 Objective Loss 0.167553 LR 0.000125 Time 0.021736 +2023-10-05 22:10:47,138 - Epoch: [188][ 440/ 1236] Overall Loss 0.167370 Objective Loss 0.167370 LR 0.000125 Time 0.021698 +2023-10-05 22:10:47,339 - Epoch: [188][ 450/ 1236] Overall Loss 0.166705 Objective Loss 0.166705 LR 0.000125 Time 0.021660 +2023-10-05 22:10:47,541 - Epoch: [188][ 460/ 1236] Overall Loss 0.166679 Objective Loss 0.166679 LR 0.000125 Time 0.021628 +2023-10-05 22:10:47,741 - Epoch: [188][ 470/ 1236] Overall Loss 0.166759 Objective Loss 0.166759 LR 0.000125 Time 0.021591 +2023-10-05 22:10:47,943 - Epoch: [188][ 480/ 1236] Overall Loss 0.166988 Objective Loss 0.166988 LR 0.000125 Time 0.021560 +2023-10-05 22:10:48,142 - Epoch: [188][ 490/ 1236] Overall Loss 0.166751 Objective Loss 0.166751 LR 0.000125 Time 0.021524 +2023-10-05 22:10:48,344 - Epoch: [188][ 500/ 1236] Overall Loss 0.166811 Objective Loss 0.166811 LR 0.000125 Time 0.021497 +2023-10-05 22:10:48,544 - Epoch: [188][ 510/ 1236] Overall Loss 0.166924 Objective Loss 0.166924 LR 0.000125 Time 0.021467 +2023-10-05 22:10:48,746 - Epoch: [188][ 520/ 1236] Overall Loss 0.167039 Objective Loss 0.167039 LR 0.000125 Time 0.021441 +2023-10-05 22:10:48,946 - Epoch: [188][ 530/ 1236] Overall Loss 0.166965 Objective Loss 0.166965 LR 0.000125 Time 0.021411 +2023-10-05 22:10:49,148 - Epoch: [188][ 540/ 1236] Overall Loss 0.166666 Objective Loss 0.166666 LR 0.000125 Time 0.021388 +2023-10-05 22:10:49,348 - Epoch: [188][ 550/ 1236] Overall Loss 0.166811 Objective Loss 0.166811 LR 0.000125 Time 0.021362 +2023-10-05 22:10:49,549 - Epoch: [188][ 560/ 1236] Overall Loss 0.166653 Objective Loss 0.166653 LR 0.000125 Time 0.021340 +2023-10-05 22:10:49,749 - Epoch: [188][ 570/ 1236] Overall Loss 0.166623 Objective Loss 0.166623 LR 0.000125 Time 0.021315 +2023-10-05 22:10:49,950 - Epoch: [188][ 580/ 1236] Overall Loss 0.166409 Objective Loss 0.166409 LR 0.000125 Time 0.021294 +2023-10-05 22:10:50,150 - Epoch: [188][ 590/ 1236] Overall Loss 0.166338 Objective Loss 0.166338 LR 0.000125 Time 0.021271 +2023-10-05 22:10:50,352 - Epoch: [188][ 600/ 1236] Overall Loss 0.166033 Objective Loss 0.166033 LR 0.000125 Time 0.021252 +2023-10-05 22:10:50,552 - Epoch: [188][ 610/ 1236] Overall Loss 0.166051 Objective Loss 0.166051 LR 0.000125 Time 0.021228 +2023-10-05 22:10:50,753 - Epoch: [188][ 620/ 1236] Overall Loss 0.165913 Objective Loss 0.165913 LR 0.000125 Time 0.021210 +2023-10-05 22:10:50,953 - Epoch: [188][ 630/ 1236] Overall Loss 0.165888 Objective Loss 0.165888 LR 0.000125 Time 0.021191 +2023-10-05 22:10:51,155 - Epoch: [188][ 640/ 1236] Overall Loss 0.166138 Objective Loss 0.166138 LR 0.000125 Time 0.021175 +2023-10-05 22:10:51,356 - Epoch: [188][ 650/ 1236] Overall Loss 0.166034 Objective Loss 0.166034 LR 0.000125 Time 0.021156 +2023-10-05 22:10:51,556 - Epoch: [188][ 660/ 1236] Overall Loss 0.165923 Objective Loss 0.165923 LR 0.000125 Time 0.021139 +2023-10-05 22:10:51,757 - Epoch: [188][ 670/ 1236] Overall Loss 0.166139 Objective Loss 0.166139 LR 0.000125 Time 0.021122 +2023-10-05 22:10:51,958 - Epoch: [188][ 680/ 1236] Overall Loss 0.166168 Objective Loss 0.166168 LR 0.000125 Time 0.021107 +2023-10-05 22:10:52,158 - Epoch: [188][ 690/ 1236] Overall Loss 0.166173 Objective Loss 0.166173 LR 0.000125 Time 0.021089 +2023-10-05 22:10:52,360 - Epoch: [188][ 700/ 1236] Overall Loss 0.166395 Objective Loss 0.166395 LR 0.000125 Time 0.021076 +2023-10-05 22:10:52,560 - Epoch: [188][ 710/ 1236] Overall Loss 0.166196 Objective Loss 0.166196 LR 0.000125 Time 0.021058 +2023-10-05 22:10:52,761 - Epoch: [188][ 720/ 1236] Overall Loss 0.166441 Objective Loss 0.166441 LR 0.000125 Time 0.021045 +2023-10-05 22:10:52,962 - Epoch: [188][ 730/ 1236] Overall Loss 0.166594 Objective Loss 0.166594 LR 0.000125 Time 0.021031 +2023-10-05 22:10:53,164 - Epoch: [188][ 740/ 1236] Overall Loss 0.166536 Objective Loss 0.166536 LR 0.000125 Time 0.021019 +2023-10-05 22:10:53,364 - Epoch: [188][ 750/ 1236] Overall Loss 0.166794 Objective Loss 0.166794 LR 0.000125 Time 0.021004 +2023-10-05 22:10:53,566 - Epoch: [188][ 760/ 1236] Overall Loss 0.166796 Objective Loss 0.166796 LR 0.000125 Time 0.020993 +2023-10-05 22:10:53,766 - Epoch: [188][ 770/ 1236] Overall Loss 0.166814 Objective Loss 0.166814 LR 0.000125 Time 0.020978 +2023-10-05 22:10:53,968 - Epoch: [188][ 780/ 1236] Overall Loss 0.166996 Objective Loss 0.166996 LR 0.000125 Time 0.020967 +2023-10-05 22:10:54,168 - Epoch: [188][ 790/ 1236] Overall Loss 0.167000 Objective Loss 0.167000 LR 0.000125 Time 0.020953 +2023-10-05 22:10:54,370 - Epoch: [188][ 800/ 1236] Overall Loss 0.166740 Objective Loss 0.166740 LR 0.000125 Time 0.020943 +2023-10-05 22:10:54,570 - Epoch: [188][ 810/ 1236] Overall Loss 0.166750 Objective Loss 0.166750 LR 0.000125 Time 0.020930 +2023-10-05 22:10:54,772 - Epoch: [188][ 820/ 1236] Overall Loss 0.167095 Objective Loss 0.167095 LR 0.000125 Time 0.020920 +2023-10-05 22:10:54,973 - Epoch: [188][ 830/ 1236] Overall Loss 0.167075 Objective Loss 0.167075 LR 0.000125 Time 0.020910 +2023-10-05 22:10:55,176 - Epoch: [188][ 840/ 1236] Overall Loss 0.167256 Objective Loss 0.167256 LR 0.000125 Time 0.020902 +2023-10-05 22:10:55,377 - Epoch: [188][ 850/ 1236] Overall Loss 0.167163 Objective Loss 0.167163 LR 0.000125 Time 0.020892 +2023-10-05 22:10:55,581 - Epoch: [188][ 860/ 1236] Overall Loss 0.167185 Objective Loss 0.167185 LR 0.000125 Time 0.020886 +2023-10-05 22:10:55,782 - Epoch: [188][ 870/ 1236] Overall Loss 0.167071 Objective Loss 0.167071 LR 0.000125 Time 0.020876 +2023-10-05 22:10:55,986 - Epoch: [188][ 880/ 1236] Overall Loss 0.167070 Objective Loss 0.167070 LR 0.000125 Time 0.020871 +2023-10-05 22:10:56,187 - Epoch: [188][ 890/ 1236] Overall Loss 0.167264 Objective Loss 0.167264 LR 0.000125 Time 0.020861 +2023-10-05 22:10:56,391 - Epoch: [188][ 900/ 1236] Overall Loss 0.167302 Objective Loss 0.167302 LR 0.000125 Time 0.020856 +2023-10-05 22:10:56,592 - Epoch: [188][ 910/ 1236] Overall Loss 0.167477 Objective Loss 0.167477 LR 0.000125 Time 0.020847 +2023-10-05 22:10:56,796 - Epoch: [188][ 920/ 1236] Overall Loss 0.167474 Objective Loss 0.167474 LR 0.000125 Time 0.020842 +2023-10-05 22:10:56,997 - Epoch: [188][ 930/ 1236] Overall Loss 0.167650 Objective Loss 0.167650 LR 0.000125 Time 0.020833 +2023-10-05 22:10:57,201 - Epoch: [188][ 940/ 1236] Overall Loss 0.167847 Objective Loss 0.167847 LR 0.000125 Time 0.020828 +2023-10-05 22:10:57,402 - Epoch: [188][ 950/ 1236] Overall Loss 0.167702 Objective Loss 0.167702 LR 0.000125 Time 0.020820 +2023-10-05 22:10:57,606 - Epoch: [188][ 960/ 1236] Overall Loss 0.167673 Objective Loss 0.167673 LR 0.000125 Time 0.020816 +2023-10-05 22:10:57,807 - Epoch: [188][ 970/ 1236] Overall Loss 0.167650 Objective Loss 0.167650 LR 0.000125 Time 0.020808 +2023-10-05 22:10:58,012 - Epoch: [188][ 980/ 1236] Overall Loss 0.167561 Objective Loss 0.167561 LR 0.000125 Time 0.020804 +2023-10-05 22:10:58,212 - Epoch: [188][ 990/ 1236] Overall Loss 0.167686 Objective Loss 0.167686 LR 0.000125 Time 0.020796 +2023-10-05 22:10:58,417 - Epoch: [188][ 1000/ 1236] Overall Loss 0.167695 Objective Loss 0.167695 LR 0.000125 Time 0.020792 +2023-10-05 22:10:58,617 - Epoch: [188][ 1010/ 1236] Overall Loss 0.167582 Objective Loss 0.167582 LR 0.000125 Time 0.020784 +2023-10-05 22:10:58,822 - Epoch: [188][ 1020/ 1236] Overall Loss 0.167598 Objective Loss 0.167598 LR 0.000125 Time 0.020781 +2023-10-05 22:10:59,022 - Epoch: [188][ 1030/ 1236] Overall Loss 0.167540 Objective Loss 0.167540 LR 0.000125 Time 0.020773 +2023-10-05 22:10:59,227 - Epoch: [188][ 1040/ 1236] Overall Loss 0.167632 Objective Loss 0.167632 LR 0.000125 Time 0.020770 +2023-10-05 22:10:59,427 - Epoch: [188][ 1050/ 1236] Overall Loss 0.167563 Objective Loss 0.167563 LR 0.000125 Time 0.020762 +2023-10-05 22:10:59,631 - Epoch: [188][ 1060/ 1236] Overall Loss 0.167609 Objective Loss 0.167609 LR 0.000125 Time 0.020758 +2023-10-05 22:10:59,831 - Epoch: [188][ 1070/ 1236] Overall Loss 0.167578 Objective Loss 0.167578 LR 0.000125 Time 0.020751 +2023-10-05 22:11:00,036 - Epoch: [188][ 1080/ 1236] Overall Loss 0.167649 Objective Loss 0.167649 LR 0.000125 Time 0.020748 +2023-10-05 22:11:00,236 - Epoch: [188][ 1090/ 1236] Overall Loss 0.167628 Objective Loss 0.167628 LR 0.000125 Time 0.020741 +2023-10-05 22:11:00,441 - Epoch: [188][ 1100/ 1236] Overall Loss 0.167668 Objective Loss 0.167668 LR 0.000125 Time 0.020738 +2023-10-05 22:11:00,641 - Epoch: [188][ 1110/ 1236] Overall Loss 0.167423 Objective Loss 0.167423 LR 0.000125 Time 0.020732 +2023-10-05 22:11:00,846 - Epoch: [188][ 1120/ 1236] Overall Loss 0.167786 Objective Loss 0.167786 LR 0.000125 Time 0.020729 +2023-10-05 22:11:01,046 - Epoch: [188][ 1130/ 1236] Overall Loss 0.167972 Objective Loss 0.167972 LR 0.000125 Time 0.020722 +2023-10-05 22:11:01,250 - Epoch: [188][ 1140/ 1236] Overall Loss 0.167955 Objective Loss 0.167955 LR 0.000125 Time 0.020719 +2023-10-05 22:11:01,450 - Epoch: [188][ 1150/ 1236] Overall Loss 0.168073 Objective Loss 0.168073 LR 0.000125 Time 0.020713 +2023-10-05 22:11:01,655 - Epoch: [188][ 1160/ 1236] Overall Loss 0.168259 Objective Loss 0.168259 LR 0.000125 Time 0.020710 +2023-10-05 22:11:01,855 - Epoch: [188][ 1170/ 1236] Overall Loss 0.168413 Objective Loss 0.168413 LR 0.000125 Time 0.020704 +2023-10-05 22:11:02,060 - Epoch: [188][ 1180/ 1236] Overall Loss 0.168352 Objective Loss 0.168352 LR 0.000125 Time 0.020701 +2023-10-05 22:11:02,260 - Epoch: [188][ 1190/ 1236] Overall Loss 0.168352 Objective Loss 0.168352 LR 0.000125 Time 0.020696 +2023-10-05 22:11:02,465 - Epoch: [188][ 1200/ 1236] Overall Loss 0.168228 Objective Loss 0.168228 LR 0.000125 Time 0.020693 +2023-10-05 22:11:02,665 - Epoch: [188][ 1210/ 1236] Overall Loss 0.168219 Objective Loss 0.168219 LR 0.000125 Time 0.020688 +2023-10-05 22:11:02,870 - Epoch: [188][ 1220/ 1236] Overall Loss 0.168297 Objective Loss 0.168297 LR 0.000125 Time 0.020685 +2023-10-05 22:11:03,123 - Epoch: [188][ 1230/ 1236] Overall Loss 0.168297 Objective Loss 0.168297 LR 0.000125 Time 0.020723 +2023-10-05 22:11:03,240 - Epoch: [188][ 1236/ 1236] Overall Loss 0.168346 Objective Loss 0.168346 Top1 89.613035 Top5 98.167006 LR 0.000125 Time 0.020717 +2023-10-05 22:11:03,358 - --- validate (epoch=188)----------- +2023-10-05 22:11:03,358 - 29943 samples (256 per mini-batch) +2023-10-05 22:11:03,807 - Epoch: [188][ 10/ 117] Loss 0.277463 Top1 86.679688 Top5 98.554688 +2023-10-05 22:11:03,955 - Epoch: [188][ 20/ 117] Loss 0.287911 Top1 86.074219 Top5 98.496094 +2023-10-05 22:11:04,102 - Epoch: [188][ 30/ 117] Loss 0.299776 Top1 85.950521 Top5 98.372396 +2023-10-05 22:11:04,248 - Epoch: [188][ 40/ 117] Loss 0.299074 Top1 85.937500 Top5 98.291016 +2023-10-05 22:11:04,394 - Epoch: [188][ 50/ 117] Loss 0.310251 Top1 85.851562 Top5 98.273438 +2023-10-05 22:11:04,541 - Epoch: [188][ 60/ 117] Loss 0.309355 Top1 85.748698 Top5 98.294271 +2023-10-05 22:11:04,688 - Epoch: [188][ 70/ 117] Loss 0.309632 Top1 85.786830 Top5 98.314732 +2023-10-05 22:11:04,834 - Epoch: [188][ 80/ 117] Loss 0.306462 Top1 85.834961 Top5 98.330078 +2023-10-05 22:11:04,979 - Epoch: [188][ 90/ 117] Loss 0.305027 Top1 85.885417 Top5 98.302951 +2023-10-05 22:11:05,125 - Epoch: [188][ 100/ 117] Loss 0.302780 Top1 85.980469 Top5 98.316406 +2023-10-05 22:11:05,276 - Epoch: [188][ 110/ 117] Loss 0.302897 Top1 85.962358 Top5 98.302557 +2023-10-05 22:11:05,361 - Epoch: [188][ 117/ 117] Loss 0.304312 Top1 85.986708 Top5 98.266707 +2023-10-05 22:11:05,491 - ==> Top1: 85.987 Top5: 98.267 Loss: 0.304 + +2023-10-05 22:11:05,491 - ==> Confusion: +[[ 937 1 7 2 7 2 0 0 5 61 1 2 1 2 5 0 3 1 0 0 13] + [ 0 1049 1 0 9 22 1 20 1 0 1 1 0 0 2 4 2 0 9 1 8] + [ 2 1 978 9 2 0 20 7 0 2 2 0 7 2 0 2 1 2 7 4 8] + [ 0 1 8 975 1 2 2 2 0 1 4 1 7 1 31 3 0 8 21 3 18] + [ 23 7 1 0 974 4 0 1 0 8 0 2 0 2 10 2 7 1 0 2 6] + [ 3 29 0 1 6 986 0 26 0 1 4 9 0 16 4 0 1 0 5 5 20] + [ 0 5 23 0 0 0 1134 7 0 0 1 2 1 0 1 2 0 2 2 7 4] + [ 3 15 13 0 1 29 4 1081 1 4 4 9 2 2 1 1 0 1 32 8 7] + [ 19 1 0 1 0 1 0 1 973 41 12 1 3 11 12 5 0 0 4 1 3] + [ 97 0 2 1 2 2 0 0 16 957 2 5 0 17 4 3 1 2 0 1 7] + [ 3 4 7 3 0 0 5 3 10 4 975 3 2 8 5 2 2 0 7 1 9] + [ 1 0 1 1 1 12 0 2 0 1 0 966 15 4 0 4 0 13 0 9 5] + [ 1 0 4 2 1 3 0 3 0 0 1 30 990 3 1 3 2 11 2 4 7] + [ 2 0 2 0 2 5 0 0 8 17 8 2 4 1051 3 2 1 0 0 3 9] + [ 12 1 3 4 5 1 0 0 22 4 1 1 3 3 1021 0 0 2 6 0 12] + [ 0 2 1 0 3 0 1 1 0 0 1 7 6 2 1 1074 13 10 0 10 2] + [ 1 11 1 0 6 4 0 0 2 0 0 5 2 1 2 7 1102 0 0 4 13] + [ 0 0 0 0 0 0 5 0 1 0 0 2 14 1 0 3 0 1006 1 1 4] + [ 0 4 6 21 1 0 0 20 1 0 1 0 2 1 9 0 0 0 992 0 10] + [ 0 2 4 2 2 3 7 7 0 0 2 13 2 2 0 6 7 2 3 1080 8] + [ 127 140 131 53 78 129 31 87 81 81 163 94 313 284 147 39 109 60 143 169 5446]] + +2023-10-05 22:11:05,493 - ==> Best [Top1: 86.170 Top5: 98.327 Sparsity:0.00 Params: 148928 on epoch: 187] +2023-10-05 22:11:05,493 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:11:05,499 - + +2023-10-05 22:11:05,499 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:11:06,580 - Epoch: [189][ 10/ 1236] Overall Loss 0.167679 Objective Loss 0.167679 LR 0.000125 Time 0.108059 +2023-10-05 22:11:06,780 - Epoch: [189][ 20/ 1236] Overall Loss 0.172218 Objective Loss 0.172218 LR 0.000125 Time 0.064004 +2023-10-05 22:11:06,978 - Epoch: [189][ 30/ 1236] Overall Loss 0.174487 Objective Loss 0.174487 LR 0.000125 Time 0.049267 +2023-10-05 22:11:07,178 - Epoch: [189][ 40/ 1236] Overall Loss 0.179562 Objective Loss 0.179562 LR 0.000125 Time 0.041930 +2023-10-05 22:11:07,376 - Epoch: [189][ 50/ 1236] Overall Loss 0.179297 Objective Loss 0.179297 LR 0.000125 Time 0.037515 +2023-10-05 22:11:07,576 - Epoch: [189][ 60/ 1236] Overall Loss 0.179093 Objective Loss 0.179093 LR 0.000125 Time 0.034582 +2023-10-05 22:11:07,774 - Epoch: [189][ 70/ 1236] Overall Loss 0.178238 Objective Loss 0.178238 LR 0.000125 Time 0.032472 +2023-10-05 22:11:07,974 - Epoch: [189][ 80/ 1236] Overall Loss 0.176059 Objective Loss 0.176059 LR 0.000125 Time 0.030905 +2023-10-05 22:11:08,172 - Epoch: [189][ 90/ 1236] Overall Loss 0.177322 Objective Loss 0.177322 LR 0.000125 Time 0.029671 +2023-10-05 22:11:08,373 - Epoch: [189][ 100/ 1236] Overall Loss 0.175678 Objective Loss 0.175678 LR 0.000125 Time 0.028702 +2023-10-05 22:11:08,571 - Epoch: [189][ 110/ 1236] Overall Loss 0.174121 Objective Loss 0.174121 LR 0.000125 Time 0.027895 +2023-10-05 22:11:08,771 - Epoch: [189][ 120/ 1236] Overall Loss 0.173542 Objective Loss 0.173542 LR 0.000125 Time 0.027229 +2023-10-05 22:11:08,969 - Epoch: [189][ 130/ 1236] Overall Loss 0.173097 Objective Loss 0.173097 LR 0.000125 Time 0.026660 +2023-10-05 22:11:09,169 - Epoch: [189][ 140/ 1236] Overall Loss 0.172547 Objective Loss 0.172547 LR 0.000125 Time 0.026177 +2023-10-05 22:11:09,367 - Epoch: [189][ 150/ 1236] Overall Loss 0.172247 Objective Loss 0.172247 LR 0.000125 Time 0.025753 +2023-10-05 22:11:09,567 - Epoch: [189][ 160/ 1236] Overall Loss 0.171591 Objective Loss 0.171591 LR 0.000125 Time 0.025388 +2023-10-05 22:11:09,765 - Epoch: [189][ 170/ 1236] Overall Loss 0.172541 Objective Loss 0.172541 LR 0.000125 Time 0.025062 +2023-10-05 22:11:09,965 - Epoch: [189][ 180/ 1236] Overall Loss 0.170946 Objective Loss 0.170946 LR 0.000125 Time 0.024776 +2023-10-05 22:11:10,164 - Epoch: [189][ 190/ 1236] Overall Loss 0.170516 Objective Loss 0.170516 LR 0.000125 Time 0.024515 +2023-10-05 22:11:10,363 - Epoch: [189][ 200/ 1236] Overall Loss 0.170631 Objective Loss 0.170631 LR 0.000125 Time 0.024286 +2023-10-05 22:11:10,562 - Epoch: [189][ 210/ 1236] Overall Loss 0.169472 Objective Loss 0.169472 LR 0.000125 Time 0.024074 +2023-10-05 22:11:10,761 - Epoch: [189][ 220/ 1236] Overall Loss 0.169829 Objective Loss 0.169829 LR 0.000125 Time 0.023884 +2023-10-05 22:11:10,960 - Epoch: [189][ 230/ 1236] Overall Loss 0.169827 Objective Loss 0.169827 LR 0.000125 Time 0.023708 +2023-10-05 22:11:11,160 - Epoch: [189][ 240/ 1236] Overall Loss 0.169981 Objective Loss 0.169981 LR 0.000125 Time 0.023550 +2023-10-05 22:11:11,358 - Epoch: [189][ 250/ 1236] Overall Loss 0.169869 Objective Loss 0.169869 LR 0.000125 Time 0.023401 +2023-10-05 22:11:11,558 - Epoch: [189][ 260/ 1236] Overall Loss 0.169273 Objective Loss 0.169273 LR 0.000125 Time 0.023269 +2023-10-05 22:11:11,757 - Epoch: [189][ 270/ 1236] Overall Loss 0.169352 Objective Loss 0.169352 LR 0.000125 Time 0.023141 +2023-10-05 22:11:11,956 - Epoch: [189][ 280/ 1236] Overall Loss 0.169258 Objective Loss 0.169258 LR 0.000125 Time 0.023027 +2023-10-05 22:11:12,155 - Epoch: [189][ 290/ 1236] Overall Loss 0.168794 Objective Loss 0.168794 LR 0.000125 Time 0.022917 +2023-10-05 22:11:12,355 - Epoch: [189][ 300/ 1236] Overall Loss 0.168487 Objective Loss 0.168487 LR 0.000125 Time 0.022816 +2023-10-05 22:11:12,553 - Epoch: [189][ 310/ 1236] Overall Loss 0.168753 Objective Loss 0.168753 LR 0.000125 Time 0.022720 +2023-10-05 22:11:12,752 - Epoch: [189][ 320/ 1236] Overall Loss 0.169044 Objective Loss 0.169044 LR 0.000125 Time 0.022631 +2023-10-05 22:11:12,951 - Epoch: [189][ 330/ 1236] Overall Loss 0.168893 Objective Loss 0.168893 LR 0.000125 Time 0.022547 +2023-10-05 22:11:13,151 - Epoch: [189][ 340/ 1236] Overall Loss 0.168704 Objective Loss 0.168704 LR 0.000125 Time 0.022471 +2023-10-05 22:11:13,350 - Epoch: [189][ 350/ 1236] Overall Loss 0.168614 Objective Loss 0.168614 LR 0.000125 Time 0.022396 +2023-10-05 22:11:13,551 - Epoch: [189][ 360/ 1236] Overall Loss 0.168761 Objective Loss 0.168761 LR 0.000125 Time 0.022330 +2023-10-05 22:11:13,749 - Epoch: [189][ 370/ 1236] Overall Loss 0.168637 Objective Loss 0.168637 LR 0.000125 Time 0.022262 +2023-10-05 22:11:13,949 - Epoch: [189][ 380/ 1236] Overall Loss 0.168412 Objective Loss 0.168412 LR 0.000125 Time 0.022201 +2023-10-05 22:11:14,148 - Epoch: [189][ 390/ 1236] Overall Loss 0.168588 Objective Loss 0.168588 LR 0.000125 Time 0.022141 +2023-10-05 22:11:14,348 - Epoch: [189][ 400/ 1236] Overall Loss 0.168654 Objective Loss 0.168654 LR 0.000125 Time 0.022086 +2023-10-05 22:11:14,547 - Epoch: [189][ 410/ 1236] Overall Loss 0.168551 Objective Loss 0.168551 LR 0.000125 Time 0.022031 +2023-10-05 22:11:14,747 - Epoch: [189][ 420/ 1236] Overall Loss 0.168246 Objective Loss 0.168246 LR 0.000125 Time 0.021983 +2023-10-05 22:11:14,946 - Epoch: [189][ 430/ 1236] Overall Loss 0.168713 Objective Loss 0.168713 LR 0.000125 Time 0.021933 +2023-10-05 22:11:15,145 - Epoch: [189][ 440/ 1236] Overall Loss 0.168703 Objective Loss 0.168703 LR 0.000125 Time 0.021888 +2023-10-05 22:11:15,344 - Epoch: [189][ 450/ 1236] Overall Loss 0.168856 Objective Loss 0.168856 LR 0.000125 Time 0.021843 +2023-10-05 22:11:15,544 - Epoch: [189][ 460/ 1236] Overall Loss 0.168812 Objective Loss 0.168812 LR 0.000125 Time 0.021801 +2023-10-05 22:11:15,743 - Epoch: [189][ 470/ 1236] Overall Loss 0.168623 Objective Loss 0.168623 LR 0.000125 Time 0.021761 +2023-10-05 22:11:15,944 - Epoch: [189][ 480/ 1236] Overall Loss 0.168569 Objective Loss 0.168569 LR 0.000125 Time 0.021724 +2023-10-05 22:11:16,142 - Epoch: [189][ 490/ 1236] Overall Loss 0.168423 Objective Loss 0.168423 LR 0.000125 Time 0.021684 +2023-10-05 22:11:16,342 - Epoch: [189][ 500/ 1236] Overall Loss 0.168819 Objective Loss 0.168819 LR 0.000125 Time 0.021650 +2023-10-05 22:11:16,541 - Epoch: [189][ 510/ 1236] Overall Loss 0.169166 Objective Loss 0.169166 LR 0.000125 Time 0.021614 +2023-10-05 22:11:16,740 - Epoch: [189][ 520/ 1236] Overall Loss 0.169564 Objective Loss 0.169564 LR 0.000125 Time 0.021581 +2023-10-05 22:11:16,939 - Epoch: [189][ 530/ 1236] Overall Loss 0.169840 Objective Loss 0.169840 LR 0.000125 Time 0.021548 +2023-10-05 22:11:17,138 - Epoch: [189][ 540/ 1236] Overall Loss 0.169685 Objective Loss 0.169685 LR 0.000125 Time 0.021518 +2023-10-05 22:11:17,337 - Epoch: [189][ 550/ 1236] Overall Loss 0.169391 Objective Loss 0.169391 LR 0.000125 Time 0.021487 +2023-10-05 22:11:17,537 - Epoch: [189][ 560/ 1236] Overall Loss 0.169281 Objective Loss 0.169281 LR 0.000125 Time 0.021460 +2023-10-05 22:11:17,735 - Epoch: [189][ 570/ 1236] Overall Loss 0.169151 Objective Loss 0.169151 LR 0.000125 Time 0.021429 +2023-10-05 22:11:17,935 - Epoch: [189][ 580/ 1236] Overall Loss 0.169210 Objective Loss 0.169210 LR 0.000125 Time 0.021403 +2023-10-05 22:11:18,134 - Epoch: [189][ 590/ 1236] Overall Loss 0.169392 Objective Loss 0.169392 LR 0.000125 Time 0.021377 +2023-10-05 22:11:18,333 - Epoch: [189][ 600/ 1236] Overall Loss 0.169091 Objective Loss 0.169091 LR 0.000125 Time 0.021352 +2023-10-05 22:11:18,532 - Epoch: [189][ 610/ 1236] Overall Loss 0.169176 Objective Loss 0.169176 LR 0.000125 Time 0.021327 +2023-10-05 22:11:18,732 - Epoch: [189][ 620/ 1236] Overall Loss 0.169319 Objective Loss 0.169319 LR 0.000125 Time 0.021305 +2023-10-05 22:11:18,931 - Epoch: [189][ 630/ 1236] Overall Loss 0.168981 Objective Loss 0.168981 LR 0.000125 Time 0.021282 +2023-10-05 22:11:19,130 - Epoch: [189][ 640/ 1236] Overall Loss 0.169044 Objective Loss 0.169044 LR 0.000125 Time 0.021261 +2023-10-05 22:11:19,329 - Epoch: [189][ 650/ 1236] Overall Loss 0.168975 Objective Loss 0.168975 LR 0.000125 Time 0.021240 +2023-10-05 22:11:19,529 - Epoch: [189][ 660/ 1236] Overall Loss 0.169016 Objective Loss 0.169016 LR 0.000125 Time 0.021220 +2023-10-05 22:11:19,728 - Epoch: [189][ 670/ 1236] Overall Loss 0.169196 Objective Loss 0.169196 LR 0.000125 Time 0.021200 +2023-10-05 22:11:19,928 - Epoch: [189][ 680/ 1236] Overall Loss 0.169356 Objective Loss 0.169356 LR 0.000125 Time 0.021182 +2023-10-05 22:11:20,127 - Epoch: [189][ 690/ 1236] Overall Loss 0.169288 Objective Loss 0.169288 LR 0.000125 Time 0.021163 +2023-10-05 22:11:20,327 - Epoch: [189][ 700/ 1236] Overall Loss 0.169610 Objective Loss 0.169610 LR 0.000125 Time 0.021145 +2023-10-05 22:11:20,526 - Epoch: [189][ 710/ 1236] Overall Loss 0.169901 Objective Loss 0.169901 LR 0.000125 Time 0.021127 +2023-10-05 22:11:20,726 - Epoch: [189][ 720/ 1236] Overall Loss 0.169842 Objective Loss 0.169842 LR 0.000125 Time 0.021110 +2023-10-05 22:11:20,924 - Epoch: [189][ 730/ 1236] Overall Loss 0.169846 Objective Loss 0.169846 LR 0.000125 Time 0.021093 +2023-10-05 22:11:21,124 - Epoch: [189][ 740/ 1236] Overall Loss 0.169562 Objective Loss 0.169562 LR 0.000125 Time 0.021077 +2023-10-05 22:11:21,323 - Epoch: [189][ 750/ 1236] Overall Loss 0.169416 Objective Loss 0.169416 LR 0.000125 Time 0.021062 +2023-10-05 22:11:21,523 - Epoch: [189][ 760/ 1236] Overall Loss 0.169203 Objective Loss 0.169203 LR 0.000125 Time 0.021047 +2023-10-05 22:11:21,722 - Epoch: [189][ 770/ 1236] Overall Loss 0.169546 Objective Loss 0.169546 LR 0.000125 Time 0.021031 +2023-10-05 22:11:21,922 - Epoch: [189][ 780/ 1236] Overall Loss 0.169621 Objective Loss 0.169621 LR 0.000125 Time 0.021017 +2023-10-05 22:11:22,121 - Epoch: [189][ 790/ 1236] Overall Loss 0.169446 Objective Loss 0.169446 LR 0.000125 Time 0.021003 +2023-10-05 22:11:22,321 - Epoch: [189][ 800/ 1236] Overall Loss 0.169249 Objective Loss 0.169249 LR 0.000125 Time 0.020990 +2023-10-05 22:11:22,520 - Epoch: [189][ 810/ 1236] Overall Loss 0.168938 Objective Loss 0.168938 LR 0.000125 Time 0.020975 +2023-10-05 22:11:22,720 - Epoch: [189][ 820/ 1236] Overall Loss 0.169010 Objective Loss 0.169010 LR 0.000125 Time 0.020963 +2023-10-05 22:11:22,920 - Epoch: [189][ 830/ 1236] Overall Loss 0.169097 Objective Loss 0.169097 LR 0.000125 Time 0.020950 +2023-10-05 22:11:23,120 - Epoch: [189][ 840/ 1236] Overall Loss 0.168868 Objective Loss 0.168868 LR 0.000125 Time 0.020938 +2023-10-05 22:11:23,319 - Epoch: [189][ 850/ 1236] Overall Loss 0.168990 Objective Loss 0.168990 LR 0.000125 Time 0.020926 +2023-10-05 22:11:23,519 - Epoch: [189][ 860/ 1236] Overall Loss 0.169066 Objective Loss 0.169066 LR 0.000125 Time 0.020915 +2023-10-05 22:11:23,719 - Epoch: [189][ 870/ 1236] Overall Loss 0.169019 Objective Loss 0.169019 LR 0.000125 Time 0.020904 +2023-10-05 22:11:23,918 - Epoch: [189][ 880/ 1236] Overall Loss 0.169092 Objective Loss 0.169092 LR 0.000125 Time 0.020893 +2023-10-05 22:11:24,118 - Epoch: [189][ 890/ 1236] Overall Loss 0.168888 Objective Loss 0.168888 LR 0.000125 Time 0.020882 +2023-10-05 22:11:24,317 - Epoch: [189][ 900/ 1236] Overall Loss 0.168638 Objective Loss 0.168638 LR 0.000125 Time 0.020871 +2023-10-05 22:11:24,516 - Epoch: [189][ 910/ 1236] Overall Loss 0.168673 Objective Loss 0.168673 LR 0.000125 Time 0.020860 +2023-10-05 22:11:24,717 - Epoch: [189][ 920/ 1236] Overall Loss 0.168859 Objective Loss 0.168859 LR 0.000125 Time 0.020851 +2023-10-05 22:11:24,916 - Epoch: [189][ 930/ 1236] Overall Loss 0.168887 Objective Loss 0.168887 LR 0.000125 Time 0.020840 +2023-10-05 22:11:25,116 - Epoch: [189][ 940/ 1236] Overall Loss 0.168836 Objective Loss 0.168836 LR 0.000125 Time 0.020831 +2023-10-05 22:11:25,315 - Epoch: [189][ 950/ 1236] Overall Loss 0.168941 Objective Loss 0.168941 LR 0.000125 Time 0.020821 +2023-10-05 22:11:25,514 - Epoch: [189][ 960/ 1236] Overall Loss 0.168895 Objective Loss 0.168895 LR 0.000125 Time 0.020811 +2023-10-05 22:11:25,713 - Epoch: [189][ 970/ 1236] Overall Loss 0.168897 Objective Loss 0.168897 LR 0.000125 Time 0.020801 +2023-10-05 22:11:25,913 - Epoch: [189][ 980/ 1236] Overall Loss 0.168856 Objective Loss 0.168856 LR 0.000125 Time 0.020793 +2023-10-05 22:11:26,112 - Epoch: [189][ 990/ 1236] Overall Loss 0.168945 Objective Loss 0.168945 LR 0.000125 Time 0.020783 +2023-10-05 22:11:26,312 - Epoch: [189][ 1000/ 1236] Overall Loss 0.168837 Objective Loss 0.168837 LR 0.000125 Time 0.020774 +2023-10-05 22:11:26,511 - Epoch: [189][ 1010/ 1236] Overall Loss 0.168814 Objective Loss 0.168814 LR 0.000125 Time 0.020766 +2023-10-05 22:11:26,710 - Epoch: [189][ 1020/ 1236] Overall Loss 0.168897 Objective Loss 0.168897 LR 0.000125 Time 0.020757 +2023-10-05 22:11:26,909 - Epoch: [189][ 1030/ 1236] Overall Loss 0.168954 Objective Loss 0.168954 LR 0.000125 Time 0.020748 +2023-10-05 22:11:27,109 - Epoch: [189][ 1040/ 1236] Overall Loss 0.168973 Objective Loss 0.168973 LR 0.000125 Time 0.020741 +2023-10-05 22:11:27,308 - Epoch: [189][ 1050/ 1236] Overall Loss 0.168725 Objective Loss 0.168725 LR 0.000125 Time 0.020732 +2023-10-05 22:11:27,508 - Epoch: [189][ 1060/ 1236] Overall Loss 0.168731 Objective Loss 0.168731 LR 0.000125 Time 0.020725 +2023-10-05 22:11:27,707 - Epoch: [189][ 1070/ 1236] Overall Loss 0.168768 Objective Loss 0.168768 LR 0.000125 Time 0.020717 +2023-10-05 22:11:27,907 - Epoch: [189][ 1080/ 1236] Overall Loss 0.168768 Objective Loss 0.168768 LR 0.000125 Time 0.020710 +2023-10-05 22:11:28,107 - Epoch: [189][ 1090/ 1236] Overall Loss 0.168977 Objective Loss 0.168977 LR 0.000125 Time 0.020703 +2023-10-05 22:11:28,307 - Epoch: [189][ 1100/ 1236] Overall Loss 0.168955 Objective Loss 0.168955 LR 0.000125 Time 0.020696 +2023-10-05 22:11:28,506 - Epoch: [189][ 1110/ 1236] Overall Loss 0.169044 Objective Loss 0.169044 LR 0.000125 Time 0.020689 +2023-10-05 22:11:28,705 - Epoch: [189][ 1120/ 1236] Overall Loss 0.169099 Objective Loss 0.169099 LR 0.000125 Time 0.020682 +2023-10-05 22:11:28,905 - Epoch: [189][ 1130/ 1236] Overall Loss 0.169049 Objective Loss 0.169049 LR 0.000125 Time 0.020675 +2023-10-05 22:11:29,105 - Epoch: [189][ 1140/ 1236] Overall Loss 0.169061 Objective Loss 0.169061 LR 0.000125 Time 0.020669 +2023-10-05 22:11:29,304 - Epoch: [189][ 1150/ 1236] Overall Loss 0.169225 Objective Loss 0.169225 LR 0.000125 Time 0.020662 +2023-10-05 22:11:29,504 - Epoch: [189][ 1160/ 1236] Overall Loss 0.169363 Objective Loss 0.169363 LR 0.000125 Time 0.020656 +2023-10-05 22:11:29,703 - Epoch: [189][ 1170/ 1236] Overall Loss 0.169386 Objective Loss 0.169386 LR 0.000125 Time 0.020648 +2023-10-05 22:11:29,903 - Epoch: [189][ 1180/ 1236] Overall Loss 0.169417 Objective Loss 0.169417 LR 0.000125 Time 0.020642 +2023-10-05 22:11:30,102 - Epoch: [189][ 1190/ 1236] Overall Loss 0.169268 Objective Loss 0.169268 LR 0.000125 Time 0.020636 +2023-10-05 22:11:30,302 - Epoch: [189][ 1200/ 1236] Overall Loss 0.169266 Objective Loss 0.169266 LR 0.000125 Time 0.020630 +2023-10-05 22:11:30,501 - Epoch: [189][ 1210/ 1236] Overall Loss 0.169163 Objective Loss 0.169163 LR 0.000125 Time 0.020624 +2023-10-05 22:11:30,701 - Epoch: [189][ 1220/ 1236] Overall Loss 0.169154 Objective Loss 0.169154 LR 0.000125 Time 0.020619 +2023-10-05 22:11:30,951 - Epoch: [189][ 1230/ 1236] Overall Loss 0.169156 Objective Loss 0.169156 LR 0.000125 Time 0.020653 +2023-10-05 22:11:31,069 - Epoch: [189][ 1236/ 1236] Overall Loss 0.169167 Objective Loss 0.169167 Top1 89.409369 Top5 99.389002 LR 0.000125 Time 0.020648 +2023-10-05 22:11:31,206 - --- validate (epoch=189)----------- +2023-10-05 22:11:31,207 - 29943 samples (256 per mini-batch) +2023-10-05 22:11:31,659 - Epoch: [189][ 10/ 117] Loss 0.262570 Top1 87.031250 Top5 98.398438 +2023-10-05 22:11:31,807 - Epoch: [189][ 20/ 117] Loss 0.274800 Top1 86.660156 Top5 98.339844 +2023-10-05 22:11:31,955 - Epoch: [189][ 30/ 117] Loss 0.282254 Top1 86.145833 Top5 98.164062 +2023-10-05 22:11:32,103 - Epoch: [189][ 40/ 117] Loss 0.282537 Top1 86.103516 Top5 98.125000 +2023-10-05 22:11:32,251 - Epoch: [189][ 50/ 117] Loss 0.287873 Top1 86.093750 Top5 98.125000 +2023-10-05 22:11:32,399 - Epoch: [189][ 60/ 117] Loss 0.290928 Top1 86.165365 Top5 98.196615 +2023-10-05 22:11:32,545 - Epoch: [189][ 70/ 117] Loss 0.288224 Top1 86.261161 Top5 98.219866 +2023-10-05 22:11:32,692 - Epoch: [189][ 80/ 117] Loss 0.292062 Top1 86.250000 Top5 98.242188 +2023-10-05 22:11:32,838 - Epoch: [189][ 90/ 117] Loss 0.295757 Top1 86.176215 Top5 98.250868 +2023-10-05 22:11:32,988 - Epoch: [189][ 100/ 117] Loss 0.291260 Top1 86.265625 Top5 98.269531 +2023-10-05 22:11:33,141 - Epoch: [189][ 110/ 117] Loss 0.293594 Top1 86.171875 Top5 98.267045 +2023-10-05 22:11:33,225 - Epoch: [189][ 117/ 117] Loss 0.297062 Top1 86.126975 Top5 98.253348 +2023-10-05 22:11:33,364 - ==> Top1: 86.127 Top5: 98.253 Loss: 0.297 + +2023-10-05 22:11:33,365 - ==> Confusion: +[[ 935 3 2 2 3 2 0 0 5 69 1 0 1 2 3 4 1 1 0 0 16] + [ 0 1064 2 0 8 20 1 15 0 0 0 1 0 0 0 3 1 1 8 0 7] + [ 3 2 978 8 3 0 19 9 0 0 3 1 7 1 0 2 0 3 6 3 8] + [ 0 1 13 977 2 3 1 1 1 1 5 0 6 1 25 5 0 6 23 1 17] + [ 19 6 0 0 979 4 1 1 0 10 1 1 1 1 9 1 9 2 1 0 4] + [ 4 30 1 1 2 991 0 26 0 2 4 9 0 11 6 2 3 0 3 2 19] + [ 0 4 18 0 0 0 1132 11 0 0 2 3 1 0 1 4 0 0 2 7 6] + [ 4 15 11 0 2 27 5 1090 1 3 3 9 1 1 0 2 0 1 29 7 7] + [ 15 2 1 0 1 2 1 0 976 42 10 1 1 8 17 4 0 1 4 0 3] + [ 85 0 3 2 5 2 0 0 17 970 1 3 0 17 3 2 0 0 0 1 8] + [ 2 2 9 5 0 1 4 3 11 1 975 3 2 11 3 2 2 0 5 0 12] + [ 1 0 2 0 1 12 0 2 0 1 0 965 14 5 0 5 0 15 0 9 3] + [ 2 0 3 6 0 1 0 3 0 0 1 37 980 3 0 5 2 15 1 3 6] + [ 3 0 1 0 1 6 0 0 10 18 9 1 3 1051 3 2 1 0 0 0 10] + [ 14 1 3 6 7 0 0 0 27 1 1 1 2 3 1007 0 1 1 12 0 14] + [ 1 2 1 0 3 0 1 1 0 0 0 9 5 1 1 1077 10 9 0 10 3] + [ 1 15 1 0 6 3 0 1 2 0 0 4 0 1 3 10 1100 0 0 3 11] + [ 0 0 0 0 0 0 4 0 1 0 0 2 14 2 1 6 0 1003 1 0 4] + [ 1 7 4 20 1 0 0 25 1 0 2 0 1 0 8 0 1 0 988 1 8] + [ 0 3 3 3 1 8 6 11 0 0 1 17 2 1 0 7 7 1 2 1070 9] + [ 120 157 136 63 94 123 41 94 81 77 164 97 297 266 123 53 106 63 128 141 5481]] + +2023-10-05 22:11:33,366 - ==> Best [Top1: 86.170 Top5: 98.327 Sparsity:0.00 Params: 148928 on epoch: 187] +2023-10-05 22:11:33,366 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:11:33,372 - + +2023-10-05 22:11:33,372 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:11:34,346 - Epoch: [190][ 10/ 1236] Overall Loss 0.164301 Objective Loss 0.164301 LR 0.000063 Time 0.097306 +2023-10-05 22:11:34,546 - Epoch: [190][ 20/ 1236] Overall Loss 0.165539 Objective Loss 0.165539 LR 0.000063 Time 0.058619 +2023-10-05 22:11:34,744 - Epoch: [190][ 30/ 1236] Overall Loss 0.165456 Objective Loss 0.165456 LR 0.000063 Time 0.045689 +2023-10-05 22:11:34,944 - Epoch: [190][ 40/ 1236] Overall Loss 0.165907 Objective Loss 0.165907 LR 0.000063 Time 0.039258 +2023-10-05 22:11:35,143 - Epoch: [190][ 50/ 1236] Overall Loss 0.164478 Objective Loss 0.164478 LR 0.000063 Time 0.035377 +2023-10-05 22:11:35,343 - Epoch: [190][ 60/ 1236] Overall Loss 0.166616 Objective Loss 0.166616 LR 0.000063 Time 0.032810 +2023-10-05 22:11:35,542 - Epoch: [190][ 70/ 1236] Overall Loss 0.164214 Objective Loss 0.164214 LR 0.000063 Time 0.030960 +2023-10-05 22:11:35,742 - Epoch: [190][ 80/ 1236] Overall Loss 0.164183 Objective Loss 0.164183 LR 0.000063 Time 0.029581 +2023-10-05 22:11:35,941 - Epoch: [190][ 90/ 1236] Overall Loss 0.163034 Objective Loss 0.163034 LR 0.000063 Time 0.028498 +2023-10-05 22:11:36,141 - Epoch: [190][ 100/ 1236] Overall Loss 0.166328 Objective Loss 0.166328 LR 0.000063 Time 0.027646 +2023-10-05 22:11:36,339 - Epoch: [190][ 110/ 1236] Overall Loss 0.164191 Objective Loss 0.164191 LR 0.000063 Time 0.026925 +2023-10-05 22:11:36,539 - Epoch: [190][ 120/ 1236] Overall Loss 0.164358 Objective Loss 0.164358 LR 0.000063 Time 0.026345 +2023-10-05 22:11:36,738 - Epoch: [190][ 130/ 1236] Overall Loss 0.165043 Objective Loss 0.165043 LR 0.000063 Time 0.025845 +2023-10-05 22:11:36,937 - Epoch: [190][ 140/ 1236] Overall Loss 0.164424 Objective Loss 0.164424 LR 0.000063 Time 0.025415 +2023-10-05 22:11:37,135 - Epoch: [190][ 150/ 1236] Overall Loss 0.164513 Objective Loss 0.164513 LR 0.000063 Time 0.025042 +2023-10-05 22:11:37,335 - Epoch: [190][ 160/ 1236] Overall Loss 0.164310 Objective Loss 0.164310 LR 0.000063 Time 0.024723 +2023-10-05 22:11:37,533 - Epoch: [190][ 170/ 1236] Overall Loss 0.163200 Objective Loss 0.163200 LR 0.000063 Time 0.024433 +2023-10-05 22:11:37,733 - Epoch: [190][ 180/ 1236] Overall Loss 0.163103 Objective Loss 0.163103 LR 0.000063 Time 0.024183 +2023-10-05 22:11:37,931 - Epoch: [190][ 190/ 1236] Overall Loss 0.163356 Objective Loss 0.163356 LR 0.000063 Time 0.023951 +2023-10-05 22:11:38,131 - Epoch: [190][ 200/ 1236] Overall Loss 0.162621 Objective Loss 0.162621 LR 0.000063 Time 0.023751 +2023-10-05 22:11:38,330 - Epoch: [190][ 210/ 1236] Overall Loss 0.162309 Objective Loss 0.162309 LR 0.000063 Time 0.023564 +2023-10-05 22:11:38,530 - Epoch: [190][ 220/ 1236] Overall Loss 0.162077 Objective Loss 0.162077 LR 0.000063 Time 0.023400 +2023-10-05 22:11:38,728 - Epoch: [190][ 230/ 1236] Overall Loss 0.161930 Objective Loss 0.161930 LR 0.000063 Time 0.023244 +2023-10-05 22:11:38,928 - Epoch: [190][ 240/ 1236] Overall Loss 0.161660 Objective Loss 0.161660 LR 0.000063 Time 0.023107 +2023-10-05 22:11:39,127 - Epoch: [190][ 250/ 1236] Overall Loss 0.161870 Objective Loss 0.161870 LR 0.000063 Time 0.022975 +2023-10-05 22:11:39,327 - Epoch: [190][ 260/ 1236] Overall Loss 0.161918 Objective Loss 0.161918 LR 0.000063 Time 0.022859 +2023-10-05 22:11:39,525 - Epoch: [190][ 270/ 1236] Overall Loss 0.162345 Objective Loss 0.162345 LR 0.000063 Time 0.022747 +2023-10-05 22:11:39,726 - Epoch: [190][ 280/ 1236] Overall Loss 0.161995 Objective Loss 0.161995 LR 0.000063 Time 0.022649 +2023-10-05 22:11:39,926 - Epoch: [190][ 290/ 1236] Overall Loss 0.162403 Objective Loss 0.162403 LR 0.000063 Time 0.022559 +2023-10-05 22:11:40,128 - Epoch: [190][ 300/ 1236] Overall Loss 0.162965 Objective Loss 0.162965 LR 0.000063 Time 0.022478 +2023-10-05 22:11:40,330 - Epoch: [190][ 310/ 1236] Overall Loss 0.163495 Objective Loss 0.163495 LR 0.000063 Time 0.022402 +2023-10-05 22:11:40,531 - Epoch: [190][ 320/ 1236] Overall Loss 0.163560 Objective Loss 0.163560 LR 0.000063 Time 0.022331 +2023-10-05 22:11:40,733 - Epoch: [190][ 330/ 1236] Overall Loss 0.163448 Objective Loss 0.163448 LR 0.000063 Time 0.022264 +2023-10-05 22:11:40,933 - Epoch: [190][ 340/ 1236] Overall Loss 0.162971 Objective Loss 0.162971 LR 0.000063 Time 0.022198 +2023-10-05 22:11:41,134 - Epoch: [190][ 350/ 1236] Overall Loss 0.162536 Objective Loss 0.162536 LR 0.000063 Time 0.022135 +2023-10-05 22:11:41,332 - Epoch: [190][ 360/ 1236] Overall Loss 0.162164 Objective Loss 0.162164 LR 0.000063 Time 0.022071 +2023-10-05 22:11:41,535 - Epoch: [190][ 370/ 1236] Overall Loss 0.161984 Objective Loss 0.161984 LR 0.000063 Time 0.022020 +2023-10-05 22:11:41,736 - Epoch: [190][ 380/ 1236] Overall Loss 0.162524 Objective Loss 0.162524 LR 0.000063 Time 0.021970 +2023-10-05 22:11:41,939 - Epoch: [190][ 390/ 1236] Overall Loss 0.162120 Objective Loss 0.162120 LR 0.000063 Time 0.021925 +2023-10-05 22:11:42,140 - Epoch: [190][ 400/ 1236] Overall Loss 0.162014 Objective Loss 0.162014 LR 0.000063 Time 0.021878 +2023-10-05 22:11:42,342 - Epoch: [190][ 410/ 1236] Overall Loss 0.162367 Objective Loss 0.162367 LR 0.000063 Time 0.021838 +2023-10-05 22:11:42,544 - Epoch: [190][ 420/ 1236] Overall Loss 0.162506 Objective Loss 0.162506 LR 0.000063 Time 0.021797 +2023-10-05 22:11:42,749 - Epoch: [190][ 430/ 1236] Overall Loss 0.162783 Objective Loss 0.162783 LR 0.000063 Time 0.021766 +2023-10-05 22:11:42,952 - Epoch: [190][ 440/ 1236] Overall Loss 0.162591 Objective Loss 0.162591 LR 0.000063 Time 0.021731 +2023-10-05 22:11:43,157 - Epoch: [190][ 450/ 1236] Overall Loss 0.162608 Objective Loss 0.162608 LR 0.000063 Time 0.021703 +2023-10-05 22:11:43,359 - Epoch: [190][ 460/ 1236] Overall Loss 0.162613 Objective Loss 0.162613 LR 0.000063 Time 0.021670 +2023-10-05 22:11:43,561 - Epoch: [190][ 470/ 1236] Overall Loss 0.162703 Objective Loss 0.162703 LR 0.000063 Time 0.021638 +2023-10-05 22:11:43,763 - Epoch: [190][ 480/ 1236] Overall Loss 0.162792 Objective Loss 0.162792 LR 0.000063 Time 0.021608 +2023-10-05 22:11:43,966 - Epoch: [190][ 490/ 1236] Overall Loss 0.162842 Objective Loss 0.162842 LR 0.000063 Time 0.021580 +2023-10-05 22:11:44,168 - Epoch: [190][ 500/ 1236] Overall Loss 0.163369 Objective Loss 0.163369 LR 0.000063 Time 0.021552 +2023-10-05 22:11:44,371 - Epoch: [190][ 510/ 1236] Overall Loss 0.163804 Objective Loss 0.163804 LR 0.000063 Time 0.021526 +2023-10-05 22:11:44,573 - Epoch: [190][ 520/ 1236] Overall Loss 0.163491 Objective Loss 0.163491 LR 0.000063 Time 0.021501 +2023-10-05 22:11:44,776 - Epoch: [190][ 530/ 1236] Overall Loss 0.163292 Objective Loss 0.163292 LR 0.000063 Time 0.021476 +2023-10-05 22:11:44,978 - Epoch: [190][ 540/ 1236] Overall Loss 0.163349 Objective Loss 0.163349 LR 0.000063 Time 0.021452 +2023-10-05 22:11:45,180 - Epoch: [190][ 550/ 1236] Overall Loss 0.163389 Objective Loss 0.163389 LR 0.000063 Time 0.021429 +2023-10-05 22:11:45,382 - Epoch: [190][ 560/ 1236] Overall Loss 0.163579 Objective Loss 0.163579 LR 0.000063 Time 0.021407 +2023-10-05 22:11:45,584 - Epoch: [190][ 570/ 1236] Overall Loss 0.163512 Objective Loss 0.163512 LR 0.000063 Time 0.021385 +2023-10-05 22:11:45,786 - Epoch: [190][ 580/ 1236] Overall Loss 0.163579 Objective Loss 0.163579 LR 0.000063 Time 0.021365 +2023-10-05 22:11:45,989 - Epoch: [190][ 590/ 1236] Overall Loss 0.163323 Objective Loss 0.163323 LR 0.000063 Time 0.021345 +2023-10-05 22:11:46,191 - Epoch: [190][ 600/ 1236] Overall Loss 0.163378 Objective Loss 0.163378 LR 0.000063 Time 0.021326 +2023-10-05 22:11:46,394 - Epoch: [190][ 610/ 1236] Overall Loss 0.163553 Objective Loss 0.163553 LR 0.000063 Time 0.021308 +2023-10-05 22:11:46,596 - Epoch: [190][ 620/ 1236] Overall Loss 0.163635 Objective Loss 0.163635 LR 0.000063 Time 0.021290 +2023-10-05 22:11:46,798 - Epoch: [190][ 630/ 1236] Overall Loss 0.163812 Objective Loss 0.163812 LR 0.000063 Time 0.021272 +2023-10-05 22:11:47,000 - Epoch: [190][ 640/ 1236] Overall Loss 0.163856 Objective Loss 0.163856 LR 0.000063 Time 0.021255 +2023-10-05 22:11:47,203 - Epoch: [190][ 650/ 1236] Overall Loss 0.163604 Objective Loss 0.163604 LR 0.000063 Time 0.021239 +2023-10-05 22:11:47,405 - Epoch: [190][ 660/ 1236] Overall Loss 0.163633 Objective Loss 0.163633 LR 0.000063 Time 0.021222 +2023-10-05 22:11:47,607 - Epoch: [190][ 670/ 1236] Overall Loss 0.163800 Objective Loss 0.163800 LR 0.000063 Time 0.021207 +2023-10-05 22:11:47,809 - Epoch: [190][ 680/ 1236] Overall Loss 0.164147 Objective Loss 0.164147 LR 0.000063 Time 0.021193 +2023-10-05 22:11:48,012 - Epoch: [190][ 690/ 1236] Overall Loss 0.164431 Objective Loss 0.164431 LR 0.000063 Time 0.021179 +2023-10-05 22:11:48,214 - Epoch: [190][ 700/ 1236] Overall Loss 0.164462 Objective Loss 0.164462 LR 0.000063 Time 0.021164 +2023-10-05 22:11:48,416 - Epoch: [190][ 710/ 1236] Overall Loss 0.164357 Objective Loss 0.164357 LR 0.000063 Time 0.021150 +2023-10-05 22:11:48,618 - Epoch: [190][ 720/ 1236] Overall Loss 0.164376 Objective Loss 0.164376 LR 0.000063 Time 0.021137 +2023-10-05 22:11:48,820 - Epoch: [190][ 730/ 1236] Overall Loss 0.164318 Objective Loss 0.164318 LR 0.000063 Time 0.021123 +2023-10-05 22:11:49,022 - Epoch: [190][ 740/ 1236] Overall Loss 0.164517 Objective Loss 0.164517 LR 0.000063 Time 0.021111 +2023-10-05 22:11:49,225 - Epoch: [190][ 750/ 1236] Overall Loss 0.164614 Objective Loss 0.164614 LR 0.000063 Time 0.021099 +2023-10-05 22:11:49,427 - Epoch: [190][ 760/ 1236] Overall Loss 0.165001 Objective Loss 0.165001 LR 0.000063 Time 0.021086 +2023-10-05 22:11:49,629 - Epoch: [190][ 770/ 1236] Overall Loss 0.165033 Objective Loss 0.165033 LR 0.000063 Time 0.021075 +2023-10-05 22:11:49,831 - Epoch: [190][ 780/ 1236] Overall Loss 0.165039 Objective Loss 0.165039 LR 0.000063 Time 0.021063 +2023-10-05 22:11:50,034 - Epoch: [190][ 790/ 1236] Overall Loss 0.164890 Objective Loss 0.164890 LR 0.000063 Time 0.021053 +2023-10-05 22:11:50,236 - Epoch: [190][ 800/ 1236] Overall Loss 0.164930 Objective Loss 0.164930 LR 0.000063 Time 0.021042 +2023-10-05 22:11:50,438 - Epoch: [190][ 810/ 1236] Overall Loss 0.164758 Objective Loss 0.164758 LR 0.000063 Time 0.021031 +2023-10-05 22:11:50,640 - Epoch: [190][ 820/ 1236] Overall Loss 0.164611 Objective Loss 0.164611 LR 0.000063 Time 0.021021 +2023-10-05 22:11:50,843 - Epoch: [190][ 830/ 1236] Overall Loss 0.164728 Objective Loss 0.164728 LR 0.000063 Time 0.021011 +2023-10-05 22:11:51,045 - Epoch: [190][ 840/ 1236] Overall Loss 0.164807 Objective Loss 0.164807 LR 0.000063 Time 0.021002 +2023-10-05 22:11:51,248 - Epoch: [190][ 850/ 1236] Overall Loss 0.164813 Objective Loss 0.164813 LR 0.000063 Time 0.020992 +2023-10-05 22:11:51,450 - Epoch: [190][ 860/ 1236] Overall Loss 0.164951 Objective Loss 0.164951 LR 0.000063 Time 0.020983 +2023-10-05 22:11:51,653 - Epoch: [190][ 870/ 1236] Overall Loss 0.165089 Objective Loss 0.165089 LR 0.000063 Time 0.020975 +2023-10-05 22:11:51,855 - Epoch: [190][ 880/ 1236] Overall Loss 0.165001 Objective Loss 0.165001 LR 0.000063 Time 0.020966 +2023-10-05 22:11:52,058 - Epoch: [190][ 890/ 1236] Overall Loss 0.165026 Objective Loss 0.165026 LR 0.000063 Time 0.020958 +2023-10-05 22:11:52,260 - Epoch: [190][ 900/ 1236] Overall Loss 0.165276 Objective Loss 0.165276 LR 0.000063 Time 0.020949 +2023-10-05 22:11:52,463 - Epoch: [190][ 910/ 1236] Overall Loss 0.165259 Objective Loss 0.165259 LR 0.000063 Time 0.020941 +2023-10-05 22:11:52,665 - Epoch: [190][ 920/ 1236] Overall Loss 0.165170 Objective Loss 0.165170 LR 0.000063 Time 0.020933 +2023-10-05 22:11:52,868 - Epoch: [190][ 930/ 1236] Overall Loss 0.165306 Objective Loss 0.165306 LR 0.000063 Time 0.020926 +2023-10-05 22:11:53,070 - Epoch: [190][ 940/ 1236] Overall Loss 0.165115 Objective Loss 0.165115 LR 0.000063 Time 0.020918 +2023-10-05 22:11:53,273 - Epoch: [190][ 950/ 1236] Overall Loss 0.165110 Objective Loss 0.165110 LR 0.000063 Time 0.020911 +2023-10-05 22:11:53,475 - Epoch: [190][ 960/ 1236] Overall Loss 0.165308 Objective Loss 0.165308 LR 0.000063 Time 0.020903 +2023-10-05 22:11:53,678 - Epoch: [190][ 970/ 1236] Overall Loss 0.165567 Objective Loss 0.165567 LR 0.000063 Time 0.020897 +2023-10-05 22:11:53,880 - Epoch: [190][ 980/ 1236] Overall Loss 0.165639 Objective Loss 0.165639 LR 0.000063 Time 0.020889 +2023-10-05 22:11:54,083 - Epoch: [190][ 990/ 1236] Overall Loss 0.165650 Objective Loss 0.165650 LR 0.000063 Time 0.020883 +2023-10-05 22:11:54,285 - Epoch: [190][ 1000/ 1236] Overall Loss 0.165498 Objective Loss 0.165498 LR 0.000063 Time 0.020876 +2023-10-05 22:11:54,488 - Epoch: [190][ 1010/ 1236] Overall Loss 0.165589 Objective Loss 0.165589 LR 0.000063 Time 0.020869 +2023-10-05 22:11:54,690 - Epoch: [190][ 1020/ 1236] Overall Loss 0.165414 Objective Loss 0.165414 LR 0.000063 Time 0.020862 +2023-10-05 22:11:54,892 - Epoch: [190][ 1030/ 1236] Overall Loss 0.165529 Objective Loss 0.165529 LR 0.000063 Time 0.020856 +2023-10-05 22:11:55,094 - Epoch: [190][ 1040/ 1236] Overall Loss 0.165399 Objective Loss 0.165399 LR 0.000063 Time 0.020849 +2023-10-05 22:11:55,297 - Epoch: [190][ 1050/ 1236] Overall Loss 0.165455 Objective Loss 0.165455 LR 0.000063 Time 0.020844 +2023-10-05 22:11:55,499 - Epoch: [190][ 1060/ 1236] Overall Loss 0.165497 Objective Loss 0.165497 LR 0.000063 Time 0.020837 +2023-10-05 22:11:55,702 - Epoch: [190][ 1070/ 1236] Overall Loss 0.165437 Objective Loss 0.165437 LR 0.000063 Time 0.020832 +2023-10-05 22:11:55,904 - Epoch: [190][ 1080/ 1236] Overall Loss 0.165381 Objective Loss 0.165381 LR 0.000063 Time 0.020826 +2023-10-05 22:11:56,106 - Epoch: [190][ 1090/ 1236] Overall Loss 0.165422 Objective Loss 0.165422 LR 0.000063 Time 0.020820 +2023-10-05 22:11:56,308 - Epoch: [190][ 1100/ 1236] Overall Loss 0.165377 Objective Loss 0.165377 LR 0.000063 Time 0.020814 +2023-10-05 22:11:56,511 - Epoch: [190][ 1110/ 1236] Overall Loss 0.165259 Objective Loss 0.165259 LR 0.000063 Time 0.020809 +2023-10-05 22:11:56,713 - Epoch: [190][ 1120/ 1236] Overall Loss 0.165311 Objective Loss 0.165311 LR 0.000063 Time 0.020803 +2023-10-05 22:11:56,916 - Epoch: [190][ 1130/ 1236] Overall Loss 0.165490 Objective Loss 0.165490 LR 0.000063 Time 0.020798 +2023-10-05 22:11:57,118 - Epoch: [190][ 1140/ 1236] Overall Loss 0.165410 Objective Loss 0.165410 LR 0.000063 Time 0.020793 +2023-10-05 22:11:57,321 - Epoch: [190][ 1150/ 1236] Overall Loss 0.165386 Objective Loss 0.165386 LR 0.000063 Time 0.020788 +2023-10-05 22:11:57,523 - Epoch: [190][ 1160/ 1236] Overall Loss 0.165440 Objective Loss 0.165440 LR 0.000063 Time 0.020783 +2023-10-05 22:11:57,726 - Epoch: [190][ 1170/ 1236] Overall Loss 0.165440 Objective Loss 0.165440 LR 0.000063 Time 0.020778 +2023-10-05 22:11:57,928 - Epoch: [190][ 1180/ 1236] Overall Loss 0.165462 Objective Loss 0.165462 LR 0.000063 Time 0.020773 +2023-10-05 22:11:58,131 - Epoch: [190][ 1190/ 1236] Overall Loss 0.165701 Objective Loss 0.165701 LR 0.000063 Time 0.020769 +2023-10-05 22:11:58,333 - Epoch: [190][ 1200/ 1236] Overall Loss 0.165750 Objective Loss 0.165750 LR 0.000063 Time 0.020764 +2023-10-05 22:11:58,536 - Epoch: [190][ 1210/ 1236] Overall Loss 0.165791 Objective Loss 0.165791 LR 0.000063 Time 0.020760 +2023-10-05 22:11:58,738 - Epoch: [190][ 1220/ 1236] Overall Loss 0.165817 Objective Loss 0.165817 LR 0.000063 Time 0.020755 +2023-10-05 22:11:58,995 - Epoch: [190][ 1230/ 1236] Overall Loss 0.165905 Objective Loss 0.165905 LR 0.000063 Time 0.020795 +2023-10-05 22:11:59,113 - Epoch: [190][ 1236/ 1236] Overall Loss 0.165836 Objective Loss 0.165836 Top1 88.187373 Top5 98.167006 LR 0.000063 Time 0.020789 +2023-10-05 22:11:59,233 - --- validate (epoch=190)----------- +2023-10-05 22:11:59,233 - 29943 samples (256 per mini-batch) +2023-10-05 22:11:59,695 - Epoch: [190][ 10/ 117] Loss 0.272666 Top1 86.171875 Top5 98.398438 +2023-10-05 22:11:59,842 - Epoch: [190][ 20/ 117] Loss 0.281806 Top1 86.347656 Top5 98.300781 +2023-10-05 22:11:59,988 - Epoch: [190][ 30/ 117] Loss 0.294893 Top1 85.807292 Top5 98.229167 +2023-10-05 22:12:00,136 - Epoch: [190][ 40/ 117] Loss 0.293898 Top1 85.859375 Top5 98.212891 +2023-10-05 22:12:00,282 - Epoch: [190][ 50/ 117] Loss 0.293908 Top1 86.054688 Top5 98.187500 +2023-10-05 22:12:00,430 - Epoch: [190][ 60/ 117] Loss 0.294744 Top1 86.152344 Top5 98.268229 +2023-10-05 22:12:00,576 - Epoch: [190][ 70/ 117] Loss 0.303152 Top1 85.943080 Top5 98.225446 +2023-10-05 22:12:00,723 - Epoch: [190][ 80/ 117] Loss 0.305143 Top1 85.839844 Top5 98.247070 +2023-10-05 22:12:00,870 - Epoch: [190][ 90/ 117] Loss 0.304554 Top1 85.976562 Top5 98.242188 +2023-10-05 22:12:01,017 - Epoch: [190][ 100/ 117] Loss 0.300224 Top1 86.070312 Top5 98.257812 +2023-10-05 22:12:01,171 - Epoch: [190][ 110/ 117] Loss 0.300428 Top1 86.026278 Top5 98.252841 +2023-10-05 22:12:01,256 - Epoch: [190][ 117/ 117] Loss 0.300357 Top1 86.033464 Top5 98.266707 +2023-10-05 22:12:01,400 - ==> Top1: 86.033 Top5: 98.267 Loss: 0.300 + +2023-10-05 22:12:01,400 - ==> Confusion: +[[ 942 2 4 0 4 3 0 0 5 63 1 0 1 2 4 5 1 0 1 0 12] + [ 0 1058 2 0 10 16 1 19 0 0 2 2 0 0 1 4 3 0 5 1 7] + [ 3 1 978 9 3 0 18 7 0 0 6 2 6 2 0 3 1 3 5 2 7] + [ 2 1 15 965 2 4 1 2 3 2 4 1 7 1 28 4 0 6 22 3 16] + [ 21 6 0 0 977 3 1 1 0 11 1 1 0 1 9 1 8 3 0 1 5] + [ 3 32 1 1 2 989 0 27 0 1 5 11 0 11 6 1 4 0 2 2 18] + [ 0 4 23 0 0 0 1132 6 0 0 1 2 1 0 1 6 1 1 2 5 6] + [ 4 16 12 0 1 29 4 1084 3 2 2 11 1 3 0 1 0 2 29 5 9] + [ 18 1 4 0 0 2 1 0 979 42 11 3 1 6 11 2 1 0 5 0 2] + [ 93 1 3 1 3 2 0 1 15 961 1 2 0 19 5 2 0 1 0 2 7] + [ 2 3 8 4 0 1 3 3 10 2 973 3 0 11 6 1 3 0 7 4 9] + [ 1 0 2 0 2 13 0 2 0 1 0 966 12 4 0 5 2 14 0 6 5] + [ 0 1 1 4 0 2 0 2 0 0 2 39 979 3 0 6 2 14 4 3 6] + [ 2 0 1 0 2 6 0 1 8 15 8 2 3 1058 3 2 1 0 0 1 6] + [ 15 2 3 3 6 0 0 0 27 2 1 2 2 2 1016 0 1 2 6 0 11] + [ 1 2 1 0 3 0 1 1 0 0 0 8 6 1 1 1073 13 11 0 9 3] + [ 1 14 1 0 6 3 0 0 1 0 0 3 0 1 3 12 1101 0 0 2 13] + [ 0 0 0 1 0 0 4 0 0 0 0 1 17 3 1 4 0 1003 1 0 3] + [ 2 6 5 18 1 0 0 24 1 0 2 0 0 0 11 0 0 0 985 2 11] + [ 0 3 5 3 1 7 5 10 0 0 1 14 3 1 0 6 10 2 3 1071 7] + [ 132 150 132 46 84 125 31 83 93 76 155 113 293 268 136 54 123 60 133 147 5471]] + +2023-10-05 22:12:01,402 - ==> Best [Top1: 86.170 Top5: 98.327 Sparsity:0.00 Params: 148928 on epoch: 187] +2023-10-05 22:12:01,402 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:12:01,408 - + +2023-10-05 22:12:01,408 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:12:02,399 - Epoch: [191][ 10/ 1236] Overall Loss 0.184258 Objective Loss 0.184258 LR 0.000063 Time 0.099071 +2023-10-05 22:12:02,602 - Epoch: [191][ 20/ 1236] Overall Loss 0.169442 Objective Loss 0.169442 LR 0.000063 Time 0.059688 +2023-10-05 22:12:02,803 - Epoch: [191][ 30/ 1236] Overall Loss 0.159661 Objective Loss 0.159661 LR 0.000063 Time 0.046477 +2023-10-05 22:12:03,006 - Epoch: [191][ 40/ 1236] Overall Loss 0.164649 Objective Loss 0.164649 LR 0.000063 Time 0.039918 +2023-10-05 22:12:03,208 - Epoch: [191][ 50/ 1236] Overall Loss 0.161358 Objective Loss 0.161358 LR 0.000063 Time 0.035960 +2023-10-05 22:12:03,410 - Epoch: [191][ 60/ 1236] Overall Loss 0.162298 Objective Loss 0.162298 LR 0.000063 Time 0.033332 +2023-10-05 22:12:03,612 - Epoch: [191][ 70/ 1236] Overall Loss 0.160069 Objective Loss 0.160069 LR 0.000063 Time 0.031453 +2023-10-05 22:12:03,813 - Epoch: [191][ 80/ 1236] Overall Loss 0.159361 Objective Loss 0.159361 LR 0.000063 Time 0.030029 +2023-10-05 22:12:04,014 - Epoch: [191][ 90/ 1236] Overall Loss 0.159408 Objective Loss 0.159408 LR 0.000063 Time 0.028922 +2023-10-05 22:12:04,216 - Epoch: [191][ 100/ 1236] Overall Loss 0.159596 Objective Loss 0.159596 LR 0.000063 Time 0.028042 +2023-10-05 22:12:04,417 - Epoch: [191][ 110/ 1236] Overall Loss 0.160334 Objective Loss 0.160334 LR 0.000063 Time 0.027318 +2023-10-05 22:12:04,620 - Epoch: [191][ 120/ 1236] Overall Loss 0.158387 Objective Loss 0.158387 LR 0.000063 Time 0.026730 +2023-10-05 22:12:04,824 - Epoch: [191][ 130/ 1236] Overall Loss 0.158214 Objective Loss 0.158214 LR 0.000063 Time 0.026245 +2023-10-05 22:12:05,030 - Epoch: [191][ 140/ 1236] Overall Loss 0.159043 Objective Loss 0.159043 LR 0.000063 Time 0.025835 +2023-10-05 22:12:05,234 - Epoch: [191][ 150/ 1236] Overall Loss 0.160024 Objective Loss 0.160024 LR 0.000063 Time 0.025473 +2023-10-05 22:12:05,439 - Epoch: [191][ 160/ 1236] Overall Loss 0.160596 Objective Loss 0.160596 LR 0.000063 Time 0.025159 +2023-10-05 22:12:05,643 - Epoch: [191][ 170/ 1236] Overall Loss 0.160405 Objective Loss 0.160405 LR 0.000063 Time 0.024878 +2023-10-05 22:12:05,849 - Epoch: [191][ 180/ 1236] Overall Loss 0.160322 Objective Loss 0.160322 LR 0.000063 Time 0.024637 +2023-10-05 22:12:06,053 - Epoch: [191][ 190/ 1236] Overall Loss 0.160746 Objective Loss 0.160746 LR 0.000063 Time 0.024408 +2023-10-05 22:12:06,259 - Epoch: [191][ 200/ 1236] Overall Loss 0.160228 Objective Loss 0.160228 LR 0.000063 Time 0.024213 +2023-10-05 22:12:06,463 - Epoch: [191][ 210/ 1236] Overall Loss 0.159992 Objective Loss 0.159992 LR 0.000063 Time 0.024026 +2023-10-05 22:12:06,669 - Epoch: [191][ 220/ 1236] Overall Loss 0.159427 Objective Loss 0.159427 LR 0.000063 Time 0.023867 +2023-10-05 22:12:06,873 - Epoch: [191][ 230/ 1236] Overall Loss 0.159238 Objective Loss 0.159238 LR 0.000063 Time 0.023711 +2023-10-05 22:12:07,078 - Epoch: [191][ 240/ 1236] Overall Loss 0.159836 Objective Loss 0.159836 LR 0.000063 Time 0.023576 +2023-10-05 22:12:07,283 - Epoch: [191][ 250/ 1236] Overall Loss 0.160648 Objective Loss 0.160648 LR 0.000063 Time 0.023452 +2023-10-05 22:12:07,489 - Epoch: [191][ 260/ 1236] Overall Loss 0.161494 Objective Loss 0.161494 LR 0.000063 Time 0.023339 +2023-10-05 22:12:07,693 - Epoch: [191][ 270/ 1236] Overall Loss 0.161137 Objective Loss 0.161137 LR 0.000063 Time 0.023229 +2023-10-05 22:12:07,898 - Epoch: [191][ 280/ 1236] Overall Loss 0.161496 Objective Loss 0.161496 LR 0.000063 Time 0.023130 +2023-10-05 22:12:08,102 - Epoch: [191][ 290/ 1236] Overall Loss 0.161488 Objective Loss 0.161488 LR 0.000063 Time 0.023035 +2023-10-05 22:12:08,307 - Epoch: [191][ 300/ 1236] Overall Loss 0.161855 Objective Loss 0.161855 LR 0.000063 Time 0.022949 +2023-10-05 22:12:08,512 - Epoch: [191][ 310/ 1236] Overall Loss 0.161991 Objective Loss 0.161991 LR 0.000063 Time 0.022868 +2023-10-05 22:12:08,717 - Epoch: [191][ 320/ 1236] Overall Loss 0.161652 Objective Loss 0.161652 LR 0.000063 Time 0.022794 +2023-10-05 22:12:08,921 - Epoch: [191][ 330/ 1236] Overall Loss 0.161801 Objective Loss 0.161801 LR 0.000063 Time 0.022721 +2023-10-05 22:12:09,127 - Epoch: [191][ 340/ 1236] Overall Loss 0.162194 Objective Loss 0.162194 LR 0.000063 Time 0.022656 +2023-10-05 22:12:09,331 - Epoch: [191][ 350/ 1236] Overall Loss 0.162024 Objective Loss 0.162024 LR 0.000063 Time 0.022589 +2023-10-05 22:12:09,537 - Epoch: [191][ 360/ 1236] Overall Loss 0.162133 Objective Loss 0.162133 LR 0.000063 Time 0.022531 +2023-10-05 22:12:09,741 - Epoch: [191][ 370/ 1236] Overall Loss 0.162325 Objective Loss 0.162325 LR 0.000063 Time 0.022474 +2023-10-05 22:12:09,946 - Epoch: [191][ 380/ 1236] Overall Loss 0.162532 Objective Loss 0.162532 LR 0.000063 Time 0.022421 +2023-10-05 22:12:10,149 - Epoch: [191][ 390/ 1236] Overall Loss 0.162364 Objective Loss 0.162364 LR 0.000063 Time 0.022364 +2023-10-05 22:12:10,351 - Epoch: [191][ 400/ 1236] Overall Loss 0.163065 Objective Loss 0.163065 LR 0.000063 Time 0.022311 +2023-10-05 22:12:10,553 - Epoch: [191][ 410/ 1236] Overall Loss 0.162793 Objective Loss 0.162793 LR 0.000063 Time 0.022259 +2023-10-05 22:12:10,756 - Epoch: [191][ 420/ 1236] Overall Loss 0.162119 Objective Loss 0.162119 LR 0.000063 Time 0.022211 +2023-10-05 22:12:10,959 - Epoch: [191][ 430/ 1236] Overall Loss 0.162425 Objective Loss 0.162425 LR 0.000063 Time 0.022164 +2023-10-05 22:12:11,161 - Epoch: [191][ 440/ 1236] Overall Loss 0.162947 Objective Loss 0.162947 LR 0.000063 Time 0.022120 +2023-10-05 22:12:11,363 - Epoch: [191][ 450/ 1236] Overall Loss 0.163183 Objective Loss 0.163183 LR 0.000063 Time 0.022077 +2023-10-05 22:12:11,566 - Epoch: [191][ 460/ 1236] Overall Loss 0.163288 Objective Loss 0.163288 LR 0.000063 Time 0.022038 +2023-10-05 22:12:11,769 - Epoch: [191][ 470/ 1236] Overall Loss 0.162886 Objective Loss 0.162886 LR 0.000063 Time 0.021998 +2023-10-05 22:12:11,971 - Epoch: [191][ 480/ 1236] Overall Loss 0.162471 Objective Loss 0.162471 LR 0.000063 Time 0.021962 +2023-10-05 22:12:12,174 - Epoch: [191][ 490/ 1236] Overall Loss 0.162682 Objective Loss 0.162682 LR 0.000063 Time 0.021925 +2023-10-05 22:12:12,376 - Epoch: [191][ 500/ 1236] Overall Loss 0.162354 Objective Loss 0.162354 LR 0.000063 Time 0.021892 +2023-10-05 22:12:12,579 - Epoch: [191][ 510/ 1236] Overall Loss 0.162264 Objective Loss 0.162264 LR 0.000063 Time 0.021858 +2023-10-05 22:12:12,782 - Epoch: [191][ 520/ 1236] Overall Loss 0.162212 Objective Loss 0.162212 LR 0.000063 Time 0.021827 +2023-10-05 22:12:12,984 - Epoch: [191][ 530/ 1236] Overall Loss 0.162688 Objective Loss 0.162688 LR 0.000063 Time 0.021797 +2023-10-05 22:12:13,187 - Epoch: [191][ 540/ 1236] Overall Loss 0.162547 Objective Loss 0.162547 LR 0.000063 Time 0.021768 +2023-10-05 22:12:13,390 - Epoch: [191][ 550/ 1236] Overall Loss 0.162665 Objective Loss 0.162665 LR 0.000063 Time 0.021740 +2023-10-05 22:12:13,599 - Epoch: [191][ 560/ 1236] Overall Loss 0.162493 Objective Loss 0.162493 LR 0.000063 Time 0.021726 +2023-10-05 22:12:13,808 - Epoch: [191][ 570/ 1236] Overall Loss 0.162795 Objective Loss 0.162795 LR 0.000063 Time 0.021711 +2023-10-05 22:12:14,010 - Epoch: [191][ 580/ 1236] Overall Loss 0.162728 Objective Loss 0.162728 LR 0.000063 Time 0.021683 +2023-10-05 22:12:14,212 - Epoch: [191][ 590/ 1236] Overall Loss 0.162635 Objective Loss 0.162635 LR 0.000063 Time 0.021658 +2023-10-05 22:12:14,414 - Epoch: [191][ 600/ 1236] Overall Loss 0.162562 Objective Loss 0.162562 LR 0.000063 Time 0.021634 +2023-10-05 22:12:14,617 - Epoch: [191][ 610/ 1236] Overall Loss 0.162585 Objective Loss 0.162585 LR 0.000063 Time 0.021610 +2023-10-05 22:12:14,819 - Epoch: [191][ 620/ 1236] Overall Loss 0.162794 Objective Loss 0.162794 LR 0.000063 Time 0.021588 +2023-10-05 22:12:15,022 - Epoch: [191][ 630/ 1236] Overall Loss 0.162641 Objective Loss 0.162641 LR 0.000063 Time 0.021566 +2023-10-05 22:12:15,224 - Epoch: [191][ 640/ 1236] Overall Loss 0.162605 Objective Loss 0.162605 LR 0.000063 Time 0.021545 +2023-10-05 22:12:15,427 - Epoch: [191][ 650/ 1236] Overall Loss 0.162402 Objective Loss 0.162402 LR 0.000063 Time 0.021524 +2023-10-05 22:12:15,630 - Epoch: [191][ 660/ 1236] Overall Loss 0.162935 Objective Loss 0.162935 LR 0.000063 Time 0.021505 +2023-10-05 22:12:15,832 - Epoch: [191][ 670/ 1236] Overall Loss 0.162900 Objective Loss 0.162900 LR 0.000063 Time 0.021485 +2023-10-05 22:12:16,034 - Epoch: [191][ 680/ 1236] Overall Loss 0.163085 Objective Loss 0.163085 LR 0.000063 Time 0.021467 +2023-10-05 22:12:16,237 - Epoch: [191][ 690/ 1236] Overall Loss 0.162748 Objective Loss 0.162748 LR 0.000063 Time 0.021448 +2023-10-05 22:12:16,439 - Epoch: [191][ 700/ 1236] Overall Loss 0.162694 Objective Loss 0.162694 LR 0.000063 Time 0.021431 +2023-10-05 22:12:16,642 - Epoch: [191][ 710/ 1236] Overall Loss 0.162674 Objective Loss 0.162674 LR 0.000063 Time 0.021414 +2023-10-05 22:12:16,844 - Epoch: [191][ 720/ 1236] Overall Loss 0.162648 Objective Loss 0.162648 LR 0.000063 Time 0.021397 +2023-10-05 22:12:17,047 - Epoch: [191][ 730/ 1236] Overall Loss 0.162900 Objective Loss 0.162900 LR 0.000063 Time 0.021381 +2023-10-05 22:12:17,249 - Epoch: [191][ 740/ 1236] Overall Loss 0.162943 Objective Loss 0.162943 LR 0.000063 Time 0.021366 +2023-10-05 22:12:17,452 - Epoch: [191][ 750/ 1236] Overall Loss 0.162877 Objective Loss 0.162877 LR 0.000063 Time 0.021350 +2023-10-05 22:12:17,654 - Epoch: [191][ 760/ 1236] Overall Loss 0.162857 Objective Loss 0.162857 LR 0.000063 Time 0.021335 +2023-10-05 22:12:17,857 - Epoch: [191][ 770/ 1236] Overall Loss 0.163114 Objective Loss 0.163114 LR 0.000063 Time 0.021320 +2023-10-05 22:12:18,059 - Epoch: [191][ 780/ 1236] Overall Loss 0.163193 Objective Loss 0.163193 LR 0.000063 Time 0.021307 +2023-10-05 22:12:18,262 - Epoch: [191][ 790/ 1236] Overall Loss 0.163405 Objective Loss 0.163405 LR 0.000063 Time 0.021292 +2023-10-05 22:12:18,464 - Epoch: [191][ 800/ 1236] Overall Loss 0.163620 Objective Loss 0.163620 LR 0.000063 Time 0.021279 +2023-10-05 22:12:18,667 - Epoch: [191][ 810/ 1236] Overall Loss 0.163583 Objective Loss 0.163583 LR 0.000063 Time 0.021266 +2023-10-05 22:12:18,870 - Epoch: [191][ 820/ 1236] Overall Loss 0.163452 Objective Loss 0.163452 LR 0.000063 Time 0.021253 +2023-10-05 22:12:19,072 - Epoch: [191][ 830/ 1236] Overall Loss 0.163780 Objective Loss 0.163780 LR 0.000063 Time 0.021240 +2023-10-05 22:12:19,274 - Epoch: [191][ 840/ 1236] Overall Loss 0.163904 Objective Loss 0.163904 LR 0.000063 Time 0.021228 +2023-10-05 22:12:19,477 - Epoch: [191][ 850/ 1236] Overall Loss 0.163927 Objective Loss 0.163927 LR 0.000063 Time 0.021216 +2023-10-05 22:12:19,679 - Epoch: [191][ 860/ 1236] Overall Loss 0.163965 Objective Loss 0.163965 LR 0.000063 Time 0.021205 +2023-10-05 22:12:19,882 - Epoch: [191][ 870/ 1236] Overall Loss 0.163922 Objective Loss 0.163922 LR 0.000063 Time 0.021193 +2023-10-05 22:12:20,085 - Epoch: [191][ 880/ 1236] Overall Loss 0.163764 Objective Loss 0.163764 LR 0.000063 Time 0.021183 +2023-10-05 22:12:20,287 - Epoch: [191][ 890/ 1236] Overall Loss 0.163560 Objective Loss 0.163560 LR 0.000063 Time 0.021171 +2023-10-05 22:12:20,489 - Epoch: [191][ 900/ 1236] Overall Loss 0.163536 Objective Loss 0.163536 LR 0.000063 Time 0.021161 +2023-10-05 22:12:20,692 - Epoch: [191][ 910/ 1236] Overall Loss 0.163434 Objective Loss 0.163434 LR 0.000063 Time 0.021150 +2023-10-05 22:12:20,895 - Epoch: [191][ 920/ 1236] Overall Loss 0.163265 Objective Loss 0.163265 LR 0.000063 Time 0.021140 +2023-10-05 22:12:21,097 - Epoch: [191][ 930/ 1236] Overall Loss 0.163320 Objective Loss 0.163320 LR 0.000063 Time 0.021130 +2023-10-05 22:12:21,300 - Epoch: [191][ 940/ 1236] Overall Loss 0.163349 Objective Loss 0.163349 LR 0.000063 Time 0.021121 +2023-10-05 22:12:21,502 - Epoch: [191][ 950/ 1236] Overall Loss 0.163289 Objective Loss 0.163289 LR 0.000063 Time 0.021111 +2023-10-05 22:12:21,704 - Epoch: [191][ 960/ 1236] Overall Loss 0.163338 Objective Loss 0.163338 LR 0.000063 Time 0.021102 +2023-10-05 22:12:21,907 - Epoch: [191][ 970/ 1236] Overall Loss 0.163411 Objective Loss 0.163411 LR 0.000063 Time 0.021092 +2023-10-05 22:12:22,110 - Epoch: [191][ 980/ 1236] Overall Loss 0.163482 Objective Loss 0.163482 LR 0.000063 Time 0.021084 +2023-10-05 22:12:22,312 - Epoch: [191][ 990/ 1236] Overall Loss 0.163301 Objective Loss 0.163301 LR 0.000063 Time 0.021075 +2023-10-05 22:12:22,514 - Epoch: [191][ 1000/ 1236] Overall Loss 0.163291 Objective Loss 0.163291 LR 0.000063 Time 0.021066 +2023-10-05 22:12:22,717 - Epoch: [191][ 1010/ 1236] Overall Loss 0.163396 Objective Loss 0.163396 LR 0.000063 Time 0.021058 +2023-10-05 22:12:22,920 - Epoch: [191][ 1020/ 1236] Overall Loss 0.163393 Objective Loss 0.163393 LR 0.000063 Time 0.021050 +2023-10-05 22:12:23,122 - Epoch: [191][ 1030/ 1236] Overall Loss 0.163591 Objective Loss 0.163591 LR 0.000063 Time 0.021041 +2023-10-05 22:12:23,325 - Epoch: [191][ 1040/ 1236] Overall Loss 0.163528 Objective Loss 0.163528 LR 0.000063 Time 0.021034 +2023-10-05 22:12:23,527 - Epoch: [191][ 1050/ 1236] Overall Loss 0.163534 Objective Loss 0.163534 LR 0.000063 Time 0.021026 +2023-10-05 22:12:23,730 - Epoch: [191][ 1060/ 1236] Overall Loss 0.163537 Objective Loss 0.163537 LR 0.000063 Time 0.021018 +2023-10-05 22:12:23,932 - Epoch: [191][ 1070/ 1236] Overall Loss 0.163537 Objective Loss 0.163537 LR 0.000063 Time 0.021011 +2023-10-05 22:12:24,135 - Epoch: [191][ 1080/ 1236] Overall Loss 0.163560 Objective Loss 0.163560 LR 0.000063 Time 0.021004 +2023-10-05 22:12:24,337 - Epoch: [191][ 1090/ 1236] Overall Loss 0.163609 Objective Loss 0.163609 LR 0.000063 Time 0.020996 +2023-10-05 22:12:24,540 - Epoch: [191][ 1100/ 1236] Overall Loss 0.163590 Objective Loss 0.163590 LR 0.000063 Time 0.020989 +2023-10-05 22:12:24,742 - Epoch: [191][ 1110/ 1236] Overall Loss 0.163657 Objective Loss 0.163657 LR 0.000063 Time 0.020982 +2023-10-05 22:12:24,945 - Epoch: [191][ 1120/ 1236] Overall Loss 0.163605 Objective Loss 0.163605 LR 0.000063 Time 0.020975 +2023-10-05 22:12:25,147 - Epoch: [191][ 1130/ 1236] Overall Loss 0.163588 Objective Loss 0.163588 LR 0.000063 Time 0.020968 +2023-10-05 22:12:25,350 - Epoch: [191][ 1140/ 1236] Overall Loss 0.163633 Objective Loss 0.163633 LR 0.000063 Time 0.020962 +2023-10-05 22:12:25,552 - Epoch: [191][ 1150/ 1236] Overall Loss 0.163587 Objective Loss 0.163587 LR 0.000063 Time 0.020955 +2023-10-05 22:12:25,755 - Epoch: [191][ 1160/ 1236] Overall Loss 0.163502 Objective Loss 0.163502 LR 0.000063 Time 0.020949 +2023-10-05 22:12:25,957 - Epoch: [191][ 1170/ 1236] Overall Loss 0.163522 Objective Loss 0.163522 LR 0.000063 Time 0.020943 +2023-10-05 22:12:26,160 - Epoch: [191][ 1180/ 1236] Overall Loss 0.163603 Objective Loss 0.163603 LR 0.000063 Time 0.020937 +2023-10-05 22:12:26,362 - Epoch: [191][ 1190/ 1236] Overall Loss 0.163587 Objective Loss 0.163587 LR 0.000063 Time 0.020931 +2023-10-05 22:12:26,565 - Epoch: [191][ 1200/ 1236] Overall Loss 0.163599 Objective Loss 0.163599 LR 0.000063 Time 0.020925 +2023-10-05 22:12:26,767 - Epoch: [191][ 1210/ 1236] Overall Loss 0.163717 Objective Loss 0.163717 LR 0.000063 Time 0.020919 +2023-10-05 22:12:26,970 - Epoch: [191][ 1220/ 1236] Overall Loss 0.163741 Objective Loss 0.163741 LR 0.000063 Time 0.020913 +2023-10-05 22:12:27,225 - Epoch: [191][ 1230/ 1236] Overall Loss 0.163814 Objective Loss 0.163814 LR 0.000063 Time 0.020951 +2023-10-05 22:12:27,343 - Epoch: [191][ 1236/ 1236] Overall Loss 0.163806 Objective Loss 0.163806 Top1 90.631365 Top5 98.574338 LR 0.000063 Time 0.020944 +2023-10-05 22:12:27,472 - --- validate (epoch=191)----------- +2023-10-05 22:12:27,472 - 29943 samples (256 per mini-batch) +2023-10-05 22:12:27,925 - Epoch: [191][ 10/ 117] Loss 0.290803 Top1 86.210938 Top5 97.929688 +2023-10-05 22:12:28,073 - Epoch: [191][ 20/ 117] Loss 0.291023 Top1 86.367188 Top5 98.222656 +2023-10-05 22:12:28,220 - Epoch: [191][ 30/ 117] Loss 0.298940 Top1 86.406250 Top5 98.281250 +2023-10-05 22:12:28,367 - Epoch: [191][ 40/ 117] Loss 0.304938 Top1 86.074219 Top5 98.164062 +2023-10-05 22:12:28,515 - Epoch: [191][ 50/ 117] Loss 0.298789 Top1 86.304688 Top5 98.195312 +2023-10-05 22:12:28,663 - Epoch: [191][ 60/ 117] Loss 0.295823 Top1 86.412760 Top5 98.242188 +2023-10-05 22:12:28,810 - Epoch: [191][ 70/ 117] Loss 0.301154 Top1 86.188616 Top5 98.197545 +2023-10-05 22:12:28,955 - Epoch: [191][ 80/ 117] Loss 0.303513 Top1 86.035156 Top5 98.193359 +2023-10-05 22:12:29,100 - Epoch: [191][ 90/ 117] Loss 0.304257 Top1 86.098090 Top5 98.164062 +2023-10-05 22:12:29,246 - Epoch: [191][ 100/ 117] Loss 0.303659 Top1 86.097656 Top5 98.207031 +2023-10-05 22:12:29,398 - Epoch: [191][ 110/ 117] Loss 0.301614 Top1 86.079545 Top5 98.242188 +2023-10-05 22:12:29,483 - Epoch: [191][ 117/ 117] Loss 0.301930 Top1 86.020105 Top5 98.250008 +2023-10-05 22:12:29,620 - ==> Top1: 86.020 Top5: 98.250 Loss: 0.302 + +2023-10-05 22:12:29,621 - ==> Confusion: +[[ 938 2 4 0 6 4 0 0 4 66 1 0 1 2 6 2 2 0 0 0 12] + [ 0 1062 1 0 10 16 1 17 1 0 1 1 0 0 2 4 2 1 5 2 5] + [ 5 1 978 11 2 0 19 7 0 0 4 1 7 1 0 4 1 1 6 3 5] + [ 2 1 16 965 1 2 1 2 1 2 7 0 7 2 26 2 0 5 29 3 15] + [ 19 8 0 0 973 2 1 1 1 10 0 1 0 3 10 1 10 3 0 1 6] + [ 3 34 1 0 6 981 0 28 1 1 4 8 1 13 6 2 4 0 3 4 16] + [ 0 4 23 0 0 0 1129 5 0 0 2 3 1 0 1 7 0 0 2 6 8] + [ 4 18 13 0 1 26 2 1077 1 4 3 9 3 2 0 2 0 0 42 3 8] + [ 17 2 3 0 0 0 1 0 978 43 9 1 1 9 12 2 3 0 4 1 3] + [ 94 1 2 2 5 2 0 1 14 962 1 2 0 16 5 2 1 1 0 2 6] + [ 3 7 7 1 0 1 1 4 11 1 981 3 0 8 5 2 2 0 4 2 10] + [ 1 2 2 0 2 11 0 2 0 1 0 963 14 5 0 4 1 17 0 8 2] + [ 1 4 0 3 0 2 0 2 0 0 2 28 996 1 0 4 2 10 2 4 7] + [ 2 0 1 0 0 7 0 0 12 16 8 1 4 1054 3 2 1 0 0 1 7] + [ 14 2 4 2 5 0 0 0 23 3 1 1 3 2 1017 0 0 2 9 0 13] + [ 1 2 1 0 2 0 1 1 0 0 0 8 7 1 1 1073 15 9 0 9 3] + [ 1 12 1 0 5 4 0 0 1 0 0 2 0 1 2 11 1108 0 0 3 10] + [ 0 0 1 1 0 0 3 0 1 0 0 3 17 2 1 7 0 998 1 0 3] + [ 1 4 3 16 1 0 0 24 1 0 3 0 2 1 9 0 1 0 993 1 8] + [ 0 3 3 3 1 6 5 11 0 0 1 13 3 0 0 7 8 1 3 1074 10] + [ 127 155 130 50 87 107 30 76 99 76 172 93 309 275 141 52 125 63 141 140 5457]] + +2023-10-05 22:12:29,623 - ==> Best [Top1: 86.170 Top5: 98.327 Sparsity:0.00 Params: 148928 on epoch: 187] +2023-10-05 22:12:29,623 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:12:29,629 - + +2023-10-05 22:12:29,629 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:12:30,748 - Epoch: [192][ 10/ 1236] Overall Loss 0.165238 Objective Loss 0.165238 LR 0.000063 Time 0.111872 +2023-10-05 22:12:30,952 - Epoch: [192][ 20/ 1236] Overall Loss 0.159209 Objective Loss 0.159209 LR 0.000063 Time 0.066108 +2023-10-05 22:12:31,154 - Epoch: [192][ 30/ 1236] Overall Loss 0.157303 Objective Loss 0.157303 LR 0.000063 Time 0.050801 +2023-10-05 22:12:31,357 - Epoch: [192][ 40/ 1236] Overall Loss 0.158568 Objective Loss 0.158568 LR 0.000063 Time 0.043175 +2023-10-05 22:12:31,560 - Epoch: [192][ 50/ 1236] Overall Loss 0.156142 Objective Loss 0.156142 LR 0.000063 Time 0.038592 +2023-10-05 22:12:31,763 - Epoch: [192][ 60/ 1236] Overall Loss 0.158736 Objective Loss 0.158736 LR 0.000063 Time 0.035530 +2023-10-05 22:12:31,966 - Epoch: [192][ 70/ 1236] Overall Loss 0.158162 Objective Loss 0.158162 LR 0.000063 Time 0.033349 +2023-10-05 22:12:32,176 - Epoch: [192][ 80/ 1236] Overall Loss 0.158729 Objective Loss 0.158729 LR 0.000063 Time 0.031810 +2023-10-05 22:12:32,387 - Epoch: [192][ 90/ 1236] Overall Loss 0.161312 Objective Loss 0.161312 LR 0.000063 Time 0.030612 +2023-10-05 22:12:32,601 - Epoch: [192][ 100/ 1236] Overall Loss 0.161545 Objective Loss 0.161545 LR 0.000063 Time 0.029688 +2023-10-05 22:12:32,812 - Epoch: [192][ 110/ 1236] Overall Loss 0.162477 Objective Loss 0.162477 LR 0.000063 Time 0.028902 +2023-10-05 22:12:33,023 - Epoch: [192][ 120/ 1236] Overall Loss 0.161586 Objective Loss 0.161586 LR 0.000063 Time 0.028254 +2023-10-05 22:12:33,232 - Epoch: [192][ 130/ 1236] Overall Loss 0.161712 Objective Loss 0.161712 LR 0.000063 Time 0.027679 +2023-10-05 22:12:33,445 - Epoch: [192][ 140/ 1236] Overall Loss 0.162321 Objective Loss 0.162321 LR 0.000063 Time 0.027225 +2023-10-05 22:12:33,656 - Epoch: [192][ 150/ 1236] Overall Loss 0.163194 Objective Loss 0.163194 LR 0.000063 Time 0.026810 +2023-10-05 22:12:33,863 - Epoch: [192][ 160/ 1236] Overall Loss 0.161815 Objective Loss 0.161815 LR 0.000063 Time 0.026429 +2023-10-05 22:12:34,067 - Epoch: [192][ 170/ 1236] Overall Loss 0.161754 Objective Loss 0.161754 LR 0.000063 Time 0.026069 +2023-10-05 22:12:34,270 - Epoch: [192][ 180/ 1236] Overall Loss 0.160814 Objective Loss 0.160814 LR 0.000063 Time 0.025751 +2023-10-05 22:12:34,471 - Epoch: [192][ 190/ 1236] Overall Loss 0.161279 Objective Loss 0.161279 LR 0.000063 Time 0.025450 +2023-10-05 22:12:34,673 - Epoch: [192][ 200/ 1236] Overall Loss 0.160510 Objective Loss 0.160510 LR 0.000063 Time 0.025186 +2023-10-05 22:12:34,874 - Epoch: [192][ 210/ 1236] Overall Loss 0.160785 Objective Loss 0.160785 LR 0.000063 Time 0.024941 +2023-10-05 22:12:35,074 - Epoch: [192][ 220/ 1236] Overall Loss 0.160513 Objective Loss 0.160513 LR 0.000063 Time 0.024717 +2023-10-05 22:12:35,273 - Epoch: [192][ 230/ 1236] Overall Loss 0.159682 Objective Loss 0.159682 LR 0.000063 Time 0.024506 +2023-10-05 22:12:35,472 - Epoch: [192][ 240/ 1236] Overall Loss 0.159759 Objective Loss 0.159759 LR 0.000063 Time 0.024314 +2023-10-05 22:12:35,674 - Epoch: [192][ 250/ 1236] Overall Loss 0.160263 Objective Loss 0.160263 LR 0.000063 Time 0.024145 +2023-10-05 22:12:35,875 - Epoch: [192][ 260/ 1236] Overall Loss 0.159550 Objective Loss 0.159550 LR 0.000063 Time 0.023990 +2023-10-05 22:12:36,075 - Epoch: [192][ 270/ 1236] Overall Loss 0.158878 Objective Loss 0.158878 LR 0.000063 Time 0.023841 +2023-10-05 22:12:36,276 - Epoch: [192][ 280/ 1236] Overall Loss 0.159372 Objective Loss 0.159372 LR 0.000063 Time 0.023706 +2023-10-05 22:12:36,476 - Epoch: [192][ 290/ 1236] Overall Loss 0.159102 Objective Loss 0.159102 LR 0.000063 Time 0.023575 +2023-10-05 22:12:36,676 - Epoch: [192][ 300/ 1236] Overall Loss 0.159065 Objective Loss 0.159065 LR 0.000063 Time 0.023456 +2023-10-05 22:12:36,876 - Epoch: [192][ 310/ 1236] Overall Loss 0.158922 Objective Loss 0.158922 LR 0.000063 Time 0.023343 +2023-10-05 22:12:37,077 - Epoch: [192][ 320/ 1236] Overall Loss 0.158846 Objective Loss 0.158846 LR 0.000063 Time 0.023241 +2023-10-05 22:12:37,284 - Epoch: [192][ 330/ 1236] Overall Loss 0.158686 Objective Loss 0.158686 LR 0.000063 Time 0.023163 +2023-10-05 22:12:37,490 - Epoch: [192][ 340/ 1236] Overall Loss 0.159039 Objective Loss 0.159039 LR 0.000063 Time 0.023086 +2023-10-05 22:12:37,697 - Epoch: [192][ 350/ 1236] Overall Loss 0.159402 Objective Loss 0.159402 LR 0.000063 Time 0.023015 +2023-10-05 22:12:37,902 - Epoch: [192][ 360/ 1236] Overall Loss 0.159766 Objective Loss 0.159766 LR 0.000063 Time 0.022945 +2023-10-05 22:12:38,104 - Epoch: [192][ 370/ 1236] Overall Loss 0.160017 Objective Loss 0.160017 LR 0.000063 Time 0.022869 +2023-10-05 22:12:38,306 - Epoch: [192][ 380/ 1236] Overall Loss 0.160213 Objective Loss 0.160213 LR 0.000063 Time 0.022799 +2023-10-05 22:12:38,508 - Epoch: [192][ 390/ 1236] Overall Loss 0.159852 Objective Loss 0.159852 LR 0.000063 Time 0.022730 +2023-10-05 22:12:38,710 - Epoch: [192][ 400/ 1236] Overall Loss 0.160087 Objective Loss 0.160087 LR 0.000063 Time 0.022666 +2023-10-05 22:12:38,911 - Epoch: [192][ 410/ 1236] Overall Loss 0.160116 Objective Loss 0.160116 LR 0.000063 Time 0.022603 +2023-10-05 22:12:39,114 - Epoch: [192][ 420/ 1236] Overall Loss 0.159960 Objective Loss 0.159960 LR 0.000063 Time 0.022546 +2023-10-05 22:12:39,315 - Epoch: [192][ 430/ 1236] Overall Loss 0.160416 Objective Loss 0.160416 LR 0.000063 Time 0.022488 +2023-10-05 22:12:39,517 - Epoch: [192][ 440/ 1236] Overall Loss 0.160704 Objective Loss 0.160704 LR 0.000063 Time 0.022436 +2023-10-05 22:12:39,718 - Epoch: [192][ 450/ 1236] Overall Loss 0.160534 Objective Loss 0.160534 LR 0.000063 Time 0.022382 +2023-10-05 22:12:39,920 - Epoch: [192][ 460/ 1236] Overall Loss 0.160614 Objective Loss 0.160614 LR 0.000063 Time 0.022335 +2023-10-05 22:12:40,120 - Epoch: [192][ 470/ 1236] Overall Loss 0.160903 Objective Loss 0.160903 LR 0.000063 Time 0.022285 +2023-10-05 22:12:40,323 - Epoch: [192][ 480/ 1236] Overall Loss 0.161119 Objective Loss 0.161119 LR 0.000063 Time 0.022242 +2023-10-05 22:12:40,523 - Epoch: [192][ 490/ 1236] Overall Loss 0.161320 Objective Loss 0.161320 LR 0.000063 Time 0.022196 +2023-10-05 22:12:40,725 - Epoch: [192][ 500/ 1236] Overall Loss 0.161695 Objective Loss 0.161695 LR 0.000063 Time 0.022155 +2023-10-05 22:12:40,926 - Epoch: [192][ 510/ 1236] Overall Loss 0.162079 Objective Loss 0.162079 LR 0.000063 Time 0.022113 +2023-10-05 22:12:41,128 - Epoch: [192][ 520/ 1236] Overall Loss 0.161918 Objective Loss 0.161918 LR 0.000063 Time 0.022076 +2023-10-05 22:12:41,329 - Epoch: [192][ 530/ 1236] Overall Loss 0.161837 Objective Loss 0.161837 LR 0.000063 Time 0.022038 +2023-10-05 22:12:41,531 - Epoch: [192][ 540/ 1236] Overall Loss 0.162092 Objective Loss 0.162092 LR 0.000063 Time 0.022003 +2023-10-05 22:12:41,731 - Epoch: [192][ 550/ 1236] Overall Loss 0.161903 Objective Loss 0.161903 LR 0.000063 Time 0.021967 +2023-10-05 22:12:41,934 - Epoch: [192][ 560/ 1236] Overall Loss 0.162204 Objective Loss 0.162204 LR 0.000063 Time 0.021935 +2023-10-05 22:12:42,134 - Epoch: [192][ 570/ 1236] Overall Loss 0.162516 Objective Loss 0.162516 LR 0.000063 Time 0.021902 +2023-10-05 22:12:42,337 - Epoch: [192][ 580/ 1236] Overall Loss 0.162561 Objective Loss 0.162561 LR 0.000063 Time 0.021873 +2023-10-05 22:12:42,537 - Epoch: [192][ 590/ 1236] Overall Loss 0.162496 Objective Loss 0.162496 LR 0.000063 Time 0.021841 +2023-10-05 22:12:42,740 - Epoch: [192][ 600/ 1236] Overall Loss 0.162270 Objective Loss 0.162270 LR 0.000063 Time 0.021813 +2023-10-05 22:12:42,940 - Epoch: [192][ 610/ 1236] Overall Loss 0.162191 Objective Loss 0.162191 LR 0.000063 Time 0.021784 +2023-10-05 22:12:43,143 - Epoch: [192][ 620/ 1236] Overall Loss 0.162507 Objective Loss 0.162507 LR 0.000063 Time 0.021758 +2023-10-05 22:12:43,343 - Epoch: [192][ 630/ 1236] Overall Loss 0.162421 Objective Loss 0.162421 LR 0.000063 Time 0.021731 +2023-10-05 22:12:43,545 - Epoch: [192][ 640/ 1236] Overall Loss 0.162532 Objective Loss 0.162532 LR 0.000063 Time 0.021707 +2023-10-05 22:12:43,746 - Epoch: [192][ 650/ 1236] Overall Loss 0.162563 Objective Loss 0.162563 LR 0.000063 Time 0.021681 +2023-10-05 22:12:43,948 - Epoch: [192][ 660/ 1236] Overall Loss 0.162536 Objective Loss 0.162536 LR 0.000063 Time 0.021658 +2023-10-05 22:12:44,149 - Epoch: [192][ 670/ 1236] Overall Loss 0.162304 Objective Loss 0.162304 LR 0.000063 Time 0.021634 +2023-10-05 22:12:44,352 - Epoch: [192][ 680/ 1236] Overall Loss 0.162539 Objective Loss 0.162539 LR 0.000063 Time 0.021613 +2023-10-05 22:12:44,553 - Epoch: [192][ 690/ 1236] Overall Loss 0.162748 Objective Loss 0.162748 LR 0.000063 Time 0.021591 +2023-10-05 22:12:44,755 - Epoch: [192][ 700/ 1236] Overall Loss 0.162666 Objective Loss 0.162666 LR 0.000063 Time 0.021571 +2023-10-05 22:12:44,956 - Epoch: [192][ 710/ 1236] Overall Loss 0.162788 Objective Loss 0.162788 LR 0.000063 Time 0.021550 +2023-10-05 22:12:45,159 - Epoch: [192][ 720/ 1236] Overall Loss 0.162683 Objective Loss 0.162683 LR 0.000063 Time 0.021531 +2023-10-05 22:12:45,360 - Epoch: [192][ 730/ 1236] Overall Loss 0.162494 Objective Loss 0.162494 LR 0.000063 Time 0.021511 +2023-10-05 22:12:45,562 - Epoch: [192][ 740/ 1236] Overall Loss 0.162324 Objective Loss 0.162324 LR 0.000063 Time 0.021493 +2023-10-05 22:12:45,763 - Epoch: [192][ 750/ 1236] Overall Loss 0.162626 Objective Loss 0.162626 LR 0.000063 Time 0.021474 +2023-10-05 22:12:45,965 - Epoch: [192][ 760/ 1236] Overall Loss 0.162639 Objective Loss 0.162639 LR 0.000063 Time 0.021457 +2023-10-05 22:12:46,166 - Epoch: [192][ 770/ 1236] Overall Loss 0.162634 Objective Loss 0.162634 LR 0.000063 Time 0.021439 +2023-10-05 22:12:46,368 - Epoch: [192][ 780/ 1236] Overall Loss 0.162640 Objective Loss 0.162640 LR 0.000063 Time 0.021423 +2023-10-05 22:12:46,569 - Epoch: [192][ 790/ 1236] Overall Loss 0.162826 Objective Loss 0.162826 LR 0.000063 Time 0.021405 +2023-10-05 22:12:46,772 - Epoch: [192][ 800/ 1236] Overall Loss 0.162783 Objective Loss 0.162783 LR 0.000063 Time 0.021390 +2023-10-05 22:12:46,973 - Epoch: [192][ 810/ 1236] Overall Loss 0.162572 Objective Loss 0.162572 LR 0.000063 Time 0.021374 +2023-10-05 22:12:47,175 - Epoch: [192][ 820/ 1236] Overall Loss 0.162462 Objective Loss 0.162462 LR 0.000063 Time 0.021360 +2023-10-05 22:12:47,376 - Epoch: [192][ 830/ 1236] Overall Loss 0.162464 Objective Loss 0.162464 LR 0.000063 Time 0.021344 +2023-10-05 22:12:47,579 - Epoch: [192][ 840/ 1236] Overall Loss 0.162578 Objective Loss 0.162578 LR 0.000063 Time 0.021331 +2023-10-05 22:12:47,780 - Epoch: [192][ 850/ 1236] Overall Loss 0.162604 Objective Loss 0.162604 LR 0.000063 Time 0.021316 +2023-10-05 22:12:47,983 - Epoch: [192][ 860/ 1236] Overall Loss 0.162830 Objective Loss 0.162830 LR 0.000063 Time 0.021304 +2023-10-05 22:12:48,184 - Epoch: [192][ 870/ 1236] Overall Loss 0.162868 Objective Loss 0.162868 LR 0.000063 Time 0.021289 +2023-10-05 22:12:48,386 - Epoch: [192][ 880/ 1236] Overall Loss 0.162783 Objective Loss 0.162783 LR 0.000063 Time 0.021277 +2023-10-05 22:12:48,587 - Epoch: [192][ 890/ 1236] Overall Loss 0.162622 Objective Loss 0.162622 LR 0.000063 Time 0.021263 +2023-10-05 22:12:48,790 - Epoch: [192][ 900/ 1236] Overall Loss 0.162548 Objective Loss 0.162548 LR 0.000063 Time 0.021252 +2023-10-05 22:12:48,991 - Epoch: [192][ 910/ 1236] Overall Loss 0.162726 Objective Loss 0.162726 LR 0.000063 Time 0.021238 +2023-10-05 22:12:49,193 - Epoch: [192][ 920/ 1236] Overall Loss 0.162616 Objective Loss 0.162616 LR 0.000063 Time 0.021227 +2023-10-05 22:12:49,394 - Epoch: [192][ 930/ 1236] Overall Loss 0.162987 Objective Loss 0.162987 LR 0.000063 Time 0.021215 +2023-10-05 22:12:49,597 - Epoch: [192][ 940/ 1236] Overall Loss 0.163031 Objective Loss 0.163031 LR 0.000063 Time 0.021204 +2023-10-05 22:12:49,798 - Epoch: [192][ 950/ 1236] Overall Loss 0.163120 Objective Loss 0.163120 LR 0.000063 Time 0.021192 +2023-10-05 22:12:50,000 - Epoch: [192][ 960/ 1236] Overall Loss 0.163115 Objective Loss 0.163115 LR 0.000063 Time 0.021182 +2023-10-05 22:12:50,201 - Epoch: [192][ 970/ 1236] Overall Loss 0.163007 Objective Loss 0.163007 LR 0.000063 Time 0.021170 +2023-10-05 22:12:50,404 - Epoch: [192][ 980/ 1236] Overall Loss 0.163057 Objective Loss 0.163057 LR 0.000063 Time 0.021161 +2023-10-05 22:12:50,605 - Epoch: [192][ 990/ 1236] Overall Loss 0.163013 Objective Loss 0.163013 LR 0.000063 Time 0.021150 +2023-10-05 22:12:50,808 - Epoch: [192][ 1000/ 1236] Overall Loss 0.162788 Objective Loss 0.162788 LR 0.000063 Time 0.021140 +2023-10-05 22:12:51,008 - Epoch: [192][ 1010/ 1236] Overall Loss 0.162717 Objective Loss 0.162717 LR 0.000063 Time 0.021129 +2023-10-05 22:12:51,211 - Epoch: [192][ 1020/ 1236] Overall Loss 0.162678 Objective Loss 0.162678 LR 0.000063 Time 0.021120 +2023-10-05 22:12:51,412 - Epoch: [192][ 1030/ 1236] Overall Loss 0.162476 Objective Loss 0.162476 LR 0.000063 Time 0.021110 +2023-10-05 22:12:51,615 - Epoch: [192][ 1040/ 1236] Overall Loss 0.162473 Objective Loss 0.162473 LR 0.000063 Time 0.021102 +2023-10-05 22:12:51,816 - Epoch: [192][ 1050/ 1236] Overall Loss 0.162503 Objective Loss 0.162503 LR 0.000063 Time 0.021092 +2023-10-05 22:12:52,018 - Epoch: [192][ 1060/ 1236] Overall Loss 0.162419 Objective Loss 0.162419 LR 0.000063 Time 0.021083 +2023-10-05 22:12:52,219 - Epoch: [192][ 1070/ 1236] Overall Loss 0.162588 Objective Loss 0.162588 LR 0.000063 Time 0.021074 +2023-10-05 22:12:52,422 - Epoch: [192][ 1080/ 1236] Overall Loss 0.162676 Objective Loss 0.162676 LR 0.000063 Time 0.021066 +2023-10-05 22:12:52,623 - Epoch: [192][ 1090/ 1236] Overall Loss 0.162533 Objective Loss 0.162533 LR 0.000063 Time 0.021056 +2023-10-05 22:12:52,825 - Epoch: [192][ 1100/ 1236] Overall Loss 0.162481 Objective Loss 0.162481 LR 0.000063 Time 0.021049 +2023-10-05 22:12:53,027 - Epoch: [192][ 1110/ 1236] Overall Loss 0.162400 Objective Loss 0.162400 LR 0.000063 Time 0.021041 +2023-10-05 22:12:53,229 - Epoch: [192][ 1120/ 1236] Overall Loss 0.162508 Objective Loss 0.162508 LR 0.000063 Time 0.021033 +2023-10-05 22:12:53,431 - Epoch: [192][ 1130/ 1236] Overall Loss 0.162593 Objective Loss 0.162593 LR 0.000063 Time 0.021025 +2023-10-05 22:12:53,633 - Epoch: [192][ 1140/ 1236] Overall Loss 0.162716 Objective Loss 0.162716 LR 0.000063 Time 0.021018 +2023-10-05 22:12:53,834 - Epoch: [192][ 1150/ 1236] Overall Loss 0.162843 Objective Loss 0.162843 LR 0.000063 Time 0.021009 +2023-10-05 22:12:54,037 - Epoch: [192][ 1160/ 1236] Overall Loss 0.162997 Objective Loss 0.162997 LR 0.000063 Time 0.021002 +2023-10-05 22:12:54,238 - Epoch: [192][ 1170/ 1236] Overall Loss 0.163066 Objective Loss 0.163066 LR 0.000063 Time 0.020995 +2023-10-05 22:12:54,440 - Epoch: [192][ 1180/ 1236] Overall Loss 0.163184 Objective Loss 0.163184 LR 0.000063 Time 0.020988 +2023-10-05 22:12:54,641 - Epoch: [192][ 1190/ 1236] Overall Loss 0.163169 Objective Loss 0.163169 LR 0.000063 Time 0.020980 +2023-10-05 22:12:54,844 - Epoch: [192][ 1200/ 1236] Overall Loss 0.163319 Objective Loss 0.163319 LR 0.000063 Time 0.020974 +2023-10-05 22:12:55,045 - Epoch: [192][ 1210/ 1236] Overall Loss 0.163180 Objective Loss 0.163180 LR 0.000063 Time 0.020966 +2023-10-05 22:12:55,248 - Epoch: [192][ 1220/ 1236] Overall Loss 0.163207 Objective Loss 0.163207 LR 0.000063 Time 0.020960 +2023-10-05 22:12:55,502 - Epoch: [192][ 1230/ 1236] Overall Loss 0.163187 Objective Loss 0.163187 LR 0.000063 Time 0.020996 +2023-10-05 22:12:55,620 - Epoch: [192][ 1236/ 1236] Overall Loss 0.163214 Objective Loss 0.163214 Top1 90.020367 Top5 98.574338 LR 0.000063 Time 0.020990 +2023-10-05 22:12:55,776 - --- validate (epoch=192)----------- +2023-10-05 22:12:55,776 - 29943 samples (256 per mini-batch) +2023-10-05 22:12:56,233 - Epoch: [192][ 10/ 117] Loss 0.286451 Top1 87.031250 Top5 98.515625 +2023-10-05 22:12:56,379 - Epoch: [192][ 20/ 117] Loss 0.290961 Top1 86.386719 Top5 98.359375 +2023-10-05 22:12:56,526 - Epoch: [192][ 30/ 117] Loss 0.293152 Top1 86.640625 Top5 98.203125 +2023-10-05 22:12:56,673 - Epoch: [192][ 40/ 117] Loss 0.299192 Top1 86.640625 Top5 98.261719 +2023-10-05 22:12:56,820 - Epoch: [192][ 50/ 117] Loss 0.299552 Top1 86.367188 Top5 98.242188 +2023-10-05 22:12:56,967 - Epoch: [192][ 60/ 117] Loss 0.294434 Top1 86.412760 Top5 98.268229 +2023-10-05 22:12:57,114 - Epoch: [192][ 70/ 117] Loss 0.293320 Top1 86.383929 Top5 98.303571 +2023-10-05 22:12:57,261 - Epoch: [192][ 80/ 117] Loss 0.298550 Top1 86.318359 Top5 98.266602 +2023-10-05 22:12:57,408 - Epoch: [192][ 90/ 117] Loss 0.299398 Top1 86.263021 Top5 98.220486 +2023-10-05 22:12:57,555 - Epoch: [192][ 100/ 117] Loss 0.297116 Top1 86.296875 Top5 98.257812 +2023-10-05 22:12:57,708 - Epoch: [192][ 110/ 117] Loss 0.300256 Top1 86.289062 Top5 98.252841 +2023-10-05 22:12:57,793 - Epoch: [192][ 117/ 117] Loss 0.301331 Top1 86.217146 Top5 98.250008 +2023-10-05 22:12:57,918 - ==> Top1: 86.217 Top5: 98.250 Loss: 0.301 + +2023-10-05 22:12:57,919 - ==> Confusion: +[[ 951 0 4 1 6 1 0 0 3 55 1 0 1 2 6 2 2 1 0 0 14] + [ 0 1063 1 0 9 12 1 20 0 0 1 2 0 0 2 3 2 1 6 1 7] + [ 3 3 981 9 1 0 20 4 0 1 5 1 8 2 0 3 0 1 7 4 3] + [ 2 1 14 975 2 1 0 1 0 2 4 1 7 2 25 3 0 5 25 2 17] + [ 25 6 0 0 968 2 0 2 0 8 1 1 0 1 10 2 13 3 1 1 6] + [ 3 31 1 0 4 977 1 29 0 0 4 12 0 12 7 2 1 0 4 6 22] + [ 0 4 19 0 0 0 1131 6 0 0 1 4 2 0 1 8 0 1 1 6 7] + [ 4 17 11 0 0 26 5 1084 1 2 2 8 1 3 0 1 1 0 39 5 8] + [ 18 1 3 0 0 2 0 2 981 37 9 2 1 8 13 2 2 1 4 1 2] + [ 101 1 3 0 3 3 0 1 19 950 1 1 2 17 6 2 1 0 0 0 8] + [ 2 2 10 3 0 0 5 4 12 2 971 3 1 10 4 1 3 0 7 2 11] + [ 1 0 2 0 3 10 0 2 0 0 0 967 13 7 0 5 2 14 0 6 3] + [ 1 2 0 3 0 3 0 2 0 0 2 36 986 3 2 4 3 10 1 4 6] + [ 2 0 1 0 0 3 0 0 11 14 6 4 4 1057 3 1 1 0 0 1 11] + [ 15 3 3 7 6 0 0 0 22 2 1 1 3 2 1011 0 1 2 9 0 13] + [ 1 2 1 0 2 0 1 0 0 0 0 8 8 1 1 1071 16 9 0 9 4] + [ 1 12 1 0 4 5 0 0 1 0 0 4 0 1 2 7 1107 0 0 2 14] + [ 0 0 1 2 1 0 3 0 1 0 0 3 20 2 0 6 0 995 1 0 3] + [ 0 8 6 15 1 0 0 25 1 0 2 0 2 0 10 0 1 0 989 1 7] + [ 0 4 4 3 3 7 6 10 0 0 1 15 1 0 0 6 7 1 3 1074 7] + [ 131 147 139 56 79 110 34 81 86 66 154 102 275 268 122 40 120 65 141 162 5527]] + +2023-10-05 22:12:57,920 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:12:57,920 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:12:57,933 - + +2023-10-05 22:12:57,933 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:12:58,921 - Epoch: [193][ 10/ 1236] Overall Loss 0.180009 Objective Loss 0.180009 LR 0.000063 Time 0.098720 +2023-10-05 22:12:59,121 - Epoch: [193][ 20/ 1236] Overall Loss 0.165127 Objective Loss 0.165127 LR 0.000063 Time 0.059344 +2023-10-05 22:12:59,319 - Epoch: [193][ 30/ 1236] Overall Loss 0.164277 Objective Loss 0.164277 LR 0.000063 Time 0.046151 +2023-10-05 22:12:59,519 - Epoch: [193][ 40/ 1236] Overall Loss 0.162543 Objective Loss 0.162543 LR 0.000063 Time 0.039608 +2023-10-05 22:12:59,717 - Epoch: [193][ 50/ 1236] Overall Loss 0.161671 Objective Loss 0.161671 LR 0.000063 Time 0.035635 +2023-10-05 22:12:59,917 - Epoch: [193][ 60/ 1236] Overall Loss 0.163075 Objective Loss 0.163075 LR 0.000063 Time 0.033025 +2023-10-05 22:13:00,115 - Epoch: [193][ 70/ 1236] Overall Loss 0.161841 Objective Loss 0.161841 LR 0.000063 Time 0.031135 +2023-10-05 22:13:00,316 - Epoch: [193][ 80/ 1236] Overall Loss 0.160983 Objective Loss 0.160983 LR 0.000063 Time 0.029739 +2023-10-05 22:13:00,514 - Epoch: [193][ 90/ 1236] Overall Loss 0.162121 Objective Loss 0.162121 LR 0.000063 Time 0.028633 +2023-10-05 22:13:00,714 - Epoch: [193][ 100/ 1236] Overall Loss 0.162932 Objective Loss 0.162932 LR 0.000063 Time 0.027767 +2023-10-05 22:13:00,910 - Epoch: [193][ 110/ 1236] Overall Loss 0.163137 Objective Loss 0.163137 LR 0.000063 Time 0.027026 +2023-10-05 22:13:01,110 - Epoch: [193][ 120/ 1236] Overall Loss 0.163225 Objective Loss 0.163225 LR 0.000063 Time 0.026437 +2023-10-05 22:13:01,308 - Epoch: [193][ 130/ 1236] Overall Loss 0.163843 Objective Loss 0.163843 LR 0.000063 Time 0.025925 +2023-10-05 22:13:01,508 - Epoch: [193][ 140/ 1236] Overall Loss 0.163907 Objective Loss 0.163907 LR 0.000063 Time 0.025498 +2023-10-05 22:13:01,706 - Epoch: [193][ 150/ 1236] Overall Loss 0.164155 Objective Loss 0.164155 LR 0.000063 Time 0.025113 +2023-10-05 22:13:01,906 - Epoch: [193][ 160/ 1236] Overall Loss 0.164317 Objective Loss 0.164317 LR 0.000063 Time 0.024792 +2023-10-05 22:13:02,107 - Epoch: [193][ 170/ 1236] Overall Loss 0.162721 Objective Loss 0.162721 LR 0.000063 Time 0.024511 +2023-10-05 22:13:02,307 - Epoch: [193][ 180/ 1236] Overall Loss 0.162680 Objective Loss 0.162680 LR 0.000063 Time 0.024259 +2023-10-05 22:13:02,508 - Epoch: [193][ 190/ 1236] Overall Loss 0.162396 Objective Loss 0.162396 LR 0.000063 Time 0.024039 +2023-10-05 22:13:02,708 - Epoch: [193][ 200/ 1236] Overall Loss 0.163431 Objective Loss 0.163431 LR 0.000063 Time 0.023836 +2023-10-05 22:13:02,909 - Epoch: [193][ 210/ 1236] Overall Loss 0.162775 Objective Loss 0.162775 LR 0.000063 Time 0.023656 +2023-10-05 22:13:03,109 - Epoch: [193][ 220/ 1236] Overall Loss 0.163493 Objective Loss 0.163493 LR 0.000063 Time 0.023489 +2023-10-05 22:13:03,310 - Epoch: [193][ 230/ 1236] Overall Loss 0.163784 Objective Loss 0.163784 LR 0.000063 Time 0.023340 +2023-10-05 22:13:03,510 - Epoch: [193][ 240/ 1236] Overall Loss 0.163828 Objective Loss 0.163828 LR 0.000063 Time 0.023201 +2023-10-05 22:13:03,711 - Epoch: [193][ 250/ 1236] Overall Loss 0.163811 Objective Loss 0.163811 LR 0.000063 Time 0.023075 +2023-10-05 22:13:03,912 - Epoch: [193][ 260/ 1236] Overall Loss 0.163627 Objective Loss 0.163627 LR 0.000063 Time 0.022957 +2023-10-05 22:13:04,113 - Epoch: [193][ 270/ 1236] Overall Loss 0.164125 Objective Loss 0.164125 LR 0.000063 Time 0.022851 +2023-10-05 22:13:04,313 - Epoch: [193][ 280/ 1236] Overall Loss 0.163804 Objective Loss 0.163804 LR 0.000063 Time 0.022749 +2023-10-05 22:13:04,514 - Epoch: [193][ 290/ 1236] Overall Loss 0.164241 Objective Loss 0.164241 LR 0.000063 Time 0.022656 +2023-10-05 22:13:04,715 - Epoch: [193][ 300/ 1236] Overall Loss 0.164011 Objective Loss 0.164011 LR 0.000063 Time 0.022568 +2023-10-05 22:13:04,916 - Epoch: [193][ 310/ 1236] Overall Loss 0.163221 Objective Loss 0.163221 LR 0.000063 Time 0.022488 +2023-10-05 22:13:05,116 - Epoch: [193][ 320/ 1236] Overall Loss 0.163118 Objective Loss 0.163118 LR 0.000063 Time 0.022411 +2023-10-05 22:13:05,318 - Epoch: [193][ 330/ 1236] Overall Loss 0.163165 Objective Loss 0.163165 LR 0.000063 Time 0.022340 +2023-10-05 22:13:05,523 - Epoch: [193][ 340/ 1236] Overall Loss 0.163429 Objective Loss 0.163429 LR 0.000063 Time 0.022287 +2023-10-05 22:13:05,739 - Epoch: [193][ 350/ 1236] Overall Loss 0.163468 Objective Loss 0.163468 LR 0.000063 Time 0.022264 +2023-10-05 22:13:05,949 - Epoch: [193][ 360/ 1236] Overall Loss 0.163633 Objective Loss 0.163633 LR 0.000063 Time 0.022229 +2023-10-05 22:13:06,152 - Epoch: [193][ 370/ 1236] Overall Loss 0.163476 Objective Loss 0.163476 LR 0.000063 Time 0.022175 +2023-10-05 22:13:06,354 - Epoch: [193][ 380/ 1236] Overall Loss 0.163543 Objective Loss 0.163543 LR 0.000063 Time 0.022123 +2023-10-05 22:13:06,554 - Epoch: [193][ 390/ 1236] Overall Loss 0.163344 Objective Loss 0.163344 LR 0.000063 Time 0.022068 +2023-10-05 22:13:06,757 - Epoch: [193][ 400/ 1236] Overall Loss 0.163142 Objective Loss 0.163142 LR 0.000063 Time 0.022022 +2023-10-05 22:13:06,957 - Epoch: [193][ 410/ 1236] Overall Loss 0.163342 Objective Loss 0.163342 LR 0.000063 Time 0.021971 +2023-10-05 22:13:07,159 - Epoch: [193][ 420/ 1236] Overall Loss 0.163578 Objective Loss 0.163578 LR 0.000063 Time 0.021928 +2023-10-05 22:13:07,359 - Epoch: [193][ 430/ 1236] Overall Loss 0.163795 Objective Loss 0.163795 LR 0.000063 Time 0.021882 +2023-10-05 22:13:07,561 - Epoch: [193][ 440/ 1236] Overall Loss 0.163896 Objective Loss 0.163896 LR 0.000063 Time 0.021843 +2023-10-05 22:13:07,761 - Epoch: [193][ 450/ 1236] Overall Loss 0.163879 Objective Loss 0.163879 LR 0.000063 Time 0.021801 +2023-10-05 22:13:07,963 - Epoch: [193][ 460/ 1236] Overall Loss 0.163823 Objective Loss 0.163823 LR 0.000063 Time 0.021766 +2023-10-05 22:13:08,163 - Epoch: [193][ 470/ 1236] Overall Loss 0.163711 Objective Loss 0.163711 LR 0.000063 Time 0.021727 +2023-10-05 22:13:08,365 - Epoch: [193][ 480/ 1236] Overall Loss 0.163397 Objective Loss 0.163397 LR 0.000063 Time 0.021696 +2023-10-05 22:13:08,565 - Epoch: [193][ 490/ 1236] Overall Loss 0.163108 Objective Loss 0.163108 LR 0.000063 Time 0.021661 +2023-10-05 22:13:08,767 - Epoch: [193][ 500/ 1236] Overall Loss 0.163112 Objective Loss 0.163112 LR 0.000063 Time 0.021630 +2023-10-05 22:13:08,967 - Epoch: [193][ 510/ 1236] Overall Loss 0.163367 Objective Loss 0.163367 LR 0.000063 Time 0.021597 +2023-10-05 22:13:09,169 - Epoch: [193][ 520/ 1236] Overall Loss 0.163476 Objective Loss 0.163476 LR 0.000063 Time 0.021570 +2023-10-05 22:13:09,369 - Epoch: [193][ 530/ 1236] Overall Loss 0.163360 Objective Loss 0.163360 LR 0.000063 Time 0.021540 +2023-10-05 22:13:09,572 - Epoch: [193][ 540/ 1236] Overall Loss 0.163434 Objective Loss 0.163434 LR 0.000063 Time 0.021515 +2023-10-05 22:13:09,771 - Epoch: [193][ 550/ 1236] Overall Loss 0.163384 Objective Loss 0.163384 LR 0.000063 Time 0.021487 +2023-10-05 22:13:09,973 - Epoch: [193][ 560/ 1236] Overall Loss 0.163294 Objective Loss 0.163294 LR 0.000063 Time 0.021463 +2023-10-05 22:13:10,173 - Epoch: [193][ 570/ 1236] Overall Loss 0.163105 Objective Loss 0.163105 LR 0.000063 Time 0.021436 +2023-10-05 22:13:10,376 - Epoch: [193][ 580/ 1236] Overall Loss 0.162756 Objective Loss 0.162756 LR 0.000063 Time 0.021415 +2023-10-05 22:13:10,575 - Epoch: [193][ 590/ 1236] Overall Loss 0.162444 Objective Loss 0.162444 LR 0.000063 Time 0.021390 +2023-10-05 22:13:10,778 - Epoch: [193][ 600/ 1236] Overall Loss 0.162211 Objective Loss 0.162211 LR 0.000063 Time 0.021370 +2023-10-05 22:13:10,978 - Epoch: [193][ 610/ 1236] Overall Loss 0.162380 Objective Loss 0.162380 LR 0.000063 Time 0.021347 +2023-10-05 22:13:11,180 - Epoch: [193][ 620/ 1236] Overall Loss 0.162517 Objective Loss 0.162517 LR 0.000063 Time 0.021328 +2023-10-05 22:13:11,380 - Epoch: [193][ 630/ 1236] Overall Loss 0.162624 Objective Loss 0.162624 LR 0.000063 Time 0.021306 +2023-10-05 22:13:11,582 - Epoch: [193][ 640/ 1236] Overall Loss 0.162435 Objective Loss 0.162435 LR 0.000063 Time 0.021289 +2023-10-05 22:13:11,782 - Epoch: [193][ 650/ 1236] Overall Loss 0.162255 Objective Loss 0.162255 LR 0.000063 Time 0.021269 +2023-10-05 22:13:11,985 - Epoch: [193][ 660/ 1236] Overall Loss 0.162407 Objective Loss 0.162407 LR 0.000063 Time 0.021253 +2023-10-05 22:13:12,185 - Epoch: [193][ 670/ 1236] Overall Loss 0.162310 Objective Loss 0.162310 LR 0.000063 Time 0.021233 +2023-10-05 22:13:12,387 - Epoch: [193][ 680/ 1236] Overall Loss 0.162258 Objective Loss 0.162258 LR 0.000063 Time 0.021218 +2023-10-05 22:13:12,587 - Epoch: [193][ 690/ 1236] Overall Loss 0.162030 Objective Loss 0.162030 LR 0.000063 Time 0.021200 +2023-10-05 22:13:12,790 - Epoch: [193][ 700/ 1236] Overall Loss 0.161842 Objective Loss 0.161842 LR 0.000063 Time 0.021186 +2023-10-05 22:13:12,989 - Epoch: [193][ 710/ 1236] Overall Loss 0.161869 Objective Loss 0.161869 LR 0.000063 Time 0.021168 +2023-10-05 22:13:13,192 - Epoch: [193][ 720/ 1236] Overall Loss 0.161717 Objective Loss 0.161717 LR 0.000063 Time 0.021155 +2023-10-05 22:13:13,392 - Epoch: [193][ 730/ 1236] Overall Loss 0.161660 Objective Loss 0.161660 LR 0.000063 Time 0.021139 +2023-10-05 22:13:13,594 - Epoch: [193][ 740/ 1236] Overall Loss 0.161644 Objective Loss 0.161644 LR 0.000063 Time 0.021126 +2023-10-05 22:13:13,794 - Epoch: [193][ 750/ 1236] Overall Loss 0.161808 Objective Loss 0.161808 LR 0.000063 Time 0.021110 +2023-10-05 22:13:13,996 - Epoch: [193][ 760/ 1236] Overall Loss 0.161742 Objective Loss 0.161742 LR 0.000063 Time 0.021098 +2023-10-05 22:13:14,196 - Epoch: [193][ 770/ 1236] Overall Loss 0.161572 Objective Loss 0.161572 LR 0.000063 Time 0.021083 +2023-10-05 22:13:14,399 - Epoch: [193][ 780/ 1236] Overall Loss 0.161275 Objective Loss 0.161275 LR 0.000063 Time 0.021072 +2023-10-05 22:13:14,599 - Epoch: [193][ 790/ 1236] Overall Loss 0.161423 Objective Loss 0.161423 LR 0.000063 Time 0.021058 +2023-10-05 22:13:14,801 - Epoch: [193][ 800/ 1236] Overall Loss 0.161461 Objective Loss 0.161461 LR 0.000063 Time 0.021047 +2023-10-05 22:13:15,001 - Epoch: [193][ 810/ 1236] Overall Loss 0.161462 Objective Loss 0.161462 LR 0.000063 Time 0.021034 +2023-10-05 22:13:15,204 - Epoch: [193][ 820/ 1236] Overall Loss 0.161487 Objective Loss 0.161487 LR 0.000063 Time 0.021024 +2023-10-05 22:13:15,404 - Epoch: [193][ 830/ 1236] Overall Loss 0.161476 Objective Loss 0.161476 LR 0.000063 Time 0.021011 +2023-10-05 22:13:15,606 - Epoch: [193][ 840/ 1236] Overall Loss 0.161556 Objective Loss 0.161556 LR 0.000063 Time 0.021002 +2023-10-05 22:13:15,806 - Epoch: [193][ 850/ 1236] Overall Loss 0.161769 Objective Loss 0.161769 LR 0.000063 Time 0.020990 +2023-10-05 22:13:16,009 - Epoch: [193][ 860/ 1236] Overall Loss 0.161658 Objective Loss 0.161658 LR 0.000063 Time 0.020981 +2023-10-05 22:13:16,209 - Epoch: [193][ 870/ 1236] Overall Loss 0.161689 Objective Loss 0.161689 LR 0.000063 Time 0.020969 +2023-10-05 22:13:16,411 - Epoch: [193][ 880/ 1236] Overall Loss 0.161482 Objective Loss 0.161482 LR 0.000063 Time 0.020960 +2023-10-05 22:13:16,611 - Epoch: [193][ 890/ 1236] Overall Loss 0.161563 Objective Loss 0.161563 LR 0.000063 Time 0.020949 +2023-10-05 22:13:16,814 - Epoch: [193][ 900/ 1236] Overall Loss 0.161716 Objective Loss 0.161716 LR 0.000063 Time 0.020941 +2023-10-05 22:13:17,014 - Epoch: [193][ 910/ 1236] Overall Loss 0.161636 Objective Loss 0.161636 LR 0.000063 Time 0.020930 +2023-10-05 22:13:17,216 - Epoch: [193][ 920/ 1236] Overall Loss 0.161788 Objective Loss 0.161788 LR 0.000063 Time 0.020922 +2023-10-05 22:13:17,416 - Epoch: [193][ 930/ 1236] Overall Loss 0.162002 Objective Loss 0.162002 LR 0.000063 Time 0.020912 +2023-10-05 22:13:17,619 - Epoch: [193][ 940/ 1236] Overall Loss 0.162090 Objective Loss 0.162090 LR 0.000063 Time 0.020904 +2023-10-05 22:13:17,819 - Epoch: [193][ 950/ 1236] Overall Loss 0.162205 Objective Loss 0.162205 LR 0.000063 Time 0.020895 +2023-10-05 22:13:18,021 - Epoch: [193][ 960/ 1236] Overall Loss 0.162232 Objective Loss 0.162232 LR 0.000063 Time 0.020887 +2023-10-05 22:13:18,221 - Epoch: [193][ 970/ 1236] Overall Loss 0.162225 Objective Loss 0.162225 LR 0.000063 Time 0.020878 +2023-10-05 22:13:18,424 - Epoch: [193][ 980/ 1236] Overall Loss 0.162388 Objective Loss 0.162388 LR 0.000063 Time 0.020871 +2023-10-05 22:13:18,624 - Epoch: [193][ 990/ 1236] Overall Loss 0.162647 Objective Loss 0.162647 LR 0.000063 Time 0.020862 +2023-10-05 22:13:18,826 - Epoch: [193][ 1000/ 1236] Overall Loss 0.162479 Objective Loss 0.162479 LR 0.000063 Time 0.020855 +2023-10-05 22:13:19,026 - Epoch: [193][ 1010/ 1236] Overall Loss 0.162398 Objective Loss 0.162398 LR 0.000063 Time 0.020847 +2023-10-05 22:13:19,229 - Epoch: [193][ 1020/ 1236] Overall Loss 0.162296 Objective Loss 0.162296 LR 0.000063 Time 0.020840 +2023-10-05 22:13:19,429 - Epoch: [193][ 1030/ 1236] Overall Loss 0.162066 Objective Loss 0.162066 LR 0.000063 Time 0.020832 +2023-10-05 22:13:19,631 - Epoch: [193][ 1040/ 1236] Overall Loss 0.162002 Objective Loss 0.162002 LR 0.000063 Time 0.020826 +2023-10-05 22:13:19,831 - Epoch: [193][ 1050/ 1236] Overall Loss 0.161894 Objective Loss 0.161894 LR 0.000063 Time 0.020818 +2023-10-05 22:13:20,034 - Epoch: [193][ 1060/ 1236] Overall Loss 0.162008 Objective Loss 0.162008 LR 0.000063 Time 0.020812 +2023-10-05 22:13:20,234 - Epoch: [193][ 1070/ 1236] Overall Loss 0.161949 Objective Loss 0.161949 LR 0.000063 Time 0.020804 +2023-10-05 22:13:20,436 - Epoch: [193][ 1080/ 1236] Overall Loss 0.162032 Objective Loss 0.162032 LR 0.000063 Time 0.020798 +2023-10-05 22:13:20,636 - Epoch: [193][ 1090/ 1236] Overall Loss 0.162043 Objective Loss 0.162043 LR 0.000063 Time 0.020791 +2023-10-05 22:13:20,839 - Epoch: [193][ 1100/ 1236] Overall Loss 0.161958 Objective Loss 0.161958 LR 0.000063 Time 0.020786 +2023-10-05 22:13:21,039 - Epoch: [193][ 1110/ 1236] Overall Loss 0.162004 Objective Loss 0.162004 LR 0.000063 Time 0.020778 +2023-10-05 22:13:21,241 - Epoch: [193][ 1120/ 1236] Overall Loss 0.161891 Objective Loss 0.161891 LR 0.000063 Time 0.020773 +2023-10-05 22:13:21,441 - Epoch: [193][ 1130/ 1236] Overall Loss 0.161875 Objective Loss 0.161875 LR 0.000063 Time 0.020766 +2023-10-05 22:13:21,644 - Epoch: [193][ 1140/ 1236] Overall Loss 0.161866 Objective Loss 0.161866 LR 0.000063 Time 0.020761 +2023-10-05 22:13:21,844 - Epoch: [193][ 1150/ 1236] Overall Loss 0.161862 Objective Loss 0.161862 LR 0.000063 Time 0.020754 +2023-10-05 22:13:22,046 - Epoch: [193][ 1160/ 1236] Overall Loss 0.161748 Objective Loss 0.161748 LR 0.000063 Time 0.020749 +2023-10-05 22:13:22,247 - Epoch: [193][ 1170/ 1236] Overall Loss 0.161774 Objective Loss 0.161774 LR 0.000063 Time 0.020743 +2023-10-05 22:13:22,449 - Epoch: [193][ 1180/ 1236] Overall Loss 0.161756 Objective Loss 0.161756 LR 0.000063 Time 0.020738 +2023-10-05 22:13:22,649 - Epoch: [193][ 1190/ 1236] Overall Loss 0.161873 Objective Loss 0.161873 LR 0.000063 Time 0.020732 +2023-10-05 22:13:22,851 - Epoch: [193][ 1200/ 1236] Overall Loss 0.161819 Objective Loss 0.161819 LR 0.000063 Time 0.020727 +2023-10-05 22:13:23,051 - Epoch: [193][ 1210/ 1236] Overall Loss 0.161816 Objective Loss 0.161816 LR 0.000063 Time 0.020721 +2023-10-05 22:13:23,254 - Epoch: [193][ 1220/ 1236] Overall Loss 0.161794 Objective Loss 0.161794 LR 0.000063 Time 0.020717 +2023-10-05 22:13:23,507 - Epoch: [193][ 1230/ 1236] Overall Loss 0.161780 Objective Loss 0.161780 LR 0.000063 Time 0.020754 +2023-10-05 22:13:23,624 - Epoch: [193][ 1236/ 1236] Overall Loss 0.161791 Objective Loss 0.161791 Top1 89.002037 Top5 98.574338 LR 0.000063 Time 0.020748 +2023-10-05 22:13:23,746 - --- validate (epoch=193)----------- +2023-10-05 22:13:23,746 - 29943 samples (256 per mini-batch) +2023-10-05 22:13:24,195 - Epoch: [193][ 10/ 117] Loss 0.315542 Top1 85.429688 Top5 98.632812 +2023-10-05 22:13:24,338 - Epoch: [193][ 20/ 117] Loss 0.320014 Top1 85.312500 Top5 98.261719 +2023-10-05 22:13:24,481 - Epoch: [193][ 30/ 117] Loss 0.310700 Top1 85.729167 Top5 98.359375 +2023-10-05 22:13:24,625 - Epoch: [193][ 40/ 117] Loss 0.303607 Top1 85.966797 Top5 98.388672 +2023-10-05 22:13:24,768 - Epoch: [193][ 50/ 117] Loss 0.308742 Top1 86.000000 Top5 98.359375 +2023-10-05 22:13:24,910 - Epoch: [193][ 60/ 117] Loss 0.314877 Top1 85.826823 Top5 98.359375 +2023-10-05 22:13:25,052 - Epoch: [193][ 70/ 117] Loss 0.312179 Top1 85.976562 Top5 98.325893 +2023-10-05 22:13:25,194 - Epoch: [193][ 80/ 117] Loss 0.309438 Top1 86.049805 Top5 98.256836 +2023-10-05 22:13:25,336 - Epoch: [193][ 90/ 117] Loss 0.309692 Top1 86.072049 Top5 98.216146 +2023-10-05 22:13:25,479 - Epoch: [193][ 100/ 117] Loss 0.306289 Top1 86.203125 Top5 98.257812 +2023-10-05 22:13:25,629 - Epoch: [193][ 110/ 117] Loss 0.304367 Top1 86.196733 Top5 98.267045 +2023-10-05 22:13:25,714 - Epoch: [193][ 117/ 117] Loss 0.304089 Top1 86.167051 Top5 98.250008 +2023-10-05 22:13:25,855 - ==> Top1: 86.167 Top5: 98.250 Loss: 0.304 + +2023-10-05 22:13:25,856 - ==> Confusion: +[[ 940 1 5 1 5 1 0 0 5 66 1 1 2 2 4 2 2 0 0 1 11] + [ 0 1072 1 0 11 10 1 12 0 0 1 2 0 0 2 3 2 0 7 1 6] + [ 5 3 980 5 1 0 17 8 0 0 4 1 8 1 0 2 0 3 8 4 6] + [ 1 1 17 969 2 4 0 2 1 1 5 1 5 1 26 4 0 6 24 2 17] + [ 24 10 1 0 968 3 0 1 0 9 1 1 0 2 7 3 12 1 0 1 6] + [ 3 38 1 0 4 981 0 25 0 0 5 7 0 12 7 2 3 0 3 4 21] + [ 0 5 25 1 0 0 1130 6 0 0 0 4 1 0 1 4 0 1 1 6 6] + [ 4 19 14 0 3 24 4 1077 0 2 3 9 2 2 1 3 0 0 41 3 7] + [ 18 0 3 0 1 2 0 2 985 41 7 3 1 8 9 1 2 1 2 0 3] + [ 92 1 3 1 3 2 0 2 15 965 1 2 0 14 4 3 1 1 0 1 8] + [ 3 3 10 3 0 1 2 3 13 4 977 5 0 11 4 1 2 0 4 0 7] + [ 1 0 3 0 0 10 0 3 0 1 0 956 23 6 0 5 1 15 0 3 8] + [ 1 3 3 3 0 1 0 1 0 0 1 29 995 3 0 3 3 10 0 3 9] + [ 2 0 1 0 0 5 0 0 11 19 8 2 3 1052 3 1 1 0 0 1 10] + [ 15 3 3 5 6 1 0 0 27 2 1 1 2 3 1006 0 1 2 10 0 13] + [ 1 3 1 0 2 0 1 0 0 0 0 8 8 1 1 1073 13 10 0 8 4] + [ 1 12 1 0 4 2 0 0 1 0 0 2 0 0 3 9 1110 0 0 2 14] + [ 0 0 1 1 1 0 4 0 2 1 0 0 17 1 0 5 0 1001 1 0 3] + [ 1 6 5 20 1 0 0 20 1 0 2 0 2 0 8 0 0 0 988 2 12] + [ 0 3 3 3 1 4 7 10 0 0 2 15 3 2 0 9 8 1 3 1068 10] + [ 132 171 152 42 90 101 33 76 95 74 165 81 289 263 114 56 122 56 131 154 5508]] + +2023-10-05 22:13:25,857 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:13:25,857 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:13:25,864 - + +2023-10-05 22:13:25,864 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:13:26,965 - Epoch: [194][ 10/ 1236] Overall Loss 0.164835 Objective Loss 0.164835 LR 0.000063 Time 0.110130 +2023-10-05 22:13:27,168 - Epoch: [194][ 20/ 1236] Overall Loss 0.164351 Objective Loss 0.164351 LR 0.000063 Time 0.065189 +2023-10-05 22:13:27,371 - Epoch: [194][ 30/ 1236] Overall Loss 0.158518 Objective Loss 0.158518 LR 0.000063 Time 0.050193 +2023-10-05 22:13:27,574 - Epoch: [194][ 40/ 1236] Overall Loss 0.155041 Objective Loss 0.155041 LR 0.000063 Time 0.042724 +2023-10-05 22:13:27,778 - Epoch: [194][ 50/ 1236] Overall Loss 0.154784 Objective Loss 0.154784 LR 0.000063 Time 0.038239 +2023-10-05 22:13:27,981 - Epoch: [194][ 60/ 1236] Overall Loss 0.158941 Objective Loss 0.158941 LR 0.000063 Time 0.035246 +2023-10-05 22:13:28,184 - Epoch: [194][ 70/ 1236] Overall Loss 0.158730 Objective Loss 0.158730 LR 0.000063 Time 0.033104 +2023-10-05 22:13:28,386 - Epoch: [194][ 80/ 1236] Overall Loss 0.157845 Objective Loss 0.157845 LR 0.000063 Time 0.031490 +2023-10-05 22:13:28,589 - Epoch: [194][ 90/ 1236] Overall Loss 0.158227 Objective Loss 0.158227 LR 0.000063 Time 0.030238 +2023-10-05 22:13:28,791 - Epoch: [194][ 100/ 1236] Overall Loss 0.158219 Objective Loss 0.158219 LR 0.000063 Time 0.029233 +2023-10-05 22:13:28,993 - Epoch: [194][ 110/ 1236] Overall Loss 0.158577 Objective Loss 0.158577 LR 0.000063 Time 0.028409 +2023-10-05 22:13:29,193 - Epoch: [194][ 120/ 1236] Overall Loss 0.159440 Objective Loss 0.159440 LR 0.000063 Time 0.027709 +2023-10-05 22:13:29,394 - Epoch: [194][ 130/ 1236] Overall Loss 0.162473 Objective Loss 0.162473 LR 0.000063 Time 0.027115 +2023-10-05 22:13:29,594 - Epoch: [194][ 140/ 1236] Overall Loss 0.162386 Objective Loss 0.162386 LR 0.000063 Time 0.026606 +2023-10-05 22:13:29,794 - Epoch: [194][ 150/ 1236] Overall Loss 0.162044 Objective Loss 0.162044 LR 0.000063 Time 0.026167 +2023-10-05 22:13:29,999 - Epoch: [194][ 160/ 1236] Overall Loss 0.162402 Objective Loss 0.162402 LR 0.000063 Time 0.025807 +2023-10-05 22:13:30,205 - Epoch: [194][ 170/ 1236] Overall Loss 0.163849 Objective Loss 0.163849 LR 0.000063 Time 0.025497 +2023-10-05 22:13:30,410 - Epoch: [194][ 180/ 1236] Overall Loss 0.164221 Objective Loss 0.164221 LR 0.000063 Time 0.025217 +2023-10-05 22:13:30,616 - Epoch: [194][ 190/ 1236] Overall Loss 0.163828 Objective Loss 0.163828 LR 0.000063 Time 0.024965 +2023-10-05 22:13:30,821 - Epoch: [194][ 200/ 1236] Overall Loss 0.163197 Objective Loss 0.163197 LR 0.000063 Time 0.024740 +2023-10-05 22:13:31,026 - Epoch: [194][ 210/ 1236] Overall Loss 0.162917 Objective Loss 0.162917 LR 0.000063 Time 0.024531 +2023-10-05 22:13:31,232 - Epoch: [194][ 220/ 1236] Overall Loss 0.163263 Objective Loss 0.163263 LR 0.000063 Time 0.024348 +2023-10-05 22:13:31,438 - Epoch: [194][ 230/ 1236] Overall Loss 0.162815 Objective Loss 0.162815 LR 0.000063 Time 0.024182 +2023-10-05 22:13:31,643 - Epoch: [194][ 240/ 1236] Overall Loss 0.162031 Objective Loss 0.162031 LR 0.000063 Time 0.024027 +2023-10-05 22:13:31,848 - Epoch: [194][ 250/ 1236] Overall Loss 0.161214 Objective Loss 0.161214 LR 0.000063 Time 0.023880 +2023-10-05 22:13:32,054 - Epoch: [194][ 260/ 1236] Overall Loss 0.161235 Objective Loss 0.161235 LR 0.000063 Time 0.023749 +2023-10-05 22:13:32,259 - Epoch: [194][ 270/ 1236] Overall Loss 0.160848 Objective Loss 0.160848 LR 0.000063 Time 0.023625 +2023-10-05 22:13:32,464 - Epoch: [194][ 280/ 1236] Overall Loss 0.161159 Objective Loss 0.161159 LR 0.000063 Time 0.023512 +2023-10-05 22:13:32,670 - Epoch: [194][ 290/ 1236] Overall Loss 0.161544 Objective Loss 0.161544 LR 0.000063 Time 0.023409 +2023-10-05 22:13:32,875 - Epoch: [194][ 300/ 1236] Overall Loss 0.162704 Objective Loss 0.162704 LR 0.000063 Time 0.023310 +2023-10-05 22:13:33,081 - Epoch: [194][ 310/ 1236] Overall Loss 0.163197 Objective Loss 0.163197 LR 0.000063 Time 0.023221 +2023-10-05 22:13:33,287 - Epoch: [194][ 320/ 1236] Overall Loss 0.162639 Objective Loss 0.162639 LR 0.000063 Time 0.023137 +2023-10-05 22:13:33,493 - Epoch: [194][ 330/ 1236] Overall Loss 0.162912 Objective Loss 0.162912 LR 0.000063 Time 0.023054 +2023-10-05 22:13:33,698 - Epoch: [194][ 340/ 1236] Overall Loss 0.163344 Objective Loss 0.163344 LR 0.000063 Time 0.022979 +2023-10-05 22:13:33,904 - Epoch: [194][ 350/ 1236] Overall Loss 0.163710 Objective Loss 0.163710 LR 0.000063 Time 0.022908 +2023-10-05 22:13:34,110 - Epoch: [194][ 360/ 1236] Overall Loss 0.163978 Objective Loss 0.163978 LR 0.000063 Time 0.022841 +2023-10-05 22:13:34,314 - Epoch: [194][ 370/ 1236] Overall Loss 0.163600 Objective Loss 0.163600 LR 0.000063 Time 0.022774 +2023-10-05 22:13:34,517 - Epoch: [194][ 380/ 1236] Overall Loss 0.164106 Objective Loss 0.164106 LR 0.000063 Time 0.022710 +2023-10-05 22:13:34,721 - Epoch: [194][ 390/ 1236] Overall Loss 0.163935 Objective Loss 0.163935 LR 0.000063 Time 0.022649 +2023-10-05 22:13:34,924 - Epoch: [194][ 400/ 1236] Overall Loss 0.163777 Objective Loss 0.163777 LR 0.000063 Time 0.022589 +2023-10-05 22:13:35,127 - Epoch: [194][ 410/ 1236] Overall Loss 0.163801 Objective Loss 0.163801 LR 0.000063 Time 0.022532 +2023-10-05 22:13:35,330 - Epoch: [194][ 420/ 1236] Overall Loss 0.163526 Objective Loss 0.163526 LR 0.000063 Time 0.022478 +2023-10-05 22:13:35,533 - Epoch: [194][ 430/ 1236] Overall Loss 0.163610 Objective Loss 0.163610 LR 0.000063 Time 0.022427 +2023-10-05 22:13:35,736 - Epoch: [194][ 440/ 1236] Overall Loss 0.163594 Objective Loss 0.163594 LR 0.000063 Time 0.022378 +2023-10-05 22:13:35,939 - Epoch: [194][ 450/ 1236] Overall Loss 0.163764 Objective Loss 0.163764 LR 0.000063 Time 0.022331 +2023-10-05 22:13:36,142 - Epoch: [194][ 460/ 1236] Overall Loss 0.163633 Objective Loss 0.163633 LR 0.000063 Time 0.022286 +2023-10-05 22:13:36,346 - Epoch: [194][ 470/ 1236] Overall Loss 0.163827 Objective Loss 0.163827 LR 0.000063 Time 0.022244 +2023-10-05 22:13:36,549 - Epoch: [194][ 480/ 1236] Overall Loss 0.163719 Objective Loss 0.163719 LR 0.000063 Time 0.022202 +2023-10-05 22:13:36,752 - Epoch: [194][ 490/ 1236] Overall Loss 0.163762 Objective Loss 0.163762 LR 0.000063 Time 0.022164 +2023-10-05 22:13:36,955 - Epoch: [194][ 500/ 1236] Overall Loss 0.163735 Objective Loss 0.163735 LR 0.000063 Time 0.022126 +2023-10-05 22:13:37,158 - Epoch: [194][ 510/ 1236] Overall Loss 0.164037 Objective Loss 0.164037 LR 0.000063 Time 0.022089 +2023-10-05 22:13:37,361 - Epoch: [194][ 520/ 1236] Overall Loss 0.163787 Objective Loss 0.163787 LR 0.000063 Time 0.022054 +2023-10-05 22:13:37,564 - Epoch: [194][ 530/ 1236] Overall Loss 0.163984 Objective Loss 0.163984 LR 0.000063 Time 0.022020 +2023-10-05 22:13:37,767 - Epoch: [194][ 540/ 1236] Overall Loss 0.163818 Objective Loss 0.163818 LR 0.000063 Time 0.021988 +2023-10-05 22:13:37,971 - Epoch: [194][ 550/ 1236] Overall Loss 0.164069 Objective Loss 0.164069 LR 0.000063 Time 0.021958 +2023-10-05 22:13:38,174 - Epoch: [194][ 560/ 1236] Overall Loss 0.164083 Objective Loss 0.164083 LR 0.000063 Time 0.021928 +2023-10-05 22:13:38,378 - Epoch: [194][ 570/ 1236] Overall Loss 0.164059 Objective Loss 0.164059 LR 0.000063 Time 0.021900 +2023-10-05 22:13:38,581 - Epoch: [194][ 580/ 1236] Overall Loss 0.164119 Objective Loss 0.164119 LR 0.000063 Time 0.021871 +2023-10-05 22:13:38,784 - Epoch: [194][ 590/ 1236] Overall Loss 0.163902 Objective Loss 0.163902 LR 0.000063 Time 0.021845 +2023-10-05 22:13:38,987 - Epoch: [194][ 600/ 1236] Overall Loss 0.163768 Objective Loss 0.163768 LR 0.000063 Time 0.021818 +2023-10-05 22:13:39,190 - Epoch: [194][ 610/ 1236] Overall Loss 0.163994 Objective Loss 0.163994 LR 0.000063 Time 0.021793 +2023-10-05 22:13:39,393 - Epoch: [194][ 620/ 1236] Overall Loss 0.163928 Objective Loss 0.163928 LR 0.000063 Time 0.021767 +2023-10-05 22:13:39,596 - Epoch: [194][ 630/ 1236] Overall Loss 0.164082 Objective Loss 0.164082 LR 0.000063 Time 0.021744 +2023-10-05 22:13:39,799 - Epoch: [194][ 640/ 1236] Overall Loss 0.164080 Objective Loss 0.164080 LR 0.000063 Time 0.021720 +2023-10-05 22:13:40,002 - Epoch: [194][ 650/ 1236] Overall Loss 0.164146 Objective Loss 0.164146 LR 0.000063 Time 0.021699 +2023-10-05 22:13:40,206 - Epoch: [194][ 660/ 1236] Overall Loss 0.164019 Objective Loss 0.164019 LR 0.000063 Time 0.021678 +2023-10-05 22:13:40,409 - Epoch: [194][ 670/ 1236] Overall Loss 0.164174 Objective Loss 0.164174 LR 0.000063 Time 0.021658 +2023-10-05 22:13:40,613 - Epoch: [194][ 680/ 1236] Overall Loss 0.164318 Objective Loss 0.164318 LR 0.000063 Time 0.021638 +2023-10-05 22:13:40,816 - Epoch: [194][ 690/ 1236] Overall Loss 0.164086 Objective Loss 0.164086 LR 0.000063 Time 0.021618 +2023-10-05 22:13:41,019 - Epoch: [194][ 700/ 1236] Overall Loss 0.164087 Objective Loss 0.164087 LR 0.000063 Time 0.021598 +2023-10-05 22:13:41,223 - Epoch: [194][ 710/ 1236] Overall Loss 0.164259 Objective Loss 0.164259 LR 0.000063 Time 0.021581 +2023-10-05 22:13:41,426 - Epoch: [194][ 720/ 1236] Overall Loss 0.164176 Objective Loss 0.164176 LR 0.000063 Time 0.021563 +2023-10-05 22:13:41,629 - Epoch: [194][ 730/ 1236] Overall Loss 0.164268 Objective Loss 0.164268 LR 0.000063 Time 0.021545 +2023-10-05 22:13:41,832 - Epoch: [194][ 740/ 1236] Overall Loss 0.164045 Objective Loss 0.164045 LR 0.000063 Time 0.021527 +2023-10-05 22:13:42,035 - Epoch: [194][ 750/ 1236] Overall Loss 0.163832 Objective Loss 0.163832 LR 0.000063 Time 0.021511 +2023-10-05 22:13:42,238 - Epoch: [194][ 760/ 1236] Overall Loss 0.164004 Objective Loss 0.164004 LR 0.000063 Time 0.021495 +2023-10-05 22:13:42,442 - Epoch: [194][ 770/ 1236] Overall Loss 0.163923 Objective Loss 0.163923 LR 0.000063 Time 0.021480 +2023-10-05 22:13:42,645 - Epoch: [194][ 780/ 1236] Overall Loss 0.163945 Objective Loss 0.163945 LR 0.000063 Time 0.021465 +2023-10-05 22:13:42,849 - Epoch: [194][ 790/ 1236] Overall Loss 0.163730 Objective Loss 0.163730 LR 0.000063 Time 0.021450 +2023-10-05 22:13:43,052 - Epoch: [194][ 800/ 1236] Overall Loss 0.163645 Objective Loss 0.163645 LR 0.000063 Time 0.021435 +2023-10-05 22:13:43,255 - Epoch: [194][ 810/ 1236] Overall Loss 0.163549 Objective Loss 0.163549 LR 0.000063 Time 0.021422 +2023-10-05 22:13:43,458 - Epoch: [194][ 820/ 1236] Overall Loss 0.163668 Objective Loss 0.163668 LR 0.000063 Time 0.021407 +2023-10-05 22:13:43,662 - Epoch: [194][ 830/ 1236] Overall Loss 0.163616 Objective Loss 0.163616 LR 0.000063 Time 0.021394 +2023-10-05 22:13:43,865 - Epoch: [194][ 840/ 1236] Overall Loss 0.163511 Objective Loss 0.163511 LR 0.000063 Time 0.021381 +2023-10-05 22:13:44,069 - Epoch: [194][ 850/ 1236] Overall Loss 0.163437 Objective Loss 0.163437 LR 0.000063 Time 0.021369 +2023-10-05 22:13:44,271 - Epoch: [194][ 860/ 1236] Overall Loss 0.163230 Objective Loss 0.163230 LR 0.000063 Time 0.021355 +2023-10-05 22:13:44,475 - Epoch: [194][ 870/ 1236] Overall Loss 0.163527 Objective Loss 0.163527 LR 0.000063 Time 0.021344 +2023-10-05 22:13:44,678 - Epoch: [194][ 880/ 1236] Overall Loss 0.163576 Objective Loss 0.163576 LR 0.000063 Time 0.021331 +2023-10-05 22:13:44,882 - Epoch: [194][ 890/ 1236] Overall Loss 0.163689 Objective Loss 0.163689 LR 0.000063 Time 0.021320 +2023-10-05 22:13:45,085 - Epoch: [194][ 900/ 1236] Overall Loss 0.163711 Objective Loss 0.163711 LR 0.000063 Time 0.021308 +2023-10-05 22:13:45,288 - Epoch: [194][ 910/ 1236] Overall Loss 0.163545 Objective Loss 0.163545 LR 0.000063 Time 0.021297 +2023-10-05 22:13:45,491 - Epoch: [194][ 920/ 1236] Overall Loss 0.163363 Objective Loss 0.163363 LR 0.000063 Time 0.021286 +2023-10-05 22:13:45,695 - Epoch: [194][ 930/ 1236] Overall Loss 0.163134 Objective Loss 0.163134 LR 0.000063 Time 0.021276 +2023-10-05 22:13:45,898 - Epoch: [194][ 940/ 1236] Overall Loss 0.163307 Objective Loss 0.163307 LR 0.000063 Time 0.021265 +2023-10-05 22:13:46,102 - Epoch: [194][ 950/ 1236] Overall Loss 0.163214 Objective Loss 0.163214 LR 0.000063 Time 0.021255 +2023-10-05 22:13:46,305 - Epoch: [194][ 960/ 1236] Overall Loss 0.163176 Objective Loss 0.163176 LR 0.000063 Time 0.021245 +2023-10-05 22:13:46,508 - Epoch: [194][ 970/ 1236] Overall Loss 0.163500 Objective Loss 0.163500 LR 0.000063 Time 0.021235 +2023-10-05 22:13:46,713 - Epoch: [194][ 980/ 1236] Overall Loss 0.163370 Objective Loss 0.163370 LR 0.000063 Time 0.021225 +2023-10-05 22:13:46,916 - Epoch: [194][ 990/ 1236] Overall Loss 0.163300 Objective Loss 0.163300 LR 0.000063 Time 0.021215 +2023-10-05 22:13:47,119 - Epoch: [194][ 1000/ 1236] Overall Loss 0.163270 Objective Loss 0.163270 LR 0.000063 Time 0.021206 +2023-10-05 22:13:47,322 - Epoch: [194][ 1010/ 1236] Overall Loss 0.163105 Objective Loss 0.163105 LR 0.000063 Time 0.021197 +2023-10-05 22:13:47,526 - Epoch: [194][ 1020/ 1236] Overall Loss 0.162909 Objective Loss 0.162909 LR 0.000063 Time 0.021188 +2023-10-05 22:13:47,729 - Epoch: [194][ 1030/ 1236] Overall Loss 0.162861 Objective Loss 0.162861 LR 0.000063 Time 0.021179 +2023-10-05 22:13:47,932 - Epoch: [194][ 1040/ 1236] Overall Loss 0.162798 Objective Loss 0.162798 LR 0.000063 Time 0.021171 +2023-10-05 22:13:48,136 - Epoch: [194][ 1050/ 1236] Overall Loss 0.162709 Objective Loss 0.162709 LR 0.000063 Time 0.021162 +2023-10-05 22:13:48,339 - Epoch: [194][ 1060/ 1236] Overall Loss 0.162731 Objective Loss 0.162731 LR 0.000063 Time 0.021154 +2023-10-05 22:13:48,543 - Epoch: [194][ 1070/ 1236] Overall Loss 0.162726 Objective Loss 0.162726 LR 0.000063 Time 0.021147 +2023-10-05 22:13:48,746 - Epoch: [194][ 1080/ 1236] Overall Loss 0.162817 Objective Loss 0.162817 LR 0.000063 Time 0.021139 +2023-10-05 22:13:48,950 - Epoch: [194][ 1090/ 1236] Overall Loss 0.162864 Objective Loss 0.162864 LR 0.000063 Time 0.021131 +2023-10-05 22:13:49,153 - Epoch: [194][ 1100/ 1236] Overall Loss 0.162931 Objective Loss 0.162931 LR 0.000063 Time 0.021123 +2023-10-05 22:13:49,356 - Epoch: [194][ 1110/ 1236] Overall Loss 0.163039 Objective Loss 0.163039 LR 0.000063 Time 0.021116 +2023-10-05 22:13:49,559 - Epoch: [194][ 1120/ 1236] Overall Loss 0.162959 Objective Loss 0.162959 LR 0.000063 Time 0.021108 +2023-10-05 22:13:49,762 - Epoch: [194][ 1130/ 1236] Overall Loss 0.162812 Objective Loss 0.162812 LR 0.000063 Time 0.021101 +2023-10-05 22:13:49,965 - Epoch: [194][ 1140/ 1236] Overall Loss 0.162841 Objective Loss 0.162841 LR 0.000063 Time 0.021093 +2023-10-05 22:13:50,169 - Epoch: [194][ 1150/ 1236] Overall Loss 0.162715 Objective Loss 0.162715 LR 0.000063 Time 0.021087 +2023-10-05 22:13:50,372 - Epoch: [194][ 1160/ 1236] Overall Loss 0.162732 Objective Loss 0.162732 LR 0.000063 Time 0.021079 +2023-10-05 22:13:50,575 - Epoch: [194][ 1170/ 1236] Overall Loss 0.162793 Objective Loss 0.162793 LR 0.000063 Time 0.021073 +2023-10-05 22:13:50,778 - Epoch: [194][ 1180/ 1236] Overall Loss 0.162821 Objective Loss 0.162821 LR 0.000063 Time 0.021066 +2023-10-05 22:13:50,982 - Epoch: [194][ 1190/ 1236] Overall Loss 0.162972 Objective Loss 0.162972 LR 0.000063 Time 0.021060 +2023-10-05 22:13:51,185 - Epoch: [194][ 1200/ 1236] Overall Loss 0.163044 Objective Loss 0.163044 LR 0.000063 Time 0.021053 +2023-10-05 22:13:51,388 - Epoch: [194][ 1210/ 1236] Overall Loss 0.163206 Objective Loss 0.163206 LR 0.000063 Time 0.021047 +2023-10-05 22:13:51,591 - Epoch: [194][ 1220/ 1236] Overall Loss 0.163223 Objective Loss 0.163223 LR 0.000063 Time 0.021041 +2023-10-05 22:13:51,848 - Epoch: [194][ 1230/ 1236] Overall Loss 0.163143 Objective Loss 0.163143 LR 0.000063 Time 0.021078 +2023-10-05 22:13:51,965 - Epoch: [194][ 1236/ 1236] Overall Loss 0.163149 Objective Loss 0.163149 Top1 90.427699 Top5 98.981670 LR 0.000063 Time 0.021070 +2023-10-05 22:13:52,096 - --- validate (epoch=194)----------- +2023-10-05 22:13:52,096 - 29943 samples (256 per mini-batch) +2023-10-05 22:13:52,558 - Epoch: [194][ 10/ 117] Loss 0.290602 Top1 86.367188 Top5 98.515625 +2023-10-05 22:13:52,708 - Epoch: [194][ 20/ 117] Loss 0.295581 Top1 86.035156 Top5 98.320312 +2023-10-05 22:13:52,857 - Epoch: [194][ 30/ 117] Loss 0.301341 Top1 85.742188 Top5 98.229167 +2023-10-05 22:13:53,005 - Epoch: [194][ 40/ 117] Loss 0.294464 Top1 86.015625 Top5 98.242188 +2023-10-05 22:13:53,155 - Epoch: [194][ 50/ 117] Loss 0.292042 Top1 86.437500 Top5 98.382812 +2023-10-05 22:13:53,303 - Epoch: [194][ 60/ 117] Loss 0.298096 Top1 86.295573 Top5 98.320312 +2023-10-05 22:13:53,451 - Epoch: [194][ 70/ 117] Loss 0.301638 Top1 86.199777 Top5 98.309152 +2023-10-05 22:13:53,596 - Epoch: [194][ 80/ 117] Loss 0.301471 Top1 86.210938 Top5 98.315430 +2023-10-05 22:13:53,742 - Epoch: [194][ 90/ 117] Loss 0.302013 Top1 86.145833 Top5 98.272569 +2023-10-05 22:13:53,888 - Epoch: [194][ 100/ 117] Loss 0.301274 Top1 86.246094 Top5 98.289062 +2023-10-05 22:13:54,041 - Epoch: [194][ 110/ 117] Loss 0.304101 Top1 86.132812 Top5 98.288352 +2023-10-05 22:13:54,126 - Epoch: [194][ 117/ 117] Loss 0.304443 Top1 86.153692 Top5 98.290084 +2023-10-05 22:13:54,252 - ==> Top1: 86.154 Top5: 98.290 Loss: 0.304 + +2023-10-05 22:13:54,253 - ==> Confusion: +[[ 943 2 3 1 7 1 0 0 5 63 1 0 2 2 4 1 3 0 0 0 12] + [ 0 1064 1 0 7 17 1 19 0 0 1 2 0 0 1 4 2 1 7 0 4] + [ 5 1 971 13 2 0 16 8 0 0 4 1 4 3 0 4 0 3 8 5 8] + [ 1 1 10 970 2 5 2 2 1 1 4 0 6 2 27 3 0 7 25 2 18] + [ 23 8 1 0 967 3 1 2 0 11 0 1 0 1 7 3 9 3 2 1 7] + [ 3 34 1 1 5 977 1 26 1 2 4 9 0 16 5 2 2 0 2 4 21] + [ 0 7 17 0 0 0 1133 7 0 0 1 3 1 0 1 6 0 1 1 8 5] + [ 4 16 11 0 1 26 5 1079 1 4 2 9 1 3 0 1 0 0 38 7 10] + [ 18 1 2 0 0 1 0 1 984 40 8 3 1 6 12 2 2 1 3 1 3] + [ 98 1 3 1 1 2 0 0 14 960 0 2 0 15 4 5 1 1 0 1 10] + [ 3 3 8 2 1 1 5 6 12 1 969 3 1 12 3 2 2 0 7 2 10] + [ 1 1 2 0 1 10 0 3 0 1 0 964 15 5 0 4 0 16 0 8 4] + [ 1 3 1 2 1 2 0 2 0 0 2 28 985 5 0 4 3 12 2 5 10] + [ 2 0 1 0 0 4 0 0 13 17 5 4 3 1054 4 2 1 0 0 2 7] + [ 16 3 2 6 2 0 0 0 28 2 1 0 2 3 1005 0 2 3 9 0 17] + [ 1 3 1 0 1 0 1 0 0 0 0 7 6 3 1 1075 15 8 0 9 3] + [ 1 11 1 0 6 3 0 0 1 0 0 3 1 0 3 11 1102 0 0 2 16] + [ 0 0 1 1 1 0 2 0 0 0 0 2 12 2 0 4 0 1008 1 0 4] + [ 1 5 3 16 1 0 1 24 1 0 1 0 2 0 8 0 0 0 992 2 11] + [ 0 2 3 2 3 7 6 11 0 0 1 13 3 0 0 7 6 2 2 1076 8] + [ 121 161 114 55 78 101 44 87 93 72 159 85 289 273 124 49 120 63 135 163 5519]] + +2023-10-05 22:13:54,254 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:13:54,254 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:13:54,260 - + +2023-10-05 22:13:54,260 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:13:55,245 - Epoch: [195][ 10/ 1236] Overall Loss 0.147277 Objective Loss 0.147277 LR 0.000031 Time 0.098423 +2023-10-05 22:13:55,447 - Epoch: [195][ 20/ 1236] Overall Loss 0.151410 Objective Loss 0.151410 LR 0.000031 Time 0.059295 +2023-10-05 22:13:55,651 - Epoch: [195][ 30/ 1236] Overall Loss 0.153216 Objective Loss 0.153216 LR 0.000031 Time 0.046300 +2023-10-05 22:13:55,854 - Epoch: [195][ 40/ 1236] Overall Loss 0.155629 Objective Loss 0.155629 LR 0.000031 Time 0.039796 +2023-10-05 22:13:56,056 - Epoch: [195][ 50/ 1236] Overall Loss 0.156741 Objective Loss 0.156741 LR 0.000031 Time 0.035876 +2023-10-05 22:13:56,258 - Epoch: [195][ 60/ 1236] Overall Loss 0.155973 Objective Loss 0.155973 LR 0.000031 Time 0.033254 +2023-10-05 22:13:56,460 - Epoch: [195][ 70/ 1236] Overall Loss 0.157626 Objective Loss 0.157626 LR 0.000031 Time 0.031386 +2023-10-05 22:13:56,662 - Epoch: [195][ 80/ 1236] Overall Loss 0.159280 Objective Loss 0.159280 LR 0.000031 Time 0.029980 +2023-10-05 22:13:56,864 - Epoch: [195][ 90/ 1236] Overall Loss 0.158961 Objective Loss 0.158961 LR 0.000031 Time 0.028892 +2023-10-05 22:13:57,066 - Epoch: [195][ 100/ 1236] Overall Loss 0.158728 Objective Loss 0.158728 LR 0.000031 Time 0.028014 +2023-10-05 22:13:57,267 - Epoch: [195][ 110/ 1236] Overall Loss 0.161031 Objective Loss 0.161031 LR 0.000031 Time 0.027292 +2023-10-05 22:13:57,469 - Epoch: [195][ 120/ 1236] Overall Loss 0.160893 Objective Loss 0.160893 LR 0.000031 Time 0.026694 +2023-10-05 22:13:57,668 - Epoch: [195][ 130/ 1236] Overall Loss 0.161302 Objective Loss 0.161302 LR 0.000031 Time 0.026176 +2023-10-05 22:13:57,869 - Epoch: [195][ 140/ 1236] Overall Loss 0.160021 Objective Loss 0.160021 LR 0.000031 Time 0.025739 +2023-10-05 22:13:58,070 - Epoch: [195][ 150/ 1236] Overall Loss 0.158533 Objective Loss 0.158533 LR 0.000031 Time 0.025361 +2023-10-05 22:13:58,275 - Epoch: [195][ 160/ 1236] Overall Loss 0.159400 Objective Loss 0.159400 LR 0.000031 Time 0.025053 +2023-10-05 22:13:58,482 - Epoch: [195][ 170/ 1236] Overall Loss 0.160886 Objective Loss 0.160886 LR 0.000031 Time 0.024790 +2023-10-05 22:13:58,687 - Epoch: [195][ 180/ 1236] Overall Loss 0.161828 Objective Loss 0.161828 LR 0.000031 Time 0.024551 +2023-10-05 22:13:58,894 - Epoch: [195][ 190/ 1236] Overall Loss 0.161237 Objective Loss 0.161237 LR 0.000031 Time 0.024345 +2023-10-05 22:13:59,099 - Epoch: [195][ 200/ 1236] Overall Loss 0.161668 Objective Loss 0.161668 LR 0.000031 Time 0.024149 +2023-10-05 22:13:59,304 - Epoch: [195][ 210/ 1236] Overall Loss 0.160964 Objective Loss 0.160964 LR 0.000031 Time 0.023976 +2023-10-05 22:13:59,510 - Epoch: [195][ 220/ 1236] Overall Loss 0.161429 Objective Loss 0.161429 LR 0.000031 Time 0.023819 +2023-10-05 22:13:59,717 - Epoch: [195][ 230/ 1236] Overall Loss 0.161063 Objective Loss 0.161063 LR 0.000031 Time 0.023679 +2023-10-05 22:13:59,922 - Epoch: [195][ 240/ 1236] Overall Loss 0.160493 Objective Loss 0.160493 LR 0.000031 Time 0.023549 +2023-10-05 22:14:00,129 - Epoch: [195][ 250/ 1236] Overall Loss 0.161089 Objective Loss 0.161089 LR 0.000031 Time 0.023428 +2023-10-05 22:14:00,335 - Epoch: [195][ 260/ 1236] Overall Loss 0.160599 Objective Loss 0.160599 LR 0.000031 Time 0.023317 +2023-10-05 22:14:00,541 - Epoch: [195][ 270/ 1236] Overall Loss 0.159769 Objective Loss 0.159769 LR 0.000031 Time 0.023216 +2023-10-05 22:14:00,747 - Epoch: [195][ 280/ 1236] Overall Loss 0.160707 Objective Loss 0.160707 LR 0.000031 Time 0.023122 +2023-10-05 22:14:00,954 - Epoch: [195][ 290/ 1236] Overall Loss 0.160656 Objective Loss 0.160656 LR 0.000031 Time 0.023037 +2023-10-05 22:14:01,160 - Epoch: [195][ 300/ 1236] Overall Loss 0.160902 Objective Loss 0.160902 LR 0.000031 Time 0.022952 +2023-10-05 22:14:01,367 - Epoch: [195][ 310/ 1236] Overall Loss 0.160583 Objective Loss 0.160583 LR 0.000031 Time 0.022879 +2023-10-05 22:14:01,572 - Epoch: [195][ 320/ 1236] Overall Loss 0.160313 Objective Loss 0.160313 LR 0.000031 Time 0.022805 +2023-10-05 22:14:01,779 - Epoch: [195][ 330/ 1236] Overall Loss 0.159797 Objective Loss 0.159797 LR 0.000031 Time 0.022739 +2023-10-05 22:14:01,985 - Epoch: [195][ 340/ 1236] Overall Loss 0.160105 Objective Loss 0.160105 LR 0.000031 Time 0.022675 +2023-10-05 22:14:02,192 - Epoch: [195][ 350/ 1236] Overall Loss 0.160181 Objective Loss 0.160181 LR 0.000031 Time 0.022617 +2023-10-05 22:14:02,398 - Epoch: [195][ 360/ 1236] Overall Loss 0.159537 Objective Loss 0.159537 LR 0.000031 Time 0.022560 +2023-10-05 22:14:02,605 - Epoch: [195][ 370/ 1236] Overall Loss 0.159130 Objective Loss 0.159130 LR 0.000031 Time 0.022509 +2023-10-05 22:14:02,808 - Epoch: [195][ 380/ 1236] Overall Loss 0.159177 Objective Loss 0.159177 LR 0.000031 Time 0.022450 +2023-10-05 22:14:03,013 - Epoch: [195][ 390/ 1236] Overall Loss 0.158866 Objective Loss 0.158866 LR 0.000031 Time 0.022399 +2023-10-05 22:14:03,216 - Epoch: [195][ 400/ 1236] Overall Loss 0.158635 Objective Loss 0.158635 LR 0.000031 Time 0.022346 +2023-10-05 22:14:03,421 - Epoch: [195][ 410/ 1236] Overall Loss 0.158493 Objective Loss 0.158493 LR 0.000031 Time 0.022301 +2023-10-05 22:14:03,624 - Epoch: [195][ 420/ 1236] Overall Loss 0.158377 Objective Loss 0.158377 LR 0.000031 Time 0.022253 +2023-10-05 22:14:03,829 - Epoch: [195][ 430/ 1236] Overall Loss 0.159042 Objective Loss 0.159042 LR 0.000031 Time 0.022212 +2023-10-05 22:14:04,033 - Epoch: [195][ 440/ 1236] Overall Loss 0.158977 Objective Loss 0.158977 LR 0.000031 Time 0.022168 +2023-10-05 22:14:04,238 - Epoch: [195][ 450/ 1236] Overall Loss 0.159157 Objective Loss 0.159157 LR 0.000031 Time 0.022130 +2023-10-05 22:14:04,441 - Epoch: [195][ 460/ 1236] Overall Loss 0.159171 Objective Loss 0.159171 LR 0.000031 Time 0.022090 +2023-10-05 22:14:04,646 - Epoch: [195][ 470/ 1236] Overall Loss 0.158998 Objective Loss 0.158998 LR 0.000031 Time 0.022056 +2023-10-05 22:14:04,849 - Epoch: [195][ 480/ 1236] Overall Loss 0.159282 Objective Loss 0.159282 LR 0.000031 Time 0.022018 +2023-10-05 22:14:05,054 - Epoch: [195][ 490/ 1236] Overall Loss 0.159455 Objective Loss 0.159455 LR 0.000031 Time 0.021987 +2023-10-05 22:14:05,257 - Epoch: [195][ 500/ 1236] Overall Loss 0.159601 Objective Loss 0.159601 LR 0.000031 Time 0.021953 +2023-10-05 22:14:05,462 - Epoch: [195][ 510/ 1236] Overall Loss 0.160035 Objective Loss 0.160035 LR 0.000031 Time 0.021924 +2023-10-05 22:14:05,666 - Epoch: [195][ 520/ 1236] Overall Loss 0.160102 Objective Loss 0.160102 LR 0.000031 Time 0.021892 +2023-10-05 22:14:05,871 - Epoch: [195][ 530/ 1236] Overall Loss 0.159908 Objective Loss 0.159908 LR 0.000031 Time 0.021866 +2023-10-05 22:14:06,084 - Epoch: [195][ 540/ 1236] Overall Loss 0.160077 Objective Loss 0.160077 LR 0.000031 Time 0.021855 +2023-10-05 22:14:06,296 - Epoch: [195][ 550/ 1236] Overall Loss 0.159902 Objective Loss 0.159902 LR 0.000031 Time 0.021843 +2023-10-05 22:14:06,509 - Epoch: [195][ 560/ 1236] Overall Loss 0.159755 Objective Loss 0.159755 LR 0.000031 Time 0.021833 +2023-10-05 22:14:06,721 - Epoch: [195][ 570/ 1236] Overall Loss 0.159894 Objective Loss 0.159894 LR 0.000031 Time 0.021821 +2023-10-05 22:14:06,934 - Epoch: [195][ 580/ 1236] Overall Loss 0.159587 Objective Loss 0.159587 LR 0.000031 Time 0.021812 +2023-10-05 22:14:07,146 - Epoch: [195][ 590/ 1236] Overall Loss 0.159301 Objective Loss 0.159301 LR 0.000031 Time 0.021801 +2023-10-05 22:14:07,359 - Epoch: [195][ 600/ 1236] Overall Loss 0.159296 Objective Loss 0.159296 LR 0.000031 Time 0.021792 +2023-10-05 22:14:07,571 - Epoch: [195][ 610/ 1236] Overall Loss 0.159671 Objective Loss 0.159671 LR 0.000031 Time 0.021782 +2023-10-05 22:14:07,784 - Epoch: [195][ 620/ 1236] Overall Loss 0.159489 Objective Loss 0.159489 LR 0.000031 Time 0.021774 +2023-10-05 22:14:07,996 - Epoch: [195][ 630/ 1236] Overall Loss 0.159189 Objective Loss 0.159189 LR 0.000031 Time 0.021764 +2023-10-05 22:14:08,209 - Epoch: [195][ 640/ 1236] Overall Loss 0.159225 Objective Loss 0.159225 LR 0.000031 Time 0.021756 +2023-10-05 22:14:08,421 - Epoch: [195][ 650/ 1236] Overall Loss 0.159008 Objective Loss 0.159008 LR 0.000031 Time 0.021747 +2023-10-05 22:14:08,635 - Epoch: [195][ 660/ 1236] Overall Loss 0.158707 Objective Loss 0.158707 LR 0.000031 Time 0.021740 +2023-10-05 22:14:08,846 - Epoch: [195][ 670/ 1236] Overall Loss 0.158775 Objective Loss 0.158775 LR 0.000031 Time 0.021732 +2023-10-05 22:14:09,059 - Epoch: [195][ 680/ 1236] Overall Loss 0.159013 Objective Loss 0.159013 LR 0.000031 Time 0.021725 +2023-10-05 22:14:09,271 - Epoch: [195][ 690/ 1236] Overall Loss 0.158804 Objective Loss 0.158804 LR 0.000031 Time 0.021717 +2023-10-05 22:14:09,484 - Epoch: [195][ 700/ 1236] Overall Loss 0.158984 Objective Loss 0.158984 LR 0.000031 Time 0.021710 +2023-10-05 22:14:09,696 - Epoch: [195][ 710/ 1236] Overall Loss 0.159079 Objective Loss 0.159079 LR 0.000031 Time 0.021702 +2023-10-05 22:14:09,910 - Epoch: [195][ 720/ 1236] Overall Loss 0.159190 Objective Loss 0.159190 LR 0.000031 Time 0.021697 +2023-10-05 22:14:10,122 - Epoch: [195][ 730/ 1236] Overall Loss 0.159102 Objective Loss 0.159102 LR 0.000031 Time 0.021690 +2023-10-05 22:14:10,335 - Epoch: [195][ 740/ 1236] Overall Loss 0.159329 Objective Loss 0.159329 LR 0.000031 Time 0.021685 +2023-10-05 22:14:10,547 - Epoch: [195][ 750/ 1236] Overall Loss 0.159205 Objective Loss 0.159205 LR 0.000031 Time 0.021678 +2023-10-05 22:14:10,760 - Epoch: [195][ 760/ 1236] Overall Loss 0.159149 Objective Loss 0.159149 LR 0.000031 Time 0.021673 +2023-10-05 22:14:10,972 - Epoch: [195][ 770/ 1236] Overall Loss 0.159104 Objective Loss 0.159104 LR 0.000031 Time 0.021666 +2023-10-05 22:14:11,185 - Epoch: [195][ 780/ 1236] Overall Loss 0.159017 Objective Loss 0.159017 LR 0.000031 Time 0.021661 +2023-10-05 22:14:11,397 - Epoch: [195][ 790/ 1236] Overall Loss 0.159211 Objective Loss 0.159211 LR 0.000031 Time 0.021655 +2023-10-05 22:14:11,611 - Epoch: [195][ 800/ 1236] Overall Loss 0.159128 Objective Loss 0.159128 LR 0.000031 Time 0.021650 +2023-10-05 22:14:11,823 - Epoch: [195][ 810/ 1236] Overall Loss 0.159295 Objective Loss 0.159295 LR 0.000031 Time 0.021645 +2023-10-05 22:14:12,036 - Epoch: [195][ 820/ 1236] Overall Loss 0.159310 Objective Loss 0.159310 LR 0.000031 Time 0.021640 +2023-10-05 22:14:12,247 - Epoch: [195][ 830/ 1236] Overall Loss 0.159141 Objective Loss 0.159141 LR 0.000031 Time 0.021634 +2023-10-05 22:14:12,460 - Epoch: [195][ 840/ 1236] Overall Loss 0.159109 Objective Loss 0.159109 LR 0.000031 Time 0.021630 +2023-10-05 22:14:12,672 - Epoch: [195][ 850/ 1236] Overall Loss 0.159087 Objective Loss 0.159087 LR 0.000031 Time 0.021624 +2023-10-05 22:14:12,885 - Epoch: [195][ 860/ 1236] Overall Loss 0.159157 Objective Loss 0.159157 LR 0.000031 Time 0.021620 +2023-10-05 22:14:13,098 - Epoch: [195][ 870/ 1236] Overall Loss 0.159110 Objective Loss 0.159110 LR 0.000031 Time 0.021615 +2023-10-05 22:14:13,311 - Epoch: [195][ 880/ 1236] Overall Loss 0.159120 Objective Loss 0.159120 LR 0.000031 Time 0.021611 +2023-10-05 22:14:13,523 - Epoch: [195][ 890/ 1236] Overall Loss 0.158944 Objective Loss 0.158944 LR 0.000031 Time 0.021606 +2023-10-05 22:14:13,735 - Epoch: [195][ 900/ 1236] Overall Loss 0.159067 Objective Loss 0.159067 LR 0.000031 Time 0.021602 +2023-10-05 22:14:13,947 - Epoch: [195][ 910/ 1236] Overall Loss 0.159025 Objective Loss 0.159025 LR 0.000031 Time 0.021597 +2023-10-05 22:14:14,160 - Epoch: [195][ 920/ 1236] Overall Loss 0.158894 Objective Loss 0.158894 LR 0.000031 Time 0.021594 +2023-10-05 22:14:14,373 - Epoch: [195][ 930/ 1236] Overall Loss 0.158669 Objective Loss 0.158669 LR 0.000031 Time 0.021590 +2023-10-05 22:14:14,585 - Epoch: [195][ 940/ 1236] Overall Loss 0.158608 Objective Loss 0.158608 LR 0.000031 Time 0.021586 +2023-10-05 22:14:14,797 - Epoch: [195][ 950/ 1236] Overall Loss 0.158494 Objective Loss 0.158494 LR 0.000031 Time 0.021581 +2023-10-05 22:14:15,011 - Epoch: [195][ 960/ 1236] Overall Loss 0.158823 Objective Loss 0.158823 LR 0.000031 Time 0.021578 +2023-10-05 22:14:15,222 - Epoch: [195][ 970/ 1236] Overall Loss 0.158878 Objective Loss 0.158878 LR 0.000031 Time 0.021574 +2023-10-05 22:14:15,436 - Epoch: [195][ 980/ 1236] Overall Loss 0.159043 Objective Loss 0.159043 LR 0.000031 Time 0.021571 +2023-10-05 22:14:15,648 - Epoch: [195][ 990/ 1236] Overall Loss 0.158955 Objective Loss 0.158955 LR 0.000031 Time 0.021567 +2023-10-05 22:14:15,860 - Epoch: [195][ 1000/ 1236] Overall Loss 0.159066 Objective Loss 0.159066 LR 0.000031 Time 0.021564 +2023-10-05 22:14:16,073 - Epoch: [195][ 1010/ 1236] Overall Loss 0.158881 Objective Loss 0.158881 LR 0.000031 Time 0.021560 +2023-10-05 22:14:16,286 - Epoch: [195][ 1020/ 1236] Overall Loss 0.159082 Objective Loss 0.159082 LR 0.000031 Time 0.021558 +2023-10-05 22:14:16,497 - Epoch: [195][ 1030/ 1236] Overall Loss 0.158966 Objective Loss 0.158966 LR 0.000031 Time 0.021553 +2023-10-05 22:14:16,710 - Epoch: [195][ 1040/ 1236] Overall Loss 0.158997 Objective Loss 0.158997 LR 0.000031 Time 0.021551 +2023-10-05 22:14:16,922 - Epoch: [195][ 1050/ 1236] Overall Loss 0.159143 Objective Loss 0.159143 LR 0.000031 Time 0.021547 +2023-10-05 22:14:17,136 - Epoch: [195][ 1060/ 1236] Overall Loss 0.159005 Objective Loss 0.159005 LR 0.000031 Time 0.021545 +2023-10-05 22:14:17,348 - Epoch: [195][ 1070/ 1236] Overall Loss 0.159075 Objective Loss 0.159075 LR 0.000031 Time 0.021541 +2023-10-05 22:14:17,561 - Epoch: [195][ 1080/ 1236] Overall Loss 0.159078 Objective Loss 0.159078 LR 0.000031 Time 0.021539 +2023-10-05 22:14:17,773 - Epoch: [195][ 1090/ 1236] Overall Loss 0.159203 Objective Loss 0.159203 LR 0.000031 Time 0.021536 +2023-10-05 22:14:17,987 - Epoch: [195][ 1100/ 1236] Overall Loss 0.159129 Objective Loss 0.159129 LR 0.000031 Time 0.021534 +2023-10-05 22:14:18,199 - Epoch: [195][ 1110/ 1236] Overall Loss 0.159416 Objective Loss 0.159416 LR 0.000031 Time 0.021530 +2023-10-05 22:14:18,411 - Epoch: [195][ 1120/ 1236] Overall Loss 0.159348 Objective Loss 0.159348 LR 0.000031 Time 0.021528 +2023-10-05 22:14:18,623 - Epoch: [195][ 1130/ 1236] Overall Loss 0.159323 Objective Loss 0.159323 LR 0.000031 Time 0.021525 +2023-10-05 22:14:18,837 - Epoch: [195][ 1140/ 1236] Overall Loss 0.159329 Objective Loss 0.159329 LR 0.000031 Time 0.021523 +2023-10-05 22:14:19,049 - Epoch: [195][ 1150/ 1236] Overall Loss 0.159210 Objective Loss 0.159210 LR 0.000031 Time 0.021520 +2023-10-05 22:14:19,262 - Epoch: [195][ 1160/ 1236] Overall Loss 0.159305 Objective Loss 0.159305 LR 0.000031 Time 0.021518 +2023-10-05 22:14:19,474 - Epoch: [195][ 1170/ 1236] Overall Loss 0.159236 Objective Loss 0.159236 LR 0.000031 Time 0.021515 +2023-10-05 22:14:19,688 - Epoch: [195][ 1180/ 1236] Overall Loss 0.159314 Objective Loss 0.159314 LR 0.000031 Time 0.021513 +2023-10-05 22:14:19,899 - Epoch: [195][ 1190/ 1236] Overall Loss 0.159121 Objective Loss 0.159121 LR 0.000031 Time 0.021510 +2023-10-05 22:14:20,113 - Epoch: [195][ 1200/ 1236] Overall Loss 0.158951 Objective Loss 0.158951 LR 0.000031 Time 0.021508 +2023-10-05 22:14:20,324 - Epoch: [195][ 1210/ 1236] Overall Loss 0.159101 Objective Loss 0.159101 LR 0.000031 Time 0.021505 +2023-10-05 22:14:20,537 - Epoch: [195][ 1220/ 1236] Overall Loss 0.158993 Objective Loss 0.158993 LR 0.000031 Time 0.021504 +2023-10-05 22:14:20,800 - Epoch: [195][ 1230/ 1236] Overall Loss 0.159036 Objective Loss 0.159036 LR 0.000031 Time 0.021542 +2023-10-05 22:14:20,917 - Epoch: [195][ 1236/ 1236] Overall Loss 0.159119 Objective Loss 0.159119 Top1 88.594705 Top5 98.574338 LR 0.000031 Time 0.021532 +2023-10-05 22:14:21,054 - --- validate (epoch=195)----------- +2023-10-05 22:14:21,054 - 29943 samples (256 per mini-batch) +2023-10-05 22:14:21,504 - Epoch: [195][ 10/ 117] Loss 0.324528 Top1 85.195312 Top5 98.046875 +2023-10-05 22:14:21,651 - Epoch: [195][ 20/ 117] Loss 0.324122 Top1 85.527344 Top5 97.949219 +2023-10-05 22:14:21,797 - Epoch: [195][ 30/ 117] Loss 0.316365 Top1 85.820312 Top5 98.059896 +2023-10-05 22:14:21,945 - Epoch: [195][ 40/ 117] Loss 0.302087 Top1 86.201172 Top5 98.134766 +2023-10-05 22:14:22,092 - Epoch: [195][ 50/ 117] Loss 0.305594 Top1 86.320312 Top5 98.203125 +2023-10-05 22:14:22,239 - Epoch: [195][ 60/ 117] Loss 0.303079 Top1 86.386719 Top5 98.242188 +2023-10-05 22:14:22,386 - Epoch: [195][ 70/ 117] Loss 0.303043 Top1 86.339286 Top5 98.219866 +2023-10-05 22:14:22,532 - Epoch: [195][ 80/ 117] Loss 0.303123 Top1 86.191406 Top5 98.232422 +2023-10-05 22:14:22,678 - Epoch: [195][ 90/ 117] Loss 0.304492 Top1 86.111111 Top5 98.233507 +2023-10-05 22:14:22,825 - Epoch: [195][ 100/ 117] Loss 0.303411 Top1 86.109375 Top5 98.265625 +2023-10-05 22:14:22,978 - Epoch: [195][ 110/ 117] Loss 0.301526 Top1 86.164773 Top5 98.288352 +2023-10-05 22:14:23,063 - Epoch: [195][ 117/ 117] Loss 0.300161 Top1 86.180409 Top5 98.286745 +2023-10-05 22:14:23,176 - ==> Top1: 86.180 Top5: 98.287 Loss: 0.300 + +2023-10-05 22:14:23,177 - ==> Confusion: +[[ 946 3 3 3 3 2 0 0 5 58 1 1 2 2 4 1 3 0 0 0 13] + [ 0 1075 1 0 4 13 1 16 0 0 1 3 0 0 1 2 2 0 6 1 5] + [ 7 2 973 10 1 0 17 6 0 0 4 3 6 1 0 3 1 3 8 4 7] + [ 1 1 11 971 2 4 0 2 0 1 4 1 6 1 27 3 0 6 29 2 17] + [ 26 7 0 0 969 4 1 2 1 9 0 1 0 1 7 3 11 2 0 1 5] + [ 3 39 1 0 4 993 0 18 0 0 6 9 0 11 6 2 3 0 2 3 16] + [ 0 6 17 0 0 0 1132 10 0 0 1 2 1 0 1 6 0 1 2 6 6] + [ 5 17 12 0 1 28 1 1084 2 2 3 9 1 2 0 2 0 0 37 4 8] + [ 17 1 2 0 0 2 0 2 977 41 10 2 1 9 12 2 2 1 5 0 3] + [ 99 1 3 1 4 2 0 1 16 956 1 2 0 15 5 1 0 0 1 2 9] + [ 3 6 9 2 0 2 4 5 13 1 973 2 0 10 4 2 2 0 6 1 8] + [ 1 0 2 0 0 11 0 3 0 1 0 970 12 6 0 3 1 15 0 5 5] + [ 2 2 0 3 0 3 0 2 0 0 2 35 992 0 0 4 3 9 1 4 6] + [ 2 0 1 0 1 6 0 1 10 21 9 5 3 1042 4 2 2 0 0 2 8] + [ 14 2 2 6 4 0 0 0 24 2 1 1 2 3 1013 0 3 2 9 0 13] + [ 1 4 3 0 2 0 1 0 0 0 0 9 8 1 1 1070 15 8 0 8 3] + [ 1 14 1 0 6 2 0 1 1 0 0 4 0 1 3 9 1102 0 0 3 13] + [ 0 0 1 0 0 0 3 0 0 0 0 2 14 3 0 3 0 1006 1 0 5] + [ 1 6 3 16 1 0 0 23 1 0 3 0 1 0 9 0 1 0 993 1 9] + [ 1 2 4 2 2 6 6 10 1 0 1 12 3 1 0 6 9 2 3 1074 7] + [ 137 172 115 56 84 112 30 85 95 67 169 92 288 261 122 45 133 62 143 143 5494]] + +2023-10-05 22:14:23,178 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:14:23,178 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:14:23,184 - + +2023-10-05 22:14:23,184 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:14:24,165 - Epoch: [196][ 10/ 1236] Overall Loss 0.140419 Objective Loss 0.140419 LR 0.000031 Time 0.098022 +2023-10-05 22:14:24,366 - Epoch: [196][ 20/ 1236] Overall Loss 0.157225 Objective Loss 0.157225 LR 0.000031 Time 0.059079 +2023-10-05 22:14:24,569 - Epoch: [196][ 30/ 1236] Overall Loss 0.161503 Objective Loss 0.161503 LR 0.000031 Time 0.046126 +2023-10-05 22:14:24,770 - Epoch: [196][ 40/ 1236] Overall Loss 0.156770 Objective Loss 0.156770 LR 0.000031 Time 0.039608 +2023-10-05 22:14:24,972 - Epoch: [196][ 50/ 1236] Overall Loss 0.156589 Objective Loss 0.156589 LR 0.000031 Time 0.035728 +2023-10-05 22:14:25,174 - Epoch: [196][ 60/ 1236] Overall Loss 0.158254 Objective Loss 0.158254 LR 0.000031 Time 0.033125 +2023-10-05 22:14:25,376 - Epoch: [196][ 70/ 1236] Overall Loss 0.156094 Objective Loss 0.156094 LR 0.000031 Time 0.031279 +2023-10-05 22:14:25,578 - Epoch: [196][ 80/ 1236] Overall Loss 0.154715 Objective Loss 0.154715 LR 0.000031 Time 0.029885 +2023-10-05 22:14:25,780 - Epoch: [196][ 90/ 1236] Overall Loss 0.154768 Objective Loss 0.154768 LR 0.000031 Time 0.028810 +2023-10-05 22:14:25,982 - Epoch: [196][ 100/ 1236] Overall Loss 0.155498 Objective Loss 0.155498 LR 0.000031 Time 0.027939 +2023-10-05 22:14:26,185 - Epoch: [196][ 110/ 1236] Overall Loss 0.154405 Objective Loss 0.154405 LR 0.000031 Time 0.027245 +2023-10-05 22:14:26,387 - Epoch: [196][ 120/ 1236] Overall Loss 0.154866 Objective Loss 0.154866 LR 0.000031 Time 0.026654 +2023-10-05 22:14:26,591 - Epoch: [196][ 130/ 1236] Overall Loss 0.156001 Objective Loss 0.156001 LR 0.000031 Time 0.026173 +2023-10-05 22:14:26,790 - Epoch: [196][ 140/ 1236] Overall Loss 0.157164 Objective Loss 0.157164 LR 0.000031 Time 0.025719 +2023-10-05 22:14:26,989 - Epoch: [196][ 150/ 1236] Overall Loss 0.157631 Objective Loss 0.157631 LR 0.000031 Time 0.025331 +2023-10-05 22:14:27,190 - Epoch: [196][ 160/ 1236] Overall Loss 0.157496 Objective Loss 0.157496 LR 0.000031 Time 0.024999 +2023-10-05 22:14:27,390 - Epoch: [196][ 170/ 1236] Overall Loss 0.158236 Objective Loss 0.158236 LR 0.000031 Time 0.024707 +2023-10-05 22:14:27,591 - Epoch: [196][ 180/ 1236] Overall Loss 0.158872 Objective Loss 0.158872 LR 0.000031 Time 0.024445 +2023-10-05 22:14:27,791 - Epoch: [196][ 190/ 1236] Overall Loss 0.158420 Objective Loss 0.158420 LR 0.000031 Time 0.024213 +2023-10-05 22:14:27,992 - Epoch: [196][ 200/ 1236] Overall Loss 0.157919 Objective Loss 0.157919 LR 0.000031 Time 0.024004 +2023-10-05 22:14:28,193 - Epoch: [196][ 210/ 1236] Overall Loss 0.157465 Objective Loss 0.157465 LR 0.000031 Time 0.023818 +2023-10-05 22:14:28,394 - Epoch: [196][ 220/ 1236] Overall Loss 0.157650 Objective Loss 0.157650 LR 0.000031 Time 0.023644 +2023-10-05 22:14:28,598 - Epoch: [196][ 230/ 1236] Overall Loss 0.158373 Objective Loss 0.158373 LR 0.000031 Time 0.023506 +2023-10-05 22:14:28,800 - Epoch: [196][ 240/ 1236] Overall Loss 0.158202 Objective Loss 0.158202 LR 0.000031 Time 0.023365 +2023-10-05 22:14:29,006 - Epoch: [196][ 250/ 1236] Overall Loss 0.157954 Objective Loss 0.157954 LR 0.000031 Time 0.023251 +2023-10-05 22:14:29,211 - Epoch: [196][ 260/ 1236] Overall Loss 0.157455 Objective Loss 0.157455 LR 0.000031 Time 0.023144 +2023-10-05 22:14:29,417 - Epoch: [196][ 270/ 1236] Overall Loss 0.159134 Objective Loss 0.159134 LR 0.000031 Time 0.023048 +2023-10-05 22:14:29,622 - Epoch: [196][ 280/ 1236] Overall Loss 0.158948 Objective Loss 0.158948 LR 0.000031 Time 0.022950 +2023-10-05 22:14:29,828 - Epoch: [196][ 290/ 1236] Overall Loss 0.158988 Objective Loss 0.158988 LR 0.000031 Time 0.022866 +2023-10-05 22:14:30,033 - Epoch: [196][ 300/ 1236] Overall Loss 0.159307 Objective Loss 0.159307 LR 0.000031 Time 0.022787 +2023-10-05 22:14:30,239 - Epoch: [196][ 310/ 1236] Overall Loss 0.159067 Objective Loss 0.159067 LR 0.000031 Time 0.022715 +2023-10-05 22:14:30,445 - Epoch: [196][ 320/ 1236] Overall Loss 0.159035 Objective Loss 0.159035 LR 0.000031 Time 0.022645 +2023-10-05 22:14:30,651 - Epoch: [196][ 330/ 1236] Overall Loss 0.158572 Objective Loss 0.158572 LR 0.000031 Time 0.022581 +2023-10-05 22:14:30,854 - Epoch: [196][ 340/ 1236] Overall Loss 0.158042 Objective Loss 0.158042 LR 0.000031 Time 0.022510 +2023-10-05 22:14:31,055 - Epoch: [196][ 350/ 1236] Overall Loss 0.158066 Objective Loss 0.158066 LR 0.000031 Time 0.022439 +2023-10-05 22:14:31,255 - Epoch: [196][ 360/ 1236] Overall Loss 0.158072 Objective Loss 0.158072 LR 0.000031 Time 0.022372 +2023-10-05 22:14:31,459 - Epoch: [196][ 370/ 1236] Overall Loss 0.158402 Objective Loss 0.158402 LR 0.000031 Time 0.022317 +2023-10-05 22:14:31,663 - Epoch: [196][ 380/ 1236] Overall Loss 0.158130 Objective Loss 0.158130 LR 0.000031 Time 0.022264 +2023-10-05 22:14:31,867 - Epoch: [196][ 390/ 1236] Overall Loss 0.158016 Objective Loss 0.158016 LR 0.000031 Time 0.022216 +2023-10-05 22:14:32,070 - Epoch: [196][ 400/ 1236] Overall Loss 0.158068 Objective Loss 0.158068 LR 0.000031 Time 0.022167 +2023-10-05 22:14:32,272 - Epoch: [196][ 410/ 1236] Overall Loss 0.158581 Objective Loss 0.158581 LR 0.000031 Time 0.022119 +2023-10-05 22:14:32,484 - Epoch: [196][ 420/ 1236] Overall Loss 0.158706 Objective Loss 0.158706 LR 0.000031 Time 0.022095 +2023-10-05 22:14:32,691 - Epoch: [196][ 430/ 1236] Overall Loss 0.159468 Objective Loss 0.159468 LR 0.000031 Time 0.022061 +2023-10-05 22:14:32,902 - Epoch: [196][ 440/ 1236] Overall Loss 0.159130 Objective Loss 0.159130 LR 0.000031 Time 0.022040 +2023-10-05 22:14:33,109 - Epoch: [196][ 450/ 1236] Overall Loss 0.159299 Objective Loss 0.159299 LR 0.000031 Time 0.022009 +2023-10-05 22:14:33,321 - Epoch: [196][ 460/ 1236] Overall Loss 0.159200 Objective Loss 0.159200 LR 0.000031 Time 0.021990 +2023-10-05 22:14:33,528 - Epoch: [196][ 470/ 1236] Overall Loss 0.159570 Objective Loss 0.159570 LR 0.000031 Time 0.021963 +2023-10-05 22:14:33,740 - Epoch: [196][ 480/ 1236] Overall Loss 0.159644 Objective Loss 0.159644 LR 0.000031 Time 0.021946 +2023-10-05 22:14:33,947 - Epoch: [196][ 490/ 1236] Overall Loss 0.159836 Objective Loss 0.159836 LR 0.000031 Time 0.021921 +2023-10-05 22:14:34,159 - Epoch: [196][ 500/ 1236] Overall Loss 0.159801 Objective Loss 0.159801 LR 0.000031 Time 0.021905 +2023-10-05 22:14:34,366 - Epoch: [196][ 510/ 1236] Overall Loss 0.159552 Objective Loss 0.159552 LR 0.000031 Time 0.021881 +2023-10-05 22:14:34,578 - Epoch: [196][ 520/ 1236] Overall Loss 0.159312 Objective Loss 0.159312 LR 0.000031 Time 0.021867 +2023-10-05 22:14:34,785 - Epoch: [196][ 530/ 1236] Overall Loss 0.159512 Objective Loss 0.159512 LR 0.000031 Time 0.021844 +2023-10-05 22:14:34,994 - Epoch: [196][ 540/ 1236] Overall Loss 0.159209 Objective Loss 0.159209 LR 0.000031 Time 0.021825 +2023-10-05 22:14:35,201 - Epoch: [196][ 550/ 1236] Overall Loss 0.159488 Objective Loss 0.159488 LR 0.000031 Time 0.021804 +2023-10-05 22:14:35,413 - Epoch: [196][ 560/ 1236] Overall Loss 0.159200 Objective Loss 0.159200 LR 0.000031 Time 0.021793 +2023-10-05 22:14:35,620 - Epoch: [196][ 570/ 1236] Overall Loss 0.159250 Objective Loss 0.159250 LR 0.000031 Time 0.021773 +2023-10-05 22:14:35,832 - Epoch: [196][ 580/ 1236] Overall Loss 0.159031 Objective Loss 0.159031 LR 0.000031 Time 0.021763 +2023-10-05 22:14:36,039 - Epoch: [196][ 590/ 1236] Overall Loss 0.159396 Objective Loss 0.159396 LR 0.000031 Time 0.021745 +2023-10-05 22:14:36,251 - Epoch: [196][ 600/ 1236] Overall Loss 0.159459 Objective Loss 0.159459 LR 0.000031 Time 0.021735 +2023-10-05 22:14:36,458 - Epoch: [196][ 610/ 1236] Overall Loss 0.159161 Objective Loss 0.159161 LR 0.000031 Time 0.021717 +2023-10-05 22:14:36,670 - Epoch: [196][ 620/ 1236] Overall Loss 0.159401 Objective Loss 0.159401 LR 0.000031 Time 0.021708 +2023-10-05 22:14:36,877 - Epoch: [196][ 630/ 1236] Overall Loss 0.159251 Objective Loss 0.159251 LR 0.000031 Time 0.021692 +2023-10-05 22:14:37,089 - Epoch: [196][ 640/ 1236] Overall Loss 0.158978 Objective Loss 0.158978 LR 0.000031 Time 0.021684 +2023-10-05 22:14:37,297 - Epoch: [196][ 650/ 1236] Overall Loss 0.159195 Objective Loss 0.159195 LR 0.000031 Time 0.021669 +2023-10-05 22:14:37,508 - Epoch: [196][ 660/ 1236] Overall Loss 0.159393 Objective Loss 0.159393 LR 0.000031 Time 0.021661 +2023-10-05 22:14:37,716 - Epoch: [196][ 670/ 1236] Overall Loss 0.159301 Objective Loss 0.159301 LR 0.000031 Time 0.021646 +2023-10-05 22:14:37,927 - Epoch: [196][ 680/ 1236] Overall Loss 0.159750 Objective Loss 0.159750 LR 0.000031 Time 0.021639 +2023-10-05 22:14:38,135 - Epoch: [196][ 690/ 1236] Overall Loss 0.159767 Objective Loss 0.159767 LR 0.000031 Time 0.021625 +2023-10-05 22:14:38,346 - Epoch: [196][ 700/ 1236] Overall Loss 0.159670 Objective Loss 0.159670 LR 0.000031 Time 0.021618 +2023-10-05 22:14:38,553 - Epoch: [196][ 710/ 1236] Overall Loss 0.159817 Objective Loss 0.159817 LR 0.000031 Time 0.021604 +2023-10-05 22:14:38,765 - Epoch: [196][ 720/ 1236] Overall Loss 0.159893 Objective Loss 0.159893 LR 0.000031 Time 0.021598 +2023-10-05 22:14:38,972 - Epoch: [196][ 730/ 1236] Overall Loss 0.160105 Objective Loss 0.160105 LR 0.000031 Time 0.021586 +2023-10-05 22:14:39,184 - Epoch: [196][ 740/ 1236] Overall Loss 0.160085 Objective Loss 0.160085 LR 0.000031 Time 0.021580 +2023-10-05 22:14:39,392 - Epoch: [196][ 750/ 1236] Overall Loss 0.159996 Objective Loss 0.159996 LR 0.000031 Time 0.021568 +2023-10-05 22:14:39,603 - Epoch: [196][ 760/ 1236] Overall Loss 0.160211 Objective Loss 0.160211 LR 0.000031 Time 0.021562 +2023-10-05 22:14:39,810 - Epoch: [196][ 770/ 1236] Overall Loss 0.159876 Objective Loss 0.159876 LR 0.000031 Time 0.021551 +2023-10-05 22:14:40,022 - Epoch: [196][ 780/ 1236] Overall Loss 0.159830 Objective Loss 0.159830 LR 0.000031 Time 0.021545 +2023-10-05 22:14:40,229 - Epoch: [196][ 790/ 1236] Overall Loss 0.159810 Objective Loss 0.159810 LR 0.000031 Time 0.021534 +2023-10-05 22:14:40,441 - Epoch: [196][ 800/ 1236] Overall Loss 0.160040 Objective Loss 0.160040 LR 0.000031 Time 0.021530 +2023-10-05 22:14:40,648 - Epoch: [196][ 810/ 1236] Overall Loss 0.160133 Objective Loss 0.160133 LR 0.000031 Time 0.021519 +2023-10-05 22:14:40,860 - Epoch: [196][ 820/ 1236] Overall Loss 0.160009 Objective Loss 0.160009 LR 0.000031 Time 0.021515 +2023-10-05 22:14:41,068 - Epoch: [196][ 830/ 1236] Overall Loss 0.159785 Objective Loss 0.159785 LR 0.000031 Time 0.021505 +2023-10-05 22:14:41,279 - Epoch: [196][ 840/ 1236] Overall Loss 0.159611 Objective Loss 0.159611 LR 0.000031 Time 0.021501 +2023-10-05 22:14:41,487 - Epoch: [196][ 850/ 1236] Overall Loss 0.159692 Objective Loss 0.159692 LR 0.000031 Time 0.021491 +2023-10-05 22:14:41,699 - Epoch: [196][ 860/ 1236] Overall Loss 0.159805 Objective Loss 0.159805 LR 0.000031 Time 0.021488 +2023-10-05 22:14:41,906 - Epoch: [196][ 870/ 1236] Overall Loss 0.159942 Objective Loss 0.159942 LR 0.000031 Time 0.021478 +2023-10-05 22:14:42,118 - Epoch: [196][ 880/ 1236] Overall Loss 0.159970 Objective Loss 0.159970 LR 0.000031 Time 0.021475 +2023-10-05 22:14:42,325 - Epoch: [196][ 890/ 1236] Overall Loss 0.159940 Objective Loss 0.159940 LR 0.000031 Time 0.021465 +2023-10-05 22:14:42,537 - Epoch: [196][ 900/ 1236] Overall Loss 0.159799 Objective Loss 0.159799 LR 0.000031 Time 0.021462 +2023-10-05 22:14:42,744 - Epoch: [196][ 910/ 1236] Overall Loss 0.159941 Objective Loss 0.159941 LR 0.000031 Time 0.021453 +2023-10-05 22:14:42,956 - Epoch: [196][ 920/ 1236] Overall Loss 0.160055 Objective Loss 0.160055 LR 0.000031 Time 0.021451 +2023-10-05 22:14:43,164 - Epoch: [196][ 930/ 1236] Overall Loss 0.160047 Objective Loss 0.160047 LR 0.000031 Time 0.021443 +2023-10-05 22:14:43,375 - Epoch: [196][ 940/ 1236] Overall Loss 0.160119 Objective Loss 0.160119 LR 0.000031 Time 0.021440 +2023-10-05 22:14:43,582 - Epoch: [196][ 950/ 1236] Overall Loss 0.159927 Objective Loss 0.159927 LR 0.000031 Time 0.021431 +2023-10-05 22:14:43,794 - Epoch: [196][ 960/ 1236] Overall Loss 0.160025 Objective Loss 0.160025 LR 0.000031 Time 0.021428 +2023-10-05 22:14:44,001 - Epoch: [196][ 970/ 1236] Overall Loss 0.160022 Objective Loss 0.160022 LR 0.000031 Time 0.021421 +2023-10-05 22:14:44,213 - Epoch: [196][ 980/ 1236] Overall Loss 0.160066 Objective Loss 0.160066 LR 0.000031 Time 0.021418 +2023-10-05 22:14:44,420 - Epoch: [196][ 990/ 1236] Overall Loss 0.160035 Objective Loss 0.160035 LR 0.000031 Time 0.021410 +2023-10-05 22:14:44,632 - Epoch: [196][ 1000/ 1236] Overall Loss 0.160031 Objective Loss 0.160031 LR 0.000031 Time 0.021408 +2023-10-05 22:14:44,839 - Epoch: [196][ 1010/ 1236] Overall Loss 0.159933 Objective Loss 0.159933 LR 0.000031 Time 0.021401 +2023-10-05 22:14:45,051 - Epoch: [196][ 1020/ 1236] Overall Loss 0.159991 Objective Loss 0.159991 LR 0.000031 Time 0.021398 +2023-10-05 22:14:45,258 - Epoch: [196][ 1030/ 1236] Overall Loss 0.160382 Objective Loss 0.160382 LR 0.000031 Time 0.021391 +2023-10-05 22:14:45,470 - Epoch: [196][ 1040/ 1236] Overall Loss 0.160420 Objective Loss 0.160420 LR 0.000031 Time 0.021389 +2023-10-05 22:14:45,677 - Epoch: [196][ 1050/ 1236] Overall Loss 0.160252 Objective Loss 0.160252 LR 0.000031 Time 0.021382 +2023-10-05 22:14:45,889 - Epoch: [196][ 1060/ 1236] Overall Loss 0.160109 Objective Loss 0.160109 LR 0.000031 Time 0.021380 +2023-10-05 22:14:46,097 - Epoch: [196][ 1070/ 1236] Overall Loss 0.159965 Objective Loss 0.159965 LR 0.000031 Time 0.021374 +2023-10-05 22:14:46,309 - Epoch: [196][ 1080/ 1236] Overall Loss 0.159923 Objective Loss 0.159923 LR 0.000031 Time 0.021372 +2023-10-05 22:14:46,516 - Epoch: [196][ 1090/ 1236] Overall Loss 0.159885 Objective Loss 0.159885 LR 0.000031 Time 0.021365 +2023-10-05 22:14:46,728 - Epoch: [196][ 1100/ 1236] Overall Loss 0.159954 Objective Loss 0.159954 LR 0.000031 Time 0.021364 +2023-10-05 22:14:46,935 - Epoch: [196][ 1110/ 1236] Overall Loss 0.159752 Objective Loss 0.159752 LR 0.000031 Time 0.021358 +2023-10-05 22:14:47,147 - Epoch: [196][ 1120/ 1236] Overall Loss 0.159776 Objective Loss 0.159776 LR 0.000031 Time 0.021356 +2023-10-05 22:14:47,355 - Epoch: [196][ 1130/ 1236] Overall Loss 0.159632 Objective Loss 0.159632 LR 0.000031 Time 0.021350 +2023-10-05 22:14:47,567 - Epoch: [196][ 1140/ 1236] Overall Loss 0.159591 Objective Loss 0.159591 LR 0.000031 Time 0.021349 +2023-10-05 22:14:47,774 - Epoch: [196][ 1150/ 1236] Overall Loss 0.159457 Objective Loss 0.159457 LR 0.000031 Time 0.021343 +2023-10-05 22:14:47,986 - Epoch: [196][ 1160/ 1236] Overall Loss 0.159443 Objective Loss 0.159443 LR 0.000031 Time 0.021342 +2023-10-05 22:14:48,193 - Epoch: [196][ 1170/ 1236] Overall Loss 0.159567 Objective Loss 0.159567 LR 0.000031 Time 0.021336 +2023-10-05 22:14:48,405 - Epoch: [196][ 1180/ 1236] Overall Loss 0.159431 Objective Loss 0.159431 LR 0.000031 Time 0.021334 +2023-10-05 22:14:48,613 - Epoch: [196][ 1190/ 1236] Overall Loss 0.159424 Objective Loss 0.159424 LR 0.000031 Time 0.021329 +2023-10-05 22:14:48,825 - Epoch: [196][ 1200/ 1236] Overall Loss 0.159489 Objective Loss 0.159489 LR 0.000031 Time 0.021328 +2023-10-05 22:14:49,032 - Epoch: [196][ 1210/ 1236] Overall Loss 0.159613 Objective Loss 0.159613 LR 0.000031 Time 0.021323 +2023-10-05 22:14:49,244 - Epoch: [196][ 1220/ 1236] Overall Loss 0.159555 Objective Loss 0.159555 LR 0.000031 Time 0.021321 +2023-10-05 22:14:49,505 - Epoch: [196][ 1230/ 1236] Overall Loss 0.159560 Objective Loss 0.159560 LR 0.000031 Time 0.021360 +2023-10-05 22:14:49,623 - Epoch: [196][ 1236/ 1236] Overall Loss 0.159439 Objective Loss 0.159439 Top1 89.613035 Top5 99.185336 LR 0.000031 Time 0.021351 +2023-10-05 22:14:49,743 - --- validate (epoch=196)----------- +2023-10-05 22:14:49,743 - 29943 samples (256 per mini-batch) +2023-10-05 22:14:50,194 - Epoch: [196][ 10/ 117] Loss 0.296294 Top1 86.289062 Top5 97.968750 +2023-10-05 22:14:50,341 - Epoch: [196][ 20/ 117] Loss 0.301265 Top1 86.250000 Top5 98.242188 +2023-10-05 22:14:50,487 - Epoch: [196][ 30/ 117] Loss 0.297044 Top1 86.132812 Top5 98.437500 +2023-10-05 22:14:50,634 - Epoch: [196][ 40/ 117] Loss 0.295028 Top1 86.132812 Top5 98.369141 +2023-10-05 22:14:50,780 - Epoch: [196][ 50/ 117] Loss 0.292723 Top1 86.265625 Top5 98.398438 +2023-10-05 22:14:50,926 - Epoch: [196][ 60/ 117] Loss 0.296593 Top1 86.282552 Top5 98.398438 +2023-10-05 22:14:51,071 - Epoch: [196][ 70/ 117] Loss 0.298777 Top1 86.116071 Top5 98.314732 +2023-10-05 22:14:51,217 - Epoch: [196][ 80/ 117] Loss 0.301791 Top1 86.010742 Top5 98.286133 +2023-10-05 22:14:51,361 - Epoch: [196][ 90/ 117] Loss 0.299192 Top1 85.976562 Top5 98.302951 +2023-10-05 22:14:51,507 - Epoch: [196][ 100/ 117] Loss 0.295747 Top1 85.976562 Top5 98.347656 +2023-10-05 22:14:51,660 - Epoch: [196][ 110/ 117] Loss 0.296125 Top1 86.040483 Top5 98.355824 +2023-10-05 22:14:51,744 - Epoch: [196][ 117/ 117] Loss 0.297959 Top1 85.976689 Top5 98.316802 +2023-10-05 22:14:51,849 - ==> Top1: 85.977 Top5: 98.317 Loss: 0.298 + +2023-10-05 22:14:51,849 - ==> Confusion: +[[ 938 3 4 1 5 1 0 0 5 62 1 0 2 2 5 3 3 2 0 0 13] + [ 0 1067 2 0 6 16 1 19 0 0 1 3 0 1 1 2 1 0 6 1 4] + [ 3 2 978 8 3 0 16 8 0 0 5 2 7 1 0 4 0 1 8 3 7] + [ 1 1 10 983 2 4 0 1 1 1 5 1 6 1 24 3 1 6 19 2 17] + [ 19 9 0 0 976 3 1 2 1 8 0 2 0 2 6 2 10 3 1 1 4] + [ 3 34 0 1 2 991 1 19 0 0 5 11 1 14 5 1 3 0 3 3 19] + [ 0 6 25 0 0 0 1126 7 0 0 1 3 1 0 1 5 0 0 2 8 6] + [ 4 20 11 0 2 30 4 1076 3 1 2 10 2 2 0 0 0 0 35 6 10] + [ 18 1 2 0 0 4 1 1 974 42 8 2 1 11 12 2 3 1 2 0 4] + [ 91 0 4 2 3 3 0 1 20 955 1 1 1 22 1 4 0 0 0 2 8] + [ 2 5 9 4 0 1 2 4 10 2 983 1 0 11 4 1 2 0 3 2 7] + [ 1 0 2 0 0 12 0 3 0 1 0 968 13 4 0 5 1 16 0 5 4] + [ 1 2 2 4 1 3 0 1 0 0 1 32 989 2 1 4 2 11 2 3 7] + [ 2 0 1 0 0 5 0 1 10 16 9 5 3 1052 5 2 1 0 0 1 6] + [ 14 3 3 7 7 0 0 0 25 2 1 1 3 2 1010 0 1 2 7 0 13] + [ 1 2 2 0 4 0 1 0 0 0 0 8 8 1 1 1069 13 10 0 10 4] + [ 1 12 1 0 4 4 0 0 0 0 0 4 0 0 3 8 1109 0 0 3 12] + [ 0 0 1 1 1 0 4 0 0 0 0 3 16 3 0 3 0 1001 1 0 4] + [ 0 6 5 19 1 0 0 28 1 0 1 0 2 1 9 0 1 0 984 1 9] + [ 1 2 4 2 2 8 6 9 1 0 1 12 2 0 0 8 8 2 3 1073 8] + [ 132 155 129 60 85 125 30 91 82 70 185 100 304 288 117 46 138 61 124 141 5442]] + +2023-10-05 22:14:51,851 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:14:51,851 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:14:51,857 - + +2023-10-05 22:14:51,857 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:14:52,845 - Epoch: [197][ 10/ 1236] Overall Loss 0.151711 Objective Loss 0.151711 LR 0.000031 Time 0.098783 +2023-10-05 22:14:53,060 - Epoch: [197][ 20/ 1236] Overall Loss 0.155561 Objective Loss 0.155561 LR 0.000031 Time 0.060106 +2023-10-05 22:14:53,269 - Epoch: [197][ 30/ 1236] Overall Loss 0.157992 Objective Loss 0.157992 LR 0.000031 Time 0.047018 +2023-10-05 22:14:53,483 - Epoch: [197][ 40/ 1236] Overall Loss 0.158945 Objective Loss 0.158945 LR 0.000031 Time 0.040608 +2023-10-05 22:14:53,690 - Epoch: [197][ 50/ 1236] Overall Loss 0.161722 Objective Loss 0.161722 LR 0.000031 Time 0.036616 +2023-10-05 22:14:53,902 - Epoch: [197][ 60/ 1236] Overall Loss 0.162411 Objective Loss 0.162411 LR 0.000031 Time 0.034042 +2023-10-05 22:14:54,108 - Epoch: [197][ 70/ 1236] Overall Loss 0.161829 Objective Loss 0.161829 LR 0.000031 Time 0.032125 +2023-10-05 22:14:54,320 - Epoch: [197][ 80/ 1236] Overall Loss 0.163446 Objective Loss 0.163446 LR 0.000031 Time 0.030757 +2023-10-05 22:14:54,528 - Epoch: [197][ 90/ 1236] Overall Loss 0.161214 Objective Loss 0.161214 LR 0.000031 Time 0.029650 +2023-10-05 22:14:54,742 - Epoch: [197][ 100/ 1236] Overall Loss 0.160241 Objective Loss 0.160241 LR 0.000031 Time 0.028813 +2023-10-05 22:14:54,951 - Epoch: [197][ 110/ 1236] Overall Loss 0.160156 Objective Loss 0.160156 LR 0.000031 Time 0.028099 +2023-10-05 22:14:55,158 - Epoch: [197][ 120/ 1236] Overall Loss 0.160455 Objective Loss 0.160455 LR 0.000031 Time 0.027479 +2023-10-05 22:14:55,363 - Epoch: [197][ 130/ 1236] Overall Loss 0.160645 Objective Loss 0.160645 LR 0.000031 Time 0.026933 +2023-10-05 22:14:55,566 - Epoch: [197][ 140/ 1236] Overall Loss 0.162146 Objective Loss 0.162146 LR 0.000031 Time 0.026457 +2023-10-05 22:14:55,768 - Epoch: [197][ 150/ 1236] Overall Loss 0.161982 Objective Loss 0.161982 LR 0.000031 Time 0.026043 +2023-10-05 22:14:55,972 - Epoch: [197][ 160/ 1236] Overall Loss 0.161176 Objective Loss 0.161176 LR 0.000031 Time 0.025685 +2023-10-05 22:14:56,175 - Epoch: [197][ 170/ 1236] Overall Loss 0.161175 Objective Loss 0.161175 LR 0.000031 Time 0.025365 +2023-10-05 22:14:56,379 - Epoch: [197][ 180/ 1236] Overall Loss 0.161409 Objective Loss 0.161409 LR 0.000031 Time 0.025087 +2023-10-05 22:14:56,582 - Epoch: [197][ 190/ 1236] Overall Loss 0.161063 Objective Loss 0.161063 LR 0.000031 Time 0.024836 +2023-10-05 22:14:56,786 - Epoch: [197][ 200/ 1236] Overall Loss 0.160343 Objective Loss 0.160343 LR 0.000031 Time 0.024613 +2023-10-05 22:14:56,991 - Epoch: [197][ 210/ 1236] Overall Loss 0.159719 Objective Loss 0.159719 LR 0.000031 Time 0.024414 +2023-10-05 22:14:57,195 - Epoch: [197][ 220/ 1236] Overall Loss 0.159343 Objective Loss 0.159343 LR 0.000031 Time 0.024229 +2023-10-05 22:14:57,399 - Epoch: [197][ 230/ 1236] Overall Loss 0.159514 Objective Loss 0.159514 LR 0.000031 Time 0.024061 +2023-10-05 22:14:57,603 - Epoch: [197][ 240/ 1236] Overall Loss 0.159836 Objective Loss 0.159836 LR 0.000031 Time 0.023907 +2023-10-05 22:14:57,806 - Epoch: [197][ 250/ 1236] Overall Loss 0.159317 Objective Loss 0.159317 LR 0.000031 Time 0.023763 +2023-10-05 22:14:58,010 - Epoch: [197][ 260/ 1236] Overall Loss 0.159349 Objective Loss 0.159349 LR 0.000031 Time 0.023633 +2023-10-05 22:14:58,213 - Epoch: [197][ 270/ 1236] Overall Loss 0.159334 Objective Loss 0.159334 LR 0.000031 Time 0.023507 +2023-10-05 22:14:58,417 - Epoch: [197][ 280/ 1236] Overall Loss 0.158920 Objective Loss 0.158920 LR 0.000031 Time 0.023394 +2023-10-05 22:14:58,619 - Epoch: [197][ 290/ 1236] Overall Loss 0.158125 Objective Loss 0.158125 LR 0.000031 Time 0.023286 +2023-10-05 22:14:58,823 - Epoch: [197][ 300/ 1236] Overall Loss 0.157876 Objective Loss 0.157876 LR 0.000031 Time 0.023188 +2023-10-05 22:14:59,028 - Epoch: [197][ 310/ 1236] Overall Loss 0.157387 Objective Loss 0.157387 LR 0.000031 Time 0.023097 +2023-10-05 22:14:59,231 - Epoch: [197][ 320/ 1236] Overall Loss 0.157171 Objective Loss 0.157171 LR 0.000031 Time 0.023012 +2023-10-05 22:14:59,436 - Epoch: [197][ 330/ 1236] Overall Loss 0.157222 Objective Loss 0.157222 LR 0.000031 Time 0.022933 +2023-10-05 22:14:59,640 - Epoch: [197][ 340/ 1236] Overall Loss 0.156931 Objective Loss 0.156931 LR 0.000031 Time 0.022859 +2023-10-05 22:14:59,844 - Epoch: [197][ 350/ 1236] Overall Loss 0.157330 Objective Loss 0.157330 LR 0.000031 Time 0.022787 +2023-10-05 22:15:00,048 - Epoch: [197][ 360/ 1236] Overall Loss 0.156839 Objective Loss 0.156839 LR 0.000031 Time 0.022718 +2023-10-05 22:15:00,249 - Epoch: [197][ 370/ 1236] Overall Loss 0.157168 Objective Loss 0.157168 LR 0.000031 Time 0.022646 +2023-10-05 22:15:00,451 - Epoch: [197][ 380/ 1236] Overall Loss 0.157434 Objective Loss 0.157434 LR 0.000031 Time 0.022581 +2023-10-05 22:15:00,651 - Epoch: [197][ 390/ 1236] Overall Loss 0.156815 Objective Loss 0.156815 LR 0.000031 Time 0.022516 +2023-10-05 22:15:00,854 - Epoch: [197][ 400/ 1236] Overall Loss 0.156420 Objective Loss 0.156420 LR 0.000031 Time 0.022458 +2023-10-05 22:15:01,054 - Epoch: [197][ 410/ 1236] Overall Loss 0.156516 Objective Loss 0.156516 LR 0.000031 Time 0.022398 +2023-10-05 22:15:01,256 - Epoch: [197][ 420/ 1236] Overall Loss 0.157074 Objective Loss 0.157074 LR 0.000031 Time 0.022346 +2023-10-05 22:15:01,457 - Epoch: [197][ 430/ 1236] Overall Loss 0.157574 Objective Loss 0.157574 LR 0.000031 Time 0.022292 +2023-10-05 22:15:01,659 - Epoch: [197][ 440/ 1236] Overall Loss 0.157978 Objective Loss 0.157978 LR 0.000031 Time 0.022244 +2023-10-05 22:15:01,860 - Epoch: [197][ 450/ 1236] Overall Loss 0.157807 Objective Loss 0.157807 LR 0.000031 Time 0.022195 +2023-10-05 22:15:02,062 - Epoch: [197][ 460/ 1236] Overall Loss 0.157402 Objective Loss 0.157402 LR 0.000031 Time 0.022152 +2023-10-05 22:15:02,263 - Epoch: [197][ 470/ 1236] Overall Loss 0.157432 Objective Loss 0.157432 LR 0.000031 Time 0.022107 +2023-10-05 22:15:02,466 - Epoch: [197][ 480/ 1236] Overall Loss 0.157117 Objective Loss 0.157117 LR 0.000031 Time 0.022068 +2023-10-05 22:15:02,666 - Epoch: [197][ 490/ 1236] Overall Loss 0.157289 Objective Loss 0.157289 LR 0.000031 Time 0.022026 +2023-10-05 22:15:02,868 - Epoch: [197][ 500/ 1236] Overall Loss 0.157165 Objective Loss 0.157165 LR 0.000031 Time 0.021989 +2023-10-05 22:15:03,069 - Epoch: [197][ 510/ 1236] Overall Loss 0.157211 Objective Loss 0.157211 LR 0.000031 Time 0.021951 +2023-10-05 22:15:03,271 - Epoch: [197][ 520/ 1236] Overall Loss 0.157355 Objective Loss 0.157355 LR 0.000031 Time 0.021917 +2023-10-05 22:15:03,472 - Epoch: [197][ 530/ 1236] Overall Loss 0.156965 Objective Loss 0.156965 LR 0.000031 Time 0.021882 +2023-10-05 22:15:03,675 - Epoch: [197][ 540/ 1236] Overall Loss 0.156883 Objective Loss 0.156883 LR 0.000031 Time 0.021851 +2023-10-05 22:15:03,876 - Epoch: [197][ 550/ 1236] Overall Loss 0.156492 Objective Loss 0.156492 LR 0.000031 Time 0.021818 +2023-10-05 22:15:04,078 - Epoch: [197][ 560/ 1236] Overall Loss 0.156907 Objective Loss 0.156907 LR 0.000031 Time 0.021789 +2023-10-05 22:15:04,278 - Epoch: [197][ 570/ 1236] Overall Loss 0.156780 Objective Loss 0.156780 LR 0.000031 Time 0.021758 +2023-10-05 22:15:04,480 - Epoch: [197][ 580/ 1236] Overall Loss 0.157079 Objective Loss 0.157079 LR 0.000031 Time 0.021731 +2023-10-05 22:15:04,681 - Epoch: [197][ 590/ 1236] Overall Loss 0.157295 Objective Loss 0.157295 LR 0.000031 Time 0.021702 +2023-10-05 22:15:04,884 - Epoch: [197][ 600/ 1236] Overall Loss 0.157335 Objective Loss 0.157335 LR 0.000031 Time 0.021677 +2023-10-05 22:15:05,085 - Epoch: [197][ 610/ 1236] Overall Loss 0.157210 Objective Loss 0.157210 LR 0.000031 Time 0.021651 +2023-10-05 22:15:05,287 - Epoch: [197][ 620/ 1236] Overall Loss 0.157290 Objective Loss 0.157290 LR 0.000031 Time 0.021628 +2023-10-05 22:15:05,488 - Epoch: [197][ 630/ 1236] Overall Loss 0.157558 Objective Loss 0.157558 LR 0.000031 Time 0.021602 +2023-10-05 22:15:05,689 - Epoch: [197][ 640/ 1236] Overall Loss 0.157697 Objective Loss 0.157697 LR 0.000031 Time 0.021579 +2023-10-05 22:15:05,891 - Epoch: [197][ 650/ 1236] Overall Loss 0.157750 Objective Loss 0.157750 LR 0.000031 Time 0.021556 +2023-10-05 22:15:06,093 - Epoch: [197][ 660/ 1236] Overall Loss 0.157932 Objective Loss 0.157932 LR 0.000031 Time 0.021536 +2023-10-05 22:15:06,294 - Epoch: [197][ 670/ 1236] Overall Loss 0.157792 Objective Loss 0.157792 LR 0.000031 Time 0.021514 +2023-10-05 22:15:06,497 - Epoch: [197][ 680/ 1236] Overall Loss 0.157675 Objective Loss 0.157675 LR 0.000031 Time 0.021495 +2023-10-05 22:15:06,697 - Epoch: [197][ 690/ 1236] Overall Loss 0.157539 Objective Loss 0.157539 LR 0.000031 Time 0.021474 +2023-10-05 22:15:06,900 - Epoch: [197][ 700/ 1236] Overall Loss 0.158084 Objective Loss 0.158084 LR 0.000031 Time 0.021455 +2023-10-05 22:15:07,100 - Epoch: [197][ 710/ 1236] Overall Loss 0.158003 Objective Loss 0.158003 LR 0.000031 Time 0.021435 +2023-10-05 22:15:07,303 - Epoch: [197][ 720/ 1236] Overall Loss 0.158196 Objective Loss 0.158196 LR 0.000031 Time 0.021418 +2023-10-05 22:15:07,503 - Epoch: [197][ 730/ 1236] Overall Loss 0.158120 Objective Loss 0.158120 LR 0.000031 Time 0.021400 +2023-10-05 22:15:07,706 - Epoch: [197][ 740/ 1236] Overall Loss 0.157938 Objective Loss 0.157938 LR 0.000031 Time 0.021383 +2023-10-05 22:15:07,906 - Epoch: [197][ 750/ 1236] Overall Loss 0.158024 Objective Loss 0.158024 LR 0.000031 Time 0.021365 +2023-10-05 22:15:08,109 - Epoch: [197][ 760/ 1236] Overall Loss 0.158170 Objective Loss 0.158170 LR 0.000031 Time 0.021351 +2023-10-05 22:15:08,310 - Epoch: [197][ 770/ 1236] Overall Loss 0.157938 Objective Loss 0.157938 LR 0.000031 Time 0.021334 +2023-10-05 22:15:08,513 - Epoch: [197][ 780/ 1236] Overall Loss 0.157736 Objective Loss 0.157736 LR 0.000031 Time 0.021320 +2023-10-05 22:15:08,713 - Epoch: [197][ 790/ 1236] Overall Loss 0.157729 Objective Loss 0.157729 LR 0.000031 Time 0.021303 +2023-10-05 22:15:08,916 - Epoch: [197][ 800/ 1236] Overall Loss 0.157809 Objective Loss 0.157809 LR 0.000031 Time 0.021290 +2023-10-05 22:15:09,117 - Epoch: [197][ 810/ 1236] Overall Loss 0.158032 Objective Loss 0.158032 LR 0.000031 Time 0.021275 +2023-10-05 22:15:09,320 - Epoch: [197][ 820/ 1236] Overall Loss 0.158249 Objective Loss 0.158249 LR 0.000031 Time 0.021262 +2023-10-05 22:15:09,521 - Epoch: [197][ 830/ 1236] Overall Loss 0.158573 Objective Loss 0.158573 LR 0.000031 Time 0.021248 +2023-10-05 22:15:09,724 - Epoch: [197][ 840/ 1236] Overall Loss 0.158585 Objective Loss 0.158585 LR 0.000031 Time 0.021236 +2023-10-05 22:15:09,925 - Epoch: [197][ 850/ 1236] Overall Loss 0.158556 Objective Loss 0.158556 LR 0.000031 Time 0.021222 +2023-10-05 22:15:10,127 - Epoch: [197][ 860/ 1236] Overall Loss 0.158416 Objective Loss 0.158416 LR 0.000031 Time 0.021211 +2023-10-05 22:15:10,329 - Epoch: [197][ 870/ 1236] Overall Loss 0.158396 Objective Loss 0.158396 LR 0.000031 Time 0.021198 +2023-10-05 22:15:10,531 - Epoch: [197][ 880/ 1236] Overall Loss 0.158438 Objective Loss 0.158438 LR 0.000031 Time 0.021187 +2023-10-05 22:15:10,733 - Epoch: [197][ 890/ 1236] Overall Loss 0.158473 Objective Loss 0.158473 LR 0.000031 Time 0.021175 +2023-10-05 22:15:10,935 - Epoch: [197][ 900/ 1236] Overall Loss 0.158505 Objective Loss 0.158505 LR 0.000031 Time 0.021164 +2023-10-05 22:15:11,136 - Epoch: [197][ 910/ 1236] Overall Loss 0.158586 Objective Loss 0.158586 LR 0.000031 Time 0.021151 +2023-10-05 22:15:11,339 - Epoch: [197][ 920/ 1236] Overall Loss 0.158589 Objective Loss 0.158589 LR 0.000031 Time 0.021140 +2023-10-05 22:15:11,540 - Epoch: [197][ 930/ 1236] Overall Loss 0.158786 Objective Loss 0.158786 LR 0.000031 Time 0.021129 +2023-10-05 22:15:11,743 - Epoch: [197][ 940/ 1236] Overall Loss 0.158825 Objective Loss 0.158825 LR 0.000031 Time 0.021119 +2023-10-05 22:15:11,944 - Epoch: [197][ 950/ 1236] Overall Loss 0.158819 Objective Loss 0.158819 LR 0.000031 Time 0.021109 +2023-10-05 22:15:12,147 - Epoch: [197][ 960/ 1236] Overall Loss 0.158882 Objective Loss 0.158882 LR 0.000031 Time 0.021100 +2023-10-05 22:15:12,348 - Epoch: [197][ 970/ 1236] Overall Loss 0.158821 Objective Loss 0.158821 LR 0.000031 Time 0.021089 +2023-10-05 22:15:12,551 - Epoch: [197][ 980/ 1236] Overall Loss 0.158835 Objective Loss 0.158835 LR 0.000031 Time 0.021080 +2023-10-05 22:15:12,752 - Epoch: [197][ 990/ 1236] Overall Loss 0.158646 Objective Loss 0.158646 LR 0.000031 Time 0.021071 +2023-10-05 22:15:12,964 - Epoch: [197][ 1000/ 1236] Overall Loss 0.158648 Objective Loss 0.158648 LR 0.000031 Time 0.021072 +2023-10-05 22:15:13,171 - Epoch: [197][ 1010/ 1236] Overall Loss 0.158695 Objective Loss 0.158695 LR 0.000031 Time 0.021068 +2023-10-05 22:15:13,376 - Epoch: [197][ 1020/ 1236] Overall Loss 0.158762 Objective Loss 0.158762 LR 0.000031 Time 0.021062 +2023-10-05 22:15:13,578 - Epoch: [197][ 1030/ 1236] Overall Loss 0.158974 Objective Loss 0.158974 LR 0.000031 Time 0.021053 +2023-10-05 22:15:13,781 - Epoch: [197][ 1040/ 1236] Overall Loss 0.159008 Objective Loss 0.159008 LR 0.000031 Time 0.021045 +2023-10-05 22:15:13,982 - Epoch: [197][ 1050/ 1236] Overall Loss 0.159125 Objective Loss 0.159125 LR 0.000031 Time 0.021036 +2023-10-05 22:15:14,184 - Epoch: [197][ 1060/ 1236] Overall Loss 0.159051 Objective Loss 0.159051 LR 0.000031 Time 0.021028 +2023-10-05 22:15:14,386 - Epoch: [197][ 1070/ 1236] Overall Loss 0.159053 Objective Loss 0.159053 LR 0.000031 Time 0.021020 +2023-10-05 22:15:14,589 - Epoch: [197][ 1080/ 1236] Overall Loss 0.159015 Objective Loss 0.159015 LR 0.000031 Time 0.021013 +2023-10-05 22:15:14,790 - Epoch: [197][ 1090/ 1236] Overall Loss 0.158907 Objective Loss 0.158907 LR 0.000031 Time 0.021004 +2023-10-05 22:15:14,993 - Epoch: [197][ 1100/ 1236] Overall Loss 0.158922 Objective Loss 0.158922 LR 0.000031 Time 0.020997 +2023-10-05 22:15:15,194 - Epoch: [197][ 1110/ 1236] Overall Loss 0.158691 Objective Loss 0.158691 LR 0.000031 Time 0.020989 +2023-10-05 22:15:15,397 - Epoch: [197][ 1120/ 1236] Overall Loss 0.158556 Objective Loss 0.158556 LR 0.000031 Time 0.020983 +2023-10-05 22:15:15,598 - Epoch: [197][ 1130/ 1236] Overall Loss 0.158686 Objective Loss 0.158686 LR 0.000031 Time 0.020974 +2023-10-05 22:15:15,801 - Epoch: [197][ 1140/ 1236] Overall Loss 0.158846 Objective Loss 0.158846 LR 0.000031 Time 0.020967 +2023-10-05 22:15:16,001 - Epoch: [197][ 1150/ 1236] Overall Loss 0.158878 Objective Loss 0.158878 LR 0.000031 Time 0.020959 +2023-10-05 22:15:16,204 - Epoch: [197][ 1160/ 1236] Overall Loss 0.158995 Objective Loss 0.158995 LR 0.000031 Time 0.020953 +2023-10-05 22:15:16,406 - Epoch: [197][ 1170/ 1236] Overall Loss 0.158998 Objective Loss 0.158998 LR 0.000031 Time 0.020946 +2023-10-05 22:15:16,608 - Epoch: [197][ 1180/ 1236] Overall Loss 0.159185 Objective Loss 0.159185 LR 0.000031 Time 0.020939 +2023-10-05 22:15:16,809 - Epoch: [197][ 1190/ 1236] Overall Loss 0.159381 Objective Loss 0.159381 LR 0.000031 Time 0.020932 +2023-10-05 22:15:17,012 - Epoch: [197][ 1200/ 1236] Overall Loss 0.159379 Objective Loss 0.159379 LR 0.000031 Time 0.020926 +2023-10-05 22:15:17,213 - Epoch: [197][ 1210/ 1236] Overall Loss 0.159323 Objective Loss 0.159323 LR 0.000031 Time 0.020919 +2023-10-05 22:15:17,416 - Epoch: [197][ 1220/ 1236] Overall Loss 0.159366 Objective Loss 0.159366 LR 0.000031 Time 0.020913 +2023-10-05 22:15:17,671 - Epoch: [197][ 1230/ 1236] Overall Loss 0.159312 Objective Loss 0.159312 LR 0.000031 Time 0.020951 +2023-10-05 22:15:17,788 - Epoch: [197][ 1236/ 1236] Overall Loss 0.159282 Objective Loss 0.159282 Top1 87.983707 Top5 98.167006 LR 0.000031 Time 0.020944 +2023-10-05 22:15:17,908 - --- validate (epoch=197)----------- +2023-10-05 22:15:17,908 - 29943 samples (256 per mini-batch) +2023-10-05 22:15:18,375 - Epoch: [197][ 10/ 117] Loss 0.271498 Top1 87.226562 Top5 98.515625 +2023-10-05 22:15:18,532 - Epoch: [197][ 20/ 117] Loss 0.285488 Top1 86.582031 Top5 98.457031 +2023-10-05 22:15:18,686 - Epoch: [197][ 30/ 117] Loss 0.292128 Top1 86.236979 Top5 98.372396 +2023-10-05 22:15:18,841 - Epoch: [197][ 40/ 117] Loss 0.297609 Top1 86.132812 Top5 98.359375 +2023-10-05 22:15:18,995 - Epoch: [197][ 50/ 117] Loss 0.299713 Top1 85.992188 Top5 98.335938 +2023-10-05 22:15:19,152 - Epoch: [197][ 60/ 117] Loss 0.298159 Top1 85.917969 Top5 98.339844 +2023-10-05 22:15:19,302 - Epoch: [197][ 70/ 117] Loss 0.296915 Top1 85.943080 Top5 98.331473 +2023-10-05 22:15:19,452 - Epoch: [197][ 80/ 117] Loss 0.296453 Top1 85.991211 Top5 98.359375 +2023-10-05 22:15:19,602 - Epoch: [197][ 90/ 117] Loss 0.298990 Top1 86.076389 Top5 98.346354 +2023-10-05 22:15:19,752 - Epoch: [197][ 100/ 117] Loss 0.301450 Top1 85.949219 Top5 98.335938 +2023-10-05 22:15:19,907 - Epoch: [197][ 110/ 117] Loss 0.299102 Top1 86.022727 Top5 98.341619 +2023-10-05 22:15:19,992 - Epoch: [197][ 117/ 117] Loss 0.297889 Top1 86.080219 Top5 98.326821 +2023-10-05 22:15:20,135 - ==> Top1: 86.080 Top5: 98.327 Loss: 0.298 + +2023-10-05 22:15:20,136 - ==> Confusion: +[[ 942 3 4 0 5 2 0 0 3 62 1 0 1 2 6 2 3 1 0 0 13] + [ 0 1072 1 0 8 13 1 13 0 0 1 1 0 1 2 3 2 0 5 1 7] + [ 5 3 969 12 1 0 20 7 0 0 5 1 6 2 0 3 0 3 7 4 8] + [ 1 1 8 989 1 3 1 1 1 1 3 1 4 2 26 3 0 6 17 1 19] + [ 17 5 0 0 977 3 1 2 0 9 0 2 1 1 9 1 11 3 2 1 5] + [ 3 34 1 1 4 992 0 18 1 0 7 9 0 13 5 2 2 0 5 3 16] + [ 0 5 20 0 1 0 1129 10 0 0 1 2 1 0 1 5 0 0 3 7 6] + [ 3 20 9 0 1 30 4 1077 1 3 3 12 1 2 0 2 0 0 36 3 11] + [ 19 1 3 0 0 2 0 2 976 40 8 2 1 11 13 2 1 1 3 1 3] + [ 92 1 3 0 2 2 0 2 17 961 1 2 1 17 4 1 0 3 0 2 8] + [ 3 4 9 2 0 1 3 3 10 1 982 4 1 9 4 2 2 0 4 1 8] + [ 1 0 2 0 1 12 0 1 0 0 0 966 16 7 0 4 1 15 0 5 4] + [ 1 2 0 6 0 2 0 2 0 0 2 33 986 2 1 3 3 12 1 4 8] + [ 2 0 1 0 1 2 0 0 9 14 7 5 2 1058 4 2 1 0 0 1 10] + [ 14 2 2 7 9 0 0 0 22 2 1 0 2 3 1010 0 1 3 9 0 14] + [ 1 2 3 0 2 0 1 0 0 0 0 9 7 1 1 1068 16 11 0 10 2] + [ 1 17 1 0 4 3 0 0 0 0 0 4 0 1 3 9 1104 0 0 3 11] + [ 0 0 1 1 1 0 4 0 0 0 0 2 16 2 0 3 0 1003 1 0 4] + [ 1 8 3 20 1 0 0 22 1 0 3 0 2 1 9 0 1 0 987 1 8] + [ 0 2 4 3 2 10 6 8 0 0 1 11 6 1 0 7 8 1 4 1066 12] + [ 118 163 120 68 104 116 31 87 76 73 165 94 290 273 136 48 132 66 151 133 5461]] + +2023-10-05 22:15:20,137 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:15:20,137 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:15:20,143 - + +2023-10-05 22:15:20,143 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:15:21,249 - Epoch: [198][ 10/ 1236] Overall Loss 0.169982 Objective Loss 0.169982 LR 0.000031 Time 0.110520 +2023-10-05 22:15:21,451 - Epoch: [198][ 20/ 1236] Overall Loss 0.160820 Objective Loss 0.160820 LR 0.000031 Time 0.065322 +2023-10-05 22:15:21,651 - Epoch: [198][ 30/ 1236] Overall Loss 0.156483 Objective Loss 0.156483 LR 0.000031 Time 0.050195 +2023-10-05 22:15:21,852 - Epoch: [198][ 40/ 1236] Overall Loss 0.158480 Objective Loss 0.158480 LR 0.000031 Time 0.042682 +2023-10-05 22:15:22,053 - Epoch: [198][ 50/ 1236] Overall Loss 0.155283 Objective Loss 0.155283 LR 0.000031 Time 0.038152 +2023-10-05 22:15:22,255 - Epoch: [198][ 60/ 1236] Overall Loss 0.155618 Objective Loss 0.155618 LR 0.000031 Time 0.035147 +2023-10-05 22:15:22,456 - Epoch: [198][ 70/ 1236] Overall Loss 0.155516 Objective Loss 0.155516 LR 0.000031 Time 0.032995 +2023-10-05 22:15:22,657 - Epoch: [198][ 80/ 1236] Overall Loss 0.153853 Objective Loss 0.153853 LR 0.000031 Time 0.031382 +2023-10-05 22:15:22,857 - Epoch: [198][ 90/ 1236] Overall Loss 0.152143 Objective Loss 0.152143 LR 0.000031 Time 0.030112 +2023-10-05 22:15:23,058 - Epoch: [198][ 100/ 1236] Overall Loss 0.152633 Objective Loss 0.152633 LR 0.000031 Time 0.029107 +2023-10-05 22:15:23,258 - Epoch: [198][ 110/ 1236] Overall Loss 0.153350 Objective Loss 0.153350 LR 0.000031 Time 0.028275 +2023-10-05 22:15:23,459 - Epoch: [198][ 120/ 1236] Overall Loss 0.152913 Objective Loss 0.152913 LR 0.000031 Time 0.027584 +2023-10-05 22:15:23,658 - Epoch: [198][ 130/ 1236] Overall Loss 0.153991 Objective Loss 0.153991 LR 0.000031 Time 0.026997 +2023-10-05 22:15:23,859 - Epoch: [198][ 140/ 1236] Overall Loss 0.154043 Objective Loss 0.154043 LR 0.000031 Time 0.026496 +2023-10-05 22:15:24,058 - Epoch: [198][ 150/ 1236] Overall Loss 0.153823 Objective Loss 0.153823 LR 0.000031 Time 0.026060 +2023-10-05 22:15:24,261 - Epoch: [198][ 160/ 1236] Overall Loss 0.153845 Objective Loss 0.153845 LR 0.000031 Time 0.025697 +2023-10-05 22:15:24,465 - Epoch: [198][ 170/ 1236] Overall Loss 0.153199 Objective Loss 0.153199 LR 0.000031 Time 0.025383 +2023-10-05 22:15:24,667 - Epoch: [198][ 180/ 1236] Overall Loss 0.152553 Objective Loss 0.152553 LR 0.000031 Time 0.025094 +2023-10-05 22:15:24,871 - Epoch: [198][ 190/ 1236] Overall Loss 0.154098 Objective Loss 0.154098 LR 0.000031 Time 0.024844 +2023-10-05 22:15:25,073 - Epoch: [198][ 200/ 1236] Overall Loss 0.154345 Objective Loss 0.154345 LR 0.000031 Time 0.024611 +2023-10-05 22:15:25,277 - Epoch: [198][ 210/ 1236] Overall Loss 0.154631 Objective Loss 0.154631 LR 0.000031 Time 0.024408 +2023-10-05 22:15:25,479 - Epoch: [198][ 220/ 1236] Overall Loss 0.154793 Objective Loss 0.154793 LR 0.000031 Time 0.024215 +2023-10-05 22:15:25,683 - Epoch: [198][ 230/ 1236] Overall Loss 0.154276 Objective Loss 0.154276 LR 0.000031 Time 0.024048 +2023-10-05 22:15:25,885 - Epoch: [198][ 240/ 1236] Overall Loss 0.154022 Objective Loss 0.154022 LR 0.000031 Time 0.023887 +2023-10-05 22:15:26,089 - Epoch: [198][ 250/ 1236] Overall Loss 0.154185 Objective Loss 0.154185 LR 0.000031 Time 0.023745 +2023-10-05 22:15:26,291 - Epoch: [198][ 260/ 1236] Overall Loss 0.154897 Objective Loss 0.154897 LR 0.000031 Time 0.023608 +2023-10-05 22:15:26,495 - Epoch: [198][ 270/ 1236] Overall Loss 0.154739 Objective Loss 0.154739 LR 0.000031 Time 0.023487 +2023-10-05 22:15:26,698 - Epoch: [198][ 280/ 1236] Overall Loss 0.155168 Objective Loss 0.155168 LR 0.000031 Time 0.023370 +2023-10-05 22:15:26,902 - Epoch: [198][ 290/ 1236] Overall Loss 0.155740 Objective Loss 0.155740 LR 0.000031 Time 0.023267 +2023-10-05 22:15:27,104 - Epoch: [198][ 300/ 1236] Overall Loss 0.154992 Objective Loss 0.154992 LR 0.000031 Time 0.023164 +2023-10-05 22:15:27,308 - Epoch: [198][ 310/ 1236] Overall Loss 0.155256 Objective Loss 0.155256 LR 0.000031 Time 0.023074 +2023-10-05 22:15:27,510 - Epoch: [198][ 320/ 1236] Overall Loss 0.156048 Objective Loss 0.156048 LR 0.000031 Time 0.022984 +2023-10-05 22:15:27,714 - Epoch: [198][ 330/ 1236] Overall Loss 0.156722 Objective Loss 0.156722 LR 0.000031 Time 0.022904 +2023-10-05 22:15:27,917 - Epoch: [198][ 340/ 1236] Overall Loss 0.156793 Objective Loss 0.156793 LR 0.000031 Time 0.022825 +2023-10-05 22:15:28,121 - Epoch: [198][ 350/ 1236] Overall Loss 0.156750 Objective Loss 0.156750 LR 0.000031 Time 0.022756 +2023-10-05 22:15:28,323 - Epoch: [198][ 360/ 1236] Overall Loss 0.156838 Objective Loss 0.156838 LR 0.000031 Time 0.022684 +2023-10-05 22:15:28,528 - Epoch: [198][ 370/ 1236] Overall Loss 0.157196 Objective Loss 0.157196 LR 0.000031 Time 0.022623 +2023-10-05 22:15:28,732 - Epoch: [198][ 380/ 1236] Overall Loss 0.157132 Objective Loss 0.157132 LR 0.000031 Time 0.022565 +2023-10-05 22:15:28,938 - Epoch: [198][ 390/ 1236] Overall Loss 0.157332 Objective Loss 0.157332 LR 0.000031 Time 0.022512 +2023-10-05 22:15:29,142 - Epoch: [198][ 400/ 1236] Overall Loss 0.156842 Objective Loss 0.156842 LR 0.000031 Time 0.022459 +2023-10-05 22:15:29,347 - Epoch: [198][ 410/ 1236] Overall Loss 0.156719 Objective Loss 0.156719 LR 0.000031 Time 0.022410 +2023-10-05 22:15:29,551 - Epoch: [198][ 420/ 1236] Overall Loss 0.156843 Objective Loss 0.156843 LR 0.000031 Time 0.022362 +2023-10-05 22:15:29,756 - Epoch: [198][ 430/ 1236] Overall Loss 0.156721 Objective Loss 0.156721 LR 0.000031 Time 0.022318 +2023-10-05 22:15:29,961 - Epoch: [198][ 440/ 1236] Overall Loss 0.156742 Objective Loss 0.156742 LR 0.000031 Time 0.022275 +2023-10-05 22:15:30,166 - Epoch: [198][ 450/ 1236] Overall Loss 0.156898 Objective Loss 0.156898 LR 0.000031 Time 0.022236 +2023-10-05 22:15:30,371 - Epoch: [198][ 460/ 1236] Overall Loss 0.156983 Objective Loss 0.156983 LR 0.000031 Time 0.022196 +2023-10-05 22:15:30,576 - Epoch: [198][ 470/ 1236] Overall Loss 0.157092 Objective Loss 0.157092 LR 0.000031 Time 0.022159 +2023-10-05 22:15:30,781 - Epoch: [198][ 480/ 1236] Overall Loss 0.157229 Objective Loss 0.157229 LR 0.000031 Time 0.022123 +2023-10-05 22:15:30,986 - Epoch: [198][ 490/ 1236] Overall Loss 0.157133 Objective Loss 0.157133 LR 0.000031 Time 0.022089 +2023-10-05 22:15:31,190 - Epoch: [198][ 500/ 1236] Overall Loss 0.157764 Objective Loss 0.157764 LR 0.000031 Time 0.022055 +2023-10-05 22:15:31,395 - Epoch: [198][ 510/ 1236] Overall Loss 0.158253 Objective Loss 0.158253 LR 0.000031 Time 0.022025 +2023-10-05 22:15:31,600 - Epoch: [198][ 520/ 1236] Overall Loss 0.158730 Objective Loss 0.158730 LR 0.000031 Time 0.021994 +2023-10-05 22:15:31,805 - Epoch: [198][ 530/ 1236] Overall Loss 0.158971 Objective Loss 0.158971 LR 0.000031 Time 0.021965 +2023-10-05 22:15:32,010 - Epoch: [198][ 540/ 1236] Overall Loss 0.158982 Objective Loss 0.158982 LR 0.000031 Time 0.021936 +2023-10-05 22:15:32,215 - Epoch: [198][ 550/ 1236] Overall Loss 0.159224 Objective Loss 0.159224 LR 0.000031 Time 0.021910 +2023-10-05 22:15:32,420 - Epoch: [198][ 560/ 1236] Overall Loss 0.159215 Objective Loss 0.159215 LR 0.000031 Time 0.021883 +2023-10-05 22:15:32,625 - Epoch: [198][ 570/ 1236] Overall Loss 0.158937 Objective Loss 0.158937 LR 0.000031 Time 0.021859 +2023-10-05 22:15:32,829 - Epoch: [198][ 580/ 1236] Overall Loss 0.158597 Objective Loss 0.158597 LR 0.000031 Time 0.021834 +2023-10-05 22:15:33,035 - Epoch: [198][ 590/ 1236] Overall Loss 0.158708 Objective Loss 0.158708 LR 0.000031 Time 0.021811 +2023-10-05 22:15:33,239 - Epoch: [198][ 600/ 1236] Overall Loss 0.158692 Objective Loss 0.158692 LR 0.000031 Time 0.021787 +2023-10-05 22:15:33,444 - Epoch: [198][ 610/ 1236] Overall Loss 0.158288 Objective Loss 0.158288 LR 0.000031 Time 0.021766 +2023-10-05 22:15:33,649 - Epoch: [198][ 620/ 1236] Overall Loss 0.158475 Objective Loss 0.158475 LR 0.000031 Time 0.021744 +2023-10-05 22:15:33,854 - Epoch: [198][ 630/ 1236] Overall Loss 0.158634 Objective Loss 0.158634 LR 0.000031 Time 0.021724 +2023-10-05 22:15:34,058 - Epoch: [198][ 640/ 1236] Overall Loss 0.158653 Objective Loss 0.158653 LR 0.000031 Time 0.021703 +2023-10-05 22:15:34,263 - Epoch: [198][ 650/ 1236] Overall Loss 0.158620 Objective Loss 0.158620 LR 0.000031 Time 0.021684 +2023-10-05 22:15:34,468 - Epoch: [198][ 660/ 1236] Overall Loss 0.158548 Objective Loss 0.158548 LR 0.000031 Time 0.021665 +2023-10-05 22:15:34,673 - Epoch: [198][ 670/ 1236] Overall Loss 0.158484 Objective Loss 0.158484 LR 0.000031 Time 0.021648 +2023-10-05 22:15:34,878 - Epoch: [198][ 680/ 1236] Overall Loss 0.158347 Objective Loss 0.158347 LR 0.000031 Time 0.021630 +2023-10-05 22:15:35,083 - Epoch: [198][ 690/ 1236] Overall Loss 0.158188 Objective Loss 0.158188 LR 0.000031 Time 0.021613 +2023-10-05 22:15:35,287 - Epoch: [198][ 700/ 1236] Overall Loss 0.157972 Objective Loss 0.157972 LR 0.000031 Time 0.021596 +2023-10-05 22:15:35,492 - Epoch: [198][ 710/ 1236] Overall Loss 0.157744 Objective Loss 0.157744 LR 0.000031 Time 0.021580 +2023-10-05 22:15:35,697 - Epoch: [198][ 720/ 1236] Overall Loss 0.157576 Objective Loss 0.157576 LR 0.000031 Time 0.021564 +2023-10-05 22:15:35,902 - Epoch: [198][ 730/ 1236] Overall Loss 0.157559 Objective Loss 0.157559 LR 0.000031 Time 0.021549 +2023-10-05 22:15:36,107 - Epoch: [198][ 740/ 1236] Overall Loss 0.157484 Objective Loss 0.157484 LR 0.000031 Time 0.021534 +2023-10-05 22:15:36,313 - Epoch: [198][ 750/ 1236] Overall Loss 0.157511 Objective Loss 0.157511 LR 0.000031 Time 0.021520 +2023-10-05 22:15:36,517 - Epoch: [198][ 760/ 1236] Overall Loss 0.157310 Objective Loss 0.157310 LR 0.000031 Time 0.021505 +2023-10-05 22:15:36,722 - Epoch: [198][ 770/ 1236] Overall Loss 0.157395 Objective Loss 0.157395 LR 0.000031 Time 0.021492 +2023-10-05 22:15:36,927 - Epoch: [198][ 780/ 1236] Overall Loss 0.157567 Objective Loss 0.157567 LR 0.000031 Time 0.021478 +2023-10-05 22:15:37,132 - Epoch: [198][ 790/ 1236] Overall Loss 0.157284 Objective Loss 0.157284 LR 0.000031 Time 0.021466 +2023-10-05 22:15:37,337 - Epoch: [198][ 800/ 1236] Overall Loss 0.157505 Objective Loss 0.157505 LR 0.000031 Time 0.021453 +2023-10-05 22:15:37,542 - Epoch: [198][ 810/ 1236] Overall Loss 0.157721 Objective Loss 0.157721 LR 0.000031 Time 0.021441 +2023-10-05 22:15:37,746 - Epoch: [198][ 820/ 1236] Overall Loss 0.157804 Objective Loss 0.157804 LR 0.000031 Time 0.021428 +2023-10-05 22:15:37,952 - Epoch: [198][ 830/ 1236] Overall Loss 0.157722 Objective Loss 0.157722 LR 0.000031 Time 0.021417 +2023-10-05 22:15:38,156 - Epoch: [198][ 840/ 1236] Overall Loss 0.157713 Objective Loss 0.157713 LR 0.000031 Time 0.021405 +2023-10-05 22:15:38,362 - Epoch: [198][ 850/ 1236] Overall Loss 0.157647 Objective Loss 0.157647 LR 0.000031 Time 0.021394 +2023-10-05 22:15:38,566 - Epoch: [198][ 860/ 1236] Overall Loss 0.157686 Objective Loss 0.157686 LR 0.000031 Time 0.021383 +2023-10-05 22:15:38,772 - Epoch: [198][ 870/ 1236] Overall Loss 0.157670 Objective Loss 0.157670 LR 0.000031 Time 0.021373 +2023-10-05 22:15:38,976 - Epoch: [198][ 880/ 1236] Overall Loss 0.157863 Objective Loss 0.157863 LR 0.000031 Time 0.021362 +2023-10-05 22:15:39,182 - Epoch: [198][ 890/ 1236] Overall Loss 0.157805 Objective Loss 0.157805 LR 0.000031 Time 0.021352 +2023-10-05 22:15:39,386 - Epoch: [198][ 900/ 1236] Overall Loss 0.157969 Objective Loss 0.157969 LR 0.000031 Time 0.021342 +2023-10-05 22:15:39,592 - Epoch: [198][ 910/ 1236] Overall Loss 0.157953 Objective Loss 0.157953 LR 0.000031 Time 0.021333 +2023-10-05 22:15:39,795 - Epoch: [198][ 920/ 1236] Overall Loss 0.158143 Objective Loss 0.158143 LR 0.000031 Time 0.021322 +2023-10-05 22:15:40,001 - Epoch: [198][ 930/ 1236] Overall Loss 0.158436 Objective Loss 0.158436 LR 0.000031 Time 0.021313 +2023-10-05 22:15:40,205 - Epoch: [198][ 940/ 1236] Overall Loss 0.158234 Objective Loss 0.158234 LR 0.000031 Time 0.021304 +2023-10-05 22:15:40,411 - Epoch: [198][ 950/ 1236] Overall Loss 0.158249 Objective Loss 0.158249 LR 0.000031 Time 0.021295 +2023-10-05 22:15:40,615 - Epoch: [198][ 960/ 1236] Overall Loss 0.158301 Objective Loss 0.158301 LR 0.000031 Time 0.021286 +2023-10-05 22:15:40,820 - Epoch: [198][ 970/ 1236] Overall Loss 0.158351 Objective Loss 0.158351 LR 0.000031 Time 0.021278 +2023-10-05 22:15:41,025 - Epoch: [198][ 980/ 1236] Overall Loss 0.158260 Objective Loss 0.158260 LR 0.000031 Time 0.021269 +2023-10-05 22:15:41,231 - Epoch: [198][ 990/ 1236] Overall Loss 0.158437 Objective Loss 0.158437 LR 0.000031 Time 0.021261 +2023-10-05 22:15:41,435 - Epoch: [198][ 1000/ 1236] Overall Loss 0.158566 Objective Loss 0.158566 LR 0.000031 Time 0.021253 +2023-10-05 22:15:41,641 - Epoch: [198][ 1010/ 1236] Overall Loss 0.158370 Objective Loss 0.158370 LR 0.000031 Time 0.021245 +2023-10-05 22:15:41,846 - Epoch: [198][ 1020/ 1236] Overall Loss 0.158542 Objective Loss 0.158542 LR 0.000031 Time 0.021238 +2023-10-05 22:15:42,051 - Epoch: [198][ 1030/ 1236] Overall Loss 0.158593 Objective Loss 0.158593 LR 0.000031 Time 0.021230 +2023-10-05 22:15:42,256 - Epoch: [198][ 1040/ 1236] Overall Loss 0.158548 Objective Loss 0.158548 LR 0.000031 Time 0.021223 +2023-10-05 22:15:42,461 - Epoch: [198][ 1050/ 1236] Overall Loss 0.158448 Objective Loss 0.158448 LR 0.000031 Time 0.021216 +2023-10-05 22:15:42,666 - Epoch: [198][ 1060/ 1236] Overall Loss 0.158436 Objective Loss 0.158436 LR 0.000031 Time 0.021208 +2023-10-05 22:15:42,871 - Epoch: [198][ 1070/ 1236] Overall Loss 0.158194 Objective Loss 0.158194 LR 0.000031 Time 0.021202 +2023-10-05 22:15:43,076 - Epoch: [198][ 1080/ 1236] Overall Loss 0.158143 Objective Loss 0.158143 LR 0.000031 Time 0.021195 +2023-10-05 22:15:43,281 - Epoch: [198][ 1090/ 1236] Overall Loss 0.157934 Objective Loss 0.157934 LR 0.000031 Time 0.021188 +2023-10-05 22:15:43,486 - Epoch: [198][ 1100/ 1236] Overall Loss 0.157926 Objective Loss 0.157926 LR 0.000031 Time 0.021181 +2023-10-05 22:15:43,691 - Epoch: [198][ 1110/ 1236] Overall Loss 0.158011 Objective Loss 0.158011 LR 0.000031 Time 0.021175 +2023-10-05 22:15:43,896 - Epoch: [198][ 1120/ 1236] Overall Loss 0.157979 Objective Loss 0.157979 LR 0.000031 Time 0.021168 +2023-10-05 22:15:44,101 - Epoch: [198][ 1130/ 1236] Overall Loss 0.157910 Objective Loss 0.157910 LR 0.000031 Time 0.021162 +2023-10-05 22:15:44,306 - Epoch: [198][ 1140/ 1236] Overall Loss 0.157816 Objective Loss 0.157816 LR 0.000031 Time 0.021156 +2023-10-05 22:15:44,511 - Epoch: [198][ 1150/ 1236] Overall Loss 0.157627 Objective Loss 0.157627 LR 0.000031 Time 0.021150 +2023-10-05 22:15:44,716 - Epoch: [198][ 1160/ 1236] Overall Loss 0.157664 Objective Loss 0.157664 LR 0.000031 Time 0.021144 +2023-10-05 22:15:44,921 - Epoch: [198][ 1170/ 1236] Overall Loss 0.157664 Objective Loss 0.157664 LR 0.000031 Time 0.021138 +2023-10-05 22:15:45,126 - Epoch: [198][ 1180/ 1236] Overall Loss 0.157902 Objective Loss 0.157902 LR 0.000031 Time 0.021132 +2023-10-05 22:15:45,331 - Epoch: [198][ 1190/ 1236] Overall Loss 0.157885 Objective Loss 0.157885 LR 0.000031 Time 0.021126 +2023-10-05 22:15:45,535 - Epoch: [198][ 1200/ 1236] Overall Loss 0.157816 Objective Loss 0.157816 LR 0.000031 Time 0.021121 +2023-10-05 22:15:45,741 - Epoch: [198][ 1210/ 1236] Overall Loss 0.157864 Objective Loss 0.157864 LR 0.000031 Time 0.021116 +2023-10-05 22:15:45,945 - Epoch: [198][ 1220/ 1236] Overall Loss 0.158027 Objective Loss 0.158027 LR 0.000031 Time 0.021110 +2023-10-05 22:15:46,203 - Epoch: [198][ 1230/ 1236] Overall Loss 0.158128 Objective Loss 0.158128 LR 0.000031 Time 0.021147 +2023-10-05 22:15:46,320 - Epoch: [198][ 1236/ 1236] Overall Loss 0.158185 Objective Loss 0.158185 Top1 90.020367 Top5 98.778004 LR 0.000031 Time 0.021140 +2023-10-05 22:15:46,447 - --- validate (epoch=198)----------- +2023-10-05 22:15:46,447 - 29943 samples (256 per mini-batch) +2023-10-05 22:15:46,900 - Epoch: [198][ 10/ 117] Loss 0.303551 Top1 86.289062 Top5 98.437500 +2023-10-05 22:15:47,044 - Epoch: [198][ 20/ 117] Loss 0.305500 Top1 85.898438 Top5 98.476562 +2023-10-05 22:15:47,189 - Epoch: [198][ 30/ 117] Loss 0.300955 Top1 85.989583 Top5 98.398438 +2023-10-05 22:15:47,332 - Epoch: [198][ 40/ 117] Loss 0.297601 Top1 86.005859 Top5 98.310547 +2023-10-05 22:15:47,477 - Epoch: [198][ 50/ 117] Loss 0.299437 Top1 86.164062 Top5 98.289062 +2023-10-05 22:15:47,621 - Epoch: [198][ 60/ 117] Loss 0.298189 Top1 86.217448 Top5 98.255208 +2023-10-05 22:15:47,763 - Epoch: [198][ 70/ 117] Loss 0.295465 Top1 86.272321 Top5 98.309152 +2023-10-05 22:15:47,906 - Epoch: [198][ 80/ 117] Loss 0.299557 Top1 86.269531 Top5 98.300781 +2023-10-05 22:15:48,048 - Epoch: [198][ 90/ 117] Loss 0.301324 Top1 86.202257 Top5 98.289931 +2023-10-05 22:15:48,190 - Epoch: [198][ 100/ 117] Loss 0.302790 Top1 86.191406 Top5 98.261719 +2023-10-05 22:15:48,340 - Epoch: [198][ 110/ 117] Loss 0.304195 Top1 86.182528 Top5 98.256392 +2023-10-05 22:15:48,425 - Epoch: [198][ 117/ 117] Loss 0.302659 Top1 86.150352 Top5 98.243329 +2023-10-05 22:15:48,527 - ==> Top1: 86.150 Top5: 98.243 Loss: 0.303 + +2023-10-05 22:15:48,527 - ==> Confusion: +[[ 949 3 4 0 5 3 0 0 4 57 1 0 1 2 5 0 3 1 0 0 12] + [ 0 1065 1 0 10 15 1 18 1 0 0 2 0 0 2 3 2 0 4 1 6] + [ 4 2 978 10 2 0 18 7 0 0 5 1 8 1 0 3 0 2 6 3 6] + [ 1 1 14 971 3 3 1 1 0 1 6 1 5 1 27 4 0 8 19 3 19] + [ 22 4 0 0 978 4 1 1 0 10 1 1 0 1 9 2 8 3 1 1 3] + [ 3 35 1 0 3 991 0 16 0 0 6 8 0 16 5 2 3 0 3 4 20] + [ 0 6 23 0 0 1 1128 8 0 0 0 2 1 0 1 6 0 0 2 7 6] + [ 4 17 10 0 1 30 3 1085 1 4 1 12 1 3 0 2 0 0 30 6 8] + [ 18 1 3 0 0 2 1 1 979 43 7 2 1 8 12 2 2 1 2 0 4] + [ 105 1 3 1 3 2 0 0 20 949 0 2 1 15 4 4 0 1 0 1 7] + [ 3 4 10 3 0 0 4 5 14 0 971 2 0 12 4 2 2 0 4 3 10] + [ 1 0 2 0 1 10 0 2 0 1 0 970 13 6 0 5 0 15 0 4 5] + [ 1 1 1 5 0 3 0 2 1 0 2 34 983 2 1 3 3 14 0 4 8] + [ 2 0 1 0 1 3 0 0 14 18 6 4 2 1054 3 2 1 0 0 0 8] + [ 16 1 3 5 5 0 0 0 26 3 1 1 3 2 1013 0 1 2 6 0 13] + [ 1 3 2 0 2 0 1 0 0 0 0 8 8 2 1 1073 15 8 0 8 2] + [ 1 13 1 0 5 3 0 0 1 0 0 3 0 0 2 11 1104 0 0 3 14] + [ 0 0 1 1 1 0 3 0 1 0 0 2 13 1 0 3 0 1008 1 0 3] + [ 1 7 6 18 1 0 0 21 1 0 3 0 2 1 11 0 1 0 985 1 9] + [ 1 2 3 2 2 6 6 7 0 0 2 12 4 0 0 7 8 3 2 1075 10] + [ 125 149 131 52 95 114 33 81 88 72 167 98 291 264 129 57 137 64 125 146 5487]] + +2023-10-05 22:15:48,529 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:15:48,529 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:15:48,535 - + +2023-10-05 22:15:48,535 - Training epoch: 316395 samples (256 per mini-batch) +2023-10-05 22:15:49,519 - Epoch: [199][ 10/ 1236] Overall Loss 0.152226 Objective Loss 0.152226 LR 0.000031 Time 0.098379 +2023-10-05 22:15:49,722 - Epoch: [199][ 20/ 1236] Overall Loss 0.166063 Objective Loss 0.166063 LR 0.000031 Time 0.059310 +2023-10-05 22:15:49,923 - Epoch: [199][ 30/ 1236] Overall Loss 0.158591 Objective Loss 0.158591 LR 0.000031 Time 0.046248 +2023-10-05 22:15:50,126 - Epoch: [199][ 40/ 1236] Overall Loss 0.151168 Objective Loss 0.151168 LR 0.000031 Time 0.039741 +2023-10-05 22:15:50,328 - Epoch: [199][ 50/ 1236] Overall Loss 0.151694 Objective Loss 0.151694 LR 0.000031 Time 0.035830 +2023-10-05 22:15:50,531 - Epoch: [199][ 60/ 1236] Overall Loss 0.154776 Objective Loss 0.154776 LR 0.000031 Time 0.033228 +2023-10-05 22:15:50,733 - Epoch: [199][ 70/ 1236] Overall Loss 0.159927 Objective Loss 0.159927 LR 0.000031 Time 0.031365 +2023-10-05 22:15:50,935 - Epoch: [199][ 80/ 1236] Overall Loss 0.157412 Objective Loss 0.157412 LR 0.000031 Time 0.029970 +2023-10-05 22:15:51,136 - Epoch: [199][ 90/ 1236] Overall Loss 0.158190 Objective Loss 0.158190 LR 0.000031 Time 0.028867 +2023-10-05 22:15:51,338 - Epoch: [199][ 100/ 1236] Overall Loss 0.159336 Objective Loss 0.159336 LR 0.000031 Time 0.027993 +2023-10-05 22:15:51,539 - Epoch: [199][ 110/ 1236] Overall Loss 0.158610 Objective Loss 0.158610 LR 0.000031 Time 0.027270 +2023-10-05 22:15:51,740 - Epoch: [199][ 120/ 1236] Overall Loss 0.158747 Objective Loss 0.158747 LR 0.000031 Time 0.026672 +2023-10-05 22:15:51,940 - Epoch: [199][ 130/ 1236] Overall Loss 0.160693 Objective Loss 0.160693 LR 0.000031 Time 0.026159 +2023-10-05 22:15:52,141 - Epoch: [199][ 140/ 1236] Overall Loss 0.159353 Objective Loss 0.159353 LR 0.000031 Time 0.025721 +2023-10-05 22:15:52,341 - Epoch: [199][ 150/ 1236] Overall Loss 0.157797 Objective Loss 0.157797 LR 0.000031 Time 0.025339 +2023-10-05 22:15:52,543 - Epoch: [199][ 160/ 1236] Overall Loss 0.158694 Objective Loss 0.158694 LR 0.000031 Time 0.025011 +2023-10-05 22:15:52,743 - Epoch: [199][ 170/ 1236] Overall Loss 0.158908 Objective Loss 0.158908 LR 0.000031 Time 0.024720 +2023-10-05 22:15:52,944 - Epoch: [199][ 180/ 1236] Overall Loss 0.158813 Objective Loss 0.158813 LR 0.000031 Time 0.024459 +2023-10-05 22:15:53,145 - Epoch: [199][ 190/ 1236] Overall Loss 0.158659 Objective Loss 0.158659 LR 0.000031 Time 0.024226 +2023-10-05 22:15:53,345 - Epoch: [199][ 200/ 1236] Overall Loss 0.159162 Objective Loss 0.159162 LR 0.000031 Time 0.024016 +2023-10-05 22:15:53,545 - Epoch: [199][ 210/ 1236] Overall Loss 0.160040 Objective Loss 0.160040 LR 0.000031 Time 0.023822 +2023-10-05 22:15:53,746 - Epoch: [199][ 220/ 1236] Overall Loss 0.158697 Objective Loss 0.158697 LR 0.000031 Time 0.023651 +2023-10-05 22:15:53,947 - Epoch: [199][ 230/ 1236] Overall Loss 0.158524 Objective Loss 0.158524 LR 0.000031 Time 0.023495 +2023-10-05 22:15:54,148 - Epoch: [199][ 240/ 1236] Overall Loss 0.159116 Objective Loss 0.159116 LR 0.000031 Time 0.023353 +2023-10-05 22:15:54,349 - Epoch: [199][ 250/ 1236] Overall Loss 0.158266 Objective Loss 0.158266 LR 0.000031 Time 0.023220 +2023-10-05 22:15:54,550 - Epoch: [199][ 260/ 1236] Overall Loss 0.158170 Objective Loss 0.158170 LR 0.000031 Time 0.023099 +2023-10-05 22:15:54,751 - Epoch: [199][ 270/ 1236] Overall Loss 0.158359 Objective Loss 0.158359 LR 0.000031 Time 0.022986 +2023-10-05 22:15:54,952 - Epoch: [199][ 280/ 1236] Overall Loss 0.158154 Objective Loss 0.158154 LR 0.000031 Time 0.022883 +2023-10-05 22:15:55,153 - Epoch: [199][ 290/ 1236] Overall Loss 0.158536 Objective Loss 0.158536 LR 0.000031 Time 0.022784 +2023-10-05 22:15:55,355 - Epoch: [199][ 300/ 1236] Overall Loss 0.158390 Objective Loss 0.158390 LR 0.000031 Time 0.022697 +2023-10-05 22:15:55,555 - Epoch: [199][ 310/ 1236] Overall Loss 0.159052 Objective Loss 0.159052 LR 0.000031 Time 0.022612 +2023-10-05 22:15:55,757 - Epoch: [199][ 320/ 1236] Overall Loss 0.159020 Objective Loss 0.159020 LR 0.000031 Time 0.022532 +2023-10-05 22:15:55,957 - Epoch: [199][ 330/ 1236] Overall Loss 0.159301 Objective Loss 0.159301 LR 0.000031 Time 0.022456 +2023-10-05 22:15:56,159 - Epoch: [199][ 340/ 1236] Overall Loss 0.158832 Objective Loss 0.158832 LR 0.000031 Time 0.022387 +2023-10-05 22:15:56,360 - Epoch: [199][ 350/ 1236] Overall Loss 0.158970 Objective Loss 0.158970 LR 0.000031 Time 0.022322 +2023-10-05 22:15:56,561 - Epoch: [199][ 360/ 1236] Overall Loss 0.159214 Objective Loss 0.159214 LR 0.000031 Time 0.022260 +2023-10-05 22:15:56,762 - Epoch: [199][ 370/ 1236] Overall Loss 0.159191 Objective Loss 0.159191 LR 0.000031 Time 0.022199 +2023-10-05 22:15:56,965 - Epoch: [199][ 380/ 1236] Overall Loss 0.159063 Objective Loss 0.159063 LR 0.000031 Time 0.022149 +2023-10-05 22:15:57,166 - Epoch: [199][ 390/ 1236] Overall Loss 0.158945 Objective Loss 0.158945 LR 0.000031 Time 0.022097 +2023-10-05 22:15:57,370 - Epoch: [199][ 400/ 1236] Overall Loss 0.158996 Objective Loss 0.158996 LR 0.000031 Time 0.022052 +2023-10-05 22:15:57,572 - Epoch: [199][ 410/ 1236] Overall Loss 0.158518 Objective Loss 0.158518 LR 0.000031 Time 0.022005 +2023-10-05 22:15:57,775 - Epoch: [199][ 420/ 1236] Overall Loss 0.158416 Objective Loss 0.158416 LR 0.000031 Time 0.021964 +2023-10-05 22:15:57,976 - Epoch: [199][ 430/ 1236] Overall Loss 0.158399 Objective Loss 0.158399 LR 0.000031 Time 0.021920 +2023-10-05 22:15:58,179 - Epoch: [199][ 440/ 1236] Overall Loss 0.158481 Objective Loss 0.158481 LR 0.000031 Time 0.021883 +2023-10-05 22:15:58,380 - Epoch: [199][ 450/ 1236] Overall Loss 0.158577 Objective Loss 0.158577 LR 0.000031 Time 0.021842 +2023-10-05 22:15:58,583 - Epoch: [199][ 460/ 1236] Overall Loss 0.158788 Objective Loss 0.158788 LR 0.000031 Time 0.021808 +2023-10-05 22:15:58,785 - Epoch: [199][ 470/ 1236] Overall Loss 0.158355 Objective Loss 0.158355 LR 0.000031 Time 0.021774 +2023-10-05 22:15:58,988 - Epoch: [199][ 480/ 1236] Overall Loss 0.158505 Objective Loss 0.158505 LR 0.000031 Time 0.021741 +2023-10-05 22:15:59,190 - Epoch: [199][ 490/ 1236] Overall Loss 0.158139 Objective Loss 0.158139 LR 0.000031 Time 0.021709 +2023-10-05 22:15:59,393 - Epoch: [199][ 500/ 1236] Overall Loss 0.158019 Objective Loss 0.158019 LR 0.000031 Time 0.021680 +2023-10-05 22:15:59,595 - Epoch: [199][ 510/ 1236] Overall Loss 0.157694 Objective Loss 0.157694 LR 0.000031 Time 0.021650 +2023-10-05 22:15:59,798 - Epoch: [199][ 520/ 1236] Overall Loss 0.157433 Objective Loss 0.157433 LR 0.000031 Time 0.021623 +2023-10-05 22:16:00,000 - Epoch: [199][ 530/ 1236] Overall Loss 0.157575 Objective Loss 0.157575 LR 0.000031 Time 0.021597 +2023-10-05 22:16:00,203 - Epoch: [199][ 540/ 1236] Overall Loss 0.157810 Objective Loss 0.157810 LR 0.000031 Time 0.021571 +2023-10-05 22:16:00,405 - Epoch: [199][ 550/ 1236] Overall Loss 0.158024 Objective Loss 0.158024 LR 0.000031 Time 0.021546 +2023-10-05 22:16:00,608 - Epoch: [199][ 560/ 1236] Overall Loss 0.158333 Objective Loss 0.158333 LR 0.000031 Time 0.021523 +2023-10-05 22:16:00,814 - Epoch: [199][ 570/ 1236] Overall Loss 0.158357 Objective Loss 0.158357 LR 0.000031 Time 0.021507 +2023-10-05 22:16:01,017 - Epoch: [199][ 580/ 1236] Overall Loss 0.158180 Objective Loss 0.158180 LR 0.000031 Time 0.021485 +2023-10-05 22:16:01,224 - Epoch: [199][ 590/ 1236] Overall Loss 0.158365 Objective Loss 0.158365 LR 0.000031 Time 0.021471 +2023-10-05 22:16:01,427 - Epoch: [199][ 600/ 1236] Overall Loss 0.158033 Objective Loss 0.158033 LR 0.000031 Time 0.021451 +2023-10-05 22:16:01,634 - Epoch: [199][ 610/ 1236] Overall Loss 0.157817 Objective Loss 0.157817 LR 0.000031 Time 0.021438 +2023-10-05 22:16:01,837 - Epoch: [199][ 620/ 1236] Overall Loss 0.157913 Objective Loss 0.157913 LR 0.000031 Time 0.021420 +2023-10-05 22:16:02,044 - Epoch: [199][ 630/ 1236] Overall Loss 0.157819 Objective Loss 0.157819 LR 0.000031 Time 0.021407 +2023-10-05 22:16:02,247 - Epoch: [199][ 640/ 1236] Overall Loss 0.157809 Objective Loss 0.157809 LR 0.000031 Time 0.021390 +2023-10-05 22:16:02,454 - Epoch: [199][ 650/ 1236] Overall Loss 0.157726 Objective Loss 0.157726 LR 0.000031 Time 0.021379 +2023-10-05 22:16:02,657 - Epoch: [199][ 660/ 1236] Overall Loss 0.157915 Objective Loss 0.157915 LR 0.000031 Time 0.021362 +2023-10-05 22:16:02,864 - Epoch: [199][ 670/ 1236] Overall Loss 0.157935 Objective Loss 0.157935 LR 0.000031 Time 0.021351 +2023-10-05 22:16:03,067 - Epoch: [199][ 680/ 1236] Overall Loss 0.157996 Objective Loss 0.157996 LR 0.000031 Time 0.021335 +2023-10-05 22:16:03,274 - Epoch: [199][ 690/ 1236] Overall Loss 0.158019 Objective Loss 0.158019 LR 0.000031 Time 0.021325 +2023-10-05 22:16:03,477 - Epoch: [199][ 700/ 1236] Overall Loss 0.158228 Objective Loss 0.158228 LR 0.000031 Time 0.021311 +2023-10-05 22:16:03,685 - Epoch: [199][ 710/ 1236] Overall Loss 0.158269 Objective Loss 0.158269 LR 0.000031 Time 0.021302 +2023-10-05 22:16:03,888 - Epoch: [199][ 720/ 1236] Overall Loss 0.158251 Objective Loss 0.158251 LR 0.000031 Time 0.021288 +2023-10-05 22:16:04,095 - Epoch: [199][ 730/ 1236] Overall Loss 0.157964 Objective Loss 0.157964 LR 0.000031 Time 0.021279 +2023-10-05 22:16:04,297 - Epoch: [199][ 740/ 1236] Overall Loss 0.157981 Objective Loss 0.157981 LR 0.000031 Time 0.021265 +2023-10-05 22:16:04,504 - Epoch: [199][ 750/ 1236] Overall Loss 0.158019 Objective Loss 0.158019 LR 0.000031 Time 0.021257 +2023-10-05 22:16:04,707 - Epoch: [199][ 760/ 1236] Overall Loss 0.157907 Objective Loss 0.157907 LR 0.000031 Time 0.021244 +2023-10-05 22:16:04,914 - Epoch: [199][ 770/ 1236] Overall Loss 0.158026 Objective Loss 0.158026 LR 0.000031 Time 0.021236 +2023-10-05 22:16:05,117 - Epoch: [199][ 780/ 1236] Overall Loss 0.157876 Objective Loss 0.157876 LR 0.000031 Time 0.021224 +2023-10-05 22:16:05,324 - Epoch: [199][ 790/ 1236] Overall Loss 0.157920 Objective Loss 0.157920 LR 0.000031 Time 0.021216 +2023-10-05 22:16:05,527 - Epoch: [199][ 800/ 1236] Overall Loss 0.157949 Objective Loss 0.157949 LR 0.000031 Time 0.021205 +2023-10-05 22:16:05,734 - Epoch: [199][ 810/ 1236] Overall Loss 0.157708 Objective Loss 0.157708 LR 0.000031 Time 0.021198 +2023-10-05 22:16:05,938 - Epoch: [199][ 820/ 1236] Overall Loss 0.157764 Objective Loss 0.157764 LR 0.000031 Time 0.021188 +2023-10-05 22:16:06,145 - Epoch: [199][ 830/ 1236] Overall Loss 0.157892 Objective Loss 0.157892 LR 0.000031 Time 0.021181 +2023-10-05 22:16:06,348 - Epoch: [199][ 840/ 1236] Overall Loss 0.157908 Objective Loss 0.157908 LR 0.000031 Time 0.021171 +2023-10-05 22:16:06,555 - Epoch: [199][ 850/ 1236] Overall Loss 0.158020 Objective Loss 0.158020 LR 0.000031 Time 0.021165 +2023-10-05 22:16:06,758 - Epoch: [199][ 860/ 1236] Overall Loss 0.157910 Objective Loss 0.157910 LR 0.000031 Time 0.021155 +2023-10-05 22:16:06,965 - Epoch: [199][ 870/ 1236] Overall Loss 0.158074 Objective Loss 0.158074 LR 0.000031 Time 0.021149 +2023-10-05 22:16:07,168 - Epoch: [199][ 880/ 1236] Overall Loss 0.158067 Objective Loss 0.158067 LR 0.000031 Time 0.021139 +2023-10-05 22:16:07,375 - Epoch: [199][ 890/ 1236] Overall Loss 0.157947 Objective Loss 0.157947 LR 0.000031 Time 0.021134 +2023-10-05 22:16:07,578 - Epoch: [199][ 900/ 1236] Overall Loss 0.158403 Objective Loss 0.158403 LR 0.000031 Time 0.021124 +2023-10-05 22:16:07,785 - Epoch: [199][ 910/ 1236] Overall Loss 0.158274 Objective Loss 0.158274 LR 0.000031 Time 0.021119 +2023-10-05 22:16:07,988 - Epoch: [199][ 920/ 1236] Overall Loss 0.158500 Objective Loss 0.158500 LR 0.000031 Time 0.021110 +2023-10-05 22:16:08,195 - Epoch: [199][ 930/ 1236] Overall Loss 0.158296 Objective Loss 0.158296 LR 0.000031 Time 0.021105 +2023-10-05 22:16:08,398 - Epoch: [199][ 940/ 1236] Overall Loss 0.158284 Objective Loss 0.158284 LR 0.000031 Time 0.021096 +2023-10-05 22:16:08,605 - Epoch: [199][ 950/ 1236] Overall Loss 0.158214 Objective Loss 0.158214 LR 0.000031 Time 0.021091 +2023-10-05 22:16:08,808 - Epoch: [199][ 960/ 1236] Overall Loss 0.158184 Objective Loss 0.158184 LR 0.000031 Time 0.021083 +2023-10-05 22:16:09,015 - Epoch: [199][ 970/ 1236] Overall Loss 0.158285 Objective Loss 0.158285 LR 0.000031 Time 0.021078 +2023-10-05 22:16:09,218 - Epoch: [199][ 980/ 1236] Overall Loss 0.158389 Objective Loss 0.158389 LR 0.000031 Time 0.021070 +2023-10-05 22:16:09,425 - Epoch: [199][ 990/ 1236] Overall Loss 0.158466 Objective Loss 0.158466 LR 0.000031 Time 0.021066 +2023-10-05 22:16:09,628 - Epoch: [199][ 1000/ 1236] Overall Loss 0.158549 Objective Loss 0.158549 LR 0.000031 Time 0.021058 +2023-10-05 22:16:09,835 - Epoch: [199][ 1010/ 1236] Overall Loss 0.158402 Objective Loss 0.158402 LR 0.000031 Time 0.021054 +2023-10-05 22:16:10,038 - Epoch: [199][ 1020/ 1236] Overall Loss 0.158526 Objective Loss 0.158526 LR 0.000031 Time 0.021047 +2023-10-05 22:16:10,252 - Epoch: [199][ 1030/ 1236] Overall Loss 0.158576 Objective Loss 0.158576 LR 0.000031 Time 0.021049 +2023-10-05 22:16:10,463 - Epoch: [199][ 1040/ 1236] Overall Loss 0.158550 Objective Loss 0.158550 LR 0.000031 Time 0.021050 +2023-10-05 22:16:10,677 - Epoch: [199][ 1050/ 1236] Overall Loss 0.158590 Objective Loss 0.158590 LR 0.000031 Time 0.021052 +2023-10-05 22:16:10,885 - Epoch: [199][ 1060/ 1236] Overall Loss 0.158654 Objective Loss 0.158654 LR 0.000031 Time 0.021050 +2023-10-05 22:16:11,099 - Epoch: [199][ 1070/ 1236] Overall Loss 0.158652 Objective Loss 0.158652 LR 0.000031 Time 0.021053 +2023-10-05 22:16:11,308 - Epoch: [199][ 1080/ 1236] Overall Loss 0.158701 Objective Loss 0.158701 LR 0.000031 Time 0.021051 +2023-10-05 22:16:11,522 - Epoch: [199][ 1090/ 1236] Overall Loss 0.159079 Objective Loss 0.159079 LR 0.000031 Time 0.021054 +2023-10-05 22:16:11,731 - Epoch: [199][ 1100/ 1236] Overall Loss 0.159208 Objective Loss 0.159208 LR 0.000031 Time 0.021053 +2023-10-05 22:16:11,944 - Epoch: [199][ 1110/ 1236] Overall Loss 0.159167 Objective Loss 0.159167 LR 0.000031 Time 0.021055 +2023-10-05 22:16:12,154 - Epoch: [199][ 1120/ 1236] Overall Loss 0.159313 Objective Loss 0.159313 LR 0.000031 Time 0.021053 +2023-10-05 22:16:12,367 - Epoch: [199][ 1130/ 1236] Overall Loss 0.159428 Objective Loss 0.159428 LR 0.000031 Time 0.021056 +2023-10-05 22:16:12,577 - Epoch: [199][ 1140/ 1236] Overall Loss 0.159405 Objective Loss 0.159405 LR 0.000031 Time 0.021055 +2023-10-05 22:16:12,790 - Epoch: [199][ 1150/ 1236] Overall Loss 0.159413 Objective Loss 0.159413 LR 0.000031 Time 0.021057 +2023-10-05 22:16:12,999 - Epoch: [199][ 1160/ 1236] Overall Loss 0.159504 Objective Loss 0.159504 LR 0.000031 Time 0.021055 +2023-10-05 22:16:13,213 - Epoch: [199][ 1170/ 1236] Overall Loss 0.159466 Objective Loss 0.159466 LR 0.000031 Time 0.021058 +2023-10-05 22:16:13,422 - Epoch: [199][ 1180/ 1236] Overall Loss 0.159400 Objective Loss 0.159400 LR 0.000031 Time 0.021056 +2023-10-05 22:16:13,635 - Epoch: [199][ 1190/ 1236] Overall Loss 0.159331 Objective Loss 0.159331 LR 0.000031 Time 0.021058 +2023-10-05 22:16:13,845 - Epoch: [199][ 1200/ 1236] Overall Loss 0.159521 Objective Loss 0.159521 LR 0.000031 Time 0.021057 +2023-10-05 22:16:14,058 - Epoch: [199][ 1210/ 1236] Overall Loss 0.159603 Objective Loss 0.159603 LR 0.000031 Time 0.021059 +2023-10-05 22:16:14,267 - Epoch: [199][ 1220/ 1236] Overall Loss 0.159683 Objective Loss 0.159683 LR 0.000031 Time 0.021058 +2023-10-05 22:16:14,531 - Epoch: [199][ 1230/ 1236] Overall Loss 0.159739 Objective Loss 0.159739 LR 0.000031 Time 0.021100 +2023-10-05 22:16:14,649 - Epoch: [199][ 1236/ 1236] Overall Loss 0.159808 Objective Loss 0.159808 Top1 91.649695 Top5 98.778004 LR 0.000031 Time 0.021094 +2023-10-05 22:16:14,785 - --- validate (epoch=199)----------- +2023-10-05 22:16:14,786 - 29943 samples (256 per mini-batch) +2023-10-05 22:16:15,238 - Epoch: [199][ 10/ 117] Loss 0.313452 Top1 86.328125 Top5 98.671875 +2023-10-05 22:16:15,386 - Epoch: [199][ 20/ 117] Loss 0.326437 Top1 85.781250 Top5 98.457031 +2023-10-05 22:16:15,531 - Epoch: [199][ 30/ 117] Loss 0.313580 Top1 86.171875 Top5 98.450521 +2023-10-05 22:16:15,678 - Epoch: [199][ 40/ 117] Loss 0.305692 Top1 86.191406 Top5 98.398438 +2023-10-05 22:16:15,825 - Epoch: [199][ 50/ 117] Loss 0.309398 Top1 86.156250 Top5 98.289062 +2023-10-05 22:16:15,972 - Epoch: [199][ 60/ 117] Loss 0.308745 Top1 86.165365 Top5 98.313802 +2023-10-05 22:16:16,119 - Epoch: [199][ 70/ 117] Loss 0.310896 Top1 86.121652 Top5 98.303571 +2023-10-05 22:16:16,270 - Epoch: [199][ 80/ 117] Loss 0.311023 Top1 86.093750 Top5 98.251953 +2023-10-05 22:16:16,416 - Epoch: [199][ 90/ 117] Loss 0.309915 Top1 86.184896 Top5 98.250868 +2023-10-05 22:16:16,565 - Epoch: [199][ 100/ 117] Loss 0.306245 Top1 86.226562 Top5 98.292969 +2023-10-05 22:16:16,719 - Epoch: [199][ 110/ 117] Loss 0.305041 Top1 86.161222 Top5 98.281250 +2023-10-05 22:16:16,805 - Epoch: [199][ 117/ 117] Loss 0.302813 Top1 86.203787 Top5 98.300104 +2023-10-05 22:16:16,954 - ==> Top1: 86.204 Top5: 98.300 Loss: 0.303 + +2023-10-05 22:16:16,955 - ==> Confusion: +[[ 937 1 5 0 4 1 0 0 5 66 1 0 1 2 6 3 2 0 0 0 16] + [ 0 1065 2 0 9 15 1 15 0 0 1 1 0 0 2 4 1 0 8 2 5] + [ 5 3 978 10 1 0 16 6 0 0 6 3 7 1 0 3 0 3 6 4 4] + [ 0 1 15 969 2 2 0 2 1 1 6 1 6 3 28 4 0 5 22 2 19] + [ 18 4 0 0 979 3 1 2 0 9 0 1 0 2 9 3 11 1 1 1 5] + [ 3 34 1 0 4 987 0 17 1 1 6 9 0 16 6 2 4 0 3 3 19] + [ 0 6 23 0 0 0 1130 5 0 0 0 3 1 0 1 5 0 0 2 8 7] + [ 6 17 10 0 1 27 4 1083 2 3 2 8 1 2 0 2 0 0 36 6 8] + [ 19 1 3 0 0 2 1 0 984 36 9 1 1 12 10 3 1 1 4 0 1] + [ 87 1 3 1 2 2 0 1 13 971 0 1 0 18 4 3 1 1 0 2 8] + [ 3 5 9 3 0 0 1 3 11 1 981 5 0 10 4 2 2 0 3 2 8] + [ 1 0 2 0 2 11 0 3 0 1 0 958 18 6 0 5 1 15 0 6 6] + [ 1 1 2 4 1 3 0 2 0 0 2 29 988 2 1 3 4 13 2 4 6] + [ 2 0 1 0 2 3 0 0 9 15 8 1 4 1061 3 2 1 0 0 0 7] + [ 14 1 2 8 6 0 0 0 24 3 1 1 2 3 1010 0 2 2 6 0 16] + [ 1 3 3 0 3 0 1 0 0 0 0 8 9 1 1 1072 14 9 0 7 2] + [ 1 13 1 0 5 3 0 0 1 0 0 4 0 0 4 12 1102 0 0 2 13] + [ 0 0 1 2 1 0 3 0 1 0 0 2 14 2 0 3 0 1004 1 0 4] + [ 0 5 7 21 1 0 1 27 1 0 2 0 2 1 7 0 0 0 983 1 9] + [ 1 2 3 3 1 5 6 9 1 0 3 13 3 0 0 6 9 2 4 1074 7] + [ 108 164 135 53 92 121 32 77 87 67 161 84 298 288 121 55 133 60 126 147 5496]] + +2023-10-05 22:16:16,956 - ==> Best [Top1: 86.217 Top5: 98.250 Sparsity:0.00 Params: 148928 on epoch: 192] +2023-10-05 22:16:16,956 - Saving checkpoint to: logs/2023.10.05-204241/qat_checkpoint.pth.tar +2023-10-05 22:16:16,962 - --- test --------------------- +2023-10-05 22:16:16,962 - 33015 samples (256 per mini-batch) +2023-10-05 22:16:17,438 - Test: [ 10/ 129] Loss 0.353362 Top1 84.960938 Top5 97.929688 +2023-10-05 22:16:17,596 - Test: [ 20/ 129] Loss 0.353319 Top1 85.312500 Top5 97.988281 +2023-10-05 22:16:17,751 - Test: [ 30/ 129] Loss 0.360993 Top1 84.960938 Top5 97.747396 +2023-10-05 22:16:17,908 - Test: [ 40/ 129] Loss 0.363169 Top1 84.716797 Top5 97.714844 +2023-10-05 22:16:18,062 - Test: [ 50/ 129] Loss 0.361178 Top1 84.765625 Top5 97.828125 +2023-10-05 22:16:18,219 - Test: [ 60/ 129] Loss 0.351702 Top1 85.019531 Top5 97.910156 +2023-10-05 22:16:18,374 - Test: [ 70/ 129] Loss 0.351425 Top1 85.044643 Top5 97.834821 +2023-10-05 22:16:18,531 - Test: [ 80/ 129] Loss 0.347386 Top1 85.156250 Top5 97.797852 +2023-10-05 22:16:18,687 - Test: [ 90/ 129] Loss 0.346969 Top1 85.177951 Top5 97.821181 +2023-10-05 22:16:18,844 - Test: [ 100/ 129] Loss 0.343578 Top1 85.218750 Top5 97.835938 +2023-10-05 22:16:18,999 - Test: [ 110/ 129] Loss 0.344207 Top1 85.163352 Top5 97.812500 +2023-10-05 22:16:19,155 - Test: [ 120/ 129] Loss 0.344490 Top1 85.175781 Top5 97.825521 +2023-10-05 22:16:19,280 - Test: [ 129/ 129] Loss 0.341183 Top1 85.258216 Top5 97.849462 +2023-10-05 22:16:19,409 - ==> Top1: 85.258 Top5: 97.849 Loss: 0.341 + +2023-10-05 22:16:19,410 - ==> Confusion: +[[1145 3 4 1 15 2 0 1 6 64 0 2 2 0 12 9 2 4 0 0 3] + [ 2 1083 3 3 14 32 5 27 0 1 2 0 2 5 0 3 9 3 10 0 14] + [ 6 0 1130 4 1 2 37 6 0 1 4 2 4 1 5 4 3 1 7 3 15] + [ 3 4 27 1051 4 4 0 2 1 2 9 0 13 2 31 1 2 5 12 2 13] + [ 12 18 3 0 1153 7 0 1 0 4 2 3 2 0 10 3 7 0 0 0 8] + [ 8 21 2 3 2 1073 4 23 0 1 2 12 4 14 4 3 3 1 1 10 15] + [ 0 1 13 3 0 6 1198 5 0 0 3 4 2 0 0 5 0 3 0 6 8] + [ 4 26 12 3 0 26 5 1077 0 3 1 5 1 1 0 0 0 1 27 15 8] + [ 8 3 0 0 4 8 0 0 1043 56 14 4 1 14 17 3 9 1 1 0 2] + [ 85 1 0 2 4 2 4 2 13 1059 1 0 1 18 1 0 0 2 1 0 10] + [ 4 2 5 17 1 0 2 7 18 6 1087 0 1 16 1 3 2 1 12 7 5] + [ 2 3 1 0 3 29 0 2 0 0 1 1153 34 8 0 9 4 11 0 8 4] + [ 0 0 1 3 0 2 2 0 0 0 0 41 1103 2 1 11 5 19 3 4 18] + [ 2 1 0 0 7 12 0 2 5 18 7 19 6 1096 3 2 5 1 0 3 11] + [ 4 5 1 19 8 0 0 0 27 1 5 1 1 6 1226 2 3 3 5 2 16] + [ 0 2 0 1 1 1 0 0 0 0 0 9 12 0 0 1125 7 10 0 4 10] + [ 1 4 2 2 3 2 3 0 5 0 3 1 6 3 2 14 1153 1 3 3 7] + [ 2 5 2 9 0 3 3 2 8 0 4 19 31 1 1 11 0 1109 3 2 9] + [ 2 9 10 17 0 6 0 36 5 0 6 0 0 1 14 0 0 0 1102 6 10] + [ 0 6 7 0 0 5 13 15 0 0 2 25 11 4 0 6 19 3 0 1128 10] + [ 85 208 182 68 83 141 47 124 50 51 128 102 284 327 94 79 152 65 120 228 5854]] + +2023-10-05 22:16:19,545 - +2023-10-05 22:16:19,546 - Log file for this run: /home/alicangok/Projects/AI8X/train_clean/logs/2023.10.05-204241/2023.10.05-204241.log diff --git a/trained/ai87-kws20_v3-qat8.pth.tar b/trained/ai87-kws20_v3-qat8.pth.tar index 21b5a1fff069cad8e523c5f0fd2c16716e7864fd..ea70b4fda810fb3ec1d78b4a2d9e41180142d6cd 100644 GIT binary patch literal 1824052 zcmb4s1y~f{_dcN#N~na2N{9$32n+KbFc3urU2`o=L_$yyc34nU6az)Y?(XjHLJaKg z?(Po!-I-l<*3ano|L1u;H}<^qo_pq;*|~S_tSywrMGOo~O%49@XKhf*Ad-)Yi%*NG z9v!KP^iJ-X(6FARK?l9ROrs37fqQsS29HTjii%SkD))t(vsQ{GQWFyq9UrAp8%0GW zB_zZ|Y2uT5rIrqilcG}-V!YFm;-g|BqLO;0`Fba% zYLZhm5h-!;u^M&JC`}SG$2%fb6Q7V`9>K?SON~#6j?l#PN=f1)(iGJoym7!fC!X{k|Lsnpit73FR19TVR@P7~2BUX!9O8OJP%D`any)T{e1@=`kT z0Dm72zJUsLX(_plKyIrew-dar?zx$yGxxqLrCS!{O@sLLpQdX!#my)S)VMn0%5c^pmWSJ|*YNk;2ap>c&z^r7*H4x{);% zM%GNr)Lf%(p;EV$PZ5P~SlT#M>4-{&i(_L%00$p+D=ATHfvAm+sI5TMPD>Q7QMXsA zJIK3G;r%-uI{u;&u1Ws>{wyj214yCnB<1NW@O07fL ztz*QwpIS9t#s`)N8dUFG75mL^P!q7(PhL$4?ZM2qYj7B|Hr5-0A{C;}D zKVC;v_+=0Xe;*RARuiO569uM8I;P12(-bY!RE>I?N>CVY#q-WfoHCkXP!nqU!`6kAB%pn(Z5i~Sa=-{2;c$(S%9b)Nf{Rlj7xNkO9jSd zTE^uX^$L}G<=+_PqJPyNjDddeXV(&c^=c{O8i8@Gj&Ys9xL(V+L8IQNQg4!ve!oAX zf3uG4*OgxZ1>*})Z;_I1709;f$hHe)JG5jwHR@d|_3po&xZlyg=NC_5^ph_en0l|2 zW?xf-FpYY@z!S{kpGALeQzu>YAJ7u!Y19W*>O=C;@2?;IhktR3mr9?20KY(DUu)_k zQqp{3Y)5rtJ0^_nxR&aKMtxGHJ|!Ri{(9kmTF3M&{Mo&bpMqT?&q$fh3QXs8Oy>or z3tFa&8ucZW`m($i{Uzalv(PmJU6vGw>0Y8D)k-tVDy&_ z|GPTIU*XUBQy{1CQQwm?-WM1j=olXgjE}U8k2UHiD)rO9G0KJivp*RX0fGKd_^Y2w z8D9vDFLjKs1jg4|#y1-ETb25qeE9qS8UF8eWQE}$Al&sR7}*CY*++rwlaB1OK=wsT z_En?)rc!_Z+ll)f{y%>4{0jd-;YPhcO7pX+fvbi$Ff=d_sC-xs7!T(SSyQ=B{088S zSY0IMjaf~@7hzJx7iBWM!teXMp1o6KW*dAlruNGUZ9)Y42ZGqW18*W6;`fw+H)U-$ zX5NgI+F|qNVpkSloRwM|O0b%Sw_sAmTQVv2!fzWSvk2abY3WQT%oK6f;1dYmS~!Bl zgp#Z+m{5w9IulBZU0J*hE43!rvYLjsV^YQ2GZ|h^&JDkAp7=6MN8bQ;H{%z``67TX zE8rytlw)nd00&m;3@9&lW$_hQsWqS?t7&*gCRMx>li{v%2FTo1@y<+JXNoSj6hwXm z1O)Id!jUAVRAOzx6jxU2OsOn(W$|vT)S6O-)ik_2lPcbWNqJLbGLH9T+WMww@A24e zGGA3VlEjp1tSy-0#Y&wi-eOl4U!9d&Q+!xW!}~I+;uTCvuhzdKL;pUE_hX7WD@?Vw z(0)DvKJZlp@czOvBt~$oEf@h->Wm<Un$DhVmg2D zvm0v`9SXj-0GIFwv9_42by!6(Aedz<%hh1kRF2c;YF$>>8d8tdG<*n?D!x9Ga=EHM zY2h0%wO_XAa@9}a&v8Efd_&<75@SMHTX04gD|OCjBz9%-jajL+LCIJ)uxIceE8X_*Sgcn$VinG<+K-ReW0} z<<5s>qQkdiI{F6q1o{fIMLz`}F5o2wv}bL>fDWwG8PHMe%Hlh*QfokGR@3lZm{jo* zOv)vQ>KR@3m^ znN;y{Ov;-gm#gti`s(((Ic%CWhjQBlk@>K+~1t2~}IEcgw4QmTlq_R?HMVi=^#rI~V)`~u? zrs30>RPlY8l#h+yx!RBE{4${M8#N$+tv>Pn1-OJigSCa*N`3&V2nKZ3ey&BE{1 zOjg$#GKkeQ{9q@;8~b z&XmG0R?bHezydUYza<<=V#;mS7EHOrN}VZp#jY&=9xJt`+-Efn|A0vq|By*}Q{=Ms z5!3#UDZcRcW7+yxIFiJaC#)@)@|2Z2Q=W-kS^RTWYE5~;Y8w6}lPdldlXBVm=X{iZ z%@lQ36lQCHAIGjd-my|=#e1~kMwqR|6+me+!ZY!}Xi??Q_ zmb)aYv3V?$D!w$6;idJsf6q>B*oXSe!aR&4e>Ow66@U_EJJuGM?OCZ~E+ckj@nu=5 zWiH2RYy!)qiZ9Qk^a9CTE;lW$z|?d$6i!Qp)pubnm#-)sLSlm>YYR3wu~KJ)v)Glz zyRcGgLnT&YGgl^6d}Sv8ut6>r-I&^cY+#G>Y}tvgA{;_ugF9;rHh8d7XM?BMmBm+O zrPhXOtj4CSOsaTqCd13lCQ^75&evmI!b4#`gjL+b znSTuU`mCD4HxQJCXV7UedLAjEIYs;{G`v5!77-#YsW3 zS)t(LMa>>kP1%tverH8b5t1N<$Yupwd`jjMMde;nW!ZpJc>j?VNg^;=3Y5(X9Z)YT z)FPIbVr9<_TY(URJ4MvjNcI1xks>fv1g1%WE^@IgS^v?6X>SqPM~ajU2MJP))N~Qo zSBjI(3g#ARR`e5L{iQJ3C@O@BTp1#0fE4tX*!Yzd14T%t6!I6(YqMgIs6SY$|G(TI z2KW#WH&lxI=XgX=mIxXq1<7WGA3t2w%$91(j@<8eR*VoKBc%{~xlr))7Ht_Ns^>`6 zWs``l3MKQSMddM4W!d=j`!k8giokJFpllN9fO<(ZUc^q2Vr5SyTN)5^VWOx%Nvi)p zjTC{CMc@=E&_&*Jg<)2Bt)D6)r%92r(IZ*X5d(X=h?^nB$tDqVi!_O5im+Kym~7A$ z!bGmwB4~~j^p{X7934Mbgv^sd{^EIU63rL&7fAL0mm9%iJC zm3&e8s8m_9DDYdECC@QY^|(}3a5?nUOpa0gt5snYb6{BTq!tr&3j!!1~W@VbODR(3TolekUWwYTrP}|)JTdOyh_JU(m`vPD<`TNEigzOJy%Z;N z>LoZa9zTevk5ZIO+!szIbX1>2&}S(~CSD3bBF`5Q@>L3vje}ot|4jsZmjYyct&RI1 zqIQ8)``?2XgZQV2GBB2`N&RCSBE*m>XYfW+h)mr3@y4QF5viWccqMa=q9UM}6d?2E zpqp`+h{~o?Wtk)q=No3Cs<~8ECJO&E+bAx=N=RWY|4kCPxrT*^w3H%cPF&$k!%Eb) zmTLbG^TZ@6DZ)xgVKPY~nJefbx3q|}k>X_HM1m88*H%Q?Nl`LMQaDx6QQ3>2GE$IC za20|?p0XmOoD?D(G{2I>K?IbS0%U!yO_BK!qO!YGStd!u>4%4?>M2!~iNgO(KB|ha zYEqa?lE_UxyhNn86e)A!3MU@bMQtCc_Wv+XOcGxarjWv9l0-6f&_%AFi1U}?Wa31E z6N8r%QIMi!lB96vpraxY6d(o31Xm$QPrg;Za1!le%>bwSlq1gN9{$?}~PAO=(`QN6WPT@oL{NTpNcHllW0skV*Y zAhp`!k=u#7;Zj}wGx#e;+l!hVq?-D!(*|cpQLU3y?Jpw}L#(r?-bJb|XRD|iA?kLO z=%%K03opWkCcF>JN3svXErnKm0I`{k!lze^Ce4rUCiDvjO-k0pC$hIB`H0^y!SYe! zL8F<}zAY<$#K(xeu|lu1SO(u+{8(U0(ZnS4aje<+{6b^x6UCT5$q|uh-T8Rd`|G)< z__&`v`1bb^p7`=T*k{rQpFLSyctMa)U=`uPr}iB|K9MzbuL<(KSY7+@Gl|vM_C_Yz z_C_Yf*Ee;v5h-e(XMOt2etv$wf&Rj-;}k((^2k-g+9Go*t4NsBShJA1H>+!z`>-0@ z(#Ry+(#T}0(jtTJ$13hUemSAXe+RCm_hEZn`!iiBeFkfbo1+G>ia@V@L5v?L^cQZo z$Yc#I{~%UlI~kc|I~kem@wcH0TRVm_1u0h+Yl~dNSVh7$T<9<4%4Q8M*9cbA@FSU2 z@uQebRoZ0mIjr*Uu`1YRI0b~0JeuiB`Ny!fxM_1Ns|fts*MazPLVw{#u<@*+<)6T6 z8h#>^Dt;1^GGoEkRtSxRPpT{Z!zwX74`K(#EL3IJEYw20)F#1JIs`$lB z$_!Q9G!h_$&=RH~rCQ3`BGochkx(sX%|fabtgfY6$!Z#Y6_YA{HIu1ImkfRltNeSU zB7LB+XJswZmC~!J36sJ6T;zwTso*rb8xG{2nG#l^z-VURL?{NJV&s`yh(rYe0h_|vTN?~w{`;3@)rS*D(0x>EYH ztSw$C&asL>uiLVDUg$5(o(rs@<-f>k8vYWKD*iH)azphGB=-6!%akijLCSTNwMDLL ztRmsMF7y|2-CzwZ*G*Q_@VA&$@wb^wRZ<3jhgJT^ScTVc0)*|UcbTq~{~l{M6*j!x zXBB~8x8d!9&|i2}e8?JF{zt5);U6=p;-4@nKh^*rwig!sQ>Gy0ddAu!*K<~paJ>-v z3%OphhL-CUt7-VxOse=dOr|P>GWfTw@;}BZyzDHzKg+*kx>El4tS#;@`oJmzzqr5X zBWtjrV*86ev6`0pGplL%FHEZVuT1I>Q{2U*V6Q^@DA=|48~a?!@}0FsmLIJ0i=}`y zv@AbaP0M1yZXquIWUCP^OcCxb#37{c7VLx0yd{(F%-e=aBla3m5oJ+jF{O#pRB5I(R~A>6 zP+BN0l~zh?Wl3czWoe}iJ1cfddu17AS!Fq;gR;D`g0iC0QR$>~R=OxFDP5J7m2S!^ zN_VA)(o0 zLAbL^j7;g7BL0X9Ut0XchROe@pG)CwnPxNjw{d-?Z{zZIqT>JhVG-V5RC5!I`tJ|5 z6yCf`eiwu<^DlthM#i!tK=v`B?;mfX^5ysXyK!D=%tFrngc} zT?h-m?ZQ|1JG1nCtcs$Vth*%Z&cA;~g?IcHK+auGB0$z%|9J(;`Cmr=c9)B&ChM+0 zUgN1GYRbAx|79E3zcWi;iK#5A$+}Cj0x5Bq+rI#E?y4dJWZm_jw?N$gW%O@%d5CJV z?)u|J3Qti})?NCqJ5>EUv-G8eYNDE~yCiE-5_ft13n1q%ZxJBtuK#Squl_Hif4j>^ zRFhc<)Bm9--uKVL{eF5VO+_7jy{e_yqJA4~3<<^OkeIVW+Vx+DO8AN`MA z(hxOejs5K)5|v~e^vAU-K-805u?uPS-$@Pp8>jR-RYO#A{jY=MuTnMtt}f@GTB5p) zgZ^`csx4~DI7oQBsr$a-gG4162mNt%sw3*jI7t866#O?%>6NLjsO0)z2gzTT>iu0^ z&Osrfx{QPVb4{u*YRWiBc(hgMpa!CnjD!BT9yJv8WE`Y_H46P3r}SDBCMvo9*Fo}E zqDFsLmvc~KQC-GC|G5e&MNJt836C`j9n?fr(qGe*{*V~oR8){WNnwkg(yLB0@pH+; z6ZW~Vz$yN!72jO^RN{Z>3ajLD+(JfAZuBihJ^dx&Kdvk)8H0Y0x0Uc|s`l-Jm_8bw z-Nv_OT^W2ECf&vR^QcswkLjg}&?F`M@@-jnXmFd+54L`^tth_q&@G1j)|&WPJH5|@ zHx>$^iFI4>!uqJvAYAw%K@lSKwb%9ucd}xi_zQu22cdUgYN~e&rHQF+hs}L!wJEoURn(-r^Qs-MEeXA|%JhVk-{rdDgwkd6LP*UC7m$}Qs?sKnm zpK%={hS9XvnRK+{U^?0K5f^OvgLC;|4bxfS^i9)};=i?^zWrZv=}#VWKC#8XJ?sI6 zNgsGt7>p8?htR2yX_WS;5A{FZiIm^sh>LDdS^e)}XUm(Ya^@k9ynM}-dHj*v6q?T+ zET{*&cJ*OXp(2dBO-1K~g|K(nfTWdYaB5jpU>4=;YQ;1q6sE@hoEZ%FJyMR%pKZ)iQCG3;ykTS!fN0y^!~XWg`7gU)+3IrN+yz`L3fH7 z){jynG_)Zxg`#`$6c?hVquqN^n_dwV?b(SMEmx7%2RBMw=T52=H)?DaO#@GKr3z1( zFFli~PUA${)vhf?9N|ehB#nNmGiX7Enm(?Hrm6M2k;Q}})T5vP=BDpZ&1ERvYMn*H zv$803)qQMS^b!fhUm^QdaSWYcg;xn>k=}AGw`<=FF6P`eu7zR^n!i~D=TloTczrfi zYL-nQ3kJ}Evz5qyohR)+;ZDJgtmwgHbDDJ6jP4rsqS{EMwVF7}m|Yj$8V2Llj6nQo zJBZfL?@w+$(`j>lO$raJPqubpdgo%m{t8sPz8!u24 z%9K6Zj!aXEk|NiDX4kn1-&>_%^w<)QdfB2ZSCkA+7*Of91{9iEkxmz_NSiKHptPn% z$YqK(#m=yxHcQJ=)dDwK+Fn6t4c5XmaUmu+tU#NBHwXxNkLAYi@GANiEEVVB8GaYp z)s)bfx5b(5gD}5WGYYN}M!O4w$uGr)2Dht9n&Upy;q_J=-FXCwndkAg*)yzaaRn!E z4r5i})Hk#vWqyjGts(w2Ah{M9y#?Le5rVAPP?T!a10f%*DYJzg)i)|dyH?-kDt$W2 zaVgii_BY3)NZL%e%vgX4i@zW^v=kZdbD$O>CKTPSB!zCWr@MKz@nVAlk)=)XUDcXG zZ-vv!PHpINoHganC_}6EIg#ZykV7{=YGxfk^Lx$2#;McLs_i(vICQ946UO7hus5(a z_I*!7R>2_DTrmRK6}P~o#zw^7-i$XD18GLtK)UeEpS~sz$EBy4;6D#SjOA=(Y?^?Q zk>k)T<_Q|E`;2RcK4Dty4sP{|uiT;m)zEVJSUhSp88HoV^mxmjpMY`OhNIo|Of*sgTkJKE3!GaH)S)t(ma*bT1jA?(k%fP2Sl(#s!#bYp*A zN?6xFI!!UTJ9Tj!6p`d9cD5-}7UfprPco%EjnBz(nZoARR zY6>dT%ZuU_)u|YH)5kGYY57oh+Fqs_m1}eXVPl^{x#cE)WO|U{LRYdr?@AM{<-#rP z5*n@9i{%rmQovUa+R)vHc78a57mJVK*nwQkEp-kfM!rJO*PqC>JzqOZGZ>c#(=;8gC+!b#koMhm!a(dP6f+K!qJEl-}+mvHS* zF5}Ml<#Snf!5HhB^gN2fNnvv$`)fCi>yYOBLcS=7F1e5EiFSqEr6~INe|_ zR#w`A@FSPdD(WsS5BLgYX?uD)(1P5{I?>nz?-Kt$Xdq zwHKQa?0E_UpBU2X)8>>}U`xsmZ85b^I@XUGg=GuM;cTuQt~GJOQyV)PbK9Oe*}0K* z5WAjSs7V3UYLb7S`RG4q0t&V*fcKhGboZPMRX*TG&LO+ef8znv9&kx74vTmhVQk;m zoa<`?OpGl>SL->`cZX`!;IbDz`pW)QuS!18OlZl2QgpI}Ee*4{3%_RfQTE%j2{)A!M}(x?r{d6^@%^RGy?-v!Xkz4fVJf|3@k3#TDbe$;AZZQ8oCImKRC zjKw3?!*}m|IKG~Q)+gqo&4*Q3U#29@3@cB*eJWCO*HoA_OF#$H0eG@YfsZde(f(~M zVEiVWowgMPZSTSSoj;`ysY5R#TTo-CXK-_Vg{W>t>73>nW}JJB$`zmDOs#8ZHsTJ> zjk$=Ip24JW3ZtTPn$sma1@+hvNU=Se&}KS|?g1went2QnQyu7F2|HR5;6xdwMX1L1 zVzi`fDY`Vc42|vLOmUwa>FCB6xZzWTdN%xyrv099Y3GgM$lc>|Iu@e|Q!S{m#*zY; zD{1DQ=G3sDIW6lj27K2Eh_jyrWm;p}Fti1Ae-=Xn%T^@o7<;m8>Y{he*=}`+dx8s` zQSp3kaFI{Ae#@9rtxHq#n6U_dG7Fb87a_?ofELfKPY=UF>GWv@`uX@_rkypGUuZz5 zZ?&TJ<2q67J7c<F89M*uFJ-2oY|9kuwJtW6OK$nX#Hel zW{<$k>Z37kbp=|KQ<=^!tU&!f4aP_3(b&B+2PclIMu-saF**I{d08#kQzR<{`tO3LS~> z*N36O*?s7F?jqU^dje-iQ=HpV3ELjl#%Py()ET!L)i&?M&2ek7`^hRSxUvaV5sCX1 zd%>}C5A^JX$PL^+BxAp~QjxnS`{SD}-_ct`cTbxS`MEUM@QR{GhTx`$^5eDsXkY;-K;Z z{14qgjdRs$+FUp4l;uiWjGn;y)(tFv^a$gh#p9=AN1UJD8gyzB#ygHjw)tc|pD&!d z1Y6ZBQ2xYvJX&Lc)5mP^^_&xOmOqB4^GoFVUWdi|pV;j50ZrGuz@?_fC>>ZDZEJty zN_k|XcK&#L>ch&T%ki1{bH(>ti0`T#>slczFxv_lMz2k3`h@mI6)F(sZze zExoj|q4}Gy!Med=G;F&U{(~a0v1l@)g3~eNP!aM_e!>fDL-JZ_hyHn$FvHveU309c z?6?wS`@)i{RPK*sJqDx0olI1lahYpz{t@R>{tow|NeXgRQ5Z995IlDd$NSnNu*6{u zT&-SUVD1z2K7JoxtiRyh7X$h*{2RXS*@s1Q_G048U8t7%0w)!35#nw{YigY1UTu58 zeM`B>WmbKIlMZhXweuOe^*Dhmz0Tp{=W}{Kx!kD@#?EezynS^rthF`X-!;IEg;wa& z`Wcto>L<78)_ZQ@nj~yh#$wlj#%S5(47`gR!I;=r7_(;;^5SO0d<)7mU0ZD}_F+aHXCG1n8{*NsGw(@0dFn2AxY=5*7< zgwoy2sj@`~jL!w(YJMCx6|G65KYP*OU0!seZ!{&0h$n{!5wyNxH|(7s0grBJ_*Qud z?lc&OJ7vb;(SngQ&u}CiY&L>+mr15Cb$d|8D~aM%6ro4RY9aQV%uc$yOM!rJk_7om7FU_W9YY9 z7&*}uHrDBM_*`G=)LTtow^gOAAwD#6N&pqP<4aZFc+w_E1=;*KiAS@JA?xN*J)ak4 z^WL#%UtpFQo4?$vMJhH=>CiBU)*Nm_dGkW4NS`2DR1ieP69UO8%7sq%?Z$p!Q;ao6q)TtePgT8#I0{&v8Vu-l8dn47FAJt=ccr6Q$4-!owX&V z!|vip%+d76#hUk!Uh5GiFTRKnHs|iwA&9OyINDd{YqNjIGC=@R#Tll88p*6P0uF@H9l~=pIP9RQ_*bU25h^Zj!tI>qh_^{aC~?eo{rlwp>rMv zo^V3e7gto8S`9sx3`Nl=Ss0u<7*o!i$AhBhu&CTK6nvh;_1Ha`doXVW_qA<1WZ%=E z_xJ9oQ_}^No)*K?oo2{gISm2dhvQ|j;pnoYGCC|MgL|DTW8%A-=)%4q?WViHC1Ns{ zTxA%yuIeQ2^r3zhPgSy;h zU<>AbSb|EX3o+&TC~W035pR}^$3M<M@G%*2NU-xDVfiMn87oSaxsNYkHx|+$8MM%tENaJUr<<7q#yn=4$xt z<*Fnc<}L)}!Eft+8Qw{bOB;n-I0)>I{Cm5f0-=Xh{O)}u@MQtWYw6PpciqDhI)nKBd05%>9C9kgVp4uxgf9<6 z#@!8Ax@j{u#UIAS`c2SclrwB%D#57AcJAKUZQOYCTyDkjc4$+?4>_DSLLvuop_7Jl z{T|KX`sU0=(!?o9IzJhWO0CAlek)<*x(*v!j)42bk(jW4j-F3aSNDf+r*?SOvlr~g zEXJp+qhRDT78h^Yp!m%4I1^@z@X-FK@?a2#7-pg4-c4w6VJ-aU?L@?iyT~|q8JAri z;r_@*G^uY(s#mNnC5HDw-(#6@HXVYm-gW84bGA;ou`RXeY){kPnNdgUahT-(0aV#0uBjPI8pwFK|$kihf zRlGW5>-$vXohnY1-AYjMvl3L=FOIqoOeCj^ed+y~Runx)MalymXoBZZxD6VHI?bnH z*2a8v8FdqD9-hI7ehqLxFA8tWqcQONWOffY5sQ7$EKz0v_>X1db5L$(xO zP)_e!_iFTe0d>Jmhu#@}Hd{1o|3_u@)7Z(2H|5-nNgMAiq+ zVb#+Ec<7#wfF)tnZfZlyo*zit@;-9M?|tCRHkhGQRsgk#tU=LT!${qtKGoh_mz>gT zQb_Ux41DkkS3{p+hv_Ly=x`awi@ij4P#2n$*PRwr?Mek_noz6nt?2ohPULXk0r$pu z<8Yz^8`efr_?t*N`XrKKMmL6Q^Y%D7sV!`?U1`stid5CBEM1HLiQzr(n%v<$n0?mS1)P(oJr&z*OXdTe$icN0&AQW^9#n0i=aVXDuW&QR9OJx7 z?Bwn_HKkmW4%B{2S2{P(8mHU4<8xCtc8~QO8_QjS`S24sTHb;rR@Sqne`p*H@(G!uHVr5rxA5IXhrqnD$p_0Q0kJ{oW?&^Qs&Va zI21P%U-DMqn6U?yg9q&k@u4|aj$+oQqewq?9dnPwkk^)Yava@@rt~XHmmivv(HBda zQMo#uDdR&Ik~}H%v=Y8yKESrxc$sTSS2|eJ;eFOrYi0oHksE!?llz*s+esYv*>jd#fGt`!qwYhYv-V6YX3UL_q`nseWor z`g*YjZ5S1XVdFxuWUn_o90pVN&n)`dVkjNo*O2ZkYDG&bDrx9oYnm8Yj;#$yKq(UhidYEM@m zs%YZ&D$qPDjhre^xUgfbX?1WkHOOd3=_yLg=+PQ}25ERy^E(&JnZjt4KSIY;!K3z0 zU`u&=c{gMrQ4718)UCjSOdMKLYrYMAV=;Rx*@MRatW4uyxzMyKFJK?^9BDOfV0kSs zDqG5v`1+OT)|(Y@nm->S*RMu&rz+&txjJ1M>PUWZ{&e_EJzAa;N{tLw;?%JfnDl%y zd@ESuV2cu%-rpX3TvRk8Ig-w|?m_Xp0}*?*DW+@)f%UzQ2)}7SZrO(9+;c4!Y|X=S z+JYX-_i*Ne4{+sYKH$ctbs$5_c2woNiYl@^wLW4*pO=-O5~g>!G_QBu)2C0kkRc)T z$f!QOA5x#vJD$XPi=#Mjem_p+wxa!Z?I?UuTjHwJqwGte)NxJ|y>oH1?M3W9`WU6Y zJVPv%!lTmeIC-xrtjc=PsZHKwXH|_}kI2TbRU+GJ*1I0`GiExLyjlda(mSx0t@-?D?@R|SDyU1TnPhQy9t|uqgGSUH zk4pZN@Xly5GU^SWX8lLe+30cfEU`Ggd9yY4*R}D+X+AX%Sx!;I7E!Oj?pR^k8;h1q z#=^;JntM8p?t6BlhU#{B^RXE=9vF_0w+RT3>0R$#iql!aZb-Y{c2GfdG2l@PdOp8A7>IPTI@NRt~Q;?*2LL*QSdsf{$Ysr8P53BG#L3W zeqzn^x8Q7z^>`~R$R@`Uqse4sI>ip3N@sxo&5)v=mL&8Owix7D(n#wvw7T{{>=`i@gJw>E zoy7(2;+CV_=`{~Ii~JyjjBANDCnIqnd^T;av49%vnL?*7ccxC42hqeiLn*eyE;#%= z2J;oC@NDfgnz?2kdETE+0r9U;*5y0)mM%gAJQmRNMJs99iDmSnb`Eh%rqZO6)Af9^ z`#M`EbgWOEwl<>BdCj?&Ufx`_^d4NR4+&IIaVT95WzYBSx|78-cD;2rrSk7iaSP|~ z<&M06#GP(&0EujE>5k_{EZlBGJ(J3k!7_X5eeEn(*`G(l9;eXjPGz)63&5$&4hZ;Z zLYtMQ^en@SGS|F8`>{{(W7b1d%WsLOkS3UJ(G-Z)T0)xw;Y33<<4Xuf(FB>tMpxX5!4=;Lg3z zh>rP;z@hB<%gMu7I^z%)vvsz+K5j6$UlJc?)u3r93Yuljf zC>2p1ISl(R>yUSQJ-pNRVSktr?czSb`I9=I-yuFLlg1e?vuJjuWto96XZ@=mJ zBsOyb&Unti-ml|OWc?<@81KZgfLyFhVryKsmys5E2b;G>A=af2PPOd@mt*&^r`1g) ze7TI^=;BmrYzeA*&w|$OW$Rr7oY|Z?0A(f?rP*6c(vwJQs&~qkKD4u;j&&+g*y*`k z_S~sli2Y=)(a(}JD$t%D_Ohnl10869Ng0~C$d#rac>}wpj}dq4DfU}*#k#OmEV|bR z#k<_XnN=@PtEwT@cDR7Z^aHpsejC2~Zs06VF6W{yF6Q!F1E|dB+7xXRLNC^pM4J+& zk<_v}D*bdMr}LKd@wqMC$bF74J6|K^y%Cj-8-qEa$ss`_IelCAgPyv1gq%~gEpe-;%_B<$qVY}8&7cVu zwrYt2wvL-V|0O({7*oevCo!$+C3HG@7t!?;m_5w}!5>}FXM8Xv#e*_p!f5&H+g$pi zgWSj+7r2tIU1<55G8F7vj)JW0XhbCkx^}8E8CH9VzKvgCNXM5rZgdP6M_z;5*W<|D zw*nsJ)}gFoC%l#!kl#5II>CHyr0}GD->T4!4eoU3a3gwpzAkODZ%P|?3_r6l%3~9Pw#tB^(s?fv3E8qoYr5553{+1B|~n*ZPRTGV|zzzZxp5O>>Abn?mV>X zybS$MuEY(K>ln4^4lX{shnPOMQSR<(wCa2pv%YzeLl18n(AbZTXeMI8og5qqX7i|8 zYv8tKIaH3@5L&bY8q{r%j#m@)&h?_z#puaJ6FTyc?N4!Z!=QQ9qc&XP=hQ3Do7iB4_ zvlG3Z?M~PI9Z(~o2rk}Xw+sm7GIf#ZB|P$Bx)HBR#8xju@@C^fplN-8Qr%SlS{2H z7}7EZrBBB}Y1tpMA6U_?=4B{pz9X%!+X$CB`olW18mc{2gjb z%BbZ_M_N5$&nI4C#smWz#MZ`U9$CrOz;_~vJzIZraxw-5t$^8nbLw*3gs#uAqBE^+ zX}xD@n#|74t7)Dz>wXnlZ0AMYYE;MM`+mr7?}MEY^Dz&r5Z+-Y29AA)fEk})fAK5& zy#IzTZUtEQn#H7JN%AjVlHxDgP|7f8)E`k1<)@ay0MsW{dMGuH4`;?||$%j$IWSymcjn>wN7Dt|qnZ_1g@@@hN=ubqnc1XJ3oHlqelOsVXu2QW^% zg`}6KKoK+W!EF`FTP?s(gQeWGXG^%JQzmokLyA-EG)pSM?(LepJ;p`5>sWH&1HK1N zMXxXO;GVh^hhFAk&8g$CPu1Ud;XBJ29m;s&o~;{OBRvG1ID>sz4^eL445T?tM(fZ~ zXi)JxDh+sykzMX%@uW;d7N3gqi4)=6<~xQ(v0Qy@Mk#C`L|SW8y0*leQmU1ve*5jH zTfK_xUeF!I+1{w2oC+usvkBElEySWLGhxD4rUG^kotsyU%+g9>+iW`=ZdVdL=Nv@O zf#=ZN_5!MZSi{{?Zs)F6*~c|%{FJM7@fP>>{!MOT>1vcVj_r3_?n^H3tmt*QlGN*y z4b8eenj11bo4aJp z0^1MMzLehmf3Y4Fu<4C2J~&l^!xkHqh_=J)4%IQK%r|TcFrWj&iqrhmV_2Pg9*t(~ zz@E|Vv3PAbPRtkx<4-HG|KvJ+sJsm)s4Q8wv!XJmnQflTe>pd?Y{f-vpZ*3pv%X;B z22)C?=Sc&5yHfkGsx*3EA`Gha#GU&s&}M2cIDLwRN1u2Mw7iJLarfby_5xGhx1!;F z>d`?zB^|RcqqrWHRPu{0o%i>kpqz@dq`V`o3Eqn_k9Q!s(-zoHt3k(-g2{VpV`|cj zqv77wssGcOboR?DgatfDhwR57IMaw$PBf&L2aS346U(xVXwoDjYQmm-9_V!sO`hFB zvD#KNWrqcwN-RTDs0BQRbVh+K+pFeYjMhE*fwz~QV~}Y$HB4?os)8`8|Hh7r4`+LI zA|1(gkvrursz#}+eCf8Y17(-DqIIk6^n5<%M^)IFR)>oH2b@ur}{9pwjZPWPI6*VmBW|k1Bi6KV~aDw*}BrdyZb!s7LQC6R^xV7Di!- zcxk3)`=EwnuvZ#3+Z3U5>|S3v&X^|6a7QJJKpb7!2%#|nbj^kRf6cc3RC-D~+$yMz z^o;gs_+t+j?tY(BCg0;C%bHNg1{12&)r`{Ky+v#GeAw#41C*+5M=z(D(V$W#$o|zl zjES6uE>6q!u0{8XZARIUZD@3P2~I_q$0s$1dRtoH!s~f>|7kWBOvpjwzAgxVZwd6i z&zmhX zeD*nrv)T8#KIa~AWzRq1Rvb70E4IhuQ?Z*c`&0=pa=&wVcP!Aq>q^Y{JRAL|?nJSE z*=T=b5)N*fgW~SnxqYUGxwaNxx%zX?a$9da=DLSHmqxIs@uxSpOn@)tQ@k~V4 zAvC&E1HeP1NrZ|fy#ErnSlvdeKZ!mV}~HmAQ_`CJ78kPYS??bI^uU9L&EWk*tU~Bt7?>k z)mMgLcLduDWKfztLn{y8w^o>5bR|r!HsElNttj4PGKNKDV%(r{c$ahxNx5fm;Myg4 zE!l&5HMU?t?Ssf-&v0u5t;E&r1K6G&gGYT@V&KboOfGwzYkch_*J1QFZos$87;)ni zEWckvjqdBwgUd(nkteZwL>6wbJs_dEL*PAeC;INsN22lyc3Nj)#?MqZ+h*e0&Usw> z(u=tPFXwSZ9$e!bKi=bfyM55}NfRVuzejVN&g+h%O(L-ER70#y(x7qPI2fFqiM`G% zar*x_JJYzDzUcplRFX;*rKA*+28znmpk$^ZI(wav zP^1(MW(gT8B_i^_zlY!7gWrSy>%Pv@d-iLez0X?fv)-S59A^@{)NRDfa0|KBuYd$! zzeXINRg=zQN-pGhl4uhzl9x4-#3rhf-NiaYEagK&b=rwxWh?Pqu19@0+wCw*)?A&_D?>wH= z%qt~FE2Q)F_aRxexrz+f_Ld|_WBGjb{Y2$aGMRIygw&rZA+Mj_C-F53M9a;9cn*{! zTIUNzch-LrS>!5`MAuzpiFB@)^d^y7t7bALuazV?wvxu|VO))@0(W(l5;ymFB9Y{$ zlS%gHNWzViQh=R9mGBcdHOAe>$4`Y-eyJW7KjM9T848>?aL)kkmU|726Ft* zMDkG0oP2i~PR^atAob2hWTr7=={mgG2mAtuqSWNu5V$l~I2kvq1FZk`Ar z@8UKQPsjCS4-O$E)PO+ti2pnbpNnTiwa=qOWfiF+lapu31?z6A1k&3EWXSs*vi?8@xjp_8IcD2N9;7@YUeccU zx#=0?#rtrnmXb(H>b8>ZGeIQXGlDeTQX}4mYUFFGHn}=Ho~$=HKzuXf$s+&$WUPK4 z^5gp_(dv=SWSMduF_gR}BZG59(^F$aKfRAgImlL`JM9svpI0aC50sGAxs05p_ej{G z6{PODGqLNS#7W9-t}99*+f&m>&)i5dx;TQ^+r*HL`bg4cxsP0^-b?()H%Pg+a`M;c z0qH+Sh8sSwgJ?hLA+3_DB0Jqzq6^=eMeV*7= zS#+M{Cbg5%r60*c(MNJ8={~{MeDZWcg_MilM_wKGBb8$#i2jEL(dghdkwaSTf7bt& z-YQ~q^A<795=bOJh)gJM5v_aqMKn7nlDyTAB~}G-q`qqiVQ+>J{fFJ6-6yh1e8?HH ze##|sX15DzaUsN~bQ~#ucb%x_ULwPHUL?obk>u9hB`GRAx%c@k(4D~e~*0fd_b;M@ubQ0DhdByKtz>W$;n4+$oUD2 z$?(OS30<&{xUAksybrgNtnjzwhx%u7C(MORZnGg9nk~tKniV7{dj?q@J(ukIQ$lpl zln|@nd*s~u!^Bm}?ahsdBYT&g5S1=V6dkKi65YQi-FHi-ky$0P|FizTNMnA8Ni^{| z8cW8_T1_hMO(lCDNXW;}Z$(~>6(Zj!<)YxQ{X}lz2{I-nnN&S>m1^#0MAl(4(F^b= zFNSU)&-X1S{|*!G*K^>Cfy*^^N(L}X*J39+k}=hUYT<2;8Na{FD>xLMDJaQRL; zT+xge@?y|oGIYvGa(m`2;&uECsi9S*Mmkp;&VD01rM{j`QqK3cuNT*Oe-`KLsKzBu z)#h^COt~w&>dCw{jYP`&E#L0LS^HXXB5B_E{wK#d zR0NZ><8z4)&Lt^Jl(=sTRJdQ=gQcuS9XVF|h-f8O5WUd(WaJcAGJTgV`O%}qsWlAc zHf_`8ek<)HlUss_-H=f7SE~K47%t7VA4wTRO$}-9RE1NSrp%>09mFjfrNxy6nsUZN zB-~Le4{kxf8C-k(XENx(8xo}7{+~HNeoddW&66_U7sioOw-mUYw+3-_vMSua^A|{? zaRIRaDHF2SiW`&Uz*Ww-<<|WCD(Y)oE3&`Gi{`r#?(9Z8&UL>7w@KnnRvmOF`%I^i zryDc~^OYkLrSBO2fE`iWDI%gdj{muDPJk{)N=&&4H}yFARtrw%ksY^ph834GT9y`oid2Be>nA~8@KLuRj)dfC>#ATtYU$qVIla%g2TneKmxSm&=Gh5Htg>A7=> zujfy4Mnjgf%u?j)R}SH-+Jf3k;^nec$?rw7ZNB109GZ>EV0keK zUgu6W#7rW5vpF$%r_Q}^)Zo%mMskDF>^a?bJMOiIh%;Psm|W03Lt-MX5fYg}4jNx0 zYoA^wqxN1PfoTP#O|OciE|Vpx(-g?1u|vsi`8MLv-9eJC^+;<2sUCddD`_n3CFeR0 zk-`c4NbTKp@<6?c3|sPwT-E55#;;?_k1ZCu5q2bQ+i0uw%3w5^MUA}2D0jp zHR+n6%$bcI%sr3Q;@<0OlkNO?a%QD7vFh7Q9Ol=P>%NWTO~pS^gP#I%dfbQn)X5`% zK4+20qXpzo*HrSZdk0M$VYsmi1EyQQqSMoJPiR+##)qUGkxMC@LMJqIjnv)Wdk&Yqp%Qcc)SxCC# zrjqO-Ze)ANB=WT1ezM|p3^^4NOX^=PCmHn%$>WUOWQOT6QeB%v!rjjP=f2CrqDg6= zXsJgkimY!OPaY*qB-_V1lb-uOMG-q2L~XXOMIEc}kk@@46Wy*_qMkO8WEScZt802> zX+RHown2`YEjNhk>b@vi?UgCITb(Id^!_?o@}P#yNNFVdZ!zL6bDNwQ@qm<0EhaZO zp8RsTMb>zUh+dJH%u#Y8jqa@^E22Xhw`I8Qf@-25TTirQJBZA_F45+VPemn84WgMU zQtffxE3$RW2U54cfmlDQAv@PJkmil`$U+ z=sxnV_nYW;*)!3ObGf3gSI?4L2YK@SgS6N1uaE?suOn7*U;i^FA6bFu-gX*d_9OSOHPaxJ%C46S7T#BgWDkM15}=aoJu$s-j+z?HX^1;;(Nc z%(D+y`bXL)a_+-<)=E8CGFOS=x>IEFn;2rY;T1XguZ4V;{@lKg^w>V!j$;F*45~Wk^KCe{WGTl@`ewuB+L&^myluGfBNMq> zb(DK|Fq=Gmm`Nr}8S$T>#eLOR=PV6nIkDnIu5p12r|?|NjrwlDz0(=ZwH(&vswL{& z>(4_tU$w!Ut*0)hS~h}fzNN`EM2_G_{nqESkBsK54-DfvsSc+RJdo>q;01ZO_&1Rm z--mm>%80wP(Si$5x8^LXthmm{(qpr)I1w{lKaoSMlV(`j1-iAg;7#7&rgkaL!A63^yibJXg4740kY2hO6xN zk6fQF)!wY#xYiTy+`o1kF4=B4XEIsp3rUgoQwYb^Dcf>O+{bgbrA+>Oks4Po)oRW5 z4I_O9n3Fj%e&plHtK>y{g>>Eim%IL2G?3H0Gn(6TRFnJlWGpANiMa1yEV)A}n%tfz zD%^?(YMheX2<{)z<}OYd%x(T~hRA0>B2Sidk@zRh+=#bs-1E6k+{N1lT;PfEoY+){ zyR==4E0odU-c1sY?Z2$KdM0Uj5##TB8 zoByvdJjomm*}r4pm0dUOxW5PnMYzJB@GETKl1Fq)ejXEMtu4M2YzSDo40I1gu;Yq$ zOmlk=otXHYN{t1q z=|Sj|Vc_@ds<>?HCN}Y2BHnpx$p3aR#qvvH=zHB8o}7w-r{7D^*OCabLtbOjpx119 z)j5Yzv!Br)YI4~3>K;tKJP`42Ftq-iK%E>$y2)U-#ilvz+<7 zf>{o?;-cv2ja|HNwGX=%qJsk!KI7=ApRvVj9QI#$gbt6pKrec{cbG7(SR9!3g>LIg zrh-8}KlZ*hTY1)#eto$ZKFzWLne_S8{HqxLHJ=5G`=i0eVI})L%bt2ijb!tk<7v*K zWZqZ2ihp0IB-x**DN!)L3JMozVqn5-w)x5kG>n)7hk}puixZ--X!3EoqwNSCmaGSJ zCtkv9)rOd(76C;EUop#MBmAMQjEeKs1ed~AynD_IOUi87x)(#KL;EOkxBOs8SZE7A z(I@Df>~KstW`g1SyO`$JJ+NS;8egPznx*F4X8AXV0c`E1=XROnY`al3Wz_|iB=Z&? zy*mTxGc6$bz76Z^a*sdlBoC@a_t?I!Tl~o-W9SvjmGtA5F)-v*Et}|fjy{zAP7lKb zd{>r+4Rnz3h_8PiYnzF zq1_4@O!QX)SomZ%;@JtRDdO?=s_7_e)#MOoETZvqM_|2sf>>y5rpta!0{(s()mYF- zmp!)zgWNXi-`sD=IQuM9b5T98pT9CmwE!)ooLAV0tvhJ}Qp+OB=5)bN@g zyfYd${`gF_jc-!*l(T3zEgFsTAK|cEQ&yHNLv`xMv-a9pSobazUSCK0=43p&lz+kY zBQMd&eLB3DXbpF#4!{rN7cjQGhFNa3wBORi!M4o;EK*!1#snYyJQ+Z@D1{Zya-t7N zUp(u$2j$vx@yX~Ito}3~Cp5KU#gs4%y*L<^eoTbYr6G{yF`Y(Dor@aWSv-@LjuO4@ z;@Yq$thnE1ocbt|eyAJ?h95k!xBd{DzAKXkj#r~Dr=+f^CL?HhCj$yD3&7RK33fTk zK;cIfobhxm)$WaBi>LI58hVN;x8yO$SBJ&hXUw3N^jx7uGZZ&9Kfvcvwy5E^6kER7 z;+^d(Y{k**4lctbILPhijWK#|kiK_0xN4Vy&y5NuDNzAC+bFm;R~@zw;`wpbPFNfr z#SHcl4EG*}dCyzuplNnsJ#_>9>2rYI+47r?d1VEnTou?k*IS&L?uMolj=*{n2j`Di zLeQBKDE?i;AZvhR;=EeuimQf6pPtdU`oqX-x8Qkm6`?Tl80NK2<6ASbAYjL0%zGA$ zE)VZuwDWov>!nI}we=OYtUW>H9?!+OJGZgk$8F+UKAN!QQxx34b(uBSD+~Q*7@$Vm zAi*=lgDM>x!C<$%5bQ8sc)e3!7<)mDVR{jd8qG9d6A{*g848v&B7{=Et5|d(6TR{r zP$uIR)<5$GN9EmU_qRWUPjGMKGuJ`m zj95%*N~N`$MQGW)52y8OX5BNzc+lPtF4WsW;PnXT<(06fp$OkRLELDdCsf4_70OgT zvU`zZSrt1*P3@0Blfo)EHG3qx&(^>y$u-({wjkbB*9~i5TcPpY)wnq95k2(e2|L^R z1M`dyVt;K3e9`-YeDDm^d;b)A9a3q6?p{dNx=hDC@_>|m=b-q?8v0$OnAiSy5dN6w ziUUIpXhYv8a9d4X*m?UY&YaLkC~C;S?g@wRTS^=Y4-UuOnIR}~nT%)k6yWJ;PcX=C zf<5JN?B=jYcC@E2uFy)tU7xqY>{3R(`wqi1qN{)=>XN|&ok4X%E=>Kd49*3KV%OvS zC33avA^+zF__O0OB(MJnzq-^UVf)LWy3q+j^z-PSO-eW;@gwarJ&x0>Z{y&xGJ^a2 zNc3}47tZ#?u=TosnS9_NxZ!_Q{3|V*2B&_b$qgygS@9&5@wi9tSe*r7)ey*?@)TA- zc5}%6VZpyMu0+37SuCm!$9IL!nEHIA;5hLOD?A^M{T7bJr4|pdVEiv?i-{l`egxh= z-cK)87=rfMMCLUXAn0K->c-wcjZeYU!J>mkrCn!MYR56}x(?4t}nu4^zi?L+WaH3^)0 zdiWSCEy;qv8t9rj7geUNcGz=!sKn-Lh4<`-)^_jW?zvmQc)SLn@_d?#tq>E0|mF{hCHyWN=2y-r0Lw404 zTHfIX30E~hH*O2tQ8P%A@2DW@JljEST+c!QoS~lzpMz{+I@JC#7M3}EVcXI&F*N5T z+wIzfhkSpc{@y0g=#Bz8+kuit%ZCcVWw)^Z{y2Q0BEaAs{a{`BAZWX?0Vl86iE1;B z;7{2cI^Inq~Vh5Itr+j z{|Z0!kriy(-$PJ#9DDGp8JO~9*5_0!zu(IhCfpk;1U51BT5t^=rjL{)t6NF_hKVI3 zvtF@2S56gAExAqaUC01KwY$t?rwZg0-KG6(;^`f458U5hR`7R8#FGo=;Fcf#S=4DS zT3x!2{-ef%chL(>ku1Xl)ACttiwW+FcHz6yq8#q#WU&z!{dsOZ6F>I~rLVsv;P$cO zgsZ#__u`L^&EkI}r@(xl3brxYj8)G2fvIsC_-x%odbhZq{#A?x8CNxMk;af!uTvmz zJHtSYk96Bt5ErOc(T_&?sF|K!d~lUH%le@&bgr}zwl`FY`}BXtTi+|Ge8_s)E9Khf56z(HJJX=OsHBMfm0?W?xQD$z&~_vALmWh=Hp7M(bFfcS0yWQ-bmre=R0`SwPN$wSarH&IEJOv` z-d#rEB`f5(|3GDW_g+et3JS7_59I(|(dnP+z=|j*8P}PnIdc)(v;z_@nJml@*8k77I5P zCoIH5<3i}?z8dRGGvV*~{sp zpL-ke%3>vf-(HS>17!uZaWAlPW1ING*!OhLV@up;t_#Wr$B_=X&D!Rb;+Mrd9UO8B z8#OF31E* zceGE0E!kS&wQDx8uPb3$dkq-*zeE??*L1Jz6zCrEm-)q?f`&oyRQ_7YjW#|4r~WR& zf4%3}s(!KT;-;~v&^aGY$jJ+n3zP8Qx)Gqi$MpKj@iDNw^%Fg%aE-=Ce4~HYhvJB- zn^>Oz3XsG`!C#QW?12NJ?&B9wB5APs!%LVgvBJG;cGLT1yTD+^LSfw-O`MgWg_FY1 z7rX544}m-5;q#gvq$CaZkIZ0hKbB(Q$X&wkK~Gqm)*AF0V1XMWKH|U`a(Lz-!SAng zU~xC-!a9h_uivPngxmv)>6yK`K)xe9GIjZ!NZk(g|Rbl(SYY4 z`3ElLn7FjBaQ1ctJvzq4m{Yc3Zvu<(W!L8t-w&Bxd$CzGCA$2yI z0cnRlVXj{j>-`==x1QL?FKsOs>rZ__XSbXNlW;SPoV|puo$HTT3$;<)bctP``y7jg zN5B*Bc`&;FWeD#(3qGF9#fzRdP}P3{jxLF0ou3Sa>$0gV*i}JtbMs@|s7Ki1QWLN} zG>_RRM^iax+w1a^vtj&ZWg(}#pD<=g6+V>{3so!|o4ZbmC41Fb<8Ny)d!Jp5DdQn) zjU2S!uA=4-mGENxBJj=+WmoD9crX4EzIj>?|19m;iwk{V_VA^`(ao7~T{(kRTaSf@KCv)1 z=`%=3&~ANAgNTtC*mkrTz3{$1T2@v`(sQq zwGVFjGJ&?I_Z4P^q`=xM4R}5N2b0l04u7i7iL1sjaX?ls@_%1|{v0#7J5*j$A8?)? zcDX=1HtS2Kx820cYrF6vuL6&Ho50xm7VIo;L3NKdG`;r(E-#ypYxffR@}fUxSWjmY zj{cJ7nz~GLP%=7pFB4zWSP1!ggUw_ZcdUqiH`Bl2SJwf%)XiKD@C z$vqsJwgiivlUX)d$p;L-L(3)8AX3Xt`v3DF>(VhOz19btx?@<~yjQgKj}LAQu13GE z^HgK3m82}I2)xFAfE&hhA=_*#tcrenqwd54G$}EM<%>$_-5&~=^zatW3p|8MsVy|D z{yto@@f0`EZP4=kC*3`675HsR$CU$}S+w0newW=ldTHhgI^l=~boRAFxvO^_lK)yV z*JBaT8OrGF(uw%%<}l_^*u#&|(*sqXv*?#&g7Ir?aUT}J`Y;!;`>ui42Bd-Qayj@k zC7OSsGYYRAJV;ly8{*N`6ENcLEBv=<9J)6hgZ{_!>B0K%)Tm$@8b`&jw(qTcw^Xy5 z-Tj14FJCLZ7q~|Jviv9(zPo_Qugvja-8s0MzX%@OyJWvU>;(T)=@?y~X$b$4#({g@ z7&wx@7;nzwY0&EemOEPu@;s)1gLIwkygL`A$NUOPde@jk&;%G6atWRFylG+S0~TDK zfGQ=u^unVsi1{8{%o!k?wYdVj4;A3~38O*5@tHXE{C?o|KfwOjc6!`hdKk*wwK!Pc z2rL{$aMQmZlk`1;)$#oxAzKc^zn#aC#UpUtlB?`xpcMvweuAs3op9^p*=Te93a%Dw z;Q(d`k>Szs=bI(V(UOO`tx0S(8iIBHUFKNffyJ8?@w@j*w6FKW(SMaNyKOJ4*PO}> z_h+!+3yt(}B2oj_gLJ_{W$bRR={s9VuE2-8)BWlI=L^wz+_LrVA~ z{W3PUO@qlZzeAR+vgGyDaOf)84I51-!LE6~SdjJ-W}5s;eD|;hR0-c{XX-JyJkUsT zZ_OBqLq8?SMtK`{;>#Iodr_O#+GjyV;YirTO4)!N7C7QfB(%Rb7TaeiW7yORcJ?2# znY*IsT%Q!U{~(Qao7U6b!mm{8aU}G28G^AxPvbW0H|%k;9cJw(F#EwT#?*u0PqA6C z__P<2PkUJOhdDH}_AnnVSIy!>Lead!7!sN~*piS#bcV1MyK)2I$KEA0Rih8S@EE`+ z#uwv4lLKh&XUS6BUa}uQROP|c|$1d0A3`t7U{jVvar(QB0i}zf+yu!%zwBtxaR|3 zAN7U5{5l$c?$F0-gG$WRrZDSJD7vg3BYdem&$E{s;NIgbylLHrkFFMgfwEMO{P6)& zCK?E#d7JS5iMcS_z6!q7jD?Er#=_;jjo97qG<%XWLK0-HD^X5Wl^i&C9B&IqobJDQ$|k>xK9D}(lD2={w*Vb6DK8WfO@ zX4r%V9nZwur}yB85GNL?IUY)VCt#_uzL5C%3vM0sgkE*2N2kgWSibHOdzwgg!+^46^TmC)%uwE+s zFdrgpq?fU;vy8C+mNITw?FKX6??bz)7P|j?C@RG#@Ve(Zn9A*GFnmW8RmxM4n0Q^s zj#YW!r66$nK1ufOF zLU?Bl4YWQ*P4(kYp=>JJ$h9#p+y@80_{Ro6>m!_5q#;yxEWx}_CQzZA2{YEXVWMyg z=Q=&btAY|sv+rfyle($bjGYKw=h#GFFX+3xi`JFc&_Q1~=3S?X+j1;Xb8`qtKRzR_ zFwcUW-^Rm+w*x_O-WYstvKAGc>evf|t=MB7$f95^w9M*ePoj3h{>3T6DGdujcsofb z`O!z%$(mWnuxNBxHjUkFKY`w2MXI;K65cJCZ9i)9M)l<=~nM> zkTZ{0?}THm)-miK{f})^;)^HoAQ9=vOXkl|mFz89$Df|qNt2zEVJl|W1dZ6UH#l}-|E%E;);W4dHM@e zu73ivHqL~lVRt~bUn1Q5lM26l>WUwI?_j45qtSoeTzo2a7r06Cg0tUI%szaP4#+$T zOP=b&P+J>Zy*7x-3^nD);5q8^WISfRR~FZY^#`3$eY$f|Ev_7r2y#nfZtM@v!PEg$ zK%vBq4oO3Yk1z>>T~46NTTk?}=wTMxBT&<}k$;uF1)>rkVA_dXs2ijM@w3ij;fh$; zaWfDMPHn}>caJl%Wq(1@r$nkT}s1X-m?oHsjwliLm3=D|Yy2E;Ty$h}Pz-Qdf^m8eiGKyLy~tQPU^kxu4PW z`@JdR3vtP4Rd5j1wdR9rv7^xct*($S-iI5O%x6!0XHjn-dpgxSg^p2P53^%s;G8C~ zWbJr1Gj0eysJjejc^sw7biP8Q+Bw{OXfJ!rzQEE_88%*jJM#>4rXL%X1gD57`*&Z> z&?0C$YgcwObuy;QwY(ES>AAT`w_hsOG+a2zovVrZR1R8QCp4oghgq@-- zuzRE?^UG2J-ZY-B_`I84R~`V3tu|ouG#ApB+QE&uFnGqLD2BTLbza8jmiOP=z4nz%TCN-pT@r{KDY0(_*Z2vcusZ(|Fzm^esi+8 zwCN_8PFN}U-5V!-J0&l~eA|QSzVgU&tHl0;PNQ;QDwQb{Nm|r37*`g+tTIPo`G-@C zJy=HH2$SK>&R`61iQs+B2jGTLoyJ6yH*+R<&6O5FMgDk%*)V+ttcOg$$VoNs^ib8P9(PMO8 zOa&-l?G0D6P4K8=A$(e+Dpu9?q32)bQv2j_LQtRz23&W6@GYYR2iyKQ%5ynCuEFvQPkikfi?3>v>EY2F{Ei+i8F|1$ zB3}Fg$AcN0);y&6rtEd*zFeQF%nG8De^=wfI%_<)N|(CZT;XN4vcRouF?_L;p!->G zrZ7ZPlA#bxhuxMJS35?CgH&}u_1GCY*-9JCd#2*OqpNUzTr_-{^AM*VD4}Hoip0q5 zh3kg>1^*&>;pnNcxFyd7(py^SvMWcix^r`}N8?oZSn`4XeS4I;RZeFrk>k+fd;xpi zk%*5w_QFPk2z(Q6gw99)z})>aKy$8@L)eWTEPpguu=v)1mer*&wzq-4ZMB5HyOz=Y zPnGE9=rr1C(SSZ_9RmO=J#|- za2!Z%)4}y{4b_l7!@EK9!t}e1XjK&gZ^zEzy%yL*y1_s}>+mb=+_*saG(Cj*t4FYu zMt{7Xl*9Ji8;J!|Ixy|W7MON(JTzr-@MBY7p>DFQ5N$S3Fi=S1UtSKx39qi>$t#Cp zed-zbbwUQW4&4X71Iop})1sJ>UKrX>Hxm~8^blkm6Iqf{9(IK#;a+zuw#3bx2Bo~D z$4qY7OL_Y-Is6Jm}EqURE~t z3$FDy6ebTk2Nm=4=o#zHbeGvZX^&72@~-!R3qg}WYb*z5-;_ZnG6sJSvcpS}c5F#p zIchkIu;B3tEVyKdvM_@k`M!n9?Oe|Gn)$NMCTV>$*Mc3{=*)hSiGs4vB#GwC9O z3uxx7haQdnh1hqG;kmmWxI3Sqh81-Vl{P%K=WG5*oL@p?Lc9$^46p z$FVj$7Cg0b@a~Wj^pD&JTL0uAZu&OJ8gqnl;p3RT-#fNlq9cj%YJ(t+CiZ;DPgocF ziYi>V#lGH}gM;ohG5_E$9N-j*9tv`TqO@l4$e9E-wl(yby$6O&Yy!vSMRcKe5bZjb z1Tk|@i9<&3r(Y)3IB0#=lGgZ#ajvlgj6X}+OsyDPXQd?R)%1dpJBHMiKLQ_axQIK0 zH!xkAf`1bunPjmT{i2$w|Amzh;cp6Y3r7my79cJ^A0U+3KV*Yj-=f*VTiEB_e%NVw zn@&D4iT~VbE*Onh6O7B{g`rLX?2zgj?0#u2XsZ`eyGBE5`|to7v?ao(2aF!kPUF+^ z3h0(+W?(ikoL|8|qcR5%fwKQFc)W8KJ$$K|MJ1SuQ};)}KlKslf2Xg|B<)F8y<#-Y zd?24NG7OBqzog}UTDZCS1g495TxgVlD;}l7&NBlg+#XXj4-#WVf|L_ky9r}wEMPn3 zy;xCH7J9Bb3=iYPFs;T7npSDyxKR)3y7W-ix+M#eyT)%_Ai zu^eW3LJccrVzBCdf_QoRaq4*J3KPFwB0g=VE1B^m1J{2j$HiNx!HCLajBLKcwocvyY} zRqAl2v5PX9b9gjfIB=N_;yYPv-U&SZc_f@J84CSm4?s*yCZ0N(3%0KmC4(wu#s3zk zh-1%XiygjB!kl|^U{C3GSQoSmIhiB4%>S*r;2;X2@bJ?+`sjTOs0Joc=S|&g`3z%Hn;>Le4MQt? z4LIyk4llRL+B)H0RY-Ea~cYX!i-%BhgzD}wdkw!?z@V|XU=4i!C*w_p4rf~^xAu&36F&Rc9O zJ|47!zD#Rp{v+DO^E9urMgflExtj=R$T=L`KhQdH;>h&O@y*TnRH-d zC)hh5N8e6IR0V17RP+Zr2O9DB7yV+nOFCI%gB9daW%m3HiZdpfqwhamTDWeRRAXNS z!yXL5j*xZquTMFB5P1TYMOe@}gSm9~q$@PWDNVfYVj&L zE}roifqb zO*sxhCAx5QC=an=hu}wx7L0ki30EyCVUEVukm$D-HEcb=t@$ZD6Bo>f<9EhGLB(l|>vNt>&Kd(xSK2`7%?kLEY)n_q7>@Q+ zcVenyGA)rc6;{u&5mq<%;p?kbqM@1x+dsk(4=kNZ>x~e9jLV}zru*@T!v+4%wtRkZ z?G1;aixS|Di9fZG=DbsDRoHGx2D_Tz0nd)B2x0Y#d|Pb+6DK}plav;+v%PT;^5+Cb z=ulea8->U7()p9q=VAGxQ>b^)45CU6=v7~1`n75>EGe2UtorpBYkMy<&F2)N6K~V$ zh1Wo$CZ-R}#|R}ZeT0!yPcU;{Noc<}K`2!pgdyt9cy7Cmpd6(wEXtX}rd%Jx8(Lfu zcf07KU?2x?a$Vrkf<*9Mo5@V|RG=YO6`KAkQZ1`Hd`&|)X7sM@EGrZlXb7WZ#|l)zi48TKh6yhWsQoJ=T+$_FBTZ76(UbdluUBy>$S@jBpT zqTpWI5!kU_PJnSc&_QXp?Yd#hF{JSdy>rrzjvuqCV7lti_BeT04 zY};#Sj(Zwx&y5xo9UQX9oT(v1MN}MgB2A6z%VWW3ditZcyz4y42{iYI%&33xo>C*L-xgip0OiM{N3eDJHq8u$$U}v~rZTpmVAm@3^{Q z_niTf8FQ|H#{E0=;rw+VtG>%&VDv5skZOf4jkP#RZZA8axDV)(UF@5EG(7pR4h^P0 zqiY9RLHv|P%j>O=bZJl01_Se z{Ni*|I&bqx$n2QF#IfmOGHw&UPj@;5zI(%tW~`@iTMxm5lg;ozArv%C9V9o`9mo5* zgV>3y580s|VpQCoOpD4>AilK_+y}m-dw;bu(*>_Vci2NPHt>-+EpHc_mkyTdens@W zM;HW)GC{TE64=aDV-drh@cN^G<_OTKbSHyiy{nVQGl#@F}aa4oxpUV&0^WZMq#U;^ZgIme9WR*FgQCVck0472n9 zVfgIp$e&mU%?rNJMfMLs;#-3kLvNzPix)^^!dbC6j|Ru8(}+=hB=NHqC1#_vB)=o3 z!?BC&;ZKe&th(BP(2#&*G9pm(j}s2=QAJEMhgnOi7+0eU?_d3*jlm1p%e*4-t|MBK zOh(XS48?6u{n2IdL0r4#EDpckf==5A|F5u%ZQWVTyqvsfaC!#IC>>2*l}0m<1Iuud z-&DL2brjzhrDJK}Bbqy_T3qHCf#SeP#VhU@q1LrVTC}4aHhHT+e2^R~ z6joiy0+$$s5wc#@8Cd#5`Y%_#icnAvR+3edVPiQuigYF6DsP6ZcnXhQD&-tK= z&h?9m9h_CbQKpzp*#CuoxVX=utwo9Xn6H7!edfTaP4P^rzMI-aT@dR{jle*szUaGS z3abhzX8jZ&vDs5k^Eq|-{3Qc{Ha(k3kIqkKZhp>ml*cxQ<9|ou^&JZ!MD`F}CEpi& zoRMD98^eMoSl}wN{%}?|0JIVs*b|c&u~XP#HY1f{_u&cjUuv0nhUP`~Zg4a2Y8=Lz za!j#t!BIM@IfyQ~R>6V?rO>_AiQ*4sny6&2%69#5rtboJ*-xpSHer!0Kh9(-_PIQY zZfHw#=<5}T8wRZ52T>o;bv#4u4c@X?^M`y@rxnZn+ZWCknlg=;Y&OuZfUY(wpq_7| z>DJ6Vh?e$-3}>3auWm0eG4Dr@s10WuM{K0UQ~_~&BmFY(BCV6(L&vAzWjgbUU`KK| zh$PRXpqei|U+=?$dwkf2l81ct4u5*^+FbVKwy*e|d$>4!X%X%nybH!GlxorCPE;l? zRJ`iAEUZ|&kJ*HoLEBAZh+LS%Vhx|tKHrAIfMwhGb^C5G%Ws!>6<-J3)2ahe37f?W ze_m!2drOP&%I;dstPu=pMaVlakS#-9(Z>2G?=egN0WxC((wZ?No%#K zYKNfWF z(ODs<*}Qy5`f1;6hx~<(tmNx$_M11s!@E6kT7w;=#V&?vo8RyT)Bhhu=N*>g+s5(s z-bI^4L#0jBeVtFCj8cRm%8Haqp=??zrKnIMmD12cXx-O&OOlijC6tvBk=2h7-uwNx zKc0?yj_aJ?@8`S@;!TI)Ur8d?7rAgn7xu8JAi|&RP&b)yo|Jm43(=S zf6H4~HKz&g7`n4kH*V8^GSg9{#DXk+_K)5FSrs&r(pbN|l|)TDfqY`E@!J7s`bMjl z{=Gj$D@3l5dn&O+U(o{!ZoZ;Fy|%NfHoLG*8XDZ2oN74n#hI*f66c(E|DnO7-{HVU zRUpYT=)2F|R6VMa-Lv5m+1I4X>s#B{`|Dn^qD>E|f}1xiWT`NVZh*oMiR&Nn|e;S|q`+ti>! z{7owJJ`A=7{=zcdO;FSTkYjd@yf|zNP<4WizW0F{{tg%zsR%l4b5M7~TV}~7TM%E< zhhw~Rafi%o3XBBD28nP1t(Tc7g_&Icng+BNlN1yzh#|zSvp#=wR)fMeGnic7Mm%?y z5w9B+`29*7sFE=dbdc}$g@y5lWn9Cxq?O!H({$L_AujmIHPNag7U(u3jv8)F#E!~b zVpn+qmj0bbx_bAL<7&gK`95V1yrx3P%3ApF!4H4EOk?h5Erp)d0aS921_q9>4&fU! z_@Rg>*b5tiqjEBO*3QE{3!=%@Yfem=V-EF-7^MdnETBfJD)6w$iax7JL(@59Xvrfz zY8qb2Zn852m4&@z!k-u{)~X@ZvNO?Bs|t_oeM0m8CNM6`#JQ1xI(GidGxT<*8?*F9 z2WxyTp0)NmLpMnYNSIA7rCQ3|PwR`Y`P@@@)qR@k?M$U_(h~8{vl67Xy~JrFMHKLsL@!fcDOFpD5 zwIWMS90plc8RX@fjGVF_h!5LC!5RhjwxSxH`ptq}aepS6I@CtJ<|MG0e6KrkXCb-f z-~wl2)VW9#DM9;}OxRu4>L4@4mkAolWgcA3#s30VvYlxPn7l`hEBAc{dnMIzXNM%W zP(q#?N|-2klGjR)rrcs}9_fIO%N8_J)}%T{iD2J)mF_=(rA{tx19*qdCtF_zli5$j zK++|HrdofpKUv;SZVhX2byX_d_See=wn;s7QYX?O-^1irh7mXU)o-Yf9)-r4V!S2g zDBGVWF0fC2iDDlE$*!pyg2whcDEY3ADHPVmYCTKH$vi_N+JDn$Bj()W%JG8R|Ga=4 z@1;TBw;Eob)`YJWugSo`0qRv@K|J_bLT&PN61HP9?U-1BagCEu@|G5^9+Qks$Ht+b zt|@F9QU{6CXW+YX2-~8&6C|>Ou`DN+VnFHGiCvt;ev zH{|mlC!BLunmMyGhFiP27-g+W>AE*xLG*z?*w1)G&&eloPcL|4&i!|+@1dnIFIpY_ zo%1R0*`T(q zG2E?H4sf#mF>Jm*)Z0tKqCWfP(Eo-qmy@`b%GEl*(w6p^RJ`HRxRo*o&ihW zZ^8a9gw)-b1eYZ-ZgX~^ciT@iTXPB1+tn$5;D*@|KR`Za9Ynvk_BbOu4tsSn z5i&nRkx(D+mGQ^oCpkoEv2e022o|jBt zn9QN(OT8dv-x^NwuMA4s@;zLvs=)cRA&Pq|;1&5zVA(4I{>$evUIiXp+bc<~*+CBM z8$ZBo`!U?#OB$Ty!;AP%DUK{tod?sWzee5Dj$HHaH0ss31B>&j=?tN4kROueKJs%F zJE>Hd+P2zMTfa>EOC(HGR-EctOs#9ml)FB6#EL9IjRO8fc_nsZTdN3qQ5$Aw%vk zc1Eqn+RIPL46$iMH6?s;KqhxZZ<#u5KvPe?7d#NN#ZaJTOiL4<-B z_sjkg)U9^rB*OYxy?{K}Wh5n#i@b@TeIL6om+~_)VJ^a36uTQ!u)yXb-fYh#pFON# z>Hacy_PYs!7sVG)XuCTmA1nsd@Z;R|XY1*vD*$gyK5}O-Mf0^vH!D;3la8@{3_X?o zur=m4HlIDs#{G4s*N*CO&pVcID>7S9S=|&DUdo4%_ut{-E*%i*F-6G*;bb>2I{a-Q z0plmQ;HCvnkRh*er8{9EE{8hp+kz4rb@c9;m7LkzQ)Di;l4NGZQLowucxmCU!Z>|j_xyGhFe~k zL!ep;{!aUdG-DY%Bkmxq4=%HZrVVdz^X3LSZb!siO;iRhp7LS{3!UZtPBKadFRVFTdg zx*yh@ddvHE)i}Fj`(gXrbaXwt8ON8bhtVnCpt|u1I&QH-Zf!C&7?m@Fd39i1WC+!k zd0=9_pV0Y(l$d4Xrxos8+?Y4mI{Q4{{U*mvyZM!!Fs+gVNoI5Mb0?zBAq#dSzYR_M zn;PC{qEbZQN*eg2B%+$~0}#R5#2T1^-IsHeuaRXAgv z{oH@A0rJKea0dgnpz-(tI1`=5eVg$IyLL%H?u7zcv*-j!R^G9^^IpQNQQ`nKm^UB@gt|oQFkh;{C6%cV`<;GDv`` z#FG$Lluz$3Zl-JJ=|fcWKJJ2EJ^ouz4B3OO-2CO|;Xl(Ja^J@u+I%&*na37Dh2?y7 zSL}debKPiZ`BOUaSpy+p29{H7Ipxpa$aobGa2bOR$E_Q%aef?ZtlY;vasSZZ``;9S zr~F#3fd4zLUb_GmJB+w(Z!^H<;{kf@t{+;?m4Oeh{c+jCx!m2LLR5|mf^UmXp-R;Z zG{{QEHe;@?B>zL9ZTySD0~i zhkD53vxnfSejK+mZzhJ^`U?-OzF?^A9&DT30>50YsKiedIU$pQz3*c!aKUA)V%c* z+2owS&6Wu#=EII~^XOV`+%r3T#C!Qdd;{?F@s+sK{vO!+wBf1iitxuH7Swjm<{U*& zkeyC@KxNe|y#Ko%)n;FVTgNZsIQu&2wy1&y8pm+rnio`d;56NOvK2dZ=!Kq zAj#Vnz*#I?fa6ZTBHwcixz45x824R`UOK&wTlcXYlXi82X>ti(%IL(t`7;Gp*3#Tf zQcDcD@pN`%AX@m9@;%%mvZ1+vPM5z1%Kyf}^wv*w=jE}GU8*d2J*q;>VtQeGYyj)B zCWs3^ltvc6U4)vC6Y2YkBwGDp4a{)7PkhJr&>fr4QTv_lcs8_!dHhO7U@SKQ7uU{$ zIo*zMOEi#tQb+*Nup{((i!-N@7lXI2`mtr{tGKaFqJrdyBM!zskJ!T|N$@M}2#n98 z6smfdO}uy2(S9yB8kmNk%39%9>or{S@9>r_Z?`xr8(LTbr4y-Iw&ay2AgC}|k`wh$BM zMw6NfZ!ms56@_l@gw3m>AiFUi3#J9a$O%18=)MC!`YF$CUJ*xg?oGs@;X>N=q#v>m zb&?5cW@OoddOFtm0;*jM#2b;8oX&C&E{p%}1E!k5!y5t`?Kz9tJN+dZnCwK$G(B#d zr##B0JSICc(#f-$Y9goNPrlu@1@E7yXjSA|xVXX%-lTzGn;sGPJi7y5Ul)VT$P7Vd z&m=*wtv=kVK1q(2#gPwQ96ELBfdoG*d}^2qy}Xw6$kVZ)-_LNLcDZxEAMfVu@@nw< z*IcYH9?O*y8#3mFAu&0A6qege2BG6&WW8tt%vk)FJ_p^q~E;x6&n4B7`M_j{2P%7=euY&KIC^w{9R(m*>;fV(|9VsD2hS3 z9C)%KfH7-s!G9N~a_o%;v`LcT4w=7%=Sws}gT|xwFCW_OahgQ;Oh(7O<@nLU0Si>J zaChZ6&inaoJoT!NjS$|+>X?22J^yH|>^(|$YRX{AIs;^ecMt_>KE9$NBU^Kuk4a@X z@ADowo7VyLsyCqE&0kpkyc5D=Dxh`u81BAxC9fITi#r}U!S=8jAfGKullhwTXNEQx zJZjBl&@H&S_y~D-P68g|bf_3Hf+!PX#^r`0`IcNxW*QIBvx}WEO7k4Gv7JLLcWDtt zYj=n(vxmsqOgtO%66f}>!>P%77}%CWh@h1`eX)?c7N^a%8vn$=4S}#wAO4fp*!ogI!Q74M@eJzIi;>}d^dkGG9on`uk#y~JU#AuN* zIJGemWpdAQQZXgmgV+}CWQ!Ke`=*7LmNtRQ_1lo%)eVQePQriV6f>=Q=?`C&SNgb;(DpXs>rMKF15DkCHHoLq>W2HkgU z!K(2(8>`Ys%BMxKak2SmHf;>IO|6TbJCH_l*1M67!IVENb`H+A%L$%t*b6b68e#o9 zS7_c9M~a;4nI(B;^hSV=V8!ebDtz}8_@s;hUk7Q%_vmT5tJ{Eks1(L2rMPm#1*^E1 z-&9b3$_C<;t^#8${K#steIOJx0h-YHbN_ZG9cfxhF3(G&J%?Lx=|2NHW??vt zzAb~j*47-2TFD!Z?YTb7r9i|F;h0$oP<>t!3Z9EYkl(3#wV+X7_GmQ`~zpfFbt0rKYasiBH z8^PJO0|zT)F(qh1>iWBqn} zxHKmPmIOHqy7x*59zAG{nKcN?|q_i;s?pDg~2@VN7umCp6z#%+E|CkRf^H9opr$v_&=MTtby?Fzok%0>yj3DB!7Nvf9}A^xS~ z;8V5}_Ne&4dtPg~{`ws(UR};ysyR=-Wu}0@HV(FzE+XS^u0ZR-ZOD3^L+6SV^w2VZ zvR7Z}(`km>K)EX47xOtzTMi8V9fV(s)*!KXGN`p@FY6g6(r5kFj>{steq*s-gG$HM%vO_f{}S zy|<`Y86*ETOXm)98N##{AE|WAndx)PCk&ykz?s z52}_SzBi|`atH84_YKlEF9}2<-{OL|8r-!I&uIm8yIc`=KB_val3mjXB z23Ka%sFzvr@ZuqIHPDhS|2szDIsGl16dZ&P$Is9K;}?iNZup=)65QHzA$P9;h}ucu z_?Gxe(i|23_ zk@++sQLSO=tXkq2+=#EwyJE)jY>X~VCN6^q>5kN!Br>Z5b$#w*?*47Kt2+wS%;(|e zy;?A*=NYVCu?YIJk5kvwt5~*Y6?TpeVb~IP`cGp7+59YYyX67%=fA;jekYeKU4pu% zC&?AVS1?CgSFobVP~g5=L-5~EN&fKdR&I+?6Xv&hFv+FAaGp&bRn_=`gR>;KHBQUf zRNY;;PW&>j>CZ+tCsBHh9cBnQ!k##tLQDD#8I3PzNp`Iw?K$Ye)@*PFZPh`l_bUc8 znKPijO9Z0};+dU~6>zMh4U~Cph1!59NctdyXPwtj>C;j;L$UxPf}7a)v*StZXO`-W zm_o$KX%Kr>1_WXgh-afbJkEJWv&DjN|L!o>r6LvGyW{aX+fBzk5Ww|I_RxIH41-eh zA=LO7Jb4-m23&%}-spBFXLJg^X|Io`xl^pd`+6!@APU{4Vtqzl>K$*SHt}<8))Bh02EU>+O;o+vHCZc z=492A^nZO%^7;&;Y0f4rz9pjETbNb31brUlioLP zA~nOV4tar#n0fEjKvm%-5j|i-+}@SYcB^WJ6I;x>&R@#!!98F>@&OL40!L+{R7Mk`}>F`KoH6PaiOs8*DOmCJgV+3ycQk@Z1v4Kl&FIlmaE*N0gzCujP_ z_9YV>FN+zkhuB+}+KAb_Nu>))fV&~!^-v8RE`Gh3s zdb3$ecEgVk>Lg&XGx?FPOV&@SsHVs73I{vr;I=VTO1hXD-5+MNh%1O>%!K;8 zr)bIIL~?tZC(L#V1PASWX3y&ux-fYOEjbxNq@L@;KvfoVZ{T3?&M);om=TNage}c~lav%BVr$f)w_QyeM4o=e2G-hlxN_gxO>gO&`8!VLSorDb`O@euF-T3a}!A#D8vYVhq2 zt#D*8efcfyXivxWaS=?V<5cP|CJpt%D`4D&+j!YUfF}d0;dgQ(diK3+5GDtS^QaTn z4!@yCObe)x_5|!-CWY5L=Yp-mTO3YHC+8{>@Hfxx**L!#L+N*FF;5YAf#(I_;-3>_sj@f zJLNq+86S_w^uo~h#5$~P^us37Q_*#@1td&Y1g#gBsKs}2xbpZtiC>!uA*Pi;}i z)?P_ImkQ&_AVSaIN?@|9R8Yor7I{%RgDLwm0oCtZYk0c4id^2mlD-tS1;txVV5ssQ zcXrOhUaRx;gs2vsx9}p~`4LDXBRWVyPdhBKzC&22JD3?A!!7!kfIeGWNb5F#IDfl| zWGaV)Xp#=Oe_ zSql&~w}qT%Cy0LVF<4lAiwezJ4$*%sc`fcB9q5oJXSPhEdt`r-kjeXynKhjo>P^7l zqAFCoaR#mXbl`n)DQ(H!M^av<;gjKu*ld`NIUnU&`^~&=?}HbN`PENnN}Ph)RiSXv z5Ma-u3`h|PhA&qYab+{F8~qyxMoF)U`XqVSe_0*!e5^RZKOkJEFd} z7c^R0NP3|Fik+>Q*1IueWq2KEIXohV8v|idc|SFq6vb2-Yti;AGeGLQDI7fz3BCoI znBd3zroXSHf6BK|#bqf(WP=GlUjBh0hfHx=Pz#klA_*RewA7O(f$_7Lq$Z zX3~32dg!BW5AM}z)N=DwxUxt_AXPU^ODvS>I161kmOKF)9z9}Sg_RMbGpb}T{Vy3_ zEsf@powPPD7Np-eK+=(iq-M7rmU_;>_~1K+418Dc+A>W^#pnOSuYmHtjo9Hk z7ZWD{d6CPaXo?Q{edQVs&Av!&O=aMMgANWSPNiey*W<>nWD0!zi5bjChNjVsCD$B3;{jG4@*~ z+>ccd%=jKlJ}#9eKlTY=|Hf_teIVAHUTd_Ej}%a1L| zp+nEf{mg9;J#Ga)X-uO2Yfpf_sTNM$`kUC%L*R8K1Lca6v3p6Dn*pDf!hs;`Lhp1|M3apcij-z{+x&g zd&-!mJC~`5ybE0Uc$nu=dQi_BfaWFd*sNN+hIHk0Qtagf--U`v*34vPSZEQbCm+Gs zsMk0{N+18YIWschgKXg*IdFI}6Qlk+MW)!u3H((qk)!?AB-D8#SVl=><^B%Rxm5{j zPv4=hjQYu)I8}OPZ4KT^=I6t?li|#SiuxmKKatXe0L<<`M;-{pll)iNu|SfN_0v)&Atb3^*8}50k#KRNJYT$h3cfu$|^q^qpiw zxVsK?=xZSgH?X(AjD^8(@w|578e>aLz_9!to%S!1bvfh#H;+^?n@?GhvsYJ=Gd?X? zZV>=332)%ZmX~yxm8SY{rb8Xi3b0FA#MA~qr8zUNV$S|m$m)w=u1YMwizuLPuFeO; zbY&3xBY|IK)rp_~HR?n!k=W%<@L7VN0Ti7iOY;`N*~Mpx+r|)TXEEYb?{87MGGM&h_7SMJtA{8O<_<|n))34L=8?zLYJ%n&WqKOfTc;WC1Nwv{OB^= za9j)85~6VPumla&jwZ54p7R`-QkcCe8kT4-2G&~$-1?Kawi}V$cAJ^Jk5LCsj#CG> zOn}4zCs-VH6Vxs?!I#z9Y+dI{l3UosaD|uI|J=p#zx_FMWULkhy|*L_t)J7(^`_kT zHc76b)r8CVDnMbmnl6j8i$YO%rg+1zyA7zK?toeu<&{9t31PxtqsZ;lt7+@U3iw(Wy;_bw9je zs9X)=FFYry(OFEW{Vohm-v;-Ua^Y2SCd9lk#dW>?G`U?2&DPaA7#i?Aj?>TB+hW1o zi0)MGm8v^ub1xJ856=M3B$gD;J%CMPPV?R#51RJPlCww%!|IVLx3_iO!+nBUeGf3S`!09ZGmv_RUPhV2;fZO%()xsa!}Cpft1aaX6%=0LT!W^ z`Z*+`l+`bKU}^$l)fkX5pmgM325mL9MMXz>^sB3-vb--{VsHxGJbH#|`z-@8mDOMs zPy)o{78GX5Fb|kF4GO7?Kzgq;cKNJ?Lk&{&{RTO>R(=B7dX7WRUNKOSyTx>Dl%T4O zaYVR13nc2Z;kbx9yft$Hm3{}PzGy~#C*C8o8&*&S1%xk3Zt%!xIw}1Y4#(F-KnkBr zA4-+OB{^GIHsdaciJhW}L&vFp)ztTO_RC!J&IT!$s>!Z^I>uZ|P>}5z&Uu zU2Y_}cL7W~vxk+oTn|T^OM$GM3`!ot=xx?V*UAR-yOb`vBl!WHUbqjpS5e$_&mXrB z_(Hjq5qb6`iG*%$wfF4vp?fl=u*yml$4VKHU)v6VjpiiId&M8DUDfRnI#Gj6>a+kA z+ljFL#WWnNxriIJ5#l<-|6?MCV_{+VJ}~w&1FPZBWci*(a$DgsoO%%fa}K@*jaBE# z+Y=_t#GGS9VbzI*I4Klj*K0t69!}547tYLy0%8Bv^bV7)k}uO_NMumFpv- z;GoTAd@7QGe=7GDmDG?hGuY zo~PEs1zsN@^64Adn!T0|zoxkH{ze+xbQ4~mKZ{{Onqcr~JG}lk9bXsyBx~Jvkr~1`=SZk1;3zU^_WE^ciA5oq!D?6(|?r%yY%wpzFh6rgV8KWcyEs zlNWi8Sh<+s{)%hVA!Iyi2g}h)ixU`ATut7U2xDM<0M7f+fGQ8HVVkB5l^S~(Vts7| z8xCG%tt7Y8aaZHm!PnBbH#i*L{WB1l@Nan5q{|MaUhV9ST65NDjyAnyZ%%fxnUuIs zSN;%Wv;R4@j?y#m0ZEp`R{l={NOY|FikFuQsS!CWC&ZnWzl2Ut?HZ(JS#ha*67Y8IVVK|Zl$7N}Q|Xr{$Z>}eVp`fu zm`B~vT*_hJcu8Ew$9kH$6x)x~u^ZmU!JY-)cw@n)2BXCyg3B_4G-|d2omk-rKX$ys zcezEl;f4{S?_YcXNAdC(RlKp|6uUPh3mRMO1mm@CW8n5U$aIx}b&GVcXI>XEsl3=74(eM9y}x9ZkPp1WQa`;g41UDj2SVvI&P^{#*X7 zpXX!8wF`LbSq%P3ID$1_s-$fKhpHS`8x5WBx1rGa7M3eQJ>{&gaU*_A>%R>bqJT8Hz zX+_Z?{Q>lHUqHXz&BD;XZCF*J28NfNK*RkvGi)IR;@Qd|A0P_b5^jbQ5b6wG_{Xyw`WjQRMV?C+9Lh+Cyhh6P8NI8GN`KKM~dtq)YQ?-WMM z9VKgaEJFDLO0DQ_e0p4s{`%@gh1?Xt>F!aIn(l&?Pb=urifK4mC=d5;UqZYx<4N}l zUpUz|fpL_U!QTPn*d4_$X+TsaTykY#O8FDk!|(!GmG+gWe7lTa?DH|mHIXKLO@iy@ zcVT69APi#p204`}JxY=I>KNh0sCLdc_8|sv46&5+$_mMONL+QghgNM-p<7IeFzoQgYBm}m+54L9g){ny0V;v79Ov4A|& zCJ=Wph&@*rLw=lIOZvUk$#*JBH`-mG?*?s1c$X*K<|;vCQ4N2K;1qT4=wiq8ETR?r zs;P=#KGpx1$g|58$o)6AV0UsSntHx)7;4bOK|NJG_-zJ;#@oRo)fVRPhG+E2c3b%7 z?!(4cmopQ#kHzU79+33z1EK2<(aQ8{%&2Rj5?j_|{?iN^?JhwVxGbeQdF5p1nj-Qz z%94mF&4H!yXK@+34VwxA=@ol(UMrjmwezj9l=q)@{D~t6SAL^R>H#E|qOdO0jqdq% z3av)2l4}aF5ZPA&)B4(=v~ND7l+Lj)$?jn_@s0foB}3ZwTLDgAYN8t#P9)0n;^DdL zHxjQ_j4d6xM1Fcab&}YDh2vUDmCqV_*L^SeY|>bEV~XmyStk_Vg~6;w;Vhm76k_0RqV^* zcn4|iriO!+4ye_%9k&Nt;f%5p_V$Uz;N!H{K8n!+QRP%J^kyUNU!F*kTB}jMTLTYd zy&y^tl(AWUK4h&8fek5MY3+7w<*BxM@1nt~G(DB`mRCwul-Z zJ;pOsvgXe0_(h(KTw|(F{AE&CEua}! z=fJjK6Vdu3(0l);a64q5P)*Z${4PJlVP8`diHZEgjJ|CF`8D3uXYDO;DKUaWUt*wO zyBUoCqKO7?{ppKYmx$*58Xmyrd^t<%saGOJ^b+@cDL}3}o<3M_f`uc}XqFU-=7V|k;?@w}vt3U2o_b1L zzGzX2?wv4C+6kH^wNXyX59DeRAvY)pvVzW_R%r^>U$$lDs%Ffn=;is2TCa#(hCbRn zP~`&VA3!gY3pgzIgks=lI(|f1K-0wqQL;TS`28sR{bnvqe;kk7Hs!!1b93Aox{V|{ zuIKe?i@@%aF}Io5EqCo7%jMtR1-6-gNSeh%@=qiQs;?%Ig>G&Rc{`8dJTV6l4^HO2 zV3j1~&p&e5zSN_NI2t?0<)wK!`KxD5 zI*%A^-o6>9jh@DE5CO5EEqI;GhyKwAWC!otQTsSTp34ox6rXmof#=8DAHt9UC+_l)BWjIbLPbWFVaWR5)bMKoeQTry^EIkLB)JG4>RqDmj$9)yeMyi} zIG%a#b`;;poW$Re(r~iJ7cZ;NpkJQG;(TQfPF-;)Cz13N7amH)u!TFxos`$qOXMZF zQD~1(#`5fd0AYegESau22!0x~@WI4E`g)={{!m_qxecTAZB;8Q|EMiUJ+~S^{1hTg zK@$<=moiy_mT0Z>hLLc8O8BynN;vX-`?*HYe$f+KukC|dV$nb=$3p07I~d&Si`%+l zuxj}#{MqhBb{wptYVBL#rPl=9o_`J%#_D7E{n-d*t+@5fy@t{3@$~N83EaN444#z| zffwxvUeXz+f?;JmS9%>UJSZX(OIE|iO%JKrR2gs))WTOL6?`Xd;_8l_L{sN_ux-hK zPirN?Tw6nMz&D0|5c|Re3B^Hg;Y`lk`8su#ib11s#)Qr|MY_s^@ttlrm2WD-1~*I2 zuE~nsl&FnSy9%ko^JGx#N@pu*3xbDZ4yek@fX9Tim=T7GR zf{%$*+*xw;+fpbmIY=!%(^2@_F%%X&N4LfM!Tfv<>DB*>ZCjp`+qL6BuQZxWUXefs z!`@Kw-R)Fs(--hs9?!d_oe>AwkVkJJ&?I=BWM z^YY245ula*Iy!BkF&Nhtp}_SxhRrJjwWUJv;BhSTX{;t1C11pmz3)*WRu8Au-JnTL znmj*60W2)6nR#!o(%i2nVBJO~xUXAH_B36EiI3~yYNHJ5uz$(Ap~d*#D*>CwPlEEx z`dA+rhqs%u=^yX&^}?T%kyLME9bbxpSpQ8@G5;vuZ@x?4oDs#rmPp2(e;Z>?XXARc zE$F4ziz7u^*ps^o4u;;Nw(?rodHogcc9cPl?R6MaWh|IoCoWjk=L|;8YiY#|FAVL; zp`GiO!S+S3X~R}g%vc%*#qNus?nwdLU%rje*5OdpxE9CV@n)?3KG1=vKFsuJCt3#9 zIDNuTG`VvUBKmnhm4YZ2K5IRFvZ0SVek2jK!(_S=jS%J&}oPr?(F0;Ja=uf#tSRP+X(}O7A~H?Xpe~ zDOyIJtMVGcUETEk`Rg=xZWP=X=WkdHjcGV=yM#1`7vTM$BRFQ~dvd_a66bX*q2Av= zq-!LHt)Baa*aWGe-Fu$#8Y9Z3IS62f@G`n*?^XxSMu_>?6}VnU2nv0!lBoqBX<+0R z^16DY{>!>9rcXV|q5CM~(Di2l{@J?#|7Muu*ypd&q~sP@$qV5dR}J#6GnD9*A7CW2MzBU$i`;54#%cMQ-0>48IOo88So5=)`5D5pH~Ifn&-`~d*(w)jIr~C` z{u%fpF_ZMpi6J|mzQy{9S*R{Q3qp3L;C)SbJnL8w&)gMZQ?_Ud(g2TG6W`9TloXxnZd%n47i& zrsQfNJA4Lq>RyIZpJW7$I>vO(dcYbzZ!RN97ivPkH@N#fqPAZbQ-l5DT(DCm?wewV zudHMQN2Zwz&S!?frr2i873@ZvGaDF%vQc(j#sDc_bCK*1j)iez=3L*C@pz1n*)H`9 za0?icn%wDlsXmcTe&0q-Uj>0}Y!$WPHMvk0PVJ`Ii=2`H0aPXm= zV6ON@`pL8f<2~kqLdZM_k$DI!jh8^vQzcri`wd$XmUHj^SU_vrT+UDUG zK%1cbc%kh*H9t`Xu3mlgd%;H{T&+tZ`1nsBiwSm1c7W=&R`{09;%Aq;*w}ZJ+_T+J z-0Dm?=OypitmP`4l)| zf-%R_;jf}Nr))n9KTXQPGI>RTf5TV!XITt!rp8!(yM%q19t0a6orbY{D6CxX!Ad?PBBoyE&_*?9Q$8a=w`(j=-d}8s(Z@MslWC589r6X!X#1NE9C^G9wc4!F zVDT|JY0d*8=ofolD3;TSljjbp#(NnD#N&9b@iFng{#Z@JGDUTrNvrFbj$+K*7>pW*Oo zM38ngWHPRQIbB%Qj+fVQjOe2iF(>jO{yiiC87|5^|DOt|XE%lZ91~$5D*nKnhpo79 z{cq}Ay&BSQ6w>D>zSHBg4}*n8It=?=rW^jugHnT32;?;EMQs_-nw&=#h&Ulws1*s8tB~_LXELVLtUaBL)KU56G#0Mb^`O zIat0^K*nGm=4fq)5WzgMO5K+(zBmIq>Q526z6h;zTw&LA89KQz10VT!K>XuD>tzQ% z(WSRKAxK~v?7KA$4a!@=$Ko14^7~YbI4=YD{CS)YJe0W>yB7kl3qW97KV|eO>@4mi ztx;AOv*SBCEinhzE$zl$yal~OVK8rvK0BoOhdfZ60*@1Y=>>@uP}6TmJoaeYOiMZd zskNb;S121ld8orzy8vXmf8!+`N$|ar+89!Fg`wLb;Hlh7ZXR|*v&qV^z3gg3z#$pT zxVi@T>;KYgH*evW4NJ**{W@ZJ{|P_U{wC~r|D8E-<2E&j2nM}T&KG=llv<5fgUO?@>UO}hVqU^WpZX9>#Em>?;MBV<| z4A+}a!o+b|@LcTxQLaz0eE0;47Yjn+y-Va=ax5{Na)x-v@yOO?DwwsclgvFXM5hjX z#-sbQiRJxleqTfhro?8EaS0LTC94eM0hpt4*VL+kdSbh`;0#E(>f z)kDe8hoH&q8Sz`6hk7>GNW|b3Y8|x;-%dM9AGFP6mb>i5)(4p|e9jDBdhqEHTUX#2 z?4fri%)o`0dWiU_KB-Y;(Qvf}9Js7tZTa;VPI+Gf`_~zeSpE^*Sv4C}3*IvKs=ZKr z&qjFt^(_79aRV>^cm*?@)KK(e9d0@91d|Id^Ox#WVea5#viM>kP7gE3ZN(R%uv86% zb?Wht`2`S|{+oo?O~v-4G??vZjM6DujI@3oyt|)B*A52Z?6}P!c4>qZ9dttzM=w}k zE5&i;KT?tC)38|J7vA34j{6?Yh8iC&KKVHh2bU#~gP#o`S8G4trTI40?cn-o?#AM% zavlWS4+PSXi9zl$aH6HSQS9sj921=gE!&eg9MrN|K5Xok0YS*stp5NTF7XzHUy&9{qjb zI86ZFxe4LF#{Dp5a1#4f=mR<*VVHg^ug2RSrAYg-%P`h_i*$bcj~|d22{NXfN2TTh zEOp)ju?NcO#RVUT_;@m#)=|p_9js>8uRcU1%|iJr7H`L!mVLxTtO%C7ZiDeZ+3+B? z0w%0p2T7(H?4nz3=(3GrHT@Le##R9^4k#f7>p9kKkQnlh7E@t=)$g8W=1&6pr)MO>o+DZQA}M&Bm~O9_zZH zlI@#b&E_h*l9vLrVZt-cJAGM_o^q<@&!1yPCVxE!G^7e{j!Q%PDMgkS#<1cY>BUBo<5G0>1ao-5(v2Ly zU`Wih7Q;p70T|+Xt?qy0I*_8a;8e-;X#8>(TsD~t{f85vasP7EKQ{}4M!U$WWd!8+ zo6`wiHPr3!O^Ew=2w#0I=Vxd(z|e>oobeLF#)r)${I3ObLFx-JJ<*E83p&y8k~;hJ zlqwy1dX+AUYa$>L{ohkmL#vgM&!2iV#oC&>Y=u^aZ>C??z=byYj3NIgk)%fGBIha0kBZ6-o>ESnq&@;E`|FD#F#hcv_m0tjg ztQ_!;H3F6V230HL;hiht9GnL-*6zjmsPimD#__s~>!T(<(tcE!Q=rT1xasR*V{7J^>h z3T_v95Q3r&XoZ>rdT=+4*G1u2XFm}?C)_7fQ&JNF3Dv+YB5|0EKV_5cbc+}U}yQtV>S zTufTQHn!h?0Zy_PK+Bt;b8)h$SUh}pJj##e-hY$FCV^9C2If64 z1j$e*SiL}wScJAWhWp6!{=SPPqTk<9$?IQn*=1Qae7`97E?45!hPlDF7&&~PEQsH; zB(QE%6#R45<}z^!s2e1WT|OPqe0!J(u8f0#fy;2w<|ZV>f2Q@(9=OgX7uNrZLNO0h zlG;;53jgHOIqXK-E%FxUxK;7vj!dF^n}YE~NFw2FZijVDG~Avnjjle`HsjTANz)-i zxOG99_mgAqmt|Su>dqyg_gDcY+3?7mfixy|ESix>U%*c&eoB1iT!6esqA1&S+Qvnr zg|QdNXJp#s@q~{dWISBP^6#HV)ww^(Ri|B$y+EJ0v8Ije+b)5(jUSVPdh>T~!~G%$8hcV1HM@^6ivqq8u}#@{=FmZO`gj$*+SAF{Ryjt~ zG7{3ZX4q7ScQN)m8mPnLV>E2RQfgkm1ZQuF#ivPA;iqL1yty2Wp&#vSQnWY@YmYlF zH~mBJcsal_9~E|oq%`Z}WWdU^6#T{e;7xEJ@xHo-Uv9>SM&ajVJ&nYQ+2=6qya@Ge zQmG%G_m8o@Cf;* z|B&unuYfvljKJ}A9{e*L0PRHtk`hDd4Bj^)#3IN5MdWDSRD#1!4KO;e=%i&KNvQvyLx-q|4jjT>c=8=2MFKGf_HFm|YbR zi;w+6&|FK4m$HGopUrxRTUMr`Wv)9rzas0RgKXOpH`y1f}U*}KH@PM`7ci_&M zz0~o%BaDxp#&OMQG(`9@`682z8g`MeOeBjxQ!ESKcW!~tZ)WiA3XahTVNaN>tJ3gl zP8A-?JOf3g(bzpOmo_;hfs1oFdbwR@cq{k8nCbx4t|})HDzZG?gnEeiod8ZFp%Aj( zll-DDNQKEoM)t7=>5d)d{4q~ZFj<}r2#QA6)bF%JRvH3~4`U_Qy;^F)@y2#+L`8cM z=oHxkl8^YLep|1#*(YV#xn=^J@>7wW)2PlqY+8>$PSoHgU13~x+7!oadtm(L8Ggr@ zJiFEOHu9`w@Qs%$TiZ8;|Dvwr;cGjg+3^GO+&G6mG1i-hdGP3Q~ zL3FM{=ES}X`fbW=Mt1>sJMB4!yOrkP@)<46S?yw4d|Mwy?i{CH$C9{w=|eH?yd$t7l56y-LXWE!YQR>7d-I9z9Zv*ee<}l#% zm{e`4q#rXJ(6%%iyQ)4@;adw(%vuf)F%$6Y8!I}#Sr~Uu-$sPbt%qhA2L7E~0~7QW zA(`7>I_N4RTak$eJCe~SHl8M3OvUsUQ}NgK=VbA@aNB6|oR7mrE2!+_Bark&9$xHEC1)KHNY1ItTu%Eu zt(9d1HHwF+Q9UL!-VP!!w4xw2tfGc8>R_ zu|28WPHh72KYgEuc=OPyehaehr%)iZ9~Xxng^>Xxn0BOx=oX%)1CFNH9&Ugic1)+4 zEwj;MOU+ zMtcoTx)_GF!=1=4zXXvjd>|%!L4H_>-`;DCXLjzz-K$Q)_ncqk!|rtA;+Rjv@+T~qyQI_>|on(Ew1N%FO4}~LQPvLX+VY} zscqE5yWUgj_QF}T*D1E4g4a)HMhihfUL+hex^2CCyEv`d^puKU*xMMoCx*Puo(W1P z4}nBf8q9e%8T2%^!u{fCESjB6E3^;OziLG!B4i3)Tzr={yQI-PvzcVh`9-+qr#{@Y zZz6kBE|90!JwRZ_EC@4lhcpifoPKN~ir=4rj$7j4goqyh32!ObCl=FQ%PQ=h$DrEH zVJaFl6Z`JT;tO|eG~XhG#|q}-%TwDxaaIA<37AcS12xF^E~JBuCN7B$#l_KQ@Irbj z))pNm29J(FlzJN3q5PPrJT-=N&Y$`1o+3nlFQ%)_)JggcS$H1X)3~Tch%QO4p;w1F zkG1V#=G~UpHu`d=xGYu+Ee2+C&mnQTbyE)IKUs*2Vs|&5y!?boe8iCNC%yUCqRx|p zJ{rWl)&vXf4`6ldF6P*P8rjz+3Z4On!T!uV5FnfmJeLXe+>R!##)H02E?}mo4e+lP zPDY=t56MunGz6BHk)u}OSi*$U+Q4AUy`T-y9She_9DufwF7OE3%!s__KC6N2Zz=x^ z{!)oxUiO&G+H#-c)kX1bq(or$=U1?`{x}5oJ>hF8WYV;JUG!ZW3N`kBh)z!m?3DhG zoWCFyHNf08Qfjdi8Fka;x zmwVpLZx>J^1$U<4$iE)4^>8xGF?EOOg{gSD!H8X-c?mO{hl$Y58Pxjt3>1IZLnpj( zqt!>9(DmGY@|W`~mHf^jTKPiQ|3?k49oi2b(v7(5Oe$(D{Y3wsFh#*~X__y8fzEw_ zVCaR{=+rBxGiGx>&L6#z&rc;I?Rkh^c zuTDx9y|mu@B$MH%=#uyAv%#e+2foVd!!(Bn)atkj1o&j4hFdK5hiReY#T2UHaDl(H zMULotz9&2CchIt>uBa4#21YJ=(usc75b?_sBIKvz)%aW#eW8eUP7*k6yDKO7-!UE? zESdgmfGjBRfhaZw_nHW>vTv8;({u}-#+n&C*;)5QP(78KxNX$SM zPU)9{9^qYJled^-mEA*kvoe}e8jsD==|m-akl%l4GJYL?M9=T|Oh4sl)1U9>G6(le z2hI9z^j3&FR+MtAZ+&|-)|E#2q;h_!?>jQ=*a*Vj-$2=;4EBE=Bqgu7JVc}jL`j;n z_PT>uQ)$Zfg$D8OuKmrIR{FtoOEM(1B!HYZlLe)NlWeZ%FXCq&e@`2^UaOLJLtemR zN#2{8NASmuA84+46>muk;m*no`t^t&rfAOOeJq~N`xx0v9xt24bG>TC>s#^wxKeDA zI8lQ8o*8<{K~_FSSrfDT!=5Iukl9 zxO+C7$JN(vVTj^1_?w!^+?Q*k&YH(a7?;z0)SV55BDa|QrAUX&#qix8BNP>N!x85? z8h(2Pj;j@r9oZ`Q!h-AgGRy(NZ*_3Jx)tBVr(wj*Ln!PZ#y+HB*sy;A<|;6}QwI<7 zR+v77h^{3(*TrVMWbrQ~R92BDJP`%IY46B%vlbhL{$z45<|`xFF&zVvx6nxPkGO~C z(!+NZ@SbKU?&p}dvy;qGF=HM|D4*rJWTIifM40|-%fMB`7ykb@?c(>Fjyx`<3nqx+ z&E?s6?vouJIU$GTPAxX)j4G)_=pc2z{0uAP7Ls+%Jk%U@#-T5Xs8{4n!WXOI!NES7 z`$z@Lt)`=Y;(GWkm_utU(z#vOJM6BXhaO8mkWCluVAX?BvV-M3;_{T9a9BW1>s~Xb zQrpP(J1Q`?U!ZYF?=1OcI2mt8X5y=wTnsQxU{4zvl7YKHkQy$D1!EKVN$w9ClQpKo zv~RZ>Mlxj3uSExI-l$>3zF)-Vs5D!!Tn?fSC*j&<%8ang7lsy>v5 zqm5NySF{K>3q(MR&?&gFMV)7JwFq8CX5gvg*`PRcIeocO9e-GK@X2OQd-NY;+w@+O zBaeshc&7~QH5OwPZ7Wdas}VS>mr>75S0G6)uun=BM1(`(ZYAI$nYbgE6gNA;xrscdmbP!2G1EmyP z6m^`w5A7l$mD}-q`eZiqNF4L7vj%1@x((y{zBY|Gnc%hx@~~7C)Iud;x#=T*q@^YD zGBT-AN(xjvwd3vw9wd;<0523iirUZK;fO&hetVokbk=%dVXQ19&TFOm&*z~o)uVZH zqG*PxAhmxUk8-Jn=;HpHmIh7)ms1&3*(w)b?ls2lr|z`j=_t{uInB65rjqqq2^ic` zOgNYe2wlsfQ^y=Zz5FQbkrCoe^XK$wPAK^AeMFx9Np1`m?xO0UiSV%dHOWvg#67AU zi&EJXBrdgJ^SdgJv$LP`8&5_7eluoh=&*{N2C%Td7+>!z2j# zjD99ne`6(}^;3i9DWohB@-X3l5C$MBCk`F+F)FObk4S8>iU7 zNtaPF*(w+=Trhz%q5#(_7Qi}}hv?!}3U>_T>4K!!n5}Su<2p1jUPp?ElJ8~E<#^~m z?V9Wm$G~0qxTI00b0U|`8YKHa^y3KUd8?GT1)U~sgp6FLkw%4V$5J)ckn8kvZB>BW zqlt{~in%zqBn1YP36?Dv0+o};`SL<5@T-zER>PY}Do0l*z+tbd8Z}!}zS7G(WD7&YAXu87!Pc*F{by`;8TO-A6XSH#t)dx-7!m zIOxEZB}%Zbwv6ISg(OgK=s>wOkyx5^m!^!|#rc{ojM|~~V0YFTbo$gu(zF`#Nvxi( zJ;-(R^0=@|^(QiL)D@Ews=!FX6(jpC`0`1o!EkODl-*x}Qtu7vSfYQUgZ_K|goA6C zfzrd^@yrKq8rWe%(RtYUSd=%|-%oyLJfN>K^10lJARciRhV74DP|JIB(KLzUoF7o& z`pn#bRLh`QhBii)N8u}>+Qz77>(ORRnz#JXA7Ef2&tr2NN#0{fLJkr9?bc2@+csc@ z)Jtk3xf9B5w5VI*UpmuUlI)wyvDS$+4B79rdDwB0HYDDH;I&RLrri#%+qh1MC*H(Y z|08L3t|GXr9DLV#kf816RR7{Hs$i_m-Bd-9(KN!K_z)cYV2<9J&B)9R#E{q>IKKBy z<6S!&>-L%^`gv_vaT_}IQUO1_j)&)#PQ~G1Lps?a9~GCLMrpSUxGU7czx8%M z#*BXe-}oN*Ml`@W;{(zCGMBFT0v$T>HZQe}>9BY@{t7EbssCEBQ)MRnTC4?L zJ%;q)+{y6j(>r3xKSS3%Y{U4ORX9Fz6la8`&@Ry$GS%OZG&FMk5?%YLwxB${^FxK5 zsp7zHOZtSpm-30L_DtdtIu&R5X7Crqn-eqFx#W*aE90PW752?sKwte?Kr6fJ8>7GQ z>6grD@O4Z8EN4@4BCwg?I_GTT>c4qZ+E$CH(u*Z~<)kn~^(75lbdo-tb&s)Bbpgl9 z3g*9SE3CJjb>i<|{vRWfRziYPqS4N$90e{J;F{ad$=BU3wAMk7YK&!oSF0klu8_x| zyI-jGZbwGE;|-H;A4wlC(xh4Y3(4A21-Mhx%zX9wOANslH;kx97_};ez*Qay4qxex0 zxvUh-4;#_{PUS<7Og5Ahw-BA~VtRAZMf#w~gO=r&@V)FJZFC)9(Cp1F&=JP{zBY^z zUm;P<;ziTEqo&k<#u@5A*Je8^@9_T!-=%YtMaZkwk^BV7EbNaxha*$p(O-E-A;M&gM90SCDKeGSUY3Pb z&Cw)luLqVMN~cxf4DDDJfxE0;8kY#DfO0t1u?UJ96pT%~E)|>MP|nUTf^; zG^osP1MuEd+vry~MCww1kuF~@%XsxJ;}@llJL-mL+dM)-qMOLKDS0INcL9mNDvQ2) z*7Rz)1S#w7V$7?qQTK<-u{R+Ke+xb2N2tH1CT_-!W5+5QhbK*i7otnJ-Ooa*{u9Y3 zr zf-tnOo!+zELf+YaLq&@f_}>ayD$u%>wbP#r$3FYxh9O&2XZApe{cHO7kr1{1RDzGg z_9N8`#if=Dz^37E<7c<)@Q1r!)bnP;WYrEP>wGHyv1YLFyBu4$tQt;-#-OLbY1q_} z56_a701ang^=C`=k?&J{tFnQ8`|u@xs8?Xm%3R?xQJr*=M-a4^Q#!xvD99z8r&O~H z0=#0O?MO5KU;Qgo`Td5!do&*HTaVy7!y+_VWr>|HmeMyd26$;>7xXe3xHEn-%n9W> zCMCt-X`VC`~#Eu$dexOxdj`{K2Xs*NmkEyDei4HWv`rBM_vw{r+a(LNS?6@ z`~ATvo-|e?c<(J!xwaZL<_WMiBm^>zj&MF%ao#(QYq_Lw7z+(^`0hVyLFczVWPSJ# z7Op73&x`KD2c1#=cGY!Y_IVa;J<%>F%xp0*41EQb9p9*g$^uO2 zS_Fz&7a)e434R~zBGpp@p>(|ws;driIi-ob4}M&yG0mjQzOCRzNBL0KQ_9f0SQZvP zk;mo=j!EjEgNq{yiL>`&w&=$I6xyWlUlv}cxi>S>#kLY$dg4HTOOlP%zj3m|Pz|*m z62aczkiROv9BpMid1qGK2K%@-c!=vT`XiD9!GfWn{Npi>yD74_g&sD3{9Hi40V2hzPn4s?zt#(EO)FuRt5hMMUB2@FWzH>Oj{d z&a+(q5xeA)smOt2jB9iqmZo}=HUCw@IG3%rSYE;Z9Fjl{jf_Z@j}yeyH=^*Z3m~Jb z&#^k|!QVa-FLig}S(QFot2Bq?#3a&lJ&_RO2jDPM1qAFC;nb}Yaev7q)^Q?s)>{_WT~p#U{Br`86NEN-xP$)bIgsIKLIbpef!powYn^bnG0L-a}DR= zScJLZW$<8PGNhvd-&(5G^`ddWRphbuo2eItUa@-ODLr^`0q(z%24QN(@Utxm$@d0k$F)$L;=(bOR%emL zj4benCWE7!I2@jvN8WQx>)%JC;a`3NZ8(^Y`bVRP*7rVo{i!kf91e!Q^@_p=iZJZx`4-jDo8| zarl6{hxh62Lv6iPn6Hruf9Jg?2Tpt8k+3e>WVr%)f){a-++JwfI7T;aibCI}lf-$r zo7u5)A-;IZ!cBGppPBs&*YG0b?(H|u>TbJR}4=yk!)K0Cw=AqE1Q&j!89{&Dy2J?L; zFs+6XF!iBq^TF;=xJuTucBSM`N7vVEQEBG(>73t2Z#auWEV^ahnZ4ZNj96w_0 zL@ciQz7!WV#la$vezK+}0mBogf#*G8>@1swx_^A&r0@Z_nPH3>%MQbvPh8iw>k$+f zzX_8@n~BO%aklYU30eBHjuxGxI3q<95?{GNx^oM$eY_A-ccn4ATif~j%ua#b>r;Hg zXTRv(;utu-kmd7gj?-tO;vjr&D}AKQ!^$5e(C|YUdU6wR%Pdz8e~ z(Z&seOAgcf7vIpZYprm{^&RPK&BAYKM*;h|4&L_ZxEnaHv!D&BUU(QoIeusPrbTq! zi@m68FTk5&wvY@cN7F2eNQhS+CdY(}(d6+S2r4h4hmRznFz4?QTwzV0>{y0}_x|P| zR=Y_|rL{0{trb{d4wMafqW_k0^7CLd2#dy}RJR;tT#u*KThg#3NCwRNSCbxPDO`Sh zh)GG4hh@2q^s{6EIgwjTdMBn~h{z+7;}byUCkW!K!O5^kEs4MEx+~iC7r|cT_hg=Z zwoOe)Ft}@b=cU_1TmEk`$}T_O%jC3tOo9+wO0X4dR6!N;vLu+B3MzkIN;8K00x!l@Eb zDq0HgpRCHShV@yHNr^PPtB0IM1-Hrw%c$896GtIVxRr3d+az^YA_-WVL7#-tMyp-U@N$JB%((cAyzCK!yg)epyp& zQ}sy`HjCxpwt;=j+g)DdZ%z_4sn@{oJ=s8X3HtLA<_hY`Qp9tB!c+E zz*194>3vRK58KlnRh~B00>|j-kWpN{v=SdY_Cya2E~h0Th-*`v&^yqREjsAIZXd8? z96BmuXg693If;SU{h++pu{HNbdPWb~qxe zcvT1;pBWh7Jf;bI-;o8E6KJmB2%Rczg1bfbGFR-&XiLd3z14CJwGLN zHzNz>zZQYro5wbnZd(z!{)cBvn|>S z*jkfpzR&P0vSVBWGeU+)^^divYR_q;B6rlEoy;hw&BB}JH{kI~5uW7GXQH+i*tnf% z@xKk?>~otZ)Mzw^e$URQ(TqR8{`pV?2~+q6-D!ojTht+N$LWV8R9H}tE3YP_>dYZ5F7CzYxvhA}xs$|Sd`J5>&BVn$9M zc#n49gek2fU~GI4CXQ<2bIieZWo6zP-zG9?72oE{hG2Ygc_N7=Qqb+7i;c&8C>=RR zhxTXTUfUKtYc>IT!Hu^3ZAZfonwY2?O|6f~(1UU+;8PMo^b}Pw>`XkJ(69gu_e~>n zHqRlky3TAN+u0~_Ck@719Y|4o5yalUi5fGfV#}R=@+iHS{yCkF3FUld%JYX*cb+lt zkNI(|)#E%;VZz{l>NdTnaG3Jtlz=98!ydCtY95$}C6(Fm@QE~=vw9g*T<8Itvx1n} zK^d^wOBaS>j*!>xzhSnCEe)Rhhnh_hqcnIK{qtWt`R~zDb|kfj-U%9|Ysas1*?|2x z8nG7pO2&}R*5HMeuOxxII81wz4F$8s!Li^DQ3~LCbqZ|BOQ}>`S>Hjb1fTLnoE0JO zx&)tuDPQQX<_L}*(C{G)ple+3(W!`|z3(y>vpOmGVM;PY~ z6)txPu{vqva7TOyMK3tA;t#7Z!Wru) zYZO!kQfv%(Yr@(XC6hH%iSQ-?anNfOGU=Y9o{$i!%l^jm@hDa?L8Dl zlrCO_E59c|>7g7<40?&R*CjD^^amNZ76GQpQXrE5h?L!lA!VyHxU`}aTdQLVI)}b9 zl|85E#llwH{OlsyZ?b3KUojzvhg8ty^_eXqmvjAdYG4383RyKr}*}F=|YdSL1F^s;%u=aaA+u^fkm) zMPZHMud)ELRbhUF7hF;A!oCNQ;3TZc>$~!kSQb-Yk8i@ilN?YjVjtcVi-3JzoL?b_ zVS4qaf%+Xmh%+sv!)JLI7M28RU!{?kZ_dt}?+mq@j$+NOEGisiLM{!>g-4Ao;5y>N zSf5`AJ_obts(qHMhp8iVs$B)Q`%aVBoWF15*L`^Df)<_&8zud4rATB8+f+qG(B$= z32QE!;?$%%IQ!k5+j$9L&DnG;7o7pGYSm#V_8vV{R#*S0ypH}C)Q*dm9me^-6M3RA zL?Vsr>4LqNF(yQfT`ws^(zc56Zufjd-$X0)Nn1*%`CEc`a|{Wwlfad8q*%Ma1t7V+ zf-%+H0qf1Y8g1*MF!=r?8naW3ZE;wMDU&~AK$IT4<9r5VFjRuWL&w2b{xnSBb|}hS zly9jw12Pt?W7q{(%$T7Ld(yvPRERY2dV^?-a~R&-uflFBLYomGA^hTB4m~voj_RO9EYUawa{l+Q+Nu|k99ndH)utLVFMK5QP1hC5r$=qHP>J!pYZJ5Ml5!W!<)-u*jGOl*zAx}Tw2kAnNu{-a8d>IREXi9 zw?3HgwT!5B=A&U^GE7jK$TO_xp8Ky0h^eR@JsSRkdUJbG8>L|=Tx*OQmR;wMxJJTwz`_wdC9W7(-yKJXC|`pg{`1B&7FTZ zArUtnJAuB_4xy%L7Zmi3;s>jK44uSfcy0^OzY(UqvQ`sb@{vH4J@t~Va$QQ4ww2;x zS!M42X3P^$+l}MiO|Z0O4Zn5V91Cnxp-X24G)?_W=S+OxAaT`^OjA9Fg-efe{0MoT zc6uQyRjT5u+%2bjMZBK$4c(t@P?1X zHO7*cv{{2yZHeKlZ45*^x2x!68O-Vwo&@FWV!XAYgiIOpf-N^QFnP~h*z@QC*R`91 zxyPn}Nf(#-3#dW$=GPn(VIuEN!X%a{>Ls~9xLopv!>n!OCLH}H1S9umuyq9v5b;3< z{usQc<#E$Vrlt@c_Fc|ymd~T&^Zd~Cm?X>p=?8~?i@PD|)6ipOU$jc`C<3Qyq?i=tlP?7$;2cHPR& zs3fEd_twthO{BSO;67!x>dqz?v1=<2E?_%@n%D&W zQxI!xhR4&*(f#;MoN%LoXkJ>22V6pNV|h4EKNW!kqc$Y?^KK@)#T?W(?q}VVLs`qS z4{-hc5ypFqZo`CqAIP2El3*>`51X{Oj&m;K;$z#zU-4brX0N9o>$m70K8WbYz$t+w z*j*gLi~UF$UCZXic%sTHM=X2Nh;BPLf2&*_|JIKctV41#o2#nAiq=bz{mFH>p>is| z`lSgswv~V=r&B%(JJ8ye`81b$!c6Z`Av^Bqg zB$@WoYYTm;-Hs@n(r1R&YLi%r!1H*v-2zKuUX%S?7xj)a#ry?jMPv?l8)@=00-HL{ zZ@IjK{F^S!ldY=br#4wZSzj64_*V+nQWJR#+uXRDKnj<8e?WId?uL}4O4QAHM`ihm zV7N62+NbC?HZ8h@yD!UPR_7HgS>Q|63)W%E>a#R2^0t+N z(G+jU?T~I9y5<2BC)P1OY1gUuSQGXwX~EqdZ>jf0F(ON&sQTYJFs(?Tn&-E`#^qu> zwa4>#+Y*Z5yk;tHeASL3=l>G#1g;Zq#ZjBon^IJ`>IWPYp9Y5(3qXimBvh{ag`J5y zurJAo9{(2sedAJaZ@i9*bG}XjfqGK7E8?lJ&UhW(>NAm&2?{gG4uP4xEmf%yUR+fjj|Ij4$BhXOEDIW6>jt|@(M8;AYRpAhw^dV0lAirw5T%C^+5r{cHsplDMpyo-H-Zv@JiBAY@| zyvva$MwY=W#S-GLVu2AMsWcz0$&`h68^gNx;F1d;_(uFw`20;2^t^C|Dr-aZjD2Dw z!S&M>E*FL7g2%*9>M5hHZ~~UiZlyA8HwKCI5PegYCL59d1PTF+X)-+lWq|JCq(BlR741pc9s zy17vMysglqr3|J9?jvC*14yf`Eh^-nhKIvbh1S~tATmG$IwI>Jd!nfD&i9KjD*8sP zkG>9cE?5my(+`n9xkp(#TNh@(yhPJBchB29+Y}4JN@cQX2 zdU4Dt+T>MFzs!@R2S4%MtRxwNE{ga~F%vbu%tf;vQS_=Aq0aOP8TuRsx7%9Dz9o@@ z12-0s6R*c$PiHpmITZ`GvlfSe zg?S&`=R3*a3NmQgw1T!eMiW75G|u=wiEH_;hcj#@leu@s;-mjc*t}a_px6`xulRRv z+QUThJMRRnmYgVDKXMTj%L8yxVJ2M_SV|5L7f`8!JM>2OTZW8{B+0AB5%ndH$?d!p zXxFTu$@{gjqb>|(0-n*KuaWppgRAunjAcI6H<1fG7wmQZ1*Y-$71F72lN!AANBykh zWapq1HO{*)$ey>44mhR}g-uV1O5-?8c68&N)br8gOB$((zDh*YM97}H9mM9cJOnv) z(GOc^Vx@EueX?l>NL0%~pN19hxX~h;8@tGri6@z+FKRf~zlTc5sKfGTQJgo{lxA$5 zgVR6k}#?>QYJ9(*>fNTV7nArZs%OmK5rMeUIQ2U>bG4Ld%)#}6rCh~TJa`fk}H!n)YN zO36aVep*K2zkBd3opp4DY&2t_^ohE@dq%ERtp)1SN{X+XBs=wrA*VV9KD<^IwmC>a z)*;@?9h?m}8ZXkR+qCIKB^M?=!HGPm)+Az5s_^A=H}x!B2r)?!aK!Hsy)dbQ-W(jo z+_ZI~Uz%UAhPBf%bLKB{e7+Mqw55S26i=g5C(OaY_1CGNQ7f_De1zCld}Cx~`iaMT zS^7>^Ufd7PvQ2Tv1nVffo__!?Mxvm@=dN7U5x%C_zvfqLKLoYPmmc}N1U&{mz z=A*B8Cdzx~e(G_pOp56Dqf8AgBI zWHwEsfL1;_LmFD8AlONadGgH*yQ30OSHBpW`5jdB5;JPp>`fj%Oh&V6RhoA_3}tPb zaHR7w&#r+ECuBwFoWE*@O90ygm_*vUZmujxKMBlc%U@w2Gv|jUz#!ml6 zQ!4}DUSAe;CPuUNi|t5_`Z?m2c$zBxtR!toN@PiN9E2=#hu@Nf{Me|*?st|Z@jQ!Q zrJEf#mGqHMIR>D1*Avnf?1u673!u)`lfJpmXKWLGWBy?YF4$U*>)tw-+wo!}w``^% zWTqU3Kk}Jy+xQMRvKOHCNGfAtX~C#}Af&?c3KM?UjPdRfgBHu_kd@14!^#H9`k3o% zg2qu;u<;I9?Qj8|(;j4c%Wd||@M!qZvzvJR5v7OzI#P>&9>lwW5WmTMe(gaKjwIj5 zXNqB%w?>G5yGqEA#Z>T(4+Vd(xnSLVhHalZ3A1?@*Osbi%zpBe35cBe|Jyg{m!2Wp z=>_Ji{c-Ym>rtq9(8{J&rqj{$gXOk{Zhr)@LwCOz0q+69mhvw%|oB6F&J7fhJ ztPW-JL|f>IvNe3a+?N~YYsJZ@caax1hv|^#EOIepHxy}v!2Nm~xTCn1le+E3jVro` z6P30Rr~iJ?{PHt2W>^i>mt_cacg=?f``sqJ3 zNu5tSt0CCQJ#EwD{F>k4n@dZ%Q~EBP{t`{@x04?yDHq8pE#1iJrVNp_gI-whU^J%x zEhAPg+u1#L^2rmvSCVTrlZM^uVnX6}!DGlL#wllM#?>(FuR6xqKjlCB+x_f1)*J(S zZ?Lv8N67NbZ)D{)1yC^1hrOcwl=65Hm6W$5PJb_uv5k4;dut(n4l(6#2`;3dfagGU zJ;B_ca-8**cJgA^T@rcCgm^Tpg;COKFwd!sr7N+TgkqoLHXmi)T zaJ)`lx$99~^JDnS&==oM6BSn6e@&{SP0{j%Bo(c#XRo}_CS!Nnf@k~*{I=-{dUNs+ z;O#>@k}ebF@eP7tOL6vM?HHIfb~|0SJR1+n#WQ9JK>nQ+(7PT>$$R~+T+0#%Zb9E_ zF4%4wbp5pi=^9%&_{ulO_w8KU9*9_dRMFE8AJd2X_M3WWy~hgYh;wgb<*Y{ z2gRE0R6KPqY%cF0!ilLU)g=KZ8YV;KfI72o7Jx=`2A*8ujXUEaS%dyn#N~D&X*-q- ziYC6$^FS3ZHXMO_lC!}#&!2vKYz>h*7fF^6rOT{Ma8s`(9hM!Tht3QzMaPm!iKh%I zy|2PULp&JlwKAvXFqw0|`vaFc@mbyFN|3wp6-{Z5U@q=R1X;rkV19%^*78ZPwL6x2 z48^jSmR%t-ZIoF!O+bZ-&CGS@YF2rF2hYFV33H5Qk;cQO@ND&180;)zo|W@^7F{FQ zrCLm?!mrZq)cJ6Cp9~c7@8Bqx<6!i5JoNEA8NZ<2bkp|=di(Ku^oY1XiZ>Y1e`O}b zbOoPdbrgS$VnjU zH%F1&?Y^L063IkPe@4nqOof;?KICSPD=E8@M{2fyCUPyUB+1?kgfrY>NAv_>tL5e$ zChpj7UWNnu+aSiTi3BeGrbil;P+71pz*mQ`##09YLhy@;IIy835ut z*3iTuy_&WPhAzo+$C<0jQE+oHWT*s_4-fxPFKKBg-ORi2(w*>5Jb`$k-Yom9qCW1+a zAFKab8kZcfMvJI-WZm{sZNaq?A&MtCl{q-Y1Dg? z_0O1mOprk5Ar8+z=kw1#*HNr}KCWT3IG2xexmT+zVElzb=&w&EO={W#%Xp+Og6}db zB`IliVyMeMZRpJj0F9n*;xizG6{BC12Q`mK?5@dF-~Tw#y;8>x-}uH}R(wVpE^a_0 zDOdEGxD3wP>w)t4Kz7aMsc`=3QFc)2JL$^t2lv(dt<9#G99@0|3d2gN;B5%wboU~Y zu<$7A?YTiTaWmfJuBF_iz0KG_eDu%GV@4+#0+ z;OU^ncGWHy_+fdAD0lhO67E&oW=`(_@2r^scz5|VRGIV)D!31@*k&x5`AiP-^f$0; z%41M6Jpdj&eoAk@JcSpi5;T$y5-9SXhN(*6)jYt0S!3|A%`H0W#(Wqsh-1?>Eg~}7 z&UAN3IC_1gZ1Q7d=1k@+jlD0D&AH?0s`Q7f(bjQP&D06MX>}7L?ag%Pg%-X&IRPcs z7Qg}X%dpoj8tgqY=&r5O?0@MhSTS%yP`v9mQ+v;t=9tVU&I?V+Djy}mHvX<_QEe*R zom&r5i|^6;vpnBl>Sd@g?1ll>WmBQv2JnpetR5!Wc`e08vhbJvR+EeCU1dEzVkbN#U*eHw_!BB{xRuw z(`f1KnYdhkA3fOefc1-hLtdY8|{KX` zaAk$}V(3$=kIY*;516&)4f(r#Eg62Tfbb@szH)yJ6DO+*gGw*Ke13QG;=gE;r>u^b z@*AjG?K%woIGfztSW9bTTxnE?8k?-FK=d1~Qa!dNhQ^%EMkNCf+BSEAUR1e83MTl?YYbOev5XUGhp`|QD|?C0 zvUEmH{U2i{=TFCVA4e*dOmFk9m-qiXad&_wJK6mWxMwtgFz+ek?An08j?SZ(dp|Ma z3)|_%hGguK|ITWqHo!oXtng6IF34QcL3V61Uk~j1WGetyg7;>6ZV< z-498$PF<4f$1J8hRvf4Ms^lO*WInm5KL-t~wou7`y426mj7m;xgor~^A!T^yu%r_SMHRj>cvGMGKI%Ap`RSrLuwS<#@l|W3#<|y({oxXXONPdp5trZJdKw5fu?m0ccy4~P+@#_@X ziEri;`HVbrAx4GI-DlvT!?BFQ$7a_1{7!f={wTJ^?qK#knnflnWzgGAp2((Iq29J| zQqy;rOt@$cN5npop7rtcd+JSgg_kc(kdeXRzNKvLV17-#ad-oCbMxe_h2_9kVsAJb6zYqZTulEh6?rM|gM zf+wRTL3z;;=1Q3z{dwmNoxE<4=H9Ttp|i%cw}jvp5i9cLx+h~OkVkD5A^q!OfokiD ziPD5MOsH!zw(3Vf^)DHrv%+gUcDNeFm-kW2jR#0{Lkc@`r~~}%tLR7VZX*8Ik#Vb* z!X9NF;1_5ES@NfAgEOV^qbAR{iFe2Gq4UvBd6MuZf1}wudoIqgbA^A8wNd?VHJ)=V zBC1}-_SVp4tXQbRtkiwX1gN!`_LN+8FM5F(v=5)x*b$ z>Rjrq<0$r0R(SWl8NPhHh+!72VeVZrq+3ivsmw2+!)yi()y)9=I%&vK)`hbfzreY! z4&I#0L>+r|kUA_sulLhL%*oWJeX~Sxd=Jk-dk}(|{~j>CErVEc(-eOd$b#3D)fmM4 zdDNUWQ2voW*1C#ur=Ez@IfqZP|NYtvfvZC4f>v)5I-d77Db_P_XBMF3t`p?hxd&v! zs#L7LB|w{B#=^C|*D+vxH#sw414iuzuy~InMDXt`+qntsweWK=r9X`}{uSV=Ra0Tk z65ioEl0pk~Z;>S)YjySSB8b1!Vv3vQf-R9HnOv&nF&^Rp0L^LfTnGJlunyV%2 zMYD;nb^z6im`3!)3mBEUaHy*tLt^}zNO;Q#nZJD@1}oOU0h6&1dLj|sIlk}A+2DX6 z9tWHb;@fxsU=z>ss2Dx~J}+C?F{k-1Tv8Y-Tr5fJiw&{6?FhsUerHzh3m2Fk$|Ucl z#D&ds-;w_e7lES#r4wH61J_C|3^w};|2?~biA95?`N?}rX}x}d81tU2`0|Kd+0cW= zb~)5k-a~jDF z+C;0thP4yQ;&HW?RCnuba+2SuhAymQB;=M5%S35xDV3*kvDM7pk|;7$CYa_HD?kj% zVKN1<4bh@aPJ{FX5oqESP)Z0JDy=V|2o{?$;e z-$||%SrfbN^VDR_Px8?#l*~=hLEEt}nFl%=bYr{O?CWH zxf+Rl51ZoTfDtMc=*Q1xbAlsr+nE=X^Q&R6TvlPkBu(l0J(0*HB;yAE4H&K94l~`4 zqma8uFU+1GbQpEJw#C~QejYkYY{g7SxkdqFda#u>yLAH3;34Mh6wR8F?|F>q)3F#m zVJ8Z&72$X7STfc040zPZKxpXz3~Tm4ystc{cF4h~9|5qc*PP_u+kx|&Q)qo5Zx_01 zNq1Cug5}f0XOL`GCnQ{qLS~bz8se^J;m2rdhVTe7HOIIZ- zz}X5}jL`38_t~=MA06+IDJk}#_j(83xH<}#?OtBn^R9{RnfMZ(UO5HNZZ@OutFtK9 zQb^fxip2UNrW|ti1yjo-gSk&8g{z&NX!tqcx)jY40=g2KC22o2?g-XxtZGY zw^GTY4t##g6MZK;kSXOKNyM&Le6{f*eKyvP#N0bc-=EYcHv2y_GAg^+`%%}4^F3GU z^)DIU=@e2;t?Oj$nhxAD;)%-YPirT(7SdG@!?2{Ul#NeR0kLJaG~_W_rH=6%IM7odoz0+S<^?GCoLf* zW?I5^GbaiQ)-_?r!6jHx$=@4d7Shw`Oiqh0hvE63Y3t_+*e?Es1ig5N2I6n1+q4Dn z;Fc19yHlgN<)6r^(XF_&V-B__8nIuMj$yZz3x+LrBFX`kbnCfjqGvW4(u?`ur3gnd zclK7&nUsiUyMy2~TC%rv-Oy@g8cvkofzz_);CRj-%mq;pBb!4e_6gY=dp{G#Y#7JM z=VFBXZrpy!8~;hC(-F5Bkh3TTjDKE+RoiYsn{_#cZ%U$nl2S=m(|e|GPdqWcA_6ya zpR&jJZqYE$GI7wXrgBsAapr16)Nl93w6bJ!C~y&|Z*V8aZDOgJ_jz`k#c}rI!83HQ zNrU`&r%TLd1i+O9UE({f9R)$>aPpXZX6)?{CdwfcT>W{cljH}q4;7%(raCG<&jk~= z@8EZOXIZhQ*Qs1*1>t0Jh!L3!^+o%bjX!&Nibt)SF1sh;rx!b46P|RlyQlAJ)^Pv07mBVipKqZKKnxUy}goZ{%TTFz>W8z|Xwjv3(923VcFj4tm>ux3~R;{(oN z?PGoH5R8XHRcYaTPd%Zp2Mf&k_jH%Ua&($pLA+@Ib`Bb$z(NEXqax5yJd{jz9S2R9 zF9|w?QRJL`3`&*V!a8%_6|(sqU7Qw9fA#JcbWRwDe?J_jvp4+~JeYZaBt4Ddy$fw5 zVGHluQ^=+?zn)Yr=5GO2d+E_&XNU}33cDA^f?V$)9J+A<&QDy8t7=m4o0=W;YPeBn zmnY=wDS5J_@hlY@Z$|Bt{98GH5B&4uyHPWAL26(z{oN5pdsh)UveJc0?h#=lm#Q(R zHHWD3rXwUGLA3U|Wink-xBy#gztV4N-nizjJ)RzvASSKZw7-8AjT?T$MlC9)pFgc4 zLpr<@ZQK;lQtyPnAz$H|#Vw+!xRyRrai+@aTgh7ySCA@vNP@%kVaK)U^m5t*s@

_^N z6-TY{ZVj=R*LvR zLYFsw9ty|FPb|sA>(bb&yPk+t$)Zj{|2$)vNZOb^jx$sq%@xgT!AISrxpyaY(Em{; z_v5E4gw(q;X_8^Euk9fTyrcyYLsDe^g0tu%Vu~9jJFsR-F}XE`V$PWC1Uo zHJml79IezOxe-@CF5y!qT=D%xN>XAOE46O6W8*^DVb+Md(?dv~jx<*$afn_}y$|NI z7l32~pApKD#er6Sj*8p_r9Aifgv4mhZ%PpUt80cE1~HjGF0jTc8UB=af57%cvY|+9^l}B~uEUrpGq%QQmouHYNQi;DbC^7y z%ja<>2`vg!X+kQ41D{Xe+1W~rPEi;AZSsRvW0ykUh%anB^&7;W1i<0SRtU}Sf`=ch zG0QNK*`j?C?Dref@~4yG`rL~sZ1$wi52up4zMp7n`kY)HKNWU|3h0@C&%s?@1{w}1 z(AXQP?CP`U`Ob?W8!VwiJh;7#Oy?M)u9O9fj@yC~zXQE%(GPvliEXU*`|ZyKIUY~?I{K9m& zJ?0fzS>Fw+XJ3$?g{Aa|OA7OLpd2e=v*2n~GnIKJ51lV=vmUEIQ^j)%>_QzU^z*&} zB3sqrdr~(XXHs!u{y)sjn1rMJMPZEnBAoeH0|OGqfQE%DUdpg%wYJEUr7ey)@$xe8 z82-$rE59c9cH}U>?0j&F`$-}kQi8kt8<^8vC6RJEhmHkW^jy{z@Kk5u_fARf!H`#N z=YQk4FJg^k0kMSReLPF(YXnB|{1A8VW2EIpK3E!NkWp&UL`w4|ot^OmPeqU7zRIm4 zkHxyF&SE|jH)RqSF5f{G$4JAi*WH9XAJ6yh`f*w2Vs6p-c@Up+9d7x{3ja;N1Y%D+ej3|InnN3QR6&fH==1GRw-q z`P$}SI5z>Tf96t;mK8K<&So^MwgH!gtI#fBh?do*!HtNUp!(`9)vOc8(cb@2!^$Q+ z9>!A34=VI+YbX6rycA9xKaC4I<jU@dFGGXdpco%jdE(A@%EZDCS%+d#*O_7Uy|LCW?@qE6crXtAOMB zRXFd{b827dM(mdq(~6*YIJq{S{`H(m5CS{dcaOsXnNHd#mx8Ms zwBWA#Au=Z>8KRc&gG+CC2h$Z72$>WKwyEm)x+DdeE`(5fQ!rgS6Kd37(>}X1aAndW zd{lT7j4S-{YOo24ONF9o9PjZi{XyrBOW-~6zsR+*1N29cAGV|}1kc%jv7li#HNQHV zE05vuaXoPC`4dq4F%%sgvT5v45qX&|AikS^V`_IYuDP~_D1E1-WnwW-9A^drrk3Ep zbPUGDrcf;do-yCPkETtHp+x>fY~#sz`dvU3%Q37> z(gb0v{4w6|T?g%jyt_wp9yg=X0Q)eTIpm~=k2Xc2%6Jvrtkz^cE#_A3Su-D4TdIJb zDkW?~kU!(_dNdR#U8c{rIbvGqUYwf!g4xICB44{Jv3|VcevRk|J96tD^&Jt%8?kZB zcFnouRML4Y`tp>&Tl^w_mPc@76-9+Vyp4o5-xqNq`l{U1I7M!Qfg(5G-~?Lwi*mo# z_+sq8o0$Bk9w+TF1egXdVFDhonaSlhSK*bSne1j2 zb74m#hw-~8%a*O@T>(w_`r1!waQ!bllIf!**B!XJtIC}CunoRksm6{Hy@wOdU#Ay_ zJkX_E3{$he;NZqe=KXUQGBnN^s`G6zKx;hbK3$ze@i|IM>pE(y_Yrb7R{*2Xfl{^7 z+=qQ1@SEf`&Q7L_gr432b9UEIV>xB~^rW7IfAWIfLu&XWz6H*`{6xc7s$y@>IQaFc z5<8QPxMQOa&>D~LL`Eou6Lk38zRzXyE{)OfO>!vjO6CV-?x zDHi+0b4p@AakkZc!Idv>;Md+){Qr1Fa?YsX`9FK$`I{0L&=ki{zl^Z;Ng?~*;3k?q zcmr?E0cXi5gXBb2s1*%?<+5w+X+e$g0V{GR6|fBg*3(P}D}bcTV=m$PAB zXW}-&efI9tY_cw)86B?3b0r?5A*-W}F)(<}zP&L(qKf)3M#hVC;rWzh zqjIVFJWF!tQ3^XhTm^1Da)gx3XxPeg&J9nz!SL8*yhCT__m%Cp5@=%E8@txjw0g!_ZP8SlEhZ2%q0T_K5UzuGsOi<@l$~jz4Dju zR!zOlRxjKSD^4iG(`ja~_PrZM8%&1vKXzcw;Se%qkmp%U%4hz}x50}MABnhg7Vn}w z&hD^iBns&>;hk|X6Aw#lJ9iR~Gkuq#v(*&?Vn& z>aos29qn%1A$pJF>GQ^g#CqZ!p;U~k&?IbtygJ!I|9BV*zxSn)#!Ehgc$^}O`mM>o zk&h&}RD-ybEW`f6Jo0hf2yH$S$_7`7pp16so&40Qr_cgI0-?&pWUSX@6*SOk7T8HdSWTxSU1tqHL{a+kV? ztMh$<32=KTfrPKmfIN9u`1w79^=ey(9kG1o;$s4_)I368=kwV-M?D0~xeyZNNL+W? z;;w^j?H3erCrCt=F3f+p2GotNVEmc3dI7C@}+n#VB*f4U{dE(#_2*)l580=wrXU7W{=fGWwWI3s?lk9EBk0f3hbKF~ z6JM$C1k&y?-3QgNBF~yy_@uE<#v5XR%ySxVDn;^#m(nv|$Kn0<2=Xy&BP}U9h{l&Z z8Q|X&%BPhHSUo)~#>;!Yt>N36Ur@)Gv?II^T&FL>aJ?ea&W8cd;(3ors z+^_Tu``?d6AFEg7>u^7Fu5JMhS04*M0@Go0@p-aKJQR*L&k>dn#*>ffV{k6d1q<9T zOCU2_N4U#yx$sJn6;wTNM^{lFPU`(=Zso;FQr$0rbyIz4j{@I4-Oxe>Ti-Kjm`FZb zN5b+50y7U!BT4gSB73-sejVHjJ9;DGh5a;WG|Xh0j!qX2fAS(rJJ#c$8-OJ`3fRhL z_VhQk(8ot6Ve{P@*pp-W8Z9bZDbq*l_G&GGa=Xgdy9 zTJ!vM3k=hL!!FLdNuON1Npv2cCVM{zfa(H&oUl~~ZySw8-KnPBtjr`X=|B~AR^~IT z1`#mXIvn&{#tYpto`PM|8X_@a3-wmC#lc|*3=Y}|()2Q`ad$3mZ?A)b&$o!FcQN(( zy%}b=^$@vUzKfploiY9)CY*+{;3@iuQLWNP`JN~?Hhnkj`uQIlF0z5vIE%obV2HTu zUZsj*(hzuXE^IvN0FANB`Oj`Z(5UW7Z)7+_{qwKzDo0%C>RkwZ31`TA^H@09k;^`8 z8in_py~sM5eoD9dkRMBJ$&9yWiTvYtG+@d;=GxX|TAIfHZ4IBp_gD@Qqi1vBqf>Ep zxh<%!olbgU^obK}0rOL%;eNgpthf|Lch4CIrX990ywrp1^YG=WMwj5%@)Al!-;pXs zZJLzsz?BCFbGu)fa!SdI!Q=Tn;!qk&uGHE>+@ya*E87xkqoSDa2F}p1S_0~dMY!z`*$_=^38G6~Fs|nrojhQRaKuJG&C7h{P-UkmuDE@_87e zSp{t2*Ab$_E*D%V^yV2uE4i-KKHQ7Tzp+O(^;~; zE|IL#X`#DTl;H8n()8NBVD7t@50|(=ffS$f0IJ=v?D!iwdU5VFOW6lJ84hOPv1w_zhRzO04|7fRq; zMIzZYSs&(aJ`8?QC#lo&680-&4(Wb%bbAF$U+Ed6Qt=e(LN7y~wG!AbnE-i(1|SYb zD5JHR3v^k-jqt3HUw6cW%Lbd+k@_=G7MsD2vXX%@1;?Rk((-TyCzC-wnU69%en*e{s0`gQ^h+CZ%{D|q~~9srq+8dQGccDc*-^d zS4_Q*vGp!o@tSS)YMKyIZ!Up%pC^%%4*zQ3iG8HsmhQsZHeb$Rauzp#p^y`KaGHc{ z`$C2sO<}%VA_>*x*}|D);bH*qUX?#h7rGvRWWf@~%qNVD+b2h}7V{iL(XRh@%bA?% z!}#A#RjhrO1($m3$Vv4>uwJ8{P26yp2xf|cQGPeGWO*SrWl~g?namxDFk@QmjLDK! zV&vf8ZgO|&8u%gI4byhy!FcH%Ab8!u=jT(wAw~;i%1^_tj-7Dl_$)}?8wOtYUXqA$ zbBXcCMeyWADE@dpo10pifFpmRN#$N`0@bsjJT9Co+mpksH-C=?g{Cx6Gn#2~o(iWI z@jdTR-gv^aN|5uX8826i;~xCC0-cQ8$=)>#OuKY~MEqKe`Z|-K@!(&O`6~)lJs;TP z?~Eb7R1;s%;+?)y=ZI-IFu&#N*yEP-z_**i(JulJo0dym!)KGS2WCW*730(^18X15 zlff0|#-guVG^;G3!d5h^*QlG$Lw7w9?p3EDx9p-Ao!_K@8Vj{BhPg|NQ+iQfUyReU z(ZKw3akx}niA$4m!P_HdxK}HS+*tmav_#wje~DYH#phm9Q{q7;8LR^%yB!=mWjPnW zCXBnDCj+;h-lt7^_Rz@lu*DjAcDL3j&aq9I8Q5zs=(025b775en%`BMzOQC)rHta7 zWQwt3S~>Mzn}?OnGqH4a7AVBzQTxHKw02@5{k5r|I2^b~10@Eb|EVjVeG%i?J!ZHy z(Oj6-^Ary6H(_+|MB!(N@7S>_o|%~W5gexM0ugap;r+KZG*vVe`Z7o1iyk4hT{aeE zO^-vT$+?=pmbWl?#zJlu$e|kZ6Seh)2)G~F!a9`$7WJLLOWudb z(HceJhNgG8c0`XmH|PZ`#b1I_z%gnNlZlGMC&||BIW`-=dxQs)!iQOdbt^tnAzQay@?+iW#iCFpUG)Aje^Y8LNm{aHoTlV%4=2}1h z9ZiI?lC$uv`yTy0JVZy{$H0y`uXqn$B7|n0r#u&jHTk9j?4dyL@3Np9$JLX^+Vikn z{wehp@HtH57)UJ2#JCli7PyA2-} zv=PQwjeb~F3=W>PfjTU@jzENG9d^Gcb%iTlch?-;7Gm#FQe@CU9naV~A(3U6*`1G5cY0v*oypYklPyy7 z!}Lb@deYkvK!aRlx#19R44FP2oEk5}mBca#m|F;?XD>5->s|3pT09=o-cB!9t-x<= zH)WOy$jWX54`+I_`-6RBn&cYwY~B z6|eVS#uaAK;C9XmQxw->VKeV*uFQma^FzVR_Xhlu9D(Gk5<-!=LdcFSfQYTJul5h0S-otfOeQ?rKTJEh%gpXY3vlK^GaOlFhn|vhxLc)z3{pir zWzZgNIXDo`SF2w1~m)e4cFub!Ng=p7)sC{kJ!Hy{v7uo)@&p00xyQ-{aB7+vl&* zMeSR`UoxMRD37Jrg64AJwvODP(%UFayD_GM(z_Dj=-6w-eZMKgm6&zmLFFK7F}WO< zT`Z?179+&aeHSVMb}PtpVQX!v zMtdPf@ENJ$W6k9CxI(PgC;-!AQqX$u4w{!rV%)U@Sk)~Goi!@B+bxH+4%b9<;=NPO z0^Bmb0)%N4w#fKHq}N@>%sv9%C7XlW7HQro*-Iu{)v#Cl`xuF;BG`F%F*I6*Q!@z< z9PRd%7QC=0E(PgSAn<|Wf-%rrtWU)GoAJ}!&-8lmXvi2z#<9~vIg|TZ+yzTH{4SUR zHNJt&A6rkfiCc()GP1QkyAP8mov$qSTSP-?D|k1r>kdA#@hc zN7R?0uV-ZQ*~?NIcXcuuyg3yfuZzPX`#7*YsYVs2U8HU{XGqg=A?;jz9A!sqpoz#+ z;<@@54Hccu9XT|GyX~UOMarc@S>0Kpo5Jq~&hz=lr~Le5o`8lQmH6FJCWcIqhkv<$ zNy0Hr?9tkWE%jBf+A|7HxwL`aM`bkPJF$C2#kr$0_esR;3(PmKCYIOng7)Hcsu6aJ zSu|!nyZV9w)a8Z2@Izf9Gog;|tG9sVsrLkq_vX`$0j~s-9xmu^s*6`&GGtw`G|c~$ z1_cMcpx6)f#2lWqIhgycZOM@ZHRj#!(r*>GCFBZE^c|PgrDw*@ zmj*}UC64!?*NAb^Q;raozlU(>wiq7V(n`Z0=Rrkj7yW7^Lq&ZTL-m^XL|!=;T*oef zH>0vh-t5OrU~w0?8}!4cHv{0TT@PQH7NK%&0sLOD4*vXDg|nPr(8G&DV4ZRxE;H08 zwxb>3Lo9=ZMUEsQ=mgdZ^qgcNIU8s8e0jdTQV0|CY z(tGogK3WmVoR}@Z%(rK8qeKji?4B&#=zR>f44x-HR`m)_J^PQG-ak$l_k2GM(y`(1 z)vD-yT9F)mGa9}&)T&**}6lW6YWtMryyDBcM%!|zwcgduO<(WccB0YI6hAo_?38dW#zbCXB?E? zE{4NGBc2Iq6d-?regjh^7SoKy%(T}LLD|EAwA$m6p67vT7(9unI9 zr?xmZ4$?~(;y{lK9Q}TtFr!zH0Pin!)b&9$S<(YjR*plZ<|}w;z5;jNY#}U)e?u2b zh@*SpSV&nc2mbA2iNq)ym^fAef4bYiz7vV8r}+`IN%%w?s!!635fq)>4$?EC>LjRO zh^Ts>Awzvpl)KK72Qzc1(W)psuXTcrns9>tKA(xnVoyj%);tnZ|CEVRK8JlX7;b;7 z0Z|Q&h4|;z_(ju;oaQ9q?u0nfgFl$P^MB)JI)RJytwvqV@hI3agZq@M1x=OXNw2#s zE!q5x*gNF&xokVmtjL73DUC<{K6$o8)ea|@JD_n&6k2n-(6C(_y>wd1IVT%B{!uSk z!_NwdRY%cw{vH})Jsm5{{|Y3}8j+0K3sJA4kIH*ZhXflp;$vt@PEFYkBNftIqGvqw zn2UieUrb0yq6hR>0SN9EL9%NDnbR?k%j@jnTJGfHgO|-@FbR@YTM7anzFs z+i~0JqvnP5W@ZAMv<`-?+R?;v>K76}&xkCaJ6wCqp%vt17Q;8uM%?;*6j$If2^+qY z;MlZ(bY}DzdZ}k7chGtcK7aWFWcTbP6|%G8&5|O4?y}{u#AP;}`F=Aj5I$g}`tk*R zm*dcNh8y72Pedqf1cuWWK-+^jTpMx7ao#2K3lDIkxcTBfJN7Bg?gIr9e2glg@CIhR6#N_-v>ET`yRYwp>}Z z!S4s1u+S0?Y>9)z%x$V3;e*?SZZO&B3RMbpB-smF|C^i_o=JSp}C4JH|X=6|3) z*H3eLAJZY|!~{;?f->tSJ|msNQtYo$g9fu5Ffx)0eb>w(p}a+~n&;jxyw`+}j@=;| z<~&P=j6p5)WRky4i_|gBFns+A{&%%epu41oo;xZHhs4ae(jzALpN14kEjL63uUTlO zcNK=}3(=?RFS!`=hTUZMRFG`JXH*AUSmk?fpmtLel&iiYp?(u_m17j@MylfANHutf zYY1}=tOX}0XEH}z3ZHeGp-uet~{T9CQuKv%9?XQ$i#Zq)tcF9Mu9HtSaZq6Q zolIED5bgC+B?{G4i7) zi1s^^vC}rNcN_0An=E>XXvZ=to1TLvV!@oeu0EHuJ_t?v;^5Wm9Nv$2hXn17M;$$F z-sP4}Pfl7u-><&_^?Flbb#FU*G?_xywQXSU=udjmrC_`7GEDih3=SnK!oK+eNPl(| zWY7I3GU~mgUL^>_o$gaLyF4cC@*@!Pv+(HczL>vtHmvAsXFF;WS-YnjiRhhbsx#Xb z7ODQi>MS$vY@#8j7X6hTJhh3Om8zglw;Sl}JSS#y@k#Qt_zOC(w8f%Ke~ccZ4OY5y zsA~U8cIKv7nyES&l$M=fSG?ods__Q6asEmoH!=m%qLoOd#Y2*CL`asdv_i3PN|(}4 zP+HUsM(gCkDESaM|C47+*HqwGfetE}Y{p-v4m2$~5h8L!nb3P5NujhcUNJSNRe#6g z>|A|xTUg0^T$50X-}miZ988Toy-3N{7!07rXIJCE({Ysr-^jaaFHj?7( zYL|0A=e?)QVL5vB(`ufV{zFh9+f82W5u8#s$qeCEobbq+YaC=?7GjmB5P@9=~u;^UAw(ix`0?|YAc zsD}bi9ycm?gXd+u@n^Gt<On|J7jl1Ap`5i0_kLW7) zmJ#W&TlOry`$QKG9qodh;rw0Zmm{(LeUfU$@=kNJ1E~2Tni!7kra={c8ojw%Q#c4?9Eb2W2vPza7RsRHVK24QR0HAV^j}p#R1krgACgXxgGCdIF^3 zY+D>Vqaz$Toz{Ur&k|XrQv^A(GO*I?CR47ofsuS+2wJOdKyh0qtuPUVX{mnLJnshP z$fT1k!pAV<{#z6QVJGRe`Cv*GtC9sWCffy@}wOt`cHe5m>qkACJeoP|@kOFAap zP8S8PdYutBdZHHWc`Zv7lH|X%D)8%_qj6ikDj$2`5x#DojP`FX!t?4joaVtnL3TL2 z|5``nWvz&fMh1DMZ^+%+8;ncO&gGA;?xU%L@^sN%LoEOJfWG;a2oG~qA#MCjh>VXX zC(O;^?EV70I%zt;K6o1H{C+}e#C}#6503}g07;|^Vo_}JNM0gaf?rh2;!FEsAg`i8 zcdcx}v!ZmmdDM8cn7NJ7N?8sYE1Bv)k#8|wxgGEBsyb(KK~+iB<~=35SZd) zWD%Uk>CbW~*C;{Nl2*c&<&)6H?I471pGli0xkB;}KfXWb8P;Cf4EMvu*f$lwAxm7B zUp`M6So7JS*svF3FP4#;hWBvUt2#V7Zxm4*xPgAJQ-G}Eph0;)9^AhK`kfO+5u#ss z`M?8ihG!-CI%=bC$pAT#R0Ll00Op6@#k7S9RBl5b=57onqkjH?pGM92VYCguL^dDw zBj)iB?$r@RVFtTLwF&iYj^p%+(NJ8&;-2pcxVNnXuj{P@)6cTtR}%$&?Q&$+5KEgf z9>LZ_DcGVopO2p_?6%$&;YZg8&}T6l7X1<2r8DwiiAf%)gw5r9j{ZU0_qD|9z8(Ke zm_6B{6>5~X!BF{cjQDmFst(%lUmP-VOoFapgSLby7eihr^d@ev%;(&!^zc^uDQ0S( z1N=4=V$(;%z_n^My%^|&&1NxBc;hYl+k27dpBK?pU5*!D8peob#|V4CWC(tz0Nw7Y z?5bID*!2K0wa^?H!P{K-SQ{;eS+Y6vgTe9MAjc(?;LNsbBrKZJdL0R#-SC?Gad976 zv^9+MbezGw#zobW<{rXf2|qFY{TXijAuUv|h`{Xfdr&Z3jonit#SRx(`I@6YLgD;h zT#Dm=#3C2)k>3mGPtXHRmps^IssLg)4uf&<2^!=*1EvR_2K!epvFuVCRP0md!|#N^ z@L#|2&ZHw?lzC3bo0kyLZ!vze(R4VHQ398~D)84?X}+kl0*hsJz%wiqvMe=lXyI!i zbM=C3dXWstD^~I5vl)yNKZwnPUDUhYnpQq)20wi>gmJf+3r&JIWRVJA@o5S?Ta%9V zVo5M@-*eGtX>rzg(oartq!(V@GlT!Qb|=m=?EwdqO0-fFasf4tcv{hg?Vaupv8S@Z zDz6UbWk;gNhhJEwU5lwAeO`IXW-Jc%;cqFwLDyX2?Voa-jJzEJi@L$}m30;$jOn5BnK=1IU_+->rJ}&ws+PeRs zQtqz2-HcXJuPJzjHZRAwcWda%xrZTb{V+D5uN~bt&gXYF9e~cN62a?Yj|WZDarL(R zY9reuoROfx-x-rZRrN1p)5Jldv%MW!Usemg5?|DuKaPK0T_xlh?g%*&4r^_OEE zsk5X%D-vZupFin zT|F^c!u*6=ey31r7ZOgh29%!N=8ui*0JCrLc>RSlyfKyHl@~om&(R&+oGn%;=`s!W z+*kf*&V+h-@ST^i(efl zJ+1?!`#?Q5m)hgKGJ)Un)DrY$ONe66IC}iwUX06HiV;FaOr|!I^LO5lVmF$I#ukAW z)HsdT-Xdfn15Uz{$Y;bn=>}<)nG2fgX_$8*4re9*B&}a1P;6|2gjYm z9PI|0@0Ed<1hC(czb{Bz^I@v>DT=XH%7M~l>uEyITA-}nh zA9mk`)t})`?g{U{J9no+v)g^*;o;3o=1<`>1NHfulb>m#WGvG?Z#qBA_!J&kbDep5 zxtW%CMU$SBUP2ej8^kYEB5(bL`|Gbm=L((~)tEpmUb!57Ce<>Y@^NVHlSMr`^+D#@ z8sYy$BOcl`9IWEfIR7@J?B|^7QIn2|4AfsRsh7)f@ShVJT6JO5ED5IMQz&gqdKQ-TcED+9GR=FgtgHV`MlCypqSN<6ScPEG9SZgsWtWxuwD~dU;V}%hVnEN zrsL}t1{2N9p=#qzV*hp!wiTSFVsqaMf2Td1eN6`4Ga-o!y(MC!6Z7HtBL}QsrOJv& zpND#jEl?uH;Jy3|)Ois`zXe%v;U)Q;X@L$7Zp$DhM&e}Gc3~zX^@YSwFcC?Jg!|-f zUv49_p5}koBd5d9liI0n7_cG(ObiCduvkZscbkTW3u?){xQ!@RAc1*~4siE#7QC8U zicx>mAltD3@-O+4uOBbMlk@ek`l2d3>x~h~o3@FL4xEDH4{gIq_N`Q{I^`FsxwSzH9;{M^4-5UMPg)$AGkoQY zy=z9ZnpLMj(yI{4ixZd-;jH?l`i_WC+7G!~qapg-3v#R7MCc8Dug>~rj0@iV1+Cv= z?BCd4##qrBUabB|LN^{^q)dwGrakg}nUoT)n?4a%ZMlS-e=cRtwCk}d3O|XH=5YSq z9UEb$Hb`#=30y+83#hcn7dA^HC>I98&21M%e@~|2-}?UqH_;24-%%s@s;`i>y*B*l zxy$*=K}A}dm`*Y}j?MIRNMqC|1Y_!h5bz_7^-ybr*huhr$=qZDL7hDmzcdtGw#K zYA=OxUB_T>#cOKQhFHHPkI|FVW8db7;B}njLZh`erqnU3srde2|Iam>MC{FA>v0b&*!tN4->KV zo5(?jdYsf>K&=iZ)7ciWIMezbebkyu?V8%`f_+ z^%L=HF^BH0bE%F)50&}TLp#r2AQD}BU;%H$Z+j(%zV9zHwex(yEzkh>tB7!CWT4Ou zyH7TMd8`8qPp>vH9h`@g<9PqB6^??W;^ZKC%;3XXXFm->@KA7bSWCiiQ_jr zZT3gRDtxK*iIKiJ0$VG5nNcbi$o?y1*lhyK$LaF~j3|-k$8;><`}U9JudUDFdRESW zo$pU!^r;aLl5d5Klo2+hMp9{)*|70+4B;9=!JtHj-NNXx>SpiZchgBsdO~ zhna-0P0-RR162|VSM3Nv8UwsMjzEzi5vyWst}WPEm00jruM zVPs({NgAQYXRrB#kwVAafXgF${FYuFQ*1*2YH9Hiwa@U{-bVBd|4hz*I!j8+{K(|m zLul#l&ZV8cMvm>XrIG>i%*ybw@Ug`K5||Be`F=DE9MA{d=MBtGdl__Ub;F=%#axBO zO#JcrB)6mf9S)_g$A)#C)hXJF@LoQK46SU!arY2c6ml4L&IXl78RB817xY=eF+Nsh zBwul$3vIWH+e|c#B#+yLx4cCfw*Hc%W9Lpmqn*IMoe%7eM{2Cz1s>Jz^BgEJ$*(Ec}h<93*i8NLUt z4GPh7j|?B`Cd2PKc9W*9>Y)FoIAB_uInfKf%8j;v36L3#vV%)CI4@|9HeY&rBt7_!f^@_}<-&+Z$w ziVX`|&Q5%D9-1FYvI^b0?DpsHsLB@~T=(MvV=<(~yQ`+-&C=tT?)Hh?+-yrbX;HO# zvn8&JKSA~%P=zj=2Fys7{#t%?5%|n@GX4^?5`-Wi^dm`U4in9!H4zSw@r&^;C)v}H+Lir z0U4x5;~j0BL9rpx0FPM|(X|67@RHkI90;*D-M zui%@6HhSSrCbzaIm(!TJ6DN-!i@W}4z}C!H+@zE_I9r%YtgjX_gf&-b?3<&w+hZO) zQQtv7nI90BeB5V(u0T+Kx!Wo(Wahn}h!!h-l$Y##PAH!8~ zosg}YW|>2l{E{alH&&9opF5$zu7RBR>A+i~8HyG4ljB`0>5<_^wDCd#JvBX)#=6`R zi3OgZ54FcY*PJ5QZRrLl_fEw1u~PWW#|9oL{D-jve`)h39imC9 zQXdV}T@5E4g+7h(8v5y?8)y1y5j;QQEO>gtPQ7|QGD{r*~Bwr1Z{b^nTsh8B3YUBBFwvtm)oX;p^qeu&pHCTSFf+y zxwjc72btk&11rAlbtc_6v4pWo?PjcZcCzjQ)!zkX6wcfZEZvWXs%>H7nzPtAOM{>K!U81RbBUZqI#C!r1aG$4LfHD>bZ31Y`MTPY z+RqEdCBwRC?V2#ybS#A)f1Lzox%&9jF9v^%zfJe2uY;5F2{g`MM55+BA-%GOe7b5f z9Q?Kzrgm+HDYCPOq$mb^NiX(T7NV#;0mXl7!b;6$VAax3Lyu^I=hlAWzq*}XdvF(> zYK`c7yW?b^sw}TlCV_W82%OKrJJs8+lmYp57(IOUurL>K)ry8$U&(c-nU($#ceH6VNIwT_?Jd%i%&Nqy8E^-lT%( zXFEYdK`>E|53RCraOdR*7NK40aqJGcg;&nkz@BLtn6vl|*7)7O8nm+)XFb;9O9v9s z$@U^1`Sw`J72s89gTvdM@tV+;eKu(_I3?v! z-o6jW%Zs#f4AG)59AXu5$k=2u=3ayGN^+vDd%N~#US-@2oI)E_QfTm(}8 zM)Ecrl`wMEpz!P0q)UUw8RTuKf0Z(WS9Csc#&wqJ1Y>|KnmRpuj3-Nn%7 zHT1}?)74||B{Khor{EQrR#K5M0+!8ujPke2FmQ$>Eld`6Hj*>pM%YVXrQ zp9v492<)DN*Wp7U#g_3B{2Alz=#lM5&lP^;3?7aLy>$oZ&`03@G>#xue;m=$Hy@)7 z{J8EYMUwXV0o@+50)7OKBM~Z>N!=AQ_66&Tms@P9&JI6N@S4LEcE4bD3;{^?7NO_Q zv!vxo$5LY(u|L+E7Qq~K`_ZkHB;O3O;RU0MzL%%b3?j2&F4Jpmyy(J=P=UHmJ|IaZ#M z!PVIhh(&iID9x{e=Lz?S#JE^^SsDk{(`ERFMqh9nZN#Acn{e3zZ~WkrOcsn$#eE}h z6O(0%a3}FN1`8~ghq=8_{OJI5EwYHNKXjB-igu#e%V6C4yI=4OdQkK4v#_f5I(V;j z1~y9^>=K`W?A2s+mi48xQpb}M>!TPaZ=NoXwIvs&PGaTh2Jk(W2roJ(bIPSLj9af6 z`yV?~sJSM-wVK10@Sr)ux$5bBLnX|3!mTg4lX>&q7uw+;^;RP zX7sIx@KR+qftkfdo$(hn10;F$4N6-@&8Dp>I69DTnb8&dQl zX`;IpDYy3prZJCfbLwOEeNSY*?0XG;ygEBBSdqN*{|`I1MxmrU71$@*Bu*tw=nmco zcWY_-F}M=GvvM{w4yRD3p&{mw<18Y2?=3nXt4W-zmXJyIx@^cLXZHRA7qTRxR1`aP zx8OWHL!|r#A9acacc!(Bs<3Zpl>0ED`|U^8zuHJn9V-LRd)hE@;S*fn7J}KFz=t&& zq{lsr=@g@@w8)Xe50_Tr51SZn(LQmsB&&&b;UhW~a_I`4-<-+^LkQto=@lV|XBJ`% z%j(+c65UI%UUxgZ(;UaTJ&a^;wp@lpcR5g2dxaO2KjF3&U>wy z$KPIUjddvlC|YXGn{}M#=2{tG)QtuvrujaVdv+499L++h1;d1QRxLe|9ZH9P%4ST$ zl(Bf&E$Uro$()F}O*_Kc2;01p@cS;|y2i11Wr&3}POGTQ&r$HzSD)MYRGdoL$>37G z4B9a_5?wOVh}`PqB>ts{E!`W;zB{K0{{n=&o62#xmEZ);kuPY`q|?lS^V(1!)s9(1 zSs41Eg?zjQI0~ZpJuVM(3es?nYcae-`cSw}kPg z2QX&#NhaJilZ>kCpsUP=!L6o;#nsQBnJsJu3r+I_gD^*r~>^U$D;%K7gGmeM1Bv8+tuk~)j z?ZQr76&Su-!;Bx}wT@gDBE(VuO{0fr>EZm=BYbJkR(|1IKRg|t3_(U)*;NmYL0j-A zy5bj0+q}@nhr=z3r_OLweOGr3qiSB^m!^hZQUChltpGCed7VbgseW|ak zkT*_l!oObA_{i2cj!k5k4{<6e#Am3`FT%-RR#1bh)?DaGWBx_27JuT16AnxL0q=g) z3hoe;SvS$LzUl+S|&p;_>AU?KIMdYS$=dn;M7C6yHWND-fu269T_B^fnx zHhwh;W8Q@H(G1?6b-po&Eo%rMeQCq-_jwsO`JV!G+#Ak+z7`?)2FJs>IC<=R&e2nY z!Ss@;DUzVw#M5~UW{zdC;S+$`qa85$Y@Uru`X*@hwS(xwGid2Ep1*IGSsjrfhewuo zFb1cdlW!V_u{uTs0}`uYdvhX|t*BtmsoK$O_X>=B)`|~@jY6rJ^HIw*5XX7CpvjFd z!nvperp^guX<04t7T!?*#wg;N6CCg@KDjtHD9GljJ2!;m++-`>qO!{NDDTjAdI58UN85N1U>UH)~(OWEVgBT6_>f8gsyH z&t*Jtr~<_27-E6RY9^tylU&{r4I{ixqkqu{x}K*Td$R_f*L)$7Qw=e@=nz&VD$&l8 z``q^xJ8)WFEiRN^!a1BR#G*sn$+KhGAcRt&c#k{WzTJ)A4ORK!FV%4ay@}6y|KsJm zj?zGN3B2Srmftt{18VyV$)$2BoEbKfIE#5R+r_luVAo+(8si3f^1XC${dzdH{0QWE z6yW(vHTbh40%JB8;`hmFd}pvY27K_x>V?*%>(rgAQ6Eb{AWA=$A`0^-@OOOhp$+n*0@Lr8wqglgv?C5jpEwQ!{q9hsg>;Vj zUFJ@&HX6?%eA5gozUX=-R<2G(ue3C-@=ONuoY2ks#N(yXMfh;GaGtMR!`zN>V8*s| z(_032@M`#5NWb@%PCPUWZO`k`KWF{P-d(*(u=jN*g4+kASq=Rft?GkNo~R z=yna`j%Q8539jwL$;TQUx2^&oHZ-mU zu0DfV`P0CaaIX95XIL;T2L5Hv2etps;D&(&eCVHmdS8pEi%A+f2#&tr ze@C-@&x_&d1amyKxsL9u%?Gdiosj7%!cF6&g}g!%;<8c{npBwHqKtMortl~B&F5ni zM0|{{I3y;!!gHfoVmxmc?0q^!yRO*c!i_7T;N55LSmqDdwBZ={Dr;f_@~zuplLHwh|r{Ss$2P z{<3V^{9WwrvHyWt@C7pe-FVhOPL8ush=Ft7JB9gl11S}gVyh&CS%LR@`c2;$cGYBa z%`;RN4-l1@UY-6MGYU~81i>m~7fV#uU`VU)9vK}r1<(ZrRGnDil- zDj5j9D4kc#YnfkU?AeJZYh()w9lQG|2m2b`20&7Rb)3DhLT^_U#y?sXU z%CHw@J&VcF{bxzLCS}UU3(VW-C{lXz1nOMKCD9)XV0Mi^eSPB)yqgyQKV05XJs+nT zzYcGM(_?qw2&-1QdcQpCW*kSKpPi$_6bgfkcT{2a}fGsEY{%E9))Y<8-i6+F7vh~G@) zgl^kQT-I(~eXli@7>Yv4+k{-v-ORNl!af}JPd^tt6ZdfO!DyN?Y9q5^c@SOKvX0&y z*pJ^!Cc&o2{h|$14%7HA^6>TACVc$-I?0;6i~JVO#MPJ2S9ewI#@Lm2acGJqqmnj^ zW=vW`IikjI|1p{mm@$Rd*6;+`pHpb|?F!N3rYext8Uv$Jl$h!g753bBHC9~V2KCan z#t+2-pzW`W)5pH3p1jr)j<5YmuD7`JmrWmFXv-jtaTNAWCq(>&{$CiqS*~)3y(Q$gNp-_B)&h8yjdpXL(jYy@)X}W?yDk9 zdhv+JOus{?#|+Rb_Y3JA`A6Wrs2bYsQsF?+JREZ+m2Mv{PuEZEK+7sE-c(t=OkRfN z+WplVyi{Qi2?8lw4NlcdmibsOu#+a;;VK1wZr0~x*s^{Hm@EDvN+WB?)Qhe}*INzt zw)b<7V{=IBzEHTjCINm{A4GTGVAQDpM$g-PhfU49K`u>>B>RqI$@zCMixqlI1F_7{ zpvlbJ8UxsL_aH74{uc9#U!cmIn{-J~1pW3dhKY2z3q46^q50cRFeBed_vw{zsYX<- z{{9WoU3-z-n`46B4Wnt2*IrI+tRD)rs-^w1li-5jGwu1jnvT)ff!pj46H|UM7}{tsb0D1@ z%R9{BE#WS4;yrOO8IL+kmZKF-0rl20@SSV~y2hi3Tw6X3ANxVbd;DXxJ-$%uPuAFb z;xbliPvny%INZGa2$#R?q;TGQ;!YPeviJQGI4Gw?PS?pnce4Z8KQ)zjXbpgk;2`~F z4KVS&JdHI-##ugL*k>~tYNn=E$LT-d_78Mngv3ovq|WoI5PK_PY3_D7BWoAP}DkMJ$3&qMKVrl(%zfPsCkeTo$yjs_*P8BqWk^y zRBjzE40j@n(}S6enR3|fFYJ`cwpUMmwhcF~J3}?LTjL{3EzG>bz|l!ccwymvn)|Yf zmY?v)2?kB*(!WDsyy?Rf=Pe}oXm53T@pw=lDaqIrJRx@T17W+&Xb|hlBNvXz(!wp~ z^uPTDq`rFpjEV_m+`}W%FQ3e6nsA`_zeZ0L%6_axcHXp?kdE8I~-vQFPs-8 z{X|alGfBI=oBB5?;Js1iXszRl72csjuIQ;~YWFDA9k&>d;dco2; z2ZnnNC$+Z4@NCr?a$DmC_jUPg&TexkliIXF^wD}G{t-G3zUR~M_=99zn^=zB_OEf) zs1O^S3o7W{{+!Cy@5Y5jGK_XCCt_WXk>9IINP4;$msEQeOBbHSmP@U8er*>0ew(A# z!?xhYBn7xvtcuR#cF{NQ+PPbwcTrRMEp*+?$IOkX6Tv3Jo9laN2%>&NGDqzv*_tc@ z89y&5Yl$X{J}kgBO`C9cs0FRaUJe;c9)L;1FfP$0oA{jZqWx7<(XB;{o%wAYjQD&D zVlOR#;c@~W*eQX8ZWqJM!m+sE>UKOkMU&vZ!(?rhPqmuDEk@2hg8CnQMl z=$zbKR9kX_>OPuI+ZNh^uXZjRYG{PI=r-_fOvMl*e=Jx(kzTr{1;Ivl=zms8*y-cL z9DA{s%9ovoI*&Eru%M4F%h-XRW~HItw-G2~U5sDnr7@e*F2YRZ959G(uI9?m5XbiY za8`03z6zd?*H^dEABsL?#Gy2p8eIYNbFLDxxoNO)`aEEBZRnRN>B6ok7~0QE0b3DB zg0Bs6i(a&mf_5qXTd^*V{;(Z-LY|a(G>4{3xm2#r1!m(p-1$<6w}L--?(q`r zut=hx5(CK+cMZ7LJq^|rZiBJEhN$AJLe!g>%#MG14z zJkJ0s=SJY7o8tHw;al z?(%RjjakCkoGE5vbpE2@FAJ)u6iY4j?h7-_5q!^6d43&V3>BY_auG9zLqeSD(t>p z!&NHS;dL1e_~$hp2Yox?S9dJ795TbS+&Ijfcb}FPh2nJd0_z-4usZq!|M_K!YBfsW zZDTE#dyNLIXCIh&f&1^L;0S*sn{if48Z1k{3;A_x$Qj8JvUIu(#cyHsfRQ5o@m!XU zO}{|ItV`(G)4$04Z6!=vV-|im(vH{l+ewc5DY{~)8U`*YL9@OH-ee05pT^{`fbgRYiE2W2g?#v~B$A!X4 zQvz2=KDTFUF6osKLG2YG@w#FKRAsM()X@S@GJgRs32Y}Gk7QwO|3uLJ{0n?#rPzH_ zP5GJ+ACM_4Lg(JuShT90?w6X3^06J&kpbq^j|w@>@wN~%-w;Q(*wE;RK04)?5q|jB zN0@CA>@r4#_nohz>+YR&$uEKDH)P2y=#qrM93e-)^dxSrQs8^r;~`XiF1a+Dg^Mo- zN$99RQS^8j$ z2s3}*1r6D$Y=z8J>YtO2&PkI=;o;+OHAoJ|h$^v3Mc}!r#-LlwOI(tigyu{bYP>lB zhU6X?dlbTap%1I$*ns~UjBrI}Dkv>mjR^~<(|MsvIQc~*{p9A)bUiE)-RkHf7Y&qo z?St#+Q?LCv7v`cvzANC8ezZkLd=?xJvlmyQ!L%9BBq?}`w%lMYe#pjlj3u*HP2s<2 zKEvinB|2~9WjuH01bU=-)6>EkVA=E^iK_lYN+ku3$PIb+T4^cxMNB6*!&S&V<8$z( zOMwmfc^*5vKA~LtKD@C-j+Z?X5541W;sdi{#<`&!oDJt)wHh-=V4q$gJuZ8R+JTLv za_Kg@c=Tvi`)dWNO_RXMW6HtDyPJ-@_yL-7CHSKge`DX1gG5tPl{!`-eP*EuClhyy z{MT$F9iep;be2J-yB?Z%6ioN_`~cYl(G7_j%}y^wp>NG*Qe39Wf+#_4bts#-L$Vh5e7r#N!rcNWRLl1 z;Vf6fKdrxQO7_`7{F)>9d8IT+?JKJ~zXQ zZ{Lp&hhwQz5Ch*G=F>QxSG4hyGMVb6jI#U0*oS8#aML*r{zS$jG&!q6ml=*Exm&_P z{rVGH;s1apJsb`%+uT8Gw+p`el?Aqcgv zIPgvI&Yh3Ot;MIwA0183rksTXg4?jVg+f(XBA(uYs3`Oj|D;|6wFwJEsh2YFvr`;8 z9{K^HZeuWam*6D15KF|Lwn6iW66~80z%|c*2}1>2aIgW9>+Xi&>A~Fhsj9rOjXoRO z$K&40Gi1{^Q@-G856!wX5tsYi1+g?$w6AF=)AKX9f_uG0GFy>8_1lSJ{(|!&sTfQu zYOpq^2fJ3K!N#@c=&RR)Pc~1;LL`ZCrWzJ#+u%gDtP)(&-Wn*cbqfPkZDE(r2-0uY ziMbj+aDLlfOjXEA9W9}ZSyCMA7rJ-lR>}A`;4pn&_nmoQw+4T%`NSQr zsKg{sj(R?v0*|v)c|E}oac{pVtjnCr?~>@FSxGW<-Fr`dP$`1%^l;-XB+irW(N;vD zo0Bm%Joz$KoAC+*L|nxox|^m^|#a9dX;jL z8#WSJdsE0N8$W*Djc8th4dUf?pWqgcbtjr|i_1JCg*p?rlRDE>y6Hs+pL^^U5_-Cz_{NQk9ez(bd$aDU~rMd`g#m*}t%X8znFKR+3 zZ>8|@rmEon#+h`h+Tauy4u^KD;Cw@IIGe~qrIJ0ovW_FBt-F~HzmxRyyCwKJbPdtpoje+Ly!z5jUuN9DQ6STL67X6pl#x}SRJt3if3}ki?PHin z%nCV{K1vg7B?nZ(P?x8)7)wvZ!_EU1i>CbpJ3FgELE!l{piWbXAAvgqALk<(}` zE;eNjZJkj~gL6jVgl{w8iOO2e_JlFxeBm}p^ta=@!kSKx9Lo47a)^5efK>Qkyj$~I?!#it?w0dp#*QQOfm$uid^VZU3OY&4pNb3jw>B!5JqE{GOTh!<3fPda zl305rKy2wJm^9`(oa@qu1KZnR`mlGf^|v#)6sIt$Cd1(Qg?_TyJ+A8fO>L6BS&om{ z*oh+~#!-WR?!4|4Pk!8Vfs3a<1)dg{P&YGaO#hL=>7@+a_7is#9>gpM*HLHt8MDg|2(Y#Z^W`4AHBhlQIO}vse5V|>o@ZD`R zkH1158B^%FvkLw*y##&HN9oYd2GN|p3uNu0>&)>54q&Bo5&Sk5aa!M6=w5Gam=XAn zc*+FB%WKV;=@^8A7xyvC5G1r77)w(H=$+jeYpMU6s%BaA@3!!h=AeO;S^q&=M+iQVwStMS^0~3cHh|85JDK^5li|SoXH>>!koaiU z6S?{}AxmEbZM}!!T4xhpHVvg^35&`0cU$qnavk1B`~zwz%3%6)MLtcY7@;cx#yy_~ zXLA3N%>xoRdhcI)<)9u6X-}lGL((9MktS`kyHQ)1MM=5`;=U&)7`lTZZ9Itwot?0? zRgn!Eya{W3E4l4oXRrm2Hd7sB5>mt@2^8wvjKn34Qf zb`)z{^o3Dkf52(s{i^s`oKM{H0DYcs=ia|NiI=Xa;#NIXHmQjRzQiB8v$o^VmqM&; zc0pw!TV0~Cl^FM|;T}hJ(1e#?sky@m{CsF8_V-MK|pu$YLCya3;F7+7l`McAzC)tm^VKViHGZ4iIdQ+8m*uTL1tft*>*gz zwTE%fmN?8vw8RjA50}C5vK+p4BvkjwpWzF#i^|zz19h4g#Dp=Qhl+# zVHTgOdIqHGX0x5$32?%f;({_`49%1g^4~omKJhfCx?>V(?UiEB1!=Ofx;^mT;wxS8 z;2w6a=W%*?1O|?dr@iNMup@d9u303I5zC|SwE^%OU(X^Hv#0ZiOegco(sFpkTDUV< z*Fta2U5I^k4cA9W!FaZlOgbiyJEAn$ALrkaaqBYSK(oLrztAAOEt~M_TMa6ISeZZB zsEO6j*5Nb7JS?ew#{~BtBcl`+64AIK`unFPa-Wr<)i8-DXNn079~-`;S(ZPrCXFN* z9)}jEAv~hA6spHxBl5yIFv7*2wVI>Liboe!7Ik#@?DG`x!XgZK3)d*`?c7up!HRm+Z=u&1~ZV!i0^FyR5_#qDIQ`9oZhUW3F!C`ei zl(g59{;-e4W9cz8*>BD1UKjS%5*9Et?j@18tf0F)&T;kLQ{m~xeA+iQMA)+q=g&pX z=AXWC<8!RtaJif`UzRx%WK-4oB75QI>6L@a>>`l9@xcu_u9y`h%kyXQ@Icy15$$Wn zQ45|z+PS^Znv{z&-tq7{DIS_Tit&?v2|Rj}N8;AH!*ovAk?e01-tvlgNA5c6-Y^EM zEvablpM;y;3EWS$S(_cAh8}Tw4T2iOG4c@ZU)G?>ag3D%(O*zfWYleI)te zR$}P>HIJX?H=DEB;7Stj6yS8huM~H93~%=}mv1vZg&E4V^hQQ9sknQGW;cw)5A`=N zIn)~OzSpIZzW?LsJREX*-#@OMR9ZAB?T{vsp69+!Aw@(9kx}Af3)va%sil;nR74V$ zhS9mNgQigmO{Hf->lK-YG9Ou$6rq>ou{sE=B{S<&JVQ<<1G#O^^ebn&~LDA_E$?>>vsilVHNsYxpC68}Z*J#!ML(h9-lK zMAN$pl_xUsI zj-dNcxq3U9>0XGEa)GSU6FoeemJVTJ-=KV-G^cFi2ZLu5z~Z$zIjg!9CHsr$du0)v z5WzF@!|dVq-NW3^^QYN=kr{+J#QU9M9MQR|h1^zmz>an97-P4QBU&cl^0@{& zd4AZ$Dgl;tY0@ffLpYJ%yl!Km+nm$fXPw9RYg`lk^HCPEBrmWo zQ@+ws19>>xrjhNk%Gco4E6VhLQ867UA;07B}%s9{htw*DHH;g8}{cFl` z$%pu!$#vLr?I!NM+e6$NmxAvX6(;IT4|%!6hPwI4^4XWq?8F1Vh;z0Mp7aP6I)5Gu zy%-NeF_zda?uautm@&CdnS_07gDr+Has6Fu{HFGio&3}o_j+ej@dgq6F`A!;hZxXb zVq4%)z)ox^o`i?OPU3$&FG1Q<1ZQ&ph+9t!txM5>?(I*7C1F?4SY!eCY}Nz4QERz# z&!WLB-w}-+S|Gsh6+5=H2o+Bs@538?%a^%a=W+k?Dcpd zsdt0uxHc;I>WOxnOaw;?T2Q|`6qgqNhF2|pI9g*Nri6Ht$u4C$sj{m~``j5$Hz^O! zMnwyD?fOV#l>_PKlA}bbeK#EP3Z_eMFCvatjR|ab$Ze*qM4@ z$GNrO+VYCbm=%P!lcr$mO>-E`n!@!62Zg=!j-W@*GI$Zc6Kfr&fzp#Y(D|and9>xQ zZ^Vo7QUKpCx9EejdhJjhQ;Wmfr*U=G>D+dUDzL9CgPDVR%;4zDWd+Gy=(OcJm&J_1 zu-0+R_SFSAppXiSs|)!2svIM~F$9}aFOaD&s+?wm2ebB!6f^yBq~LqmS5nN+F#j9- zjeqz2CZi1Ph&kWmm%HFWlrHVSdw&-KTfLNZc{UbvuIs}A7v3Wg@c{*vqTJf6MzFHG ziLYsWueuWWCkVGG8Co0$w=JInVU-|U2|FLK#Fhh_A%Yb#Fa zi$#ZqQ1sflilbBZAW>P!xbJC%$IWH*)9WOZURVcL#@pfwR)*WEnt(7lKEo&-gLSnH=QUX^&uCyGxg~((y@d|iCImw z3}uk9*TF~H`DA)%1^q3XkNwwfv5VrIz{1c`_*_+#yCywyXG=>utP%A9!)tkE=SE+{ z$&;2ql<)$&AN4>NWoH-`UZw8&rbPemF0^aRM}s9%sC{A|aX)|JgUC;1j;^}eO| z!>`c3^^1v2r5<}IScZJHj=-yO63o{ZPuZdQe3fIqEY~mfAI$e@gv*9DFr{rASbNOC zg@4v#jI}Z*+4;k^z`dN6*$&#YUy_-!<2j~TT}F`$k%Y@nr&k-MqsEF|c)PX;uJalE zfs86#r6>-Yi(7H&cuAOPJp$kVSW<9e(C}3N^p)Sn^zUoiS%P$k;t{(iCYYmJ zKs9_n*@~J0PiWJBQktP@MJNSU=Rp-E)xhJvD z-~t=BU<{nOJ_m1^y&{Jm?1kEt9`Y+IQ?P637?^yPW@~qvVSMB{_V0xps5yL%=N(;v zMx8YLD<6cZ{Pk(J-+?{jOW9$)GJL=!gY1rIm}xW~GTcss7#_vYWK+!V_M*yiuSo)& zz*CoWxdWB9uspj1eg&4&jR$TKsg!ikdRa_^lyq^T!#7ASnJAQ66O2`Vg30*Rt1zx6 zhJvGj4d*?U{4atcafmvDyJ70T%Mdpp4VFrw!lM7SqV|1%vSQ^N`suhE%Z}pN$zL7F zI~raX1;~_4a4l%s1BNxc?6F)##0-TMCY}r5r?#R$kEu1V@?Ot7qU*^ zS@4p6ALB+e41R&Z>JF(IX>fV?o9bCg za~=HL>Rh5ISAKgsClh!ToGm^;|NiaxH2Ev#RFt^X)83q+#80@WS&4i7k~mk@4!Ze2 zGu#rriK59>GR0^$YUnCsp`SVStEA!LD0l8@+9+m;wJNxqY~n(Bo{28sn<#nEg}6iv zo~-bMiTs)Q@}Ma@^kA;eVVHb(H?+Ly zreyqEX!)!TxAVTD+L}--{G!eYl~v97bsBRT?8*}EspF=+ zR2$RLZk&&e78f~Qll$3_ffX4ZO#7(kZ2igx++`I`lCEsW)s-sjuW6=C+OI`;puz)R zD#YLk^J4OeXIFhl&_ShQX-wA7ryipn;1SQ}3YPENZ)T%#lF(lFF?n2a3Ow{xP{PoFd)gdJ6SEaa?)r1snmLJ?TPcUa!52g|Ocz$I z7LpYH|J}s%U`rM~!~;&P)NYzNcKdCDe;cH@`-_L@h>9x=dU`UA> z)d6~k<1szmjhQ`iioGNC4_l>BRF3GhW z(qvTTPNNRz|KofXj%L;!)uDd-&$E9M#2K%3C3q|+9A#dPVO*>;&}L^NR`l-#ZToa# zDkr4c^AwpPmS;KYIODSesvvKn3@b8Eak@q#Fly#r{E&Z|3_RLH#FcuGncaw!xBNum zx@q)XKsNm?KA)Kum4G2%2*MjaVO$iLe;s-n(m}@YBdMsniOYSjvKQ@b;oG8c_^;Cp4&+#H;}3Mv zPotM3UYy51RoIK(lXbc7Mh8^fA`f;e5Vf{fvX(lzpp`NNmpYT_D-%VGy}B8u|2Tsu zhN3X_rxthY_e=IlT?C{ieq=d$C&0)#{9L9Ck>?D=jRfFkFaFu{C|9sz=^5BF?-cKi z)5hgb`mw}&ELZU~55E5vgCplGI9vGwvc5M5;*NYH5<7WD!sWTl+p7!sd=>*q?)>iU z^%z{5>Q655eBbBZXVCBVF$jyAN{=?`akHZLkm?qIJfAR(OjX9ZBL?tSw3{_~%R4Im zDC6$mSvYL8jg|eQ&d>hdW2>$mo?bnT9{&^#w~B9(kQi4~__~{WqPu|jpU`7EYecw6 z%{Vw#w*aOe6^Fuvv+yGSMOpCJXq1$SAZ+7*DDmwS%zHH+4UE6wf8$fphW{>^_gR^1 zo)v*9%6{aveLPHy`hguAmGGi}G-kLr)4>}z320xU$ClU88+)5*wf_^4)llY6zr7E3 z;nqyYgc>7jN=Q^e&z*G$x*2?=NzPPpoEANWJ-Z!IR zvSt#V8;%0@-beURGn=c_dx}r`Mx!0;0n1~}sdeQ|dcaehTl-)s?bYpqguPyfxAH;p zsx@tX^9`z(`A~!H(^2{FN~m=T5K2huWBKl6ykM$WrlnuQntmF@4(SBA(zzVWR3FmV zm35s089&gCKFc+pYT|!-LSx$PmgRa;i!s*Xj z55sf6!;uO6oMo~)`2HA6?%5a8h>hRLlW;et!)ydKeGAB!!+KyUGM1|b6}0lDsJ_<~ z^V*akN!?O7t?d(9S(@Xn7u%q=ap+EhqA2UQaTCn!2tXS{GZ^woB>S|EVcETH!bc5u z#5npVj@oF7x!$YrAx**|oB^>HOQ6lG1-YXLV|UGG{=99*<@e6QFd4;|y)-7%zlFff z0{+{5^BNeATMv(}^4z+Ood^~As`@kZsrhBFI`xk{xzR^vyw|`v7G2chQY!6JDX-lCA;6aI>$t~-Sq#Z?krrDFK-Zz?mn|2Dk;)CKK!XGmMyak{PV7!DY)xL=S??XG>s ztc^w(WVnhOmm|ZpvX;!tKhI&2QU=I6C4!@kAH5o%45!}2L9iRcnqE<$SH1?}Ig35` z)|k(^r~RPL=iwmAa{M%jO;6x>}BK z-oNH@i$7h*m4Qy!^`67x^E!;lczH(Ox0~%sp=9yTTlCv)C(KoEfUVjqK^Q!XZBNO9 z6M6>B_$*bJ>+z6q{mPQfhZ&W`w+}McM(c+*3tT$C>#!RKr%*~-Z*SSXp=k| z`bc4wY!?3dyp2jlEWyQFjG;F?8k&4|gLP^mm^~@Q26ZjSJX;Dkc}CEw`;9o%Eva)*!<~Ed1-TceVrM!39vFU$dM9_%&RK_XmP;ZMMH?dEv))fLEg;Ke zIz|V8%udZBkk?V+9CjVT>V6e&MgAT*D6nU4^}mKar47QP=8pv%4%lMr+FP`^1cSX+9$*761ezv1<3xg`ENu+k2 zBzij)(KV_n=-SD59OW{|r_JB!l8NV0^~_0F-!um#Z%hFT$sD+sc>xYr-h+h~d4Bod zKVTf#zG@Lj7QCu*@4l(@*s%@v8p( zGvl?CnLkeN@cHF&=uhedU1}Ae9oGe=j(UveT8`?^-9#fkj-iq^(cm>T3KUhBK&wv7N6$f++qGdZ~KX7d^Cf-+hw8DIVp6J{D-|&Vr0JVc!+g7O^pK@ag3V;S2Lyr zKh9eTJGA^!@O2fuR|v$uJ>$8m#qlU=D+SG>`uI>a3%!>5;Fc@lIGcClt`x{(;yxv^ z(qt3KhCrftaj!7)Wh(bQMwjm_d=gw0Oh#uf4lemAgJ}*&9t_2!RO$+JvGr#!e)pi4 zA}ZO-pVRR4TvJp|{0LqrBZ!}hA=hwFne!}8=3c(Wdfgr zobL)&|7IZh?<{_=6$p+l8OKHKm4n354E~)sj*OCs#p!&uVSfKnJXGpN{-pA~f+1<# zsB<01sK~*Hxh5LU;@M^D{I#zBzcrfOqd`mtz3W^sU-a1J;}Fl8t9g;LP6Z^wR4}Fx#y}r85>|W7Qx&E>T2( zyD>0*TQlPG(^m*l_M4qKkv3OslD;_hkc^w5&GBwgzb@5-42 zvMUi(1|=9fwXwLU?{JxMxEO6(ITxJm6>#wyzJH#k4Q?}+Qy+E;m^AhYS7s%hHN zN%!+epSL+Jy8jUNG)pqmX8a~?yYwa7 zXki5G5mRCw|Er-T@+R=AX+Q4u&@PkhyGQ5!jet9n!BlIw4a*D<;nDmV=*7Euw^t?z z9~6YLM>OMvrLN7GH(pNY=Ffoo=51UCOu%zKHCXDEOXC`a^s#Fyyiw zAkW8@Z&H9OV@lW`GW>n&lD+e7{N5trmZ{= zdVec zk!1%@(&B2~2e`mH;J@=kSaH~zl4?*=WRa^x2OOj=`mTxdQt1)?Igu|{JimG^9>3lFQ) z5P>>w>Au1pmx=^vxXfO#nvX-1#?um)3=qYpvfQv4__UAj7}b~Y%-X}$uW|!*|IY#9 zJ>S6vjk)-@iU+^3vE-OV9o?$i&)#3H3shMKrYsIE`>cM2RsA>%V+uN{@_JLAiQ9%- zEx%xjy(hNF+0sARlkw5$B}5#;ar8t*bhhF>)Q1y=>76WoEE-Ke_8*~LE>@Ve`zIYS zbHYCU{2rMXkIyqsVUN~2T=;eYbMnpy-amN}PK+FcUn`0*U}6-St0&ObF=6;*paOqC zo`;h^?PZe>|DlCXmoPGab(poTybmTOle{TE1-DE)$hlRSkU8^z{D2V9TN4dGkEi1{ zHVbVFIA|jr^jmbo``S+0H=%;YPfMHb3f6JL#M-yalyZEy$qq}Nn_o-VwyQo*wZdc;jx z1z$)1Vi%<7V|wu~NM0H)T$!(qyVhA_$T!LvGM!g(1@npa#z1IHd@ZDP{Eb+b7%6r9Gfj@FFJDsg6_VmQd&2_h#= zuF)G}*)+6w1iok)Ge7t7o#*#yRLkx*eu$OhiZfZvRkEvaEzVe$3wg)2xeMLZc&YILd~bDzEKPH4VXY3ZsZModrJke9&NW4Eil= zqUTR|qn7spI^mxLO4qN)lhXt7&Rr)m`fe*Kw9iDhEv@JnY>T@JteACC?a;Y60~$)N z(|J<|@w;y?4K^5o%DbuXFhU>h4VB;rPJpSgTe0Di0c;MQhhlk`@ZRuf#_!!u&}=vZ zGB0J=+v{Gj-(`ikVy`ADtyzT^?bc)IKs`Mw>5Gqj`8`E#3a(fXfZMgFp+W2loc6bc z@AD~fCZe{ue^WM}yJ^Gw%DbV$I)nV36D?$#zCwV%8uM$-dO)AfvihleG4D|W6AVN$b(AhL1e9#Qq$^t)ZF46yl;yl zt)qE9p2bXBIi?(UKjb@Eax%>Dr!|b}{p0X(@EWAJ$iN!?$3!Jj4X#wx3r1&VV(`AN zpj05k^Wmmr#9SR_#7q^Mznvx9hd&9%D){lUg1cnFq{CEgt^sI#+67vN$3Vo>aInix z$A@B<*c5jePznm6-$tdNvauFcuH#U}?FxPs3c2a?pJV)?OK`w87k-}F1sfmNk|Q&#o{yrJeMcRZT%y|eVcX>dmi)bH+M_+cVs3Wbh;yq zo)nBW`k#q~tt-l@sl(W=T$&iXUpP;DE{^l&h)z8r8hfgRBEI~&?l3SzYU<3mlQH=B z_FPWv#1I~|{6jDGU%9jGZXss-XF=FWWw>)Vo31S~gFgxrn5U=pnL@EO+=gcpxbx!T z+@r>7ocFX7d&8u-L069VlIzny1y@njG!iDX0du`}I#cc(fHT5Mgc8vr^zHaN=%Vui zRXxXXDZ?VPF~b}_G2ht9udl*tW5U#bb7W?%SU~UdP)(19PTYU*8csKTLz!_yIMq0r zda9(|Y5a4K>iJ`K_7rjD3#j73#&Ttwbn>8s8-0Zh#*>M~K7mgE-p$Abz5HB;!vQt?ST&QhsL^ zWTnVFxFx|nJgviMjuha;E%6vf_9Hhko9X@K#OS+fFx7=>Aa}!+dfA?a@ct-N)fB-> zw{lW*Hjf%!Pymmb^FnX;2k@rRg1Nd{k$I872oerA!q2+BP++!`W*(eRTYrq<^4e!} z4YO4^7r9Z~zO_MQRAC%HQ|%*DpcXHU=3NJVg{Uwp0+Nnr!Hg@%Y5NQxD)XzH47}vs zXdy-1i$9@Uob(ay@8ov2C)1oRI}--lV{e^t)D}r27x=JP=TTAxy&u1lv z4C*;2M%WWv59if#K{nfsbRRXr=$w0Kugvo=OlDKvBbKzwHI0buc_dh|IF@wl1hBPV z!>QUw1(H71hWC$*;^yY)bJ9a+=(kmFT!MTEH?3wmC+BDc;-1dnu-zCrrz*6u>Og7r z7j|4m!V&BMs+sxE!rmZr!Cw?t<4t`YNxLG!z~n_JO?f_vR?hkXw!0|NBS% ziwlIG-*=Ix9?>-Y$pSE#lf^1bwI&;4kFg5;87lTmBv&Ipk6U`th*Kz9iaVP6>3x|c zw5M|l@s<~do`hmyG5thtP3?!zQEjAQ(+*VsEro}J6!F)WXijkaBPzF*VE5+fB=`DV z;U^=1_H=eT_o4GECzH31Q_X)(_pOVd&RO9!iis75YG|U$hb*Eo${0;_p7Q(e-6&de zo)$L6kfkPb;r{QlWNJthEaBODUAn8-=IzUAqR&b4_D(kEGg!ctzPimBor$iWfq8AGX8reF{!@M#6;+q|KxH&}_622ae`!^9) z5n-wC2N}EI5Yeb+@ zS(&_f#DL9$aabg*rTYX^V5ql2nB`>#CA)o4_1jIlvZoa4o~(ylp&~3gvY0zDsKR~t z;=+k5>2t1D%~#^K4;EwSjE@-O{zdq?_L0!& z&?Fc#@Sr`%ct+;3ATqW3FZ=%d3fOPt3hRGa!U;hV`6w8Rwy$KMaeX$`R{3@(^EPGY z)BQT0|yoXL+(-kZ>jszZI-$c` zZAhH$3O#KsXkFEU4Q&>}XND2rIWC-}C~9&IzaOLdiBEXw*+x+6UIht`0WilQop$^_ zMB*IZK#`{(RsELwu+NuHoLeDFVeY>M?JaJJbkR0WrGHLu2buV zcv{hOhd#9q5!w}P2B*c}iIQ6_DHW9@UnTqK#i5$AYpW)JiNtfFYM2FgcrJd+e*PTl zyd2`R%E|LqFVMUl2deq0FtGhEG5azWjy*dFqq|GNCnSWt>sSh3wzUdc1e#E`pZ98q z*YdmbL$JcLfmqx3(UqT!>CM{#RC)PJuo+Tk41dgkxf_m_w+Bp%M6j>(BHgnoM@;aA~taE?-hPa=atVRJi~R5ufhMYglD%>_iQ zI1mOcPjhKfp`8DWiCnjIDp6-7aGcv28e!}}_OPwOxqA0;S<@Ss`SKDBovQ_-CqKzw zRWWc**a@H4TVl}8XJq(g3GDr03(uC1;j?hI@IBO+Yp-Bm3yS)V@ z@6Jb3{ltu$r;^2)?-$`b(>I~RkS=s zyqFM!ue>MY&FS;WA%)pwtW_@YkIut_CIg-?DS|~);$V|OHF#;1fz4ARvRm#BDgLgC zjdudM+{X-eJH~)hYLlauXH~Im;R<3IXN@@ev(PV&FK4B_r4Oot(8GF&uD5(d*Q?8+ z+8k@T)BP4K!D3<%XAVL9%(U)G95yYwO&vnk!`8fSto_H`z+Y^SW+;OVzZN~SI z?~o824I)`$pb_zc{WtDEGHc>A*sU`buRCs~nw9+f>tMgl?5g0f|dWVvjoO(5B4UaP^rat>^dE>N9O{|AC_r%?M>H56y`^C1Ade54pAf~Dhr#>BQF7g^k?b)F1Yw>i816j; zE7ye4=i=eive2B^_Er9crIBF zLw~Cv^o|aRzGfh7sG9-}qgKQIQ8ILklLEWyz!2|e_Jo_)7SWQGJ=iSm!if#Up`Nh? zH@0;dcR0_0JEfWga#bd*@_A2MlNm;TIzA@FAIH*pWBBtR@66tAwuTm&B$1fz@jxve zk=X%SWoe-+VIZ`O7_{1PrF1uEI%_%i_RCVplGh-yBRL?ccd4wvlWryq`xWheilpUu-2W{e-k2 ztp?saFyx=*=h(h1X`K7nUYNU4g(g000&(fF43YQ+*4j5{O{4@`4Hr`PmQ1#@aS{qS zS3b8f3nrs7*tMpRoGLpQ0)Vcc{tf;6u+1?01F^#jSmu&4cKpfDCbR<*He7dpw$%Ke~y_5gaWyh}DEoPgqW z)A9ZHMX<499N5kd1%J%L2PQXRXwFYuvDlE3ee-E}i688W-vRrjm%xz&(%dxP$(W@y z5B?q-&t2mEfkunEsp{}4BBc2?SDXyNV;2A2^*aFX%8rL`8{Eh?n+R6HN*TXu&o<^AyUv&lgH;YvvtuXW$4G z#VyaRsd$qYP9HgiieYj%ZS7oPc5D>BQm$tID;>{n4D6@*gQ--&&m)`$96@J-8@$Qs zrXeyHfjyu}&7Gvsal(AKV7?PvR8ZJ7>J3dBY@=&$Rg+#Pq^{nj$qm!H~8Ki@sNaF{*FKi@?=Ue(U zavzFrO@^;KUt{)|+1%|WWll0;Jfux}2d&Kk;JtJ)I)2Ee`=8hn(^VzJtVtdu>~!J6 z4tsp=u#9z8m?wOm9RWpBWn=}!KvDzGLVCS`+wsPPJ252-GY@&vSW5GZOdG+Fo(BZERcj}D{w{rGDqIYOEWM>_q!+Uf1 z_wF0ws5nsejo(%6clZx$e~U5`7M;gZ$)Du$s8^W0VFT?|{)I{h#JJNjwG@Nwsn&ud z*y*apj>&omHR*irOJfCTk4eUjZUfZr#7>C(oB+)3hfr2H3eR_6FJn|)ATL6TTM+nz zHd&}~0>_u6AB@>JM1t)@@!0n|*b(@Q zXgsgL!$srKRgwi`!))~S7vlG4+p+YKA6^a7Mw_#hIP+3CPW>jz$u-84=Z?DYtEn0$ zh)#ynO(#)riWGT1=Pf3l{YQ4q*Cc`u-+1P;6Q(S+;q!+}!RALJ5qKQK2E7|-zuTV% z|8|F+PRk&oicsmoU2o?ADCC!%j+ zkq9uj@+(<5T8-{I^%uTQTSi=I5l$b$QAHIev>XQ4~9IG69gA6slDaudJnah>9h-29gv^oaI7n*wP8{K*&4&Nr5z zF13NZlh{Fy_i`|P!613)&Sy$D8dHClb^?*dNWl0#=p&gY9Id6mJ&R$u7hz6Z;mU1L zabAV&-TjPs@m!G&qeepkIRY=Q$3u_HGq(PAHmThw4F&r|(bFy)4()JZ+>B;0GhR9| z13jPViby#qR-HqIBi0zTKUg?6Jdb+I9Kh{$lOTBD0SQ_14)e2h_~;jlvXd7Q^WN7~ zHl&e!xFUs1?(sWm|JQWlzGH&v_q{;=t_u`@wLn*`o%D$%3VC@G2JO*@vGrfdREi?X zm{ls&)h~jrTW}s4`ikl5?`yBjC!`2$@SoS}zJ1Bg#N@914ngNuIW z34eT_N;en&gxaN%aDHthY?s!gnj#u>V7wnLKGMu?$qpv^na(6+lQk&ko*_lXV<7HT z5SLWp%za#7&OMni6Mi_JBO0c9RDH>GSZU({GtYb=1#3^C^|?RzZ)qA9YaFLZbt>Tg zcp0dv9wo|0FA_zyLh@yx7Y-Wbz>y3~i0oJ3vMNlu!Ux72>Hb9e#uu})wG*LJQ=F0I z^HZKPRzt$84kFn*hGgI8XX|C!)Tum+Xo;ynGVjG{ydc4Mt<31{t)GeL!_oA0Wfb+w zS|jY+ph=BBIAUIl7rW5Cg57ARLdwhJaOKqmQe-ty=G&2qA%ZvP8=inE_hs4mYp==r zmTIb3$nS1vB(WU>>d?*b|6@M)|1#qU&vO^U-P;3&3wquXzbO_ZIB*{Wr;*xJKCLiO}n>UVA@-EQ%2CX#(y62_F&vb{-oW55IU;-h zCD3`gJUiw-uC{XGj9lkqxJ#H&@~0B`C>g;a?<{J*RtvIr1(Hl-b@s=?61I19W^A|;vzLx%ZDx1Hq52m4Jdus|j2521V*!nZ`jB)iysU(* z$KIj>>i(^f9$&N(%STu;Q^t_HW~8NsT__^d`O?q*C1x8 zsbt*!403aYA1m@?k+3uAhfrc!0BzmcO&piVL%{Jdry6mT zOB$xlv4p!}LFmvd&Qv@Pf*%6jTe4;hw{Q0?dRe8A$c0=4^|Sj(*!FZ>kXDR=JTLM5 zQ#Ya`|56ZXl}tZW>0oBoUhK(ui_OknxUE8!QA!Vllpq_}RJsVhF*oVrwRv=|yA2-r zEl-?-{Hdp=6^*p-hQiSUw72i3u&^=^661YXejx>^^Dcw5@C=F{Sb^eN2O+m=fK;7P zV=8(tq1&@-7(A(%M(!7&>4GYH+^LfdIvGq#Wfi!O%B8gQR3Uw(d7N(EI0A_`u95c8 zaWGnX9?y95f@Ohsu<=9@-Lx$bSDcN+x$|@I?!G=|YqpnNO2{e1UVtNqD+< zH$GjjKzxlmFlpW(ybb$Ai#{*McP_VKY`}HgdB&XPZ@DG(>=#Grle0nftOJQ~4u@dd zV63kU;wF`vaTUx=EU4r=Q&)b_sCVJi{;wE!@3A(9C{Mz6880F~;RXNAdx7L6q)@yu z9X#Bt*_Pc6?EME@$k*hH&^lbpURBrT#)VGdTs?%`dLu+Pu9^sDigGe;RkZ5xSiTIM zJNugK6)NU^Lpbtc6a?<=U{hw*(wd$R!i=Y*@roq>`LM)~I};g0U;O@v(?6~!Rfxri3><+UZeU#|ewNe@tP!h_k-bRTZX1|nH-la!rL1`nGm+@c!>*`u^!Ws?Nn z{Mk(I)-1!iSz6#=tOqr_=74lf6`Ae(gn;EyTtEK~ahzC3HzkR3P1lw~|5iEZJv0$6 zE_w{x{fw~FqfmH7<|!K!7D@(^#YyI1C23h13bSPWz}c{#oIWB7Dt_(4Wr~yF%$l2U z+D8P>E!X1x6Td(s><6iuF$q7*hN1LWU(^?5lJh6mVbt@PFiz4Hm0FFNSYun5mHmp8 ze~E&(r^Si=lQ#mvws)leO$EI#%e!H>w{nYzZ_^W1FNEHf0eIBF9zTYR!dLZ`II{RN zz4%Rx-8LZy4%Ayh`}YExsdtYqTi{4s`c*J?{2nN22_i9Jewe>TNS-Slr_O0rSo(em z^~}FZdaC(8`W+V-JXg$aDMeZ#{ZzQ{(|Mu3(HVA%`EEw~fF6eMISCc7Aih(=Vr}*X zVaLN>)^tn@5gQUhMtu$O?LA0mv|E(fCYWdsJ#=(D_=ESl`Wvx123}=%&Y{NS1-ab`kHtw zQs?eBj^fn!1d;TIPtdYd5wjNw;g+5TQ~&uk%?W{EF{fpLHA0yPY51a1K#lj5Q9I>4@?@?3W`V1jY05s zc^q^U#L$n@d+@9O4BY+w1~&duz)^}v(dS_?F4q)*Z(*=-v)%Fk-}$ebslx;-1oAAO zYc^ZTZ_xPCCs^d2%Bt+((CKVB-?ic0Ce{qy_ePWYo)`uTSy5(UV>n3OS_-WZ=H$PK zWZua+j>xbc!Xv|*abx&SkWG_^-pU>LJ1PmziiW|{edBTEOEF5EH)G{{HOyD8g=yP% zQ73gDR^-Si=8lOk44%}$6rB;_inm^9vuheX_ed6p+jrw^&u4fbS)2>G^p+Y+o8n66 z^{oBr5WE(274vUDAW^3-)3J+mg;D9sP<}TOGLr~eMjNq3_rqs zzW~!JEa22bXIw3|3A)T@lPN26d3M-M-V=Ec#gzAA*Yu^h?t~TM2+UwcP7cK>8A zEjfsjODtid{&qO`b_#OKb1>6P9j!$yi2Kw>Fk|j6n7eET&ilKNWZtQv?_OV_H$4<^ z`Sc(dT7Mn4wr0Wh?Ni}>R|5NCWjfTqzd+aNttS6+pU~e+j*Jx- zMn2n#=g#Vp#rv&EgzQm_3e_Nnnq!mW}Q|D{1}_O&r-dACD?cMhTgE zzC%16@BF%rx7Sw_@e(0CZ_1#1`H{p>!$$V=19Mchy8-oCrR<`L9qdp?5!l}$V4cuS zpPrQAGA{Mvth!~~(Jun#lyMoHd~ggEgC|4Wln3P6gHXKLn1S~_W@6so7j*BPKKOLi z5{CH9z?lQB!ja2Ww0{GCSZud!?JQQ0^7x$$xHDQV7Jd0e$^S!OvezU!j(9X z_moQgl;rmAQRBqyMLD(lT5K?Uhp#3nbI;ZN$->)biOqd+*d{*~<#Sfy;DnE;lxaj2 zJ(iNcJDSPP8~G49D9;>nPlM-DF+^8RmQ3jEXKnJ&V0%$4-n-+4R>vyI){VS#=A$%O zZ!nJ}FP}haFQy9n)bj)<&j+%nM(GI4TdpyWD=HY>ge)fhc{r;(`V7&zdxL5|8KmpH zR?!<_p5&fIHtt;=f=3VO;5I2E5@1hA^dxz9mT4`2&bWvLtIy-FKgF0ZVF#M-C-l)` zZ*-Pa$3-NCVqeR(ZOg2u%2Hmb5gM`|=>R;a$wvA#cWb(F4diAx&zBEr^QK z;{T)QJOjCU-!P7h%u1O>6Aj8r;XL=Ze$gPLL`l+8iIh?#GO{W&*-|1IWkl9_?h`5^ zr4(ssC@n>lX!)Q2doR2=-*cYlzOU=^!O&h&dfM~ms}hbP8)}-rkatZykBjgxPf`7 z{+J%!`;0Vs8IX;3Rp=qD!mU20z*%^l$40EdwCMtTsFp!rzgWUX=$FwW1!`cGp9?o3|@&V0|} zl6Ef5J|{|Lj=9#1_x(lMtoFl;V{;ja7q@8G&pdh~Du)p|ENkkgSrXu^+#b60-@ms2ruvbDG{#6~24Wez+I!2)Uu;PS=NT<~xP=6#jqB5v29)aw9pVPq2H zAVrznqEGna;1hh9`w1__&P4ne%tklfV@3F9B}v1P4vkeId%qiClRfV#th|6WMn$GO zZa!n$CO)D~^}h7-usXX$zJu9x=qFXnOJ%@sAx3AtH#JthgD}u^rR^2{QEFjA({-Ee%HBTp?OFaMd$ar|QWdhG*m%;3Le`%m;3!RoZ1N}#2>EXyWdiG`&b$um8dTeC5 zdy2nt|8ZF^K(Ca&Yd#6z;5n+KV}j9q7Q0bqADaETM8)=Y^-NQ z&9Sip{FGPDj+ZedCFf0Xi2q#<#)QGj9n1JUls7py;}7Y%Cr2O7Ji}-%UjUQMj4<_$ z0AF5{gSxp7>8peQ(2Y3=N59{v5m%m*zZ1n_y8S{ljq1anIr(JhO>52iOCk`f_ZIYv zwdndo)>K#JEODLr5{vGM;;ffDKy$w@S_a#Z)3!&UG-o^)nSPxdPZ|Ir5o45~&8Efo zG8n-jX-?I#i@04?5R5ncL>CY2AfE5W!13*uP%F`xt!Dm^f)qo@xc8p=FOY|egQmD_ z_8t_0>#+P?16tZ##NqkXWPAK(6tx?Z#V{fh-E201>(A7@{9a z&A*<9(T`VqCvD`d<&#E|VDCiy@ z#Dm6G@GVwMV7T52gL)4Urvqm}@TmzOZg=H7Sl{69{hOFJTM?sl4dB|I`_yXsY}D}k zO}?j3p~&~^#$UZf$d6kXV(bE1+1=prw-y(!b)b_LM}XtN6nvxWL&d$r!7i{GH5=Vv zYgHj6U%QCuUwf)gPu~t>f-Esc^A#0KISOkRcH`e$(lF?uAvm65Nf*m|qVBye5KDi8 z+djvl{D-9w9Gy!)EDpqjl2(xN_Xq}FOQDJ1=YqrxL-gD8myYpX3%PL?%=)T#bmKHR zdghod7Cnx}AX_(BZBu~PgT}zX27R6pk%8Ym8i{=VdC0yige7zKLr3x=a8dN2dTRb4 z6&4N)MvF00xR5#hbB18@8lFx4?-L`pX#?z^`yO?-w~<$eb09EG3xD1I#%OiAgSGZm zG&kV8&zZ;b|MEqx57ZKvY^L&0j@smjE)^?knCGd);=u4wv!&* z><<#S#lZnnPHbYLzdJ%D-z5&6u???AHDc;d4bI+H1&94@;N{ng)V#QeTo8Pvdg4)d zO>2aT=Vs8M;*)40GLhdaJ%-4&0^kRBkUTP5pu1;0XLY}m*hLRunAS<|>a*iGdT$nT zeZl-LeFqtL(gi*p-B^>bNP;ZUsH6dWrukomD=GMG#50=LkQ<6Nq-EW?n!%tg@GRvZ zMkE{Hy2WJ3S8;=9?f^GP~Cf2dT=?+8GXyF%K69VJyPiAv|6&MY#MwW`oP-U zbHFV@{GR=0ApLI7?<|g;CGw+gFp_ZxEA|eOEvNKhmU|vE-`$P;)QKSbsts`M8D$Ku z-ORm?EJde9C*XRB9+>&5fd0;Uu=v&nnY$v$jf0`|NQE}a`L0CgW2DN*lH>6ZO$I5zix!@5h!rc z8INmD&3?$5%JCh6H8x!OzR$4D)(NGWY8i1&H%2hW&gAgv^`yAv5+1q3pwjIZSh;=- zmoY&DH?PWN>_Qqq;YUB&>~xRZD6Qiljjm*K5y0 z%)KG@%V7npNs-MSZ! z>-W;xAu-(KH+AIX3?ER9o6a#jw?tu8B<#a7V2a@cFzF7}0r#69hWo zUXjGwEaEv!na@#cr777t&50cQEyjJ}9eby>!*Rf{q2}X?O86lx$8A&^raEn+Ak{6! z>HbsU&TJ@v{uAvr(!0dC$oX;LXk7?r?elPlZFZH-qp^aCbEkv*$TYH0YbVLv{u1ur z?Zv$Y(a`9UjcWNDSxH5A%9<3TYP|#w6`LWxo5(#E`$1o4W#Wt6Wf6{1hay(_x?Qb#QttM%VwC4hQ&C(R$$&==RryT%EI!tCtRcXBUxp|8zQ0REEv| zGXqaDYOOOed9Cok-@a` z*i@SFrn+f8Iv zdX8^Vcz>B_>(pt1~v!d=e}5Kop6&58eJprEA=6v;u);re`{XFVz{H? z6sz051f-qz@ENI7;3ac~NWZuSwrN+1k?Uf7s+fpX1|z808wG`z<}+9Q6`;XZn_F5h z!)j?W{l_XwxR0+e^ za!6~gkYK&zZ)WxNbn+oYz+nu}hL9da+Y=%nkw8gG=6BYqJb|05@R!=WQT z2W)JgKwk1L`c=h;z7Xt0gN^}olitLIZByrZO5!oCkHKJ<8uC1DJ^jzQiitVIbKxGV zK;-28aNU%Ux@pQ>*-1D4?`v@0ubv^Q=mG*Wb*K<)foYWu)N;8FWPH9%c0YJSCT{#de)koV(8hyg@_YV$ z;TQvoMVCSQjvT9V>=ilutQ{;?>qEdROT1jSkUr8aX0&swC~@|}3Hu%qANfPr`b_rK%t z{lB<&#sZigI1R2fG=j4DDz^UU6=-+-#&|FT4;5UQ_7Mz4tp%Fmvjk_R1qsfraTFYR)Jp4>yr_k+m*bnikWL()zcDi&ID`Iz?wG_7Gc8uN4N$6>C z8>19y&Ya0a{aq)qms`!;5ZzfVmUo^k`o0clr=27lPM_p?K%>;$_tv5y<4hv`cP3Yz zF3g3VTFkjd%fqtJ1iJpgbY|7e9o)@xmYj6EHaE-T0con$CD-~q$xYE{Qj|O%%_D44 zH@1qt6+O$ljxG|fe+TJb-Ai=c*5R6y)A<~QS0kzN+za_TWf=cI^6;N|FgyPia7wP0 z91UB=(Kp3#!rKsDe~n_DWg>n5U#W(y8qF6f$I;NqY`&fy+4VPt z*cgT~Z@-#A`~(w0+68w(Tvw3b@(p3`MtcW_$Iax1)Z@rq?_kop|2gpxDS*oDEu=C% zl=xk~gd#g)>De{w>6vDR-=!@hyCtgFV-^8)q5W&}&7p_7?%PjZdS{U$hb1VwWFdWD zcM7ZOH)2;p7}A2htdxZrc9t5^l(c4YchN)UnNK?2I8=q&GSXa2A<*_I4fO4`i*!nO z3Q6hVc@_UkNq+A*fx6*FfmGZK!5FfI_q&8a)kjU}A1q}ADgUALJv&mnrxEMMc3{<= zDV(BdH-Gkvr?w{J@!X0>RQ7i^<>Wp2J>fgpep*{Fc=QVx%+lnuq=q2caF9yOK7uxN z6Y#+dVH7+G!AHY--1PG^xL4+eg2|C#g6={M!6ULo@ULKjz#&#u!0xf5wWftNcNQE3 zywZ)T7OnWy=qWBayO?|_@Pk^pWb*TQJ-bV-n$^tmqI)xi*@e4Ih{4;dY~9O`HBM_q zXsoLR__>#oQ_08KNis%cafKrHN$WO-A57uTD>iV>FrTn8fpqnuv#_lA4~UNmq(V`~ z{LpTSfMgg5st32CtoSuLa_9@$7kPrni`dZE9Z#x<>~sXK(prM0ZpqLzRg4Qz;91;y zqgb%@Lv{aCTgV*AAjVoN*$*4GQbo@p@@;$s+8ku)YnoS#L6`^uS4?4dZ%b92C;Dq&ZxHQQh6N=_xFu%rKwCNFHhbKYFRuO?y=}*{D zc9YSvwVC5lO3nD?~)4o&VvS{+ya{w<#9K(N)3#iYy6WGMxSE@oy;XwOd+)%rK z_`en-k}5Cp@4s=JbnX-SEWw0ZvZoXMifnkN&H@zvdl$wU&S%oKO3B>uS3%!Illkay z4l^AyNl4oi<{fjJ-}8@{)|ch5M<1A)l4Azs-CswL)e9j~3b*LrkfY3Ekg?37`_Pyw{+`VpZDd3J0`999n$!lP}S zWXdulI>xP$@%>&nAk(&8T2FZ@C@=V-)c9DWHWH-{2I})r^3+0SpY%%8&h~U!D9} z*CW?hX`?e(uUk$^6E(OUtMs}0k!jqZZ!p-|E}>Og$u!7fI+i*fhcivq%=v)xxP0Xm zS#iWm zDuB zJKS7#h`N61ftMy8d^Yv{(;2nSS+<47K8a$#zIlwF09&kjb>%_ zkd_R8M(>9#hU`cMPNx-(#tfQ1kMe~%YYx&`iqlv@v>ue1y~RS8Kg!kklnc zm;DIB2U(VQ@U}i`N1EaGGYjY(dL2e4tp~%)nXu~FTOuyuSJP823!`6u;yi9Pv z_SM}0L+NQaJYfbIqZ|W!_ie*piDGzm%ze~Tk|t72D!Dgj4ahENf;0I?LB6^IBDF;@ z;8F>mw%SB4g!_}7ZaMU9t0fk>+rr8F)5)oCGkn} zaBk-+(wGqrLj@BBpW!|n-Py&~$4`U&enVKWE0LdJxWJ2yBqDvO3KNeQ!rjkET`zfH z?Ywt2foCQN;$J*LN%`NT`ky%3#tg7Wjw?Y($SPR#N(-L`e?`+^CzPD%gMHGuM8QHt zu!I_Ow9Wz#f1SeR7P@oo_hb1Pt^wWBq(|ceqy!+f9j26jCyPQoA(&^dL}aPJh<^{> zE;eAC^7;G6gjKwQ=m$fNtY9NT8{pOD)9}UlEq3fr$9Cmqta6$hjAWgJ_E%c$bykvw z4STTTIxXO3i3<08$t{fDCe78pPA3nlE#azo3Ak~)Xkq*n)ccbp=ukwntEG>k}X;^=}eN2*hBgf+{Z zLxhbDsq5Jf?6a21ob!fc9J+cR?-@6dJzO4D3({rEqr)&E@-)d+-T?l57R*J`)ucY8 zi*D}=;rkL?wT)4IwG@a8hSwaWI=8a(=OUrRTTl=V+!+gk^7sUH& zmEdsKE?8xeM=DfGh~fTDHm7wy`MX*YcFis)kJqWdW!nQ_@A;RQ^khMEcNgh$mH+>) znC@7=l*SA4>E?A-Q1E?(cwI9#&3qchKLh-CV!{iya*_!bCo0F?=qFsg-co=Up)l}h zB{+Rw1!ndhWMkVOGJo22JotmfL;spE`uJL^^?L?%#1|8X4~4A$IR`SS{vuq74@U0j zE!wQnN|#BmAxZltKvCUB+Bkb6s6QMFmv&zVrAHHB`kZsHZRZdyZ<>!vit@xSR)yMb z=XY%nu48))-&G$uMLX^1ldWI*cl5u9kY_#vH>_5MZuWucXT`g8`nyuNyd#6&UgH4o z6;;vdn+sDno_91kMKfU!{Ges+1Jbd@3MRR9VV>+1j6Jc6epF?k*D(kZ0+QiT)F(1! z`DDm`X+~Wy^Bn&@DX_PGJRCdbiot3Vap~6u5VtL!{LB)CNqc3$I=dK|er?iSI3C#Kla=s2(TDbBsR zZOwSBZ6Rh3nh;tjiF@=?%{ABQW>p^&ediJ^%jWkUgV&kJ-4o#5h25~w zdknnEcuHDo9kEqAf_}Ug2KYh}K5KQsnmcOD?))VBZ*Dr9AurOi z>lf*~?*pS6gXB?Q7~H36bVY9}#pCyxv>CT))4VVE;NE&LwWtF!qn@RQB%tZ#IHIkz zm44gm1{xPXGBj8XaMmN%S!f=$nGBFe>&eNOHlk9RNZyGrp@Y?mOd8Jy+Nz;LPC4&| zdd+Dt2A-39840G>`Ljr^TMWc|^GuTL5~_RaCq9@qmIG}Ma7|qaxuI7`a`aC2B0sY) zC~k-44?i$H;}*iRS0eE9-xTggS_2;29E*Lj4R}#89F1>}LH|r|Xi0v;zEix6KIkeFFe@Rs2J;iGLDcUJsqu@VjN5CxWxpN58rDIc z{S3}}8Q*J`H~^Ebe529+XE9$n1-Hxy!E+lY!L{j6Ft$4iWgZ^}(Mgirg#rFMDl?a< zZukIY+rN<~$Ag%JCoJ2Y`4SJGSte+*$|R^5iW&+_1b@XR3YzsRFhQ;ipu`cl6OowJ zV~N%LY*T$p7u5?bql=z+0CVg-#vc+Bj27hLg`CYGbbv>nnN&m?vRCj;N`*!!D-@^B}L}V5i z#8%?ohIn+x5GdQF2(|Nn;uV+s_$wd@rjK8TbN$PCel_n3_E6yZTwh>Jhbgt@ojH!u z=c!|n6y6x~+H{OzJZ||1}CnHHC;CmoK)*3|8sMXaMB z1Kxs(^;#kkya`Kx2$8OFd=J)cz}+m42rtEZP9~cA%54465n!ut6US|r$5xj7iG&NKRZwv*sDGcY=37gXi^2hQs*(WY~Y z;hb0uM)X%Bwro!y{1dwRm0&k_ZV6|@r_473b zyRBonJflFS`{r5tPUs*EM8#CgE5rfgsDXBdsoec3pBd%y8z8?s7|UDcqRoq=V170U zuQY81jr%!t__YgF3r&I6r%LF`>Qi-{XJqg6Qe5&Y0(I_)vc~TA==oclo8fQ=EoC$z z;^UK=o_8PMtNbAIlZ0Wvh7chWpO*RECykP;=Kw-1%F7Fw^M!So&Ff^;DJ%%}AiOqcc#wHw{vy z50ZC+?bQ2ZArWhv&h5C^#$KR_@Gr80UgCSa>6@40@aq4V`^hE{!ZV)!9sEFJmy1Jo zPcLyXkUnUVT7*)#TN* ztLhf?oRJ(bg%aK}`Pd zPEk51oqUmRr2bB2Fx2`ViWN^~qI)et>9-1wF7(G8;DIuME5Y#81&9k56ZC%XVm8Fi zM`@2gD0iX(e!p@f4xtxdp!z4f?BE<$c)T~Qx|f7%v&Lg=`BQdDVLO_>ufz|PHDvw# zKJp_*mAldl+^qGjtnBJ!*u7UB`acJQ($-pdCnF30gm{c|o$N8T`>tBlnTg9d@10u!f`(P%weo%tj5b%m;#zygdOB1p%SrxYr zIxrH6f0*0;V+Fy><^cco#{<6?Lw0&3kqS;`nh$5t_%p&7V=jhIE$YzWWqnOhdllVm z7YM3`(=cz~2%A`@M|!@_gHv(WnKyAPn9Vtg)3&|9OO5*6I+bXgDD;XRQQ>)9rE{_T z`6>Ktafi@QC9T)_uI9qsRBg8+{QEUQz=S*^j~l|kVcZE4+T@FhQPXi> zB%;*O7u33U1^HPa!p&S~0#DZ71$8G+@Yef8$Eu#dy=Oddi>#>Ny^#`p*D8X~N4wyc zZy?x>-le~G9|X(GS0ONP5vWc#fiA^tkXZO1NaPm~!E6BE{A4iy^8#mXY9xM&1$3N# zBJM7|PNv8lpg}p0A^l=HbS&+Gsg|-}R$B-sXFjLCUI1Nnxj1y>1>+omKrNq;4KM0J zQ1b>OR)0dTPI-)tcn5dHPJ?da1IX77rV(TGFlwLI{27K@=}^Lr zU{ZG#8zrN_olF|Q^rL5RptPS%IWP=kBPR&vw4cLWjxR`g#3pdFJWVnV+#$L1tLV** zC&=xC=V{T#hYSG=+{?3gy%OZ8iPe7otR_P{wfvxZ;0FC38w8(5!pQf!N<0|78s2_N zhy2PpB-bn!bL=+pc}U(tXgm?y&L=UCGrKU3&$sXKvIE#vjSt34ftGzC7)ScYhi_&>o>IkXG=- zz@Lp|=YPI*)^=eWJbsa-au*=+xioj$)sgP14FUbH_wnhsWcbvUKyIIz42|o@3FMY* z3LZU}OJ?1cf^S2zP`OMNqkQhd6~#&9nS(t_*8d}mYsSL?zAIO_eF<1kRe=IWB~CbU z10;`rrnmIh;NBlc;K2=Ni0aP3-_QAbLe2vCb0HnPQ_ES+_-iD@N0GajIKbX^I7t^6 zsiXeN18|_Io)tS01**|E$?fYN?Ao6}=&f`aHVC`1&etUIzh-N^Z)gUwo~1Z;!er7_ z8^e37Lr{0>X)wM+>8D$lF(j>y4*r=3qo*I@54}Sinlysv zlg9(y&*#m33oy5GI?P-aPakbCWLFJNhUi;+=*nqzI7L(zP6)1(r+N+OY$hdGvpt&C zG%mn}L+T{4I2mLO9$>r6Q94%YG_CRu!Bv~YcpuACqB1iB>-8>?o8yn0y85`HMfh`m zcdQL6Ti&tKehB&YRS;zSlGuOccO>D3bmqf6bpBHOe@;F7G`j?qX9tt@LC;uCbDk;q z)&N9nv`|iS8#yp2PNcqT(*-BL)410r}&pV}%`ja(Ic`zuV9R8xvSG ze-TE^^M$J~LqKun8yfukB-$tV!K&-g^uH1Z#&oSIwG0VBj%Nd!sjHCgW80!>wk+gRFW{$lchc8#^k|QX>>G&kW%kzeG>)k6D+qn{>yCUHJpad+B z6voDHSLxI07`%9Q0~1hSMeiMsq=Cg_AgkI2{AI3Ejdp7=6RBgb)$x6|v4835Iysz` zCqg7tr?SEgV~L#487lkv58IY_mp04AqLlGVw!Pm0n~sFj?AtQb@w*1xX*8jquZ=@) zaTk>~T7mwfXR-IZBdRA?GH)+00@r&RS!K@$bYXfm^{X``dMfvr{Xf&loYT*U)nDGD zBZ$KqBXK5ITqXlna7_S%JSv~zqsz307?d`y2tZJcy5qHQ(Xws;yws4b<= z_wDG=hwboXg$*1uEipZ@jbkm#ub}vC0qWNbQMaxPqTt%Zf|mjq@5to&2rQ15XIW6b zNqGAtt3nX7Q)}#QUx&CO2^~Y1!Q}A` zaBmo`$$Xl|idNg;J68?7Y5b)6*57ibWzSvuX{4FV8Ti8vk2DaSUIiFx&LWXFuVUGx z-59p!j>*TL7U1+>9hj87hJPPwp=E*){JtnBaF_W^v-)SVN&hCJ?0yf@GkGHFUT&Zt zbVq1Zs~#D{bI_90n@EvNJJCIV6)FTvpsd6J?bLecP5)ceOv(zLjSkctX*56_ji)lB zzp5(V>BBYETzFa#2Ep>BxOAit|M|>B;kMH?nqW;6Oe@&hMtA1>vPI-rc^X-4_mJ*< z=Uh{8zm^6(bEdf!(QHhpiRmQm1vul$M!Ls&9RBWkOJ0mDg)M5Y$&6F?X~~7pG{bEY z8JaDHyI&;XEAKYu>$j6Nx<^XrSjX9*aCIF{yk<>YGZ&%Wj&K@llE6$mvXDF~yg_6f zk{N}f)usj6j$qU`XmYYF8nYY}&{WKfl$s2XR2^Zw;J%hF$vZ%g^BIj4E*8G!N(x+j zyZG5r79+Qp!cR+OX!+(xKjrCA3EgO_oVjaP}l-3|6UZf93_ z43mlt`EX$V7M`=S5FC|Ez-wh2`EYg&%*tJf8y`z@A3xNPY1b#xgB$1L^o9(0a4iaE zSaakI7f9ZPaOfz)_xrVHfQLvWMu@aiqB{yY)jF^^+zGNS`hs8EauT)l6#hEhMcVG4 zg>y}|&|7o|ZqAWK@7}Mh;=Z@c3E^hKhkBXLiP@t*5z+U)TkpH(`6 z!$NOW`ZqUTC7iJe;MU3b#&BoDcp@T{*9K7!VwSS_>yzg=3 zsq6tJ+I>21@H$m<+Fy>j7_SXOH;(c->6b*v!nWqjxe`XL^A^VMR={8-0y>vc@s_v` zK1@DN>XNI-uyq<_8_lH6D~y?(cS#udkaxu#ibDB+zPNLZD(>%4z{Ssc=-F=#sBnD^ zF8uugKgy_5>M$AazC43z|Mu|R*BiKOMhp6yyOLG&T=DP6oz%~c=NkEG6OE}0;Z$82 zy~L`c!;8zrZvQygeCj;OedPfU&fR9tpECtD!`;wx;WAe3WpT^{Ause+3(UZ4l>&J_ri|R4Ux4fP?Zr1gBCziN4u@l*(DG_KKT|3I z=|0|_d{&h||6c^>UI%he-UEL+6Ohi+1ea0_Gx71CFQVQBpCAq}v)&#bth3nM~n@OQu|{MP-E zoJ;;rd|KjJ|Ellw&gH9gN#AZCP~X*jPxi!}C#Sk{#m3K;{ z#=#Hwa`JicPB4}V2C=KpV02_Nd{_QUEH~KD!-2{)E&UKaY8hZIbm)Q7LK|W^$$+X` zkC0z&TLGJ!Nw~{6yuExPXZz2BoA1wij$F_1u2f;pUNOsbYy50@l%)gd&OL1FqNfCF zct4ugJyTe{5>V`41n~$-p zY?R4;V;xv`#<}KVV=69v$#bWyCJN@RtpME_IrPeCB0YOejc)0^M`UiCr;0qY{jg0e zB#pUb$^yiRm0d5NEh4S9w+%(uP5WX=P9PXCb#H?>i~x!yg+o+@;O zjJIzHmTZE|$2q1Uo6Zu2b?4~gQCqw>R}bbfPpMCrKW009z@vc;)Z+DG?nAU2w>@AV zSgcwIVsiZbVr>Ld+p1qO&9U+*-G4%z6i7{_mi;xGZ3wG z3Hqt3VDilnU|V;S84Q2cZcHM3q9w_QZ7XUreV;*!g$Fp~@bknC6=O|OJjLTXbuyI8ocH2-! z>B3oTe7O@nLgTBq%Sus^_1DN$#YHu_@)^`0O`&yl0ok^>jS4!iQ0VI=S9;{gXMUS@(f81} zguf#nnSxX8l~MfWAOkM?fVXIFi%}>rAg|5` z!|uz|1UUoAsHSrteY(d$<4Z?c+Eh+sdmhqgNon3WI*nYc)T1>p4?bR%dZ27tYeK*NdPop&Sa14#JM@$-L*&lgKQ74>9sbiLTU4 z%=jq`T0i!HSjS-~eV##dGjt(iv=DmaatS{4hQ#VQ!0?=9t$*|JYhoF5f8}dR*IwXx z9tVku{8sp_!n6}JlwRj=!4LFIhb8n)J^D}fwlI7%tBY{~xAJn=W;MM*}tax((<=vIRxG4rFyPlz< z&1x`4I}L9M{bHqsu0eQ+0TzGAr+W{j!ezB{WZ3c~R*vz&2ZAoflho2n?I+1LVOeUh z{RZ*u*&xY>{XNlq53q)yCNb8 zbg{?M$w%l9l_^w9;U#lu!Vfy9V^uOQXSP~6iNHY zsJzbyuMh#LTC<7`(w>2Nm+UZZ>J+YI-hNP*&V=Cnu~7E$5ZSSj_Y4X6djcoN8Gqq@ zl0U}7r`=WXS!g4?nD1C4G0Pk)ZpuUJhhs!uV+SZMbs^OwEmYWl1zr~0Lk7=?unRTU z5#vT(Zs#M0Gm8`F!lZmii+eb8V0a0hkP0C`-)Ye&qjyYw*SnKbY&5%N{%m%z=@2tG zGl&lJS@V@spR%TY`Lty&&yAcr9_ub$B2%BJf&F3+9sMxv^mS%8=MRxI%ON{!n?Hc#NDGM!DuwFy)&HwOS{M^Wy%IX*tI48-&cu=04;@QgpNCe?xY>_t!i*ZMJ!qbYyw@ycG7BMNaL9d z5Wh8*6m{>we@o-xUQZ;K#Ovt&oF{BRVESg$CTNN`?zxw}|_W4CMo<}YNSLOepFb% zZNA)@VWGN@;S$>meb`QhsGsPt5lmg8iT8rQ8BKY@$3)t7Z<~DW`(=D6XN?R`oIY|IXvu0!@@Y?n$0_#4$oC^Wm;n1RUg< z6JOfpVL?p;v6p#4Mb#SV(y}i^?n?>1^}LRL*j7c}ciLwoXIne2D?Gqahs2$+JB5~rAZZ<~SryP5QGX>{}RY;bUC zhoM(x)a+RTqkHcIb@-En8a_&(i5GZ2PXqBt45z+zx2b>IFIL3-2acq7V7B!IlvlcA zdVbe9I`mPLcAxcxx@S8f@NFf1sDFx@rFhXxKIYi;<{F9Px#nD-8)i@Jr`_$c&_a1A zWd{pfm^QrIUdYI3tAR1waB7;?uE{2y{`$)t63=j7WYSn?@z z>oClyL zA(;7t!z+J-5UPkYE!OfRmfvoJr^F4^Gx5aemE~Y1wwCKm`a|Eo+>a$q8d${7 z65Nh-(CqY$xOA#4+%buVm1z=OMCmThSV@E^d1d2WxiK8wEuf!{ghP6#Bp4mOgpZn3 zIIRtSwCrRDa3Mn6f^Z>jZR1Y(X77QeH79F=CDTd3n=s5ekWM0t<7yK9a$y%&L2j&J zQRFGle^UGp|A_7dsuquXhi%B0fz2RYGm~>uSV)>*1QL%wg;X?eGvT~%)%aYF}JoUwxgHG3jD0h_tch`;f?xsNTD0> z^kIIdu3Aw2+u|&_@^&KbwLJp{*K_cZpCVMG{e=A4&xqE#8IT+)CDpG7+qa9HxLZWDwmbOqb8}`gBlq3zpbslGhLc5fd@h%k+mG)5I>+|>T zzu*0M_>2GU`@Wp#alBs7=LSp_xO>BM-`E-{AE)cZYpCDYqwsb9alEEENIsrCX1i7^ z5=GM@U{ILXSj&Dz1zo}A?Cpiynj{!YIE}Mxj>4T!*NKnYIG7jnjAW>~ah{_2@KO3b zxBb$4xX~L9jk=d{#K+rY#it1{Lq-E@?JlDFwue~SQG_k6gcaBXPUvPhJG^DARTw>_omI)xPcg!XQDL<+}n54OWwLUAo})?Uj9Jd%`GeCCHfeRBymULcyhlv*eO4=;7UuY3dJn*Mg&$P5Do}sn zSuNo`3>Uacl1Yx!Xi9w^HM|&$g$+(rqe_G)R8PQlS%Xwn-9o_bQ6UmF|SVXQyC}?`E{rh~jNSM!|V~aZ!VW}mcQ3g^z;}09e z@ib*3k#7ew>krYz^G3qazR|p=_colO`Ht?m`j7h%wgC9xfy&iS*0p^6y=;GF*qHk=T{J3kh|(i6vN+LUCHx=j;mWuiGa8;-g1%P{lg z8aSkB2>j?%^n2_AICmizzeXm&xc!oleV~a*b|mtqetjOPccAMx|X5d0MA z%HH%Ff%;BAGBGBD?+B_V72(xn=4~~4#z7SpU)xTW6v>mSa52(Q|C`pg?4=Pu_mOWt zJ@f)C!uho#!5_YrY=3tFHwf(G3SmY&eBEElyf@=k+XYl|nIE|>bBVCdYTDvCi{;r3 zVNOR=iS15JOs>r1T3c)B(bfg%9r3_6!8V4gbn&KoQ_|^0~)rh^BdnY3J4 z!iNrAweks@PA6{LCv-SmxS7e8-A;h;JiuY{tBf%E2*l|Lal||>0sq6|>J|sb2ee|P-izm~I zW}T#1r7Ns~t%+xEwi$Ydh(%B0vqpdKpQSf^a6{=P}mHClD1|uZY=erVTGYE%4|QRI_6>3^zn?JoQPM4MDoqawy0VYg6;2vx!)~G`0dv@ zu=bOO(241&{b@9a$1Bjz79a4xZiXJ#{_y8mHn=^gCZ>1IvGtoK>iteW1q7Aw|a@*A$kc z>gAzq@6cF+(gFEdDea1F&(D8RxL_CDj(X#mC;qgZt++ zoO|jpQTr4HZyjc_hps2M-)4VEk?T1a=7QY%GetOgRUMgsUXt4SW|I?Jw?ogUTpZ5b z;#IpVs5*^?@?GtqS}%s&=!?vLK?r-eH=cCwiNUJ!Ag;(c{+a`(aGoGn2%R4B;0)w_{l^ zglyEMeIlKPMle-<%*|T)6_0&bz*Y}G$6M*2pt|X{%-cyFWP3LX{Nh=7lUvK_mycjO z92XP&{4VmlYYrGi>#~F$v%zVPCY>iZBpN5&qv@YdFgw3-Y+Qs84KEc3mDh?aZ}LZS zztlx^Lr+p<`K1L6HNQeeU^>4b-;`NA{)62n<3+jV252#SK3h_H57##j@{^}(;hYo0 zVDXy&h|M1fY+O8rjq_Yb*Zen+{m}JdP7A(Jv#3hgFzGhEUk2pf=7lV;V=>B2E95)! zcpCBB6W+hh_)fJ4Q&kA;z|*w5eg-rAHjJeGSdE^}qiJ*3bz<$=%Qe4k0wzC= z#p&MT#+&Vc70v@-K5hzGkuAb2iTk+gm-S$bP8+-y^0%`@QX-Y^Siyyw4pEbL!jRP8 zs494V{Yw~B9tdOZHFh+td1^BW-1e;Z#03_<>j7E?hZR`b*!z8QAG(^s8% z_1r>HW4s^hI5ksbaPtBA_p=>Ll%4o<(VM8JD>_3`u_n>&g}( zm0iTw#TYZ~3$OWm6>;dO?!wmAD?--1I`}j)ft|l(0yUrX*^6{37Q7r`{M(_RF(U|% z$V76#bZua>j|Edw*1?0VR#20X1|>m9F;VRA7r$r3K$!W#hrU4MF9!-v1-CttO|RBI>|b0f!-vz zthI+uzbcNZ`_wr71)tFVzc6moaar6|c#m8SiJ}gY!$GIM7(>rVu#wK{STrjcWlg2Q zqsksX*(Jii+rn=3!$lhA6V58u-{7zBTtE)c0a}?c2adXK=C*8VgwDZ0nA{dl5BzDw zB#j_?kt?K`mJY0J<^mEQYXBR)L)pzm1H9U_bh1oS2ivFSlAypybo-IX6f>sbVB<3) zwp5BK$i5~$Gb*_GQgUoVauc8Deuk_NX_L(XWrz{l%w>%q44!L6iAAY6Vyrh?6s}Bk zE-ye`MKzmi_b0=-`KxJ=#{~Anc_Vvgo<>bROcH&cl10wUyMT>hH}R6x3##LxPIJ$U zVbjN7W&TsN@aXDcAW}HRe?BuGZ7zf8--8&Ot?!J*Uzg#mCZ1Rio<@VfW*aiff`r(- zCVCk>uTo(OcLSG@|6WWW#r?0yi=6HJ;epevSNjqt6_ixvsT~U+4knP58?;0-*Hw}N z)f9SV;&sYO#aMw_htN5712!dy+&7)aw7-vH83m!tti%`QY6?A-&?tP`70NkDO#}VE zv*BK?@VBE*lPy18uN4&&dblHs3-Q)w0beGN*QG0nwz@Nzy7$oQdpv|QhBaIaze&_8 zbjbea8)4dqKJw{+o@lMFKkhRr$MVgOxIMfppLVf;Bo;2?KAr0(@9QlvSaJ-J^QmWJ zY9rx&uNXe}a>AsK#$=d(K9{5uN2ZQQMVqzKXx;0E3f&f{xOF7mzrhv9t=!Ies(X<= z+8fBxNlQ__LK?wE9p=w5!}!zZ$ea-^cw_&=>XOYJM7tu8EfV)&E}lj>)V&e*pRL4U zA?tALktCFJjKF=n2I-5{UF__~R643e9xfcvg|xhTw0qAYx~4Xi5lJ(AvCD+Sce_LV zpuR|MkFY;APR7FJJ9*RMm4uB?VM`oE)tzf)fnRx*U)H>YxgTu77mboo!24s4n+-i+ zpvHVxOX3to2TYDR0F8DQ5Hul+cxgZ2@@+#X)BDKZ;Tu?*P9`b~of(h9p=`d_GZd(WVYXFZ*{c zQnl5dc&5V)UnRJb*DqD*?SD3S_p0N_0A5Q~W5XNb_TE{=+~cO7K5h7g6#0ozNYhKwUP^hmJTgoV7lJj=U2N z&0EfJJ7OJ(=88vjbY1|z>7y-TiNId@y$t)NXuvnqP2jxJnXVl<9m^#fsZ7~$O1!d2 zrOqI+IPsNgWh|!WrXGgUw6l1>YZ#v`BlMiQu3%o$6|~uQ2Dbj(1_ysd6D6y2q^j@< zSIEXSNfALO-EsM@ECAe?#by7)=~MLOF^FGE4ZpJ&@G{Xk8?mXq$)e~@DX)fht$DoD770&!nGFVmv*sh+2xoSd|ZO#(B za$ImNy#7s|Uu&Wfv&>=rlqX#L_Dbw@QpUmoKQc#K8Xh+}VV&JjER>0chnmN5`O+F{ zPz&T{%pJPn_6WM9OqqG~o`9gJ0vLXN0xIo&MwdM>!$jNDr1XRnMEG?>u$dG)lrop* zrB4CvH)G(*y+*F;v=v;vv^ERDqK5rJ&ZxH>dn^@jRXe?sNJE}Z>-Vdd-I0$k3|#=Oj(xH@n(G!)z- zE^38f^Y|VqZY?9>vgcXf9xv9%;&GVFXyFmhkyH1L$V&+oEH+(^bIQN-?PnvY=A2nv z^6v|9sbD;tI<5jY?zAK}{AgUTL?4DE*360&lSL5b%N9T;?XYrDQ!=`M-uB=xb3TS>93jy?((EV z)XdD9Mk~Lf?_%GQjbDa{hHdqsyYHq_L)CD!PRhkoQZw*;z;SwhtQU2d|X(Q@fGbk(M0=IM3@rY|kQ_xd8HbbmK%sgVYq-8FO|?Jy2M_=cov zIKp~uPtaRs%qApsV)qz*ymxvX+3E3*7~k%I&Wb%adVwBu5cZbibI(%M@}Kl`Pds@a z(@G}W+d|g5mvlyODuVld7~g&d-Bkq_*{duduHGW?v~i@Y>IUI#GuctSo%myYAk04e z1?XsP)KoIY|E|VEp7dt?7Be5c3npUePBYx7yaEb!#9)fE1r2{O9&8g|R;$h01e-f2 zl5s}X@MqL*F1X_f9lkf5UjBZPEVQvB5#wH?K%mik z%>3qqqxiFY#|C|_Ovt=x4h#_Mtozva^&^dopAOzr9$EbN(2EKmij!%o#g`Gkk`Wi1`r=vCa0=50@ zM`$G)yb6FFZSu5iqdPqO^p3ncnTu^?QT18}FL=X{YUFiQ%a%UJ#oLa8yO|oBy-F4* zee6Ta$R=`zYvuCGd*GC20N4EfA(ka;px6GPbRa^+X|_~Bq_A7UkJ%XYFOFr%h-mgQ zBQh~p9pBEK%wE*VA@`#K*S@-orCbCkv?b#)D-%|tya)@mhp;LI1LkI02L?5par7Qz zT5`)4RQK)!sa+*x>bX-8GdYg*r>=$JUpZM3C=R zMQmP8RyPR_K%)x!K4K!h^f?o~9lP+PP8rE)ct-Z`&8NE$ouHpv8_B4VU9_@p8Vs3P zK$<3Q#S}P$A(=X)u46}6nd7{+Z!qG_7l@MT z#-VSz@ui3CMaJH1p#+s8(X+~@GF%}nA`=7PhGhQXBn?=-}*f}I)_juX}i4E;5> zuvfbkr3Qpq!g5vII^PJ=XRHD>&jftE!4tN28Lc4L`%#LC2|_* znnyv{YSWAV744&CQ)i*)md6oEO%{31YFk85nah`0rO^Q z(XK=aeJ^vIPpW#o6(ScV>^nUwxE&~!&zJOTY5#-j?I0f z%e)=W(c_?BR2Krs_wo{@xRsB>dlvHq+oc5^fW3 zYhQNiYzWgRR%W$_Et%}|RP?*01W(iC(Es2U(rBUu3tv^joS*L?LpmFl|GQ6o&nqEc z?M5qW5^;1)7<%mZhN^AuqR`CQbfc3MK08v$M4TeV8Loh8-&3@BwIUlIaEzXLl0wX6 zRYhlxVzqk1!RiZ5!d)>#h1ImmvA%Vg_*ri)K7HGU<8HmeV?tl@K%kJL5hY=A+e+5j zTS>paO+fBHO|bUAzWAU#^)T2iU zOV=C`i6#vf@~|pw_<@CNOY1jO*<4MIU;EAZU6N)^eX7Fyyc9}J)gq?;1WUb{nYFKQGW)l6bVZ7Az<^qz~LaJs~wm`aogzd0b_& zlQ!KqqvzI7hPi+D;_+*GP>PS}1*yYm`D-m6_q&VzrT*0++X+VO6l0aIXRDL;1=qa0@qt9ZMWAWM(=}aXf`5R$M|&|J6Vy7Q@78x^#uGsja1^dZjuRC%*MQP9U$CDyUg#xu z3Yjct+z@r1Dqc;6qhj`;6)yqj&4!EKhpqwEvmI9bE5T`LLAd&91wH+~3FG1CO%svY7r{NKSAcgUC^W9Rjfd{%K=U|p(GLeP zJhxGuIZs=GUo-Z>zz2je)`_5hpcMS|j6{PrLtv%!WpIpWhw)bv;J8;LF0tNCdv`H7 zZqSEqvbHj;7AxAh$_V)M+!B^aQK1H{YhSys%`UXL5=cSq`G~+X#Gl z(hZHq=A*3E3XtyDM24)khs@UoEP1&RNcowu?N#$JuC|IkTcN;Cyh))isNf;{k%6;2 zmyx_V0{43H7iw1|IL?(ByH`0npNhmfjT!L5dpd7*B^bXqKEwEJ z4$wIwj2I7U;;yml@UD3*=sdFng||QHKAl5Yd1(Nx?KPSBf+!f9tO0jd2s7EG$FV$K zhJE)_Wdj1wXVc(rZbXi7|4Qw{iu2iMzvd6s)z*QHnlAs#+~J4wrLZwe5@X-Sz=~WO z=yB?z^S>NM`6FZKl7byzmzRv%L)XKML?t@;@Jp!AI?8EiN073S>aclS0xmOa!c$w{ zz|6p}yvxDq5Wg&$er||@3F|-b(=^NQi&Y&=P*{M;Csg^Ut?sPZydA&wi?M+3kyWLg zK9IHA1~gx|!6~cJpej`l^6E(J4~L@=bz{CMRop#D$Lz&>aCurRZj6v*Ehbao_i4ew z{Von7#b1KPx1*Ty?;`G1W@wZgPZCW0Fn#P0wn*m$3-q6i(-kAwgiawFQgDD>s#?mn z*p5e;D|zS;`xB3>DWqh#HWqyUO#@e0^RbKGP@ST^IOdEbs1CJ(*N=YViKI@9OGlKL z_zqiYQUqXb1{bh!89cR?XU&Jo@xvcuIC5;hXyfU6$TrG>8&ga8c?J3KUxL7ZdRY#R zyDDMGOCMChja_*C8uABP)Y+ugCKA8*G-f3-q`uFrgO?=qa* zutCV_FSJc)R^&u(qj7EV3Fyx)r_YkJV1Lhdvi?;tH6EWp0)!o2KurqFSZpE9DN}PkHjFYP3MOJ|VEeFpIPp$QFU=8Wv$$^OSSzC0!zERUQaYQ|x7{P+Q?{Ok}$ z+3zFgM_97ACyugBKh5CVs|oPJfhWH$(wN0B74|IV1Dd@T$0UU<$Q|p&^eYwcZ^bxP zs_-61FS!U~6s1Ia9c!t=`S;{;evowP3mSW3<0qOou*~VCngpII-D>t&LS>WO_61 zdi{xumM^D;YsyH`H*uJ|a}7P|_=`*lkq4XQ!TkPk8T>apk8JF-psCxN$pI~Y`fIc~ z&5Lotjzn1+(KQ*$+J!r*ojfcZ7%qxA{0^qdEEdhEnZOQ>T){>K)?w)Ut!&l;H5OZ9 z&c>DY5mUq2WY&`sa&wn4EO9waf<&(T&)k>9RCOYW^)=uEgx&VMjb>=k^p~uBYD8+^ z-zDj*HgYPzi)qYrLsWL0i{Ay-`q=2x_)I>PTW2~R*K3Z&OJh%9Pi+fM)+xcD#pm(a z?n6`|EgU8KbEvp}7|l8~4AP5S;I={`zPHT<9o-yQ>f?{zLCMG*f6!O*3`>oY=vAR> zVd8w7tnydIl${$OQ0W;yXzJqf3&UXXv_{n_d=MB>0;4qW{z7)9&Hy5?>+J}d-Zn;{yh}xZ+Xytt$`$Mci%B9wJYtlM!T^#_QANdh$DKmI_Xgj%o zKbKVVIq(2xVw=Do7S(KnaZ+cA%;%xFUgQ97-6u%nk1@D0Lx(K=>kAUM_g7E)l}eN3 zH1JHSF%z#lirFjvp|Ra2I@U;EBlS&B?1&Yo>sc%gdIKn6I>+bQ^; zF5KOk0l6O{m@XGpeQ=s3ck{<$h>Z(HKHY$&6m`-s4v}+6o59_^EHtK2M58Itqu^BVnidO?jq;95?6@L#UI_- zVCUq5otpCC>bM0zCw{_!1@cgpIh?()JwRgmvT2!#Bx`$e5eHIB$=nz%<`iCt1CNB> z_ude&Np-EZj;BOt`+u0QV=@l9t)(4%g**6>x0pO8i=LJU1M}-~iJr_;ruV(d+f6MC7~mDPDswUO9uVR1A{} zn#F_*ILzIjO_tB4#4qPEsR$ZE4%Pod0~BMHPafmy*)BLj<|y$Ww~$Rfw3B_eU%|HN zpQqQ#giOn>D|FL;J~UW#2=)H`;f8$l;LCKrlF!PSxM@I@wFaKTl;>$U-697Sjg3j) zG&gW8E#f~Nxkb+l-=`b*g78dR61wzUD{suW|+%CrhP0fTOUdn zO}va_5o3#aJVE>W^4-di7 zZ=A`$z~|)FH{t!YISJHgG?rM0aCUH=R0KaH6NDv&q3L5x3HgX&8NaCCQ!UKd;Eei7 z59vgaG|VY2<8EzEA%8~X&;l(Fs-mg~X3z$QA4)Y(2x+`%4#$)sY~mGGD=%H55!M;?%aNA2myuVV0@v%p=bb%zl-@ic8-5)2_GP_a)P z5;ho;-Ph(I>WA}R9#_H?`Nf!f%Lo6iEWi@A_2i1c%U#>%MnCJB;_T8K;y6U`hK?Pi z3-8XyS2EHtLO%_rcuJB-gT~O(<8K>YtP1`Uf5RcgZPh2-tNH%-vTSK9qxuP|L~esO zsxW=f%Tb{pHm?$Ptdne8)`rp*v4eD+7?Am{!k%-`12)%gMc+@pbh)V>`I!3_8$4g& z?B_!nIr1MW>0d(QnoY3XO<=(C`6XD*prD<-m=3yd2;%NI>Gny`jS*bNUF+aqYACG} zvZV!?FL>8=54nJr5?toF1~U)y^tY!WZ@ILOSVY6~&7z0n^5Hmn&(9&nnpeb~-i>{`%K*u%^F z%d+GpageOk03-Y6qV~!(%(^y+$w!pg#3mo|NxG8{KiK3Z* z495*gQ8-=q3_2>jrDOf(pw*8i3=VjOc>xvlo%we@H9=r>-_OUUHxa@)q=^jw6eX~? z74eoTVxj@<_Gh^p-GSu8W1VWx=Nqtlu%3*)WQJLY z5qf?|K_k-?^t*QzvWj%rD5E3z-F+Wvd%uN9#wqh-$7a%DwWo;gsV9AkuYHF;CrgGP+u(2=-&rwCTby6mMw=n{C=$qo3H&MdM zs1&Q3m$IG_YgtiA9+Um(2iMM@6y=vu^_a zM&d1be=`Id(q=LV+Yj`6P%64L>wGMI|WMTbr8q&y|h+I9KTPS z3}c1P4rFV9lX1Ql)y>$)jU*$wfTBEq;m&*9`Y#i@qqzdJZ zLZ2{F9OUYU;jSTjsP(tuEaTW160jo{T-GGQh8=Y<@|Fq&)SZRd^NpzJ(ph|^YK~p4 z_elABMZtgmf^^+_Wpij^JbK$K#zz)*WOB6w8TYq{$PZZJ$Wfy}tKNVlolB!tdam$t zTMq3h+Rve_A*nZ2gQ=^hL(i%_E^skI{N=AO&nX(7E%3mzSU07J5)Nn&dYZ5tCuPv}dE>T3M7#r%g>D2_FMtB`M@ilwZW__N`R^ zlpWpn_Xp9GFXvtI9k~lealF#w8eVM30W@r?uiACC9JM!f@Oz}5(91*r(f_s>bNZgA z1*hkDc4b*D{`z|zVjtayUhfewvEn&ZO_~9v*JEL*Z5*uFU<|ho>A~#<3e@|F4@6D) zLT@_Gq5iE5j#-YgjsMp}FMaW#gB zGU{}j?=pEdaS@MDs7Aa&|8@_zU0Jk;l=}0-0pCkvTr*>A? z`=;BD=khUF_}%u1b;MgDl05dzAz?ThueqP*)!e1XnMWUSYur_!sn6%yaT5NSa=%3?+}OqTo*ScV16!Jt3;HFqd@I5Pmh(k&+!l;Kq~)9N+g8b+>0AqjfZ)&>6BG zM_1RcaUg@gcA(Yw5onhsLI+ty# z|3$z|eKjgx_5iE5%qOZZ|Epfp8^J$V8bc-=ilDdqo^rA;(#eBqO?> z@QURdzOY;qy7hX9{*ssc@OR@_$)x|?B}`CMG=ma@Q}jaAGOE7PiA?f{fXX4q@#07a z_Tcb(K4qa4T%g%B+OvxtnhJyw*5LHc)H8h!5~V(Ajz z&Z~o8ZTX4nTil~=TsT+pP?}_gh(Ys_iQL*C&aRB^8)6=bW# zV4pr~)Y4;Lj@||*@0UbNHxS;ed4Zu{Vz@oZFX>9TrMST78on646K8AOhw;XRP+~9{ zKcNNt@^CY~+fsnOA1AW&i|xtE)@07>y^t47oCP`p*TOk`9Xq!)1D^=l<{rnRkkiz` z@7LZAu`1EkeCr#Ucj+;illY9pX-eVGbXc2!*p%vuU-?km8GKy8y)e|-0KEG4$Xcd3= zA&CDPCaU@E0cOfwyo;A6>$AK;UGA))E3ak4MS(N6{dyDL*bvEY`W((w5(vmeCljrc z*D*e>gKp>%m{DD_EWvacDRJ74O9l6n##?e*f;%wE8ARI{7!j;{n7}ILO21jc`d)Y9&dBB)urN1Xi`q^Y;`*Pc}eWP*L ziO=NM+7Zxnzn5#sQDhIKV%f}}gE&Mco3;-m!{iA`^sG&j%{GfG^pEU9blbbU`fAkx zzIJlN-!G5SC}DQoZd3y{e%n}U-E<~j=8S0m04oPS;OO2Nbd6gIw!Eyy&GlMP>d;KX zmz1JHaWQLrX3fk+Va#rE0k7eW2^8j>7-l$+$3N~}MhB#G+};yL#SbnqW8dSVeZ5o|6mua_AJNf>oqu8_7XRTXoIH_qj>O@x0(tm-9=W+tg4JD_ z!hDxeJocg;uL^hTq*sqg{uX!UB;p6>`QZa{%j^Mn`GGt%zLX{5uf53iRwJA%)keSEI)k>; z!%#Z-0T%r;!X{rQ5?`hR!zONmA!_ZUd0-TNoj)5x4fNTLg<0%EM<(4nW)e!xS7swz zKH<*(kGO8%G`8sUBlJ-_j6{7Q`l;o>xs?XgUn;^h6KnQm zo;iknl*UPahl)mZyydKOg{-4WF|GE0N((i&Ll~V(rgx3EUAID;teR+x(izSAo`rHJ4lbqgbY=p&AnxeDh_$9(d9l{U4K z-;P_yjK-S|C}g`b@q*wk7=G8B4g6FB%|tO??Sd5PoH`2XnQWg+X}_e>p*{@PB$GpyJJsZtz+x6su7 z1gZq9<9-`Qa@i&a{Sr&?X5VWzL??i}9C{cIjTS+Gatf1(&la*y8hB}g9PSFbOboh0 z;rNsLwCc%NURl`reAsba=!lLXiZY{Fg8nd(Ou=QCmsJbts3;*@a{Mv3jlA=z*6d!GY%X~?jbxBhS&Zyw>N8J*^R zBJ^2msy2Jy`VW_EyoWAj>dgG@KDapaH+*o+fQB8{$gUI}^16OM$c)Ls&O;K6MCeto z8jy!rJqaB9s)Lkdg~FLF3bds>3eO9B(1>#hr1gUcYDXV~zoW*&z@24ipWuhpQCI1P zw_{nW(8&U^nQ-dw4N5YVV0e@UmYNC~00Uic)w+#;(fP*;bQ1GQiYT`%E9-CkBI!Y z4cySitz?$PTe^DVNVfD#Jle)Y!Y95P->N$^-+4!HdUqy#&5eVEim}A&V?LQMDH_(j ze@{)_MY!y`JbdYY2Y-*)!`3r-xGmrm;Zp@az`GpgwPrW$V=rm{kQtofFaqsTRpd=i z9s0hQ32%!pVk>tRN6EQUXYz)etW^WmGe&Iuy$>ksP*07ORq3J~>Ga+T+0aJvNotE^zo%@|C-R>~zr zFA;K7d)U(J7s(whJ@QvKgZ4O<(>WW4Kx(8U>sqNJ`WLy1s1&zjLt7h}(_sZG#j?@p zTOGe3S8K?Z z4239P5BlKARn*?oK;K+&AOk1*36Yw`+xah|!HVB$*~9Bt-`fcJwMW5jvnos46ph8Q z-$>|?gT(LHe{eBE2JZy#V!gfYpm8#Xj`C7xU;D*HdfsjHx=J~o+4cmkJI{k5+b)r} zdn2K-JCx#$G;;Tq6I-Z~gLfBFx;t_fI48IAt8pr3?^qAlraQArd2f`72*q!SnWQ@) zh0NF z{iOv~VVvFL#YDR3l1+VX4mm9BOYfzYLPNJbNn3ve4_iLQ^rBR>lb$T_lFdl|$8OSo z=LxSN7Ky+A2IKsu6KufhylvHJAOjlnV8r<6v|(r-U7D?k<37}Loag{ued!~=f8%4i z?ZZr>b*qT}2p^<7Z%o2#PQx+tNGUzu^`DLUo;5T_*lB$g*y3|oCheE|kMO-;d15;h z-;X~-JYqvg)ImuSH8GduPQ61^6UK8v1!54`Hvt1H)xc4h0e<~q2CF-s5!vX!+*F-v zu08Js^(=ZwLNjgYbJ1GPqAi3UYkh?{TzpP%{;MS?L$hg&=}>G@t>f37M{N#(mXH)Qbd9#t6v}6IfZ^Ls+_fq+1$#(Qa`Cop+ucu^SnE}Xj zBv$*sxJfrZtfRAqqs8mz3N-O-2id(>jVhf=tM)o5k1@j*!L2q2(l)i5=r6XV!*}xh z)5WFM!-gu6#r=82!$@GZwxn{YJMyTc^>YrFy`u>N>%&yfjx~@q))8lxNjcC z-edf{&S>7lKp9WUR?vf|wTbIHC)zsK7Bko6(#q9K1^<-;DII;1PSATq#~1nH z_wJdfT>p@4bD2(~UVNqz{iBJ5_j18Ip$j$h_EN^o$+~H?;Ax}-WVPjy16Phve=5tG zelLXE=k~z6>`0g{^8m+bjAJ?G5;$7*7imiSM1F7HjA!KqEM&NXsBhPA+kl7OsJPFA zTNLDkMXjeOFU-hKycc0}ZUZ-Ex+QLK)}#HuPI1vYRKfaA4&J}@e-xdEJ67)>$0HeK zkBZ2ONVXKt{dt6hNTeyGR8-QUZyU)T2_d^8MVp9oe@;m<5-lYuC6xxHE#-H9f5G!y z*Lkkz-1mFD-dXT!ii4ni=6wj%vWAa2CYW<3fIR!%h1XuHovH^Uxw++LEiWC&A?mfChVrL4yquvmRQT( z=AYj~o_#YB)`TpB&Y68=|Ii4tyWtU?y)px>$1Fwbfil$2P@;Tx8m7I`fIn+pF@9Yw zs-*C{d%@{88#r93$cRVPD!4!>g3SALl2{G%n}FU2 zSr4Ai6u{?@rk?JhTlkDsmrFQ1Z?p>dP4n+-?$_4 z68?x^i_OuGQ1{j)Jn&kZ-QLho=keT(z?S!5DYye`cUud#+>eAFIXg1#vNG#7*_HcN z`Ws2tPin#MIJE8l#0)1VL4^1znzKzE_Ds}gS{zlFL*XCEi%oW*e6yTvKa`6hl;>^^ zo3bXh%V~_CH)p#h3}gd$!FSV#WLfq?w3wy1PnE z+QG*6m{DPIF`WN86Q4QqnOxbi7*Q|>4Su|<`4(}N$T$`g^Etf8|4o7$>(Yf8OGdDJ zMKG%%*@<^8k8o)#Utr~b8@YH7D{i~RHSW}yDco(pNPig>>qw+-ZRc?0ODm~~UdX=#E~1ZRE84VVW6rotBE%ZfhndZc!JbxP z^iYv#eVYjt%XpUkkMVTO(x=3M=(fJVk+z4CBSiE*hM)0FgMzGf_0X8OPqoTrnvM*`@J@IE2niP6j zzj*f^Kga$-_i>iogs;9_&~+8M+5Q+9iq3-c#BiLH`HS({QHg;+eaNdf@i09%3EFbR z;I7;~oG?;{D_{ub-fhJjEt9ziSVP`@ae^VIR!F~Bk46D!@Q33I*ds9o_Q5$)+$4$Z z!@c~rd>-`;(1cxJ3b9_g%-275X#cYc>TDN)lR`iFB>NWQP5p79?j^#yov}Gz^odyP zU5b+XWUzN@Bw9{gMyEDCK=&;VP$WN`R5boz{yeqi-d~AlE)0xful{k~Gq?uzKb)}Y zmpeOc#SD1nu>iJ3og*j3Ot@*a_2lO!XVzNDp4$0Op{Y-lky`YAnvR3A=OqyVl_8+2aKdzFYyTmcaV&~4_ zvA^$)^?!At`QZ54cmoS81!aMB?SEXK%W>|xNl}f%8f8?EIYb|iQ-h=Ch$HVMxc2-K zZei;sJmZ>5SH6fvg|zeZOHdRJa#iS&HkRy}&igAhd2GmFElIlGOFm0pWT(45uho6^jwTRRYzs2C{<0NEm2vW< zQZ75fliAkg4HI8CfFuqa)$O^+y%O_OPH#$ z+tBS?GVy$=%-ZP|kj0L!%(}&VU(#d}wVka+Casmi(-pd$y@VR0c1e}oJ7PdzZ7j#t z7k`qpgE6q~-)yX@PXQ;VVJJDpGjzP=aetm99;^$;KX3b3;ZJKJWO5@eOA5w%9X=zu z#*OnjorWYtLa<-@8!jme#!rq{ansWY+?ikQaAU?pSST_PS6vmSG}#FL3`fJc_lj^^ z%aT;_p3oMhP&y$~md|nMVUJ@lb6);2@$|Bz#t$FYPWHcz6~6n)h?5=Nt^b9}%kWI@ z__CU+FCn;2ww!#O-B^2m>Sf~FX2BgBHARQbVjwraj^5QDVJ`mAhOfW+$%>zW&@B`V z_4a&Db9fbcO}z$V^*6$AHh?_zOl7I1DR0^Q>k$K}*7Ul(7!O^3P9Djy@rW z+ZaPyqAxyKVH^BVh8V%<}PT9=2JV?UF) zdd(9!Q|vtWW?Uuz-_9+V?2MV?9R;_mb4ao6JZ@3ME#~-F1+vp(FYOas%|(8AjfcWC zK#G>5aJ~RvKTO1=Bag@iDHpJN8_(WO?Wf5edDPO-5gRjG==b`qbmNpmjK6*`u4vhb z(`1(8t;7-f_1r`3N?nEH?{4FIxd^OP`~a?9(%d5tKhA&RC8lqh2JwF?43n!Z3DM|5 zeIq4&HI@Mp)zfUo=qvi5w}L5D59K0kw6S-&A{RIm$ws|9M&d`aNz}J|=)D%fE?C=$ z8SxTmUZ@Ib(apfUQz8Efc^1e-T@pSk7!LF+z}(-jadd+}n(3Nip*o+j`*8~g`Mcrh z)p}Y#xrlV{o58JG*}*Ap@xtDcCEQzHr+qnIg+%T>fCg^4IH384d@6WH$~t)O<@Hq{ zr}&-diSBS>EuRJE{e~sWkKvt**YSz>eZ2lS0%Uk^`|@v9>{*wOcc!l-<;I(kJ<(0I zG_1K}77Xj9odOeX&m~X3rh)8SSCnH#7`DxZdwSpmy6is+=_1M8^$$Bal^<=e(NCS* z`SlJ)Ms2}8ffMkT?F=F+yoepV;y_max<)6r`_U)1uIz%20#<3Q1IhUDg=`daqB%ED zGu)b3W`FN}I(1qI?Cdv(S$mFP&<$-giXCR(Y_h@~>MZoh9_0BDOF6mj3L^IQI(~D{ zr_oKH0N-8$p&!3+-0TC~=cv{6FPjTD=e?uvl>XAV!!a=7aS#Z2M)QTS->_%KfX%1l z>zFTgQG|^=gVGzbA@SS<&{&jAy~QFy^XEb;B>V|Yw#>u-?B%e|Sp=2`hU3-}2Yl(e znKP=oN~1UaVplS)?DKVJ(E2NXR<-}mYk8~TTPW`<^UrYMr6VvnO1RwS5Li5Q1{_Ub zu%e=reWvZi{5PkTefumJl^^{ez9+Ki;g;<%=C%ZMv0p=NYhK z?;G;!-WHtnPmWHD&LsB7ZRmAGYs8CUUp4Y%u(6gNx!FS+yBA3QEZf*qf!=@8b&gPP-DQ|DZeh+PHy zOl`o><^lujli+YlD3NZ^xNiGxW8f z#{>yMD}D4nsDJHGiOwVJ&KI;nTt~?@t5#z2;87 z9$C!!9leLCs%7|0vK&Vy8DU34Cy6Y+38ousKy+yz7Ke#K_Z?4ea9tWV-)aihxc5*? z_ePxg;05nbFUHX2=itU*DO$Lv;D=>KAn}aP9lSIJ9nWAkgKUJP9z|&ME62O*wBXSt zKa_QTMGp-BCYz(S^V&o>`+27ae!g{=yxpaMQ!?A3kmvH1Z!<$lV?RhQV(BBHCv>f( zJ^56(g!%g}4IbOe2(pa%j>Qn44|}AI;qE=`E#p#h;J<0)%9beTSk8MzMIS(CaS0r> zItq~$DvLJOPQOnUw#P(X7+L_}TtZV)?;^TQUacJLc)~YC+jI2Dx zr0s4Xp{IV+TIo2tdQcJgB`(g-cq|wF>>#IgGZ&leuhN$f-eI7x1-B&jIDVM59Dn3n zF^ew`(d*a5K&XKCS0~p)uI4frnG!*74Aipkv$k@FzSwYL9g^I$<+a50!)89l-Z+}AOrw@{f zu>ws0EKIKJKf#+8NZhWE$1P~b9xl9ypN(IT-XWgpsaFn)@maifH3$37CqetsaRP0P zOJsX+3a!uVrt#@bsJHnY4uyDeYb365eFrr$FWZ@$wE7Zxcjy<4vECpsc~ycp(`U0@ zQ``70ZfC~QXn;^_87`};ke#Zrl9vB+gu*$ww7R#B>}EeRasK*{l+Tb-aZk`)<4wA+ zi3rMG^Sp4oMmkfh2G8zhxN#L>wMzB|T=Kzc{Cem#>NZJGw!st_O9!$mPne>fG}K`@u(98OqlNz)NvA(CKW$L1`(Rq{=fa zU&b&^50l~H;y8Lc>JYQ@R~fz0d7q58ItZfoB)HdE3fwi1DX605gQNG#uvx_%E1Ej6 z$vGb-n=UcmF3*Oj*!QH$wU#!yhvB8yuH0*x`CR734En`T3M>wf5p4YU0;WxKBtur& zP`H0KJct>kE44o2H>(5O(r9&9d0CTFYjGq-AChTsizsKB>j@IZ9Bdkng^Gf3jFL*h z%rV!<#%~(%cJKu`)%X_^&MDxP-8J;Qni+P*FN3eoPhyE?JQlh8fXhEuo{!SccdgIT z{T_MbK};RC`18MWnMJs6p8*^>TMzaBOlXzKYPe_q5WgjpGurY0m0F`qf}?!5L3Y@b8v9w10UXEMqO$1vwYs=p7Tv%$*0<`=n5{Bpi25 z3q#XYLuAjD#bEChM#jKZn&6lX8?fR0S2aF()Xe9ey2y^z3f1EbI!WGPeY*Jt20;lR#Ks9D5{j+KX zJ@hsaEw^Q1mu(8Z{KxYb_`JhoIWe*sMZs%~7+iRH1tuk)uyHEM#!>k&PVZX_+67r- z0?&~!y2mmh?drHND2FWj&j=lJi3>3%(={uU1>0hEdOOqw>h`)ch+OH+TD!g z1!vL)p{A@|Q#EZ&ETM9xQ}8LT+kEy^rW(3)sPLmmEUmOg>4fcc?t}tVRa=MUURs<4 z=R+T-a`eHBcwD8UN4Kt#B$-1Gs9GnVIZC=iokTuS_MR7B|EmP{lJhWD<|Oethcwt# z3ye-{pm}pDQ(Wi;q%4~{z1+kFq|U}KNu%Uzzb>+g#WdLOK-?b-&LhSW zzv(Q%{y|SP7R|ydXL4}+ff_WuaSTPT2as_ObII<`G_>kdMoC*y!FR_A;P~w$&z2SC z&WQ?>scz$8%C>LdY{utrFQ3H==j8D5#!2-2uP%CPBJal>6z81y?s?gmxgb^cgwN`h zz_75a;87p}k+2LHt8f`)zw2;{Zu$6}_b*z-m94Hce?uCe2&&vKy z+!2tvst?wsyg6gmK}^g3pkB^yu4qyL@y{QjS8o!uns*vw*Xg4CiAEamQUeB}##6@& zX(TH&3ilkpL%-{`kk{+^O^Ac~&|ITJ(?Asmq^{5jd0*)3VqtQrRtXJh9sB&`6|$N4 z$7lW;%T0KF2E0{%lb>#}PiFeeRvZqkA?FCb86yx`|I zTS##p&&_IhO;3wtLrBhX;#}~Z_+C$d-tF~Jrdo}M=k;R3*&whDct&r{-Nht*kA^Mt z4Pd|h1n{$R0=E;s_$l-sB^UbXj78%hl=tJ0_^9F0Aw`mNTa8;TBS(sM>!6R-G@#z{{?5|J+guHADU_dSb&C;9sY33NGFBu8S*m9f^R+at)}Xeb^$TSsrW6j8aS zm+_(DHM;K0UEINUFH*|+Oty3eUGqzV=RX~Sco9`DyX8BIU;N2z9plXBpX-Q_ND%H0 z?L!f_CH$^@3#gyv^UW8UVDXhU;vTS+M7YVKrK}O$Yd8vTl+#FRp)W@Lt0gPb?~s;< z?-}*YkwpDhFYa~>X9hw}!_k9Pv_pvBx-?MXmRmfanXgWekAT3VG06T1l5J zc|BtXRxK0+u&5!J%}^mH%{^e(Dtzf z?rnYvZyKfG_pGN(<$MKvd3+}C)1QD6e=s6h~d`x%uoaCBp zXVZWHetUJJIhh^WOShTxeUeX^7$%gAV*iSvbK!E_9TULW7YXCKc}f`kDWP`Eb~{== zUznqo>S#mFP|bZKOso~a*z|bV(U(Mh+@JBkXBnb>^D}+3LzW9&{sP|IvgLI57=q^5 zx9rCJJdh<)u&nhUs4RL3elJY9eFk%Ay!KI8m2O41t2yK4gdNa!aX;NH8w(qbP2uV` zXE8h_ohiT3N}I#{==9aXD8zS%guXUlLhp3!l=;9(ZA``_!z-jGSDR~`GzY_+2+gV$ zhYLQSb#Tlk}2029DDRTExE&IijIg0Oml^?{mnue zFvbx14l^`;<+=Up>$xpz3S9ZP76^~!vnqEqcz*R{##qUX9&&j?HGUk%zGJI!isl94 zK4%(4r;Fi#x|gX=NgG<;G~gVa%;DJGc9M2+JX-Mi_8ITBP$WbNd{!h=8BH&m`SbzC zYXy+o`=8LsgIf5Q-(oL%)qpmJxr6he~`XJQIH_^0|Zj zDcHAo6V3%ih%hf;L$CMY`ePT_OBh)@XTSw)>MiLF)d%n`?hGk!Jpj%DQ*rx_r6Ae) zk$5-tLSIb|u{FFz)1w$jGs|NItRt;@lVgMVT6k!~VwUXUwW~WN*ichOcPOkTo)d2q zhwF0uIdv|x=lYu3hnqWCeRW0ndqa$>rzFBVOJCUcWYlI|p9~gB{$@N~6>x@*7yEuq z3~7@bB(vQs=$VM$I6d4K&vx<8YpVmd{G>GJG=xlc;3`NzD1u_jziFv|X{}f2JP?!^ zW2eDpNIps=oW}aRqDQouS^VR>t<%Wwu1`0?Gv{t1LJ78T5@)E5NrzDy&+m9X&hFz>HK@q2rHGTe>fgh(xHqKWunAb#dO_Fx$?(SMHEr4# zh_Q)zHJOPgN!aT=%=&K?`6E%nMmOZclN(=%|JMl+P{+%V{KlWs`g_#O`3Ck!zrqU= zN4d>%&3GbE1!E1f_|DW5Dzo)Ct+_l()zYo;G%UidUv;!@j}^=-_ki%K1)w;8EDWBD zBNg0c7}7|Bv?Bg{Zn34xB9ZWf9>6r!K2mf00*o13MN<4^iSG4LX1ac3(v-0u~}=7H(NCioK@c~b`ezCDh&Umm8LgiN_{7ABlm z?OEJe_l2=2P(eFSE0iDjNX|sJl0CA0B!13(LE@%Iu;#q6U|XCV`8hoReL}*qquP*l z?L3d;72G&kUz2MK6mHWGpDV#iWEPk{EhVqUZ3FGN-AsPkXU131PrfzWCF|F`p>N|6 z>yoGAQQcq+o;nvD?Jv@_aUhr&`Uh0?G?<>7A4$F4SV5PZFb@A3M;oT_J$@%moF5yF zj|wl+bI!(??miRm&6L8tU1gXz<|7pphuIYQsDqsPQKrMlj2z}Sd#V!bvEzOTSvHgJ zXNrF&n)?=ziIz5O&pHu2nAMJ6qa*a9=5gw}TnZPJ%*5~$5$v<{CulNKPI|lT@OeC; znHHhwxYz@)4W$sDslwbDdo}LZBRNhrFPq$++fDxF#IvH+eqg%)80w3Jz$cRojEdff z8SD5_s3ps}!_VL1@k8~PS<+4`Oz*S%3(k_*`R6fE(iu<5h{Lp!NJ^4tam4f*b9`+* zacwrhYQ1-~GVdKfC+><8%^xx1)dqe)tr+$M1!85LqQKwoAL*O<4Ho%i(i=aX(&ABB z2%1}fTxS;Tt9U~%nzpemjrVbf-EwqGR)VevFR$?~k*4Xoj!=?(pw4x$TE- zyhF~Cqs0fHd7~DXg?dBrbXm?%`#t9F9iqp}MXqo78plnYSHK;Zu^+>?RpLq~4OqWA zoLPF>kUOl>fzovc(eCvD+&gnRIaBLRTA><~U!TV6fqG(GA3`T7&VjU%8FXaA6hU5s zw4mqGTJX5Go=CfiLPJwH*k73mjzMv3?{gdSXxe8Q>2-qYK3h+ZN@(F~JySfgAb|`l z%A(3r7s-F?t2ocdbk6I{9q!V@N4&l|8^8VZ124r#WYQ@!X5a6fWY3HFbca$bs6CT~ z|DXPwxdlUwlTx?@~&@LagR@+8Dr#)5FyE8Wue$Y z@&fuWZ|D=7vm%Aw=$p^>nhFr^n1E95I6>kIZ{!*@85drA+Pp86TFkJ4DXVgA%s%80 zg+>D)@841{OMP5>Ql5V&|6&LHXM@4DyX=a>99*aMo_33$qZ_IZf?7GR+4>LTlslI= zx;lZoYX6*A_)ers**vSHHUsu}@SJB2{$tfNF}0)z{|FP#^N$RYOgmJW>dC%bS`N~y z%c!3vf&L2~kn-jZ`B7HH&fFM?+5LC0tbY`jy*o*|;@lwGHVw9~o=ML>J;x@_FyS@I z#qilAijfyMl8>8j;Id`M(0jHXq~5zm4&}O{{5^i#T5XUdjx|AP_`puTwuJoFl!2P; zrQD%;r#S~I#qI7$f&H~x@GhSLTvU{Tq2`}a?y{ml_<*G#*tQ6t8>}IJjl{rfxf!hQ zGlhb~qTqW&jQcespKknkmDKt3S+1rGDy^`eUikW!oL}inrhHt_jX12rO^1b$dtO|t z(UHMjecg@a2`tEe@aLj7?Bw&~FL}RP1#%U3L0MQ0+2E@vs51e z6G(%b34AHng>kp%f`i;vl z_QR9vo3vsFpL3RZ2J+gypq@Ae14$J3>gPG`KmL5Q*Z39sO8@5W#je4je`Yv*A{0|5 zZzjhK{t=l|rtoX*I5;mDrZ0|-=RBrP$4M$@Kyht5*gsT&y2DAF>~}xTcHub)ldNQh zoK}*-SKHA~a~XH@sU~-p=-}J#2E0?H0_!i`r_)rI;Un1?syaRbYWJn%($ujuCw4Mw z{+tQuC(gxh^hUd!MD&DznAO{ZW}BAbVTEQ={g>bU^4*3bu`5CSmp+KJCUX0O#&ADU zl1PzK5(<&~WcT!);IAbL|7|sf>IzFrVpejQLdtk8>@wua36llt@~Bs=h|@0gQP(sH zK>+g>dfac+o_>%BLelNP^p#`GpdDwwQH9$W`;L}gn~vW`RbcrJb(lC#8r4$YQ!5>aB?C0McNxcOmzbH zOHl(KE^WarsZn%lZxJ!hSquX@3bc7}4%OneV8+x2@@{Y*-mlcav=s#3y7M8JGq96aiBAj*?f1>JToSjC@Lde?{1!qb_x zwtpxIf-u@5FU3ABm8FvFCUduAs%zOxnq>CP*VIyWA{>P2(AvF2?gX8P_5UHj;f1wdW5x zo)7_Y=lkfneS8OWe92PM%j1Z2PH1m7IrZ!>wb~`egzz%CAVm>>Rv$aS_+!v=D7hXJNqi3=ki% zCt0&Zv01y0i?NSKNKk}S!yQB>ewdm*NW&?0o5{fownY9@I=@9Xk(^8QhX(&YtaiK{ ztQdIA+}d`IENT76R?Hth6@ubfZhiG@uA^bgk zGLBP!O{%$ISiN$9v@YC$f10bvtd6fF-QN^!&Lpt~6HORXaT$2;Yl(V)Z1KzVE;97+ zJT@rc-sHT*q#nsO!HOuJCCywMoO?$Tz{Jr>ad z&v3HQ;WD$pxr@HB9-;z$VeVCcK8`+n%zC$G(!T@Bn0;0oo#ZshDTR1;rELtecJu+c zxoR>dg=V9Enm<|X+DZ)jH{g0BQBZ4e;u6FzAmghBALmx1%h+Q$`_^IH|E7rE{K8=Y z1b~s)6*{AgXSU1;VUHC!kQMct={q%3)Mi|1?l6CMKC*!vdfvk(FSVy`j}}4Q4g+TC zxtBETmLj)Hl2ZTlOsJ00aZBjz)aZ7IG|O4cXi9c z+P78Yp72Yk#tkr@mAerokk*e0bE@llOhoqS^YbMc<(<-p4{UQxs=6sFsKw_ioWUMEPluFw%|-JruZ1hUbv)&4W`56MzWvsHos1N!oObo?YsW#Z@Xe z-`ctsOLLjm*%KKn(=`5GlT5!jPK2dnLrGAt7n{-R!mae5K*HmbKqo}f`rp!8CbD1* zv`aN2ywTxotKL#yzNe!#e3Q;p6Tu5{yf=T4V~T{sQU2IvFxHrgA}uFLM*4eve{&*P zr2m#|l9)#wce;|=Bb7vCr77+%Tg-_nC~=$l%<|;(QEWg|8aciH0x7X+W(4PLAtz=h zDtGmeQ|m@Zr1D(4HgpFHJANlF8I`DcqL3!#&7sZ{C83GuwFg<0GGaD-rg8$GWBua; zVXLFL?dEx$PTzEH_z6dj*DYozS;pW3&8IM-$ca2(eVp1Vh@rKd8|DwK2h~?c@tX8z z>ZZ^^hnnM=7mYRKD%U_7!n5H)s~={%|A%|iB_VQ`fIc2t%U%_BL*1(9So7JNYfYLL-APB8tM7{1_Hp#DK+=v@1eW+;lWtN6@( zuU942W){I*qnC7STu$w@s^k2o^hs*tY=*yg>(S$nh8WE<3GQFOZyHh-#k#WvL@je2 zj_;O0JKY&vx%d!u?oS}4Yt4`!8bsZP0y1s#1X5*p6RqAyvy1(OxGA4Y$GfpqtbwvUp(27ZUM%Dc^_fgazL>W7q67 zTrePj_H)8)=?*&-Iz5*OpS70;%io}5w&>73;w#~~k^xD*rwrB4vS`E2ZEX63K00pk zC-%?21j25bgh%cf<397r>@kr-d^`R;O7GAkx8M}*7^ozNHFV(Tv@G~oYftZt3n#h> zM@XC_qAR~IJ1zYeDm5L#uN`h6`6jh?blC#TymOzi@Q$KKycrU#Y>&4F13;&43{0BD zbLFp`L}TNN>?GxR*cFq+%swIsH};Qbe45>{`GYm#t{K!uZ1AAX!n1%X8o`M^7y9)b zL(Z?zWm<)Yq2atG)}0Cj%NeW(jtXFVlDPh+_aGP|L@qzD`Z(Oj&c92)Q6 z<~^scjP9M^crxz;dZ+vcA)a>`C3l_^p&gHfE|tV&b`b>SD5I_VGJJ9@jXJu^p+g4P z-1RXBAqNL$vBFgBXbB<(qAwxccP3~3(}t*YCKK(lJ9sEZ7;BbKW4`lU1MdhAkji9P z{WMt&I9Nn(_?$*roxAynw*O-hpC!UdFjSw(u&46uOD$||CKc}U78pezla1l#>P^cWd{^S(S;rcMMB4rNkw7!YN%M3H`4rBFc1+Jj7 z88eSL5SOl(WC@uKwR`$7o1aixv|=Z26LaQp%-i>L+p=VvgkLkM>UL$u;PMm3abqF7 zJhOtXlM#ghn;mpY$yponP#TqIN>fh-OGf^aGjmtno;iEefOhj6ChLw(C+#y=Kv#+s zhMwWOzq|OZZG=Lxz>OT>ctP*~QQ+|H??VpApYIl)(vAZ?Ie`5f6WPg^O;cGk(s) zwMy+%P^L!;Ht40XPkp4=r7pnl71Yr1ghi0r^p5Hrbm8vw)zJx8lF8qsYC05ukMal- zPWPx8cV$5`r+%}QD1E&`8y9>fd0GTByOUY-H8N8qa6AY=c-8EwKS(rQGP>4=@lKE{HbAe!e6_&i?_ zoQwGaaWCYUT(NuP?BkO_)~V1lQ4%0=>NT0zKOS3W?%}RWt>rTIeZ>5c$()G8d{{5B z4aJlKFnq>GB4aB6v4R!|yn2f|b?jwC6OHkD!G9Pd;=wK3^bfVJYjMJn@nqOb3+hMb zl6kc(*bojb_2$D!^BJ)5Y9S_G0Sj8sa+xR0&}C#k`E6?s%SJaq%3EGM+3bts zZpGri_bHr?OFd^7n~E9g!rZ{FGh|c8XWW$>z!jePgU|DJ)2nh3;1T<|cAxP_W{h|+ ztr5M+?oGCUNw(SH~d>2#VYzK`nlP}=11jkNp8;I(#Rwr}=o zD3^R|Y*%9)9l)pYf za&!1Te0C(`O}9eY4>P9YzZi7-X^uNr#6Vd<1X=d%GablVgd=5F{B-1 zb?KO&rR4d@S}Zab2bnuvB+{su^}Q6xdB^dry<_KaIAT1odgtlH6MSzha5<9|Ilu-o z*Z2(016DU>6A5+grZ)fg2TT@{FRU#oJQxdcZHMToWA1p?+n*eKoM!W4Sqr9u4NGT>eiQDRdvzxv5;?k05TE za~wE&#G&{72?W+g(!YI2aqw{k8*jdvIFvTfQC%Sy`MgQhy^A||90iFFh3+vIR|<_E*t&1(}$pNbAm@!M1@ zBbs)R6W%j zU04?kQdEXG-^cW+OD=s9%<}?`G+<-rVj3ejj28o0I-zwfN{iPq``qVa$>=Y7cz+$8 z0cUB{kQW`(c#dp%!@89cP82}4G^Nw?4|Hf6O0l!W#}RmDTl(R@QU{*%NPrUevre3{t$qH>O?r^}`7!AQK zn|!G1KFA%Ma+)Z&trGMlT%yfxSFkV*Z5BAG(5Y@IqEKCh{Z(n9dQ zdYNu*)gi9g6LIFcYCK;GoJ>MIs=U*JWo{W{AjBT_OCGlgyl;g5^)ZmY!yEH-FTqE? z=T+(0Nm6J&6SMO_{P5x`bN`q(O%aKq3;0a~V}&^GHa?+cQ%g{E zvnJOOZjF1EMnUe^<0N1$C0Q=923VcG}CEP}JLw{}?a|}|ox0Cy3Uu-BKic+RjGlfpV4GwMiS>~t52=@oE-B{ zmHRz^Dktxi!imfe;tFP@vo~k+{1OrVtaOFXxVHZzsc8dz-m(b}xSxT*^*p=1SD1YK z)xiE-T}q1o3t|uQJbTwY7vP*L!kEHAvP-?6qLegh+&N7p%2J73MHJO7G~k(MNhm)* zk5s>~gQ{M2PX6XwJfD$>i7R@UW0HK2wI&`JZxcSZb`fS(P8Lit*5}U0%!dlS6nb)^ zC5RumOlB-j#_RC}Iz4JYrEEMX72W}jT{X~|mI9jxE8)nF6YwlG9bQcLp_Tvto@XpT zNKc0_ji=}uBYA3lznR{Cs)JrPq9J0b7T1ux8YeYmVu@@nHz)TR*6RKyJ4A0%!>^9u z{X`gcJ&}N_vT&@*3d9Hfo8bAAPI8FP#n@lU;P$(|#H4$#aka4>WF07{T5EZ{^M4YY z@tR;%<@1%+T9W*k^dxOKqEA`z9Ols)FCsKql|1c!Kz}~(!N;%{^>GZ>Y2d&$NrrLG zCW_pPpZ}2q&kAvGiY8s(B@e>c`x#CDLOLk)gsn}oC2RaY;M9FmG}!qrJCrAa!zJdN zxs(~_WLbbOckvqtO(t+LuB!GR8sg3KznBR=GGKmL3+7ha()`LK+IK=6gRQo6w>BK7 zPN{C>aIAp*JuM3b;&#vz6Hfm<%fu_Q_Tq65VRYH~gyf%IZd0jriA;(6N}J~}=y277 zTJdb-gv1&;^y57~=JV25{Ngd$l%xMt6Ct7PEXdCg$HqC0ychqQ6dpdoj`E$l_ixWo z@65Nf;^uo)<@20}^M}x|u#oK4-^}<#21DliDd?Zxi%jbTy0CE#R!(+j1^GVUUM~TQ zXG!43=ViD?LzkPs)sqx&naF;~D&g~TX81Ne8#ZoUg;Ld#;OKH3RG+26pNIsQSDHt} zl3eL!PbnBle?UUQZOPONo>X|QAN@8MfXkGss8ZSv$SLJn)?3T4v-Jh;JGzFNIEKG3s(C zi+=GnBfnep(DVsIZb;4r+liv|+>IuxeM%2EX81wLB}G#By&6BSj)K*B>u}MtohX}l z6fQ~1;L^E@=xeZC9H><_7;Hr3!@fbJ#>r+bqeUMKldVh%PU8T6mSJ0|3-6-c>e^7@@9 z)r&j~zYd&7&5ZFp)8Yvft1rO$f6T~ThbiElcbBeAQeqpATqn1h4wKlcI#_bJ(B|uc zEILy*y!K?c1+!?HBuu}VNz8@faVV?-?ptcYX1h|lKurl!Qa4#wjTMK`Bc5eKv)O4fc;>`u5}u?0`d4*`)ieWg zY5Q0D+0}rgSk^)PB!H1G<57I`c1%r}ipopw(^W5zz=dlv=#it$jclE6?V@PK#%tN3 zXA>#{z*86lNKp^U-1l{TKJT|H8g|EnhLYeS8hDA0 zPl|}Y=Um*oX&ek6s>daN`AG#BD|vLG1mBgxm$=oz*duD_)Rv3RI z-$Z3p+-WixN{)p!S(oUXH(r z)~^i<&rc+;d!Nv)N5?XmkQJ=LHbyetdn8y!qBjK0i=_?>}4k53lqfn|Fl3X=epr z%tZ_es>GZU85nWqI=H;w$Q>9}2=eE4zCwhuJ7wU^xVI0fNiyWqaRJUo&*L+WeH;K4Qp8nVI!{~AssBZNF}kKGa&7N1~i zy8bBqXKYS=rAw)NWj2HbX<^-0MV8#1Niqht`E9)m0a7e5*>V_6Y|SLb=8kZxKn+}a z!mSrCeoj)g7XzKE3Lh%o(y^d|2X+Z;g@emz*0OC7cgh%tpSlbCs}tz-%xD}|Xn{Lw zcB8*a0=f;|LY(dl2iFTn$PUj7F#1s(9dq|2=@PG^1}d^F<4Qa`H8BpmTbqd3(RqB` z*V7}y>2byxiGAL*%~xPk!bhXhWaG#t5x~sQ9>AfxVX>bhjynB@9O2?3;iDKlq zz!peY`4O7r|Ij1B=g6My+u--?b#;jJSQH8TukV*5@RxWmaZ;Yl$ECQ?_mzIoaf8tZ z|I+aA?7f({MvRHQU&ua)$I;U(6EVtdH_p#oL-+nLgP`Ss@ZW&2M+#j4y$b~wTazmM z>Dte&+m;N{F?qPHG>=?{;yh)!beB#`WO~z)A?{t~vA`Fsq7EYz9*~8CxoC@@jj-Ke+B;uytU|6cgdNCY#QFy$6vGwhVK^J z$V$UaxaZt^vM6#fZt1hco2#;6L7^<6N$0R~YZF}kDrEeFhLTzZ5Bw1s&xiUJ!qE$M zIAvZW*IY3P-9AK;h1fvi9vYE}>hs~-E;)g7I}vnrjY-d~0z8vz#3Dnw;pi+CJZ)u) z>f?nz$7U&X(-{JBKljjkwii)p)_I&O_|+X^*24RuE+9jX!}&4|_^fOOSGhF0p1k1d zRv#n5cT(}v<7re&w1S+u8bnqEjDUZUPl%%bFIq#=$*ld$$O*4V*s%2jsgI1{Zq3ri zLo}6o)vL1Qtxw1_`B?g-tplqa595&+r?^*R+DX^AAKuO(_RCDcz1};K-N@p^4&KB14`L#!bP7_wJQn_s3f{`8rPBgV154Zj1A!Da zZSjYA&G-BpzhA^~*RJVe=iRBp&xveN?-CRr+=Z%+?$mjWJ>AuLnR=dZh02;UFidto zI!lkkUqKd(8$3!=pZy`79^VL297TsmwcDDm+{nk)#$ai{Vvv302FdlS!Q|f#Otp#^ z&Mje_<-2!~KHNt1(>NO&ex`%C?lH7Bp3kisaTZV8j$j$9Y+&fVLvUbLA4%S_047gX z#+FhyoLiGj_P#N~=Q*dz64QMsdyTJdpZJu&-g$(z_)0O0qDcPI!TCIBnPTo)J$`zK zExgD$3!gv72~}uC)~qg$ulq-X)3aPOn59Pw!!MFqXKv9k70FawU7gz3UWYug4IT#= zkpsEmxG_|UDO5|b#OfsU^UoslBR0@ob7#Pe^{4fwob zE*4$-iL!So^v}}8_eEvcxQ@^xV~im0R~jF4JQnZij$s)qWSHpLJ2Z^6Wa<@t^p({W zNNvi2T!BBh3ATdlyj8s1lozxtTp2_?7Q9*JDtJB11-H+9jO!DHyRYY8obcg1^`7fU zN_Ae~=n5sFCz^^S5(aFhu@trMRfM_;LgFY|j}MqT0(6%KBi$B*fAYMLTjx&7w$J6= z-;~f9xzpGWg-@iPJ3u6c`(eIhGi~*~fFqYu_#QMA_~}Ma-TxaK9=@Yv3P#}M`asa- z_rfJ_ad`Jc4>vEnj3GOLs_pH>J(K36r0Z$4+SNjcMkMO4jzZ(Kp{&tM4*e$00u#S@ zNEK$7#__@#ChrK<(wE~N1u9^Axv*zx8HWyO1tef`Ixf8;jVlG8{k5-J>`O!o9&ftC zeV(d8)0|aBPI_?k+1sUA%C{DD~!N2~1S-60qQN~kh{oJiDE@3XcE7$vo$m?m z>+!p=>%?qqNEpj}cAE=)pjx!Z6VBy)1U5z2;pBT!`29i`@@sfJ{x!c^Vf`_fq}qyo z0{0_m#&;@nvk1Mz9nnEI1-C^1qA!ot(I!(d)c7(PU2_xQ`NnauG;$Q&*!-3BoPJ4& zn;RtEz79UG9@2|<#o$ey7S5Tn2{v4t#BXaEia(_+@S58f%=)+(2Njo*^3^7|!$O-B zg#l*EtVX8~yGhoKTIv?54r1=lq~v*6O&jZ!L&o z787343u|Sj!%VHc#O9R=Hnu48m&e{lz56TC#Uv6+m#%@v`FeQeRtXN2zk#Pj9ev;o zJL2=48{J+^?&@A5QWL!4(Dr7my>^POS+oEhY&P3ol9QxurUj7tdJTPdZ#XHNWJ}hr z$R?HTNAS_ogG@J16J9q7F2l(aMGH(q`7wVV;JvP)BBOH%Ic088)y}~C!~*DhJc5ny z)Wz!3WTGqAO|I`DWGX*=WdxY1#-nXoHh$$?ankNZ@JR9wl-abBGaGs^ z;rc0V)${}WU6%rEb=hvS;J|XIk4z^=yj5}d^0~}0H3D3mCc%)6&cvcla9*B2PkSw7 z5ON1-x90&iF~^Q=nxMek&s`@gf9xeE9+}g6l&+4J9wn-y41&MzbxD9gb?(ZUWKzUmHH3UXvh{1t!OT6!qPWP8AM7M>}FfH&1yXg`~uu2{J zx35P7XL&Z$CJEo0`@yEcvtV{(B26vJ;a@z{fCBdnnooCB%ZGhKL(3J|xaJ^RSk;Kl zcih3MZ7W-_X9zKlY$5^DQCPgD6F++%2Dyw0>|5|`_E)Zo$X-8&`O{95H9qUOZ@XnV ztB6Oo{7!2))_RP#L=Is`n>XR1ta3aU8wxsyPeG`^h@@MD(eWb$u2H-t_MLCWjpu#v zx5XUxA?P+4S$u@P`cgs%0+d9xvkNKd?V`QwqQFGpq3k*8iGPJNxt-uS^B?*bWBU%U zw(~p5=FIb?Y>_o8-Oz#IPi>it(2c4}^oN3uHZCqjP4sS~G6sFCfo0DtA^*P^GGsQx zyYC}VyI6tw)?BAn#b#`tW(N5rw;2yIV;1(f6m2h7kawGVF}LkEd9u9=R@rwFg^PLU zpkIbd1o!T-ByaSNc}pvbU!df^T3R}*iOSN2BG<|scoZ;#^=Y_ZOO+keLpr>%6dd;h z-?44?b?Q|$z-wC?!vep@ z5cG~5%1p0^8Si zq7Gc|Ok(X_PxntfL2qBakE*3fxM#04R9+R>Hx?(bu5&RSU1|#9N4zooP&a+2BMZk| z?MTCpI+8rQ8#jhjaYJr@pefY{Y0Ap8G-u~-I^20W6o;jdUD=XwP*)m8P7yvExi{&v zZCb*aE05eOStLrCXU~RxK7;mpE5Yv51ZH_I64eS7anlYf^td8M!q_LS(Yt{&?08Me zKh1@V8Ate?M)T1^E|gfMHlgC&NOZ0iy8S=(AZ(N-J{X(@o~BM@&5;!RxP1tHpmLJe zaybr*rp&>Bf?LQ6>(N}whsO6RquhbRAZ0q8dUhsMKSpOv*O{_k2OWU73R|6XEK7A;a50g1gnc z6N}wXV#Br&G^;p+f5x7mt$TCu_(D54l9dMD*C`!;_XhELc@7HlU(-Wb>Tt(M4t^G< zQ-1si*e!Vjvyc5{r2=b;tkY(<8*gH4^g^^ApT;*={iN|FY25RYKKkkKDzvb?O^q9q zsmuOq&gh#5@toibFJ9S_rKTCAK0Xp>zc@q|9TNKHZ=cd~u_h|k7|nf`zecAFdk6gZ zChnrItmxKD3sJXb3|;S}2*$4iG5gLD=qj-X??wf-Dt#3bDVKraHV)4Rr_#mVAL&#% zN&cdACl&d81^3Uc#Kca6vrbwGUo5u4o$$$^mV5^9{HtWHFV--npl}jde5QI`SQ)*R zm&08P*ogOAML6eeBD?V0lLe<7Vb#?WSi|TNGIG-#99h4ezt&X86)aPQnb&g3q}W4b z-Ro*%AM=E}mi>cR^pAzVM{i(|v=JWJxQDMf8%>v;eQWFW<~yz^&Lr!*)9FFINSN@= z2);e^g6v^_&^tz+e{AIpI>%0v>ZOTnHW|l0pL1X-3Nz8^_hI!fzKoW? z83K_4>m^0gn4rx>=;>91jIX{pvHLGM9$3U4+50hrw;If%D_@MY0m6=k2^Gae-KUnxo408wN zyEU+JNItyS6iS@}1xCZowNy5)luU^jGyB?1U-S7O<*tDVZ>C7(93L0KKu}NYl*}P=5M{&iKPn{quDCqbUN! zvxHfH^m7t1rkYn6TZJY+y20GN9CIAZs8w8$umju-CO16Df-M>J_ufe`?#(uG{@PKx zcF9=MEbzwNmfF(tsAb^4NZ7@R{U%xyW#MwrU(z$EBzPas3a&#v(IqmLsea8zKI|e@ z(zIY3P4?p1fxFnKAx~Q7m2t^>;ZU~cDe-Yv7kY%HMEcGy?(`f1QFA?+yq>pQH1otc z2oQE=U-M_+iZ4NA_^aon`Cb+{OD}~GFCkZWV;GZgd5xpxhOzTUc{-{0x6M1@eVDxc z4k_`uiC4-d$ zYlU269zNJnkF^urn8K4f(zri`KlEFU4mj*$b<3Y%=-xso&t3=}?_DA5l!_>(>Mv2d z@5hfx3FRiuD#Qn|<%HjrB9H#^PJnle%556}F<(JN^#2Hh8yDj;J=#{PEBh7Bo z@qLGJq@E<#V~|QdR=JTivl=m=w1RKoc9J#d3md)dIBnNyLjEJ0zjgTyCs7(kqJQs) z%p)W4(1udT^Y(Mwb7WkMVDU z>9?jx>h~!HRWeF1vkjbmkj=WxUF>2PM2F?8(96+D-d zaj9@-+O@=tJY8{y7GF+;wG|xHADs*pkylA_@K;hEAk8Z%C-Uy0Vk{Atqo2JsjP8~b z{D%t6Ir|}2h1B5GIkGI}PCmXnI2&`jh3wzOA{@6v4Gc$yLABm79Dh@sDJUu8f`W3W z&nlwKHv>*d2|gxvj?TE{R~@iK4A`ZeWQ*b#x;$+nPLdPO1qTeUx$O}yGSp}NpEJl| z1vTRPssK04n!gHu@e&hot8bw#j8u>_)6IGZ>SHs(Le-?uLnA zUf~4|`jfeX!6$G{X#lk6e}=Gel6Yz032N<_jP~}`M9aVc=N@^D2j4pIC+=M2=f_9# z?NPa;c4skJa7PlAHILDq@AsqiqAJG!I?EXd=kUKFHN>@XCG;;;gJ-8s;rFE-eIxnAzb=^WA&-NN`WX6iGlpw#0m;fK`1;x{+Wd7L zjt}00S@$B4a~1CB`6Dr8Zp@a5>!DeT>dZ9OT`SV>vRU6Lov*aC@31oQxcc?~FxI&uxKL zoh0_>z5}j0ecwhaa!5_Xp&VxBug>y%-AP(tEG_!bjq;tdv2d|2sdf^FuP0UU<%CYM zyC@Q_{d+)F4lLrnjJ2c31n2ojlL#1dzZ`yqO`<7}6m9p6%A;TJBR6?X92Bf7gaDHW zw5gcI^Db!+#1m-H8v((Kv-tGUuj!<%Y1kUD8mH}^gjb%7L~<^gyjgw@$D3QT-{Rq@ zx$HO>Xy-tztm`N^4~5#(56De+of}L%MSp$nq)DrlNv=XFx8(CrZp>>B7~bnot^WH; zZmn8HGDR1t+umH-`p=Z}*6P4(j}CJgzwTF;R3>ASl`s44<&NV_)>7B!Q)&F~e){_M zC7Nv@3w5r&#C9MC!nZjjW{RfHWygM-E$42Ot$t4Fz#PbF3H*@X=^U>SmAir&f zKImSl7BbLzB=bi;T_((&Tta$L#_l;iU8;g2!6nfZI3B{|q|koLCdh8S1gEB(;q&xx zV(`6(EHr&WzRSDdl~3QPR+&7txgG`nUWMe4dksbzq|@7@oB7~B%gMMZ4|2_P1m?wz zuC7?+S53o^Dq4LfKKl=mih~4?JT5@Pl|y02au--pzlH{jFQqhL1ZwXJrszO)sF7`D)LuLS?$*>U9Nn0WfP&zzQvM!+8v1-=?9ZR?)Kqhx&qxpD3! zcfDX3oH?|Y-}=Lli~3PXtaLp>Go*vyzp2{Us^KS|8bZZIDb&tafpCzeOp+9cx zQip(DV?lTBFrd#wbpL@C8a_IcH2h2@%SSYD8zUJxd|Sw-s57!?&ra^j_C)Sn+h|lB z?gRn-p|IJrjwJ19uzgktR6|1<_rKnTIVYpY%mr`A+vPLqj)xK~gm=V`>N;3?#e**I zTqa~8yUFHrnJ|C$H@KiR44*d`kZk8!LYFuSHgAZeazCHZC)-N7SKov9;XY5uss4Cw zZFDi&m1cn4yEOjBaUHNslmp30f`|ReebRd^h5x8zip~4eDBZOke{YYa-!)?4S!FHI zo*k&-{;K+q@lX&w%_kFQ_fh>4!3BI`CYNzr12X0xME_D3{N!DPRT<&*c#0-0QnUnj z>I6R<E#87eEm`lqQVy4o6izfQs@I)NBbGn|=d2BGf7Y;I6p49D8I6Bm3`eY5Hu zTd{*<`>i9HeP|ciVH*Q$1?I-}Q|6F);y82ttI7U6m1jkN%W=*PMK=7!L@YnN9rndp zkn=a~VRccWa8|a3Ny@X)wJ!{_4?pG2e#KT3=}~ys=OE5`Zd~oS%$6+pF0i0K&0|)d zj98dQ0sV0Mgz!3r8P6G>upQaNuT}vDR`{WWNjQ9rnGO~Q-Vx1%2-fx8B2F&(-w6}Ai?cq{ooot*{CGre`AM)9^XD*$KfiI^ zYZtbQm(}jP|6{O4{fXt=AgK zqS_{sH;>PA?Sa>+>%WWD!(%?!R+be*Rf#2gQZSimc81^)$$Ti9@vd5X*GX8h<}dx| zq(!^rA5(>j=k%h30dpC5h)sO{7u6qi;O`!F<`Q8I^6roM8{=C@nB)gEtgAx1JZV-a zy&Br}@`=Me!OcBG9u_rt6G`0p?3 z{qkkY?yHD?WR8ZOodeWnk`&lKwx=7dD#-IMw)FSB7u?9ub!@lY5hxNn#++)R*|8P# zP~*-ISgAh|Uh1v~BkP6WlkG_>$Gm|DS~^rc(w`Jvb)z+FqsbOo7d(`m!&PhO!;k;G zp|Sc4-Qxe4UpH5ntNwRJ@X;QDeU59m%h$%iXU7e&-#Lj}bT66QQ%c888gf{XIu%w- z%i~|)3bnl`ybrgg$g$ND-|_AHF`&pTA*QXmRIDhUc{r-Gz~pR}8Ss^AlosxGNGK_EXZ55-4GpsRPYF!Nc74wrqHKZ#@?ZU(bA8aRIrDN2X)Z&Zib7K`95;>NAd;Avny^Xc*{$T>z8wbh6h)AkFH5NOi zce8WcR%$#Vn(TZlxGnu|VfEkQuv_a~SUVJ}aK(CxX|`-yJbUb@7c&I;S$$b(>!m+b{e&91=J%RQOI2qX40 z#~9W+%8B-j7Kb>EiKJyhKKVIe7~TC~2K-F@$8{>Has7XW;MK}EkWC`t zvL}e!LTftJbCdAAYZiECe(+}BRCs=MkW`6|!IUckXQel`xI%_X*uhK?g@hfw>u2vUrXx3rfaUJ%k zO97Tghfbx=X*Hw~G|<%Gdyu_!CKgC}&g~>P!I8?<3Zy zkHD=Z8?kWje?&LC0vv+_$ySwN@S#=`t1~VWc>fCz_6=d{sx~qmtwCznpa8yaqA}?C zY3zI?>~%dQ_}A6W!W^?3`g<1Qnf9stJ%MlVwZEX+px_h@F**t|C01lcbvmvzyh_da ze12HI6;v%Tf^7%i5TAN8Y`?0C|6WPZ=xj&U{pk!l)hSq+_CDs`F-6hxb5p@-Xq-*- z<`ZyPzl+4&_)E8*>mu76JMsJcxv0<=M>6itq>^{DSVDpeTf22V;I(3)i)%@t#}#r? z0~qtE$6kGT(eT<6=${Zr9BoFjU8|H>f|w2aBJfHx2L=DuwtLh!We<&PMpBkJ14Kw;c-0XaBPGPgKFbvl3{(=1DqvQZxzuBxI@2 z7eTn?1t^u3hBc~xNyAEg(9=(Vi*v7H9NM$rrt4U1g&~B`%^`IO>$yLt^07327mkoM z0V$VkK&eHMM~&3)@NF+`($K=)???lG$j|N4} z@sRq&ntqDV!t|dnxjT>EP`@di^hlK_9+{U)1M=^a$}~lE7PuC{wyW^21cJWwW#RzO zxdD$9m?Sk-cw@z3q^}Q68s|l(v>zs!+BIZe@j@`x9)ex>zL1X6EBI&UJKQ|j2n)tA z7{7WeZk||8$8^cV^1*&gF12OFNvdE`>Wwd3$Fq^bytu%44E(!f#FiB{z*IeL(PO23 z@TXOtcBHF8dYB0{sx^VZt?$X()hV>#y$4qKS}^H$6})#Pj0BZt(@}cA@I+T6(by@4 z>s={_QK8imKhC1zJU19z;DS4j?W~rfZXn!u6q#C`&dblUFoPMI zsdQTRL0TGbk4DeO@^YS1Oyxxpqwif0Gz=1PyB4=#><(?vSeb@rx`aJut}?rJptbr(`BLVhP{llt z2C(XTe{oxZ2S%Mbh@)iY(&S+dG)Sq4{0X^=LuAIV6UIXY04gQE1?Kp0kt2+yku3gW zK0TL{j`ULsl-Cc~p3S&tJJZz&A&3?nE_3lP|BVh$)eTl+^>Np@-;y)@Qh^#x*QvwBQ?gfLOHuuTYVj zgZtYq(~ci1qWOpBa6!M#AhBsYRyJQCWPd(}|3=oCr&N=>WHr;6?S^IY;y6k8O&LDj z6e23isM82_nsl|9)4Li8OMk4xj`Ndf-l2MK@sF)AZ{|9_P{_GUo2TQJ@S&nPogOGL zpo2X%KWVrJ;!2ZVa@l<$>Wn*qeGi(bX7?)oWy503pF+u8sq6frU?-?Jbe#_Q;g4cN zm(%AD@ABR@fVtDst7dC-L6BuWELL0(-H+-Z(`g^|DwGqnI=*0dGlvdL5*+kbXK}85 zXUN9erMPsKC7Tnc1LLBukL@x7emc{)f&JXA4}sr>MK}I?AqBz`(IPx#0V@#OdODB0ldlS@&x)k@t&$z@-kb zD?5?PSXn|d#<@Xz+bL|j6@jfnpU_3k5RcnL2n>5uI!f2H#-&ob=C;dK_R(lBIXG-5 zreS^Lh}v(eV59br_gf2en=oNPECxV3#kVOKKYSjM|R>k`(FT zWv*cNM85irVG>St2&Gv?74%5&MtX2X4a$lgB|ogUk|M22q8VYttTmk3+4RfU6C8&= zwui`-w{PeO{wR^{6Y{izv*Qhv5hb6^!VKvZU4FThSDo_)1lu`TSa%spYA@h(w`pvf zyFZqm^n{q46qvk^r)_%akk{kKEh>0HmRlxMn-zdh8!O19%saTdN19l!%7h#J-+*&U5rj^q4u$Dp4T@NF_0?D6C#Jggll zJU`|$@geqfx{54*P)I^u(-e~4?hXEf>g;Jx8~b=h6AF}1f}&W2D5riu@$R1mv2!@M z<)O{G%6T*&6Ge8oSJIP<)%b)_6Sz^c{MgfoF8m{Ox7M8&?p0X@C|C&)^4H+gf0-CP zR75p?w18KlIZ8LXfa-rwa1~r2`Z}RdxZ^Cke76Ok(bmw^rz>Q?&eO=t+sPB3t8hbC z9m@u7u{1@2jozk?Q*5G8CVUFhU$T5cbS3nRaMY(OnkF^jzH<+!oSFgGw{1qO&-r_vjfahF^w#rxNgeUmJHW zy`H?)XoUOq1EfpnDmaWw!iz2|plrPqIOr>|sne9$E8dn_$~GduY$(0iQVhSe|Dp4M zzmSyB!OIV0zN$m^p-L>uXMnTSi2bJ5Pemt_76fCvf z;j*X_txPSjx1|70LY<-Wa~u8oCyhLEOQfd<$CLk7NyG8AHTdCEUUkJhef*n|fS=|X z@P!@6D7WVT4QN%z4Q_it-)lICPu-5<{^s=k#q;FT%nO*8=*ebn_rfKS!Y(E)3gaX< zU}i@N*Lt9hFTa%xw+-cRTdNHXbLb@F|6IcQgeq)ZBmo^3a=6$!5H=;pgO0ujowMpA z->UT*)WcSl+dW4wNZv}D7}ljH{>`~6Z(9k*kXRPNgMq*IF}YZNrS2NH@MuR7Ib{V zAkOg#Cs!THxlzfTG`QqGow?!I^izk%feEh<4)Zzk=S~}%)NN(v_V(jh*(u8Hk_!Okr0maj>|9d<6Potg-PMbzLZ_dZq-DmKAjo^HJ8cDXfb`l78#zcju z+{OS)7@?g&tlpiY4&#rZZkruZ{+3CUB4*O-@&5dV{2Tnujm`AQ3>PpBwZYC!eK=0j z9e1mrp+{N^$hRlO)~-iKiE9o z$RPQcFbnNR9z*%8ErKWb7B>Cu;AEbd3EYb!eEx4geA;{sX5_ZejyOG0`SEpV?AL%b zn+`zNGfDJr&BoujOPE)5l8|yWP*^SrmnOFI62Y@T+boQ`Fpz2UV8Ifs_cHQ$ImEo-6V4t!623cK;5)~J!usS8^6Q}!lq!6;wNR3!`~RCxaab7{ZuFDr z?b$;etAlt&vGc@Zt_7%GZN%&`Mz+xtLimP{NcRr21J6sBV3ENtHhJ%ScF}bwGtJRu zdBPN2UP6v3^jish)b+wWa}NJ?_9`@7{gsS6{eryRoz2TF{!O|II9wi0srrqv^!2|{ zq-|FdPK_&}R<>g3JfxCdG>fe2`SOm2%B$nm-vT>-fs9RaN;!&E$MbjMLaWp-4suln zdSI_t2wL^YWVX^I_?>nhc5SkSIhh;y$5q>KmsS{0R+$sSWp^Mw=mG@g#PaQFZuI59 zDEwU4hT0B8(6Txn!Y8kV|E5lXcEJJkYwTnG!tocpYUq8s$t#PzoBrR`mJERzkt5D* zUvf0vRgX0Z-&@CjPxFbRY&0uqJxZk{K0t1{5By2GkL$XHdAZ=m&>eZ7mY!~b%lsgV zS2>IRHWV}@5wNiHGyc1*%EXr(Ate#P*wVHH`+T%u|Cdv2 z*;#9L2r1d|FcKdg3Sv?_b+K`4CLVP9hWD1eBs*&A;X$P#lj-!McADBuOJl0&NtRF5 zf!jjgL3&h8WRoGJR_2qhuhxU{?_|<)td}g!{LB3AW#e(rT`+Z+0-tQX63Dl?7I6b>#Hfw%GJR5uvC2LTMRCWf{#*KnN+Tt3o>i$$fQl;H9wDzfJ^P_ zq|s?6E1c{=mV6iq1?nM8J3Nl4^k$&r*K$^pl?gqPi*TxjKMr3y3Ih*E;F!u`@ZZbB ztibyK(ULPp^TawF_4X0o^Gav&KH}*0<{{4HCa|$Lo$2|-1Ov=g*Bs~>sOpqk2=%`w z*YuT+fYL{~_%Ge8MzKZ-t-lTx4X8fk-0z+v+az}2rH>bJs`V`5{(dIBdml|s#7%(Q zw~ZKim$7VdMd{Ta5K3D_TF(ITK~3j{pz92RsA@%4PKAKwafvv zV{qnoXDqY{q%SoxsR{pxTPNwr_){yf=tdfLD$28eZRN1`fg(mZTd~u#r;>+DN@1@_ zA=~jgkNmnMNj^Voixs&6qsm#XGT2_Y9cym=*Bd?kR5Tv0CKL+-|JV+u3X=#QQ1(Cgj8{QH%J{iG$C_AZxQ-8&hYC%B8|i|K&k^)aAQ zUr3CVqS=NM)4;*?9A;`ZR4d?cq6o3%@Rup*yP*k;4}E|W4)1a28$-nYZ2&9x!r0ca z_&YFyd~@%@4fj*2*20g%ckDiN+d1NvhyB#0?kKjZ^bq&+efTj|yXKC96rCq@%Ws_! zuW?>)LchL_hP@Y}*aXXRqT^}}FA7H0Ox;t?*WHX^Gu0f?x9BN#oEA)%UU;G;U8LwLWCJz!FVUKw+d7xbgMHn zZ=fQIVC#^pAh+}e|2pa=#AnKhGFIqwcex^XqxhWe3=M)`F=goK*hGppg;V`?R`f;3 zENpP@hPOE3CiArr9Qy`D>Z;Ymu=v!Kdgny6OW3^s9} z$g^eRX{UlJIA7WX^^FtQ=Jqkv=KUGEq)Lf4%jROid~Gaq>abm#z8iwn{*nnFHn4Z+ z@=!X!j5+eW?eD{L;FMf5jtLXIJ!8z7{_bq{>3$QbOSS|_s}zvx@}QqhN5ST2uHfNq z%p27ivWM-XnR2-s$Xu%>XA|_XG3X==?=!{5*$W79UV=N$9mfr;)L}WV%zRYt;#M6$ z_98YPybpPx<;*%<;JBW5^B&LjzcQe`lcqzAhditk@`COMMp6rS=Y zsf0#)*g6wBO*vTc@g1MYmg1?4fiS(?0-22p6jrQ;rgAY}YuE~|=|B$2+2{Z&SCDs_ z=FHhH(Sx4{`?*^`r65*Zim%=$Fk+Nq(BP9bv3gzvsN=-`Y!vdM*Eq}>sYPUx)S2&1 z9h~<%4|?ZNdf~?@;+$bBOcRO?I7K(DTfY| zZrEB^^Gz6|1%+sW=dGvScXgMXql7UQ$?h1tL>Xo-GB|IjX6wK$IbIJgRvR}^E| z+gL~l=p+6_os)HrgZw+IvAFvZG48sLB?DjSjk3S|>E>UE^UCSK^mX`pg%^4oWWf&2 zZL}}&FI{Ys3@gr;;^Aet@mPF2ot-%p{-k81pR6TzPRqpD+ANxgIygBq7Gyo@prfgb z)98xBd;RM0G;TI^@+<_Cp9|RIt~C5|EgbX0%E@!ReUPS-OKrPPV#bqHy8NCJzgl4D zt}#1<$#d$kdj3Tkf3Jnsz1PPD9nsuYf5CB8w;cBD98Fi2?4ri&cM2}X!^~U!0^aeN z!pzzi*lLvQvn8)4^N~AmlK-UpIQu@~K0J3U4F1idA&%BiX&gZ(<@kf=SdwcC2ZyHP z+7EiPczF${{aTy-a*$>t4=uo~W+CI?mrBI#Q%R>BCq55Ubcr$XSHSljYGDIqUXWSb!whNEO((fMG9 zEB;B~sbN*LEW?EAU*3l??HzPyqazQYDfF%0W%!X^N4ugPVf&X@Jil3v#d6v#RC@{1 z!Y8PIbPf)8XhN|t4YoT*#MCz1FxPYAz+<)GgiollU7`An&<8uYXXB2LJ*LfcPgyOM zx7BC22bA$DZ;t1O{-g$yJpCBvgca5@DEl@Qb(RZrfKuV_+eYKgz#=X#K!(X~aUff@ zK9ednO=kJ5quQ-^gs4jBdetm@S^aEiE4h074ZZPY2J2IIVau)N;{L@$aq9Vad^c|; z{&z9}eB9jNq{(a0>Uu&h6$jF@{%UM+Lm;}I5o2yC6SxHI!wv72qFvc9obSft;X{6G z*L8j7e=P>*3qIhK4l^_@$;Xts?Siw;jEu;ZqznH|Lldb`aG19MB>Op*c5WmY+q@Z` z{~LmfCrGf4)>1e`8cFxee74|B9jacMMkM#m#@XLY(ZzTbGTDy;-_?t4f4>zxvg_#E z&8FnKsDrMZItNdL6ykKnx5Uy!o-UrzP5h@!g9q9|PfzU>Ot>$c?Ifl%UZoLnNIt6H zyN&l}oTc}P3pl@Lf^*jg*O{ha!11HRI!_6W?q*=8K^mNm>LED;CY+6P3rt@6oc|_q z76vAq;?}88WGVhjseING(yTKX%(-ErxZ#8p+is^nQncu4dkIu74a6mlA^84_C2o}w zXa8wvi3VS!k!ioZV9m~6_!vJQ*WR|LmN~7k)*}^OZEXgrqwA^5rt2s+HwNpq`|(%K z2DpCYHI?6T1l8=5pxJO0tX55=N?HGe`?ww1v^5Dvr7fUh>516xI1FZAxQgqHPm;WE z+TeHi1djU=59c=cf#sX4u(P-hQ`db!-?_)>Pi1>*GUO;!1`-gKt#GPDp{BakR+T2)g=}!>p0hGyYg@N7s(5ptB z*wYh`to9C{JmK(r#~LoS;Sz+uy93*d{=f#v|b zTZn;-=6q(@8cCkq*93FLD`=l&ESjkIjrK?B;?uXWm}NGU|Krs`#$H?rCL^UqYfa_A ze32QQ{cZqLZv>J}QQx^44*s|=Mu(SnU4|30Tkv0ZG|IUjz@m!b(0M7C-a7b!=q*BBz{LvA$KhA`G=aXFc!Km1)QDpC+?ZSEiN`nmb}P1PL#a{XlK3)yiAx) zha6nRKM`BT{r5G5=pI=ItHXj>*{Wt7)l~vjb7IJQe=9gTsR`c+KJ3cELcVUyFsuz6 ziH9VH;g}jz{BU|bIGBCnH`?W*SHKgpxjBnwfB8&f-qi^(+L75@!3)h8P5e+^ZPm+u8lyYcXiU59gP&1v=r z7qAb$LY1zchgmunxH?i6H$)%i=f(*{-9Z)Cg=wQ;kq|bJNxbjk6mdS5P1GE=)%Ljc z68}CMx|L}r@f%IS`>+?0eJ_I74(g%RX-~X8-xRU=5bC{J!)D(9&PpbK8!G@cCdJ}rst8qGqe2|7cKY);)78fD>0{1r<`iSrq{x~Pcx9PCO1oYc4*TpC`!RYqUO@@Hsi z1L;UOk8|tCBQe#)87hT{t^kf%Q>n`Bc^FxjPTN+$rg?lM9hQE(1u#7a?S^{sxgsnDOfp_ERyj430@V9pL5*$!NYl4Xqdcg7~fi!94Y zCI4ym?Pvl3jk#Yi}dmx4!;W-Xnmm=_W zBv4g=y)rI+nVJdW9AU4VRP}r zG5%+&q_L56B7QHjn55tC^hxc0wzJxfjzkIQE$0v-x-N|=hApI+Lq2cph;f6 zoQ1L{Z%`lC0Jt?{JhRk%IXJ|IL7Rjge?Oi=qQsqWvCJiMUl@i@hnGP_`eB@?=1-*r znk1=p9dvC}gnenxNx4`LpUqW*L#;C8!WR~vZs6~N)ptO7?QznR?}4W-=D}n>Q>hG9 zG`HsmnQ~o@svIaF-^bhV-1-{$?HELsWOJDI;yhU^ugyemK2Bf%?Pc#-N;6_FvdEWd zA*fpzj-`1+G{x{2du4|RS+;)$9kYCh-5Qfgj(r*rB|q&L-38h7%A{4Gy*7>*E!8F# zX7wbwS&Q1|g)sfC6QHlVg(PH{z@{)AxKK0&oo-Kof{AI&f)`f^$mD^+)euw}?}kY{ zpUyf~4%}N6K;OxMuq6)}vDmeEjQ58J%O}v&Ge%qLo%6vtclC*`+HE}LT)`Gh34(27 zhfq6FoMRnd;-Vi9>EgEn?yrjl_j+qL`F5wCF?*{<-mkw&YAdGDZwuV;!NF5pf%^jt zSg0ykRncPkmS?CZ)fr&n^jugkJrBLRsz|b@0xVoN7VdC|`D|k+#2T#RI!r>icWXs( z(XH>Oe`^AlKJyrRa#=T)?F{B7NPXsK*ddr6T0=TV3%C*aMqJ)x57RZfamH~yj!C|O z?y*4qcwT!XJ_&lfo#)>OD6+&YZNBA9N8KzhlAT%U`%Pm6; zk5)tWGJ?RVm=lU z(y8^fKa7XGKixf_=jGdXl7H}uEWDORCD#_9gyRT#n=*y@98(1o4?QM(7rJ3!&lVDJ zR-FqI>A~u*L^u+tz-KICnG0ea)M|ee!wGbmZvIQ%w&SY#wi!))S5XoIDvYtFIiJ>9 z%3$@q~v9FkL5L^p0UqmgE>=-|mM6E zV{sJPO`ig~`QliSA-*Xy=j@D8t6mAG>albMjMggIoc2 z{_ut3Iv?XD zdR$i2gt6B|xuwUB(J?o4Xvvl$qMJMcw7>HV)_ot@9JN7aV5=nAxnV3b=eZZ?s4r!^ zMHO*h;3OIPjV$^#Qy_@BKr`+&s5U z`V+I!h-dBpbcki$1}ew~K3}~oCj#Og@;e&d@7-$omFd@eAY8q`2%ou#F$&UaaE#NS zaGM~Gh<1!3k0U>mCR=rGnx+J2uWQCV-7N;juSUpPkM%@s(@yqDs|OLoqj+K5S99WwQYc5)E?4yr= z8q&nQdaU=tH1K}2lziD%X?c|AuC3Q`g2Kw@%&kAc)KuRA3;m1GD(?>ZND^}5TOjLM z@5^cw1wqC?Z+hhUZ@R!hfpz8CJ4$AaG~~Mj_{B-X*wj>b)xCgTzOGMJi;O2)Q4+|T z-(g0*ILvEEC#OXYGrrqXShJig`a0wi);hRi+2a}#qijpWgsvF4VkPbQ>mz*sN*eck z^27iiThJOCPs^^xu`7QZr0=IKqYZ%*Vefx`sC@7oR2{E_AJ3nGrrk5~nn*1sFtc&s z^a3=TnZzEP?*i2=b0J=LF{v3)q|>Hl(fsQ%cymnynpjgDyT%=szkfy-UQNX9cMQSH z%8S^oUPx?C@_Ckg13WhG7H)2I#KDwT^q2We=y5(rQ)SaJq@f>o&jK9(SeM9m&&DH# zy6A6|j~QdM*)w~Mxl4B}1cLAHh|XXk#7*Nr5C0x!v{NU_$C_|W560GI@Z2kllm+Ne zHH6KJ*3pHvT~sX917Eaz@C>$WvgdOn+WGAueh*`5M9D>ZM3%$1;k&T5`fcrj@@KF? zN(F02H$zx<4XM9ufF{Fg7+n<2KgzsVm^>2lkMCppYmm9ASEs zO{r6R30wVE6jkOdht{w5@c5uFFaA`ag;-C#AYexNH@OJe>zn-og- z2*u?NVUSV_Ik+K!2KMb+6ct*Emy&vV_13at5k$&r0&W?R1 z4UitkYMf@1Du#*nqEB>W@=YRmIE^-54MoKsd8pNlWX>3*;Fsf(xFEGy z81Q_YK&So~H{;k`Z2mn=<9=r_zt^PDzRpiH{@HwZVWdwKba@x`Ep^P!a3PGC4{4e4 zg4pHkXTI^y$$5#E#N?Phs4Ux0?QVKwlYur<^F=`VFNcwb_hmVqPm)-Fjv+=ChJfG1 zNQK0HDEXyKZ0{X{1)99`@QN%taxUO2C=y=V+eHSVA5#evUEHp0!CCK-;_hF#07{)j z$mKr9x1*xL^7|BAGw}yby8e-HvsKXN?M~iH=!Y@S4q$DnCDqN{gAFlXc_e`Yo(nm5!`ZmpT?AN>xv=!e9@w};hW8$SXE)mjK*Zhz?3+cod9zDx>S*$7hNxoYXhMZ-oNzR5*znoV;1-YcgQHZ!G;|EzY}dB;e}mefVqk3kWtJ!2Y7o zP|)}Sd-h#NZQGUjc8wPIp{p0=ixRMtq|%go{QR@6iVfGf>n(S|<$yw3( zY?q9{=;S4G&Z`$Uk1^u86MpKSu z%@qwe`%nVRCZuDA`Aa-G+KV<9@ciS*(YU%q73^n<;9ln@p`PO~as9LrHv1hVx+@>j z?X^75ZX}brces)~<#@Klfe7*;a5gNi&SJB#P~ozAFZ$!~3Uc_%Eo%9_5Jxqp)4n4d zsv6;-PD;D%%+i~)3nL9rBY9-O!QmW%<0~5XV=%rg? zpt$!v`)_*^#11r=R&RU{?i%UT?v@x)x)8;92xnm8F+b`ZvXYAB^EtA;|6z6f0`A7t z9J00k9qEotf>moTkTIuJK=a%T49F2>cUQVYyv-wW%~)C>(q%#BB|741y<3>yQb%Sl zT?LD_oFTFLrsU!C5;pv(GvD*wf+`xR^roPZ?o;k(^9t@`cl2r`%N1bU?tAdEU=g$w zOrs?eD(KR4Jf}>ukmTLijXSoK!J!EPveqS?uDmHj<@z_%AvjG!YJ*T^c^5UmwG7X0 z)r4XD%fxlw8RlUZ&q_N`Lg(cgLqksl9GQ2H+WFbCnFHrY?!`l0JkkYG*@WPZTI@?)-^8X=R!+9xs`j9$57g+~++X&A`UrZ!})!}T* z5wLc(fXI6;VDPt#q^I)l2C^@%xnFrjH9Vix!k5u-s8|-nBMzEZD#VlSD+-_)qewZ& z9^zkOOkj~C?3@>lra8yRI;Cp*!Nwj6=dZ=Cbs6ZVp~L>)WA->~DyU|p5;xx4>lXc# z#vDD3CVn}%L2Ma@`4^L}Gbgd?A)!yy48Tw+4$UragutOJo?UN)vmNK5QrS&@Hn@m) zFQ1?y&rh-aS1&-bhYFYcZW}m705B>u$evkMSP~IWf3Kg!^|TJ~ccNsf|K6P1Om~4w znUAFQ#(A=}{V$o~vkmgjaijs(ld)~#!d3Bd=vOF2HE$Dcr|ewrUVSlHEqaztn0tvV zt62lp!PA+{FcFGrCqe7QI7l>61Bv7AAi6^e3~R*%m#x&Xy>~V`&*a%{Qn6r`dx@qV_Or9tr3{wRLj1#qHn7dvT<3)w2=2oeY8W2^54 z0;yfh5jS@%JuQOVIt#Xd@Azz;*@ZUdW}vdWoOn$#V<+cW;iuniRLrfFR>z&hOfC#V zk8H=nBd^KyzDy$Dlu0D4oq<_uMpk+zQ-{xAa7242SNdcXw=Ta1myc$+SwZextxOeN zHb0(u=HNuS|62*9zL$ViIGw-E9?}o*A_H4OVfFNX)MTS4`rQ`gbV}Z{$M+18PQyOl zzwJP_OPG_x5%PHcnmk#n>BBCXtBX&yC16J4PkQ%f89l8j3%4DL$lFDR;B+nwhi^8a z<>U_B9^b?qymlG?6S3x)n6V`0I$d`a^wH~nVgcbfQT@SamUPeiYB>cH+LmoAHU?k}Kk_m2aI+$JvD`PLd}uZ7vK!BzNdv^+B?f~Wd562wClchC3ES9Xpl&iw zaA|!Ww*37>K8?!-&EWgA!=RpeYShx|Onpv&f+P3Px*nC%rVGlU2jwR1!AOyMl3vSa zNczS?k<4veswR(F^_fss{gDxdoTh)z+f!opoTz8F65k#RIN%VdnJ4&RR|#o$9{X;ZN+6vb)676nS?=cQ7onY(c? z<1!rhJs0h_@4!K61@4{4JdVCg=Jp4L;Mu*ih*NDbni~GN^nx-${ zOsNWdm?;4rKl^B}f+}v9euHQ1hr(3;EGj9A!R}T?E-egj+c`g2w_O9+%k#K>bJVzc zyE-g?W=XgI^{CdJI7p5+r@_KRHZKL>Taj)DGHF-83qmj5>metu4(4VpX~zHkP| zt~8_}e*cj-qe{Ui-k5FHzX$Kj9$~0cKRr6k&mW(Jq35c}U`9TYSsNA+%fC8s)v}u6 z!MQZ7+=`ggYv57M*VjfX=ir6?id@nmU&Hc(gYfhIwzvwUZmcZ{~DPLt{3l(Kw$( ziHdO(C0234kC%g;{Uwlb%fX8}1bfPb)cf}t`rwT*ov~VlOt9_)`%&k~ESpZEJpMf~ z5r)7|(d%T~qH`>hFb_1N4`9r^vD_ueICyc1_r5L^#pF!>GoLOBTaTZDIirv8z4KPm z@$CX5G0p_GnUul1BQwEiCO;eQT}{9Km;>9D3*nusGCJ%WqVDISpmMtj$x994-`kdu zQxB|&NLT@0T67a01{#8$m=QjSjDuX6=a~HbA1hJGzXM7i$E)A`xxR(H*k7T{zbEd1 zjEq`VuU!(gF5cktcK_kNkT#-wZ3&!?dP35&bZMIB9ynaALKdn{f&D!Ta8l7%u-P*e zGWS+OVy8N^OYuAVF&VIHMnB2cUCFv8%Oe~2o9a94rL1>|@W+b~LC|0fOjkGzbGz5z z4^1uXu`Pm^-ns1H$rj;zA4y0&mH_**L~v6}FMB4c0P>}Sv7~brzu&LI!HK7F#r+IC za`7iS=rWPIaTnPk+cP*XVF$OQPZp0H?%_<|3em1O7dV9c4yeEH;}rBhiT%dL{- zhF$vc?v5rpHAWBPYA?~(&s_<#Cx$dx%_9CuDPVH54_{?u;O^{HYBqZ%oVS?*`e7;X z=IcCeB)XjL5EC+At`7+r&1>ZD-m{pK(LtvKtO4IOeRPEHm7mLwB^if~!>>1?oZi#* z+;C_X#(jA~``rxD{rPtKhj+S-a@mcli3jPgf@Caw6vc{e;e>yC(y%JG7T2zsh_ti} zs_tZy_3JMXhj=&o{KIklu_cCHxyo~OUTeYnpxs3N@IpFVrYI;jb0+phgMmuRjZW*&oceVpaTcI}e!mm8{;UCeqg@3(xcY!76b!zTJ5N zt9Rd~XY*eRpNBRQ8@X?|#&!yQc<%?@@oX0a)fr)3=TT5xJ51FSTT%U4BW_q=$MqE5 zXMD=6z+>tS{Pg-WYdYyV*mjJ8equ(|4y!Z243Cooy$i9iEe1d4h~k>pQeY=83C!?9 zXwa|2Q+r&wO4mqSL;A_?zLlV6odWmsck!9sG3av9548&&IIp>moZ_qmIH;U}u}Klk zNaI=ZD!hr$jK`9zcTUn9de89c(Q=g0OTmlRq(IL8%!JrpkvMklvuW&wBG(s=aAE| zMyeJxqY~hUW(}<8`JP`noda}wzmB5Z#JzraWc^R_=0hpV3Z)=Ab_^G} zSB2Xg5{aMM3(&1MA5Z8#WlcXx;E^R$;G)DOIQ_?x#5ytPGo(TPEu6@?-gd?Q_vtXy zRz}s%siCxsI5oR73$LGvt6lTZft)_miXvV7yk+i5yspo?EyQk; z?|nmIW-ROc(-6d8EG7obIh4OpOOCh|)2jgrD00i2KJCtDHjP_K1tSLZ9>2@Dub71y z&z7PjD9|H!f*?ljJh}785zeS4vfMg3;fF(7%&Gu?s-HBQbf&Bpw${Z%(D^1RcA%dQ z7aZp{R>flfCJo?*3n71gI!N9ZBgvCHsZQDwq9CpY)A`xaj4x>xaZFdJ_Kj z48x=M&ABqK4)mLRo}Bx^pFNZE*_$~Jsp{|N#PRSw60aFZie4BoPwltEO5Y<~Qs7hc zmJQ_G<@8~=yaX;*RKk;|DMA^kS4>UJRM@#woQry6!M~?kp;WUYtzNtmL|pt~ftxah z+}}+98s?MGuC0)<-j00pI0bvZD|5l038Q-J7p|zwp@Vhm&>^D;`}O$z)29@AXX6<> z)I0}AUfOUg2c)=$*2&Dwq(G8AH-(lZ70{wXA(U!3z*)ag8tofPulq-l9qoK4=$jY4 z;NXiWx(EH0Uy$b4wX`n1n&(T-gY(7;AakIUXwP^+f~#7D!|yYxYQ;&kTDTf#i##OL zcH}V1BYxyRq6L9IQZROLJhS4C9C_61M^>M6!owfu3lc}BaesQ3fJbUP_|^^+%ezL{ zmGGJz7a5@CJJx~7?%C{4-p}>ptRv{pSd440reg95oIN>H!i4byC84S%K2plxX+YG)g={TU;eJ@qZ_ z{>FFM8F%z^tR$5i1{mvwBSe|^T-Uc9!qMulXuR%tU_OhX<;f@}#dsf?q5P1(vsi-p zw?xUXnkYAZ?HwBMBa58c^_b_b-e7Y2cZ187cCw>3mKF*AkVkK&AU_)5Kf6HGS|QDT z^&OAK+gr)g`B&Hqw*_#4IRg(?1fgMe8lCE4O-CswVrz8~e)i^PzgZ`-V@w#kz%_^b zRu|*Hr#nD(-&I`pfoK2757EbpuCQx<4f**XkIXf^L!RVBQdf!fcp$h3_ZtF)S?nXC zYiD5dH4jEQV9;WhwkSre&!%UyUA-G?!8F<|%kuU{qUdSV>G?szY9e9Qqr6BRShAq3pmTwqt22ll|a3Ly~-O`kb{W z$Fr-)mzqM8q^w}v{MooOJf6z%eCOB+H>u}xCDnfI@fVN3_K~| z`E?N({y~JB9>URm^&i1PUIu#JuY#xCO3HM1!|8mWmg^+x^WaK&wWlAAFPwmPn1vp$ zCS2H!Y21@@?Q}x!SZE(Uiw~r>VX>|ow_L{%BU2vGx-sLyoX=uEnXkw-KB&SSnSFSt zp7#}xO-0}N(fH%SY)(FWEY4Rmp))()(x?2qq%CIn`xJ9a|V z{uoeV&c@Zq6c9H;dTl&AiJslx$qM>v+*m}>&(KV{Q= z9VLPLo($^##g>UzPoN9mh;pW*Z0V{QK6s!z0xVwn!LoMR$G|W=cfJ>}0R`W`;7s*+8f0N9NFP&3#!bnA zktPN!T`q^*T^&OG1Wy_>`xJXbau2*|HQ?6F*i2UX z$&J;a-UDGKTMow`sVoKiRtJ0=4f|$NODXxM5cqXkOs&WgQ|Qwz`M(dKkk5 znQQd=)nqc~=q{+p=9$0`4nTZ(HE11?rt6kH#*DK|;p6SMv}}0+Is8o@r?|+!itMkkPoj$bITs4g6jHUvdHxg@BGw*RdR;-V(B`1 zXzm)AJluyt_iVAT+KKy7DhhNX?}|UY8Xg#glT7(QI3QO{hL)I-d0jH#SlnhPDBc{#uG(+(@`)IcURo!PwUVpSi zSof5w7P}y9J_JGLIb`XclXSOq9+MFlOT0d=rpaYJIQn-zrYkvPP|QoxmvDet*nWll z8LbHK6Fw%wWcG%E(X~Ft*$ zyCFe=6Fyl(dUWMM7%c*2yl=9{M;+p7rb7^)ozS^@5@o;4huO(n7~A+y^k{t-d3w2z zd^4zGSB-UK@4Zf9CT_YyjU*%?Fn$x6Epd-JwarDNyNZm-?bl3(t`0aR&!#ti1d%ZY zHtY_5&(fvymx<4ECJ(QEU@~%_Td^f@IH>KI&d^L6bf>^NYt%!bl;g1B%w46mqTCEVk@9HK%ZT)3(xt${t z?-OxM<~nLHk#};(+H$Fhb8tA5cMkm6g3?fqayox8XRQUT+t)&r^H?|-n2Y-^SyM|< zHOLtE46nN>(8Y-q-`7$}4NLWX z3BEnD3$NUZW_LcypsQuIsq?x5tWS7?o++UyOwop^E)RJ+aRB^y`HDQ=-@@+I+X?4B zHqu=NhtY7i9P?+)G^TT06Ydy!#P?WQuqOB>GculcCeU~~c}Rk3dUAx^^p>M`?%Uw) zfn*3-CIyObuaGxNlJLD;R?t$_h2x)w(>AV$>AG-^L{GAWl))e>+LA*mcNa6}c}|!+ z=Zet1MT=|{Q-gCGjp(EvY3!L|2uIvlBBt|+k+D@H`k60jsp5Y6;r=l!$PdGp-M+Z= z6rb-6R_78vZsX=|D1hF$`(#^oB7OWQ9m>Al0qc*k@ad%_K0N0RF>C(N-@k_}&67Yo`g!olb$~m22qx6yf&DVJz#j$Nd&Te1EMmAy+Pj<1uCwR}o)9A169;kH)0p(i5(P z{jEja=Q@(b%QwI$^C&^(1!Y0v`6k?!=1VUB(}xpx%IM6?UudMh0(SlzE$~wxWcD`S z0JYz)m}-6sC(0S1CZmQP?QhXQq!%AF2a&AF8gN_m4dzTf0jFYhNx>k(;4JPvyzMO16!j76h_%cwJdhdKkN!xGAhBT37DWz}# z>>=REBRa`!BUS&p6aVcip~pVdfUU=3`rh_4dcXg``=r~k{Lf9CFDVL-uM&_xYy?A> z&oP!JB_w-I71f$L2-i}t;hHVsm~Jy45?CP~-X6_~XEo#2yJ}nm-!}-KGag+3Qv^X< z0;vsBfzRQ2IQOh7cmLlwVV;>GR!p5t2E!S_{57jVF<>VAYe<1@k@qnF!4~Ye^pu>h z&!ZckBM}Bu^6K|K=5IkYZZ%fJ7sJyy#}R2d_ntSd5t~OIN2IV%HKpKdYby1RlE$;u zO2G1Yxy?Q>V$XVR(e?7Ot~<2{7F4&3=tkMx+&6nvdcIj5R1?yf;H zJZeyaorbn_gP<4=&$|q}tJVnmC6b^zGzwkDtw5D21~`4w4A?w>F3jpXB-k=;g<$jk zBtZeaiyr;_zFqnhI=RdP?a}G1@Y6O--LaaPEtKMNHNx3)e&6ys;Sx>{D22rMzhG@5 z#{IqRz&T&>0A+rT+L!(mt%A}(e%Ul(?Y7YZ2g{joJW7Ln?@*zy+tj(%FkV} zS@Az2a_|K1jS9o1t)b}BR*D1P#8I_CgU=4DfHCi2uPkb3HjBSQ-8+xziffa&8AdBe zrtAquvGEWQ2FYWj-W4Kw?lRFc_M*?SPT+!bn`n!=EF|iCHuWkG0#A#D+}O_vSR_)# zdgX>f@%Rv8`mu(Zdc@-$2W@QJx(BX%I>N%=8Q^TX4sVu^2J5-X#H=};2I;1vaC9Ci zNmYb)3hmMl0uf8PXsXA0!8qn%7v1~)NE&1hlj#j-1#fsta z(Ef^N3G+Mf^h1uc+eaVOG~z97y_Z36H%nHm3?nYuvZN~UF$r*4%X}^5d)VhkLxW=r z7gStLp76ZK`IQ@B!_`W{hF1y?kW8ZYpw5y+PX&W0H$ZH^Z+zyJHp?AB8mbbsSG?Blb>zVF2>N8PFw zhE$xwM}Gdq{)`UzFFV9U4wpjh^J_FPJI(U4f&^Ev!H8DzET~Vr9v~Yqnrkjdf<$XY zLGS3fxMub4X&uh|aLWO5iUZxz86lPl@*eI?9~PE|}=90@F$NuDfmB?^;M z$%nF=WZ~yCY};W|Skfp*)!i-Nn7Rg>F-XL_mhUJ#Pz1j7uaN8!V=(8lD2feQT+R}6 zDw_2eD?0OqqerdAg|`OizE@+wvOo@s_RGV=)unWT@pGy@a)+Ef#v;e`VKxm!f3*ql z>)b5RUDu15Mzc7zX?`T$%9!+wngb_wGw7fDXE@DGt@tSHEc%G?`TE7Sc#WyT$*;$9 z8;;+AUf~cdls3ho8QQeaQ~}ZpOzF-SKJYYf8FUTWV8xgVBu6zHMlc*q43luz%`PH( zEeRiuzDT_e_S3a)=b=PP4cki%P_Qe48=;Dv@`@~cm6t{T3(y7epgLjP`iXd)|GPxK z*2mkfh2-Y*II6ty4BO(H2F_=kVN$LE9X#JpwMQv%%R^XHRBa=(k4kb$W3J<+=My=5 zu?E;+*-hq#j>a*GnP@2gh-5`iCXe`>Xs>ZOt!RHhKK0$DqptAJ!BG<}eYZH!zYXg^ zw5o>O@++WEwlXyPaR%02xB>kURsw@5vLLOVPi=nQr~Wr5VQK3CIlxH5k=Szjm79Vk zqyCWBw{BbRIimzqzIm}0#EU3i3?#eMevzX0D~S}JXUtp`h?*8TL{UnbpUaDLErHp% zeu4q_id4w=^y`0OZo||g|XD>viR^gAGw_b3ntRPhxwWJd~-;hdxYJ6X%nsHhh!=| zOSa4`oql+A929~KXypZMe6mU!2P~Yq;d?&tFH(xg-AJcV)25+vjUV#DaNN&6A_hNm zSmly(n%Zm*i`{Nu=<;;#@v+OiLzw4={Qh_JTDdJZ{6LiZ@xzi6-$hx&S?}qr&^zn|*-xbCl@m;8)nnO1PcW9h zV_Gjcfkls``E#_CcFNzRDZkvHdTc2u{Hddze)1?`%ySw9L1=EAjvH-~Y3A+y+|ZY! z?32TH#i%PqYAEi-(iqgcdq*+KEo=evQFJyP3h`0hL5Wc~$(N?H?LtYTOwujVJSkRxzPHg1*^F$@XpWSZb zjd$m1ky*O~aovP8;V#1|ochmk+^&hA(fF&6a9Y|-?wzbLcWko``6u?2WDH0`e997h z@pK_x?R$u-zD<%&HRq2Cz5Z~DDfvoq1uQs8WRRhG$HYMC6DbDBC47@bhDf~A>8CBw^p-7wv-diUR zH?6LdY5L)oF`c4V^EVNd{yRXgx;KNHtsKx9p3E5ghdigX5wC1D!QwGR#C&N4jjl|k zwzU?_PqRQY`e#N*+Iz@Q+E(nHUaB4M768C_P~T045i0J*#?gUsJmM2Go%N8$t_PJLHLKWcn~ z;*>S^y^-b*bbHNiND@ke=|cA@wWT$cDv%%!ix^ zs`~vrD{OxP+kbSD`fO(|q(qa;ym$y2T_=H;i3V5jQowoo#gLktHMn9$G3n)>^>

yxgnndIdcFXo8r zcqmoQBK{IDX`O@x+|T28o6&9T44k{05<94|e!lbN+w<%FX%4Qw!-kSw)cxTMR@=d67C6`sz%4bT4gDr1pv@yk_ zw&Aur$1u>^6vQ7NqC|QlHZ^!s_o+3wzcUH&N2q zo5kkeOJNNi-;+gKp0UBt+S%(KV&vPnOtRo;E{RKx2ERTzJbB;>%fwe(9;@eF!2I9# z)a5*q(56LVlI9X;ezYgYcip#TCy^hE;u*8+Jdg541bXcABX?tJX=QE_YxCeYp_@0; zg1B6W+4U3b=S+s~pz}MB$`oZTo?p9FcYR*nT`LV}9zQ7AYf6Rq^=Lo#m@t9tJ z%zt*XrV+an`gF^i6SPj@4EnZCfv1wE>AIVBq{4L%EvV0+r*@2E?LyXqg47(cF)o!H zENUT_#^z8Tv2qd@lL}WKzGssoHnI0AEI7F+BP{(Z1q%}_(PDZD;Uvu9i<%kdIF!ug zmFshU9f_Q8b`B@4Hi~-@ewquLug}etnJ38Z`as<7g^C=3##>LIy{{GmiZWW z3XD~zK#E_ZrNRymcJPQ4niZcS(r2EaRa+9H*z)tF0EXPq?@ zdB_MZwpkLrBM~g<*VC|T2dTKiQk&X)~V81E*J|- z9#&&pus7MK=LCaFi$JqjkG;F;KYGF9F}tnO8T})EGLJ|F+46Te2~7;98rlQg`Y)5Y zS#R9XQr=ah*tL*-+SES$83Z@Q@gD^WU*jqQ5_XL@6 zGozL2>Sxq(+vGQMx@8@l&g@X`!>WJyX#>v(It$qZ4oQ?+>N|^=tdXMdP{OP^fJB|^621FHB8_Cg*xbugAo&o;?jhU^;f0I zQJM6;`(ehTYc6w2eG4oPTta24FEHX7b9iT;fDNo#125c9!kfygWOGbBv-7JyioV%K ze2%NoWh<7j$KRGSx(}SGr*#~NPx>oN(znDs#s@nk?Jz%c6FoNg0WRBE!`w(+c2kfS zz8;Z+e)BA*x9KU)vhZfFd?g^SAqBS=Y@~&IUeRR@p(wRGpX!8!!!;jdLv9WeS0VdGQU`97_~V=V>*&83yqiT!31@owagtY5Ij6O$^kcUY zeKy*jWGc(4PYJvzvK7uB2e{$uRbg?2zR(BS~y}XGO%$e8r_)5Xz?v zvF|-@5IdPTvbXImQI5`JRqs}ipqESe`Dzz^e=CNH?4OSflUAXq*+m@r9?m%po#dn~ zcvXi~0M%g4eq!@Z5A2HENclYDYIb^73Xaw>QzDaij3uUY+d?skh1{Ca{j@rPnLdnXet_Cig0OqICh;ojn8p6!p_tdOpHECnpnAlU(w{ z@jU&xSQ77~=fd7leKc)zw;c3|BR)m3c>KiI|8e%F;Z(hE*uTt?44KCwQ%T0eUiWRN zltvLkC7CLvdHAM@B4ZgEh$NH@Nm9vP*R_Puh)PtXkyIk3IsMn~_5V4Z7tcP9{ca!j z+Iw4T-S>5#=jX(rJqJ-VM+vO5y4d$h()9U6f7a`-4TdK>6OF`9D&nR@w-s$t>?=WaoRCdreDVe_z9qmxnJvxGff)5J#dYnH9?XULIuC93GRlW}ngrNPd5 z#4ucwbIaDn_kXTpZc7ym@3N({bLY||&j{S1nFVgY?dTSR0{Z9LIR4(<0g_kN(C}qP zz|4?vmRlqp&87KiG)OxRV+c7vwvVh{vkK;GEu!7E&Sd*%11kP&CXi+O>9jk%m-3pB zSRGa*Y|0uMWgmu%lrkW$>JaNa#fHYT>}Nal8p$rbSaQ}(k$Rt5iDfr?Q0ue=_h6L* zr#yRf&6v$Mxuu&bxZj)fs4i8+IN3aQFpoJ!GKl4PNf*!1Yb}%&ok@r_mQ*+dz|!E6pR0~VARSz_;!>g^T9Wm9#oB`)*SC`-LVgIW@N$Z zfFfM2l7*bhMl4rx;k1q2xc?sCgAVy}6gnicZI*mi;=p6D2t0=Bv+W$Ge`ax$nH>K0 zv89umRax0H2k7zJkFaRUdFEMX6MdD{E4ZXn%bYng$YyQUuCWsc*f?50cRXW*N|EhJYejcyL^C9^L}kRzUBaQnh3n6mC7 zB%kAXyv|byKb%5ym9c1fX$@wN{!RSe4)MRc1dWO97P|6Y=+8?UKtWWL&NN%<7z--2 zN6MP}P`Ma4JLb^+l5c6?ENQ&+y#;h@b?EywPf3^c5jcW3sMXAL@>qHZm`A%Xa(x#r zKd#EGIl)|4-1`UEszF^m z+AfYOyN&2j-ZP?Pw*_j}@x9Wl6aI`=!^^$xq%^1%z6LizrOX*(lI2G)4J42wnTCRg z&v!Aha#19s_Zj8{@cy$0)2R8o|1e9mhkV(hjMGxo9GN$ocq!VKE_;zBfrFaa+8Jwsb5$%W|Njt-NUa^9~qyAqy_{FND8(3Ut??79C8jB9GiE z+10!+`)lPH=5k?}z^>L1OV$MmZUlEg>i0}MVRZ~2pU8ub-}lIxRqj;7u#Vi{uocx; zeuC0bUf8>dMZv8(m|!xOs&=@bl2QcC7#M^w1#fzE;0Lj{bwaMG4LzElqo6&V6%X&h z1C^(tKlKOE%Gd%0=XXNrj#Dfp2yJtKw<9~m8~gJ@YF4b9iLQZ27Xu| zH0mRzNSCa;#b!;XIIG_jtN%L$Pe*!ju}0MOz2^$hX2mWxLzqY|M0>(`XC*WhR>97P zd3Tr|Ff8k_xOx~-2;cd@Uq(M+6s%QJaq>St4<%RUhIqX-Lj&Bq7R zB*=tUE->_0220F-K?tAcFm$TJ-GdhdBgf`4=M6 zkad4;X`fCyUoKKFh;)BXJX3GLvg7Tfarb;`F(VlxzaPcu&(FxrOgF}C;0iNY{vpVB z^)tOwcoul(ambl71E0++2lrDM#JzijG<~#%^Zy#?b@dh3e-try>S;IjkG+6*|C>&b9y z3Oyn|ohw$@pkSO7+uEvuV1QE;DEM6!4m@0t=WD_~fkaxc|Mf4k*AqG3Dhd) zIazn79yBec(&FJvGWo=4lvuV_5Mb?3-Ja~FUb7zHKV@Zhc1r{e-fM+H-*Q<6!#L6w zREA!A$K$$m2TDe2nZKfb=r?aVvwC|ww{&+IewLdJ*(RY})ZQ@8M(QRmH$8`TR$ptRYKuc^`I^F_(O@1a?wU;*@!czWaUxwyko->|;G()qb3`nY7Uj|02h<4z_!Wa*8|Vt-HKWC7JpAVZ($2rd#;hWadvcpW+fv( zeX&sVh!0EavS{)R1=1;!OvO@RH!=DY65&eZ@4iz0{7q zzRz%I({g$wJ_T|<-k}qMms4xz6oAuA(&thH6_Prbb~hUB_4K&*QXox%2iUZKCiKCz zHe5db39z!>RAIe0Dkm&J)92;%_}r~%WB7_p;{Vp))@i6?bC4P6TEofdrnBqr`Oyti zN@30cAKu?qjvJ(AW5*sxx*TqSiJ}uO@hgB|W^ItnGq9>RUqmPS(J=128ckSX4xN#U zQEAP0JZK%mJ$jLcKUOv4A@{R%<<2^a+SZZM)bG-SgN~RRh(v zH;GZ;Gt89PNpH^((ufHPxR>{(#q23Wud7o@-egU-dG%Cw`6?$kUf9dbO0y)P@7sxn z=m1`Ose@ZrEaH69a!5x;3NAEP;I5Aqfk@Sr;5o$_pG%*j_fA}-MWqQe(kYYfbyjxV zFrY*`=8qxj^0SC`ng`vgKuP7A>9BLXG@TZi!*9UEx$>Qh=*~kH+@p|(xMXAkv)8I! zuPWqT!B;V;G&S$%+`!D2*rzXHGSAMLPZQy;T0c`dic=^Ki>FGrbO z<+!DPF{#>-L*;uC@w9d(+`DiRD`(}(Xyy8waB5-$2G*Y70(xZO>_Y`|OiLVN)E01e`fK2-`~=M2{fXS1 zM@dJS3Z0fwPVUQ`!byYeRJwcyk-GGaE#>)-t&Wp9?F2n^JTQo=2d`l8Kg8Ed%F%dY z5gF?_mO5!|g*VFQh25ngWNdj8Go4w7c1DdPWSM}9IZdO!6?;)qJr3^|jYajlo58m{ zi0qH?z@WpwNqlYuE$v~k@fhJmAKP*+-nJNR!(p|WJ$CPX#K^4L?f6e*HT^XEEJ-TP zC0@cbI=>+ZS9_-82G>0p{4th3SLh@bR@dq5v*mbT@j|}C{wDe@wL^U3LC`ng+Rc4vs4`L_?NMJIEjkreixk>bKjY}je{ z)Yz{kVlcR%oraG2E3 z93)VyM`w&VPo9ktusQYn*{i3;xJe5(!bnpAdWLRc+;!7=AJGM3$c6Lw3q6!uWX}l) zuAooUEAVvfWiCE3qK{?r;G|xS6>Ec zy?B65%`l}M!Oe94%_QP|aV!pPDJ5z9p2KF}v7A|)7!es8fiLzCF#%D|ROo1n75W;e z+pri4@pz>) zw{(RlH!1rnmSwNtWaB1qz1!9aqQA}}JLf+q=hvPh`%fG|(K-jR(E2p$skXz1QS)bn;m@kdX$n|Nbi`BU=MZXE8jjwPQ9bpi(!OXz=JMd^uc7F4ErHx_2;Q;+zI%#({L z;4`ol16PmZil3Zj?{7+=I}aYFORUc$TWG^I%cyblR^4QaU(RBVt@pww7s83>=}K^3 zwHv+Vy5RojauCU6$kawV$9Fupz1$&{w1=t_n{F2xc$W}W{(n1_&u|)+dt>+lArWaD z>zMPf4t!pf(i!0mw08M=I@_;`KaZ}X(E=pPq()%f#VAJHFdX+hPG)yKcumeEx?w@! zS-U->-00%zTWIm83EbM*BHU}#;>!1U!b#o-({yhq)gGEk!`GgrANqR*YK9bkdOe1@ zC9(9+>na$0RDt)^R7q{^EAlZXn}#2hrN`B0WBTSeGu%6qG!)Z+?C*ltChHMGSxWi$aAy~;yZObCR2-bL0~3-1|n;wQO%v-$$;-t z>`_d>yZStnS7a78>_yyg z$vg_15*+E?3!ywK?J_-acooflo&fn@VyNMTNUYvqLYhl_aD00lU1~jpROj4>|1|!g z%dcGOyqPj1H6^t0?IKv3kwyGAydWXkcEWT1mr3KDNaldlbn3(-AznOPO5=)5SoaoH zR6bHk4ZUv&+ET^1K?7H=E^jrrVzCaFbk>OK-Ibw@ipR+1x5KouuxITuG&ErU0Z5ZWCd&v*)msGg0 ziKY%2Ii?icr9MVBRJK%?Hsm~^_HVnmFrOJ*tJeaqApaSaO7CR0oa(~=g2QRyk0Vs_ zqZwTmf1D2F{pV0L=_HkmDWTU}^eKDD8E2UG67BKhg(u3R@rmgZG{bRF9&hC+L&hLjR0!!?t@-n2 z9H$|xfs>|aIIf(WNplt6IQ*SkjNWQL$;F>uG(Pq+(`V;T-OnO5tQsapReALN(?+_n zbR9{W=}Fcf*@@C|m(bw%PvWoH2cz!4#d9~xsFsW+b^9*v7;`I?UJ5U!1)eR$Ys@L; zr_@p|_*@DX_2e>n>%?boE)AjBz#aIS|D2y88Kc{#vv@@Nlz>?1b2FFL)5HbJ)GNRp z*S?rYuQzOm*SW88aE{{E`{rDJiRbo@L&2O8}d33+&(eS~8?x6!_4WxAr?(9v$* zYf?Y<1Ievh&lN|^XBAaa$?5GGj`_cOsQ;BOcy@dWhb@DobZR}BAt<1acF&~8+(Ot` zHz}^Bh%z_+zG33um;rpc3{9&?2{I=*aXBt7G(FtM5zKl5|EHBsJimL_q0H~A)=SkS;79Z8~_rWl~&=J8yrPdYZesbO57 zpWTnPsqiy|g;Tq)YM9-(iiE~hJK1^4c5I9^C~1jp2y3w*mBKqPG?|0wXxm7b4H7~kj3ekS|&%AJxSuGa_!*q*@Wa89Oc+G`tZxd zj7mS+j)D<6zPxBG-T&N$yWYZQbpUd&m##eav_GLD;s)=k7cyn63E-uouj2+W@lN&O6Pn~CJ@qB=(j>04l64F{o zpDGt~B_Gd{K{=lPw)iI-Gp2#6TIkbWQDat5M}c_H&L+Q}xp9Brb8x-ZoSqcU;Uu@; zAuEl66%Aa(SwbN=@0DeaeCwgw;VjpBS`yQuNy zom|HLQ&cUdkTUAO@$N=1y6o&3V$kWvy*i30Z?}z}+h4$WJu&BW1u?YObYfB)Q&2Gh=W%sdoAsR>OY8){ z+4s0_?XHfJttxcYb~mo@V=hh&R}dV3ahLP-oyc9B{RQ+IXK)px+VFSEY`Q;Pfm8ft zLpNqR&~WA|w^+)EHbqyEK-|JD=zFJ$oiqk6YzbZ0*l*s>{?(@XNMMR{fMgD&eRS`xcaC6Y=HQPg`1_y2poN8%p zYj0;`Yh&Z;GF`;)|M9`Ujnn^*At@)rhks*{|2_Y2$1wf>a18dA)A?ht`F}cwiitj) z=RpTwBfl! zTQw=}i`K(e=jOmt^##n*bsBU`+jaVP^&0qPn#RuYJ4oV=s*q{%he%4F7K{uZq9?Mx zkao?#jK5|QaA$2vWaUv>bnr4cVeSX#Zw`YedrjEfxP(^T^`+WHQ)$SH8FaeCAMzx* zndqsfGqbAaGWN+itW{tz^WaOSP*gb|a@0Qv1POAmFWwe>`BLyiT`{cr^9CL@j>Q$T zM2X8XZ(0+-o;+w)z|Y?o;#z)>XW}J`_vD9(*Zg~=Xvb$VsiGO$_j(|+Bogc_6Jc6L zAG`a^amK23I~_lbL4S7>e%8Xzmb8C>SXYd&+2S`LCo`C0e^WX$HyuoMukq{t3BuzQ zdtgJWJ-F2tGIdYn=?UJy{?A1l0xyZ7;eZtW?Te$&47R;;hpNj+cHd*7mpqy| z8MGKS?mkcGgg~lRI+5G0r;V={U0~E~o-#H=XBbmH$G7Q35vXpxN*=EW055qUV=s*d zx$J|O?ik7GJ{{oIo)*&g*`E2#pNk#RUr5+rImjeB>;>C7$f`Vn(OV>W7U4Tc-D=Ky zt8rw{_H$&ib0za9M!>#J)umt7USKzczQmK}%iwLr447~64cT|vxKuU`O8i=w2Q8IE zdhT83i*zB5sXYKE`7B*WL@oI~-^Ssy^JEoJ{c`L0RcUDz%A(+0KKCtrhSy>@J8G>_qj%V>qqe7yJ19&42HT zVcVHcP<2BD4f@YO{PGBJ=4lm|2EVYoWP{o03*NLs%$ziM$fI@kKKg2U6A7YaB*{9M zE??8kRRq0>7FuWMo(YHoTU{BP}6N z)9i$z-XDp!6yNM-ql5nQzry?l#pL9iB~*_!ar{JmaPKjBO6Jd@YVVZkR4X;SPA3uJ z{CwEZQx22dPch&BT*nFLwDB}|6V{x)&aUj50u_3*v3W6zz8kbrsy-Mdx}0OD)|_KT zPus;z=z0%ZH#1~7>K17esTPI>x}fZoKs472WvxR7p`+gq_I_!B(Hn=MWvx1P`byE* zn?D4rdyLU4Z%8N>@f4n0jukrCoo2VlD}r0nW#+eYB?gY|W;g#$f?wN~k=yE@Nm9jl z>ZbIU`Epi8z)F{(#B)m4Wfrq9QWxQqYbng%2l{kf+znFprG^}u>r1!)y~K)*pMyT7 z#i0CoBinao3_1FA5(!Tp1^b<|NNK$X*);1b*?KbqR=(iB|G5c-{!PN-9%JtA%Co%J zErhzAzD65G$5S+#3vNBX1u?#pn7yloOxAM>-2x$tZU|?BzT^7Es%&*kG3ztkiI`hU zau3c1(M-<?Ia{grtR1~JcLCswDaI-94&wT^?7b>D4VLdI-yAKnL&*Ed%Idr0~ z5?!XUlXz8aB_A~0NVMHENcue$%Xf%jLg_+G`YX!%&UGb{M;1~0;!W7Cya`SI#o@%H z8PIT|9G1EtWCzsuFjoU&rayLhDZDw(nED1##`$e4jJ{$`7u>o9)AnSuE(;TIVC5m4 zIL8LRhaV+3PuQ^^A3uQ?qQ{AM>RxhvyAIygnuNVirjn9vhQxFAH3$*4v z7)4Fy^5StYrdI(M1+68!KiJTVdqvS$qm(>;lFSxd9~3y6FJKPxIg4G7rej0OC|0+7 z9MO(Zz@%7Fht|8Q!vC}yNu8Y-6)|LqhEEWYx_bz6nm9q}GC#g!?G)@hbqq@$uY~N| z(WJL0MWFWSCD|`If!i5=kedE&q#ts4Jl^s&#);1z%KS)Yy56jzqhp^5E{vOxLOUn2 zOi$UdChiPLS|mmOELzD9tcoUQtPkVCBTXne^)RW_-c58;)Qi%v2!v4w&NeUxwITn85r#D8p{3 zJp_j%+hI?_SLmBI9`B}}X7bw4k%EF)GI_>B*k15}xw1F9di}XdGP1&q$a5YvBVrF3 z8Z)0=xn&#UJ2wOEd&STuszvx|d^!BpngKSxKfv9nlL+5sn;PtG}k zy)56`-Dv{b9`SxN?Jr=Sm5FI>BUo_hC{y!jH5^)bku|y=O`oPsqq9;RsO;!nnB1<( z1+5;zin=jqw)`fWUh@UEO(^4*)F*Q<9OF>&-a*J-5QW3zF2R1iMoi_eFO6z}?B{RI z#4>9aaW<2t<9PqY>AGV~xF{d|Yz-pzlNK>5f31m9awNO|-E`WwY7PC~*bW_q4_V*P zBIvJvD_kHpinbp#qqA1e1^vJl^6~6Ner;O~0hPmGo5;fRi0xSEH4oM#96;fTCRnSH z$M7C=MucZe9IhY0LKPKmx zHf)*>m=+;AEB5%X;@*}vVV=?R(ycVoVs;lJ{#2Y~ z&br6w1`X`x003#Ka-xqxfnCL2m_zza-FvY^j^LeeLoTl=7kn?kDDxe z`@A&WhStFSjV8Zsjloo_p7<^vU)|ichveEUf`nX6kh#BwmF#a}^<|65$N7)h+SoVj zZnHP+SnmVm#msu)#pv7YCsR+VVqoj=te@xY;B?yf&;q631PHtC$rI!G29PhJha-t1 z*b;G{t@BS2GFRU~T!tEbA5%@v`aC0k#bt0;(TVY}DrR=dpQS+Gk%$g!`skISz<1aP z7AxEWtw|do%)17D4lbos}1+9WIQK!`aYU%oCr_+i)l&LdNOz8IpRF` z1@k*149}~X;Pbb;ak5T48svr$RXs~KY@klK-1IvMNf#2?eNQ+R2`SvO8C?h0MhKMGC@@=@hOkx&>>k4e*<%GpD=U71zZbVgvXt z#)p0<={twpv^-yiY-!)my8ie;ddE*BpJO&K@9a(zgCCuk`TH~(P$?kyc3mXm-`k+F zt`f(O3l?--?qGHD9a!VTZqfyRO*tG*zA-cAjw8iuqnIyKVsKLopCb{=U}Ig2>C)+C zY$|^)I<7H-#C`UR?vf`&;(k7yoHm)9@s(#ABbw1!lR>Qu`LONjKX$46I9T}N5Bavk znCNjTtX-)D#%}i}Gc@$!$!-BX8&QpJLMMFtWJb#gE3len1TOBD zF#SLy^R#j=j_ZzxE1_L5C+|Gd_*|9roGFJn%PQfkjsl!1qj+>>1{}7ohCBQ|EO6g- zMrqaohZ={8pgV34`mNHzcdPSI_O&FeH9ZVwL6$HacoB-ONC*ZPFHjq+U7eKF&sb+y zv%B8+l7e_+D7*EI4JbwOX3R1&_nRK^@cS&Zzc~puTTCI@ZJU|!!BW<>>?FgpD2V%e zMWSFS0tJvmhSXe0fyE8BC_RVx%jL7#pD(j2VI?4&u#4$(+0U*UrG>Y8XL3sye8kOG z-@$#wbLg9r$9ipz$J+WS_)6{B= z+IvaT;~Asq?F==Ps|kh3Q^(oSOOG+jQ_c#nSiK|l{mo3>yioGUV3#0n;Y3o@stBX* z@%?8nmcwKF5Ex2rWE>8M!?VBHP!n~S-FjbyysNUZT_XwV?5+>KJ(sccg*?tP z`Gbqc{)4H9bP#r_Gl@M{D1?|p+o@4x#Qh^VDj5d$uWzsiOs27cul}>=eoKS@0ugvo zGYY=M1+qJerKoJw3HF4<4WZ+eTCijj$|%tgV~LD@}bSYhN9_-cbt=j)t)L zRSU_|23yG85DMBdVh(YOf5ClKQL5P?pjNR)^hDcTwzYc+>|3;ei5q63qFw4JLxds@e5LGEW5V?dosZEdA{IcKh0yaDT8Jp|?BMIhDihh5(Gk_o;wg)S@HxMs9@Wo+Gd%SVh8armV3%k1HxYWa|(V;*Wm+dTpzcRDA zhu%VrdYFk`b%m^aM;GgRDVFTEXJ}K|abo$n!yz*5A=yy7i}Va+2`^t0&^7a}l5sV! zi1eoeMB|P$*E3};c!lmDkHW`8FTa6f!<+y*gEi^9C zq%DSPgvtw*Nxzy3_ZoC(w_9GW1uBE_R-*CpmL0 z*#}A|;n|D7!sSWvq-lC1ocA$e{m2}gKJx?fY+Wlc?#p4Xi=8I3x2`5~JH(*g!-g0- z+rWh<^`Ja?A<;MPWbQNs5_yAJcw;aF^c!Tb-+nwy-dey!%)CKXwX3t~|1#ivlPPRd zx978_P7wO)F5{38PM+7_gWDZha6kSVoVSYOrYDY~oi{D{-qR4cZF303HF8jC*G5*o zA(m&dFQMtu^XZJ>7`lH%6%I643RhK|G8aBCA?XLdvPr8dcr7>uUi6;ClDBE7tuqV~ z!_knnncz`gqx<(=4G%nUsIE^p!%lk<{5MmBte;fO_eGiF{Xu8i)alRpb*j-)|2}$4 z;Um1#$N}(6z%3yeBuslXG>*E;`UNbc0qC8Mk3bM@dx_~vOYlcd91%TnoP4Vo z1#eeuKw|- zyO%iR78Np1`Wn=XpIx@RJ4Oe}zd`>)4g5kS@y(wIoT{pbXFUGHLix8)Rgw-9l9kEw zQ7w>BZUNO&TfjR&2w%^N;G~vQpw3?}cS$}bW*RZ%*fc-d^owW0br?~tL#c2pH4A1} z9fvNN3`TFLpJ*oqvRO;M!FI#N?aA;h=yXmXlIY_fI0z zx*?oQGCTphuI`7k>k8T4gX=+(?;_MY7Q+4%brIx^O~RS;_vm>O7x;yCPpe&N&mI*@qlbA8?Cx#hEC*>?YIoA&rzxxI{Ksdf|uaSQMO_ zM?1nKN$Wp*@<#G92{w#l3yzet%~MUNR%|I0U&_O5S#``8{{{w=#bHb9Gh*Dbj!r)? z&hc)?3#t+{nvqEj_?eB( ztImN(t_uhwlgQj|3A$&)YTC)`mD4Iu5Mw)Env>s!W2U=eOZ_`|Sg((_4~@lr=ORE} zs*0Rjl}eWM%P=R5>d?hN8(t+uF)MN;@$R8sVcV-8Oq#7dIlr@`dV^USTgvky)-F8) z4;H7e%2Whh!o6`?A7PU1(!i$Y4C28j_-FD3@bBcW%cHv4BqIqt-BJNpHU#oLe@H%> zr;?w4B?`dqtp>kPH;`$;~JIu#M1}|3^6^g@h z?GboISHRE6 zlLx<#5C=I`^p{u(uU(!(M;V8Ye;BbbPL#aW8;53NJHbEH06k(nDW5r{WmY0s#71&U zc{cpgfH~--JBjdZZs@W$9VfiYrhRW!sdcv&F?3DCuJ_S!HR>Zt&GR7B>hsvlljGP$ z<9kt5b0bCtVtEqf3B zQF3%!E~VGj?1KT@N6^Omn>NYsgWtTT?8UapOsD5#s7v%_qWO9fFqFevgR>E*4dG}p zUzA$39K!hCnyNoXcptP9N;>7kW9LIKFaI|zbzt#N_!e+eZ3Vrzi^mNGLoj z2{)FH0jE!@!q5_XnlkoyqWWWnYqUd0w%DabH|F7 z@V)Sv<#U*S!}`SCtDZcnFc9hv?IvOSGlU1uQxsi)2P#8{Vav_kSXy_7%nBB#+hls! zo{n@_&hsh%27CnD2M?L#{NE0jPTVBZ)(4O;6Xa>s>nHfE;xyf1(n0206q24zZj_Y1 zA<_?)Q#ZNMK+?t%?X4Pg)c9VqsO&OhR40MqPrX^%c!dluQKu`P#IWW0LRj+B1|NiP zhxB<8c#!u+c2A$pS@#{leB}pZwDW(cuq_dqj~L<$@mAsGxOk|)sljDjN`jmK8{y-C zP-cq7bw;l5HnV5Zi|Wgp-w~@{8q_h<6jf#<6F0p&CfD;RYZg#SZf&Jx)>I!D0ToWEq2&0Z z*k0yD^$+$llko=nTX=x-pcu2jeJPt;@E+S2#>4N#LHl3&=iu+i3^*BQj~|>~k<7~* znP`o4a@)6x*PxQ%#aa!%`!pNnx35Dwdl)9?Z-&@?+OS)_irx3hm}Zexw5u(j_RaeX z1IO7 zq#c5bu1!Qo+7BIxBE)iBqp&KbfYsf64{P8hsC4hfr`y7yl4tIzIe4OHlsnlp*_>Sc ztj>mv+K&G`l|cPV2cOp}CHs7n;BAl*aXDr{xIdDt)q@FSp2`Kmj$f8!=ei;?|K^Bb zjmNU;vZ`Cmc$c&IeY^=*b@C_<{MSj!>ob|d{5sr5N{M@!Zh+sbm*Kk`>xh1xI(t(6 z0u#J88PjL10F@nAK*swMST;-35oQ{BAZtx(7%`Z*_-nOL`2yqoWk}d2DuXeSl63v7 zZeqUaFQcn&N&QcTlPUi_CNjkjnBj64FrWAht~P#R?l+2(*eUP$?5YTcY%Yclxdk9; zJOD?#C8@62ZX(%n1WT;>`Qb5DOt|QSL4hKqN&FTWec=Seew{-5wm4zK_;66vmPC~$ z7jVrsJ^`a^PxooR2H&x#g-(W1#3(_LPD#%u^P3XLpiUC$;JKm)(s&m5MOQ*U0$C-l zOri=D$(onFf|noGL(JO(to(BtGOnM2?Wb93F zxTgt+KL^94T~l$V(^q2gqgoKivZ(KIY6j$Jvt81kp83H3MIanm&F zz$3kw6}_j7>v)aszu&8&_2zB%@n&4T82=18kkYCH_Qu1>y&q~uH>W7NqZcsl|SeCLrkip2d z9fGlU@}Tx$GE{dok>goEnc;C$V4~D#64$MQvnst|zg-|KFX_Yy1o zEd?QbmvD377W$v%aWeUECeOeZ$9)k=Xwgx@$meatnzs^o$g>BYEtrLkXLvq(8Lt~$ z)yCUblE5JUI=QDR#=c|z!G+KeoT_w=irha=J8cfp6*0w3@<(yuZ2vqnK=HnrDmK&h#>oE8n4h z?Nsh*>n}9Xj|awO2gUMe>ZMeQ`j1t}T6&ibdPsq2atpXF?+4wIAaXY00g>9?2CZM7 zkfPa%WMiZgeV->rZ$-@`RpV#i&ro~3A3q&i?u{^?uk-UzYcG<~^OY%ly^p&nBII+- z>G0K(?~m^*63T@xqqj2=54-1LbYuf=irNloZkO1*s}jkQ1Boc@d<~+kZ{l3(7UtF5 z6X5cFKfHUp5F?C>$c=!d{A|7m>uj=#88?dV|0zKSMu{P`bz!V}1(-Ex6Q8O~x|8*1 zW;tBIAe$1*aZ8~F8b$P_ha$PVrkk0b0>b}H^VwIW8Z^mSoyvH5#o5xqWa4)fiP9S(|Jzug+Jd zH0LOpnwtnoQ$AqYnMd@f<3Z{;V=CQvFo}8h<2HM5xP-(k(uHK>CxW2qRpi^?F?RU9 z9~|y2V+FG{a4;gDWn<=(vRTLB=-R!)t)qm@faD<5OFn{kS^;oz%M@&|3THeNHw*Q4 z#gX9aP0UK=L1;z*^e`ni|0{N&U&DB&V=ruaGXz!3TbaZ9{~tx?9f;NYhVf*tl#vx_ zn31fE_uP+AQWDZ4i6rfBr9q{PO3^?@BorDdNkrar-*3{6(x9!Op`|6I!teb4eE&b^ zyw7vr*Y)`fg>}A%P%^WQA9-^Mf3Bs8_sG2k)&~u^!@|tjZI3kCecFkOCfQ)Sx8Sjf z-Gk-r!L*|?7-Y_zg+o)0!rWXh*#B4`uC(lc_9|gsQ1%T2eCx$4t+#R~?N#yP${c*; zY7b+4Ho)WAg4?l*tgJ{&Tt24 z;;t@m`&bJu6`!Fk$e!&=6guu#jhXX-PX1c9HH(VOp_=_CDc>f6VuB)|l#|cq}NtTY^rRy<&r=s}MCii>-gB43|{zbG6%_3B2*g{H)9M z+?IVOz;&h%xSkU-6SvQD#=pjq#Z@yw-mMn!`IgT4^R7vT~W~|=m z!9U+UkeUySq@C|B;0nnIm~^^^yZ%&$rhYcW9M==L^u<-KpgaY}=C9^zJU;Q!8_$bA zZ?lKW^J_4_VlSv?2zR-cVHEnwg2sx#Hs?&|`1~&fN>c)P{E$`)dSc3{DqV zHFo&6Vj2FIHUc*-5_&v-shD#ynR+`$(Ytq}aE4wU_F4eF9u_aU@jL;3-Z%xJKSl{> z?LjbNW)o+&<2zi<=*L!jY^A891DTrcHQI7_E1899Nd7r6%)hi6e+BM^6+U9l&as_0 zw{V5$tvs614&M1s0K7Y#2cbpktXF#++mwA9<+-Oqe!7?+6*UFYnp^R&$99O2F#?0G zJ-E0w83!6=!|1D)?4oogZ1MSulO*++r+EP9?cwo{K^HpZThXbLmqC#JK-LyJ!4Zi3 z?^i?lCGQ@A-qGQZzjY1J1v%D{(}T(zZ1L*&H!d-`UJ`ki8<^Q&3##f~z}TsaMsRvn{fZ9u19E3o4Oy5R+PjkE2Pq7lD;Q_s#sQQMZ|&M)K!q0bt? zV3QQOe&zu2+9@#1Cy{%4OJIV1e!&^&AHib_zw!=ogT>+IqgZFW1q&He1)jH4>AX!4 zZo77p>whPiG#W37F0GnN2igoUDOHPl7$?b+--3uM#ID6*AP(;-aAOpID@h*u8& z0B?ySXvSH?mo=tzlk32jNlBc+^R;}+igO^R^8=KY>?QREbAcPAN$xHCXjXq)sybca zykPe}T)&`76!lDw>c)4%?5hPZ=jJ&a2OHqO1rK1Ql^=I}Zw$k0B9>ai7zBniJXdZoyRyFFqA1w zuA^j!NBHIaKX%{N04n#nf&R%>R1JMY-w$cvr4iW>Yp|ag%pL-*&ON+CN+PWOHwqX2 zO`yKPMleC(5LzW{6g<)?IQnH#?dOaSplWDmw1`$KHc`kCk|xWJ9E1 zPq|~6egaeYHq2zdxJ&EQxM~?=3O|32jj6QeH=h?W9R*|flx3|jGyWq?Ip9vsqs{32 ztsD3!Vkb<0-Y8@xTm&EI9n2W@ALkb_i0G$-#P#+V`YkX>LKD4Ro=jJRh1Z?v^6bfU zM`jUzC~ksRhVqh@NICY|yNMMI{syv9G4SSTGZmT|Q=XY6KJ0e@w9*r61zlOx#< zSWdJQp>k_JSkKDE*H!Pi1{Evlm=Y&B_n*JyQdx z7o|TsKndLgzx>na8*vU7YdEmZ+a_ zoa>B75SVENV0M<@to07I7A?f+zgl!}+F9Jv?<)o#|B7Mv1UB)kBbZ)X18y&?z<$s% zoW3lRChG3zuTF^PmZb~Kbv-k9Vqqlly}bc_)|;YHX%-zw`2r0y1Bo9bjZuC+)c^hm z_}w*|y3z+Q)z*oi?cWTo%Ox&~> z)4HF+zV=2a&i*MHd1*XXHlPCJgr8C@Xs+TzLR z>}clYISekoKZX-RN~q@7NUEq%VaK0da}HUMAYQX&AKq8qPaA*M(6rVh6#TY6#YbCR=+g9dsEYIz1v?C*?&+)e$LogkGa^2tO>{A~I_}2DbJUqiQzYGU zw1%JEGiXwE#wZD*Wl#`3xS1S z1*d>PG(>PdnyP8B&tjn)vvm`D>s!V8mKCG_gB=u5C@(3?b`yG#N61j70)DM1h4=GY zV3Xksik@ytZ=CzP-Ww`-YM(XZu~Xyd#jWEMu|)yLhWCS%>1r4=<0!vbm^=H^PPREw zxTmu$cr?FJc(({u+;7sR3I+D(=N|A|YfHYTwlSRwC7hXSO((OG$vtHW{&#r{8|^Pk zZ;do1rC!b`>USQ0OK-#N_=~GLVNMyJU+_Ay56QaJoP9Yo6;j2k__A{&@pkl5wsEsH zO1J8QCeJPCvP2jDZieoS8I6I*mXK;2VTWYL`pfysZl zn*Hy=deD5xDnXpP_IRoD_R`XRx!7OIUZWg>x$Y}ce((OO4_`b3qt3y|2=RH-T z{C<~YC|J^+{WW-N=2kcpm(AKa3$U0t6ZgcAl?;q{N3J33CExzOrp-<-L3+(rF2Xzu zzp9Vn?;n1O7SYMHRA8#)&)0;0l|SjgXJ5&ie#6+CE8R>nd_NsBT!#CPuVZ!X4jd1M z`KWCTqG>_0jKu=3DW3$^;VJac<`y@@VLUTFVnJQuW~A)u$_nv(p5)Y<)EbG&g$Xqv;IKL3S^_&Y`d}cx3vs~f3aF(hs z&;&2l96XjOD-IjE4jxEtz~0_Vm=a-)=PHk2wN4AxO|qcKn|IMG(hNSane;I24%7vX z1^3#uxZ1^&e{dj1a(hlD`Ml}lrH4pUz1~-x!Ih)syjyVHK87~8Ze?qt=d!L<`=EmF z==T2x!I5>HHQuF@Me>RT>ZkbQE)NrJ4h-D`Tc$O@J_S9j#(tj zT7>g--h-XB@|tEiShl-1&|Zu38yf|F_ho)~X#rVCMpD$~rL1{N4j$?mBgyN!f>vJ< zoTJ>?>{%rcXbCvxdLe)7!(MjpwUn#>!q37vAsOxTc7dZo3J#Ka4Ns=o&bLi6vl={o({{`0oX`sOY$yIcYRb|%#GaVws6$mBmcwsH|} zK_n8mA(*$2eNjKpil1fC+2?WebFntla;s+BjdrmOSr(GjNkd_#q?VTrdkrQX#_Yzk zi}>G(MWi@#Hx6lk!2j+45H!R0(8EiUP^V0u#g$Elb#67Z>z#sRp|>+Pv*r+7*nfqi z$Szp6cp$`w>?4`)p|rd9gy_C>HTSIW5O(`d2F1L2P;q}Qtn;j)Y}rP-IMAJC(MoJGllg@q&rOmuY0a%W{JM2 zJwqK$pLb#DrANTlFUPO94dFrPc!|!ewbXkv9F#ofOAf4-k!<>Ghtpfq!1V82C@wIN zEb1A^_~y}UScf6Y>_3T=ekIc8w4GGDeksi0n!qA0nN1!OBKo&%APzp~K=#i!uxSoI zfSK#F#geyZJ#;;&M~PtYcY!nEI3Al8e-Rmfh=Tf9d8+;6MyrIZ=y-vHG~#?4UYh7E zQKb^K_VaEBjxLKiFf0TiLLpPB~U!2DHqT=`uKe%-nv z=2xe~{Px;l=B=$*$H}wNqbATUy96q?j-w|Z55R2clc>-TMbnR;pq2}nXnd#=j>?$P zFaV}D}pMn0X=4wmLC1*yD<9<~evEZMFxR&p60&Aj$ zTi)*t)O@IdpVseTYWoMcn6eo%mFv-_^(kx_<;~lll9&8*55yOb=96vSL6nUjC))KY z0tY;;fGbDKxg)2mLAg|$X0MzAmMREEiKijv!6X>ELWzBuFp>GTET%O-^d*D8-Q@e9 zFQJSL1*9q-D;XWWoxJ8GP(Z+T%F;Xz&vwsb;op*(Mo+6KNLLxz^m*J_4`o#N0u*fB zfHj+f5v-PyPUa{I_m;4yCOXU_O&6wk@;Kx}AZVO_#$Oa>QaQ&>+4!DCaOQw8KfkI( z2c@O)yw+u~zov(=HKT=pgPknH_yKFKbZ{LnI3LR5b`Tw&M0XTS$a~dFnj`$|De{6( zVQ0884<^D_@}mpSCe!VccJ#(NU%2;G!qydo*`~kCDe9>aIvjInu^!_n>tq8d`>dkq zlu+u7aG@P>!O1v+^=@<{FW{4 zR2tK_6WYi%JV%+&pK$D&t9)hadXeYxX>h2l1uiNjfyW+UZnFKD*lk(^oJhUJ&ztRw zTk}qF$6^wpROoU}epLmd|I8$Vzw>FvuJ`zebpSS~v(sH4_!8x(IODWC%-fp9YE89a zR{me{VWYA5RoD-@Z8*TM30fhJx;2|qFjK~_^L#MzaJWQg`T)s&+5kpIDiW7b%6!?F zNuu$h>tGjh02|-PQ)QegckIUy7_g%VRJKJ)rua_bpX+pTbC#P>^kF%8(SDeXc;~?U zJD-Y*=N3?IdninH4#VQ^u_PT?2Q8rm;B3vCI{k6I7M+wDPjl6LLR+QpQrx)nn| zO@vd+oY5}w0+o*6fD45D=WE|q^g30DGhgcpbFtsJG_M=;dOzd0oYkmj(S>_;`=ej+ zF5EWM39J9>#7k{~+^LPjQE$o?lD%L@VWlr|W9>AWeNAA(uF!W0`Za*wJx4x^O!~v$+`H>>I;TG-i9k?+<=#NYGAzoLaaQf#Kku# zD62+v1*I-dxCC zZHdbLeUxr=8`};IVK>$d7dt&3NAE^%ldPXTM{+?pW6${Vk0etPY42en%d}J<9~Ack zaetwoF=sUujIF_Yl`ewkc@)8r$&@zhH7E{P&PAO$!a6ouvCwI{+`tw~c$i+p=@gHm zxv|kSWrikA2wn$n>2uNI&qe67>=1HfFZr?8+pz9>uDEvaZtybQ3p;&1a9YfGSiG;A zuiDZBx4;$$93H~C&0ByiA@(>KXX2{W-%!IZ5=T2=P<;72yzQw)6beuu>Dw!^SgR6N=?+KHLKciXrBs&2)+m3 z=jJs2-zmIzFN%_WzTi?tebC(+%QkD+Ktq^1*I3`l{nYQ`Pnc~-&kHROacv?m_b3BB z`V0Few{|F7)C?t;rCF|?CL3_TO7Ke|xBA;}{PgJzZdUf9`(sr|CUpP^a2{cw=g+>) zt>de0BVn_X?(vcEh91;~-@?rz$?eJryvISFfCEyRZQ$BFadIbYn6{3~9IZzW>s4t}OFJ%@5+n&8ts_}pmPbXw zNpNUJBfN;;$1FZ5OMbX)ps`)MNoSNLYoGH8;sSteQwgK%h9a6-=SRjL;~{*RF4w1g z6;7ItCZo4Px5~+dR(foM1)4UnSuq_V)YK%7xlZJE)tT&a_3(_~NA~^Vg(o(*iIpdL zico(DUSGQd{#Cw()pujrjo^cj6`;V;8$IgR-vOq`8qr~wp@Q!?90e*bCf7%B_s;C( z?PQKX*uHv5RM{kC7W6SxX#_s{jNI=cX&U}?44KQUfx_f2?v~jH@!!8UVJhW`YU{^~ z$E2D;pun}z6PUtzCJu05krualSv|6DVnxB7d%_Cinl z5NIsPfz>^7Y|KV|STA%~Umr_B?=?EqAv2U#D5=o)|9+5HtheMx;dYD@_(=i&>tRt$ z3oZ;;Oz&@AhJPcRSdhs+I67=S8`*3O4;Qxb1JZ}Xhr1%wjGPO*Ru5vm{g!iQZdGEP zzztWkIFGVphoXDUW8SSu4KJS5roiFaG$!c+s}pi!mCmi~V)sgZ&uRzg^7#VmGR-Ax zAI+dQ7MZk9w1r(rOG3RBV(|Ma1!cqJ^_5(7cmcbSJzA&mL?6Gx~X zghkKpLtCo{9@Gsc?VIK}XYe&H`S}mXT)xdYrqziyONy}lMKMY?2+nmMW3Io?3Eulp z3I)G@0ru;wY5PDkYP6fktRs@xzptz4@1&s;nf6lfS#p!T2$7=i#_n+TOg=Zc{WN;H z7;`gqzwrBV9hrRd1lI7X0!BKt;p4&O?A_Y&w9We%@3BsU^)!TG{b_rOIz64PtpI+y zIE&arY3!bt0&%aV^3f4({Pk0%)Vn;01?SYG_}pFW$vgy_WkZ?6{4T6so5UtqPGyT9 zWiY3aa{T`4FL*=tBNa4U<~{SRVNT{cX4&=!pYC;}^QvEQh38iIF9i94t=5z$pGeOS ztOdJyzrb^rG1&w>#my7WQc5Lv2Kb*S3ds_W5M{-Q|rKI<-Ha(m_fGz)>%hx65(Wkab`jx+irbqte zJWTqte~HuC?0z)}bz)52eT@Docj4HtahN{tAh;b{!1pfMg6_xjXrPu3E{-*(I~9o_ z^{Q38AXtimc6~s#_g}abhUs8%^b}@WAHexZrX>4$9G%@SOQYu{F!Mda{7=0ZBi81? z?!;s``a+din#!@N*QZ%nuRO*TOS78|+dzNQ1lZ?)1g=?`QOm?A)>*BCFZ;<66K9h* z^W{}9e#X6@Wl4T?8ScN;iPjFU!0Kuo2bm>8SIURp1T&IanvY4pE^z=<96q#8r~!3Rd8ZF~Elj1I!#A;``cs%{iW&PHcbwK=Zl@9M_OMN62Ag)ng{`eM zW!?^ke6FH9j{8rYKaw90f0GShT&M!Odu1|ps`YZ-5)P-`d4XMyrfkxE4Yq1@B&hwc z!Y&0fDEFvGb; zh2{&{wPBeoY0pP4eD+N!pD~jU(BCd_VMpNNf^B%&tpZvCwFI{FCGMVqFc-b}noII8 z#5XVeX==;@3aVo8_{3HaU!Q@`bspmU&_wcf?4f>o36w768XxTVg8#V-Y?1sJKBzJO+41<>K~pq*#7vf`rU;wI?B{$g$nrLc zBl(u}vk>*d4Q>qI%dRG+vIToW1x{`VdI>v~YiL zf9O?L1MIdhIJb>u1Ln7)yYwBryW5BkF5k}wE;SGs)*Ot}_r?iwMVRsQGzK;l^9keS zBvb8AQZHCT%j}ojo^@LAZ%rhwYcIk`^~o?VUY!{Rc0#T34XCTVC_LYmnBAt1n%R+( zf=!obntcZC*E^2u*Poyt({&^rVJYxro{VeQoBzm0$l3&aPa=|=g%3~5&_icsUi?Rf z4SppBY27uVEfH`8;yjRD~N? zZ-$uBF0{C7Fb)1~3+4}UaHE+PMY>ht&6ZJMx9h6ty^pZ@SZNmmQI(ILZGNt+tK%=xA$?5QG?Wnt7C$*6P65TU0gO=U`5 z{Ir6L+zU%Nvizb&2XAx{m#2W`zLsq6nbr8l=@K3u6ir*A`mo>kkLcf#k0~dmp?K5= zNLziBw}`XnWA^*Awc;#MLRuy@x99MKuUNA^4{p=R-8;bI(mj~E&Yqp=3gd6M??vqc zK`=I2pL)bcc>|yG{Iya4a7${e_`uRJFn4}G{9b+>b`KxP(li6<`iD$1>ny`-#eH#oz4 zaDlV`ayGiD?VvIcp6O|x=zH!Z-l`PbfUBQ#Inkpr`EBRVw0b&uiR2TK zG+EPIbvCj~n=btDqOVrjXunc~2dXS!k(W97xE|&rC)^R2?lHiVfzKfGc{J5qF5yicmF}S3HG(gRf2B;IXTJm%L_e>7lIJD9{8+;>)V*#j zt`J+Iwo$9l5xXmTQQZoX@D}{7Sp~54QSG^m5^No>MPqGjd4riY)Fz(+WBxk;jq`ei zonk7tWGPPz^A?E;6&~-=X*KLpZp$Nc@B!NebiLso(7jFjZEC<)pl#_X~3%sH_~`mme2d?NOtR z0;j_xTpLE7ZvYMN2EM#*A>PWU;;&qPi07?iD1hs8dG$&SGR>Wtt&9nC`l8G_U&_%k zyCH0yTP`b?@1tp*Q%Ux~EL52miU+sl;#TW5xOVjk=67lx=j3HdPbTG%^!*}cTj9gT z2Nyu!4-?7xOP{!>hwosHmml8CD&iCNT7$~Z#aJU`Bdu12iu-GsvSH#ia8|tmat95j z`nqMjM!*sHGFpx)P29;9eyXP_rk-qQqXOQ#*F*uYbm5Ty2=R4ij(@p4VA0rZ~v&y@P9&?&59d`S>l^p2}P1;OI6r+;FiFKbm%kJ41F*#;H*x z4h`U58a2bf&1Rte(Hd8oVE+a8g{wHvVxf5fU#46l5_Lzzd(S+mHot{=>CwE+s5jspkpYtFTC6_b zjM-=C!mwj!_{>6i>Pj``=Nb%RHLE9ZmmQtC148+E>Qg;9dUFfRR@VWsdmI@ad?41| z9?o}FyyiZOeE{7>)Gbn`_zzuZIW2@DC+V||7e3+jz)3W2O+1Qqo#^X?>E!M54n{A& zE!-RQ*v{Jn(Z6aZWoRYSnT}RIGXFWx3mwpoPfo1#^deA7n2+gZ`?a@p+8_dxV2_=uY{^2tT|2wiYb z#^CgsxLV2y=gv4S`XD`4Gx%PRpV z&De(~FRx<2Vtrg$eBAl0ko7tTX6)aSv3S2%mBx)Xr@e37>H7_V)pp#Qsf9ptjVhL)a@ zpk7%z>=5n@CrY+s^Eo{>UuzN*(=RCJdg*0@9nKw*&hJR{!wrhM6lZdnQqu=QYu zJ*9DJOQqO`){$6zUyJ_C8Aj^aw@`dw5XA^t;pI=ya`O${|35E)lih~wU8Xe~xBMOX zh4t_z-C2Tz%#iLZ^P<4w>)20ld`fT12AQ`#;w1;-z<1SK2s>fSB9HiC!@B<5lAdny zighFC{oC^(8!yk|U+=|Xwt0BR!vqZeFg|a%0`Bhfz@QK)#h-|CY?fArMGCfb)vxGcBNr%)5#`d4@w>$g0N6e+~+)lTxz^TGrWuO;Iu@l z^b5wbb7wi3yi2Kl@^CQC+onJ#K4wz&lWO!b%%_{n>+tcs5}cp0m!1j#iAN&h!ED?F zntUOiJZ@G~utF_WhHfN(#SF;0pu@5pPQlC4wd{_b9NwI`55_Ex6B$=t#XDuKw4-w= zMc3|vQvLwE+VNiSS)Ro+A?GDkU+g3sJlElpEGfyzjx*%?PmWqu--_D)g^M4S4CI%E zK7`{RFGIs)Ii@-386Nv0#<9cVFxFrtd;9w&w{x}|eII(1Qk!(CYlIxV%2%WLb6PMU zq)&V+vIuH}g22iu8P$*Nrh{xT6}@c)^CAz(HBI1_ta0J~80>^tfeGF?N}HWhSH_uE zPtbax9W`k$U}l|C%q00EU$QPv{M=_Mtmi(VqDBh8_n9`8M5W`3DxvEw^nT<%AB8pL z6It(&ek>?D559Bp6r?>1xX;^pn}xOD^*RnW3BGuXfW>(9kPf<7BHSLbhQ9ix((jkL z)Rd9{kN2AigRg;*bnzPcs~loaoHbaX;CHCMe?r);oP`tC@3{we0%2U^O}HsJ&6iwM zAg8MeRJ=kN&nXtuPg@iESF;;zLelBqTQ$kqwux-aEM?|9P>MY5K4z>EFRT{wntl_t%ad>B?oY{5U1;#r-3IOGTK z6^&at7Akb=c-y_v5Fo#g4buGsB^F;`M0bZ;T$yzEg`?&-7@B;wY$4w<7n4@965Y zi=3L4;I^5sD)4_^<1~fyyze3JYAX$SuIic+hz69w=+EJ>c&i13$d$pn8`}Kl^~L

Jvyz_B5qh@d5-ADZ3C*>~&{gP2>{IW;$Y66i(071xKl@VhD^--Q9ma-D z*#x$VGvVL!1Qx$1fW4lI@b~x??(y3qZsjq3@_Tv;uU#tSv=^4a2M-6)&O4?!wW@~K z(36*}J)B9;?hWDQRUX4h=^MG3m!FCcKhh(|qd_#ldSh)&vJnNyWU{X(tXa(&9SAt0 z2+6mU;j{ZN$;5%%$@gHfxaA;HO2=Hh_f&ytrMAKLUJ(=h$Yw3Ru_&GXk<-8V4eOk? zf}i|-&is)nn_8M9bZhow-kKk{U-}bmoPcED=>YS-4Q3(ZLvZ4?Z=#BPceuOeDPJYO zl1>)w1Mg8@*s|^z`ZULoga2-He*6VqW#(duz9S!J;J|eRt5a1+Km4Im3zf;j4t=vD z{&1p-IDsp69;CzQ#m~y;6>{ z8iw?iJ4oaCX`J%*JlGof2C`oeSS>H$BlhLM$Z=}aSS!P=in_#4u~eg4*-6}Kfvq3$ z$%8(~{K1nczBEB?2tTLf3}nA@VR=$9-0peP*!>%yxOYEp;JW#*aoa5gT9B#4end~k zw=ou!#7VKYg}E>?rsBMvy&Yw&um}5Dsx0V-4s4hh$U>DgY5e>Q3SK&!{z)8e!zk(3>L@oXPo=^4v*DBKS&Ifp=q~6o(3D+o9TS>xuo}>ar2~~dEjSZEL(Jx{OPM7^c26C2s-z!&0HgaYi z_4Tj_hGNRxYQCtqh||`a4qCno`F_)~DKtjl8Jdj{PhOJF4-n1_vv1DBh``aj+^~yK zU-y!dPF>-b2t2tr*TdO*i$6lPd>CuA+zFPok(jB|i7{T+fbIAXtKxT3)3crY6#o!B zaVQU`nF}2CxY?Nc{4m$HaiipXP!-vNv}F5ke;T>tpJ?c+K)CwH4nmHPVL>H#Ag3si z<(UVOnsaI}U3X->(~L4` z9l3+7R zh8Km-frjMBLNl5>cB(|ZuaQ#i596KwxuPtaBDkdH%Q`*E*t1r_d!Sr_il>vf_bF5O zh!GKZ;>%ySo07(*ob<$t(M@0)7De)!+H_c5hpIL1;@_Ar_-D8V9voxEF0R(2VI3Ff z&8;!y7PtgM(n@g4OkI|IL7ELSp8#g=XYfJiXGm?b2hdSr#riVr+`&lZm>LSrA(Q+yGg!#DDK#rgjxeN5X(ZD?CTY*=-^>yxUzk9^r3B_%YRYt8kuMxFpI@j$9{=#|6bnROPY)*E!jW|I0F~fo~gJu>;kM1deS~&p?KobW%wRt&-)t3V&+AAFgWfHwoMi^_w`v!JZjA9!^AMP zK#@GZy@kC8!@wi$CqBCWf%~Y|82eF}Ct3a$MYdLn<0S^%&jH@_acC@-R`^2A%4g(uLdftro=5Nc zmx5DnCqCduvsKPWZw<>)%4vka1bBe`&6?2rQ#$I+tfU*wSMkzvLz=dMz)AH8{4jpZ zjd=P>3~zo@roMqhv8xwmMCq^}-C0<-uL~XxG=qZ;he4&co*R4H9kliz7Oh<2FU)=% zXn0!+s@(hqXCuSluHRMh;RQvYV^s~?RQ_SwdoO`&)FnFkV+hWlA-D!Cdf>C|Y^I%6 z%O5qXa~98k!u5N6ij2SZr@;lr=#x8`b>B~gkdLZxBY82gDH*hVtrfYs?8Nxefi&UN z3^EVYMK>>B@x#PyaPT~VbCY)CDurq;;K+?OSq(I_}$?)5#6`Vb>@Tt_1H(li>$+Cnw634<}IOyI&pp+|M*RhmZkEhU&;`;atZwXWrXtuCZltl z6gz6Mgc(npz~V=xz<-8Wu>DFobY1+%#mA-MRv%T;{#S=`bPL0VxlxF%E~#)eT$)!J z{#J9Pk5*#wHz%P(Xr}@i0cohKXvumKbeXy1Zl*JR1}!N!qWdQjXo|r%_&Rw6PIWp2 zfrsvjYUGE)+uO2uF(O&eWk^%xy&-}px(qU%=h2Apv-G!UJw`=pVfPXt%UH1lb{W>Rd>CH|~)BDm?sQgpK+wnkUr3aK*G zxYLZwZ%ITGl`i2!;eCNcpYcM`7q;TsWwvT>0lU@Z%n!~r#dV<-d|OO4dS4n2M>@=D zc-3LDuqdI{+6el6MV$sTWW%C;27Io?X;{|%7G91X!_KTx#^LWDW0R1_ZuvS46D%TY zbs9gy$_XA^uc`{H$yGy(!dtNVKrLiF%>%9I=lln08~Rpj&&?Qq1uhu1i{{)jC$q_# zu*Jwp{9t!AZWz?ce+tonVH16^uIC=wKG&x%)2Dc=XBg>9AtraV!mLZ*a8GFhF9L^GMj8*`iG}DcAXo2?$IWF{SjD@C{p!%qN_@=n{q@ z-X26|<=FxcQ;TFb-{OAUO%-xT$65OiQ>N@;LK5$C3`jlACC2h(DdeO!taPTyyC1>F z-ZlK0ZE4)G=uR?`s}`2}gK)vCW@K@Bc-C+kpYTkI8jm~ElB|#DnYfMO(lpqEn1L)J zc0Sv<*_2+7aiJTHLuk}dS3LUiI?l?Th>sgZU=e#754jf8Bj-j^OVN}}4(n#`Qq;+R zyKoN2>BK)+LgxZ%X{wPK8yoNqMjbniKM$1P!Gf?5Uq&6A`XkSqrQ|f%zGUQS332fv_$AvCzs&(-?>mA zb%NHr83Btm!QquZp~dtkoJlx~V$~*aP#Q=N+~dhEWeI)WxDu>|yT5L&Jb&AE0VU~A zfiEWEoKMsr*6)eHnU(ukD_2%eH;#oxWK>U-Lf|&h8^RnO^=Fg(`uMj6h14{s zl$OR7GTYV`Y$;(B>n26-wUWr^@OG+qtOP}w^We8H4Fa#$fU@N?(i*A2?^#g6y$H5p z9}k}g4R{aoOBRCi1R3tf+1t43Ya-sU@nSVMmolB53AEU*4-@qtVBMwltfoDhJsnfc zPHMW+)yh&#ZNAQn)ee(M{s`9fWRc{Wstbr`90i4jwGehvmQC>y<^)xWWcxzs-b$~e zQjb7UvDlEhe~zH{o*~dy(4&x*59U zdOd-fuS}#0wg+F_Hl$UybEqPHKY7^-d{fgHQaJk_^oors#Zq4KdFou4vpk3k+aAt2 zxL${A{fBYemv5ueU9~W$L7yEv_y~rMNTfi?9e&up<5;?CE9yN}rAZsE^MCJa!0uP$ z#g9h55NYK+z}xu;@ruB~Jp8td;yj;{`raa1cWO4&4awzt?tOq-w~@3XI1QVpKF46c zRS*<;4AQ!8;^MvS@OPXCG?zVt;wgz-)X5Nxo~B8wG}hte4Q9-EnJ&5iH<(T7yaCpU z-Ymf7J(TCp2GdQ6VCEP|7i;$7Nvm9}_sjunqhc;eSCKUu=!$ zDnu$pC~1^5XwaZK`+gD{Xf&mSh$!=r>2FAyq{)y58c47am?4rg|_Cam|woesVOQ$yTw|XC82Aib(rW?O7^1}fXY?UCP;W&iXrJ!_0 zB5g6;$^WejBCqB(WM*)IN*o?mgsS9+q@q2t}ere5!2bdj|S}D;x>`7 zElbx<#KL_4vrxNgD>!IgmfYr)dCk@5nNHFr);1v(H$3$O|I3U!(3S&d;pu=B#1we4T%?~Q1_>)^l|KOSi347229$*yQbYn#f|sy-Puf9HsB1Z zP415l-T9z!;xq)5yI`?gI}H6gu%hioiKMkngEi?4!|SS@lBaR?(#u-Oq!Rs#f+TUm z4oMjp?UmM?!$pES8zZ3?Sdw+{%q@#i=41p%Do=D8bpgN49L=8k@jQo z-}W57VT3&D=#GOK`ms=(9S^JI!&$VQChK$cGukZnBi+yO^zZFr^3Fa=v)_ynx>nai zZDb!*Ul_*X`NO=I<#<}QYb2ZVuvwZv-Vt0B=dr<4)lv>yt%Y$;3W+CYf6@f&eO(&dDQkQ5UP2OEsTz0_TMBFwBLjl?(ai~ z{U*b%EiS^Sp)Qm+wheyd)>9&i+yFge$@1{NEJeF7*jWCM29Ggh7p7){!R$wvSgI!+ z$`d(BI~+;h>Nc8IJb=jCmw_xl5sq;c{>J{Ktpf7${br+5c5zJw0 z9p3Vc72j_wY4&4#u_x9A&*dHFd&@W%C5xe>Nm<62y)# z3ZmIf!K9AUXo6BHnMK#*aaC9D*o{)`mER(j79Qt~cqO{Fq>-C&DPA&i3~>e1KX7G{ zXYpHI2VSqOfnQ%6P(x4bOK+5?p25#6x?GY--7pG+J~&WQNe=BbJWFd7QXzBZFF1en zJRfj1l!|mS=zGI{Xw7lJgOVA-knHOioid!wU9gh{?K#0))u&O&)U87N&no(~XbAf| zT;Nh)MdHfzLdjguaJDD&7igG>T-Vae(xn`BCBTq|(r-wPNXm5O0B6ogr>d~V-lD887+ zA8GK%6f-x{)z3hA`{UH>hYu{-k_LLN<&fi3NJob^g64%UxOU-75aSixf<*^-gL`v$ z!Seuaa{nZ~y8k3Kycc^#=4_`^-?Qn^?ECcbMJHZbc?ND6>Vbc_tLRmYV%MMaWy(pv zxG;8#d#*KH?5A0bp`B6myeEfW@2yT>w$6ulACmc5wjp?0@e%h~55f8Ia2BxmA(XVl z!`N^8__N0ng)e89q zc4SgE4ZVEaFlPQ`%#f+WiEq!7igbqHQ=3nF=4sNGPBFuDx)Czodc)kHJE)#@9A^5z zhiMiY=x(hgdMK2_B41$NH`T)Wmm^r?F?%*9U{h4pVsh zHYt8R_z@GVM}uRW6`5&HLY-is{-JYlhtC1p)V77rMD22T(KQ>s)vB}BGZDO|)imzW zL`@i&V#nO(s`RJAx3+ zd~o1WDf)iiE||WHAhGI?_HEmRA8owgO}CiM@pEDWP9(!jHE$eueH-6em&O+=EX0vL zRN;Pj1ZhRlW@_z3*h-#;A^l%7_6+SCU& zhlXS3o!8(j^8_N!FC%5Qi)_64GUn45%DR`Bu%+KMSVLYGDK_zRrBanP*w~{?zpbot zdM7kY`GLir&cdE-194{BOHk{s!^F8r8N*o zHJ!nxizU$g?=-m1>&22RH?fDSVwhm_i5ZWX&Dm`H3EEBx?3(E9@Qm@M!@fWH&`nq9 zxPdJx7f9%xsI9EsUPBd&&r@)2GLHH^6lEKILElS;E!v{Srmps5b6zaMfawd6Nv1OO zr~6pV)Cf5IK$AOq*&lC)n3H;-xGzh$WBNvaxUG|F@c!T;P+5P4oBizu@P0CUYODrK zi<}2bbz`Wv=R(+<^TuKJCPhaNNfu|)DN7r)q9`vboOcqvWQC?%=-t9;G5n^XDyxAiG8F&#%PJxDRE4%-)BfqENL zxGC-zHz=JGzR4!z_96$i%xV^UaJq!s{6>!C9?MJT{;+{f>zW|p%Rd-n@6G*{mP;=@ z+y^a^I&4}M4LRI=O8IjVrlbvo@qgvm#9zHxj*Aid_Q_4q?dnYiQQMhP#c^Ey+nINh z^`H?`j37O)hI=%954{3ch=0Bq!e%*f$iq zPh5M`5-B=Qjh%G9%D>GEpaK3bVB}{#X8OH`*%?OS+WR&b+^+za?@r(yt`BBq28nFX z{Q}fqQ%dWLZDR1V*l9eb6EN|6!Se4nP<&1A>rsX_Ah7$_qR$}SSp@Fey)+AM_N2x|Klum z2|Gdt=_}~(wzrVF^b=>HcpCOy)CKDsr#R}0magtf6^SC5oQ$~l8pgfmZXdBDOGkUQ z03UO6tg=|=Swj})yo7X*9l-YwD!@DZ3L9B>j=eEI$CA$Mz|}do@p@_*T|QyJmZ;Wo z1027CpV2;=RKFYc8ef1Pn9Z(iU&J;Bc{1347~?b)=yswz9a~?+_xpSVRECdcB@rI* zV0snykLgcE>)pie(hTT4B2T6-v=}`)DOq-x<0EF-Q2yrxF2%eIUhOt>XPndf zWRvl{Eu!u}5E^9K;g+cL+K$>=)lF}&;!08iN` z@N(2Ss5z>Iu&Ik%Ek8+Q5M9Sx=AqPH+j0K$D{|+GOKWE6X$OW22{`$^@|h{6iFAi6jJE8gPWSf((B{ zp>_Wr;pMKW4UOJ{oREZS{`$ zG0Bd6zJ$Y$oCPGPZosV@swv!j2>$NxK;x4Hy8qaL9tSzn&O=H}q28Sx>m|pUyG76M zscdQyuLrrrGhfk@E4_M}oj=mV*2J1t%%}^6A2Su%?roh?_a~~-d9iY=&&CDx%T$Bq zm5W>tdpSI4A+p|jcVX?CG`f1J0jDJ`hO(TKs1#5L^LF>4n^RuG#fi3XF0>xnUGBny zeTjJSj@Tz#w!+DOkc`uVvv1j|(r*>ojtM{*N%#HUL4vW^w=jMI4(vHB^%!4D$#YeO zO||ck_MbyVQRn<^+6T9%#qk@jWzz9aYo+For?KXHxA{?do4LRV8Q3m9?~h;2M(1Vy z>1}y1+7A!HAKN|g*SN`$VR#=GXh*@@f;3)b2(iTZMa)?+cG|en0ycemh^vl96T0TY zKB+C`id=yrcLN;xFBLMr6rt_lJbXO%JqF(}qLC9{f^&mBX!NXsk<*gVt+%aP{*!Iz4{}!$k zy-0@m$iSM9?JN;F{|3T-^-h!&goGd-Iik8kPK&=_2 z6(b(|kcq1|wUzay&x^~kS86FW4nHOO*@(s(hS9q-R`lW1OR(Cw7FySiVWWytC5B%n z6L+!@7Ho27qb?1=Lyuxv%+%AOpJOl8ZT4hSW=Eq>Mk#(5p+^E?7ENgP9HJ zfe(_m_-lC|I$m2yZ+&a1uznu;|3wVX?+pq*H{q-8PqaK&LCX?6(9WZk(o+m*+?4fn zaJ3)R#%-m})2nG=tu`l@_7V0*h#Bp7qv@r0H$F7Yt~j!A5W6*b8s6%9f;AyxZsX87 z94JWG;G+`+6VdB|YOZA1*Mpz@_cXq3-vob^YpDJ5Wj^f1IOad12cMpvLIH~26j=NR z22OK!@+fj}`V0Q__O9rK*2=<#zPGuO&^^5Svr(9MNnj!KVpyNwK`(N5fi&_p1 zLb>1aFksGQu20WZjQug0+bbhxo&6F>%PSN&TknOtlXQj63xnx(i6(6o`4z_ZJ)i^t71jTE7 z;_ODMl0jb|=BT?DhBpmn=VY_l&=qqfHZ3mr?ZZe6DN$k9wg+;Kfg4a}k^|W2j$?Ae zz3BUpV77GdSv>H%AKCamf(%iwcu^QbC)yUEfq^2+@~s8?mAWiB#{*UR9N~w#lwhIe zarimHnZ-3*vCcDgEVODPvoLFdr#013vFrmhOn5=h|HMGOxdR`^7ckQaXPH~lM`q8B zBK56rphZ=m6&-)V1%!%xjM0DKT>M)+zd96GR%cSn39;|97@)7`Qqr?emz2w?@po?a z<<_5Z<#tdvr>o#VIiEITdfd&5-P2`3y~2(P$2rrPoE_wrxrF`6)~69IHl*#n8y?$6 zg748j?CGZkxNcSr`T3>OBHn;KFQ~+aCMRLVj0)oxtONSmTT6Z4erWIk8aShGzRAw^2`T?^rrZ6AImaY4M5!U%M zJfxDn7*qkCR)f&zxgmX5JISM7f%Hf}4Hj#99pFO!6R?LzxpkBBh*=X zX6#?c2zY}s;pOz&o2RDVxfR=8w@M9vv`TdiRiJ34$i}N1$}$ROQIh`+oK|CkOXK?p z#k+g>&FgX?my;93cp94b&W3}w1DS5KH#;EMuqY~QR4{JBkvLT0!vGqi|A zcJT%smpmiwHF|V7bQD|4_M%GMA#OvGIA!Z;&ZGtG;dn1*UE}~0=QhGsk3=@g-GiOVo6gp~8_m{~`?D8I-od>4 z3q@_G9TjEo(D*l}u;xe|$_%lh*QQBa-k%+Ct*R4eo5-_`U$Xe%@vHeSr#tY?UT1n? zQ^$EkmEzKKGx?HB`WTir7*Drn(Vq@C);7Esd(#t%#V;N4(^?}s*%HW2F8U)K<1&}G zcyNcC8)5}Hmw&U=tONye# zdBNipNdAc3yblfGs^4T%DZjwy9-2kd7Ec1G$d@=ZFXJ~QdSbNFUD&!HiOqeS%@((3 z(uS@G?z~Yuza~P4j=c^6(+dX7|MWgATa(Uzvz`Yj!vOxfYlM4~`ZL2xe#};WBbGhe zi9d9d$ciN&b1Ht1|~MU46Nm3!#}{};&xmf6-%F!cTxSHdfc&WF7|orz(QSz zliG^Wcy>!9=~W$~1AYHu`s1@OazZ4=Bz2?Z!TrL7WH)kHwin+V>Wgo7Wx!Z^27@Ao zV^Hy0zU{0W4e1CZ)g^71{rjrpROw*o`^8)2b&rJu8WmiNcrF<^q8Dn;DdwO@hS|wE z(xokjXqQnHE*PZ67iYiV%U9)do)vYtaZVS`5`8mqd;a5-DwCyu+N&_c+7$-6$}pAn zH(=^!SFR=@5W2-4arvuF@J4WDR%RWzJD`BR_q#-Te*J0J*>3*7qFxw&H=BQ#1T1{p z7+5NL0%zCzVY=H59QaRz9`+Oxuy}uvpPI|z-dl3zO-aXL#^&~g`Q314y z4DPt6-}&bzXJKWGK32|lWHzJsun!9FrHaooFHxb>edaL?8yfkymTnCtqXVzS92iMH`EF7RnPD$Ie#KTQvtAyNq$2e_wXbdlKmG>`xcu`Z0x+DBQutl;_<4CY_t-jhesr(b?8` zivC}}q{bF_ZhDyW9T`IY^T!I)J^KmgCThWp7F(=Hh`Hpx)SX=l*a=AoHgn75cxjv2 z0bG0Y9$%;MfvQ!ag`&->f|HRkYgUS2!tYsZ&anPMzO1&G=ZS+8eZz3!%?)^9$wX)= z3SxogdiYAi7;jfeY5bHBAz|1il5C8jz7A<*`qPij=+2_pS0m|iyc#QYn+P?&JE*WD ziN2;v1XdOffv2wEdkb&2SGSBiAD9Urdt1=eUgao(Vr-tU2hUw}B+K-xxIE(y8p&>k za*xMS^UcbTJyB#34ttMT_m`o3ts!l4T?3{i{UtskZ*Aw>_ee`#LrME@Xk677wBP5z z+vmC5EB)2*bLc&M`Ed)b(7VOIpCBVLLL;fvpo7wDtU1N6esrn~$hzY+c4=1!$?EjosDOxHy^rXje(82@wcflk6b)Y?eQr(IbNmAKl69abKUkA&p zV)&EtlTbHrC)#a&fNNnc?3{TMCyZK0V=t)Etbcy=_j)|vHTDRvUM%4I5Py0(V>2xs zY9z9HSFw%9u3^l0cRp#&VEk`u6Mn1Fq(v*7vHF7rd-S^+O--~&c-4cO3v*yz)O~K! zl1cn(mmb*F+RRnG-ph9kETog+&3t9zdt{!9?Cr_p>`>AGrwmV7c3$K(M*0n9+Y$~+ z&9aLyWN;deel;G~9_dfDYd)c^+Dj<3N##yaB&GozKm>+d^K!JLvuBe46Op4{KCq*^oIeU|{qI)Y)o7S4-Sza@q#6R9e7F z##XZ41|J-)hri~dKk;bX`vU34w2-EXw%A2AoL{-90BjoVnWKdoTcXwmo|Bwe-o6iD z-Fye6jgPs{YOZYQw2^G_!#gm2TpOfnD=?d`>E! z^;e{Xhhk5;sE4@jm&HnzqY|sV5BW)MCph^`H}3A7EqtzlEv&!lj{hyb%xQMLfJ05o z#q*#%taO@1$!~{|RKtWGDo5b&Z5ObtrGeLSnT?-ja0uIB`lT)$u9iib!7dVm`O2as^&| zu1-Z?PT=c8W4_i#ndLrT29;#NBGxHk^^!X%4&^96?i?PgFhZk0kHE`oK24PQkN(cd zApZ@~)Ls^c_D+k)!K6*v_jm`FbRrGd>@*4(Wkb$+e<)0KHZxtU%le%zgwxx$qx!D* z=z6n^qakU;n>LZA*=y*wl&6I=&*Im@Jbqc_8FU?ygxj6_(zLe*U{zp`!_pVhxzUs9 z`xsy9(08JeoI2dMLzDal_lN7#*U?tbt@Od>z2w;^5r5W~W1;3BFzc-BWCaqcrEqb zY;evDHm>fobf|&o*>@FZYbE-mWwMBAyv^e8HS5FCrNfx$WW~S#6@bo!Zg378N}NMJ zy_@SyEUFhZHtj`DW-nJ){84Hc)riBt(|jLiT3`FhC7 z%g87wh~NItXNU|ZvtaGQ(69|57K;{yFR-w(wso+#wssinI>BCM_W$z}`l5q6{icv) z+yEMN_X(^W(T*IiV=v9CzyBb@sS+4mvgZP;qh&&Bsw-Cay(G^V=cS`UMq8pJHcv zIxU{LiawPMgrbz?OHdYK83ACrRiALrmX4 z1=>HfgYCl{p|5SQ@M70ktlj7)q%T~?KUp@FZI@}l8;jz2yFnvG-n=W*y_`iBZr$8( z3w0bGH5$(6XTd&yq}r28Y|zB_0P7c#we2Bpm-lu4PJefPJS)V#-(9h*R}N*i%pl+0 zW^BX@8+PiWCxpjHnD)#I^km*pFfET!@EOR-{0Hg- z9B9^V9k$`48ha}Arq<6PSUckoK5JJMVswTGmws!~{=Ly`-J&C4XZr_(^H+!|aFKiT z%8Lw74J9_ijxNSm;`#48;mC?$xbJpOWHy)6+CWueC&lxJ_kHa5#1FJg+?ek3wRn7! zIxGF-!0z^&L?^~Aq<)?vJG*+Q;O6=YmChVs{KuhKWVsU7U36om(-lcu^t)c-OQDY2 zf~tD`Ski_$Oq?`OLvsrM>8ApV-e*mh@2=#9YSAaOSR%BSl<{2ZGKto`+2VJRPu0&J z>rTXR_S4ES!S4&NRJokqj|nHevl)2naTLVW_Ttvu(xm(E7QngyC2SS(=I8t9Ktg5% zuiZL|R+BxbCr-nU^D}8Ukl8+SA4bjr1YXl$~?TrJDD?(Cwgru}78Z zxU#6hzFEZY*?Sbzw&vqWHkJ?jI1S#2bBBKuqnX}w(I?sQmXDALrhkFwvH4R4eCdeA zp+)9oDc|bYE8re(SYD6(q|=g)WLvT#v12_WoYG2S*~zg3pz}Wq*4kW8a~r~F>89SI zH*Jp4=jM!x6_sJQP4qf%VFqO3w}O3fmZ#BL`55}U5#^%h+4Z0o{PHa zyH*vROEPCwV@j#K+lL){)R(4|x#GvVZfSJRUtH1gNvhy84Xoz)(#p6-G_$+{^Buck z$JDEMPd1JsYLCL_rrvCIZ6fnB9K~*^?&Ic^UxVWCKh!atr?Anwc>d)?oD;E;zKi$5 z<25_k`$>uX_*JiAnixuZAbS4tpb$fX#SZnSzI?*?+0uVOeyBb7E-Z_*WLB4K1%;$5 zq@^VKO;!cdafed4BxXCBnq%l=^0!;Wh`E^I}c7c49K#P#epAeZQRDzW#^K3U;)(%9b2VpI}1tQ5c(?jz1;) zLEiWXUfB2#{8xXc_!M_qm0t@}D%!y;Rpd}SzQf=7P5hj#w>j+*f54|w2j42CF+tT4 z(|sS}Kav|BZN4rWmF=mWfxjM63DdTY`8W+au|yhFp%_Cw>XlT5ZI1x6h>MW0tMz|nJ8b9Eo=Ve9P-zEH`WqM8rU z+_nVPoMQ-$p5w(lc|Wl?x;J&FSdcXA8Ca^ta_Lho+0<1p;j*DGRew2&3lp+v-llro zGco`?oMUky59|!AYuR=+D9+)2pU?$r{ zCa!uA9{1VC^;?j@HHle?_p=1a%h7Gkm|k1C z3C1NGAz$49a$lx%iD~J$PxB>~*FA^D)fd6|{W<>bxqHax&Z1jEp)lE2Ss3Xcdccdm zIX(~i0_*->hTQGL_!m~y{KXYsuyt%Vm$Z_{l%5w5F6^K^MKTP`_rY-c=O}xm8t!$y zhacm__w&UD$yWEnaMMqjX}^2`5)V0?xiOIAwZxtT^`oHlXbfIIc99KG9SQy~Z*b8{ z)-)(&DAS0D71_qUDDmtxdXh4jUDnTF@~brQ%D6-5c&v}O4yp=Yeix$pB!Wec_pn|6 zcEgvZ0P1b6$o{H)l5!&t;{q2|u5@AxoZfT|p7AQw*|{F3?I?qO&v<_3()rY40m5?` zRkr@V5zFq`%$iOX;bc|uH)ri-k#f0^P^-wzThq$dY`{9UqOx#2iIbF+Bfa!H_Z;rCdTLISr`Yd5!TTm(gjyRsHfMvr6vV&E=y z{QNQoM&CA~n)F*Z&Mce%C36Q?8w_IfBMH=ux=}9Lp8ldGeSCI^?`yG^TUVOFYZc^6 zPCd+pnE5YYX~7P%O#BLS`zX-%`3Gr@!Aj=Y@0)aXM7A&xX}mOt$b_M)n5-)Zo+ z_3ZSPefVK^CBISa9bL52qTz>&sP()JD;TH4`tfRP&+w}WpUo(Tw^Y|=oQla*~)^6nLMqvPbaNE=`{S;DzaZwi+d_8 z>0XvMs#M!E?Zy!}D)}P}-IxuIB`)l8S~YY^9%Fa+ZIo*X--3%30{bwCqeH9Jy|JpFhBA>;7UCxU+9k|`5$As-C zDg912f67Rc^H4OU`G4J+sqT0@bw!!Q?5x3p>$}*m>qppb#aMPwyMe!Y*^23_D3ZKl ztJwXj&O%cAkZJEFVh77|N_!efeP-^$?f(uzW9WZSm9IuJS}(CU^CRll*22p>N7&Uq zo+4{q8Hbq$lKPG~p(oW!I9{p>Wr0I6^Ir=7T>gm}wHBe`5i9htKEqF(U?c38OQj!f z!wA?7kVbN-W}l7I|NB8(XWqu5o~59)#DTxQIUP`ou&TT z&3E)Vg?Wt*!m2f&Xx&~DLC(AjKHdn&6AQgC4FRbvTu!+X2F9PCzywk6 zJrVK{d>4h{wT4WQaT3K(`8S=-Ryl%BE7wATq80w|ROOZ_&xM+fN|^OT>^oQy1owuw z1I8P&^nNM)){_BnDqHL$)h&d#Bg%<;`x5hh)$p4(+=qXACt&#;FV5FwD~7c6BJm1y z4+U#xIM^9qe*eMmw+d!&V$vihr;bK_UIms3!(m%}43jzL&ECy* zDxBKUxx2S8GVv*DSe>I$Yi|nov!+l}&~7^TFj@#q{tBLzsif)Cgz0yB!64ByV0vl_ zEwb!ju9J-6(3YPxBA^R{o~3ij+H*POjXU`OZ*%C>u7q%hdo(x9pGJ4(;^s>Oh0+lV zN&R#Jx?emkl{t|ID)a>2boCQbmMszto2FqdhJpUo&cD4J65?SdAiR@`zlU-IQ+ z1~X`y&MJg{Y?XJcWYxbvUG1+RJYXxZY*Hm~AHq?m zIu>tiHJ~47t%c;rw_tyOoVfq4qJ38@nUh^M42~NhYzPY=r%_()TT!eKdvFdUJ?ckG zb7X}(eMYnOQI}vs$pE~0S=DLXn4#2pD4ot!{=u|MBWYCATZ%~C32wU!X~{lMVULgj zEw7d7aoaXpTT{U_Qu@KB^}k?;Ofzw_$MY}RYT3M>581kv-jtBIMmYW6jWrH92wM9p z;gHKd)-78k`E;ujcPWRllilWwGrhuIZ0~^6n`fY2;eL8CVH$nEp~|h9xDnkR7&B$# zr~LHKlZDm}D}J8BVXoU*zy*r;sb^^W`PKP&B}?_oGDwwXdh?geASTLV_Pp&xx)HWh-?9&pD$E@rlS#OAV){;dDrO8!QD z7=GCNiv6dfjxSsPlFm5=r>3}%lBIh*1-XYEVo&uP?(6aZsuE|o`d05m-czUm5e@L- z{V`Z;BC`HB)N!Xi$+Ks{wvafT@xqoae0OJ{Fw|`myOMkcFKqaRSAFg?+l8Bi)cQ)n zecUl-ZyEr{oV{q}p!H0rq56=iR(#@BWx%wlUK z-jo)g#nnCBN~KT8?izu0f*KojUkkF&7JV0|kCd^~d?t4|9U zE(n2P*A7C?P&@Wa&5>_RzXF!qRG}elG?#njF!aRF;I``yWD~g-G%Smz=FUrWPCQ2) z?tce6X)zso7DvNUE|QDJ942|11@RNJXtqgzM&I{vsRJdjDPTUdhIjE7pV{Kb?fck_ z*`JZ!IZS1fw@IaY#2ys2ICwM036loI@{WOl*m&zaj?>ryi%;+1U!Ofd?q=1vC_@T5 z>w=|jsXkcmXb(>69mbItE$ytOP3xH+!wmC=FlMV-RpvfpO3>~ zqkE&3v>1B2dVvkGHX_Ik``s51n3nXv^!AH$K71gw(> z?%>NXzBF6Pcaotr5OQExL=oQ6wcw_lP-LI%Jh??|T&V!WSOxI4n|DobJU4iEOGu7G9uxoQPEjVdO z$v;;^%oWkcb~1vgpB=<9i(+X}pL|N!-arpDFNw_OffUqZ$W#1AYF?Me&Zi!gz8t-e zEMp9De|mz%pR*RMj7N#RI~LNqFp=lz41$r*T`bz+&$~4ZAQlk?K9e5d*UK@aaoq#g zN4rDhPAe+?rAj8VF5n1k))gmr^9(3s@U5LI*wK88NTo~?t0|I9vO zf4f+bWq+R?8&(3zcb~%JfX|fLs(?p0RTh}|iwpN!0W+-tBX5nyKcn=~_;EVvef~;| z=7n$%66{&fr%JAP%Sc!orbYj@Y6vawJcQqA`)QZ=dpcxUM`e%n(CbAu^o%(NY9IQ2~-^P8zvbifRG)J{hxM0jO;Gz zC9B9nRBy8fpBD?C?wSjILWT+3?mnU4Z!b~jNj2K|U7eLBujfY|i@c1xeM zR~C}Gs5#}YXh+*hOP0B5Fr9ZeMwQ?0Va{DOjOx7)O;5gsxH!UZ_xcIPHh3|eguY^z z<3YOE6vzDHoiKCyOY#nJWIyI!0H^dcVD{i66Ux)FT-dcL= zok8oOQ(?r6&(ix*c68OljJ&(U!Dp%HM}J~Ltic67Iedap`Nxv;kzriRyL~XorZ>|! zJB#UEncTG#TF`Ua3;%E#Tzfw$bmt~;rfJ`(uX;XC`0ptOd+y|;I`VMc#zM%p9|9l!%A*F5U&n1%4-l()jPZ;97gJyk8!AD&enUz|9y9j0T&gG5|gJQQ{}{)b7r1G$|i zHoRGZxSUtvzL&G#V$H95f+b0aoVCzVf|k--JF zKY{a1B07BoQt!BQan51`|Nb7cSAb;Q%PW7l6tc%q2{8`D;cbv zzj6m+qNwcID9BrVRr0c-i0)_wV(y@^FtFcdFf}V?Ll)|?tdB9YXni-<`+TElt$~7n zS2Js|-A%)!;`^xNE5F+BD!zZ{3MJb`zyH@Z$?c3TOip~v#l6@GKJ5!Q-&-c!{61+= zx_md;xfMvp99cqXOYU&eLzmI7D+^ED8v$!34Cd?0m!pr-XyKda8RAE7;W8%V@|W!Y zgLSVT^8Af?{J?^E{!8O9I<>ACTL#!r+p*dF-L4EU@4ZjD*?uBld0&k_JGbx;j-+z0 zT66dUO$i&LXvOjq%c1Z7WIQ4mLC$|~Ir=e-bi&%u^Vtr%-G3Dh(8&|)YsLzXltMVe zp#m=InT-Z#yQt!Z9=feM4OMDU%=Ut>m^aFSr|Xm%KWiKK^pVBh?#}p5B>6_B?!o(- zQ^|9Amhdv5Q&6~^!JZy3<(ib1;^44>%*az8Uj65e`eCW`d(jV+-(xPgI5t3hNHOf( z>CQrj%2AE1BI#Udq2GJd*za96=&5l8W>{IWk%=S7!6kw}^zt|*E!se8QSWH*R6A%I zse_^Ohmxy}Hx$pbB-tX7zya#;^yaWL)!MOYBCOX!#n>OrsNAs6Wn0qXRH~udZmNT~^*};UD5>a(9;koK)@*O1K0&|u33*}%cw#*29!iGkhBn;G1cve#V0J5Agv`^fs@OZf^JV5YMu^ieXaeH`H6c zhMlnwx2yHuVfXD@3Ywl&xASZ`46$XNbiFo|8EK097n2eeoH2)8`E?k-X@*m|b`|?2 zm%wH_9pZY&&&LU3XEdqc2y2^slogxpX9|%eVy9TdJH(ECxJ<8>U1iV8?r=uIVuo$?82D01NRM9FR9@gX& zdKJqsi?&JZ&EDgXzUL9l%qd_8)HUr+8RamuXH%H)GSlwc+qqPe{1tn)M!*~Q2yh+X zfa(+1kWlGGS|v^_^yn%I_diE{*No$n`B!j8bD&T*iW5@X_X-odtMKT}E!17-&ewZ7 z(~nIRu#z>ydv`rL){}|LdxwgA>O#hyHx>Pq8nkn05RLosfg9ywEcVNjNk!mLqhBHH zubfRW3x-n7E_On7~g~o==$7$Dgu(CHhSobA&w(k8l3bCC~h0<^R%d}6UjOu4ZV~yKP*6;p7u1jUA z5MAph$b2;wPISo%Etl3Z2a8np^r|fTeEcdb`8}HCBkt0!8*?%E&`kEO_iq|?R^%dm z8Ny=M6hq1)2{bm%hM<*;sIy3*l5;6EwYLf1`>-Kq(P8vS`Gad}#hz``UhF($46}S5 zhy=DU`Vq7PA6yUT;&zP$@8@n@tJ+(9V0xBp{>*1K_beb{;z%gkunda(9pkshiJq;z zElh6P3;3_Fo9kz>nJ&MRp>t}r7+7vWH;oeD;4W`|evT@gi|OJHI*;JTRMleXj^*se zVolcDcnyTCJ1V|)EQK9I-%_oIJhbHO1pPzfz%;-gYxc&&w^mDBzyCYx`1cXkmvqCw zr&<`@GL~JO6e6U&iz1`B(WKaW1TVK%oy8Tb#%jf*?EOt^P<+3Yqq(>K&i^-s;Rhk~vpZk1j z6S$9-W9}>S(WT!GHrd>iTU6ocZ5= z2k-!IA{2^ueS;CnB_`*wVfs@HnrppzQk?X3H7!Pvv7EZNxIxwL^hUxWo!cnNy&^P`odV ziKNNxjx5Acjr+dki+I0KW0t!&GkKlkti5ELuwQSw;8B<-+`V3iUxp*SUwg^6tx$$9 zSgIx5t}YdJ<(sk4_y2)Ha|O;WpUxb`JJg)3Ss*jvvCwtLK&TBh6m-fjb7S_+#8E?D z!@(`SV$Nti8!R1<#uFT{m&PR;o3oofJv>TRpY^gEr{*U-xu+)bASYseyethrEbg<{ zo#tkYjUWvJGq79d33~7L((+>QJRBP)bs3w0^Wr-&ztfjXv0ckvbjw0usTsFB@uttVM1XU;NpX z-=7Q`<>77CL_Vo2P`GNiS@_~O6q@7blY;YO&iBh$esD|!Jb1Jg3q+>E`r#w!=LR+5 zNqGeYi;V$m6IaRT)3aE3rhxyZJCK}?3H6kyz(TJe`fEN(xORIsech+Q`e-Cm?%BS= zoQ>=FK!X?HQ?r-0b{fIG_#jxkwm*nhaipH6cE zIhTXA!ccZ)g_{P7SXv}H8|i^2)ng2Xixif zx}_m2*jDytr@)29msWv}k2|gpS_-8ak3qd{12MM_UOUc&S$G%0CehKMHhL7bpAz#F zg2)&sGGs+L`?w;_D|q&`38jd0Q}@7S^k7^Qx56x#EBdEOTfB6k*hro^{9A+lq>nJb zsV{%9>JPt9aR=Ad(ik+86ooH9743${v1&p)6}z@*FPDI0H8ys6hF?R0xo7lnhyGi-t^{W{vE^UE9BM zzV#aHNzq%pB9X;!K9S6C^%cH7BMncO*}-C8kLW5kf?S#DY*l7{`ySlr)x=-_A|e#?#-PppQJ`E| zNdrvfVfVs1xV(BA8!-EbZPIt1%oRMyW^M{!HY6CPDchj)lY^Xv;uKgE8bWunhtah) zb#VLmYl@H5rw-+4_V$hmY?~g2SEEJ#u7VEf{!wKc{S8=PMI6|y4Ccly3ZW093+c`W zTd0)J2btgEZ}ZfYtDZC&zV|$Z`LhBzv$nOk@mU>h@1BBpzh9J^JT?%DO9#^86b=n* z*P@gjt7g=bY=H#-k^&3jz2ef z7;8KJS@NaX28jodH8!DKVK01|BFS=CZ4wl%AURpfNyqPlg|D?%6R z4@`jt^+E8`(TO#x*f6!TYhhTXEK~fhO^?M{z{j2IsiRIFikq+TLoW*0^W+f*X_(U$ z?;ULHDjhUw5ZQbl;VjK$GHYJn7fg#}hF*lgT%u&qAp_m+cnoo+XG8Nq zreYW8H{FO?zTIdS`#uh9Rb_?d4a#;eXOE+$N*n0W$9h)Dr}`18ec^k^$`UoNx#jJd@6vDN&`fKK60l#mZ$G z;E(cb{^f}e(#vz?(C=6pA5|lIxc0Y#!h2bg`s`#6x~1XF~QL1pa5a zirtCKUaa)48FRf6&H{77LEpOrpMQD>2TqTr5$8w1f{ukE6F&n#uQ^E73o6)#b+h=J zRn9OWL>F}055djCfzrVx9bCKo0q*(lWLTfk$>y&-%Bz%CaHo%rpkswEIQIuG7_R;t zcI7N%CSxkO0+AJxG1r5Os_KI~);RZ8 zG+f`871;sWIO@UGKVlv=y+4~Kx+_*^p1gD`*Nb0y(g}1LZxdkmlSi9j*P6A);~J{U32njkK0}_N*E4b7M!6^BT8^xw>p`-r9p1mW6<`Q z%q3%tT`bZ?bPCN8+K<- zr;|c6ryzEhk2YHH*y}y+*{??TwNCK1mSI@u+y(dYIMC0rp^D@AH00bF?&j@F(%Gp# z?8CJ~sK37+mFp*Cf_Oi_{P7I${K$#iz5mjna51ltL8PB-0X<);;JAG#70v#IRvu=g zz#oJ3NxDMi3j@lqQeY=Wo#JL#Wx>U$sSxK{#U8$C!5pRigq4H%_D_RY>O2NP*54%! zOL)ni+ZN2_T{9|c<)fkK-`zKRiLmz59b4t?AK-zHB7TwDV8K&&)+TmO^7Z`TP|*dD zIn~MMR;}W04nL2XYjoL}15bb#2?l#pZ&5ay(9`#+FsA(i7CFc8wk{i(xyX&UA9Du! zB&ss!vm02q^$_NN)}Lat*V6a}U7Tr@5v?C%&0d=|!q1T}G1zN79&r-iBfnF4vwH~g zjaGEVFIKwu^Kn-7y9zIgf18XqWM0P4QL-rqC$(?zKY3fESb#R95h}%oaty26pUN+@E zX*>m!!iNyE!i$TjO@+dXC#jc@HQ(#6Hh%r*#}-~UVu``>_;PgznEetHkd_nCA{#iCCYt{s($S3N8$HDv zPljXA%4hi9-ktnn0aw0oK?@5Fc9**au5+azU1tU@ckIyFHG<1+9|ybd-e&cLrV3wh%U_poED2fi~2 zgY-?$c=yp?q;n!CaLY#M;^m1u@#)oY)-ZA<2sT!b>pYg7zLt*eUoPT=iOARaw(_es zJ_P9&Rn|0ZEHtb(V~HCVpnGf$4)e_6&)Xi8Zg+jdNBtg7i-u36^FD9TxW*aoHha>W z+CogTGy8wnflZGLhIFq$l<(6Q$2dmAfb=HN{2akgR0@*(XDo6)`3DeE>%sl!YRHtR z9ULnT^BQ}{(w6fdxce>L5|ig|fy`=Y#Khj5N}SkL+3duZFX3?ExvWSJc9sufdMh(v>FfZCId~k@Z_USHXAa@uM-}XHK?W{1(GsddQgQOW z3YazLDzAAv5>NKgqSQf=)Zib^D&L3G>5mooP_;(lIOzdi@LkL;X}k}=AI4$I+NE^P zEeg+->(cSgAoRQ6hshaE=Tf$x<<`7wqK@n)OuWfap+YuWpJ>YBhnurl#Z&lWdxtc> zK9urI?&9$^{n)ENuIMPE$sW{dLss=M=IdCDPNmUe4)6@c)-q-po`;#meTCl1Enxbu z1DwBDi3_x1$jG-~PP-3*hR;K=u<6ZGlJ4SdhpkZFt_Go^JM!tc>r}DoC!gX_M%Oj= zGm41h3%nKJX|jy)_R$2KG|`=us;eOWqziLUGNr#^muYHL6PO1n@DGpq(y(X!>4QxN zPM>;+YwbUsd5PcCinfdNB=Z+ek5&@8Yh~=pm#VOt<=3SBl3&tr?mb0#jS+tD3S~2& zn6skMCrNL@K9Uq~A*Zn!v^Q}CyJGIimJUc^mN(tlPPtoL{03XTMH0j(s%na^!5UmP zYBhWExi2=h)qs5aP}qgRU|a3UibvT9@sFd0pUb}q$8}fmBJvNfzBd8go2OA>rWu|q z{)yIwujx@l3V7dEWoa>oxow(KEbfUFa>SmRx$}M))AO3IE$YGkP0`>#s12e#p-dUi*{U<0s%T2@(ngwWjgs;CxcTqow`Ko@BW^Fz zt#A^poG^#ltv}JVg{x_sw+8;_oj@;t54bdSPazlB{{#-PQ^Q`zo1x!jLl)9DfGHlC z&+l0kPbvTH;Dh{TNV8lNNZCaVnoH$a>jwiN=8!nw@)CV;_) zPJ@hy6s)~ng6*qJn5lsp^*TF;dsgbsZm!5i%b%ODd}9{3K=K-V<>YYgmS6ngA@d~N za$fA2zAc1^Iq53fG&nrF9&{VEu-*DBw_?nAe5v3~F2CBzG*DUSHEke!V`~hSWfHje zDU8k8t0a2lA25&3ELv(8!2iCtjFN{erWubX;0)Jl_U7+?x*+3>{)!jb@W!!pclS2h z*ZU7`5E?N*H3*eITk?GU7Sb@wf-@IZvl+>|_$hAV$l%g5T%4oK)@*Tv^p=lUoVJh( zOp~GI&>Bj7^%Jzp`e4@jLlmfF!5oKe1?#)+^fspj%`}?$*+Mo|*=6F9#iP)9ogMr9 zRu-cQ%b-Vj7cKC1WmSq1xHVraS@nrZYGfAS`{EJ&~=o-W)a@cBafo_Awi)?#C2$RdgYpdi?&3+wLKB#Al zyH3!{yUG}H&lVClT-oLqILGNp%xlRwBMf2A`@G&USsVE=mxld6HB4hLR?|ToxreLiol^!GA&(mNbO&Qyfy@{Xz${ z$O$Fmk#TGi8^}(nIl%Of)!duH3Vy9wJRj@+4NYfhK>U$Dp!s13r*=7!Z!uS+elOp0 zIZr7cZ3$;2vv>{Ev~R z`Nby;g&EN*bSmEBQe)&R&TVx&eiw7Ro8@Pq;rBR6bhzjZS;Dagb$Zl3zy#cvsIgww zH4uJ(44vN{NRkg8G$(Ec9bWYm>{1^>Vzw^62(qQi8ADt>HQ)u309NoB%E!n4nE;I7r?J5PRg(pI0k)-tRFxuNZpR{e#VYg!m_SKG{^~R1gHZ+FU zTYr?5$kk*0eHTt4X(;dM;mKHh3K(=xVq29`v1jNRcrzoMg(!&s#aqLvw}q^5D5Q?g z-Ry!R>M3kB*9{ha_h6%}5xMNp!}^sq(3GykURs&5v5`liulIC8f7}R;t5aox9vgA@ zFFEQuG6%JLmqO&Kabz1m7l#b~1>%B(OLT4v-F1={p|?mEgZEmaB2?!ELBfPEFDii=1ur;uP$A#j|8W4gK4i!3cP;2lsEcZ zgNG-agsPqb*y{HcS6sRT8LH}+w8khhuT>+6M7NlR^=GW{x=5-sj&Kd1`m+&z1Esx| zTQWXzCdHp=B%d@(X6FE;rgeaejH-^`h(Bo z3-o!*7Uoj0nUizbMGXT?ZKFfl(8bFM&hK%en--<;+saAg>U+_`ix#x4U*t*l zsv2zx7VjQK(`m+_TwId(499C8(W z;k@sf@oY)w3*4Y7&fhY{!Ow~mW^S_q2Nld^N>0%2Jwj71K_J)g(zr(@}b+9G!9ef%j=0Ds92>Cag(R8a4xVKnQ-qj?D)4qeC=wAp= zY{pPdyM?e%exeXCCx%^o@t)VXT_ce*+`^1%p8yv$O-Szd1AM>#lU&xZVjGvJvZ=i` zkfALnYz`DN$;qu)^Z77S6yLfpcZxh%xF>o-zGIX@7oH51Mb_qxtwl-LF1narTU{gb zZ?3{aZWe`^*|3{Gx}_RnMdZ7+Lh`RgoBgxfNS3`2PyeySS(Q`R-tbQB%sn8oIj<74 zwxG7*#DAKXL8iAhu0=F-`9D4jmvcwgGsQ#)YIFGiPJlb+q;18n$Sk8?Vae!wSz*$uxq@f;aN_Y8bL=@#l82nV6t)67JNr6 z7y0@&_;dFa@>o+z<6aeFZohtzW~WA<`weB6cLg!&p-BE=^k05v)H!zlCuVf^ux6soZmVk!-=+M*Q=r)IH@fBax{=@3@tyNv!eIKo#CZD!x{6iP}K zP;hcCuD%usTRr|GH{B|VQOpIcFUnj=qX*qjPhegCiV(OloTB(+V)o`S8NiBe$XU{SC#xXE;^T96)?<9s--Goom9BBNE4UjqKH8=W-DQrqDmR$9p zg0dehsO3-&A246ZZs)@@-0eB$w7$ZK?nS#Y!SNW@E?P{7Cx(;zq=`bt!bWo6>@B)6 zzwmeTRH*B80qaT|BYE@83+DMovPIPvRPoh`l};5q-}N79&R$>P>u*yQxG#j4s~X4k zd}tBBNp8|p6W&7CQWXkZ?1h?bE5UGcAIjRBMT&{~`~Y$9<0PJK_r@pUcoCU)_M6Db zdTYt_U;W}s%G!`|-+?J)K+KL5*wE2~U3v#`RYnX>&6>!rwEUumqdcDe@s+8?y~b6a zFY?u_0XpXdUTPj<&AQF!2}XM-!jJXxv@`byJt#X2!$%I`*PS0oOVaP-VUr7ZORa?L ze6Q2V``@_4ObIt_Upmy^JqIWF)8zY=XSWN?VgFxKl8U{^4=Y}Cf5MKU-Nbt^^M@kr zl9-YA_B0%NNSiJUyoLpMdy7ZD1OpIJpII&^|w-D z;TCew9Z5E{1`R%lSRV(R^dXBwBLM zRJgmPudwW+4r{4D3nrI5G3#j*yOw#B_rKi+kG8r@?>*ny-jqWCVWQ! zIcv!{@1)3bjpKKW+`!-UXu*DAS8z^g1)tQe#(uk9Mjg$qP`&RcJwEb~H%$;3Z`vN{ zCc62aD5gN8_jqu-kRtu|Wgc0t$Y$LSJ>bUsnQTB|0w*`2H!fOwkAnAw(;iPd;n412 zDmmGvsiv>oY?k4l&KY!h&@0JI>6h%y z$mADMY4KVb`@|haUhaekPF|97$7*i-Ko6GO=Lzf$ug0aC)97qPGxnY34;wQ-KRAX&Z2UGsvB8<9Fh?fGcLX(UR#eKfZ-QgIlQZZ$*DS7zQUCddY z*aUOFq~haw*U5U*BW%dJ0{TCTME0sB&i~#UYZMfm~0xJ#TBzbOGutki@^molfBcEG1Zx(mL?d zE@BfOIr#z}infJo77o}J^ObWLVFKN<&f;u35h!C0pO`j4JTKH?Oyxi}wqh?ONtE%? zg;sc{#laq@Rb0ft9lYd#2|HLL!(yB)*iei6kb1R|Gq{xnqjt7(!y{Mo32KI{bkJwQ z9mlcnv21kc7sp-U7jskXD!8JB_i<6^W*C}h#KJ{SGGC}n{@KMIuL1mF=xBr|EG4ko&`sm0R1h8D`f|Z|-BX_#Vn4pJ~9z zmJ6ttsKiw2JH(FXeNN~VOL6Dbxtd4eob|%jpdI=a$CS^4+j2#yJ}!?&Uwlim6Lf_) z&f{rMs+bXyk78wuEK#>o0as~{Wp$(G!qyNin)zb@_jBkJ*6#e8`E|v^Gz)XlcN)$e z%-KxdPX|(ot0r^?6Fc%apKE;EfE}+IF>PWKZ?rm@y)-u!(&x_ufZW;FV^GE0t_hiW_Ygqfwi zNn_3;lB`)qxd(&cqv=Z8(H+Yl(HoBgef{Y2gID+@PM+q?tL3+eUe9Ox53uR!d}>`; zhZe~)68silIy~4xfX1*sUw>{ z{T+U33}idsTt(^PO6+BO8P6#%7p_~}q=+{XQZ=5%a%YFIvbxPIPMndAS2RIk$|!>1 zFeaH-#Z)#AvD^KriPs6zA+CN7IBoHU*Z14SeNii%Q#U40*$ddW^f2wd-{;-c79^-*>+Dz35jX+}(Y$zdG2)tHz2EWV|N zoWmf|NBwc(Ha2y_JKW-No$iFG(CvU^Zc=ENWL3aIR?W5W9|GrMN`4IG-VoW5H%AL< z)21<9ha4u~Zyg-Gy@Hu9iGU>!E}_l6BedsvIe1J8V%;r9Z0(@YaH}gB7CjkH(?(Cn z_OqjDueB4Kr~OfMncLv`apzg?{uuW9>kl+(HL^<>x|uC}H;SyzMKR&3$oE%$LxaW) z;w-oOlY?QTkRz0!RRagRTO-M>>@geu{XDZAF3#IpMRu8_8RyKrEMJcbJ)=h?Edk!WINNwSz4&LDsS#9ZTB zy85<-l)7UuG0>D2&o<|@4vI{Fha}uu){jisGjSj53A?Ow@a5q3%zl;{iz{Bs6@Q*a zQP~_-4sVCP`D(cEOB}O)m<6vi&1i@ILeBlAEepAnj9(iR7+;ZveZ~6=pOp#0&7mNK zUF6g$7t;#YQt7-{Iv1_P{@Y>Ck1{A`b;}2_I~Qvz+UGsS7IuKSjSs^wTP*NE{3q_G z&pkdu<~l!X@kuD@4q-t}-zhhwl0NOuMT;PH`cG>yZ7@ECz87LyWSAPu%+27_>VlbT zYXaR6|Lw-N&MaMcgVm{VaA&_bvs%~*+AZSDw{8Y=ipgf-V{f5Uhy%kZGdbne-FPbg zKUQuO1c#{>zG=SZ`aNBPTVA+ATD?T<3CoI|J3YLgJeBCq1ZEkaAncsGRe1gK73JqE z2+tz@1f9qULe9^8gqNMSFzQh)Y5vg?z9#k;UbC~(==pCksjiT6)`*P6o9Flv&s@3||D01BaR--J zlwr?;wYb?|fu>mN2~%Y)**E(>uvC2z`>$G^E4lj|qxWRs8Z6{(E0V|>9;EEL7Fe=KmJKWz0uBpih)%zL5R^NM{Y(CCd*xmZ%-gKX za<{fZi}w?lI;s(F7e>?YxNu?oh%|H#ixx5tThk-08MJlr2HLDO3=ccZWM-Y8@am5` zu1``dSUP{D{JL<8T$D$F-xe@qa|O=rSr&N?0g_o|Pr`q?LP1pn%sd&x&l@=!M=8(3 zDxLRSSAzk2-v0{fb{F%STO_czScNU@5ZO1vco6)m;rRi1$QpeRH)<5|O4;X7e&23B za_~Xk;-)oyi^*U!YF1Nr$qCpkFJ>$IZRaA7-M}OI?{HM{e!Me#17;o2#vh04&_nhL z`mavp3fDS=^9f(fl*IC>va&4K$rg`1Ov1jQrWE_J1mAtB!~>>Y6g%%Tj$Tp*v&|-g z(W)?h!@mjiJgJddD;IEcx2>Z_zd}xT?`rOv)k|oY*qe=bF%p;h7cia8K`>%-3s~*) zfKmKasc$l>NF}E%CBiif4V$RD@n*8fKH}9kzi#0Q+Hm60f zx#7I6hs4nwXK_L3zJM(>>kYc)Rax82bAG z{cX$TBV4=r!9U$N$2p07{wFuS<3JWve@rIF(Xo{09Y*ahGNEUQ8NJny#ElaS;kla^ zd*vku9fp=HKXs7V^>>d0Q89`fQ^Tb)cGkgBub-a4{7(VV< z46OA(9CoJ<{r)i&Z>t>zxqbF_Vft3=?EFqv_qrZ;H=M+|wk^0K;sK}H-NNPiM{(2A zjd0BUBaq@H!JFc&Anv7FQ&#T#oKq?oVoVA9&-Agu|41m7ZFV`w!gR;n?dL&l)l$AgNQn1j5h#ZN6O;iC4< zpj`{Qcx^>9c_FI)e4iDPkua=Ma@O+WF=Iccaaa9J3w?mtjCP?74UEB zICkRSU8zTkk0e^51s69x!yQL%;GDJzFm9V3>@1i-x84MD{Yv!U()%Ce^4OTRxLsnt z3-wvtd3%V}F$2Fvi!c;Qa9*|@wbl8P|G`Jl+b$EH#f6D)oG@@7e+f2ToB*p{^ksH4 zdQsTu;nM#9eStY{46EbAxUb84b6fvKVCJQAS`c}T#^+R1b@L!nnQy}Pv0uk^Cd;vY zo=#xksZTd1pX9=e&6u<6T6SgG462FTj0;TOan>zU(AH5&bUHOswv93?_#FZ(#y*F+ zN>OmLS&99=%;0TB72DjT#E-hTog5GKW z;fS`IZ9ySVg?3(E3OjO}fvcM<_ z>0(2;0@Gh-Doh`w3Yr=TtX|oG)DOAP(?e6JtV0dPUOItFW#$BX*Wz1o<~hoGA{Z>b z2(FrwVWh4zUaR;AZ?6oYIgSCi-69=>j4NSvPr#<%sR{eZL-`0AQ=4}m3a7=;f zFZa>?nHH-xOJPU17_xO%olL8=hBa;ychu+VVdt2`5V5!v*VioP&U%l*4RMpvcll}VGD-nXV>ZVmiGXou=U3G9z<6uwokf{Bwg7^MCO&nwd4v*ims+A4Be zcWr@76{I>^>EXYUKpWc+CpGLb)OEIwM0c|>|0c(2CVumAsf>~`W zmn^uk<)M*G;nOmdTb4z`wLascX<693&x7@j7(tV67*c{tDNQh2K(@tA_}AJ}5Xr0f zZ9oj!J9pBw_3N3@&Fj#zDh$Ll0)9$7$9D%TV{unFx>`0BBkQ(9x?LrFi}0n+(lK;o z!8#1gnGA6sPeMh1OU@v`2bDXj$X_;&-49v@w$lVkN?gOroIgnUgh8aX2qw4%)sp$&0y-O$KFtXCObcz^15fT*Xqwu|GgtSd#Ic}J8+bX6*1;Jb!Ay* zK63ZZYg0hmN*p!N4rl*62^|Ur7#{NqCM}S%f-k9XfcC)VoSBrl{)JRC!i*U-EN1&! ze&aShC2seKe^|79f*?Pvi$?Ox*{DnN0isTj=K8_prjpF0IbT4n9-K=>v(j;3P69k_uR^WGk>K%CohH~NqP^`oO4FMr zTuSwz@sF=F=Qj$}=3b7~=EG@xqXiXR%cPTcZei?{FfhXua;+P~Qm=?VKxYd-#CImG z8e)$o-@>Tl$Up2gG8La)b%issMIu}H07aZBq43>nIp(PYZrgi{shV^c_Ifz;ydZW@ zZU?ecyC%|>{sSl+U0J$)HL7L*$9(j+-W1MS&evdb z!3%0jhytnmX7;3V0NbP^M^B7G0ZTFm}&*7GJoK6_=dl{O8_a zt(`~M*R67PmRBP1)SKPJ64Q`xR)dZCKOp0u5{nXNNJhTFtiUG&jSob!;EPX%J!6ao zWA7)V@!pxNL^tWBFk9$RbYt}$a+GQx&h|VXPtViFfMY-~yd7-HY~!OKX44dG{(gzF z#O|lOQ(qc&|2Me~b|L$shxF$Eb2H6O_$dRh?@u(Dwm*U8!-wPg_;i{#y9iQ}o>S>? zZ{c|C3MMT7giZF|)O1=ylAbx7bb2JD6yw9rW%g!%r}PB_+Yusr`vqC1E@O6=8$fb= zEDJV~W5;&%XKP=tWI2BZ3N=wz$m!j0dY?K#P*E8!ytye6bVg>eB(IftEcO!48s|nk zccqbwotmIrZ^!2D_XCx$s?14i0#@Z;m!6#BfFqt>lq}vmjcjJfbH9or;f+xf*MG

5K2Dy5(!o>CQ8{H)giTZ{5k{+K$4mj8nMH@(g%>7g@kFbLhb#WnqC|FTqaq zR{Ka%{65Q}&)pHQ``RuFFp}WN*eL#mV-&94cbOj)MDRqFE`~nRj1j z{kIsQtB*Gf{u{x*mrn;>vCE*~GM!~U$it-6`qCO#Z`KCXwt#Sfq-%0$BKEZJyp z3(DEAghPCnfOV1%GoPwRZ=2Mab)N-bS1v~{HU+Xic7ORGrRDgaeI)zSIFLTGK+`Vs}4tn9s+Wk1~T1GF0{+5g1$?~ z3mx{xLP4uN<-V)NMSdG;%IXVj@VM`s*|}Ceaan{w%u$#WpeM||{DKOWM{(IYxBti4 zd4~1)zyCjN($J7XiG(H^()&7J4I{HMLLw0=Gs((ITPhL_X{RkC(zwp^3Q0yH3Kf5vzy-V?-2={$fK z#SpgXoiw~y&?M$J62L;d&uyF44%$i6*x%_HIO3oxyCrXi2R3*~x_9Q2!S8l_*tG}E zO-mq7eKfcE@i->!nZPQa@&t4d(r3D|94Tk8s(&rnk-UI?yq*i%J|Z6`Q&;#XI~!C_ zxr2?zar|lfi+eUw9_IL2(;;sgy4RydY5m5tC8Bf7bK!5vLWRpPZ{i|)|0f@peHtbD zX1!VT=fPm{U^~q?;SZaywqV`ICJZ>l)9mS|co=(xwGBvP#ubbC;*a5Ap*V^T{pttF z+J5LgC5C#2=AlgIY)0_DniP0Y9`VuuWe>;IwXEy8}z2Ab#6)IH!7+)8FsF zr+@E*o4#HolwHIvJTn4ub9ELsUkbK#%o57`zk)$k$ABJYahCZ!v)Lyyu|CHMCXJ$- zn7r_B$zF;QcNwE1>Y?AJX8vD+ioi|TOffP*d%wu>b7qU5mA89PcFsZUSaTecG#NO) z?B*ZL$V7?kR9xnNfvHXi=l1D*<7e!A!3XvG$K`l#B`Xtev6uUhA}hq6*F;BztJiVL zs|}oX>?oS5AaDai8F)Rm1xKB$+}UM+NIs$kj>T>wW!qGW{E^3x*L;g_EK=BsfbS^3 zU?>y3rKy$NBx9Tpa=~py=vhCIfyzwbR&F}wj;>&8GS|p=qyy5*>;DJDH=!8$;@!$>&KM{^eZ#P1#vjugC_nStK zo$$UwoF9ulkCD@VQq)ivs{B5kT<&Z^UNM)xIG>>FqhHg)DWmYr7+cc4ti>vQ9zgO1 z8&F%emC}9~z*e8_`0Z~bxj3au21ll&opl%Pm_MD}={^NBC;z2`-nFD>6bqv>bHL#7 z1~R#+3<>#D@YBUD5Pd{|#9xOo+N4b6kQef!uFIg#%sNbNxClD`ZR9VEeM1W*@5r@g z9a)T3#zl2~!Q(>0x)jbU2rg@`aieUl@W!fHup(ZYx^zn^V&hQg zR~m&f?m!siCQ;&uXj zbvg%G(V{E%Rx-48g)oip5j_9*FmB#=i2rZRRc>&3e+Vwi<(})Svl@LFcH~kt?-VG- zowvRM)+q8DF16D6VIgSpOqBwhK7-@mK5Rrq8F&BT2(jzZ0p(tb^z`~cavJ7^cgH+{ zYr|K;udRdWt!5^+Op&n{3ZApa^;cQb_Ceg2WE&bW7#Dy*qZES40$V_GpbzKvh7p(U6pg#-WMat>sShb7b9g?k!3tuiCW)| zVt@J?a85`2&~$Ds3#x(uj=tCLr2qM-wfI&?*z%e9>Ujc zPAsxmACeMR+s*p7f(5Gvp`o+_vp7`3t|k{V^=E6Dlx8|=&gl|6H|IzmCFtsKfkiI( zAm&%h+0xq4SoAH7ZCY_f($O8nvbEd6#NQtr{D!m5`Fq&i!f@swNo6yA6>;CvqY!#O z7|u@&gOUj@%uHy)j0go5Zeu}quRU>DfH`LDJw@f$8foa@m9)EGBK;k%#|Ezm0j;hj z>>+=PF3t$S-#eaEOJ`2RV+~25p{vZ!sSRWM<%i?U;d9spy(2WYK_HJ=qsbuq1qSQZ z^ERrYt8$4vO7G@aM7J@-UMhn{++j#*&Swz|u9IQ89DAX5bmGLeSQb)i#9rq**(

oEmnc*Q!jTzZLPyyKgLmq$l7ox0N25%hwp55xFV_&ZP9Mj(F(BGV9 zlpGqycFl2w#JaB}=)_QLei@bojONE=DZzHJ1Ep#tGN!9lnVM1s99gjh<|l{QKltip zKO^z5z0#Sf#Ku}+k2o{^>bVmSj&=v*kGt52+uG!E#Dcs^HqfEfPhfX_0XVpd9@Y>g z<`(FL8kzSgXk{=QkX?ZfmmS29?a^YE`YALAiv72|$MEgQU}mf@dTV?dq4&>iktsg{ z15dR;eug0!u57^@1LoE^PM?Cqez?<3?+9*O@G3Ak*%u5QhXWTd0sq;*$HpZIaALDn z%|9OtwrSu~bk`L#k2RC9uf&{fxn0Eje3=Z+xyP}>Uk)2y`=Q#SAy|9o90VN}=Q^8C za6_M(LeP=j)IPgRygwYotK$!_(6t57aM+gAPvukn8?mRbWfNE0v>WRW=l~o481k>L zVcY@>Dt+^==EK8YSZmTm&t3revX8X!TrF5h-a=34Whk^t68Ezuoc*yvNKfqM_R4m` z+lxy`+PWJO=O2X^wUNA0@BlXUbQ(mnB4+npp1QtR;&+GXY-0O+F8}tnYWv)^H96>n z?7;=jD@TnLP5s6F+HtPtP*EnA+@}Nbzc}KSDdu1?YXIxKxq?d-F0er#d*IokbvShU zL9);tOB2r@g`N@}y2lN|?6GI@NFqm4A2+h~>lfkbs8igN)8F9F<}8@PM{>G2hIvaq z`i@WoBq=CxO%9?r8`(VcLwpE{eMn1&y}dC_zAb_ku6hAEDRDCLn3 zoHE+MQjP0qa)CZm7;z0AOJXU^$O_F8P06EKl^x88=As^VNz!Jy;4rsOaHcMjPv2)l zKh=4DB_B=$`i~=JX2Ff{5E)>?bh1{FqVW2CY*?Nuxjz|+voA*TMQ6W2=F0Efs{SS* z{j&^apD2YftAm6=m<3|s~?N=%)>#+z>B4O4?%aYlTJ zzBcR8>LJzaeCUcAi;ps_>$8h6R#AhPI3J!XW&m5S&w=85v&lg#j6E$oE9om6LMEET z&GBsk!xUvU-&*w51=hfy*N!wUQgp^WPR5;~E*N^|3sqTM;iGboVu_BfknO&nbLoA= zzc_ROZH`<(hnTU{KjICXD4YWF?+>zzA^NoB3xkepZG7U~>G*SFBW_w03^seZA;Wbm zn3uJ}40l(O8yLX&^UIjYW+fW4-UFKS#A~K#33#uXi_<#(!W8)A>~54uj2Yu+-6wzSlqHzjBj+by~0uDpD9J{-0$^ z=F+Jc;G`Q%4m>V9Cx zp0oXo5~V?Ek|rCO)Cwjad|^y!AQ!8t29ffPa9?UX1*|-T`3Y_qnzfaG;kJez*NS`X z(wQJt@eE_47tp)ssnjW#O7D{*cyHyo%rit}UC%j!m1~gnS7?!Z{7v4}@*_XHYLJlI zR75YVYPs#Y3k1p2S=4eNgO$h(hxjRbc+;jd{^_+?S~1g_)jUmRN{{ZtLx~oP7W<}G zPIR(?nlB*ENsl5;QZP9)1eVH(JDslkIAX+J?(FDc?0%99t~?S823rQP)8Pg9d7&}8 z{B8(cedt1qLVIAi*qkUhWlTjF_xv&VXHuoZw(>sw@2A_i*FQK zV~zJCZ-8G?BMKY*`C})O*-6Fo6svawssjXxj`H{Ftl!T#d$mkl6zW1>jt--cpVl>n z3pCjd&XU!>I>rAn>ZFTuGL%2GWG<3Vk14(HKbJ+myv2&6D{G!c6!M~sfAU@hf{G_$ z{&FcM3{+l5QJGpXY#;C#b^^^A6+ zq+(rkYTV3~FFnJ*b9oK<4~$^g;0fT+t`ClnX7jGPkDx{F06vb}ja!1E#rxwGY}!-> zYh;g*&WV?J@QoH)JsQUJTV2uEeIIRVTg39j*AGnB<{EEq=Xb9&hB!k@uAu5HRSl@Y ztR7b`^p{TL7DTY6V*wTZz|h_7M`h#ZG_jWMIxVLg?oflvyx@c?1G> zcg%*>Y1-^>a4;MAZ5worm0@?bhvNm?WN7xk1nu|tQ|Hra{CB$^?G19d{zuH=-};06 z=Pl`?w^@S)%}8d68CSSeWeNBWnGdIniveUmab6!Xad3kkQz-4k9xEQFLJ+;&o(fZ6 zr_%DjN!+5^SUfCONSizoxQN*PY;Ns9_?MN%f^DO)OQ(rm-A^Njtec!^R5Et`UBJEu zysl3Bxd~-IujVb8=#rec2FkXNWn*nuz_h_BXpTPFHNY?p|0z~^?n;k7$3 zrt%=R)V`*qE&VAbI*bOM?oZi;4&Yz6m>rlaI)eip_@z$Hq;q2^voDU2TvD6}Atpv( zeDn(c@o5kC$*y2i2P?w#!$xe&zj5?cMulB}(n@2aHAy3>1Fr3|femK9Fy7_|*SBUb z|0OY=JFB^v=I=U(+wxmkM&d|H@%at%@nLWx1|WW+EsXmS#<+$s{#DaT?$7$y;Q#Uk z=soP>vULhzZ%Hq%7H4BUw;fnUvKjuowG|r^6zRb1!_-uuE$sQEhvk3=Y_*p@Bb8XY*Qfafh@$13K3Quh(}4%aIVv=T-<`>;>!Mu z__zMaP(Qj2wR>N~&rwMfw`e5RKIkIn3B6D_?<5Wwas!RBjtW*{~5KSYE|FZlrGle<|!7`AAtp_U1@ZtM;rpH82n2C$vNT!238n!io#ajzCcH zfzy)%fc1TfZtrs``@#w#!zF-po-D^@=_7EU$QF8f)f?A;7|+@EJIf+&*bAS|MF_@q zYMh>59woGv;QO-&Yl2Sp5q@8^WX6kinZNdE)*h2V{~1eh#V<=?lxri``cQz}&T{^d z<2eusxlq*SsgNt?0F}|?~(GQTO_Nh zw8XjP3^wAB5&3MKL3(AS;9+jh)VAirG0TS;zELl!emb0+iAe zy~e*>7@lw8H}^{*>%Nw>+$Wyv&|d-fyJiSqXNY;wy~$*WIed|JD5A-E^V2e#a%X3hdvujz-_f6E1wWTW(Q)0-+eaIiA9U3VIm0QBJ5~k+zN1K z4{M~BYT%sdvshM5e<8M1SpSM*EtITrbEbn z%PbiCTOH#RZlgQxXRdE-vFcuwcpm%%jSm2(z8Oi~S?fq@+y+WJ8V-T)7Q|ItnJ zZNhR{d7;jI7Tf;Kg5~cY%~0eS_7mObZzmoCh5p%m)$3z)sq6u{+Fz>fJwKcc`+19- zq1>1Lu0MnSb_*EvsK21l-U>PTDy$&IL722N4zq0E0B-BUv?9jg)2UCmu!6t*Jpa4k zl5dK^cWtTBdl~(B`JY7U=ueQ+^Mihi-003}WqSLh0l&Pnq+enNCh)3w{%!jXrxj;X z{H4{ROWgpTpUVYjiFiL8IasK>vzY5Rd>K=sW-t@o047(R1}SZ8=y?Uq_ef=yc;FY#l6M*<^?s^@q*!A5t-1BugWL!Cf zmbjL{CDS6Rr>9JA!x>61iKFAb#(4E0wizK4#H z;f;r|#`z?SZry-O+l{GaS{rX5TMPp;KEZ#L8(F~^sflvY`=MJl8?KCb1^zD{qM3^e zTOrO#Hr3gnSG|A_=O)sgzfM9>aX7B86M5j%?!geBXI!9$8b8|k1bVN0hdI?^hvVfC ziKhE?`n1LvhmWqJ=_+@*ppEf@&B++x_rGVM-j6|rMRCy@1DU9X$K)?*RCIJ4gbnbf zw@r6w+}(1nL|c}bw&cRJVf$&djUI!HFuu#|JuY|4LDeLADlT`#!#2mMZJ(O(=RX9s=j>wxEiHn1hX!dRoBHOGp^F=3V& z>tHgh`bHxh-772fDM=^)JD>Qh^11M~^Eni|noD9{wo#KrfbpC;D}ENiMipRs;E2nphNcu-eIMpy2#-HRWV(7C&Dt z*sC%9n7R1@{@%|@(ZzF+`K}xxqy}iwtmw<+Eq{a7>W0*eUKR?aLll{NUIBFI-H~il zyh-}mZk+U}v8-*uF7{779q;8DV!d~qpqptZj2D@p@4ou;F3;OB{X!wwEI%vEztBdq z2~(h=MT{^s*xMT_r8F=iw?rz8An7Pha%oG8_Sj-?7|Y&a3&nc zfz{g6#N2s*`Y)}B|1VQKpL;|=nqe^+HC=~4L6>2`PYpWX7QmNIjKS2yMJQJsNK)Oe zQ03fM+O$oLMLTU|9-4c&3+JDTnYA?ftk(_4-QU7zaR+-i=0C6-yo}9{+>g^A4rjlP zxN)l_JWXDvM2lC7dHfxd_=}6}n8(v~?8Dvjw6X3Enss?{@j5=#lplc&ttzle%!$H{ z8eEZ|gUcQFaz)~qUSYPYFh-^uE$*8NGsAfLS$qqdic4r_#tI1gm%zq`iyq(RSM+86 zf8?4s0yPR;sK;eEKmXr(__=!qZLeaKb#6Pf=nZAnmlolX$5B+9IE-0g2AjL>Ib^>( z1!H!1+6}#03o)A~lgE26(mCnM?KtRGbK5%(ZVmh^`W0VdiI@vtdUm8>zuyqPD=c7( zEqI*M(FPy5b6|F9DynEGVez8}c&}cYWleaCtw!^3 zkp%X)3NXF&9W1s!L2JG%3H`%e>3V}HTlwZEccx$w|HV)pQ|_CyTNV4T&h9WSI2O)Q zpQ{Lk`jdnU^>$Fc;~;HKxq{l;$4X2$DWlxLiQI)lH|fK`HB@~?6aSo<3z`coxYL_T zQ0D72<`!ARy|G)%{Ebq0ox~%s?9X8~^;!wckaeY#wcp|F5i8-yC|RL;*#q+2q>s18 zZee?@hp@`JyJYmgY`rj!?B4FB?*=iH7T-en5}++P3LgcQL*zt#Zh=EM#Xry$cW9>} zA@~zGiSw+|8L8~UUk7~txB|O0B5Bx#NP3gLqGpeBAqAD6f-1X3fXf*=<{^}`Jydi~T|y;!G4t@YFEutV<~ynlNU`k;COSQa**p8#_tWpBlH5}? z4%LOwi$^(mvn!C6J%&Xo9fsr{OYUFESNLzJ349B9j+(Cf=tS-uko#>+Z!8B2rir58 zaN14LS2_(`kC(E>BTr~%v5tU$6WRG;e_^Xu4;~2G0sV*ksPW2P%7(v-hH{N?I5IK= z_5LML?qYpm_%&r=q^A$3U7#sw`z#es4b7yWvnRM6D<*N8awpirPXk!Xf@oGO@}ACu zBRT|JgnYXRAp1{|{ScbK>D3^1{me_W8R^N6`LDsb5i4*VTgZ98bA)Fd>Z~sA1^8z? zp$AHX$zJRLzaP{J;Ukw}&$*4&=_>a8!z=s1^3Fm!VBG^RmHg1@+gI4@>q1tgW^gd@ zEB{k^IIHiQ0I7K!$o;-DJ8*Lk1V0Jqb+>D=%{w1r=pq51P54Ru$5_!Ne-9k~eF@Gy zZVBzbR)K8rCP|CMdq~yE2UBg4BlYMh|9bxze7k86iy1M65A7Zc<6;L3p?i+vs8kb} zvxkSgxCs2_9K-D?*(y18CLAKfK9gl;G%THD1=73r!rnfOcvF&tPmBiOHP>Cd<=c6j zs@GIJZhL}zE!R}DIc5qDUvQhV_S69y)B*#7{m3)=2!54|r1jU-D7~T?M_f~3Z@#|= zOW8CCP|AP?CpWgD#T1n7pVl}<@R;l|4g)=+!013I4Tw08P95QRmBxYL_X3!d@CMuq zBUy6BFr3@t0qW9;^hFW?$DEF^ssI&=Umpd%4yLSe^C98PqNk{B-H7`-)hSuk297qy zq3YmHygmsb%WWvsJ{ve+si@f%mLBQt42nvKH{j!$`mO6y&9gqSnu{$lGHzsbGvX4 zjE_{n53LpaFwb^qyc%59I{h2E-u{3oL;JFxJJM`%tv%_bhI5?@H*g#J9b{G$643Tl zH&^C-6J>*~;rxL$XjB)#TfM!^dwx}8wzZM0Q}qnmoVUe8#|m&6ZvsWvGT`z!ec{pI z;X=}X-{?ZfNq+sLy?jacZE&-X#h*L!VD^~t^i}gFhCVyY|CZ#y&lhJ<@plLn4|xv5 z%wtj0W(l2s>|68eiKZ~~)&;z|eQTY}EgoDe{oRPky?MsI4rzgzJUw9>z?Jm)y!WpuDUi=687~{!%XIZ#EcF)$Td;Irv3Q@5};t6zGkM zO4m^fodl(wInR_hnyS&Ed>0s1LzYTX&dL_$SLO8|L=>#kbQpHc2 zr0u@YZ+jTu{bn`Xn7J3dMSkj?vzD-Y3MT7eweTy2R&R* zTX!kb?M{)i^38?j*P26L%>dptdoNl1ON91W>!4(ODpNV2#_#yyj^EQ8(9Y5T_8vS3 zjvCLX`-48Di50iI#|&wuMG@~<63qrpJx9Hpe%4f7oy-q(4|b)_N%wI_ksO~k<2CpBKoZRhN`~JHvM5w} zfMC8~f>ZW%K)m@RI=n236sF5l=NfB@c8%sW0v*7tRLp!-v~UrVJHTsNIc@hW#_PfW z*3dP8^>e%hgINv(}%%`9Cm5-2iBny3-3wX}k6OC>4*mSuB`0?;D zO6#@rsoxCPGqc-t?6(G~i0|nop5Gzw;V?|8j|P zUm3HA?b1wraXhcIx)we^ErSck4MDDXAJ?_yCfpWXS-SjtxHs?zKe>7kRW_-y$`5|b z$j^#deJN(8{1E1yTu9D`gK+c2=MXBjlg+45ValK8;jIx;!nB!&Fzxhf?s1>fgn9ol z-@|U~sGK{}{Xm#$7$YOO4sbLVlM*x>g%0bJ^hGqRsg6 zeaYGv>F&fj`u@yC@GIRWF%h4o#xe1Xa^f!UzPt{sg{iP+O+M}U7s0P;5;;SU%VF|I z4`@9w7L?T6c**-oyxr!9wCL6&7!Z+8`@fGAGP2vipWjryq@<)Mi;FdH8iWo`Q5v#&bNr&6|L~j4LQ#?jYM`)quI-o2kLKfVCvV z()2l1q<&}~ja_nuU!@+0Pc|&Zk)Orqy4X6B-~JR-vmVrRU0ul=DgLG~e?!>O=qK4z zIfFgOjpbYi1W>Q~I6+D#1)>dFv9zI_N zq$jBmt=LYx+ANr_Un5nF+d&eGo%pHo3awk@F3e0bp^nw{{3pjR+YN zK0bteyl?O?#EhzZOA>D&rAChvg7C?qdzkHI&iekG#b>3dGTrn9KA^ya$#>e~PE7~c z(>ou_jw|Bw7Io22aRfBQ=K_Pde7}Y6m235A|tr^U=ep;wRar`r{H@O1*eIlMnCp07r8cU7UrYYjnv z@;>fbQ8ItgSq?JZj%9((Equf?8LD>s4K6PyahHFL!?|ZunSC&byUC+$NZl)`d=f0g4_?bUWz}tWowBm9Ly=rWx-2sjwTPqzs-B;0F1q*h( zE{e0wG^Pec3--)ijrw^#$C0UakXjzY1#H;@MQ1}&PPzz($lo;<3enB5zX9v z)Tr$7KYXdBPVu^a)aoGmpijL=gWS20aj^)=;W2+j%xXKv*P!-cVx_JTY|{0CthVwP zcBdIJ!veuQB9MG5Nl;W_On~K#Fn)>g~41`1*zD z+He}na{RqTyFi~{9i6$33NZJh2uBM#SH&DN|73c6NGHI=#T)h z4fAN1y^nBr!5;`dJ07iN-qD0*D_ObfCvebN$ba)+&b~I>=54=qk@|&VHYM^Av}q-f z&BY4fmzcs8ahEcCY7V#uI-u6pR9atd0DptSVXNm@`m|~{+j6iF&pgjT?SRW*v`&{k z-Ca%#u9$I|%{J(;OLWAiwsTcGKH>Z`mr!TSV(PI@kyKon0vkhSv8d#D$o{XC`FH(6 z?{%(VIXV)!-rG-g5tc7VW`(d9^H(*u4KeS&OD)L2l@*VyG zH0I^>xWs|G>3&+SSPD}$gE=j`Vo>yS5c)5F%71a{ zFMLrr!%x=t*LWE~+p7$6{wIv$GLxY;3YAdJc-hK4p*aa##o4I+Bg?e}O+A zH4cJGMD|3sD{ELjQ207BoF6W7rleCYumktH__k&KY__))N!>F-pPhXqZZj1WE8|sn zNga}s`ro(POc_}zDMdx`r~l_`pcE&yJkZC_f9ooXmCJ*cTa2@`o@i@3ZsPcX)AbVqY(U_yjK*>TH=o&SwfRTHzOF z6x+eZiB_~KTZ214em9=TP*f8i<&6b-Hm@Zww zfBCu#r+E*i`u@$RbgLe>E)3>;=O|O+ta|L3`4cN;q)=#c#;Ln5;?PSs@ctJS(z!TW zq=IV$?dtcXKN%LZmLA`>q`a=hw1Z0L-Uj(O!b%mmwu*IJCFN< zg+pa&V@C;Ryz?8*p54uFs@RV!Ju1*3Cjz5J-@tF>iMV!OKeE5^16O={$R)3r#edVT z@C$coQFp}>nqsPlRaquj=scAs^%MEF2fF#btqRn7*#}Y&Xwx%XhdB{<@Ora6Io3*1 zgO4V)y;CN;ytVwM+>4TQ?>zMTk;6%bF09cg55>Flaxu~T1Rp&k4M)B|fggqhz{Ipv z-cjKUZeD9cNrSa%SklU+^1Gm{PZ^w2WeYULtYhDJ)BEDjmN>~br1S> zpTe%tnN%X_z;@Sme!*d7JfA;?7HJ(rrB`YAYQRhUbEgcYH#eZ)<2>|%I5gd~u_m$U zAJ<$)Xt!%He#*<|&vp#qt0#=duiIU5K*J-npQs8k&!%A9rhSOYm-*t$GSuxOO{*W4 zV{)Q0UV!HqC4HCw?RW+UIxeHH2i&Mbu|FL?cM|SNw5j+7M_M~7na_)PbVsk9<(ddo ztD+#(c+a34yb4RH^&*FQEkXKE6g^N-!^>+8xH$vE$16py6rkd*G|?2alJJ7AiMJW3_P((hj}xD*!1*caC@sF95-FS z`8Vd%OZfwUUFIxkzZP3~Q;OAWFJ~>^pR)X1J*MVch;tj|;r%gfwsraqmg+r}P97`a zqLwT2M^5Lleot1Ap=vul*$@U_Q=ai^AKZ!S)52!RkAXP9r)=IbEw(Yy7N^I2V5e&> zgyqA#_-Tn|?66cjjhz36+Y%z4A&nE*hBbm@fOR~(J@f|1*2VDoTpgy=I&tGplbkeUG<)3TP=+qNJtVsrKs#X`Ko`|3ptNmn2VlI7mhos=k zKR$Wtf8f?r0x>_$=+Q=1``qzk_-k4BaE_J(>s$PhTl0D%Rt+9W3U`{o({&}gqqLdS z?7FDx=w4b^>;dC%$O@;E9BAF(N^0wj;*`y$YmBN+plQY0BW`})X`6$%-7$ht`22httx?Ac21Sq zb)9AJdllH4+C^+jkTJ`(7xVI8cSCDxBmJJMBkWwCz!^7qlgS_*xH2gj<0Vblcji>Q z9?{G_R?c9$Vz%<-;#m4rGnH-k^g{nzQD~C-0{dJHBj+`CXzFSy5bBz#&Rdl?)0;WutV!P@OvT5c?oTa}Ug(sY(wAtUW zyu?hrUaZMa)1Ps!5#%MSz`L29!3=TsRW0U+A61-WV^Uv%+qD$(J#(3)Ryv?Vk}dlh z&e^N%T#rZmdf1_p*;FKBkKa$5LQ`!AG+aH#AK9Eny3fSirQuw<==qQLeGnn67~#N_ zpIET(4l~epwHqJNew5ml0Lt~v2M<{vHran6zv{t2nyP({K6L(}-+M&=x}g`_71NJt z>~mqCsc?IRw5Ma+|SRls-+%%!q>|W)9r3K;pYz{I&fs0$bOH8^{djT zOuX*9JzeOiRRKkq+Y!VX&_Bn)HLuJ+bAOedvzwQefy|GwOkOdTT@PE#-LA_b_XV0l zidQ1DJ#oLr_LDA)n>B}u{61lH+k5;Z0U>R?xck|pO42$X=}E^ZI^1H%4|{(dp14eb zk$p!o(;cc5`g|fC&&XiA;#J{s%tq$cm@WAsx;>8ZF7!9HlH}^5SoS>=e(8o}Y6+4N z3O|N1srqB=Ue#`<@0|%-CFQJOzZrY$vWfyy-|~kFTcJdKAAc&vkWTL~r^bGXkg`%5 z_Zf*?!M!`!L5=xL$w&s@yz0b*`x3cXW6A`zBXYvJw3$pT>;e04Km&ZQXlJh`{D%r! zs?_Qv&Sz5Y;DqQi_&GiZJT7ezo=sd$pM)skOzupU?>wC?DV$1GH+wKyL5p36`zgwl_VrrXC}L&-1EKh>%%D2t}#jdM>C);z6xJLe&O{x}{Ed2<5qT}gg7 z(`ljeUD&9S&s5foWkEsKY>$|Ka9x$ke)#!7*`hl1F(1G#IUOTYbrU$gVWsF{ctt&| zjPe7Og`Ba!s_VxF;>1HW#%O5q34J!G?^7$m@CuY;GRZG!>pyAL_qt88y%VWde z7*I{1AsdmIM$?{erFkEv=xUQM>%(uQ1yj>NH>{d+_UU7saxYAND@{57PO$anR?OD7 zl6_GnoR!`O+C677b(4ugy2mkI*C2x_8ty0E=%ZAjRY#fy5yH?sH;OV`DXiJ`T9Rg6 zgqy_~{i<1$IenQWBngNjtzH$(Y@9@MX6BN6p)VW0tO%F)Mqu^L!*t&+6kOL%5PitE zsm4x@(xjhCtol8M2D=p4R4|R!Hl~t(Ycmwo%_ol`8`!Xzvn1af%0`}DC=@n`yu9o; zoPqRgjP3o*Xa5m9?T6g>bf*`v{CqLocG@pGFq-kN%_THA7Yups8z3mJ7oXa!pjIy( zR~*>yo6*03S$HD7ob!a*;&oZ@pnbexBlevdTWFc_He7hQ8UiXu;2g0FGhQi+ z9Sy&St6xp24)9pZUhdN2Mo(BzJ8p~+HW}U$oga}nEWR7RTx(@|FAAV7ZwHiE&!ECX z)wJ!}2aq}W6HMifV8;?QHoPc>b;;C%+2nRuyKfsidFVNY?4Khj&3^?N;yk0}bpfq? zSkA@|_C&>tJluMzzz@Fn3aWFCYGWI!r%ZF=_`Sye)3GGLCiu9{7RuBzo-4w1e8hYD^!Mz672r7g#C+` zfmde;eCRR4_KqXWOsbIe6d2p@TQigGd~~5^g~$)N8_^G2f7aHF_?af8RuJ_5j%EL5 zb+OZ5wAtQ06WNE(yTtivaD8TFLwZaTYWAqI>qZ*P!EhMe-R(d@mhsdvV;=aV%!e|w z2-p!(Mw`k+=Vo#@O)67kWgq8|jK7%CuJvWfX3>^~NE8d9&4_wCL(hGSajYETe}?j`yqLj$ay2az02EaB|nN4AAEx|uLj_Y z4Uts*ZWnbP?MsctR`_|w4rtM7wh#$!MuB?cv~)z#wnBu zy#thlb2Swt1dm{2BWy4t;5c5KGgKI|M_1UhZGa%Rcnc|>^2NEgve>Q#vivXpHXhP9 zq`}%1=u)>1z2oF~pSl}0N_!{cUdK*YoA(}k7wWT08E;C@IM2>?41(F;B1tjKg0m{` zfRExXZrL=E!}I$tQNUs3oh^i6GBT9z_8iGZp3+m&(Kg4KB`PP=zeD%&zu`+UL34(n zEPb3wZMx2`O}$BXN|ew;HH@7cCi2T?d$Z?ZJL&U@sW3TJ9;Aw`#LePTTDj|PP2U-( zV0DL_{f0y1S$Q8LGU0C0^qmK}`j7yo>)`-Z{4DAnJy6)TPs~D8UFTE#{=fG1KC=wMKiXWk|H1Qlp67GUeP7q__x+l{$JOcZNt1V7v~hdR-A^s!oJ~qGFY+4Tx|5vDi#{w~{E-;? zIkAK7V(`K=5=10}iTG0)%PYnf^!X7Du02!~EQJTaW3M8{1f=5cotm6nQ591ia+yBU zsb+WeMZs610wp1RWO{QFYgtv(5US(F9K3Bo>$)ePc7hZB`PNJuEyuyxlogOXZ7Z0n z+0(c!d+>0QG){Mr1d$^WB(X`0oBGe54i@FXU$aQkpc_t4EfMFQuZiVq-yg>Aq3>w1 z?HbuXP)hIZ9l(#pid<96R1(5-p#M3Ff_n2T+%qs2-p!8?q$PFZ=6DfqV)Y1|(UGB( zwpxMv-B)Nf?I9hQT|-|DCXhSSf>jc%$G(v_mUna?QuC~jgeL6gcx4Xr?T`YjJdwp7 z5)Xn$pC#a4i4HWn7onF%ElQj;CJvkYnQwbHfYifM{2+W3ViWe0Et5sialA1hdh+!Z zSGuwL$|Lr?y1eDN5)YW*qJ_yykLj8%L%8Q0!??->;xWT)2>Dn|ioG|o1^u^ay1`NA z?J;$absHFnVhE?Is|Uvp9-=?!7rJW8H7c{BmnygAgTBi*=FHQ_ zM2E}bp5;u!bbbbLB>E}w70hDh{r8cGEc!!j^uyWFjdP&T_6r^_%7*okI{4<*b9y3O zroQ9MPF(YcpEs!6l4CXLD2)rD``0T{r*|H93*~UL2AoN|ivSe*+Bhq#U#~Afsd5~s3Cuwwa2KduRHS=;rLxcRw}6h}8QN?% z4-Q{Y03ZGRkgQrqJ{8o`#ZW%bt}mm zhY5I0yNqtrw!s@i^>}!e5I5@l8e{)iV!r+b9FuZ}^N)*!BAx{!(e?_3X0C=2-s@f) zSwSz<1<;CUC-7|h3F~Xsn6j}i*axePxW&Aee#@vEzW;9mcV4rF(O1)hq$WKm47kc3 zRzK5Vy)Fe^mjOPuxkX!o#&H8L9<$41^)NyB49$vAWB2n}m6)h_(C;qChZi4!kH%PX zt2CGIjp$)(P9=Gewv_h@I&yk>R1?FJg7sH z-x%X7zvs-}((_lLis|N+!XvDde5k3S#5Wl0-M;9LLY)%(vfVB`(*ayY?^CXi$M$S!22O&2Pcj z@e|3e^TNJTQSRG~KycN{!O!dK(8I_DtW+0qzwg~ded8pilxIz-L@$I*Q*N*icn{<8 z<>vUW;US}vlT6O4hr_?!3>Z0=vdOGDXj$I@n+64#_iipV*Xw0Huh`HpAN`nzA;nZ{ z;!n%|pSG;X&O+$;8^USsyN(?R(P(t{0R7wQ!jv}Uqg25PQnTv~)B5u|`Jy=&S=ly5 zk2W*Kj;qK)p0D#Ve?DEe-kq8i{G(+0Tyk=M0Zpw{!j(DpfV7($(zeh^Y_2L56RlQ8(QsszFLJD~? zup4FONwUj~7jR3LdZFH{C7`YAiqrm!LZ)C9Jn8LX7MM%G%Z{h`Cb$=7|Gi868a8u} z^=Coue~Y+I=PEX}bS*yo8G*xI^Dy{SI}`Y06Nb$l$6dDAf*pGvm~Yzil8gA~RTRczT+y5q0wK?JVX*dD{4~Ny8oAaEA93k#h%}r{@KlhH!$tN4O zJjdo;op@2_6CVBXi>|(Kid0vAhY#6&W+z!39BQu-v&4IJ`MxAvv)vAxuJ1utDveHo z#`yCtV)YjnOo4ITf?!10RRWUsAfgLaTUeAqGrW3wNT-aK^_ zFSvrc632pt$~nRrhG2WT5+~ChhgpULbo2a3m^I@ck$cVa?J5#!$MXPseqt%T{cDzB zpY18S_~HtbDjOqM;#N#vD|>)XCc{`b1=0p5N^1VQ4ljK+K;P1f0LP=SfAL;CV=BUx z96Jw8KsYQsDZyRd_=`Q^`;@r74I|#>;W#fUo$mDB2VcLxhHjWE5S7su{JYm*Z( z(DnoV9Be{KoJLKgvm$ z)wSfPy&SIm(20%#3-I_hHKz1!556g#%q{2V$s*M%@Y*A>p-WK)wmAz(;Eegu8>o%n zOFt3!+m%#D_ddjE$yJ$^WRbE8+EhuQgypt|i<6u%d5I$U zJ{_Q2XU(ITtAd%SYUc3PWG&RV{6zaDJ@l=H5@t!3k@!6kmWSTy#rU&+$a9otKrY<1~@!tFMLWI4~D;nsqEwe zR+o3CD@w{k>*TFu=A<-c#Pbqxe@VS-XJO>=vinu7& zkcdwXVUB4RG}ON==8WEn5e-QV@H!F<$4f2Y^X#z#+eHR6`>is)pMMWEF3;m4k7zMf zVOIDcA_5KLTiDKE5%%}u^PJ8BX>^?8ic@5i1UCoA5&Oz>RP;nP7jG3nnxwyyUDIS4 zWB{!TD@0y|eHG%q8 z#9!bXD+dP~n|O2WUG|@+J!mA=kSi~rk`tnnQ8n!h^*k~J5_620DANO*IX^5{ zd+r6fKXx; zvd){)t7qk4xj`)MnJI(zBHi@E=Pk@PpM(6&JRLv#Z-G}Ucn-6)7tCr@#O^|26pJ$D zR=%>pw}!eXB)$Omdu@cyv6soQ%PNA{>?9g;`vqTY(BmwtS;kOl0uD<5#jh(zq0%;n zSZXtZN9vNGotVP-P1h5+XU<&cg(#%TF*+^l>5aKmC5PI_h`o(W2#pQpu9 zvD7del-)vS*w3LXf4+?fHH8&@yw}=hE7{-K!FS^oxF5$ZVr34WFEf8b9M8ycCIUjY zt6abcK9_8^ItQ#9y8??E)qRFYTv|D-&-7_T=H&Gn6uN=qHYGYC)cAI%y_Kcp~F&>Yr^)pp!HF*EBEliuRnw|}5rA@c17$x10)KIPj zQ-=NnX4+b;pVWgnb4IDhXfAyJum%1M%;O|DH3++Z8oVrr;my)7q`BW%AnS62sVfen zX+)bXRGcPg7&5Z-v0rDYxs=00e{a(Tf4{>wna^~cvj~YPQKwY~5vZIjL>u@7?`y?c z9Aln}uf(okqn#^AfB8lp4~t+tBvQR9S$tWViYp_tU{T_8GO{ZlB0rbo*Ejk6NyM1D z+hPRpF$VS3BshV04|`y9JW29jjowApX!EI>_$)6(|7%a^SB7V}?KsXXc$9%dia!|1 zBTKM%0pH`8t&jCP36c;^sQPV;O~0n`IoA>DE2Z2J?&8m^*8PHBe%{nRy$C*wWMY{~ z0p?HJinS>@cygsUk}YEdKkfIT?xtws*t!s&jpKWcs-^URYaRJzHi5fW^_Pragg7}} zjx+5GCx*P|&E`Q2lx`VIqg%77`l&-$FlO`RRdCCb?`dg| z#i<^>xYRQQI&SQO(wJ3vj*$}-$H)jmyv{&>>q|Ofr-#nR`e5!;Q?Bup5I4A78s0uL zXQj-S;sLc>h&{fHB>X&xQ-1dojWd-vLHGbOLFGOHt+&2WEe`KlwB?h?|{M!TH57K3y=#Hv^2RWk45wKI;iE-~6c6hF;K`SPDLU zZcwK?jyqW@3-kZY$1jEYSfQVQ$5s^+|Lv8iu4%#)bE%-kJ8g%gjgaJtf!~u?&_AXh zg!5KGQ@R8(*7PSc1GizJ(+-SV=)n1;H$m^EOzQk80}YgiG2vJ;Xvn^z(XOkg(24|z zn;pm8lK{Hvg#dhus-b938YwR8gjX*R6QyiP+QZMt=E$PX`^(G$V^B7V$>hM6CiQu(1^ zBAGo6QkAYS?QV7;J$*X{r`6)73lGpkEsk6|FifW8tDr) zoP<-MqNN@j-Ucy^OO^3kju_{h_ZyEU+-G$igmHA=4K(5h_M+nen~RS!221kA7HAcvy}3iRUp!z5>g5_L$^w5z$|6 z4;gSA^#T(t&n>@6%l=D1*Qa-AwJJaVcD@Lz>3@iRVi1jfc9Y8ZuEIDe0XKVxJ3T4U zPXeSONnpn|5)+`trZ?Wg=^Yo~ne15nXc7X~XGP+UgSzBb@fncM+79L~9+O+DMbM!y z26tAhWEJWRXglwbS-aJgn^G;!wTceFgLWe*-F6$T#EPj;O(wY?KO6qN(F7r?Msp6y z;jvIlYU7kcbv`Y|bMJ~k(NdRsUrvKL@@<%L-V|L6B++3vzYn~u#z^#>sK3W#z{Ukg z=ZPPpyZ*~UZ!s@;yg@*lTMTH#?q1XjJ%y>JTfuXFIeC<~p8jvd55k-Zbc6gi4FAYf7NuS;|6lOO&WY+Rk(-0Y;ngSGfZ_)!c9@b zpnI#G8F^=mnUv}>aR zrz^f5$Cq!UvsF__%RE)?L5VO~n!f|OFNxzM^{Lb&(THB2S%>BI2QZ-Dig%GHLCh*4 zMxq8;|8qPuAG_fEcr$K=!BfzBHi?*9r^CWQ3RkqgvVGBgrs> zuBRqF^Liogq#%U^S--kzH86VM@I7^;GgAdS|&McDdMFD$Hl_&(3#1wNA(0vlj$Rg zFiPi0Ji$>;0^7ds$HB=P*&IGUvGB=06xP|yZIOI|4*OSPR_0wa5UXa_`fFiWg&i~W z={bydJ`K8-_Qc1-odzU0qwSF#=>OAARV=QMeUIZhL)sJ8)0*79dj4ow#efV{{!TN$zS4y?!+t3gO>5Vh#M!q!tmZ| zIx*TCGUQW8Tahc6d_QPe`sOBGaCk9_{Mk-z4M&Jo=L+a9ZGn3MSGmESxmcDwKo>6A z0j{r8c<%NBs0+5n4_~_J`!Ub(=|Wi$$~_93oz*yTyD(gy?S{J24b-H;0(MJ-n~PohE)&#PapFXx>nY|GKil>z6B>mtBUg(~GFtfpFSOx8lc> z70~~ABK!+Gih@n^SnFdV-0(?5TyiZ5|Lc89-OKv;8S{30#*{E6)q5f0^et@Zw7_4o zGN?0`|IL<$qv5O!)R2i|3ijm?c9Q__T6>`Eu1y$qW;VVKn}e-0n^;lxXJoyzA3AIm zV5v=LgZ0(J4qA8-kOjR^pl0 zQEZQ(nNHp{hYK8x#FPvhsN51y24-xq%(!na4%A+Bkd$q9RbzfBEr)qf6nJ#Hlvx}WSjw33! zeuEnU)4}1xU78a16@9-nVeP4Nq+;VfwDA4L{E9B3nhN>wz(5HagGB{NGta{lEhDsB ze+NTlGsu*n1w=lXrO=TEcPj+!p~J!G8g!D0yP?7qfBeFBgfYyWP4-Yww!@0p*6V+{hWk@0%ZBOu}p`^x2M@($nOar3px!6(@#Mm%u;u$M8i*1U~=01v08; z+|b+?G@+f(>zVpnJ}de||DDmnM=s;Bje81nw}ivu%tf629!pf(`HUE+iUYaQAC z{9(h)_BLGkQWG4f0efC9WDL50Q0qm(sB*Q5Ox$^jO241Y{f*b+hO)cp zg(w9WU$4O>l}b`-wY#R9R}oO_`>-DIx-aRPqTGcN{9XU7mkjKyuZ5+bOrUz~C~Z<4$8}2u;g^H^K;?)JyOO_C$+CMO z)~;KqyMIxkfxWQxr#0V&yozsi-J$Vb24pv7W4UxJw|{yxYAJ4l<|PW~o(aTg zzaf1buTE3-tkA4n99`Gu(ZtP2CFjSWSBE#Q_g1EjdtAtiE-!H7`R+?lj7?O{f&6e` z?$5?PqMG&q0{-*E2^$j8WU2r^e!T+wMIO_@q7aCkKIk~O9oOzKA7 zzc}+S@DFD2TK|B?%pZeCcV*DdB@_Lr(mv?i}e4ZK~p5FUQ%1cQs0aCY}B zyr5|Vp?tRE;B*bLWzuEx{#!AoFIDO=Aoe z5%;QaKGXPt;WoSgkDpCsZEh7dQUMp5Bn8&fPeYN(Y#1Bt13im}Kwf4l*ZKJzDpWM! zU%7NL?fX_{&+6l_NhKV|hS_7#A%FCip9q85U&x>BLZD@M2gH{2)6W5!WaMHn`%x;X zVa#nsSoUBF{8PO~+p>mHz0Mo2z7ghBZ*;Q_%N)5!QT9xtW-yslHW8%1U2V7&BhH!k z^h4BbD`;_?z-C$`V8wAa`o=Dgg4;O8*%dKXG#m$_@*(S@2={t;3g_UH1r5)}purwD zc3bCkoG@BOLzE*J1KLOq?#d=|pI^`qpEQ|ey;B4i7v+OVVFLcXGM5wW{(~N7Qk?U3 z#QUwMY0E5a%=6~Y;*OWFjfupAIxM~Y&I%Vz=P+gJeLC`^mL7inh8&amNkgoYXtJP zYe~=Yd6co+|`p66;41ZHkE{U$Xc7u)y%fU*AY&c(;Mu)0b zVa{MG-{l(v_Dl-t-g6D?=kLdsBuzGD!(+m2hy`)(8X4b9$$|O2`z7`n_|3fxxk=6B zZmcjZT;z|gL&up_B`Wa9R2A2r+>e?2m8jy5bNoHA7RHJUl0fekoaNz zRAP_TrZ0KN$V*tZJ{i`EiJ&o`Dh^)|t=*_3mmZsCR#&Nh= zwU6A0oeF11R$JOH)T0~Arb2*lJqEpNqW#Y-p|K|c`&+E>`X@8oc!bY`ZMw?TCwrpN zGIMtNT;6kHG>bltiG@cqthq$CU2>DbEcFa&|LKKA ze`Gk#CjxTNDje0Gg`oHx0~jz-f_3gHoc|%7Z#i_3Iou|O8Dc7e&+6?!Qq%-3Dn1|| zaHrnmbv`|A^&Y%UrHQNUeDW0{*wU(nc*Me-5m$-9I*oezc;6IGXSqznw;g>X#?K!= z1_Y6l9~R-W>re2=U!EO4?FH?x*TtvZ3%MU@irAm^g;)&>alfX9(x$U>@Jj7C!D2%_ z(5#C9J;OtIVVMa=&b4PfpSI#76+0%Vt%Ia#2SH=nGa7GX1P>Li@LX9bnjaC54?O_3 zUn(XoKYf|a7v|%u*|iNaGKc8Qd_MDfv$|o?L_-(}3B%RhX&~by2hk#t5L>pHSl{@{ z+{@buvV8`Yb1S03*(jT6IBcK~7VgI1?N(IL=n);a*pCP+TtJJmCrsjs3;4Q086{rm zfRjobW1?tHCaB4S`CI|L_s*8?zH859Kdq(8J(Hntk{#ouT}zr`#Y7Q9p?>au3kNCxhPiR>S)RTexfXpGnc2IbEE~3B+syi z{eE{l->)r#UGMMk-roXZqj-;wJlz3j#Ujb{%+qkzIfhNIFoe7V@38jRd6M}so2aRD zq29GfwAH%JFMn#_s`7oB8np)3uoDHdWdzW8mZi@Xi(tc`<#S@X>uAc=v(zEK9wvI}q31&jbld8E`NqYc4H=?h=^^+)dd?^(@;$^KatODJH}DYc^7^Q&!VbF(=5&EJK(-DL;qXD^UdD#?3=IuAZ-6>KmBkSO-JFFTI~K{g&BW^V5C}!;omIeuhOXH+JhUR>*g9d zd}Dx~*nSs;yj=MEg}R{1L4@6Ve=P0v+Cj=)yWrpIPU7Kto3v#GQST&ms@%8+qW9j! zVa2_;iaaDzC*4T-;6>^;P7zp#9(cY`lP0Nt!>SNBwo@n(1}pF2@vZ8d^C2Y+U*JRN z5l1wMGsH!Khj0=<3t5*HL>$&-lHZpjQAEQ77e7wKdw#9><EKhM_v|c+CG|F-hcGVNS%_IrFYu1>N$7gg z786feaH(RKSTEiqa@X$(?VFw@{ubvwcG-GC$gHTxQ#c3lcdsEYN6KS4ysa(L4ajxB$!@kgr)^;VMw>0)K9 zojDzDaw9}PQk~m%v4QAke`BJ5>B5KDT=*gzMej;}hbxuCbexba92$R0_hpA4*I9mAFTuXvZ9{UVe8!)jVj!yI0X?hG zNdMzAkJk5!ac0Xke6`4mTJ!w_-P5OW^M_gBXL=Qs)RIwoQ7jsqcII{@n?V97&>Jq% zcryeU@gtc$$J-KK?hd7Ae&jdY@d>9l2DM@C{bVTm{vCte8_ft?3ZE-V;1R0>r^`#J zR&*NK+nPajuh~N4(Rk8g^OmGOHio$?r;^O-LuB1O1+YA>2NV1DapL~}8s78YW%r3G zT)ol^ykoK$YOu-I7K5vnV@aqawAU1q z1#SuWp-l_t*G|KByW_ax@l#@7QUMnWQb027InVDefE(9ru=3qUa%%2=nBQfCOQKV$ z@Mj6!>^=d4leBSOya(AiuA_eUhJ`S3E9VoX;k`YJlYn(dKc-i1NQUS+!K;uRcB5%oi?D0B&4X4_yBc1wFsBw zP9~?0M&r-&Q0~XUatQ9>U1PgzsO|ANu(UUgH8B21B{sB@eV+4Z?UfTatKS8UZuJtA zd;jQy$X<-cYh2NmF^%q%q{!mMBaD*33qscC(1a#^zT-8OJ0$WPGZ<0yEjP!TKXt%6 z*A?c4wW8l}8MS#l1G;$U0$o|pX4U55zGr3V@kJMX)az-CO)Qn0vxl*LdzP;CH-aIv zKH``i42GMRLWdz`+COTebd3SmVHS?$ryiYqT9ljbbe5UF)`49(*^pb9lug9mE<|(9 zuWaR1HIRy4124byK;8Z@7*p0u|7-Zbq}jDK*jO!MSGh$}GIjwtD8xfl`8?*O<{6CG zZOuKXkm58P+NgiuJ9d}G7&vo$DSbD)1?H`shD(jU5XEP6NbSFD81{{`Z26&0V`55C z`c4HtdeqNON!JB^rvT#D|EBKAXc57l9{MWuJSKI^p`?~E7C8oz{o^#TJyjOhRK(ze zNC#M-WJQw_iokTTCQQ*SqZ=}3qd=*JcDxQD6aD1jbk7;GM=FKB|IY}5Ttm=gl`&e* zDTmS(VuI}YSInrz9Nq_#Lnb^v-LOMnlWdKi%dvxpXvlnLoY@nPVyV|yp`}vXM2!TP zvDKdy(pm?rZ{MQnDGr!1Rg}A&n20TvCU|pM2+to55?nv6E3jRoCs?tf7@v0mms?Sb zMc)Ge#FDvV8=l~{Ti3}|ts8Xb(;Jo@B2m~QmPOttS~4Ej7P5Ec`l$1u0hE7uNB8o( zhIM;xqgeS%u=$Pjb=V#J`$3OeYs60=nF8+E;ni@g*cmhrYYE=?p2sJL)N!^9wpJ8cNy3|5kHrCBiDTbq;WO`$U@EHO2EfPMUICbv5M6KxxR z4Q%9XdA`CVoW4E~y>4=3kMU(O&GiV}s^qxZNX;$p|##hv}*&X}5kJ6dTYDvWgDG1M&!#N|PGz`q~(9_P2GWcbHs^!yx+w#^;zwL?MRmUo}1 zpIF1$S{ZPqy3?@4P{?w0RyQ-$GXuoBct4?VIi1ue$GOGxyzpRO?o#VAE`Hi#@;D(5 zH9zFxx5`fRS(VDEtQ{sBFV*1N^Y<{a{vT=4av=!@a&Y`v3*++i8Sfblw*0d337oI1 zV3gC#Ks2k1yAka{Te9-d`%60Uc*+nbm&Hw4}wd9GMfx@~5@%_bksa%YU6&FBz z^D@ZU^b1eSD`Z+@3t)8tf#EZ5#CWd^jGGWg=KHOoFWY!t-Hh3SwbnoB^m+2|MNuD* z=ka$G>yxy8wGOu?dOJAHm`ifwg27^$3EQc#gl<(A0eDlzGly4@X`zIg@Li2g%M%GecLBQGhl$@;~c4?!)jbUJPtu~;e5a{L{ z!Sdgq*D7KA^d{o6o7*dB}lgivgrwkHA~Ycaq)WUrFbUlVrr~GHLiMz>8B7 zNO8zQBIZrtO1>%lUJdM=9ceIE%Z#%S zB%LV)5tjoHo<5fIv5p@ByKr0oT09xd&q1Hh zM*CB5sNapL@c#QGJm>p}e3EP9E{hl7kKNh0%={^SJ)4JjG{XX(_ENs*G1GvN_j$s$qXbNuo||?xK#TZ$k?_L1WoXNDa=*#_xnrNZ(v934zA|&1k7sB!Ogy+oQB&9U__^p ziErPN!&lF+eanmpyWNem={ShRwmqoz)fsnvNM^-KVnEhc1?rmq(qlhNxwSe2jFZGP zxcKQSDlSf16quBK5w{8i%=afR$s=iJJT~tj zb+qiFXUq)HBtwL1{@|T68VUUF^#Pr0^OZO%XW@*YDAcVr#@tE`NEEUrZeMv`sKq!a zdO3sY@ecS6?=qM#Yonm(OBG%6&zDTSKSI(Pdzk2%mfXC|vDi`Uh`(nAv3l0=mivx% zkww>4alz{qWPMi$$X%R9)NIbOO##v9n_|r$Nwgu)wTy95Z^Z3+aqPbfqA2)t0BK+p z-kGS)=auE5b5TFL=EU*Juk)L%Y*b!=X_E%M+eUNn{r$9+L)>uSJY5< zWB-H&u}40A;^;6H2Xl9`iFfZqjbxeI-iqMp4Bko}<$(0TK z3{!m{ouR)5#8d0&jwLGiSMxOuj2*`Q=Q>=sUOmbR4bVyj0?rB1bi{onI=iI7)b>Go zUU)L-yE|~hnxfpzJ;UUdTNQKXw-E16>c#b;B{ZPPj^2LmWce;s8;^J@^ZpzMuyzqbfwKctYA|^Evr)2|@S2I?{es z4$jUT!GjK?SY)8f)?SZC`N+9!WS1iZrU55yJO^crg76n3PI{NzB0o2D(a*JC@XoPY z?8=|h>HTqu+_{-!xvvd`YkVp#xG|tgr43JVAy1l^lY;ZC-k(IvH?=@OekCc!LK3f( zM%2%p!l~=SNZQ`BB+*b0n>~}6X<}kr-R>n8^B4D{oBCBEs{Wa2>$ZU%U#D~L14_|w zLK@yHoC(d+nM5Vj2jbdtiN?-C>ThNWjSHTUuFfyy(YjFf;Pk09IJX2&9QpuPMT1~6 zBS|~7j=;$01MncXj>gQM04?puIQ>g8sC%^t$Np5t!}Wk!m)_FEOG8+D$p!DqZo`0F zQ9L|-A+^6*$c~Hq0nHObm@V9Ga&U|~%t$fDiAF^z2wH`1OOt4d>~f<2$q&6RM`OF! zY4j=?3-+N~+1{{Qkh@1+p!WDN{U=sMlRhfK?A!NfO5+{+xv`r&ZD)oXQuDAqd?&Hw z{SG_6XOOsBUsnFnYLciw3yn4&#kDD^mQnRKoXO!>lrSkJXQO>F=5j3F5KSSn8~5WA zabGg9C4;&)MO$W^KZdOuF1T?ie?EUAEa=|uh-=2kamU}WoHOGCK_UD9`AW=p9w-9^NHuW4bV8@3t4O{5A8D_(TgKz zY4MC$sv)KXhuZ!Ug<@k!Z;$1jw}1F-=PJ@_Fdr^y>wsbLGCcdt3-|7n2HD4yINJbJ z3ijcX>=zb4j(Boj3Ly|7D@Trc{l|@~ngiw!?1;Cn3XHG=6o^l*jdEVgQg&ne2aH9dv;OXm2b!k{(ryt>n1qw^bkegS`tNt9Qsn+f(9BCLyVjh-6B20 zR`zB=LgPZ}v5(&a4H{L!Ki`IK ztXYS8jgi=%TuRa}Xf(v9p5${-3EVA*%!bbJD5`BE!u6iIjxJsuBsnw(jPg6!&b%7- z)6XWn_&ODB4O#Lv?HyeCkO7Cc&BohbW@E`!Le>vO!r$&tG+y71Ds7dtSuq-!XgN&j zP9d9Whu{hq#^jggLGR?P*|z$Y&G zWaL0Td^&AUt}*Lr*2)ER$4yV%=kVna*-(eRu<0sKSyOq-I^0R;>QD4FEg`mr-H<7Lk>qW51mXB2sB_Yq_D+?Dv4Q6F!@Ny& z?oCM&=b9^!bN7Plc4mTG2j_4*-$sE}{Ww9&#Aj5T*+;#k&A1!Vp>SKzmTYh=0J4bS&j_L)$pwv^9(l1@p{$-5@f#pWw^K<7kVB5)uBsg!S4N zLM*B$L!pcU=&o?44_h{4UzG@XHDbiAe=ruCuPtQ*8_d|U=sLW4sSw*cg}4g2gV1c{ z%sOWAIrb3|*!STsWb@w*XMU4rEtrg_{#kIE%2s5z%o>_~@g&Yy)5K}kSCQR5fjc&T zC2m<_#SP2!QkCN-F!rS@Cw?FoukF{yqs7+1JaD3mkH28sdrvbGOKkb`XE3|o={dEH zw?J6fNep)m;m@m)T+73?oX3e-+ze3zJb6ADBp=_W?gAFx3>?Gt8}49({z`n}7tc)6 zUkDC&mXSjrLrK7`Ptam>pRyNEp!wA)czM}6{MMI=toLEgiD%1PujIWaU*5n-_*E#r zP)%Y#C&8}6Bg7?PChbeSPKVyj=WJz8GgmIK=ue)n7S6SIcp=>Bb{cu*{ZMt0J%j2>zEj(Z)NW=Oin+eTO@5k?!FL4%Cr&}~ z#2W4`VC(N%=nLAeDd3JxeKh^iCUU#u49(MO2HS~~K}M(uW`x(m=&G~eAifP3%(>Pe zw3d>3PY?XMaVl+{kN~F3SK)^IWmHIi6ptLO##hqssqalMxK`Okmp*z)P1~Q~7v(54 zU*m*cZUGQ}D~;@*Q9%1H1>yKbAJ~hNSt3(jgqdy|P%vkdxLlb8wzv0U>-nX`RL=aF1$;RK3a!2cNXEC21N`^ zDJ1LPtCR4L`DA~M9s0gn$L{)G-7s+93&w|QVa6nuzIVM3E!P$zyDFPx8=IrVsl8Ay zbd)|-io~6hY*9LH6a8@JH5%EJ()-HkaL%9r);zg`_n9xKR>0qz!;|2x8B4FcTTbFn zTEn}`Ghjws0Qj`c!l(BDyVtlue1JS89T5hZ5I5$9{9P>0OQ7t}IK03vVwOrCrWT9B z8`h@kk@w3&Fek#8d+fd#lcraYzjKz5b5d>ip(KxvyxfO}wd$c|a07V0E`a87g~&== zz|6!kTs-Ajzy*rn>2DR@v~66{Dk$CzJ^!Z0TV_N25JwysWQR8nop zj17{sUsi#We6j++o9@HoWB8=K&OFq8`?EoA!adq+bP&I$7a%v`2`#LU&Wh-K4MXj|&&$}(x_ykUi+X^C`K!Wnvdn*wo}jRaOy|hA2flX=zb*WpCM1RyGY$&T}7?LZvNKXwV`}4fU(v`P=nguGi&^=f1z+ z&&S>cY*G$`jcPkhd!vkvV^!hXNC}>KeUJQbw8UdB!Jymzl>O|a&kP#bVUy( zBKV#4)rw(E=Vaok`OnC_;$%4D_X6uSKEctqUy^NRuW6RgMK~SbLhhc}4R$%(xu5&P zQDM|5?xATM=zN+3rro22w{haLydOB-MBps<=#XWP<3(Vd94Lx+^JwHGQ8pQ zY5`Z^nTtxjXShXj98AhS#klY+UBM{>^!=PpBfkr{QIB?^Tyz|5H14Kr9+Y63{Z+xQ zj2WC=6$cgEBl^Vn1h)A|a&~#zSUz(X>b_NOu#r@wt^IP?R*_71Zj7bI{s+-=V?BE! z#!QgAT#}ohC62uxUeH^P&&wX!;>+LG=7ITSKIY6Q7OE%(#Z6!jcP(cZS32AILHBE8uV0hBaaCSk#{Zoub;HwZ{)W zJxb;Bm;WGAgAru3eKo#PtHLRKhi2}SLO6A^nY`cBPrn{dwk;s*S2-=Rj&!!!zGcfkfwU{0-P=LMp+Af z59=k0hiukSh}*;74@o8WMk{k2x?kA}ixikr9)fHgEDRLAB@<-&NhR-N_124FSEo+n z6tm?qecD2F+8GTCFIB=;!(7lUk{}89b`Xi3;~@2EI1GN-0k@xaQ;qjp_`#uvE;7i# z&pQ^Om$eq3m560sw1-LV6BRgnEuXGfnTYY1CxLY&OCz#Bk}>5qWbf}@lB&58LL0`y zXDp(6+crbvs~vP*$Zev7Z_p!-@7OUyM~$t#-O}W8{3wv;U(VDH!f=#O#YpU)u~aWe^`~>X8fEq zp4DJntOwkmZ%X&4|0bPd-r=o+SNOYAiaR7#1Ru>?plR7}cz)t`gF#_DmAiL})O=b* zr2X~bo8*%Q_E|M|;k7(!+?Ycr%r_L0_kmcoryny`ilCniL*^Lea|gPQL;clRy&m6}rLA4{S3Xegh7Oua!z*r{*= z=MK9-(d?OUgV_Uq{Y#;a&;H$*Hyhn{p22^`KS*<`DDO7cqIdk{g>Praz;uUZVx;6l z*lG$!qi3?^&LaFXHVu0U5(vlej+cORlv!QCtZMy4Z;3kMN#6uAX?ZBoaN0~lm;Dfg zZtA1!Zk)h7PbIL+qYUq~Zl#t}w=>D-%gIQK5fG_T7{9bZkm;Qdw~MPdznLlApT(oN zfV?^u*Vog`pjzr+nM^0E14F>Y9XzQOdG z5*Es;l7t6d_#e+5N+5mzm{34R&8 zBk6c0+#esvNO=viP8RC8TYWys=`f}$d_JP0Lx+^L)G%Y-d?J3w6QMQwAG_d53mId< zbH6%GXmFw}++BHuWL>>N-z!hWt5Zf%$$hgRz%T*Qx~xI#TP!9@-Ne-=?Vw3u&0i-L z@VGPsuf55nMSCM)R-GZ9^tnKcxH6C}c4d`&8i@9YA4a|3%elR~$uR#`qTxM#6e+8r z^rIi@8>N9e-!-XXJ&CTy5Lx*noq5E2jpLlRK&>I6=LW`s^sjW{@Igov9Hx`ck4-SP z>?)0)`V6&ag+lb!)gVH758T5>Jf>eyujoDND6Hdjs}6F zy>Md61vn(?#$4%q2MuSIVfVTTxN+PLd=REaM_MCjotHBAH7JQGuiQ-~_P?bymseAT zfiUuq=ZdZjeTZkjI>XiW;1<@gnYE#&IXc24hB}J(Xw|(nGciV9K)%Zr*)KcRB4R38}&K z7#%BY+$VBaVlNYN`5T44r*@`V3l+fo;}-319;|r zdERkmci?WgZSfG7^sUAD$9BR20YAg^{LJ20I|kE^OvFnv>KN@EhCOTf+(DWLeBhs% zCd^TyqWF<_INhShMKmxad^%mbOA;>s_(Zgohlql^0l4-)g>Nq&!tqg=7|i>0^^&b% z#jOS0;&Y$b@d_@~?EYMM^*(|~SPzj!+T%Ib^I{P6-wmpEC6?Ph?g&3W>?SP|0hm)! zO5E#L!ISF?K~nQF@lL8?Ef?tnO&wx>AKGVeAkcsd7DR;37{pZTJzQ9O;b(8dhsJFJ;}E3vI} zpb~%Dh)RePQ7-`qc-&1hrGC-_>Q?l$ff?wN5Gc@lO#+Ub!9{BBaCuM`&c5sw1Y0-b zL!E_C_~;Wo{UM!I^3NhCJfo@oH39AUub0@rmB-m{)1lud@3M#FDU=hx2OFeYiEvXb zMl5=W%@GD@*EyE!aorEuy)(EuemSH&wvpJKZo2%cNCrFotmyaV4C3k0LT3G@+-S7$lf+%)SL4b95x6FT4n)`ZGE7BAe<(tEV;~B8qIGIkZa$+oV&k7eJ;h*Sz zA6fXPYYfsspUACj0rchi(6->;B?@fo^;1Az!Kqawxnr1O!%%r-3pE2jmT4`oIfMo7_=kzz5;R$yO9~@*=PB0 ziGTiaV)1$w-O%%%-uAysuGF0dR&g6S`fnDvj*DWJZR&!g``_vJh7hubhQh&7Uj)n2 z4-*42KxKlSvxiiskZ6&2RJKP4b+=!pX+5u@*6|{w{+&r%hCN7xH~&uFbA$cn+ekH} z?67!^6?(Nqko{JR!N->Op>}P;e)GY@cE$$s^EO6Di}?|W<20|rq)D%Z6+$;e#_pBCl*+C6GIbG*4S5354WHQnziK#o|@cM8BuvxYAp2=r&cV{AG8_VDy zLtV~Z&5aaKc+X^X_tU0X7wc)&B6^ljaWJuXhT=oR0;%Ijp-es}_U zj~ZcPSSz#VtPV`CyM%kZ`MkO10-obyiQOv7(CuA8?X<f3XYr&vJhbakK zjPsWKK&7Q7oK>GboPH6-Mm0HMj7>Xf9d{D;)acX23yP@oyK+q4BFlZ7HWTb52H>z~ zIqZ9Lmb^F3f(0wL3ZBD*EW=(ZMZBj z85{Naxsi`8=NhX8gR)a0>+dpf-gJuZvmD2molfX>RvTAexI?=N_Tpc~c|@{l9u<8x zMi|;N7o|cXc(?dE_H~eev~JxjolC(`~7g@+v_xb zQ8`9z^uSAPsi-}7Ck*mg+&SUz@s92Y(NPqVJqsggwqp^M^h25zw;Znd7*b8~VxrcP z$Pj9XI||gH)H#m*dBOq3m|l`MkcO*%R-&lGC2CzfkIt3U5T@4DvtJBekq=SEm_7Le zkzZH=>jI{sr?fPAw{;U37_7!3BMIi?T02s0W(~u;c#iaxL$GJ8H;%D5fleliQTv?= z*_m;XtlzB%jFv9^K74|A*5qMj+-&^sixlLX-UbU#bYNG6JGRyw!C%izvHrItcD~R9 zmGPz+ZKA^|}Nq5&NWHki|S5e$bNiECmynVbC%pT=gh zxMUT6xL}M`xh%6oR1~DbT<|~3Bx*iR4G-oYB~v{=3cByO!mg3YxU2!BVA5}YzIa~Fan7nf* z99s9B>VL~))BFRt+O~tRk~s?r?{-4GqZHF%ZboK_?}i+&Z)8Z}C|x?xZWVa?AIbAI z!>Mxdc*QS*y%5xe){6tVN2Yi1n%Y!O@8d6gZz@Wmz6IrMWH`rUUAXX93Z6Sj3O%P? zB^{v$usHV)9kF^vs_srD5*A~)mh@P1;`@A5ASPU2UK9w$9hn7_m2ugDIN~9uNxvtC zQ{PAVG@8%3zYN!byrl-X@aI(2VU>lZ#^&Vu*^_kB1rn>s(o!+u`uBT#FXAyOhj;GlP&W;C zH0*jyA1=EAI~^5;P4iFStbxTa+kX}lI21#)^7OIrmyFP1#ZlUGtCsjEWU-@ya%gUl zDYM1RgS40!W6a5DoHaif2UZl}k{SL`JBnv=-M6D})8^s#FI_n0d<3286-X)rt)k< z-kQvf7}R>uOqCtu=#hI_L?tHxekIt$bVfw@?LZfuZl#Y(nlV;8SFC`QKgMyVT-7+4 zlzeidu>u^p*)Va)le6BiiL>5Zg{xv^;alr(#;l+hzxW>q&*4-YJs63)IzDtA(<10C zvVh9j#~XgFK7qNbTe(XiN~kkD!tPeLz*(P5n1aiEkLCI`vgF`dqIo5i5}`EOOj|@N zr3=ZVVGWduyiE0PleT+tkDH@tIU0g;W+HfIO@%Q=8AR-; zJ9m6sGtSYiBF_)2(uq}ryiZOW?RW;|?J8F?`pH5rL~pe4W#LWwE7BOWA3UdHGu=Ua zl{>7zmP{{L>l3?szNodcne>Gx5x4v@>mit*)=xqaX>C#v>#~cG+&f^x| z{DqsnK5~ZrIy@)+JSm-&%ihhL4d=dHB`Yewvij>@U_$B~?3IcpZvtA$^Q0Asz8`4V z8flz%E0wVqj^iSXbhv{#2I!TRkH_qO!g`0JRHTz<>GyeqvVlCb{1*sE__<%pK)>L* zvK$=HN+9*;Oz3r19bZ0_VA@CKkas)Zq4WBySZSsUPpz+kdGBOKJas27u-D>_b7MG5 zzbK4fu@)zlN^$maHPmHZDm>1T7U=NY^Tvue@Hn>}JHARnN_4j%|H?HsVc%EsO-5Yk zpt}*C92pN*xyewIo@Nn-%sQQ>j zE}!~Dd~8ebtot7tDzlz8*BusgPvM^#YlU3CD%uD+g1WbgFF-N_z9Swl1Uol53#>3G>OcbA$Eu7T5@PoC2eQIVZ5szE_RW{ z&Pi!xwRJq2xyaH&Ls!l=Jr&J&d$WI%Gfo#$OLfMoDGcf#ZWk(MMS-IC2O=GNl0Tzy_|f(^It`vBw{C62(Hcu> z@nI$Ke+2OQsT5Qk)aUb1e0$`eINaD5i`H`TFx=_F=01%_@tP3YGf;z>-zSr6yw~=- zQXgABKZ|IoJtKlKowS+vqZ_!2gJlRKJYTEB?5m20_U>~;!(=-C{FO(KT~Vac7CE%d z{TLnovbS}RlL;0aPbI)=;*84F)~Cke-FPQqC0J@l5# zL1^sz3|*a0(C%t0RLrxn{H`$*jQ(6Cbw5S8!>wr;Vz0vczl>-%IWpKqFb)r#8msw^ikY@cz?n^yt`mK4x97N_E&3ZpVBjwcv*qpTy0sq z@ylr)j)7B&TEYrhZ4&Tv3O9RED`qdvCQsI-V_{@9-97Cn?GtY&C7B*rk}U%gfsv#= zOkLRK!S5-@4U*ME_4Kmcc_NZ91GZ@n(c~-3a9uzz{dNB$R;}7DkXt_%?cTnk{hNyz zkBOzM*fBZbr4w7he47PbDW8CMFU5jaoDL0jQWaXeXpn1u1BBXJaUpX5u&(GfGi7fX zM3*ZHO?>#?PgXQ0I-I6V-b9cW7w2w0?Lr$pXKdGUh6Br$ap_J|v~Qcl?QGA*v(CG5 zN|h@&HD8A=RdV9@eW%Gaqlu)du2k@Jh-dOBmD9;dCt#8MdF-5d9Gow@5gW}c>UnZ1 z#KSbA@|ECTb4ASl9SZ+F`bR}evPq>+6SHsj672Q)kNLf!3J=uR<0kuB8ZY^|!6B`U zRx1sle#0=a9?U1(!&c($MiUy6L#Rc;Fo}0^r{yCRfs|50;pV(jG9c%m!FIxRDZqLZYstd2Fw9G=7H)w?YmbY-~zh`v1 zhZF{nImSd4-vIP9LbVo=hMEacXyLtrH0rCtk;w^cfexj zw@~O(hGUD5g4^;Y@;y768?!hY-zNIvId*`~oD&B3-&8=il@^gHhytIro#a?x74CXz zLGH?IV$*ofZv>y`8M-=^-p=kpdl5fyT4e!xFXe>QtOPlIRSEb#Ao$zugnQR(!Q${) z%quei^@kHMOT`;5mI?6D`e9OWA&CQ}8~hxo16_-Q$({ohICO3mKoTV_4hzkbTGR6&T0GoXV3m z7_)R5X1*B4O{%w;1%4jPQ1ES(6wfAR(V{SyM1hX>KB}xb2KI_MVsUUjJ$k5!vO0x0 zWoVGj`#Fk+H;8c4bBw_DdL;?kBEzZWuY;n0)ubow2WRmKz+O!0@|718hg=%tm$xl*tWdL`rS&#ojQ>c+f1u5=aOtd@|xJ9#6iSDSe zjGoIBCNU}noZgfW)$1y7fy+j>$#JA@>q(j#FUEBhMAG6fPUKjZ8Tw}&qAKQr%!i4J zcpxZ(NGH$09=SvCZ|D_y+c#F&Y#`2^aOC~SIrB+Mg%Qk~;6fu7nv<@F$51qg@Avg* z()O$n?qb{l*d`SXZ=T#BS3gTY;>$QXk}iM|-LvGbH4B4A73BF)FthzTN7T&cP*s=p zD9w(7#qAn&!>H{wqit%(w;ou0bZcjQle`iBnZ4!JW=Q z(pEW97?fLz=bCHiq11954ZPdlTN=&fBsoTACikxI0-gLtoZgQSM_ZE(0_O{Q(D^c# z%s4ti_gAhYHSHp#pw)->ja&tb=jZ6`OFx(bX(^t^9?hg)N}w%1gw9jSgn>*s{yZvz z3F8XzS8*R~QNM_(bE+ZmyE(kpbw}cJ7kv{$$n|HuPwdZLNL+5+utd$3ls?V_%jp%+ z)v`eNz|sJ|kCA3NG_~k;rST;yI$z*!$ezSz*HGGO^?)^LQ6Z|0}Jwq)4Zya--I-|dj8OgVzvNJOCMxhXY1!yk^g5wT{NH|oURtcceejr%5q9$I z;)6OQN;(SyCWiv^Z98PH(Br7R5stn862nBYq2Rv+wrj#?c27zu>@n^qeqUtC+P}{5 z%;zZZidDL-Lxa85IR_gKSaXr&3ODvb5$4a$=4k0K>$#8TM@c+nJUup1lcf#hgue!) zeY?!=*zlAZw20HA&yi zvBK*Y$KylMvD}cxG6}82~R6%(2YcHvhIJBJ3b4? z?9+qwgM6Nz&oD*jS#d=Xg}6~^J$^VGz%Cump`PvORA@C1wX;NVjrBhwv#kMNEb_1# zJ>o!HO`_mXusiP^y@&(Wl3=*~IICoyM*Ehi6{#p7(PfFg$3OZr1UJgrTQIXT}Z)UYY5hFf9KGxqV_x|LJ|IKSq599jzbck zv26VthXszNoSEcOQt#MGJ%3$hYjqTed6W}`zTohW^*yq+b33m66N4K=oyn21Vt63= zojjM-1m71W%nhw19BV29!t>#%J>wr}8D!GBIz!OX^CcS;wBVj$Cc7tWBeV5cHtDec zhM)Flaa+0r=n@A>sMkG*TkZ0hn^P(pjylR=a>{5J`}iQ%d=bK|HVrP+jh`d0naS;M zImuij!Kl3J1hLDV2unX?k%C%9((YlzD7?5sI`6m8`YDzq*ejhWHM8gL1dPX$ebQvk z$~5#I5T(u&`ml4TyTQv|QMj=5D*U+=O5gN0(s{empg7qKn|NQq_vv-`VYUNCE*Z}~ zzQdq+r5&!mFT!WjR%4#{O6=_x;M~BIWWRSi*$@>4zA+u-Um*xTwPet_`lGQ!&lb*n z5~5vs6{EjO8dL7JUlF7ir4yPpEWKiknJ+YFq7e~`mn?`hHTGuWQqK&xH4=-KA`*uK&l&h8F} zEt|fB)^B%Q;u(tOgZ+4{V=8GP4E3$ozMVu)Yp3oUNXKmrZVtq*lUL0S>UX2+e?2LQ_YZK4Ih6iUbiteN~(s#)AE!Nb)?l{kxtD1WMR;_Q5Y%O zh*_fb7+2T_jT)2HTKpMWZe@X!;5Lyv!*jUJ zdj-nlwJ|ez7R(B`h8n|j$wX~_FFAV<#nmrToOO-3%(zPnJr$vPR4i-K*M<+IRk5$a z97AVs6Sk)+2^$hd3kPin;J2GBL}k^%=%#TfB^660mI|>lEe?ZpT3Dt<4Cfv{PcA!~ zg4>@`-1O-fjdy5d10AlCTQ=#$D`7qjdYw;4Qa6L`c- z%46)OF2aDfkFj)*6L{+`#dsUQ5Z??M_~AI+?vN;0cFzVHWn&eAVkpqEAkEh_-pLq zr{~07dJ>f15#SQdJo>3A3WG}&fNB3gn$Jtnxjs?!OkWC&<@@K)-g_`p{3mg9r^`aA zxC6xQ(}F|nC&or?C8+I{yl=3|_r5-D0)qsXR=Y#2aw0{B;^*Do(Lcj~Ir$ z#{P&;7%?KkP5ZPNqqk`=r?sP?`i~XaJG2Yto4An>*{O^}h7F(pNTSo^hsbS40Y~;b zaT;Js&eZVtynO7ZP?g{Te8s222k^gq@g7F8% z+3T|(lm9ME#C5ZG(+RF-xUz|#F^)WBjk`Vx_O`vjtHJj$ZRTW*xO)lWenrAd8*QQT z&mJ=0To#r%1%mvkV=(MX==iUW81-8dUae3BOZFmc|9S;MVFg&$U7}UWSLq4ccwD^W z4R*fV$LYR(Nq<~8{WZxXRWF8gf!VMe++9Y?}t@+m3 zQfk@tm~QGhOp>2p#y`WgXfW0iIj<(l$UMdU5m%vOT{~!(jZkmJCMr7qJ07pQgU5XY z=q$YzHpMklEg?%a!iUjk?_BcH$QXR$O=$kcFmCJSW%%%SK7DKQldL1NAn#Za%n4XX ze|PcD@XI&h&Bi?PN_{%4_%K6w-DeH;(vgDsm5R6{KMp1ZiJ@Hk4cu5-k6V@qaogdm zg0U5%q-n_r0lqJO`olObc#<@UFj_%t`|pw`+bP+#qM6BA>ciE6=AqG#ehqkr%bk=UI3@RkUqPGmtAH7Q{Z9NPZ#_Et7(JXQ<;sEpU-x|F6OF#<#PUAjJwd48}vdQq` z*X%-}IA*uTvQw%)FeEYs7O99ZK5a9o%W}eqp51{4UL)v``4WOI>A}sMSgaoB53Qqj zxFAL|_}=2ehzrWlI!~vZt(u zz`G=m#8^*FZW@v=un@DmuHeNxKG5pV){ZZz$4J0g%WnHtCg)JvvlC(uLI8B{GqTyzQ zom-Bx^KHw3xf2b&+!=5$X2{fU50M>x9wfh+!Q|p5!KC;B{!HbFqgCIM)hqAOQ%xO^ zwksL>9?s$JY*|8b_LWd>#!5KWwvcn>{eJfSJa5fkpBuBtf_w=dhnHoaQ1#+O+>&HX z!!I168hR!eGlO^9-LD}I4oY}gr=6bM`U|zHUQ!R0JhHGU9<=wDvI9vs1d$cV?6H%C z)g0BxSX%fnyIo7jBcnBtIL!hXv8&WTVKwU8FF^DEis=T2BBJ$s9*qf<#gCfK*meIk zL@l($rq?UbxAg%JHQNdbGn|CCBAPKvx$=2x|JWP+Q>?*KV1)>Ehdn$7tGO0RENxceH3iNRfPj31gdXG zzRXB2-Zo>b6gw|K+m)y_`3gBbK8N|7l8j!H++o|dee9V-O3>kQ6k^@v$oD5- zNYJH3Sg>R*XiXj@^EUP4jG5+`H#vX}__L3Uug(F>Ygd4O;X)}l+w7xds(8Dx*%l9tK2DgCx8C5FD#1qLG zk>u3t4Dwhci+S$*4AX6+@%or4GUXQU7U$?*e<#^#|+Y~ zD#^b!JimM6Be;F8tfA?FC%v1xk0@W(<=5_WYh$G6H>!PA7tdubU=E8)qr};CRH<1gg}>_pUA?R}&1%=eogQ-Z+pGN`uzHGsJkw8fa*UfGHcEk=p(gSia5; zB0tJ=s-_#bYToVGykj8M>~O?=<*`__G7Hy6i@CZ35IUUx&@5MFFuH?wa zJFsPs4Q*4+p^~Nx@pm(l=+PE}HMSzck9vBrrlp@z-gBL_*RF!rZF*!9Q@t)l2pp^d(am*si$irNUx5n?+cUX^4;w@_m@Rv_Rm#JqJ;w_H%W6UpKA&8 z_W`;@hI2Waj&LXKCZKJ=a&+D%&CNM$Pq$xeWh8tH=)i_782>MVb>Du7?s-^ER5bYW zQ~xe3vylg@>zO2A)i^w`bt?opzoL7dQzaZ-r;%V(i9Vi{!O8+veu(Mz(d^b}kYvz{Y zeih!8F`<*Y(O77;?Ev-gSh!|y zcb2iF?=GL=blHLN>%U{{`i0zOG2MoT8z!^$B$yp=zD=!Oh-2|Z24vMd(2a_Om`>f98bfpTEcbj^{~zGJzbO}O{=$#;;s@kNEvn`?O$Tq ztfvv!Qz?n&`dXxgpR>trWyyj~sW8uOBYe?~ra!m%qil{o_vdm2T|X)fi>;0^{e5|4 zsb>rfUcE^tMGsI*r40L#imkL8UOTO3U7?eU>r{}UZfeBC} z_k+2tqKSGVfv{%GG<;LQd#_aESgF=B($mw*zLB}ZimNVViegqm1DVfwMz&#(zX{~N zVX)dn3fBJ0gWNL*AiOdexJp%_ez_Ev(f^a@yS^vChX(PE%nzDm??KIqUD&|@SIj)I z5qAI66}-AQm3)Z}ghNt8U~g$o7$o4&FsF!w$O%@2?6823F)yV^32wGkJoy7Q5n^5KZ9PMB#>f9W#<- z3Ac($XlmIQgn#y!b*~TJ?k^<2<|pF$6yg=Om8%|*VH zUGa{|$vFtj=0L*iE`^eS8Qcnu37p0aV@`c~0=6&8Bw*u**K1v9;gUe{x&9#uWe2PGOZWhm--beSy zRKX@+GuWhl+(YKeuSLGBW zw8rsFr@4^iCo5cImJYu9;mo~?O^|gm4Kq&IlblOV)chucu?H`qm;{FcH8oDPMV~2_ zl0^%Ty&zTjjt2f$h_T5rs8ewt$3OaxP0zvw209p!tzZ(M*|Ylw|GMbt)12mGJOaY@6fuz5msL;qhldMx%e zTJJ0*8FqR=Py&xCe}<1|He%oXwe)a-5BT-IJi-X6V)u%Ow_FN0Z+&T*Ia0h8+v1(j7l7@Z2;n_$o8RlzJS7Nos~5A!APV z-sz{6V$D>gHJ#4M$z~^a*Ps+t#@hOS^kVE8a&^o<>g!a?mM)otNyg{U+rOO<+i)nz z{s^TLd&o)~0&cwb#C+d$THn?Hxfhqf%hW;8$r_DvvQ{MGYd(2z=*W2ojpBZ8SLWhQ zn81$Gcrt^Zw*?MKaT9l>q5QKET;rY1RhW8nqn)3kN|!6TTuNqm%pO=JXW}1=BwW(1 z!;Qa;WKOV%ATl)=uPpH7)>}z)RkA|PIByy%*PCHn*#Xv8CJqPMC(xp%&*1AF0Nqo^ zaCbawxL*UG=rxN_EdHm9S4*@wV}lW_7qQ^<`V;}1ZesLcIF2sV#{Jhsi20@e89MKH zuHGn)S0a0p$V$V=sK~m{`6eWVU!tLC$SShh+g>Gv20~USq>^=?bE#+|DO%c7ik4E+ z@BYv0#rJ++&wb80pYwj(;+ri~NY))o`eD8pezqutK|4hpTnV63KN)r|877Z?vY_~j{ziMN=?;&NbAoy|Iz zCbC&el(^ZsinLs}iu05kgL6b2h(VtxhWjsq4+l<>>rQ9!^4ERftr$$cs7@tH2Cg`M98a71MXp5x*{-(V!7Oft&hPpVJ+Q;nte+ zE|X0Yur9$KH>tGHsL@y8o0Ue+NJYS;%TX9HYRIYhEFo7*f07eFjc|_PG0dCrimH<| z>Nc1RyMDf+gANrazT25&r&w~~A9cB?_kU2Ps)ZzM@P!x^QNiP=A7q=Nq2T0Iq=&(X zr2M>u1Lv}F*1G%~`f+d5mFv2GRzBh{rK;;(C|>qVE|b&xeDo1AmL!8W{| zhCR~zNxVD1pA$M3CTxqtdv^*OH$|tRNst{@c!lF~4Q(vhmXDuzO@s~cq9pT7Ju|(* zoAG>}K{{pjF)5ODpk=iHx46lmo?Ze~O1i}S`{xBK4*!Dyp&jsU(lKkjP4Zwk#siyO zH&CA!(=a^hSflI8CK`L$7Vm&2_uohvSrYC~#wY&5JedKMt>0#SWq&YC?5ZS5r!{fC zqch#9$ywj;dr8a(LqXEkfle%v=USg#g=o`wNMEZDzQ#W=C}1p=e;_Jv0%0FJCV9>&Rs0CK<%;NRw-?n%#8)b zv}UV0mSwJheYGYq4ZY!k$pk!+{F~Tq86szYW{@||#?;lM1>OBcI9dNF_7ry?7pDB9 zQ*A|<&rv3DXoC>f{+}~<+A0nIE7!vm`5v4t^%`%t-ordYIb4vNhMrSiq0&k_R#w*n zjE$wyart<3eH9M|^bg7TwU9)rTa))Gsd!uS04>zGjuLiOAU}2+&bQ2Go0XKoE+By^ zRf%AqBvoP5!X7eZTPO@2cY&kHQ^>r-Hq61?Kcu%xiE-K0gx`9bNKB3$YK7gS#wPDj zVU_~M{*mDQxSP54u6o?f&24y$QNhQBI`H7?YOKDW1#=IW;`nhk7%v^hMIF6>C9TTb zqnA9xPUktvl9~-m)16Smz8Pj1yFu$GUAU8(P1LGOKp?XPtKWpu*p>nu7aj|HqQznH zo@kK1R!1*F8LgUeo7msF4U#9v)0ronAWB{Wq;)&VDn4Jj$6JiF4$7jgtU7HU(?t>+ zo-*>^R4`F07EM)3>E|~wtgyEghH9~VhA0kN4~2o}JvZnbGKGS}$@p#dPdb0{TKx00 zkWAWN#meiXP*;AZtSRXs*l!ahOEaW!{k>qCR-3`vbqvAu6}yO(Xgj&SaSE1nj$!o| zrBf%DG)&OE)yOGGq2B>3ycwVXiLY+JLiw$*@@@msxK;scPM6W{OP^@+GhuG>PbUbP zdW}f;ghTv4b5wgALytbv!{zz>-@SMRz=SrM(_Krn-ts)PkGrYO?^E30g$z1s#6mz^ z9h%+?C7!+sP%W~S_DJ{8yUUm2?18CVq3jTy_*xDPJ!cDqYo0T0N^=BSJH!doxfq=N zd4Aa6TSUxO7L~LUdG@G|AoQ^j9I5sLf7h2d?vFCMq?mEiUDHu?XB?B}6o7rQws>*1 zA`S(vXdFTg`&{VX3Q`<~3)G|2h z(M-&wwd`kR4A+wvfI^`;IMnkOe;q*jd03Qlyst>b+%;jOO`OkW^7n?gvE;>1D~2f< zw4UR&8Mi-56$ei?pU23`$+|CQb?dc_OqyH^&|Hp#v9v5cuJ_lIq ze4J;O=FlY7t=s|QAj*1#krVuPEa9CCp8CSG1N+Y4aCJOZs2ae?yc^cT^Z#M5g)}{y zD1>tId~UHj9gELs+YnYv()MRb=`F7Ko&F!ouZpsJ6YKo zd7yl-7KE&q(^XFvLjC2d*4DAdaAWgVX7H*pJv)8|d7XWg+UQzAlFDS>X}J``>(ApB z9Y?&y=j)}VM8%g)Nds?c2^e9

m!QiWhv@JTv$M45n^ z6f>sS0F62W>2aozC|9VUaGM#_TV@kH6A5N>t%$DZR&Z^J!MCqsVWsR15__f$#BS_H zCC_oB*-Vj2sPI0xr|HytX%%0WW#Dt22pXk*5$pEe#{(J<@#jHl?r2#IRPgm>*ThVC zCuG5;PuhUDWrDcn+p^ekbJ}Te{4$UTyoQ=hd{%35J_!#igJVu#$ci;**ljnaamn*i z@CA6#-k(jB+0{n&?zu--Z8L$L7HfGm{2Zzh>dyV}-QD1q+k>w+9<{cSo(Cn z6Fp1wnBm(js`B?cztIJ7r*I9v%Js&Okx-_~TAJ&47YCifVoYD>Z?eoS3SNHV-G5I+ zprXwO?vy;D1B>I}Q-UZQ*NtXFJMzg@^{Zr`a6hYe*N24gv)H&vw)Az}aXKbUpNse| zXPuYugE4qofeHnQq~^|hq8;J}GxpDf)=4r@YQ720=Kn#ZN)gW7YKXa+kk8K_;_-5B zKO~|?U ziYNAyEhb_Dzc~?PSiYT>E}l;e`X3PkMFW(dYKNCkE92_kR66ta9s2CE5`L1+r@Om! zxV{8gl;0*py3T%~YL$1W-@0bT-NOOSe^y{B)eMRD{6cb4Op^DfjfK3*92z99fn&t; zaFtCpy7%gm=|>0|*s~JmxJOXW8abvB%5^K*J@kSY6!~!JVL3SMZZ)^O@GkDxKZSS0 zZ{VM7O|H;qE()A8tXC}W!=vDd<-dc#@ZTYH6WfWB28k%18H1UfO~fYU0IB{Y%iS@_ z#}PP#svo=Q1ko*Q$fYg(j>It4%HKyH2dHq$mNxWdPClKxlEaM(MKt<-8Z)5Hdlv0a zQ`vz`#$l`@)oQHZHb4Hv=k~hkh~!viVaIeChWP;1m)HvFSoR&UOEYN@_n)$v6l9POpQ&V^*Zb@Blkkq@OxAZHJ=L8&vD} zG5jRP(C}T;U`18}6R>O#IwyWXzo0qR`v!CbFWOJRn-`g2G+h*;_?;5d?LU}AwJ|)4 zOcxee#6a-DlcaX#Lc&h(f}PglG)JU{9RKu;*bgmc=1JF*wQp8~cc&?oKk5dT1s-@I zXPCV|ZiLR+w*WSKO@LGKnQW!VO}ga!X>#C*37sLohAWeP#$=EW_@UGnc1mocuaxA$ zO*TN#S78ADYn%kXHXO&s`}=6*(u>$IG>P|YR1o!mBN#IF2YTK;MD9m4)3TNKahLlk zw!fwf9WyLxn@f>p;EYAEMMsBtsQVN5rF>m{xSIIXYqBT&pJ2xT!%4mu;Y=eF;nI~c zg84g=;hbg$C=`~_k-H0M;I8d>@XupV`fX3L59UKcVF%rEx>opfcgF4}}lE#LLBpHlki6QSv>7hrd_TOy&zJXdLp-0P=uQ*e-+!N8^1Ow5>V_EM zXM~M0yJ?2+5|kA!K&M7EeA%Q=#_E3~o3A{B&%RrT)66o^@c4(`8pVv*v|re|Y*FJo zs~Wn0={I~7bp>v8SHh;lX_z{r4GzB~Bup=#ZWoloyLbI$>BBpuu;e`yR4`oP!MR*@ zqa*jKc^u680^~d2>pJv(yP-WaI1?b=*>S)ec5@;kEk|mt~Uo!t3*b1 zb{6UXoQlK#Ei`P-H|s-Xv$1k ztna4IV@~{iMs2noqZ6_R@#SCudiPzzP1kB+Vp$^Fn3{|#4*X}fHINnJ?_-L)MDcO^ z91I>-z+f){eb2uSbhl{2QmADn-O3<_|9l{*xtyjyub|fsp1{_pkJz)1j?l`VMzqLz zE*@B4N>?k}(%O#8fbYW~+Qth-=xe+;-HiSmR}J&-90%2cZXotk%5_)7CwAP7Uac3OZ`gv!qx--Q(T3ErVQ8JV8hjknsUCf2XU#f3Dpd{!an;K z&*Uy=c(`vKw_%ePC*?Jho4s?GT>gqgtKkfUTy%$L7YFDY!G3m)>MYJDatdU(e1<6{ zlkt?FGcF9zB=q8WT;rkJ@Vit41eP11QBXsL*BNuZ!FEKUa|Ol;UBLpZ0=K`X$hfZu zXvOkEqNt|;#YxIsTGR=eyKfpse00YXBmA7<#Aza_UPw~+eIzf4JzUJ4Megw#b(4m@ zbORg5zI(KQbI}jPS;I~s7jlBm9y-USM%P2b2{*F-c?0z-I>Ni(W@E=JB_`}cD~Wj1 zkLUEJsov<^#iQ;H2pdjZfkDNE8Klt@MAn1S}mhvdgWY0QmkMcEAz)NykoeZ%&W z+Nlj>lGaJ)6E8ZI)p&wDh7W^nSuFT2&okb9;3{>3R=!U*`))sxx0b%p}hK{(j?0KH=7bMhsUn6vXDXpqT*DB~LPuRMhozSo33yVHn`ttXRdDbL2q zG}5!-KhX7_K6g%G80T$~;S4&AAnl3~o{T8MXK&Pz3FVy=6ZyWp$V6=VYJ$|X2>w2b zqxII=)=d|lpptwu`P#$Z0kZq~b0Hd2>erL!|9;}-Dbn!GWDK*#Hjn018#g+hh=Q`^ z8(A%Wk5OSp4rlwIfQeI0XBua9ldz#iu-~BuD|hbYrvBLkI5mg}ImLk~BShlNW}``U z3m*Pp&+XVf7KJj_absTOP>%(*DqJ|){K;s@cZGDHV*9wESrgflyE(}E;x}vgt zI&GeK8K3tn;kf@s@WVP4eDh>CocqfAcawst>FxzMaiTb;zpg-aHDkDPb|$(7n$u77 z-qDRbyX0w^lR#Z|jNs)Pb3xx~X$)vf5Uz};IVm_{ z&)>Db7O*?2Qqfdw4t9B!l5x{3$>olhgxSXu5AAS_)icE6LqFLF+iZHqlF!tMH(|#% z5ANsVFJz*RF}8{73T!vegI|#xJEPJblfFKt37f-c)Wy5de%1qK27e(DC!GZ%ih0y3 zVlsw)ccY^&ro?yZ650{E7~<|+#tU|lRAlFATD!FXk~EH!SGyY;CuJDIg`w@#qdgw| zePl5~v4Z~N+^FdOO}ORIT57N1gbRL`!XK?wpnRhMp5NL`ZUwF-yTce1Di0*nR@z}k zoB<7J;j>cDw6X?Q8!2rATSV9ciLIQL&Z zt+Y*M#=a9FmQgD}u`e6EZZyJX{}}q_zbx1vdXI>-D533%AIz+2LC}{e1RvH+L|L5& zG${Bb32oGe+Cew)@P33@@!F8OzmvIVEKlw{QDkyH8Bx2%6Hxn}H?(Dquqxj>cpi)o zNp*e-yOSn?ykiBvNPUSfZ{DW^^T)7<6cb^*e;eyucMc0LNHHm`y?AAy9DkhJjhB_9 zA>8yy!>V1kxjVuW$a(l+op3w_1KC?7{y+rlGxjJew09!Ce)2WbkfX$9J;@}sXvkDM zYvPV6Y4n@(XYzu7cKgDwp>SdethW^iO3uB30gDoLQM5Ob?zyC1SPzPf&w%E{cT~yz zA+6OaMV+!OP&iQn>snm7zc)-Vv?GMsBGv?9k^&m|T@7U&B)M6q?o<0R2ez$K2WDiE}H9aDwO z@S?XNx9)E~JEv?qxScdd{pAfrfz!j;53iwGkT>ej^o6OqKQP_j1b1wI(-6AY1L%6n zXvR;6!UsmAbx9++Jizk{20oK_5BU8lA4Bl%)4>N7$Ixu=cjj^a5FKesrws;jFyGJ` zzE<$m>7uRV;KvD=RP9M8h}=avl_U&~n+2M6b71Giolr9*1V0@eplhW8qZJ*@PQ8@D z%G|p{=KF7MWG>pl1^=};@6dGa=;c8yZ8ZaXg-H5-P>%EG_n2D-9dNR)6~0O?WAecf zSC_sceIl~BqhA*OsOO^Iv0X$i^DFtD5eB6LL+nAq=kR`p!ff+u)>_#H#AVLo?mQDJ zcI6aFs*9tAyid^B=oDBfN8|iS=P=^vHq2oyAwaAQZkai9AG?gW_D|-_2K&h<=~)Er zp^Adt4;H~e*O`!RvKzZ6e4sfi&hckL9;8}HgWZuH^q$JFf6hk`Kb@DjX!JGpWD7|` zk_(jZGX`(*qZp(89!E6FA#Z^Oly@z_c&QIe#e`3np6;H>t;sqJyyy`F;-{cSP#Mn_ z{Q?uq50UAeVt8}n6Dk$=l-hn-jx~HEz%x0Y&e~u^Bgk!I3nuHt_DN zm>_*~CG0Xff*Txna^15j#iM6Q`Fv6C!|hObyGIeb`o7|fvG-XW%ZF6zpB7ouw;xuA z$1vB?N8r2G6uLSRtb_6?shUE#156PYFk7T4P2)A$?3IJFPu9_2|Alhv9t3c)%mDV5 zrBTle%VI(3Nd=arhN7O(EUY=9kGIs;aeuC~V$^MZ@6*(ZKG#@a-THEnBJa|8=J=hH zxaW!&9=xU+Q%yOUy|UcnN=4i?@*KVRY|N2d3p6>XOLf+5hlCpvSha04|Fmmha{fuE zoW=;M-W}ma97MRh9Xi~MvNCMm-$lN3T*iBA3Q7IrN6h2ALvZ^3f2680o^JT+gC|q} z;*Xo&-0({^ff!#K%-o5rcHbCI#Tl_-^$J?+ABCx*itr^xf~pLy#@&r#@U~wU<-84H zn^YGDVm7{KoM_ArXE zU2zaIz0Q(B<|C>5Ery*>t1k1>D(o3?)@!Aik|0eoCGp!8c=gU(OIY zD6c~8Kd(gr;{$`o9dMn&R~izVh#7NjQ1t9^j9Ns=v#0OKj+dEWqA~{7Txr2^7t+bD zBQcGW9MIMgp+Mz=rlzXyBTZljOxwPdF00hBTReVpM(q!e7>%rW`v3PLmQLB=#gd zG5QV`tR(cma4`%t`Gx5xMKOMCK21$tLjUHPVMv!RH7dN1QM(KAqe?Hm`pA!4y{-o> zt`Ct?ItJ>hj?(qt-cqkkzF7P5pEdj|r9HJl#C=RBhHg1Wf2jzQ&(#s|;aDtF(K3;1 zrwOEPeLHLLH;?X(Rz{Y)%S<~R!(4gXg7wnMR6Rctie21b63>$A^dDgFfWl@Nn#gBjlXv3INDI2AlYjs1 zj>SiNvbpchQdsfVl9PDdLkm^6!cq1JUT@7K_kPquRmpBn)4B{FsUKqM_`Jb`w?f>{ zYE`Z*K@XR0%%;oO+4!gI6K#S-%ZB{*RBL-S4rn~5g64zh@be*VK9moBnH|K==r*3q zILq&qW>VS4>sXn7&{8GIm6WDF;#OB%z_kGv>aX3$=St>KQI{~LL}?yqWS?UKQfjHi z%ezF(F%xyStwYDSK+f*sW%}K>mpm+U6)Rgrmo8IT(zOfdZUBP{tMTfBx$T@ zwftPYgsXd8#x;M4#@eX;7+$s)<`&e@COtha$A1AdR~eHmmnho8XHo?3XVZ9(Ll||( z1Z8#2=&_JPOnL4znl=9xJgIBN^mYFY|ze%cLpC!D0!E6Uj< z={1ak%w=43x0t&+shYE^>Bamt=5&AZ7$(q736#dm!oxr6+{L_CR6Mba&o1b2krMKp zuli4X{70Po>mSJx^*C<2sw=$7y}(R!dBeJIe1LDJPr~ztVIVfE3m5Qv8~(AH5HEZM zvMu$PqSpuTP?{^0+etUJYW~I#{W5}rIkPzNU6)D4g8BGVvzl_Nmea#LNAex-Syb0P zORMQ~_&I3_EH=-F*Z^lPKH?<}XvhFHm9gXUch8x%W8BJOwptmqWO-r_O=nR#zs}F_L{WJ{FxKgPG0CqhRvcbn?S`0riwc_Wr9y9IMZB z_3Hgleby~HULuB(DZb8Vjs1WVtRJw(3F+*hdkeI0$i~fQ7ISYd3Nr;(H6(aRJdyEPH=$TmT{L9fdEz7)mJZ^A3r=#Y1ApT1GzIG`Zyu0F zc_jXpMdVxB50da42F8bJFEH?7dbD9i}r;M8=Qz z$t8oUs3N^Rn&IfqU2v`{hRoZ2iMzWp4BN8>^hoG(W9pQ%T_q3&@ycjCK84bp4NCRI8>H4+s2Zj#?HFb(NElDm9MmOV?wf z!?$q1QJ2QM=R=dk7AR7Sr)N~oGImS(fSU6IY-jnqcKb#yy|{o4ckX3FehJeJIeH{; zw2q`ay~@m9wjPiIpOu1}SKTG2VuI`hj zhZf7g#`!G}V1I-#-LKelKQ^#Q6$V(__K0lV6bXllRk$OO&rt9;6IIVBalei;bkbiF zyf;aiGkR(Vj^i~1|H%hH?W-D|QKt=BXPm)Wxr=@D?Ho~wIL>D=YG9Q)l4oagaIu{_ zgkNofSO3jI<>s4sK;Hm9>^8^Bo7q&Me>wSttDyZPl8HAI7^R1nbY2m~o$;&So6m9V zzgkT%4^Jn)s=$2;RHMy-X83eOk<17-hUJeMsI7Mnep72=mkXT9!LU!bzgU^>QO$&0 zhdX4J2k&KzeTH%T>||S14rs^U=iiY&thl@g-pMP+IW`XD1fvBb_BY{AtT&?eNg{9J zg+9r}K4_pv)&lM zk1dO!SDUZz)#7kl&}WQyTTKR1r380JG~jH@BVwUjMO>x!z*`#wT)o?x_dWFR^>+_- zxb_rsCf_HX?~W58BR5cXUqJDy5_EP-L5$u8kVrZS@5lb82_nwa@{%F%Tfd5@zm4#H zAPQ36!4MgH0owJ>(-IeLTsQcQIUiDs3sa}k+oPH^IMtTi6;_39GoCT8a2IxzHnDb! zN2qb?0Jk&T2NQq&!c)bG{2q4%om4tNwrVsGy|#T+IPMU(1@Ge)?h3-`1IgCy6<^r! z{=?{53<5_%cfDk-Fh*qa{Dj9AX#GqWZ}|1IXQ3P;9Q9#rxHL2G;BM&oWeDn%u9Mt_ z)tFmylRWhK50<2Kg5(Kz@+bKq-t9=hmoq|{%oX#A0!iVH4A^7G+BTjubq=TBJe^Gh(msTr5YeEBBDNCT%XKiL#4@FB2>0X>itR0$!{w!GWng z*c{$OQ)FJyPjxj|ptB5z+Co6&y%LdYECF*k2Zb_s;pzET%yWt55IuPc*YP%ni*;Is z5o4ABV`_zcLkzjz+QCS63&HavK8~A+> zcTP$~tv$`;*EDUk`J7A}PM71_hU=cA7l8PpA)2FOKU4H-8xQBO>*}kM2svI^LXmPuiS-?__owT{OfNosYOq-Nq z$+Mssrtm;FeYK|mOQhFxgZnKwZ;gjEdS4@~mg4upXMCliA^wIev3?U3t?j3SdulyHu#^tAT4nb@!skq8N5NR`?}ZxT*`<%YzmzetWh1z`jFQCjSBVx?L*H5N@%!6Q`p&PP zZV72*VyFj;*FH49y{?V!ZyFgWA1jFadW<~Gkig=uV^mFNGU~-dLu~vV@;K-{qvzK{ zp4L8Qy7U*~yeGoAYDz5JT5OB+4@h8!^;~q8_)6#h4ksGxSK+ajIW$AlkrcWt#jBFB zRCgR@)jt%mLe07qKbF%lb7`*o)C{ikAH|S&Q>c=C7hfys<6xkXK#ga}C_7mSCe}Rw z|8+X};ni8DcI-C#?A!r#zY>J(k5~$i7h!l@40|xMjKp8A#BdUew(c?LxMK?Z67!{7 zieFQOj2`O!ng5+9ZRER1D30O(-L~r{*zNWYtv>IE#*=xdJ}nzB%ke+^rw*6MadhLc zO!|3TC5#LC#R{bd;=Te~c+&R;m6G~!nPdU^*ii|u(x0JNdO2Qol0+e?X#D5AnfT=P zk?fm1 )F8sta7VL2JW^Y2sXO{u2FP^oP6_|pi^d&B5Vg-65F<>oj&vTWW38-h^)#w@>uc{)A!{l4zKYg6+Xp8)6EE( zka3`}I2ww+iE?He&G2^OH}rhr$h9~okqct|_|G&2Jj9nk<_;(DkdA<_&ieGU_G;)T zbigmQ+GNDyob|GK@m$=61ky28nM-_sil)DDVVgEGur=O}j^4_}6NWOJScW7GWH`)t z$+IGt1(6lMa-lx&6f##zu~j>ty{LSbxJ&Xg{a+86c^!AKWYCiv&b7k3S_|>x0z?a) zSWFlH!k*qN4R1%|>G*XC>&h=dbo<0Za{H)ULkT+zC{sLFkT>AN;J^UR#40cvN zpxvjB-oqvQ9r1nR#Xu?UtIh^w&DB6{LKC^TZ6#h&w8z1X3ix@E0o+=9kXiILhq|tj zMce5bJRc(i>BIQ~d6`(YYR!2X+n1X(x5l1#}w2=y*9 zjC&gI2kdPCnDU4?uX13Hc<9l=c43+yW=U?iUWd{nPw4U3&*YP90{7kgBcGLfM3_Pj zy4hnS;_6B8^E05;dtQ>*J2K#vz>+zAhP3>UJ_ekOrv*E%;k8Z881hXPcaNT7g5*Ac z;;XY{h;D+rrrEUj_5e&+KN+q+dJeOusR^W_XM)0-Y}lQC6zddJNQ&rll=V7~Nuj2c zD=25YcHOjQ#>ztI_06D_@ROEXnOIw|?H~`;HRu6Feb~o8gVl8cBImpb@1;J62E#m3 zEw4`Id`M&FH*MjzMaPhOo?~?Ub_mHxRU#fe&3G^B2kRx?N#jN=Ax-Qkzh7ZbJS85| zamvNixHb_UJf1Zou^=~fKb+6!n3;aT*3N;4ar=!FytaQc{_a!4 z0Z{zG* zS?-J56gaizJ%sny(v@NdnBQM4(OBp(%eszX)F|Rru z=0Eklx9%``{)wpevxi8wD?re5D}l%0Q|8R@L?~2Q1>gTO0ae~@Y8A7C4ZJS_&l8Nv zju;ElAgqdu&pyY}j@k76umv^GiN=T4C&|k>8Ek#I80_rIq08?*rBX}AG|e6VovMWi zWAuyr*5{Tga;cN#adC_f6z!eEWpB2^g3_&Mn_`W3D<4y=v&a4tH+1K-0{4C7@XuXs z?9&wncFt~WThqi&(qB)Vf84~Q$I@wmzA7#DD#JHTr$F`SCuUz%6SFnnj2rxDgOwT^ z=<8a0h#ir@*s4ri?tXz9gl5p8dooZj?*s~)llXbkDa?JD!ESa7!20tmNI}7F?B6EI z@51iD{IY%Q%<88(uCx{~D7o6|%rv;QJC zQ!eb64*FWuf#Xmt_$y80j_>$}9)*p>HToOXeeDHR^MW(a3x>nP+A-X;R`1Ts!dbKe_tQ7 z_1j!Ap?(}lyq*CMrnciUK1=YjljmhUkHLCn3y|m%!`~Smf)`)4xwnr!am}j{B!lKe zILNlq#8IBJIxNcNd5c2-_Hj^PIv;J^a_LLX0%{y$aodDIaImt%w@>&P-1qexH-Z`Q793=z32S zi_Ip{;)>`~eh!U->q+gheUQd8VM32y1xMSh)#mC?zub zJS$8h^C<5@$tQlnqJnUX0&ClO=~OIm62u?4f^QDwgYvQ<*nw;D_kD9z_Dd(~N)F(m z-Avb+xsorf{V1j%jm}MVn0vn!PL=ZCDK9rPdT|!_G_Sz7y`QMaBt>p-QaB!H{fy#1 zE;M@H9cESWTz1mwTxMo#3)#P;h=#6o;A)*?=!c07^!Qs%+%_P_&!`k&+LBgPC);Q6nEU6&W+!9jx3qpL66RtU~)(P zuwqhTn3{c_c;AkJB{g!Kk##=1nM!amS4!c<*0I>}cQs7jm`UC+cOd@cL3p>Nm378s zRxjl_>Lxws`w}TQcJ6w7FD?##{yT~NWIm5{t`miu2XSNe33y&w1D?Iw)K0qy2S)7xVqNG=t2j^|bIQ=ieab$_Ucsw}f8b0fR{N-SqGdopCL6vgf0-x;Tx6OdKI z&xNd575q>}^YN28nJCLfoq1;D*o7(3TVV(qr_zZizn9dy#s_TV zldbPwoCATP$3W=KLDu~WBI`7hee+`h23T#w3c)cNJLrVv^@s7E!c|)JQxEdj?4+Ix z{{zpz>u~0eI-FAB02w>vVT0FU+@h5MM&ka#&P2wUBHz*N5wXZ-^G7)07VRx zASAm43u1fd+L8P8;fAHudTJrFeNs3mU&+8C#b{>hoC4;GoDob8mcf<(4zShMDb&$3 zt}!dR+RAD;AAVZ2l8pR#(%;mBVgEE>tndKZ#ffqCdz0ZKQK!$qi_Zd4qVlYO`0u&Y zn4>?A6E;-EvP+V<_QzEeU02AwOMXM0%v#8cz|}-mFCKSx#E@I}z36X;1olo`Dp~S1 zsbNm!TKM}|9Ex|O;?pyF*bK+XpJicClOG6CQ7bX`a~Jj30mixT8|WO+ffSi>pmMpG zQl7bLx9AMot<~WUPwb{b^KH1L&PQm8z#IQuG2|xO)nn<|Tq>@99&KKX5{Ld@wAn(k z@p6Y5-B5cF_g@&#xhtp(_Sdu%KZ{yylavHwNfF%7t)!p-B;xpY=NRKRxwO8s zmtJ)VDXG=tz7G6{A8M`OqKzXapYdn)FG#}S>>N%fNesBO4diy$RL)Xfs;hg$co z$HX~F@FGnN#zaoyytg!?acUKcoxe<7=HKMq14?vu)(z^}*o|V2b2(kRa&UWTMQduS z=`UU0_584$o>&+O>Z|+d-mWL~mXkKTR5pj;xI)Hi{!dyQ1oUpmYbJkil&l{8)Tn)0 z86KVgi|-jD&O6Hmn!m<^Z|o~{jXF&u)8EsO^#*j=f;J+yFO7WHhyeL`DYVRr!{4j) zP%HKd$@DrzRbJK;IcaBH?9oh(_2%J;V^^71EI&ujl|#8XXW1l27u;HPo>?T6O>fLM zCUf~S?@?hnnSA;cv*~mKO2rmY!yG3(@rlEC6&761OewBP(+_*M$nkE$II7V_@VQ(Z zMBH}7p8|7w^SCMcd0BwXKL&WsEe;-h#)uzZLh@MyFde98AMv|*on8YtEL=jzh-ZP8 zMiNu|J&1Tc7!RS9JrEDC=)Mu&2_bk$b_-|md3v5>_9K;1bE#%$I;7&_r784=;YxBi z>nZVTx50(=OQGALot*FInNYJfz>}gGn3dj1W(lpMN!w17w;P-ww%M37GQLN9j+9_w zuM#@El>qnk`Pk3%CX^IHV0P3;s-D!xEU^8`-g~JG3yKw>za|QYch!*O??JRDG7zu0 z&qcoHMIUy)Z^SFn@MG{D^R2N3R`qNJv9@AJxvC&Ia9NuMcE{lX%lmj|sW0m4+i?M{ zd`8zPm`Mo&lzniHYFDnulh-R5@0In9K5JV5HD16bybX>|MA7<|AjZ=7BUUJg^ojpLWnMqQachi_i3H@;Vx|Ljtv( zLYd@=&K$Em1=F+noLrF(wP<&sju43| zIe4YP`#Pjg)3S>t=)Fdk{>bBbNrxxlH?z6K;En@4F**km=gGscp8!t`UqkN943un@ zBP%pVsn2K*)t5L(o$_@!)u+K|Q#F(Cy-2`zr+%m~PynAocd|D$8GpI`BnCGhFvBAb zAgn)|z&0hpc$d|5fBPJ`HGd^FGj~SY@1vw{-XUPi6`3AqI|!INOy!mxgZ)nh5MJ~H z{py6dVE%JVsW9hSZ>ZwF0bMk$j$&@<5`gF}D0V5DTPykwU%&6?k=Li#U1Qprty~&i z=+=n8cCAL~>SA`M{B2_Gp^qVwQ_0Gj_t@860E)-|km_fBcw1yWEZnw`t=WEz281iJ zH#RIMM|mEdb3iWCm1&^b`4!xi&S1EH>;&xj*hZ?orU_Piw3FKVLIT&|2{`#eGAnF! z5^1IcY<+wReWu>0vSSm-u8e4gTKS@9&`B)fdA#QO2hpoK5f|$7dmZmL(ehOi%&m8a zv0)rR@v$PbO|^uqoH7P33{1nKQ!2#jM>hHTZZY>;y!UQ}i-xms*5vurs8xhC?pa`^x{9y6B}~|lM%l1n#v(>Fu#9B>eg@=B3cTKu z0%8Ht?1>Q%EA1*-cI@4$Oib--VkCVEtOIA@;~I)xd1uh=>|}g2;R*35kHrZo>)@Fx zKzz^^lpN8=`pIT=`1^4>s@zPE-4Y_(W*-MTOvc|9H*meSJu}slqnm#Hp%=E!f(lPt zG~QK*kNE!L`h5U4Lw`xmp9w$MbP1aCKQdd$!mH zr3Bd+W}k?@)%w(iXPj9r@Wb{djs|zWq+j=2l1Ed1(5T)x9yig3~=SH#YKUK8UTWk5xZZW&NqlSH(I~GlyeqnUMT@vNnfpr}WdJkC<)=LSt z_CLXcM^eGvF&MTsv6fGB2WY0@Pps^3p@!aNG+_zZ$0mUK+@F{&KA9t;MPxJ$to;AH1&!abe{>(+?T=9q5sT|$!1`@#EQ z*BobvD$f!ACM4KbRbn0a=sZ;3Ukoe$q_em#8BBOTUF6&}A~n$#1|IL^V%>VkOs0v{ z|8(Q3I2L6h*W$|;BOn&;4+b{*Fk5{FbE+eem0B;0)+5L8oU1=(bg_0IYMWrMqmC`^)O0qJNt&ob0i1U3;Dh))a zgp!O<+Cz)}=dJ7a!o~TWXWjSbCVnPeWUoU!DL>*z@>Z9>X-hkigy2;V=jz|PDvE`)|5mGwFGXfvcYONTYwn7FI3lzEj>Fvy6eJ4oX za|`;%dRd{E`FyQxXZrI7m;--)Gwgwyn%Ek?M=(=)= zUSjyl_B{HeDWI!)8y!9v0-JT;H)|| z@Ipxm{bwTxA@6;0&D>}z>*@-6ZngaFK+58pX9aUZL5o(1PNwBohcIlUnO2`aLhK!a zxI+n#@YPb@w_(xDw9I)0CPPn{+lq^T2^k?q)6%G#mJObDdPSaw`#|OAaI!qh4U|(_ zScN<3IQT{fH%-4l*Y3Q4q_zjs&1AUa{O&Pw(-!o8uL6(m&*VgIhvTXGc(C3YhvnIe zIEhPH+_jn#qN{P1CJTDtk1HZ{YF;G`!b#i-pGQRcyC<7#E(9K>V%*0lRc`erSFY5= zh`ZI&KsQ|NB#HyJL`8TzBal(){kx-nV0+N1f@X*3=-0^LItU4zE295E!Gu8wH0)Emfoe88MWk2cP9!=W# z8nt1C37PQe4jljZ6i@W+#~o87abSZzdIbL-Q zpWMGqLz=@$zTtSNH=hO_g27C^o;XzBk>EU}JUIBT4j&8fe~Xo;*@bs=FniMsnqn{! zmkt|%`Z;S{r{u!t!i945*aNeBr^ZNYhC2MIe@*mn)zN6x?^L%xm6x{JVb^dlyH>6f zA_Nb>p0HY~da4f7uQRw!zl+cL2_fqjAEu|{-{7YYS=7wcnOi*jE+;X28+N7e9)!Yc zD7L!=KJ(n9_5(6-`B6>vQ{Pv_Oqasku0sGC2ee6ePv7&|phy1_ZdB|OP4j!h`w7P5Mh8)RQ*xSqSn-@L`=g8dKE7gKK72yI zJiZ1I2UB3~##EgD_BkwjKa<0AzPNLGIx+ktSN;3f1>9fjOzOOJv1^^z$dA*%n@$8%8nH>noc?J|hK27Bu~E(0Xm6lK zEsa0op(`cS@XrZae{XW7z`zsS|Z;P-nleQG79-F!h(YCB-y?l?$W1~lQ`DOfn6l@95S!;s^May9w>BoVLcM&rUAPnendry%Qv9PgW245M4uqgVG- z5)r5mFLsCF71e8GhItl9ez*v7G6EJDu0-qj{x77)7ix97s7>{>YRA#JSUgLJGgr!{ zrMs-Lb;%|=%QXy6f0l#I#rHrqJRbDhmAIz+Z=pY?4=h}Y*pnw}u|sl{+)+xWVj3o7 zw)INPQ08+AV|U;V%d_NM{TX)KtMj;XrZ#L^Ux%9y*i&*S2a3Da5rO+txceWIU~RoJ ze)_J01r@63XhXo&c97J^Hju~qb_~vX+6wauO$Yd{AWJvI-=yCErO*RsZCLh{ zBC5$Y@GYDTcV}Qc*S=;I1o}r~#)p$IN6;KKFJHm<_{~IQ`FM+#Fh_bYd<)1w$iqvP zavTfuxT^6-!NKG!4vN)ay6qlV`F#ozNs!?_x-I95^h)TV_NgdR#F1Mo9$>Yn!p)Zp z)R+V>eTb~EgJm5X@k`V^ZtR^E0gXM+F*~{i{RW=Fixn@B?kJ`kjuoPaNi~)z_t3olB+_Fs8P-A$Z2B0FBALE8 zDdZA$UEahzoczH&?r9#awdVP{Qasn=jtyfh^^jH^SA%-Ol@%FY5*iW{jirho->tri(Bj?_9~@5yUZ|sd`xHh%H;Al~tix@+ z5*F*f^)88wJ(fjUWCj;yQLx)qt^eShmtb4 zJAVpi-=~6RR%`Ki;~mPCufr|r+T@3wG+g+j4UN+;!=FtW0O39;en8m5_rY|Fjn?|y zNS`3C?z4mi`Oj(7W0n?57UH|(2{i7))9PbWVrf&e0pvMFvgT@sN#P?+l6z(rym}N& zv;D>vT9SDmL?$ zW7J2QGNOi}rT?)D_HTff@4MLj2A0%pRUA_@QcIsb{y@XaZjopU9hmm=GpW#i3gr(~ zpxP>pUU7Sak#{>WWSaonvfc*g-de=_)s(o|CkENXXCl0>;|p4r@caY5j+WiA~| zX@nXL>})1Ag)R{1k_b6wU)Z(N4Pb{tH2$zGAm6qOk(2c&&{6R>?zp{<`#EhFw?k_L zyHa0b!{!ofip`^@=jxdZy;7*@AEc{gEXd;{YdF`O6#hn1iTwpJ?2bYStjHb3C_c;S zZazo9@$W-B^=J4|q>DzrpM#3m&G5#h6*$LIh0W~q#Lp-OkF^hgdqy7gXa&Q@(_vt# zo`X{l(O&$@*G9(=`cS#~H5C`y13zD0G@tSI3YCnjpyRI_Lw}$i+cKbQIUb~)J69n4X(|@<_0NF zV8$N|_e;i^>jgo0T0Z=I^nuBI@&Fwu@vI?Ik^*>|of*3LUggqUdfhxHD-p z+AbNPt4Foz=*({PPS=62VrL+6b}X5C=oB8A>41~9M~K3o6>#M0D9r!R3`ZREKwr!i z$>$h&<`mC-A8w*ynk8g@_YPYIaH*;Z+hQM~qXF*&n<~qFu^LaN$F@T3WhXZ9 zj~clD4hBnaV>lN51ru0fE-t91`k(eS6n)F*_=rD3Q8gp3_Bp?2jjx2)<$FPN>qSP| zw*>?8twGx1Jj|YY23}V*(WT`th=oi8P??2XZEGg|@8BKwW4a{S(zy-t9sZELWdh$Lr&?hjXZc{YUAoOT9o zELy~g9+2RYT%GZnEuZavq6)Sq4^o+Lx?F9N7j*Ai2**XgV87v2j5GnxSF@j~p>g!_ z^i0OONt25`s7-^5S-i<}G)LW&VfzO^V!gwiTQ_KreP#*Fow5pSwg1MxA{{hje;6J) z{2$gWD@3aedL+pC47vGoJZv0stxvW#b(f6C?OBXI$4rucc%oP3FNntrF! zIZr_6mj)i$w;O*A3Zvz_>!2(jPAj#p;PpT~Ds$r5&D%X+QP07Scwg(l@mj0luX;JG zsWr9OS6WSWeqRVLt8S5_?&G*ytIxsZ0|wlimQ&w~T7KO7fdne+Yo#C9%glbH_XUfraftZ)pfOB6j zUti-p* zRiNm*6H@1Ol9gB@(Wb*I2ISuD)i_*opel}UwW&PBCt1-M||pt-Ee#%isoGgS)%wLlI<7Qm(s_;Vx~ zKHn=K>BC+$+(#8Ol~!_J?k;9LgIFTglm{9ejdXZxG@E`Sn0owFz`Hx1k|@st*vq$)8j=EPTHHW1eR5GvL5>p~ z*GpmeIes_e>rnnNNHTgz;%t>+&56Zy_Pczz8G8iohf^Y7tizcfdO_}P(Sosw2RXJ< z1$%fl&81Xy4F@bN|$%Qvz#lw5pW6ur8ztnbx4#q{E#I_=i26 zmR!!;1Ke$+@7Nu7hQGtOvzDJ?8Smr_{+&F}!pLDGdpuDaJ?bmSmFy9W-2D>ICZE9) zGi@{((&45~5+C%m5tg)637+@wiS z$sMquTaG#P$_W=`jM2FGhp2JfiT^*JBKz%H;O^~YNS|N`wkwrz`14o1Vc(3(MISNi zLLDRPH5H@qEY>|92fe~qiNEhIsF0b-xo7M#KjWW+ciY49;l8bWUT-D%GA`I@rw<2* zPs2U`B(zD)Bb&Eb5!aG@-qY!a6LM#At-but`^Q4Ep>7*JJ^cZp)_XvFjUnhvmV~CV zbxg>TrOdx=QW(nnov$TM;pCIrV9Yhk-27KMX^k*uH%aihYag%Db3ThLZY?Kp{Bk_o zQECP`>&Ixeu?GFvzKm$LS5(#6tOqwOYi?8|9;fP>!|S&^6ESNFz_2O2TNgzV9!6pM z{xT}FM4vHp)oBV0glH?8fQ@x;d zsY2D8;NRG8w-0{1tVKWPYK%W0hdPx7urPfpDqkz21;T_`Eq4cJc7FxmFB$N>!x!%7 z$H87r8+bE%8>}S0vr~lh$fHDOs`aA*-IcG3ru^!B^ z`;0D<V~xY|5oZU*h2*h|LL-r?kd zQmVYm9Zqgo%Wlie!t9~F^kJ|VY8$z6v&TE*CdWK1opu(L&c=|Ccpn-%Wev%0jK<-a zH_0XAkL-*KypL;bnR(Wf6Bril$0g3cjM54dQQBfW+*0K0%-UV}eSb2(dzOJyW6wdg z+b`7Lv=^Vajo~rxxhPQ(kIid&yzxzO^PDGI%y!Y`B#O1bq!~WwR=SvL?@L1^#aFDF zdl!)`n7|1tN8$d_0rKA~7c7d-rd1Ai8LcOCaQ4ti^UaSQW7VYDSTmwSxuYjQz{dfW zc%_gWzgH+&;Lb|#bHNFpQfa)>f3OJugD0K2;1R3?-5Yko0&Xot=DPFV#v3GMdK!80 zFMu^}&tud0$1n9&A?hv8q)u~gg2tf(Fk^TRJN~p47WOv6)4`dbQoj&i&dnm4YhW4x?)a?0|nX0HD(E(uzb1cfuFT*YE)1V>m zjJm=ZWclniJg1z3p?|BW>Dm+MJAFR}h??TS{6{eLlQ?epFom0EdxF!s+KbXBl(>nD zuHcCcQ6xL?B))sRg)S`ZgmnWGuz#b3MahFD>?fV`kkQ(K1$}c!LtQuy^7l6xBYE<% z)eUlf#zBerS@;-`3x76OLv`svcyq%MDvy4GU$YkjGck*45r`#ve`N5(Svl^8ixH}> z+=dY-!4%1(fSgBfpt~TT_K6b9;$H1_x?v7ZB=B~Z_4H0n+;GBFGi}u<*0Q) z5DNV>$MApU7$_8pDkD4bY}jsG>)eVF;YTrHhM9#C?=dwERmHziS=jc~6$kzvMdhzG zvv*LvFn{fT$X$%^^iOKv7z3*!V3h}yzfYyiTuQXzob@CM%n)8$T zjoqZ9BO=^>0Tbw*K1S-EbU>a|HR>-`1TXywAbVjNC-cXY6RJ5)WT!oYWB2!whF)=t zSmr&yt6Ptqw;l2R^#UBn>_eNE{#@F5V+>t+gqC(MAkT*?z)vil)L(f8w?ZTFo|88d z-N3W1v>icKFrD51;61M!>4hT}+2q6+OH_q6o7)F1=j%WR)p0yQv%0R(md(1n#-)v} zSQLq~wp!5YKpjjvumJJ$ZmwL!7*y#+;#K<(#8UVjfR#CYqA7v_t+jC9OM?iyP0_&Yo-!_1*kG5q0|ctU2-DPyLZ(+Ox;?91p+NM;@% zI7y<`xnRfPEE?68jSX(}-c5|~4*p*I;{~%e_XCOM8SL-3 zwW2aT!EwKW$V#Jg%qx`_q}KH(Uc7x4V_!Sd7=si@zdo02-~y8jPI^5$r|}-BEO*XsqaD^|FCbTsji+C%BGDzk z8QU(;A-YS|V87*2_I!j34*YP#$G7D9{655o$9c{3j|-S_EgzyPl1O)gD*V%u;6{g{ z+0<_$n4}t6t>e-}G;(Iaip^fk?3JQC7G#<3AoxtH(B_0x$p$4JX zVO2#O<6Yv8|E$*2KmMNlZOo(k1KF`K$)5YM( zo?E1wmQsg5HMA(Kk-6Px49DLj()Yi@aF(=-#g^(fv?V$ zQKmZ*g6xz)KUOh%`(1XQn0ncvUr!KcG!Sk`&>$ z&`!waxp4!F?=Zc4W}$W1SL!im2dI}M;M|!-=wM;bE!nBfy_IxC^Na-gRxcXn@|hzS zDFfY+Jqx71Kc-i&wlNpWPr)jIaFXk_6!lwr=&uNUY`$@mJR45oHr!C>j8v^T1vgL5 zJMuN2^?J+v+MNZ4hm6TIfu~?7a*`E4D^4Y1SK-jZ0dnAv8|8-fgNt)7vGIs!doEr; zxdmC!aAq!_vvmU$n}0)=tO-2w+y;FzGQ8(7jQtWALkG@;;OU`n_-Bm^b6LHV`7N)9 zLPMJ%u-}LM-7|=*eAKzewsT?8Q8DzF?;_;DL82(tO5$fHz;bZr-nfL)NVhKfDt<1{ z#8x1o{O{T@WQbNdEu_}W7vI+3gG|LUq^#k$Nn_@!0oX-Xbo`hwI#siJMKwaaAIUWL8Wr>lE$}uHN}H_KYR! ztCr#Oz29&`*D!q@$usYUdg$lZpHTVuPaI7>g>w#HqOmV|U4N_<{N`D_Ya1jvu@&P< zqn;#2K5WMQM^ih?ISmTJYNJV(krKSN)Nuc4w7B;eU_F(HNqV6kH!vwib+>iJd{Ok@rd z|A*=@r_2(Jr7Bn%o-fk0#*%A)q{xx#dDzf-k~=sgf~CjR@rQsZjQOU3SIi`$9jpab z?i=tRRpInaFTmn3L-Tp{ynf-3D!DJ1Pq8M*+)P*qN_g(7w+=&g=0$<-$)nUPI|7UH z9XPS(G`#dA43+kE6A5x0+dB`!dR}`ZwL2MCCtoAyTlLY*{V}`K>mJzN7R4pbmbAtC z0T}DewFq4PjM*af2EQ!mN5SZ?OtwRqS*v#zZgD(_Yg9LqEyJR`|I7ph3`Nj?H9rd$ z$KhUe9em*#0;P+tnuT{-VOr%No;Q`Du2YLii%lC|P&tn0pERKBNfo+B{SO@oF@X1z z(uvfY5QhBNLnYt5qu$lx&{^zB{@m`azN6GaJ#uG(NwEpW4*aGbPi@)dLOmpV*>@sn z90BqR!@0;_1MZccBeg%}24|1YfvGoeLK(E&`P6YihcN?CTj3B5RV?Jf$bQsnTqh);l*)nYxqV87$zszE(fp$IZ7I)?* zFVcf`$1y%AbXmFMbmY;d#eb0$u5FSf`8kdlRcblqWXOj#L;gAX`z_}o%X zJI@U(Eq~I7-sx2D<|w@VAPo2BtRUy&>&ZY^KlyR99LKNN0G2_8bXagI>itQh0f{Wx z?`jX@!sIPpsx=bJqXHH$=nM8x>_Ogh(N8bFt|!42t)N~yk1igI1cP6qG|=4<82tfu z+^R&F@-&N_eRmpdW3{;nWjwD*_7yRgxB+Qb)94K$K1*kw9F=N51Ji|!v2=zvukl=j zZspS;Qu#l;_2CXKbT6X58E=TZwH^qohH=G$QHWE7aBRI4cezpt7A}f{&aweoKJODP z>za?Lp+OMRw*k0659x5k68Q0}3}ern#m%BlxW&$m{@q~%6X$f8XO5~cS0wp4|HgaV zH02$(xm-qbqjl&I_?%30zd$QCH?u~kOz48c>&svbu#-=p{gSX&JHn(gBjk%gO!yci8aZOJwtm zMeMqUn=r-Gjy(D>AL=R{p+LzGOv74;f~q8nzSFX3)lDN08`Hr+#FQyCn~axgMQI$* zt~yBLscY~E`=%(Dk9CR%E&|l&(UFi{}+TNqJ22`Zz7Dnp9ge~C5(NXg}Xh3u_myB zdM-;rrQ=1Iks*mTwx@vg5{Jvn#K4#;5ZmY&UQZi}1%Cpt-2MR$hN#2zq2p$M_k`i_ z0x`JjGZh})76I9QiY)^es%rz6pNPWkYmZUr-1q8DU(eud z{U&-wyqT$b*hyza??sd7vsCu%1@`HWd{}>41D@RpA*l@_9LoQ}=LIn|fbTWuFFr`` zyIK>8zmveV!ULcz7(#Dn)4=F+)n1xfIQ`ciSUmY7dGux@l8H;W6@!6I4_*!Wp z$9!HvmJ4??|83>7WzrUav*u*fIbsAdf)*s@4u><6e-m*hfaekkIAS>mX0GuEmyz#K z;e3ij7a3s8&rEdsYD4?ActRzL4oFD)-`hszb}5k zWEY39J=kLnbSD0Z7OIWV$8%X9K<_9r=N{t6A?ye?CEQX&RDcNSpoXFIsHG84PpQc25E zTL_w_PdZK}Qnu3^1d^+%&*LgOZe1H3;!e_ORr~1)Ya!g_>4lSP*5kyUa4<|d$T|J~ z%nEX+d7pX@-CTGcZb)5#hxxCtQs5WcL0QJlEBVf1;2%hLHZ9fnq?=0 zX?(tt+Ei1l@8%hdz4O>Jy8_A1E%R_gz$u*jK!G!zeFNG`(_o6uPnEC_k5o5#t%0*WH^{2ev!wNg8FTXQ z0E}iVBUz8K$tk-%U@1L;o5tS=KmK&!^mX!RoCQm_=y1&Z6{-9_!IIgNXpfz(uH3O5 zd(h)MpP?#roaaGCW2Ak|77F|Qaob#z7OsWy!3Kn;uO8rlAvrd@H+?xlZb@$DCw%KSX zbcPlDp3QrAbJ1|(GTc324nLR8q(*$+&hD`~OpqGS6>R?xR|M`MPHZ~8Xh})!C0ESc ztHiDjuB4GMyO__a_i5v)V7NF}3n~JR(cp4PPUlWB)!@Ps&mkZRs8y$cFXeowAn4yANcSRr+>Y~Y4IM| z_hu)}(-y*2{2pt=Co5>Lv7yrn^U!1SZLHk*s@hjX0$+u%#O}@_G(&UpTsI9y1bJfM zYAq%_HG%wm8%!MhZO~}PA{=-l!fBVy$7wBHbh<_`N=V$LZPvT7Uat<^)_$(u^>Y`j zRqUk-#%rO?pMy6S@1d{O6g;rl3(fjB(6z!+R7qqdzAtgXycLpgTt}9UHi*D=2q8T` zx`~~67(MJ*iS-8l+%BD&SbF~;DkR8pqdNt;1n=wQuaf|M5R*&ZN&{1sXaPdwB1z6# z6?BLVr0enyVS4`svfy48)_2@RZMPICo16vPP!rV5mzdA}^O{Un_{#G_W`j?cA+}as zBORx%lZeD>=*m0|gDR(KfWJD8zi5gRrVUi%zjQiVtN~qKE6^i7aqORNKTfQ%1ASkP z(41)zWV7=UqS`+Tnk9Fll&~dP^1cIqs@?_TIvJd~H-g!*CLUF)e5vA$?WjIi0P^pB zB$My1COX?03(q#*ZzE@A@zwn;+;v<^&exYRxszYh`yOsAxX9rK|10qNi4>`F{?7D> z5QQbT)g>sa>v0gj37J|r3?cqI6F_kcHL?7QAp3HMV zMrr?lpGjBH7C?zQ#&V>MYV4E3>yq-A=2gbuh*cn4E0@H4=2_DhXVD`%*NNg~N5amZ z2*=jEBlgl&Amr&!7V6z$J(tw*Y}i>4BDNZ&gB(D=jIizROv%*$Rzn=GQOZ>1wSexM zxrHmstK;vcl4RvPeBbAXQcgvr{AN3S8f3`HZEd9MHJfR&rx17S-FQecu|>}KK9hPBKQ)`~T@jhW1IDC{q{oup6=iYjZEWL#-f|toXaSo2w{UGC4E~Yl_jCp26ByNz^ z!9S}?$e{_{Z?S!M2OWw#$7o5)fyt)1 zFlj{qx^=p9G0r!zPGSNlw(1ojJMWUdvaRgJbbp+vG!JKdm`ZaR;vo9H0_>@H11f`J zptF(Bp;Iei&Rp6Bf6Rba^s4czOc&#kiMe>bH{dpCjQxJ+C3ENCWXwK``-ug3iX(*nMu06exQ! zy$_`<>bu+U7Y(3~6|I?$4Stv=w-g)BOu>bs~04v6D44S3Hju zI>F+aggx}ym>Dy1##D5^It7#d)5Kl}CE^?tf#gRDN?zFsRfX+rInN>44m~igNEbZ! zx|4tXE{y%m)A;_n5q@k=h6Pg^QHST!f2iz2gTwJSUu`#c(Oul)GCzY|{Go>_2TtSm8m(gl&Q?>1zr{rH#F}dR30Lt14WAvaLYK$uU3Xz81W#A2FrXDePQ_nP?u?hz~dP(K$^onUb1B zT-8032`NZ~wnv>X`QURfy0@1dFFKiZ4+vx?z$v}eM4d5#YUdsl7FbAU2wouf9!XdP zr8E3{G-oKI3(MOj$>FkQHhoDtNmLAB3$CS;;$ zBmJPb0*6J0$(q3f%!_SlWP12VqILBfUDB=0-M@Dk#&@g0oUY}Z#f^IS9W;@u$WP?% z8|AV#?#{4Jq>4Nnl7VRf3z8ZZnvKv+Gf79_cL$ul2n%m;L0DerUA=fnjgZmTvQB6vgEC?JlS1CP5 zULTZ($4^u6=zDQy$+Tz8goc0A0~f84hid$ zw2*xLk&y3=#N+V;a7@i)=Oq3ly_yTCZtxbmeVh#%J7p3Xz82lfF$L*k;@rEhzlrmK zaX7RyjoLcPK#r0Y3YtaGNy5?)S~8Ko=-03&zxOdPOl={%a)^oXy~Sl~XiJk>^pY6F|QXb9{S#CWZwZrT+2r8RK(T zt2Uf+!pQWSSkn3*xp1G?&R9$JQxtJCm?c?3>KoK7c)WYO9* zVLlVM4I9WsvSMgDz8<}ev^EuGA0-fviN}cl>1^t}W*N?#S3})oTS?iNoJCIUYx+

Qq+76zjePQ--PR^P2$_?W_>ww)chIdPL#r( ztFuXPM>ep#)nLRy9QQmOBAW!lp@x}iku9nrTm{%9&0u8Ob6C7Acbo7bz~PejFF{R-d!c?;(ZfENRUTZ7#=c5yU6g!>7t6 zV6QpK^frecrt+k^HSg9UiE18v0x?YkZr&@4>t1*Y%8t9>#iztiC&7CgWQhmz>b z8FJhf`9HUAe>2m+{SnW!xN@Wt@*}37sOE2BzPx(}Ma&YC_~l!ZTBx z|5O5nr1?H(n8n1F^F&PaDr})4q;2y&+OO~*%v}?Nrgz0LgFOU)j(5-*^%pU)UlGVS zdz!N64b9*AoY|CAjHy%0P-6O0?yW)sLSQyq7cq;Q8hs01DrkVho=~_v@`HxoRfi3q zqlnF<20Wm5gNO)8!=dLbkRWn4`%C?G}L(P2&Qy6CBhuaGDI7vwXF2W=Mlvc$O-|JW5Nu)9Djr>;KnD>@b%(mig z{#u7Fab9qM?;SsvxPn7YBAK#rCzJYI1vCWrbD@3~uxHX5EL<4Em<->8&$;QSc_t6n z$5zt^SAOBAx1SjAqiIa)H3t~fT8a)ltJY>L6%)mrxqVyj!qDI)FlDE5@{WrAgCEyy*WgWga>vM6K?eWh(k+dr&<+)dW# z-q1%)PwXRG8)sobN-VpqWDPs_`$jyzT@1Az@w&t38N~eCCD^@f2DHspf$txbN!vgj z+3Y_R{V$uM;$a6WEM*N*h0n>sz!cnbB#KQ|O@sQ3AQ)~cB6+&o`OHdLye9mYICn=g zOAHloT74)pFo+&wv^^#IQpi>h2Xr0(LYMIr9dH+sg`tnNbnqMyWk`Zh#&LkmH-56rq`dYN5!z}+S9;omuJa6(4o`;1qG#ca-6{OH=q>$tA{l*_D-a3e z6ZBb3Cw1~J$K~C#P`FMAZ8G?qI9Z6=W?B7Nzd7gv9^A+sDwsfkN(nilmTEGe4 z8B{R73W5dv;p`Cwxb{yO4nKcJBY&L7RPUp(UGOAc5$ErM2c7VDi3Z*kHRN^uX0Vn~ zWL!Jf;*~rLSZDB^vHT_jZSG6ZZht7sGTP+L_O~?GaR-V2dyfR3aD>T*?PjA3G3qTh@UL*Kk~qSPc}aqx>7T%S1u@`HzQ;M`QAP|`t+)g!PaVjar5+jHh0 zufliz-C&^1g0i|aEZw>un>U{XR-}t;yuKa5;}(9gIKiFEvM0;<`}c%>$ZH~-=-F}# z<7>a+LE}_1_`J>Rmvt?g7=0rLAOjv|@vH$+RXW*m8~QJs$gVmYMmBP)U{PEQzLVqN z*3c!3AQ+Tv?sFuIGJB$}#X_+(Wabi8~|RJIFK$9)3$ zhRwm^T6=6#{7VDQ3S)ARFuDFvhYW7@z$Y#}R9$T=OgQz9kyW#1axV{&n$r3BVvZ1| znjJygkCZ-5ramt!xvYNAko{x=; z$gCA`)=&!{yIx_VkF2MyMiSg4dqM6Q&ut0JK8dM-vB zH~4HLOzvnVKC?d4h|MQ)+l&IdDE$NvPqL&!lkSkhw<_%PYd6rhpqAcjhy*p~3*gY< zMwCRRLqT#BS^Il7T6oQ&0vqkf^1nOricSJi9_+^xHAeVY-mfA+pG4%~0Sy+Q%U{?8AH6B z5>~3~azo5bHtBOWHXXT1mZj>`v^s*0p8#d1Wl-rC2A~l)&SHgWGZ@WZZV{_7!J?q5 zkO(xsqs_`YAoB5QnrvuBv??cY<0d)dTzgqysXVvm@c*2hXTr6kg( zP|fu|Qz?-)Wy?-JC@D)KOG#R%z^}YSS_n*hi z-MKMy%{AwJ&g(p%(@5k;Q*`NZhJ%~R1QI)U74!mDfL)baggT$pzc3%5L>_4CWIrXmK8 zkBr3nbOD)raXdG)FAAmk8PKc`>2N#9kvzDIApg1y*Z)|-b{dyZF{5O(nOuvF+*O=) z{TX-~3t*y27-(zUW;Fyw*gBXEMaj!Rr#B2wJc)mvNf~$t6bRJuTH*56hx#g0g(@Rg07kzD{VzeWZHW&yph6hMU z;%qoMVvKOBjpS29jE2b#8FbVgIcuggd4*ziQ4=$QhYa*yY0|LH5+PJ zDz*iFEixj(B^PnV?Ke5OiV>46%6~n!FD%gdM7xS&8nKu{`T}Er>49hN`1vRNjm~3ELp>Hd@X-rmp;Tl zxW&|~%!31w@?_3XIIX!k9wPrvf-yf;V7sj{{Q4NncJ3hf!8jTBF4QAKCk=3?L=c*~ z@OeC=5%|$e0Z)vLW;{X-peH^66Urnpp3ev=W(d){w1~dhC5{rKj?kthfnYLyE^y0U z5Ut~50etwc_eUnVdQFF>FEK-t6)uojqfQ#!N0Hsn#2Der(PX0YVR+fZa8XV?vn1OL zuFl#`t$(NxyZx=K8YJM*?Elzz^9x9di$4fnt|Eue%wU^x5w{utp#RprU>`s1AiE92 zVQse|8~I%atDNKT)B#Cu(&k7$vf zckjf(HZ8!>_7{n7qA4i~ae>CzNVYXm08u+?sob^{>d>yhWM}J;8jp9R;A=I+4~|5rjWdE zcBS5r6G&+AQ+E9KJlN?OjxB6AJ@X@moV`8>qS^A?t$qzomNbyotGkInM98y?x7o)b z-E@*su86}8{D1fmd;IJL_Tvs8sESgB=|{{^Y~wP{_unW^@#`~;YfOOXVRty+5eC&d z%3NMn1Ws$2!lBn)h|{cR-t{iQ?3OT`+9-=BvZAQUzA4?nBK836Cf zo8aKQ8C=)b0GQ(;rsEJtfR`!DT*du(ao6F z5RY+ZCt;I$CcauQlV|?Cq^hl6+@&^O$X)P?)JjSURThg2UFR)kWeZbb^4=0Ud+ajo z*su*S_0ttA+cGMa;()6`94D*|W3BFfWVWSdqF>{|i*J>0Ag9+)x*x)B2XTQNGv-c#v=_DDiuMaCc86=0Ma|@3xq<(WL z8IiOQSKf2ydT=GOm$%bEofXvV+)m6|eP2+u^FBIVt;hX5Yc(;-8u}exv3^DC(0$=Z znZ6l5dScom#QyhkUnn6dCMF{z_WwM_h%sX8cY1H}3Hr}uS%k4|Q4@fQk-dAec_UHJwM%PObuNpJmeLmh#|!RK%$- ztD=i~E}{OCFk)l80CK7s=-INI=IfY(q)HlWxMPavR;0m;YsW|%dxo<&KEUbD<9PrZ zi>TjKNq*Ur;A!o4ye6lJtEUt|g5x0We)524XJ4Tmy#MWxLN#@3pTkkR!|=uCExv!3 z2fJoJ#7ez<+@gFKPyGBvcT9eXDY@?4v%ra{KV>_tYMo4&lPB@ZYa!0u84D?yE#!B6 zG>ID6w6yK~Ou^0!A^OL@(+cuH0&lg(FnC06!xXI1a{fAn0ZNj-uVMOJKKHO8j z1|6GAV4>SZ)IXpL`*L}2Wl0scW%vMR+c}1kloJ@5a0F(`-J`W>4@ky@6xe=r7MK3~ z47zWcMP(nHARXsMlavWn(7+r;N3R0fI}pZ<9LM*PW;brvRy^H#k zf5HN}Y20GZGTyychoAfBW3jN8Og*DdPiS=!wf;7m`?nB%Z%A`B&*LyYubPC<> zF|aPLkS&OR2D;vRP}_Jh9II=_5i*aN+8@z4LNOkjb(g^OrFJA*`3$S_upFHnwDHBg z`%LD`e6n#~41X3V}+e*Jbkc_tf+cPH=S`Z~AMKl`FcS7{Hd>u$$I3lbm4!3jEXIb= z72I8oJ~Hq{N7T4P58l`qbH$!sQn1B8SJ=C!k0<`?-ILB)C_^5E)bg~52os}oo zYW+#hb`6mG6HPw^-Y0J-p1|h;d{)$HCr0eGAUC5@aALFs8?IK)EHji9%<7pz=EhbK ziOY+q;!#yDIe9)CXVXo0svo7+*LK0kYe95hXb>`AJIQ{{KBE4>6C$jyv(djQMH3At z!pSv?aAl-33EI_8!+TD03#7Z)NhkJWN2)&86*33))umy|f>Ri{cRb#D5)OMlQF=8& z62DB-=Co}}sHK-AYh(q~|FJwa|J7@9-&vE{S34Ci9V;d$=gj6*x0>Oe zwgjB~*$y_HNy4gHhIA{+;;k(_n_kR={TkRpj-($Gy?*%?zdL1!+W6T@hkhREi011d z(L4HL)(iS&?O*(;`JTKUXGxE(OT+T<+qsJATe#m#jtLsfl}Tbu898|57nm!Ah$M>-RBUuZ4I^4@U!`(4rl_U+eJq`LvL&RF^Gjnx(FNNbOpmAyh zm`erVr0Hic; zZr%r?SY`!{fo7n0(}+&8mjx46g;?;rOKQs`!2U@-du^pT=uTNl$G=Y^$_EMFKKcYT z?apDSP9F|=hSKR$5n%Z~hgzmDVfPrylgW47aMa*Pj2p;h11p9Ej-PLm@z0%Ml<6HN z$?FE#t{4Zg+g$uv)d4rZm5zVal;k(r+rv(T~Hu zrVQu#FrKd`73j-219*Dm6p44)PIC-4(g5>ztUbLPSKAKJ;FL82<7BQn2kq$!NTwvS$)a@_UiJyPM!zt z`*Rt0@??cbB7Zy_97{mGn)jrZETJQ3t3Y+{AM1DeW*G2tF3)BVQiG>u^p(PFx=qas zyK|o59G<~5yK52L@t#ek{L>(CwWZ*q(irl~;D)thGT&P*Xr}YJoavM%NoaV&vP#@% zo{PT>#=P0WHv3hP_>~`d=B5jVben-D&vq#mlcNh7CXy^x67CgG!06l=^n(03m?tL< zqYAd->p?LzH_l<`iM{aPr5M+Aly}K4DZ;Xk|J&~v0{eneQn$&0tbe0K=4&J{c`4uU z@~|(6h1?NYIY)BcYF3=_GA(Y$kP^3a@Nj%d`b#w#Vd1hZz7fpAqY|96gjoUoVq-5?3hE{jQ6P$`qE!q+s_Gl-w*CF}~x zz|p(Ca6?oYS2#t23)-GU)M`g@zfP_JDNi-}pUe<>-&IY%{c$FD1J2Q)?|Dpv^FjCq zuV{T&F4=$6rS`=p3FuoyCg1o+Qz7G$DA%1 z{}JzulmvB8{_~wy3w4vSK)UJ`w0n#}rL!Bbq6(m%v&CsL&!`QFgB?9;MD%+vq-D$o zx5{tma%2ix{JMz;D?Z{Mc`Z`wu#sCoRvOm65F1^%pX}plVuXkSn+~#kr$yq zb1K&pGK_~@ZE)S_v2gpeDhvo`Q~#GuZ1LQ9I@z_Hc<=g3a!-fSQ<5{ur;E=;^}IsF z%J2zMcN&M9R{R_QUAZr*BjM3TNqF={5#QUE(#XRs&6??hu{p`K*=Z7nS{x%yc_*lz zyA$1@^@xr-xQ4!q`%Gh;N{MM_552W;3oA9O1yRMhqQg@sa{JE+v0`gI>$Xe^%J>Z2 z7_Et1^1ll>r{^GkoXp?ve(r!ag&44z-A7E79?|D5o8d^)3Wh4^WB*y+pCA=a3@hgm zo*e=QZ)Z^hSymKsM+4RlloAayDXO^n9?8=)f`|%FT(s{c{HJ{#Cw}00saj@`#tsV} z-N<0p+hr4-8woUS+#9yFY=GGCJhd^09qEQON>I^d2Lk0WtVzNdc=SSDR2p-YY73r| z`vaj|fMGPI{*d7;yq@Are|!9sKyffR7CzORphuzvG1mNuvjUG1pVm@p@WUHEO7Oe| zem|kE;RjvWBF|2Yj3&KX!{F98ej)#lI4k>Q1g(}Uqz7i1Vd;u6KI=G4I?BdDSehS= zd^Q_0@=M@&qZ=4HW-yKtQq1r!ajHM>HeFXz3y%sDQD^BEEL-;tj{MsXd)WPKWau=gbO zzF<$26>Q10Tc((H_XOMadx*SkP$pX8!^F7h4>ZrYL)*R|g2*SEz&l9}YzkPx&B`0J zK1m#RHmKo%y&HV~xsM4MOheUr>HPM`Kiv%AWi9RFt)8vCfdM+dk$}?PH z;j%is*iixv)vBPFk3 zu3v@^1nnq#x{oXks-f8rdTGv#(VUgxIt)slLwapm!P{UHUH2mj3MPM`0X8$RLpL0! z=skvu`<8;Y$wuZp~g0Us;kusNlNmW&%g&X(t(|L;hCciajOS=`43 z*>h20VLs`;lR|xjjZ^^#Xl0)?G`TLJ+(mN|YADMsJ$43X?0bvrivKZe;#9mimL-)^ zlH~ioSU7(E3)o(ihgNwd2%GhRY#1CblrvI+)!klXMeJh?|78Plx|4Vo<44gcz1sr) zLl;=l%tGq4_OC#${}fz!wFuH6mpX|?q?-T+n9xKl`@H##w6PC zX^TMV^HKU?t1-S49A_(>JK6RT=h^+8Gf|?~125lBV=gy8Ci@dSsmj$`WT(0U;puLc z2I;{twl^5#vw>y4nDTS6cv!W!4fZ)7BAa7UV7TxMuJ_7BDUU+5C}kuZ5eEbTcu4Lko`$0n8W*!If?KYx5m29MsPar4wjoJAMR zj9AG;DxRj}`F!ES7r$|C?>ic&*$#(B>Is(juVE@|N1~VgWRj_23OQL1sAFb29acI( zTKMy9-Ku$9?JOOPA*GmxL3n6QAx_Y{kHUsO0!6nE%#jNp*@cJOFaUyRiTWOR`BjOV za()UfU$dT_{nr{)4IVLL)-1-8f}6DaM3&%$);G9y$PlK5@;wC`hQi-iCa2$)b)VKq zGc*~hx8^#{*`$m$jqbb`?gBX2SizGoU)a2d?rhJ^2Xy>JW3J0k9rLo2VAr@;q(55$ z-A69v5^jcYPF-5qlH!WJA@Qi6B*eH^{W#Z6R%mtgE!hz?o+fjK-0V47=&XB~8*_L% z-Me==`R#2?ueZcO;OR0!ij)_%`ZSAE2u|hB22u=fJx#~lmqBh?@* z_yoSn_rZqL`Lp2*jOAUQn`he)sV%+a@8ii#%bwqm9d?qtJ)FmT zPhxRH8D9tUHT=8(wv&Q;yUEkT5@dlw4PWc*Mb+mI$-$cz_)7IK{$2Q)u5U2mRz1lk z4li49@icjkS{|p;c~7v$tC1L9)q}y}1adYsi%NHCaz)Kbpw!w#>&*+1w8YZ%zpjkc zRWW?%bbygP5Q%|456~{sj|1a~AU~hAD6aZ_(gSouFZ z-gX67pH;!tlfqEDF$`BRNf0j5#PQ80e68}HeCwB|)$x3m8@hsahxb z@3)+AZ+$jVn0K4~`zV?CS9H>6lT64L+bJ+k=+222$}=tPs$^bjFeYD@KN!FiDMP@>I#MCs$c&^9NTIh$#i^KZFY*DK0Io$M!7ytS0#booRWbDRkWbsCR&bI3rnjXkz zCHNh^Oxt=UA~u=+y5>hGy?M%KI9QX1I#+5A>OqK?{JWI(HbI^-PM;n`I!nFezXpYV%vgS$`ovJvN z%Q@Xm{k-&;{=1sOK}HN^`>Ij@;x}^R!hf*8?iDmPe!^u3DD@FvK&79#p%^_)%1F~Sme2jlz86UF=f%v|f2q1bGJbbF&7GThh#7M#iAqb)qxF)Du`ODg=*!!{ zP|AF2@i~Hq?hvCZoX^umo7`EwP9b(1M6(|z^?~O>W$ybdN#T~nxwts7gpB^0P7R0W z5pK*p;nlus@<-hQiW@d_we%g0mzU!{4W}}>v+U{2BMZR)#0z$0aVXQmyN>pM|4ltM zWa8EFuKY~t4*Ow`ECfg`!=AV+oWgnDiSVL6Nfe*rHS;AkRCr;0eV~?k(Jym%(S$|nUJ{Z-+hEV~E%mNC>kZXH}{OyU|XZjij; z_t-mPAypSEM{*<>uUhf`In|N$+A)5HI=38tNFT=FuRU~{Py{JY?~<(s2{iohb851n zgw%hjAT91YxUJJJqeRe7JjUq>FL$0`9QThVX07JnVEKhLGY~Sne@Ji({(WL@-{bdh z`#DJbnnI>jSfU9%&O46EAb#g9Hd`=CxI#Q0%Z&5EWn}{O_bcPI2nEDHDZWb6oz@+<_zW)sAZsnMo%{4^zW-QIPbQI01 zD(N?SW3vCAB=xSI#+9Ku_aIc9^AL>S8k#D}uP{8u0SkdSp{Ny$UfG2M?_lr<0hUid?lWg3<| zDI#?9N@(+3kNzvOVf{lBw3vSxVY(dsWAT`sWE4d$zdsZ#$t`EA;>+>P>poa^+LY7; zt8(%F5vaeA_eGouWAExF(s@RoNb`{~%-H)U$!fnidf#lXsLFFXOk8jg^3!}l<(wk5 zTV)65ru-xNTle9Ml5kA1Fy?yV>`91PK5g2S45hzTz#Y*WSYUAjM;aKh4ickCkx`N; zJNr0eskxv0)9Hp+b|*=~$y%_Rb4$?r+lu+Rgy(b>JL8**8o2O%84RCVh9RHN5e-`p z=vU+gYVGq(ImchgWFR3Ta*e=`}A7I z!0j7dJsd;d1!a)z`bf5Vl?H^>XJPx2WNcgE$;rl^$7eGwxUDVQxnqnYO{_jf{&V~c zqA}f+JTwARp8q0J(Sv>8_7h#Mk2M>cMi*XxNNV`GmQ!^R^HcAiXmrv`flR0e$QG4C z&izD4xjI3#XZa5zzGy0EGh~J~qs_>rn@x0IL@yn&!IeIkV-2y&iEL|s0`u5P9HX0z zAp{KsKU_=ktM*MIRwBW3hZQ){;ePf@R}%eoBAP_ph-VM0+@_Wbzi_3RlQ{pCuc;6< zY4d9tGS)knO^N`~#%DJft9WUgx6y+>?M$P!=NHf)KisVxj!uN8z;kT@?5S*JB=9qfOy#Hprh$}TCS?b*o#W&p0Al?!NMJgGBGgW zg%9M$DMN|16q}H|8rwq?MbnF`aJ%a^JoT-XH7ruYzThSJZhkwhQ@x0j+N1@=gJZ$t zuQE43CP4JQGmRW-eMM$H+DbB>x)D`FdrtaLH#!-saQ7P&K%;sZy`x1*#`WW5NcJH) zJgJEM46MYYg?;q4#C&$(OB$PN?f~@@D%m6Vw}X+^460}71fJ@Xardf9G+HJi2i z$N4+3^xi*IWbTlrx;o`G!^6f-*G!cf$z zgH3sJpw!Zko7!fL?`kXP+$Uo=7&uM0+f1R)gS%l*^>1>PZHD8~-$A+SGKujWFWgmG zO|SiyhGODL#vV#x6FZu4)jc&X!d{9quMP!y2MT?G`ph`exG(+YD@uzK;5i@;LFf8~ofHX5H>uLk6#I zq*K+G!-{fwp`_+R>g}o`TEp?<_2vT)+gKdS>cz0Z6I-5oYRXK z0vX?#&~&t#40LKj%bz$lv1A`Z8g5xz&Z@(0>gU;iu?H~lOg?e_K^WzB8#*B&T)6GX zE%0BsR@lAtC8<bHA=#E-Dsh8{3u-gwdcw*mkwGY;|)8vOopYe3fc8oM5cB1QM<(H%rB>U z-0QFlQVkoaE%S(-G%*&RyGsadCp+PuCNK1J(+BG{mH2)1C#L@I8pyq9M>4<8pt@t1 z(u2Z1j9JiE7_BG^Q=F@X3pUM!8S6EKUTiwMP{|lW$P|cu$a4km{9~!~9Q>;O8y!n+ zNTG!lNGldnt*<#G-0?0<|L8;9c!u&Ny&=_`T6d zM}?s)*0GaaGik#X12otcM6+_7`Oi=T%qAAX4&5+t8*PM5GAp^K-kUL1B?D&_`2hFK zfbU}^2`eV@P6fRj$m4t7wvR^BvB&o@x5p}jO8q=A+$>Jrv`&%O_nquu29nD=dx)RM zI$X0s916c31_gBwp=#=0m^*i)@TI|VY*Wx;9u&sW4@tFbhP^5o+-wK_MaIHIBNT)y z`_>9a=ld~Z7xuE+Kg;NnG+CkdSKj?6{Qyc@tAwGuE8+C#Ny3(;Q|al}i+J@|3T`=7 zgx7f|TW}ylqk`_BuY(R>LtbLX2N|NM+5xl!+DPH02v{}C4DM)03;!!yDtPtB5RKh( zm?4XQqH>8G`c(M}{1|Z=cDob(w)bpGO{Q{0spvoAB$2FmSke3*M#- zvTY+YK~~wtT25y!^LONYa%}Hcba+|Jl$g!Kvj?-uy-*~X$6u3ovo_CohLjB@1(G~k)b8hL!5THyp)sA7))_N2h(YiUHoG=-}!G{XmnW5}M# zXYh`;4NC2+XVkQF>3_Ntg%-}ybXRx{jNY$=aSEyQd^&#$pIL&RCsdOo|Lp?b_mA~$FMqg5?RSa#$CGjGFr z8twXotedfx`ud;ddL?bhfqyCV!BT0MPDawh*QH6(27Z5G>l2LSYla^~nrKk}2nW`d z;oV|2xce)TIHjn8(XB9ic197h#N04=tqQ9%rv<+cW?`X}F}0M|hk$Vryrb3-zs<40 zEu|xAi_SHo_Ar%XR!-zROje=9dMy}mY34KZ9#>+1%JS!+YFut`2p1J=K<@!dva7s{ zG|L=gMcwQ1RX{xJ)Owz-`037W<9$efR3FpQ0DpSvd@|my+Jd{%EAijtd3X|1!B1xf z`Bd(NtE=)c>TVuA_OFxQE0f~(ER)8+pLyop{cxi1AdhcOXHkjko#cXAH@&y{EqUI3 zirx%7M;2I0lI^RWfldPNQ>pXDLw9FjK<7izx%W|Qh58AMeWHO*M^5lgVO>zm%OijL zt>K`NA{SEkm!`c^p>;fyWO%PEm-v1X7{-W0!0g2|NOcp+E}R1aJpqu^^@^K9l3VC$aYVNX}}xId|xWAviSkQbUz{IKri$RQ;Kc=iYoIll~}S*y9#D%iRor z=hc8FXGT+`cu#_S23~A7LFtnBcu33Ly7a#&++B8<&onN>MS0ignJGg=YS}C((uw3} zSyOSK{Ts^d>c`V7Y|w;p!5%OrzON&x#)$yD%hxQu)t%O}C*Q>lhj!yLo`hF&VPPUlB#_4CMmd-qSXu%oo&;4*LDRskVOL?YX+fMqoy_!54SVTI< z<LR?T@E(7uL+j=hriF<$-2qgx6R0x=cH% zjdr3A-m(~?BMHsNqWSl-AoJ(j(HV^)Sa*6KW85VJ?$fleU;2kqQ7*cjJQP;)VnD2XSMryE^K@uvrD7n4QH zuNdH(+}SAh`~dF>ZYOK6s^F+BQ*wUk05Mw}P5t>DQvI>@0_&P3%(fp1+#_K**%iK( z`DYh~b=yp+^X&6@d&r&(TAI)01@laqQORVHBF}T`mxG4iAzbSeeNM)49yjOfYbdT) zLs4BX*?cDhzwOAPWvlc^`o36vRA`UMZCk*q*PQ(OrU=Gv|6otp0QR(>L36nj5tj(;qrYtIQkt4c4)zS-`CNz%L?ckolML;#`o{f8F228ylMDa zRU+Ye5pNYo&?BF3GKPwB@I8`}s2A(tdPD-O%+SEY?#av*pKfZtP8l4}DZ;9faFCu} zg?P%9$OYsGoY$094>!CFfO$@LmqtigQA}kQ7ylL z7NqwxdQsk_MYWzd3zf-|XSbQ;{Kqh}qmb3EoFbaxYs)Nihz9qbR9HP0$+jPgWOd{f zHvCo*%!ocnDxDR`w*vyyN;`xJe159Io??sj_caWX=j`a_v`B+OuWkiWkh{BH1d8 zYWfapn?}JClmvzQEnu47kE^44@o9_!7j3|v*1&VEV8=;=xCcWqCcu@XvoG~xcz=48m8(IDY*frP&FB@ zUy{bFsw*MqU5IGZpaEAV&EdUKRj7V_CkD=(3S%Vw!A#{eY#oopCDx5Cu~r}_LmF}V z)?AeS@rRjJY)Q0#zh^SGyddA450Jp0=BTcG9~$zE$Q;W~(p&U|p79)ut-qro*T;eT zvD2D!cDjU9T(6Kdi%!#D@1yDR?b_U?^|wgJhZW?;s*_+kTTv+grw7lq7=xbbUr}4@ z1bm<py&cUK<23HwXh&+7y@}fRDbR*vRQKAR`BW(a(ws*3YHX+tSae08Y~v-GSZs{@Ya7^o`C+so{X7kR zl1~R>%-QXSUozq`{#coDiPnxdFA$t+XQDkksn$aWTC%l=sJdj4Sv9uMQh1h8P^x4X ztX)VC`t0R?{??~etaG4;XDp1exV0H&7m#j791PN@9B5UCttdcVXhp*_{-cx-D>{I>HokEN=+y7 z&f=Wrn=g3h{UDtaAIx#XCsCr)n{LalAe($Bj9gL=S4$&cX80I5t8^M%Y7U{|SY7&H zLlyPX8AY1#G`p4`Lwng_qbm9eWj9>uy%LNss$Dn_wGP!b> zkasQr=zX8bTvuTz8nnsb)`LSBQlJT;)~{KyBaJAXx`g;|Qpfsz!>HL(N*!Kw)BM3b zB!3j|Drt!1#$+#`re*5LS$#luKnW(MZD(SqtK;uN4{QpYi=SnaX~)56*#CJGM5Y>G z=^&8A;5TgJ=M-9ZI0nwnQN`BYS@@q#5!ASCViLcfXQLOJ(raH!NXD}$y-2fL3WG>?I1Zx0NeHwD{Yy3kpZgplmiM&-(nphr>!ed#-s_)hj`R{uCp zlzsoQO0{3eQNmgO5!F$f;87q54JB)5 zm)(;BL;} z;hhnZKbpeb#b@B>+a37I$bm+5{l+pZB_AzQVFk~aZ*>ddOu7nbQ`TjA)B6G&)t*aD z?m6SNG$WCteIGsbw+wGo#bIj5U0TLF?e@fOfx_VhgtdvH!@ir*Z&xSX>Y|BjXFbLb zk4|EqIu&JS8o`N_H`sAQ5&mZXAeBk$h-K0a^kSC5$01LQYd(+r)7H|)$a*q}^8m&! zgQAy3)arN~89!`~7auNw=~Jv=WZYDu8raB+pIQb!^S|I3rjPzDNntgQo1k5DQN3gM?63oiJy@S@Jn5g{)cJ0M07DpjvXBxI8+=yp0u~x7f*$^iTXl| zm~+1i*z$eRtb1206EgIIntbk}(eJO*&~a5@rWVcwz1qiy{yR+08EVk7Ct|qOcq%d> zEu>H;fELIkf&KV?rhjED*|<1~-R`A~H!MQw6Mlzm{8S@2I420_1Xf^b=0~!qRGzt_ z^5e3_wi%G#w~b7uGTZ{q3K%covlF+qxjW2Paxzzjo-t#%H(BZEv%MaMWxe&G||jb}YmD7qiLh181T9*Z5S8|G=oWf}6ZRI6qEi`6zLpbiw(p{Q zDtk!TlQ=XsyUvttm1P@4k6NxA`Trq9l7Fx%wQD zGh2u~nGWcCs~2PCra`>bQ|8UHKKfvyCG@yoVt4c!qs0#s+9AH1nOfL~`T2ch-mVkG zZ222}rcgrut&t-)s>fhO={~wd>km8n-fXnC31An7J|~SCY2<~`ERv^N#ALmQgmErq zwB9QP%!*dRQvWG%Bg&Ncu6P(1m4T-Blc@3vtSnCCXF&781{flMi8!b0?%al8$#tJJ?O@7J8^_k6KUMvidsf( zMD2Yt9Wr=}ClzPn_kbN_w`noHG#>>Ix4Y86-8;GaghA)}dw7$dfzR5$9FFYmCUVZf zoKx_As<_LA9N1HYtJj5+ecJEo&rUu3XmW%q@108HH?CkT)sDfBXJ_G9qozQp(^t4l>|=L(&8k(X<+}sd6>;@2a{|{$9DHopR9C(S7cf7;U-dFK1{ot%gIs? zA1b+|j#QFpdb>b~Nn*NOPWo|T{pSn0cwH7RwB^$6r`71CI900UJ(5D=cq~h@K#A-Q zx-x$h4U8^i6!sm#V-K~+k9A97#v&q2a zEY45dgK-rzh*`S>bVNO80$t=V`qxg(z{Mgx_={o0lJiM8+O9rEI7Uzbn7@^i>+YX9QH~JmKyY1XG%CCanJB&%JzVf|GkY zDfe_H#AQ2xx6>Zn=6(cyzo^iIpD&Qp$^K-NcNocSD`ghlsOI^(dT3s|8H*Z4ZXSVlS!`EVcI*v9Dev_!-b#< z;`uj^Hm>z%s0Z;m6qJ+8ormi z>^^nX?MGXGHEwsAK4)4gPPe>Ly)4)|nfblwH=|Iji#p!2TzFn2>CL0)ccqn4gR^9j zd?iUMNTK_`I>P#Sr(pJj_kw{)HSXNMOdPpH#TG|N(l14{_mbjd&)ebBQ%!Wd>JrYOb2Rrz>mQBEcS5DkOtN}7o%!#=RMeV3 zNMx7P(-U{@;A^EET%2T$h2D4Z#mFR*`{M_l@jZ(4s9Yl5Q{}LH>;kH);zrt3GRf`F z!RX*!Nqog@SuM5!eL`)?w=WlH((nyTI}{7&mZyPj-BQSg`&zKb=@}$==z_`B`|xY+ zTRKit0sTMtQ>_Xm?0A%hb3R|g)KRC{ulkZOIEr@!bTicMTo@{aCh+r(T_nG1CN|EP zgzrx)aW_`(XC!sPh;h*WQFPvMIlXTjS5!(`ifCvcX;2aMock)HzKNuvWYbopqGYvc z7nSx5ApOx7vASWY>D7F7 zT4H32p5Mk(>(ONPQRgS3G({b=7Gx6BR|O zVGqj3eyH@~E98l7F->sVKuXs3&^sdn5+SP&Y9H;mmMW7 zyFH(5Ue-%DAO3{1c1huMYdOpwt-$unA+TnG3m(6vj2-XYXruZ|8rhytcFx#>BL-QJ z`=T0_$8RH9kHZnt!&m`h3;~Uc@MNwUG%d9uAO0oK9MxCE#>5@!<=2q|Epy@I$~CLdG{a@geT{~$0+3vK?&A`MTVEf?ov+L4W%^o(7cXjB9hJ$Q$dJ^akS z*BszR-~f$G(c!|(!@$G~XgTYSu`814>!1kcO++?*R-6qJFZL6o0~8&0kFU3yUqJWu z+{JH;P2tyJe$M7(3io1diRX@LI>bt#$GG!?zz32L%PPT(8GIjd=m(XIazyvulQ4Cu z0}PGcBcG?Nz)n+RR9Gqt4w|kYu5-QqYu*#)`}tDprq#{fxDo|7+MS5(cPqxC{3s58 z7v&6h6;qGZH`uewcH<+XVLJF>7nvEY28SLD!XfDpo{MybG#>>lb&*8*j51u5DFgbo zs_?aI4#@7=#lOD^x7AFWJ)?VvJh?p%)Xrs)V~-kGv)`fI(Odis^yDd`auCS86Z@DC z>z9zB(Fkn#9mQ>*CC|;-+)C{)N^#+v-eR)BM$WuG3etTyk!xz1V43)v&%EDZ1rPU< zN1|t#_GwcvL)8kCzD(!pCaEFrR|fTiF*JHZ5f!;U3*LR02>}y7^81%R#7Q-XHteM& zVWtuLQ&*C_U44*=>+@iJG9qB|Ok=ix>osDRI}QUE#(bP{7|82cWJBgA>dfDn z@~k9a>&J?joF;3+XXBRrpP5SsqM`Wi74l|D80HiR(ICB8a`f92j1%smnhWj`&+GEyK0+WN-9vU*jVl%$KicnEwG&0 zN_(Rwa&twKacyP?wXRGd+f=mRu3IXx9VsL6UAa8J`@3nJR2)R=Ny6|iYnZw-5gr#6 zvu}(k#v09V>bGzYz0q)j z4*1L@TWbutM^>4p1C#D!NACqLGr0qMrmbO8U1s6mW3JFG-NqEIlA*>;`Y0j%jWm2l zda$sAo|n-<^}!ze_$!SpUsptq*c9Oy6<5|JWdX?y=J#b6*D}pjj>P0=Ii&Hq$sFR! z9O79%L*c8jJ*JlIt7OT2m#g^OEC*NheUj0H0d{sk+=8ok|KX1SjPHt4#vDE}C+I)*XZ1&$J5430F5`XMtz?~f z9?{D;fSj57=y~@VN#-KaXY6gfJ2HaJW3o_W!)D$;Y)Z|8is(;EGcqfpp8U7Mk=SdV zqw#K1@IA7d-R0Xt=XFX$zf&3L@?F#?>o$_cRk8RgdMOURYRAFz`Sk2gXCnIbO6|K9 zCGe^_lpWeMLh_eLm-B?D5bS(M9~j9zZn zOuK`&qO;{KQr7GRABGI!S-3E?C2ByhTd;si%Y&!C7nS2;i15Kjf`wZnaMr7_U`ZOu z54&@5;>_B)NPkpjJ5C-Pd$0+Uir!G$hE?K!3F!;o0gI_5TQ zPU$9zXKW!f{tw|)4uOSYGyh!|G6&aJVesEx9Fm)l9U*Z7r_6^`%Q=Zy2j63gmxP1z zeS3OkniQC$GB8K&$^PBrP^$3_OfeVBF2g_iA)U{zL6A?q8{$&a;B@bFP5KLFYVvX@QKr6-wD z3|LDxFO0%F8!gac;TQt`{>UEKn$-r48MgHB*>jv#w>oUw zQjCIaC+Yq>Sr}hi&2+FIF)2ZU%UV`R2Y#x94W3}-^~3nP+A+bTe+VJvHQ3m3k@wWM z(yB;#@HsdJ#tkTfz1dv4toAZo>wAIk23BKU;tihr$v-cBvjyL_-JzE@a*)_@6Y_TM zC84kHqx&go81>i)yB|f6t&5|{mMtpi%`*$Kcy^0KZw6$k<>8&&NZR)~oZg7!XPebg zU~8*RqM}v~b(_%v z!DHSswX}!$SR}x8r3t{cEhE7zB+N#muE5&aPoQsA1!X2_QMrdRQFPZ*a!=_Vc@o|w zXwJCH%B&HkbirwQ&vrUy{>~)ML43dE!)#8v<17}KnB$?V!!&4$6xQ#!U%&5uF3p%; z#)LVi&>0DbsO8Srw88j2tJ60P51!jeA1{1^`niet$yc7nnCg=d>FID&v4g1!7hwh_ zc2kEd;n@DM0oR*6q&jB~@SfKhoLEO88I4dzC#e9oX;l)KKWrwSH3I0CEh+f>b1PoC z!?V^t4x$Pf$31uL#@!iN^txmO8szJs{Z~uc?ic_QB}GW*lFy8R@(@k<^Nw)aPrvR2KkG+BwzTH;Af36ZXGcqSL)1w>C(XWGTUJF z#Vm;T^5FZ-_o%2>9jc8ck#5h0)U_rMci(?T4GaC~&P|b|H)j(hO#281@AkmnxfQTI zsF7xiT@d8l=Vx!X&ai1@nM1k#)c<}mJX{ma$R=ttjkhb|^^=q2&7;@+ z=j|X<3>rxHylZs8qmNEm|}!u=9k zLD!$$#9E{tqdzasgm;dU(QvC0I2d1nZC*@=@dE-R*T|E)c#UpzDg4J^L&f1+3)yW)i6!d zPR3cIW}ti~l`IZyA<3_LLE7pp{~g((vcDQ`BMa%9(2V+J6OG{O!9ciRI)(nt9RSyt z4{3mU78I87yNt{NQ2A3z!*3X4#MMPK&c+??J?aLvwjw-w>nAz7@I3C0Q!}fg51F*h z;rQPhU?x^p6SeMROt<7i({uJhsIfMgEIFwQXf+na?H1#&=ZnFT=aSSpE++TBCgJti zBk1_D4xOw2v2|aQ=z+(*5MmXBi+0qZy~Yi4`cWk3Hxf#6RHmB6Do@6WvN`-+Zzl1# zazVe03s||%j+;Np6E^W2C{m!x`APWDJ%@E^{MIno`-|rSm-muf(}Q%!tTa4Vf0U}6 z&^48~UV-<^b*PbcI(vI;CGL5)4p&&q;`lvJXwgz_n8tTSVP!OSa2x5J;Bp+XIZZO> zcQKk8<|I096SP!jLqFq7S6w(xj#${@=7gyEYM`wF}18k?y&I3fiyJGVjjX$D5V%^_>=nWNU5V9w=O z4D+ArN^ZuFmmD{a!R`n#h_N!I-6}PBMqHn(d(6+gi^EBmT>^!t4EyXP7U(D#wYZ!W0k~@7? zf@?JBBi=oxoa5fPb+03)!-V@nAW?Z2X1dwnfv9=Z{Ea42S|@{Qiydgf$U(?HXN2Q> zmqF#sSQz=u@8gyBf)jPar>gmA@@qF&b|MF_6n0WZUl*^HFteyH&{1$9MANM7Zy|W|eS2s5@H}pJ+epvuIv+ZDSe-IqI<;)rmU#9bV z2??lN4Zg}t;MdU%`ekYe?o=LRZ;TG(-hn16W}Rv};ejYj$e+z;&;QbHKHHKxbeSY8 zHV_4^37m*j5ZAvg1*BMCs=am&=dt1lR*pT&uFpk!{^oW1?B_Dvzw8sPa|}h-eI+nC zwjVb?73EG07BDLWLg3w^$jSXQfR{G{q5RQj!K2-!)H7uQ{(GgtU3t2T3>!>=WI+mi zjyp+fqV-6A$O1F7-hP2!_Xf}`LT15|FXXz}VH#?;NWZY zSOie5;XU|;&!oSd90${)WN<)JjT1t+Dh``hFA$ z5Es0h4U%jEMt-cq&HVaeYQ<+3F8IuCEP6-+*BEi8niC*jR|YtV1S}6yfTRIs!5T|V47+2){fgl^PnSP)1>qAo z4{2Yp`16&yQnw7& z>=fOEyHXS&e!)yGFjhb&Jki8v8AYP{GaUz-ZOHFk(oBf(TQc4`90i8O7^$krku+(p zN$Md)NzB8jF|ssNVHKGbX9v;y-m^~qC)lE_GlIF9(e$>|8)BU+#d`sz@UKxi{+#fF zUgmrBmdR!CY`h7vZ97OlHm+d5ZEvPF-_v;B3uSMOfw2w;>CM2KsG#8ql;><0?vBB&`=Uwngfr;-W+tX@dr#FWWHCnK zFD;$+o8FdNLNnUy(Em*~%@#h(G-OT$?MAC{$t_-P}D9^houi^6UG}5Ry zhs=520Tmv`bQRBgKYL~Yk@PdgCM%cvivg3_kfGb~NVJSDaEa%F)J%xgljEdCY9GD% z4CueX!*o{8JiJ={md-1zVO*UG$)$4xf`J*yxZ7_9Du0Y6f4lX_>{3a(zB-9X zR!qZN7mMJh&PuX*%x9+I2g@4C{6Ypy@u&G=lr!$9M*{xCdRKLFFhrOv_B~9Bnk$%d z%vItv#e^OC6-jpVXh7^41(f?X9}-r)qdGZfX}E@fDr>0I3r+{A+EjnCWe@L6o~wgD z&Rl7t-_-H5smrxV$*I4aYgLKjupV?)Rz+???VC1exWeM4s$N138#_M-8WWe2)eB9C z=ka3Lz$L-v{(E%Eqx`_TgCrLJPb|Cm>NtzxC!)y)SF+DMx{d%eZ$T^k& zGx7LN@&{PCRDxlTKVg-2J|a@5Uqh2TfA?rvgrBA>K~G{4{4@MVLUu)SzMYcX>u??J z%(_LTbv^2 z|H=y$#n!W5J(ksXh*q-dgDG%qI13th=H>2xaD_@2Ore}ah>tWVUW)c=I`N*y^P-gXROV9;3df2A7zsPg>xiD8zfy-@f zLb0hcIREd-qCK5Shqf z{1y2`U@dimj3!^DrGf3lR3iv3oHz_s%VLp7%#MjyL`@7$lue?Z*$tmyZYYTCPg zE0G%Y0z!oM*jRx zsQ;eXM$Sz>OSXN=!}Z5PQFmoMS+KSfn~fLIj_v)_=IVM9DrbjZHqADx{rH4lH|$|v zM&y!!0N^V9knUkesb|(<660sXb3p1KM^1~X@J>7?&HL07cF18gI2T&9{vmmD-W_+A>Z4`nbFiK2hRP=`sEKg`j&Dza zkG*A#%f7SFzb_S!Ef?Y%!5x1JNfKsBClNamgMkfm(Di8$+5PM!+&ZSgxhieuuC&;b zn_-D4_G}_7k^BuWx0~bp=pQuZ#x1g7Svqk#;7JaobP=632jJdf9k?~_20Td*CfDj- zQ2U;@%*Z%J_IwMF)zh_TjpKOk?Qk;N($x$n2gBjm;2L7`Hqq?ZPiJ^{f0&WC_z$Aq z1~EPb&9pvY4rs6Qfo3SAmwhk8pUg1G+NciFUZX@#QjVvwXEVy#JhQ6FsRX=ZUlu2_#W_rp&cN<}@?<9ICT4Q_ zm>^$3%$EzHi?1hmOg=&u$M=zo^Cv>vh&5d$;!eIN8Nom}f4_Trk4@(cxtM*Y!7c16 z88fYjsz>NRsBs)_UCF@MN+o12RIu6athro%@A|{KohDcMz}4}ooPX6Rbh>EGd3>0J zNjE%AZv_jRCAc=?-JbX8yEmCLA2&dLtoT(Q?w1G|1LEkUeGq?!cG8v^u_!8KMjd4x zAX-0xXkHovF4vADyQ6}&HCHCamyhsWxTRq9A)VCJcaeV44o30HY{5WB3O$*#1}D|N zq(|ib!3M(uC{iHUz~|n}GHUQ-!cn|r<43BGi$qhRlw0r5!*)qqmX3K1DNy!7EIdMMmxggD1%v+5Tm8$IeigEaGRv>e0 zH~~Ti+!@^!NM2vrOB;`s!y(lsYS=jzW(Jyoh@T|(IZJ=oX?6Qy&&n(^WQ9H~KJ@xa1FO63XDhkx}CMXeHj6*@oQ) zGr0~ngAT-)QU_O6H0b)lY_Sew-!+b>bE1+Vrr)pr=lcTOT=9w;2RX65A9P4x+CBI^ zMF}0c!|~gFQEsKd2=lexgKpO)^_yi3aNX0e1unfzjH)o!H132Jol;^`y;E?{Km$VqcfiA23A85a zC#h7~%kDl~OU7HQ;vCm@;AXW!_#`$7{#ZoPXyIhqwErqzIjcdYr0C(_zy7RvlO?OM zZZWBuyBvKhm%*M7dHCRWENPLDh2jFPUM=De>-5(SG#2xJzw?XujZK0rdKJXMPK2x7 zcmkgNi-oa1hq3Fe6NuP`a}5jrl6i^@CkWIU|Upwhl`0A;k)f7u4dV^=YuWyk9VI@GKV3XSg*h z5Ifm|>XIHiB@eaFm3{y~ngex*;%CVEU->!V8ew8`haVVHc;Pd&~9s=&{ zz6x?uE{yT-n2-5pYD{9E3C{miO>34!alS7%ab9jC$l_Nb_|-wYt}enC&Ah9f_fXko z*Ksy3dH9f-he^g`a7fhyYVZ9){mqJK*S`+lYKOq1edo#Q`a|T#o(JS$(KLA8zKIsE zlp=O|(Wvy_CVWZeL&xlX@Zu!RgmXTUHD+#*dfy*r)*T{G660~k8yCDI%y3;bcHG|w z<@i|QK5bZ>NF)ark{zz&@$BqEY8xER4df7SSt8q z@CaUO9Y;~BfNMNUsnhdX97>SJ_jj+7%^RKyrY;f(cya=zZ^z>n=`mcv+*|bt#04ZhKRA^-&v5yyL zkGO8a8GhU0RJ$ca#yY{v*SkO`U6d+-!N07Md`k<*0NZTu5(%A$41G5N zFH5+h%)xoI-|IQCj~il#jSkhz9FjN7o0$rE&CASEmYpDO%day7Yb(*=OC&k*O@h$f zJohTR4wpU^!>mIrFvCq`d31^3@b~w`M8cMij+ug+#R<+B>9v&tE)Rm`Es`eGO{i|KSj_5vt+8c7y~>Ohp=DI!@d zK}EwA_?%e?E17zne3ZQeiW$MI=B6054e({oi|`v4+ax@+tb$&f`x0%kZ{hik9kj^6 zi5k?s#(N@@Au9Y7+5hhdIr&e7PEn7dakb*y==nrw)|UaX(M9Co%faIJ zPsohU=~!338?0+j!aVa>=$AVU!Y0M_L00-O{*4756CI{<^CrQ!*2!e{rCd5o+md!3 zYQj9l$Mxrse5BX4Ro#LK3~X@ntvR%q=Pl$~rQPX~MP%1zz+v&(pg{NKy$v8<`^ z>QXdm^)|$mq1BAb=H1NZ%|4L+Vm&&^8c`=z-T|JaN#UCs>Q=eX#ED_p^CXSh9h(Jv zEIyNG{gd#evm87&^aNKWWAY59GD^OO>EbDN@TE1Gq@TLOjAvt*DrOmO>^cUoyeHF? z1}Df{9)+W`vuRVuW$NJjS`AN~yQUJq!p@Vlsi6wX^ozJ9%qc}_|=vrPypj*U_u^%$z)?SifH$@GwiAG7B7lfzs42}-6u2G82ocoR3+BFNUth9m0w1cJp6Pl$n$jA+EflsyW zOu%IeGQBQ^^RBUAFP_zhUx{zXw&df)eNPc-RjtEcGJ_Z*uYhZxzY~zkN|IZXOI`Wh zn(aHFy^i1L_?3G|*6KoH;B5=Z^}8TBNeYsMOi9%jHyBGPD;bkc{9fIsVPjb)Z0;fa zwc;@z=2^%;gw@FEc7689f7_{J?=0r(z_=5e!RB2YIeJqY7k54Z zhoGC#uPh?QWgXn|p*YD+T!_HG<6b*3~u7jzOUUu+_lF8bW8F-Oq~7ZUZi%4FH<33%t& zN{koX2%U0rz*i@USz-j+Fh-Y~+Ibz#Una08|GQ5&yl$f=Hx99@whD89#+%UQ=6m=@ z{tGc$ds~n$SORV3$RtT$XY7Lu$gpfG&h@!QFj)&6UX`)ai6@cvzQzW(Er1g_^C4u# zr}}Y@8uZIoYg(W+$`pw&#~UaP|E#amCF)|_yrpxvt%_;&cZe>G4|`6JJT@Y(FG|RZ z+cKQf@5iX{Hp1N928l&I?J61?xKLUpr0G2P!h@ai*rTF1K((>_0epL;Hm z2h*L=YuQ>b*Ip$DqAWm(|pd!!@}d;pDm{0;zkm zaAQXpj-4yS=f4v1q#mN>xnt<`-2`uED&dxU8;HM+1zek916MAbBK>oh(H5Z(f+||kM#Ys5OS-ripVz|XD(gtrQKt^FiQLZogMIm(T`draNn-SY)>*Fe#NEa#Vl32 z_@*y;UhYRs3M$Yins;hWwquUv?ZWe6$EeZu!>p#g6x?Q;$n!OK$ep|<)>lE29eLZv zE?j*9XXhNpl?$4&x~)%OE;9;G<s)FUEk`qs=8!NW zDZ$$<`Ser%MDFhtNf`U@JvA_>A;YWYfknAHdnM;9d0k(IjzOv*Y|%pGJpa&&6h(4` z_EC|&O>D!H=~Qp~Z@gn01I-7EnL*yOz9eT1cX34;J2){DcfUN%)l|J80iz;jhn&LM zih+am;e9*ER)s8br%Q=Ww$bIpGEU>~-yF3%H38?h&c`C&_jcW19<8mlnbqoNm@`(P z#Q&EG%<79^y+28E^|hyP+I4RXHg80(E~>uqH_wc*kHNmUG7>3e12-fmlCS-H$mm6$ zxx2@kYiN$cl;wT&`zbpvqf{Kmy=Tb#`HwKo?LT@GWVm+)|LB?cByx981k8SR3iAHA zz>r=SaXH6vb)g+(dKT|rJ3k3a6z5PeS6_N;>t=59pC@>IZywUMenh<_pD6J&HvNsx zWaz~$+`q<>93P4%Et}iva(Qc13!Vd}((^&cK$w1kN_J)T?Ww3gVOW{`Nf(CA-Bv1HN?yf^rTr0>5^Y`V%(^t(C)J6qwumgQ9F z?;NcDFpc??RLXvqng!Q`aObj{|)RQgmGS-1QH(fbm^LwBmlFL!aeE%%OT z>zY`SmM=saV<&TSERDG!IS%QSP4vMvGnjb4hsao{aD&2<5chEb>{s-rq4#}h-m?s1 zqBxU_Cl_&>m=`x^VLz4ga^@YM(_n`CG}xhTPRo=k8IOe*Fhp7m#Ka2O)0_x=ido9- z^k?y|r3MCN>*CghD%_|2^6Z?UHb$*+6^!qA#N=I=O$VE1qD|yUl836WGEN2KbF;ap zA9z0Q-$If#{vZkdo<|G(uQF}P2MJuZ0UIS39NwLVtF$gLHV2lFqBbL_9urOv1RkKD z^wn|MiWoY$`6%p?m7*>G^jY2C&zN-%cd4#dH50m6ld~_O^jV<{@7j&Pusk#Hw@sn- z&E1%EUk+dAXVUy{&lqcV28@VmGiP6(1v$-TVr&ylJv0{(jq0h8`Kp;}*}NkD&r;E- z_zPRTBb#b`JqsTf2JyT2R6JRFn0b20l1MT_G;kb?A%hVpngFcx*eQ5hcvPVNwt&Vo zEyHWK)w%P0A3?M18r(_irZF-SD4Vs5w#K$IWm9u-aZwDu`j7%IcNnm#@=w_d`SGBp zEpGPf@OrA=G>7Uf%*QJoW6c~3i$Fvz1H2rDi8PHOiwIjk+`Wpd=eZEdZAoUc#TT;e zyX8q^dOH;eB!~}QWz1`}$#|`KpfjsOFzwc2@GMe;MXy8Qgz^O{lUfHd^ITz4t{+!C z`3uo*wFa-#Yfb;x%$0WVIOHefxnVBwc`wE)h`tvyzM{$cKWgwe-`oE}Hn@ zG*#U)8M@!;VAF)x7}%!*U28+h_Ns6?ReS8uQ%Y9_DMlUO%g3~I?UGeB6>Z~LwUSx~!!j=o*d%q5I;f*qG-h7|>J(V!KF(n+Fej}{#j3cG|vu@S2 z2bd791YO1uVWy{M(urAQWcys0&GU%NMAg|Huf^g0 zm^NmI{AaQxi}x)Z4`Ua6WLu1AY={@Og(c;~|iE4kf+N zlInswrF?hx(qmrwSqviOS)?QJE=|U{DOi$#ztNcGb&USlQH~;($H9E)JNcZ(cgm-qAlt9pAU{m~=`F)G=+mf(hxfE# z_h(mpyx}ojS~!k-;Tg-^eH2eHYzO&~HUp3T*-JWX7hrW?Ck`GvO?Lkig&(o2VA-!B z)=JipoB2SB>&`XBjh9~F$1yQDG_(vJY|w#Tlgs#i!!0bfT+GdVJ;CgdT^!zzh#>PZ zjt;j}Fz2TpAws%~s8e1u<@_(xx_L*@(_sg8q;F$Rk19g#q%?AE(P_HZ^9@a05+e1l<~r>Dkjgzg(ce_@%J4OGo5xd2o~9lM^l1fv4lGWolJ$U5huD%SDZ^<>4O0* z#Y^fcoa%&DGLm70za_5XH7q2b!^V@)rO2oj$YQytCMKH@9H|tBMLW|mZFwZdTg5`l zwWIWz-T=m_Oeat46RC*vB5Z36hY5?*m?abTz)6n{+~ev0aJV8F-?>XeR-8L%L|h>u zZR0?m_p&;UrVD&CE;4l$<7xNtMACP4lxNe*;BA9+2%Md0a`00} zH`BdMT14N^pJrHGqV7G0oI%VWo;ebUQzT5dq2)ZgW1cn&&WgcjlE5yt(87NQhp<__ znRd>a%tW;$a!U`Iae;N2*wP?@MKc#dVeNMK`(FiTJNS%u%l=_eBT?YHKNb~cRuK`lIMVu$nxr>0itp2l!qmeip!ZgR z4s39O!hKROHE<9ID{LV#Xg+-t8-OYG!rc7<8#E6IWpuCJM#m~mRPjAaLZwBxQ;pMc z{^pq=_SOTO2X9lY_^DX5CV+lFnna}IkUUyG19AmTOo{CvTV@@Fm0}sxzE+YOzTU|6 z2(821qFlN}ww&tNr_zAbYsWdhIOMqQNb{0~8)oZn5&rFMs3p-+tw9P8Z+{t}a6n*9mlcv*;a zJ5Qiwiwv$Az6|nH6R|Ed4L@DI3DN5+NwGx)HTd_O{?%Sew?DhZuKCPihnXH{t^Scn z1(%Y=0Ufk^PobcEFpO~mS&}xHp|t6N-~^XM_k39lr58r1QPMNAGujJQmd$0nqYcUM z0dHIzJP#$J+fn-Q0luFr#fq&JzB->&ppDiXa#8*zX;;d@1eMdw>B~)c%6AKSK6Vpx zPudW3_K0(`O{-Y#ywmu|Wi>!=JlSEva5qo#ImlJ2+_(odoOPoba%>ygTFu5osXg@d z)Jp8{7vpsEuF)47_sF$$Klpvv5SFdm4K_iosIGFGt|;FLH5SHjeOQ52pH)ZhbZ3%o zp1Uw%C>G`=d}X>9x|2A?LGq$3otUIXqIe|7eD>~R6rLT%F&$^gK)n_m{HV$o7$mW3 zmIO}@n#G-Yb%aJtmcj8NZSY{C7Fk-hklVKX873a!JH(dZDAbt+b9sM*+vqiLsW(J1 zK5epmZ#MVFo8XR&m9(v|6sC*o;7*DA*s!*WyL;P$)xoV))HD^<`ikNFXc4`$JrR5- z7E{g1`Ly?&6k0l_!QM|TwA0=96wxt==fyyFr8A_P9qyQZsPYxGC_-V-k2D z+6k>}KB{N8&;~vim7L0-6{bqTT=TPJ%Hu?MlNbQUW_-X&{B!KB|F`nHv0JE?dVh7|Dt}Dnz;`TXV*3zv4?e^zZ3Q%Gj}Cbk zR)G-1YXBSyxbz=I2sO?mur&cS!gvmN&u-Ye_8tw=xlG>oD#Ja6b=+*e=i>AH3c4t7 z0Nbk7u&Vk9E2z7$k8%auovwn zkGfW~!*&y3QJ*={k&_0wcxyD2R>9;z6Z+ZwGHULVy=CXa5;osVgld2Rm_q$=hG;QH{`*|v+!+VFzBa6!u=Q>W*~+gC90r3EOVvk7l%eXc*SaT`i|^*@Y@hBBr!VvgDEevkLi3L~S ztwbWZXGuh7#_3cpxRg{k8i{LofR0tX`D7N0MAvFkVp z+-?)c@k`MmA{l=OwxHp`ej;xFovwL4Op{11JNupvw_RZqRfx+VnlGjCa&9?2*eQwj z9{+*IBOfAtcN4K*AxlsH>*ITG!8kGXH^HYxWMiKcdAn~w;9LEJzV|!KW}n&2Ez&(r zKbPJii*4^=WPk%&`?|UEE1VhtGN!MN@^f*T`Q0F4n zH~u2hg&HtWor2|?u9JVyW9gDn8P0m+2zg)nK`?xDHL&OO;MJROK5L{!m8HY^tj2Qk zS^pvPxmpYK%5CxCw|25D%8rb-eW#5jLs<6t4CV5*S$p>ccr05?Kfc}$2jwYaw&^pC z`BOse7AwQDBWL*zN(xp?I>3xuG#{$Izkrn|vpDOwlX2U56JmZkiJA7q703NOgDa2p zVwJTPsOTEQkd73$rhOBaQhh?( z-VIwuksT?8|sQYzX5rKu#9lxWZI`ThZ(A3WpS_jSGBuh%G^Qf|cU(MN!Gq>(ql z*YWrro@KhQ0L(QKQCjRfRDB-B&A$?0;`9W%WIzhDi}r9=ek4%c+cGF=e1!@JKEuYn zneaO89*u8!#%C*CxZYLXxOu(|CuevYt=5@yA*$v_Nqh!Xv>Lnq zz9EO-{sWUWUr@@&Kj$m?l58&m&=75dJ?oI7UpkHvfR|7yn7Y^i{2 zofZ7=LOX=NS`T^}=`b)d0d_=p(9mOcq+TF|8#jrdyZ04(r)VS7X1kcU3Us)Yk3tdo z%zkU05ATTK8~$RO@OOGUK6;^mvl3R|V6O>0r;W8yxG#Xz!NZg;wI#X6OUUUIQ78vxxUpsoCs@dzFDC@TRo#1d zRqQ`vD)x(Pi<6>n%E!TThpX^o^>21mYAhL)mS9b{O(Yd^Ge~nI16Kt3aNyNJ>c6B8 zWrw5bqCMZ)^%*6M#C$EBXE+`Vzu#nXZ!;d=?*RAgMVR7st=7N(Q-)@<5}N%x zfuBwA`5(UPLSJW6@0B^k-SRq{b`Av{&vv4uKtu4L;5;O?zot`9*MR9QJE;2|3+P#a z^2wj*7d{KPR-u~480mm@~4=W*;ow zyouTLpoC~7W0h@Ej~3>n_)g+}~N-AUaH#{NvCyY_BjET(3i z|Gg@NjNCi`-Q8`dT{D2YPdulYWDC9AT0lIOq_CIDwqeY-Y#RMGRA6DRj_3Sk*o4Co z?Co#ikg`!9-R_CO?{-__mK03SY*4|4c}e7t=_xWwXK=sEV=>y!7(;$4k#KcOJoEb@ zjb1(<7EIhuuIQL@)+g&o*_m+S>uE+GUi``8jTov}U&lP(f1k1V?GIa@B~ktCJut!K z4i@y>B}X#$g5}fz>TR0^8cZtO+Zc`qebSlkTN)()#6A3y9SfS#&)H(RQgZCkS@B%q?HW9QAG|;=@mN+l*4BjtX2xbO9iFraAp$`mL z?MW@n>*!C+%91qBGOU{V=`><>j+}I7&a)ch? zGr2uPv7sFPMl0Azvl6>FT%_th0q2S$0a&hH2 za`|Eznxy5Cu$`u)^!YGaOPj&;{c+5nh_6?mRcxINuX~8URnWWCw|aJzf;J}6TM^~|9AUu zSE9oJ?}=-a2EV6rR6aQa3n~-w$nHeYoKLV!xflIcg+hPoGI;)svQZJi@c8=_5SbTA z{>4}^<%ns);aznlxd*9~ZFa3iDsYeSX&6&ZpSTuycbd zSJPC4mVXKa-DX~7$J-=aTqKQ;Z4cr0kx{a^SC^hTtjggvXX}#tx9M4(T)6xC5K8pa z;GA2Qn13}G4}9?h2WkyLGi#XmBSDOV{aJROYXbYZh9z>@Ur5(~GQhGi_+5J%yDzw& zcusr4J5luMXTva@_E?IJNu5qQ<|g2PekXf2ua^!QJ$FLW`_FzdI;hBAj#xNH9x zit04ro>>Qe-!ld$21UV6$=i%{xhze|(?u1PboTDw?<7HTgsn}PL>Z}6x}xI=F6@^C zr@6Dibh?n55`IbyWPARz?3lP@RNH*D4 zK)>BCZlkLk{ymV)o?R%w#{+z3(MOEU42a^|n%ii}LKC{K#Dt$`-J?AUhv!8t62%8kUeSip z7CInlLA(3IsO0e)kk~MVo_HR@%m4DN4fM9;_^ z!dj?uN2afaZFV}C{P6?H9}|n$Ub*6@?uXRFq!u@%Im34eW$ZYq1*y;ViNo&Mc&yAA zYlil;+^jmz+@=Spkbwkb&(p@^mPz=g`6)dgoCnKS15+XhAQiQ5$YS0-jd?T z#IGj-k1n$h(!Wy4j7zZa;vx35&t9Swd>y!3U+9S?u9y(32WMXYAu&5AL-X)8YSg3z zP}D-VezygmJu&QlOkj`y3BsA(XUWX}&cNfTX}Ctmnix+B0Bf&>gb!Exm=5HH%hmaqSL@d{DZUj z8qrIdXMPhh`R^{{4)CZW1#&KKB4>MJK^lGB0JhbNkU5@Q*iw$pJ}k8KBZNRPlXW}5(8eI|Dm6Bz4gfgot31$`nc z&9Sr7a7)z}d}4EoW}PjZdm$~8o_S~uhYDh?YxBd%IftX{>pD5~vi&TWS~`S%!~S%B z?gcg{u?D4MXR|FQXVYu<#MwK7)og`q4h>06rPIXya9?939yW6zojmg-%eaZm+kSwY zc_d95-Z-;v3uh56^Bv3&eh<{S@gw6g?hsp=DucU26e0Hd(UuF9HqajBN2BL|rP{Bp zV10HZCz8DoH!NL<*UngQ#n(R3vkENh;Cu_mj^zGEoK+T@n~!mz!j>og#Q&y~>UZD8p_SJ-SnVDW3@w1->W3aVeH zk{Q8mq$4AM^JiZQG;}J!QQ8y>J0+>X0ylhmRGu?BwGDTrEJPcHP`G>fG1Go6ky#zE znJZqLArN17fozTM!%bHTXyafp-jdzQRX$I|tA{gjtL9F0KWL1Dd*;!hGCY6FP?u^= z&S5K#h=XqWKQd;&EqSwF7x-Hz@d{2N&qRF4vSTJ}(jRGfy)uJj^e-h!VouoS{Ltz> zI}QzuR$Eu`{G&=mo(+^62N~uQxD&^h(Vo_QP~8%Y*76co0V;lMfM_+54REJ9f12su zm9AWFk38Pl8A?~fe7e+TC0Q$Lg5e@ZP`y8&jI1f5QN0S>HO;p;c7qsauC;XQEW;62FK^tkrU&e61%u( z)*8=s!E^3$YHL3YWg2Ip)5pC`(Zg9Ja(OCqG$G+1W2oGX47zl#6Zbh; z2H- z$k>^9>9`;58+l2TJNyuiN}&9&$8=wHDI@*u1Ut36PjK>)4)(uqCA;VEVq=?{a8u$I z#xZ6(8h6B_&eyfLuc8}|>sw&6Zw1+3oJLP9SHUt8fsGPt@amF6a%5Q((`RJOtel?5 z?8%B`o_w&yF+UAaT=XDlPh^A zUFK>WH1@*j)hAIaF9LhkoaQ_Cp3J+tiyA2z7O6Ea;UgJ1tE@*~3-ly=O;*e}}T z$~{SFIU$Vku2YddO#`)nH$ zc!f+ASBCb^o$yZVJu_B=&xITkvUxRy3f5WQ2laR5%&M|MTH$F9lOl&mh{-HIUtkE> z915r1w-D8Tllk7wQf^P^dD&NXby$z=w{ZA0I(r?e?p1MkkS$EBaWa8NUi6-yr^8D?^jI!=P<#2uvc z_#kF70qD9siEDYhjx*exhuxpvGukav?g_`pY_~M>;AuFh`M3}#t4%a}#1k*G`E>vMYMv8OMDj@{E1zTzNhi;+ zZcT%1U+)3ZBwIkc5_#03s1&qq-4FXMrJz4b6^&eTaBGMxS6EcSeLvia7mFpKr^6i< zEbpWaGuOd_u@X?rHqu+qWq9A~VzS!Knhxl8QPqZt#B!k(23g7Bm*;Ani1Ts`w7kn? zy0Tb1x{mikc;RxpYq+qpi(K;>2OF(=h{0<`&UZIQ(+4!+&fs>uop6w$;&Qa3AQgVR z+es?s2GHgO?)YZ66n2i9z@DKVcvN-(CXZExz2q3pDGVo1rgL;@xf$Mh;Kl~#yVLFS zGVsm8Y@Ay72`@(fU}6e)!i`WV?x>;-H?{N}-sv&HxJ$wao06FBE#_caHbhhkasi$c zf%f8TcKPH%CiJ~MMC|Z_la|tMnb`Vc9PbT)$VlXn5 zWtts-V1#HU4IW6sN}*^J;rs17Zfh~-=Q2sh+d1@G-y6E7QxkP^^HGa-QpJ_0*w%AT z=*>KH&{P51tF#=1*Sw<}BPL*SxfB_E*~L~|SLJuAIk>;&F@3s0o#@#ng2Lx0(&aG? z!jB6CoSq_`LE5anOr~Sm!kuK|3{ULJJULfc>=ca+lfw(&j$qU-18b< zkrr7v(hup!iTlzFvYIDG`CYt77L5H!TI@|>!{rI&vC&zg_=)#XNG)boHtvUX-EPu3 zT*Dc@zk_z8I=J?P7v7MT!k<3UaMfp&ZTaz-tkd>k=j2iheQ|(#RPMk(r zVLbdRwCA0{+Bi7&D_;M49W5s zr=DvH1eYZem`W9K?t|uCCVD|UWWQZY`d(^*SWqh5IM7Ev<9U)%n1cab4j6LEi7_f@ z!OwiJ^2~xFy6kBVxxV%`7Cf{fW_l-Jxm7hT+Rpb~Bo1*O&UlimO%F&?wJ7Xz+l+lb zH7Og+Ge&Z};JmIQ^{+D|g{z~`-ZhW3-fg4repQl4^>Wg%VhQT0o@@D2I}2;mN3fjl zy_A;mT#pO_*=sp^xhfW=?EUeKJb~ZcCSYuTmHyq4&ee&;q2W(ST%tT4PJ;_;vgQEh zN*W-jm}8vWOWw=lg|Fue=y~})+(-F2=(tLio8xy5ABVT#oWQk^EZ$4H!X|L$;=l0k zghUKokjFmcc~|#g4Q4tN(;A-nD46_>95Cjy7{Yw+civOpom@_btB1*`LkT@SE(*qE z^%Kz*zZpw|Fm`Ra3;fMgBa3HCgSJW^>%3kJnr{-2YgA`niYak*3cHE&QVB3;?$bkY z{Qr9EApOR(5iPV^So;gYtbex;$XtxXO#aUA#iW2<(rQLd_&LvYPvE{Ly+U_MJG}W_ z7DLYoF!J*!QuM}~J2MtQXvR%GNBf=lyROG7$LlTXw~i8>@$UTjRhpPSRDKPI8KuhZxqQFxW! zBV_3j)}r1G{@ni$L|RVK<+tLYwabQ@GCHuJT8P@G|6r&1tROGv)}fa72(=0;VoCzW zp@xGkTyB)GUb^8v!+JcYc~iRBv~@02+RY!!^}>T1wXe?xSLCc~|o zMjIEI3-)Da6FsYX+z@jPO@D79FAFjV2K>Z@e9<)MV-b1u-~l;jI2pD&PhqEf%z^Ku z17F!3Z@IYq2(2(svpIGtkQ`c723sz_VCB73Kp-89bv(EJQmHG3Yp%r4cg0}3`&qWp zVJqA^Bnhp4H|Rf&4QMv*D9rBpMBa`Q2e!Z!Yz}QEIY$?vKanABt=E{vTJxCrQgiZX z{slI1TQrmU`3east`uzB+)s`SNnqrb+2rla_w?adMOH{AjSRmlqkHFuGMBu0zSc7X zLE8C2VwrcBa{Rt8;A}aK@u{Z{k@iga{fSun>;>``oJQyYxX?dw&!6iZcXth0XNr?;J+><}X@sUkG-_uf}q# zX>fMEEUujA$c(1U!-3v2)ZJ$}PSkdSx{o{~d~G|t=ew4##0qfEuS+a1Q*8XS?`L}1c&$)_&S8oo|oie6MYJi`z&GpOG8MX@Eg_kYT}0GMwC4a zaJS?*z3a3BR{aQr)JY8x(5;I9?a3ggvX+BuNgp}tz6CqB1ri~pVJa{lLEpy}bf&BU zx3l&cc+=0qZ7vuD*u2V0M!_Ya~luXo- zW<=|AvE^PKd$v9ZJta#}_^%Hbm6$`p~I_EYPS4@PZT_vR9fgH20UxE8wDuaPahuDTa z@6a2ckniymL6@6I+C@9m#l7BA-gZsba8|yT&IAxb?x}FR0pFD#F5>?1<+>^i7V#| zaLB6;S1Z^+k>(yN-^;FKSjQBm=cmv!TA{F`%$vk5RRyzirT)?n4GXC!#SEH2#r zE{Q6h1WV5-;+|>mS?k3=1))uL?kKEKF*{weDC7x zX_eH)>pUiC<#QI+tGKtF2XNw82~@BBCb(BrM|8LBAmOl$+V7hPnZb5oSlvRBiWP18 z+Wl~2^Az@j@DFF<^GnN$Nt( z(p3ZxlqP_7i~_j&^w75-Uom4k6miuc?>|*-g0T~Nth9Dp66L6+ICT9!>#y5RgG4XU z*nl#~N~tF^Os|mExr%hfUUN>low=>%%jb z6+dc$TKGild$}2c(rRGF z=j@gF%!bSWdRO=heLty=DjAz3ZlPAJT*{Hd>ju9FgKC_6PxU>j-_p-#Z=;o@CzT$)ZAQCCw<} zGwTuNG`jo}j#+F+;=cy4&lXMu`7`IRtG$z~A2Gm;iM{OH0cAFl-x*(!*o&L)8)HkF zEWKjPUl9T!t6$!U4&a?fO~A#SfDip06|4)yWe{t^0}>5#{n2l-@Ir^Uh30!QpcBXgS0^f9H?3o8Qy;)(~>*$TYe(NE7qD40vaT z3paWp5gTW?vfcq=5S1woy|J;>%}p1jM=RheKWl%Ppg`k_`M%rTGU}gtoF<=HNff6r z@HmPi*~pwTFsiReT5h`1A?IYl=olRkdr(Ze@1KW3V@bTI@RC$a6oPTT$MN^_Ec)c4 z5H&b-ig$G)nZ|n9owdd?Ei%@R)XgGKD-b5~Bnhmk}t#=U~ENBMjd5>wxe7+fdLx7)1^=a|5SY+3m0b_P4-R+2%n#qdN{ zgR zlp*>-b$0!%914zmVq=%gA&8q(B=xtEHpbFOo^dIz6lhF+ro+kt28BAjB=pjW~L zoS&5Al!ho~%j9P$$XY}_-ruL*Q@ZK$1*h=)#U#BUzseJIT6@_jUm<*OPK7%rBLXLwIk4mGLK3PJgx1TJ(<#;B;BoI4 zNy~anjHQ%d=!rU={K1dWJ31N4?gZmQpJIB%?f~We=z@rxnhATALN^pI?#{qaA?`eDsHg6_zANpsJg?BmjTd@TlU64t| z!^6;|U5Gf=ePJ#+)>3yi1OBLOLbD(GDC(U7)#BTD#^E)PPt?VUHlt+YlqS0VRvK|X zI2Sn67lIYUpG>**&Z?!5?;Av~f*?&>Vq9h|__JIJH~wcrYzxb%E0x0{j}mfWQ48%n zv4P#RBbKo@Jwr0IjmbmDcp@ir8>5WHIS*@9Zem3*Y0{BL@JhEbbo)x*TPJ|$mmB1_ zpB#+QenY!%p2DKXgEY+F2<$5Q>9Z3XQTtXU&9_|wFD7q=h`6(i%9nU+>n)bJ{=rM) zJAasljX6aoz5b3tuRU;omIYqxxlTvrvSPD!4`a@bI$4zhkdOu;LPCE zd~MG5mpf-*P|8$hRKa(t^JMz{Qaq~PiA#RhlCk!6yic^S#VhZ^-1=<^v|JsnE7}6+ zy6+P~qwgf2H0(wrFBuTeZYR4xzr}9?NzgrRMVW-lXgT%i; zqRXfrXa8S29y<0EQ}~XMP?8GSJ7p@nw!4jtPRu0F9vpW#_zNt6Uh;lbmE z{5*Ftws(o6v+iRO{$e4#))V8t@8p>{FB!bjs0M|_JOlgwL0nQV362tc>2uL(`uIgD zd(^9(ReZ zGmCp>zXn&;sl!Ht36MLxgGf4SzlZ)pD$0_pj zG6gzDNP+8Ec%1oGF@sAl7{PzaCZJ#*jg7@^)W*n=+Z{K==-yF*$(K`b!+{6%LLSFD z=nv7o>uLpSo*1&nY6`J>O(p%wx`BV#V|x6dCoxVx%Z7KV3Fb=o(KBoLjrFf4ep}fp-(R8R~Fx^ftN9 z#zNU7Q8cbk!WTA|>90t6*t7Hw{kUxl&PmP&ud%0?315vtO#Lg0KM|rvXC8oI#uVQ3 z+e%K{4kxEf7qi=UxItrlIxJVXMHR+dlS`Mj!$R{6JXL>~hFVy1^1fy0G-eUi9TNlh ztv-;c{eOtf98nr0CyX+yWSG~9Kgi0{>bQUJ3o7_In|835$*V`w=yUBR#NNu~w%1nS zI$c>CrcDk#Ld}Wtg<}x2QVskE+L+7j+03ndn$YiaiDms)lOLCrApEo&7D-%$-9Ap- zUGrCT)%-~yAGU*y*?SIj-loCOys=#Ss~fobssy)bcRoyB)=2m0C6U9z8n~%V40U%| zLHR=ojsymC>mDWHw!msURJYjDyChw{}>cSbRJDlW=}H;dznSli>cO6BIYvWoJr=z2;T>V`ZOFEd zg%m8tLF=)0W?Z*D>$^dc*z;WMQ`eWHy~8!U&_97|c2eXzZcgA{O0^J?esBEj>IC5} z#oTHyX^u7zkrRoF>GPMD7_Xfh*aODVIB5#M-;+t<^bP%)e&Z=LKmQjRB`M(K_0`no zbsP+QcViD|9Ve>eRBfz{c^9;~B%5C;jMtt;aZ~i7F`T^&4r=LSc$9&4{l#!0U@U1$ z%b;$CQ#tvmiA%e%DM8@R;6l}mBTA+0-%$jtC8uqv#i z|6DmXxw8rv*U#m5yTj!1@)PhQUZ0E<{>O8`vmxE5+^X|g5q7x7V*T}7v@azYKdg&_ zZ)vgEv+_M&KMMI<>|3Ptry1VZTw;L(2s?bFI>&D|y| zr6^O#j9iSHYZa-BYBaa_#zj`bOdefqC^_WbC^)mHhWLEmkE`wVsl%IjXj!0+#lhKZ zu-!5YaGr~=-%HVJ+LrW=(J6SbJ)O*O7l*UL7C8S%9UfPGPrtq13)1~3v1jQl^gA>T zG-_u;&9kHA*f|$e>MbKNsr{tX@+>WW>Vz&-K&yJTz*akV+&MXtj<-|C#@T1BKaq!| z?!r^{bCxl@8J|kxPv&CorpvU=^cZbyX@;L)jEL4v59{N37a8N36F^Hr6m65TQA8o0 zbNq7zZ_Q6f%jUVXxm*n>RvW;E;mNf1_Jaq{j%>J{TcFMM=-;|L(6C;CbXB~Md9g~VVr;(p3fjMrkRKc zGMF1YH*_0LX6oEFvSDjl$#&)!OnD-Txr+rjZRZ8t_jERHINd>{KA$0Ge9!S{P80cU zUc%05xs8sC+Gv-MD0fXU2HcX3G4yXWz6u^BAKfP~L}M#DPOhTY6LiSo-w_(cLd^P6ZH z_u`X-yyqrq9QQ5cJUz0FcTlXJj4{8O&{n0B3Khoy?+2voW$%KrRRHO&4JFHXX5*KE z4BngeiXI-fi}Wty0qg%>Q+0<-rft*<>um0E)tzcwoL>VjnuDm{TtlOl8ADL1EIT&5 z87IfACrLv!aQ80zBXXy@eK35gT zpA)xPfNGZrXQ6(a?oZ*F{3AQ@rT0`UoK#FQPUWHJX>Dwmd`2HE_zmaFV$gcIEp}ha z<*o#$Fe%q&10JLx=H`o8AIFoc!J44)el1>}DMBB%>v6F%gP6|0_m=;*Lh*tgxN|%W zpOhWpx^@@hlclmued9flseFPO_l};Aa5S^oxHte5p4pH~?0ZhQ#nH~#&UJWB%FcsE(q0IuG53?II_ zMIHUc;g)U)yb#xiOOmZ3;ewCl5}$on zsk_p9>mc7P&^$L7XXQKqivGBZ&-70%w%~hmzv+rO4AUko$uo4+IOC9Ef|h!8|D>B( zuY;V~3QE)Y`PG}zb_{tV#f`Y#vuEl!i#gt0 zNu?s!cw!&>PjNA(e4LHmRa4=r%2X~ZCxWwjJ_BSj_}$yUD*Tu03x|8``7XynU=o_B z&+}pQP_5un*Ipubx)!pB?|&d0=5{kvx|QI!P9%MfI^4sMH1wH~!EK(^#n^lCzVXeE zF|t2}{*5_Kod*umM6)LF5QvZqJxMsR?jef#UEzH`0o+Z~3TWY-y%jw+IBts=$jqEb z=Im>O4S8~;x{;7#vE$^?P8W7c;6zOA-GSvTt~l?^OZLBFS1LQU7_TiDAz`i8$XkIeut9r*CTn*+ZRXgd@{f#ewy(5Ry4Nkn8n?go2M@31Pkpw=Ijv^{wk zQ(SxBg^N?Ofr{V>yyyHS`)s~Bb2WPs{|qf4t-*ElM&VJSeBvKTlRCl9)4NCaI9TDP zNEvcYTpl$3ykfVqulO^MGx-s^5tdsN<1v?P_UGskIGnP9zXQCr{&{da`S_(zutN75 z{i@bO7e0D}C!2P_!*5pbsd0b|bDx<`vnX@%SN*icFogzXKfR}8sFd-nOgi(w~Ra)nLuTYf$b2xAP9Ym8gpBM_SXL`cj-1IviBbu!Fyja5aTcNX zw5{-|A-^X%F-mQ2? ze(Ve!UC~4)-v7dOtTti3o7Cav&PLkIGjt6#|B|J$=_IDVnI!#~&O5WdvU06!>8F;l zm}K&nc4?{eJXQWsy+9e3UoL_rb+c`xHZMfWN0V*tFWyS*gs+hL&RNiYbqZ0zH#(nC~S@W-HHygqU9>b7uq`d>u>TSRi9&o+GYR7AW}q zoNl-`1Us(SpuNo>GWMB*O{vRl*!1oYJsXmYZjrf2Mry3X$LGV9%tQ!VDu?>UZ>h9Y z6sxZNoR#pAhiZWXOzxTh-|LUT&R#z-n;Q*%4hditRnJJ>(u0yk8gNap4b9!>QR}RD z`uUwCHdkM#3B_AM`@IAvp5z^vqAQtOVNS*dIZ zC1u)0RK!c09z5VgolKrHLQ4){k)taf+4Tx%Hx1F=ow<~W-p6c7w;@Y3XTZG$C+X4Q zKI*PJm&R(QGh1;#O!@bjtlCscFWmKnu3;H=PoSU8=iK9H(4_-|H=|&`U?QCQ5eGLn zDA>duI!LYG%VO-PDEBa^glwJPL+#xwY0!0{=E^8t%1x2MKOupPQp6(qOxFjE_6{9zyBK0xOs#(93t-yBPV3Kwi@-;^i>~cAz{f6m^m5{}xqq%KLLaBcbgkb* zqE!BYsh47yS(_K*sqi5>-Zqxm`cs0~Ea<{YrMFZ^&zQ8gGB_>crZoiS(3P2UF;tIZ ztQWgueQOQ2{+U6_I&DbV;WlOgpSO4tt0HI?)5gMe=~OGi5&N%}lWi}auyLQZz=wo9 z2CQAU;EZX+xPF-Q&&wwHNqXeu-EA0c-Ae@z+pQK|(_wY`gz-j1D{)#8Lw%o)VSg&| z9D%4g8!yBIM}3L^>0oH|Pqf}` znIkwcOB7}$6rx%1Mqtjg(U1mfJo4b6;8wXRF_u4#VLp!3YV!gbxzCjZjtDXBmnQJM zSp}S#EeFq}WqHP+3J51`#|o=j?6^CgZc*1JSM`0^;j)KZNM|~8r9v4C!VQ@@ldiDg z3+HmpttCWeS_h*yHV$mm)^Z`cOlfcRZpNm?lRNhy886$sqse?0uW)+>yU;xb-6w8E z&v1uxUxQ@e$rFAy@=_M|CR9NUKOULRWl>>~Ly%>82tIL&WZp(On009l{5o)&{=dMScE;0t<%gKTpSE!IxHX|qZk9=M6 zotbHrKz^rM@t(qEboK-x7&zaKiq6%{6o;*7G?0u6?|0J~{}oWld%UwcK9Od$enk7< zY0NgEPI7=h>%@dbkp}&x;I`ruzDfPUxcc@oCk?EaWqGei-`UC5`?JPzFZi=ze69nDWAm>lUf_wxahTcn=e6b2T zD{>t5->!yU*TQk|RT+PVwdS5h@8EgkXNg$09@<{JN~~*(1?=nuSZ#k84=*>x^=tC+ z^q(AVTYn*Ua)u*U;CUU?KTe|i!(WmI$NN!xa2npUm4Rhyg{a~)mPDQt#bgI{PB(8Z zu07aC*Hv932^;j7qjeLR)i?ZrRkDJPwL0XTau)7MosTDdhuQEs-)V-26uK0L)643f zG$QCCBekW7Y}&3yp12)i-=}(md2v1U7mH(t=LC?_40X`Y_HKzclLb?0V_I-6pDry9 zf}rKA;LhiFtd?J!VDWw{F#j@-Ry|Du?SOkEgZCqA)T~C=JI3R=*IOXMYcZ&tIYsWAQi9k+8^CnM zGvcWGp8PZw$9ab?V$)hpsuWsGq&M3!Ne}w0cmFh?It#*ZEkD}|eO-d=jlZn+m?~?* zlrglYxD@x~Ekoh>1pMJudw$p=5gsNalSs)JFg$XEG~ITAp^+Elu1R1j4<(C2T{{ zVY=ZXaHZKVNo~>&a=*hK93P8t`p;evGp}{9dzU+B-!%re_*i3%Sq(W^|DJT~X3@Lr ztJs-4>Zpe8G3yn+;kYGSSnv*pXsmxR`>CvwUX-~`rPj+q(}8BDxpfbW9yi3DH`n94 zk!9Qg^Fxqp_Jn-8W^3;>F*(8iHhj!qu(MtTGsY<;aO5vmBTdXVK&%KR#uyejW(SBJ% zI$Vo!NNp0`yEO-ntjUFavTf|JusTYwItu-2qAcq)i=BEZ1|(m9W#hegZ^j*OGQ=}E z)^LBBnoTu$Z`VuGadWR=e$zYnDLx;TZfzjV(UvfzWP&>-d&%p^+OYA)c2r*9M#*3d z+34vGUTd$>lbJ8*=Tj3n?M+*_sa^S;Z_sRVEo48v__+|rRCXW{kHfugEY9dCL!In& zy0f$#s$mXRS&QKM&{*n0F5&Z;iDbOEXiH>72>Pa|<0)2_li!mM3WwC7NM7D15k}~^ zwlTO=^#$4PREIzQCeSRk4t!enKZ?%7AFKZjv=uTIrnv4?{^9QG3WPjefuzV zY#Ui1CP%&<5a5d?Ht zP8DSw`Nspgj#m-uo)Xd);Y()otl+9xGq&pOahfW_@^`;nb~-Mk{>nbIU`!5#)=Yp9 zkI&3>2{qA>^8wC?zqgv6aG@&>8**R2%|z>kN}TbF(VXt;+0@mynb~TvjYM}wLXgb_ zsCzvhD-O?qeSWVgwLML}_9=<(R_McG)48xXSOFE@dUCUrl$kqsHqpun=a^eX**LiB zI%A}6Lz%KHGz*zXY1b3FO@-g-UA{&R&a(jx;JLZWvIV6D)pW^&rS!+rI-2HaO>|<^ zXqH9-rZ%^ur-U;0t<9sG4*sP+BRp`KHt&>v^o_(NjDoeMVX$V`9#}jukJ)WtO)mPV zp?{|*>buS1hTLVj8?ODR8mL3h=^lo;=5wk0dKVZwaV%<8@R>HzgQxXg?3B}M(Gdd4 z^EXeivv(Bt^IRleTNuyu@~phERD%sy;^7JJT

3N9d?yKQPAdGU~q1r(07ZscL@) zJiP8qZt=W3$vaB8=gc_f_dRV~cWE~rb7@F0|EU~)x}gB0@|57%yi_td>MHLC=Ce~O zf)mNAoNbL5cI9hwM`w3~hj{{)@?5Qqx>aQE&E3q>{ziHr{~``7ai$8(R51P5Y4S5+ z6LDOY!aSH+N>*p=qwS-m@n-EYQf@s`U~r?B$iEIin`2VAzp@_xTOZ38{f#kaK{Q&; zI>SsGf1Ns}9R=f?zPNAM31Z#RN^9=!!KodF4yxvd(C5bmrfsDrC>-AfeSfZF?^81_ zE8LWu;=PORocEMyw$?FkB%=B5j~X7*2CmeX=a${AL}UBaR5LUO)^Ro@e8N6>BH6+` z*c0d&RcL|n`{m%J$30Twzl5O{uW(p;4La?QN1;5=TI7dKZN5oZSY-yY1@}p0y&~G~ z3nL{=8S`sY6}jgS$VmNGroxeviA$wB_p>t)_ev-dtxQ+wi&PNgWhJ46w;xXZR>s^I z23Ycq=elf}%Wd(xh7mFRcKD$_d!+;e>d;MsJAVsvP_=)Ij|3}IseD3X;Fft6Bn7uQwVJn z@~*WnhncQw37pww!)^5WL@VGq<*Eu$PwfUW^G2byh9^9XYlD3ryXpSHGt8@sF(_+U zz-lj?LnPZClHK2b5Q80-5b)5M^q%&h1(|OI)BJW*wTDNsZv9J=;p#xzH%h~*QaxDx zXem5fsR&=^J{QQWi6w~xMMPTd2)?=_gHAl_2Zy`p(z`|w_f4C$$c^HR-#o-0D>Wdv z*^=*VT%x_9BXLZW7IQ;8yKvF`IB@w_MKh9$$YP~tcH}xs*groSWPJ2-$4WU^ z^msb8|E+{t-Db@Ep^2okQrppGs}{cgI*h?d1lu?HQst-~svt27hko<-(0dl5(VKQL zBc?}EqjQFMV4wo?)el14xG};Iv+XEu7(>T7M4;-;Thzy9fZ4vQfdmQ1IU1gK!TrVW z@v~GpO)`*W2Bm|!rHe11_0mH2{W1q=FL_Qg=Q?0=n;184Xbd^2s|>=J6`cI~e;Ao` z6Az!Y<6bT6LXTO_SRgYVSKl8`+NWegvd&|i!C1rm(H^w9#*N&92u{Rz)z-H7;&W|F z?%<+-XqNVkmflf@^4hIRmrQr_rW?WZc}iK=AIQCb|4J z4m(sF68}6WmeJ!e`)VJOnnav5)Q+cZUyKk4Qxy> z8V5||W=l5GT{{NJyqV_Up*R}9v}IAbT#ooJHb(2^CM12;U(#L^MEut6;J<^4>5)`_ z8r)%tXBIf)v*l(DtGepZ!X=kxiq$a1%2700yb(1AZ(N+YF zwI9sF;DQ{y)7ph*RmHUSL^`SVw8yi13~<$k2z+Cii^pq+kf~6?FGdgOrt1n|*;$8r zPqJXjpc#&~mce7^zvEmhNvvF%gK<}^xp@LdQY!n6iOpSriEWlpdWbUCj&k(O$r1GN z=twrOcQkR_qJ#nbyGcN!2rG`>B2BprIoEfETAjFrW;=Ds;>im^M`IiN$TWzV6LE@; z-}MMr%;D&TtXSf=@GuitmrbMXV=<_C0avzRCQb9|C*}5s*$dU@=&7-@xZGduFz&f6 zs;zG$)y8gI&$s^^mX5qboP*clsnCnWY}gn*(~=?0R8I84qnQ{ieu6XhhS1>Wn&{@E z1*TyNK)>fT?jdPnpPx6ffL`m6PY`~$TFTw z_Vzqgw;87X0ou6WlmaG+@%vUWRd~C2JjiGKCW%ssRQT^0`{UkuQP##N2>)SCeP-y> zsgFi*l|Rk#&`Lty@W1il5=A6>TnSnCBa>tVkK}&3YT~vKcdR}VPJP_>VMiz5yB@N~ z(+;}$Qf|9oV(}_!+A@bO^697NA~Prtr6&&7THN7gMLOckZdUtwI>~E@gl^e(R$fVx znl{8?RCgL$SZ|CGT{+=?L7c*byDe*&z>axRyG*EJIU6p^xzhB*TeZkreIa|jQx98 z0HZcKFBVf;YV(q6hoQWvT)z71E+8EXOoZC zl3#l>LHI3(e>N^=zWRj{_cM*`wEwc1bzNgnIK>h@&%dE+>$k)BxjoG9!Rw?sZ7yba zs!Y3XyawLn>tb!A0kPx3E$W+0@RmZ0;8tT9dC>EnY1*Al7i{}NCC+)itF z;-rf%dSj^RGX)SC>T%Vk;f}g}EcqZ&Kwe*nBu_>!flI{}T)fX+%HBb`+J;5LP#ajG zSqU?y#*x`m&(q)IZWF^BFKDs(F>?QJ7p`cZ!EG1RQSG)2{VMYT2;PR&mO#|_0|@*i0|bbAvmIpv1~OM8g; zPkA&iK1Xb=Z;*LKN{}aU2e&sYVt!lu5XJGgM3eRk=xfvIbo`?xe808`gIlJOarRCq zR=yr@ZZO4ZQ&vLqv5~B0(HHuBc0aPA<2nB`iYT==jQJ|Ng-)k$Nkq~o#-S@ol)SQ( zePKF8o}Nm8Yuh|F|a!^}oWl?XzK&B#nf|x-*0ie;3m^>!tZz z)QC7eJ`UsVQ}%_TJ9_(e!;W8f`0Pj^95eSBEU7t(A9cd<-ac`rx=xb|pR3QQ@7=*? zgko?oe=_rEM<=uAQzf{pq~O7KJQ^1SX$ez>3M9;3+W(yJxuxFS+ILUGqjv-{eQ%jxZr2 zuWV?${e^kZu7Q)rxPamCaTpTj;Pzj$iNP}+_S}J+^rws`I+syKY1CQPlTjw+-x*<` z$zQmpTLH{nCs^Hjl=@Uz2!;GJ$%6$)QN17*tAp}k+)-;HZu%UW$E;x16qms!mvo{g zOvBETr|G>1^1|s(639!`sMtkKa%DAzFBc2}A8=%oe-tCuJVrQrwyLn<5`en23%0kY zvs=7Z;^&MSQZI&(*ci*sZVP0*>=r}S>I4$!w454Uy2r?xy#*EfIJj0dSD4q-2M6-g z$-Fh0LYXXC6p1n5d(>FCNp%Dp7~%pky-~u$c!TdVRWd!&5wOfR8MsUJ%y7TZ7c!zoGIA2^#eo{AAuJAParwVNT4951YWz=(|LOB4reN*POmg5wRc!b5fKa4y+}99B*i#w2NR_h-x}XC`mQTKUJM?Pw5`N)|wG z`zH9v^C`P7BJ`Jy79J1Ggy(MCsWABz*4-ANk6i&ID3-&^S1q`jcPbQp7BQd47m?2F zRH57mH4J}$oaas_&}hjmpf*zx%)Mvf^I2_Z5|+Z}fHCl4C-3;kyaDDlXQ}H(2Rt?^ zAJZKlidviZfIrs_S<8R1PJC*s-jYkbXBLp#8DgAw?Pm11s>IR&mQ4KB$~NgLMYjqHUFIl2(V_d>-}&9Q67N|0(_4bc~`GntKX5VfunHvGXsvJixiY1{}HQ>~tRS>+{1cRq%kW=dp!}97!#N^Hi2u@6;d5>Kn>rFC! z^i&b`s5{fz79B`5$EwlhZbuL)8XlwTvX{8?uN-!uI?%C3EwgTt!(6JgGDkX zt-Vi2yzD^NCTI5U{v0N*<04M*li@kfzVvGAG2UzSNVJgOR7J`q(W#;#W`Ut8J*hMv zIyO$_q{WnR2YV4O!73`9o{yu+XWH0KS z{Kf&QwEhcuRQjB5FIz0!lTg=ReHlb%kV8Ydm{#XSu*`y%<&*v7~EL zJvJ;V!M_<&aDA9S@$d1lSY8X1_Sdpkq*qb*H~hBjLl&oe=PNm}$riSp48yETd%^> z>>-mgwz1D!zmmJR73g4<0yu9>hgR{i+}=^8nBe|^S+cVles?#s1AkLFCO?@*a%@tT=Xs?D$~K$(;C0B>n%A z>q)nWu8B4{Uu$C93#?(1TR8m7{K347_(ys^>|qAm)v>jG8kBr41>dDV$dxnGu*X)4 z+EpabyOS278QYE9XUovzqs(a6!DS%+8mU&P8vaiIM^>C1q{jxPG|m?0g3R9ndZ{&& zA&X?8OyxRt_%|1hrJsPh5B((L(lXp7ejFC0ds@m=aH@ zN_kJUKI(F_eyxSOelwt=|0nKw!85WmpNq0Tp8<(YnUIyANhbZugXa^DU-E^vFj|d&YKh1S4Omg+Y;{aYlzemZ)99@F|nH zD`AeD?OH3?FIz=oW$Z}izFO3HltKDxqsdvv7IfL`4bDdCkfLmg&6ig~$&*y7K|kYA zsV9o}#^Kl5ZFu*AB1SlQlQsPpsF-pRX?ILu)pPeZis_zVdu1drGcCgL-Igm*eQO7V zzS@KRHA^sJO&VBUct&bB2&qT2HrV;iCHYao*ma7}N8guP2}?1wG5 z`oA<%(H%xll#CbdQ=AL4#yq1NeyaeJ=uI?NNpXRbZ8dufQ(l@N~1en5vq)=j;Z{E8yu1@E4RR3qFAr;p92 zLpxs)>lG6OS~tXm7aq%?_NY@N35+X`(bC!3pdB)W z(|^h0B=rDjeJO)IVVTr2CIvzz9Z68|W|)6;94F?VOAN2{)X#3zG_1s` z!~CYhRTBo5Pa?%yQy}SZH#A7Uq;3g(PJeD3xxBfL4c=gZIWu(N*oQdC-g5*>N?(wf zx|^9>3p#1E@oaX=+Do*#{{(%^_YDKysDR}D$*^~VobZU0JS@4U50O$fa4rOB?0YG$ ztL!}P|7J-`Zl{9!l0Zz^FdjF~4rk1cpMs#q7_yo7OizB_2{A5mWWw0VB#2I<|IH~N z|6bWay!8blw@ixL;3LNQI-CY$R|_Z`tff}Uxx{AC3wD3bD~#Lni9P@9CF88^&Sh8p zhhpi=VBu0d(t4{5Oh$*{h~h>X#BFD0RJ(&zS|+e7v+>U=MfzK{mp&Mi2%3jGko)_I z=Oy@pCR51>g|2>Q{*3;qW;$Sk9-~PC+DJ8Bu&4d?J`hiqXT*$Hf_dpSBi&zV$ zUh(imln*i6qoLz|8@XQ~3;Er3&`o#3ziFv#cj{U4H9dlT+R{Lo%1O-aA2zI~Y(Ir< zX0-g#4#6Gg2737q?=q|rQl(r~-k}&z-i=TYK9swRr|ki3y!**<)5~sMvE@-?a{inrgnOh1IezO>6UPE-4yo3zbdqEZdw=U;9 zSHZvKabap6lSC7+bsXUiHl|Q-e|6k&e>cd9OYt{=D*7|pfGjb7CrVoV0xTa5;nR?2 z5;yA*bk^LXGbTPJZ>O1q+Q34}1m!~Rg0IxbuarHj@r_wn&btCu&7Tf9G{0j16SkRO}A*BQX$=xUc-EkeSpOVD={W< z3UfPYC%HPShzR#jhe_>uaHf&Z^WW8h<I!w8oQPrFRmpcbg$AyMOWFc z4$2(2;vjl#dWgPFhTP_by+m^MN!CtX5v8|Dk;f{ ZtqzAWbXL~ecb`z}S;eAIzx zzP`-{Xi8DtS+mHLu&*Q~`5aVeB3vq!5W3zzk53#`$gde?I00=S&Sr=mJ#r1Weba{g zcFt80SUG)76k6z&+Um=fs>~mp#>c)m{)z4~huMYmirIaI{vjZ8zi9SM|f9EEXwZ{7NNB*yzIaHh8=z+as%5*nXFLko67^DR?QOg#z>eXpos zLI^7}--?_*^oJe@SAr*jo9WQR(^PiF2(onfbc{;rXWlirk{MMmX+O{Fd$iyIE`274 z1AqSE`<;{VZ>uV&xA_DYT9br>o-gpOuP%(a7EMPCog!Y%6W|8_9vCPa4?8Mishvb4 zRa{=lXUE^k^^2V(*u0XI+g0Izr+ex2)&Y84Vm00>lgHN!N8{jL0ly`xBtN?&;Ynr= zNewzcgSjknYlk}y8LSf7e@eqUe_TM_Us|BEVM7j5}?{Z)ehm z@aww<%xvKqO}g!TM`k)#_-Yi_ZtH+y_YTpOQ7=e`Qx3{VYC-oRIT~Xh#+1CX#C1y| zXk1(q+Rd_r|EAp}11UbReX$IeWG|p{e`R5Ca|Z6SzlaTXbMVxZL}-n7h3V6sa6@G% zc3jWK?YBm976X<71E*ya;$?U)gh24(+%dLobPY`|(}ZR0GV*)16}>Qg4OR9o!+{T) z=+RKcE=o;B|I#&heKZed~yhVs`oQnefa<0Kt{EaQIh6lpjv!2d+rj#9d9K2rVrqG zyJD=av}TQ>YLI*54s(79@YI)Txb(ChQ*!Gt?t>mzG`N=iu5UuZ`ybPIyB>PPeg`_} zr(pH>Qre+nOmsLQwLa-dPgFc5V%zKS>zpD%-JwG?%I6f#y>yd&Mn(G5AOK`+yt(tY z?K#o)GK{)fOtpI#(GKY@_Ml4+XiYXCm)~2W+p=7!`Yc91hBY^e+{L+bUBN`QzK`hL z)8IP4xPbgeEuzvlk2>AVq-)Q|@OkG{k;l<+cG~(UAXmfhc)z=n`(wlyf7=AisvL#q zYlg_%FiW(pw&7Z*X>!*S%dpCN9y4#tV)PL#CK*w|bn=xFl5C+*eXFms#q*vqr#75M z3-rXjz0%z4mTxp9-H|*{Jj9rfsbZlwhWcgd(e|VAc&t$kOY0TMg<)N6aFay2b#65B zaSM@Yx4|_9`?);6=jQrzFLQR?4tzDm9z8mx@wK=k6tFFHszNN=@>UkLp4}rxE-%4J zXOi&Lyys-C?Kt{%DxYPatwov1z0|_(KfGjfom$MYAuve0&CS50}bon^wUA%>Ah6tc~ z)nzz!%YWo>jQ!;c%ack;KFuiTw+;Q%Tp-1 zmvib86q;!a90o+jf0J`=+VAra%> zF$X=bk<0H!Vo7lUtI`mKi+G;@$mCH3ck7Un&EJ@f%M$p{f*I6QWYWx?HvDtKe{}pJ z5t+dAqe48InZ(nLjjxuOqjdRUNVvUAbl_Dh%oZtPP)MWb}aS4Hj4^ zRf}0e7HIn@i$)qN(psKPoM&Z>!JoAtXwCtsw6K9~`M-!Dx0k#N$O5d}{v*$5z`BsM4bs4-LYCixH4p)5_eRI}vtYa=`QAxpc1aHZ1BB zqv~JAqRjp`Fxe^>N2jtOhID zO=ldjJ#RecoU#BXZ5be6d^eEmPh-g4duC*s^DjuYR0eWn6FJ#0#Z9}hpV?|NiCd=n zhR`yeJ)88Ft-^9<*}gy`e`Y>}scfK5vqQ)hupS4$F?j%N5eo%GU8Q(SP5xpI}7cVZK&Y5{c zkTX>hwx544I+-8Cp4l&p5B2z+)x-%XDahj;nMb%cb;ev^^-g#r{+;~mXeFN(m(ze( z27)U?Hbho`K(wd#Drczd#hqNd3vIt%9_jtV{c_{?b%ek{#R1? zZSi`D8jL4f#&v+|mhE&~=PPEJXCD2a`X4n?suVe2I!SJpdZUtP0a#9{g8^G3ZtGbU z%$V|*9t*MMh98aOTAZ$<_!4PB%!PK6P_qTJjLQXYswt}_B?l+>=#$&uPEnK0uV{Qy z3)fCQOr@TU#*m|t+%}oUi^i6=(C*mJLaaV+pUP(r;v;$I+g19GAX=lDebWL~#X%|z6KG7-?f8K+N!r#Kc#Wu2g=n|Q8Oor*o|I>IXEu5N-*T!ol{BHYY zJ@wss7)-4$(0eASr0}mP?~ixItBX>}!R?0xAxqz*@`4Pm<_V>=yAsxWN)y+zc@Vu} z6~3*EhCOJ()NEW%3L^K>k(SmFY-Ejd)f4H0>H46+=iXYYA5k^GM7VRao!<6tqr&(n zcztCF8uwhq-gj}-s3VK_F1qm@VkvCdV$9XIe!!rNqj+LeA-R8~hi$l5C(=qfMB8t+ zFxw4|FoxBOsVJ*}G)rAXne#Vsy?g}Uy^E%5pZ!SP%(3uFwET@Ow)!UfukGdG$d+fAFmEHuIHSitlxD<20XRzw_Zney<|sW+R2`c}`jo-%~kO zcc@XdC=WhJPUC$)dx);wGpxTGN+MO&G5KN=sTo=Te%*#(t(Hv_^~@T&K4n7f%uM(d zA&EsX2}pb5P%t+e(6vPTO(eQM=%#z3riX@jMoF4%;!F&QqbY%K*c|rl69#7=~Z3Mg21#oM&GO`LQRM zm@gkeqPg3|X!}K&f47vJkGEv^P8H#i+IZA;`hvL~GTfo3H)++GTKX|&A11v0MHA%f z$$!cV8R-T~_;-4}BNh+9xzVancX1`zVJ5_#p=;@>bLC7f-w%oPdQYSIo{nAA3?iR9 zjTVb7M=2)>PFEw8X=+x&t^^&dNN3^o0)9*JshmyY!a?uj5tMAq1j_|8@qNQm>N_fk zre>9MYuh_;W#KbeQ!2@Zz8?iEer?DJ{l2f3xFrO%M{O&8Xm&!7ZFy?UsT9t3Zt+#UNU)2=6?kq_j z&*Sr<7&EliYlaoK0^kJCfKv08fHtihSh6UJKCGTdZx^hFl%=MmdDky0w=|4ymOqLn zrg4n))ECT^?lKsaH3lAk-$MLH-XK}yEHS5JBDIe0riq7&>7M(>Snw&ASz(*Yy92Ut zuE%YR=5JaV4(5)p$2-Bwu5!33>jkkZ>~LfmlKwc}lDN|qQccZCfXs37#7~CYFWkzS zb{3HC&?)R`O)-+TJBt~)D2(y5Gv{u2uEe!-H1I-mF)BPiLhZ!YLu-35r*ZZbrZ1A^ zUah;uB)-vLPRvUIuLJe8Sm7MX8YgqpnPW8Og%~~=!NI8aUS!kRVscS&kjnF+zR^Gn zX>;tRQ!11~_kI3aQ!`&ccSC>lQ` zR6?553~->IG+xoGQQ2&0|0t2OLnj@3*a%lX)Bu%>amy~|^0 z-@McG>Pr(SOZh|o@;wGGwQ0D{^DRvo{h7!*dE-z>rbwwg7$B*MX!AF=ovZ;2ehtRS z61!+zOB;@y$imTseCI~N3vE4iV8fhmwCaKk+!gm{R3ieZ>v$I&WxWidE#2{e_ElDF zdm+vJ^pB)HKMMcNOu-O-PZ%+DhW&Fq9ZX+%5w~OcV6jLIl^+7xyyY`Fy!;)*yLVU> zpG&w&`XFYv?!>GN9SmAMle?a=2!adF5GDISylxyU5}3XOxnqi4pYaLkJW@}O#-G57 z4kLVR-;6Kx+@QwHhNgaTWF9^oCOv6=WX%?REUKMNtNwZ7f2~@~J&7YY?vxa&)QzC= z!wc!71qWD*5KUaFslik{xJ@6HJ5kxdwRpXP|12r|Lg($jsqzIER{o~C@!+(r2fE_-#iO|8+G!G0T8E9g`LwW4i7x8<2&(0g%X?9Wnvxjwb1$_{E5)ddRdCeh8_4O4aRK|Iu=->v%v&&m`dt-=FlZuLdBvQ>tRZCg zw{SPN<>5PTGc>!ng54Z9OvW<5SnDr}+`hwaaO~7a_|bS59r8KK-zCfO_3BzWf6Ydk zequ6xG{7^^F7YgAeQP{jF`t>XjPg68%{b0(4ev^u2a#__LHyooIPmc{6=mF{@hx)j zbb&Ocg%8qg>yJaMT?8neiGwh{n-}}+8Qt{Pf}Z_wh5XX=hpXqU=&E~1KtFdn9owD5 zZi(25!rEQ9A;gtD_3@$N+7IyK+8Nkm^q$6q9wCNTyUExfhOsy?9)co9qWbu0T(@^L zM(#QQIR#qqsC6nQ?sFLTtt>#N`zqWsJwL3M-432bR^YjK1$rN@M!6}8*liGnb9^Gf zh{n@HVx~-|@>4YN-weNO`DYJ<94yrMg10NyQ||j_ysmA9f4ciI&aoU-#U?>ypnz1_ zSV73uC3Nw%8Fc0i9aL&5rOPMil8?Td8AqpbI4YgSmby-XPfkbSlFM4uC#!Mm%ikDm z`J0sP^CHpuN!+7@C88NA6xbQOf&0ll=DkA^PVv1){#&{pnhll`V)_(cdGLNTuZ7$f zhj6m%sUqY_gS~xNYuxM%k%=9GTxn<7cji@*9(h)1KMn-rTvECcgyw zI{b;@0Y{-zPz^{vR6*M-S-gX>iyHnYpyC>f_)VB9Iir0Hrd>G%zawDDE+;BJzV!1fw|W= z(WgI*sQohyuK7DUKKAqmlM^HH(hgs8{*eXT)zX54<2Hkon<93_YttWk2S}~F0P^P< zLMPA9UKpc*GcN>j;R=~tTHQ)CL_M-j(U4UuJ&6yV*U?+E1nl4fo-Wbti)*_iV0p|K z4$iuA`^T|F+Ydj3c z$dVKHDro(12HP1F&ZKwQfOZaxO$RR$(Ts~wA1Mnz-Vc)xYmK<#urQiCO@V$`dW&u= zScDqk-h`}91l=@CDEv@HUQeGn@GD zl{ED-I$vQHj@us(+hpR23ZF-;%yT7S^|{cqZwB}3VKfTFZqRiP?mL3&X}swr$;s@T z0yiS8@KD^=LwP1p=5;mOnyVON$@an zfRx8cfyD7=q8j0Ga^Y11*E&LmTRRxUUH==!MEMEG?yNN=A~cQ~cf~P+q6IYBY#HaK zAB}4^sp71|wlGqno!%ML{jQT*~)b&QbSqA#~~J z%ZN4YbZ+`8yr(*uj#mnx<~m^{tily6dmht-!elNzdOmbVB(dGr1 z;+s+xnDEbEG&IED19|W8bze!`U?ayd;ZdlPs>cP~BHUN4t-SA-W2dZ(X3I;CfYFo` zu#A+XMsBaknTRT^Iepx5eQhsp{m&kMe2Bu8&n5A|WqYpwT^>#@e?p(Rd?0@oALD&% z^HF+QBBM}U&e$Z5!sP2~$z@lTc{wj#wAWn*=1pxRQm!+>ZHhek-c*cg)9M9#Z)%Xm zZ>^|Ao(Y^XxrVkcWa-oQTS>QO53UcJiKjFInN8-k^tk^7y!-P7X5H6>;uZ~jJoX7` z%)LbCdxmf)&M;i+Mpv@?#uQHWV-*eEP{uxv?BD2hLY449WDAs?IKI_ruzD=5eXCCsLGS!cw#id7h&WizX$~~aNyRr`CNT8$X z8j|~cI+!Xoq59r8SW?tUb$BLmJ$FTbd*z_iSR5ILZ8)#Lo)rDJ|Dwz8CN`+fhEwPB zfX=LF&d;=%?Ao15zQ2=a4>IYEdSA@x&&VREnWRG;%TMCm)hp0;u0LVSx6`ye8+o?S zIO?@y9dmFc-#K5n1@vMS(AmWZ9~#!vk6$OzTPfFQO6WE6(V>gl z_F?5;-=xA<7SyY46E@9Eg!Kbs7^}!Grt?A@d=S!6uaT=OH7L9@QVJ~9*auT(lSxq_u`e{V(UVLHK zFW7b`g>?ONqOUcj@p+dVO_a-^27R`4fAuBSb6Y+x`7{E1qLN8p=S0}!okCL#TuD-x zCiXcOLU*Fw|L-#N3x9w7ef2Y$5>iTY_E1~-6Z}*h+9?vFAo9+pMzblc0aU-y< z!w-jPH}s6lWer{9h~1b(Mqb_vwBGYsbLR+jxBe&6`795!9b%yIPY?-Rq>eg|L$Syr z5N=gkVcjb$YP&%I1srL04gV%3bM=Ms z>5{_ZM+!jK-Cg*qeW6I{Y95`s^fY`nkYHl^U(=pG6^>bZjvT3!$35#G;2PCED4Ttj zxL*l^F(q%9mZtaAB&U@;)M#LfCnZp}-kVBToyKcl1o*si6c=-25)~}CNV7kiW9oDd zs{HzlNMJdUmYjG1ihhqtdT0Pld$0(mAGyth$_}B~R2z~oB9{)jG4OK7c#L&_MYkWf zr^P>(L*5=wY=2-xdS4wv%gygZ?D=4nVsqfZi8RrYDh+gYHpDN(nKZ92kB(LgMA=gQ zWPfY{x!BZ6|4Q_c|Bmy${ipNTZR2EU7ym21S@|;*R+IoUI8K!}}E*A!^?<_0fQr>F>>y1yjXP`-*$U*Q}VXc!w-vS!^q9B=~Z=e{mQe-|Yn{r)^0+<%N0WRhjGmGn<4nFrk&ugH^f$N7m1%*NJCtPsr#pJ*ml1j)}1|# zU*aNg>$cyx)Z&Za)%t_FWv%!;pXYtUc7o4j}LJGep%Q45&)+)?SaU+wXeL=TmC2UnwCgZfn za1ZU4(h8$Wa55by^f$4E+PNu^bS#ojG^f*-(n9)G+lHN!coAdmd&#Jl3=A^Qg3#cT z`0PzPjGgua>KH?zAb%%k%IDkmV28CfQ|Qnq9cIEOF>b@f4D7r2kny%UieEm23Kx33 zWTRFON%SFp9(6=6e65Z1pQ6Y z@N!xRl(ih8pW4bAe_3T?j;{f?)cz37{<{tS27O~$S)?4M#u0z%kJ9ZOp*dT&+e~)13 z6h8Yub^z|@s6dNJ7CQVb!?6|Hnc*+e*gS6~w(N_d)yodiE7GdCl6PCa-|I@U&yE%v zoSO=>E+*3T3F}~O%}vHxBR>?=D= z-DwFqvVR$wc1e7qwVTFu>w#?kNc5X*MeUbtgarfI@YVl@fOm=tv*#Vfi^6=O6L1jc z7#iaj`v~Ix={_u55dc~@)zRWw1}(ah3=4eB$(5(AP!aS$iq12j$L|f}Dy1nckx(?W z)1dl3=Prd35+Nf$5)q1u2%#ZqCrQziQd&eB&$%C4G8-tPfoLEU(ei)(FMHRE^F7Zw z_qjfw>pxnw`~~gX9Z6im190!$RZugtjg_+BigwR>d9ID`zc?O2-N(gN|M2`#C0H!R z=n4NJ!dcVljP*U_aIq7X&P^sQiDywPbR$u{!oNvbG?P%dPzbQUO2#hxgzCTN(wbl! zP;uVEeUA}>Il8iBeUuKw#-@@D*ACI{FS`7!{S|eyJ$%mkoCktXWH}E8?GuD zp)XTMZ|B(3y$RCRb>DdpZ%`J9T-C%sP8=itNCeh2&afWvv0|PTjD_nhdvVcqZLGUh zNi=zf+14E|n9Gq8@I0gi51QtXZyPJgUb`e%x9A+W3MAmFlt0d0xf9H!#G$MFKQxK) zVSl_+13QzsWKp6K8zB9e9NxAbb~YgsGkS{kvS-Pc#`8m zQr4Xp9?^+jJ81Xj=d^YMLmqsHrl0$_LT24mVjLFAyK>J#@8&Mr_frL}{*1%U14nUK zPZVmzCZV~16*+f$EZ$!#XD#TkWTv|skWRI!n4-r~mUlxP$uq=zDbI0vYbiN*IFzWa~7oCv+%nfY3NgJg_aFuI(99t@`~O<7Zk^y>0(O!c11SzERu*KCYHcJ2 z`XS6jW&^~XI89O=#L-7&4KWn(Yf{GmmRwo~Px70f_ENuKX!dD(^1**N%g(n__Om{Q zG&upI^OC%hEQeX!BfyARN9R~+GQ%F7@KsX|4s2aWYn@g?qxwlEB&NnH-@y?cbRUL2 zAv3udZzvYKvoz5wk8qPBh+}ageVW-xu^=DM*_hDa4-#mWA`Z!IU)k6_YPhjj)w(0& z8GEd@2E_RHK*JnD6jpbTmz#3w<=$fAHvSKJGdPiBhD5keF;!&s`k7>pp%jXz|6%=e zHQ~&TG30=aG+2e3;Lfe5Sd%1{S|yF;?kjhbhxQsg!+nEzFRP~+xCevz{bxpgJ#(>N zjci`1L#!1O$;9RXsxFqqj@-*2=2MsAH*+Iq&yGX1=93Y|FLdCJPd30WeG~BE-Fkk0 zj3g>t6%OXR(bQafqU>ly{GEQ0;+YS~Y>8QD<8T7=g$^@1&XnkVY@o1ZGBJNYjV#|0 z4~(rBuD+=b6@i7MxNJGSe@Yr>g&w52L-OF&_(*VQssy>CzOcGXY9p!-9K%fyccSC& zQzY?u7gy$)hP$i_h*`xN?paAJS(a4+Hez`o=y+T`Q^pLwU;In2uj;1VN}9xXZWeC! zQ06LT_+U$*IOb^x@FNecYR{>Fdiiaj^Gh4Nty-u0>X(o!K?}6bZUGleFdOQQ;D2dakz#J3Q~fIcG7l zb?^inFEQZvSiWmg_bIhaTZT_dqltzg0oR%i^6dLfdbqNjzI|B(uUF`RyYU)2aZe<> zWu%2J=+B}WY2(4Ycrn>VPY@f^Ui_f9obgNZC);Na(WA4os_SIT>C%#L$cR+I-f5a> z6wW)w*Pq8f!LxDnb{wv``&blo}?$PMY6Z~Z{Er!ZDibrqh!_4u%NNvHG5^Z z2;8gkg4cImp=Z!p_L_Amx-R&Qote_q{f{h^evD#vW*0LuA0LsK6VA{Hci+%Ab4Kvj z10=yY5~wHNK&%=)(CoG6t~w5@C-LmdhXg90A>iFZ zTC{0H8*Wn)rK4L{&{n%>_)}Fv?!`~Ro@07k(2j}Z=$$XrN2!<=E4vBOD>~WC-+b4~ zG#!wf*vA~!)u*q=w-b@H3TpOoAFAjZlktU-G|GQ1TsM)V32iUQ%*lW0j2oq_;;9r= zcFe;U`W~#Q{ti@RebI2!L@rjY-%|YhFfBZ(L_gUZ;%WX)cI}xzx$*uqe0r5d3>OX4 z>t7yIHIXRza3qWQJ=Yh;eJ!MA3LogY6~6G|)EAn_9w*7Xn>Oe4F8sC81`A)8<5{b} z_}qOqbR7)^tDVxAsVaky3=3JG;qfr7FoY2~ejd;4OTvo82)t3;KyQ0xvxgS06I_!` zr!5osu7GEU@I|mD@Am&G=-Vnp^_JLhZi^Ry^R4^rvWe@+`HnZZB2*ZcExt?icLs5{ zHvs3(dyL++=YnJS4lDQIbFjZ!o9Wt1(2~~EK>j?T&Fa#v{klYW&27SZZYR*w%&=?{ zJ=Re{2R`ayQnxiIgfy^U?)o!cyZHa%MHmh67Kguqe%P_=H#QeefNb+UB;{u`dtXlx zUXJW#%y`F`g}Cz?xWNfC()JO7{V5!3_=39XH^JYi9OiO|fbr(nmR$=ISNE=Zn_mZx zUfqViLY|y+%~s6k|K8HNkZch+VPcy*zDn>UbEAd&8$4ZP4WTiS=;0;G6~C*26h|Ah z=b1PQvq{7~+YvwdIm7!6x0!~WX*BenCbzfk9huh0z}8VsF1~FHN5l-kjJbk&*% z1U=>wEqjmZG1GnM-S_%*aMuzNwDUV2_X}Vj=q2IOpnS5RuY$(PIgrPzmT*UG4x-}g zN1&{FhI9y3i5LCN7$n$JplB%u- zpH|jFtg3>wl=&%c3*UG9X0(Ydy!e&$bQzOHM~z^AWG&7<8^(Sy(&s`Zdtqb~-%oY) z3@-YUhpsj!=_HjOL|a~g%NQAE5+-YM3&x-4vZw3f7q@7fq7eaSr%Z+%KS{ilR)#YV z9wcjE0A(L&aewEla3&thV7|*TT2j}$+!nZeq6vSLQb$rdNI59@(4)_n?~lZQh;kg(U^Am7JYbaKd1^> z0B!Vv{JI?8-IIW=S6>qC$8ykkJ)28ScHkN(Nps9ZDSUO>1(tuf40n&T!DEHFQ2T5? zF1frLYL;GtUWY`+X-uEMR+uvEMpsTda)=a57n5h|hPXj*6m6&t`>7y<_KgOEKxZ*t zy3R9xmGXGYubk{2+d{~nK58>t2vnxH;zCyia5=IL%qORjb>?n(zxFRx2u&f~Z_a?b z>R;$?y2Gl>H|4Hu{)L(==3w(Iaa1-*VatldxO)?{XyRtRo70>(7;=%gg)!nP6$jDj z!D7zf-ezn*5`{kw{l-rlAK{9oVVw7AgdV74FuK7StBz5cd_)GOj7dW-cs6M;d`#R| zw-NOT-*D@}56qIuE0|opKfDs`C0n1K1Q&;g>A0@~?E2x00Xrq>+om1lpr|4~R_wuKEer4bWvD}3IT`rEb4V*s(93Rr$@`Opb{#h*W4s6H4CzD?^>sCw zo6|xbn(X8BdNkSV@gEZ|Ww?IVLezYz4apS-P+HePcG||^(YPXFvA6W9057p$eCKtwe)QxZ`qYFOszw32~D2DaGVgQ6$9 z+1#i&XnEGigw{o%(1to3bEX_eG{#|<)(U!Rt`u?ePNohO@_4H97O`BFPb(tLaJ}bq z#y5T=v&VB1j+OdKqmORH*m!kLC^v%&U2dnkcVF{t=mHujVMS__mk=F3_rLFmITY~h z_}TZ?{it?Tjdl| zzGDuqh}FbXtxxdV?_``Yx`0aVC?WqU|B!&3gFL^J1OxwVqEnW8L*0V_{516zx_-RJ zv^i%pjw@r?s@*kc_cjA+qup`ytWz}p(R^xoYZ_e`UQQn-)Y2|LBz_iFu&$m^(~Tln zxnG~Hoj(yy@V)GI_vbR@%D2JhhcQ(SZ3prFTgV~HJJrq`g&B#nckr#n8Y=#58tT1I zr&g`At1HPdlCn08%q*N&9eQyPnt~>C!)~(pa4YZJ$n36OwZsOtGCDNmW-?i=Gy|mf zS+f&WbIAFxe#9A5$*tgc(xR+_{XPD;Kj1tZ>~w}uQAM7!lIOztjNX&QvZVTl19*us zw1_Q6K6hc0nGn=AZ)HKlzk~O4eh=UVp5U zOQNw#lHmCy5B(@>y~vf^MLB#Z;`5qL}E z4KDY+0vohSiQ;N0bUyfrST4OygQgUcjWUvK;gjW@R_ns5@3-XfbJ0a|z$_mPL>ACu zzN>)sm1R8j(n+#LE)K{QktH+vxpAT<^p6HmH$^#^IS@u|r9`0iZwcQ)KL@)Gxo~6e zHdFUctLY2vS}aSui!X~LIE&e)=<{nDxtOwmR2)zR3vLz~^?pXPm}OWhGeB*_=F>eN z(!ogE78Z`JrHZzXX!aJ1D#KU9?B1?}G@Z}T+D-Vwtlr&%2R4f0(}GJ@>3;8UC+{;X zR=C2Ji>-n7+!VHi-!rs+8^IIjXt<=2MqbtHaLGU2S?f#3A!0`iF?h(&RTf`>|K{%j z%$i9B8+E}o^b$^d;Y_xOtwQ5Zk*NGk9EG2)2m7&~U}2>p`Oh+l@l257Q>zoWi$irx zXNx_=4*Qb@)_R~NatE{Ll)+Ic6;S#S3JI=nsoCBF@_G9dZt-y;FulmXB}GkynKcdu#it^_nb&CrY zPeRNxDIf>ewBx56_R!mP3soZD)7F6{z4JLDOLhVI-QGJd!sBWUk)R#4uY5~W4 zG2F&I$52bP7~AtKAZL;kyYA@^BImvtvNw1zs_P|T@a9Ll?`|kOQaC^_bcN&d6KlxS zj#$>^&k&}MS&Pq)bNJ6b4V#x&}Ei#3C-q6Zi%N>t?_E#{X z#g}%=JfQ7xm)>&}<9gBt&~ru|wiKn4uM5AU+n)<~;L|+nW*08_;_U`f<0MG=7+-;T z;vf2cd^TRNmgQWQP9tli?b)@td>?6038b5R5e)o3fe}sHSef#V>?Y&))o;I8abK^h z!jgJP>*E#%%%&H;?4?t(_$;^?pNff+XPswQM$?v5N|Z2U)X_>R(iNqnRJcK(wVbI@ zDf@3(3|ZV|4rYGBxF(jrqe-`rF2PJV8yL-QV@8>_{0+ovdn$2SBu7JE8R6-&Svav# znO6R>we)%~$4=cOPqY85BvBV+>6^?I{254yDSu@_O+>w`8(k^gRjfsQzxv|@-CUIK zZN#`+MW`|+2|RbIkm|TDD3dK_E#Bxtf21M*u2KPZyZg?x* zC5XVOeFHEjOcK=`bEw3YFwD6c%Bscfhw}a^+!-{F)MS;jM`kU73sYTbTW}2=$lHk@ zn|iQOFd5xTH-OAJ5eU|+$F<}+gq?XQ2$q{}J!ZWQWZWMQpCuMR$U=Fps?rqw98;_| zs7!$ro+WtswGuln3{ltR^Dyq*3EcnA0G9^J;p1`<%nY9ZR(=&E&0C+InU#u%R7LS$ z#&&S)$O7ejf0DOY07dch(00xwylUvp`%G9`eO(_`Zb~3>H!JW(OeUGL;0=uztgZgS z_g5W$>_itU1+ho@{mAfqDlGde1VWmQL`O*q`5OV*zNVU~RvdxEwc=Rvq?9f%UQK28 z$YRp$yYzv-2xe<)gL-Ztc`Kxk&xIta>qY9=ZntzA?i@xGoW=1;?D6V?xB^J0cDO&$ z2ahT#lB~O;czXT_`NHR|KNK5NgFlnV&FThxnjZw}=Nqa%yXnxgr=Fv>cPdV`^CKU0 zU9GOg6hWDY1_XGVA~`0PY1w^G%nCE2c@yT7c_r)Mkw*nE^MYQ0&Go{Bqn9 zwobfAj%9~}N46^YazqiH?yF!P%#NhDv{FFk@=Ta&!*jhFJycy#j$ef4v5RFbQSi+w^X)_JC zI>RbFb3%*7O;l3lJau3Dh+fyd&6>9sk$~X~)aptyWMtp0dO8c~){|;<_4igZ@#>&E zc^CNQPa@>{h8Mi!uZ*7!e`4)VtYu8PqG@LREr=MYp)wcE;mM*{di3fmGHyu_*=DHB z1jK$~eCFR`wsY`g@Ys(UH^6W3Dz4I0lN9>?psh{|6OCs6!RkXuc6CC=U zgR?{x>G5RMrnS)b@exgtc}QOikHzLM0DDF2 z=%H7`gj??n%fE$@m}zxb)f0s`Q#S}E&HY54L?rM|)rUmftsj4W7lA%YYp9_!xnuFS z>6BIx_)!I%)sV&9Yk!Zi>q;{6Tn`dV~Abt2E@gu>)+MsVNRlgf%O z09T)RG(4Bjr_U0Q-C>Swit!#4Pf5gtH7~1+PrFq=KWt1Mxnxtr-7JoeI4!6zw_)e+ zjUl&vZ^5(NTB6dTg%vkWg4gH=67WbE6C1N>fYp7nePB*?k-k2x`OwCESg#Dq52|ol zZ~*F0&qe9Z4R~oBKVx-PpmY7|P+P&D+1xKm^sO(`loi?Z;jvh%`8b{aJ#vm-f3g|x z`HONkd?v`w;Xm{`uF5gbfNp{DKsY`gt&m{_(Jmfzk2w6 z=r&XTsEMA)Xutu+2P79IF|#%s(qkIU?8$*wp!?$-!T|+hu+@hA$S?msO`d(Hfxi!VEgu9gL-p3yJ=ijf&5j1lMXe)6&NkB*?%8180pe9R=C!Kh3B3 z(|Q^zuS%hv$=_kYHDMU!@BbtB^01+1JE`TJH2o_U!Nc<7q%w|D3ED*Gh^-=O&)q@n zl_@#Fv@kPwXjWU;?FR0l3z`QiVDJ@L^f?$p5ruU>M<~;Q>~_3oG0^yoAGAsH!{E9feG^3N?+N9)BA=N!PTt^+hb}jYK*z3!6$rxx!cfXmoNBRem8y8bq#wU&|V}u>3Y&*(e9w)G7r{VLPBR zYax_k6>S=PP5bj3_&xFwxa5UU^IIpu^42)GD~P133yit4(NfgZ&$Mn|o`X$>Z>Vyx z1a5D6fL2>{aOw6NyqhB&`E(BGY>oxN?;?0NP92Nh<&y3iCwS@|2yI=DL3m!MpkZh` zxj1^2{d(>kJnLLUo;G!}j{?m|hwMJ;l@^DK=CI_{ao+EHwivbycd;i<{AQ|u{3J^@ zJtY(L2q!-x&;2gmgFUm45$V4O4eHxyQ{PJB$?rt&wp-|m(x2pr`a7C)G#(%F`$nU2 z1+f|mgru&%s){LQWWZb?ueUDpPPxMT=7Z-}|zN6r-&k|K|A)ZC&-$495JJ-*2pF4Kr1+C}&#ehbMi_&`K|AEj4b znvu)Cspz;=%(`f-1ZZjmQr)FPq#&uGN764Sd!*aK0ExSsPXV71{T*85*P>+`S%Y|W36^%Da?Tq_&)ZY! z!rTKhNjyNxNlE*9a97A9LVh>k=9@f5cQ^0#A2g@apH)(g>?f$nHq-OEGf*(>K(bzS zlAwacG-Pf8+*N3_y5K)TufB7JnNzheGJ6>H&&=SU=Oq7;o`Fi2q)~f1((0p8c*!}E zjJ;S)r>ab|o>20dUMTxOgEX`0E%8Az=ei~CJ2Or&@3A<0yrPWF@9DwTfi#>Watg!O zd%#S$UDW=#CJA1dj1^W_$cG<<-bh_$9cT3hXUg9}58)u%(sZ7fXehym#!vXJumvOP zkHcyA$Mk%(CB9zzjUF`>u};xoVC{EHkgahE$kZXm^VoOtNk11AcXVOz=o4^xq6W57+iCf370%ex z5tbayVU_PQxbpfFbnf&dZw}^=d-qSUY6CSy-uWDP%zrx$ayGaqG#e84y@n$WmGs2H zJZP6tr>A#MraI;sf?OGIzHebYDf|5$PS$O}wLfpM$NUwz6+KDp^;Qk|!ZUpH#(iX* zid9HXm?k-YK9kfs*;eZnhtO)9rO;WE_t?W9KGX2XWfn?Xrjd@DN5OwX6Z>sN z4LsEmvCcQUMipEkC25 zntyr2zi)hIWS4DZpA0X9vi`r!$(nLF*F75|y5BGk%m1P7vE3xB-3o(t7c#K{VW9qY z8Z`TSqU$x6;-SE|H0bFSTI8R|cz$1n6RxYU3T9%~M#fuExoSI3SkF75#!g1RX@o2M z_5juVgSiQR2iO!_J?eA#3>?ro3H$8oXyMqgxGZHg8fyQg$p?!ta_A};n#xqK+AYh8 zz%s-d9Ud{ffrcIV{9E)6?ABbuRvfmXZ|?H5Sl16|dng2)O+K^v`Iq6H*(=EN7qNP? zNYlDGc^*LK89ZNefZc9zhS;qW#=%xo9NOxFzY=vou}2Z7hrXrCA~A5;ZW-3wHDUDG zSUl@3fj1t_rceAP!n5`ajQmkarJB#Pn_8w4#Y1kG{lEh#;K$2nlR`%CyxGWd#QBo zdFH8g0;XkB3|@W+7Ii%)#QzS}aNP+huR?KSHSeMOS%$|q#e@2551Ov)2}if(Q2Uj4 zsrVx)ZZ3ZoclogyE_gk}P(ur7d%Ku?K4nigtnOwq< zOUSfMQ)vGKf7F{4j#pgPfTCt7Q>%899(rF$YF37!wMrEEdn<1iWD(4_xk562CZD#V<{SLkVO)T_tGgM zeAes^pUp8hhkOP8ZA;Xb77F(=l1Exe`<;5KZ>)?N*B{c^qJJ?kRvmlyB+;=noXg?) z^%M7I;upWmJbRl=3)(a3SCMOM-djHFIPVs|jtvH5-2%2ca2?51w#NP+rQi}6PXo<1 zgF}lnj=T;h1D=bC!(l`E{enIfHJ?oj`nyS{O+NYHb&D>{TL;_s$AfnEC3;jWk=l1e zRr~Tk-oK(d*&sP%Fs?kqj_EK!zt4T-u2U>&ExtekhTLIaX%5^twg5^vS@?PA2fOs} zE21^&EonY+5$|qM2It{O_LZCrUA&ILRIzXLon<}EH(NqqJVQz~&=#yW6x zlZJiUYU#0k7a-fdT~Mu&P8-jq;Pr4Bp7p&)SC!7f!;PMBT6Hr2F6PIOc}AGCpW&8U zEMexQ*`eFE-CV|I4|IC$nX_I$J-qJQ;>g@PqnjIM6P> z4|Xh0gmDdec=CuoZmvim!8n^0{dI}%TV)9*U+qCeYA1|8DZ;(~eFapNW1&8<53M>h zFtz_U`zzL+4rw3a`M#_4U#}Lk6i-)gAM*^|?}>s*Z<|0OznMxoNy5v;OQ|2KlSR*F z;L|(TseIlsnCktT9^UamP}Lhry7#q_rhxx|>iwWm8_tqUhiK?{c?a0}>BOVq5*feF zhEyLIPj2fzqW{nTuD)W)jDFO{fg?%e_ttR!Y^y>~e{;pVBb%t6gd&)ye8rDoh;B(~ zkQaIf>X#SNg0pAoJdZm#uJ9$s{yv7U)~C=N%X;WI-kX?ey_Q56N02otdV>3|8F1U% z2^Neff|v6FF!-j6Ilg7=eM4{3AS+c3~+Z@wW1khT>C-ZgSr}K3?GMu50;}ve+`|tbU8iqvyj+7Im>F!7q!lA{SO{^ zrm&SxQ?cOkHd48uieXci0TG(SXX|)JOp+Zp#&{aYHFeWTU1xCXn^F4KYZh3KTStEd zYeQ^z6)AaRg5Tz^*)#__1Q>@gk6E+(HX ze8Ji-0+%u3gnfDz=Ve^u_xFWl$r?3W{Z*Qoy7Cy^QWy&N+G?2V-|Mixl9vB#51D<7D3WIF>{tG?Sjg~wpey#a&)~*So`_>2uCY}g=YoUWDt2VqTUurFg?0A13=_)JnPo=yP{FUc zt;KFQQ=1ijxtsrR4jaHfniQ1P*+VB9;~Z(KU)I zP%Jo@mH<0=;XSPhe(Xwl2|YJ?ZVvpSbkF?oDB zM%zwfcI5_8iyj5K=Ux$!QyoVfqOVh?+nThCUn8QQg9u$O0T03?1i-gj@v!ytFE(ZK zad;IYLi*1MKy8l%bPKxKZ~EnQ^IKzdwB^ri-RR+urKKhvC*l=EfVQeOkpu8GmHdvi;Lt(<}T}l9%HD(!UMx=Se-8&$i>^3KNE?j4hl{(HPF#=sDbI9L08-&lvJnl=C5{>DApr zM9edfd|mknewjrJismPgmaQt(Tt*l)8p}yQpF9qxaX3N!1-kXb(UWOf82G}CwEkRN z6%rl8tnC&BO`bKhNjl3-ZCuZh;5p3md#&u3y!mv-pa*xrd@|XkEcxg%}(s)!l-2v^h#6_`Nv~5jHEuQGMCV?kIkHQL z-3%Xc;7_T*^V1nCwJt+=oYq5T&oZPBE7Fh^nhW*DwbZ-D1kd{y6T|OfFl0Lumb{({ zrJ5zI{+n;?mhX+kU)hAG(=CXdSR3uU`Ic?eypB`WRx#UMM@c$YK))=`ra#6{E6({(``!`XiiB2q!qrayI(Klj2 zMDqJebiQAW*CRr?u{jHAD<;#$nYVDkKU>VQoJ;QUxxe(=bD4b7&CC{!sP1rWLD5@@ z`2N><@^i8`iTry7tDKxTksXToL%|5%@)?&Ik{xuLtu6tjD0YR65~muIijB9HbAFkG zj(*z%Cr+*+f8z(4Ce{!d8&vTAw4?MzuRK+dLVA(4Z}8^q^c#1@>RaAaeDNpc5ScNsC^R6Q$!o5#c8gIJtY;zO_JY~U`Jd(+IF|LA}EJ@i+|9Fm@}f+@PZn0Itk z(n+J+5b_q_$GKL#2Tg&WYb=GQi~O1SrmbX0>3G!MZ;j__=TX(C>SWTkS|a9jmM+{L ziv9^VAfeq5$__rI7u0ypPf->FisSfBljXSd!zLPJ$-6YlGs*VCL%33LJ2s`v!KOV~ zWWky^vb9Kyh)12F`lYg{SpJdshwp?Bzq`rh^m_8F`#EuGJC04fBhdATKl#$9LL7K6 zT>#Jfrljw|+C5&lr@fU1EUlqgZl+}VlN@IK4OOsp6e2DkYsq-qiJW1S8`oHPmA&-x zIT60J0PIr;Kf@@XKEkG)xrHZ|v_(+8l*=HaAVj?HBs26(B<71I;1WwKE<*hboibkz zu9f7mc4Bcj>0SyA+aHU8uX3QcO_bcfvJw8wG{L0s*|;oOn_7>LAzGg(IX`-Z;L;FG z3(zKS=POx@=54|qABsuWe=+piyiSn#t505K_0WsnW4T_wlekl4F((|L4tWnwK}q&h z6!|lVhXf1gc|j+=Gh+jsoW{SMiHLEVbV4A2T%etmhneI$G0tz74!5@ZEElJ`v)Xa? zY%<9v1Qa(Gz=8Q8j48wO@J3yv@y0dk^fLyPlC)uJoCA|>X+*Vpo2g6HK8P%+V7BFj zb9s`3INH=m_r`h=zYizr=)C(lZY@jvjd!Bh#$R;yh!Q-M0C4ha$LX>noJMB=yW*u8 zdVm@6Z*nH&TL7JXYaF-vh7k8~z=|t*HvycyKQSp{eAmkD2wdD2i086jVWwmb8p=Ck z-uUyl^1^lA|8$Zb&Q{^dwwQ61W|=r~KpeV_1DUMAFtXD8q1BYFH`xSsEUYWv4*atV zeavLz=s#_4f<`Ul7sR{B%lvU>*)n*rPl$S0P=+|3Ac2q9a%X3)!?@8Q+GUwV{?mU& zB}BAn-1@Q1hT9r!$_pcuFiFE7J);;jC`yezZn8DW%lWQM3z)pinLm@BVlP@p;k~u- zs70JPoxkpQ+t3Kjl2*cx^F8GHHf5L)v6jpBR>hLjydOeE2MnLDpej?3;*5v;v1p$L zZAS;pus5tM7k`iMn`F>#;$xZ-y&76}OToa8KHQ=z&ur$}sf6}ioR|Co^txr>UVt-f z_G!btBAXaiITiNE-o^2H%V|I&rSSo8u&{SMZ%sapI-NSODB1;A)~Aw*7Ms|6Md#sC z{#ab|g`a^>5Ccp#!2jmU!zYmu68E|npKV)%olnH9&9YB|(#}Lu|JDx1PI*fgE2zL@ z0YkLXwMb4Q-);T;FrAkCnOw5xV8z08_Iz*&ZfGc=fwnPJvpE4}YE5y$^0N@`6HfaU z^PQ~e!ual%2MY0=L6S;2E$DiH2d=4czjk|4`vv0YZN3lgZ8LynQBFAdtQ(xxoD9a1 zJK%xVZE{;80)L!+gxmPHxICFqrnoL1H9q}8+eK0kdeIAnhazEYSvmO|fy}fU3{3XA z%XaUGL6iOMSjIEdEz_@IkC`#aW#p(t)?=K;cO|U2-NRap7lLJbEI{>25FS|3fbLUQ zpu6x9+PS|I3~!$T?PJE&>gog-|4#v>6aK?#hT$Yle*!(CX9Lf~&FJE>mc%Q5Iux&O z!p>>w0{Cw=zS>)f`jYW$4+IkkXDgj}X7{w-`|PcY7HNCl^UO(Oc$icG1kX6t30 zuz!Xy>yw@@Nbu-kXSUe`Eh}NYO0%r)ioLh;5IX^Tu9gwy00~l1(uy-aCgC)XHPHP- zl9Ni#q2|sSZ#67xJN7^CXm0@#+Z5_oxNti z09UfLFtD-$&K!>*|Gj<)V;9_HpUk)8ls|UBc`p-Yz1IUctrkQ5Qm^5`m}_L5!!VgV z{|m`@`yR!s-TDK1UUnA5n%WQY$&cqxp$m*4^o7zC%>%6-{q;~=;uecag#T< zNm~^m1AjqEyEIhz0YPd^6?;U}( zF+q$C)#6^uM`37&1+G@r0JV_<5`S+C4i}$hN4#f{E3S8-S@aPGI~rm8%oI$?72_JW zUlm;bsR-{H*KtX!rqSFoz*+|etktrKsrg#=$-rguan3BPRv$#0@HyDnP=#Y%0=RP< z#kqmBTGE*7M^43VgF)+Y^w94U@YW=oeP^VJee#b`_MI($dU_e>mtDp|(~S_S$7fDr zRPgr9Ec)798kqZ+31g!RdpAXqTRDluc!e&{w;n-(?q$?+KS$zbSrRurX*hEK8|~cP z%vh*s6Ri^yplD$Tbyr+S-mTXqFJGM|);A-WxryItsm5d)dO#Sn+`Hgha3_`ObF4hc z$Ji~q-T2JD3+ncJ)9DB15od*3G(6YK{(G_sW-lqj4-qNwyPqTWWu;{M-dZ|I)dBj` zQkdvXt{CiLLNk5_L8t`ZZLRa2nUXM`Wz0AME zx5=ETCN$ufIxYxnVgBgF(?Kc%P07C4FlmTB8q+HHuQv*(u4M3V?LMxvWGwB;3nkmH zCc_NvLl8b-!vxn0aidG31=CL=G%sPGqHi@0zCKH|mi7>5cV`%nVlaDFG-@xs!$#(Z zv%=BUIMM&6WoKvr8(@8k>Wv+wKL;BGD=d_7R}$~6S2aY(zPr>xSqD!WX>%(syds6Z zhPZ!WhM>^*ICkANp$GO@pykB1#J}EyHvPAQd+|3KROEKj@L7C*uBZzurgY zd~*_8EDLpEW9dNNadca8iUf{x#5akbNycqS5LUfJVzgzTJ*1WV)>}#PLn>%X{6*?j zVMJnrd>Fx1AWNKj@$@+ z&7AqNL7FMvJ;eK^UejGwf0>_i_^jr~R2WK`0b`%*(%|KmblC$Jn%QQ9amq3%*_w{u z)0cwRO@5JXZy^Dqo%~rSpFXw8qT7R)5amQ?xVrfmF?cM=oj=d#@#~|gh@&RAW8n}v z6_`ji_D!JauZQUG_m43q<0&@1&8B0$?8q&lShSK=N0~9Fajp7cbXzZr{b&RwPG3pT zT@lo`7{*sg6PV``O4P60io{xO2T`>O-p@2dJah8V_w_4Cx9BJPqrZ{K64zlKC=lU3 zdCu(-;Hy00)$>pCOsSb`sK^|4C|CNZxM@>t6Gbh=15h0Zvo2GV;==Hb*qSFDOveY=P3C_cvJCfmXAyGlH9)Q3z=8zja1 z!=TJ;9Xy}a(kyA9R4SEb zvrJKDAu=SAR5*JbNl_@GNi;}GG!K&Kci#6ez~}Moea>3+X<)+pW3g8H@>CW!^!225q<_az~i;kvi~7a~^!!K8ZT; z+$_QUap+P$vHIuZV*E8^gAenC>2)0qW^1AeRsJdtJ4<(A&4kD78?|imkm27+!=fP7 z!igq3$zZ4SW{64lz^m*6vhig%nfICJ-FS`1^^0=2n)FDNy8MJmORZwX0{MNESv2*| z*o;SL06w4Tg%KCNuo>OYX^z81c)KA9mHZ=!|E?d@`}jc|vX7&h37KStq7m=8)g!hm zV~NIHOBBUfcleQmU-F2= zzQ1HmtTOieX&@^e37}s77(WY^gXM{HP%UvTZW)(|1NL$dC@jnl{W``zZIXmLAB)f? zx&llen$QJbXQK0g7bHP1iF_PPry7&8NglifmA{KgyhI{#@IA%oTZmZH1WUoOH_NGC z!87{RyMa!6*UJ2TU=0T&-wp#mG|wAf_3mFoEb0? zH1BBOA!~oC-)l-^owGpGjqmIj5O$HmFxr}qVe&O(xa~F0P-r^~UL5_$9&Qf8a|d&% z9vecfzqP`xhHP?$&(2Pp;=uViPGZ-johFLvX>g}GmX=Euh53}hxk%%BG-IX9wlqD&B;rnmCE z^rAE>ztWntSGWl3y60g_!VJ1!ZIH~{-bn(?HbP8vG}$M24W5jRhtWc7ZrxrBeCbq5 zp0$P0Q9Q$IjrkD7q{DM@;;=Qwl7W+#lxj>7C$@fdUB8sz5UY zoqCkWc|?4~Z-$Z}rT&J#Q5<0(`3PCeHQz?yP#$sB{y=ArEhR32SJ)_TQ!KbQ84Q(P z&<|f<(|z|w*%tMmv@7Tid!SC3J1^`EnTIlojJzDVWj{p5%xa>U2U%EoZ3-N|Vhst? z%*Zt5DO}xSHSR;}2H0otj%t06!F_ioa2LngkkAzi*@T(~qBEKVSLU7~(_@|x!Kn*4 z-}oD@S2XP>!BogqlHS>jWL%lKNS3sw3Z$=0M;JUHx%ufHqejO9aMoU=$Js-n8IrHBNup3GZe^ zac5P`IrFQB$e!0@pe#lYn=a4h5;R3%@8gf4_Hql2aI9aYYe$VN-zEAYU1YwdBzh*-6L`zdHVGep;_y<-Futrf5ZFFKX$m>0~m>(|lMK@TZR9c1q+ zX+XL^?+l4aq%LAPbfiZMu&s=wUo(Q7bw*HqSpx4`*3$d$g}IfMnmDZK1S=<8hn+f7 z@HjSF5ZYIWbCa%N7Eg4mk^h>iX7O@IGf6H^&a4bfYMFFv@q>xC+$&^?+$w_LNb4qsU3#=NYal zL-d-BE%wU1##Cuxu6e#Csd&V~-ITj{*5n>+9G(f^7MvmDtAa`Xv)zoQ(oy!&ls447 zupb}Q31jmZ9k%<>Et+2)fZdF^1+3w}>sHe(4k>cIA=>04c+jKOcDPB7rx}u8_$ok4IB90vF{p`v=d@_FZ<#8Rf_Q1 zFccilg)(|lv&cT>z0l%aMlBz5cp)nVKR&#UN+W7q+f;(4$1kuC)%Z-sifeSV$O%3@ z&;jSgT9EA&4LS8QA^GJjNNPGt3hlyS+YCdt>`NK#dNM4?OB)OP_l1TjCxVx@F0Or+ z$9KlV(RpAZd9dsP=x_tI7q`V7mgBL$OOJg!PlQP1RFT%JUBDSg zadMHPRKuQdE7sX#;r6Gj%J=VhJpVkob~%sk`7I5JQp+KB<2ER9*$ZdJ4H7q5M<~2< zjLx0sj+czqf!Lrn_v%V24%FJ>6wig&YaB+_iAh;>tIVWSSpxPdq><=^)p+=1Hz{=z z#na2*5V>=N>>f8_p3W4|PTPajCG-$nc%2B_&*l)T8|SFOtXA~9#rIP-e54B^_j8@o z3-DNG6dX#-fRa^GFsDYCY%BW-AG7q(Ws)-%E8fRv>akQE-=uMOMW%E8vL|VP5bry8N~0qe zO3bx4^JjtXBhs_v2{~E)P7t#`0Pm$t#ZB3Z>H6gd=!(QF%;7tr*tZ#b(_iBm8xhVF zE(z7FS z!1xT1munL7pGO2pJuX$RGaXg_X*Hpc+0!)B@%OG4~!`)hPTEA@T`3a>$Ptm@o_N+eg@86 znWV^>l)k~WyR2z)SSos4T8xQ{d#L?``_;os?~%tZWVx1@gMaBgTzF?E0IN_~gF@@}}5;CpWZgm2n}qVt>h z>`fn4jtm7ee)pp6@Rvqx>7!;N%IKt3jcZblki}K;Vrfu0lG}o%eLF4j{SR>}Z{ZAL<>NTBW$&=HW*t|&-k-}^8-cI9;30r;=_F{h?z4fDuj#l{cTgXK){i`Ky{we}j2iC%#jhBe7WhpIvu2J)3+*v#{ zx0u)o8$-^=@qo8J6RBK|c%4?Ks=@ye$vg>eqm~I9CS*>_d={~$OYCs8a}_r^eisaw zjRonN>$Jl!6O-(n&{M$gtF-50(cf|+zw$2G=1^R{Ss@vZIcFiO*F`*aKJb~UHK=^x zCz*D39!#*F20Jf0!S6k@!Pj#vo%1l2MjU@dKh2$n54RuSinb_nIbrVH&5stqKlqbl zFQzi0*5$OKW(-xotcyzd6S!&P_tEY7*)-=#KFm@dA-?NPF)fXt?;Hol?fPLN&hL#x zzcJi|ZDTk|yCi1ThXvF%&l3zgUXn7K7HC-{f{7Y&%s8J2s&ulAvWE3kpfH|%x#39{ z@7qqEzN%%H6&MrIAyIB1={KI$Sjv6*7f+Wx@`u910a%c!1@V^CVC2LU@?Y3$GhdDY89@m2s1B>N$Ok?~{QgI&|8oo%s#~# z#qz|wCyOxgJ8^>%lHv2;Nx~CJym#m;b3@GlJ{^2a16m6}*JLh!tI`q_t@5g_lr125 zekxX<9|z-Bgh9;i26vg72=v}5WW;|5lOy#3@cE7jCt`D-)G}6ZMfo}9 z9S?NOE}p}c^%1VkiXi5Nk(5qPp~40H?~}pzJT-W?r_*WD-%vztN6TTLbsikMrwPBV zWWs_ZNhtDEhEIkDT#u#@*S~Wym$6@$&XmriRQ7{O&({_DixwvXLZ5J|*t6!m8ic?83a^bXF}=7wtPzu;4Qic}oShGgJm!H}qoB zd>eK_LIg-%w?q3NZ**Sji%Lth;ZfRMqAT%$JZ2Vg)r(@tg>(tBc8?|Ij5p$Ynv3Z* z6wq%M-!knbCx5_u&+VQt5>PwxTzNRqLabp z>wR{6Rz4%HdWZ({T@H)KXHhStgxm;DB0>qd%>0YGfNLi+SC%|sgeG@WtM6$7(bgOi z6SW$S%}1W6xD0phK0&S>DWd%k9MR8G9ILLqpa)ak@Sv3nSUkESepCk&{X!abahP4YMG~UoT8Sp#vx#{=j7{CCWVrDz4W4-n684Y9X&<(e$ioqg z*Iz62_X=gAFOIWl3s_9Gr(YleSr6!!XRSnj#Wb9|Sp-hG)zC$k#o&*SI5_sq#DJ!E zsdNxGi``+7oM4XOY_?{2V|dp2*wYC84Ae z{+ukt0rnGF7nV=wE!aXl9n;A_jS;4T_c=;Nx6utJGDzONS>W(5zuIBiG`vj|@%6It zxaQGSewGnP{A(+T3D0_WUpEiuRCeH{WR_GM&wzJp-AG-k0Xg}?7dvYFY3bo<7D3_s ze)Xlgz{c$uVP<-A0q4G9FW*s<*0{r%z5M`BVrQD3c1pwNy%(9q+V9BCf6Zh~7Yn=g z9)Or-3fz}kHE#7bhBV0f!tjB0Xt_iNLT;JR@PXg-@Qyv~rEi(c*(1fUXeA3vr`zC} zk5Sk@>j0++c4Wx?A8iua3b$@Xa=}9qc=^*H+-22C%+}-NkLDuY9sG@*@tAk3&z?@| zpLemMQy6%fZotWprqgfwi>c;#XYRY}2h@KNjAf4U+-#}u5PD}mh=<>RiCgR;p*Bg7 zx^@yS`8$p7y7YlLd{Q2|Psf6#I`3%gHeq)DtRgBep5P8AmfEBiS92P7VM=%^omr*L z-t+J!ujd5Qc;7yJC0Ngt@pr|nN4EI!aT-Yst-voWXXv9-glQ?8hvg%Kcz)w*EHWEl z6!=}`6Q@N~`@;|wi+_jm+A(BfR1}Q+Fb;1V0Q%{N1g#NYNL7_e3D^ygbF0>qgAZmw z|6U1feX356MjXL)RSdMPyGOmQDAS!Dk(_mVI64X5kuPRH*&sf*=lEwSsjLb^neP{A z*R(v8o_ZQysjnu*dxuHy-f0;3EtphhHiOgFB69qs1GP9T%R04&!Feeecz^Rh(BT3( z#YeK-?cTHWIq!;{)HH@GSUrW?z99;gL{@WqLidp2q1Uvyrx-@8!}%=5GKhcAXXK}A z)3AT{$+pR-F*cdvr|XUE{U1-h8G-kLk6kVpNEUXbP_Th8WY7v_2QV_vodXW};j1^%wU+Po&mh02+? zgAVx5^BRcsNW+4PePq)*bFS&kYNDj_3H|kZaMg++QufrB6zt6cIk{)_TmN@9)GVC? z{bBgx7t3f&RmLlE5v2KNsrl@pqi~`!o4GV42WIcs%WmSkxJHe-R6nbQmib!Z&9)Y5 zrSlM_oJ6Q^##RWO`yb67NvZBQkx6Z5M9^g-MtG~g8=WK#U|*q1RiXM}P-}0clEa~R z)ZH6Bm%JjO5jogzH%gA|twQIH>!7BZieAw|oQHA*IkU2ZNQ(B;)8~ztnSwja`gPIh zapxUlEpwjX3|cYmhdqi+(&S`b4v`UcY0&vT2|AV47&>_($SQs!-qv@?rMOA(-zJ{< zn#?h&!mp_NcsUwBH<$JVTan1{Z|t-`=J07)67Hr-lH-%gNL#}>`e=~^-X9)=Z2b_4 zcqIW#)WUF;<^YbjEI{Lyi5TJT3iA)HA)B)s@L%LXG{|uxci!s4yC3r*Ni-X6W{k10 zeN##jk|l6-h~J0y8{v=7>iA3}m0XNf=4=*Z;)@CVU8M7x34eVDwWrR49bKN7#_%ki zX%s*6XX!QjL-6C!H^G0#k~~Lsgq}}|CqLd=KT5~Y4!cH(%o4`yy zM`v7{gB}|^aH``~j1m>$tj5aVhsnb5=y)AHoo7blYrXKPdl8=798Sz@chG)kJ&dS( zM~{qiM327l+)Swf`Y}tH>oeqiZ5tkwtea*;l5Q!I`X`fmAY$bO%?7$$b!rXj4!q-CMoV#uLILi&Urp9t5yOK#({$p4 zcj)?8j$27wsOIJ)pcC_&yfo)$5gSI);1>-{)M8_B(mtsf$K}iPgXNSpX%YQz7gIX z(k0K=-hc#MH9Tt}Pc5E4W#z@|$$}Z8_}gd@kAWY?{e6iQe_r#9`(cd8i{Nr3S8=C~ zSOE7#0@`LLlJ3{4)W|3hHXoeN9hNi#MofxKvT@<^EcDSJCWM>ya2|JP=N`gkovO;M z(t_Wj)gX)Nu&;d@{JtWK8cP;aVmAXXpU;Qm%Zx$+B8)+;GMF$}fX}yRaVHHd z$coY^;^w}VuD(AXZD*gvyE3oXGS|<7VKEVQ)&309a5kJ8ew$1ZXBy&Ie&#BB){*MI zzeKwSuM4`9WYLKkVbU`~@uglf>tH^`;{5Y8dV6?p^-nLro0*a1>+Ac(^TEr2{s#v{9<|o_q^Z-uwiNg#>9o(AH05?q~A>5*vs+;9N|Iw?sH+L5{$vNVblQ;3d zS%2`9xDr=B*v7UvyE8ZJ7SVBPYq^aMCfKwx8INCxrB_PDFu6gP`*b`FOy@r*bx{i- zFJv#qoO^{qOE~;(@CwFuEZIm>##WAAEWZsie3=JqJ2iCPws~A!4sMaogoO!W&!L>;^A^JKqVUAf9 zY%+rR*BcmD11-2(I~n#yb2wwnZ1hR?f(o?&cz%DFO6fOJk(y`_a!(>FM|}mc?e(5||2>zJU9UG$lF04Fr6#C0xKhp~3bKyI9*R@*YUzyGc?e`TfN zcw8m;JT?bMTNfrvG7N@yhT-w6M!;i?2v<=+*F4yQUNhEk&7R&ok0BmbeSSdayjg-* zyq++#s`jF{i6oSm=CH5xR&o1XF48Ae@|b$!rr>+(Z#)**%g)K!1u1*p5X1j;$fDdj zW_i&HObMsV53BXCxN|CisUIwET!oFH#u)kFHtBUYz(WO}iRPHU28B9A3837j< zll2au*6PB>SXkp&=LHrGV^g3|MVt*S<}+TqCu7z}c{Ka1#RV+)VbcFx7Odp6k9DaT zwDa(5>M`*eyGbF8?l3Q)=QUn2;z49{4NCip2kT2v>}1do->b`3f?(51M~RZ zy`jlUD)LyJ27JmU?$g!~jb0^KDE*uX?G9tdNarz6lc#XHUd150pp%u9Xod5^Nz9_x zpV@wW2fAMEB1V6*{QqjD?jMh`1q z!ohk8vMZ{OVNCbXNGlnBpVLV$G^Vh76*Ec6cr}XGTIp}g_jo|5hj^P;fX`Srxc~V! z(Ke}tY3~7RHBI56-Fj@`zsHwOOMwRor)Nf_G3Vl<) zK-yOoiza#f>i*7ZCQ&~HqOx!D?*rk`calP%)khL7afNlftj0Qik)msdBhgj(EqU?s z6`6BL57cr)aAatJ%xv5XA&N=l;dFNheA!GNkQn|CkYo4knaF0l%D}Btn&AJli->2- zpyZ9W?B!Q0P$sq=y{~vP1-a7H=Wz^3tnQ$5+dfc}UoP-xfd~w3c49Mk$dceKJbcry z)a-f`|Fdi_k(^6F%+F5eJkOk>%n~=elf=)0rp$-mZmy8NOCC4N8{m-xx%BRicryQ! zKNZ!K;a<;kC+iCQLF?rJsVdDOPx#Eie0md4%g&>XUluXVk^yvk`5t=Xac}iMc!*Wy z@mRs{DLp2tk&-!?@N9-IS$x+4qWHb&NUIPte~lyV>K&nH4Sq6b_Di$2!7s=&4NW@s zoCWC@XK6+G1M*VJmdby9!%W{=Nq@_#;fxhGa8}D=RHHsP^iM^wO?w6W%MOI^V|h2$ zz+Nys!tmLvH$?LCd6L+{JI%%XI95-e$~+p6UNTdOW6^u^dYJdlP1{d;7X2biQfI&? z@7dU^UqCkS`H;eO-^t-6AL*W8J*a>40bfP&eX+xJ^zGd+I8wF*#lJr!^-l`P>3x=L z!UZ3Ayj~NW%=u^6!!yaId4cp)&2yUg-X2}9i=)TWEKCY~M)PAjX}W0<`T6J;IeD}a zd>3A)Q!fnDF=>1+AuXK>-fl%XMH!A&`_A}PsMFYSsurW87Px5tEP5s528l9^Bg>Ec zq-!_maZ}cPqLcC^xf$ESVB9=Mn7|D%<8{}N*4CfcXL%Y&H*SDkBO6&m!3#RH*`4{6 z-9Y!B5w{2o{z-SlFJyl^R-m?FJKY;`iXF6^M6!40fpG3Zx_f6k)GX_!XJwv~t=j_G z=x_IE_wO0>!_x6E*2f(;nsMx&EnU>B|20EMmAUjtGw~EwrNZw5@!lWa8TlfW4OA(@ zCM^k0>;sDy6M|V$qfMx{|2?~O!)r3ycvJAF{2cD#a~wC7#$f23JgC`yf)?w@(c1^! zpxa?}GS>Du+0g3`Yd@MpTA_f%4Tgf2p*qhv4rPmk&!TUQGR|_@jF&^65Zznn=@(aF za&i1wY8{YBw?yV(oM$@S?WKoOmX?_EU<%0khf$6HEV;UrXSAp%glycXi%_l$^$+#= zuFiE<&2I&hdF>^=ey{=iGwZR-xsM92iQvDfv&hMdTiL$64D!hBDRVe=kQqKxLO$^9 zs%7fgFxR*evyaxW(fo7v1j_-^F=_z$UQ;o^BMPStKc{Q+cZ0#MYM43Y9&@&}8Wr?RxNFXb>94Z=(0rhq7J84; ztV}EFJ7*U6@NzS3T=Wz!KCJ0p;B`apg3y)H++tuD|fk| zmSr%i%@RT<6LBtLMGRxQWGpF=t*5z@t?@?t0Q-B3DahWLhBtkCnVBOpD05B$TH3?F z`?C*B>rJ5rdsY#n^eQH&SqA@8am2PoW-uN1vtmy_(a!}N;WNK;`Fv9jDyuhQ$9Y}U zHQd7YjB}`uy(?~8*+SP!7h>qgVC?XzpkX1t*c@QXsSfGFvZioo|9O+!e#8z{^Ypo# zS03EIuqV`OY9tsu_(C>h-^RDib?`Up23o~#!zKEuXyP-9?gtN$u2Z69c)Be&=~*A! z)8qyYV(E0N%p&@|VH!8nSDhXioXI&EjNys{IC9Q6o7AkG4U9mtI%(Bjl;MNT&%6_= zo!bmR^!6EauQH>;tEy?>vkL5gqs#rcdko9(A}z1DLbsHq;S-~4+@m^YZu+$#?Be$g zLwCx^nGAb+&+sQ1u?a(~!bWiBdvtnD=6KQhB<4;Ra2m?tFgd7zoc%U~ci8q)UciKN zWEJ_IffOeqcasb}Qe|eSI&j}RoVkW1doG|MoXGQ?=4=q2i z?T?In9!o-(E~ZBwJJQs@yXnPm!I*F7hl7%8RT&X6kf|lasTWyrj`EiMwAr$+3XzM)Lu*YloI)F69Wg#7to&}969B6oh~T~FT{0bW_0a?383BX zOZt^|Grt%|@^W%Fk>>t`F>6G5Pafa%wtPZnc1O^$5kYk8s^es;kRAlo-lgxRBAB)X zQ<*Od5?SXGQ{IMk_OJF%yopW@HBfU|{ zOyav|bm$!4pJdeVp?x8Fo1ul?_iY4cl^3G+!CUNT&OS2Kl7gFpWvQ{N6dH?`qVcpj z+?S6lxrKp~xPja%e3NsSm|d0N6t*{^uvs(SNX(;~7i`CkpS?(j8Lzbg zPx3X`>u7g;GfCgI4%AYzSiwhM>UAiE8PcB1I9=AJah9)9NvWM(EpTRixvob#-Xo6ZV;#KjxO|!xzVHa`t^5Gq_3urGz4=L)L8A z_3H%nkyys;Z0<%ALl*r~HPJLf41cW;#qrlii1GA7`fqk4^=LcIG?ksfc9)qDR(pi} zaW;pW7nkD6+umrS#&8dQZN&hcT`>MW&zOJsjz;G?aj$uY&rTy5Ze?)~n}11^`y{>^ z?&2!YT5H|>A5+7sEmm{ zcjH4b8t~5_bALsn>s&XG-f2orW|k9QZN5AHq>_}yb<-&+YrxOj1})_8z_nvTB>4iN z@^+sD3qER~(BVXY-|r!1ejo81N}-`4;;>FOnObeWNsFar$v%;dmA?a(1j}y=s@)pQ z$x`10_9lPFhun5Yo1>lV@--UVPpJ}mLi`4{vVW>h&-B5fEn~SO2P-jduz(#5SwTB( z9^vNt9-`OPK|J*X=>Z!FJlc@K=dnys#%C5v^nNAAI2BIkXVPbfUXzi9iA3SB0z8U| zW6Ihq`7>3(@7knMY?U7jO=qc#YA#8UQ->VCKO`|VoxL5|g+b{d6k}IlcsU$2)z2yQ|LjSOISaV&$!&eLV*dJ>y<*Z z4y8k1umhJMBx@mc-x|(dOl9(>&4On4G2}*~2#&uoiYc*Ixca7*1dfiexO1)?%*9nK z_SkJ=&iBQUUsE}@f2})QzRjOMZOh@(zb?|3u0mg`7UHJ!Hq`ZUAjwhdqE=YlpZ$>h6^cN`z&y~2&L>xbmQbnAZ%8)w zITPd8sg#?FX^Q` zB@F4Q6~vzpBC+phlV8j7=u>0L=3QHzYQ1n zHH~rFq@9xW^Zu??N>^laIjK zsr7j5^L6IGrUdp_ff3=hJSI{_i^+Z6ZPa1HU)p{vhY_-S#LsRo0wza$|gsF&SD(issoRoJl7g1Br@WygAcfz*8g zRMcILQ{U%F-ksx}!;S*-Tw|E5w8^C}pNElZgDK=UW(oGrX~5#8DcHRt7YFs{GA}6eaa?M*Y%0`I&Q*3V-s?FxTor`Q75fY7p8x= z6|pfd?w~;4nRFdfC68ADnXH=2TpkR^w&`ba>h@J!I+;x$eOXVZPAMZs@@>>`bsxI# zXy>_2(>ZyeW+-->Ou~Y?=q9bVaH>THJ#{0o`d&XVxm|;b2@NQ}o1fpLs9EIvzCZ-^ zb6~Zc0&(w4ggq0Yn2GXZP~%)9qp<5bZ7HsyNs0dS=vpJ5cMr5EiuZ=|&zA~uTglNE zspN?AKHBa2i5~3uLFcGl0@2iLx>&ZIs=nTdu|I8acG?+kMdAgRx>gd`*iORN{0^!_ zwV00k;RheLBrqoWS)}&D84@#lKR*WuM()lIeAy+%$*gn7g2b_0=7qC#{_KOaa!DB# zoWBkIW!uSL;pLe9?jL>sa}{IfBWv+C{}dX&nS%v5UO*h8I#Yt#=JCA6emmqrt-o(yuE&MKxqferi zv+g%TVR9zVkJkqB@0KP%U-?MhFqN2m@FmP_`2mm4x6%09nWSg-3Ibo07?qyQP~7sf zI$Ada9FgMLUb5s zM(05uqaD-0*;@y)hb=c!u?yev`h;wLcj`d%)%Bq|Ee!suYC_xKII6ru)Z)Z}8tCZt zN2gi*Jju46-6U5+bF)Ob!a#5QZ~7%pFm5(@zcj~RuO(p2wX<;d2FLpYA|ORnAC_gr zQdh72bgmLtzdZ)%xwoWUXE(i>W(2NQNI$Il0n-m%7MvFsh7#j^nATxM zmfbcdJMKtE)Z%ezNSZy`un>7wawJ!3U%|fpD%Sl1O z_v4t=qt5wGOdyn1!9{~CdM)fEKE5}Qu`$Q>R;m1X&Ud-4+i)X3iS)y9OP-&3hwQ#z zN40Vyh@DS3xVcBbYTI)5#sfoC9hpHQa}4oI*B9chwh}BuzA|PD!}-qGM${X=#|ViA zVe%g(@@M2M{dnmJ{(X3k{An1+D!H2zg|(WTk3m1@f4PC1{rot%1_TjN=dtjsyoqGp zI!L`H8#5UL2cgR+l5yz{#}*B5oM3W-5M49$X-}rKY%QNZp3epO$HFF|Ib@3+|GA&o z%AWDym~Fevz(VOS{;aV>!_E>s7M=$Fw)QaD(v*0&?1w31S3!v#pIPc|g_rfq=}~)K z+*7<8-Wz(J*qAD3yp_2Dp+jOx{7N#+e>Jv!BrIR>aVwlEE1+M+O8xv>i zNGERQJ7#A?(Vg-$mw_i->2qiF*jA2hx6k64wqojeW(&DmYYKOa{a^=|Lzf=D1gT@& z$(vKs7!p&CjS>q<)W|_HVl4x^hfagsvES4O;?d!e1ihD>KyDjqV4k=weATcZgPW3x z=>0S*QYeKTWEFNSy@;h_Kcn)zaX5eS73}yZWfAROj{B;V`R>|jno$+QM9fO8#-w@V z-k8M@cw3l@KI((3O%~R;E)~JbR#7nNjwIsxS}0pL4$W4)V4M!}_Y~i4bMbJ1z5cjdo;IQs~ldmwNg>uzw~z7E25CEMt9u1LYF%dXzjdIxrbBc z#7#K*--Kpr@L~)X`h|DrJXfHL|MD!0^WSOvmQ3&i&N+9=YucRDDj<(Tj3mPTdOSsW&+5jE{j_8s5L-)(jV z+XmjyGmW|AWzaF!Y_$*l?wC$%qZ(+!{6mm+CW0(N2Yg#_f{Fj+P2T>f#P6l~7#^~M z6Fv7IeEu@RF0#7<9&u)jcP9?#~TEKRZ{4)o%se!fFzC|2_HeG@Xz|n%Hvl0j90v z?=`-&X1?hiX67z}e9bNzV}6R&%~zssy-wgXop*i}t>U@@PSH&_*TC|BcW}RU4wAM% z_+Hl(KA%n|%6n3&i-{ke_-a5my{VyH;ZwLxKmW0P!@@8i%kvCAZiL+FJnx3@>x%}| zk?H^wPED+Z3=KOli|T`5W}+j>n_EN*ZQnu3ihP{>wvY||XMj&Nav^nb3hi07g?At> zgo^zV5Gr5BXjd)-ow|)6C%L``}j#0mD%Xj(mHCReBb442L? zz~zC2w&j}9k^C51w@DaYi96G4Zx3QC?`%Dn-Nt-g!S69mj6tQMj-7XD3vPBBVBU8Y zQ%T)@f|h;t1kMF4Jx6^#4eK`>xcuyYtNy4k{ zofxupGh^;_1Lp@MW6PRMTG=Q|cAWOZFr#uZ^vV?=WHsyaD;~>yZ;>5s#&EBBF8uw- zbNy9V=KNp?m5NRx>)pn}+|pSj@$V1je(F|Ot@D*GdVc}RUnp@iEbr2We=SVRk=s14 ze+h^+xsqF9O3n?mk&`5FSOtzstDiLK$q^q%!E6T!oh(exq$G zy3yQ5o|G}u$nrZYF?6;Ul~uk#e9!xV%lhLm&LdD@yuAVrtQey2tK|XMw=~B-iFxq# z6U_|P!i@#HaGYf*J@X)#itH#MjoVhk^I~=KpVcXfNi)#qF^4%rTe174I4%=ni02A` zrCX1~J~4g{wos8Yyl-GNvRJs#k}oQ_Qn?nMqx=d(u8tw%dxQw5(8qUR#)HfoN&4c!bWZ2oLpt$v38Q_xiEOWZN3A@< zsAr}bXScHu#U6@rKeSG80V(`lck{J*{N(pp)pS-XPyI^i5y zdhQDmiw?zcp04a8r_0r6Ri6_X(R*Z4P)GI8-e%HUluBzw_oIu*c}&~cLz_4K#2+cI zn8us|_Sw-n)pt*5f!WV{wA{!MBY(T0WPBiT*&|Lv4!ot_a$E41X(2f_jep1YFeKZ$ zezTu)bBU<6IOvAQV%A1eFxB!V376dH?xk7mXvY~+mL`o+Goy$M&s^Q{`2h;GX|dJ* zF#>jd3VJ?oCbRZjsaA+oBR(gu(B1s5x`pgxzb0kU?=YSi?sX!y|GDva?7pgW%}Vm2 zzZXvG%mJ(URFJ`e4UKQlKY9#O{DUEm{z)|yBc0(~FJeyw8N+4cvAE& z{78ks=}$T~ZC7Fq8Ybeso$>U_(i7lzK!o$#HwRtaf=NBnZ99Kcb^68CmdXze2WUDsKG7jLlgaj*~XELaC|`@ zbE7eXS^Kn`o?I|MKXjbMBZ&d5^Wc7F&F^hhY@Q)FD{F&Q>0|t)eFgkN7}%FLnHvn! zMm+6`CFe`bHwmS~%FGjB7)Hp=l(De5#GChmB_evx!=)R>Q4<|)jBrt<&l;|yVN*VH zL_{76-5#?RKeuCP$z>XNriHORCyp6452@%HLuwJN4EcA3!QxgHHj;lt(I=aAHk9Q; zw!bGzMdzvFZzrrZT}m#OyW;VadfZ605bj=U!YLj$=4SSu!mTzJxOXA9xubrq_{KC9 z^i`fg<=bOWy|$Jt->$$Bg*ayH!Ba%<=my9O-HSzvVu*jra8Fh$Ln@%k!Ka%^%gfYE$;pK6*^qT(@W?zVt|3F8lt zZO=R5M8!R_p;McF_l#i!&C>|8#*}D$`N$l|K2NQAclXnRw|p1bg!Bk@RKLDu!g)OW z%(mDz6609_a5@K89yq~@{47BQ#a8OI)(k7Z&xL-OEF2h}jyi#px%>ZZhfN~F_&y*A z(|y0P7l)$Az4Q-s>edc8wmOBhxyLcS8YcK!`zzCzHka>-e8i45ZPh<7*g@ei|ISwI zM#nwM#D>$xSen*|oth7@*0vYh*UaS(Ix$@8wG$*@sD@nTGyNQug*P?UwC7em_Q>|p zwL8|5@0NeaJDzW+x4DI+wskT`re(m@gm_}JC6k`ET}`ilxk(L|O`sy7&xqmSxp;T) zSNvQvNXDN#M-1McCksXuXu-aElIL#*8m?|=Lkz$)?*ns7DjhbtOoDpD*${A89R-2s z$jY5liN>Of;4$YDu`@XXZ;GxHHw_DJ_OkmJkp2p`&wE10W!wOjo@DI4SwIvAbg0gT z2=gOJS4mlV3U^Ybvs(K>21&@dL*AxZqKA=*AX8VL@qIrD+|p*!gz@qC{JSq*ZF-Kh zHy*|MZCYgCl?2!}l89G7B*Khq%OLN8F^$@(45g-@*=lDOuEb!5#iI;WICDzOB5M2z z*xfo5cV-oT} zfSzOYU`Kx<)p;|C_>GjqhvLO#PxJ@4eMOvm%+DNZzA%j8;8{{CaIJb%UB~SY7;mv} z@_3x*(nhsScf&h}9^5rn2OS-)=mg^xq`b9&29tU^vU>&g4d+n#(J%B3zn5?5(Bu+d zSrOR-yodX75(#SY#o5sVc+0a3ogXSN-0%N!cBbJ}z2P1Q0S2vIPqyq@$h)6nNT{&`e)Ael_xSA* z$sX6D^SZ6MeZyNhX6Ih)^q7Nna+R<=c_Cdnq8fW%DPvZu81*q6LnYLUXx7hNIMHnn zcH9}nuJDocxvUoc`%wb&)!H!U-xNBfX(IJc`-Gl(%eeRZb#dyA6uc`Ih!#7%ag30` zm$$y7`|?WKoEnV_x_6VCoIOeNJIv?*%}DpBjck%;C;hglil2d3usOj3TJu^CFWfam z#~dMb{d<{HfW^cxp_48uios(i9}o+b0&2WDhpF{XheM$oasQ=wRE57}Fy4yYQ1q7B z%J=Ete%VA@?hRAdBa?6w|F>UZxt^0&U|@6Y8T_5dcMkt_qV-JPWfxS4x|d2do`4AHM}}f|ozLQF`Z0x-7{GbzSXf!k#YneAZ+-|5pzw-(w6C z8x)~;(iyZ?vSPQ^)euX|-OOrf5!tvbj-I$0C0L+5i+-_=0*|7F_@A*ONJXEaSHjb& z$%udGwq-1*$&KPR{F_N9`1O&qc>^f^>=3MP(t~*rj`J(8p_lGVWK>kKBs-SeR48NP zpUj1#7puwm9XUi@-5V{|%5puLHQ2R~;VS0%V8T){QZ13k#)ruX8-tud{n>5soUDc? z-X0WPb4o<)Q)e5eEM;#_h@rRkPG-`qv%yVfJ{h%nB~4xwMP(YBMboP9(jc<}rXb`! zU6E0OiQ5Qies`L=I%g6|;CWaP=QW{A?lk_Ltce#b{ct8f|2d=lLNI}>r8|PIV%o10 zSotf8y&v!arzz}b^eqC}0+;!Cd&WlirnN`V_(v8p+7?p{kHhfwRV4TOV-Q4|Oah^; zJIe2TN1e{-!0r`#EOTlynHpWlH0a%-Pap@iY_;fhy*zemY?i?J#6BA5w;LPw^r`>2ywOWI&to1aR~kunKcu2&rK#l5azie) zd>cf*yieWMSMUt`L^OBXL|m?2r%P5E(;nXK9N%&YSDSm$mDBo(E1w3NGudgh8=oDD|rxq|daI%(@PC)1ySlN_W6L4_=dP zt$c>*^lkbqMnq32|6ymG^&s~X|I*+8>$Qs*WIlg1r4j42a9qJQvS#lU*my%0$8J)F zRP_}#lAGe${VSf5(%GfVsZoAZ{aOk;nB7aaK2ByHzlkEhf90|24lU$D!DmsJ^j5rH zn1`F2e$$5|0ksMmVPD=JLEqP#bl1WzFy3|@0-Yj|jTnXF=RYQGSB)Wamjz?u?M~+T zug6c>NUxjxMcowuV3lkP%?FV=Es`OW~<%tXFbyA|lt3ae2XNJ!eZ8-P* zowz>N9m-cMf=P;D@W|~Kb8h-iymL<(-%fT1|0N?yXTkw2o3okIeJaLObxY&e`{zmM zUw-y8r4iVFX595g70&RCHj1CT#B{1XMF;I`xGQ-i$b3r$r+urTGB1Vkj66$KeWudk z4j(~$J>u&J{cLxfI&^;H-OgV};MgA)P*LF|n!58G*?uvRZgCmHnvxv^A0iFgSpl^! zTVP?$b?9z5T7Bvs>Q3SN@kHwdt6remr0rc#$ z@!9oOx?oK-@+EzS3zC)R-v{WvzvY5ES7jM)dNpdAs$jIgF4Vo&Cn2A{Q;G6y z)IIeDX9iZ#-5LD-`;G;UD`}$@^%3yj2L9b}HH^*=6vts1GqB8jOy-;Qp~Mz z#55tul#;a2#1O%ZZa>6m}-2I)=eX8YzYglpY5N&av#d2zpn zb=Azn^Qu9(_{JWtyuF=X=q(_t)|`gW-csAU_X|bS|HiOWV$I0PVL9T_P(%Zk??%zE zJT^M>2V*3wS0l#MO6iil5jztp%rvS+X2#?dYEm$3v~GN zS+F)%0&k~iXj>r$wnszZx?>V)j9db<`$oW~ZM+LMAOytuX5xHq0o@#*hz_cC{I1EA zs;lSW)yg9bV|tvV>aC}48Fge`mjYb-ewP#=2m68!jlxJAXq6hz!keRC-=p)C` z5LxF>j>lcV`afd0MM8tTz8ueLm5pUqO)DWPzZPJ`EE_@!X2WfZTsn69ab8&TmY8)c z!fQ{g@n~;5Eq6)hM{4tNYTH9%+tL5hdx@ zkTzL@g*UTUt*|6!U+jNGptKGq)h9r~pOsiyo6MMKUM8~sS@iAw-^_q6ur`ygliP0$1bS>$AM?fiAT^!viS?~GNejkE!)&I6Ja}ae3-^2DecgCw!kN=}$mRsSk+Yv^T_4GM z&60t>(v5M#&cR-%ZT}FA3D!(3n}X`fNN{?u`6dM z_`O;L^Z2Oxx%;uuwrMH&Ok98)R?g=pdmq6dpQ(5xI1ioqdCRC>71-R6LMojGc#n4x z?B99{BKsE*ZKd7ZL!V=?q>N|s?31R!M{}WJaVjW__G9ogIaCb_Chc{SoPR*HsKLa7 z3^V}k$PeMB8|Gs!&r~(AizZ6H{*Y}7=kRogH-!B+os4t7N*0`10i*pBv7vY`j=2=Y ztr~hpqj=U#47&*CS~-Di;45mM8O?mZggTUr1YB5S4CxL|LQcB zd#uERv&P`O^}b~8zZ86^BF)L4?qtiSxuLb{5|p#C!%)0T{Pcd(_FGy)kxvnZreDDK z?%vFy6e0WTwgB>X0xiAbNScaWAa-H_JM^=V$j=DJ>g8@YFJGFBF$u@jp>Y_r-I-`E z6QQVY8#DE}BaL}eiRW~hh!7iTo0TkA(_4dYXCFi8D?sDH4)C*)r8~!NheF)&?GqXNmV?#7^P#c$Ac#*{ zOOAKk1}ldkzNf4L3#>S(xS;`D%6GVSr;K)XRub2^1c)j*4113^7uQaBbb|HP~dN3bM8z*~Z(M(DQ@{-NnkFdz%h9 z`|<)FoG--HV^};}QAaAuM-p9&cQi4_8BZTMCs^08nnoyBV^dob`+JoP#u#bBb6vBV zlSfYA`r=rke8(F6yd;TqO%YtL&8LRGXX&97F&z6)j@uXa94{@+gRA>)5b69c%*sOE zr=)wDoXM5Ywv_w;s(Rsy2IcTA{t`hh4r#qGkTsT$jQ&9Rv15O2S)YClcTJjq) zsLJ?ApBy!Wdc$qZ0hNd3K(;<-WL{3bR`!s;RgElXt`4)?JwWtDpK9+|MzcB@rbo{U z9_L)7mTiW_&b*iuj!hw71?_Y*&twQaHV1|e-KH}C^l8Jsi=g|VnM8SvfKzr#=(leS zT%D`NS#Gn$mO10eL5Im)mthxHW@y9VNMop(#4w}BOaPshUnm$i7hT#%awmdkVNPom z8DksINzVaVpnVW`EK%h&4o}42!3)Tvanzh2R@S zf6(A=(zdr}1lqay_$@|#ENC|q%ph=|gD~nSnGTitL2T+P} zquvH-_|W(twT`?(-UW?=fP@1?d7%*;>~+KmzRsxPpvUEA+rit`UOK4PM%{DO=!WoU zI@j2g3*R9IO$K*)uI)qEnRbQmRi6Qu3mRPMq1g~<_KB)@U4@_Jl3eC{2E|4u(qh$d zbe5MD+jS|KJr{Eg3{S0rh+U(&M~^R(iZ`RccFigNEC12_!@pwu+TO2-)1D;!cq~L-1nQj+~>t4Ux)!_ z?{hW?WoXbNGiH40D##wy$llF~hl1hh+_DwzRL?sNYCLLSOXDxHZ-O&@-|fwG23T?_ zKlrnP!U0S>GeDL~nW9ZqB7Ngs&M8mDd_)e5fb=`GnuhUf6;laBiAF1G$@O1urTMtK_qA2W%;ytqZ z^nuty0+%D{1usqfy*(AGs!ox75<L(Qk+pn&&~k+%kv8DiSFcw> z^Joc_xa~zJDW!`9rWNG(n+kd>J|5T4qTuB}0^<(KWA)N4*P0b`vFlL_R$5u&ttUHB zZux(7*&00zIOc#?pNDaLKANbMJR#c4&k*~Ex=6*Q;*O-HG*ELD5&Ak}=Fnfny=BgfO^!AtC!CGE`OF_VeY(`jH6!OtCv_V9PSLwGgPlfyy- zh*j;P0SgwxSS3Yx&i(_>k59=9Gbds-%aBwBt!4T~UnXV|P2}v2RbXxtjw*N`E`D3X z5)E@aalx0FKNyF(%a1|bc1_q;bQuEuo9TD4k@#$$ws2srJ?K95cC_c$019t~ZgnN-)m1X7<)!2HGHps`LBrs-Cb->M3b^5iU05ZB`r z>$c$cs@2S{#bxl)Tpxb=7l4?DtWfRqWp>rIaag_MC;T@nj~4sXllDzk#G|pKrF{|I!$;oDPFlz+rEVV@qsQt@FDm)~+0xsFOq z{!zvrd=vyL1OHL+XA?QOa}P+!V@3ERu!T7ir{Tzbc^Y>$0XipbW2F<^;E0PX9)2+k zCa;p0Poamv^sxe7pc0vSIu z8;a&@;=C2RQFrScPWt&9y7$;Zd~`kr74~joTjILW>uxmu`ey}9buww)T}p3>ev0<| zQ-QlpU)V!g7ofnag$U(83P-p21l&+$4b@rZ%S z!3Zo)VMQHSe=Oj8$MvEMiEWEQ$&{^ z8Ot7B7(!lj@1|k8(m0rG2F3}h!fL-|#Hng445l3*hj*{U+-2v9{i!i@*Y{<3R^}(# zEH1$ngGr#|=7iy;HwfF{gk4k5vAbhR;drhA?ojZh&6SzVYhzWi>yQulcYniu4|Z_Y zny0A234H=NQTX=L8$2#Q1>dguNkhVFVCM1$HnGA2rf&U2&&_;96|bujmz)jU(fxMd z|5%e6Et`aEw)~;b2Lmw2b{ZF}oC5(*4l?;I4pf$R=w3ein0k$k!e@7r(0rv2O<7n# z%!cn$B@auGKP4%AClO1gebfc*_8^?8{{m+Z@a~B;7f!bIBwA|}a*y&?a!r5k5Fe=; zhH(vt0E>&TtB&_Y*Vl;h!I!+XyTE1}JJVpL5}2@R9mM;%;=5NV>@;CM{k8Zm{#eJm zM1s`_;lZUH^u^;* zodQ42gUq!}^PpR_9i*HeP0SN@i7b0{rLKLed^ihjqW5@vr79;yL7i=iFM z6r};XSD(R%5;@p-oM&`sI?_K@KbTRzi&2JWDU}#ZWuiq(V9%*TaCa3Fx0$gpPelRT zg1tcsrqHzm(xQS86(Z3wo^Bewfo`5NlJv<*3Ed{00$Vc&cHn9rBpT^*XJobM{aOKd zR)54Be}z;DU2(sXG0ry9LnZrRDszL+V_$EhQ`9Fy(-A$)FMPPVCXN$YRdv@GwYiWOC4p`{k>WktNR zVKUyU3Zynats%vF0sW?X79P#jWuzA-V?_OBQvX*DFAf{#2h*9AulFD}^Zmzw?Slz4Q?+7=*!%qSpx^^-PHX35WjXV-JGlI6iG(m4& zK7cyu60nWD*GuC8_D4M;y#Rt}Z~je-XNQsJTK|!!VYlGtKpZEq z|B03%gP8Pg4V}Gd1g+wEa~X_?dKYddb2Ip}<(F~zaegr2vik^EW`YWWZn|<$E`AHM z!cl?UWa^t%qOpDsOnH-xyO-Eut-~Q4JYs^t^VvWD_2ze3*@BHbD#?@UVT{47FxdQI z6xgToy`>vxY3(~c%S2c-USf|!4t31!zxMdk@Dp1ySr_H>^g%-|l_*Q4z})P9`nau- zxlwA%+&OEH>fq2po3#E*>o_WOdX3&QBB1#OH; zQWWdM&p-;Je-q<=du*%~htkUVIBEMKC@7l?+lOPQxL+z0$b;mX7eA zGe>OvtN7kbF4;WR383^dJ^Jzzj{G7C$^NZ^>%S_DCH_T`xO1W{yy4cdo3)MbY)C)(Sk%f!1Z;_XUonSk!$ByvrUU#a76AJa{Iy%x%YP=EEA64_KvN@zD+swaX^dR z&5Kv4xne)GKOfJZcjR#S5*f&O)W)PLEG1X&#Zacn11>*}W23evl1QmNbjYxR`M5X} zZmWCX$T#Bjk+=>${zixEdL}@9wFf9Zq7OVOcSEtyXkq%Tdirrq0$Kb~hU-~#1jKZu zV7i+o&E1_$lcpZPl?uG0Va`S5hJP`$LX=T;=PA0H|6SWSZh*q3<24d>_OQw;8s;nH zV`+gaO0909Z9Dn}$|KVua>Pm;E_%VVU-T!fL(25|{utU(pb7WHlZiyXGqrz{!pf-x zq5M=k(!Aj{>A2KKTx|o;LS;5dsq;-_{VIf+5q^|fs!Ou<$v>Dq;%{IVDC6HcRs3yj z2kN_Qg-7=P0x$8K18 z@d7Mj@<3{&BKT)|Q^SREaPMz6O*yK7t?GO6df6H}yyrJ_fBGo$sQf(X_Bkh7F?omZ zM{YiK>$if$bp@nw7oc=)8u7~t!-1?9koe{kYV;tWPP-D#sx6qY)8<fT0{rTEkc@LlrWa@2#HvXa zu<&^@d70qICbZ`8`LhHF@7#lykz=7&JcPS{?+T{OI)Kw%d(dw|6@4_hi|r`4 zN%Y&kAbPI*A9-eR2%b#5!K#jrB(1MoV9PQWIy~bSDUU57sX;<=belb07v)Td_G9bhIKiUro$ZgewWc{{ihFBgYdy;~o`>F~FY1ly($6drn)mlK=Sa2*K zq+wwy_@rt#UA@B>LnAgbaOVLjn^cMSe-_i5Zw1)dDlXU^VZq;>D)3gC0k`VE25Q$c z5zMoGiZ*Q@$<}qMVr{K1R-D;NW(4d2b3QB=`sFCy;B6#db4T^X1dH+|JEtyts!3%~?bW zE0J7CVnlOOcX5JPWsGQ;h5kc4Un{JcYI-@a$$IWMzF!KmCC?MbBs1>qia99%z#nKM zP~YEUNYO_nI+$6>_KCfrp3zcR@X`+BdLFV3qcU(oVlk_q^_;cQI*f-t>u}pt>&a)$ z)x7^kfjpXOLyE4PCU1h~(d9QsGCfZk$lKlR5VHC-Ewrm9buDh(_V8$Yay*8f=}#wQ zY9R^no`PdO+k=0LHMD&F4fi$|T5H0arg}Smp7PMTao0m=Fc{^jM?+aPd*Ue`wC2nKChZcUjEhZ=r>7jz0 zMsj+?0q|LE$jMZsU|{P^dg=Zs&dBZ>ajQ8&?G*AMMCT3(unMOp@A&>lrGV_%cboo5 zO9HViDiAQ{GUFB?290MGVcT3s;@t9%wF#a<2NcrjpDQ*f|DuALR$N494SBqExR7W~ znTleobhs&lau}=SNap6l?~I+eqMBgT-uj5jJG)XC zKbh|Q9YO3=&k>a`xiG?X3N~EI1DpDTj9HlpJ$y_9wcG27#keHgd~+`DXb7Qw`zx5A z+w=sR-U}J0YY&ObTHa?kq)R_9&0;Y73cTJb0UulEWBQL1__6FE*z(+fBdf1c8Z-(e zeEwnQ{_n)UtAnPmEXVulIoRKI3P0NlxiO+G)UCFio+&EC^O`Z7r+y-S(b@oY%{OuJ zd`IXv_NIpqOn@VtE1V?8T+Yr&I$0--k~Oz*r-}yT?d93N`qS~roof*I+KT#p_JlLW zv!HNkIb4g9MNP{LOy8?cT2(9Pi;j<^Gj%()Sg;Jv@-zTJTrqLjoCv0SdA7Bd70%{) z;&!PKsH&GihrE)=W8V<)o>4)rNqWNftBcUcz!V(%uR?B8JV@+$LzSP#Q{9(SV3q57 z*z4mBjt2~I+sr5Ig;5)!&iFVCR=Yx^*?fH3mqU%RzEhc~gT(E$Bkt~BE*e*T43EZ{ zli#NDn77;zZOv=x#bj-A(&-@e4NHYzpopL4lsM~~W7vyK1@Hgd$DDZ3iF$ciaQ#6K zx!*e)z5F(_(`9FYrCAkd9C?P5exIgkh7EMxv^sFw_zF#ACvrZX_i)kIO=$Qc60h1H zMp;PQfFgON5fQDG&LGk#mk9ju{%!nlfZuE9e71U z4$p--kQX;qF=z2STKHRuR4g)M@9l|)!?wqHx4>?8cX~W#?s8^Bm3FY{2g8Y6y)2*G z^h8nF9l9?vjOc;|$WNPug?*RF(i0}IvP72aTKbGpxS;E2ea<#vRm(lcIkl zRN!S<0leOS6qn67Mdj=a$x^#nbYP%?zA?%}DTRq#T#E#IvuP^~J-G`OpN4$_&koyop;52&`Fl$3A>z}WBut48nB zO;O57R1U*fk9MpaAOZh5CJ3fC#a z!Ie7Lrazj?z9}iXReuV$eiGv{x2bW>5(h9W?Gl-ybA$0CqoHSII0k1_!ok0{V94GS zJ}BAZuGvNGh#@IX`JWB_bkUcXlAR#2NtMs)=_98e&EX*#I;wI4nvz7uuje^?<54g< zz)q)A)@X3Q?dovt*?fffd{nB+Brn&V#-Dy_;Bes}W;>X`y76*gH!uaCuR24Hm7Sm& zAy26&a|b=?J^_u^sNz}OLfn2{#6O<5!dIC>T9=%`tPTGpaNV(iSR}6|`SyR=WEXAN zK43^PO~%lJFcs}0rsL#=9#}yfVdSXeaOrIU#C?f11>?5->a$xZ@3r2j3FO8T! z1%Cx`p!3lN&Sl*s200mYUV#HimMUTo&3j3I>TE`tHP^|Ts^4T>Q2^9xr9#t{dWNVj zVQh_4;f7Fzv(DvkJ!%qk$x}zL5iX#LS59EC%ww!7cV#Mnxsh+-oA`?BNjB5-Ar=07 zBDkkK%*tGRi`Lq$aMFARRK6*P?o2OKn;}Cjti#Pk*L5 zUX%7Gj)qPC=Ws=+6aMbj*~cdzH|7i488eZ;p|{s^H+WfBax_hjoqmK z7{0e~e1I$NbxBN8Sw> zvk%_$bHgcf1>^H(v*QdW(K3bW)U0+9)${#GEt8FL>7J$R9I_E=)pF=Sk2&eGx`oRN z?ohQ72ZR;*tkgzly0F`xX2^`DH{LFX1)PJc5EPPhcuF4+a2`yKh zL{dY5curY~nfFXlT3diRAL8?v=&fn_|240$CIKR*s8ee9vD9yXwKi~)D%YBl@ScN;U>WIJ@U&t&?F z%BfQLUlQ{401Z29L;8a|>^zr6k{OG;;B3hp*c?;C-nfv+NR-^bMJWb2;e$LVBq^iM z=4Z_|Y4HPG8;0b}Lm;o=Nm2r)j1vq#4=%?h>L2pM04LW9pM zouO_!V_3h|_iR>-I@l(rbJ8=Waq+K*ak2e5`lFZ7OUnBAy|xwaf6u1h-?nPEYt8CtqR(Al9Y@eaVNZDSL*RRei;D9c|FpPlweNuSxm^1Bj_rf#K== zIq$uT;M9YuoXJ!j97_5@&R;Zu2!lLawbcVZoEKo!@Nx`KJte~NYEbm>H-*I#P?fzA zOhg*ohdFVyG5!cqP8Xx%oE?36-j=Mi_u^ju*G-Mw%CO?rExLDr;M?PAoY@TnTEsZe z`S*@7&0*zaFyI*C%6{xMlY}vc`pG|i1upR@&$tmaqUZ2!)X~w#-uRcKBKij@d0j_~ zk4^@M>moL@Z3gb$F%I1-c2LiIp{V=ToqTOkU{~raCe4dGNbq4EL*ER2%)P4rf{qgsszI{YU~&l?TXAi+w!_V_$j^SjU3 ztMZ(Fa$=Ad!7d?vW;%tFUSdYJuahP1Nxj5j${UQ| zK9&B;;8~S-y_siA_`HF=Cw=f-|2JBp+&FPIAfla;fB>MPU=IHwS#Qbk3 z*}BG#noh2x<Xo0c$Fk zWoMO=wl$`x`ezb;xTS_^A`fgl{Y|tat&@1osD_kt<6N%aQR_NPJL>~uIFc^oS}4QRm@CvMZiSjf%FqZ+Mxz#Mg? zo|A*=z;Ri!p>I9ttq&w$7CF+Nc~|k-)M5Jkw=~aCT0jmT-p_Z(&a&d-rBv&^5{6rC z#_LrH5ZwBg=8dQ%1)o>5Bx){RX|Th!Iqg*ZwIU5MkRYy&jqKsAR&aP$85&<54PRue znBe01$XN=&Yqui(voRE0OJnHERdczMV|lLDb0s(^xer%~>T!WD?~rp{frpB9(C8Hd zL-&&)ZTTb8{Z0cVLyF0$&->Xd^Gon+@n0f1f0!7ryoWI=NPGST(3(b1DkiT-wOqns ze_t7`mGmXWGILlT<5=?H=vp?P@I9p?`7~yxBP`nd2BS_P8SvkNe?xf=*ny+;-ez|) zAfpVPx(mRp;}SFUQHJgyRv_9w0j1?8Vg1MDv^+}}nXVpeJ~W#HIswsd@e*TB8SO zTd^&8)NDh~{%Txgc7=ExN{)8@>eA?y&((0Y#>8VsP@Oa+2tqv2>X1BHQ0VD(O0LdJ_Tj=O5{ zVq`9LblpgXwbj{_yU(cl5hXMlvJ@n(nu}+jPrwbPab&&RNG|i|0SuL^B0rw^;T^+Q z)OC9o_5Sk`NABQVfM1VullMBHOszGNv+-=y>uXSaTEI@aw$|54nVAIS=N)AC&l+Y- z4QFtRj(@~m?#bvAa23M_=Ws7vO}Txy=AgdZI9$E7f*WyFg`0)eM0Z3t-ab}BdQVld z+h?W`>$B=;uAE_ayfdGpS1%EFwFE%xtMKW?TueP758C05xIUBb=1Ek-?8SCaz`H6A zrcGm>ty1HTj7mWBvUIfloj~RDw!k_$TM&#K&)*M9$cwjqaPrkmjB#b?tew{Az+L2a z%4Xq~ZK-rc@ODx-;VMb^Q3`s0(xEy(o!_-)!Wrk)_$^W%FE=w->%R>W}jtWWnu=@inS#(*zUGpT-9dIec3-i)Qj8ruRI%z_c+HmaOZhxydE$L*KV- zfbs|U_Wr%#$>jpbJGB6VdDj1bH*?X?X%zbX0e-<^LWO`86^K=A=*;ihQrdy*IuRI<_ z)dNPN^ZPb-XW|67f7*pkzm&kH$7Qh*Obi}z4da3*Y-3(0`9m|$sPz{2)t<_kMdu-nxAd9KIw;j8-gz^f(U`O!|ULEm%6(?uy}A zOUR1<2AS#r7tXebcOj1qWyzNa;?4 z!5=5cvD~r5PP)mnO%@$z6BnKr zSpU{yr{29g7Pnn576(QuUC#)xfbWrjfv9lN<{A_uTsNa$2 zS(0yQx_TP7q5B@avDX};w;6Na4@|%r@(Cp8^ASw_(aL1pbird{AKBhEe>yNBji&lY zVuhF@R>s6~e#|XSr=>@5IcyW`TvbN<@A;AmYZ>}4xEux)Z}E4{Y7`Ddz|gl9F#hvQ zRNY;U6(em?uh)R?)=8khe`L|tA_vjjN+)0(e~_&o$AesdGI7sY1{&K`gbU`4lv(p@ zX`fiFxY+-GEtn@MB_<{#Bldqj>SBzT|JH!DAv@Mg3-k~3pJrh-!`$55(sH`9!%VSd z|Ig2}k?-L0re2Zg@K&Zw*AhJMe4~~ZKM3>>K1R2ZmcpwujG&p|?AKl2LHZ!YfN@uL_PfH>(P7EXBK4XMNOD&MjiwFDHmFS>eikX4{ zFl(}aTjTqPW{EwGYaYdg+{$KmpIl62PBoL2bBxf;Qvs(`l;QNY%cQSVm6Q6DhjX4C zBOaOoXp}QixZ|z~vY|&{M)EA~TmLBh{8xdCH{T1YOA6VHP1; zr@KZ)G3FZt@HHxq$edq7o2T2ubc0F4j3@I@(R&J`oDfT8ZF#rLi4)AX2zk7I(}G!j zx&yvyF+%geal&tPCuvD*7`8?#;_&$v@~I|(tD5l?e1`5bb~BgYmDNV*ul=1VcQwF= zKH_*@yAHENnG{A$M3qm4IDW90%RXL-cSk#M7VpM#w()s1zN8YrC(nV0ll@@PsLNDM zqk#tAIu1adkl)CmI>{W5`>lyDKn{I9$2hu#^Bg?d@1h)$E6&I_a;yH zE7GA?D-XeGHGo^k&%mmZSLEntGvVj@oe;8W747_40~s-@!tp!0=`A^TFmM}(uRjz* zz03>dgNu&vpu#75r9K~mDnlVeMA5-Eg?U&7@ZjVq#^9HtaH6L^j9$2pBtOWa*M=v+ zwHyng%Xd*f)J>sNdM9CoaeJm)MCk7EIu|Jb}uX=J@Z)< z6Kc|L$2Y=>10#eMD-*!Kq>ReVJ&$IA*0A*PKGf{$f%$<~(BkoE7&-bjp81!6&v(h; zKzI^$yfOnOpB`i*1Mb3$C->n~*8?`J=OkFnI|Fmc0l|g{bCNe%QkcT=nfN$ax<|DK zyaIY@%Nup!q?5zoGh+@oH1hW-Q8PEB@*h^ZHDPbTd+Ib-9SV-+@|}1^Zalwdjs1{@ zzF&_Er-s~twG#i}S3!X2;%GhtUZ02kuVtV<_!WJZb)VL4(SrJ}0k&~x87b~OENEzn zp(m!CA&c}P;r_r)qHfKSC09-`%O4|SZ?s?d>(Wb9GE;(h(`c~t?j(x~%;`4wBe1LF zBt7{j2+I0P$aA+@u%zJ~$OlHk&YJVI#_=3;h40Nt?e_pKy_h)uPR8U=aUtv9PW=A- zVN(|Na;1|QdcQn|?rn)=OlB;CV_P<3;3W@mxg-VZPrQVUhnC`m5$EtNoz30)zFqkB zXda#St_xxoEXFPGE|A_%4-)y#7PrkgOuPi|{)eLT4#e_%!+6NZ$c~UvNrQ|c%lo{~d(M4d*XKk0 zO+u-uR1FJTrU6qFHVO<53s5$lhsndri1T|L@;gU~JnXTgJMEO9@A_zJA&MuHX9`XY zk8I+(V+kw$UCfz#9H4c{x3QzRnw;-i!)_nc!DTZ&$d<#ZD2mgEk`vZi58L zI7)E2)LER|ycAwqnc{-bKumZcWSC6_zTUwn%xtU|`}ZT3>Gt1(1+i7k$NCc%D_WGo2>0QM~=Fj4AB z=?!^$t@pU-zuM0vPB#LY8>Udxk)v(eR`14z)zz@HToHq!w}Fdhv1s<1%gpq<5gy(X zPmaXpuvgKO;M?S2*seMQ(iFakZpL2Zs)oKJL6>Flx9b_uQ?%pCq$A+eli_@A)@RYC z${5@QC)o=hW75>pf?jjZVWmV7biVikBMV&l1-1d~R`5km>cDR@eE>+glo>vq z1h6?rsQ4E9IUrGS~qtc$=`|Lih8ZFH$?I|G63Jzq!ate88 z0>M(Po%pgvz@^m_8@Jb_MWPWd4K=`ZJjZw=v|UU+cQL9Bt%R-YEa&#y4e#2y zz@FqL^42|nh(PsYzkzqw0!rYu}Ta3MaDAV?FK85_$vs#Ax<3U}4{s>8^_T=d4lbRuvg#wGDK9qq%mEB#4;R1VLB3!Ttwe$IqRx zBO;Upta%7Q?j!Nkz$G+uJ1APaR*&R&S@DLNA}CWo1>Y9Ak=r#3q1IoY`+9vg9{jVF zws~v;L(>-e_`V6BTK9@Ewe1)oaQuG6<&a^Ep5emP>U@)1Dc(=dBofnCajtBr==sSK zfhjVS7EjZ~0l`bK!s->-(Okf4U^w6WOA*%^3z>`E2H;;agf?k8lZihs!GUpNwC{Hu zI7~fEExk68UsH!*Sy?cS{x1^9b``L?nZ(47Pv<{w)xp`z=8(B+&xmwIFrNAtLXX-0 zi8jy=)2Emy^t&Is(!>q_ z$Oxl!k`dYgPiGC`D{c@J(-GrEl4|r!VG6NdIi7~WNBCV z^mzhb;*V*fd{R+fu=mMpAny^3OHtt@shu8@7&5?fs z@YenevaHRXKbDyR69R5?`%DyJ;dd>#wBHI0?beZ&z~dx~_hM?t?V%<)9@ZEO9MK>aP0IpNWv(-uq_ecTXT|LM;HweAAG>Xf-9zx;BJEZ!(9E{m6+&3p?aS!$$#DueM z^tZwjSU7DKJ+R^mUjOf-sAy0Tw*&}$VJKk{-RE)E{Y`9~!#nJ|9|N}5&$)~11n+OH zE7V=l!LnJhtapEy=yACSRLM;;)cr8T$Vy<&{l}uo!4=NWhhmuhZ z*0k}z?POLP!wXy?ZLGJT8mH9~H60eI17O7ISWO zz?5Q>sB5`+y-9^XEsuK-qaF)ArP>^N!6}{}H8GZy9-jk^@~5$Q*lakoMx9y}Pea|~ z1>}RdFlQYt!Q?mvcEI}%?Dp7$uahH13*GBUamsvp!?}#h%0s$+mlp1SaGjmJx)@wl zQ@|}n9!ySe!R>dFi0catvaGrc7bIknM3WJ?hxCb>KL=p{;BY9pSqsWC@%Z&k2Fvoy zfVW0jcwyybb|v&F)~PgN&9OU7#&kB`{r|J^C)yexHHUQ=@6>Fub-%pXUuRz5v<8t2|4>ZI9Y$eozQYq z=)7zvnHL%$&#o6<_s@camybcpC!g$ow}70Hbtf`fk8!u)9HB?Aq2kd6?2@**;6Y0S zn;Ey+F5!D&w{j1u#iw}epBJ8O65bDPlc?h#GtscB1=Q|WFMB)p2xn`r#cpRU;6b~e zjkG$04taHCabP06+@MW*6eajsYZA#d`DT&bgI3)1$^=vI=HjxS5g2~of_Nm{gpAde z*mL3>>HZf-60c-pP0?=>5@m(W=bu7N-!jpBb73Z3^A1Y))Pm0K%Qk&BF}UJHFiHk! z^Q8~vAuUi0Pwl(F?B`+SONy;T>rlU!uU@H2cKaJI!ZeA&fNW3Vd4m08svZ1{(|)f{&1;4*NEp z-u6hsDe1>(wY(25@Y06@msI35ms5wX88G491LC1^8>ie!#$nYeJYJ~-J4p%A2|Yb@ z86vpSUkMB`g)m~F?7>jp2#3Z|e#f7YyzFWbXOgs-?^qhh#%Z}hm5RW*vxp=nGLzT} z^ypmp;c$)~n_Ozh7nT@I2P4*ri$FCNsU|7_EM{~|#@ZT#?{YuFFR;$qaw{ysa zM~NhFk_~3R{s14tCE5Esg7-*Zs~$Zv94du7c(Z;PTXb9K*55W0S)8atCquy{Ier1i zU$UV%PvD&$4r33_AA-C$N0>^0J>)7Kz`QG_82j`EX^mP2J++efaI!4jwp*3gRc$3l z2ajP^!2oCwA?Nm0$RtHPVJF_z!*kompjfKN*Y+i?NNo6hYYi=h2O2g=9 z>mo=L4?wdEaj@oFCiD0A9nNCVEG&=Q zSfGO!Zb)OF$5D)_&arVB+z*Xr!oH)=2ttn*kW(Y=$W@`6uw}4`?dr-CExcjLZ8AE} z&}A8M@Oev4DRtM!7e$lu%Q_f47>`pvjNr#$FR2`6i`&{NS&5DU9i4AWc7{%*=Km;e zJTi~Hac^TMcGfZnQzvMN7V>$QzwRdE9ZeJu1@dfrf zm!sXRR19!(6uJZJ*-$SD)NX%6VqG=6xTnVVi#^5Ht}!D2_*ArB z7lkoKmq>BJ0oZhStLTu94EhLM-6?^MWJq8P`R)3Y#RogmAP;faI(0pFVT2LyuN}b( zm)Y|w>YDXZKfkl+F)1)RxecYwmtl^01}+@FkF&aG3~NUJ;f$XPJrnPbZ0PeqD${SmlGXY z@PeI7O@j7-LekTq1|-4@2F66NMbQ_b)oVVBOLP>4rIrzWr3H{T=eB6-l%D!?+Lm;( z-w-H{+ySX{l4!TzaJ*TN&VAps8UL(3Mi#uRWg+{DxvC?-MCA$#_)BF^xklF&Fzo0d zsQ5dd?uqk-rL!Vw7hR0EL&C_spbMmOr8}~*{M}N z;kt)Ntl38p+g!q4&J&(1sf~Dt8xE_KjrsQyVbC(NkQ*56Ad$w|tmI0ZNL{=Jk4Qy` z;w$nPS2KcdwDBgt*ajvx(VKqB&Z_r*FqbVUnZ{hFKV)Vl&M;%D@CeW&;WVm+Lk<#!a)Q8<%$g$bXBijbmYzCO! z*Qab9QHRuLVmQOq51em0(|vcZL-@-On$fiz2UMmI>7$K!Gvgr0wv2;ycipLXb2MrW zdBt=(f3P7Li`hM~MCK?g0ykBw6H%)T#@VEkO+K6HH+wbwEEa3yp`VE_7ub`E@5$VS zC$pgcZ7w@8y%3rYMq;i_8adhqbpP#X=;`{0Rk|L=^I^qOCazWj*d+a6?r ziJHI_UoMB)(Izav{1YC3VF+3;=7P?$4>;RuD8JhK1twX{1dJHM+ue&`+f>U~vUw7g zE9PLGA4j% zKr#3~2(PuWscat*b@ZGN#m^p1wf%%#&S7hsJXDYFut*g61X@ri<%chg`?7GU2^Uy9F(IPu4fEN8DK#Q+(*}cGF~FUh#UF8a zAW8@r!RvGwQJ>}8VPxK2u&h_QJGM1F4_}3R7hRi@0Arto(9?g@VA_?%P-|O(8PM6d;@%aXV1NF2t@A6;`(2y7sp3n{cu{S_n?>zT!8)D!7<)WOZ zUF1{j8g8bvB~Lw_q2uNi^2lQ-m5V3>g|WLp$F?1|&xi)a1-(RE^j?%_^@p2!Q-PU0 zSpl=(M8FuGOVG+qr5lfqBku-(fqCXTjA;pjr}?Eg;Yl{OZKz}+HZ%AiCr43l3q6p0 zcO5p!9mVEU3Fz)0gUh25P-Aoo9{*^}->xZv%Z8h9>4IVSq`i%l4xNUnp+WFz#3A0J z(4I)IH06iyoJAK-kj4LIOW|CNOAt3l7Qe4F!g0T*@Ga3+a4OV+w_E!Fm#yGQ!omc6 z@$Va%e)2V=i*#s4$O(v*n1X7uVvzeom-4eDL2*k98F45E^5m}*xtvn!_=JP=dn@3r z%S*7VxF(u4koE)N=i(U;(m)i%*kHmEwm^L3)WyN1~oXaaK>!L?XI($uangvg9D@jfvrh6kLXmS9PM4 zr5R8mR!1+72!^uH610DIBkn0vhOKwANbmY4bSD8gB2t2u_idyZcNQ}9dL`O*J|BLc z6d3LW4w%;^!AF$8!I-`z{3;oXt|=y@VvZ6us)~T>;bkyEybNyHZvayPvY>H7i);$h z;`_&XdLhDfI{y{cxL7Cr=(KhC6_ zTR9f{yhjVDgqU@~?DGu{fBPSYg88z*pFc!OkLfb0$~;n0?Zo<*U$K#Wa|>h#r^9V- zG+p^Q8VW*``D;x@{9t({z7_iF+5yL~PuPKO-C>2}w1epR%n&BFx{g`aOoEsP;`Fc7 zP=32l6{=1R1KsE)(EC+|&G(DNrDb+-@@D~j@Y)0!$Hl3FJ4egEjfBs_%qh_13K?UQ z$J+c?!RF)nyseiAt*w4RY19Q`eXWa`?6Tw++O(p1?-4wq;KK);@gxtVPC`*|yy!f4 z0;&z8$*o*Dew|4UnYTso+zj7|MQ0q)L{e}~2|n4MbzW#a+ZVGEV(O<{EQg`XE|7** zBf)ONW9$xS!Uu!9_@Km5G$>l=*YZODdhc7>T{MB#x2BO9*2}?W$UXK_=m|}$e#6=) zuOxq}zmPSrwbA?6GDvNUMZ3fy@X$pMf6f9w4_+ZUa8;9+%JwGWdX!h0 z(aZKa7?8yBVrbn{kM}dqkck=wNHUaQRFw&8TRwqLf@{OAQ3rRpDDZdnbnrLXjs?$B z1U`H(n-&s;*@3G_iE$&`I-*ROsX?W4i<_$@#H+QmC)z4BLg7#%u*by)(D1C6KTTx7~&M&2O;XMFsZkXObu1X z(yPxQu4ofjoYbM)i-yv{h_a-m*k_YkBgj&$K!=3A{e(o3bj%Ij)w?-AXg>K z4UneQr6=hd{X7_6kOVTZir8y5nH7nqf$8QSm@;CYD8geKtvBP)uBwVS?ht{54#U=8 z18kvEDd@?WfkJrzo1bUQHmMM5@hN~T?X~1DdiTLF)$3?<#RVtMNQ08rq44LA&|&kl zVqW*pvuQW`xjRa7Y_n9NP09UzIOk+6`S4g);E-IzmbjB(8o2>ne`xYqPfcK4qX$GE zlz`ZThqxv=QS|OfHkAF0!qbLnqK4o&Xzia#XO>hDTchDj&OsM;8L2Y+AB)+wWpim| zZy+u$x1fg$hEWy$RL-Zl9_D3k;uB6g@P%KO^1I~H;B|N^(5hTe{dWbj7E2N*p(i(% zloJ206p`f9k3#14926S;hwP;)-(WV5KR=)j-Z}5F*7^~ReYb{9)0d;*^8+Mon?ORy z#6J?~F8ksymtv3@!L=LTX~$r-_ili^P+)}-RiO^V>4kOcAjbf~DLTIeFV z;I9*b_&dl8RSmmI|J~o<8pmVc-Fes<_Z@X_9LEGx#F>vR;q}tlqE)Jwna|?qpi^Z} zbb(#$+Rp5V>vr55ybf)nL$PI@B^%jeg}07P;Pt5@=>GL1 zcb*{gKa!7ybqCm|pQoV4;4_M;ROhW5 zE8l+r{hnsAqks3apZ>e?ty49)PG;PS&Ta7ADVWKupNono2hn}96VVyD7&HTH;rMBL zOv(NV-+o^s;a`D|aoq#%#dk?^uR8x+_}e+-KjEU8kyNhRlneEEOmEyc!|rZYqs{-? zak*wXyiZ%qIySiB;YD_$e$|`I;^Gn#-<6M}{s<(S^qTz;(aMS-4ly5nKX@4a6JiTyedk{*y@|KaGI*-9q zU`Y-v+z7KL6hc%;F=$1^L*cX`{IE&)$*O{2(V{;yu=&SXVMv_M%A2a#lwxH(GA)9I zzF3I;Pc3MpkvD7}qXPFm#Bu*6ZIsGw!C_gYpgqfw^HTmu@?RNJ|FY{i!RH~G&RGec zEVr{aC$#AL%M(Sd)yr|YbpxhqM}lJfN_N>k0Rn#$QJ{XJ$Ok4sLB)AEwtgIZp8J9Y zwAJ#*WNld2tWmgFV28flX2XY8e#GNLJIEc;GF&LcBfWhhFmkQ%d!%DTQ}3sN!s*2@ za$Gb_SO1B{vohJT9C^6o`4YBSyD{gxZBR8W5}UTW^ZK_n`4vGA;AhNwyf9aVCY&3| z?=LMTWu=z%-#vBSPGdCh8#$4cd^p6uT%8XSy5#uEh(v)E)Wc=9RM$F34xz6@qnN?e z1g@{W3(s9zCvcD}u#8#}{p{snrS+8TKVL%@|60R7o9u%%>xQtx-+8d>-c;WA#VZI{ zzpQTT+7qKeDfjTe$S?3TR%|NPhm! z#&bI&_$j&BWY3?W{J_&1%owlF44WO$y8kWisxaFEm;SK9AbVoc&0IiyaC0BwY)*e-!Mq zjMW4Dppx$yuu1p<;l_@pclGge7=Tj*bn|u--UaQ7V z5FL0M6HDHf4;M}RdIaqAMpNG!Q;1`GXsqND81pfR(SUTaZ|+OT&Piar=`XncZZB0; z`9d!0{R63<%S5c-1f?fi(t<@=a3G}+2al=Xj%%~%{e%XnlAcYijOWA5>8|)``EOCv zdPjC8Rh0x7Skv8>l3<{CkIAf3g2RTvxYsKI+Qyt0MIHXi&P_ap!xJ8p1lKog=Dm5$ z|6>;{o^p#A)`hbnZDPVXMMT^*>q+0*>D*eQCeggx-mvrKbR8%3;APn-@{F5@hp946Q6-oU{xsgN^GlZ^T`6Q8UJ$JrCE z;d7Bt+f;f_Mozy@;a#%K0%`h7O% z+bFc{aO1tsSAbP>OMQ*X8L$g`z#>2028*XDuyoBRxL_#sShm^H^G^}}a^D2+%VaQK zG!3LrB*Vh4I6D4n09zVW%)-T5z{D+viF+zyT-0F_vRspO)!Ac{hXgCoOsF?89Kk-F z&&51VX*w^!2FfpL(WM5*Se;=xdDcA~{+Rw_#wSk_hi?+VdwQ}hkE+NzXA{wd+Ur<9 zDuG-U-p3po;i(+w`d{19P{~FCbSCS8n1n2C+Orkzw_L;%Eoxx%B79G#twOvQ?ZwuVe5p+RD8NB_%-fhwIxeMT7Tl%Lu0Ri-jF+K{lk%7|LJpc*HhZwV`B5IIcXz*tM2W!6VX4V#~B~!_X#t zyLmO6D&vApOOwImMF8ya4Z?SMUXU!Q$M?#v$F<${cxt{0{XY9GS$OI+%%7=>oAztM z;UkC9K%TK_5x>|EfuSJfl!mhfci+-(Q}kLcaNx3}xD3hvpmq8w@@!u>JI?)vnJ?2h z);k)%Y%*fWldkYjZyjT2LRRDF`T00>ej(e{c^N}?d!XH;0ruN`B)`^Z6`C_0_>_GH z*NnSM=f-g~cG+s0YS4{ipXQLU6#;BQZC+Bpz9F`nc}WFw5L!Elvoyx_PBGgeH78oxyRF)7me;lB@Cxr`C3F;4pvOCQsN0d>NAR#J@HmzPfX z`<@jp_!gpps0$gAD5f^#9A9owu(W&M;O|D_q& zgu3w!!RB~>e1vGO!#}do^ePyL9fX+Ii#WZ53ShJ-8^>K9%kTJfl!^I$$Ai1S6J0Gc z*qree#4_uc+!c9p=STuP9GHSXx11CB{J%-;dOMONZw7f&|G~OXQJh})TV@#lhs+O0 z(YJvH(s1PlyUj~N;({gURJRd#@4ZfTOb!Ayx7Wu$dz zT2wWZZh1O`pZ&L(rH>m*S+^X`$a})f5|!!ApXAbq$K&|L|1IU)?^NQ&Gh1M>Y#Pye z5&(hTGEl#v1wLDhMefI4<|uHx^-@c4#=0Zy!s+2?@Lk0EG}N#{Is-D#?ZMoj$K23e zA#7RFK4_SC11dU7>d0FOurgXfGMl}*^ubETv+a31}e2zqj~{q>f-+BqIJ^`?<_J$2~+AaM3<1U`=4VJOCQBwv2grQ)H=;v5?GW=g9&JN)@qBwXn%7ytzP2)EboEAk{h@eN zV|nCM`xNNu8p1y@O2bJpTSZ@*p5T+Kjcnz&BmBLM-k?}#4Cy~d(q^@BG_^~eSd_cd zuEDX;uBHt0>p5<->`K&rArITfxkGViIhrRnaMD*}NRh%e2$(UKefpdX9b;8t%RGUx z75JCr=RScumQC#B#;G(UQG&V_Nb=$5$MQ{{di09ZFE}3l0-i6gW?u_dVA>Knd}lb1 zeF(K+`X55rjlpQq{#GM&vuk46&G&?E-a6R3N0r`H7{`xS_{YY&OyeUAgwBy1VM)8g z@kwDc37jUw`K77Tu(N^>JVO?;{0u__tR0ZGVmFY7^AG%Zwecj5bvj^&LG8k3}L|Dj|@ zDt4@oW?mU5AbKW+cf~!V&G(wXD?9_2=0vc7t$*sjUN(eb(OKN5=5cu4D~WtBC?xX- z_tGU>Qka8I5o;Nt#jCh_!I6>rFm_fnSzr4Jb;fJq*x46gWta&3zDx54S6A}>pT@E# z8&kN}ng(CrxWSyO8`u-M;jrWO7INOo5arX(;VXG{*1F7q75j1YO;;6|bb%73?h zGMPV&)e+xBo?Ul{`oIYAyl6liHpa89#y#BD7E?6SiNvp0r$CUDI@0VGR-U8@ONTqw zzk4nX%jzz0mwPR^LWgoPJogkXn7kg(O&yM3#%p8jis|G{(?O!yrvkzcEp`b0oR}JG zK4a%%*gT|K*llXliQd0aXKV#(%YMX%Yn=q%&u;i4Ws028;}`iBLsV=rdmEep^G+w@ z^Oy{9U<vV&P@s^7!AdtFARrR2W_@N4hZ=|m?-C3= z-;J40rdZNG3Tfm6ShPfw?=HTL5!;&x(V7d^o@waUGXN{sIm5A>MC#q2OVsqPFf0GN z_*?jXDAy&CzPVO-CNY#e56EOMXRLum7V~H?xya4%84Wd^Sw!nqBH1%GiupcXM|V$c z1*edm^s&+?(75rRF!R)eMK$rF?sw(5SD_JacYa30A=X$K;toR>FQl(#ZGm>9Uoh~) zh#t7)0gADUnL$D|9yxQFL_GUK)@w`hzudm!q<%#{`0rN;T{DUm;2UE1vH?%u_2(2V zg7{AXXCU!XnkaeOWC)xo&+Eyh!`gcyuuc`u!*e2;^b>i|cs0wWQQ&w7o_+?($K?2u z4mJMR8%NwLD#m1odl;IN2^%_>(VrVzK}pGt9z1@Y%nUw*TUX0~PmvU!ww=Q(kC@88 z9sLiC#P-wSy~pSet#R;fBF{|A2~*f<#D`v6#Ezsy!X7;p61OxI^S@@|kTdqY-V{&R z7?;5M{{=&~{vozn|1b+Pdx`xg19A1eB(6GtJk>fl9!w@agb4DKS<6Im5q0^sU0?+U z+J{lKLYZ1ABKz011O-+SAJ?%2r#`FWCXW&4E!WnPWjl6a*1afjS{h9X+&AHjzpG%> zC!VlthBirKyWpa`6qQr%f@@#faCAyBT+TZHJN{I1!IB32@8oHMU!V{@ydUAdzduOa zhM!!=wOusf<$2`T!iZ7t#-^#vnF1 zkeg~zC~`Re1LJR-;Nk{Heu>jQSg^H)w9UT8l5BSx{gr5kUcU!z3Ghs%hjIc(&|PA!A6KHErN%6l&EdIYBNnyj%~>PY@;_?genE;m{zl!6{y}q|6qO6-#!l%e^$^Bk7TL;>q1fcifpze z=bbR~eu4`9r@-vW1QZ=Dhx{;MKIC1BM&Y{rgx1AWSJsz}F3g2!x7nayFp@hx{4&nd z?u7U+#lS7Lg6G#J!ETc{)~jj2uO4xTMBWj6Qp68RH&XQeAb0?^mG~be!}#kfhhx3k zQ=G9gSyWS*%K|;5(S=i^NrJ2Tk=Yb5NO%tsIydn`=Xmxf#fKkz%9T%EcLr@U#@E~k zdP#CMmVt8oci8Xm08$&b!A-|dFlZl!(psLVcq$M^IbLJmCT2h**@}UfCi0O9WIhvk zYVALj)P*NNW`r$n_xMYQV+Q(GpJvKnN!*>Ul84hxFf_#yc3d9{bq|`+^y?)XgXtl- zD=QTbB+8pOX&?gA@C%dhW1!I=&(~1#RlQ_Yj{BXNcI}a$I;lk0sh4B)=xFf{wEX38^dx^}k7!m6`JV z^lS`2D$oBMHi>_D>myn{J3-#q-4Hy%Z5S|QJpTbI$tnfG3H5d_h#9C*F5m!^?UJPu z?@L%;_C=CZe3taLeiG)r_et<}Z<>7846OJ-96I6-yX|0wi94d}FDM3q?2Zzcdqm*p zx*DMTqr))v@l;Gpi9*K{!JKRDEw=hj82k6Y1n>lBz|;uZPpYGB;?Oe-E~N-iP`pq@ce~kJW7u!GXzfOj&*_YD_6) zN7@I8%(3yfI!lH=JZOiO10FcRs1oM6O~hjdOR(jlIDOkb2F`Ysz(%dyixM|0VA{A# zWaw@ss$2CJ_n7{}@j16_OvbIi6(em0Zu3?&c2gACXy-8Z+a0n*tDM*#*+A3ZMl$hd z<8T!mV)qm#V%@$ycw2uMyZBRrY8#BK2L6cF4=7Wa1#e(w>L{DN zOH|;}?miqnrH^x}3n05UwAAbFU5~?z)%cErExeuLDduGzh!sPQVEGgoIM@`#PxZ;b zlAXix>Al0a$LtMM+&M;8mHvYv77EDy^(E@}rc%2PzkvB!l8>#WnANZz58J*V9*tpO zxY~qYEj9;FtgGSkLcQpA?@%a|6NkfZWLn1(3+P(};} z2SoFig|pX^>5_m46ybv0Rkp0>xWLkRMQVOJ!1L~-BBgP{yP&rXdY;9K{sn)8K97?i zWu*ms%F-Z|i{Llv_zHYaH~b~sH)Y!&vSB_8aEIr5lCi=LYLiOIhZPc3U!w&~cUjR! z`4qNv(knc&Y6)7O%7xB31!&P@iRBF_N9g0&K0N9i1*dmW$oVTx?yWhB)j7g!ef2O*j+lpqrWSO}`@Qf@E%8l3u<40)H zr(2@AWt*RXw#h%T+w>;Le^sSf4>W0%mImG!Ue4y<65)5R8a7M?MXkgTq@#>!LrV^P z?Pw$^D-Gz&Sz>tLKfz7$tsi{M&y)6_RiNx{fyeEWIe`jE`@6qF#+jvLpY(k??4TBZ zN3x>M&hQD=r8@9p8x{G#%Zhmm-B<{n<$){H^{L+5lQ3L8eDzn-@aMOOQv z^It2RYrX;($t-s2r~?zvaO8h}Y$9I-?zmg$aGL7kPGV|paoATi{;Aei_NC7c-`;3O zr&aZMT{D87UMx@Nj5nkzGe*-fYqRj{Bpp;dUy2D&5TcG-LAvoaTqSSkA~8U70sl9`9d3=#11kgdnOidY zpd>PjKFV)r@&kWh%K4pS!Ok2m^i3Lbms-wu&mGDa=H4Z1o89X->P<#_l0UsvwFGzV z90{qV8JxlZV@Yce_Z_RkjY2oq!fP?e)Y^dM8f*BPGy@h~m4-Kl+gWvB(UmCRmI_sV+2sV^|rb0IwR*#=LphtQ!7MdZ);yV$xb1Tgk7 z?hyZjuk3wASGT>!m}(Ldl!5-?me7pvp~xiI&gjP9of|{oH@)M;}4@Un^Dt? zY=BFEoD;|4vdtEo#yxp(VB2UqVCV+3itJ$MF-00!a1q9Ajf1PQ=6pxmGI$)Ufo-oQ z&@p<>DC75&%-tsqR3ih%eOZFn1$I`!k06v9HoQJz`BUibb;Vmk7GdI-Ts-(loZhrO zjewUc$#WT1PUDZU*u1;u5u zn9sTg|$uA2-Wmc+rBQGemZvVMUvQvi!t0&DOqfQ7+3 zz-Wp!F}U5rF6v)}2h9;QR{VkJ)~05-l7PJUzKPs~s}qDi&1Cl2C=T02ia7hNDO&1} zN87n-xGzJ5>kBTj6Cxvif8-6wbeu-6tgZvm(s)qL+Q&*Wl3{TFTFklW0GGpF3GcI& zsJv`Fd$DR1{_&24@;QWbv@a30)`XIwIl}$VHXLoNYPc1rDk0Z%G&?7G4Ng@Y0*Uz( zX!i0t{QJul+$@ISq0lZON}fu}?jFZ2p}TO5^L$K{uEslC*VDZr2Vk3OIxT971P5n( z7(HSTSI3nzdA&E}`XosjRM?Jj3oPMVV;7{I^rUCnMo_&`3t?Nsdo=hm9b#wX+MvbpnSc;nQ zN)Vh=jX_J+7+D{UfTfh#=ET;v= zCS?EPG|YPbn)8^v1tM28!R2&)oDsDa)pg^*SmXw$KdfPnlfJSQE3aeOQVTGuRN<2E zX@ak@H!j_;3+l_R0tS?`yitqcYMxL+GtnYGo|i=@Clx}!hq1uhYGTiua&UKDH>jwj zKwbM{jD9cdqEHf z3Cho77ask`UOaMOcCRc@XWIsZ2^Zn(iZuK+EDIc^dr0@I6v+2gqracsg1HMAT>HX6 zJ719LS7gjUlkZks6rv%?%{s`8Y7t!uy9!|H92gSCR>~Ka62~4r#hx}!Ni>H}I ztG6=0+kK+veg`3(*t4q_9O25Qa$+aF9>VKBX>y zJB4CyeQal9KYQ|d8su-BLzd34<(nKVF>x>z4=NR+SjleDEx#Eot1SZlS{@+w-F5h9 z$0+EsE9W9S<)~bT22Jo6T&1;dNRPc7zjvkqaSqvvn}tVVv_>@dTe=KoE*a3SxM8Gq z>Qf*c6|n6g1+06kQ>a#C&J19ZORswmtjV<-`)X*9W(d|ZLioq`ysqd3kQ3R-jHpPVxVT4 zfWE0!SafqDWLull+ww=qPz7Un6f^-grSiCDW(YmLDUeV7mw>@kfe-K+PfNZXgrun5 zaQdvDz)PP>;yUWtJ7EviD|B;B7wSQAcMN394F%6n4RHB=Ju@>L!p5IdW-iiG`O+^N z@Y+NDx(^b`qJ*xEXn!>iO?Aw1<+o%U*RTnsWUdqCoTHp%To193c4z;i=sf(f{N6Y& zWMoDuyM&Yxk>YvIb(%&flo66tD($p2>=jCsRZ3}!5)tlm-KC|JjFQrzB5CT|(%^Uh z{sFJY>v>-1zRz`iKJWLUzi!YlWGIdaT|^%)P=dJh-Bk9~GwyA+1z5Gn@K(PXp~fCr z@{Jp8#Y!LX$Ce>*KU)V$*D*NlBF0YN4=^Rp3*|`#h}T@@e3seL(U}ieE1QA)TL;r& z=cF)J+6#vgCYn}HiC2WBun(}ocMFqXkZ|`@7abvmzoYT*A36S(p*%mUd71Y#A zf-msAM*cp&jpXV@+0UI$u2%zHUHlc@tI+|O*nMfiEvP_AeD46!j0^8N~i;+d1j)0pU& zBtI%0UgulFqmxBw^;H#MbyFl@!hRNca^A()_!xn|M8?Jd*0Q0$ujLB&*-;rDJz3 z1P|^M{0)2tSL`d<<)C%Y+MLDRYgkHjT~t7WE5;q4Q(51VR@69{jW@ibnCpV4%csaw6gL}!qnM^z~=MB3!dm`pkXz>dy3bAtQV4^Qq zif5ag&@lQQ*%NREKNyI>Aa@SFaQp#-hh1Q^bRQGF=A%S+!(0d(zmJ<0@Qd7Qx5Tz3 z0mP^yP~;VzL_5d56??u{qg`RI@uJfsruSeI|82#47|5}K&FdxUgzFi2bFL>SuJ<68 z*A)5kmM|Fgpak{H;?U}rC4{Yu1Q{X6VDqDf*@ewx>1C0aw{Q;pln}G*#$^1NIJ^3Z z!Ap3v@G*u4cYz8e=sJF!>8Tg1m755|ps zd-(QmeHeOZ2VZL7MF0JoF7Oc@;IfkglUrvYu+dKAl+*}NZ=Z`R0u8~kc^aMbUGTc) z#lp#zx1pVvXNHC|=obABcGI?v#Xq}WZQMkdN*yDvKbLV|Bd(LKrVeK8bB-0io{EK; zLT0MygIK>F;bZ7pQRssN2r*Tqc|sP%P!q`hE5UUBX;0y7h@pd~O%TP(@hmm39V5JL z=@4aih|8#e1siK|OO_QFcX)wbtpc5u^@3P#5C*IhCE(1U4))<VX@&56*;h3BEyJDokb;jCYhgt1OCq4fVBYW#{ICH3HD^U%}0R?osK)sEQRnzhq#+xoZxdq zDzn~}FEAt&(BWGuYDXxrthI9FT)7Pf7jA<+meXOob~>~ia%Ur@a^O>E7JTdNAe-ji zgYs)fG3a0lE^e+vhpxrrcI!BD+N}rR)P!gkL7zrP}j42S$LHs zC=ShRXjVf^q#k?`?`8>XF7 zq#8P5AV1WN&W|6-ChtY^-{?(37uz4Us9ItnzQp+|k0H$QAU2jCgj+T%!EL1)u2M|I z;)n5&%FDprU;& zYn>}VH&_QM*UJ+obpTR-o4^hiGqS_?A^UbI7SCp`guSX0P*eUHF;0x;CMASm!asFj zdp5wO<{f;lWCH${u3=uoFF;?_cWlh}MFXu@sHUbvfArRn%$d_fIf|*M5t1Y9&qs-j z(vPxQ)46aVmymVeHGpuDc;Q7W*6Mp=_WYCRUXqJNJGQW8&WD*>Of9%b7ejxN6r6vy z6MFmHAba*0OxkmX=m_5DV|j>LchY(B>QsJNktthfae(hMyH6%JO=bQnlgaphP7pa_ zfU}Dga((hw?KR%b#CPfEh0od!Z6?D|<$yEH7hGz(k0*nhr30;-eUwRPoMX>SR`3_^ zoP|9?E-Kg3jb7g)use=qk+=(&Fi*FQ$nWqY(=3Hfl(QGeJhjGgJ6uJs_YBz-lPBmu zM~2#i9gzNiLAqL6egQZeZVZh-mM(pmmiZ>U&`#J<=b}y|7n7t?6Dg zhx$>AoC7Q+;3ga%%aENjBEJnNPTja051sTO`HSMf-aG)jHJ37vDLJ^qpaB1H^T^0j zMgGzibGrGS7R9+Lz>=IX|5H`<4ToyHam30;QDkdv+^L+Rdug8;6kDOApdama&Hm;dh5>u)2lGuvJDJ;#H{B2 z446A>EH&Og8GCDl6Esi$ACW%DcW@k!XjlNAxJl{cGuNG*ekEb&gr3FUh zALbR3MmG1Yh1(-D;Fo-7_0bWQu2*m@F5 zFMS22OX+O+@(MV<<0t!_5lOCCpN9O?3Yb^rj_X_Nz-dJ;8@l@m8PXSzQiica*+i0D zp6Jh~{-;P6_3VL3*PNm9ggRB+p9+f4lejk z^rd9LuHB0F{MSm%>dH`)n!@`VrtGK839QLEK@LZh!`GAF*r=PYU_#m~_Un!CyR9e$ z^IxibOMMYLYxNHK)HC4eHwnzHhQcTBD=b2Z*0=7Z^57=<5~}-gk(Dc>4Ny{QP7e`Ohm7Bo{=$+kO)| zjz_k(rc8Y4{U8!DBL!z2Y-Xk&3V5PB3~N?g!^#bj@Gbo$mpNi5R~OL>4MIPwpi{Uz zsAjUb2fW}eiH3%RB3v!(5%tfCA*=ZvF+0~LIEVJ(gTsW5D6!*P7whtIv8$PEc`4p) zlgF2S+Vo%8B6#HEAnrHFge9XMvN=Qcvi@rYa9xhl zD$1Ko&!tjN#*%^=9C(_aW=~pQfVJ9ttbI7Zy>L9sZ5rZ=?Z=PflsP<$wJyY+1yQKH zxr7vH3LSmTNdB_N3v}0i!dF&i!E}KM7f6<6Vpm&qHo=D)OwL1c5fEWz=e;XX5&PvpE4gu)~nE`XU6h#oTg#- z)I8)Y=V6Zb6?B^C0^f#ef@a`IJpAn<7|(wU%|(aE4XaJKaM576m^&Es&5GHxKNFzy zau|wNFT-o`47S@#LBr!A5SO4vRShDUh(b8~h>=^$OKqmRQUwBf(s zUQly+7I)*A4G9i^0%TAWQ#E{orVmtT<+3YqLg*0YmaBo%pb}hmO>ioO%)suQ$zT}& zh$YRfWmkj_z!Z>&fG`c1_{fNV5fDL+r~C1n{CJcf{9Ak;hl*on9O7Ok$Feh*KEu)H zqv3mIl1Qp88#{_}p=S}0J0G*~N`4MbxHE}_oRj74o6nTQtJ z{mipILYS|)k@Iq1aNwE=E!}m!>ezER+Vx(TeQocCNwNp&9TkR}Z;oU4@VT({xCEaY zdlT1`xX@K0hlD<2he)*PGMnceN5YN%V>?&whwY1PQT^)OYCa^FJR568>d#DLRVuIW zn!YtU{=CnsYm!NOZU;&(42FRA`Q+f^`@0Gz1fB7FO#Jy!(AcbELE>FidHgoBN+Tgoe zE&e{L!(JyG=+-)=E0V2FU9@F zVa&$X1X?d1BX^e1#BDQGX-ifpx9iIZc)G;`)BZb2w0vIRlKN>Y|XUhSk34h!k< zlS{zo+-f?oa~kGNd`H$zDPlXPjGz*Y#}TUQaPB29y5o8z41euNRcrNlqlsOpzjq~m ze3p(y=HD^?N~Pd`tH3zRHF)@?BK^-cm&&$BLSOq#x^s6qDCmTe1sAR8Ciy@pvNVKy zEqBP@kO<7b5DFQ49r?Dta`LP2BboHyDHzw*fXyG>+59n4_}Nym>&nLl+omV9_|o&LA&}3$gJ{WVj8@HUL044 zNxHL8CR~-@wr4r4TcD0prYW(`sqSbU+Dr^H%z3kcBX&okEXd@uvV6sZQh5GJgM}B$ zbHX5W9Tfh>p>%a(O$|eU!TGITMmT> z6Ss?#8_!_$cNL!dQ_g~NcMzA5S{qz~TL45*nOBer{^6E_={|2Nf=}wNpw- zbX6R=sU3}DMp1fdzb6Kq+y>QTCMq8sByPGkU3_(nBWa#u$_2SF40&n52Y9%W<^>B_ zx$u4VC+G{=`z*pYxRAecKgg%`OHuUt9tm3V5il@`>^ZN5rl&N}&)%6+>m1B~N_oX> zjWuD+u5O&<5&@na`Q&c6kP{eC;djk`k55-7S<@wF_Sja~G0eLUBSsGgfu07+ zubz_+f~&&S?i~C(e*&6kFA{e;nlhiEr$9ID27C3Q6XyOHh9k}9g4TuvxcgU&DamV* zWf#@y`a!M4zS@R(d%IExQx!B?Wrr(!?f5$*fOYo|fniSWgDWj-J;82iH+N zPLUM<{NqcK4|+rEtav~SsRRc`Dw1OzlD`TJfdeQXKLwpC-pkQjm>>p$iE639?xk4TIjdeub zp_PrhF_T;CIRPDZlW+Z*>bB4b)Fo7rhF^7>Xkxv9>_qw zPZIb~UYcGP&&PQ`fIghFN$@l2k`nz~Oeook|DJ?$W84Ji$D3v{^wwnF`L`ZBzm3Bu z&k|woJ1fkcc^_hy<`W+^YZ!B76Tc-gkC%0o;orWAF9k zN%|=%A&)^&dSRHrL6G71CwYozD>h)6vog*&Uj=LD9S6UbBk;P!0J^Y~={=o>KHMId zvSSnzdoI9Bo`d;&PO9Km(ShCHtl6vGb%e5tVECSDo2(OjrUv8PWy4%YVV! zqHdC!E=!;MXvT{*=~wE9h;XD#0A6mqj$8i4^Yaw!VdT8$h z!yT-2Jk5^$)f7Cx&aoA(eQ$ES#IT~82`FN1mc zVmazFuuS~itpJXFUP#IU+{vg3B64D23`~<3vR&8Yc=gIw{O`wY*6!B?3%zsMU?m^e z{ADe)4%Gp!WNvqt!uVCwX^FvM z>=|~7OAGI;j(5L_mhVzfR4DT_>H^~iTTZm$1xeLAB#z#D62+TlvWBdS5Hm*+ddB4v z@9-LOvndYmop;A;j-&X*)?yNNMT+!KjN+G9=tE><8JBZBlCDxa0WGr4teGEx9-UD@ z9hBgJ!4s&@%O|%>t;jU>=S*r=lIX*U2wdfRl1Bd-Ll4Y^qm^m* z>0Yw-$vm-+tRAG54#Q5-9e(BGICSZm!|S{#K`m=bBD?#M_-+1YFxFcRMdGC}R&Edk z89GCJh%w4N+r)mJolbt74&>Wh=F#!DCy~H;^P#mUvuZ)N9U7)qcI3DCMeaEe&6Y+jtDl&;Po65{G>F^gL?k7=V5Z|LG#BoE*CuN715Z5g?q&rp zXn!U)n=hvCK#p1@PUM$O8A+_QZ{xD{m0Hoy(xbI z7N0DHb*U?;anD{nw5c{8q8AUB`6=1w$1{u}^FeT)KE4K0i2>ZaOh+O$V-Xi)XA8Y}AKo2u zpuyJ$a1h+WtC3^jyYF~tmXk%<^9$%%Hx=4^`vhro`UYNWas;Mz0^1W(fKsDgkP8+( zsW0`yS*NW*>&78A_v1AF#&cP&ygrkZu$M6ZTscW*k{GaiIBy{IZ{G!uhl0ODK;GaU zId(k+4~8m$**Q19q3SdYT#=jlC-e#adLoAt$LaFv5o%ns`Xba`Y=w2(QsB#+2r!VB zq$<;DU}9zg(=1g*iGUm!sUC+L)l|tN%?dFdNhY)JapaNGK4L!nBJu9)C&T=Mh{wsv z?7E@`EEjxaA2*!AyO&asZYd`#`me*c^QWoexs$wNi65Mt9?upvdWpYYg!jOu&I;d}?44mpYyCrwz;_({~^tqmTtOr}f7R=i_&6jxdp z@aoD6^lD5tIz^phT63o0`uumKt6(_VS^dL~X)2ud!dw5AQ+EdcJDENvi?8v>S7L4TDF8ud2)*GUZY5hBh6{m z-QoPbl7kp?e<4b&8BhC9orjKLW>oZtk9d zUZ3|+>Gz*;>guWDXCF@B!u?nHl%W#Xd%O{M_U4g2=JMp!mI{G8co9F0f5G%*MRR0ahJL$En|gA2C1JAoDq-oW!U2(y-D$?;Wj*E)EM+Z-JgINtt zMCtGqEN;)KS~+4cc5Rr=uFT(yTVMAhl%4_gk4bnfl#|NZ@og#P@5JsLC5zd~`QP$PeUlEnB6?c5e{ZS=8eB)OI2@Q;W|D z8ARH26>~7< z3dLgm>E!qk84A%>P!>N4c6OCO*Wg4pPO=X2+=oyt={307B^X9XsZyJ1S&+Qr9sKnl zLoch170!>N)UWkCKHgyfQt`g196SSlk%g#{5`d8>@?q}MO+%Y9}Sj(u%n^yRA9|2zg-C9B_+Ul-%d!Ye?z?d#}Lh2LXQehh{z4c z#QXCLa8uC*ezb!we>)?Zzk5iT&giNl8=~iejO7e$&^ke)j8BVhtXd+tSB=5$d9T=2 zcupSl-^l&7$N|ro{bZA+E*Rfxg#Z2^_tR4jy8e}-p1LWGI?`J$IZ2pzF4;@IZnI?T zdd;z8lRf)}Hsr#_e5h(4iJZGTdnhH$nU^ib-Tg!P4-OtAQ(-U49N9)(8ijXJ=aF!6 znJv@#dl7fpP~NS*h$@s+CeWbJlm*b&za<_*Hjjqy z^CyEX7n2nEeDU+yzsS2II^^XfRXl#Mhdq49;i1a!TvhHev}=n%Nz3h!cVsSaF(nkm zOJn%#rh2q%ih+lRO=12s;XXfV2B}|gmpl7vFB^`#$vL&=42PB3j~J(+5+1lq?` z=+)7Q^k{?)_zIxSdWxM>I2f%~!C{hza`v*9Sn5IRF=5j-p-_q>}nu80Nd3Xqa=28f+yYVW1_&SIWIx?PmDpf)P zdjvLlFQAmgk$Jv%aks`ENUcvNw(F;1wbm4@Ng!a}JOU&_GHJ!Dvg#6*0To z4wL;l^Wle^I=u1Kfz|6(vAsh$mld?2SymFI43>fOiid3U>DwSN*Mz=Z`y0iJ)N%XA zz3?$!18VK&G70-=_!E2_28(Lp$vk~}OW=$&+qy!Jl^=@ED8VUs2$n7hyv3+Y%;F~U zN5&?C(z9|Q$Gw5?`SXo;-8PgzrKQa)P2fqL$xX~El1CS5F&ud(0Uo`hsQUQFAhwm6S~gR7kN8JgS2i-64Qs)z$dybI`q7WBT!FTs31EKVANk-Y zIFJ6?gGZ<qWdm#aMRl_rbh

z6%8x#)|(Zya^NDFwP-jznzx6I+5ejRt)3NcxI5*y4-bSXHx(D0q%=sbjo^jQM zru?l#G5qg}YtSOFay7^Q!B=Gq$p-6v?2yF(Jo>pDDg<6&;`q7fa_1J>+)+V$u{@z2C9~r+5vbO3$Kk!!HpYrWRm) zI#<-YtOQJ_t>Mq-ABTZC6S1T5I@$ia3-V-g#I>R*EERl$uY-*lOS6T#kk#KbvO<}+q_Vgt9QLy_+PrY`onu}geAqY__^ zNPvLDL&?O-2x607gdZCPf8$MGetLg2s=OPGJBJ71AL9(X@%%ccJG-51ULHx!pC;nM z*#*qVySU2p%wDn7zyTQP-w$~kwQ#eL(_Q*L2+s)T@Rv!}xXq;yk7-;ME#}uj{VWM+ z*ct&ploP8t7n{jV|#K`E<;fz+1#l1 zVUs92Y#*k4bSL|{qaeKV(2s6^$$HP1uu`)d&3rc?iMWb6P8K-s-cxore=>V9-H{IN z@C3PGBkB9_;e71L7eti%g7qFr!HZRWY(Q3;Z+&cpKCKpT)MzU%-&~6!;_)y-@B&^P z?Tyw)@<_qLalGkBGwNeJ2&(NBapQmj?_B(eh1lxx&-Pp*sC9u;1Wg*AD6oa3cEFB5Xr7KR-x6#pJ4wN(F|>c} zAGFn)H8(l+xTjdVLk{KDcbNS<%+4pOG|o5DTR92 zUB*`yy0~ukMHUrYj<1V~nOfFuEL%N^X$8s%&cBN!!YvZ#I~>E_PjA^6`4RM@E)~w> zt<vT%P6gl5&lcLr(7nT-u& zmI}SPBX}!hZdGw)B37L^&FUAVi2VYH1Po`s z6&_XN*GaS90!^`r+2s4o)nbQp<(%zJwA8R*zWb!P*a{vzWmWOxJb-`zGq*uDRF=-+!mrGB`$Hg6eZh#{TTOknaq$a1$7- zmZZy4gKJbM=YCz$A%UM`xnd^AB37r0U4n|a9Sz8-zX@e;tqWwu#iRKdQEo?dnXn8@Ei8 z)7HPobqvcDOP341>Kt|Mh0{Utu9IKb+>QZpjan=FIZTDgZ!#b=^hS_Q6Io75ynz!j zGq(BXS57@^3%5r7F4G>ph6{Xho?CHgGOLMMDxUUle$^z0Oio2gpDBe5h_t>>XTfHd z*{ZfjTzKVHd#8;bnc|Zq7Q62tbL!c_xcF)oI#&350mG}3T|aVT!t`0Ph>4Z_Lz(NL zjjSX-h;7gF65ounBy-}stD^ijve~}-n4i*lBK7eOcWqEN=l$#lCv#4dUHz8N7Gsg< z?qSL{>^sLEFJ2|ixG4tJ!EEZrsvGS8rbn{92PTT$uAXLxSAG^x zo7TfMs%~a~R0^16j4B!OL11OiH)R<^CUO%LXR&J+MzQxz`^7fCgslnu!TsgtiMBoK z5j(2M;nKt_+?~V__PslW8|_ypRvpvHsi>H+`4VxQ{&~r&lMYF2{o+CFcji6ugxWbQ z{6HEDXLC6I#zl7Z-#RAyql9%dD6&m$E7-x>VdV6dL>8_o#k}w9Fh9K))pt^h#HKxu zL>oV-G22;ZIHjfXT;b&)@xGUv*_v5J+|{8AxP=WD#NN+>+4dna%*>M;I(mmdz3%{aP4a?^I`*KWX+e#0&yj?rfA(n$=c3&KExzRa7tZCj3g5nJ z1Db#6V|QHhNMKt8OKVgid2czv39%dQls#r!(;gCk)iJoGNz8S|Inm<5uf!i>9EeO@ zIFxRU#MsWu?CbMpEHWmTTON-apD|SUyG8Jvej?*GtKg9xLz&rZ2coI(1aWOzbW(W+ z{2UfTRTlri18=uNao;q|+;Wi^CaVhb6*VH6(glWdrTM!H{P3lkGgy6I26BIrF{j}> z+1o5Qa5~n)(>fQh&u(HvR|Y`5z%wmO7y6aYiZErAB50{8Lr{eW6i;6X7v`@AHQT|| zL-<|3-!=}L6XjsxmH^b)6$~cFcA^?>C5ruq5N{O2DkmE9zwYkF$mx!}+lLMe7x<`8 z)_Fqgf}dpM(|E8-SW7QC8KCG?CetW;SAG6R5cPY@Rj;srAx?|ZrFGv&vcoN^bkXn4 z;{17x7-e3AxdTJd^jjJxLMghO)tc&9F1+)V4Wk<_7@bgZ zQ74+TcZ4^n45$Dw_*48Juo3yU>C>@qD|td?4iJK zxmepr-YQrTM{a_6?Lj#RFet*|wNE%3vIE8GmEh(gW=1ZfalM7J-MYuUFycL7w|=}} zrlc8qZbVhzv5-Q;>@snYX983g4`ttOsltXL3v4P_$n2~d*lzT ztSEE-{v#6zb1R0_mNERYCT(_P!8Ej#X+fL1gP6W%6EA911nr0)FyF&K^x{N5bGmjI z4TC=8?c!zh`l2gr*r6XJ=JiHwI`tgm4p#9!6WigbVj+<_y^H!ADxj&XFxO4a$2pFI zCpV-JUX_Hw!;xvY>p>2yYY4(%y*>!Hw}aP1JlT2aD9--H9$XaS#x@yv;!Zg);&^ca zuhqC7WIX;d^W|>zSd0`c@5v_af3%5$T{`kMUMQ*BiHGu?!Q^Q&yb^Mc!Quw|vdkW4 zR8|l_%Lb474b41N$oN&5q55vC%zD zW&eE^^kWM6yR3l09`i{4_GUaa{{+aM%ocxh>|vvnRPp(D4_vqP0)AJNgry}(;$+qf z!+#vYM?1}7c4>g%+t?%a4wqtklD$~8%ox7XPzp^RbRp%oFpoRW$ndV?ptf}t<~)iZ z3F+6ECm#S04jp5KH}hE?>4%$Jy=Y@d4s}~S3W#ifby=kb?Lc+@gGv_U-ibg?;uASl z7Y8yeMl?;x$arqqhr_n!qPwmhd>dYgs)OCJC&7Uh|_e^gXdh8Cj<(Ck;*m#c}!KQv#)#eGep`rRTk?nIMm&P9D)Z_-k#N&5K@ycJ-g}a>T z_VBm(wOmWwK6ViOY`p`9yGGJRr%kw7HcxCb?guVjEO1&D`{IUJ8;qM+i+ZyiVIVe; zyz9*erHnR`MTgVxzs6E2+nwC;u1ZiF@fXJz*n&uP1;5+8k4t!vPIgCJfP)R)c&@gV zv(mD|%EPs2W!y%54(Ah#rGD%UOe1>*=10c)hveozS4gn6rbF(e;x8eQdcW8P9SnnL zO8-Z)YNsz&G*!Ua&%-dZVFbNC;THSHN5JwGG5ExEB@{lp&7wb*fwzes{Wg3k-1ar6 z3O-Y5oy9rqlF$=dbC=1{+!G{dat(ybq~pI>DeP#QE`E1nIV_*<%*Eu2n9Qg|*l(+e z^m{Gn@81o}-rglyhqJ+Fmn_C8d}rO`4DgrAX<}2kf=KNV7`f|4;}^kuR6b}9Dy;0p zp1cBfd~*e#(WT8Fa?mCR59niFN;vqu$iQ&Y0p9j6$%~u!NmOwxK3}Q~%B@j&XM8Dn zcxeOv)6>SPgzuP8lmZ(QggMP072c}z4@R5n08KOEYyXZ$`!C~(pQ{<>=H21qIbXES zoI&16oA6if30ym)1kus41w=D05obTyK_on0F>7IGo;0uRbuH&`bfGz@DhpkX zqi1nb?p0iruEYnVD8cZK(eNlVV;nea`Sl!F^p1mr3#qtvXf4p;{_50x7p_WU05Hw5>;xZvt^b4 z@EG2}GN%Q^WYs17plVGE%$~BzyR2!*>JF^A*UEkmc})yIo#Q^ITHstG34y&J%x&&E zlG|THNq0;O>zXJFmo?Th6%!r$a!o$0o{8`}x}Nw96p=WA1DtK0jfdapVQ6kDK8>`3 zk-}_5y3T?oUpE46AAR=wh_85RM58zdrokFt3EWhBm27voN`AQ*ik@Dz#6QOl;Nz}h z;+`YSbffp7<78Jj+9NP3=Ip}5>?>P%IUdKAXH|zOuYfJ-%_y@ol-WZnHuPO!FLE5k zN?L=6;>|AbKYI+{9QnjbMTRtTcO2}VA4L~NPQ?D;w`BL#+4z)}( zp}uv0*|c{{pu1!_YJBK`$X_E^m(M@g^+BB;m=^`})byE2QVlcx9YM59WXYvZXP`bc z7XD5whk+S~;p@TGB>sq#17=_Lq&nmWazY=<%A9dYE~ZD<+<8CbJUa zxi5Vg@W=51biTSFYQLC<_Y&X0=ust1G^zvjg#NQ_@OI)t_i%G`&7mgm7P}rJe27YUO2YLRJVD3~O?5gdn z-uWYj3!FX!9KOwj@EfMMDm(%%21T+%6Mdmf;Xk%7wuH@$O2g-6H@U1_DLglJHh7mR zLL|6=dYKybD0>3jisk&A_TuKtSJ^kY7i8P{U3@^865c$!3WL<_Y4OBn_8`WBzPurY zS2Evmhj&GQj@EXt+VY!8E?W#9Cw|k^Cmm3wzLzeudkl?@BVf7z3J5U2fJPd}NS00q z#tNPia#`3{w@l(c%&)+VL0LFaHi?|9>V)j}AlMtcfj*yny88Kpk7BJAW@O6j74U6f z78$m!lUy;_#Xm`VBAh4j*tI?#6zu(bm0-`u}3#lrfk(N!kGXnz;K<7ORx>pX%3N#1i6><+Av74WjnHo9M6W!{?W^pli4%x8UGB@_PFKxQ~8IxRooRKw!H@OxOt1v__N3 zdKswl)1v0(+1z3eBVtn{N7IA%@~8EZa6!Dl z9SYROXxmt@outm^uUU&XRv8zv0S5Q*rFqYK--%LZ7B~ zw!N#9n33`9$(qOP?3g$_u_PHZJu=uz*S8?9DkL~r7yqQ`;X@q}#d{b{gY2yfL zR3dPT|Ba?TZ5wH3>Nqg1lB7TT*0I&*I;dAs#;*S~fWwb$*$3hM#YsB>C-ki&vUSO9 z``ztKW};EG$7F#?t0^Je@6^H6xCdMv2iUv^uSk6C0sh^)Rp2BaPZpeAkIM%|1clkDxAGo0dt#rU6%npk>3 zhcC9VMZZ&b$OY%Sq(}1!*`^~$hddE-ZF?_q7v!zrLwOtgH&qHBXiSBVVR=v=5Dt0S z3+Y=Sw{pbdGCC!d;lWkgxY@D^+{lv0Sh@ZU>GHcIcFTXnzC4wsxA;fwkDzr;JtWQV zzqlTD3?G4IH4e=1%4kvamHl*u-vd;-W@MNyDfnsr-VF8n>u}XJKlcH%n9t|=~f)}g#R~C!{ z%l3yaIk~q5>uzf>8T+p|5(41f#< zWIwb)&zTq^@irH>tw*B!*BFj52mbvlA-6ED0{rsFKwyW6-By#OT9XRMyZL=Yrh75o zXk5*oPr8YZMk(-{RW-om_zehhQsEDtbLT@%2y`DVB~ioo;v&Z)@msMM9vE>9pCtE* zaKi$qSTT$*dmzOJq{;KvE-Kg^)6R?nA7haFZH#?NNPSv7wSTpp{EQYDDqe(kx9=yL zR|k`QjDwh~WBBZ^QYe4HpU+h>q6%w9K(V=tDX4%FGcW>x(KnQYvHx{2x{k@N5jy??D@L2SaA9nNvrq?NmovTlw%a&w{ztE z@&E{#b$}gSbrI&h>LC9Kxth(c}keS9)m^g4RuQW1hkF2ajh{y~XNlNj#?{h^%TS*#1OND5vXxN%4GYt_U(ok9VeXe$rNGYLIB%?xK zH05`H|N4`Md(S=R^?E+DsmZ6qs2MH^6@{_FoY4tVW6s#sUKDoNR&x9qyAAZ+oBx=H z?FwY|X)Shkw+-AkZDMLJ4+FJhLO;L%6UM&y%&44yMqPrGA#}uQ_SN2IfiJNdU-}i$ zH&%zJf{i>MnH`4L9|Thc18d%*uASPBcr|_F?EyHrFdSnH^hBa_gQTr;H5*-D#B8Kf zU`pFDvY^BQ+WtC2)dq7go;wwuS#qqDoGBXa$R&@KOyD^pxhKRqOK7m(VevPoD^Z99YedNF?Hyk_c zBcA%Gj-k%ih)wn*)R4Rl5+$+Fy7V3KzwSdil%&bP>08*C`-hV=f5%-9WZAwnb>5{= zl~jx$&M#~&gQ6WRu;26-9R8Dw>(fJM|LQfo!Qo^&aX<=h=1+iTsbVfYRuxZcp26ZR zccI58g=7~zC99XjFnaF4m=#-;*#U^)t_V)#_2DL%KT?eDZj9jS8_muif1FvZ-Br3lOxbvNZtoZ$ad>D- z5UW0U4jU->jNPt(l=e5Cz!E1%{&;)}*gkC_U&%x0Iqkq!oqI*orKH)`>W#4W=tmCaJXxc_@yMSKpz#hu{-q?8#yHK!?=`b=&ib!ZCP-dzq9wxX zk|EUJv!Qtl z=h621Yg%g{jUNnGu}7&bpXRlIAFSNV{}A86w{GtiW!lGZx*reV@k13PQBh!vJkrFI zYnot$P6V8OFcO~W?B;^6g+O^x6sAqt#%v6%r?G_r^xeHe8aCaLCvzoeSdlmkys5w% z>P*d>?vPEPH|emW8`$hiJo=0HzrclfZyn?sQXV9jv zjkNjR3n(>0k{thwo{K)u?fu#XrTOIoN3Rb1^j<-v??;T?69WeZ##5jF7LkwlZ*qm_ zo-p>&Z-~;CT2627QK(Xl2md3EEEh}gl=Mr`=$3$OhSxAO<_z{*4CB?m)Y3~jFNHl! z6wG*%f{mu8?0ESE*nQ?1cHATEtrK;0^;HYTY|nPK&Q^h(1JRC3k zRCeP1XL6CA!T(CzQ&AM0dK0?ml-*qO|2dVS4X&L#+ox{!p z`P595%sS){*cWkuZFVu_RaVU5eq2`Ni(j23BU1-)jE;?6#h<@id%q#L4}V8JYA=$* zBYu))w@=gN`aYbc9Zcq{Ey4XW7(T8|p0tT(^KW;I6~6iJ)3R@gVpp?+&8K-YAjdpk%o5D zXY#||_R%dChIlK|gsnH`!Lt7qL?oHvqgi3N_k|k&HFhKWaYcf_Q!9os^Cq*;CZB-O z{)br9vqF#6)5zgH#ZcHVh7GqZre6~>M7K8mfOY40qEe3*?y4F^U%T7F`ui_vQ}=7U zs4Flnx>(Tdxk@)^P9*D#_M`8)Sn}}YXP7?qH*6fC%b#9m4%wG)Vu#vpFmleJRdpP- z`*sDsI@G|{a(A|HlqxS((u$Wg6KLjG7#Qv ze%rt(dmL>UqezN^ju4ZJv(PiIojT_3;FYXJ(8jnBjV*^EzV;HNzKF$5Z-&7L*$J@p zrXo%fjb}>@>PSXKBD4?x1QzqX*nv^Pz0JS_k9tg}S||L;hmb-lYkLQb$BYEMsR!A# z-M09|uaL$(*#wp!+NjyC0hG;Y2bU@_K2ctnhdOugPUl9mOV1a>*U$S|@jzwBpDhFD z)=06TTvYm!<*uwm*fa`sEsjEf(PsD~(?mQz%CL_HUeiO{ETO#oA=s{q zVEgk67=w+Y*cqq3kb6rS!2M4m$v0NUH!qdQtLzbs+(I9StGGsd5~oAsJLU|W;{ZR`iCuL&`O=g%knN)MZG&ro$ zLbv~%1_zFRB3`cZz{$-Mmb*-(OQdX}yeSqcu2`Z(=skK69?;MIlQ3X!7W}&60NryM zzqC*<41ud8 z+?1AWr$G-l!ETrZHaHF<_P<3&n&9wFW!N&T1=g~Q_+ZA3UHE7@uRJxKANOmr?UrT* zY>#n<@dq-F4u`;u@!p%jx$1<%Add7Ag)8|?c#gIA6Kv?`{s}37f3>RG<;~^gRuB2U6=O=yS=8vuF5

`rEb;nTd*Q@~AnGqc8yt0(bEH|W3<%Hkk8hp9WyBiZ#OW*QeD{L`!5_G3rl&Ca+Y2f@ zM!?NHmN>LUL38{Hu2^mbF?DzfQYG{Gs(6aZs=}^h&J1=zN(N(fZZ9^a%Yo$75`p(< z$1c=vA*13liPpBSgz;48xA!#>{W4_~3aD(tS{qmrFdAG;9%Fgnvoc@)BVo^ToaU&D zTqqqP*4LEz5qn-yZOaD6YRzqqx0I$E-mOEaW--1f;t?I;DF(4;pO6i?!*ST=B;G@0 z$yRX5%>$g@eh`a$vxryJIM|T4j(^TgVzrnqz^C(Z`Sj;y z8;y1_-9lGm*Jmf*?fnO88h@WIw=IFWX6kTnVLUTFbtT?8CV{F7FQFzhf|mVv0Tuhj z$OykDNH#XwuDEd%_A7rNLF?9I`Y%U3Wj+hiA0C4wv12%;`x#m;7UqTe0&Z@x5{%j@ z!rQBk!iPz~y9JL%%lBe@=GP>wahZ-^+N$Z+sr4A<9*iwL1$<_8C!Ah1mbEX|BZa2xp)w_;!4YS{pIbkAbFD}G08~-vdK40JuE&{X&UWWCb0G?$j6N8XDB)r{> zr|Zf6Z(-=~&+66P0Z6ewUHke%W7JoOMpzl65HeuPsjkV_-r~}{|GxJVIQk2Z;wt|cNmjHU#OMXS7A4L0TKnC;fOymWc}2B z(lh&{9dq6k3-9<63;jM=XqF6Zzg1X$em<}GdKG?+5DDHiO5;nE7_o|ENaV`-3j>MN zd1w;cPFjbNVWq@-a}-@_F@w~S95}GZja_jo98)9S(c_#czFm0+GVTj4?x=krI+{qD zejJ1H-Vv}!(HpGvbl|5+I4Zig66;QZ8yMuxqir%ed>PJ5twItV8^Y|nycV_31=6w2 zLXZDoEXk7H4VSilU_#T1=$YCmFnw|so+?#P$e4Uba_l_R7Jeo@o;eqP7uW2w;D)!vAzHtqdM%r|^U@3XCjshg{Ka3WxuuPHW7+^8 zww#6{l`gEFx(_2ROs6MP|0{dZGMl;h)Rr2DO{My2OJL-Ya45BWK)uD&sJQ)VY?8F( zFaK;Nawlr&!X+$)FZ|2Hp!jZR0y` zz!@GcOL^qPzesbiaC0=vc`a$TrtN<3i`;XpH%Og7~ zb=k_3DrBU`Vt5&B2S5H)Q>dE8M(g|lyH~};EcFL;*DUHcVn#6;}gqGmDhP z2S9Pi8Soa|rkiGak#8;SATu`_gY(tl{Mk+XjSZP>qx=i3+n9=TA`MuvyKkwJxFUF5 zEr53#`{A6;c{x?sSOZ&_%SoH~e{T11RK69aO zjRAX5JRRr1^&>F?r~CSqeUPYcfH9-H$l?i8Fyr4eYU%%uE-oyh`h)W9(Dv;_GgN|| za6%jH-y%`|;=*qXcc<+eXOm8WQPPoGM0@P3$+|)*UVKI%Os=^{56D*#>FgCyrjL03 zUls10r&6}?`#oakzLhi%*Msj~v*FX(w{&o}IE}dmARqe#LtZYVr!JXdSw=3{tIp+C zsZ>Ca*GM*Gh+vEAQdm*p$=b_30Hd2@`KZhnc=oa-p0ds4Wv9fk5IG+fho-`;gEq*Y zk%xT)1vp(}i1uloq}}h8+0r|dDE&JE*JsZom70Vf*0vVajveDQXSZX9kX!$6_;)xH zSx5ppqj8S=1=`x23@VF{gY4HZm{9SU9{cs37W);WqKP&?^>GZPo04(Dt0TNZhX^vV z>d0oxM95K9q>ru-XIHJt;3f*U%b$kcbYY<;+gKfi+4Ma#ZZyMZJZfO(-ap9aR%nB6 z%t^@FI)k-aXNx04CW6ZR2&OV;Iq!W^1_B>{rtQ}ph}8xW{++RpiR_pL$HUH0g9&~# zar76^J${l4-<-o&Vx7NF5Bo1fMi`r7DeA z1N6w3s@)Dg-T@N2<4uMc!DJzy8mWyDs#HBKHi@Gkpb6JUs&S3dhoS z<PHhL(% z9j0Yf(eFYx$5cshyIX`~XjdUFA18v`3Uze1xXq1xITjzE4~Le93cR{pA7d98lWF?5 z=<|*1QR&-Kc<((Qy4=3u*W10g+;}0hxrl^a=2`Cj+%IS_wH*aX1k72Tj!uKZJ+0$7 zn3_hy9E%I=eUUTH@T!BRmvU@whC3Qn4?#+#BX2eOG37MQ;tb=ZB<{g3`l|F0xADwJ znDl2fx31?dX-)eNOU3MIuS}BQvMr(gdv?Q%_Z{dOFUNnaUnVMD*#lEWr|@o(3kjYg z#mk>mV~uuH2|1EJTsc-6^6QJh!ZHg#nwFNye$@iY40-rm-%E=YhSTO5K{P*O9&X|omF^^YqUGcJrsNa~>8b&eh+8K^ur zfgJuGL5(?K2wvAh<^N8hrv^&s5~ovKlg%I#`&xpo@E?H**(#9#G?hNq)4B9{T?p0a0`l zPXDBimIldm@9$z7I6sV58hTRIScc|LT}y`&+PSMbmh|n>y$oTE=q|0%Btyp%H%thk z^`C4}Z|EX-NLTpVjF+OnW=W8sg=R$LZA8{+snX_3ZB8LU1>O7(igwxA;+;A%{5)by z8SnChdY|3Eee)O;J&!)fDL={~5oaa18iR4fIMbfWPp;)s4~(F=f7oo98a`20kgmobAL zZF|9Zt@NRa`O~XT;tXVM&YIw(HOmfYCoS&H;a|g zM-!iM*FsA;?nLu~Lsp9M{2;8a?L1P(Bs+IKi#8$)PS2S7SrL zL|n3K72qZ=M5QOP&IMQht^X{P69(LgdoD^@9R_hY^@pT0j}?0s^XfXG?4 z$7Blx#&Hha6!nR^XZKN?F=d=>Ng%yvYe`Ork0U?dY0!A_3@*$54n4MeEte(ae1FA` zpx;k*GZD$URA%5J{k9~b^zHZAjE%%ZvZ>UNODt?)c32kE_JXaPYRYz6xkQhtxi^LW zD@iANSDfQ)%W|msvmMTv?Sy<7;{#-`iHC)J;oTN#|wXizD0yp2S<$mWt+2{)bGs|E-9cFTaPI#=17uL_D z=8tq}PiY42z3olog5rp#raY+&EES!d(Z=20mdo{xlM@-<8=#{tiYN)4N{=lvC))zz zm_GX`sw%aVsn*k?=POrIgYuo+^2ktnVz8F!4g5ss7#iU9T~FzQJ(X1c$2wXsK1eT> zk70h)&ZDoMCem*wS5WDz2brPZv0T=QVf5gFeEQ-*J#})Hz!`Hl(uaQU===L_Y>Q*l!mj{obC4KYS?SFF7ztw{Ot+7!zz11+u%wZe^#Q_=oiu3-NJU6XuKw!ulPh zj7`L82;Nr!5;tao9kUhUq+UXC#TnkIq z<@3lKe&y|ve2>L4qGxBtyDr&{`eQz#UYr>1*MC9Gn8VQWR|2(Fk6=P6!E*}~`1bxS zkT7_S9F%BAXU)U7@U{(k=Bz5*1Jd#0hc%EA;>(X4q07eXRDgy;Z`P!~6jsIc)8KKj ztbd9a|72GbN-s)>H0xIU>}`Z8Ge2>>`2!;HCiJ3{Mjb_i+!%PNKbk)l*-_@Kr~yOUb1|~rmhH=qCq)PD=11Ll4BRwCJ4(kOyX6`l&$x-}bYRsF(e7QPBtty|JhUglr|x@Tl=vCe4Jv@%tQTNdng&yk<`H;yTG)HO#3nWo zJN8|s(Uyj+(o82>F@HULoN$#SZ$A&2r8-<)!awSzy^geJEAq33KO=r>3B)c;kv=q< z!yf;3i@Pg(nD@A&!p@1S2Iud4xVKii04|k8^>;RmKN!UZq)f#huTx-z-CK@5lE?)V z7@*QxZ??rE6a1V7o^1O{o{nsUUMULQ>`O<9E&y_fL+zi%IUzMLd zzeJSrVkB5Ou()>LGrGO`EG! zvOqnOrN$K2+(~6K*Y=XpTT`I_$7D7&^fGqdZ>A{^?5KpmD2`sd9W&MTlhOJ;@MwK9 zbZ@eN1!V?&S8+8hT3d?e=hSddd+c!NxR6)8@D%Fna=_@pQLwa-=f|6GMCk>i*#Z?Y zIG6K{M$k$)c|nG6R@;U8JLGx&(HZOkYe)8T*FHAutP-3}d@l;E*Fvh>X_I;?Oaep@VbJ2Tc}?VZ1%Uvz?-9V)`Xxo_}%SOx3Ru?aebo~Z0I8P;Zu z3E8Q)7~bqwfxVL^W6g~`&-X}S@Oe+!EMo*yab}H=H#nRl?vrRPJCV}g4 z9S$Eo-oYs`4gO&1Uhs8D!tb@ET&F$}`kg-Ho_ZrJcI(EAiB)u3h7O;0Fp-=Kn8~DG z$`@VuRLX_!`Aj=c&qUYyGVG0dj%F>7N%x%+Fm|%GJM7;{*8iRW$?{c9TV@Fj$~lVr z_mu(neldOaQjFCpEEetV`3ircPotQtJpDZ3Ezwqf&ONWnA%-1e*|Yuk@Rr41Fy8r) z=~oM4&o1lD~q3 z()(T?m~3{&ZlfzWdUh!$TyJDnk2y`M4axlXp z&1Ws^=b|3lW3X5yrRpwWm0kkSuvr~J+L=Bmb!C<;I zezh)zacgek@GgJczkVz+JL<~+T@i~5Ee3>r@;+3#vz~ojXGU&hb-+g3Tv+WikxU*L zfPcHLVyurja$aI$gld3cv7GTc%yO{h~ z$Tu!u!G<{BL_5b6y6>nwY6V5X^uxX|tAVH9pWeW(^&{D#-Y9`Dw?oJixzeBES!BZ8 zMqK#5i`*%i&j&fZLodlGxaEl!+vwi`Y9DSA&8epBuigUmJJe29O^x_|$4KtQ4qvb{ zG+>_(9mMZ<72u8BEZ)UB5GrdvgXxbWU^wq9`SW2imAEmSc>Spm_L1K>=L1<>-Tfgp zeR3mc2;9Q?E;X1z#!)SuIjHe85kJl`h4^_#XiY#dh!iHlx40ZKc1I$NXcnWNQ--sK zPkl*gvKkt>{h~`!7V_uUjfC0_S-51nGHi%@3OhU%VWpE6N%%4!*RFa4(>hnNE$y$k z-dt7o_o^ZEoG=wTX0iC^(=c7Z>6rLzgDpLEcyH7zJgmWHh1er2zwtwZOv&mk`?nHmng z=gLDeNQtr&UA*BJ{hXaeU49McGuPUp*33$Dx^9Ojt_I?U{BrdEdLEh_^P#0Hh2Eu4 zX_n_5_!PVcZ-sji^(Dph$s>I-UdXt9FmHmD)+6|g&$noylPqLczT!6g>?6<5M>E51 zlj*Fqn`C9=B$&3K45!^nfwIZ_v2er!Og>pnA9_rLFI4!ZFlZ(D!u#Ms{vpQBK*)Ye zw$UBxO6+b?AZ^xAWX)O{;8WcV#{6su7cb)Avhe;Xn!FoYY?WcnBpJ9nE)MQF=kiS> zed+4FP+oe`8kC5O$43FnQQ_VLYGy8lky9OE%Y#dJ{9_|?d7zRc7KWjL#Vf*Uh-gey zKlvS6NA|h>pvqYsnPPe!He9&|<_qpa?(zjtuUE_-=}$)c#2N5lybr%~R4qWs5c9|F z2hsGG^^@E2YZ|Vs6 z+4hhYF0iB7@6F-W%@*P{CJeEYqMIq7vWH#CN@0X0rzsUvFDpTy)Z5 z^pYQyHWRNUovg_$HqX0yIFQ-!3un98F$6L=8Py7yWb=_smLEyR z91UX|y-S$niHY>`b7U5bX(Og_S83(VU!d`O8@f&!14~!G!AU{g!tGATQr?T>mHNcV z!B$s@7_pE(_-YDz!;hm%Sq4=o>cZx+we-EfYjHQo!hc(J;f2%$eouTGXy;9V^n#0E zvRfABkCTJNP0xY%67i2G_0f|Wskow3)}A%Ag^r@d?4i+HV zB%W|frOc`Ejc+2uUB}SDRf50H+=pUKfBN&hHY}Yx7NgVZz|P|$`4+So|AtEP4pq_o zbTJoXJ$sngAWdlN8-cTSsgm}|-@YD|`at{AdsR+uo@0!e-7pjrq z*F#`aT}F=#?fZy5>v+W(h4$Xf&x zRc_LhPdxM(?)tw>;2;)zYOnH}*Z8v^oXuAG1MZ?u1jO#?yeN1K@ObG;gG?Lj6Y^ zA`f+yh|Yy2@YuK$J8o*Q#kXE?-!0X_=F|ji9sh%>6-^`QtT9{V5CvE29H3(Cau#Hd z05wSinZ+MSZIuz)Zt@29OAWAc&!PGC6=-mCgtLAuejT=&Eg6A5~d3r zZF#!n_X~JF%amJRJ2n4C+70(v#Amxcy3WS@gS!gz84ajBP5S*;msr zXHy*gTYOH)cgpbX2S=A3e7OWgYOC4SACqWr@G0m?`GL=4En$S^y3+ZM;o$qlfoP3V z#WgNRphj^6{G4u#ui|7;4Eku14-|YBfbrqfq5$$a9phci>YUxIH}^#)>kAyu?4s0w&S3{Huv6?2%&QC;mR=;@>FRPS=QMJ zs+h#?>B%BfIcwN;wg&o@9MR58mrYovPr6?p$E06{c<)my&6X9TT4zp^B9w%WGea@6 z<1ndl??D%3J?va$#{Z1zrMo(>QF;3;;U>19&B}_!Kdw5g_+tRC)HU?eUt@T1<0@Ee z>tzCSk7C4q2~cWHM$tB1K7M=>H_`1k_0S-^YWgUCh~q$MivwKSYRQT^*NPtJ5v%xo63Q1sh?J zz_$C)xe+{X^R`Jt3Fzf-iSYv)And9gzc|&JUD&XRc)8z!{h7jEKK2v&@76o~+ZaWQ zKb|D{+b-kFz*jKwkrNCx)P ztAhRHWHRjP5jf^%0|qte{KVnWRMveL?seTuKY#XxbRQn~Oq<5MwUno$q~^2vzXT7s z#tN#_bCdX;(%`p;w$ihoPt&p|L+~-V3Ys*=l?C()j$Ow+@Tj1i)&*)qwY@Y=J$SlI zeM$tH)e1fXs}!tmY{!N%&jc>xEH?2xz{ufp;B@vO4r}^ITEsM=sLPUmmRxmiZU~ua-9dyB8m!-OR+V~Fq zxAhsls7)xtne%#&EKpnEqH0Di5qcoUFuyng$2*V0s-3;$^eSIeihKkUr{;4N%X#kH zo$0K@k%{c3#kUC)vQyyA?+|(sSvYNBI_`YvLE7FvB+sTF1NS2x7-lsBB^*cM$Albc zesh}`A8f!{sYp_pIUe7;)nV{`p_|e!_- zws!|Y&?&lo`#4f7Wa?cfj)V2C@k~*M6Ml*QNEV&ZhP&FfC}BEYcz#*)!DGK;_PmMY z-@o5n^AB&*IPN;Kv+e+1IaSi**++ae9q@7FThP9kBmB>Pfx)ElsCeN#Lk;Iqohl9N z&kGXz(hJFm%r*e$R3g9MBVIt|vv|ULh#@Nt* zOY?A@$0qo?)C#BX`b#w`mH4P@(r9|B2_1bFiGH+5gNnZ$bpLk)LtlDgv}rT_IraiG zFzY&1EAs=l(Yw(uJcOL}9f4ES)%agi6k+206ExjZhlTKTD7-Dr?y_(Owe1Bk#pEbV z{h&b47nv|RI_mgNxd2z3zQ`nxpU>KlSA}ddZMb$ok-ZWj%Li=uKy{OI;k%d$X-qyt zLM*DtNTD-!Xh9nkI}YH`i$*k6Z6*G31H}D!BW@Jj*@I?#a7;)X46YM?M>UlG*t=Vp zr;JdW_`t=s4${#oi88x}z=3Il6SKTwMoIvCv&0E^ysyQOGg53xoE}E{mWbjcwa_Fc z5zpDaq?;s8KuF*iHvY_5{@eN+{ObI4?qdyuwckx)j)^?4*Ts_&CwF7tnL3hxPv}cq z3e0}DA6O}oK+VQnCqE_2AU0fzD!hq-y1c~ z;COa9L%y6*fCFVW$&vAsK*w`FwC_I%oZAerko!!sFKuAVg#2k!kvm?sJdEC2H$hzb z50n=RUd3ak>}AhmPpOx&JkGBI1-B| zUxX;qj6oY}sn)z^oa$P@OsJy7aAP_7>p2HjWq-nnk138>mBf8qJsT@7-xEB#kJ*RO zJ;ZwUAk5E^V!IbzCjGBo!Gp1cb=ocgUr?HD>~Rq0&N!mF_!Q0cV&KH%G|(~@c>Py? z(KbsjsIeanChl`U%`yS3R3zEKHGf4k-kDu9`ZbuXxPe#aPvC;$(#viSkA!6Xj{-C4 z1%@?Fq9O4q~g#V_&NU?Y%LMK(C10n`?~g=1#>;F8>3 z?$Y*!tiPW>`{;u_)gSg9j%)F(j#;MQWn0d=s@FlW>SE5{ZZ331A3-lyGu%uAuqFOH zdvFyCvD58YWB<{-M?w-kHF*K1x{C231~zzT_;z^UHWk8kCX2oqq(jq~5hx~^2G;|t zaOFB39KO_veidBq>9fwloP$=Nv8Nryr_Q1l8{U*z4^xLh$DbrqCzt*n~{{rY4~NHK&;hm)jV_@S5K4Xj6 z1qxS|qRrU5MC!gW&2M$!AGFltYD+iF$hyq=d7lyPL4cd4AEnXL6`{9FP3ZMY@O}?3 zqgTXDT536pm-Qf6rMDX2>HR^iH81cZ?}pCZ3=FBv!l!GdvU|d;X>9N*8fIM!Ukau` zPwF3T-t%kpcT6v255I@%znuhr-wK%Z=M-eU`U6@YOYQXPqPX3@Vp$*1sZjWe@t&VRQ1qmA&$ zp8e!qlmxB&wSXk-7nL#}5sQUO{Q|G=^p~ZisHf0{KYiF7lU8)%FHON0&}|NdpF{YG zbtBoe3O~t;HalK3H-H?<9*v_Fa!^L-u}QwY06yGd3`omkXH32eZpG2;a`8E=zlkH3 z==4F+SR=NbD~C$w3v7_ydUF3k9toAoLRnQ^>eg_(Og{4q#t=<7aj0H2CF2Y2^V7$! zX}VdXpMbxOWolG^If!|lGW0YP^hX2&Y<4>6}sPt|QhOAhK zo0GP{jqDBNyo&<=mB!?Z*$6JfhkDNH+U$p$!Tu&2jMvBfni z>`EUYd!g)r{kJvwej$niGY9iBB)KZJ^D!rJI}asVwrUiH1bYjzNQ@) zctwnNJQNKF)g-X3Xg`FxC<*g>GwGaF0zW4z^96MR+wyrQj#Ovi{=0lSVa8IJaPl`6 z>qkJ$s!YHNasKLt8kqFXiQj0+!j_Tg^iOpTp8VmA1NE|Oxz|Lzv^EdNMH_OjuN`B~ zYm8#Q{qV*7O`Bltr3CPylVFi?FU0oAumJZ_%PQaokaS87_EVq#t_w>5GzU?3S6~P!uu+qq$@VKU51* zUzZ5ENNd*9#hShqa_BD}-h$q_P1rr-A(psF^P8IkX;7X&63-A^fADab#}p&@_INzE z+GIC&#}*JnNx_G>@eZjUCqt@vZJ4RIlmBcPhDR4OypEI?@0+^}f9-mYSss?I zYNzz?U-T;;B;W;Dn-<2t6nYlpeeEGt#)o_lSOymVK7dbFJ>IYr1Jm|FdeLYjK^b{9S99%kUV!q`$$<0kY^;<%YLj2{VePJN&qEQs~^MkaOG`{F=WErK~=o z-7^{9ds{Y7IdL{wV;XesI!ajA=TytqkNNZ7n8@A~_ImY-aAs5` zm#`@Ka?cN(#g9O<>Kth4cZUNs8EPG;q1tW>c45dRVNUO$8`Gpfc4h>)WvHUJ#D2Cy ze+dTnX%kuPpAdc1kdK{pkF?xe!2jnxlYG2z6dgJwc)QwB=&XB;wqIC}_QwC9X-PM8 zSVk32#~tPV9I}DI;fm1e;ZHP!{qexg>+@XRkFo*7fuqst1eE^WB>cEjc{xKl;p0U{{0oOKTd{MFMa}ipFMdH zSjVP~zYh)`n^|SKXW%IxL+ZbWfr*w9f753guBi&(Cupt4#y#tpA0Na)?cN?Ro7Kql z{5r(;d45AKW-`9`FA~?Utc1nZVbG+g!b%Mq(HyDsLNE6=n4F!+4+`wZL88g3{2fhx z?K;ma%k#lr$+ygkrmwV*E@J;UkLSA;Tu^QJPb?TM{5qu{txXpyK)mFo>JEj)Q zUkt#-EBSOz=_43@;;E?l^;Xy_k;|R&DFuVj?WF!#Dj4UtatYDuAUi|o-P-GsuE>5G zE#r=Up;jc|@(SpDChUd0wu=$RJotn)Xm&T#oWJma+W6Z~AiouzxOvX6xr_dRnfvbzUO2=-PO)fvy;_G70!W!#F5!{uqf$EkQ~LN<5D%pE4bE&@$oI|y*q2k&M*Xi)IOe`y6=va%&HUd`ju zCzIjo_N7>zb)E=rVsddfOPBZ*qE)~gVimf)to7Sze#N#QrKi?P@~7QbvqLpj>~^hq z*30uTG>)&Mm$wf`l|BdZX>cFi`oI*PzxhDAE?eL?%W%x~>Lt#L#W1i%_@>z^4Pkep zP`&IZ$^@Loa)n>q%B?$K62F7d{aZ_C<~5Q+2@C#sW^Ea@X$FnjPej-AF&&6~3+7Gg z7`Z@%KD%af@p@+5+EMQKHQoiI`o57_VN=n5PcLdn=t1GVERuh%ja@$HJbcQ(WA}2a zxX@!d0p#)=X6nExyl*#(9kEggfVfMQaP9n36HaeUDJ56V<>Fjr0Yczqh$F(yz9~^+{ED0EX+aEG7IkB_n z0k!=UM2qc6Vc0*!uG%CN%?e)iRk4cQ>H>%|`OhH;K#Hd5T`UagKJBUj(Dd zFnl&70h=uHV2|e)+WyoI)@Q4*-(!?Smgfn4ITiNgbS1$HF$wH; znNgdY_Hg^pRXpn20Vk(epy_QjwrT52u-Siu%>H1`-u9AV!~WJozK|g}zX$O!1X<;Ur{S2$6LIlTM7Im7mj9>Bb z5B|Py#81B|jc4TjP&esmk*V~ZbLbraMtfmH2;*~`0)vWxbPk9kYP2sy<4B>TPs0o z_j590@@w*R>lD_1(~{I^Y2p?`XUA0^B)Foj2Z{FVjhy+rcy@V5GHLMA=CWIN!S@N) z#O?f7&@$4+;z~tu&dFsPo%+ZIyo2}lhH|Y3S}>5@%=eGz+~Xr_9@8gW;uH6t0; zrH6q{DenN8QUeAlXW?-5Cmeot5S$*>uOxj7v}F1C~lq(9naQ?JC4VLxsAp zbEOwc%kY5LOd9mL1y(z*Wsm+&rK<;9*x|%bw#<5uU_rJw)%?)L4mchsMl;Rnwfj#X zVKqP7sutzaI&%e6x4Ph09S(Gj2L;~Lgf7@+NRJ2X0-51&;H2#i6Z@aAfVTR4hl&Ai$J)YFURe#mcb&0X4X3P6PfxcgFT2CA{lYxIUP|cZavR8ed?{j z*o$$%9_iCT8+q9IdM2h%Xb0iX-NY|l1VhJXVWX4^cvL=QJur^ijnrm=FGV43*9V+& ztqQ=L@BEbS!9yZ$ob)x4wrO`J9!HEIfxIgW1yL=SUd7lCAzxl%A-frUa zb{qL-5sA+32bsIaJKQ9_ml~QYVUOlAXiBex9m8jsKM(Y|dol@H4Qe5O$uR9!FU9}< zZYPFgRk{0}W9e~)yJUL53BBLv%z_R92m-e-D{nh0>{!B)mP)l)S&6u^I3|1# z8lN%2%>WLG25e`A;y9 znL^(*M-t0zhb6ZU3jVt;NmswvK@2tJApAc)ob%uXre4V9Hvc>*T9kC29yT2v=+cGdsQIX&6SH}C3lF03rXXMOR-dCEIgm3x};eRiUx%hHF zcrkhj@gA!|Jv~Oy1tm}6>6lY^Wx5$_wDBP!6aK>`*IuK2*FiRNBx6IXYz5;a-m}sR zi|C48Y0h`g4^kKuz!EMhaidIkV()5Kyjq@3eq5LZAIj&jkdnW=Pq74xHYKyt=lq#m zSpZKaZbDUAzLT?P3|o{D%#AR$<2Lr}Ws@i9!{`f6%<=JZmYcqqvDp%w;qRA%`#;9R zL#ZdsYT`jbb$U+Ygt1dG`1N^K@WUQM7OtQpZbu6K_?giw_Qt3qX^Wv`759}p03S>j zgR}P%SYW5mX{G3bdpO1I^B2HOjnx9%HT*tvsOYrqwtPk@>q*P+i2($M3Fg`SfaAs& z@Y_Rk_G7swI0x8qXtWAmSm@)0k>jauK@;4~H>6ezlsGQ21i#9QQ@7VS>6G>a3SO~K8x@1g&A8@$Y!6Xkg~ut{zPyKQoS zeKd4J_&N)^4%x!5T5&4vbwnT$=f)=Qkww?ZCN$pCp3V^Mg=5yISxDkBR->1Xfv-ov z0r!bG;jpx@y*rccEI1BvL(Ab!rvcxUh$EvS6Yx*c3#dC}2sv-R!|S!0Xc4ywsd)@& z`ZL_r&3CRdBshHM3pqdd-e_Dk7&vUA#XJk;v35S}`JRu`KQd7KL>HSsDTg|SO>hjG zxB|=c9+224LO5{#2~l(CE3e!A)?>?60A#?qF8E3mz=)_Z;Y-(%!f1F^$B%E>Shkg z9*?1ct{kqF%M)Xmo3^sv7rWESvM51-wC5*D%3W1rt5dzq zVrMB_yio{GUh-YV(c5s2LnfI}vIIuPqzU^rJ%vJJ3C{A-2)2#S1b$eE(V2ci_WQ;u z2>WNx9y~dSUq+aM=HFrApU7Hxex?%Xwa*G6oSK3_bxDc3Segh7<1ylXRaCmWMHQwr~Lp|S8 z`Z}NQJ{4I~!?(Gl^o0fqUF}IjV-2_~le6LSzs2w`_P5Ypx1OX1sxfVqPQ1FjSn%%s zBS@TOL6<$4%Pm^<7xDF59MXz|_D}K`d|YWo`=_r5|Kb|#ewZmR+8cq@F9^E+kR~UN<-@MO1x)Wl zFnp&!*_pfRFn{AV(DxeSVAV05E|920qegky7cds!&UEU!C)4rtvDuhiVj!f>?@NA~s8^hIVZQ^R8Hp8dfVTjz706Sl76g2-CX5~f5q9&f? z{phQ~|Go>WjPe21ZUb(M?N4@hoC)r~BLU&l)H%&*mhj9mnfn+N#MIJ}TrTV+yHvFJ z{#LS}$Iu-H5|)A7kv+sQTncZcxC?{E@tNgGYgk{YM&0foAX0m5g(YLg(~AM^q`!9< zrn%k0`muTJMDk9SMv%78lE4#tVYu5@nnSQg&36zU8FTy?W2498!K$yxm@F2pp@G}KIFJVJFWd`ON7?9qBtI+$z5jDV*L7FHz%P;u zyJySh?ib*dt@SYI77ODp8BmV*9s7s$l5K{PpsGuK8M08 z{!G~VaWQ@*ewa|k=Mampur%M7*fo#mm&WvvgLD1~=h~Rz<=7H@s7)cDZ-c{*V^J`; z9oZQpUFJIWH|cG;1HQk;V01w|@-{Zy-$It~-RN~2Tw z6?{H>7Sjl+CxUl8yVD_!t?*5QUyt>m-g+~%KS;vxrHYW&8if^q``Eh)vDloYiC6&` z(b)uvAKS^i^Ychr*f};O%boQ!3ZY5Mk9KxzQ`VJ4D>Gt+>}iUiUq1|fSDNB3$xCqV zuA@NL?-WFD&t&Geb@Ao>5~8|I6rE&L$;^@SfG?)=u6idL1sxdweIq2@v5n)8vQ^Br$Fs|OFY>u_ zjN{9*dBQ!XXEW&NgPpN~^nBi1cF;zh4h8fJ?mJeph6RMNeg`stmKW~blq6giCQkRb z7dU?15+~f^?}^p4{{6Aw*@BY&F&7fmhkwY=6N?~4YBv33u^h(BkEVam_7iSYEY{{$3FaSN!Sp57K(DTwB%QoU z>Ly8$zNZVweYZxn`~g_kx^t;*)taMmZJZ}w@`J= zho<}gWm7-;K=d{XPQ~OdLw47(biX6UEX+k62_-fqz?@sF>kW0DFIj}+54d>5iW}4E zft&5+;lBg5P;)r~f}hvIoj#$UN+|=5RevEXX35b_g)%rT+5(kpf1|CFGHP(gNXV4; z#BJ|R)SoZHJvn^>jXT7#Q12bCS}sN1;|`!z;0)+f>}Xu@<`0qFErlZT7wK;2Xm6hxZ@x4MWi=h~u2CF%n>katA`v*ayBMxCE$1!= z9>Vntr*PYL&*pa4?ZhR6k=!4%FX%4FC1sK$xh2ZhI5BHF&hXOUbl4bbm*zoyU#P^0uX#H|N+8%yNqZU0XPNsoY^zev&-i_a>F`8uwl9aCXRW8k3%k^YikyC&DsT(hiowQS`;=l z_Oq)l>fFYPRa|uIQZDCK9{0O=lE6SL6k<*1fJMi3m^1n-OBT+h;_=_%n{%#Uev}9( z&}BUD{XOfHxea?h1+p32Y{B~TLR3=gWN)7iz$brORI}FNa>i_9?88A^@7^HL?mPx% zZ=0d#$`m>?hCeG#*3eAnKd|R;GgNw7(AYs6I`_;rqV8Tsq6VT-(|ZcOGwuaAngNBA z_OY!Jvv3$cvR1_pj$;hZf~M;W4!?tsq5v7!EbPw-${s?u7xey73A5m zYFO%|Kn8Y8)0JYA$qX4eeiqTlW^SEKXLV}8BJHE}&RZ$e=swTxTLuzQ<4)nUtVxb@ z9_@j)Ip^_L;A8e7364V^oU;a{~2j_7>HLTV?&GSQLry6jwb zKk_N_oz;Ua$=4xGaxPh>pN)ofS@3&744%H{K-9;i(37wE9MMCQJ{h@|IeSLI*nbk7 z@7+^ilAi$Mm$gHDyejq<+v4QNYV;0Cg|MWRtfFEk>3e?)rmu2YtR z*)bSfca;==vttW64|p-KhE&~hgf%y|3SYj4A41Y4U(Q|MF@jdsMyiVX5mi1Xg zYui5Ip-t9sH~KhE%RUEx!+cmPX5qmlT0HwU58L1DhlCv>-0b=7ct?tIUq)Rq^dM90h{3*Xt&)j^|h z#E?4e+n)fci&wBR1uIzc$DB($@CPS+lH~g4`(uEuIy0B86Z{@JOM1<(It(o_qR-F$ zfNrfwCeL@f{W8Pw(NZ4}J9(B346AXKyL8ZZxhdx{UWyx|Acki`90L0-!1+` z^k414$745;j@Z7&tx@V^ApSOJ*zClYn)-B^-vrdt83482g~HU_6Y$yLFbwf*(7nz* zOt|C@e0CCDh;V&yK1OVbK_AdVlR=qF>&{ zX7^vl8}`{Od{GnTkD35ZTYnN0LnZhS)lCkLRwjonlVI7%RN;&1?^&?ZHHiMM#ks0I zA!W~%aMTo8u%8kvELRx@{cLp@{3Hi5Dc7OYZ7&C@&3O2c31@P*kNg;WPtZ|SK)jRZ zal_#eaPMLRCgpi@r*cy;rCFbI^OIutI=xwbW-g5Sk|>PWv6{^KphgnKHR#V<_XHb! zQyRS=N&&5pCn63#FmQA?@6VXWGGyKgzAA`7cG!7{J(v%%sag&u3w_8@7L8SHJa6Us zK3H&OAse;28BEG+gzM@`pkU27L1Nf1Qdrgr1N>}!-MRI|TTzEgsS5zD)9nQXr1MvfT_-={|@0QMBrrYnczeDnL*hdDE z$9{pl&>y(Z`MAKyCJjsNU=wZe%5oj^kwF-Q`Ch?o7D-+78jEe z7jM|)w4V1g1Pb0*#FG=dHnBxJo50tzP$;vw6VkVKkrxN2 z`Nfg8`W-BMjS~$@vZYJwJ8|sIXq@-k7^Rzpg6o5s7+Y@(XO=6X&(IWB@25n=jmG1- zanVA(g?U)Cs|`wjGMw^Hl9RipiJuO*aES%tRCh%c*;=}ZN`986Hb(=ERe-T*UTmc*c~)8_2xU&FsqURLG4!$ky)mgk2iBFrxD&j&xCH zF}r8etHQVMlX`n?jUnz}c>#0 z)fjm3Pm+#^Z)DNuTj|s+Ic^evuQhk>V5O({Jv~7*iKrY)Yr8X0-f}aXx_$zE|5d`X zbv?{}dOoT9s|K#x(PV3^4pwIt3NEY*=kJ@1@W(rq|GwWdzww^PL zuMHy?@Lh|u2l1~g#WvmBZ1KQ)4Eb~dU3tFBBHpvt@i7kW6c-CZ1gAl6$s{V1xF7VN z$k46%ru4v{HDsC3M*4E&MUc5X4|J8=$hLjS5bp6^u--+Us_`zdTzx6Dex`(l51*m^ zwntcV>>#(I&;}3W_p{zI0bYJN6*ZjB3s&{~hsRzSu+J66!dJHp1f7|IY|p)>SkUMT8C2Mj@7X!@h_cKOpU--%bIYjtR0&lT!Y$L#jtz?-{Um5r|R|(V3y!1 zF_xF6V!=9OgX)SV5XriN!?f9AhUJ9!wU~W zT`r7(O(&;Ov*oWK|D6&9&%B3&Q(NG>#}3e&5CU~=>*#`F6>46sL`OdQjdGJhg|-=! zg-`fCTa{BLp8ab=Hxx$V&lxc!IH-`ZeN*9U#7nYry9IcR9AX#rPqKaAUlEHOGxBzD zD@mZEnC(Rk2ooCvwz)$<1QKLr*FEOHDT{m_tiwf7U(m;PfJLYT3H=s%z~if3!r+67 zRK@LmY|5UzMy>O8?3T9&iuWtfX3rhCKYbiDx{7j-=@i&$bV68a7f*DS7s8m6 zGW3<*C9ECvfJwU=iPztYj^beR>s!b?KochocXa`QIu53=DhJZtgLh7Y)NLlmw> zYv9J^gAUK6%KfxUP7%H>Z!R8f6UD$zxpysx%3p}eI{_Kg1_wGJAKScNrcOl@o?{`7~R4< zN!}kQhh%#b2*065BkV$9*fb5T?3!`U^ZA%;;0sPkh48WBD0KhshPms-DHtn(gwTo! z-Cx0+RdLK?p*pRz%n;<@0by$T5dWS$0=?lOxGEahGM7kFo1+99`EGMx?Hl$zgwOo_ zP70#>9uup)1F#bgK~+o-OX+Ax^k8tOHl{t>f%Z8{zQYh1~jc{Cu~r3WwIe!Q_?ExKZ{IGy0f-H`Pj^u%y`0 z`A7)yn&t_gpUL562XTBSkqDq`3F%g@pmSaWrzOoKWB(Fr;HQO-QHZL74Scs@Emt16 z6pXLM2;Z!UXXJ$nOo*}O4nKN@Uq8HrXPwh%|KtepO5Kb(w-#YdM?AQ9y~W@kKk-SL zEBsEmPE7r-!N{kQcwg2TE4n;kk6tELIDQpga7o1Mx02ieJ}duos}H?WW3YXB9xhy* z3|G2%E<@}{E|||c?@x+g3JN`BMoFbW+4e7v?voL=UbTZxDK#Qu{8TX8s}mM#mQgLa z$6%6ZL)%uy60*yd_5{1pkGsO~^57GEFd>;+K2{82$#S^WDNdhGQ>OJJ=Ro6AQ%BjX znfwj*9@>1d#951;phx5n^7d;Oe%x7rmWj!*pU-EKCn@nvN+X=O^)uOh#f0p<(#}pt zOkz?s9i+TB1dk+Eb9d|XP;rk2SS$U9r=JIdtbQ?h+qVc)R>-51(@deV?q`;g)dQv* zZ?Sp5ddYuP<3aGrnfx3nMQ=oop;2*D!25OwoSLJ+i5?grXPhI*_Q7o7u~#)v+4&U| z7k|a|T2ph(4Kn z+xAqa;3jj=n*?>75iuZ#F<|x(MqJHDSHCpUzGKj#XucHv7V8ue@7GpZhWT8C(j|LzSMnG<I@87Ff~LEbLbgW8W)G1v)3o*p@9%(CbhDPWCMj zmR3_julT^xOwrKHHQej;=X_r$3yiydqwKyXc>mCG3>jg; zxe_Pzo#_iSDwY}Ol1-%q%Y-P`q`(a9Jm zK0b`{hy6&#<#)8yI@0XBZog0DO`#$-_(`0qSK{+){E9kgo9RZ$t2 zf3WBK6>eCgH4P;+jM>|fX4FVV1$G^YgsUGfL05S{+0%6nnpzK1EsZ3Yn6{D@J{w{p z+MD2Rm(qs5zeq!^j~dZE=FR3ly#OhyV>po|JdgazX}s#53jdyzu)ysc)SVb$ z+ZJD9KA+M+{pLhak82?E_r>97=Uo~p&O!Nu7BF*@;0)H80??m?Tv8>p8m0^^_U?CzINN10n$tfBG=bjsgi za+7rFPOZl{aucIpja<^MiyIr2^Oe=?*&6OlzsXWY5lH$(4cZUn7{$l=xa@;U%$M+U% zxGB4O(4yN2PCu=M^7G;l^!+Kz+ONi4p8p@yuhXP{i9B;+^Geut(GO#mi_rOrN)TG1 z&6*FrgvYMDx9nJ|U~9rk@}=k!T#P;i2WB{ttzPkP;$AwFJNlMH-t>ri#c43>ZNJTb7c#{}s*ycg-n5=J_rOGh+yIj{sm zda`KpN)t@|^N_^)PG!%HmBIt|)_3 zlI38#SDh8qtm0?2CCqTN5gT#f7Fq|*W?Ap`=w?$Hx}r}G6-}-%zbX0T?EYIqJ3m(@ zm?euwA~DdPT?fsPFUhqvi)hONRcJI9r?>h0O7`s!WaD37923&R1k=&41 z!c<7!v=Oz}UWMWZvYb+73TO3J9QM34h4=3-fXUPW*!OWfe6mgjtsA^QqF@G07MVdO zOkayhyHDdEr$t03;-bJQYy*TG_}Z{?+j*k-T^}YzDpKwHGg+`(vp`|gVf6XuPaZVS zVQ$Tr$f~vCP|2yogPcZmS+qh})-r=_dA^pUjgAtwB;C+!^VX$Kd*;0ZaGG9N+pyUI@3w#d|*aW z8&R3(DEy;-5Ywb{Q0LhmvcMz)mwbKB%7bIr__=cM%O?@+5`KdEr0rzx7kTFN;x+V4 zHH0x829Ulc4%_#Rhk+wc*!eL&G+`^x^6fRDUwqVYBHstpUyy>wOviJQ|CG>khdr8{ zm-F5xaigh0 zexkrx^$t|HN^>`7i(zQ145zoh2?AsGVnl-l92HmLlr^MCiwqLoAQ5cNk)TH{O<^xh zAm?RCbdH0vYMT1d5=Ud^TyG2a@Xf^!&=0M=aGNz$<5q8^) zQFW^`aDPoMv|W6JQEjqxRqG*kdzw7Illlm{s#d|#QG4LxO?8mCqXU0l1dyjo@?rAm z<50tAfQmVF#KupQ_sH|Vn~udi$1n}n{Z+%-lz8FXqW!r0{A6~|%)vof+maJ+9E&Ud z$+50qoDTL3BkFU+DtKU z6Dc(DqH)jjK`G7@7Ti9H^G1Aw&8dn+ra^(eyxh|`hVSr<&9i}H8F83)tq^y88^DV1 z$JphspP1L#H1fCexG=q;N~nIgh^$K%;22$ZGU<&qvz0kXR14P$`o|@qnuUN{aVv(c zI;uvdMLSWG;}fa9e;%Ft`vNKJ|0_I{Du*-oEPxH!jNBNZN4!FHVd_FhqW8%fj)6S) zBjGS)uaY2kPv?Pwf(VL-mEz+&3Cug+6uduw0?RqkLNDLLXc?@-t}IbUQ?FebAPha&c|7^p7#3^}NAIUe!mzPZFj?LmOn6@6t6*nHow<*GpJ>j6 zzw+P%pRbOV&!kqtHN-3ZBJ>Yt(9iu%F#dyp?mQ-j_Twh8@Xg0Tzw|Ke+@A*fEMnp4 z*m7)3bYkav6-d7SXEM0{Jgn4Ipp9~pcxvZSob+J^SHe}nr5UG~@7i)aaoz?M>dy;8 zw{8LEIvRsFo?;6_OWDgQOC8|g0=5P_~ zr59tsb}??E*DCyKF$MQry^4Qte#cMslX0NtBiTyNV#~`kWNjs6^?YS;Nv0(h}b3$Vq(w9ffuI;&eTMA84a8n5zOv- z&c}dlhj7ym-o?9tzn9A1BgefaqeA#+^qCTiPDSp#*CYeSh%W~>;W%ofxC2~6?!#km z-XGt?!DsPRv>_{xWR$cpHzB2u&xwH|RzYR!b5@`JS}6BM5-p?S@%z9GcInq|c$J>S z)itWX@IXCy)T=;_;T&9i-4OKNnlfLhOCYi(1x_8m2lI7ZAawjCSn9A0jQ_3_uInrX z*)xy0YjrNd%t@1pxYaVS7FWhH`+hQ^HydQHt%vYunQ(4M8S}q!jW6!{f?h)fxJ#~Q z0oH$Eer^K{uXx8Y5AlAh>^WpU&l`SQF8J?LF%JEa zV3&PN1-rbmVYPHIKIq~7s0tgoc+VC%{b#&j^$s6W*WJLqY8wp^r!=_04?@`WDj#6qXi2a5(u@y%mqK_!HHV%t%L z_g)pa%@BrK+S0o1-{APmQ9OJv8e0CzW3X>D%Rj}DcLw6L@$f2v*AE5Xd=&*De+bWIBYuN}KYGdG9@Q~bHG{Btn289a!14sgO!CW#C@~>XtT~t4W|7FxV9&`E$U0Pc}?MV@9-MRqg0%%Eu5H27ck_XlEJFH>Peg8f z40Oa4v)m>#aPPSbxnJez1_fo5-)qK7r_N{d)W(rTU7KN)%p-DLPZ1pjZ36kL@~q0d z8!SgZBbFO$QElmNc1dC($=YTIH%=onE;qoIdGCqQ-)Q`i!7=wEBHTBxUZQ$dSzxJq z5UQ`#usv_`vB>*5F81bmos){m@|07sY>O$zzIjYqHoU?I4I-Si)n>G)c)+Tw4wJ$A zGl177@H831C_Qi2u+m{O_^zdxLYQEoV zvl%u{2@rfr(S%QY&UrrU4XcpKLkCYu+I+Zy=VISriLiTDA@OtA;dpEC5V@CF!q%tMFckwCZkCA)-}gF>=l*rV z)RS*9Kq(bwO{{b@q0g|!+X7PEjA41s5#h5lB~1L4AC3?%XSD-og*~rh(1l$jBl_Yo zSyPcar9Dm9TX2TxssF=%xlr8kRtcYPDPzP*22W17%0iakfO*pP#Co|myH_mDx!l^w zTw=uWROm5Ca&W~>wbOAc=|VYuJy`Ry1Rw!cs@!LW>&9CgpQc5)Lt%ZP9t$5MW2NL=HsQCVi8_Em$Pyt!+?nPUoFtWl%V>(nM+{n}h%2?0!AAEq7G%gfe3y$;rB695sc8&riPz-p zCQjiQdDReYN_qE84~efSW6R~NnOz3YwA-0VQf6smv7r=Q5`LYLKp%*R9}TmI#DrS^ zF5r#(4Y;aF8#vx|^L+V3@N?H^dHOq`p-?~}K7d^MrA_BsmgBIc6}CQ&hoWy$G}dl5 z{dM*sblq8t=E{RCN?D&twr%2s>-mm&=wWm|dl@g)-EutXU5b6S0=nk48#oz^rR^E& zbioP-wzV=Ihn-Sk)DkCf-nJ36V@+TvIk|CdxI9%j!Sk<@W$?MFGCon90m^@Pw!qv& zM1$`=zT$bE!`cRDGPDM2lWO4f7g=g?MulddIKl72@OR|c4jdg=AsqXn1uK?~nz_r+oz8-i0Zfh2h-?Bwa)2kHCR+*x1vcE89Bn9bLE;z0z8LcKF(vdZI z)A$mOOWurrF9qD<)1~O9;DGU-k03B0ip2GN0?%!FpywqE8lQKfSi}_+_?f_sb@Soz zw;<}h?jAT!|4%bDT<3d@ekK4=l z+;71bS%8nuV+36_<>Ya)BS!zQBEdJhm}GELzwwA1#5EhAn8^cdwD?&F1b%5_Hh|t zepHGnUV9@5e=rxVS`zTyKL^gcel=d>cNhzKZ}@+zDdeoL8%Vvcfj__3QawIPED1Ic z)LxIHwogh(|Niwf((oMY(tJhsx>ljBb-nPLnGNc+K7>1aACs$AR@gDFnB}+@;FIZh zgilK&&?_gIc^`rL_@eIEvT)RGtbhT-EEe!k?j9*k=JLHhjya&TTE zyV5DjMaBo9?(uR=b2r8gmvz|uLxUX(&k(M5_7F_F_y~gRT0zq6m9W5M1Nq+ilIWk% zWbX&NY1+I&7MlN$ow#)Z?H72^L;492-6=^`_lR?nUv=>V|9$38HUR?U8VvX3?_rL_yP2lOs1(vbzthN0oZ@!9xmM4Nm2t!NU||M zlkyS8#j}#hEZxgsSziQMtyAErpDrCrC}c~4Mt6r|_|44>NN#g2yU< zL4EC9R6i!^xNwap24^Rva=q?QCIpn0r* z@d1HVEzcTK`9>bssWKhuS{NZd4^E!oJwh?r^ttGBlGor*_o&Q;W0T{+dq)~vDnAap zPt~$rN(L~;dOpf%KEv=RS!xu~N@`bzfa#~naHGtZZs<&hf+d6C{MCgH6=lKrb*t&B zi%G1?)fSiDDaUKCMnPS{Bhq_qK6vcr{qeR^+~v2oQ2)^rm^UU4`okpYocGGucOwip zE+BN>G8t~sb%sYb+Htw&f(ECjQRMZV8r&~mhd*w_vu~n{_?_cB%;+j(iAFV;<0S{1 z`Tzay;VZDBVHGMaiDk;8DO=F`lJ{Pku^)v-;Jk4)7vo)lKY!|Sn=flJ{a$l8HO7+j zoqvXkTup#oPQ~<@S{vgwTC$^w+SJ-Tlu9>RP`l^v;HZ@*U8X4pXFj;ov04X(vQNda z;pakJY?L7=@0x>=?W3u2ZxmWxnT9)0dV$J%EpDZt9e#0!Tv@#+m(}==$VeS!nvqfP zNWqS{R1vCvoRi?L%f(e=U|RP!%E>GnK9%)_@##eWw4bUXp7 zMYHhy!5~09f%9=AX|fd06@TZ-bjS>%YWV}N%#Xzy4GX?^D*-BAhWLfGVfYglco)}( z_nxmrxy4l^NIZ)bZFPpJrb|&QM3g=YRe{ev%Q1Fm1)qVOA(>hu+0s@Sm=+ojpN9F_ z{eGa44X=65p&QhzD+;$6@Z7qk9n7%*A^fbn!#f9JF=bF}JjU6PQe+%37%fJV%%mFHf3nS-HrLBg5UB*DCkJtCzGc}z!JWP6Zvc=d3tv$RL

$Ono56bi3oLu3&EJa1yi(p5F0dASHV| zoqn$derB(x!JHIMcq7ZkiX@=F=+?%QC(6nGC*`2yWANIHjki^1m46!>jK53ci5=a{WJMWE&|JV| zUGjv#uDxL!LYFZ2HV3w<M;Fxx`OPu7{{MF$Jh?JwJ?7nY2ao*~hOe_^s!}e3H-7DnhEFaCa~oUP{ki?b`$!D=a9E3deqPH= zHhyM_Q)f5IcY713cZtIJT^q=>weHN{-AR<#KW$5mG_EAKJ%#;x2tw`o6L*M)P`}g1Pu66&o zYu)?DU2C6n_I|ydug8N(ZYU(|pgT!CC+6-M^oVaH>?gm6?<3#;)e_zQ0#b7)nDj1C zCT(pV&J)9DYlv=vG<7`ZR`07bk!X+8ArjfB zc=DFFqRZc!h@aB{$vxCZE{!mu8DpoCr=3rT;bAZGa##kr{qG?Oi762W<)mC%zy7z# zC*crri5^GWHmHyqTRYM6UxMhgO9eUcJ%#9e`$gm=JiXhdlzfgDPD3_p(G8w*^o`$r zGU@cmdhPLM)IY6L)Zcqb9MvCQAGoDdJgeuuc+~0y@rmlk;(2BViSo!SGAJj8NJ&P2 zD(}o9BZK0CJn@hWLxYdad{iK%zXx;=@Tu;Vxv5fr}=f) zm?i~oy3~HrQ;YFLq3tJ08nB^r7H=i0-NWe9nG?t|zq=xF%P6vbSb=L&!Y#5>`lQI; z>oi%r{}p*MwTpb3)-PJKh9|M7yT#v*UJ?K5NfbApdMJ)ux0pPd+bdZ+W^{(;QKDPw zL5}s_ca5vQK`w82Oj`bY6`gptg0wwMBTgHlxHX5K6T?Gm#8R13w9F!h{BSFEbq~uG zKQG)N4iJoqi6n!iJZ3y+U{)wDovucsb>9-nkQ|peIai$497`S>&Lt5eyojZBGP#(f zB&ry3hg|;bM?Q^OMQ&;ABZhh7$sX+%@viPmWcb`egy+9*;{Q$r4DqLfc()tD^A~Zk`#U2O{S~N zCDuxZxwUdCr2N@KvHT}x^3p6>bZa?B*57~Y+A1Z7nzlJ{!>oVPu&rZJMfM~*tWypB z2-`~|&e$ViJf;PI+PMktw?*l)GAOh#S_Q;U@adreJ~+?3H3k z!>5~|wR2U_qH1*%t#pv~`qWZ%qLrRm?~hi0K2025H%VrcqR^X{=2+iGp4P}bMuFeE z>t(tNxP=C;_~K4UW_FINIEl#O2yO-YfsXG0!3+CLl}-aVdHNcw}$3$9VWos&qA zdk_5}Gpl}U)l-5k)#zyD;i!F0KT5w>Ooz$O$4$4QNM@iVHvLda@N;<@_PdJO&M)N# zAByH|ehfst8$h$MJ)<_HgHp$I?_5&&$hmG*oSXW;$eXp899Wnqlem z*P2Eex9=%Er73y-Wt8y6@Q++=P~d&@W6&rgMRdJ;1a%Z&Loajoqj};pM1AjHdL*@s zlb$`DJjkDlb0mDWsVd%*41z9kk$u9wiMvO~N_;I2b4#dC?{=iv z7f|AG6vZkCboz&4`i&ftc;R1B`}G<0wpB6lymy2iuz$g+e9RVabeqkO&3J>#KW^uy zZhT5Z>p9LuV=IZux{g$zAEHTn6FCpbypWP{D(z3v=LUi|QN^@c#>9^w% zSdBshb~n*IMILRPAB18SIr7?TFn(~BqmC{fNNeM9E<&$^8&SE9v~4ny%vwZo|F$lq zcRaSyb$MoJLAZoVw{kmOw!jp5$m(+^t8QX?;U$XN(!`yxE2O#Sx;dqBt2xh)3D|Gt zdJ@db5Yh zAH2t{c2?y-sZ65ZW^{7bxB#x}dnmfrdlG+^?0uS~TT!ZB3w2N$fky0*a9$n{L2l2h zh}y_@QNizdI8dgKJ5`}Um%kf=yz>fBEFVmT)CBH7mmcD?{W4u<(1uCy2lRSE5mwwC zL!1&{a@QrEcSCb!oPM|xZIoS)7LN18DgI?NbzHKz@nJd_v1c8+bmcr(Jw>0-sQ4_( z@zJ5vgCr3Fi&So&w`BhJ{Udt(zyNp0E=~-3wY2b}NaAljKzk1zM)b~pqTQTMWN!|~ zDeELNe#Xrx<;D+^tGu38hWeqQQTn+0?geye>oc@a{0-$!{ey!0UZE?`C-Ba-uEcj( zi#X)E6MnJe33qn&Aqm&-5O=vKi&!4nN82~aqCYk_NpGGD>S8MBNA6s)t!}kwZipo^ zY#hNYY?@1C76tIk<1{s1GJzZTrGvI!wdW4~Q%1851UfDLDW`FK6_vI7Uex%0`=_GoMG3{#fwS>NR&$ zF3;8IxFzS3oKtk8MA|1+wq%I*9#-=!`f-uwQ9a_<^>d5?-!esI*G5qx&j`JdWC)^Jx5VdoSi%e_ zBhDMji9voAbv#>vgZ2$0nl624?=DFWJf?`dgV)f<@n%Sab0)E+4d`2L4!V74Gupjs zCwhA0AsH9E7!9!-!dFKW(3G{ZR3l#qj?Z9Ya=M{z!h3Hc>ns}q8Zeu?O8 zV*{RFu*5a1AQP$Ao#3(qc&;QSgYNGhP0d&P@vwLg$(1W0=;rtOF+YZ)j$^;Xk9{je zuiTE(oBNB!RlFZni!P-rmTn=Rq#P?Fy_^e*t}c@R9V`Tkv~^OE@~h2a-0Z5v4tOgOV*HIP10&8neiRvV3RM zy;}v>*Gu|$B@#aU-ft+(xrW{fyhOfeOytp&^Z(21leyO~QDeu!xcB03?!mrdPSCpWM+)LT|sEC2i|HIIA7WbarSQDK=kAFI@PABC9KD z$ls-WihL$r9XuB07KPF`e`M%@$sD8``Hm(_+$r0-E)hXdi+`n5i@;VJ^~P(`%hyjK zZ4+O*PP`1K^I7zl))H=aX(_U;y+gIGo+166SLh9vgbvR*%6W0QXv<%y!!xtDI-rTVAHR_GGxD|M0}2~XmD@6PbeeLudm@pC2b%UtM#bVutV6$w3$w5j-#6YZKa_bV`;CBJ^41NUR>uG zBGGerkv6wd@^e))GN>*lzq2J?pkN!KZYXgIjy!;5pUR_4+DW9l{R|m7UGlxg)m(3? zI{h11fqv|5M@toYI6i5iWcERZE^0`jHwUU{h<6(G+dhbnMl;dJLC;;+W&J>TvWL*8 z?RQAex-5QPk}p59po!jckEB_tL$T7R0`A;?1w7)m240haxq>NyTp5xC*zFo7z8>j; zO1>-T{DuPf2RsyxnG!xa0kyn*(q97ius>yuNN%jo&2W29AVj^wxXi`u`u=ET#a=z)PT zWd8k3w0U48w*Qk(W_<2MshtziHaSHUKh^?SIA0_k`>K$jG>9*7(#Q4m7)sZ6z@)mC z_WK2seJ15(<&PPB*0$U9=OAs~ZK@6NzLZY48wZmoLqfR5|E3VEXv?`j$;W5K&xocq zLqTIx(BJw#vE170+&!@xA1v`Lx=Z|@Cz5?}+MY6UaPCS;E$a}U>7|JRx-N>>&xk

y7C7Hf_9CN;1#X z9D^2HS@Yt(jDE5+qjTDK)1J$AYv|SH1SJ$9f z^dyd{y+WS3YT>_kq^Y8VgbkyWhIHztBCk&uxa$Lbv?jX{srB26F9c^2@yl3r)LkUr zBvn=a;N?>8OTsAnTrL<*xRpY`C+gE$+Xb|$vw?ou5y@@3@gK4k56}U#zv7sl03LhR z6YHC{)OE2JN-xS0m8H5P`Ko4egwvwoPNgLAP%8C}oF=Kq7IHTVOX+*>YqWFIezDH! z|7e1DK5YzL%l+)Hp)2=Ga{u4Q)2O*4X)V7=lDi#>&hC^!pNC2^^o1-q?vQdRu9OqPiyF+d#NZ?y`9TI259XAMch5^0jV`IMLv5@BO`Y^H1E#< zcYI(8+Wl37dSBd))h4G7+24mysJ$^geL)88bu}P% zlA6_8ZX@UGU4n|@y2z!0b(}eS&Ur80gpSlMB-v&$=#4~DEHX`@??P9gZws=~xyNrla+VyZtYk2`z%D;d~Z zN-9mINJQ{N8XjMR^iJI*wOVq#lkaw#ctnnlSpAZ_RFg=5XHO!b$MyJYIZ|kFgezC* z_5eMEVf@xT<14 zIP0RM9;HvuMpqEI={Z+47gU@@6YOu)rFps9!>|w?3lF+V@exiFHIb8l%Wy zKeXBgM0*5k22+FEHb9jRE?I-z0NgCuKxx7ed5~+lzz~)!PD*n@?7u)E!bNq z=}sK2_coa!Ixq8>e7n&rP72AO{)-Y_h0B|`+Bh>;H3=*9z{Mjddg?4ZcYZy|3mu1& zcfR0OCFRh=Jm$uz?WM!a)3^d#Nw0OrcKqMlACwETpdEVC_|#1e#IbTFb^ei09L7s(|@TKRpvGV6#9P-&s zUhlC*XByv=r;RDpe)oW=kEDxj79zYfzLnO!iz6nzH|Xw+&B*I`4ps5l<+?Dy0I45| z6D>^B!WB)1)aghO)tw{ZGv!}I7Y1EK8|r#R-1ghAqJEBPqI)S;;snS0 z-0Q0N`V0O)kb?C_?yu@WkwT3t9c@u4`nTpbacF-{r}y3OY%%Pq8r%6Rj7xdp|9$Yom>rB z6!U_M8FH5{HLj<9)%Qeqw|?N(bmXF}lZR3MGzU`dmr4vfd+LWz389+()6v*fKgp9Z zs<`j_Xudf;gqtxhlP4#_5xE8RttE z8apA^!KrlU@0C=p@i2*c+)XY=O6I&IIc*=uF?5Q=6@EXtgSafRp|-PS>2MDd)Jva= ziHOj)3r)y7W;U*VTFf1s@JxgLlZ4>%=y(dav(MV^E8&2=5h7!ZJTx2sgmGp!-kdaCa^vS_;^s7N1 zy>`_?ZQoytD&A#rjy1*T&i)H%$1;i`Y4fD+EC4v3-tQ2(fGif1GG=|B3WB>m%AswlEk*2LyD7N8iidw zO3sC6i(DNPsq^)bs4OOf8$Q#6n9k@xJKyKit^Tv9XOBS=(NMPY>uG~@L|4qKmy7N-}-Epnl zsuuyA*$Q8DMC%;YoqvO@)KuWh^0rawM~+ zd+eg;=j73vk^=g##|r12^&%=R`3N@M;kK#IM*cSoAbBk@u**&zX$Sb~j-3DftbjV|SA z(d8+~IQtwC>&zEBs86SHM3H{Gr-2*}yg^@k3N`1iM*TD zsEycwEA3d!JsT`dEBr2S+uF5om%SOTp6$s)&IV5rh2xDlY2q)ER+)iJ<|yFnMf=6^ zhBlHJk8-qhe4RMqyBb?14*`F0%cW+Xz9K#QHkLQYT%?mW4~zd z>4|FmlIUrCV5247(W_2#>*nySw=YXvvYRD!dNWN^PfU!|wNKjkV7e4(XxUUf{*;7U9e=F}l`BP|o;AsIo31{cD2Ji!V2MB&@L`}|$sqg1(diLg9+LH2>iwcoN8CetQ z(xNl8V0%41(e#;gu9D)X-5NqW?x#?L?7yVac|Ur8AqyE7T$f}QY;c0q9;EN)h;?E; z(Md@j>8g7GO7|MdWn5UXq>YFy&k5Be+qdRW_Yv~^yFt6TAr^DF zsX>ySso8hVd%OoaTQ7^^L(*t^nm&=YSt{xE1HP#1&yBX;iz9cIBHe>IqOeF6yzuHV z(kAhV7%Uw`&Yu>cY|$6g^kzMmqGQJ`({n{5B}`h6zrm=?`xlwk(#H*H7=;3R^HIxV zNB+<5TJ&t7RV4Lc8tRrjJCDm}b0gNO(&xJA*nVROH!VWK;J!GYSFNc=tx@sxtMhTu z*Tl7#;?2iVy5fj@}egWASFo@TUa-kP{9Z-wt`c*_PNjPEF!_~QtYx>mwc^Msfeim0T zrHH8K&!;k}s*;(HI3#j$!8?*<(b-Yg(B?nqQCQnWdSg*LZIJSqIMH+j;)LON-Bf>6 z-Xxi2+c%5-IsIrV9LLznQ5)HHr3!MmM!gWw%#<^9FT!F3Xc_1fqHERA$xuhSVm3W}5Z zf?A};NK5_iJ;O&vR!T}yQR@GB=}3vBmTX)Wx+-$TxaCVWEg3i7euAy7gUbY4FL!6D zp#RU;tq@P<-X6m|kN2_byW-dq*}u#`ew7d$r!8D^R}xafGzI1`0?<5yC)GJZZ}>!E z&K)~pt=BRkZnhm{lpEt?XMXT54XQ%_`BY{*IFqHPt%k1u9^>xSskr^ZEa9YvxsdQk zQP8dr0Ojin;Qb>V?gwYH%1zVQW7TxFlT8wCb3+6fX>Flm{SX*CSsiM>%EQ5uTwH#B zC7#so4ioNOgcPS4pq;oKHf+jcCN&q>{dN;!DAgBwc54Yj5mwfNZ^slHh4O)4D8gkpv~G0+yxI!KZRlT8kLnEB-lS5%< zIc#^N6$l2s*saJ699Q@esg)^2daEzjqI?eAB=n$Z>zXltPgO|0EPv3pI%&VN=rzu;HHjH)L*j4NJ6%OkoQZ@&rXwUJXbPQUCJJtUl--)cRH3RRh=neYgBN+j;n|BN zplMWw)di=ilwfRXOL{+2p(vLB7h;I~8h^3R|H%mNR%i-0KP<&K$pDts z5Ew>`ke22X?Te<|!jygfTL$d{!) zKTmF6xxdhq2z%`ZtGn>S*l z&^>&J5Hw$3&=gs~?*n&``~6R7!wFX*bef$o{OvfQTC)zix>B)k|(db`++?=dm@cy6BiU%6z0{3f;ao_%tf@m>e^GPSWcK$27{b-QjHD4senAX6W z{$=?5zF3z2dJhf?F@j5f#BdLkghkn=f{RO};GuV$-5RVbe5zd{j5NvSZxb;-Ewum+ zoQfyXpPyqdD_^ktRn7M3xwBJRb9nvK04UQpLFBX@JNoJlzHIUgJNKwS-O=OZw~mT< z-mp2GZjG4jSy0M4_8ey^IW}Nb;sO``c*3j36qqspAUMsf1MgU4cyY=W983pG?$>#+ zVtgz_ZODO=$!j5cUM%FfS;O@a^~^@6h#zMA9aGP3IA?n}UH5D@eqN%(4y(T7MU!lq z%t#|TPID{0UM?0dE*&ML4{2qa-+pG;e*!%CkO-mUlym zW$UFgzXRKu$+tv)qy1BUc-k7U8(M*P=iK0x7c2q3#UHXy0DREzhEt0YpmF&S=;jCE zgob@MG~*zCQkTefJvHR7sQbeuvANKs+srzT&Sphtb)leTD7YlbgU&=Fc(BqA-o^UE zkH@c=+eH<@`<%BBy=XI3*Uo^l%2|-U<{UaWrIR21(w2Gs8!z0pH5Ia+E)%@7>R`>L zec(8<33i2L2%3{;3oDg9g-1dhc)9q%&@BhSZ;cgsyGI%}zYBmXqs)Y-Lx&23?~8=3 z8Ub|pC`9`#C)1zhonUw93@j~=g|{u`5Vay4Vq$q%X#4=+Y!1l(R}N?UYw?C_0U)Dv z8)6f!Az42K*L~T~`M1PE`4Jy@by^9wZ;c0Wp%R=mbcTrOL)jhw6?~x5A9{TJ7=ik= zv7`oNVbMMdP+Jg-KG}cbx5?=VTV`4b8~eh9)m9!tvHx?nVL}S4uH415uqxj6WC{zN zyA9uSn8#;3-s6{@ddv_05yzi>vWKr3HHR7MedD_i9z#yQ-|>m`5O#824j%%vpBcB9bfc#-9;D&iN)?giWaG3<%E1;W90 z;C{(K*pp}_ynVJ)xY;rchSo2GixV?It1S)oGDr9^VH0SUn2258+!eX3iK~yh_mYi^ zC}uxSo@2&!9JIF%gYND`ARlCfjmJ;13?oIhchE4X8-E`E&9BGCCw?+F-98o_EfUIK zX~9CrAuv<|<+g3<$L;nMR2 zW@08{y3u)jn7)BfH|i#HUZ5ld-^>T?iX?b?@HSl9*w6mFQV`U>+6fLb&%yUGIbb?1 zAN+zkA#1{MSZ16LiRV9{h|Ui*!Sn;aI`A=+h^t|Y@@W{NtO%>6#==0+CTMX!3ODz! zgU_44;oAIIs6S*5Gg%?j_-I4U`V2g0VlvNfILTDD-e>oZ*a#1nY!+V6`N;+~?_{#| z(SqGC6=9^2mH-{8;Da*Z#pnhIOBy0vGC#xi*Q>LjL|vhA_!yy4%~iOYg6PK5M50=L zmOCNrVDQZr>pqBrgzQkCnar`5`+xjV5UnFUmj5rPt9YdjbqKaVF^_{I)@Q{V) z&4LFy?qJm%1hZwj;gD21tP>rAH3zS=&Kt|wpS%J#dFE4AQkKkqie|FQ7KyA!hw}5x zWBGrbF@$uaS~F89E7y0BT!bX0hcTf;+D}Rc-*%*cx!wf z+o#{buY&>fV%%L;Q>^|Iw2hQgllgM?476Ies` zc|K}Y5`4R?4pl*v?|Nz|yeVIg2Q7U8M!x-UY-J`SFW3cM{i~>3nJ4u)UqJIdim_Fn zCP#KR;KGC5_-WrD2(`mtZetC(Q|G{~15-f1(;v6&bpU?MVYn6W7JrSZ1bKsZ1{MAgjJH3i@`1!@fjT_)HWk*p?7~mRy0FvB z8u%X*6q&M=7`-Vo!fqoEa@JQ)Lg3aeP?#^}X7TAJY=397!IH~SSc@VBwN!6HACU5VEJ0!>_1X&c^8+U-LT)qMJP7{m3$SaO)21 zITp`WpKqm8HYLF+vKW*Uhr?*=0O0Cpz?%A15G_Pv<;U9KuD%dLbva>GiH0!m_)qqJ zo4a6lRU|yRJ4?7SZvwXNAHia~PO@v(9jv&qf>m~%VzqBS-_g!#Itdubp+XLTcKo=v2ZT~3xy#uU@7km z^Zy4jlcE~saXj#4gWxq%>YSL1llTHM9piAg=4}kwk(HdTU)lW zF8aU_=fBvbV;UGPcY$zkXV{T@nDw>jvkyNvvCncd@QEjdxMhPq zbmT}2v&W>e9np4dZT??cXu#~3-Oj{TO6(LT-o?*Gn=TRt2^gRLr-;LlN?Z#a&T-dmjuwRq3 zgjZ_UV40p2M9tH~9&O1e;hF+`JiZN%eppPWbUq@bW3$QDMi*hiqyjeoxE3?>Q-St5 zt~fl?n3oH85YDEI5_(?R2yZ?0;cogC7&)aL!Vep=E9e2gd+{TF-nab_+2#RlkB5M3 z`7D^y7z}=m5#Tj!IrDlX%c&fWW_agPJgqslN|VHG$`7z?BF ziy`lzqR=Fp%=`l{F?eLeBL0Q*gQSjte~$)C?59cReoX++*e8b@fH-Y-VDVY4z01ERIL1URSERfV)=20z4 zZ|ysa9D4{{wI)EZ+)`lvGQx~kgpGET7xXIK1j~))g2G87;isp(&@g2m)5$WzN1i@n z)@>)*#T~8e`labaSM>#1ls%5~dH90`4c^HfdIKL5Fdj_1$3jJk8+Z)s!!=hd;n=A4 zU>4EMlrp`9HMK=Te`TYjX0-zrhaUnHk6M->E@hUh^@X9T5;W)LYN&AUgCF+Q;N0(z zS1&lg|GD)KChpkgcejqIwl-kfeZz26R|(yB{2yKVcQ6W!c!&evO%?14RD{wp1>s85Shyc3 z18Eyd@a<|#u$+1xdqk;$N^#l`!HCczxoTTC`9ZsxG`~?PlV@D(K7J1D^jF99cdMME>!hxcv~s z-B}JPvw!2vIsW+TDNj0cSufdWF^D#Ay$%O`a$!O3a(JoZ0^7T_+3ey57IuD!ko>lW zt(@|R**hM^VX`l7w-)#LVaNj*$-XC(vTZG5;pg03Hp|`OgCZ9N0qxyBV13XY8qdeV!hL(#=0Ou#w0t3Z z6=WrR>yQz&Vnz%9csa0J#lh;N9niT_49}*Wh1dtPVW34{Xg>Xh4Y3(uyIZG|s3k;n zFB}1b(P5BR zXb&%{`ms3kt$0B83U9qf5&RcS0~d|mkPOdFx-b=_FUEo9j&U#( zJV3Q)CfqjZVO8DM{A}-)aR25P=CXe)v-3}6>aPf{ayWrc`tO8AqE$G4;Vry+(pP*W zV}XGE<_OxQF~X>316bTF4~ErRaI@}!@NkE}uymBRa3R116r=Sat|0;vPF#V-kLSRQ z=0F%Q4iFTMeqo!1(=2VJoG@dSz)GhK5}GHg3O25;Y`NoEdV_w%qRTxv{8SUZ@O%{I zp1RW>b!mF9-$F2+c#KKqKVn-7KO;pSdw6nt4@4H536+Yf!mgeDEP8GYv|kQ}Dj8$& zFi3>OevkOC%d#`TC-s!oLdRi9b*{w$X z$e7)Py=^9Nv+NwcRJe$5YB+`y z<@2!BlY4kYVJ=R4HwglFror8|&Cs)F1njH>4!0NJ+S^ldOCZDSe@0-scox29i zx&I5U`A-A>%Nz@(g?FKF^LFT6sR_?t`GHK!SN!gH7FKcF#PE)B?6!k4R-HFkn6pq< zh;K|3Dt|;kL`^jG>urJ}{WPdF+X()D%AnZ(Au}5JmsOe15uDXKu%A*me!{lk+XLZz z*u(8O@$PSIv)%<{SF3`9FbpdXFvE!?9p}&q4ULI$+-NfYB36V7KfRdPGN+#;i@EToMQ4 zCY-2?Ou#f8u>LgvAlVQM-nsH)x+yD5T`0&eUM%#)rwR^NEMUmPZ@9%R z8eh7)1Tt28!^e*!V4QCcTa<4otd~j?J}y1WMwnb;Z7u_BQfxavX6tzV>XeCa+w2UM zQc}X>9_WIfC||c5tw#D+{TQ=YHaA89WZl!H2!@{mYFvYs*$EmPei*>5-|}5RL$Vx7aNdk zR)ZwJanQPUFsQ@>C|{Wde*c1@r|d1o2a+$iY&xR=g`k z536L(hCFLsQ1MTI$ei&oDn}JAFSCI6YjSwgw{LL3uI(_fIGovzuHx6tkKspD4dCR2 z?f7WpLsYnVF5XbCihh_+qvn$Jaq^I=a5ig-aI9~HklN76rlkI0>ozPGM$Z^7+}S2M zTc1(m`?aso!@6_vli_P&cg1NOHB=fFnM(=w=gQd5xsL4Xoe41IL@q8I|O}k+&Ki%*?`*L}t@M6k*p?;AJ=? z68=Hd+wb`6ji>xCPQ`8N2Ok`oo5hZG%h9C=cGEF6g|y{&{;^|;GhuArC@L+JLfxl-AM%#O1DhOZ!_~&|(E3*sLJqHiUzKKp z@`m9;<2X*Zu2}(_PK3aT5FglNeGu+FjR*U|D`B%kvhdYnk`P!nS-@dVP&0Qp*agT# zcLPB5n|i!(&S6X>ACwQPLr$z5b+>mAhR;irf-o7m83b+ zi0=9Qd;hr4bMO7*{CS?U&-3iF&R(DOe!pHW(f4RG{&|vz17T4s-Y4qKvW+4sz%BpyB0m4=4+vuTH6bClJ6 zjIyw(qejx=p$%ArG^P_66aE>pyTAZrRw~A9{~Zsn^%T<^^KPQmuFue}fZ6soD=Tn} z$P#kqn;!Y=j8Tj8dRk&Co>>*wz`XqQi{V7{Gt0f_vc*Cln0=)R?1s)3YT#xxO5WCk zoNxM4J4?>Pe-dx${zy+`qb7~CmIoojtzAsX-^a{kr#uUqvrtXJY2?|n07ZDZu+@Jd z8ztt=#_Xx1tZ&Mrq%&vH>nSO8eA7EB&Q1*tYmYL1!2(R(vPDQ!w-mf7*;K!dFshyXohew<%M`u($5*S;d9(S!Edwv}WEMB$KuiNqZTQ zkP2sV==6M2+j|_C}E)^7IRh<;sTI=78mDeTaQEjY#e1zr)3!-Soqv|tk*Y*p!E zr%c&{FJxGoWMj77FNRt>%Nj}Lm!pKxf6!xZ2X*;%0^0mRpV5D6!TedT&Ro6UMFo18 zQ{8c|sZV$d>s;!^mZ?Xu`*+wKF0g$4SBl;eB0`w>^u0bS54K(9awW!=W5j?eT*QXQ)Jv%M`2*mWA4&>84j z{xhoXs0{UKb`(`K=|_R>J(Q9DB_^DWWJj)CA9X-6h0R> zlq7pD!v~LTL!}|Q)PWZZY0(8r>^=Srxs7YUX2xSi;nxK7pvHiGdYAwFkuAlB_(ii^ zXEV5Aofowz&m36^3$YsYLahBhOLp~w`)IjsHM)9eDO$X(nHpQBiKyk7i2F#HeXgm_ zdj9ocee!cqxl0zc_l5vt{I`y3)JUL8)$^%EX?Ez#`$kIYTR*j4$AXDld7O5qKN6xT zK!19;2K}rXLxFw+4A=fNlk1tye2dXRy*HJR;^1~F!{Y!Hbp1(L+zdc8&0|hKNMlsI zk288ZA44PhH}$<#0-gBT$HWa)G3)$hu*e;;izWEaGA_BSQQ}fI%g%ybsTsh=7i5!- zi8dsqH-RL%O3`bsn$jlvKd4^CFf=K*1*skBKz)DiV@>{i@T%Miv_;FGir>73TG8@= zaI&1y;0$>b-hUXG@3@XkYrN5o>|m6*_!G0|Qvu_8CxhvGScLM-5>R>7J*wc?HD;;N zS?0-`r%aoPIMQzxM^3j_BBO&M_+ZjQYIi|7HTv=ywdckdZ9YJlzA=nKG_=sr#Td$d zXcfBod?8ZYWsDRg_%kx!rRezG98{sWo0@zxO0}98Ba0a&U~p+awKx76<<=mIe16I! z$$xjK@YZ4a+Mjilzf3BsI&=)hOL`z|XNh!Hzhj8!Og685HS4L=hRU9v<jZ3nm)W7-#-y*8WxAd8ohrA;SsnWLT~qd*@&_^CVlG&Ev+f zKfBYZ!=~>jR%09T-FOFiTLhu&14-1b**(a0PX!uZ^@RHJ>Mc4_ri`3EcF;G7A&~wh zO2S|7CvKM_*k$kiShvDZR%-DdDt?zSQ$0kOpI`oC@-7TBbk`*&bJ=pH&%u(^_++Ar zcRV;Codb2d?HG?ULg<0sYf9nlRjLdZA$1)JEww&~K8Ef>!5seD%BrC6CHia{*Obkw z&|~j+N28Te+o*|xJbG!e7V17@f&Q9GAuZ&^*4&)2ivJULgc{*(UcSlx!$u_F;ls&&|u$a0s_?}w!>@t-%TM-2=bY?%geqrKM zuQM;bwxRiv9TXT&qUzS2L8q7cpx@&u$i~YZ#SR{#CMHS=8|^@yKYR;z*CbOz%OR0B zjVHYaC5h!_6ZZM@d}i*T5A!QigLN=F&S-a)FzNDJkW$4iROXh0`YO1TjM_S?bkhdp zeQtvJ)F{e~J~@o(-;zwl)`K>G7hLsP!3z%XKvTKUBXNtT#1Z9g)YSJOJ|xAqn6EmwJVA0^Bl;(S9Xl}9P}_l``c z$`GYDIRkNAd#M!-GtusBIrJiA2U0xVg@O-fA~^+?D)V?muMg3q%4;)GM6fQ>AGbzk z0st-7%ZC?p*HD5>uToZ{=}i0?AtoxUgw{w|ffi)yqg}UikY$u1T9JDe8Tx!gX<5cJ zGy0oadUzbI4%&(g(!Kb!=nW|OraXK2vN-$5X)b$5hfwV)RwxynK&$luYW8?Ybr_D4 ziYwAcR74Hc^roYbu|Je`@@*>X*m6`5>V^{GUrH&YjaplB!EWIE8XRSM2+tJHXLegO zGhZ`aGrhaq**SOBSajKrRh*;B;@Mj4Uqv_er{q=YUnQnuOJ}0SxB;53|Bq3AV#~Id zn<0f3fUd7vj($tuLCrVkqrG(+bZ5~z7KC4C0=-tyZJ{pgAHykzkz<&r z_%rQiUS3Gl?gUBTQn~{1cm1AWw;wExe;#5P@KgF>fj+qB=#yF z8AOTOo6Nh7hIY?IN{J;oj@KD2&K1V-do-KoyP5Tm2w>;VmO%$S zqN&H8Kd7vT6>Lz-TsE3JpQXPO=vH{1(iL|`S=&D{eg5iffYvH@l}r$OWO)y>^vyy> z$7?NGD5Q<1m3>ju(i|ify8``pM-P<~XDD6J0LL#WqL{z?82_`kn1S7M*iU!Eh}P9b zWG(j(HVA2@4)(00Y`;7NV3rSC>DJ0TIbFgel}XdZJM)+w`;}Si%Y@nWLyTRtCxvDI zoJHS6ve2dt`Dk26ljYv}!4Ul`j6r7vJC|d^ZaNmh&OTLwXmho`^jrA>6Ms63sd!w?$Oq=4>%m#bsv`!i(&?cR z#SIZ&c>}Fe8evS&7Bgxc&zWO}Vyv0?XGYj)3A?_ol5%X1r*3X%sKSUOI(W$)oa(Dd zTfJD%e%$24KKhfycHfp@7gbj=%4K2vJ>erpIkJVxYMROVEXl_Odh@_Br7@1*Qy*kY z%|bV{^QqvNS*#J~60=3pl9~9IgvjEfsGws7dTr6bgoOyR4~sXkZ`ECpWy%rMzLW5C zz$Y0_lsuE?Dn{}G)6fg2BqT18ha6YtBd6&IbkpfH>TLhUaK!g9`;E$&zkSK1MB*K$ zZQV($uP_SN(4e*Lqrk-EHuN-i4vMbuN3Vl^!2ol4^yP9En)VGtwMU~+(DkjTy7m+m zvO}DSU%8bbyW>$0=LGun<}4}<;GeG?nxgtnU8GiO-(wE_yUpzSAjRI@9*EdA%OlXs<#QW4EZRgYu~5VJRB3 z3{dBRNc4G zurL^HV(n07ktKEcRs_*<6LiWZ-!`bQ!Dy&t;Vz!*0KU**&%>HC%u^HK_J3cW&ps3zlY?IFoubSbStPLL9kuNZ7r7Tm zp>}0IbSgX>ReYReUVQdoo-sMh>-Kld{N+cnGQ3Wr@Th52eJB>ibjqV(eGz(BqXY^z?niPBb!b%Y64D#&M@lIvjOm$S z#`4w?_Em{9dufgsI|GZbBHHT6nInnphF7BJwJ%XbN(EA>DMH?>K2ga#ji{T2T@(yA zL|qmJNOkpgB&p<#K2JLFb5FU*=H@8lW{#N)R)dW4^Kl$5ybhc?WsApjtI*|lTam{6 z-6;OwEY{6Rf_*nUm-XG}ixOYRp=7%tw4OOYE1f(=)o*L1axX^GxXuFEg|wp8!n&+j zyfRyC1lZs?qV^JzyHQNbF!jRy7jt#)YQ}p`0`eHQW20X5FddPH7`ZoFiQLO3%v%vh z(mkb+{XGjNqo9%z&wfW)6wE`HZ@QxIzeAa-!<(V|CH_ACzg$K-lAlGG8pQp=e ztLSbAcV@E53Ax{)`S5vc{Jaflir<6i^>RZ~ zU0<2IuirDq+5EiH$(<#eVZ+nLY7ptc;y6TJ08wBI;xKWkUi zRPyJ4W|pk=htG`eds()8y)d$9orgBhRX{g-_-6(4v(eqB`6&3!6;zs|hs1s!h9^?9 zQRiw&&YkPzlQm2*1x%=*Y_?? zs9QxYOD$#7*2uG))w`LS7t7JTfAJ_mC>6P0ZbHwEE~1h2Mzm>XIAa~%&v@nRWIOwo zB0UQWBt&gN@cjqoQCl@#GpCr&%1dT1ow$!?zBM*h>r{mvfHBv`Kst!=iduvf)kpjJT<}h6S{3tc6 zMuvU;{w)*JX~wS8^hGk0YUsdxS@c^{g=i|`=1Q^6;zP`V%_^+pn__fg)E)V(%|e%d{Y36|SI|7SH^_N@ z3gYPRLr2U6P-v@|{U?4d0(P|_#aT`0@lktZeY1|*-S5iABup^f#ZJsz?|9-f?<|R% z0^~(uBlW0Q8|{C$6xAwQvD=r=VNc!w?4!2<=(y@8SsI>Mmy-nKeq1*3a zY)BckJ0>0-Wv(;Jo;_mf47AuozvrXgBvYgvsEeL`X++Z^H>tV8pQuF=^3PlV|1c>9coMhOl4S4nkX%5hy(9fcEw)5o7is z_TmqrHzy3DICp2XcuyyF%Hk{|WYo*N*`dhZ3_`4YwEq(`(qVy*?N@n6_Fw*Jnu5E7wutQm;^F6JZpe$`wbHIaV64CDMt1`4N+Dxi&LCU zQL5EN$||yoI>^rk&XGW9kLZ51E5r=F@(@7zGD5Uk!d1%EE{=LIai8*9^oyz3=Pt ze9vcUim^siP(BJvX`vRdN2sk1<_N6YhGtJvD0j&+RAo?rt=vvsV3 zWQ*v^3vJZDJ5Ql=uOnrCI-hy;@H|uD>CV_|twP5yFGh-M&rsGrz0B9e>g*ii#fDWk zAd8T2ba<0Axr7C97(g9ns6!wkk<>t$;l9e(fY~=-*yh6fv64o-kR&TJQmc@qyZ`jgK(foGZ(T5&V>T*RjrBOdGAD^N(|h6%bEPnu&EJ`e za_V=JZ|{?7vk_@Dntz^3oFj{#tVpHn{vqmnOCHh{xv} zLEg1L7&(n^{BGj_bGEq&z5eKqntNYU`|6uf!!;+wJ+K}5w(MmjjpZ3?HKf)hpG3kM zE>v<;1#wn?Lq&WKp(c`3sp7Z$(Vc`4B*V{TTk!Y2Z)-E@UzJVF{z7XsBdd+7;0mF6 z6${z-G5l^}+dC$-;UW5SXa@3#MyO19E-MtK$Ubg{tW`ucS`(dt0!+P8Wei|Ca14Fv zvpZ8Jz8-x;0M)t-AQul|^s{d!61=z>6=r&&yp^^nAo(%1^z>r(lcWwC{*>RX(aa?y zt79qi*&#@u|NR7gFcajE$js7Pjt)=oyLV5IBh}hZ*!sgtV1pF#QbjLhcE5>=Rxx4r zPF7O(3m;L918b;-fBliXgCX+aCLrPcI_!Oc<*fUocy>#jBYLuQ3z|Qfghcj2_V#RN z*23M5joAHw5j0=K)XdMJs(NyOKzsT0!PI zJ0ey$jd@l(i%qaz#md+(V+ZH?vl2ZL?C_v8tEDBtIt6GTyQ)R#2^D}W&b^@f$2FK8 z?t7SBgTKhUULmBa9F9(hTCh8B8nP!eE!c>`QcAI=o>I6n8(rtmM2qI0L@U2mppnoh z^ejve#kkBwN+(vZi;8X8^UL(vknNr*e4>sTsBuJhe|(?yGNPk z&w{MKsW!_0WR8-h6OmBGar9o)hq`uk20i^_7IJoXK>jz+p+i>vsJQhw`knrSvOBK8 zP754i-q>{$uIyu`MM0F!Qkl>GGC745hsDw3*THDT$T66wWkOk7Wua$&8^{1pfT-aO zPOElmk%tT%_>rEEkmX52<8MhT4iO@_H~&YoS_{{S^C zNI+5>#wj_&MbxQhvecE_G3wG<1=jW1f6S=XA!duM5?lRJi#>eRmJR)FO81YYF)7Lq znImR*P!s?E67SfEB*FrbvHTMBs9O;|&#s~a`fKSCD|HfBJ%ZW~U8mkv@}KSN-!S9+ zj$^Nv81vPp8YvfKq9B(jRMG9h>TxJ`^%_HVR`FVNJ2e^A{c1r1^;XQkXLFfMX*nuR zq@3w-TfvNNoXyOBpw4U>-Ou=%yr4BETtT*hAacDHiB=b9pxaM!P*hhGnxUhNc3<3s zGQ~R4s!Sv1lB7A)>*35;wI4!MbT&#;$V0am_kexg>&T169#qJ98r2l=iC(nMllk{u z0DZNXpo)4oQ{n#v=y7dJW)(+|N#B)^-U{tN4uiAMg4j~tB};M6ITvYgwYC+>y?=pH z^4}u|^>}pKtOmKBuSDOw*D*}NXU1G&gy}f|Dd7vb)JH!582t#NbmY#XweF_&H&=h= zUB92myOO9vX}k+#f)-d(B@!*De`G6bY^%#=pXp-+h%72Oo=F|)cuPMQdVzZ-e&hXY z7_RgwrT<1MLZtl@ub*5)J?$+Z->zRIe8wpn-spxrKEI%%ugg%IUKLZG9ihl#StI4S ztA`G;ZDc&0Sc(zsq2@gnN7rOjkYI-n+FERe)PggZz(Onff>;z1jK~I(gL(8X%?I?g zIBUkU1Ttb7VZrH~hZ9=wi6 z=>oARoX4WCYs*o&?O}AMnFFr}B*3gQ{$z73ptM^DsRvn8lzr}J`h0Z-)3I5SEqhkT z@2;w#s~y_Z+t>f0c`x{L*~6t&iq&qU?AuKJ2x@~TJe`;`;W3P;!71j%^nW~2Id2j- z*@Bs|M0TgV9edqclI3^?Go)rVGZi|M5%{nO2?ivguO?1NVV4|heqDflrs>4yzC4Du zY!^YXHdm>>4JXn3m>tNZz!UvX|6U-p(no;e7w-A5|EvDJ&633ntS$e)`uBqWa{+S( zW|j-`KmUJT|ECME``-Zo{0p$Qv|8fqVEz9J0C;?LIq}(;#FO@9c*~<=L1xP&_$+<` z9$an$mkIGe`u8-liO)Q`EPI^<+z%tpQeSYsw=6un))AhlWZ{`IF07nn;FC5Hc(EZ0 z&v{%==4O8+kG?6=xA^PY9uGk>z8XO3tskKQpL_iDy%9Ar^CfI;Sp-~jBgse3TslB} zKFuoF(jFtHNkEE=-TZ~3l;KMo%DZqerBFDBx_(s-=Da^i9?$ztk_??`wLCXEW#}O> zu+_tRZ2I68IX-W>-jjNht4RfXuZ62fDPFPTHc^#RrSqm&(O>yIljU#Mk~(!ysJ&wh zW{~;R2PZWu!S@sN^*;+d-`J2hMPju4pbYJj*g+Z=+#%a(!h2iN3}hc-=n`}hei!e9 z^H$eG;~A5n#hmYK>rNo{Lk&c&36rSb*#w?cB3j-dK;ln0ygn8L8Q~H*`SCQ=XAI!w zYy-UhNe-SIUrjnsW|L=prmIl06uB+ujxBWW0-;|VsCr}_Y*}CqUH9yU%U5aRy01k* zhxQ`5t3*k;lt1~&n?*L~U**Y+it^oWig4a`V;HdIE2xrFhSTRifWcQ%aN0(dy#3Fd zG<(aCy;n?08NR?3xantCsJH<3Xb8YJH|>CqJm1~4PaO1>(D+et9`0My3|_w0Ad9yo zVbfnmL}rx`5$y}X9nmMi-J^SP_nFgp?W8%-9*n}(8** z*WEyE+JzSu5(o6X*2A`M{NHTuELgfNo|Ct548On73|7e;g2&vC@ut-7;~m40Nb5C{ z)aX)NwOft6>u@AeNQ#`Z3nZ7vFLAz|p3iMFevBvD<$2y;mw>o~<@k`iJc(Ml47~5n zf~NKka8sx?#Bm>h?baLM?B2`ZVSOcV5(ePku1Rn)Bb=ySTu-JX!pVf0s$Ha}6He~B z3G`Ky@jcvxml_Y_@^jgogoQJJd~75Xmp8|0N7r$$`+4I71zs@QPzgUsTM6%cHUSaR z!uaH$1K@{Y672Z6fX_^B!MzD7`};rqZFmG6 z5vv31Rb-*#>P0q2~m3U4EKDcDoE25XoW z0!KebAbhg{J6~$%efum5j4ZuD_Rems*A)*o2ux$IA`#%C=7-bn{Nyrz`oL;*9k!AF z215P%@!NBI!D~x3kltv8*J%8~vDUqu+-O~rP8q_Q%D#5N|HOfKoCb-jS;6^srW?21 zNQSp)eXw?|0BBrd2u;*wp!kFjM1roV%CsRFIiAT=Ig`z~W9J9iqQl&WcP`o~faURQAz~R0`oF3E&T+9-%=VB9L(|?R~W{+d3 zeH`+IZ2L}1FYG} zXW`vX!xs-n!b5x=L1<_{c)xxrJisX=o+h*44ch?huhH^oc7DO>X9y3Z{3BhJVmxs;WB{!#{~@6#Dl8ubl_4;vBcqYDx6 zGXo#@EF#Y9u3}XKb5e4^hO`|R02A6tL?C`YcKxnQR9+A9?w$I_yJTud45G7vz~?@! zbED4gNNP4-H|r91rk%m1p)(wlH`~bN_zggA^?7iod=J)kEX4JXjmXXsIq2_V0#$F! z#S^aLq*C`OZ})f$_))wTc8`hhbVkGQ?b!7|arPh!xb6YaG-iPN)gK(Q z$tKP+i7XJNEDHyiK-{pc6@1#;&r2@Y4104#Z1r+?z^S%zGO3wJzCSJ_$FD2_tClQ> zt=6iL&d7l=@^8TJ!1b`E{4)G;vl}i9La_P47TA|K3QVdx07>}D3l)0L%^MHHi;XT2 z+mJpoGSxwB8ncKc-$lEs>;fTeLgaV889C#(g&g76K}J@d1RW_RFvT?8V>iS`xo6N69QsJ@JaYN2X$g zX-$O*GRxVH2x#vlr9Q%>F=!5Xwy1>^E2Z0Qm&gVGiG2rqw!MYl-{-=bGc3$_{|c&p ztb~%m$*^*b8lMYY54fC%oWNHB;QCiv;-wr(>JR_G=l=qtCUTjiv~-X#sly~Q*qJh^o zXt-keCS!7L1K$}YAPMqrNRpe{Sv(JQd2&QxHK%Z+8TtNPf*j(U1#?YQh>{hHH!39Y zG(V}51g&bjn#+aQ|IQGY{I3wKNfHNh)>m>&8A&`{>jy`K#lZ6iW}v`w4y=3|2!5_^ z0x$YsbHWx2lB&iRJo*TqEz+2T*A?rL`kEQU#YY==35t^Lgl^7}Pzjj1N}b0BiIFWO z!DPv}Amz5?2^rllOOj(a6nCcyef}+hKLUP(VM#$+ZdjlCG$cXa?&QI{S9E~kh9~4f zcoOVP(E+j3g&ZNf0y61a1pR~wc(y=--ZbJ!`Hx>BjxdS0@!bXRtFnVESj~eM92b&P z8xkPb#gOPHZvq9)9DL@+Ih-JF4cq_bz{i=@J+y9s#9sT)4&K z0&%%91e)Cxpk~lKT4Qe%WwJ1x^zQjbUM(4bQZ~v&v_TC2G2KYQ|GkD8S#iMUodWGR zSp?5I3c!S;6XeKD3+mAOEbt^-o93`r@w&eMV0G{YlKAch_u{w;P>B-(@9rzWzWh>b zBsCwYQ*Ud>tV~ zD~M%-qNlPjea#AZ*=7uDwLQiQhqy#`t0B2QCP>=aqw(QIBYb}{flWWfsc%J9ERW83`9Noo?JGMOdN|PX09{ndn);4>+Ld9C6mZu6J+3=EC={I{4zY^d<#|| z;B#r;c)};`_rdiqn744g8KI*BvB}IoBvOA9xwU@?n~rZLzN(vXI|>DE?*gEcGYi++ zoB$qtpZd?LL16so9(L%HBMYT{$d@%LWSy=6Y4`5I0uGfR+T;{?CGG(?Ei;3qnO4x) z^brucxDv(}?JxGigz{LvFok#UXi1$O+R|fCzbjY=vEL%fv(Y4YWX|s56lF zAq)O|-o+W8VM#_`b`h=m*Mv6uO$0ywB_<0_6SIsYJS=_~a-S}TLwhx;%K=K%0;`{J zv+4xQvhjm&vInt(LOZG7t3~?_2+)ngtLR3ZH+(1L8S;C~1F|klq2(%5s>@f5IvA@* ziP=Aa4^N(j>YA0{LroRAEj37%mKF2M2@@8BM2JGJ9E4e)$vTjU-;EWUtUbG z0j&0T1wQfr%eMaK1a=lhaMGR4;YP(Rut-!2>Kv^F)fc*e(S>xNm@^Bq4w+!@vN@#D z&Y$ltIEO9sA>!!5zsCl#hYWOgM(5kc=fY#LBWe9 zr0}jGQ4gPucfA<1t4v#p4?o&M*oFEe=GH~t{Y!43w&)!1?vyh~=X~PK-fg4o!QA_!L?JEXwYR`J!&mVGlub2u} zun{7ix6Ht>%UKZaAWkY2LvcpkdvL4#IY(>z9nPCL7Te9P#4Eyzc+RpRoa$RYKn&+3 zNAQjZF5PcO5(mWbsTdC2_LOV)_dz@edoPC%W!VESzUN50@&WIX(;ia)d>PpxyN`Ee zfe+4n$|o@z3lb@{Ex7Z053VneBRS=jxLpMECW9OC{0+glq-PmfR4|A|Tu$&zkHq7u zW2Z1W)qvm2KjH{no=vjG196SH4B@glyid2E;Ki{u04_R?e^A0Cc&`{w^SnA36KTcI zGSjfpv$=S$-C}OYicfY|Z#8jde8~mvwa2+l%hQ16kvuH@i~rua&p?T70Qg$-f&10% zBJj_>4Bj4Upwu=BSk(}A_U|+iopecgTcP88Sus{C%C(_AEkynR4^0G5*NN;~S{wTxadp17d!%-TzH$EP} zGvuefe6HYBX&a14Bjy^U;~$;(a8P3lt}YQF?H?3?@XH!}%UA?^Cbi*7js~`Lp2u4i z{s-Sl$R_4f3*n>S<$M?WQT%l8HM@O{-+}ndGkBzCCm4j<@E@O{aVb-PPJKUzdd_Fp zms|^@!2?q`q&f`6w%x(aza{9)r>bF!nI?74)sps~(E(-ijo~h9J`Z276SfG1k`=a{ z;6zLtXx_Juv}7zN;{gKH(xz#m7xWvvUT2D@pYmCSit<#Qo-+A9vVeG)e1MURGYQ$nhTDvBjRA|r|OQxMf;n1oeSU*^S?PG7l8-o!<%hnCns7`ZdB^@Q% zPTBC`_M61VF^SyxPX`X>-2<6&m4sWKhr6wWsl~|>#O!x15qf1u`P)>14&_GDqZNf; z!zKt`){)^gw|Pk;vG9^W9hq-Do4n<1;R!l7fk3m>WY;@msJib87Lhs!W#@jd4L{~X zYNTF5QQZ~9K`)N;dAlEJS*bw0_MhP~6V`A#GZNPDxV(tIEYh{7j=T~7gQM5_!>C7B zU};l2G&`jR-wm83@X2pdwfG}BnK;Ce<9>h_X1G(WwF7Y2wu$UFNTf$Un$v58&7s^C zU&{6NeCpF@by)i99Z5Q;Mt9e>5RoclD9HD5Mcl}O_hi~ZO7#hjg`+GfUG0nSiytF_ zZuz8JQ=N#0bwbs*veZlZ0*rr=O7`1n(UxE3>8r=}Vcx@0cqDQf7OeFFvx6`;-c(G+ z`g4d%M;UgDn}yvIW5O1`tXmL8Pt*}1Kas?5|Ko4^6Wey7Tb@K__8^ql^lX* z{sZv!r3QF!+@I{(tU$*mA0so(65;2b_t44E4sNJ#Ce>nm&a+7kuS_QoX3fin;;;YU zm6vtM&DeAhoRP_Mx_*WP&X?v)@j1l8R|?@a`&H2M>ps$5f0Y=vZYBM~sgSx^1AlGj zS8!d{fWJA{NuOT@37yx3^HY`Jt55>x%DTWuC)e^GX5Jw&g6+gxRtf4?oP>|~_x=o{ zc(Ba*ASo#0`vwH{@wd)6D7QErT4#mue1B}^->)outs(^DHh7W8)>FjzKpANZ34twq z5A^<@<51}QQL^EdK5b*CP7j{1fm_?$shj0Ol*80Nxfd$dO3 zrBD_x7!jqLMn90p2?F@uRS{|^{123yG=RE+JW>v4)2%x^$)l}_Fr-0^%ImFwu^rAt zuG^VbRk=@M4|u?-i_h z#L11|zeUU8_A(t9uloq|6-GGkpc3{xnGOm~`7WoBcrq_h7Nox`giG|l!^qDEz!7<2 zy8nPA?fm>MZcCJ>L^6NFzkx#V0=Jh$Cif9}-y@{$?h@!!C`Emh=!SkS2Z^_XCheP1 zNn}qq!K$N5)PTocC^FefzSybI>e>x>k3lE=oA(MfUlQj9ZmuV)df`MfL><&uAA_8T zEx=IUfQY|J#7{TP1Mh05K&YV(kp3q{Y;ziM%wq?v7cU9B7cYQ9UB~SdeZ9$uTp2cW zujB;!EQDX5eYb0Fio}H@zI{OL%+j0t5rz>OEkA?Vw zNii^+eFn@A5Fw?VZ^7M*0#GdSHvU^426~Rwf|NmBI8x;WZsaO~`p2f=!Rb?2FQptT zYB&z2e%t_a7x8_=`N6!8uJ&ApE02$z%fR>DG1%H2jrV;&iZ|SA$C(Lda7Ef5P6nyM z$-+y>@@`SG+AIYh4@&3tPsx!AUM?@s;V@YEa4t`<>j2qyYb$r)bq0{W`j~rA{t=$C ztmHiQmII5T)`3rYU%)PN4s>)^g1wUp;DN~z5ZxaKs=wC&1e9Rhs3U9*U4#`@w(!1L z$iSxo55S*X7Pwcba12y4@I2ofpyGE7_j27hM~&P;itjT#wKSCIg(#46BmO=Sa>%zP zbKVzyLu{~Lo7jc|ob{=dQ)F?35H%UH?q>>5s<{ng?_4m`2;-6M8bGVEoENy}1t<2h zD_l2M9PG9H4F-Pu1F@CO;CACputrD~7AYNrp^wtQ`MM;IVe%w+f)!!EK@)DVRKTJG zp>Xc~**p!`1K>*jY3#aSF*v&`gxmFR7D?LBfX72Pyf~+HT-==T6{t@Ceq~PGOf_i$TN5T>R87iF1ol!ZEA%0grie z;W56OG*WBgo7$v4Uk2*85m{Lp@m>s|d0iCeTLIt{A7KGdX#e>08HCSzI z36nqVCx#cg>>Nuz0IeH{O-aA*iiO@tcmY}fwpgeq}vPNbNmZ%IQJR%+h#-c z%70|ET9Ep@<~J6~aHG%9wWM74{osYB&!q>X2Vh@k4XFB3Ldw2)!m<6m9MaBbtq+6& z%gVzW&mZx zGqaGg<4zjEWv^XG$ZB<%`1cx|&8Ke{rIq7WP93jx`AVRjbc?sbO_vzR9>w-R3J(m; z0Sap?_;ZEJxS-(D50`p$!dKer>fQ`tDR^cu%)z!5B^ z72wrwCD=Tr47Poycx$F(f%ic{xN?aXZX8!^gjHK;%_4KKpA5AD;fh*w0=}lN)xfHL7)4-1tjfmuUE(ui-BGXzjM4@R8*=EuLX3AJn5IwCz*bwyqE3&xcslmJAKWI z6Rq?P+bQee#vpMro?C;Z3?JZbqb{(2eG`7!3 zOa(bX{$y3mIs)UYh~oi65_z`;8-MA*-h~hf^5@_r`w~<xF^V}*6_Z8zPkCxL|3Rk-0m{-0Qo7~}lwQ*nc;~eY37tEWKJnR>{$S}yA9qrw zd(DrNJ3H;5k)I;<;JzER@010F6)dS7{(L0;Sut5yD@vb!=u1luIME$`?Sv!5Yk&8e zfmWZTDGhsNYWwgH7^+$U56ivhU1H0KZw)`ym-3B7-`-3F?G%YrMFT$KuL^A*$iZ%b z68Ln>Z7^+73qA-+!s5a*(Bpj)%r1Dvds4js$1c$!h52Z>t!OJ>rILz{$SZ1LSVI>DZKDP6>hWm0A>gc;{{Q7u!~$a zu?n#!FIEm<;h$0*`8Ffo-$U{+XL|#9b!|BeR95B3b)Dg{E@`k4z6XXbn)prdW?;iV zr?S5@i@cb*7{7^FN*a#ZVdJv)oOQL4yGc_L7A+pedJT8Q{yU8(=+u?b#b}7yKBpg)*gU zfYy_xP~Y$k*w?5APKM3FhC9~c)rJGyd}nnm!1Dy7;0-=I6-lB`W#AZL5$tm2AHE?V zO*q8@MAc*t&Ri=%?5j$^UYG-fnVH~1tSWpr@fVmbREC2{7v9*s6`qk<2=^bG1%pp^ z1Lc7$cw8|HpH;8J>y|=Nru&;GwRk@%pSX8C8|2jF| zLEpi%u~%S&+83}v_c|x!#WFY)o&a`GIZ(LpH>l4M;wCJ(2KJ0^1tEL*uFY&0ykxFE zsV~UJ(_b)tGjtuBJDZc@z53*>%}#!9bT%h*-EHor)H+V?yEV{Z%UQTgWEU8qHbP<7 zIk0!{65cf@8>p#8K|A+*q@g$xo;VodH`L*S^t<@|;S=QifCs!KAVGgdSD^X7qogw@2drF}3VY6t;ff-4=(|20 zuJ*47=gdubw@;YCXJ5aPw!%cPu11r%-<2mhQ<+fv-&=Bt@rApCAHcIFNhC$!0~mJl zf!ov?I1z3&Fcgu>@fk|PAO7%rRiACipO7LbWUNJ`YE8hZKhdy#^GCeQ<1*A8 zy~R7FN0Wj1Vo>dqIad4oiRZmX5_FiRk>H*GVf($Ic)#{NTzkTRJc)jR6J+mjgnJ9| zpON#pv_}~$Cw2lu_hoSO$S*K)$_mKrGJ@T0d>8EhqUbyva{9kGE-l(BB}ppLKvpTw zJ?AkZlw?K8-l0(Vicl$OXo-r-C>25}sqQ(C$Vg;_jEsdAv*Cl4C~9 zdX%i6rly$7XP-k{0#1K>fp26evL9UqrgqP~{XD6}KkKw$jUQ^24Oy-Nf*7C}R1GJ-4k*sd3t{5^T4CAJ^!!u7e zSL2rG@K`II_2YUv7D{iaxgm z^Nax#Sh%&FSNsa#RjV85H3aha9V*ZoFrLqB)?_hQms`0rU5s7={au&iz1B{+`)vXQ zck_d=swkn$U@MGsdI+K8u7Z2XAy_LVkK`WMh}B+aAv@-nXz@m$J||o!>uovWn@!H7 zn)(xd)maJ;_bBt->@!p~J(&)TejxlPtEJh0m(i3-i^$iNaT{==QFF zXEMagbLQauZ=tfs->S+d2e(5bZ7p0;u7VD8Qep11O_V;VLfrmU5YF9SPD{;~gJzG1 z6z*3-dcGz?jHKPkJhPgQtZJhV55xJgi9eZsIYX((9Jw%NwumECpy`|u!_ZGOtBAz> z;9^>6tt5;;wGkut&BYx*ykKuR;3$vrILtl-51bhd+6z8Hp3iAqZ0wIwiT{K?!R?{e zqfKypz6*4HpNh?jSrp>u&L1-zdG$PCr8P04*9m71JY>$h0uPYlr^p+J^FjY|)aMc_)vRO*VqJ#}fpqfU?>NwSW27Lq3XrO#b^fo7y6 zZSYNmi1))`XTd=VxTZ>8(QY&af5NyyGbo^%AhjZkPQH-x^^T6DM=c#WqWwrX`>__( zhJF-}ru8C?QGxXCMU1F6up^ya;)74Wx>l~YtdafhUMq@Qn#3VLoy21e)5N)hvmy11 z6$c$V3tqP?#L4A@$TRwo?E3Umm1SYZ@ceBG?Un7PCEo3DZul#4(V!~o{Ou{|$;@%V zQ(F#H&ZO_R-?Q?HY992W2frhd&cqPafqBYx8zt=lxowg(%`$fnA$!}NRRIJlN= zMa3H($#nQoI=+Yqr20S@HPkjsq^-x){6p>EXbjBjUc<2blIV6=f&d zAiR43T$#I``~ycYXpR$4T$EnOhAT8z`4x4ose>=?YN@x?1jcFy)>Cq$sr^3D15v6F>2~kZN9P0X+)(1W~o&Q*gyn9%1d)HFX2&@pL>`r(w zzR_izln43W<$rV|J(isBbi&R`J1UQl8ZFD~kq9Fet{^)XWi&6_3|jTdWH2HOat~R7 z`SKEaR5VKbb1gzBHTJ@{zvjb>-4U`eiL2nAhbE}K46N8b=Ywb)U;^h1J5~fungdPV z8Sqe}P!z4B(W6^?G|M>Pyd>v7y-5g$Ht$aK*sKf4%pS|0t^NobY7@aPYza(-D7wCM zB7FB$WjoiSG=5&bKm~K5#$H2KdZs5ij2}TxgL}h3wc?8HpU+a%v71n=t3>Agr;&A& zD>Pefg)0+2Q})sJEW;g8H1#Otb@0XFqv3GW@;05GKc36`9OG8AwY(`2c-ah1bTX+x zO^Tqw;Zl0F%l*~P75j2xme$`eO%Ip+- zl5`bMCl+z={EuM%@;XMnw2;F`efhH6M^ISuQyiJgENbd0ekRuN=%g&3cX<~rRXT|q zDo>#9j4SAn>W4owBjIOcG`FRt)a)u+AN3m6cu42|C z8^vk2*PJfs@m`_fP(Yi&k3Wh=9HvviT8T_-?2uw0_F74{hT1;w?BD zpui~mTe#}LOT77P1~%;sH z`-8o}GgCdd5`6A{N_x``ECl2(?64C?m4<*Wr~*l^mw!LQdn`I z7X+@?M-{aP-t*G7XAbiRVl ztfS&#yJE2W+?Q+~>7d6pHTZTym9KsqiZ?w^$ucJX%vw^OQ0n$}=BU?i~OZ z{<-pum)+w%(~$S#cq@WlACiUM zR=a7-#7WMF46QkQ%pq}quLO$ueMqPnHXgLnOkjQ8Z*WTZ26105!`j9~Fo`zA6|e&? ztGtH1Rc5$t;~RJsei+6YW(mKmY*_iYBNv5da)&AAta>RxNDMLJ`oM`i-6Df_zSU&U z2ZaL~TtF2a|gSHzHuy+zxr-t>6R zKKT3WW#y>4WLoKx3?1x~W$PlZgS-6;TJ$&s7W97!{l*^@iu`qCeFr=R9Ryh2&XNoJ zScnTJ`(gGdpsRDeDc{o(cl28VGtAep@0kK&Lf;4Qbc8WKo%jue0RM`+VkV_~TI0op zqjYt&9UY!v%z57)LYUVP!NbmqznQsLE_!y3cJ=Q;@19i%e~n*JwbwR~BAaQevooTz z8?Sq*!Z8lE*dkt~)0M^0PK+15=1X&S?P#$!Yy`cHG~x%RE`s)5f4p*0o%NQzp-Yv0 zar8GeoFF~Zm#hmyc&E)!aorHN4*e#a>7P!Dy0K97aSUEO+$a=CzSHTxh49Yco@`dD zH}%z3$E_bK=!!}r91blJpUljn+zWQ>elLvT1MWfFzn=6Yz?gNnw&$~l0@?n(9L6Lj z(lmb;o|B;hdZ*WbZ}3gZeASBwCq>ZX-d?DwHXRl%_TsU>gWyo4HU3G`6EE~WF4`Sw zko6wh0WXYSg(gSh#GxZkK;Y{Q&_}N)(3((++><6O+A@#Qc3b0`Q(N%*phn@7uubSv z>d5VNGRSyh8Y_}}!uiWNc(baQ`b_;PyZiP(hCioyb8x+%=@0|nF#9|7?4e5fifuUg_#W}q zN)cyQ_QDyR-$Gb=7!Ce<7aks5&eI$+$nf_b@~`?R_T0Ku>Kocqt$!N-{?H(3g!Vu^ z55RkhG-#%4u*_W@JL=Y;yK@&DBl}2t%XQgk%T&JBdJWu8=y2z`&b)C%E@!6pkY+}g z!Dr|V{2&a+UoMXD#A={acfWy@wl;V-CFRjk*90~PdH!2IlIz1 zE7VAII-iFBmNvp}L2J+CBgu9tkS#$Ul*qBriz{7KHsPgnLHS}` z=@<=Hz5r>(?Q!<|<4`sIv^eM2Ythdynw~j#Vvi_KPEv_vlZT_pY}|VKTVao@Ox@7n zaeoY*kw9yo6o^}^d|`f{cAtBieVc3_h~1Vd6m#;4;xyS^pUDR z%!XTfhBQ;%O*r1{3J3GHf#s9WG(6z2=)AN8J@LLE6s_DMx~*}iU(2=N=Sp+EpAWK0f%L5MsBmS2HJ!S5UW_)pPsej=MSH2HDv0aG(Z(@U zoD?sPJo-TJ^;}G`UyS*B#v6*+lPh))b>_`xMjY>xM$g?QuybY%wGQ&22Pr+c=cJvq zKrx2BemtPAG>y7wB~ycjv`60zAg8)`dfu~+B7BaJQB-HnxspdsHRe27{}mk^JegqV z6nZgzD4lfCw8ETnmdeKtP^3k`vb~oaus)Ua1#3VHbT{w za5@#24IA4*<)quKpwq82T=D1xe?0cX5}l9mcKwy|Q6*=gthO^=^=g6e3m2hP+Pl*x zZWCtBp9wkop6G1;5&FeO!~bVH|7}Qsk|+ZZ*tf0iW+O_j@+Bu94>)ZTzjdv5R56)-$8jK<1n!Z?F~I~VLv07 zFNQ$Qi|rgfdt+%Oy{4tg@3U<>#vj`6@Rc-cJ~`wLa;pm#4(x;1 zXRc(O(cbJ|tEy->`W(AIKh8@tuFxpG0lZXu1p1XuL(%ytKC~Hx&*q*-&(nQz!qtJ8 z?5%>$z8m=E&TSmvaF_l5yTivOGJlx7P12~0h84#G@RD0QdGB365GFmwlwQaGi=Jy zphIU2v>Jn(#cI%bm@GDr-o!z#9T-YIpf7Hyw_BLwp>032iEOx_!_G8g( zLI!1a>B-7Z%JGw=*-M!9j6O*JgVowz{Kuq@E%$cDdtZz3WuI+)$ti=KLf)aSu01XP zX|K5Xs-67of45+T>su-I2y{*?8(Uquk#ON;^Gc?pMD zC*Y?4wo|(unymILQi#}=B@WwtPV~z;3ASe^;N{<$;vOk0cc{@`o)K?CIwv)xKIa`w z4^&sY&*`Z6z1v@Yb^Mmq(9KYLA(*C&8qxB;HoNY6`g3-V*CN#NoX3+nAPUC=V*B#^yy+asIUgnD9BZvMy}^XN5=c zh9l2-+(tPM+_#*c=NYheYdgr;<$?>Eq;J-qAY2j=je^I1{CK7#EG7o!1q1mi~Xnn687WOy}b)#cQd=ySqiyb*( zU>--V{v|%99t>0d$XdI-f;=rt@YFttE=OL&0pD8j-0Q;-cSwrE>bXO_CFQmMij=gY zHt$IJUmr+I`Z82>lq_&}_d@r6$(ZSN6t;$B!ymIJFv51Xq>s=VI#hI!Sc3nvg{_s# zUW|o;Zi{H!h-NUb{VJw-f2mac7!8W^U8Q-+U-~2I-wPsVl0yFvS#~$#iN4oJDSr;% z&Ktzx`;O3iJ0IpZS7}g8B>gjVfQqliytt1AcP>$YQ7>aSyZfM6{>ceqRoj3&IADYQ zda(Pr48D)(fd(OQ7+8{p_&WYzrnC8I$GKrNYbYh)bT{ zgMqnsghs#pa5}!2-nZ?illL+qQ7w_OH;?5mx+}OgY6mZl3F3btf&8wkoDb!j^XW7r z_Ssdy2DKYm%itx|>^(}}8|U$k-+7egsm1xnw}|!0UxarTUWl9Z$HDIF^U+6V2~Oxpaoe1LVQ8z5z4 zwP0H?8eAuz5j(BhMT+oFtWuQ@X6;Wlis1^m0Q0FE5H zLbuD_(xz$SIQQSH%F%c43fY5~vQM%V=fs|bAGI#6=l7C&v@7vK`DJQ&Y`}ii6{x;x3A?70Ko1QQi3`{u z4mo9qT4PicuXYz8C@W}GfRmKzJ`H_!6co@h8Zl}%o$FVQRu?Ywk*y=hZ@sU)ZCrna ziGMY%SE$MN?D@@RXVt*R=Paze1H3A8B?xMxfxitxvBH@*_}s+T11l;|IJ{uvh@bfK zoy3Z&nkfRZZ$OcMBKGvp;0HSWS#3l#&X3Y%_qGT$^L9l=qeLcNK8sNUT6um`M?90H zj;lwhDpLBoTldpep!sr zDd1r33_ANmL++z@nHydnV)alJc}`?k&U$9Yrw6K`yYfYtcG8y9Uo64ivutU>okZF> zsvYWdc)~(~s=SZNMe-&aZd-1LL+|Yck3Y-9XI{bTvBP-Hw27SGsw6*f zPO6b!zlS#etEhJ>iPpa8#h(HR$vy|iUUpQ}n%Lop^FMIQ8dJrMw!ScS<8rL{rp!$| z8yA%B;i-};$Y`;N6xLG%i=hiAp8iHXHO}B~7pXQGvJ>(htc3?Q_Pl2PMBe;DnH$Sf zK+jWwJ)SJXMZf1`^9yZGyRw(p8eZVY(ESWs{KQ=Pm-rVoo<`EGfj~n_uyGa}9c&^TZR=a;P$r_@>%z9_fCFXI&acb>3ch?Q#|B z+HAp;(UEvxHC{Y?;5=*DMYCnY99G-ck;Bva<4mKS7-!{!F>O`wTb55{6V`H-^q$kM zKO%MWPQr|o?f7|w4131x!>eP?3JTw`yn1CiZ#O*2kMe@qJnI*1pI(GETmE33b2Xai zU#+`Qc*k%X<$XS6H?COQrzjOE_bnhh>0_{q0)I>zMQYBI4ajtnB0xwce+Ew zj2`#!+B!4&f{|A8s)o@Zo~_}Vle;Q3POB+&Zk^?^CUP7xZa+TCwva#X(NUga?1okM z7qZEPE4=EFrQ%%uOYT!uMz{Ly#1n~n^8V>M@;%!QVz$~(3UAuX+3Fn>(Iz(;v&M0! z=4x1a>pwjFIYG*6jl&9AG-lU(LjOxWxbWFL4jB4?mw%tkbNb~_?#V9rRpKZ#qI2;yDo zYjE_6>0!A*y&_^)`f;-z&5`R4FCm>cgqK1%zYTL`SMmOIEydxQLiF0Bi`{KEL&IwsHx}-J zAqOJxyNQOvX4ZZ>Wp*Fkgakh8w2Up&wd4k|VT_%&u&Q-7Uc8=y4qgJA9UEEMYpDzl z9e+ZRT?xMH=qZj}dWQ4&of38FEyac&2bBRUaPOFpbYOAr(CWAxVfL7Jeb?Z?{v%PZWVh(|!4%9>+F_J;HaISrPB%;s(yNO`QjSVK4VDt9(RG*zYn34#Ks)YxIx9ndEk0^r>Ht`uec#z9lf?z3s)?g=|j#X z`aW%s*sNa$=N{Y8pHEl8vhg9j^N~>(_aKVV)rWPvqGgqg%on57>_=mh84TV;wJdhRea8< zj51<-^VWjPw03ZRR!%l2`T84=Xzl!KuG|u$_M#p2}E@ zc_kVcrg96kGi~we&RQ69(+m3~XyBe72PwDRH_@(-#QFU);IR7lr1`9f|3$>|rONi~ zQ0UH`mHM)^W-5>A>C8&sUQxJzEcATkFAGirJk*d5zYWarq(?n0((i}L^OA7MOC22O z{SeF>N1<%9H+H;jjC%9pAbn*=?$YxMyy~*I(!0imk2fiCT5mPRFLLf+GoF3sNAoHB zG~V?igc%`;D)nUz5mTK3eW{j?(HaWZJeO~*?I2B6D*3yeG9h@H}2(e8i9;ajIN zNIh0Hisk&84qWbdhYpyPgG~Bn{<~3uY}bfNhqnpO&iKImKqbhejLaF(@t>jWA|CP^IP`aaFb`tU3g%~ za={|d5nWoI;k;i4@~GtYa;Hlm!x{2(!)?;^{|*n+=-@!kM*)1--=Op z-$T5xrX1gUzr}n1HKKKTx@hs;+i?nfLRyqiPb?^5}~?jIZ&Y@|4AU&;N}Z(>vbuDIxS0yg-Uq1WBls5JE} zl+E0WH;!tFRRjF_acwp?rVGa}Hlh(R1@I`2pyxV?rT;buZ%6e)lS}H< zFtLc9ff}!fxhk$uZYE97cq&Wm#1X4W24u+xuJsh%U9ITuq&!gcdqsv0a<0uxq!k-_ z(xk8}WPas|?0#seP~*Osoeo%1$1x#fcUzNPD+Z&ct{F{h(uEge7xUjOBDCa>;yDtF zRC%!$j!gIf%68}3ESMrHihtRv<3K*C+Pu9h>l72-0B;Ud}bm6$C?CY#h zXzFT3p9&s;_vYIa89!WF2iDA!GlWOdUQu1Nk(#c13o56UQ~0MgaoFG9T>Q11oYG7= zp-5G@U%G=or%Z>#N(xbT?<43Rx*8^3P-oqc&bW0{Ap}Znj?Iq^QXPL&=x3$^Z?)n% zspc0=4V(?dS9h}Bm2*@%&IcoY42IO>zi6fDPFmA7S4ds(hWc!aCjCElG`&R9C>vOD zFS}jbzPcl)f4D#m)nDk-v|PHa-9Q)X)MWz)T^20AYH;wj^>k*SzF_YcKpWLRQB#kN z`0d^sn)~}GJzG1B?S2Zp`P3+WE&EGhdD2}xe;AKlCdc7s2{b)+DnGY)29A##X;4)! zRzFPQaNq81Fk2Nq_1H-7JKv#mUM@I&@DdJuQXvkgb);uu$wIe;$y66wDIWU}Lk)ks zvTx#E*{(gNv_7=7(rxQP&g*)R&^njyPrm?-aRjd~1=6q}VN(vC6uVtHYd#GE=+e(ez>`ZOib zq?CB#lUZP{S_+}J?!s8JKV*Jf6Y3kFazL;aj0{N@I&yFFdmIOPTWwH$`CT+kY!3(9 zyQ7g=GIZ2=MBOW*=L-Jj({XM0$PKg!O7$=i;!N^LdgPjsebA3xHC#WyOu zn+h!YTeKU^STqR&NDn&t}ePyTStcnBvaMpi(oxI0rqW+x4RPnp+JThihSa z_&4%ZTPs|0wWpCT_N;o+fKnEJCE-jIH!mz_z4z}RSTm07x~6kWq%U^5?Z(v|w-GlO zqq~Y8FRwSlm7!_E;D%YyZj_x+^s5qkFLI`oO;+rG(h?8#PT)8d4bn`ugy93!6mwGB zp;?0pPm|at5Bv&!o1=NR{$1SiX&ovrA4|%^`(pm=y*T?g3qKu8$xQd17%6!?Ry?mG z3+*sg8^4wRn`*)8MrY~A7elU)xFe^2doj6m3cXu(2VV|uq2=n`A!TY2PCBmPa$>I& zs@z=8?h-3TU{kRJ5%=rMF zgt26DIt#xVAI1X>JK4bV4K5vIEA3Txj#-q6a|-_Qou#X>^LihiwylV{oh8ZGs??+n8`BX;w{5<}EjYl~YXJ@~%O z`h4%OEm@iD;;A~1D!mLcIN_>ZWh3pO8B<;~(qaDHAI zj_K{%3s#rc^Y)(mNk~O1uIa(OJek91HDKzQ6xkP7J)WeI0$i+NqMK3R; z@q=Fo9_v<#a@7(T_5BD;w7CJt+$L7UC;F2`!$4e~G6~NQsuCV*EFh)4ZagflFRWKo zlJW~_hWW*VmmNBUYOZT}jYa_e6y|Vo&uC0Jki-YBZbvs$;Hs^!VBo_tYU=V2wlrR% zuC?>&h5KkaR_ldZDsBq*bzRW$gG`)QazWG>uR$(z&%v$8%Yxc8XIfuTKr>$6ft|@s zqS`$({P}jFRP*nF;WyUNgQ$h5QIacO?Yj+K8pn}Qaw%9$Hs*yf=GY-ApIlGGVN}!= z64JY1)G$qciK&wQ*@C|%W@5O~0J;;k1B3mEKJ>9b(l4S3Q8!@B*<(=b+=v}|H3&=B z48|hUS!7W<2;G0QqZeMT_-n&cT2vNJcO_=Eu>TB*Tj(Qu+Oi0Z4D)4r-vnG$s9RAU z<%08G{}Vs{98KFjj`8A1GZ-J_#X;6f@aw{x;@Vmnu6=h8%Kww}zI(@EUF#`mJ$M|( zNzbsZVkGyUKaVCmJ*6@7COEU%M9frl#&6|@ct5)pRv#=7%x1>|W>-T0ao^$lDjhmz zd;|*L>2u(@9q{{CFJ84*(#6^|0aY8*wU&0Y`N%O)$<2a4^UUx){)M|_C?1mO@cmuZ zJZ_{h=i8p6p|2$F{QX1Px59uHrVwd5nsV)pSClg#g|=>SV0gJn=+kxsGz#LNsc8T# zu+##R9gCrCn6@xQZ34b{w-^U4sE3&9Jm|mD7+X54M+57YaB z(`Z}H>E|x^KZ%lU@D3yA&_QhX&6!IsszT2E?=)}JEjqI77yT?XWzTW09N6qYD&417 z=I)B1`4Jts%cNqVTgXK5#k^QhuR@&QIT^CMso)fAb8L723#^WqhMLtw&?Oylf#JEz zfRXxGxkw+Yu@tuKxgh)bqmB+YJHf9D*TuW$jkK*p6HWbS&8y9=sBBmxl`6;c;#Cnm zkH6CQH#fxuCkI|MXBi1A)Ho=@oLc5)(cq=Eg6DN#=vvqp%cisme+u=$!PNrmCY9k< z_Y>f~N(YaruE6VIi!k|UEvz4>k2BX3+#b?i4BFTOa795vL%FL~7WKQN%m3;QQ`V+b4u9H}4^;GrMfTY==SdnpPKzSza#M29Ef?~Wwvfe( z(=e)MAo_b&2qEVWiZ-K9L(dMGSlqu1Cf&}1(3kq~@Q(}nABu(-mtDmfC&OU%kL_&P zZ>_Yh%^;=G0>`c0j2^G=h~4$VY4ts6zucM2G%1h2oqI$}25-W!9@g^gA=mMkQ6F03 z)m~w0WviH0BC*kf-O*k5H#!s=%h!K7D4W(fhpTi8SbME0i+*k#`N>JPVMG@^UZf<) z=~;;NCeZZ#0v~Dp%Ec+lia+9HH1cgPUsBdlzHP1{F8owQZ3W7TmY^DmCCX#<)G~0_ z(UmUi=$f$|eM--Re=lue;;aU+cR5JCC-vf+ zwg-4qQUtFn>yJqT)a7agS8#&=s!H=!_qkzEN5#Eai+Qu}4J_~W1;34*j8vV%8(utR zpS?CLJW9o}W>4^1SREY8JI*`Jo^XEiOX^qh~aR@k32M3Us2ie zhdm4a@XakE5vPXoi0Ir>OSPLe}fh z$akJfo;W{>dC~rhJbc%79^NdS`?)tM=yD3qJf|w}-y;1!w_7;C`2i+Ly5pq#1QiJy z@c89kc5QX!$fBb>xBfl%){JL6i`P_c7E9(A(`0=z7T{t<9*&c{;o)jK%q+PM>GwM0 ziQ}nIS=R+OPrD=xm+Itwg}d0eG>ZmU$MMovG5m9N6dwpm< zH`vU_b@83$<2>eKed|tK)0#&mW_x(zWIaW%q*^ZfCR3#5g>dTcK{P8SO#JWTSiEGa zA#Xmn84C3qHo^omXPlyBfVtN)po==d)1t-vujQJkk-D0VvrdT(+v@4qB@nkCTS)76 z`3dcwdEo=^<1jsV0M#a45_?yzBa7@nag<61d9HON&75-57@HxazV1(@0h{TC{x|X7 z{7CU~XGwcMIgDN!U4TO&tu%a!y7+&#Jx5JO+lp(1J(S2;bugl|?l$C}|A#G&4ZIpc1eNlJq_+b_yk4*}@mC z3t-VStFqHXTL`?<5n5NokiQrrm<(7*KR(6NiS2E&Z}$wi?~(127fpY5GyVjm(MYoD z?z~HHs?^))%0BC2NIk0~$IMq_pQ-z(%r%x01}vs+_pNx@4kcQCZ4fM%^iGzzk`@p8 zF8h4NQC!*QrTE~oK6XEl4{=LoOB(hu@O_FFKK5yc!<_HKpEqfhUm88}?hvUKJr@Z5 z?&`zD(2vk=NFoFzb*9eGYrxODT38%pPmlMyW7@U5Vq450@lX3FbWgKX>==AP^x9-0 z)O;wRpZ@VQVrsD1PhKkO3tc3iyK!R76?Jy{HW2RG)K<V zvgj4t8INkkP{+={An@f*9AFRzTB?$TPzJ)cL8`*B;Y;9O_Bl})6Gu8e2WZ%8YaU|e zL{G$8;qU!%xO=}d-gZBPD&|ST`k#@!vA&%59&h2Q38Oi<-&L5t;TQVZ+sIw}eM0-P z>$2ab+xXh>9*VE>0gC2|^W42#KUDAa7MGej%ICAA{E?#tT5R0JdB?gaTJpv!X6(^a zoSXihDjq(?$Insj?%*W%oN^z(cx+)wh|ey&x+{ztEfiBd^Z8Rq1wPF0Bi|fjE}wDu zE@lVxq#3Vu6+JVDDmt_^^TK$jXoPX9#K5OS2yloyKc@VGSq|f?_2hM7WxXs;p zmNX|Xt$%}aPyWFk7kzQDi84?6n90YU?B*s5V{To%S6r`uP1X`K1~nCP@wrJ0tgyTf zCr&72F4^j&2%W*oCCMDW<3DzJdQLF0G)B!qrRZ4y0DCRzh&d}~fMq6gTf-Y}q(hvX z+#sHC50T|hxrz_JyvC>NcH`96@6h|oA~sofnP+Xx=T{o{Y5B>Xv_WMXZh%aDd0mF} zZ+vmT|G>)rQxEd0KjA$0r!C)1I7kDk`l3bVT2xa^zCm0P9i_;K9k9XdJ=clIdvpvIvYm0R-Dr5n9%y5CAr5E7Qy4f)2(Oa=#)?JvG zyA3wBT%nntLj_aAZsO>^+F*9>DwI{Qct6OFa&6v;fwj%_s!tM_PiTYQE#~lZqyx-s z$%ao#i|JrlCsyw10I@S~2%~Mi@z6yl$?wEY)=p~>c31th_tB^dSNm%=o!8&^>XloBbZ>v-sG%5j=j^)4wm#xs~oB|V%brJ*K zn9}a(2s~A*2e${b7n-Yv(;Bgncc-5x*BJ(!lX;q^Dq1RM{<=fEL+oki*3o>X|5q3s zd%m(});`hW&0%=0~-+U1b`^MnI_#`;@X#qJ) zK4V&&GQ{$?)9K`cK&a`^SNt>8Qh2o?7WUgD%dV_Qq~MWSq%}KTCUl!mZ{NJ2sJ=dx zfu}6#<P31J!xhGJ%t;W4RNIqOv+odhp(a>-KcyzE8cmXnv6BHphl5 zOj5-uMYANIL4Pc*%y2-KT=Alzwq`4FYs;4t?;?6rNN7!&ms z-XCj$g$712`TB5_=tt7x9q`AzHrThJ3(h!t05WSA;Q09h>OWc_>0-Sr zkH7r_NvDG#X#Zs4xHujX$Bl$?d;||iDJtWvx>E4LYjnO{JMQ-4ADE^;r+crpd2jxF zajEMvN=R75&cj>js97WZzA;7W!6fmN5;+T>Ya_mhszS>@Vr23jPw?#p0uakmP#e6)V z2u%^s;r4bfariF>wAGl1mgk2co9J-l%sTK(+=Hbfzd~@(Y}_)@2Rsemf>C4&xgY2T zX@+u6ayCXcc`j_s-UY^IchJAzyM-gqN+IdtY+37?5bAm{1}s)3iKbijvV)c$_Bp#u z@^?5&E=qBvj3eWhaiBq0I&p_w$YP$sDQMg>Mw?J1Gkf}rF= znB3=tczM!Q@|d_#XtDsvmaFrB{*NhkR0!Fe?oD@n!lnD*S3zUQR#C%c9A-O>6fCT6 z!1%H-*b#6V5?w`@u=uptR5^wW10IX3s&pI~HOB+b>c;@2q# zoF01yyD0gH?=r4~m=hruZdBvhRZc?VsV6kA?*&*FkOi{ao$>RfE94m8i-q4k@mPh5 z7&f{u=(wNp;O~)ICt&vW`*mp2uR5-6WUm|<*bryW>nL=%s zc7XmsN@m zhB`yl<-U_<;KH~lsI~n;`$G(9Yh5~9{S|p0>|x_#JzhO10p7~8;H!ENWm-$?Xij^4 z6T1%v1QmmF(+!xJW{QV(JWyt3jj4|2pnqvT^|70aGo^m!RG;4B>xZj2;_V(@yrhC= zFSq8mMgOSe%10Q~?m*?b!ftr9({@;SU8LyiIh2|-1K+yGOMYh8VdlGF8gVob%Wg~m z<)smzR>eFoxIb$v+lxgTtYPr?fkJVZ4Q0*=;J1@fVd0MrLa*0`;8%E=>Lq@)-|zMK zRV@62qv<<(M2cbhz)ZqA8i)rWR$ z74(Eu7u&-0!#hOXYE#ipeHQFEb_HsG^gz#qZuq8M0`yCIP1BCO7v=@qVz*!g{!=r>eoAE!9UdzV&HYY!rtLU*rbr<(b?}qZCsKZT zgjSn3QRY22S}1uUUiB;(>Ry~#S$_R?<*@MURNA+QrhKX+`Sb46HQP2wfhQ&rT=ory2EZ%6GL%8IZ@Q!rt)7DRP)=Ye-KvAJ>=XKx?xqT0h0gp4jIqCmHcD=Mp;_Ju za@1bHb3W~cR9U^)ar_^dl1q}18B|SQiWj2RIVY|NQ{kMZC#XNLCy#d@#Kk1HV&=5X}Q%FIM+#kLOBONZMA(&!k@y*63N_+TvF{F=Gel zS_a^t87XX>a0_jt&hTl;Gj(2z5})wc%?Xm$d3@MNMck1=@_R#KvCGtn3QbRKDKVr! zpSV~LmM<6JzJ$x%bA&VUTvawIHDuR2WtcyH3;T?pA^DCev-Tf*{ByyDscaN}o0`f0 zlt<*_uSunZla z{^0A$&xGaA)#a-{-{sIryI@)h2#+>@6ZQ3~s7Z4wSQcy)kKcFW#wax$^|}YGHtmk? zv(z#0&mQpA^P?Y2Ri*bjMA$Q`ip*o~KtcEo>N6JUQHOSzJMlHi=2Z$E9PDBG>=U$M zRRZaKZZ8_1$fc2;CyGNnJz;(GL2 zDP;)R)eRNHdz>Jvi4DTNO_pfroI-1!M-06lyMt7ooP|#-7s%F5RmMrC!x?1$@O#M~ z!Z&jv(s(|NNbL+sv*ti0Yvae|TghS97HG*{Lv_EE=%v&Hk8Jn}bpD$#dBA$;`m_=z zPMQy&M?NK!=Y{k&QPNmuougXaW0jZZO~LWc6rx7a3)=X6Hua9wX0vS)ySQG7bLTIj zyh<6>Zcd?q?~*lLLQhocdRkm<-5YY<%(2U)t~mTa3%HuBkS;Mx#eKZDa%%7(YBbOj zJ-m8KzI$P``kX1l6{gX;*{U3c?dczw#4;_QS zUAp2$BN2N^eh8sK=8#yQ2M$&az}4|!ZlN#Do1Ti_-y2ZFsS?_BRmLGJ_mThq7&`Aq zD%&uQC(0JGH<5;h6mg#WI#NoSN+C^^(jt{el9{~{C9*3dGwVF}b(Bifn=~{eDV0J) zqCvgSAK?7pjOV_u>-+tD7>W&MP8b=Vg+;G@aM3PXERfKLr*}f(;u%AJ__!v)%1iFR z4&P-4mc7IMG5h($Do;^6QbG7Gy$8a-DPrD=M;t$_4%G}a_=zR+acxiq`oHyrN0qbi zP)soRj7>tf8EfcUdKJn8ccHH1jk<3u-&5w>Vdz}3lX|vi;K~IP`NmD}>GZwhg-xC+ zeBOgOFym(j-j)mm9VU)ODbzBvU)K|*3$nO1a~_uLDTV%h79=Kntnl@$Sh((#Ln8d& zqr6!WT-m=AN;9hPdYm0*MN@F08K}H`8J_L<3r*P<5T7IxHGhr=v1W{LnWYCmU1IO;-j@0JViyI9k(1A1}Q_=H5;t zgPY#b@=uCt+HgBs_@tjv`)PNWzbB_&#+% z-j5#O`P(0dxf9#jIYHiZ23bMPEGL1=Rwv?NxCn3Sy@6cMhxjC|n`9YBfXb~D4BoIB zW}DjM*rqw~e$Pv)tyoAG&fr+9K9WcZR>9AfDflTu66Cgsa`#;!_Bp&Gal%Q!Y*@%U zQh9}(nt7ijXGbwZx+}?9t&^7L&wXJZ7Szxad$r)xouxQNtBH4LIFrsTAasa%Mj>5` z4dxTbpC82&dEG-cUQ=W9*6NVu5m(6Il#TSg$uR4i!LrLWMtBj7{)ICE3BB`iYsnqqG{$q3_Pnz_Ly$rc6b74#m}(P zS(QI?gBE|>{vHTuy$MSyLtu1wJvq{Qkp5nfhQHj;;Dzh~l(G;P9#Wnl3{308qkYYY zRS&3nq9T5&D1oN+MbI8`8dQS%;Cs(Q$hPc&zo`f>#rHz|D^Ygx!(ViBrM z7R=DNO*DAAsQTTNyy;trqkq2;v%%@aMQt4@^zUIj_`9+9lQ?u7J_2!Okt8f~A8nW` zN(zk(nXeXUa7(U{P0N4JKCqFY=Z1Viyn#mZ>J;G79=oC43vZ8nd-T_$n36@?BH}g+BR<; z*TwgxTmL>_^UI#lFCiu{yuOmz{@Q^4bD4o7CTrkwix~Q~T%^|Vhltc8fAVtFk(dSC zp&lDH(}0o{)ID@IY0uB0A1>_#t0(tp^J6d0-(w0I<5!cvZntTQm>jyiF@^d0{-j*p zfo8t@OwGgmm^%4h@*<17Z|g|oQJGRmTDqUySG`TMr9G&<%6S} zI(qR0yg6dunb~JwFmok4&_*wtB<9MqX%!vR=wk-VmH@2^#W&r~xUTN~KA*-5|8aDcZR|LBLwDr8Nh0a>K#kB=ru z^7&H~@x`JgsOcF;-khrDdR}hWyUGt$1*UwhWr_GR)D%bO9>6833yH#{3ev593T>iP z_$4Rnu$o5V+n)LSbu~spn~SaRZ6pY42w(WFW(>b`PZ#sY!KQnh=b@&aB*Ri6>oA@#vOg2TV>_|!qbJ<+F~;3ul4y1y z5N;}j(}lAPnV~iv_VpBB>RZF*Ic?+6?>hGxeeeSOy}Z%s$XXz+*%+~`9Rjw7Lr~ZX zVWV+66dnzS+i{3%-XCB|aRJ$D-9^&3>5%94&e&X|gU))*bf!)&X}ctZg$_kfoXAqe zZ%^^)!%MiCoq?&3++mDPD`@7+fcLbGByj&1dRN*DuXk+3YkWKWy0C(IztkNzRAz!< za~*`ezf1#@U*U_HxADxVE$*#hKqqYgE_9!Sxw2z1kmGo4Jva$BEi@C%zmd;$tYFDe z*98zFzn6QzUyvSAbu??afgQcgcz8=LS(PEiKbGegjb(3fTm%J^Pq=vAYlNslW=WPBfB2>je}YLpY$#oh2W0JqGV^tW zo=?MZd2t=|t#aYNZ*Bw&-!Q?|!5g@a{}*4F=V6iIbNVUz9PnL+py^u;gsi`jXx-Fus`}tEwRE{l?_Bnw zci%~|bCqVZElwQYcFtiuyC{KLO%cU{6m7bvrI+qpd4b+KCZzq!W%S;@Z?wpCUAa$ap=LZa+KSo*wlSptb|aP z{%S{-;XW53&o7dfl4;a&_7T$D+(Ys^cM;u7*XRi@?%F7}iF`k4#4H~+BNc9AneUf> zla}C&)ehiT1&TdJR z_^m?iid8hat(aDtDdO7xQug0E5st%umrj!CX9E_^#^{gJF!bzKL0zpGPC4+FE(mu* zr4Q4nt&%2n5|QP}rzj!2@Dg>gQYHSz(r6eK#}n@q!Ch;HXdUO($dh|b1ApexXw#K6 zFyJD+pR|bR2mBDcG@DJ*O0((HaBnv2>02tub;Q@2Edsl&E*i)A3B9`r&gqbYbdg1H zyGH|Vbj^TSaVB6oI-iwNTa1H&ce&poYCZJB&TkZ((u{MUlpF z!l!IEM!2_O^|yoc*gHFz)-aLStzHN<^fX-U{sgAS-$0dQ2Q1vo%`W=Op>IG8oLv>j zztx&>vt|$KTVKGd%2#pnfQ&GF=|2={x`(rpTCnEMM*Nk)dB^f@Gl{lyVB>-qAl&mz z;z%QGy!8!q4EDl-dkLWQY#Dr9dzxG|nuNC(o7R=A)d6qiqnL)WuMP}ZK0ArsGF z=(`PgMYormGxyS2ksGNG-v|Uxr$fUp3IQIu(6nVD+%kAh2Ajme-o-;;{B1Iv{aVb7 zu~g1;b?`f>V7&Z!gz))t;5Yf+pa zI3N(72v0!2`@tM^F6Pio4l#nzTBD``c+aY#hV%#OdPI)E%fOF$wuv zbC_$qKt|ME3H3fQa5Hy=bOZ{B-3otjGM-G7Ug?9Zl@t+aOJMU2BABs8y;#J>bn|DV z(%tUaROK3%flwC%8|@2VePI&JQci|o`hb*i_oDS@Z&SXQGad*w!S=1k@N$3;-d$6M zcc*FN!_AI(wPTPjOXZTOP{Y=LSP1Hl0Z_j?8}t&-fFL~=tidr}q-%<$(r zvD0wNfk*VCj3o*}UZ9B;iyN!c@W|_Y9D9w+em)IG{nA#pI7E~x`$W^^13SPh@HQ00 zvtV~7A0F+v3%6~fp;c4??)=k%j{I2Wv+6XuwY8l(aLxsV$C;S0M@)F!lItZo*W&c+ zg{WkuhP%9y8J+SZw#YpW_=VxH!#xf9D))n$?o}wwdJHiQ(QxsV6=bE~A~p}x>AMLg zxMuoH3_8nYMpVl2kL6^{$vBNKH}6NwXkGdsx{`*c-6wu$>PdahT$pJQ0^;h8kbjy%q-~)BU3s{YJ=FD-t>$>M zj;$L=X`c%{j$dh-;v_uLxe}$1s39ZrfX>Krh9}!DktKnt%!PS#f%oMsA=Ax>Th~+! z@!_0v1^>vTg!9Dhcpv#TW-Ym>r3b5GD~apVg?ORy1{U^BglD%Bz^i&3yh)P5vu)`_ zOi3J)9!w>o_ay{JCyt@@FUskDnPkD5pAGboW*2Xsha}u+o{!VXZ?eJ7oPA}o2dq;# zcjFb7$R3+cl6G>xfMP3ror;t8AZ}Ic0Qf_Lm$%b ziQ!S+3zoObn5x#lBns_XV9dFFtS-r7Z`(%_*ybjPdoW1Mt>!?wTr@lM-$aZ&W`k3G z2bc|xgjr~Qm$9=+qJb{XSeK|k29zWqeaU2aV4?ut8aq%WKFNyru!0=(8%O?2zQz98 zr;1Z&myvg%0=sjHKzw0|;H|?P+;Qg`vuAM&(;2;=dEF#SZkSr5qtp$?O_4Hzg_Trl z#}cqCJVEyEbEG0OS}4Qi+fBwB0M6desvYB@hv-GR(Iy4TQ#C+0hs(Lb~MiR1NE++1da9CF+S9uv;MPs}2!8VOjGG80`> z``DGk0n}~RdUob6RTwc4;aIC}$7km{Y2it->)u%(FhSh7E!(y_s}$MFqD1dC6Xr zR=_Y@6I5ta0j;Y~pyH7}C`*n3F_=i}WG|y#Yz)@?JdTR%EosoXnPAl-!q-bKhwuXe z>iU!GRWd*Zvi# z_O6*WSFHmx)oR%Cvly<9zM@`NWia^tL)-~tvx_Z{sG&ByoZziS6&9E8 zf{QvUU>rBk+z|5twyGS(gu*4LoF)#>yc$SKn*t``%mN#t^7uPxc`Nhauvjk6M><_Izrp^>U`}JZ%A$PQsG5yMLxe)5jIAfVy5Uy z^l!LICl*J6Xp0orLH>-JUG_th{wBC~(+~3k!r@@lZPdPP%r|jvMA;5~)+rj{v4f^i zE>MzhbtjL0xO)$Cyo$i@%So(i2m;ylEKyLsgECh+C(h?1*gJ9x<5o!XYkOkw| z`H3g#D1RKEZ!9hxqrC!-%o)$jS(Qn@99s>{mU@ie&B82$b@=6z30MlEahYr@-inaq z4=yex&m+I0*4#oUslEb}BBu$r#&f+v+cJ>WO5%i#4lvQMgC;a86U~gLRNLJaCYoD; zzS{t9nm8SHUXCDdez;Q(g-1Sg&4&5UA5qs@JNiyHjIK*Mgx9JPu``g{p?aOB3nCuU zM*HoXdMxYEb-llb5GZ1e~Xzz=)oQtq)sj@^%g=gUK3g0}#K#&w{vPe>-b zjD$TOZ;^oN?IgJs#l`vTKE(UgfaRre|QQ{?alK7@xB0MuLKk!Tvwa$fM8#gz5 zC3XsR%6!l#nB$Z<{Gi9mztjF9Gukt84m8NiF~%9Qz+Qhh=y#MsL+>`2dq058t^Oe> z5C2MnzuDs0)DY~>H^Grro-J?nT#5AJg!(?cs*eaR3=%$f`hmQ!G}+Clc+ zHa7^*I8J9gxIh->zhcIkxGS^)O&Umv%*vq?~53r!y0#9_=pNzHTp!7 z6b->DYc_~yyk(7sb7rsq za@FRiF1F%_`F{odQ2|LVF2N<2EQDi+oP>M7{y~usgidXfc-uVF6z^h`=e>tujv47Cm_9NFU
xz zYiB><>)6ouEipK6@gWp_Z;5J~G|@a`4s_4l2Q#bnXj5P=J(OL_R?0=dgt^^J)ueXv z$6_h|nsNmzmef-HnkpKzZ6Z{(a9#;R0r>tLBH0V0=wZ$`Hf3caE35QV)ucj>^TPojkE@%V(zta~%MqUlWwAF;& za`T0?QPG&XTL>%sjrkkQ75EF^wUZv#@7U+_5!dAA;O0FO=o2R|xOKk-n1v6(M#2ET zZ*d_p*Y0C<7Z;$k9YW{P8fNtPH8>yp5rWnvz~syVTGBZI|K1Ehg@#c4?dwi8-@YSw zU<2qZ%H)1SE|K~v-E6aQFwWc-iXOkqY4=D4x$99vhVsSW#EUT;yGfRYa?^ z&@yJp9X?9ilo3u#P8_$z!I>+jaDVL*j0+NC_7h*+?){5xdE*BkeLGvK;J;N#WWK5PjXf#Cv9^4kmNYvZo&`#*b0Ot1vDI4kp$!*(7ucbHUo~~qNGiHLwTy&_-R(1^3;?6*rNgdeS zeL+6+#Nel5Fwq@cfQ2ppwI7YqN`5tJS=Zq7cTIT0Jqf=X*W!SVB9e^7Y_P}IiV zf&J5AaLfbPc<=#~nPkA1(oImA%Vnd=ZBbk<1CtH*;eX3yh4tOR*!ja1*_g5T!_L?0 z@+>V-oBjtxbKgN9G=jwRVmP@Z57IR5ldpej>D^*PQSI-z=B&K1vAGj(tKY_RY2ldm zgCPe)svvvYLkJwB#}D!vhG;J_eiJ_s(t1VMrcd5f`J$BY5|I<8gvkj{S&yKn&MTZJ zOk#Ef=8?{CEu8QBKS=a3TQ5TAdYbO6n;d39-xKl#JkTRIiKijS-7xyL6X+hy=(q%7+7KjE1^vxSDC zUUGw%NTxbJCrRgrSrG|U^0vH{Zv2}hSY46{9y$83x??Goiu=O8Td|IQJ2Qs6lbvUO zT{y_L9!eyea^mXP{umngBZDR_527{0AsjoNuO=Fr@A$xWn?~g$~bt;{PCIKty zWVt)k*Tx0&aLa3OWDXk}jr4cQ+B1H2(Gf3jK4gB)GfN(?yfuhi|y%O>>g4_Y<$ZCIyr>NYaVkQ}O%}uAdRBOK&?(#0Lkf z>*5{f(&AmBII(dCPD(4GQ#+$@t>a|08s>cE`wx(SKMUBC%O}%65pJya+z9fq{2vEV zI7b4C?8qrHmArXp0&YD$WS`DbY8$?lIJR;OeKktDrRA{GYeaD9!elD;SA_m}tIgPG z-6Yv7_+-9iC|$K~AD&E^LbK(I>COK>TRz`G*?-;(=rW_(q>sJGCat+mb9{G`nzgE| zddw5HME*Lt>GX(wzVbeAg7qOHvapmfYMn!-FS|zy3@!`g7Uh$;DeA;F`y(&^n=It# zZYAo5!OZfxbJ^rONp$I>M zedk!~r4^LMb}@VAoDzH%c~3W`n$k^6ZnAlK7;#@_fKwNCQnv$yk*qBw zKPsHb__OA;&#sMx9xGfQ~SxKY3 z%w0dSu(Xht6pBH-Dds?HQD7P9Nk-XR)l?j-Qc5yo?z3KME*f!>F7_-AVkginoS zAa(IDtr+Op=vpl6`p~Z=F`}k7f-xqg~3TN1L23m)_m!C)oABE8N>I^;{SLnC)|btR`TWv z*eTzG0gpaI{FZVM`?WwAskasGO`FKS`1lciaPFk*=S7q6Mmb=>WpimoGL1G@<|oM9 zz}T2%dLU^x%`PVnPbdZ)?g~2N;4OqBZL3locV+irM+^NP-@;WRF zUtKF>My7Uxh~8yddD0B&5&>H*tHi%BXNs_T;~=@ozXaP2OohMSAA;3SYT;JXNeuI! z044Pc;Ng%L$?YkiJ?}N~u&X94pEwl_hpO;hcsu-=X@idLG924G16q}A5<$mA6{+5C8lx#5^Nbi9Cf7FHNBi!L+ zZwc2k8i)RKEm3ZsB1uY^fdh(?{G(3makFw6raFD&7>sk^WY}~3q-hJkTraZ&3)Qgf*K z? z&8Fu>w`(^~=RZ3dJujB%OFPhjbN?6-^;1?#W)Y-0WIC@ua~I8+oJ=xDN{F6f9##K& zmMTdlQOLGOLR4j=y**h*lCdeyuPw+kKM3osD(XXv?>?S}y^k8{W-n75yVZhs`j;2s zYqha4Oa~3EI0F^m#(?dEog}$s1C)72z+Z_VXgAe`K8H)B;veTe<#>z%RU+W z)rgv_UV*nyRzlc1C8#^72WAzC;3qKPxYh$iN!%S~oYsJVfd;}RNTGi22kiU%9X&}8 zrE-x}N#;J)`*D`sbl-}PIwNa$SvAt(kLj#*lmrs(+cfR_1}ZY-0m0Mc!E?`Za@uh$ zZ1l8-#L3MxxFiY^|E?mbI!oYe@;Wk=<+{j*^WnGnad?$t4d-ju!LEb<5l0yvl4Ed{ zET~vboNX^Mqspd4d{CP%T4(o;_0-~ z_2ANS819#463Zt=9GkV2-1!#@*=gTMcv=owV6hr4Y?|0vpKg)?CvCds$#sFfxDFFB z$C=bPDB+-37QOpp3XLl`NozKxQCWTl|pS3Mg%XE z6iAGB5AD8HPqHo5G5mKYTUE<>i~F8XsK}>!gn{G99+c;QkK)ocdOdGE`|*J^m=65q zvDq>-KSPOb*f|@ttT|8BrRg|WBV>Gj^^!3cE7@X)Al68>h}3f#j1;eb^qAHd9H{eS z%jZN=iRId`WEQ2bxp`6bUzXV2*Jt(F+w}ICd(5Jfk%EShndI>tmgwdWvYs=~F&DRM zfYZn78SKYr11N3*Di`aV`vx|>?NOY#00zZHBg8X}Ja zK9f-PGvhsOI*2UMLW$H2vVHvllCrajJ)U!#2!@UmjjM@N&1@#|J^q36=k~J2?h5!+ z!-RgC!be5YL_ZDH3#Ku;+}^U7sU5$QyoYS;5@94~1cSx`Lr_ti z3*X0QgO9i_DBE~IjfWxleVRw|E`MS_<@liIksi8eRWy2)1>)WlLJlvMjRj*a(lYLu z{xNn5QI+rq-T8|ka!WkqbqqrEm+Rnkb_p1Ne?j7cSMwYl5VNt4FrO|R%F$G!N$vIPfs7oyp!xoCc94n}AnVy8c{gW%KC;hwcPRDTn| zixUNKw<`&AOl?tJcorDR8m#Abepl^Jv17HKP@n8i91LRjO&;ZNLyur zYpf>2tdSf%d`XF_kg5FVDt1EHa~J+~w8LDL4Z;p(Wqw|PJ+*6?gl4MmS?7B$G&}Ya zd`UABcK?W`ce+LS-FK@nwlW$%b2Fie`M1#e>J)hLa|gPMq{E|_AoShlhcg0OpkaIj zI>Z%{$Q_4h?u~<>mu|@G4IV2r-K59&yvt>mjZWY>zgevOUKLWf!GbuFYW7%p4tssb z4$7V$q2bpvQDv?Ryj(FBi(Nwn%?^%`zV$Huyy}=>N)R9PH&>(W^_%efRXwYn(~dz4 zT0nN)Mhq*MhsQ^i`SU|0g%(%($#REmaI{qvcG*makAX(~(;Z`l-?Tc>F3Xg^^qDji zY0eTJ+|>qE+e@hte<`|NdIgW8PGO=8Qt#tCnHsL&ZB%d{gW7e8!-Y3+!@w0!kVHDc zsEk+-Y^2NNWLrV4o*gM$7+;9lx7-3ep%=r=}UgG4WNG?pzUJJL>#-aGd zE_8dCi#|(Yp)60GUmLTWQtqx_mEp#StM}nHlW6RfRphr&6Mnh2A|5Uh7a9y)LLbps zVye~(#gA;r9gi`L$d~QhS@1O`e9gh}h8jdZD+HqT75VdJcF};LU#P0vjcqSQLGJ5# z{`FPCgz<>QgnE`Z#l_HF=cf?+J#MI!F@`bWIOOs7&w|#ZUvw7fL9@JfG!=M`x9E0fxMwx$vjuh1`~fO;*nfY@nK;5=akVMpZ=x0TXMUq8|j-LHN#550hu#0jY*(@OJxa2$|(e%l|cD zhSpJ(yR3s-;6KVsJV^d2Sa9Ee5x%Q&3cNFTM2D62glblX!uQsOD5f!szxu)ke)93V zaP!xD6xm=ajCms>G+EF}{CJ~aGx7|sKmWq)ofX60NiD>wF3Py+<^|kR_lU{AxEL;I zXM-Al5A55WL&8FCvPrBHW*na=OkER&+gUy|e=y)TsH*aJ{B@yEgJp%IolAwHHN!N? zFddlZ8vIMW_hGooh8|O{!tsrb_|v~0dlu+`R_-l$y|Npe%G#J;%Pde;`waS7mZFhu z4BIg4Ec~83j-L>4kE|G2ho7_yknvDL@3k*!)`u#R&Gl=}x);GksU$M^A_?1NT<{0y zko%mznt*H&@RT^lc-vdjkaLcfYA(Y~lL0$j^f7Qq6@Qn;!m#u!csIF>Fjt?@9<%RU z?tUIdzKs#&D96Af&ls5Tcohk$-h2R*{do4#daZiR%}n;nplAsAR-A zEL8%P_4-6=mPXOY!}nP`VGo6kU#P%l3;vvU5>gJ>!8z@jC5G?&z<5#{xTjx7c|)(r;Ig%(H`5S) zuD*x%Ywy9L(Mmk^<0-2<^C_=AXdlBDO`u&z50RWUju|(Q%u19R^R1#3gb!Nf$)(w~ z5N~FLeov;6LhBW{c0(mGmbAbZyioixP6!+Kt;V<2iOhr6qcFW;JT^`Vhs*gUczu`w zwZK8LhI39H|5FPZTl;ag_BnETbUd60nT*m6V?e`9NYqPi0RBrwNxLzywkH<6{p3*X zQv&%MMTBbQ91Cq{6C|j1qpHR>8j_F#CugpL!gV)5w0#%su*}8Be-vQobswGN%<*Xs zM$#MCW(rk9{<0suHTZVT-|$DvYj*4^O}^^bH87@kIRv$8(B;Va)6cx8Pp9sL=HEl; zbh!+_v{=Iz@%T|v^yL|R zy3vl!Gs5Ubtr~6*tN}g;(oo7J4!*U=X;Tb7-o*4Lgn`DC3GgHJB$tJ|ODnw-A<3tPcFw;pP@Gmpdymh8w`({a^>qdn z3U<*Og=*lfDM1fyXQ&G2k{nv738kwiuqNgzcyD1dUHzN$t7%x#MM+#fJ0${y>i3x8 z0clhSy-G9fq@XZ!fK~-HkWD{a$&4HJbi{uu{uqu&%XRDFoPi8%{4I~G0^mf?v393*{f=@ua2!G4U@U-G7(ic{i4I+OQDZ{lZXR zSBI_t%jL8`b3E-C$8c=E9IV@BN8W4O;UqU-5~?NxBfbh~``d@f`LBzJ_iDgE&Rlfe zl8%}4li}*8?I2KJiZV%EBzjst8E$64+2JX}uZRNWvdomD|;bVH3zhoJ9yQcCQd^KQM$#q=4el5N!i-1dU zk7=ve2`rhn4Ea8T^ha6%-igm7?p|R`!R&0jV*Z$!^)HU&x`)xlG>0rYAxU+06Udzx zCrF`T3bVq!h<)#G0WXtJFmJ;zknd9p1&aGF5VN`qjEzz;+-YoPcN{h)Ip+oRO1wSX zN~$C(#+PaT5(Wq7Z^!$UYUohn2C<|K%#;s+WdCI7)``WDcd_ z;0ta{O2e=N%|z@j=g}H(#5cVE9v-^%@k(Vnan>Il;gKRa;e0;<)9RE8aY-`#D^o{c zt=<$k$vIjgGQ@@Tp`$1Kz`EAw@XxLx~AJvvb>1J~a9g&S}6py)|`RP_1{ zeSeMlu61`{&`OEMk7nUG$?K>!IE;DMud_VyG2n3V1zf2r1KY?-SeYilKw%qRJo5@Y zk3Pa(`>oMlVKz)E8;a>b&r&g$Asyf#`zdv*~?BTiBYBs5%;`z@kc1ftc}A0Th5hfy#*`zO|aST9PF(7MR?aP zb7$$@^o+tiq*;Ynwelj$9=S{xtW6;gCftH$%75VBIbRUby+^K`iosv~0xTT!7*#%< z#;CBXRKWS7^QtbvvZXKJ^1wq_VHOU(YM1Dx++8Sd+l5Et)rHMh&*I{HLV87S4=n2k ze*P28YjA%XXr0#mb}9nAE}kP6z%9)5H85ZKC|G_6K;+ z=k#ET*-b#u>H7)jySx>fymHXzs~JlkuCQ12GA@X9FTR%b&(>{D2>TokW2tR@z2`b^AbLpt9t zq;7wP3*GK=mKbZkp~e0oym2??(vqa}B*RA>&xGA)uTK`C8=f8JoWXObf9@CB)}{?z z%T;Jn3QJ4Gr_kIViJYiPXb+RsbD0;Jf743!r#7AAHwm0CPGV+%?iG9z-AuK{|6x_T7|b8RyhuVjBDsNYVk-9N7X~<+BKn6@Y$Eu50=32iV^zB#{yLP%g8J1 z1Uzyyp02VOB|o3fBfI~dBT}yR@c7Y2@_MQgPu^G(>g=LP(Ean|rH==6uUiZJ2ga1d zhB0CrMo9kI7-H~d3|;atj&^r*J;6`9Orw7cg)$qg=$Vdh4_cBqjt4SR%nrBwy+cRm zog*^R)A6!#5MI7CgX+ICCRW-p%>Ee<=oSMJcHQX`dU=y4d&#weVT!+yC1;jGW#vmU zsX2u%>stVm?8HHe^x?sB*3ZiPMW4Z)Z5mPK+j-!|nL%Qm>;aMnrc93Y8hsZ&tP<`gez`*-?Sx|Dse`YJ2%_(o=W)KG&FTPEbtOkUTK zII``+U8+Ajo89_s9gUJ*Ex@MFtlRh~(w890^;AS)vAq_NO1(<9riKuwwKr(ctZ`r{ z<4R5~ki|W2SD0Nx>zJVSIM%V_8ci#g08tV#ulTz=bjvcwstoeI+jPaG_B^veo}P8(CmlaX)gG>akk;4q$CDWRpSbq`imK_>g+X!<6Ot4K zl_Wv(?CzQFK@?F5qN0KsGe#6JVL*Z+NrD1q5R{-|Kw);zbdNbJC<=n8sEB|8#e@O9 z6}JJt(n^h#TZMsW1}cz+}`mBU3fYj7tFnlKJ=R*8nPk;g)PpaVz*C-It`@lHuvdT z{vFK%_i4PuQrNBRiYQDngZfDI=%KFdGFL%@~kx&O|lFSm2!6UWYv0f?WhQc z!=ECJ4oG7c4#zL6U!e#^UgyO4Wa{KILO3y_2Q5gsMql}?l=b=Cm#K}r!Lpr3vroT6 zMK-(^ettz099`K9*6o0F6Ai1Wr>>xrm){a#t1@_eek zXcDuRM6-zBE12r9CCt#XJ8RpqP0>?#t7zqXYw7)acaY&Kb$pgjz))I>q7&C)o0>$N zF!urKvvQd1-r+=6HWDl9p^md^u9WsZWX{}vsk2Q3w$uIt4iReO+H;r8D!sX>$Mrx zU*o`jZZT&41An1x>2#@mz%fyxR*p1AdnQ`-Oc7h3I}d;4ZXl(TX~L|D-RV8q1^T0U z6gs%9l0LI6?k0qN*Sl*&88mrq#1#wcuUAOQRfqmHQx(h zMn#IcjB#ElXy;J+a(NIoc+;D;tlrCdMvBCb6fI!BpP#76Wpu(CGe@%EC6ig5?LCo8 z+5~J-U&>~;KS19Dm*e&cU(w~qwd~`&GCZR9Ib6jJk>8fg#1rji&^s&ov$wyDaDZ_f z)18!!e|M$G8oPYu9gSPXr%MOW5#m{_w7(I0Q~yzv4;Ny7d=iU|v7-0RoWUc_-mv*m zQkh^J!Bp!PM#_Eah z+D>D2pocujyBg0(8%BSb4@U)K3h~~^4%V}2G`8<|1g-BL!mjt7j3e#_P;sUTTHrZ~ zMQL}Sr-2=4T0Yu<8+M{s6nnZTQV1Sr}YRf%WcbBQC zf5xY)#B#ltPx00%6}Z;n7d`c5Ict7cD7s_6A8F4kr_NfYBK2!0Y2@WN>{HcA5A~fy z+i@*z^zTA}TYkt~te3N-qs92`9EwjSi>_>rujY%Y5&~JI8|qS&e#G+>sR2Lo@w~-sYtr? zR66rMevfU`Rg*V1$20E|3;Z}d7bhjObx9@m30H8Qm}%l&wVB9ff`~mb&1F9im$HEm4zfn;XhrU1rz}mr z9qqn+6pxfWz{3tT;?v%Xaj$N%v`OgBc2rzpftAMct`f5 zuO(ja!A{&~(=f5cYDF)Bynguiy|3)|UORb$rI|cyOD(IMl!A6P7hvBD8e*CIC+snr z$N85+(9~5UnPt->RG;d;YIIZ<%?$j;p<%Pi@x*P{*tb63%v~uR`>Q1~ z&)x}W1%DpTtv*E^H)pcP4YP4);8_|ij+W`WPs66Wz1Vh3U#eWwhUTf2u?rusp*1}f zc^+GTP*?x0Xoht#JMl6|Y+K|j|Jq#*FS+a^KD{+n9&M&7=3ca+^;cEpD*mDPg~F#> z5TJ`aw`ZVnC)Tm&sn@Xc)`Kk8dpBOn^icc5uQcMz78X`;AEkevjxWz$!HoJw;kC0Y znNiYFw%sKae?3!)xV6(zvAjFW9;qvy=AtK0n!bg#4Yd(V4!O$*JnMtcd*8z^LzU#c zhHb)ash?$%%?YwcFK=Ltfmv+8u&MO9+fvLA?8PP=*J2N{WAT7w9Tu&$lO=9X$FF-! z(JRN1EG?=4P12vNaR~>g)hYV(Uhw-a+rRmutSQ@@V!_gsPg;NdJql$1}tQO;sOKT>XtUWZbbjOy6do6h*8bFIgOXCE zyQ~M}oFzf*=A2O3%S~@7=-r*Ywd_SlS0AA7R5r3D!P}|XjPKM`U%@jin8e1nIip@< z!(_^bo9IL72dd~3hbC*gv(I24t*A4VwX2x1`|4L{ee*U^%sW$bRpY$u^uAJ5Wu}Nv zPrFLH&Fv=hz4;O~9{6)7|B?Q9K3Uc^ULsprsmNKp5G+#BbZ5_>>EbHKGJ3@Bj!?@> zgDpSs7@0pRMJcNaWV&}^=w)$>Z1%#3$ZIX4d-m?4`O7Av6`uEGS9*U#i)LJ*H?G@L zMBhY6LIW0ag6eUKwBa_Wn z(aBArG-7Z9+H^;a?Hlfm#7!l#h-a#@7qdWf4mC zo=zWj=!^Qjia|w+^SR2a8XDi<2hH8$iZ;p@Q|J0UXr%sNsYP269dcq5(h>Bb(__Dg z0z@}akLMfcz>%+!wSp@$Yv~TUF7_r0J#YloZP-t*t+T|A>Ah%CJwdi6RY?1L7_+f+ zmH77gol)T^ezHy(Th2F-A2?IQ64dYG(jO@}rEU}IU2>g` z-mSPU(o`2ucDjvc2$kiVi5GiS%Fza>C*H8*D4jdCl`Y7xNAZf{jrW9NvKj{vNL$$G~dbMcJk%b6$G|Be7+lw|7 z9v68yso{$(L+a9dqfGro405nh_@V~pxb11CY`>{^-iLB{Uo$&OszO;N-JRSelR+#e0j1sRr+FHH^3H?;r z^kd32CiOKMb}i%9p&*^))MXC)a;ej4#QMd(q<(R` z=*XHnTH~w1qN{aS$BNH%^YSmWrguCOEfliJJHly@Ll#x3b!Hp#GU&XSS+bi&yXgCy z3Z2Z)5#`-Tq4Cqki7qsCpbxW4WPZ|oS^k7G=&+kA9(=@HRuh$qhRFNls2%svs_K!* zFf`WTDIT(ct`W z>|^K6CMZ|ZWmCc_S)n9em(#}HPc_7Yu4JG!a&P8&I~&_5&eiKbJx2EViWp<$59GFB zyD-?`6#F!MCVsU>!KJ8{&E(a(SaV7&n|@PETo!DG@>I;_2d7lxi0@-X>uwFAy{uzs zaW-KaUKHTS=Xnawwwr9s=~L|2@nPbto0qVtv0v#5>q)rvU_Tsx;Yd^*1GwiZwR@`4#=4`%ni%@OZ7F}IhJ4SQb246Qe!qH=^i6=hm)%tzw`YRB2vTYh-efpl=6Vna;XV3Y-{;4`c+g`~4k-*Q}H9h1GXh#I7qS>FQu4 zNik)Mq&-Cj!*!+JVeX0^ybZMFz+|darXthH%#eLi*JEwLX0oUS&9a#GpU0KSLRJ*h?dDZu`FerWOr}D;B{l7xdks7tN_x%95s{a1a(Id&V07Nm<-xip}_AT6}!ttATd3RYYV zE)X@kOr$G!>9b|3sZ#yE9>h%O{z6Uvz0H>&OKjfpZRBKuNT7+Q_^CoCrz2dM=eqDjE`u} zoe~F0MUvds1UaQH`Sx~SwgdNL~r zbL}H(SW%_OM_V7~hE5P}xSlO5w3{PqS8)HQ8Lz}Dfr>jM`}Js$g0p#TLm-+r{R-$;8Out3_~*2!L~cEgTm%4l9)19CDyDMPP}@u0dwx_3zuZ9b$) zZ>~Hm^4-t^+4J+#J;yAlPK6E{8#j<lCTW?>&qXw6ampbS)g!v

?|zN<$TV@azCF>m5d`@*BW|zeU~r`(a7#G^*@lM>>z0@`uABsei#_ zUVhRtc8o=l;E9L$nJ2xW>gpBt4JwIY@(QAuYvyFzs6n0FU(>RS2g#+ub4kX_#dy%a zSEMoSH07FZ6S-=CYQeuG^H*(!ZA-Jt4NeRt4H=;@!pIAnD$Y{VmzGd-evsI6?FVXe zH4dVuF*4LbU1YQFqCiHP()0S+;L}fn{bHP9iEaTH8z^&*-``WCJL&YA`g!=2P)JKP zT}W?m0w$aqhlBEm;oaWOT*!E-vy@Yd6WON{s{4Oe8)~y-Q^WB z*0F>q`$k^R4!}uP#kA+uSKj`FB0CUzm>ulCBo=k{1AUhi4EvagYC6*JpymY`6!MLa za7lxt`fH-<1+{E#Od<*Ne8$)k8@lgR6O;!>u=b`}aYwNYb=)NbZObJ4&dXbvRr->h z8n&2!o^8Zj=SJ~`;TFWDpGNsTqC&fJa{1Q0D&DcTid;*-MH2RRvsK1Ja70Nt-*Qs{ z*X(tmd-kkntA=~=!R|nw{B0HY6rQErdSEVpp70uKKOJ-GG?{Q=5vy<+AsDWbbeC8~ z2rUOWZWnT2q0(HT5)3qXNFYF=Tyd9Zb)`m%lS^8!j(j z&#>VfO?kWtS`#Ij$SBDr9aXEq+jg*|f4h!VzBA{8LiX_n7Z!?NX4dn~UfRO9u5A*H z!Iwpvxx@9#RiJr(CY6enfd~GRAnuTq==-H!vTM^_@z`^!*io<@O=}|W;HCi>XE3L+(d=BxLfYK)iIu-l z!LIs5l*{#|CCfImwZV0)Lgf%1U0KFttxn*@p)x}E_*x7I{)Klkta&Ew*`dky=R>>xj~KHm7DeddWAjs5T4K zI2VT7^sLci;w_e{WG)D=&4o0rT+~V!!a6OFvgos3{B}86p*!;lZZ5cn7Q8=pZqX5X zz8T~F$OIf+b{c0N)e+Wy*pJ`W-o~<^a^_=u3GaSW6egWqAk2~Id@;)9Xwo9-5z2ap zBSzHVy~UBZXJ0idP(4qjXeN$~$t9X`#WZLCR(v?}6i)Gq#iEUwxRD-*KkJemACGas zh)ZMG@4x~ado&TluUNAmmHs&X@G4>LaV7q>_Y8LHvL4Jha1c-ByW-FI;h6K&0kscV z;lGvUgbVq`8qdv?crjk!^Rt7|BhEy)H!Ty(8;Nkt=NHS98jY!T{RP8IV}-TzCgA$M z#h9G5kc~WT&4ypeW!>c`*@G8D>5j>!!f>5|fiYtu&boA8WH zw`BxY79s!Mj$OaA8<(dJW+Y}G#@C#rMQSNb4oX;Ynxq$Nj-v41xgHmf8;;+W6^L^@ z>sjvaGweZ1Ic`Y4&4z4D$BT|*Fm+HC?hm)XLT&&?{aA(vuicY$CEPkqUH|7b-kj^~ZLNFw-1zX(X!3My3$=EZJ$mCFNgBqb$OGRh?pSXv zdhsfbbZdz7@^T2Nx0NH!vk#&&p)2^y=P3OC^&9LnT1^B!ju7{~U0C_uPcoVkLY^j3 z^u9de(Behuv0U|hcYdNN{vE3|R50rsD5kH(5@l}MkMupAc^$_x zdDrjm;hjy`NoE(G!qdDv@PURDbfP8(<;2}a&0!W~{bYpEUo$2#|Ckzw zFy8v~bG$LNGd!oaqPQyCgtYt+;0d_dlF6#?xM=Q0Vuk18XT!!g>Wwi~*x!#ru3QEC z(%bm#Ofzh^EtsURyx{P5JK}xIm<0RwusP%qVi}o8Hm|*l-HuA~0xpR1&gR+k_Lhs2 zOUI+h%IRw4$XYovk7G?bKW$^S{mdoeTgQnrtNm|_nM2J`h$A<3d~mXX5B{Jso9xdN z_^97U!qZ2Lt8*~c(A`FIZ>FO5=9!S$ zafRXe383|J7D7Qt4a1VlaS+RPl-)OI`b1j<|Ezh*DVFC40ZuOJ9;x7TCud>dgQaNb zpDkQrc5-F34!KBVCH4WG~TBAxaTB=GEUto$aGqw^*P z>r9@;r#%i(DWs3mGcv*9X?Ho7g^oe6ZW@&=CrRQCwV^|{_fduD0GD<&HnYm;V#-Ub zA^dL!x|Sxwc`T*QeZp=D*so9J{Jz1RuSrT6@txpB=>R#SgIQ*-BfLYtbSad_;0iBYFDGl&DuM!b^<$VI)MB-ji#E z+~f7pic!W$EZLqc)%S`FN^`E~(yd^Q@Q#e?DKR(>jx^E`qKJT8*WuOq=dV>Xp`A_A9xxrkO59wLT+ZXnsQQraw= zAo}-bP^z3yzYtvl@7tckkz-DDVv7}|)hj~ySx02**@!)&G|A0l50FjKJg~n`BWm&& zSrV`k@2F0}f0P8t@tu`8a-kT}3`pio|2&DTUol8QON@N^mWbCSi-wow9O<8KkL{=6L=tbKvPp8tj5&x@g@B?tEQ6riCO6s=or3cihUs4uY+ z5=J68PP>AUsLXa8cWMECC{a#@XMRKZZLXw~zQ;7BaOgt~I)sy;PWI<3;k(ufU>n*) zwb}~7jknA2%_&iO-!%t1?jt|beNBVrkADGoEbDMteg;HJ^Rv7jZOHMHBPKCX@X^MG zeA8MDKW`P&@-_Ex$g?skWp6iF9~@?Go+&}yyWju!{`H^t51+cE9C=lr&#Z7pDqCR;x7=oRuS&h$J&8)p72`&Ej zko*aHY3i4qiTa<%kX}t)B-&Ag^vireeYhX)&P+rwia$$Hkbd^|Iv4F?DismmjvtoW@aB z9>IETailH!Ci)(u%o`SRBZ;rs^SkpolFfGV#z!NtX7OQeUq>D%df1bRSQ-M=0bXDl zEsLxkt>W#~C?QMB)>2Prz9C}!IYb9n!TwM+-fF|OyzSTi5yAa6DDK!vu>WUIYP!l$ zQARw}bZtgU3n={fe7V_qrCMs`gB7Go+yyJnn?*kRbd!z`7!PLILE#yS_=XMRm-r&H za@iy!9Zm28+LrpX)rl38wWJ76wmx6os4DWBy1XkJq9xfF2 z;y+<}WTg8Zl6A^ss;lLpujmte&t6EMKji`Zd5uU{>lgfss{~5a8%^`xN_5Q3@yH!h z>WXJMwi%)+SG9IjUgie!aphQU^%U;wsDr)RP4JpOo+u*OkeJT(p*nk8nFl#h%%IV6 zIF?64?2(mZ?gLD81Hk$GU zlU3;K{tDz@{F5>pj=^pLLddy66V0hL<_50UBx}RBqwW3v)SRpsl)rMAI{lcB_x$!v zBI9}yA~L10(8x53KgavQ;5MGoX&HFdO zZc4;bu}}A#JwB!bQ`UNLJz*hAIS__)tCk{z{+HOUup1i|UM1x{0&r}`T}=LN#&^26 zAlocK&>Khsry&c-Qi2pj#Xoe2zt`lW7eKqm}C1d7Q1@)@(ITdhZCwq3&fa_z!RC0I}{#ukoUR?h` zEftPoe5?OeD-#~%|K!J=t(VBmC9UB6S`CVtD&cm!JgGQZM3Rex;ncmmc!T94>f=iZ zN-ba<>xdqpUcc$VHnU=3zO^y9MchCgNn3c{BYSwq-oJvoBFh-kXhj`uPCzPUUEG`I z72NUKY_8TZJ#sE(80F}Q5(zB?k82JQ?echd*_98Gcg|8z-Kwzh$~Ta?(vI*itRhQ1 z_-MIaF?vm$$p7Cz<=eGzE<7;+;LB$9lRaq=?<507J98m(*&*`tmovxez$UcNGy*S`l{TS$4ieeHIKFvvx?Sg zRp&CYQYV)zJ=V;O%sOFKUwM(tkkmszyQa{RX{02WVAR8z$b(oJ$+emSijcFi9)@;0DrTO(2U8WCKZJRdvt`(p9V z&BQ`kl~hhILPy*lVd=~tsMhH?30tKFrdto**?b)l$?QhfQ&z@d+3}G7Xb0KUv=X=? z!BpTcJ}P;5kQ@Iv1o_M~AXRla#Ith^`Mh}x2*%$;HkV4^UCMm$DIEc+;beF>SAbZ! zPGGeWZz6kOHOynPVV3fr;BaR&tdW?8PdLhfNIt@jiIsS-%rBg2xt%h-rGfd9vq`$Q z0e`1jL2v;UTjcKO=<1L^37k7ws-}z?)O3z@P33b7gEHJJ~V)wSsVIw zREMTK+0Sc;LK}n^ll5ZhAd_&NdOckTE^DyaM*Y3yEZ=XRLM>(P|hZISxk_1|t-iq8upW&VUv&q{H4NT%Ado~MtmV0Qv037zXhP2Jz zGV51gfWmp~+358aI=r|YBrb;H%u~@2FC0XDcBq7s>1=;BbpVv~oauudkLXuH_u!mm z6t!5x4M*~YvOMPrv%&~Dd_Q;}#TS}}!;REA>to-d$9f59IP4hHxM&T0m=M9qJ7-~@ z@Leo-_aypq#Fk}(7I5=!Y-9Vm0mLt60^H&pNW0`eYT`~1BpeQ-KLpaW@V>3|!gK~s zZr=#HL-XO@-9Qkp5a)&`YSWKWs|gBK!#yP`Bu;NGD$?yiM>RUY{IV9YI9m)ev>NcJ zj6AHij%8K`I})o=8zMGz2;NdJsk(^(+`A3I0lPNJS>`i(e{2bF&2ozFkzGT(U%UpM zOU&q6mb135xRJ3|D5i$uEYYfrbEp@d!rAm&Xld?K20h8ZQwzV67wZOzbYDNy-kA$s zZYeOd?Fn>v9Y)$^x;WwOUr1DS$IQYv=uxjPSlZfP<$Lqsv`^Xp$seljPJ^i@>Cj_H z!BOQw6cAoW#p(icXtO6oJI$rh$JubLK0eWguNc;Ux)>CQ+ zEvF6;l4>B^Z-2lu>n|~qf64*%hGF~5Qqb7^jqw<%!F}hJK%>MCPEFGgv;X@}=3mTG zlzySUI>RS{@*K3FrTDC1!$>i^hcbtrU^(Kc@*80XrH?4`}h|~w8>tU0O9(RxbFuuoV`CW@1Q!&n^T7@y|tn6`8Jl1`-7w1;SQ_V zOwuCF1iU@Ng;Fu*qB1|06Y}l@CE7X8U2BlQDN)(L_-be{2P@vepZ8ighta^+`uTq28|?}RG-YgM&E7V(H@__ zLRI)DuBoja^I`@E7pbOU$D#%J+EadNZ&@|mzr=0^2yk$8WgR;ISP!hz)M=JKg}zRk zgM!!&pPOa^cScYh%O%>w+_1QS^3FVEw1%c}rp|muA3kh?-J<%q-zbNBGx9$Aa$5~u zZ<_@pc5&!V#tZI_CJlHv=0pFTWk3sT*a?opb5Q5E81%=fvvN|NV&)6wFsEELLZjYm zB*inqW4X&HAvRAMv)hwuI*jOn_Xf0L`3-m-{v7e|>%oro!bsRWh{kv2X|-?>6g&C? zn4(N%r@Wc`FepW7vtqHSR{|p%=!xau36WjShvDf0MW#Gn1`?iqXD%e>afKq)nNbpo zgtElxgyX#+6vI!4ZCpufZ(E0KT-oO}gY6^rRifE{euDYWIym%lfEg53Vs6I`P=4Ak zk>jgmRBM*PaTk}WT-GT`z3r~0E^gV)`sp5&0~ROHvWYV|=MaisJBfa~?Ls0qVu85S zfy$LJZu}E3B$xRF#>&f?4-TF1r$G@}{ZK&bW0o;Nqp@(WCLi7*J7_;rh-!W&EtO90-ydT5=k+JtPf3Y5VC78EUd(|flB-x= z`bK2?Zy7!9M@v=OMO`W`TCCa;TO-j2<4DlehK`v3+w^1|4bBHl=r-L@Hm9=-TMbHM zZo3H_*OJBikF4ix9`K~RWM`9f&nJxX$agfh?*No{^dXmY1%$_0ZEUY1{xcXz?^-5I zUtVMldyfjjfy@wmX2VlX*WX51ap)z~tTe|S(F#mUUJI42CdSOGDWVE2%F$*fjmpn* zLRZ#pqPkO^(F)~U7XYQz;bG!M)#2HKJ%rjjSiFSr+i*YdZj^gV4(#N}0JGr}AJ^1MiG1Png8Xh)rMS~g@ zq`RpAZ#dqK?OekcwILx`GwO^_^0VyMu`1Tm!F!X+Q0LoJ*t;UTjDm7vEU63w~{1hB&U-f(QK~%lNc_j?!tyF>ncqul>}@% z1(C_GsNeVcDY2(Drbgc$qn!=_UEk_Sbb+8BnVlOo0UO zeGo%$C*ELlzL6@$1|gd7q#iK-t3bk#{rSc9l=*cbCYxKv=D!%UOo(EJ4=)51J(J{0 zwBw)OZa_qP6Brb~q3X=EA>8m9Q9quCj2C_91pm0gXeAF|+3ZhfjYAHFZaOhHYFCjh z%T?)^i9&qfM71`%guc#GF}p%1bt_;XeuQF*qAn7g#1 z`B&D$svaxcc*X{J!w5G&3g(!|rZCN~uHfvV4oW3Tl+ZR;%^s-5nVx#{!tA)406Fe) z6R&R{MR6}CQBB4{B-zjA4#Ph&!D8{`i>U;e|KbV;P)6;Acc|Y}>EzhrqogId5)&p0 zjrvTW5AG>c_Qh@B>6yu9{pE>;=4YZ%GeTZ@))T|0&ZxXsk;t802C_3_skKHDc#G$B z;@sSas{F&K;A2Tl<>*h`e6EYyx=S9vUlN#ERCEuaa5h(QpuPR2n7g;-BNC38i>i&q>Nk<|bO@m;o2oiqiK-Mc@_5 z#1pyE0N%rRXY&f2`t=^(KADNG9ooX?C66;Pvv-nG>s?I!nPT`PJ^*VzuKGXvv*eMt z_^K4UGqt}R3%iKY8YCGk#oD0uaylI>`xrD!-XoL5Mvxge4@=+Zll6AtjK2-Ty&ZV9 zddI)(lyo+Sntp8_lI(G!BQw3x%UV~n#i|FWtOI69#eP4UQCLmyYDYAV&F0F6x1-kY zEKjFX1jc3tV43GKyuSYLgse4WrfRFPo#rN(_tJ_gketm>VKvNxoL$Vz8a0?Fg7Ke- zDwIfqnC(jX#K5An1LX<8(@`a*|ZB^3oSc}YM`W^gRFy0YAeYfRTn z1+eqgU{@w0u!NB&8nCFR9wwEV4Q>Z}&w&_$Lc^~+WK*R_vp-!IBkvK%qBcOuX{ zo1rI6Z0XC%+wtg@Of;VGo@;e|fO&j>1S!4|=Iv>TCx`FEqPL#S*5 zQllL4f~jC+8si0n1vli9#=RKVZ+bo9Pl!9s2xTYh-+DI<_@*z>8mV!E&h>x+ay4sAh^d zE*PbDrm@T&_C9jKZ9AA3=Rv0TX1KSt9!LE$A~Pn=k?+0t&3@^OLcXZ?@^V=k_Qp_#R{O zb{c+|_7ShTvKlj?-%#oG3rKUC8rWR?R^7>ei(%OJUsK@;wDFk`G_C;B1jhJK{W;>K z?az>dV)SC2a87}z9FtX>ha~G`7zybhLOXo|xo<*rZ2mW};CD3}P!d5|B2D;;-XQnH zfmPJ*w0R)I31yCNVQoG@MHfwCSE5F zx{PMPclmnMGulN)edW-Jq_t4Z=CzC)elp^>)8NGWM_`qCgx-}VPC_q5l61jo#CwK1 z9kQnlDLW#{65n7p%@sg}yBENF*Iaz}Wf~si^WywRcmCfD5MM2u!8utXjCZyzCgErJ zVdvLslsVOhix1RM@8?>8SlChYd7}dwWV3J2tnZ^EOKi}M^hB&`UyJ!p4l!RVn=$k! zNX3=C9TAM2hjT$*{YGNg2PyayK&n02zHvpTblCbk&KdSh1lZxC`g*m-eIImTl z+V?dW8!mf^?vI7w#T{AryuLaV%xIt<*yv+cE+u!E^_1FbMRf5VAFv+}X!$zP2fcUD znk(cPrDMiaqts5^v!oj}SIi~yb!QlttfXoIcBV-0R}dB*dI92ZlPQ zz`H7waP>5{!|h`O*}pU)@8$+0U-D2pR@PlT4Q4n5j z41dLx@SFirJijgpF0aU8&hhynsrS)vv_KV>g-wTh`ZR95@&MXTe`T6J+{YW=>VrtN zB5sXk7)udf>RQnP;;P<_#WO=tnQ$g7UeN+OGAf{ZM;BVsG>=o2wGJ(lH^!3@0T@ZE zVD=aQ@&Vy+)hrgCX5?|?rkzBq9+~0v1q#?bDw>(B`NZ&Q&x3i0H|UIiM4Bb(Ai94J zGA}y|qM3p4YO)-Or9Oh2)Lo?2ZUxJ0Uoa=-v!S>l60JFO9<@Aj|3TwjTDF>qm z9A1(>dZ+e)yL0p_NAK=c%DYq^e-39ir?bz42u0kA(@WgrSuEdJHLj!asi;r76RYpF#Jv75NQuvMz|fW zP`J_>9luk_?)<(Y;XP%zez^^GKII6NeLx<6dvF~Y*ZMHiH{WM&FHwOrHyfxG!xNOU zT`o2M`X8*ipa733tVA*;eAtzi1abjy5(7Q9F+WqV2ghz|q=a_PG9>+4k&e)ZiXX z$*O7N@D7@JBI?LAxBkYZ(LQ9i<}~6TI|I!)Jr5R+*)x8J_LCd01Ic?=9cVfu0l(7M zA=TrqhltBiIr5;}w`` z&-p;@?sh~+yyKWBEvrat+8B<~oI$d$KO+NQb!hqd8_0&eO-#yICKSK#hBqu9lpWEe z<5LrGkJTecc_TtgXn(|o;-5(OtRxZ_a2x8!8Tx$UT&S@*KxuVmqlARZjEnYbR{MDX zvjzK6__>v2`mUi!yU9Q zlOb>yuM3dk@y+~Aenzmr`};6ZUV0FCK!E)7e#%I__op02gvrUt&1BQYebl4Q1&p)4 z0BP1N#+GYtFtf$YP-^RRvi!Xj-08jq^xYwF9~Xs!wngNURyrKi4TD!(GB|P{#gP3% z9;%&xjtKoYLWGUDEOUoLyaZL~W{q}y|5!FU$8u%`R#VJ((8nCXV`PA3-Gy_CIag5( z`kOh8-ttrjVz>(Q^4VGV0p|lQu&x8|JWF_ego|GfPvWaCnMilHJLizP9HTAU%S^*c zaCf-`9CcgF*mJ%jp9&4)*ItajcwA+6k03_e`7(|+6DJn=(d7Nc2l#Vj7(Vd*Io`9i z7VWfKKr>IB>0GfsEFC9;iUqDQKYXIm!GFhz*Onj@AoQIJVT))N=|bps<-y);pd%Hp zLutq`Ns;77y9Te~`*$9)yRb?0tXXla1u*HNwr5Dbe{6>Ug zy=c)c1K7qUjz4)eFk4gaz>G+FWVNaA+&h35mnNgj%Qj)hJLgHaQ5;D< zokJ3AYhlCH3`nid1MzPNcAL9HNL3#yTO|$;M`q)OnUh#~=653M_6~V<2av89b8=A2 zh0NMk0LoXQkghWi2j=FWS$mXW^I$rpKU_%aYwzJjADc-O{}r4#C51-X4pSrh9x+h{ zGH_mIF=@O{)8%jf63EoznTZPXB<+WIg>xc#I`0qj@`bu+fkW5P7RUKyuadr*m%2Xo zQFf(=t3RUgHZyhyLYHZ6EyFk4K4J9Oj0{fyfSeE2qi+Rju-R3X;Yjy^&g5s3LRXzQp|49*mFo-^Y6lKOvizAfvm(mF413e$=7;_qdacoLBfsLl&46#hP2SsSJE-$#)>Ef)3(?WAm+)lOBkFaJ zH$8IjJ(!j{Qs;)R;c+uV{BP$fs{QIVs-sF1j23oMbC$5XWSs=hjsdfw1 zUTKWE8!gZ&k$G^?D<04LwFHVv#Ha@YEDwHaHVN7Dj3k;YX0j?4W7!NRqOf@Z`SKx* z1o{sUuUsj}Zks{YCU;^OJc}uXXigDp?8dpV>-0V(rke z0B@pd9!??(3Xty!Un1S+1Z{5SWTs0HYTj#wr?cI!P|Y>O`Q)|g)2eT9#F#S6c}gJ? z%TI#GnO-K#Xa#&lresd^8m4=+i~8%0aqRblcvGl3W4o+{Osg0t)ZIwlcfWniwjWYZ z`_qYLVv@*mO)l{cFhZx+t8gY`Zg9hNAET}479{SJg

f(C}N&;T=n7EF*HU>7PwZ z*0E)X$SmSK4j43JHeRTHyJj5KJ=qM!78&5z7Yk#Yc_4noo=A@VAooA*K+4fSu#oL1 zA}V7Ke9w%~p73yD&F%s!{ctC?ve#Lj%}Sonv>V{5Ek&lCy2l)tFGn&S2r)koCy+&K z9kh`V;J$SyEyt`M&JXefBjn4Dw*`AA_=Ia;El!+m{eUwPP ztFMECgRLOvJB|k@tGGks+T^fzD7hZtNIjH|r0jak%MqLw-Wm*xn#NgcqSXZDQ3|{$7R^i8*EqrE;NE zM+_qPWq4h7r+Fu%g?aPD4xwX$44k?Ui~O6L?x-Q96;_Z)P3yCrdzM?698JTg1u200v4M`q>Apj)@6p;Lvm;G|^*!=k~cV={yY zo?J$b=(m#X=L*R?qrGHO;sW$P_W&kWAD_wzMYUBGB=h(y54mhB}#yBf*6?d--pPhrvG{W{P+G9*na*BF%`=o4tv-g`-@A-gw`dDC(jWb zwo};sj0XqSH8Iq*N6diK3dBNOIMo{A=!l3pxxi*k_PB2(iZxfbKB|G_mS!DuCoPHN zWh{wOFxwl@=s>j_R^y>q6ZAU95r0(t!Au{1&BWe!;=T*Ii+qpBfp6tnX5X=OhZo0Laj2e;fx7U+r5qPKYbhD9kGJ#Cssm^t_E@cH65uRX(ko7;&IiETMYMB z3v$r7MkS*p5@9+@9&b8=O)koi;?P%2{J0b6>lc1*)mqW&c*9i0nMy~C1S(LNKATsJ zYa>X$hNyd`GoIN`@xxzX$ln!Xt~%>+m~`NUUHXv!LLOAFJiuOkd+^EHAnbfM6@UKf zkCfK9A@xyflq4(!8(dXL=dxP(v-Uhr@CYG0=sjeSWd$?i0Q+UF4J9J(0+dxCKiydJ z6AdIOaZ5JLLIL~E;xHHXj$yJAvaHpZxW>OsOVK#fX;w=X?xKm;+-By}F&o@<$(s~j zmm{7N68K)oQnFV>gNzOw!(&HoP!SjZqMh&8k#)C1HZG4E!%4nI zknfy@9<5$V3fA3ac6lwsx6?w_Is+wk7a|^)oiWE_Z$gOL+mFnOBNR+U?1ZxR zYbY_=mMQE|0Dc!Qvd;28eqQHD=0Dkt<9x^P+Z9TzUuQR%a06tS?S>BB zxl0=N++yB4Bw?>|b<8fZf;OqYBXbY20sn?V-oYG&MGa50tJ8(UIcMubL zOBMbnu1L--PsQte9^w-_H{ps;e>hemf52{mD+zl&#BLUx!)M4zMKGeD~Z;5#%Mf>W1oYsSbFL{ zNiDp9_ry#`e;3qo7fECM^Z5fHF zM=&|vSIP+a?q|F9&WxG=L8zF&9DeK$LBG$G6BG0YZzI34M)hYVe?}iJIJ}Z1l=_hH zDNCH%oPzhI4xzw!J@~ul3x$+FqNoIg|{I_dkV(#!-L|GN= z{`tY11v}nbghpHJ=pLIb=|E&T25^o&WjvqjQ3!(`!B#+rU&mhejoeD zt3zdU3_QGg8RQNupr;S?L)h?6xRg>w~Cmof6r$#h$1oTxIwR(rNdmjdZtqBfipZ4*H>y+%Re`fsJ=V(O(rsdhfpoprkK@AKMcR(h~;zsCIVe z{w$qh#RIbq@;G~<6I{-%f$XY6+8E>_k$R_%DB z9`!A~!|Vtu0)?9^Sw@06y&-B=IeVf29M-f0<8b9Jonmt(2mZw{gJU=pg(X zx{G%mC?pA$rsQx>DLyq;O7U&+g&8BlASLu3j>==0+?fTnIuF6fa1+fx;}o5FG!AMF z`r&hfI4xw_gs#+OViEme@K6v${9%P?>+UI}v2ZujS8#*fZ+OQIX_-y#IP1dj{YfNg zauCGL;vwyBHk2+~M0Z4#!%;HD?zYO&fftHkSMUI?J$i^ImFk8}Q5xeU(jNGdNCUwxj9-<8ZHDfPCv=2~tqHXAs%0J%LcT zAZ$3b7+uUZXPRx6qtp-yGI7$AD(sknUR8X?-y9n7$Ur3QU-}C!W+UL(Cu1%Bm1seR z0Mw6mQi7ERkhC`xMTjq9vd`93t?boA>x>vAWLfQ63rU@&4ylfw!v9p2k;7&cW~oC5ZA@>bZ3icj{`WGp`o~3dc-K1$xm-h~ zp$AC#X;<)BG6SwI*Mg3H)o9-=cTguA;Nl}S$laIB(W`g{4QBg^N&9^AjM_*p&EzmQ z9}hvuC20`)8Gwwv8j!DO2=YHnq8OzcEHuY<@@Q4@a@QmNE)dr1XxoQYxXE%yxNwsCw@N0l^$3v&4d=7R|sX zc+*Mq^ifV@&|O^ISw+GlW~0~7;<0_`R^r3zC1%?k;aofgD$BVj?@c!x=njGHVbSPo z1IrbiHePksYy~uQ?!=}S_^{Q5dSto3kj(1gQt~U8qlKCK$-R_UlykHhdHGt5gz;k{ z@N6%sDO^ATH+4`mgG0eue<{muUjsTr-_gTn2Co}APi^|dW^lKtgEHj>YxY*7p#cZ{ zEHE5a6wf1jJ*y4v0>m6@KuKUK3+wT|9P_h=Adx;>v zd8@e@JM<{8`Zj!S$yKy4-k(&|>63a11cQ}#Az}7xX!dBv=yELUp>HR@JVxMa^hY?K z*u>No2&3ewC8%?K9r{VK;Al<(8gp=f`_?qt-glb1@sMGz#I`UhbDI!jyBk*DSpq|I z#IWqlPG;_IUAR2U4ent-N^eymk~>{UHF~w8yqDJ)9b+}p`XQ40!{8$^UYy116iX4K z!)~H@U16?iJ|dA{zfy&>o*`R@x43|1V}E)%iWZ+?eH*R?k0+alfA$uUhviz}SE&aD z&C)z0^?M}C?K|9Yqmf{O9DMyM1d}U&vRNM`sJoPfKc~NCy`>s%f)?R?UCxJ28_1*E zLOL)Z6Nko+`%@x!Rp3UM7|e8H83W0(u*};R4L6oR&{=-^gv%CWV&%o`3JOC(vcgnR zJ*(ZmuScak9iaGaabS8d)t_y_=oI)c%fIHp+_#RzIob-6vkFm{8H2xcxRDF)AJO8N z8F1ZGl}6L25mhaH;`;9g%=!=l(-!T3sHES-->`z5q}G$NJ_8&9bBN7{eB9)69_nTL z(QS)Y%>Q~{|D!+Gg?rI_x*_zu0AX|geX;ZzojUX|SAbU4GltL9ITCX%h$L)v!AX+& zl;QU8Ort<9{(0^mrYg%}(m8~@)Z2#ImfO;=?PSn9!E7p3ei?MFeM4nFV9C9S&D4rj zizx4gOy*RyK0MLKcy5RxHrXM-{ru`W9j6>kt3}>`S820p?JjTpZOKFEnE9O3bX<;l z`#GBH`dAzPIHLxVriWmIp(O42rv+YKJxh1pR!5ffA5^!xk+LH@IOTDdF*B70ZE{Ai zG$RBVs0?B)_XVgms*L&&W`S-c%puAKa;T%w0Iu)L|`7Vk?$W{R)Y+r)nlO;6gf;w}jd4xJ(e;!F4oykbwnFB}ec!NF5n*Euq z5Ap;PoiE{7A^M@2YSk|krqRPu-!>okX0Arh_pC)D%_^v^y9jvgSMg){OCS`T2AWs= zKt0ZtzHn59mR;M9(hkrx5{ss1D9xhB_s#)D%|K1x5UgGoJ`3z#MB~h_qv-TMIr^or zFVsrxhf>c-Y+aK@`cqpucYi6Oy59j%W*!aq_pli}E)R2Jrqjo#q`_e*2My)V1Fy54 zuHHXAW`PJ5Wi)Jx8iI8c*Hk=$jfOwqj}k^52Q*jkOz80Z7s9ubS!@2 zydDJE|KHcYd@&mv6Colab1Cz@)&hHKM0;5J^;n-zn^RAQO8QE34 zqav#C%O-lpypIrhbr(3FTux_Ro<}UWA)xAg6P0f)=YhXKW8UQ~mYz=-5lv+Z_1}ooOyegw4ml z)cgadBps;VHC0Gee2i-9o5?b6mXe4ivb-S5m1DT|9Fw{tl{^dMLjkKkqY>MAgfG&J zEY8Rz^Sqs4eWekMHf}%{*3F=eKmtWi)Z<~LK(MKqkG$8Uq2*qhB=KfBQBLh8iQDz* z{Kx{%WKsh&)INc4{&t`{GiK9b8_TE{hnrEH&pp@`SP8e>I+^b+w$!(XZhU514p?YF(8TS$y~@s1Snc^E1%5Z`RP%sYlOJUrWc!ZpJ3styITf zerP`0LeB30%jVV%X_?h?LD43+dXe!Q(j9XLdaOHPo5UL!54(*|EuI5gv-Ihvcejv% z#0sQW5rO%KZE4dZb7&!ld9;4@BqRl7Fc?~Dn?3ogV{dBC-8f(MXPEG(9dr_Lo14;L2GC+tdEN2Y?%l_ z-@^_P-n}qfx9S5{Hv7OdDa0^6lck)NRj2X1=O@usbQ0-o+m14lgV6!;CUj`_YG^qY z2_gQQ;MDmhcx3kjZTXpkEI*24-GLUOmX`>53y$)X66EO3KLu%<>=uv~NF=u+gGf@* zZ}j`Zbue$#15vFen9V*sy-mvr`ZI=|BK9NCQ3cc_s!d=1Hy_zvDnB^|$yeRH9Sc7I1p)t%+IO zKHB_vH`odHL7;;W#BI!{iV7OZ>jV5yH`xLaZXZzds*51&e;id@$wANMn$Zcr!|)o6 zfoGBh9c@eCo02H6T&M^<s929 z|G9;v9|uCs93#@Pr-1$&7Xu5{dyvenZ0ASw9{ir8L!T)ufKxLcLddQnkUbL#d^f`2 zeE)saWE@TUkM%O{5|aPbfBwh+$|w+lugPqu)}D`c{kDnwsALg5U*HY`)NWGo^fFOt zlEy3FQJ_)(n>iX}ObOi$!Y6mM?F2Jc1Y@_oXP6-~9=^TuI5E8T5Y3W(Lb&pAS09XyJk7Zqcz(rC1Th(WU`jh!u`aPTe}XzCN^ znB{bV^o2N-mo&!F@mm4bx!2)NgB_hDX-@pRgczq=QCRHcJ+ML){e68MIC+`DU$-*G zF|v&DH&$W4>pHN(ss%}i2=V}0-pA>LqcEW+t6`(fr1vS830X>o@;kM@zrqx*rJdF+! zA!#|(t_pbg;UGvS@MCG5iRSxVgV(hQ@c*cK@30=@HvYS*rJaU`5=|nb@AWw^k|Zmc z4H7DfLWnZc(4whQX=jA8N`0@-c@>G0Q7MG%gvbin@m%-){O;#?et-0*1f;wu#RN{}L72#Nf9gHPG9( z91I4xit0jdlD4<8!riL`lG2WFvqeKdx3UblNt+;Qe+6d}DGT=$6;S_rD=i96gd^>} z^q0{BDzaPy_tz{Usc&UbPUAM&^MS7tZyUz#a?9lWmLCN5IV})ktPa7KI?2WNQ|X>@ z&qXg>mg49Y>ZqIi7tWPQ0P&v;Zz}FFFDvEQCBJUMz0VWbD@!c-#?_~o$Nsa3pZR3Q zWk@B+ZL5Gig|YB&R{_qmF&8_&Sps*L`NE*<3#qhg9Fg92tlIm4CcLj02M4a{<9i24To zR&0j$gbHS;lsls^uh7mdV;PKfJ^^nQ#E_%b@}&J@BtA9GQxnz2O^J6H<>Uk0jE$9C=CUYoESkpNx?E3_^?Y$(!aMHo*%9nD*{1dz(RU<08_|u?b63%Oqno)i zdZ*y!q;O7r#FH3ypXHw3lfs87hj3fF8kzYd7q`U8W7DLgAT?N(*?;~3SH5fk)&$Jt zUhW-33zk&V95H1&RuF!FAdN469TrDR)Kh~Yn`p{-Wg;6C0w;!-lFvRKV3*71o^-Z9Y(mG1^W{#6vfT2$5h1Tq9^;`&MG3H<@0!E@DzFUQxe+ zw=`LDB8j^>m)QC3Ci61O$qx1k_4v?2Z`H+-8G9w!ua5m>(BI*(+wwWBlQ_#ARjXz$ zD;|YqaoJ$Hd?o1_rvg@sgSjQ&D@kbFZQ8ecHq#nr17Y883oPR@csDN(I)3{>__sFl zHzNZpv>oZe{sOu&{vdwQ9)ja$jD@cb>uB@%{WN^50(rQtoGyzxCF-PAC8*+CK7RTp*oU<+YWBz=Y>9yoQKf)z>b~R-9suv z&GJVlS+T9ckpx@?Rh!5c4QaL1h){D*Km{Jb&(lV=(;{H(*I%3B}8BQ;>&Hwn_s zZx*kdwvSO*;)%wCTTww)j_5^wWPXMT%t`Y^i0r;ZO8?m6^>yLke&IZp_h+Mts~rvX zb>aqOJIF`FK(ai1wOCWIM{SnA#tfu265?`D(C4PKf6Tz zhn&QtgI5z{$pu8$^eOQWvh@9Su2?pG6n{O&i$Bs=ODZp?qS4gL%%v6E>6EKm#6(k* z*WdIMO}dg0rsrd9)OqIZpRVfobtmYH>=N3yXA!pU5V}CW4+huE>7+!LCjoQgMFaYq z@k)mk1>as+;k6sKdJCO%D;JO#jl8HM>IZrCSm2_)J;028+et1Rk-^csn{fE833y%5 z<>D=0<4N-_YBpSj9kgefSoXyLbjxN7U3MN+CwCV-`p3b%t$X3M&2R4apaHJ6ZV&lZ zYD#9RRntJx8Z_4*g1ObkDE{nU6{IkXIvjH)xnnrdVBzjHaq2p-KJ$dcsJ;}P*%>M5 zKb?#!TMUCHc;a-4Qql54X|ibX6>fcg8!~%32z4ojDTls7nbb8pWw;t2yL>cMjOB^x z&q)H4Wi36C5e5H*Zny(`ELmJw2GhqqfKuliRP}9!@fx}80p|dA@sR{NW0j7e^}oev zQ;x*!lOc21C;!Wz{&)V#TA&G8L5^@UO$!cpk z50{o4B!-*M5xWO_m`SWU&HwC3N1dgt4oC{w!w6`J?jr#&yBY0~K}?H@qqypc9o^EZ z%J~V|=5OzYQrBQgrk_Wq-en~Dqvl?9rR(R!M1kehI_iFvqNWC@@_WhCGlQY{-Brfm zQ7$(;IgR;JT1w-$h7qV$BZhiw=w1QVH`o zXAjr1J)L`f<2EyN`Z`c8(t|x+@gnnw*O<{qlOdbyCm#omCFv2>qUlM|OmU1UP1oKd zj;YP!6q*+?ax;4wtI+e@8LM>A_TI|Oa!{s%sZuo18_5~FTW~TthsEh`@!}n73(dyl}bA+~Jzj2$eV+sjmq-v9}+YhjT4YHs+8r6i-to5VJ{ z!4$roWPE9_mcH*y4KlydYx`=+Bx@()G*q9weq_ff*S#RQtC!QlehczuaRiZf*I)oV(M^NXSc)GoHsuu3ta7eaHqB zW}G6AOVY$Y6w_cZbC;RBH&g5X~H164`R66pZEGJ%qxL%w+HK;X2#TaCdXRa?yTYIqSw8B9U!C66F^X|4|jpB5PNXWJI9d z3tQywRawwHop^BCCk->|Z`5bvaFN0mN_8E-Q-2nwD z@l5UKQZmWU0Ouapt=bjREY5qdS@iGI9-0>TP2^a_lAxca>;^q8&>4Rgwv6%<`&3L3 z51G`){rkFvS$pF>QE<{FMH8-qvw4P{*w2zNPhSJ6+YBJ&rUB8dFXMUwMliFE=n}Dt zG&q-TXO5i`-k#9Gu>X@XO&ps6m|O-Hlh?u8={w+4c!g+cm>(lKb_cU6x{9_LX_EWj zCosB7`{>4Lb;RdlJ)LkF>D^*m`0&LB<~PW3jrQ{Na9kAkYPk}t{!y0=FrN(#d?iz5 zJO%vO1ONa2Ly2D%qp*MRN$MzMCwAR=PV<);GaE17<0e%Jdu>4k91{AXb~WY`sTWJ= zDaTFZz13lQbqb}!yPKjowZd(KH94=c#!nm$qmLcX_WRr{O#o^Ydi!U6CXm(Rvo6Oy1C=(fM>u$ayM# zGne?N%!0AA$6!wEb9#975a#B`KKy*(7uNWz^V^&1xEaH%><;MbafJotRA=9PZu;mU zu(~}RKbObgYZE2ZvQ5Dg<&u1tX)AWVkH_vJb@*~*7<2Em&>>iUhB1FL11g`npzN@* zOoUjGb4qFAEE0cGacv0H4)MnP1KW{^hjXikC2$$KZ^&<{CnPXJ4Y@cY8XkX(eC*y% zW+$BxN#D>E?lDFXJTsH=(jaiJ;hAX3x65Sqj{~4`wnWrw`nU*bnXou!s}R4rtJu)9q*4>iB;8cI~OqzF1uh^uN#&RT?yMF z5@48>JvqC3GrFHUh>q!vblhWooG~g212&GPd$mkylaDh`X>ZVqhyMv$h|6gnv5$?bY+B`hjJHIQWy@m({^C)eX>`)pFmXRP z1Wx?kz}47HBl-6G7EROOa!3L9N>%Wx-n%O- zi@CH&y@wQh&g70y^COiz>#9>+dudCX8dUKqQnAaNsSMMn*S$=sykj_B zCS^<;URq)Hg+LBt=VEA^H0yRX4croLU^`UcHU~+*{LOOCwX+7i7XJhS(b4;uDTB=iZ5}W!~;?~pMT)$)<>DppWPTyNe9l0X>QvF#x5Z=cbm&?M` z$eU0lXjL&@!)Rz{Cw(2O25Zbys7q-qQ@A?%u|n87dDd5UaIuitj%=JKooSzo~Ap~mZHS(!87>b1u( z`ll&YnC+)WZM3+|lkuRHI|kw;;)vX*C_48`2Fg}nr`OI1{%pY)a;L-{R6XxNZR;-h zx%4P!xkZxhe1DnFN?!sm2DxKOx-g6UafOb1cSuQt;14)q2%ww+OON?7Ht)pr?yM(_ z(o{3nWKSe%NJ`>PWv!=Ni5od}wwJ!O|3;ICWOE^)&e#_l;m7mIWJ>d4vQ+Si$`#09 zaaAlq{k>!cb;g~~kWL>P%#F?5Pk(D%h3>dcsPBJF2fQR{`vV!gqS+yOGTH*OANF&F zbB4j4Z7b;8>=D?wXDsV=%@P`aE3n4m1gu`333J^Z6SaU;2#}fqrOVQ2>yx1*LBbv9 z%Mh-vV;`ewEojyk4w7<(Dw^xHikh^jR`-rp#o0mWu=JuCtL}7>wmwLu@h(MJp_58q zp_oWnM?u4`Ec(p-Iy~rF1p&?G7`{}AT3uDZD}!`M_LFg-DjyE&YZMrtr z)u3T(G)Z5Pg&EnOaH{DC-2Q`rOiVPg-P;g2S8y%W79}XB!Me3_cuQvvjv7A>oe)7I zj^R#uS5=p|?IlC6TqZ5?(KKa~Gm8G~r8}P-LhqcnbmlBeu6a=xv)1jLaAUknze`Kv z)%M@i@Wp)c&G#0SelZr8{@#XKVVg1MTst~X)W$UdvTUxv+Li%5sQ7x5T(kD&c8s;h zodrp#|1XgGih@w1O_BGwx`K(S*Jm0xEvGR>k|;WGoi?WVQ!|;pv}4RTTyTCNH@1U- z+uT&xO8iL8WH;JacZGZz(#5o&*N0ob_6a#kQ?Riu1I+`2A!xo0dR#m~PMb_6dqzZq zdqpDMs{D+zme)tFa1sjAj5uu#rKy*flGR6c!?Wa?r+$I~u5XlO?N=vjkSR z7Qw6ZVp5u!#YEhSMq@iA;pQeV87GW@NBW1DShZ1X#lAdt?@cKpZ+!(eo(+H&aUM#I zdL;C*?SuVxdBpzSUcB&pJI>Y?nCr3?)Xe=j-F-$De-wL=35#}PTb~c|S2v<+KnD5T zHAU!$KS%S{Z@?>uelvzl6^`9qL({PcURv*8k?|1bab zpZzOh)rCj&?%IP~)yyQEvzhQ$Izjd&Sil+@fphr&Ie(fvf2%^hp#r zkjCB-bSaI)s@k98;`6<9eXuJdx42OJe6lJn{Cb|AUY(EX9SSf`aRH4tROJ3>J*6uw ztWep$-fqLm9x;DBiztef`7>i2iO13~oZ0n=^c7tq7Y72dKKwP6S6)jxv;C;#L6-LB zZ{Z9xRM0YE3pqP`D;~MoiKVguSlsD?FY=$$dZp@(#CagHSu$nFbg=mm#YxN6TWs+2)QeQ#ed z;ScM%OQ{#>*8J6YSvdxeUDQIddJ{Nj*`m%=5#Q;>bI%ze@1bQ(Oag~O_Jv<~#B?}+ zVfqO{x3$EC5mWi3+F!J_tO_#^d_(7}ZKC2+9~kK*HQo;$_$h(L2niRf3}LN>G}2z$ROMC!S+lMj&yG}PQ!7IG-5dXcr>8Fe z?w>Xz-KfD^t31Mj>_|@f;uBI@>VWO}(}nzrm>SI8OYJv|M(;HOr`78mhThwQ+Cvub z5(^9Hx=DpJD?$kk2TtIS&>p;V-GtY_BaK%7%wfmw<8XTDe!EHUW>Wv`N~)9>Pg9Pk zaZ$<3nOD1#Xry;8`6SY!n@WC>2i^Nc>DhId9ebKPbn_;SzjuLb8JfuaWY=PPff6o_ z(FU3F4>WRvu=(=m*Y`1G+yvgeHuxZFE_^TFXd70;4wPY|Ad{~mr|0L+RaEtMes%~Z!v`) zL&u~9oWTY%aaxVkc1u0Hu=Sv09(|#?Uz#{q#YKEw=}11gcm{dgS|HBrs-%a%YBJ+R z3vokND!%IyYW%+6Vq%o)P-JC~7n$jpWgrb__Xp5DyP_~m zE$~%s|*eUf*}r(65NvO>D+4!E+UbIk@s!Ew`zmfvyZ_ zqCtx9?fyn>rcQ6exu!Fl=>EC3C?By1$Tb!CRQHkwDm&uVG+}qoc15v|9RFy2CYr3% z&G1N zTV)wenlJp04S5LHOE7-E1|7Bd0QP;&#5d21bgHi} z8QqDf)|mjLWBoCTYM8$Z5e*+h$u4mlCm%h-{r<1aNuGM zGDLxv3#iIvMO^wY3f~^m=Jozufpr`0amBPP>}l5xm=|>$2Sz-?DvvIdTKXQ-6*dUp z1S{^~JF&><<$LaN<0>v7;1d?F%!6b9TIkOY=`>A-RaQD|S!o`;zaU3Xyv#vuP7m)G#o>)x1owDZ@)-_~;r`+MY(b#`8+^cl zp7lwlF0;3bD)Y?vSRHr%pQ&Wce{`?^^Aq6}uz(YDtD+)>;ohL`$nv3(l zJ`(!R_Cn}g3HEPq6rS89##0tupc^eiN}sP}YHg=;m8vEvXU&r#v5#rp&KYc};tLHYxlo5Hy2|h&eH7D)&fcbYSRxr+{9U=qGzmKY zwmS3n+#E89S0O*6cQD1z`$RNE=qcUnz`JewLY}u3GB@XqfUlodA*mO3`l#Pd7q_Q5pkL5=6010v zq#yc5ZV7$$jw;6i_m5^>YOa7y|4PitB@n$PfsAyjhtzBpRyAfT+3}B|i?c%UnY05w z{2+^iJR{Nh_8NNB*NNDRvgk{pU#k~Gxt-NIFw(J}SdWY_psRYwsDtKGtpk z{rp+vt)(7mJAaU_4@eY1K{<8}9!=!|Dj3SZaHg2dVlaK8CH$<_!X3mf!^e$8-lGJgq^nf-_+t{6)0 zL~kbsY4#+4D#z@z4ubTP9w5J5j`f*?$R9#?_PlFkhLBxOc0bGA@=swZ5@tb< zVmJ6UWDw^2Z$??Fg)SbyNqo9i9$$HP(n=wl9Q7g^mUyJXkM#`8T)hoWmMZMHi$7Vr zWxrV2&@8q(N5no~pUy2idyI7U{bHWD-H-YQJ6;bw`D!VE^AFZ+v&|h(}*zs%-)0@>x?g)F=$O?IC_5B9j z7aATVQ%w=prqN|9x6q~)5oAf3DRHoRPv+)!6TLOrG{rTLNgH>RW|o~4 zU0v=^`ukTg*-Qlu-+QBa?K%U{nQnv?BQ0qC=zrwtRUJNlVj#|aQina-`WSS<5er4{ zXj8uqleE(d5sLT6=gpgwC|H3w9&vEyHVeL--%Q zn*6>R#QIWo__=TkZM>%iZ=w{rNgXO!K4L6>P8E~#Q*Pwzj)}qz>K^(I{znU&!>Glj z2h|;)H;|#WiF9|%6?(d5D-<3Y!j9U~$8;}$L+y-HiTq)A)JwM`ibJm9*u~?xB^#~_ z{1gjPzGxf06Mq*nrH$;0Z%^YFJYK+$o7IE%b5oh7p)=?)IELSMxJ(Qs9>h=b3CCC! zLTB+urd@FrnU zf$`IK@qT4;y!mEr{4i%QJ_zo>Ka->|Ch#}S-~0f*{hx`vhFGztcQRpb%p39^bOg;q zjmWK<2~MG>;fDNgI__$P=<|gTjPy^$Bh@1?XCPPfQP+sNZM{JYuh-L`C5t`n>}jDssUfW-A6MYvGfy8zRd|gLwVH({N}} zF4H?}JF3j(nEvCZx$w*?y09?BPC22Tt5p6-%X4$+9W@~jI?jxCc+bN-Qy3grvz#$H z@B}BSHDKVHI3&}}aFNR=nm_+HF)Y4WebnAbY&7aR{+!iGSKfXFBY(NT_L`T3IVSxGN)+R`xEA@4)f`t#}ciOSS2XBgJc(S+`aZ<$*!4*c)_6Y-xJsL5B4W{prK zY_S^z$Bl=@MlP^!tU6n`cM$tvqzi=AkA+4Be=6B#N+0R{BbO8t$OP3vpslLJ+_5gB zs+S94${20d{#hyK7kiyKlro4(D4oe{7+S-u&X^^V89o_mq9#MuTFSkgFWjHq-jFvE zazs|xKZ>VaB6rs*L+`t#Fm;kUXp7w8WuGVW^U$S<16lt>TUAS7^2}H8y?-Mp=>1@( zu(`x;u_D!T)PXt)b@E3tM{GRI5^WdWW%??g(~TN>)bI5wM)c(%y|~~B>3`Kmj#W0> z$?*HJi+iy2St)_#p3Dskcn zGO2Rt(Oo6z_m13b4_D4*%!&FPMZ{zoE|;+@#lA(#e{c-$~Yu6tZ7`Hd$jp1iq)hZORUS z#=XDkrMN%LkH!7Wl^65Gz3X-8_r71m6aI1?541T<9}AiK7s-ZUj-=AlldOJY1iopX zn5f!QWP!W^{X2Lj1QvCOE!zv2*#e6?A$T8I)_;fjm{~~`yEV9BN8QmsYZgwZ1HmsV ziTMwA6MYwX!rmPML+EjkyZQ}w+u9KAzvDpGNtrQx+zHM-?;$Z?=x=h9f|-HR5V9tn ztc?0c-o9N3o!`@mf2HvMy|#!rEzBowhb1vn_*gF);Y07QV41t2Iz&$+ozR5mB)&)m zY?duzyex94RGSqUp0|seb7!BZcZn6*PKxkKCxF_1y-I?{yMbkX zEuFqdiHUyujynC+Lch-akeMdOc5gmJb3CNr;_VjLXy8i=rn!>3pXJ=m*q^k*;UD#1 zqDxcDM=^y12e=#C6e-2)xepZ24G=thhuJ^QW$dw~F0jEuVCnq0N!G0K;o8!B7@bF{V$TK%QorOYv%%Mv z4%6O5)Lt}!K~0p<0~-&;epV1Ozl7Atq|ykLeqla~1afgG*lT9NhTXs6Z&fSxm1}3p zUZ^pPkMN{KR-K!1`iWi1^ucWX7IU^K)|8#sH4T?a8Bh;{$8@HbGPtabghh>~N$nAr ziCYYxbB-al)UIVKxwEO9EIa89E*A$u^Yt4rwKg0eb`dxoTM8>0(Si1@bGL zSy_)U*t763ZLKR1fAOn>qgEPlGWhp@{6GKOza+TS3>YxAC%MxOK)-`pgjfkPYu1TUyq#f-alLoREs^PKwd7xw8 zlMA!=!lZr2VSMIj=#j1xt*%s|my%zT{2zS;|HRXglgr`TS5uOY0!PYc9aH?TgPwE! z%(QwR#WX>;tC=goem^;1U_D&qRL0kn%1chLzFUSHmY|J8ZzaQ)v}G{0Apn`W!{9pe z9=Nt2XEx3Iz;tcP=gMA)nA^M7K+Wn)K>l0>$Fx=wq$~rRN<1-a7{iQXr6FUaEPPT6 z1Vc?ds*y36ROwFOURqD3AG9pQhOQU!!Rsx^wz<*DCvtT8*b8VGSAcGvB@pZUp38UJ zP8T(upq4-5L>AqY>^^Y}UMy2!7Ffy9Wk+9wYNs_@wr>V4wJQc&Cu{OVVje4HpNI46 zPSDmtlKeS|99UBG5VoE|GQT(rJLfm!=v(8cLt`19?1?7vAAbRoX3uPZ#+ zlLT+=tbl7!q$&;N)jn%xnpz)65d+h_Eh3qp25e zX=%U@x(wo`NO9u};<#-ir$JfXB%HqT9>%-NVz283;ijsI3wPb3O?Q^Vi|nI*Ge>DUDNla|~Ae842TjKT*v)>tI`*2>d?WX9AqEsl(hv zZnxihXz=^VsVJTy0&IiKGJM3@>6C(Lt-y+KjH1=q#wZ!oPNt4hC2fC~3wf?0j1K+- z8-y;$cP_T@C3P)<b4}4+31@jp z|JVD|E$+o%#RqZA_=TL@D;3s6J&F5$x&+m~To$w&L(uXvh3n(0Mg9GssjY;t_baQy zr_2m`c1RFpw6CV`7c~(D`zK^}P5}39w>C~U@8ANSdcleTH;5Z2=;2)lanK{dmpoL9 z7|yBXT!v{tVdP8tO@0g(E@?(VoyWM!kEDL@5ib1G6j&LsQ=DDsf|m8`aPY`zobksB zTL%`vsO|g7o4_jaI&(77ipfLE`;kO_(JDHod?q%1PT=m@D5GQ1VVpTtk&i$59(B`S zF~w1_oVv$Y(eGvXqJqCuAWn8B)tPYw53Wz5`U`tNQ9c3x!8oeFNsW8^UYBVZ_=K@r zO!+YfU(s34xny|fLLpZ!L*2Fiay3B|DlGa)SydukP{DEwr`k}fZ6B65HS(wB$+}R&3V+LuXwguSIB2sv}-pWhii9eiPFnufVDA1e{kN z4LI^Uu?R|oM!OL5x#c4KAo1X!b9NZxl|2yZ^IuGSejvLM5Gral0)yE=xe<+I4$%~UY4|=QkP`m zY+i&YMmd7MxxWl|4wb~7>NQvz{g`|8<|H*6`GEwh2wnZxG%@S5lE4rcLLYB6#B`rO zH15{{Jh--)aR|9V@A+KE)TFcMAhnau$!4jIve3hAD2to-OR)h-G2BGALG0P|GMcfG zLaSIF_RNcho(*!4xp@p)E_Wt_9Oh7Wp@(ABOa_{_r+~tkAd-9Q62uJK1ret_sdJJU zy%C^>oBMy$y1lFEuzjoPiX-v#_WBxbqH7T?Q`Dwgz9&+f&wrWO`+t*XeX(dCA*M#d z!Wd?J9v7oA1V_Z&AxnnX;gi!Nu-1Dn>Un8lOZYR!?_C2qGG#pkNQ?yEfbi-(2P-;l zaU`iLpMly}E75hzBy^pi!95}bN+%)Ay{^K~Q_iF`ER0@@OQuD|&7^3lCCW^b!J0#{ zAiY2d25t=j|1CW*^WhKC$t)S{zP^i#bo3@&j;V}zUTAgSwK%%kC;*jSxl+E(7kd%{ zq24bQZ-pMmJ!7?my`z|f1)1>~#agV>$T~2bcn}QZc(L<8PjQ%GgZQSyOQIRRhO8f@ zNGfu=nb&88-Z@P(@N1K#CzdnRbbBW1kB!7RO$$N(+#jLucqpthuOV-i9wJK>o?y!9QF_IKzcnq;t%D?pvcGn`c`=%T?#%{=~C_uDy${Boes3?l10hmB%or61-|G z$vHes#h?IzYcw_l)^NV?`BVsHa$E4OLzvm^9f_9G%q zj*$s13$V0zC>vpv2m3b8rD0NL_%L6It*{=||G(aV3`AnjVMG z$XICl?Tklf3G-g%R?Pq1i*NHf@#@qrQL@JtY)f>6C(~5f6&rT4lc#jRb59r8z25@W zn_lDC$2nBOPMX&|djk(ftKgl29-3Zo3j=L5VbW|dv3(+qj7ADxHjjoJy?O{VcE#r_ zKcoG$_b3>B@ZX`suxyGxJ6~fX+)FYXPnR*U4pf(E2-$|p?#vfQRH4Dt!3*htX zKs%QLHLlF4omLrIF!ve@Xz2T)V%;{v$ppJmN%QFyb%o}b}<5J%0J%PAEX(y<3m<3I5f$Z<2{AGyB8mUrLip1-H)cKHl& zov@ussS9sIPYK}!pT5SDVm$D0BG!+$B%6*dr#YUgMO(@Fxt&nsc> ztK-BewVSl{y&^G3gu8R!Gg1KgsNXOSmviqJm2quUGDHWA)?b)dDDW}T4bovCwu$U} z874aM(;pYsF2i9}BHU9lK+iQ!K>osZIx|gzZIu>SUqhy_#lBVoV>gh0GgiN4Z$^y>S>+fD?Y z$ug|+!ZUCpzy-QCe4#tkB53dFMKtGb8Em1CxQ6W$;8JiUgtO^5VsR$QteeHZRLR8U zvo4`7`-^%`^Wst~PSe0zZFYBfjp(b+H9Rx3h7R%`Dd^!_p{y|!?|rnQHxz?Frc)VA z3PZWF4i0(~N1}Q6FurQ0HLs#K35PZ6@bmVK$2B23LdWJ<&9?HdyR z<3IfW_OGOwGw`zgecbUk6I^c(A|-upoN?V`c=ucx_ioX}5V>#MhG$aX80-&SR|m-2 z#u+f5h7h0S$;9Md1KIvVmU()5Jb&d}kI2$s23b0wV7H-uKA0LaP-{7jEG|DRdc9#A z+?~?_Y+4EV{G3v^z%=l={S-ZZ#F6tlNo=Q~0yZ!)2&Tv6T)x&5t5JZDx&(SkCBDtZnr;7@zsBj8bhgW;A12hTjvp1`u z=v^dts7HflHDyx!n~5aEIGx62-r+v1n+tx^?O

oV!~iS#x9o^O8K_Tyuty`cJdiA1e~cQ43=XzG+XqLkEa&%^x^rb{4ihokQ)W zCDL~vH^HXJ-S}2&IDfpa7xmT_V29N?Y7{#Gtk+!S-aJ<(a@%>xUsOn}r!2&oi;L;P z5i(?=rJCea%$ng&lT)}_#}4Qjk}`x z89NBOQ5pir_ybP)3Pb8t*nrKwIB3ow-p4+Wy7;JLcTF}{C~e~_&UN9yLO*_7Oda)h zZ|1tzo~J7WpNjhT80^$C=UY-=*vY?ZA@XJ?$mD=C5#BhsNaemi*Zd=eNxRV+V zNewNiQ*6(TA3Ge{)(YRsxJ6vZrCRE6@E=_)>|<-=tvSO*H}KT`9&+l)WcrOUKzA8! zHhGpio2lT#svKAekH%giwjqu1N?Y(?Eu9Xl&X=f-Yjy^vKFr`IDz7K64u2DQD%=u#gx7Jj zWDRYTjOH|McO!rABK;)j)@qN(@JCf=;IUs4Xy_Eia<`YWk6{Mu+q{`hPJPS-U%mj# z^HNBAdo3uB9>uopiR5yU`^AOx&GG#1`S9Vf1KbO}4_-R60WaMpU6FNkOZa5;S{;Eq zw4TtKA?rXVcna;Z4TtL=>X}`EYG5*08(NNQ^Ng(<$e7Eq#%6*)yxJKv(=wUzr^A8! zC(Y)q^ag!TO*G6@Aa%Xjuy(dKtX%9)Hx(-o&Ea;;@e)BFY|DYdh1K|D`f`%!o(1Zo zj*-J_CE2LW4KT)kF=>czVC0m_P}=Dpu~IU}{aWq+^Z)$!{;c8AgmMF7oN45S394~q z!fZ2c_oGy9_oZ!^b?X^N`CuB{Sx>q$vcZ4%b^0}-lN)2R4eb}(l3#%~{42|&_;JQ# z{9tZ^^$S%okJskL9oK}%R&6u`J#b`-DztVqk$Y8#gzjQBd~R`-1Dkege|I0M8CRmu zz&A4Z^lMV=u@GYVXTVrkOXsNWr%tXiP<>${&Rg+|=+*8gl1cJ7I-?E4Mzm7_c7x|u z@8jeT+~=O$I?J11%i_ak|Dqw!L&?rLH!<;ZCR+C%r{0gRlaE_(F!v6JVAqKIsDJ;X zNWuRNx%O0#I^^oY@WuPV)L|?RXtj_T`}=wM^RxKjJ3b&6UWaq`YJ*S7ROXLT9bKVT zjI|lvWYmW>n5Q(B*h@X9?H$A5!YFyzC2Wu<-dGctSVR7{Y*&>lI@4e4TSR-;O{bc1 zlDt|}2==cYj}CfmOy~V*vhbo76glY%8cGvcboVLkTeX{B*qDS%lb_>lPdlnCu(5R0 zPucyu=Kver0^xNL!ydkR7}d_j(6*hWT>sfprhnunV)iMXkx~y5MGgH*iuIJaz&j?q zQI{88ADE0=zWLEs&s6mKsgBlPPmcdDan)x<7-xQ+OIkGo9fn0g z&UNljy0E=9 z28r1LdU)=HLolfbhG8-y#JxJ3ex53E!*<|ujFEscd z@P|UJL1O<|vNgevu07f>N|l_B8)7D*Z*DBztmlRN&!;$RU=%*|OvH}Y6eJeS<5s%= zz_5lZ*!zA0-#6G9PYe>+JBODM)4=26O*>UEE%i31cPp5dH0IEZuH|T1d5jK84kNxU zpE+gkX712_4PrP(1?y~8cq)5_`{uTYhJ_9M49 zXwb#WopJsARN8pln1l)1@wuXf6DP4e}OHJv5v;JhWWNVe| zpe=Ml5{hPfUgJ%6C*hTcFRCok1Qs+Oi=NMB)A3ftm~#6S{+Mb{{yjkw_2MRM*mN0G ze!wzVZ&(d>o$c$a0zK=%4-8qGO82*7A zK4?uHPN?BnMHA{e;xR49W7K)uKbjogBbtAsh7MM;!aqw0W7w?7rEh*F@;_k#8#gI| zzf3Y+6svAG*#9oqe03>Va$S;EuMlp;$L~{tyNspjg11Udf}gm03jVH{jc@cDkuDmG z3U^$vb)hPA=KgsmWtA?yHB!hOy=tZ{_1SpEy9!6^on+RvEWr_V0uT7oNn-JoqvMS7 zF+D$CY_46yod0OWywS{|Zu4eQ#hE`iMK4Jbe@KL`VM_lGRqq|oRs8<{+hk-VByB4z z^L3urNu}&Wr6CQGA}OLF4McW>QYxb|qCrMp=Xsqp6lu{`k|s@Ug;d}3e&2rY+wJrF z!_B|`IG*QuuIur*-|1>yIg~#U19~IgllX%^5|1PiNK|FIB~^^8YFEa9S*s0 zV_rQdo*WKs*>Utqh96ZCtLGyoN4GOpUW%Ne;D=a~efVs>UJ!RDu$9xLM`)0C9BtWDN3A>+IhGM&q2V@~ zxFds3NVlPipC*Z%pG`%VPIo&0O%;5cAc2dF%c+M{6*4Br=|S`5^hJC+eK*^XvE93y zo05KyarttHGV6EHH}8E!Zbu*6?`@pPxW)L>kiYklpW9C}o6h3g2Um$g`BpeNG!$~q zowN`BEzPvp^wDoG?{TL2>WqCxGRzX5$4VD&wBxR0F!)H!S(E#7I014EKg`RqyV^L$~ zhMrQNP3QixqPMSVV)X(RV=7!_HB6;Rj(MB?VMTc0jxsql-G^hH?sqn|@ z7IORI=o|fd!GV1Q{d#gWNSrr?dkevV{qA~wRm$F<~ONIvb%%A)aC#*tTEnk4$@ z9!|dc0Uf?&3LZH!3QGOXQP+ZNbb^1Cn8cUD6gVyH2<_l)>KM9Ynk%{vufWl^8YmvD zhoN8mM2e4;sTBK+xjFNwc$B3zNDPa^@eQgp`HL|&Oj<@x(^U|XQ!2{oI>{LvkO9kO z@yvwNE-0?L!v!y-L^FyPeL181lmJ_MnXFnL1;Qj0Vcl6zupc zMm=HQm-KKnSt2jND~XGQ`=dJVEB^2K^PjuVe}6ign;`GkSg5*_4+A&bxsu7-xLmg( z_ja$FyNsVjc7-G;%*({i$C; zHltmlf;o4E9C+(o`aRBqoO;$Rs;me`5B=S>`)5+3IOP$E=2lV7)9G}bULcho*H4Xe z9l3?d*P%*aLQGur8Uk83(6Hnb^sTPI#=_&ZdclVo$=`NN!@h&`_<$$gC{jlELVZw_ zx=o(zn&F>$Y0T@CTx92!;-jfeM8_+e{w>;uzK7-l+iX(fVzHN)o-HGb1eMSnr|q=P zbS7QD(U#21`yqPta}Z4LlmO*vVzO1}Xpc)us!e?8hf_{zax<00L~$cWqu$)YE$=VU zHC|#8*dar0U%nzUb_sc-bqndQ^lH*I^e@q^Y~~bK{Nxf14-1(m6EY+~5AObzAkzJR zIMtIi^n-9u<7!V)HqVA68A^z*JU8cd@0G!)MbBs?35DM`{b~F)Q@W`-js`TjFw491 znWNukLdV!@;`L)K&6%!?H-E-4|0=_{SIRMBAL(zx-`YTNZxDH+bb=lm{EWUg&S0LX zj3XJJtZ7)I9#m|a0lgi~;?sR+nEQRrS*FeLihc=GRN4tzKbV**fz>6=oZ~nH6lzS^M(?%Q;|J@YIxX*e`YDe@6 z88t~X=>JRJDTOiZ=QVL3_mKJ5uo|`;Pk~v#{it`qL~{B?KRtGAH~P26b3X;EK~Jn7 zw_3p*cB~vq^v@4xmc^usOP7uVnWd@Vze1kIpTAGucODhn&Kd;^(1^LKAj@p1u%RQD z>!V81BciUe1lF!AAXkUia$g54;+VK596SFhy)HEr>5(d2`FZ;|b?d*NH3KK(q@i6nqCl@PLB`=bv~i^)siup*(mMk&+xe>kzXq9vVXq0NN) zSOISo&xGvuW`aFFkoZe2#O~Wr65VtJ0)`i1$hTsu-W>|Jl>`S^do0jIRz{zbVTsby5?ZE_We!H}PD4);)5lZa(CDg@gI4 z9^z#EhwOjvfT@wIsAP>4x9DF7*(`8hR2Dx4DYGM--Zc+K^8HwPZAS(-XmKMq`-u$m zEY28hoF(Zj%Or5T^iphKkPiquv=XXTP*m1}6kM-4i>O`2E)H8FlW`ndk_ zyU6mVp)hElA~+`eB+KPfn8XGxqO$dzC~{K)(diWK>kWydJ$yLz-vGcZxJ>*${~)$S zY3TE08Z+KJlaY~3fIH^8AZzjwUaV<=-A*p7yX7fn(dFSVGCr4nyL$nYdd=C8CSc3= z2z{Ke#k`}3I{b)gqZW@PfHObE#4IZy6J1ikTgVm6Fvx_frcU5AX%+fDV99)+o1){c z_eE=iKavrfg>L!m!8q+{4SoIZv#2}VQnaGei`woAhfmg*nMYQFYsn>p?s86HYF*^X zhO`WD9bkYBJ3!rabZL(BS7r;OlV?wzgf5Xf7(E^YDYqQCd9hQ8=xVE|a{C?T&!Qs` zCiIa*GOT&~}cGCYfO``!dn*caU2no-u=)hcS&K zc8Kn*DixTL7f9)z2tq`~B=xhx+4OW9W?AkTGRIFumhDd^>sHx-UAh*T>{?D`-i0t> zFO(q7yhF(MzJR;55dxcs@{VI4^S|x8Mb>grOoM=>j(*=I%#TWF_H7j$G&`49Qs2R^ z-4{Yt4EmTeZp+AhNpsM9kj^Aat%OXb2E0D@F^kkjPDD$Q6%W{lT_8D~<6L+ESK zxr1joi6CQmX&^5+qE?bIoDz{ZH;uWu&yT}MS2FRwCa+RAlQ(Wn#;DE3_~x_sJtbz(k53GLwU!Co1gcbmN>X5AmiZ%@LH_s6*1u)G=V#%- zWKD|k{wy8;7iwF&N#VEcAYQx>EN*8poo*_y{=rM;ceRLk1mv>ApF6Y19rK}OF+=P;fJA-U5#``E(vIc(0}3Hadt zVfNReZ1%xSU%2kT5_3OUxRZB@IO#NTZ*R?lSy{@Mz0`?W%IkpJ?>MHnagi`fI>;m{ zeP(oJw?p$HKelIK30(VK4)aDz5^J5kT+Hn>Vq=*@he4lstEd|KC6t+pjqzl{jyv>Q z$N@U`i9eZLn!{!-pF|J(yRn+zZ`j+`Dv=ijckS<+9Az6l#O!1{>51+}&5X}$B*OLx zPfj^n+c@JLT{80|QJfXRW-HHQYOIaO1?gwtq^$(k1Al_?u|LIIP{3&#m@|>R>uF57k0mnFdWYCE1y-zF=yPc4 zt$}QXyIi)d0kzObM?>eS+`779T*8#=MCG+6etqDNzIP@OFO#Xv6gPEJ#U~1UklEz@ ztcLfN+iN+lSVgD`XBcF=X3X{gPt{f3X}x-kx$7R&upmgdoP~yV>}$5 zyA^7ZKark~xs1`u*I*hxj(UIJ%JsxOg;T3CNqojU((RfBTJKlVoBHmwT+e}xA=+e; zy)9rtb|qVbv{mNBGqm%@tpX^V~;F+OFMT z=sgk&Om~aw^G3pIn_=uAaT}3NnE(>%$^f5ANn=Y7w7l$vm1kyvYONeSS42?aVjcAn z`Zv=&%1PG~DoigCU)6WoRpUGCfEd-MvV1X$&8J~27$ z`$-{Wh!6?;lz0NfV+a_ZvLW}J){<|lztZ3Fw`%{-{6``!_6@ncd_4QjWfrkMeugB3 ztHShIc^JF1guFF2AYVVpksWq&lCpktQqM#piRSl<5#Xq)HLvML4AHgJ)g`4ubu)}x)73AAwg^nbvBJ%=9=Kmu7f_LHjv&EP!Qly7p%z{xJ z=SlA?9f;8=!r;S;aKuA--edc7>N064I9$36iw7R?H5Z3X zry-n^;LC1J#+;^7Dm~2x0w#&bgZNYUrpgbFzL^3wU-jADWozgTHWe%ud*H`Y&v9dm ztk}w<5BLqa&@tYDKE1IHliIez{^7#^zIg>nedUDpj~lqQqG9Yn%uXg~;~|lBmn!wX zqQw>%UL!sEN^oDVjLg*fPAeBYCQV~b5tpHp@J>!PJ+eZLEB<C2j=5$uGaRl4_NgFx};>CX~n(QU2)L`b$3h*iT2Uqe= zg82d2IXtYBSx4o;xPmi3yy4Hj>dVZ3OcC}-e+Vf-u4dzuc|KNk(=k%VQ z6~^A3f``scqG@+JK+eCG^VN67QA#_p=9wj`jTjA=ovxCeuckD><{VAhq{y#&nT3hV z4bW0r1ZTfmGO?w(aOxw0DHiqg^JaZs&0FC5%n77_o1Adco8!1vsfauib`2o`jb#0p zmDnyy#^qZKz;>@3tWqo@Q{;O@1y{=Gnb0w)@}{2d-xooeavYHBl0^x>^Eh=?8_wC? zLTmpWp>}^OxqXvE=|k&Mx}xVBHzz_CM$=TPo7qKd24dLe&zh{?)hAHi5(G;fUV?g{ zG~7CvN9pV^yn6IIPMs+uWCUJuf@J{L?s-oy-PLEDE&%x>5lV8uCD6xzY_Q2h5i_lO zsA6XYi5#oP?+sSwyEHdav2POzKRJuPTB?Sx+JZngF_)2Oe+foz^8Z%Ezc6)~*!C>9DiN*{%B z?7Oy^|GEE2{D0qI67>puC?3tkouhlv+VMBO+M~o*{~p4xJ$M8?%w%xMkSr-#fTj#sSd@LSTf=wb-n6+WA@ zkf}w*6Vp+BeSw)`D*akI6MlW3u7H13GNtOkDCyh14Zn=hhp6;D=O!ON**G z?X6CACP4|(ucAmoGRBp894Adx!rc$9I@?)I@KV?-Vpr_bGW^2&*_#0`E>Z~ zXmW0Y3so8 zW^X44Dq?6?@E)>MUY(d)NmAKwE2;7aOL~^~7uVLrq4-H3waqy}R$t4Zk4|Tc+}v_8 zbG;e=!$Xy`n=HcDZMJ0OpjetSTm$z^E)m`#@noE}2i~@tfk*eYaTy0S?9VOPg&qHf z(_zypm=Aq_8STr8@bj25+3~H9k=9YeLp}GzD*Mt<>rXKai?-lqN^T%cel>K0-Ac}6 zYYSHpS4VZf8)3tpVT4bgOqW}`+ov&)b3|mG1$ofZk@kBpXWDemCh56~N z)4Ld}0l}MO5Q>u}N5LzTU37w>Ec!-K>Zp(V~cY{;Ll{f${0Khf5N z!T2mLl4>VwkrZ1eDvwC(l_XPRSDO`4r5*my@FAH zoU!#+0yE+AE?iX=jAv#P;zM6AO!*Kky6hN6rS$F5U&94II7$DHKmXqu(2WgQWS`wR zxX@<_k!sPzo5epj?TryYdm4zV>!NV`d&QYSPEOVnlQB<+VpbBSZKIa2~$S)k+Hg^aP!mw zl)0utDvBkEy`}JMo7+OwMI*R>L&VIhccIRk4q!&81Gq#h;1$ars<()b&iHcLWFHn`86IuedgP&?OnVY!!%9L=A`SAp_PX_z(loy_se1jKRIdahS9Z6l)LH_D) zp#_d*RI4|Z=9@i2;%JG_rw+y~I-$(bn5U3@=^Ar(q89osSO$MpU%~yL2Vn5l0sOYt za`F>K694;qak^G5@*C=~_0VgssedOf<)2_xRwnGZSPRF*su=z{jt-qNlx&>h4DP*0 zF|EL#J9hXYEjF1gxU_PZ_S6`{=WFwe+P9;@{Zy20mO=U^5)a=WLxKu+a1!o<^H=B% z>Rnj~r7~4qpUY$JTZAfWskjkr+I8{GsDDH>R+#JN7GR;v4yb-_0^35vAz~zj`s=d< zZk`m}oVgEwO;E&eRsSM&+X9^FFc91`J8pyk5Y5IAi7M>$>N#a~Nh`N1~C}~^) z@<4_JWmnQc$Lm4*jk%b=xt+VaRF&PMUJsW?55~*Ge$%~a&D`5NKJ<25I~_4j26aRI z@MM58A13t#e|lsH@6tx@qk9up29Dy}XBCq2_j>e>jvjtoc@b2a9+Po-8NwYqgE-0L zllKR|h@!WcavQX?iT~JOa&cI+ec$$6y6M^#a`k8!)rq_Z)(-?8%&b=!r*DO=KMupL zHMV@4^97ut?1{}0C8Rd}9y!vj%QzjYKxCw7-a>(yGTVzj+n>M~V=_n%Urgn)KhU77 z7PxA+7IC}KN1yC?Ld9n~sJm1LBhj6V%}=gEan4cv@r}i|R;@%SL>p3Ms>#)nyXl0P zcj&0INNeQtP;o~$O>NyvZd)v%7mjK3QF2vispdsins{KLejJncS`LqdmQdY;ViMYE z4WfD05Z02-9d$fF_n5pVuA|)GSovj2)=2UfB#Tft$`t3=F62EmuVU9b7nJuGIv(4! z!DC@DXf9sCo{PEnzy8I4&L4?6i{$wUCVBWnd=%z?UWt|OTd9%IRUW%*6)kvr09JXc zz?o2GFti&Cw>D3JwJL>7cxfLAh`oy&B&IQS!rmlz&3vr-b%w*ZskiF%jh3&vlp`whSxA z4uDRBK6_bnKW0x@PwWjnh-*_8NzfY0$NOH#`bICZb88^=tm`7}AD$O3#-kQh@eWa9H_gCAk5AOT+SFmoYolZm@*#1s2!_@b24M>AwS4V3NS}z42u_9aAs6SC(8Pjp1Z6a1!VvbDX#Kz);qC}nd4nV=X+q9?urzf&1-pQ*#9!|%At84Uc58UnxU zM(}r}&0z1I7hGf8dbne`d{3H_Kl^aOf&>>DLhQpSf;ftSk7a zU5AdmW^P*g3E-71;ZsyPnJ6K{w?#MM6{AbgZQn!v`b0Np8 zM#zwQ!mqdTkhQM@7GG$f>tddB{>7bC^}!4DQ27bHBTiGBmSRw|dqy?vqKZ4Qf>s2!5=+I?XbzP3z7RF-T~i~atJeH4%yRCiZbmUeAvJ! zW?OIqvAt8loSCJ-H|!53!v*eNu>5|es4E$qsyRqyU&DgGHE>Y!D-*hIIDLOrnLV@U z5$N5J0aX<*m_FT#?J=}sm9@sAUz$5^Slvi(owtC=IWI}?7@-5O$Qbn9gTeiR85@_R z!p$v6r0NZONNf0c=*#~Cop#rt_HHA5FI)v4nG5ld%tmtGDG+>Y;6LZjf6t!(oIi>C zedz~ssizxjBAG$59?(HD#3}bmxVam1xO}0fG&^V+TDiG{+J$1qV`C<16D3jOR~hy) z`(B=%CQ_|kvm+O7-ZPjMVm`$CAd7IkyjS;a}81I^{Syx$HVOcn?q&=kee*$(&5OT}(~p z+L6C^6u98mbJ&|FrP!JuI_&sfN1|^!nmkGtG4gApnDB;~Bq4S%o9%0coie96>su2k zcCIGVJ`BQJ*{0}SRY>fo`9oet8RI_Kn*94{%xs{Pdn@Dx`djLe{*t8ze`;`ZQZ*TL zsE{l+|Hus6E_jC)Pp1z~)-YLvi-o?w1JOLBg9Vozxr^%lWGWX#4`rE?6+W`eVzmeK#DGR4wSZ*|sD(UkkR4vtu>38nQQ%U*L-EjbyMe=X+rMl`d;r2o+*G zF7Hw#z1}*D9jUkumR(+rF=a=&N8ZL%+BBP+Pm;K!2A66R+SSRfxr2zy=_T~WczN`? zFairsm#^l=8{PG3JYH~`gQo`{gyfL{^1c>(Xa{i9QddO_pcZ02#LFSji~LX`D0lUcND47sbE!gRYmB%|sK zU_;1iv>v!9nz6T>t{q>9ONMRZOd}YATHYYhnu2L>;vjy}3T|NAP0~J^x3>{ECZ%_e zqf2x?Y&9N_V?FP3|Ag~8+A0Xw)mJQUN8k52w$Ca?3x&E2Dku8{vN+UAX$oaf;}sr#I4@lK5SwO!!Y zFQ*eG+n|QkGD0_o>XG>rO5?#j`6PrcIS)RfFX%^Z zVE;n}y2ZSNxK?i=tTj(|L^aTRvUSiY8^r$S-1*a;E&CnDo$P^oA+e;U_cAec9zyk|htO#oo#@m>^(3u>qaKVi8>sgg z#L}Z_h}C+c!=Is3uk3`9HQL}EI}sP(7@&9KQt-RI7sQA6ljJ2EaLm2UB=GZE+Vrx!uPZ}*NAHfEr>zmnukK|=3EgGsqLc8cQ5pt5C(x?! zS7hFjGu)x5$;ABLT3nE#PSY;|4m&@B2AKGOA9Wx(o#9X$SAouz0_RJg+ZY3aXoJZ}%61<1NCW7CJC{E3U*)}Zw z8>I!&<7;UAnP5I{i#xfn(E@BrpFmgD0PcjMG&xBlIrW z|LrP_Qyb2%-W<j*Dl#%k1UX6s91UCGr^;;hS;T1i z<+0#AlGTNdOJ>}_%xXM&^%(>ectd%~c_?l@L$WXH(O}Pobgz{W4OyLw?UhcjDDn}^ zfA$lLCElS`lOr~+^5yP5IFI8l)Q~9frd#_Kp%B-{En^B{-sLP@I)5a!xblat{nbhb zQulB>2OZ)*Prt)N4)D~c<~l8}20Y(TxKzg7heQP{qBJUXDriEkz;s@&!hW`R z4H5ZXShzX_l?8UxnDjN+_U|m*5xKGvGLj%Q&KCFw8KkpwHyG>*hsex2=2gUK7}_Dp zzX*Sazx_gq4C_VLM0AN#YDbc}udPLEem$h3giJU+!3RJ6_yn4Vt`X_zCgy9g0ts_F zgtiJT*mL7AK6XuotV(${UxR{dDo1r62+WVnIQH@H9B{f24+F_!_(O(bpWQcdcyQJK zoj=DFv}zNVqy$WOD+>v(XFw(QC%BmGfC;`bkgpelI>9^eLFEZr{Ud?i>Pz92ZMHLg zXD%_Pi6z~T*#cJYKA_UdMX>g}J8c~lO@?+Yq-i>4f|pi}u5mjn?!EYePETs59nL;f zc2^~tfN3~rtP}P3H^Qn7hScoebx!fEH}0<}F3ZQ8TZK zuAi39OkOyM8PG~%Jli_Bhu>%8iyi|Svimem(y1X1fgTVby_dR)X0R&;Ph^*!-Xuy1 zbfqtIuF$}W>&#ULLt0yBLa(K$;ZkE+GFK2)S^S&FJ@39{?;Lv^X9pA@8G4ZEe4dTs z@^IQ6FF}rNNMmecr$J_vh_n8FU-Y%EM(m<`mk9I!n6}QPRk;zSuwKe>DRXnUZ7+?I^FG~Ek zT4c@Ei?=(+(;Qt__;OeuZA_mqbp_Pu{_K;yRd*g}(%H5wqoVxZ3{P+}c^i3OXr}s84 zW2+R=-=E0o`i8;Df_QX@Jx)^BmxI3Sa@zaO8uSw_=*asun40;P+&i2hFdzpDnM5T# zJ#!t`yllJp?4|(P%h_U{`Z8LrIZKq2dfEPvUl3~_H3zcaz5$~rgcin4rf;ljXnndR z=Du~obALyWIj2LJ+^_q&uIW3e#8QE2?A6?1054ay>AeE{OBP<%7S7moAT^A$v^me|I0@cbvkY-}v_DSM(p7 z!ap7DhmPq<_-A>G_{;Smh6$0S&Mhja8t#fgQ7-5^pTPy~4cxMXopjC^RXi0Wc*;z2 z;g0eYOnM`WmQDQv*YPmEwP<33{+X&cF>23aKyw4DqChp@VOjDy-r`z$zwP39Ns*J~EzR@{muH?&>Y&!o)9%q|TfP?>8 zVo9DoFI6W)r|7Jp0K4#TWQf#2}r zA!^N;OLSE2aOhDT6pu4UY3ZAsMVT2j$tx$D`*z^c1AbUP#RQUej$(R)QM7B%a+<5! zORJX(OcM7z4E*^GU-_5g@R4zNmRiy11O?Fhy#s4ZW%yIi(lD$&6^VTcRxSYuW_8k0TZIt0L1L|J7UZmC z-22PXDFHNaWf9e${GLuyUW>C!FX9*DAN0zUbUeDn6W1-zCAkMKQ#FYkS~2PnI&Z7O zd3#=bp+RAR*R;u2pBQ@Br8C!P^~6OTlT%yfESSt+hiI*uJSs$_5RbSCi%%gie<1sY)t z2@4y@$`O9>WzKlipM40!Tm)`sNEIeWu0U^(!!%~gI5y31fDGCY22$I7Igd$u>EA8x z{FA&NG|({sHp;A_G2<=y^No+_tu7Z#TV#lbtEQDdeK9-Li4<~amnglYp>o~8gJIX%doxQaX+ z^^ML>@27b+9{#Yn?nb9dc#XK4~fdR90aI;q}_BF}zey=O=s<+_X zEG(iGsfARn{s{fFI|3?;pOI6BUvcn71%CS@Ia*fo7Jn)p!PFoF{T3?dw5k?Z-lGM>3#%b(g)gW{?WFhZwvbIHib0%lTWsAY&w8()L&^s}!}0uh zHez2lG)&fH`6biXRYj$&@%!-*F0fFe+~mO7--i7$LzR8>H%S<`D3RvyhvbOtQ*x?A z2P`~`;MlS#uo@pn;;pjDkp;s*)nYE4og7J54gW{e?f#IGqXlH;`?auXi3ADG*Z@C< ziioj=HH>+|vabWoV9L8ukaUnzIX?}UJSCf$dZdthqwhe0@?A1ws2M%3V}^!>(o}8X zTCU?mDzVu%h$f!0gP}_$Sh7No9QPdwtt-9RlT$~tD`LNb@r3i@qvr`ZdgLN`()5#= zH*5-=UuOWE(oFCEDk@%^~crC~t3i^%M z)Z|C7X#HR`_m-yf+e)y$T^+Cdv4QX<(M+BGO1yob7U#Y$#mlX_+@=dY@Z$4q+RWX7 zML%L0r32FJq@i0`#g2M5E7FAZ+p5U6Yz*X{4qZ?Gl|-;F-%;4wF_k+wyntR&*^gF6 zmh}EVDYC6^9=CE?Im#{5BA#I>biM3RTCUQ?l?G~yFFlzLZ#Kt^c1Eq} z?Y99&Yw}y>+atpMSQ5dm8}X9ep0bs+N*-q_n%2M-twZopyo9|rrw*h~II!>U+Oo~l zzOuINJbMC6$+z# zQed88BFMH>5xJZk;!?q~KZ(8Inr>nG4=m-DE?Nh2RlP96TEqrNg|V}&?@{Mp29VIW zlh4VLWyf0D!-&CR?u%n0{A_Ok=jRzT;~r0P&lWLjyvKsAwKrQla2p<6SH^jZ-T~_+ zyw6nR;K54|QoCsxMF|-)*}Qa{tLA5|^)n(wT4BHS{VQdEf>ZY-{9#j@jWlJAsGNLmABtPvCB7;s5f_|D6G) z%x$BWzg@?|O`*JX&1tk7$m8y59>xdW8Fce=%OQ5V#656E8qjSVH`pZ9<$k|eE{-&{%Ic1HfEKGP~4>$g}Rt~Q3-$~Ne zHj%FvtlnH%KOYDMw7gSm`F>T?ur_$zvyJ3tx-9ik)03rwL9D{7kMjuCmY zaYKzitxm|mOS_Et9rZU*S1$*4Ui(ev1t{U5#e1pOp3|IFzXE0k{KS?kXGq%hIb=f0 z3Vy=QM808NF=|U5q#2Q?!My7x(Q#|%+LryM%w%1h7&%HjKEeQfbey?057KE=dlDQ} zd`)E3^!X{1NAMFe#$d&;MYPg)6?eLK1r1;5h+B7GqAf9JXq7@dz0;vin~O#0d_|6~ z6?Xq|)p4jV6&advWBRf~g=rgZ&aeFE%ok-{LxXE4xrlC||1znJ z3s|w0OB47CP4qMQ=~Mc*8|3YmT?=*KB7RBUJ z^T=x2W?;+624zzFjlXEof@-R*tHjSvIDipOMM9@T9}r{zu87ky=dVss!55Vd*o7@j>p44E^s|bQ z{pgHipB%v$g|T$&-ecH#&KXxb+F;tk+w@yl44F2E!P@W_0S&IO)@7pXy)ZNy zyNk;jZpo~(C}%dOKA>_{?tVax4gkcxJrM+L9^lcX7N>*sc2 zlkbE2>PB$Gvz)uqG#bO&2GQLc4>5H)Q;F`s5VFN^Hr<%Hl6c5_CB9KGZBl^lxPhrWICX!+qf zO`X_Ge}67Qi{+tsxSTs$7{UGzYk3&mVkrxLw+&;sYY;&EPlGHyAQ!lm2Krg8HW z@TAFEfjuDjdfN49%Hrwd3bPch&i+VcE*kO&owRXM@^e&OT*-CSh48mOccVkhcD#1X zU%&^9M=soz5BqVQNce0)vum+*L8vt!HQx|7?G$#}lV;&qmpt6HDH4V`tMhY{Re5|H z%k8`%gAq?|QuqCnsmcT`zB$~2e_*HnfA2pMUlil$eg&%6F zXM(^khDN_5WbsO8VxyBJ@?OotlovO+fC2+bD>X=@#cUXV(ioy|OY@5R8p-(i9`xTt zWmJC|14_4CV06_;oNxMwr0G;JX=ZBJ6kJNy*De4f(_(V#pe;Z5;S+R-%D^=CEPBj7 zMrX)b!NJ{iu>HdU`eCBrYI+`T=W*#h^_(osEE6A~hFlk3+ILKFu~^fC789sp%2)1s z{5`xq`T)KkL(#4v7ky=cFo#o|wrg@G3Jkc%2)nJehGu;64qZWus5f`8OtEbE*k%HuWsIZWc;4o_K-DYE`bg zb1hwzmCb!oF`!>I8Byw~0$%1*a5483>7Q6qx2sP4=JX0~&ZvE)uQ7oJ#%trDV`s7N z*LwVlnQA1P~TR{c|~o=9jJ&yo+J zl5qZoAqMGvpeLkVaKRD}ayly?{f{Z5*dm2aW_QyUg`*+E+zV&UxIuoj#Xy30q8Lw& zf$SCO-~-E`axstl%m(xBbFAR~vi0y@;N(aC8OK|s7UG^WcX64SI%l_@$k|k_Y2y_C7gkxuFwpjM<9DC*`p+-Gcq{ zQ;OX&JqlDaY)Srd13ENI@LSI8q6RawaJq~uiceO6p-d)u$*ABH3t8MPbU_MiO<<*6 zA6KiIN*~ES7yo-#Ao!%hFvxK!1L-X%1@NUAn&O(V^j@}4!;E&SqCBW<7AjLc@j3Jl+(}mH{tL3A&?nY zi#2Mhbi?qSm>z9LeqVIJ&wLZEy=lSE-*E!BN}MNh8dHTGPNKj$)8MJS6hFzl3)F^% zk=nVxaYOkoS|_r@dt=LRyywqgX0ib<{QO=3ZLz|sEEX))a}Uzj!d#DQ<$ zN75ay!WSO2{0aSkdqANRjd?&Hl0fd<-Rp#NB$-q`9UD+a zc;1jcmp55RyzX>he)d!RqsbHUN$L*l1K%re9l988sFuaoQL4F+DteV>w zGomT9)ZdpncyBM?ae*zGFa~CY)T6DnW@M*c z0VQiTi9l~NRWfZdyyEwP+8R&3_DgyC`R)=Zj7kROsxo5CJuC110?DnDmmqFi5B1Bq z5#eAzR4Vb6s@>PZUwvv8+QIEln_aA61f7NHwT(WQ7cz-l6S#sV(UNG2@-A37`8afP zj=$$I_OQxFj!2a)Bx3uFDYK=4JPDVn#CcN|qunS@3r<=9ItSmv;D>xLek;MTz&(iD zo6F2O!8wrYpiQT)e-A>5&tPhD4ZM#3K*fx3d(GQt(~qSXaLC;W-Y+Tm@@*6SvuYK7 zo;njbs+GeHZdcWN#vA4YcgeWJW${$>G|AB$PGFEH1-zT4#Ct3bG&(Pm5giepdB$9_ zDCaC(IaG_U84OaEFO6WqtwKCz(2VX_Y2nn_53#VLHTfWMltgTNhaE=Va|mu{lp8h! z+QOBQS+*R8+{`~srnCi>{(fn1};Q5g3+=x>h9%fCW-wJ;rGUI4*!Lg3=BFlO3m zBdmD4538-aM0VF3;r5*^TwbRhD!ck>*6ej+e1it4jii7PcySWMMpbDU0V6v0Vk-nK zmu9zsQD%_XA*=LZPBo3X9s!YYm zA!2`>CMhevGpEibp{GOsa6uxO=N@trgq#k8^{WBK#CI7A)ZYtBekh>YjTa&0tr~FW zH+T$Gqdmb|DC(3V9B6!p>;&WQLRUZN;wAl-6=%8Y61tpvj_UvZz|PL zi#q!*2hCl2jk_PM1>LGqq%c&$tEvt~V#|y1u4@3V&F_Hra*mfWZw6MM<&1YFC{eSf zxYK@Sg>bJ(l0K^y1qIVCL+kfkaPQtl|CynQ#^%Q1#tub%lNKaq<-74`&h_@@E62C_ zGl?W9CE}`LLGo{xF%i4LN$q4P+Hn?xw6>bU9=W4X^tv5&KVO1|l3p_(tY?BvW-#hX z52Xbo&f~F)$)LBg1k&EABBh5$SZd`Qvi7?JxwbnTn_S(FTrOLI+v#Qai1SZ0sbBK{ z&c6iG!apJ?9wetby0|`*7kkO#44X^h$iMeul;`bVO#Ws;c7J*$j#i2$ZysDGzf!L; zS-vqijeB1xu$kFh{tQjuI?9)un~%40ZOoH9Ram_yWwv_-pX98sB3pjt;MqpPZ0~(P zc1y%Lto}P4%j!y!b=3iEe5pM5g%2hPR60)F(2MphF2Jf4Db%7RJ;ak^8|~v|;&E$z z;$(b;%-Iu8{N;q$;$y1B`G_SBej7^q916(TuLykjax30&$P*p^2JGHPX2fLCesW2! z2!A;-A741V7rXsEL5lXCAon4TI6Tb72Eu;)!lLKsOw$LPSn~yK|052snn#%KE6q4C za0Dx_sl&ED2g&1dAns#N$(=XJq`uvi6f>u4)|`~Z2RrAI$HH8_c=sdu7@bbMXAO`B zon#VZG=_RU$B}fwIjF@J25_@-GHg zE|rAk{kthkr)|vBXJRC8(Fgq2bR~YP5rCJ?kR{z=*5vdz5BAWp5z^8YhtDx4taoZL zd0!LBuy45M_r+o&!D_JEa|GFZpKP*XxC+brxbwtqjd=4ui$Tmz1}CpdW!j!bpgD6j zNz%M9JaLVC&lb9T+g~0y&M|`QuCuF=eM|VhZ`&yM=QL@)q)hy5&r!3+?_l4XKgpoV zS~gYV17GDv3G;H5D9E}*l4AQPva7?Fe7$Ck`vU^lT~mh1{fz;v=bKhmpl=!b;BOGB zN|eIC8XoXh_ibh03S_X8UuKi@d;T!T?0%q#sB=`m(qi(*Vh+jgAtbMV3$xw!4_?v~ z$K)MK#XsiXrh+>#nW&fqIlIg8v30k24Wa7%8#~L8%x+WE{jPzpaMKV^S?o*r6DK(~ zr8n6-H}iZ4A#tizRwR=pXyQzwDX zoR?$I8mki#yqnZt{=;9{zdDOguJi9GOL{>~A zHO0PUYeh2=zA;3Orq+_&WtWNg%v8e2zrdemOt8uMwbUs+O}s^pz&Xk~Oh*?z;Ed>^Z4Tr|{3MXEb{+V(N)@LG%wmainjvu^O_bBtAJI@Gf1AFO?}dlR6#f0w zny_vv_=`UI@^2j)9-H8YmYyM>e>}sReP!76absfQtwRn4m!ca>DJBkYMmO>iy!CA- zHumxPCOzD){il&2d0sgY7ELfbBz&t z918ZD1(siSU1gnDY-B|}PZ1~d6ieFMA@yc2^s;IP=~{P>oVj)Uf9GHS)j*;6ZZqnO zxrM&zzrfNrf54Zr1o-6k2&S3UpqM|a!6p9;3h6xo+B*sm&CN7_OGwdQyz0T_s5PWU z#?*WpRECr| zR^n=(+2iP_YbGW7Bd2<9o)&+rq$cSPRR9P7^Tc*nA9=mkoN9ZU&g@vS1s@U-Km)ns z%x5uErfFe8O>?^#dT^qj$!>pu17^j zOgfcXs3=?veJhrs?|1s}ff>Q*=i3928T1;Scu}zVZ6uZF@Etvk+==ke2+o@JhZ5EC z!X!2y-(8W2a_6hUilX^c$H!@?%R-)h!evEg+;Qi28Gpn4>xoRp3<2JWKrv*w_aU;X z?1J)*Vf5F-t-v06OQqb-Mk~&%aeV%5@J+FhC|d3%uT_%KuQ{u01Pom)(wHeoS3MFb z{-_}n@4qlL4}PHqaYd+Pu|Cq&7_khq)P|P*XZel#_bB^vaad>>0fn1FXnUQfNcuFz zc-w8nb&iRsP`;nR=VjQrnw2E4XA68@V+9U%32;R}fzgV*WGVCdF-m`HLvQE5g>M3U zyrkR^>sRQ5OLY)>amt8(XRd*ylVzER#RPq;#C(V9UBp``8eN~7QFDCIgSxHgLCz`e zLs+&L)#@FFcyFLqUwJBfbMK`EYjuoE-8%{E8I* zrzRnO>FGb{%2+k})F;k-e-uUrEiCAFzE-qjQc+$?rdj}7A;19KZhe{#uOFY zisD7-hESi%g5lIlX}Tesfxnx}sGU{kn8F87P}|0h{H1g9NvYU#{AX_?SW`Z<>qsgs zFRBib9qO=BJPbvgs)ePManK+rKu%5F2=*!t;7N}luR{iKZ9vFUeXD-W{PTA3Ug19M z=@g*neb~k^f>$$ZckV{Yv5icG_If6wcc$JcEa}R*q#ZsuaSVlf$ z8tu5%#$R-I2MEihA%nb+Se0>R7qk6H-Qs*bx~4;x9pVl6*L!K& zKcg4=q6^X2@*-%olcU!pR`Wy?P0>)rEYOl!#k(>6F116F&p$8HMqOC2g6a+^LF38_ zq&>SG>vd^h4n|4BCT}3UrB&qM$!Mq##`s4{EwBG+IaXS*3hi7L#_&JS#hES`>nZR! z9>#KXemou~ZSeqk-X_@j@FLTBVk++YD$PPPB2xL?_;}rYav@2FeEd4@|PdtvOY*VC>(uaK2Myc(YR#-@Jh)H()!#FG4Ckh1<*gE46 zQjnXCGj9K2nBO!xkQj{a&R2r)@~bH8dKFGn9%T+6QKEM5E=T%jCeVAuIpopuQ8chl z6rHNw2$!7fp+_qQs`5-o(m_5h-G$J@hw`Y0o1-67N<+FGZ?WUXgD~UXefVH5MsyTz zFtT#ZX0NY_W1S`))c9p5ldd2O3%MNQI{SO@^w!xLHu@8COFc|HGG(#f;XoX+YBv$s z5edx^Rv;KEk7B~5h+nWex#S- z6((dtGZt0o&4efKCxQCDry%z2GK87rTPE+CM)r&+5{KAnxMSgG=KG#|So;>kU);&{ zF00B=;R!*~IOi9--)#?#%L(=aoA!36f6FZzmrJk3D zY*$lwY9mhVIwFKBo{pji1HvGu9)dF}mowW9Z1Cyne4OlP4F8&4$N=B_hA3w4!!aQ$PM)kTD#4KHo*FO!%|Gdvp zr^;;Lr5pt&Ts|dWSOtlDE+rmUVA&#hs;i`n+@GvP2=(U!wqO)LLYnTS0NoSI{~xg`Uqo zh@15nlR1Ab5Z8ZG$giXNaQ&JU9_83`oyO4+Y^ciWioHZPe5Dyp-9hyEoF2^fcLo`Y z*Oaq|Ad)%2@#6l!^C#32CV|SHvv`G^Bc$yVBEMdi5U0%%pl7E>s?B9M|5H7YoMBF) z?~mb^1!qa~Mn9acs6@ZneFmcRHSxOnG->OWB+EK4^6M9lG2?>mWU#W6@XbU>(?B}o zF_=Le&(S25EmTqy%(FV8+!1e9G66 zIPaTDVyAy%{D0fQ}%N_pZl+#pSpphvPw7T)`F>Dk=WDF1+MgKFqngjlR3i zn3WQ@fz&%EKsIU$GY}s|)%-hS$sVi1uYZ=17F|QK>8}yUGH$eL&||2(T8KSAC6L-k z56<^P9=KCzo=UoWj3~xKX;)EIAFhy1k{$*-|7ss#v{v@4Qe<_?f z@^&7%9U6iSZdsF8E&i~0k}phd*P%|l93pPhW7wXf9jI#K6=Ya1iLLVgq1u`}lp4pe z4YobSuB}#ZeyKGDis>MBu#4j=@508xpK+3{JM5B;hVvE6QAS1^>U_SD$zN^EE*{uM z+TPuVp)*U^bN)HRb<0h3a_s^TdJ_o?jJ;r{#28K)n?(j+Ns%`%uVd#)3(2Y<`N-yV zDL6y`ZIrr{UNBV*mjuP*Nl}Ji9#H}|f#z^zy)ae|;dlV=V^PG;kIeIDri3b=NqyU# z!YtO~IMJhOYs{_fl`P7G7Vl(k=9@$QO&=$M+H*;hO9!JUI*k>U*h2Dkeqh_- zN$gjhK|F1G56Na!AVH?^&V%HP+{0ig5|Z_(q_mc{X(2 zBvrVhiDAFbC-htK2DA9rLVErec`~LQ2%`^gfCm#!=lOE_%4yON^XN2Q`d}75dTAwE z=_iA)4($R{s}OWv$)4F3a1zxXnGW^Kl3-&{D%N{!4UJ1WfYI~^5zbrGG#H4oPP9SH zfF$IvFoNNqmTYRlQBs;2f@L@FKzXv4$?KR2vghLkvP73}s(-9#z3y3oBWFSxNihnVf0#MZ`}peNq@=>PT^1ieeZ`_?;- zBdLqS4fpXII&PENM-=W$;$}_rJ~AIyFQ%4Vdk+_nS(8Q2)X0*zQ_#+c!*s=oOqjY} zp3W5#qDHI(sWzzq$T^t;eujK(m)Ohy#Kz#El}hZqUM@H8qzcA#I_xT|VpKM#Q9s8= zu=dq@+^u*R?!C~YJ4M1EEjABnFVKK{bJS=#Qx(`OlMc$6F0eLaFLZsR;r-1kP;ot# znv)v}hN|~qpPV^r?uvkC(NAIJ)g9FQ6~>UhQxK+nT}eW|#u4j@dH=urSK7Z196Nsx z_2WqiMXOn@qSdIPk5AXd4wi(8FZ<8}hNxqIpIZolw{If{jng;LngJ_Vq3eGexg23@MsAW_k zUf>1}$v#64zQ06STR4HhP7x+hKANWy6ah0|v_hfUZTNGr0+I$Bc;n$hpG2 zrj1Wwr<%2}eU@g(X_hO<NN7k(|!kwXZL`Lx$SycIz2qx=ze2jzRsUbx^m5a8A*qAiZZE+1L`m{Mp83T2)Mm|NOJmyzkLau+khh zISVoo_doNecGdyuK*Tph2F2dU|M&%B6CAfH_l*_m%j*#Y(0%ogc0 zi~-!kiT2w-&NH0R7Y%3badnrYD$Dx)@nydmuVM>24T#GD4Rr9j5Z#w5g^nFrP5i$Z zu`>iNl9reR(EGa%r4INZcG)|0cyTI{anR(KY?Op64+gQ2=M;7lDk8m|chN*?1c+8E zqixX%AX*hn^5*#xGr?gJS^kvxlrJJP_=S{JDud_R-X}uO+!1?ShjCeJ2BxRYv2wjO z{Cbefe;0?KC#C~#ZLdL&&T@FMp(!bOPm}!yj#QhT7jLh%1i$##HFj-7=UN$-6vxH^#K0&huH^8%W1~0YTMD8?b13WP$dso`i3I=^-ZpTNmK|+&ElQ70! z+h*`Ri=Ob?lxwl~l|CLgrBRs@G}1mI$-GDuAn{(^=<=Flpl19O``=}_jz%(j_QM}? zvBeg2-qUaeEhOKaSbl472AuX2```LU;J^Equk#~vMdLG8dDuugD@!>S@nP2Jj1W=D zTuCeb%z{=9*A?Gbg$}yZQm!VOs7Ht2FoG}hNZ%G=d_&KYa}Xy$h=n-p<7F^9Ju8uT z|9m7H)`Z_`arsS7vSha`pQIH_Q7cTU7@0*TRN04gymL+{(-0TS{1975%s$l-{e`E& z^U)(5!~HLVUu>h^D{nVN6mIKOG`CF-UH$M0I|LM0!|M@89_ z87Jlqbx^tjcivKGZl~$e1}$o|snu?zAGiX`sD#%z9+1V|(h+2RvpO!zNoTU&-2g?? zztql>*=SIz8BMt@S5wfEX}%#r2#rQvU?h9>Q2V?z%hR)4_sdZRi;Kc5aA25Vlca7`=~xbGkcyyC-` zb#Y+JWuqk*pGLmA)@bXI8q4R)-H3dA2)=H!0#f#^1YL=0JkqC+=uLZ3Ozr`c7Fvx) zi!0FK06CPkW(qYov>&;xnFWhGaIAJLh?5 zX2}-j^n49UU?2?nT~pyWDEfr<;&N-9E$Bm#6WC-0V6`W!(OPXg^rbbOk-8)cd-s0; znRsn5IC+u%p3T+3)6{9vhBoF|ZV7Xjn-3QhJU|oaiqsg#&pJNmeYHsFFhjl%5WTju zl!W1Z^auDhp5EAybp{sk_K+D2l{XQR_@hE(O*QJ&$@ ze6(l!C|MDnzzokhNSyD_VjnCkLE+p!*R@P-%brRnY`r>?X>X{dzP$NIDKxb(t2ck5 zJ|CB*FAH9V>gSClOEwHdU6#NBtw2gIQI-ga=itaUOOe)C3b}l( zn(|rs2|rfoXa3LqOJEwU%hmp#$Ymj?=Mp$V|9HFrhlIz2(6AZIX6~VA!Dw`{-4adA zeTYR3WEdtZ44*BygYA#rf|Av%sB3rU;2&LS@Kr#W7ngm3X}fon(YpAVnMylT^#|mj zuj(@OG`9t%K0gMF8vbENpJXPv*`6G2+Xyq0vO)5749*qlz-yfcaAZ|6{`6x%Gx8yY zIX*G9rhngG6#pn1zW==kUUfap3tv$hsVoEL1rH6#_M&^+8OG#9ucgtK3AFmlCoHn+ zDq`an!ROLtP?Km(x$$(6zM2eDFY87Fv2N&Cup#L$I*WE>%tmv!m(do(B4mxFBgqYp zK?dTx!MD@|e5*o;zvm0yN$L#V^ztC1y}lO3sQ5A3G8fR#Wzl&2hyu2_rVL8S1ourc zU{!vlpeQQ=a$|`e()K*ZF>qw4i3?8XmUtPKSsO~7sar{}D+{BYk`hr@jxZGIrjuUZ z5HRqoq(6NPr5&9L$&OE-;B$2+eer-P#|xC8SI9=ejSnYrv172m}GnGSaJe-P$wkMaL4_zMJXm*O zGj^Oc37MsyMIXB;(tPhZmHD;?DejBqd;zJv_c2X8)j5JhmC`_W>L<|Fc5xgdb`c4G z(Wk2?tws0WN0IVjQK;3q4GtIIQ*RzmCbC_tLDpV`RW3^>R{b||^0ry%*7_v;aoJqz zuhSCbU@t-UcUki<9TP>8e}w6*x1-P~!p$xxd(d;_a#6&?a|k<}N5=(o@V9|a)MUJe z$!=WETjlnGQM0~;t^WPCyn1#oiMAO>(we?FyY3?CzcGys%(r0=YL(E|IuZ2k-??<8 zhzaUb7{xc5S+xJ14N37)AuhG`*dR$B1wKCnS8Lp%_F_6E)3Fi!;)c=omvbl&=bLaN zlnbWstvF8Tf@93P zV^K9u?^CGH;`QisLo@Dqk^xaMMKELX1rU4JLNb`EFt~jHE$G4E{b>)1=@~*j^KKJ! z4{m<5f>6X4_;MUY))g*#1SvD1|^#OldeCd_abF%)n5-}~2p^-t(m z`A`^NRzT_lHV~uL3rNVkZ`gO_Jh}X02MkRpf-~pRi%$JRRoLnw|K%lQ=`T|hR4hbN zEv=}!j&S1b8bKVceZdFDvzcugj;QZNJ+gdRh}T<{Bjv6}W~@LL1s?3g4fQj~s&xTm zQbswsv54zwbH6?9;xu>;Ct&9cU09Lo$n6Z3kS`uH`D;A{sLR#HOwhj?UVwik8r1w} z8O5DbQ=_(1pU*k)=1#kgYnO|YhN=3f?bITitau%{#vLK07CP9hDvoXHU%?70mDZ$6 zJ>m_8CLre-`*G=)DpGN1HTyTrkIc%kAYc>vNX%F__ z?(D22-%*zfIQ zu%9c%2JO&fRoZJ%)Rpbj9?dT5#40|%uOmll_Qmp_)TdyjjjH6BKr@buk6~6VxkfBc zH<8gtsYt$iFVV>2>epTGQLu~@_k`sZZU>S*NG9Y508zLH1PNQC$|l_k zC(6f8kW1Z|7%tbw-aW-=@k0a1RZs-xOAhfo*@kPj$Kwe1pLha~@%+l3F}lfWM8re~ zowg3O^mms-7c#aGflf1`+CG~UQgzsCaH>U|!5nt&oZBj*cUP*Vw8dPHteyyU)=LLo0U*_%k0`D z&K3zCWwp}>@n7Cos?t&4(%oP$V`O`h)YV#2)3~hAVmBAg4PQ>?|DDOMK2b?(}|H$;ZL8#SqY~ z3ky(tyEIl*l_j0InUKpHMN7v&5}*1CSQaEs_YJEO&m0NZHzy8g_h>ZvTq_*@{FFpj zoMX3cK2IH}KMtmAPmzl|D|yGY8*0|xTuK}cBtYvEA6llX6y4Ezfb^AADKncMB8xA= z^xb+8urUdrR9Aq;-#dwRZ#Uj*b(#0^!w%H^NR_#;GYm<;t-@c5Hluf(Pi&V+5dJ+m zA0#JnuI6HC`o#QSly{IiZF+PI?KG>HHe3=4CNm{qAw|)=NHI9KWe$0pk^;XEJm$Ok z?IDWH8~jbV4`QU!AWP#E^JAj`o$Vk^w!Uk}kM9|Q!-_%Z5T8YFAMC}qzeNzCMN?SU z*WqAVyqYdF8iZzR6Rftt1TS}80Ov$sGO1%j$nkVd!otx zdVe_5Tn%;oDzL40Ia2JBf?E=8MEXS-nViYT@*Y=_K)gO{@!&j4+UQSrUVF*pT~H!% zEm364_YLUyvM}^u* zfjGAZBrT|c?=G^LgnXGzBBS&`>E3m+T~(YN4lJNf?V5nk9}k0+h7MG;w~(ZySk_8e znzptvrbVxup=Sw;lAXD?iHU1BvwLVM*zf<#-CMJX?sQ*J;`%l>l9FKY^LO|`*C8U1 zaEpvuVoFtdIqp1u4qaF|!gurc1-<9DsqvDTq%KU7Y$+NAN8dQ4FC>9ZQJHYf-h)`& zEhi$bzwrhSaay$TFvp_yhrh~2C~{N~?rm6npWA8F`_CKX*8K*}#|%0;=b>eaNGgdh zaV4jJrlY95DIhs~3x7N-ht8jMq4M9&MfzqZVC8QUIzoLpDD6KF$MwF$5wR6?bu*!- zw(OHJatnUT%;Xg3fiSs0_nnPDq*usmQNn{684KHo4!nD8ydaZm4b^po( zJoD(+8oZ(yE`5$CA@5Itt6V=E{MiEr($=u4+aD(8CnJ;ZCRkKRflMDN$3fI3h+O*t zmsb7){lX{c)5lG8=fxpNiPxY(`xaK!{DyV*tRexwAE0tuU(#~mJo{txA2Y3M2T1n2 z(GHj2!JDltUQ?F(|EqsmN4AlfH~ffL;$^5yx<(eGBvO1g5@OHE(h(C1^qhwV^sZ(b z%-f*LSiioBW=0hA-W-!+*jE8))%Z=E*`mh%zS+RW!XUojx|ir1eGD~ROrZ)?z7u!xI0!fak>*)6JUA~`pU~xOz}4 z^RmztyJ(PCK96H16sQx~_biv4RUoei{u0f+7}6#p2HRC`qkIoNC~LBUg3oWT?>Q05 zVw*0L?W+Y_WHz9VujbJ6VH{BzG_Fs=xc$LgaOwJl&QCQV)o~B7K=c8m7@&bp+wf4t z*HvU^k~T>`90U(8TM(01mr>8HUFbT;YVtWB0uF-;VBMgME+(#lQC1vpOpFH28N=x5 zk2ao%S0?iM$Kv%F-Y9vy6LEP_iM!4IOEcYdM^n)X* zYPF%GK6>K`=PIOa`v~8b^?>1LPjQS>J@yN8#*gk#paNH8a=G9InHs8%1=HM^C=!F( z&!;m6;|Xy6*In}Oz#h_iW;U`_7r@s!*ZX3>e5Cy-i%GN3=ZRNoQ6^jBsSbHTY^EBE z76y;v5gA!n`dEwnSs)8m&N7g3?;Db^jKl|zsL(%mghErZ5VUN~K?j3r zq;cgWqAbkyyA{oe;qwukGqoD^{g5GF2Wlx7pT7{`GXR%9p5XkN0k~T&4`rT7gl+fM zGE-*BVWG$Yyl_aJJes=`&rdZY!^U-J8MhBFqH79ezJIA%Vf>netv4vE$U>fCb~v@h zp^d*MdJ#X>(~Ei?caw1`RwJKzW-#kf0?v_5h31X%@Jm${jRa*=E?W1Q-)g2zo@hVr z?OTBr9cO{+k!d(td<+TaBtWTZG=kc9)CreMRK`0ObYRXW=q=ttYfZ8yk5v=!p|5wC zmJMNOK{GeY>^zS=^xpA@jEZU$4!dBHms`=lqm#+}Z(3yi-*9A=GlDygq(HrWo#lBn z87au0U}S5Afm6qU=aY-fBa#T2r#8@GZBF#xuje58SqV<9kfPhIV!-2PFzpbvgB~rp z!WWYf#9KA*!498#ta->Ew6o%jogXeOlM2}^{lq5+&(I|2{v6!B)(&jk$C-JI% z^Qeeh3t?jGc`W=Wjmn|*(7|{Qbhts2T-thoe_XE`{dn{Smnu&t{Tlb7zc8B47p}(J zMOBHMkQO@W?S@{TjYI7p4Uy`-ZtRqp4C|kH(NdpM$+7Z6@_EN&@~V6_DEVAR;&T_$ zIr~l!9W?>C=w=FU;ysA>T^agjvjT0DyP1^tr0~rrD)2>#3G_)-9O*FVzy9<8^q&8_ ze{IvVhfY@-5|vU&CFjGNq@xA1pN&=Ld`hO)&EJYz^){s69NdMpPj2Hc z^S+DT>*rIQy_-@`OKQwrwl3!#6E*jw#imwMK3aNsGs@bG` zeK-0yxdT)~#nF{N%W=-8M!v^RTPn8YHE(@UGFl;Wg*l`?ALTmjfp<%5sJBVh&`Zxk zqMH4}>Bq&l!MBC0w zkeG!#_z$)t>}_id!kw{LXnP{Mo~uvWJSl|vtEThjtxCd~^J}43BAa8E|6|HmHc}Ra zL+J2=P@JdIzzOne{B`|GJlsPFN!Jne{g<2l!30nWQF;{=uF~4pZqkxSx znm7L(w%d>a8n@a|be#ywcqEQDPt~HL*~w`CzftOFKn^PWu$}9JMnlfLotDGxznS}G z^OzO2sSH0)0nK~xoBHMEkG(VJ+NUS%mWyE}+Gx$FS=BAjhBC1bSkjV0~Z!R1!QGmnJcA2~))f zBc-W->zf(D-M-8NFLV4{G!^+OWYugCRR+JJ^*E(85#4Z`3UzLqLC@L>xuuwJ>DwJh zcyKws>Q2D#cozw24!|S4P*lgc0SAi@{{KCHEI#@22FWy7lE)&$ z_DbTpR2=P@zaJ)l4n=80>R_+ahOU0q#TP^h(D9BekooDz46eIR34|%YNVW!T7W@G0 zS_SBxt?AUq!`JwNx0}&zItp#EdqRpmTS=ZqHf#_-XAVO!HP5S zeQN~b5gDKzQ-e1L9mJ{DI>W9tHE}Q{bnpN`G7X7J|j(aE)FR8s5}_W{ge(&w*5A zz1{>$*Qrwzs0Q0DxQu;bb0E2F8&c)6-$%xc`F;h1@O0f=dQ9K~JYmgX_+m5OHRC5* z>fMAjNCeosh(kp1ZJvLu1iGNL7(~u5hw&;A`f*eXY;LT8TLICqA&xc{yz@=Gcr^`d1^`SKsxvHA&< zrQ%2?PC8NE0(Yo=UV`9j5ypc14%YsP9=)vPAO5K}1_Qwrxb%_=P8p8GLk1C49LFNo zyFCeY)MXPt&k)N?#yynzZUr2_;X36PaFq#b9c7O1b_K%St0VNaVXWY^h1=^obp6&$ zXkE@_MY^isyZCLGcY6xud+IbeZ2iq#SnEUra2M_gy#{CeKR~&GDXpVB4_%vHgr03M z0@che;uu4L;kQ(-zT-Hrn%QW!VG-yb>w&1p%h^9()7heher({@$6(Iw)v+h!=^IuC z+&N+mw!3X%T2dowGyBS1dGZb%>?YGZ(uck&g^*PxA1vz{W}qxn6NGk8rJE!Zv0U~a zYG%0ws7;&z;eL+aZW)6d_ffnP1stzAAdzpg_W~Z@;mQ+HlY&BH8~9+m9_1U^A)gHq z$ggMsjXtf0X2pj{PPd)1jebu_#4KQ+8<;bP8zP}STZ)!FKMZ<@zT<})@ks5+dKltd zV4wF;^pfIT^i~%hbk@1St#EnT=e#k}+9b->$#?M9oz){dR8|sE!yja(>3+Pxm}0%A z8L+KKoKEt{doV|itNh$eYRK3_dZ+kDr6>*g`={m?(?*v~@Jb4fVxrHs0vJefIo zq5;Lui-xp;WD+&v3k#B$d$Gz*dU4QH_YI-6zM zsyWuk1hM!r3wHc=hmL_3Xw+RuXX(6xr$R0eI)_i(TT`L8cPhG|+(gu;?Iou#dy-hw zJQ(g8g@&j1|Lgz!Pw)A^`lryq6rb&DAXbXaINM(rult~mH_D~sC7r3Dck3#^-CA%i zb_Lg;9Y9Isp|BtO3HwHztk|GRw8uu!hSM|P!t}??=xz)A``t?{^EHj9yx=*QtFS0k zMhTja4Z%sD9B`Am3U?MBCaKwST#sCWyi~LY&DwZOr@X*wgY|gJs{0)KS`Qm4@*yf` zIxT)xgg!ZAj9TVbgs(l5$1!gv$ja=mnB3r4u!U)O^Q~NB_j)Ed@=1W4sd8+GX2rL%lzXMYahJ2@ldDnABwg)~6i zf{j3(9S8m2lJuGvKdH(L0V4KgA&EXppj?;XJrY_#)J>=Ga=Y$Bdb$+aeD@ueNbdvd zVjJAOvI?85tp>*_>FE8P$z<0(8G5@8i);6s!uO0ffnDzcu(T~lwk}R+W78D;@7+w4 zJar!Q?9Jo$%S(A3r>#MI!+w(4^b8rcog&F$8mLQb3a_}(7Iod-idW7iNM%MTXc=a3 z8P*|0S4<{7Gt)q0LpFrU7?Bx(E_PUDbCk8s8Jy^h_uGO~0_=^txuqX}-n?W@Nn4(m{3NrLg7#CI<5{D`a zC|Dr_7O$!x_*@*-;P8`rtZ4?%_bmiju20uwR|?U~KH!uVKQysL3ZK{B3sP$9!M=YQ zc~ZXtW$n5R!)Hr4ZeIlO7;c9|)Ej=rXP_^(Gl=_@Dtsi!ie9yGoGh$M!K;F+@V9Bl z$Ta(-Ovw*7@;3bpIi;_PmbA};nDCkOuLKKNGB|^_5j#u1>+U1>*Tf(A;Qf<9B+untOnB3fa(@}af31bzJy`*J0!y**70&e*J;7)? zMq9Q&OU5{RSw$?(g)=Hsht#drvFdg)*Bn>?D4#NFa`tWNR z0s&hq(ps%d$Gs{-i^ikz&*?@`pH&3YFBYKL>a}PO*Vi14_9E)x_lQuwJ@e9Xf;4)a zgEL{pVAh)gnOd1-Rl5X5#Z-`|@5O0eL<|HwY(t`Ur%>axEOdPPTevu)ihb6lQAd=1 zV5>Krr~30cS<6=5d6^<0g}&AAga%T|7EBVg6tLC-i?`s~z)^g3&mobt(hz z$?Yeihi8%JVe0TXwGI@f-$4HNt4MN$DZ7YKp_W&?q-5-kQ1$3fm@F6#?Zw?x|D6Qf zaw`_UOK-(mjZu)cje7=c3Sr>uabC-xXtHNek6L-85r?;?lf|L0aaF@8(w&+`%1#Fp zkq=(L?Ei(WJyMA0oB5bYYvrlLO`?uwJ>%&}bo1?Y9A$nd7x0!{)u5JcIY=o~l~wQE zdWbW3PevOeMY+D!6D0D-f=IchP}k;+Q{#hU6#qIynZ5U=swbu}wiEfxu*)|yMGB$9 zGcoYRWCcW=t%1c;vl)0kflbDD;}gD`^bYnEE&d{o&YV$+q^lP}_4=96z95IHR2M`3 zyfeIG_wS*iuVW}GZZeX4a+S~2CgJ#%54c?zV?0vu-EvvYO_Y8vh-lhLv-4M0k*Lh` z{F#3%arskG$QFJIH+M{uL(_*TzKqJGeZ9C6aGX>TaSZ(YKe^6Zbi z)}WKTnOzoeG%C6b%!~T;YN(-hL9aCT9mQMcIt1= zBDA~F27QGbWVHAWerTtF8+MjZzl5ETW?d2aBc(`RYvJ-cv!${9Dpj1=xSH>{(4UYj zc{F@mfS!8OfId>u35M-iu$79v)-NXjKn0XUr4VQCY1 zp}M3`9)CAW;O!a+qMAkaLP`8%Ug(<=URJ%N^?<3Bh7P~C=xPeNHWinqMZG{H!4z5(TIpj5=t~H4Zr<7>+`Jj`~G$Q zv6k&P>pu5zUDu1vbo$DCOjai52O^n$I%B!oodKMDTpcs!-dOBCnTb`0&N|meUE5Bi2h!5$zv`m|>hE&DIQ}ncd2axw+6c3k=S45$EW&tx}WLtgB`(EL%P4 z(cMk#wb&Z?w`eqb!{i-NiQh!>HvFM}u7ONXwken!W->!&g)z4pQ<#t;TZwIpAKjJ% zoaK2-E}*Xijg1_{fBJUNje$$S(_}QeUm}E67@(r&`&L}bo^9NmN&2KG{SQ%9UqvL| zo?)X-gt4IyyP&Z@j~sGQgAG#eNkC96Gt>PbGkWC*?p@_|GRO5GT+48Rkl7l%q{P2h{3kbqdE)zp>^6yI z3_rNRaHqWg?0^1OeW;Fk-oWQ)F$DUp>ll`9P1`~Ug1^>5}$&_MMGoYiIxHBZJmr)58UC7>i*=k1}&pMmA;cBZ!*dE z7gFqxi(9yDSB&xPSqByU7>1M42F5NH| zE$3O{gQ==iPv;Z*Txem|+-($?*9GvyaSWST@fG~D<;b{if5n+Yl4$l8C3>i)n|7(H z!Hh{UocBZt@QaKm3p#IcGIcSeyu}mb&S^tJj|ZvMilRg_k19*$zyTs^B)8aa?Zx zP#8T^fsOsLlx;I|XTSO-&_YoIS$;kzW5Bj4J=_--yg%;3m%m_ zyfJ4k?av96b@04mfs?|-^j-TsxaR)`Dx2Ssm&)4I_XS5QU9`#VY8y<-;pw%#8fffj zOTOLL69NZrk-}ZEG$+vjy01K#zr z1-#G@f`g`P7`<61TC(X#Do=mcImlrj<&d`da zLHO#_SaN6YQR-DanMqMwN3Dp77Y4)4*DnG!g#{B$*|7ZRRxiX#C;^Mf| z`_-_x_$$#>8AP)L^_ooQN%F(e0!~&W!x-LzWM-`;GSN%$=6Yu;EwO^0PO%}&J{`v? zHK$;a#8BLSaV%crHj%9;uCclt>m%o4c4ZE{FIR&epLmFV;`5QAG9v9;P6=|X+nbM+Rz zJKmls$o&~b7Sop>ylAdVgh=$TnofVDf_ffe+UznB@7&9# zzOHjWSYnx z9c&eHxz1;aCP-X>QOjn4y{10sD<{)Tp{M`nKaMyY-%BO(UQ<;aFQVbl%DByxqpgeN z$=6prd{5C}Z_eEgY073y!lppHw!#(5RuRh0@@JM^>$ZD7xR*0BDI~hB0c6V#H9RA$ z#7PWpvCF>GME99~qw_bvVD^ry!6VH^{8pI>D1QH${Cs^J*7ds6Jd-x;cP-?W*eg(} z6-Ka$KBX4^v&r(2LhjVG$=t9JKRA<^Gjt@E&fO~2BS+$!IB`jvNagKM(e8RFh_e4o zyAInE_#)(?7rP4g2Te|`IhwA^bjQ=%M$zipS@hzn*nyo&-PQfn;#jgB}3`!M2?Y}vjjJ+sizMY z-{fXRR*E(*D&oGVtf#sSM$~v}F_rSzLU)dtPckE|=>Ggfk-@){I3;vFaUPmUvt{jY zNatgb%=184?%#_29Xa^#ZxP-aeHQD2r{TW&8|m-X0^-&cB@(;tpf+Rjge)){QvR=& zWcVH@yNu>R=hD&WHh#}FKmMZlU3!=(Z<;%Jg-F1WH)(Wl(+2o@D-)DU?Qr?g={UQ) zkY1a4s-$#@9oP3Ns^rN!HBM)lB$Uo&X! zGd(h7{A}jUpDW~{l?EK8r~zi`OP`V(FEa!8o#-^pB9ltT0F7ZS?|Pd47+`v8!dB+6uFiWt2iwGDUFyYA~Ixaqm_J(eQ z%LTh&>GB3Hj>e#CHVa0I<`4|F5LOlDXL15#WjN0bIcfL zXfmuK*@a8(g2E-}l_W{lpsfcA|I_FB1Ln@FBelVeOvU-X%)iH~%&MgEB8LYKkQH(e zR=7Q79%V6XVZ|jVFb!r{(;>Ll%?4+C|KsK^EQUuCTY-5~$w^FK%3G>d;!?k>T*s2p zs6Ao<8nvWj!v%M``NnK`Aumlnbxz~MidA^SuS&LKUIlYoQntYK99_IwF`BRJJcUZ{ z#^CDd)~MBMj(!VIfZ}Fh#w;^tlhxk>e~H6c+3(5VmNR5LKNH^`RYYUEDDZR@p5exE zd^#)~me%Kh$)ptK$HPq`LxGU;ets{g{@M(=5$nDbOW|RkmXHJ1 zOLAlXG2Yu*=Iy0>5I?;HCg!eY4o2C-;Kd2BmGh+i71spK>M`8Ar8SVOPx=NM86BrWd+1>1#t#;3JWXt2dJckAI=eZfhLYp@+kzs^QkK z$>39(LJR^+aM06ueB8E@`JCfI_tlK20Shd-g-IjuvVjF&e)Wh{W^ZPeO{uqgS~rWE zS*C=WNSL4%HUj?Y4z7AGPt?jCX0h?zU$5cuIxcE>#1{ePIw5lz?w1D)EW-GfC>q zYBGP)M8?Es3KN&E!7cQ4w_90kg9}`9=!1;S)a~XQSpP+d{WY!uT>P{c&)mPFdtlPdoUEIDNoM}IxbqWcoB(P0O3 zkd1qXt6P7Oj73spZki)JmAFQ`i>ZhTeB==;1kQ=%X=rY(hxHAES=s%XIQ_B)54C;& z(;rIwXAbKfn@$XzeL&tP1;4fH@O*zKuKaV5Z#Xgn<%9lV?9LnbNzfN~hcCpr9b=7fxz~J6>OUkd)cR;_Tc9%yN@W%!NdM#_U%ZD&NsX)yh0tQ)CBWIXOh> z%p~+u0q#QEX-2W$g_Im=#x<(#_{&$C|Ew4wu1;GF{fbeja;$|J*P8;0cQYVxXcUBB z7z`83N1<!;(G@RGvF4xDnLDC%uChQ5yQ5{$X?DGM^id5a zG*p2V_dTJDZxo}dMiy>;attSIDnc*K`84v6H|h0Lf`778@rbq_9GJO;>+-PR9op}p ze*Scn_8-E(R?!fhyQ4>KhiLLI_hi7EJ?89;ZEa{gjgsm59oU!?BxISXu+4M;q=d`! z#uW|FjT>0I5=ZP%8G>nbf^JP`?=VYhT}>4)7p^Ku=f1)Jwl3FxWKfP ze1yaMe~KCLC^%&IhImyOqP;{nHIE*`O72j>fPZn+^z{_D9<`aNeWps{HszD2xp`dr zW?zBD{#|6UY$APQc7yy(>ZU)(Xkv}U07QI?VAtylcj>v&)XiKKtKL5(J#DKY`%x24 zObIX9yYUK!uGipa%VuC-zXE^Y+giv!wE};S*oNB&H{ioXLG1Xt2;M5`J}){{DYB{0 z0|(1ia(H4Dw&;f9W^F${t!^ATP0k`lZ?cG@l{q8Ii-hvHKj3a?i!t)s2^=uQe}g7) zJ0rRnW>h2-b*F^0KhgrxGa}(}?@*XGRbYD>$}*87oM6w!xlH!vk>ax+gq+RkWF{Hp zQ1PW;uIp(N=9+1v!l5*xAF>Qoyrl$=Ni-)dcmbAGMG;riizIp1NA7h>Jh$wlBERHj zHQzGJk=Hr=myQv#XZ=-peEhMMI>~wRX)DXI^o0d<#k>_wTs{g1Vp>6AVmtF~_!;my za1p$&%!g^p`Iy{s7u47Kq3h#cT-)TE^ql`BGV|jSI)468ve90bN*9Xgi)L+H*zg0d zW*xx?KT2?^wE{n;J%U#h{PS<=N#1)?2QS+%#)s9}c%J1@a*rAn3-{}Ud2$%jp@u!~ zQ+RIXRGt^OWIudHajpD%(LSwsJf&d9&wVW9eOcPTAzdkUOHeJH^iz_F5q_uT1QV?P zn@T@k{6~l2cL-XYO-D3D;lWqNSm9{RO3YwTVtY70qHil5)HMqiLlr;H)eT0zyGA5^ z2Ix1rFHAxFJKQnJ6K^b667jrHnvofp-1?8Uj!UEQX|nuNt47?lX(c~W zJAg0kDda02JmYrG&ZWK7(XnIXM+$eQgJkffH2i221qZD=sQ0@e7$!;v&yg!YJ^nqhzxNm}6wDzf_o=h3 zh3nvTU>Wno)t#(8Fd6iFt>D-FcjPuar^%XRFMlt<t@pY^)ncSJFyAIQ((|yUC!(6Z=!g9A`2$E>_zV{P^KYGRDLBh zQH32)7dRGr_YZ@LJqgVDRAtx|eFCyZ+tIz}WZ6IW+Tm+WCwbvzDX`w|z`K_e;=Qdw zeMd8i|9P0i9&;ssGQvq+$vLw7dIT9{VMX9+ICE@`7aaIK1@rk%2X)=o)Sa{wJ5< zkbHSCJ?9OZW-Gvwrwd`};WKd9-2_+t%An-|s$8{i61`*ZPgA2c$pZ85G~vP*e6JXc z!^6DM)xeYW7(Rxbw^szy&4Ow8%5*wOZ3KE|2;AT8i6AS!LI%NdwschxNROOATD^;l zbB7dgShjR=3Qd0h z24>5H1Pd`tqaz$&tpCV`ut)Fbc}IIS`Rm4^f=gEA?(8v%fb7^Sm^fe zBQNw7OV%h)B_j`wWhZmNa5eG{WQ7d}qlZ^S&qM{#E~&=e%y?Ao$%uM518?2J)FyDP`h9)^Zdd(BvwyJp?NddzPm|}!9n&3t%pDzDF~CwX7mm@ zurA{B_(3sZT1DnwxuNCTBCTiQ$E)s+*gB# zC!p2NJnDUOJ6LNt(#@r}NMv_9yfK-E-a#!ybI81U zCDO=@#C@(59C~%w0jrA~vp)nT{!t-qA_?5SdMFNlGmh;!_!?4uMzAfNE7-0BV_35% zk&qo}!2aL<`M>q2Sm3bVxt_q3K2;^0r!!k@{~631ilMK0GYr{LBX&y&Aj_@9WZe)S zE`_rdGL-c}w&4c%)9W8o8G8@vZcKyW#?|Db*;!_0?PHR0W<7mbnLvuO*U+fFgXyvZ zsh~Nun=DWrOk_u?lcH-2n9kN0wndk|k>;V5wyOBiuIEcBvrN$qL>iL73jub$#aW_F z5uVJ=wfmuAaF_UKcm-D&&v1QPyv0vVZZNNwo+Mv4|73m?XoKgfg)C@fLjWrzZ*3UmkN|dw@Hr<^$6;%-Iw^3tXt&+>8$DdIl!g`N ziRTPD#Kx#gv-6%FC2g^iY`n%fU`&(BGj=5&rOt;kH=7cha zT}Qbg&UoQYRvQ%$L!j-4yUhEgo$yT z5u4Nd0is)v!SBrPaM}Pgu7DscEdFVQH9+O~t zXg;@Sn;Ls-rW89S%^4h@egY$vYfMpYz3A4wW2B{3ju=;77wxF|OSYNnV7{l7;9;Ie z#|wUq-RoC!&QsRf9@JjIy^H+9bQxSH5^3Ipet(#^ z8|I4pqBFQbGo847FV2>Hc^pBGy_i9?zHQVz}tx{%$7Y)I3JdcNsI{ z;wm_POwe|@Y~vR3O2o$06o<44xuyd9G#dt0pVy&7b-WoS&*aMCxW9ICIx(o<%SM5B%Na!IY*$iBt{jD0{S zzKbcK(}tDxQ)1XRMiyqF>D0AHSJn>sE`^ zUC@?>mor62g?@`pB@>DqNkPVG=J(=w?w5QDDSol(T1ul66Y+8_$((8o|NKSl&exyt z&9U#~_3~->sCz$G{=t&GJ)qC9zaJCJ=qecfLzgD@M}p&-SQ0VI3Er-DBta%wO#Op7 zB)PEHW47R!M!du%^+2Te62r4ln@88*wxgmF$-aoP8b(f4F%5(wLtP%8d z^%Crxj?<9U-%a{Of=>8n1F`)*n7#-&#kNkI1uylL;FO27ko#OkMjIzHotB}X<>LcU z`BCKm_CNovKZdohh1_#%UUlh2K6|`77vgaSUw^5`qG13{l1$FF=~KM zuFt_YFN6+1eUb^@s0S;*$*m!{0~r-Muu6hed|)H6lsoYEqgu!c*iIJzmnnN`!W$PAT3w7&8v9CEpcn+SjC_FZm9LlXULxtnxLvtZ=pCSnnMnWBeq z?(u73WQsjtm(G50)fxte7py1KQ_JwJ?I)b~%9KBM_B!nEJp(%&zA{E$=di@f688qr z!KK1(L6ve8(@qleRnDN%?k>K5BU|hnZB7p#zd}sh8p*?JJ;eQ}7FMQN(rkYjHce7i zU|Ag}3qr)yy?X~8J^Bzg2Fco5_|`E&kT{3|e#J0_;7f!Fs;nMNW+usqLjG5?vB6GE~aR`&zr7JhU=@;iA#!zG|I2#cDXJfx7km0?wSlV_SRrs+nnKG^j-+6 z8jW{G|A4Wf*FeeJ4s{HdBSzoHc3V?^LCbVL+r67BmUxfuB?<7gE(`ki4a}KQ z_eE4%8LE$#Q9bowoLW2#hbXGz(l$4k6fIOwT?Pog8F?~oBhAyEBU1BU&{y>LcM^}Jr1 zMrXm?yE0IVsc`>*CdlbIu+3LmX#d+nI&r}_(qiPtzicbSr(TvXdXPr6FP_JwOV{vi!4zz) zdP_Yv{VBQOkd1}sUHMa8i}{p~PPGcjHm|%?FK11>0kCo`_aR?9A zb)ja_SiW|+8$ao{5xjk&K)UOq>4Ed6Xk_`1u55J`_;M=z?Wj{!eWnpjijPOR!vgc@ zy%cW#C(E0TD#PAemi(ZIa3Lo-nuce_VECP6+Ulo|FD_5!|FkXV_kX51bK!V;@4Aq& zKHnCDpIi~{GfFsQx(b@s{z5-1Q##`7Sn|A70i!nQlJEa^;TaQK+f3#Jeyrm5h2FjO`pl(6&GObo=#3; zn1Xmm>=aav4x*afmfYiz$rx_qj@vtz;hp|jn9H8v|5RA=p2Pa-poyQ*TJ#3Yx1K;q zjHG|=`~$@+e@W2lHTc{8B#u;hj0O=osFTo(1Kp$eg773Ti}#1JN`c*3EG_m^*$mH} zrBLM@;Ni*XBsFpy$YyWF9b@ib!TxI)b^ki(s@4$8{pN7^a0c~Oet|W9_Bg>~Hc=em z&MN$9B4IPM@l>z`T-8;hDl7M+#62;|-8TswJ%qf;Z@ciss`pe*<{4=U%7TtJ%f!ZN z+i_t}3hr>4jxk&EsNGpt{AlELLwWK|A3lu+;mnxXGuU=1h2qlX~rO!F^qR zQ%o^7XKv?Je3W>Zjmi-4%a3VKQ0C?Ldr|gCHPtooz%^|W__|>@?z#MznpG~Q*Bt`D zxpNW7I##nW zzG)VHn6?&&%nU(9IaZd4J8Mv^@ z9%HXOBKjAH5nfzDg4!fW`Ltv7O2q}{^1e8n91;my5pQVE>ffZ=D}zQf&Jo)hO%-=4 zXpyl~gJ|*xDY|-jG1;j;k~Z3oCQ0>?Tys}4mv}-Ore9x+*XD=OS3ge(caHgF#i&kl z#jT6Dgm{wct2l9#hACPoC5v`VoKL&sGT~jv9BPoDM4#KNyUYu+=p%N$mXB=>&C_>Kqn~07llYEJvg=`sV={h_?;x6{3BtYO z8`*LA2RZ*!8R`bl?qBoyKlRzY$H^{X&SJrZ+9~ z-ND^{<3^Vc`pArrekV?FKVk<0ho(90tEg>K3}~HF;{HhuhL5`xXu6CEQJS8_%$;+P z7J2HC*wYM&kKIRKCH`dO#~&BDx+l=j@<+g+!-D+ylK|C;pS=HJVl)u2>V4AY05?H{??yS}h)!e(;)dkpt>hB?T1?I3;2 zl1Ofn9aLX>P4_-}!@SS*;QU?=5eqYgU1fU@c{buD3C!e)LT5ZZYa5DsRUXt;<1)xO zE`@C=CsBD#vG{`UHz|t)Npr6rVoN;HOe-a`vPOYaN*pE0s$jn{pXk)RCnMI*A{`bv zkor#WjeaU4Q&oY~&Q}u{0<%d)ays`rSwzZw1&{Eyr99hafUgtFXw|0}Vt#cP9lSJ> z(@RYdyj*3RPT6`+M%ZzT$*&P@?@uQk=>zn3lsSxguL0eQo{=AGwV5hMPqOJ?CAVU{ zE%#c3#kZYCG(0g*B!4METxsG&?KegUO#2_iSj-cfDKf&o>nRzyt+%*lx z`YGo8c_}rlQVbxI+zZH!!8zCLA{Ck2<21?JpAp<#s|O(Way=;>)P$=qpC<19Q!x8+ z4ed$VO7@4}ELn5cRXp;E2rCS=Y2AP=`INn!Df=poM}8^tuOHjONw;*G9x)Z2ye(kQ zqoFWA(g5S%gyW)tYFa7jgPO6rB=PfoyE6yEh*iK=A{!LJOO@91gRU6iiPCBuQENm8 zCC_0p{f5~(m9}&9;(johdq<&FXfU*&?xTL{+xQpR%KV(PQu=*d4wE`4k8B-2pI2(V z!Y4(V@+&75Q!5`8>R3Jl1AE3`@{QB@EUFcL4Jf1136>H2E~fjQZXgT&hr`42sdU~U z1Mn}Zq1HQVQ0@JA{8wO!9UmxpW%LOz&OVFh+x)qQCCf>1#ZU6^-2{@oHIp7TtfAk^ zUCARz#citoe8P;&__48{T&YO}t?k8B?sqJiGy4`@;2?;P-45zPd>V`)_?Cbz2rO#$rsXT({PIqGHSrUKHNV6yp9FT* zkzNW)_n9-FF3}&+Rdkz$0_tqYpj+3SK&DTduY9l@q)%Ib_tsL3KU9K;JF`*MMerL| zTXSm5zN5|3RaiOyFrF`v=5^lfz(lT&ld?WBes+%5BnS&}_gT4oUm-K1rG7gs(ko<>>Xwdu^GF`TblX@Uz zhOVBBj|x_x*Y*+Iv@IccW5GvU*;PgY?mZ-}*)pVk*E!;?G@9?;bdwewo=A?yTou(T z4p2GW-*`gsC!C>EP&>c~)K>G1+Rw|>&|@Jt?s*>*)bjU#_pkrfpo0Q;WbcYd%u!R~ z)dn-Lma&4e`$n*Po(gQp9svJYJ?Ou4fLQ6jB6hxI+=|-*ujPCRwTaUwpEsSQDr;Wj z#vF>fn1f``QNbI?3@0w~KgrP-M@1Red1lLW8P+d41H|K8;3^wKUJX4?D{E(CNTELs z_a4I8<}`>6#OpBPr8*(!1f9Xgmqa0`h-9m^G7aV{;lkSwoMyf>7B~&Z9Jfry;K*OQ z8mCC;wI7Gi|HxC98J6I&EFAt0Gezh7m5^~b3f}2T!!jXX=h}$~GR4`LCOt`~&L_Um zhFj&>_InHL9&rRtM6F@Sq)NiYifOovI!&MH1f_GVo78sBf-B!oQ?>s0;P@#Xj2{iq8)>G* zJ)n-5Ezst6oYz6w%qIFQ4LfP(#on9jv-X@9Sz({{uFBe)35iJ6D(F-;*J35u>0h;5KKUS`jdI%aRjSU9tGJI>iV5kB3DXWRn_5o9|QZy1+xgMVW|$RsU$e9bHswb<3Zlm2b6aQ!%c?>h2?EF75N!7uzzHi;gXNwk`l~OK zUi@`j*z=m;F$*)%!nQoLyQ|5kI4SZ=mxqvKt5taCt`jWl9}Jh)G4QZiSJc!TOg`sX zqrCcax~Nr;{r7trJJm;lt^P8SU$0ThZL%MUJO4am97k$HsH8XOg*w906Y@em(Z@xT zW6{&z1g$%s{y+Mk)q!V$Pj`f4TVfz^(JpZ4=%kaqv&j8{xun-E25O?Sz|?OH9T&D1 z*X5@|x8UE4wS9r5>pv6efEDmNQjV2S%_BeiFLUFf>ly2R?`deY3tO3>!ny^>vHuD$ zF{{U~BFDC;(905ju=s5!F_`HMi@z+x+Go>)A*8=B#3#Kbb-IhU|)RukR0b6Do!I9WckA8(PJ@kbRc+FZN#xc){0|tCTuv@23_;d;=`h8z!y$tEshI(+M<2b zXX^mtC(J)3SssjL^)fd2wFz7FHJSXpFo8eh8ies6&u7VPVQvTairNc@z`dc{X=rT# z8M&c}EZ%z#&P=!hO<{^`(ijRAf4`AG4lOXQLLM??6mW@x7K)xy#zJ7^MULzwCYS#a zU43bGENZe*n>%QObU0QHorAt_+K7zQPa2>ahL4{#5$E#@*cFq9fzc8_{(aqWxHTq* zSC0;7CYbt>-`7VFvB2$oqGZ4yBgbK0#ZK_@9|@0~V&Lia!>~g61Vo>8Wpz#*2epxl z1^!Vh@e?g&?Q1$AHFOGi=bGUjXESVdzE50JgnGV*U{Q~7!w|d-&0)jnvO5LPcO?q! zpDXZ}wmg7AOKviq&Yk3Mr7C$*7elM&?n5mHIre3d6Wek@k0uMp5!tLXvSOAI zEO|Zx9_LPF%67P+{3msqaDRbtQ%mF|H!K#7+cTLpDK>-}&l%VyQ&R2Y!XpVy{QN9Cz|Z8 zj6^tUF2;KF;CDQTWR=5A*v3optj|krJgyM~<{e3}nLEiA?qA7T)i31c44O{Y8`pq& z_6FAGfHoLDn!x-1IRIDBJ*0_smiViyhGyG8p_2E9WBQ%?|LuPy{xg65dRPKiw^tKJ zXDuWBXcRin71Qm?_i@I}a`AghU9_GQfCt=mlk)9;%&iY9!kj1pUV<)D>-B2x;v5O? zis58hGdT(MuLP2;ea#Ge;S8rK-UyRsb<>**&qI}0H*ja3(WeR>jMm6lZqO|W z)V$x#jqeX=S+YP$3;}fWuU4+#oBJO2ahr(R-~PfFteu0k z<6|MP)rQJ#Zs(ky&LIOr_R_cqT6jsniP;;`0~N$D_t6HZV7@&n6f&Q5pt(AeC{>8bX2&S3+OC5OhL5B>#8osm!~*U6 zQmCi!I{UyRuxa@re)q)&6W5+V!?ZS(44z91APaRCg$SJ3U{cE@Fj^y&fqRjI)l{kuh%yDi1|hg)!!RW0VVCm=se4WEyXAacHwAau)p`YBbGCOp>= ztxwfPToFaARG*TsZYx3Wmo(n!brjf50@JBUl6ZzBl68h5kojvYJ!|UEZH?_0U%j5g zJ)WwAmp`0D1NktRnJ9+9;xaIEut1A!X&R;7O|Rwv8CBCLdK%S8bKhmq28$8c=_|&C$u7=l8N66L zpgr1d&J@98){@A0>-Ex@$O|N7*8}dUv{7VcLzCY9#6oBSKDf-;{8*v!0gw4elxy;^Tn(cZ6bHf*5 z$8|+qJVT8%>^8tZB^`9b<90~-90rb`H&KJ3c6dy#pg3&jOI9Uc-D6& zU}GU>e<`HXij}e7J{qG_Uy_r18Hl$|!4)nhM4@CZ2HP-TVjs#yzL6r_(HBI=PM&p$ zT!I4$cKqXyvQ+Yw0=hrQWtM4{l1A}j?s;P{b4te&&sr#eyYTszvRFcW8f)mX&e3pe zcr7{3|J|MDA!jtxS%moiUTCP44hxQENm9O2x9FHC>`)W@2ZMz5@$3VcHGdS+iciQUq zgkDX{#XF0;=vq!5Qb>%bdzcB|Cg`cV^`fx)o*^c#4#qEK6|jAtKP*a`DRAkKc>k;t zZ`8OX8uclJ6e^g)_3#(eB>f;ym9@EKLxGujcq?~hbujZrs}Cv;Mlk`k7pZ~eago}D zB$|0-BN3U{v+6Ul$sK24iT+@=&vP}aTyT@xyf{YljK&e|^r5h?%oNIob(C!QGnf~- ze<8A~b4cBy=P>c}5Vl+JsE-a>#q@kIhshJhiSzE((srNA+=;wp*udE1qF(`M7y8BS zJKYXuFUqOULcwcAPw_eK$N8s{E79xZb+pQN#%ZZJ^kZNR<9R)S8`+mH4j8k9ocei~ zMp>2cO}9PyDX%Iq{qS_~i+w>-Q?C(4<=Jp-`DIue;SaldHRwn2A#jPXf@udA;Ly-S zOudp1-hD@)#WaLEtjo=uw+U>L%`Pcrz-?H`CK5U+L~^$y|5MSu*zLZ`veeR46XX=Qdg!fb-Fb?D4f< z#QY0c(qC~Foc+C-2l*;+>)}$eZ=na)7S?eGHU-f}FKc$`16wx2Z8Cc)-;}RyG~(YY z+VK|$>G3x*+{pbTcj}FT+7&}co5owo%0LtPoB^c>-R-w>G#v&2d{eMGhjVEn;4B8DPbxNj&< zGk3<038iScJ)SQ8EKQ3(2GEdl7i|9`@IQJziEQgR+!zd`w84Nr^iaWe!&UUMhCV5^ zm4U9_1oHV-1~am{k>pL#hq#|lXhrQ@QH|D7lGpf!%x=uaE{AM}K2;X0>UrVlmI?SR zt4iFSUe0WZ(8NWWr^(N=8&GqF5>K}&@wY9?F}Jf3->X#M-PmumpedCmjZGCTs)&NC zBNo7q!F!3XPqZlfuQsXuWx@Fdq~dm&PxMX7DX<$Ch^qS+@I!SR`1zITTy2mm<|gT* z^!9SHj;STXOZ>=Z^*F3fwLuHZ22qQ2Ga2MK4U7+Yk`3Y%GPCG8d84}?T}&t7y~;{5 zGcytGeih*9(CxV7tpq7cy-sLcDCuajreluz;KE%aY1Ep{%%j6LsF~!A?7eQZDC{MT z@ow~Ke7C?e+fAPCilU#6cr$-LjiO#!BWZkKI*F)?Y*ZdES1?moCQcF5x{m#A2(f4LMgOLtHj1R^cdAo4Rkn>b-eH#5RS{dzbdi{T-y?HcM;oCn<=41>BMNx(*!`at$ z?`;UFG^uEiL`pOdDv4yC6Cy%MhL9m5&fZ4@rAbMVCQ*teN~u&*?|I&}p0%FuyWaKt zJ&L2A8Ka{!2xo`~~BAk5wM)sPsgy3Ij6z7|=k(R1H z!db=ivPx#|Y(ta{SMH`R^n?v35#A=;zSp9hscs%S`e!FSH$H)N)32h(7wll({hMhy zx`$^uW9mbrF6p{d-STRGgCch|Y)O6hcT z%|g~o-;d2TcP0BwKcniZT>i6f$M~m4?HRcp0%89+hhC5*#ZO7nV*#CM=}(h6eXkkwO|gwt`B}rs`#fS-Mw_w{ zCmYG4GI6H8aUxSYYQV_eHfJkp;@L_kZH{h9q;KDv%F25_Vpl{iX1*U7q3z9`*v4oR z%bjZsEX5*TS`M!`#fVA?HG_6HnNcxwcJI|z{>|?NXpy4Nlxa5bHwpMO75b1(*V!f1 ztZ!kx^AqS3v-0`nU$$Y;MJ0iD#W=zFnm8`^%q(WeT%DVjvjF4g$q0<+XbN6U=;h9> z8^SR=#8~!sSKYhslPrziEV5i970X=9Ur%s*Azvh~kF}nf$HHGC)SdURnJIOwy)Q=} zIk$~Jzqo~aSha%lvrcBE-MH1XW4_|I`kw14zJc^y$ zV8%3^S;d+=>(N8w)oH2yxpd5|-~Y`&{D14uz#S<;x=`b;z4;QeT|b1$IIL<}?BB;0 z=1MchLjS3id-B<;2OX^P*L_odjBH}iIlTmoU?h*DlW1M_hhm1N_e)~LbzU%@<>4gg; z*(v0FIGZr_OVgRz75~_g*Gt*U?QUH6Q6o-i)>QU44zk&ImAK-iiR{vBKj!OaeeU6^ zvuxg^By#DqJ6H3Lg<~`%Klw@#{Cr#*IdVkr7Y7^FP ztqnh^t%v=wsfXO%NVA32sjP-mDE-S@h`Mi7pbr)Ku$w#xcVp;2_e4UAgYFovUBkgpN z)L*?%vYP~4qMNt?Lz20Re^zlz<8BE2gW&hx^hZ!tKyW=Z|4^ zSY3KH)AAykRCYRYkC%$mCwG`IbrGw%pqr;yqZ{YxI#oXHx9Ygi5A!>tbM7|3``Ql9 zWMu;#HK&+WWe133bsbZq^qCPgRj~YNevFB{R6{SUKFc1Esb-VJLTNkmGUoT-Ka_u$ z#ro)XGN<{K^fb?SR(?)8v)1VvKbDK&J~Vuz|0z3Qs8taA!+r;yI6aO_7Cv`k`p@as z(>WaUjb(PqL~>3&ub2^&bF|7=+R`I_9McxgvuxGB&Abk)qzeuu()${X7&na=_VCRj zS|jZ`=XbA$v#%KB#;Sx7ug&Sq>+UmHHZOv^Y^cCI+AqibsuZX9ll|PB^68e65x<$J z>7LA~EiJ50+%oR>?6qW7FtQc#g~W-!fhlU}q;I?xW?O$K=E4-?xl4Y{oJZ?R#&yjX z`rZ7U^qAdd^mwsRI_yFqeLL(u;r`Zf-SlB@B-(({?u%s;PF*0kx27=(|KgeR_pdM+ zZ%w)1RwkBv#A<1kkxJTDv4Oe#z>DefmS7(DhS8tno&V?gyGU_~y=Ch@Ibl}sE@n>S zRkF}lm5FoktJ~$J#T`Bo&RJJSbIVJf@ZT7TaH2nvKI`h7fcyu1MTscxO8eHj<=(3pKh_h**BLOM{neT4 zzCo78dtGUFem8cdXXE*D1>VdERkl>gUjW$g3* z@`rawLBz;(&ZxkHyA?B!d$RukpK7aR_HDOf6q`IaO)G1D^^bnW%dj4|*BOws6&-ZI z-k*$nsSKCtQOy;%SAg}q0y?X*uI|Uc48eSQF7v#@mfNsV8I4}o;W~$Dma|$fagpZ7 zI9JP~tgp#*OkRJDRjYJl%NAC%>3%agIk8s$2i<*~(Y;3D8J;cw>gaK`AsHKdq1v6prGO_v8&1etj zFvB-uY2R;K_~XptNk9+QnKfLmyRbBzewS~?&y#%04F;UxD(2d9pKN;Q-C9$HddxUl zBkVi9?er)sacUpuGA)MeY97bty!p+z6%ppE>~8vM-6FyIn^H_t*8$pguPQw<5W$28 zT;`-7y=2X@hWR^;47jL?D%{V*t1Kf=`ddEEYi8=a?y>70DbasB`WU~#`>d59lsoU= z$DK`E#(2KvvunGvN%oN;#&6vM<}@41_>FXOhChUy$+26Rj}Q9!Py7|wxWhO3Ewg4> z_G}ny8GQH&o$e~bgd5rlb0Veq_li=vp=UbWx8y&ZYKsMOzuG>vn{$_wXNHdW=19Ik#3549&JPmY z_}R|NRm3q#Pc&%#0g-16thukhQiGsh3uh$IA&M{YEfmx54u}QINC50-7ehTX98CFr#!oWJR$cV$8eX_3%N;6 zPlWoNQhYNeg9}kow0v2yi*Xq^!*4iwi%xWWO2iM3#3}Gv}R4nNB|i=6b*!K@V9fxbQ|G%)@Y`w>EOzjV~WKc(25&l_qg2 z6`#4D(r~6bTbLWVTBzwQ$mX73Ud-9iARdzht+l3~+r%j&bak*)%NwOM9zi za<^8|f~9T;1@F#Bvgb$L7{N9HU;eiRr}NX5d-#1et6?>r-hKT&d*HW#(-af-8-*;r z(npFu8orKk-gAyF<^N@iaxSwMlkC{n2L|fa=h^U^d;hWL=M1r3AN4GU%Hvp*v*RuG zCeIMGCIR=yWhsfjEXN-^bsl@|>2WqHUy)InG{C)oJ5x}vo5JL5(c{Ku$q8#?DjT&< zmizuok*+h=;4V#{$^3H@^8VLZa1zlf+}NMjxCM1@xf4oDxo=Y@a!Kq|dinI1WN%}U|TWE5R-4*&OZr)s*ZEQW2-)TjLBl|Nli38q1v3i z{8x*f@vfDqZ75}^B5~G&$L9xRH!=@A*D%)-gc`CKNsQUOgGKH4tbvv%vnw)-o!%o^ zry$kBy_^!mC0e#|PxsghzHKbzbT)lqBNJQMfGRCko_B!V{AC_HdOv~jzkh@Y`?#DY zQn%?PQ8Vd4t^2Im&g-nCmK$I5lL{OEWU}DzYehlV4HG7Bz?yCiZQu^bPNbE#l`yF~ z#@t59KdjY$jvLC7;-dWo+`NA9R_`>t)_q&Ard#GTtYDZNG9dlNnW-CO6m zvvuCwnU@xo84_n!FY@JN_ZkT2&tUd?tTS0&sLj3Yk0P`0HPKZ+T$xJi zE3EzMZ}cY5IQG^49Cn980jDQDO(6YPhCOrS6#J>zf-9_VqksPjnge`m|0$NLMdFq<8qdLhr4eh`5JwxterJ~)e8(=2F_UNV8vuQHligg+2 zWbartaaYSb*r)AVnd7g%Gj1CvGG|M!qx2^o+Gz#NKPS-Qk2{dY87dwXYTUMQG7k@v zJ3SXz!&!Np#-##QDT<|sJo4%s+v0>CuoJnDY(IC!EuPa=SEnyrP35|}^|&dwe=+0# z-D6HiRd4~OET_9ijkVig$sL|;T5Fc=%PbsC=PzG>irY0fMR28WHvh@R4%Yt3O?G_E zB6@H0B>LB^V7}^v*^IngI@_^5jw>-*W;yu$A#+A0!t!5wl#nB!Nn+)DNpMA#Ab83R z&L-m-N2gEX-p&?qu1!0*!rzYE-e5hhhrRP3^Ow_G4Q?D$O2cBF;L3eZ<_bgzxeJPd zIP2rAUPBE3-(D5&FKfr2(K?2#8~+IF?n{D)axD8n*MmeDNzp1tTZo(IWX?8z3GtbB ziyVP}!hA+GVI~j71i4FDIu*s%dhQ>?zoK=We`>WReKoe29>{#dk1cq}?NkZn;ufPo zI!v6CDTrlDwSpPrx=r-g*SG1rMzOTnw`26Vc08uYWtjcOW9pVs0hlw0v zhQ%`Y14i|1mMRd{d}rD@tD6pcD$LgP7GUS3qkOGbQ#qMkPV~-^*(~?1fElr=!v52* z=#*V@%!rVuexZo)UrXolrB!dU&1WmQ=|dx|jcg}*<#?fPWc^CoP<9E)vbe_Q{oG0) zQ0->FJxb(uWyY|oT^aN^Ek`a|*N(fu9p{e5-seUgj=k3!KBpvLAXpxn!F*T6y%kP{%LSS(g=XtNu095g`<#DSeCKt#2XG6;4cMN+mPx zwD26yZ3TP()(vjkkt7m6Q;xoUDwn?MqscN}&)7IEp$EcJH)ff06w@%Sl+n38Qu{!| zh--Zm!k%9A5WdWO%-_7?6IVRsRHs2rVaKgc;?^6AFm=8g=@mjBjtN{0ePYHEzSzfz zIx%g+r9Ky7(rQZCol}(ge#07cmZ~dFKB?1k9YPLHm$=~hv@rgUfupSSx`>O8d;m~~xkdKqy*xWTdw!?HRz3O{E*~KL@ ze|B!8t6@m)}_^#52xv zJ!`e;MMt!`n{l?>nZdD~b*r*K?5C!HU1r7|2@q;1a#yl+<#Kk?bm9CJBjl%Eh^sv* zIY>LzmC|atS$xGQL5!x0B-7kDiCk*ey=$>*`!qk@%qd{F2+Q^y2;qx~EN& znc$Gl&I!6f`}Kchm+j1Bd|ie8*pn8{&z<55V>7t}m}YJaYxIhK2PLyB_@+)w{GDo4+~8_2#D zna+$ZG$N;$qzd=t9o*WnW&A0t#&E_@B3O;P_KcfHDifc4Snyn~pZ#Fx$G-S>92%n) z$S)-W?!t6GX4Pz2_Qdg{+`_w>f*QqH^zn-uY4JxJnEEbjGOKzFcOWQ^CD(Pi3lidj zjcddN7k0HUL%%%eV;wqzK`Rs5ZEgyq*L;Vb9a_jR$sKi{i;r{57D-8)Z!`TR(kLqO zU)QolV#1b4T3Tqw@&EZu5U~{T^4qw1i~pus-d_8>X7S8?gc zwE)d-i9+G+MgHshKP`ay|62f3M7RL>UoPPP_Wno%&kX+e_x~pR&HsHV_fNv_k5ZAm zkS9G8o|C=8EScr+Qs^Cen#4*{k2DCqo4$G;8m{E#?! z%N_&GFZQUNp~%~I^A@FR1dV zy%b6FBt7zTyCYG${)Sp(^8_Af97Bs~T|6ZVF=WRIu*{;HGOCRL%Po$0?)N_2x%eaM ztUN%ZwA0|A_F1&3`HE?Wm8skrlGLyS4~!=&;py&M$i(Ur=a%!-t#RX_xB4dpwVr~Z zmO7v$$C2loPE)B;vT$BYKrI}XNo|yHq_o#DIPdWSUi*PKs_WSs_~ujv=lpA7x^o@W zMVx^#&(GtpaEe@a`%c-d^nnj|vcP>s5x!faN{SevcUkaQR66_=+j3$^Ovhw+nn$6Q zYAX~BPk^W1OR&?Y5j%%M;PN63jJ_{|(Vs5Dc{?R+&#l1+`VH_sgTXu2B4qdaIxuKTAAu7-sXWxb4T&$ z*k{PQRLeX5Lk#vFtj0Yj=MjgOLu9X&7oGdL95-yANb1BqNu1eyaw@)^++Gzxr*FNF zezygoCF{Oj`z z^PEzt?c;T+k9FsHlIQYyD~}n0G*b$OMj>$HlNmlWokB z(IcX-DJ*-afJ-BDaGBQtZ}Y=wc=>lYjcOJ*Ow`)*2}q|d|!s;|I2SQL*Y8Pn?P zXVG-^bb4cm7VVjEf@d8aux+RYbb93;Q{4CO97fMQg zyeAKQP3evsCTMO)quxP%jJae;EW8LAGa5#|k1Ekm{8!Ou%IaxX`6|kNc^$PqEdvra zKZ13u4UxC;7yftc`LAya=~!YJIErhpd1KFQhJ3e}LxO%zCX}eB(2Hk0Nfo<>MyFTd zz3Y{@sNgJaI6DI`g?)jVL)xgDPw327n)Hv~&&cko`@HKV$Dzoe3NG7^2L;uccyNC) zX8rgLGe1p6%|+wMPnp;F(?S+Qcu_EGW;zLJc#da27NfZ1bi$}b2R^Fy!I}X+nSA69F*^8yRFqW_ zr!O94$Bkjso|3|w6oS+*KW&f@Xp;ulsW{p^htyr0Nj@(Afal!yyeie zpJR~X3e>RJj@#3xkbX}QvUc)2^lRTv)+a=gyEEsK0jssd*YG(^+3^_{HU;6GshjX~ zybQ{Ui3<0Zm8gHI1@o9NtlE18CXAS4|Ky2un(RGNZyQHks)umj<2IaCu?fSxSKz>` zo0!+&MzV-Li8!xMvadcv@pr2`y~PS-MNBi^cMQZn-*I?;)=d!Iu^rxB{tmLwjZwmi zA^6FbELmPlt_5b3>Z<`n&>=_CuFfF4^p=vE)f-5=o<7+@a&eOBEi6tr3Y~W*Vw1K# ze0G!~rySRy*4-5_I42TTd!^$0Q+0Umlq){^?T-(}D&Tv!Q~36F0eGskgL`8m`Ov$F z)Rca~WYZ85nOKafh9{xz@^xJI@d6s{8pT}_3(5Sj4t#j>Fz)@7g1?qU;OT4*$F`S4 z)*BCe_pch`Xk~gy=zDUuyp>$aJ%*d>=He0$Ypmb%mn`1Nkqv5Vh+Cn6$TbPQK<7IX zXL&I?MlzkWRQI5JVlnkP;|w?yI^n^RBk*jp9@r;cq9#g=BdOW5i0OMDVpu9gM^s7C zbXE^ZepC$&ua-h`%VKKDQAHAYONo@VE0OavV=#9Ab1c~`Mc%{fhOd9ryRMNtxbM;d&94x0)XokFgX4Sc!iI{E1@E&EHi;DRuUuocIc74rPtue z_X>D@r0nUhv|2cl$d zg@?*^qxZ}xEQ=_l{0_aL4zIsVT?q7mpW?@1@LM}{_8r6(p{e*-ArnovpM-kb2jGYa zpgz-=+&jI9bkuDnhnpqIV@{g1E_#7`KDSYJw-vzA+7Xmnq)^|m7PY*M$hZh`a^7eh zNm;H$zIYj6vcpzPwOotq8jq31Qpbpi(A(v(^jQqL_6|ST58-WLm-`+!!zXDNM--aOe3kQ2-$)Jrb`KW44dOSo)$*nI~{Hh3j z%lcven_DpR(in_6p@OULOXJ(RQTTKD3WoL`M9+>YbUbkyXM4w>gUn2v9o~qKnng*R zXFA?#RU+e;x8d%8$B_>6M4Jb$sM_g(7iCMaNq7!@=Wrd)YeM1_t3dX`Bh2zhp!}w6 zr&e}O#G=xpxc77b`ae65y>C{4-WV^mN!KG4dJA#K-DfzNO2ahA0E{xvLG`YDEL!~- zCm-I7*+r>%vN8{go;<^zhaG6ySdGhMLb3i@B*KE}8#9B0 zl;Voj2{7`a$7hI2K>Q;eap<`9@*!m}Kng zZU-};LU<$Wjn6dKV(!u)B!A0MLOK=a?s3L%i@dOnF2?Vxk7DSZt0;OgjS|)FqD=YC zRML(?Olx?EwHvNs+TbR#-fbbN6mnO-u2H1JpFJZVe5L3s+95=LvlB`FxR`7@=7ht# zQ*r6_bCg248z%nUjXyPK;uVD+xZdZ2s={X``%VbH+Y*hOY!uRoQlunQn~-x_WX#4G zyy|cU<6CQxnWI7eeH%wsJQ`1`7Nx+v_3tRx2Qy$u=wnl5e~jFlS4FO$w1KOYpTOeJ zEWFgCM(!6;q{YaP*e?<8Mg6z%t@8kW@lV5Sr5QNZunDFQh!G=$u_W-GIiXvgqRWFO zTxd6l(qjjSB-2F>mbVi4=8wyhiy%g7_!X@lM@#bJvmj9{P`_zXxNNrh0j7*0gDQ`9^}Z? z5E3oiw?m5K(PefD=ASu)8y(64&d6iR1S=eoJq06P|6ugx7jAuBOHHq3#Rte z;P=A|@r;fQrl)zMu7V2Lds3Z5+J47`v6)C&bfU&8TeAJRJIQUDN9u1!6EBBb80x-~ zNclPv=?~*@PhB6(+&BUcIoCCUcm*DO-mhd2Z1)5}j!neL7xP(zBIKiL9iwGh7 z_dAK`ur7HveK9ePDdKezbs`ZAcAzwi8 zdpyS7xs8?A%kbv#9~}9%m|WguMFs|eTq#M%IcqLq;o4_-VeT)S`{xhd+)s$_6pF0r z>BpvNlZd&0Ig#4@ko58EN#V!Olv=1Y8oAuU`)9R?tARc--WE>gn>pgUEBi1dT!bt` z2@G7k8S8FUwt z+$NIEdu&LlqaEnlzNCJFBgCjj!Hb@=sGxTX*QRfR@Gr-pqHPVXExCZ{J09V**haLN zcNeC0Tn32>Gc^CS3-|lJ#|$-5a>-4Y1#wWGERV>=w%6B)$DdqsdcqlUMQ1%Zax;=l zp-z!6cfX+7#x0p9_!YSp zIdZbA$u3b;GGXexUFtI`E+0e_21mWb$4SiTv)%?rrg1+z22m2;Ug6Nbrlz6 zTtmSRQF8CjSn{blAN`%Kkfv$lXcMn#w4Po#I2S$#_tfLCC-60#8J&(vV@>~~{w%(r zin4|@Ha^I}w%UAnu*DXAWqZ(d<|FjlRg8C112Au;JbG-o0Cp!Qp>E|)-3 z-_iS0B8~`$kaN*`yM~;|BRO>iqOew zVkAdU04EZc!i}m{s^YmPm2>V0tbLk9rSDM!vv*RMP;w6roOlNo+X_)o9DyRXI@lE> ztgrQ}Q6&Bb{+oXc_n*=S86`tXXGJMdIvIR8W)m0l-c?zeBG@LLn9ZSx~bEW)stGbXBgW|K3%o3MRm zC4^NxqRP_xvG>;pK(!p$5`7ocG83pzYklCz_jsb1VMzw&)S>u%7uYDd7z<+#VzQ(r ze#*4Ro^SCu+J$)FoR7~+n z*jLCuRLVO^GO52Z62$BAGu%hyNT%*0T;l2giX{c$JLxphOc+ZlWHQkw^cc1r90Z@6 zQ=mgQ8*YpgC1b8s;csO=#8%Z)4(}gRag)xykbSd}j_lYXPMLJB^}t4F84P`r9rjW*d6N)&fr#oZ&?q}A;P^2OHTNX$+6 z&~Op2xvCLuIRP>8+(ugCHAvg>OCUaz#=4!Jxa@}_xnsNm^P;b#nZ{nSRm_3#)MgR8 zfrD_~xCk$XYyZdmwLq9#`KO@{Z*<99V5Lobg3CPdeDD0yZ zvGw`|_$T!mwkOF$frKa;2gu-j?=!gPP#rd;RiNjLV%V6}2D1b6VfoZa@H)AL((s)K z$*GUvT)7oQjupV)vIHQ5s_0i}hbjSSAd<|0;uH?75ADF3JvO+1l@hMHt_o#2yWqa< z4OsTV5X6M z!X;xTV4YYluKN;=vI?$f+IAXV?TUvyl}=co{*tQle?}R}Wx&8rY3!IlkoMh&^Jm0D z-ufpH@JI&d42t5{%`%u#BZXy$W>Ess<4`*C1+wgXAeys)ahj6wy)X?d75BlC`*-1D zOFaluuEUeqr@;7{V`0U7w6AZ$LK!XCL}}t|gWr%8uz^_l>Jr^j7c%9>NBHD=8Y&i^ z0ez1P)cD1j@WkyIq#E6Y(uq&u#Rp}q@O};Q-D`1LoNz|JzYF>VxnS8ViAtR_$!=jk z`gBl@)bRy8of=J0dMbiaMus>~r60aEWP$h26iV7N8`d2uf=-n@Ts!|bZrhrG!Cebs zz1VCxD#T^ZT#P|S?*;gs4TpBo6d02+7DGd1F>m}{ln{A?jmKmNf1wU$oSKefO0@9b zGY{;t55~5gxA8!WHK}4PiS)o)5^l5`;sgGInxqa6KMTWZElV`K=YXp8S`b=&!Z@82 z$o*=7(w-*xh*rl#yju8l(i}pR8>wxj3OF?{0NnS^h0DGTl;`D3)crp~jHDz5rfTO? z2FuH-ebv9HoWlmf|JM~>^pAn+`RbtI#el?@L69bf_-lbLj$T@XbKai7DMyZ=&w^&k zsAWCW4Ch0$J&h&WSFl<|2qGSQOE$Oi$(ic=B)K;nCd@nlkN?<0qkcAg%k78EV8Ex= zXP`&R06zISz_|3=@cD%#RvggBJya<#+wLSa>+>h-&$vWru<3(_S=uN)<}nVWrx z7x1EQDE&HTJAE5MY3rSXq^q=^w*Ur8`ch+Y!1S>r|P)o;cMvdmc+DWx(wpm`ppMRG=(wQ`8n-6*wm?2(?xgWB!jkyfPGk*M*MRDq}>!_+<=)j1@t>?bD$0 zSpxO-Z#%Vj+i7xS;4qPJjwRyn)iLsbC%&<^L&Z;$#LOxcZMzVi!59OTcc9)IqO^niW#Xeyf>&(3Q8VWs=6+0u{o9%$ z)hZi|f^Sflk2zz;u3UT-T!OZ0kI-ecaPDDwV7>hrgbhD{`|q@{N$Lh(EEOSZ?5^S5 z+;)t%szHgTlGI|Isg&CL=~VHme5$uL1>9xgFxxK^XPZ>uL$9kiJhU3$ir&DMz9&Mg zHy%fvrjWT0XmW4HO57lK5j1od65 zFmn79%zMJ&gQeebjSv@4{$~Yoaho9ETQyaiFG|upb8&0n1k&>GFHDFL!Gcn8OkF$| zk0r+7d!4-~lH!M+DpffDb{5W(JA=_b%JD@12`o^Lft9~BsGlDxm|#2$&dTS(i<)`R zBzKpZ$Q{GuPo#*QhCd$aJA_Svhp|vyi0`<4#!nZ2V1wpkI5kfLgSV}~X)!XCLHILJ zIT3`F&VXb8X5;;#JGk}h0^(h;kp!MTNNU>?skrF|c;wr6^vPODM#Wc?uH!)7mZ%W_ zKm8c{dk95oS+eZIF0%ZyG0`gs#W|q^s9GUK^sICMzqP^VN3s~3Zb_KQ7Ua$2g~Ym* z@GQ)y!cA{wG!grWA9jlq2{Tc0#6lBwN^S75vK4CF{t6~{J#gaV<9O$t67Dk(!v*f~ z==?+uYN;JiXWBz8%((#5R=6NUrs7kJ5;(x)Vf^uRcymiAIv#Jv;2GC3%5fsuCO4fV zYL;S;c_JR2`xPJOiISP(lSxbX2LxFaayUW~?MB?OdeLJ9$4zi9`4BwV`x|81j8WPt z1802gLZ_osp&}|5_R0zS_c|4{DlkXhyUFOYWHuFjJ&mHa=25$BjY!iw8)r%2JuZZuruy1=1vzSOXe*kUo{SrHp?+& z*)9Q^`b1*y7**(Be-us*U4+F^uOVM|7OqwGL8AjkWQEfN(zV)}EZZ4QeoRd#&3B^6 ziV#yQ|CNDYbQs;YIg#Uc-xf#jk}8Axmq#& zRS|)tBR!OD{X0-1*3)@_{4isj)$z73&HzN5xiBJ4DzcKVbjA*aR1Rs)!LZBqt6Mbtjoige|l(_xfyOW zP6D3sL^yOO9m^WdpndmM)QR;2p(g^YeZCYHILUx~zX_ECb=3Zo>M)JiQNEwjsGPBO zh<`TVrp65z)3_DZPx?xUEdN0DhKghH&ztb}!5B<%>jKY?sVJ8u4DSn`jjk77!=A+q zmN~nCO~yIUxvh`rE{*x0C!%?t89E6jkoqkWB-B!bG`M_(FY{~RRPr{M->;mPA100p`3Wnd;C$VFSTXAjHBaL^wcD?b$_}OB=FJxB=U7?rYo7;k z?*T@-RlznApy;a>bU%DTInGOh^?PT7#f=r9&8Wex1*O!je+>08e+=IEas~cW)&uNP zM!VC$z{lPSJx$7B>5g^KGIj^3t~>?f?-F<~l?AcCBZ%GRLnPJMn^gR5paRARQ6tlu zsgLzh5b!kuxGGbq4YPpqVQc8Ci-i;RXQAxa0+3~$At$7UdJr`kc23HoB7)OlON|TM za#{yg{A1YOxCS5n+J=5xmFSs&r_giDbZOm`sM8Acs_X;fz5zI49tc&+8sO~V3y!P<{-x((-|wZE+LsTfN(s2m4~HEm z6R0hVeX9j_Xv&hV zoR`>r?;E=AzXoFz9YFehGPEXnp^?5E+Gh^J|2+Tt=1>J&S9|03kZ`QHzmFWu$sxag z$k6#N`oelwiSb{1&^qNR-dlDF-)&7pcTHKCyT6(88M#53sOFM*<7`q;F_GAw4aF-~ zN~pS4mG`6LJ8$HkI~@iSo569{r4DpJ5Iut`RB0MMH|JO|BGAGhI-pnNcI5% z4u%`ivk#RJcIIn%5t9x@%{9=w{5D*?XMoe*LN$Zji<{XB{-gxK0QLxcKDGMOZurO!3`>* zt&>t;q9oM9TG5?zP3XR5mYDwHBq%v9MV%uWBy74F*=c(cx6XS3CyTNnN00<)*g*Zw zW~uQlPay4B8fp3Z8^4G>fjfnR5HK8w&03P!Kdp{>*Psku$39aE+vLfc{Z81bxgY=D zHbAf4eU$5r1h73bhGf6;A!$7`iTR%!cvCG4s;eJ?0L<{pnY&=VdnRrRlEfKLqhK3z z2Fxv%;ZFUVNN=zqlb6o{)8Yf@VxvgPoUc;)GP=+%#M9*E`f%=rBWM9HK*1P@{_r?* z`A*?*E@$NglmN=yJo<1v+Wp&#TH4#1Xd2fVQ5 z1YRpMMW^xuU@mVBBFgvCkPL%C{5WV&+=;E3$;f}a9>wWRm{EQLjc3{8ynS~->~$}Q zhn|9;(%MBgQH-PJ_5mLnGf zGRWzhV@dYjuNX5&4Oeb>3Pp!%Ft1OMx>+!TTDv+Ax<9?8Tz8t{lqNYeldHnR zc_m>g%1F4}9J0GdiX`rNfHL109GDUhjeSGdw6Kx9m~2XqB&MSSPYd-(9_V!zqyAPN z9=KNmO=lkv^g2vtSDhq*|D;Juy&dZKi^1ah_1Iooh`O6k{>S`vrAh>kAFse8<28v| zy$`W{QAz?1q?5}xPobS;DE=1CcVR0faJ}N5x`pwcNg~r=>QgyymhvgzDax%uN+htI^p%N7j-NLLK zUy?IgNqYM$ATr|sM(LOl#epEGESrQC`c7!ed%+XmT|luzZ81Af1#!-F*e z#Mw27`tB2qW%eP&T|^b8CR%{KPZT`%w1BPyS72U76UYV~f%&f{;Mu+vxO~QYc=Too zeyM4|Z99%p^UC(Z`LWt~&8-*V$?#8g zCiHGAhn!SxycwP%M76JC^Q~lTs+S?=`Eg|Sl`IH+RRk}Ub#Y4AQM`9E6W1PfBHi+H zdCMl<14jP>J{~?pK0Q;Ty?@x!B0L)~`&a~XjH_T$%QTX(tb#}{lcn{fg?sl!C0sW# z8&B*3op z1L1rfg%Y9*D6c81aN^iA;61WOu}y#vWP|YKU0uAB;1Bs;$~e$6osJ&wPTM*L(uZ#t z)B4kO=^xfNNRp2Lx4wCVCY(As^zR9dzJ7`}ZrS93OCJ?=TOJbTg~PcY1vq}y3w$cj zC876=Av5YFUX%|bbd@Q#Y@dT+LiYFVQu;*DfX1ZCD=M&q|)ZCby_-#TQPke`U z!z}O(j-j;1CEUBK2c>GQ$j(V;Na4cqw624YKD4P6#;AK^=-^-6RB#ErhWDY&c_EkU zfAF7gM(-yvo07=kurxx?7a`k{Ckp>(1@c`v0gi0Dgd-Yr2+Ujz-%ieju8dX)TAU0c za>MZSz<&6Ds5WQ>76Il0EGwZI|CGpV? z7Y==`rObQXFh;v8R>aMq!6l}`G_|+nU>FaZ?%7z}_c#Xohaf^tMf!elBd6`!#@@gF zP<}TpY@BG0xZl2#yoUu6r{}|jTF35~vTr3qB2HjlM;N~=o5BAE%JchEwZfPK)2OC8 zofaEk2Ya68xCu`=wB-<&HdgVJwl{-mTob+l2iX2&T@dFZ5iY?&X)&uybQWb`WQz1}P- zln;|k{v+x!1J7~Fka?Wu?Jbe2=aFVzCf=+t;R^4Um|s&#)kojrh5LMVONin@Ml*3` zl|5^Z(B*f18|m4qQeo#(N0@KTg>khx+pf`Lg>NJIl65R^(2rxMfw{Ee&V96bx^r?z z5RdG%-s%f! zn=_lFvSv8%w~rm_GdO9^4`}KoW0Ro)^q;Jy<})ktUu6~sioQDE+b4xa$r;IqdQ;(? z@DXQ!`?Jc#bS}Iv?r=+*#c7w~d8_}ZqjGv%vI8e#`}EaeFAi6yJFhT7J9Ay zS@4SvqS<#w2p&7isj%uVDeZiYa9J@cK7J}YS4%nOPb9`}J<01fe&nMqk9ot)AM~U8 z76x`-&$sn+IqUmxu6+B3KTMDjJSyf3_f9(t{gO-A+36m9WCqdQogZy3Rt>QUNmj%7 z8@I4w*+z!-30Bcf=X2xcvrKv*d_8Mv%~E$#8)Hu=H(epG+KG_gCdZq*zJ>pcuKZzg zEPmB(koL;yCrzAvjEzS-^2IOu+(Y^YefLPLi|UV{!&j(nLk1kjE`t2#P%IDaFP*JyD}CHK?ft}y5>S(&{jYx_K5*POM&kaL#Yd~z|nUsK|j8iP2u+j~^(-bT|kpL74S z2CU#ThJF_J=Ag~jv1NU=a4Y>N%?gR74SMplLo*TQz!+X=BEfVB*X~$`tV#C>^ z*m`9%s;Y)VRqH7}mFBXx_j0Mp=yc4!uo&}ayW!QIQ9NGjDmEP1O1oE1;E2Qd;vD@G zT^xLY?)iSADjf~_RVC#yr5WgE7)Z@gDcmCaj&IMkWx-sV2eB@1*Eo!-o?%;mbP55$43C?4u!tT&*s?JQ{h=RW2^S#8fshV6?)Q3~GWcZq~ zgjMY{c|d=Cp8Wj*iXMH(^G!$5JYyo1F3d!Aha8-4>+$Dpv3T}3UHH^~4&?eA7n%|= z?1}+rzOO=)_#NB1{=w1t|KVJ&=)0Ef%S{p6;eA<`rW{q|Gc(4cZtfxBi?uRaBztny zn;jf^T3mpv83bKt76O2>VYk!JOJhpkghk?*C9=%@GDYv@&W@F?Qs zAG%W4x@;aXau*L9c848)`tz@sN6<57J9UpRq;G24!i68xSjl1*s|bNyE?WV+RW+EN zT~7-%2S}fZnpCW82k$!i0E^d~p zN8a}14OhjCiLnVgc6*7B6Swh#lH**b)lBO5zSF8bAIbS%9C`2%(G#dj*81Ntxg&!2 zzy8b$twlm%;ayrh-xpuLtl=^Gvh1@zm1YhO7P70ZK)L8BkKAF4;kvIdq_`a(^^o%A z41V@98uK1EV7&fB*tEVAM!j^w=^AG!hD;ZN^@c;i${Nl2fmju#$Ll0#F+W`vKk^3S zue<_=>$t%qyAhZE_;LRy_gF9BJXb$0<@>o}-Td5JGWBbi@Hl8FI!A{Kp1sFmQQk5R z-yY8%X1n>Ek}me0n2*5nTX3}&%C;Qv&vqTBRxWV*G$bKr8+%?o8Mn^+#4lESg= zrh=F6I3e_o4`J*j(scHwsgvc%V$5F2)f)=pg2MrPKF6Iu{&5kXyBKy}2Aud~9e-Z` zj5lk`^OjX_@Je5-bM6m=m$4~-=yQ&jn``ofx086AvNNZMTtu_Pde(iD!1hbs5K&mb zmo|Oj$Wj&Q!%eO{GHDhiay$nW`mw(9Sn<7lV|(cZ&iZ;CMXl*@$-9k|gjbMpkEM*C z!MJzgB9$BF^Hce2JgIFeZ@9Zp(6*|kZm-tE|Jx8=Bx<$KRF&Ws*B8Hkh?-Q(0et%q zAcw0t$Xem+7z*L7LibtVR_m#1#uk8<_(cO0`WpC7e`Vp+L|Flt%3O&4B>MP~gm z?pi0kQp=-@-(#uvl;5KCh|KkPEb$bKV@A(?_C zAyBBq06Ma;l7reV@Sw*bB11ini@pruy)XN-Z$T;d7*Ne!^_md6{TYw^r@=BSw^DEO zJ2<}9kG$>H2*1z7;Iq0FJe@AU;%zTF|4^C5SU-;`Ud5r}9JBAqKfX}KBTQSyI9eJnHIaI+OXUUXQ&Dr~1^wMLAGN(aSvf9>F9&M>PyO?v^A0}muZI1@ zQ>b;lhmrGS*(zIypQ{>j$0ix>DK+GUO&M5ca}E*LQ}JiSKFTwf#f7d?Q4_hx&S_cP z89IiSPnjZSYqSuN^MdA2D~6GZ3C`yike<30$88PbpGi?%s+j|uEsOA2J)Rt{RHH^? zKNU6J658HB#oTLW(7I|J28y1x!t7K!X0c3g%esR$qg+hzibQR#GBSJ?Qq+K9)bb$! za}^Vy>?3;2ePme(I*V&-j^kBNQyw$L1jCou(0va%yg4q5%)p))Ir26aD1G8JYd`VF z^fnsWC7*QGAEU{Mfjm4$gP$Y>AmFw$cPN|jc9Rz{?HkR3&WCx8$v#%Quo~VT-4WcY z7uwHF!YG9)csErS8Nbg$=KX9`@9#kwYEEn`W_gtJ&GE*~pEb_yV4L{4Ja~94<DP7fwtCfA9{m~g5e?>Xv2 zvSuIXt+1TRVh8Xyg?RQl)5^B!#WD2>SY2CAmw&%N)Icemb;eVHdI1tIj)kI2Z|b$d zpYnCXs8rburZt!F!M=hzle*FI87KMAcNytmIbG?||77WJ+6Hv&sE1<2X~g8brLeSL zbZO=s1RGza-;*uK+PH*O_MYaBN%vT(eJB@N9)ZH@5A<~ZO1@xFfG^`}=)$06KG$8` zy!&k_Hyn3jQ@ybmaL@olOXuR@;c(%f^ASNtd?#h^J2A3iINU6Q`F_VW^uDJLd51Z? zMfA>IIBJQZC+d*-cM`r2n@PXCi*ZnQ6kdy7%Dzq-d~EoBME3bauEobOd154r6+3AB zW)pFSJ{Q%MtB~?)8Wa@W@Vi9|BRBpQ_Sl}Jd)p6EOOYPeh#9>jO=WbpNKe$(f6$k% zEi~-?NDdi)6KX%zAUos>G$yV=OKl~(uCn5?>J!Vx0_mj5@Dhxtt7172FZ;LJ`oYFp2n9yZg<@_uC08bH0e{-x{wbC9^R z48qJUZ0n`NL)5mxqH+!&F^WOqv7hv+iwU-j)5pgcYqCE{yy+FO2%qL#TO+x|%9Bs# zt>PfPK<<$n#ACC)_=SBCZAuKI7tQ&?xJinF_4w5^FS693J=WFkwb3 z8BFtKO%r8acwSw6p3jt36^BJpUMRk^7N4~>FmrhbEm5z+C+h^L{tbkm-Z$RUF`Ivy z$gtO2Bb**_fCl`@7uIWf3Yq%#!lNHoCHlL1@ESWqo=|&5%zs8;nsXclq`FEZU;3d^ zdX}bj60KS=8qdYrE;!r@=3nQF-k&w}?pT5_>F_ejbPA;RtBlCtrzYA;T7`>JS;^-; z3y~-0LFG>@B4g-DKU^ZNGIeK*_Wiit)<84-Qt0&kOuBE>6T4EwNIiBn%XW+BPURq; zrE^`x#mn$vr%wo&V9W|T&*0TwC78JW<0CqIc%r7buf1y;H^}Vd{_B-_h}CZR&yq!- z<)VLRYc@G#C}D|zELFArhlc0@VpcXWOZ`azBZuBZTddgE88oWrVXl*6^BGqKCL~Z_DT!z4R_PlgQ8r+}9 zLil+IE)GAb=;M8%RPj2`Y2V8l58e5F&&4R(7ea^BKMQ8^^=N!*fI*ruPjyv-bu=i{GxT6?~@?z#NIZ&r}UZef83rH%O`?M zsq>qbn5S)~3yBtZ{*Wm$PMmu=i?{5dzZ5onD)ryJ2Wi*jq5kb2SxEO`<3T@MQ@j9c zr>l6f!5F;?RS>*wKN2g-G4Q}4^gj54<~5Yjn{&N{BjW;)xj+*`I^GMn<`06s#Rw=I z^ug5!_i#zq48x_}Y0lSQaJ-U)hTMJ_@Zl%atE>>8|4?vlQsMt%ZQ*WsMNo0=DJ=}w zmR?*|!#_8Ugk*FQHT|6;X!1qa7lxp1ff0snnvQ-}UGOZsj(i)AlJSh&v~%u!+FS2O zV>q1V#ordZ-mSs0ohN8s`S~#HW5n&_%=v?;zo#Cx#;t`*@Ge3Nlf=ySs=|$=q9b9m1#)~;_YCyjMxdl|B3=)*K(NnB zzL=GRXt`h49O@Bzy3ocXDQ6Gv^iRFfai+Nh< za{g!9fR*nSK%sRl`Oeel>As45LSqkn!uRr?ceXq$qd)H$vyY>TJ=ry7F5l8j#e}ab zFhxwen!UG1oyB#!ztECC^gKtw&z8{8tTI})zZ3u6SLCtZOEDt!C~JK&@dc6>pfzgIhGZB?&fV_2jE>k7Cl`0AUfbBOsosg zH#7+M@AN|Wltzj+8jF&nM|fd}35TT1vrOc-O+=3;ICZMb$jg^Kp=CKI1Fw3aKe>iE;B4h+T)%iU;QxLll7 z7$CXFAIdTBjh9Vk_|f>2Y>V^Bak?c8R+Q4=!v|^S_71vboPvhXqbQj@3-iNFuyN*I zTKVb%M$but=f7;sInW2G%>$5d%WiZ#ne#Ajv93I9uSUy8BgXdqm=3_;G#YvfqD zm;SD}OhZTMbN#NrP}54p@VqH()u75@`8Tkt>?H5rwTJ2Jc+uA^eutrd;rZS2}}P_Vd!jH9^+3YFK^SEdBbSa!pjsrM2-h2U4YM%adM8cv8c-IFxBl(~+Um+z<0gH^QVr3sFzt0B-~8nWES!MAQ2&Q2VKm4lb?o)5(w zWfH@yLUS>2O$mM<55?HN!zr)NNRrmBrHGTFmsE0uCOUqi%D8tl=FbXz?S29OrFPK5 zxpmZ3Yk-J@x8VGX_+~~g9`p7Pyq`|Q`>YnSNU0PW$2=EgWsGV5Sr?SWB%xCD4Bc4% z0w!ILBG39Y>eS{y#br81=S)C%DDso%6Zo#K9zTuUh+J{Uw#ngNXun-Z6GrNycI#M# zuJ9-Gb6u&%;iTv@%*AWtELd+T!S}vnpf4lp1y5Jwmi9iBshmJeVKv%641%+s3+7mF z!GUTc%vie+K`Ddru+JRYled_(KR=-7x(?X1sDu1ncOWRu24Uj2RU{vaj;L37z2X!4 zyFY`+Xdg72hhx~BLIm$?q}vih_*KOKou{yR%VBi=G8^%t?pxabGP-`6h*JM_N&Ba1 z65kA0!SkRwFdzg|aSWb@T_~_!0RvWsL2KXw{3zFf-Yy>$7pUnWH-4<_5bDsy(oi>=UXBO+%g>m!tLwvPwZybzFghB6z5FF%q z^;BoBN}R*zy+Z+wSmbT%#{FVNjr(FH*7d%R!WCXbee5cLq;>($kU;NCM@>D;c7clZGf zel%3fvnpcWoCY#+zYM1(HNwaR=jr5n83bT5E`C{wqr>-Ndv<|v=htl^`s_-fVfr~( zCoMuk*hNZkD#3B131SA?kKJ4o_>ORhf9ww8>;@U0Vl$qz%w2iAvn7;E5^?6@M%1Ns zg}%2fb~JB6gSkBKw7qMD0+~3p2ZL%_gJF95JrIBQN+FuCnIYZKN zuiEC5W2hu~&T+|kOK%}rZ7g5hdLAbBHKJE#vgGWk4dj2<8^u;mGfLzHP`6l7U{sN(YPO~sz-Vw?7=0x(+UV!HVnRH9dsB5Ao z>DNc1&yp<^zife!{4-s0uPsJsJ-$zP_9=wY1`QQ`da=SO*$Zs{q6JN-tEp_*C!3_Z zx{~FtwZfpSf%K#NnPltd0Lh}I?g%%?5eydHmHf+&=J>c*h#P2tCp*#vm7Lka$YsTn z`scxvr`@0I4m~7|drDLnuLh%8ld$Ty4J|mOC!uRSB`#Siyk)~+O#Ny|y~3*`_GxC4 zb*sf(r&+Wxith^VL=VAe zFcB)GH%Tb;K*ymeuy}TbHnz^8am)OKI`zGh-I>;sjuFoVPuoJ7t{u zQ$*3?9Q4$?$JC|0js8dnOE$_{3z;8V1?4S6rIUTaIqmBwtPg*T35Vt3vfvvH?rJX# z*Pki){jQNTjW7_xWad!EgHBpn>&C01WAXW+JNaaT1ghYiuH1G9-ofa;P+LG1<2G2$vU6rKXAH^ut33 z`jbr1>*aKExgI5YI6=|z71l;przCL}sr2%By^v#cRT!-D@OEb>bxRWprp*l!kC>bQ7Jm9U>_iW-hFA zH4yF@j39d7M~7Cd_6;`utQBUw(zdAs94H;Lp$uMO|=ws=>bcH|=dD~QID=6FO6ISUKJo4oh5A8(Yk=aYJj-QsUy z+pYd+zR`kJv8z#hM2d|MpYfxl(bB2zhSFE5HMIUwJH33PNvqZO(uR#vN+%!T{KN!K zYII|Bop`J-V>Vk`hBF)bA;(dJ9h~;#hjA<}PE%wn-%Z%^Z3`vL7Ux8yLG6lGLZDF@ zx;nl>k3C^1@3RtB$vPA}bEq(GV-Eb2qR}tc9g|1rv+kZ>th97JFWMi6tQBXGl@^J0 zFZE%$p_;~v8s5x4pT#{Tp*-@?SBzJm!@owpLao@V=T};a{ncDvXFP%5^(f`14$nEf zB94_rkA_q$m|uAe;6(Rmyx(hw1&{aA_1||e{l_}2m^F`FYiFa(`lrxZnJmmMABeUQ z`{A<29D%MS>=^7L`bT7VsEiI*n_k0>J5Nb{*8)72ABBah1E@^60R1hth|l;(hx%Pa znR`4;elHSP2CH$cZy!8M)k2Hk7{2&*4z~Z)qD^bunI3Ipjfic$D(^iV(i=}>ZLR3Q zpebBY{sB2AH-s%7OT-NDV7&imjbB#R*h$5Z{<4iH}&QEJS3*PrkdVk zNv~_5$gWK0J8>Q&qr#VS8s_o32Q!f7_@N5L%i5N|nU$j6r+6t(r+IIVvgyOdqf zc|bgq0au~#@ffqFE#dR`zo4)%6p1dHd`n`+>Cu;Aw^x(@b|yo9VjXoh&A`I+8<5sX zP%!hNur7W#)oHoV7)?F)_Q*%iA|>owFc!%(Jn-b5$oJ4)3#DgQDYfrj+J3YWjy-r@Hr+M_NCVc^7#Ol3T<($1qRc zq$S-sQC8~kez4R)I1a(~6wY^C$BopT7_j4uaMxK?2+{C^&jv@t?zhGQ z`(n5_vK8Yow5-9wU5y>y09T+a(1MDMT7sU3J2ahf`Jdtg_{ zQ*<}apoRn~98%Kobw&-X9Ninc&acLF%%m@MQj%@lOUEM5Vf_e^rEsMP_jV4zqt*oU zvaLkN&qEYAU?t?}7Buo1OA923F|kC~{A={)SP9wBeN1bHe4-_aA8D|;8%{(xApU6> zHkRE&;<~|n|FIk2$Pu%AF=sKTxD7>XHsgWY+%>O{2Wc%bn4P`I7Rp+@DYlC91#64&`1-n|S{RJKnYR8cZ$P5Pw);mG7^i<}{2a zJtrQz!HRnv@#LbH>$u$R06s({ikW|V%Pql2N_BjLJl z1>TL-!2FLl1P9AwwCuw`th^R!W8k?-$XO{*( z<1u8`6zIMP#i!)WwAw73rmgIbyD@~i>EUQ-xrt6MC+^f)z^cb?;G*(DDl!~}myx_lTM#)iYb`y1FLbiwb2Ulfu& z9bQQ*;9Hj?J$WBQCnSj5~$ML%O!Q5nt0 z-u^?xHbt)g*_~sf^I`Ed9Vx+D@GuVN&OIgkx;>R~BMdHsL!i)DjZ(v7&|iHMtwA~e zbN~1`hY*yqwGT46YFh$N&N8N^5?@*|VLqL>AVcGq{F9uoTxipL!i7(*Rp)~jFXP5d z(f`z|k+Kq<>3)m~|8oCMdS^W)8~TC1HzpIZ7O@DN}E%UP@&a&q2u9KVM^37Oz(OT_phG9!ik@$!QeUVYj`94 zp4kr;bveR{j^mQZUH16x6a?QY3)~QSDvj}h7(QC`H@`2XE55UZ%N-LW{*8O3)2+uy z2cFiJe(J5rgU^Pb#cQ^hBbm#S9~yFK_9Ilee;^e%DYlH1L&dQ$lFf|6#kG5&{3INM zJNhED_NNfCbsJx}--G{@ZHKnl_j{%0U}kE6ocl0e2+7bAbPy-pP}Jv)l_{uJe@Ibr zJ)pGCNXS>vkTe|7t{=9zI#IWbk&|`1h-(C#u(gy$c6SQp^{Ep zGL3o-*E^=T`t=l*UmAy^*DiQ^P8J282?)rSqVfW;^7B@Bp7SO%(;32>KwV5q7W;~L z0~)SZOa8@;G^s9_zUz&H+m~QcHvJ&@eiE}P+AfH96?0y3k%GeNZsh3{L$24Y_=vdY zB4aqv)2c=&TV5jc(LOEAIc6`Nr}u#^hAT^bmv_(&(RpqBd*^nvAiW1l}r~#y6N)CzcaA<=2zjq;a+|{;R|FN z{ou$y`9xD7`-G~5niuYSQCyEaX@*Jdx7JV;LHy>1M9^e@Kh*UR85{l@KQ zM4#eh4?fi$D7Yh~aZ2|jviruevz`tIHD86}!$|guAIi5>GGLy)o=e*mxak~WNlTir zCfZF0Ef;n5X=iDDp_H_^dd0i1@aMD;54Cz1p1C9!U|d&O0rf z?mGdh8oz14HVf?CUjPHIZa6YV4Q`j$@qCeiRd+58Z@WteV&Nf9{{|&vn?Z*H@g*m<)xCR_-ZxgNHO<=Y;+P*v8h4i^j_FkjibC zHP0C`vSLj#ClUtYTw_<%2g-NqjplbrSk-^DxG!%Q)u{N=FdaX7BOQa+qq?GNRR+yi zGn2o&So4b6m*BlJ9nFh#a6o4xF3Sz)9lJWA-_ILyt9tRY4S^INpag>h5$QcmhIj89qq#A)9jB0p&}KHk4Zah;VkeAQI?kR!`4{-+*T}-8#eLLA8?F;Yy6N@XSyz!^2 zn%HBiioSI-&Yc(qv;0vAK5&ONKkkRa=Jzq6;t`~|PT1A>0&m@AS@mF}IICaF%A(e6 zC2NXT+0NW@xF>&AaD`x*M)e*E7-6Wy`?ejx1Fc$O!5N-9_1v)Z3_FP0$5Am;F;r6@ zi`5nI;&>8H#3Tbe;F|e)@y#zB9$kYaao6zI&}E$(>f$| z^M^!hE&V&0hj|W#=o0Y_UBA|0c$X4XZ_~q!e)n*FfdZRqS+SnEm>p1dhU0aSb04}4 zZz`8y;lH`i)Hy+3oAzMmsWYf^cubipk#yqm0N96i;abgnoVhy(J1&f&jARvJVPC)t;?#pV7GtF$3XbfN@S`xN9wD29$MJKlKnAME|;{ zia82wSEJYY30T|XC2h$6LC=-?;>TP&vZ&dJck7pej(nuDRDaaYRp8eJx6ymu5$sGZ3@_2vo|`~5MgUX@2#)gaeT;C556+V)uLFLIxFb*)gh zej&H!$1zn6;f&W2Jn*XPi;M5a0Sa+irJZ8zV@8peKuIrD)&;!VqsNizJRy+(mhqilX(C}k3 zb@!|!nFu@diby8O$|frNU5xpz9)fD%8{tgMOiHQgMzTeUIP)z6yUSE@IP(H^EjC7z zvYc>bc~|NhKOO_3!rAiSXI2q+d2gNDmvkre!ko4SwAging6wzlvUNfI-{(}9zX|zM z&SB*qCEkB*9OwMI%j4&a#HEdGlsOxyGhB$=QP#Y?wVYp^7tchqJ-UaclHAK+I_3PF z7Ed}$L83RtqEdzB9_;0!n+}}sdVnMApYz6J^@8tA1qxFSrP~e#Lg3Fp^zZWv`SU(g zt5z^JTYtr;e-ZGx6pWWQYhalYDJY9^xLF<{LcZ9C1V6orGEZlYl})A>HwNJB7df6k zZaBZ563Fd~4zp!;vM@7Dmv&@@Aaz>@`B)wY?-SqSq(!vvOHag`%VSAt7Yq@5!s3__ zd^8}I-&s84rO&!ZpYGj(vd(^-QzWu**Z7E>`FH%n<{|4IS&4%~uAnIB9q!%!jTukv z_}rfi&i%O-mLCW3r=8unuB4xE`g@SXEkw+s-m>NG+Q+!(L1XF8u}je@vw%8EW{O@b zLt5nFDX3l4piBE_kjY_dJS{B14y{Nm8ztsA_t|num@jJo6vFpv5sVz0g~y#k@pqf( zLs3=blQWmF?TwS1-Y|ny2M)*m3#Cv!v66cq`pe_Cua;(tUYv~fhgfu}0Rhe;^WQIk zZ>1M=wU}Zq%FseZOd>Xalk(2qYj{%PM;;M84|h&mu*Kj4b}*kueQSy+a&&j8 z)Bygl+(3H#L^M1kvTS_#E7qwu(d|#4X>{o;GTbx^Eh9DIXI@Lii~IQBUk#m$cJ?cIg{EOq3v3n`p-#zK0c)JVF=NT1$q%OU?d z7wR!Lg>7^VrEkVfm!|vYaYM5PS1i=vfuY}p>Cy4DK;-$C^DUhJti-#z?`4bf6Et$0 zI6pSLD)M??FeC+E{?jN$1Upp1>S~n$WeGhw>tbB`vrSrM=W_O|M2t`_QzMiO|4_V4* z(oCySbjI^1%YXHhp3vJPWit(!`&h$u+%B|DxQscT6L?T!Fprm&@>cr_u83*n|Kwg1 zbTeQtnG2Qh8`N%j6l+}M`J#}H)FYxV-sU&=%2tuOAKEXSrd=gHYPAdhrYi8)OdlSy zGnR6aVi4JL(*O9M%d5WgpdwYNQpZC++v_5kCiKSnl_zQYl3(=d)E>%^&!Jw@19WK3 zX0pmN#!AZ=K5=so>QfGiyuw0Ux!%a5ih4;;sP`aMZ!ve&GMVa&Cga|e&os4SA8k5o zCUi^|_mV8L;u?tsY$lCH$Iw{ON4=fz&1ytth8H=#(Uoqmoy`hkd$awQ>oDtAPK&b^ z(Os_^RBqb94jcXWqS7+nw%(Hct^g)+?)Wsk7Y^F@gj9W&^uv$d(nE8j_&zsz#}@%axouCWt^fH>&b?xtty!|AH$12U|RA}_nHY-bXH`LEo_ zqhAsO6&jnGpJBA@ufHrvB5azJ1ymZ$oWNyi)v`}C1E?@*!Qs(d=t zw3Tw-NoZ|JfHX^^n>2TnJAdk(N0OuWA^W2Yg<;n*r?DP~RsDI!((%&XUy^x%NgGm& zGo>r{M@u(6zb`efis7L-mr(gsl8pV6OABv}fVty8Dk%9%b+rn(2^l1)svR3@=BbG{EMs}5Y^w`HA{$}&#rWo$sO^weF zFyYD1(_m}SPUGAMNY4*_%Xa$qoTb~17jqcSFDa%7oqF2YcmQ{5toYxcjrb^@hewys z(gxXFVQ23d__{co;+<5apJGaQ$oLQHNpVl38@k|Q6f)z>Uo#Im~PO{!4BYrVFl=^NvM)6;BX_NPCPA@#bv-+># zd67xnq1D2RZEkSNlOs~CJtw7djS2QRnD*x4?*UsNhkX$r?E9p}JDdwIFRR<5mz=d~kmVUpw#9zBq7 zTeX&$D=X(d=ZtyIWL56`)hj;VPvZOQhL87wLKLe5lOZfDhJ6&`Vwbf9Z7i1v}wmNH)R>G7(TL$L9Zqp)?>A z+lMFOZdV_=yZE!h7n&cQ};pLu;O%AG@rbL zZQ)ztGI=~+D)fPB<4`=DqXO#R4|U*W^vw?ezo3I4Ha6L()YiqEh6OC}D{v|COY zA4MPIWDQ%4{b+`m=&K@wb}yc~tblKSF(+&A8tG-lxb{O4O@&k0ohSHEbO6hDjlf;8ukhZeh>s5U=+KC@)H*PN z-tMX-$K7{HLqCCfeyO6eakZpMzo}(O4>4owgZj0buqI0d1{*pl?P+hs-Z_CeE%rQD zfjE7`CCI#LqIUN{T0AY8lq+RWQ!bC$(+6Ve?Ec6d-WM{C^KdBBhQpFRLtf!D6xNzk zI?U+om@~9v{y(-%dd<3L>v&U-U}U_T4X;C1@M?8}+1++Jqx6Vmt@5b*+zg^M3V55^ zkM-W%Mz5Utm^tz+Nz6n>T9;y)tbP+ZfoJee)KS-{zr)~jF_1MFhR3D(IJ)8)c5A!w z_HQKWc5Vn#*TatfNq8!E7{50uB5wa1Zkn6P0}mu}k>wDKKdgg_b_1j?aL2#!fjF$! zP65?L)Ve;0Y|FCgq+D-tpN$e$jGBchlC@-0zgB3=zAaS9S;1CzH7P_5Bo&n+8u+W2 zehfZ9QP{iu zn<$~anr2>lOkw)>NjGRGl^?rF2Tm&E*pn*SbZ!VnD#p;>KDTJRiUr9yyTEmrC*(4| z(YKGixkjxEKiVeGsyW?kb5?drZ{1Fj9yHv?8%GuL`nv{F!!swzskn_+l!szkX8}%E z2BK)tTIg)-#`*)7Az!?6C(Sp&bq`ZK82^)=7v`YeJpip8`?0!L7;+=b(8p(;$RpZ_ zrKyqdZCrvA7S2T4_XYo+&x9|3pNSdaOT5waBR_K&SUpsc3;!Qg=N(t`|G)nVm6oE? zE=fkDabAz>aUv5cU6Rz3N=C_(fW zRWANfD)N(W1&_>z7ylz(yvi55-wAz7p%1mTy7R!@3-M6#G3hAn$Fsfz+33C(+x%?6 z%H)$SfP2xhI5U`FR`h{Y3^I`|AeCoO1yo$NZ_pf-j}r1>A0)3t-{ zH1*>o`NWtpa=mh8xpAB>U+wC{>z@tc%opqMcF;GxSos?p*B+zKeU2b9tqPOZj^dvG zviW;n43`IQ7JlRZFmP!dMd^t9)`{IXJxd?C>tn_Iqs0fDgimu&dk!6+ipB?q2(?%X zkFb2GjZ?>g=`2V^kI2_d2j9}iBeS83)Xe;;_aq~lq`H;nj+;VH$KS!BS=!>BH-O5% zt5BMa8&x&lmmZ1kk>`EwdF86D7$`qQZJqAZ)QEKyI)6VZH|9yZ#u`YOnKyaru01?y z@-$9fl0ffMe@dRM7LfdGsu#m0l}pQ1(_%eBGWW=DsAKVZOMjRYiGQc2U5c zYN~tQlim!L(Jbi&d;Pw^Z_4KL(*bjMeCN3c9PozJ|8t|A4`)#B3T^5jvX+wKASfCjDX-tQytNZ7$?I?tp&-v<$uZ>N8MEqUHM1Ku*$9kLTU+2O4h zU-~K7P|Mb1;rrQGwXGxOei;Ra(DRzLX~-=DlE`!>N6?;VD8Cr21Z9ig?I z4^e#WX&Ui+q2zuhfhYfR}XjFgIiu?8#WeqYByl%SSNhhITP-l#uR*8 zm8{MLvZQ9h{?UTZs~Lf`sPMWti+jPwrF^KieBogvvr&>vh_FQX{UVT59+Gf zQY(^^#~M?(WtqgQ!f9`eA8aGv5Zdmd$1im`?wJ{1wLXFPcN?kZgfdrW#j@$Fn~pvf zM;&`DY9n)7{RCaLb+|acBUd;S%DzVSfYthHqRRnMx<>(r_nd`R@Irr(`O62h3VHZVbB-)7Bu=uFRxM27(U%K2<1l!4Pzkq26>-N&!`Q>* z6aU<;B43kM$Wtw|@O|C~nzG225-a`!ak2Pu=@(7=qKh-$IyBSihpfZdAYPGOi+R(x zOLLqPal=KCCr-(P`U6|E`;|$-6v7nM(C#0S4efeN^ zBAUkSkaC|FNi!Zpw&wQ(M-Ki=H3?lMrz6{BV@~!(U;CG|w)Z8P;koHW4fzAjw$t*9XH>+kM%z7 z`Qsc5Od1v>-K>4>*yOa{@m%~;*vY$ar=-Cg=6siKX5XUQBI_3*tH%3nD!lCD5Ul2M7qzdS@foWu)?cK; zSB*dN?0IYWcJW_i%S`E17db6|J_j4tkEUOC>C)`nS2BIEXP%~4idvCh3ausi$DAqh zNXHKH87sn3x-e7f;8rMG`zni@_K#w>ZaRE_pO!rI*(o0L>NDJyzhR%M6WpX}#43qz zrPjzH(narB*#e8s^yPyiJzW1oQr|dAXh~J5O8F^v_$=RLq9d2ZM6oF3!1DR+kbJX5 zsyL#EVfyAGQ}LFv+8^K*UxGQ~f#3A~JXs;rce2KCxCyy=EE|C==rD`U+#_f9)* z)T#%o4wbL6eZl+BTK&)Y^E~n^1^=rN+CmwNPm5Q+ujM@H58qRm$?gB>v2tEAz9{&R z%bC|CSqZ+dG|v~ zyQzf;3l-F5YGaPf4$0!~Q#d*TW7S13;_T;?dv6+iw%DMvSvi>x%%@~MP4OM5mY%2t z(({Jbbn!wNopnm5VnW7hIvK8>~=wH;RsEtfScXmZ%9mC0)E?L-)_zmc2YQ zLi9-}kahd7vTJY7P}HVi8opP9l%KC7TLs~pKKF!nhQBBOT`TBa7hBr*yAu|-KaAAC zBbf4S1N~NMi}1(Fu-Ik6nMOO1nQ8*tjSSP*&7D_)t7SZx~dWfF%j;zWJ$v0{v z6$P|~&Fp$9w5gQ(k29zF)k*Z!-I8Qtmw38!rfh&HsBnK{-U zN_T~q`12=Hc~&X4oERY+eX3q6(z2t9x!qBgh?`(%*4Fc5SH!22_5NuFlmG9datHZTOuo?lZ$`haGsj<*gK!Q7kD2b)n3R<8*QH zc{&^x1G}$gysn`?8_ z3}kzu@AhgMjF*w$DdKlEVPsFNd7z3Vtvl)Gtw7rMuA01-IO0jSb=bcl3H7DNq?G|< z$nmf@rTaS~LK+SYdsTGpS4UQceehU25xwh&Q>=Fn1h{V||5?_&@zZ%cS`$PLLFLf0 zj)9VEGJ4YiI=ACG)jnyY`Sz>H^4nn=YNd$LCb7^N9fo6nv!&k$L|4pJMO@vZiJ^ zg?z3dj%diy)N3N9U5G|WeI+V(l_Bx&MP!$L#n1Md?AfO~zLuL|LGB6qlsZA0V;4(H zdS_Cz&vBfeRf2{iB9|JK1r>**FqS*xTj?b1I+TM7e+ABb+ZV5LKTx@14o%k7rCoIe zblz|eweK~YT;rBY6~p?F&a6d<)Lx8)gtwIL6OV}XU-75(7Dly?rODf4kz&09C-3?2 zwvSI&*+c7dUgys9A`amy6C_%7u4a2VG(; z5qf+(C9FR|4o=A;pL3Qf-fhRC4{=cYYl367;j+BZ(KN+Bf}*7ZvPC^NNLLTPr;?F{ zWTBo<`AbG2wexw|-cg-s)Vpv=Yn>emJwjoU*#%Q;1Ek@4Gs%3W5qiey2p`K{(Qz!i zPpt~P0hQ=-c@(bMJVeX2+t}eU2J=_w2tRf+9#-w7nCaT`^-(q4R46zj4F{#q*CtYB zcq(-cZ;||GOedXN%joCi+jLq-1FBVL$m(z#NP}iT@7EvNZfHb5@0_P9#{={y_6D*o zGsXKWdOP1ep;Nvwr0AkU8p}?~Tk4ACF1^#`r;l&t>phon%lk$A;av^6q(s1UhbFQW z=b+lD6hVQzG4N*_si4meDju~Je_Au}u;3o8o)jm|oU1DK=6kVh+f|JC@rlBeHsWIT zc+~m`A6obklxQVlXXP~VKc6lNC6CbD4N*MREQT*ulwq6NZ>&Wx)}3U;H7k5X*WM_peM$-pkjII-lnn6vNF$v{VZd@k~wsQ8YK4?3gZmRf;E~C#hvcnxLcp-7XApFn#Guv<{+v%R=(CiBwe0>3{Ts^LX2yykI7A90P-Xz_cwpXRMfBi-g43TddME($|STkr7s%EvIq5V%3&0hcypIf+pNtY|8mb16g4j#XMFb|HpLqmV< zMTqBh3}5j{@J8;_o@vn((NhOE_NO68;TMufV$>e0$&qT%3|b^T`Z({R3gOuq_Ap zdUI`?>8v&&47;&abTw|rGYu6un*E^UkH_d$r6x3t-XLXkUpBXy%#SnOI9|_z)vhg& zUc5EKjGU9$rsRxNt7_OBw&3pCo)jncsz>*Uy_1z69$UGJteqZq=A9zbv4b!$Iu6fV z)rb}u;oKHuyqJ6drd?Bz(*78J&6vYsuOj(R;&gs;dNW=!G;A{bYFcR zrOzqCY@KdUsVt{p6-_QHZ^PIBY4FEqC!je$9(EfQx%`G7{ZlTdJuCN6R-O%?@wVgW z-NAfs`yAsrJ1f3QB9uwpz zjNq01HF!{&BDd?C!I_hO<9)Ov_6{|X+nL(R%SR2AA9{KQZHtw8$&EMYbAA+eX&S=c z+nV#uzDhXe;*ZOVgcoo3BdF(TbE=`lpSnBYp_U(l9!vr+_mEyLIZh4Uvr+11h^76s zQ29v{+Yb(;^7)4;-cNArbmbUXcojD$s`9C;YM8gUn|KB{Ak5E%-(NQ17Y*-m@jyOg z*0bOt@;E=zjgcf+zpbuUpe^#wT}Yz1=02h?VJeT%I?sIy3i)wRJG_gkBmZ~$c$TQn zDRKo??*9+X_4g4xei(L@C_(RtD}rXtL1$?^%I}|)qDOhqqGLy;XWu?z;sWNt4a4}? zut9vqbr=84KFmdvU&wzpG|JWFW%8|KQaS8Q2X3&NiGP1CNcpR`ll!p@@`(IQ-G(}% zY?%a&U}Xf3{DA9gyRh>eZO#ig2c5_Mq6<8kSM{9Be`k&4pl4lSyHJx&Q!H66V?AyT z)8~CFMzY>n4|E=}lrDJQlg7u|$cy`RkasrLlBYYS(J1}*w0L+Ycs7`Ebof}-*ze0t zlO1^Lu}27ha1Nm&Q>=2_5E?g2p|9bHA+5G-_-_)&MT~;u%v4E%+LOx^Ji3O3EL|^=akR_Zt1Be{~BJ;hkxy0d(ZW} zd~qV*SHH+7F6kj}Sx=Zu+at1&p3**{A02OVh7_Mi(f8LEU}c`k{9&iGaLZwGFtx#| zg1D2^lCK=w8A%p7!TR=h0le zb|GV31?G1Vo!r5#m_GL@IhU-%=0n#p=k{Vgb;^$q_+LTGqipiBYmbyl6I4z~p+r+# z6l>l^;nRLx|6)16Xm^ZFtK(>znY++i#-aKBS`0Zdn?vX8%IggsFw=ZJ3f#uio2lcd zsPYZ0d!_TdjY{Y+TAz;|iROk1H#RoQ=I{~E8IEzVbv%O4Q|_T;tRDQ8_F_T#D|lS+ zW1rv!Jm!8LIw}RQ{rQb7^SsP6&j{^4DTj-H9%rxaiaggWj1M|Y;$tU=VA=ycOsHCj zyIU1;e#=X0vnHNK&L71}Gk0^HZwjC58_(7*D*U6{254oqN2a=jYg3L(qjg{Llto)u zp;pfG-V0sf%^sGS#PbmSRkZoeal{MHYwQkj2H2ahag70I#i`2X>ABFqSbJ=FQib!U z+Hl~$EBN_wgETPrG^A~pIXKXmPU&1G`3@&E{v9C=lSg2=Pd&^R{s;HMC?sWM!TfqO zpD(ZCAB+ERuWyb#RrZuQEhU(6eK#s{)!H=P4gZ|BYM}9 ztI*{Vk`$p5dKe0M?KvdLk~c+o@`sbx5j5x!x{YebA?>SSOoOGC;uF-P_kQvXkKj$a z+Q?7bQj_ zt}%9sJbcP?-n1qZPNkjju1^dGE?>!iUe&V+b^0IuU24~!SIUjJV}TWq-3ZaUXNuEX zwvxs9K6qia2gB0y5dO0p4fGyB)!#EI-dGjiZuZ2$tbdegY>wbLcVH!QVh49`lP(`P zAh}PNM!I2oI5W!}`+}a*gZmey5!c1t&%QlwmWmvrir}-Jau&S9`OIhhd2Am&_MTpW z)q`?yRv{XzR7$Dk$N@^~x0l>!JfMtcjWj2T$-CCN|H$v(#@!6}CZkBXyi zxA&6PCZYdrOvbKWzIZX{5M5rl3^yEhA@}}%$vONOnItJ8nUh67TN+MT+Tq8V!0Bi-nFor>*Vz*(F(r!>TwwO$`iDNm{Yy&TdnS}(;uT;k{xS1mg` zatX>aOHm$JfJ@cMlvnD6!uWIW75`&DRpFJo-w{-pNxmA9xOn?MuEzhSUVm>>-wk%8 zt22f6zSvC0irRRs)E0GmhFJ1P@Hb|5F2?HSPHFy!QM!_yV0B-UH2p2SIgaq7>$xMCmuIp=S7& zO2uGbFj9P$Y4fDitB+-kz23<*E$)+QPCD#yP4tr$(TuN}C>gDVIjg-1;HJkPr22kCd zBNTfqo16>=!sJX8viAhjUfn%3C2|Ork1W8VMcMEgwj8EC8l}ugB{JUnP%1UbfSFkp z_Kkav+kQ7ulB~y*)Qve{L^r;ZJD=ko`E!W&4Q#LsghQwPWF-qkL%8VDjVzWdw40<< zHCIwSb(GR;GU)oi95lBv<0HoF(S4to$9-p`)9Ay(XRX4U_xG7Cc4#3od`KZ2d60LWQS^JIH#VhQ>INv~iFun(0 zSbQ1_n>%51pcmGcjzxf0Ii$=+thUzvpZUkZZ9lfvb>VZbf8yC}e|GuNm#;*s@cFu2 zdhh2#ibWfxU1b@FbroH%`-Wn+*Glv%zJtOs5|W>Ga_XGy8ya$Jf8Ml_Yh{q zb@&)wAiUOlDcZju7j_=b5tp5~=s^U^62)F_$UfG${6SCJ55beDM#;HL3=aDC!|VKB z_~bec);*$dSgC|krgmfbUWrHMj$`}H1K53>6JM|{$B8X&Y*5jMJ9^n-`H?uT+TNDm z?&*$sKjsRb#3`5y9`n?h`}p-VYrfD!i+wj~^ZWa~d2HS~JWp4ZwYfe^YIV|w<)JcO zxG{| z*$y@jrb69rAN5tJqb9d>S=zxYHZSqub=sYH zerFOn<6akjH+K?k-jm-FQ_^Z40_@jPe!Rn|9El)wILBj24UcoMhM_*u+7D4E`eXQVIq^*(sa zkHgvRT@e_cgLI!5jC@~>pI4RH$yAy5u8O35JzxH;bCkCa{>_i>tmj8-kFvp`vwXd; zvf#ny!@%!8ChANhw}v`$PTVPS1&8?df&$(#d^hft$71*08NwT|71nFo2(GDMRUEv_ zQ|3o;-qYJS={X&d>!xy}k1lIwWn$HJnI9sP`rv9C!8p5q5lheSLS*PM%=jLH&CO%Q z{k#jh8~D+K8wsS*Pno`(q}l! zIzczJEozRL;#h_e-U#mFG7}R-zSqU0ZKkjqVTJR;BQ@3R1e)~IvC>tK_j}6t-_p+9 zvG+oLn{Gu*Rh{Vg?p-u(cb1sXiqX$A8~4p0;oH_juv)znS5~dyBH`nX>X^j;j!!}9 z4+q%l_n{h*TX9;D2#q6l2t)KbHz{f?M0?h(e99>U0>r8w_(1vaJma9KPaS3UON;Z*UxHhd4i zl&i?`*aZ{oLpa)4MOW7Ap#EoPJosD6)3yimqOe%@{u+Yf4~r1q`!-Ea+kmW(@kpp# zjzs}aXves_bR+x>?Ra>Krgi&G(F=(1=YXt(@E4!|wM;5o-Is2B?Lbi*zDnl~|D+*4 zKLrCwj`Gb8G+Z7@ukTrbvkr*;_#|Y17vALip z*SoMmU38guY|EpsZopTQZ8$gmDngyh@Uo>Ay{!}3MKzu^Hy`4fk=f+?M4yhOeWA>n zTs*iv54*#&>A#*wX{C)!^qqd`&jD=0ik9s8{ABB5;xF32-r zGT3y;XdI6ZLd8@)8z{>YH>Bj^*W2=HCYIL76x7JgabT-(!?EStC&^h z=X^%QiLH{~zj0DndZ5&LqLzEuMe>61HgcPD`)KpnN_xCEjYjq_Lu%O+4AnEoyGwQGxX6otXJqn+9{MdTV+aIHxo1qST|Lj?2p83+QNnL*!F(`&ly@!w(zW&M zy>%rIb(zeHOLfU{u_`5v(?j*IY2@o(Pi>oRVKQk4hHY%dfrX>E`;qCmdnOh``z!H- zVsw{ zC)89vk^a+7rO}<8aNwOUrbNZ@i0Gp{=c2lN?$xi-q{Z9mVA&H&n0pjoryB9{grj)% zcrY(Ejpk&*mx{gUhb>b&^PV^(_7*eZoyh}O?}08GAGt|0cg__#s=qKDrOg3Nr@5z+ zqI_e;iyItIv4-J?gcB@t-yno zo0a(Tz;6iM_)QkDLzyNl5&7zUx_tEEH}rOX2j3@UoNFsG|4SYVP4pqHczuGl3g1BR z;eiys)fOxI?1an3OeueKK7I9mLD9Vx*h}zn$oUVp{`*f>GNq^N?HzNnXnc;x*GBQ> z;v9Z`E?wl4li6sPxqQslH2J}Le|i3e6Fj2XR_*`MMzKK|0qM=PcAZ&k_0Dn_mt8!P|MZ@2vY=OqX-D4=?k9LGJ<;$i>ym|8NMeQok^^VCXS>HYJ6=AXhIlQ{5Z6z?mG;wPatoZ38? z_YeDuo?{Z=k$(h@$5nWrpBfOWic*gg(7)LUE_=Sv8(k}*Ee|IPjb@Bmm%=8S9!jEn zPBJKbAWd=nNVjTy;A(1#(#FNI`0puFaOqvje0xb`(;Ea^S0*w-_xZv6LxOYPn;i}( z;)=!-tnAm0B~4ArI6sg6992Y!!6CZUaGIRE-l9(=^!V@&v>NcI8!jd-IBwIruO*1E0IcLi5B+WY@P~)@I_$aKTyM zUrAL}r%9_i7r~QfK;iXTeD|pZo)6}KD&l_ksSCf^wGZlzLoq6@9WOiHgad~L^Yf|~ zc=#+C@wb|(Wkz>|fAT{ok*R(qx8>#i#Vd0ftGKL&tL}EZn5M)&JG`Wbcvv*4A2}wC-1?HP)pdqvqmAg*6<(#c0eo`iPT{vPpr8H5;NYokd~%v_ zk%^wWUY)@ee;xR4`>s5(@-__W+q29#2OZ{t9UY0Me)bYvqYc>JatBj`a?$UqBRfor zVQ*_M-lftP4_15N$;S{({C-8ug_=BP##+R`eoBu|y5Mn>J4{?Y$`pR@B;^uKw3}ms zB^OLZPIeg#-)oaYp5Uwe{X^%XcVfZA$0$AVTXcV0AS}rX_k%j&R=)=9vhKta4mD%4 zi9X+dwTYG7vpB865Kb5WVWs1AHt-4Nptl9=zv?c3KU0Ta%WCmJ+;fUd_fYtQS~_`V zCbs>R_gx_EL;^TLI_MM%lC!`;!NsLB4dbgqkmV0<5KG~bKEC%zt+1`An`zdzp-o&2HCpmStk$lCPC_d3+JYQHIg=&Q)crP^I8~)$v zV8<+KzITno!8AIDt62dY?mUg# z)wbf^y>EZRu=yCJi>c!gtrt;uPhP*tL5(O;S&z)srh>cPR-IY+ZTe zZWk`qSjLmSDeyzkA@MEfe5{&rOi}9*L9kw3_LDqbgmc17Jx*t9Y zPE9<&IMU2Y-%n9?o-x1GU%;V<9i@|Qp3=z4j~!1Mbmp-oQ#qr)0><_AJYem2{wH*X zsds{;?sEk1=6gSO9GZw17YB2q=zCngX9;LnAphCzc-MyIynLk=KX0r?uzCdUF7W5X?Vdd9`cW8qDB<~| ziPWl{K>eO3RzwJv4NH;o`_C}gR;Z`lZP|V=|osLynE{xTOGt-`UrUWVKZO@2E(j$KbI=Op<)Dp*Pgm^U09uFrP#vYRSRT&xJ6 zs#6%YxCBlAwvoxYp7iC3Gi6)o(3hfg?B6$r6-u)Bq}RezV^p<= zQe$plTj2~ITCK}r8pC;{&JzBUtQ!q$a7czfJI+L2u? z3-4@!u3xoj;B&cDncIQtV%y^R6IIR|Hr;&ruVbm%E*xGy@_qVa&uL`Qzqf!UU zKk5AS!Znf0P~&OtK|(+4CFMO-mIAzfJ6hK*Wy{-XXpC==_C6ht+sFR5nqX_7(a(z!CxHtYN>Yj#AS+u~s_))!Js{nf--+T-gd;>p7BR zYk>6I)}2&O$S|u>^tC)mMzvWxnaX5!Y1z0OV;hQilS0imKhv|weRvWoX6Al(5HTp0v}EUKL-=J1 zEA+*qT6#C4k2EPeMmj%L4ZDAiLEy$}+#J}MgCA_- zZ>sxvLTj_5r|4+T$}mLszmw7-y15yD zRP=fKwnp4Ch=Y@(=r!LCSOjjx?Uat}T;|V38FBn{Ydp@|EWTE0732TI4$Ble)h`v>KYhksv!fJsuSaoACO5#sl@fY<#>Mx%eNVE{CRI!4o%V zX%^si{~yq**uXbt#)dOW_=NHAs&;>z=@sNJ+1J^HGkn}RL|p4^0on-sZV=|S{$`-Jp5RknLt z%z2)5d^YSjXP*$=PGauuQeOZEg?IGHG#yjF8~xAu^KM*E)|(*qW!q0u-iFiC&RZ?g znR`m)t}6WJFK*K8)(A4Rb4FHT3w2doN3SY;X{X9&+Ww~-Sq3$8!aRvDPS?cw!drCV zr8eaZijgS^-lVXEQ0VPb)Gwj|?h6F}>O!R8eD%W(;a_~va#OlnXHF(lWl}|?x(w#S z(6go=cHcJ=T$&i_cE^Df#U7$=k_(fi+_c@+|8ZWosO3*%@d>p zmWfinZ#m3`4Dm(eHXbCUGKp0uZJ zrvqbmOEt5t^wZ)>wL}`L)jcnc0O4-N? zJIY>~CVe(^m#&X!CyfsqDp|H~m7VI^3r|YJN%r-#RCg{_%FOkoYjZA25i4IwzUK|e zt0GeRc+?ZOwpb%@b2+(ne&wiS-{|=G%^}&R&>FJnS3^6Sa_ETTbcAN~p^WGp>Bqi3 zJj164FPnW0mHV_Q`QU6R&hC%&;efY%@jnas?qNrG;+j|Zm@CeoXA|J=?ts7R)=_1( zrF8auKb)QML0Ua@t8C@Q&FFbE0d66_cs5=U{{6R+gMDAh`mh-b+s>qx!BMhXL)BTw z=OEm~4);KtdQzOCjN-B?%KCbh+Fu?eo&8qeSbq2o1?;Gy8J(I)Iyaa{Z|jV+{mw{@ z-XeS9P)l0!&WP5$M!`Wtq?Z0aWp5h0qeAy06 zIa9h{ANsiIGW=!^2I@!tQiLT9^({@QWPi&K_YZj1Zhi>HX z*PfQO%q0CcIymvZNiv==6yv+hCimHwq@eObEHt`EKEC;qd8mebUH&?D-hU7s4_uXX z))<4MK}u}$ccIiTz(BIN?IZhVEqc&Irh8jMU%X5?&BKn{a#hSw+`qBgaoOYBj!ngy zvUhp0D7CDh)ZhNJul@rj#=n;;Z8RlE#}>?)+ySXI!>Iq?94fhVN!s{#pR_Buh+g~@ zT@#xJNlWaO@V7{Je)yp+`=6V~4N^7spVWblnu}yI!dq97}ulA<-a%HS`?k!Tuqa=qux}3H?RCr=M$o*Qk!Lf>(`Q`Ho0c>dUv$Qrg*X4P6a-BN6Z?t7UJZlzG(RGB-ttl z(()B2$@ESFuex2tYdmi70KuIu7rUH8!7-HJUjrZWGU%EYqV1Yu{vH~|Hm@D|$(AW> zpsR83&A-s2W zMc?V`aCXdHDik^UpTqpIOY}ma#9;d|}yCHN*FIH=fhQ9A#deEkrUKu1%TE~Al;=F*T+O7n`)e-S86T{OK&e)a4tuzAff6 zPEFF~lWOSw{HVxXRM6&;MU*+XJ5{xwz~*Hp9AeXzrw@e^A7+61ZO zw>8a7bz`;Uuh7`;iqIGpxPG_|Et_hzmDa-V&H|)vo{jyk8n9_Qi#)ToN#(tEX7~n9xwD4fg%`4Xjl}hTeqnl`9$sE}Kued$;YLIkN2-3fa*va%-Exbze)qK==2oW?9su9|3}q% zKji$r?_Ww=p;Bl^l#J?rp2vA-B$TZrTV!NsZy{+=Nm`^mP?1q6C95c8WTX-yqnBBB zR^RLM+xPtsP&fDUd0ppu9FNE8bCj{{2bDzc$H&k{j9+KPDp|htC%K-CX4H{nwGs*! zGDzi1DZQ^X7J0R;upgO$+(XytXjc=Q?J9$#V>B)($oaLHw}h+;0Q#k(FfkV&oUYNH ztA6B{v7d(fKc=bSU15~IC8J^op`FoRm0H0`EU^o^`P`)lpfR-WBmymzO}*I)oO4H zvca_3snEW%0NrLq;`D^MP{~dM>e{n_Sis0Q2ya)GBHyPIcU|kw{kw*8-G}KoQd>eZ z2TBNUXhO)O?bu(w04MJX#*N5>S6}wQiY0}TRhojNA3K@mB&qY1ct?0Eh@RfE6ZqTk z2Se|5WB*0Z>8`&qbeFoKdgC72AK4L|R*Zz#7NRXaN9j%QTl%%PK+J(kVL0qN>;ijW zR+t=acNrBf^RfF<9s0Ll%HGL8V7yz0*VN@;msux%*5ezT+G%siaRu!2TL)japBQ%G z8TPxKgx&Q*c!nN8V(}8x@7KUG(TUXv2|~TYG_>$e%zV=a1@60WewU`;>;4e$>MQix z@-k9_<8gLb9Yz_(;K1KrIOw$w1JYt8jrvBKn9&C2k%y?FV?8BKx57ugcI2=yhLUDq zqtQ$D;Pe$&zI&7upu5AaFx7I@(WR6K^$Sts02U z-{)Xp(kzj=FU6#SzT(+=0uv1mz`))b3wu4LiMy5YQGC}rtq(_`GQpzd0A4&1uBx$W zsCzOP_s5(>>l`29#E5{t-c{+VwFMOqcOlQgf=w&tC$Lug=ue;ZwKz@Xpc%sO7vt+lDqg&t^QD#XG01 z@lj-l2u^L?YrKAT4SFeZ4B3AQXIAy*ir*tS+-DD)wS46tTH-u4P676jtOO)+cf5kSj| zy%Crpd~PX*>j((xbGFl#UbOUTx3J_-?d7gRC;pd=q_xRU51oGJE?MVCww$3 z!{iy!QeE9Z8aynK9{;HmeZ?%Eu6qW*1zX_k`i{IQvma|}9>kxbK&+`YM(4C<8kysX zX79JyRyu|K6&*QsjVg~zwMM70e`)AmM{+**7A;4j`RbAaj#$zkv9sn=CtRnTUw3Fo z?*Qsr^g}Whd9W`1hOtfRDW3K>jIF<)XaBPYn8$}nzRUZdbHFcr{boci)`yW%ZO!JN z2MK0@2Ai2~^H#hBs(Yjg`=2XyiGehC#65KKW<77Fg1v(~zhHKD@7uLb|#WH}ns{yNwIUGtP#i^t5?gSOiQAcXCS6 zPWEm-$WPC?VtK)Hy5#VVo|e9`w>fGkJ2p$1p2Y6q13#m9X|f;tUDu?_-6?Q=yn{>j zDdJlCSdoha zoI8_iM&+Qotq12FF65Dqr|`n70sQr2C@a+-;4HIzUUT&!D;BIm!0c#T%rC%_%?r75 zaR^68`SLuSSy=8@fV($NprhGV`*RJM^jbU%t)=7C{@i>_FB}P*IW3Tjx8Cj0MrGFfap0nfc>5Si+Ohc~P^L}v}}zH**9 zNJDn}*o@{;!g;&F78xI|(wF!Q+3?caIJ<2bcHI%&|9ui3l-lye#+%&p zwFXZT9Q3`_TGGX%XZhL5zpOj0t^A_HWK>&u^S^gtd|+aNG&4g1gTE|*$%-FzL!5jCK$_vQ;;oIWav8 z9^+@Q>!|Ging4W;Jc3EWb&6hKNuj}z(AZ?RU}tw zI~7&V#-8)3xPDNZwSv5PM*SiFyzDc}RENpOS{=qy=N5Px^krqY?WFfFjFvteP3wCP zM340A*ykzZua;x@W#Dw~|1yxXS5(uB#ZFioSt2rtg_6$AOLWa)EL5@NJi6-YX zQMq+*;fv9xl56|O)uWci58DPUB}M+_sK{?xMJ^=Y1cz47f|JcK#5df+H2D==xS0W0 z(OWw3wIk0Az6`ZvClD_>uPamq>$;;qJ6osma|+E{T})9+ zFJhb6<1Y0K;J@Fp(V@B!2jYyl*BVvX?S=1T&EJl~X2EuPzFEvb-mRo`9SuaxJB+J? zUAfyMYrK!|%+X>t`+1BZwQc!Jhs2%aztPsny%{NT@`7c3c>#uxHbA4zaq028Gqlm8 zGvtfCkm0C+V&SY@)@aIS9QI3;-$d@=9WbluHjnAhR<7~ISe~JH4X^aoc+xy09)@0A za%dH&ivHo*NAoejSdN%QYmt|?2D5_8v3*t(!glOIigPEv_(bB;LK|8wSI3i^ahUy8 z_`h5#pgUBJ=Rrl5{XI&yq5C(f^S0MAC*dI-Ib;*I+Rmh9$(vyECL9?)M`Zy^9Hb+o z0%>Y>lT>$OJ|*AlPEXty(SWE$uGSoG&*yszXcr@ts-%?@^a`otXea~#I^25ln6v8QPGFDC@98jNK&KhZ^JBoEFI zXY-hUbh+&s%xHN;e!q-y*|Z%z%@5I+&j;vnL?eCgeGBbgDDv(CG}g_ z9<%DQ@g`;lU*0o<+3O6~VGj~hthwi8IX72z#Pif(v~A;oZp&*h;DR47@jb`EAMK^r zRmUWcRSTrovI8QIdzFo6H*s|3bUgM~g|y`$-R@8)RTaA<>vl)cU7Wzbvj+0?v>bK{ ztiYd9xiGACM|S2~6xqb1Pgw&>x&>ikNjvT~b^vegFXxDX5nTRwFh94|$F0ikIGDTv zqn9t?Zytrbty=qk`d?bNN<=P~k!0Lf>=Nss{H7HZse#bA(3w}A+00Sy`?>#*WSaf? zid6RQHm!|PmOrn0#EsXNvPrHvm;2n|-(w&*Ja=z5ESCajZ3X6VZM1|8%BZ4CIO=>hyU z?2n4l49ViAD;T-AM2;%UrBv0*4b}84;Y2FlQ zxup?3Uc%;wU+|ggM)r;!$lFJ_a9vIz7k4y}>tBDuZ=Ssu9k)49%+^HQHFwIpV28{s zU7SkYP5Ik9!pBvSO9I;P=tf11SuDD8L#<^ya!TbMJKW?6aZA{9;8rC6$irt>3FUt@ zP?-=#A1kZLE~1}&w~Cou?_m~ueUE0Zly|%^^(S8&dzV!g_hpmR9dfsBD><^#i#K;W z2$vtv*z5LY{*&LHQ*?I7pSpLIKe$lE3U!HG?^3`$toCvDZ>jjSIhrGvEaqZyPo6-( z$tI~1pA%9c3u((?+G^sR+nZ^ki*BO^84KbwRHtnn~XP0HOS-s>N$_?cv1tJGp-VIHP=V;m2PVoPE zm*m-9vm*wPeX~&N|F?7DX;o1rLH}}c%K-KExJ;k(hI?RQfVbb1*_FYZmryB?B8=UiI6 zc^0?J@Z!@NS{!<7xineuLz9CB!=xku<1J0$uq=%Z%@lseq-^{a_kNWg$*i?CpJg|W z^TM+u*l^Q3C~4e*Uc3*Tj||1okL`IxXW<8**oo`xm*S!5Z)}qlqu9$^yz}~UYwLZU zoqd%%EFFyn`UMNr?DFH+^j=vuYO!| zI-rBsLn?(YGYsP%cjFae_WUJwA`Y#2h&Ts#9#Wf(8qK3H>K2MICtRuj&x2H(<4S>s zGg;HqjVImh$j7x8v%~!yHt1H#M#lqjeB1*h9u-~7q3`hRo+Z~fzlO%(uDEzf6$e$m zlCIh(TG-|bjaji>n%WouWo=#dmK3<;u|BI6m|=E7wd5Q(TJElXfo0z$zS}`VuD+x{ zFL;zkPZWOPyyvJk|rMI*h8M}t1VX#v&Y%6%gDX$z~+Kia<_{KhYvSo#gTRF z+5I0^MLpo{Ra0TsyqcpbU-CgCKQ@3rkM!-rT{=%j%q3H(3=>}L!WcZS+X&W7LfqGE zc1t|K4j9V|_Q~O;&cM7R55L@(}M z-YlivHYQW6B+{}y&B4bi`DnvY&f78__eLz`9!GQe@Pcx-Ozb50l$*<|UVfm@50a7L zcLO>nt7tXliWz6Qcy68L!jIorx44y)=iR1nXMACiS4p)G|Jgga)llz>5VY~U1da5k zxc+nt05fl#h(v$v*X-c(S^Gt*d9T~ zurzmI-I@eA9W%g)0q3yC>;*=RlfiU+0XnGe=NPR*{@aUnYlDlA`mm940>7BJPp+07FSj^!T0ZNtow)x4KTOw1?y0*(?>`a`oZm>QH~rx( z_JYQaWf=6|Z|WbBjf9YH>}$6VTe@|?=eKoIrQA)NE3e|$^H|*27lM?l794&jf^8=T zVaJ0)T(j-ZzTM1WyLA|J*pc35XK@2fNr1wsSs9j_t+yxojGJO_L+|!RsU7o;WR|oka!RwFv zFppnVhVo^>La0t3A7cjHK}V%hbUpDG(*);b zT?fIL$(Oce?T-x1Xt7H;8iH9ppW&Wqk=9X#FzHv^{{Xbo# z?ek=k!?Y?{*{S2QH9LagySRa7X&#}7yjubs=}xw*0+sZf^k;<7Bo={>!&5`Bsb>*#l>Bek8YM0I+@D8b8?a<*KP!u{_H!pC{M z_dbfh?)xw$Y#Kdpv7q)|8)(z$aw+(}ztp4eDrsQmQ>6TiMvcopTp?35cNG296<=ua zT5W`!Izv%rDKx4z9Aow8qW6>m_&WVNd}$X-n**`GqbnVYej#}s)+d!$9q8?<$&!2K z3~66(TbPY%rOBb{Fqgc@>tzSpQ#Xp%_nV6$7M%qztAXTWt*MXt4C**%7kz0HNNNxL zDSDmoVZWYCFD(p7Tho&a65Xl5dIs71PA5jvHD5@EMp$4`hL~rcLD7R3(wMtuboa+x`t++EoqV{IG$yV>{`OKNo13$q zS`#gq;YQ~luBMlRQnipm)AMas(H9VALf_QVybmmituYJC7fMabOj%7g*9CnRYw0NgwUe88Psyqo=#PDqi=&oP(sx~sVci#@)_O^2mEf5 zneSP#Pi}`7L*A0D${5Uju1doOzFK-!%3A4R7p^P-FKJmQmZ2oIb= zy>xPOx=`W42U7b@SEQ{=uF}ZW$<%IiG!2T5knEc;NL5pg zOUK{Wl6z$yxzA~$jr-5SudxaC7Yi`r{wDdrbz|he?v0fHdZfl(3Y>Y+2si$$QX?IC zwp*$RUMmHjwTDHt4DD;Wz_cuqeI|JE_98KVGqS>4^%1ZY{i49-pD@+52Cv2nud9wZ zKe}Iqxs-t1E`^j4v58iRxxiqxUQ(`uzLXWDBwfjm=EW_!T#ymZLyly_s@f5%=2OJ( z-<6@%fSl)|W8Faq0i%cDlbsh9h>UmMlKD^!JcKR6i7{qy6#AO(M9|dj2wm5Pue5gL zki#ylYCD}5y_?9aB*)^OeNeZc4@SI{a4=Wo)!%>mpZ&X9kA2X4HWR+$e*L}SBA4_F zW9iHa4%NPkNhXS{zgM_LEXSZIJ6vRv!|?v_4U~v}W{_DN?1i_)FLfhEeHNalQj$`O ze@LbdPo!+Ir+xAv7!ehhaK^_HkxPY>p>unrp3=ruZE^3gTuTdtH~Uw@D?F&l;2uRf zd`ApXoQhPq@3$iC@NvW|zbI&yEtiVUB$LUUY}#0~0yjpPz~R2wS4=%d^{-?os!@lZ z$SgcqzMCIy_2(9+LquUGDA0Z&wp>_B@naLIi(?yvgxd3fFYavcBal_ZJT<;-Cl{$( za-ZXUB_|b$CS4m3S$Q$?#n1a`h&Ia{o8T0v#PNMJxP0e(x_)OmX8(xBlw)y}KQxnC zS42?QtGTFrwF2}18ll?W2!)2i0a`8mj$Ze_InrPv{U{X@a)c1W-NuF<4j zHn@1Ml{VU!;hx4ae74X>l#dhIMW$oA=xgf6jpQT?ViSd{ymrJwj&@(l#VVpBZdJxp z+gJ0c*>8DmOCSOxfF%Zp5mGRT^HtOMrFRS;G`L6!QfIswsmfd>z7EItcw=E;5oxQ3z&oiBBU*aEXsHW|(mKBG2FMa25LvGgyOMq)Z7<6%CFCazgUBRiw>sK1RcI|M#k!!8se9MSRJ5{8!<%KR&JF#DZRW4?b+EWByX&3)rpIW~*CT{QaKJFZ54VlWHgg+oduQ%_x_X|;jtdS6@$U^ePLw^4!{Kt71 zd|REHW_t6*xH;_oW(RBe4q?OEzIa#r5pT14^TaI?$UYf`FJ846z4svp3=tf(;kNSD zbFb0mqXn;m1J4bsq4oC)u)U|!|D1o69^OWW$v@$2qsv2_y-_#H8JjhfAm1M&_@qG? zA~LuAgF92!o+wfp^^?XbtI)~J3)Hb`2gN3rqV=N{N8i@tv%9*I*n}j$jJ?cXblN4p!6X%woE`@j8vYEeR&vRV+Wh z2kkWz1gB*!rfjsqwbbph#oEr2dhsC1oYkoAq8=7GMd0#^EIb&n7gIhSgT;aoI3cok znLS3rL){;zgIe&Vpc-dvf6!T@QNqy?g}91#2z>dSI{wtc!|(Al@mOEV(pgMxx6jA5 z?tAg+LnPvq{4lUqd{*7Y;j3jGt-Pm)2{X1KsL2buk4IwJ&;dAkFNZAcawtF}nSA75 z>Fd3H!sU5^=C_37)-93wOJ0Brn)_+gkv7oOvBlXdH}Nf{%Bnu zfz$`*q>H!oU|^eujoG%ib>a%$a8X0!$w=t5ZiUtKUDz~U1>Q|oIG;Tje*F`u&(i|p z8+G)~w~<~JIOC>ixq%9w{J&8`! zcA<|(9D48GCHh1Qk!B-0{y%r1kCy1ePFLZpg4umMYYK+CgrZDUl@I(^k2*zVwml-e zLTPuXz~M5@F%lX2A^*Wnxfuo5x}v0`8{H|+p#hay^tGg(49X6Zx^P9?>gAESf?%OE ze3IH_&Y+yc!&1nhqP2_Mq5ffm5+eyc0N$?da7_W<`N?T#{_XN6ouSZJLf3WIu zPw=v1k#JDBCKdbOnotKJV{znFoZmW!Hlb;5AZ<^Uq(V^`8uI_{ey~rHek3(s_ClQAw8t z=NT#&@nGC7{M$AkdqVc({(xn8$8Rz0dkBiZwxaFiFR-ZnglO@-9_H5wliiv4z2zv5 zIbM?FpZ7{BmL5_M^&#x{&WF_m10%)rv9$GQ5k*88L;gyUuO2HvwDtqM+8l+yGY_Dj z@kxyAS3-4WPpI$Y2h_RB7r({x*j@iI8cSay{)Rojn9+>`dlD!2T+LQy-dsG+8bfUT zq14$A4m$$ib~z9w7s~!;|9;-&JZefS@pol9I%`kl>|-J9qnXbMU9_;dRLuIuwdI>3 z_UO@g0?y;}(9?OB$c4|tj--d!_w@^=517rx17zGZdNcVnnqlgD51dpiz<}0R*se*( zu5N+os{a|m`O4Cv@0EtnmwqE@CdjJ zJ4cx>%cUci-caMUKV%qC#-FV2D#e6n3nP30>#xmIg{=3DE`&%629=(GxRbe1Dra7WKbvJ@sj^S2j z62^TAz|kZRbTu9)7;Fk0t?9;_d=%I`dlC1wh~Qjx;mLB2=75!tS?SDO9<@3Y=&Qmj z);@tsKm z+i`##3+~g-8>T$Q*ovo@&E_8~rgKt~5eJeHzc&^0%kNq|T>cp|?@3|?9mQQuj&RM- zomk%G19lGHj}1*>aQ!7Xx|WT?ar+M@f2Z@Ggj9~*{FI+e`Yl+v?d9Ef7$JY@HWVvP z!r0J!wj6qe7e|+|q!x&Z2l1$BT@B54Ds*!aaNDMddab=cwbMj)`u89hJj;;-)Cj*($c0vlKix{hKc1v5^$NV^MMrKin$F?k-WcqV zN{-7;(BGO&vTRgEaq(*0+b({uuKpPJSLC(E48@zJ^KraRAA3HhbBC?hxOsUPt4$rw zqaSSG^I})$BLHnbhn(TfSv&b_T_!$!|An9D?RibgL;89BG$pxYkh1Wg)P$!%xBeJx z4+>Apmy#E~8qz8P^2AjDa))ty*-z-_n{#Zi z8E4ij%B>&j$~(62E^o7qH z!TfN0AvDKcfy4VNC`{{yH4~5U&o3)jSNO8GjctR)H|-GDzB_r5=%bFD&+9kqvfG1h z(0S*G-`DLh-KLCJ*k0rj@?3r~W*lC(pNF%#spwAG+^#l`%a-is#Kz97*|(c`H<%zr zO>juXzn$*7LfnxO@$y@H-jF5ZxQiY5)ju0f7Fpott1|hcrcLtcRkP$C3p$I>{+`x(#LuyvicT|b0YhsKYSqE zRw+xKJ=1CY&ppT#efV`F7oz^!SQuy)ajWp^o*j9B$;k|*E{9Ry;}taawt;n|4vHH! z;HveVwjFWD2|9^-mk1tyWfsF|GQY6XfqrQZ1Ox}8Yx8y*b$2Qc)@|g!bwjY&O^Z|1 zhVj~r9I3&p7um)6(YT&s8=&jVUp6*##P}=@nfg=sVy5wb--1ZiN6eunJj0|{vk@gH zzPj=VYhGSVf$Nsz-(3%$eYswAu`OBWQy(7v=L3b>D4=plE=@b9%bnUQak$buc!wTB zUA-%B`?7~!RWIXYpcZdzYr#!QF5JtoD=&AcLfr3g;XY7?|5kVS#Ok9!wF_Z! zr&ZHDUS|86UABJ`5XD%SPrrfx-q~`&6h-OKXA4R?HWNor#BoUb+x+>?SvK35B@JJ# zgh!FHxLQ)tSWZJ(0k7Ca8T-Sl}_^+Mh!SC-0M{*!(n+e~8s+K%-w+X9s*v8&-dUBDUnDZ_CPC8bt zwC%3Qgb0V}jDLUl;Hoa9Uc3dWn-x)XV4VH)-rFR5OBt+m&!g+azAR1Z%*o4pO7|~$ zNMU2oQHA$Ndby+xr)BQtxu5Ou=9MPj|7yWqCln%8|2T3-Z{oC%b)2qsl$A1fu;rgE z==r(_$KLhhlJk1<#{(Yo_G^tC=wioV!S(nMyBi||tE9q+-%=IJXqc9qhg~w|onc0} zDYA=??@fW{QeKiHpVch_dskd53Sp}o9!Xa!e2ww9lYp9ZJP z%Wxst5qeS@{zlE?`Q{h@=l>0ME3BW0K&B@?;0hj5-h&hT8 z^w}_-x~@0Ebm7(f9=`*D;&b{sUN}1zSo6RkGO>#>=K3oyX<@18^OXOCZ$w)Z+&xS$ zwm8#|q2)Axv7&GSx512Y1z2+`7cFULaiyd?V%>~T@O~$Lr#)k(!kX>%tQS z-?;ho7O`iIk?cMD(67ITrLI#(L8Y`2^0Gn9gGaOeQexGkhwwU1um;qN;HmNjt{vZT z>z6}3cB4NZKfM=iG=-b{M+2_9^#{)R!?k7v{;KI9xvw&`1`R}n*FrRkcTa)y9q6`x z!bUOcyLHi!gTnpTy?8J$wW&g#ZwWLbXX5QEW4wJFPKylhQq{Ri+8jI#zlHB2cbXFF zr&#dd-QOVpC%j9SCg5eSt#ndlw`{9%2GYtDIW^CUefB=Z(Wg`KEmIG}rfFcq+F;auH}l9MG^yKN0Vy-BAT{Z%-?D+`x3_M%#xcXoLEgIk0@u0D4{ z#j0-T^id1*w?u>E4x&P}0Pee=!hb^%hPIr`r{Q@=>29_j8;%&ro2|#OyI}+Z#=gV^n+BL~Uy4hDG0?Bt4ab)G;{C(XsCp2J z+gAk($>Wk%O$yDC+KdHB$Iyp9cm-@ckft975&_@DRbQW_0ZBx2> zN}0xoJb=yh4@jKXjSEMNh0<9oBx-nLL7znQ70=WepGB+gOm$*xwk|C^iI!4-@z&|nP`In#tBLxR+ant9L(QSBY^b5?1P~t1y zy|7sME?E|AgTts0#2y5~?#jsFsRq7Lw!hpyIz04t6JSTbyfL*J*Hr z<}6mTTfx5h1Nif&0a)(&TWT-%B?((f@L_ExI&K<5`Z`f!b|X3ke@CJ6#~*qgx`Ia6 zF2=ei71&$08KF_zcu1=m*QmST=KDC(d%qrIj%#t^LU%TN+E1`JW^k|nG%>074%*pe ztW^E39R^MILsQsp>UlQ+w+?N>-!XF$D0tcaK@vOa_Tp#eX&5`E8fywq;pyvbc>ZQM zZLIQ?jo0iXxY3%t>-IGq_&pB;Zvf3#uOqbIVzv}MJL8bQ?1(S$9~gwE z9sXgz;En$O{fn$ogZbl-7;YP~gCC1KV#6ChB!(AY(=<2f?zi5gcI6bDFI^}9J(W~m zdzxwoi!)E%K(xLP-ZO*aGKYi@Qnxb~#e2|3mi{zDmRy@6Yy3K!YOHRO-yU(G2yZXb z`0glu{25Ffu@$3tK7{SZ4!q3rI)}Dx;c08v3h(z7v|HZ-y(3y2zikzL>eUYBf7;_} z(G+s%W{b5(Q&8z<%wt~J@`mviT$1#H{_AdyUO#%mM7CVY%GX4rW(STh_LK5Ah>p4Y zXLRhOL)AB~NiDw5q{bo-M4dhkkDuCnEp#eg)d)IWmoWDDW+{EC)I_^uYW(%iBB@%t zJGM`0$48%xmW~WuMSsK{Z=P-fX=>YHy*Qs4JBfF>j_BfwpJ#8t|MlDd!R&1d>5c0k zGSG^6Al@GiPpv81A`sJhy}`-UM3kkP@y8OuH4mPOi3X1izOna?H~&I4Yd)le;xxhLPZQm;~~!ZWywa{G!;n)T_9g#XzQ>Nn1n*ci0WQtP_d3aUCH!XrTPalg% zxq0X$-0|lo3#PEUv;A(CLuZh>^60dlg^)Oi*E%{w9{AkXZtr|aIJy{WdUB!1)fIH?7pMns@C(zJOMswI(bWb$oS7M&~DN2hwP7>as;rC_B zox9-TyB>W0{%P)0Fo-wn*v38s6As#Xe5}3` zr<^X}xsK_4dV>w`>{7*tV_)*jU$a?JI8v{FXu>v|_Y|q}nXK$%DdvD3{b3VITYr|u z`CP)(i@L0E`Qmzxb#h#b;vA$0_1;*4;Q0rUzHb=NmdM3kJFg+V$o(X1( zF?zqwq;YHaAhNy>*7S)&$0k?ER<~uvx{Z9Z#|qiti$>D%wTtNGMt{=k+ue3+UBS@SJ1bD3r3%aZdx>}@gT z;As`8^Q@K`o^_I5IF`sBE=s~TrCUfdX~5e{o;<_Xg~$Fp&QY^<5OVe$+L!-?$1G>~ zo3~(zvODkhG{)bB!?99yZ%Y<#q$iUC>8Z~IYFYh^0;gS~+n?IvTGBOWe62yRv?+N0 zbph5Zwc+r?o%oYRC;0j-$4{w7%yfT?GxUEL+5B8MB|p-t@Ttfu68G==^2RoPsZWVPQgdba)a9bR)oloVPR4+9Z`)CIms9-IM*yKI<;PcUl=)U z?TM641>*a37ZoKPxq>(H#Gyy{roWryHPRI+xtHLSyo~e2PIuPkHoWrdNp$<8#@kl+ z;w6Ls(Q&1>Ma?EE1@9prc~PY8tKlNgV?pVsIj>%l-zuv)p|#= zRa-~ef)|`**PDm#8zef#gSn~j5p@!u(SI)f*e~3GyUwU{%V)uiJ1HCwFXyuRtqmfKpdU|Djc@T*OO^`R$}y3GeWTs})}+?&XHa5<$%L*&&T;;#KD zUa3D{a9{^hx7eqWiOm8y8D>De*BZ=DUQDwmZlL?M{YlB!m_Ljj%3IwHxh%Ca4~|{L zi)c5S*Zv3VYqxPhxFniV#&XKyF#cIKhpoLoqp<5&)a+h@^Jy6rYcG6arQ2zKgcHrZ z=!vvI;nZokkIUQt!+`1xSjKuzmL66rz7b8FZwTg4Q;s* zLmBrBu*^LORWcQ6S7r#smFeQ)*e2?Ft(XQ`UZXQF_sG06rb&Z?zf#}ytEBf}7xu2u z=B2p-vc0W~rQ)dGQkz^=QV7z6>+Cbg>70Y3=R-LV5l^IXMarmS%D=ZGo>R5 zqo^`pk2+|mkdCY9iX|vw`Epel$!bL|GMwLLuBNgYBQbvnL@z7hr`bN79?nWazO`@+ zOZREX`%tuQ2#}`TA5W1BGU<;+J3932Dg79D9jW_1)0%)*e7Tdv`3hdL`gH@O^xUs9 zk?oL#1)j!S8iumPgYa*5E^X-~9Mcmd3LN;3RBW!J!m6(nn42bLE^MS8pBBQ^NwB~7 zdT?FSPRd#Igj^05VRT?9t+`^3L1#YT^_?{8nAIDm{u=Bsdk<--1|UB6D2}(3kcGSt zMj1z9VEH4;>z50KHo`sJq{eo;r?J!82!6ZB2N9-LY?CKu>t{}|OYdYJ@MZ!(nH5Mu z|8;=5SsaoQ?@Mvt-wK9<5^_%%<({GSBhpaQ94u+HQ^&kx?btT- znQVQ0Z`xwf0p8ELQ<6CES!-6JxO$hAD()R63;N=q{G05|&R|(rrzHDj(J!U+FVSSC zq=V)A+DeCG!^Bx9iPkl~m4dQQO21vz5Y%{0auXRSXZ@}i6PZOl&xhdWWqmQlG2(}) z*+lUd ze&POz1>7>cKMmLyPr)vEq?5P{-@nK?-**;g4bKvNS_4*D8pytK;dAYw0%?gq>`HxP z6Se0{YXi4Iy)sqWH{PAPw`5Y?_(NQgq$OYXe7U^NN&#BkXW&8l6Ee<@lg_F2$2`vl zw9m=Fc9lu&(sY$?H1}lNcb&P-rpK7(wi0{FR|wzoc3ifMgl4M|&%3aPwQu{gudAWF zHa$xIc6kW(tB=Bl0(*8})(upIqvBSQkHDv9AR?YEEFM-70+Z z%BGqR3vqFaHI?2OL1oU0@&h#ra+^9g4p8q%x!DS`<;R?4i|>r175hV_fU8|)my{17 zBVY`adGh_0)WbZRdK4#pFW4_FX-H%>@{ zy_BD5oX0KkE-yEZk@lZlBkj5}N{V=|$-Rv)a;@n;J~w|Leu=&DMI}d6X?NuII~PIm zgM`MJ5|>N4G(fqZ?6=E4xq|ut`Fi@y3^{i(9YPs*gQQMdOoW@$5X-+zBbzCG_vLk8!GjBNlv-BR9xbUMf*&k(GblQk@qQNQ<}ZqiAji zU@A8|Fh_22tbmS9SwdSozLjp*JfrR5JtX_44wB0&!Kpvc6FZh4q?svwslB+L^nR5= zhwDefyE2&)+pHxo!AO%IkCsIg7D-1>q~nb6^-k-!7yA#FaJ$wbc(E;Huvv|IZjEv`Ykmeighfmk%lE#){(&HbKxxrx%YaO`4 zGqfA{)Tg6@Q@56vzZfI0i-~53?~UlPa-QT?by}*vbWy4q@l%|eyT}bkyUDAU6|z@- z0PDOLGwcox2yxp)`86{MJ*LQg@R~0vT;WUk^_W=Xj#u)Fr1Eb#kAEQMxQ5T9|9}3Y zoEgXWE0Wk_|4!B}pT+Ul1Pk|U07p1=;RVe)++XLVxDTbFlZFHKM2*0gkiGB|@BY-= zvDmcvG{$&jsP-$42Bd<>0MVBI^X z;k$1mu754z3nlq{B5pA&cNxxAI}KO>JTTQj4Nti%c6TsB%*%zC+P*Iy_!?n>dpy2v zi^JOY8koJX5FY-^P;=E2?Jk|<;P1;=y?HiU{7XRj_!2mnxWasX6%C{qs%I&3{>Gvh-bt?H`G|xZhR5_4el-kEe1-_!Rz}c!+JM`*4R@9vqu@ z0DnrxVtuSJEM5%6l@YxVrt^YUzpB8*zfbYH+!cS9y@XPEBR=^}N1$#$o|`C~q$Pd0 zEbcV&3m*MHs?PhL>-P=&lv&6MQIwHVk$9cQNg$=YKIG#@&8{mZp$`SbY@&HZwbd84Ji98poivaWH@RX3WDrT!OT&nrpoGvx{;Rx6Yb{ zt4|JKhg&IpMW1+GvjeYM+J>w>rMOpp5j8Fc5LPpZ^^8Q8Lpxb`q5H*El@B#TY8`Y9+dsTzqf<9}BhM zE6y7+VQC^P3-hrfHjg_^ThE!o-{$M~5*}-_QROE%+*bp+=aP8lZ9_S2jUAhJAI_ut zDscPP3neq<8PYSS$Fo#=tyZtu!kYf2&4cmd%_qKCcHjmH~?uwlRk1dZ+w zS%+%+AwQiR4^3l^(c%&QeNeEqI|f*8gc0is-@^g=VWxv$vY{BrYoTm<7jx_5;LLw1 z*wmY{|NX|<@Plx9>Mll<>af>I z!OEDg#kZ0lVQ2D6zN9=$aHUgN{R>B8{__Gl zvR8Cpy1u0j!*0;3vLrJ4`G7n=^vAs~`|vp7C-o^jL&hhrllI_3>6Sy5L}BHUXVd|D zmpPLZK5ZbQ2a)vaqc>^Viuc;7Wps~+(1fOGlp~zjQ{6`*_((5g7U+{CHK31M zFG@LYHKe%>`I4o56Y8SQV_EAnI9MoQWLXxOJWeL#^ccKeWRKV1?$F}*!ll+@A&#ye zfT0rw^E@gICtt3|w|%kXK7Bs@m${6>&j-`*Z^xvNz7HfBn_AkSS4ouvu2Z0P7wq2u zfvQDE!DPJ!*6QCUPS-$p!+bL6+?CWGIZ}c4XDLr|r4;);aD=fadFE9uJ`irf z`{XTQ<7hx%{@eEd^1hnY$}oMCMO%^sh^Fr&gU?6ki&6qK|LLRcU0-DT?xIgK(`oTf zFUlLx54k2b2Cnnr8rkVH-R)6K>Rl$&!-{p3T*c(;Xio0~Z&B>u zb2Q8T4r!{7pkAI{wBW-t>03ZQ(*7@&HhOKQ8Mph<)zx}bDI6qLm#xU{Uw2Ya?m^R< z{?M@*k>s8fMVtGrCdHJA)JH~}UcJ?Y`HAP0{d_u9McI(7vX zYS(3zo}I?24T#t-*#DZ-c+&DRNh|a)y#R{wEj^yjLYy58C)&&)M8wE=M4L>K#aES1nr~K@#>KY4%r{0#1Ju))gCR;$wcXB zt1cg2X2?Sm*Yk`KTd{Im0D_*Rh;y}=>sYNszUn~qaLy$;?`C>7eG3xjw$u29atMqu zN2kgRs40!Wg#V&&OE5{K1&j~k??F#rxJ9nXvWmkP_U&hlR~Np}0=KT%S-b`1r)Hqn zyD9&3|8ft#h>>a<>~QBM{M4>t$9s9cS62$t=2+y%#X?u)B(D3ZAhvoI?Ejh~_rE;4 z`7N0g6dsXkNY3h=}= z0fkG~V09k{yt?YZcY|IdO)(Pk)#p%A7X#N*Ik1dOp;6YW$@|bTdLD9<+|&o+wY>!_ zWHR7n9*MlgqWdtR7dosSfY_62sCN(Gg{9wcK}&|^KkmiZ$>RU}`ve-ZHq*NFuGIhI z1iIxQatb{@qmQ>6&ofvB*^|?7bZ9i5o8_VP;CBpnX~6_}J@yo?^r`BWoOU`4o7yyZ zxHu!s44;B7?ZZ&p7KpcDWz_3)DUF+Po>CJS_pkb+Dt0@L%D=^Sj}~+bcDm(5V#|C)vl-vuw1Ahb+9f=j}^ z5^oiZ>ys9v^ZPQ|tq?%H24-RL&zX2LXcR7==nqYqFod1gU<;8uZ|h`=)kBnF*U=ZZ z=M*Aj%L}xLJb^{W@mOIt4?S`R`vMP@3_5a z*;)bL7bkFj?Q)bpKY%w6Ekt+MmPgv=VAN!DJpQPR+9q4<(i9mhCgz3Hb_)Mk zAlA*bM0O7s$SjBj`x_wjgp=?N{-SQ8>$#X0lKu5~y6Cf#LgSX>o#_hU;aY_SlNSN$ zrQ}>GkJ0CfaYa@58_d%1JX4c%0!r{9O^$=23-C)ujko-g;n@on`Qer()c(B*#Z^;r zwXz5r_Voxaw1=~qJ{I~KV$c3cDE%%%k=$0qj$eU+T3t}zew5BDEJKm0Kc=Jz-^Tul zG$hKIa$?$~haVSUrf|Crv#r3B*8}+Hn|37B*FtYXXU?FNJaq944p=aqHB)8y)G)GqWQ=>qHJJ z(2UKNH{k5Vojhh?4EHkA=ERIbJbG~hXI34@hVoLJO@E1hqrd;p^QWfgJ}5gxBKEC5 ztF9Sl*TL-b>izQ`q7)scIf!*E&j>vm*l66lKNV8r&B8o z_^WRM8aH*q(ozS$HamouJ9}`!cq{&-H;`R-cjodTa#*-N3rnMlu;iXD?^~6C&w1%I zzU>(z^853uU_+KSQo^q)b3XKPH z;KCAJamM^4smzEWYfWX0yy(F@n%ubls|6RiUxL3pv9+TQFA2Yo;FY~apKm0u93+pb z*=Eq^v-EhsDhAiPU_AQ3D%yZ24Cut~qNYM)pqTfM+{X3ttN63mPkuL9jrB`8^}c;gqz@rs0*ICfzS zQtSVt;k{O{;=pgb+oB`y-R3Q_vcCM&I+iuYE3p;&^2YLW(mjKtR9V#-CvGXwe_u{g z@9+}oNy`k#Ke<50g2z1MkKjPf5I*!JaV$6JBrWUgh+Q_@F??Y=mDZ`^hnGHNj~5XB zbQC477*@-5u*?3p-gpvdO}2J^n# zCVWEY09zDoV~<{m$gdE)lO8%aWHScKuBmd{fEIM!XNju&Vpg=qn$O)ig_L#~wvEcb zi6b)7cXuzkX75b0>-IusiI|J+NJqnn`?f8*f?m8JhIH=fqoaFAo=e*7;?+QUg9QKF zz!m=gH?y2`9cr4>S!viTp3=JuTfQHTqNWG5bI z=VN@rDw_R|n{vaGe8DsjKA=7a@gVS$;Mm6Ex%*hVO&hMT)|qv@|7j6TC^zQ1_Pv~O zr#pfs#ZcMqXncs9WhbwqV5e7J&TP0Ii3MIT&iPG2ej6cwcPlPWeD*)jp9tAAR6o0b z(zErD|8*ySd|b&6c@s6^kQBS(mU_Ih z=tJ9jEC7xkw4M7=pMLq~NJ_fX**Myh?>rM73)cg*LM4tq)Qm&V;r(r9`S!Mb^wdnk zn4#?EsK+)v^AIyPlyXHMSSL_{FZu69Y^OBXsvN@ zXsc4BQD?@}F466ZR_lm^?L(-YLL^K|hH7xTWHSrd`H=q;7r z5L{f%g=D4GAUzGcN##eqDJ9y42AKbr292aB12XMB9|nx0X@(d)^9*epZm} z&hfCSd_~dj>ujCsdcb`3B@ETsBRt-XXtO^LhaK~!YfD048t%th6Pj&y-=A*VM@F#Z z-fyHf9a&OxU5qiKmQY@PBbDDcPI#KZxAeBLYu-ja`A{3Kmt^oM!A>w-Jvg)z*i4*{ zn&v#fqn81^x^6w|sO4~H?+SjLUd)e#7k$#tG^l81lkU@2{!;OX&+om)hw?5-cW#%{ zMsenty{{b4CrfOi;>B(I8tC=*T^MuA6q8L}(tjt*$;v|k{d)d{(+OLS>*2%qcga)N zQYCD-7rpQk&}q2c(F)JZV953?;b=KR~-bTC6IiZ)NkP*4jPTxj!3OEwyl% z*pudX=1}PKUG&&AR2t)MfhfgTOc%Y8JyuctBeomAw%JZbhhK@_y9cSP6P+NPr?{k| zi2Q^ zFkD+pj_-7Z_ubsqE~HtyymS)o44IC{&GYc#s0xSG8v)m>ad`9vZi^akw^q~E&LupY zL)Kh^*NgX5wo-%Estl*{Wev6^Ws5lRsUlx9J&EVpE3hLmQySW_1}XA~gws11r`G1s zv4kwxQaauz?8O-W&2;0V3DiV>s{QX>+pd{YZ9kQ~mo{!y!yDuN$ckSGFLP_Vly1qK zal{V(ho{l*)9+~7va5*qT7sVWp)gP{L6;g0%}rU&{;1tS#L$~sfbC(h9AgXb_H z@D(!B^!e|me^|e9D6bP-$@kOT?0P2|+b#GQ!?_=N0O7Og`zHdXGY4)Kg+F?DCws+WshKWnzFklwqeU7v~UX)!t1f&UMOhrNbFj$0uF-XFN1GnJWR||Zgt0k zFFELV-%E5#4Dn$rVUCgwa&~T~jKVE~iR{gr*BPK@rvvIY=2GQFb^6|+CnYWNrtJ#^ z!&E%qMr4P>H6vee3&-$>Q-Uklxr3D1>9p;IA!2_UbxJzzTSP}aXJC7pA1&MTnHCw1 zLsj>7%E{M;={3QdeQ8B6yDz0RLl@GVEr$_2tpvVLPQsyTK2>j4#N=2nZ0vL!aZy5d zHY$`~3pRv<(-$h|Zs=fqlj^ng;J@G;tt@hb=3xn*)+x9q@~HCPmtvG`BwF@Q;=S7T zTxk=^>Q`mBvDTXR>@j6m^GZHAv5w>Rb+8*5=>|g?A6V=N#DdTyTo)bmP3g<{uw^84 z-_OC`uyhP{cu1GmPQ>M=+30#H8_WN*;0s~vIr{2CY5I=@3R>}rIu5s|dA&O!M9BgU z%NLUV&!6<+a$ig<+=Y4jr=amm8gw(hAZNV~U+y=O3%nmm7t+e8$vujGOLNdS;4YE^ zd-2n`ew4jPL=3kt#iW4;d4*;sYc)B z2)BnzJYe2tOkMdCs}Ae))@_M&QS{YioOO6shl6xrsV$O>&9Nymi4txff!6mGsO6X{X{_e%t zQ~#R!{OyKmM^&-)%Xb=;9UyYMtDzeh#olRa*tLB)TdrM3OSfi_>lBHCW5ixW)r0S= zb>_N3Rs3M6tX;y)Ha@;MjvTJsq%OBJXt?K6wB;-FF5fOZPJXPIhpFSV!D4Lp909Y# zUYPLG7J=TvmD+MdWYJ}@cI00A7qXYwC!DGp4A5LM9UFFa;mAYX?7}k)?N-fr$!Dj= z*!6$yWXD^U*hNmh1h?5GSeKKFK9fcFT{{AM&S${+&Ia;x?}V%)$?z9B0GFZ~Jiq(` z9t~>T^^Y50NZrcUq9e)rbu@JmxuQ607xElME^&+m-ZY zThgEzs&><=-D2z)z-3g4#SAI10M!^vZK*kcXucJCtQtnDqW&JuI* z36>O_ahK-jq#```E7Fg)lgl~<0Shsp%Kn!$*=by7 z4)6OJt5kkS4s{CXKdG7~nJOdB`zz7=12nrLkq*APA%zEYmDE)_ah}g3=;s|o+2B@6 z3(LZgjlv;P{StqzCZN;XK?LWUl$y7N{BIXZ^~s*l+Vp@bRW{JrT`p`~CETxX2Vz~* zB56c@71>XIM4RP4O0w$Jw%Q@-HgWbjnA|deFIt^|h z_ZzM2*_mgie?{8LM9f=!g0?sg#Fg4Wl#Lk&ovOa%GU<+_=)Vfb(k;2}dk_bd9;T`J zVxRE)E5&=S6cr~kcU&Uv2|5DT8pS0R9n*7IaX1CY}XmT}$`|{}&*~bhoQh(9T16n+`B^s~2 zkJFFof(>5vo8(^@@lkjIPZG{MP6?* zhTfWP?#tD=xJdCDVTU<2$D1CghNcwoU8x2wz%oh*aa<<(kyrl=W zRU>~&LAr5Lz_2XtEgQ=Ux;y#gmmW|q&;Yr9rr}FZ2`9`w4Bg{}1%CrbDK&{6W~`-q z8%2)nW)NNPltj&UVx-~`vJ`k&hS~xqv**$8IH#6^R@pSJeBj2F!BvoZCpz$2$sC`& zoAvkYL*TlZ6!NQ>lCF%DUR%mb|6LzsJF~tAoj$vVzJzr^qU|#dlP~A?f9tqg+5$LH zwKOC*)z)(GY(78SjJJt@5Ph$1#f@S6S!Q|$cS#t+-kv4&w5~~N8$Lz~$nb}C&qL^M z+=_JhWVEYh!lFkRl4QcU*yK50UtNn!vvb&Eo5cT3s)1tU5||uFMsdn9_{8nxzZ}Td zU)}zn=a20Nd%oCdAfF8IWVZ=Dd7k}06gC$k>r)2lioAYuxVV3J5}oxdkq4fZ0*>*b z0Qq5f_Fl03C%Ev)TR#!==Nj_TVrY}YWV-);E*afS!S?qXQ8A|ndL=lLa+iBl(|HEQ z2Lxlhkt4ziHev0-4SeBj6xW=d!YpP+5tC|RCicK|E*pPeN(dPHhORZ{(ThxmM!{~J zy_$gHAQ|4f`5kKdOo7U0HT-(I4Tmla6YPy9S~58uSu!yQJ<$`(za*34kvUXqTqebt zP9|mJWxTvkI#15q#kZYQXvw?{cpE#6FRs0TqvC9GZAM4b?w*W@RC^2;9jA#u12KK4 zn2TjT!K=BBSgolE-K+rk73NFMOXFy6O9x!6dJk*$XZW~VmIs8*$G8W=XS#MV?$-UL z(EHWs;-SJ8V`q`abz9tUb;9}sFDbl6232C-R3W%$fpr71c=u484oE_|e*$dF*28mT zCBBGffJ&QS7&@H9htMrU?hF(j$D!@62EOjl;hqZn*EDgK^U~BPzQY83P7# z){hH#RyT+hW$@NI?{N9T1-RcOF zPv+1&oB_xGDiArwhHv{X7Q6K;&|57Mpph?uerKuLxrRRbPQdNFyZF(j%nz=bayNTN zYF+A0<8OBnUAC8`Gpm6TgxiDR6LC^D2d@UHaQJ4C1-WL3$1~s2(t&dL85n`x z9nPbBlP>&wexv{EU7Z%V68!B5WL_S^oEck@Yj75u|E|ShvqSt)%ziS1I@qn!Y27T3#Oq=H3lD#AEnI^6LGXC z702vt_(D^Ej;_fR+0Gk{ zq^o09_-6Vpwl7hJuha?KE(8hY%zm^kdWnge(nS4#mpwhBF(>_%Tcoj^C&dZTF|561frexA((Tz8B5Id_Q+i)7yg0)i& zzu|{rKi(5heB0^qSvSrew}Dq|pTsAGdpoMnYFIq!kH*k{q?IuO4A`+7vg-MeQ0}A&2=@SIo-Sx=dV>Ec-%vLGg0Kvt1MZ&MUAhOB;n|b263l*gwS)Z zNMm@Y=YP_DBwzRw+GOe3xVsmLp~T53+IY zeU!JR1;+M~>{4FByG-2p{**`XG~R#?_cx=Tyd00~GL0RYZMl1e7RPTmg4W}QAyTTS zUUmaHH*bJK=Ca}Z-K-s$##xUdF>vTFQn}DVI)B3iicgD|G&4WlQHGq3kuWqIL{>yQv>Kaz)AHog$et7x8f!8=ag?jW& z>^4(EFZt2<5oig88y$GkiwsniTku@Z&irR;C(ie5!3n6M_q`mfo_QSWro1MnEed$` z#+Sl;&P#XVIwALE15FLyMH{oav;CzkZk#U9r^IYqW5zJv^i{|1j<&j;X>TRFrGr+% zLdyp>KkaGNA_M9!yeLzhMd`y~KL$6(v77pRzWVYZ+>9IHwyitVRBCvS$|a7J>u9I6 zItq)dJos+u5FX=RL5rJ0;Hedbze+u@W`wvezX%uS!87PCvyCgmyVwowtIn_P*5O4` z28$sFfcO6@j;?G3`GfwQqt?G1UXJ4R)$cc|k0U5Yl)7c-;|Nb(&<3aW3V!R;&9c8M!5 z)K%w$d-hUH0vCcX2c4;Lj58 zpFahL=JzBMuNvvC;Ur9CUxZ8_ftB^IGiGhM$Ew-=QMYcCug6 z356Zc(P@u3N@+=@;srA)*`*8gW^Tib$~ZJBrD91)5Io2E@~KW=p>IDN?L2L( zFcglp-G~d=$Qw3T@@zY0E*e)$4*LqpJvN3E{5QkOttU30pMa|k+3*R~hDuC-WUN%= zoG){^OiSXe7tOKY-Zlz-=}#9!dkR*wv;QPx@lR;`SIgQRHe{WxCdyYnPZZ~Y&9*Yl+>gbpoNaw{FAl=Ff zv8Qg6yqY6sypib5&8`$%phg!qAETPs3Yy){NdEPm=6U*&`}d*vVR@S7UhIr4F{A0K zYYDr4(;yYiMzdg!wN{(3%<6;CRk4Ts+v9Zoyf!lX?0{CzN-nL^Vcl5d#9w^{WQPc07m z-v?sA>E5E>)e~vFh2IYt&whKqVB+PoWHw(3$~|&0FUy~M?7WHYbEc!XzhiZhLqkhXs$*6PHNc9SM8SkoZI6*b_E`6$d-w|eWzOWe zPc1o1@GTN$%IUiM1Bz%HgG@IQE}d10$de8K^Zbcgtij_ZJ8{^aB+joqgexuQalz#( zJ?ssBetj&Tp5n=^zdumA=%6;KoT3?{`lIFd8FDe&N_FCldV01E2b)-POG-Ss-0DOz zUtH+Inr8CoWsW$df0VaU25(-e;*{iz(|K7WtFRJzx>3*z9>LY(%rZxKxui>_(9v6m zc%L@t8Q-G;#y99*pm2sAjf3|=@h-Np!r0Lc`1E@Pa>{-1EkuIczuUC@?Fkwta$$7q zHNv+2Mah(*xOH%44oNU4^<61G-HU+J&l~O6mb}HR$ zET?$#4y8; zC(qk{JIjO#qUxQpS@rE3edD143Rg1en}%PG{~~X>pYedXU9vc$3tBu z8EtvT_{#cjoYrwJ`+My|%W*XXbP!%=O+N&MY{yC6ctlxTpn31cA^GrEIy}f07EkT) z>9&}|jq8giT?sF96~!FYnma6Z;=hkabCaG4X6YS+^N$@!vNU47DRw;DX&ub!LqI`$ zFs}Du#2q<`d1tE;e)1s98cy+>9dW$WX9IhUy$U}uqk1#^GRjID(4)g0j2`tBYLU0d z6T8T6q6Ll}Z-V2ia_EJq@G;T5+GMXLxP&ECv}X*)ey*htr+-jEvl<*$$Y5IMi&R*s zfupbL$aav}GrG^irzS^iF7?9uNHZAMKc+_6N90%pM;M)FVKMZgJ{0!Fki?zftCO2kpDt(K^nGmi_USfS6qw8K^&Q6g}&%T8TpWmRT zwkj={u!1fHFQ6&IzVe_w^;}k*#L>3*A!pWrZd117#K%uCI?;wxOIq=uH5IeRC!uqf zO{iaF0_WvAaEOSbpu2tfc8eCv7U~NJ-6&3zocX~BLoPK*;kVghra!h8Q#+nR=M_;{ z|7SgR%&EqU6|x+E_t*bCf65++S=_W|Sn#M0eGR&CVM-}VEvqrWUED>)WV$lCNc29( zpr@~m?1ZJmGeudvwXa(KmQi{or==?bR$9VzP@Q-;P2tkJ-UQ! z2LkP4R{35t4=sea=zx|dToJYrlrJ>>uTUW zemka3R+4t^NTAw}=V-pyRvH^G%l7&kICz&ffBPfLYi~cp6i+K&uF(SX^u@5NZss0k zcev~B6rNQ-h^Gp!;Zs#VZq%QGsp1*&M zgRtaGD^^;Zg-y(DWGvR>&pj()`Y;{ZsrKZ);1>N3I1Saun{Z^?99Y#4hoeUi=&Krt z&p3(uv=+ck^x4nNGGlC;%*%QV;C;*Mkh1I$-V~zv%BAmEC3^KcjfJyHSFqEU+=cyyp(0y2f(-m4kvYczhcBEE zUeo=U9CIAA{L?HlV)A`=0;B(xTS3~ss9E%P>sd2f-OkcDR^i3B13;9UAlRpGuPUSjPx5zy8Cto zWozH22PZVCW!?dFifKYzd?h?*SW=(SzO-V661}_9g)y zm?C~8Q3pGD9H?4^B|GlGO|b*dE)a|Zk+YrQx0Srlf8fUX(R{R}7iVj{79I9%q{oiL zk^5MZ6TN84=~a137oX@F=9~y z9p9Ho6#?sUZg3RM`xAwhYQY=nyH4chX7F)?!?-peom@5@M}CS0Pui@21#$jp7W|P# zCUtam5MY+F0S8OQ@ZKN7A5wlEDYH(Jr$=A(vW`LfKUv(Fnc6*0bMg`qzPuOr>|M+)#a%E(&XOPZpT)u7c2Myb3mhm}jSANsNpZt& zlD!uT?~9dmx*!c5?#c3zs%-o_kd5b>M-b%FpUiS2C~jsn5*{AL!71IiyOT1XHE$MO z;5K9*)#pcLKdGUMFNTRsbBHJ;=v41R$YVtw`!R~1svV|^x+9cqo<=Kcz8KjG`) zP-J@UM3$H}wAO0jMd#CW?~5|-=B427OJkM`{mu*5w)0iJC|*%?nYzjJl6KuVD0%0_ z^T5m$&Kfa5m3i_c$i~K2;$F&$vP#Xp$5#w@GqPyg`osrP9zbjbwZ3Et30`NrAJr zNp&AblfJG3ss#(rL0^>zb>B_qkJeGr>KJ&b?S@rTJ=%p^CVcP_wpnb?6S|7M`MteV z>=Q3unC+MlRLD~`;#fQ7KE{hVceZr`?w{+-?n4r3RGBp zW+Ng+?sTQAGkqH}8Y*E*Y&=+>Un#1y)9@UqALwH@V#rh8H1rFW+v!6)ZXGv4rF}=ctb(e>zzw3$efXTQO zBmg-<}L6wu{488n?i*GWC4ZpBY&nP7AG{wQ*xI-<9{V>n7)iL@6+bm<8UZWwsN{+^c1d(N2F&=4JTD)+31uCA);haJs zEwd`46T=MlKC_$CHq|%2Vt=>!E2b{KFO%JUezjS$Y%swcfSO z`t47RTMV&fzvyWvZRA0VT6jpeTrQbAk7d8ba&0U4g~Mv;m+w_F|0`TK|1Lu2UjW-D z1#(92LKr@D=IuAe@QY5PAp0N&&I1KM%BlU0~UJmCI4OrUYILki`!+8P|D`9d`}GTXSq*qHv>IW7(YTo=n0tLT@P^u4 z-ktgn#!cQlBjr83-PNJ4F4!-YOW?e(FQ5ODjp~bP)G4-@H#D4Q?O|Ja{uw>tp*6vh zk4#Ureo>Y1DSh^sPjO+foOd#r@AmTH=%1GE?7Fnb!(?gs* z-~dg?-)+Yl+{W1$#IquY~2et24J_WKFm(r zq1MckG#@nZf~l$a+OMCj^KfOmj0YEZ_@hh?uTP^ftJ|a-`CX(Q`s!RY(t%$MIEfjK zt5`k#FphtfL$|YKi0gkqoGmT!@Wlb#zH}MVq5#34YNi&26LhNmk2LjPgiX6_ywp16 z9$h!-X=fRl%Fm-};T77^ZrsL>{2))0gO^4yl+NS&w5PVNRYxdTJqCM>r=#bhe8^5w zV#9HLSjF(Xw9g{gwrFRr?W;c({HXOD``zBid%KLHVV;q6Vc~GnS+|XR#vG(9#ShY} z;tu@4brM#1zmo2}a;Fom2~_&|Dx!Am;o7BC;T~w-crMUmuUKT zB7jtcw|it`J@%g1jRpOVAo`LCrzei(h=K_542+TDPG(a|^#eZhQE+4HPvWVrHoh%1 z!HA7M2zD8O?v}Fn{?_Mz_|Mf3jrdZ51qZwL=0w+Z2+oj2$(wG%bM=sp=WC#+#Rly2 z4?#@PF8XJF*EUlsrnfsJxOX+iJl%O*|6Q54eD=VftE#m1@g7O9V4L*y)eK3!@wqf$ z+ff=MdOteNr3fGBM84hS5$L>>N^7F&rEWC)cHhW9Bds_g(;U5LpGA+z=kSu(rfIP$ z6q2=^rX0wl!M_gBxX@kHEH@l^VaqW7MjyzQRge^2NO{(MX=$<-U#>J`zd99O^h*w7 zjb~A}`j3)T>UZhW`!!VWu|aZB?#1^iJUCKxsOx0Y(e0xlk9cmu&IR*OHGdNI@=c_w zgit;^{4w5+v4e+99cC8PV!g>~{I}*E)$Q+!hsz2m?!7hL?>>~m&9!OI>UG!^H4KfT z;%LXoYU#Vz6e_IO6Iqm>ah9XG(W$`0sguZd+dDx`*$AEZ_ zUSu)8%j`gIj{+Q&bw=+q8z`@XBK90Og)PPVsaHi3_Bj5<_YARX@>JwqSG4$Ec)Mio zxt}_{P(w(434SOV@y4-zS;gcV$-kdLDaA4vlTeGf8%35Q$ccE2q(-QNmW^wVHDJkV7DU3o99zfg>Y;qZIgb*Y{%N|dDt~N z0TcJRQmFDo8f$P#n!c$sE{<4;zZM5@Ygih+e4I*$i!RaeeiaJ-+3k|RQZPS|AiwkN*z@R4)EyR zAnnUpF|#1mmjJc{JT!z;BGA`Ot@cJSgQfn+{yTSH+yzGgOzaCcnhD z_Wf`n@v}c@p-)N0|J=X+UUugm z@(a{g$6vf(I-o{*8y@r-fRoXiq-NounfGJ_uX229lTWv$(DQrUKeC zOjeq${*a1yZbdiGT{!wQ9v?1^!dn2Ym>jW$#?Yerp99#~5N{pcOt@Mqpa# zCAu-!|t-BxM{bL|K>n`TB zJ=s2TJ|7+YPwHN_9nGVaIMOEuZ*1am;h7D-$Jn5J%w=i`VR8&R2F>H9c;xX;O0({O zNx#+E#!BM8qC>f;d?sg}N#$9QZMf#V5<9m#($7t&DS7Bk(&!aK* zPhkCe{%n%S7eYs{X8BQss?CJY^#shT-9^^7dt=BP8)T`8J;GPPBaby@-!6uT_dAQ~ zf4bbI-!o{$-onVik12H1K)l&hBRDG&h^sk;Lcd_#ZTk<6+u~uqy*pbvPUB~3YlL(3 z3^o*M^3wsz++AIlbL3j#)9xn?`9D;hXF!kd_s6w|Hl?AYA(cw{+~=H|q=d2+3WZP_ zl$AZ3+Cx&JGE$MWDN>p$qGd#hWF#|N#{c>~_&@kQ@JvW`-`91{`~7>!{hG z*X=714*ude_{?1g=ku-T^Hh(#Ba4~C=_v}av&n}BOFUT~ zj14l%^o{kMuAJ5+nTO-?wj77SKV)dBK#HOYy(wjQ3gm{wayqOZDh#dc`?CmD{rtJ= zMh6%*|1WIl7p2G_BD8}!W1sJ@2HCkPsPJeN3NMi)p=)i})L4!Ug`-(N_bZ$=UJ7}Y z=P)cmnXJbQ;+h?XxXA7ph8-8DLjkMkFx$lhNypJJGY=}ea2ju&v19uimV+Ipjtc5I zf@ITM*!=|2A$1MD>PQEBzjJuiOp`8@7NOs^YcM}on=`G;q1D?{>HYI96rYg~@4{-) zow2xS=XdNs%W{6~{@8HxCHFR{7@`;0z*wJXQVR~C7<+x{F0H_x!xxxK?J?4W4)l^c zMDk9F6c=_CpD$X0H;SIXsu{P?LDZf;ZPp=+*DUoY+=RLLlBBl#CU!bRVSWJMww^Yc zHu)^Ih-{&CUxG2KW(k^9yW&REi}-e}FRfo2_MiKg>tkz*>`)=QH6QU(iz;uFA<4fB zS0y1g9*?!lQ|7XX^v?SL8Z_O;aV@tnO2v*-F`f;FXDedU=DGM*v4y#VZFnpZhPxKq zA!Pmw2>x*jvqWxT&(}cgE)WId>#tF}K8~u>f*?PA9Cp_p#QUvV_^6#JBz7qP4vQYA zF&*nDeeG(Z2O&87R1|J>NXF(+LrxqQpx3@GwAJ~5zXlYjXo(r__%z5wuyAM}J-j-RcXt)W^1?Py z&6@<9#pEGM`yf`1&Sd%H7SOkqCZ|iIh%dWB_qY$VucMoN{<_90zQFO%W}UBSh#$pW3AKGba%L4(E+Dp}r4Z-Qpyywvlk zx7`VAj!&j>?va#`dY&eGPoc?k7SWj}LZr*3^7U&B`O(UU$l0+9V;#@w~*zGXB3w-k`KyArC!%>BzNR0rU`}P^u-%cT%2*? zUyLK)Vd3_3HTD+n^ z!Vl@dbt7KvWHXJA$c2Pl4d}KH;pELy;q3Q5*#B4(TaxF3kzFWOCP~ngY7c&!vkKpM zIESv-R`A~AP5Dda7nx7=9_k+$z=cmGD8qdgJ=u80-EVihml5 zCC{V5eWnUT+2(S$wOK#+Ts61br93Rsirp>~| z6_pUI_7wJ}>w(RuX2up~Tv|Onyc-e?yQ<&AIkg8Md!P!gj7@>jE+cry!b4P0GJ=G6 zYJ%#`LvUoL3#?sVioE1aXdje^zT2J9JLU$oMXJJ_S!d|_Y8Bd$5{$X(=b-vRDx{dy z!0hAF?CkgzEW^WK&QBja6|oU(3t};{+7?f4+Q@vITRET6V@Sxm0C(q2#6|1$Ky^Or z95ihOqls< zFg|L(i*4PB7|ikpUe8TI?%y^jpV$n}5q5mtyPK3PZAB74TXBn%6$Zw>1nGhx@EP90 zb(qbC1}S;S?$ib+l~eGoGl9;iZl($^Hg}o+6HXm-gFQyUT(N;0E^YbEnI{OKVJ}B9 zYvNJ<+a=I>YXI9;4Z#$hTAXpT7aqkL<0_{f>|w01$}|%=cGCg(bK~i_nJ;C3m7$;? zOOP-62d^0qO3!yV{$Y&h+vQ@xLb> z@?UOU;cKf}xUe!=czNUvn1o5vVB}S_O#g_JU2b8CH)HsD*yH z{|z`y+pmlxook?@L%;n$JR)yek-%mWs}Wqd{!(Gtj%k!4JFBAoAiN{3~t& z-$gMvGkpxbGF?l~J1n?M-O7+{VF{HDt6_@F6xeg+E?4!MIbU|Yh7DC&AZ`2uQ`Q!u z^_@~I>^#eStJlf0_Xq71e+gnK8mv3W;goZJT#>ybSJv}T5Wql`+d2Rxs}j*h@(*6O zWSIxXdJG*M3nm|WVcDjBm=XV!o9||eOE2^EZ-_C89UJgL;5*c_|0xJq5XWhJ>*Q42 z+qe-f=b&x68P+>QV(-vn485|DBt#D2nw7WlO=2eASv?X1rpl7m&H(D~{Rp3AHLxIh z0)At@q{)dASf?e5m8zd%`QRWlGyaC{dT%@}oQV1h4r6^s7u<}P&AdFjQO7Qy`*~Rt zCmkrnl3wPKv=pYR-*joxA93Upm*Cy}Y)lok$G>}H81J2RJg?Scm&AL#`bUhW3U9*R zo}Xw?egzd*ccRI;9K3tHfw?Sa;>6rqoLJk)y152$(zzadRC3|IvH+8#LNJ+iQY1oV zpwu#d+_S$4Ee&R3tMyH8sPHIE+x~)k<|uf*^Q{GUMwc;N2;I(RHLJ z4Ua9M&8cbJk&u1hti+u6F$OespCK6~-@^~5nL8qisX-4G(Z#$+7&EmWZ|#V}jPld4 zB7Y73&AHBaP{%pv(I;WLK{T#0QKB8ONPpNQBwweEaiL=%azrfL8a*77gn2xmT!M2n z4M1Rh1{O`y!)aGVxY1ZwFUy-G&O0hk|!Lf4C1l za+o;H32A`_Rvq=nopc-@`j+5~BcIS|2I~sFcf_a4LCD{pjB16YSS`4R8*jeCNf%?7 zdqW!iE)eFgF=6+dAvh|t5LQhnfybFwpu#2@^+*1|vmZNfSjr}-Unap=0}ny>o+MU% zw#N^S3HbGwFsz$Ouy*=RZp6fE_&seTHLn~=hXWYzU^h_l{S%a&Z~#C2lA=%9$~4pG z4^1ljM}BI0d|`q=F8MwI?W|{E>WSY_t8@)!-y4SRohIG)b~{lmDOMjR=XT%@bX59!)hWxiqO7%a%m!yRi4NW;7uud{iClIMv3 zoPXA8%krbzYCP;d9o}wPM`o#VGL0{x^?pDbbap&(^4A##;kDPg^n7IyW z*}Yk=^A7|}ju3p$%0->6q3qvc4i~~7!W5yG7&Xt09wcm}hn~6g>(^De!Mq+lC19Ce{()(^-|(~`a(R3tfejy@gPM;nVx_>El>yk?39g+w(H zQdd3(!QLjpMz=MfI&mo}4e6lEyuE_n);{hvO@(bCuR$<1AHz@Zc;>ta>&18z+%YBH zsB&nk;PCJjQ3~1YL3*{RoUTz7_i=uqAmWcEXeP=)|6MDt+inbrY5b!7pCjZ^!T%uUoaqlfXMY2p+MG~GW6qU^1}+xP)z)$$2Q z;u`9NGsfIJcgh@OJ4A`CcxIg>`Ly;uO;CzJo+fq?}(RsXDnT0$4%0q3@GeOGlVKDT@4VCwgA|oAhoEl$^2`A>!;O`2& zbUdH?H&&iYnzNsJRZM8V;z~SuA)CBH_tC+Xv*@`H@>jF<`K^@$l$a>Z!ILd8LFpZY zFa}81)MT_D62_Bp8F;E`0)4T!BcsHF^v3lS9ksvB-UTK2#Z5#bk22zmWpG>nVJ3 zF_rw;{-5X1u&0q&BtM;OW*wsa7iZ%X^*#7TzaP)-?Z)==Us&JeH?FyBfa`EN*0d$! z4_9w=oH-n~NG(RmT1Bi*G9>%f4Yd1#8$8Ke3afkrVRd99xXpS8b!Sy^xb+--t#N|Q z9;TqUNi5cWJ&As2i_l-+kTO-1XuRKDTBPz5pAO4KyU!n4PEr&{)n$TPn=W1)|SI2IOcLhZXix)SY5a^vDpp zzQ@Cu-vA<+Td94!8hH-4r`fT#WHm5?ytC#}Re&yrO!%LF=L2pV<&U1dKQXb+o}}7& zOn#e)u_MGOt8X^GNDjcGw|x*Jo`8h&Pbk>`9dve%pxr1z-Ac6>^*RiLy)#g7WdvF8 zA44BoUSa8VZ3>9HhW_@#)HF?q3Rj7tR+184JGT{`oBu#D+h?D}tvJc-D0m(@%KYzN z@Y=ULmbbo)-?RRqZ1xb&+xHJ8-TN{1x*Q$TlOnB7N&2`a72Yq9K~o53A0JAq(san6 z)tF{E=ir+O(v(f+q_@-;%*KyR3J(Q(uV9piVtOZ2=_~{e19O`{&WXL)RHmYPMtQ#giu*U z47E)>Vwc>&3YWWytG}(MRfo0@d$KkrCgfHzB=ugf&)Oq7T z$Fs+w)}C)5IP?uHK7E1$7cYEeH5#L2S?_BX;bg}B$}G+0ro|qF#W{U&_MbY<%xOS{ z4L>m9(E%KWlDKid03)^u)9$a@R2|APhTik&s`)~?cs`okwCr$~?E@ToYCzLpfQf`IbwN!m1omT(@o# z^$D{aN8=j`SSMPIGD-?CdL^4%uQU74 z^C!t<6j`{+Q+miF;_@`;xsM`Mo6n@zJ(1KtHkyXLOQf=^RSNi z`=Sy^a`U5b!5pgFph-vfTB4c#B+Rd9LHB)tTgEHnn-V?d42s8ZzVa;FyB=mu<&Y&*722<+>xOz_< z??=6ban5tFb($8Wm9@j3R3mu4P#3HezMiDtWL)RfNppGlMtb^T|idZsJ5o=f9fhWsdtED?&sq#zY~n)l4&U*t3#_vbi_d&}XeEJyaf7oqV3<#gj@8vW7hqNS1je9h@j z{+E9%KQdztt@U-Kxw;={Qo%^>Ji9Z_X**1dcmIZ~f<~M^b{h3mwBkGMYcPu4?N+-8 z;oHRvXs_uI8BTCyT&@a8h?#`t*F7-HQ3^%we8sol;$ZahQMgHoWe(lOa-W~R z!OxwnAlKr|uU{3xhtH8hPnLCV-F%TI5rZ3Xz@Z8npqo0YzUdqhjRX`VlTkuVpHkv3;DHta9+G4X-|;ah zx5O7$FZ=nQ{rj{@S8#)rGIo}yz*(!ARQh%bMJE(6XU{=YWSI;}pJ%vCR2@6BIVk(9 z&%LpJ2!#$_P!rh7wLKrhSM~~%@uH3RMmiq&8POK6BqJ;q`o(a*m&1B{ zw(ncK9ex^`2!y=MaFdxZW)|y$dD;ea?QG-RqxN(0GSze^pZRc_5nBVlL3v0l%bOdL zq=Xjk888RWIcvFBKTcq0<8rv=r4F89YWUz^0ER1f!=gae8A?h4C3EJ7I>mVYLk@Uz zUma%O`-ywB7GT5d)8KvH49+yia|$L$xUt)ZW1E8&6d729pg9RVM+ZU7#$2v_$xTkM zz7)n5hhX>rvH+WxfUH~>x1%@;W79;ac7rtOSa)&B^OkU5=1K{6s{gXkabguij78r>c5eE#>02)P++EL= za?d!g3VoEFa20L7?ZKp1H?Vx!FRw2 zb8ad(uxATrY%au}50TWOzLVCd&8GIUgsf|;)PwA>KEwcn-R?OtKx7&jC@hQ+eOBK6e2@IQ3%_&+;0rdl?Ij@yE z+^kG9&Mtm9UfBKxM``83rSEKZX~S{rkG-awd-5r0AA$UXQK<20 zJT^X!1LuY8-Dvj=wy!+~I`0BGN11X?d7L`%kUuv#9_zbYEb^tg?wc=Y?E3JFS1z=sCYU$0U5a%B^&fxnesqr75&47I;R6&M$A$&!3 z@XpH5xY~Fc^Z7(#Im?*4#tG9l=9`LV*>1<4XqenJg6|1ENV4qNS}3!ec4vvw?Mf{y za?^*5VO4PP&I0T>xCDnQCEy;`?f&}G9lvHxWZlPJs9JOwOM;T|2-~gBS{8^>>Iqoe zm4}bKzk{OiG2CkP2!~d_;Z#>0;Rdo?;oqC%(7s6#P0R}Lz(pTOvKa&Q=VRc_uShH& zCq}!z)Z?IY2bZHPhexCxaUAytBhpOCd}TTcYQ^zf&T722{U}b{m&|rCi{PG*C7iNm zfA96Z=+G)hU&kKA_N&GCo^=Y&ta%HUCo+c45j9HqAcoCyMi^5!9CM^b;&k&;c(Ps{ zt5_x^Wkwk4EnS78!+xWsT_bj#e}orVo^xr1HC>EPp~58Q+O-(~^^XB~=#mg0yHkW; zSZ>NMuu>%dZH*{?W&oub2k5@f7<@G^2mODK!5-hc=)rm_*=wc>ULA@Pm|NY18^TgJ zw&f|Xg$CqRh|%fFmngj?8gClcb46+vkkP#lv!#LC+sl#28{n^gh6nzBIHRS%RHU)!3Y|kOm*B z^0_3*OE&aSh6bmE*{fk09iq9OnHBL04-H{ z3iq_8Gs9b<#%hRtQ{ zsY|A4GA$P6;>{svRVIwj9|Z#|Z^Nt;KOt*E4h&gd0<8*BEbl3X3dL<$^o4Pg_iu)D z#`K+63RLS;>ECJe`u+_Xw@sdmkH*kjk9#D4QIq8rq8Za^ z5ms($0o`TZ_#Ee>=4gA?Z~li{H_f7v`T=aF=82AY9FgVF>K?1WjT4{Y{F`+8Q)^8% zk}{-cZAz6Z#Yv`m5~hnZvCftp|M$pVK7CpZUEAM6PAN)!(?BL@yO^O^=OBvA4?zj{ zAWVL#MAC=V`1!^Q__2?*`N1*fl%N}crQ-ScB(9m#+Z6b@*G2h46YC4qAq403Bx4!eA^Uvo%mW@Akf)ZrrQH69E{aT_xBAgw5 zvC%|Xm130779kD2gD9BFcBOG4xL4g0ZO3gx^Zmsj>zIp6%YA9>LN^+5-50z&f+0Pv z9+qmq!*Iv@xV`%%_C09E@L7!ivPp)z7dK*xNHpqu$KcYnYpH13aI$z3jyponV3Nu& ztP}L3jHn-qy*PpVcUv@O{_BN?6)5yFliiIBNu#h1AMFjtoUmdPo9u= zBz|GDna_?ww97vmMKX-((Jn<&mS2l`Gggs|cR0;f{?2ykaoE?yI`@y%(MD$ie*7>B zZ|01luM>08;mS754MC8;{nZlfHM}hp|-9jI!HBe=RCjQoHds)X6^vErcNfK zqHFl4Q~}c_m|`@`P#Jgs292MQlySnD0;bF)=^KH_pF4_8zkE^k>?3S%A4M(~OYo;I z8-;xrCZXTu7#g;l_CKqofLF!zAXkXieQU$JS}u6^dk)-4o`l>-XPj|<7kZ|%`^la0 z*mdX)-t93b^&}mVayyLTy6w1XeHr7MIl;NocvOxEMv*uJT9~qxHpvB0Rp@0Z>s>>i zFZ5vw>(ak(F~_oF4`B9|OnM7*sas!xRy-Spv)KD`$;%IT(=L;Ei^b&LwuCxXD$;kQ zK5RXC4y*SC;r4`W_-MKh%GBA>;4K|8_I`si2Xo=G7xS9*bwN%b>p!jv#g`Z3@L+=w zC9|HBOvM^hPkRf-VU1Aev=WTnS#R2p&3I-B(dBRVuy^Ttl)vtYJNwjW?wzsp<-$w+ zm*s`Y5qt6Qr(o!^R;I9k%<-mwpUObR|%2=XqFgN`~D2 zmoHtaPjaz&xWbh^`{EVp;Ov=vgzYF^mlXLI&!*9?zf*9>Q%iKv3&i*D6Ci)nFV6P- z73egG#f8EbQJ49{HXUC}vomIpi%mLqOm{yv^*i&abb!qYl2z&6qG$L>avkn|ok$82 ze=#6#C(8!ih0CAEVws%>ewlZa7QIcOKAG76%zr#T31R5uizxiZ1BETasW~E?v{y#Z z9&HcQ*jEYeizDFZ8AUwQP=STv;-uS9Ay{4<%z2OFP&X$Uj(>d#(a)~K5$mPgSEcEk z%AQdS191^;n&y&-^i*2ZRwocT7|)HJvlODf9f9@}0XXhV1m+9uFmOG)=XvR2)QPol zuz3Ma8qfCv3DD0x?hszu!euIeS6 zPx(!Bj*aDpT20}e+IMd3sbi3-JQpe++=slX8z7~74xaQ+fV%CnXjt!rQf#(aC^w9Y zwn~8=|6W1RqYga1lw~(d(NeO{3f6;YPt#3|Q(< zcCj^NIx~!n>uRW@Y#B7~=mo=K_O%VH<@(da@q1w_4kSpUwzNNb&t#p{z;^<*&ODgY zc^7c_Pr;rM0Z=z02@U@>3p86_b1#c~xWyik7S(fA!Iho!zq!tX%eNxonEe6}^1sTp zs?+FcL%*zqli@K;ollCkN4bVIpm@K8vwJK=8jt*FZ{z{`9C!xDCAcwez$~1wAd)*R`2u7v zjG^U45|9}y1UYJzf)^2M1wokt2!NIN>}M<#KZu8&4UF$GhZD?Lq|f2C+nm?p$z0;B zO@cJdg@P+elfg#p61XU4U{B;&fyEvd!T8zpxX-6$xFZ9fxru>@Ctq}1_-WM%GVTk3 zhBN#9To5D|vOn8;34NazLoS{2yo5y=7a4XGMoQPh`*myR&Hb~aWImd&b1kCkQAITW zZ~+C48qdvu!*I;o9k0$WpF2oInCD2nh3CJcKUl<-jsqkTV zKVKYmZB%faMI1IeBnxC_8A8>Hr5I#C9^$9Cq1Ve)JSEnSKgSfIr`iz=PEN$@Z%0zg z^>yU1&x)J(a|st5+`z@YKLDG}uE4JebI?(+i`>3>Qk4E=wr7rpB}E$0sj7q$nGSgR z$xGb&_ajF3cA=7WE*2M@hl02d0;_+uP_;1^+l=bqZBG&BuA|8E7Y=Lih0`j+@WtgaEO80LLDmiIuZtp+ z@&nYs`WH+4M$w$Yb6C1>IQcqGq%?~uq`>;l@6xYQp*QOjew#}9`A6v7t4fqGKZEC% z4x-^>=2{vniKmvD|953`L~rWFJDAIJI;~Zqf8Pi9>Q5M zU0AiIAK$K0z$ljU9J%ov=w1rMPce&dyYPEhQazGtrWw;VG3I2pdBM6XC-CzG1sFMv z=#1-kESDOGNn-`H#7mSk-PI_5Fc}m4`%po`8Bh9tL-~on(Ml^DU%dN=4pot;Yq|{? zMAu?^3M{u4fg1mPaCekEee+w7&+b1&iB}Cc=UFm({H?@| zd*sMn7*JR#9OW55ZV?n>ik1`|UhhSZYmed43(BWBJx~=Sj)*0v$Xwf;RU*VeVSi@ltf4(aJH* z*`3Q+7)Gcn7l9dppRthd!T0eWpytdP)LNT{57#wfExyND!7CUSDFV|{UZTOwoA@VO zgj3sbmfJB$53UTH#7G?#Isk7mxoSC__%&V7t-1wfq?zHszjl1q6@pqfwXyx{c9aV7 z$D13asQt<%yd>I%>D3|h;MG5zbkz-_$4QW+>QuUJVMOxjx>R;>E3TV*i3{HCiKAv& zp=w(cntrjuCEB%s0TCAr+r%u*W;p(TQYjxSaWW!;>5bOUz`~keLI|1L#io|iwf$-+- z7;foM5qMvzfRfj9V4K$ssw`Va*Tm1`_2mJO=sW@9Mbm&U4Fo}HHD~lFQ6Q++qIsi6 zkw3h z8;iBzSN;o5_E$dWr1ya3-?gwgVIo8)M{o{KPXwnFQ_;8G39q004KDJcWNbMO_3qr{ zR(NDUzH$xx5(;6uZcoT*t>>0H1#%Ag1#tN^xe@vHUlxnO#lGP93A!jPvvIpdkjmdI#g` zzM^#Q9o(dI98a8UgZ%9>AQ0Zixls@fP528^2VZkOE!)x6Y#hG*lm&9fh$0=H;V->i zsO#8_nOpL(Zp$Q`vf~zr7IX@7MuZ6(EW7Z;eNR;VIUK!4N#NS4odVB4*C*Y1Ih6v0 zbV$sl6Lr`+N$=BNrr+Qn@47L6@~kvKyz0<@(iVPc&v7pp3Pt|E;4yipA7&JSWZW3FBl7~?H| zJbLceV`sgy5Rl!0SvIw}t=1MRA6P+I{{ybpH5>}3`r^d`6%=ZG3EFYLxsLIPg79C9 zA+mlKRSObm_lrOZ4~T)M2aiLlzXF7?zHFLg7~O1Wx9*UxFunX@J&-M4}c z3@fDm7smWIh3TB>9%BZ_+RPnkZv>&WpBIX3N_yl?$3My8*6W#E%hg?i z5u%ZFb5|E$bX|+iE2HRV1knAK_qef1fm|#H@ts9B`V`+1L?+f)SQj}6JZi$(443us zA0$DBG4q5~ZNN3cV^Gg7mtHpj?LLr&Yjt#S#KK&-5_SlJ*2cixLN7S|d@r1sEKl)^ zv#_ba03)Z`k<5=$JW0%OKD2z-`kQT4DIe3#^aFQ4xKi7XpVp>3U@CUgq*E%< z7$}&{r-+T<@5Vi#_6j%L?BWGQwL*fg=2HCP_Dt$;V7rtZv-u}a`)K&jFf!9x#6Mmr z&u{8$ph;WL-~j!Fv!4$@(AgR^budGfS?Q1z(hD1OqPRmjYq%+2;%I{pW9>Ys#DWjZ z488Cu4ONb1KfOQjz35(C@i!Uzwo2fIHaD0k=;rii=;6|cQm8HyhYMpZ=%o(hWsh2g zSA(Ajq?d?sYsBgVxdX9~@~MSeA3RN9@@x%`@n`PB@n2xhngyt;aS`P2YY3W^9B7${ zFXkqxL+cu63T)`bF6K!R&Ill-Q>zf_yU^Q@Wm6}O#~HhgxS=cVywK7ZKE_}Nzl`Nt z@?Hp$(96$Qu}YFRV_wF{SS?z1>? zZxN_+tES|g%SqAs5BgdB!%goC@O^hWtekS3{-jmokynwh>{1^NKQjZB-uA(UN?mGR z5RBPFjZk^#Ae&>W;vHkx?SIuXCD(!r_Tz@<5oR=d@$b{2ly=lKH%zrZjIl zun8(-ngr&v9W3rw$I`$Hpwkkkv1iKl3_zI0r$Kj3=L(EH2p}$YW=wgjKNm#8!sgs(xMpFq}4tq0Svj_UIbIa9K3-VrU zOBPnX^hHt;zT{Rz^L|yBG876Om;Pbx{U$o}@Bvhg^TV{{$54E07bbqs#!vQzSeO-s zo-SIHc+in1?%4oZrqA(~U=@YivG@2(4Kk_`B}vgh`01F%_IqO}(Ot-(x!q(9|-eeth@uYV!zx^O?En&?1jzYmc zzL*o9T8F>Gb`=aOfh?7otQV@CqpJv(v~odFA; z+oHK0>s!o?MWF`9gE+Df`2FkPOg>|bC(Xi&0)mN$FL8kxbyz?8Jnq<7g>DmixF)v4 zPx-11M?00^#^^v?DKdo41va3omk#0AFGH-=DI8<79>41y!rKr;nok;NSDP5`xL%jm zZnLMqPs8Yqj3K3khtT%Ft4U#!1qG#!qFrfHjM*{`oTpcT?5E3M7|R?D_b*_X4r4%E zszI^n44mQ9fa(WMG0#yW>*gDio+QiVI>tc7GaVS(CPkWGr0JXL7Mfjo0_*ZWGQYeg zP0fqTx4o6CYQ=&cMnj zoz3P!^V$1x%UArAa|t(1V;)zup@z^PoZ~YN#@_tF<#z{gr8jQlTU8l4@@*{HEU?3l zBIey)>WZyXa_L`AFWG9#@UNYMQMyK!=8O8%ZsliqvxDtJHXUVd?vwbYxDDq#?t+g6 zmzeK4lthMD7k0h?A2C;#tm#C`JQ9szW0a`OXcdVl8S_V4m-8PVd+~3P<&Or0P;f^D zYetO4FLM@S!ax82+`sCI^2qw10`ZmhyAyik3 z(>?PXme-uedL&9@-Q13IcBSK3kDus2ssb-N-N0L)f8p1KKe3+efTqmb4+9I4&3m_l^LKou+7Avl+ttCUJ>A75H(2INmm~g3NnO)IU;>_SN;{%k_t`ORoik zPmZ9PEgWnd+66^VGX)*yypcHte?|O4+lp+QQ@#mjl?-Q2heVKR zCG>x8N+0F!V!1~hR(VA+H{CIOz`jeXpD`A!(Qk}+c^1X=CSOLUJb6mV5~233LZoSW0Hsx)<2=P|mQ8fQ?zA|( zu2zZ4?$0n)C5Sc+TSKDb+{kfL%-*}PLFAWS+k8< zZE-fT9~)p>n;bS=qsYoHykPWmFn6tHZrxXzbzhs3k35~OUms7-TV42>eK@PtX>qA$ zHyG@A4?XR+95>=3*Q=TFd53IhWQHO_uy6-V@Syd!Q53SUJ^ea3RXVIVi+i+b&x8C# zHbP?yUp>$aCtDQ^&=h`|PW~J`D}o1$+M?;RBb)hl=7;^$`O3zjoW1Gmf1Q6eM&`j{ zQ31rA7A&oFTZ%;&!4h=(cD2e=|w(qS&b$pRdx|v@M9#`=!j$QHqzz^{ps7# z{`^p1itC@~3=}!@E9qIabBZQRs(>GOf+WIO=dY)M( zGcRr<3oH9TK?M(KT(2(J3DdEw#|ak`vnHB(~$ftM2d{fpsyiT^u=ik{nMyq z{S^t^Ywuj1Fe+NO%1%mwPEj)FZSioqAB=N9`;Cdh zW}SjZ3(M)d`&FqlKc6nesgr{35E^)M7#;s*LFM|rXvE*g=;b>Wc`se19|0b`tL8a6 zdF_W^uhIB7bf9!{>=aqgr}L@BN=Y~!lBImv6dWwfA`{08)X&%w4(A8b$%7%X8ofo> zo0A4_XEn}0zMcGMZIS)aTracen#I$cH*nU2{rp~VSwoK2(ai4)sVLhJIhp}9caJ75 zsox1&GgtUsODQ`kguB200oUIe+_m)^2@%;+SdL(_vbH=ldOsgtHkC`Q4OzRt*xTJu zOA}Tth3t$5Jn!6)T8&nRahmXdwO&dwp9Nb;>_+VK?7=-V`k-C87Uop_kkpI2!*Eb7 zwYF8EUhTW`!GwP((ay%+p%uvQRY*RaMoWE8RY+S_G)uDWrPSnm1+NTLkQ?w=N=|u; z`Gmln%GJ=r5>E~4Xlp8U zb9_x1lk24WpRFY~!NW2XZ19PyW|ZZVh!xf9Xq=Z$mnN#CgGnR3yX;QxV?*&PO%2{w zUgVZS$UGcOF4Lz-OWL=F(fNIp8s{k~+7H6{zw@Z;)>Y(up%u@3ei-v2mlI8|qZG5t z((|>|l7{vud6`RF`Ss*>^3qIgiZs?Ir9P?B9JP*EeLRFd_~goN)=$NV2!9%rbXfXX zCoL(wzKUZ#jNs%J!`b#N`UdW!U%EEW&TwJS4#>c~p*y8mOWZGC^WJ^X&A^&JD z8CYNZul@G{KYl|^E1d%qfAV0R9h@Jyonv##`Sl>-MAucsZV+bP8i>dd(I_(!A>-+V zlF23|N+?%{7PqHa%I5$1f3ZDYl9D!zCmo&rR9(3ocF%v~)zlt5&h(@#qKaw%op!L9 zH=L%P=!edJDm-sqJlBo-!4WrYP0f>S?IgWeA@Xj;4ed^ipp|;3Bb) z)+&_kF2?ib&^4ki?us)BM-Uu8kaaizVy_R?{Ql1c{w?NjXVg8>^~Gs83Ra(n(N;w3 z#qzUu`G{f#KHAlbBZC!rn$jj(y50oo?Tq18_W{|uJ(-V0f%|_V<2~8iaih@i$*KdUAl@d}}4& zKeru6ZR)@kQ(BO#q=K?hM{sSE2ltp=N8Ntbp^v^EZ|q<$Jv1Lj?H?q-d5v&;<+b9S zhrM`KNvv%B*xQo+if#y9E2%5b zSKq)fdNthbwxir{Oa&)gzQL1^B=Tp4S>)HT6?)e7ftR=^IsN`2GLQa+UaA`+j{HEt zk4{`Zz!(24wb*U{5T3VRnf8?GVpseNs`E8P%5x1K)p0rhj4;8T1A+K*X$qzdog+W6 zJw|?U#Wi_b-R|-p9|y_bdRWVk37*+Am23E}@6Jc|R|xjccEL7xMd0oG(n|l4H0J6o z+Md>iD*|TlpbNR&TUn8(hOFn)=9{?TweTDt3`c_`*eC~+rK6&LxZI@A-KV~m&IvBf zhlXKLHx$0Zwaxr=LnIGOJ|>R+Ej;y7;O|HEDEz`@vRbFXjlFC*uA@7T>7B@CpZoF? zEoXj{+mY$VU|4;ejFG#x;_wYku9|PlHx6&+2}P>%ADcVN^CKMOMrnZ*rhJ6Hj5H^y zBn4jAdT@jIj{N_HPccP*f*$=WBwD5P9f6b}5Uj z+L3VfOIXFxiGlp?^)!}S9A>FwGPn1L=`PaNpXFHM5sj`WcXY+=S98K{g5A*M>X!t9Kyxc=(zdmMhAiJ$FI#fxlud-I~j6EhOVP1EukgV(FxN7yOtxhIh%Y z@MPakJS@8hs&bw~9;nQPT`tjN!4jxg63xz&ejt2vPqe)njFUdM(Y3!4KhKyX_x>mH zFOvnq=XyBLsNBLX>DheZ$^*`fRg&vmOy%dtgbVNMaa`K_ls&7S@TgYta+3#M@^klB zbDFCwcmCne-yZw2W%)<$TGd-#*{vh*nU}=XA@MwaIiXSP&(6y}$1hJba_^qESv7e- z+uHp=#>?)!zDVqqH@I+{E|0mBY8iidu>nSP?NFc@oyUP zXV2xR&Bsvs?J8XlNtd$78-MabIcH3x=79NO2Ebw1aK-BT6mhoL&J`JKq~vUZ_N@RDB7*5wr`yZB)0Ts9EAf=~M5 zciz0gx&1Au$f)Bx>s93`k>%W}&3iZrzfQ{Cy=+oZixbvU(a?LM;Ak%5yoqOdSD~?d zV`DMb+kR!+FBmN+EaZigez4Ni>6~pmgjeQ7bI0S^+_iqHys5`7uH5*L z`J3qPXUI9JX)0$uRgl|1R+0OSxXRnP3!7wn@N*?sHcAxSiL5l1w9hb?2FPb^nJ72a z?;+29Ie}M>6)qm%c6?}lKHIg|vE7xa7OZL8~>* z8s&)FZGX@obfjOPVnc0Ytkx6Yju48`{(YA{|Hj-p@VAETRF5-it= zS^uiHR5;a{I*d(~^}n|T@lXBm^Y0+cE{R3&0mG4+;7)-qDwsCAiRPc_N!lJqIAvA< ze|`KB*tLq3az;|^=5rEF+)w|~(x{E>ht$bmO&;ODkI$Tn;@_=I*mlu;WNbCVs;KQ$ zk|5Z&?L%Q)8_UjdJ@}SlDb^cS@%E7+92zo-v&%y z2LX2$;6`>lEQE*l)u}kiCBTL~vu@+MRVI2%!=M>gL>Hg#CSTt`c5aM>>6N~iGfHH6 zv+AMhyhbqhRONL6O7fmN9~lv-9_~A-BsG&Nq9SLD-hPI0o?-? z`Te>VR9~k>0W*5zyS*H~qqI?~b_uA=SNwDj})aGpMLHk(YDz$ERWb{kHT*~U=X zp>Ie}E~#PctroJJtckx1ACY~H4eg3m;$wSvVCIJz@GHGYKV_$BKTwUglunyT+<+&5>F1Cj599s(hqkegIxa6@iS4#iNiT-n6;79N44Pthd1Ko$0Ux{XEr@hi9w3q zh`jEA=G?o~ev#-qnq|edja2I~;LH{79kPh$v=KFa zUp;(1m`5K*N0Rp~(1!Cv=csH z>6l?L7|z?0ko9;A&b{f*Z*z$!%@t1i|84%&yW%=d4oks^n`;nn8_S8Emvh5!Wj<~V zRISZLg6$@Z6f^$1xAAzEu?@4sw%~261ESY7L$E}>X=I%s|C8!1G%?&y1^>RQq4myQWZYOGn=0$cOXe!TCuF~LP-Mcblmf9m zssUAp6nO14b+&!^5YLwCBEq~o{9moan!)Txyr+D$hf11 z8tr|MOc~>rP|yGcZkRWo#~uskLRSUse-wt!E~7AZbRc@24#SvRYTO(;4CyB2)N;mv zwm5a?+tKCdx4@o{-BCl_v)*VsB?T?1op?l65bIqV!(JWxb3ZS4d{Z1weoFwDFy8O#ji5<^ru_8ON#|L9xaeEPd6i?#)mfMgTK7yRy&E~Y` z#hlrznxEFJf=68j=0rXsmpQ6@MC8mJy%V|rvUDWew8NM|uj$Liy?9=qB-ySoqw>L* zaWPMiTk8(t2QR-!+t(^1IBp_*SNgD#>jQjsHK7B3T4G03aDNZ|#whPo=oWdLVppe2 zo2=7u`?v|q8^d^ZLk#VUE2H_r%c<>vGRSua;l*~rpiA9NNyS+RFLcJF-D>g>-6$Tu zB#v*J)?mOLMnn4v)a_(6XPleDh9d@WTKg>OgH()KXU)SLoAEu_k0-6pW9^mWa08(t z2YHF&B8@Os^sYw_`zCll6R^wl4GOe{<3n>Kl7F`4`&YzFBP-a^<){;N(6fha8+$&z z+=`#fa>T_csnYMWBPfG4Oo)%u%3PW+<(Ref-eg_Tt%de>-4{DBt(OYP~UIDIT zEnwX-iu@q{797z7>d$+?XL7w{(Y6C0uQrAIR2gmxerSx60f)XF&KrkpCk?+sy3}PA z&Zn93O@}9f>Y*fCdc-^)y{K(LSUBq?2(LqOWW4BVC?gz>1{e3#u za~A#l=|Fa)JCVHkA1U-x<)%>=scv7GBUkXJ6rVK)*DUbCfHG$`d|B}&TCycHAt1)c@)A;YYQI4ouQf#GkX1S zDDUlN%(YWp_;lV@@^G0?MHop3+Fzw;uW&M(7(iAp(zxtZ0;g!q=56}zX+{6;H1yYV zsqXwSy7_Pd4cL|^S6F!842>Fx zU~^6UGANcus>3PY|=aVXD>`KBOC6 z9WV&*ln3E^;bd4iY{bx3f-`A99%1Q9Jon>KF~3=l!rnD>r{xT(AC9KBw-TtJav|v+ z>qQmQ)9|uo4KC^1;!DUu@~LX)Y033h2(cIl5Pc$rl&EF zrJ-#OOC<)MB<+2>q@cu&($%iQ-8p*?{k|DUTJHnM{P-40&p?G#zv|;derKAW;vp5T z6PdEeWK6b?7FlR}ltjAG{c{&&vQI0aQ|B!h@YQrfXE&`M8$*9>1F6Bz0S-k$cn{ut+mXYuKSvvFJ73YXtBkZq0v4>Ndy+lCv^z5Y0D@H8RW zskf4I$0FPqAifjbOmOkeBhsk8KsrMOcUxg7`zx1Wrdc0s@eIO_52Nw;UOybZ6oaIk zB{ZSS4oPZ#1Sf=F_I|=6;$$D;NL)pF&nl#=Gp_NP^b7nV>kxNcl_T~?)tGwf4=pX{1P~} zdsD)yTbC)SVZYQr@ul>xU>^;;UPOO(l~MSwW}3fTg^o6NrFky5xa0Q*ep>X1Ym#?R zpNbs%{CYZR95%*+RfRNGeILEdO%wAFi5|T*qP=a$;6a37$7Mez&C`Rauhk5y`{7JB z14?lG(R`?!>kcEoDeReM!TrPRd2;1Fxz@l^c|Wa-a*x3ZIMn*PlLXXu?u% zdLvv~S(@<8O@LZVD288}ff8#CV!cplspe_fxbab3dU6ZPFBEd!gqwJG^dz+U?!c@$ zb9rw{S5}WogkHuunmRR~EV8E4jN1J)$WMp7Pj(~oJ%+qc_<~!lFy!-nJM(^n6s+|i z?CoyD2{$6JSM1;TR05`MO;mO95cztv;g4xA5ixbqf6c#IZO~+&@~P~9E}thZeuG%M zQG9xAEJq#Og4@9bh;A>kU8>PYz2}Tw1{>hK>7&TFW#C|+9IVOTN>}xbIBbL$%YP_J zAu*Ewrtpu*6!^AmK>OAE zMK*mfPdRjmuYRgyqs#{3Gnvb!wSk-z+*2BU?!Ba_JDScn8*}!eUVPL&fp6U?WxGE+ zICEAi7X}8?<-3)FWpfA9gm-yzT$ISM+~$=+dYELP%`XRA@k5SA#f^Iy7eZdo>umW#3$(x@?Be_-_APp8o3+ePUoXxTq;a{)nMe#t=y$@A~*Q% zVa?3raK6)n6R(xvf0cSpS=ie@CaHZ=GXc{ZB=gh5$ zxqlD8TaDpwZw4V+I6TK@JjE@C*2p(}jKybk`Bjn;-mVfnq%}Vf`0OsF9BISTkF4ft zFV*E9>L1uQ&`LhA;=S~%v+#4797EJ71Mb{o9N#gs=HLIe;GIfGj(*USKQ}Dnxxpd) zv}ZE!Gis*wN6RtXb1QZ{vw@EPZhUnZiPa)Y5d3~RZ!B=&*Jn!bNKyC$6%r71Y9=-L z_rdv){m|{8%=2b9;ZV^-gtYXMZj4MOhj78SELUdB;vwv6gfY&f13dECk+ zSQsMaf0@I$>*Hen{O}^jM1JJ2q5{kBvx3+DUB#m=pJGSLD}1IyKe?UYaVtLT&+C@9 z<4@f#(bwijG<9ibd|6crEB&!tHC>hC2gOlNIIwAIHC?e5e8na;HfsGBRY$*2|3l>z z@a{Jy3@U_4SvxM>){axOpCi3+9@lo*$g@W`a>*)XdDfU#^00U6@{hq^xa96<_UW$A z-m`Y&kLGx`GiytR{mxOvvrgdF!_XmCjXQqq%Wj@2@SOS+35xc->$wLX%-PN#YEroO z{$K<~X!FmdhCJNm7;>wZa*AaZfBc(ATelUE&s@<*D%RzcGkrM0{Vsos@6BDF7;wTI zH?eaiJUZq_nc|MK-^VuGQ<}}X@jJQeyf2iyUWUEDGLdmz?7|$==EIfk`GL|Pc3j$Co=rU-1_kT$9N)&sgzN?{0iB+L}-FO+xP|S605@{$KZByUO{0^XP316jE70pt{A+cAX-hs`GHP}im z029kgVYuf5=Kr37;@1JtX-oK9U5{xC-=o{KUwC0z45!CiaaPoWA9t9+t%p4hYubuk z@kUaWd(he)e)MeQR_xlDiVk_zq7L#zu7dETENTanJ*9Ymx)w#RK4V~x91A-sVMR<= z96OgyVUrfps8f05(k_?gtvx|EeeP1iYr#SnJ%klShpA7vE>Rc|8zMZl ztvFNFmRDCEqr43oczRSHI}6J&zONG(b~ECbbp}WsT8KmE_d(mw18QD@==XOd{@9q| z$shoE=`ZLecsvdSX&f_2-vimDqU+F4_e35!;~Hr3SNI>tNS>l3x6E#Z3PV z*dQAK&B!8JYbAC<^Y&7->TPOV5lgmfqH)i8AWRna!HZwF$a+d4`5AOU)t)t^q4-m< z#2!++mM4_)ZvoD`BqKXu34Pd?NLTxbI>%}xU0GC4JDfD3Z*-kDU;ayHt5rlktS^pf zO~Tp&eH?AC!AFKq;*f6(S@i&L`Q8X@DK^8fd^wjt^Wv`oA-rw5$b(f^^Qopd9@Y4d z*1RgBhUbO!;g=6I#;(Jn#lxZ0U<#*bQF9$UNS}vQ(0C^$+zfJt?W?Pjx}6ey4rNd` zLv4hW{-A{0+Bk9H2DYCKLBFFL#QobZegt*7i-U&Ts+GFjDER^F@2TZo54WSPO0X1z zjw5;ENUU0_jZ;cXaB{!kBq$z&+w%$}z00KFgF{J2?D|y?(clTApW#o@S3JC$fxgrH z(e!Sp=xJrZ+(mexvMq6^F@)ZDyHd*hfwbZF1U}W!n?F`0iCxcQ*sQc4|5gi*y~u6Y z-hF`b`JZ4NsSb7Tw`6=~2K=iI;7VUL?)u7_?FCQ!i_sS3DJ{jZ+duGa-ZrewK8J-v z=5f`X&mh51{MI1|p2e}E4&RIsz2k9UUmBXUL;kD&xgK(uhQ8i|S;C=u$jOpdPHxYm ztfCNX@5uhWe~HX<8l0xtaG3T)to7-Rf<0M?y^(|pkLj>pqrgsLr_biZK0F&)hVval z>42WdG+e%n`13lv)?bk?&(oHhtQf~Vm+NtJVVJDPf?;$aYX)hqZlYfk&R}lSa6Y^` ziQ8WYc_TC7(!DFs>98N|?{9;r zvMulZ{0C``cHFMf9T#2Z(He?D;NZJ(ek)kPUH_9I+ALqbVw5~EyOPg7`ih!=zTEb{ zFL#Lw!QBUA>CecX@EU8$=}+wOwf9_vm8i*kh&kY*7yWs_1S_8G7J)Annee&2Rd{Eo zW92e?bP{Y<)xj}Pn7a=(Auc?+y(-T)*Wg>(o#m;G`?%%HAnsQ?O*n+7;z@xf&(}Ii z>LV*@;g!R5{8Ao$FZ9HR39*R&tO51(l^8i75|*o($1W=sY<44Ds(4CA)~v+)e8Kv< zuPz^?xsMwM_TtAqI^oTOIncd+6<@CR;Nn&K9GI!Xie5v=;!`1=ACW`f71B`}*_IbA zDn+b<@H*r*!e_`?oa)|IUcALd?w<37J(F%?P^B`57GA=gfO@IMCQ1skWE`8emHs-! z)7>xr6s+kj|GGy*9`|!HZyc$N^auwUdN`C6-mJi#`rG(dXu~mKrMyT-&LMvV!&YZG z=e!-ykH3##IpkRL`wh11o<^Bl91o5$XX_1G?7Cq){c3c=GO;Tj`YVhF7>Dwb`!e3V zJVw&G7eTI#uBbiv9$)0XSUJ-PB~SCw<4tEi8aaRqcef_vg+1^%Is^S~$S}{kn_%Si z#}k>D8SBQPef?m3Iemk|T1v>S-w$c1(={xnoq`-JcuSilS~5o)Kb2hZ{zVNG-zjqG zxq9?l^aBP>hY^{73jMFCvF)@R3L82F%@5Vddw2$Uh>U;$@N#)v7~9|I$EUr@F=cmW z{OK)rGIKVPm^9$Kiv8 zE60u#at=ak;zLY(DO^xz3(-(^3htd3p|qntr|cTaL31x5OEHFid=EmIT#xy1HTM{L znLGGzX5-UkcwSnIcPm9MHtew!dgUv{B**{P`R8twCA(<0Vk31oXyta|F`jMN_wy$_ z7!rZN=f^QF*A@@=)KJ-(+w{8L6d!xfMEJhxXfXKkA6qzw^Y)x zgQnEFIGU8Y2f_TcEiOEngZHmzAv?M^md+{1*>!_J3R~!n?GD)eFooyQVClI=0xB;$ z;G;t_tvQ<}$=7wIyU(Umd&6vUIAe+56OK@7dNS_aGZFk-7iprNxE@Y_A$+5;vkD%T52kCyECG~2YgV<={U=Tc`N(}>^F!(v@w4jM$*LbI+F1G$O!EcMD zB7s! zagKBD$ot2O9HXl?zV9?gKjZgkm~;a@)8lYrL@iW@Eys~{VMwtkMu?RqU;3g9wfuRM z;9o|m%BRWWE+a$iGNhC!vgTz2t{1(zA+v5%Q|V8rIZQwkjmFu3CsCBDf#dd;$X`1g z4L{P6vc?}e0Uo$lGYkutyP$KNI=6oC4BaLi#m9YRXgekvWAuN(-bh8p$$w>XiVPVJ_t@W8k3LorWb6DBV4MX28m zs&=zSQC<(on!DoOkIpzSI~a$Ptnu-(8N7S-!u_!pXtnOU%*Bhyer5|fH6DR&X{g{w z4uX=%-W3}+lUez3y5l?sk9TdxBT}T8(FlRZ5@B;F69ebU7OSVlJ-5@Klh?2{ zXg~Iw=i|?+>lAi>2);NslUCpYWWUZw?IA5(+4`6AoafWwy<5p^#6EbADaGWU2{@ds zzy~M4#`vp=P@A=ruh03&Dv@?_%|`8zwW;_+;idf`F&V^L$Hhj?fCNy6>h)gC0e?tVz@>= zZbzNQ2Imo2y2ux=UrmDH&tW2q-4|9%TEVx{mCpt&75AM1?B?!Hs^)VjQ+Yl;Yl)&o zFCSCj(ipMtbes^^A1*=bu_SaiET2w9dT06OEm7C>I#hxGYFIVlSu&hT)2%g1gzr9x)dNiB*aw_}KN~?9NKO zV6PW)0>4qN?HE+K-jRCgi#_g(?leO1d*%u@d~>oJS1$W4SPjq7bLkn>MqfhDuxzaM zI**=hKVYW5DG$bb6h|rY_NfM(bWa?MPi_dQu*BZ{Rq*bzmn^JO=t%otWY8xRH3@zw zvIs$%(M~LOTZMX&>vP<;2dep3aqQ_MRO@=fCe0IPhl<)TL>FH_=pepTB6XBnNT(?f zg{MZ~ks`ZmxJDb!IkXDn z722U{tQ9uj`-7^FZ*eVAnMko?wc8~=&lbUxdR;4Q9(vKKyzMF?H z7h-YiS7&s;t%ZI2r*r;xJ3iu~%_ryk9=Fkr*JnF$M_F%vs5Trq!r{B+f+k|#_T+2FTd}FwC*0(z0pGVR z)aKIz`jxNF?ysJrwnP zI$P99%iAYW&Fo^>itAj0(>Nr>H<0UKz|p;)^xxdZ;^B=b(!GXI)mpLh{Ra!qi9Fh? z5DXvsmfp{FgZ>~Temc*B?TjONfsq%74i?w@w^g{|bDb={{-ngeKZv^ggW@B>@jp=x z%^S}s$;%R*FPCBG_emIBDb8(~V#Xgliei7-(Tx)+RQz0(ml~YJ{o{JfR6-=#5JK2+?#7bpv6^i3Svp};Y> z7n4C5ll6@0)YxVKt5@pq`OTGBnlK&(9wjh{`i1nDA}=9#=aX$rSU=$^^y|g_oMjcF zjjQ04`Vg%>j$-olx#&MF1xD%paILWeUktb4KG%);M_MW$)8EW0s#n-8MVGb2Y%A=8 z0_QjDi5!?AUmX49zv`cX$H&N>lJfZEptd|F!wxFKJvt$0H?o#?l}Bf^aQ6iZ_)M)I zLIS74|H2G7?A67S#fHKQt%b{GihT<0AIJ?9i{uNh1k$njWz_o8ZrXD8G5z&fPk*vn z(}vyg9Nu9LCqL8VgBFpzE>7Z#F0b&+teyPza7BLn(uj7w-NU=L&w*dlO37s8S6Pi} zob0q+s8qG7f_wj3B6xZ)!P-%XA8?K~`u3EWOcH)beE4gE8)9m{79Cc0N~G#tKar zcx7cfj&;i6xoXucjWUzpIIxvBuXSgO^ZltT!a;mLhhl5DR8+exMugK=T4Uh?pF901 z_IH!)k8u*ZyIQeAivjoew3x1sc~0G$`(pPPeYSDa;6H(LF|(%`65IVDDbbA{IJ?8= zlO1T_N%9zG!$UfK!=9P9;IF<|Dj(}cDjOz|!7zWE-Cszr!@tYiSH*)*H=)abzFghB zk;%aV8D3YU4lla1?}1X`0u7@4ptq!R)d6GT^Kd5M551WO98)Ssm({({r&Tez^@ZfO zDx7tK&a>;xYwV>kQ`Ds4aL*Uc+)?A?FJ3K{J81QmpB}i6jth5I=Wz?g{k)ZQqq&l9 z?brp=`$k-;asieLN5cli@}N*1xy70Uo;kk)$9FEaxrY~FM(j(8$vJORA+%Z$s{G&dzhja$^%ev$2;u2}q>C5t3k^P;ew~8-a zDuQkIgHWz_MgMQL9IiSG+ZtW)JK`h(~F7}0usEa0_?*o^I+B~HDPFxOj#s2Ka82(7G0d*1(w_=}E{aHmC zW0NE^I(!iwk9TC3AtQK_Z$0U$PLr%N4$9_kQIgLHRg$;8Ya)L+$rJhw;#z+uS7ece zZ&jf++o~z?u4l8Q4KHTWhy#^0aJ)L-Nlrt@40XW)CpE+> zY4pbL8fkjGlV9|HC|{@(Di1_6(ssB?Mb&p?cin^JclUOeC*92AMW@cno4?GFKiPPY z!xi7K+~+C}@=oJ-mogmPEi)WTf9#T7%^ScI+*`2VmL)$)*P_b3Kj_`tVc4s;L4IVb zhukVjLB3+kA~a;&rwfVQXy>-O80o3ci>;k`qNx`@HXFpVN2&3;h9J3n$x>GGb>JOk zH(C2n7k)Qbi`}=}g*0E2mA{*be2%T$=PUiB-4cUBW>QJfm z#9C_V`ixplz6z(t&8Y6~NUu*6l0o4KgxQScKOIGOD_0jqvkK_C{~w&YDnpp#S!sxW z5bhq_PBl6UF)>t~Dzc`KZ?}DtZ)gZDEWSXO=_qBtPeJKX14`T4P9|SwLp=ixk?*gI zQ90Q>z0#LsUPLmB2+P!42CP8EQ19PMn)FbS)~W2FP#>mty*psV!ATsoGFZ%D_F-76 zTuQD!kDQ(w{NSlRXD!a+i~$X7UhDiL@(k}Ct zq%r3kM%+)N5l-sT^L>SaHMSdf`&qK%62Zmy*$T}H!PA-Z2--csLlm_+;9mqz&*}z? z^KB8l|1r56mf-v_;@NdIFf4qD9&LBvc9*u?;nqVjUrEQ9b;@k8B%XgAFXq)=aU3yQ zWPyhdqa}d}@N+YT-s}DJwLG59rA=%xt{!g&4&nJDhI8-Cw*1!UJp8&FqwdTSnpp82 zM$tVG5Iu^*UJRov4zFlQn|rkOva9{$| z-d3aN`Ww7ky$$LIe^Akla>>r>sALkM$4y5Har5Xlyvhhbzv^K4?L0wW8>(>W-)W?V z48+sm$LJ6mP7%+uWleMXu=$_4?BMRiTh?_!d*@?PSgpR}FVPckBla?W90TNkX=)R@K{eNjnVt4cyUoY)` zZZCVVVE{cj@|}!x)kI&i7SH?i;1WFpUcWtt668+Ox^xRigNL1XYLywQJJq1xGl>1~ zwBhI_ekfM5=LL7|**@|!u3ueA`X7hjNq<#LtQv>IInO1OsgSPPcWKt}`DijVV^0-F zKK6D$f9Sf42gKEqL)sWg|H*A>Q~7&omXF~1-4(gd=1z1B&bxOy;Ksbt`{sTtHpxnpU8Dt8Tb;AJ=S{_FgcGBQWxDV6!P@Pt(_GUiPN zMy%cMFRpw$PKox}lA&@kMbmR>K>aHzK*dOMp0-WaxWk!3* zJH*Q_P1cbnSBFTJ>F4Oqt?y(r%?|sbOCA4w@|3#n521!(Gid6RMtWTAgoMIJm_Igw zi}p71qSs1rJ`#wXbz#t7U?*KqD3@|p8<5`kSn|K|kS?hkKt*d2(k`5b$}m%&^4EZ# zo%E#a;ZbC;Q%TlKZz-8(4@HZ#NjhaIeCx6b;eyhk&I)bF4ddbp9Mun7Ns`V!`l-58`qCE^WNn78I_1*c`Po!B z;wSBhi<0K5E+Oynm6ZF;0;K_ykb2h}l`)5$V>z_8BduY)TdW&5iJC7)V#n z1YkgT7aqFc7p$&t=R zs9g3Il0q>y*!AVU8(VSB^>Cy#9inl%zo}$ZG=6?cr}otgFjn+VC+dlrVo@kM+}@3f zR|jRqT{lS^FYcvfg9~YaPBvyt)#BAly=X<(hjjImgwG={$Zn6hK>>&v7u z<8x@Y`(7zz)=5fKC44ZOFWl84W7Y78Ud>h(e$Ol#u%eNc)t)2US39Boq8GQHS}0{S zEg@$%79Q$K9G)_Tlm11rwxt`~ZYSc5s0jy|1|Ui8Grn)>&-`N|MGsJ=tg$vQFS|$z z9alkXYBts$x1!}alkj+(Eo;C2ESrv-vK8NhNjs~9$g}Lg5_bizo$2Xl{#-*iCa%$( z-hE}c6C_EkQzo5$WGZdjVol4#HIbPfA}y<#MH+t^B~Lp)X_xXBX-}8qf@3<6`aVBI zK4wGVmTn<2&q(C~$wv(6t6q1HaBDdO9DX)AAE-W0(D&0k0vPB!Uj>GYASYNuL^O*)2j)$F> zl`Q_zaLFO_uGHz|4ocB(p?${-pjA{u1};Z%EAbCL2KQoD*Q-1&3J5J8h3joAu~vt3&r!M$l1%u{d-!7x|4hVLjq6B2!mFA?gxL z0>`lZb6shh`DIdjQ3yTN1Cm9%J~&-niZK`|DT$u7?b8la<=u;Kx`gw0>&?7kh5%fA zi=w=cA(Ee$Otx}B1+BXB1SNz1pu%qmZ?AFTNiA(S<YB0t0jM#S<(1=kfn~4fwyl2FaoK(1DKslh>g5pC>ncsJchIeTDx#{$Eet=l}NP zEVCB}fc_^>E)JQ3!jAs`c?q*RZU6N@e*g0S?=$!xy?>QqPw8#j4sdQ~3Yl9VMa3kp^`9u=hr)|aa*q!JRHvsL9PeZu*IK11fOA3EaOWrOwCHqzFSw6^-?+|l%LN~NL zbB30<WtL`%<{kqO34HwOM*R#gHb}{gdu= z9fY+%hGAb^7YzGpOm8##P+Li#wtq6gg?6e~(MbnwEqBqJN-=M{lT6YwHN5xGM)}P^ z^x57Hv%8wWFj)orXX;>e$^*LIltnd{D=GQt4mx|}8X0ejf%@qr{C18+=iXkJS0cyu zZW00xc0|`O!HZvRiP=}fXnD*w`gs2cjecoP*Lwew+*&lKf3F2R<(wz?ESSU(O72Ma zJnu_!JuJvSw0Fr_jCp<;Q|vWZxn>&e71!xUeWFMv`l+s0BFKMw0kyN=BwO&JgQTstSt|Zp zFV#L!qjk?sXhq@~c0E(Ux=znIqb83Ii#oIZ;W_d!90ljO)1W$Y8aj?%LH>5>Wag1f zvHkPu)8kTF+@*nLrmNB0Q|&2#8|aJaWU4J3M~;UKsp{EYn%ujDIvVEDA=49-;dy|* zOnpjO!;7fWHIr6cili=O4)?LQQM+|I-;NFBS$?5>Dcc%v+Bl-$ojy1`dW)n} zWlI%DqQ!2q3zgjrpn?};Y1dXaBqc{6JbojBcAupA%5SOYvtWk}mf>8H2cBH^_+J#A zcU+Ed8^*OYwY3ylM5TJJm5{8`28BubG&*Pk#fH4XH{-vE`J_uZ&Z97IjX9SW znxU+X=PrVuA+kV{iYKX)Yos&{d81BwZ>w!?5-e z8Xnxm+@MGFp*oeqZiUdq-P+X482~4gJ?X>AD$&ss0&9<0%4&EAyry6+J@<;D%^oGp zI^h+2w)HQo$r7q1mssc(n}TcagHeAd5pzZIbf-g|zC0g6 z?bEp1%_tJT=5XKMszJ!^+>3{rXJC;3kJdL2=1e66;ry9jxb{JpV*0eHPP!keVt5bm z&=i>7vk_X4nhEY(w1gv#UgR(@m7*>eQrysHVQhk!?eo8qwi~XGWN)w7F@vm?Y{YD5 z+UUQE1fDV4*`3W-tt^R*1z%Oo3xD_*7JARd@<@eBggX@1E|vBBg{MdkvN~H+1yFS$;L=% zah8RTZ-Njo=9nP8O{c zB2Ri}zlF~IuEmy|cc|~LM2aaj=+aao)@Do%;_>7re}X3Lt)WoO6qJlOi9Y4Mh>GC3 z|9o`{J7-P!l%Og`ZUt1me?u%;4~k2FYYt_Poc7}IGvfnJ_n zLf1E~qwrbd$^0*)slEfKx+0N@tGs5{3vDrL#Am7v(G!I8!-dYOF~X<|&Vt9P6~drj z2T5YbOhO#GH@6?k_gYY_%Q!MSp-p#J*1(Ow<1X9I zq-EZlY0jp@B%xG5-O|y}ydDW#&Gk5Yc^&zwA0jEA>oifKo6cO26%Lvj36b*)F(!45TsTu`J=!115i=iGANFP13cllrtxm?E5a_`}tOQ?f8qgR}3&v zVJ?pQMPRnX4s2-2!=d%P*#D#v8y0KQMuUDNVcEsPpoVV-Vtf}J3K`5U>&MFK7jHXbxC)EHC2E2p%nI(C$>Ho}v`E4Sj;^iZ2i}Q=Yc<3grL9 zn#AfZVNakDiKp$Lk}Ha&#j}4i0^ehf@>C}M;4V9)`jagWpF{`$`jX7L4AOXC%69B) zXKwBNG5MsP@cEIQ@Fjn_U=kofo8}o(%eN6Uo9|h2bhCy3in4`goaerv!~|RVjYZ82 zKV+?a3h_`?lD5jA`sHuP<>(LkUSERiS)VcNYbX9I-3^(<7L1&&N|!wsP?gkX`o6%O zHYx;S`_)|-;F*N`{WGEIdJ)UVa1TxGaN*5Db763bjo_?3j4qpvB0YZp8LO*>3z7Yi z+inht6LW;sx-*1MzK71~{7o|kb&>wa|L4(%E#>&)yH?-TTlzG@kM%Mv_PWIW%zh5mfX$1)PsVh3o)oI?S^LE40Z& zF_`dk9^ZwnqN}$TQTo$*n!a!yg2P7!HUTJuilNVXtaRZ<_(`&XTE_+5V) z&t|RA)g}9zcd#j~4WIirA^c_-h8M;_GByJEf3)K?H>rd(TN;| zXDS?3XT$NQY!7bVStwke=_*`cqlG~(fdcIc60)yF3z<`% zl0)A!Dh#|un+Kd?@tO}6^P;jLEBOEW4r>3wvl(sC z^m881ZtqDajZtaT+3kWYnscGtJPSv3QfSEGTzaxOj}!-XQcWgjDLKmur?xoL?u0pH zbk~PoeLjct>rTRSeI_n_5JUFK0g(EsfOn?`Qs?VolxE^dgM%+&+V0P2yiP(-tDDA9RAfktHXBe8lD_VLhfk;nRfB4I%kZA;!rRa4@~|5_*8q%MyV z*%tQf+IC@3Lzti#xP!CBqiJhYB;9IQPEnaB1*NzmAt8)+Mn64c=^e7rNY}>c*+a-N zU=q#T;zGQNJe6frkB5-woz2ws z%9WV@b@n*;H#0i&mdT!Ig_c%3baR_A(!dmB9RI^PcPC6ZGl$=scwfa7y)cuauJ=lO4yaA7-iB0Itw+y!-btDYS| zS4q$XzdbmpT?_B4hj9MSAR)<5To}In2boy@MMRP!HGfc`YoiCj;D}f_0q*9qzA7MTzR9Dt4qY~7j;rnXH6U$^)Y+=KWc zdM23{ZH3&?Xh>X3#?^B>sIPr7Z56_3o7M>`E-Isumd9z4Oc)%vi-(T|I`(HG)m-9t z&~3UjEr&ZF>;@#6zSfINPj&V`=1)!_xX$QoGa{ev7YAM z_(Q8Fdy>IzQ!+d*OAl=unfs7XHc9;id%7o{t`8}s?R&Vp?t>z9iuCZwp&u3vHxL|# zjuOUCpD46O$k6VMYSb8{N@;tx3*)7Ig>3g3LeA0$n4*6fD|$X7(XjyL4>J&YY#)}5 z;XYUYO8TL7fR=3)$Gb54p}l0;1IOGok&W5MvGrBIr0Ah;#R&_h$sb*$J(<5enQ?(zu13$9~n^mO5@#e6}( zYr3%Vzon!X5lfn{6Y0U@VYX5MR<;UHM%&IT+sf?Z-B{YT9n3Mdh)#bvN;XG}sc*_& zRGVa>d}9g*ciWL7@7+3Qdy;X40mNTbE?{d~PSz@s{{eiGO&U`2pkPb?LN*0*ME!lKnIt;cTLspeCaz zTnhB&cZD@Hfirfq90KU3_W}yAn?k$RY$6YSmv+3({S5w6bn?|ms{1sUtQ-2-IvMl4 zz#VDZM00V(SN~#J(f`<`!P*oi0iBWb&f4Ugi1)~{B^gxtB7vnaO-$Z-5zut~&C(ooi z^B>|`UMa%YT!fo1&ya035q_&p6ucU@)6!LYN$g-4J-qx2%S;>4wMUAizb7;A4_T~n zkQ`cGgkwzmLd*-F4JUhR+g9_@wj*xLv>muCghncFC$Z~pbhAxGNah&~1ZxTRN19UG z`{7jk!hxRojwG=vdzut#LbHeWqhrro5U2MHYtI$&^Gi0h6{b_*Qy*%ZJco8Yb)unP zPt#eRr`@SqO(}1upv=q!(l=e<)M+cU{B{&1FS!db_X>p-LAiovYJu?ZyPXi8J5Kmv zF+m6@@?m~*0*miIjycZ$5A#>e;Q6*m*e9`#z7;GZvnE$kYrKd1**z%ysY5rrA5m9y z11;?3;=t@RkkXn71iPSD{~#UFKTRGfwY+n$Ei9LH5ESet2=x+Hbat~H_j-xQL0O$X zY%?K|#{>$0)yS-Fw6pDl@|diA2$}isCq3O1TBF7@a~Vr%?<^ZSnmrn;L!*#(d>>w4 z>rYcNe_@f80linB#y?y9QL8qmhF2TlGd&Ume4X)CUQ(DI{g$*!8%R1Zg#JE@q&uhA z5S#XoMeh8;-euIYr!jHZ);|&YQHj_s`5*Q3a-&>Tp8J{e7Y|PN@whA;(ySKZ^yHB`;~8t!{J?T9Sy2~rB>Tdt zlOO zm;mihwitZU9wioY(8Rk*g?si>bKeD;f4>P|N4`ezc5!OsTmsuA1*|9Z1sgZ2k<_){ zlE&LloL_E8`}yu?pt&1uJLe}fTZRhbJq`*7MLS>u?X>J>@k|Z&sLWd;}qp#?T2#AGrIt%?{%oO zlqB81Jvb9QmfQ^!(Jqq(W@%=wc@IT33LJo=`f> zLZ};ADJrVz&+1OwvWp(d@O84rN*aL$iVGm;7mI_HXJDk7hOhxA;bnXjxlU{GT@(#J z!zl1@5N=@|4*w^@an7K)^5-+`+{}@=!~p^;o==oxV;g3?#5%&$#3j@ zS_!M&`+}{solf8V7t<7;O&RX=ij5{olqjgc?}8`J-FCybhWYSoSix3U&0{R&6Z`te zh#fvKnY~GhX94@x37Olr2)bIFr&KE-@wYY7$2h~1yIc$1{m4^sE3uUf1;q^O>?h*! z@j>`eFcg;Sbf7Mpgkf=xIQGmHHpk}UcWoGy(^Fvbpb5!E{NE-oMz@>e@Vo9b-n^_v zV%B7|uad+5nFG+XWet5h;7#s6>nX5JT2NX0hepoR7JA)!v1pzoshOJ4JLx#)RB@S& zA0PqW728?omS~pMGmj;Hl%nyeqo`mi|BScqf$6g&__DPEgKQg6^7S*mT6H62^cSZ5 z{txrc)4=l4Z?OKfJe^r7NmI;LP~E&O6#X`y=SZ9I-z91KK3I(OV>_7HvHrMYZwwYX zi9I)4&$h%Svp+&2VggIJbMzxZZuhaImsVKVWP&jon^?}YYwWcBE^NzpPM8G9*2Yh<1XoX-!st1-6`=mrVE=tsWPKul@@|dz;EsM8% z!`xO6#q=$s@O?9|>+ojiJMY4a{X6k=vM&CX&V=rU4Jh+`37_N-Q19-=w9WT%_fR=9 zc0IydgGM|b_Y#WJp2NG(4BC+foSUYKopWBmfA&9U1xb-`>O^VUzlO{P-9rHk;ZL777wOb z!{Op^Qj8GMlS(bhe-px;csrpQkb&wAM$93~o8_+C%w)c;=gzU6WHn|tWxF?{ADW;s zi~C~68wt0JEro&8ZG{)f*`mVS<0AL-XGE`OrQq(_3~1+6BXFY%?@@_hKIS7!d=ZSe zo6*p5KLGPPd5mm+vib#g**95Z_;-&&t+6K@Zg0gX=LzV&@deu;dxh<8cfg>(e$a{81G!65H1Mq|NsZH@^?$0c;q4WuzHh+r zb=;$VBogtWSXfz@qi?AmzScLf)*Vw=>!WlQKJzjQ@V>~O^Lo0f+seHklbKx2LUvbW zJ^L!@go6Enuz7R5zwaA%Ac%;LSVbm2xU z7(EPMC+Sf?cP0AU+y|=x+3ePyOH6V5Pu9%)Kx*M42p`Ym>SI1ryDyEOv=gRM z1(?QnRfnhVU^PcCvmd6r*~P65C`!HytA|$*BySAmYIy`%iqP~nk8SqJWB%UP7>yW4 zsuq)}^~y53+VYs)mezsy3@J4ITmh%ZIS6X?gMG|d(YvfhQK`WoR{TDP-FTAA&Z3H) zoi|iiJC6ym++px6T?I=kfrby7*xV-~<6onxPr`=EN~OS#D`1AF8RlN@W!`U>Zy|hzBXD*_aVW+g!M8U+G&~UcvW4vR)H3$w z>|NG%)|1+u7tmFoaMEny+|QZ=WT>Soram&DeO zOk^s_1#DI)f6iuK!2CIdkR3l7C9`L7-()5}6*uCR3OP@h-_M!&z0n=avawIdJ{vP1Fj`W7~@Fp6CrJ(Hb`<{9$C z)3MKCJ|tULLgr{Bs-oktXsQ(E%`!!yC>R|}C*V=sGIXEW0r72o1{|La)gzaXRHB66 z8>BfmX$F$VRkAZRZqCC1C8%6wLfwiY4m< zM3*B{MB1IIF@yDM0g^*W{%Vhq<=`WW@V72nr?!SnGdl=eWATmrQr)!zr+ zt})m+wSqNPio-!!19#swAYgDg9`4A3wPg`5YUyB^4!k}oKwz%f}B1Z5@F9~e*GnyqqY-SX8Z7+ zpXomjj$+;qwy~Qlyx7;oaC|f03703S*c^M0EjrT=H9|95^`R*_hZdLw>eiMWWgC#h` z{NPuz1NnnZaCN8=o|;JE?kpL+JgtfJ1_S&T(@qi-K9lYr-g(xjVeW5AS?H68takMp z{ODSMe8*Ls-7U@EsUz{LYdJK}C{xLoL8Mkbg#LvGVq`}g$|Rz(|NU^p6gxrw?F1}8 zwG%J6E9HpO9+yR0uoCq~Fo&D{@3+%y3#zh|IdYzlVNaMpX>O}1iA3p3AKP0k4m zs31Oy#)NrdpwteOPFV%_Ju{edn>EXPQzhhwn$f(?YxcbH|x zDdzU!BJ+QwOY_F+)7q)}R1@A`VE3eiaX(drDvw9lajX@Ash!C78jPfZ#Ta0<3x6hj zXQAhm@V@;n+Z(P<=jXW7Y2&ff&vP96q!7#^1H)OzwOq74*@e9k3vtftH#=6&b9~E9 zF~0aH6i;Mhd`=PWp3tX|d`qg^X-!$0a(J}g9M0S&QtPgXkOChlJ)McLf!kQpq6FsK zcAQ=Evc}0*GcYs26w?$H;BZZV>uO8%J$cF0(asWr#ISr(JT8T$K^Sxl!$e8!WLh+H z`8tL9-Fq*pv{Gh1+->?G^duseCcx}Q1XtY`W6`nw=naU(z4MFMy?07%#N5T~)%#^| zH{6QlqxM7Xd>ka-$6{~(AaNi9$HyY=bZh^(~-EeA< zM=2k!2Re?yd9NgN^^e2^g-!4Xvxnn%GbpZ_hIvkE^fEx64mGOMtc)C-$-aQA^a6=< zGDYnV!$hxF?-Xsym!%RPHJWN?LSK~QSmeem=CbY<`&fR7oH%pN;CDHNdi1i!@k*F$ zWPnv|_4IR5J(Vo|Os`(YF!2TXtZ8o}J30L&`{mUi%cJ{PZ$Ew}%xZwpDNSX2?O1N5 z5<5BGjLlwM!iPxFnFGt*R46~HLfXn|D798{R~_#T zt~^a|xLabdo4tSTd+ei#_Zs01!k%@oSJGa>G!qml>0dnjvW(_nxYNAJ+7R2>PmX+ z+W5|OB`%Gbg{~kg1jpWlQtln-s`Bq0Z+-e{G>mqLE7RYZ`ZVpGDb4qdrqG3swAjU# z-rdusQTb2MW4{YtuTvm&nbMkMqQim)EyxI?&Xz=)$Py{&gb%g!&7p`m(Drau>RqWq zF-@A(ekTTJ%n~rE?<}tL>M@tu16a}L@yt$Xvk+s<`{2Pr!h`x8))st-ZImow3+(pd z?Tkz)ndZV#_ZX6N%F$}u2lXpMX=j@i-M2KO67G~yZa;%W*JC^%Ax(jneD1vK8@$5P z=|@N;8IO*k@_Vt!xEO<=h5tcZgFibOXQ4CXE&6YoLgtHRP+e&-o#MNo2|*yk@v#*5 z`z{NPdCw-V|HWJl-D&LjF?3|xW_oWJj(K;sBazS8<0Uz8BQ!1okl{ej9> z8VN_iC8-W!3;c`NLdg^Cf`2ail6aoQF1{!n zd2~g%*!o?#l^hN66QRhA%SL*770rEhfqb5|P~j{II-f5^wZo-oW5W`}1TDoQ9X=m= zI{@ppS>Sg3M1;&Z4>tce#{X=?-v8gly zj-}>8v8=9;zDr!V*)dUg>}DtYEtn-3oJ*pWwJG%SM+W)uTYc{jjPaBAB)Zx0azR9j1-qJOyWBz-LM;QZOX!{1<$c`-!ow4Ph4sdr(J)S zGv>UM>HIZeKW)>=#4drBm1fe}qxH~k=KDnM*LfJ-!hA2Zv(hqg&J&cOtb$(j-1`VM z&GYknjGUp&~`$^ZdY8%)5m~!@=|$mL2-Ae(EJPSReMj0 zG>D$$^LgqIeh+Z7#f)wHp*gGwCzGXxiOO<9e6*bKPf3j0r)yB%CJowk=!UQ^_nGiB z|FbZ=_MPa@m6>eB&^4@k`FG^hNS2z}zSAG34H0N=Vy`g}ggpe)J85 zAA_iCs4;C(5u>SUyOHzyDGo|~!~5r3h5W?_gqiudLcQ%jHnBnmgFlE6bs`^!4L5!1*%^~_2MBE@^}#0&Y6gBfypSI5|7U9M5Wi) z&{DT`B)H}CjA=3AWa_ZQvIf~P9XM9=7kYOKk@V~syv<_Z+sS7+6%lmTFP!|=H{wZ? z6e&dCz*W`7)H5oc)crlF|BcPCd|rqNnliM{OBV+-VZAjB3+F633<>e+QuHY?PVS)W*!c6(c+qhh&RWe`M5&JIs8rtStwzf z@vOP|*7TO_!U{&g@U?3!me}r%ZV6X~Ek^ z;`Fg^C7tLOPLY~+WV6JCe20vp4MmnTDRMQXUT`6w)8@3{?Q$k|`T$!PCxQDu=cvPs z&qWu=3KxFc2?5%(g;uAX!sE7D>aU_8G+J(CgE6`rm*y1@$x76~|Ytp*!v&Vt>(2QjZFC@Vk0%tK$Y zJ>%<{@#rg55Ld-{?iFNdHJCESYf+5!P`Y|6j%7St!>+!sV#-r0;k%_1;z2)fZSpS| zt+S=V*X}(3s!rkoI+R65)G_)4jGhYQ6hDsUYUq=D>jV;0456%_vHa|L33Geosd%Fs zB6mk)<<%t2>FmI&Vi^i_9ZVl<52GLV1niqUjcWPNRL>eoWtqlgJVb@9J$#9Eym`XX zGi9mUdl9t;@24@5qcHpODy+5%Me*$hIyPNh*r_y5P&wj4MpE3nbSH|E{+ll(4__sG zY+olhh}AOQhQhF@74TMxLHCd{aE=2qDo@gjm+dqmy^G+Ygc+d%JYL3Nz}0*dF8B;R z(?MjR&<)kSVsvCt6YAu0*h#ZHEMSy4kS~d+`yQ~`l-o=`N|h$uQm0XGq-g55IcUre z;62O@c(o^msjxWqBsP!bmg$n7yBZa3l%cspcW`gkVuXf`!rtpGq{zJi2X}PPl3%H! zF1u*av+x8_O+f?>Zwtj1@t8Fm%2JrfY<;@a#n_?B*bL z_0KFw2L~ef&@!C9st%jqRv7WujI(JrAYFDVmMWdY4xdPvn6Ad@AroQWb&y@}|Cs6U zewv}a4xe+0Va1>Otoc(Dmhzr^Q}+$bntBqq{gZJzXfrxnE#be=9BX@rpuN8~o?jn| zz@dg%ZrjQFmpo&4f3&hOM~~2}P279TzjF)Mn6rRK!`Q%R#&X#rq+Rzy-oWwLKmQVY zF#R;k-1L-1XLw^>f;V<6j)e6YBYX=V44GfbP-xSET(=Gq#x%31s(FaKeGp=&wnDP) z20JLx$Y$zRvBjYk?0oktc0E-bUNMIJInzMc3JE-U<${LjagZ~#M((u?*7{}_GfYol z-yWZWab7$wH$_A7(QRg_uYs0qElArHGC7Cm%yUT@%WD}#j|^4lY5G9=zGFOY3dXoN zaTttVykc&3)$C_&E))N1F0zpwC)zaRoaor{xy-F1f>plQz$$8*si*J*#cUT7=8m(& z`GTofSj;~r-Tf%!S0~<>y+BWB2aBBlflcTwW3%f2vK0fxQ2O~3v%AvG3f784V@?;F z?BxR8HU~6RbH0(I6+f37V10LgByYFFWoJvy+TuCFVdt5#bTUg9+sW!uT}baPcdTvJ z=ZwG|RNxj#qs=&DcZLeBzQkF__P_A%_j-2fu{+aTK9h~}E)?B<-64_+yDz$1Hw*nI zuY}*ZNW5lV*#BuTRz4Vlw&crr-JK4()@&#&Kf;<8MzQL5AuMe~E&kM;$I}sM7^>X` zyYJ$p<|Rct|4xLAi~#cq=Fs~giT+Ll;PXcsA%-68?+ZJ&`1=snqCFXnnv-CDYA#$% zM&i|NZCu!=gJD^&Ok;-;`?O7;mE~P$;qL8h@-q>{hKXaXUjrMQd4~0C3TDrw<5^2z zB$FO}htD;Ou>9mHsFVzbO`bMq33M}+A&zX{96Ppc;5Si@=0t>v2IIT5BE0U$;9mJE zJT3M_Z2v+w&N!OgR%B?{Wo|6%9ZR2cp{gaMYdvF?f#Ii`i;f2&*Z_eQ#4tb<;)b_(4efJe+wdDX^MRUq#*} z*;wy*8t<26!NxX_T@YroKOS>fpM(R60%y~8mzmW3-4;W;&G0LJ80I+crfD~0NO5)= zy-TTJ;-MFqsr4cDq^ylae(Gf9{{5hSN}4uqm8VAPga1hd*!KNpfA&0P>sk#F0u9Iv zQpbphu?TrO2ywUitdojU`{oGE_2_6qOlX=K~9ZC6zH&Ld|RN9#7imZiB zuri$j`n{5wJ6N-JpK)xL`3EZMc)}fEwRBQ$3eR49(<07Fog^JEY_ALx+K0snQ+l_G z)J!&tqRZ1nEe#Q9-y49!efw}jB@MC~5$HKE4mJ@F5T^1Q-~ZNN+^=`cxy}@l3AJqI z+mFmu<_622{DPJFWS}fG76%^h#kkHBh@?lbmmNo&1{K0RI3F+m9z~q-OIA0qjG4Ng zWvM&|a8#@U9?7*Zc%w&U%lR|Awg;}^6G$z>gRYj3qiYKfAX6dA*i$7USCwz{p@f|Faf|;CrtpJm>nWzZqFv zEoHy9-ep^}6WN^M-e^m3!2+$F5ZlH*e>%fxlje8abv=YTn#JfbsesxzIaYbohKX%i z!Y(H@VW$5jyc)9~@jMT(;D$DawXcQC+Hf|0u_ue#9>pv+n39UN9_0;nr=HLIn7=fi zJI>t6HfY?#mfxHYusI#S=NpO?OXEZZUu@Vl6%}^5v0k*f@s=oT#YUk=*> ziF1y8GfQ)8W*HNcsdi!vl~#n%RHN@~@)m6*`|4xKoH^`WS0GbSTFdgj=Cjq87qi@f z)7i*?RWQpH(EY^%>pD8|?%@~2SZ89Id_3MCTM3T=kyvJPkqNbjSm)3P_CcbVOe0(9 z_Pr{~xSvZO{IY0`{uwHH_z}it;r@&z3p4G3~~q zXj_$rA9FVIEY>%?IW9qK-Lz@nTwOX_H-uWNlxWYqP?171_^VB)k_*y5YV`c(I_NV$DXG20B5YhCg4;S_ka zOhtN?7uM)az#E5;Y^}EzB<9Y;ivlm+1zLjo#wE~SsYKa>RjFvK7P(fq@tLL-t_dq( z7Z=WMIhwG}P7T&PdNr0kn24iC7GO7f1%T}KK3Ff!SrGmHV8#w<8oux+ zRAvsM(pN+1U6K+>-ucXYmy6@e#TI7Y^OVL!yd|&Jr?m9x3S`aO0e@`~6sF7-djGiz z_x{ckMy}2kjg`J8nxS`9BvaS}x2fV3vPFvA<_=-!3k+D>o690&Q3-6eS3v9WIe4pa zKlP$M_I1)lrZOi5C$dZMXmJ{zS~=s8#axVzi01vkY+MUDgohV?;X!f{tXCCbY|3e* zZsFbZlBqb{Zw}rIk?4w9i-73exW+TZR%TW-apq9k(VLcM&enXE;HwJ2b zf|AyF1WIzJ^W7D|=(j9v^hmV&Z9svcj;hwM>-NUT$&vG={EtTQiRn33r zcQ9#i7Pg%3W)ZgrBWS1^+(&X=r%5tCJ;+6`O(E;-m&g`r2Q!zpoiMnz9|nIi;qr3< z%bPTXHTDC0b3lXc9ag2Iry7ybG7D-cb09LEi}i)t$h@V2jqw`TXOu38)z8Ka^k*-#!r=Dy6G#~m4;?kB2&YLe?QIaxwjGACwOL7 zimYYEL#D;`|I$B|GsNVmtuviCptPNqu=o7?6%cs zCiAF)jmz+YX2v|+pFIZO{FG5I(GPpf%bEG4G^|KTgV~9FxOrz3emu12`R$ny-5o)t zc8*kSV1fx)kb z&pvgWp?u5(i=Hlnqf$N7{Hutz-Jh5b%SA&~CU%t9z$WkyYf5ipSLt$H`s)}-O+NPdo-ZHu>&)6qC77GGldXW>5q58cM{%;i$3 z9$ta4R0F7OjzDm56viB#h-jBJ^t0KCD)qf+Zn*?q9@dGatKL8>W539Qr}b~VJSJLg zqD6i)hR}0&qJ``eyL4C?9g*s=UU!NnmgZ8a>LuFG+7Qw$$>+2`(ab$Ux|or{Ohr*hZYsARKV~xD#-1HM+i93aJJY?oO?|A(>NQ(_zXTJB&uTckL236M_DZXe?H!TG z9?hSIqMVsn$a&(^o3G>Al_Oa9U7B155EiV-?#R(8uy4 zWpPe=5$Y@_qM>;`0%wTm+gUr3d2UT#cz2<*WETeIXCf{1GakHp18wcMcw09B7a#RU zFYgJ9#hUcqb6Z7vd+oj)Wr{Mm7yf`fA~}4C|@eJ2%-~~J#2EA1+0EJ z!5em{JCP1FZ=;IdXu_FJ znx)ClTAX3K<*G47@Vw>h@@15xTql|y>ceC{tY?NBuCOqxBdpYH3tMBWNtJ6ykmMI9 zid(t@KMzcSt;Gx+J6ee=zkj37NrL+GeqY88BaB&Z28&T082R5vEcx~c9yct>ajzMb z8oQIB^e7BG>&bfr(@`}NxO&SQHu3@Z>OPyjoU@v}zi^tZ{lxDE1v9v(>i`n(tMHtN z6HFy1A$DasYf^u}#wA>0UsjFAf+$Z^$)!SW(`x*z3Bj+uOE9V~mklVn!T8T)N$*w3 zqeGH*&y%P1Cr%;kTOs7HoI_y33-;MQo2{%_$!2|4!Sc5{81zsM+zN~0-+^#%-G=O- zv9K614O@!ckilm%wMDyOWD)`Q-}jKaE*1B_mgA=L0wg_M#rv_*@cjFXIR)KgK|z5` zOEm=}2Z!PCO<&x4-Jfdiy+)y-1Qp1~VEWQsI9i{8dsdr7rf$xn`SX^EO7j1}e$H!r z^H8KPiFjr`?>Lj(+r&KmIj1=E5)FyEM}4=iU{2#RGzR>_`u6_9!!NSJqhuZ7@5u>l z>zSp@UULWYwOoSKUrV9ju?L_0hSAAIw$!$aJD_`)vm%{ytn<$?W|DG+_0Lkr3RQLd zp1cqLjoyKYa~I&XbpcA^(qOtY3g?yLG5$v`K8`wyuzV{)(h$PA@R35zJ0t2z)TYJj zhSTX3Q{1`Cb9SSqATQln(Bi&Ui|N*aVvM9v;w3JW-03AXdv#mc5;fZs-5R!04G~P! zbPBVd=*Xr#8cA#Nok^<5o$7-(A)$N^)~0MhWb;Yt<&1LKwRZyu zM@Ky8F4^f&__4RNvETt+uYF2;dahAdb`90&)RE7*UhYRyCDVn9B$`}@l{YG|cTEw_ z>Mw!1g(XyfJHh1LHDTE5%R-S|wXouyA>^!wL9fdhKY2gCtNlA-B*ZCqm=>7^4Ww0v zU*oxiF4z{dzZ4U)Ae!Vw*W(5Zxo(y5b8d^6-M=a6;#HRW7mnRu$Qbu z6=&5}%~cZgUMmVeR`dN5p9L4ceTGo$C>Xy8#fHD@A-z8mB@w%EE^ZsPNOfU{iZXev zlqV_8RC@c1bF#BS>DIJCH1dc(c{`1zeO03o`*{Rf)J-u|AzJt|Yok#6%U5uNr%=9r znK0n;G9kp{6fH?RPE%W|X~*^{`1>UkmbHgC18g=n?pcVFQT$9{XGdMn-N+z!5>1D<;Elr7spOZ;xp=loHSPxeRPfkUVt&bfOMoNsjeGHGs66{N$OZ4_#U^m*)SBp*4GVP8H<~&Blg(_@zJ&uiHU-5X533c+EM&9bdbZ*jpR{J=UtrylXyQ&tN z@AZ%ZLvPdbqC>R0G@hJX!f2AUDo%b>!IH5$sDCWQncJEqb6lK+2MW|wZ%Ml=TNvu=vfKucLVdx$%9O*t=r+F5$pR>zqpWteuk1utI*^X^a-ewY4- zXH(Uw_5VF}L7I0HvdHOiFimdWM3K4tyXcQOIW~@X}?N^HUg0>%2|wk1o)! zg!}Ymf+2R_pNaw_YixWyQ#k&9tZ?y=x3JdUigcHcp@>G_7xKF<%y6vX%+ET3)TOYi zrH|$C%#N-~9roMRV7KxcxSPgMkJV0EcOr=jrPCpb{vUa78ja-}zHJ*ai;$E^iG-AL zpZjr{ijt%>sWb=;8dVyklq3<62nmS>nG&MXh@ydJ6q-aDL?u&$pWgHN@c;h2Yd!CW zXT2XvqXn<1{u-^nt+cBhGTbaGYbnapj%JJkor6g$}Am2OAi@wSEC6f zbd9HCTkfCx9ZKs%`jd^H7VYU7P8<22)bH>kL=1Y38EukuMog8WRon5r@iL{@-lA>V zcW6ME2IV#FpwSDQ+TbnSsDX=*cid@TtRj}|~fD+|5r*5UBPBUpbT49`|Y32nzt z2n#192r*{LP@4FG$zG7c3x!dnx{OFoLyNv@l%Qwb2khT0L!(#nsEC# z4U?nWzei9-^jCy0cm?gIQqbkoBz}=I+h;`(@qYdHn;D2)k%#lw6)E^uKicW5Ot=4a zqWG0K6&FfV_Hb{$cU?zaoHOw8Ox+X2gztN6^b=LfG`fVwLQY}uKxrx6B+Y{Q!kL3{04si^0A_S1I0>w;dz~RWWT{4yX^eX zIxiOTJQJ-Hl7Q-0NAcs-TO27aN2pFE&s5lwY@R(y^p4>BWqzkC6SzB2ot$^`-aF?B z{Nk?P49`$(9yJ$ZROaELCU+@)mlN90D+_G_p~B{k`-Bk@;oLP_Kpj_ZlS$A6T2z_M zrX1{H?Q8gX?tMPHw&4zwY`e*V+Y-pJ=@dQY-|L#-cyw;5h2)&~=pM74)<27+V%{m* zMP8^s8;RcHbXczA?zYU66rFdPh6G=t^^SQoYF#Cz7Rb?}cv*6@ZNensLRjOWaPO;M`Y(JYT+^DBH z5>dLoI6Yl5ki728(f0Q?__N&<+UR!x|s*@Yup93iZhE8tmucXC4FD?1dq%UaO3?Rj9ScRR;vfm$UL5<;tY&} zyqz?h`yb~I2Dn$}ziBvQ!oNNa~(pza` zUe6mzd-TnzX|)BjXxYi$ADYUd&+VmI7V9Z3*`2f58raxwN$%B^#E6T6U?{Q^r1fSB zjaFvlzRi?Q1`np@nFobid!q%1b;pDzJr{(zYhgu$0&2thxZAc0hU*@|MfMFfta?Z0 zzBSbBf1S@lT5d{OdoHaiLa-l(KI6K~Nk?mM{Pd!2&o zbLoxyNlMtmXnx`t+QyyoV!IpQTlpQ|PD@hCqDVaQ3B;1DMR=@oUTAGE5>`}X3iFJ9 zGygje*z0BW?4v2qZRJUlXh9DoP7kHRbT=}78bbR{kELVtO{gWvlp@Xyp|j=v>Bs{m zT2d-Q9~>2_RDU3)SY%SEZ8rUQ@PJCs^rI-7W;{Om1otnFrmp%?RQl44#6R35^|&|W zw!MR#o;uRBgz02Ows)!udgsRrU1uxc_4WaDC*FYB>H9tkE|L^{Ertja({hk(nS&`B zm-)N41~U&b<_Q_H#<;+(eV58S{s`Q5^V?GN3)G{-e21)LqyxrDxgue?- zW-OySi>F}p>Rm|Ov>98+P8Iy6<_IS7bA$;`LxpyYC}HQt1i>ccIUO@9p|iR-C;>bN z9vF^{zN;AfbT#~&XQJ9j9$!Y^w%1X!SR;j06RqYPN5h-5X%D#DS*btV z)|tV-RDzoLUG3Py(WKF$PmhlCjMRgnL=R>|PA&v_&Yw`TcRTs2y3>mJOX+!DG)WGR zqR}UhP~Or}r1a5+j{4isJh!hndgL6w_Y`1qxfK--oIw+dCs2j>M!IqHGEIGVor~8r z=tawD8tF8NzM9yw5p5nUCwLkw>`tT@-ZgLLuKL{13h4dxk)`SVW4)gz2oIu{2rtLl z2tRn=&wZ3EQn4IDau+XM&o||M@E*N zbgbY94gNTYR@zHb!L)yf?op!L?Vn(h(u5ArK+=#tLg|A}QQxh69E|1tv*TV!O@E5f z_bzkx*&S#L&nPp#n(`mMqM?Td)9%*6bShJUg1#L>g8UXlx7xv`K~+%tHb@xpMMlWb z;N9&H`^jtY8p;o@6XuGZ2%A4W6K*=rVMDd&vaRo?u;phNoqau@s(0?8pMAbKf6g1} zd)H(2q7Yg?aTOKMkfT>|A6e;gSv;2bkLe#TLEqs*L~~Zhmey$`?W;)#I<)A*R!N#L zRfTKOr|I6+ zgS3Tv$Zm8yla>@etJMvswv+)h#Qq%G6XIZL8;pVj-&o|{cPw;gJNvxgjIdu6D_paP z5Du?5MU0$0)+*Ux#Gfn#<0$fR7A*#e5RE;}*&*j4dTUCfdWlvk>r%n?6PWb21YfGZ zqf_k=CVh!SBA;80U9*s%d2Fa-E@<7Pa~SHG2eE1E@R|F@Z|yC@z`YOo>~#q}FP})G z9$V178*`}N+O@Q5S|W{0n}BwUaad~XhKX4Mjht&uueMmxiM|IY7cvkvvL1y!`%yAA z2v+YRv8KI)9FMe6LPZy4)a2pjvyTW7Z-l0b7e2oA!H*#koHg)4q;1(GYHMy_!#51Y z$hkuh^=~3pJ+2fwsyS1iy%i$Q=O8{V6T@^;u!}$6GO;<7Y;~D_PG~}J;wSWk--UMV zB81G}jF?SBxA&*gS*k>Z z!$|5~gQ#iSP`1WUo<;Oz(}W$-v>`N}+M<$RzjO~Sj^cMVz41cj$|1tdg+qm}ir$oa zWgfjU+Dp|BcMDms{e&&5YlUtPE%b{N$3bs(*k$Kn@|{aqy7D^a4`0mrJ)3ClOFwG% zFo3bk1pK_sKcBY^L+%R=+}%0>gDMA6{3%5`HDUnm+~tX8TOWk(U4Y6R_po(q7F4~8 z0LL#hCZ&b0RX(SNYZJ+-(2yzy7<1R+cs!jy9onHLP~_*hD`AbK5c!pw-mRjJ%uSS0 zv4R4h6bo&V_l4uO&jebyjot6>!D5a(GYtcMT99DInL_)?_N{Rq7MUvOZx`8_%~UPeUZ1{hhY| z%=*44QF*KinJw&qOWG++o56efy^;9y;Scg(4I*9R!89Agsi;+kW=Z@9xyb?4mJ!7L zpo!GqF&a}Zt%q2W8!q=#BeA~zG*nEJ-nDVYr*{Gt2@jDU8;)~VITxsPC+=$0lZjmq znd$sT7Pc25GRwz^u6U&U%Yl>383gy0o9+m@ zbXRcNQYADfX2OwYWG(BDKq?@gzU(fb{)NR(H+QLulMPeR7>Ojt+%08u*z%@AuB&X>usM?@q!# z6E%ScD1-q%3c}Kl4#JKZ(*(yu3x(X1_59sZM^5*i(WGK|9J!>2#9wpKqU^(zE`>4Y zum0?oqZ>t^nNPBlrqHxCuMzy@KWw`{kS?@2Q)tX$dbn;k#Wfs9%GM+JIOH@wA9JBn zd(OD$JcG8h9aK9%jP~hppKq!X&FGOO*{4c0yY@JOUS?p?^d>0UhTvZ0Z0N;~#=GNx z;CTNZevZ+n|L#0Pk>eXIoLj)Vic0)`V1ti^zK~2lk6{0^c$ZZPnJx)Fi0jhLz!Rs}7nXHQ;tli)W;GrdaJRw$G`@ zl&Jf-_v{g_J!wO_a~`tgKI6<#9on(Pj=By{pwHo>NX>aH%|2;DXY*`ng*4B(hV@{@ ztB>d`@Xy6^}8{(r>C>}-nu5lsqKhbZs$Yv>%6qR3I&R4~(=RcV+r zkMZVg`L=0P@QrhTg{4%G>x9u8t@$otJi1o$0Pc~2PiF&Cp#Z?D%<{qhVHpZn^Z2-N5@5&8$2HuGiPBW z?*TpdWtfl9*nM+a!=3xDcrSWaiVUkLL` zJA?@qGUj?G4VY2icag-X4pEwuCH+daqr!v@P0kfv*U zB&l`iLo9sw1g6cExUldEyKsIOw&k1R)2yRHtm9eXuH7Z!`OO@r^{S9n{K{c*cieGA zAqhtdi_oo{j^B>ly_yk;r$$4_=ba)Aogz>F_zY+8;SW&QB1U@*H={kz6YVQ(cz1pr zwe*`$U*DV1?!cRPXnqnN(YJ8vj5c@68qv-(v2hxCF<==+0eIyXatnIBPQUpjPIhOQ(uUEZ0v*1p*59_7;%90Pt+ta zeeR0<;Z4D#?&GYC7^Nlrz``g?cn@35Gtg?d@V~gqe7^kjc{7?!Z)4K(PjGE~fy!Ic z@pbSXEIe$B@ZM~?@iU+MO|BE>O{dFcQz`7>47#k_%q|7}W#Pxl*_D*_=xOvs^jJqo zJKV;M{7&3n(utp&L-1T9>v}Ctd09&IHhU;Z-djozM`x4H!r9cT*@q?(?^)S(&6Q+`v#@?kJW=Z^c76dUW*vGpoY1x#w^EzGDE1@`-PNwOw6`Ovo z!{*3z%%N*|(vrsWN{0Mgr9*{pq{((_GgG-<#TLXoW#iYY;Jf}XBwrMRujymF|MUqz zua?1O?RNY~al)@LrdTpt8Cx=>k(a3kS@TFLyct9Divy^i*i9CJO}Ojy2XPIzn>CC^$i!4iEYT>#i0gZzcTE*@qs^23V%Q3*w4%u`OXD z{C^W_{RC9R*R%AmJK%L99HZ7HLUz@EwDr5R@HY52ZR)`BLB$UOeb;vWCU?9TTwdueIJyY<8Jv`Y4Kcn7-| z+{7NKJ;B?j65JSN6M;Wd!FMk6cBu{Bt$b@E3FVUCN>#j$=Qtht2r9 zh~)&-pw+AnWBRV)qG$-qTkpzFnr>ugte)dUcLgSW=ZxmEt8BT$16FHM#3WDM$A}eA zak{z;c?}-)>){SMpR|SkNy#H)tuk(^%VB7q5sVTX@Lz?1?Nj=gRn141qyC0zEV9Fs zE~`WQD;N6K z&fOkH`s8aE&EgkFG55GMcF*Mv`#1j)bBesrT2T5QRjnd}@WJeL znJT7l(1hQ4ZMgexWoyldeJbq~ZHc*tl;yd&6V}ci#igLXWh!PY$-oX9cRKoK7tPaI zPD|grQBS-hnJ_tuKd_i3e!j*s<}2X%m?=JXJnHslTSjJU|gNMB%f@6Mriauud*_C&S^ zKR1g#=jWUZ^t~>?&9L_{Tk#S7tzd%&83~t9TMGRmrwX=N*39>`2dk(KVuQUmvt75+ zS>mKVh7ZdSeQqiG_w&IT>*18})`SK)=u->t+bEl!zyXg-n0jyxa$MG8@aZY2{JtO6 zfBDc{GZEV+j>2~-7wqHj`{#exQRtxUwAjs)HmjAh9b>E6{(?pp@pL|YU`uF6&k}Nv ztYk*ApV-}tpP5pt2I-$2OA|OxwIsKX8AymQZp=a4Nw7dsxCgxEZARUUyHu)K!hJrO zq&g;!)zzrrU$hLkYaVk2SLhx0#A5ei#!8;C0h&5!lP_TV;_k54E&0r8Pm+*mmMKWO z=L?1Ftr7Hi8g!pAv~8SC>FZ~b>smL;J=u)`mlbJFu>|kQ)?tWYD_R#`#P5TK==skN zfzuYi|E>%v9ag1JrlADSf_*u z^-gr8=Y~e~)kO@?X7{j%lhv^`t{v@TJ0LNw2QSqp-aBtVcguSq9;yL%V zW{|SABsG6isz0npn#&>}W0nNffMD!A(81ic$>FVm4E}arM1AxJ6C6JIm0QbmqSg8_+`isdh zTD}h&2gK-2=oiT6J;2AmTS><#f#zMlN>h#ck>{>KR5AYp)~o#$t-j8C4*_yaSEUhW zbhOCilO?Upyud2o_A&DWBdj0wo-(4Pgq+n1LayHqI=pWcY3MGcuB36on0jNO;;NG1 zoN|+`wi|%ON`_EM9ZU}Etf{-fl+xU8kdJdIsXj}k!o3=7u~R>$S+BreD#}w?lP*=X z45HEQoICq94bu7XQ2iK=^PC^P?PLVr`Dx(Y9YdV3Ger%bm28`qKyl-fN#61v-1wZL zsr(@<3QO79!TVXmj$$VNLsM7~HCUMKJV01uvXXl{IM>k=By;|#u5I4L^XdB z75&W+<=iO{+5gKx-sdtT4y#1nHF4Bkkj4|KB*n;^=umHD`ak5*!n<@G z*}AyKaghHtEJkWu0EXQUz{<%7QTn(H<%8Z}&s9Upu=xbNj!r}<)gnH%jb%TU#r<{S zSXktZ8|{I3;S>zDA}b1;WJk^_)2PB|65XFal{^kHl4;08!1t?gdY6N{4+q209wB&Q92DH-=r;abxK>B3S(p6Zk5R!?$OSh^f8;xpP<0 zKeG_CpG(lWEwc3dl?ugPTa0bh3*mY?8s({5@kcTc_ihE_uJUxMn6;Rm1qpOWV+nM% zXY+o!HD|i|;%I#+Y+eUrT+S}~w8fbW%JfNR-6`B&o&kf(B9yO;VX0HnnaZ0t%)s(9 zrmp_SnY=2bQKScF-ZA{3xg9*I$HJ7hGsoUSW)ZxOrUizRoNF>I{;7*Se&g|HlrJ`0 zY=TAaBCL459V%Bl%m%L=Dsp((BZ`qtWGD1)vn6MWSzXd0VX5X(!8kulFnBPXpQq=- z?#yQFKH^KcrO}jXafl?mKC@BRhoIvT@2s4!VOk|UY+SYyFg=HrUAe-}#@8`Bn`KzH zayOnmzkv3C5}YyBi5I6|;)eyF$)CN2lkeXnZ&npcA6>ytH9lnO+8^l8syDPx>ki3x zb<-~g8G(!k3;U-?khU<0dJPO{bIEBIt}YD?nu+A%dX}1co;hjkV~_R^qc7K-xFdH4 z8Lb#bi}H-fmh=4#>I~?-@l-P8pF_GM6mVj=2|D!c;r&3K4h~nKjrj(o${9{~Lc^(b zdMZh9*W4b-zpVSfC+0c98l|Ros6X$GyIny{t@si%`g)3;IC2+blZr95Rgq>!{b0}1 zG-2w@5UjWmYfX4YbiWm(~MOeVZSMGk1>QlD8u7_DEC32m|_MI}&5RIb-Iwjqo@%0@Jld!)uib zqF3(3o_=ef%H0@AihuDT@Hsx}iIIm|40_LA#E3~%NZEc&R5i9hG=0wnQCid>(rW5Y z*FOxS^rsR!`kRuX#U4nuOPweY0UUaUSrLM>Z>+G;AYbTK`YkH{ zqs>0XjApO>EhsZ>HswkLlf8=?RCtcQ#Qr0znso&VIyJ}@>JZvm&)U4du;x$ktcz!T z^}Sw#>fYmU&J*rlIgdATW&Crc2`dJ4!s`?xiK9)J@UI;~8-Kt_bw3uqScA;5Q(;kd z2rcCaczX8)ZqJOS(4R-RV=J2E!nUHU^%cUWE7Q~9HZY2@OkD2_3>)mNfDOv#Yw!0)zN7pYa|+9be$|a|V9fOoU#@ zVo2Pogy-K{e9q(VT$Q6p<8I!}$5GI7*#L?9RnXXLiC2$JC^>aJ9r&V4(`318C~YL3 zA6MgCtaSL5oIsvO0iNqjL)bM7IQ({m2k$C7SC6LLPmVM?Hvj`42jU>l2RhswML*iL zsQ3O5nyWGniF_wlMg`arKN&0C1vI=j$6~JxQgbh)DJ6|G;dd^+_?*VZq(p3pyN*QO zUER6!1g1uUedKqJm6b7Ue#LquzVyWDgnj5f7bU!nKPT*(cUMSNF2#}DN9f7F0)v(# zr28U?e0-zHDQzi|y=TDU%wDLvtj2S>typim3l)QpV^d@p{JdO|ps7Ojy#pw2wI=x( z{KlEse&n)DgYJgCglknD-ar#WBaYzWmv~%1P=ufSS@xJ8iplpQQ98Sfg$BH5mBG!dXYPJ-`4K=xd{6nr z;32O2-ocvQS~x%7%p8A&u((U1Y(QEVq~D%Hqx(y&>h8mQby+glEkkvw4-xc8f;Otm zrvsU_i0k(VAD;1k(d)}-Iw(f_6g6m6a3-|H%VCsyAIopdfEMo@T)MRr%ld`lVqG%K zz2M(V}S$z7q5Zo5w>>l{ZqTjoXGWAubCd838?8TP`%Lj&N~ECvt%9(Ldv z=jOjS#B=J0u|;)1+3Q7-YkL&GqsimIjCl~+oFN%B6P2>N;rc5AGh3=LZD%2l(G_%Z z=i>i~A9HKd#sBB;dHlwH#Q*&Ie|}yvc!2u<{QCdKpP%G7DE!~}b^qV|HUIzZ-~4~) z8TsG&Gyh+I&x<<5|6hG({Ga@qJ6t}qh>tZawyTX52YSNon>Tl2uR`Y4+vJ>kjZ&4f zX>{pZSl02~`^7puJ0*=XrzP+@RtoFKrI5YhNzyQiC)X>pD7e>&oa~1{|aN#DXi2ta8^`}OLph}idXZn?uJ9Nmv{dNgS6xE)DC){kp~K6hXnAfM>gs~9 z_;W0xR1Uy%!)csIK8Q56a5T?dM~2@Q(K7kTB>jea1J+y8z}urpzl85Q%i3UH`x@Ub z9tC@O0&g{AVYz8JG6q;-(Vtn+(pfK@PjnY-&dwGVs|2zuG4oi}TzfWX*frGNI)m;9 zMTmasgxXA0;LBKJ{N8tpoJx+-zBfVCJL3$t zX{MsnuL$dXXW+)I8ECH6g69%fCN*jiQxRXrF6j2MIcjg&8jTMu>_L?9@J*P|QxPb% zEG)vARNgJ&y_`*c5%e%8f(9#XqA?lUpkLyNOykM85#oh=8~xxOw+*`%a85AK*DrhL zjllo@z=Yp7_C)m}ZonY={Ny*vlkQ;A!7ezZcR``yALfU+;tRigTVx9GmH0#M`?_hT zHFrdR))a2HOA9w2!)1f^82F=yV|;f$*Bdff2KRwzz_S`Eqw)A z-%o=q!Uj{rv#5RKpt+)G~zjT5p13akWGxnMDZY$u04)ljhU#^$%fm)5d16Y4;Obe9Gbux zn`{rs5Ah@Qk8{YqC>Hh+$*2g{gz@YrOmzMU6Rzd35pPql!}T20>%(}Tt`JjBreNER zGHiV_2KJ$?EI;+S=#69-KFR*TX3bg*y*Q2**&0xW&Tz86-i`FyH9VqsXBW-?r9H#a0i0*U1uM?x0yT%}vE-{dW46vnf zfurQ35u*fm z?nJ6>5rh0vWz2gY1DELuP+l32(%=)MDSw_ebmFV&Jt{)nKRqhHWK4s- z)M=CIRMdL$`;B2N3d@*aXFXa7UNb~cO}Ru;y#=(oEsrEa70vI?lQ7@2|BG;?L&TI1 zjAAP?Mz9_q5mk&*q4BkiFtHCt_T6Zd%{qWxO+0riV?#cT}{Ok3aR=qXAX{7O66B4(9@NCU${LVPY-9} z;-*3v^Z#@2u6%&bbaCO~({6fY$GwzC&d}y#A+%gGlHyDHk!iCOt(8=ujHjHD@w*R| zs~>?1-N{^RM_&mTrl5zRS5bU<Z8v%_8qY>9FK2G00^$$S5yQl}O( zd;Y`jnio*su15Dq@mz_TGAYFg3 zG5qZ(1RjaSjTe(COJWNhYFSK<+Q&#F7e!*%TAm3&FHOpv z--_Y8bKs|Q7b5-)rCiv?);!wGtiA5BkFS2RyS*yBA9RNu+R-iS?GQ81Q|@Q3RQ?mV zE=L6x+!ywrq7alSBm71yxg1RZdzuNS)j?Rm`0m;ICdSxhAZO|=jM#J+ar4U&yVH?A z?OaR6$E4}G)ef>N_o3{`Hgva}XATB&|DFAH&fPS|x`EEnfCswj1wrYhvC!g5!Zxi< zLf2R?LBGjI*g5qTUG8e4JiQi5jr_sR7YsnwVR+As;bElLG!b*AuSP@vxrojRrwzhU%1=5>fA;p!4NXNM zrFMW288)28Dw&h>-icJBJR8e3J+We~Dspvm*t_a%cKyc%HdJ{s{dZ^q4ZZ13(mQ)# zYCf21zbVo;Gil76WPz$J46V)Cc;s~kC%!+!r~T^Woj-==^>8=ypLK|HaD(cOE!gz; zCF%J-r(YJw$Rueg&97KSE{giJT0swbu7=1EH-Xu=PR=>5#>v|+akQ=+$DUS0bypXH z%9f({^hm_Uy<`1WTk(v{Lh7xbLOv_KY4}fHQaTe)yO({$4Le!BH&Y>J&-oC)JR3c# zB4n)wSu8ar$x2gNShb$}B7!M%?=c#8!ees<>eEFnYblrT0kT*z-% z!}-02SoBX7zk6HZCHV_Ow-2GvQysKT?KjPT)=t?!OYzUR1jAo8psiVgJBie2yR<$v z-u5JCxKXpX5eeQ_INsn1-}Brzr{jci;}@W(#v5XMrzZ2`0ok8eS9U3E$=5Z-tXg1QZb?8j2ZdyPM)Fo43ZvpPVf@DD7cQlA=tJLW`AU5 zSi$`kk(JtT`g?UO9p`<&L(aX-;GGIK^^HR{PNC@371ZoXLzK%^mXNEBzD1nJT~iA7 z<|2NUzQTu+J_LAI?ov&bTCP#O_h|}Mk61-?Wf~89vL&Q@Ga4{)IA8FKe=pLn4OrE>SQrqM9&Yhv7yHI(Lxh2VQs4BXR>BSaWTFH_BE z@{6hTacu@#Y~nH3C>Mjj8*!JR3C!~*!niG)j+NXd)3Mj6zfUHb+HT>-xhf+!5;3p9_Dq9%GbWE6h(zaqjU+ zijyuPu}5Vz^je zDdC*ty&hp&)h=tB>wv^s?pRE2Uoe%_a~8oHj>CgzOy_1##$5x>_9>o-u!DuT<#-<# z%b!5*^G-_4u%rQ+wp5;Z3lC3}V?%Tq@1R~1-S{WX@@B1OoBxVa|BXsC{LgS2$qy@& zY=1L9(*`zAN?nMHRuT}R+3Cvsk5LuV_8)1&>f5d3=% z`h8r3hi|JX`_fYih`&!SL-t`up*OyIO@em*VIfB+SU4uWUl}JNg{{A5!ZIIi zrPD8WlS}(ja&h{_QrjP}Z|Z4GTe=fFjFiYMy9J(&V#qZZfEB5oY>7DU5$r9&?wiM9 z{)AeG;nAd%KX1$gzsUvoSK24Q_lFXSBfmZ=+kjw zByBvrk|e&k(Kuf#ia+rH|6aa8y-*7&r;!vA!+oo3G-={=7sNFLW7Pr9KebfDPQ%~q zZF&`(b=yqXJJ(9c)Ez5WF3o4Bz74|G!82j*${jBXcOdI|3ll$$!LMb`=+0k^#UrQk z%2^a4g9cVeG;PGj=@!`0G5|y@Z{1e{9SKB+WM9>JsR{XeK;DI>B9DxKC<=w zVH~;#6ZUf!`%P(Fa*%}8oc`D%_6j@PZ$Rq5Vw^ry!G`QS$u6$m&zwh2rhi?gbS^|h zIoh02v#J1Q1KyxiXEshaOvZ_!rbu_cK`~oPY2k)as`28HfF<0!Ct-)}+s?vdY$gsZ z&O^k@Xi+~2advXUNH%WcWGv}j1%+25u`wl27}0e}&^>=ixU@9|sv!sQ)Y%=Og90gY zj2qcLSW1!GQXq^ukL}UXSRyeMU1FA4{lW-|dd9foJRNQ?=J1(*8hlMS$7j1YEYpLb zB<uhZjhn~C&i@p!rs_klec_LLpF z+{NO>>}laHYf4;ZO^$M_VZM46(!MOh=3*y0s=kCIG`y+sgclj|Pu`>b_fclpJl?07 zie&>e@gp%De^pL{`6l4(um_l5)r?CIUZVZ|7TAdKuHe2)_&y>66D7|;vnCo**EFH! zFb<-FEuB$Bl}w zzd+A~yRevY4ccCD%+Tl_d-br64UH2cEwevJ@@asw)*C2);T_p{DVq78I9WVx$JQoE z(pr5ACL!mMxj!4@dmi$BClNc+f4{WW3qtdq^Mr&>8bZ@c6Woh3303?;tPMz+H}a!}Lkz zn-Qhwhrzct5P$8A;6CmoeU1&J1euLg?!6w<_Dn(jWmUwMNpg?kC(I~)jqFbrG-8zr zl^Ktxy*K@n2&9^at10Ww0a|_CjY`#ccf;i_n*QV6C?6Mw&-;3p8*6@bi`5c7(oZbJmw-hm=VyGsa#tpe`uuwIi@5hIbzpn;0 zPA)?yKI5`N0kox$!?K@03b)S16Yi9qducBfw63S}pV#s2Y$gJ0CgP?&Kle?RCx=vd z8n85l(!K`McjMNX4$0?~3o^-Gfvl4=%*RT@RpW9mm}e9;}yt z4o2NQ!X`D^Lowa}t6G1v=vU8$4DAvjY=5cnan@fP$!tVX!~-brd`u>C4=Bdq2DR$; zqulNn78-cSgI6C;_ZD%yC&5fYzYt~Ztyt}meY%7%%zoV*gx9~yrGv;mW#lxA; zuqvzu^10>c7n{ejnloAKzNM@#LYcllQRQg)50Qc;#LnfV?#Fc+L^qWfy4V;&h5kSwh;6yl7B0aRoKw7CTHqVUr$+q7hvUh0}=Jr=#uoq?nDRZ%s1g~A4R(PM1v0Wlcm<`VoFv! zLDT#K=x|*h+-v`$F;Iq9w69ol1q z81zmOVzy5b^dr~Ol8D7*;5mn+El&!~4&g$3QH?csl~>Hq^~5i z5aT4aZA3b=alQ)We%JA!UoNC&b(zNuu=vyeL`5n^XngS)vEQEJy59@L^4VL3UlOu5 zy+F;cKUffSl{TLLgk%o97m-hgei2IH<1J>PL`t|d&a{zXDFr?jfNP{cJ5z4pVEsK(WMDFSZ|k& zR6ako`oXi&E36^NOQFTonBru3XHHq2{CWCbqhTq&I?lrT9v`T<$HLku67}x4SpS*l zn9guN-ix`4BNch@-EajPqb8#Ep#`#?^bwwBPjgMIDWw^tT3(4fjVc%y%g{$Bdp`eO zhQa>!h|;|H|FHLF(Okdb`!6yjMMz1MDMUp+&vn0>ga#5Rl_-Q#8Z`JeNt0Q|G#Cph zl}v?1lT;`wMIxm+(LjSVDD7t-?6d#+pX}e>`(z)qYSl`gB|h)xx$oDrm1N4sN=LK&CsD*kVr%2AI7Lu#pwR z(jx{mM`8wj8eu`%oXL0p?>%;dGyL-n#!$n9DP;HDh>Fc*u|J3*`MLw{t+ytd*JG&T zA@5$-pQ0-zm+4Z}CGzI;r(f06;Zm=MkK;;lwxa}x682+lkriqW6GHNIQTAUpq&zNR zli4No6WfXcgJbwlJB()r4sc*(kr+C`O{ zq}a{oBcjKac9fbli;4%EP`H!^2@ABT`LzxextwKjCCAtsV^0?KE+6$B2{<|bG+yri zfr+E!>Bpx&EVz1~&#-ba^lTTlO$*1%yW!Z#JC3>q$~353oCe6slUhy+?7I)bpydFp z6u7g{e=YCzcyPw-VLaZRj0gW>amgcydPm#Ps!hBj7%odvB?D>hS7rLQ$%viF5odF= zrP<<^{S>+;ftEI(rsIZQP*PcpudS-+YZxh9x;aEpFIN`a9_f?efuU6Yb|6)LmJ+H> z#RTm^3c{f{CBz$##`5=ru*+AUJ4%gkjqfkUjAhhSrcX!Zbt(154;EaX%Jc#Ynf1~4 ztR`~|TA_hWC$_>zXFKM2`{QrQT9}6fA#GPO2HrRdU7ua}yJ#O)TVJ9tTP~4(=tbHy zNSel9`~`PGiY9zfKyNwcsf#~mVR~bP>E5G-3F#Vw_*YAM73o4};#}$Pf%C%nmLj1k z;l7aKbWYS?r9l+bT_7s|qfY8tBk1=aOPZ>x$=Qtz(=SazD$ibB;Iryx&sQ*s9*VR~ zp2g(;oEdMdDYjuO<@qa9N1i^Jc@LzOf7-#EXVcGWBKeEF6ISyTv0J3b$w-Mb?Oq^Z zp%|%rxrbq%+-v)+2;qv4VgBGI6;8WKonNoe);e1%d`lD&XiD8x?-1Si4lnr`yI4F1 zHKQZ2Sw9_F{66-=)*VR)JYky_B8-a(67tPLge%$)S+dk^_UBk5D_`FW#Z@;jtMUpu zw`F0tO#-T_4q>(0JQ~nsOT$?vE=gvrq?qYT zXk7pGa4K_x?^QhY18S0Ak#8$S3q!|H*j*K9h73de z5Hn~bxX`PMi|Ck-6U9w=2bCZhYAWbYJ@x^Z`DX@nzt6%6uOfOaSxCdZK9W)DK18k# zhSSs-WZ${~+ht`?ThDv7fiv0LuTiXQY7x74vzRp|ig6~T5*&P93Rm7g7M{%zn^>~2 z1UEv@tTFeQc8?S~_VzBqB-+yyr01ZSY;;L*7&m^{3fHNBU?i1^!VXZ}lUzuSiuGsWnY ziYdfnqw&|!8{r=uAb-RSy9^fq`r_2?^#do`22;+j9B3`d$N3Srk@hweKQFn%bC4}Q zUk;;3hTEuqX#m|mJ(q6qIrxmXhID*NEt@v&IlG>6lQj%KhkaTnu)y{-_C)`Iu4O;! z=OjbpOm`s0^bq`b9bn@%iseNV$B%AnA)(9(55qYj4p@qjO~GQ$SCSp27s)#r@sF8Ea?HLyqSI z4_FsN+I0k(8V(|_wF78E*LIWzGR$?8#=XQ0rlqjNYM^!7pj|^y8G_vxr`v z^r8VlpKx$j1YQ`$pzXwd$XyZ>zNbH>X@@S;@sAEP>eoj4a$1GNSKSu^vvP&#=X(Vw zIWv~^z?pqoI*%1_e|cfM6scbhVx#!EuHtwyygBPFN1*`k&EMlQ40Yw>2QmY+7;tx|X z*wF^3Z3iM_wLp7(RVcWxA9p2ugDvm)574}W?T>0Poo6-sf`+2;-yAwO!Ha%3^gw*$ zNu15hhDnSQ)>&?)t6#PeJz79-KkdaMCnFdyv|@p}X=Ga+L6QYh-22Mk&vw*d<-y;~ zP&0&FoHmm_XV5%!x`+0r&uEfYrqg>bgE>6lXVvStJ55Ts=#x$nTAX2Otwv8?52Keh zO=zS+^r)}`-z)yIHVD5g|AM}jvDCVA&DhF7-$YIG(z6w|7HVHc%&Infb3@A%Y zlDZUgQO~(t-UISz!xC$1Hdm#=v!y9za{|muG--#FHhDzffK}Ojcua4GOvX7FPEVt2 zOTy{owsO3FWKXFDqi8kv^ZgzbMHZHNw0Ma(HoV@>8N3+?Ey%=#f@>65@P?|Eey3gU zwh7*%EyA#E-oh*MK|(^ZqVVS8ZOZGt$?{Vgn8(LAEYtrG%hm{C7yZN8@SQU$Wq>WU zsA9o2O1_9JCj$6C%kfCz_7G-9gcOk2+%Q zQ_3$Wre1kpn%>59tC8G?wM34>CWuj$+jP*mTz2wk4YT~3!Y*_wQ`A)*S~>k4x^oAS ztKkGHe8A7Qb32%lQzeVFd&V4|RUtj43M+a);ExvPLJprvM-w??=6n>!&AtLPg(zG; z@`Bb}&!g`f#YwqPoPtC68S3L<{7X?q(OPZfmUXicRu?d6L>%(A%5s*t4ITLAOa^>s z-)C?c;x$VswI0iaUiCDp`?-QLovy)Ef0K}rGetNv^&@Rk zFo#|8R!lJ$Al0r%hgVD?rSO@gmZ2aR9vmP9oT#NEH~8J|#Su(;UIrvgr$a@abkK4> zzZWq&&pEg$Nrn*5DT7VV7brC<(RmNfaapq(6>=Ha5p;xFJTH*T$woS2HrbttMv?VS4eHC%qYIWNad1f)UUjx% zPJuUU3ZmeKZxSoqwRB%B!-$FB0ZiQxH&U5T2943{K?PrHCslXhZxE9a)% z-G}m%0-3#aqVf_C3f4If@i`5c^`;KZm$NY|Hy;^KYOp4WLjzu`QKZ)dQeWePWc3t` zd|iNxFB9Nha0ah%1xrt6pv~nU#&#R>_h=j9ZfC;f5W448fHN<0vDUN*cXkq4?6s%$ z0oEjSsT>Mgk0H%ja2}2hJV&tvo$D-7=~qi~?_W^uVGZH8Y9-EeenY}QKGQo95B4Vv zj)&)<;+8F|nh?kor$n)Ekrn)hZ9+wS3|iOT5rR&)3IksC31S1pXt=g4t-2sbn}wtF zd0Q$aUnrxYPfYPD(iAU$4M60s3phIb2tQIJp;L)>$rq#{urmxT-c#thnn0v!Lw?z# zITwS`s{Q8F#l4;>b}HP3!>Dk!CVp(x!5+==nELMmFXTf>^Dex!pidas8teq*R` zdpl==dgN2q$eXlAN(qUR2Xjt=IVvuXhD3@2(#Q4(`?i|aPl=|vzfaMLBUxw*PKBp^ z8Qi7TlE}rAO2&p!wFQIBpINZk5{4z&I`k#oh%EN4qXj|xX{d7!=@nh4l9B`1JNXJ` z^A4K&oeG?gEXI)BD@f|!$*MY@vZP*lgs*#njV*uha_kt2)6U13Usv#IWfeAtOvA)w z>tO761#{-`?9iOA2>JRClQa)t^~6}tAdf@IUlnLb^+Oo%Fr_9YQB>ktn)aobI&X^8 z??5f$s3y7_+R5e#wQNb-LAKP5ceUMb-RN1^X3`n@8=Lvgy{~yRHT2VDPM6i#HM2qN(C3*HxojpypV~o63g_5*UGDyC ztY&tLhYGIShYCyn(-T}D>(aN&V`$&XEySwl3gxw4Lf(T=VcYp$7HQDL?q=~j`-vF5 z(n{u?|8p3X!{@XrxwNC`G?_Y0Lpjd`7Hb;gW3>|wx*mYvg_BT7d4={jyf+1D`k%_e z{?QZv|MTDfLtTOYssEGzH$ToNmH(NZ#Q*L!Bjn?V|J|Qge>MI8tV1z=+*FzW)oXkE zVWRy1_4{X*m?`{Uzy6>8`Ltt!(*K>0`C+-l|MK&S|4#g)iJ#t4 z_k~Fm+OA5GsiP_5!FgeYbD5wPeovTNe^gY^@JCd=L6$`Y=+Fnwc^|@OJ7t&B;pCGC zPbFt`@l4h38_LxEcL=>yGk{~C0cLG^&T10kaQi|!ijE#d-|HVZ&->(Udi}`Z^LNCr zk*75~r;;W6gYeU`WNa%zyYd!eS^ONV{ppLeoj>8}GlmAlnb4Z4h16i0&Dqn(NNwjR zim-4Xzs6v?6Y>NrOIn~i@I5Lz_M=cM9;()nSaVwi#m*x5?i_$eTFZpEeH(GXt8u6X@L& zk7v^N;IC8!zZ)wMq?tntqw?u!;6dsPGeF80J(z1ZF?oJg+wk)|tn=&fVp=xwK_6KO6A#Oh~OBf~&kKZ^V4M{CX%od!7rOofUAs@)?okM$la{ z648SvprPUv>0L>o9ox#tiUq)qMZj+C9IUkUq^HxyQ$PnJo0Yn3^M@g9d7>lpiSJ|& z{qM4~&F@&r>q;S1tX2p=^jtXJdH{iGhYZ)CFJkSb~%lgsC4WF=6_aC-e4y5YeeVD5E2R~N!;KQ!xaO6Ay zb{W(8 z>$OI=AX`8700 zzhf108gR_E$IAbnV8O^|IQ+dA`BfWm{vPl6eTsm5*k$H7{}&r$*TFJRu7+{AFQ41Q zqm%c%&3|{I_1a71#ob^^;z!ukl4Q2~b}ZQ(-$mO!VkqU`NIE-3gVO$j9Bs1L$L@2? zWy2lzOVtmJqc-zA3C~ab<&G37?g>a9LXkHLAvd10PsNAPKCuVL;9XmtbDXKe=Q780 zEJ=NaBaL}>3;X*$!;t<@FobET00eg+9_8*@Ol{isHjWU8Fwcuqw2vUU#f^nOokYuDR z{JTGwu5fnuOWujxa>!1I)|)Q4)Y}UU(-oQG9S?Tu_#t+ws1)gj=OC4pik08iP}-@D zw9(mz{@p0X{j_fAR!LLk12x#bNc?h@uM%f{Uw)>s-~fxEY?vCZELN%_8LvJFD`;b@vYDU6mZTua$TDOfor z8+*KSV59vHDr?{I#Zo<_Dptas-%;-@G{A>WduZjkq1t;e9!{GmEIDf=81Ld-6J0qB zU#^4otNUZ%y>rlfnvahMkHPzsE~aSd;QoAftk>oq3>h2JT&PLA>vtkvDTI4gc+Md{ z5*3x6*jKH`?`^+vFZ3-e#+SgU@);{_?P9VYhT{FeqtF|b%6XWFkx*X3B)rp^*Y7H( z&a>*yx=~o@vmY|^{<1kEf3d)#^DN)QjF$HdCpV=gm~GvIsWI_zDT#qhp)&VO>Eo!I z308$3q?bFRNm3$<7L>+e;jRNPT44g8;4iG`{s(5cTLD=Tc_Pp6pG8NuO<=D~U7D=w_pn0#`L_QRf_X#d;@y7}aJ@z!CZQ78EtLr8jCB~cHjuL_ zeXwDV1G-GSVRd{9z0TQ5Vl$W1rM!v4`QuCoXo1kJv7CONSU_8yOz4T43)bA44EeI{ zs2x$sWZDzh7o8-gLzX;;>qy^!I#WUdkKPYE!744Tv6@>$D7i(3dj9sK+8#4#9b-`K zZwBeF-Mml!4d#3{H~m)%Rka`DyVh`;QL`RX0z+Z?bv|PG?%<-_4rm@*fm{DHu;GV1 zHu?@jv))duJnn%L%Q=I4;yUCtY{2Di6SNLh!K3;8uy}nP`~8Jy5}PvN@%|dr2lX(^ zn9uA0cMzsd5V7y~tk}_P1(tht9En=BNv&`I9oeKx?k}aOWRNWFxE{;y%0#ls+YYdf zS*>tudV*%hXV??;4$BR{;6y|Y$~%*B@>4v2M!&+T=c#B*;I6)RJF)9pGmh4lL#w+I zX>-P)ZTnL68iqh)RVcnkMPuh^U33TRf*0?uHSXSmQl4AaRbEK1syC8;+gDUPe+6YX zIf}O8p3C2_M7Lb-iQb-fAYbiev|`b0!k#gRjFLr;S%2J-eoey{egTOkUedf%|??P}NwxEuDzt8_W?ZSmLOKJN~kHaEvsfM>Pht`IZ62icMtS z=ZCO8mPgo1(L%JFu10LoH2hWBi`!#tF!jILSjM{vlH;7=DklyJu?TFL8HHx;7%W#$ zAurhsn&BBo!b}CuaaN=!`Mn5q|HgJcYhr4{E;5HZ+>c^WO}U&m+;&Kd!p%)+d?|Ni zy*Mtghlhl-OQVJQ*a}hPy)&Z3XBR{d-&xX4o}v8)&b0Zl07Lp_!NlY%>whZ^rCYOL zEE@tb3mvp37@?~~3OCcu$Sh$rsYFj9n;o|hv8xO!bpTK`-1yB{@q8o zVJjvryamZwVR#?AlJBgYVftkya>AVOR?-M(l-CK9jg||kQLe(=sc~!qXEelL`_2w& z{zTJ^HwZLug>`{Ho-CS%`KuRT#0AcEjI*Y*PnS~g$as|Vx&M^LeeijtN?u_Kv^rFZ z2Ap-EN1m(bQ;Ud#4(TE?a4Lo?*0F2jZsU=}Lo6Bf7|-`}Z?)TY=vxUWoyvQLMFYvF zj&mZlW^yj^6jYy?gW)5+XdUyUzoTc9T37`%);Hm5PCxou?ub{4JK-7>jFiW>DMPx9 zhA3z8XO=H~)VAVk)DnE3REJSAMG%~`p{^Rl3|`c;(+Rto!+JwZF114Tq*;)PxhV8| zcJcr8(g@3zCLo)4BClA9w>NjyO>3-eAPUyC>ah!*j()5q@`|ykP zjH+f|l=`ly)XdP>P`}IeT-Tp=L8)t2QEL**Wa5{mat&-)B4go7tFM zS7B#y9j5~BqoaKl8olE%>&_k=U%wBpw{b>=$7-xScN;7Jb>O}GOMJ3B$TM=O_}CSV z?ae08e?ASVzqAlNK9CYVc+t0a3rOPpXiC02obEgtM0YA5vXR#^Sb)n(<~HgPrc8Q^ z`lfo!QCHxM)R813!FP^d3z2g_3o34Na7%v}{bJfQQ9+JcR1E20oC4KIb01I9U$|Ck zk;!;acUS}TLSMi$|0X`I8%pkzhEeKxS?Zh=K#gAyQeQ?q{j^h~sx}ouZhv|=ZXI`Y zPiMoY>a%O!>u6Q^atbT=r4OxVSRiMx&8~mWJdTeN-10SqNp=Q8BKJI$=?|x+%Q!z& z*+kI)XDIA-HxR~sUV%<&&V+dD4ZWpdcpOrUqdZG1D!xee)>VA}b(McVgkW&$NzUs% zhS_gwp;rGIlWnW37O~*3eYLtOpInqzQ`L^`s;8A2>~L<_^W)^1-yePKtuf(}WT8 za)tAEt_v4bYD8Zj-W82nF2P3E?4uW2F;sl^0I55iWXq?QF!NuLOgg0&Umo%MT%-bL zKCQ<1Y%93&XH;FYI%OGXQ}kO+^3}VEgj^<_z$jQJmolA&l$fUEMd3S7e8pU><=1%%ze5)(O>lxBC;;jOi ztmb<{o0SOP;fkegQFzSVsWB}h;h#1J>4j?r`H~Gn^O8Nn@Zb;Z$CEZzWc`Kj3AvBp z(gwuJtmRpvFobXp^N`7}@W*B)xs>?Q3@a}R8*m0$Q_A7Ewi>$cC8;BaKN}A}M$Srk znrx{`EuZCRS49$@Ri>eF%{hF0^9(L>FEB)*0YC31V1`K^if0EwvSunhEV7`6%QI<+ z^*(IpbL{;}iC9tTK&d+x(=~||WRzdTS-aT?ka&ytZd>5>I38YByi>aCJ2~GH6V7hZ z7TkAAV!nkc{)#2!a_DT+@$Md}mQVUSWZ5;qh;Z>ljAM{s|(=WkXu za2@rJPUMb-`*cE5!0mUOBWtr4GvsvXizj!sJhi6s1=jefoP@H<9`w#h;I#anow zAW3<9W8lJb`&%NmL9%f_)Kz>r*KIvsujzz8@8n1P&?fJs{}64v4_ykO$mTwk??d0Q zrQ_GLN6VtgD=US<5)vui;W2y5GRzE?N<<~D_ z=QEwJVdDdWGK|nTTOrUgJ>)NnoJ!&TNpHc zo$$=?u;50V7s|O^68uNjvS4)nyM&SQEeKe3jl8%U^r~3{8BUVNiwuD6wCCQ8CX^Re zVZp~sI54sv341t0d1oWyD>T>_%`xo9zwyj*`7EaBJCc2#V$Ieb-G;djeGoT!9!}W3 zXO4RB*}`SdS@lQ>cBxL9cN~VZJ0<7QY?=>e<73$AdyCDKzQ|sWyvr*3ucm&%?sP?N z2dy~#%k1#9LbK6RQ_Tv}Ua_kd66js2gR6soi_A?Pi^@OMh)Qmzvw7dr*)?Tv7T!CY z6(^jS;x7=lOcHdx0cIoVQ?+12XxWdR)4n^+QnaMOx7I5q#+t*Xe zc7H8nW^rrTeURq4BaqG*lAHqor%8$@Y8 z+S$rs`OJe7nWB<5`|P96oPRE2dKOREjjSik{nbrY#Q8<{5_d2fzM47Y^kbgKRoSz4 zVs?kh*;v<0?81^$?DZmn*{(BUFZaF{&Dqewbc!FciMwC2=Ct8Vx5k*w7-q_>y1H4@ zzd_KNV2XJv4lL!m3Hv6Pv4wd***u9(7QXHq+y4DD3o(gjNi_%9txAD)d1cqpD$*X6Nm^$F~(?-4fjjVjqGUfedwCm8L&`LUH56X0&eE0IkwVqLDK< ziHhd?i7Zu)iPAJvMHUYOMSIZ39$NOWeR)mnc*Hn1yjO$mdO_@}lQg1997C_SFxNXT zM6=r8iV8NhiRSLmLBn4?yfi0#y|jy;&wcRnrZ|`Rn~8ZHWn+C~n8<0X$mQE{QPqTg?2h9h z=I0l~E?vrDPTnWkBFQ7nb3`gj_LgV4Dy^cY6V8bAzfOh9%W-fY0R$AMu^V}3na7vA zEIFu2)V=k*=-THz(bW5n%qo5%tBBpsDpM{oiOVI-W5fwIH{Y5)RoKgpZNAD}SJg2E z|4Mdm`vVsIV;H<{^g~;t7%;Y)>FqnpTzdC1cL#kIVXDfkGhNu!;h$K@;arv#d4e^( zbbyAtD=g+)!0r1%wzDOTNh>{L+gH?@jSv?TNv;2C79CG$eW!u#ZL-)fV2NmeYrM#I z@=}prkp$gt{0MW8Mi>NivEe@5>~n+?yv~c$e>?v|>;?C@=8Lg~mmZ5oN8c4?yJfP) z^=H_MubIqBJrtogI2UHt4m=epvnyJ%?7u%^%ye@*lXfm-fBPR{7j_+E5|GK_81NyU-Q^Htk z(E%p;QbSlIs0bhKs0iNk7SKA8E$RN_cO0E`;qUGYVP@DJ;pe#+k-h0R(XpB{qNZ;K zyf>{+9=*ZT-E)eaHF?4U^jlbw-(D;|bOa-x1o54%Eq0A?!+}wH=ql^R>4bsgtk#Zv zO(|M;QkqPRrD&`402))LPAq*nHT=iV$@gn;_3Hyf{yqwQhy50{*4)@I?&?Q1UueEw@7&b`w$tN5hT!4>WtNqds%*W9{Z|$ zjD-}s3h}KDf{)}1AtG!e`?EQMJuS^=n>XEJUETfB@AGgBoUjIZUJ+=pzlhgdeOi$A3lWDa4T@UqSd9ddBQLM%^lpT{!McJcMXrFTx zl{1r=nDHsL;88K_SfU9{(O_KLFbz{?-VkaY-4pt|{ubmMrsJlk4;HNt$G?b3+Rzq8 z*UZk)S?e^AS0>MS?}vxOE@;jR#=(`Y7~Ho2_YZqxD`!;umsP{zTp8Z$UPt>2Lz0Wo zq+qX^RIv5}o|qLNcJv$g^_fH3ScL0W=E3?>JQWX&BKvbOR5BAnr5_VSWebEBizS@9 z>_qER7SSBjz0BZh9%sbAV}T3%qj~33wrAo?=FhnTh2wu=^!qNX6R*OohRf(ls6}`8 zU>d$$o!mB#qj~)9`_QC{1u56DQGP14X@eoTU-qHsc|nw}u%CidGYDcG@b{3TgL-x7 z+0VPVyN|>A$_dQ-5MRDmx%vLXjZr^pzX|}}5NmJ3g>p3&$ym(bw%jeWF^eto(xyB3R z&F_ZKLdw{*Yd2Wke>qI)a1oyLrr?Wm3EDjK5ifrko39k$$g*OLC~biMG#9$@aR|x9 ztB^_E7f6l%4RyVzd`8!cQ1cE<5lPa?E-AXhGZ4CJ=dj)5CUSrEV(97j*b|VCJ_@GY z&U;C$=`gup_`%r;FVW0iXG8F7@kESuR45>uZk^C2vAU4fha6G<7P}tnf-0QwE&x502wV)n)%5M?a z)`n%;<#bi%9JOsYO_@JVpgblPy`9_eHuene$<^|{{z0C(`iY$;9dIQ{iuD$wt@oyq z>Z(oTzGHbVFM z`NG9Qd!c0BPqe5?k*eWSzKbaoG;S3Nqn6$jo@Yk0R~wvJ%!z0=DA9s0C(fg#antCg zYbi^!Xk$BG7P5K2htcKt6A5zmRBm>TSzC3m!yoUm$5Rxjk9$u)SGpTJ`8dC80e;8~%ltkOFlGV*d zidxu0P20a>RZlIx{d$NFawUU5^QgCNC0%_Ii9^HF@!Glo#y0-UZQpHXn84qcDh>!2 zwZet_iTeb5X2f&f!;rqEKPG1TL;1@#So43=nzkbTcQnL;Z$q|n1zE%_rl2@4?#Q_X z!;P(Yl=~ElHrbe#mkj02oO6{ph!STFrpq!46fr*xH{3%p*E0gseu&XCc`+*KbiH*nZ~;YVz2iJoS$(5 zgDxk+ehhaFOgMt6UEk@+l3p4#L|L%-=!eM9{&=gf1C9A-acfKom_-f_1{`5CRvlq^ zcdA*szXB}&%Hh`sF_fHmFT6S0Ayj+x3GxRLAYPgW{Vwjk>B^)@lKGU#Gi?r0wQ%Xz zfl1Qe@z*3C1`D!~cjY{0{CR>$dM6=kk_Fv3B}!`=LRHNQRMpan)U)(4l^=XfLFQqCOvN6d``R%f{dWcBNYzr%?0d9+ z$a>g5+ll<&sW_%Q3a8^GaeF7<1MykIHuqBQ%KS_YoFihSI~jj#-SB#$4O#rLAdMRS zIedHs`4diK^!-ctQ65d1YnD^cq%C9;?#8)2ysulELdH%1FnarM42bN(i>&t;s`rNP z34gkE<|j>^VqE|DRoX8#*EyPq(+4kX@WI z+0E^N)TDv5{GTSZeEJ5%b$<}W84jDvPvFMzyEyIIg3?9x$k^Tjop052{d597E{>+v z(uVZHZXCICe^cUIKlUPW4fn{qG5HC;^m~~FWv=IJaqe*Z?iG$}@!@!-Z7WRV3=R1f zCd}*P?#ErBWcy<`84p`1w9Z{GjI~M-cAg2uuFqEJ$G>*#eTGqp%numPktdCqV$P?@ zrPVG4wB3cfU!zALa}%&;L?+&d(qLhF8mB$p;!o)d9DjWQJD=%uKLF>Wo+Dhib`YM^ zm&5VE8gx83PT#!K$S^jJzP{3=WBWB|#?670cG(ZHa&u8_VT;0=0fNl9Zu)qlnF1d( za?jA908MRbeOo1bOf3=iCfyYV$e6OjMqQ$;;o>a&z7?e(Go`$%=Ja`V9ZNpN?*oJQ zJ<_fR{yXbo)O-uxwq9_YwG?k&kA=xUWeQo_pF12sV3R@^?W(b(cea+K&wF5*dPAst zwIAs$RigP`;uO{U7f#E2kvH->wk^7day2P3p3#6eoWn2p$_l&pat4;-XIj{#Ln`~< z;qi+WWDhW-p;r2olQN9R!3mRnH{z`QMjW?{!?=BG;J9WfbUIE6@qtOg$no()S8ouL zo1V@R_NB6k=W{V=VHB42_~J#l2AcU{K2B*8{JQjL!Ub*8c9o(J0h^KOnhL27&Q^07 zgC{{o_&cGQ%|8AJ%IDhfhtHhGXr-ZYfG?_wmO!O*CjEF|LlUMo)O<=A7MT$dsaTtc;b7OEP+G{`^O{#R^&|tbB zkcrj-VX)`t{VCtSki@=gG~&Npl1N%7*e$XZ)QXo2^~37ukKr8}oO6dXJijo5a~ zYLI)Zhm^7qGN@ieS`&RK=1m?=`CLQ|o6E@ANuN#`@O}PJHTs})4L>dT{`piE>XNh} z^HiH>8hcq*QX!7I*W;&b4=h*NVRqmlJh+>L(;*jG*Vkfp#poc@61$F^#bVZP}y{NNqNvE4~@;>SKxOAeydK9ebS=|r00Kasx7ltZ$`S2k-} z2b=A`2lb*wm@&>3A5+`0tF{3H>|dggdvdGiS0S{8`|!SvqJbmFQnj`!^-0@M*&S)l z+~zrOZK42ORWcr`NSSkOVQ5dtf85AgwvVU4Jp;KiQp z(%b*y;KC-6_~e74m+cosj-FHK`N6SNH+&pD`>uqic07A;ppB6iR|y+C<_d!zS_xB6 zMNmUp5bsm%re%LbLRhS!V3#~m7&6iXN>v8iy=k0bffjGv5#vsReS zGb{2(kCIh(D%syVLgzP+B6VFA(paWXO_lqYymSE@`|AoTdpkqe5ib%}+ZYRzUO1Bb zgIN?~3wXuSBfc{xaWTP1f22Ucao)?!SP9sl7=YKn!X-52TM|##ZOpvi_yHLc~pcU`x&w=w)-lydEgeL<>;KN20oIk%%nElR0a13%4Ob!_1 zjr43JsE)^%>O5ThQ4Fg&wHUBJ5w_2d;cvzxSaOeq&87e7_T)`;9Y;m_{CvG3} zM(?#tlpKGPT1(n!U27D!2ZSKA$_Ljj^k8*x9X^lziAh)N+4${?*}mVqnNv$R7X0ww z&)qSY+g>H;)-(%U=X-=1I($!Fx*1RJ2V(x5bTT@TMe#E(Q1emVlh??`+=yV@`jds5 zsu^(h+Jgqp4;2>YprkSsrK{gTGo}<7jptCuGkRYfW$EZXWwO1Vi*e1xhzR}$X$9{4 z)*s5fR5PG=x{!Px7L#gI30)sxCp5pb7rX=~Atc|G?vJsj$$M<6Q+q#byQ8r0%vSj4 z`a%4SfR1Tuuo-Vixhaf}Og5y(Qze*o>LTw`*WziAF@5M6M?O8XDWXpscSjF{yPX^? z{hX-Mat7}gZKbwZVN`iCmcHmzP_ezb6?NOVcT#4E6qkh_2^V|uIfjoVt-(Hc7m$WTgJcwvBK1t<9a9|x?u$sm>-->8jPXOpoIS9uiN+wQlW<<3C-_?r z6^?l~QEd)qL0X!Th7xz9S2(~#VK%}BxuQ2tBrN!0D4ed)66{x>q+8}WH2?ZR8e-XR zA`Xa8>b<;T46ad#RVuD%?);=lX^NSiB$!tO9xq(7y|86?Cv4VDoui>*Jo`roFPEu0JH2$_E z-Q}6gGoHT@Gyg04hf30mM>Uwc{xoKBH|Uj~vqFw+k?>?eo^Uo|FAE-ameqVKWr;8Q zk^WmrI^(EHKZC_cto}SyvL-?-_cyd7JMn?{Dvn)l#pMgVI4J%MUk;RDPsdqI`M4J; ze|FQ5Hydb*jV(zkmZ8xk4RsNzxY*`Oi~Hu%zBx{0zGDX!rOu^olRsgo&Tfjg2&4h? z_mRtqO?du01Ye5R;`J9dvhT2_R$cDbJo+4#C;p)Anlg39Iw4up6LVviUfcm0S>lptyKk9@B5820;uS=0U;W<382O9?4pv4y2yhqX!;B&>45(z<&?ufNc{k9n?bq$#4In({KQz&muArju+#4)j-*i*WLZOz@mWZigE zYyM2C&|g63%Vtx{78Tl~IDjM@?<1+}3(HpLtd6EpxLa8O(dDZc``<%EiO(db1=`fB zJb>pwI#B!mJ`PD{;NYYy^nPO!mAqL=hD*27+ObQhJ#Yp^SsTE=L>`-4q_MG`?{P0B zAyOzn(XEL*|Eo_2Dh;SKbvyiK#6tPg2RuLKNJ^tEC^eF^w6`RZBF`LlOo%4mKoLzI zKc7Ck&ZMz{;dr!oDZ1?(QJ}$@0hv=M+FnGBw*skT$tKDZ9ipYi{i$bgf6|N@LoXtW zMCxWztj$K76@;#(7kZwwK_-x@WbU$Q@-LaJ!9Zv?%oEBJJcPje?!rWW9V%^`K%w(} zNbCUb-p)QOEV9`m7`Pr|hQ80(^e6SK{v?t6WasSi=B0&IZN%Q$aa@FtLJ-(yY0)6q-cUe z4Z~PkiauMDZO_CmG7irCK7LAiH;O%jVYM^} zT7Lbo#Css3l2xEEbPkk#2-o)Sgr|`|m4t?p)6IqSI@$|zYke^=APj{IgE3_b_qa4B z!#b;%Nqfs9Zq7IM_3{v0N*s0_yS{CgB0h!o=5lVLd>c=iLClp zOg{D-JHZqozM+)w=`SPp8h0#ypNQED-Qe3Cj%j`$`JDD0^UWTBJAU5en&3`@H(1iX zt&5R#Z5v9;Z+!%9KV zVT&L&cLf5hZDG5TGgwBM5)O_b3)>-N#-tExr-C7FpO{Zo1@n0RjJdnLXLhoVNPOyv zYezz{uW<)b60_j16^j!=`(fHI4mt+@STZ3L4!omDm)(&SR>sQ8PqH0xzUg{PSP1|>}2!FL+` z9{jRSmh(O2P;q1!G{>8;eXkvu#Yb%>*)kiJ3msAOcr8k%+M{KVC+1I?g3i3R?C+vF zwnnLy-M{b^?%Rh@)1Pk4uW3b&&K0a2oeM*;)9m5kBP=XtJ9Ay{g)~!NEL*u2JI%6j zU}zaE7QMiH`Q>=d?gD(5aKj#LKgJSom}wj;lrR?nDxX@a$aD_fU)r<(zTv zsW3k=1kq21A^H7mM0EP$!nY6v-~Nm0Bln=Q<2fQE(@}dm36GTj5BA%SXqW9cQP?hx+qBqiHlkt{) zy4D}9zm$-uQa}?~8U@rB)B1hI@O)l{OIp!LQ7NIXW>@IJ_iVbLev;LGNoCJH64?G; z?!Pfv2K8wna6Y3;ew&QQ=FVvPWY*5~PqeUMYJb_#N0OK)8iXJ5Ly+D(1+V!G@~!?* ztZPWXq34;{x4#|N7VJcfXCQw4&g9OgdYZMZigG7kp?8;+X~U%9r1jUD+BRQcf>$on znwiCR%2^9jCk>vyE z+Lyth=o;cW4#F^&pDET&LFZ=)5-Z{?s3ltD#OK!)npe=eIvGE$TaX$04=ZlnbuFBGqo9#<=GhesvLjs?}vXvAhoIo)7JzC>UidinYVr57vhJY(@tbL z#)Xt69BJ2~;k12;HhHL=1+R~jhwcCVzMNE2QI$QnWm7sI9O~~2%Ly)-5`D{yP@XxAQxN-3UEgMux zTJcHr^UzFO`Rsww=^JsY;t*1lg0Xd`Hw?K0puIK=ceW;B+%ieh-YZUHJo=yyO;TB{ zPmixW#PIFvw7Ei_{^gp`@HciiH4r#(XCriOKBO@nALzHoTT-0J`_6Nh31+1Sg&(!+ zY4iL&w8Pq$79Jab%E8kyV3i+M{4Hei5AxV@T3?z*nomcO1V?4e zP$v4(dqCvzU7O7IRPT zV!bm|u?or9u3w9dHJo##t3V3}=#bH>KbSLZ4=h42qPx`%?tI1_oO*~3Cx}RVs2$lI zA5Q_hHOPLA3Uou%@n;oxWk&{q4cdpqbKY=BlcAtA8l;S-?p?eG4y}g6&SSgQ|U*Z&< zr%tM$hEjjoTeRYDJ?$#KL)lpsa9CZ5Zu@+Ej{1z~TyYA0-45qN>qxr5icF@IBcLr5 zo|6NS8b1M>Ms>hK>J_YHBuM(5ixAMWNa%cGBM1ik+_qGfUUvS&;G%x0pYV;P-j~6+ zB)-Ag@q(PK-_hkknnGNFnlRy`gkak$X|6I|+I-#3Kf<|}jG=HkYs3*YX6h!GDlOw# zXWn1;`H3s`KX6OxEy{Slro4F|Sq$sJ(&m5AmEh0cKR#oqyeTaz)**xA>h$<&2rcN~ z?vm$zlyW7WW_U!SJmmtM`fBg_z8B0J;R}^1sHR$8QaQkqj7H)s#Xpa zG!t}$%d&Dp+o~X%zc!ra$@$RA*JiZk>?nGD&V;I-#d3}_cgJ={p(!tj9V@bA#)T1V z%cyk0vh#v)u_#SA16`KuvWay~v|>|TW6*K^KX?_UqJD-RMTL%|&XxUW+~rCPip#;6 z-?6ygAx3lgES5gisHrI=#Rj=K}ij>n!Qbiy^a58}Mw!E;#J*$Ls^e^rWns zZl&f>=ZSh6=1@altS-{hyPLS{&xfCjH&bK|(DF)zGkne%+NnVM4#?2p^F3&Pe-C@L zn_xAz7k_*d@PKDtv(r17jdT{z(xy_;<^cMx%KhCx3;6u93>y4dIq%prOrP)=`zx~9 zpr*&{dafd54E~7*48ARTGw_CJkLOe2$)1K?OnhQR#u-^^ItHUi!Zn2bY%W&}k)5$D>j7Y|tr^IuTFq zUjC#zW)3O$I#7WVqu@8^Fm=E&taLd6FNeW`#$hAjaI~>-$!NV$k+oRRnC>8S9`z${ zo^z0wNhGtcQ(0TcB&N}{kEMD4$L?7j<$}RA%oIB)BR-H`yBs1bpJNoZDVE0Eh^Lc^ zK{W1u6e-LLApe(sNaswuQu!m;!=k7OQ52k?LSHw(qLss2DR|vKa(7K54Y@q}%6El( zCoherEHe>tX);N))ltc>Yt$GNN8!0nwBhk& z${uM-IaQl5y2b*a|*rwLw7$rZ8CbvoZY0#nq zvK@4Wl-^yUlFdG-*K@=4J0^Hspvd}yKZ&}>8!&NA8LBANAX`3?xTK3XAb_N zSif$Xw{j7hDmI{^YdWqJPb9{}8ryEz5LZaype%Qdzt=%1=X9pTt)-j0U1`FP!?aj4 zh;k(kQkaVeRqxzG`?p@C>Gu*S-RdwcJsw0k3S;T1loqY#nWWbu&PbP-Pv=w?(bsiK zLiGSe!Sk4gFzAFCzh<^jOeXJGHt!^(iMxpm!$@P4ErOsDRp{Yh73wfoqg+QLl9n4t@*Db5ufkahJetMNMR7Fa$pA{5)d|mOe4cbpoJLD& zlX>TGYE>CcK946*o%C<$ZoN$7=H8_(Bfj&`+6grE_GoGkx2A2ydGto*1lG@Bn)>Z2`p14k3GemX8$VNUyf;f2X|q&tyKF*hZ3UV;9{jz*cNp9~q^{YY zl0#cruSqStwb2P<;~zt5Ycq26dZC$NXzt}Y+`P_dyt(1brL zwG`S$&JbGaXA5!mepsm=iOyzz-*}?}!#oXi9&2Q8OA<-GF^@BY>d1S7Dw*2q)8Dm2 zXQ2k0=LL(36C%73c0(LN%Dgt z**wyxfJbBLyb*uRoQdYl<2z?Zo?ZG|k8FpnbaC){YCIN057$f51KYtgRc1Ja^`)R+ z!&T@;@4}i&B^q~6k2X}#C)p$eO4b`ebtNV|hg-_7cidr9{#|F=nYM88y@K%Kg}gB7 z=S=R1m`oP?l&II*1O@+%LSF!Pc5y$4BSzDz;AwQW$C}S!oM~SI?{tX{(ehuhq~j7! zjv1Q?g)Y>$JeUkuZ>Q~aoVw#v`E}<&xB1@rxBYtRlJCRrz7Bj>{f?zhqp8Ma25qz8 zyhraZ^l9@;s?TeoS$%8h%X=q^uRTC>LJu+6d}VsSirKgmv9!3{q$lStz+>_a zOk)@DwZE@WWfv~Q-ajH_*!NPayM!=*n3`~RY>JR&nkM86=|Xy&H?oTk;=|o2)aR&@;=^LrPzZ_ZIYJuI}8VpzW zrCVEq$=27MbaNk4?Z;D;+2lr(4Lzve4(_LOo=0zZ?%8~JJ#%TVVD{5PXs(I_x%>ph z=MNy6_+hjn-kg5yJB#ASvDjb}gNgq%gxkxyX@Ex~1)U#@TeF8^M9KvG3N|C>H{!IS zF&fW}|9A8tMPd+9KG8vd>`E^Nd5{&*GNBX-x`g(C#8h) z8J)EC;5nL_{{x}JI-#sw4W-I9I4B7@)<(EZxq%#Ef6A8?S$-@ox0t~hJZ1lMtG}QX@{EH4bf6}Fpm%MK`uLT=?AL4GvEo>d! zih{U%u#dWhp2eKcrZ$jwSU(m)8{K_LCO+ZI+sy`!);b!b&Son zKFh=_Qb=T;O2=BbH)3KKxoz>Gfy$n=rC}9!dbv?qga;i7lcrYde^?j(5anqH$t1#; zUN5jCiMJJ$I=qEG+-j%0d=9?gNI#P4mZlN=-ZABPF&x|6$d=|LlkB`u8rJ_9IUBrS zuVfq8Rnaq6C%sKb-?LL#ZM(#t(q&#}99D9xxDO?MYfB)haQYV}({ zRoj2!dG;Nq5EsK<8Aq{3Z+}!&EWq*qM&p?B61vi`g^mtdN(;0TVCjDetpPCz81a_1 zZM)A-$bMnf{?{mKd=l-fiXrdC{b&eh-aAMX;%Vv;;bMA#aJAc4@XZMjm43b?dfc~7 zBwuMx6$@vO1ZU;N^s`|j!ra)!=OIiXA{@1Itr2NK$orMSqPMMO7YqED=BJNP9QFd= zR=vX2of#W#jibk`30%+H)^FCO59L<#zjO2tIpv2_3L3FTHlNFIoRX~BM6x{!i3 z*UsU3<3j`n?i6xgxCl1N^Z2}d3yYHoWUFe9u(ic)nEJ00&D?EY(Q3jDW~^d+)Mv0a z3JnMqzYnboHQ4{~DJySL;9M9TtUt7g-42+-PTN^9Bj2lt9e53qgWF(sqLc~E-Sa0Gm(reS7eHhfR^GL_NX zK~&Stc8q<(Hu^td@3Q3)F4#l-`W|E(ZAE!ry0F(QNjR1FpWqQ*hUU7v`2FE3GOe~y zK>k7s^0FlBy>Xb~%N-F`o8ivgutshtFnvKdmPPuYqa+kAMjq%h9!RU-Yf+)pcMR*Q z#WcG+csTMh+LRun@#_t&FD^p(h%XGB=Vo^}pZzHMqGq=k zPTf(0(@I@Bm_3+gne`_h)etuO`+w}$jY1}KM2`loF{B#xkrcTujig=A(d}pH6vO>v zH&$tp{1bUvX&HopuMgr(nh)ANPcntpbu3$MDO2!EKr?5#&hITln$I!}Qagg1{LhTe z6Rdpi5hm_*h)p-&1pDaO7%g1te)>)k4?>EUKm5>ZC#ku=d z2w0oUui*&9&xk^-t|r|HeTS669hf%xE9mWIbPXi~@=6?Lm3| zYd8f3(e-T(H2BJRl6sN>iN15#t;*dqeq%*JHohX8X*)&o&AaGpnI*NPxsk8*8K(E~ zE;ETOVh{8Th0$h`LR;-$8n9+Pd1uU~Gof14!_Q&!7bVi`eOa{E)}JkPIKhI-jx!h8 zI}psU{rvIH4^}sZtc}Jlz#qW;r`lu0$TGuCfQa*^KmJX&l z--pqJ>r==id@^{-55XRan7TL`!%x&eh^$3es}e3BHiqrGcGh^}Ii`EO#^|gDOr7)s zGw;e%1JAbwPm-jEn#Qyu)R8|w@|pl14aMK9Sa0(gc91j8 zH;h;<+_K&#JYBz6IIt>-eUsnAeqJqSDz%-gtV#=El3G}E-)1;CJ+cV1lR9w5 zpcj89w87&kgn(M0s!QX-p?zu=rA#=G7T zSU$cG`uyjM{k0OBj@w{m>khG~e5#BprljxG1|F`@=b7XY~PuvjnQ2l|e6M zPLs78&%T_wkN3eb*gtO%e$Lv8<|A{_5gm#K`~67d9e@EjVkBwb2jhs}$m%c0dCk&P z@1Q^uoU>|ZD@O~L4T zNh0anC#cRt#59gMvz~ZoHhsT43k_b*UhJ94W^6GfMSf@3({ZHe${o18CJGu0&-0v| z0WH{SN@dm}`f6bYX=xGE9!`Pg-BI*a)tb(Zw4tD>yQt;#cB(c=q>%hNys`U(nfog7 z{^>zzGzURs?1^i|-`T(koQW%#V9S5}I(vT|OS`1#^c~KF1CO^Shi#d?{Ws30ppRBklgyv{OP)ubsJyJ3tsb#}R zLRp<^4DYbtC)?PQ!tX4`;xzJhox;`F^Efah56eR8p)S>h)!)x!ZYSrF@QljJ+rwy3 zqc*t@QKS^PT8w!33}%{tFt@7%C;n5Tu#p33^7e3Kz1RqwRW>-pJ1NQKt=Q0;gS169 zbgE$ z*$X8>+dxw|?xHPhDzv8UVRNbIY%<LW}!UY8TC?H8RslGszX}YL6kXAQjbhotWy`gF8(v&=Vg) zerc1b(YP0*u1%oa@pI@Pn@2?w_OSVixCCq;cJ42q#>-=zv5ky^{$dFYPZF9X6OUBXy`*Y$5M}8^P8uhG`$yWs5RR*ckma z%sNPe`v5uTFTet$o;C|}#yk`1iU!EL9QSZqyYvf8H(p$yqPlu3RSM>D?lwuhnE~(?qUk zF0lfg`z-wCOLo840s~7`&@yZo=h{pon`5@r|L+ubY}DYb5l1? z@qWwgxv$yh!r#p6Ni&-!qe8BJL+Eg%DSdj9iJUQCF(%u9v8$LB3aP*l__Qv9}v`J9-*9&U&uk6kB8rF;&>u9;7xA7;ut$!fq`ZsL5n~i}IX&BTm6D6EkwZx$VCMxF; z)3BS0DlO>q%T+Y*=wsMglw+_&8?5>Fxo=?_qRZ?cRn?E4noH8T;5JOMd`#~Cr|Cbf zBq|v)jGV+}sB~5!65I{gb*9ILuel~_dRajU-ANR4y^M5XHBh16|XL0fK3jznI%Ftb_j(G7(mK~{~_GV2$L=Ju_DU=E7v5! zyUG_m*Ej`Q{5TD1XaJDp#+I3Kna|;IBT5o84vQhs zbqvoKf59EILX;)(=Y)!2TvYJHtQJokIHJ$p9q%zkuL^Q&7gKlIDB2w?M{=A2^6lba zak;627UN{JsLWh}|Gs!WqG%h>Wye7j{2v@+IWuXHIq%;Z;dNlR@bUdxK`nZ|5Zs%? z^HtF-qs)z6GyDzDIZ1FGxD^F&3?R2Fn9Ew!=%GvRIxM;0Y&yDc9E8{Q zXeO55z&6Vku$`G_kg??s7L2`vee*|RvG#DxA`KYtv!MMSXH%ZsEE+F=5O+C$?ZBa( z*ckGc`)jqR_?ikOi;uw6JsPkdZ-f1uS2&>WA11t2p{5n>@Of|vj~1N5SI#y2W26U< z3Vr-c3a1Yb&eCC)NUKNLVDDjjY{)tc19x-mDqV_6<{SCMQ=iQnzJfJA0w-)7M& z@vP9gj#<8~6fUV;6`Z^?1#QJ_>>rwhSt;rG=jF>k`-4bAe*+!9yB?q7BCtq18sD2Z zJ8V=aLhf@Gso!b-K3b1wru!k!eGFT4x-dtiM&4PKxcR3T3W+rkUn)sOMp|@SP^S&` zJfqt`9%EMeBjDCoDw%qj_IaJ8&5pB$ws=z^>E1-)qh%ybx#>8soLkyXsOXJp=(N?7>SRR;S=h)Dyz*d3+_1H|Z$P5qXmHlK}drK8C#h7?Qn)5-pl%1N(Ug5&dC32Dx2h z<=iJd>W36&EVztlIiBIXQ;G-nq1?H442PE{au?e?7U+F}Sw6{Ol6>aXI6e*%NvcJ(etPk&Aw&A_0KIR$CD(NXE1kGFP8s zo2nu2AXlQ`T`Kl@PeocBq2ZS``YcuH^@o0RSF#7X+o~zVHkK^E`jH>+GMJVXLiJn$ z%Ium&3o9Rsg3NfIu6+kxtr$vaGX&~wzra=mKWEzqbTHq2^8~Yzn!*@8WkI$n0j}Ph z3l~<5pA$`ltdnL!gQTUPI>s5dLVMYX@^9=_Q4d0u?!e~7dn{K;Be~@#XyEjCYP~)d zLyJ_gc=%Y1wo{~f(R1WKYT|i>Gstha11Hl`9A2Avro1h#q!GN1Tw)B6^i3Dp0_O93oZxz08Y3oMZ`H|#KFlWONN?++`PN~Im0{56ZN-&{%sYiGkdgS()Dw!&NeB)N2@ zkn?_D?%)o<x^gq$xc4~6pa1;J^eK3d5h+M%k|cL0s*Rq_vrSgC*&>+K2Zqv& zu0R_8ct6?XAH5x5a~6L*J-(?EV+dDkRHpB_CW^W`lxe*H6=8IwrwJg(B^u_@dM zHHq~3K2XwXJdNMV{g^AyLq0GGew(dXN&OYJy{m-%8=ETZx6c%IwO{1m7fwT)Fc4$b zC}MGQC@egZF+DmJYI=N!qVF&LiSC(*j~sQrOMd&{4Qn`KMdX;;}Lho6h76R zLiLSKAudeL{OkJb*g7f?YxNGG)$$WLeSAl@oqy=B%X^3)mY@UgW8Qrs{Mcq22G+&iw92i5hI{fkEtOq7? z&PCJaIl{z=vxPyEZH3W>yM&Ezy@iNlhlDl7>2#&GoQ_?;NBhSvgM5q!>iz$N>A@zJ z@<|(8cn|$d(-u+`3#4hsPtxe?d+F<~MAGD4tnll+7dRl1L{+Eg>c~h;xA4bW+fzt? zd4li8Qi$=~hu6AGlvl*}us*k`kGtbe+!;cSLybwU{Vw859>RTH3v}1(!!cikK*c3k z;-oW^!*F`ODv4IK>5}P9&~jx5I<-55 zT|E%bekZ3f*>`T_8Wl>vHieL5js^NPO-93(X}J2xR9L)jzVP#(lVIoSOgCFy=*p>J zIymK`uzpFV(7h>F_|m=+PIq=ge`*X)7W{Bl3co3uS9n=8jB`SL-Hph|UY-p6Cong+gRG!&CVMtH4;rg4VDX@1 z@QW#9d%Bxh%&2B|Rks@blz$^K@*a};9C6mQXr3ud!;t?<&@=4`{1go6s6M|g9P2Pi ztrl7>3$XFcVtfiZhQtD2v=|1UM*ld!K0Il>+&ZcrID>*LuV8o11+3y(aSeNQ$`n(f z=MVdk7qSOl{46lOdO2dM9T0rk4u5B?f!qN@;lVC5q5s9vLSE|v*qoXJaNLO0E8e}7 zIf<18nFu^G2K}dwN63q5cxp9<9xj_e+CF*|>+20ihe&i#Alx5#aCVb3-wn)z*OR9h z&Cf$4&b7h9lOP+Sjn-9K_%}BbwR!iUv*0xr_LQ>~ZOQCUc_2F;_zuY?kDz?^1ipy9 zV&kO;LUUVxe0x2f5-(4n)v0PUw<{TGFV3Mw;~~QC>0?yMY}|f9xH-I(YEsYBi&H5i z-v2T#R&hRn@--BtNW)p}4bv@)VLP7w6D3QpU{WhH*l_*DXm#KnxC>kOtR-3al$M# zetMCG^j2~oW+Ox#k=_Q)&ed)15{>`s!T zdCIP&sO3v>*CQxoVhHs9q;j`x9t_nv75jx3QFbekD$ZDz%x zm2CW|2~>5~fNs40gS@Vrh-R8<)*AOEuD*6-M^q(YSP{PFFhiMVv95-V+= zqJ_TWadiSF_jqB(%vs2ZzJiEvmDqQ!0RPUO$HOr(SizY+Q?oat_Im)f9L~Z0)G8cZ zQiLbdqM-fo0;Xl8qGVJIE-kaCkR^+0_uqXKeWX7b=Sz^WniwTkw2LGkd>0Mf@ldpB z7Vk<}dywB;o}rlfg`MFY5NkVYc)ZCdAx|SD-uS}tF`z6`8>mue<63O1KSc#m( zoiJ1lL(SlW$PL*9slVHiVQ~u=0=^?CN0MGSUx5;5lHYPDh0Ng`a5 zuLCazH^TD~=Y=)Pl0!v5p}VA)#E(29vjN+&VOBY=M8862;X-nXu_x^xQ^>t}5p3El z`Cf1b#5Midwh>-znS2QQY5AX^_Ag3Ur5YvNvnyr4E`4BT+O4d|_a;;3&%_Til<}x~ zKh&zckhMDqH%=wdrtCP<7!%Fsm%;q!ipAK;VNkfNKnqm4<9h|qyPN-k(;#WuI<5he zv)3b>XZ#=USp}s>_Eeu{OV{*fP-l!d&D9a3uA6Ph`@Dl(x;^OIbw4V^Uskez5OTA= zvR$@QX<30Tl{<4j?(b~unakNCd>3)AcnMZGA3=dzFqSsFAu;W{=@NJMpS&$k! z_llwS#ZkQB^UC6eM08H=VWoxYNRt^3!~XpcvhFtvKF|;QzDb(LtrIsdPggd#_;3nA zjVICeEDb?YY4mr@3GTp3r9rv;EP-a2K2O0alNQ7;s)6LRdnmkp6m8o7VZzYkkc*U{ zv}s>(eZn(5d?!QwC-tGE?G5I)%To{g2HkvR%FK<%x3qMuc^n3%1J$HsnolD;OUP?$ zupn{BUwCsoRFJx!OfOcRrE{EnvM$906HXc9vZ^H(WjNz!og#9#NkG-bnG$O_pGAEG z$zVUm^{>V0))%;7sYC-e8j*YWcnYet#DFQzP-s|%``@NeZqytyP4uNDh9~IxuQN0t zAeRO@@SM2DIG*DIjb8W;S_d9OTCE!fXIJ44Erb4$aX3?(3pa-86`et#=me(S zjf8?<9L7r>;d$b-$f@Oi__PsZKgWs|oqvJmx#D!Zy#~(t9yt6U5X0WMVF%B(u2b`; z&IP+jHGU)Qv|UNJrMFXu>KL50(Z$CD+K88EK*f;P$kl4a>I?PQZSWFmAM3FovK;3F za$$drd%HgJ``3g?q_|pzX07BrxT!ax;_{1oWM%05k6~o1G>*G52ass47?u6&#iVJ! zQPr+OVG5FTHMNEsU)&@koh!7?R)ak1l*z8ff_~RO6#W^0Pc*+&g4yh~;ryc2be3mN zKQt+$CIU!_833oNtA&^$u7YN#lQ1MDkWBMkXi0?^1+;Dy{;Y8mUd`Pj9O-X@Nv?;W z?&5{&*{b9?NRu?T>X7TITjZd5mj>`$m-5Q{Y+08frkl&+^Y&UC9NUb@)|a@S@(~L} zWjK*oj{lu=M~^v){qO$G|NHy@b1tjuSO4$z|Nq1FRVHy&>i_2Vd1gbW>i^BJ|9|pp zN6sE7`~Us_C&I5HXy!KjP#=Y{!xQ1&8HbQR=b+*qLF&W*qpU{@X?KDVXX$Ct-ua*L zY@9X*mrLRIRsP(V_m`^eKa=jhYVuz!OPh?f$aC^HTun*k%Qql?iBxr-o=BcJWrnbtBI4lmMJZN$MY(Ear9{ zw#+}6t@~NT-WJ;kp>wQ+Ym3Z;rNfJu-Ar+e9Lt$@_Jc9LX&0(3FT=5ppDkj7P_gVX z6B{N^EzV=eOi`00mR!T8o>y3Jo{4jZGGJw0j`i&;vFF1OI&gwWKSzS*B&^1l<2%t4 zqlg6o#ppk*8orv>ar#vP$^xR%>GGYOT|JR9RjlZ9K@VoBK4lB%O~7}jPmD4rQbibO z+_wXnO3fwBz)$6qJ`aK zgp4Ie=+2%DN|<(vrf1s1^7L+mq#9vXm>!Pu^P#JGJhSfENvrspG)oJ2V`l(X*5H|oMHHRzzUllU&N!euW@%;45AmrqTKH?pC6{-fbT2z zY=;U$yne8=@&yp{4M(45F6#Q-!gMF@&#jHek}uL|eA>wZUyerJhyC<%k{k8@`3%(; zk6<+NJY@Onkvm5W3jcL73l(KZDz-tl@G)-955lJXO-RZZKvDboJj1>iYL+|j<<)U^ z@Wc=b_pl&?KqD%0Q=$dxpYW<%p2VpFXEICBd{l~LtxuzK1kYH^SplhM888hIVdPfz(YQlD_*(B>}L8~`Bz)6DzLbr;kaICtSUR)my$8I2L;xuSAG$WW^ zLm{dYFV08N2JIw@_wb_KIm(zcVi9^0j-lhiY#2P{`zEteMAfO#-hb@~ag~63S`3Pn z{qb|tDNLO`9S+h{G0)KxO_`}A@#6v=*%3rW^VDgI#$ZzIFG-D>_t*{TG`1$OiLqok z!DX|8P_a`+n2|n!R@~wHmjD^E{d-FI*B2wmzl#wQK<2!{>R&fM-b`|92NhIZtlTpW(3aiNfJaSy(o)0A7j-n4ggYHGCo6 z>Tc5bb&r1Wtd`~O7uffJ?;HO8gt^mC>~47h`St5@|IuvBiZ@4n_afFf!G}4Hif2Qk zwg}#ayM;$qKEk+!GwkfGVuqY-w)WCGY>b_Pi?R#R_{9OE-;c!Vbz|`-YdSquoX9m3|HU3;&)hwjkg$)15jPH$RV{OXccxV{a?W_i$e3 zyaE{KR>4xgn-xnCDi=E9uW24>zCO;MZ;sI9xty~VHW&TAjmO0op4iG66VJjM*p)&N zJOB4Trs?{fZR|;5vtOTLS`OJvbRb{2+@3F7%D5@qa1TT9-Ho`dd=O$8p5$V@i7xk> zOOHP9#k9r8F=}1_&-rY`q+@>gR_B93CGiODJqWv!O;{XWjzcOB@q9}j22VVPKVME` z{eMyXIrkrmCw)O$cMB%3I)LYWws0O{j0f`PsN`cbou7DwZa!ToJe*)7ShuebPOji{ z`AI8je7QGCCg!o}sxMj2t6ui}2IJtRszQ7RTYgDF!p=XJXut<*>-}#9(`G{D>Hfsoo5J zI}LF2S`qgr)u2j3f>c*$W0Tq$Je}YT)1H;M@@fUo<6HB-_$XTX(U9KrXV#y3&NQ)L z4!LKKq1n&Bux$w~Ou;;cy=5_+A)11L2V-4X`+@ND)t^XmXGoSxH}jm-iK92k|*$7=KM}$j}>WF?#R2 zm;5#NQ5k2TcbIe`Vq_g=R#qTP{h{dl(hkwLqd!EmDzwNxeJGuqWKL46L)hpy+3d^c z2KIX#69S4w!U=+~yX`D&VvC^j`6;xv+6(`xrU+V~ttCV$41kA-BLTuK{v^1f`fD<$%Lm0R;F;w>@gZ+Ag?wGZZ&+_7e?=s+mFJsCyaR>`z{ zge9!=Ct=P+W7IT`q+hNhXj7#&sXo|y z2$H6gMWq!v{)i=8YewU*j-t*LM8k}(K(CN9p|hJYBV3()!=-70w+?+tzDSdbFVg|n z6sky=$#W5glx@qiKWpZYV~{J=RxhT)X+;>BmxI(nRj7Gshrl>H3~;lB)~k5oUwVr?~yHmwR82p=R$<%+yXpnp89`b&8<30ipCksKXE6 zdIYlDsP(VL82?s0xOJWPVW(5mj+rD;HiZpkT-@;+snUO9xm^lL_Y zYANb6b7AyoI|Uu}B*(^J(l|&MtLlx#PFC>G+CYsu8|dr+C)#tSKMfb_&vP>dr1I}1 zchDV2u+}+jydf`ao%M^Pju{FgwjRdJ9f?p;3`Wn+Q`n^$hri)P$ZtKy9P^sksAMMU+!I_V?#q`$ZNMcbZY+oDLL*=~J}cK+>NgPp5MP+7dOC z0{Ph{W#3_FH}Y#s={)w>4ifVJlM>c_krdWj9u!tv1qja@;)IchTIj%#BrE2)DtN8N-)WqoT25{66k?L9wijkQTS&?nj0lfQkHVG%lI|S2R35P zOeuPEG7%bOXR!NN96Z(ypt??N`u^RB1idC~zx)#AR(ElvWDE2w{P65$06cs7jFM*t zpU4?gZ2C0_;nkS(vkXjH4;PNCMva~}_E}t}TG_)w2SG`wnZvbG(?)toxDL>_iod&z8V@i z*qWAKwWP!y+?V{$nmO)PWqMbKvXAWn6k8KPbt@CeroK?rB>DH`(=y#a;rY-5f)fu)P8+ctf8_1)k-zpiBF+W4D?#_uq-f%RHfo>r zfL8Hr*MhwhQEy>^SUxZhoSBIsF8A@|Lp|(&44|zSdLfMkGpn`n@biBR**Wvz-QEZT*Z0`$tSGdJq=iw3s_CUp z2A&L6qjEbPTFm=F&9k_#w#uF+I`pS2Zha7j{lg?yhj~}UDJ|k7-j+lQP4QO+%ip&I zxeM2sRNq^+WOyUHb*&iH?L9DkJCy8RUB_LoXGmARfq-SQG$BcY41eoVeQ_3SrfRt*hk+a`z| z2s8?gZkdc{@>gb_L_VW^;WE1R@EkJV7>AC}h(WQFPM~KOf>63$B65)qK|(*mQ2M@^ z%->zF87qN4-*sa&>KFfk_N|}F>h|wu?`g%d@5d#wLq-~`!weJljmk{+yoV_xJkOKy zo_L9=F}}x(b-l$K2+-AhYav7<%Cnjg0oipdR%V zOhJw>)1p+xJd6%NZmun;`(X=0zU3%S`~hK z4Wkk_#CzQ-%yh>XF}i|%)T-VbO&{-!vJWYs?*eVod_y&K_S722{89iTX&K5i=fp80 z0Nhc=klqNl=+ z$bRDvG^gkx<9}}>G8J%590au+o7YZ3Z|ba&_|pm|*)o%PV426PF||dG{Ef)BXDy1) zn1FPWOOdGk3*@h+g~Udtp_PN$0rr3jeRTXp47)x$kuCX{!Mfzun$^9S%S`^fixF{~ zkGwAIL#_9AAO~$Zws%;E?bTCbt!wWH&iX5v&2J7cwkKTCMXO*GIS_!puhB>0_Y~2T z4{~Vqgb31Xkw8x)gaviK0v_emMa=%x2!USo2?~_yMorh-(9Yz^$S!v#>L@fsS5}?j z9X8A79c}%^v+>ept7~-FUp@kk!wqLa9N-Wdv&tD=&AiU)+FW5b>k{_n!Y5|O9|xFa z3<~ki-!ehpUd=>j{bwR!5agg&Ml!30+?nNL?=W)L1)4(lcIMCLC5*-CY(_Zkq@Z8* z3Zx&s2I-zTfKjSEK}4^n3n@#=wMJ5N=(c{W4ARhM@ODAw^lx4V#CLyn5!F6(OnZ%bIJsz z!YZWlL!g%tPG)+4`ZL9H+n8gUe=>%G+8QdWf=)cuN9Kawt|I;;Ooh@P#^$voia0bL z<;iX3tr65%o?G>er~6h=7x*Lw89zUS_6aya9lIn^@&!%w?dAf;)!$r=t9=uh!}BgNzf1T`;_ZuqJk&*08<~!TytJ5CA9pij#>+EU5O;9< zQ^Ytr6K0p*L*}(h8zcMYHbYg?8Q*cajCx}UBmZ*@3how1>4H9ZDpMPoT@AOHtlxy` z-~}^VWS24Lz63MhPE@1m85D)Dyp5(8y0R6+F0950U$)eE9GcRskCMlkqhhrPrWEHg z=IOai!J18sN$yD|$0?Rc`zWZbd1r^-UE~RPYEtOkt8dKxQVEnhq=tg-&p;;xjD&st zB23n5Gp5*6fm!1>4fSj>Kn*Q|-cENyP{`ysG~xMWbfiEZDfVfhNnb^_q)cFg8;K>|?8NG(|KJ1yl#1PrkX#!*O*C zNvdK-?If6psa-tY%v#>5v2rN+fIPD7mlf0pG&0Vkf}HBUDM%u1E_3p*EmQ5ghEd71 zLcb0jL`SbUqPE}j(bjMJg8yQI_N?$jasqZkMEnU9qH2!DeVUKH4p|_BHzp`j(-gg^ zosV9eFGJ5KoJC=O@=?`8eN<7ek0NDeBD6?@=lgy!Pq%wB@0G++Bpc+8rgSHwTvtKw z@q}7llUp;-*xrtHKeCcly1a$8H8{@n3wQ>fA7nF8Zv)u1KZ4nq;&}FYn>X`m{6@y& zWi%sX@Qg9lXk%XNe8#xU{*APqf1|Rt?I+9SM;ylCEObpv+V1|1f?Q)gYYB9I!TNT|IOJ=#mzp+=7V`Qa980l8+o zMEPdVl3UF_m1Qw)U3G$buA5A^K<{|@%x=bS+hIoPkTQBJVTiny&5+Sm17^0M&a&^+ z4(4K*4oW>N7?%w}&i(ay%uN4tNJ8c^(w&ivwl8-;vj-NVr6NJ7^oxLhocW1az4<<) z(z%>@W4wiNI=`2B{v@9FN-3SEy|0+J%KIeiA{@%z-W$#858M#cMr8A#{~B*#_bw)u zY+}4NhBNQ}JYxhMY?wz^;u&otdzAHX6MB7MJvx@WlSvfBba?|QOp5bortJ3!lT-Vi z$xX089P&j=gSMbc|KptMBr(+5hY^n-X0Bh5L22hCk=xXKw0`&+8qSMG!-uMw0LQ1y zdY69Y;$OsLmYw1)Nh;y_Z#F<5-54ZUxDZ+YZDNvcmNB;Z*BSkoW?qxQFJAjSCB`G* z3St&lqKN&aD0aCPbE(Llaq5vYnW1i-VaN%R(6W)d47lcL5c=O-3Ib$1`u&t20z)0#g&< z$Al^;GCSgvnJD=k%&Ek+O!7B#W}mb#M|7EmyV)mqlQR#l07Os zy$?+;6lRE^r~Fb&Z3Yi&paxDKZQWfYzp%h7y9Ys&>J86kAINDS>*Dvn->3AkUw z2~6MIBxa}0LY{-)R^G`UQM?7>gn9Pt6BFm~hdCQIk1^|5#TX7BU|Ixm`6vO8SMeBS zCiV)WkD8r~)2AWkR&YD>{rM{<`sYXHoqb3?v&G$iTg*T%t z<#|ZtjV{V~J%L%TDb2L}7H8gA6*E#Z{xHv;3izHI6POD(PBS)AEsT+CAn%#x8Q!|@ zg8FOeKTO7nu}JK-EZX~B9-X}>iuTT*jWUMy(P#RTk@zVo;3Bp%J5q$u*WaTIui+~* zB~=)Ga{0xy4CSJuTcgoY*&sv?JZ4ITzcH&#hM4xCF}#bPCi6<(1F!l+08+oT4%s># z7ib^Ud2+8Td6~WTJaPXy?CX_=Y`+U)wI>@gU4NG_$@7$%wBcY@-6@(KGcJjBT6uyQ z7-gCL)@DpSTx7hKv@sv<6JmEnuwofP`;sJK}enJ3_JT?)Xj&6awcB$Z?s(}@Y@_-lM zj2cbl@PIYovByUt=jAD=TTHP2J}*3!|CaAv6oM~Dq~j9bPAI6>#v5~^adLb$c;|)V ztAAo~^%#~wrfrOtz`lpT#|*G&w$Sei7 z#HZm<`Fz;L2Uzzr4y2r}5VsCD34{jCX^Aw^JZ_QW9(n zJ&Fy~Bk;=D71;LMIV>%E7yfG#-jF?DeKBgkXgvPzwiMzgUV!-*qM)t+1ynxJ!Jj^y z!&yFy!P7k!%Qltb)JyhwmYFh8do3Jpo%OYvQ=Ni*Rd$B8g=V;=CPRSWNIAZ?|p3Hb2MW z34Y$N$VePdxNsBmOev;kUt^uI{y0~6F8fUl#f0Y^Du`L`Q^F50Tc4T8i&m;KH3O(GReiD;beOM@y28D4SS=E7h zY!%gnpB%r5t;0;#|FfuIy z|Lr^Q=`F$ds@;053!AXla6FEEdK6c#2E1e5H5}(BOL7Kwa(yQ5&>rN39Y+2d*#Bz(@5}@QBk!+%1w0vr=wB zc5gIv2O|6_w-P*$W@D32g;=6K1wWaT^3PiIIwK4`joe_ep%I80CSe6;Ikv1$!?!PA z0f)bz!DrcI{3DCc-dd!9e?3*iU5l!qu&*0bR|H~r3vY~XR^r%2URde&ld^#g8o15X z9MD%2E}>v5h}d0TX_@0b_xLZcc?Riw$v$X*x*MKD~okSv{T%)a+~3xB!`!1~~1 z{Iq5qHd<81nio1lu|E%sO}$grV55c?`bq(Dkib4okKpqjbDaCFog2Ksu)W_4!DC?y zJa{64H~X8B6dg~j7(5@(G3taF&g;mNT< z=sC}g{O)2o*-==%Sifw>hr2L!p%#|-t^^BqO~sQ6)G$QyHe{o zu>=h)+?xQ;FHXZIv$XJ6-vW{@bDmT*__G;W;iM#55t0s9LX%AzTb4fwJR~Fp)HNwc zxigL~S~5a3r5-`I#2NNrm^D-fR>3y8PsCEklKlUo|_u{>eI8>qoWv$Zxq9N2hvhhY>8@v6Z|{%Ho0< zA^f&plgun_=l<*e$9$)tc$`|07Xu2M)3^yP}QUV&@EG{{O+#2%N; z@cn3Y_PwXDGmE{!F}9098xRnTgAe-XT&)KESP%3t%VcHiDA1 z%YMeLB!$}7LDz0Q(I`lRNz*#n5&JKklfrL7PU`{O_7s7D`#WKpN+9eND}i5;DR6e+ zCipHojoc0ygYIxMJ6byjaxG@EZ}OhPX|V-FDCQRgT`(s%45f%&RwCy!>lBx3Q~J+X z)HO{aBHzd2<4t2I-5Cal&&HDD@;6{z#}qs+PYu8SY69nQHT;>oiTGUUByY3h*v6R` zA?V{UtF~|*X7=oWNjr9ds=pz9cIFuU{`4bllTM@CU+R=MOf@X$XJ2Mjmr$~O#dP{$ zT|Jrl@f%5y&!e07FQz{)nbK~B%{1h$J^7nA%uNuwK`NUj5TScZNKdknPvXoyg~|9fDf*aepud)s(F(C1dTC+@ zvA$45!>?A+Z7Z#5PKpaMO{-qS z(KMNCwuTF)?rX2mOQY#9(&lMy(ftHtUn(`zg#z>G4{f=pnWR+uODWtqeQKwkruW}l`p6Abo`D4dT>H1 zRerdRo;fU3?l?R|J%oqI&oM3JcTX2Ro)bzw%;{l&EtQ}O^IT}b`xvsN?>trMK0&`d z%%FgzW)zW^O&}(n@&_f!==1_Q|I7P--aC3KmNx4zr;EkzQZEf7y87%^nvm8_ zP3q?$X{*aL%;hEBJ|}=Ix>HD>hL=#Y^b<5uOp{N`9YJfEKNWTwrs-?G&>1?e^!A}C z)DV?Z`FR()iH;FutDiF^i}%rxCHv`G+DvcGsv+NOE>WX^UUKxu1S(c>ji$!Dpo9JE zXn#c%bvocjEh2V;)G=MUt*xFcJr+e0|{$GTZohG+6>KG_?0-k%AuN1zS0TOyP>w?FS}2Ww;t%3OPy5T zQ>*KBR5d-2x)gL#N85paY;wl@Dp}S!hHo#KLV7l;k&8PA`C;?@NI_mIy5n|}ye?WF~^*x^EC$(#&RWGl4TxhM>J>3YogO8j25U;F!1-FwJ>9 z`0Vi~QS~{*sJ)v6KDhv*qkSaD#v1NNCxQ2aaYSkISV;P13U{BCf_m#ZaIAa;TJj-y zO2}d?Zz_t{rx$>QO*Uk8Vw|d61UAQSKuAXqwC@V#s?A@qQiCG@oH@JOp0L%j>0rOh z5gPaY;E#KnLsIi7TeW!-k=$1Ws=Z~f&G81@)))cspM(R0%-BkyVs`nh2Vh*b3wQa6 z;4&j|nEaFDf?R{Rv6GbG=;jNo+lvsA5xfLG8CY{&zX==v`0zh#U9+)0zOcm*4^Mi) zYUb$U6+1=n#}HBYM3mV;4`X&`qBPdA4&bI$>aYvzWXT=TAe^0ShDT06g$3K=*~O1t zq49n)d%`aP(uayjmqIQ)kBVTAzl&tQhZKU>*;QoIBUAdqeL1vm@F9*_F>K$r5_VpF z3A~cgW}l4h1>MX&tiDksYujE3_J7@=%qt8F75U~O9e->#M8B0K*+!0R^fb&*wo8NSJ)vM|cp2_657|3o%%HiRg3^{$ zPGh?o^80jxlZjA-F*&Co=434AcRvd(nNWyd`h=~Oen^yCCqnCB0sn~2ZZfyuAM$^! zhNs&1KxCfaoMBzUsz`AlF)0yp$Ap1V+X}cTuMXYPH$el&aVeL#LDwlQC{^r2pWFrW zMpu{r-}V2RUpVxu^svsZ!)QCc3FVn**)Nw=am!VCtTQ^JY+3mroKI~M@Pd3`wOs+M zY+MH&0~+++^84iq zSgW7wVZr?lpf_S+nq>?;vJj^A%NN1u3puDI|PAtRzumrf^?^2jQj7C0G+LfcG@sB_fm3*@&Dm|HQ<)rZmIG z%!RCSTM-+moQa%WPjZ`LWQkVQMJRou2*2J<5eRoTf$=^ocJ1rmY|#2b@KM?g7X+Gn zdnOH(yH&~@ZjNx8QHyAb`5Zci%cEg-E~GJ+4;uCh*yxfC^a;0vD(=5UZVCI+GP5eO zL%fe9zcPlabyGQ8nLH68liLVO19q2x&)NRC<{ z{h+gzvilCwt(W!4m&UK`-jdnSEYeMydhAJYpa>O<-b*z|2g&l#qOYgM(vsgAbnu)p zb$Xn^&5*gy-s6w3)3U_quI5_KGyEp&sFlv{mfZ=$wH=V*v;&RXB#Zk;gz%(_nbbl4 z9$8+0j)*KeNdLxZ)5O1nWSj4C2+BUliN#W;Fqf{B z4X1&pT zMMplx)00__G*VfY>Z;k$3H!oG*V%S5r!tO66kxji+buFVJB&Z?NfRXJeS?;Garm^f zikh4E5ShCtiG}GB+P^=Y?%OB%&tC60M+@$)Iz=k>Z=eN-awxxZKAliENRrfk6A8a; z^87~-ddNFW+Z(r0eEBqK*`7#kJcDV_uq*9*6i-JtE~T@|JqTZH1t`09vTQ*PG-RJA z#i;X zb-xx<-#@WbBNov?C3C8?K9d}6PzMrN0}%<`oZS8rHvWAf=V= z=`HFsVcihvdhSbB+5&x*>P;6%$>XyEo&Vu))3`MUC1{0rDoIZ7A#$m8t zLX~2oCu%`G_Z^|WztXuSHdUl={(PD{Lyj)rR7NI#c};9hLfNYo<4LrnA&EF)MebRC z<#zIalFm{Yn%lpJrd#LJ3wlfF5!Xy=xZ)9c(j){PHxo$9)UC8M>=b|C(HJUHJcm}! zmw@T*Ipo=`0&;iBY^pJPIn4`OMEy4(qe4+mbjjp*B&P8aIQVGNyx43`bb~UjA3jX$ z*DRzlo0Et{j0MHPd+4E>H+=QQ)ue2CAPv~1Ks(dd(~m!6{%dogA0veAsZAl5n#zgw z0de-^@DDP@`x)PNw*xVq;ml405$wA7HtF{Wqj_ud>FW*=HpA*H|F2gaS6nVjC;s)P zCUZi_iI)2?b7nqk-+788rbe^r_2H1YS`yk8>u{H>r-O9QzJKQ31oN*%VdZpM&oiOl z-3K|>kF%)r{o^#pLW61qtPB)pHpRM$d<7!0FQk3iz>gRvjaR*L_dJ@sqZgltXA(nl19wg+- zSo=0D`bSicmRN44KdnPaO|&X~n>k2kkC6cRh+x?JemayMlcA-JTj}F&59GMVYy}$P0JFys4e=fpp15$W;cqkhs=E-i+7>0oT1<-re0;{!};HTDmNLa}w zPEtpdDD?{ZtByGd4T_tv&|L+*Q+6v}ojL>G(rg&0!GDD{Ze8|= zbD!~w&yr@4R#w4fUv%(^mBFxIeG~SU*@$D_H^8R1ui)egRa~}p2d><$W z!s6u1Y?|0a=ri1pnT2z)$wLQR-XxAqlMG;yhBY@U{3VxrRq`bK;_qHCUipgL zUH%0se#_xmNhbLBPgg7v63kr@HO3DPRfGLK5qv`VD(CcFkN?mrkF6WjBPU$Fi0RS~ z?CzkW@VI*{?rU(xKf@hB>3}5;`EwB5^Eu$BjwLo39C#)qf!7r&{NvScpc#tX+U#s- zY~2Ze*Omd^b%h-1?PYgvxyV)wNMOqY61d376!i37z=jX2Ns>>&KWjqW&U9|{*a|#$ zTo$)hE)Qt4KP>530!{Zl!TXCO^cPkThs`nA;_^aH*U^|$O`8cfTbBV(Jq_Y~e6isR zJDhRhG2BjkM@rkrkb~JKXU6O`B z)1trFm9YBa1L9v1K-*p%q3z~2)S+9M-X1WZCFT=o;KNW7D<4LGy1gZJGE=Ghj77vR zLyV4fwWN_dCUg8(iR9hK5cM@_SI1b_VIv$^2~YT}2)D zTwIkN5uoh9Egp-d437R|^MZ_USgD){wX8j@_j^hPUkj551N+JC1y70p;aV~c@8kyZ zBB_h)CGOxvXR_zy7Xe!=gtPjd&DuEX5T8e)biv~&k}u*y_r@=xZxz1~*MdlHR@-t| zTC^M9ySfpHQ8~~mD*x}i3)uf&7aY4o1#SDzKuy_x!XI-LJRA~E68<(R;pKi9CW>sMJ7*uG`Z{VfaO+EUfISCA*Up3b+N**+ zP@81`5u>UBvN;pz7;`x~Yj`4kk$8t_>zVVV zj_qR9Ra#0_H}Jq#VL$ixlO=HuHK7lVj-%Bfs&w(|yCh3omVT7CqxW;(k=>P@#P4M) z`QqI1Uz-R;DP93%soNkLq7AK~Hsrl`JKT0s0P{*65-epzvYu_E!Nia>zi%RAUuTl2 z{pR#@ml>UNcLSB4szVm2Rg*``9Yl2QFxby4gzj_FFvj9BN!@yeJy57X<`2AwpFwA! zc#PnlC0h!O%gn**Js&2zo8a{!ig-v%k3I3iliU^PiLCY)qCgi}E^OB>ye4Wt9+v#Y zGC>CPA{K{BwdQ!`BwbuwyAJ!weCN=s0&<||K7VVvH>h5kOiZ`0z>UTkkdPyR>&j>0 zet6BEJ0wrGcJ@QV;0Q?P{)GuH!l3Y^6f``ks)apFzw>X5+VD*Y5tRoqW_$K9%oCz+*`FErlSO_$B5uRnYmb~ z-WpHYvxB^i&md>%GFn1=;Hv#{m^n=pN0pz2P0=?YMI{I#`WE3o*5m2*ecsqrLvU|Z zr2sAEgzHv;E6-Bc*}`%7;p7xBY*fHqb|3$lcV3mA!&>R0a3{G6C(m1)!scV8;V4+u#0ps4sMs{y3~iD$7~5+>vIYAI8DKO z6zU;ow>5i90pp~DC$QupT}a&82h(+1p&|Vbm|UHU7yIbr2koDr+29zvMIsuVR>>xs z-o-FJ#}q$T6vG4B%lLt7l{w4ayC_~k3umph!@0giP*a%*o~eVPbUYt`6J-6*)U>J^lnegN%Drg1?{V=>$J8RYPNH`QjI5HQmbYTbtC}-Jd94^dFgKk3S`2YQnJWI0!!y zo957B+2gdOWh~i{RLMzwm_uF8ucil#Cely4XHfAK(sUzFi{M%@*8~-RX=S$7#mCee`1Pb^bx|a!@iwH`F+8s93Y$#Elorrk z>}>j0?I67&#?ZvU8|({@L_v<`52-4jN0uB+BQXudWX4HtqIp;0pR?|^TWjghl_!b) z0v$55_$v|LupXLpDoJf^9g$y#V9aSRHaq?Ynz{Zwbng=3&t15WswRZdtp?(B@5#Sh zm2Nn!nW{uOboW!)HWhL*Z#+CdQAWpGI8o^+LB2#RH!$qIRH7ieM<>C7?iRP?qp zIWvAT)%+qso1C^#JspnDnXr?(yO_}(U-aqBXRpXJErulCOeMP%)X4yCCKJB5ku}23 zRR8^Hj_xz3{i-u4C$36XdF#`-@a5ds#-p_6eGJrRr-E7E3Fy`KBQ}Y7WISk6o`eHg zQgI!E>z8t4>h6<<=pk<1kt1x6%rum?Ack;>eD=TRf3X)ls-}9AJnu20(;mmtIfD4h zi8RHjh3jve$>?=;HKSECi`=Jff|I_|-^Gf-ZUMsy8INoKMqUAJ%y zVO2t@r$Q**`qGSA7EC22-nMk@obj~y;(qFJNEnX9rqSYlXZl?K9vLxpq&a@Z^vGL% z?!$a#VpmkajXeI%J!8~psDcYM=#8TXN>bWXkw5&n~vtza(Y$bV8qkwUmM|;mMsYq#Fgro+I$~Pi(M*Jv zMbNE#fB)nE77EKar(%i-LwiU}!yKhqL#p5H`e^m$ZSjMAcX78pcqZ{c@ZXwb4IRcM9 z&V%`uf;iR8zr^*S3~?wQVJ$-~$d0;TQuJ~R{o%TdyuMRUVtXTKLehSE^4fua_G0mM zQDk$aGx=t3Pu@PhNNj)T)3 z{9(uDd;?>_`d@gjj3^E~g!5m5`MX{T^pyjm*nN2z{-7F&ORkiI(Yr;UT)&iU&7T1h zK%W5R5N@J zc?)+!F7fv>q+CG)f4dX}KOAaccWoTJi`WkH zZ#f5cV{+lF_hYD? zFc)VtE&R}gN32j=G;C6mqRk~I{_(@f0Y)GQPr{E{zPBSr)h?IGtRX=Potsv;XYh7!0Ft^F2iZgvUmNuQa4xv93$gW$`c!l#)xLx>( zUFfrg`;lr)+=BJ+VYy*$WY~tpbX;ba4(MWGo4N2beh<#ie+r>ZQ@GTE*XZ!x3P@h8 z1(Odb;I#P*CqtZa>+v(N^$7I~(3LXv5J436RrEfG2Cq zh^AXCYZ{tQz6EY3L2K5*l@r^j(weu>`Ya511GOBDF(OkQXQH$;C9Lql9q+qw1&rnl zLSpAad`r3s^t)2n8rw$Vd3QExv6aTdb^+LBa4DI3OCF3+I@FYZ1J|^z0-tyf&vxd( z@JJ+FZ%_SaEDWPesGB!S*GL|QXF|qUA}fr1|8am6$pjLovA*cY69yk2eMEk2^k>h= zB*N8og*x!M;$UV!j-?P5@h z{sB8vM>$UT3v1nJ3!@ds;Npr2xc0F%?(6#k#lq^K8W#^=q&~sf*Qc@0jn}a3yDa|t zq6~H|m*ti^&LCP*mfWs&7x49(voL9sD(x5RPklpSQ3m!15)vrkbOAE%L}iwal?Cl9foeY2bv#*;&*!UutBRiPI~kN zWcmv5TZ?P>yk83bYG95#XM5p0L>9Nn>f?!4GjQ~u`OuzNgSUGw#C;zPu>K<_++7=o z-RC;vHwOsb-d2ultl&yj% zO;myUqAh}UZU%Uc;CCE0KY$h8t;qs=J-jLXI6myS3UAu#g!PL8u;iwEY~z)JGqyTo zxu8F6vqc+we7*{fQw|2PmV9VeH2TLTt8oH$NJ+pFA42i-ARaVECXqzLv3N~^0sCMo zf|jLrc%@J=?r%@VmyQW^zAtrfro=V&WrYV+7`R})u<6(~W)Pf%uk$UwPi5F^H=$!% zHCy}a5oo_h*h&2)r*!N!q;GglCSHHb1$S&FI{gpfpq3Dpxgw0EHyi~=9|;_CMxL~N zl^|7O6S3X%KCV(G5`t&XqDOX$;ybZnB-f>xfO#@EdM%B$eH#R#w{O6W!29gU+!{F5 z)dBk(rLjrfY9chcoolVKAcv-EbKW60q2hTv7?)1O%soHs%SPgWt6s$ZUJMD`5)FB} zOF_hNA5<@^28A9GE~8Foovf!$Jro@{MD_@MhU*ub;oWgXthiYRFNw(lR-_87b~W->eON2dvn7+a zcqO=HT!WE!#Sr={8EhA8lB&k3@L*0eDX-C_K{vvINUCF}&Ei;Zo)<3mHiAnh^U3z{ z@4@lnLwGwkiG3*Q4}BM{Ajx|?G%ef#M@t?xT$KiLaDHy2m9`Zf*<#F zdmmV?%Y$)@1X9^6OwJYQ;UiD-Fe#70rs0$DUrl9vusa9ViYwt`OL%;ACt}0ptTRc+upH2Q>?Zf}pq^=BSSd}6LW>e2W z$DMhkG!kRm4~FDQkvZPxSOgEGBe11r%=KX|6C4j+ha+xR1eyU=+<96YTdhtbx1NZC zyl)q4`m+oU@TS8!_5@oJ*~uw?5hI6Qq*4WQj%4`fCX^b4v#-NHv-9;mSXH-X%6S>m z?$RpSIz5z&>x(5`&Vf|f_YcW(R-nfwrO^FXrqKy5@wE6rBVDxWBIy_VNGxAOkzb7| z+M525bTl8K2M1E<^{X5mvaKgkLT^dowu^Mi!k7GVy&1%OiV`8_qSRCNE}8uJ zF8!`N5o$JGh3Dp{|MA1i`wgk_+odGSdWd9ISkv+Laa5~aiMm^dFeJMTfu|VUZ66?6 z`)i5I0xPQMpH1(~Jw<;nDWq(50$tM*N|h?h>7?!Hlx)kS@?$LNiq9ofR2fkt^%&x* zF2!$=nogsp^ppRNbx!a>x_*i)-R3l&${e3g{qzQjpX4cWtoHhgszherYRfk>8gddsCs`gJ)?St&N+RJZk=t9$#*)tmXl8U z^WVFe5bGLE_rx1h56+sZ-<2VYwwI7ETp6IwIVAe-6{2lXMY6iA=)kXo)U>ySUT`<3 zzr^MO$9^Usdd85b-|d`{%TrkWK^31#`wEW*THU+Jld<_9b7Xs;c9RP ztRL8ePwlb54?kT76J>!er>z+NAJX15tmf~H+g53wlSU0HDk@TYt?OEw5)BkWAybqg z8A6mPG-{+#G-;M5mHMT<6&W%{k_stvkz^(m_xHZv{a-%E^P<=LI6C%T>so7lKIb{+ zC&$uw5-%81tf*6Jiol2SHnH$Owk+4nh`9%Mpq6SSPDvXEgZ6MX!PA2M?y+DSkDU}~ z82(qs?BD&*`tRoT3TAFxBgkm^3)}bVu_q-{Sv7wJ+w@>RyYqe?`(^tHgEm-Us);Qc zo_K-n#;Y)AuXo+*&0I(7Z!fMqRe{#=s%+c#8SICfF7tiWgLMK% z*|WR0Xf}Qk&MQblnVFF+af}};{<@Yu+;kk(+kX+yQSnH|1Yxjg8IEqO!Mn<~nBTPn z|Lo9Z&t$tWzj8d&zWo&cU3iR6H%_CxWg6z+8bb9rRd#FZW;Rr7#g6>yK@}54Hls^i zFlCvW;Fc1XSKr)#X$I9;A32(xZ_;NsMx5EQi?i^vzBQ{XYs3*_MfP5*9;%ivMbC9d z;jp|jW<5BGah!u$h4&kymtN(ZTvu_xcP<wo5-hTF$b zxt)OLd3y!Hi>m1vuVBI23=M(dlQzMk=BBz0hmUgI6emGT?s+`vDgp{uWrflg z)P(Ez<_f~Sm4pV#Toz?6_i{TwT##KmNfxw)jc(7=!FnzI{Q2qI5y6bza;Kz>;9RE;^342o^ z&izuIW51%%y-`|tYrBlFt^R~S{AYrowQHg9tZX|Pwr;=|+GkO7*+YS3=2pSv{U(BW zBCiDR@{^e&m9yDEfM>l_5Zg^HnvzsqlSDWT1r*p z59b;>@^h2G^xOi$?o6Q|m)pOn{ZO6${hqol4>t=0B4hD#nX15Y7o*5W8?*@a%0T7-#=){y{$XFKpmGJFy^!nY}24 zmtmivXnz_uIV~0FpVkzF_pSh?!4(37^viWko|bjT+h+?FOrI=J3^S=7RF=R=z3~Du zI}5?$Cq*c?yc1hanPUIYA6)rwA{snW6nM+&)zzP; z;O>cQ`1n%?rkef2q%dV+t5%R;%+nG^PK|ED0AV>m@T-a;-Fx1 ztgdjleJY$b2*z>7rGmCu<5{7=3!AQuVfVVlg+IT=W0|6Y(6R2QP<4w?c+ek(%PM1p z@26iADwg{QQ&qzS8cQC7Lcl(uQ_d0L#GEAI#MS`egSrgCid^nVnvE6Ke3cd6nAt0E z?dqq2OAe!$mMiYrXe=0Z-6E8hTqW#z!DYBc3Iu|PZb5zx4^{R&SIv_e+6PHl#;K-5@x>*fsMO`8&-FL&?zJhM@7SU6E|X#Z0@ zqkaNAosEU#cJAZy$%_Od+fEACy*V$Od})c$_0l9E1~>{Wq}K>{{+=t$_m39pi*Fa+ zUePGHdup1n)cdsHqr9E4X!~>_3HpDTtN%N5{=d01s$iv1{NV;+@hvT(TzG{red7$F z9+#KLNOPf?T9~lPu0e1=f38rc`>gQw%#Feq3dw?Kdn=*6=v-m0$O++ok6_`iQVZeG z>b5!~w?LkW{-D6m_X5^7JP;K3L<*M5&lD_kg)I+rgpnV&2)nsF zjc-(j@XgV9A*Ux6o-Uaq+~nmYw4CQ4TsTuv*d_c=@a4xmp^LSna1LiVWJTMBBJF#H zGBqjO{e8A@ykUf}tT{JxABI+PAL~6M7@MX z>)ix5Q~L$&7sZ9|TdoP~7o-Twa>ok3nkowC?$;IGT{=VX_DhVQQr}W|PUfbdN4H9N zS3Xo2WZ)@CkFyuvR}K~GuFexSx=a!-Dew>$sRh&}a=ELmTHeC+xz7X#S9S<$YL5uZ zxV3KLp8LXm4*5cQxlH)8Yn9M)MXSKQdzo$ie zdl6Tx3t`stD_O_jBZQ8tc(bt*gC=LgZW95n9_Ybiddf(QC*xYxllW;$2W+o<1Gy@N zcttrBo!<UOpIWG-|%o5P3tQY?6 z*u{zuC$RZ3Gtf;?hL4uXvk!B^vGHv>P7&LVf7`a=Z5cjlpYX(CM@#T~AO{QL@6)vG z@9;aP30!9AVCx%yFwJs-cy&!&Eb0bd<0hcOAq70|xR*)0EN81-=dcCW%;Cw~M-V7~ z7p=aH!ri^mxY&Ci9(^T`a?T;RhNsI)MypXVj$KH8htTMF5!|r(CH(E4fXc;t0DV{D z9ycvEW>GwBzvYKYQmJV9*aNRsXXBjkG~9XcF|7zPfpPcpAXYgG6p!bCsCOJhs70Yu z;8eECqmFqCB2k-Ph)a^+aJeT-aP|$R-=|36-0PoM@Q@<^s)-(d{xfSF+{I%tZtK`? z_fFXU?>M^XjAix#2eE026F!ho$9M6~uuoSSPiBw8`4wltLE{Qczi=E6NuNPUl8-C3 zJ5l~$A8fp;NX&P4l6 zS%CL5wsqlqc3Sr|bv$kewI6$^{T?nG_h&9%2pELFS0BNYGzm|*Uvrx}bj&6?KYHke>>47!!j5SA z`_s!D?`^`%OsnO3$3ZbkiWb(#(T6uH@yR@WEWGiKbL~#0ZO4^(uO}<;pz1qKkIN++ zf_OwX+=_KaUH~`K1~N^=y7>I<9@rmY2p`Wrq)T^Fa^#^7(QAwZd-GRhPQf)A zl~@F&x#@K2bXodxWisq9-$Pb@E1+>IRj{g2nv5N}!BPG@aqI&qdb z#!{PPbvMAY4PkJ5!h0$^{W@#(=lV~xB&g2UZ74Wf2@$#v==#=5xjRtV0! zPGg)!p=81g+P8TfZ`OoVUVp+Dkl^^f-Z9IFTD(3tQ%r?KS*^vqx=9k|BGeVZO$@VmD#)%iZ{^-}_JVF}p$okk7|XOhAf46S%wbaeP# z-dD94n6O_FG{Uop`SzDs_Bjfs{R$+JBkN#hwgk27t|l%eadhG+U8?gyhfJKNgT>`vIj+`RT?D124x%e!0s@KD7RS!J(Rt5tz z5T4s4lG|mJcinCyuUIz{JoJyz?aTNy>HR8NDeUL@DCiPNQ3)*itN~e`zlc}UV`@8Z z3@JY0&-2L}%iS+mK<>C#yr{2%`2N5vDs$>JiC9nvcMQkS_PgU?!{8d26ZnHT%`>rf zAGU|RsUPXS_)S#s>Iyhrkp%BK!(_#bw^T=I74OHn<7BHK3(ghj0QgLVZ^t;NDaV=D zRsBPr?eK*?@4r#Qtm6dg)Hxw1{|4vj~5Vm37WZpmh+8|RA)^DI5o@Qiv3eq(IGGJCN{8pWiUK) z&!Q`o)acrghxF6Fa`NV=8CRz?W4vlJxKjBK;(fm0wTJG!0e&DE7#tQnO7FQrc56#paiX~I32rgsNl=_Ww)%;O+pFO54v}(2*rjt5 z>B1&_(qPFHM0IFLK^Cn0Fq+%<2SM0Aiq-kQV&W5)fcPLn`S($1uqp($g)L%Ivo-iV z=^@l$>k{zu%)mnf$xz_)1ZuJr@w)6+;#=KHM2h(EXwMHAeD)1iXHUgPv(n)0zX)(H zb^uAEUwEPG4|@53#<~wtRBW>%s0#?XejE)2{gYvZ!3m@(*D_wj_%h*D%(ge*%Z=N8b$9UH;^?B zmq`8;6_9U9qwD3b&~5dnzhQ{p9yT@#H zmLJ)&r4mG>N-^@UHBDSTL^SSM)8c!4*n89iY}Mw$y@}P-KhyS zxi<}0yR2d9CyJ?a{3uxeV+&kZu^oz(_v7B6am>KS9N!y`!jZXS@M+UTBL2sc<6`VU zi7Y=R{@9W6N6Rn+uTb>t;&SS79+*^;%JXKONJao;Woz_@wRzFJ>*8n+(>Uw)wRNcpj@Z!*?MTr0wS1J47m^ta@lo|(=i*AR{MbOjmfxT^j5OrupH)DjAwdJ zuHprFgUM1}Y~^i^_o}7@s@EKFd`}p37QY6Uq*_=m^8|MLJVK#oE1N&7l-%;P#G9+) z*hF1vy6UVJ)vMNn_ebTh{n00|I^K+5u1%(f+DqsW!#nWh$SVk#vj!JAJ;X5Y|5%oG z0UN42#!RoZ^2VHhOLXomr`+2;)o^!%{~qjvtd3E5T<=AH`uK;^}Rct;df@KTF;CzK=aL2p==1+SI zzvtSs9tC^$T*HJtJ2M#r9(F_SA92k6S4pwZnK-XvG&u79EMwsFQZEQNk^_J1kp(=S#&WdHn9{xo+&p(bZm@SVes7> zn3Foa(0Sq0i&% zsqR>0dt&Wa!4l*+N148fl!r!ST%y!O4oPLk%Bkq4gvOaf{ z#nRJg?fXdB|Mf22e$yPHe^kLz{pBDgwiOhw0TCxa)FY6&F=^+&U7>}E-TcDrjblhq&lROmv0(#skQ{nK^}o4Op_3xKNDHNvahxOmQZrq$ zM+|SgGsfB-k`Trp0gbSEXnQc2_lZl^X{{ciD|;kRea%nIN%CN!=UuU|pL5n~odJ*J zB~)T+B&f;^z?SDvV7~bR^cj_dRjC6I`QkbBt0mxV$@E0%~7)e;M+*m{J#I>)Cs=V(L4By$+sA%OdXYjAK)1`g#FVsOhd?2D6zw2gnj zQ9Bi9^)E!6Ow-`e^ks@Ymi4hQm5wG;;#l$u;7Re`$D0`yob- zyFtcp|3IU~_CcXfE8O^IiI+>-QOiCS%Sx-!)L;fXTRw~4OD~5P!*b};9t?M+?a=$# zAj(-q5CcIIkyrgr2i;3RD|Z(B^|HYC$5+ztHsK(qz8r&FXVAjQ2|z*&P|4~7B-;G~ zorp1bNA3`e)9rw~XS2}MR-c@!+D=Ek(*Q+b2Xsy?!^*jq?BBS2dh5CyNayRK#8@d< zl95h_Q)F@K*GV|XLXzz|$8iDOWpT4qJHoEf?ApL$Ot<@t?}r1KZE`L6Y)b;;v#;RU zv13$LF@&lV%EN)|`?TrQUWo3ThJI(;ux4K|?q89C!Ku=8Z%iq7PTmXr@A~dGc5@? zTBt*9zO14_{=HQ6WGW4;odw)&7aR?VgFrPuuBTZ^;ZYzRtECHWzf_^QMjdAieTUw} zTM(FKfk&R_<4+YMIQ#qr9sCjkSHorD+?P90SXBbiR@XZ`PUPQAu~)da{#eo&JU=ot8#k?s-Zs+4qqTPxjL2 zN+a^*Sq=H77E3q9ao#ce46@B}AsqVC#*1zUCZms~ah?DNE3?_&RvGEWynt1lw>q7mx_RdKwXVE`{k*LKY2<*aUVW|1>_{6guX7do?{+nBZ{9{? z-r7$^wj02sN^_XBIs)9DBZ0aaPC4VuGt(eC%5)M{Za{pWLy z%#ll_)l;6srz8*S#Cxl)|NfvvVz|h9O7IcD`N8LClxo&GrUh<=IqNlUUuNc-F)M9pgqt@tTQfsfN62)}d`%L`e0L>BK@_jucpJ^c>DFh{dGlMH$>B^5r;JIx*QQMQt6ZSkH&Yxab+eWML62^`Y@023mjoVVZDH|V= z+b2g6UG-t!owN`tSv`*&Ih0HUUS-5v%!BuO)IHv$x>}-k--TnkBXQ}nCEK1wkzR>k zL}Yvpbh?-D9*j>S?-Dd2r*InN$S(!!*`jFYWW&3m^o^JM){TCSks`Zf7ZZgTXEM%z zBHdJKP1(z1bf_Yp*K(?e_w1({U7q7imPSR8iE>A&?c58zE{7U2ZS*Q&N(ZRl;%s`^ z^cUsZMnUTOXlOYa0TL>QtdH+XAe+W|!G;JzhDVmr*M;>Q3rT@?cCMl?9(<%X_eM}x zpLcZW@4Im0a2~`2Rf1c3DvTL>!TPPND$gNlEvfU6!Ru=^anV^FJi6)+G}sTo_^Dr^ z>c$UitFRx|A6kx(p_8j<|JZQq(s+ehWqFVu7(@0hx2F*3PjWxLBy%QM!Q&NCWYS2<3Tc+T{t`9VfBwSF$Rt%;#~45ol|Qw)wiA#Hu~u7dUarX?^ohjxbEJL4OF9^Jlou2fMh0?c!}s8ubeydWv~azzGe1rcskxlT zBJ(_z{2WaeaBX{w16Qc#;yY1oJv%s(1?)*wBbCExhKnN+e^g8rA-! zM+St8N%r;x>giw4%X*T=>(B4wX*(Cu>c;Q%$JtsCzd8Xee}=-9C+_g!b1cu8a~!T) zl*Dn|+j*z#oUCWd9I*bg_9i_vzk|r!vm*EA8N!Q#bF@ODb!~{j0Zud5auet+s|!6=cW}y?Qs|QYL8n|9ww_m>YULx5 zYvq&JiYm?*!Fk7Wu)j&rae^dDHh+ZkdMUgaE+?$q`NLLelP6<>&NxHANbKuga!@pq95ejHb8hjX6@8uL*6|>+sC^R^vF{}B zEgOmM^4aK}T@7>dT_EjAJo%e_g13IrLf)0R44+p0A}*s0N#4g8tU9HFP1F8^pmQ5w z+b%uGdZh>-`!2#;CqL_2Y57r8x^sUgS>gH`f|Iv^ zQ;Q6QmKjo;?*YUta1@D2Jx-lh+mpb{KY8O0eSpwb7btNmpk5Qj*rSwE)P1`F1vNIb zakU&RZ;B^1`?cWu0&N($e~}Jm{-s+R^Qafcd=K7}PxX!tkw=a|x(<%Ugsa!VZpjJA z)(ybxeq83*=r>Goo=Ml3CXhEfN0aA^Z^>J+Ytql)m_+M&D^ZCj%Rup}(q}4*6EnXsIHr)AjSLEZ4lT%3rI@TOpRk z;a}>kZF0jweQPX)*74y@nhBNmPa%h_*OIfWoacOxG;&@%y~czwn-8* ztG9T>Ka-nX;_+vYm-YIp(bjVy-da%bgy#A-QOj5d`lmFCBuL!hm6k{GDnhL|AKYaU zCYnQ1ew5RTqmXXvbEYyuVz`iOhQ=Z-P~ta3f#XzY%SohR;jXaX<{!P(eTLp_jpG<* z4`H9l9B6*#Lr=XfrTR)r)PkSJ+U{hsg@whe<@6HX#UJ^+zg7gT=Z3;Nihi2C6*`n}eL8W#?bZ&LMSuz3$C*l>3~fjrchoIxMwG}5zW zvcNm_2Fz0WL2Uz<2?8L&ZFJ+|$g z3tD|NLD_I=T=^}8m5^gh(3i=?d>!!5^P9u!$K%r_u$=QC;BO;*IR# zx>PoDsfLj`QdIWBMCyOVhMvz)LFcy3_+Zv9YzXGgr!JBh^H2}y#d&qp9`~d z{psttpCr57jXqc*Ml&wR!!_p!m_rR{i>LwB*``P5ygtMt=B;BjNsdgo!Jc`KTEaHl zd9w0-&+$}?7^``r&z6_wvF*=Ju=hK{n9Kbroc4JEo@k7~m9ekbw~;}XUL(SvZFiJP zUcF7Veg zA02St1XW&7Xkx$`5QNmT_<8kg`r%?$u|uC+(_KbpXgQOS0T0M)FQ%6|lOVhK2R*fP z9PRP^OK%fnZ*GFC;TijN_mnV6oa%3jinLI!iDu+VqmDR9%T?%w{ z320w*9UbMP2dZroh?D+Jo^a(`a_{s~RMYUlye*uWH)Se2*Cr@#)@n&Kdc&l_#AqMQ+a9g+=0x*j7D`E$_e1LZ94V z^zeQ5{Kh44>Z^f0BUd3dL5pQRoWcTgU0GoHIHo;m3{x)DWr2H6(e9;YbjNLgiRq_U zsb~pXy->)038z8B1vPkMWeAlY)D+e8y!T9l9>E*y6XWt3`tUqka!=vsi)qI{YzFL<`!x3H*9J350F?ahzNx zDvd9|$BHk|tNj;#+n~yHt;67*Y5>#=55f1Niy+6s2sS*S@Xam;d=Cf0n4m2*!O7g;dOPI*nDaM~N}3(RsEv z=FdHd&}eLlSxvMc~2JVYvQNEX)p$0-i<=yf5~`v0XuU zS0x%xtvt!Lemu=QH7+oRlGhkIU4iZGdxIL%r>Smp4_$WmIt5KVFuE@awtK~3lCl(v ztNehL<9|S!>u&z|@)$+Ddhzd}OzupQfp&&TSQ)7gc`s#PR(dyGQis^VPOs$y;|e!}|amK74|u-t5Oj`41S) z@vZJhyTHHM!QfWu1lM*w!${*`=9_epi9dfqWj_bfpSR}Hh32-@?t~tB5voND+g&mH z;TY7H?f~Dt9e7%+4Qm}LF?Rk55K+p2M`6d{@asTkniR$4=WS%iRIb8}P-hr?F$Gqr z6=7*{KEB{s`KLIJS*c4FPV9V)zrP-4L%(-1-S3=traKcyk2b`s8ONx5zpeowr8_E%wcjtJVqU$JwBJ;!c#3;`pfp;Y>Df7i+t|f_<%>$=;v0 zW$6i;tiscUN$Lc%8Gkmhp+hh5h-Wv73A=HqlQ3bq8B374jYh9`vWnrgEX~uMHDV+9 z@5lkM(kjrDpTdfpej&O239XFR)2=p0q7~{!w5c*vP0Yf&>lrAuJ`XJHr(oHGySVv; zKF7eDhEC`8u{N`w{l`mZCZmtA1D;mw%}lPV5yHJo^5)@e`DxfcsDZP&UiSG9OW4LE zd)V&4P?$*9!iibO;m=o1Rvmr`)5LCL7+A3A(ZIH@=hoM4*Kna1*ZVe4#WNlcVdJtI z*k-p5J_M9e&F^EN>%d_ea{Dh#=oQ7^Vy|Fbf+|0#&Vc{rvl9Q)jt#89d=p!{*Na_B zHetOIO6=N zz_2rDH)z|fguWN{aB0zLoKjzoU-jsBc&m9UIzB7X|Hui?&PQ<3$RO_15FA zVOxAqFbB<|{a9*nG)phYV3CFj7;!%fW@g&K&5>ceZ1f!GIR@hWt&#Zl2G`BNOkB;K zDIF6V!FgOExcA+GuB#V7gUb)f`Ma=+yQZwfhjYZtKs3`7!>3nn!3kG&e5St@E_!Iv zmfd!2(=;!Zb;Xa_EdEBXdDKy*4Vm=cfC7%vTmzn*-}CWA89XBTmG+->B$La$$O@M< zvNv{^*VMsqYk&(jMZW^ygQ5`g>jHfy+C(qO?ZWNX{qSY_26US71F9}Rg!|9`z(Zpz zTH%sMQllElcIh{mzHS7yzJJ2TTl#FZiZ!$68?xD6VmN^}48OWw!3TLh`%*rO<*7Qc z@K$?#+U13(+$`~wiZXNCC(5=27_m2BPqFpq8rWm`RyO)d1iPL;pM|7yT`LT^mg2UL*U-54IF{Rn;R0JOqaXbo%f<_EaaK0YSX=_O z&e5<~W(HU;)qt*nHoC67iSkBGGji!ZTfP)!~%W#AYcP` z_xQ#7R_J8vU_4A(^c(1@bQ27`w*+dwD$}XakyyM<9hb#;g2wBk;NzU2~$wG@n8XUJD4yRYsZ0j_#th)^|FD`<8 zyBSnmT+aHBy0L+58#XFOfxSP<_1%tZqPO7)i>sC5@0WSSL=|iB(v3{qAhjR={iM#3(^~Ztvf9%2iJz9({ z*uzYk^f>#hXyYyXYSG>{P$N!^B-NB01opd;ZE^ws`2zD%sR}&ACF>ihuA)r zbC0ki^Dbhgb`r_Ex^M{4 zK~;W&!!KqpEMzZDfMxFrXQ$qtWNl{!STaG6-7DI|Zd;7uzf$_k7CaQNW0y~{?xh}V z)V-6a>zKjrJ`myC+D_%4X8%zQ?Pij(z5(=N+Cg8A7(I@I+A7A!D_``3>V z8QG~s_f;w>6?;it_c;2}v4U*Sk-l{e-tm0~y3fmOCJW?DT=HakeuKe_s3jpsOcgqd`X$`Pt%qRncp z_pu@USZ3aEfo`c#;NAvYDDmEfT!<3v+EjSEa`rN-%uM#9tBA$@=MP7w-KED0-jbGm zCsEmTBFgk8^FFBAVeD5o@HJqxp~aR?sLdr|dpt=0%I&Cn(Hc36GCj~YOr39r(@td@ z@?fDE_HLBIzKRnsl;Zn^aag#-lGT48 z%x~j3RyD615@ifvh39%QjJsJP~Hw_&dX22@j*J!G*&r0n%9ow># z>^kda6@folL1rIwc;~{d1Y7`q*d=P!5KWT{N7HNb9?*l==JL8#8>}Z)n_~Vs1@upU z58quTVd2=RxXrj8tF|I`ya&iLQh)^4aFAq;RBKBL`8MhjeK-Fnk#3YC`jaY|rh6rm zKDdt^7kSJ+v@*UfZ#jSKpO0{+$QzQ2%IWW?0!W$m6Q^uWWs>)AVbywVmTv3BE>2TH z4Hs|tH7AU>c7`nfafKuQW_}o7u0sr@Cypnzx3tLDTZ`d--Z(gDVFlN2jl;QXWYNmc z5iA-e!FBbSQ0v!6a~9vR&V9d@7pXCsr{sMH4{kPL>Nmw%wEALbmdvH{8RzK=&cSu< zv>M(`=m1BHZWL&!vOl9t*i}PYHqpnMC2~%d1TTKS)oCa zE`1%on?|3pg4LHIc|_wgZ_8&V60(g@OYSXk)0MyEtd$*mD7~E3(t|8pQ;%9XAt?5z z($Dw)F%dlnzS+}-{I8}%w0-@5uroywU&-8NzyA~Chbc6(wL#l}|8Nk3mDb>ZXdVqT z+D$HOjV1mwj?uBrKk26h#uyuYo!&_3A+x5%l1rBr8G9hjq(zsqWwXw+ogK|=`8P5C z%SF}j@uWQN@m5DI=M!waL9Jem<}v-j9n9T5lZreL2O-~+PRg+3UrOA~ zf7F%C7rFD5cjId)(RgP_vm{!ee7h(@WgKcr%0uN9Y0PPwf+cm9^hS;u?9f{a7hQwl zKl7jTw2LkIk=6=f@9pr3{&{fM45J^#En&r~brAP5lj}SvvCY$X>^dxBj}Gi-yZ2PH zn&3319#PFY&x!Cq)g-bnBDdJ{FJD=(tR0)#7QymTX0Xll6hC593xCx4uY8{q(I{21 z6?+>_VDjTslAj+#<P^4qDRc&=mZb&e;fe( z(-ZLNvH92|bpdi9fvwuqz|9xG*{2Vi$+-+0qVjtJo!A|KQ%|18z1)3>)&@^B4RFNw z;w!*m<|V4!e2tD?5=U^&H89Sw$9`Q&T>Yw<#81kx?)WH9^Bx-zBa@@FNs$o>j-gY{$RH^7;sUS-RHb}dyFqqM)`i8nC>eGm;mPAEVz`K0tE(?q^29@fF0#TF>Ja1S`FOQE4$ zM?LKI73?{XLYw|1knHOlt+zy`(ObHzq;>R7>({}P@$!=o5Mz`I>w>D$v6JJK+U&*J zJ@r(3nL4CDn+Q`M^C9VnG_(NJmre-)gYH zG#h5@{{YDb%eV|=9bPrxh|3amkmqn65)SInhN66Gq0~?Ex=g8IV;EV#_B2^OVB-rL;l1cFNY;D+ zd*6JcXWxa;^9gdqYjZR0e5eSkoo7LKunMSaMZ$5nKkzwG2u`dGS|8tqwuV-0SecBQ zQan&vr4tq(nTmIWktjP_2_GFX!cW88?DqT)G`a7?H5(dGzv2P?v0N3lbLXhB;%ni> zgcESPzy~r8c-dqp?shGw<=ubjve9MmV`4T7o?Fiz zmvVe5embtdbq8IeEAjmLP}s8Q0Icq)gWljA`1rD!wk;b&U&#t-`I<>!!6&d@Rsf>X zrkLTg4o&n%!@ZL=blvWk^eA`sK>0i1dLkN<7de2)R4qufK24@+f2;Kw1r=Iw`WWR;e>gx|olS^UJZibs719?*( z;s=R#n04?E9CfLOpFg*uQBWj?WGqIB>Qp$;^$AF$A>KDKge~^1w7Ov&=cYD=x$hj| znYKB2ggE0f`w%qSm5G{fi(pQSECjqhNB(3=VS>SPn19?4nuf)(NY@+(4I{CwUkf_^ z&Vdhdy>#~a>tJqs4HD$mf@j8aP~Mpa{I1E|8}C-y^t zE&+)dU35!P2&KO)P}XlDj(AMMCc7MveG>@If@WIF?a>R47SMN3xm^E^0?^+53?5d# zg9eboN2b$o$;J7o6DNl=#(#%$?wwk3wg-0n$iNz{>!|(B9UgNIMk$UdAo^|;XulW$ zC8;Q!r(lb-l5=r(Nk1x2v|ty#+?Z&$H2XgkooP5#ZyUvpAwx2SLJ^5nhO?jhevZ&2 zl2k%zAW@3`k~EvdLZm_}D$OdDX3D$Y59gEX@@1d%?E6{wTECUG zREh-k1rlZ7dSbHt7+7RY2c38ka*wYY((KhyM$Vkg5*EX)7fN6=>o@bFEEDGzYT>>P zUpRNY5^^Tjfb``KqSG}UhC61!#>2&A-g{l7ol)%lip}P=ULY>-o`GIY2&Q$Nz>aEb zELJka5my7alww8fb#9W9^k`UNFNoT^%rN%tY#h#Sg{w9K;Bcv&Tbtv;mDJJThI3lU zvc;#MwDCRs6s#sY&)x;6%4Mh)$TB-y;!thVCVa7IFSYb0RP~rJy#Xd@Fl`#%`y_zP z%VX%>^B#2U)hW78z77*o?jqm4w>UCSiaKokh=uvjP}pdI#uo=u*~b>t{ACpPkAMl+ zf#)^V4HN;vy=P&E*=P7TGm_+67ZQQ3E6CiM>8N(Y4!-@mOv>)vz>ZZJ*myJrKX#0u zB-anmtuVuZm-D$t-$-$H%gAzr-p0`+y=}B!cn*E`G>mSY(4)B%i|7r0wqsb$i*mKh zsnl6(l$&%z#SNY4og>VxajT=f>SyWE8{TN}lJ!OZ*pC9@>$rj?)48#4*nZdAD8jQt zny9SQfFte}+=h51ZnyMZy0C_it9!74ZsbX$%KOV$VuW0u=BtmQ>#39D0+uvb{!UyD-uTfigf2DG;J!!PS}rupz2MxFKg$#CxB z9_>c_Z|V|O?vCa5=I`a^&YREu@R@LBDkZt&anI?kbAeoUc_Xerj??aA+DHEDa$T_RbCW#@wdD zC4uy@tR^+s9Z#RGbf;15{Nwdt2K^k_jVAXC(BfM>M%gD~*Lfdw>iUV#3Y&1xsSrG# z)l3UKE2z6-6m9xofkD3IAo@B7R$A=Bx>xSFp~4BXERaU+Y{rX_g4_l-?ghtLT)Xyu zx}!3H3}2fMA~%y@@xBc3vFL&iYY(7Q`Z>y%v6_xXEu|_y{-E2ZmuSDR57+gzqpe&H z%0FSbFYC8+1FtXRrkF@@uh^SWZ?ScB#`5KK)4pDk^!^O|Tv80;3m0%xtzT0|y&lRx z?FtyV_`&=18knKy&U$ywGS4eZ8GFwx=-8PGUHk0eR^?*mx|KE2nQsXqg;!yXRU34B zZouU-yr_3G49@<12`7WP@$Al0EX#I6U8z#sbM*&`Nw;`P#(}U(}Q`lTt!Mz&q#%(u~<^EYN$~|#$oa!h)rN@q1aqo4jaaA2>bJyDL z<{rPK&Q&#hO>g$7gR1_0h-T-=tD8UJle`2R`LzQjepo}){DrVbOanH4P{RM%ER^xH zak!B<27BAj!z-yE5Rz~~k)xJqm`riUXJH6hE&xi?%Sa8&_P@BV0za)y#0i6Ks3Ec) z{X~7RF;4_d$KQjAUjwYVErK8SQuF_Vd?Rzv6!WBdRU4@3mSyz&!Pj{3LK!N~VOf)1JeXwk5R{H4K@i)S zz9^#^Et^sfxJg*Sfuu>pq!eOU&d7mCZc;NUfNbbcqzay!T1 zx3fLKw0ZE>UL6WT-*L{r&}8ZrtC;>vp3vQ82_3SFA%EEx`t1BV>d~M}Gow0C&AI{~ zZj8r*A1SD?%pUDBq;a943fG4t#Vvc!%Wd8wguTyW!6_>OuCqM4fqguf@Msj=CY*3- z;xY{9F-A>#2|88$%Ol-UZ zX0uBnfysut@OzM8o&#^oB4Jxj2{!n%8QZ^s7-Zee*v<=MWUeVQxyrlX*s_Hn&+3VP z^19hph0D+~5Dt4U3Q(;B_i*E)T728C4uQp;ByV;uF*ub4gM<9A<+D6kHrt}!0u2<) zoP@Kr3K+(D2K`-4kT`qEq?Moyl(-<`s@O~dFYVuT_;l<-cuI3DTtf|?C- zu)F^cnLkyD*)5@%u-pSBxJQW8wH(5|>p={5cMv=Jl(T-146|>B92^7_a>nNYBeA-a zb4IL>!(Xh*94fYl1687ss``o)jD?VY7b3{Sp$PJKknNEPWOa~FsW7-~26|_$#@6T< zRwSq-@B2zf#gQvyn$awDi516kUG}%eavfXwmf_JSi}9Fe9XQ;Jhd$8=ST_3xYS!nW zs!lEn-6#hCt-;_ja2~$eo^WZj#x}O0t?!36iQX0AMzd@+$BQRY301KDxLS3(1xERXIol*Ig8YH*V z`>c+?&w&|VN3e@a4Gf5naKnh?aD3D*X#$@Mv%oQfY=oDb`VIS=ZANoFb; zE!#UpmAOe1-qy1jwrVQUJRiqSe}tdzk>Iczxf^?gxLJDF==tMaV6pca@tLueoO^tZ zzIWB3m-netQIz`H$)Cks$cq0Izl}#{^?#%v+>Pi@ z3Eo&mOTIs(>rQpiqL!K5lSgH^2MI_-+cDj z^ZaD}>+O(aG6OS$r{iDy2x!)K0ZvaKXCPCLXsk>oqJB{zde)M32i1|dM^>Js|F zb_abpE1h1Hy1?d7b7*O9Bgg5CEOTWB zgv*m${Tk|Up^I+hv{0GJzrY*%8$`D)#beJ_z+%WCyO-3E*;6Lm;#6hsx2F@dxS*92 zc3qCsGZn?ze8UwIPrl(;9MEB|=UyWrE_^Uq_lks5MN%ij%ls?TVic2Rf`ym{r%N}E z2@k%+*>kOrv&8oZqmbtdkK`|qg^Oy5(6?<&pX3YDT6P`03j|2_scc3o=pExaKMQ-u zR^Vhw1sEm;(C#!F8kww0$8-#su7VTHh07aA-Ufa!JCsl2{3IdQ^(OF`#KFR+5|AVq zg3I=qqtBlSFqF!sk$g9)q*oW62z!G+%$~uyvv0xR@l=Mb?A@|^0+!A#ra zS0Z&?l=&C5lOZa>WXn}C?oB5l?nZ4M?hk_`T>U#9ts{G}XPY>r6^+ zggyC6_PTS4-Cuo z?pF)L{@yl-_4oyAc@p8jaCcT8sKMO^$1vFN6nvaK1DQ&DfM?D%d>MBO4F_dVtf~|^ z+eC2790A-U)kTy=w~&aoB4))t6PVX^9-59##}xwNuvTmyM2!6<3$;DS1tDwTST4Y; z?c!`!%?cNEq@njxLD*T!16Bfw#GLI&IwYNsfBCMXIm_yC`mYN|HcQjg6}B+E=P;Xt zn+>ML`(f#%94>9#gj{lo5f1EUHuX#}wkHFKRN#6T-6M_x&MfoeK?@}76@!{V5G*&p z40Beffu!Slh?dJHeFd4!%abvXJ~thBECNX$%cz(z|3M^`vdIo3FLbxBKy&XrR6k`6 zI*rR=Re>w;nyn$K(hFewWkuWwhBRo0AuW0@O63JNq2EzQ49e?-?qv=bJJ5hePMIgNzL%x;u~$I}e~Z8ve9Q!HbSuog;Scp`)8Kj5ACfEQL+sd0U!V0>$l5Xs^or*~VcByMy02F3o{FuE3SK5fM9_a*7q6Vqv)@Hp0e zEP$cTIb^+!-U==kpS}N6X-$dp_6+Wy0eaUzW>l zfofCo=nz9A#Q#%1IJ$`qT1nGSpVlT z+=1Y124P>h)P;P%LjctwXpkfkd6>89Z1 zR3Rt%cALR$y}kMt?L_gZ5-=%IP~QARJjPDfWwE|_yg{1-<) zb`Rr!zJ=*MzmUk3N0Co$?r^*QD5&%WfFjFBshfWahJ-pm*{Kz(Z7JRyGGS-KyPPW; zgPgm({)}9@8VViq!rD(-7}N8J#P{!pJ@VIJ_OUI@gX*sw<;lMWZh|*Cdv*;ne717X zn3=^CSIRKmAYl-!8dxGwRjC&Z4|}kC2LVEvXvOIo?yE}v%$t@H)+*n&;3_T97VpjoX7u8Gnx@I z4EE%?5l5r%WTUnMRK1Z#**AeG*Vux0yLF+yLkGSqs)Amd7RcH7z;eZ5NFUgVJBkv~ zUA6>AepJD&r?atW7C(l(ZHLBC4NT%!!^L7h7&ik~#_1O@OPWqH(;a~EWj(*DVe#ad z@fr|VUjp({{3PYZIERx~#(8;YIy3boiV54LN|K{2n5)0rIJsT*9Eqd=;{9$PY3>gp zh0EFbfG3*qwuxt=d2Sel34#F(IU8_fesWwStC+BdvK-408K&I0lIdK;N17xfNn?FD zoLDH25t@(4r_XHvbj1s@H1ITB4}1tYudKnwF9N1a)4_TdyQ?{eoR62A^;xR8!HwAi z1}@49`VAxJ4F=WJz*0~PmX(}>Rc~dO#CJuUzE`}A$*+xYTUivO&hH@Sj%5=8XJxX@ z*Oye~sxgAKFF48TvYEUs0LPm&;kjo#xsss=E9<&R=A}xe@!Ki`-z9H2&(?|PE&2GD z&2!5@q-Z0NUA2SZ-FA~xulbYHF}<3!S$-t4rOM!~Vn>ARq!^2@5eChBEt$X(an2#% zRQ+W-C5%PSDds8fH%6NzkmBYQq^5BjIc)ib$^R-$mdOq<;!}IchC5HmSgr#Y-MdEk zKg&YmC07U>5FtHg56P<0R4`b`hqdDw__tYrGExr6FI$8+a%@r3r^`S=HI~DPj^#*i z9O9&$O*1gqTBU!UYswt%j%CJqc908oQV{Z65xxe^f}Rtm;Kio_Zg*Wt^}7*fWRE`i zuQG>Ozf*$=(dyzP{m$jw`T7F3-`#>+=WNACZfdaPPX;_%!o{{oO6EwIko}4qiN=`@ zqMPuFG;k)!k?{^D^|=C3_2z|Jzu%BQv#Uwp$1=utOpGff)J7ddW9X@r`S4^y8Y7DSvwpYPRUX zrdbRmMyJ_!sT;m6?tF52;{vb|Pf*tq-=bX;u!W6xV4&zDbyb$lsP#n6((ouDl| zn|OKTGg2!TK`+be8n?dAM0#Yvo(etaiqs@$gXWR0|76J2+jFFC)QYV9C`w!`=8%jH zwv;)mMJ?sTsnv;j*nXbvwe{PKA6PG9^G0=Q9Q_wh?=yvWJBNtaJ_$H&vKe)yXQT48 zb}(6BiZfjlaPvO}^u3x*#=`oD3%l>qq3$?!cN}br!=c9V593lV%=m?fGr^D9J4dAy zHL+uv6MG`5({}|7G4PIjo_a6$mPo9jnaKLv`N~uZ;Z2_9C4Y+Q0vl{_~mvMZrEFb;&!JoG;uYC ze0u{PI1FEPf51~#13N3d5Tw^AfzEq7ax&{TC(&J;*}i5oc(1Sl|7HfZ%U`0a`=V*~ z{!>)0yq$)Hh0|NYn`s=|0kTgemwFv@r=I7#>FsBi>De`zbYKa=FR!Lxfd=a}D_+ap zW2?dyQU5@Pm06yAVhNayRKTufBP8?95F=gAahs~D# z{NzR!7Ri%wby;HZLlGuirQqh0`((a;96VA+X!KD52b+g5I4cuYEQtl@%fj&AAAd&o zaxmc=i6rqoa%7L@H>NJ!8S~%F!KEsbkQ&rWB=0m6y-Izsn(>F(8h)11+*-;Aul(;OQ+V!O+VbUPazxQj&Sd8uhhoem0vvk9DCM+h$Y4 z1;L=vtOvY365zb&3E}9elAmBi?)J~*Rz$p@H#}>p{9GadSHb zQ&K)K>BRQ`9z6peMWSJL*-fzWdrY2vXLFe+_Yz~v6R`41DToFL;^s7Us(e_5o)ML% zLr2<4>ToF;nn)t)jYF`(uMr028$g?P0LntMKGqbe=E(%se4M~GTX%)+v-OSwwxg2Gp>*WO-QGU%>te#I5s`Bu`CpFxm zIYeLAWYdM)L#R(?9j@U`K<+s|?2f-p{Upv)=BgZ$5HfpWiO#q6;GiQxRnDl9()IDofz1Ul9G1*Hm{-7**Q_Ch ziEN5x`#nx9{l$#d^5LWVLy%m>kH6ls8A+bkP%{B&7FI@z%&f?m7B5y-u+Lz#BV7FM z$M$tkkVqzqvAVL8{L!0)Z-UxPTEB zXaT8pEf6zl3>#hIAT}))ItD*OhQn;aXU(!38!y7pj%mPs6AAG%mf@Zq{?Iv74}7Z| zN!N>CjG=uxqs{Y=l++}H-l|M+&1}S5Q!AK8X$jVUrUQ>H;$h|M*B}`X4-an+gV*X| z*gs8-_OQJ6uZKk;`NB7>n0FOsd0YYIG>WI#8F7q%A!@pd;^XN^=n^C*4C_(%R3s5M za+oU#%|yy89xL~s!L2L{x2ZOOy_+P1z|%fh)ccyG1_hE?dw!CJ(PKojDv2EGP=}YY zOW|I^0GX*?1RgHZ;D2Bv#2gd^!()dTu1^VPB+8WtC{L10gC^vS*ch{W=Wo2{Fo@D@ z=ia_grlhwZmBhJj#-pwx(6smjIdPyHYh)Sn(&;2|%}Rt)J#(^vebx&HUU3HW_k*+5 z4HB_moj~ShNXXa)9gEIG>5+LPa9$%*ylM%!dg_rfu_xrSEt@q+C!np6PI`$lizw~Ir1GBeP9wLj?-IRF;6*YSeYF|w^mnrKIyWghd`!^(6Gs0ts4 z_+|r$-}RNK&R9!c+}%R9X($tqcd5WBEyk9qv&dU!fZH!QcAs~g*fotC$SRrY zt0fzNlNyIev>ajne&WEgynQ4+ER$R=PlChm74i3vTX1nX7kB*|Wd1x|K)UR#VP@PZ zc=ppCerB;8!yZAZo;Zhg4D!?FC%N#-Mh2_SC(t*%c}!ef38&(4AIB~-6}&5r!A8ak z0yKCqz2_46Ctiaq|D}^blP@I7Qw<0H>Eg~chAev{9#@-QU|xnqGMaPz!2anEB68*- zX;l49+CL0H2pa;7REsXo=B)gwdpwH-DrcZTMY2JlM$8_7h}`>OXx&f zvHg%4d_TV!q82QL*T$E@`(yz8QH+M5_b*V%{{qGvoW&*nLlDr`3lBew;Y@i~l5*-L zGiTV5OyfceoXulAwI`^K)7(F9yg(9#dJQj+YTp-?kAGGnfQRs1@wQF2y*M|hC z@^A*N0XO&)DulLiUa!b%WxRq3^&jZue2)O@GgD$RW!It9zD3)VP?c)|xaV=eX ztlop&4QasFVOj9H;zQ3xucUbm26Xih4|R=wkM}qK!m)vLD6>d}^wwfHhbQT!>x$H= zvIXS~{mEqJ9@4VfgT((G$C-}VIM2letBY6$cGpTc{6PVph`NBYrV(tc-2)4qK7rk6 z4J2ewL$}wdaBf~XT<56;^{<~v$Vo+TS*H%m#?NEs4qNm}V7c8k4fv%p05uJ*anw_s zR$t6RkG&Syy5<9Y;&zKxXH-*nvtG#KUV)Vxon`W_Q8zxyTL185k>3nK(JUU=)PD)1zqM+vtl)R)p08v`nn2AE}FuAMIPLLq!R?| zJK=qRAB30g1rN_42;lmlrJ)VxG23uDuRbi`9V2PtA4t@sB&Z*s2HWI65$l)_;H=aH zJp30xQ@jAR1!Hl2U?_&D{v=Hm*NJDtd*YNb!g}2Ju*7#B=2k9Zy@_2gy{HlrE$@KQ zk5IVq?JPK6a)$6ZF0fALB-~uBMrYCqe73R|A9#+yk%&Cl__`e|wmHMc0t1-jjRITW zZtO8izzbqA=yXa6EF5j2cX=ILKRgrqxiO?NR~jV$M1pgT4-~6ZfKTpTl3)9s5oS5F zJdfi+`p{|MYWhO=Cl@HMP6fl^e_*&s2*$e}kd~RFgsiuv4ckxAzlllo(B)9r+W!t_ zj?G5-fhd@A4Ti_2zA&n#fZ^-e+{LVEIN$mdIEZ=xk1eZ__@tm1yDxQcWj%`Jxn%Rn zEKphK6JrCfK3zslysEh0sY9l@kwJclR z9~O9=gLQTlpwssk9%R0RCkoTBYIZOhg#i|_Yj=O;+4T9zS#&eR5_@5J z@P48J8pB$Ur0oYL#@={XQV&B+=b~`q0vu*mL;e*baQ=J|%Mw_Ic5EQJN>kbu!%yXp z_v4+n`FM1fJ7&CIjZU(O08dllSWrHM5dm5!q(TjIRjDldTR8Pih=kbpF<$B&i~$W` z5=1+h&xfX?RbK}5U0M!yn|!daWi$31utefz18)vGK)7HGgq}{qN`pweHsyuwIp)xp z)=E4x6N!l3Qy3k)3$}ZOaNMyR^IUFXreqDiXTAIXHp)|*&#E-tLm8KKo`l%3^DGbe z12G#vOiq17(%PuOJWGlp^L%Yd=J7oE7h4D+m$D(_;v;aVABVsF|5!cpH`^C8flXc? zP;hu7mPmS{K93Wo)Fj~GhbCP0tOtvi_~3~gA7truW-n z`WEyoaDy!s6n6G$LiVd^m@f7gKzBOcEafG$&ucO7jcYke9hLFmk8v;#l*f4=j4*hv zHG)kvc3z)LDwN+b$DUU+9&2||`_qmz^{*3k70oE5ZyGH3V9&_#MzEGhMQ6=&EQzW|GhPW4^Akf$b`IaZ-yJ@z zt|Fi2doVA4OHr+}YSeFXA)TshBx$E!kl_Ay@^tThn0vLEsL3zmw3CPO`JkKMpNm~%QBazj}jXh|F_ zv0~rBCS{~wqlrX)V_92i&M4Mp&vtP-qAO-#`SNr&r&@tFXD&m%?^^Jwmw`+C7jdMv z3=LhX(M=-)1aysH<@yW=6y-s?b|sW{G{8r332?xE9kEG&L|!f)#D%RFk<;dgdsLqh zocET9RZBp^TpqZ;CYFp}_9vU)dSeAShsxd&sA_x;`AVIz>XI$a*p!0DAMC+ox0Tqr zbps5ww~^Gm3Su3XM(&Reks#?7vXAdCG*(W*np%EL2neMe4d?0FovE~Mu`!N4H%0Cv zCuHn8(fU>$o^yDN`^$N$@-abbJ|IaygoMCiITd)^XG3i7FCkZQcEU}cBN*mYL}J;Q zdcFnAfzc=>8ul!MD0wAtO5|{#kUG8^am3aOHrTQEG`5;od1w*pXslEmzB7C7KG6ANYOMyJ;F<2p?0H`bm`Cf_OTdPer zpLoYujj(4&+avOIqJ%>W_aXsk^RBpDTF$guAhl5~^xK+XS09N%QX#>O+?a_leE z3b-LSHN$l+c?>Gwjs?}HVWjRHgblxd#5xngbFY^)@b85Ccg=`BUpKjAvJT#gR$$|2 zDH^!Qf%=9o$2&b?I9<0H@6S~w3Pb-G!-07C!FI#`ReVI}AC%>WYYX6(Xn)KqV6)#t zmx-2*0?S#vg)4safOgDPSoNVVVbO1Q(Uj(i#3M>_NJ zNo49_h!`q{M{y6JBj`Lx8mEC%9m{<2ii2y5yTS2{D9*Hx!y^gT&^q}Q>TgwLy2eF0 z-=;(u&_PhI)dtJxoxnFXOv1*+!SHkpm?^}8P6Z!I41}Z6=PgiHdj;C|&BsR*a+r82 z4OnYC>i4w};30uW3)kVREs_x3ybPkA=fL4|Z^-;m4~w1o z@cZ|Va5(cKxR-Q5)lNGQIG6@kC7(lZ?GG3qW7vD-7R=i81y&}S;75mZcywv^EY-6E7a{F_89-Y9Cs&X7Ae zkh+ovo5a#ZlU8oSgvmBbRLwhm& z!+BJVW7*|OJ-99>0Nfi6u{wVaOln;su%U-o$?HLgs0g$^*TF*78>n6x1C9cS@6F%f zE018jrJabp!8fomq=bA{DJ6|!Z(-Cc6rRUDgqAc@Y#2NZ-%ospj&N-pSAIjAYrfHS zoM+V6P?NLzYdm9e@BkSqollOje5W*%9B?b2&BAl{!s|diDEpp=TfP8x>J5NI-z$>p z5J#Fr)0uDYQs|`vE_AX`f__zV)iPWV%C9X{4LQ-vGb==x7K^u5;v9`w3^-S2o&I)$MMOXpMGb1Kx-WCG_j z44|=g3984n;QO;i_#$Hq3NLk`e=icc?Bj3jKZJ1AObPfp`Cy1wiQa#)gl_Oxp^=h9 zjQeI0#^&8!#w8-wNe#9S87i`!z8Ruo!Y^aPXdKgsg^0TLPJMlUZ5rggJp zXnm7Co$B?V>#loJpR20)c^`Xjo!$$!qTA4aRRx~ez(*ZtEr;-F%3yKrF?m-ZM?X8T z-dNrrm{|hU?)V41_%{H(ImwtCxdKlQw}L%;M(X-`GTc}J;>9u`p6Bv1+ndzLt&DD% zG2I%hmi3a17yR6;u5!9}&SBb6EJt%LE3kUBEd8V(i_Ogixcf*Y`Yki&?*F60J*4}K zI<1;TAG5RltRZJS;+ldf-EuT?dIa4e^q17K-ii5flDM4RwZ!Lg;W*25x^ig-OD^5c@J073%zN+v!sU5cp7MP{4J+BXUxD5Tm!y?VVpOZ)B}sG3 zCod-=2>($rv@t6s?q`n?4c`0kcjJFV}G@MMn z%O#qI>12&v2+5kUA3cYv;LRR)__2r~Ykr!-*IlJRZF8B;0bztkr;SKHd5-m~x^Qn; z9?NKbfW3xyaN5>-lyK;Qd0z?ac`gsPP0Z@- zraT*tk&AnBVC>#;wAT3!Uf(rw&h$21t+)~;hqfS(OA+o+%Yau#Y(M8cD=OjB&RFce zORB!BqVBjDDbEUG#Q5$paYD<$?)Mrr_^FRsP3peF-F5_W^7O*|ge zcm%TM_h9Q^2E%1bF*aKUYo^rjr+^Ti2vo+wl?#z6$;4is53uUpFq;?JMTNHp&^9)Q zFljVGo|^9j{`@WwOVp(#m}K(8hf!n_9J z^U?J1ASjYI0)Lx2&|ItlT9+P^#fl|RUbveE?+K^Hb%}JmE*!QP)IjV*9n4xZ2}!P* zVD86*aa!leh68`dtWagVni9y)2c}3w($P3ciOoIwLC7s9s12Eh|AfOKYtLm^yDbAw zXzIe95kKho{26y-QW{g2N%IX%VcH&Jy#F=_8((^&;{6Gf+8ROAd?P`vjqMx@_Ci5O zz%e$X`5;Ig?5^?%?8@oA$X7HYPLOL~orsk! znJkOx7QVV}0{R6RXq9!1~Xe`qv zOAF7SlVUV}3c8HG@iBO+_5qvQQKC0F;qZ2~7)l;a!ck*W>Zuzhr-S25Ek5v&GDCT4xcDG+pfS3j|^#b zeJ_5^PQ~{}s!?FEC}tEaz}dlBxJm8_Zp@UYyVQ=;B%2^g=bokaoLk7U%H?GF&=R5; zo&fj5Vu2$R3TDkRKt{t!4xb1JoUMkBr^6sQI|K@zu-=eGiOlrFU!T%xBYO` zFAmM6b5O|S0=6%CM@-lLCef?Dk$uJXxa@2Q9(&NkdKHV9`Rk8yMhv7l>vpjo*AiA^ zd^-Z#EvESA{Z{mSb{s8-)mhKA3Y^+(4#F3^SRdbe+`4NzjmYc4iAfC_b=HI~y0H?P z-z!0KmI3Tl6QqM~Japs^FAeupAs@aQ5>wwm^1faI^XIYMI=tp6but#~-3xK!gG*@H z@(m=1U&Hq^6X4LK2%!PyEl2VDJw^Jt zTa8{CUqsgqsKSkj&7e4P2~rh&V6p3IXj9M!{uB8qd#(yMxwE|m8yV#hb!5x zulR$qxZzqUKB+2aGpbcscNMhVw`1o587fhapPP z6>=Wrfx_T2BK~U*DU?PM>a>KU_%|@N+e(-vqsDBuei^E-;^3@;5|s8W!kkw{c)Xei zIuoXYZV(?Vx&9vQ4~^oS@n862MLC>kDuRCJQdn}?A8l15k<0c?)K{&eQ}=@C_Gj62 zbX1&1q`buKXQOcBpc;vuF@sE(Rwc_k+KC?9n{xAM6Wc8#hTh%;jec39u+?%f|4$4w z7id7}xghvM*e*BwEfBgdklb3focwE*CWb1fN&L1bq8oXLR7Dgak8C-bmPOZ%#yuaQa3z<+VSP#0;97ycKsSkIo!hiuV3Nu0~R=_@B@z5eg=8}So}{W z1%<}q@r&jiIIb3QAzZ`YoL6|D+{>Q^T=>3tNxc-24>t?e#KXa_KR>BZY0DU}tCpC9u zb(r!l4EdOWMW21p>~{kygs{Cbe;nBN9ft}`h_n5WgSeK7V10;zkR&<_F2u~G;>OeI zhq;4TZ750dwl|X+V_Q(OvY~4pZlDr*^QhZ1Rt(wGgW-+Bv@0nK9G<9v{7F%G(<8y1 zboohbB=6D8`}3*dibd4QQ-f~&1?X1X1%)Z~;935U4!j$pDWTo;!A>QZ)+G;{AGJG5!MHO6j8rG6s&PH-5YM9S4{uQEj0_sGK8O~bJV#x3`_tX4chPK-1a3*! zNv>t*IwX6^q%V_y~H1Li{@C7N741!RJZ1`2jc6A(Qd3P%7(IadYMt4inHuaZy z^hXi)H}b-CpHgx>{|(Vi2|%l*E3l!>8GHWCpcWRx_|mxtg(ZUV)2B_?uxTm2IvLIC zFjnyS&KxKcxlfPIYoH$jGU)J|46N4i!$;SbW8dv-u;HyX%5D7lhN3 z1%Jp2a|Lkf3WG!~Mc8{nfn;BjBYOPjVf4a%;*cW;YNFQ|(-C`OGkqGF5M}-89(FLc z!U!740CKR5mggkK(TxIhLPMj+Qr(1sL8GIXk40$M*a{y|uR=^WsQ5;d|MBneZ zhuAw%DpfcfL|^AlqjgtL6F`cg`evw`34XoA1K*J>6g% zIvv-@)xlxB1gBd{@xh)5EVeDf7IAUvyZ;MjPrS$O(JVMKUlraS<^|38DH6T0nQTka zf(l(edN1S#=Dogw5fWlt9-%k1@mmYMZD&HOTGVJ++gF^;8%itm6R7RHSlTZiMK!eD zsrmIC)X%3Ax}INwRjY1+>-#m_T5TC_8%`RN?QgPDzt_;EIP=cCYc77A_fK_d@E2Uu>surq!pcFvr|YUOjOQp{rvSs#M0rwK4YAEr3t`qciT=jHCS?Wc;v zjdX7OHW2yZ4=d&@g>K24EXXL1W`FXeieK)~ndKL#aK$-Vca)E0Oc*j>PkiJgzdgw9 zGhM{ZJHx{jNYf_<*1TlL_9u+VQ6JiIn2&B>a0QRLvz^N#ndH?7ml$gWke-LHnc_ni z7^A#j#Or1^8GEY>OZv}~6JM)H+3s!78^jBCfkQ;@`E|m{Z^bjclfXK4;4|-enoZ1U zWZ-P7Dj>zR{o6y2N0!h+p8#5SMU--4S>M{7<=7wl1SGBRz_jKPZqR=*T)k9puGr7N z_$2H(Zn*FSjkGS4^hZPFNi{J4MN86)KkwpP(^L56elpcn}pYq6G7+6y|5`}rD`esIGO|YPcFj5jA)YW zVM^LBE+pNkkI{j$cuBqkIu039OyZ+)bGlK`ll6{g{EwpZ4(s{t;<$EHl%_~TQHs{* zKIhyOQ6W1*_J|)Tg+fL&v<>=PL{JC(RvtX7RRd0fbX*4et(*W!8~3w; zhKbB_OA079PK4E$7PCiR=QHnzUXc3h1b)slrb!axpg{8|sHH8#vG;!Cako?)`pO4y z{VF5HVUc743&}0QkmXgTus78!*t?f5bknUr-CA}Zdq%`V!kk}BaZEjH{x*`I557p( z@9m^~+c8!`sK68N~S=}$hm{ucP7BtPu#mCI5WsQc_3(x6%$UkX5 zvE&Zz`@9y`dg(wz@_jh{Ld)t-hKiNLPDQKJ`J($pbhn0v^|Ik*{&4N-R9H017>Z+V zp(+hB>q9dMqut zW=kVG9jWxeT5dkpi-*?FmME+9t zxMb=?)H8eyMzQgf_{WlZPTSGer>EGw=LzimYBixDZ#b8pY|ih5tm55AVtJ8C5}&Jo zoWC7EiRw)(sB`#0k*^rXCDsIR550MOiGt`udfcC{`mchvB;Uj`vkRgR^Z=5XgVmG?K3Z~9?=d3PE&%(UlcWBhoJ zc05mMJGi+(%)2K6zh~qQgd}R+akPs_T=P)s1{t z%6Mu?1E(v}KVNHnuQ&v;^%LAnl;!E06!{l@Wu7YK%wG+*<|(ct_=+=6F<&Vkp>q?e zZA-_^N?R~RW;O=EA&jm}!&^=(amkPMVwQ0ON&GS<<$A&=+ZrMJ%s^b1XpOID$YSi> zdeM{Sg2USa@ZRWL9D4jJja9HDU*RB*{9uUklPnewad(e2`63IN1ZG9E9Qdek`{Y>PYc=5 zHd_{QlflG09zsrdwGig9gpC%x*SAxif>XOC*4pg@Es^c}WWOaTD__9lVy5!vZf*Sj zU_7agX~0*%GqF(J3)OzRVn=)uR=d?w%=PQ^E=7v(bFj5K?5}P0d0RLCyHt*Ax0KLZ zrOA|(cNUTZ)1gSreJ&ESNk0Nb+SSe&&=@!e=TI&YvL+;o@b??9?TTdU-0E*@Az1&ry_jA`Ph%ca$ZXxPR$S z2&pWBzkzp2sv(xzj_;uO`bAW^U-S%D%%`vU60~K=d91NMhs9x6aBDvoT--VYy$#=s zp1W9@rtC#h(H$7JWIr~DEbx|R4T5yhMIkA(-11}50BrwM4#zvYSejo7j+PR0zuCzc zb47<1-D$=)%K|i2G3CN5BR->X1b68lMP+_2)S#h9b=^9+U-Z3d$IF9d*?w}83Lu~E z6(r|*7*hvt!UX5(`14UMZB@EN6|1sI=f_#}ZBN7)lPHwA(2vK5mebNfsnoT12p$$0 zVQR1q3j0)(iGNcM8|Dhj_=|wzXtGY z!Ijh~gPTed@ch zuU$0meKZnFmW!^E^egBTl#f1o3G|{coV*)$Qo`*NtT5S%`GbOSR?8qd_)6^NuRV?{ z^E636?=0>dyd9OrCxrL(VRW-wlAfhXQ~J&xTrAs#Th*VVrpqX9g(*b<{Y$3ksk0_b9ky>XsQ`^EJ9}5a`(l zBO3Epn+7gy$I8;rc(nO0cB;#8jl>$NNj*mQrMIKRw0`(N=@8s{S3!|#+H~iA0Y+Mn z=bZ*0sOwb`O}`#Y-NS8YmV*Yp9orxKZqmf1>@TeNH=K+nD^p!!H)d!C&{T^xw490W zWAS3^&~O!%($C;MuR0pF@F|s)U#HG!k@0$Smgu!1T)A)@DtGqB8>>ITH1x&x*V?$c z{Vur9lBbAAZfJX@A4&yD@+PIe{GlMnqjt*CX?taga2rdbE?lPtERjn01<=2iWBBOx zA?)c{k1=Ly6m_C6&9wY0dL9Gu-{VAV?+wM;J5%sMc#z1e%R!0ck!Vudg$mLC(aeoY z@xFcy1`T(jAu{roa-}iC+Ug+Y@-H9XuhymXd%2{2w~gF_ztdg?X)gKp8bjHsFjs#$ zCTAz&?1NQEQ@XG=-Gv-CY@nl>YYG3wpoH9Fa-R2;*daThz$8dpzYaHD)L^H>Z6J1L z6dbv4FVt@LWJ#K#aO=koTvR_5W9uF8QMfrjuQ;9?hb`gHEPudaPc>XC=3;|;Bq^zR z0Ock}(y&kI@Mp9dHtfws2f2?>d!PvOiZjSw?9mE~*Wjwi61)=j3y0{{W7glNXp>e8 zzum*It@#{|`0^S4#7;+Rk--$VT9GP(6si4@EZw;4&zJfJ^PsB%T)Lo?JxLc}Q9~7+ z%^gOzacjwSa+-(&?8}o(RJp6E64$zY6e6=EQS!oVG(GNzDwp@*QKwWiYdyoxk1JrE z2W6mE<2i2FB16`bEGTbey`VEnVDT6Jf?Jgu{0`X+MtP|?WOf_=SwPfkvy(!%Jj8-| z4Yem zdkRoLOFJVb@?)7lQDKiY6-C{liCgR`@!tgMJhGpb$oSI$@tmpT5%h5SQkd8E8UAxO zM(gEva7#46wc!mx& zp3=c*8Kx-vYb`EHNXJiJFYt_RUrJwXkM5=MSf}<#ER27Kk8;m3^7%&k$hB!%w<-Cr zSxK`41l%xQ^hw*E#A8dv-FxJ9d|N)5X583{*1GDXuyrqm|GC0G1iHg%;Rbm74#C%n z6VSjwmX4|(LWkvr_%r$iDqZOZn^G@BpQ*d?jL640wfzD=v$33aKKI9U2Or@hu1C&Y z>R{iK3@3`?aoJ{jcy;;(99GJ~Vcwx&IQSmC*cpe<)a>ZG+iq$cnMn~z3&Cy7QgG6- z$Ams#K>oY|UWqcoznh!!YHJtny`e_4HG1&bv=ln}Y8gKpsmdPh?8hebR5H)3NPJ+I zh4cA+G@6uyZg*c{$~6_rc=-#qHs#|3qp@^5Yb5{kh;b8<*&iDn4B=t~O}Q!#i~odR z!P*G45i^*J&%8(5)IV50avk~9K7{u#dcp3GG_4&ijqAUxL7U4Jc+xL{yk{Pyf}RkH z+u)1tf3q>YP?~}+Y{D3!4jtzVr>zu~no`2&1*(6zzopYCcc^OZi zp6Y@@tvu@0c~f4Qh3Nd%Mg2faXf3vbS(O*rx)T8X!d2K%k4|P?S&#qf*3tFzDl{)^ zFfHBV&wjToWCxcMju_il?D&h0q(^4_k&inz_|C$h!vaR6I5O4h7TlzGfd3Hhl2TL8 zvy%z_@S`CBf^7Fe|11kg61fq!>(0YQ7|O05HlXJDFHqLG1+7n1BQ8oqsp5HH9XFAN zFWf*wMjfYT=ZwI9&}8SX9uap@h&|%e^6EG&h z7v7VYGu3s*0foD9;TJQM%PxhJl4DR(@)+DN+l{3s{;rJu8BZ9cgE+Hd&K+5r19V&K|?T`)$rKWy5o0nJ-f zKyJ-oI9oFho-~%Sp8G@TLB>_oHxoU2w+GT|nO!t_)sHF09<;hYteY2`|4*091Sa-ro6=~98(_rDc>K3-a-xjMZg7C|0cl>(m6aH0wF67KQ zD}3qu0Ln`AaQB;5*#7+l&e$ius<-fF|my6bs=3epswbqQ>cVDK(uk|2xjs}eRAdd?dJ%w0p6*lj_Jw%3S!`$tm zpldoB+j2cwwX`?;_Ogdr?0P`A3O`V?l{^oOb*1{keKambWOp{Kp=R1g(tjS&-?icR zd0VEKga3=iK2M;<``41B{b!O zwvZGZ<@)eOwF=bsj3K&r4Tr=@QtE&2;l$v%Imzs1JqR&u2cGsh0PvomtL( z187Rh1iz|U2y_9yX!Qhs2t2q@+Co)lTj}h=zI@>BXwn(Jj%LcuCF#d)TyW~-qkqa; zS(IHvbwPnjua}TZMh@+v^W=X@l258SibFqL#Qp)**ir2-_6OW>cGeWk)EO@R?shx} zQ>gM#2$<)$Fo&PT!nAeHxHfJ&%1I~SyQ{nK%5gV5SD%d|vdZvs*eUGQ{)zte$0$xJ zp4P>kps4NAtf=9W<@%O>LJ=Q~8gD*Aq=UGB_!@*uo=V`3_!*c|s|6N1%AgSD3wQol zLtUWgzx-T@4p-&Ky4jxgUu~wsKjvhnJer~&S&`wy3^c1hgx3Ggqpj9EcFRQ%sw5I& zPI(gbi-;gn3L}l$XV_`^0p}zQr4hAJ7*7?3fS{OgK&*^IFMe-zt7#(o$S{J{@w~C2>oZGG_r+u_?NF2=?L3r_`$Qvy_go}2)Wq)gifLGyaaJbco zhUHf7Rs>1|66 zdWyb)9v@{oY*2|xc5b+6)OMV5X9+SDB|P6an+bC@rVwoAs% z-=;zI#;xG+w;m*e-Pjfd7j`?`jQy%zz;>UCAT5X?TIQ z8|PE$4O2dO{$`$S8q0&drMRiY6|tY>Pg8RINKx%G8T{)aoh`3O@6;12ON*ngF8X4> zb%@aK^?qUZ96y}Ynt^hzo*4Y?4co3Oum^toSV_!r)DNl?=a7bUT&om2X1qm1+a4Mg^1lGxC3D&5A8!E!#<&Hc>8+vfCPp$NWOm@EaH~@DaZA%tPaZp=c;|5hmF=vXXC6 zEMk2G<96eCrGy2KX;kG#b=tgBdm+C*c?<;32IZG_;@dW*sk7?}QFQjzy8QpEFgm;bBxNb*0 z+}_p19(uS6>M_oFaUO)jK`PuuYC*$cQk<2Zhtt%dW> zRY6uq5rq8u%!0pZvz339*qSeiICt=5+OY8`san@^kBe`4vCan`Jn9(kGr5jcol=Aa zcJwfDAH}KG(URcFe1WnzU+{D;H{?}t+v7Sgi*gt?P~=c#Pvn_~(OgFKopsH9jLOlA zaj#q$mASc5|NfI`_OvZToq<$VlOXO~Z&QS=B>z;Wz?a)-qfED*xI>+fQk7%)h&3yD zUQ`f&cH5qtM@#Ualt%J$WAy5~H~n57D0YhH(Fd&zntZ5{GM_xAyT)DA=sArio)q(# zDhXt>x`N6wbwx(dG}NdXj@h;{Y;y7sq5q=Mpzarmr!gHjc3;KTCqt>PxSR5w5=Z;p z?(*h+@A>hkojh(t45ceyrG;j1Xxd#po^>dY2l|Bb6n(Kfyy7&ik$Oz-CT9Ha+W)xR z?IrwNeknyrN%6OrhH+OHBJE#5qvb|Yyo@(}`8|OSY8BD6xl=Lc>lp0z&cTXRE6Hx~ zCEUC@izLP-^MoncJY!-pFLYL-_GVj=d%l(mI?iyP+ZVaio##B+-krNGPUNrUYxscA zlW6p{#WdxwC-s*x;nI?u`TdAQK5wu)JpQ4EQmdw6e?xMltcz~o{FR=3sm`-?aV6CeVWxXXLqak9cO8ZtrVBu)sJ6) zdzho|d#?DQuT|Q16@De-2ko6Rn$KTdhx2H7V}+yFZUl9Kk;o%)w%n z<(MCpg>v^_q295(=#}sa3#kcr%$A@9@9bzsohm=Pj`2P-V?{3eP5gTx2j4Zf;6?v; z6dx+bzsL^ee>W!3;$@d;VR{8^-Zhg7Jq_s5f0E=e{5AHuS&ai8YfyXaNIrMD4xe5+ zotGUyN8MS`^kd}#N(voDEvLuO>&<#1Ct03bpEBTEi(UC(YezJ^=Z7BmzTwc+)3_@! z4JY>$p!2>yq}S7kMQd!x#I6`_CCTF3&+>R+lNn9CznIeWJ*Y@F6bJZ*W1Q$b$^G>K zk0c8;V!$klFDd1Fcp4w?pU2Ctyu=(^NxJ5|ihP1&$SSCWRz0hx^2&0`clkwaCH;6= zs4Z=dpFovUTuEQO4>xv`;bl6bxp%-3k)x(bak9gR>RzKqK`d@PCH7Yzt>>@ar}5I- zkK9msCtvj{i4R?wz^$`I7UOy&4EA7nyI%z57_B6$oNc7Oa48)CZ+c=^K?^p0pa*vq z_{BIQK2ScHODrtmd21_qR{v_;cIOj5T-S+S*`cVRm4qrm1?c83&od*p^WFi=`R1Pk zcu&m^{vYIVcRgi}D8mP}sq*&GllZxzYxp_O9lYRFF|Ad+N>38h_$3`@3jMZ>E=&%g z+efwG*@{!FvJBwf8xsh;{Dd{<)v)^(6NN7?WQD2jCxw{x8Q5fPEOtyaQS#FmvUAp? z@jKs&yow&Ur12c4={0Pw_Hcglsv zMJ1NGqcent+*UL$Peeh}6YXxOp~to(LfJGa)^=}{P?3J$%xJoWS(K0p7M*Dz@3fFj zTv~<6W0UZRz=(b+N0Yaq<`(x++Cq z61e+`I9x~=Lr+)CJO0s=P)$kAVxM1#!$HcSX*z!*7;NjHgfsU-?|9&Jqn<5 zL}B^ZYMGik_!wWzc)%3i}bcm08PK<8|C4P%G zNAo}ZFt{g?ex}Bd>9n_x$RQu}yQS^!`|qb(f<>Qw9?z$kO65chRO%4j=u#%wiN}*$kcjWDz1V zgGXoJj*pGtiXo6Nn1i$49Juku&0^zNW%H6p^`PZ?#nQde%5s*vv}JU(j1aV9xMkE< zH9Te@iLQt5L&=mkkk?}feU`|BejVV>_K7I*sTAe}S!0X%gQ8Wruxq6!EL9UTC5KM3 zw(l=+#igrwOeP8UEB-+)*NSZ-OT^$_DAwIJ#$Cg^;aZkGg(xVJ#iRRpDvdl&R3>+H#*Sp-E}Of} zvR6UuFT26MWHqAemnt-^ZN#1W$s~KoN9;eYp+hSsa;4INJn4@tuef$c{K|~OEp9sa zbn#L6v3&`+Kk@*#P9N+TSO?)>av@#O4NBr>K*N6nAtWJC97l#@_1XpaV##YVxNw(t zrQfC>8^tqMbd&E14IyW}mo)g}V=CQpk1EV{spYUPh3(WL*W5|8YrtfBE%w2dZBC*a z(<14zQUuxORDkB%aB!&fhvUnOX=-s0DWtek#Az`vub#;E_{?K7evaV2w{oaEFPUrx z8Su5+q zx5lX_6Zwr(>-hT_Gr6+zWpL4Y0^uH2urf)BKOK=vvLOd4G{cwXx~?Q0tL;=hOp=z~ z8%YL^vb1iRJmx4k!mWq-?AT*1=&kw2K6%%&4(}mY_oo9AOMBq7=1#tK?mAw)SnTF( zlH=aDKhpAtEi~On;IsF4h;xC*^v0xsPxngZKUCBCD#h8&KR;f# zcOKuM*d{ufo`8DUGBElU3$N??;IaUFEXcY{=f(Y@%b@_e{?v!6q&QU_v!sa|2I7MY z64=ooi|?~-;9S55A!!1!L0=OvYvys>*d#LO9wp-R@iXyEQ!w5#uLFC9tN19qk(is}7BVd2c2`C5*){!P4kvdlp>Y1!KO- zcerk^0smHd3bF+)!k4%xVaOW7_)kV6m#IG%rnNwCpEck$VmHh;s6n;AVqfBPH4YYi za__%-;JR|rH{@pv(#Lz)_$*cM+un~dLY3(I2{pn>Z}IziJ;^mJryH|Yp~I*3xK8wf zP7e|p-H&Hd`|4$M zl6#~NZ9G*39v`>B!!ggHaMVw>(c&|I@n2%SRhmppq&7zy82d0Eq zV|13tfbbfQ=H~YyZc#coy`2SP98y86as{N8jDgh96O{fumgF;HD0E;gU-~PJckT$~ zZXX}s{;IOHWxO$(#-@T-NF%G~7cATU+_Et8oM~~|ZM|iVUjcvC+XbjSQ=TT&K-W%4>4b6^Fg zu82o+HNnv6UicDs9|n16;EzAsaOJY;_~5lNZt$1J@RkbjiF_ke>HZe{7uN`rjnkmZ zHU$cPB*Ce`wUlf-iEe2Rr1JAo=(Hdmos|Re>Pbnq(f*WB;_;jP`Ok;1EOh5rM6a=a zhP8M<(!t9%6;LUCf;~Ahg&m$>&D;)UqS5LI%t_mi=4NhqS@h5~eK)~2WqH(y)yIAp zUqSk^V?uK5OkwE3iA+h}6L+Z3#WVMfP(t*2J;+Hz%@c>k+}8s1A8`=t|LLH`$rDuW zpGNl=CD6Q5JX%RFvd8m(HK3s3BnimK>h+*oSNPTE8G&XdPNTEZK%Re_DQ(lYdALT z4@dof|52~pT1t2?@}b@gpy93Rw9QMAYQ#(8s__Ys@@F6vW^2=tE^Uh2s!3kw4nyVY zWEhry5GssyAylh|ow#+ya>3U~nE$w)#kj5 zd}g`xUb6y`rFAl11qX*U!I>Lv5Tkh;q*~-Kzi>DX8fXL-&*s7GyDQ+0!$yqxxDLN0 zo8yDaO}KTzGrTmpN@VY4;OU@rlrLP1M|OMDT;(;i`h_cHUkV4O^@&jCDFbI#zJ@QG z#J={n2I#r|Rgf^+AY3}-#N-D(#ni$q)beq~s;fDor!^n;4%r9uo6nqY#$`6OD1 z4#~wfsA?Y262Dtq4Hi4Tv;=fO3LQauDNEkw|^_FsJRIx$Bw|JNsGYvrZG6b z7z5BDz!%-Iu;5HM8$W6j1WdGtDSiB*FgcrMt%{>S#Ry7)4e)n=EaYZ%!2_uWIL783 zhWkpqM z=DU2zpP2_1DV=zC>;jlzI}5y~_Qj>!wP9Q83fL;nT{{mKqkhU`9B^m?J$6)Ldnd>X z9hx76ilBd%8{<*dZ+RY*hTO*oJ%gzFMLCvUdW;hn2GKTM3GBOKKYDoR z5-b{uXXkDcy&6$?KKMCam@7-sXGhRn?Frb-RB*$n%P?`*W7zXa9fKScaB7ei%o==| zz0iHfe&~dtV^%!&JiCik^9S&qx`teTp$$*Tt%RbMepq<*DaxP|+J1FK&)L6G;-?4A z8tp|Vv$?qAY7`!hZx!0JK8yK`Kp1OZ4i=iyco?rkpf=+D*OFvbbPKL1&c&0LU*hL) z#;A8_J`Vk_97mXCz_K0PEcY%#w1GZ*dcuQ!><)qAWJ^r$`i(Q+pQUL9lW0irbh@dw zn3~4BLAYyuhfA4sF3FQ8rBPF!$$BpK?A z7W3<4X~pJ1{G%3vE>mkTG(H3GeGSBLt!RYb!La3~It(}e#pHg>!aI85=)39-KG3Z} zJ-esqG3qy7klP3O%UWRVJCR?j>r0XCxin|OJNg_JA$TXw75omKWwlqvqmg_(?z(dl zKdP=^8{Ma}kKqI0`(RD4@fD2crX)#izcN{Qp26q4p5g-Ce|TdoL%E`n_^+cMX1@Ib`@f3r+p65Cil$YxLiUGov%j1;71a$lD&g%^Tb@cR4!iKl!^ON z=VNYM8jkhfhI^(A!;3|u@#rlZ9NHa@)-zJ^!tt~CU37&mF-XK~kx#MHpa)fRMQ*OG zDHZLDqoLzQvcgGGOgT^qLmn_zvp#@@8BB%Q(i^9@$L-G@5EG-*_q_(u3;LJO6~(w%unY4(!KB=fe6 z;#CHr+OgT#S1Sgk-<3e@>TA$0BhDI@h2h{`VR)uama^>Ufy1sW=Gb}$;+7t#(&snn z=&WW6*eo)6I|fnvD|rgF9gI2)bTBtF6@RX}iu*HPVhS5bPF;C?jr4x5ZnKihIux;2 ziUT0R^ANO&@2$aG^u+A7c-}qThjCq7vDwHI57k?xdZpJuYi@_ zAl|dBn_ik!k<`_8>`W_1)0CTNxA%^ipE!kYCDO3MyND)?O`;}sZz^nA$}3JgaI-1H zc;dur$R274y{G0v_WJ!oSic2A=3QSQVv!{d=y?If{k*|7*A`tw!AMxXE1tF(O|A<4 zX?H+h8uwuszud0L4gL+}S?ams`#6%is}|9Lq&#wtSw?gI8%>V$MBh$hHo1lzrn#F= z;-<;Uc*~qQ$s41w?@;B%wRG+s(>n~3OI|kLCuFKo)YcCpTDu?a?{uIE7!;K zuyzeT_~{e8=Xp$YVU^+$_L;atF{O@9rR{w-<1p3nxO8qWsO!k`kCNZ1XTc>hwQ8V* z`qSjsbC7Ck?Z|fOAX+_7g3k25gGGT^FgM{ac!wPT1*0LL^uZN`(_y%)dpiDz9*nXI zNjyUhO}C@QJTLi=~>R8i}J(BeTxGwDcg$Ezm37&x@YKHULb9|;7S@Pduhx< z8w&f$$o$f8d|%dz5{GV}ex{g}9czIHS9#%u(b3q)u?*{{M54T60PQ;;Ov`+CQd_4q zPO_83x&Ix7$A=}*$(Z5Czsh*bPl4RE7T4okzx@u{2wA75Ns3ccQrY_()+VPI!F~zrEd$CZ4Bp_Z2ff|GO$b zRnnJN4t<6d6_0R9VLd)bO@Vn0aiBhVGrYZ9PUCaV(t?BxDmBQ1#E=r$FLvH8i`kg5 zv&RavyoL*32Rp*VPwSyxx|O-#C}n1s-U<^ZNqWZ`ojoDH}UwCB6eVfD!9Mf0M`%ArR@%*Med5olAQ4$1?`zh zgDgyFr8-ehS=9}{6?vRI?y)j5(cNnVo{bZPthF6%WI@~(Cw$Z z>TMP`4G!SnyGN7My6-3x`3xugRzvr+kMLdZHXPaMjz6k>QGNSn3@!_!R>wW`(A0yX zM~30w95FMac@L64KF6NKdl;grPmhetVE?*0@a%1X6s;UsekBLyk9-H4O%3Tq(;Xb~ zE(gOmpWz!?qxiLR^LU>nS>&=ZQS6x?pikxPM5Dtfy4Zz2n~dciSNrjXqxY!MRtr^B zA3&Ag9oV?Elf3k!Fk`|);c%}BPU#L6VskE8zVKDVJikD;*Zzobq<1O?swkq@KQEzc zb}ah5Q)f!Lp9GtU8c;9#YQ{Ktnrnw0#t)t@f|7v&YkJ>E(igYUujUl!73V`9Mz7hx zqaQ8Wd`3}wfiXnHytFJlznTvZ%cN;nl3Dq6TkJ?1E9@Fk%#JA}vv=hYLVo)M%iO;5 z+-OW5eU8{^`EGp$%zqaw#J9VcC9EmNUtf%&^2}}aM?QeCk1;N;3x*|XVJH#59Nzc$ z7yfd3PE#(GrVc9amA|cdRgNiVRz~YchqeO@uKUK3hJ2b{il3 zE{5&86WD}`!y-3Mnm3i_TW;Qy1Zx&8V1Mk5S>w3hTy?ECf89ETg1E=9^ zYmu?FY5{B9eA~iM)sa`F&7(BOPV%+{)M>*h98 zXnrC5eRmYMS=YkuTVOe3X*aLku$Xrb2xU8+AENRg9WZ=-z~X>^Up)9*MUdEV$MU7l zLG=6h6lN@$#l|}>WYfDxGle^!*(7O^^>j;{U3TiQIO4OLzKNVtz1n19@8B$Qowo#) zS7(^7oEX7#Up=CYPZWe(w)=S7&bh?YW7x8XqjAnxb>?nd$qFV6=V58{@zw-0m=lvp z`+tuh7p;lV;jk62S4hJZ=c$&jMaQk`f)unKU~73UtPWRg(Zbszi+iZZkhZWn0>Pz& zEq6Je!}DcUZ#zse-VF!h&j|(5!|`Gg zL0Myk#aMCIv$tdi9WgvDIKMwby(#$^7-9#x^H+oUMI}qwb_LR~!VV3k?P*6@5qOw%@^W8Sod+(e86CPsh(k3`9 z;U`zj>uD@-x&U*hRkJbj{RR72z@U16CV{twuyy;fblyD}*0qVLZI?!yJ)@WqH%X{W z3KjjE_T-iri6w)c!N?6N?4D%^dwnefB6?R_?qRddc7G{ApY=D`hb1i*uio~-9}gC= zKIT4_XVgN;50+!oh(XLs%Nb)MoY}LHOxWvb%-cnl%nvyS!Fc;$+9un93u3mj&y(&W zEo~C?CQ8HZRT|W{q6~0)p855&ftWnBiJkcHLFk&~N9Gsz;GRPrLfGej*!yW6jQq7k z7~|QCv+DQ2*_WT-cW(i#^>`#`u!(||emVTN-HUOBDNI_-Y9?a{d;4S!t7?@Ew`^0_aEDb6&^TCV%Zo=tT;b4GDEOW7mpr>e!N1uu_u+QC)UELQW z`o4geyQMIs^BPkc8qF@IooCywkK>D+wE2sZ;!gKfAT7Bxof7AaB_oj!tkY&h=U%?S zonsT|*o;^@wj+sR#97d8l|X#?`zJIs9H(U&iFBg>N6o$Dw&P^@gF;&RZNi;$#gW_j_RiEhe5|rfce-I$gwg;3#W%*C>aCQ(Fxpq zQ5;w6@Z)7!``}t|8dyZ+L*dQ$tS@>STbz7chX*kG5K zQ?k4(&aCScs5vbi9-Vv$*Y#%M!xb60uWu+`c3X+-2Fud2b&pXg`3`;zy^K>9#NkBe z(^$-8xqj^ndN-|-`W$;LOmB=6?sdBg>%NDJK7@5R^4uhJZ>~nmu2LNL_&T<@$YPd` zA-evWjNdzIK>bMob4Xs!%9nc6tR6XfCw&36zRjSyZhh(L-j}$gV+lRg8cS8%b!bv^ zxww}tf{bm~;pM}pv_f>I|6QFzBqGf6CFwIgn@elr8{<6?hUHhZ)?a)JUKRoZ0tM4gH-*g=$DrW7 z8?5%6B?UK}W9+{1#Lz@}gyL=CN> zvK!~=t4S^$IHSgYSiT^B#@#)?sB!x4myUGE}3e-e0 z(Qk~vzY}ahE_c6K7|4?LF{F-9Y@?91ZW8 z=t>%p3wgG_s9ZiBwsgK@uYOPC!_!CcB^MQVNVzumJNuVRcHSZbmwcLD~LAszV8I*MoK}y<0;U(P6O^K&W7n<3*e&Od=^<4!X~#R zu>#n`RP5)oVTB`M&!fH&yf6_=?H@w)WHZZv-t(3%L$3&BHP4`P@^=U}p9vFQ%|xfw znYi`69BG9NXMda~vdP|q+1wM87&dQVeGO~byRF3->~IH7@;k6c`va73RK!rPA_(}I z$)3Bu5zftX5k7p`EEtRVthCSS?7q04N_+VSKI#ueiJB!CXpw>))3dOt=K>1PWN5Ie z2CW{W41X#*nZ5M_So2X&+(V4UQwam`-$8AhqqQFO)6#MI{!S=JIt!A1OIVruV%F)B z%jBOAgw2=y*n|b)taZ2?luaoWmdGixWt&o&q_YwBUzLWp7x$s>;dN}UL^W7!9EsNX zKhXEPHm!VPLZ{{#vONc{uxjf|tYxJdj`QpScMzE|;$7v*(XF^L#|1z6AbVwdmZ{?_ zHvF+GTnp;}meCCVqv$;RseIo!E@ZEik*ySk6wY&B*L|eShP0H3mNe0p(bE|6rY&8yS&R2VIXycq7lwo;Gb_vSQern#FmhcCTZhSK`^#jaJRI(N-Gs}p zFM`gA{UmjG7fF7;nCwXZ3H<|#ID4-WyARSJFHH@Is+1|2X~@fL|Ip8P+Y++joioc@ zFM|hG(pc3L3OXmt;QaOf@4D-mO-2$V`{ZvXax97&3>IKC_yov4wm*Bm>LB)q`C?{V z5L{3E4mWL=qMffG^Zmd8=f2o{X7%(VOlf5VbA0bolEKgRM$QBf$JIBOpukC-9ti$mweH@qc5zexNy^OE73Fa-&$4nl68e*CcgU_t-l}kKo@U$2j z+sxv~DVuZLR&0iyo>yR+!V6Hip--~k0O5(e%&cm=1D6-GSwzKj(o3thVDs|++-%ZRbw`&~R?01ZSQ36=7yMqxGL&!JyYS{QVj zgWs_QVDw!c-78pc!QHJW^Q)GjX*tZ1&Mr=|K?C&D8ECL?7N(WG;Uq09U?!@PnBmDc zWP_LiU+NdkcvP*}lkP8r8@tG*q$CC}7c_ide9$Y_N3f9{%!Yp>SsMwPR zezv)gcd#4=LW>}^VUWC&%*6xgt67FeKkT_&PRjNTFd>n=OzYK)I2po=8y2y-gMDH6 zEiV@3{$0e48D*fuZvZoG@=4#MAG{5ffjOmVWX$9yr*;3RO|~{E>Z!ed#0(S`I|HO`4uxmxLnHl2|xn6YEji3crKB$t0WkSh0U2 zc%Ka>4xt=S6-@o^Z~O)|^QO2YEV2$=UZi+ngGL&U>AqwE|3yqXyeOHZ@-;-UP+S0X-Y!OcwK;X40Ir`zVABdp$$9IiW00t$=?pM5HE^fCqN zpMY8?ToC_Q;_nwLFwMFf*17t_QGYG`wx9^Izznvg3Xp!*4Oq$UGY7b~ctKnhH5yZ( zZ?+GI@rB9geLuD=$-Ik+%36@`OF)_ST~0;b?U+>bfyHR zzPrMiw=VcDPy#hq-hhi!?^&?9FPg_HW6H(tMES#V@#dAD+Fbf~qx}jwEX115BhI=yZf|Q6p9X$IJTLy}7vHKK8EvvzjlaXk7 znGd!%ECavbc4DaGjQXI6)=%Dn#Cr|UeEOaY|5k$t{R@T?s!5FGm@Vh0-Ad+eOb*Kj z&P zAC#?3!(JstI99p~JToF8!IcM_f0RJ@xsCAXCl_yZO~+?l4UkZ+jYn&rLa@pK=n0<0 z{eKJa$-~{Koudhis|o4a?#2k-w!<^_3Gmuum?W>tVVz7Lps?X6>}_d*#;h#Zu|E*f z*i5o?s~zxcSqEQ4&*6#+4}6`s8e5-S$M-iJ&_h`kuN6i@zl0;yA0H+e$3*Go#p8Gq z>dkPjJZx8WcpwUZEn)n3MAEn{yzzi(%x{Gur>&a`{1>tTR zAW@>4cKj-)ot9Uq`;6ysl4S|xAGb!f_l+<6TrekiA-;Zp1lzv6$HifWG^6Y(r457h zl!PGn%ghZ}=hp~0paa6o|BxBwHsp_cER+6g3FFu^hcPiphsXa?7%xRW+=u}_V&yc*|3ZPssiZ*Nq>`1RPT81m($@?R4MOO`PT;qUKbu!^yeUGTKQ0*D zfC@+0Op576ocg;7`(HTYP2b<+PeN09rT^X z?Mdn-&ke7V70!+1_}<^}W3d-Gn4?NeR|e3}GuP7}i{{a9&_-O}KP1Tt(h%&H!Fgcb z%$!kB0_jz=xIsUs=+6Egns3((-X8Ig`gj^<&Z&lQ(=KvpGKtKVl%N&;ZLrZ)3tX*? zn9kBl&gn@X=2ixKRt#xjs!!{2IG)8&^mif4Jv{^k8&5GUH3c(6$h_tlKSv&2H?!5>L7JHr$1Nvvb83|4a1UmE9(EAc|p zgbBI1gzaXs{)?IAznI>+i^#5}Z(+;XZg}Z%i+yLa{fpQzyu;2JCYJR;hxap>$b123 z6>RCl>%R15L^{6Izq>1P`zF!7~o=M6@QJ%=xAQ^rRA86x{&> zTBqR2-^a|v6fYcaU|Ff#__EjsO_HJzBVh$=7~s={Uy4~;57;8JZkQ+t%X zD;kiYk|dIJdI7h420zzmX)#r|cIHa>tl)0hxR(1>t_kgWZsEY!R&2_5q6X7m@FhZBJKJSNBKA^W~D=f=egO)Km zIFnBnqx@dMr?<(_X%zrTnocPHYa#Aph9S4Olo3~+&nVGej`9JZM?6JoKtT(7*k@zx z@PEj^WCu>Z=mIId-EhLm7>=C3kGiWP@RN)-icdR1O|vYh%HMf(d{6}Kn@d46a5pR& z_9i0jokWz|Ofn1#Ac1vz?0&BUv;Q`bb@^dz58r`wa*9FUHxly89YA#16?)|7IlAR! zFn#%~3e?qKlP{(5q`5PKF4FR$y07hN+V>!A&Co+9)?Iq^?Q)u;EJ5e`|3>9cFWAo* zg{fu67$IE;-`}OcjG6$rd3z^4Z?8?`#sukB%QIwJc{S5e`-KrZbe?v``qMub{OO+T z2r9M9kNvF=rhI!Xk*8M&*I3D6%XS%V%2q+{-M}w2C#jK6dsa$sO2$!Nzfsr_=K+yw zK~Tg)xYC7sT!|lxxXmF}WNz{|X5kDA<~krvl@OzOIbFE5x)wXV!!UVt71Dxi=sqzM z)R1R~YM)oFXw}wvhbkH>v9K2I(P~f7UScRlF=s9gLtg zuTIk^Tis~4%10{q@jd;|`W^imNBa`@wM4LZJEhVa5b zaGvK4Q`UO$J$pTBPM1L=$tPf4ypiTCm7^^Mtti(e1`)Ze%P;64N!?QgXOCThbCNr307g@62k3?E2lJ;8A}{=97?Ts)9LF95DD&tfmjx)f zx>69ct3Jc!#J})Tz!;`}NrB|qUXtmkgX=&4hWP6>5IeOL^<k@%;SZig9oG)yef1{H7C6FP-L;K7C2GV~O&P!| zk(nsJaTD%g&(2Yg7lDQB42V^kfqRRd!4I2aI4i#bl( zpWar~q{hSiw4Keko~h#HuJ)Xyr3*e&9W#b55}ihWsaN2L>N`;XoDNCP9N|*`F=}L) zMw=$@Qz^}8GF$i^Q)yeo)U}5JkHv8)3ONP09AlxyLk1poCNl?*%HrCf3P9Fpo`>dS zgY8@9Jm(Xq_-zfI5FBUy2H){NktJXu&kwdo*u1~TS>!w<_>uh$`;i&|EfcrcvmomM z@*cw`U2&T2p-V^4EktKO0aVE>03McUQgz4@ChoJD)aM@==7}=nRi*$n)oV$q)_=@= zEgsJCTlM5YND|rIc7q%ZSEXI-d66nkVY2o~^kUt0_olT%_+<{=(Z8L36@mg{je*20XC91z&mj;>K2AbYZ=l>&1MDc$*)i#kxoemK}tY)uL$q z_9*GRW6Oj}-R9iM9EVfUOYl>qFCMey!5|NN?5--smhoRutHrvq_FshBXL0B%>W^|V z@pxuvA$;JEf$gt2I8l)f)f*#4(9U+5BB^w z6NQJ@;?~VAm|xL`()Bgy89D|6opsRr(VF#!heG*kdB8C<;_4bk)|IG(Tx%}WJdh#3 zGaizV?Fl4RJB9YNN75JjJZZ5$hyLfvp(F7O-4ew*M#?^*(m4gXu)mWwY22m?S0ibK zMH0F+Kt_2m&>ecQ?9Mn5L& z=PXY6u_g5WbQ7w%%8DNKUI+oL+sNB*TS%&K1*BE&Ww~tuM7!)W&MRPDO?i^^SA0KN zdnSx*{bEF>TNXoW$Z7cGnnWVxv*G3rK5)GyNyH9YG6z`>PhashVn$tAFH1iAn_>fw zbBoBA;XcMVe2_^gN`|_O5D;J0L&j>Y!8s*@oKn6=%yJJAN$ua9Ke^)MUVJ0m?mY*; zb2Xu%FcSG(cHlHE8T1I=#5%N9VO(FI9PFEoi`(V!w!Fi)O%l{%?7{~+-sWM|5 zqKTM>94R>=$+Fbm!wN=?ou#cs&&}F+rQkD|zPthtZTP-b2TC`!)_Vs1WIG1o;lvHDnc$QQpXK5uxeTt ztP1}K`K>2_+8lu6yX0Wnl6JbI;x^?wn@3l?*b1Re8Zi1<9#-Gz2L;_}*u48Mbajt_ z>cDjTx=#apcsJlx!&Nxhgjx4G)sH5pE9?JUep(zdHqC&&|#FYrK>k-JikUZ@v(b3~RbGK#hj%Cnwa(MFH378MVgvqO84?}g|u>k*uY|AZ4t zR`iO`B1#38(X=CKT!~mE?qbetZo7RNDr_mhJ7(2*9)!6=%@g$CyT9~CS|FaV^2F__ zp7`D21Q@DpgQNGYLF8s7>dq*@Yr6{YNT3oe*Plf_L)7TH%{E|kaxFBU)dVHK`@mGj zL;6T24EElEW|L}=U&``HY{KyUy=_>&#txrO=zvs)F06doMFfwnfFTV7IOZY_N-Vpu zWt$GB$BW=}T?f1^JP(~;%)>aAJ5nEF1)Y0$!o6`16zK}ZLy9RZr}7DYW;vFXDerLj zo&f%EXPNRz=@5SHGTGifn;0COPO{9csLFjp?~E;`ewR%#cT67FOc%y5tH-cC^ExEB z2_ZkzgDZ|@;=<+0=(|A)bWjYUg})Ph|5(_%G#m!~pFqaR$6#aH0Q|bhWpD#S)sz}S3Yy#JL4SIyAJ=Q~~SGk*x{f}BsR`1cU5UlIB25sMQtnRq(+ zIr?Qcuzigt$TO*hZWCvClbZuEe{aCQfg0FNn^}Kn7pSmag!T18c-~6_ql6pahEz8^ zU(pWJjvQjnE9P*v=VTa~ZB}7cRa9_JHQr%#`7Xkw@l|jt$b$I!-{ISVFYrxJ48IiT zri$vToha`w{2f5Cg-#EhIok1&*bTK>j@m0Q9tG^avMdh$3 zO#)x2Y2&}Cop@*71^5tm9@1gXv zk6A%J>~8B8xL+88XA{okgwiHrxhtDoJ*`i#tT|4lLN8O#2X){(V+3S)gmLL=L(u#y z2E$@vkow3Fean}lpQr`S05_cs_Wdw01U@`Bvj@YajCDlq0$N#E#r` zT?XBI?SMCI8#oF35kBWY;&k{e(SO*2KTi+f8^50@n&Jd0y;h*27y<_bBS2R>42DzU zpy>K(&~kKz;8h8*#`-jP-A;f%g|%R`;34RwH-oD3BN*O@tbh9eo>_4YAFUdMKRz<} z(1JmAV}IB;lmkgDAK2>WH^^@1!MyhAcwpg8GSAwX2sE&abL$x(<*NoCW(q@8;eLpH zdJ1~$d|CE(J#+kJGY8y;Il}rmP^X#}f5XS*E!BfU-#l14_Xf1z zZ-BYY4q)dz4{Ua3lB?R1Ww{U#6ZV0{S(kvfKN2N-}A&g#7!P#Hc&f?+-kzZCb=aT@_qH^`L9?B$!_0VXEKl zWO(`qNCWp3CdPipo1?#QtLI*DUhe_HyLj;V{I#%owlK6+_<*o}8-(S*g#36ptlp&v z|5DW~!*T-bs=at_^JuETO?S8AKLrfYpOu zAdpcD&|(W2oIMaYS`Eb&u^@XY8LB!S!yyd~w!=IdtUp;n$udzem(hb1c_(<$G!uVl zEI`HBXw>tN!r>roEcY?Mu6~xevsV?KPM3$ugFEQh_))szUpS2y16a(n0YsvwqqM|M zNK-rw4coY|S+@iyw+Hq#urT^>re%EK7P5;lhRk>HURbllAe_LFKIz#P74jIqj=iCfz9%3H-=eKVHIl z__~5gS22U$lLY*7oS;T7oS3*KkX_;rh}5DX;&zkmVC{5C@-amG0VPS+0>z|K;ZK_v6`D-t@9BLzV61~LWS`aiE z&w)9^Ca_en8XBzG4&UEmX6tuz(h~5KXtE42r4PmMShNY^pLaoAWjFL4EhKiuqTn_1 zA7ow_z%>pdY#y`+`FuoB^x;gb*PVm2E!b{AXF6z3RKw}`={W47jCQ&TU;o1uW z+|}rUcjKDTe!mHESfxjN_+m)o<*TqK`6^uQXoW2rTJSU71~ScmLfV!KpmZ<|3Ze@k zvc}Mhy+Q_LMmWyjX2kIL|ASH7RB!B54XH-gv8QYD2SapqrE^jA4 zmAc3QT}Sj<<^u8bl9%gf7n?m8|<~A5(vuQ@_ zDE)Q8o^HwILcwP}nB^x8p91FN)^|F{E2E2iKmGwDGXXAfWsqij0;iSu;+uuO=y$z| zTwWy%u9I4DzOo1ZOBAEW*ga>Y-)gKGHo}*$4x!SoR)B%85G<*H6{nt}t>Y**jmcA~ z`svh@WmL~9SEpP%1q@PaozAQbCIxkaDdvhh$c4pzW#y<2amZVcO7XK8* zVepjO&G!GH-$(X)$HPH9vmFclAA;4_uV8Ij2EpfKSSHT~u-kSOE=>L<($Tt5 zS0o5sO6GL4_da@UUl@&UOvB|(Ip~{m2V?D{!RZ~_{ZkKtz6<|ohYCM;P9zUktRC}+svD`Gv(*n&gSKJT5?kCWvj^(} zv+%^$EBHgW9BurXpd`KpT+QEsz&&}quv8d*8hLR?VoxswGPo5QICUU>VgTOnpN8i5K4HxF zcgTCJ3-cHomQ${R**zaX%&8u3>@EVw`)gRH`Bw}q97PdJaXMNyMh=#IBod+?` zxA_gZ)_t(dzYHIh+`y|Jub}hfQJCIf2c<0IE~c3uF0ncFBfhKPxkxqaKgxE-?ybWs z|Lvx->{NcWboBUh5A$!gVfw{SI5(T0&N$0QYor&@n;R=3>+f_- z{i=vTT8gy%4Hw%@+@ZVFp9qQN5&b3eKy(`mkKy@D8dM~}-1;W@GcAU!9#kMp?)Ne? z#^(`Rjs!`)@y>8WkIx|ffsny%B}o!*vX=OYBiZ(P2dY;};#vtF)X?r^ZYyVUlHFoB z$6k+-@~}YiA@B-We@~jOpX6du!!}s$xr4N?{Y<33-64P0zk=;E#u;OEKO*;|7v6^( zL8eb5QFs>3nHj&*FxdB-!PaAJ-rx^E^Gd{?Dc4(rS7fwsL9#ub+>*xBUf#?EZ3-bu zukS)L>z?Wq<6%9K4~Sl=6!C7(Bu*ZV?D_c`F;{2(R4e8}XHO0zAMln5es-GVxXvct zQ57U1zZpCF*t>VKJmj`~Bm(mj$mLRoT!b7t!M?+~3$5_HRuw(bG^w!*pv#^<*l?aTd#bKUxe| zYu`ilbUs+UeTbYf&LxI(PBPt=GR%F84$j1^5!kP$0Mj-K5YHv~$nkD~%M16w=O>rw zi!+lr^NKAh_8iA~;}#G#Y00{beiNs2VMM9KoapP1P`5w>D&`uG5}!kHPx}nqHY*sa z<;{rNGG(%4-7ThJwg7Jal0=T{bu+Ph#OSpJPth|s4kzmZNa$2CXJv*A^DlvUn)55s=r7&xBVMVYK(JqppEfrz0ZhEHjsS^gG6KFZgRN27Tgw>ks`b6 z%++CMxGLyR`tq)jWv_0MzOZxTg!Kc)^Upv?RKKI1R*B$1!l?_S|43{_*$v-e>1A2c+0#slW8;Xx z!WyEs(wNy*-_HphNFgE`xg_jrE*Z!jCStwDMAH2}^Dp!Xh88Lhl zWT@kRi|#U;8v8j3-8Gzb^SelgY8BDgMTut427Grb0r&#+VTO4N{#zi3!IP2@I%7I! zJNA$hTH8s>zx_0SvL91Z_M`slS4gx2;A!`B@S>4eeKh~g+e zsOEG}OGeR-80=)v1RJ(DV8+HqG)u0;qw!|&v3-Jg)Ukaeb~maXe+Fwx;w12 zOvgpr^KfDAFy7MOqJND7%C6@@-u&sHgep!6IsZc*5`9DH?;U5TWMUw-JrzaV*6R`f#vaDfs~Vs6WwE`^P^{gr zfa%93!T!xKIQvcmH`=*5m}1hnpet z97da+#nj_w7JXJ(NY_Y&pvA8coF*KMr?*YAaX7btTla2 z3R<3$)jukU{)`XIN#1hiML{$Z;6D@fN_lZ8x&>|e-+k?etLr`e#f2hlLW^d19-x5-G&^7Zn%fJV?WGMkAPlUkZ#xxo;G>$(j zzTg>w5QvCe1b(e5AkSk95>Z*?;8R&p>Uob6-fg(3ybY&ndq~A)*2VC922_P2XnOdO zlrVX)s&a;+Y6W>;O*Q zX(Vu)ABkNQN520G#2j```LS>Z3anIwf!U2C^Bno%FZa z0ZOeCSr74fOxL-OwP$@uK+zK7()pS>B)k>mDy_lkX~RC_3p zzEnC+*SC1kf_62uKKBt8S{A^NFQp$ji|H@mQgiRCC|MegFC4wGDkKe_v!7El@)p#2 zzk-V^_rbD~-YSPH^ zc>UNNxRM&XuW$F_R5^Zu_IxdPy=*Tus9B=(kUgGp+K9P_vk$KBY9Q|K2 z=brd5mnxE(R--5~&?HW_KSxsVqYPP_1MXaS9edwrLt-xboO^_mUmpiyT9Oy*p%d$9%v)#71) z79Vu;ejx|=3wm`WgJ9DY$!>!lCV8S{YN2FPolZzid_27a2yN=|PL^P3I`3K}z zJ%b19>w$mz8;H0Y4_nSh!TL=s>(+A@X>)QSTK;O-ew%fN(MC23_ zDC3PE8fbp27?e!SAo6Aa30z#s42f7lGbftV-LNC37i3|X;cNiN1*yj_P%fYgbDQQu zA$Q?@~4lL0ish|2d87=pmRk7EU(=J{2liR-}8OM+ruB;S;@e)YhlFmM+wm! z%qAwXBCuxVB?wJD2m|K!V82-gq-*=Mz!$^*}tpf9JpTYP2eQ?md0^1I- z=9jBBa5K~zvh)uU_ZS;SDpCNtn=T%1=}g3XadsK)lH3e|OC`PsQ}akDZ#V>OrBY&u3e;u=WIMLC!$ zc^d4qw9&$38*cdOhXG?7iNqUw;+m8IiR(RZylW};%-2QJcgOIRJHeqj*7$QgkGXRx zhPk%ZnDk$$B+<5aNyzcV(L66|s|cW4DwboF*=3ZvZb*kSO3>tDA$~3| zNAZJ~piem(N_Vo}PMwDcA8(=Z?ih3nk0E`gj~M&S_t@UCABw!*h9P~Ds9|Rech9XO zGU_r=xLF#-OU2N1hbelyZloKschLHGPV@?o4~ERMz)wOfPvw*f>nm5ng5TPhmCm}+ z{+mLZRg367w})iW=X|oKWj}GJ3Fub38-K@$Vek+SeXli&LE#gqo3{n!ISx22ZaunB zwt?ULT)1!J1lr-xIh}lGIdYM5FxK;)yu5Lrh~B$Hi3k=+5$ZKKhik;7Ag zlDH*nA?~Tp0q%`l=r+oSg_k_gQ*<*5Zg<5+GvkR>TQyVZSxM^nEb+d;cGPxSg^o%? zpkFWr<>5~NEO_x7%eac&YJqvPCFwimnRJCYFP)s@z+75zozw19!K8^_fCDcxq45Ii z{aX?MQqw&^_w!lMOgf03$$mIFkB2V%5eEDTzOaMCi!G7$PtB-?p|ts*pc`~eBY8Wj z5AX9vNOgoBJh>|0 z1VRp^lg&GM;GXUuvSf;vIXG0wq2mh7DwdgIzWO@Jzx&T{e^E7OZ;3f0CoRHu=r0mE z5q@|w%0p|S#<8;b9a~dcfFCMkQ6OFhJwGy3>CX(hr)(NMBP)!58YQs9VQ6ZY` z`vX_B8^gVZ9x@r$!e;F@vuqxgN3&j>=BT=o{kFkGF(Q}nxShbUv=p3D{}0!gu|2W( z^WoN!gKQ_X3;w-t1OI;oP!mwcc(bfxPMbP&TA_w9_g!ajlM`on-N+n#_bh>$wFn1f zfcB0oq={{^^yw8N`Y}|C{_2`Z->zLpOI-A+Yo-jXY$~GjjC1LVS-G^@XdSk=ushk< z&mfX5%GGM@p^^VpQpdU@n0;#p-gVPQyDm+p!0R6K>d!aExFr<5_ITl9k#Jmd)D-zn ziero0W2lwpGESJw!M6%bBz4sx&dg8I9Is=t%xkBe z=<>-3BH9$#j^YbC)mTm2i?eCq#dMh6Uk-v_Q()Zm6#bdBk;>VVn=8^7{zkcpU|-G2C8OILpQVc2N~xO2zjNPc3BcG!ENB&_K$(ySC=J`P>`s=E5wRY7e2$`y&q{7%+bOQ>q6qFRMINqHfdY4{gD&^t zP&B??KJ5`3eB9pix!V~bE6z=xI*vsxbq!C@VSB| zdR-Gip%rW{>JR^q-dQ@cs03O!D+2 zP5Nb|*3k|={<;a%;@NjY!A#tkYz{9{Cs`-me`xUPC>&bF3mbn6kW9hlV0?Wp$XweA zwF+19z&C%)-jj^$Uw)**cP40-x*+#?^S6gxUW0`ZZ@dk%J?COnBNTU z`{Id|z(LY$!20Q0<8ZumGphD4!-P|FpzMP?{P*r1bRAfUUXQ)7e#?3M9I^o?|2gCO zr5A7~yH7khS_IGjX`yMfJMFQEp(58p>Gtp(`qQh4_A3t1_;3z*q-8<e4~e3SB6OVlOZ-z7!7TS!sAv6C7Z*#@BL-^p^N=!)u;rySf7;NcH%H^mEDB|gMZ>_DCJuK^Sw92&-_4m#dC+&OUKZ6e^*j3mqzZwhxObOYj1GpY`6p) zL>Hpwa1s89ETpDoE_An;7VV2HqNff&rH{q%LQAwI7&-p$7KL0&;t-CCL-LsbE z?!7}4qU z5klzwII!nE0D%&fALUgpU z1Zb!+gNNOvVU_v^p#J3!=-&DYdChx`WJ;f)+NE94ag79eobQF^Hg1F&C(PjQGDEod zjuxE9o-jWR`|>Wp9#tVUvqcge|0;}fl5at?;tFWE zKLyHV6@$xbR)K*R9KT|mqU?6fr3aHdK|A~g)I(l^uYq^CoY)<-v+O>ar*j|KgfB;q z`nD)fZ!!9x1H!>54ZgUV0v)B&uuNrqicEV7@S0oI+l3}(74g0aDe;mF5y2nw&l zA6aG4r>PaLPZdVCf$O12i6sp7(}vvS9B$PRhC&%KFur{U(k|YQ2JMd^o3AhaZx8;U zZyG#NJPaz!E5OK>!=PlPIbuAmkcQJ5^u7^*3mg{@sVjmtQ#R66qaKy?#E824cZiu?n}Gf)+4fR7(&qnwv>kZPS8I(pd_Mnt;8 zGI>2Xt<4B27~9bgXPrrc@j^Q}9+oeh z^4JJH>()RE+J(@Gsf8fq^MP*Fc4+pa6rP%|iQXr4!;5E4 z5s=MBqboKb-#!PVG-iN~W}k;zIlVC8{%rInM+Jo(vqB@kTu`ip8Wf&p2Inksg4*M}PSP8fOwzYj{Gh$qckru23&)?Wg){iy=yuf>x@GVVoxwSqc+=a!9HEOqNBBLQ zNB2R|HSKWfeQgw|WQ?q{7oy;#GVq}c0!N8=bjce#T6t3qCExs=D!1-1FMqU`I;OFU z-lp<|^K{>(mzX@JD=TJlK1E^ZBPIm>L%N~p$U|7gc^6gXxy)6uIkYX+gx_+-P|2dN zaCPEyxJTtSn04VYIC$$KSak3n|7+4_6-Ui*o(I>Y?K(wk!etni`4Ro!H`Q!VXe?9U{;q8OauaO@~;Tkv%7}QFHWWX z^4e+XJ2G@u%x-Ez^#GV}5Cz2i_k->y6mtAZBVAz=w6*62oK=4ve&2E)ro)eMg%eXx!wJJ=SeGqE z*XCZM>^HI0{Vs0q@biA%85Xl*&bB=wWCl!j}v-DhOx3>Tuyz(D7K6(V?W|V-7LDPVT*fF|H zdmb&GN8r?%weT31f#9FzQ6)>{DIVvy(A;>9E`QrY^=@yVZfZ{lR8c*x%-toJvL^IZ zhd;cYi)VO4afRTzpBP|%i-Vy?TdKOQjXKjaMrm5?1GA)ef|CO~faRx9&KqBk$S5XaaTc!qW>NbOa5fFSCUkq9g1IJPfMjW`z_=RQAGAPKI_HEAoYN`~J;pdDyv0vgmHZNpJnwM0yOT3LM@htF*-C@(tZJGBaV%*wT^e6V`sW> z-lg03r=b1Z^FQF6riv~Z@Wx;i97zv>V|E50TWAMZtl0~G*=|PT&;k|9m?4?$JLt%x zYiRQ4d89I19lDMtLQ4?_*1m873zKs|*rygSM{6d%u1X6OCkg}oqsQovUv24a5p8s3 zM;iQe$^kAOTLDKrPov+3MM&dADY7Zo0IP#Mz&v0BySiJ@{AuU8<*O1cn$&{trmH~l znM>gtISaH(RUeHStwAPVqM$&>76y=`&^=KP_zQc3@bN^TJZ_3~3XPHYqnT*coF*#c zKX*FQJ(r%DxC6be-h#f*TZ?8H)}XD8WoT>AX;gAk0N&YN0sfP{pmft0^fq7=CI1;m zl~zBI`|EM^D*G9_H8Mh<+kPCZz3~9JjLpVl0gAX=X*!PGVoI0i7tyXYVRRQaGqj(j zhLjBdz>|t{FevI5=op*{*E_8O7BMxzN?RVz+f@MSU(bY73)CSv8UJ7?G>LW> zsNs??ci^iJ_hI?!spz@)?%_y=xIl z{3(v+zY{_3>-V9`f*mNqa4XW|@+_9i!oXS1o9TTb4_w@P8~AZvnJiWam=2qOyp^Tk z%he)aJ8J^$Z}Eh1$~mw!EgwAmejZegC(tER2I$%-bFldBDKHxJ6c{>`1H<+zbf7K| zbw0{K&(f7au;3tw0JlM?bQyG-&cdFYKd}9>CQ5ajfy7L>9@A)^yio3LUX> zJM?O^fjb`hz_Hb0$f!aBZJj$EtzIYuwtrd8dAN#zNah%9=y(qOSyA-D#2*;sT5(Rk zZa`0c3-beBz{2?#V1*aLswxX{T8ARG9iED||7U;_1plEaMNjGHlM!@B)noe6rYN94 zV<&iSp9>t;AA=)>JK@@dcxbn8HL{g;MD+u1sMP-rTx2YQ1`B2)v!%Oenb`oyJmwrC zZbxa09s}Cwdp7-A^$2SD^*^S9H%b}Xh4bu};5h*XxZsd6ycg%kxtwz0IbT(*peli% zeU-!}R}zt8Zva{{vH^YC^$d7%+)AM-!=Uzy671@Y0_#R6=vS#uyl;A@T+ptUs!uC1 zKgl2}K&6oKoZAFCRX85Y=mxkq$P~uhiG(&+DqsiqU?r(eV8r_fxSSLUea|PuJy)vW zA^SSYqDPIoYIBNmoADL!6;$8`O($6769=|Dv;{KjL%^7`FI4>K1KTrvVdOw2@V+Sn z&mVGzQXgr`s9J>HHKU5sJU3~U=X;U5Iqw{$w>OgG4A+8Ba<723d?~nT)&yEKTY!e$ z9@rh339TOH!P2ww-2T51n*O=~=iW*L&Fzt(6mV_%s3#dD^# z!MpRcm(F^S^o!8{UALgWaBSd>a&PGF%hSOz!)T7HU;;tnahNux9geP92xVs|!vy3fXqE_0kbcmiweWQEM6$9LL4}?no=JL!wu*%IJ zs%%Ptc57$CRudzre9i^R>pKCrH;F)e`2e^!QAeF|dq9=v>rop*EWosA9bjue8w~uK zjm{o2MF&Rh(3IG-;MU-Gpk;3X)2BbBH5+^Bn<9;zTZuxV*J6?NWCeP-VFs-KqykTy zdcoe68))5BZMvpc4E*>p7d%~L4wlzsf}%YOY5k!@`knR=y=*KBq@O1eV#hxVaG=}qCxv9e9tL1c)lR5Zop94nztpm3t%0Uh-0qZ_Zhj+WZ;E%a& za9m%JvVTy*>r?GFCt<(n@P*po9_N7oc8RpoyL>vRM;!QV&xegeCGgO>Y8ZZ17OdeM z8`&K9anIXW@cx1rJoIfg^!jNBT7GT@k{k8_!_sK#7sue6M6EQv%O04>a}=$c{orMo zI!slzhHTm%xbXCG+N;SH*lgvxjkBe}%!_A1pZYX-zd{FYSmF#HoehA0t0L&iKWX$j z=f||$R2j~@>=?_4Lm^C?%lL_=V^Mv zvXQ=~^Mzi2iOU3jGy{3PH^70(;~<;s>xWmo2W<^b5IxF(@7u3asdpDsFaG}I#apYw zTZVk7_1Yg!d6_{k)|d&-?l%TwBcj;!vj~nn`vn>7+E4qjC+Nw`PiXN7VYtokEqHzH zF_?6n3eW2UH{Td+_&NtH zOE3hl4pwus=OlVYhYmIL3V{hL2~cTvg?H53mZ#xb$5V=u1MulTI<|rjRx=yX{?t_{ za{p4a;zI+d$%qG1X)8d1YB;j2c0nn^wkYKUAI|1Do%^(=!9LD0aqeyvT(__oJ`}$J zk5(nYic3*2Y`+41RzZUPq{Q3@pmL1uk4X4$iccp(61TbT%mu@$3^(o>?j~iONF3W}K5{ zgAHh^7^D9rN#M?`uV{JeGqid5Ep@w5oxWproO-k>7v_IqVfn?IFtyJRDF`WP66>?B2fhH#o!=9VYaNzV3 zym!D9r@dW(ReZByYj^>~hY5^O=G}DJJg)25!3X;;t^lD2030%N%BReDz@lJx+4AUzv^Q&lM<^V0z6hlh|J)pjQQm5^sj?NxXgQB%;QaL*@ND-~l;Lc-^lR&6Q!0S)c&f2CJfUGexAa zQ3at_4fNIVsr2m=8|m?x#;`(j3am5`fr7?y+P64EYEi zi;1_l1ykicu;=KpB zL=<^tXYDbQ~YuL1?(8-d*{u2=e{8a$V*2j$I$;Jr)@TvU+@MUvv6k8Lze zc^e1sqjk{tD#x@Ef}qx67LX0ofULR%Z20~f9IEL7#pm*X=^7qb(8J|Q)&|4v@;hK& zU=~=nZWq|$YXcHqZ-X}HcEXK9%J6Q@QgCOR7C5cgLnp{PKm*5Z@cGvc?zODAntfM_?y>B zSLX|Xxy7&PZ4$G<%R^sivEdW6@b5G@x+5E&04i|T%K30XqZ!co!|rX z`;K(5tU@15`#nq_5>B83K8#@LKZlfz6Nv#-vc+k)qroS;=rQ+nt%^U z2V!q8gEu$7Lxsf8u=mv^=%LK@icL5_1Lxoy`oi(D!Y#ppd1_ooU=$qZ&aH!61YqeO z&Kv6=118HXK-o|&ERVkc)pLvC!vErcuId?(>Tn9|^|ggN&3Vw^umtQdFovODJm83B zIgIM&GW@fb!_6KeU?Aofo$|PaK2dUt9(wy2{F!8d7LEZG=BuG}{Cb$J;0P~reZ;Q* zGCJgzDm~Dh4S#!k!;IVou*by;y#J&LMs7^d9)=zuB8CE&;=j^$%t0Xfg4<7rR@48& z6yX!Y67XIv4m3&ULT{@*@Zy*;Onq~dz{fgoY9ReCC_3E=`@MB2 z%{D&AT8f9aUX_|YolvH%hu6|Udm^aO4>9mrvNz%%69&_zD(FL%xgfgqrFlal=WYws zhQj^oRI|!mAgmS&-z^!Tj;4O0W>)V9`#8_sh-?OI^*@N1seL>v-2=ckVg(E}y~Mli zx1I`b$)GO_%{NO^b)`c3OX)wIveZym22k?Qf!DLI(VH3)K=QRiuvGgr?~7U`{c4Qk zEJUSIL(R@`?EYfZdR}0zA|y(`?F{11R~I<9bUXBUk&kYSUgULU4^S7)M4-5H8uh4a z4&6Pp7Z`|N<;8WaqFyd+19k&))JRP~J@J$U4wvpxf65c74~11gyZ!>$<{$t$oHMtf znlU%^&!G2?ioIRawIbM8JIK|h=kEv#`#owQYzJ1RRt2lB+cJ)jN?nzI&fs^LEgU&7wJ)rU%)cs zF*Q~694I(*2r8>hp~VY&z=bzc;m3uA=8maC^z?Q&aN(^cHT>cvShQ9Vz3R^6NtxfJ zGB+wfNo_@HWoj{Xx!4)}=k%8v=VtY98#bZzgNf9Gk>&83+#2MWv7Bx*od;VTlo4z0 zPp^Ha!rgla!{$Q)yeUlu)Tq;2`a`BRPf~^v~_!Ir|f2p8ROu6cI-! zfXl#q@FMS0nI7%gNrQh1cDzM5+Ni~f-oV<;kC!EqMf=TAf(5&-*ujC ze)U{F^*MWj{`zP;_!#*p}U?}(Io(y4@=?Bf8&&VTqMjn zy#-ly>G9H_2-taR2SgS3%}q`XP}WkhwCodG-f)dH{h1dGZiVt6w4r3{<29nMGT+Z-CwsJOeI#5X|dSkD`5ab^~+vJx@iS%a=CE zqOsJ!ROB~nc-KG_trSt^Io#F(k0dbAu*&0&bO^zS%bjq0sx7t3C7w=SyA_E1RpOb5 ztfSeoeh~G%7Z^={2reG}2KfALp7}Qg5bSpo%v&SHYjdrqGn~0jzP~KB>aH>{=XjbA z#ebPA9Ll2U<>$a(_gKnVkOKS+BH=zUN1krjP3pl_jw48dh=CDk+7mKHdpV_N4zX2W-mv#&dBA00|EBp|sjE zYI9!~Fwrc7KMEB9vQUO;FDv1LJ+{2VYBGR^9J}eBJViwE=%FHY@MWZzlDJ_2GgNF) zn`{ylr;rA2>2$%)tHHd>TWTq1Is|;|T1$Hb27nMdA@~TGQ_^G8=*I8GU`@9?)z_s0 zYbPt=s^t=r=hh596q+k6^uPD2<)UIjLXwg~|IbTV2n%fr+_HT~@K$5rP5U<)o14-m z6gMs6ncCUVLhJwkUi$s#(THp=YAm^euS!hAzI}B#?9nR3n>(Kk_WVO0S&q;J-StF# zcqvxQe+e^<8&Or*9lZOHD1Udo4zdN-Xz=U?{K8lZzoCur;y*R)*We_wNtyHAEJw(* zWhe2t)XV+uf&bHtz*fyKa26ebjWoxR^1mrKDkcgaNgG2E@`_l^(iN>)h*3h>VsyM; z7EOt}#N-BtG0|P3jK5bjdL@5?oZn)MzFI#4rxq*VK|N)>$NCih;F^xrjPBxpr{5!k zdzt8M$sX(xv=u44viQdCpLpeNEo>=gf@>|8;Q1+b=-TjY)Uvo1A33do*UfIi2KC>_ z6y2R@d5$(-ozzIaF6Qp#tG=NlJJzz#ynV=lNMZIxwHlG1l8)>HQ}KA3EXcqAfHd3B zL(A*FAXyV5_R|g(+$(FrmV^oP*6HgHAKB^Syoe**i7<57}1q(*6O?eJRTS zLCK)Lc@;!H(5Ycg^9^*i=>VgU^nxXiKErJX&G5<&3wZOjD_+zYh|^!^<4dtiIoFyv zK35cizwD8OpJv2kcXt7v+-Zt-dZv@hU$RNUeHX0W{U5eYcf=c(CNT#prlXnB$4SRf zA4)Ho0uHt15bY-)N&ks3R0fQ3O?wOyuV6^!g~ROZi0Me=vn5l1@Gag{-HXnxX~8oG z9^$U8j~Q#vZRFp&8>n%M1b*8x13NsPgDgw+k-Ee!l-|`S*!t%nUMYGJ30B(Rt?kKZ ziOxM__F5V_ooYh2p8bT&H(Vi?Ml&(+>cYQYq~PiPwq&~Y5;Seib+pVphUvK>f%|5e zvMImZ8Dr0KGCd%h7-cBpm3Ehqp@cZI=JFT{^-Um)+@08qsTYWy0k_|eyw7~Oxerwj zsNuPWC$XW~Q_{HgDKX}EL77E!;1AnqOgT*tCu)&x z5q0bfS8WW}cjFzV*O9~g{lve0K9a6%1+&+_BT2tQ8IuNMw!60lZI~;Ja}N81GD#ae zBhwo%Dyl%Er%Yk?!3w--UQFZHJ#j>3yc|jNEM*PlvXOnZFqZOI%%a#S_|W6C=xDk+ zf6KuJlzMMDKC9n@FI}Lpg9*g$L-VojwCz~&%qO((P#sqDIgRCfS7V<1Rd^~%8Lc$V zMFl=*aLSQ*a*8oTF$aCX`bsO*J!}n;jux6UYh(&%<$zK1lVE~p#+Isxu!`|QFeBLp zVi7YU^hFrWaa_=-+hNt%yI4xFcu5Po+f*;uU7ifp%1)9Sj{}I4T?+!^o=EeTj$p7Z z5}b{_&xCc=gQ3;YbdZ%g9<?DPlk)Lnwgma4eXx5BC`AR5$1#KWoQ$V0&;V_VRnorR@g3uQvURz zNbMOgahX5T(wl~UY{+60q~e&heJ6k}(Ii3X0gSCm66LmP6P)?VP4J)d27&bK9uWIW zkwkJV_+nu>taPjgb*WxKnb~nfQDiaGGIUk&#`Plde4bA{JI;V94)@5m-yLvQbSLx? zT7hxlaad7)8eX`46fc`v$2`d&W`17axZ=HW&`By1h!(XnZn8mw>wB$%_~3qaLFF#8 z+EE8e>#Cuwl+S_EjF%!f+Y~xtcktDKPoCw|gO^hX15ZqA5Ttc;il-Ua8BgfdW z`T7Cr^u9+UMv2Tw_*^#iqXx;jet`MoEeY>WGhr%7$kL~e2BJbRPb32=+eV<1XBH6s6I+=BQxm{#ds~5^RSN3wy99gL zWa@0>H;!MYhx!WanByNZ$x@|SfpnM;ak9*x^06RoEo99_eSvD zF9;+bI*4yiG{C>FOyS@01?Os?26l`K)!SL3F7{7Az9F0 zgk3#GiQZpt@O-)%9*#^SsrrA}6DI=5;h&aRL*EHSzH&ip#O=9tWIj=$MVprHI0!97 zby3}scQ|~q9ougUWm^IhncK^1$(eJGXw~{3NM*rK!EW7^Xm9@xe5Yds)*P)BD5vHN zI+i8E9sL=Mr=>j6PS9ifH6uydDg##k$}Ifleh!{tcMZRfSqe|9y1>NYS6H0g&YrRx zXZ3t$vgJ=kV3Vr>Q5&2BEp_AAuJ zn4^Dk$l)31xnBQO6l2*>t_n}0_b2O#W6U&GGEl8)u9|#fn0f@`;4w^$Mouyb!EH?1 zyG(pAEskwEbe<_caE#p*cYs;Ub7{;u?+uUj%!ig0`|#G?Mfh336@ia)BDqsbv3r}{ z;oHX>*-T|&wtu++NxASFg)CUke^H=M4psbSoA)i?uQ(sV%3OTJW*>85cS`i)n(!RK zd&8F`c}5+KcJ3DBERrI78a0XLwl%Ey(zjrblLk&KyoM^iy+l5rrjcx=0V&9x zA{MXbpfH6Hf_L5L+>`d?3YW>fSN{=BHfSMDmD}jW1}(BFQ;vjqOd(5BPjT$d3Z^z; z9iv-hf!)uXCi=#0WRZp+Hv5@{ehlcM=I%p8dgl;>=H)Y^R{{w5{(-TZ+{0Isacp#6 z&`plidtgteN6cl*H~^=Mqh~AE;2Simi8=Q`k6je|nD-GWTFxR@pQ&NT zeX9KM?3wJDXEo%hj4WHACrJ{AeULc6pG+tSHYW^F|hYZF?W8f8sT0)CyQNy_0OjXWb^-8<6NP z?O+#pKF11j(X?C{$4T<`BI}b|QJqQ!3aoj68g_DL7`e^p$L~TMR-nq_4kLE*xgu8m z{1dHTBueb6La>E%ShB zts0c>YO-HbM6p)9J1LH;W|p@EGotOucc4gdeG#okR^nsGOJ5tCO+Sr&b2QkibywK0UJh_0M|V(BIOB_CSNE)^v6|D`d&A5MX86B9+id74Ts^4 z?Vp)C^qIXn?G?`9%kyVB)uA))r&yo`(j+kLR<<(6l-D>2HXPr5P@`FXN>IVbPH86lm z_mm4hlxLtU-B>7HeVl}j$zdKj&XhNV;QTq`$f8FQ$1dJZ_U?8hzE085v^AZqX{}&J zhB(Jcu_Dx8Igb(jZqXRCJ0AJ;Y(ls>7QMfe41_EsnTj)Rj5u4|*t1)aS$;X3NftO_ zxusj!=>rOApTb{sEnk*@VA6%(c{h^mb9#rC{X2;Up1mO_|7D|??XQ7VZ74h1W{r)s za`DoK51G;3H<<*Tbf(r-0?8cGX~awS!h|0}O`I131Gijk+i8isms>EguPl-0*4s$0 zUlmW+JV>5xI|FCWvq8<;nz-|w9J1VD49*9=ZkUunn7VTrja}42hgY9R$3y$ci%^#7 z`tky}*Pnyz=Gm~v)*D`0*pIZ1oWVZZW0}#SeMI)wBSz)!29ndJfi{pBl=XTc`zS(- z4R2CpGfc9Xb={9Bg%#7_4(1xX-u)9zUVn#*_nCsw*ZJf&qs$)jGeE7iQONkkb@KIz zJ0ti!8+Q#+I8xyesS8(UQ#WA}q4|;Bw8En)^N2h9euqBWYd5b+t;h}!a(}yx5q8-7 zyb%9>-eL5h?<#6au^^dm+ZzlH0!W|w3F~CUki*j=^jj(s_x}k)-c%l!WqOCAQ;(p2 zB_+XAHwmn#=Y-Z-{)J~YOXD}yBB)9?M9{P3Br{chIcw(m0sUThSnwf2xoIeFh`swf ziX5>XWhTxF;alzI_(}N;Ovn18g|TJCXtxNn984$91`OI(CPxb7$Iyk6srZmnG&^TW z5nE(ejWn*bGdHriF7MuA=DEopX5qkQQnIO)nAZRtsBl;CG|HCj-Smp!HPMo+$0(H(Yw;5;NnOI}v+biFiXJWcFiu^tH(Ztu>Iv^D3)I|7v4) z<2q}0^|CZFl5_!1v|HoQ!fojLzhhXbtDPvVI!m7K(ji+nZD#l`4Zx<&0p2U!j=6K4HvaHkjo3) zScg5?+@bjqGYwWS24QE3W+%XF9`INrg-r5z!)bK0w3^ZW)J5hWk0L4&j-<=)5_2dh zggNS|%%Z8H*!gQM+A`*eKE8Kl(&Gr1Co^JK-E4v374obJ|28|CsmZjT2!;H8ZRBMB zZpLBmZtVPU5#F6BK(G9*k>0#|Qf|S6e(x`%=X)p6$&W8kcA+S?ZIfdA#dHLM(#?#M zc{SQu`HNhzzkt&AUE%rwzu>=^MwC9=21UB3!yXZDl$=?OUmJRm9=}wgB;!s5hP#mb zoe}iv;R19G$T9YC3KH|o2eNuwkWE7z7C{4;3ZKO!ba-H&m&s^*s|jXCuTzU1t5MUb z!?;yvD*s5f9)Bq_gE?^59xoeSf#*E^#mG2e_T}?2cA?{aXgr$2d{wd$ymZcmzpwkC zmcjcZ{?Q2f=n;g>ZYE-(^};Maw3FGk>o{rMJ}J<%+(n8Xm63zT+TriR2{3EF6ff#kkIQ!s5w~4JbzQo?fn^Ms?tf&JDecU=NNs5#@YnQN|#BIPXWw+vxq#H zT!eeh%_aiwv(lbu!VbGMT4>mcn|aHT#=Aj`$1fq=*>Vl_HmI^T4R4vNvZ>_rw56nV zwJxb>OeW*IMM-0aHH>{2EtuteoAKhhzjG9>qCKB}z()(51%rRfNfDRrJn}jgf1aTa^YIr76L zxiV?+&-VmL+LjM(3ZIea&y}R**a5+;`KhSPuml|wN`iMnjZmjmFFRyR;pQAoJUOn- zX8LpNvI7-pI+q`1OhO6RWc4X@|v@<)?z6q2E9>Tdg z(+DdRE;y?&4))JG&xrHpkwGG0?yNb?{Ca@N_p`<1#o&BFw#9P6=1w#6OtM(8j$hYM zWr@iS*J?83$0Z`+c$1r7ePgyCJwcA`-AN8yZ5Q|&x-iF|O@OJl+Q5R3h#b(qh^%sN zGBNJ*I;hm!;dtWfv^EV=+hOAn_pRiVb0HU zTD}7oS|l`lizTglu85bf>da}*c|OO<~SHBgG^uDJEnO1PQl`uNXACwI=Ezai>bfs${c!mjd53q zVjA4Wn1`C#q{gUHB`ZbzjqW`*T=n=6a$_xy>VsW*_#VY{gmcL-R+$MC0h%^+)Qn}{oy~s^tZJz^XhtmB0Yui z^s8yO?qtsZVnxjK{~`a?-OSe0x5ys{8P@0E7}34MbwJ1Cq3^8wOwV2+vZC-a*b#Y< ziPo6Mod5oejvOpw25}HmXKK$_WNa7gU3^ZE+93sf;+2@(t#iQUU&7=j=aSvtHBQnE z2T1l@6{gj_8>IFfL6f(-NLFDxlPUWEj_CXnC{N5Je`U^+IU#0@QcyB^>+it0XWe8b zg&)KB6anrx9+Lc`R8a6vk=&ZWGA%iX;IH?Y_?07+k-aoMmb0IHja$Nog=n!}Ul);E zW!&00rbDc21t?ryOmKhNcUmaoDqOEOo4k93q1bv&w6%PF<6p&P%%I#}#`A53z}Jr@ z69Xb_(@cd%&#tNPlye$w^6g`zOp=p8`QHcdKAbSF2Uue7=E1m=-9NQ2)cVzOceJ6hhyJZmz5o?M?&bCm?MWc_(2ubjKXkQyex4_+sQ z)2Fh>7M2R$NWBv*Y1&LYT{pne*j{osFqVvuN052 z%lzue29+*%nUQ_pnfda2;VIcojN(&ycz?K`b`&mV40pdGM?R@Bn?HVJK%h0>-!%`} zzwtmL#*>$FgIvKR*Imh$9T6mN+auWeV`*cXRW3@|I{o7BR}bLt-Y9gdr3~+?e~700 zBw;U!6udP&5(kBUM;j%rkf?V#^iH`0v!-5$!>>h=;H@aXzr7L7*W8TuC=y&*Js+)7 zR6>h4HSwPu_=jE3relu>`$1qZp~F(b;K6k~Px8C5^N)dif?nw;ot z%xQf48F%pOqT~2Vs1sjy)lVF45`v!}oQ>68dIi=t>de&{)dKeTXQU9Z8hQUT!IBEe zIQTB-A-q-0H|(2-3q&8|KebJGz*rxR?UF~&jKy*6lP=$8INNl-Q0Aq_2l` zP%6~Dc@Fi9l>&*4+NfsnUG5$%0~Bn?hgruDqG@T_n4FMC=fv}XtT9HkgDU8KFbg#s zq~d!`oOgC@IY383aOIE?KWRfV{uV9_XKMvO-GUgwoBuLkS+oYM6qyC$2E6#~#?2@+ zGZBw>neta=s_^@bjriH56Ll_^LEl%cMsL2FGn>wx06f1DcwvJiig|nuT2#D*VpeDH z*S`xIqLd$jy=^o3v3vUPO1T#75iiY;iIe2>emuq%uAHlG7#c>PfV0!#Cs(RXw;H-=IylW(m zDiv*f_@8~lsF=23^LQ&fyg1^LOpeN>kB_(D!n#Z7BzNBXX?Fv6<|Z&@^83(j@i=}` z+bn)f|2zCM;~Y*^@x;RSO3=CNeEz?~AF$usdVK7-CBIhvBmD4AjAO!T;yuc}u%dPv z7Mf~;0}u4Wt8*;S@xevN)QN`!cEzJ_$B*)VZED2XrQ-bLWvc`(?Ui76dzipyn*n;d z_BE~I{ zN6R;ucTIrD!qadEcenQK1IIxUS4Qhy=JI=E%5izZd3>vX6{O`w;q@!WP)T|wp3+@~ zgZni2@fW28Cjt)$=IGfo?TLmc{%$j{<(OG^_lMAqu?t8_SBhP|@)S(nQ7iD(Yhbn* z=)>uptJrrc$0>eU+c>jVh&WHt#97KLx&B0yeIMq_svX-%p5?JHW3&T}{dq**95p22 zq0bntmUv+Ds~c@OS%yB`tA%DoameHQZ?eI<9@Z;{!fVT&2)}p`HWfG%9d-%m!ct)EhtC-2m=&Zd`e*iEi>JR zCIa<|`>`12rqdB(8c&Q~0r=!2Q>;b zXSV3akRKn?@xqQ~^vNp?U!2^{E>}IzerXwIpU`6PkG2i`S^Wxl79J*9c5 zj|)*>vYAknD)Cb1d{u`B;C{91L{)kN^ZnE~=(Y(EsJ#q?e2;%%{n#17=GZ8~S_21k z)M-2MHT5ErtA966PqD$JDr#h=bP|cM(q+E&rQ>}zEm%(Z0lJ`VK&}YqAdjPqasJl- z7=z+G;Az11{RdAl{oBtq$_XE7TvM-&OOmGW4{;p4wD_(>O?as z7Vf&*)VO+U0$dYPNGES;XY%iDVZ;yEK#O<@r0wkwcl$TMeS%jo(XLS-a%dJinZXdV z-78tOk0%5NzHAfdE6N54 zVu;2ZY8S1B)B6~Ly+$3~6A(@lu4?gwzW!YOW*9TSyk2TDE0Dy;-z~HKO+_(buY*;H zSy^sHko3^)BU1A%zv&Z`VZ!k<9`r|YA8y~J$o(&b@)K^0NS?=9j4D!QzdeY!d}F4V z|7;(*)Kvt+u!C@CZY3UTwiH6PC<+UEY0wg>TnutH2kZ1NVqR@A?fe@?wZsGTYRv~I zSDHzN_AR0x-VOuJm3^3wy)#R%iQ?~nUgFBz*6}{&_aQE0DM&1hAmi0{`MlvD92`*s z<7-D?a&{9Oo^S}Zewv64Dbr!gXD_P!^jf;avLq!%lv$rVsx$vIlv7 z<~UluG{vy`p2XtpY-adb6VtDaq_4Ve$laEfWK_{%@K~J(c^{)8ujL^%7X8r2(LicA zHHfKA31b`7-RR+uHPnAk2$!B2$yI$e;plJ|eBEa#EBw@Mox zV{CrfKgh>N@Kslr+R4w2*e5DD$!jnzw;DpN`V3;)uKmE*c7U|>@NBC6ES6dY`LMy5 z$tCxT`SqyRJX@<7Bjo<2!2Ae!f6|&Cac;p@&prIaYJxRtzwl>fI5q8Rz#E17xO=4m zL>rtW@f#k4OF(~8k$eHTtdN8l>;%m(%Sp#^3BBQ40MQQW{P4-4aI0EA|J9A-uYI?X zksA%94KDW~_va_kDp??3zXWpondkARp&<@)9t(l%B`~L78|f%-W_9EAsHylL`6#pJ zCk?t#`-=uO3y6TLN*{?F)TKk(`^qv8bd!C~)Rm1N_YwnoMZm-F!>G0bgY%Oz$kdUQ zY^2ju^26*Tyt_P^_B)q~pYJI`576Y+vD5g0>(Si9Qh|GaNr5>&(rsq-GkK~>BC;t$HRvu9bl5a zi`rOBrW^DQ+W!SRQF^1XWeD` zw+FDukOGqIISy8=cf#*+lzdszSDvwAk0B!jYGHeVek^mtkmYG)K!+(wtZxUemtUEF zgEf7=yN)!c4u_{tl3|$kR5l_nkC?{j!@5@#Zdgg!(c-nR^vyT2EX^06DNLo*I0cU% zOd^s`Tgrrm+oZ=oX2|P}VA@${gDZlSNVi{$sClNyI&wcUn=ORKCL7Vq507E;9Ti^R z@(;U9#=z*~r>TDWIgx9>AeC-ga7=p`^j0i@wr5#vtKMYzHm@FRcjVKOTnBEcnS%?| zeeux#NZ$WKB|Q8yknZ-#M$YwEz=Bw~d7=j94wtdA1xPlT+=5BnY~kzGAj%I9mhFzw zlW}?bMz^X0^XWpewE1UQo;A(L%`n5%m7I)pSVTW=N|v#ORSqR4-#2Z5|&*XC4|& z7o8XA=UInnYx8s zW-_s#aCPTY;cQ!fVY1g<+D);C5al^ncsZ;PoT{eLrmQHs*~Ay>Xf+M{X(zNd%*C18 zX285xlW6baNd9xwdAM_80+cS9gh}%Dclf?y*!^^2S{u!^uK@h!o5>*XMVHJ~KlY9O1rK5)F(i`{ALx9B>Nvg|gK17+& zrw*0WqwWV?(`~-cpg2i5y3T}t@pPd!>T<5_`F`4OuoqqR;5^RQ)1SKjI6-5&d}wCx zp>*?%Q2I<+Lch)4MMtgB7IsG}3X+T*7#XD|%>MhChIA#;0rw4P$>w3yFfNSBf~rZ2 zoJ}aPP^aHkKZT6-5p;EOHd;BIhxA8-Xu-uw`e)`b@QohMs!F9a!QOzzx>!KOjAn9V z@MW?7M1-g~Oh=e`cbaf0#!+aK_rZ6CyW@dZ8kj2Y0X{b}#Ya8AhzGY(erLJ@q#F`i z($`b&0Zo+tHBcwRx7biIyau$#?x34lI^@3S&L4E!#TUMhqN+j1X+iBM+GBej-QTZ| zpuX5iNImbu2Rg^YF`E-GD6f?p^~i$Eztyy-_Ex&nVw%uxwX^Ui(N1^?^Mv+%;n8qkR9fsockGX#GM!({ zVsRfR|G1EjU1&?s#yQdDW~R`YpMg2gw$NB7IhXQbKRs|Ifg0tlhHo{<2mieS%Jn)z zke>n3SUpBstJmm|Q51yhwF_|1pfp_gZiaBDpBc@cENAJjU7%NB4RwDjgK>*B=-z7& z$uPBeGI#KOI>)HD5Ix+OzfX6Oj(3&&Hj`UnVEiZg_Fn^yo?S~5DY%Wbq}CzQ)*`m`xOCUqfPWsNght#KC9bOHz!YVg!i0N6)^1Aip{*|YA*lR6X8*unJu804+7!L&Qs~X@7g2obOk_bnp(XGh?V0$TT1I|^!h9wAXF~;> zFl<2CweETF?nN6_T5T@;AXRkjdwWsAb_AWBWGY9W2H*-ppD zr2ayuB(GYPPJ3cSTh8iFFT}@hp}eMsM$9^fCYk2^T1W$*`l^m^R4ii-wtv`XJ97;EIhB5&XpeE# z3Q)IWKUZnojBabIsL`LEQ2D_NM=yDWr7@g7dRzqYHWlLfp(`+av7dN*^@p<3BXzK= z-yjS*znHp@OQ+Um+o`yBstWqUla!ZCt$DnhiHK72=+iQPYjXXk)qD`u@rgghBdz;kxeFZowt%lPUm>`)eKVK>}-ow>jVXWrI1YF%`0@2pD zWAcK4)*3ay&0IMLG-5UWg`K3|78NJ)`oGLE6LvzL^(XW3M>4o$@G||f%+((p%Cr_7CP2;n$8}E>n zk3vCl`$Zg{Gk`klHR71CQLJ*>Q@YGfmEIp)O7|Fjr6<477SufKh2csI>4@F7^xU_L zM0)aSsZ8M-yGSpW%{4zH+3lJ~avoHPcjYYL&kks6$vCzTDJG#@J`j z+D;9A^W9SZPq~iwsXGsAij)N_>l4IK=?h!3EKtr_7DY@#xKr9|q%C;u&S(F+qQT9}h>dbT26mRuS#7YT*mKEm@zTH}VoJ+$ zoRee9depxos_ff0_6#Mwiaqb)P#XC-L-)sgSyPF3taa zhTgw*5~a5y(Es={`giFZnDPiwPIdb^}w9cQsq%T|gLm6F9f11C_!Z9lOIaRxGKicFcitTo(qz|#dt#jM z;!i7Siatymw;=t_U7-8vP&R1f5!~G3!jyKH$(ol2#S{HEm7S5l-9A7EA{W z6jb}w#^jff6Y9sfbxL-L(o$UQ|Z~K z3+c|cugIonJ@K0C4Jzmlq;;##!q#eMs&)1N9yvUSFWA3>XYX<--TPpd*lb!S{dD4t zL_P7lm>S-PyimABKl#McMPNiPK5CKbv~L7QwOG_QTu%e^UlZl~E;z(&HJxc;Ovepa z0HLRbu>H!hrLk5s(4+JrRdh?k8Se`a#{^N29Z~q7$3=1RkO5*xfShj(a+aD-34u!2 zaE!RpgP%XxEczE6690LAV{^X8kzQsS#B0NQi{C8ePT_?qWL9DU4t}+VwK<35?ZZ!T z&x8Fm-|qyysX$>0)r|d1$1jSfacdqjlV0sOs$qtZ%~&c7q3E;f-DV zZSq0xJ#83Q@d)82k&zsqDT3*i0oXdk4R0t;;;TmGk2P@d(PHw; ztOIhVrINV1p(ximv*60AG`w1Rl59{M;R*}@)>)c z9L??K%pxsU!uZ64HRSn8dpKntjBTd*c*w8^_!VwK1K0WZYEnJ;Xr9K)-lO^A7tz?} zmB#P3pXMn+n{eK>8T{Bc2Yw+cnGN+HNY~pu2Xpx|y)7(5F?$wn^d1P8ciqO$Rg2+8 z?N$tbIEuf>TfwK9C!m1u@rB$csc35ne?#WW|3`jIc14%Fxn(nxO~*tFozeV((__3{ z9RoM}P2h(%Uc>3X1F<0TJ+ZseLOhPk+2wQ%ZhUSE@hkeotOLHFQt=q}UJk??4{^d* z{YT(Rvqjk3U!P*gCYI!z3#H5Q0@uo=vWL=)d z(-z#tj+{gjGggc)9L8-HD9AkZK5%=jr>t_)9~{=b7aCdQK%SZ&zxpH{ zXO7RJt2&0#idV1TPt|#TDFe7kKrAiRx`!9?82O^Yv39)_oc1E*4{CRKdR3FJF#5nz z{TH7q&mvx2I0Gkb_7XeQ4?^CJ(_&}v2~_b`;ayXDLcrJn=23H<{ApVv^;~g8I{4QB z-s844ZZtg%&D(#mH@pQMchbxDaoEU}NR6orhG&1kj=M{^RJ)j; z9-WEfeqLnhB|C6(!90F1xR}SK-l1O8C@JZbV9*(T_97fe2f+dpMcmH9bz zhTuuZ-`X$yuCx}$uKq{&=_S)SPXcMYu$zjSV}*a$rVFa$s^~rWy*jW?8*ThA;qz!; z&}lEGr&9X~vojRw^A`rN?d(hx=lasIbI#ER9lC-w-J&DdI$qXnZTF92jzKB=dfKd6`N-1(n0-;Y40nCx%%G6 zU|g_`_x-t(j&(dlOCp|AJ%uX4*6^S3z~i(q`o9#qwa}ZY&M#uM7Uya6&NnpMv6a&0 zZ#Z7f;L!mk@a1X*T{UqB4V-)#Y~M_zHuE)v%D8$O+BQoV>pw>rByYPPeX1!;SkprY zYkN%9Co9lhHeO^;<36@@8(>`TzECZzhFe!ZL)ha7I76w49Su7z^s7sepJ(oZb4n$B zp=~b&tvW17Y(ix;?gihzYzo(DvKq6z?uTI3RU^zO=}kw3Ytm%>_vkjuLpIZCvMk}; zUcMv#7A;=bpTF}FNy&t2=zD4d{8T$c|L#bF=Bh>(b=#1BTAmJ#A<6J``5GMO97I=} z|Dqd`T4?nic7YtBXMh*$tm9Biw!DG1W+!S89 zGm386lub2XSb3oK2XjIYr z_?<>BeM`%#D0?UWjrW-BNZq5u=yH==)c4~fDoyjH-}XxAj-KwUx?wr1&^gPVZvRy# z1pkGCVv#NN?a%){8;-kej}WENMcDW$n;P67Ph(P>!*p#m`a`h^yxwePmGcUy zN=XPi8#9I)eO^lq{DU}OdP{nvJwbf&_8$B1-e_TqqK%O7bA;e=U7b!^|C%m|7$wLO zr_rCa4`HgsaeCots4!qemM|`SgwSjl1J4%Q(W)5(Xh{7UVQ#RsP=t4A?<7^>>9Y>H zO*vS|Opl{IpXAd~zAYfs&!(@d73tE#LaD}mDPBACgzl?x zqHB(t!``36X^r;=SggNGxL=+vbjM4=t2MQJ@KksHLGc(=$k~s@K|zB56m4PG)Rz!E z>;?>7X@Z*b&rzHFA}Sb-qK&INL}{HWzt;W>v(CHn6Glh4OPn<~yQ#%{KJLxWk_x`f zVIFv#Oob2qm-Bb})A>QIidFk_9^%ue8rU^06Q4a9fx8##@(q^faQw^b^j!NfdP`PL zYx?el2Om#?^w?M^iQR?k*Y@UTHjKj9i+<9uj9aXK^9n4S9|#V?$FWTL5@`Co=f5}o z;QKb}$YvfN&!VoYvYSa)nTg$Rtaix9GZq8~uF1#E17$F46sPJ+M)Y*C0B%LqVp7jK zoCx8dljT9@_fus@yiV~F`&azm0}DRiyh5Bj(2Q?-U5=j5mFW26{*Zs}8)??+!9Klw z0vkH_(4kZ0dw!RPlY{TBfcg9Bl+Ei#f9!}MD+?5upYCz|xOFpM{lJPh**wB!8=CO! zyeWLO-7-EYONyVHH{dq%7pIw6^M~%`xK68(lk5)BB0>dxwHAkd=d>m=d)%< z1NuPY1v#&_2*3Thz*lP2;`ZzSp5AG~Z!~x3R+x=T*9$z)_9TwmV9V2Y%Ndumjd<7i zKCAzv%Krul{A)!w9u;3LwTT;rR}MeJ9WaAl@r@)Z{SS)UrtTqwOV;t(>b~5>M41ac zY*F(5w)ea<;zCnS?`~6;Wvh@SoWSPv)Moqxv!Y!q?UpMmDJ>xjG1@h!M3aEK80_96W{NQ&hu2SNF znzdv2g32j8VL$}&>63}eLI?8^b3S9vcoEUn1Vz^*c%|D5KSw&yqak5bXz--nr`T~8 z5yeNCoSLnkvd^(~=zN{$qQ^vD&vGa9<%5J~RWqv>RP^`wcd1tYkk`n(^4D zGhme5gUT*%rxRTEFvXkSrQfYi6V)42!Ca@69DQp?wm<$(dVkn}`?_X|(k*FHgKq|; zGpRd?fAOx&;q+lrbh7=G-P0C702@%#9T#bU<)H zciP}sOP_Q||re|8a7c&#kBKAcUPGkn>?wH-_`;g%R= zQc!lZ|5aqN2{iRiPx)6l2lt0OgWZp;=<4D&P`g?Kch1L3R1Ni{`!_9u;M!N zhiRbp?+@NiSkE+%T0r@0UBMy0K@!QQI^ycwMTIKtL-qRc;Xr%OpEX6(`76j89gD#bsdwS5+X_Fwuvl}&A z+gH$=X)hRNT$WB!Jt*yP)4|w(ai|k~0RA=wixGPh*sAk-+}uA3{cKvK?tVthGs2w= z52_`(`jMhfUMRa1+9p|jcPTpgY$5*r2a|l82vn83q?7Hfh?dh5T$4SH4|-_`E~Z6< zc>iLbn$Gb?$r|1|_7)1Q4{1(#4%Eh0(`~(N*|M}q@%ex=tU#KAM}yiVverP_z9`HTZ}sA3PMJ#(C>*Uy$(cYR6U4iB>jSnp*vCMz#DYU7*9;3vCJ@HGWg8v$&JelspnNSVY`+ETr)gY(Ic0#-?;>NnZ$Q)n@$nFL7W>P^nK(O$ga>hLwHR!T(D9Shk`TTt0h- z-CJb_+Ls(Ky!H_NGx)S5dxQx+=(e0TInJf-ZqsSgIa7Mw_^~_-H4wv}Unc!$G>GHx zS5QECrrs0z-N|nnZRv4?mak}}Z};jD^ZKda=lGX3y1y2myss6du1%~aRF`MeZQzwt zrt{`yGr&0XHX515bCtnsP!9TYg#$7288n5bcE_QgZZN4*@!@mNRNx$`{64I_0gjZU zVzI>zlCfQ&FWa2p?C9g-8nb`6pwDQO4W@kL9bedc&7HWo{S|fgT4C9}G)zdF#phoC z3ORe^d0|5zz_aDcu>tDp93Aa_R41FpgS9fee20x&P|}+ zt9sIbEQf7&(4h3%Sh>da72A~g1PVVD;EFqbyqDjN(b2EkR;rZXX5#9q&3wcklvXE4_zLDR;~%n3Gk$|ZuWvJ8eI%r`Y^+~DH!m0B-$*z zBqsfdrm*KiKxb;$hpu&&8~jkUdz-0D&1G7%j^g~)YCK0%D)R8dVzHUN#7(+Zy2L4(tbKZmoEvyt zY-||@XJehEm(%s|UQt(B$h-mcr15BSwQYjr&4r81+xsd0T@(cmhXzxd8WjlMZ6zK5 zw@CW3KAn8Ou^g(ZwCRfpZ7h0SJ}lTdk>(UlVs49=sAa7J-=mk{BgafKSC7Eer}1J@ zz-zK!&!3IVtSrm*7<&HYu%V-_TiZ)ExPK~3dX!t{ykigX8*-4m8D>JC4KBjxwQop{ zYdWMWrb`wrE(D8R=OAq`hfChw=*F-DSTtiQ;r|Xx12l4AZQgp^XY`ybHQi2^-qB#= zb{%JTYI?x5^@XCLYd>P(puwDf`;rRRNleKr3s=9rEVb-21N72kN{!ask=h-4DpgUw zF871(h1}3qancZFnC@#sD!$vYwtP=GnwE;uA8PU7gp9HV#d*M@a^a2T5pdQGllW_f zOMc!J*#+Zzvg(pQSgBgjto(`6CpP6}mcQ-AdrJ$LW`3i@qsWxJujo_yd+z}1HFhL@ za3PU&eY9c2mOqD8TQ_3$i9GhuPm$HdZzf08?n+P2`&}9+_hfbyE5qr_JIT!Hd&Dux zn(WTZV)nzl8(S2#2?pexAhwxFFyOvFga*p*O~X@g&y4|?E}qArU2b^cSQI`uRSzH6 zJqGuAlf>6E=aH>4J)#n~g>~)uS2kjcE-O|zDkgMW!0N)2P<5jsU-l*qbp`|z<*3PE zUAbI(>h?kDj1hgvTA z%XaM;C+WB8wDe;Sfs7lulGIh2NK`_uuybcWvLfF)@o8H+bFtYD4t@5J)sDN%mdV*f zr(ylk)l3<;e;Erg^BbjGlY=GNDFxz=yf3(=Fpl~C4wG^Ni9<@Pmpqq!lYK>PLs`b| z-|VI7t91OAJYx1-gM19PD@zG|C!OK2>%v<1SbQGiPi7{$(*niW z(mBmDMSbsFsb$1KRx&fDOv6Zn8l7;5nJ*8CBL}@Nt1vh&EtupaUhtnOxv}_jS$g07 zOnYNK+x^o@-Up0>jK^D9kH7sf6Fpg8-8}JFy}*j(%+Hc%DbUM0LG0=NicGQYASc>S zvsq<-SWvn)m?qbdgWr@`uizFqQy+u&axb`@Ssz-SUn0KyQZ3oPxkMD_bz|LI`bk#& zyHomZ_6ADJ z_l1#?7rh@yhQ0g(5m&9q(DGh%wrYj+UHbE~3(J;Z{;EC9!KQ}EJmSUM<0>Tgg7To{ zb|(A~o5_Y-D)__CT{=AR0H$f&VWs<9#bM0>vcx!Lne@vDnL?kHY|7m;{E@RV=2ZY zUT1M-CrSVM2E2V@7WT0^haS7%VYfT3{Bp(^o;^MnJL3QGNw0g!MjP157UpY!(_LVl z1?KEJu7}#?fZx13oEpYo$6>p!kUuG#xV#?a zZ~ZC%8e}MIWip;zv>8vN+4Hze0k%%Pk5|W@L@3xyye@RZ4+YWuvHn-wv!@wRZ7P3h z;lk;&P;R*B2Jl!7sZK>b36f_y)M;q&#kM*)AuTx>ekm@G)6uKNt>=J!^-*BD7eX z$1^dDX-Hd-pAx6Vq@Y)8BG{7t6j<=^Dt@ zIS8)unQg<>3uw`ikCNF3Na3bLw#_O7ANCxCJwG^ON8Lr{6?qN)ZqMNpXaaXzr7E-B zu!jeA6vGp5Q)&7!Wei*TA6v6=A0Om1od=Y}@XP)8!^biEK~;Gg-=SnAOMZNWH&0WR z>6LutwG$T-$2YyuBQp(+TQzys^uFBe!8S3Ya3dVPenFaI@=1Em@mrb8t1K3KdJ*ig zj}ogZzpzckAza03I}ZG$1dS%q%(ZacQ*{O~trly|KT(71v7$APN3+nUX+uz4NA9FT~So*Bi9qNjgOPoq?@4M$*jr1KF%AH}QpvsjLN75b5FG z@JLVw<=@?z#tL(ZQBDK=^Xm)cIs>GcUdzO@Hbrc}*PhgWk~v*Hr-f`#c!+wv@A2({ z-Ed_^JM$@FByX!7Xq;j=GJO^8-CGfhA|l23KNqB~&wt>t-&3(i;%r)77Y*cM2mBtV z%;yyu&|h8U5LK#4FWw$PVk?L6TCYdYt-}(>*zQLA*f6MQJENGDL{uMejP{*4(t!k|Y|Nk_T~_m&wNW$>8)@ znN$tzMh7L7fXC9lBz;p{sYz`y8__bI6r8?6<_)kE57a1d-G$r0;B9yMxnc-xS&%6m ze>^~P!ssuH8u|^V6pw_NQ?$T&>JlJXF|<0!UikiQ3;>-5DZZs7r+x`c+2DwQ$Ipv1 zpXuV*8*8Q0eN!=cOblC=e_Zb531A=GMnT`tS`cBrmUaIV4i|l2kn3uhI2l5u7g8Ho zjh8uh+hGfm`L$%n2W?s4_-n9xTbZ=1%a=rl=86MW4Z`i-2gy?H%jAlcludA#@n36R zNM~O^!gn4l=Z9TlNygCCaDTG8*uq{j-8+BD(dqg0@vZ^9@oXr*b+{qzw|I-hdqNa0 zo%MsANHyhCe1`C~8~ahCtQqV@qbBAR$Y*Hh%O%gc`$>ZD<&&@rDKH&=~j}D|c!)xdUpBLiQ z8Pjm;u>QQ{TPB8kKSV$I9@yT@Qle3rCcgJ~V)c0^AgflFmZ|x%bnhk{bD$Oo4>>`+cI1LWH)j_Mf^7rDMZcx=d26N>-gfG)i z!;xFZpl*sKJ%$sh))q#LBS!I=F1aL6={m8~pG==M<-&rS6=iofA7O@3LCjX61;+0H zZmOz-f`1R3H{A~UPTB>BR8)BK>s4qj&V<*$7Vr+8X>`9+6x~=Cg*|mcaqJ!!GX3m8 z{vi4)5l1UX8b^iE^u|_DPVGm<^qJJDvKy@uK7s!FB4AIh!>hPPsK40^1C)Z{#SI0T zVY7+09&zWLgL=@t_CH`j-vP9%<{&e>8BgmatuSA)6Xxt!tuvXbDz=6e`9HVATmMM86P0ywN+ zKpo#+#gYa;x>IQ++_910eeF1;ZT}2UK!X;~=mB;=F0xM8MZ+_?p!i4$eLGZJXzu<5 z-e(8X6aQt?(nTslmg^QlsYd|c-5v{3N~tjL#|4`GPfOTpR7?$I%P2Lw0{gx<(@!4l zq~p|ErZ(sRYJCEFF<~q{%F7@v?hyW)p+_R_jstg}T=+fD6waC$%XbB|`Kzt9;_bRO zkY+cO58SAZ5zZO7b>#{?s51-0r^Z3otaG?X*BqVw9B`3mU|GW=5R2sAgZAPka922k zrselw_=P?EaC{7IFhW>5;}~0bxC+)CT@OOyEIcrEAAHJ+!t5yusJC?iy34M?qOG>_ zyeCuEOREB6<-4mBql(xJH7Qj2$I^hR~$;a*~^)1 z>eEX0Iq31VV|&p-!2pX?1L*qjZnR|1DLB1J2^<6xW?ew-G!9};Kx|#cpP(mHzAo8!teaD0k`jRZ-@!-l?mseX?7U;Mqh%Bhf1l%J0Gbbq)1j6 zZUr4#G*L4;!LAIw#X>)tOZOGs#o_IvfgeLUeS0lkn7Nx)_wRse<^*O!9{p855x?25 zA}Zzt=bX@l;GWI!zG@Hj`l-&Z{Cy2$JnF$_;VN`cIu6}au0YiB7|?xF497>$qX%mz z(C1?ksQ-@{)c%hQwqJUVJ=DigyW3x|?~`62pQFIpnFr}e^HBQY@*XN%;)P#bl30R; zIUDlgvvk0u1L9=&8B)hOMO?LK08S}bBt5!Oa^`up ztuTlmHZtb(BZjlK2U_e+-@a_?g!N+o^|vH!Lo&N%uyM4_*RA~N$$ywT8Zo@djK5I~ z#$AuSNx_sQV3xj*>ZEDY=_#&gnODW0M8C!Up3fkyN}z{g`_Q&;UL@EmlPpjU#vz^& zOk8ULD=RH9Gs6|0Jy(T)Pq^IWV=Ps_vzhg5E)}ERsdF%%M23A!k~<>}*y+{^$#=bW z`E275TrsX>>+Q>>SAMI}*Ds1eFW?dAChD>);fy_-xB%5*I;O)W|?o1Vj=E$=NcH6LsUx#r=)@Hma zjs@9ZfB5_AI`N*)S?Ay?3>NM zru#De&cEW^Fne;vW(#0e9ynj_^@vT z?vrWqo04=`df5`rezb+&*zJcNaqWFO^~VHRu`ymLx*`4|)FZ=Aoc9 zFN`#gx0_7(IKHkowz+*wJN7AAS#kW9KxH7`Y;)^bOUoaW=sx-slk8tTj1iUIb@(x5t-I=k@Q#VG3IuYVZeL`yxyRT z;Yt5+CJmiW({qPZl`J6=XS2*}?Y;ps+C&4AshD-;KGvu2h-kH}@h$WePgZ zj>j=-Hd4p6Ps9h3X3}^nic~(fXI_uiupVkT!2HrHH3r zeLyAo&Jda%W@rBtqvu9dKCahY(46N8sUHKO`_x9F@wW|)yPfBmiQ~zv8*#9#M1kJ2 z+a;GGDq&2sI&@uK58bbLLGLM5FwVUml-vs-?oTk-Z3|_xJ=aNTln&Wx*U1759HIa8 zi{#Neb<%Qb8gATh9p&jt^kw!X2#at3pP}>ctN9D#xRg?857AacNJ-suPN`I+vZX|b zkcg~bk|=GGR2mwZG|8xY&-2^}sgM$771=9W$o}1b;rqJZbIKIXTxvzlf{~GurBWqni<#evaT-Cw%5L7<&f<(xo`n}|E&g|R_ODDnTo>V1FN9) z$wBh-Y8a`He<<+3V*-nfykK)lzVKd7cCFh!M_|gmAf-8p?AqD}mfOamOTZfF5Ag%@ zwU>y6a11G!zl2<~z0aq;?j{ZYW!nVLh$Y0YMX-EBG=JLGO-$Wux&Fz9n-mWAKDb8$)Tne4~b_BiO97&B-p9w~s$)*mv2f(lXtH7+lj2=*~ zqRZkM`KhJ5K_)~Fnk3v#?@!@iYUfK2mdioF0bMvXcs+O++JZ}UJgEHXAZthk<--+N ze26?#S+ktp`KE!)Z3hn@kp&AoKGUzB1}sijkxuhzGVB9e(y?GOi+)VJ4Ock9x?HY|Nn~jkYhD?0hB-&h21fBEt66ulE)YrO= z6f{kxs}_nuKW7|Hocn{8>Je6-;sGn-=ktYYb3p6a82WPgP=WKvsqoa@6@Q-1r{kP` zS>d8cR-UELGvqm#t-RcT@wz25|kl0ej+ z$-`1h==ZNdL_IxJ(i5Ks=7-N>&eD8w&)sh*U8^mwx^o8wF;=Le5Cua=YGLO~b(q_( z%!kc2A{R4rSi6$sc|85dCttB)=S!{FdK)Pwbv}qKOddeB>U-$lw@bu66N<$F-!#Sd zX6(a&(NWxPtr+@MJ7_|5D^7@92%95JaZX?wZkSY-jAc9@`^Oy zTNOVT`N77cLY}s?ia)#(39v*?oD`ajqa<_oWxsrRW~w*Q$d#}Wq~yg)^12x4G?UFZ zVa5zs9|r^1jTrc9H53hfNffulh`;F^6GwV9iT@jxOqIzXIB-x={QToc95-<`C|!z$ z#NLtOi$~^giISZgv^ufz8&vrh^RqT}&V^*5@l9@j{~lL5k%|Lr?!*2@btVuF1jWy$ z%*$a2Iz3axycc_Tf3zkkeMm@L>RwT3K?+H5Yv)U!P9kyt9fd&&^TauMH}Kn4MR9=F zFPL<57_&-m;4lB)#*}nh@mlG<@TgF^ZrX}=@tgdU;tr2dEVpV2O7GMX&$qgaL^A8P zTzOm^QMnS{Eu?z33+YO0V|M%J zcUlsW1$uJ<20b>X_e~d*U4087EapFY=r zRJatH2o<;j>UwOUTIU|sx_%5pA9#t}e`k_s{Zr8>!a;m5$N)C@ctdHvHpm~m0ukes zQK}(;R*JK+PyP_@-xVWPU#n+OMnweo-CUXsbr^|(TMGn~9M zo@Rx_!ODH{g73>4$eqLuv?p_aZBx7^Od7CT!p{6jpZr=v*X))O%$>U(I&{q8HH?*f zOI1K`jz22(4XRykB!#oz&nE>znqVi{Nqk=s#Q$9p!TaUaG*`=kEe^=aMJt;l>yVWUIz-^m?Z+?*6j{<2LExF{LBiLiaHLH`p2HS)5*hCMrFd1d|mRSJ_h^R+*$nhEqLLz5*&+Zq8gqlOnPT4GcAaOHE-X6 zMPLk8b$f#9^cZ^W>kE2z%~mEIW6mVxIWqdiTPjI!@L_O_T}zYXINKTo3Rz}UmG7n< zl{*DeGPz_*KpH=LD3Pzv&|;t4_p*%i0ju63=GHokpO`Fz3mOyf zsDmtIIL<_aoNsXJq!9{DwK4gOGJf=N#Hso^xIG~f9!~oN_5&Po-qVSgW*&ql*1EW- z`Yb=WSqbYF`t#tdA|R4JtKW7JXzi&buM&R2)JQ_tS~pS)^(QPg{{oYm?npB}#z-7C zx$ycs;NQ#Vu=@HzlwVf@ua=*pp_{XqV(NbOx@;cxdn^k#^xW8&*rDvrtuXp;%^XZu z(WPfKhEup-ZYgJlv5JKYZkMQjXC#=8 zc|+%<8A3#|1tSV_cr3E1_S=MM=s$ll3m>m+TlaH48$7xb1{DlP8@pWG^klO@H~l$? zw@rsMOGWzo=y!62__Dbt>anFd30Iulh35BGQ1gXxv|2LvxT;+MGp*H81LcTp{#dp{ zp_A`k;ms@lsj^XjT=?O?y)fNjhrq(`4!pH;!tNiZp-MFz^c#OOhm1t>HTw-r-k(BZ z%cL+-ZY^p0;7Z4ym2|tSM-x9+H@54!F}>iMNDgY0!^@6qSlg(F6Aw8+YyDt8v$_Bp zwnehLvp>_FvNr7BgRAVRWDa;x`Zf9Sbr3u9EQFM5C6F7<3z%m9D%w4}lUA?xW)p+| z2p7DWMoVJXL0#D;`0K5VWm0iyzf_rpx!#1yvj*UU6&Y}Pk{&(x)dg&_|1iaGQntSy zjbr&IoFpuj%aqL3XP?xi;D+HxVNm06rrNuQMl~zi6fcdYcSQvf7n~F=-_-%h87XW` zqS@fRyV!@Mf*Xh(lkA?Rh+(kRDX6kuyKh9n@9+h1Q2EVw- zMsy2A8+wgIk0#8f1__p=uyYVJe5^!#slpUm5mrB6&XfXGQDB<`OV+5cV}2J|%9jpi z(k*LSxj~*Ts2c#E&ON8|-)0J`X7A^l7M}wDP&uZnqDsw!qEYGHP`Y_ei{SEa1se_3 zofz9Q8#jj>1^?EWR5LK1l@_*}?F( ze+TC68o)O9w&J43-}tn+8SIVah}NPPU>aJ>m!BR$61=-=CrS+w~0q*~Cm4g!m z11H%*;l>(j?USMxuVAMSlrVbg|bTfp>f?p9FRLylIaJu-6tgun2?P(vWx_IzTdc>%Ttl*;yD;K zUI;fIiLs(L7I&y!#lP26@KjVbzjLcp@*OP6(3w&+p12=oO57xOW3<3}-WmR==@ogq zB7-ykpFFoq2S#S5;^;%?@z};+_}VrRYX`Jp)q!%XY1HB44Vomx?JcP6w_I%GwM1+v zHCOzh$_hqbdw@pbNc?lC0Hvf9ams5qd~&;w4>gT~+F5$&y+{a+-KtQa53L79q4x+i+IwA&f;W@7C}dg_NuRfeKtwdv8TpW0+|tF7Nfv-vj>+NG!S0cSA(vU6+C_Nh%e zlo5R98IgG+Z@2k^b-HU=@&!joG4rRRUr6pF5^k=;^p`xGyd_p|ySVq+55(!(0-STr z4;$-E@Q3CZbSX{7KUZgBh5bWG$EO|EoG5}kxvP*>UWamT{^9a?eR^y~8eOws5iaVU zggfk4>68@b>ISST<+@ta|3bT+?;XM6C$rF598@EKjTrw8J}(K0tkcD6TrQM{?#7 z?zATtbBdo~6h96yem8DQmf;?@x5Rhdj*5c|(#4~^_t0N^E^#TxRm9_D4bkuS=4rG2 z!EdWNjdDw+YP$RAZ0R~`S@MD2o4pmXgX($ta6Qo%3l+iN)jtI}*GKr2*mvcN4@)pQUhw;w!So!5Fe_ zqf^=pt!fnb&C2Ue}t2E}6+ z;l-jEI3}-*X8rp{J2$*W6PI_mrC=Ib4wj?MSp%6zmIa+>(n4BK%m)218}7d3E(V@H zjWc&A;nXQL^iGN@NS#{(&p#Z;dBy@5@wNd5Wu7L9$CpC0xP^NuONo!pPDNYqANcj% zBz#vj6PAb2!!Au`U%DtN=0*__$F`zgA4WI0QEuFK{Qh`}*7*Lcpl2>AZ-C(GMa$y`4bu)<63WLU~fvaO>P zJo+uz{(s&qW|fwNAroh#QGFF}B{*Hsx~!E0tS>Gn_rZ=c7F0*r8%gmpQmdDi5WmKBNJG!H)w)tMKD$-0_n^I|tRXbPux)jdZt6|G0Lsm6q zBR#dt3l_hcf;WS5@j%ZDoc&TrMs4UY zW&_2uYvzI5Z+mtQ%l^-Xz|Ne=oyKCB>So4OpzaD|sWt;H4o;|*OT#s4j^7-%A3-Qv7`E2qn zFOg-!A3AcnzIa9PC(PP0UhJ4Oq$YN03*X&xf%j&(2+SHU(WtEpiHm_TWL2JqVG)7k zzt3*m6eLrWgt1_kZ9E$6o;35Jx(9Z3 zSAxlLM!ac;58Mi=q;pXVZ9n-i%h4k+YilFAe>ul{#zk-){eQ4S()Em$Fr*?byyP=l zW>RUlLlTakGM=^jEXrJQSajTViH*<4dNNJvHov8-O72Rhu_5<#F!*!@7!4~WeQS}& zRCyDbqRW`savcZm>%+R2)%^aSdgySAho2M9>uopGj!G1{5@j&Yf}o`^&HII z!k<{fuR!0}jyQ1I;L1d%jN^<{TC7KKj#HEd5JhA>W+79Z)ZevHW zw{$z1pevxi<(5KRQYbNWPo+AxrF4Lu9Zj2mjBD|5SZHQMue{z*1T`jP{Dr~daZ4wP zzs>K#=lAZ?tg}~%-U4fwec6F-x_Of17pr5{@OkvaTxl>$t>*8WoyqWNYsgmFWKo<| zt*CUg3-5k-hOTmo;T<>3VPo6_5h)x5#uJkGP~B8MreBV!YaAedJKm9TY3}$cMe^UT zy%&#cI*1uzn%Fp9R~+zl2+LM!hr05k;9i?9`rX`23h(U3gHcZ6;Ja@y=QtBj(UGC2 z44feBS^G34Hya2PoF_M*py%Brr-*?45TPUo3cm))s4cx4w`Y?EjDF zomCMe&W)j)|2C1Ervuqy$7FKA+e7g2i#9VKT0nNDSn)Mul<|v_Cc1Z-VSj5C&rI7% zP8}~H8%Eiqu0VqgIJzBcPggOOaus&AUU(wvI*CgY> zUvBhkk-$1no_y6i$Zs7chTO&wpm`gM__VXi3UA(4=3x;=A3bKv7d7#6!nm^~WY4lM} zYb~deS7CpsYL+s$?kFOA9Xt7tk81SE@NT|Ru^gs+*acJVcOk@XMU$RzJSSoDpFdw_ zGw6ew=xLMzbN7+tolAy7O8j)%wP%N*`^Gpj+RGC<8#E-GQ8VmZ{}X+pUC{5se|RF? z10B}bh&F}o79PrL;65coxOuNSZ+5$4Q~Q?@4-a>q`|J>XGEkdXp1mMg9rGGnvrnSe zmCM*a>k?ckIY6tgrII6$Z;28X9~QW8{3mk6Fh2XpEfOjz#!o$X1dA_DK=r___(RK3 zERY*0J{ORLIir+Wz@-)J$+1yRZOCr{w1ljY&$7(=)j#4t$5i*{Zx5}n<3l(;a7C-B^B9__RT!f(q`vpd_+ zwsAYmT;c}-LH=;%_CC7#KV7)dpMa~@G_gPn`qPg^(JlDR7Bq{jS*+3tP$U- z@)J|LVIs%>bV%YWbxb<32(;5VmVF#6esbU({?@(@VuQiB%waYzI4X_7PCMw_0UrEi z>j=2<(}UJOkucpaZl@*M#;`C`DA?{?#>J7Bgo#6@iX-JE9q>B~uus85oEm2!-uS^q zY!mSa2boV02gZ*T|M*-jzG~$y{v(-Xj+uLn?vYu;0)Hp*eV$r4wcQR}+iF2~*&xZR z<14?@cAcJ2Y=+@K51^{h0ZrCAh|gS471v~Vi|gj@pqI+u^I7TjB-Kj@o&Jw_{}z9G zKe-x4htQ+rVflseqE`zi^I6w* z!L$94sBgtym{?WJ^G8MCxIQx)>OGSsMt1W2@kR93rd*N#k~~q0rJlgma2VRPmBH@D z(c(8bLUerC0|!&b!Q`_~$-MCEAd~rq_r{nCQxYpgW@3Nwo{o)T_v_|jz00*U@5>PC z`(!mZ%{WglW+1ebP6V3=b6|bmKQz>zB7Ts#m%lU_56kZM3Vv7&WH;x&I*RJ?#^K7pP+>mRhr>*Y_)=72)q_e46?8Hev$k8x=VykeUji%N>2YHKJOlzqk(k%r>rH;UplC3i4Q(21W$ zI$_zLqwspBEpDo;hK7X%V$XF8#8rtg&^_q@}{fmerVw$(D)sBcc6P1uKzvp=Ew zylUKcc8TOH6yUo^TS;ztB|U6rLZAQZ7b)d`B7?S9lA7v!0{O>h~Xlgi)RJr6dmCFzYEYd zqeS%8GX#eRY2YP$Gk)N?KA!emh~H)$gL%}Eu8X&0=KH@=XZs@hOcY0zye3fTF9vLH z{7x#rK$kl8`QQ+rskqL-5Vyx36Gg8WL=VnNp}SVfVbr$*dQWnn_X_$+oj;FY?>+{= z$p&e$j3!|1s^^k#$ZDQB>Nw2Yya)WwCqivQ1YMq4O_O5=vZkQ@5cw^FAD>-K<5X|b zfm`I6#;yTk{V5Dy-doJfln09!G-RNlR0p3v{Xox5c}0&a+=JyxzU*ejFTuDbFDAD_ zz*|RFLRf6HsHNl#^!DxtyT_+Q2c{BsPyRZ+^HiPfd?m@i8wFwl|4c^wcuq84mh#Z| zLvUu#QV80xMcAsmQj(cyCo_ak1VUvmqUoK<+)bA<@BMmAKj}D4KH|wN{yMR*+q>at zfHDkN&_oA}uz=dr1uV^ZI1`2rq=OD!{5-<=w&ega;R#pyBS*aQubgXrH`} zZeYK_y`&Ok7jC8n%64ogH6hxsLcr_RL~18@mYz0rqx!o&P=5SAL9>c3amb5@kOT(9 z8a>Fa=&_iXJ%oLWEa4Mpy2Eq5D0rhe5O3e~6y;P0lA=#DNUAfTVbcV3Ws?yJ%sUKb zD^qdpTv?n*5xey^V1I5f4)8aDed&I@x^*#3&y>ZHmfG;Nqgg08p+{wB>?4Cc8hK5| zbyyIm2_$?fP2XmYFUP8&caQ@fudfENr#jBA{KW6p3?_EVW8u8)Cf;SefxfZbN6m7d z@O2k<;G`}$)RfAHjHwcK#bP~_{T>0@8Pzqx>z!$)QX7e#n@*)Bw2{`ye_-pBb@1H0 zy0-DYlfd}fGjQK50K1|f9u;^QI-dUl!>$m}8nBY=%zwfF5Nd2}jHn-j5h2vS?pPYVD-a3}qqwSSdV}*#(6o57It}A(Y?2(&5K2cG)d7 zex$}tZl&>uQWJ%;VQSD`Z-?GfmqEax7ntwx1qNSLz(L)L;A5e{KK-|w-jv@-RhH$` zoVRN9jZZ6o5K=1frk<`n@%srbiCB;N=Xc<7t*Lxn?0CAy;3%C?e2E{>a$;>7X6*9G zczFNho9LOm0j^WcC1(%U3MW>-pC6?=aub)?v0eb zyO&Czj`8HoJxWx&L7MG8oQ67q$N5XgqZCEv{!Re7DEb(RaZi zW+2+vMdL8{cxZi83AZ2VU}x|%*jYXb^8D1C(L8TJOz-gh2+_%bzUaP|;|+`7P5CA1LNJ$iU3 zF9sEJlWB0}F!9wD8sbM9@%ZXm2LC>9A3V2zK)t5Skj#qL@iD!6sP*9*>{-|Ybxs@L zh+Km}+9n=)CZ@1W^2=c2k3<+Dcb??Od;+J3y->PM7xKTUL-4aYGHGiM?7RI6P8|JA zN+b;FOT%wLjnyG=7&Hlee*Qp*q0-`Wh3z;&riI+z5y8#Pb7;kHEBtyV6TK@|pks3^ zvKJaiFDGEcyC}YJpd~e`Sj07MG(-5DZxH9wM&*x<;O7^==aVAjz)-S#ta>Jc8Gp0! zo78d<$gbcSDJ?XpehoKU5JD65cGIqT+NkE108-XQ%(yxg-u)}*FYD$(d1)Tc$du=K zDH(9&#{vj-)qsUES?u*dXYhJ`66$o-X^XxWEm_{dBPO(v(hrZYcttv{_*n>P+nsS= z%_3Y9e3=b-YKjK)LNIV?3@-CbguzBybgK42ob)_MuxtS*Q ze%(x0z4=HwoX(QXr51t?Nl&)xb~I!g|K*kg1!$zX2WF{1f{kk~KtrW6eXpVn?|+JU zpUN$cCT8q`!5kRVokuf-D9CIc`IP}E79<^F9c^~y5aE6A&82UO}~&Vtd%!|{^RL_M{4EZzE&L- z3lG&^+@%i=Q3&CZxlU}f4!U0*hg;68*`ClEM-Daji)4p-@Sd_vcs#-pEo8dsvXd2( zd$c*b^K%!;-2t#A+CcO?Z6p2s-)|gS@D!_tlwz~@S$OqHjuwe-(rqhTpr>~r)|`7l z3fC>gy-y3VUUF|awaJ=pSThhqNg!s$WkN#lQ+U#8L3(Nr&=jRC+?K$xyR-;wx&tvr z^)!rHdmGkXeodCAYB9CC$)t7TE>X4nExI;ZSu&@7D@rM>Bo8JBgVd%LK2n`?n!f_% zZ;ge&%FhHY59Db5SXJ(7>MA(5=b`ZPy-}js^RvkE2v5-ut?fL={4i=3M#DfOD{Kn! zMAc`-Hdkv_lBjR*NbYPxi~f{RcHg5ly$C=(;k@7&}imMaodZ1E9u)x&X0Il@ZtlZN%5r1 zN0rR8D&QyT9z)J9Rk(5F3|YKl3#*;-O0@i=Kc74B7_aeP!>>0e(vU?ewjsMRiEGw= zLb?-3jl~LT;O$H=ZY>n#ZD{1f-CQ8WZZNkfdkBLiS=&S2Tj7c21n~SY8pmH%q9+?J z5}izE7_YgBX^z$hlh6jzm#RS{?6X1h>Jjd=zgiUfd<|%~55dRp{=>}c9fFputK>nF z0%Z0`x}0Cisp;EqY)G*-K$joSS-wW_*{q&hy?6uXBs`^`?so+{`y7SY$E{&$*JOU= z(h+W8w4Qu+90t|K1$@(?c2cr*2eeF&BvuMVtZ2Yek?}LhcWX>EPoB|Dro43ox9L-H z+UKvt<;^yMR<;&fI`2BT<_rO~BZkz^Z6@q2j36r-fRB6|Mc$w35VW@&v-3j?=>33h z8gC}?CS3l`-{oBaos%kX>s1NQ{GrM37hmDp$x{Rs%X4^-)HWND^g9wz&;+>(W=!wy zM3SgcD}43OMYJ(~E&n~wh9C1$ryg;@&wXDh9QyR0wKdlVzCoS*32zZtU)v|RTjb0A zb<$zm);1D!@Tfrlin=JQ{}43g%MfZ)O-sw;_<;JE@U2)uT(s0#@FHA_OrE@h`#r1R zS{{Zxq#>NG965`Z2Ly`(a3uRFJA-$7PbCd&z6nk~zDdpuD;6D|wS@Fu5|O@4WfJP4 zD;gOZ3Ddke&plPh1MOajmL^Z&Lz9ZxqiGRjYN0ND|7R%G?0!v$UAhlX_YUDQ?E}aH z=?+2Bg2ODoG?y#(|iKBux;wS&HPaGkm&KCo~D zXDpC|J9 z9mDABekp2wT${agTPx^u`a$HS)0uu(0ypk77OmQ|QXu9qz zhpiy%`JG$+QleiVp7w@RalJ#=cv4O$6d!Ox`3ckMER!TYY1;}IvZn>+<~dQbDF&=$ zfep63QNr)}mx*ylx9FesEkU($DzDWYC3KV)0DDygVf(e%CW}C(?K+gL-V?%R=T*~5 zu$lG6?S+=^c-X7Z1ESnkp3rMZJtHI>(xapB zoS8g4duNF|mPpozvL$fv-$=YF69lsT4Rp!dG58|13l@2B(cJL}BM+s)iFkWFdDjfj zybOaK6Y{~nbtwJW9)?>_#got4$Dw7aJ8@OehaaC4X!N~SP&1z)nMf>yHX5 zWjC7pcIt@6-J6BR4@X0WPaf0;MTpvpBWrDfc5oGqVB%dSAWmVeJk36gXT-?itOLPl zE7+)G9zoHs`w)=bZ)e-E7`U)PVrfU!<= z#$+i_TO$h(q(?FjuP1!X! zaAKVTPAK1ivwOdb(j@}!$;0x6!?QHteV__45N_o)zEU)Bmkm{2G?1@c5J4J_o3Xix zt7-9GO+3@KfM-Nh^9`eu$k?MxMAvo{ljvd3sowllI^8mxZmY3}%*Ae8EAu`-B4N7@ z(mzCv?tFuTf+O&!K3}*Y@f#fO(P0XUzSFw*yV=i3#@b>OSugG6UcX;aJD>67*jtI; z?B_}-`O--D*`4aBQl17r79ZAF6Zu03}&p^FI;*Lu9r?C@F zY1p8n^l*`wtmyB7|F(~()Nd-i^{XF1doViB{%JMIqeGApdlv$`TF5zxBkXwI2uPV_ z0s&XfGNp+Yq#~^z?woA1@wAAt`F4Gqup*?4AFr+vHq7kd%UI_?VCsn&_uV->HohUT9STf7`1NF{Tz2 zM%9Z(3MI&fH4?Ab&BjkZGR)iM0H|9IGPwuVo44G_fKw1|&Fo-vv?JqQ&fu4)o`U+IHrl)>M}74UrUSE5yPlT7X%##E|$_-xZ! zo)TX2gNFDB$gXq1Fi|%?F>65`dka<*S3u{5NW2!lI|zxepTT^VI!3zjlo1^F|SRyAKeLomz}VoA!dsiBvkWdoN*7cqFQ6**b49QR&mgMt~;aKMd`^y8#nVf*Cm5bQghy2xLo zAKx0m=d()Kvb}`p=!}Jgl0Br)Et1}PJ6H5RVYcXrMJud2c0-sFa@EFnUWCXr=QJJu zsb1u`x0-BP7)uAl8-vqpUHWiuCEOaKiyORqV7|K%*=XNnQyG;;)ts_ymJh2Iz0Y1M z+AxUF<&Ty?S&tgJY(Gc({Z~@o>D9FUU@(p}nuMbUTEQyIt5_2++0$*g3scRF;m>^u z4|uyA9zL;#ylnpmg&!LrY)Nl^M;2#XqzfYSLO zoDELc@XrUX7_H`yHcrQ|8}iJ=X@(iGl{TN08m(3BGCvad?(2R`gWZ zys-RZUGw51*;chmRJ$!0t{(`5$oOgSJlRh0Uj8tD|6G=bf0yhy+te{;s|8NnhqOP} z2n2INcyaDgn=ksC$S0Wtf}}EYTsUn3o^)P{K6ba^+$lMXy=F^q&6v%HevK1-d1^`> z#O>r~5QYAsb&z6eh&#S+z=LoLHm6^rrumZjpVWPx@l=bNv!}$@QzY`2))$NXe&Lox zH{AVkx-e+nZ}Q2gKxnB$iA=we$lbJ2Ff7EFgdW?;S6^Q)+84GEkN&d2l@>s6m~4`q z-OqW`084f%dI7EMnJ3aMpG-zLoDgYS+~P%kk)o?f?QnT~n&8Q$fz&59g0X%@o2?5N zj;NS}8h#6K&(2awf9xSOIg2nkd#qr>?b|qQl9X6lJ``OaI&;e&dGH`6@K%y*+CHhA zot5|_lugEw=OZb~t-A&a`?To#hIM>sXaenh4^aEx7qmF6&D4wk(#?BTp`vdYRdu(e zN9J+((k=#{ea~5@S~KbBK8uQWrA+;&h|NveMtAgHrpFRO=@5e?7WzVly-SjuJGlIom>o_#mJIy6pw*mbNO3jJpMUn-akJv=z4Jx1#OO68!!? zpMSrbO9N6&X@HXryvq{Pkex|1%11<=KqF0CRL<;Q9A;zO_Or@o)7hC`O{Q_WAHymG zaE6i}zRNVh>0jQ{mv*;VL{=v(zw}7LgHU0k*KTC_)AzHgW+_|Z7EL$54~Iw(Z(de9 zoGEULW~;M9*@vJS7_BMlqb>>d*z^sM=7Bdom7ReNwmW^GU$QS4Di0PY|Rp z-HV2NC$=VMu)tOq+j$=b+nQ$$wO!WTPcv^ngG05WagnHvUGd9dtE0qBVaXD~S7})+ zS?P={ObzQUD>L2wK`hiq58fVkh2N%etl_jBzO>rOH_2#nhj+W-+>#*4IcSIBkpsk1 zsS#N8Oyb}u*dpP``QmI{gd3~pFn1L@T=Q@#T5Z;#m5VBIdyolM=9{AS=nv#w@PDv* zx;3~in@H+3_KH?5$ivoozv%nI7-o@_2dYhPAi-3Ysjn(yn>WAZeL=>k8B~wLic0iW zkP&Z8pTw@OU&8jsEP|`W(J=!~MwG#343Iak+{ zV-KxXvXA%kXugsuo$zT4lh!DNl&cfik@K;v=jSuF-rJUaIdX}v8CpypTZF*||I@tI zRDrA@E&QF$9D1y33x+%XfMNYh*_At8%u~Lcxh4%@4UXe&S6hyw zdYXg-Mz{!mRgK|ahsZ(kRcVn{;UOs1JVnbd3#r=_6*zn_9#n7U6C2u2Q>TBaf zLwxwVB{Xt=~TFnQF>Ge(Q(>9{7OjxL(Ik(WMJyfJ+mQ4Y$a_Q9ro3iwRV!#qlpD+v$`s zwlp$fB#H4h0p()@u+j%XZ(|(z>KS9_y?N+!{FTVeEF0QOqseO(vB)K#(?rRhdgIRv zw6xIy1_h+rxLU=)(%wFZE0mDDByobz-eu6I_JCYZJTCCfyHEFbtC6weEXas+gUOwK zY0+Z`Nd~s3mRe0$33D@(+F$n0~OX{CF@>JIha`ACz4GEmcw`yA8i#uDPqR0&=^{TS#N8|Y8 zF_X}t>=6WeEJSJHOst+;2Z`B*METxPfwY$%L~V8wG+y{%GpW%_*km!eW?bKI_@wg; z?hP0Vd7}#8NtHW_7jNXJE+>MP!E7>4#Q-!{hr<500ywBSN%EiKLyh*wP~(6`8q<&j zrjMi~EJJB>vo)V57;J&{K8ZZV#FpLeYK0+Rlx=Q*Go{)a1g!3-JDstmn+|t(q+8b; zu%qQ#qR)DLLM@vPoAe|FRQy^+^wXT_6y2NrRFNzB;`rER^5S7ME8;S}(H}vdZB$^l z>MH1_L{n;D-B-K8EUb3TRcig{@lW3HceF6OMv-Lye9FV82hdXyvAj-EnGLPT=0k=> zQg1^wyfLefcb3Ex+sXML*VD(BO4ih+t8FNFN71WkIlQ#Hir!u(*{^0r@zt}|QlE&y zEOAjdZQ1;kx_KR@a*rLU42)o#?HG|*UcwjM5=9e? za4^8q_W@WO9!1hqY)~TV!j41{%^!M@BrcJL1sMZb-qC8}e@h0&pIQ#mM4 z=5~(f?2iyy8U!#Ub_#L#w-JeT_JFYX2X|XGTC{zc6D!?RL*_|%bqjCVac%3PBFXE@ zH*U6K^S7wb!6T+X=Rxt|z9` zMchpGBzGLQU$EkxA=jBB;)Wr{=&!N~*Ee6}|8~pKmhvLLN-{f!kA`gM>C<%1X?Zp* z7pUjH)uQcY)({ceP5pKKiRrdF8un=sm=8Ds5AwUAYm2OyO58Y&6ML{$$%`I0I!j-C zSECz}lW3hsJkiWe5_(r?60_TnAl0s#Us`WND_;+xB7rlxeP#k&Zi#||cvHG2#0eX_ zSD>|JW&KD?L-lKK$)u-*%2Ap8F-g;Rp4gWEjs%4x}+BpG$P; z*9PdlttQFFhvL1w=UmpM9HN4B`1UC&sMB{6RCS$EB_m1X*by!8n5HY3`(hF9`KyP2 zq!q~F+DVmP!6kwibc#U&&R^(^A;D#^L+2n@OqHT! z)L80knZzpN=Q54|N}z7UNN#og20wXKjnX^0!kVslzVFQo{^4&hnC6>X+NU!Z5dpSIS~>Ei!w&wxnh*>q>xb$WaE9Qr*+lWtH^Ved@>XoGYk$&xrU zl~1l9>&JD8q=!Bi1h?oxkLGV~Z5uDRu6}`9rbhEQ;tK1DYZ`3^PdLO+4%|yE|HX67 z??WZ-lVE{+%YQ_11B#~hIl{Tu=8bI3Eklw%DZL{-gB&Z(L0&!0M?Ztfa< z-ja&G?P|D9?SG2S!x5|RkK;B8*_4nKqB7#S=bVQkNxPI%Xs4)DDuv3XVXx94BBLQ0 z&pn@`2#J>VR%xoF!MCk`_xBe(9?!k!-g7?h_v1H^sf9pC;?!)8Pr^*1E@7Ntg?pc`>N;pNVUZ8j(v)Y1A}x3k%a$ zV`F^3;ZHs@$)X!cq;YFHs|?IwV&h)enWIEj_Zy>){neatY7p|uFGs7_DN$4I1P*S~ zLkd2(k=K$^(X)Q(I-lt(;2=8#tA69?-11r)8$6V(ytNLEx5|XomYXqbJpQ|64=U0bPk$#3#zo(U!!8I%CTqq@sKWb@rU#u zg25R!Adx*pi@fIJ*P|N2aP@h3qL@h2e@L*xvT{5p`wF@}D*(zeljxg@6}-{9JdCBS z$;bOHm~CoBOTwb5t|EiJT~WyG+aq3O*aI$Kb0228CBix-#hwN2#;PMO;aAJt@XXHz z=*%{EWY(q1=54kSeSi6!^Aq+C*XN$6oJ%~tXYqyBZj*&m)j{0PA%SeAZx@V>>V-x5 zj{z;w1xdvr@HV@Qwq;nsw5csrptMjWyOA(F^9mgwmj=<;L6i~YigeeYk2Q`oWDr{Q{d7>qK_ z!O#DkMsvQW($9(#PWSF@Ze#Tjc;S+Y9?o&##~+?4`7HI4?pBNw=5K+LFU%Aj*XgAR z!Rc6Z%A0$9Z!h;UD8FvZ#dI`L=%#O9EkYN5OAF2d!tD!OO4VmO(c$VHXnM{;^m#-R zyw=$c@oJUSgQ-ipiKpb^+5}0V(-*Yv&wjk_;9oj>eJi}KZKT!KtKgh-A{g1l<8&%Z zmIYtKhn*)19YG~pWAPQM&U#OOS0~bhc~xli!==b@j2AxG?N80w4Ww$KN;lp$VKJL( z;d8nI?NJx9(sp;Ld!aH!H(Jol0cqUeu}vgyu*8e5=hDF++C;M7mr=*z?)1ER7_3d+ zDRKQ^L3N|mL9@vPKUQ?(OX{`xOimqVxUZ+SL2Hopx9L0?8OkkrD21=@Gsm*a+qiA+ zZO$uAtYD^FI~^OT3qd{EsB8Or?AIVLh)=x{u7@+U5v9S-413hq9Y6y=m-FL$uOr)< z4RGekJGwixjoyx6u{~oOsNsdZm~ZKzi=HInf8!^R+!h~V_fHd5lsdrIieYp{ z_h>Y1%ons)+5)?#bVw>KSI}XqC(te>9duywRJ{Az7Id6 z;Qa0ZnxK^^d|m#?9}J?U0;GU&y7Gwv~-y1xKvnR(*#xTz93_POD#7d z0PC%4f@Ai->MpGeKx)svpsqGoyiNZlwKOfk8|zcxd!j7q92|?Z>f)hA!wE93S3}6$ zbRb8*Ldu>hkbR*@CfoFo#HY)Uo_&(Qe{dnkwdXsv={(^NJ(R1HLrYT(WH!C6&G7;w(rf}jJ z-5_k~V3)ZRJ2F8T?wOXs2H#6q_KPa9?AgZ7uQ|*H3WJz}ofcWV_7^_c?}-&hox}6Z z1wMhYIW9Ppj`Pc|pp$9`QIy7LK4V%m?a$K~FbL!}0Jcp_rc9EOeA^ zgrI$6&)}!Bdx%%18Y!OM#Z_+^f|H+#SkLQWP;fw<<$l%zK^~6fOV?nHWi43wVilW~ zoWh@y*9G$Z}y0oSXPZ6!CHvemJ?7d*7IiE`8HwOH$NG8oG=n zMaS2_K#g|0@#5rBxaY%VPQCUjed*Fc z?PBcf<0EYAzg+BKUK4zoc-d1t{=Xn*5KzyH<{m))pXG=`Ni{!qz(^9)K?SzJQLg5y z0$ySD4aXiV#6N`luUPdcNeMG14;}J|;l#(}_@xmdg@HHJ*4Kc%a4N=Ve*&(k+01J_ zdxxc1J1p=~U^f3STezoH;Q3{+41JC@-(SF#?>NvcC+2`>cn_=#UJIMwN28N2^MNfB zc&ZQXaqr)m!+Fy}$mq}!FIZPZ&rXw_DDHmM+eB>Ok1&gw4&Jk+$%7u`9xFvqOhrx3ZLh9jK8@r zfDVpRL3t4#I971y&Hm&-G^9DA^Jgy}r!EaV;Sl;mM>O{JR1iNYv=UEpyhE$g+Hd=|9PURlY!dt%cL=gvJ-O%YwfI3K(3Nd6cy9+M*5BeNUL9>B&bvQEY`^d{ zdC=}pvXifaorX1gx@!#<>0}b`liEaMMFAe=VMgk#BDnCWhsfVuS77^;UC=Vego*V3 zNZg!HKp#nlH#n4xiC>AoFIkV96+GFbPYb}&{|*fP*pJEsXFz_t0+S!Ai?YPBKsuM< z?4D@uEa$%D?ZeLyXn<=9ah1ET*v6MAm-&>W|E zplPJYHolKxiTleyfbrA1z~}I${}2im_~!Oyr*MzdBf8pRB7Jo|nT=epgrDaK-hn;p zIQCy6obtR!iwe@iS-o`nCGMX*&9dRHjn zUS#oDXGk~C2VvkPbh)!30F zyEqW0=|@4!@cppN+6a{lUyEGVd`8O`ABVXflkiyu6`0_b4XgGP(RVgq>6EAxiDS)T ztbJaURjix=cQwA%`9IwV<85|e-=KHsq`Nm*IURtBa{Bn~qzQ~mDudIr-or6D8SxU$ zC#2!UWbx41TZ9DPJaYNT6ha)6n7h($mT^{@DZgqa-SZ`+LTWfs9Z+YTZA0LpM;>lF zB`=;V{to$TQaL>?1m=#*6*U{UNcQ;X!K@8o+>VdYJU{*d^(<6pYSUNJxU#u$ST&HZ z$WY>59R-o=5)0AED~9}_|GqB zTweZF8cMKQDKW_Y?WEk}3B7Z#@rBaK+;!^B_TP_#k$-&Q%EBFZp+DvIb*D?}PKSd1 z-dFs%hw-%Joh)1SObfSps!0YKZ_)#wgk7 zA`6kUIRB6i(;wT;w@f-i_VHPpHVmm@eqv z%WtaG<7S3uFol?McyxTIFr%12vz47$iSc$SHm{*SHl63sY83JdKAI6_)f$-l)r>bg z|AN@e@#KE~^OKxk+eRkl>|{AM36fnGbnwlkd8l@LNL|(Z8v0zfS+vA7M)dsgcc?t0 zN*^5CO}#g~alZTLIp3_5!fkc=NNdYsC2o(Vq8na@!hLmv8d^0XEj4|o zr7cF{7ncQho{cRZBKY${rE0KN?pm~dj*xNvY{!56(avxEp~oJ0mx68UPO4n+6?zw} zViS(KP?hrCuy%zjlGT*q4^31gVO0di*)5wLPt?+ok5a&o>Vj@|1|cmlVuN9?Ln(M_QRSXZ;))2 z$T?(LCjGV4fcBdC(QW=(5EGG3z2@j5J$?zxI$j_goG#{TxBufkjK=ZvuBGvnM_e&g zR$?7dF*Hp3179z>L)E6HaL7cM{aaPjO#@|I*OeGtE^SV$M3}xzvX)G2+D!AtP2@Fq zzXEmD7>H`D1s{hEAm{#xPjoKj!_J(c;;NgXk#c(D0wSdF;x=>cUAVodHKdAqn`{*o zS$&X%O1r?p#j{|Xv^^c0Cgie%4TO90gydqbjrj4}d^l|^N3ZQS7N77H*my4T+;mYh ztWtZz?^t($L%u`UGjBt135j(M+8rt~cW;!;?9Snow%^6x6&mnHyqLRs)DJ)M39dUb zIExQic?cc}|5vY*88l|@AKEnjndHV_W!m-Dg_@lybvkuHi+gPQ2IjTb(e{L0-0<`R zWVpu%QGKeRXmgM~`}_JZZ}4)SFOUm94S@mY3Z zvxGl+r$SVG!<8O3$fMHd?uqJ`AE47?pU|)i^7!3|RDSBOGPq--g|~-A;G_3z04=g( zQ7fC_$JjQk^!PQZpPR-=y(g4u2w6V=opAJU9=Pb2|r#sXCx`GLDS0mX|jzLJn zCidp`dS>}7oIUYU0?n%Fu>HI)v>A3VpSK%XbHsNTUXsc3oz6o^!UQIFz!3MGsR4T0 zlKvPwj?I=&WqKDh*vPLQ$h%_~G=7RiCZ7`Vluk`r?(&|Gwadj8b5-F&^jX+YV?rDI zYx%dA4v8k5xlAqc%XuBg(a^WjRp=`@I$OLir+MnxxNl4()7=yegSsox(qs#KXV`68 zeBuJNwj4>H|K|nG$=YniV9e~yig=p|3Gkrm02{1k$_Ap`@MMeOaI&@(Pk)=v?8GBk z<)ajM)S<~%d7R|J)K^2?sQK)xk}7jo8NvoP7xp&+U-myo*Qb7k1Jc78&aMGT+6pN7hP3jK5P)3A@uyBSyAUqwohj> zd*-7K%i?0W9WC9UB(CT351pky;{L+PTSH-JU=&P~*uylMMXndFA=%l6#Lgp=QMdPO z{_)jpL8uV2@!5@yEgb09feWn8yO?bq-vjjvAD|F_&P2^McMCOESV4G2!tSvrqt zRD~=oY+J$esnPtLt$(?`(Yk2Aj6K~oQ=OlhJceJIxtdP1lOh4i9R6Z+Q=}?WD2Zzu z%lF;XzynoEOZbs}e!n4Q5 zk-dwWu#)sw{=*s>VXYpFl^-vpdk$fGFv1?~7v}zfajkq8jHI6iiqXl^HGKZ&{m4Rn zG9L0(2~R%95wA2KQhqRu7$h9WzPH*D*Y3~v4JyRbO$%_fvjMu_xdW^DE}(zC!?9*W z9aSCs7*85<5>-BvMmNew@n0VO;f=mo(@doTS~l+spEWL76#V84+LA1fN_sYN>*u|v znTuz@sKP5?x5t30=kLa9TE#T$+e%{mCm+AcY$InY$I_{;X560*2ly}LedyTo25kOs z01r-=}}Wk}u!n45y_?j2sSfTL-)14~|~AX?_h#jXK5ow#?)j;}>CZ z&R9~EM8T*urFLEN2_^P5rc_2X#zYD7I^ zwMo43JUpf|9?kw8PiN2a!1eC-sGR(I=gGfZ zQRHn|?4ad^+=`CVo0m%H$bUmGIJk7lgsiv7qc%1Q-2V0IwDJ8??wh$ke{W6(BwY^;2ib z6VEY3R?37k{k|8ee^`Wn_#dQGLI*ir?AcE5cx8+7D>w0yNmpxJL8 z^FKmH(Kw${n*K$a$OkRQ_cZt7k9!-C-X{edCmoE9lccF~oP#I3v&k>&RSIHp7zjPtHct zu@T5qii!p_3x!#m5qn~PlC3;B6?&udL~a{Ok-pq&I(^|i8@@okT zOvEg@BNKF{>Irtio3JXu8^q1a!CvUSy;xg|M_&9)4u&H#<+%|a+X52xEj@^SO~!}* z&4Mqk^|bU(HPYeMvZobt5o?kFx^=|5JvK9Pt_S4z7 ztkLQFp73o)8ryDokw)m8ha`VM^X#(V^7<;+|706rI~RPWeHHGrO@~Li3qgPWZYZog z1=${Pl6x~d(YUugBGn_wqEF*hkcR$v=MCa)f1~W zjqYNt*Ir<@Zaz1oqJ$P%?13LTTDW!jE_6%!1lU-cF|&T*yO(sM2H{@5GKZq!y8=P= z-#%mNWM;tKi;JW8fMGk*rUHkf=igv^5f00IkFpv4^R;PY;)-Qp3GJmCu<5 z^`b2i9w4b(i7nHt!FXa4@(Yl|xlNULYuW|;=J*&KnYtfXehIvIKcA02Hyv`(Tlz99 z3I(?`g4x^kc(CC^YVEB+#=mnUZ*CpKMnBu}d?RIY-M9*?FWCk@-DY@E(^fbhl!^V- zTJWC2QADLTmlx&|*kjfX+;mw9&x%#xX5dOXw0!hHSWika`k*R@f^pcv~$_&H5}3I^%=anB{PR3Ez1o>r5P)?~RuT zvl2O*OVm$&1xffd1}14ru{y5`&Md25frC5*u9H!hT4tq)PIa=r~ZOU;1>(mjWvy?oI?v_7eWS z(|{^&BB_m&!IDFlxV=^i`2Ln?(wE>P7J1f-<1X*wFFi{}HlN&ay2e@-9HR@jr+p5yjKb#7#H#Tps1!5X*$@W#J0TS@&`$ zRx$N2a((DYI^P?R2B{MQD>VsbpisPZb2OVXu^PV*jKbFqwAe$3VIX_uCX^mqfiy<^ zg$)Z*alyzAB)e)RopS98?RdY1Jm1;`+a@LwN2VhF6}?ET`gfu@uxK;2?>))Y94?~9 z?|#xhw*Y(ntCi4{VdCG83&nFXjKx!52GC!eAM77F4dV2J`KyOFp@kb~!gi|*Vvj$~ z;+k_(E)K6o;V;)-)B6@1@FHb3{B+|{TsyIiK09Pjj=btXWf?alzdbTU&&wIVb-6t! z%)I$O!u;UEIY!T(O2HF8Ho%t$@8C_b3M%QArFU#kqp&5K2#vCZQ}0aiD~oJy$73-K z?>_@2ezI(ez#jg-^Z`z-xWYBpZICD=P6Y947hdsa9&b2f3*RJv3tJEB<{oS*;SGB* zWFebk+Kpx> z$RHabf0?6lfWFEUe5v7zkhSzD1YLRzYg+f<*7yOW*tZf=9+cu^&p*+pi*wN?_5ejR zHj@MWl_V*_oK2aOEKdEJDqcN1NPK9u4C~w@X0?jB5TB(BX?stRy~|7Jv+h!`P-=%; zW&W(`wHD`d-~tqGF~a4CD(mf*Ev@&Ny0Cughi&B4DFyOd8K}jG!>Cwh85n)Z0BWs3 zj($7_-Ex~@$o*C1pTkgc^57%fwcwIyP1R_8Y*ZM(DB?Nj3$uIKkDWBe=?2MqJD#}y z90Wg)T*P-vVsWWU3wD+ZV6XiJj_5ZjrqVHvM8z13@}C}Jwz2Jao6!T>wA~*0O?&`Z zXaDgZoOD6=@FbS;t)2AVSjd`ZyJ34fTYBNL3}+HyO%hH=6CE5*_j(qv)mQ3Sn1f#Z z&h9R{)1VUH+j$7Fe6#xhr25$qfgxabJs&HT^XOM;Gd53DBq?*W0gwBL`xHFT89o(9 zlx0DCw=9XAo{H~IaD}zKW7vgT!6;_#A(oSP7;RW#ga#E}79DH~C1rc2GsUPv5PyL4VQmj#aCpLfT3voF!*zem(*iJE-^}$o<`W(i-pZN|dJF~%N z$!4fNAw}#)xgocCM&wf##z|2pA!n^WQuuy`x~)?q*4UhkZheABD(qqA$CT=WC!5r( zejHi9Rp%aa`0p%}T5*rvOdm^*=yx-vjUU;)Q0ICH_kr!75QyQLHUwsVME+c0{n$Uf ztjRfz{dt*3eD>56&5!2T;NT<@;J6mwP)M3Be zTHI`=Pv$SG#v=9=U&Wi~NmNagC(AL5rxN5hGzJP%0&tkL9WmTnfQz*sA^D}c{-D6NoSBjkw?g;?>}kEtTI$$+<`1QW9ddA@4CIRktt58Vqdy? z$*1Z@fhiV8mb6AfgQGmj&YjAyT|Se{$dM%zt7XVu?@H`?Nyrf#ScEqEf2PT5xpy_a6|(Vz?cS{7X>Psif;IK_ zI==PuF4Z#qn>S#%rz3IyU_fXa!mlnwk~wz0XvvWaLY{LHZWcR}wWk`eZCx+#Q+)<* zJ^FzC%#N`QiGC!Y^$UEAj}tthCo!J@_^!ag{F5??>KZsed%2;gM3^xRimSn zPplIn(EjxvTO8_-=m2q;PRZ^xZH|2XPEBiNN$B_(_V{lZS#9``Bw9VD_mZW_;Wg&u zMeQ`)R_;dLZ~ui?d`l&ZH(rAz%bgH?GM~wTW*8PACBctfw*-_vn$Ihov`!^D$qiCM{7a{BWOd^{!=t}O9|Jcrw8 z{L+1Re9;274aUOwxk~uP^l|9(ta03wS_PbMtVb5y72%CvYauLBo~?Jh!ntI)6Z`my z`1RKh0w=6R(sX|au`u39!q#cvTq8F;N-3Yt9`OMC_iB^(P1<01VHugUl99jDZN+D= z7^A7Xt;TJoaj?Cy4Vpgvgnau0u%*GC{Ft4ND=ZJvO?`!s$K4|nJFT$W^k}RY7=Y9j z3DRFzN}o@8%-4VWPKV`5ai;nfc(PG33cI8y{;a7*L=#g8|G5{h^ZJaZmBx_Ir2=2M zml5}wd&nr|6Xe8#qoiLqi^!ZfftJJ_hB-n%dEfT`Amgunm$)(M=7dwUG{Ntc#m0Stkkx|rg??K4EV<8yH4R%w#{^aYY^cf^&K`*42O z2y)!#8LTSz12!cDUtW{T_vIZzxnRUQ;vBNl@)S-}{g1dLD8bh`y42_WG*D3IvE>kZ ztX1y^71BFMuzR^^`7}LtMaYV_GCy;&?n zqixw~^LKFWrV`;c6d*hhhx;53K~kkIM7qh~%Dh@aBHPGCGgCs(E7H)n`vqKXrSpO<% z)_(Cj?0F_+Pj{B0f4jCq^N$@uzGg3uH%%x0@kfYte*^w~^({WCF`n#t3)q^KAf?j} zQRa-pWVFCSoO(W%+%`yu^2qVv-Kx$~ZMG0zQUOe&+#Qo|br^e`rR}ao3M?4qfuZuFFWx_(pnve(@H;u?xJr5Fb z;0H3xionN8zv1(>53u$q3o^Ax22%rjHjUVlj-5`-PA8m|YL21*Qqt(xz+~pqxtv@P zIIk;xTS;F<8+P<7CYe3MsCs}1--~{Wxfz(`4h&)ug>88Hl5&u3nuotWd<0vUD6v^< zz2V)nM{s70iul3%b!6<4+vKsk6IP)mwP5z$`6^p(vKp(ZgrKC0;c)PmkePXO7JI}!;|+TDp%F<* z_|D2s+-qq_swNq+EYW|gar7QZX@Une_N)+knS)vJu>`ilp%Q#X{Y4eiRf#O;L7&J!P3JbJ|{Qo;RIT>iic2qt>CtR+h7NPIf6TgLCEIfRE@pWBGqSN&h z0$qR8i#p%=WA~oYr%x1ExNQrB1{fm`*M_$nukC8#*osfp3Gi#DLSSR z!g}T%r*Zm1Z`ynaesQ&)OBx}9^&4e5=eG(>>B0!);Iz86dE7ZYqkSQBmgbAwh5PN8= zk$vyq(lxgn8I*h%Z11sPn0yHTH}e`4HxGqNRuU3;K%ceTJP-3Ony{v+GuR8Rf@W9A z5|6{_qLs(b(c@Fsamu>~xXls~4WFfjrWEDkwN)E%=A3ZE5|;;pP;~=l_Q?BQ5w6 zxQTrpSV7mbAc(T0P;qJlh(~W`x!txb_)#wyxg{|F?grK?S;~ATt!E7;^%!$k(cceK zaM5fL8)+~M2if=1OB=$$KinR_n=u#1Y)imHhTgyqw{^kfxC!sjV#4};BS0_1m-xGF z;deKR$o@yWiS(Fqc=p4VRiyj`^lvn*5Fh3%25Ncx7R)b*)Wlck>65;aM#HW|NOZRS57O#a(n|$D+op>^p0V(FZ-xQry6KCK7)^j4p_r8 z2S>eLDEO+s;BW&Y;v;p9Kim_-2N&AYve*LrvULH8_R$ev{XCNR*xHCTt9Zc}DWBA`D7q}6afc+=OfW~)Ys16+g59~AH`}4!l&JJURnG^BD{RYfi zH3Lo)XBL^Wj&-S*;bocnD7Ne``SL0qTt&5z&+R5fzTsfinZV8T7TE0vYuUk_AVzNSgyc9*B=n924GTjhuk`nMsjq`#lxo0WL@Tu z;PO-*Hb{J&-CXa(h6NvH<@M88zt7CY+&>pbGrhT+*udW-435TfE)Q41de58SI@+5Ia;5NMivp^%=%zRSbyBkf zhp_*HXvp7g&TiVo(ux6z%ej$62xSePqX~*7?b{kikCv;gpFWJrR_N_;) z_iCti8)Ctsc67V116&$aiuBIi!G4!t<8=>P$;O_s#RDT*XDDq`~I?>>6 zKbDL<<4II9MWl9YCO1RBn+v(L4}J!07S7~%Luc4@{5ZZ9ncXr)&!d_-o35o?#MLKU zm-Q)VeeD59-@N2It!1FC`8TJcRL>W<-opicSyb2DitO>rCcPtzVf1&79egniY$B!E z>h#&7%N7u!z*1J zP?yUEy#1RjK|B89GUZyy0+NB(9H_?IR93*ou1(Bi&>&`5SV-qhx(JD<>fl`N78F81 zLakN~zH}{*{S*Iz4c)WZ)arZ43XMh1YJ}XZQ6o0Z3-PXV+0d9YmYo{65wtX$!Qp`o zlxvKGr9sIMRvL|MdJDO^f;04LCx$8A68La(0QL@31_!Cb@NUC4?h=#mS*z1vkDfD| zOPb(^r5&A|Bge;t%0bexDn3x~Hk*7fWd+TV%ujA83k)n~@ypM#V-Fiynmu6)=O1Uv z69m8aH_TQo&}F|rcH$R{M{(YN-0-uf19W{|gAwq{=!ycLq_*?i@^mW};xO|6~1e8d@d#U|!#y$eS zu*!nQxrylR>k%;7xs0WUShAScf?xZ4qGaVyH!g6)Q>w7DgdTFQrhRWuqc=h(W?7*I z9n0+4`&~(BMp!MLyxEu7Ea*pHl0(oiiy+dl*^`}&i{kzaxr;*0K4TqYdtw@BOtyU5 zi&c$>Qp;Kkvj3hELjT;bQ)Um+i93n|j|yBa)hoEub0$5?5bk+5NpO2Iy#0MV#)syS zwfQaVoMs$*W1h`gCMDq3H{0-0O9fiy`I+nbQ;M$^t`;>{jpCxu#?TzC&GgUua%2>z zg+g*)pq(_ITdFxzaMZ0L+x{)aLnn8FFy=)wQm4TI3j-*!*JcXuPvhpt%~(q|}eHP; zja6W~a=bD3OUMG$=wQzm59qHQXW_`wG3<=MAm0&n5~=s-k+6&ntaGSreQ5CxcB*ow z(3ex8`fJlE7v@c9v?p18L5d_4D}m(!e|RE(jsqM^;a*KWj`I}yBz|esQf50oUVV&r z9g;<%+#Hi}kC~oN75VmcH#x5rPZFazwo%CDW}9Zfq@*HlhpY`VtLbOkr&Ti#^(V~b z9mnSFy3G=Y!^+C%_)Wb2cFUET0h9Gx_DC4h1i4a(!-C8yHpYC2P zc3y;yV|M^KUj<%9BgiomWx8^qIe2Bl{lKjH(q0N z4>mL9McRUMs2W_4nUh^R3k1%-8LN-kM<%>oMfAIpNYd#D?D!%ZWq90x_lBx$tNAd2 zqdtmN^?$}2&4#nW?<(Yqk{W6Cn+@{~D0+DQ8yy%I&s0BZs>N|gx%)e zLgR(&=+{hb`0u?GY-!uWg3kR1@dKVXe(i6#kT!^QK6?fC6dcHD&WgCZ4##?}A!zn{ zh9?LtlwoTJ5IUvI0@Am#$b)OiS7UjSc1VRpNcpiJ6AK}!a55eUAk4JEZg(Pc-5X2G-mk@W0mpEj{~D~jY&&{CHxFJuodN$YO0g>{ zkKpD68FpXV7slR;W|t#lS#0zq466rluj^3U^(lot6xfBLXyQp#cCh;(({J00=Vjl6?`L@-V|tbq9$AO~)uv&mMG|CZ7)>spjmE}l z@+5pYgHL}I$kbUWcw%z}+CMrMMhAFNr=|<|QS^K~T_%DAw6BHWou{ex#;f?VX);tR zMB}%*7wAaMgCH*v3rwyOYMYtFj!nD5{6EApm7RIy;F=`T|7!|)UtJ{l4G1agiUp4! z&tZ;vDolS;i|v>mK5OSp7DT;8|Gs52rAB?Qygm~4UK2ynOo7APu@bum3GUz1MlfxG zFRa$i<@FmSm~(VQ{|+@G{nab5@zF3L^H3l9GWKo)c3!3xsT@PDsp*k;le&ZE7Vf~gG zR?a~tBf`<(i5}=&!VLkF{2Pv*l7qEv`z1A=Xhwz25awspk0pTK32(F)x z=U+a7mM@V-_xkst=T)+Jd+=X=y@NWa)OpjZga1Q2A7o%#j~7^B>2kWodp6B^V9pmA zRl=e#$I-55C-J0|IDVO3Hwv<>$4kQq9wapamaAsaj3Q7-wqz8-eFTNkGPuw=gG(JQ4OMeTacmvt zn@`NcUW@0WmcMtnTMgUb)|V!%x+0mTNdcA#lqZgN)ku7=J?y!k#@%$?gOpPe>F5K} zRywhEw@PeORU>Qi%!;N=XsP@a(mvlu~GrMp2YfnrI$0nk#=K5h`RV87iD-uYDS5 zM5#282B}aQG|@=!^L~E6z2~}IzVn=O_Fn7vyYF;#cs2_Ywb!80?xpl{ZwnfuT}yh# zNgRa3OIb=vHoIIkg&nWq;*~``;B@4}CgrnQnKz#OQNI7)|GGQ+uZ)dFBglQh$ zhYvq!VOe)I=J@E7ZEVpYahp{>S(z!#3v2=98cT8f^zHcVwI6PK)QwdjI?GRT?vB|npS zf7xf-ap8%m8Wc>F_#cvy(aL@Vg|d5`36+1&KuBmf8Q>5h=&bx|Ymik%8x(fZ&Ku=) zqj5dCmU08$uTz7b>Vrs+E?bDNUJ{ux_8xxD{3yCuw~#W{o&7ogRxJ51lb!xFkalDg zK;D<9kT@cc{b-i~Pp32(Gh{oR>jRr?ZD>o|;FVv^J5V@u&EO5L7|9bBqnruwQ3);qbv0pkKx^hf(FbrW@;nSt}V-;KYzCZu6B9Ytm669)| z_K=1mZ>og={Aq89%at&339>HsnI!|4P%Ug~qQt6PS7`fXAPaTU zp?k{JDebnM(x;N^Jt3PGJ6t#)DRe3LQO= z!lK;KaN?;WH2qPBAIs(n(d!iXyFQ3HP3vIXzeAXEekE^p3dZ&!J>>22*ST8cWxi+r zLOz{!=d17ah4jJtkf<5SmNrdB-N$43^ORuj^6oF}O8y|*mvDsXWZh+3o=RQ^Wkb4r za0=Zm?e^C2cAUNP72H^vPx4MZ1e27#_|9&M#A;W>-?O!O!e{As(jhoHT952pLCLf( zws3o?5!0G75pV2l;5JsN{NXT1{?xV$-?!747c6+Gx<^<_JDw;=ekbj|T#u3Pm&awHx(lFs?R==nu}Z6+`Qi0c#IAUo3~NaOFc@49kHfgmp>-`XDcv^Y1`q__eQYYE@%4b zhNSt<7VHamNpXrAEbX@)V{8&xK;Ko+_OBh=zkiU)R2zt~o&`ErN{pvslL)HQ6UBbjlxd|kfaB*J4FBatmHR~l@v-1F){jLO z?_*SZ^$F z-mNgB=hv>K(@vOCd(}bI=;l=ZR62Y9Th1`Mq6eGjmk$oPH{tbEYY2YM;iS7W(`p<+ zs$l`f6-D9nm_K-Xlsli)=O|wj)4{VB$MU+*L*)wFm1!S&Ff>WCNB@_pnBGI3IgTE{ zuZ?|#UzKO^{lkyqWknmXj_VB}7W=40>?%G!RoX=;)-qSATR-6{&pgBM*2`|-(%Vp!=Fw!$jIrcS=_J0e)B`@esA0eCrqXA|mgpw= z-e6*!YjyHGvMmC zS8)H+F>#~PGWdvIG;zm0cE-0J1`WSQx^B3``Y+i9bFQC(E@kb|mK}uGPtQf~y6*hF zQXw0pFr*+H{TiPzab+r7JxPoIjQ$9Y z2WLW;)CHjFz6PuM9YvQ@hb6C{Ih)jEN-s2Y!|{KjSv&TncUD=EWtKHmd+J~zCZ-DS z4KNXZYY1e(r2$Nr{0CtZx5FgGo-lK2CJR%NiA@u-SYUuBXw*5wH)s(8b9G?(xLXQ>BW;Ao5% zJ>Nsx)PoTFYdRdTPDhol-(YFfPIw-?8=d{51*;Od_(tzKsCFF&&KD$?O{FSrP#J=S zs>(#+>|Wa^t2Yb6$8zlTP3oFh_s4?#Q1CcF=}pTRy7zB}(ADuh{C#*7`dKSTjHE6Y zI#3(eIV=~iO;N-A-F=`s{3Y4o?SP}ouLze%Ji!O0XR%W=7^19;@?c0&(9*^xY0Q%?1%@g|5p}y~hIolhE+3O6@{$|E~ z?(gP`-xPW0(!)ZSzK`vdHBqE8cn)m3yMs*f4p+X+O!4XBk-mE4}}NUCaZ5 zmn-p2U%WV!NPgZs5&ZCNXP$4JhKoMN;jyQ&!mEe-uxG_(G}NfX37fTO-^+>6CHE&O zS9OL_4ohiA=~9}o3V|xTWZ{R6arl+P*gQcS>PpuLhAk=(SKr30C-~semyfx&$_Q3$ zatua>uaF)78Hc|mX5hWDR=)g8CI0gC6vs#GhySc4PO^0NpIx+>^;O?4i&UI1?6+IM zw{KIRanu-o)Hjg8!d6fdb?Ay!(PXT(1{gYo!9`Ow`Pp}kT=Uc={$aE?rcF1a!@dWx zz?9zHU)l@Ssy337KcjiOswK;fQH9*Gglfjb;5Uy{JXjIR*OZsS_fIM`I_7x{Ses*t)RjhEj#FK)SP0{J%UctXq<{1aS;#T|R; z*4ufs!|xDW+rLR{ujxsn-Mz6!E_Jg8snKZR4>Vzp?d@I@=&-&IA@zJO`J6qIwe@m2C^J>wB23(4fP*x)qT-cb!dR<8qK?NCuK6Pn=SJ@3^MwNbXXj-MkBsJP zdj#{{|C$jsXLCht=Yxi7;h3XS`PJ-?xHs-P40jzzA85uiIADpZm-_PeT9HKMpd}4j zyb~6iHKXmd531wUx4X&bsG_Nu&{*r=r3RSFMD9xy9^M!+*kh zr6_*D!kQ8%37azEGE8i+qUFl-=@pd^aQJ#OH2io3s`-}m`aNrUXLl4>ZdE`N+LteF zo5v3(KH<^g5RCGYcH;TQLRgXs>FT?Zk@9)+YmN@`Uv1}j_xeO}e?c#p_NI=l(0Ktr zrZz$->&0!1cHl6#?znsAGo0f(4&c&!c{(iXJsO`1mzE8}tRW1@1Zi!M4;KY~kI`)b*b4ucY>xyiaPCH5T z0;+9$I<(^D@26pnZU|gzJkC_#WDEX15>Rh>I<$H!(mhkB($ll^>Ac1*puKPgJ!QX- zsz11ZU$#X6>!CxB&59@8bAOQ54!KxU7Yq*vbm5oW7h|F(rEHl19htB2k&83k+I)g1 zEK|YZ-4G_aXW%z`Q~s@UCQm$bg1EIC-oBMKO0JbCcWt;b?I=W_o{W(<{;;npv7~V& z+B(lYA*zqx3%lOsiAUf4m3g@@6Vf_;u}{WBmee(y=!~Bvwjb`mE%OiK+l&cJxFg5g zdFt}lXXnZLY#c0i^}Gos|6)jmiz4Z@tuGvY93eaIwqLBBP%4~s^}=+&AncdzPF!D{ zlf2jRuNLpdEsbgD)mi}I&#mY<2UmQ1bvE=)cmh|Ak-Okt?%J{fM~x^Enl3J7 zFP{GecX?m#{4$Mqw3Umo70US4*Z(}-p)p9)Sl(?RhrXVcR;SK4`>lLZ@a zxzPeTbiOeyND2|vwx~+^cy+$@rkCBJI90oDTf5lB{bxpZNc{Al3kpeUi<_v_eGsJJp>{wzMJzbOHa@v6 zDrWr@-b|M|sXvvd+q6F={?8wY7atC>b-x}@g*Q z*VmBg;hC~s_xspR`aOnue>+8N-M`9eK6vwcEQIGeo`Kf2mtm3Bc$jQf44F3$i7SUE zl5X#N@ekj-^U79Ty4}}7oKg2yRMFYYD%NL1!`(_*`&|fTWAj(hW3&zS$@zirOTIE6pRN$R>>!`kyMsLqn?l`PN5SnLC1P1fo9JJt27_@g zuoZ#0?2#Ep4yj`X7c}^X`3n4gZ3V8~c8u?RdYC@zeGH3-=+ZsA*3&fEV=^%#8B(n) zKrwp;%C1}$CJ1We--0&TlW&ovafX0}CU#IpyTZx}9r|-*2`n3@h4WvD@JCIbEwu3< z@0UBlfR-wz{O%#sp5QAKHW$hoZeJDSzi(l?j``sA@fTR!xRw0em|X7IqAYy7`V$t{ z)QSo>kKmu6b{N!LMY?{i#nsy{qOadP_)lV}E2e3XEl$T}b3T;`zTH0Y`Tv~c3sp4b z<8EbwgSX%fyhtG8oi12&NxcE3S?0LaZ-X zLSF7@+)e^)ZYhr9r^d|W>T4}&gikeGG8jY?Ycgr<-}REOejv5-`3Uu~3Vib(3qIaG zn=39(=Gvwid}U-ex_pufpH>tKUfRYqpxFb4S!I#DZcb!DLU&wUvJaICKeHH#CHX2z zN8(bP#Sw#V;Xh#n*D!OX*G4(peO}o{?e33(lYcU(XuX%N9Wo5pX&>b!_f@&p`(yC* zXeeb(Cux=_vD5XR@RXO2;c$VfJG=*BFz^8?1Wr{5jpQ$aNU@@MtGs_?G(YQn1Tc zE1=#_+Id=TFZsX^Gv%QChWO9fg8V9r7ZceZq?$uB@kpPQ7@KQGa0Zad&p@-VyY8bPMJbNX~(YZBY540H&>2 zY}bFSmfe7w=hXCfZ|=Udhz*+&0Fy(DnU;ASwjBuMpEhQ~THV>OeAx@wnX5s)cTE>+ zOI_&cwpp?+*YxpH4_D}HohS5qEM<`|Y#~!e?Sue`!G;~$)ZkDqPx*X=%X|7t9IQDK zbL5e*e9k?pw=QSc#4P#& z)@o140P{o`-nTDCbbBkdZTdkjA zT;H!0RGj-_{KWKf%ejksJt+p zcF!%RcG`+|>i$E(>6$vVv-04Iyn}rCdL88hS3+~Ay8O3y3;(cjiQJ%b8@NO)BAOd@ z#2Ynh`L#hhe9lRRw^At&XuE|$<(fi3({w)ibUANw*~W#Q6dqX4A;C6L5HGP%w4!X~ znhI+2af2gaX@j-6dtnB=c##4JcD6AW-8H;hA4OJ__Lapbse!?~Frhv%omY1_$XB?p z7NFFGEAO1gYJcU*s&c0as;eDw-(9gJTwyFRzLU+1oJZ2eHQF@b#xR;}6OC{0SfD)T z1$z8A#$7x&%li(TAz!j_GG9Agn{M^*4_RCN`O#WKo;mF~H=YwfpZ|%Zo*F&r@iGTK z=dKbe+-bwJm-_Kd-oJR3<81khj6afup$}hLd=mAVQ;2-kwU``Plud~bPm!a(`{u{ZhO zH>p^ls7^OM_2ws>I3nedyQ*DQ_a4@NuWyNW{?gEVMJ z02V4(aAT9+^8d;NE}Pp8_t_ZmQ#bASdJSFbW~Rn-;{~p`Rlw%FTE6H)2PgkTAk)@!`ktr?jqW)KqmB_ z^G8sa7Ap9yHxYk&-vXO5NKKXQxtdPb5XNw=qyQSH|UOrfV9aRdI^VcdWv?{)k$RwWR(Wn;E!&rkF*_-iA z5eo8~$41G6I$Y(xNjv1e+vB8jZW|}|l5^4}3F5W4VoUeoY)wIp|evr2$6k}+8>kn}iOt>&@1>KR%AuY3zv_1I*U!z1&(R&2l zN2~Hdt>%2>o{jwRg3}<>Sx~(%ZP1c?6ffwlL9c{DT%S+}k30|Iim8WSMrgk!kz(CL4d%LFJ{hG%Ip1-R0~G zyL#Bd%N!%>E#(8(1nz{iQUkhXRusQ7;|CtPo`Ui30oG$ zWA&;*e9zu1D4TK)y7cNHfSn>L=I>;#MSIBd4pq9X;|Pm8oGT`(IpFcnJuuc!4o+^0 zsF0w-<1#eBFJLNLs@W{F{!}CkUpR^VzBHP(%+tiOHBID*Oird%ZKUf0Bt}L~8@}!E zdbE4TD~P5s!M;GKWK9wD`Z55S1Z$U89$sf zPzy}-W?;`zW3gw|4bpaMj93}=4L+xN!1jxcY|s3W*iCOS?z3xQISwD;)ci%%*v5=} zwy}cVyUO5xW)!^Ah-cXvg`#a}J;a^zhRvnJ$bx}3AkD>u0pn2J2wGjU(uO=|P(RPJ`QdPuS~AR~UCNL8}4k5O&sxO|C8=lMXfrv)#jqS<^vw zO*RJ~1$v2*ht%Pu!Ey+{JcrHtzDm3)-EmBJT*SI(Cb+Z08Gl`HLa&|m;-g8S zKct(SeS|%$redo01BmM1$-cM+;QHKAC=6dDxrtF2Q5?%+!uo@vO&$7oHRsVA4C&4d zU3ln>N92yDp>!wHpmVpaBmg*?s!I^7IgdsJ7o(a^H2gPP;^`7U>YcY6yhQsF)VDhD{v~AdF`sHN=ojS{eruEdMBMfY4#~=l9 zhDAPle~*NG!?hCM2Z(RkM!a?TEfmm5yNdrd+x<$r1fI6-(6rB=&MMgk!~5sK5ano^ zo}0lRN`sHV;z63ar52vnC)3rUBE2x^7z(|++wCa2N=Fa(5v^Y-(9etNAOeR9{fm&g zw%fsR;{mYg=6Gp-ato*M7vR`OgPzcS#_kL*qKjM%?Mk&?(7Xpra7_LYIC|j}2Bx3k z9&TN!!^slr*8Ktcg^cAU%|$#dK7*Y#y-HO()a+(Gng%!b1k6^Z>?C``#!o5j~pz7EJ%PSAD<|ohDJ3TYFy`YjlU$1O8 zy4wJ|1D4Tt^BzvZ5Cwmoe>+pU^37 zF6yvvL}OC{bbH?bKgbL?b;}L5Mf`zQiS6cPdI7>-cBRLjJR##o{a{z0ycb){4=_Rh zHpxCYPvT!xFsD;XY3`(f++{{jo>)4FKg}9s_j$;#61OFR&~Pe}1?^eRFF4(!e@0}3 zXXF>0@MAqqpJit^uC5t6zb=G?p5gSoz6D+!rHp&xpJBDy0lQ0lzFokkqjr-WYKTlw zrn`IpA$I%n1^2HVbZNK-olzbDl`q5Til5W%NZ@3wZaORGDNn(5RqwF3CQ{G2(r4nd zIJ$bb9X+IVA8u%EgNX$vV9q6Tk}%Q|4_R1AX1 zEvw+elJVG7?l|Pd>#n{?o5JUo`c$4BY!d&(52`oo7?U@jIAm zTP|4H#Xw(^8PGQ0fy!A6ss3e-HytK}f74Eyd!UlWAMQ!lRE~$t)FX6n(FnUG>9=V9 zkpzHt8#{K{!mc4q6drvmfw0ads5PS>Yz$pb><4=91sOl+ zKI^;mZtPFG__l^!c*=LW;QTWi?JOr896SJ)?z_e875#AMUdhwhIE3zL`AOeA*lw3t zq04i%2jFkVNPPd_EBcr#+3jmvMQsjmq~?oE?A$sX?KT%^;F0sz+^J6`G;a9~!6^-( zkhTxXB0PlMQ$DlQ_XDA|XNovx_esX}lkwW3Rza(Qi|y2e+*{d;AHQ&k9ay7C4=WZi zd{)jM_W6um)IG8JyAK{LR>etDUw=ul8y^+2fM074mY9B(qRZh>QF%JR+XiVD<#r1n zw~9Dp>3&qT8OY80@8E?;SMwwC0Dex-o(Ja7rQ`jN!;1A&*pzmCy_eDjPD=8dBbDS&I*jDeJ~~+Deij1u$Iu6JuJRX#`}wtzbv#nZh`3)UX7d(0 zL*Otyo)|NdFW45(JbbZ)&K8MW^Y z2K}9m7Gr**vg=81^QRs|K3(KF7t^s$GDbZL38^o%`s}!4NHYmH9+IbV@h=Y4-(wsca#yaH|CU&;5|rO%mSO3iP*LFqR#^2*%G}DkNFkYx!i$XDlH_@W1iVISnU&goykG{mVC)kogp!DIdgLKz^87L z@qT2b%|SmkYLKSF9V8yB+x}VHD&-Z1GfUF4O-bO7f!&^T1Rr_n02{RfN5}T$t83j! z!3T*!Ge4Fq{u{>EcE7`S?Oe+{1}UM>g>M+Vz?omA2bfLZec1eKF#UJ1QW#v&&IW4T zXOllJ#BR-+y!`NC?p`*HhhN*zhsn7N2kcs?j*3t|MYO0RqN*(|=?PJt^gz9bs>w@8`3 z3S0NMg{=DcD7^7(30HnIiyH2v-H8@|n2jA)G``6EZjoA;T> z+taP(>YEJtrDq|c(fA2OZG$PVT&BjC*onN4)kb*P_8hJSb;8JL6S%=mXB-_;Og9!C z$6kXI*p2(6F=j?Pv-!7&GPon&mR`c%k>byH18m5+^6Endr5Rt?o9e++!(tt zg#ofhoe5;6p)sohFYsEm8Y3?&anJBgZ0_nL68Kp}V|yDc$NMcH{Z>8R78oC~?z)aF!p39=G3~|7NSpLR4n4g-yop+fT zM_h|Dh}PBsQt)I3eO)b(fg#7xdFgseBR|r8<%j6o=^pU9p5lw3ET(d`n!9Q&;o1wY z@(HCmPd1#GONLEpc?+Ghi0*5>1QkAx<^JIzUMQ4%ZnpZk@i?hnO_BbvpkVXj@ z^2=Lm<>d4p-nprX1!=air)C9oX3$@H$uHV&#lU9fXh?YU^0iz&?Ijs&egIQ82J)=H z%{;DTC47{hCp5Bej{bt9;5moEsOV`4T)JeE&&I33mvG}@5 z{p-oI9jHldf9!F43K|~Ujl*ZRk+M#8`f^eZw7m-lW3^2Fr%Hi4@K)?7TgGp^lpNi6 zdb02CXT^}}?r0FwU!Ese$VY$LD*rq5Ah$5Qz;&83xlf`h&saDJ+v?Ad_q|W@y1_4b zmjj2n%>NKZ=q0n{J#iQ?XbQ?4!kAJ^ISV>u&W!eH;cv@mEPX!%Ev{2+d^!eWTHUZ^ z8+7mP)QNs+at-VhSI%JeoFZky%`gJvvZ&e5CvL8Jv8jDTca`A;*5t<%3DCORM@*)R&+0m6{L|;Rb zO)3f}?wbym?6p1x{K-s?{`!1J$V|b1?q0fT)HUij&BU(z4-I+MZY{YbjFi`Pt)WU6 z2iSF4c8ZRNNf^A(n}$TX(aqhP*xaT8X!xQF4%P2YkDgUhTz)l zA9y`(9Yl;NW|33o!TCeZ*jRT?(3phm--e-7Zc>C#uUT`o$!BTp_6%z2{tpD(d${|` zEjFt6eNpd$varYRCyBeUPFOY35C-4x4GP6+LjJA~%+@;+$L&+Z>HQ7p_CIs+#hPR( z=N(T2>ibcZgcIz-HCOm^Pn8~TSK@f2Fa4S~NK71O3KM>Y$Z|t6NX4@QpgXSu`DgQh6mlWnL7C;Cm@> zxnGEox-|eEYCZ>*5t{U`^#o{^{J@IzC#?M@1@H3?h(OmVp&^}J&eyA2}OnxFm96`?cb^mzY4G6wV2O1ey#@(yP1u5j()>QXW))i3PI8SKYl`be+N$|V7@5HS>_Gr9&9*oo4l$Gkp7Rud9Ke1+n7>g z$e)(Vv^@Eg#;RHMW3}c{E|*TUi09**yNigo#m35 z^E@4TPwOVz9Bm|yuM5Sd;!?Kx(gE!5U5sxW7qH~)H5hR)hX@tcxMJmfR@2rCHW@4P z%fD2xYF95P8e@(gL!-o}?)%~Rh4*-LKtJwnmPFzzZo?d>6GDbgvh?0hfqexjuz37Y znD@eyKHPbLhTJW|WUl~THbex3KSrf9&tYDA6R*lIQqF@rP?~ z@Q(w%a73xOm@_?%ePaK?dSf@tl^Jon+*pY4Ya&xhrO&0_b79`)T4pi%Aw0J%h0}MI zqOR}|l*b0pfZq%f`DeJfegiH!znTo~FY%OAV<7d27re`kfrR;cXt#%EV1D)wuINkx z{rC;|t44B1r$>v;ArhZFV;KIrcaLoF|4(#rHK%*sg2mhl#%|9(Le&NwWP=>1vp)l4 z=;m3o$f@+_!d!Sl}*pKIT^Ou`0@!ahR#ACu+AyRWHmW4R8 z$M1izXW5d^nS3NaVo)$yXvWuPJtAkyd&2PqBp*GPu-pC`lD7(WJ~g_+12;{fJa;6t zeH+P&D(}kD28PP^Y^p0+RTCf>Wwi@sdAh7@-YWL=XD>b{v6}g;@5ZKO$HAMNk>b$R z)dYTjz}xXie!NoS9`0-S;1!dIe^sNHB=IhGI6V=jS(dR?8?Nz1w2gR}&lYEB&10nl z^5I<0GS=gw0Xpa}5r-=ok|EcITfGOe^-P5NkQcJMwy$lzezJoMRb#RZD_E+%f5}Sq-jWX@m8(~`iIzI*_|+r~ z6XT4TcJID;`^qS)GrELbJ}Yq-vlXGeYbdkce@a~c+Z#)bT4i@V-Na#@X57EbNUXaV z$#kBW!$SF4vH$u{#Qpn57Qb^K^q(}69jjSH)eTF@eZRRl|I-rQbwoMQ{nHDb4}6B& za}xyX3&-KoobPZ|a?NC$d?jY8YS8>9PHg`9i3|%$W!o+r!iepUFuPEdZyk|Gdb<^r zbR_q{K(dVO+|h;H@4cBSfB~~jdM3=bH^K7QTcUrzaCqRgjHY)Efh+pjY}d2ZY{uB5 zGT*q#qBeO(Y?aexM%!cAd)p)Mx#f&--)uEbTlAAZ$m(tDwN4qY4_D`-6+_vuLqmAp zW+Se8%8qSFA7K0a;c9VbLNPJTt`L^>xGVd-X`-mQaF^JnsXw`qa#~#eGLL*3^&afs zWr=5hUMJCmlBvy~v9iAhGtI@eWJseUd(}Cdt8Y;jOUzqX z!Qgf}^w|M6?B*!4-+luv&HN+N@k_!JT^zt@=WrIeLdt}6yU&|yuaK=jhrtWIa@@`e zh4(s-nDWIBICp3#Sr!xq39~#UK5U)1VZSlHNcBaFq=8~oWjGm9vV^r%E@q9McG$+1 zp(3wN&@xlTvtu^d77fx7cQ({B#T%(Sw6~7U(4x;`akmC^u|Fs->o3p+L^3(uB3f;aOGz$s!B4*Rnk+$$@DfLGzncTg4! zX*ww;O^8C{UMnRwU{|&8^IrE*7~DnSe}87VC@Cu_sHrLZpI2W6s^GI>*~(R0mYe(f zZ1OR;va}sD*3!~qjLnpZwhA8qznAW_H!L#upb+ozker%7gO6F?o$r6U7tOo*Fr9Ag zqSA;%#Oj#}ds}q@yz(otTB!^7>3R@-q&#=Wy|*?6y)%WK=9y$mc`CkW-;RD$hJo(J zm7rjHm|c0ZK@>L6$cTsndOZe6!*xCBfgOX@ycMG ze2(n-Y)eyKXR&aP4`Pyo4)3`-2W!Be&djesE6=mAyXG5=mY8o*wexV+t)uM3tQ(|0 z&lF_+?&Ha{2KM0kHDS-gU^eZpbjJSrj0If0NlslfM3pElHumTkvSaEpSa8je=y%&j zzt4CCK2!9le)(RQG53J%?8pjSebt;?_TI$Je3Utl_zd>l8CLyilHLD(R^}EwpC%qy zO}A}MfbUYiY1X7exKni;9o=UGm+u&Z`4fg>ONAV~PW2+HTCLdF;DQ$W7vq#A6KKS; z*TiXQ66C!80!{%9BumH_w6q(TV%>ZEH@*f9N@7vnE)(B*bA0__8_pj+kST{*@`L^B zKzHy#Sg^4k{)-wz_dZ^N);;WCYi$QZ#$JtvK0yh?)XQE zPVlILa`lmXNmuE+c#jr11el}PIlnXeM?S^K*2Mp5o(P7kaCx$5yi|Xc zt>`%zw2LF*$5=OLUXuZpR=4m?%4KqPtp`7wIgFQSN8;nFFR;-t1~11v#9QxDCC;%m z>5XsGjYNc6S< zcjZFC!Y@d?(rHGbY)3=?*em41sIe?*qb~`XI8k0VVS;>C`eUwJt^@OfMzNxdqj2b* zJ5ILtpemJ#$aeK+*{0{&jm18&=(s(#MlCG2YJiI!Ti|(YwYdLg5Nz5YF$<S(r4T zEB&G}mF{vcqctvP#kW&Jg@t_&3#G=VAv1OJ+68kdjs@SDVI9u4;0e|#3FRZHQ6g!nZkr9jTuv^k> zZ>ys{e`_6!w-ip0<1ez|h0jTteD)5wF3Y4Ivk%b5qcOxzXAkmm3+R{6Z^4m0ht}E0 z;bRh#D4YH0>6nOB-}S(G$4N5s$3S6GS1tL{JX86_VJ>oYEp^;~wn1{X-z8Pm%gD<1 zLQ-d?&368(Vsn*-kw(Q-xEyhk7PY>o8`kODO}udkMP9~c@0W?$Vbys3ybGTHo9?42;6*WBP!=;FC*mJfmG?saC#nF$j&$XBQ zRX`yP-w{k}9OUHlx*=@NOAFY8LHKdEj#yLs5g)4~HmRS4<)NBP(#MJqx12+iyljuN z^GTBK0lL3bpb=J8R3pIwHz?_WWFx|HZ?`bb2_MM<2|DDlHGu?3&ytr}B{~+olvV8V zwXOBlpkmWw>8z>@n&!Ib_jW&iPCJbcmM`VbJ!&An>X91eS zW{4NEj?kSux6%st5%k^@e=J-)oOhpN&#N8h;)3iP?0Gz#O*UC5YVJD6?hc47QBl{R zVGS$5d&C~*urvo+PYea)rfD#=_^iZ9>Lq{dyPx;#;*D!g++cBXA2k2?g9KL|pf^Q= zdXv_$3R;3TV`4Elbr5Fd??C5eFHz;E9+!>p!(TgP@=B}BkwqKkvox3SP`yBMl5Vo5 z6XkbUHw_2zQECwDe)>Ea-Bp|Q4oYY4Q}$xP!U1&5;ZgWx!Y;vL&tSSnUFy*8H=@?A zMKrG{hi(Zt1ui>xK*qWP@@qpEx{(~>11F{Pp>Nvx@!Pg|-=`b@*2kWoO&!c%OH8xQ zHCBAf1%2UQRIu=GzXAl>FXesCX^9Wh&mq6NpY>XiAbw6C3rG4H&_fzwRM@r^?1O&` zZ^wn>kBnZtV^=Tv>1ltsj{bW2JCmEF#&;cd9oKZNrW#sat~hxP9_#ME{eEhk(CUWGoNSf;^k4TV&3dqI6C$Lb~1M0rrn2N%?QeO zs7&XBZrp+ee>q*mFJOnb$2{lYG``WIBVUuUfXrY2gFG}Ai0yGVKBVXa*888wB;{m$ z^uQh`_}qojDRme-XFj*=XM>|sGI>R6IbYFhE+2Md3Ql@_8Fim|;sNH3>1yd<14HR!?Ewx)0!`l0-AZ zEuhQ88gWctpoY8_y-bYx^G{2-nhBSAg|y2iFD!Q=DzXc2+*JA2up40c&y~hKi-n@C zm1w&(ga+2N(?^*dmCz^t)VquF^x1C8=+EYS{nZL?^t>-Q6xN@!VX-*RJBr)7KHwYe z^Z1nhL-~eIdtq7$;szXOZ%5+&>x8rG)&{G;&RL=R`f;_%vML>xTkfvY5E4(%~NH|mKq45 zWF}F3n*ftPc~bG@4f@OB1P$*wM*eQTp?uharDzUEpo?=5oy(%YyU$>F<-ecVtx?6r z?q>wd>7iKXeuzAuDe$DvJ*8Z1A2b};#M(dQGND5vUs&44&-m@(-Mc-7?!KMqcaT<| zc4{Qy;dz|jH<9gBKMUqXb~LurPl$OsjtvO^4_)R;!n$p-G(I+v8h73gyjY+07 z<3FHJEI!Hx++KzQ``AF^2n*^S(GT}7Nf%n@NWIQofwUH;!o6?NPL37o?D9|@EWUD! zJqmtaR6DjG&hY3$7k~W*BYj+D(>3dvtVt7H){Q0QM&V?j_eFLE_pq=w2l_Om7d=`i zP;;Ld^u-2$Q8mkgj?0`$pMD9Xo+EPM`}JF_{{RPU3@;OGgZjvKOtX~VbnGh6(0uD8 zj{79Oe4(lUNPzDW%}_t+jqvkZgb0B|_+zU#u^CtQW|Yh<8reu1Yy^5b<` zE|T|y(J5+A=)7@Bu{uoaw2qw;{q^4-<{#yk4Ho#)KZDpENya?~R9WH1EuEtDhpm5u4UW^PN^1dg-2Ica9x5WHdo$%O(fj4c zH+`a~nK*W%3&6UhLxCOS!bE}d%EM5DvQY3LM@r+Jh>5*a~rEw(CT4~`2h!I5})+c06* zfe!p-xO5+m= z+zGdz13vQHg2&6I!HJ#PyzZYd>-TA#aI&v{QTNaHVaoQ4TzA$+9u<25EB}=_c6pn@ z#*fRn@VC6fg^SZY6+3-YXv^&;uzQ>hRY#q0YN`{$m40}Ao37GjyqR*9_c-Mg-#PrU z)&P7=5A*UPXYkkcUbyW}00io3^B39y`0z-w!k};;fZ_ssRw11yZjy8J8Tq2>j!@dq zcMLDsn!-=E$MW6X2<|sE#GrG5Fj)5|!5dQd`@}=m6kRFg=PnRe|5tyZ>b()p-Tj2g zYDBOMEEBCdpAp;7o@T`dldy#>=e*`Hb{us;2>P@aQ@y+L;DH)^#F1DWP#zA;XexIz zP2j;(cA^AT7yC7!j7|)Ip4Vn@*Pc~;L0Tr?QZ46ia zg2tHafXjCZu|6vU(o}v5|3MM{*!)Adv}`83Kf4VXcoQ4KA5Vdvp+Cy-^D+z1W`*PCiPH&)iAtuNP29XJF%g z7}25uJ5klthL7zAAiK>^l znJk%)Pus+8N0)H|OvfFx$a87^^j}E|U#}cTH7uai8JI7fa8brmRJlHbRlYY0g z;!D@wa|#_8M;2P@@)VCkYJdJFjhFc4+jIYdk4}VO(q>5uzdfh_$Q8QuwF3_6=|vmt z?WlidBs{B=c+!*ng{~8IMDHOr5FB+0yxYs+oM|^U_)8Yuq3lMVtnI{~?Ek`UpUQ*f zyYiiS2Bttr+I0HwYc`qar9to54kezt47N`jMvJ9yO;@XBkmhy;hObq{6+c?gAm$N7 zEBh0J3#rUaxsw!DdBGL0FH%lu81>v{MSCw^0-@ta(f7UXLdTI~AV{tTDo-O7fyMi1 zx3dwrNY;e!YbE}w>^$pYt4?p0H?jm%C-A%IC0w1SK`&Jp(!RHnc-S~!D8Jkp@_x19 zkhPy^>XVt6b+`n@GjZ(Rka;AVZm08KOyl2|4aS7(YvglH9sXuN(PrO$X3|rdFce#Z?wE| zLdrGh(`74?pjV!${MiRn`97C;ZfD;UoC^DrA!UQe+NSZ;b=E=ZGpURG#k@S+fgjlx zpKVp&2aIzEs7^$f61o2 zd@}02A@o+wL4z;5Sip|~Y`Sq7#!Nad9RJZB8#~7G2FXX=u&)Y}|D>}LJ3>{SYS1Rn z1GFmdDII!dA2sQ0%9oT@vI&po@Kt`7z{t8k?b@>oJ=nbq1lkiSOxr;-r(2<4{Bn{w z=_n2kY$mPqY#~&28_V6bov!*6NI!lqr3dyW!+n=-c;}ZNX4UFTGyW)Sm9pqewU_+w z!oG69eg^W3cYS!<_;Wa@*oH64;%rZ{I-ga08{Yj^fXc{&I3miKdu9wL)dS~>P5TSb zYL5}ziy0!Uuy)2CW^1rV?N3%dHUp-;e-E}9>G7nh3Wa||JIRD(1?(=ge->;WooK6DqIH)fv8XsU$@e??8>JACN;ZH}OJtX-&hVcVU zz4(!TUTmR;HQ(^xNxnByg?83HW5l)OotJE-Mw&vdv&l>wf5eTMyie63K=;3X*W37n_;)2BSS?{L%t-SZy`|c9)jn zru23knbSd;v-UIh$qbd14oeXiRd>b0$Za4WmPkp0n*595A?bcn0+W2!YZCw~d|#KS(clMdKM>nXlW_a>X0qD!8M95w#+h%Qup8bB;8MyI9534^OHp>9qjaZWvHCf9 zdUOt^UJK=i){KIbnPah={so*JY03{iu;Z>HZnGP^3@~bSJm0@Bjqly8BVN7lM^;UZ zVr{9{$+iSz+1>vt@#f?QctXm+&3@j*Upg4kULp6P)ysqDTrr0bqXA@5r5-J9`z*ew zh2;7-TmWVgj!I8o3Ye%)yV)w5=Fb(6&H+8K#|hm6J+K?Y|+ zYoWWbuh6%~n14$h4>uo%g3ay8c=6s#G3{?EJj}esaxcFG?~2pRCA1g5T5z1^Pxhum z(o%8Le2I&Hbs0WP+JqOzuV4*7+@OBQAM{ID2J7Nx!>i(_@W}ps0`XL8J+!&y;3&m%{PNR`8wAF2wMm!oOXF6wQXhjIzYT=U=QF6IOEeiq{IGKtl6R(Z+8rn)^59m4&e$sFL@0fReWWO zI{(G5bKJx;T~o1O&0WP#hf>=2SrA=5BN*NsPG_T5PeILJ4eWe#EUX!2!5_No^IscO z`Tmm?SokWPYu{0$k1N-~(54H(T<2ovtVlAr!WRYxe8JJJ19;Q(F06N-aPlheq^Q$U zk6(rlg#@fc;baR|O?-_VIz7Xw( zIk7-ulFi|-XMShPKds=sj7Mh!|n=i8F`-hPztmvdpl>ThL1NuL7&}zkGuv$F~ zbz4$s?42aq`HUgm_6y_-KFj4Ed`hIRAHr*oVv+P~8+F$Z0ksPAsBa*wOd8i(m2zegnbj5sE`qVKU zy|EYhzFeIwnrVdgJ5!m}^$1pcco>WyH5bN79O{lE)9GhZi7ns#Hq2dUPggh@Q1AIF zPhR%`ccr6qEIpx-ZJOe1w&9hrI^h zt2G_xylTW26C=30Izf@Uz*jte_pdngXAC^~(g(6V&CzC(9_`lh0g4y@<7NCWhfYU% z%d9lIEWr)Dn$_q@t!RF>_iFy2Pr5MftkiQ^LivYd$;?qwifX><$YS?AAg|_JA@5!= zx}-x4_x`xdr_o| z26gP>LT{!gB4KEtV9Iq#enYH)%xM}n7y>=)DuCo+-&5qzbhlI1* zP3y2Y%9@^RzrbDNF7Z>B4zlIC%khqXv6RFAjjEWye@_bGJ^Gjtmq{wh4f{57_x`Ks z!V{|@spbM_=)t?*?7}Y$JxLyB+d$@|BlOJ|O47#1@W5`9`Ptxne)M0y;P-tDIc-)# zMi>}z`%QcJ{RxZsv*L?nA9z8Rka7n7)A;$Fp}1{GS4@?%V~I;wLmz|QFgkEJj~=ND zwq0tlZp05ZIWm;(JKDfrZJvxC{=wXE_!s=s+Ahs>>M$uqfiAO;Q+@j=I;uL0=b!0A z2ldsb=cl)`%IePKNbF?dzF!cxuh>b4-B<{Vnlnk_jD5T&qK!W&n}O5S{JG_`*YLKM z^EBy9-t+7aZg6iX$zHFb92$Izr}j5i-koYLl$aXwZ9~3bw7KMVE*ZyLYP<3XySa=_ zUjRGxmht_|)Oh)7A9^|M8yx&$r5q@4;_Gz;<5K0)hjmU!j%W~_VJP9{!UNqWpWD9maZL8jsm_H9QdDU3H%>P$6L4rvSH zSGr|ke&IBp<`jgEc{kxjU`mnm>fh}88B6}ScLJupDaFN$s>$N%Z`i#H1F>!XA|7<` zFkeu8KR&a1!Vcqt*@eJ6pf=O|LsSj#SC zo8VCUWU*@gJ*G7n_5A`L}_Dv+Co~shg2mi@5C$Gg^sKY)>zi=n34$27YFTB%_HrdAXlZyHc%lWe0 z%d%%CbK(5=cxdUm9>|d@;_+pxc>kUyd`MY>;UA{4jM<_5OHw3Xm2nW;e}pJ6tX`$m zzA|1Jbfy=-u!3=8vtisi!Ik%L$!1sE=ShOx-rP1TPw2n1U7Wvh9u(RggVfbGpv2Gs z!rh|@TNht6$2c2REH>if^8aw&Tt}r|?i{7+=ef#{9a6CNaxZ?`JqK=Ddcs+WPd-$r z#nH)baQU_`dLD5_yI*0vT*$y-n;&6X*N(6@eKTLA;>#ZNu^ z%P*F_<^CpRpc~^rPoKI;dajtDtje`g`Wss)zf7|wA3ximYV0$T8hD&+Uo~C~$jQT@ z4gF9nQV>)cx2w3WpL|H%>c%OHr?hkj$<$X$NjaCGE7-B7t z*sz|ya#N7izxNA^W@L$Dj(%qKlOkAAfsFav*ONj5@aOh+bn^G06MvniUENdU6Cd`c z4Ibg#Z*L|)HAe-i*B2q-b@bz_zxeX1#AX)Dss6mDaL{EJ-E%61#K{sc`$Y(*+T7>G zQbw_DP6Yqt^@O*_=~4D@EXq2>_lzI+hBtV^})hRlUsOwUM6~mImw5A@Re&x=dm4Q z#_&5g{^Af1PUEZudE}*9R2=QgwjC>=g3)98(d8AZkJeBcni(rSjsXerjN!*T7jyr+ zS=`=i3?5CCOB|4HuyVyqIB;DFYn*zVte7d)0BsNLD!;8s-c-W{L!UPQh8o#u#*Kv0! zbsmk2BJAOE{u0R48OkGobR^5QO;JaV?NQ&KxZVQDl!(2RKBZ!Ifwio>rP zhRM9QAHI;j^_n>R@@CRC{tTJ3aRE6~QN&LB4W+Vieze(SEuT9<;2*|?@rHl{bj;lb zIN3T-9yjij#7J3B-_E*0FQtBmWn0_8DtM{H;gGzzofkrXTS{AxCUJue7CiXQMv2L= zfwg@NCYy|QV4?F93|c>zea^{6-7x7)E$|-sd_Ea2ZoG>5wh>4y>PYQ6SAIQ+v4~5~ z(0l(ps@JM7H?-18$T>1ToiZH7a2 z^H`>G*OAZPIv=L*?@oJVbf#|Ki@~k16@u??qMq&jY4FtxbladbI&j@W=)SxUYYei0 z;3j?SXL?@HiC>1RFXsv#FB73~v%j!FrymZ8Zik)OhE!*82lz#Nia$B+P)5`hTY+oWN7d- zoKW=+eLXJ|y>F3XRNgW$(}`g!4pZP@=uBpB5l>~^_Rv7N4LjkK&mC4DVzO^zdEVAu z^mrdDdgTC-_EFQQ``L|@OpB-J{;Dzuoj+u-Zk1S#VK{xEE-u@xjly@y8T9cS z(Kx=4*m>y7Uv`=#-*asajY=Cqe9nDUgh>p@H%6g6q5lKE$;*gZT+d?WVv}g!Z6553 zai#2JF%<5t6I(ky_2_*|n#3oVWiHeyo3qzyAiItg{jMmhJ`?4_d|UEm3Tz zfe+bQn2Xgne6XX398OC8j>S8j;d0nx@NHSo*`;be@xfM9I6fdD1z|YmzfdF-Uy#Dx z6Hv$CHa28-f@;-zG1p(3$gj4gfiGpmq+mZuoID%ziM)E$$CE3(E>&} zt8>Sqcev=#dTQcr#izEM0xzKl9iAVLc4tzUOPeOM3xCh1c8);xRVQ%dHKa{122($K zO}V+2l*w?bV=cRpD5s_#a&`*6^HNr zw(zBQH2JR!`EcR44}IC}2jRw&cOvZ)Br59gO&p-ja|>?d{TyvKB6WPgsJNQ)JaQ!Ag)yR2HwpW%{mWZJ1#6@U5}NpY#7H1#+BgH zV|x+n8u8cr48bp9Cn||Lw^*tU`;E+LyV?vap0SZ%+1(Mwbe{|!Pm6eU+ywq+MQ@(` z`y-aZb7s{&iwF4(!u=Hl2Xzj@8Es3k*D4P#n>3f78r+RJZuuls8@<5Uo+Fi?XJ{&S zZFQFY3_1JogX!;;!`4Hxy5}ie!5pa8rSIK)6g`yaVeF3JDbJoo=Gf@ zpE+Dz`yBr+`MwG!MGHEHB_w;|bu{=bkQ zpvQE&l(6Vj*9k9``H(ZaI$KKIzb;Z zrtkEF{Q0&N+8*YTgw0{JAYcg{ey%q?IdLgoDzXw%OSE|YxgY%Dyj}eIafxRz;X3BL zm*c66Q%LE>QDoo-#2AleyrcJsdHJq@4F&34Lu<6u0reF_4`|?nMGi{ZW3#eFGeN2K zbwAWwCqURnyKUi>Egs^Sp~%74YMPVfMhZUCjR|;{5uM^7-(0VU|{yT2U$4;u}T6;q28}s$-LRuB-?!F0Jn%(?0PGqk-FN2mBIbyobXSTOF zho$r!gh^=1G=szWUY+e^`NbD7pmQ}J9Qp#s`Q)>Sl5_9Qk~sLRSi!pY?G6du6NxhQ zjoA6wQ8C~}Z7kZ#}YQ(OVnH!Axy_18lK2-dPqtCmqs0 z?3eO}C&4?=js2B6p!V;g$>rD`Fni%)5>&m7^^No*X0k3|pKzRX_>jvkzu1PWH)!&x zUinm>=tr}HU&D#|Hni)u1SVE`(7oFbVgmMIfy;ej(%=NE(mUY~_ce6;^o4BhjSsLv zxe!LY4io~c2Qt@P)(WyGMp9y=pxSmd?(uaucpprJ_P`F<>*^;w;oJ>6CPy>v#y>K{ zj-y5LD?-6=83|}{p|&1p=)m)nNR!kFrDN`i%3hV?(Um^rVjP7&SCi4Z^Lgg%SAp5p zQlGfB43vosVkbKay4KIgVoMdM>DyVDCMaZKrY(Sythj?q+{YrS#Nq*QrC^j9C49Gjg%;|ocul^q`m?oodaODNr`Z6T>#&7{R_AYZ`eb#K>BiJB3c+7!5fE;!?_dZO?2HR}_*mzg3s?NTLwS1Yl}prg>uVU6_c zRl~2fZ-nAi1i~CoGa-MrFg`d>oalT2Qusd7ZD_7&b1;qk;0k7u;e#t)mVo(_jyO0~ z>gn$0WYy9Ny6Rd1^|WhcbHY^V8~y3@fM+Dpt_vXE%WcVkiQSZ~dbguQ=6CLRnDdrmB6esdhTRk#fwt!%+P>oSGvgT3iLnQYY}l^ zw;0)d0Sjs~r0nN#%snJ!KzEz-1%_c#|6vbZ@#{N8Rrkh8nT)lJxW>LE{E-}?<>>k~ zP8hU!H2tqjrC|T$1<`qxOqMMxCl-YbxY4W__1HUH=6r{}TO24>XeBwtN2U+o`|lAt zeAPs?_nLxJ-!;OdEvjs1zY*f*_B<&3Z;IkC8zM8zJ3jE}bNy_FlfQy&iYc`CNUr4-w^+Z)h~BGAco?Yg zM|t+#PvQrJR)onWP8;EPV9HW9BifVxlu17N*?DY!-(vBi`wjB#L6UbWw3bx;;fKF&cBBhN&LH{n zhwvlWlbV^vF~!yuZ0!MaZaw1!w%tA^y7pK>y}Zq_X2o0fiKf9Zn-;-4wbJqQV_RM% z@59ri4BmV1dgfmEjm{fxvzhWNwA+ydiDgje$(L9&vd!{DnF*) z-0TqE(%Z=|ucP?*zyjHD-KoUU;28_Q^bN1Svgdlex8jmLR(#sg6f91P60cl71ReAz zQPcSxuGyWy-7kCv*X@6B;g=8`NElvi8q3u^r?3#69ys7znWEBcFlerv#HKxZAvpDN z<8k8yp|f>I{(4b?Fy!ZQ*yZI21vXciXg-+TF3uL>wwmGjU)O};pUq@$-@C)VuSG&d zPC0yR^Apkzxx!_iBVvq6AIt}Ed-?RCFSvgQVXaxP)Ttl4$da`Au$Tl+| z3wSZws>BHW_atN8=Z>_~Nj3UlsH)V>=*1>!uOqz&yMuG4l8xBBUl{&m8mn(!&Q}^C znjYPOx7)X2>Pj2_K=TNRH2Q?IhgtKaPij2tvICi-)&X{@hry{8i|LH}p5T|RO_P4k zr3%Z{FmkgmEU6t0KF2zV#wTvE9g9qw)QcbN;fxm6OG)PSJ>W9W4)b(P$v>TGLeYV5;JB>> z48NPw;ZNjr?Z;?n+(QNs-eZqj3#N=!&oCL_-64EOQsmW zvA8f6ro*Kfs7%m_vBcOtk+{$;hME-(rUy=sz>oW0GeuDzcB||x4AiY4C+0aq^t2B2 zLSPbmUHb^m<{Oja4G+kIzXnvWi$q=Hap0`eNGnVaI`v5>579x) zpRY;QUAV^7KWdTNdG#FLB2Y;HX(#tjVm_PfMV$guY;ENIy zmj*z?;n%Fycdt-<$r2tUcA=yyO6vM>viIF7vbH>e98fPuI%)!0a>$LDe((eJaoefo zhg9;pPZMqN%%gsDZ^Or(!=SwC1hC%m5SlcUUi{QXKbJRC-x%CG zGoZ#1*CEzk26H;+z-x)gzFzAR{Wo|g3EzJO4W^89>f{>E=1F<8CJQC5H=0eBPdp>m zRK*DIleW_tf7)rWTP59k&Ja2!n!}ShPeDWI&zJ6h$+XSX`5DQbbGSAO4vwyXSeuhD zV9__H2Se|R@0upV+vVqQ=KGOwxRrs<@=$1;REB#j#?h?zJ>iRXCPXx>Bi#pU2+e=T zC_Z^Bh;N-p>I?ixUHL-1blwr>8aRUNUKzUGo-8;NzeUYW6Y>4#Jp6oo2mQkdbKifK z&Q)1Q!(0sLe6<3&JmDkkk*>K2 zq2+@H_DHD*v)eKu$TStc7aW8I+l{fKW44sT8G~al&H+Q$Q^c{_3_82cqGa45`fl@P zeAZ$?Z#JKXL(>P)lK&(p;6op}YU(KJp{0#QX?y8*hnqwUP65}fWKv`7&1O8HIO5!J z=CGn1zx-;1&kG%C(&u2vIUYhB-(O-KUc|B=g#%fc|eq&F>==fwdcCTG#f%Ge<2tXuCc&*UM!2J$FFcdx_s`K7t?7 ziRZWN<9Jc@Q8F>U8mcOsXiwi!;_^otg5J~9(A;4jT<@t)_YSv(QRlvpcV}k8n;(2QlapkA=^txFWaJX7W1~jgxx_+f} zN!1f7U;dhe`Z}Z6TZYXc<|1Am#IBhC5cMBfoz^woGE=oOlV(ezwAQ$#p&SSqFM-_jy>U9{{q9TX1Yh9!%@@OpN$Q@%@T$wr%%C zZ2b@92_>rXZBF0mWL+7vWRu8ygZWVS{pm@%WYIa{hf^0Q`)wt84D&_#(a&J%|gs|D}l&ppf7Ou2t2wrp7 z3x%fp$SlK1TppB$+Xhwgjs?>Db*fBU@>(6Mdw+52X}elYcOM6+Ncku}N?d3LB z{Y^t~seUi|G*^h(bersmot98}{W?D06M?abTX@jz=S41yjJUXvwV%%SUg>e9sKyYOw8 z6%MON0DotFZn;exH~olU-x^BA59M*BNu5J0t7RUW3yE{g8D9z#k~QQLn!%>TNW z^_S)>6~0AqtNSTpvSA`u|DeIYr7h;KH)%n(=~W`ch~Q>$mi1`5!w!77k9~4B(eLv& zLeDS*unrn7mOk%BY{YXcJ)|>?9RgH_*Z zma+Q>o^H4!sul~#l+mo8#F-257({B1rFsA3&x;D#VmL*E0T0yC~>mHj%aj!u<+D&3R(otqF1`lq4T=U zqMHo#ME@%#LYv`3@xzj0^4e!ITh%+8WWO0GWTfpz4e=1D`u!HTlNt6j|4goYSHQ*d zZDe>Ev>6B!;@2z$4F2Gz}rX>~^(njjp-G~%0R>2;sHm#PcPMu`yS9syT^UcsB|Ql{eSX`Y<>5bNS*@X*ntusL`gGr5?@ zR-HQRbWu7!7@s#qFq87m&8lfMAtoQ-$`&XOjDpq_JGzr}qXphwX#RYCKKzF^|7bFa zhu^zLqQ{P;k3(YVsF)P;dYUW$J#!EDDJuh)QwB6+Rw4BA3**BLxph|*k9`4hiJAHh4Nk05115Pf%M4PCoc zg{t+v48tOxLE)DeSmD)|WH%fqey>$%%txd{l+5e@wXh;aI5Jd6d~rQsMre#`1}c z(e$VD2wHrxj^>4%a!rpg9(Q*HcWq2$m1BNGaZNl857Jl0c^~D&J;w9tdINcC#B=!V z`G^!)Cb9b(=g5Sd2&N~~6%O>N5cKbM7k*r}WTka|SXHeLtTAbXD+TS~W4HtQdUs+4 z4_jD7=jYg=+ckh)QQ&p+J5&e72y0)gBTctDz$Lj3TyK5|+kSQ;zL_f2sK-t6WZzK~ z$KFBV>^STcpD8Y}7=vf08Spi;;!qh|i`zz8;C}&SZ0zN}{CL-Par9neOw_jlOOLmZ zsj&k@M?YGp%%NpRBWY`a5;xQuKv%sYF%OLS?xI_4L-q?|Y^VXIpE5|qwmRAL%rh|f zt2Km{6v4f;8d&mj4xSVSVSSVd?eFmegAD4h-;*!I6; zmGE%oV7%0;1CPJ*AF72J!(VxMha2}JLP3B?)WtLkpBJ-82! z{PhJd|2+rGKIDUKUrY8}(I1$Ur_DA7ez z!@~{7*;KX)0&-#@y}pfTZr(1|X5=zbxD_T_OODZRPR#8562L#}pd%t|zkl{b``Wje3je^w;@BR{Amwe7truM_U{5 zLvjld9_fjjpHAoI#{LrLPY)ZiOHt)wFz&dKK<&mfLp%1MR%yP}ebFB%t2hDQ@i2TD zE~j%gXppbML~8!H4=9Jlpx&b67&|?W&C!2>eMVa_d+qUXqaqY5`<`H-J5}J~{5ZDD zB#teK-H3;ky?K3YA@ebQhZO+>P^l%kxQet`?rkI^jvZqvCjDqk$T}#n+(E0YN`za^ zy@XDydvFiUhcM956~hYN!Q(6A$j?h^^xv}itm92B8hc_kYAmV|lGYBuH`4j%&vOZw z@zsjl6s*K0Veg$Td1O-APZwHOuT6E&}r5^2f%f_mAQ}FW3beuP2IE+eaWpU5i(0~+^ zKcjcz0E@k#BVDu24JV*=<^XBCn<0*H_lAXAB4J_QMtpYWJ^rV3!Z*DY&Eev-aH#hjfu9YWK*ew;Sh_^Rh4??PVEI#+Wg|VW zaXsP0@|IX%5Jj6>X z&hh25y{P?P9)6oOcN9XX*g9~7>3P5=AnV4K?qS2XRo!;hbBWIQqimdm1dC?ijs;D70J*XN#gAF zu0tAVKpHejNCTotNk#ST-{;Rc*WTAYd%f@T+&6w3^9?;1WC|ln*=dIs|kRwS;% z*K)|!q$5^^h_|d+CR!LCYInJ~n5MK{;{9*C^Jk|VB>%OG$guQj zy_5z0wvRA>YAq-))}s67>EVr|YtWvaS$LQ2K@@G#hi%?Aaf4}g1aGh;+9ux_?*!o8ZMW_w|14& z>b}otM))JF_GJ)EJ>ZJAEN-D+24^se0Qm)*VQj^TAoKaBfN`x8-v*1f0S+gyD&R{ zXC89PU(-Cd3R;%Cnd@#lhaVjknBsl|xHW$i`g=PM$*oj`IISC^cf1AS-(JA+UNzKk z%n0zUXoq8?FVS`q!yW(cCbuoQ*}ouBdFlF?+ye_e29*F~1S zF`Hd#O~RLTzH@{B%K)bt&DhswHhXnH2@YjivT5$tY()74HtzWWnr4^8pL|jQ#&*Nm z((;o`TS!%5T|-030eI%Whs~*xW5?#}z>xnIvzq~`tmyAGRwns?kA<2* zb#^aUpE`@<6dsk8?x{k5Dh87&26OP;!VxfKRU)Ju)&j@y)jZj$fDa9w%YUhAwp(#K zo5mJB1wK%QDaNfv_m|3}q$YbjDgP=2?j8oOlWY0w#cRpK74i61*$bSaUkDvH)}!da zDR8f8F?wiqz0~efy7=!};p}%U3l}v1Mve-x^qaE^-}t(L&RsQAn9W^-MDsJOiu{M0 zD*Q=9*#`2}$roQ7@sO4*+=A4PAh=+s$P%+W;a&I#Z2PPSKl?tC)Q1!zew{q`S(u+r z>!^Uw0;@z;vjMqyO5yrUJNRB)3pJk}qn#FC1;?Hte!MxK``Yf$9ec5!D*v4Zul5e6 zjr${LU6Y}}2TkT9g))8AjT_L|avc(+|G=|Pw@?=)SoO;z{^P}9xYJV#nYn~U(iD6 z_pBJQb59^?rr&YI$N9MG{7)1()Cwl1z=w zM3tjW@c3oYY(CzDe`P+wvA4~5WT!$L7RHm_owJ2L!)al^Yk(uhd6ArXwe-3B3EZQZ z#SIG`j-0M+;%_~7<)54Tp`Ygu;plZ^arf+UyxTI1h#G|~kcPlk4YtCm$G2jiLpgk@ zg*1t}byDDKj%4}e>zUJfP4@5FVOXZ~jy8vz;mT_Y>|BE-Oeb|v;v(QC;&pJvLV0qh z?}SKaQZF|pHwV3aQI3~6?jv#=1{2%a6LHdU70}smgpbWXLu~qY<61d2^76%I7#(&O z%xAoTBvUth-P@a=k7e-`aALJo3w9PoF|+4y(a5%Qu*9ksNldj#i<2MLxjKk!rTa+i zac^QQ`Nw%E%2LxO1=v#OJMDRXmKelpkljjUXzmwv*xG%ER{Ev$$8ammG^m1g3RdJ* zbQsNorM7mj4sw-$)cc7*VI78a!Bv zVPY1?*QmsyT}58J{{&w;a`q(5`*%^?76ZD+hDkJyKe8KN4N#n<48BK>W`D)^A?tb# zs_cG>HU9mC9)%K^-ZGbjd~w6}8&2cpU&i3mX~Ad~{)*q;SVso-HUf+5!aIkpBU>&! zg?-#pV8J%*y>M-=Y%hlP+eR$2<2gsor0~{T2Sn#A`|!DM9Jv~_j+~iPkIIcrh(qsq zVl^*`{?mSmR82ymX@N3)2ML%5cfiSTO=kS52*2LBn^-G-!-35n>dAbS-(HP+OCgkxE3y$Eu zQ*+smyC!(Gs*L^7Iz!&eyBsZvbm8}JP$JLAXVb~Ex5Jh3Z^eTg!ssf~d=6hTWYWWw z__Eje$YjkE{@S*eI6@~EyGLu5EfiQ>3sUWg_0(7#yGfUHq^`woy}tO;qeEz)k`*>q zl5oFOW}utHiqWfT3;gxub8hxMI|dWOSmFKItn2SRFx)AyzPFBn>i6Ba&wL?nUosEe zp2?H_F~gWu=1P{Kv>1{%n&5HksZ)U0#6qwraR-Ebzu&4h5aj+RXOBdV!no%=Vcmk=w7L z1Six!>?FBPYo7X`ey1u9jo*oL=G=q82~*gQc;ViT+(`yEnZlc^vte_v9`|9U9Be6^ z%TjM%VCLv89j_qxtp`tIyI-2Jt~F`6dZav&-ToUdXt&^Z1ooEg;MUNdrZIfT#QVrk zx}2&smLnUB92hdQgfwiw1aH1pf_`RoX&8T#TJ_a|_UY@qa)mZ8H#ZvM%;ovOAw#ie zR|i_2wV4|4ixFmup7_CWefx%JIr~2?uSlc&LsH}zvXg_24#sGjuxWIck29xTRH@X%p=GA&4}uYN9gd#jpXO&T#^&N6sbg;k|7E; z`0q+3EI&O8ClAer8$Ahh@{M%N*rJke)*yxl9!D8f8do*(?_AF`8|<=7<-s zm11V{dZcR4Avoak1ScH%iFco!OiuZ3C6l>#Vu|2G9&u2EtR62$5^G1U$itl3$e%Z#lBy$bRH(l-!$O&>E2ZL`h663zm9u)bsu(i zutxt{)Y+p;xp+XO8c(T;#(Ya5ws8N3er!99&sQG99i1nL7M?=96u&`X-Y@uNJ|35@ zwk3L*@?=}b9`c<#EV6su#r5(1{8asJeCL7}zAd_k!#hj)>Z|u@Xo+4M564F7>YysJBF^$MY?jSV7WK)WMBT~4 z(lZwjhwb~w)iVj?vAUSlh&SL~3l)61HXpJF2JnB5BT1o;DH-#yi<&AFVEcV3Bq30p z+#-sk<+!Zm|2&;3EA4?zD;K+8KBm zi;vxs!;!w*2shy&8ml;%RgO!d$;?sQSbwSP6&F)#Q1p-=bkm1RNNlyu`(!r_PXwPxYOL=hxD? zeKLRfCcKUQ%os!!%v~AxYA&0jrpI2ru15`uF+wiC23vj$234)iOg`3;se3b+9GE~A zmRTbA>RI^U=?6Hl-GV-FImJrqN3&y&M%ezQHZs+e@Q=24()nBkT`zD$=ezg8u-*RH z_-{1cd-V}Iuyg`-trYWbB(>bKWAEVM%zTJSHN(>GYf#GQLff7(K5*9E%zm7ey?xzs z2mAgP-*Nb`S~}NHm6I3T$tn6zS*Tp0;GY@?DHi%LvP6Wc>Ijw@wHhBX2?t$|PA*pZ z9UbH92MN}BT%DDo-Oy2a{N09ZxcULJqlRzbQ1eL)n+>^t`wggk%RNXsce>vs7-!img@RuihElOcyew!@*6!Hmz>K3)Ax(W zfp_(!y8b2lXU|CRu+?ODOgnis=L*?g^_INN3?}8VDlBV{kXIU_!^V%9zzvBF6W1*X z!NZ39#kFq7`BoVfn}){AUo|URIYxEHPH`zGJZ3$E?{_> zRvi28pUNgzEMsAX4*0x-8=Z9M4K4fiM7(cWDZLpKO>bRQuumW7VgI^B%l@%z1vg@< z4?I1o4(!8P(7pT(yYl0ZVr(H^C-mK|HTGbin5k$?%y>TF&rvRv6UzAze=6mFrOeC`lQ)WA#=Ca}6{$ryc zhdLTb(ZishK{NnlK!0TZ|M32LqQ!B_tGY$53dl6=Q19{XYZ@+M580$K($wp`0((wwnykD8L%uDzG4(1K*n_ zY-#5+(sO4DsZLu%{JcN%qE>f!6cG)Xck9TIo;qS}pG6MYO(ya??jfmBYsk8SiG0pJ zMR4bI09kINJB7dE?F;_!BX=x78?<7O*Utvn0^PK?)DTKX9ziW{%izhkOz!sUcQ|t7 zebhQ&0A;50@C%Db7I(OR~?KpKB)fA1RUnibME|rBe{8JhoTVX~cD@Q?z(>!+K zgC1NG94WVZ9Z1u*ndHBkTev^mbkfck*p#1#8~#(l`}dV2 z^L{ZNh`9?w>=su_eZqx9oS4jE4}R)HUFztV$T{tKhTE?{K(?`Nd|t2|G9OWnth{I9 z;MOwnp5-URvLkZoiH-WWHq#1IT|GL*U=e8zYQ>gb6NKzYyT~O}o}{Iv;J?~2u)y&N zdh7gz<|aNwKFe3L=y8O_vuU_B>K4=q(Aq(<~NAJ>h;iS@|SdUvGDqGgf zy;rV4bDsP1$!!a`#L>OtzVUTn{L2rU2Oh%vw-ti?SuJAw$rF32+(%~DOtGOPo=&>@ ziw3+_qqrr2f0QwTre4$IFWuFob4u59H!h7NX6Mz&ebcje3pbjoTfPBnkHI*gy_^`l zzC_+#zfV-Y&cK^@&VU}%t=!$xgJ|hyLwYndgnN|e56O>%;rQe*c4h5-(W^xRT=7OD z@d4#4+^Z>XX}gCu)sr5e`6t~WYVvG&{bfHc%G*v2_0PeXHf?r$WDx#0&zns7vJnmn zeFN^vK30@t2)j1#qk&JXxv%~yWVzE&a;|(DG3y*jRGjbQr)DBxK5C#Tw;Zk6e;2`S zRjQKm92YPdynk^Q(wpRgwx_y?Ki@K9q9uFq+mdME8IpjX6uyJnbw5DktTVNd>_YL) zW{{fMhCYcWfx+a5;566{$A+1RCgEn0V}1aO5B8zE-|fLy%D&ObK0`^YuwV9@G>#QX zbeOy55f*A53$IR4Y%k12XPtQqIPyGSYbIjL)_RduYZsA+3skY}*PFCzi7XxLwua90 z%*J8+U%{4!8Za$)B~Lxe2)6#mri3fQ13w4R(VUT-`rqM_FVBr7-v35REJJPB>sy2Q z?|NsUeS0)G%fz7M)Ab$PLuUh6)}k8B^HTaLR*Ud14ZO|y~YRd)>2_X=alt8LiWS2@&Yu;4lP zHo!ET6(mU$3fSF0pV-T*TiM6FZ}`U3Tg>ULHK@&uqW^?!hvA&5oZ8xaK0uPg`Nu}` z0}s?#Yh(xusn4L7weE6OMfd2(oHpifzlVL}d*M=QB)F&Ruujuhw7}g~vi?UQn>Pqc z?g&m%m4F=DQM{jq)fB>!FG?)CvkB&}{?4wCjbt$V6ub95jU>kJXZdaEkap}FpEAsV z$rDqQW;h8iTW|v1bkFC*H)LUp^DZoFhrrp5s%OJ$6Ik4)MJ#m3HkRCEBI)ivA}J3r zm5hlP#wOOuN+Ny7N=ojjOFDi|ljshMWRbTHvE6NhATBu!61N=&w^di*=kM+8)5#2$ zQM8`jQ&)%Q!BY%!ap=_A>hI$iSk>t4yQzq`26_5;+wYc~s8`<`8&e23*d(vZl^ zFoA0UlVNbkdGKg#V{1)A*rL@UwsFQ_$@e{d?9i87_DEWd)N)J6Pk%QeS)?fW^lO2H z2AD~7YO~mtz;P_fa0a)1qP9daBw2FG$w#uq^9CuVr(o|@cV-$UoUe1Yl-0Yf;ivqU zNS+@OxVW2(@PJtuDlu0izuL0+lV2*RbA1r4o_CTT`luXd?wCnl2s*pQN_S%g&u_J`foyi=S zwtbMav3>e1InopCCw%q?Keyn>+X??*UgJ?n%2)|bLZ82^ZXs>Zs^afSpA))b^GKP+ zer&ZVN&NZJMJTu+aPzu7?8i!**iRp*M1B1(xk}S zlP1_i62TrFS^-|G(%6ErFQMYYV7A|=0#|Ue@SWf9k#uziCjDt-|DQW}m@tE=4;jG@ zUcbq^Mvo=^z5Zm``E{hT$A%aj+k?CNO0cH$a5i-2I_A@+ijKS8uU{j}y+P>93K`+P8Rw(Oooi?_RjnJ^{)b z-qVR&wo>>w6Nf$jO*3S(A;`p3pws<^k_0tWYknGCc99~}YGkm>sx$1XE|04oD3NoE z+wgjgH(Z9^A+-Ew7jM$32xWgV@rV$i58w6+I`5ohN!JN}F|8PGy;rjDuHH%dre&gO z5<9`e`w~}wDj*|I-z3RhDa6%nDT=e%0c$$L$&>s27UazQcphiOABwo5{UFD@eqdOjPjYHJA>n1_f0u zEaRAhnj)9u@PHa1qYmNk#fu@gXb=e;mJ4am! zIhCoB0{t(#;N-iO*+-eNA+d!ZL*qBx9TX|esKt*t|ZayLM4{C&zkKjJ&l|@XJIb| zDYhBMqi+lKY112yX6kFRw1j`qtTh!rJ1+wFq&881RzAHg|Bi+Q>tW+bHdHtC3HrL^ zIcWYDhWg)`OZ%YmnMM1Wj380*zM^xD>b3aN0YO#6R3X62$Yd`ye?8esql5{%zyi4lm}G zT^941(m&yg)mv8Z%97O=XHheK1)?Mqf$znCMmsy7qW3R+@xQBLk#b%Rx5zmbu5Que z{C;$nZQi9yAKM>g+kE1g^{Oqf#9#tA*6!r^2ok_0n$wH)mqmSMuqw zw*B?kEE>~Q0+F(H@Y7%v`(~NQZv9AP*Y^#GM!i{r!u4;K{W4a7{EJESQRgD=`Qbb& z@~`I7dY!<^Z4A>&z76k-Wm&y}oZXe+aDHT%G;&pKhZBGEXl76*9r*Q(KCSnoqkI|C zuIMD|xVL15iw*IQS+S=gIkqr}FweiQA-nlLz9!d&%VJE(_|;3uC%3bR-zZJ) zgc=EFfdqCxC6Z10*~C(^^O!~A9aik%4QecoQwfbha%uoD<^w&DqX8nn37n5ZHZ%5} z&E}4GBR`KGfyPJo&|IMpdHtC+U7WR_I{00LJ4I(u!}c>^vP+KK`@UYdUsTv``CsUm zO(#EekOI8gBg`WsCP{SZWQoGKG?tdE07oYM<&X3Uy!Wr4AiQrWZdB}pIZs~F&%>vI z@eSa+n(M*ufIDYh7R`q~t0hJ*y5vXKah81cFgx`9H24@7Q8|5q!MN%sv}H!_k2jZL;9K7OZ_32EQ}DGvBWB>`1Y)L|b7s>?7v)xqsie% zaeqTE{E3Rjqjy)qztUGMqnKmcte4?+KX${A?K=3}qI~qb)r}ZGUWYgS{)6YNz6JHF znIL~Oisd%!#Z439$<0t&_djEa6#S(S8udIXncnfkiC3z>x_?YPe)M zhrN+{#G)DuB_CF#urEiIVDX_=dRFZY-Q@fPPVW}jeJ_=m{y|lt%QS_?KGUZCJyDP} zQjI9Lph&eN6m7J_%<1NQdKOZr__A~5Rz!TL$L;8Cy_bC2&c(Xv36wqr7z z?h?q{d`0-io4FvrI*8T8OhaFVe5d;3x8Q319TegYvf1^!(3pWA^v-`W9O#;}UoE$o z#k(Xn>FNfYd0$Ih{dha*T++gibPnNc1;HK1jM%%Huh2&vU`=Kn4h{)oPX1Tm2}m(> ztzhs>IRmAalG%~SRHpsy2&>t+0|$nGMuXG`&;su)7N)wEc}L{28FCk3ea-?FT(E+1 z_8qLdJ(ERW{L5x$7J|a^FL?7|e;lP~g6nHz$xEC9{$@G!%n*S+b2SDQ{t@`|?ipBN zOCrooFe6>B_ppbHmP6O=I9w3sON{-;k?(gY9P;dkd2{610-ZDb>4oFVsv~2eP!fP; z?e8I_-dt+>%a}C^d=HPe4fycJ$MDWS0Se#fk^{-xMCA_Q;5cW3R+nA@ll$qHw2D1DzYbE#zBm82c-WAgV5eK?zQtudN;%vopBz?W*n>p zZJXu9tK|@1v!)rJJva$Z50FL4a!Fip#ag;%+dlqMe=7aEsUOW_9@rpOmo4%Z+x#FfNIRm#rr6 z|0vovMSMg(MLX~?R~^_WK8-)${=+p%tJ6&~GIZ(tNBqD4vPs!VZ5*-Aj2@jc4N1^( z@Jh{qfOY9MxA!X81<8ejYwl1KAaTKN*VT!~?UPVFJP!>EM99ruOB9YTi*MT0!#e3Q zUefXnJ&i8L$7k)Prx#0OP4hB3bqS>bJJP7~gFXlTX! zV0zMcc2MZid}~S~WBNDX(13EWiP)HSFH1mu8XdU65|OFNV~NkR23ovo4Ki7_h6Weg zz~3TWn)>HFT-z?eFXt#Rb~^i zv$Ac&;8Ttl2GuzHF7*vv@6m?cN~e$wyZZTE%nZkIO8lNJ9Pj2E0j|MI@VwRveEU;8 zHuCaC7p`|h%89|Cf9E4AyRio3en|0~_mslWv1eF`z{Z@~Gn7>rDxg%;O(B0oi|%!rHpqP~PZ(Gv!avCo9)LS!_No3%rgR<10nB7dRYu>?Uq0m%?VJ zbGRz2kKo=loN-Gj6lFS5cYZypb;!UOVMXZ0V`=cH@TY{INH2sv$2U&*Lg=apdh(^f zxKaItpM81HPgOZi`=^J}ZCZV_yZiya&smpN_l_lA`;LqQ;fi>ZmnWo+{EP|`1DU7o7252mCa`_9$@uU>Pj6438b??z0{2(4}qcps*kPCN~svV-Deip$6b= zRSYV#jmWUODrD7sA9S;ga)}Q6F+MhobRWyarbp)!rvVM}e2ND@zdjYt^_B^I9%s_% z*NGD(Uc~-WIPqCsM?7ux#l=tT(Rk~X5H)@-^VGS@h8#J}E(r+9H->YeqGmH*7p;#Y%e^}AW7;C$*?$9Yxz;%E@Amj3^kc+V6)40T)1um*~)1%8{@OsIK>!tc?$19r62i}5vBw}|Ka1@ zMQU}60P#K#k)1cnULCg;wKz^Md!wd~oZ|=K(!jI)q2v_Y|Ii*9(hhN3%``XzIa6wAQ{2ttohlKYpKsKV-NfL&suz z0X-B+w*P^j4fX80N*Rl39wE7J^&3XC#G;9to^fPSFCPE%$J$UuV z16WPT88tLok)et4==3BX>{*(D9j5qG&mB&XbS;Lxvit;pVqbs)jzhLbX0l7MrmX*m zG8t}QPJ*KAh2nPv?i(HtEfY0BVj?i79-qc#L+=Z-`e>@CF2fW?Y-7`IjK@gkF$(x1 z;dQl-z_bqz*fOs{^z%?G6)T2AOojrs@jQ(O`H$p&KUIL`m9FsPo;^%W7OLi}Qc3%c zyX5M7FT&|mL)Qln_988Z9ofc2b-UgmC=A2)UlrM$!(FhW@gP%e8b+4>{sl&#=U zcY<9$ngIO<^=OZLF~0UQmaU3DXV+`|h}F!xBHkD(%dHRE0fjQIY-&UzEGkX`zu*5@ zM`8<`9n;NTo?HQjgZ#)Af#rW%4+MXr7EU}6Ew+zPAwyPNq%o_s;jxn~c=&0vTj8T1 z_L4T8Hk#77H@j&3&tjUrD-@=jJPHxVM?>*XUw(r5L{^x%0tLSt$*Mn2W(_5WVc|Jv zCatuH=_K8SyJ35%Y@uhR{t3Um&DnCFb5I~a_EtKogH`_rkpO;U z6K>DuXIn*qdE6kba7sDt7;VLe^&|tl^}=2HB0gnBA58uh%ND6&_E~l_%ZSW@z@K~2 zVWC$Sspc)tpAsWVo;g!Iq1o@s`T=6muYD33BzR!62ZAb@bRc;ozR;ujBfC_uBrOIZv*70Y1v{+bOG#+#r z*w$28$&;W2Mzm9C-EKEXt#;8M%o^;M?7O~0}UU? zP()rbq^_wz8!lDS=FltbbwNDa5|F~SEVL%e)c#<(7x%FGk@w6bZVzkI=z|&b9Infh zC2AanKAk^M6`>2AAD;8OBMivN32GpbRbhQO7olBmJjn?aoIZ0NqP-sP=nq91T=XRp z{=5%`l)l@rW^6y88Zm6j+J-kMUIt#O7tVM1pz#$WaN*1bKHcp#m+LbC1@k(9H^MZ^ zN7x%JPM}wwqh%W4sgfB9%^Oa*Ta$=I+8roO83T9N zL0ET5gQYF8$VX5^NePd%sPL0-~n*thzRum>ATBr8_2+tor<;rb@_@#-$_t}Wb+9>PMScKewy-n zSI^QrPt(cMqO&A0D4z^=JOd%#Tlmk#6KJHR3eFqn%C2qpVi9l0&}G>}@RcX0=mSqb zxYgiCUKve-n=zvpnc={E-Ah@lvK5#Jo#R;v3(3TueJuWqCL3dZ6)kUhB>F4zz}hMo zIQ_cGY*g(@xG>Rz-OKoiZ%YgL{Zbc}kU9frYvy9{kN|$H_aZv$>psE5I~hB?U&VRY zz2wlYYEg{d;}K5}64SP$WbTr+Biivoj#P|DnJ)uioA*^|#dahX46bQezfsIr$6HZUXG{p{$9F6=W?$g^hcfmHHm}QZL5dPKn%V#fvmCGZ}~cT86H^8$=V^Bk|6@llWo2 zzu@o~irxu4Slb?JaF?A=63y3;^#wtBa5%kQH9%c3Bo?(|s%++<| z=ju%)uyKfS|mhijAUXfN{PQ!RhWKau9Hc*&cVF2Gt#3|VrK z5&S4Hg)dvZfsTEOQt~dq)dUNqm5~ld)<=u9Q`{7u_W;G;teWQy*by0TIJv{H< zee@1}#6=4Qj)8X}&K@nxhJLP~tD{fb1$tPaqM{SvdUOS@S*1=wZXM!^p5EpPH44xl zxq0wov=RH~(Fj{qma%<4(^+fk93*Vn1-9Bl=GE=M5adEqPdS)I96}$^V*dPiX_3sc z`&H;qsR7=1>7!VY-nHX%FFh_3qQpv4*6; zHyxH|wm}1{5TCKR!LK}Oj30P@7H9cn^F_%okcZ_}E@%1?YH_&{hL%mCN*hXOreMF5 zZr+P6Qzzn#M1N-7Zh`jQ@&KJByCMF$Hbt5x+^3Dl5Gzpy?=P=l`lpj%k}l0|O8o`j zU7p}4+QB_f$>qmeZl)f|%VD@k3s&lCvCk!1G$isI9ZxV*{9p*>Z3f`*pCe0KV+)S! zPgCW-jcoGV0jl?X80Cd=or>dh(H-Mr9OE(r?aWkw=r75jcrzN7l?)OOIh)D<{ZNTs zu1&x+NDtc_EyI^>cJue*g_*I{Np!^~o3mfK3Us9$p!Hlj-I;KYE-HIT>o(W`G>w*o z-_~W?7nAwL4)vu4Pb~O&=Op&BuaF)4?!k)3r0~%JPx;U%v9vO}2$1{*_F+ac3o%{* zW(_BRI)%f=QU}gEGg-9xbuv13<}%kUWI*(%iw*^F zkEu7@G*7}#X9n<+y$H33`?8J?H|Ok0|eX@)G^#>BDiEiEPB`K@zQ1!z5v@f9b>rb+q$E zHl61F7e!?0a$S-VJfOCM-g|8hW0Otr$|g?aS#cGDZ(V1t1(Vqd%@UNdHW79#F+p=) znc!Jll-T*E26z|v0;b6+u>a;gps5!q%?jFuzWUsSjDX9qrSTYydH9J&$j;%;?ivp+ zo8sW+NKMwCX3YGZHn2M$hxpN#ZlhI~q|wx`pF|oOa5=g z3V(XA+A6`ncdQj|Y083q%oJjxWKNuGyRo=)7`wE7KiF71(lxi`G8Cu&4g|+h|FOw7t5CeVFAiTTPiNgwBA$V3C~9j(X}Nh| zJ2aNwuGvgyJIlZfXBT{6P88SDqYmw+J6WQP9($N!PrqkKfrpYZdHm9Xdsvi!Qw|uT zhEFR|__7SPwlL4K+JoKH|PK%Qa@wE;czCB!q zT~_;Nm$)eg57^J9bEee7FhyC;y-$VZYkYv5t={a;0|S<{au#Ydn=0P6)*jyzST$L} z?fCP=X56!TIzD@DG}$7ZjsxesgaS{>ozwP$hdyd7=|DK-@6}*GKE|-P2`@nEYCg@d z=%ky~b=bl`Dy(LA86)wDildhBDr43?FYBbmc_iRjt7$Yz6^*cfqn;a(FPZW_? z0&Z_%S`cw;P2dZMdVt@H)3Bj`1vKs(3mK-PQD=)P4bBmfjiMN0CO%AD1P@k3nu2}R zwe2K&iJss~i({%Uo8W4qDp{B@8V?i`Qh$0q=&t%EHvZ6tI;8)h^aP>Tc)||{&CH~_ zU%XMz`^Tag>C0JeQKsm0stJ@-_<&7Q2DW;lgC~mOuvvo!?emTRQM@wFy&J$S3?Ja# zyFFOtGarEoIe~>mc#^)4XXwJus#xWYGCORwjctqi!tTY)!2co*@h7EJ(#PHk911bM z#D3GirN-dAD1d6J%m!(GrLgZ##_}JG>5np@lj$2zblnLt5Nnb+i9fk5^#)IDm&e9# z99d7)$-6P0bX}`8_WZOLb?XVPXROCqjvCuFWf>g)eGazPe@7?nbkU)|wfw(|80hfc zO_sXM!zw$rz_G4%WL)TCQohRxe+v^j7T?2(^1^ER5T-JVqS>%}P6(|3&zQ&@1R^yl zla%ZaVO3!xnQC_jWIb#~oYHw--rfn$499Her4wYb<6ok)FACS3-A*sF2AJaFLK3H# zlaP305;`^n{?oU>hZD2-cS_CpjlMQqE7t}2gssF|&ykD?nLxH}e1gaEdgNbMDpox| zkxc&8RJJ}!1+UdTNtP70k`tS>?MF6i!1?za34Ol?m*k9OQSqDDm&qr<(?g1`R+S=} zOQXo}V@sI0U50F{{XspC7s9Tov&cm2xispU5gWIE8{_`I0I{PJj<*~^;^&Vft-sEa zfYM)N;{s*-Wm`K*!%l1a();=3aLyh4c(D}T;%5m_ixr7TX$z4#^MS~&5R-kg>+$E5 z3;5PG6YQhhfqRX$NQ}NLks8P(Kc>wU9J&zacMU4f0=^AoT4jsLqV3?6TY`R!}$q_f7_~FUv#O zqOnnM_n8YGa5TV!oLXS`<1;kt9ftcmvoIOH9{e3@Ai7RRGG4=3l91vm@oGq8okMT2 z;gzqL>htNPA{j)Dt#e#-(XHlDz2 zO@+Q&SRqRhd_=|XMnUlNVa#!T2R(3jFIYH>;r?nP$hepdjn%fu_ShckdjC9{JnbBQ zeX&ll*vEU{Et*fZ)24f*0B1>TudWu+4aGe?Bjh6dQ|@(Obbh3 zBUk)n!Am~U%|^RLs4@)iOCJt-XEVTOlRC4zn#G2ncVzXaj<9%v^S!Y8By{fQVar?@ z!OvXpV zZq(91xaKztNheeRO-f}EL8=lJ_l2xwnJK-BWZ6Ao=XfQkk=-9YQDXSQQZo3`BsTon zF&Mtz3<7QpX82J%m?~Fb@#Yv)0RL2%h$(gI_dWuS~R@QG?S@C?`uf1f<0J-bfe zj}d3EQ{Xd{+NTJ{Juh%&r2!l*iWPE1A=E;WgU|Wo!q&nO%v5?j`( z?l#Xh+p9|6e!S13N6v)1O`7P@k@27=`wlwq9)>_KP4ua1jO66iPByF9lIeRc;Z1|H zsON7}`kHHF9hchJ(mrj8wpuUNdjFCHUs*&_25y0gcQXtQRi`)aIfElNlx2SkWxMOU z;KR);EOyK!=HC;;)EkS~&2NTGCO?m5eLTaoGYeSe5@}|>JQHlj=(B4}G?<6I1+Bc1 z&u^dB%%=QGXU);6OfFcFy%gIpr`&D$gt`VR(c6cT9V%hr_owv5zd)MdXTa9%+{Qmm zO%(6xJ8!4*a2rihA+%hWNp5vdBm0D|W~2RVLd4c|xpgqy8t+4{Upa;PAAI6+txKrP zi4Hh1@LbFfk`{GtJV+B{g6Yo0J<#v+6}ujAC1YQ!5vd2VtZsoj`XD!$O`d#}PD&5r z|E5{;E5GSu7yk%4P3NeAA`8cyU^-lOpQh-;pr`v?1I~LRykIk{rNHl zo-Go34nuyTd-Hk(-%k@7vFjj~)ILLxwG!-JY&e4ca~EdYE2Odb!UdetIe5bQftTTCB{n`dg_~ zt1o-gavt>Ng}|NrVc2YtFAlqS9KR~DqPfa3T-1xxobvjuAk385%#0%J*gT29L%jLh zZLj%4-G|g-nKSf668zZP5uDrZp_7e|sYN&A<9$r&Z?8O1**Sz3raS@rA5tJX(FOg1 z*Qi0}OZ;f4D#T0TklpfDJZ<$hv2?vP*I<~>g{O2=ul!Z8K;Dk--F}ZV>NP<(bPM^E z0x9M)?Hg5?<%927DiQo>JlQ%~iCA4u#i?B-WNY&hVz;u2TW73_J#t3kdG3$lU(;e- zlbeCVdIQiukrb;{vt=HeyXm3hn(XdtAAyNGiq_l7N^}vrz@DFA7a7N z_tAvdB|gLXOWwo%OBZSH&l50b*H|*eVHerH(hF^hdw`I^JUp%DKdfqY33p5nVCdCE z;`*%(bV}sN)15P!+K=6M{)Hm^Tv37Oq};^!%RIP42W^R=r3gCar9*XlGGAaRL#npB z5#_Ju_5n6UozPyz1tK7(^|5}l4rnBhb$_V<~cr=-- zP>y9|rHRY7o%q;O4FQ(R6O~tr_Bk>)Nk@1IZZRB05_`p{Nw*J0jL`-^&86ZK??~zO z9p};Yva`sk_93pn+=Qc6YuG#cc-g<2wZlHka;M-PV@Px2L~?(o2{CR9Ap4W$$$X_s z;;QTjx$k7z4sgc{LieIo4vxqv+YZO*#^RYCX{6R`1iSXkio_{ALsP>a<0AQe+)~*t z{Iu?Xz=?Onk5U`4PTvH(z4PjXj8z6SXFkINE%{(rV2q@!c0%Ble>8G(0lal;X8H!j z%%JN4o9SQ9JH%yigC~um%cZpJA6VbSpM`z0>@6qMJnkh~a92e1Z*CQM)_@F$U&Pyz z0?Erz4SUsrarP^eqnYurKv?!=INS2~9r|=?9x7h(2KWCv2F}Z0p!~TSZ1_GM{C37X zZtDTU`wWq{U;NCRtSKplw;8s$JZS}gsq;QRXlVg1x@p7`l>GP`#hXx*pBf3a8BMP5 zH@1J;xs~itHzf;#yy1JRt;P87s}LngsJ9Ig({cdo9Y_7CMxzT4+<|f3RO!_=nqN0fVGV zvv*1G$%#p4WA1-DbE0;V`Rbv$<|Dnc%MM>REXQ%vve_@rmlicxl=eBk9fR)|QGex? zSi62k+3D+QB@1+!`TTiJyilsH#BgpzoZve-tX-!;bhuK z^X?^|g^UEDY-W+9*gw)OhH-5%*UX9I#tx0>vALkk)pbIdPvg?kZjEiFY7&m6^ECIA z#!Rg(+j39Lfi%E}H}zRs=BT$=h|@e@7BFvj`SmfI$|a_Gm&bMuEmOaqR65r3$k}3B z{qpI7uI1)i50wwv^rCeBE#op#29?cidjZYk{Ywk`%@T%wlo1Ye{@??Gt;>wsmX}>z zlXk{iPrp217V%eWo(Ou!j|xkc)L_Nx1ip8|K7O^Q8ks*AmDT%%lukT1*!-j24s*4o z1IwHJ-OW>$iSe;>easI|`y~vU_~7iTk22+1!?%}@nX#@sCorY_ippQyw(~Ndch25C zJFc0h&lAURrskJzALv{@OH?6T`cp3?>SmP8JEBpR{3VNPtiC8H$($Chn>w4nRBA4} zG2mC(i|#jNW+cWT<{6gD&lAd)&#e`*%(Tp<2bl|ZQr?yA&}-&LYmEh$=Qo5cVh)xc zo%`tgy~X8Y^Lm%h*fy~IZqsygy^-_Hy`M*#?}>U}I=;QSZ0<>X6y{2nZ^TSDSrZzp>xX$c8=d2#7K zUK$eS65fG6zDw6G8nw`Swf88qu|%e1CgNjlEJ?!S|J$omQbOWC-&I$VHUFuzIBTNt zPyU~De}T06zw!t5g8u7VnOQCR)34s`_s{X8smta5$!qye{jd6$`X6+DZ&XtJr~cp5 zkJ+hx{wdGP*xL7>{M4sK1007|N&i#+pVHtd>vjI8`v0do8>uNO|I;zqdGfvfz5f69 zx&O4C8|$p~PyRpCSwi#wc5nZ==9l8E|NgxT5HY0>p%Qg*}J&?zrN3)2`}~jIsUKo#)7^o^CDj9 z|10nHZ|DAdn}5$|%^ITi@8$o$`5%F`vj4Yh{QZB?=0C6T@SE~~--ioV_CjHQA||+= z!}u8i^sRdysnrzHpSknIv1BQldVw~J#Q~KtgLMq2!kb-3#m{&A<*KUJL#X9;k z{VK`*Jxn$`<&po?8iREg;J~tB_}r=)lGEtRs2`{j~L^X^+R#*Mi58TlXC!pZvnJMtPx^w5l+Q`U(4wil9@|0}%_4 zQp~($^3d!Bo9q_4sg?-m#$?>;k%`Nb17Nnx1v8Gzqu)SD2!5l{_v>tQU73aHJLY0| z^EBiLTaeYNi?6v7FqXcK+h>kKu{#moLj&p6;e|Bw`&wFjas~AbJ`?h-JNS*dNM3X?o+}QCK)SW2>+-0eGbPaY3PvIQ-1J3uK;zM>3q?VQ7dWTA>Pwv|(x}qAZuNv}G*Ub2P zFL@z+nX=HT(NlnSD1PPFkrRH>_X+!8zCRi&B^z+SaUYwJ@{Ynad%>u)}@Q0KXGD4?R4i!4_MckAp3zo#;-X5?~1*+SHA`7`V%49OB*t`HwbYn zA_a$~+l2Tb=WzdfCe8#D!|%oMWfBAvHqfmHOgh@*)z@nq+bB+-rt#G|2`=*duhy11qLv7&i4sxVbjz?&J%!ud|K)^w&jquhUpx+ziFu4Vab?39l2Ixz6W3-1kr- zSCQ(8j*h+baq~{9b@FC!d}h#FpYL?_$T9RTD8eDBEQA+!@$)B5g_gTPLPz&v?yk3- z&-Ne2cU3$=%8J!cGJQzDw#cvrtFMv!%MG~oSA{dv*YI3*4?!NO?DmN)TIOg0`G*?Z zYq=^flM$~$4$Q=ofvUjPDNwO;BIo@X)Xt{C`TIT$o4ErCJ_qq!FOQ^aJ!yCUM)o5| z9&_Anpwn;&!*x^sB;l-45OvyMmi}1oE@P((w3JIIa$@!Pmk5bTqFYsaKq4DZGgU zhg-B^>2ETAp~y=Pq~mTsb(n0>pj40X^eOxy`NxK!@~$S>T#K%II~l1PEBSi z8BG?G-oZ@Cho@yYp>-H4Vj6ZNT*q9xYY!?ld+9s0lEr9Ha zb+{im5IuJ0z(HjV=&~awrH^H!lxx`-iQVk;+zYhP>^E)LYKY~&dzeIN7wfBhf;qpO zj-=-y=z5!sg$6I!vkXbn^$DUSDnHnWrWtggpA(rGar`w=z@n%MGL%%%Sz z+p-#t*+($D?h3{lCeiO+E;Luank6dd^EJJW@t#+*`S&9-{KXd~p5`aZCtgl~+jKp+ zMRm~Mr9;r?a$mIQM4)@Q66_Sel5p_=y>8CuKC#F6w4xh4C%rGidx`ZV3PZ`*C4=hs ztfK^PRq|8R7 z8D?bepc`kel3q7HV++wn=pH{&|`nD}m{C;dtxd zgOOLaplJOCrs$kOi<<)QuJ9<^nOebA^~TaQuNqwSO~Us6AyADh#U#(OQ2L&ZJ;yJx z?GY91ecKRPVIxOtC)$z0HG3M6(}tW4Z?IZli90N=hQ97nY;xDcLO%<>XKOdUshmPX zsH9*|440zOen}66iYWk11KPnBLP9 zZ+BbZ+wg%*Zg-3*ar7xRK>a1XRa8Nt-&jbO_JHGF2b$=T&DQPqfyUA%8qhkNqV4Or z!o}U(C@@sa3njr7-kpQ({V1GNFCcrF5-M1T>H=D3oIw!> zPtew)cJlcAg|^`jH8p-_{+&HhIzJwRynNBH>N@o{{lU^+DD&QzK0)5*8=h(o5*;iq zV&57isHE{1osN));WHx)iQCQ`=L9l;D^Af;PRJH>B+k;g3*-DLRGd745>Uf5zVG2b zj}GHOLME&ohSD0>ks%7tB46X z)3}&6g1gKVaruYRGF zUP{C7a&NR=O(WZxxpZgr2e#_?bT+iTT68R5Ums~)KtU@$8$=t zz`q`Tu3w-vIGBy8N@6Cd+o+;XE+t7npgRRJ*dS9tZz~Sa%3gtp?y-$^eY2vLC8`iz z%SQ0CZO}330f_}sRDEtM*?wC`-%gE0>C$2xmaRv^)Cwvz|3-zLQrH%_gL2e5*ky%i zvx+wy+TyWg?5mgTW}*T#>w+kGV+adNNQUzLRMd`+LELVEIX>D<#%E=b{qrWCKmCOL z8{47T)q*GffADj4JAC>@h+IBbigt>sXxw%^9yXkKu*oDI_jf#57iTle+)?cKyiifc zD@{79DUH5owGhAX5j{K5AC0e9Gs}~s*&hcPVk*-7qYC4ezh-f}^|^>HR>t=EZ|L3X zqjc?dZ~UHo8cEHbxUzpGl=Su^TBHUq3m3R-oQ@HleR$xsazqY!NJpPPr&VF2QKT>$ zNvaxXo4N*Fx63il<1|a}ZBD|6*VMi=QCK}GLohy&DM*y;q=WP2P&_mQ(;xLW({`Uh zsi(wLBvpTCsdoUnH)mqQew{ZQ8;~Vny`EP6yd%{2tJGZ_xOM*>~@wR zkNnk_`_|XvpwldhwE0bW2?}^o%lMqB^I>`9CsX~(_@~L>FQCnP+_K`D5jEKM?jW_b zpTl6^=Wue%Lf>^KV3Jb>yVQ6*FwA6u&6h>~Nqf!4-b$cPt$pdysORkDIXAKRs2^_h zk-=(DYsf591t~YsefMzWSr4Jmh-m8AkiiEm`OCWwaN&V&51wJupPv}?10F@XywzEX z8_jLRhHD*6wM?6g3J%fS*(I24_YJd-enI`2DE{GZC4YSRC)e;)=6nAt^6D3|eAV4c z%<6|U$%H?mugl%c)P~#OF`3(H217TPQ}aEG|v{Mx~N{I`iDdQF=S<8iC-xYCgS zvHlB3qg?QVb?6@V5gU)3LQ3dG{!(inzq3b%Z$GZU7m+v5*m0OEj0qPVu+nDz?$#p`{^M*zXV|fQIrX1~U9iS_f7noubjoJ190IpO$-S(3H3DDJEV57wtM|!k28a zxO|vOPYS{&A1&eJcV%J0vty#>rnl_Xo&IF^z!&V)C_KKs2=n?mqT%@jjL@DeUgt_v zS)K`{J~yH_0W&bwWDEwSX<$ib4RvZ~lHX8s%F*A8jn5NcKPnyTa{bBCNfC0s(_o!5 z$n3A|e9^4|)$C4SC5>13MUDsB>4L*+y4-k&-W*FJ-OCd(#M>H8x^}R33&*}>Ey(dp zbgAMe&ys)4M@rn~J}0Z->wObfi^OwPqY`GEvWJxJ3b=3SPs8lpsE63LQ5E;o#WUWL zbl_l|TYsKEe_O(v*IwYA4W^jNO&~o*9lkGqlCNbAjcHAzHmfKs`!N~8#xE#!d>9X0 z0_ z&SMIHYUj+Syc&$64>w42;ZDl#XNl`I!SHH~#qP`@FzY=CUcD61{a2r&3a^p$mnWn> zEs9TT_T@6&w!ELc3E8>zq$HR9OmxedVzg7pD!huu=SYc^i?yk~%Z~O;+d-X%x5#h) z4N4L3Bg|2Iht6%c;9&X-e)4ajdcFdlR~NE(?^;;I;Y_x%^SsD!-bVJth|#HhBkub| znm?E+$0slRM!Qyw#srDAupM~-9lp`{AX<#B=4w7c>H@z#@)Y;Y-9Q@~G@17yEpq2e z5qd`wk;W<1P&6G{`8~06qyf=QQv?S{KsVWwJl%S8i~VAp>ZcZJ_C3RNcZJcA&4mJ)W**^?*kIc8Zu|9Rlb%dEhp&^E3Ks6ckD<1I zH6||d#=X;iqQ@cUMK)(LS-W8fj1#|8ikkyU2gh!5RH z@NkjFDNZpGY}>7;DpGb?6Dr+cHen9^oF7UNzvq!n-cjs| zt4EE;QOscWIH;~knf7NWZ^LZ5wW@}F+u}hHl`DCJ`*iNGvjr;ybSP(fE_?fJI^|C= zrK3rf=rtk;YHj6a7R3>Gy0Y~>!vYoof^L>`6rQr*cTrKv2oS4n8HMgVkvN<0Z(~0Jhk7$PeTx>2m z#9Rk2C7nYj@buDF{E$tBX4iH|1t!Cn+yD@kL*R}h_kMeLeBF2%fNiJK0y%eRM+ z+DBc8E=+@BaRR&puF%fR4>W#TCH305oo3q)rAt}r^ktMTZS_Axnja)F&-Aot)0}u_ zI?I|8c0J}xM)wvbW2ms-q(2tFv4?W1J9;T7(j3zm))aNvtkbAej3YjV8H);0Su2T2 z!%otZjaKC1n81CW74Rf74d+4;o=&?ny7i_@$t*;RL?zyGrg75}15e8@^|bv5&qwtmTe{=-^c? z{(X@LxBNSYN3Z`%Ipag2m{tj zi9*5Jdn69ZL$2Z+Hpd*Gdwmq>qVz7&katb&>Z=Jp=H~MhPsGnj&C-q75KxCXkwZxjR+puOLhu_aQy0mCAbZ>niZ5~SX>Ee6D zYw+ydDQL8D9Q#^FaUK4&`)@hfd@7}<*A9`?)gU3QK&%Z@)y_>*;_*5-6W*VWifzI^ zsQEG)tId0&@y{u&eUk>YWqTlfB#xpFSkR0wn`l&mv(RlaRajqXD6E-UPmdzQ*w!OD zwC~+$ZWXM}uiN(Fsu#CHg$!oseo0hU(O+bJKm$G@cWK;I6I$MxPtUWO z*p-eeEOh8~aM@CboQ>Zp@8aBC+>I=8w&gfG5bTXd${GCg2{r2+gl{@$h`(0clF$yg&br9J5 zI;A<@V7Z$|vNdKsn4mI_V)BWdd$@yb6ysj^_IrnoXK$i+&3OzzJp;zqTj-BUJ1hO( z$`sD1(%-ONWZNJ|DT!mL>gR4jBGggXm8v4dZ(hOG?wj%Z=Dqm!$Uw8S6Q@P5ewnbV zmL5E(Y$%^F_6yY5IzTdiw{dO<#oM#@k^_o!YVa^ zy3CVkmx(l;?ifQOc8h&TJL34v!Q}P%fQQ9ZF*_Jz{!8risUAM~tr?fXexMG0idz z!Pk~zitJh8XyGy8w8tUgW66H>D$ar9gcER@oWiQFe-+(Oau!WA?PjuDb6LZf)olN| zkK~^hL((Zrsn4@Z+|}m^UvT0auMV%FHIjcRJ$?*EjcjDE3b)cb)m)02RE^&s{IDoe z8qb%Dti>;ReaaTj15Jqfj||50A{g;E%yNC@pt~ zjovf*QGA8n`gD#3$}eO#_is|X_}VE&6_gYW1YPZq9i!!8Hh&&TC8txUaXKK?a70S=$RY=6H}nHp5PA2miwFX>I(+0BPa; zH$_2qcmfV3?}LVF1^xsy(prgF`j9`7q93kBTi*cu^>GCC+d@jUqv*?6bE;Xr6oIP= zZ-0vIXrmS!9@2yFxzhxdv#q4%dxnM_0O?x$vCDJKh)EaHn=@u;3f-d2^g_mVl5_Kbx6wsaOcvVwIE9ZJHg zrMT{HjO>%DVr^Mjs_KxZnX4|bn+xT+nPxjWPDP{oRv4c;!G}kl8q7CuGlE-;Eauzy zfuuq|TDh51ovj@W{Ocvs*<``)rT31~`~&4(XZ;X^A^1eDM25oHH%VugFIGE3V5xIEzylIFI+>Qy8ynI{3AW*ZDD z9SMCivF~WC3`Vuvkh5VZ<-Uxeo7;0m(#u~lRj286ZCMoSo-u=l*{Q?-O)OsSI)}oD z9z6K?Gd?WjE$?i9&JBIU@$z9Y*ipP5%YsAM20aT>`w>VUS5rxlk-(%u{ZU7G_^P@b zW?qC}d4Fm3SW_I@;)$QF?ubd4h2?h{o-BON`sX-P<&ZQQl6M4Wk_=&@KN+?6V(HHj z8@kB#$#e5SeAZi!@qIHe?WH2^9wMThZ!M@-&^nCWUJmQ{%lQ7`CZ&v&L+L$bxTLxZ zbqe7^%jI<8y0~Ay8sA9s4)sNP))Z4|(J-@j>N#c-&s}jb+7KZt8t7z4BOZ>+M6Fd4 zmS~y^Rq3OJqn>8M%J)|&_Qo<=SNoHlu`a>X4f~L|Bpih;7s$E%X~XZzu)S?H+3(x`#yt1z5aT2IpaV;||J4 z_vI7zIrI5jkI~_!juhq8LCQMbIP=g0i}tR?Ecac^Y5G0p)m}<1gD0@f28UQlqZ-?6 z8;+Zr=kYY`9pZcLWQBFtX`N*-s^Xn+`Hvxd51L@qt9|(Hw**IKY{!(3XUV#76@A?h zM(o&LJi4+({M#oZ@~J;Rg`iC6gG)Mwmgft)bX?K)}f`9z-I+Nsaf{p`){ z9CqwsIJq265RIzxWUCWQ$>8lRgwEQEH`do^zd;8rUGa^KWc0E3KEu^$8MLjbq3m!a zx^7oYOB`!RPD~71IY_MWGjSxA4cY+nonbJPu)(_c7}{W6MxBzTWFJvP2W}jr)R0EH zHTE$x)-5L0a3xI2PhvyD^ogrP(@tAYo~E73{iG}TfVR0}97#_MseMRePUz9QC#K9w zcs=%_MJJY&8u3mkNAA)#2ZgTUb>E&C>ersn-yEyvvrg6X-+y1AY-Ba&eaymwYIRJ+ zEPNUiifP_DbSyfX8C36KPd}#N=I|DJmFr9a`)A>EOv;Y=x-?zWsxju6LnB^7vW$4!ybuA8ivlh{X=u^3Aa zhl}wNS-9?G!)sk6xQp)&c=T9@!1iOX99e*C<1KmAmMC5@YCpHHG(v#o4)m(oiSJvq zDZoP+Upq!2p+=ESTe(s+azcjL=YTreJx2@ezjh#c)gIbaSWk6>_R^SlQD$ov^d{Q@ zyGbn51vA?bIJZ5BbWTbMes|RcBb`A)!2=KU_&x*ci{@hRB!(`xQ7|+R^9pLmQE-$x zW?i&I`Mwc|J-+}!leS=AYc(pRrMUmy^SGhfEUF4zOg`7_urWm%t@cZ4RMvXtwDTzK zKR1h7U;k#oW85&Spq$h`H;9^6tB9Uv>#{Lj&a6hl9wpwXC|=WyhP}0BUYCZ8)~u3X zcMm0zbZ{YC5M01&A`fDAl^6b)9-;*+S5U5O9eY*xl(lqkBORG(P}nE-2~;QGT=fqW z_PUF&*7N!NF9Ui1rc}h**3m+TJn~P6qFMm2><`&YI z^Ychg`z^`Fj>qXG&QQn)HH+DXsm}e8nZH%dGt)9dEY-WUbBH&8P?Ib_5HDJ zv=4;z5kj^6Z{F(pfd8GPM9y#9*+W&ajgb2qC;YbKVY((9?(D~y$G$jqGZ~v@uTw&B z8TI_?PNx(Zy`B?A6U>g&gg)kY=I?^W-mB0n*A&wuwQxc~1#SbAaHAs_SAt?Nz2v01inhrMuEm&N7 zKvVMP@POw|{JORe4_T0dvxOy?v!)uoUd3ZuP!bNyoWdF#d6sXH%(?=;u(#fW`QKPw z-e4i{?>mNLTHy>#cs~7gp&&LLiK?*-nBOe zb<)o;@0Aq48Ms2Uqh$>0;Q}$m8k9!B}x&6b6}Yft!02BK(5U zWv9s5tkN`|8Oxkyl3{ac)9)bqtQE`? z?CRmsx(0Vz*5a{pCcS;AOZiWq)6+LgprewCLmw9Kh5Zll*7`dBV38H#U%R35me{`Q ze~TR*r$O5rBWZ%{XzJ>_nZ~caNn=FXl<9wxRYWGTLxK^O%J0S2eh<+d_)Kh%%|z|H zb!bwQV&@Akvs-yvs8OoFXnfdEHc|c|JL|cPY$O*^xRO4pSDq!E=c6#UU>lZ5PNz-R zK;5gINILsDnO0iBrfxfS75!#%;o-D6D3AKL)y{7 zYdbM#Z9WG1HX_zQfp&!trp7h1X;0xmo;PqSR~}NqKWqx75Xk|M8nY8lC5|NC{XvDv zC&?(unWqn!$=5D7;t@kMg(n*(3MngG1lZf*LBdKrII;;6!+TS}n2qeEw-(L6KZ>+J z_ro+h@i?&S6Mnm?a4)qgEa^ELb4K34n02yz>na!e(?^VrxH}K)L-cTV;z3eNwx^<` zU##Rvo#>mNC9`EGF?2*2&Uh<9`QSRyx-SQqp1VCo4~#_e=Z_dI9Z=4Lq6`#B3NRnBN{MyqyzCe@kcJ#jc^;(EcavzDVF*(S#h^*l6K$QjB=Xll z)yU8S@)_9N&@^$Gt2p zSfKKI);w_$ybAYfo)Ga)Z&FkH;uG<=_DgvKZ5WwHfj^GPb4q&5?C(bTW zfKzffmzcJm*Tx3&5nU@_^I1H%432_}Q-5@A+)brV?CAb78D!;rWC{aEu&%O)D0=z= zR&V6^uiZ`5sjiHOKg+RTPe1ytoy7E7!$jYr`=a>IC7SwE9o3#%zRpLVJk&=8}k9P?CQ3~g~`E)~3AeE4I>h<^m^mUR^-f|Ej zAJq80nmfpPx&wwLedtC;F!j5kf;+1M(R1NgjC{78w2f=gQ=%4Lr%GTy<_75nOrf?> zF4X<3U<5?=umBUA0B{Z5RQ%vj((zoeFc}eknMQJEE zS|iJUb?0GJp7_~Zyq%Waio#5qfR~GP5IZx7*D4<3_9t)fEafu#Vx38fWI@}%?1hzG z6rObKMPH?z^e(u9JiA+Iph_|PEz_~J*C80a8^j~`zd`Q_=WyjkENxjJ_9q%>vReJy z7iny~CTehxsAo76eFjUa$H64X1pD_|U|rHpT2-NdAd81IQ1>#n$}GY1^?E4G zkmGlbeMVbgGvZe`<4uq^o;>%0^V{vhq~n3Y`)8iQ@}XWRpS2oQ8nYnkuBU1GW3fWX z2;o)nDB9$Mn?WNn&@&8^n=~*r<_Hb2w-R2>8Y)zLR~NGHT%j?K@~FVTg6eMPWBoKY zthE6S=Qkmt;2J^9T9hqQd5h0(NwX7Pw1MwEhH&uiZ?c+Z{3T;b_=yilI*z z6OrK<3GO@yx1$IA(59CU&1C{$oyY6gC2_Qy}R5BytR8DAHW!taFSqKPHoanpJ@jO?PcDVms6 zG6+@@^vP4`XMP8zEM0^4A^5fM6S*3 z&K+;j?w9&Fn>~(1gVROYG9moW4l${)nms?ZXeRnLc*9v{K4yMM;CAU#dHnDDXuVX( zkM!KiYa1r>s8(ro^{l71F*#H;&J&~7?}MpD4$KWz(K~AhTr(YE;VXw(>y44P&l10u z@4&K3C$tPz$9%0(jCzMLT`rE9q^BZeQE%?ESBi&Nt*4?qFRB>PMA8o*LNDMobg2wi z9j)Mr(=U4U}K&CYMjcDf+ z2|nv|KC+^>Vnf0xTBp#?f?aIM>&PqE&b|uc@!RmrHcy;p!p`C_F1C$fnjx1&&qsTT zLO#zXiGqzJG)m!OfDzi;Q)t+yRCf8BGfw%PZ+`G-7a76Z z{yTTwqs*;W#X(m43VqLA1c%0**l{uqj#c5LzuAVoY-PyY*p}?>o*>nG1$27yLq1;O zC-0#xA+)@=p!=uhk&@0M8o%rmdmG)r1~v{Pg`k~i77fLP_z=3)V#4DaCHbzLX9&2Y z#8=rIz?u{_I8IDq`$Yw8x9wh*-MWu%?b${N1-&4rn#`rOZu7lju7U8)68xQ(1lRGs zi03jp@Hcn^Mtetz#|<6mN(IsR{h9P@LM0rkl5oFsKSEl}F-~mb`1|TJBSRa0@WKWz zcO;s7e;o?PhelXsGX>_>-E7=FZ+h*MN?}WbXxrNDWZ|-ac&a~iK8!`#%G+Wa?HGcy zGY~s*2S%(ez|EEjOnIb>uqElNUTF?B_Z&fw{|==xr-kgAvkKdFW*UB|?m_eQCr}jo zlT~*`VO&*jUVp6(rxU#3o0d&EHOIN7>u%nq=*G(g9olC2mckx|V%=+b*4wWqS=igs z{R?Mk_HId%Q#Tfs8l~~NG(FyM{Wyk2N%ILtcVOY)i6E_K?C68F)W>8P-aGH1m0E+D zj8p_WlE0tFB-P{2r`=S2Ob2z9#(2|z3Sw@=i*CQj5lxPM!G7jx@vSe!@l$$4J-xHp zwhP9r<#r@3TOqdDR_zvrq%3C-TJw4KxN07rXe_YiR;sm)r_`OB=y3mYl(;>RO(_gv zo*^r^ldT~ynB0QJep_jmt}!h0`taYyYX#quzQUJ{r}@Ai-|+Xc9G4#D#-00Y9v$6WOG#KIAoQD2ybe@$^|BJ)F?N^u{aFGQAA?;q8RH zzX`lG{xpv$Z0Ehqbtr$^Cg=}0t`S4a7E(>XXKJgB1iEF1EvxPV=eVq@xna^*2yud@{Q^nZn-qbLlnJxQ!LS&_) zNN+_77(Fmnv|{s1I`*(1TGx3Z-k<{V<0ARRdTr`pYv@Av2v61gh$2Pdgs8Fdu-VTT5B!!H%X_Ui;*`~Z z4O1(z&o_hoY!YFcyn)C4a$)KVgG58jm04CWvC={HETKjlYN?}0(nXKqPKRj8$cbY7 zkOz+c+J@sHOITREH}xs%gK26Fe3Q<2{yZ`P%YEc|fjE|*_GTGfygrqkNO?*tHe5x? z>OI_kt%2}OeyebMe4was{wwynj~(g8rHfuvdVDCt(q&ng zej*Y94L#vmWD1{vY3M#0PT0AOZmu3oH$Sc=!vPLd9Boa8Jhy8gV_D3_@e+?!>;-M?syg^v|Ph*7>ydsU)^af2JD+v*?AB9F6=VR@KJlsm(ihCcz@T%hv?X0*)&dcwRURwbj{!v1w z#@3MZn9(HvO_%D&3e-PO3Fij_^JQIdEnjStq#t6lbE4QP@xI3`(wAU6q#eeeyr}PO zF>dBn3=LW@-nTHehMgE*#vB%%U>gthWRvr9S;~dIaNF$#SqC%poAi!4K5JuVQ!i{2 z^!SL$FON*6eFi#ovR;Cm43DqTeltyn@o?V~c(JtP!v~nG}{J)0fy78e*%9DM>4E%sLXcG`;! zFKb+o_#~S2x!vq(P#iPT??*K%TGaaW2n~&yji8G`_~M+1+;2V<_x=%+)JhTUmOdkj zON^(I<(8<(lpznXD8=#_RxD-O61ukKGKKveg6W*hBHC#qVtZ&>V4xl zAzM)@X_6Tc&V7CELwiWONTjW)($M<0D65hcg^Gv>g(ON;QYsBeW+vy_I^``VmyEu&3s4@oBaBki*NNA{zd>B#X;>UPTp?aeBfAkCvp zBb|}IJyK$DFXGiIDL)(|vB>X3=+3HlgfSvUPPRa6f9ZU1Y704f>5_&{I9nDI$$B@1 z*nG`A%O02g5FFN=681UwfL6h-^=gILUkH?-fw9A~DC);!QV3-9fMVxiCF>pMDuW;QRKT;zpxF_yW^*+*g!x z_W!hb=G0Ewy>JZF7kgokl=+=C%^t!?OEgcd#>+p?acSU9xDUAuON%H3Yxn@`%IJC5 zaF|?F=SQw*vk}kB*+`QuR2x^x{JtsD1>rnwl2PoOWvTzCd^4Z`w@RH-_!> zD4?_Eiijk3DC|jTWc!ir5o8!tMZHY1Bl&??=+G6N&7jK{CvF zFm@`sW1`hloUZ(a_a2Yo=PAd|OW73@xkZ@h(+%U!Pe*E_9TF;~J;qxD%FrH6_LZqv zkll>D{>Avw?=;mII3Vy!3VhbsVV3Ly{OaEqYQ}MxyKD>!c6Ctsv+LAjOf;3hyGThS zv6KtKaaV0Bdd>WdC$dkGb?`G5MjjAXD7cAc-p1mL=hoaPK%4*Dq{-zkNcY1cOER0c zo#utr3U9B=(41N)8eDjiPD}_V{V^7_X@r(I7v03f+ zc4}oTbGbUXuiQd2A_HmBFAcip>`reM3!?lkeX*Ontf=eu5oemOz&Bti)IEbxw&MoQ z2Na>UKosm>$}`PC2e$A>JOx-rl8tUCIj!x_vpdXqb-FQM+*d%G*k5|@EYW-KH`*}U zpUgk2lit@hY&enzxhx0JinF}@@(%8&HJ00c&qii|5BgmGPNg=TQr7Y$Dcuey<5k1x zL~JVSB2BPHykO$_*P>`M*G?Sz#ftwjoXz#pr*r)W9tc0AfXm~L(M-cFc&fGw(=0|{ zp4}-j^NgW&-<4>URU(-*>66T6by_Cz&FyRRx$Bo4K5>IUBcy#NimaK}yJBi1L)c8- zjqPVA+l(qQwvl~2(Yj%b?-EB!)?9$VI1L*EkhNJdZXvL*? zwnr-jzibY{QfVBjoCSP5`H58XjA&Y*KGMfuqMupqZywMkx=wcd9bn)F05ELtZ=b z5#~Fe$13|{_^RrGIh*`g&aQ22=k;{fFnBqpG~b~EJ0G&zJ{sb*s1APXZyoR5OM%ZS zQRGvfYVo0Q=2-q+8-udysmI2Ju<0=lDH*eHa^@C1m^>GaYQynI^(^FGh9ds8DT?xT z^NL|EyzJH_uF90zQl%)t*m{gG&H4fj_5DRRp^1lWdu^=e>Dc_*Z)9WK9!%2=0?F{G z7g?`2!2bDWC|quczbnmX_2M|DYopFuobFLpw@9|*kfBoKMEBNG-nh-bcJYmx*I{-LWdu1*Nyf;$G8V6dpZ*^=F=varQnk zd2@?2zZrmhX5!nBsaWkC0;RevZa1`uAAMWG8_qF){1fAsr5XC(2n!S_bwz4eEIIYO zk9n>ivFTwOehoM0!~J{mQ-T`r@hg(0_!`L(26y{OhOczYfnL^2`hI(rnE8IRxO=s=IBoc8p_j!D*1Yv3+qy=9 zPrLpJNq&`Br;&%2k72l|HUjd&cbRul0Mp!MENJ!Xi{Br$5itKbxeA}@nk4!dFg}r8 zU3wS4eV(JohcD15^OMe8#*o_*L+U>pnf_W)7`E|%O_lRl>^!~%?PENkEd*g_%Su@6 z7>>Dm-caxMTFPdBrIH2vF~U;^T9230p2Iq#-(f@X>OO0+@M}IQe%(dUl@e*s)P!yt zRgm=C`1<@Tmed|XeU=V_mA~Tai4u6kdLYI47grLJxQfpl?$Fi%xtTxVuG|1$=a)FGk!6Q@1lyuXIQJndNE<`1hI6lwK%<7E?=Kf$W4Bq=NsPk<}Ni4U=jNtk~tQT9T*-EixX4DN*s=+|Gl}iEMKhEY@vUJrAGv^W{~8tU!sIyb)rH?U@fu#H zd)d4BBI~}$lr8AHmDEmIvuN$>Lf6J1KCRJ{FW%|PQ!ZHY>R>e@<&yO{3iz4XBCAT@Y-();2sBsb(JeyB0Ls^fSpIaywIo}~h> zcuJ5{$BR71)sp-0y3ly+n^j4hf-7lnvMzpp=#Fj!hM~Cj0>1Zno5{rT*~o)-!suZ!Ic|F2MS8 zggAMvuNe8kOMI{=3K0t%v97u&_x-P(mHjoPCEpU+uKG$k>?5Mfz%X#@rBGWn14Vk) zNFLKyboCl3${(I8F8FB!voj*p1MlF1XAhoNXvHl|8COtC!9HzY$Qi3(P4rL7`IkY- zJ3Q%$`ztQ?~+?DHJOrp`v zoSa9WqO%p#*{s??niH#prTQFyGu}}Dw&N__A`L(LWamsuw_`LO`obiPwvad*=5%|Lyxrx+Hd#m^nz6=%^=8++<}{+vRt?7EvAK}q3+NJ zD83!W2PCcLR*_%WO`jH0lQ_h_F8isvbt~P?51@xdS=9AuFH}GON;(Vg(fCYR3@kH3 zPsI-yCpiE%&(r3KaSvGLhuO4+MbdetG-~>mM9sdZDEapq`Xjyny_fp1ve`E=U*g_x zJG!I1aSNu5D?~tdEpF9hz~BBin5#wBLw0y5->aI!b6O7b9Udy!9Na{sj?JRghS|bm z`;Wr-hyfJ6`v`?6#M7XZ6lzW0iQZKiMD9M6`P3H00Rv&OWe7OVa#oT@|jazKHt@>(L_dTCR_t zliB?w*7jYU-lPtp8>V-d-m@EQ#Tr{OQyC@}$?J-B<9drLi;V@RP)&*-H3%}n&CIOF zICgkli?GaB6)O)Mq~Y4*X-$^{lvFmA!oNPF5n6r3WgowB!!7aB+;cPS(tRU1kC|q3 z{?SjUMo+`@V~y;N<^eqISd3&RYy30VgMx=TuzR+cT>JS@c$+PZrp6)odlZ_2rsGBK zBI+|If_e_#fIVGKJx{;Fp+iL! z9X5#~hnZ6Pm;vm9%T+eh@grGHwxnJIQ`q^6@hoFy5r(%VplP)XuYH^f;p|qNnm+`q z5sOQ4Q?c-BDs9bJPG$O;wCbM>rmy`2oBDsawC@@$LMG7H3DZ&a?lRtOsb_vhTGV#= zk)SN?H@x3@!T8HH0?A(Qu z4(l8a7^x^y&MD3oOmc1JM!OeqT1 zO3$_5YEb0%9tJ1+^Jx=u$C6cri{qKZLa*r;tz0G)2V& zyK>6$qiYa9K-YP7>o=}&A`lAl$5FB`604^hp~-0@s{b6urIH%0(XQd%^feOaMIlQTwBl{CBV%(~5PZj+&ump)Kfa}%Kz$*6f4je7_GNxOe* zzIM)B8tjx!uBY5EW1|l3PB}yF8YbxWaWeYK24KF$MdaFDfpJ1TPHrs4q7ZHLT|EZl zN6v&%MmJvDX~Ms(7|+wn18J3aCw)mBg4Pp%X=v~D(0o<{!|zfaa(e-dwo}Kf9vRrA zti<0XyYO8y0klTB3j!8P@7IZaHX)-dgQqKxTIJ*d)U8;!h?Of#2- z(%1=pbhY{#n^;{=$2z(~`*~kj$NQ0eo(kHg9KyBKTfDTpthnfzoahy=j`b%su-xmi z#5pa({fS3V`t1TV-S(2(8tS;6!DP}A?Em^*{%zYTTl@Sq-%HQOhkPSr*7x}+jp$`Ivm@fDXW@fEXe{Kb2hH)5Y%HoBd8iU%bt z>ALP9Htgpo>m`~A==bvy1{r=qN_7wvrkX)((m^WUt10e&(nlOTxwrUF$}dMJ?84~a zEZELHiz9EOdwl(E*ak{Do~u#xse3ci2wX=2@k=O}H<4?200wUf#}Y9J@2pPJx6zqY zyz?pjiZ}@IW&nOREyR!aTKwsovHbh>rM$-I8oO3?iFKVbo;hZGrST!3>Eoh5srq;&e|H(3Z25FnAa( zUw#UqO;X12nIRu|@;6SMFUPXpZm1c)3$2SKzURvV_^f`+-`%^(?W=OQ4BLEGAWy4at`kbjO}|Aa&QI8DP|#BGUJmOX;H= zapZ|PT5_#1eRVSCnjFNE%3!SCI2_AXCehzC8=7%#GT*Mfl@Ij};0H~+fTo_3n2#y6 zdO;XjJz7Y;?4HoA5D^z`&(T)JDE8avF#m z?>dS{GYbT}%`aKshGS$a?aMbU+s=CTvJ_fl4?=C5I-X1JwwRr#!KPe;u1{CK;@BHh zO<|~bC7mzM-ryLPLWgTSZf?v^oXKc*PZC? z?yFeh(E_yzJ-Ag*b^cjKfmgFn*ipYg$aBAh zF~j39cFj_xxHnUIVmy1ZrI<}jSi+Qamg45;5aiTmVbih^!UWqkVe!g!%xTt8s0CSL zyQ~@hj9exjJ277L|7av$SQ|;HHl^faRZreg&a|;3mhQ&IkfXjjA095{QbwEdeBTCg zFHD-T8_k-MQc8546#20=BL5z}UDQ zm@n;vn)Oy7W0sUneyk{t*8auIRPxBMbp!2AT1I=^RA{E~j`;<7vjsy(P)oHo#Wm`X z=jVg;(fXt0H5rV3Q*02satK^*%izTIu|l_OW9D=FAPXPz0?t-BaCh*9QS(mm@2aJu zR*tQhXgr_GnD5||ZUpkkje`*OR}N+CbLoK93tZ5-f-lQruwknl-}_dMe;L&P%j#GJ zzH&nc(}tH%PwuK}z$f$@%I&MOsMnuDb~s8$=;K-pN5di%eTz|E2tn0%|^V zhR!cBbD+yP|>#&TC`;cbf3t4I2oA9|G z{*3xYiCz6@Piv*{_p&^NTo{PQN0gEOpELbi<4yy%IM6+F6AV(R!^IWUaXaSeiUo}*TF zEO)vTX*1u>QgF!KD}0mVw=wsiBIXr+adBN2aA~9 zU}=s%U-Cq>I&j7J+B|-6jO0e>FE(E7EgI-+h-T;d@c0M%++o%bZvUqUTP62hfV(zt z^N_<>Z^^y;Z2(lHeUVSzZ&=6 z-m;~assD%o()r!SF0Nbxk30=@|G|je&u50Nx9RhR@ANad0LnFW82jxbbq5*9w0O#f}WiGq^(ct5X$3>Woe`;Qkgt&yoTut|*;&GloKR;7|tTmo0` zdYWgi2)18~{ zbm>u7a%t^D<~wz{dq97FLi&Hcd$mMjtj3UKO%|C3oS^kXUs3eA#jx)Z!rmQoWJeFb zVjCxoBa68`$!2;I_3^Hy5zB|r61yVyuj($vR#&oc@jH6^GuU#2-%24c|V04%$E8QELZ;OZ~{Mf(Z2vP)>LygS_-S;CedGbX?1$7sXM zn>55Ai(FpYK<8%(R9`AXKkYnylqqEg)}Lkdu8yo_Zh)Zg{+bl-wvfTUFwFWI&O7(y zb9c=$K4|x3)TP{l_Rm+yU*St{y<~|E_$|rZ<8@K;v#Dda z@mcckOEmSaAJg)*O3Oh(+Tplv%*tXGWgEj{?4%g-Uqdx@lT z7f`O@#vC0kkzVC)R75PcxmB~3`D!cE%DSPvXH7kx_YFeG*Hp=;od6BTtC;#I8m^y4 z;GXhLy4GkUWOqih?e|ki^?DTknm3c zFxL+HGuC0$)d}>VUpy_p`IR0R_o5P4Iq0)6DE{mr+HJKJzg$=&o}ciL)OYBCNq&&P z9ojgoa*@(LOs63!D{J@p2KB_|%PzE9Vp=zBJdUETw=gWY9u+=` z*zs)^Ixnq(1udl@cIWJ6`;0hm&$&`KXFWV4vjosIOMU|W5==+p-($9Z@ zt7fvi^G#3w>0e(ib5x7FwrX*mb<*?AnMFwYI}W@097o^NB6pmez#XM)i&zth&ldGK zk~)emB>Bd3>oXz`=^%jw7ZLMq$SA4)PWFm%a1nEX`XKRUk9>~7a- zM%gF!dvpuU8o!1H-5N;EaKseFI0W6lj+d7mrE5tjSNmMRZ)ps{G#-uFEhPxCsui?j za~N}JWilbI+^nA!pA>1#ubD*flES?_Nr>Pv{xujE)R+HooWnKF9-tvt?$X!z+4z6n zs>|CNQakj4eBYl&Ze1%LziP#;-JufS7mv3|0TLH8ncSk35dFCfMiEL_u4o9;GyaI; zav0+nNyhy<*vIT1yqE9{Kqc}4ddu2D$hDQt;qqov`8shi}O_y~U?XXsGok*dyZ4$tBP zZv^lhH68J_^Kfz4n{i@VS|Use>~V8LZ%j5bz|Z&)Oqx=Hyk~dWfEVvcM%a&wNp2LE z&|UItYp}o4=YG`CNUZmE2Bm*yj%|+2Yl1FY>>Z95TfRyj&w-rl{}q%j88f?)x7gfw zs`zqyEP|^gcJEsP`k!X#^!`IZZ8c;*K8zB7ucDb=npiM-2DXe}3|-8_5R+@LpR2_e zgn5c(*QICRe?j7Eo{e1dcC?<=;f4;XbZz!1rl-H&=E}QJ@K;L_(SI1$6mEsv&ndWI zBa6-($^6abdakA4T`c-M6B)OSrM;RmbSGS;0CnKpzT+sf*@&4B%`oA21%0#&A;)}B zShrZ_@gogqBWmD!?+Lo=d*aw0S;+;Whu+SeI3j#Ta#;>S6|X}#xe%}8k`U{l%brbo zB77Edh1=oLlJ9vuJ?e0vSv(H8n+Wbfxzf372KE^{;dYmq(3}$%h62V3t@NLa9Q-rR@DHY^dH$yJ ze1VM|_w4;1BRk`vyRL$*`+SrdhE@@VZGf3e2Q9K`l724bp84;1z>a>Rwor%krZ&uJ zXvd_@N$70OKzeL0KKb><-@b!jXWS1)m#z7URm$8#t}Fi?{Es@E9+5}HCJOvLh~HW3 z&l|k1@*i_Pl0)ANs%_gtFUm96*=dg$`_F>5T7^T?WDbV!uBM$v{g7KT8?G(0@Md~9 z{%O5FU$}HIFC4lb)~SJT2na^{jQ|!eIi2>~E@#S({itqUcgplvCC33)6!C2u9sC|f zchdUc^E+R>>3tu|eyrv1pN`@|Nj-SqQysiK_dXv|e4Hn~97xBvTqmz7+8BHB6dQDT z0Mno6!rYzuux&;SHrMN7g)f7SVJYv4H?Jf=+inl!P0GS56=go7_yTDzD5c01b)>KJ zgsoFt$!0pJviuxB>@vS9an0fEnr#|=?w3cGM#;j%TyngGd(h3yI=Z8JfYwj&rciek z%%66ioiC3d?X00_GAU+frk4r9HU5I{Pk;WQE|-T4{L5`46>)KU7O6+qkY$l4>lD7Q zu4xu@zO{mmI(JfhoE5_S9P#gd1SVSFLeJr|`CpTz{E7WOzWv`FJpNe#kAl~@Qeg%K z1w}mQ)WCx+2WaSLMRZ*mfH7`1P&ZS>+&$+gqGTyK1~$^1mEACLSsD)QX~M^;uW|fy zZ#HJz1ZHS*nt4iD1^Y!3&lUI_5nW;lPH`mTS3;*9f*B0|t+kf(n1{3Isn%{farYf1xo;%B zBYn`V=MpqE^`>L@X3&4W^J(9OXqc51Lq4nv4@yoZzYuf04O)RkR4muc2gru9=0Y9hMk|v$T4v z35MQU3F8wvklpqhF;ZyOd3uu6g;9qy8=}b}Xd0EYS<>3h?r3o{##vJj3}_KR?aI(F zHN%sFIBY+04*uJUQQiDdIR8donAp858y_K3!U#p2c@hk_wGP6b^Y`q z!(i6=k;Y8CN}9)&C9dx&W&PO>¨N%N~m8PAj?>@|?}dD-<+LQ)#bu61b5Dubp_0 z+iaH+*G?TQYGvJ_zIRhe{%I*4_h?~GF%MYa>x=CCe-DH~o*#va4|_AY?DKTocNAS& zKc1ZXs8g?7kLa=}@v}yK;k?8M)<@@HWR`}_l!tM`g~d;83{UG~U1GIb$eg&E`54lFBoi&@jjZ&l;!fBFw*Gk<|MSAnwf}(nJdi?Yi zUmIvHE-~FF27DNg<9k$a-|!_VYbIJ+FCQ_yyT{w%9#zNL>#+vJ9rsYdpKdrjzy?iU z{qX9sC2#4vnLis9$7f2ejLz>%@ax4;teBfh-7_mGv#UH(_FnW`Ow_D z2`5uB;j%xA937P;KFkh#4q2fja2eEJoI&!67}7U+O|!QQL9b~?F@E1^%rV%BKM7}9 z$E5C5GIT$w&W>l>HG`Op{3X_uZcN3qys53JicUY4IJ{;*+!-PDvZPHy>XcBaU#Awe zTNJR}DG8fIExx6$iKeEFL|oQB9M+Ws8Vs>FeKPvj-DO(}qgl?R?aWgvie?U#M^o@p zl*v4n@dgN8e<6yq5j>ZorUFHx+?dPob^)3Yo0c})pn+JV@{L8u}!qBqJzxpTj<)fPeNCtg>2HH zFt+?n9(C`ohHy1|gwH=nQG@oA+s}M@@nkDA+zXH$IciilHI)3 z)c@ji@^@WHv!==*cG6)w5;&dqOf})#9D4JH395Wjn;d^poP*+!qv%rfo=q9h%Ivof zA=$r+;BF@m8Ouxx_44F@-fQrm-S^_e{#a^CFQmtXauBa`HsRADVWhp5P`maeysDxw zWV{2uKQ^VOqv}|<^oMNlZC^UOB%SWc6p-0)CsCA|bGu)a<{OWXV*jioc&nC#Y*i=F zw_dnket=RH2EhLEM_Msc4cERX;cQ|ny_oxpP7U`JUp7q>kC>~A86ipR>FfaV7^Vj! z-iv!k&jCAs?Za1HJM7O_!?XBnR5j95*wxX4O@5}r20H6YIn6L;TRVUr_3KZ5LJC{& zVL^7rB8|7nVbP<5*oT1}Q|B!~;-_q+>U*)gSA*Hog~{w#>Uh$)?8$cQi)02#=CseK z3kk*jSeSSfCew}}JnJe{qxw*r`)pD%*hnWYUf>1+^Lh2szxd&zhSNsjWI6K*8#`+a z+O>`0cuWp&%3Lw_!4hbs&&4ufD|%e-A@#Z(rdj1$Vr6y%KYgTtFW7A(JqO*UFpUm& zqC|#WYHwl}HC@SgZ31nZs!E?P*0JF`WVrs#`&j&94{jYKQtL16+e!s`n%K;C&P-&J zgC7ZA4s*#mUW*-V0tGo{_@q0q1 z$E1Bso;B2_{vvzZ7`hskMt!AoxL%Vib*g8wGP7mO{<*}OOwJ~sDTnCm;$t*gQu7m?@<}cVp3Fq!ejRk^hhyIm zJ)~(LpcL6{Tz92E@4hdOJBQ2gwVR_+e_#Q6OpT%(V*#ns4#hIeS5O;}NyDvXVTQ(X z+`9ag`uh&S;z`l?Y0wD6UR`;?*$_7OunKKxSw=cOZZKV%i>*P=CC_ag8yCzeYQ8NF zcZtN}D~0%2br-MLPIwZf-NlUkg8066f;cZ|54ns}#)f1O-<1{!i#_Mjwve7EJpBagt~4UaS(dNb)`uUG zJf#D>e*!K(fa}kxJm6&pzZ6o4+_Q(Vc>aEvXy0Y!wr421j~(0-#$efg6qE>mm9xF_)%EY=&yDDlDis9C^C{aqQFJQ}St~A+TSXFm zkUvYi;$?V0PO=F>Dq^4;agOFN3RllbDA%U#~I@l)A@ z#E%X4V71^C_^<}-HM;=qfJ=COHVyGCkkyCYp}ZSQpro(IM+U0#fIdci&6r74B>9xw zd>W`*gfF$19e~#W4X(W^9Q~zT{9^@KiK9MBYC0JdG)f+)yd-YycqZ;dL}L$pXhlOJ z6K-LS1(gaE zHNX+;?WJy#jf1hQvJ|bZvfMDT2RFSX!%MnVAmo{r82`DMPl-+8AH8=U!!s8Fih78_6$Gub_mA^hLDXHH;did!5m$4s!86hwFL)NWQrSIR_*xqTw-Z z-7e*UP2+JSr3`OI-$zla9hN!y!|P@=gr|ODs&#;9@;6Y-Y`B5?S=X>SuNcEC$Jv;a zL#k%%7UiO|hHC>O&pHzF0&2h4))PF<+Qnc~`o zp73F63A~n>=+=>DSDu$d;UFa#o!b zIA}H5AC`DhLm3t|XdyZ`Z-c*DpyXD)!W++i;f9W?;+mS@h|2wnC05svv0*$s+Xei) zJ{PUS>zJHo7He1U$J9!XqMwwxw;SMqx7#;zZ^<>hZl47=iO#^h2^qLE|1vhtm-eG! z2eGp<4C_K8;g}VNK`%~Wv3Uq@U2&8rCx!9x@kjAoFA$|ko3YX3An}S)N*N%FS)vc@ z-prP=qW>ZAq`G*zT~_SZOIe)0;eo{RWI~eMrj3WUx-d zBW&{lD?$7BDdBN?Pn*K~6DX=~18p6@mh9j6q*r2M{t-xP`2$$Q0( z%^O9dX`97I%NO#2=Z5n2vtOaeHkJbK9j7?QW2Ba7OPaAgP&Mu#ewQww=Buk{N1;D` zd#o>7+X<9`()mALxtEd~- z7}p8gIU$nkWf@`(=ix)!AiD2zhhbj>+ulo`x^4r8*(!47`ZC_*hMrjBzfs)Q>pT99 zD@Sd=IwT}iBK2@GHaYquznZZ_OA@K_r6wwJSE1LW04)6T73WS~WjF3b(GOiy$xTqm zPtH5TS4|D)_ht&1ykHY@tWKkNuN|v;s!VzwpXpQJ0)+OA!TvKP$k(_@hYtDCgFnC6 z#(B?as%1}XwBCuZz^Pao+#RyN3+PlvGPP%GAmOz$OwLbY6WvxYyBl}dm$DnM(~;q~ zP1X6{Ub1Kp^~Bn)XAqIyhymlJ+0luaxNUcj_nPpAZ|x~7&MZ-(_@%2^^|nSUt3Z*Q z2Hj(?UX-$TUuV#^k>i=?+}cTJji0-NOe_}B-S6{h$rcMN z)G;BOl?#OT+b>hWz7?dg=O;VRrG=8FE}(UzF0gM^U3sI>4Jz8|5@VtPebrOc@2m?= zI+sYDd6mN0%iV-JhjKdEK8H+NidgR_KWWQ2f9mslFEfn$#dDQ?dFucbUfx}wo`|}n zb!9M>Yz=`^*LwPt`;=|jmAue?$(&f<4yG}C0Q>Ac9hWmMmd)Z6Gn)nBkJy1zbJ&~We}Yo1E#I#`7xO9) z5x=2@8xcg_L9%Slt`b-o?Zj*YJ4`*72*-hvC;Q!ax+HNgS^r5M+lqxyXqZA)!Hu+7 z!4$Sn4+`fK7La?@XuL}}M2g4$vg^u@v{C&7HlEesolpAmwa%-lbFdku1lO_F8NW`DSa$5LP@(^EL#$10rqpzweIN7fY>MOpYt6nO?bR95ehBD=E z%#pfgd*i{)S`_pB|U7oEpi1!RU3+-(w*wk$UX3zgaA?}tK7rhSy z0;ceMZzrCb^$YnwZy;#VC76y(#Ky6Au+*|Uj}LI*S4+Kc(4~ZYUR-BMOY`VmO&7`8 zDNDmPJ{h}sOHXI+I%VnYb1B>_hfUf*729YEP%8>3rLc(==p!6 zS=S#C>@|18;Jm@`Za+m!d!Az_0^+G@`a!I^;Y@mABhfSY5w2x=^B3X%e9~1f{&H$} z@vYM<{%=bH_w#mVH-_I4ocG2F)j4u_5Y$BZr?W`SI|6>R9R`DdALZrvrF|w-owED4?FQPH}vo+3n-lK!lTrj#?32LMa}F{wYZT3j(4mT@_gUB%4R}eOslfsJ==9i=G&f`y$&0~sVCx+| z9LKqJ&Qk9Ax<3{W1+BEj0*L zh{U4d#Z0-(hUKT96P~^^h1;R7a0||)y3Uu>qLe`;J_(fEWrXM_l?ul#Y3J%c!54M` zH-0Q|;fOH}_i1BImv>;{I33C1_kuQ!ekqJhk6}MXXwcNl+K`pqOZ8qU%u?Tv zr|7xz8*KttJgsC?GrEU>lM`*uXD^rfJSRx!66rHFPf=)bmlckm&=j;(bA-F4!E8~D zFU$U(i|!`=Fq}CNeOe2!e2?V*d=>?_nrOBybt7}!tHrL4Sw`MFhSEm-W@dVO`Fu-b z-ZY{eBmAro_%n=Vo|r^!57Thv#!UR71`2=v9bGc-Np7cesFZrWo_Kar*^lGYKWZO) z+fhX`y8Ga;O&*=tC`S`6?q%+ktr)Pk4Abqd!|~TGm?*48`RQKhHdJE$_Gu$+>jAp$ zc*~~Xh9&hL)ECCGnOr&Dk$2nw9VW7usPKWrsEpl!nA`4nA|rX&b;EG4Y#e5F7(=$<`;yCt( zjhxhlIUg5uq~Dp&spZpl81_-bi2qS^o^d(8Z5$`sX=-SYRH9w?bsoogwF%i7Nhw0e zC>hzKy_HIPX`@I|M4@PE7$qdDNc1Or>p7oSz34@s?$32!=Qw`9?^mCPMQIMS{1xMs z6&#~0DlX$M80*VEbP~?v3mzF7N6~?vm$OxT+Wf6;I5-p-+*n zK5R(+_RpeU-Z4kh#_>46iuKRODB%Rzb1*DB6I@!x!uVOfaPaMMh>%`DzN!BHD}Z7iX47%wFA z-5p#gG#0PV5HP2uG6qkqMW1^+@tesw{CCz4Gj>iyedfn|kSt4wZ$HFocZu!4ewdVzgv2lu=W2&w{h}4u=D)1 zC`=9JaU`S#ZmklY|2?jo*d{umXBNG-U^2a=;D8?%Dx%-Q7tr{okJzUQAW~Zzq%#9y z9b597e;Ah;KB`om>s(kAFEUm-}|dcw^J|q z;1i4sR=c99lNIiGAqyv?@<{&sHmLWQNAGS}OQ#RH)A_%j!O%&@`<%qOW6ukTV8oIn zmR=(Z?rtQ}YzE<7nL#>2Z^G$G>p*c>4?>!xP${zu%D0|_u!|k2lah?1uFu6&D$U?$ zy9BciG53t@32yxN8Dz(B9GnqTrwP)Ubi`ANj`a9*OBa+7pLjP&k@|-Jb}z;W zJZR(uP1>kEfZW6~=<#PfjoW1~M=l>@*6u*1ZY|8SXT7h;Y}gsL1$G4ihJK2`7k(Ep zZ_Ql@E|9|r!(-TPCyu9oE<>~WaD@5y(6;9s*1WMtsY9YzMOPR|We#gz*vMSegzpHbwdx%$+lZACc4K zuX^h6((#vZXn86Y7s}v{kK)AB{U6!3u#2ScNd*4@B}_RRh!0B*@#W`0e3vVWY4uiA zbYD5uI4sWVkG+Hj>98a}j)#Q9~X$UXBZAUsnCO_y1q@SapGtlNv{ zJhBnq&Bhu1Td~WZp-=?f*u}lWQJ>~wf1onVdVH0vH4Mb3-nVddZyBmzDu(HP$?X1D z7!GxaK<<$+5CortCGP`CY^oIr_2~rlOK-7mc{%?4y&Of@8aUxu4xW9)7%FQsX`0VI z+VKeJw<&{AKwrQz*J&8NeJ5P7?uXdLCTO%{JUrx|6Xm&k$&=7wvR0xLzV@nP2|YsT z$7s6ccO><9dQL9KoPS#PS3o;gsK=)GS;%0l)r%PhN^7Ka%5jukdX2?NIl6D4Cd|2S+7^Df*(h5ZxhDnmt@_F9;4KV-+>Ubc{5;Lf`3$~D-n znvU9MjImPF03E#hU}cs+Y0kR{%4gXB;=dPfGOo&h13kzG)Pl=|-6*i@{5S1R~SS0BYhFF=yS7&tG%`WC+?z<9S`oSTjU ztPK7IwxLJxsC_3Kz03Ai`hoDhNdnKPE(h7w8N|%w8M${&eA-t=OcKUG%U?h8k+J%J z%`t$T|IfER7Y0rQVYOZeRaJMP-xy!vU($69q|x}ndMPdsP=Zr$k|AS7G+auO#@qIc zonm(k`V1{;asCjV^hw8=?EFynS{*IrCg6mPOUbm2ZQRnlZ(QcX^XN1-8dc`R;9H?0 z(qu#|;gj5UpJXAtH?mkc3;*#jB&NSLB$8rFN*W!QYj^gMV8Eie&4Gq7x zLdUso+^@rWQVMTyjM#TjdNB?QN`&xRfjKV`CBjSV4bi6y5c>)RIQRHEG}|viOI&L( z-98y}=8K{IqFa#rvkQidB|$9J367;ig3y^@P_!5aeBx;mtYZe_nrFk*(nv7&O5yH@ zjUs1d^vKsMm(k_;4a|=Jj@4b2#4GXv2`H0;Gl_>FXJ9)toehC9S92IP8wd9iEZ{=@ z7rZn{h$^`%P&=0j5^ggVrq@%5h<<}J*;#c%*%XwJU&tIN>EvO^V-k6w0Oh=nwy6OAUHO@Jsq1*!1ri$k7wTN1y?b zn~lg@<}>-QD}Y$6tR(|ir(k%jAxcjO$7d@7V6>DyNRCQ_e;HozUN4XIMn{l#g;&J= zZY1a?UxFnE=I~h?EcwfW%lXD{6DTo$jUQW>x+lLiH1R5eK#5m97DgX)%fl|83?;}6znu)QGoBOQD8=1i^QFCY{-}E+4EqhVIh()|&UtSo*A#M^*iV@a zy&E(!cCRsvUu6WMJ?8Lfa2x$jC27Wo2Bc-r@#>^gxZqPUnoQjWvjpO>()ScmV%}dk z+(~fxd@d(Sr~d6iQ~OzXwlfYFO}m7_H>^pYy)IN+@_2qg z6cayPgJ|6V%pWxxT-Psy4QImO_9}Ji^>Ph;I&~!-nk^9Qi@L@Y&Q`!(38g3`ork^5 z**Au}33)T;f_>6nFx==1?FuR2C-Wb8{c6IjqzkCL?*w*4vol6ZBl5QF{aByL_y0=d z_Zvj=)xL>1{?jZ}CW){~JC3B3mibLFq>$7 zW_v{MdK`9LhK~FSsB?FO+hv6$mUiR(9c+GgRShrlz1Y7$9k*tAVVkBlHzaXZVE(UO zaB?UIoK(i5#ku9^`9>0|EUn2-*9xK?whxZi)sj1RONoE89WI&`jyLCIVg7W&Pw+M8 z|C&wW^$-69hkF#}?FoUOYmcFqxCDKA)sl{1q1Pl#_pDZ3Bw!&VWyN=Bj3w*Medxs&w28s-KXg1LqO3EnP5pO@dYfS>IV)5T);`2<8WE5HO8qO2U{B*=6A_}^cN}g z>u?DzV?Fx50bw*c7YoBCzsWeyG;ZvUMuOrp7`;4?NRFP$J$_|u?w$P(#(Cd`V`@>5 zuHjEVq#dGrHZ7%PP3+!jxDp)&m+`}&K5k2VGUs^y5LdFhl(cvgvdbuyyU6Yrwz00z zixQU5uh&s2{yOR$n}kblpMaDl{*;e0r5hW&ah+5KzI_#fcRp^%g=+RF zF~u4+**dYi;yOfTzJwzZdN^KdJGPxUfW7H}JC=QdWJ^7~b(eLtr&_~uZ(o@9+6eY- z*@DqIxa}Ts}adqQyPoK-vju6Cb9h9v@~AjZVg`iP{!^g@95xsJxu8+gpH3Z;J2PI zn2f#-ts6B^t9Jvs@3KYdT1iZf>>;T>{}E@&*JO5AIJU6)|FbRWIO_3bc%I2JIklSb zq-r<9Tt{4WPzR4${J~da80WRJ2LHqxfPU~J;_tEqR&G52_I;P&*LzjeIGX}SKjOhn zI|*u}@^E~90p`cQ$LN#3m_cXbfpv_3eK3wXwEh3NI7c@}Ibxy0Oia<8kJZM>BzN~8 z2-9Wx>L>~=ZSfFp5eB`hW`kpAJ}7;dg#N#d&>Lcf)V}{Zb?Yo6%Z1j$%hR{Pi!Q`N zSG2G_hj?!^Llz&Kb>i*IIEm7XnL7twmQ>2ZIzjrC-B+A{)KvG|d8{cX|p=B&+F1-rT z#$A{;=QcX)l_8%JL!G*$YIA~gd1<_~FyC)tON2x<#TR!;T zX3lg69WdQw1K;>42)VcdJA#YxM^!HBhu1*hzEyZgzZ+-t4B?uoQgrLFag^DgdAZMT zsgmIzYNC6F+qR)dkRiRo+}eIDk_#80tyl|mub5zk`ab;r^eD;)xI%waEJPcXK*d%Y zM6Ct5)mRr#B%UBDYM0q8cNuxTC>(b^^uQs>ci>e12ac{}PQ_JWxNutw9B4QU4;7!1 ze@e$m^0#L4FXs(0l1c_0tDA6q=~Kx5tdCQ4WpJK<63DpigtxV;VY~JMXmFC^qbDUULfSUh5v3G2=8}303&y!Y^&wgF2+0z& zQ#xhlwpnrT&-yTgr>_PR=OprF+e{Lpx|a;zwI{l1mE@Uf035n{6Sf>U2gMzh;B)^3 z^%x4J4!`Fx{)s(ompllOYf7Q@cM)#5Q-TwZ9>d7j?7TCwi7R!Cp=ONgnxe$$~8e#fm5nQ^u5=2MBz@$Zni*q$4t~X^!w0aNj8$658 z7Wp#gu^gtpiH5XS!zBA|IjHc}aA`&~%m^M3Odcb_jrjBm0tZL&KNpVWe_nY{2R+B2 zjX%Lh!}hpNd_GxWR3gZ2d1HQ-9>%2De4HqsgWX+U!8>FwT$1+Vp0%VyM}8o9Y3_jA zzeH&x^GUuf8p5hER@~0FGN#Up#+i!vU4#CdVL0I=YKG( zQWsm)gUB)URBnp0Dg^6}C2Plj;UtXgNSI6!g!#RL;C!vq zt)!-L3h|z?oa>)*9G>`8!s3;`A#`5=?AX@_?VCl}9$_I}@3@#gLoX^aQy#O2MbJ&& z4C@Sj!%)#zxF@2BQ(KzIXeU$XwyuU%cpFTTzrd@bk|;>=1n%i8_^(GAU9E6wSzRKz)!9N~m)pQ$ zC-z0Nj?F4# ztoJ7bCXB=AsS@}zSPFubwt<>>JUGZ+hb3(lWXoX}5-P3^?Y(Q@)~Tz&f7nds+J-}V z+X7rzxSQH+^rz?FSkd%@mvM@EHV%25!d=_S;B%H7$~uSQ=QKw+@xB};ueD{ibEMHf2$MuiC%$i|8GBz%-5`F(-l?mQuMza@&ZULNHv##n*na68K} z0P)?W1JZ-~SiwAyb3}Jz$NL}*9)1jSzKudP#%A+obBM+lx#(;bg#SWyaMf)Sv>o)s zaGN;Z?{Esw>80^5vZvvRjXy5_-G^0mtpIvE@s==~oBa@mK9xmapM4PGf8J#b-73`h zT#bXmWtiAb=$Hp_^x>*TvTj8Sxx6F?{4TnX0EZk>?p;NCST}q1d!1a+f zP-krf5}W42x})AOeZClI?C2mSo3EgQVg`=>xC@sZy^sH5|Dv+pbgJm&jWQxzu`QPE z1vKmFMw33eTfK)B9!}w4PzAR2Tt|maM}b{vso;87F86R*7`|(2!5)2SS|_|7V#3&a zrza0??2@D1uGZ9bksqBjzmE8{s)Elyd5~-G0;6o^i(fw$?<5z|<$I4(C)qhP`_47$ z*4s!wFb=KFemVRy(+6D(6L3|b5_C$hhIy&MFyqHB^pm=QP5;&4z2+%YY`P}(mzYQw zJXk~TDlVaW7FyAH;&HgcG8&&~GakTwJLnpj0F5t_$g9>49AvH_wdyvMxe`viUTlVt z1FBNV2NUun%FecCBkk_eqqWaR(KXr%P^xAGX?=2#!1jed93x=(y(26;c@ig> zPoj?bmQ?z{es1y)W6m&bzF>$WG%a=p?MyYHlbj5At84+Uv3MnadX_FGUJ#&3_#`|! zT>!fd?}p#WQJ~Jww&%o4KtWm+k3AZIG^HST6v?ujb86w|M@@{%B&f}NVEf0mKT{*%Ic@fP@e&k?K{yBOl`90jGy0x%qyK(^A; zr1Dw}`1D^Qiq%R$q+DUsdwGb6@dQoL6yVxkQAs-;{*;ItpChY?Pxep5pF(qR&ytno z=)rSbE2s(2ylho4wn*RLU&u9n%3KLavRw&?-x zG0g`nMaI0_?vGL%mI9fZ0+pi_&^m84_8n%OaDP>-3S|8>#V_z=>NND$Q^4AaiMZ>R z2fEvuK%B-M2y$lKume-5kkkzNJKc+R?Xd-&{#aP(k^qx`EQYJw?7*|j6(rx6kz2{F z#AUQR*-&1KNev_T%}kdbOPz#@0c-J7Q#7`{Sc%_F?=ZLM1ez&1f%IK|MxxZLK=zn1 zS~n!3cX1bH{n&ulPi{t?{YOy!)8ko?GXasR}Y}WWgGSZ^Alt-mWk@IHgTgT&`VmPQ>kAjR z&j;&$o%HCJQGEaGar~=ZMOd*$7Ph?~Chn4c_}RoCbsAXDyL$<|6`WumtAFsYWEgj3 zkD-5W2+>y`|6sYM44rHC2p==QV~48=M2M&Z>0N{t8typFI+oqBhNE7(117&a4@EKG zVWR37JS6f2ilRdy#dRY@q`oC%W@n-+>%OlQo`JuNitzjX1WdVn3cvKl%(+6I{mWfI@bC&_0 zX+YMMjp6tFKFZH~yMZ^7ZX~NXj>yhxBO#wRqD!3!21~cVJ`E9kAeODMg_1bWaFAp9`5sgz=o212(E3Xa42$nDMz7-M7SG zXMq#?)+q6MJN_`2%_I82Bb+%fBCyRo5^v0Qg~C=h@I5XDc8@J_p{X7UpBsx`Z{DTy zdDp0BV>wk*J%EeG%*V%}3OMbd5Za3W1A*T?SROhXH6!=Jq$|>3Q5?nZo)^OR?}+3f zHJh5sq|o4d8T9RfJp6q&1IzS}qpR#XC^mEh8tnz;KmTCer#S3sT8K8!&cWByFX2k_ zS?Ek^p?l1((hJY)Xc7~1hCLM!i7XS2Sf;{>KfbWB&L1w*3Zk;c1Y%>op|MdM{)%Zs z!F@D0maOT4UGA|k@^z3b9Yv@? zSvhX)3&q@cRlHDH57%NIK#RX0{ZO}zis}i|dqWC*c$_@{{^c0nr8+;0TE*&`@p{|%Sly^QO&1=4hlXgaSVh-T#eK@#|m@vMgN z(wZ7n4GX2nDpV z5yjkFlc~zX7>pQa4Y`Zscw^}R{&x39-eyJ_4m|I`EwdlvOOW-=MPdr@sm&0s-UQU@usJwgC?fsKoxpcj zBdE!!p;FxpT-E%7q@Ri)$}bL*?y+Bq^=N0vT*HBN{v0|(*@B*^nN5TJq9N5$77Lyo z#RYf_8a7yi%d9TaTHuVmi9cYa;vujwXeE*0f}5pB)9bIw(9O9A{e7-WB2lX_$F~9x(1vCiG!o?q%XmX*C!Ii zN#TA+(R=p0;T)CCQz7_(3LG5phb$hp3K$tAwlY+!{C>`_-URj>!VD^f`|<;o6YDeHf%;C)gRDZ z)e76ge$u}%v4et@0XYcLf zVmUZ5bsfsa6ys#W^H}AQ&RBDz7aYh7eWAoR4~n<>YQoG zL-&w(_{5|O50+d<(`G-{#jgWL$U2}p{G3Ya|_t^^d zwu?Z=ItZ$Bi_v@cFXUG$(&OLm;8XqEnCjS$v65w^NHvk`wD>J}i ztFfbU6rJVq7Hu=vaRKi3WN30Vxv8>&s@>R5D^nNHkkM^)fp82B5pkoHL-MecF%fkx z?IV8zL{Mg#GCn&r3H$XNU_Cht#Kr(K0!47~;2iki(n_i(s?d8P7jcoYCA#04%5p#z zuypoPcs8XQw>^G=VTvwnmJx($2dhDM`y)85%GRr!ci`HmMwoEL3+GWuT%MQ>Yh}7{ zY(h3JW9*gDAw5KYO96zQU5X(&_Sm*B7q{Bp#ffUs{4W!Kc23#EyB8^AZ|or)ZPtkH z(xx2edx5)Y6+^1s$55R{=KWl3h}`ig?)k~Zq(<2uhPxN=_?hzejArnmA#$+H;i=%( zqarSD=^^Gs&p{i@HtbN{fRC1y!U0=_L^C5iVGFp4qksRJcRQCaCT71zZ}F;E_}@ z-t;uZ%c~=yc#0$xVi8o@zlT@y&*5CqcbK;I7YwC8r?j-4&iGkM7hK+q=Ug^n&JRCq zu$ObaaRH^)A;3_;3o zZ?f^XGi@_}aY$xsXKA-mV62sZTM?WDRB?(n9MG2Wd=iD!p*(6kS*BLGJ#XMz$Hq zkm7MGn0M+s+09$P(jU@-6bC6mj)pjwaC{?HDa}Tg_5#N2BJ^vAHr0MBLZ5{tVf&u{ zuUm3BAs?X*}_rw3AG5nNO!lSkh6E67;``DG(NC2dW>H316p69~p|%zo|nw5?aFMn%kSZbf90l6fFBS!e*=|NTa7L zElYZWU&hMPP0ADbf!jQvzIh=Zy;2ILURFY@AQcAh=HLp0d-(3=eM}uGM$3#-=w5mb z!#7*t-umzGJ=qx&UMb>qV-EMP`wI(R?7+Y47Gs2!9VV5qB565OIQc~XWfin*3{v|^bGuJ z8i5x5I}mrz>F&K6-5)Y7gLR! zGSodvl|Q&ao-f(1z!#>@L~n0Rq*Z-T%GfeLt^ghtTE(&)W9a!8cKd^{iyIW(6{lm*OVYLRI&@`%CjAs8Mk6LOhXvb5-uEm3OO?-D(eO^(7C9T! z?tcd_b$_~O`#HMjXcet&n#&vvQMhcYFA4?5($PM<>G^P1+Wt|6_geXdewo!mos_+C zi?j*4Z`=;eVJycfx(Flv_uvNKCh~lH5#Sdo^qU_AX5-F)599s(*eOG8Hz?9&?Fw|M z$1-puwxm6L4Yy0Z0dH+nr32DPgAYl;->9t+yL&aNbp)_ZLmm{T<-oGa>FCq!0Sz%P z1cBMz%%9K=V=85^r8*wP)i0nEo7*pm`~^k1mN+tCh)0{F_=$o*e#`pBe8JHTIO|>x z{HQG%wYP(z=btQ2R7^sPS37V-P8Fpum4Z>f9|(-y$Q9DWRhWjO|m&Da7%yYW=3wh3R=usPn1=>f%&Epn_kk`9M z$g#g4NV1p_ZuaVcrmFE!$NbK#RnL=x#htKhFbR6sZ-9kDX%HZ-#%C=z=j*5F@x1R4 zYTU?1vfm9IpPazB>O71+9DzwM$B_PPj;Q%buzvRmOleKVS6wSnFpFg$G@sM-Rd;Ev zHoK!5Bab`oyawsgZ7@nl1uVh?NqFFR`ZM(vcJ`lU>sAJ}+jxSG@`|9NCk&&rR0(E# zS)p^~Bl0@;GvVfOkl6Pc-a`(VXc8j0*b_lV4kc36-Wr6W^J*B&myPS@<5)*OFvJZIgUn=HG+FPmq})H~=-lw_GTEfmDgqbk?+sI197 zock~dy)g?9(ueR06R*I970XfBb=UB#6hDaRe2!Kku41e)WT z@Q~*T>?xg(W5p$Ltt!O~mc>f0x{Z#X*?j$ZGki=kBU>*~Zr2P8F5qf3o;mRqKKQx7 zZMuSvAAE;Zn=>$A&oAO`#WJnJcgXVJSD@o`3H&oRhn})gROw(BY8||U>eg9!YA_5o zFpvD@_+a+8cH!4$olv_ykiVHZp4YX`qhG!T;B555i-JuUag#B|R4(J1rR%W_8sJt+ z9=zzj3fB@xf#KKx1ZN+F3#?|Z;8V0GGk^SN`tYHlKrVYU=lTrE^YD{c+{pN1%*Vua z|A%Tf{BaR;)Q0w0ng1MKZ!W&h-R$2M<|dOk4hIZaCSvMoG9tg2EISg`rW^odKK{4sW&*c$43b zL95xF`5Z%>_`V&o6tZB0>KYhVV!}@uxJh3sSkh}>I&cPm1c%qNE|Y-3*ogz=p>PBF zGEEQp^&U8lWn1S~bdeS7zi}Ecp1gmlN^hSKqxQxlnBBS)#;I{+`tJRlS4RL24@qLk z)KvKR?mKF;+{$AKC;0BHMth&^#4CbQur_EWvt#Fw9dq=^%LEl1(%lJ!aVQUXt8w! z+P0)Xhr(`RZ6{53m|Y{=4p7c_QaxvMGJ}42=R_m6O`>*6#boZ%yWBgvK=4|_0=HB# z|I);%P*-RV(|50jg>xOiq>-(6vlz>5Sqr(NnS%z>|3JRp7qtE+(WVbe=*oo>^uyCk zvy|D7%wDzSn@|4sg^LL6;x1fb#m_ z>tW^h`Pe(y$~^GT@NT>)O&dIl$KSq!_d2^^s{2h?v(_8G3pFv{U=f7-$l%r}P4t!b z!HDogtQS3js~26vE59>PrxwSO&ZRUBbKZ#W!tqnqz=XAcou7ntf!V z2JYP+0Jr~sLazf7RP38Py;zluf=tFL-kkt}Wz6{%at*sYs#DGPDk#E-BcOczAosDr3}A!x1H!0s-uvp&FV z_-7nM3*sB-4#{tHO?M}LpLrYIE+^y0JA=5Bd8pHLS#D;O1lIQIV^~xke#`$0OXrNn z%FM^0z%n&&POqeCC&KCLo$2uLTNcP{wu86L`4Di)9=^_>KrSk+0_Ek8z-;zE_>*8x zKLkors@H<&ha~Cg`=WHW@moyOFe1>bK}7ttbz{Z>=qnUFYN5b%*za~@Hh zm*+89XASQ*((h)>gV9N(0JCJ%;p}9VGQH53OLE)HFP(P z!ZetMK7<~6eJ?_0FH*aY!WJ2-=wIjFICA5^?Q1|k14p?`}CsAyP$h@1i^fe`|C z6J^eM)OWJfZ!C!yZ{o&#m(x7sOnlPaK<@URh6T<|B)8d-sAnyLeRlFBuJJJW@NNc5 zvbmi|+FvexrZ!rZ%R=ik3sQTz46gV@leS+wxZbKUeEOtP>XOn*b5fjeiDooxXTA8- z<7{ccBptk?J(Ya4HvqFumx=m*E#mfj4(BrFB*}I70ouL(eEFMt+Ef1!cRS0H15a4K zz414ZZ7IXb`E`sgv4r)gpWx#oAE0062WfuYBdGLx&INW}C9Um~iRJcE?rXy<6391^ zcM@Y^!p1=O=Fo_-?;gV^;jJ*ZrIb4rT_<>;=s*-AZki9j*~J~Q(E^j%7QpM@A|amy zqU;Bv!P<@7ahE-0>kC<&bU68 zWQb)0pYDjnk-N-!)pw6%zN;2AUs?`{Qv$K9!w)>-%((rx^vF9kAL`ogh1mxp$iqi3 zxtabOK8D1FYZuNEzcCZMX`(aDnvfrGjMlD$f2hFKBW( zDsb94ks2#GU{?De*>o$0{M}OsQy=Gn!>b0)uPB|<)(Ktwjh9o7^f9HqcXTl+{4h%aX5zGzWCwp11r(uYB6;7Ny5Cujbx2iDb{>@hR;6y z#O(%lxbwhO$QG5x*l|40ygU|dr%uO(osFQmKNIwi9Eaa!;<&5-I*4fPfrDL-P^tbK z?n@M--hC06Ht`&mu4rT#r!)MtgJ<~BBL#er@dd<|>v+QF6RxtR#3_F-R~2xH8wf5y z{%-)fTwp$wyb~ytyBk{yW${Hj$IosuV*zjJ>m_I9lLchf~n>PZqY` zXyEQ%N8+H<4|QihL%G~c}kKI1cee!q+UjX8(gPbT8~Ia{&nV>zrnD}>DpG;y5YQQWBN z2uj=Q1wR!O`LDI(`Ajz*{?1f)Iwww#Dt3!d{qf2)B0++#t9XwwRnKA5RlqZzNAQ|Q zKAupYiT~NKS{#p6oo>2;0tWW*p{`#YJ8njH*FzBaHuLLQ9MD}F^R_awo--4NSgjgoi2UX zN?W2YQ`5}})Wv=@e4D?Sbt9UIQKKxJU80Vvh3;5xF&0*3)WNwIlThShHdfe7!YpfT z6eNB}g^FfuP|3wYM_uH5E8yvb86fqg4c^DaVayP_=TJ38t&B%7FMBllsyhI#_yb)} zr{GC-e>Cvd!WF}8Z-4zV^UZ|8t%Ap3RI7-&Y*z66lPwk>@+9}&lVQ#U16*mZ#oIbf z=S5;o`016aq3!%fpy)>cf?KmiD34=y?#x_%6 z>@uZ1dT)9b)oHl`JNB!ftRtI^mZX7O(q*WvdIFNa51_rH3hsF!jiYjlpbHx@guL*7a_Vz*LR-*na63IOSe|)e-9<-tG>@cd@RJnkz;u(ZSLWBf!sZg$Zt> zX`GV*9g!nc5reU~*bg_$+2Qz(M7SNc9zLrk!HUb#+=Ckm@aLy3qq+4O9dKNvh<|rI1m)vB*i>FXXTDFP)?#<)#?2x4@>((8 zTGfa(OEkfTvDt0}zJrROb`&ov!+04l%v?4MQ`5!pr)dJt*r86tx4UDtjT<~1%X%Az zU-7sl(1wlOTv_#hU}YEuU-jhSrcw&PZY6k{A_hY|yDwj5!Yz`1c@xObBw+2Z+9aNH=Gc;)ZG^QK~08t03SRsC=vdKliV`Uta&HSx&9 z5h!qa1^G=2L0nlJwnsIBo7;Wjr1y%<-#rlmg&o1VJrx$ORl-y2)#1;rB65tgCPQsJ z%Gqe3S*0*q_1dBOvQ4O}v;$j5yYYMe+rlfA?%`+O-veuG+CfslIMwf%r%OT^)E~Bx zsasDF_qI;rv3my0>i7p~b2M1De7gH=wo#W!WwiEPBz)g_2}Vk8fbQ4pxb#B= zuD+s%O^maVap4%WGIqwU95EcZC4(Qex}mQu9PW2zLXUeBJV_aiQx@i-|J+lkW$F$8 zx)(!&+F7t)9?QB0SvXZM731bVAxHG5z-{=$e!d~<%We{3g>S_Auo|3U+5McQU!Z<^ zH0~-cMRU3l6Qwre#m*vJWK)H?v$N=@g^_giiclI>*-W0lI}GJq9~@nzjAsnAF-_kM z%M72w+JU?9-GSwlWyJB}+5z%t))=VWY)(@>`>boja@X8e*MLl?AgG0QL5RV&eVOUKh*c@cW-mN`6Sdg=Bd0c_d@;My|@a(6S| zI`iJ88XiFn(Rs*DYEa?ujjUqNZ%}zL?r}PV;|iAH#I*!(+gPxD^$K*#c?SX1mvvXi zfsctaBwV!vFXip&xKmfm!(4Ck4x{I)FypQi!|Q2#wcLxj(xsInzi>;<;a)>K_?R`PJuH_wgAx z#d_g2rC8k5u7KrV$Kn1JQaGX|LjG3RK}+&9ENl2kr2XVzxw9Bl2Weuztus2`R#fO?&NMsaM@Z4MW$g~q+Bhzz2+u7A3u+L>1Aur^ednx_lRoLYw?bm z9(;6J7%r*|#iyrFV%&G_x!iGZo4m(-MI;+jpH!;eJz(g*$Uzk(_#FDgIF3G$e2WjaaOA| zy6CUMwD^3Seb*dwd{*JYv@mosE+C;f z?Rqd5`A3wD-1Z|IE94=s?vQ$u2x&YDpKr zJW2OE_R%kaK9IJr1e{lnfJid)piB{mt}*H0cVG!B2Sj2lmyP>RR^x{bJ$lC}jwWP( zf{P9duzGVCcIO!j9$0t@Lc{Wj#B?P(O~sI&{ymr8$oxu^?X`Kg>#O;LhF{Qe(FE$^ zHIF{IqKmyxGZ-hKAAbdez__(f;rkg;9Ge}1=NtR6=D`%2-5Q0z{L65G(j`=Un#eU% zJqVjN6DN0{#qYj1u=s+VExz8w<+j`;+wWvxb;kcVI`4;^zxR(fv{c$tgNl@-)azW=xg%s0 zsYr^3k=YPhh|pGPp`nsAR3egSm~EAnD5O$IWTX&!^F5#M5B&po_x*aEb6wBp<8cxn z>!;%OjJ2345sD7?0`dB%M?_up5xM!*0Mx}xNP27lNF82)PEp5jtJzd~FMBsF;WT-& z->iA>to(R3H#k(kE`l#kO^0W*((&pYmNQa3iM2_?=zNo(dO5zuDOV+l;hQ#CLPME5 zbP<|OUW(cGG%@{OKlqPVfJwy_#_<=Yr`@IKjL=d16{LpB*?(|?ARkQ`mY_@4yh6Vv zVQBU*60;nq;nG?jxL$LE4YeuI#9XL%mG!Z_Nd@aw9zY^YVIAX1^rejgouYr4moxo1 zFQCAi*QX~=Q%3}8@T#k*b7d<@)06{2aYYaaEW>=kK`hA-p*wa?#LIurfubQF6uav3 zWIwIsshhg;9uIs1|CX&_+OUzN_bo&H{~n<4bpyNcL4M~BiKuJ7;1h7!JhlcAp8JX&hQ^t>Q+Ls!zx}< z7{}{;Fp1|`u^W@mcEQb91#($NhVELomL8B~^P$f#NP?XkvH2ZMghi6!o&FoJQW3{H ze=gAE^;LAIe?HZD^nkI&RMA6h3r=>_!Dqdbak7q2F zwg`jN#Oa=0Ss4FM5&a4q$c`$O-7YqUTTLHeY0gcu(#M~8C-3BC+lKIBq$79^Y0GIv zwF-?nAVOEoorWWf_bk>M15IDp>bYSPZnqCdy$2e3r=8m#tgK`Yy_vU z^VE~wIfi{6up#oGzuJ~mWGuIDf?lF0sdi*?Y_%DM5yx3zh&$MBV?y)u4Zo-G! z-mhWz4J~S&ra-M({vvzc6zWx9iw%<~+BY*4`iFx^?|Yu;nnO#vQ^EWpT8>c@(@P`e3)>A-rw21h&ur3)i~B z@miBJNm|X^UMpGWI;j?x?@)vb3p0RU_buE*RZMui7)2uLxCuS`NWeNNbY;J(1x}uv zY=|f+50E0Y(`z_iE7of{Aqb~;X+Xu9B`_^gl{XN&mRGm+IB#YcAJw?d@)iCMaZl`t zIWH9>Of(KR&B^kQB86eLkSF^OOrk!)^sSCqppc4fF{bp)tVFu>WgAV%e1f+YO{75? z(sZe!6}+x~19MH5VX$%=_hz(<+!1Mm-31j;u*eoK(hQ8A<3O@XsPlzgxt$mks!!)n^`Q5~13_YC8teFsfRk1@=Nu@YYtW!wW70>3XQsz2_}=5!#NLIF_kqNP z-htSIxnR%wk#~Bsu*XdtjD93ROn(gseiEjSs+ZA2?x}R(ry;p>Km<?tao^L-yNu#C59IR zyt!U2N?d*Zf>XzJ=wx1yk$sbB&VL(-n9D!topuxRdov(jy%p*hdwld@D5g4`L5t+~ zcuRXCsM~!5v1bdh(k2uaC6}RxUm4y#eS>QO11^0ag>x`8g0pT_FlnI=c6x?!^W5)n zt#@o6(c;SZ7VSgB!?t zGu>__yd`{FdWzHc=n)&nnVrFDd%e zV^n+P>Z zeWWOlAKn^U(pkg5@kR1$#`X8aq6rrGm1yH_%cE$JoyfAE0j$Thgz~ax(EP>h{Z}l3 zJH-~EoVyQh>=NQlV27$5x#Lu7xhgJYy*Hl^y|65s&D5)O;KR}q@_ph{a4Omig|$j> z{&0$f>KPVz$^;XCW&XkVs^Ij3`Zl>*0j z$^1uneQHlA*U?NTnmwULXIW=A$_pa?`9nfr0%kpo!tkq+IHc!;K@)bPUEN%K<#-O2 zG;8tE@AKGrEs!dwn$g!M#Od8#xx`F%8{s}=bH$A7CVf~BKdGIAt&C@F>?nga24!$1 zCWCw#Ng>8oZ%O;MDL7(15o5y@U9GL^G+=~2__C} z2~Io9_RwOqB$UmRLsrAVd@p=7`2q@XX~=ho`8vO=;g$3|%zHn9`W9ct(=TG7?0y>S zz>Pq{DaMcxu%J0Uzj67sv(U^p0jJ!Gh4Kh(HfQ$0<}a)}wLA%wS5y=KU|V9iP9BG^ z9AbOaHuz@ohba1-M5DiZ@N$O*>ex@9Ws++2S?5*sHcH{ywFmP)+a>Z2S!>Y#Kw&E5 zSb_t$TyS5*Ibf3%XnWO&=d9hCtNS`cvJ5X~EQ9?uqB!TS4X;nmjVJeE7w=GyGIN?n zp^U2py)v1P<`f#!wn0iaT$_nse-6SbHXq8DX8gIfr5G|h1Vve%V*maa2ru~wmdDiK zw5%x>p0%L4J9=?p^G0;4?1AL4Unms65r4drME9~-u02VBeEa;C1f6Hxyhtw?=$a3Y z?aE=oU^$67LP7SC6AC+UbYEIB)hlEy=-t(rT_Hk`s+jS7H-DfST8t6sa2A5p%R%~4 zI3BVTqQi6lVduxcm~(6tmA@ausY&e2Gb%&nGTB*BzZx%(ex%wfS(fv%4^0O>HrKtv zav)P_j~;ukU(m*7jC;*kL8x1xhehY|;P%3Uba_M=b(blnTkAqlYA%H`Wf|Pz;tGzt zjNsvvM8MBS!74rmFPS7#lcr6)nVAv1rfuoG1(`>vv@=IFWZz=Tr_(UV7&!`$JVD`U z1oHnb!YfBV;=7emSjIa5A-Aj1DJy`n8Y=KsqcmNpbeyV_AnKp&Oci~Uz^2Oy_?SQD zKuaM^I$Q;}RF+_GHOqaSJc-4+19<7wTr!i#xc``oh5D5=zhpoNq=u=r$I`VlCYjrQ75`B$2<0EK)+Z9bJ zmB4gLJH&lFN(VkYq?32vqes=%u z-6HiTra?{671*eC5nmTw#7oK#F-YGTMJFvl{bj0{cVj$Gp)0x3IsE?@akO=Qe0Pv4cZr#awvTQ$`BtBri}3P^*8AxystE#?Pd&d~$V?2?SB@sY?kv>6*#Wa3?)Q#j?F z6h5?ZCD+@$xQ3UVrvJt=;lrm&n0GM_j2kbbAKQ%zIPGUMbxYiF^ft5|4uo%WT`_y@ zD5yP&1gDfi>i=kv+RpEw&WN~|`AZkAU=I6ZA0eql0>ACG!D*)=;Ne*h5)SR0=ap2P z`?CsvvTlX-NiQnqeT<&dJWh|6se!`l63BnD7$0}Pgvj8PAf{DECV~%c6|=$hmh$K- z?M@pyed+xrd+2ZG{=0TD3YF#eqX>IHzsU6iJG~oFYGzC)1_;q3+hS4k?l#^z*L}Q( zJsvETahfJfE~PFmWi)8Z8<^I46?8It;qb^S*cvzlbJZkp-dAb*SoIe!F>pW<5QTLA zMRZm^iwC6bXqa0Hb@IJLujnp?V}GTH@mxbvxNjniy=Tv@zZ6ps$AHeZ`Q&GDuxUu` zXSlQd1-!{*&MwB+oW#EnSO2zP=gs$McHUizP*CM|UkrrWs_us(D; zc6S^>r{nD~wfs6fSauu!w48$z&+fpNP<{ru{EY9mK1L7z)A(Vp7F87(N9)a^v^zzE zbicSkoZi~Nx3~Rhw6hC6Z1b4A%>{iMZ{bw`%{b>I%S70wqtD(#^od_YG8Zfa5zk~$ zOWlousqf%i&qX+A*$Lkn=T*7kE88(YfvSZ@IJn#o^EV&nJ-oY*H*n(!Z*yEd9<44y z7sVuO^-{p$$0y*5UjrAj;Wt+8?Z>ANUZcooWzGZo#%XuCs?XX0$Y`xmCnE7dQ@ z-!9j%>P;`Y9TK7nk`mNSRGWs6l(YL#8oqCe!=#e)oMJVFyto25f6x&6)8nA`NGBM~ z&B0}7&$B+tIdtkY!}^8F7&1Nx8=hRjGiydrvs0PgR{6xuv)E0<4v&)$=hV1%o13Jh zPZ`ROOJUchcDN!G31u1zbc3!G-QlB6-}pIUw(wzGJu4C4pO)YSiG8GANiS%Y!WW3- za|cCb1SQMOI9gSJTa0euf~7}k7`u~qyz;|fYe$^@mSA<6GXDLn%Ud{Aj>l8@kG30o zK>ge(7;IV(P4jPoRIxqYIFyOs{`D}nKo{v9S_>LUF*MB2lWx_TPmRCzV@$~{OqpH6 z?%bJVeU&swD%O!H&%N;F+SRE2Y%kuA;_=Mmoq4Wo3%*ix4t@1%I-MRhpB|JOCU;kt zfOR0tZn{X|`g|qE)U`#AF+F;WIkDfks?a?Xa)?`Y1aztlL2J`Gs*|yqcI(fj2U&JK z|N2_Ci_pV)R_#Q2d>89*Jc9MhZ{p?*Z{%euqF$c^ivN*9kuH|S|H61}M_YM4- zFbgI%E`)&RH6YB3AsT7)abui~=L zZ*kqg6THG-hq}MBkbbYE=atIoT(b+*@|6U*ZJ!9u3pJp7rXkcP=R$$H5?U4)b2^gA zt;zncX|z+?^DOcR@KnBMFQ_7yoICp1z_XMIHFN*LDH2r;dHSq6x#Iw z5BR(yZ)z8Tn?No!OZ)PyR>$-1?ak#GT=xVi-9+e@kAfn@I@Mcgq z$OKe~uEp$~Etqec2p^=k!kw@QL~+=I)Wq(NC6FxZKdINWiUdC6< zOSfkud*=Sxgu}}gpoP;Onu=%WL*;H7p0o-dz&5<+uo*iZtc4TBmp~(53N@@G;nD89 zWTTEMIqDRF&Ebb|->U=oEg_hCrH9ktQ;aWFn27;v?_uq(4%{b_1heM7W?cQ5=%RHT z9EMJUc;yBVmI$O1p5;^1vn6!)t#C|ox{MX@4r!OAAQ2G9Sv@6>$ms4LZi^;DG@v+#I)u?OEe-PLvq#?cIb1 ziF`CwxgL$hoAGl7%T%&1+Gu7a?g-(dJzF$!^ub>Uw@)NT`mPb@n;a**l$iMb`VDeJ zUm(}HjxyrW zS`I;9`7m{RCiD&{<19~UbQGKoJ)66@fFrV`C*>H~aJi1-$?-VXCyS|*w+j8%wS;!+ zUV*y?tOwY`vekn}2zMtAY67l9!B-_T@KMK|R*qP9W`tOGkC7$hGN`K_r`rPK>2$`3 z+E8+g};{yC4nq(-$9ngN09#>RgyDq z3*q(W0vNU}BLRI|VQj@`*t;th7Ffq%*^WI}?qQ5|iv{sRcRK9zIE+?nyI@1`47|Fu zl89`Y44K415rtLZP;veQhD8SBRmoGx zH6%g$pK!JVI1kzEEHLVP4W`bqfcd}LxXFXJxTcAd$*TY51<%UEpSyOjD8ve;O!t8N zd#9m$Fqu3mvg0lmDs$tH(olXw9dje(;xK(lgT7W!VZUU$G13Tcv<$%}YjGUw@F3AO z6NuY(YofWr3$=bJ;;X=DY;W5Pa!a*fto{iUuwL2i9TymDlaKD-(*VZ;$BDrBUGl29 zggh#E!(C`JuUho*`(I`|gBe>c0Amird;W>?Ctd*3E2f8-lo2z}WOMNEJn_!PW8B68u zoFBkDX^6_kX54H25Y9!mhC3S(2D7!@Ve2;qxO+gEybQQcgibAlDT`mj+^|jfg7MzG z6OMC>h26MtUyie1sf#IrOOTnwkk5@dVOgfAut1od_ve%62aL!zz5sGH+y|2{Y{RoR zm!OK`e4O_n1Yv&}%NT6qk%#WQv>aF77hVXqDre(V!yJ4yNsY4=Il{#)cxI})G!gC! z|A2pu9DcUS!!3Hzm>R`l{l??8?^_9dzo?Z8s^`G2dRvhC&zBUMP*^Y{4BU72f^))5 z)^*Uqw6%Xh>P#{eYVLxGDZChuI6WNH7DQw2! zaD%bfOVK(b5$ktrLyZ|fykiwygMH;>_l(uBTDk>}8|{R-Hm)$EYZr_O&O^V{8?b)a z50XD~9yYO#!_wb8eEIbh-SVuQ>Q+9XXL}Bz=Hw`}E{?_4PZUP=aC8Pr1=M#vACznc`UYP!kiyb+FrHGahkTgG)Hpx$LP& zDakVAFJkXk;}{xF18H`r6TQE04bJP{g`pq)(7@UYRyde}jf@tkFMa`^W~77IO6C+8 zZlzb_zS6~?gn8FbPQZ-~W3Xr01YG4y2DXl(@H%$^Y-^CB2S1q8jZRiH-*7Q5^*e%xax*aO{5Mkg?*&l??Ns*bYEVu`=3)e1WHWP4M=D80@g-1Fy45aA~jvn#@C? z!a5S}Oo=DmZEECKc?$PjF9j@@%z__R*MVGM5H4R=5AzolK)mloGM7vTgSETi^?}b& zUi}3G>;+MeF$Q+8%p%>8#;Il;N5{*n@Vtfw_T+cK<<&1?RniZbobAoKykQH^Kgx$E zcq<71kO++RiNu}ky(`|r1Jq6-ZIRigR}b27)7w3`S5JFk)N=+tP)NX^$!}=k^K>fh z8bB>%M!1_Jmx=R$3UCplz^*{xsc@BX4yRzU-fM^~YJ=?RNwCdd7e3Auh22VlaPC;G-06VLsfWO123>^yLulxVgQ55*iJA=*Xk zPhBMQXH<}t?)Grz@n-N}w+S3PZ)30Mb>_J|i(^Y97)O~8^SzPMeut`I0(y@LXN8-p1l1Rrrv5ISHvEWJx^zVaFPpt*0IC` zH_fqVQ35K)v#iXs$#`N=2NJCbtS}aWB|5)}W1%^`m6{LNOZDjbUB~b=HwEkMc0+wb zDk*6-Af^kBqtQ?yrrddiHRk7NsObYL^`x6#q*I7_!V6ARQG)DSpHJd9MH3@hNZJ<& z!?Mzep!{Gih%tWdX^~{c9gjfq9S(pM9`HOZ5)7spkQ|Fl^70w`Ub#p@R>3&=@B0Tb zvREB99<+mbw{76aY+aTULpaCS1(_||xc!SINZuWIqcPL;@%EMQ zK{OG*JSzZ~;G@ucd?5_I_>WxCkfl8ZYBaw=pN{2+{4Wy+*NslX>IF05!$T`DbvX#u zd85R2tp*&~`)~+SdGXA97F&iSx z?Lm3H4D`3KF7OorcGuMdu@ROlX3w?$q_t?gb)0CiTy(SIWH9m3hW$22;IjX9h#5N% z<;@XrBWVV6ql$v&KRwWhTn9!0Z5SNiiltBPp_a88J1=X1$K9o1bMhDIRT2W(*~*}` zN|prq+~<~6#BlMfx6{4E9JO|v;_t8W5IVjNl%K4HLMK&{{9_u?@zEqzBN5P5vKHnA z>cfOoe>&YOf;#snQpo`3<6LkIEwiKXyKFqkpB_Zqm&_z*``?pNGhI0F$Qx4IUqYQ( z6BKwwg6kSK1I+nI4$8=b_Yx&svRDp_i)8S@b|*qSj}k-Gi{z(Q3Mu_jOb$D}AnVG- zU~{P>sgg+LD*Hch(Q7A>_P-Y76KiLar zP)C|-Wm;kFttjlfn}V5EQ{jE32jq0xz>cv@6U~uO5+U`N9Gpn$&Tl-ro3o~h8bLVi z(n%a@dxWtS0J*Y)a4u2;_)_vYus3JE$ zS$)&wl@XnPk;^caZZR0`KugVzes=RLV0!(E2$2)ANAR!%94& z9DyCRv6wgYIBvME0cEFkz@ncAzaIocYVJL-4xNP+F90X_MWVr+65Kv>6L#Fai=n}? zRQRPGF5POP?C0R^0!!>3ZH4XeDfnAeo?5w8 zlc?b}V1M}z*bmimb(dMU*=7{Ja2K#k`4aw0tikt&YvI?YQfNy42X!mwVC{iZ7;#LP zKI)=4V4v$;E!{r8|)F3{RDm|~I(s{!)rAw7Y3R+N&aCbWJHWXA# zUc(9j1?-)*o->{jNz~r?LwM^JJSkj+ho28%u(m2F-n-NZ2Wb} z5|tyixwl$M+-wJK6(D)hg zI8_ANMfS{1P=t3s^V48USzgmUZJxgbkJr*y1GjhHhtR|tNLwEZzD0+@rP>{`F0%e! zR{;@gXN(CY4cI0$6X@6wG{3Zl1KXxTec4nLVRN7@C%jR5s0aS^WhwTikdJ3&>8o2l=AH!!|uA{Du;NjF6fz@;iFT)krv z>I?P}$2C1>lbWR{(+aavwoHMMfAV_6(>(Rg;l@)pe4(ce4cR}x>-(1 zz*Y~H(wedAA0OSkUxB)@&dZ-KVc-=L4oYYB!0~}Je0Ea+t2ueZPo{u;&h&u$MK5t% z#XDRvdIsMrMWO4~YRrt}r@|l(|9FT2`OSFq4RboZk-;3pc&tqcLyHpTO*~PK`{(}v zV~r6wy!0uo`*xc+bWeb$Y66$GC!kgDJ$!jooeF8>ltCy1I_cjJwPgi0tAtK4{RlU9MDR zRXEKVl;-tSuj5IM#`03!3So)UG`ul>3#WC(!`(Vbv}HX_|C5^_UHK>6E?A9YGd{yK z`y_l3EJ}rz#N#dnQ5rH}Pg}TjqG&xrrlkKOKg%Q0Wh@;xxbH@;?J^uuuz`PHML?m? z74Hur=D9zHv#-LqZucnCsA>V9&b2}`-&CA@a~Z0c?S;~#b-DpmrQN15*51NQ^KV_h+Rt$Epk^t{M#>9#CKnYPD{JVBLZPBo%X*$02{a7fj znRy&9Fs5{x=?ntIkojJ%q)2h%oXcMy8!3f(;fW=TuYH_NV~ z#H48S{T&bQbR6KltT(JZWI!uzr78c?o49uEJurx#j>Xdo@M5_jo{Q;*`BiT~dKsll zk{8g6o06#8wl(-NK>{B&Jb;DK8hGc}Pk5_R4lAldp(m*kEOaKJYQ|)o{!JEr>3ZxC zbmN3?gcIjCa`0wf1vv68fZ+>u#QkkD*ky#_oJ+E_<;pf*Xr~h|{+2gyuZIRrE1yp{ zU*1iPQ`MkD`VL6<>Z9YWTX?)8ABWE!#X{y|@Ya+^$=Pg&&A*%ArBV!?b@f^7c3WgHD5gBs0{U`^jgF{`Lkq z_D%@%8bUGW@gY>6vIYMpmEnHPaU6Iuid)wI#c6DAG?lnv|I%D?RCp=4=52(~-65#! zV1W+P#BsIIC*+TOij_Wv$lopraVGbPr$IltohO0by?4NQ(s77R4Z&}*bFe>05Tm0W zqeyxfKG`pi&vu4jm>6J~s{lINMDaE#*z#@~Df6=VYB12F05`Ik>(W|neCOB%Qqk9- zk53Rof~R01I}cpjHUrJqykHS;QD{A@!LthyB=KA;p8i*;;b>p z6;L;AGM0SJ#0r&maCMf0VEG}_&kkao=shZdGVNyD0ZHOC01h>}#f`Sfx1Z!BysLkE||_^wHi<EZnFUX*AfPo zT;hnVkqR}<{De13u3^;6JZiFJ8(pzfjehf;10OwKl9(YySdl7&G9RkI@23SsJ?$d{ z9}_{mfQLHP6L~AE-_k+-CfZ#kPhFb@=zt6#71pxH{&;y@^RWhY&)9)S@@?=&jRdNw z{ex(ssf>xE2kW*~lltq@kn*$v>- z>Gb`4XIgD!MlE)W;kDFA)cV?v5|09i;fhr-)ATLNLuJm%`kg!HrsP%Ah_ynz<(-Vd zuXB`Ipc3XQnqi}A9y*M)lKhM2a7`r#jDOAHre#YJoH!G*h8E+Pp9FsI>L&aN>2RS@ z7cGs_@pEMiG1Z9#oiltGTq}gvH+>)>5jmu5^*v%55CzEsZ{XIFN#yN?T$1Xs2)@8v zI$oYdcRv?r%$j6W7(I+RJ`p&?UqI@QXL9Ru;z+sUSNP2Qk#Y{kSpUNjU3=3|eAQ#C<$TJXubw^~G#-A3KV(Zo83lLPFrP zhr-HTZ%BRmD&RkV6MFP-Lx%QZ9AM6X0Rd^UVAc+Dp?L~p3O0c7GAsP4$xqdMBq4K& zBQ)*}gK)bD(*-_i%ul_WB%Tt6H4S_aBNYyEm&>tZktY4@wTE{4y5lU8flsA+k-L!1 zX@y0@3h9HmZ_OgI^tBb)m!S*X+$*L!?fxdkr}<31SEyap|{u_4KZbI9J) zkI3)EPE_zTkJ{*Mr6JkRI0egN#Iiyj>|G>LrDFq5)X2en(>zd#mBNgN!RSc%QSiG3 zmdYH*$!ui7ZxcrjW(2c2O%=B-Tb%5Dc8Kr~so{%{r?EQcAL<8Q$A)UIQflkpOJqHh{CE9Jbb6 zg~-@`*sANmwQOxMt@pBIxvWlDT&j$3bT%TRmGXY5@8oq$oa6Zmc%zp3O5{_OLQkk5 z@6;0^sm2+VUtNdQXHP=yuq2ihWx@-kDcE&z7mjC&(3&7w`tg}AZCt^JmvmKN$Ss)^ zUU&roCc5~-HWOELvmS!2G(L0H!F7&Sj9uypn(}lIyT7kB?Wq=@aEQ{z>ikH{h*^%^1Xb4p*;Q zpk4D6oZ;_^Au2`q?nV~oJPpT@g%N1}o#oaclngbVMw{P_cuV3gCKh($#;?J+dF6Wm z*88pe7mw!`>_Ni8v;#M9oLX!NokAIH|>e(6IvGCcEffTZ8&;27B?9^ zfv}H4Xl*8r{+%aL@&fzp*B4{EUnvbQPN(tA=Vl=C3}>f*K;i$&F=%HxocK`%vxeR< zrnndtzM6nbG`pdEnJ!PGQik`J?a^P&o=sJhmQwjJEBd>25CpSSvE=qDRIiLj97{!2 z4Yo^?Rifitm1tFY2PVDcr(Ibe@$iXktl$%%7ZK-~12xPfVgI+b`hsjC|btz5owc z*O6I|KXEJ8zcOtN?}zBFEC`XX1MQu5beWkf4J?9QW6$?>=3cn-`9bPC8cMfK z_oas>|DtxD_o=Q7bENM(jxpDF;b(7Sbm?D@Ph*|YOVt!l)v@=Rl`GDh=7|C^{*b-> zEzx}`fV2PFVh!uy2eORMPS<22XR1hYzR40tPac+*nxSHXA@WEH?3Pr;-^ot6+pU49 z{)vV4(^4VzkPxnfi5Re;37mqrlBXA%iBzc)G%W4G`OgbjcP|ZBzw|}@7bo$2D)ZBc zh48GzxAB^-m-5a%JA%F2?qJugHmrWoNVE?}!j@mf@YkA;3zDA0UAXd=QhtkG~> zz&nCRw6Rq>p$&T8u)-U`h8St(#R=7Raz{QrASa|-IO+eU5{JlR zvP`~_39Z$*HzXP#r>$h&^8?8D7LiUFgH`iSBU!s2J-@HQu7XeaZtgWS6i&sBp)s^B zc@N#DzJhk|?IYH85|FFGI#%15pV{Cp7=8Z;hXZOL_x3S(cwQUj;;+*8cZ7ICXwJJi zQIdqFu0d6xHVcrDLPapN#8Bj2Cs~E_-x{hoA$pT|JBYUhaC2EqTf^T z-UuIk{&No9_si~q(4;)LwZ9a+1UzuzS3h+A5`!T+iR2gQ zBEJ%}A^nR!x>lKDB;nz7I}1z|cEQM3yRe~6m$yk-pSQ7o3Quo%DV`lR#d*XLZ5_Qi zyQJ;pjlT-qI{KaS)!$59ZJv=2!*&?P_-{EWKR_=ejrz=pq^Eb}(iqm~`BWbWB4&&Y zQZxfjs2ag>=Fu&By%*EgA3*yMUz{H;i!0a7M9qomkbThv^I8mWm9+)l^3*^p&5gL@ znG{|&IEJdFiTLH-S-e_rj(dj{v6Bj*che-8HJVR;Fdy6Qg_^LxeFjLG_Q+%Veh(aNoPjx^3bbvFG8Ir4r3VWq^MCK`IBYAxv_-^ zzdA>P!lTe3I)Ra_KTas%9`zSc>L`hs&OD6q_CcZN{Iu_l7?o>Krw%ieQ02o@u(-P$YTFDz z$5k7IUESfRYBAV~KO$;{HFM6d%7%f(M?q*)Cd?ONo+{Z>_^B`nb&8oowz><)L|$QO zV-Gl(M}wjDXRhCEH3S|Cg$Kp8pm{SLzO1D<=E?k-N}{A+MFld4Zo|I2=kfE~quB7{ z2EI#orL{GZw9vE|b;pvyQQQLZnfH8>(M@>si3f8(#*&#wj)CjXWRQzF4T^s!ApK~E zO&QgAdb%r&tmM#aa}jd93DBR%ft$)&@ZNd~1cnXpm5mlMS~?_U-G`&TM_67t9v{}P zLQnqJFz)lk^s09b_mB0??@ImRqFeHbOVl-R37XAG`EB8*otNWI%smhLMH4_o*AR|s z8&cDHR~oP*l=d!<$0N#&mDpehJKYa)nVR2CbCYDqxAiyh>6v%fqNBnVtx3_9}p zaI5Ptd3#$KgG>x@Rr@e}W1Z^?UJ8VrI70ZWrjfQ#XHKEuCAWN+5w%R8POnPc!MTt5 zh=GoSsi1hZY0m3OKJi)mCglSWy620%J&I!dEl85gV;pSRGdjj^AdyZbbYfbV?~w=MrB*Q`8^GmtBTpUQFY*C-0qe_f$*)_3WV zj2b%kze`}8O)$0Q82-D!N7m#{Cu2go#KqW%yXlbw<*Tz;@A5vLou7(7voAmw93{7W zTVdKsMQmuFfIZBWvnFmL-Z<$>ek(sQ{XRPomTOg$4a_-YyDAh#Ci%jb{u?A?$t#eJ z-bT)BJVDO+OagAjcM#QDji0`3LR*;t?pyy3vPI%A@nl@_Ddz;?#kEeTKY4*OZww@V zEnk!9w_mu2w;zyW4jWi5|0QUjslpan1v)eE9r=}22L2msp!shVDpv5}{+crEIBQKJ zY7fJkN$;T~B7x-DX^?kYEr^td2`245MUE}p$N8O?pp{Ag;pORSRP)zs*u?e;T#y~U zZ#zx4)XV^t0xK{SGbhcBMnt`SkR*){F-A}w)W)nQZV{<))p!%!n-l?-d?}z}?g%Cu z55kuMTbwH~2PKBIaKni$k`|3h{EwVd12Az*d1Us1^yEw(5{;9(o|8bP;x33n5R`?-7A& z#&Wa_2ThsJF!uB}tQg4xo3Ry8AhR8;FNouU@g;FRa1C0c-1?2bi0YL2(2zTY@~(Ga;%GCxRnh=J?{97%UnHB@RWA@WlEK?DKjHR$&aR^8F}WrcuDJ9Y}9`SYzaTMG(uK51Gy3 zB)Kh{Ja}ydg9}v9TRR3Lp7v3XOJY33fLT1rkD73%?>gza8co*hSqXvlF=W}66P(+N zZ{W3C6~8~#$7vbN7p}_~)Sq?fwt-u?;!qO0N6bN`(O!^?mBHUq2JG%Bf9BpDILs<QlY7t7t- zG3fAlJSfotkG)#q)XT4MXlOFEFHOd1KXViuP{qqla=7PZAFOE6g*v-5xLUIl9yRb2 zm%C3mw?hTouW8KRcAy9Dn(l*@sRqPQ>@lajqnwo7nFZxd7O-3HJ9!-&3zcINa8i#W z8cvVo@lsuQ9}`@4QShU?qbB2n%+%839o69HJviJ< z5BX`~bQN=qSYnEAPu+!kqJ?1W!5oVnMIiJo4GO3219Jso%)8nQr1~Jtig`ebvl#z# zUlDaRSP3c{qDkO)mTNebhxRX{v6hD56iYq)KhDlHoT@MU`(`3SLZm`v4q>l#FDj*h z2xX{fR8na)m!Ak3qY_DqCRCzSoW0hOlx9Ve(p(x)(V)@Pv;VK3=XzfJ&vl*G_PMam zy6^As(Q#`)G57YNhpy^qJ>Ot=QNG!B@itlPGNBOh?=Nv>CPF{N!IxZ``ehEO5^Pwi zjL!^VsFcKDRdsh99ls!3|)=eyvAVuZ8Vw@e}F4krjDNZ4o5vf z=TKU39=E3V3P%PN$kyHEdnddxup*0IF6>Z~m=Ip@+LKn65 zn?-nmcN%^-do@u-K!5WK5DdvA{?*&aFa@ zdDqdw-+$1gByYQ)jc2`hYg~c$O22UUHecrJ~REmzY8CZXMV>L=R%_x!_H=eek6zLHMGetz@fg z#?jkj5&yFhyN)i!TZI3C!V#J1eU=zCAJ)Vo*IX{v<}{~fIswh>PPZ%lBWZii>y2&d zEP*Rzu8BQ3Q&bpo0sr~?0B8Dc!ew`!oI)9YIr;@o# zi$`;2HcvT;eGS~N7JoFRV;g$ft%BtzrQotUZT$4GpwIvC#^swTIK7b_NFL<{Wjhew~iC!scBd%iGnmt9R{sLj!nyYTrWNj$UN z8r2yH-D4{PaGMOps*=Nq6&p^H!ZgX1sxdfow$R&R5|71#@7eO5EY=-04(D5T3oNp6 z*y_kqe23=|3!IKS^P;g)nm-y<(8rB_*v&l`I8K^@8}Pri*YMPDWt{ALJ9M_P8GZaW z2fu5bf(!hu1>LEP^L%56Z~;S7o^yoUUM@=dcpdG%?SR(`UWW%uf-&A6k7aiYyNs<9 z!AhkNfBt+2S9eK5^f)cB|K$R6gLc46q0>It+z!8a7l!}S)y7)ZtthHh5e;u&j6)h_ z1plElUafZ=claN|io;T|zg+^_p6sEeJK zM{*{OddTdCDmLpJ3VsKCA!C0DbUoF>;r1qY>u*`C;V;jH@E`2lzuKYfH81hKxS_Dg z*B5x<_#c>HOSWCGCn{S#h3wiq2xy)K4-ATtxlB3=Gf}_|f`7j~+8)bSIN++YNATra z%kj)#6R>!iAD->71IOBA;koUX@a30Zv7d?-JmM!KOOK;yv!geD5;hh67(9he%^8O; zUMNR1h5JwOwp#paVjP#*vVi+-R?QjyF2lp(#kh7uJvKaX5^1jgi5|~!!qaWT@&3*f zoNl#M=+(ZD2CN^Ty|w1};bV7n<(@y9JWm$6hYNG$L?^Uw32=VlX0Wn-HmH{^gdMf_ zpfYd>@vL(t7x#p6b{C&7m@wx3ux%&11M*0GAhZ|;QEX&aXIS>u(`0OJ*#AbrHa!~ zgW5-Aryz^{w3P7>_Zp;eNeP{vn2z6TC*afZak#A57u{N#Rzn!W(Gi*f;hXe#C-t+|XONQ!xO?3YqkGe+B=oL=fuQ--%L7WWbzj#}5T9@33i; z-MtKBMF@J>!l7g68+!lWOgwFNx1Ic-9d?28`)q#&yhV%O7~!vX_3>dP zF(i%M1s09#!2Hk=Z1dI!L%J(Y7CMM0PqRQuI2~;|^bGB&3Ps(M(ouMI2wL5ti>XNh zex-L1NeeoaXNN0JKkyH$e>x?w5HQYCxXmHwBb@i%({?sBamXRj8g;!9I-!v^jz2gA z2b3fHNbWdS*W|;+2EXOXg7i>|n;Ry|J*X}mW4EW ze{h?yV>^o;=RbixcC#Q>V>cFU{$?9xP;Iv{=Lsh}Js16WIteeI=Z(k+Li7;ZTD)?Wlp=K2#No_K%pl(&;cQLXR4@(7v5LLk0vd~-kAzGpj=`0*xt%)gn2486UON4z)c z?WjcC&bXkQ9e`E}ezizhV;nA}i+vtijQjB2vmBDFazI>OjLkdJaFO+Fd`|GM+-@6!@5%f@eGBWjH3w#+4eP(7*THMR zcY!Cgyby9I>lWh$wirhUU3VMIywHSzJ;<$7$i6Xo1yUVcbrlPen3dV>?H594(+Gw`wB9m2k+5_yi?h~~#_#=B2Q z;+(-7C_f0m+i(O7O!|#~HOg?*AdIW?+k^JB^dYB!yC@}865or|u=B}KN4}a)_`cdR zj^EqPU8l#mPizGA2;Ac$nO}Hu^+-H8Uj>Jb7P?_OmAD&9x4E{|bfg;B!>z*CxM_as zC{I=wNbPw{j|um(O*Qyp-(I|;a0AxAT7@6{4!|<`R#=x`08iik2cx!C!obH3cw8@3@9?o{hxun+2xg$O%|W^D&Omyn#oIYr=}ICiuvoZj{k91^aKj z%c-vMM?%O3B?`T^f3$z1z++d?7Ka}EI`slpJh21kw46g5muI1<;j55ERU3*s`5N6^ z%Hj3x+Nghh6<7A|EmuCc2haQGk9W?UjH{xRaK#E$yeL2uYvk&Z%!x*%OL+xZVp4&X z|6auw$%nAZYIRIcbfI^a#puY_47>k!sc_bPL%HJI7JUBH6FlPnYkcF*MVO*{4mOq- zgKw)2UeFkY9kla=eNCn?JA@#e^-ThU?IVhw{1}~e9E(SuOhy{FN8#?8gZRAKT9mnU zKDv2HU;*sjkJLWMV~kJBPN%o8i)Jlv7L?i5y3Cp|#zv=t|^qbVj2J{iqbW>YVnYH%X)L*$)CA zY-|G_ns-l#oz|lbl|)_v85ahq2S09BjMvJgyj(22Bo!LHc$h ze3?2CE%p5meWQ<&Pn&B(*w~9=?EOFdJ2bNRp^pxg!n8ea{Y@jF@9|hMo1Zq z>-AvF4|DwQo<3Ic78r#d_Gs;N9kfWN5Ebk9Aj@5=@y%mfFwZQ|@a!Zsve+57YyQBM ze5;_>U&oqH;;`ek{{*JZQT%+3Bu3?D(D>q)NZP>wpKsL1ok4;4apG(=DyoH3zj==P z-K&q+2z~*{J3+X&e;FYR8J!Hhh7 zM&Of7-}wO5-!De0p3O);?m9ZquoXAY;IYm1ModhOV6?LWulrbyou5CmTY2iR-L(nN z?EI&`#x0L>@V-ZDaK`YhNO9^+yyadMt}Zsk!-PKafrgF+I=L{b2(;CKM^U1oII_GNJ#xxJ=S;4m ze||M6j429?sSDUT@t2eHL#1NQ5`hSJC65 zT9kL>7aHMLi1t)!;_qv0u=DzSTzR7u|BF0|WvUC&>f!QupN13epA!aown=d8@JD(p>Q{>*8iz< z{I4jtr;Up^e1^OB@c`$jKLJ1d3Rvb-n9y_l8>%~{NWhVPnDDFt2S#_{T} zX!-<&?6`~?-bQBl%6D2N`gXbpiuxH72 z-0UNP>jxW=k!C-tZU{%ukL*G(;PW^yNYwZI%jOJ(B8ikTQi zR$$9NT5xK;1B|_<4Qh$s&@wF>{H$E)fLJt^Sbyk%EoKxv1@D6YA5E;&xf!R$)Z)-} zC-AbbOR#3bGi*B$hTko1LJ<)UFrNMhOI~fpdt>Tg*26IgWzH{zW!B8NOb%6dR44 zik}%B!R;|6m^{hFYonv^p3Q6U+o&`=Q@Io*r3#?sa~injr{WI<%kcTZZ)obLOgJEO z0Cb*LKxl;|X*;JwqEhw9+x{>-`nEMr-aY|y60ImrB!io#^rE-LPto<~CK#v>ZgV%r zkCX18)}pKEtIrv9GIs?2B%CQ#jex!nD?}IHsp9wlO~4wG3vj!@k}MC(;$9aSBAvA| zICMk?dLg$L&G^q0Ew%1IuUqSo&w+>N;ItCErp@hkp}QY*8`CUt>s%YW@1?-xy*?4Y ziFLxKjcxFDe|Ms6x{Mr(i6k@3-lB?e%6KqVVD1!8was-OWgAo<#HnxFj2cD@xkrKD z8!mje-q~S*&01vejLE6+moIXM501DrVOPjYsZuK6>az6}@QzMb4FqJl@yhUY&%}A%q2=|7$2%TcyLLNa6 zJx+GuzVGlxkx}E(C_`CfpR9rRzrByb+IJ$A6=C>5Vt|m5n1`)Wy|9tw20ZC`F8;4@ znb3{yhkdL3aeeb^ynpXweE-RP?3|#67JS!4Q`VcKBe&C0=v3i;1*cHC$vHHBVmUgf zco==UE=jHl{o8LdXONXw814;ez}f;A>iaY~{M^3~d8as_t7iqzV8mFw(P#<2tv3;O zoS%X%T4vz{rNt=tjw_1!=Y(uHDR{F=9ZJO`AS0~_DJ0)TAsVUZlC>PVeKru~3Av@> zz-n~e=m>iGT3^Tjo#vJZdae5FgWPiH#$ls6@Jplb*t%gE_IS4&j|-~6zF)dIov%{J z=k`HvWqmvscvgcmxcR|u{}m6;Dfgm~aomkBMGTQKkxtF1CamM17Ltjv!-+Z%Lhnh*`mhf zibQ(xPWpCGirqP*3J*ufkgpGq!69`^nmWdg6@;p=VzWTD+^dI{CNF2%9g1Y?hxPdR zyUk!O+lCwW6qBGOU2tetH&yP=W$v3U(~YtVaQrJB2&fu@jpvPmAD3?YEQamZ#0Qp_XhbW>%qn9ov^j;46nW|nwlwJW_ngJEdAR#s{AF8 z{&R|^fg%~I_d0+(n)nVzRgEP9mY3;_pd>aux{LZo)xz8-7R<$;XKu?YV1<1g&C^e$ z)(SEVKNQpFCpFpXLw@ip`Z;o}Kgtivzk!<13HVk(DwAvA`E9Qp**c{zTJ>Fzorw4j ze~(v!>T)|$5$KO^o;%MDdJbo1pM7{gL6@wOR%KoOouY9WHt5pU<1zXwQamhr#vXV%4^-npaOM$Gj zD~#FPa$u-iAI!cOaKpe${4k1Xec}t4o%fJ;38N zS5ueyJJ`eNH`(43M96aN6$SRKr-#2fk}ZY;XWMru@mZ)X)-l%;k8#(dpOhZ(N-8~~ z_xh)+c9!ph@$#xnqcWBKx~t04rdiNdnJEyWp~q$$1T)7t3wr0TE&sjZ8ko4wW#iu_ zL$kLg4PQTle~|ePrngL{+6J?kb*vln+UUtvJnL|9&~U)&@iKFdlY>m(ptOHoCPyojr+PNdG(ELceTNBTHfpSa$nQ+WT-k z)G7t=<^ebG%{Pv$Y0G;)GFFv8SUsOjR6ogl6+Kw@`c>5Ii58Rhe8u1Qu7*orhSOV4 z>sZW=)9_xS7VAh9{OrI@sAH9<{ULD2u-FX5>o0oy{ zr(jm2yNd72k!Me)#KAGmmHg--@nCRUhZ^KtGxPZM%)DQo{jI5kkfuZyxpoqhjVVTc zM^acu&L=wBMV3vOugG4dD=^WgkKFiii)q_`!`S?pL&S!UpR=05Dkc>ItYhZ{X5ljg zXP-WU2M&1Bq4Jfyig7<)A+{j0r+Z+0<4pAYksC~YCc###C$ZLJC)kCy6c}?F(aTDb z{Er--&2@RptmgDH-6e&1PT+3pykU^G_gK@=%n#_$rg+h^j9jKNAk9W-ny?$k+rY_4 zj8~Okr_3UoM!pH7!a|BSaF<|4xSe&yu46NrB`19o>p?CcU*?=vBjEh6mNySJO%;neiFV`cz4yUk6Yb#YXmU(JFSRIsj*9 zP2i6=?m|0dQ|a`=Ll7_Zo*ue9kN)VQY%-OhC+l)a+R59r-N=DCewjx`PEjUDA`4+h z;UV&Sq9WNBpH0SyQb|g*E`RG|28(RTASc?-L(u;5B)qnfCivFVh&XA+3{{!evGeSP zTscdAK8=JeIZt~YEo1^;6?VxTq*Wy$?9TjBBHDTo_SsvIdg)9sA07dt@oCT+^9xML zb?m)64FWu^Nl?mKBJpo79CDdW{{7X5lPhnDMg+FQxu#gQ)7XLSCd>J--6?cK;SgHO zk05x2Dvdod4HEf-B=ACL~M0V`;`-#Mp$#N5>8)!I^I4*Lzv|jTP(ck-p`8mAyS2;Q~4~|M@Ey5hcWDn zemzWGa9Lzv*8!)A66k-_fPWuvfP(W;l4JK7B-dmSdjlo7Gsl-a9p=q!$~@p$NF_Kv z@F%Xr6^X{9G~T*80>VVb%<4}n`Bc_VxjSuCJxR>=KfA=-9=GEiq9MGp>TG=aYB@~V zYRG(QQlM7x5jdph6NCE?U}|gy2&Xq$r58+RU!FnQ7G}eS1vc#ChXpul-9moV_6*Sr z)mLzO!V_cJ*>Z9%x3xMfYp^=uHf`OdiBqFRF!go<^IiQVF8(Ji`E3H z?{$Q}dfg#%oHiQPRS{U*SdiH@N9&6?r&l8b|pA;XC%$EQew(_@@&PXM%)WW zneDMHEHK%VZ2H3fEI7{r1C#mbclI?br?RDq>XI~l`#JBWP zQNZgk-s`|L82Ria2{?L^C=MUwgTmA3jbrDCYS#x`nJ3B54Id*4ojoYBG8)EKG%X^Y zS{D4f#sb=S>?|&HZ>GCz%W33yYZ{WB3vQ-Cbgt`Da6QoiqT}HtckyVbeeZ>5x_9vt z*B+(y5gu^M#2Cs`R7r>e5BDD$+MalrMEdUNve^^P&^c#HMMm%6K>GDeS|aU0TXU~r zo14lom}p2|IUS?v=83R4+nS7+(+wL|x{%O_4`iTVHO+f`1U9d1f(3Uk(R#B5)WzL} zTvI$r?`%H={%(~leNPQuVlU+T40XuUd8T+yuPiyjZ-Z@}HgvwMB#GP_M>2lr@BwM1 z@MQW1knVH<1^L6sdEhf$ykP_ScK)O&_t$+mIbD{NL z5*?V&D-rkW>7>AJ5t;wxElm_=nsKl6$*eJ1qC2xZs}5!C2Y=hI0NFQTXw7sO8M7A5 zk-q(i@Mim+e{3KwG?yHjv4!r|b7I%ku9Hm%-jZaeJUTpT4BIsjN6)?bfPULlQ~Riy zq^Ir(J(0a1Jylys<2@&0&OM(bOZSs!ugloTBS!&qH> zAFS(ZgTP;h;Y0maqWth1?HTRBM>jjLJb7bQ(_ul<`)labo{J>BX@CsX8o(`+hh1U*`MTGEyN0% zi|L$OBD(Bf7G3i8JejTV1k6XaVa1`*%Jus2jiB;&lDaRyrcHSS?@1>=x%w2KOB?7`O~(H?JdACb9mqmFbJ*Fj zKdDn?0{=iwO6d2W!jfXf5|Ms7n2Y35?2}MhD*mS=O<&42@fTj+ql;D-vW82A z?9>Sl=IgA^)+;|l%fd$SXRWQo0c(`S%9CTn9*cT;3)>O=xh7-Uvo4y^pO4tAYq89z zWE>4L7*353I?xvrwouikLVEVid0Hvl^>bUwM9Q6$s7~x$cC6c+)mA1^_V^%8eyh!9 zK9yrF+!C^H(_!kS_m#ewUqzqjonuRW#IWukJNWipTi6DTg>3jI8*?A7-Zml*D_Un}|y-6=~6;QmTJ#I(6+grJnnK37W_#w&IH#{h=Yl>_+Rb=?A@- zmjPn&i~nPzyqf5m-YmBIgFM@A{2#kn`I9F0yrZi;zf!f24IRO5^PMbvzKXaf<`=tn-J6Lo=~Zx1iXrrPYb(kp6s{}9zF?q#jRH;cder-<{D^;yaO-E2pe z8oOib04K($vXSvGXu<*E+}2u4-I_Joql7Q~`%EMDdyxmzk!+;H5(X%%xliRxqv^yo zH7xJ@Nfs$vOgZAkf6LaU+pVLR$M)@P`Q;=w*>ygdE0Ig9GYaVMao%iF+(Onhy_^MH zG7@i3-7X%@R*2^ZHqc2Iz1Yb!*BFkKVWV^z-DS|imnTnw!PedM*q@pFy<`i9ms&8l zfcI?MgR}JQqhkK7kxG>)_5ypBTgIAxNsGOfzNB-e{-Ci5OWE>!he-XVL=c}ENo&Rn zP-2$?_Hm$)ea5MDnDa1ZZgGqT4Btrwyg5;O9K|esOX!K;#%xmgBbqaR4h@;VgW6nA zXW4dDR4MBiH9vBY8Q&862tzKz@%$O0>iQl!bLT1+(=EZaPwA!J@3Z)G+(>SW>3_7l zF%i2ccj5BvCid(-X2I=4#cO;8-^*=apykdyEpEZCgR<1Ga}53dHiF5u8?pGB1?-;p zARpH}24W1Y=$A3q=}9ven({^o(!w(s_f(HXzkkEa21(QG_q*ZT?gHlfRD)jqJ&)bV zTF+`*>S*0mJ+`KJJl(&@h~3bUWzIg!=-(OVsr!grW|*PFF6@nABO1@J7Y9n1y8T$% zW;&m`&$}z~db1MlD1GN^ZmnV3T2jn-=sDUd+z(G%bg{|>6>RleEpcq_Ga9h+Fq^Hk zi#@V@$qtkb6EC$`CO+G*M%EXPV>vsI(}aX`u&L#bNcW*e)f4kXHlg}1Ti1P_{jFS2 z8?$}sze-(t;D;Uyo{>h~5*p~_AZ0fCtPOKqU%-r;4)D#dC0S+jIhywKE!-SFoBlN_ zp@DS|;hAv1?4DkaoxOVbYqDpUoAY+oP?U&QU3DQf86CWjtex1?z)D>F$4QLhLfGZt zBP_pq4;$SZD?XJlS$z0yHh|BWAeQ6ziPGAUpm5MnVq~o>3On(jGc1{B(7#MlPCJjvoM_{D%bJpc}e_* zp`R!&8A;9=7?Gdr3gKOip7^DdmAL9bir8oMKafCgLF)Y!k?JbSiv2U#?wfK{e&IFt z?rJAH5|qbOQgm^6iWytPG--3dd{KXFAsZ|@z$E+9>B(DI{B*=C7H_5`{gQiz9qn#SOYDjbkx)cd!_dlK7&>MD|U3GRb@nBS-Q9L|Flm04HF9H;I8_VF1#_cDumkW_xE;tSqj zX)W~qzCwkfNcwN5K649NNA;GzXTMLDuu`AVu+VKMy>@RLT=Y8x$4yk3T+J+~UpGt~ zcQ%7QQ8#5SF1oZ?{F7gSV|f{cb<{rDnxFGmok|gijm)d2Hg2{Mp)#5r zBM+$wNoGD7;q216MyjQNAlrHl-0M_f9$(&rn#FxMQ;-S4x3(d#ORk`HB!hb28i2m( zKcIJF1Y5E2C2a`1%amrUXFttP;JEHg;k-G{M+_edgXTtbnB{Zi8nB&hw^c9 z9$4G$CA=}vC+KP~rH5wH@y<^~_Ix~)KeGfrQ4Y?i9fasp6XD;^2coi4E4pTB57^3F zy=XMEleqg%x<1xET_sBd$H2Rhl%Gf#uO{aXt0 zGb2DHJD)0CoWQ!G?8pecLC|<8K^{FS;U^rA#$Ve` zeD)g{k`icJD*+KFAJCPD{)4pO75wnwx%9t89~jX7Xp^2P&HpZ$N>*HXN6!SeFx#J7 z=;!-BI8SvV)$W-DgE7CMHtI4YFXzrXX(0}dFZzHCaik$0Q_lEWdo8K_*TYrTX?NTC9zJU;~)dkXi z!Q_U24H$0_c3&eD?Y9or5jW8Q=^pnOlG>w0Yb$f8A>@k|1vGL^H7d-&a4PL-@8m11 zH0Z=YWBzT=5QwpS$5&O#2whY|S?Wb&a4Mk?#7!mpFKLs2ll8cN_%*z!s1|x1^oi(j z4y<1E2FCSOlgbCu_6|w~IYA#QjGmdDy?2Jl%hfTr$ZbWv5=lx9Nqjjw|6mtd57-PqQH_ zV>dj0*Fc+6t;xet!@zHj8=0$B4V^dd;x}9J=`zhRVD!O|By{Q1gyc_PQm)5*egBC5 zJ+>!v`W@kfVGXap_9GuCna@Y+NU>dtccCO=DnIN+3hg|6mfZZYorsS`3W4hh<3WguXr|{8$$0!a5Q?t+x!fl*HR!@3Kj*n6x z9jPM;uW*&GwjB$96;zn-`xdlzLmeEMmxFcky!l>v37{U^`FBr7(NLRjSpH-^Ran+U zH9xfDM&W#Y6lMk9mZxz28+}^(U^pwdaU4xGNQ3=r-_ljrlEHDnmXrm(6nXTj)A)va z_}AJD`uwW!H@^B4t(Qy&{h`Cjma&9fv|Wm~_yqB}|Bm6X#tv8*`<{rMdq}|rN0RXH zG)+z1N56D!<6mjtr^*j;GJ2aiigDuVYd2c@V*g_KQ zM7hXwX#ll2olasLx6l<%Cd7W3CxnDC(bm)Elz3F(;K5|NM9_)M_PwSNTkWY+gA|+9 zz6-pcuEpXA7y8*zmKsVn=fo*G=3|A*Pc-l`03Jh`2hRq||_@}2T2>Uij;@3)GM znUWs{KyIP)qyJNi6-&sLKZv9o~OqSynsL;UeBE#fqqW8QJzRP$;b%>Fiv zUU<8lzWy-Ce?3wR=iSO+Rdy*X3TZ_#a@XO?H(y?T)j7J!_7FbxXdm3zFd1Yc;^E!F z4!aFM@AAJByy(=OxffO&kD-6|Rr=2|LXOzR|29-f;dko!vUyjxNIZ7w2udCc-PDKk%FY?RU z3G*yY;CIBV0=LL>{H``J8ng3*sO{`I_HL|4RiTF&dD!!Y{%8G|F1Qp=FBB<~^)XY) zC-iC4%diBxQ$JJ3ZSMF%mM<@{HkG(WTav}c2Wa`)3Yz~!iT$yYV~q!fvsl>x zGPS0Re)82I+S`s&uaWuSl&DRuZmUtXn`?;6sT$~Y7@7I6i%(Df`XaEYhJbp}-mQVZxq{Y-3etrh0_*Mp9Q0`FY2 zoS(Y8jT`(c%mOtUbRe>yzjJ&X8|!$xD)PV;usXW}?=am*qxRgPj$6jEu5syP_ty8+ zUGV7GE;~-0`}KJBeQDHNa)gM=ALg}7Hc^tXj^V2Zc$1J_ko8T0MJCll%1m>X@#GDz zn_hw2>OPBR{#wb)D?9?%ckkix8D}VUJx4tDe;Fu!pjzXd*-kj zZUl?a*$biawybMpI-Q%S#HNi)r!HHp$Pix{rrDT89YP%06f4Xt)~v$M3~pf67zfa= zy$`Dlzrn-Aq15uH9W~{9_)t?1g9O?=3yxBP+sGWm|Io!o;zeSCcQL%i!x0bh}C2%CDG zsqQl=_W9{?_^bbjZ2xnTn4ae88>xNBXptsg_Xm;BQet`_#D;tb9Ya<6&hiV92wSgw z&fo0TB`Yd@Nb&ba{L1CswD0al{H?@~R-M}{%Bh_N7rPXx`JHD~=ahr!=b7fb%8A>c zBwG(-#(xlfH&CW3W8>*T`%bVbSOd95Zur|ZY0|{qhXb5BQ&tr8IHb%b9Mpo*qrY(N zs~)3CTQ-s_O@}DCpvGU(IR)~zrl8f^4Dx*$2*Cg@0ZNi?_gJOBFdQo6!#C!XCk1AmM9OaEJ$ zOJl4)fZTL%!2PQHcojk9|88kB%sGvJ)aQZTt?3%*Mf^vWg`_G8RkUOG0N=8mKEm39<8TOiC75}TRK zt7QJm!ApEYOc$+*_oEiYvuK5hGs&O(lFRrN$!%EL#aGGv1dopGaPiU@y4G4o6kjYy zr?g+@%d~8WYM?FMB<$&eGap0wztuo|P9V38laN?aNB;X>PgX^5qYeW1aFVi+C*H+F z(0yfCpEpeOLsLWeOv>!SrK1o%e-c$$c$FWi<3(L0SJReFTY0B>E9rH6CwBYcURvh& znK#6T$=wZaiBsQNV!k($ro^ez<(p#oU6G7lsQ(Nz(`sQ^(F)@A?hJ`d)u$;{%kldJ zE4q2*dXZMNDeIQjAgh%okzB{CkSvu6;bCXcQN23q@W+dVle=u7eF^PcGmOekJ9xVEarLL7c+~dNh!OzqfY5UVl3>rpy8=MifJJ&t7Q! zQ-;&T4Wg0JS)5MmUK%_~nPeHpGPjL7ID1$gj;S`HLxMth*P>)PuJjuU4l9ATNSe%D zd>DFW=FuX47IS=O$Ksl5Xw8MMRJpg9WX+3beQh#qO@J{k+M)!{a&vgmqM2-!N*hQSOO$Mdgdj^wkAhrrT}I`De>LRcAEi-#_(rJd8?(+v;iF(dg= z&{T%-K;R`vTJ;X1$#$({F9$-2 zA~fU5_FCRlrwHyHoA{ zI8`1cC(kyL0r52auk{t&@*Ko%8Or3(1Znc;-fr4)Ql9m9t%ZHpzQVT;#&qPb99FXU z3N^K{BDvGVkTriJX{p*ym$h0Ehws7+r?!<3zFsOyDq z(|D>Hg*{;}!8^j0qz~4CQT{93K7TZ(5*1ZPp8As1GpE5qtA{pM*^}HuktE#WI(Cvy z6K$CL7;m040&R>q&QE?%=&6z=Z0eC*`uFoCex80S-V!pD)-Uyjt)m;MujWYhC+!j# zPaY@ubrivLPCb9);!rZ{#}6>sXUD4QHq!FJ%}mAJ39h=i)3INS1#LhRIzk@eIZmIr zTU*s(O6dl=Pu3Hs{+LFiQbs}QW*uf(_(_yhFbcn(J%?4~g|hW}DXb{Aj34lC0P7X2 zNG@6@N@&bS+Hq3Q`)C=Hkkx{K9T$P?aD``1!4SJ83syN9l1;5De34crJ-1Spe9r!j z=Wjhj$I5;Nr6_gsXRI$8o?pcOCsWCf;0kF?Ydn8cxe*@fjwe>9D5dshWYg|YNZF$S zd9IUS)6|9Zmi{Z6X=}yKN5*2M-6qU9FP**}b05}QnUaMMoyp#x&Z4?eBC_8flA>)_i1kWBrkHLZo8QiWXyeV$YOsK;&rTvmUv2CgPp`8dxTj^Gu;3tY zrgDOY*UrCsmtpU})z!Y}Nw~e0FM$e~MlADx7&;F}s@^z^TN&A_ga}2Ufw<>=FG8q< z#*d;MDq5sPvR8IlS(%YYA-V7ORfHl`+M<*q(v${;`klYvo_o$c@BNPF`8=0hIWx^? zbF&jCaie%8im)*;i46dRFDRgJ;V0;1w>9>wI)x*y=fXL`7ifw4O~%p9jd@gcgU$NW z!F)4BvHDtHv?gs4-V`?*g6^cFamOf3{66E!MiH)remF+|DUiIKbBH}DK$Tg;O#PE+ zM)#}+Hk*vV-M>Geb()sQ#$+cov`#WHK=pWB<2i~B3&*`1DOMg)#rvM-o7_!NqFz%= z!B2M~9=MZ-Pj|24+9fXKR^8ddb(rWt0^c^^|5^+2ja8@NEIdL-^&4=n!B;W<30P6w5n~B7;pm*z}?q z-d_J27y4^+|8*`wk|AX<|5ZBLExG}x_sqZnN9|P{o+f!_7{Q!@B zlIJ?F?8Qv-1>AlSW7kikNU&x$V=i%%ci_wp#!FHb9}9Yqk1zR*U;Nt7mF~>p&a=zl zj-~&$*AqbB_#4j03V&uW@vgR z{H?sl%$h2}SF>;6H#0i$ZOuD~XZDrnA;aKSox6Cqu@T;zP=ezd)v(I#8=1b}XX7UD8`CePaVai(J% z`lVirnm-w#8~SqCbuQU9+s#a`sO7}|YQx2` z^;pA+4>?afV3z+Uf_APPW2WUSK#66E_|q{N?&^JZI68b8X3lOhnV%|*k7;<|874cp z1CQ!(>y$J%z*&-OtZi_K8<#B z0%!TZ94Y=kjO*qHFl(kY=h3AJ!c#6p;&2XWwr{37+&(5oOqrziTA~Zj)M>N!Q`q1W zL0^|Xu5~rO!3-Z!MOBR*z~}Q5nr}wY&9kRLQs8;AG-M%p;D3%6z9tv$+w0Q8C%%mN z!Ovt|b~R`()&;#mcT`*V6E6RjgYnl5%!0=m#3a28qzczEYkmcDR5q=m8p~e7z8-N_ zH?o@Ef1d?ia>7i0=td-8GR*cId?5Bz1R~7);DBc}EIHu?A%A2+;N%%*bV&|55MPhZ zkJb@gb358~@iKaMRhf{g!|-2L7(^KV<*nMf3jH|hgQh|;DeTBVeN!9I+m`J}Cr=cN z{+kUttsP()Qj4NPt}`F?-ZNc+F32iE4aeHdMzuGE5UFByz4h~`tCBYS_K*T=ZDDGp zo=wgbIPmtj)u63@i{Y6?1F_UT$7J#qptHf^C?chm`IhjCXI^|7-F_p99!l;eoEt!% zANs-@y(Ne|l4MX~_&(@87Q!>aE-0z1koY!7@D@&=19km9CZ{jUK(bsO>6ojCUccg} z?Fv^(cC|P;Dz1k{{ZeXWH8&x>{S3;St^(^_lxf$OL?orHf)1`lP+t2MB73T8Ctkkh zRV>NnJXo@T?8v=VTTaDDf6G6j)VH22cx4N9USG&^)DJJi0)=^FLwxjzcqknT zDWmI+&Cu=iVswk>L6hV-Gk%00ZJ^?e$n$F)-nMG8@#22tfdiY6{W?wF)g!7@zhf(% z>beS#ObN_67G7)bvX(qG)TQRq63B0zFgt7LsC5mhhZ~tf%s`?qqY>$f6_gFwbG!v^ zYAuFjr@7?mrVi%69}_%WkcfON{WwjtM36k2D>G0LPf}%0kSE)6n9{D-VE80~@%kl& zHZvNWG|xbu&H_pBO-X^zCC^DVlLE(kv}m4c3nVY25Y62K!RL~o=e;QS>HXmOPPno8 zV*-#+TLB#TEJiS1s#eh9E|}WnfG-k(BkAg#kWVYYrg%9R@MXYgDTjCF_XTFuVzz>QQLSa_loY4JixrH}|Y zar-NgAp$PG904RO|7a@AzdvN$_~bs>6YmDM^9DFiD@Ev!s#KoX*+r1`pc$krL|A>t5U1~<2mIdD zf-DpFqv=uK$c{<&{n6h8Raewut|En-Li1A zya8%%XF~sxIMR179d2i~K~Sg>b^GB9T+I$xt2~3OdEf)wDlL#Wz)uf-C?orYS5cMK zQ=~RN3RIHoVBy|VKf zQR#$wI5lbvH@{}Uki9yrnm> zVi@{|&Y?dG&%^C3dk*Ku6xkT@3Z;MUA|ICCAsW6KU|Nz5Zf5hTpOplu3OmZm-Dt0~g2!jq8N} z<~JG;?@*_EtP6bm_~@o-;V`9^$40y-lD(T|P{nKaL2Z5n>^u1nIxn4rD=(a2u}Ly` znCYN_OOddcKM|^>3y{T8YtWsa$lx(mSbwz(e!LT>lW_op=?lQ5wE>#{mH-c}1m2zI>U6Yjsm2>+@78N3_o2ObA3xWx#?g4*l{?5JoR{a&#R^82L9&@adKY zH0+uOU$)5*-$*68MB9qS%qiqaFHJ`&Dl5_S%Wp{FrW}xPilZ=ZD_Nktyo?NlGY$&S z+201sG%}eBdxzkOOAj)cnhwk52O!F0K0Imq!W8A~;T>yCBVozoVE#>w25vcnLfwrR zn@M4MaMoc)oflT?{>6Yd88?lFp4(6Hpc`%6?Zud>{f1q^vE=FrUF6_XOj3UofRMjA zlsyxt2kah0-}kw2KQ<6l$2y_sg9wCQj(`b?9$Ly1 z;ohz1kW)QUyG+FZH8U~1L8rs;gEN;(gh^wKA10QBpkc1NN99EDlfVXN8;LnV<|!_ z)Du8+ARG*uvzaf`9pGGLFL7S43v)wMATD_Z)p?Q+ET9oh<)23*C*PpKl^dvh(R`XQ zTMEUmn}@3A>}5xT{_F; zp3fzNTMS^V)(=M|DPWgNQP@08lhHN)%W&d^8624fdt6q-@r~-#bH_$ho2-NI(igl) z7ky-}R7Gy5rXu|yc_dQI^6V{&nYnVRU=n>9j%0K)p{*|PJo_Ms_&1_OhR>ka|2CWo z`pyJf>7f&I*=smd$NttCs5`dAc=kW;K}E(SC|8S1KpjplNroI*}t2xPN zYS>Vf^T zLWduS+?U5bqwB!eEtNcndZKQ67mmKEGRfW@0?wReNb0|7%*@gnlr={lHS9Tv0*BPV zRA3a0C$`f?imZQ8N(t;rdeEA4?;-prKmB^bky9n_0lGiup|xTs;EJLRd^=YFu`NRI zL2-ilw&Eb{*sMuq2lmic`>Oz+T!Slr`RS{UD5iE?ooNk;=0!I@N49b)9wDIFlMJTc_e0_bWiVdJ#{?x+!sF6vW~io- z&xKNwJi3~uAIyTk(VMW?=QgIvDxI@O(;2Srb^vo(7dmEomQJ$!LKX45P`CUB>TXU1 z<%Wgucj^WFx1$ZkjIYQ2_io|3C>LJx`G+hMP6V|46NoXW)*0V<4c%^QId=1+c`J6l zyq3V{O-_BcrMx%pbh)A{FT2tWuCCq-MNI$>U7f`Az&e~fWr9{N{g#NreK|`_@LBfw+;5$tK2K$SV>t8kadt`)|_W40x z#}FK|d&@leZRufI73W|}34WK5i4(JNq0yT$=C~s&VE#RMv?2jHLS6}WXE_e1q zm|h*n^43jIXiTAM+_h9eF$!7EZh_U`$KYm91zkl8d68y2A^XiYP+pYEWU7oa74MF5 zyw*xH4U@`rYgj2`|9d;kh;gn>t2RLq89DHP8G%_rVpMzAAI8b`G1K+$C5ez4ApeFk zU`^F|rcdWtZP(LoxFNZYW^m%r!($w zNoq?hIO=_fqdZ{3)4ci)1$rcdkLhKQe-aHjA)UliNR$TM{tbM!pFlFz65efj3j4@4 z@R_ASN1EiWv*iJD!P ztgEvBO!w!PVZSq{=-i0wjIF65b3FPx$>eV(nVt6dzC#_p#+YIezcikMk_~by3TI|X ztimgotJMwd)S{vKjx=y{FT7pDdU{<>lBq>%wECQ`>2loP6CPdn0;eE6Yk-cOHH4T?DVA-Z97SXrRSiVJQ7$3pnasLAgaC z`04&kNN!q+t_LKeTPF+9%`r{bcYH6bOnOm!%jp98c19l86bse0nnlrtX7=Ql?{84g zQp4_5W7r=VQ~tR5Xy@~49x>9Twj+-8hvQnbN{!7eXy1=*wx#1HK6Cs>-yi8t+yd=p z>$>SB)9cD7ufT_pH*Bw;if;0yVE)XD&|7p0jk2u8f0>u*U)K!Gx#@z_P!D>4$qIk* z4u{k$d}PnxXf(2JFW%RrhEqqQ@h9mPZ1lK+tGLy*Hv=pyW zRitHci>RcMGS-bOCpWLyQ3DeL=0i~|+{%{4EuwKq>&!y%8mvU|&%?0b(PV1TA5P6% zKA}GXKhPngV60@Gi5J=VpuBraX`qA~{j;>2xJJCCd(l^VC(wi|@X{6=rW$ao&&Z7j#f){k9klt&4jLJ~~dcEHnOVLs6aA77lk&nS&l?SNBOp&^a zrV}8@+lsV*AHi8)k|3^06n~xfjmZ2wk6n)(psCG1bR@(K2NitB<+V@o+*LQ3+J#RT z@mC5+LHaN)9h{9Dg|FkJj%93J8B3?KGT^`yWqd@k9rmdQQ;9`D{Ky3epCO3X?)wCw zsYR_b{ozYpIL?3b0MOYsQ3j0qK5x2mZklU?$*#8vwIPELuD37`mfN_ItM=L)-sg8P_+J7o5b*}t_wHE1^){}qe}m_p z4y8fhPfyuSM{|BY!=bxHxz~5bVzZfXpy@q9{u<1nU!ub3h412ZlG656cy1b*3cEpO z+?(J`TPMxv9XZ1sej-jEu9Ko3yW?QjBNN`Dx>P20>IZt3`wWNooMU887~#{;I!V)q zAle=I7GBzw(D$q_z_w4Eu8@8Via+Yf%D+-ne48gey!|{LP7lNZvn;Ts`B}KJy$RVbieBXX&2c{5~4I-{DyLfmgoD{%hr z6ZrM?da{D$Zun>2ql!8a)Z6YmeG^hgx0&yT&Or`c^`9C=cI}+JqZQPfzluIu=tUKH zx*)YM5HlVcIL=BMcQtV_^LQV6u(%BZwTy6{MI1(gRw(67Ct{wSrV^$(bSLi_`gSfB z8COc;O)Uy&RUnt{{#*%gEE8N)$3d#R9|Rum5K2Gn4rb3C&8Thd_1HE{6AH|4YMj!J%u{p`bmxt~V_f&qG6`ut=^xCRfB=bWnb0jYbDSjP9AGO&m2JNdRJEYBVZD=ul zs$B+yY@g;aAdZ=dPLw^k2fL6Jux zmNtD35m;fiAh&MA8W?UD?-iT zqd+yT^}S1T-el0l@RFWiwx0f7BtehK&V;_7v*3)sJ6_4HMhh10MepmRaEhH7W@CeC z7W;jj-4&RJe@ejhUN%SS-yL*Lv#0iCy)rtaqDYf_j-u|e^Gti;CVYZtK|8vP>HeK3 z@!m=f%(9&c3nntqW7RZJIF>-yU3vhoEWEL%g(%I_2W9(;E z(We$6l&`XjahEKEVqC=T+!!aMJ@xWjy%TScnLjpJqeaO=b&xpW}&H6 zVKgvL41b%Ei#L50=VqIza+PdzP|Q?4$$!2O$wjcf>;?)&SK3g^duc4QDuVg)q7p}V z%Fv8>JTVDu@|4| z-HXe%OLJG-Bd$x5INorQt!F3q!gPrv@XzlodgUvJKDEpN+Pn%&m7k~6R{o$B>oicF z<9u#`mX%90nMogT>pM3X ztQV#m%VY43z#mwp$q0`uh{D$I>~K%oXSjGv0XH34jzV!LNPT#TDv$Ky6VJU4rifgD`5$j;E_NOK!h;2WsP!w`LJUymj4F5uQZm^y|f*z5b~?yIjT>7*lQpFfK>vW%jS)}lBjYX+2<4*0_*O!NPXfbp08y!Ensux|Kg>~!-UbKX&leq52u%h_Rz zyHcJaL*Hh+hq_Zo8b@1Re1bFsTXI;z3);o!%W>wDEX=2{o3747G}NV;xSuS; z2|LSS*>Y#P_ERfGWABl0mN3&KkqZy?7vZ2q-DrMj4zhS(gEtG_MGo%L@cKw$t#v^o zHQ~7 zg!P;=(pC(Oy zyW^zsQmdIb!t@pjPx(W#8W-Xg%WzoqWfNUCrUA3pRUm;eTk0{)_S&CZsWYb&4rH-8 zNR!=6(hdil($|l#oe~Fyd_~~?4F&#Zz4)j^2h7`-iPnhqU_Q2=Q2;OGvicQo|F&^* zOXCz$JhBY)*c=qy&(^4ANQC?K{aX0*3(;JQULs>H4$J19W<3qMNOtZ9z(^KrAHIjh z&YU4p!IzjB9%CRn`zO45TZD&W-{O_s^l}wS}wPeaCFoE3sGoXU=moH$4fmTwJ*_?&{nnc?Y=nz0P9s-Ll-^7wz~;Y!o+=cbxmvJCVD|wvgGe zct2^3zR2qn6sFb<7r@X|7sp?JOMaPK;t&xB%zJhZJJj)Uvz<@iqROx6+51r>?*4|& z6Yht5?hC-#;4SQJyU$6k%;O}+M&W{dIlRqn20kah9AD~rgu5H^vA*OXTxu(iDxYqo zvT~wyy7xjnyrBq%F4e?)rfGBAbKJSd#%6LoBDcY=wq%yc)PMz=Z7Kx8&u)RI%eREZd`VS)xKS<$4H?Ptq3m01%G?+#HX#)DP=cS-f$VCz&?0n(E*hY7Lj)4 znRLeRbnw%D2g08$@bbHta6phEPCG4#-et{2-uwxe|9Lq6`nQ=0Kca&F#2)4t>a>A< z%@S1Ik%m)a7@~pi;XaKIgzFrS^)AfB{wEcPbMPak>gaBKd)iSlxB5E}EdiGE(a6X+ zsu0sDMbvq}3HkSAqTY2s7`-`>u%0P zE5NBU&tSVP-pEC9899G+KI%Frj1j$Xt zNF6GZ7pFB>AE5`u8u;w@B%0YOOj~2#fo!o5Zd7SO)lb*rU%!>`aP>3XZ?X-`FVmuD zRE?=zl0K3PSV@Om8#qQDKgeb2F*tOK1M;=!Iqx(@t|1RW5Pjqb zq!_Pb=Q2zF%Oq@M0opD#n;W%RfqOq+8uABOzo-6ISbOgv87P~9lltG(o>@UrY*{9n zdThj;-dBpBABsWir2Ns6UGGrl*4404%$G6X?82#;jbtcL2y-s0;q;^Xumx*GpR|8V zwtk)uf?=98x<3Mfoc}P~uxcXVFM-nZlb9T(EPC#7F75uN!~3dW!RSPefx(1aU6zti z-MPqg+T$(<7f$VmrMl_hP_0&XGBml)wrNQnZ?_~l-pP-i@_X}k9$Z29drnYf5Jfd6 zdQE%`cC+VD5cBAZEFR0$fp!xyh_m&iCxCkN-3 zf>UEOxTqA<6}Im5xPm=hzK}=7p9LfJEtk>2MFlV`I0t8L<#E)=In<)hgA-e4(q1V+ z6k2zNo~bwMS{=5pJ>svQZjvmBLuH`M!mmO(e#FQ zv^LI%%FPj?H@s(~uKTgn?Pe&wkCW-urn}Ic(+{iazrbGe28b`Q#iosQ=)iyyTso~t zWqq6|zpoTtEVUd>4qri!jki%xZXI3msS2x0OQ&XCwCP7t3l~K6y5ztN$pbwS03AaRtMn2j^S49yf zQZKaLwXLA4)?)=|Y$n!fWH$86gyI^%RYl|8M9 zRJy15$G>W2z3GB1u>d_dTR`N{Mk+#>ZW3pE&z$M$AbldY1`gtH9eky0cw23Zo zoa%==tUtwJP=d_+ZG<;AHPTmu*7O=ag(BE~-mtio*>+?;v*FJY5-4E}=XCEvvY$$w z;N1x7r)NNCeHE}Ch1DeXfS<$tM*okBFeRN!dn!rc_*rRKx>B}jBRIU=?Ph| z{>>RUuuGkOi4%p?XStx8OqL%}+d&W?34~eM>tcztC0Z-_ZT7SLuH)VRUk3Emc0wtvjm;RD7!% zI@lggbN0=p(6SnK)v8gI@i=PeD?sP0ze_z*e$seTN@lH}hVrd`)CMI<&^3l{;fa_d ztqlJSyu;qqbl(+J)E*9wb{9F}qYAJ?Y&J8oO`M1Yyha1YB@p{v9dAnI;Z63AG{kH# zdiu*0tM0#r()~V==l?bUN3NJQ50#PA>O%NfXAzB`Rt0No>~P?MJYvzhsMa@0h^mEj z@eb9kAe;VfAWe70@W#IhIA`4f=HtR6V9`(p4|fTo@e4j6epi^@9lU`)*`K9BJ3oNt zvSUbpu{hOilA-34zsVf|R$r$*hdNHW(a2J18t&mp4c8`Nr5gv4KFbpwlWamhI_XH} zV=-vImY~m-LOCH92jQ8gE40etIvI;`8VYfA?JXf{D-j8WJGAgCHxK&IVH5RvSB3VZ z^5K_tVpzPPh#A`$kBF)nGGCoZ-Z;F4yx1H{=OoZ_tv0lK;0D?I0+TvjDIC@&ZnC;v z7K6;tuygX8NJ7hmzS$Orwv8pAFrArO@q@+k zHnebs4o(bfMl?$VZ}YB2&n_p?_FqDve!U(nZ#+j@=i5=!mLw>vGXmYU4mket4d&B3 zdj{(!BAsJJ=p|Pj36_T8@>9V$#N#SlF6|;}&P(yWL+&_x;xf}0bqTpIWmyr&)9_s} zLDVy*g;zHYp<DQxK(je>94VRdd8HstrFZ6dE>XTLK{@I}JgIw^)zEC_1^ zHZU@rJ&5~V9Qn8ZAqRr|@Z(Dd(bHGv)Y6Xy=&&r>;q!-Za08pU)Fcd>muy9hlMr1| z{uA&pKUGvIh4S2fw6%yIab~TeYaf21om)5J%?J14LzbR+vx7S^zVn47?zqkgSrrTU zEz-~u^p`P+O@?5-0I+*wgr3gRq`Pcq(i4>>sAGu|`g1x9%T`O^u-Oqvzxfp7zO@uZ z)vtwP&F5GTof_S=B!hCyQ|b8Zr@S+_H&L6J3>DG*#WTvALknIBG7FCnGn#XLqK{jX zU`4AGIDR_~k4+|!JhDJ~N&-|dl%gmt4crhP4%-&JC*wxe=w@3PnXNn*{D|ldFrT;WajiOu}dr%&YzY4}%U;Yqdn?)3gZe=cSD!_dO(+ z3o{T6tAgQU$xsu10|^9tVb2`|9ll@1qt zKaA!qH%3Z=*GYX)0YUr5c|Y#6UaDy-Fr4}TmA^d#3J7e^LUEek9=Ea~9B46D^!x zwI7@Po(b=2r-8w^B)Ye@8@jUR&{P)}ge3Iv;|25ZkJ(}9`@)@^!rKbavNx0b7ia)| zd5=KYE|_FS%iuMxy}XIr??B_z5GuOMk7pdOXHac5=Ty-Nq%b=IJQniR6>b%-6ZUvS zE6!Mdzw$w&bCAqcRlvKvZV;`uHWakI z5TbOi@yflW$QC&<=E&9gbTZQue&3u8x%xWz(Y^V|`owFP(kmtEy*pX{#%z4^)k%E3 za|WLCT8R$S3d7ztf0=~qJ#aWrka@7h8FV{SL19%1!MWF%D>`!E5+{y^`Lxl!0RyH& z{3hBvRmSnSx~Nvj8AFi169jxrBR&PoATLFNRv3pM4^;uEzUT)1C5lwW&z9~gw4oZe z7oxj1T`2js2Dp3egMQ&;=y!WToC*Y}L?4^~*dPPnz!sdUWs9u!KW2i_Xiz=HLsqg*}BjZZOW1ga7ujSzHnTw?R zXJMJ92=wekI9hNJGgXJO$ona6@SY_CrAE`}VYW9K@(UsJSw>meNIk@^yH47-$-~F} z&E&)0&G6fd4=>6Wpp*B`)^`0Zg3!?<_)zeU2v6E~HkSrvk+$aco`ie;ZPJ!PM4@ubO(pul$l8nWL09ewo z1vyPh(5)$+&~?lNcJFtB`!Vt46~8M~xrIX7l69=k#;11cjLq~-Q#qLC?WTII734R> zqOBPLuxvpS8T+11^n!Ntdcm6K^WrY*JevbjQR*Npb_KPF3&6cgOF%125#E}Qa8Q3G zb7ScmB(n<;C!|P6>W8e z&uZe-)M_(XaP1&iJ}D$2f-=;~z#knbID=*5S|}|~p+ONE)WMX$?)A}08uIrs9XR4Z zJ|?%2T{~2Hws-PjhvRyh9sUwl#wk$ULn`$3lQ3x6cpb?XrsK@dlMKp^p#^6W&~6K1 z>NoL}6zzG>Wa#KINv$iW_n;T-%^iWepE*>kTO4^#PXbeG16r!eqs&}sIG`E{O*U3g z6uXTkMQy~nA-*``SS$oBQ>FRgGIXLQ8x?3SCK2=f@u3sLWIT?Kk($lcb^qN5TlQSO z-SCSLjVO5kd@iGzY6pk!){uu!(&^N}IdzhHWAtC|9JJ8il=$g%5Z^_*JW*xYTIC^E z=$qby1eO7|Jz0x8gUqP<0$r#+HH}6xFL~YjxCp%~#1GfZW4)47Bw#X@Cz|VvO*RS9 z{|*oE>_#!werJXg6>eb8r#p1tpVz=a=OAEl0hs+aMp`$7GFMvP!uY>a^mKA7qx|AH zxM`gO-I#dTzc82@Ym3l*PVSh?m4(*^HB7kAU(yyS&K&Bu=cswlraOLx(=FDwL9s6m zo%_Ctye>ZvvR_=_+gf)V7I_)`#ztV#iEqThA&`sz+8) zZl;<4u!+9nO-(^Slvwo!|p7){lRVdx!eGR_42h!n~Lhz0Xht3)WPz(94-cE!;`nt!2B*nrP9}6kIy|gn8nuj-Sdg|VJ>~J&W;>v;X#ApQhGp> zt%KEuV6fsROuxSswr}f0k61=l%Ardz*U%2O)y6@}qXo3R-43&nj`&+aIj8gW8)kD* z1&Q|Lqy4(FC{bb&O8=z?vL=V9)m&eyHLwt}%ieKzHt3spZ8M?9ce=o`jl(ZK4?p zrgV4S7!~V1Lj!JKqwcrv^6~}CkW{x2yj;wMIsP}0N%=u$$0sYI`#SRp7^C?B}^ z$&z-*t?gsPIW8ULG#+^u@Yxq34W zakuY{$6s&#!n#G*@jBz1+@sp9T%oOerrNuCoK>factXC0oG%adU{ABNc&V~7j?^z8 z*VHmebwwD0KY2t|P?RRFd4x{on9?V!${Bwxf8;9chIp}ku70U{M^LoXd9+$o zoa=izox9_4C09%^7+HGMFmu^Fj={q#;X0oXQc>h$fgV3>VHJb5l%o*;RY`i{#2mcQ zISs{)-+}yVHN0WHNGxzXiHS=M!C~83uc~Vm4%#}0duZw_&hfd7{LZbw2Y(*Nhp8sM z`gt8bx>*?s=T+c#V`s` zG9B>~PW%Jwp8NYr_kR#oP2Luf!wWe2 z+d}RLvBB5ezu|T5r?K}H75q!;HExkjz_XPOApaUGY9%xZKHWBSiC{SXKHY@dm*l|x zv|&Ft-bH{Lwm$^)O=NIohB|j(?h>Y;W&x@@x&RNH{|~*=6TtcbCCr4aH;%1H!PcG+ z@oQ-%n$&Y0^&0iyFqdW6+M3}OcSLZ%H@#;Db6z2S-L;74Aw??-b%+Fl ze$kqX<$1>P3pNDy2Ib>-7t2yI*g|d`hG~{&+-?x?FI_xfo z;`S%Zp=WEb*U9HB3rQ2+6>mbN>uvFKWj~N?JBS9A)UjELFg6_TLmF9jq~J#)*<|*e z_v^rOEXTQt)l}BNIUzpo-CytU^_ds(69u+6XrD=1MT3~b+X2ia)Nq&f68yFJBqXez ziybEG@q+hDQD0vyXZ~kn)IJFC9FL=lKf`Fo&n8HGIvW?a=yG>y%mk>Nhc1Zh!D<{O z@Vb}>BK(H5<%9y52kLb>@K@#Mmd5_nc!3k`eZK=H{wq*5l1L!{>6bEhuh-*SJk=GD#M>d7+QBr5R1 zN;B|Vmd3eSxE`&!bq@c~{f_4sRASF3w^7?jIvMM`0ymdNpf2k~5@jQXwdZ_9-1Aw*;fWvWC{~*m&>$fcDwWVYPx3w1`uVNad-eH#|9I9j z?C0G2URGY=RFNsdL< z-Q+g7>*Di%rC|7~i%4n@v|ieMos>vk#<&z2Ry9?X#n?GPaB?Pwzqkljn+j>iiEo^@ zhBM3<_8$La@z^jG9&5D0p7hr;CUt^WFz!)5wD@C0;B6%|894Kr>5RI0_hU!&`=gB0 zI{4=K0H@g%GBf*0`NUIQF< zlOQ{X+i>GOw;?B?LZ)3(;x(AoQTZ|}`a8Ihdv-qrmpkvo`YjC*6rl`u+)Lb6Y%WaS z_!LYYdq8`59(tq{h&0E+jk$#QQxMp!J0xRo5kVYl4W$h8b9qew5Qr zn#o0-n=knE{8r&4?Ev_k+DP+Tx(ZbjE>IQq?)*Yigy*i3u%a!B>$*6c&F1;_7Rz+e z%Q`DDjF%7kRyPcOmr0{2MnLApTfl|3-4MPujN5gv4<4Q7z~fJb3SJh^21~9I3t00Z}=4_VfUOW^tmQ?jyP9{FhhF4yb(fHwhx~1x-@c8)@ z!F;7{q%f=&v=voJ_k{}XXqp(l_|S)Lc%Ffgg7cIv4CY$*WI*Gcm$W)Y6W^>+;f(u! z;XYsM!=BE1Naj_YB{_Yc!B0sZqc~j_j5=e5{>|6%g=_+jayW#K4o2a=oG378$p8Z` z4fRgkc z=Vfzxf;BXUf9H1eKMTgI7K5I9E_6KJ2^unpeyfghuS^?Z$ z0ciP9qYysc7T!EO6^3j1K;NVq?nw3oE=)IFaLvXR_ak(bKZZ2g83(++iNd{lc(5fJJ&$L?ql2qKCu}I;?i}LxadOGR`b)U% z;YZjv{T$4*+y!?omr?(Bwb(avCKx0Hq3z7ObSjRfJIoYW(jj*o*Y6(sHyRa099>Q8 ztLB09bzXm5YCYt9ONG&yr!aW10Gqqw1cCirpsC}NkQZC@&^f6J#nvz{nh(- z{P+6bac@s=t;hG&dVNp)ul#@2di($7egE@WJ^6oYVt>_o;(uB1HGi-Dx7L&Qn%MvN zoZfm*Ur*fQ|2DDz$8|mHz5Rdp_SF8)d*Xl9{@e3+-qYV}J@uZr$N$&#6lxZ>`t&)O+I|-<$u}_pbe`-m})b z{WB<3F|F!n7>;6^yU)T54d-{7! z>@VxB{j2`pzW+bpa&rz_)YlhN^8GMu>{Pt!BFDlt>@nw9Dw{v4fmOfT!=x`C$B4y_ zpsUfwd8SElcH%Hv1q ziCPVd$inG)$>=-T%u4>67%N?Kovo|TW;cJ&XYGNZ@Tt=Q7fktrhvoaQ2{nj$>-(`K z>Lr+0tqi~HkzVPV1oL8_(NhB=@_atEvgJbui!96Uv-0X8ux|kY1L0UYpQCw>uSyuM;9bT=P!>-NKW|giZSfkSo+BMmSJu^`U@oK9r@-c_)V#c}Mx^))!7V+9H?o`if=Z#FQ=h|T%&5JLwlvb$cn zux5`DaHF5&@X695|EG)DtwI@5N09?tJ1v;y`?<5b>!R7{4V&2ex%0SzO=`Fx$6F}A zj|sOVl@y*kf1JMAyoRN%OJI}3^jYA5RP=}y6PUE?F_oh$&@?XsYa*jr+RBNjaj%(9 zbc_RcgxE zm%~btD|chr3R)pbL5Gh#-ds9Mn6l**hHTH^wlKz(Y@C2o$B*Q$U;Ir~rvlfF&b(fp zB&!<}!WQuFi@I;hgh}0s_{3ubF18nEvdKT7QB|MJ*EVEvyI!Ha_aQnXHwUX{zQVc$ z5mp2x3hpXy!nmA|{O6^?sOa@V#Zpgf%semXF1LWsEs1on=LWVnu7vUPBJ3UIwa(}v z)Lh!h%p3=?nlF>MF+68%jnPqTe#-MGCnw{A^ZDGDy0PdKHjtG?9j7@G-E{kg0v2tp z%mi;fW9nxWoKj(epAIj={3V-seBdC~VN34n~VuYxirE|B%ngtw!3^HJlZlN*9jYGnY+u^JBMT z#Mrs}KJ3DWP&lwJo6A)hk23N?bb8{4YBGRh?zAxXpzG|xctz0D*F|^BQy6fO$2_dF z#Hi+OJY_Qww@me7t#w5#bYwhxS5wKXJ)hz{X>ZzbhoYAo2a9qRupNyHn8ztUX1k`2 z1+YgbX6(Wo&kka;n;XS!WhoZs5yKUmwK0t!J{xqq@v1WdwG|wQzw|c65c&eC)g96?Y2;rHuD@H>B2cF)il9X;OQ_XCuL@H)&sHcMFN<^!xnf#;Uq zzW}dXGh^=48mYPT0+w{tmwk-O!AW{~n5Y`bI@jAUN4SCWllrpn>ESGB#06@T%YVms z?Jk2^Zd|PUWTqQS!0YuKW-4(Mtsfh+Pw|t$_+U9|?2`qPFe!GbB${zH;fHcKEzVgwP@}z zC768f0ylQjH_m-gBG;ZYgP)gIW}n{rh_vn9MM)#4h^*slxm5j)>{f^>^Q)_8u3!9E z@K|T2UwVMWpEO|eYHrfI(++XX>i4O^!Xoas|7Sc^;)!Fm_A|LX)!aDkOgg=-5MI5{ zWV7o$S=PyuC|yv35z7kbt&K|T&&rLMR18f2&j%c)s34*qmTX{j5cBTT7DeCD6;-{n z6TO}(#zy{fRg-4v5RL&`s5Oy(dg_BzHb8+;JQMpoj$8HZR{k-A{tC4DxW zpGCei^&WN{Q)T{ZR`7h-+n`h{3!8ULfPNcB;A-RZ_|%8feXUS~E1fI)Hf&3u9xx7H|uS ztLO=Z&A4~FB75XB8*llx;2-^XcD(H@d)qXHotS0M#FWeNnV%wC!E?F@{1`Xx{e9uy z?^zgd#2NRi6`(^xhoG}>F0S&uN`;#SinJdP(VhiQnB@vT=8?Rd4fZu-=A$!N*SY?p zUAAXf=CBF$?drE^b>%2txYCL*8!zF-s~Om^T|%@;7|L9964|m0aaz8U=V+3i&a!!| zp|>)}es{)T*AGRu_;^2d;GQBIU%isbyx{qWq-1HZk1X8F7lKHH=Mf>?$Cg}Lw00Cb&?(2992Hni(UTig6lH1OvT%oMLmpnM}^FNC$GuVYiC&4W? z0%T$%xK_?!ob$OXd~mo9X$U4M>!wFohJv127Z_>n%F5PI6R(>TV=$S+!lv^Mo%H`7Qg>y<%^#`P2}cSEGFyM+(y$OE>z{&0@QIhh&Pnd z@NCO;vbNt=#-a{ll)XN$ICz}y9Le*`MCovU&dg*xWv8=mX(_OcKYO=Y;Kkh>eVb}O zw#2xT>$oHfbDp=gU(tp6RYB0z=XyeP+4;y0v3-DuHK)D54%SfnoXz| zeCx8og(Wul>P9PhX=w<^e0ude5I5MM1UHFp$6V6}h|ftY-cRr0R8f-0BB0 zmhGe7;Fd4B(6DBZXoLC*9=l3eLL$#wbGjCl3R3K^S{o*!lED?`(PvU89Ic<{d7aS;-&1DxinZlj>J$~7c|QZ z1TM*e0Y5|De^Dt{p7{v+dSt-TZJ#i`pE%nq6~Q*v_7x2=J;AQ$Mow>5L{vef~?xOE@vI+3=Cn7OZ@Tkx8ZRAybH|X=LsdfUt-peovgLdovl!x z%968-xS`&Q;lW51un!c0<-xOPW|7Zy(S`47do5j&j7;LW*u( z^jcs(Dgakap2uY_>Y2k5Ld=UycG^mhY3({hYL4^sH9=zNVZ~!u&-Q_jqqpFtHM->L zS_$ra(`W{xby;QTa5#A0ntbm#KwXvu3aTso3WgtzfU+UV?ADv3G}H79Yc3CGMI*1^ z=W&(n;I#y<&NrKz(np>ftsV`Tn~!`3goiVWO$IY2OcE9 z;?6FUAm<-T(S)144(q$0AT0O>B}1mx-W=ZsRW#`+BxSPkCSG?9p(mb zvvV29;S+f5P{}+C?L?;x`iYi*J`XhyjY#^+X6zd=l;$g`!_LlhSe>Iv4$Ki)ggX<)|>qt%0Dxt^kV})YcnHXZD_rg$z_UcA@I41)m9Il zCpEtv^}VOFaRd41nWqi7_C8}N=dEDn#J5nr`z6(Mor1M4>KJ$HHJcbv%AOy5gSURl zkQ;8>xiQ_7cr6Wa(Xim_?D+6Jwm|GNUFVq0P1djFIg%dn+@kxyVyQ7HJ~^8XJ|amI zmvvX-wSqnV76u!i^4wc5Q@LrCvBKSJ^Kg6O8I~utnHzfOIJ3Q4f|?%> zfoIk=d~{(x8@j(;SlxUaPCYcEu5G}A=0)T04Y5pfaRB?k>(Z`pAB(Zk_1tcszStF^ z%!1r`4YcZ3;Yd$Q+#RqFer=;Q2f~MD;3b z5q^!slk58Nd=FPxpV%iXHL;eZx$)0lUhbZp z2elUh_P!Xer{scK<$YlN+Bvxz{QT-=-~`bMhL7guP(Gn`-RL z@PRDORGW2+U*Pf%G~sM}e=>f@3tVlI#-*Q(XP3H@nZ+`1RxYZ;*JDZri@udY+&5mk z#%L98;J=S`b7Ej7uQhI{t4L{7E3c3J2{P&r(s9*&q4wcMe7!pbi>G&CkWCjZ(rsZ? zEsiYQaSrq4^*A29c*G=)bVTcEG+RHzj*Xb0jRS{`gF4C6G(=jKefeWw2=e=}pz0JB zPB6knW0gqfIVV=L@(w>!B~Lr%%L}A$IB}z`=3u{VK6vZY4E!0W&m?q4;;N_Lxcn(M zIJKud7C9`G`K5m%_Ro{Zx&zZ-ytj^3YrHbPT78$Z6B=OiY5sTf{uAs|qr-k2kYdvV zwAhRthjG!%2tniJY0Uq!h+fj87}M}sIG7H@{c~ezEPWuz@-(Dl%`d{yg~yoJs8V74 z)ppEPmxc4o1K?4%2PoZA#pS1b*!K3*^qaysa?C6P6)S8=1i!y;^YIU@SDXgjK1=E4 zx*Kfn;Qh?9?Ib=aSPq9|mEpdotX1qzXSfx=o?9I}2mfe^vx0l|?9bH`Fu1=niEaE1 z27|`1aoyKJPbwdrW)5Or-s{=c0}5>0Uj9W5NU)0Eb~ea(IBEMZ^2%8MME zFR_Nhm1q{ShKqlcMOA*6(UO+|sCzz?%U&SCRCgX?>z*v0<3ivamtUvA%;D=_Ju1AVwSpL_D$isv#Z!e3)8iAP*EO+7CS0gGmk zmtOLiWiyzl`f0=c_7eOgxc?mEW-4<^j)L;7ydqU&DFbxukpBL(=s2B5`=(L@gE?!?!X0N!??L6`3|{vavlo z)ii)DpLCAKrgp%sqeED`c{t`LRe^h62JAWGj9D*(pgp^gP|F~qk}U^XgQBSK?!&nJ z?i2Dt{wkUBYZXzwSPJew+~%J{$o=d}Vt(%()Jk(`E#Vrjpv@pW)qv z)9~VzErj-2%UTB?{k6Cm>F)JIqT)1HTeyGoY|Lz zM=Fr!HP(WUnhkWzg>t%PeiZuf*nan&(VXj%Ya}yKmwfs@#d^iAD9Cs&!QIK(O2p>M z!v0Vvlq%^U#S^C?ij8E-&7(l!BEsbzh5}`t2Q^~c3HYoc&Fdn{qJ42cl4cQ040g$r zAlnRK$hue%IQa;N>pBoW|J5WjER`fn&m`nODN|*pQP~kn9l=R$Q1s}X>ap- zqSCOROh_~$`c1L>Qv_CQ=X6$1K-Q^5k7sJR-{j;j{w^*QjPjm9FZz8!x3G4a_jL(L=lLNw z{Pc&@2M&=ze*iabt;hb+Ze***eDdqJAr83q0aou%AVX9K5yx&NqG>7v0oMayfx0^} zX|5xsqc;)v=2;|^-#=LGxfC@w1u^%)W*9sTP$KmWw?DiQA6~L1QVa57{IrR1D@o0= zE}Y@S-9zx8aTT3BUqDQ5>_a~hAy1y3Ar{U&PNdqIRj+*kH;#`d3$h2o=haz)1%*%P z{_lHH{_$Bl=HgIN@#v?p`}t&aDBMGK<(WaYQ!4~+_Q3IKr8w;ND_GRI8<%b#iQ2OH zuxqh7Nn1$R;*lpo!ABeWIz6Co`bFV|6GEOxKa9QWFtX0&wNAH4?|}9@vq0QnJyGw_ zvu>W?XFXRk1RmrZWM| ziBgI7?b~Q(KZQxg_Ud!2PVb4=yQ2*3Tc^C+O7IdT^g;rri{eV#DOvs|p>K z?3iFV?|B8SXtD>-;A`9v4^2{g#Lv21wXdG!7iCq zTvL}Zj1N4?>k4coFHGIY%K;qwwKfL6#A-9~A=&h&C{SQ^-2&R4dXW&nBy@b9jqh7p zg@KzM2rdqKh2m+vM#61Pa&k_ouxPazN`E?!W$99+##WsKg+xHVH3DL^Vm6$UR1$iP zI!)FuRwp6TX0f_+iMYi`nfQB7hG>?-))u63!>^{o1DU?;OPnNgUrX53cM@a<|2H_x zKV7)7GL*cNm1S0&8<}MCbauRRDoc>P0J};vAbd_gY;p3#i~4f(*YaR`_whs+HOrIT zFzUlzotEVyhO5HUnl(g2*O2Tq_M_)Nt#%@AUgfGssG6AtC(! zPKgPyyfI60`ye@Z>UxB8ov?;IJ!-_IH71kLG!8^MHQ0Wpk2TCQhas;&kkay#w0xft zn|~sVl@?2~-#l(;s(}gSnq)$MuXCJ>vJ8<>AHc5ezk(O5ccR|IK}01e63lk%ko>4L zR1-G@t=uELHV{yg*cvFf*$U=Q*Ab~%sbpX66%rlED?4p8;<_?+(M@TCNSo(0h&rGQ zMn+fYE#-T7X{rXSidBWq$!7HCcX?74f0R4Y*A^Fw_7j^U)ll|pAC@isDXeXmAhRp_ zuqjp3@tk-jI{bXjwJ9DZE8QHaokFP4FLyVL<1yQln&y&Um5Su~;xOT)@^q5BT%DZQ zZ_VW{kU((@HBw=7lJ2g(g(99~S9nJOK;Z#(naV%6`bI*n$9KAS$^=+D(hy#L>*T&) zTS*SdHB+&d5-c+K0TwYS>!R~E?9Gl1Y)GUbkufk}*6XAoH~$K}^$x-}OoF{J+-UW^ zbT~V|RgNXSZXfup{7lZ`JXk_Xjx%ylt;T=rF$HQ!$=9PoM@ zjo|Maclv4&)p19NZHOlJX><_m86r#1pclS-5yjOH*F@Ld<8btpy=0I6X>w9*H0=Fu z#zw1pgCMbnW^|l~(l{$FHKGsQ-L@WAo;pHKU>Y1AHJ7QVRMYS0c^*&Ci6pwCuo60EnL4D4IGSOGOr6@v~C83Zy={;wB<}&xT@DV4YZ3yFw?1|iz zhj4ScD-_(Xhm^K9bp3n^7oEAo$>x}`)M<`bE1?4NeoI+{=mTdzK#iyrE+#w2Wut-g zAzVK*kvl*7GxzwMdrGGN@&W?mBehim}9>I>qL)mW& zU)pF809l%&;a#^Enf`ks&W$+4nyPKsg#vNTOTa(7eB;=c1%n{3JCW-=yo0Mexeiy^ zD6>gQlH7K$LF_|H6z)83ikqG-!%@mL_-nTZUTjUod4cgLVcH-1AKouay!{omk8B0+ zC0;_qPnB4{Rls@M$+D_VkFkfBuX%|pXdhN(R{j0y3~^QHYf=pxe`}$O@mD%v*D=dA zor;*Y;w}0-@@L)Cg1{zb7i`+>$pfACvxELBaQZ?9m-9x9MUOeheJJw72A$<>!2DJ8 z!ofRe^ShgNZ~K5dUmV08Y#@YPyFxd+Z9%JPx>yqZ5cjXI#)qyv$KC6zxMmIak-zO=?a#;_)fo3sm3tcgh3@#E0=I`J z!3q_7tew7+t+^~u(nm;&)GEi2hpR7h(|?;|?w}OTsQLhi5)ull#wk#_;xTyhiX3|8 z+Tjpp$htbO;H5hOcq6_ab=s4U6Aw#3`Bx$q3szn*~U#jV_Ng>X=sCe2>0 zi(qe*XYg3@M`*V2K5ECN!i>kq;Pr(CVDefQ{J#`)q06@6)f4@&U3?&I+xD0qN{Gex z4wLbV(2Qt{*MsGnLl{!F8F#Ldz=q8;g`afqa@sCq@wpqtL6c5e-Sr;CONrjN)yZuk-n%p99ffB$sH zBA)ZoIN6?=;7Cqn`b;3P<0uaHxr$<)zv$4ho4DhBzJvD4aOe!K!pX7-N2F~CYx#rFJfMp! zUQ{YjaXY2C@z^igl%!o z!i3FXtnWBE(IBrh_M?9Z*5&SCe+t#1yh4tZ4jIQ1*I&YvpaiISdW1#T?qsI|KSJk} z62Z#S5!7~CplG3Iv?$STzGzL?8_sp(6K-|6G`G+599yWE%SOMuhNwM|uD$6G&g-Y( zqi@mNe%2ziqxr(odPZ=>BvbHS57AfEg$hLj*eD*4-|*1`+;8OwR`Z&Ne_jtICdH{F zfyNa&VJf)#0E><72(3}p-{Q592~Eh2)l0TvF#pL@k@jz z{)u15mQ6HdHi>^g%wHR;EM&>&<438E>kOv%p+LCxkQ58Ld5}GgO-8Bb{YB4@3>R5y z%ZVO1MPpkJA7|ZNNIbm}`zTLmY6DF1S0L`< z_fit3+!4OIti;4MRdJkFCb*Hktf<+WEuS37=4s4g_0sE+e?SD?)4FuBG`wpdPX zm?mi5{KGBzbeP?8p1~eBPNnVJCW%^V=89g`EfzVys%1+HMzRbqD~$h>%M^!BV{R)F znUd{%vO9A!i5!+JR9yZH4~0D9yw)rMkCjQd_U%@-Z=?ZBwhm-l)pFT_OVKPa`Y4-` ze2A4CeNWx`)k5x!k+?tnE9%TNgHc=Lh?_pkub#_(zF4#$CgQe+GdU)+&xGpi7pT$mpDokw^B^Vm$mgF_b*>bi=EW8PNB}H^|R(;&PHBnB}|)T-n*x_~A=0myLUwo-~<~6J0fqLv|ha%mb6HJ%8zt9hLS9nzpA!D?@sVwHgc>-)}8ZZc!{PHknfpFPNLj~V2aYBVbSIU-EHvW->8j$^;R-GHIbGr^NHfzqgCh^pJf zbq|{WTeeI^=|*1Hss1)CsIFkoy^pe{#l|9`PCT=J?T;xx4?`CjOIi=y!hvDeSl-w^ z;D z)k|pFiWx!=Swm``b{m=;MluHCOmpZ@I__YSkhSl{Cp*93qeE%>I!x zA1H-ac7LK(w>H84i*hjOQ4=?@#0T)6h^;?Y#2)uSc7LKdP4;cGI(tGLW@u}mu5dYb zGr1Hz-abXS*Lk?5H5IDrf?%guCzoo#bFxINg3Frkp|QRR*2dY9>!lK;OI@3jrCGw) z8*<^^N`E3+oDxE*bwe72?9%g%2krgVxI{(5AZ-1^WiVYP0K9-R&rPx9Kro z$8ho_>lFVyG?07r>jDnt_rj%)JVvMSb7{A<2lw?x3|A906!JC;;Opl=*wn5el!a@I zXQ^fOt1U&k>J`-fWB@-v*aHVF{AW`IdYYA4fFC9d6eWzGoC31Ps}+B ziSd9}-2;UCez@V9b0xH1pODG7uECoKcev^{0PWj!agKigjIfTu_ln11tFb(3T%|;^ z4@$BLCZ8e7J%Tem63I=vS|xOwF^Av(HG^&KHe_I!4Y(~R6JFCSq=T-0V8_-}va5O7 z>_z4bW;1dlyLfItQ-0nCuU+P1-qRG$&G-&%czTSR*^meO3*S(;>^FjhaZ7Nw_i*g2 zTF;i1>ac+UUU=Iki|Xioq<+Fh5W38Yu3RGpMnU%lKaE}So3sb~58P=Fp49g*ENU;l zw71sddtPu{!!ueILV-#xu?ukY#k zJN{SRQ|qnw^!8X!{xAF2+P`YO*Y>Razt($u|MtDt^!lE)zvJG#r`B8VjsMQY{#yTU z+>`%Z^ZifWyZ(32-?jhNegFEmdg8x(d-Go3yY}C_r`}WRF|mKx|EmA1x2N{s{(t5F zt%?10UC-LT*8lG5iT`DPU)LM=dNHfZUO3-z4s3l=nHSprkQ4PAO+3ZTIWehX@>bH5 zOy3d6l#W?&XEt0F)N!+!R*5~x4Yr2Yr+BR_=u6h6^(9~3w~#WsKBOkeiDYn7;nlA) zvd`otl;6JuX*xFw%$z#8Eu&9R`}XT_>x&%M8dpv)IqrgE1=^%4MFN%h`(n#!W86Az z2JCQ(!{Y0{P#777W&An%w1_s;zmp8Ve*S`pR3W!!+hy3ky#l7(hyz{i7Py_k&+@J9 z$CM49qYL)~s^Y?M)Gw5{=op4YSB5g*S|vQyKOS1`BIp;36!I(EpProIBiz2(0axzIqS`wZ zxUX&JXoyTHoR!>396k&q2gZzGcm2(T_Zyv9*hF89~tFhg`>)SupeAQ^0C#NFG(f%^H7}?hVR;S0&qVUgrreePktAx$U4OiDXJ;t)-KK(gU+^OjKi?q_kMD$eS>wSoeg@8eu$Wl=(k6eVt3lUYX?CymEbf?p zh)y#fLbt?63U)PRptZUZX|X7Puvrh_V@LpLE0TsI8*X7kssa8m8w~XuXYl-Lp=2$8 z4(l*}GV9aS2Yg#>v53cepZmk$WbqpEI8u@XiDkl(VY!gs_X6AJVnk+tOeUIU+9ao2 zg4aZ{xIASo!?pY31Kjw2^%^J88 zSPO{{s^L-odtsDv9sVj91b+N8;`-syIQOg%s7$>JaX@~K9Wi#V&Z|*i$AM4D0v7tiS z08KaXpYCWZ5C#M4TlQj!Zty4z{{ ztEFV8h7y<&QOGcBiiMjktav=Gn8zlTxT@E|# z9+6+USK&9WQM6p+DO_KX2}(vMd2DwrQ6sNOm7EEYKXV2y@@LfAKUJ~%*LteFBC@et$A>h0h<);)Q~it4#pX$(*TZ^#?rxXl0_6%BvC(*Bp6|h%dp2dV2kv|h}f<}!l=GR1UmGyq) zM4>Bbl2}gc3nYo{umnqaZSFh=Q<>)feCt6!^c(EhM_k2*W?gLC)d|ZrBldv_7_o3$o_#1Is0?!*UI+ zU#jYob941rP}Erb^TMAtys@*KZKVP~E>n zZ;;(Q=cP(@1LlUDfT|B6a5wIp;OW}YL~&*+O#dl@$~aSE1?t%Lg5aj)1QQ&SsF<_y_Pw5bpVX%A#~=TByra zFJ^E>o)l`_;EvGYcc5k^-_T(2-XGHKx8RaKi30Q24FZk4b{cAKMf*j%kh^jo;5yxs*CO!5{ByN* z&(57t_Q@K$XDuLw#%GCqxF1=yC=3UgcZE*#K#79~BMcnrxhXl$Ay)JilfbpLRcl#IWJZAU}sR@d8{VT&0#JJ^Ni zqn4*?VsgyN`=>C}Jb~-CtDKv*qJdKfEly&M0=u+iJcN(4gc0Eykf&NgV;T~nYH0{s zY-@yV-<4s{>OFA0uNt>@t3F*b!4TatC2-xl+u-qiExCCj0W{in5T{XbbW8gsC|fPS zJP9>)N_Ysf5~d4-Zs+o6BJ1$>%+uiZT!LLKIfzH~*O41fE)ePHbQ06L9)?c0$7Ly} z;KshY&}Wbg(>SR^3V!{dLu4j!3!M_F*P*eT#inMuRrxd*94QaWc0a}gebRv`n+eA& z@VtLfDNwRmoJ%Pi%wwF5s8UuTF0tuPX4g*y|8QySc)SD~lwNWgL${HWXjfAuw&aIY=~Y9E_!|Wz1dLgmbwQ0ckQODG>Ylrqi!(m#|7AaRfJY| zgGj?b1Bh6DRM_^)jjQ>Q2f41>s6#>k zwyh5$JD+~U#RvPc6C;)BhO@y`d9*DG`sDC<*%DOTQ-!hjuaL`c3`lg-K=NY8Jn~o& zNn+|UpmRVBtW4xVFSpl-^Hk{?aW@8 zc5?=IW1-L@8;S;YkgF>*@VcE4$D0PhVX6L5?^28BUbo`cpYvfsS0Pnh?S!j-)`0E9 z90!%tPzM>!76w<*$Q^}h+A*$JV=IB=rO4;_vx_FKVe2o z1erZfoFo>6fvV{jvawo|S{rK8KJ67&lhljg*O(Wy^L##-E#jZcv$S|E%2rTxiU7m9 zrx3JB4I}QXB7dB9S@-KTuse1f#J(+pN5>EHShXnOw3k-MJgvxW+gbGHPcb6ezm`UJ?83+u);K-< zF`SSdM(jRRk{iXjP+z7GtvvVB+v`KfHicGhz$0fY8mCTU#sk|JY)V$}_qdU2^}^4e z7H}kGiSYNXB7C0{MSYG+@%RW0+*@NxCMUZx^_T)uF!2@nU>gdDr+AUHfk`B^bvwA8 zLb9diEw|<7Q*OrM254ST3+hAk!R~H8S1G>~1_$md0wxDwl^cCmZ;z z5G4%X5eAO@bJL)jKhUgOh})J2qNkZA=N5OHtaxYu*^UpPOtwWZHDflXX_UoH7gSI= z(=7Nn=pug5y+T^=$CBa2A|g=Kq;Fk=X!4)8_Rjk}Mj!OefKV90~enBMGMB)~7~Fl3m_~Fo%EtFDma(zK7@$Pc3geZgiLA@b}e* zofhQXmQ>F3fd)HsqYFljm%zi)nwaDHkVZ&c6lA4`!MRnBt^9|du^QJ}i9ha4XI*<9 z!jLXr8!I*j_UnEDozRP9*B726j6Y*ourvs8b|T!fctMWbN`Vg^&xpJ2G<3)xPpnNS zC_M-UZey0!?_w!7BQ;j=+P{l))X~DSbHB*U;C*E1SU*Emyl4-Tq;d*5d+}wQ!-rjd6gQqHz<0)4g=Aq|5y0X zkxM#!esgM1cucl;DCpl0Cce>PByn~)7DP@W2hLmv-VJFq6gPw~Hz&90?PqjW`X-vwbkKLyTCo+~p?bbwK89 zPuR=r1F0R4qs3hsaA&YG`8J}StgEaB1zRH+zIrSv4_7B@?XE)#~h^!>#FFOk;!yc=|`MsJOH%LPa?jTB7_6g z-*a!YG@v|f3&oQPf|{~w0VJFLbxjN>is zqGc4NQdCGr=Y8&zj1V&Wm5fSBB1DmhXiufRCk;~3(0T6Xkd=|hmW(osPzaUzd;jb@ z*VP~Id(L~k&wYQtpAV!vI&(VBKA`x6K%kcuDunR-^~<&7uWcjI9XAVhN1w+@is~qE z^dmEW>>_HWtz4T=Jf3=agk13+PJWhYK+cx&(Es2ZH_ca$1U!|(sb7onxuqhhmHLl# zACUo{@vF(MUOTdDRstTwR61@T8XvmtCj*{axthjtWSDyg?h#9~If1wEq9fz13$4*S zJeh93sm*@q8DNx0H$HhdR%EPpj0XLkA}G+$K|SFx;yZ3W1|>+)^)Ay$|C2kMt%&EO zJNI!@WFOMZ=B@a$-x`~?^jLa(bn^MQb+~BH7-sxDgcdqZrS-FOMPK-x&4vpSWV5L| zhFo7n6_=HeCpI%tChO;!rQ|rC_FX_<=7iu@$8+=$-+76?cb!Oe%qO`YRKS$){XA|9 z#=P6B=&|Ny_>#5K5lc*`){0zs7^ske*^ekw`sC!GXaeX37zUPfcrW%r?FZ8)N|BWM#((-gs_&e@e=TU)U z>wmc6$Gk*uF(JkklxK-L1Wr-|m{b;~Lkyoi;-bqn;x)K?3J%NjIyG}>6ZJ^IK%|xT7&Gh6$ zM?tsdSNd)4HZo$v1Iy@ICBn-X1?^|_$iqbuXkOxq^Y#y=EAC#R(gw$f;MxRweXA`> z{ThQ2{ku`7(Us>Sbo)xI4u}VylST(=$ka?Q7vKN{%Hh)1q-*(saCd z>lDqJJBQ@aQ=G1h1({xZ03ST|$70V4qTa2p=C(vQ_-Y*M z=}kt9Pimmkz7BP)W?*T}P~7=9jJZ4w#~zSoZ(N^p8}wu0#w=-};wNb~?2?J_vwf1V zdTd4gQl}u%+Gllw|MW()A&J#^G^3A<>#XM9-|!bIgS;FWfM^daPc@cK=J%~2;{8;4F?h4P7n=At(S*P z-ffiH_lYbP^A5fCr+9zwDXMTkpH0&BT zRh&N*!dibrchPjTO}ro~zqf`AStn1_)~zAQ2X{cJdL+nbH$cnbURViA@J=Uzypzj? zOAem}K2>i8R;%4HabyZUyLXV>D!K?>tEbSaA9rci_fp!Ekbs@K{3t(nIho-v4Rhcy zgllgV_6J|4Evv(E)PZ6WuM)$V7)OzDXKJAAS(hlSb1_XB>xY_Y<8l0FOTqI4wR~2y zk3G#?$PV2shpJykQ2zL7dSZoR=ff-Brj z-w2UjM;+e{k!C$(c{by^MvZ68Yg zOGndx64y9%Z9!*xj{6;~&cElUv+4z-n8dC`GCf?DNZn}VF64HSP+u$jygCD4ZoEND z51EkTp+C4-@pxiC7%ngk>%$)>cj9vWqxfKO2MJ_8Fy5?uZv>R(9#XO{1ulg4!HN;m?DU)%8gN%14tz|eYfGy* zF(ToZ57t&tU|Nr>p*zob8e9DF0`DeB%cE#@b#LjIM;w@ zM-@8al)+-&?OO>kZr`|Q#cuAnyg&7Sd;nj>NHNoFHCE)RhEa-3ap7%WQPNLyar_5U zvAS9r8&-Cgi}pK1XYW))%kErAf4*C&pLIgG*}0SFNxATDIz3UB%Oa*Rn9cK*E|LQm zUSUqwC>XSzMsC;?pwhh6f?G!5~IwhJhkTm)|ST)0Mw!|)+F5Mt{$LE`99c%Y-48V|^e(v_m|{QAl4WoIJl z2Q46_Ut%EKNI^8FL5^AXmyn<2PiG+XsX3WX{^Nym(r$EzOW4y)sM z^*)2e7xutD%_s29w!(5}_gSjq!q4xPO~&uyI?<**7og2%Hz>aM#05$RMZZo+um`;F zsN?PwX770nmKgqqiy|TUrQ?rt=?Ex_Z^lE~67Xx~X4o$7C)Xa94EQ-w zFBo#V!b|zhaKQVzV6pBS-edg;KUt)6ib-Gbp4@&WxAjTA_JI@Jzru%bYICqW2dHR9>#by#FpF^YU)yTjoA-KJs z0BaT~!!A#KlDOCj#>&Z|y;UZm2d<)ix|p=wEkvX8>*%msmyMIj!rqc3ay8C@w8hqw zjXmeldGu?(uav|6*-%W6Sd_yOIYT(O?FCMeQGlZ2c=% zMGYwHunjwV!-;%tvEbr;1BCKC@^X#5Fnz)hp?7~Y@lR?)Et1Zhl$AjB$rXN6^~)E`tFL4GIdzD(NrEs`qMP5B&@uaM zh=tn?n7io&m>n6xJAapB!ogu&e>4A{&tF4w^e3JF{x)%%c!T-Af50wB=dqNIsp#rr z$F(F#Fbj~Pk}A^doz4(uw62Cp{jtV3)9bKo#ToS39mCEo9K!81g2)oOI?7=B0W0y;!PGrnxc`|3Q#U$^p`$tG_R|P|JM?javcrOJTliZc zoy4BLiNQjnv-B|0re_zNA?Jtffxyh4Bv|D;$H`4&3*0_px2+l5Gv1r?=r7}Z zBZofclki0Q`}(efMV7$@->|xSG~4kcm31zY5@TyDdw*6;+MIi-NwyiqcOzKWuv<8) z;|dn9NW!PpGTcvAM&$?Y%cF(Rx+Jc>9u3>%uV#qh(lz-A zwOM-45vn6;$cB6F!+e))^tgT&PXq`>H;YXzH>bSft~re)qel!UCb1~@vpC|&d^|ge_Zg0uOs*Fxv+CB9 z^wicge0VGgTO9vk3m3)4sXH*MIW07P*?(-DY#BS5W5bf(l(4rpK2`b7Ia zyP@6kAi6-sOz>agD{S2rKxFy3o|m5qHdj^P=?B@kTqc2Zetj$0Rg*${?xkX^nh*BJ zgb+EPZAiAy|PgX-Ff8Ax-onenqIsGLXsq}C;UredY`>Xg&r;Jlu zF2k+7t4cDqOR=!F(e#{&3mYa?LbA;@(eR_ZVDokfG?s3q(i`%rwMQb}-h3M$>kYx9 z%}3E^X)8`XE`eVkDALw6Mkg0Ha!u{?S>LZ;D8tYGKH9w^`p44gj7n`hev0QSUumSi zt9EgN`YX8sX(e*=^=aOdnNNh{PE*Z4o6u?VQa%R~AUMf*?!aMv_GFa>&vyKY4%Zym zjRZIL^-2tzHO7P)>q)UtbDk@|sDUm!k;#9zc{jTECEj;h!e^|d*>dj!c5ZVOdl;yL z!%W7os70yhFULE2M(?DT5|!!j#3XLvb9r{(>k{5qR>mzs3rSh#H*V#)Vs7MZ%53D% zv5ZNB*yiYnYO5DBlNBx0e&Iz{GT{@u{_!HaWqge@XzZk$Bo(N0{!+odRjS-KVQ zgEd6^>JO@tn2V%5h1~9{;lgj*lMQRb$$(b{D{2ZMCraLvk*_A=5sSOTXR!(^($N&o zAKD7@KHY(U%OI z4&)~8j)H7O3E|G9rBHL^H240pGTA6N!fCv*hRH8=VNcRwa>Ue(EZ8-c+c8CwIkldE z6Jx?*BI^?jV4ZEgg!Cuo5M9W2vr1jR2<^vPS zjJn6*LSq<_yN%y5{3VMrTWPg*9cLz*jDM7)Az{iD$aps#PJa(3RmbiLBxe6Y-R#v& zzTO6cK6pY$w7p>O>04*KKUHzJ*NDl;Fk5Wd@e^kZH4~2c;U`Sqy1?zU=s1dm*d`xY%NN$yW$aXZTL*cqk3{m_dn=)ti`S#Ovl>n+5C4umdQc40*pesLbpr7Er^Tdmw+qx1?o=W!zXpNamB{CY78&8?8-Z{#P#e~}tD>_;86=s; zvcLYC7_xi~6LS05u>zErNlM= zCT5u3A^k66NUpXP6!)LxhCZ0X>UPW|I@W>2<8?Sl6f7m8fI!hXooY+f-)Z>L$@}E|CXz;~ zjA17eL2B+vP9w$(y51PjRlF(5 z13VJP3Ad~bBQdsy@q&5iIKP8_Hr61!C&xh5{oCOAQb#x> z#sQ)%R>0M}Ghj;ce4<|~i#}`bV~bojE^c%X-fD9bZh1CO=z205Ryj)u>psf}W4DwE zi(LYRp8gzl)@X;kH9k+>dgc}YNhM*uj9j+@c2mh`W@HMD`>Mq%W zUDhSshI(n}TpWQ zgl%bsV5h4>?i%>Ph8_8^!EZ0|c|DArQa41oA=+S0f+2WAJ|vYGK#Qo36rK4@tqa`X zYORLwzO+8Xtel1B-$r1fyEo6>;(HwWB2f4FOt#y8C7YgR3iNK+gM^O-XQTRuJBl6H z{ZCUkd9Q--yuT4VapL*^uZqwp@G-Tz5lcRtNdc7~S^+=gg9PFY4C)+Hau z8?JM?M+#$v>uIG>_HMH9hDvn3_Gk$*yRC&gTbc)}y)=Ych&NIb^m)3H>+x2Tg6-gQk~!L4Dp97(Zn%NZnC@I`4TvX6oP`b00yO>nV8ETL`Nx zhHw*;E)wemS;9G$5|h=E@Sz3Cu6#dO<+4@SR|&!`wSL0byEn*#DRv;)bO&$V|BR0h zy`r~LW^poqov|xFjuWmgBD)vzGl}*B$kW{dBNoiY#EByEYd*7;6GRg8ik(;Gp{ z;Ww31I?C@xZqwoKRuPA_9T4UzC*1uf3gjQVz&&_&VK{7g*!1#>2aKJU3ygxGrtl&9GTx-QGd!$b8Eos6rvKsJW zMIv#~e?<~@B>{}Q1}?IEMzV9fz+Ekln0kEYR@R+@PgxdV-am~g%H+eS#_MF}h(HlM zx8YjCM8xk*0||Ps3DIT8$ecI4=gfKrY`Ff5N_EGP6xK){l*VGWV={(IjfCfKTfu;5 zD_ts7qAv@av9 z0?FF3!cEBvv{OV$)&Z|+ep=@}#47vhL#_&GABem4L7 z)&mE#ztHE{3VIJC`5pgEZrtCYxH`rY2E5jQ3^xG+&If>8KLgK6vxKLgH$mQz04QIs zL&{%JS~|0bD6BVyDI3D+zqoYVlDY@h>|YIsFXh3ZXZ4`!^^b&|-GDRwB*>GM`A~g3 zl6Xm7f#!)TNX?iN#GWLOKX*H*@yh$)-`WfDr-uuNPAuUH{%oNNP5ZI+<4st3?FOh- zdBJmw$z;Ky7M|RwL)7k9kwF7v{I^<>tE|(adUa96z~Va9G!x*WX@tG|RZi%GcOt{@ zAL!d*g~U3d0ZVh`@!G--IC#()+%^mci>zH|uAcP4#vyie(azoYdC3^jshP!O&1(tR z8oLi%`mR8Hg9lVp71PxV7(U?h51G~fVXyaT?&7&5FiX%UrrGVpp!+BePF>CaSSm4o zjT(P{&eMe2LNI#x9g+hIaYauaxf_-aGiS=NOD`^CXs##JMum~k>k6>TLl?iyR%E{} zyOK2xJa6bj7)h*+g}zDSp?%p8L2Q2v?7!AU8_q3)A)g<>kgpBk=;lYvZ5|Rw=Omc4 ztA(7G7m_7Q_+F`bHJzEKhoyfHQ{l^1u&N~z3P;C5z^?NodWJOIeYlSdO4q^pV{OpX z_MMakbl`LCIZR1Lgk~WHxV&gQs(ea?@cDlreu}D4F<>doFRLM6iuS_QKn9%|TOn25 z6BgCikw=jSXqw})`kwsp7;{RRIxmQYF*g}e8x+BJ1#RJ!xHn+zsso=K!@)(k2qqtj zgzXQe3m;j#fC(HYE1pJ?zFAh}_>*(g=#gJZD&LYkD{HbW;4ri_oQ2MnhhXfU z1mZu@iIY1OLxVopkk6Ym;nhM-_`0-}vsitOScLV`x6WH(VSqElncc&Orc+siM>M4S z{)3gN-vuu}NOReuWw`%SGk4f_4meF@WXd0Dx-V!A_i`YKOrHOVypob6gOB~-lGz@? z-!6t>{}rQugCFnn(G*1&BwIfA{mbbo6`+@#JDugM1s+D@ptr^r#+o^!(wkgN`2CU? zv_B`eybU0)Iv2p&ntJS3A?c3ax$I;80*hC%oF7XOr0-r1+kPG)@6!FrmaXrIeNr{? z{gwqQZ2A7hnJ^gsKJm;Pk8&FDVlI@D*CI;wzLWC z{G$Vj+WX1%>__Cn!LB=c=y(3DDJq}K{?*W$|Olq$m=H50fA?0{W zG%_(Bn?J~*?%E3^cdI^~%Df=#pCOt{y7 zve0938Q9y0^1s=5D7h6)#RmHYbB_rD=kTJ&AV*<(+je0_?K_yEdkL1vCc&$yHsPm@ zL#*QCrL3@5hdZ~ZLSXPAmwRr00)F|I!;`0nVP5|oQtTfJ%YziSk<~R^dRi?t%38|} ziz6`k#zr`QW-OsEQplf<9N_eJ!=f*vIgJ~!(C$+OMP_eBkMa^hH}(NYj%X)~#+k#d zKS$tFcPw{yS{Lb(k_T$pPn6pi;F=&lhZVC+csDvkxV6k75<(Tgp>08xB}}* z2x?NbBW^{H+-&32+bE0z`(Ky zT!P2K_HY%U@i8S~LV_v`nRlL)B`+sU7edLE(ZhtF{F}fq%v8APi3=PWHXFQZhT}Ky z46xR2hMy_z@Fy_?Q?uOYki4NpBTYfryv+{shopkqmh~c3b%Mc!#qemzdm6~+OT1RA zLhlkq;Uzy^;i&+1p~lq$(rlYS_O;8y%Bv=j_1+fT9rK_hTY_F`{6QCYM)PMfH3%G~ z1=A*;C%HS12y4{ag=V`(TW!cIB-hLy5gR_UZnr{8=(|!)s3kZCJ!|9O;HMBM`nU$F zrlvxXx;^ir8V(mqT*;C1%iz&Ao<-QS50)!Bfp%{nw1r6tUw^6w&+3!pR8A;t>)HhZ z0}*Hh^pP*=g=DElIPkheB5C;-#1DG-Y}am~?8FS=Yo!jM{rVNC#AocC);2<>QXy%Z zSwjY$hl5kqSa{K1NY-}5kdjk!q%Av&j0`CxY<~i3Jf23>mr3wUqKV)a@dYy0SAxVE z1r(q8#`$sw$gX{lAbHCRXf-kuuD?1@__c-;F55Lrxc%E{*w|PIk5?Rks15dTLNlNE z)eV71OUH8?Yey0Zt#W~+tUTPa*M$lDh6>9wo|Ah9{Cq&~I1x|L=k^bLt2c3d$i>*2 z5r=>VK5Od)^@bOr%=<5#9WMe`>ITtoO2QbrKBjSPOPhS&$LCn?$IVK-tL~ zFh?O1ZZDb+eRHpp9SdV%WUwcwnB>8lt1+OyE)-mD@UF9Oszi_SJ2!(vu%~_@eD_g= z_mdQa_jUDz@A{kJ-itg~a3ut4<}c*;^&WL?d&)%}{Tgt)a|9`<9SfWPGvR8BGD(8h z7-H~xhR{eUSg5i0q)@%@A{gHp0L4vvVfs{kVeI7P!cW?Jg{?LG4)(qr>H3mF*hw8| zZz_Yq^;h^zu)I+4Zj8`S^;l-;IIKQm|^h|h0^J8yl ze$_~a$p7N(ohz{L_&8E-z6^uiEwQ(wTHyD?nP~;(vof;>*k~1nr#G09+eey2JH{K4 z_7F;qLQ2{F6mz!oOg1rYC>A|WR%9!uxwD%KrsJ@Jkt}bGF?Z*(3N3%FMLx_3rDiov zs2Y`sKA+DLlUcF2E~E;zey3nW&=GFEMGhS;dWZAwRANY^7dqw0lmCXBf$awql=Lk{ z8QBcjx@!!4J{3VL#%nUaa|Wb#UpWi?R)muqQt?7x4S73nFGGv-|>bTt^6`Nk$P}31NajV>B zOGBmOmSJbSu~4!A?ASc1R6?jhSqkeo{uFTxt<};6prYxrvF(-aI zshPo%MyCzXX!$}=TteBq!m;dCSpkz-e1Og0{{jtkitxA6Tz2QE7p9ch(-a#Yu1htX z8NCZdB~ce{an0mZh907t?Iox;yINEvJq=x*6FJR9753A43LD?a=Z>%ipLJcw*gO@{ zu_L>2^;$#LyJ|O0)%j0=uJUx-7HxE#c@Vv_0d*_fFl19R3sTKw&lm1xD|$Mx7TdJgF2hZ*|a;>tZ=qq@P zolS??n0eResqhk%pKHUO8ckt03(WBTcn(cIk&9Hovct+XJ+f; zm}>KQCWu|fh8M~*Tf0o?RT=4GG2WRI%-XXl#@+3(4cV3q617+O}Duv=s()i~L(bcW@X6t$2wq z%;Q+Zy(*g4CB;_Ac46$MG<>zJnI3rimF&Gdk*z&@jK18^M3Zb9iRO=E0?dlR%OA4I z>an{>g_kZqd7I3nt_QKKO$*r)i3oOI-u)+l_)_@G7}H+yd%Gj-lQgHGV|4aK~sesexs{c zL-06$w|;^fZ(T&W)?C&Y8_q1Z&tO$s>`78t8tz{j&u;&VC36>gp_x}0o|AhAFg$s1N3@S#^-+x+`R2f)D!Cw&LYyHz;4-#OsIo z@96S%-1^@4C5)Kr^JsY1vK?m4kE8Xr7s;(E z7gC}n=AQMtLW*b}M7-WhmQ_t>vvXD0kcYR4|BQ{?yi;D}&kG?uyz2>4R=;t2uQg2A zI|`Pjwvrn&C&B)))!_e>kdc~q!N6q!IL|L3lip>p--A|c<@PO1XgL9TJp0H~w*vHS zvH~5A3Q}JiNq&6zN@9Okk>hvu@!z)7a3*3Sc-YTizdu_u%gZUa(PRxY|82KSU%mo) zY#A9k=R3Jkt;vm?7KV%W*utd=h2+v1XLc(mic7iETfh6h0_!OLOLxw_Otm_A?xfui z(7UKe)|R&jR4=rX?iGHbd*A$F^5Yb^7RmF{OFrZAOe=2aqFbc%MK}}<7|@W-ONj39 zFQmbAJ_=>$v-v{~vD#c~Sa$0n*(cXVntn3!$+rgY&n{vHYT-Qh-UgIE=!I#wsIfuyqAWLu!;z#X<7qy3(;*vxrIva|M>^>6XS+~G+Loyv$ z;|puJ1CZRQ$Nhf&ThKghIhhcxjDcCEFn*mH+o5-!%^#Oa)k^ll3SrarxkxwO4sk*9J)hvr0(+f+uEyo|wg zC0o2uevo-|t%CREGD4#GmmU6@EqadXC71&ms5KY^@1oQ=bKY8Uun$3oA%V9q;KN z4Nk>d&K|Vky(If%5ewFmy9B*ILfDx6OfBi(z`m*xM}WwnMg zL38gnl$$o0igYaCmQp^3=l>9xo_xuKR!?zdL>OLNdWbZCH)mmXDmdP43^~1Y6|TN3 z#pW6C{b&>j_vALh#%YJi=HJ=e4MSDpED9!{i`KFgG0)lfVma}&ArDY)ydSM)1Tx1xy&H{ zj>?K_D;2~B_K(p8EXbDLwNRzYXMBTWIYH-0-m%?AW-t3CnEEG&VR#zA5EYiYN}XK{ zUW2j0UMwz8n|<||M?XH>LK3&V7ew5XAmiNcQym)*%ag4RxWGX{tUg*p}#Wlhn`9Q<21tekCP~icM7kDK`RcztY%$(Ev*>ZDt z)~|e=%h@U;J`_^J)-Hd5J#SYKkz74-WZPL>eg^jOJjVD%>csEqACebz3j4H%vHsKD zOrwkwCzI1maA-O8bg9I(@<}vveml$m)x+-OOn{Gny|I*?MOz`oed%Q=vGoFrm#b%{ zRa-$RRs`7#wBe|AJgJG?1NW21!qdeX%-%2qz2>IlbIF-B{Zj$EX2<7d4(;STK0alp zA2-nx5|h}#&LKR;U`4}GUs!-l`v zaBSgwdeKLZe)--H9-%Gp%b=8XJ^g_*Kj)B?w@OUoD4~x^t2y%cFfp6-g4&DiSx$i^ zbJ4G3M)W1~z2eQxw&%0FO+MVDrEfUp%5wA?#=Gze`l2pOiQWyNRzm~3^QGBOVE7{z;PuR~#RAk%9(aKj{Q{ zc8O@$rpD(=?~vT`T&(sS#R5$yvu>k5n4mQo-kpnPQ=e;yuTOMi6?b}Ag|?^o{J{`X zb=erZzQ@%LHRQFN0w*m||U1P2rpA#q3zbwUDo)xSdA-4J1 z#!j8wM{{(X@#BJ3cv?C|FyX=_k;l{@^|MAjqzfG0@qM`|#5m*)#yxYzqtBFRe?&bq z+4WYlPL$HSD{>{|)VU6xxOyBVu>)rZ= zefZ|XEX`(eHJ^^r(LW}kiBz$rY+M8W7i>eG2T$j0WDn!RsAQHt^)YiTib8JD57FcQ zCXoV(HC$!O8QiNKkDkf@#^Tp>(=vK zgBDD=tk04yOmS7U4r&=iV}Rd5s?PV0loI`zowN#*+Iya<-e1NZMmWP4&o1)EC>Kk| zR|@tmT8@jf!f52~9C6Q09dSUKlGrxjIJqzT1_M+Yh|Vh_c9AU=-#zhJ9PDwFGyA3{ zc9;!;m*G0FP*I0S+kttEcX|R!f-@%B9A(EZrn@IS36SlzVFmW7(^mwlpH&aRp zYJ6p(vwt0}7nhN#@^b9q!P%_v+gRBDcP+PX^*XZK>p4E&-^f~xw8c%c+=zPDR`UD$ z8S-ZAd+zd#jbQh|1tur_Vhaj-@xjLo)^F56(oWtdPBKGa!<(;U|8`emFxws?-we=* ztC6hJ&X6et9-;Pi@-)5uG46dZUF;=A#6)A5*xKEKgkT;w>--$D_hU5`ZA#*-FFDY| z-d;@AF_M`aU&CJRl;=5#&FEP^LOk#!7~KL?nDY16IC|JKa@KP%d^lx=czGhM$kzbL zQG}U_)CK?DY)0?32CTkd99mZ7kdrFfOljSBZdF+c4onGPD}T;nyXQS(3UV2&F?%f$ zH+ZuiY(m+Q?(qBJD~vxOX8X2SlQEm0@_oBfDth*i+buj#J=yraU5ZCmiML^vPH|J-W}Y6W1G-%@{7&fLYrwp=w(d(=^N# z8+$99efrmuv=2GJ4e{%zv#wS$OX&i3Ip{09qmhe4Cn&Py`$O3o)lKNGr_45}nzAm# z!^B;ANI>OfyuaCVSgq^3f?Yw#|()yx1zF-xfx4|AK%~2EjYhc6fk52jn?U- zk2n`@<=lAgeDqBA!6Jd(l3$LSzLk*`r>~PVy`fBM%T)4RE*OJ&_pDly8oe_8xaj8y z4i!)D#>nXH^g_T7cFp4=9vhy`a_(t~2R?ozzZ{p)zz8W;vg0FJl%fie1C}iExtECj ziKVL)E2x251d%tFVLHy)OeQFlvz>0l`tPk_^e+e=X?Bu;ybN6Hp+)>`H<8MMVc@fW z8G0vdW9@|$ep>wDZa=BPfq9`YB6J}byV$c0buDZeac4L7-C>D)cd+vEDo*3xFjnts zL+_4oAa{1}!$ac~S^WU-VcMBObkq18SN8`D?et}72Mbw`y%hUXb%2@NOeJY;v7*p` zTxdDz%3d7$P8M$)3x8b@w=8|j<&F4;-P?w;Pt}2tH#493q*wquI2-MY&Ju?qyO_4l zBW|JJW7K_F%=Gl*>DL<->q9|WG6jgTeeJ(Ze*UA{WCzp5UU0seY;wUWh-o%s@ zymyKEd&JrRpG z>XrDY?=rKZRs8TO1NKed3kq_ztR(;Uv< z(T=z_4MDosoUOL$;*$RjidwxdvtRo6IT_E5?B}{+q$A1*>d%E^$?sUMZciK=a z$zX`|Ev~9%HVpglkomqof_vt+p{%+D6a27YlP*@mO^F24$qM zVisgSzD7o#vlE>g84elJr%>UbfX^wbK>vnJqRaDHj!Vwx%4ZhSl9A@D>4YgWv^vAC zpBZ2+JeQ;4eFlu<&oVP6)e|qzN|b(fk~4FVVQ$~YF^4_d2({P%SNs0W$n-#Bm+~DH zHKZU@R-Z+g+rfXi+GJw0I#HgQ4BSX>Sgx^)%ywDFsYy2C7Uxh_QyI&ZTPb2vs1*|# z4&zQd+r<)}1TzyaC%Wy=5Ntea%?7l~$=={MqQ~{8iPX+0k~HfPH#EZmgOxhyKf_Jf z^&y?hN;l$UMcJg&BNdZYzoD8%&dl5BF;giB<3g(hSfi)IaB+*MslAGhKmV8S3Ph9P zo8@3k*h;8w4+ZNQIo$g&fal3HsIOj2?j> zuN@RU7`Q;gXDMPWeMwqBHjfs<}vf5DmQWYn^A zJI}$si_?jWYdX&!pG4078AI(_tI%O~8E5d#kA5jzk1sMNu)A4J>}<;_RybK7wk@=P zXuAYKPih7RJUD^UOT8hnh0iIkqqwJG9n-w235yCCcj$mKBYB4)V3sl$Pj_MT5_kBL zYC$|CtFdEq6P}QsLbRt&hyJN!L3>m$UAd)J;MACgWyzO$*PJ(AR7gPNy~?LHrlG1# zD>vcOR1zE5i2kZ0NTb6|Ec2THnm2_kF>^EP{2R=!SZyU=-G6bnsG;DN-)~O->M_Bb zYr#11)smUiW{S3)55~?Zk7>J{8hLtn2zhQ9jfV@oa7x5k+`HTpG#-8-c{}%U2Rkm| z_mTp1@%%|oop0ct%Ln7#mN_g&A|2FT&V!-J)q=H)f0JalFT^&mjJ(j&63slP&Azuv zF!k1NbD5Ye;=4|xl4(i^;cCFv2Kdm3H`>TtWj;UZ5 zLS7LXolACBS`m%93w&O1B9SaMrxQM0f`fH4AWI>L9r*D-X>S5e)ffH`Bc&+HP{vfs z9LjLd-Y4@^XcP$*5}Gtg(qxRxVyBS?I~yXMB~tHn@GZJKpta6N=#c z#m+aiU~(NB9q1_IPv!LKeyVHWsmCJ0u=oyjXQY+#&DP^Ebz|tePbLncjM0wjQO5ku zd8iYa03vC{sQuS$yrkO{Y3HKC_X8doLpfK4+l%wJWlo-^la^iDP-j9$!DkX-6pT z1|s$fouw4<12;eLS?Ujy+`aLhAJ-_S9D8u+;qY)*s*tX`H9B}>5FPLhgN;eK99P|C zs2kkHi&GVcZIVgQ>mCn#EF^J+LI#@N9|w0vUl=)tsLU{H{RCGNP zEpwR5f8g*DT8?=W&FG_GKQ0=X&b`ZN`<%j$A3mWV)(*eUQO8pvIId0$2J@bV;nAYj9?5vG8r{Wt|}JS~MIP!X%}HIOO5?93U0r_5o7-Yy5jdk;g= zv)?FX_bTRDnG~Akb{*c$m8D(R+`?{l5|rhj3<*$OYjp924s3ZOOEjO!lA1Xz&%MTn zq6XJtdCpH-AS?-wlpOG+Nyf~Z1WCwMl*f(7T9|LH$M6QBCHP$DJG6OH6gvKrfJ03U z>N2hd)9z@bvS%e(W_=OgQ_SJlE2Tjnr?p^lAO>&BeZ^E9`oY+|$VGE&I~j?=21;|T zB7gr>N6xPuMQXDQutjD*`k3I(tY3Z!lnZqs#3LJRJ)V#6K7I}-`~s=hCcS8#3&M-? zr0~nGK+?IAi<9lMMtSc$sa1v2IDN%SFly%f4NWWj-ZK*A2H!{f6h(=L9nEN2Ayg!* zfYxrUhqND$d0OdhjOiV7vX|4ktxc0a7EiA*U+>>xbV~1|!cDi(CzCjIE@B>fRoINC z%LL-jA3s2fPaVilkinrNlW|PY3_KK^gGXo5#9xQ=A@zljK-tSWYefMh?`p*eK5b=Q zx*jkjy7{QEUWdG0mqoor(pZq=m$|higR%jP1590gc3JD2%$G>wuztjvqj8D^rkKBL-1a?#3q87SN;PCi}s#FI+J zXw9jKIOh>}?+rGi3LV$;Th2-1Es1<6{6LeDap}ws;d*Lqju(=(ISq?{tt8iWyWlg^ zwNNB~6aI4`otjYQhi_*_;P*#l>EBvgQFxXMlPO#Xufw<)S-c@>>=Yo8T<)Bku>+Yh zBYTvD0;(1tSVT8Z!dI2>qJAb+|Ovn_>T^f zwAu@jiVEadqz!S%o=Mulf$($7pvGJT&koWffd*!9pyvj&a%&)TRA%9t0133e?+j|a z_8z{b36N!{CCJh}i-^8^8fkIUWoN&%g6tc`OwET5p1=-Q-qvM(aFy4mDQiMt?t)AXLHJ z)jtTUq-=@8wrFJGK~tvdIKKLf5k_Y24s>+Uc4qMMUg#tjP|@6r=y$gqvvprEwn%=C zCw`(yLV7(qr7J=1cbV`CgT%q{&w6+-upE6nq>1X0B2%_#9kS9nO*t|&L{GSmG?l!O znw}|Ay~ka{_?K`y{3hrp*`Q4yHL-tyASeoJ6UoIoP&8u~npAg#I${+CcK2?>o*(l_ zhIIzjA*O+%i`0m-*m|<=unuWPS~$#cCb{uFjA(aAqgc(4O!A{Ps5k9`s*t0|`yx%$ zc5r)_N+YuHb{^SxnV{v5<55ts75;QJ3f1lu#bKIf(bJ19C@|m<6*(44c0THcl7L6> z*e@EA!ejWa4aR}8$sC+~r322ooFU|$3preR7~}$D_&*N5gwx(nVP5JbNIRrJtmQva zZ>!>w8>5N7r{sYYr+uWnsR`;jCg7A4mgHvEJ^uUN%jlx+3F@?y7^iQOhTlaTA&yi6 zb(MDj?ZyXV?TB;IIqjr*0%oL2dQS@Kt7WFG2k)gsBV}XKIR04f0oX zgvzRL(733NQh%tEHcofpTyit_@Aw^=S-GQ9dmiz1y$F#XqRCMcCD^Oug95L_5a%2< z6rk*mlZ#_vI_GoW4%CCgdyati!7S9~V1;Vm4x)s@As97zL_O9&KrIivO!)J+z|AW> z(&d|iF6FjC*_3fCPMZrKPU`a#i%+7;)vQa9f5Iq6pV>K2fOuw;Nr9gWppMJTDO=PEa;-5QnZoJTs@+`B7iVj zQ&IM;%l}Oa7NP+Qoh=V2&q*uo3gdx$2J2L&S8TmuTI48 z<91@T&IXu2x{;x{`}v#u(jfjh zeOB|-d#|2<0s`-&+QA0u8;u#1a$t2Bl6$ph}}QFVhUILpf?9Km=pf# z*m}eXd1g!{MFMJ6yNw5RTH_$8xZlC_?=oZtwidy;x;RSMb}|~x%0{+#HY2fbfzW-P zB75?$9(S6|!0IOO{xnaFO&Tyhn;{d^F@)nrj( z#bh`=Zw@?tF2NnPGW-mVyW!#ZZ8)_yo{9Bb2fr74G5y^L&oqhSd;66zuYO7rnO!MJbGa6^(j37V4qLc(mjsuC z>wz_{PD56|@1te~H>kRh%(T6CKu7#-QR1J;uzJP>BK`6!BXHm<%6wdc_5~lL7U~|S z%7iOXQpR6x!RzokGLZ3)-v^1Og&1oERc5h68)XvPL#4TuBZt6C@XBX8_W$!6DVzu* zG3kRm&poc_W7jn_%5i--aKDR(+avII?}KQk4tZ!%p>0%g2b2R~5vL(_%@i1_sZ z{W{|f)UWyz%odGUAXlam`iL1kc3y@)ji(_bQkR%<+D+eKGjuU3iUaCG#%8FU~{SK6-+=qMiO>>8!M_j7SJ z)h96P#trJF+e|DYu$j4|RsuS9VQ9?9owAl#Oad&%!#SP;QPuj){{SI;dXz>}XSSii zmju0TRKq9JX5vxn5Y%TWjS%}AMaa1l@v2YISi6$UxpIst7?=lTuWI4>1s{}P@Rm_B ztD&5{_G1C*>x^>Y4!lKI8NGds80FS?l=s$U2t7s2%G@`M;YWM2YJwlxd&Y=(eBi@G z%XDb16M&5NWOTW_lev2+8xEYB0x|;Em@6ZKWa|notp9E`90{}}A_sP(1gpqU6eqRbGHqX`zTCL&L!kl-^7%(`tiDc1!y5bYHLTjQHR=#wdbeTFdV zA9EslKA}vqiX~b!GK4Y=VmR%(jj+ZysQ#gV9kMB&MK0K;lihoSsRi$?sK~1VaQ|2a z^W|JNl;3{>NzabJAg*RU@#?A8p=##HDtGSe5M`X z>E3?B^t@Zf2uSonN=XqF9Cwx3QkVuk($5*q@!yza=SM*MUFUD!Pz`P`)5t;bEOIj) zNJlP?nMjsS)x(w&pJBH29N1hGhW!^E z#h0#fx?ah#sO+OA*ai!cm8;t+Mz$TCA>(MP=f5VP<@j{wMXc>_(yTw&v~Wq8*PCFWlD7F_EV!1x{70PC04Leg$$ zG&d`o_}tNDpXsME5(5U*r4yQz*4c7&#;py>$8151e$NsneYx}1N{OdLU zvHh+8vH!dMkL{oO@7Uk+xBs{PTLk`9{%iYh|KD@}r}jV3`&<8=_wV{|``>+m|JDD; z_P76kRsR2u|98#5>-~4lfBOHe1^zeJ^l$q=&-uspxBqY3zuN-;TJL{n|M#4KxBol; zZ~wnGf&Yi=|NFWBUE_Zm`@8;s=l`|+e=Yy4^-ulZb^dquzw`b*_V4!Jv45BUY3%R3 ze~#B9ML zocBZ?A8_-)%gv(kpx}Pw=r#dMN5_NkckW#O^E%$aUL#OTlSb#~@X;^H9_niSVWjG0 zj;yjG$mQ)*Nr|f(^r(AL9$p?eb~<;~DXxX%dsC4$&zm|@uLScySc2-o9%i6&53?cS z9W^}X76^aKfYFa0_)RhevsdjzX(n;?7l+L`9Ig>ejV(szD$Aj#FC9$%ZTSg&L1M(k zLuTI{L#bgKvE?yoTo<5DHa{w)3f~>V^`5gKS$!?+SfYXlN(ND|lpT_}Ek*JTK0{H| zeY9e$0P(qWoU;R%$D=aJ>nxh&_sn z)|SIQpG2ronF?|VDp0lnsOpKH=%p5my!abXt$H&Re|9|DyyOvM()tcYPvtN#^IsV0 zDreO%mX%>D)HA7DUm5DCcOXMQoI&DjqLESP1*1q-3%Ql@v5B=?kii`G|6*Phn9mPWr@73GXsV zV|x2K(Q3tYxM9W{G})*X5G$9TBMiY6)r-pMVZ1$&qMt6=I`t zoM^ixBFtgPA9!y@xiw44GL{n?{5)`6g z^+5@et@;!Bw{M}m>%;NFw;VjQfplq%+UvZp&Off4o+RptMSFEKI^REFfS$*2X~ z@%Dg-S*h@}_6*v;vkQHxj6x#Q<;cx%UNEiGmv5ik1Urn(KvDV%BOH7b;=MAd$u~k_ zg?}bYa91F`H7N*lx*PlFt>Ge^-+5J@&dgr%cE0EBEF>}Fj-EG-@l3n7qvJ8wOxe5y zG_Ak{{dS25CxHRv8()n2<~ZWPj4|Y8-vfTH2hbcBDRkf01}~M4gNpF0@X~z-)?zuW z^~5>&(nt}Cra0f^L<8babVPwqx8pyXOmU&{bv$cu6_ONw0ug&=kZ+6h(1!lm$nn}2 z>g}xqC}g+@8Hs10voEy(-fTk6I$mHJ_LF(hW&zs|%!M6t#~G(ymRY6m2@A&J@n*B# zxK%xYdfZ>cmwn*R^o*a3GFm6&vJy`!f6X~~F{2ngcoYlPdo%c@5g6fJN8sX{By_Xd z02Oc~xjtO{T24ygqqM>jJDRL zFfzUfVt=eeg8LJw)6W@dI4Xqk-8+HQ?Y|3_nxVuyzY?m>3!xvnF-Y;!e4@DP3zg#h z4jq}J1hP(3aisW0NV_qp0ii^5#t9TOFoC33UVYrf+Dc-#@W~>@zlb8T}J~)p^OxGsfh8G#vW0jo$osV`TJglF!#h)a- z-AKGXWkK4+`nv9i17Ptr6;2d${{g;-DUHr4c%SMO#Eve-U!?Y-M<=ThFLENv<8rv= ze{wlYDl(*P;zO*z`82c1SQUBNpGGT|u7wfZbNE8mPp2r*y<{_1EA--+i2j~e;MoV9= zz#_pluwg+T|NB&q|8m?+bmMFym6Kg@f3ovJpiW7f5BF)zvy4k`^&JuZ2mq3{b(EY66Cd$g`@0raN46HgG}p>GQyjo&j_9i;2&SzgKmxtQpfML@rzhxCN)14t-GJk{8U!-o~`sGn1gGZ~|I zA?9{2P;(c;3=d!Qq_F^6d+kYo*+~>qRZXQRkE4E>_b?=lf_ZVqMl#LP#BEk42&8#o zms<BRP+fuymsp(n%TLNR$G4x+Q$2l^pm>e%N$c;U34A?KiY>9Yd26vykaUj;WU4a zm;&$?V;C1P6piODEDh$XgU;D*pm$i}yvs%*N9r+J+*yi+9u zC@+R;yv6dDi)^Q)!iq@g@HG4lqL5i5r}ZV(Mm1*L#Im1kAS|qaDPDCDWs7Q&AiG28 z`bje&Ut^JvRwO*k<&lfEb9reS8o@o^iF{hkaouh=gr(yO(c7O(@lppr+?k$%6Wu-{ zjnA{lc!5GP?T|33u)0az5u8OFlX3}b(9B3>-etO{L_>}LMieS~94$RF6Y0;?LLVnM zq_{Q#z`!G+lcH9-h+=hZNN8_ z1j+XPuM}g@$M1D)r&jqV!rpR35+4gQug-47bv3$}L|r6i6&>Vy<6Tl{I*BxzI1mFz zJ)UIW06$o05Ip%|j=4 z9x1w|iHydka2iu4u%XigF00K3O!L5C|1}u6eG)eQoj*?D3ydvokJdsb?`R}i(sFPB3SQ`16lsm z8JCBQqa7Y}QBU4P;Q1U^+^ylGaLnQVXi7XJZfX3eOy7g2H_2^o99&eKr?+_MXF! zkc-9q;#H(Y$(?EGx{W43NQL)Zb4f>_0ntl24D&f&v-R#Yp5}LzQGA`otSQyucd12@ z88JT0+W;0^?wzNmX0{>63m&+0OaRvgb-`?dbjGa`qYRF-ZG4F|2|6iF8oVT_9`R%n zImrUGG|qw=%Xx5!9}A*ghj5bGZn7e(uC6}u9$tUqFgVI&gVfOV`kf+g;n?f}UXUY$ z9Vd6;(9|PvV*7H#&uhn1b(@onUPic3nPT?n(2o+3@Y+DeK3zw@1|E=r)(En^NDC_F zO(ki2CKG*SO_JI3450KljwmnSy-+4Fk*R_e%9~-xuozF8e;Lj08pUtC=ks-FAFO`1 z46@VB!T-k`xSK6$Xq2i*-u;>myQ3QL`WtedmG}Hqk*a$RO68oC~zxHO~3ZNo)s^oDjH&mo_8ZsyDx>4WsT90 zwha`#Pe&U+9b`h(KOxnQa&&9FIml;uO9b z7Jv_)j)GN*V(3%wB68|oCc2zz25;=kU@(b==r=KtF}@fTh&Q4~uO1@WZYg%1n*zM&{UhmKokm z)1+3cmnxqbfUJ_>`jcovmg143(yspnvgRwvBEquggYyto1m42=NKHHT?A(a7mmn&ZXQ+E`lF43~N~ z;d5FGvE(VvAMxVjj&-(Vk8v%NbGHzs)qX?W@}YS1x(Kv?ml00CRm&Kj@T7`3Eaj{R zG5ni?c=$+D<20gTH6NqOXSlq7i1KnR z!Te|2>wlXj5sO`2khA>@Pt;r)Wi31pC$@fN;-wt0GSVjRD&%mi5yu-fLy>gL%;)sS zqoDpt3i`J1CbHV;!rNxhMlmPfqKH3-(V}`Syr8!lO%4Izb9$%N(z1A~jss5goR9e{ zl(E6$AFIx{!LvTAVCN@y(NqU#bZT)b6RKN?{l7Tk$?LwtsZj+y z_3LJ@3xa*Dx9`T%wz10y}Ofr$=ehuvU`V9Q2T}VQIW`e;zG5k4T7S_Ky8=bT)hP4u^U%nw+|1IBj-B_gsd>KG@5~UIw$M93y%`XwLWlmStc*o zWIE=fT4b@1kM8YwNXlK0ptD&E`ALx)%;>TZbYRkH;u^7#=C3fKKa0M>BI?6Tta)hgmwt!jAMlPqc(mm@mi9EaPq@26^2ZXnsLR3yW3 z&t>T>#)`L28fmz{2D_IPD6{SX|K_inz8rt@E>qzO0%nikdc{Av=`toYg=|gk+el(lk|>=w-$Cs}9=SZu-bj6z!<<&uAg{Lj zd~@DeW=mihyt*I&zs@~?Uzfs>#R5TAx#N^HtBT3a$PCreV7ZU-#HtP*s8!`S!X8YZXrDL zoJ1NLorzMb9ojTaAIny7`15iV@@B6!8M*7pOwPQCHbhOt9VZzJ513 zU$Tiv5K&U%)CJnRe#4z;Gt%{C3sIWYNZ#*Mff{-jsNAaHvvn6xR!Tk?_Ewh2K;mPA{UN=x9E=~4WMKOsnX=s2;fN+*SH5Ckv(2?r-lq#mca zL(!g4xb{qil&$_onMX8`eJAV4vRxO+#hL0PT{{!iOb&w;AB9<&ojRml*B>ohk%iAc zE-sU#^`s8>D;{m;y)kEv;lW@{>%KD3h0f8Pgt zQd(iwVqp@hKoh0fUrddiHtE~jgSIVimb~`}K z=&R&wM>RQ}mqCgZ-!L;oOo)_j9K@3~h-a%quD|tTK7S78@BoFp+kY-o=DiId_bP@| zUfoaPT~CsG-{hz#XY-L@st`Q3-2?8vZe$AKGKuK{nh_x3pSskz8PWAxMX*y)m zxE%285{EsxIn22CPUK|6SGaC;lXwi>MIpJj(YwH3NZ{IQrhLv>$SE2`HjgjDGTQ(y z&eY6t+tgFnR16WbOCBXFOd&xfrIeAe8a`B`LDuJ7B2!wkiRS(hUcmRMq}1dpv$({X zH??vTh<4lo*}nCN(;daWr>^7Gxl^eSuZbkQbQ;+0ZAO|u=cDp~Y$Uh11>YZw!QXsl zU=^!WbV<1m6&+VVK8XqVljCpX>262Alby}SIvL;d`h} z4^i9Kg;FXT=5ank0%Zrj;Obw3R%cA5R;lMhi%b)!ELH>G&u8J6k|XnVawK|{s0H~N ziopI%N3jJ-h-Z)ve#v`4J+}a_FrG)M3^&34_-xAWWHQuPj05vY6G#BH6i01O0V&1? zjZ2q-wi)fn!8ZwF?wvwr-}AuB{TPD-CU9}bL4MhCMf~f^d`=820nt-FqjLkA82lvY zf`&L8+B1wye~!m%1zV`gT%N!bpA+cDqkLp+DS#KZyhkg9*Wf01d#t{_5S3)Uhf^F^ zm3#ej#=m+#T9Lkh(f^~y{a>#pAJm1>GBH=Y=e#wxZ&XJWpRV$^@p34Ic>?&v6^?&y zdM?^uqlmQ@sp59J8xv_~u)7gN7>_I%j<~=)G%Nz0ZZ*b!P8fu2$N}?TtI6JmuV`50 z9eP+jj=AD?kNI)kg}R^{4MVDDsSCM|1cjZ%Bg&=B<30O%OF#%de>wzjb_o#AhC^Va zd74>g{)@59R)+59CX~${Atp^qm?tZ4g;?WL@cKy)dK})xf6nbcKgOM-M59(RE2g}L z+KngClBQm2X7V_6yz(l%(6vL&RiEMWzFV+nI1PQC(Tx%cj7ewPc(PkP1HR4^!*<)g zqc`^yK)1XQ9l9sPzc#HLRJ}Qk&!sVt?k$5d-p8Q8y%I33nuo0oQy}WqJ#3~gh)qr( z$MOcUaB15Z?0;HF9@&&5iPQtsrMfmo-kHbo{ly~h+!%C#1C)sDSK$jki9&My-$YTt1!W+BHhab$rl5s`j}430S97^86fbbAW!xAnlm zgM5&DILf@Tok^Bm%_q!~As3GGL;%7!o&k&X+T^ir6wl!Z$0a&N zgyV1(Wy%j6gS*%Bm?wHd96$X=lDFg#aS>M{(cLp3J$*K=Ptqjv6TCp_k_jpH{J{@t zH6+WAb;9(=JjyHXI&Z!_r&Yb_BGPRa#^`nrL`8^^RA+|7**t-`>>y^#f@us-cpE58 zNryXiNsJOt3nCY^Q^RLYP|vpCq9TT3@hS;Ps=(3|Z{0|8ph z=afNa>K`knA^$V_oGVMTZ#m)>_97GKMm+2m&=wTC85c=YI`e ze^;8A-apThtR3{lGg)+;YCqmP5YP6yE@Exh9b@;X2sE7Az-zcTD~(P3n8F(N1ki0| zCJh$jtZ1?X-;tl8pHR==j6jTh}>zc{wDZQmr=-@$^!K~b8cUjIVAC3>=U>Rc`hb&GY& zaAhBjjIj?Kwb;EM#TuMF6<9gHMxtAJlJ#nJA?y9aSn?*C{e0^phe3 zsx+yFf}{J{?hw5OIzE#P;<%6cM&jArnI&w`(wPmP1+TEX^4r$MtMSil1GaMK4pe^s5H0X4lC>2r#WEAJX<6a@Y-H{j zeZKh(dDY~>8ds*1>cKBW;D~O+^?4<%#Y#Cg?%n`}r7B=UMgZ$Y%-Ga8v4+Y};f71Q zqOk6hVYd2J7&+ON$1c!~BX18rW9KUTATO=zSnq?sSd9}0*yoE>8`hrYu-I$}Ngh2< z+EwJ)k19EA(WORK$m<5HuT+kNvZWimN=5Kl1;hq8*}%uGnK!x+3xw|?ocUNyDt&Pn|;PF zDu!4WTVeWx@^4mSM;Uowah*;3X~@2R<3f^;tY?k1KC$@=l}Of>nRNIMRW`G70d4fN znKsxWwCGOGtOniqJM0z3F803t3AW~45xY2YIUDC4#C~atVIQa!unoT#vSarau@g-0 zvBP2yL0aJ^`Fd-ZEfQVEnr?qVHtd#Tf6N%ya6@j0eP%VrE`OfFPDw9kE!8ydcYwmLMO zNSEwo-y2V77wR?R>8uqSZN7lL>amBlEgIKwwsaMHST~ZL9I1@%|Bz!JMn)1@18w^J zlrhp!wUS-$F_)Ekf0(^@`6PRF(VPZ_!@tSux=ZW~b!j$U;0fvck%6lmr?W3IXR_O3 zZjs92lQ>f`kG=K&H+k1D%nlrpX%N)tAb0Mku|N4CtD0k_ffdvm{R~u7N9C~7(7><}{>|+nCVg4}Ju@c`S4D&kM%>p%mrj zc@(xZ7~pza1*{d54dPY7q)>JoHkJt{o4L5-&*K(2>lw#U*Du0KIUgmCQm%a7%1LZ} zix&Iyo*$WPH3DKr2Z`#CDzCqi!)B@WqS!}*)Wb(yywC9f{v7Uz`s-RDt3Qh3S*@hM z?i66hZO_C@LbqV~C^e$%G>ar2T}%!aj^ZNzJ7~Ol3Ig}EKzQUBb7ag4@444ULQ<89 zataL%YeSK^Ts%Zqm6H^wRYK6gAnH>$311t}&(hihe?|(3!rR@%K%W^T>X^cG`PRGW)Db&wDlI)fQC#i)8HnH4kVudH=Rs;>wPbbdmc%>v7Wl52CjOK+k+{w`Bg(j& zH>1`U-O#*_Tm-He#a%bXiHU;PEXWa)<@x9q7Z2}?n?`w;02~@M!9o+qKvzQo1->*v z;(^s*V@g2z+|PRBx2MRA`QEj6*Ye1Cwg}8GZNuga3)2*~l9TH{!76h;*%&{Qyggt~ zx&fIgadI^C7|uS#;$PcN5Z?C~GEinkEpeEMpQvOqDvONpqRx19Yi1#GuI@sV9!XBALzCa>koD(e7O7fg(%s)r;eoE(Ox$J* z+MxXeKIJD#UkgyCYXyGeH1!dD1n{c$$9ox(_$$ z)#FJI^y%x8rPwqnl3o+32wn#M^o)CtQD~eweQ#SY{^-0OEqEM-PiX7WbRogQfjj7Z z^PK2s-GA`r=^>aJ703Q^Gico#w@L4L588h$0)B?Bgctn=_>hqpz3*-e_7C^LE53@+ zM=va;6~tcPvghYequVivcvXQ9yfUG`yts;^PEMmM^Z7Xc_eNUlS`vLi?t)RzrzbeY zI|@%T-Hjuaj$uVbPL)GH4z~qQqLy3AA%lc1bhoKJn*BV0uH_%13&Leth0*&spf`rD zvbLnchWj}GZwbAs;tXwFq)$g^oyAPuB+UEhNdHL5#&;~=;x%s@aM_?D-Jz2~P35#k zX=PpPCfkbkW!TWmR2s2@TQZ*V%p2clUty17zmYRLJ5bIQF9` zZL-Z7Cm9FgHBDdfnbz^-X@nGA|Bzt*MicsA;uzjS&&O3M*7)_#PV_qT3QFEN#+<91 zMIVcrN{=rSr)3OR(5X4cj097X$+xi8ILYxI`r~{SuZ_Kc&!72%x209$_?=7PgWg); z$R@FOQX-ZWamOoNSK}QV56_LCnRt)9AiYIJ6_@vgVo5Jc>@T*74iQ3hd}#rG9Xkih zxs0PzPc_}sr14e9J+#N#$@Hbf6~th&E=HU5k-v~J4lB!`hT3{Dvp|wQ zySRsXmO6_b%0G#WQetRvLn(UIPAQCEd?F^}d}(VQ(7}7t&Y5IjE^4#+suuU5^a zpN74}0g;&4hRVf`g^h4!_b{GNxB{n^8)2Ol<=7zo3{uKZ!AZrN@ja_Kw81@n zI>}iJQ+aps3!PfrwW|cHF3ZP1KcwKfiPp5o)JXiKp5O!fv+xErV_ZJA1K;zI;^wsA zhi4qfaB&BYxl&6{>Ua))*&p!%g z1pn}{&55(vJMsXTEop@lEk~GRn&NbyrwA^(5Q{#Qb|Dg6PJfcj$EH)RQ4v8&_}&+9 zB6o2BhwXZcuP`I%#r#%0SLQP=4=Be;n=J7j6*)RtvIP|{v!EvrZNr(-YjICrAng%a zO>UKF)8BqP!9#ARaX^v@9p)N}_jawupQ^=)s^n{2GJ!vN6d!Dcz z{eN)LW>K8?W&jhd4qQ7F#Ml+`XuT6DICJPKJee;<+Y9yKS4vZ0L1YN4)Ln*E=K5nl zp@+C(#bs=pVoz&KI!ZSucVW2+IXHdcG&Z^dPYAo>41R+0hrXMa5+)JRK3;@;$ELX z^re9|9Vz6a0o7FtP%z272iJM2H)ANYSMh!CMeLj2l=?Z#H*#ok5K@fV0 z&(KIwoL#>^l1(`gPwzdZh1$L#dfTUqxP49=naAmn{AfEi{pvs|IZ`XkKCzxo`~5mdKK=Yg)w@4}jpL6I(Xaqo z`h@`dzCeN`^;gi^Ho3$g#~CkDZ$o#}o3KMh7RYAgijS#`@H=f1J#!2nd~}`uaCs%G=QV}B6p@dseWRGd;&JrI^YQpc z<2v^3g}JQhqfRpTzMXzQISQs0T9N%xrC4^W8QtO`f-lZ@px?4(&}#b_OIlRoT8~LY zezqE(6UuR3Ocoutk_24b=SOrqa`B$i6lt6kPtLDO#GAz%@i*oOd3Z9M z>RuPd9xpXU*S;vTH&WxtyY06~-t;(}IbcfNv|0se%~2qBcsp5k_B?)DDo5vSm_uhT zQlz(?+l0-_z7ToqF52mX6K$};2ZgPBk0pF`>E?oBEOq(?*3THi7+o-`+1Wuo5MP7~ zjtS65mK2liH-4c@!7b$MpC1&|RuFBEHB_OX7eu=GkrBx#B5mCama|WhT-n*I!@*J_ zqpVHxxAu{`#ydnqk5Bd|zQnKgRN(B&vx@eq38OzAZDXwig4kbeMI>TK zggk3r%DTuKq0RT>Q47|=6|OT`J^vW2_ALbW(S0N(h1)Mb9Y9_(@2EtXn>f$coXnm7 z8!zJYUWEBwh^M6${*(V)Too^NDEo2Vp}mZY?c>ZhzS6NOYlGAb=u_c`ZAB~mJdhSJiKcA8qh z`>#KEyP=9$2`=yMJz57_?h|xU6J6aU(*9tqNJ))odCi2(E3-H0VaIz_MXKT*JFokGy zZcuO;o-S^M$BM3a#zmcm9F2nuMKuuo*N=OBNM<$DHpBE_8`L{<3{Iav$TL2xz<=pK zp=(Sb8xZjT-YI7T7SE-~6{m1q|L5YtHve$5c^RHPyA7*+?2z@pC8WAKq2rzd2ls0W-^WM#3B>#q8?i@t& zU)i`c$00 ze*)h;a0_f&yqeGZ){JdR1Gqx>G*rFP459tMqvq%kh<&pK>vq|&@j6ST9I{KGY^(t) z=f^^6p{{sg;sx+ck&eXkL83@gZ}42a4O#nIA);S4PPn7a9V)WL+aMMBh#lhkw}sN5 zn@VeY2a>CdAswIHD9VYe7OG75fSRAT_)JR*haMBIetHJvHQo{VM_R*bX?%6$jU~7! zdaV{%Y#1Il))SMci8X0n~-LDNWA8#4ZMFo zmz0zyg45;!0&l&GUv0mk`Kd+x_QqG@+;vm=x)+;iXpNQdaOp@|c}&DllvJ~dMX|Uh z#~Dh^HBmp+RyZ-`u((?c(giP0%w?9rYsY%jc==tVJ^fYF?@f0Ww%7*#F z-?LZa{$h<}G9Ak+6HjQEOS;21!Dx`@=Ld+e$jk;$-8+jD+Us~k_d_&$=f&R~ljqiV z&aubq%D8>57Z2aN6PKyUU{iAy$ba00cVCz@{ptqXdp3&A$T|f!Va52tIg0oDeGPR- zHpAkgb8zEOE=={Cg0oyV!+xD-LPd=Q9}_Ev>8(Tff;~6!M~WO~PuwW&k17EbKY6^n z@)N`tP2# zcW?OA6hxK#%KY8QR2iLYRbZYev@%- zKNp;?vkSuSEEcuAJt+)Xa!blNUWo1=N6@3jBuY~CmS!z3L+UdVG_hBPpaH|F-X#JhK#{+^eF~No>SOssRs8l|26m20 zhU|^HB)O~w{}#oudc72;-xeV3%Ra`uer{y5LsH?v+4*1@F&ZNO9TQs43S5`%;ffE2 z^8DF5g}ZxgAa3?>co&;ZH!5XGVdG9%d({&Dc1A+?Na+ZiwN;2*V+Dgd6yZqaGFavL znVOLyZX!_w|ni=;HUv3#lQ|=TAm+RADvS%NRPKX7cgZ;(Yhh*84z9V3$#En~U zYf{!9ESUR^hTaZG@!ifG@w-25!V(=-Xq~r<6;!pdv#zz!5Yvk%jaFyF3g5E>w+e;! z;E{0Bb}R(l`zL%X3>4~)sNsBPIW|V4Oz5)H;hScM3!9df@XP+W!UgBO>|F_f@0M8D z-gsH`Z-TT7U_m6vwD@5+K|EvxFNWit38Lvc zoO#6jL11PQDV`OO%eJ^xLEoL(@cuzKY+wAs{_wWN@Th1gEQ-s(AxTBTddrW(l*@Pd zy~fjgOSmR)bTb3@m5~tgY8Gsk_5n>BCe@>s7*+iUbtB7Fx5O=KLC{z~LDbUQ6s~@4 z6rKJPIv_zP(2TP^A4ElIRtdgrHtDcNANk_Ae7F1 zFO*F$VH>NxV8*{fao8Va-nnur%Xpd3`sS?>heSJYpOr_%(O-19wuu*n-q?=U2b_n` z%MXfPt7^g3A@PFD2wNDv@ii;xtt1r8>JS!G9t63K73`sZ1KaW^4o2sxi}sYLK+^dq z%(EhunFl*SO5s9w;@)jAi`^)stXasEH_Nb=Nsoj*WSwBQO(h2~zRd)Mj)nYms~l6XP4WKvRl>#kTt#sp6joGRby?ro}mxCDRkpbzE;q_<|cde zVI6;Qc|K*jAHl~p>tLJd8#eCX4S}Wi<);2#P_8c#)S4Bf%ol6sb^3rX{#m9d{Br=P z9DXL4E}R3WH|++aQ5DQd&aa5<9zou%i!r#4FCK z!=dH!@OJtu7}fb&lz$>0&i$5Wj@e>x(e>rOJsPE-kquKG_yz8q-UTzOwd^b=mcu&p zRyHH_5dGJ3g!=4UM`1c?@OowrTXSX({hYjsc0G)wv6a6m�N!jT_4{)ZX&R=QJhq zi&M$9>=5?AXFR2P2(2DV=R6~NF!c->jaQ}XxRSCPcMB_fDkwPChwi8TCiP@hK1**X zDV7>h=-W9Yx5beB;G+!VqLS3#<>il!o(m-+`bw_)3j;Y+XA%A>B8)JA*5MbOyA$%qS*O;Nj1C3 zZkeqd^s%y%9RE;2BL8&yQBy~Y&Us7nUyLP%PfKXRs6=j3o{DCxI_YTqJ7kvz(@Sj= z`Zz2RpUkZgPMPn7UjBxJo=TEE5nZgY%0Y5$q6}?qlg`la@g#G*Hyu;agvAdnY1``_ zT>m2nJ^sdF)yktZ!}=lhZ`?^Az7MBYW3?%z$Xncfbv)gkdLQDpPA2OEd-09%5H(Ks zBlGA(u=(B^j1m9+ndr{?E68@_qO*eyZQC@Mwq#}D#s!9OqU{oPy}2V5oBU|Z zZzj?8laq|S{+MhmUQ(=9l*B%%0ey~*pdQIb?EbfyjCU@V`onUudhZo1w|Xwb^AS{c z&X2b9u~I*$wA*r&F^PH-Y4+?%BOits(^#1EKdTpd8$w;kX;=xI5GNZ}4eC&KZ9onZo zgH)EvGc#jl$t&j$S}yfvIg|moEo~DG&sRanai+8%Dm40^HSPBA51sP!@y^mS6lJrA zwAG_1Avgy|K5XE%Sx-Rog(2ntvZFiYU2sje3*TMtE&M0#t`Ce&fl`Y|R(P=zb7z|K z^%d7)!`}>27_ZM~WIRV1Wd=fPqI4#?QmF3^Hp8MH9T@XO%Kr}Ip6?@}_D3vL?5xBy z%D!yk#-s45)J`b)PlnW&jHW+5M?j|eEXG{Q23h-VrkavT4sTUxhMN|p&09vx+N??8 znHoIRxClWz5Avm6br=Pk$OE?hUWX%BmO|Yqj@A(makp?5Yra3i7g{#FW{*7?ZH%H1F&UL7 zs|m|>_F&2;MM~Pb2A7w|uw|9)OncWrNG_a6_E(L?M{9F) zy8QmQ7W=#(Zv3(9bl9JN4$cpFhkL&c=BOcS0~|}mgOX&0g)a`_ z#rtJ&F6tz$S+Ib+-`Edk1Kg_4{Yn*O={-Zmx&;_gsR6c0JMrWGDZ;5WzVu+247qAI zqKnri7;$MZreD34S$sJ~kF*L(nC^GY9<mCsI91Jzn&_!;RiHdeX<@ zkK(dmb>@;9hugR1pk;?G-(nYuAIAvn&{#RzJnzGk1xY*3A1q03?=>}`Z0Q%1<@J3 zXDnh&Ax_L3Kwi>Y-uAg3^f5UZz0awm#wcfeH`a|F*(M9o6_HRReGmMw%R|a=Y{m&% zzF?(#q3YJGmpFT435xYoF}>Oc+ZEqn=qMSus4&D}z*jT~&(`hu?IS;dMn!+EC z6DSj6A_S&g6uw8~!194bcqzO^EGN;0d-*#BrgaQLVpf6Nsc~c|8N}zFPQs(Z+QdqW zm$R8~w?bY1D1NpeTzVHZ6CACF(yx*KAaU6o+7LS%$I3Me%}qtv_k1(<_3y(a2o= zR;jqx=m@)=EB$(>nOJGFE(Q4nKt|kXFq7LVRD~>Ny;hW9?-KyOMuqZ(zgnbo)LQtt zT^H&H#8KDLrPR_-5i*V#aMrI6!M!;AAbST(2akZ!H}j?Y{1nzrR$L8S` z=}mLmB|MW?g8IK=QOlzZC)}LNcS}8&$KS<^wXAa3x7pH6ds-YN%{wo&M(!3Q`=hb! zaVi!`GZ1zvD+Kwy=Jc}KpVt(p(BTMqN#@5Vuw+0bCa3MK%1pn6$DT=VJF!JiD<_KIgOEW$05j46I%DAk?6&X>(_CwIv? z2)ZUMS#KFga^I9F`^9idwc0OOPPApu-`i8Nb`hF?dx*0>J%B8m{^YS}J@nc66jIce zQNGGKUm#A5UH1!s3=VNU>!uI3Vc*2DquxN{!jrPXGKv4YZ`zfq#(QKurtWX^g2O#?H2)k)B#`)~OxV->-rnXCmoqpgieankR}cPY~Am zYEs*rO>}#}Ybfk#z?QKJ+%x1K>FkrCp^9U%b#RNwuBQN_@4-i7OT zd9e)FM}YnC)MDFAIFw79puba#=NN(xowLi4UTq5tQ2 z+}zQNX?R$|99AKCwi`;ks|@+DR$2Z`dV5yuyM#(DM~JK@WfQGV#r4OvxSF&FFS{UI z_^bFuxO2-F9*tg(`}0TBVht0D)*MKaJ0f{y0b*416;_h_5Y}w&Bj`PwhkK82>J^YQYz!XOdJW}E!?B@LUX)uzbVjZ>j=tNM z(xbZZ<_Ifpx6yzi<@WG?CR=dkG-u9wB<$DJTp~{zlJiM@rr0PCRFf=w#idB*7xZ9eAw%X?~ zV?{CC-aUz*zj&Lm5Bor(agQZT-b@SmWgM7f0-?VRu}kg}%8N#Eo8LpIQaM*RvFbTG zPqRl$l~A^VpBGQO7>NqXN%&)Y65q98G-;$+Qf{}jOXR~leDLQHCNyb*h5Z%Ez1>6> zj#7Tjsjb*&;U7G0T`jyV?!fzT?p2i(NO6i*@L=>IY+f{l?u3-!1HD8{DFO zW3A68&3%Z?r!uhL%S34OPlHTPV^Sq0>1S^vq<))+#>)yI;{8r={#Sso+=$K$7jdtk zQ}ExDmE<42mL{(qLmy1*xPc(%fl40b)6h-9N5HPQtmhD@sH@Y=pxe-0(x2W0OLe_Ti>dv;3FzJK!~348 zg*Q76QcQZX@Iv-1d(_+mk<)+<>&91&yraxNgg;~o_XdhHN4!JH*CDw3c|OdR^3I*7 ze-I8F^o7{qMX-9|7$H$L3p}IuVClj!JjnL65a%WlwTQI%r7~GEscfL4N#DRs{v`Gq z=!b5vBzVTgiwxt2)40`geB_TgxMj#!(cZ@CP@0-3Tu|92TpuRK$A{#?iBv;!n^FZc zF`;Uul)+m6YbN9*Ftzqv<|hTk;9K}W7&L68>N zh$mq2j92*j*naNX@CV;0tBu>H(uirU+BEsSB`Qp=A-$)csg0aym(>j#G1iMJyc}qu zmJS7oCZp}@5Xr=Vo3yj9k0jIO8p^r6Ausuzl>V|1GbIt?yHk(gmFpQ2(|4W{xBc0) zdW?wP&HYKgQnbj}Q6HanX~14x4Zix*J2H|vLj8P>)5x$adhBCFH|J}^qq?c&FgRKG z+1g1HwnovOVmFd=GL+=M)sP%aEuzo=teM+cpwppBw03(FZ0;N)NmUES;`}Gg?#|4QdaXDA?>Q&6q3Qxot*J#$G3hur`?@_g|&Z zeaez;$!}=2T_H)jf( z@^P|U`<-4}_L3NO50fMw-$Z}aTj}$m)#%@?Nfo!YP}uq?GHuh~H>7@nKYntg=osyteTj6;ALHu2&2}9s=VdUqfH=Lt%C6f25>0RPs|^K~h2%q#65>Yb{rldK|=j%HWxl zIPVcz&Fw?d(-)jEzQiJ}R3z8KCW|M|jUa=5apa)>7E`ZFe}9qT^ygeGdT!iL4sB|b zZ}FN<*>M%mDdkYcA#bWV*h6PapV4w(FPu-`(OPdREq-&J?%H)%#rOM!E^k#Nb#0$$ z+5Ul&X%FNiZnql9p!^}Vp5KK513e|p)3b<;QkAUMZve;*r-0{y^ggN@JyjRd>Z=XZ zs;DSQvec3IxO!4(gNQdt{ggEgJ0#4(M)IvVkJj29rHUJiXx5V)PsPlB{7OsGXrf4#s!QPc$r4KW-9(9}^U$O;obrYiP}z*Wl3Bf^e;f1aN&C02&B(5a5A5CU~$5DH+H3e1uLyPgfDO!3T4yrJw6X$P3@$<1{5^|Pq zuRP52UrnZeUrqUwtRKS4-)G2f$5br5A=U2Hy(n5~IQ0+uOc%||Y1AVvy7#JwD%F$e zxP1W;uc8%(MKt8J9hF|ZOpj{iBoob*skG@V8D^PDUi7_7(^H?)RhbyHeSV5Oyd8L5 zUXJMK*fMY%@}IbUot7Y9xdRr7mARz9ERRXJ3Ky(X=+o_=9aJR^thSWNBXHOJV1icvzE~1==r?_`vWQm|`FY{>rZ* z#HsFCBS6R~N-|Vq)nGbZ`#5=J?7_>}I z`W&f(hf0Ye?K6Y;)j7f3@4h!&YkkIhzp00q3;XziV*@#>R^%lIKd_+>7LkWuHSCL) zvhC8wL9E3_q4?Sa6jGxpe#Z&kwK9-r_Rr&HQ~C-qZH|g~S(KU*8P)n7%NbFC9t!Zs)i5OM8hP7ji7cy(7H4tq6#+fPa2 z!YfyH>5LK|y6+8I{EX$7qeio-lXeMaBZly#{7GmV6U`H*ykv*u=D{?XKf?CG`(WBM zCH~xNHurDHVLEk_vA}T+8~vD+)0;abH*PAFoyTZO?V2GSiyRwz>#b zr|!Y#hw}LPUL0@qXy8xM2U3Na3XIBfW^%eS;6iO6fW26B>+LZ9f*%rdZ$-tjNrFek--a>CM)`8+oAg0!Do20}(v z@O=H{e1L8dKUcAq*Sh!S@0RF+%HyG7z5%K)ES zDns@J70~oK2?j+Qc!|tOzJC1|OgY_`9;}GrGoQNfWTn;oh^+p&@)HHD(DyUG8PbCjT>!h)tDrRJGRVKn=5CsS{NAB5-aR-KD*ya|S3f+Uc&P-c?(G#l zsu;+hS6sm%(T^axe>Pu}xQnkUImlI273u2DC&KZR0qnbb5h@N1!K}$~_$cW-cFvzI zZqM9-PoU(Uo^+=~CP>dpb3E2Q1Ht z=Sl$wpb6ePaTR-|8=%XtBVP!`x8*b-rNvSBokg4`Ic~J|x@d zhZhn{u_|ByN}b+k zm9@Nl4w?;`P+~TX;u=!Wbc7Q`txpw#>n;hthmFWQ=!Q@mwgmeJ&%;ZjCZXl04Z`YK zrWCm&8nbd&qU}{_N75%fOf1|5HO`CCJ7t&nS70#4>OR7KlVow>W&_->`&wvu6^Lzh zad<`Fi^es66MU;5ij6yOK#`v;)p&dp`UVtNEwX)#DdPTk%V-~qX`YPMHFl_SwhV0+ zN!j+7qeOcz?8Ot_v1qHmh7Pn9q4lrTg4J$+Txr7bOf@o$ z(EdZvlQaq2d))D-#&kC0;vO^@(7>pw1RooQQP2l3tj|)$&n@?GV*glSkn>vfm{2MV zdNGblYi46>$#d+Qa9kLARwS~ieSpJkDuo^0Qr+m5SxykefVN z2=d*9p+y5xe!7xia^g4sNb|>UwPMsWdLqD7Oh6T!+^U3#iGQND8YEoX5r35H6hGHnF3EJQRI;nRA_#L?uF&T z?-yqFJ36vp*_DA1{8^UfpE-!-%i^%6tr2zAT7|#OJvj4IANW-`Qc$ayicK;G!kRb9 z!Xl?^q#a#&{`WE|&s2^_KeiJBq@6ht-W52m{;_y!pdG$DKah5miAIN>J&&J`X9#(( zRrSm^z_IyHFTWbdm;c>P_QE zPDA*390&CLWM)51ap)9VbU9*1a}U-CSB#XY`)#=JeEDs5G%5_)u?Un^Kgm`%T*c-S zGUTWGUKrl9hGqoG(!za%>D}5X$kM8CUidv>Yu{JG2HR3xRP_{>r?lX<-I-|6YZP94 z{~llDtK)h5`?$zAmJYAIAh^U2ARYG*T>sr!2;IF-IOz8h_kWS{2xO+=@9<*g|1Fj- z$a0KnzJnjuPZDQFYT+Nli?}d96VCXG*}Q05!Bm)r*_&+Wop26ckTX^_8DQ?>edrle zCQ80!#h(9O3!i>IgXyEkiP{H_V3Pvxi)8Lc3(O*^YHFk(4%u{1kl|r0;X^#}w0faC zPY#R*e-s@w55m~+hu9SR5blM}!s9EXeHL4#Y+Ai_LW`im_O5mXp<9_%o%0c`QyYX~ z5(Av`xCJ!6kHt~FzhRPXM^(nPTv5lZbe1XYT5FbSI!{!c1cSIoRS8q3VUO}*k@k-h ztcuS>WkU;=cfeez_v;iohdyI9;b+jPCk>aHY2h2EaQ3=B2l^@eoE z$HJq@jBWWA$pWHu#g{L%h`9O~*tp`bC~859xS?<;o3}WVE!upBDMbrJ>Ugf9yb8a=KjOG<_&_DyDO6&a9cF^ z>>N>>K_&CI_*pgF@V+RlZ!#1)l!|S`yg`3qJl@Yyt(r0;SQzvDCgfVk;FI!yto~mz zoZU2pO%G9qufKLyiQP(AQ`bCfbx9MOYbd~~tf>sC-M-{N2D57Hv`|}J-1Y(85Y8}cK8d%lY2y8*WG8+b&f$FgAjBntHs~^ ziRj2V8=-Vn8rzg~&%S$JI>@PfW~OgqMcqpj#f`Re@SXEKp=BFmL%*fMhWX7PXI{vL z8AhU-dYSNFwLb29(JAab{gF*LJVbE6Q!6|T*(uU;Phi!ndh`2wrtn@_Q`%R07%sN1 zfPM4g#XPfAsQovBjn^H*;2onrfhs-?^(FCWH6qzZx3j@-R@ zyD>uhfyl0HB>r31BltX;!dfl92`-+$#hc7Fqqp5c(OB={u;ck>(VyHkLVCNjb9q1r zd)F-O?sJbuDG5Y$+VVVmWxfp7rY~k;wSxW7_1!|>W4T~B#smv4D)8mEHZX}sBzyVf z4b&=*73}N(FyCY9(6d!VxP?-#%O^ijOYe{OG+i)O%T<{B&`>zt^8j2&kA%qZ{j6ok zAaFKqz`xFp*!+4g4o&lg3Lfd+Lm#z;3RXz#In$>t)n*XY`{)$z1K8d(RvE3$*7jSLqG4ZFjJ8;Hg8Ft`G zl$38gONcLdCHy(cQE~r0HrK=k@3gFCT1Tvyd3A`;|Dp-&jK9ZBCxzh*=79%#+L*04 z31>FCh=;9<6a6`}8!mZ{!U2O*SzgFQo;<98xt&R1tDD_b!T< zJ2qFu^*aZ0u~11brQh?`V--BvZz-Cm*|4XLFU4lANjNJW#cpr%DN4PK=UowKm)l)Y zKjRZP@YhRZvwRS5Js1C1X%Cw$j|j7!^f7aF8~g07iE@LsfNbC>xUT<(MX4hC8C}CS zE6Rl;$#i&cHxsN^+R&4P8}KvFlg}FYoo!lJCpMk5SIC*SMX(RlfCBqy7WQqWV3e#6 zo{|OJR=x<_+#iFv!9=)SYa?z*Qx7?jMkzj=Utda~H|mV~(#_~Ndz8n*J#Y?!@0 zU0Bv{JNr-Prx5l*fghb{%~J1m3Yo?zth)O_q`5en1u1_QYbg1WN6t>McSbH(?_jjV zF^b!J40{G7nj!s~~h(sR-MI4{WBFGyA^KkWE_3 zsN(1jUVqgNo|I_usxEbSv&RqenF4&8VF%(_JF3>bKP_lJG{>dlKd3G@Ms)S}Oz2}$ z$vh*MF@*A;}(?Y13`JQRKIV%dym-e7NpD&JV7%t6%-?87>zaJP` ze-gJPy9%*V{_frU6Cl2>ibBR(T%xCoSC=O9-a%RX?Im~KQMa3&IOPSC)SbDf!x(P( z+nj|5{edVCWe8naDEheR7rXIl5LeCH#5>%MLRv=|m|JS`I?Wk;p4}4e6)==v@%aO- zhPsd%EHn12w6j20eF6kDMuN7I1)q`W!_9lQ!2a~R5dHD9=uWympS`7=o2CcxPdCm& z)5Ijm7^u$=m;3R!b>=Xjsv6QHXQ6me0vE|%1LI@;d1&$z`~R%<`1UAONKozKx~spk zX#ZmScb{V+D<_6i`vE?E)>3}Gzb~KK9xe_`E96(lr^Ed@7GU%ErcmycAkOr=3HCwG ze6#Z#EK*s|ccx6@_EKN`tT=6m&US$6{_QLc5fl#l@$FKbN%mbQe6jQ7?unbhj4AWQ z0YObv5R7=ptu$nowk<$)8Ips@#d(aQzU=0Arov~}XKyVk(2mwkCu=wI-O_y>=Rd-2JE8~DRlyJ4nf7gKMa zj0cXlvY%(?^Bn6NfP?yS@$l8$AkLVly7%WDAu+s!_vJ?29ehx8C6Dge!{=0)Kvtd- zU!bMT{jbaMN1bDNk^5}^Om`?hJt35LJFen06QzEj`S;k&QJX=(b*6Yx^fP8se-HM> z%JXQqIs9LGDzm9mNsN_T7ahUl-n~5Y9csjqo?}0M}Wvl&hxA+kdU2#Zp?gk)YnAA51@VcM*d|R9qe4TI()Oal&%5|bOMyo00T#N*bjtk};Ddf@> z1r2igAhUM|>8viLrP_|P@UDoAN2Cb%E9!&+ih7a_*2}s4y89F_Q6u^KR^n%5NNMbD zQs2=}()7ic&WyS)`1b^o&w35X;EwGyVA5pz@3}G^pBhbLHax*?BU*6hE`SB+%uwOt zS{xuBf!PDat_q~SbK~gdY;zJq6sV*910@xI5t5I5qu*XL;mGM? z(kLpRxJk!omWHgv%CinDrdmlJNqr^fZl6Pz@(XVsdx%Tog4pgzRr+sR0M+)hf!iLg z*zeb42n{<#r}iDhoBJ&!sqGsnc6>U%ofj{}-LWH&GIhzr>Dt5+MRfG)Lh@agN<%^w z=wp6?aU`C5d9|Ou9uYOE4IJ8sg*oj>*+(NtRx{Ji+_&MmKf%B z;Fmq;1*L$#d`Q?n80aW<30K8psh^RgT*gMS)vbn-hE&jB2YXtSl!+tH-zL+k*{Cg^ z3r8N7!SJzFOs3nAR|Z^{=7rx&gJ!q!ZbJn9GR&t|O=JG5vJY7Yo8bDVJE(QjLYmf^ zLcs^#p_!_zWXsF_H0YrwFB-6nWc!rShm+cp+|M!0@@o)J5#p$!XcO*w+)H9U`U0(r zh{mHqC+TXh5I(cw3E5NTn{OtKU+pf;%~&0!S981xt|YpNxeKd zT#*8HkCixmI8WKXO7ZgfeRx@S3qLP!hIa-iN(^7iOWgIO?~E5V(aQ!aact5rbd8?H zX^R$jG^t}pTp!VqfXR|4w`B0ekrdc>*q3g(ET>`0j^I6DE5>N~Q{DBQLP7ly$!R+d z9~Z5a+?e-)N{XeOW1kFYs;r4*pNuPAyHbmplMf4leV^gp|JHG3RXv)#xCSh1D)CfW zCo0~u$D>a-3G=Hvp`u01KT#@KU2kJrw~XojaYa~2#=u0g+1#7bd*Z($I_7whIZbGV zfEk|rey18euC&B`QEt%Gt4TVQ&Y;_lldNHm6!QbBn;mP?6P}H1=vOCKJ z);kL-GzH|F8=(J;9q9MWo?a9?pyd8&Y<%vDi=?~V?EAnOeDug5bZD9kZr?qb z%deaG>-BDyS1Y~0t7>54N*6G#G-pbu_n_$4zr<4s_{yq zDIf0~N3T|#^Z996EgA=;Z~Jl;R-e$oAzkZ|}krBc*-nwIUoLjN#s`NCwe1nD8~=-X%}v4vX+BM9){?Qx8)`6kz#(DrkrK2$@eoUs=i!U#))28PS7^5V zCFF|4I6^y(+~;l=_Zv2kCbisytYruIkWa-JFew8s>1mSPlh=4Oq8?7weuU>yo#?Z3 z1fEIJmwFY7AUajr6t_|g6-oPqsMD(%9$J91livun?XFyTNN?foGJQV2%neJ&j;B1* z;sNF9+~wtAc)YHJWS6<}H{~*tg=ac(V|gOweRu#4CTFlB@*Bi=ECdZDfajkh5afpO zzIy%;^ z?sG*nkd#%TlD$G9d$yO7CKWPLkyJ#_ea@4TkWzL;8QGB#D(ijkU-|=->$ z{(ZM*wKEJ@Djviu=RS+GWS#NgSst#E2D7>ziyJLQP^05i*4nMgzW$iTkL>sZUtYY! zou1zKE8l{$EIhfq?kG-rgB7o!gJVYQLyr$;U{X*Go)@g84442bs_r*cOySYpvQLxbi=c zYIy`h_wNMFuyX!pS|({;aH0hRV<~lr7p`CFMVoGB&;q`MKA$*9Ba{=#;>B@seXX!R zoKt~o6n*JU(M9@O$5Tgq1goE#4|j&@gJcS0zMaQNv+EqHf2$(-jbrH8(ZlpKWheUl zi=fkbRoFFnDSz*k0&c`-m|M08UtE;O8Bs^=B;>Ds+d2s zUYmwI%qQQ4D&)Q>iMG_6)23YEURin>Rg0SNQY254H2(uhu_p>`P^P8sLSLH&j*L$- z{n{OiFK%YzYt1H#JMe{S_YWnh+I(uel!aNFjY$9G5t{kj8%HY1(W&LxjK^H zZbqOtln4Ev!ru8RNA)wiFkA3FT{vHXAC%wX@z;&aTic3eH9dqyqqpEbel!{0=%hA_ z9b`QyiMHMLBrk({!F{}gLZ?39ixLt^)vlNO^0x|BbR(O2;~cNC{t%_=c#%ZWnA$e~ zhc8-Vg?*Df#(IW<$G_EVD8`ao<38@0b{ZKJR1cQ~X9-j_z7A2@kRBqc4D z#-^jCP`P;oMpzh-i+MGQAFpNKyM=Y*$XG`a5}eTwYQ z6qaX^AQ2& zeQ(>czd}_1^s{o>w&NnT1&(FsUrgaL?XPi9qUKe(5wiRBUt@J566}TA| z%Hf(L|GoJ4L|YJScG9ZM#8(w z99}i%P*0HzbCa9S&3xP;{t{Dzhs%HPR{gev>bMGic<5zNdyz=VdlPY*w<-R6Uxj!2 zn~~-DJbJ3I5%=$z!BoHY@vROs7?n-MOW%f2@TnF)r{5j!VO1{wEm7#Wbj>7jl@zAh z64YL)g7l}$Sl!2=0=H=&3^47(qAd$Bc3&FZ7x)%I?h8>tdL_H1JW>3xEEZ?3_zsyb zb(n^@j1NsK<|gaOz+8dEUS~mG5G{t+F_Urr#}_)+Lu3&q4Wc?-S`;QWfiF|TnT#PQ#m)+M0{?W zk85^s#6}lj53j4TxVCr-ZZ&4}yZ7K%Nf72Q-G$1^zxZ#B`D>fUA7oivYqhy7ERWfZG`_V4MDeCo#IQM z-TBYA!m*W`3suh*VDs3U{MBWh{MRw53}mjbS!qWh^SleQj5rF!)q!E%K~C=PRPo&R zm5_U>1PZG2_!*}I`5Bw4VfHi+mfh>bcDRLu&Z!3e*VHyVAq<B3y7}F3qI!1>|v~xr|-Is9{emA)$uc=JUJ5eP6qmhl$ zD8S~L)uO+%-9Te(Bey#3IA?g_gRqC5iub4QfT~;80)Ik-X$KzWH1>CJO2L;!RyNVJ za#=d;eavvphAxNDHG>4t<2_2uPNAQ2?=i-S!-u=~j<)A*kSZXUbsaA_VO+3mUni&q=vrJ*}p=7T5?g5y7xIf-l-^6GCw8jyA zO>l0lKHK}o19hNW{JNu3TyjXB657AxXp60+_2m|O7`6q2hb+NBja=~hxP%*O9(W_*~{ICiFJ8vHF7 z0Njxa4udtaa6SZMve#zF8up5xGsO-Cun6k6|AWbygP6!$j?NYfvqiz3ydmO^y_8BC ztlBt%8Sbx#;Qb-&?~{G3-BXrcHLu~Ss-ozPnGEimxe1%BmWuN{ZsL(K`53p=8%G-* z##070;2U`W%2f_fw!SjomwrUN_UKyp@ohFM)f5$^zg#Kj@9g9X$m`DrGS&RuNF_+Q6}DL zE{A`m+PNm)qtnMf1NH0S#{ACkTD1YfTf_j`N|M?~bFi()b9d zAF!aOV~>gwM}EPSCPR|^d5Qg!Ybr`)cQZBVO#Hs(3G9#c!mKIZvAHgcHb@VL)}gUv zKll%-4t>Wi1WCif`jPlUIS;4Kk0zbmOXA(HqPV%GTBJKq1FnQ-!io96aI5_za5}BZ z`Y&L_^knI3q(7VXXc8^Z-ba5jO<3T${i5^7-gCdyYRFwd5yuU*WN(T)QA+j`iW|~! z@|Qn&p+lPa<_;wzzCT!v_$X%eab)1TiTnGx1a|HFCGrinVA6d&ZYv8!)wTMxZ`pVJ zWOfub&&}qIx|~?@;VJCc`Mpfa@&h_99Z0FC9r+fI>-^$X>uJ2sQHt?COHFczc!NS0 zae+l5j!X~6E7@m7nk!;NE9+WNb@USqcUlOW%!EvS^l_L}un3Z8g;AQYsH(fX7LS@MPMN5?_1uL4fC7xRg84)fzbsLO{Q?ESUXDhC* z=!2J*ckp`tDZJ3PmSjf0q_Bh4WM!@e*`Kz;iL9ZbH5y6e{Cf%S95aCw9m*lzc>pIP z+#RWFma~kz#^h1Io@zFJhG+5hI2Y_FR5FKf(j#%^scvj+&w!DE0rb42feU^)jA>XK zQQ2W7GF|T#4^(?!MFA_s&+vI*!uH``)a~)~i8zuJc(Mv9R zo+aoUabw%$R$*#X5Epvahy|%D($;oezR2K=khklk(+$V?rZykgEgJ~~{`6y?`)yzv zv$v3ONffjlJBia&Q)segJe+SA`pGxFsKwZUS|+=Z(wiq3(-cv$tKkObnuSwL&Re`u z?g)+9bKy%}81|N1kjQNxXv#04OLh^YnV|;nRlo958*bw3J5j9kv<1J-TaGo=*wDlD zCYU-p5&Fyv(CAP(2QX3PCHQ1F8-|?9;57A5)*I7toVsRtx+%}d?IQAVL>Sn`E zf#r5K^bNReszyT@1vYD#9lJVpDXhy(!+#&%z&x!;mZvu!jn)b0%lqm4i$M{1!|4T# z&RNRk4!pu16-ZHwvJ2dCRKblUQ=rBF9f;h7?EZhJ*jBIk^k3gnw&9yTJxTCq-@aJT z$owLBzP%lpiRsOC~WQ;^ufgZoA_+$M@Wz1jhz(+OLRjM?f)X~!pzGHAt1U=u5~ z>BM^<}3?DDnbiUEKt23-@ui=HF!X<)c{pt_UtY+<T;1 zoD2ac2eF|}PX#vM4D$KXpIYnXIrHXgFwVJ%6;^u#CKclM@9WvJl339DrARS{ZJ1(( zEEt(g#?Li+_^K$6OPBX$N)BS~aA71{YilCBLCgdWbttQKYr@!>?=itZ0v{DMS(osC zWYNm0%&6cr4tK9b(L^yC{COviKR2G8{ZfQKWlpom zj(NXb0(*D<#uF>$S-_uh6p$)7X)>o{RiX#eH5E8k({k7$eFF%++XHRa=d$6)3gMn*w0?9#QF{lfSGf&Y1H)(=<&9Q%dV{B?tUl*vBh-GeEVEFzVrqj&U*#ll$BZS z6;qU?Ncd|@^@S~^Ed=J7P`}Ef7~1y$qt+!dBjNk(dr27km})~IW*l#@a{{{_Yz^Cm z$kn`ZPrkID5qJb0K%1Vs(9)K~uMLo3h>S0r8?6It?{skI*KWodYk_k$ZwBPLsnTaP z2jX5_BucL123XbOT4^OXw{Rcc%sGzj36r>*Fk=`R7B1vrDma%9&$$B!ogpUCfK5C% z210^WSjP%=_WFRS!2NUvmrr(-`N$XVZOg{+fBl*0xiAmM+nw%@Viys14=O2b!a<}nT zvcQxm=!W4pj|+UheS-Tn9&myV%lbK*+59qwP3=Q4{6i}@K1A^DCCbwsb4S{e^#&K% zo?t~nUwB06LgsxU3&x(W*HL4T6>n^Ad^*)x^=9088?j$60HE`(M?eQ zWGi|tiKL9rt1<0MEAEUQ&iJ)A9VUFMu9%i^0Z$C_V`nt4qjL9U7@yDv#RWpA(K;R% zTIGOebp|+vNch%e9bEg4*_>vsA)C=Ym}aTef~wbE9APnxY}-2dd(Qb{1>6r6ht}e) zz`y*gdIPpd_|`u*bvyev`WZ9`l#_#9!_Ztdi6u>o%PQOffRzwuNs+IID0SEYe> z9()qw=9NK3;AYX$)$3u;m*?2IY&Oiwist=m1|hqGQ5xS6fwGiKf9 zHeDIR$F}M4hppankJkO*&UeVtO{r%N4?6A9?!PMDXUJ@{>ht5S2)VkhtT(**k`}@9 zCE+Ej#E3?X+|f0PxG!Z5yS7;m#}1Ie?Z40PCDE5EWKQ1Y>U)ei1KGEH&U^-;TVuG_ zX|-I<_<3;bkqC;F(zrjFhxw$x0ypiJEMFC9j|szKxQFM4Q^4f|oT9Nj)V1b_eMa5q z-z(1KcHYp%*g;BI`C&9D_-b;NV}IfI1?Kpqpil5@r}F+8ju5V}_kFz(gmKDZs@PfT3Q4=uOm z}&AQ*yc9FGk{D2GRT)`{@uo;)(cMqo{M+V;~l>$@J_bI+_q$8-qr0a=bdp>ELyP%cU3u}U1JH_-aL#a zq5|-u{|Zhq!45-HhfssVN(dV-ht1KSVQGI&41BePU(l*e`dcMD_gNO=#vkJ@-c!RT zZDFF=-ds5MB}24%DF>my!Q-W13-V#wAu4g6hO( z{676QpHtGtNeVQe;^by%On-^NT7-Vt+2VbHyUFI=P*@ncmgg-CQOKkjEQ$~FvQ zSDxh~%8Y?)&7oAVb_`B_eg*?aMTn=Fr$d=YV6p8q!F5ZAklR`hO4O34O&?nEhtm}d zb<>9x;aBjRTod-2a~X7nTu9JjRh;+Pind0?;=ZIN^zr3rN7qzQ&XnVH zcy#a@zbnMIs!Q;?ei&J755bv!-8g9w4^QjQfxkv6g+8f6$-fb_zb6CLmfP^_`VS=A z9yNM7mBaTT^$;{L0hQlMqty8RY?F=xwTY78hJBhi==(Yx|L70Ci5Wlv1ApSWRi3Qj z*)pMDtVCHp()6;-n_LC9Qnj}-9y8m6-^Q-x-?S!k1=iN+`R^*%vD|=rbIqALO$i4@ zH}k2p1=nU}KL5JtFS?DsgF|LdrNM0jN&d!gj2yQJmz26wh;0mV;}4B!4k_|E$^48b+xTkkEH3|? zLpIYs;m4Ix_`cbUuGR#?z$x?S@8e7S2Co+I+c$}=DT(G6h0Q>juc7Gjy#OuyMuI4$ z0fSsyF-^xBQr;Niqi`d5yk6ikj(o{W-xl`s9=?2f#e39h%%w*UN+~@-$Uq#~gJ&}k z{e~!0d{_jk%S7VVu@3ZS@lZ-`lqaS3BCPasrj%jCxeY#oMT>KhZPlPLtNdw_=|}Fz z%@aV5+tIW12Gz_rqW@mZBgrIh3>`3lX20l9`#&Z?_e)_Pkoyl`4opE$uiO0WmveAj zKn~Rz%2JcMB?UaNrA1e7^7j?x2vu+6W-f{Dc`TyMC)2R-cMHF{N61J(7-?U<0kiK1 zQhxvWRCf6UzxVVx1lDmbwnX z{)0@=NCQ~<*+|xRG8tSzR)3T$0uPZ6$LJf=S~6NXj&wBsq3iHvXb@A42aM|QP}qI=a(xOr`!10yI%mucAASz+-YG^(N@oTh z9xURODVx)G0X=N2=&j}j)Ku_?Nerm-=6q0{u$^A~jm1SFEtr_Tnkps>ypHPzwDOrR z&6qz1#ba{tU?Y(0#T1-4S&r5^OZW~U?-YEb0M%Cz8Q00MAHv>hzUzFN_`(4@2hE`% zrxt!y58&BNH!-|B1q-f|!|q$>vE`gB_aIO3?F1`HIu!oH-RXo{^2O1rU~b;%f*)zp9iDIMhCp2eGRM5d6Qv+i{L4{%5BF&Z08l}Ke0Z>oh95o zWC+#C+0kSEN39qFB3Nbj|7k69wYC?}eGoU+s5}l|}prwNrlI)cpG>?lUquFYZ@M;5QDPNX!rV|LY=H2F z*1`cXA>ihE0bAqb$tp~rS`{bI%+tAide{SAZtDoDb6L*+o!5Yudxq24BLm3Vehu}Q z7}J^ZIi%k=l3eF}#yoC7g@Zz>Na}$z&Z(6~N6UP1PW(I?;R%~C&bsnNh3)?QOapUba!k<)Q}e=Y|f9TON=JYWsVp9N3mk)_ipyh@RnxF2nP{Dc4bYb+&g zQKeV)&KNz6@a6PuJP$I2mN&5LV-n`vDHrbjiPV~~9b=NkIK9D<%%958^d-jy4&fa7 z(Yu9j-{MX`N_?rmN)2+Tn%#i?@2%Qf9=f)fQsbxlD|ANyG*A zXHY-#G%m3dxFoW_#Pz3?>2tg-&9pBRdZ-nsIdMD=&WjZr^#6u0e&|x^AVd20PoG}4 ztK-{gCX!aUVtTazpwW08pPp<={y(-uy!Bu@|3HDB{@em5TeHafYaH3Q9HyAcR@!-5 zmU2E?Q^2QLWHCR576+=6wNfjY|9nEFuT&&`3vyBA*k>@(vB04VIw&AihAMu@Ng{=8 z&>!niTH1JuM%qho@z7S$0*6kqqIeuz^51Fdc%#a<_#cB&@>wKO*^YJDf@gEaYFeqh z7<2ZO2@a8+?BR~Xc=i4xH0Vl4w=MdT3%WWq=Yzlj(Aq|I*oo zudrIAOdl*V&_ZOz96l|(5xScz7F}XZw@P^(u;RI!DJh{^ft|5mSH5qck9RFNx}fJ^$JYUiXzW z<OI2YDr3>N2fw$^s4f|KYdo#iDxtHzmXb@SD#`GDE%dk_X4?(+;o%8w z_`@Zcn$tW(zt+@do< zF5v=cy)ofG^8+QKvU&7+t$}E@aFMM;u@tRUn_K7R!$k3?;NpO3IB}Hu>kJBPv(9noe4moE>aZ*K- zvBWO|Jkt+S+T6)#6|PB9N`J_B&LsM8!3;^FUOjyp97T04HFS9NJ90c#PHL}jVQ9>5 zn(FqCFP<(UafUYqx4huL7er&R$1?H}oyOgbtLe=?Wyyq(lW5*SYuahtT*q2hyueuzC-;`4@#cd6SY8lY7b?0fSLO7-Te2Ve& zU*TYZ*AlIJh0-RuQKO19-BLV73-cT(_+1yTHp&-M73=u#LUy22<~ZsbFX7dHO`x&a z3NXV-TD1F7HJ#jQMkD3&@Mqy!db&xP9;`AT*P?#Vxn4%df8V6i*4do`<9W%yPM?qUC1#^JAtZ}8_8L6l>8De@yAmwDCJ9% z!{XV~_#66xWH{;*l+QkcTG@5D;pb4a`*MczEmYy2`3yLeZbDC&-X){8QzXp0Kjo_b z<##T8gKygEadqK9$=Q87Nke_PBt+q$kU}w_N4G>Ida2KuMfp>H!dKAz^yVC=P-bf$L@x<3{?V%?UM)YFO;0%Jsxy76J27lmH3 z1heXR-r8CZs-_bB>)V))Qk&hWhKFWeBBk<$rlalWJ;+iSbv*NiweBFY7( zKU{>zLyajzBZ5=go_T53ds`Y-SY7CjcBnYtlf_>h*UUTko8f;~szIy03a4oMae*J+a^kNA z{B~T2e~&AXsZK18Q9OdrgdETt`%1LA+@E}vRx+(ZRW{}BiHee8=efI2oVcRhi8y;( zkiedc!JcLr963gr#@_tRt5#`Z@-$`KV*ZcM+j|5G8m z!O+JOY2*H({8cGEZh_e%%yan6JwDxz5#29Ez4a*q_ez5+cshvM4P@B$&Ii1POD)6~ zTeBjgQS2D~0Opy-6l-JnKh5&2x3V062u{!1Pv<#%=YM$8xe)7SD}l>MVrvAX%!K55)_>0Edoo+%2*2}J#eyRiGP@I4!$LTZgCLDlp&ihKQOzj``v zc&`DMH{3y)$~Baz@(?FIyMhr)gJH>RPYQ2LCi%iDoTqsLzjkEchVfpXN8;e`C(M!sP9d(<=SybG?q zUJc)eec?lg=YwWX02rLgLItM}Ff`5`*4dh4_q-|eI&T0qeSVKmoInPe-hCJpV#hM=DkHYQc{IEjCPin*r4m$c7RCJ)JTMOF zIKH9)ZPy>bk21C}_(DAD%6p-IZ7XkEavb+eI?t_)%%tKkW+Yom=+dD?E>}DG6;6I^ zndLUPu6h(KCFjL+1h&YXn6a=cG6+3{tev!*67O76jsLw25``s>Wd^aA(RY^|w@-Q} z+-#o%s&`EAQJFuhd@1bURjtLc#rH+)b^1YN^Z?kkXCyOqRH7+2U*fXdJ7n`#3ih1r zhxMziux$Es%5PI;2PfQi&^Xfoa_YsTaO9v!GQE+)-U?@<*(~~c_dhsd=Et6m*~aMp z1D{LwMIxH17`bFG=|wUbyr=^P7LHG=InmSP|455n@Zy|7N|2sFR7 zT(FjI^J+O!Vc0WG7m!4X5tu`Im~?NOy;VU zD9puH7XA`SZQ}r0Z`0$_B839s8JR;%9M?yFF}OxtLW(sQ~Ot1s$msEJLz{D;fG8 zhAKxv#RwVZv+o2b>)EiK?~Jg*L2#U&vS8D2Dt~5C8vM>_1GzJ=VSE2*?)S%ZnDsRd zGina7v-3l_hxfX;ey@fDC#mKeA4Ecn*9%bGWDL;{%lQirpTO_|%EBAf6B9fdVAxt; zOdER+zOy*o@oOminAeSSzLmoT-9>QxUxM81%#-T8^D_$R}@+}wszXTHO*7Y=NLj1|N*XXC;NIpJe-)((!-I0K+|0g{$VvGkRbIJ*`5!S&F3 zSZ%7xP|cOID?cHcc*v2JkBGkmJDQ4cfHuH z1_oEYKZW&js*Jad=8im2XO8wixe0kHeBxOrH1c}O3YMx1d#%4*;E*`$#-)#}m=)F}?6SBam&;jDS(61!nuP8aVf+88%)Oa@msyu;USTxyIBixN%kj z;+%c_?SZGcdQ~xmi^4el@KzW&xdiTADusI9n;Xwx=f=)m!NoQVhjf1p$O`3Hz{Xva zs`Y}+p)PLFSbeDenp!a>S%+=sWaz4^zN6B^!61q?fqP?g;gYGqf#pvV#V?<7V;}jD8ouSh6h_-a-sxR_SXF^^i48k!{slt z;p-#Wi2ixZ#q)UTY>aQcr94F5n83{g|A@i`NB!h*=++^>nV_8i362|5iv03?%kp8NJ zY5ZQmx(Bw1Ttk;Jn;+J6c(*O{J$3_hz6qI$lvT{*i#Nu~8ndZxdC))l9}Lv#;mc#6 zf!g{omVeNJ9M9dsS;9 z#d_iVZuUR8NFI)V$zKRvhw|$ah!z2~%R$ei9Ngjb=Yam0D<0-7?tJ8}+a} z#GcjvjA6A`_dx8}QLy&lD8`(SShM&PGyS=X9sckgVuuZ2^=Y4A^OB29cVR!y z#ypkHN{D6tPKwM-{Upx2Q^Z2dQ@}9qD2x&QJuHob*`O^8U{SOa(`}jpInScmM1C25 z+Y`X7&B9^zvM@f(bQ*ipZ04xi?+^c0`870W_^>ZW$3SAqB0O-s5N;imVN>0sSdj4m z_G(5n^DdhRmJueDf9C_=l&8g(s^5e>y_s05Rziwrn-~k0V_EXqaHviXm9CV)$mRl2 zsB{C%$_s37O)z`zxrA%d)pvx0&sgmUMKw%yXNVJ+O;pt9M(okE0#fmRT`ioUX#&W_55N z)t}9lQp0}E>CATEaX30ki>*EN2nUZ6`pvI*!aRd`7M1k?Y_wwqpWOkb*1MFATH69Q zMqT7?3=y-#MiV)+htnxg-31&{gbvsi2UfdWnT?(EP*5=%vR;)8*x0B-!QP|U+mJhu zI;I|cLw$Mm4sX_(ynrRU`LlN+=h>&P9*ji~VPE z&L_!Ndz?LFI8;@~ve|iy4jgtyXRmMEQ=9wj$<#V}m~o5d{j8)x z4#imLQBL2+r_o&7zo<5TBuP&bJ};M*gWH0nicSxI+Pp6s#_gFxEplsVa^ZFU((M+A zl_^C39S13Ne=51Rb&-AjMH-wgWc3;{sazAubXGq}_1HY%zgE#j>z63{H-Ubq_2NIb z&Ga?+BedT;i@Bcyu~lwA{YZ4DjY9r0Bh!F9Y?q+h&kHzc@*;esBSrbm6|`H&MDig96 zrBRrP3Hb|s3n~3-Qd{tX`5g$SFJFJ7_R(E~Tr6_;mIrmR2C))G+HM)*Xr~kYg$V4iVN`=0y>!1v-`;I)XI>UfI8);I? zu6N{DAI?unHkYJJ+}U>FeXJIjLdD`MWLsfI^;h;$a8W91RNMw^8;yY(gK4h(9jrJe zL)w;;*!2n3^x0O7D^`xfTO?#v2alALNFH#4jRr@=4WXy2%u)VeBszV6N7nJv$ID{)I(I7QF3Wd|2e6V3tynm3~n z<1|#+&V>i5PD5x&gnR`(7dh7dN0#7=m__TQ^68l1r}pXJ0Bg2nQ@?~ZZnuZMB%Z7! zYTv9SZSCsR)i59Jg&x8B2|RkD2p9bs32L!Rab?kMu#1^WPeqM5xS^a%)0RobToBw= zrE6)Yu$Q`4lT0gL7)d@1drOPHWl(zAV97|ObM)t%h>W~%V{Gv%3}5>Sn{S`Ss>=gW zYr-rH%)W!Q6J%+%ZYwT6P)A*9?pS}#i%B1Lp*=-`G}teB6X?>lE_W zRbTU)L{hA7(3ZeCmn#icc)tXwP|@5#hk^-=7i)?9jaPzuWnFVo;tjO-3iBK1*1urS0% zm?;+1;aA@L#wg=~yngY@29P{%H|k8;+LA187*+9f3zTn%3qS(6-OY zBp2>Oy9`d#h$FtVsCuwu;Fn-{duu3?^IYCn&ycFMG-ztgW6rp=1kD0Z2!Hn_^leQF z>U=Y%3GT_Xm>uMfeNDtGd$suH4`--QYc%t}nodUwo#;igrsPqWk>uCYa7o*pO3GFX zqBA}siRT}|GZ?WRj=XurdAX+G^SvtE?*sDaWo`(E2F``09YK8A*JP}<4}f&7Wq9YG zHps>wKvio4%#^+)+BmA6^G?>~C#>5iJ|q8!J9F#~idz(LNntIg6Sf8&eqP`nPMQUJ z8eEE$Q>=K` zyj^JcCqnFZx!Q2yNljwlY`B2A3=?#wiie}o^Vj@-u^gzB=kfLEye*6JbzUr2d=yRyI$YuiPZ~wH8M) zJ;@T*)82<2Cl<2V`&1k&wwa?A9A)3^lpK$=O(FilSngZNZ0LI)#tq!Fgx7hjNJZzB zng7(W^w41?$xj|YZ{ls(p^JX(yPT)+9@fHbs(IXHODnQ{?@KS{EvKyf88FH*ftpSH znN(F1ZT_^A?vH$d)<)rsor_{MnisL+YXWa9xYH_i0wjB7qJrIZr&0U80 zx}uR{3~VK}wg;Hg!8Z84jHf~OqT%s~sU+VU&EB{lWtKzYX}7YABy#yxiViel)7Wmtoi;S%>%I%w7LT`)Ao~R>{eJXSxj!< zohW9;5cHa0K_y*tS$VbK5Z%5N3-?JxWA(Hht(m;zQ>8>!HA9talMj@Py*r;<-RDlB zs&bOD_oGloH;OwZ8BGnl-a^$1U{0fW@O-wA0(X~T*t+>NB59e}2@&3=3kj zKiDunSC?H94`p-HZ{wyR(^>t~0ZeLxFNt&;#194EL4)EO)=(Y8GUAhI;@Oq#&0HZX zk(dEtW^(1bY(X&vbVOlk z>PdR1cbeWPO~l?irFcquJFN+cge7OQXxEi%RP|4R3OuX046}6Tyn2A$x~qnAS=VrB z+j^M!;xV)r`mE~qBnmmft=q~`7Z_!Y8zy`0(a%65#he^KP z_`qbQceC9=qa2lYPH@~P6~%Jy7Sfs_*&Ndp^K1HDqv@f~?bUY; z75a6)d|to)T!PTjpazNdl({ab4LAavN7c!n#^Le*>IQ4cX z@wtRT?yc@Tu_ zBv`@+w~?sQ(kLFjNs8$xN_gAHv$>x2JooBZHh-O+E!~^wL+ed+VON72k~C2Kx=t*T zmc@%ETXECUTrOg9hxo_TFQSc(hw#?Oemy85!d#Ug?T+TaQfjcytUm1J!;z_W#(tJtUL$W0vGs}VGjmN zC-IN}Y=t*3C((aSnmjitOO#plt@Oso5V#)w31Ds;Z2I8=+#wX_zH`JbyTQ1A>Uyg6 zeJy(5WI~1>;czO)6mISCrA5mZi>tzt`6uOWyr6{mRdqL!RLMl;GN?|Ueo)j`0V3GSef5n zIQzchHPx-rH?W1zOH_eES9+?}L*VlZn~tHX{#RpJ$<_IPvhUV2s3!jp$0)HeCh(X8D9 zljQ(ieyJ-OQM-d~dWhlXt4{uoa4t`tX)SK*+eM9~jodovLTVTq!P#^z!U+uY%c%pI<6*HfOR6D+YEgB$Iep`ygY z@xD|Q{1@K=OAQxE&P=Igmi-i1_{j|Z$l0Or+C>UAYX^$v3@T+)XHQ`PMf=#5lIiTS2DHp%EMEPQ;PMxM>1lAv6+ zXLkzS6%UeR$F#AhvT022>=ib`J%$x;MRx9fI9uQ!&ORB~NLCD+$%5@IF{Ip(r4&|T zgsZ*e$m4g+EA}?qyT=HZT)c!q${B3tBX5b-KX=AnYDw<8X|wf5{;+{H(#%PHq~xf) zEt7gy$2P7?W`EQcW7I__R(9IZ;hT>oIIdjBq@FgBnUn%s<=sq!)UL7v$7Lj$oVKJk z(}qd!OJ*5H%@kBS4}Q6QhToa{Y1J_uHqFr=T!*(Z=c}VwkHu(-Mt(limMRuzJCE4; zuurTwFPnW|I0fcE*v$Oj1xbwDh4bjISQa5Qmu18hz_&daEHFjLKu#XQ200#M@y{GG-UZ{>`B^7fri%y8 zzpTiVXGt(@q%O<4ti)6Wmfmb%Vdfn>f^CtTEHO7(#A4R$WNO`K+0*MoB|!!~Y{5edm zSjx{=Y;1}X8>wf+^g_n5YTJjb$+(1l$gE-QQhfKvt=2qWqv(-F1Hpo%AR4Z?VY5T7snoM7%mYRA7y(S zEZDg;o=rY;oH_3i{yr4W%nS2aQMJ58&9RW3u)f94HQgh&hta8{=UL$-1to+XG;_F%4xLh_tej(R3 z)t&TL?}ca%X^y z#(Cq7Q8O*VhYoKvbp{ep&KeIDJx88d)_T!u<$v(Ty+QsFIi457l^Cd?j=#S z{>jv&zq8Kms*-iXMdSuk*?^xzCG%Hpk_6O0W8+O5n10i3_Uof7lUi%Z=B?d9bIYpO z)vIs$lh3U2%c2H;Mc*#=)$b;^{!ajFRF7rjPYytrL=8M1R)JDA-(km^Vj6toC>!yi zh>h&L!lwtj(z12uVS9|Cs9e#3-H6%Cy5`Pj5B%;iAQ zGSJ_BG4m>Dq;vl@3jK@GAnR#@%BL*Yj_KuOIa-n3b{xUWyAOh#nl&s{@dMqvzl;ri zksylN=*7?SaAiZYib;J$Fl|k#rt8OIQTBQp{ZUuNS({g~pZy=eV=q5e(LbIpty1JS zSZ^V>@DViN?s68eX8?0g|3Q-;9f$UXb3u3EDe`kvVN$#G*vEZo{Ky(3rdv_S<Js`H~H!wrKwg=LG(U{7M51S6(QRxjvdb8147{Mz|*X2M+Dm}%!Ah! z&u1Tw^~Y-GN|3|L)E5^naA4ipl{@zA#jnRy{CO#RZ0QO?uQg%oFdKUM@-=<9a8dk5 z*mopVWpE!~nNVP6INu`fB(n-%Rvu{1G)wlgg^@oX*mXWldiY)Z$Xk|t=Z^t9`#hF6 z-ayFLp5w|*g#Fc(38=SV1iRU`1`B^iqW41`$2%bsw(RvHHpeNHEzlfAKH7Hd^u_V4 z!dngu?pcXN1GiA;zCPNwmt)hN4A|m5)40tBlj&`DnYf|oKMvO%A$!{tVv}7#%|D9! zd&?Zx7QN+GUwuR6et$X9f)_N+<}_7`Rk<6Nzfs4l5^8evVQlp}5|5n+v(slY?Pob` z*UdL@`D8Y02|3Re&T?ku3X1GNfpGaZQDHq=axgwp!k&x!OX4#3u^&;tX!hkK`ZDn~ zn~{HnooNz!3ps*QX4gBa*}sUhemX(nfiK z#XX*pjzp6;nEi_P7;}O$=Zs(*J}u$weebeS0vmAVmgi!h=R?^rSuOUY`8>Ud`9ym0 zjTC+MF?GizL*5*J*3+TF&W(OfVEuw_ezb$k?7zai?kOEB8Vb3ymH1s3 zfo=E0$whF&j>++*5y2Jg>@IocyrhJFdu34jrgplydk9VPjH6TIl?89?H{Q}hofX{` z{@>IRr{2B;HMdr<$_!5mD_Por`Qck~ z=tXA+?f-jGV1+%P#6=u?^Tdgwu0Q6K-W(xYP6L`ggt1gtM-0kwWa6j*xc;Jy_V=xa z`-@H3Z{ZnOtDFLrZLQ);HDiu@{sN4HC*u0#p)C5S1|~<&roXB__$4Wnl=3pjMkW=5 z*Z%>jvjq-+s#9p^R3BesrP8Ck2JH>k)^lx2WfV&IcpT$1yYB%kfVYI z8SOM=(e^Xx$JiyTvs_!4(O-c6l9lwzRhyk|G~+KlPvABtEQ0*;ccCUMk}9>9pqIeI zps#Ke_TEP1(@@11=Ss7tEtBZ%rArhqSH<&}qiOi}@oZ@NY^s~Ool8*e;5x4S;coVd zNMLetigv6t;Xw{JXv%}qXy5V7>fdSl^ddQA8#C1c`c#n6LOtdwJ0-l*LSKp2F;9UbhZ`lLU%zQbVkfee?wqJp-vZutx z*M9S|d5+{`;=%5(b*1piDPW>^hc3pPfXO4JDWTPvJ^k+>H_@+$a|}AjOP%-!b#=cW z|DKRwx7URI3q#?sb1X%hd%~w<@~moi1pOJdf<0slzfUf(yvbL9Z#RkXoq1`SDyf!Uk>n2u;E|p&uIS=~@!!H;Emn01t zvgRzj+=F`KjVL&344FAA(7i>I;i&%t*i&u@(+^&z+VFX_YugI2-ZPeSKRJ;7IT6Nr zeUN5j^9GSYat!oqnhV=>Lq+3f7V;sEb#(8;Yd+O0f;2|-gT=##G3C5NqOU%dEJx;? zaGy|TRX^%@`D{%ZS`-fVwUi({rB5_ca)6FKYURv?;cJZA=v!r;NJ&DlKF`9RZxkmdd8H6hP zvPllRNzF+g);lIsRNpw}o1)8i9W3Toy6a%Y(O4=P^%OR^8{?fBXQ61&DTht3W-}R~ z-_SB*J0J5s19bk(L9?y(^mXl7p=Uk?npRdo{PcS0)Rn>`KjqQ)pOK?}-+Q?AM%QtQ zd?<8hc7tWr2q7RPKmM5%%Yp3_Y(MkRAtX?7ZsVeZE z@onV1^CvfCn*&apI}lDcG1_l`kW=dPrzd~EQ+ha|eavCmkI)eKiX8Y^T7D z^T{9@BODiU`e^a~1axUULVpijhK*^-(7$;WYRrEMZ;Q4=kWC$&nb+cwm^2zY`nSRE zd8c>_$MG=Zhdxf(o=B!UPNVCKz2b$-Iao1jJ6cr_7fpG)0~c;sh8s)vVMf+?vGmbq zaJux0D?Y494JV4g4!9 zqFp%A&yPEuser!X_wa7(D!zKLK3IO8g7tnusAMx1@5zsWmnJEgbyNeS&3(ajVI4o@ z#tqIxF$1LJ%DI=lN@%oH4&Pop4n0dtkk0iN^5aFEf7m68tM1~|JU_zUH(|Jerb6%R zQ5bkgfzBsKp>g_1^i;mbm0p&GDU(OQ_Qkg#Jm4^QUwte%y-$ZH-k0F}15J_DtYf9W zg&opRku^BEB*HzLYzUaNP4E&pL2phXZ3~gXWs+ji_V!2&lDzz@C!85CLdOqtu$3#oqNnv-eT2Yz@eUD5jGu^`)?9?Cm+IlS+ze>Wm&Zv9 z6!DYXFi6M`!m$JIiJ}K~z?IS^xYcDK`MO6z#i$UJX*dCQn-1Z)lhd#$H5LaS?2j1I zAfA$MhK&{vMHLEi(3kd>4=a2LG@zSXnbi+=|JK1t=htJ5?h)?5qFETL_XXs~DPm*4 z??9{NA&7nA982eN>!MtsH`|*V{65$43R)yT_Y5qO&#f!OsNn$ER` z`5w`@bm3W1a$W#@QK^JM31x!INCcv#Bj9FE21l1CLD^Ad(0jj%{}#UtR_oe8U&9Be zNWUg9&+ozSNoMdUyBbsL@1W{OMcjB^1**@~fMv`^v<=Y0l3{AtwMhyeEDy(RS8qf3 z=|yPisLx$>u*CXAQ<~zFg-u0*=PT70_Ox|_aibp|S{?<9wd~MspeFX+iROYs%F*k9 z23p4rfQzTK(LAyrbguZvyL&B1|8Iq$a9|?_O^(C~71P-Jz0<)#w~SNeC%}^*rqH!? zJ>5~b%>8`t1$EgL@GLe0YEz4#_~Bouo@B`;tat+6HYV^!HxkVM2J$`0f-m5iH^v&p z!H`3t#zz5)RX#(#3Tm`D_<6-i{ zlh7O<2P5_j<|XGh@Fg!)A+Ucs#)+rnfwoDs#x4s6?E5Tc)iWvI_yvDxLmOOO;sIvM zWvH)Y7@cwsA>~i*l#y?bUY=PXTXjUV#Cj0-Ht3beW^OcGc(Ipz;MWC1A1;Lh1Ka57 zI(gRDegSG848+@oPq}r1t1Mq7mVf*@2^O3j3W^3|&USEu=+`o3_PRrXCMXK*#^#Bn z-}@5mXXL@Z&)%4K^9Ilb4%eG#rOilZZx5uwi2AGCkgGSj`CBJb zX0?gH5d98gm4*p<|9+%jzM8wNlnRYMys`hkuOQ|noSuyumv-wc_45nn)_vT0*GtUMxNDZWB)unN!Lw4&}z4aBzMqcyYT1u|XpjiO;Do0-rBh7@?gDH^!{O z;~zL0`>2u+UUr3lqjG>h>9ZTIw;vTeLN%i0D?0c|GZ!#^ZaBLTK9)K>oVbD`JXl`c z3f=3Ku-D%L8s!YZ>9`$#RQSC+4b4D$f`jww zp#Kada2UP{<)Q=*x3?H%nk&fHOW@RZ2^n(}#6Z9O5M@1CoYAq5DvPYyEd0%RSsmo8 z2k(F#a${lsrAgq>yZ}eDL{hn_g1cvw(d@9e(&;?{2RKQY-&~OlEjef4QFH->b{oK^ zFPY@88q1F|NQC7#w78Nli2@UJ8)WTMM9uA+;iTLbZervrIy0+EROWaSetC|;hq4;z zZ{Y$()zuKFtk1khML_BbDN3kmq?(1oyRGUp=1hpf$JcYXGmD$JJ82JLL&F#G#xyA~ z7(XA(@`B;7Wv`GOz0Xc99RN?`iCcIvnB3BCa{caPL&n__-l?JpmT4Pd)2TG^rQ@Hu z49&$<_&SQq4{jD6j5z?iXV^fnWeMkL*h88P?rh)PmlXCyVEFhA#G7~%(;JuoQ_^ka1w0yf1p0%m* z>-=2VV~YsRsM?30I44jvD@Gscd!yj8Un*Rg|B|~eKt1e6Pv?B|>-gJ8fY!^MqMYRu zFyXc@pL12#Vfy55?n6X9_vzza&bRym_wLmN@OfGZQY)VEQ{|uWyP&~=Q&|Dstvm6{ z?jxMZnr!atj#SRo{yH69dLPlSG)uVfqrT~ZuH^pU~t(19(QQM*@^vd z#&reIi_~c;DK;>BO|pZ@HblvR^H3umsGlv&!k3PLu)_m z(yip1Rz`qx?Jm)sw!I{NIh3Xvts!sUZt-hu7TMZqQpTZeqPAm^e3j7$zMzYP60X3mC)Zgliw$vOcA$_@>er- zVQEpbkY({=afh$apEYBd`J|)##`F=C@cBN4f-?*DGJx_QHnc2g1k`?zWf_g5>7QZ< zT{6*g`1vTD_c1Oe(aaP4lHL7j)4Or}hvq>g-wfMCFOEfWi8ESAxo9+r z>lUA>oyYgdErgkK7r>l5hrw}NH1z)*#FsCx;&fk^L6=I2IM6l`di~@O>|Z!&Nr#lE z_^FZ18#!20|5jvN)yL-;J*Nzf4Ya=45EGK!&`)q~_cy-*@x%J;>jH%y$>2Cq%tvro zQ}e1c+oT1i|8VA7-#z7R4CN@~;zhCCwR~<}vcRTp&47!17$?3k2cXIfRQyJf*}I3J zJ2;!qP*$S`!RawWX^BI{;_OnNN=A(*+_)dnot*NML?P?(#=hF%f_TA;U(}4bWWD?% zZ{@d{uNc!!n38#P&k?*`4shcS%0aBjLTbM2%+;>C#f8|!K(M+I zTr7>@13um&>AJz>Q58gH|96MS4~ z!8e5TuN&cR{7~BKxq+^lM{pS-8N7e_3DIkvGSO0Rcaf~X^Prr$e0}|0-gfWLQbqkN z^1Z}!_Givhd>hC1j!_ebL^bl)3bJ6A<3-WiI71qpcb~ge(#emvdk5pAJn?9Fo_OHq zGeV~29M~LQ%*_muVPWG<8S|;4{>f|kLODa)I98F)rJ2$jgEqST?i05@RPf-$4M1-C zAF}8>#5vXaaeLR#fiAUp2(;A$(NJf8LhuH@E%>N-y7qi7rm1o0X zhZq?9b_$fdlP2%_H8dqW0+tLcf|~q9nh`m@WJq%@cW~TSeoOT^%35CvJI$6;p?<13 z?}r{mNX-y^+dCPwOE!!D-V|I*7iHM=4Wk7v(KYdUEgkOFifoAZFBrzix{=)fdFMW< zG{|Z`{?-tD#=njUJtu#t8gZ37*FA!7I&Z_(IUl3Fi*(^XOL@nd0(-|OQ%z_;5r@ki zys>%qSG?USEuJ$j3-9O6#UisLT;v@CX?lYlzh66qF2~Yv#y=hO{JR7PM>NBaX@b)* z{2}(!SjF#Gw55*ckr*Zlr^Eg#j`CiU*?H{3JCSj?J53wkc?R>34Gv@Y{-bFA(HYCX z$KbJ^uawxOjE)X{{Odpu^zsI1ioS%)yL`~TU^=!4JM@6>->`kVHyVrYVMFI%aQtNA z=vsXlyN=(c_%COn({2;ZTu_A}f#xXo2*;N{f|->35^VN*iKg1AXk2W7OI)|(t=J3b zSh@p0_om~cm>!yP;sE$9y^C)`udylepZRyY(y;A!Jn(9xMRRwGK~>;wCOh8|eczXb zW|az#eleAJVE#cI?06h)R$RrTMZ-}lAsFs1J%Cj$W|(C15X;Zjz{~GfQ2pT%ere?j zw!5D2gU?3zE&CSM?wo~XQA4r*&TwwOv;?;uoQB!-mR|1=I?9EKIOx>ra;^m3~Gbgg-+ydNZCQbvC_jzw6I--V{-d2OZg4v4jW9P zL`R`O;VG6z?!??OA*uPP8y>Zu!>!l%a|#PK;w|+f;*7ov+}U(z7@_wGEer2}37?2@ zi7x!vE?GSLI|&CpJV@sXrQqvNE&TP0;QFw5w7&TrWBOHqv+hrpc38pje5N`c-B?S% zG@OVxts$qAwwU9JX>V&WtLelng!Gd80pG93nz)u{*^U z--cmI8({pNKE7bgLd^ZDju|SIc=-8Hw09th?UuZXl{2R-y_ljX*aXsR1Wiurk zEeA-9eWTg%Y6Vs@e>0h1c*n1}uOe|#%U$B(|y4ij5ch zvwyge$q!8xnRhBkj*l~A3hr6#$tVxzz9X7#sy)d>0}NOI1+z;|>8vEHi|zM5O;)Zx zED#JO6#*cz8~2+9pEH%@U!KIaC=A6}uIE_ij3OBC<4P;+hqB+rE^N`gMXc||ez;XQ!}(kyq$(y$=7{t%T~jS_D6F1|MN1&&mdc zk-@&1Y?pz)MA~0)>`WUVx&P!7>-BYKRp$ekiKZNz7V?^&{2I&Fl*N*!-+2B<<~VlY znZV}Do6lF6t%ZLdA28kWG38x z;mqU2BRc>7JJoEEuq`Rs?6=1yrek;?E`(iY5JSS(QFR6<7#CCM{3H_6YO!>zU*43%N!iEssl;rdM-5ycR z=oh^5ePh<@`kofQJIelfnSr&yHS;qZF4;ElC|M}&V5>+~QoYDrqPc6BL~r?EiAm~l zw(g8C8@+oqyELzh#%B4mnSw9gLMfmAbq9b?P3{5S(L)=K!w^tXeq{Tpa>mu9a<2~21oJr=&! zo3cjM(Q>W_7QLFr9&O2HH`Lm=w$Eqj!}8wJ9nTeb(iGz-rp!JHj6AiSZY=Xp1}j?| zPXlUavZtqv8JL(8D->q*S!&F@F%O=Ojb?L?%x0-c>D-B1#;_n%0Gk>*vFXne>HC&Z zY|`Ritb4rybJ_ozJs1{3v-DKiH0KK}V}LTdw?3a-W^ZLvvc;^!-hk~fME92LvbW2gP-sO+Ywh|Y%#)D16N|xTzL;4&2`C+fGkij|P zRO+9woO+&x6zfSO@iKHM_XOF_8AU-erm|uG>B7;I#nkPq&e;_EaYLTkvW1@Ic;dSv z`|?4CYAu~`*{?fj6xxk_rK{j(({B;nHUA~_%wy@!1^3*vmb;X$!x#GeXo@(7w%_O<^Jhr&_RG9Vu+L=4cew(3F;7?dxv^14r2XM6d|iNlzUe;JL|DP;#DL)ezY%`9(~JR20n;mKdx zkg)m$>)d>twpC4Io1Ej>NA&^hTvP*#ymy;1%NHyXVNmHjQ18 z=@qMK=}PvEyhSs*T^NR3XU~N!ak{}b6PKUxqy;Ce!CTsmJiU*@hH=t(Sm2|% z*R937-V0%XYz53#>0s(j+t{tWjr6D_l%lIo;iOZY#8hI}3A5JmY~GJde(|quG%~({XhI;X4zVD& zJkD56j@ziR2Hs=0v(z!|yd{&8#9!u=3)l0s5El*9R3)m>r zo$S(zEV1npXMuHOg0X2aw4HxOAC(>GA82b` z1%<{5%!bJ~DDYt`v)}!V-J383H?Q8$2NaG(FC8%(ZY6LbjK|`cr~ToMMGiYW!vW0; z2C}kzQ#QS?g!PZz#r@|HM?)VWe==$+X>JN9xk(}Tb<;a2Q-2PBgG<2v`Dh3W+0F|6 z2S`?AOvRof1K3}Y6?>_3A4XK&M zntdd1-(-K!GY@jI;&~M>X~JMD_|Y{Oh97?q4t2^f?%Zw24PGdG_x8ez*YRW}_}an_ z6`^;VKjto4jTK97z!AM@#+CVT(W19>?Ds`_-1d`1cP>G(S{ADbnF9f9T4|u~L+Vs{ z$E{r%%1ymy&#oMZWxtEQ(EMFvneCRNxKB7c7#IAY-%n>i$umRtX{t5Tu`UKsc4XU| z=CMak+HBIcf$W3NAV_V{?ZYepi!xsjk>{@?ze3}Xd&N;{a ziHj(2fQb74dcbe`?#7aAp7Z_hTx5?V{e%woW(?2v;rnV=(2kXlspTxN&KFsjA@!9; z=IBed9aLmi^R&@n{ee9`KLItM~f}-m&Aya zM!$vH5l^AbE}G6X-KEkCh3sU=adMzoW>p@>?k}xppXZr?LHjl)$x=f1vH7gC@Clcp z{hX|lz1gskQW)J*RQlNVB{cmEX4(G<_wTxL5HIbA{Xf|-JN2Rn^)N1&^)wLVp>-O=?ui*zoe0xDU(Meph_5|@Di&1RKl}a$I zYNuZBO%5xaARjNt!MqHRxJ~!-@VMbRRuR^Y%bGUobAxEQUgve z*h<++l1iLg!sRhg`H(_zzdbn}4E$+CT&&AGKE_&>Bn(b_H zVTbMwgpr?AU|N9&tX#GV3Z@*Qj*CZV<#-v49~(tqI+AEQ(_nw))0tc@5?_!xTF%=+O=3UwrZ0 zZ{(%gB#y6f!gRq6+bmm4i%h#{#)rvhU^R!<-q)utr!%aJbA^rif1yuC z4R33!a&i^}McZf9;gu`0xN3s%eZK(guZo^+ySqzJe)G1J>icZK1yM*proW-gk zOkvK|MEPqJqyMt=~)x?h3qUZdxF#K>M+!>ul zUNMSn$V)x8|I|&mJ$fkSwm+nTO}F7i`60ncsLniP4lqm0!6di&AT_!x;-s`r%HMm6 zLT`%6EBGY$SK$g5F+&G?R!^hM_WR=d6$>~?T^WDKCY*&F83LVeFOW(7VVtxjn$&Gn z1b4O?%oM$*jF)w^d+u-?dCiFHHvI>Ro6>MuX;}pK*`M_ks^S;nbWQQTbf} z9*|ndTtjR~YyUe?2%3PADML^%E|d7x>9{3O6Ea#xu^(?DXy%g*5F2N~;uJa|uPKVW zrTm~n=-tZnTQ7L+5*TMN40T-d(CSPrKYe@^1tt6hjm6_J{{1X^9Td-=9GE3?n|O>4 zSGLop^XJ*bS*kE?{yQqQO)br;jbu1KmrGChLkHcOX$)0yrV8_zbwN5Wb-W*j)d`-h z7>Rgw#b|8%IR+!lPeZ$^JXvL&WZ9D}uyCLh9G^3l`W9Tq*&zeO|BlyEpV=Ciqi>Fi z$4kh;P8yv5Gb3?=3I@Pqan`p;hz<&RxT3OYJ*>#TEc7R@@=?V1-~RePnAUKCww$wNZ&Ph)UidS}RhPqr_Zif$&KX>{ z?Ssu$mGG%Gob*qbfU@w}{1O@b{8Yky)>+9FiYmcRaQ`ZO5_36*PGBu?x-RPc1sUmM z6foY33U}wwFEckVPx%D~4+HS>x;zl|PvGaBUL=lKGh6)oqcQ}{S_w7ha-ro}JAY=* z40f_d#9pkv3we)Lu|*$8z>$xiNW-#=vsti|?HP8R^4`^u+QIXjP?cme*UE4X84sa2 zcnkVY+(4=6JRBd)V_f75l$jL+8#9WyrGX4fruV~xyDDJQuy+vb`4BVTxUqW6H~8Yb zk>k1TaacOn6CLbJVb8;{Xf$9E`p2(F3sqB0QFL?M!rz8f7mdYXgD+!3_C~Z``xup~ zKjQNaeUYqcG_+dZgX%5~RF-+oFRQqO3w8)+{!;>{yRZkRKDi<)(A){u3a8QY^BwrG zMFKPG^>Ae4NAwrwpPsJ^VV+G6_5~*41>a&EW%VB2>(g;^^mVNHlLxOB_+W|Ucd&2& zfwi5Rak;0JqsHLVD7&*B`&nLsy4D*wqFCB-@6aIZouTenD0KmoYiXM*Aj#dv}z|MQGkxv|mG2h+M>_RCjJ#Inoe?jcC z^L_Yca0xQ54#JWzCowhUI=&q|2|w04W6kQ@Skjn?JH5MbpNFhtK@{Q!%^PIsD)gG} zCb9Rw_Myj%<+$QzBo^!CfaSdf_@_MpV@t=Oc(VoG_*#YsYNv|i<{!q>H!@J!(gZ>c z!tuMu0(^3OBc2iqF4nF+u(A9o-g3WzS`|e+w?7&Om#)M7g&DX%XAvHby@#0&*U)zP zYbr=hqo=l$aY&Xd1a1q#j>(x2rVx&eOBP|A-Fa-<5r|(D58RTiXSPHK6y}I6O2w8e39KasCsA z7n+oC&Gz}IP&3t0);$-MfBI3>nR9r`B>|H+7T}t>Q!)3~LbS{ugRR#C@QHAz_!5>0 z^+}oFcv>Epx@(|Sl5m!Ay9-ZFc;eDe0;_m)3K~{uI=U5kF`7sTF>*S&S|p<8ay8`K(wc8M@y~P}k5Fhejvh$s>au=S2O*Q!6fjug-at z&b=iZE$4Ax<5iIsx1%qY$k`M;fSNUT@WrYKVPAUzT}_9f?iNo7JDZKuH}~NTQ(=!$ zZ0Pu2SIzP6??X7Y;61Dp+|d0^iN!jRz__-b(+4m~sp!(z@Mr*epEkB-OwEgHBxsRMmc4r5g=V?bdtw0d!v zy}1-Vbk^cVD>Lj6T|@J5YqY3*2vUM)VAXF;)ZXuiqomax)prs`(py}=%oNx3yNx%E zHezZ{04mL#fK3I$Y%OcNW3Kff{HE^;quk?g*wpEm{`4drmmbbD>G3eJDw4X#js%&j zOCe;3Gsyb4!jF%G@ksP~`oS$G5buPiU&i64oV8G;T>ziwHu1AgOu_SedU-yhhMzgj zgykt5fH=2ufgO7fM(#=Cl=fZcn)7_1SW6Zo>s+z_-HGt4@fcJWUtlG472XKXZG1p7 z*k}%-`%4Aim%cQY7B7apB^J2$rH=TK%0NnNOr|WIHh7(AA>`HDNcH9bcFboi5B0~m z_`^!k*7t^Nys>DZs{0t4Zaa1Oo54@q2Zh`LG+agi56oTOvIh z7u#!rJr;|DK3;(hT7BaDFGBagA{xFwwq_5$+<~!N$?GyHjdf-6xjBkR^TP^eTS>N#wN z`@GfwI%q*s*b;oT-w&rfSqO2B21J@$IltcJxKsKAWbb~>uQRvBOARStKV1n;ryoKt z?G-jExL;WXf6YDzx^M1rpQO^@xltv_$(M3+`$3;$lBkSLeCsi{Eokru+q?76UXqV29;HCSaM?3(oDN zBlfrk;?)?$^FP&S!VNRXx~_&3BG2HQ-JxVso5`8(ybiIeY`LnIuaIeSi+{OtHTaoU zVc++C;9RCh#*-F9S|v1jB@b^}CgQYj55$tRbDWf_0==;p_KKCa zX~!!+Fj)MVCYEWFLD?!W8nGXs!2~@kd-+~%4pmib@UrDmPTE)>w<-+|$@<5|VD_lHk7x#75BKTo{ z2s+Ykfy=f@XuI7Tb_~wuT%@)`H^=xFdyBY~zrN^v!xdUHn_!23p~K8K^WdKOU;5#w ziLQP@+;m@eoEkTjVsa1A{XZ?7oMbT{f93{NU&{fDL*BU6PT1iJ-a|VfSKXGdjY@ld zg3sR!{2km5GhdFw4(<*&WY#$PA4lim&*k_2@l+BCnNemWgo^S$*R4TF5>Z-;hKh!k zs3fvVM#D@r2pOS)_qmR;GEzpQ`6-o=mLi(U@BaS&0uSEa_kGTFyz^_R+Y?UnJe}UjbTOs2WhR0DciPSC9@cA#@KQp>Gdmw7eWy; z_uZb*UYU(PIj7i`kK(K|WeT%7Yld4*7qI+q3M^>zO5zcbO?x87fOyk3_Mqtx!y<7e zVLFioh~?8MHcI zA?E@ice9X0SWaXKu7j}ZmnlZ>nudo*II^B;ANY6hcN)FuCC*vvAs9L42sTJfWk&?^Jq0(a4SDB{74vw; zvAN|VNaIvd4Cbuy#``WT8Vtnd)wQTHa{`;N`!HLWaE{E2S70^fHf*cST=vR~k^G#0 zz~$(1r}UR`ImW^4gP$?Ti`3$W8SS(oKNt-r6r)5*9G>2~1a|()!=`=(R$6!(lUEO7 z@6%vb94c1&EItfuJ@zs8kq-F0bq$1M9Kl{uEA)G|9oMdTP9=V=7Uqjeuy;{QaKAJE zH|p)7p6foMm01m_1|GpQm+xc)pG~^Zkc3Ht9yI5Z5!QYy#J^isF`GI2869k)9gg>f zalUDEtN3Gbcjk5s?p9>kJ}&G`VLYoJm{jW4{F?diUd*DM_rgP!<7gwFM0317Sh>bD zOkW?3ucK43H;d{T)p_JB91!>T@4PG|}7qIXG|bBIf->nJMshQcKqu z_|g=~78NM7+WsiEaC{-!_1Urag|bv-O%jHV)x$4KT<}$69(`s}h&N;wum%4uV)t&Z zMX4@ZQdB3zrsyMVQ@JBsYvc$^=lwyGc}HMI$|I6D`ZrFDmZi#h`evJxgV<&MJJYtz zkx8`_;EB%F5SzS$eLOq{TLP!z+e3(9B|@G@;zRajq?sicRH2XEGg^LrpU^JNoHZ|9 z2vX_JcxaUbv#`CumQ>AWpG()V?bnr=1)p0|)i-9j(wcbYK`6QXdpgw0c|n#N!_Obu zaHix~Hub-`H0#L}GQ)5_9vErN4rPyGW|L;HzyK5Kdh{3$eB){RT{rOciH*?fGlykn zO%^Ojt|2jLhwyf2F@&uQVR?K{(n|FPJMdeGOQ!0xe_ssPYxNyWclmzyR_z-r{3^#P zcEymF_1Ed5WDz*_)EYHazQPJpi)VFZz&);n9+sa7#&hME)ln-JX1p2QmTR%A+18+4 zsDiJT?Pe|(d8{KxvqXZOhT(}{aOEUPCcd6xUspV8MDFGeSFT{grk*78pMdA%^kAIf zBqT=mtXe^a8R-1L8GBd5yObgBSiuGEEv=`EqO+i4$5*zbb{~UoW2*D^F3!$Ag8xZQ zV7q=^7FNx&;+!V^MIFD3p!hk7$`&lc!=>?X?GN7*tF$BXiq9~!Y$UxX4Ciunz2S1Y zKNq*aiOaUz0ipaE{(I&v%#M9emX+wSlyzZDdgLv%`VxU%u`0~4tsLxT{b2roby(`p zXmB2R1b?kr%AR}k&NlyOZ2j!U-d7SfY$#5opS!WyX3p4odc=EnsS? z4Z`qxP#OG<*S1Z^fth01x@ryf7>kz1(d9UIV=ktj-7QoxEd;lV8BF=$FzN;IXS<;= z*wE7qw`+VLUoQjnb8gblOGdIo6S~o7;z+9aa4A`&*$d(y*MNcLOPc&shWqGt48vNK zOS?Reu(I;6c^x5Y;|gqSt|MP8FQT%-T4rxr zL8C+-(NdE*R_S^j-i*y;;~sFdqVEq`v{)AD#~n1iv}^F{M0vT^x_Eu;nt}uVSX!n`vWfCU&G{V07GC(yn!n%)K4M?eN?V z9xF^>uc;4;CWYX*HS_V{rb;q6kdAU!9cbm%lUQv&53g7R3KgIv|=#lKat^|K?UcmL$X|WJL!t!I6F&on$3`o#r#%{HES}PMCKTAZl zJ$fwbxIOpAr;*y;e#&L-t!55ojyTK9jy1*Afyc(rxTSOp@AC=Ag3A^x@xL;f^WB>n zTb;!jt%~f~h+ACPd;yF~ieTT&x8ipVJ$w@$MTN~vVS|1l=FsWfaNBY+ErZW>Ji0?4 zPFE!XM%v`H)J@PgX+y(=WIVWBo$H(6#SO@&b4d;sEXl47W^B8Ki*(Q662AMq#_JpJ z=#2*H<=3$LY%Y6w^8{#^X|nWwHTG#r9-oVfEQyeRMI8#HSjgRXXtMYX7%66g%?vrV z?Q9rRnKY4!PT7rO>SG{O>>;+^GUaNtOlUx=4vYFF1c#EzO!k^I7Z&`AsPN~&&=z~9 z{79G0dnknDOc@xf9m`HAB%@EV9aBr~pjOISm{+37>OX8@?F%Qd*>0C$Tb%{Fcd8k8 zN~$rX)oN__eGz7KcQn5{%)z6k%^B}+w5kJF_KZznT(e%6W50>sbn$=ijaW1`78frVP3^Z_R6G^=cmd+;M7odzkEF=m|N4k zAvL(bemnaS>&{9hOK_IgEHO9dJ5GwYfTPStvpds`aMfHTHs#2CN{@;%;SNK5@p>`4 zIr9uu_nt*(=>`nXFlI@=f8s3fQ~b<+4zo@=2}uvc*|o9<)FrEw1)W7Kcy?1T`G*+0 z1>@N1)FRY&olhKnc*m%&27bA$M_#UR#MrhXCe^3R);S)>d&d_DTUiy~pH9Q-^qr`E zKpi4PD+O8spUKum`RF`z9)`c;xfi?6lA!Wts-iUkoP9WUHX{sAX*H3V28e#?&+tav z0!;LHkAArt?0j+yF8aQRK5d$WHYfE-!o?1}<1(8KS|_o{;1SHpG>=SeEXS4iFXNVi zUCeYzreM>gdd#;Q!qCZA(CpbqoDj1as}|X_&CQd!WWQRl8TT2Be1gGtLj>D$|1+BC zPs448Us1J*o^0Qx(QL)!4Q$1dcCd*JV&aY3?0%anlej&W$@%bi<(ofo_*oW~o=xFd z^%`t-LMChHJ!LyKw&J_oNVGHw!#SlvY~b}5`sCddmVR*sgEyDVXDfDYK}Y>ktl5sdSC-*ND^ct_ zI~(`#{nnT}AFxAdC(C+tA6>?3u-Iy8GOtLLcSC;0(+X#3lX@_YRtJck-bhM&!~-Qy~Nh=**(G8WwiQ@JYD=p4Q2O@ zWtp?~!lk_t#7mGSn7?jgLcR%A5RL2LxP|S(^wjTo&iT|J{`q>2o$`=iH1rl}$lr{QcJZA7 zuM1|rYEg8Db~IbCs7%nk|2!T%SwzF;9YPI<*?3{+Jf^Q|!JgX~qwAwkjCnbVtu?=e zP48-iC-4}q|G5FbJ)OylXZW$?U@KtA4|6H`#aCjA#s;y!}`xnAu_XU``bS?SZd>Q&aRKN$vMbt4=iaQ4jx#xe7M1LB^ zV)s2`z%rdbBnQHiz;7%pI8*sIUOMCbuGmBy%-m`@+d^TcY!0KlH|(VBf?0Z>tJ;) zfx8@i7!z~{$Y6dTHJ|E4I(0E*}wUXSSWj7$smx9XdSmM7qtwhW3 zFmalu$VKEC+D7wc{pdLnu#L2hxmsvf$_PGzKnJwH`b3@!|FbDFcox_Ch8c^{E zgz>-l9^~?Va$aF2(KJ{Lt2Fzm-lAV*Yx#HbPc3#%Vtgp>y}S+9ymN)oAtSh}(nDmM zyc+D;{~n&6x&^7{uE4ai9+E^az>4lj$o0HS`sT{;If#DJvPm6F4pyUoyeaMP-OcGN zy3EDaj|WxJ6e=9u2udplp)_|pRHQ}2Ov`X^mN^F{|NOY*u)9#{e1-)6QQ~$kd&%{< z_kmx7Jp8>q301zU!#*WTc3@}|8vmHb4Zr_E4*czcHiuBi+i?fHyZXq~WgHX*2;hoG zZSkN>zaVk?OgI2FbgB3}kn!Mzb)RM6b43RR%{mVuN_EI*PDdb)W^cu{2XXD znL>Ws`*I`865*KT5KQb!gQI`dSnS44XvFuScTO(h#%suPE#WKidZ9LOUkl0TbDzkd zdM&oZ#g%O8X(e6>XE~EMCM8Q7mvi2ef?3t8hxmtg^=L$062$fALgUzQc=Y%I%rEu8 zThlJ`<)yoLKg=G~I_x3opB@sItypT^U!0TXflpr#nOR+#01r(mT;DsE`%x;#rT1Kc zZ9MBy;jRs2M>(RZR~Wqi8%t+@PU6N?`cW<_fqT%igp>T&4Zka=2p1?KOuAk~BuW#Z zr#*vwzi01Nj*GI)me;LBTG6Z~XBl7|ah&g|fi0T;a5N^xf!E7(9FlYDP$MjoHo6bl?J% z6?${0Pt1l5_2yuyJ%WsQd=JKLO#(lcP_Xg*3=67fVA$~}3~1|sl@+b9|9u8Ovzo@G z%-+SVY}H_46;FuSpKx3gdWjC#Jb+`vUtpj@i6}C8G%&~pIJFG3z6U`00}(nf=*CUe zTv~GC!W+R7^Llz@eQL=p*=YP(vKF5Gut1;0HE6g_pWC(W3nVG&v!R+(gv?gr=ZN=d z%-02A=9>Ux$|sPuUz=c0SRh(!9>fh(!imkjp(bKg#k^{+{10AknDC%Xm?T+ z_s{Yo9UqOj9;L<1jDOd}ce#^yGnLuGu4h!HSDd@ua2uqmUC4i@vuJeAH7Lzc@Elrc5Uu-W^5xH9OJ7<6{kctu(i{?RpZa$53|Br^94Mz3XAE3bf zCsZ72p$UoV+-wIUc=7Hl4i)?smT*;|+ANOBjTYp|q1}*hB!uo?n+bbgS5w8TmE_yT z^^l@lON)YE()mY9u>9;PW+x%Z#urF)T0GN{bIii=JKjOwQ6KzX_5`wj9p~1p*#_#P z*5ISAP&8Yt!bLus%N_2yMnqpp^zM9^fup2FO zro;SVO;9NNkKT#fNsNQ)Nv@(J^KIIXx+|9vvr-N2VAC47;yR9NcKQYdl{s8<%M1Km zGXl0IC~`@^Lphm&;9JQ~&tOxwDyKOv2xToqP?_wc$){}a?rmczdQeDXPtFFH`3B_Y zY%Bb4OBmV|2Sc&A4{mGU2pJmy()$LeoP`#2C|sZ`M^*~2Pihcs$vOi*3%%fi`ZXHj z7Ytz;B}lK?;wb|W=63NTEtx+LuB+Vw?{gzKi!xPtM~W zrWU}9nF4lT`{+`K^j$oMT9-RscopRho)gK=F!cP@2w9;giI_+e)x7mZ`0?Ty~xSHg&!EAe~VbjVokhx2TBbGq|L^2nWTqkZTyn?pH zayafyG$fy_BL}?(K!5uXX&Wr%w(zpNS((n<1mhFblkCPrjhjK&zDu}#vORv@C4&tw zZi4Q~0Jt+}969lOJ*8sNc&YI%%zG-zK%)~L?Kxkf>$Z~Zqyc!T_(RE}SZ`FG;7Sbc z7LehvduTqhg}S&?n(`=&mGmeGZns8(%C>Ph;kz~V+gPFutAO(cleh}Ct8n_{SQ_)_ zJhKih!mmU6+}5v2xZ$`e_h#-L)LX*ei)Qg%0i$NB_k1I><6SfZGw)+;v_3sNYlsdx zZASePKbrk$0hacCK?9q)sO=}mUCvU#y@nt0SicXIyJ5}x^!`$p|7I2E3A8~sP7afP zYvN#OvcO{F1xy|siJb-^+{`y|@IG3fONzWpx@UgCH(q&Yqs!0HaSD~1Gl#BsFyX?? zr(>jADZF@bj740^!D{JBJQuu}i!YwdnO^Yb-t&C!9bK!yJTC(rU7PTdMUl{d>kC3w z{3WtyW^hunxpXYEC4F`0As5kf-^z+*f=>15|$ zoMbuzT1Icf8!qbjNXHIezOI3`wQJx@xEr`0-GG6iZi3DENISC+Bm7OKzP$5JA!aZ3 z>W7(4EB52VwOTl- zf@ceJK86Z^3>MmhV%IF-c-0+S&VDJkv0a-FzL+XSAKUq1ZHdL7w67&m6bi`4o4uRD@NQpQ00IlnKY2o``w%uZTln zx6t+M91L5x0i8?SY1N8yvhrjE{O7Al!c&}K)r%wioX~&vrC0x9@yt=&szV7M z@fOc-KWPpD_E#a${V0xhE~hFpFYqje+4$RBn>|n4K`J^r(UpI9Dt*YLo)HCDrMCjc z(5ZrZ6%$yH;1cb%T*_)^NAVf=0=(n2iiC7Kf_>&9*mKIcq`J|T>+5Kt9(O-OpMw;q zGWnfwws8vlZ(qz074c^5L~NsK%;-)TetF}IWbO%^(`<@S0dMKD!-wgSqdy_D&K1QH z<=Oj(Z;40yByPG`5t`aRg{Jupmc?+d$t`p%u8SEe0i{Z8w5LR&-irfHph?)s1QZ*oZ`yEtaX?%8o)eBhI#3)p)V!L6JqF1_HFxgQ8M1X_3Nca^#XE+D#JGK!RLr}{z7#pKpPzM6B=H6l zy-kR0fFh=yeNAoFodcH@mhdU5o9&)Gg|o@k<2r^GvUGzHrN7T^K+%Blc)wkWRYxTf zH=Rz_AsE5ZH>Jbj4pTV(g23(iC4?|uYKC)Sh z1L~4Yz3?XWtR-xYYAVw++>6u7+n9B7CHCqbKrY9gD-xMqx{!P(mu^`Lo3tK+`pNZR zB2_^y9LfOy-kb2VU^5EFOyzDr$;1bpvzS7pJ-VDr1a-YMCURJY9glMrj_mmgaf|p~ z=H6hqZeGS6yz~~Go|eH}1H#Uv>JaI%Ie0tR1UBruL8iR*N8x`SaHlnzpLbow-U+en zjGP`k{ka^s24#}P13yTY_Hop$JHwo|j%59hgSgsRqxlS$6|iyRn8g_-Vmo0x3>9>f zNb;TOblFphbECPPH$RYZ9~Id9h~pqWuN#{-PX(2zJ~;Z!kol7mmFDEYab zocyfGmbQ<<)tyb)_FJhMIv2 z+Wymub1stL=AY}pg;yJ>wd!p=P+x>awX?86KR{^rt(1M|Gslw8qcJaY9qg@j5yY7O z#;jq)ecf8nQ+EZOR#u|-t+8Nt&47#h7Q+I!#9(e@JZS4p1oeY^p|5l$G09QG)~?NP z=Fu7uoVTXUJHy$4gar%va|biZWWh6!qMeA429?cVlk4S3xD@Xf{N)9QdUCjUUv0A& zYx9}1SUGan(pGe_9g$n|^gZ?z;!wFXMjzF1#3)|We&l>)mLFb>vth+vijm>|6Q#RQ`*4->n zX!!;+Yto7Li$aJPXok@q9kjzX0{`+qf2l++K7LG?cep$=&wR;R{I{^!qzvH&Lrbdg zN*?6DZHDO)Gtu_-VeIg7V!Fp9OWnu$@w<~__yO1Azq&!;>USrpMfW(C*`JQij`_IL zqLa!rl_5z6x;a$=gaePU&My{L?D$)S&;c&L6%aOz;aDAsfv+c)>buGdCU^OpBzWj_WZ|=o;-#f&l;Hi z$tWsjorNvDa>1bK7P&Sf9cm9v2eXBburfNBwREQ8uPSRCFW7@wZCz~TEb-F1$$+n0 z%88G|T$W~N2B{CaX!61L#O>xmIQ`j#i5(b;2W)Rp9p9P74ZJre$jwU7P^K>;Rl4>|8A{nVLWzDA-KAiT(&sLM{?Y|-WAG0Uqs%3%FfCq;`qY^Cp|EN*_*Y{=AaW@VQ@n*Ys~_(_ryf_)G3GEi^hR2 z-(6b*qnMph5|*3nCdV$U76$t4Ap^U=!}jFR56^W<8H zpU^3^&(nZr)721qUJ9Sat5LP_al-nLN_6fB!m?Fq`24FpyS>jAwM6CMRq+^f-i>5* z>=k+_Z#wtmmOL)|R1YK1OHkFy(`4gf0{Qt0L@uUWu=^+9ar7>yzuilOj!EG-e7PBK zeZEcR6x)#Q;$y%mFU8g)3ov?vJsb#`f|}+5w9Q!x-`@%6=Wh$xjB!KwGdv9oy!(iV zlr~%yaDmPOx*+3KWeP%a+{wRI)oN23I%rqx$a+FyFnErttm36NdCG z|3>HfuD}<)>GUkAqJK)qy3zrzi@Zw!z=! z?|3KOG&=w3SPhTdz>!j z-$weVCyWRiWU%&E3$-6m=3+EdVDIui@=nbTrIQTkjPU8qZ@xPUF7R}~nrM7glF4($ zoG^0L4EEBU=X+_6hc%KS+?@egI*-qd2Kb-GnTPrF-6&V`t3e=mBphxK^qL0 zT|*>BpTh5RmqVyiJ#}%hh1#W&U}|GRx41>4lEh>lHcrZWKo!g!Kv6?fptx%ah>?&*8Cs^8m@eQ81!AM;N+x7u`E~ z2Tqn>NX|tHg+$^UdF&wI7CqrJFnkv4?rM9;GI)SAUI_Q@&!MNCTS3>kh5DwAAWwxm z1r}9L1boiRY*_CH9@sLOhONjhxjE+(F6~{5OP?5Guik0;KJcz^-HJZ^KrFCtl_LJ) z&*S?0RAH;zA=2t0pt7<+_HWY{jGo^^d;Z%D!hPo;L(P;b21*L*4~#|6eLAqPbqDOQ zE~TxVK?2{^PXvE_XVAynj={&KeK=s1PRdSxAQjryuwQ)?IX?WB(B0>RHX;o|)xe7= zh`og>t&;2oe~!7K^pONpp@Um-d-C*F07kNypN8>eE$fnmNkRI$mLW?H8JO2rRJ;STGo_~mgs@oSn3(-dMLX4*Y0;2qG9 z78!C^{~jQJdS{aEu@>~6>@(_fW+ye8^h5YJy_&pLI%_tw^&0wINT>HF+$7m{?r`yT zHpf%3*?aT^e%G`{gMiEAQ>2z)lKdUI?{E)Rbx(plTW`^Q@tv6XTM1tJ zp5-bh&P6#-3xU(Gbr7JrkVdHP65iC4hGyep`X$E=emCvGg5_tRLH;s)*B6C`#;auX ziR;u*$pC)okD>ma^Qg`Q6ExQ`#Mztdpz`W9`lQ^Byg) zjc-PfXIc4lnlO*&;2$C4%hqr%I`g<6qc)>vR5{hHswMXfE(65AbmXP1pgpyr- zVDYLK9!~eiGdW{{J(3V4e$J#`@UyTdfEy1{-kv`~u=bRH&+oIDGTk2yaV9 zQmHO8Fv>3`W$|Mm=TWjiYsxWdtt2A|Hkv}6^rT?l^Twi)2gJbK{;zOH={hi(;|wsr z5mYuVB$qE96OMT)1mC*zLh(2ckP6hsQ9EvsGajo;&SWgat}mhxzB+--e_Mi{6D8SR z?}x%*qm{5${FVUZ$HSCc`9ifz^2BhH8#(B@iCn!g8rS~~CQq!SVDN>9nZKwq>O7wV z!+)*lm38Z8>HS?PsFE-xTe}y-q3g1+et0S-az%JBR~hCegy6~e6?m*b4=@1*Ux&is z!lXi)uD6|>JQ_|dZyzM>BW<8@zp-G=`Z2I7zJWRsak%zcnVVYOP1fmLgRW$0*fiiw zMigfX{T=4hg|%Zq<4q|)TX_h1rize%VK#U8T@~&Ql_nFn-!qfbzeE%_tfrApUV>B-=7&EXL1FlM7zDY2DhSVf#+k=HabmE9mI}H^E3n8Y*4!))J6Oa6K z^6T+Onl)}bJ}Evg47g>9;tNjG%bx4u*i>gwo^ur1R;Cb*r9mK+FA#>t-xhAE+(0JD zI+Tn!9WBglo{DD@pOY^dd9)c<(uXolXqn_d1444>loQEx@t{5SjFyD;%cJ2y9KUh2 z8KfJFPvLDDKcPY_|2NDH1g9Tkxp|TnEMl{w@M=dXxi~OP?&eOSIS!-2H+nL6v@;Cu z&%Z2qzupBVKb3^y!6e$dR2d$ucM|ShVFa}%*NDaaF}Nl3IuSkbji$c~pl@~?LFJYt za<^^mMqne4<+n!*Q3xC~Ew=ir6t(rk+@Wao$>H zb-PzF=_PMa%&8C?rX9l;19cYXq0W9TsKX~6sdzo681?3ea5g)`K`DDIR)?*`eh+z8 zxWbfGec#E1`R@QW6w?)NMv^{rU@F387-&(2hC(WMR3acV&&9H(y#nks`3LRlJIN<9 zlPqe_Ko8q6?A~RB(wU=#ziMWYwAtOTq&1q%ma<{n)_tNc`Aj>zdkVF^{JHvsUefjZ zB3=vj!Cy}!g?rYY!W7ZX^u@GpJTDsss{V7>s@X}-fU>mDk2Jsmo=Zh+E(Qk1dW zi|W(mq5SGvnk$`69IM8I$}=@A)HjBo3)G<0!3R6voDtORG{R3G7O`z9axiC!0#_Pe zhGCzkGOw=v;4^v#-hW<*bMss9%V&mj=50jfQ<2zt<0DJLB zx&4cRaLt`ma#hih8cK<>OsZqX3mBo8XW_4(yH2=RbcPrRj>F%woAA`oA6s~z;PRg5WbBJ{ z-YKq29$)-~7yo?0)UbyrI)l%Lj%de(0C&h}c#W-Ej&pJ~XTgCx(cGgzFP4(@93;}u zg6LRrCdxYy6#S19O+GJjcYiewUeE^1zMPT`ueZT0C)4^v~x_v(eL_&!nXRAV^{MA?#gd|z*b zH)pBAg-?ZVvu?TuXLkY`1#Dn8=Bjyxm!hP-C z%pQIdVK(ZG==LHCvKG!`ZMCh%Gv6P_RNYKaFV+DEmkdGQY|3Xba`-I$ZB(`sVY_pF z!t9_3;?i4)2M1^3A{A$-iwY(^RbS95?iZy#=kP+cF*E&rM)>rh3TUQ(#TTZbq|;$Q zkayphnC?q}1AO*Pb@3(+0-tkT&pBAMG=j^{TL2Xv=a|)#)9k+48oX*zB-(vb@4s4rD3L9u$#P)gF zu&c}B*_!wn%-GunkC051GgN>s-A&|vvNl<&ugg{kcw)HtQB>g~*ouS_=C?1Li;B6= z6;&)|&&PYC-mgl0>}|-}BGi}z?-$>$eIIu=q+*WJZkD}c0=vCRi=7T=!n(7C%-#PW z^N1;BCfe_*{n99QeAfjs&PTxZ;1Qe{F30MEPT-e@bk^G;Q+nrv5}VyF!HjoLf;X~! z#z7#;?f+|zTaSjJ#G5HhbYC26kDJVV&%VUpD+*A#*^J#uJcgZrc&C${2wFclgj;-O zaEE-HnZ_JN7E&n1-oJSUKGD4U^89#i=k4R{5gk!_&9H?nPUSf>({dP_P)N<4!ZBxa z7lvHTV4^xIoaeklw8EpDob!FaGW74VrAaT@thmFtvF#;3V$STPh$4IX;R@TcU&u6N z?qlGtFOb?ff%9LcO3jjr$sxWDp>_2vR8MUn%w!Lm>MX$8-D0d{y98V6{g;Nk-@poZ z-`1x&dQ7(aE)y+SNE2iV*_d_t?0ujeOG#|NcC!(f_qw0nn$=70AL&Dv@*EVOokbQs zAI)jLGlBbyXEUSm;%tmz9h*LK3}}7lonyCd;8%r7%xdOTtcp)#4`-@@1Mg>V=lNS( zW=&*&lT%p$?%07Y?jA@z2a;*&5(U%*rXiNqF z?5`ts2UXz%-G%CMZ|H2}La1^3Nfe_R*&f@=_}b8s%^R=EcIyPQj`ci?RML@VMO%TA z_gJQ|ON^cCo+|E zaS{rvh4fh66gKJaNXGdouv*aow$?WkEidkXk0HivW`RE%jLO6Bbw|0@M?Lr?DwG{w z`vhn0FlHYPr_mnoQ(U@?7i-;l1M77)n5VcF^NX-$H!za98Y%MLuLejEX(7H5P2A0O zMVQ4OCEDMnl2bAnOvul@yZIQxxjd5~tV9o?wCX&zZ2ab~Zy3uT_?mTR&T8!!&zrfMY>G<<;HA?%P0Tj4Z4@nB^s?*csdi6+)Z)ZIR&Ot-GLFGj)KGPAly-yjc%cVP}$2jZRe-qLVt0g zM`a_l9?{0!cp=lC`iH);S`PcReMC!JX?pxpEKUrp0x3&A?~b~vE?CR?=pPa0(F1Gx<+XF+<525vJri9+rf*7^<#Qy(UgQeg#^xZf^> zwi7Tz?l4^}Hb<6V+7aqzzZ>bWh4RUVkZL{Gc0)rpRR zI5|T0`Bq~s#=sm80 zkZoDC)N3V)O1O&SYOa#B#yi;jUpsn_8bZtYaj5xtF06MdqxC~81rR)y6=!XQ!Y$kB z$AD8P@}m?&tk00XTQ~9c+F3Z;wiHV8SEF;j3M~{71@E7|I9zuc^)6aa@#)9VukR{m z=w87$i9&jN_egeq(40*;5XSBw{DpqM+tEf)#CL$MlT5ilP&^z9bNnYT63~JXo2PN5 zLFZ|;#Z+V_HW6+!}`$uRepp!h9@WAcH$Y zEAW!ac&@YBNBC>QSG+w=7cT`&V21O~g3-eu*xW16rYzQGrhkIjK-eM7I`ST`PEEwQ zZ%twEr#LW+I#2(ERpU#^kH{T)Mo*&?s~9 zrC95kf``VHVe--GkW!+KvrdOmrI2dO_70;ZO|2l=&`X}STT@#u*vyG{*v~BRg6sS| zQsa23K;yIvms4K*>N<%foiE}2EJ^VDaU|&&`2ib}gcvxmm>J3` zWAej;Tzz^n6nkr6*ziFZQ*s|Hy!24EeHiR!T;Qa>j$nRX&aC`uDcgFXjkJBx;SA@R z3Z2}x3dZ|CCf~SSY_zu~uAQ(SE=wi?4d9a*VK?F83RiX|(3aov+{Q9rjZ(SuCQvM$ z!i{nfWBLwCY{}+$j%GhBK4rTOeXrM}Vxt2KU-AkLy;Oi->K^DQ^$?C2`hm`aJKVvO zqgdr|nqVP>vlA!1G5%*G&iojRi-vpfVoDvyHGYJWsCRhYcq!GY)JNlJdp2EXI=8hZ zoEFJj1Neua!oD_Gy_sR?TL;`P&oe*2)ZmLKLqXN zx06AobfsWb>`}67=ndALe@ZR0+eyUKv)HkE4AU_c=cbD7=0;E7i5;_pnLR(3JGQ`# z6Udgsq|g~$qpLJ=pL3TgI;P`RX??UBJd5>?o0)yxOSpP03m4kzumb`1Fr&wW+wM7z z{hU6FTbKGA-)@=1%{`a`o0?|uD&i@e#gHnl9(e-obVb>|;DdzzoDBbM?I-O=oCKl4 z>cTC{BFUCH8({TuSM*w42B$Kt*s_xGc;&)K7IyBGz|lwwzD`qOhh+}p{zJ|X^!`oB zs`{to_t!rt8L&Ech_Nz`z=s23+D~~L_yfGNcM%?9ROcPkj&Uc`8M_kbSQJT}Yz9_IVoku2oE)b?T zCPP7E7YUi6h|9yQVP0M%Tn-t>)m+RMw5*B1`3rohrRQiEUfNl5&q)lA`sYzkr%u## zu7xN^A2!eED9hEE$-Gx)qQ`<%(k3&Bf!}#yZMq11q@F71>CS^+-vV)ue=xU2<0t98 zY>KjmYtiscH2N+d$w}PIfH>V=dgqfh{&^vXhqql529hlH9D-s$^Wk2JHIhf%7Ki(o+jmm~ZVI+?eP`+Cv%yA8x8) z`6fA7tg#ftm21E$ydC9kyuo{C#F(P+8%6|nL&(-KnER<3gvZm#jh_l^kN+Vy!L*W& zx-x+i`zFWjYrV~7@0!c-u{$lE&%39*hKTfZeRiZqjw$!6am`Ay-1wk)*4b>xEjV+W zHqdM=b6?2rE9&^W3SOuk97^(u?dU1lV{tDgjAz0;XnWWRvBn-8gqlQ_i}_pxkbCYWg7f>*Ly z?6Oz{XZ<=CRbCshxP^QN`;;xaB;JW@##fSSQD*GK@jhA@Ysbx%Tg^Q5=CSG#?Ejxn zbv`3XUBhp}!22H5mHKStSe@bmk0-9v9K&@CDZP>j(acEppkQS$%1`~ zaP7u2624Bu{BnOAmAxIu#)Q}4nNxgsOs@ux=+}13vU;76=y}=FLgMFBr z_7FUm-^Mxk{(}cgMws8LtKtkLUD1`FH(CTO#tqqZT=niToaHec?o{=1;pM;kSk=)# zKxdz3i*{6DQ}|2zY4K@%XjB3U)u!wtw&O>YU`U;E7}o8oqvI{CSjmciOxae24OEzL zZ63BznsvYr0D70gCkZpThpIT(EN9;_xTVCCh3d@l17tStTtTSR)m zLpcE)-gnYX`ynd2G)M2fXvK126Cy2;<}YySZKuO?^+W zu+|8c*Rqz|R2~7FMHSh$&&}-av~srnl{4(@y~66OoS?2+-`piPl*=43p4(^1J0bgw zIrpT+OfKdhnK*cf?I}FZf{Su+*=$R^;V_AlpX$u%?Rmht*e7%HE2nbqzWeaowyR9> zSpb{*rxHKxc!l#bidd}qeCGQpgf%DlabmS1Vwo0aw;%@AW=pXJKfVezVvn=WD^_FAx?Pw#qXz#?G+>FXn{Yyw3afRN z<)J8BtvS_MUi*@tBoD5*s)@v2K*Fqgmr$4 zC(qjwVPgFRfokGMm}7p7m8pBM<{1~5_%anpeaWz1BOa}Td4J`B&XRqvO}Ut*v$$UE zL{9!W?}yNIg*QJ#SyfXIT=z=g>Z^Vbr-m5L*i{0HlWl3X<~NFoCU{MI5`Cg?M86-O z1AfU3r03Z;=+Mj(l+8QI$^P8R^z-epByl3wx@7{kC;vfN^~W^hP`<#-e+OOgw1-xP z=2QE5vdr4?0xilNCf?=dAg0W-M*okZGYzNe?c%U9WljnygfyUm4Ch%}Nkkdaph?n1 z6BQ+;nG^~|sgxmPh)kLGvz|jKic%>>MWLeMUn&h~^6vMe51eydXP<39YyE!rtyKnn zF2xwqeH3z+7ZIlwd1%ppo;n?kAPuvmaej=DiTwVXZhdzNe-{~oD0L2rzP*V2JD*N| zJ~$1MaV@a?TOCZAqeKs1Yv;TF>*3Wi@+f9C(AalLMCm{&S!C5t{Ux(_`DRn>UNnX| zHpwuNUn*}cZ30;Z1Ms>#fG^J;D0gWxV6K*mSdp=h(w@w68<}vFb~2(R@3n|UmpOD; zn&9;EM3D(pQ{&hDxUTCouUMW^Hgc^ZLheM?z9kN8?-gRyw8=R7uQRSaSd9DhCBWZt zW9h8Dv+#TOYFy(ki;k08`OP1NeHaK#Gp}oOcAhO{K4_%Fwm+d2EgXiIh2UWQL$o;H zCzB)NDB2WgUUJU1+`G16spA`Ag z^9u3KD1Xp+a2bC{n30*4t6{jo8>$=Bj762YxX`tZ$WLUj#43v}?;lH--+n=_y>Xy> zWbV_I0yBMW`BpIbpg@-u-Gz(u=JBxt;H5jqnH-rv4}4y}B#-la(7t~M^*2$4+7erK zb?&gB!$?w$bULbmhatNrNoOa*Tr@T8@#n?wZ<>WR$tLnJ&)4+e$4qLreBqHUpR zv~BbfxNzhuX{jGgt6c8m+RI*K_EHtTM70#p2d<$mOH!z3Oe&qHQHElLHZpsy6fU)0 zgnIhNQRbjEyS+c03sQT9V^yv&?^D;2P8tnkWRKzF@eX9Q?i?ssnND~ANg-{vpNN%D z8a|Oo#hwRo^hI zCJxUo;JqV30m+iAI98B}WQkyX1FnKx}-ya~GSqDsHXXy3pa*>3~8&crHi^lAh z!Jm5q$oK9D+$a|acqbfNgd9&kd5&P7M0ZryP`7W(uvb2VUaagCX^1A{pz;UgoQW5| zJU*3KUBSZjN%_H`5UPObi=UPdl}S;r5O$Qh3P@<>#v5O4-4jA+N=4 z+YyAv=horFopQv|)rBwhy@t16Mv(#S5U4VKO1J%+2pvhZ{6pn1v%(H=uI3B6&P$Jlsnd0CmB&_hqmY7ZBQLrD?5$6(&zb z&RNBXeQPmoj}}Je$z#((X|65r8<~H526gRK#~JTj4h?$eK)^|%8~1Mv zT^!y?wgv{$eX*~IPuq3W+%At*rLvsPHYNC2B8~3SvShZW8lL~ug(JVJ!?3H(bl20x zXeR8|E*KBO%R$oYtO_IjW?4{kLbxCH3(g^l2K4M%h}pCCNzPY_FD8bAe*Qwy`pyK= zoef3w$v`bFt-J@VZ}k8Sg3)i(Xx>eioV5P904phped52Fyp1@UWY4D-HWya%@ zzEv=-tCq-54Tl9^EAU*l2kOimi_bRO((axHy7n3+&&Rwb8n)?lhjkb4(CN>8)V~D# zqK0xZUtFq!5o0o@gKW33f;e{5HV0%P|7H9;k@LbYROPp$Xod4;QNZtBU@5L3 zdBqX7`~9V1!snyfmqn#sNwaNLu4H<^Q`!)=0ObAbA>~@HX!%+XTtqIz2AgMO_J8ek ztz;IOx~9UVY#m7PaE1e3E%fCC3k>i&iKzpJ&~siQmSvcLf{GKpqLNPcKA8x+Up^3t zDv!ZE5-aRMD#`b|xpewQU;1C#O|mLa1=rFUu)lXa95|{7+Lszc$G;vYh5Lr%6yNt` za(Wnqb`;T<70-B6G8;5JgNRAG8OU$71B3Yo@J6sXyI8=m(@~fEes?JszvCY0G;P4R zX|m*tuh_~vM;`SjXT!N`YLKwvx8URx_NAlMK|1g;yxEk__rxXO1|xZ1Lsy%ZP#y|a zCF=D0<4oM~qF1zcmoIz}eWvdn#Gn;pg7@5$$>NEVxJSw#p>wJdX zhq~a9doOVCn#(YU3vSYXjkG+#4{N$jj(O+|+6-3b|icR7>I{#wgz@jbxCELIl3J#|_9uuehz zJ8318>A%Tl`nzDriis?DVWe2!@h0o;%%DS~@301q686gV28)&-$F}HrGljlG?5JZX z^J>_HpOb#kp6>~4-{=j@U89qEO}@gCewVR@f?GRWRfnlHIIyzSipM zD-UKBrp1DtEEU?-<`3+F{obT|?uR$706IS;6dU$yCqn-|E6{W4XA z3AGI~*EQu@Lk-y7N*VEs-lfclb7FNj)7Zh>?`-Y#S{8ZRmq{)fAIsRzu1#439(~FU8k7;>)q^i)_s=L5zQuz z`h(36HCc_p7uGQ0B7^+9?AxtOvD2Cb;yLn9nL$7ovN#Y|hUkgAo-AbXmZnVV>1R@+ z-^teJNQk2=!kGTPV1avIOkO^+#3@~e*lt~Uwjj%&Db6w#>&h0ebAC0mFQe?vfi!j_LqlxlnasvN%wzJ9vKHBx8o2Iq+^09qv+tox|F(!uPeR5_6x@}BqQ5XBs zdLR3;quAgzz3iG=KUui4fGyh5z=D(0S!lyTCYs%izAh=4`Nka&ZqsLpCd0+KL-JTd zeF(~F++p1sL&SNnguPRvH4Hs3FpPS<#m;JsWd=HnH(xp=)>Ts#-yhHrhlf{Tkf@Q} zGq7Xz_l3D*`f=9lmC0^Jzh%=M8!*Cc6Z@l9gWKOFveJB6Y7SQH?(rbz_9c#Gzer@~ zG+)8*vx?aCR*z}aR${^AEM{Qn$DXGOKBDm}#rdzVFiRO^4lT~&-{)n;5&j8m#AAE0 zM(`bW=-&@qrlrMBpIyx==6#@PapC-}$`BU&@hbKV*}#?!Qxy+!-zL`m`GTE4Dg`%R zK7g5S57{JYE7n)p!vZq`Sa^sxR;ssSx7=~~%fQ<8pB z?iM*+yeV=D`+QhdJZQogv6b9(@tTvH#p4fHi_guQAXfagTD&{_5kz@ZLy*wz@mko7 zkL4xtilseWxJZg?W!><_ycpgzti`GYY1~i46I_E+S=lV_B@kZ!6v~2@!^o<`fZtTX z#4-)vwS0vmLk+-R{Roxm(12t?Ixyu*Jhx!eG|uV^+JxVo4eOA>cp>@SYp>v#z%Y%1P<;Qa%zH>ac zylyI}2(0s)>BIS$r+eY3?>90qGYoX@rhuQ24gF<)gXu@;!o`{eu=9;ARknWseWD4F zcic#)pm$0aZcW}o?(E9~^Qr&e&*$^R z;p?1S%T4r=YIO}gI_d8n54IIlO4o$71-95ki{J2SeM#M&9uEQ;5;ApwHxD!J#)0Ts3=eYkDEPOtuFPJtg*!FNQzg zPr`0TGu-uZCfxN{Mgk{zgUYz|IJDzBt`*Lds+mACJ-hs{s-p~T5~)q>yk9x`r2BUzdlMh)FG z$t|HH@Z_c|Ji=6~L(xO<=I&aM(EZ0Jj66v%*LlFm;d1oj>nHpXgCRgXrgK3%=fHX; z6K;z2Fnr%L6u#{pOPhurh3A%Ju3wb4)us)erk(qIDw}+WCd(${6=@4JS-Yq zB8!10he6X#N9ct|b6=N!2bmN4<>Ov`wwhSnDZ1-3zRXaf7vNYqs47n7c8Cvy&yh*M zOl8pRJL4}7{XjGfT)013hV+fBz|&k3N^S0cf)ttUxNV^X#5;LHc4`Bhb=2Wf5-u`b z*`2U9` z{TQvU!fBcx0=qZDeWp{73uj8)P=$0}{CW(h>Xl92MIM25GjGE6jf1!YosY2FK}?1} z9n5WhkRu^X#vPAd2D#1C=88aJkQBa8pkklV|nXu+V9+@ZNu8LQgUn z`EG;R&RZdQju9udE}H0;j^q3`Ua%Z>vXig5xSW$X^M!_ZeSzOIgdTU~NbdbUJ#I`3 z;WU;cqF>NYs&?}$zrM{D7MRO}%5Yoqe9sMHm|PFkXgpNfJ_K5mTyA`N4EJ!%bRzw_ zkpF%+K~#nAT(_MJ7Z#(*dD~xs@Ku*V=0z`Ae$NiB-y!gFZxTohvxXDFtq_#uL6=)U z5jh&XB&~izj$8GUkkj4FiC1Pr>9&pBi>wIl$EKy+mAd;7czOULw6}6+O9pdA3Ac&U zjtkth3!}MlL;HXq(E=xh^X`OLEzz%K@(^gY1!6DN!iqH;>2pU(lDbKTQ;{9Pb-p#@ zwEahtl3zibmc}%0P{CpPL6<}*Z=V|jU~C1&T4^^PzlwuW`I}saJc$40F;6inZdIC&}?i$ zOVX?f6llPrdFq_O-v(Mz@SGpJ!H$%z`E2!DJ`VV%0YD2Gu$cM;66%G{!|r{=e2)=l z8*vrl_uYZ~nm&=o5EV#SzYVTlygnwGiAlEk{)kZ(1^Z7&pi24_wFy zf=wmv5al=umTSwsK zF=&AAD2+;e$ zn1kLG!uhRWx^x-z85N^sjhM>&&1R40r_mdM+T6X7f!x)nEzEr1Ry@@gTYmnp6#mG~ zK)u=!mZNtHJ~&dG`zC@-9Hzt~qKnwTixx~h5Xp@XZUbf##=Xy#6q}bHyaEWwas8vMr?r%d&6D2m{^I>oa4uFNV z={WJ3EtE~D1MkVv=(9?Z)s2`64?D(SpQja;_sYW|!z|XnN`*+({UFYbS?qpn2J758 z!8)p=fUV7|WBotuxcFNPQXIB1X^%4?m_At2iwRtgv5+?uoJQg2e_{-CW5-LM(ub4M zp*-X-N?wiw+363N&P-`;agG#`-TjQd8Xdr7x5l%1-?uZh)mvF=!FM1Bo7u>D|KWjx zWVH5GW(zZ7aI?v6HvQBSoOmIWO%T5p&Y5!7CtLqZTNE2YWNVf6_2)WWPe(oQ_Dy0?Bb$m7CGb&lRbO~yfy4$oo^xc zV0Ssby!0O#*s03&CraS1>#sy3Ps?z>`p@9Vx*Awrw+7^Xd+2UiYkXBc zjr|(!$ZoGGhwXup+?bF=&UdI0qZS)klCuc3$4;SMAJTEl^>x_gB#T0s2mQLF*_75% zVl|S&<_Ctrismr*)G!&eHhssN%F^h#X(WqzO*!ebTk!H!13tQ1hq|ln$$_~o#AA{v zggkr8*Q^-K>eJlB=l-QomBkX`s{VTJQ(h+bbzCn=KQ)y*bXtn-5%M*<50gdqAEvNW zwNAYAXc9&%NVD(5|3Ypu>GEh-SyfABRROVSjjT2nK&RMjkY(gpl*6cSNydq=sZwmx?~zmqo2O%%|8H&LzT4=MP5l$-iz3sJk7jI!SjbG)mL zc<1uHB7+b$miA7S+vpyJue_C5>a03k>2#J|j7{eLHa_HJhE1atr9H%AZyl>YVI@8~ zE1jW3CT$w=mt?i=furriahcHZ>sIq&^*c7hw$H_MiF*io{WB+r)OWFQMuN}7PZ8Rz z9>XgM3QbbAm@W9r)WaNM_=mf=%JL}Gt{1XMo3vT=H!(iQ@#DW}-=d)fqu|TMb~In+ z3A4YRfqjExMYdn`VC1gXOhP9FT|`I8o^2mdLAi=aLkjk)}?)%j&CdnI$Vi&;I=Wgs99ZxyEJ}UV~8MSmifX%X(bZ(}=O&^lT z4BUk-_i6`FJSHrMit_OIS}!(TV=9gmx*VZDt9XOLR~WBQL!Y?ZA-*x%Y)XCy9*tkm z=4uh_9B_vXjXzL#M;Go5nTt!dNy4mwm!N5|j`LL?1@~5t!Hqdvg#E7>H{s@ADiKkF z^J6~?>!dUoaq=uS6&xXPVH{hvqJiqh=)hcyKI(Bb9fR+1Fj40TuCQ@NQ@LKl#sDtsN6!|e*w&Uu`cnpXj#Ha5k4q9tsb>yBS8(6FhAY4;7)1C;Kdd5vE>LUT+ z`=Ka#6U5BCO(s_!N3Dc9e7!0Nv~1o(!OTq%aOMMNB>5S)o&3r)CA68I!4c-Gc#&BJ zEW@7ps|DM}et2Jy&l?$Tgs5{e%)z~n7-&z%?bG_vx=WtzvQ48ug8pL0)Q{wNrVPvQ zc7?Z2JdVA_At{)|3=#&YirXc!;<_VmANPPiJiA5Y8+`#2sSV6J+lKuYe5rkjGM?2Q zixRH&&s$T@CA)rNABd+vNP?(cPcdUzBy0-EID`yg_~)NmxG^H%8pRD5o2E&s2+) zw3Shpnsg}46*!wFJ~Z&(Q>%IMQcPj|MAq|f6qj)L3VAa10rl~IgHqyZvefGm$SizL zI;Uwfi&=7PsQM-JcZ;A2V^^{Y=d)O5a0hgiZe#TmW$yQz;b?N93QvA4=Br~=xu=6Z zbUQU7(K3PrGUkLf{=UMn>gd4_en~A3v7qdT~v{{tx zY53fk17Fwwq2BAfMbe&!(BeWcj^41GrP!q6630AT)f*`?@py+e3j^66%PV9~$`P0~ z(Vz62SaIgRw3$|r4n2IN3CFlbuv`wJZ(PRB^jOCQrsc&PL?(Qs_s$33x95yob z$1A}qvw_^wlmXg%f!u#~m?p}r5)JP@QGxvjG%);%`BN^!3OOqxPKw2Q;L7YSE-afV zv73M2nT(FHTY*=*2(t22WKq~i6%}q;^{Lg70H*`w^T~U3cS8di{X+?k?Tn$kzL8vk=c(=n81rQi{QjMVqw`##P3svPFFs#>zx5;uGF!>Mt%#-`A9ZjNy-kEX z7@eK8kRR>8pU%DA1SzXO;Hec&(Cegxb0XT{@!T1#(ozZ|n~|nRsG{lY8Tj(}704)^ z$GL@$N6q3tbpGl-a=1DGH8T*RWruS&`lL~}&XCBYx5(Np8V#9 z7M*F-!NnR@t|)@N<76@1Z3;Dcbc$Z+ssM$BbK#O;m_HPHf?HMQjVtO)aA1}ccjeMa z;x1K=@q+(z|E5H`f8tc4vidGx1+F5Gl|1s#fmJ+zY?-3Zf#nz#Zp{N@a_naW6vhq( z-AUH4C0CJ`Iw-}e=UlK_E%+`?bDHSr2g3i+f-ODi{C`mf*+Cp(#K-NkbVlk9?}s< z{c$Vj1tC|Wuy)>g?7g^Lq2mYkNiA@1o$KodH>uxC`@MJJ5=qnlL+eFb>Sz2Zqr}^p*T_da3CFp}U{c+Ng14 zf%7el|1%YLDjQ(osxc&fLJ42;MiNy8wunUcF|y3%k<}x|t?*;43N5)B%8#}Bg>eDS zyqbAWd1%HIGVXUWy03XbighNyE7QC1H$elIHWks%;kumq8w;Woxf37y^=)ckM;j{H4CN3^%1tkM~)C!>mK^5dCwvtuw#yS|0K8YDx^O+Mfsr!&YF z?m$C<8FuxXk|<%l4yi7brUVT<(`^52~L+~iA z|9*?EdS-{mv!c-}#apV@QlAMzPH1;J9!h%z-cm!$4{Z?` znNtQ~*5hPiXHpFF*BY=@M$(Wh{*PE2bimdpL(#3(ov05frTydV=mygYB6D>rz8|a& zSsHqn{6+yM|E|W(2LH$x77uOhBVqBsZNz`61&cIY$?Fb?;FgwfZ#uCLu337(w--Xz zb8;Fn{?A0TEy9TCxlX5llJ>)}_Atm#5Lj1RkCD5RW57qZ-)dyXz+H3i>B8KF~suH5i-oxmncqf_h>P^B^m zKaEj_&7(DVmla3o$`VhwT$=$uRK4KT^Hcm3e2Ldc7K)x=stAD0GX z!4*3P*i`o(F4apjQHU+MKXjC3)CO7dUQ$LBye&hdIzfi5&Q(E&oLW(G$WRvRAOSXo zQ?RGL2q%x$Acx%l!p0yQR@^iHYco?r zHO@692S2P%MhR=Jf6brZvR6re>mBn zCS(0YVJ;7?I_ZF(CHJ;%2nVaH_*jagRQFapkjAt%F=rp~+O83({J~Y1>sn zNSHlW6tIl5@R8#d4^`yK8Vp!epeJ2fFS64AnhBEMRawigK79Rh53@cei`7>yh#V7k zVb{ISRVq3SMFYpE0M&^*pb-IV6$3i8;DqnXh0#DqH>XULnW zPQmJ2D`A#zz)jSv#F-O(ab59Y(6!Hodi#Uiu3MV)`7{xGwq-QDS#E+M+it?5L%P_y zSc09uP>4f590kwCDQK%HxLZORq2se3^n@zHhxQdr=M<8IcOGK(uO1|l<2bnm3f##9 zOYyPS8Q2;%1C!TnU|&v5Wu1$^;>ua_5aFiDE%sT$t!w;CUya`fTP)4E^)Ygo(|QbO zu{$?^bt;Ip!>~oyn6uh?k6T$g1=cJJhnlTRxdZ1%lIKid7j3d*{E|Ps!ruUH&G%9= zZX@!w>uSMJTVS#%WrDw33mh^zhYC9fu^*Xkn9(vHyoa8JEX@YEzvm?cG|6LolN6Vv zDfEAH8sVdSD#R-+;yw<$%-QJd=IZBZv4_)qnMqNqz+5xqvTd&6>@TuhftdoiFn%~Y z;kgRNZt)O(XnnwE&a)@p*^@cfqIQ^oLcK(4&CVXDpYtl?t}3ClW)FBGz9 z=A!ZrQ==xgx;%R8gc^3*JN zGITQAq_~+gnp#0K-Yv6o{+WQM{}C+PdH@EEZy`#CAK{Qi9qA1UMw0_uIptYF+?V@i zWV5FXZXMyqW_rkgS>PPu^3mYlE*%94RbOC@SuPjTyq7y5?0d_e){C^6~O8Z>99ci z7hjqxOAHsQ;+7pYwD!gwy2Vu&&;His=18cJt&|hrCVS`tpZ)7u0dC(^^=TM zhjDe{WVm9af(C^MU9H8U#bG)EC)kHq?)^e_M9;`rqZZus}OJLg^EhMLt#)95|Nwls% zmd04Rz+($-GRkv+H`TVK_a4V$)$@;J%hznuc;Fn2%*i9inkJSXXLTgJ?GRCEXcNi3 zpF@nVZKv*!k68T_Yh&+IRR}ykmb@u^0wrHEl%EQTj~(IZHD7*L>R^&`+?OhT{Z>BdZ2&oT#*7%f z7(@5$k|QfFMe`#{Lg0gQ7QFDRBnkV2p({86)-U`;IG5|-SDQ;_#mbVOK6>!+sx{G< zln4LnL>Op#49B;mqT7@=aBcm6@H)DMj$bIk(b*UI9dQ}N)oGHj@3;yt=V+7b)`I8U zdkWFo<7ttjv>9yLeMKu>bVQX|THxQoL5y8``HI#V{Am&4Cc0|FiRaxSlW!5Ek1G;k z+9w*?rbQn1w9+B^LQrLACtSeirNa``(S2x z9m>uhDE}R_n~6*H$grgKFlv4V{d(;&pW@%bhws}DON9Mc!5(L_tum`rn1GOFj>kmf z`qm%}Qy@2XXAw865F&Z_5U76gqq(2$uugsq9EsS2slpk|LS;%#!U&5;!S(IGpJQc;deGM)Bh3Lo}Jnw$Nsy+j;V0U@)+#Pq2xv01d5 zE_>%o;LZuFv)#@_`HwI+lGi1Bo&2fsfChD5{gmJOw4Uk?|4U4U4yT7_y#f#Yt$g)) zfBs6QeW^yycC3LCvjt?l zg}{cLqDHjD=lGrXt-&{nh-Utgz#X0TfD@|7 zeJQFu$)?++)zLPriZuO6B;-;8oiR!d%Piw);m&r^`tL2&Wq25Qe_|voJ#mf<-Cj%9 zeAfYy%V)CQ?H(=YiJ;M@TJYLD7hFg6kxC6p($(+q@?9fH$Cx^@e4jMWfA9nEf2WD& zbz#>vEgIJqO2CXiPvKTd4t=-Z3FX-enw)rx=Dqf#V|VB=_C^YMEeBlxD-1(ZYrsa` zo4owm4yCz|MIOmY?7$^MTq?>#@1eca^^k{X_nd9u1DP1Xr?7Iz7nVZMwq)B_SKJyG zN6sHt5;&L-*~0zBMB>O66x}Z+9yT%f=b8~-Fr!%c`5>yDR)wk8-AG|^4fZI`V1_2M z*k~nb_A+G$(Fw9c)lHG;{A$h%;r{OtxT96bv7=kMt&kJ)(LOL%?{_r%1A2CPn_@MD279CES8@PE^p=PF;6 zI4ZtAyI06+n$wLZh$+WcY3$x~U@;_6BUjWiB`HiK7flD^6gu z>PEAu@r%jFK?|5s`B;=bSPzyUxq&~H?x;S`&qB( z%SlHt-~AgsJ6H!tCc1;fUJY2$b=%Uau%Gw$eSqHzPZMHihpWbpMys3glu5Kv-1vb` zq{+B$>Sa7Gp~3j9i|}48fxdVq;KbnVaW%=mRSxCo)6u=;( zD6Hu;!gUw3L9}u+uHO@j=BFRwqnFc}n(}qNaE=t~_SC_2i7r}hyh_OFo?(?~r%-X( zZtPJW$y^cy=i%~)sFU{vzt+pL_QU=d)-aptw8!AQG2uW{8?hjk#~?O~J$QE#cb!he zIos3)CT}H%_atqhHTG95lnGz6?{Q!(Xa11e(AGD{k0wZ$ukL<5<=k9 zsHx;(K{(lRhtSvN_1N`a2-aaa`Ql#y=)4irGiRaV{xNJ|tdL*!JcqYaP+)No=$TW& zZ@$xpnX3J(gw4Q|4|93=MR+C;s@_~uzFm@2lq9Q;Lu3yUe+RV&uing z6El(5=6E4Z59;m-7_#pojICRW`IE9Ca-18A+{*dc|0UAh2b*Y%k(jQZJr^&?CD6Rf zomMj*0`fb?0%?=P9c#nz@w}s`YPFD!Ju;hYkTYUUU+!Syg#-><#<2$7A9TTi^Y|`6 z62~pdClSLQVMTucxp++m51&kCtM*sXf|64>*moVPS69GS4=XU&={YxT&|l;$PO$L! zsa)u?JLH^i46aqaM!oIo$wrkqXmsN#q)p2N!&Rf0-K1^8>qNk`yO%^`^6y~T)B~*O zyah7_EiO|z1wSn(xZOsAv%PM~Tvr=0iE54ocZ9Q}@=?sF`yA`M^dG%+e*wE?KA&hN zc3Hj8T}wwlpNo~pjIC~^Z$O{v>Re6QI}BfPkbMiVM4Lg2Fed#NPAZm0V&lU44|fRn z$~bh1a%TStv*f5ag8GjQSy?vX^WeoO7TiYK4~tOj>c*XPS;6c4XT_#}HNdUc1!jP$ z1e;>6&qm+6i@Oi0vAg%GN$swGbbS0#<}f0L>p5VD<^N971F8E_&A1HxUIqBkA=s zL7L#NiIIVV_bPM;8nrjFum?G8?SaYcs*4nta_Hex)7%C92v!>;}G;@&(6D4I!U=a->T=5f3n{T8e0uA!vRU)0#2EBfK|otLRz z$o^UFX02s=+0mKWsOYnce2I?6PGJUk`hyD#{_D-YPOig)tIW`PsRk3ay0GBvV4AE| zf+wziMm2-+EarRx?)YAhM*@ell|jqcx_WEe|0o$(yH<#f#+@ybJK}@6tqqvG*Ob%d zCV}ktKzI~jCa{cu;H-QDrZd2kGOHi_Z_Cs4_mzccsd*9RoN~sO4Z~Q%RBsmhlhCGl zt*Bl90M{I=!>)J3Kr>sF`5ow^M!j|{cBLu1wnc+2GFE0&Y$w3m*ZNG(_YILZ*@QCR zCt~rrtt{Ihk>0XdgN|dKLWaPtIhp?n^K(qtDy?PAG-ED1Y;Ozu=3QheXVvN8jegkn z!;i|<|D=OfUBl->ZnFO6EGGBtB-b|1fgb$%oTx31z)M#VG7A|Wswc#(_lHQMV z*B@o)vWN3OC->p|Y8!mx+Kz)t21udKO>BQN16K*(rHfjEFEVd0_oeF$jlJ{`)s56y z>H#rUuD^peVJBctUAf3B=nXnc%x8Lvfy~O`0-Lwln3GmHf$fp%sGK#5UA=d`y!`J$ zdZ)1)r}PhFnqzO%B?;Q#>-3#ijfp}Vm15$@pMk;l!mjIGGrs5?%#5xgKK>Gpaf|n} z(IP$O7M~Biu>b!lba�me2(yS-93$33}(+!avPyLB7#G6%S8c?P zTds<}MA^{I9ao5+#YxD0+D@k%M1Z5hRcQ4Z&rLZsPf~VdH&&2az zz?q)qWZcL>?EPPL=$;cLy0W>8tnDuXQ?H5eU(?(wgJfUtik`Pl;B8qC3 z#-{2Q@SOp0t=qVwUl(B8PE~m4XGTAr{!HdYf1t5*<>5ut3z+7(A5_Y&;mBKW;Q4h0 zRDK!{6SEwlt;rC*bfw^0LTyhITJ@sdJNtsIM! z7dUaPOHH_+anZ2PQi;1;1W@+Jl8e#r=Bo|&Lgd8^GVzoP&MvJNb>9T;Rbl~Mwf-n+ zEd4}I>5as}kM-Gt8YQL}qQpk_%G16dUN}VMG_jNJhJ~nsuPY96DYp|rvSlV*j-P~? z|1!w4^TGJdWITvoY{!3tUFkey8@{9EG8h=&hT{`{Vr}J3{_H>%#BOM%_eM<=<(0oK zHCn4flon5hBxxl!dW1aP%=L<@Z`EO2-wDG1*aAt)OK`REGtfMF9A^A?6_kHVak<=Z z?r;4D-2LA&DleQ<=PqR;anuozd?Uhx!sl66@(C_XzAH*;>Y~qTZ7@4jj@ddLf}7>X zA@bxjOpwnfbxW;87E8vEDQ1q~y5|J%im{|nw^memb<{P7i)gaE)OM>mHBy2n!xMzWJCjTc<0sVIN28Es#C2vFRj}HR z>)T_6p7LEVUNFY0#&rkrteuX>bzji9RT?mOQ7Ml8b{6{YT&B!R2ZJ3Rpz5bBaAod& z^6`{0j4^WJo&;P zao34U@NH0kp-#^Vv(eo~+wjE4S};xgCfdAYkI;kLh+`i50lXRpKJ7Ji=SH^I$E$FUsw72MD2mwdOyO3Xi-jvIb2q$73C!^gV6kR3G(RvpeEQ`aBIxCisu zDYFccug^eqH37`ki)r^pYuLNDU2ve*5TDX6_~NhyYZ``u*Hx?Xg?k%m-o>SG`06H( zzcCVb``_ogy_<1JRZh8~?GU=5+J}s!X_yjmCK5 z&@PssIh<5Iv%-48c@0VD&@p5wCfGH=UXKh?`Yaut)oV!j)*>jo)eJK|$5K)MBrLyv z59as9;FH>HdUiq-x`aw|l2c2e>p&pn{qhH^Rk>i=Y69s4Mf~;h6S(-8Jt`lY&DGZm z8PpBGNlEHia!0ce_sxzc3c@{n6PE`!Q5WRT=%L@G{~)u`i|DOeM~9Sov;X)J?DzSx z>`#?4tD5@|KJaXVR{KBYEI!M_Y#p_bSYW;^d|m@Rb)|K4OCTQB0u}}0yvUo zhAu%%VO4E1ZoU{#bOw4c;a)R*?r((1!lR^Gyc9BLFT##j0)z5i_r9Q zIIB_F3hK8OG08wj{?zG8k}8+Ve;F4E{2X^uI_mVY! z4uMIFf8hFQ)iCp%84MdV6~i_!MCy8kDVlo2BT`B#=Rty_BqzQku zxd-aHEV^xS&|cN2G!`8!YI{cro}4RNSQnr4t4$Gil#@&nP$!J1$+ za~OG>uZ9MXrsB5)C^~Id2(LRLIhmd>#9j8JXnge;JTWc<_h_cmnza*P$MSr(K|KqV z7p;TQ36nXwD>j^5rZSi3+72IH+$SZC8Qfz{RcoIw_ST!Y-*{kgqNr!a6Kl?3NRG`OoZOjH$T}A6#1J}7JVM^TS({5m=t4>Ut^0Cg7sm01iuRyXNPA2BeeRRll$A(Ec8DljKa~ch$Ox56dr(59@t*s< zMGJ{gRx(1ecZ5{G=TCohxvuBx>OSW_-|y$cZp?fPYZdOY)toE`bF9dfl1nh{w*uL{ z@Gz%4>MlO~E<@&Cs3Gry4v-Ch5N_@j(rNY=;nWG9BW1IYNF5R*&gH+sJShl7XXV(` z9&yrfrG?9$coV+;KAW-D^>LhcToX;pN+xmoPw1q|LHg&>TXg;w4$VW4X(&md0*lL_7*+@9wu?)! z`%M~7-NGAtBgno?23ia9g^Nz~Li??^=*fFF&fL2KF}HWY*)=1a$mJPijqEgXQt1Tf zd`Q8jb>ZyiIi5G&+9p`NLz?YOEQ6FFU+lW6K-SK^4z3c1;fsY73=4H9QZ;=be$^fx zXGxQh5AFi->*69V@^6F{9>n#DKRMfpkfr^UdTu()U2pgV= z&|+?xq5#tW>k-b#$pmo?Z}KYr9z80?Amv9XH{Q4p(69(g<>Ipf5g3>U^Z6;9N1+G45Wv|~v>psFhQ&(fB}lS$mD znfh?VV=KIg7=jgpvxvkV7m&KDK@{zRsC`c!%t`-(=NnDi0 z@&_s&sN=|sW7(rUrL@$A;^MspnEB>Cbvda3D=c;JOmVc}?a^njd{`;e*Q#)>;|^!t|u!MDg7lsA<^D{vHU#^X}(qc*SZM_;Zg_C@+WQ zy{2UBgfBEa@;UT%%hLtnx9I$EISjBmz_8Al==Jg0Gv&`LA)uCbizP$qT0%17_Q2lX zJII)`E<~R1!1#qKaMo!S6ubqp#dXib@;{-16?lEd=T_H>_i$dp=61F8*iEJ-!$Ed~(EXwo| zPLDXjonAi?<{waI@oN{5VV5Uh+p<(H^7DPpfd9SkJTnHC#-|X)_t!90S%PfVdxkU3 zGMM8`1M+->CRuv^KXi^=DNxWDhPSr&1~ab%`0E_k>Mm+au@`T`ZU;Z`h;p0$@+)U2ao(no2)x^WoJgcw;YL9X9-f#K3G zx#+MWa9M2*wQ>qnY;_D3`#6CU%Xo~Jn(PF<`u(7{YqoIzY$G;l`fXg~_X+(DyHZWH z3t(%LO*2A}@2e%l`HM2#nVMPDDz8US6`(@02e)uGw*O#TT?5*``NBzY^Wb*H9j>@V zo!l@s#;EWDICo8%4GnusH{P9zrgA}G`e`!mS-2TD~geU^L*>^~m>+u(j!e%2T_SBwM0eTlgH%?F`}iXpZA&hrnB z{-Y`CKk53L1=PH6C%rvGk{sDrL+{oufu5mLX#B^NCgfIe3N;F>{>UC~QKU|_GV`Vr z^a>n@)Q}|v>ta1?PWInS$tmZ1k$oE9K}6i@jq`(R$p7owLH%i zgx@I=)J!6d&M$>%_Q;ug3G%r8tS&B<>83Gt6FI+q`rOY`e+6~{!^nq_Dw;N96Sd7% zNBQJcc>45f?)1`I-1}!vsP5Gu2rZN(+1XJr)S!Zf3F(kNagdvo6N9@S%{Ti-S$kwP<&1j&R@jop8=wnKY@L z6#QH%1!pXy!QN*F&bfRA)=W4I6E{5Onyh$s>!Vv(xgs0fntA8;)E=7QYzO>z*X7*~2#qQz-G+n4+g_Ia$Jl_m{bN0~aDN|m4mzfW>2oYhG9bzi2a z_8d<=FvXF9uCQQr6!>p6X6MVslGN-SEKzj|q%{AdCJ&wg;WLK$h09oi{0|fs$dZy` zX-E`J0pYc8IMd}4Gc_-y4_67G^Yv5i#YzFbKQV^{t~fyN#q4K~|LDP^mBsM1t&cjK zbiz62k?fQ%&kEmh3x(Gj1cmKBAa8L2mrOW_1_$eCj+_A7@8pwey%gv!{D&3W{?U8I z+Dw9X{``En5cQ9mbB{hu#&*jd=y<0>J_p_-YwV@qca}ME=(}7k~iimpq0hDI)m%x}IobSK}!yifrf^T(QPw74Y<7JrMTZ7S?g=ncAO zgFV|q^pMj&fx61ZY<8kFTXs*I{WejX6*%n>usymu&FjxYfatmOyrwUu=ImhA~Kil`k5;b>pI9T zT~lNarWN6gj8rZzqCud!VkCaKwhk`G#X-8^5*laskke8gE3`W!&dOtXckiyX3oF8F-mMz+m2mIck* z$W*=d<5-Oo>|^8yc&Idk=QOM(-Td2OPW~R=y%7V~d!Eo&xCAnP&xZ+>pM;)s%i&?c z9d4AG745!b4Z;6DaD~M!0*!MD?DBMbZsXQg@S8CnUHj^QAD_HC1v$sM`PiaVfo1)@ zkm9XPtV&ZcFi;u});WODW-+3A`W_d(!W!Q0%)=e33=`Ul(Ik;SNA}I-cK>(VA-F)3 z;~G|CQM4^oY)augu0Dow({=(4{>OJn64);BR=8q9*wrf^kw*$qy!eXwPw-}As&^At z;~F^c?gP~v@0;Pi!?f+{+|(!w>bYN*-7F#O^I2rHfzNwwI$E%{Asrx9zi?`AI%a<5SxA*n zxn6%4aNIcqEaYUt@}oIwoH;~vv_e6=c@vR+F3r6OT7%!Tf^lf`X+iY1f3WcZ2Z5)W zg;{g=a~je=Xz#B{pyPJ{sytfo!q5RQPTRo0b3`zs$KPpysvo4$yWCZk*BG2&3&VCA zja0=EXksqjWFv;ER@{p;Uv;-3XP6#|kyX-_p6Z2calU68^o26{v4M2bcEtbFcXxXemEm zo@8}Fc49O!ElDT6OEkeE&z8*LDg_-O=OKRF7|1_q#eJB;&kN;~!hcf1oZ80;WPff5 zvAueh4XqxICa>NL0&~~Vr>&{N)HVXSHKA}~K@?Z=w*UhqwW0JzozTJj1pO{=gM0Z7 z-d>&882V`hlOzZdzs61FmCA)dgY}8h~Q65NcO1_3qRvP5# zA!m50C6BTRhjGKi2rRm%#5;VpP|-XuSZWf6e_z%J)(r(fN7z1)3W|l&!{2Gm4i1$T z^I5S^=N;a@--#Q~Uj^sGl0e=>z?AAZuCQbQ_)igH_t|u+F!=!&x}_FU-{o;DR-XWU z;WaKtMg`Wng~8n546fpIH9d0p16Lk?lIj$BfEGI-{GAyBzxSAvb5}}{R|%0PuH(tL z4XH%m^d+Zs_lM)bDaBP~9$(PCcR2cY-GPxQhargbfg4YnAacD1OpiW@gHm5Xr!o`F zx6XuG!8|oTgp|qt$-_Uyg;$>Ipl8RIdFOx;dN>lc3wyX@dM6qXKxjx zJvu`FZJ7@no{s}__cyRA*o$ji)k!BjGlWegYrrH%9ln}3W6oc5w!q;t>K!@)%fko3 zW4$h1{eFQnoM;B`3dRVZfBKH5(j(a?l8VbV+mh2~)!0s-@w_H!z2C8#i$Q8S( z;jIt6(=W{ycGu(!_N0F0I!#rf;Ho}$+c*k3Zw@+^9-R*vuL>cp-xKmTX2NF|F_L^P zk$kb-Oh?6>hRMq(f$6kq@RzQogAq|&XX#6hug`fYbE9P2C)Z`(=j#x8QIk)7w% z$@lDcsVs8RZYd&r`R=0As5xxS$|i1GggNtu9mjrL4l1Po>L0ml0j^h?a zp}Lzrdurgt;-lw7n9O-9^Jj}-_05xyH!H^PlPY>NrfwKZ0a>GULm-sAYhP6Er?4zR0AzBuE~JeGae$7y_lpHs@0 zexiClj2xV~gp}JDv!Bl8pgcAUJlj)n$`3Ex8T**)S}=orih4y9a#u5Fi)EZe)gghG zZ8~~8RYCW~e^B#fJkK?_!fE<$BWGvaA!jUSbKKM^Y;Qz5lezsJ_PXsOeF^8l?%@dX z!zYM%&gDIq0oTww#8%jOc|EP$6$iqPcSxYTB~j^r34=2@r~SLMoCZ#WkU0TrWQ>Y9 zi(e7Rsqc#?YL2l)+2J7OdXGTqzKvw`sTfS@ji=+JKH~CMs$}Ggr|_ahyN;E%#|#gCHQ(7OQ2BQXSns!7s7- zq+@je3HB@&v?uO8F9Y_+K;S(s(HpFqeX?}?NH|)k}V`Z`5cVR7!Ku9!R(X7Vo>n< zC;0wp5*yik9m(zOaQKuSa~;!2)>hpFy`;5VwyO$>xZH^*Zauhg=??zvvWWaEE{6~I zjLGK(1=TA9g21v%M<~DTIINkch0=eQvMYDL0wyjXs;ax`qo=z`kmM`Q{`+-$uJk!K z)nSmH|JBPa-TZ<*Dfxz`-S<%i0@#h`OWB;l3|tweM?N+uqeE6CSx~O$l-@Mn>HBjj z_FZ!YJUo9G+g-b{rGdkl9g3Jb+LY|kP-O?%8F?o4N*q3p8A55_)vP$d41&@_$b!8J zq_X}zf5p)tBezX~u&qUc_1XYu{npc`yT4*)-F9Y@H_UNvr2;-@W1K?2o`&g5Vqla; zBJ3$Y2GcuJ1<5;P$j-o9?Vvn5GV!FcxQ=mah= zrU0CyJjrf-cQRmpiH^12gx9v_;)gOHylXXy?VZ(3BqN4_!}>LBbJko4seX$y0_KtM zuW4jmRx8dAiWj(iSVhXrd`V^CE_AuzLneYHEW04af-h&0In(w#Nt9SS`JYX}MRj zuoN7~&$}atg_9o3{H=wdr`^GLXDkf!^~4Wi`Xn@cCYucD@YAmv6QpB@ww&k6Q``Zbm2|M>2^PJ|h7}Kv;WtSe!QqP*&^+n@E&Alk z`Kc^sr%t%Q-LD7f+Vz2uva}S~jZRRj%Y-W1K-BP3W!pOc;9j0>=9zQ==YO4y-+E=) zlcHDL3#*@W$L=X??M@{os3WwC=b>GTYZR!?>Vk-MB4+;fA6_|<=eTK6IKQ(9z_5l# zT-&uPAoji&C${tNvo_{zhQl+#+X_b}(c{c5xtPr+%~ldN*+gLHk{`m30T(`#yATH) zrou}1E)=`e!TmQQgVH2Q&3E{rY)BM*D$vBQ29v4VG;( z!2|;im&!(x1LP36lsdw}+d(kb^ftC-2B2GD7PI;{8T?*_3*MUV0{gHSd^LU!xJ}!| z6!xhzy(zZL&Up{}n?_jo?hud~KN+Sd%>(D_5hxm$!Ftc|Z}w($XiBdg;;@ex@!u$% zRw}~ygwgasUNV2bS72)A3aMgdC1k|a9q3$f?&8WyjmO};eCVMo({kfPW{wPt<9`1TYgqa4NR zCe5U_)raweY%E;W6vKMI9blh6mTt|}BsV{pF^TJi%=@P}^FCG0ZJ)d!j^yxMhd?{H zE^9!hzPE>QJK{j}&5={{y2eGuz7QJB4aRA&8|YoWG;+=LILGV8u#ZfJBW-O6vGG`` zGl?~-uf>@D5J92qYPMYb9Q;}%#~x2hgf(Y-LF{)etkBej@xPp*=;~kYEC2R){+a^3 zk9KF8XGb%cdR_L#NgcMlIfn-AH@NFoVZyAgI2iSQ4s4M+f`Qs^Ij72EkUCZfqHs6p zdXvj^ujR7onnEo6-Upl|@772QU@;DI?A>ZLd~J1~D+o)#=3qO_4ECZ4VsGGc8x@xK z?_n)vvD}RpnSzjG`{^GCBi@5?1x=U#6C{|8W7YF~9h0oANr$E~2S<5+!kjUvaD#)x zn)8VN>`W**?nIm;jM4i2SoXks0LRU+pqVmVbb&z@Ec`hZVv0^4Er%kmt2bD-zivatPtA)8_oNyfc6h|x9? zKsN-i+IRXaLe>y}IDDr!lX3;MAxgOAwxUqNFOl13s=!ulHHMt|&nW91q;E4`(Ag4i zxj?>~n$?xc?W~hQ%S8*oE4q}~o_PahM~hhF<2!7w{AOnND}?BOdJQ*D3UE-+&W)*h z2OEN;F@2yIGLmi)A1)Gn?tbI%R)Ls$#eo0!chI$+iMK3{)3eq4vAVFIKC*m)OQ)-n z7uxeVGtUuhv01oql=ee9Jp2@G?XpoWmLIkGNJ?0WvZ<@LXOSM58uyW&Rtv z@k9EglJsH6xKzQh5fzvz6-K%}q+wTd6>dMRK)f9LF>;F?{K!sW0R_d_YA8+Ct=r1p zwW#B-s`X&wYC%rK-=}wyO-RJv$0+Sw1gk#9Kv|v|y}s9to&M#*MmE&K+r}g!fATRM zuPn*H{HrkhODp$J%NB}O_Q9U!L`WOs#m-zQ5*`?~9~&4p%i;6zDtOvzMCKmMhw%k%pcVRw zOFg|xczjbH35nxko-x z+%ulRn*L@zvuW<6vLDZ}rN-S@C#OIHP0LUbIpUW29~Zz@vaf9eu-d_Z$Q+mloolba z=5G?@_nC7vV+F&nDg%htPApCNuCUo|4Y+vgP=guQIO}Ws@Hn4Q&*GWmL@5e4nVPcY zW54qpYYq0xZwLadBq4apSeEA84@dV;fzwsd#O=aE?mvE?Ffwc-o}TQ$b~)&>J71S^ zmPeA=R_`)2+^hqxXFZ5-!FN>pl1}&Y`JEHLdug5GQ<~|%9(?{7k))bDxMZCO+xcCW zy|##(-M)aBTIJAZPSc23inQQkMKQ)0Sfay^Yf#j)6c27PqUZGtp;qh|lx<4oI^WfS z`MdXmRT>|m-q?Z_|F;o(8hnM1<^dm_%xt?`4Lt_~=e<_m8-yI@ga7fRL5!Wykg^l4RQYekV% zd_WWD_dSE@((&*REb;HM2uIf^y39B+cJ!8d2EYw z4#}~io`a~HDvkO+cAT#IFKV)Q2xcqFlj*H1$+@MQ$;zdftbl2-Z`(bHsb?Acl5Gij zKkKn%Y8soNu!v>6PG^I*0ZdO(LiD0;8GE)Oml@=35lKJt5FxVT3fzSY619rK_X`ne*=fTd5+sR z&LhXxtC6Vnx~y*RVix~iU*zV+|BjZMivD(=M2$s>?EdvrY~xln(WNwhQR8`2jNnw* zjUh`W=a$M$=G!pUos&d!$pBWSwS(`#Ij-@C9Mh<2X3K`;MH_?4NoH>k8+PFXyQZcs zQcvm@D$8q&n#&?E-_n@%9js-e9&P~Hm-%e}uK~73_z^JjIR^SzvFb7xB9KaAi-J_q zOzj_^)^KO5Uz&;RZ+^y$<0vz!;b*#qkeO_jMXAn%+>vb|%-GU{hPoRuuP`OH$WKEQ zeQ&%7|9FW$dwzsq>r{3<)tx1+t|Y(1Y;dJIKZjoN-A+LyJL8Qg^K2c7zrKgn)~bls z)%&ojp@S^DaT`bb$BH63+CTYKD>v(iWt$@@8%+X>0j*P z@!_JR7pGX$u_ktFv#aRt2?;#w;LkSPm!j&Y1Q?PRCAu2^1r|NiXXmU(kzKD;VS(gs zcCvaU4#*S`v8i$J@%A#<#B}hW;|(@mHkrBaGiE1kdB&deR%Xn0uw%ihJfD0VU7v~Q zdSDfMJ>?(JhjZB?r&?z3Je&m>c(OXz5umr-irHGPm)&hv{mq8Vx)hEjW}2dJ)_l%sts3iA zP2x@p@56&)SN3&A1{Z88!R+Q2GxORsSbW5tdAlrU-{s%orPnLisc&U$d%`iiAUw_% zcx{Gj%Y#Yp#nWs^`w*LZSsbzwexd9q-UYKT6_zOH;5((iXdN||WxO9DdOD|&Jzu3F zI@OfIhNgRvtgL9VO7|g>(dpDtrV!BMD7%}n4OgGv%{lLVhCvpip<+cSyCzowM*Hh| zzxf^f8#M>lNVZ``kU8t{DZ&#{^6W`u7>=|(O{(^eWF;>+c5Us^_CcD_I@mYTIz)w>9M8ml#kPuPb_$JB1m>i-`^ln}IJ3l2}On4?6Cxis-m% z6W%JFDf%qfFB%~wE9yzR$Tq}{V7jkX5b>A%*-%oJO;MC*nE~oB)p``0UF*YkjLcw? zFC|1*UH3Bs7gr{pt0PK}Q4=xsQKEW7Dds-&H_xToCp1VrjQiyGurB2U*5bsNW`G-$ z)(>P)w$(AOhkw}po+kGAcQyAn3QM=oCt)%+ zn7aER^IG4Af=^SRgDzxK>nZxJlNIgW@E(H?&lH_h3KM?16TpgZAAqlu1#EJC7W;Wf znSCtIgo}a4*b6l^@}H52?G|ii8l^4l^S>lkC&N1z!qr7eDw3kF(~o0DXd7Fm;0P8s z&sW<_7SR{pYuFJnBepa|L9|;vnUzabV3b=ldp24V7D`tty?OC0;KCB&c4R%G?fs(Qss(XLK}Ot2XYl;?DCX1}#YXOkVPUr}GJVk|cIfs^{BP!EChwTTS}UDFC+Y&6R+>zd zRu5wx!{4&0Hp5vl#^Hy|=O}RYVN1d<@caH^_I(79Ymy4g`@0TnIvUP=&&_A5EoSW1 zj|?`A@5gkGZsFAKe!=;P7n$PX6U-t^LDU(cAcXc~j{W*s?7Q$QE*?IEt=wKfJ9rMb z__IjDMqk1Odu7-?c@0tD!jtUMnGLvW#~$>EH)IM6EZMF0sqA^`N+!ydWXlBEEU%%L zlT`^9jFmEEf<`?w^?!|D_xGSyraEVDBPLp_KFH?qe)9GHkMQ-eqs)A}t!U~7Sk4T72AD7{!1NU+6*Qw;-g(tiIQrfnx~EAJ-(Nb;cU_0SiTI2p_2os3vtl{L#3jA3=V!dOAr zA{MiG2b%vKqW)Pw@Q%QS{VA|y*OMffAayp&x#Y;khi_#$TW*7y+j#cUc^9tv(1@et z6huw-H`#&j`3#m1$36dcp!l~|9Is8-gR~X6!s0Z~xoT&>k8eV`8H3Eo*%MY6Yq9<^ zzC+wwg%QdMth4DEdY+ieTmw5D)+^^oiTCNF5@y*SIlM_}8(jr8X4ktnk)NO0=V zBteIB3?6f7phx#uaSQSyIlGs<7go3rrZ%49ud6bk`$h-SLRa(dt_gyKfCs!gZL9FF z!2w90ThAGU#KWYfr-Jm`k>p7&(km+rI4|+j+-(29+&YzZdTGWUNSD3Ao%J(@J}tg; zW$?mrgJ~u=cjp**DAy{e?~9?HA6>|Y#pl6ivI~s4tpdx}71JbJF%VW+P?g_F_&p&} z*fDs8zVmLg|K@Cn(^B=J?y>@zxbruS4$9)@Wo;1pW;#IA4{`pzER8D*x+lzhyOIt# z6$w(e59jO-v^khGJm=^XL%5MPg8O#v8z|2x1)-FrVEE2VoH$v26*8 zN&Mq}4aR`NhUIj^;{;(tc^!ShvkgMFDB$P2#@x*D2AtwDeYk7s3Q?YMu%;{?hg}E< z1MOqn8MQnpnZ1S6IATgiNQx2V57)6I;y5R|KR|uEKhVogmRz974+7qgqDzO%fUjW^ z>^j?whYftFz*P3{zYl-kQHx4I`QN%1;cUm-+=Nzv-2G8G= zaKBfN;l78Pai%HD1v`DD1s#V|VQkq(;hs;eoTK|b`$>yrV6;{;{qt@*x1cH497`3 zJTk8o`oB*RYN{>ejHSfj;0Ak)s$UEXZFivUgF}L9S0m~sCJi6cmhi67;ewk($GFv^ zes1@DEqeBn71e0u&uQ>Gr_bY;foG~iJt;DDLZd^`#CavAK8s_!yqmL3D zs~0v((ltNw1XK8aT~q1;JZ(0}{q&zkXQuRXEoM^aEFS=ILzT48#Fo?Au8YO0%Wzm+ z1eF=R6&z)F|7LbAXru}_d+rYR{ZbnDSbIM=M{22KKqkOgH8H5=IKgHIYp&Ytadpa? zX#8s-54UCi(70Daf-apTK~RV`D659R=w(GzEi{@6#Z%z^=43&ctsI(Mlg9u}Ia2>< z1ICYw5)QRr5l&WPpfFaNynRt9jH}6l7Nrt!JL||PD4W19g}c?)_Z<=Vy;8?X8uGB# z;5+~Kph2J$e3**6w+mhl8$)U;nuI5|b_nA2p9)UqHwfzIsF3YH<)Brf6n{_E2miIY zY)|AU79aEsV`m>k|Ducd?qMpot2l;TDBi|C#H6tY#1JMtG{@Sa7{TRr&rmtE3HsZB z99GIlml=9sx-^_wjLb*7`+=A|;~_ZALI2^PD2|>SKaR;79KeFb;>@G|GivuX zVgHRyY|Zz@DAL@_CJ!9pbbtBc^nc1MOw9NAQt46ZT{EoB3N{3xo8|#>GvkDnA6l#^( zvGA{aU-j8X^!Sv7;%?DcBXDLzTYIQv&^|7uc^v-z#WM&uRk7|7X-qQd87^&o=+Lp} zDev@*<#dNH!yBveF%;&ry#xBpz4s#;=IP+$xySi)KozFO4uE<5SD|obDgHAQAucPx z?DK}$C4P-w+fj`xW?5kU-Z^YU{}fan%RBdWIih;#3p_Gc9-B>CS-p-eGyMZ>Esm_kM`g3vh>`^6al#E(or-0d9=V_r(TrMy z_u%)AGOXKTi5=^nR3{!v=dL_dM2m|l@KJaOlN%dhJe01>dW5)YB4sX7|Lhv=QC&R*jMlZ zUVdAJ!Ee6cl`s)MCuiZP3vz7juy8!|#1|X;CbQ_N4s5}=kF?=R0?w7uUbyKEv&qy>&K5-*sUganc-n#9 zDXoTpggqpS_aU?dzr>gkPRuq~F2I#& zw#2}bOJmu-Clk?IXBy=0Gy-$JJEeAE7}BIML{YDjeyCXlkNsD};<#ksRC@%QUoIh< zBkjp#!3msvtDN&RypKxC^GIKA7iY-1fy+uQEKv3o81@>##izOS+AUR3Der>OmD5OW zl{q;yxD_^y9?jy`&Syn6YoNRCK9mZ^LDG^OLDS+KxR9fWwq@7oEB{_l$bZk-508Nr ziTg<2gZXUCG6}N2-G)8s9|`wH?; zKk4pdaEUxo-`>a_Dw@deX8MGp*AMBhJSmWBxd7@VUL>(*J&R4XCH|!b*u2q}XUSwi z)facD>57C+KO;%C^hqeF2!$%|98y&&UM+Kk=Md~N;I62jCRVdXpg%PR+qY+_H1c~! zr(8CBhaFiyMTI;rFe3-tC0Krh7r8T9o%|B#pBGD1xqt@=ATKFR`->vDPVH^P==CzP z`%wxAbDmU>c=C~ZzdHr$O71}323evnpNu+71f=!lAedP5Jg(o#!rlAbXjIinlC}3- z^?Kf0xk*~g^lE3w<3r;Ab$QTr!cDnRD1PE-?kFE=5sQ*g!B2#d*Wot*fjcQx)nSvxeBx7r{GCBbyAw^ zftp5(p}s$sJS^=O*ja~>z!}b1I(QVqt^B#kvwCAQ!dlbU12!?iv`~+nFxz|(xKly z2Lz2(pcKhSgN!ajKaYbEMd@&1AQHxX&?mkv>-h}fMYwz;oeK;-MD+YixoxVi=;+t> zY-P$bdMa}f4P5z}o{;$n`|6BIoca~P=(C^b>^J-9_NkGW^z;ZwY%wE~UsZCK!)Bo3 z`v9)(2cOxu84pHj^KqrwRGb^kpTx@Rpt5QV-gue=84Fhst87W9!&Co)k@X-w{9!jS zd3R4(?Vv|ibVp;8ge*pHnnGjVxHG}Za{QUToP3>HNWNUX3q^m1KHNt$%5 zoF;oy(iolzXJd3oMpOYjmYW8FXLHF;_mNIXJ`A?OF}{-&3`4OC;nTbpD7v6Wo*cUa z)5U6_{jn9PKAcU&H>i+5S+%75xfXFwG$Gbzyo?ec@%8yZhejkh8rb}uBJWmzB-P)qhiMZiw2H}JRb4Vn5XjTjcTa~(Iu zoKkp~&4s^Gq@!jD8K0<29Ho|VT1TXv%7@B{fahvAy-0(ILVL)Qx(7*O>QvY59P#o` zCd+k?le>@ZkRt2DWR6ZbnSQH<>{@jJ^8RFV))n@I4&;$3uM5c2b(th!R}jHn?_kFX zOQ+}o-ZPWxNA7f0Lg$-lu&^(eTQ|K0{GSGbtleW+eD@J|evJv)C3Ap0=uPDN7DzJw zwnF|feJ3NWG;-z!-xoZm=Ct18DG?=R6L+I`gsTZA89QnO%9GcT3flxy(9}Q%f?UYK z>L~bcpBeGujdU*A(oQKA+eqbVZKvepe?isxHN4>O_F`T8iRG9d;J0#w)5%rO=!AbS zh@0dUV*X+&nEjL{U+rJQR&fjAxWjFvFl!!p>9m_jc=OzJoeWYo!htk)0ugpkA}@=U zkmDmXNrQYNDS5jee55bIe78j~=Fm%U?lU9Xe)BUS@-6h#&Lit4y@2+k*<|aPJhDY- z=;S2h0~3NS!u@JP5|cNE+}*U4%vqlgTN8&7Q=u$WzkLJ}qd$O;dKA%f2ql|arW3jD z%cT6>U*g$%jd*CMlD7F8PVNt0lXmeh@0@`$lp1dUKK6Qu`q%H%qLleja zdpDBoK8p;^$t6iA#mJ82hr}sS)9KqT5m8jJBU}A$!AS4(WQ9*U+_|JjZpG{&m)CZ~ zkJ*cXZ@QAp_YM)05+(9w#CLe=@)|^P{Ci{3C9-!(6A5bUA>6MHa_7!M_^NaPl;it3 zl@xu_wWJ<86wSdlG7M(dErG|M#0k^?0CN2Wu;gk!d07(!oaAN_VfT^rtdE7(L|-EQ zu9S2>KfodN2@FUhvSCuhY?X*8h{qFi4;@nX zG>=pVMv;2Ax!|jP4l+t-!k8gX?Us|EH*-l~T`Gz9 zTMfU{o`IvGFI+6yhArDZkPwBl#No>b^7Fkk`t0lm1@G-7V7&=xdDH3SIE0hHygoR+%bjHKp67$-NYMO&{b7S$+%Px?wjlt@sFu7Ed_43p{go?;&zdW+!=~IEQR|P($WzS0a~m zLxCIDK`!kKg|Bn3!Pl+)&pIw9JB|EF;QD4z`E{6xr>c`D{I`6ev7eMLy+^DTiimVe z6SUpA&p8y{fVz7-h;`Lgl5c(-lAa3rF7GUIetso;FTDyLD87XI4)Jj5(m0O!wQ!a9 zxM6wI+!gY;G*d#KB6}|cRQlvrVuQ`GzmdKK?4~7#jc7KJ) z$>!w61y^$DZZNg7;auZOJ4a5f!U@C zaQELxekkSg-{dv^&=4mx{cj1maT{Tc{v=|xTa8G%|A3+V`(Q-p!~3aKxVemPX}iti zd&hU+`=$^|OLEBVm&WA9<08^LR)H+fup)JXsc<|07n-OgL5Q^>6x#Bxx=RUUyhR?F z`oe=S;|g+V>Msx+9#2wUdC*+VGGaAY55>$4CsuI=5BS}(r6bdn zr$o8b8#`~ykWs}5w;aUC-_HcE^SRoc&v~BjRXgHaw^fiBtqv+nbjY-ymw3+(5TmVG zpt(ntc%3?r?zRf#&E6zX=5r)DSJKJN=RoQ=ttYid<(!h;5=d9<8$Rzog0z}WfQP#% zch+GJsmym|c}8j$z#4@%Uq|`2RsYm6>kDaN!zp|YC-BL`}Qd@|2S94k(_6fpE{RA-zQsi%| zn$vA2qW7(0$s(y_xG?1qnO&SloUOMJuVZ(K$UcB`Zx7%#)t96rhJ|7)S@%eG)w@N9IIY{Jc}`)OOT z6J`Y&u#_Fe)T7G=x&4ZCfqo^9GnmA-ZaT%z&JTp_qHlt_jmtpsh9bUO+Kl|D#`k`M zxNu%9Jb#nKL_BzfCw#Y#Dyr_ytwNcJK@%Q{m6N0W>~w z8qLRYnEH1D9ty98nX1+_Ld%i6WfMmIlUrds?+TK>ZNlc;cX3LJD$FB&8;sFbLQl0P z+(-8qp&i=@{hF6KSDjH@NV6loe|s~Ar@!Id>UOkrQVP0vKEnSwJMVZdzp#&MY8Mhk zMs~``{G219WQ3OXAY_yxr8M*-WbYBOixv$P?sFtjM$y*P(%w^=dhX}%=fCIsdU?J4 zasP48bFOop>w3SF|1~4+fuVUqKmM(vRWnt!1-!ZRvU%j~Ft(Q57S{dfC#?DT!Pl5%tmvYQoW?>ve-c7lR|#p`P1(A)6)bIKB`oqdGdo*D z1K5>6>?C#P}BTHyL0@gVNcef7Q%5(uVlzePtOBD%h9K(@d|QA&y?p74Cl? zS5xs=g@4_(pEobQ%r_flF{#%n#6IvB8_-F}cZ)IALxv8-=o&NDx=Ds@+i%I992>Mf$ARgdcnGkD2VW=aJ{7N&cX@e1zEhhOe2Zvfjo_ z%I5KHpYID?N1I)r-e7JDLw5RQGy8b17PU$O8>evv(d**aAoIBHe4`4E9R7l9=;Y?2|j}+=ilFV#(RvENK*tE=>_-$=AmlKcImAfQd*C)gWGw;ma_*2pfe$k#)X`#ac_VeeJ#^N>5F=T)KfhRP9G@ zEHdLK1BbGC!W3fmpMfyL>U&M}jI+#P=mSB?GK`J2dsb6lG8Gd`Zn2Q+2kg0OoiIV@ z&t_auAa5kcuw|+yY?#6=q~8i4xi5~H2Onx?5)U7;jcWg~bGoWb>DEDHCKj>0nd7+q z)rG9=gd}?&)yUot3S$4x-W8q&Pa!LV#Q5=}PW9>vJLYdC$?i^gB@}8ViS_St%r21C zEckB`v9?)XLL%BhLU_ct39mEtlS%z zU-&fP-0wz~XDut3>~`R~B15+5_&u)qvzEnad5JX&gSf@Zb=;ufDf`5<*(A9M?3$cC zTPkDC(sjqNTLUKYPLCO^I!l=)1$PPW^jEQ6-=1+9882>ODc0vN3lo&e`my`Vhw)!` zwy-;?&jhVJLw<48c{ayYtaF{eh#M`t$;N!R!=wad<{o!~DLy*Do@;$%hUe?~TXk6; zH*pTTS)0asr$n*rax0;-xGzh3IgroR%oY71(;$EHym`P(2XP*oFHSnk1^e~y#9XTZ zubXSl?KTdrt@`E6P1o$=Z|$Wyx7^BVo*xz}zbs_V>gqi7uLH08Qpr5KO?dYDlWb~A zGPBGX!U{``g-364`LCo@rkE=7)Vs>qZL@}|_u~F|%x*2RTScDj`Ev))gZdFPH#7T@ zv8>1ND%a3h&XygX#qSK3;PAXEJdd5re9oE)GQL+>|D_RJ%^-=*?IX*aN4#cL&3BkY za~O9XdXo*&*uts49M9VShfVO5W(P+O;58ys?ET<;e$MwD_v-A+4|rc?BQoc(C)18F zquCKcPl_Mga72-9`x`}Wzm;XS#p*<@s);uQ&F4cGF6F5$YRu)9k?>ij0X+4Opl9`i zwWY+Gy$j6dWB%#$jW`m+_by4L)B^lWA?vs@bD;rzY%Fkl?gzvk=&uAv`~LN;n_mz})3_vJ}g&EabZ! z-}Y`h8+`T>TQ|Lfv03|T-j4n*yo=Oe_sxQYtvUOckpD+@R<#i|}@5ho>GB7Dw!2H~fz>$)b~NS6W}Tmu?{Y z&%8&a({Gmg^egiaeL6-S?ZPnIJmK}T7MQ#GC@)^ndvUy!L~0uV>aloz3A`@XU~!*3Q7Nk3q2$)l2?zzQk#Jba}x#1!Q={HMjZ)4BicAkPw_JB|ys%+)7M;T8 zBUMcH_!G85wpPdqtzt_eKcJ=V2Mf{aVx=QHYnsM13$s%obW@S)Ru}l$~cYhOFYhtVH+S3ni9tE`o29 zZb#^*2`qb`FVVLtXO$ZjS#HmIX0RkuC{?#(?@m4yrtS!%ZAE%?^l=G`n2`g>+qnM3 z)lG`Z-?AisrERJHL=$=~X)X1UpF!<<))I>zOKLZ5utjfn7Ta+eOWGA0>zK_7TN)S(>JP7zczberA3W z-SW?xW+xlb(2{7ns%$qYR?jE-Ug31GWE1WFT|>V#OsBaqzscz*`)SD$H7YS>J>48y zK?fD))8d;qQ0QkxXaDC3$pfqD>AS_`eAiH#W#mQMd#4bmRVp-e^iFa>MT>qGt8u;* z)X@wX(Ix1TRFhRRi#EL$$Qs$c7I9{pt~=*Y9Xv=zO|sh&S8QYD~=rpC@st8Z5*`%!`? ziril9Om|vuq%xa!(th`Bh@H!OI_S+hdb#%ys(18qt+)uXrSC|pJ#Goj|71ZI8}y;B zXQtDwVVU&l=t`DS*hSMKCed86^QEt1e`*^fNk>OJ)3l%F^uB}~8**p`J)(1v53_OM zN1yyh?WD?xZ$Djez2iqsWY$)RYB)M{V>7Y!>?A`2?vU^o%2cJTg-VQXqnn))Y1pIp z!l(NJ3$fFqX$R|x-pe?mth$T-$?0nmvK(aVCQZ6U{R{Ed@np}2)RANT|01?{B;61& zmfn~*mTI+|QV-Ssq@~fC);sT}e>)@S_sr2m@nZ&E@K>9LhG$caHzTN;TQg}Axg5Iv zRj79NclyqwTy%0CrlfT*l{|4$Fvy+AhuoK^1IGH&oy8G!lkGA}5`F2#zByE)=Qo*^ zwSZ*LiXrI|W60}{XQXq13H7;GNHrezAvXqXq9gky(SwooRP#X!&7Nc`)KwQ!+Yhtp zBD0Bf;zR`-|h2-0<4iZ@!O}b~A!f8hnsTq?={k3xGpT!A6 zSLjIE-O`ty*+XfyXCOVTHJfhUdyjZFC(tp*Gijh+6(s|=(1~Foq{HK*umpwFzVS1g zbAgcdf(TM==)lvmPLOv8YRRhOrNsJ%5q+mTq{c-!Nk6pY2diMw0z@#?BJ3p zn$`CVIq;&8&U(Cut_V3#T7FadCb5a^Xp-R7scV;Fvh>c^5 zL`S;hu-Uv%+cBWNB|$J~2@LCv3Vqj?z2PY@qGH+V^H@%&qC5=ze|4 zkM(7w;9k6M-xxaqy2Sy$fNhs{M_m$_Pc)&T{yg& zU+s8}MTh&)g_-L7?5cno`5j$^3@)QG=3c}!youe@mE)d;62xwM8=GG+oH(<+yx5Vm zuA{cBpw)v&%-%}krkoNvk;Sa}?P&N4j@-~;KdqmzojKo{&Art&bNhh)Br&X${7jN1 zq0-k;{c{x;A;4tv$hU09&r0Fw+3&(Z6)V~{4SYNSpKbW>E}|mCSy2l z4wI)c+TG;(o0sI3=LN3hcAD)R+{F$I@5`SkG_aK#3H0KMlkC{S3*?RVQzDg-hn<%P z@UQh_gfj1MwE6uN;;-447{tz{az{ML<_G0;!4j$3Jx}Djue;|t(!pS z`HjYs5+l0WcqVi54P&|u-x!h&SmlWow2#ze+PAe*nA4rgn}UM*jO&%`$hqetLs86U zSRdl`4^yk+H!1Os@uf9$R%-LM4FUYZ%SYso(@Z`rv>d-CfzKa!j5w^l#ZTWJM5`1l zu=USyDipn_d-Le#>u&@;OPX$R9>_u!a%spgd!h541<|in zBoF^96^;y3VH)qdSb>H><(xS?6Zejp1^aPd>sdT%-v<8aZ3zDr6+;$#`15|{Be{ii zqVQXF0QVbskHqerMhA}4Bb$pQXs_fxac=#=-i*=YGY+q(KI_$}zj_g|T#_y_h;8_1 zxgvIV*&3GqB7%K-YC*eOv-uXN(D^ba*xHgJT75H*zR5GDTX!CyKiAw6G=A8S884Oj zUA6DXI4orvH-7^v{yMSoW4^HmCabxtaXeAkDCVp}*U|-hBiPg#tu+t+?Gf^t3fPkV zS!{oDHTPfhiM>9S&G#*nXA`Z$$s@mm%&=LO4jp=!Oq*HEey-Q%{rp(X-0jZv)V5?c zaG@NrZq(#wA{P-sGlP)O(KLkmkkR)KvgO|=@$(bE5toyL`P|KO$%I>?6G$$UTiOn$ z>fe&NORSl&O}~?wMh~D>`G@Ff?=U92EQekWi7|JQu z6`^6Zq;lH;9v#1(pNJVj@A#ePb62Erwab@@^@j|$sL+RQND}vqF88NeW>I{A>jt)> z;VXMH{v6%?d4A0anRMnY|C(g0_)_OJS!BWc=_F$RR<>=^1j_0cnccpe$gUmnr`@NL z=^NV=YSA~7_G#Ckl|Rc_-%+>ueE&G4l$N5)a1TlNcb^1=nUGxr;>;pH^rbqwjYPqC z9vgYMnXg!t$fD#g(*A~{$l^bLSaf$g8Ca2CF>a#XXzwhiJUy!+T$x=jYmtX#CL!Y>UYiqWmzDbew2lDo6LT zM=e`O{wxWSu)&o0Da__leUIVvm4htE)Q+4jDqyd&p71M!UXaBfO4z53W_)I{10A+< z3?1(Igq&FJL>6^sa9iu|>}2juvQ^recyd{?;QHp8BbNuz+Wr^FvJXkr{9_Fk#7OXe zUuDSGrY>@E`g`*4$rG|7JwQ0UWFL7JmBUhZwTg_9U&Q!tBUgx@NwTX|_&x8^nw6W5 z3jG2VSx{*VDfYWUGD2(Nz9a^|2g-1)e;64MwvwOjx`D85Aw(uDfDbyqi=QP5f@|In z4ATvuwWSH2rw9l0;CWZp^BvB%g~i?_S3Hq3Ek7qYIH4-;!Z7G251Re1^e`@=zUwDS>ho|{ar9WA2oM=YVMhgYzFZcglv z+^?G8^+xoD+I``TlQYfL_7!w(l!(v#5hSAetT4;InLbEJrMmQ@$mcEN??X@WGcV@S z(jx(U<(_rWgNd*e@Bm8K&v*svEXu94`} zM$-H_lHQ+hOoa)x_^;E9Ug#gi`dvQ-OR;w1?Y%!_$n}jvnC&`RFm^f7+PR4~cHU&i zJk|(K3w{&tXBtewem&bF_NUxw(B;!gkMYdDEyQ)r1+u#DP7>Rvj9u9rN@RLplh}%O z_R?}5-IRJroL8&J&rznVIO7s`m@BX-6*Hn%K7<~w{36y|ZYRIZHK=Ec94T8koCLHy zBr*&0=#1|pXg`g~%x;1>_h?B{>N}d=7}$@l?e%2y`>y6!_Z%nVR&AzJK6MC+XD<=k zA%ltA@)LC4Zd*E2#+zGQDkeJLs?ADlrtpBUF~mGq!hBj~4XGG1g-$TpLnX7kuGDo97mJxiT;9y zY`$Y<4MHZ&WGP$5lD78^thdaISa%#JPtGK=6qO*FbYLLSd3Tcb$sb9hCI~g^&Ob#S z`DB(;wvXN&c!EEj3%=OlHIML(6b9vla|P#9TtZ*pu!Z=4BCu^pvbU zzc2TfnEBtwXn$q0(qRhwx@H%9@lnj(cQY~}Oph0Ajbed@ElgN?fv6UGQ}!a26iwVk zwnk52)#L4{|0oyU6Oc$XoOaU3OLX{)5w7g>sntaHlsg}F@e`FE(M$UG$zZWt%S0}? zJbTyoC7kf9LSbKOe?y=0#whO8x`32T>SHlzjXvF+5=DF5 z+W4tiBl*O_VCJ1Sj*3+Re78j{(UJK~gvRSMCEA(~`EM*)qGiV?yUEeU`c6`EU5$@l zJ6AZi=$KevwU}Hj?t&as7RM!YGqY2|Fgl=-(yE9j|EL>9g4Bgd*0~V8A8mA->XT3HR{Nr4pfu zZ18@2(x+0E&21yRZPhonaz&3IbK96KNlfJO^;VSE#j#`kGX;zJ1DSh<8hbkPK7YhD ziCcAhb?+=L*44T4aIIEy$1;m<&q(BB>%K#OgdzPJFtGNlS1hSMT*@>(#M+$UDMWWh ziSS8h0xj`Lr$H>Ayc&3you3aHtRck@EF4V}-xrg>6dkIZ-^OZZ?POj&f$eJxWPMVd z_{ig*NweK-zUqNBuj^3bJO0>N*l7QvzZFLCN80Xu<*o@9J*la5Rz)}ONjV4QT`oe& z4HYs{JBZJDP5ITvbmpqz&&D3_Lq~q9;x7013p=a6lAo7fGKKagVYA(DHqB3;z8XE$ z!oDPi-YCB$d~09G1J|$NCAM=}tubg{*CPJXdlPH@6UEnQICA0QFE;6gF>ei>OWU*T zY3=a$bY9@&8n+2jl+6FfeAOn<4(S10d7h%hIQc~3v$_Ibn5QnTgOrFt%@f-D`Z(_H zj1ra(i{;I3#&q}Y`6OWFTrm##OKN(qkUB%r2Ua|icPzchx2-5=b>}LCVf}2$!=DMl z^Xdq0IA=G%=@`a;S=mjS+BYBU z>5-3(ENSN})*ul_O*S`HH%ToM{q=U-Uq~T-gQoGO#0hkve@M;voQ-&yH;tQa9Zy%# zDE_AY93r2s==;Ve=+J_I)$IVp+k@ zB=6=;p{8^XnrV~bbJ8Ajij)kkLC^C+7Al(U)K*&T0i>UJ<)i1+JLnEe6&cAh6&!fo zLy>JgS?ojGBXUBr^-qsDBkp}ZuVjG}jm5o^N92t5HFjgZ5?|`x!;fhW=aT&;Y9k&G zsSO#**u?@bUaQU6w3GjF)2pKQvRjwTb+G0mt#9*fJG{wD#b^AxK|0&uC3;bv8<>n1 z;pPkG)wF2E@&w%?J~w+6U+8+1N$2IW-lMn4DW$7?=<^9=ph69wJW7I-9h3QQ-ABUW zz(jgK`4l-m>@ii4SjtEK9?6fW#n;@F)Hgr~}OCZtq$5tpSydGu+42P&q}68UJJxHyO3dAXD}UOY(hBF7S!gO=>K(I=)i zJD!hT(64smyc$6>cQko#dXC?`bA|1c{Kbx1Q?p=;-NJw>F=sG2j(e{^%j7PH@$)(5 z43ZYS(XfHHhJ3*liw|V+_rFZBNrT&dso}F%^sBuzGKyc&dddHYEa-p1k=*HG0w4Hj zJU={iNNvuRc=qL8Cs#P~hWm(h{2Oz^`0=_4+(g@uZ_eC9QjIV0gXa>(yw1qlCDWht zu;Y}^*f$Lma%*Tvb1=VLzlo<0R;LMelevz@7@oVLi7$S-lBYFq!0vxlG&r%QO^HgVqn^O@Df$FQ_W4L)I*5wX8f%A$Jr2t7e_%o7WgYOVWr^W=P)+8>Hh zd_nkimMJ%YZTN2z*ElzW>k1y+_f{gGH+K)YwB3$5%f;~VAq{N!t3_Nb@+Noes^xj{ z^U1Hhv0`uURvt31i0?mW#H6$(Y2drDH9Zm9#AB%zw>&A!_6^?6-2+w0r0r_lXp$~< zY1>aHpVb z|3G`$(KgEc)RZ`v^^>oMXg7xy#S@phbrrSY!Uq5S;bEVjzah?|V?;2YkJz&X5?Y?h z;LH;X1YI(_Oaw`GFh4m_=K-SHf$ey|JAgl+k3yRmGf2jv}@@OGvBbenC2& zQ_G!VUxJ!HznCY_$EHl+6HWi}haM_C(XW-M%@1JH4XtR##ArTd>1FctZ#JJapn_*r zohI=E1K5yXgKJ%Ve(@8(B>BBH@jTPmfyO=(*XxmQc;c#xnmdKF*$4MSR>g-aMQr_FV--afhlf^J_wl!h_MwA}f+weMn}PmKU?_drq)Dx=Z+%uf}A^ zy*U^gd6twFNK=!9V7$_AVt3csk%EvkSZ+Q_e!DIrhl~4HFFy5%tjHTloAPVOmg0Y8 zq4i$!)cyeJOjn~jBYlO#Y5ln5moMf{0mmU*zn{GraS=L|NkrOoEUmW~K(3DoVx}Jp zO}SwVt?O|lcS3^Mu1TjvSJ_x0yy1he(ny^xP&!Bw8w-g?%m>0E1rj*ci@YpvBU)Cq zOlnXr8>q0Jrp`Z8^I*{&=D6b?IVk2K9u%3ex5oz3&@dTt`~t1u-LW<5mGq~gieolf-X zy>B>Y9!g9+GFi(~Bb=PB&a}tabH^o>7#nhid=;OOrN?u!%=aXi38+elrjl=A@fhpZ5=IibhU88neEAabkIhh);p6R_F&l=Awb8#|KmMvy#p6cW=8$oL%E|MPxdF1f6c)I9U1G6^LWCKQv z+znY(GGk|<$a;4nPi_n(Tg5)F&$W#t_f;!QoR*W@(?dw`&k7{JKZ#eTquI176H<{P z#jgFQN)q1OBYLu7Op1A8=T#*1{Vtq-vXYSXX{dX9hJ4qVN!lmrF>8fN!Oz{0sXMz6k9A{- z7wt!Wi~jD@LO79{u$xI%rm#gd4}^@f_2kxQE3)gUGn0ynBCfsb$gC(ij6brBv}xi%8ofDE=2l5P4-~I^qL^Uv*f`t zS>oOWGUbgFbCM5ZQ93(8>)g!(KNkvE6m9zh!&IFm_o2{v?aC(&q>C(>{F z(`~g~q)&D%>N0GIar-RheYb^3x%|Wziy1UIaUp+o>a8$Y$%MxzekGMlmP>B@bHuXi)4j>jar`R{V-v2Y|;SlL6~jmlvxz=O)2K1I6PRSDm- zogB-nA~6q(5pJ=VWIe1U2DhY1_51#0&Me76m1`fnN;LG5_}{-dv-Q`j`G;2#Ub2jQ`szV0$+?q5iHpd+-7AUex8>xwy(du`=0!r~ zJV*lZC8KWnkS*G)iO0ruB=YcDvS^6toj&4E=3HG*NY8q5EY^p_R(X+{qSa(_+8WZ| zV=dAB>QDabw~lP{*g)JO{fW`(jf515??HN@}lIwEx}h}@jBfgF4nOy*wMRz==aMp{hFTN-Imoy<2ij@&Sa;4A1}7-C{k6j9 z-_sG;XC~gAw1wPI8$91ekszdOR>h#dwUb^yIU6C>VR;6;HQ zqJ5`g-3~igci14j$quSxr-PidgPe~eBBr>Y%G(Z?Q>P+uw;hWA+CuZUBkoVP#Izg0 z#;R!;G}#i9k6A-b$sQfkZ18501)~01qxYr_G&`J8*g6XfY8`NxIv}r)H9{xZ;jMuS zvJTtgV!v4^v9?C6c!E4q1Xnd*>oUbfFpMd#L88jEw z?-s)D;yj!m=8AJ$U10xZKCCXeBGlL(_XfGb-*qW2WG%$TH*+wrS9BL+HlBFS#3ZRX zi23M*zUFq2m$pT7$6OesJL7xT3@nayf!ofxc(rpjdQVJ2DCYCAVV5>5`s+94XPJ-D2HG~vS!2DQsob^)0iwIqGj@QAg0xcwl z>f_f>J&YQx0|Qwj_)gZxIVC*=e9*>%XidDV)W?{GBop-z0P)F=QfM73pDO;3SN%)y0A+Jv{I*gWGU@-1am=!Dn;KX|%+LUxsM> zYKg})Euk@f8uI?L!1$9>q1tVRMZ*Y|w3=Yk!l@W}%oyX!EfF`@5g$5dVTFP%4hPww z(9#C4uTMwFUQ02_GaVz&TVl3_1^O7wgz0M+%usR0vp;rtg88^*?Epi=S+KY|Q@oxT zI46D|vAxrg{2xX1Dl=RRazobhW!R{=9CJ_m;f1jW&LyqE-c(;mO!Ps!lQ(W^xWRYx z3RL)bA>p_$`W$vgZM`=tr~2W*&rK-1vJ-lp#ke$00Y3ly?@Xa4r9lg=> z-3MQCJaJMv0BUL*pyL&UTN%Mv+`bXro*OYedjlqo^@UsaDnw6R2EUOD;D1f5_3JYS zXT|Sm#*39uzr7ZL{Q|Im>UzwsUWWD&EAUCz6E!mHG4|bjfd9authMQDX&A`t_|4HFAzsH*W(6Vjh#z_Q68}Y z-kyGlZuY^iWG`H{SpjR))i4sj)ee_s@N8X*gKw5YB5?%<)Oq3D)wR&w;DccS>mc*Y zA3+kq7`i5*!2GY_$haGh=K;GgU3VLBE*Kx2)?x0>4QQUb0qrLP z@u+V)_Abdq(tlY{ev}Wl1(_JIPW)f37_Ud=Vrk|cNG}v+=F79tJ2MvvVVU5KS=b+) zi7%J-;`E3td>5G&-Ivm#a#3V*4Nk++m^4%`PR34+RJ^;LfxMU`C@ZBRw>TCh<6<%W zQyhvV5)n2y0p{n^a5p^#t{v$x=t;-nDWbdbdoF6~_e1yMe)Q>6gn`q`@$Jze7;P%Y z2%{tT(eDs0wv=OjX))v$9>Lws6;OGyAKUA4z)lw;XWm{kZqCD#y}2m+E}rY2bV!*b z!zL;TnW{)t1&GRT$ zeJ;n7J;yP~{|M?fl*3x;AZpY1Au(ki9!eg>fN_VB`gT9A2kwJ&)jp{FD8(z$D|&fs zA+8yeiha!acrtt+R<13;4CQ?|uX7N|(@L-`sTBVE_FI{~BX*ntVinAwc!NUXC@?tB9JLos zP^M1M6lRFm?q)FFI0e`K8e!yNL-@x`f<&tx{sx+3`qQcCZx7t;V}r;o)=1B>#<>+X zC@nU_nB}I}H`oYg>MdaUWfHcWpMr5SbTL`a6m2sskhjkS{xW8G^3WU$D$LR4X^z#p z4$#?eiyQ?Dw0ca1Nre?m7S4cDkRyU~?J(+&BZ?KQ(e=p|Dy}w2QL%?BoryQkAigIA zWTZL%-kXXEQ8swA)CQi-Ru1b_jEw~bPMn4ZU#FqZmFdviYmH39sUW3O z@pGIt;+w4DqG$&n*kQAX!{41?jmQ6HVo;(LVt+bdVew3Cd@vVM^BiC$j^8#PZE<$@ z98}zJ!6#=MOtQ4cwuIm zPM8yKg%m9ZNUgVquKH~_UcL$+)q9wxb{m{4gaIp@%#Es(XaDBd@t@{ z@*f z*ai$aUXKl%TQUA_Gc;STLNWU?=BSGvv#s|Ko_h;6?VXU@*&%+rx8YFQ0h@!@&~4cP zm-V+WYV375638cR1LX5cQAnn&S+PdL)+Bk*WU6aue^^Us-0$QZH>?g+v9`C({t_D99W9aycm z3lBd<;rO*s{2UO9PyM%HfqOVio`mDTsVKxdZAH`IU1%{0hGXP*@k%?O0;F0|sp&xb<)jWm_T?W=FwRbvFh! zMq~Bg7>tOGz~}c7=xB^WYDg3y&SACFVqkSM4t+GEFzMJXbh$@iqx5bJnHdY3kc{AU zF*sl!jW)R`_&bRAD=`8g1ETPDb~G+T#v$!=EH0R2z;$p2+M3faq+WD5pGtt)xIFyL z&O$^-8mdG_Ib+$l@n#R^uF61dNEW^fPeq1Q4t5zA!>wj79JBLrCn86T!Lxy;G)&P* z!kbqK=z5usDDOS+3dzCQ=sY+)%s`K87W7~4h4O_Q4-#wB@LyII-a77q)BbE^AId@U>QeN*l#LpLLUew}h2w%kgh?0R z&e9wVdX@>ZXDRscCjnIzqJL#wI_`;l@VC3f`_L~PC*S9we64tIp6$UhvjRN3U4$>U zvf<=W08f#pCdll=-V=N9BqI~fBD<`T@4>@oSs3J=iL1}E;1!*PTkG>tIb%QW%{~Da zw^AH@T!z-TeOOzPhq%W@I2%}iLxT@t%(o&;+gb)kk8)TPmE-#G14w?Jk4ygL&`2-D znnn4Tai|#4n+{;^;Di$1 z{87Q&-*PC*7>C93av0HH5ph$M@KH$-XL1#xdVDp5B6J*14= zZDU|=KOS38slczV3?v82!>2_CXJ?Lv+ICq)%^m}nOJlM1h&=3#l*JfV1wJdpb)%IM zps9>?WE^}hDc?H-9`u_EkdY2eX)Mcj{9L4=JGQadL?Wz9qk$yOD|miOWq z--8#MKB4}{S6oy3io0H~5#{ztjM+aRQ|<{C*}TG^;~!w#{T5@jpTRBsJr>M+iQ*5> zFgo}Rc5He9w~?Qqzvm6CUVp)ovA&ps z@jefs-}o5Mjy}TEiZ0aoyn}q|3(T^3i^#X{P*eRDN;#jQtMm?re?DTm;zz*eJG`qt z!b$cWW?lJ+eJ|fZt?zpn-}->FPd{Rc`WFnG`5CKGck7KVIRx#!Hy{isw=C z1y3EnVSUnXOda(bP78Vw7S#i_8J|Qa%vVIu_<|qOUohA9x5#?zfq9=FSfcS2<;LA8 zy#5Podd2e>pG$FXH>&P@g^}rRH1vGMftpBrv4oxT)OZ>@*5JS_2ASu3G(pFFD$&*i}ADjki5p}WMKSM($j24zD+eF z)@?@QYsX~57V43&at5S$#U%3OfN-x`TrP6N`}*N~L@O(Zf! znqqyFCOPv{i=;l9NDKx}B4?+Wkf?>G)HY}VH&vZ}hoX0i?$F+z{DH0u+m zxGAJP+LX9>8awur`^iH;D)phQwi>35m0wLgsp!6KiX8axmJ6*rgK^ZfZ^%sx8R( zKGvkqK?@QUVn(*!HX_OCCgi82Dd|u*A+^26pY3<44Oi= zG#QY8l~ah9wE=nj(2PjmH6`ErnUne3%*dd1g!sl2GWVkydEVE8TpkT#wcL>Od#Xn~ z8w^O%M_rPgU`TG(nTXC3Q*tuWl(;W7ClMu6iNB)_xhyl2jDI_obe)(+Z0)b0Dd7qR zq@2aTl-p z#g`_~9Tx$cb5LA$232R8VPMt(#Wse=H|tO_hvA_~Jz6|#P`cnO!otr%@T$eF<&EfH z)CecbCMYOhN6455O|C}dyp4*Cb##gaGsSQDo zuR?n3Wf)Gnfd%KTK=)@0w5B!TN5*A5@{NY980)T0+l8Z6VfZjO0$*(-k+wQk%=d@m zeC}#vp%sG?c>QK|145)EkRQJ@Jq@5|0@s@p!vF8iONZkggDe#cI3JcS96j zhwX;HR1~)S7l|nQ2-tUqSu0!Jx zE*}dLC&sR7k$CxOH>%b}V`^|LqGJ<~8x(^)jTqctAA>RFQJ8ot4%XdKFkKaeD!ph} zisw5fG8Uc>lCZNq0Uvea;o3hDZO;<%>thUtwxr?e$W+YjOu*^oiKyNZhp!uw@Wmw& zU(ypWRy7&l3=?t6ECC^|NqDCkk1KQI@i<7#4Qa(g-y{i7rBjhspMqq$R5X~TV!UP? zKEH^Avq=j6$fcm;cnY#dWg_KzCLTFtW8=kicrQx@CmDEsFAc8O(y%i&6>n0~P%twM z17Bo7K|2#OuO}m1B?Ui!B;(zYI3#K&;_<)~)Vxo_w(l7TXivuVr71YHB?B^-;xYeZ z2Fk57pfe;JA0DN`@k1&)%~Eh|Q-+uV&BE=fWTgDc6l1(3EQ?G+VL=jz=w7flNJ5ED zBF4Wt2F;`cn37c{#<}~@x3d7re+n__M*;i_@^SWL4(^8MK(1gvUbPh?E~yBcmlt6` zR{;*(+6#s4#du#s=v?d<~r~;VSm*Gyo5(G>w!vJ{W?%7NF} zQrNLV3_r0K55E<_)>+)e`BIFw;8M)kv=5=Hitu|#A-23ci1O>X=<(kVE1xpxh85$~ z;k~#e=H7O0EB+sf&O08f{|)2qJ*6RJ@4cVv&JHaUDm&R_gpyP$*|PVZ*-{7@nMrAW zQ)%zLNLotzeSUxRs&F3XoaZ^`-1mLGuj>Q;{pFBtJ%MF^i!oSHh3DgixSCvn^Xn?n z@q^X)#-(_ucoN+vmB{MO#|-^4+$zmLuZpAq8=}Ns&VRNH5}g8p!fJ0gr=Ru=B$&Lt8fZ$>g&*F+K6?>Ps929DNGTn zf%IN$5?Ho_S~u>bx2$SSR`;g~uq>K9cXAQkO%+VLoo?<(9U)F+HQ#}}SGkb%J}WZ3 zXGM34H(&lIDq3*Y$_F5}i8EHeqcg<DJd`6NTMxU zDcaJO-uBwlb!R6EPHtw*+zr=O?oNnFE{?C03fc5XYeh-cIjd;Nv&TdC7wE9tdc z(3c~YbVJ07%5Rxd>lI^q%56rvGmWX=awAnv7}4U1%_RQKlI@2jd4zAL=0C3F7S0%n z4kz0CpB=F^ZRn-%X10E&8Lb-MNMFC0knr#ZN_oGTyi~W4LZA%=e6pvO06R*lwxQ7g z3#R{WC8tBi^kj)CS*%}ASzPNWp`BOhX>IY(t(hTiCjoTWNvK4tmdY;7_?O zG_Ys~`LdkO6Tx=mDYb)|Dw)Qu?Ml$KWg0fy7gJkGQDV8`zZ}Tl(28KMu_i~XrmWPyenaF*e27R4m>|CD?i>0aHQcFY7%1k)?O@!yHB*-5~0C#N++RwB5 z-I`^7*bCKBd0$&eO`#RB~(OnS-g(}zQm>JyD?Ix!fU#AX<$!yzab2Kn$H>}E5R z$Le7yOFoL?tw+&x=MYR=qwtD7140i(V&RJz_;KQ4*U5U-uSr;v?0CjStm&OeIXvqF$25Qe0Hc#RI{y z7_d%2?wcf7dMAPJXEI*?ibH-u3id=NqQfo~N7?;cQaBlWqw$!|<{kylQ}OOaDrTHZ z!KI(6*umB#^M9Fzz=cJO8A^xb-YmFpPQ_X7WcZ)Wf}3Uv41Xj+KP&~=Co*6K)P1pd))t5R<8hds8Vl+5;jlM}!Yp<6-gdBhp(qr^-61G? zb{rxNF_8C4#7CiI_*f*dZIbbeeO^kWQ*dKt8fK;($C0RL>=Q_YoOV3g{w6~~oz0=V zQjluS-t$$lxZRY9@Pll-&IBasq+!>bG}IR;c|F;v3vXU zA~>yK_w~yKh@M@5NcK!H_*01G)g{<8xfmfz6_{#Xh6%L-%xf+~@Sk#I8Wv&Ah6)Vz zoj?|w>vZ#EBR{+l)R+y&iY$zWWMRQ0HnV$?hWolnxVta`Z{}oR3j6%uypV`a{$$jh zOv34rB#hilK#O1+3d*vuHZuny5jnWrQGhu!3Q=;S1T*+cu|A^^sV?k!GP3~x`DSCy zQTCj?nTmas4m4!IW@!fUM~kslp#ryi%3x7M}p-WUu#Nk&y~DjxT<_f0wpvuxOF38td5DiuNMDR@7ThFjN@ z;O>-$M{g7Hj?MpW$Y;Y~MG~Hm}EfaKAS084;OB-inA&9HS}3bECUVe_#A=7y zY|Oou4%I2C7%0ksa9t7>#>K)pKN*JK(h$=TkNEl|henlA6$iS;v znUIt!!zZQ#-n&!`f&2vj29@Aa>`5=jx@GJ$ub%@kxg3~GEx_gT zdH7Vs&X-3XzE|bKQmetR8BHc3Kk_-a3m6&S|6toPx}g z3S{3ng;u^wNb%NycgjiRB%XwGZX;G!w;*=*1^hJaKyZ5#%l2sn*Zf)(^qj#Vkw%=f zX$IF|8!qZ~A-$jpW#zS)iaG@L)?!J;85qB=!I}3gQ%|K9;_g+ruTsg@YcECBa1{d9 zp1^}Al~~YMj(`=VFuTguTKKO4wzp3q=x;rEFP+7y!g@AKtVTHNG4Ds$UU7rs4mUrNSK^NW(w`0-cOR!bFjOT|g;xW@)`uYY?|M&_<3wtnAy8}L~m#(mH z#o^*6G?kylgxon+3$|hN z?j0CBXbTNPcl_gZMS-;kHp%Wn)H6rSS8&I6T`xSg^~7lYUbtUz$DJV;G_1CViMTta zW?CcjpB-E4z!{TQdO~TyopC3gSbE+Ib0qgbkFg)868sU%j&bn4A40YEL-yPr9NMr4 zu9rRWA>1F+ih?0p83e8A-q3XmLh6M;?8$S-rjY>ng?r+go-fu~9YUvO2zFM404G9m zInf)}BAg8v zr+d-YN*CJL=tOxTc62DxmO_#oiT>;&Bmdpx7rl=(7%wVoy_-I%`_lcXA#`_t5SeBl zA^#Ug$j&&BQr!0tXR;5Se(Fch;7Yp~ANj|12Pyb$CGWc?bSrox89iG^B7PPma=?PJ z>$XwPDO>tF-Id(49qG?@7aH8{PG5KUvc4jamW}Qw)3yWDDG*HeY7f!c_#ler3#KPJ z2goq-Ac_AuK=QK=(5L*p^v>Uxc4T?d4DWsPwB3Wm0_>@2a0iu~u%mA)9VyMhf&Sa# zNHe3Ysm0Ttmb%$c=+$ksXSp@S4qKDIv?Hx8b0$T5Pf}IdOC3XgY)y_mq~z~GbtNve ztl5g{*>;lqY^hk@gwhgC$p4`!{h7O&3jS`QEt@Uq-h4~)er7>x54MrBjV(zW+D?4J zZnS;JF5>#-P5W?_ts^H$=_)*4nvoe*bxh}DW>FRGEHUWCKnPayC}6~3ibAxe}T?`S0g zi;B>gUyYj5YWVBbVVy@EIyas~2LBoOE;$7;(=!O_IEh!h)yU~6!O{2S=-g2Sd7(;# z44#0rRv{WYi;yA6^gUKDifqlsw#r=GyPkum3r@gKuN*5(s@T1{5vfwA(f+UjE^`|2 zJ+T>ztS*$|oJ9JON<8MTg6`~6@S2}O|C2^opE`xAf(Bd)JcA*|6wS}7#VGqZ?qwC& z%#LgM;u3UgGQD?UA_O*u<4w(RI7!Cg71NF{e~Cp=a|}MQdDE1@ctj2+L5wRI7k|e< z%RB-u@oiA1|qEOew?^X zsW|~dOJgw1G-lZf7F?;t`j|Z_n9F*sJQ+5t%Zq^RJ;n#{M8cEJ8s&2$F@8S^4vjHr z%!@(u`8aGXjf8M(B=UogBXCVDLs-F$Q1ibMRIm9cL92VKbh<))33Y z&46@>WHQ&}q6|1ivAS&uX4bqMx%hGCy>ICiQ;K;=g)!W80hEIu9l#}W~`CIQWe!wRohxWz`}rC%I; z)tTO&kcjB4OgulE0d7`1uhGuI))A)77qVKH^*mRO6eA%oAM5Y1&+Nr=6xElY?-r~7 zs0dSwi!sb*YQwiu;mnf^nSTk$`kseV-Wizcn}sQi*Bc#HLENL&=sE@*c2~xEK^4X` z0Pp*?06sm~d+IYCdMzsQ4Y1zX0OEJ`Fs!Euu}lt5Sg*mGCr0=cs}H#^I&fXQ2G87e zAW^7}d4p=G9|xYKtKt4qCHOGTtRspssoxko_1y?9c?LLtd=o4lZiKVEF%B))N8@D! z{8AyL*sn&^FD=MDS%dHEH6f*~iGq_rzrPMN<251vSRd2Z>tg3RE!+*)!b2`yj2men z)1L6gLlYTa6tQ@XIs`>kp|wUH*I4VlX^J)~8`hvrQXgEoM!4g!4rvZr7_BhI-)AN` zJZ}>=2yTMIBgUS~u4DPIMvUECgFi;=U^vYPPfi#hcl#Qw4(7o20pJ&`fme2FSgorH zI|T!{RTx6-f+1@5Z$wVbS`-$TFuuqb#@jX`TG$j<-Sm)kgYolw*WggDHe>yCF(q9C zsm7|PtXz$@aSb@X*2UEcU6gGgtafAE*epXf_h-!Sr42ZhXo5vYH(>T9eVD)1!3V}~ zJ?7PgimDzokL$2rM-NNvOu+lg5V_79apxvu#UEP2Az~}$=~>}w(iWycTi}_tIYt^y zF%i589;dg#i1pR7bF(2fDIJ&3WT5m)21XbUamz0oi!Y}meJz_Wmt;YITP{+pQjzYF ziyIEP=#MNydO{|qu^B^OK|YR|r{ic(3Wohs;4_*AgUgxt@g^T@&9iX*b~ zB&;1u!Tjr)NMOD7pAYFUoRNe#-%{{qZ2^QX=0V>!1@@YZMcoh!zE3f*{TmCxoD`_m zG6wWUDk4v$!F5GCS~GIczbXxVH_=rB%)iEj$TY!gu+orRtf#Zh zL;3G~Y{|*PXn!7bTnf?UP=cur`FP6u(PMVSSXG&Wrtxf~u$go6tX#bMoXI>5c~G5` zg~|IFQ`^La~avcADk77#U5GIm_A-MNFc$g`Q2nP z$@z!T1(V6Pbpl%QOsC&Ej$w;$=;8f>>9_x&rfv*YZjcE zv*?`uOlnPKnS6SaDQEEn;%fh3`=?)c;L4Bo75u0X7X&9z7^E%7_XgO~0=fy+_ zV|-pMLos7sK7AHIkq!?wsjb2_LlK;j;6nl9e7OAiFxOWAGksQ}{VN}m+1IKR#gUdP z4OL57q%u#2*K}zpycENRB@(#7BY}N+qKu&v1+TFr78}T8YnKFyn2&|$v<6JscAl;( z*peWRxn5G>ejo;}Q3))YEQK&`d6e=fpwdSUPvRvpw@MNsl7i405Wv9)t6;xX5Km(T zF^fwWf*On=VmysUhA?FQNI>(sI4-t{AyZupk4j`Aq$GsQ#eyilFMyA`R$=yk!g$0Z ziE7B*5ot@%1G((IrjS^^46~cgsFt{{?ab=GHjNQdh5H5&k;yj2S;6vJVe)h}| z!#){dtP>Z($z=jK)3Fi-?6tlIi($btc~t*Xg-5P3oGNAEaYh_XjtZ#SAqEvu1x$*T zfy)m;ID8SqygU)8EAzp~UlLA6%E&$`3(I~{jGBm{YNiZCKT5&WRs=iV@<7v^A3a`* zq|l*AF^$r+NmGFy^vF>Y;~jpPs?hp5YUG`uLMIIw)3HUJc>bu;S2rgAo>U_ZZbeFc zEk`QC3MAekMLa?ZG}lm(RE_1xeMFk>^Gnh4sj}o@D@pBlq^bX}D0SqEF-=9Diq|Mn zwx}|dS*cR3su~IY)*vP`QNNx#>HO26YbKz%%?h;O|KEG1M)O?N$lXze1O${ROcRg$`YNYnCG zdHP}~N%HODBqlFIP6cvwOH_`u;+5&MuN=jN$dDwL8rg+u5F6Xj(}P-c;nZq69;QPf zW;&GmONU&nR@3w`Rhl%aM0$+Ld8Z>zreSOy$qq4ERw71&FT^NjfjEg@7NaXJ(lju^ z-a}1A%F9ru6VA%iH(8ZV@TyZql`^F=X6JOeEIs(DM1jStY3cR-C^7a$1Dj8bsCwbN z-%gnQ^Fd^aD?|ewAjP~h5BBWB^67h#{~zObzk6f&h&Q;Gc%V40!!eLQJ*_Tvsb$iK8UiW=i9^v?Si%lf+dk zF?@X^h18Rh*m#Qh2^WZ?V}T^@c?u)SP#S+Ws$h{L^X4s(VeGaz9^X(xXQmwP+>=7x zR4M%2uZT-h(%5)S6sB7w(Ro@57t&;4@JzxV+YL`T1 zpA>%O$>Hr@X$bkrp?<9h9^VpU%#IYSg5{t*LmIx-!q{&og8Np&;Mpk&wdDeg4HRQ~ zohWLOgz>~k67thzaLiK>T`AJYc2K}~dpRU)$imE94$|_fY@HfK+`6iWD|eOP(yR#G zO^Rq>9+z}(4rYw1LAXl^v&@z8M@kDBteI6|)wrVS??q^idULd{aQM zjwS}i3G?#C4wo1uyQ?-emLUlx2~a!9pQ!>T%Yr0c2U^c4-<-NS)Lh!%{i^dZUo zGge!bVaR^wd7UOA-s@strV38IQ(^Cg3WOuHF`e~$`jgc_^HspxE04F=MM!$3FzIHA z(~)suTJ%Slx`l=49G?&wYVZ^PqZM?on~(Nd3X%9%QF8ktOW*W_Nz{^;THEAgxV{(NpLDRHJsxjhu4c}j@}}Q&{#y9&E}KT+gVg} zegP%f&ZG0fvq@rf7DZ>xrCXmC(7d_JnfH1nTl1Uc2^{06UI{_+DC4JG5fMt(7p4~q zVzlkF2nA{ilZ>nY&Hf}nNdc>9Nt6JE=<(76&sCJSXcb{Vi1sG(QB(#m<(}aqyM|Tt z(uj|~4fE16c_GTm5+dne0whmjR23pb)jT59enEt8JQJetJVNAiSDXq4MM%d$jK+f5 z>)ht2d0uRP-to`^ucdU%ke3W+@zQeJ<+Lnu8QF}lpz40M@A4~X?4BTPHsYbAdCSPH zkdJsDt)x|6d?b5;pH{gElDU{5^{f*h30Z!UEL%m3HHBzPk|?E!NDz0qBuR-&)2a+9 z`nyzt2FoOg|F9xCn=%G*zBnCtC`l_m2~pI20h;?mlwJi1l8NyulCa{V3k&#(lP5^? zSBlfJ)NUV5saelPp%YOCW*u_lCPsNnAf=Aw|( zz!G~7lerZ!YO0K%!m7A>R*q>oYH-t0#(Q1nhxJv&ZBGt-WMvSYtN^V5ML2|t;$nvs zLR=JZwOtlFMdk2$oL!5aium2HiUBTJrhkg#w~Qzrbx5G0OBQPtWbwzAU1w~YFgY1$ z7)fJHL<;q0@)&rkfbpXmxX`~E8b1`7&qRqa0!k1yQ^2$_WlWD!#5^ZO_)5sb{=E_s z&Zt893-g_AQiAqYDVUv7!rWAGY-jZ*heTk+YP0Xhq|hBF1)-VZ5D1Wg*gaXi=Ex(@ zN&(i#q*3`n9)@lT%!43@ZVh%n7~&wO6j+|5i7C=rh>Oy}Vi{d1FV?|}b(*-sr-o0m zns_IugfraIu&q@@M3Xp{j3_`rP7SiE%FJt`gkSc+yZtI?;n6~)r7~V+DBy{?G<4>9WGLl_HUeq<3qV~qxEuF1gsrwa4LD(dmf4h?a{b={21I-LQ!OQ1jmX)AzvGTp2#p9{2a#Sxv{vZ5`v}7@8j(n&h&17 zrkBQInnV;lbdMlkAq3SSk<7P$fcf8#Fwey?XbmziOKLDi?L)D7K4UcIhTvI0``XtC zeBlejtvbdEtc*wRP$YiF1>^hSXjC2ufi;^o&iTXmLgf&I9tuN*Y9vZq!cpZCh7mS1 zj7bWG@TpLY)iFhCxI)6i1&&U}R+|g5(&7(h`Nk8L?P#D-s#=k6<~|SI#K~ zV{lynj%o$NNsqA&e-7h%Z75bcM8J`qr+r?*;GZ0V@EZ}}u-TsO*+_(FN8;f1I0)w) zhxM{}98iqKQR#ThbV`69)7HHPBcW3nhoPHs$T}W}SqqcVQOa|}t<5>7N z6wb_p^~mxVCUZxCXEoztoR6b)^KmrPCNXX$6%jEhI6OTbTMop+`*#%j&!^*?QwBaJ zBtvba2&;{9aA$rl-W^Ya{O3gUEr@69WhdgHJ7Wx&CSXf?1eShl#Dkh{%!qEm;VQ=V zE;)xEJiVCq>=G_=oku9o1$+`akN4-g88dhRhgSDv1J7lc^>$&U1mkPZ_rosyGNx5MtPk*Nx|J zdSMqFj(6i6(`#1-Uw~e5Hy(RlL=N>}bZ$RF(=X!t$1C{ykG`QeFG{7`_VON5WQEPU~c3!{4jY2Ke;C`o%M=orVmhY`2+53 zcn&|c7YLeu2V!Q-N68sL^ufEBa`h^# z{>9x#cf)<;FL{6=gF$>@-xIhp00Esl2>NjcmWp=};Qa(aNl)Q4@f;uh9^r@jL)_1M z4DbK$qpSH9Ruw)2@?Nw1@JsY5529v~Gs%=X)9h=zXuQ*hn*Ml^RJRv3XLwW0Ev{c`Z-sU>d_#r3SHN}n$rdd)c$AStIZRm-rC2e=Gp_UJJboGQi z>3X@*M_Fg)Bd{UutM*h{W=YSJw-HI2QEbe5`fO5nvf=Zf!xN4)@XC?w4IHWEx*K(IGmWmtk?Kv{$Y9S7dU|v#aS7Rx zX7n}+lifN($?OsqTId{bgFCWt%@Oo9ahad!0$_lN0^DvYqw|TGLRC4T&$aqq(Ja zX29+>WG|;Xo-{?I`xM4UH_dBx&~hx2svv(S0@)$By&8pEIeg zcO?mL7wX+>M>CS`XyI#T((2kpMW;RJZ~RVrdclWu-2CYLxCcGYV7lr87n(fiMsFUu z(suh@q;cDu%%6C$e1A83xxkfT)m%wxp$q8?yHM4Cj2vn3u z(<3>AEtH4y6M1Y~qK1fiRs3ajZQxt>ylap}?G*)VQq}->stUNB6kza60iD@u;7eVN zRQ4=(d87h!_T1UOR}rWCWU+(IEAHQt!R$?L+oC3W*{0gCt%Hv%0`l4W)OL zaY}ME@J$uM4e~fUp@G8lD)@X!4nNp)S!00$3i(xVy-XQT)s^8oM-}6&PHwA_gXkS~ zJp8GP#t()_u~>^!f7ilus}8s)v)5a%jk=wyaeSE}uDcK>dJI_|u?C(MY|h@e9^nak zSY)Dw$l0sm^IH+U>*c}O&T9LO3V8fZ39;O(anoiE2HNy7?c!=I@6ltLBGW4_>a#2n z1Na{{!f9V)1WYkN#@TiFDrJh=1smX>vI#O@H$%461n>1spx3Amzw~toTgtSS4{Nb( zs}}TqwK2l*Q>yB0nV8W9-OhAjnckZ5YgyRZg4I$n(-hbHE?Zh(Sg zJucp0c?bI%@tC^{8{9gu*U~1Bil*OGWXZ{O;m=5gLKZoKIEijdA$K%3U%t<^44TV-{F#kza zO(!JTzZJJOLrbj*`x-lNNwWpZ*||u%)`gpa?GSp?j$J`rP?0)^Ce>yH4YK`7X~Ej7 zU8v;gME}ihOx$9e&$1RgN$bSH(GHADbb?p76EFUq!Hd{NNHdPA+oTbF>&~K6r3Leq zI z>V?eCdEAEEJt=ae?Odiu356!OEVRPdWhAVqf=+X%x zn^xv8?ZkGY3pn7_!+bPXP@HiE@yzd9c_$rNJ{d@GVvMI&0Uj7QNJ z!E~mrXT$Ss0n>MK5%7tvTQ-pok0rUNJCY5zf0-~&W?tQtJV+-LFkOf-@{#%2xv2={ zrFp28%|VTE3I1!&ME=Y?yt$qM1?x-*+|0$92bs9NBMYMkGthk@1-l-nAdWi`dnGe5 zX&@7#M>1g-nSsu?xp?cDgVi^(F#TN)`h1F!r^pz*^jwf*7N*waBR@9}XBOl@*CifL zuEsOwAcc9FbC7;D9e?UF@gF-T#kdr#IGm13;Ve9KE@pgb0Rn#Jfcr}U+${?5C$RuC z;!5!Rcs}GdXF*Xs3-yn3(YdPtxjBq2JjJ-((NyeJO2qP%G_3oVhSQnZ&{)m5T8S+D zvQ39`dp;hwXX0UCHp)X8dpW^4z_4t5`kRBR!o`^PiFui~r=eOV6T!t9IHi`!7`y`5 zOwGijpP4A$mXCjXa}i&Vi6X|WT1J#X*q{V{HI*oes(^yS3A~Ta2S+&@nXj{;T9}96 zmTb&fP=E>V0xUBx#?S8+_)yPsLY&GlW%3!MtDQvZlNy|QQ;KH)BAjO$@VIF%wq*U_ z2uytE9NYGfbECEA=C1&Z1Gy znCv91Zu`TLyfqyO-^MuwrDL2~-QyesnV%d}?suFOzdmyMygzYtU%%rFb-m=66g=Wg zEV{?}V)u%(UT28Y`~4$l1>bMZ9NTdYTRn;MZN_Bq{+{4GHXP?WzU4&OzTvF?_=I!c>@BD7&u7lhlu=G~>?r5g z+;1H6ALHCyJ_Rv8GqGRyA4g;88>iKEoa5^<#Ch5GnWHT5lhe~%j)9wuCoRdrGL;I< zVJx+;TPeFIpTL^fayT^>;mKRZQ@fNS{e2O>6qG>6umES+bz=EG2Rw@lu#j<;)iG6Y zGO2*q$tt)`D#vfFd~B1$(EseQ8{AvR6=WM6>37N(0Z;M<~JFq z^pvqbfhTc0?N;+|?S~9%orasdae4j(;W7apN+I zk>gs9pf}7H)K$uu+*15qbP{cuEtso%2Kranb7Ory^Cq3ftDki^=*i9-W0~(&oMLmn zN=!am%KE`F_<1pQ{dO*PXrDyU_-TB8(1KcapC6lg4(+Pvu>4RfLRMTyt?wmlKG6%2 zFIN~t(~qfpu0VR?G9q0rz`W=Weu{p;okDQSq zmOt|qT;gL$v>$`g%`d>;VQ5yqLg(%G*#GYxL~1_b*W(Yko$~=#jov}+8q=roA?WxCip$<2lk*4~9rv-e=^@kx9zjC(1{MblLhH_5JXrJub z@)r~uzG2*v`MCMN;r*dMh%x_*nXP|VX5u&K=>LFt81u%D{)R2{+*_^xfr-p7xPSi> ze$BXo0M)79WH!PVsf$81Tse-eRR7B>|;W@MD`jttf={k+dnSVR) z-54(Mjzeke7n%nrFtF$sCUhs!mGViXU^|KC)iJI3`wZghnodQFXHpf{EQ*;vhvtk= zq5Jcvl8*Q}#QyC<+2Jk}Kj_7h)=Rh+)Qe>=dtvC$`dp)P=+|$-*q$~-KRJsgYQwVW zXRvVKEZ$wOhqUfl+-A)5A)b2td(nUayJm1-tb>h314Q`FLK$bE`=$Y{4lRg`sfJVA zN!)qP81k|@Oz*41a>XjVW7}R9tj1)j#l@xuaI3Xp{96s7QU`u7~mC+LAWn_cM5=|aV4R)@whe)8N|h}bs6uKNr^xEgU~ay?X4&LH4; z1CFp7I^<6)(uI0)^K}oTPPL-Hy&JV#I}o+C8;YS_SR2G@NZnQlm!8Ezc06LS?YPQn z+hv=3;AhHeXy=JLB%?GT4HdSGA1o}yR8~u^YfQwEf-L;ko{r*j<^!tD zg2>(ssMMzcGg)qmWjgZKXQ6px4ytD}9_Us!6s8qn>7{h+^~%AlovePFnu`rnng4NF zA%;S;u~)o+&3rSFF~;_{IunYS1sG{6WV7sanDv#Qb5#jE*&o-&Qg$CJ#?FW$sJa(I z?_UPiUQfrG!c@5Xuw1+OX=q%Mg-nqg>x=Q{VKz>fuxyuN#`)xC!;JAwhxD1(&Z7_~ zXJq3xs}0Q_F>k4NE?O)q@$GF5T$#_o`pjvZo66YJ+l459RDv+ZTTL4;Wc+^#B$gN9 z&SU1&o5sAh{)IRdT8X@O_4uWI25~Z{aD7EJj!IQx>mKG8x|R>)mI9Q_&OvEh9vYvP zzP3R&6P6bHAjXVr6ee7og`boovm*fAxZTU;*?V;N|~Fa=wQ@Jnurjj zP8R`E=Mtd)i~Ll?bdHu*5o)XxA=#@U@&7GFgJQrAd%ooD?m2EJ;;;^3;1yni8j|(nk$7 zx}3uFe6~It=a?dO`zuqliW1#m3`3xy4EgMkp+~;T#Lcq!dy~Z}I6;^e)(Dc$RDQa0 zo`yMo@Y7a#== zKmF@oN#eVgQZ&p7JcEsn|xExaP`IjIt`ZJX9dRSViibt3bk6=mGd442|JL7vwhbF_deIiVUA!63Kdr}1`^}h`Fl9c_ z4UE&=0?&wTxEHtyVP;!FZ6=swvKcQcwy+*`JsO8LVA*05OkJ@Vjzu=e*|-fe(rmGI zt_6PPSVCIB95I0=EU(ZS52w3h>- zyZz?iUcV9b32R|CpKTw$2|;I!;mh{%!#Z>P`_Be{+H9eA#S(-6obkxm1D$`jV0JM( z7cCp%U1y2smD|BrUa52zQ$whG8DW364(^2$=JBw7 zxw`=(116YxavNqB*6Bg;NB!i$;fk)E;v2Xi(v-P{nm5f+$QWB}XF4Ja!x$Khw@ zY!+^TORlz%4%*Bz05&4*vITY}n&D@i4V0p7;40+;ejyjMy$XQ(>mUp|24lKG1P-q_ zjAH9Rbe`IazBCUUdE*VOm0s|Avj^L>b|Kwv4_^M;iERTcf7NdvmWeSg;)Fly1$^=4 zt{apS8OOlII1+hlw7qx4KXYH`WxC?&#eGQD4Z^)qUv^L3j|X{sm=3oG*0bF3-}}AL zJ?RBkUKebxvBxKtvzigS1M?=iF>j_51O{B8t+o>ew_V^-wF7G<9a){}2%Ba{+%L06 z$~RXCj&6tDD{FY(vBrlKOC(8c0;g*Wq5zG%if+Mma82NP&4yr*gIUk7sTA9`pd1K@uW2z3iD*l+e>n&d8+nfjtaA^@jK17PxE zKjaDyfOE?O-*$Pza`rB$#P7l#ivR>fc!JjZA~GuoI}Z9lKQ92rXZPZ{UNAm&2cyD0 z9GiDAP3rFyvv2Rxx=9LBGomeQ#Z`==y z#o#26)D^4#JG@-Y`oKqv?8ZK12}sqKAX>Ymls^i=&EKFfGzS&^LYb zTNqi7E*`Jiqpsxn^ zK{b3CQir@XAxKmmH#!LUcbHcFSPkv>)UocIGOBXaG5S>zezt_3dD^)4X*Hq;l<@o& zn{EFFT4oub@}fG1D%9b`QNrmuMOvY3kvfi3aAmF=qbiYM3YLV2^?! z8vbY@IC2Acvo@l9t2V-ub+AT37p3}y*n0Lm47A}eRS#U>j8J*Z6onb4aLZbcv~DBV ztlS9gT0OkouZcWQBN%2Fu{;?d%2yf58(G}fT~*kVD(>{F<4}|)=#@H({58<)tAeg> zb==%aP}#7W>13MNqhy3DuDbBsuZtt|)`EJB;lss&7GvnMV~y~7x;PS*WKp?T3XPdE z@Mg3AHOA^ejPf#ORc@Vphq-KtgidsPr+#R53+ zgb%Y)d12hX3h5+>cE})k8{UeAuJLmuk~ zM6kqP7&_%FPkcE))BX4`c2Edo^+Nc*ObBGbve0*mBiBj>p2p&6W6Z-44`W_kgdtWh zjm-1HIJRE_)59cT^N_Iski;4dDeSx{4=#IIgcUM|R9qIVL!xNt7R9>dq7XeGit`^N zutZn{(k!1&G)e~lF*bBE(`7ub>LB1hJ*=6(23&lburb2`(;u6_^!_?*Yt=@c!y5cp zvj&|l2Do=y8{2MaV&R}3P8`u^x#oHZ(bYhDB+Cdss0(XdUHr}h5}&X~U9}O)i&)Rfe7Dqr zfy5eAv6{ht)*1-c>cYTCA74(J;GF1MOnqpM*cKD?C$7Wom@QbbaRbgKt-4|P-&jc+RLJ*hOL`dz=j zh&S)^KJ&V->zvPdqho-`4I3fzb`$z&MpH9a1^#?0C{I_x(+CxsQ3K*8Dwsl_@$?yn zw{ukR=d>DBf2~FLd4@NK)p4ys4Vg8X$d4m`#jlOTOWcTkU)9hyOBGp9lws{CkBAII zj8F|x_gx!LGqtE^yctt0iK~Bm3l5}_*MwgSUL9MoGmkjJ4V&>WS0C#?X+d+E1{9BL z;^qVcO!=pQz*}3;Z>@`)g*#BO;s5tL11RmInq?_H#~R6#l532ZMf&)vXN*0XThSt5 zjt|R;<4@22ITN;E#Zx_GzS71>ygr^y)WhtF-`@myaU9^cgPJaaP^uTY0! zk17TwRWL+4G^xE_*mj9{7!zDjU*ZFq5myYK^hD_dH!R!Xf=>FHneK#Za(1XLI1OdJ zV@Mbx{;RePY^R*UyE}F;s5yr@Q!kRg%?10qJgHaWg)2E;P>b=zZcoyS2>N1dwI5=o zNdqF|4W)%{^o%-%ZBH&>+Xn{-o;(SGf41O1Vh^49Ho!`E$SS+yn~5X#|FcI)lMCeI z91s}cgcp;XAgO+V*w9W;wQ|Fu4GyT-b`EPpPr~T?5!CV>hu5=X@V7aQ$_cjMLQdhl z=}A0aVTFAwPk~R&4jXvwA#&IeY31&iqviwOh2DsH=8Kyj+;Pa%AI}@T5xL40v#1`L zY~umx%8M}W^~B@lu8?_i7Vb)Zc&y-zyGNXm8gK!V7TRNc?>W5CI17Pq&X9TLfzK&k zm{}T#vx)&&FBXKrQvvW648|X-FY3&F5FFx#(jadvsqw;mOAl;z^2Fd2PhuIkp=y^0 z-c0bo^llfJyEtMgZKs8V8)`K#;>q5N$P73Sk5CUN-*v~Q@m`2o;|8-~SM14jgTfhi z;&6CFt&ub^Hr_~o=?o`+M}(TV606o5318h2_|gqcr+o0891H#8q7bx@z?LHtsMDE_ zTUB#$$9Xof7Ko`iERK&VQ}JN^Y;YfEKxvZ%4y%Y`W(D0Fou=U7->InjD}k9U(_lVj zJdRq7$GW}=2z)4r;f6`rx=j=+-4oHUb`p7FL@~%a34hHdV%GdgFnl>4AD#%(T3-y$ zHVGp|Mhr`(CsUrn6s%n!ivGx{ICE4K55LX8zl$@lpjiU8f%DLLXgba;T7c2>^U>!v z7rCG2p^P;C6Qq`4nvoZ#5880ZHWY`6y~BnP68IU zr{VAR*{FA>XG_;im|9N7xyC6N%M+#k)fBXdiQ>=s$>?_#Ma+shh^-Qb#pS7}eI|-H zlUa1nn2pe;>G(1*3rD*rq0C|$_P-RzB<<x6)Wc~gw)yjm>fO}?!q(iCT}`Ae5PanZ3(J}#37Y44@_zfZZDaQ&2wiW zS8@tnFVhgGH4W+O!0tJ{-S%@&xY?$<>qGLFZdd=CG-j)yk{w(a3$i(N)Tv(asK*}T; z2^TW)s5cfKpU4B&AB~MS!tgab0`(hbqLU=s-u`kSugZF}J#P0X=yxDgXqsy`H~ zdP4Enk>-0RqVd8l9I`WsBixe+u|+9J7D+u3ZJ6NUxFAp1i!V)n+sNi7bi{)>m8 zd;(gjH@4+_8hQrOP@a;EF})P%yiUWXT?q(jPKQC`CCr|mfZ@($^thy;iZ2dDMQQj# zOcQ-i@^0}Z!HPJ;eyfvlTZ-Oud@{V7lA$1!hML34uwR$}@q5uQdXz$3heTX5B=6mk zR5*kr!9OAevxm|VQJIF}Ls?K3%c1il8=KA%+h=PQwob^y!+~t5YG-4&GV#W2av^j! z7(ar;u=q+OtPat!$&H3Q@maF>g`jL!D26}z!REa$9AW}6Smh7l{s4SC5s30n{wSx{ z6^F%RnnW_@iN)gW({MQM4M$gb0-9e$BFR4tx3rTn^=1Tw?}j5tEE-SaB8guR4*lN| zct}3F2V+Te&C`sSJgr;5#$d*4;^mzSM!}?5eEk)S1Fz}0q)?_rNhHqZ(oA?>7^LZ# zB-8P^B0?Dw(?g+e9Ry4c#M+z)L`VcfZ5;7|1OpM;844kqT_1TKjrz%nm`r}cu#Myc z?hhfYLNM|#hQa!+A3Yy@@T)us(&IuPS{j5L(#b^;H_6L45bu>jaN>v$LT>xu){HIb?-LUJq;k+xCTb}|lT@ex=}eNXxIA&~3~rF&cuT&{;>!NfS+ z{uK@JZSh#D5JkGYNZkJsLCmI5s7D6Vz6?c)1oa^|CPF$r76H?UhxafJ0SnWyn)pQ< z8sqT%TqMn`h>d-j_&x9GSY1iNA%%DtJx)eJM;2HDF=ND&VWFLYOi4N)ies>RP88PA zwbZ_Zw1>;0kW2eT_jv$(MZzFs7>bJnYH+)ujmcwMaK%{%i@df#m+}k7CvT!_ObeE; zRG~3l19J>W&vKaZSdM5z@4PmKZkm!;OCL9Oo8ZzWV~Bh;z{*R8n74f=^ptQt90BWU%m|d?zZ9(!C9sHSKh(~*k(ABPuTU~m1z0m-%LniRuVvOIF+6XW-!n`6K)D)=U zL&#=a`A-F3JJ(}s3&Tj2CbA#t;CG4^BpJQ$3~e;o>H~pVbp7e#`5$%UZr_AI+8Quy zScmUxG|^-Hx(vJ8^8UA*w3(;)8%5hU9b*^WQe|RqH~t z&j91@8sLk`HY6l($HDVkvFM2|d8oCp;L=u9F5d=)8bf$^nBz<49?b0D2Ze>Zpzn78 zSs!=8Y{ga>xtgNl+9m`$Z$^Eh2`WPMuzX|#Ca=;$bH_%^v(v(}JKDsL(Z=vST}aN_ zjK8&4p%&kYZE}}!W?2{IAGP2?@@1SatjDR-ZP4Ux!iSW4w7b;f#*)jZt{}aSK52!t zuEKV>i~J2&Vg9`v!eL!ln9&Qd`Ca(>?;1Tn+Tm^6g_&*b@Y~o9iI6LJO8TCiG;`Fv z+lY@6P0;G9!D*pJY@S?!J+8GdwyGp8aU*`DQ6{lSEq*+!$FJyW6t1ekBehES%qxa= zd<|Mh%P|sLgN^G-@mH@LS`SO`;3H+M@a5r#6LAh#7od518KN&&AitbellDz$+1`QI zG_T3-Ye2$l`ix2~_^efr+wHZ8yh|L|kQ^lImLsaH6vHXS;Ja6frTfd#-$NXUD4KQo zRAN$Z8LSnmv9PoPn?_3^ky?SXYpO8Ky$Gvblz~$z!E~uAOp&R^6+u!qx;NtX%4SrL zHR4QuBgQ_pVW9-gWwM$flTk~xOf{YiRwMNy<-$Z%!th}c4jw8e?sYLnwCeDBaTRpd zH>3M>1+J*o;d5an4n~w=-X_{#%1sz2TZ^DQjS$qYN0d_=o_948Q?&(mJR4!V@fsv* zyYO@L8m?`fOuV`N#2zdkALQISo9(mzek9#>6nC2nzzL53Z*__ zD0uk8v3+qA{MW=lZa4v&^z2OaO~6j_H47`mVxnstjEHU4*F!U-kw}b@SGU0cFMuy`{5Y1T{Lvi3n6y~}`Q6H2%-c%2?c_)FlB@Q#SNaGY6P3*rIO!^ZC z`F-I~Bv1Q!L*n-bCL-+-J#UO+31SeA<$*D<5sO3Ml2{zHBsPD03dDueP`u+366(|8 ztxW!I(mpw5(7pHv&4lRpcgoYS_+bjv`{)^WBngkkCqlD7jN!dS(V?q@}^0KNU8=;-Tk|i0h8@^%m6v#^lexnU0Jx(o9Iw z+-H#ZD2652-kp!)eT8_xuo&Ew0!Z6tVaF$8n;b}^e2IASm(tv4Q4Fkpk`97D151_@ zE9A|7%8xpLEtihKdF%k@$sa=0*?s5;I1b-=$57p5i32aKkX611MgvEoHun&uzna1S zg9*Y@4nV8oFs;4q@UHhTaz9v8%}VQ|L&vcz`7oNxtZ5x(kBFNm(JkzNZ^ky*DSH;z zzg$G`7kl)<7Q+wDBEs1gA#G>z-P#`Ww>m(t&<5vR`r)gMFDtCNsRI|Kh^CoswX1Z6&)Mr-&P{Pv^XcGwvdMxBAPqdl?p&*6ZJ8(yje z<3eZ<^*jBrm)9FUYrQch=L4G_cPMvyLYdj%oU0Y-XKkR`bPPfp9bi}ML~}fQG!HtU zQp62~qxO)`J&pkb>iw=bjS|t5ux~wqBFE#9etrzYW+%}eehjY-ZDok355(A>vj^*dj9G0*))X;mihm#J+O^o9qbQ!%pC@z6keqXHge-j`(tR zXxVxm@iNx1%{T(VLkDq9;uN-z+7e&nFhpqo-%h&#!;|Ik3adndaS28{3t&!qo1D#s zNRTRk<%$9{1QkJIRRInxF2YimBI@-Q;FU-raSL*B@{Vqq+NV5uz^TVMSm8rXMK4`xS+FAX$wmoh772tb)g*D$1j&fSPYTT)pd% zJ-3nO6D?TqtP&@eR}c%k45G&i@W46?&Le3!JDdWg&r~mcNy1_~V(mm_5a%L|&Vx*d zz01Mj=sfJ8_YqObg9hD`-^iE1ST7fIh4PR?Iv-5P#jpz|~vf3N4^L?m$!^@!IuT#TF9RdnC3gt$TlEG;Ur*r5tZPIZt9f z5i798GY@^=Qn7wX2EyhRf#+-vekK&4{A(5#7?Gw~hw$g!w9w zEmMx`bnLDU6(E%6^4)U_uwY#dvZ8WPvLPFln@dsQo{wgeVywJbOpNDzVxZGH?Ry^5 zuNJ^{dsh zyvOaG-_Q9RzQ=_~J>^_pJmuCtz0U0$?&0pe>E*(Yb#QB^UgvZ}I=O-$ecbVeHcsPI z2Y2&WI~Osli{o)_<*ZAtaVMwU<`N2fxsoMq+^M1~-0G@UZt}PWZcTO_cSo^-yIEMn zDeh?GjJ`E;nWe3qfK?Oc_o#>a=X{f!yYxP%D0i1zy6Y;J-P^n|HXc z)Bocx)(&u2zdqp>2@G&UO@kc2UmsUD^#&Ih-o*)xb#V7D_Hj>Z?r<~T4sg7)UvN?` z54bh6A9BT+yiIm6I*H$ql}5=9crc za3}hUx$WA`+>_c4?t|t{Zlz2&H|_j2F4?$;o27J>`}*S+H|T$dd-doEXWssTbB=h+ z`S%WT6Qn+HL-qlfsN;uC^5o?c^}*&1eppoMPkB&2Sobvqv;O!Xdc+q;5BOnnI%Nc0 zrtBzj>ciH$p?#-2a_2iDe7hG+xBw)$d&BvVKMYL*aJ!he^=;wM3MVgGdN2f+P@jf4 zJAJ!?ac;mLt&*Owsv}m|VL$w_@WH6F7kLo9U>HaJSS26oIR&C?g&%~DQ-8F_7e(bB zka_4uyy;*p-w+0m^dS6^3c{>me`Ng$#0Y(k6Y~PFguHh?2L4z}eSl}Yp@^&thel`& z+%JSyrc1NSr7t@1F6U6i)Uvk2dvc}bM^e;&JzyNJK?B&6N%-~5m5Rs3?+&Y z2v?)pc6&5>sn<|>ITWJwzKZ5SkW~+WlAS-ci%~|GzBhgcQ+~lk^6Hwka45?}kPkBqchZ88Fd+;Z{{*9U2I+=~nZ#F1y$r_35*(rFz*g z5=%&Pb-ghhtx^$4MF>DWu!S8FxMvy!HOo-)y+q(aS{QDhjDn#<6y>W$6E`vj!7j1* zqZ{Ve6G=RQqay&Li7d&^cTuu{Xr!JD9b?D zLk3z)lyRUy7F(9e;prvv{w`e)s}Y)8x+o&&=UOZ%TnnX#s<78m#Mw?jzFHnb-Rtnd za}Aa`C=feF4I>LVC_Pd{2F*Q>=*eMCyaGam6fpUe48DC)K=t9(;M0+ZL98^M|7PeS zuWtHcS*+T}V7g%)ED{tk=bt*nnrQ~RPyy!T@!4*)628ynaQMhdM7@?q_9_Kb98jWT zxEkX^6`;9u4SqTTF(=l+NtwgtTQbB6BM;AIIVeag!j_l#HpPla{YJckTzM$nSdF8n z<)BKwrGtOT$Gmq9#I%5Frile|90Z~`5Hb>5=PF>2B5}zYX|^q~3LQ(N@qOKT{5Z4@ zqrcZ6HHCwwxdIISlHVyq0a{PiVV}G_j<_?*Wl(_b0>JE+3Us0voV}Eglf%L41Bc?9 z3Q&5s3H^)ZG5CVRDcbLsD;2OYZXM2*D}cXvBNjMO4)q~bI45&3Jf?yrGt`Kwp$OlT zn~-9$76wTmzb^Tl6@m66%1HRB275ylSOgPWEO$Nf@{|$xiDqucl;OITfyymZ38^6O zTPU#|D6>M!50;UEsCA5hekJwmoP+U;&litw_@i=R0IXIAKq5H^HXZ>meNN}pPj9U6 z_Ce@YFGN!p_n#xO_Bp^;-w}3w zXK+r*9s@LwabM;D_Q)QVVvaZ%;|B9cH(1G%R)Mb1g#v!WunEAy)4n)S!`>Oh?Te$#4w}1ssqlhOvNzrjlE1i$vQq-X za9S-GOIyO=^E?#6;-syVArDDXEKZ+}!vcNkPd|=9S5y?f6%cRnd;mHQ2cfvu4@q@F zklPrF)E@FvF9=8RnP4P|hJ&Xm6xHv^j}jLGL-I6>c?P2AeGtw~3&$er>2BK;1SR_* z%(n=l`CBk1h6TWVHr?kg_`qW{1TvD814`PPSm)E+$NR&2|ZcF;kC?XS^f?e#HzZtV}8oSw6frHGx<|v!#a+r0zvSJ%W_OpYLyV!r^=^W1)uruNMOx=46 zn>1kyJJGm>#qTmmu;s^s+~-n z@-AdHnzO6qr?s^)WvNg1FdxpGJykGgukV^M8;xCT`zuq%tF)7yh}_Dm-)XaRsp?FF zSB)he*uZ3y*0ay&H!wcIO{^wDn`PM;ut{cy?30u!<4-bYpG$|SU;PQ^5{4kz^PBQK z-(!!+dwl=-hS;eu$T#o^4io;vkK}tWt9_2(&jau|_7eSRFJOE2DFQWLW7Wau@XCIJ zbs4XzC;b#jYhOS%<0Upve1}68Z*h|H)!tPNVYvP?IuyRbD{2_ArXyH==rg?NXEJv` z(rn^Alm}m-V#iZ7$vndV_Z((T_i^d=0H*0Z#NCGv@pS5cNGg7WoUVIF4|@dNsZViY z$x~bwdIY;`k6|(KG0xw5jH z-4}1MOzRbXnZCuv`R}0G_Y$8P-{Vo}JA?^*z;>R`h~oQ%OO~JUrT!ff<_=@Zg_p!> zdjqcxcj2Bpi0i#~v3$iKd>=nUg8pk94S$62?*sUI>JAb^2Vi`r9}ms0ECKA1FTI4He~W8rylX>!bO z-9qe(fe_o%Bh1EmO=S5Cg;ivlv4u|jY@kqxIX@C)JJR^rN@ISu_kbX) zz9hg*m+-Tu5I%N9p7Qs0{DsTwU-r{HQg3JkZ+Upw zp=4eb-93txuXx$MHXb&3k%wh${e%5`d05&r%Hvx!j!nKXj`8mnU}t58*km06cK3n+ z6WJli4tnx469XPre^7w^E#YTlA>-I=0YTt7_nmJf4Llqj+nE>4IYy8 z+^3qrnAQTRWLxQ3wSSo9;*VqsDI!A{%6$d+TxC5yNN4I&-4XXNSm_P7e06VQNPOv z!y2AYJ>dmi(n#=zdBQ%=2|LpqAs}f>-09;G&pw6^7Uzjaat_y>&Z9DdYJ_4>G-i4r z=A0iA)&wB3h1Mcvv<`UegVkN`826o6>}Hf9yu+Pp3lHR2UxfddBc3n!hJS}I;*#7k znb;!^#QBJu>;c|(4?I)##9uXHm87{r%$I5o;>bUveB=ecosoX#0!AE3pEJP~y|j%c zcm46?i$7&(`(nFaFkb8p#l6SDI8NTz+}1$MpX`nK{yxxDAZEzjKs1L0BAe=)I|uyn ziT0ydj336U|G~C9{OrA|AaRE1x_cnRydU$k0BZr}UnIgD#e~@Yf)dLqK?>iJoyDL)I66=2`f1X)&vFq@Vrz?5b9 z+1GkLHdTs`&6@Qa{tmoM|Cb;;rq9Q0R`M~sP(CJ~$-}Iy#<7Z{yiAxpTS^1`%_$k1Ign8MoX`>i=@fA`fBgmZj3n!!g;?KwrxXS&+9Gel8FC*X7n-QFJrtDA8 zQM_M2N;+~L7BGpI)ftE|0cind*G~LR+7A7xd`vfhkG-aQRhQF;7V%NO{*o*0w?pccf9!<#d6^kehH@)(_s;pg!i|Sz}HN3jTxEv^*#&XgUK-KNyi$u zbm;v_L&sP??T7M#Xs2RSeNR1z1fz zYHy+$(RRtsNkh@wbS!+9hu?dP@aRJs?zj}=WG=DqmyqAiWudGUO^F}33tg6Rs{zg=Ptj9wO;^SW}g8YSI)LYjh-l+ly^=XEAE1xpPvk)7Y z50A)vXiUvROCot&X|}Yiwh~kMh*|%t99xeT!^x%!U2|!MwTq68VhJv*6d>SH5jv^2 zX?BwK!`?y|M--9wBO7Nw=3s$KDSnn^@{8q^Us!}4T3LV*>OrgQ z%f**>xw!cy4|8kNP<6iy=XFc)_GKwvloVo+_UF{iG6o3UqXJxpWN@M@htWED1{T1+1yqZ=vvZ6jhe)M34QBW~YQ$1?Jm zbgtb9sUa;$4eDV3s6O0Ejj?i(9#m9yp?hBk$yOQ&svv!SD*1i3>QZ*BCca$Uh#-D7 zOchkYjs`{^8FhRZR>O@Db^Oav$6-lzOh2rLa@|ev*|!c0W^KfVQDp@7sv~8M8g>j( z?`ZhnOF?^ln-T$Wk>a4^jRBww`=2vxe7jYZ$?7oW{egP zGkr`GMYA{K;f8>SfS+=e_pb7a%k15X zwqWmw4ie*ZG4rzt1XgGx%});>6O17rU_iWMLzqw91U=H6OnjsQtI68fp0gebD>P8& zMU4F^q?tC~j6l+P>=@9&-DN!e&iOGR}~3T9r*!jxkPs9R4S%j2;Keh`Us zU6I6SiNpC_QCP`CI=h%yj8ln&@uGO>rx2@rd>ZnJ`QfFTfGad}S0q+PsaFE7E5+i# z5M_V{CE`a#JWe~4{$P6?w#P-I=t3MZx)UKjIUXq|;$YpCgf;7uNoyX9=w|x+5^0$E zn`Uww@?n{hN#2u7sOw0?$`;B?un&i!ZVY~&iNIb@%9o?w?mNnAn-m)b+xfwW8KBIG z(NJtvjD@ySJSKX@V53$n%I`#B^0P2pvJJu3C!z3Bi9|+MFq%jsH<|pbiv(jSV~?^q zI!S-`G#rssu%xWJb4TLPj9|q7^Tq8Cp@@DEfI4E*Sse<;@y7{Ro=%^$G!=S>;=xNg zHfMtv*sc!2$6q1fj3aUFZ6Fc?BVZX84e9j}cw86(G1(BD2@b`P=|MO|jQ&fHDN92# z8mi%;_-UUE-Q@|yRZ7OCnnc7jCqkYlnKY5{So|-BbjvXiI1-6*bE9Ba9tX(`(yxWb z!kv0$|40+~?RPA^Po;ymAR3?JW6^yk5jigLl)0BcI=)2GK}2J?kvzA%!;vCHHF;JL z!d<9eCgF>x*+DpAO4+>Xaj>u@4P|l&N{&*Fv4B4&>HEQo<_5`!iKllw0FP^;;66P9 z9#7&SF((@K6;aq@8jt(3G1$^gef_y9XueFja3V2K5lw{Au1K^zkAuyFSk!%s!9CGS z;5{1;v!N)ArNt9__!8v~M#4cPguHq&s7NHQwN(OM8OCA04P_wgjzG=q7#t5JoyR50 z5qOsdgAmF`&`8Ax%B_Q7m@Ao#E3L%7>B+`Rs>z16QxWD@00YuJIm_fC zdVV&h5sPE_4$A*fO@rTZ;=TB%U{Tp6d~%9Kiy`SWA5iYywRj{_?Qk$IA9CtNFqvL} z2%kcXOen^soKi#!7eRpL#N#)VV-nRCCroKpJ+2a*XBqadEQg|0Iq8V2p&(v{A5Tkg zmhxv0d?wx117dSrqIthdE>3x;;~j1P)!P}c*_neoH>mCtBZkJgR9MuekRKxjC+#Ws zp)UoTdO8-DQ?1~gf%7VtD90)vaaW44Yf24l^vmFQhrIH~s&Lhz0?#FBp3c*N!b#OA zSE1+P^18LSy0{wpbF1K*L34SLGVn)Npd+mcJiL`qSyKul z!D^cCm!kUtJ>UOQ?RBpNq5_nS+nA5}C3z6LmWx=+Ohh=OQg$TOz3bC)=jbJHLW$77 zoCZz9G#vNK!G98ks41eHpKZkmU0Q|*`)V+op0_b6<*-sF&4OwL3>1oyPBZ;gMpTpN z6k+6hDgJw(gIz{cfB!7Nl~-kyt4tYmq_0>IT}ByQh4|K0gaX+@gofrrmox}({^elL z_$;U<(lMfXc-8Sb%y~j=u7%B%q1+06Lt+4|t|8uiGsJhbqp-Oa2`4I%{R5P6RtsG`$wBi`>;$HyaT5WTL1HLn;J zrYcj8lRP%7$q-9nEp|t)g&MKa)+?<>hs7$aSSX7F&P(u--s>g3*K+ff2+)y(kgE)S zOjCxbvkJ_&ucJQYdd$3}fZJuO(Ya(bj8)}fwO<~|YuCX{QysgMX?C?|9ZXIuB6e^+ zEJWpyv{V6G4=F%DOaZ=jiii`D$B)NKNcp1#j{*)OGnJ5eo^+}|q%mbZ&7z2jdoE)Q z)*dBE%^QXCl4BL|UP}SX zixlwdnG)&Im&+FugHwn26M_wRxuqTkUnw{7M=j=DuZ3-G4ZhL$|2i9? z#I?Zob`#w5S`dD#4H6=4H0LGlqIUKQB{!oS;pa=?aiAR4=A{_8Mf?%JR#X{Yp?t+QSX*C#cUC9m zlo4>>vjta8nou#)h=o(@G0@%wlM{8=wxb$NhZ>M?)QI}9W++^0MT#J4NWWacz>iMs zT5=sap4VZsvWs%dyYV>V27OJ_zcMZnc-B|McGQJ#dLSJeV zI)AsLrLP9XUm9RyPdd?EZIGLNh3dyH9O>vnfN&=oe6J$ldo6TWLUp^8nyExG|-Ejr%EI*A{56^Y}o<6-tq zWZb)sMe$vbX(fJ5)-~|9T|t!bHB`6vAThX=cvM#~+p~^xZkwRjUxo$zRhUE?VI#UG zmR@V745n6y1b5)yw0_(?(gnTZE)0FXg}Q)i_?B}6LA(1=@%SoauD8Qjy#@027ch@z zreBO4pi=IFZ}I1eH+dc#c+R6{y%UsQorKN$qY#>T0>O7|VJmbDTYjHKh53HOm6ASG z_81aZSYwax3DRjEfy}HUNbnOJ~oP6+}r?7SG2og^pfLQP$$S<%!)Qh87yxW2piw@WvYzrGvJG2QopiKEJ zHf%jZEUe?Gd~^t`dM`8@_Q8jAqRxXB@Hk|RGcQgeJpUkkS`LBt(S95hwS?vzOYD_C zg+(f-@WJOWW(6IDOYj+FRUOAh@srqE=ZINX9Izn44j~t8@xuBXtf>F%cF+z=v+Rf= zV2d~>>fcfy*G$er3eVSIk|pxbgy8 zrjQT+lne1}eDLnR7akW}B;CFp`0ku1#(@KIicVoT!Ui)Zp2i=&Q^?LbgF`WAP}+JF z(1ecup^RQc) zMTGj1H?i?F=^9QzA?^fr*4x7GsRNd6vPXr68v@CLRV?d=mpP&IOz^{Os{Jc_eX(hl zFPe(o(cx-O+7mZSy5vTzCr7x>^@KZdv3+@5pgMX1uGcT1$d`BmPn|LTm@_u5@y674 zcZ9riMRow?s5f}v)jsNFN86&m={!okT`{G}2@b^gzWK-voepkvT;0H0eb5r`i*pX1 z@Dp%^O`#{=O!L50@-2E1$6&I*JHn$VPxX-}vC%w{S?NUAss~bJJ#cKED?QizVWHuR zkTc|?eBzFuk{2k$_5zxh`r+PtS6u$)PPLUA;^upiUcehWbOT}j$RBr)`OrDxgS@+T z2tDG44x+YXylLJ*eDeYibWQcf85LJdcJRX^^0jW#C5^$s^OPs# z49!{25Zq!9!C+^&_&FhPJ^4n%J-|%dF{kQbKBaVKx2GakYnodsTS z8*qcgJqHX1yP?d~4Q(ABuq<-Md>%LGlE&dloFgJy?2t)5&wEMUNbC2)aET}GG`V9u zj|Y;C+@av$h@100Q7q|0Yg1RO_~8X4X||S% zx}uWUU|#C{>{b^qJ7Fcv3SSB_{u)6h@mGL#Eaqc}2gjhU@dp!c^D)CpJ~pN?j#iGvi}Mj)KhY2_G9uGnf8B-&;PGATP-BQ$(2UpKF&}GG=Vjhkc!|Hv!=A3>Ws0q1NWR3&TB!dY*fffTdwAHdgM6%fl!xtC`3plm z9(H*LFR_aRn3x1V`+JgRg8jcR+w?b%%KpIoxWCxr#?JkR_NPpM zY3uT{(>xQHnf^a~y78BITS9EDryx81nD+bH2~5IyJjXq*&N6JE?VnoBW_KzSxJPL7>FB+b+-R^W%k#=h#x_c6|<;966soUbv8XjxAs+lJl55 z-(1QZnZt%xEN8EZWZ1x5IkrJlh7AcTu>a=Bu|LhS?B+{ZruAVp>u6Zc{!UoJ=KBz@ z%4iXLe0l}bIlPi7wy$D}O{-X|!V1=|w1SQQxs-KLrWDW3#q7N9V)jpZ0UJ2FoMzoi zSnz8}RyDYi_55DR=6Nq?JZ`egdBrOBHByS@ZI@=h$D~=ji5x3AEyW@$rI~H?3id%; zhMhl7OegQPY|K=VNlf6_Qhf#X^Nb=3@{wcj&nmLLeR6EUs4P3WoHD*ll~~_PCDyc6 zf$a@t?Ate0CJJ@tzN8d9qzx|1t-z`SZ5UopT6*6)>~?EHs#_!6G|REg{4!#$)}hpm z*4Ky0x0GFt%3sBZPproB>`JWrQ;%NNMz}aML1dWv%E7&8u)7T5#%g#LS3zv37>j(1 z(Q={)Dbq@*uUZF2w9N`M8vvi4~d!2vIGBi)$Vx zJ}X2mc_G@Wi;>0ZFp|@NvnsS*ROcp-6+w4K4a92OG4CL8J>N8A)zntB)z;!Pe-UD% zDv)E=f?AVCRH#)W?|Ut-kT+uF8O=e$>Tru{=-qo;VS1$no8~$~U&#&ju5Re#zew43 zr@^Oh4{=8~(9<3j{3qeF(;By8EJ^=C9EE^mXqidg93k@g6k0>B>jcIL5mV{)8S)cY zBKO`gC^sL+q4ln?JnM=rhpo_`Z~zTfhmc(V{~F02@2k#W&lb}E%{q;**Im$}aTemC zt}xGcLyw#jQf?C?iRzev_b1`kaSWm7j-Xx20Z(VxVXOC9bpJeypAzTrc?D&riyTAd z33p^y`C$AzH{=f3qivWnu7YfEUi1`_WiAr`)(ry-oiUf+8D}53AW_i+Wtyk(tnm!o zc$~rijdP4ZT7g|pHVPhLHA0Bx@ z^^gY%H9bgQ>x&h*AoXqz?`(6)t`?(-ujU$Txx}reR1A~VzLT9xX zQf_$S(pP`jXGXw&FbF+6Wnj5*Egt7l&*wa4k+1_itW{#AbLv! zn<%d)bIS%O@iBO`aCj|3^BF${T#8YE^Dh}}-m?n5i&sJSpESarDL`P9gMkr8+0u$A zU#SFZHC0$@twpq?GJ>y4VQ?X3V9k+(JoSwv_pJhRQ-uHUD%=nvZSklA*3IFdx`aa< z^+=`;XhOYO9on56z!C9AO?fQ@jn(i=Yy(0p)L<0}*iKP_aVo=%4N4H2$bpZv%$sI2 zd>dMeZ+mH*gGrCOfnnqaX@mEz#j!OC;B!&N%vWmI=BkLc3Ki5YP(?=Tddz*c9&8=4 zZ|L($AC;qg1UcL=kcCXeDwuRDK-gUl_tjV7j))9O`&Piza5a*2Whq2i27Fs&&}h91 zhG~?8B)k^9i{){lK#_V?Yf&^s0f)=v;83|1o31Ls;`n+v+irj(^^ojKwDCt+3lp?- z5ag~0M`a^CO4^7scQ#=6b4^%mqm17J#E5&x(D7j%Ue#_w!u~CgZKC-Y$6)n%9X@|& z5Mb+ZSZE!VJs|c?oD$R@siE#R!;;!{=oV1HvA3$YFi8_qtsAf>+XC^=tg$oM8m8y> zVnzk|&K{h`juJbZQKk&JNE@7da|*w^Pr}jaIOW$IM(K)UI9+CoxlKoL{fI64LeJpe zuhZE7%a(kAXK|VIF_9gIAd`O#GkY!3^~(ZNZ4Tnb!$bIV$O6wa4#2bN0QFGzp;Frn ztG;f>N+TWoQZU5gJLdSYV;9nsEb(;vQGBpGj8}Rm@%hV9eC19dZp<15R}bRT?*lY5 z-GkHz#EKKz2~{<72pb>4|2RAEcr4#H?voNSB75&yQRF;67hkC;il!DS(v(U|yCh`q zj6_ltMM<&>QHe_0o1#TZdnrBV@A>oj^YMCliSF*s`?{~ob)3iX`Mf_Hmn?;=NC2Pj z`=Zg(kMAK?V1rTsC@>HWw^w1Q;}S^oYx+p%B8>RY3pY);=do=za(B$*KA+mWIa2obeFym461IeXg?xnv8b+idv1``qSlOORn1$n$%Bp|Z^f^7EEq z-jHGNQd7X?5IJmdl*g1rY;hE6VurO3anT{gHFI2&H3l)^ySH_O*YPjDZ2ZiTy z*nUD8-a)De$kf51(R#4ltPTTd&X0*ug^h(g+FtU!*Z>Wjs8qxDajN(@OB1WUXyN(< z&g-&M<;PhO7U{|eAEJr`Rar=`83vDM(l{(BgUL^Z!Y4)*pM6yEAk#`2T8_#*TT%wMZp-j~njDUG zNh2>*5uxvgV5_MDQhqC8(>-N~-_t;li!z^At7DtBIxIwWFz$*5#Y$9by z$*E(fyDEa`>he5jQ)J~E!1@}0{$-p!w$TvDs=6>s*2Ck&dbm+zi2H}M;25up^Mmy9 zCQ%0->vdtbMu+>84YA8f7hOxWaNocP1EaJNaYG9SQjIWI$^aLZ>SOLP0|ebM;re@H zd|7CWAairr8k*sToDrr4ahBUbV|g0M7Z+6T&OlN5z11{ghF0}e(oABsBE+l76zIKnIgu5yrH3>rEDy8{n8UGJxqj- zO=iNUE`7nd)IxA8Fcr3LHx)v?jRo83d|%;eD$IOrAzZs?A-wRm6vj6U7v}u75(aL! z5<({r7tVedDX8&U|Ib$oVd{5N;nrOfA@G)wVB=yU468E`d^ji8f0CBa^-e?BQ=u-H zl&J|bB2uT-c_~b896mghlg532ttqgln>#t7>2& z7+*CNB(50?XY)*ixI?DGIc6+0YV)i;MHAu6be=7kXDqDb8LojXhJyD~Q(^N*&ZXVO z_Y0r21d(gHLisr@p`t)T*fdN{Fz!$lE?rR*LSs~gK{J(vv5hLiko8(Zhmnrpv`t@_ zGsj4H+HWd&sGACH4|N31d%8lExPhQ(Vj!&AWhVUOtk$xA6JcbViICDhT#%h^A&k>D z6%M$Y2rsqy>(15{mY&iPrbcTEO<_91k{@cq-rpL+EMs-y(S4rB_}54n9l>*EubK;+ zPYf3Z95xfK?iwMa_zV~H?(%(6zok&nXC=&hW-Yw~fQ{GX4(qe0fV>TbvHAAC;n znXr`STF!sx08@ntJoA(5MYh}G<$7y;^5=EL(NWlPb0p65OuoR63>t+FNM39Q*L?o- zUt@61k7xE-j)og&H@@3K_<3Rss(;(Uu$%j3?gQNm|KnP%k(@Czn)4bMpCbu~U*`bz zScX>@9T1c<9?gDsxKQAL{zv1mmuHx|?6AY~AP28k;yeM zwpXS>Y{qzG#5(g{tqaD5IV11~=Rr2vp&GUv^mBEZ|(SJ=*%6apu_scytJO!M#|!`I=SoIP*k?XL=?vY~N~)3;e!5)yEN* zgNEa_hb{iR;JKu>JcntHHI9aj;o1Ou*lPo=oz^H>9FMSv&o#eSB-l_W_9nw zyl>oh!t2h&?@|2saU@!|MdQ}n80^*NpO<10XA^}$t|Kkb;yEi3ahTK)k6&?d7#+d0 z>^Q^e-s^ZMmF|P%zc3W}N21C%68pzTp!01Q<__G4rvLHQGWon+Zy(pFalP*Fz34o$ z1OJWPh7C`5;lZLv473WtL`7b!@*ds?{+vg`W8pkK7W+G6u>22a@$f$7KmHu%)d{E% zi{tm*{k%@#9=rZTygru#X{!XBFiFHoqeM*GlLXhEB;3eKL(#@m1U`y|$(v}L?u_EO zLn$+mD4~;<=wP9&;;_;3=Ac`^x+Ap6hF1nuJb+M7%$e zfYGa>v1n2pW+!tl-P}^l{jV6FBB$_Om9xDnPT@;P8BCJ6C;dY$cFwQF(^u74zMJps z#VgUwXXqzNxu?Fr7&V#4QE~7j29GF)kzoOJg+iR+e)~0zCtz2|^ECFAKrEyP3rxyz za8?<zR*otWKBt~ngBc+eXf{00{T1~%rF;Q4C(q+oZ5`%h@ws+i4J=-j z@%pv`t7leXoACuyBv<3>`*QTDo<+^hbExLC|CA#ak^ZO_drot1uQ%Io=@igQ-99yUU6^9}5jzJlz_moT-x0os3g@4>PLFT^YG z>032go?n1+ZXJ4huOlPpI(GiK!gEe}reWwcxX4|{pedJ;!s@U!=mKXvoX6y8H8?P; zhI63nu*U2nX7pDhpZ5;@voCVyAfM~-8C}Wn8u+fLL;90ToR3%!z3DaBrCW_fe9a{; z<@i~50e%noT>n%JHrdqT*|AzY58!%1&f=L>d>v(hEpQ)j7Z;Y@LHvzojMr@C8E0p3 zB<&<}vPwBiu>w;fk3mJd1QB=2kaLN@uR}R&Cz>+=d2@1l?V7;dZol(VD_^*i9#|3!QU5*Cs%dc8liRbQ>n8DBW z{Es!*yQqxs>1yELT8Gq_T1Hss&RMoyY9#j80WA+Ls$RBouPS{w~J{XHn8|?A) z{Y3P>9gA5C)A;#23l4W(aLUXTC)=k$clBf>y>`O-7|tLr9EZ{iz==?vi@WweDD54G zAps8fsmW^uzNW%mPMF{7jB}9_;aD;We>x|@myL&Nq!YTb>~UUd5;oj&K$our*4^U! z;qQ)Eb!9wc);r?*GA9@`O~7W(*Gk!M&wW|ULlZ-abj7l`hK<0^5hzF(PAkWRXpU-QN9We;`5QqEav6x#N1>1F@ zSd-3sTT8=nvzO=pf@cdWhvA<_1RAbKVQgRw{@R7&Vm@a|J`TpB(jELhxDO)RLXaoB z2VGNR(2oe*x)+BL{LjM;qOhSR0p^|Y=r&J*(qQhPZRXrZsWjwYNWmSx&g}9CBpIY2 zSu+}yGdQb&_uOinqY>;K$8)fwpk*F|`oMU^I40vk@j+af{o9AmjgdIRk8R$YFx-+4Mc6Oy!{I-h{x=$DM}?wKcOS|*SLnbOp2c@7 z5r(f*vG^3%R8Ha?%f=K0?nuHr!xV&-a~?rq5@aG`A$L3l!?=b|D?I@VrzW5@EQK?0 z(~x?JYesh;!lpaN@jHNXu6+3GnZ%DlUp$f@#p4lwU-x7rVxA=D4i+b2;q_+3UTOfX zt%Uu^TIBM3>x-i&xqs+1PJ5T&H|M7;vO0spgksDJt$;*pIiA>^L$1OlIEpp#tjHUv zH@uGgkQ?}!brVK=n=$uw6TVHkiN#-9F?!lnoX@$6?iI~w*xrZ{a$Kvv@G9?ZUqSd+ z?u%$`L|4pZj979Li_$Kk^jbYc2J!5i|LRejcNJ0(8julm3CR(cU_139oTD3|@VgnA zidS)wYmP#mH{dbPVQg4djm@@=s2Feso5L?5RO>2QR@UQ(V;w&AT*2iUo@qRtYhi@* z(B=A>?(1dPF{B*VpOsv(qZMf^Bd0bBj^*im@_k@7VN=C#hN)76;W zavqD{)nL()O5~Jr-P5Kj=ydbkr3KZ{7J7utm8F|D}K){3zoZsM55HSD>60Zr1i=$lrD$F|q7qLS;7SPQR%IoD170aAY5 zhFJMsey?c5$FUE%SMdclKj_AWbAGKUKr z4b6q;eLO4X^aw%sorO?1!9v)*+e)yi93`|BSqbJmm*k0onXr4Hsc`?0uJCZSx{wm8 zBpB%{3SQjTDS1#%kg-q_QnlrUyDloik2@;Dl|72W&kS|J>93y9*r6{dx*7}Gk$S?U z;hMs?r`p0Hc@1Hzy|xf_K~IQiN6IBO&*%iO@IKSjcxa7izwk38%gp3ge{=g`0N_1^0IbLf=_aA+laiK#hU0 z&Pqr4;;t$5+3N_}+j(zgy@BwyKv&pQVIlwVNhF_yIFv zd(S*9mhpyV#B50T%!OFgJiORBi)+Le!ej3O*eLp7rTGH*i!4C>=6M)lN#jInGLVC^RVZ$H}2Fg zz~F_8(Oc_-+l!aM_dg#PtoK6V!=*f{!y6%GKFBwi1UGkA{Ji0Tt0JDe+P^YCcgLY$c82YcNmJUhe>V%2M*b9^yeRXlO#q%WcZd7rz~7mwEY z;QOZK=v>2l&KuXmqjm%GML8c_Vhj2n1mb|_YBX{^l%7Ta)}C99VS*Q;G<|W{dMT>+ z1t7-EANJ9Uv4Ho_*WBI4^~fu+_u*RHP~=Rv7Ty!@T*GsbSHY)u0WR!Wg(`(r5S z#(;GY4d4tuF@Id;$E!GcA+DDB;l}nwFwyYC*`!sxzp{yQmX~7p2nk_+#vmb<`=m8D zhzlhqV#3I8qQcwqKfEXN1s`v`$9(l(Xus)!PjV01g5JVX@df(iU*L`2Q}icwav$Y0 zOg#Jyh0mViba@-%V;?}JvmIUoU*kaVTSTgV!tLUZ2-*7yA#=HZ{9!NJjk+Oi`VJ$P zb|UB8Ygn!5K=zKe$Xeb3Onb+(>$~vkP8Wp0kJvQpE6nD9Lj0+3aJ%*nZ^e7?sPQd^ z4t;@V32$&Oup3GBuP{g99TvB|!PVXuINJIdjk+K3>hC8UUH=8Xw%=jU@SQU;x**!v z4KM!h_xiuZ;G!P9|J(tWo3C(p?R#j(bs@OA3y+0&oT1kRl6r;KMIF%l^AMjFKZ5<= z=Qttz05+c`%nze?j~7&*(4yg7QnfxVZNl&j5Ij z@;C3XD1h$>w!h^1&>q+xd4nw*USY(b9`vq!gO5kM@RJ|kN~A93yM zS6too4IjFG@GQXJn6dl|9tDXCvQz$I^{8LivFQuvocChes87f%<9QgwYj$irQ?ejsb$FZ|m27f1N4aWoR3Jv{-k85xi&O~U$fJfC)P z2Cj;y;q}2RXo(%foxQ0r)8s7BEy?&BlZ*+?+ylaUEQ&MJ@MvKYf1l!UH82j}L*gKJ zBo-s)#o^Qac({6U4yQ~KvgUBV2-lYV-V}nvBO|b`It(W1A)GB0iA6Ia&^$g8xtd|P zRuzGaCt*;Z7>z8QP|O#P;@3+Q_fK$EQcVm7ZcgI4jGT+Kp0ke*a^HF(pJ7gqfb8H< zpdbYIgSp>(8LzpzLSXE~wO)He5g4%#TS}r)G=MX0h9#icoqOJw#K5aM5#x>Hk>8Vu z8oeas^4e;aVJdFgC1Vx;e*e#d==_t8m&FO#6O_vN!7u<4S?BeUSyA=V?iAgB& zPv+-MJjU;d!FJx0Q)`aHAzgmFc#U2*g)`mXrr@V1_X1r>$M2o#pke8-|7d zO2rv}&ZTPQz82mSb3Tv;b;oR!_8)>@!!b+`$VJuY!x(@1B$f{^z@>%)WW3^D(|3n4 z>+(TNdzgvkobCOvBc1!^vr#9PjLCaATXb|P?!M=1ew2(*t5irWN<@-9@2iP%zPNcB zG%qFNCa;^9TBjm)R5H5ea^B5}1SG|AZ`EwxZ!71uw(NdradvjUW*i*3mukh6|8c$f zx^4EujvteNmSmjo=lhVGNt_|I9~D9>&k5q0e1j6X)*}YjzQn`gSQ2J%4%U#I6pV=9 zk11`bkm0X$m^UJh{JUZRP>ayUfOmiCGxt zdI%o*IWTTb#hmOkEDqa`9_@H+n6)2M)l%U7BN;pD;_*o#6&JsxAd1)PZ3=1dIFwebDG%%}lEWDcz-8X!XG?)90v<`C?Lna;+rNB)q6`pcgu;aD<+E>}o?B*HR zJ9GbUhHugBBZ!PXiWMrk*!MLTN)_4Yf1eGdIoU92<&0YELzq5gutZPu$af-FMMVDh zIem)QKoJoM3I5&x^PwVQC*r%_Z^g=>rIt&4H~L!ISX*0kCj=y zT$|4GW+S;DKa6K^CGN#wx4GEXZh|wK(s(348YYcig7v?Zg8o!pVULHG(06&U@MKDW zAjSO~BCtaH2t`P;Ui|QG!=s94xX!Q3qkgBcw>k?MDH?(u?{hu%(iECIoKTX*bB(Gr zv8j0mByzXmYXa8=pH0E=aobSJvw3AJy>T<&2O$;S7;fo`#g}b)Z^Q>VxnD8UriEvJ ze#ZgB_egiRh1@BZ;P52|xs&*H#o4BXYKyS*q%-{F`99l=vwH5$71lqVES$AJimF{X zn6{z{+4>i_kFfwJ*JtC_hm#0xeS&RqFQF`c1G~j9;mOFiJYPzddp_Gpv9gnf!Y_Ugo4Q6)$n|Vfxp!oo3}=4HqvH-R zd+KS74=I450%!gm+={LD_Te$-bgdbh4c*75Frw-N_Hli&+B^>|zv+*?zqr3B#0cBq zgjF;8Xx0v%nbf2PgX9o&_ijX#Y8-lV)iFO)9H(Se`F>ATc>Y{LxZ@@-td_O|)^k5_ zy%~0^j>YJ*#YjvXiBm44an0TnF9x~b=h(yekrj?K$_rwhN*oQXn0}kuf%@ha zCpi8+A2(X_@$p|ZwqH4mblD>~mp2f8XVkb4MGf^9T;sKxXB2o^;%m7*zSeKWimT%h zkgmrW1{TSwt3k<{2qfPiauL-K==Wy$|6Q8#^;pIzL zEUH_J;^8Y0F5BwpGQe znlX?ar+`B{bZ{}5XSXy};cne^=!o;{Yu#4VeP0c?%aQ1qI0>s3siOLtA!5Jv;M;wk zKRudzeeSCWp1s3_KNWl?>~AlW6pR;^&YmMoju}SJ7D-UnQ!)Bu(#Y0a*w0@5tz&I_ zX42Y{5i}@%KiPDO3FoQ@3(}Kc;7qwRIcQta!@S4r-rW6c_2YKd^UQ$8t?goQAG%p= zWIoIE9nYrx+rngJl$la%C`@jf!M$i5&X2##+_y9_=?Y1T-1(02_lvD{eZ`in3}$;K z%Fw8fe_7wp2sW=Zj(whVhSjMkfM_}Z?o~P=h(fl!^}xyF?+i@hCRC<&)(lIr|Gwb z;^z5YQeN?jO_h|OroScZ)dB-{EocsVf6a`kH7sJ|{@Aij`bU{g+-No>YY=;!;m*Df zI74{{#&EXPHBwS;VE4v6W0%a{vw-4m=6O$@_ANCf+0L)*Qa}OwwXu;sy-F04z*eyOZvSs~}FvnCsJtj;#mT-Iph?5Hmh- zS7+Hrrm!Icds*yWbK0gnj%L5v&kmOFWKy958@yDG_Lx0pKIt;_-7}NzaIR#Ft#VAh(-P}Xoiv4M7-RD5JSD0xSV?x1V(FgI0DL*COkIBc z>S_~J_tvVCZ-4I3VEK=El$;s@sLjfzbx1#Q?LrG=u24aO3%rE=|labdV z_q@RfJCRJn$20UiK9&VF#ju%wlbD{zOtQ}Pq&as&XnDX*TzY#E*V6ysXx?f1_H{a~ z2whAIK1Wc?l#lF8MFu@gm4eM+CAi(R!DM@HcKpvF7U`M8=5R*miniT6#~_PmQ!Sz8 zZE7^QJ&wK~PA0o=Hgr~U6KN$m({gEnzGSG=og=yI+TsK3U1Tl$@zI)I$NJH*6;X8k zeg*4q9z?yzVo7f43!2$fOQM+uI98%X@#Ux3^cM}xDS9bYjowXp>RV~?cq7vM8-4X8|~`+X0|)l3&ZhSaeT<#u+&v6sb6HDres z*Ri;T1+4R16cv>Wp(QJ9Xl98w3}1M|_~s($yf#G60R_$_XV^d1jCu%nb$^T~BwEYZA^wEM_Hs-Kljr_-;nJ$CacX7C6) zE&Y{gn9XO36G7ZEgC}c34A(bhnM5-u-fP?PD0-8mWpw zYdh${BR$ATw6H1D4>GU#m+VKh0>&T8qzC&FXlLa?HuUUjHtNb@);~3XLVaTCgTg0r zTdzvO?S9r;V@`TIG-%Z7DP&~$nx%+uqWP*G^v$@RHEkbG%{h(CWkw!ToAi@Ky%C}1 z^QMq*SPeXvY2%317+lUzV>%6!*}jjVEUR=lna0ngMboxX@=Zqb6o=Ce_KWRscuprC zJ|XEs1)eMRjdSB#@$Ake+$a-=T9F#mEA`=PWRD=HNK&7jO3D%cn8<`EmRfayzjv!> z5SdYQa0D&taHSmsEU9Ld9W8(3kDdW4Shh2ZOg4X`Oh?3;?BKr>3ZvX9U|fr?iQ zBTV$7A@&>S(bkVlWMV8!>WE{3Nj&@U>vj75T^r$B>)CrTXI5|R%Ph9~vv*}FY-X@L zogC^)JLwxsY4@kcM=r4KmJe9xKT(?M#OUH)1v+GDNhgZUAbC)g`?&7WvYT2oTVoy7 zZ_uZ?Ll%;BQ3Tagj3Olo z3UB&-DUQ|}AEVd%q_FBm7U_v@W{b4_*qx_GSWV4J`l{|u{c^#y-hBz>TWQwc?}%SIP)~N;9Op>EG1b zl$oSY6SVodu1Atz=|;-E{eV8ddrNjW#Y7!xx<> z?oC-pYxd5d4Kfb&R%Q?VrK!|^a0WfSrb-{~kD&RtoykO^lg{-i;5}ywFI~Bdnr1Gd zHILnB?~!9Dn4F7XnR~E5|AB6;?56_{Pf&c%68f<#oVKn!N3sq{bbCZFeLX#w#?Cs4 zW%DBuKH>nn!c?)nrHdX^@*Z^1EEEh~jl)XKcZ=0pOgN5W%7A$Knho4=q~fdXPx25SndqNk^(xvww&(NIpU4G6uH=sCDreG zG$Qvuw2qX)nicP3BdvTTNrO8C z`srIltJd$N_J|znz5Iz*WUMFmJ##6g?h%~0pO3nQ0sI$$=FO0vUfOinIs znBPuCn)cY5{GQg#0py=f!2H!kq~m>`KJE*lZoMCD*2#IaZ0$vo>)_e!vn+6HF!$D6 zn@8vLZqjY%&lGa@J?my(q~#q*SM^8I>5feHa&93z`DPt$O4XBN-TSQlsoh5PcDq8+v$@yeLlvDn89~}(b4YqxGSQoP^l;Nr+Pdo? z9kJiaYpYo#zjG!{NYkSi@8(juwi);Bs&Rk45~`$oX}*^_0*+e1>beL%hl}8B;vB9U zT0t)}WXV}ljjVJ;C~IR1-Ab8*E60Z+$#6ev(hef&v@Oq8R>k)Rv#`#=jGRe@mN*O} z5my(Q-M)gIaGwS(=cCwn*zeJg^7d4H%PsD|eL*hikB zMO5%X9a8<$(^iyttDwH52j z;rq0BAAbE+CgVl(aXOg#o>YcviA2QZa;jI+XDRYED#yG?ucEgs(ItVY3LH!IfEQvgdj>dd=>^YBRArGlNH;tkq zF42}!CCL7iMdSOA)Smf>D&7o&f4M0(dzo;}{2DBlNXNFvQIwYUo32(KqJK4iaA4qD zY~lXcaop3PO1AiDz6}1#y_CGYl@2#tCjB$bVf?LHG|W&DSqvP*IPi3=v(`Aja2oDmysf}N9wV)Iu6#KtF4 zj9v@5wk48&mo@i$2gA878`TH&FlhWM8sTP#rvc|^Nc3CU`819yJ;%cAxCi#=@C>DT zo{c8)0vp{23iExEsP{n${Ww)fk|u*8y+99nnLLa3cm}EWo}>78L8RWMf<4pJ;luU+ zC7M%_XucHn-6^nbT);Vg!La^gf!U{`A%%SqJrx9{cekkcfg{33sbD>S?uWh!bbi7e z%BdR%QSV{+ByEGQ(p+c6eT1)jdXeAoid5Vb;5R%^9X`3mQS!rV2#|hB->;s3a zNZvypjf_M^Y>st>L8ur`+B~2}IdiU$sDt5~8n~`(#}MvsIP=_)Gm!Xf_25cUGh9tA z4OR4RLNd8nnUTiEcycy+MP=wD-)DVP7uijhx(<`sky7d&%X=@ud_*RZf+^nOco;eq z2diyR?%6v5wsih(SC^oxcTZNjIDWg>Ss;7p6-W)f@ru3me5`- zkK8Te;qX!!Z~xd~?mb&X)to13i%L3k$C^sam%-AsrhCEYUe z=6;6a2u*1H$8h?fJ#NGZn7-*Zl|LGci?6k?Y{o-=zyC&aR*FEpN)np3p)?^#3vSNJ zxc==DrPwUPjxF39!t=SFjmbnxj5jJi2jJ6ybgY-nBPFAFs*S9ny8A<5V#%|;ckagp z70%b6cm)H*x}d4mPu;#fboZCWd*E}5efNdF2g<_X zkQf@`Mj>>ODV&ZkLg>O+xCh!nvcUi$&WGXgp^B{&+sblzos{(I4EwWx5PkBAqCGaZ z7#3x-DPD8g3cs1`e5xqB=QNKsAJ3z=Gj7w%Q=z0kB-!?Hf(m>5-h{DgB{Gh5qJsIO zX;s)Y7IZ9_ZZ`Xq(zFD!xFSul-`=w2*Dcr-$?2@twZV2r$r5%-N{-H49!z2JnzRB2 z?Dn&xtorsqID`!0TsK$nvX40{$uaSEU-nS)F3;C2N6-RA=$*`G^UkF(M>#(x^HZCW zt3}9joh02$oy;6Be7DuoueZJ4B8t@)(^32|78j=&;_pR63_lb_i5tWbxH^qSz9=Hg zU#Hmfn*q#hsUpq3(?WK~%1L^U3Yisn(6U8)XzjfF?7poM+uEARPCtIal+;aVdG$J4 zYk!d%zJFu8{aV>Gae)+unzKpT+)UlkWjoF@lpV3DVT#H5Y;L&%-(wWwFlU9oU)W%K z?pv|#MkO`2S|*vDntzM6U)|2i)uovESXpLNEyB)<2hp_^LrCK4Ev9|rI0b8^QeA@y z&*!jZqiUwHMHAatgGd%*LnWB6^=I3=eRa&rPnll7n?lPpU$Npt*VqFuM>79yz}&Xa zWEC5)*tYb1K=(vi z*$0W+EO~YteNQ14yVF&1?-Pv4*o* zEN=8U+lcUCEdQD&Gn;&g-T(fI=?wahT<=|Cw>KxU5|2N&!T&6%eUBsQ29?u)P9~&j zdXIE2{Ub>iSJtBc+BW1;m+hg+vCP#kpAFNlWZq(>)H&t~DQ!DN*;+GMSkZ5mx#SOf znx;=pn<8oL2yHs{+?eWWL@0pOQ;gXoTcd%d?C-7l?C1Hz?8f96HteVbiQg`@HBO7M zt?@i(8@NT5{`>ob)tp;J(=2isYm%m-q{}RIKsl3Bolcc^G^s(-oBdkQ$;$F1seN`L zTm9cbR`olJg%y{vdtc2+Urm8p$37*U!K1O9>wn8~wUM*_3eAd4qX{12Ofq~ViVX$?T#{HG8<_AFB>% zW}96NS@A##cG;@mcG1d%Y^T^Ow(z_t4a!JKC4cjA!c62ESh{tu>lXt~$vK9TiEs?+L|=r;~{CP8!ga zM*3HmVs!ar6l~#|04-jVoppnMzAn}W$m6YrCAx3RBcW1R$WE3J7WhdBg6;rehLV)< zY>d7zwlbK~M|jh!QxUYQ-~)*~ltaqs&79q2kF0BM7^NG4lXq{TK2BK(nAwXLDq4giu^<~4nz(LpybcQD47sLh z6jr_&OQ8zuX{FCm8hUWBaCz1vm_~d-(sdaO(pJU%#Ur6{D307!3Q2W!JWYQ2gKT%K zgV%glu8kZicxGdRwnaAkI!E9P6_K;>C5P|N%L0BmD9+}5gg#w8Zxgz*@ zX9zy3h(n5V7Egyb<4DUldU!}4ntL@->B$+T-xBffOFbM5!eNnr2-*^d;J?cfw%n63=p^@^f5m%IEg@S~K{z?1n+Djtpxg&DxJG9^X-TD$_4y$Bv-%eG zrCg+Y#!u;FY!;R8zeS%TZ<3ByG*vW(lGd7P8W%Met)o|Ag+Txo*S1kD=K)AZXkpY# z8)%&#iYcFcxd$x;jT({I?2(I=S9U-+5DW3loe=46rPF8b(edA-k)#qx*EUb1vF_QF zcw-(dir-7$^3$nPMh}Y)^32&pFFd<7gG8Q1k^G@lDln@?)UYzFOshro;V61J^&E-u zdhc8C0Kr62S6F5}NVwM?OY=|0P~x;!IyGe;zPOCX*W2?kd=t-CDmldcM3tEEr;@5F z?$f5>qL{0r0Au&@XdUi=@GrMXW1=LwwyR=}^FFk<)WA8e9YrsX;P}}Oi0cy*_En_d z8Q15{w{}CD<6LrjZ9yHEC23h?A<6z)L$4E;)AjLMLcf)^km&iEXFE$^Y5QHWPkclN zXAH%-{nF?z(SeDmKa@93heog-_Wji(Q)46A?emP~Pl-iALO6DFox+Ae!*Eqi2Iu6u zD80lJ5B{^ow5qY#XR-u=L0&Kj91WXtMqk9m>H9`aaubz-xvT`1bPmK_lcTiEHj29A z7gL1N4$MhdgzxeK{yiB>4&{cFzl!McCOs^vR>YZNb*Qp-a@49Ix$-SUqAs*EN0at+ zePIPBEFpcJ`?ob7lQ!2Bm~vmilR_WNFmxopJ|*hiI)KjZF(m)Ee#R-Wtm}(l8nGnH$Oc6y$&;rd+Fk> z3#1}`f}ZnrMe^VMdKO>k_A4J)@HwN$DDImu*GAMib*_V##^oq;+P7;MCDv%slZI;A zV!n^``CeNx$_+35rohV+m^GAps}xQm?bSgTJRV7$s7$R(WGGTvAG!WA==5kI9jOt> zP*;MBkvaDWmQ#(wX7Y3&MAdR`@KtreL)~F;c(M|%LF3Vt;fL+6X%rkeh2)}@$=H4i zik~?^l6yMU>L-)WJa-y?iPs^^M(`|TJ;aY0j5}M)$R#Y4)H7dRvRhc0ScmciX?0?!q+!Kul!q4kRA@*OrtrDha4-U+5@$7AUDt|hR! z%^45&W8vvxM(=Ox(Z1~;*$`nwJ*he{{c4Fzj(Qm7 zB8v&<21Bt`1m`b>QDadQ{ZKL|=hYrui{Ondx#OVvcr~id`odFTBpS2CDM#rAbBXC= zYcERSX_6dV`ToK8ryC{xnMLz@B?z@4bR#o@e3lzinZ31l@+7)Acf&B04u+o=?rp{~Hs>=6*v!_F9&0v0v z3^0CB2=#>RC0p}VBvLOyZzf5TD4$86>hnSCG7oI^G=TW>!_@a+1-+J^OX8(lu~2IS z26Nu-Y0C>_*jP#`MQez6ZIE0n4%_RWs3fnH(kkB2+dG3WU{gAMuE`;vXLYniQkE*s zrD;=M94(x#h{Ns+@WOTk*78Zg__8%n*>wmLJ_oTt=Q!52sDatETBAvYaJl_I?EaQT zJzt*E@LeN$X8A`d^}kJurJV6U<1_pFmCxuV8c;;(AlkcNF!kDm(Y+P6w8k}L_zN9g$@>!v*TI{!W41KShTn2!^#7Ve&-GJj z-<(zo>;6q`Jp!`%O!mK?^Q82uhK}*t-eJNdYVV&zt%oPkY*!V|u+qe*!XD~vm%@Kb zeR2H-@2ALoWRv7CGp{&NI{!}{56Y&Yb>#?5c^8ZJZ}r?Gbr@%7_c5&}ub6Va4CQt> z()J;1$h9klT$QSrm*Z>J=F!6r9gU$c*ACI74Yz31sV)3o%`^Q@&BPJkY^t17M`xc{ z<3PXys@{8_Ow^~+hU}9xw6v41FZoP$igJ{EYcaJPKS^=XL1ZkeNxK3L(8quvOpyrY zj3=&t>>Wp1Cfa1F<4${|hheskCp_FNU|;i@RJRX-?Ut6;_L#{1;+5^0Q_4A<#D zAz9aY_N4Y0+gsSqG_3BjWBW#t!z>Hhp&!IPT)EB8dzCOR(Tj9S^D)hC`9Bn$c{r6{ z7lzF;4`nPFizu_R*V;%Sgha@YXcU!35-LfehzvM? z0nROh8nZM&X;b8EkipXeTXFT3g=iJD95=1dL*H^4%$uczh8i6F$2mi$H}#OoLwsNt zbrw{5KR|!`9yDToY=xju=7NhsuI3m<-(C$`{M=Y{yoLzDeez@75Avjy3pQxHC*SrD z5Q{aMV3-sNVT}ZW(%zG)D}P9)rayRo$%F&HFM{3H5y)qmn4Loru$`UJ@e5hOe*eR; zzpxQ_YGtu(i6+jE62Pqj2_QMU2)G5_f=c!EgqW9{lvP65~t{^U%&9X(~7}i#o25qyy0}3H>Jm z35kMWwtglSc*~&<>s+|9S?yBBakydTf_$v6UgBMW3vYhF!wdGoptUb_IIw$^{eN)v za|?_rvTM3yE4-PE1y18x7&H9?@osexZ~X`4I4_CFPbaYd;R8aaDS8;|;~l8SA?_Xs z?i>c&9ekM9vKxE;MdHQw5PUg(7WD2jgyHu#pnJO(ZSJ<>w$9gB`BNR|sLe&5&7oLV zEQUhe2XXQDV5Huz7=igVr_6wFExyKMSifZ^N1Uw_vjB zF`Qbp4!0Dg;KGHaxb*HcY);X`3=PJdm^l+Fzs!Uvqcxyp*#@~OtuW&~A0{%chDB>I z8uG2gS2}ax!NEy#>#hqdH&(--!ZBDaJRM(-aly)i5@2&x5`Mm(gPTW9(7;I>KaaPe z5bG!(ol}AJpSwVw_Y4%U*_-^x7JMKXjO{I9s9MOlUajUhcESpCVt>Qts_A%YLJ<$u z3`0)+bI8nj2W?vE=*km|U;3|Mo@@%tEu4h>h(a);3(>9I4o50j59xRqM7(VV&e#CB zAD6+Q+*!D4K0lV$&%~;C0=O@LgQA;xu$c25{Fr}rsgNqpcH?K~icYBDFb1lLLMSzV z5Z+&WL@d5?!+_Oyahxj5$dFTXRhFTzZxEIV_2Vh`T873s<;mzu3oHuqF_vGEh+MIIKu^z_H zs|L~Uz$QC_XrS7dUd)*%|&U{4rA)6fFM`?JO~nm}DK)gb0jc zS@F0d>|STN3|p^n!YTI<<}!}K#oC8Z&`=aw4oShzIz2cP6bM;H2H@2m1x2x?Fw}ez z@-uotk6Q;`-_n9z>Zq{|%PagO0P{VsdJK&#~Bc%7og1wI%{-NyaG6zt5 zQVTwfh(Or7*^qsG7c?EX3f)P)pcrk6AHL7SlHyC~7#9k*ox9*Pn@cuzJ%s~{?}O`L z5s0$cSf}m=#-Uh;bG9=-=}Tf)ZPxzyeqj_H zytRrN2WeB&`x0~~V*y#t{ElCn2;KR`knZ2i*rut>wH^|Uf1+aXv`q~L-8zFS6i%W{ z-95Z0`v7@#&R{~N6x9+_pi z9tk0|h;fYs>LRIqrag`6TtL^J4#%~}J#f@@6RsO|qfg6RXxo$$)zU4boJ|FE#36}B z6f&1&LlpAA@y9FMC(vx91-DLK!-hR!IF;#(Kh14WknbZt`}hlwYW_gc+(cSmd6E{C z#!^vBpgo(F>DmPvbn|sCT4c;eE!$q#+e1d0+%5Z#b zIxd}i1~UgDXqmufI=IoB+If|u)Q2;y8y1gJw_<4X8V5S0wUCCdn@*FMqwnIhA80lA z6P7D{LGRs+Cu8**kF?)M&o%X^Z*h=b(@UiJEt@D8>x$iaKY;BWk1_sGDY_RGGfzbt zc1tiP`IH+P3?TN}3sFPyX;fp^0B+FS!F&r*cwKl4KJ7n2cPkvFT=xTMgkb}6$F$P1Y=+e(kPQ@{c2PM?OYWlUklO*-j}u}chb)#7xNtb;Y(D-NTAHsEAFMeb5J z=73#JGkz%2=jmK@5%N((?UyL)*n)+#((&!FGsxRqiQT`i;Gv{qjGxX$FEhtsA zni?TT87zPfSZ}3YZ8k7=uMJ%kGMDmoh|&x1U0COP1@4 zbQa~hGL5Eee~G5|7vb6GHW*aliiTXX=nv4Ocfat{FMOx4{AD>dX~d$c)lN!(`O^8d z3+a#$V|eSiU~;P+F1=BQwtnZZPU$)>M>G0og3wiQ^XPS-EOg(;`er|pvGiaZuF7WK zr48A5D{d8yvI(W5zIIgQXaWszJV6sVF0{_BhVH(aN_BRp(7&=(*i7zXw~-9xziynL0T3Q3K#>UK$Gzo zzTq#e-#8&6M2}d#hke(~ap+Dca|Q5I@r*Vs>GP+5I)B2q@oAVUXTkPzKcTr;6c>k{KCUS_cW`~CgE6g5bZ|iP7ZYs zr~w1Fr|{N89}CmfQHmI&@7z=jGkp(D9`9g@wjjRbehd@0C2-Wx3&ACsbvm2zo6a=4 zqaX!4G6ZRCct2ZPeTFHInRqYG4|kuKh6>)RQ8>p7mFHMtT|09IsjSDS`iF2mX$=1Q zD)pAV{UjfYY$RkZdxhALtCw1gZtPQFm{VYq2=?iZI3bbiX1`rhp(YA z;3X(;`3`Bn%u)AIG|JbW#e`e8p|SHK1al8V&jA&BrDZnNOx#A(4ZWe%tPLu>jL|TO z3yq#fW380|Dh{&w<7#>Qn<0Q-Hf%$lBjuPUEkZX={$rUuZ7TYoCY>xpStz&*4Ue>>^1nBw*eRg8#7m$d8XWG%T zkkH~)k*GRr1tynWM7#7!xGDJ)X5ZO}6Ilek)6~#daUse){Q=TBvv6XGHCkM#NAuNB zaNmg;bfwxyFuB9#2jly(-{mC)s|Vte?l8>vzXKc2)WZul6PwdH2mLmeV`f(p;_U_4 zR+5TobH}hi^&P$yQl#J2#p%&2&%tnnA3ZCD@sm0?x=(Y!Pn{;Xm7~r0Z_1eL;(>l$ zcj3lL<}jLJhJ1gY!<{i()>n-};qv3`I%IsiIyNimRYU#An>h8o3b)&pz`~mb7<$Z{ zb^5=-+s6XfofV5W)R&@W&10-)j49_l8?;~RglVh%QM2ta^aKxrDxHSKnaug-)`+E+ zsx;+SH5|LnjR%7CFxhzuCha3oqv{@hALB;nbCMY1?})E{9>Za;_ZY>Iq;;Gw7?GoADXEy!Re3mQ2T@nRt5hDtE9idVzKS0DO-zk zgjJ}$zcxL(zXkm`@3Bi&mu}(@pvF%Q)8|^5bW|*loqfB)cB6K9@-v=frEVq?Bi=CT z@Q$;i>OOhMvdpcYt+DrnDQulOjZO|FM1-dU&Te6S&MF6zw>p4K2E~(&4MV^)!;^WG zbMRh%9>+n_kf`acgtAXNh?dAFvdI1|>{Rh2v4V+Y>%JL~JH@&t!%1X%MID)&@|{@R zj3N!a-$`D(4o6?ro=irF!pXTyVCu60Bzh}D#}FSGUwNN=kNpChZ6sjD#fz|gtrEKM zMv>O$gFwH_V?*&nGA-aB6!R2-zg8^SuyzLUyjzXWVv{)fYAZ5&$D zTR2%@12)Y|;jEQB{0T58FG`$Y$$2r1FgeJH>XRaO;|2*{_)G4*dQQ4bmqJP92jVj_ z6UY@Sh>A^sJq8Q0a?uWSEhr{$n>>l*i4rn6V2LfeZozH7DOhFmi1X%X2H~9KLq6|L zGF3O5BxckT{kQSt*um4JVC_>Pm#9IaL=TXEaDfQUK1}4FoglT-L*PQ$Fd0iWf>gUu ze4oIBV;S*~xXuXsU-7ZM?M(c%O#p+xE+Yy>AIUAN98ifC2j0LrFxr{{ndJaG9$zIV zqz^*KV{@{8<2xc0@&;}+Tp-Je4v}Y0kI7RZDct*soqt3t;3Jj2oYO~UlQ}J0;Koz| zd3iUCTsnK3^z$wya|#;B;o~uob!RCFczBB37v2X#M|H?jRde#S8h}S%oS3A<5l)0X z+*#m5y1%#*qnr8UR^=u#_{th?&Upf{0sbVi(VHZgJSM*T&q3hLd3e+-6PqmgI1j`3 zbNt0sK+B$%t!f`K>z5W*oa%(1QEt-gd zeID?KX~1vItuP?B1tlw@IT@cSd*RBx53owj0cC$i6P3Rvr1tY!B6LX(gdZ1>FwHZ> z&}%N)0;c4>PYpSvc9}e6-Kyt>Jn-aHJSW6t2brEu;B4_iVrn~^Snkz@dE?Wd*XcNU zRUhd}sZGLaLrXV*$R54l69H6a;+ zn&j4{S`r-W0_FrH4CkM*IuVQ8dDW9v1@QcZZoTc!- z$`XzW-fp<~w$+<-W@MGr|h;3{q>+Xocz>7R$m1fV;>&xe8huA<%)CTfj?KJojnhe(q zPjWtaEh5&7{Gqr$hpe(CaG@az680By{1Pvc#%^o)lfVTo+;@q8>3xW6Q6Qd)TtxFS zJA?VWhcnBEhs?fw1Z*^uIPPzraa;~ugI_6^$-L?VV7BKCsXUUz>5VrgizlMUfjMrR z&uVu#X|_+ux8PyU*)#KrShEz&r?H$r*>^Zk4kp5v3rnC*&IleY3I^eTTr&2~5DZ&C z!QAC_WTgi$Y-M@cVf$Rp^za#Eho=RMr}+}0VE}DcmYvDyAi_zT$bt2uuqU~L=sj|R zAU@#~u)`dH9 zw(RmGk6-M7X0b%_{D~}V7pfv=vc()H6(Ckgejq-}k3*_v6PI>Ru=sBy$z1xBEYQ6S z3YjUS`XPHR6g3mkz&9pEH`j6o>NZ2M)O#YjfFE4NT}aaNxtvWp-#L?iY~lXiK2FtN zJ0f{N8La#ka-viim4>SE??g1gF+T{vK&!a#)FN~aW=p^Bl z`mj_djuYCpj6~%xf^R3@a7g7lPNNh*@(#t5M?-(e)Y5J^!Z}WqP1GPtVIvtFKgNlQ zd&}81v4@@aUL|>~TZl87U_n|e=?&imUe49-rUJ-KTgDL?Y2=to zWs{T@Z%lk$9FYiaPB%Gn5mOv<_Bm`EoK1!k ze-oZ=Vb)t1COl`|K{(?gtot+#`c}lis*5IYDkqJY6^ny3I}_ZgS3};9$Uw)t9b|I< zG0wNeqQt&xE_4qJlCyJoV8QEnC^{zt8%}FOTE=d2D~)hYad&V|JBmTI=w;4pg=s|Q z$tB>goJIWluaf5{1R%No0*8+Al0A>apt0PLm~NMWNz+Zxa*!Y94|#*pMV!amt{XAf@HTVLjiGLAz$;6N;w`dUieTrx< z)Z$EJ-sBVwN7s6+ZlaD@z&}pGc45DN?CsYLpX#WJ0IgLbpErQRb z0$ASYL$v%#h)x@0pz}qNkGEIALnAj*|JjG6gd?eIWI4saULvM@lYNFVkhyCqT!`KU zCfgM7?4}`Tnp+3vHe9f9kpn!cuYeEU8btMx5)oQImt3%tAX8I{L?W)2Q{|@tC+`Qq zd}(t?8aF_PtyZ}0Im;~^7bENAEXnB<590N)jx_VslX?HDNoVj^^5*w!=!{c_s_Apl zwp1JM9+zGUS>=1QfI7gVmX(2!l)7KjZT^&7$E3_U(TE2gy%F8eY=9Q zK6IRudAkVqzYK%5Jsq$qAQoy)uzpQSD)`o3<^1=fileN1jT6Pr6!{$H;#zhl^!T0< z>3;T>GqBT~tm!>MGzwP}zdN~PLiQ0SRbe){^InRaKUL4NBCYsR>>Li}E+EnZp=8r6 zD{}mx3#p4VCZ8h{$oV&ooQ9;O~I3SHdT9h6EhxBJQ8wbEMYtk}VcMo^QNL5(~45+m0k+ zIjMw&VTbWQX+OMS-9dgWFDBfYmBfp`gCm>&g`-;}LcCbc$jZYQ`KQY;ciU`Y7lbVD zxtqApY9gPLB%yDc9%OVMM+=E{DEH_by3CA(+tZWb#aJXfSo(|;@#ZgQ^*e52#Y2dr zuqIin2qbI0Em7VkOSE*HiTt|+vSfDzF?`TP!~~^C=?^!uw=bQ{a!3Gnf(uc3(a@q7 zLt-pL$Zp%6u9?#GW8Gt$5*$(kIec`oK;8&Tov zB|-mB1~gj~ z!_xvobQTT5v~I>t8!Y2E{k*`D(G(|_FRBn%#l^&UaWJuB-Ds{^gQR2X5qTI~Migh~ zksD*T$hI_D^6e)tv5r_q+`oo`$iy&+JLbconF*v&_%M;nbt9L~Nt5f|Mug{~H2L|e z4jgr#!si40xU#nj_^Soq z%<{*!fBsn7ZBH()rex!jj~rN42Kj4pz*XxiNLqBl%hW;GcElMA?tdnGZ{H&i5Bw%3 zBUzx7djve{qCviAj8k{x1;^^kJTfF7!R}|mket>;REolh@w+4frZQk3-wKiQ)v@RU z^PPm2lE~_6a)ISY&d)we#Kv!th@f7w{_i!idyW8%oSF;Y5-vm7tGB>^_ye>e5BS?I zfEgY z^?P7z!Uw4A%7%YZ=K)qs!hY64H(H?tIwdzC^s*A1XxC)V9$nT4>;XNSzp#DRFL>~d z^?7U_Kt7)$%8@pya65Pg+&>irxmE=Pj2;lZ&i}7{TFKS}Gr_t~ z8JbF?NyB{&Nc^=O)DBZn>5{|Rav^--Ee!)rsu2Hs2|NuShJtkku)mjeHQK8|CSC|V z+b4mq;0G}u_JdNZoiGV^U_iG4rv2*%m75lD+e{Lif|Ovn{&n)tV>&GDSA>trPe_mV zIk>J(@xqc~IKsN1E0@f{gW2oJ`-k_*tbhsPzO0IT)e?eLtwEr3NeBKcp9`Z=%IxP_ zM>x(&q}? zv*(5@|A|1idKX#l{)5~sy+EwfCyCq0A&BUHLbA*h;qQenfUm4 zX9pk5S;q?_PW*61N*Or5>p&~f4~ELlL3#gqxTeGp7xtHtTdxJ-NWnd_sQwwr+!YM( zOxA<7>RFh!@C{tHRE87T)1lya1>82zBwlY1li6&4+2x!74k}k56mP=oxcfx8O9?bA zg~2m>5mX+RffxC#tH8Md{kHRQ9nWvbQmcU>$2;&;uNta1-UP9db8yA`KbR2MP72EF z$P;cwm@je!W}PU2Wj6(xFTRKL({rHbI0jF963EOW!z6i5G29N7g?E?V6O~DIFkjhE zdU^=l{;v$aSB{eVHb)^xmN}6I+1|ErHSFEl13ps^NjS^7n9mFV-L1zU{z^1hgvNtr zZYNaPra-@WI?NN&fVAt<(7tK~w8(!XAIdq9u`&?E&fJDCsR>XiI17hZ2hRThA5PXu z<2qY+poyEI>!2OnR((d2W?v&J>+g|q<2Dk(`Tx#ZbbPLO#?t6IroYkjy$sntQH5;oVK3=Ewu* zSD8ZYEjBZ8oDb?wqR?kB3)=Kn6PMvg@>{u*814kvY%v3Ndz^p`*RDa&o@V$gJPds? z)?`xzH!-dWBeqAM!lqUekO{4a8q$g@hO75Ocs}6s{hPwA?U-bhEss~X!s11!qpW%iY<`H+TMVV9mxS_8S z6PQyYYt#eVzeJ#3xIJ~!TS;B?b!pL^02Exl7kL?XD_his?mRS$5(QQ2;$6%bfQQQ7tYp0IbglOslN{5Qs253kVibT#W^c4B_8Dry-oW;jS)jEj~OcG7SpdbeI>EDe)>gVxFU?>hb^x_|f8~CHR z6u;Wn;aybw`T^hcwxI1&1Js|?zqy(x)Bz z3Y4cpkAAskiGs(BS+CO;74{s&APB+I)YT|^OOmes^&KOW#HsttC~S4viGA$-o!Aw~ z{1Em?@7d#$Uuv{@s{?J^;z;{yIn-{kDUBO4r+J)IdVX&#UA5Gf`kgL?qG4gYralR) zT`$81n;bZ>dLN`WJHTQ06;LL87_M>{Q#2|W-vy+h&FCt|?hM9T!?S4Xbbd(IT@3N_ zb)e6x2NsLU;^K!Axc|#%FmdaKzHue25$8s2E)k3m5@n99W_*M^#AAf?&@$z`}iZ7Zo zF6fH&EfBV`7&aFUgORs4o*0P2EOT#syX7#T%rcmDRuVThR6)nQ55N`s9A^4mf;{Q9 zaNdn&c;@XvW8KTlvz>t-5;8&b`c7zE*$x)B2B1pZ6s?5TVPa$v0PA#1lAmA@_!0J8 zD1v9Eyco_o0u@|bxaHnmaCZyD5|16IKM{xSv$H@hxef-06~WiYA3nrv2a|RuFny;1 z=Qhj*&xcFkXu}!U0L74bob`rkOrgOl2zK4w0Z}u}adhD}>~yckw730WYMKe$t=~Xv z(i-Oc426qVQs73xN#K%*2H%}$A?EFK5Is(DM#xN*uK5ZHJG4+sX*bIIufPL?XYiRD zyDsj}$6sUH(Rs}wT;IVBXZTx4$u%vQLn7he>=ZaUvIkbxS%UeK-^6*L27LN64GsNe z@bbcVRK1k}A0Iq~Cyy?|lIC47ekqgj3G(1VY!+;=kArP5Z-d)mHP}Rt!@Fw{uscZ% zbxN;*t&KcdewmM&_5-j|*a>Hp$>QbkNc?K5j7eEOVDZi!3V*5rZV${mOk znU6nfKEihH5!fLkijEU`Ak=ac0`0}nMJWmb+b_Xj$SEjwGlH|nO6DfyNaa_n3d4y&7L@}1}U~>9SFotsyR{59U-Lfj2Ke`5mT@Ij+VIX$h zlfnJm0$A^;jg}e8)R(^tdrx;`n$I#i;AcXEciK}9>s6O6(ZH_hv(b^cf)skcpu@W_ z*xOlw0Uf82DqKV3NB1$qw+QVR3p(+$Eb%wZbCu8}dN&NdjiAEk)qXV7HtH4;h`a)On zp>89F+1Fs+*Fv=CO2W)s=Cr&Ojp-)gm>GHuYxaerL8=Wpd|8Acmo{Regdlw-uR#~E z@1^?J4v(EsMd{VcaIm-+H7B?z|Bb)6DLev~Jzay>&G+Lb(ftT@+fc(T0i{-i;c@0% zG3idicNNFjediqd{C5pMU-QOUn8f-;F&L7oiN&$VW{wfq;x3MFr3A6%AwM2B-Ge!0 z$>_242ztHy0oE5}u;9fsb^^p0+qToFma!U@5Rb;5AKP%Pmn-uBpy)Pj7B=m)MB#Zl zDE50Ho>KC}0HI=(y%B+n|AyjAenXmhh^=RR-=W7Rae63w2Ho0$G;iA=-u@s!ITzmJ zYR+3QFdhcqs()}lUK=kC*kh`75U$r5hdHyT3r?)f-ST_yjKF zD#jr$U5x26K(BL|77I@>k8$S5&K`hIm2cu%mk9GEI zFnVMadg%C|ZJ{kLZVtg>+v9k&h^l)w`29ES(uss4BG6> z@X3P`9DRBeYac|QX8jHfHPb}F&kh*$-#L7;BNngnC*tJ~vrzLI7dBfmPRZ{Yw%7dt z28kj#U-BXDOS{1Ox!;*nd>X#kr-fWe%W>HXNj%igh0n~m@MK0acBc!`o|8kkLR*yb z)NoUm-kG%ZpfpW))uL5BI`n;>Ioz174C8vU;EMGe=9ra2EylBvtnY)ir=4)&Ry&xq z&ZqCvROs#8*;I0Y0)2jaI=!PSOE32ls8pUJ3k0NKoZlVfb}oZC!fYmGvIW<(&hJUq z$Kuy{ixU_Avi!w6oRE-W-%}96cK;)JJ?`LkV;>YJu7~f{$H0ZH)8cHS;luJXD6*^w zXRyD`Z|s~=NNbd2%Bg@|KidPiGRNg!UAm$A2gy(rgJbs<*ll7pZudYupu7jiAB%u@ z-gMyW69J7WBf2MFk0zhDqW%l=FiEWpPcT=O$o3IjwV?si?FU(xR}y$sj39aO0+0;7 ziDymYarjpYuCr9dNgg3QDAbb}Ph6fx4JW|5;PI2m(m-HN8TiwE;;CEPd;UAeTtp&;cys+o4E>dyk znpBZP%|UVIYc#;6zjQEly8^r@*bP~(+u(u3Wq9!=1KtR=K=+4Xj4a`!o6Dr>yfA0T zk#Plm{T=XV_64{ey#!=pcR~&0>WBPy9EDaM!eu*;KwOnGM2jwgXcfkVtx&{QYh2L& z8xO56ts#wz5}&|>togXY!yS&*O2QEvKX|-; zFP!cT1DiMNAgyB=?4o<&@1GEm*EhgSmR}ND1jxCd0iBt%LHaW53Y%X*50-Tm^2*2a zGj5^Mk<+-_pa5@9=s<>*IylUcg6huGD0liQ#z*F1jIj{&&-});93gr(%?O)@Es!hJ z0^59L;fk07d>dl3I{73xR%j1>e(T{5JA0j#x*V)eu7#$Se)6^dF_GdOBLkb}u%3l1 zs+gL4?~O^v#RKCu>6SQr9v(=N`m*zf17io@qEQ zPZ5LT#IXJ7JPAgJ*LIX0KQ{sS&h_-+Mf4)VcKO*Ig_Ck|feGr@P=bl5Vl z5`8N(@!z>JOl%8-qb|IRAZJ@|XoFD#nZ zfuC0kP$7d)I2J2T#oxd33IxDq6?2QNT`mPjy$bNw@ zJ%;f2%g5M0BO9h7yWpBn8)&f3>?5sKs6R5rX4mW2j4g20on07pED_e-ErMKGF-)n6h4uP(VQv3c z7~66PO>UfIl$sB?^*EzC!dU5-2|pRnlR09Jidz~w9SafWLY*131WrwL`u zZ}CUfx^|e^C4mbpOfhUtAviGJtEGJ-YfiSJU?y`0a$aIjfg6@p?!xTyGuX3uHB2@( zg0sX~SOu$5-_#A)9lVIqMhmE(tsr&&YslE*(To<62N}F|aOo~DYFL=yA`45@x@ryK z=7-?k)o>V3Is}RBS+LJ;2<{oO%vPl_M*U90@^fOiFFlaC9S)*W_!f+*9D<(G8Myg) z7F=ZWbN4oWynW{{>@B?k9hDkrmK+1qzm7oPoKf(8VTzW`*U%?69dCWU0&-8!z{ytD z+qoNouN_mcAeH5y9Gq~;))?%*AC1L=3OF`rAu2tKz-mP|2%A|01$zB(Q@0<#cC+_g zB1$7Kdqh+}@XP8T zTwYZIwE^B(!k>Ub%*FB8a0;Ng1N;Y-u(Ea^Tw?!%8XG`-aRJ0MUWW1GKVSyS(rZic z&?ChiC^7LJGo-uFuuzqD&YXjCWr9g$)847#B&gh;slZGM~a%wIPsgmO?JZ z-0eD70AXu7;Z?XZwB@IP*~W6{cif93>~|nGa0cg&v)4`6z^g(bxSQJw1M-gIyXsSD zd^;DCSZ=1ku@OolnqdA<9rS&zkB!@R;*cflA2dG1xDPrg7y?*#ED%#Rg}^WA8}Oa& z?}9|H0zdCD5cD2{6t>3`QyBzNxd~Wnau|%3T!Q+yY!7Jv2a+`mu`qKkCi_mq6gPEz zzIhq`s?fyv8EzPEwH_Y_iDSz?#_{?Rgk{8u{ojQ+8nO&;UP^#X=W{^g{UD_1+=7?& z>@}}HgL90f$>q2Q67tVOeA*{C@Mk_QU@k)CvzibWc9=0D%E9SpF4`qBp8xwajF7$r z{ZCWjV?-W&94|(Rm;d4Azkl$6b_}R^CBTY`c9_pH?vJ7ru>8IQey{xpSq}_QrDiLt z2UWnu@Ymo!^$2uFILP&E9{NA!V46Y*HoZzgVdaZ>{Lv4vS>lNkqkHlGX&sdDTZZ*b zm+*!s>sfEw%eo;Ou_R;-a-JCAI}dYoGvq-BdtS7cD*?;X7m(Yk6Qg(RL^KeizhahB z#pUbhU&5FYHJ-5dej5DE;zzp(FMQ9pA3v2I#H~#+7=54#>zJD_#_AL*tTDvAhJB3D zz5=`iS+*ry1aG^_;kjsTwni0TE)+5R7ATEGH3Ts$3Zq@Z&|SzLj5b_?cj9rtb9ewd zFa1XK9 zBOX=eUPkH5*KoUT1qO{@MoHT?yz*`<{>V6p)zhQ!ka;$K$8*?Fl7^8@f;f|(4;8cq zp~d_L%r{N|nUK$rpKuWy7FeSkyGKo%>xpfbHSx_cB^>IGgN^qcVSVu-DEy^``%^dJ z#*bTZMKW8HxJlC=>cUhy)g2yfiic`XJ1|(sg&u4Vko`a(PqhR>)9W=bb$b(p^;ls3 zrQImQ`Xml1EqFL$2Cd!t9aqcA;_uDeDDS)$m->iS0JRcuCq$2ItK@f0lNOvf9$?}MYv613iVeAUaHCEdaBJ$q-gCh{d8; z@WGm9wDVGAf4j11KFs(D9oew$cP#K{=E3VVH{fCBFr>XZ3yWeCv1H8~Y_`io+XuO9 ze(HpJx|eWg>ve>1F{;7lsQefCDc_?H`1|00=oG^;NLy^7)c+Ki`KaKw#!=7=D+2K< z4|ulA7oN8`K^J4{dHfDU>Cabw8Mf zCxT0E2L8y8L>=bYH5YBgpu6oDf0A`u3+7{gcM0A!jKM<}EkQXw7LL}3LEi3X@X5tJMM&K7-K3wP48a2?^t@&zz=$_e_e!F&TljNr|J zblqL>VnP)*2yDZswq5wA(gmX=Ljm8~!}*KWkiEebE4GGUcvU#6GWO!bq7Il_@CMo) z({N=e^E`ct!bTMbC}(G;!R>(%vqlbEu1lbfsVrWcpMz?}Y1l8GjSm{dS&!@)V@gS& zsorlE%v8gDXIQu3&5eqkn59skHhK&Tqbt*1c_*6?*{|stH5j zz$)l|dWfBsJFtn}IimAxge;IKCVw}5BT6dexJ&B@uGpQ7U#F|VFJcNM{t6&^g)vf# zl1bu)G!mgdPPAy9f$20TD?4vixr~D7#{aO&cZn9a`yDTCxbd@+3E+&zf z2C0`!Kyn#{N4Nln7P`Q7VM%yg9}I=DY(0Is6H-{k7JyD8W8~G^l_1-<7yP)Ekb9~7h{-c|u$?Cb&)02$^X<>cn|*u8 zgRo>0Ty01KC11ff%tgUJY@M6?p1j$eOIn*MNv0MLv_CydKI`R>o%Y{|x~2w54_ZPW z+ox$e{DdDnym0ZqJhH(#i!rsM$gwFs_^7l3PDk#6>~+7uAty448{pL(5rBRtn1W)5+OM_yYMW24T#2{HYd^BJd8vKb(7jeW3q2QFI*fK zhApYB#M0*)*(9?N-U?Zhy|!}Da8d!hdin8}iV_?uRD*_!#c(nJp~{mNjyDTI$M17ZX~=1 zuaa9HeDHa>3*j9rA_kw1k`Hy~N&B%3($_EtZZ*t+9wiw#yWuAU{Cf(+9<3nBy1o7d z$|x^216N276EP7T;QOoyTV5!U%&us%vMP~qSu2okabt3NNgZd$-$AlvR~VesVI1Kf ziDXWzKZz9yCTBn85WX4r$eBCE#N0X>%8fRI##bMfQE&w5dl6vF=D`vZKEM-W40vW8QwjEh*(_a+}Hk0Vz*b5xR^nr>uC(i?)-2p-x$)c5=R*VmBJ!8~Hl_nsIh-X;gf zz7f}Lo#f=eV-mbu1a?dE!@Ug%oi;)s-|&SrT={>tWQwTItRV4c+(7?kJ_s-3f~7S{ zAUC|8u|0PHm?{95p9SpU-w2P2-;%)1jbMKu8Kk`0A-8JO>g}7vb9}&ZrHwKxb-*wYPKc(aRq+t62h<_g$eX`4n<-xJ*0jLP?%YCDR#FlpcK$;f*ITbY>yu$qz*8 zE=g?T-5YM^pzE^we16lzzB>*=HJztRoNu~a5ET&J0!y`cs8zmaqHmksIpohX zFV`FjXyx&V6jdh%AKQU^&yz+ErwU6Q`9N8mcCVb_US5_1h`CT?LVunkaKVNPaI}@v4h6)215Xr1Ee~%IYV_rWQK=;R8+S3L&?~ z6k7VR9x3|8*yEgoU+ucEUD!eM)jp72SQTBF7EPU%6ZsxajIJH}%BE{7(4s(bY&IE; zv9sngV8TX__v$nq@4 zIl8VVqWy^(<%4Fl25hriTitwBwW)_y`EAh;v1aI=@RCw-=vY!s#?6FZk6Ma<37Qd=x%bPEe--Ur# z=eiG_XWQB11=4i0LXA!gzCexLr|J0gWEy6@iGEhb(%!_ebnd4TbEzd(d1MzGr*@m^ zIh0dzXeDU{?xb-0`ILBeJFOnLl%@?1qJAe=G89Q6SErercQ&2OMKnkV8$`7)vZ-K0 z82y!wrmP_To<4Al`D7hn)p2zsWtd8Jb;%U6<1}R`s<5c_vze>tEtc4oMpyUN)8XV@ zq!jvy8F1H_QWSS^sQNG~n}aNLb3HTe9ZKpYiOeWCjPCgcP~w28bS7perEHkN?B5+_ z)4t50zQiDwm+sBxa5l!zhW$*6=cczL-(gc&DB1j*NfGa*v5C*@a-|p3txI{NJ~WqS zJ&P&dIG(;OnnN$Llc?m&OsaX1NMGJ(Q(aaag|g!$vtc%Ei!7)73$u6-ONZp{#nba~ z{Ql$~ONQTaIDhm86YggfGV;>-O52}j za^6iO^B=CmJ3UvKly*KzmU@%>qnQ*n)Qqa#HE77bbPDfe^lU88eP+4Q1D8S6VrEI( zBm8L7V}5oRY)7e+w^H-GJxpe0B76PJmCs29(rMpHPp75Qq8Epmr+pl|HOr3bH_B6g zxeu*)stj-2^=!uCMXXgNnVom^VEH>2F?Ej>Y@t{x?T$N2LuVf$7dv(Kedu(iJT#ZZ z%vwtq8squDdn=W;s*%Dh1=8M7OJXBpSYBle%gTMrvX+&T+tyN=YMe{jV+|-GeLP9s z)}>bwC#bR9j>G~@pjqe#x7-^PU(roAnHtn2L)1T3ol3hSDX^rExhwo(s^O>D%g?dQ zxc@0L*l$ZqRUK(mejvT|JVLvj-DrDxCI$Q&O%6k)Y5yJWq!NpzWj|vn)3J~)ct&sz z{{$*+@u8oS7jbvUIts8HPd$ee=}KWd8(OMP-r?r-Ai{}OjSHkq%`m1`X3letiZuJy zIkvDRoqf{ft`MVJL|_>H;-WD=B9M9a~64PxwG0;W5~Ab2V0Z*n*IDP zNo`t5G<#$j&*Ge+EADY5DKUvO-sg~%+9*m`@Qz*kw2Yoa2sE%FhF-4-q?;CE^h=_I zh5Z;qYddC8jN>F~xAmp0b-Ss1P85xo%b>mIG+4d{m`<=3_d4xnF;nzOaMqx;$F{SO z6%*K2%QANHM=ARtKbuM14`&-S{kbD6f-Ht+@%K*xiB8F-EtZ^nYrT^y!xJgfFNIcE zZK2!cf0@g*gWT~onVLu3qqulG(wcLJ72T<2A)8aE{ZBb52l0ERb0O`yJ%a7z?-AX$ zA$%`4gH;wBV;=^4Q23oy#C_v*p!_J+?<-;npFXn{ddukVv1Ho+*OgYj9zcn!^Vo&- zIyPm_5XzP=pcWf*I&dzPL|#Y?7HWv<`QxlGHhvmuYLRd-!p?SI` z#2pNU=8_*sue#3}Wpkj|=?M>eeVF{&iJu9{SfAm6;QVwnPuzxz&@K2p?HYn6e1fj* z5&U?z8~VdmqdsRY=F~0`)*TWTN}pVXZKI;FCE*tq-+lp;h6hNKJO&f%$&fZnN55eT zs>HWLEq8(7>8m72zx#$KJ--m-d=$|=lOdz8Eco1%72cP;gyk)9VcX^BaDV<7Mk3uf z{w)WZ6FjhYM+}sUhT+fV#RFu;>4s&OS=Q*d9KMGJJ=KT20~M*CB$I#aH~u z8w9-M@vY!nVHo&X=i z(a=z}Mx2NgRM!YX{Z?J!Z}dRnuW2wkmrg<8X`Z=D^2O^T_Na?Dhy0t9_ZTeem-PMC7A;W|p&R?*EXPa{lc?o~_s0kmlB!$rv^n|Tn2McH2I&k)(B1Trq zLiK$mH7G>FaqkMiY5|rU&_dH*Z6qa$VqvJJP~ssj+%c&_YoM%<(NKaXvd7TL`4q?J zW@2X37%Y6a7t7^#B8l%W%i5!`zIP@LOK z1%v7{&`g+uxY1)UI70)6tEI8z^*+RPrepGv1oReJ2>%vq38oj%;qEL&A!P3>7;UbF zd4mJS$jjsVR$##K3|u+58oxxGk^CzYd%SjIOq&l5F5=wsxP@30!FgLw3owTnL&4n= z7bqSy?br@6h3HFtbU^=P$Qi2+;v z`5rR{N4k=+_vduH8*T(u?QfJR-$NZqqJsOIx0qA^A5wdqal6Tm^N9xIZRI;$zjhyU z&ZOhh`b8MCb}S0iT#$UrnCD_zF*>OMdJD`kHbn{RYc0^991An<&vx4}6^DmZz_xEM z*36Yc?ZhZVeUHP6BTMnYYXY)=DdWRxZKQdP608URK&I|TZ1hzX5>Nk!Dzz4DD}$id zHB5Lv(pcD*coa_2G0^(91Dl2p!riI8RB)F&#)4;)qs~(P$(uzn@5Q0tzM0|`nn+`3 zCyrli;m+sJh`8~abbr4lSGg~A%DE1!BeLKzs1)6L21pP8K?AE~P!Xp`-`2>`k1cW} zohAvMU?8! z{)XS-%+Ry6QM8g;Vp)R16`^S?C^E3JlVvty``K85I4 zCAd*_pBfU6Q11SFbojw0BL5lmc=TMVI~;(jHtr339|oiHkqC0|LXp^P#NZoiTPs4! zvIA*cPd8;MiC~2J7usqtgnT4_v&#e9*_AhWsQF%q@&EGRPA}-i*=Aaobc-fA=b~zU z2A^jhfTqYv^4gkDTm1LaoUTTCPR=%U2FbjTXhY~N6(tJH! zc-%vu_sHTfXE&AQpTrj=Anhjik{Kl-wmu5OEML)J`P*c5yNtx6hQsSbD-HjqhDzH( zXiOOeyA%hE3n`)KorUz|ULoy^;B(i~eq!4OL1K#m%=Qd{vAr~cb&g_?K`xM%gHb=E zsV3t$QM-%<%KQAqJ^N{Jc;i&X~LG0y^1e%H{S`?*xmcZs%d>*3FSG1h!| zkL&e|sI17IpKoj_w(K`uWxpu>J?FAK(ZE@rv-HkTgle`EB_6Zp?jbv}m^_)jygW+3 zPf1bUQ=W+!c^R{IzeV<`O6bO&LXF`?+~Mcd^&^&%_5AI$%kCppXNtnGV*q07o^iJK zJK8b%DSh+`N1jO}eqGuQ&jyBNiZ0M%T>F}8j5iB);T|%avyjGOK#&d&%`nHp1^U=_ z+7L6Q>ti`9Lm*2-Dd)nj|o7W zyeV0R&Z0Xv+(}@INQ{5}<{Np^bK7y;f2NJ`1Nmo8We}c-HuB%+NBXe0n!HLbP`T7a z8u!E(eaqaDS>uI4rk&JLbC>>0dQLC1c&^!JAU#Wwpc|?TA(Iram}esj$NSUK3yyRv zz>41Rp4BjiURvBQCt5?H(c{BK(%wzWl zRq8lEvG-Y`heF=qw-FkfaC`YTxb*Six7^xdShR(kz=-FPz^t>z7)Tsr- z18@7mR*PY8yYZw?Rz|# z#pkb!YMJCL5jxgkLF@CE(xLB1DRkl?cJa|w)|>c_S&Z*z#^<`2x0fG%2pdC91#hVG zvuBupf3%DM#b3~F{yW_g9Xnf;f; z?EBJ9?C$GZ?Aa=waawbMwcQ&?7Mc=tf7x4>_>8-s>#WFij{#k&a3K}v5c+P)?@qm) z)TC9*bK44Z_Kg;2FAt=V{@0k2T{9bEZcaBii#7F#5k9|*gZlABc+8n-*Y4!foD)t| zaeoI<&RSYJQi{fZ2p~(|K?vJ(mR)QuVH<*uGM#t#n7OPb-MXPfU+!qp;9Cu>Z+1Dm zzDJc3qLum1*PXg6a_MWuRZ8Y8Q*Du(%yh;t=5WWFnw8?1ee!kY^rwpTEWE<(x1MKH ze`(X42>u2E{Je)O?<14`J;<~sK2z)p7O+f#PN z?l}A0{)u&F>r;v3ceZJY4vko%MQh!QNwMcC?T)?AI(J)8!&oyC8R|pQCm)g0w(I10 zN(CGC{iAp0V%TvipMD(tN=NjhvEgGaE3j!{<1Ih10qdT!@vHRc;F=cp(BvdDQTfd@ z)GxAp?Q#;clYohrJhV)nuwl2PDeLtECS9+MO=B%lzRe3_e5QV4R}LHb^EDfM-vLod zE|8eIoo9~DGt-S7EG_>M>kNz~tI!Hsm~@(YrnORfem*VDGlj{EYBp|dF=H}SOys}i z6eKo>y1lp1)9c4+@!wZ;t@0D)Ng2?6t5Di@_Z*e;N|v`_B)!e$9*#sc!Fbpgn6*8| zm5jY;lS)NHvKv1983~g|?$~*LI5s~q5gM-Q3bO6y!fFj~!NnLtuf31(P0k!EKg!~+ zt~9E?DhW$AsS4UxTJin%shq1B8o0D<*!jhHS7JceL2x-Uk(7nZa*#zLpkxv^|6qD=dY3 zpUs2;k1bKsXO31j8d5JeB4YCy$ga4G#lz&Fyhah%4h-dI(Jgpg5Dxc~3HUI|Sg89Y zC0q|0%z3RpaVx(S#nZa@UVSOg-mgXY7!OQO-;S}iE1~e#AL*|yq4#w@u85@~jqmVe zTaNL-Pm+r&0I}B@$#-q>76Vu*`3%TFkqo_hs@NS94r?TCMp0@%!9w*>e$!z>~@Q1sk zJlwv>A^tCCCmUQv%GG^HoVFXia#|QaS{uW22IF6}v!HC~C49-AB+P90$HN;_AbQ#n z$4zbEaZVQ-t~haifgkqD@E*i4FFd&}BKS`I4v9CdNR{O|=M4(Np6VgO*{Uz~cQY6yu0XtfQa6x|n&QCT+Vfq&`JgF$m z4i*tSbrpne{ZB}Ueu<&+pJDwn6n;+IQC1m+Jl-2Ioga^K z&*K;;#Ng)94XA&*4~^IPxxN1opVyy5^pnjn8@vlY`#ll%Tm_9jN?7(r6t+5#@bdiw zO#0OXx5>tD+b_|f|a6wnqNv!lZjpTWkF!lU)Oh2=TpH-t^a>tr`ifoX5Py?ZVEd&?G zfkKzuAYu7xBVn7rsSukdEj%0HAe7}y7X~IS6UG-F!6%Pr=*u?R11YXZdjnE6p@uoHP;H-nQ`P{*eJHqc9B>X#IK(_q~+` zjf!VbbpJ^~(|D(S*hl&#T7^6Jb20m6Ii6}>N3wT37Ulim>`ytEsW;Qn+uX+`=MNWm zCx|w##nc`B^e6BYy(s=d=@(RZW`sLc-Zs&VT|7tpxfsoz{^%WB2FXABG0m3W3AN8t zclkB48~%t+j_#);q0i`9qa+TA^0|~=IBN4;@!s_ut=xK)26b1_h73=v$l8OTYte9= zF%kzZ@+`(HH!Ksq3auz9!FTKpdX1G7q=1IY;!k8cgqf zpzE%aaj-lZkJIMkft4aMdtH&Wh4)^5n?c5n=e`1fT@mler{@O+RXnGXhqHNZRtGKp z#+cRpfex2^qB|yS6lkC$9N5zfy(DR2+vZM+dtX2*LuI(bwx4RQ4S{sZaQuul#Ox+< zM7!DGoM#-e+)ZKCvjwikC+X;;ca(u!bT3a%7!qnA%nchX9J(wkm>iWAHebDmhkr_G z$jviUx@ZqA*fJ2qJ2kK>+!PmF-cg|CEz)s#PqmzLyx{ju1m8V}rpJkJ(~HN2&jB#+ z8wW)P2ed?uMqutJ82Ai9U-@X*R4j+2I*=$e7~7613So|S5NfO-Oi)utYyUsG-Drk4 zQ)F;Y#vDF(#qdg{g-iyulj@(_yeGO2z6-}9ZcH4{mIq_t%tQF4wg9&dH6dmh&*&Cj zL!GAxH0(s-!OytEq7<-BFvH0mTBsbY#b^V^H(;6@+8<-{h{eo+9>nPZ}JGy#udF6 zlzLbaH95-MKdgzl>lM&(lXn#2rof?o1RP$qkfpd9mUp*O#kW+_-MyQ3_SI9bxdN0f z4ud-Xe|q~=3ziGrVdkTXf~P$9CAp37)pBUZtxgQ0Y8IIeoXXX@KjxM%Ayo1F5IYJPgq(WMTE8UKhfZW2v3h^IaGZ;*u7H(GK- z6SlA0*}cbO=-~2VIw*UR_gCgp?InGhs~SsZoW@hiv&EDcT|+?mw_9!Ev|f?LgVziN8fl*6C*+qssI%I~3ArD@bZ;Td)OGNbjSV@O|dF8*x( z#x!RCv5OJSC6xjvTDv8YJ8RZZ@w*%Jr@fPQOzNVs;w=v=mABtfYd^?qnD< zn*TEgGS$|hks+hVpxmCGo0*c!BU4%+?nNF^S~xE{0=>8PQ}7ob3JfzN-)cY7Z#6?o z);BsltD2P4kCNu3;h5~AfPtH;Nj0F2wjMl63ir*)uy#HD;Xl`B>{_^)MAD+V#VN ztPX|J+Bd)0+o$3*@Kf8j@qtPpXwpqOA2xDE?w3<>>9=Tvj&< za$8LK&*Ev*_6ipKsFbdU7tjNxRB~E$jvD)ourj!eMmF7{rJs^$)S)Rfa#}9?y(^h6 z{8gh$Yfbt)ypioI62YADO6ZBJr>gjORCmsqyp@So?^7g?wnV;fJVQI5y<(U8%xS{Q zQl{(%s$RzXFvi2^{tX9;-DFKR;*8EdP^H4{rliyVnH^YGO1JsW|9ZT@-P_}7tz-b@ zF4#>I3--~_%;WUM;3GXV*+45r&yq=_3VH@?BWwS9e!uFYYmJfYqFX#Q-?OCH;!Gp?D?OIyw7otF8E8(=T*vNu=fa4T--;`(@xQX8}n$y zt7Rl*#~r&CMkIg86pP+U;E<;{vU*hM*+*&Ww`Z(Gqg4*g?^&Hqn%}N;2f0 zss|CybTG!6-ksGU^|B$9+@;H}^#L^NWi|~j?c+Os&NJdWDBpr6N^;?Uulx@3Y_@^@ z)5$o*cO(yZZ+6GyDr}i`2T}&robhoD@p6fHlC}i}K?`6S=pYmfQ4qEn+6uDkPY8## zQ-pBsBZ87a1zq(nrZRCk$e6#w$IMswEqp}xgaN|MmUhHm`~d?oPeE-8XY+qn60n&& zJbGjiw@CwYj+;RAz8bcUQ^4i#Vz~W92N^#nVO?A`_ISR5Bj3$jxi|s3nQlm$rVZ~> z17S#osbKouRgh6C!qa*6+_heT3hn{jyrLPmszn5ud5$>fWQJK*Gtu(5LC- z?*SCPzeeH2^Ehc1DEJ(oC3G4t69z8128D_ecze_!^?n|PeoI6C>D}1b){Y?xw_z;x z0`Il<;YQ2_f z;4UvFyySVhCDu96^*n=#a~<(%h#{s#a&K-;DW1)}!+VVTv2ycZVONm3@OjB#VSDyj zoLu(_D<|#6{^iT@_{ugIS~=oQLLW8k{y@LE3shO;1`JM~Lb7@nqRfV1#XnWJwkq+C zwWDzR)ij~wrKhmwXddkOow7GJ8Vd}>@wadQMhqVZ6`l|3eEyz1kA9&8?q)*IKLtU3 zl7&$8Ge-EhAx7v7n=UkodkKkWT!hufrV8e##|eu0VyR?2+LX00;*vFX z=jhXVoz?G+!*Ni(3BhbLx@#H{B6p0Rj%_v;0{Cw1m)}$J8)Jrbx7|@?wG7#JMj%>8LijvTLx|Cm7c{yY z5bk7vp%&AT{B=1l57~e#n1IHHcTl)9NKp9k7yoiL3k{dUg;tRS!C$#RXuXy!boWFH zPq-^BvnYs(oUmo*c^F=0C_)vE1V+q|*a%H1p4l<3)J=9j5PLo}eXwaS0 zY-*exyI?hp{TNikJXc<1TQ<}%zdh3Qe*Xn#;*`M_smarU0zJB(piO<*eC`^k%gVCi zSje%-=v{Atl7#2#$;Is4sN>8nKbD<4UcxrDZ)Y8UJ=n|@JK0sI1I$lx zCmZQ~jn(n*@wDUf$Tm}(Ui*IJXGAMpF_?@s8}eC$Rvdd)HH=*w7Dk&@vS<_k4(?n& zlz#r(O`<#}9+4f-*7`Lu8JimBvd)K$M7A(3QE4)%JW6FvPI&rdG;SsSX7h2KXX`F9 z)h9X>`DqmE%0I~>EhDLThy*?7jQQw)Cs?9(7#o)|g&A)dKy`Ot+V$nXXGNRDz=E#Q zzoZOGSKmc*FD8*?-VPd)HHSVtoXjS^n85;9rLcsS4EEukKD%#ojx7n(CFPhv3Rlcv zGo1a{ry+aU`JK0@a^fvon~_KKFPs+l&1LFwrYvlJ5A)DE$~-qNVK>8_>6>dXJC?AH z4Wdk%aQq}2qgTWp=ggzdzJt`Oph*u$F2KI8n@E3rG7XnAr-Zjk=w7S|L7RA%-maqyo)w-_N4H_1gJeE@gG=SJRca#pKg?lod26vzrF>$*t6Gw-NGbbwua%V~51?zE;Y@kcOlGy@At`s3(cO+H z>g{pm-84n|VwX&58JC#X=U`@XfY=2$SL*qmM%Ue!QT6fnypJ{%T|s=NmU@e%NB1$6 zpa|xybd%|MDzNl_8?P?qxv73(hka?=1G_mI^)?_0iX?Q?z}+Rn}G#!m>PC?JocA zVip$;v6YjPSkKyZbY|io<}}ZPUX8F}vmB1HRew`h&vhGCbbf=KZSya?m)r8#i_#W0 zB0i06Xf39zyyv&fCx}}0hoF7s5fbHR^H$rhFc?scLwUKF`y~f;h=QN`B-H*|i7M;q zpq1QGl|R}Km-C$V@i@Nwl3iJpasE#^8a~fQWio#%CSuLG~|lXQ7}+SxFB4?%6EHF zZm|i)){Of{tf4BpRrqiN!N%vth>X)i)%Qt|=q`io-*{Ay*^2Z2?!u0l62d+Ecc>1mMn%YVcy}aX0q^lO z-l>4o{rgzd`WO-gQFzUsCttL^St#%ZsW7(YCtm(e zheg&zSdVe$x%r{O*S%LU(fvOJXjS0*vI2zjT+1ep%kUXpgjZRmF!D=68?Ay_jTa82 zSqm1?Izqyh0YcE(6L7rD*(!^paeZ_kI!*_nXO|{IXHP)k4SCeBk>T9EYD`}q4(B!p z6t=j)<@E^64jhi*u0xUB5&SN(MdyjF@Hq#RKc0q_kvZ7d6tpi^LxN{;KK`A|S&zT*TW*M;6D-d8n3BS! z?syD4Gapa8Y;mes8Er+8h_pX|x@-1WK8N>z^ZenJ=Zk(hWpwiUhV|ff`qgp-x-U~O zCEuC*!H-~DWG1?QdUKC)3*2f?;&XQ_zDMLiZeKAvF71MdAKfmO+IO0JJoS~ur(y=G*SEa1Gal~ z!^r6@tVYa2XviYu9RS)~6a@`4IpKtF6YltOS7Yl)oT{7-uS4tbkneVyEM2f}jl2-F ztqD?>>Tn?OJpB0n-g5UWSiee#{jt@YH7G#y>mUI+D#GA<=MWTSE4a*(7p_mZjD#CT~usePYo8gDQs2|PP#tEC$~57 z9dsJCQk+ZkP*>O$cAUAi-e#e}1#IQ1O=#b}9LGjQBVA%8j1oO@$t4A<2P+`HpbCoN zmtfc|MgAe0RAVd0-JEfh&$Hj*Ukk~wm%EeyjG>WpcpiW|^L{;t?NZ|o~GdGZ7eL_5{lmX2$bH~jV1NmXW3y2&oJJVJe)(1mD1@#SskfdHx=@g zTm}7AvxL@Tg^*1s#m4_?A^x`-2mR0BqsSfP2J7S7`k9aqT8rBEyAX8xEFNq-0~M8f zl)~M(vi;nR?Zz3no*twdI)`q=6=Bwf(+DuQhmR|!aoL_L00URAzww=pd{Bgn#c=LP>|pFm2`gUq ziw&47f=6lguoat$6>E9l>6w&}|8}5octk9kRCmCa-!%?dU80fcXK8oD5uTMc#$vvy zkZj@1Q01|-W|}9>8XZhJxsl|oSVW3%+G*nGe(oU`6LJTM3;l-abmdDd2_-wpzElR0 zD|k+G(0b^e&%$Jp=_ouo8$*-(`F`61CYMIwOtw0T^R=N_JP~1Au3+QRAMlVA7v^X4 zXXdC4svPD(SKOQWQoyVk0ud!(8 zbv#wNiZc^W;QQY@xIf1L8d`o(o9BlmovPTWC=IPai{W}%kyO&O$vn)8Ldy2!V4C34(-kL39bs-SK>LcQg%vUoX0*KNli$wT<4oX>k;d1(H#9+f}#LqBp4(!RW9 zQ5hO^)60~$86Jh={oN4nssyvT2+jNN;TO@4@7>R6@`P`6ENdu&6Dyd)y4&o(w%2S| z@Borml^~yV70v{iNJ@X_(O1*Gq+d9R`cLho#}}(8v33f&O;ixN?cw{rXzkv-&Qb8J%w1`A9j0O z5P8jVA(zqI&%(Kvi&US{h0DgUYY0KAdH@{ki?A>B5RyxOV!`GZNJQpiLvjig zrl)C6UNRMino!Mx@672?I!SFFPPZqNu?M$@(W-=vaGbUs#fo8Q&Jf2iF@2ode1T>K zHL^}m8wy!?p1mqPOi8V0C`F@yjK5By)YcuO9(;m{NOe%R$WuxgX9{OeMcTV?FxBh! zu-e{WDz$f_mIq_$-G-s0xy_GUzPXcG&Q;o2>qfKG3&^VY8xtLSlZ|>YnC#2Vz#b2Q z;pGLW)EP@L8z$2rDL0bYZA#ImRjhWYFD;zf!rG>q(P+*+ZRnIhU-MX4rjNk>MW;yN zP$gOO&S{VG6DsPjr-jd!AhLN5#RxX^<;fQ6%8j81E81DmpF}k>+sWjF zJ1z9^qqiw*$n6bhaA+A~x#b}I=)J-_&V{tDljmEEhtRy2BD6wMf*fb;roWrE)5e3D zR9|Zk3$LBXYXK>=Rx_)9IV$uJqw_mwk+O?7MIRVNnf?8&^VA0x&wE1CPgv5e`Aux# z(V5iOB8D;H4P+g&9D%RAXolxfdi?1&*`z1Z^m!-f%Na~iayyHx(Do`09JKLh%v(--{=$xRXZjSx)_ndG{o28h=jK(A#hgn!9W!edql9 zTPr2V|G!rD#W0#2Blc6q{t9~Z@d52iJwcCrdTFq0E`8t(3MJ_$Y}#Z+T#Tz>eS0Kn zv1TJXdp(1U*;NuYU8m}Bakz@?Bl(XN<+{s~_HZ?z4 zLABLaS-;9TX0)u0*p)+fawyDH;=@Nl{J-G@3iq z6w?jp?!ZVg8(~Z{&05%m%GuP{%;!kwZ&AePmvo1-ch8qIxJI|r=g#MpmU)l<^H_+c zstW41dqqQc{i2qBNxGB}$IcBJK%R!2b+BL}B^>1rFpsC;6j=&g%bDklFT94;vBe%hYZVYf-Yt*?PWXWvyY|JaRb&m%H zFBwZFYd$i=gh?d#LkBmyZu7lu4UK5qMd}`6cxN;XR!*7ZJK2StzlBr5*6B2C!zq%P zw2K1bvw3gq37L=WC%gMR-*UlLo@znpz5(IWq>aWv*r14}Mfqi2s? zsOOUgnZ8?0RWboo($&fC96m)2$A?j4k_`pV{><`ucQk8=4vl(iL5eFxu>PVg=S2A; zW|An1&_BpJ6H8vi1R1};88t9I#AUDA4L-=R(840MRvrO97%JJocWQmat` z%~5rt{qq;nu7Q8plV&}-z_Xa&CwkJ~d$%c1GLIU+D5K%bKJu~4qj`zPDOhAbS=o8e z6p4*wUG73ApKPdNGIvo%U!jcyyl92$HJUjroL2s{rjr%D)HD}Fa^LB{E4I))=0@4S z=h1s@Wh%^XWR9~{Xn#Nry;R|SfYl>0VJ!cythAv2vfi-X#gl0L)X|joO@mgCE@gdI zsZPKM)9vl zQTKFL%HkOl>m_Y8{_a*pUFJDwOXDr1{gznbe@EG`saIb-zxd zV>@Qj_M!mtAJN0!e#@tQZ?efI;yC%*B+!RWCHato z53j|*8+oMvxte;SzR*gUSejI;gZQuWuw%wL_Id9_ws;XJ{@-avA+OlV2czlxS8uA( zzf5Na_fdr8eDa%liZWlSVE)lk3Z8p`fpi3+_Ql2)Q5R%ry{A@gVT-+nQ%pep)j>_vIkN@?!L>x3bEdj3Y6jLr|{ zXXN2bV*DtUKI0pUx!}uATs^?jiu9xj{#acV}iWxzqY|5T18dHb>-#hq-$23gEl^pE!Jio}i7ZM0!X5{J~k&yGA7;GB|%- z8b0TPaB`Fs^D^~fiE;LnZBj)~ryrrJq$@Ong)sBI+u0oAe3f+;Y}NJs%w_3ta?$6! zAk7>`_qEBd<_0sZNM`qJY$A+@hikluqH9X7ORkoBaJZ__wRB5a*&nD40 zDa_txfH$LKDXYJpSmP_YToyui^gC(#F%`Tz>&reH-(ZOuu@sjUMam~QgZH&5MDHwQ zx3Gvc87`nrU8YR@+-x>-y(rx_wq*Y8c_B1Rk<#^F*#G_v@5m0mqLK_f>PQ}`mzYf$-1 zCi@KN`T$ds<+(eSxt{%a$nUtKQQU8Q26OB8pf|x2OBD*)5epd#QZHo7SLV|7q9D4U zdy#~@H)wCFIs9gGo`wB1DqJ^_^LSb)VX!o{&QPTj=A$WP#7YWWd5Khpr*dC#F8!@N zN-sy|(QTXIv^6-M67#F+@6Mm>2^M^-p zNHv7?OasXKl`}QX?x6#%4B3V$cx1hgQugd2%k6V1XL$w-aTh20sYrH(qVTG}NQe1+ zyy(ts3dp)izTyRRb5Ju|TdhDQ4IQk$rJOe34Wn6kC#hFK6)I*E&>_b&2U!E?=C6VD zs(1kHXs;%9@q;w_R0m!0{K9JDZn4D$`n+#-o-9}#ZT6_A30LDmzx?p$=0!YeI7cY9 zr{9-9k!4T;3y!_W9_)X?p3fMEhPco4mp_+2R{dqev>ver{&iZKzheX5j-#^N*X&)+ zd_+b^!zkb+Hi(4M3*82CsVt+$f2%09`UMRgeV&Yhp3;c20g&ASqMRgJ6%kVGhk04a@c9vqg|j!)4AU6ucN>7-r`4?dpDaz8Xi7Q<1v-842{Wyh zrr#R5H2;?sy-P}=)^F$7*-K-nN06lJ>t*oASqq0!WUy?f1*xBppqK8yNn2TzM5a$5 znXt=DHiBn7>i5%>fI^y+#1I)52P+?IysXlqe}W?uOKW2<&*f8cwKBG;j)QXgMY05TS z{SSNGC`~7gt4N~#ISJmJQ{x*)*8-1{is2htzvQWW{gw)-1)lZ6d zVAW`nbP}7Hr9#fPYH5GP!0#T>*j0S!Rr;Wpp(uZOvD2a;-UFkphoaZlI#bsmjr9CKrpNNrb1vvNU z;)$a%0;44a$+8~oS*I+NoINWHIK5riDiteul>A2vt6!5HKWBvLTMIf{Mhd^*kl=E? z8MlsHf#bb5h~N7heRa1GIsXS#t(7rh%LLquvc{whYq4sLBNX_&+qYmHJSXqLxN>_G zUOj>1M&bDTFAMT-R^zoC=fxH|;ckzl@OO=>uxy;HU}4k^#o8{M$>ffn`y#^0J*CLV z`-x!Z?R@s?hocFJyyqAsY&{nyyouW?9J%!x@^9~fDgVOOh#E+;R@k^a!hSIkK}@rfZWR&aE#2&WM~5nm1ieKAg$MbP!h-t-g5(=*!7F{Hpt*qaTJ>b1D*U5g zi_`GaqzGq{a-jJ|5mhh7Kp{gCF-phb%RB8-t4nc2Z;{Y!Y%kpW;36b#6BQ#|%kraz*CIoN zC5MIyH&yL$>e_zP-*&~x?Q?`!ZD~PeH51OtMq}&gBZxVk1drB}Xz|X$myYMCkjO_K z|M#`rWAW3c0h33xb}q(>uPzXkap7$9a=P!=PB-*pEyoadY5im22hWU>u?xuK6&o2*#!-65$?=2;qG#JGF8VbULJ`uscydQ!~52h-< z#`mv%=)25&Vc8$(-oF88m~@W<&0R5Lx(<4WO@N-*G+bA=#Q1e{a81b?TaAs;ykRCz zes@Df`%G9Zori1NBVn#}7$@aIp|YVKo$}Rqk$o0Bhx=ew^d5ND24I5!9Nzh`f#c|% z`1CFs9&sUfe=`BoW4+*G6ofAc!SLGlm)4wpK~hS6ROeTYDX!0OM!pJC?nN$;+(G^@_BnLA3@6=x)9p4AuBzs(f( z)wl>(8dnP|*KHNVGB*mVx2+N?RjgpD?2i+{0Z46e!1nLT*ut5~MN2zr1G`NIF;6K^ zw+|s!zu~RZilUV|LYcoHxc=^j(%eR7XI#d}h4adUeo`{V<0NX$l z$o00M-MR#7b6cQObc~i{aOU`4U-DGF##C)zuM*K?^)>RSb>xuPS|yF55`>0z;XuTzHw+T)^7o@jeKsQVGjW>tr&bO+i)PO{j8knu%8G0yJGE)0VCU-D)PZdRY&q?fkU5o4H$8oUb1Pe zv5j3F*u;*GYGvW0EHO1k3LBFqVBC?fNK?Lnz#rnmC$~H(`7FThsvOQ}n1CVXKuOOG zEO|8@YCAkIH$esy4jMpbtUThgM&aP;e4e*_LSECqlBVtryx@P9bY6v}YCY3hTfoNu zl_90-^N?AVj|asC$hjH_^%bFb5OxHUGc_r1cPG>5-$`(?BEnJ>V5eXVk!u45f4`Sl zuib-B<;x+%Iq^!-j=1777&*~%;l5TIE}0M6rKTF@qVS!a=(5MebZb=bUfSY&+%xfW zINes*Kx3Xs2&J_x$SwSgmrF-rU$8uO=2_!J9KX~3(<9wl8MFeR4`T777`QWh251B!nx6>(CK*#MFrcbWZ4{Ywp~qY zT>oP-8U<{PTp^Qp9?9>Aj4%But+d043VyV&8U=xA2Ujuk|F$B$_H&ONcAynP2;mX`>j965O=lhb-<8qA+ z>aJljx}~h6?G@X$@Exj*4ZL&_m9e7<^Wf`v^=ina{Yt zOuWh@7X4z?F<+Txc@fX!FQQe!$LKT9BQ4v!8aKmzzzIAUJjo2BPwm9jx2Evt+5En> z`pD!Qs7Ja(uzhE1k5M9%tvnVj7o`tc2p+9?XhHr~L>67$Pqctv#~-$m#?;p)D8XAPY zg?piK&Ia@MnnAHy6&g3>aP8S?>WD9;|6T->bagP6pP7xsIlRk1_z8`#$)|@Oi)pXZ zXkqBeKG<>Qx@rD=EEUzmOzWwbeNYsSvYwOQ-N8an;1@_XmqBT79_^r1GDymz?kk!wSoMWy z_5+HsNu`7Tp3p4iGCF_;DCypUrFcE>wa25(B^7heCt>APT_mg##rrqnc&hu02C0P6 zs=v=k=LGLwPF#&++FF=g*9XJQeoVXG4z*WpSo1Rpo~uf5d3i9D_n$;-$0V41>!yf~ zGScqMrvvIY5IOA<@{G!Hm*?tTLUe>><=4<{_!vGnYLIpAI4cvVel&IOMPA z-tWO#m5?Ym93+JbMVQPC6KAW0o!JKCLoEt@Z zCOpBPycR@iUBmphw`jWR6MD8gjlS9O9^;X(l;)rV>&mmxomYqx{0>x8{(%;M&nB_4 zk#tg>_dBn0Z(!hI5;5pUsLe=WzI!L^7aIzB2P6gK3Gc9W%xodHTTK{O_7F0y-#II@ z0V3)7=+E~@p&n-djdDWA3sEF?s-ePG4WA|TgbMHB!e;BA*z|j>&{1M5gymZaqw@6e zzSaPD*P6mPwFw`MFX817?%??nj&g;)SUfQV`X#sVU~L6{?5l(3MW)g*H1)p>cp820RJCe@{2! zQqB>a^bf%1nR|F2B7^4`Gtpmg8UwRaF+AxcR)`eCOqu8RRkP^U)`yf~WrmhBa)`FD zLMy*(S*}Wh$=VFanX3x2W#2Gh;ZG#1nZu*h3Nlk?;MX_KETcRmUTKG%-)$H_IENjt z8=#@DhYx{1c;7u5+AJ8R2N;xVeX)*@rdrmjLi~&tEdJP|)pHe!%*>JX_Y3?Q|6o+g zAB@X+g!QQ>k=}L->CXoWPf!L(I2`j&(WB3avIQ|4I`+_jh zgtM@y08Y~@F!f^@T*(&ayM{wA(g9o4zTxMw8eA;?4^oc@qoMp4MVQE8L$D`;yMo~x z8qS{?){s^-<{5$+xM*?|mNyTf$UPB1#2q2ISO@8!OwhEA^Rlj};G%;a93GCwz^8Md zB&&eLh*fZ2ycdp5S;%sm3}Y^r6xk>PU(KO{fvbw}_OGI_KU@hO`hQ5OsFU=!I^b8s zOi1kEzTInMz()Xx2$>CGQzi~BIf?DRv+==3L8xw*5vDsD2$Pl7g>`Kb!o~x-Ld}N} zLa+TR7%rC+R+itOAs-)*sbUpv-!TKbwJbSbVm9|}jz!BzU&Ien!ks@6fc0$|CQpCbY+!!zN`c z{<)8aj@^6|`1wPOch`2=*u(kuFywH?*ZR0oSgXUGO6QFbtd)$W^b#~ZslhTGea>9d z#odJA@ON;=zM+#j>Cu&cU7Q~?PmBMbOE6O<24O?Jq2d_@iGvGq#M=o^$4-YZ=RH-P zze{_b)YGc3aj02%7?L^}!1w@Y8ZSrrZCl)%%KK4Up3vyGJ!Irm3qz$Alo)d7blW(= z;=hl`O;8q2KGqdd?vEAz-X1HA+GL4Q-?riNzBnYU@1~9oUXYj*2(@k5G%qugGMzhU zgF+#qK6m2tKRKb<>@S}DdCR$?1BIIYMm`_SBZ&jG8sQJ=g^(0X7JCWS76ptczjHP+}U z$U^L6eh=*_B>9TJ^uS#L!`_yVsP8Cv-+oRT`0wxLB{N|VD~Xnei*Z4G4^+PtWBRdZ zTH#bdlOMe&yX6Ov?Yt1bK1IUmVKdK6Z-uUZ8CLIZCygdCtX_YXI+stw-VH-JZ*dCV z{YoPJ^w)I%(GA+NP8YeMS_tO(%xwV$m=#)rDf!<~w>Somug2g>+hSDNhhkP_4Q?%u zMOw@ZWDYaJy$L&U$xH+H0u^A<2V_Z?QT^R}^mQxm9VGJi)X+?fTYnv@O8=<%j|lpE z#9_(YF)`u{BxNFyG{q2kF=NsFY&_2LuKx58UobrD8c^tTY8wU3mu0e^!1JuDkDG9fF-^3_No+ge22hc#8fzp z4oAVpQ%Ey91et}i(XM2R(Aj%Y7F2-)nfKv&DgsCEg+TlGbrdYRO}F(UNP@HO3%~1ydh!oA)Y&MLX^HaPE2>kaRpD5`~&x98so+XZAjMN zrI=YmFv?FC4l4RMbTj}*4=%>y{p-;ZybIMG3E02J8y8b&q5!!nvapK_o0Yr0dL!gDqw(Kbs44?E@1C2N4hr9&W= zG##aP+(0Mn@wht#_I%D+|;e5XI&6fiRe;gB7;E=y_;`<<(j^=X{2C49udF zFRb7g9Yg~*29tMEKDk)V#)qBru=1TJ-knON7t@Pr!l_O=S*3?+%OG*t zK)g^Li!6KoS?c2rm#hpFLyr**XPO$#}HZ=zF0v9$L#@3s3IBJ|T-w8Euse^FX}Bfb6c91csHaklV2?%(+hdtE7EZjqF5 z_w@x*7!I{EHZUrCmA?NUirKxq1#l(QLuu1OoEptJk@CMt;-dr_jNjAiU_LYW`<6~! z7su-_x5=mS4c)ms26;smD0r;T_ZnIBBesN6e!i!DI;F5bbQ7zm)?@s_OcE<9qR)q~ z)872=Bt8BvMeuy_FKI>m5gUuSyGL=CSurl$i@@W~Ds*4S!|y{IaF^$w!$x%B+0nar zlk^n}qb{N+zZ@Gq&LHK=UJ9{^r(UyDq)^Pe{m=Q=&hzv0OU{v5Tqk)Ki1T^=Cd{>p zhH>x#G$w07VQ(rFp4&s}t~Tc=%!Wk?cQ*bRf%P>OSf4usUSkx{GE@ie1Y3OKdAj3M z3vhBs79#t#Q1e&;W0z{IC0-||3pYg$)dze3%hoianFVV zesq1O{`((k|KV0LHL9bQO*gn_Q@~BN`PgQs201k+Y>oBCi;6IueAY&HEP83?tD)#f zZKEn5HB9qU=bzKVQJkZNsbNMi4*yL@M?WJS&ICPXm`mbmw`k0vcjPjCC#Dz2peB{i zQ2w-&7s$&dk$%3a=X1#s2vIb_ZdEIc&eeu$*<^h6n}x7l)wF5nMG{ky!roiSSeG9M zsVB##@)96hll;icy{hEj01dec=17UADBoZgdn=YyS;uL zkLaf1S-SsUAoTd>fYt5o=nUpL0m(hs)G!$pFRLkA>>(Ar$fG^?Vz8sD40*jSC>U2q z*JfnWb-g3>;*AT3QeOb+?na6Ae^h*BAYK+-rj4e`7;7wr!fF$o;GLrjHrikhtZ?T* zF`dZ$Nb|P0(GNu5jpX_p(TNU(KE6evnL_iE7Ro%JO-@`@iSt z7pd{?2j7){;mfBs9cVuqjMOmR^L4*L;}4vnHr^H3^xX%gt*T(V7eI8FC&sq=V4KNi zf>l0U(_1ac$4L$GxRD;A9MNnu3lFPi;nqYX}~a4w>lxh;u_8tG~(=9 zHz>P(CfBlhvdDN&=l&EzXHf!faSr#Shdv0JyaOXPOhBZ|Svtyh`(k3vw8oEn!R>4~ zpKvyA{%xUpmov1XtB$73JWc8$A!L3-k5pHDp#96+IbZoSNwg%OY)Sy4KkSC@zR`HR zcm^!0q|noQj>byp(1@oVG>NnL^8K}tK0zEUU#sXv|g#<{*)FRQ7nf+dp<7NwUE=GD7v(?oYY5d;|`+zXt`sKh5oCkLwO_3`xroZ z?+vglV-=QHn&8>W3`*QnK;w_Zk*57M^4C2>Nv5@Qr=W_AWo}Z=l`|CIR79B*n&^{B z5M3R2fwIQmrrhKN`ttGw{!kF)Gx)u0)($e)wB~$TTe|b#0Q^1knEov9p`Y8-(DY&u zbnEpH|8_kxcn+X!HRl3u3BbboBarthz`5pQXo|>%dx#y*eA|V-OA&|+9)`2^+o}FQcEd(-g<|M#+TB9g*RyEkuYecyCU3nBZjmD;1B2EXGAYXvfMOG zNuPo3Mkd^`YYHRI3p*?-gI*R*CtDn-A#^^qe$B`1{t|rQyy`UFNV5E4M7Er<;_~+j zSvP&7h33+bmVHSoz18HYe4Y+}eo0z9``J|-MSbbY*l#Pzok5}Sy8Wi50el zG*f{N_sg**R-FrCB!`jVCADD)b!Mq#?NUY z5#7m19cP2Or#xRR5{dTL3o!DXHLj|~kl6?=GTP`zS4@iO>+!#&^kf}rhA89U=o=Kg zw}e)EC~^js5e~g+H1vcI9h$a@^49VkEBE$fUn1t}+Q0%Mm8o6Df&|4{WXGxz}MaOo1#dAXXV%?)B>XH~E( zFE28~oIBKPJ|0b#S14_TKKabIr{VW?XqQMBD^Lh#liFk1s$snh2UIEi{To)3%exup zIgi*Yjpjx<(~|p(s90qaf8KG1{NlT`I3S;n_f4ep=fX*&d@}DTX0qtYY8ur4nN6DC z%GT95Qg6o#)?aAPEJ6k5I?0Y@J?h|DGb1#lO@i1;FS=HkLxT*~(Y29rY{9#$OgrN> zi(>(_RWzA$NBU4pEpXMKi@x%+`&Fk%rr%f2rd=s#S2u%hcaEgSZez+c&0{x2`q(Va zoOl&}lC2|ib|$8jO?7P~-}^@7`N5X%T~egV1)4PcH21!@4(7h_-E{O+FY8W@VavZu zGLH+x$a$wbJ9RaZWt}?4Oln2Rv>}0&$7izPbKgn#W{pMM*<$;H8D2fZe9A|Xt3)>? z@UGzXEe}Z1Cxr8>eDKBpGNi@Q*)G24oS4?i{@^|v-2REp@`+<=qsB15FG}pI{&KdK zcbwcpyP4p5i9I$5W;LSEnBwdu%<9W!Hsa<5w!o@@^=9%+*jqo6UN9JEWjJTtd<>na z{>99H++(U%QdC{IpEiAsVa@A^Ww=T(1^*E=ZPEt%Z+j;zI<}W{z=o2A^n8jLVL*oR zEmRw{h|WB&VYL>=*@-AOwq^bcrd|=jw&!tgQq3UTF_OorWx3QPpTHIl%wlOpH<-gh zzGJWLV!QX8VOde#Y`E(scHzuxW+Z%Jayj<2YVIp`y*HXI!5vm97Rmf8A24H`0hG3F zDQ$fBfq8}b)B4{k-l3Ss z!n}R;>IFqwx<#5) zGb))J=ZZKhCeX`KrEE!>KlMaTp$A?4EQEJ7rmy0C!^*AH+c}?R4jM(tj;~p;ohH3F zw3;*}r73=FFN<|_p}onr!aSpi!judv!QQC=b5l3t_@Spbe!~aXi+Lt0DhL|WG=#Ga za>Azzl8^o@iGv!)62o{`}3!AKaHY$j;iD+p8mr9ihT z6SIb-AYZORkX!32B&>)Q-rnxVncX??|L_JLbH)nxW$iGMxQ*V6qXa2;aiR8!yzqJG z2B9XwQ5ej4cYk^)h9Cq{nf$I^T7ywh=P)q%C=B0>7tZVx6Argj!LM+iu)b1Vn6+xD zpuzLbL)V&M#2f|0Cs+zg;3Igj@q$vcpO9U*R#^7ILU7PjgZ4O{t4SFO7td#CvHpaa zYc8Sb9nVk3{3I3YYgA}*3L;bZPR8>h7JS?!{I1+Bd{SN~?2%k2G`fx!9V+Y)T=j&7-ikucBy}MqDOh;lGfUV$Se^4GP7A$p@xpS65@AkSmaywsyl~;t zF5&lq$;jHb5_JV5L44-4eQOBRC1>GY-6t}L8xB|9VF(I}7ZQ>V2>!Qb2nY4PA~Q%< zXld+%vvC|}`{%+WH5fhH1F%ux`K_!)e2?&$J041~c*1oId$mKj(CsD!hE5mm&gkR5 z2A;>-EGB&K94O=r;mqmyn;1M#5_3)dQ2V;S)Y5D$oU@%NTq;)=45p0}wsa~A_)k?R zc5xLpFYpq!sLc~DoNT5~=5=H-={@(uJ3*D_B<;KwVelJYWQNVg0y{h0&53}t_zs>0 zSc^}_(dennLc{Gq>`aM)b!{wimQ=#%o{F&5LqbS5zK3ADGRW<|kLf2)@EMm2ocOHw z;fiPsKQu_Nd-{;OhZ?z?rUfSBPvC2QE;dgck9+z2UTUp~RnxVE1?LTg&PZ2b$l-tV zpl<-;Ar426EZFnCf3(_Lyb;@rF2~*2!uwe_B-I2H2PSmi8ZJcdX{FLl^)#nN1jBE~ z=Z7ctSc$ld0M?4w({=`7;M=ZolxhrSlPTXc*sM;MWv{}0eqZGCa zwbJ3jGU-FYBaukqY+R#Ysef8Hs-QIUI}qGU0b^wG#(5Za?RkPU?F#fCYDVmu57cIG zkp@IG(ofF%+46B zpcq>CegEb>o=shEMyGgBazp3@vh5Uu?FD6?BOHad9}m-~fJ(ah>@BT3RF417mf>?= z0p!x;(4DFd$8a%BZz{#~vV2q}XJSihA}NMeQ&aF8+AGQ#!WI(vYVwn`Ev>OY))S|- z7vWDw5B(dZ0QVY692(a^c^d~n?4$_pxH=(C&l;8mOR=>l9UmXXaNlJiKJQA#kkJvy zC`*GIzhfz^ctOg!-E>J}Gw0Q;!Q`*=Ff`pBN3!Pf-Gm8dbR4A^&PDxiU^+z|mlkdE139}vA5~JnX?vC|t``l!%sWo-JYfpyC}+%^ z`;3k_zb4;=C!{*`E+=)Vm3@5;x#p}{EmdvPo8{!miI6TCc#lq2{;0a{X<}X_Z?N*DPmEnGhNQtqWXlx zR9G?ts==lhU~h&|F59WAb22S?bC9xq6rnm}5FTf$!*$dU+RZzoK~t4Uvde|K!j(zBnF%%A!p?<9(mT z_bsANy+?Syu7J7utY%v8wzJ+Px-{YS8y4i#$_B6HPFRicz}h%Bf?m~2M_vrVF(SCx2JU_>CD3C0^8D)z&xhBCy#A|an4B@x0cng59&u* zowgk_Ru6$kHQ&LeUqe!61!3B58rB_0a-9d+u9y|<#-}PK?XjDEycWemWE)wdlspEX zXORCq8BNzD&?~c_GFK`A@8_U4#R;~f65v=d3>$byq40Pi#SZbNu%sKz(e^Tp$7 z8A>y4C(*1*6UquXic>y^@qJ(x%(PCh!RL>$Vzn-2UDQC8uk8uqMN~5B20OND9rMV~ zWcwxw^d~BtwT4Wh)7(M(dRsVMdN>tvg&DMB{0$a7=r)ValtfUp4pOi1JZSm=5<8bL#)fT(*P%AcA+-LU2l?IqO0IVd5!7`A3uh?N06!gi@p~?9$^65X zy}HBJAC{*lT_QO3xu0?k9C54X9Ce6g)0Vqu=~ZwI6Eu^VMX5P$8{`AkF`M9gb``{w zyr?KOghUTF(?ofD)?Kict<>7ZHV@s!y55GeN5hY^o~I?u{-r+?T{w{Z&-Ss=m)|iF z1AX!x^M&c$f6f$j29f6Dc@+BnII+YzbU!ScT?p)D;p>W6mbM8cte8Nue}{7xvjvoo zEJEYzC1_XA#Jg|HaLjuxmd=u*Lh}aZrryWA6Th>SzZbC`$3HSp;}|yY<|$^j_&m#P z2_e&GzSMHshN2ZuviV-|Y{btB=B~Sq4fz(#X8o&Se@o?2^rDeIZOJ5Ui&)z6&z@56 zx{{T_A{sGIj8@MSBmdzy*$&IiEHUyo+dV*+^2ZFOw!fcQwstJ(TS+0*bQD%U5W$2G zPpN}@7J~cpSaqxe``Lez?F$&inJtTH|D-^wF4u;B)e5|9NJXvv5mwM0!^EWybH?QY z7PTv$IV{O$;q!CZH6e(toteRmIREc-KqFK9Zyder+0Osd^5 zI}OA7IyByBqq2^(lyR|$URKGH$IcR#ohwJi10T?yL&=aZXW={gO^}i(N zZYio;WzW4Avq)67-HXzqSYsJDY)tYpMCby7p-d)Z>5MPResl2{zX#~n`n!F4#h8dLMM1P^7tB8 zy14%?`(t~6)Ss`Q!Wq7tTOmbuj&;l=OMxOZgV6Y73d~*zXyYDj>G}kkx^Ey>^4$pG@ocKBc&Q72W;m zN%@>Z-F7sRe8fM{0DcCDRWpQ!Z96^gO(k{i3GjTDO=Yf56z=FolPB(`s*?+lsiTRv zeXhuwF^BX#W9fDrpSk?%V-EZ8uwl!dvM~GaY_i@UT2pkH)eI?TIlXo4yKx#Tb!=l5 z*PXd{`ZiN{NugESHdE{9OZ4Bv3*;d=o-PFj(HG4g_CiI2o`(EmOz{M58Oy^+eo7R2 zkKj}7fvG|P#;Qaie!xw1c;uioDuu!hdQ$rS-%Q?i3hgRz=UkH{de-hv0o|N)K72LZ zPgp^N59`y3)h1Mx*v1;mjx+r>OL~+_q`c6SM7->2ZrwLJ#Algp^^w%gd$_K=r@YMJ zJQ>VNAb&o8Y1(#>9&B4q4fTuYTtFBV&l`ZZHMw+YLo<22ZzkU(lPEnYobt+J=(zf3 zdiJ=ESb!O2xsN8VE@P5SnoEZ3$MNUQLVEaOG<@g%qm36d5TjN>o7Ve~!s9RsN`K9k zac9=kc@@maZ7)5(Ek`4`pMI^iD}A}*Me}wnr?zFZkx%K4?q#x1 zO{Gthy~v(c(4^WGWOc%cI$h()(WHs2uOFp*bINH>=|c+a7=*5EBG}tyK`rSQm}<&j zCYNAJi|!Ag-^a$#Aa6Npu6L#V5V*mFe-}|MzHP@pai@?DPzW->cztvfv)eY!JAoIhH(s9j0W{Ao3XhgeiAN zP$inB&z$FC-pbHSukTty!ahS8T4DY*4I(*XM=G`**g-5Lfm?8qdo z!4dR!&{yhy)ZUCt+ZR>BJF_q+545o0#Ou_lFLBt3E^ZRn7<;c8~hS8SSXlrFS~h z$%CJBTX#j0sJJYKU01};uov|6Sss~XttRDwR5A)*PPbGeXf^+xjcT1j7GZXnB-q1Z zN*bL0HdAlcB69d6iLbNdaddwY&wuNZRQGKTush?1a!ot7l!V?edxp=9#WowSeeE_=oaTC8RONsrCk!6QY(`Rg5_Jc%51b7-Oa zIqI(zLB_CHdL}!D%AG|}ay^U^CDW+%gA#3ao=ub5YRKx%db*%A6_YKNqIk(BBwg4{ z>C>f2citPbmG=jdmxw6Mv^>jxO_QS+H#noIT8xf4Or_q7x>V;HOdk?CllNd0dvNSB zvvGJ$@rtQrnQw@R-QStXrZ7^!(N1no!^y2>8tsXaq+y(obMv|vRb1Ijhr+ZmX6j_* z^WEe>?z?I~?SxgACnINMKILj8@@LjzQq7E{=qKl?XxKTjt2xR>9p=oG237j;j%bLY zC&eb4(Q;XRI$Ye#zDFC<#Qeh~Dz%2FWf48B?WUOniz!owJBjWEQ>gb)D&{Phj^I?f zkP}5_E0@yA&!b6X*90o=aU(g6layMUO&@la(ZLWc8u!fAB1hPDJ%pbAk*QfW4%_{*lWTET!6caEn8su~p2IiA#BJYhqRP=@s%mZro%ho7FJamf=& zSZ2wce|>D^W(A5@yUG^Dd}IBZr)Xtv46U$hAaU*{ys=h+8chn>fWa5Ysa^%wCC9iW^4ZEC zQ5+ktO&x86Xqt@{g-RyS{0$c=@vk>2X`Z8YcW-je%%dTeH%WX$0KL`LfZ20zdg-Ht z(?RR;qEH6&rle5Pv|Li5a5`iL@=M%Ex;v8T>aLA6cr5qw=w3kMwGy0id<*3l-)QE% z!B}nFLCu~A>BHJxwB`OzD%ny+IwJ%cID+4;s>4WFa)4%k524ua<@mlf4xgqzM!%am zo$#ASGnTKX?^2yq6?}zCgQ6 zqxfw|KMrp-hLb1lN4JD6aFi_dh;KW2;Qc)QSWv^qIvVh81wB6cA{v1aEzm#PSe_fZ zV1jx&)?V$2@zGI;^bTWBPw5e?+{B(4KQMCrZ|qr4>^V1y-^pC)U?bW2si?#0>^ww% zFGfJrQkjDa$B}n`u-UN?H*Cu=v8ECaZoH!jk&o%&+D6hC63p$JPIGZ+oP70%p(@}q z<}TOZbc2I%iVntpMGt)GrNJ#t%V6K_83I&pa;AvpyX%#OlBGR*bz3Rlhima}ur8mt zS_Iz#IZ&IBh*x!ouyarh&j0I=n(o8lDmiWLI=5llcRx-qzmMON4|M$cW4Odr<4EsU zaOxzPujATrwtGjk*(vvomqy~m*AOTko`cO7mtlIVd`$Ru2LT&zqjkeVe(C1T>jT61 zj%37*bxOm$@yD?Kco`1-@W*nyOK2}-R&-~3{=RZPTl~$#>)L$S9!`drr72SWbH)9< zP}uEHXWxQo?y=<%t39}eOSc;lrgaOW=G-D*_d5D``7I4N69K0Ue)v4w52hi?Sm9~_ z`8dNJ^F#zCEr#>ENSrxy5)(UkA@i9FdhX4|G?#ry{QVlgtkO^zu?scc`DkpNiJeC; zVC?i@Xk;E{|8WQSR^>B(P?pQ-hAG^%FOoMn#NnlC9K3KH@l9E9ox1`q_3@bc!wA!k z4#$^S-SPVVDNOtn2T$`5EY%&4n+6WpeQ76LF3d#TLismP><)#M-%%iQJ9bAtz&%fu zjmE}F&%*~iI(`fT&sSs9!sF=rY$)8@Xv1@)HL9jd&YNy4KJzgZ@f|f~cl9N7rV9xK z9L$G)%51{R=^WE(J-n;-!*SYN$%@;7YjOv%aeFZ?zTC%_`Fq%9^Ii^h>4lXo`r%Ei z7CIh@#DVkflA#p`2lM;zdzOZWk`Egovrk4VVo>3qg@DuZfKDgSNz)t;Y$|Z#K_z1E z$_(_%zZkjlHV*E1jcF}@Vur^dOn>`Ka&MIR#)xXnI6stKE86oDnFTkwW5SKn2sQz6VkB$}!8V2z9MvlID#{NLy{3i&}sKNr7 z8CLD~5~~IancZ*1$k?4|9^Z)5o8O|x`WuKJp~4=T$~?1u0aktThoV6OMm5DDQ!^K? zBa(1d`j__)K8g0zN>H2>2RgnN=a)oK{>?D@?QoACFJD7BZ}yV@79G0Y%AAgtETzx& zqbNY;gHCfVlqZ>B;Nd&6XK9MH&YS6akRf|2%j}>+Yi{0pg^ZVY)8BdN)UaIBojK&E_3%5}Xy<@d z_)$0-T{gEyaPK%e8QYUgObz6odK-S2WFURYQw+KBkXlC>qHpjtTK{+$3`cjtt4h97{`;Csw)26!UIiyz*PQ!DhXXfN}I1TLwyRWZk zjoSlKsLZA}779Fhi9Uatmw^M#<JJ`|puNH&hGaV%~mHS{=+$t}H*mDwL3o8>%h?qW)} zT119b>*(c{r`9btG}HP9Bj|h|i3Nq$s2RIhB9=4hdH02s=Bg{tU<=8r=mS03F&4E| z!Pqt53L{P`bG`JzeLHS}@W+butE?@(b;-1zxAHKS=&!=@Xvve&?MNM(a!J4Dy0ufo z5Xz_lkuiu)T6o@$Mixm%&|9ru`?s3NjB8s`&;W*YmZWoPStcs+nGkGDZ@pc zNwyeUAZ=(A_3&Is)vsSryUaCY=RlO}oJj|p9w2yM1$-8chJpEA?C8G{P9+g=KITtx zTgTFi(XP~~<8th-Ovcf|%zj`G~`X52l?8iu17l|7u zg5lcF4Z~UnAkB6(azA~giHnn^LwYdXs<}dL8%I!fpa1ZrHWR2(l=&$G^0Zw_!gN11 z7^PdMKKpK65UoY0rnEzS&L2`Wy+VC5{b-kd5cOJXi`-`_wBqG13eVq4N9DaQxpxe@ z^lHGBx_eZ%^Bt94dPzqON=Wmv8S?t8VO`!_xEXbj@Az z50yA;(?o}SoE>kAZlQl9-}W$i&B#a6se#beYej*>%B@$bnb6Zv*(lx7nX(oD&G| zdOJ>m&c}A8+EPcl)MW~CR+(duyAoD51yE0;xzv~yO(TsS(wnm~!!V?gj&FKp9sH!h z+N7%)X;yfX##wWUEc|Z$eo6yvkHDd2|66Md!&2*(x>?qL-BZAJH!wACxSW$t zMd`4u^m3sRf2PSW55EtasQH-Ztt9tw>C`+fielEr(O|DFwEpmE@}4vlHIfP0)MW+@ zchH8Nbg$dW^Ha{Xp4hEA0$$R6HsH%18vIyL%hpTjjav};W<*l&;pl| z$Fsw6!?l>g>`H0asSpZVlZl`u`RI^y0N4D2sqe&As2KQ&c2=&Z@5|d#=P5g=VMG9B zjO;~U-1pJxR*7i3asbn=yph?I5zsC*MMPLz=qo%#lj{NeK_8wLqAGJKif|fnleCgw zP`&FHG7Wf7KJAZD#l4qQ8&*tf6NXUN{4DFKUXy5Q#$7adM_V@>Yb$ewujrEU4eGRI zDcRlHL4!R;(&`ZSZvya3#l?sV%f>QL=IlX@=+rEX-7 ziK+^CHSr|Q?1^DW3g^mS8`z62G#?zd9?)D%Xqk**8_ayBeF@fUGgi(m&aC-4w z{%qx@wg(I)cQJ2Zz5EiIth%%61ssS$R%UMH9AHt5g`GO}Xg1tW`41LTkAb@=FMzZR!EMa2kCC;IlAjSmTavLP~D@MlzQ+K%{rV+@i#}3$w0~7%D+Gx z=M>P)n;+<2={hnyxRx>}KBpOf@6w0kd+AohI?^4PMjsnPsNnoA%K3Sew9NL<-hX}Y zreOutV21_+ZYcwe(<409EWI8kVb!-m}_Kg3eV^TosFB{1^5!>t(Ld zp317XQ>!e|$qgh0&rXsrokh*ZkJ6yB{d8sMBl5TtLvvFM zsa>Z9pt?~|tFyrX$2eLzQD(9V3h2whd^-N1M1C&|s43DB)l)j+Lq89+ygL>DT9r|O zlFYcB8bjY+)my*4EB}TbtB^+@MY0ZVOMhy1kxlwc(vH!gle_<;tHDu}@i>RBe@>xA zz0zp0!3-)ZU4V0XI(%ev2&~H6!BXbiMn_m-WMv0-nJ1Z(7Xnea*nxhP1ks;0spQwA zJ#2rKSnKz2!uZL5X>7wM+OLs8%{3=zn%Z$nTC$h6nJ1IR)o}b%H^GIg@^4PT?WmkF<`0C1(KoUl zqz>OhrI@v^8+vTiM`d?4(fZ?0+4+X3H+ID6@+tT&p97{U(|G^dj%=7_&z-I+^QcXk z807o{o_8+dUE(6#su%$!nu|>S>3{1zoX_C`k z)dFOyDE-4y@cJ z844dZVt(utWIH_IP7wpxwaWsAQ5XL5qYfAAnK$k0&bM0(V!<> z-2nIP)s+j@b>}L#5JaSG#D*TW_&P5b##5Z3Ej>fOyc4ldI+Es1GJxNeY3$r`1}m3$ zW!0KVhQ9gE;VTbR6bLByPNthw<5 zc~N_zpkar$yL%$3To>WDYT$U|H3GkA^EH*#yh|~bhv!O0QA!Khe^+InZM7J>xtU&! zd_!};U8I`kQY>Em1h+C|S7yv#;q>;5TdXdZ$^O&94(&@<|M#DWv1b>-5)hqy9v zCk{?d!js}1_?)+c2UQQ`rcbX>X(o z`a%e|u=|3dRvpK;HoU2Xw{T@K5k?Fy-Li zrZO+y8?TKPVbaEa^1i!-U%qM0U58BOqVeaUe^qi2Klg`QXCEj$?TrJHJ@YwGx?2Bk z!^WQ_xRLuDH!W_U&r&~J`7jp&Q`(|%5Q;}MI&cRD|% zkG;~V$*7#Nwj`mW$}FhgN``*1Kjsc@jqU?Y;8kwOSEoLQecTV&?7oO&Inu2cejU+G z_i?v03SO z9U~K;VvXGn49Q-MCt6-eIq;AWYk=~19nkVbE;{Jng|^8XXb-W&5|553{N4^1hqmU- zm0^gz7Kk6;!%%*GF1XB7cF2Zt(3+npdh#1{X6oSOsUg@d`Tj1Y0T^-E68S#_X7tX( z%sC4o#AW=BJS1m8uTTR&$LM8}!+7yfC$_AK;y;bq+|m0AyIs{1 zi;HFMYWXw1?jrl7&j#?41;hB!@r|r~yS4ONb>(C6eR$y%nM=$yW-qrYwr{i$=OW74 z)VYoDI!~g`VTkx(8HD~y(MYcM!}$Zovh!ujslA4?S3)p71cT9;=xrwN=9hbNu4f}=>TC0Z;|cj5N`rt!b}xx8am9N&u0H0(J--yXy!wHN3%v$Ytq!B{->mFxmm<1xz2W2-ed zY`78k4H(1M;vCsKJb;t4&vF~r0+wVMK9Xz7A2K~z-?Z))=&rOs?yE=046{C8g1 z38!A`IQ5qczkk1w4Z2mK$E4GUDO2G8CY7RYXb^7NdgE|}HQ(9No0FAB@zg_pQ2&vG zIoD6aaB~q~d=|$u*UNOUR6gXgk4=ru%oMt5p(cbT)$y?m2X zgL{jCs{c6u;}!NAdWL)5XemryzT&67ukib64>&6ICMQbf-1|#OTxvZFRnlYP+0`E& zO)rrer^MNjhLWRl3U>+)W7htic)q(cKTPP!2WkfJg1;j;D$tLQ{dMJn{aXAnM~i37 zAIKK>f1qmlKn`zqG31@jPfz zE*tN@!m6)=kx?0fpIc(FQa6TUOiy#x!zAuLv8^!o_{iHwv=iZ*^m+UjJzjKkCbyOE zS$(Iq_|zvJ+nipbch!A7q?W90*+RMs>oCd4h^L7ia__wxU8-_m;WCAj`j6#*%cgOD zWq)4nV$7Fc&f;%FtoWVEXx`$rnH6V7^5T`t_G&x3gE+RHFmeHrUs2O(|2DSjv! zjAfq^xrM=1%r3rw>|ru5zCu|{@!rR))mw|!+A&x+=n~TJO2<_77_R*}oj1-7;P98h z{HEqQD=L@p#NBP!aco;Y<=l}I+O*^Rx#ReK?ra`@vlp-GY03*zC(GQmlK62(I^lF) z$(1G)Lqbt*r;oxGAN=tR4I&^|E`+XFS>Jz)^WQCb@kNV zPm}8h$h%m^I!>7XopzXIVAe4;J~62no3?3iahxK*30jVo{v|NUQQ-7@T1d6*k2xiU z@LuCjZHw=cU#HRd>bqLr75meV^IzyxloFd)b>%J+C+K?n3FWW)NKZa1W9qjzl%QTg zIk#2dcQ1~7&%Ptme%5g8*MqOMe~0Yn3H-Y5B>mU+0jb(5U~T#`?Djj2g{2R1Yugrz znVL-dUY?);*_Rk%;EMb$;TRRU6I-)#5p3GxnY`E-)T|kf0 zv(z_20fS~8!RqCg$gcPtZP3|8m18d9^&|P7-Y0vHRW>}U{UY8h^M6rVP2}C@9bIW3 z4sXw+q~7`o{j}bS69=Uq=wdYPE{VmKC9lyWpDSi96lM8=nt^p!)?=`h5^wykGfeI$QnmDU445f1O^q!vKIlI>GvqxD&;L%ZQ`X_s zx!)+6EB%gp9N1>iWHzx`%55B-dE@YYa%Q80_1T4Z8Eb}r?MzTI>o_`2y+}SI(`kEU zD+IcFnMbQl36INt;I4F6^6$UF{Yfi%{#uIk zcfm+nZ-)=FToB*54{jxYi3_ywGv*+^l;0=y@ha%u&kA+l55Q)rD}v?zRlh<5*$rND zNB04T8r1OOM;om4s2~v+g~jjdu_wrkou0kHH5W~GZ{3~KA6HPW!6EAQ&j~wAODM0j z4GwupPll%rwq*pO&7RG0_$@t8ZvWAyzunNLe-KURd4g2dhteUB?P%QQ4i(SI2u$%n zP3BH$w0n-fSC-?{hFjRJV94>?`k>FI&Y1G=K91<^#+3PTUc5`0=O5TcnRorjX+ulA zouWmJ!}`fPZv=^dRb=RVj<&wNLEr99$4EI}53WqXm&gh_BR|Jwxq2w{-iA@*%g}0( z&5-0b^_pYb0vu_GgQXTp2tF`!@ z9E&cO+H=Uz3~V_ipQ*;G{Kzg9|L%XroXO1)heGi-tS>^H_9M-wA9wq@hlk(F<>QA> z@L!vg(z#vD(P#EaSE~=MNPm#VA5A`xz6F%28)w? zXYDKg9MpxI(Vu^x@#Tt{t9j6`-SWM0ir3^Hz>DcY@ceoP&p-Cz`8|5`sR_PZZdQ&K zl2vV1^A9f;M4>_ExQF~a0|lE#lpeI>voG9uX|6Ii=v(vchhDs0s}d_EOvaMdlKI#59Q5wLf;iWb>*sc4 ztNQowk1*h_vCJ8o?RdQ8+g8bY@c3QMY#XwbHEb?$(VPGO94FwCp9{YG@ijKk3MdHi=qJO}2?V?&fUjiEGW(_{7V2z8 zU-wWb1}CG(t#BCqTfigaxo^^uNS-rvF3JaP$C+~(a2+FeTc<*i|K$;q9pX{sn1Pv`H6R@}yG6|V?71krK_ z=G-`r>fUbV|B9Q-W&fws+U(sic5BsHT;{yE+xG{I3+)192F0X@F zW>x~Xz$f@NW;f<5p26wKHxMj&srz0mlbMQBNU}PGy({m^-MluN9d^W^;d{`!ycD%G z6){PNF{J7(HW=Q(`-L6Ybf*J9{gi<$*II0i8_qXZ@5CnUKPdd9&(l3lOP5ds9zJ`9 zZ_@dtK0gRK&U;Y*IuIo{W3hN>1#G{D!B*}cJ~btSoHf~Y++68y39maVR%R&SO(b9CuoKoD1=e7OCuGQ6CU#cq(oqfPV-Vft}&*bdIcs8GI9wR!x^B3BC!o_g!dhx24l1=oa z*TU${D_lu##En(j9DT||44E-Q4DTN%`ucPcb$y45{vUtI{&_z^o>t;=H(lY=#+Z8z z8_n;hjF-I&8@}kfi5HBFWm|`Be7UXUoR*#7@xJHSaq&gYZ?0oo9c%tGc0BhR7{aL! z+lZ*{_Ci;Gf+&={H{X}DM2LlzxZT@I+`nWk^m+^tBL~Iv1BzzzrziR0#V9ec_j!>! z?}RwCu9djzHADo=H53cSnv24-reeQcXHjp|PJGwzCz{d>M4Oaf99C{Behjt~^9W5^1?l4DB{>8s?-}5f*HbVQUqv)VLRk%NK6U%lea8Y>^GJmz? znL|5?%h%?J;d6Y2{SRlM``=9QO@xcj4f0%8Wx!X0yYU5s1N`636u$D|GTXR+#zOh5 z*WYZ!?mL=zb{lo!y~a>vJB|`xivxwD%}()bbZ-&y)?WN5auSsh&bhZIkuX&PU z7DMmYh=mDHd7gGXPrsWddAbKU=fiF85SGaulK)}vn`_WNrot5=d90;zgLhbbV#BAe zV4-Zt!%RD~v1*iXE{GP9f6_yG$(-)RnN6VDzS#3n)&d(JW+$?ZdfIpn-Q zE1pmhyQg&*e!XO3^i&^_vc^KZT4X7n4N4R>#t+4-$>+tfHDO}S(0t+Ok|fSaKApjN zGxj<@mTRwQ@{rsS+@j+MJ~m#L_hj9`zPnG6`DDImJ$r(j7pxa*b87iVgSrSYvJ|6w z__4?69Xwb&MV?RG`PTMO#=8sbqxFRSygLYs5N#28$yp3tzE+&=vreq)p2S<#%Q@hQ zhB!4?a$#CzaJ}klKK1J)-a9CAW$+g?bW;+lf0V@cvcV$kS`Xpzbgt;*JX(x28!LM3 z9V%W`x{0dWI{b2=GWYf8MB*gQv@GVME0jcA&2A$3h7zkh`31kG_ZZY? zId?Cg!k@a&;3d5jM9MA$ac*f3G4OsQ4-SuJwSVy(vTO_syPkZ=V#qvd3^HyWd_ zErpfeRyf99=HF!}+2=zb*WH)?$DW(9cj;tY{#r|W%CpJ;^GfQy*BuQueG%Z@L@nO5 z#iOaxvDIfA`9@~*fgMSbJ08ZN&aIHJ_B0t-MbOd+6{OAEM&@!y8;P|?~|cu5rAuRzRBHe6kny8B5{i7xKdMaj?IpgBvaCZIZ#yxU-^Kq?@q@VYW zmT9NZ_dbM2eJ{|UQ_gZHW{n#of$}cxu>SN_nqX-~fs6aml0T9^t+$QK<__lILm8-R z`;iV#YYBzC_jFI=D`^f0qf1M>VOgLGDyO*PsoQF<`Ol5@KilxcsjdqwSxFN- zWap+jm2Axxk>Bl;v|TNOtOg$?-$pa|uTP|dLl={6b~I+hx5GxwQn_pHi{_?pG{=1t z_1fSoUAJ91R=5-$uYWhKZjBz4?q1$A(_{QQioj`7`f^(0=(pXcYrBZ`(&U{ zdlw#+29n*{oupwELK`ke!J=m)J>6pmv!CA}axUOZKrvWf_D@8qoL9%wrGgJMX59mN z_IDd8`3JJ+o;j>>dJrc&XCcZZ0+R+y&)cA#G^e!_VO$yTM-Rp>!Un(Dk#9NR-`&xyjq<@i@M-k-tqU?0l<{0)A6eG+!kQ0FR621aSv||4cj2}4 z`EeM%50p-qhyZHdvx#0lNT8@2Td2hEBrUzv3TLZs(8|}5r29h|Ga96KTvr7%RcB#J z=@?8fFh$L>dfY7Ejc~;c(vK|fK5DN>U;7-Tofv@g+dk9K!N({xW+RgOH7Ng^ zO*`Eq=-Yr$8q$1{*6uAM--{2(f4C#e?EcY;>4~)BmL9&{+)cLA;^|cIVX7+MPEVG^ zk(F6E7Vh7Oe4Pl`m?Tig?R)6Nx^S|O{z)5C;>qdJ8B)|4i$87bk#JTQt;-ZypFZNR z_64|~sGvecUpZgUq)sm@Y3BnsDp?#zlLuaxS@x#j1+p@`Un+0ctoE0 zDRi=93axrJgztZr{mb9&c*rV6tZlKM-uTB+%wswK-7DQ72gh=kssB+g-(c$hI-U*= ztinU5WL(S)$Fj~YSXR^;Pbb&W@NZ@0S$mrFx-6uCw;yrwYYdLB^M_?tXVf+>MPs!( zx=s=3W%HgIEgsN(d1q=`_ltI|SAfM_6I{2G`(&A&ja}3Y`5!u=*gS-0}3wNaw%wKvRadiuE=B1oP*oHts+YwLp{zr*BCSlmre*`s0Wa)U~ zo<}GOW=Q6&MiJY#m%OM&rs($04S!##A?V-}SZ{rSfS>yy`n%(mcQ#JT-EghRcAlDH z%>^IzdE&rew9(VS_a2rw`+O4{-#^LSKVRay%AQE}8jML(TB2X8SM+@3B~mxZrsMK^ zv);%S6P?>5G_4GJUMigHJ%sI_AA|0xKKxg&6Gt7FZW*UQRvl9y*|5^zc(^-4JqEx| z?mv{8r{aO*DCuI(f$6aSa51$Gj~7MK0=p;*Xq3;9+Y4nEa2s#-J;>W)nyAmpRv2+v z<{&1e;J<+u@VVoJ9&hb%PeluT`i?}_oGR|V?Ew!R^N$DU43?dsp6Izv8-1;&ASyf+ zHy4M);njR}+@FbMZ?+;jLJ>1A8RAQ;i}Y=L4ElbSPE~{Mh*$qW@kdRtQU4}At2lwW z@Li}hN=9@1FVb{ri|XYUsM6L1#<5ChKGYInE@{%EaRdo%^6>YDHyd6z;&v6PvfH%> ziU;#BXx>h|J!+0m@tRP2^Oiz7_mRBcQG9NNWd6A6vn_Eg3n!nPX zAC;UOah_KN9p$K=W$3ZwBp%gY$J0yx~LTPtjq&@#jbxXA2R(6tRzfQrvcCk2aFQ5Ix;;?Y>WLV7#Kw-^M1V<+# z{@pSde39<+Rp&U^W;4$V-p?;%C7V-Qa zHjz%RCK?EL!NPW=*PocdjInM&&uQNXLqyRZ6Dsy?uc}>oFlEc zJ@oV10O++Gjk=`W2!0rY=Gn!_o0^Rm*E8_xlP&Bo??QfMZzMZl2h(TgtP z#HrV~;`SM52K|Fcc{#3(U4(5%91(GUJrZt4qT{S~9JJ7a*J|tY!jxWIlmt#OdrT%z zil|4{MOtX=LqFOj(p8OPWN@$@octW%Uh0S~zP*rtS`RrornvBJ5MG~hz_bn{k?A^% zcWxffxyq&-xhe(0BTgc#uRRJHw(;-koXoC?oV2XN8K2P1qW$FH&m^BNnm$RJH-tY47d#4Nfm9dQSaTcTi2Ftpw! zNyqX`XiA5A@2=BuETx_J__&2gRDI70$Lsmm{PS$4-pmOr7W1+N)A^m=I*uGYUFh_l zDe`A+6?J)+#qqd8p+7K1xNOtrk4uy|FhuTV4Tp(IODx3Ee;vh?t7f7@t+t4BR~7Tx z4-u=|SO^7+@#6Q)*1VuhUHY4J`FPDnj`N?uyQhrfYpu5PXa7*%ShP+u`|8+f>pS*a zrY_FSnZZ44M(}@$o!O;%sTdXPCGOOW5l^hu#V8-?xcZ?cwzeeE^+GT4@OEp_$|sJu zuK+y2Ju_kG2NSQ1I2jfB~UP+@U>uDIngLQGb(;xHdIo=4K%HGKl- zoHpl?LOLh5OXtydHLi_N;(?Ez@ucmoL^t;<%ppMiJjkT<`XNb_8=_L9W z%@WS82|VVG^trqFveQvhF>;NSNK^aHJrio#e0eedD!R$NmaONdhS9wG$VAQydBb|H zH>9hqi1#lZFYLpega>SdhU6>M*_NO&D zdhj{Ny4P~KW(Ub*?Zq?Vwb?;^rRxIN-m|{9fCHOB}zzQv$BVI|Cm z*$a&>nj)=GN&Nm$%MPVVd}Qq#%<-*3%R7#&5j>IoqXu!cjwg3W8O z_V+v-IOXR)ekq-dx96vF_XrOzOAO;yYbJ4U3(33nP31l!jaMkV=A6!Fc|*7Jd@Z(+ z{yh&ZVr#?G`lgGyhY(Z`v20 zucFF7&VNIbcT3)XK}EQ4`oi8D)I_0*IcLr6&BFrhc;ny^BB+<87_z9ZsN$BqbH`Uq zu`uLi`?v8+|1^G*xlp=|t;Ij?E8?E(h_2J;u=7mm{EhMEuelrfkZ~Z-njOl9^Gt-V zy@{A>)lwMwhYAI|VDZy9S`1ipP+a()A}%NRiR3;3P#GPJNmWB|Zt+7hYW_de9wh@%G(#6Y<2|`PIDxx1-sf2zdt+x^XM~3_{*D}g-3Pni*ySlOx7-FN?7x6ZZg1y_?{{K+i?($VRUEZP1kwAIYYsr_|SEJ|g$^1rdD?f_e#R;P~ zL&2>VhO9k{F+T?)%1jR?S3~fi^J+*?1}grgBBNy-Ofq{q2UL&Gs7@Za};JgNI=@rIT%x`i!J8LSTjRO&XE@5(Cgj!T<3Uu@z9S~;*m=~)7gp|x}#)*qisX62%)y1)3Z(xi zK*Hi?%oDNr2<_MeRt?N2-Njg zgh7*%lnoP$?h8m4?(kEa)8)*+=2UO4x}D8EisV zM)pSi%CB@xW|;C17jkXVAx_V|%ufzmb5Ukb`8R6HA1fchbGh`0K2T+qodc1&O)>+% z?m(}^G@AG44w0}wJ|7F3Y0$gDvT6-V+Cdzoq&nK51LhHj*pL;$n#+o z)Y|XGnfVFwGu(us@7AMVuOQizcY{@u^p$lolbOLin4MaH`F+YT%~y6^8!UKmzn=Vd z#4KDC^|W{KAu{n%#Iridd91D^i^$e6`F)>K23q3YgR2rw*bkeGM#3$bo!ahv1_2XeCH)UB}0;36Yk%vErJe)K48x^fG%y$z)U zQH8189Q$_0q5Hfe*k$5^KkSaxZ%4vBCIeOqCsB3(t@Jei#Hh+s7}{|uW=yxjbGLL{ znfnFyP8;~zvz-_kD4F<2ozQ%G0D?`Je&eKt{654QUz~yv9g~jl!HekF%RN-H_cZQV8o;ghDZ2cp zEv{@Ejp~g%kUZ3r4YwSoMQx3t@%ueA!wp%{vJa=u=*Tv9!}0r(bn12vK$Pt~N`9e- z7C|x-_avQ`#MRS>XO8$9q7Eag1iBeH5P8$Pvmqk*%+@d-r+*mUt&YGxPmS%vU2t?{ zG1b>TCUlyDU$+%;tXS?~%(8IlqylT~4(53S-q3_K`skZ#jO9l)5qZT4;l|_ep?m;_ zC+;NUYuWT=O$go0NT=VwW+Qm71BB8GYDntBC6ipi_S908jqg6OQUh^q`9I zbwwizTga~dozGY_@jqG?AbavP8BnSogKt|xvH3+Oxhs>iGwmK&d3FkxjgCa;t6jN& zXb{gnqlA5yA;4Y9eY$H#i3@wtaHady;p8vs+u=Oj%=tiF4y`0*ixITp;%fLG(m~CT zG_n|Wopzh|$DjAo)oAw?1A8td!-idCp6G$U)8EpqW^H)Qtfi=#c{1x-MN9TM;AE@u zeENC?c6xW>98Wur{eFm*YfHGVbr*DBW+}6VKj3YwjW%7+(7#D%p#A3@{b*56L9U$prtYX48YI-g0hslFBW5z~O8j_F^rf^t(bg>i}lhOh)*uacI`) zjUU5C<5kzuI8)FPPkP^@J3(iuYU^+$Dk;H8a!7jau;6?PRjk_FK({WHV!33D+)HJ3@OEJIgso7&UeogR!rdBITaU z02TkCp8c1?t*seWm<>i({gv=4&Y*X1-jMsqOW3@lA9O~|LY&GQ=tWMWy^XQ-^s)}Z zX2#H{%kE@-SsS4%9@2?rC#ZkgVd`Hs3b}<0&()rY8ULF0#TC#L_m~ z#J5KflIFao`F@r3ExR{vE~>)mLRYr=XNjEOs?e)CN7a7%<>Yr;DDNuHyN*sX`|? zO$1*~6GuDk6aVI4g5QSLz>_ zwHNW>lR5U@TwYi0#RujD^XuoE`NILZ!~L!&4%O<3L3wS&ql4{5blYhx-dTi@L*aS3TwD zS03{jlPi2ml*_r`1FpNVhoip4u|xD>Zgd*K+qOFMD0@5Z^wve><;)f@zc`5dHf`BF zvkPxX9?TgA)*^3fU!gm>ttfo)jAK?_;4ixi`DRiJ-V$%juF@G0`yzus?mofp1LE0X zS5L7o&t3F1w-QDQxA=|L749%Cn_svE@TLKK_~QJ99K88IpMO4=ARVKsQXS(TR${tF|2O}1M35lho*V6*2vw8#C2J}vddWq&nczCv4Mh3j!$TO+=3 zyS;QH_mQ3$9a;ES<^#(nv46KAJj`PPuPb@ZhpQ|3;o>~5ohiQ`*Yw0#4MVYEi;{4A z*379*D&oDN^yn5o=k`->v!?SSoN4zMKDB3&bJ<+v&hISdk1!CoeR~Ssp1nl$OdHX- zrL|aJ)lx)6s0$stVrWU{K!frfywd8#s*4gZk7#hmuE#J!yq?+7;+OngQE_YsTz(!T3ysPzfzVI)fgU_5})4*i@8&Jd+ zrG@NxqMYxhWN`D~R9@C)K8O5W$&aRP;h)3KBK}n+5(?fRrCFct`}N~TbGvgxfn?vV zyADpgg_3CsqF(n2Kh|j_j2n8eedJFZJ<*w!ry7dAE49SiWGgZCPA&8%|Av2OHO@Y? zMLIpCGb?8azuVtMM9TgAoWM3>XO8UF*Vyroh{-%D#Gcc$q|^GtD$ew>5HlvL3*}E1 z;&gzY2yHP}x&bXGF$q2w*W$0?8=PCZ5%k~`YKseT z-9#T=SI=X{{xx_XKMN~%N#2^boMY#V39&sTa87D_)YblG7+k}9S4iIqV@PyC^T}0 z)ytDK??1`fTec6~qLWDJX8k`x9db6dnOW5x2B;;=6~7-EXM1cKuOLSh^F10 zaC^c~nL*hJHWQLjZ40~6_sM^O8Jfq2)2r{EFkaDyyB6Bep;T?jr|OLMPc7i`UP$lR z56qo@5Es6-!MwXI(0IHx=qTsv!C_mdyf#rCAtd1Xrld{iQ z;=9kxli!gKb(mn-yKOksz6Ab*W@D3jERLPq2E`TO7_xF5Ue%ScFo*G|)NW?0kDKG? zA2F;DSHlO@3S?IBS-8h}7O|3Z(|LF7b%HqM9xuV5`N!~k|13;4EI?LsG;8R~LH5mf zED5@frT_Iq^}$i*mL0+#XP#q{Q9Cj7dkg}OZo&Ck@z}EW4c^uTa-RMY*zkR$_QDw0 z7VUp28hG8OZgx!sh?G%cyrVYCcYZdwdXjWi>E$H8xf=J$r z?%tma2S-=PwY9LmE`F|Ct;7rrn%MKvwyZ2|Ju8~Gnc3ydMx!!6TjtHjCEgJ`aFVca z0(e(r1y(5fA^y&OsI32um0vmc|6?%zrExBrY9ktU9mc+6St!~*8M|8Fve41HA(pz|c2<3f{!_4vdNz}ib3D00%_`bKtkxl1Wcf>39|e z+JHG!ceC>wtvK)L6>N+hX_QC?S$FJVgO^={v(;UC5*p9;4U1?0VjWoHibR%bKM4nh z?x5a~RKejlDeQwll(aa%VeJD2>T_I2-d!=AcfW!y^HRo4-sSv!C7F#g-iio5>)00( z&AynKWB>j>!73vi&X*LYaq?>H%g-WKtNIxOr?pwv)I?;Ngwc|>hEO~?3!i70(fQ3% zEacTB#1`0bS9S`EP>W<5_fJ7ne=(CC&pS|IU&tV!P_UG9D8DRNPIK0a;9a% zTr5^n?7lo!^G*Uo&b@-ef)HjbwvRQ2zh>#HlG)K&1I#CDJ*7tbv*jMu*nIv0-J2Q8 zuAZo1R{yRsD4t_OIVURaN`xNTY<_f4&s5|H^J5hQA|D4h?+4|ShQUeTk0~Le4B?;+>lpT zn>33~Y0PH3Z(hQc<=d#Z%a;xP_ndv!nuq)yXPK>)G4!*Wxk0S%-by*7V^23auxnjLs3@&q#u-iQTBHch|7=IidNB&v3~WJxD$Z1}{~6nHj0TLj)J8EkivA#UOxGe32Ko&B20!Y=(5 z$eB#WRgXR_ioVJ&+?B^Rql-{-Kg}*rf6J0T^Y70pFShyIMV4YRzykb7!Q`zfYPj#} z;*P64-?0WdvFdbV=^z%#+@a31T~P<6SGVc}wZ>~!ocK^Un(+#(VQ;>GNSj~H%i#^VUj?o56tjlU)paBRLN z2n;a50U1W`twfopy*{j#RzYuKpdfDiZPu(HOK0c(CotF5V>Ur0*uF)SWk!UvEX7>r z4XY9I*1<%NcAY$e;j84EH77Cdo!gL8Lgh;ip?T<1-W}vQ zfVgJqpJvijcAF9Ax@r=$lx;vyn^_eIP}eUjDqxd^c<3*7dkhnM0o?*@71B9 z&JJW*9?RKT=1`U#N7DNLuxnR1>i2PWXU=rGH2Oab4me3#=f1KjazWS{>P?EAZ#XDl zn&wIQ(2lj96rL$UD(f_9p2bf79E-yA%}c3gjTn5Lm0^%r1CdV7^UBb`p56l}i#36- zU^q1#3Zp!mMO0=xlr~HpP0|e^SSMJ69i9j9dZinwte8RNul?ye&-%VTq(f4h>u9J~ zG-gfpfUS)!;3G$F6~!oa`-elZad6~2?-_siv+24fs?6T91GW3{IELqjuk!Ds`!-5z zl&1$glRPimgHAT+lMiQqNqIR?(orQG{w0GnYXQ$<{f5*9X>_-jq5s-3%+&r5wnono zH7^TEH+erGEu1X5C-L>jlMs}ZLEkQtSe6<2rg~8F)Nd?!(*-p4i&JWf8Yy;Y(4P4c zwEXTbmgGMJ&IeD!h}WEwlG>n>l17z(4Y6*r6sp~#aiHH8%gu)InY|^J^c>~Stm)9! zb%R6v4j5bV^VxEF+OeS!?kY=>F)t7|!?LNEcX1xxv!PtI4%WN(98=0V1xLjLFmK|! z8l|-u(^?6MN&C_5Bu#7PzQ9|5J$im?KlStsCYkrT)Ucq1R?E9m+>s|J-5x|g;zOyt zA{kKwD)c_;J3I7P2Hm_rQj#Z2A2U0VKe89s0z5Hk8Q_ zA0cY4P+vu@mh0HVLa^6n zGnAM6!n4~I*fX3o_LRVgO`Y+^bCaS{%FmZVV>w zO{P!36Y=lc5Go&>!j2sq!bT1pgyW}~?9Ug*?wuQsy^TXL_wsD0eHcZleFs>+cY#1F zB%N)t;jZz-A=L8s7J63OviIsj=6`WI=R!@E+pWtJ!fd3_XkOL2zE7D;SN zvShvE_1TI!YaqHh6XhC4Y^ZS;Yi-FDoZT{35KULuy}&N^S&t1Q{f& zuET>UQ>OdrhhVg^9C8lT3zpn*7sR)Ru{6CsEO&zwEB4cZ?&5eD*M>4Pmx(C3;>R9L ze!|x0zr$v=--0=He$3Tc3X+3LSe>5$Q9hCQZETF?(q2qlaDkC_n_$Le6PESWkl#x- zGaqMlmjA{8Bi~JD<4-uSN(asqZA%xxzCqAh?84R$UC4CmYXn6nWYDiu$gCRVu)eYo zTTdF(=)F$(Qsha~`*H++f13nKb#GbB11mQD^)t40_hG<`GvJoSqWhRRu2}|iM#m?? zm;-Ov$jdRzhQIfsw>aQ&)<{@*Dj{%UJ#<~fn7?I};KZpAwnQh2jp5m;WmB~9?sPCy zJfX*?9X-Q59%td_)6aqzilW$F=EI)-(O_%pYuSRaVywzGU2tdETc)k#z}&+3uwnCV zvJVTTSa!V{>$+nG-Ty?`hy%p}ht-#vsbm_Sd=$aA!%b*g6~_jD*(_LjQknfcrp|6( zA{Mo4BDR<4u=X=w1pT%{al(mb`>!&qRYq($_rqn|*)zw1AU0T}gQ?4QvT{KlI*W=> zU{t~Sy`=d)khAE|FK5-&VX!f+>852EdCpZwilxZ6uWjA{Fv0TeBFu1ml`TL&~tdehI3qAg^ z8wW-+r!8r$T|)zl_9id|zK>eptHY+pg|c(bvanmVo#)be1wGTJZ5^&-%Z^y!E2-;Y9nYZ|U|UQgE!zL#3$L+cidq0UaWd!G6@F ztir}xKNwzTB=fF;O^S?zxdNkkv-Ih~FCj5zMZ=4qGI_04cyn+juB?(k``<{^Z$FC3 zPYUq-Vg z#9auraQD!rGy4{!X0jLw|D;okP7MAja_90SIVwJRnE9ZJ)lB5imMf;@+c|-@ z*YKULv_FniIiT`g7$oKzA}rDtpC;@;^_n5*Ig^ZuY7(?zSsV#w-=me=+DYb*F6jEO*vap1pT(hNqX*NrK%CyQ4QY01P>a6|=YDxI z*}j)(LPMB;7Q2TXh; zid(~@kz724WDHa3_Nx@~|8#+BzXzaqmOt(f6hN#uohembWv%B7q1pYNZR=f+Yt|+Z zClLfiZi0!$|IgTZXuYriTlhPqlIOjbjtIu$LE}l8m+4K+?_iUcBn6L}f%(;Q(c+wm zg`0rYoqW&F9XB(Mq|y)9-86o&CkeP?F8hrSwN08rV{U6=szf-}Es`atK2fx841sX{ z6?A~o#d5Y+49_!0W2wYuY^`jB;q3+{bH$w#eZv&|I3NQj-@z!g)JBOoFl*vt?4Lb>4#&x1hp9J&`RlN} za5<&QSW>UAE2S2?V4m6u+!_7|@twI8JtKos?`PAKx<$AU?2fPD_9)zW3g+FpkgKXe z7VjrESt-&!$Ngkl$a~qwXQ0|~g}Vi^pjO(BZ?~_(qnvv03|JH7@HX-T+0d>vY9PGjAjQ*c@FhUwg%jKjQ} z(fMW?{5cD2a?)|6|K-oX)*@u@D2Hbqzh_$Y~9epLguA1Bb#bA{6i8;$Uei0CWNr9m)ua^eh5kJf$-b@2CVEXGRK9W zZqN-h_rzoWhbZU<@DA(`-t!yRgNF}eu|M`a7X1i;_UkrmEXr(#;p(GkwAYeQPL1yv3tufhb)55?(iU;^mxfHdkK}50_jqNbudvVSDH9A$WwpJhSxe=4JhtIE<%|&EKqk9^ztP_Kl;u!qsc!Mq8SK*s>2LT#8;p-BH)Lo|_Yn0FD`?)xv z?So^Zf3gJ%-`ME&kJu~26r5gm2s@{2!a440+q0z~{U^6#a+eC-^IeXPRiYGSm5Z@Y zPC#;7KDyea>4Z}fNPRbQ>mw;_rW%PBIgt#{K^f*hWuJIHY@z)zOyxhH!|*g%4%9(f zPKrL76=H=;4(yZXBRItv5xY`Rrp`Y%lj->IWe2j&dB4D57NoB8?_h5nUY)Z*@%!xv zJzIdas~qw5axyNJPN0Ja?eIQ#4dUG^uv5N}{T}<89kY|eezOdAB}I*UII@_~&4-<` z%93RF(dr>Uf z`7C$X;bACT_Kzv6%wsb*48?%G3hLv=L9l5G4RANg=%W%O2;lFSr+?8Tx183*-Qahv zdM2xp%3!k-V=sM$VdVckIqWKEK# z533I&Cy>uscy`Zfp&7Qt&c@@WQh0f)BR?^WRk(AHjG84}XA4-H;bbhEqk%u;2SeIk z0b5oaf&YywxW7CdM{>jPob%RBPESK{* z;dwpeb{}Gvn$z)ag#yZ?R={@WDsZ1YUfwEU`(n?st55UUPg65Ey$wR_u?Ad!UyA($ zlgVF-D4MfjMHdv~uU8g~j~Bs#b0GGex`4vE*^o;M!mEE_7-VO|XRq;??yUj;VsSYB z_m`EdUXL%z$~4dR49)$pgogSmqs(ds%D0V2VDBB6jflmW()&1IYXp<02cfb3FGTX= zuwP|8jGt~n{G|8nN7P%EJ;D}~=JP&wRXe;oj>7%6BmX`|!}3iH&J1v-&9Oi1cd#5T z4xDE1=Or;)pCRzH?qJMiDAJ#fLyX@)7H$~DgvxyGGBy+KxwkRgE(^COkHrJN&&?Wl zS#V)qGux4HjCBp%VU+cil@3=!iSJT$s08C|m;=(Lr{lhbF)cbHK|&MGAIMFG!>U6N z+^9jokHbjWwh<3~Ip@HB6H{$WViO9Yn0fF|=5cu;yt6cL)$AnORVRXZ_jGVn$mfF( za*^HS!S}c)P`4%rx86_0L2lQ6GeZyY!jX79>jQh$w*?7lTKIgcmDSxHgL)OD>~Q+ccb6993bEgPFeYBgdS&*H{yD`rq5EKahEOfzyA}gG& z&A>v=(@#+3T>9?^aOloUw(7ecmX+}x(&6K*JG7nkydMQg?jYl?Ep~nPRralRCi?67 zv#O6fCnn^vHdf9is-0(2he9y-VE|^dXtA6QJ*Zt2!9wo9QyugO_U?~SZT=jWWG=An zzhAH#Wl22Tq7GdH{#`odgNEPbIJoQ?v#_qkoQb305;OvDN@v3(@gG|(CJXgA130^7 z;p@&6nEot8*N+GAVwwD$xfrg)i}3HUDjjj|gIOncG#(gC|4GTxi=L_6!=`~W?z-;R zTZD!i(~vw>8==}$$tpaQHvdqgS(!@ICi4_hOK+oi;Q;Rr)x&0H3z`lr;d~VtTBSDC&SPA2Uz6;hBG zMKmcC^2OhoV~Qw#pBJY)n)TRjBtkmVzTl5TDL$8!VkQ_#AJd@&Jo~S}Gu&Ue)1+2T z4EDErv1G~{oPF>FdnLzG`C0B=&`ZI{vMj82v%}LFia7MJ?#7Y8)05+|P)-4v4$2rG`x;ap@5hEVSz2c{pH3wAAc+dF*|Y(B9~6SqLD2na1dTrX z5mKD-wQXG+4DbAg!MWFHyD328TUBIc)Uhj`)#%Azi)7tZ=#Vrdk<&wIW2+RskkX@d zA=5b1N}uni%+RtriHSwLX4mFD$7=HoO!iBLfk6@eBt#{%wGv-2Zp|NvIN-V|A@2sr)r1`JDHNN*xc002QP^Z)THv zI?>h?j`j^X=(*tlm)QYOY#WY#+c`LwqRjmyVxYm_c!s$K&NJVGU)fOZq=6)!{SN+h z4kD3vv1CCtUT=Sd+Sg^A$ytGC`b4Nj96~1l4u5x(Kz6t& zhL;SYU))3bYVv!u(r=_jbYgGwS6E-Zi#@hEkkSw1cZAtwqHjnYmlfzC9~LCvXa(~X zA?bs2$Wc>_mdzba`_w%+BS;^)9R^789z@O~n=tnMZB%q*KweK3d)KM){M0^7cqzg2 z`bs$XsUOLbuV8yK7auBp;h{eP2PXHi?h*2&yKg@xZVZR}$2-t??g8&-!AP#E#skg; zTCu4G<2)2;_-_^3YSRT%-f1x~`-%L653qMy4eCGI#vDv{;i{U$~aSY6LC^fbA za&OKUs)-&+r;o`})K3jkn;4DY8ZC@^-_P&C83+hV!{$;~G}&x|`BN!anTw+CGth9A z&#MeIaNzSb%yFwg%b{w>7R=y05;4lxSD-w%MwmoA=bX*!kQ;m!W-DXyt-uR5BV;LZ z!%#XZ`w>ke22(m1tlB;GfPz(2hZ^6P_9*m4Ah@@HsRUkTaS-=PQE&G0{U3xzTN zq2Bc~B!B$D>XqG~5#5N8kH&x9$vmf`f;*WCNb4Gmu0& zH$Zd3R=%4*jwPmtkQUX3+TUehQ?kMMnPFuTu#&s_I^ABPG5aNc4(9i#o)tJH%`?#o zHxPdACc3q6;#z$oD%U?}EsiQU^l>s2reDU#i5KzgRx=jp@%`HhaeDqP0ZKa3*mZgo zj-DF?pGXC&-Q`Sc?`cxs#AmpF>M|~UJO%athCq3f9o)Snxi__$4ew^ijF^ml}cl@LLm$%s)sjqq2zKfb(*9yatqPb`u0Y(+67Z#*)^hQYbx6h53x!V#Ym zn8fbDnlx8>^5r5r?(Br_j_EL8;f77Q!6=qb0S972#4rhNkGPvI{xsGHd%^qbZk*b< z6GL;N`E&g=OzJK|X!e11c)4O~l{k0(@b9jM2JQ1%Lz1?MU_QI?_+2R$r*^STv%a#C z51UwYc^`LGet~_?Fw!df#%3tpW7m(1a2GN6|7hOD4vS)Z-*W^8AJQ72)}be@5{3!Y{O>Xu-G{Ft-$n^~{LCDdt&F3F zacFglMR9l*+(NZ5eOf2;D<5E=-kTsKA_*%tJK>eVeQdSr=d(0Tnt6|5xW6u1{>gJr zR0*^XTt}koeVjB&g6sULkhj%F<&t-}MXDrwL7Dy;Cga6qDT=@J5PRhpqt}~f0*h1d zvGy9q*k0g&XQ#2EH5rDncDT{Yy>p*Rv0f_+uYFJ9Fn2P_<<}uFuMnHvZm@I4LvXT% zKf5PA!tA_Mr1-sr^{amN--|Z(Q~WJEHuW*fzyF)rs$63iURC14m}|Ij`!P=5}Z%oCU&ACv@Ekg*ux6QHCLtaG(oAbOB1=d+`+;SDG{vE^LW>o(pE8*~nt|GmVVOF|>3XXuF*$sm#_Qr=;_2ba~gYVFxlB^b4TlFc2(Pgv-1;ahRrkTz4wr+$e7NaT~Em)X3=EtS)5-yi=z3z znUeNMF!7lJ-vcSIs&`|l&T}DUX6VzJb=z_AQX~|Q#Y3feH!Mv3pG^))73pMdGkI?(z) z6MC~Q(r)jqB=P(Vg-jU1YG1z=ls@CTCJ#<9^c+u3Z@yt$T?^C}^<(_Xn-GbAA&5`; zAz1lQmDTE$VM};B>I<%5@F_1UnHxplKCdTJ$>(rw>OkCy2lzc>1B!WO(s-2xmVE9M z9Enz7BdUK29?s}PBdDOeYhn_twVp4Zbv+q5}SVuyY;M&PytZ$JBi{GTqk`Ihxo)Poey&!Y? zZ?g?mUNWGKmEQ%8a4}kMj^d^YD>k9niTm-o9-`^r6HN4X~_?;oINX`LD?!B6M0Om&Gm(;ha0N@S!d&VL+z&koayFBA+31q zK<}i@nblPpwnBU{tD3@{UPG$T!QZnjlk7=#+a{8l;6m=&zXd5T`M-N6#}AbVsR zqMBmyrlbjL?KyWo-JJaG0x9{GFSRJ9)BCuUlsNe~y{x)IFL{nv)nz5NWQJo}tUd)) zs&nqxbh>=87T4E3g0WT$1{+EXFPO;)A9wsDiKp|0Tdeei+mAX3pX-m~vv7F~8DWO4 zwTJ1-(QFzts)256ts-rYc)A)GNmlL8>EE3%w6gO#T~3L_y91dBds&RRHannEYYo5C z+i~|*HirFa!o#&^;pG%fxgQ+qdr=~}T<3nmy_`?4@ElzX_9`Um8&AbJcuiXBiQT|lmavs9{on-=f-gozcRG;`HooEoA?NAx*+twx(Z z{<=-KHdc}NnnrT{8qGUSsbn2}iyrE%#nBgwp^&@`GaIyM@`)ML?7faO`XSW6tRd`t zJW80e(H)YijtET*#*;lmv7+-QYttJ8omvryJ55BwMIn+cCesF<;ZIizpoyN!!hG=_ z+EPANSo(r@&OAbK$9^y!S|LdZZ}dsW*p$q^2XijO9J(JKLRHfjl1b16y2*VIHjOFx zz?rD&+$j`VAtDTTaFtqaiVJ0(6rma)hJ#mUz;Js7eS3O_Bs)HlxnT@NAB`pP06*Sk z7sI)rA@GsV#Lu^!Uzt9NYE6i`W2FF!uXg_}-jE>#wO%^L);3HRkh+Sz$o&(alK7PqG~C3RWxq#nevYS?o~TFoDOJ+l2@A&=Wlsrtyq$60$#5-UnC_Z#5g-oem5=!}%3PAI!q z2h)OOxT~@RA$#6nAgmEZ_WXBUWI;Ei@|eQF5d3zMNBO38*wy=%tyr}SjgKF*`nRqy z=^TenrbM|@B`Nk1BRA78-2e8LnT?dh`JKB_d`^W7Wc%TyDdg_Z6jY@bV@roK{OgAD zd&xT3@>%49U<3Foe`K!g6XI_e(A|e8`0Q;ZlDn_Nr6LmgE)pcNxC)02O=$eLCWJQx zVvTVGDnFk=cF$AHvoofP<@edqE$b0%5CQMeL9A!%W_D2HA(IQ&q`Rm2tB~(>H{Kiz zf1^0OJaGhicV)0+d7@2qsQI}@#uKl(e<+w8->d=baUebvk<@dV4@J^#I> z+Q_M~gYx*5_-C5W-iMa48#$d!arifOxmgrbme0dQK{#%V=3e=E4JZw|!dzXSGO7Dx z;FlzePKjxFan%4X7LKIn=hSGg=zQugOQdHy$0%!37KQlPVB7i;$W@Afv)TxZP!GkL z&gD3q5z11Qv@-R{gJHkP0wTQYT*mzXxBgyZFXMUkh4*L+JjS4I^-&fl{)atU>W90Y zhf#eZ7w4T-4qE2~vk{Im!lT~!zWGcV6A@?@-0ZwYcaT76GccKJsZ`^3v zlR{>A?*~gy8;+}OJV&(g3+vjXj2io1w*T!g^l-n?q#wHYuYNWpl5Ak|ULFoP;Y?Xx zi2lxFP>SPR6x}Q|?~a4gkuhjpvj9`lmm{@w8Ll@U$GqL2+1uDq6pfC+OV48v)60jl zd<@1G@+|F-k1RXQ9jdZb%r)c|+vxV3&FPI~d7~28kr7P{Pm1u(EgU1eT+!nl$Cf^B zV_E6F%(q*Dc9{*QUkb~}jn9G?d|!bJlT^{m=aI$H_u({OmzZubcO;%-`)vNP%w!83 zo_QS_shQYY#P42vd*NRoL#zAkVPs*9l;$*4??{FaP3ZT%hO_rl#Uqi`@j zk9f<@e|v%SU8Z&)+zCBh%@oV8Oi{!CLO(=-{%)l;U{IyvxXwOTq^GS}c4oONIG=cqUnvLaon{*0ihi zt2LVnFDyZP>1Hf-;%C2YA8FVTN#QbaC81>2R~Ys_hQtTX5?*RT>sAh<36;9!$#V}6 z#b=@K`VMr?kr$4gI7;Yx1ww4QWU9g%FOo3$fpZQMBO=8o+j zeQ_i;*@Ki6577nwnR;^u`q+(XG+S5riEt{zKb| z&xj8B&VN^-uM{wOEG~wyD#g+fzw*$Qzor=RVC+>Lr)4n{Y-yAAju&;j==5 zJU1ASyhuM<4`|c47A>;ZG~!vBG;ECY;4}3o%y%~-tKBN3bH|u|eH|p6t0peo{IG)> zU-V$qx2t&E(1JSkYUnIUg2b=OaC{U=#?2wLqP3hhSB@be{lILaU-+Onj;u|U=;AkZ z?jRZ>tSA^Ij4RL(dPz>DsXrExr2TRlc_RxmH4}M%EDVMeg%4eUXq&JY=gqqC$Mrd8 za_-GXsYs~#Wusyo-+4dRp{dqmX?5rvGMOPmqMJc$GX=D;{|S=#zBtMF4AMg0;8*As zQ21lS4u}cs>h9Cdkw5752NB^$*I~l;$RBjz-AZb7SWEV9&a`ylBD(ohk%}7a$;E#I zMgnkaL&$=MqPdfg0I|#{p=6idC`FAJ9lC5$Oo!_;;`td2daY%XlITo z1w3(~`(-lZq%2Nvn;v4o+>kbHwIB^oB^sFROaJY(V$GjL8w+(v;HFDW3eTV)cLjeh--k?BF}*pJMf*LH z=#xSWJz8*z+&S|zb&4T66oawrS`%JGzGngYQLs~*hC;s#Hq@tpsjhphP3>L1f^WMjbHqI<+3##$+n)H%dCzwX0Bl2F7Ox=!9J!KV9T(H7&}yt zepP&>g{-J=#0VXwOQD0)Pj%O4kB^mam1WnjdP;h`#Oz#akVpWn$L}Ve%^rb zNI-dw541&ZGOvI9yyKz;xhpal*uxz@yCX5UKLiiuy-^eIf%>~A&{^KcmKgW(ZTM0Q zWLx81S^{PrI0x6y_3%E=In|Q{G|J^3lX9C3qg!b#{qA_26`O!OH6t{?8G|6PL69Gq zfY?BDSiQX@hZDxXDXO=VARZo~@p7!==lfmNCZF=^6L zH14Tq_l9!U?(rgKGqjB*OK79+#8AvC_`~FihePWQLF9lx(jGOl`TAqA|FH(t9ZWDW z=>aAlyNA)-Ww+>2J9~b)kw39FV$h{eto#npV|frI=N~Yu{hQENyc$cZ3UNxS9K+&j zp_BU(+cxFH@}o5EY&nLNTfcK(X(|?*Wuw^m1#a?ZW$U@!?8y59=GJ+Ft@pHpzkwoB zMEvlp>n|ID7tC{$9Q22VyXprk}}Z z8{rM_dRa_7tA<@>Yd<;;nv(sE4d#}c8})w^g<@HR~wfzlMuJ!En8T7isaSp0P<3{IMkkH5U|Y;zEd!p6a=SsuGQhQKv{94uFI|F`B@?zI?ph z!^Y#K+aaDa_D8Y)17n-=+xS2PA0A6USQm`eZX+)E_I2dxzDx`<>G<_N!TG88tasS8&>x54x4 zK?pC>#HBkksWjV));u$$@Ibx~UT7hjSe_=b?@12Krfnj@^nKf8S}c1T5%&CDZIysKqH02`rYqE6dYQIr zo*=FJSycDn0L{q?qC1y^Xx!fUWcvC!Oh;8gVq5{wFWU*boMs4Tca0X-U#q26RYz%z zc04WWPvEQ?UpnYyLr%u)D08GbdGyaEgZ2tquq~gm!y{>6*dS`PDo0WN6}Xh0!|#sm z=*;9D_lB{w!D}h~Zm=cm<1rK&ww?kM^~tenKN5}h@Lj1N#3OR($-xBruwXm=$^T55 zDT9Ppbp{KImOI04D|d1J{mZ=di8Ry1Xno@qln3s?)%dlzXR3}miwFw2?M?0#o9IQK zv9PpPRyev}OenV_m;(OopjvA!s#!~9d1x%%_Z>phALWsXO(Z45kNjW%!B@Rulpt5m z&m{_!t2lyuU%keiL88LYqhi8pkDFA$?;|I<-_xu39+tk}LyFo0iXXm;5^v6-?l}{= z`$LIZMi|q>IDL8|FHin!0;zb18%e4vkle=&H1F?L`gF#PO0Kz5a-}^bKNwCckBHE{ z&IUBKzDC~933zZR4_mG);$KfN<<54YI|uD4?p!5)e6xXP^%0a+Jf;^piIl~&&~=MV z>Gv9Qvfx?Pd%teu^gL%cHqM2#>Q7+DPgoo~hX9eeG&x6w`X`Ja&(v~y^e2SA`8Csw z7oB9g@ew8Mj;9mKugD_kKFxPtLe}ew@YFO7o=v;a&Tb-ubESuOB*NA!7;m3BAX;@Q z=JH>_WKSzP?thEcVWRzA4Hrs$D+D>oQwAotL0DPxpO-< zH~he)F_n0F`zz8^%HX&E5I&fUpkn%i+TI^XtQDnObBEAFo=vH+w56uipcRsG5WF{9~&G-)comwZpT0ion2WQX}9HYyc+To>|g;VyK& zKS1|>p_V^eQUq@lV zt8mQLccgJswvl_XJ$;W;7Y>ygCNz0oN}&r?gqevwv?8USf)78V?y3hgSgMB3M5oaW z-J5j9-~q*~4I=;k-IS&tOAAx7g*&R^gbf8dg&O)r!Y4VA!p%>zh1S2G;_$y`2w3s$pLO^i@wl#VahQnk)jtCv3mZve`R-IWL5{{M1`7X2(Rl|_{kCD; zC>cqSO@t&Nl=IxzS3^S+rL;+urZkMUpP5mKmXs06%qSJ9kfJF`L!~rODN3XwdY|`? z{;d<|e9!aT_jP?fMGHfPmrD){QETtx$HAxg5%CC8m!BatMpnoed<&~pwGbU>10|Cs zD5{DP+@2H)vvmW6^Yx~}`VYp!v&+`P?sOgDUddcR{D+af_YX4p*mHB#VR9cnBP~ex!-L06;$8( zK|?tkF;%ZWI_%$K;JtiUS-ip4TNi{k*@;5g*vmri;%`*+d?2=nZj#J9KO}_s;a0Q_ z{B||b<8D5$tG!Dz8%r_1B^Sv?7ct3yu22?bCDh@(uXAC+B08F5Z?`* z#uW6v6&Jdm#bfr7Zs_>UgSO3HXdH8d+pjG+#P4|eVr%hex~(v*eUc#k##Qk0ArvVp zBA9n!e&ybWfowF^ML)vL-yiwzYq$^>I7qmj9gK(UAimhdz+^xSUY$)smP#lN42%=} zvO|O_)5SvB@O0#?y^E){x6wFcIyU9Zz`22I_!;~@S=qm&b*aBek24+(V^Yxdr3=w# zl+dtNhPzMIpc6DfIP5T4c-reK_`9YeVERpXF3ZHGd3`iGNDVZ-oiKhV{EwR9QI7=d z8g~hKro3)qzgd9UbfK2}jBYI1A`}f+E4Y82A)GD<5`I2gB0O&!Ef_hv^Lv&Uq*Q*= zwvBgb=M-fekJQ7)**2Ik|2CyGU7^n1zi5I(AAYQQjIgKQ@yjq4{#_@aD1RAIQabQ^ zW{368ny7R)$J}3|FABv<%F&%0%1nGZ>U6Aq?fega2Or!0r>O zLYvud96M$tEOd~D&hmCXD{P>#^ZN;QGT-@LLQ|OUwM4M1-XJ8b+9ho3I1OLhpZGTR z9Xdi*mAf+e;ALDn_^0AvfRvbpTeh!?F z%D;D4;C|T_s2*^{%};Ug7?Xn4^U|TVLQ;rWI!M@YtP$q#WQ2hxazdx(FZ}L*3{+Kz ztjH|Rk28nm&hcpP#}LAvleX3e8k6{mYUA_J#F<%R@-O4}>3sUUvWpTHl~H-Ag>d_| zgz&ISU085c746cTbARJKwLW$gA|Lt+>e0J-O{A4ZeQczsnLP6*5ski`r_h^|0k{8# z3JrbVkm08zMDe+@`>;P!)N*M2z<|fXX3)QiHbtR!70FUXUzH+3Yqg;a91T5&3n&a z;0s;o&XC8jWD)$!_L1hp4!VEkAMLD^;@oOo!PZ1Wh*_3`17(-su3wKq4NaWk(9Ann zYLI{Ljr_@lcy9U}QhC2%epp_Zwbe%Om#Lv&3g2n<+8*xnP>1hYFUVb34K4Xsl;dAP z`x`jhUHc`Ca^+d9U$c-Km=0gd4@mDFDp1OP6q*NPZ&WJmJ5}xsM-l|2zFlG8E%=DXzb~YdB zHQCV1|A5N@^YFzf7$ut$5HWlP4qAAk%YQ$#KK`IpibjwS>0_oyMzEf%BZNHA7sd~D zg6~@wJgfDDAIqWSNOAP(-l3_LwRAJMnqm#L&`)S3yUE75oxBVwZMW!VZ8hnAlf}m; z!PvU%8Vt{U04|H;VV@GRk4=W!MoAb>@1z1bH5eXOfro@KR+cdQm?|NxpQ|UhD47aQ z6Z8a&&m#qcZ=;0et;bOxR*J^-Z!p>Ch=`{=vte3|Evv(5{PQ%La_8C*#x)SuC~UM+kq>4ynHH5A@97^?j< z$TI%{O~3b>j%T@If&L089Eir^z*f4O|A^Mx)zR1P6hum2g@kh{UI%|7#lgeyYPu!@ zlqCd*);8p-_aWZw18wc6jj;dsQT(l;nX()4OM?(mIT1rH1R)^$0>UFiczIMAM@|YD zBQXxj3vIDAemxcqj75u5BK$gUW3=B^q$Y)-b;)gDbs6@)yok9s8!=_E0u_EZLGiKy zY;@m0mKC{}4a|SYR6h+S-`q^L;`?s)y67XF=_iBvf6matvLHIttArLADN31qU!)=@ z$zs>EiH^kFWXVQa^x#<~yA>u+{jR%FrOr1Nn_+}yU(eFLYa&v$H)fJN&r-SEf!X}f zqSOQXn92f6_8>`>K8^@wUMrN?44n|RcE~2$^Ei_J`}iNb|HX-^u3&81%XI2J=tal5 z)A*}mIZL%}XPsks4`d;CEgdwYIjT`C)lQGqZ}ec+b9C81o^PxTd__uaO%&r4Pr;jc zwqv9W8_j(K5!LTm;l?v;-I_Yq8_&D<7pytYbqziJxRZ_y)}ak=2D7)G??tiIrx>D- zv#!mytm5EWc2WBpWsE#dmHH>Cdh{)Ju;eMLSQ*61-mGM)_4T6HYt-2AY(Mfb^I;Ra z#3+6C4vbX@!)VudIFvV+XUbCq0>&)K!tlu~L*T-N$tN z>RE+e7@M(vznGrt_4BS9n@m-cV39H zKl5J3N}j2>Gm)knd}IsjIIn3`8{1ZF!>0Be5ovwx5M8}`ihZcfX6El4nIrE9)hOqY zky98M9KB4gHzFzb=W$w*c99aN@#l!;S8}=&PiNoj(@&$5yw~f;T6_neuO`Tc2He+9=_a|iw2Nu32 zj=#rZ*!Lah*x1ZScHx`>9bXc|hHuShj`@vjaaKB;`11qXlkH89&MA@OE+-nXCQu~V zoXHyhPNS%u=IpB1HnwzggQ#WuB3e-;qThaZ**Vc>+P1xO@;M-^yrxZ-7YDl85{J;R@R9EW$)1aeOKwqyA)2(a!=CoA8f;CZyIkq zgW%yq?a!}a|BHI;d^JQ^8?Qz3$9kE}elfb!6-r}uHq(^lmq=;NJX*3qM2S8TWE^{f z{h~1Ts*=xpm9p?z@i6+gb4zgUIPTmz4$Twgg+j=mb$uGFU z7I|+a#WS1f!+>zI)nX(yVH({EIze`c2kG%qE2W>`d3BUKNs>sZd>Q>XvW?y; zU8nHOr(}QR0d=bL%*e2{q?lz-Gsagiot9L#;r1?OUYN)_ite(ZmhG%dc@*u?@ut`5 z^Jqw$1>Ik^l7=RZr)5fU?BA#VnAiL$<{r?=0@}usZA=dn&sTvlKg%(<4ti>9NIhN& zR5s@aIk_j3P3%!>417raWi05cn>)E`Xwt0R_4LxZf<~CWARFygii`Y8Yd<}v?Zf%^ z$W=pX9XW!&Y+i#!DPCBzEe(2nKE8X=GD=HXNMk|+m|pi`<{ll*;&kRv33oNb`R}1i zkNsJ0PYl!G&)#*iyNI2$q30KU=!3W{y-gTMcjbE7R)sI@Qo{?jU1b~n=i^JSvzE|6 z)c|sQ?MLY;X%yOXmAdDGTfQ z(TSNaNV)YZeSI^H9kt=Sr4 z)4lJj=}Puaii+DsITb^&t#Jh6nHwb4bLj25)0D`4W^vI=ad)~Yir)EPWq)@XxiXSU z_lDE-lzkM@eT%*^LW;+Y1}Pxb%-OUS1s)N z8lF4W8A5F-kJzyxxy)4gC5tKj%m%R@_Gz~q-Fq^T;`0|$k$)I<3_DNnrWMeX*b6kj z?>TL8;lAjTT@+lSCy4WV*s0}K!eDtH*f;LLjD!qKe`-d>&NIoK@004v5AzwDCpDg& z$+-%tRHFHYhFJEH!`2=?A8CZk$}hP6Z9YxljEw$2c?Y?F7s^?Pja~R`2?Qwa&BXb&>@IhbYtP3N^ap zaGCvunlk!TDwkQ8$3%mh6?SDDq| zMkiKmr1B$C^gQ%C9eg+jj;XqM@2!q12_sYtG{oDlZ`lWf-^}Sr6H9+y0^|3saCzMV z*HVzv-yQV*^91@-9ZJ8pT9cd3T3X~~Po+MqC}O@d1vhRJHux~8dl0NN}Fs< z_rEz%chLw$IO<_y!FT%5TuvWb&(Um!Ka>$;Li;`srw)l_q%D_Be#T`KXY!G%7Bu3A zLNPS}mO_6I5kVOpEj!1h=09g$w*{wv6}proNbq{KqPI_^6k;*GW>8^B;yb;cQxDNoM^!uu&_T+Xv2ZQeN)g+~ zQ61+3Z0^UqO51bj`Pg$b?Wu`iHN{HkPO=c>&X&OLY&uSORpF*=JiRjE&hz10=z;Zm zQoNc%VP5;`2`ST$86zlsfFJ+t9VE-yrztFlvl7NGq_$x$+-)#}z88!nYZKnJl{!uj zmtUZ%!-8n!nE?8{Fq72mACuSl^Q3z46EzNtCynB3bTrhH{7u5>O+hm0?q5e9T@G~0 zU?J^UwiS!Dj^b^64sMJ%2ispKP`u*^#!T8og*rToxL_X1Bn-gm*eB$(U_Q?R4-$^a zDhe*cb%i9YnIx8`NtXjmY1YIQG$(a0$vZK+&e;fS_LlKo*?J1kxk(EapC!|*RMMz3 zqKU^8$*I(yym!rmUCVBWXGDRDTd9T5>ZGn+p(>?a)WaDm*B8W-TfZ&%6cmZh88JBR zoQzMM%OL*wBmxEwq=;MGtMbF1hGhR`F`SoWYhTY?>_nv9GoI$u8PJG_L1cI@fXr2b zDI|CV-Lki$eZji4P=W8HylUZ|AtQ{)3nIz%Gc@GSHF`DrF9sWG34f$4gs#b6<3C%sn5xFWX*AUO&R= z)6*9;t)-WuAIb}7=UqU?Fc~2xe<%+8xCSquN}O8N!_G+C(aL9LwBgWEpvQcd>$`F* zZ6BFRj>>$;v$C5NoSs0_yFo$9v9w}g3(1{oCAHbFN$(=h!ddzts>m7TyI0^{+(!H^ z9?F^UD^Th_2hxvkQB_+4B?)z$zhViK2NqagYKR}dhrl|CcPHE?LC@6`gN&T8zI-p1 zwp)>q=32xD{T4;n4m}%Ipfnse%fWf-4{!CKRI%3(e7!*P zwH8v+ry4SQsYJDN{HPyiWxw9^kL3P*r|eR5Jejv2^QsdtEc_w{D+klPGs)a1YJwk} zyRi4c0&KRMi@6^(Xiu04@4=oT4L(=S2pIt-Qw#iy2}Sq#6jbRI1P;#^B*1tqbyrqhu`@>PQT?1!FO~r?TX1W@~ z=Uj6Q5LnPgaivli$>*+T6NoO`jHW%ZX^nVSD`E zGy{GraZtH1hVK83rkqc{l=3SWR#GYGsjo%>=Rj6m;7n>69Wpk2MD0%oVOz;yWGg4r zulfDp+1W%VUbb)tmkY^R9j0RWcf4P&f?PfyR^hYbrZrJW8C(H_#R<^rHH7uE6bu%) zS8?nYh_eP<`V)?l9oab9?Rt`&7shQ$CudcBzNXW&4= z?}0VSCHJE#*#Wh7SLyPW*Oa+7kIMeU@EyoejPxl(hE_WHguSHF;*X?R^$4GfYY?&E zG8|Oj)7m$jKP>)-ij@q}xM(!G9vI?#g(V8zG@(y;LSc-+|)&` zL)9S2atCC^Md zOgwt&l;c^RlY7Y7-cqRf;)bh*(Fh)T1#gr0!R}BBuHU-{=k|KegMCL!Hdk`Ta}*9$ zCS&sVi?BJmh2Kvrf40~TareVoh z)ZagbLbhd*>E}LLeTy>`Mw#(E$VnVgOv0!1sZm= zXu=$bIh$dfr4#BlTSH!N8Frd3hUESVl5UBjC5qg^@gfm=PoH4z&r;m#Qpc*zjnpnR z8P$fjNs)JBW%hq2F_m1J{hD{vJ+DxL;W>J<^b2}~1#hp>%XgHsOl;eibS5(CTZClv?r-|1=z;`)IG!Cew z?5iher%wrIDq_XdLYv?bYAZTsqVc&rkts;Il7ld{~d}R-E`VN;49A> zmD3M6;B7-BY#s(cs=9{SAKoO_u4T&q# zC1Oh&cPD$*l6}GlI(oB_&oM@$YKQ{bd}Q!qvN(!WK2qr4(by|ka1MhBbVm2k?M;eE z9xR0`e~Ii7h}O{Ur~o=qQrnNq`I$X+p$-E-+?nr?@f^*$}?a<(Ix^v$$w zjw9ucl%^j~%&B-2#x>*u(7Blwl+&-AHHMj!roBMZ)(j`p)g!6Lg7*We{aIgY150uVV+!e>WVrq@ zjVLVPOx#J7oqdGHdAui+=f&)VHh2GV-|VV~+;ccihk7fj$al2_Mg%FqBJL16)|Rtd zG>nF+WU=_nNwh~_i=N17ayIx|wq|QDoB2D1?F~7=1Ov{7*zUu!Lz8KhdOS&QeN3+k z57B=Qf+=N;H2n@pVejrwp{u5OtRPsOpr}lz{|%r^qyDj6a}z3iyMm^t<}>G8PE0AI zfjNB~O?DyQC}!Au66<-(qA!gggQeait2mEz&hDd>K4)4|dx zHZNi2ZQof!s2F*7e&t;3H6${)&fIobk$B%Bwr=QZl8T5Uua_0{!&`!8#Kf~o$zx=D zhx^C74avK?lWjfB$dTuNb(U2!gF)e>{i&3UX7y3`_xYS>!+HJ#YWZEVfc)~$kz+ckua;2p&1DqTtqX50AIf(8OZhi9(;6c^y1Q73#6~=1i?%&xdKk{n5<|%6 zXE<3b6oZq@5Dc!)W_7kxY0c$Cy82#$ch;o&J+Ozp7W>H_409vLf7$GaH&cjFpf6t^GM81unCtK&X0b4qZWP{RjnBGSw&Wp_+%tiiTMv*`577ZWTYUJ) zj1+N}eo8E)+z>_jxm=Pe?xrxwOdYzBGlJS;rm>-^2ieL;ZOrTTJ(dzVj2=ecXTN@4 zWuG5RA%p!F$)RT!n|E$Eb)OhU&f=Ez*83Feym*h!PS)hJ(8)AwTMH9!UQP#^Jeixp zFB)xC!0w+NL1z5tM-x5hnB^V1crl2+UI}C$B2-AQ$zs7@F0xKFb9!Q;N-?jLC}{Bk zihFBMnR^|HKAX~o%y=4Z6-g&~f3w5mH05esAh~)qY~{#GYL{`kifA{g{1}T#BMT3%I8Y!Pu#)RKguk8RgGu;Du+be2E6VoS{Kw z$^B^m0ejx#dc$g6I$4*@MHa@rlxuByHrKOr)yS;Qg#lAg5U+tTj?9Uf0K5!s~IV?lT2|Y9&;OE70T}*hu zJ+xQ)QmXEjTd9!M>s6V#i+9Td&uNr|**@F8proX-@KFCt4Y5O z`O~6bPBd(wJ;@AFp=-^Itd}2v*3QM4a_l%RMWnL%iyhd_x-V?#GHIHgDN8SQ3p8o^ z2-@-c5dGD;!=6b@q_mySXvKg5n7B)U@8)V)Op85D44FnBd4kdLXc?Q39mnUk2T2Go zqqbMPSMzxTt-g7T?w&7V=H_3RWwIrW_Q+xPTob8^b0E~MX3)i`2x_-mN27A>nW1Jl z%Nx9t$@|Jt>VJ$r>2@riua&7 zeE~^x_WsBO1$uPqI&&4j$R3Cks8{6(+iY-x+3$&BPXa|GlH5T%a}RMY%Tq>gwWu+j zzt=5l*%G0e4Sx59eT=rEf7NwNe18ztKO0Y*x;Q(udo&rEj-#=294MyMgCvY&nekB_ zT7NZxb-t)&)v9T158r8-RSx2u?RKU*vW$%^H>UMBPf_e6d31i3qqRzgG=A)FmQpL? zIZJgq=ATIs36JTBfi{HCCRkGH&5C}#VcwTK*c;c^Oi@ap^?xt1pQ9Od8ikQolN)*5 z9Zj2E7=1NVqt>CL$ZF9OR+?%>yHt6{vR;u+aX~5;JL5yd~!|8j#_JD6-;v`0sa@GOJd5CO3IAbGtuh{6BJ8n@VdW zAJC6*M#|L-$a|kAX&)I&vpFyEc5@w5-!_aAmklS+ydgZREC$bPGqQDHG@AEYodzYd z)7OtO|H-ao@!~p5zok#}N2!w_=12deD%px>0=1U*C$@buX^SRP@$4dezk46n(k~Gv z`HqSe?2!LuI)<0_NBvwmY`kxTb60DSe)}4Vg)b1hB`56cvk=Z!7zuCE?$OK1Kgr;w z0))8RXct zHFtdu##)O7I41a@t4bAvK6~JB_iF4oF$WVQW?{CYHXap**05Q_VsL|arT1;~9Ts}q@2enuT1S14kVrZo~=r$A#` z2;}CULUvXeb}oF4P_b6L(BTf}VXYM5HvpsUuV8~T@3%R>$Jh286ili^rKzSM{l1#= zwEvLGF;z%j8jnO3Uo@4kLXn#!dIn0sK~oj3GV;*08;hm&)6wA(iS35vuzS>m#jWKi z`uGU%GI(b6emc%XmY`a?h-Ze(Fg(^Cw=a%|#>nZANshqmrcH=-|4rQ^6|rx)G8{{s z@Z6#Z(^p=D;?oKEQ!*DuBj!Uaq#Q?H@&Ch${(|jG?&GXbht+3a#O_c>#>w4SRpW`V z)A}ek+!%`=O+oU1rP%tm8Lv9?V6mZ(c3S|_9X1G%Frl$t0>^ zi`8f-B)DJ|&zP}`7wEQ01vR+#(#zN|1c!a)|4t`Tldn(%zavDY0_`W|F!P=qS`_+Y z&kIF(Jk~_DxH-BSLOEAyF`5H|u&~n?dXhOPN{K=Ij}Oq{_ru@DeRx}R8gI+*@mX>w zr0vpaM&Di1YvH}~1>!iepZ`5m^`PT_hwOe!qR>toH6Dj>_UB!kc#(^R!Wz_g?Z=p$ zaCCB3XarY-{L9;jL<^EHzK)Q`8xfx= z@7{^pi3~0ySL33Y2!Y1eNydru)x`$j_HK9VD=UDxR5Gs0ufl^XmoP*!AB*>`ghFy6 zZ2FS0W%@U=`_sTP$EPYv%%m={LZ~w4sSo2!1oFer=|(_E(_G|<=HV9;)@bL zwdote<;HYWDXGGJ=qhCQ^TgMbFr2SGh>B(JXv`%gbgi<5<1~957`X%?R;#e%_9!$S zc0p*_idcCS$A-N|P4i78H>IIybrVKwNeM2q z>fw^4DmdTjCpd9_%hF8+==_(?xkjs?VLulU%Y$*+h_F3f5#>R)XvvGk9IGpcSQ&tu z?$ZUQrWwMFBQCxx z{1}VhXkpp{ZCnT)k8b{1yBR414V6d)3|WXb34HFNwHT&>{BtVLcR=kCC@fXM1ttD_ z*PQ^GG*F0>>f@}skwW%=B{<`97*$u#@jgr@>MthpjzkuEB8?H2#XV7e`Y1M^B&^q+ zA}FNrocQJ(w3O$fJn|X5&zwfUsyN`+4IDWbj;xR*yg70hqCK}Tb6Y9ab~Rz1uR0>Q zpK#_+W5llK?xmd*(5yQRPH(f2m6?XOCC9PsSUSX&|KglsAza3Nrj1jcP_?Qu7L@1X z*^e0P+V}@ECtSa(`bJ271uFj-Cnh$BF z%`UWz^FwvpKD;SX7E~n%2i%? zY!ghMuHd|D2e_Txheqa(8_)PGXu@GEZ(k1y^&qJ44#o4RI|%#MUualfg1kZ}1a&Wi z-lu)2w-RCh(^+`1jn5$UIr~qU^9OCkF?NLy(hWC3V*UXvJNt)Ho0{n?pHr1~Xb7^q z^@M^w+QLej2hoa^c$^_XdqyL98cJiXn=<~+lHlI5vDlNPi5qs681DR%&!#IeQY8Wd z4~Ifw#5FW|stHpMjTY4I4;PgGsS4qC&k=2;EyN#uLVA}==v^+~PfM6X-BAo~<@$WS zeVtwgexQEG`sh&UGq|ow#sHfi@F+iuT`%}E{aQ4Vw9POedp&j>n~MLEheC3QId&>( zV&2FQ+%OJAzP2aA^e(|a;})8>yvK0W8!)yx3yGV#nED|I7QUINH93nP2cD8aM;|TN zD-OBJLP+0!j^v|t$iMXh{gZ!UdTcKca2=;buQ30>W&GD!PQDVEl<)eNOse|v=UF>> zL|XE_={P)+)kE<_-Y@4r8_4S7gR?RwR4hdC_R(;E-~=y`8{P;zkQOu-;{y_r5q|{t z4VGfGunn405g5~T90tm>;C;#+TVLCuYTO=V4fjJvWH92R$Kc!Dd04p33p4nvxsvz0 zeQx}qDdiyue02`}kHzCfaWQ=ryF#MV7igKU&KR}U4OI4O(y_EmTv`WlhNf3b3hHQryI1zY~Q=bit{ zzr(s{-Ar|8Shu5BbRKEdo#;9tC#;Uv68bJF3jLx-3wmL51kWN8cJbFXW!P$D#=Br{ zV++-+%pslT1N2Cwg5frbxWapCubHLLR;er8e>_aciltECeq{X3?^KGwI-E z?ppO}!IHrBaMp{)n@il?{y-Cnd7tR;j(&peMj63aOiEBv5{K={KnmzRLc`@NN#grk z`f%ea1x;(G7yNxH=gs%NU8!ia59fWqR4nHAHoN&z+*OiC;m1j&_8^SbzK!I& zxH@c;d5-YuHxM`LEW9HwV&TKXUTv=F zFHCS!eL2E-AJ40KEPlCNqoz~I-0{a<7`EJhaxsO(LgOiueWRbRUz5UsGU#NzQQio`;{!Gq%H>3u= ztt9WagQg4&r>#>n_?$MB_HjnO>9J&*t#^z9mPXLV^i9aG8-aCgme9#E;e9Z5IJqg{ zXV5+BIFU{L##T{@?>RC%nogfMgJWOAFu~&IUsUsMiSiwuL6MXdR&N#)wj2z?HtSJ% z?(&Z=Js6L@v*l1QtIs6_f8T^!>~%j-*^ z;Kk3^V$oRk`2xWeVhgNNd^aNb>YvHp^)o?1&cjlGJhJn zs`;5#2;p47Kx9Neqtxg)^1mHJvbqv@G%1oAUmhf*l1Q>zwSXqS-$0|M2%!Hg5&3KY zj5%v(Jq6N;q(EB0duw9FN%-V(4VvSDL?xumG123%!VeYuJ1*ze>SbPzfbOg zSyaE#0)t$qqUyU1`VL(}(6C^P{2PLk)85hFFR|P$aGA)+6d!I4!yFsVoUG5_IlIIB zZnptCOM@W(WEe_c_QxjKB{n1brA z#klPkfQDcfjLKR}rrdpMKOsBAdL0K@LMbK zv#yCey7})srkb1r<#E_k0yk6oQoKj8Iapyj-DM1t+jxy{%AOu55&QIJ%safd4jGYRwvDd zRLxH`ZR6+tyYj-64`sB+J&m$<=F-9Z^+6-}gT17CE zzl&}Bchw=ICHyWPEZAHcDrhbqEDV1+P*4pWF0^cWkMfvrSbp~#=i$ZD-sRWn&z=Ih zd|M57>(t;j%M>w^ze#mB&#L|SN;%_;P!oC%xD^GBxBIcsE&+o-Ct%Fw$(W|@%=g^O zc>mN6{Doke#0rSTSR?kR9mZSF=lheN==)uS#H{bUKYEvJhs4oNK0ipBUqau1+@+#P zC8YMfm(FacqI4rkXmEbO?YU-n5IzAF(K?WusfYXfm5{ESgON5z@O9%e{0jI)Ilgri z$-CC|{mc=WWrR!hW}JE153~8}PmUJwd#fCdh00^qSOE*PUvVZn=U@Bh(&p7(a5S8W z0e##DH93->d7^R6C>w?JI|%pbxGvAfaL4z?-M8?!!xJk~OAvVB5$&AQM0;|7a6gh0It%9Gqq`H9EtSUr z4{azs9fxz%cB6Q406I6UgQL3yf(tkoXss6SS}MTh6nEe1mC~4JQjnWdM}xF8$iP*A z<9jFESv?EGgi5~0>!5FMo|Ak}Eoz^?$DWcKuy1L>t8F)+(Rvr|-r7P#Wq)D$0Ug0? zxibQLZ1CqqAjZfa<=>leh};!{RbOVq_qGj2b~_^W@&w_}aWg@6tf{awV!xoBFjx5P z!&i!CX*8xhk23EapdFcFf~mN!utcsAY3jLXpY#rrM(1!(Ymy-Aq9w?k8Y?J#Uy9r3 zc$$6ZKs=dgiV^v`(8v^n-$XO$Xb;BXbHnjV?E!B5DMhKq6^JjJgf>}m=yREyb;lfG z-eY|s^~OL!>Rc!kx6a42J`vnG=W*oIXK)^S4r@I$@%7(7>d=>kWN4`HY;uUO=zy!h z2K5)*n_gkqx~qKNVe8;_P`waZ* zW+=5if)kqiaOXwuX>cwthjq3~WuLl_Y0!83EKu)AwMocJ8iMlKQw?omjZlZpSPgkk)j0Oa~+ zL3kaAr4lExMr#gUn~jE(soF>B$=}noxAqFUC&fX1S0I`us39?=mMV%KQ~g5+VY{QF zu-|>Qkk5B0T1k80T(TVbR8Ku8Rgk>*75(J*#u4H}VaUHfZNHFUJyTD(?WHK3`Cuy4 zhm8*SH zdvuFT_ZO2!TOwtD*At$86BllumJy^p4jZj71~h`is!Fj{#u)_ew1LRE3qO%?&e zH8EVO0H?klKvh&U#;sKpuKrOHdM#U#)!2XyOYdV=cnL1a-oaX}LR{wGCzDn_C$$yH zv_1Y3g>L$YIfEJyLYFXJ!Auxmr7A2+{R)43O<`oCjBq31FPtsw$@IW2s+qHouD_MU z^|V2lT`PmoN6MUIrh(-mRoHIGqI=KdXj;j2O7!23u1IfuaCJh@`cOD+n+aQ|MaX~S zh)kJTIOgI5tu;k>Tb9V@7l-jGF%_f#CSdx1ahNG}7183M*xbDzc^MHX-+m1nL}93? zD&n*IIG*RdN@hD}qL=%pH6AduX`QBdKMKfHCzUdXH6v{9XT&Y%cXYAuv?z=7RGePW zvhi|4uBW=7`BqZ6%ejEvTRzh7ap|PA?Kq4od=NRo6LI%{VVBf1I7-~d$3|V~B*c4|_ zd^$)kTgD+I=R6CuyTql@{7rf_G* zJ`%XU<*xP#dTCfg!D`OjwNuH@#I>aD5=wb%-jUWQMJ$?PL%T1V(GU&*lDVElP7~Yc zfL9h4mT1$6)hU#GQwf>Bqu3nXEVe7HjQcXxXjTsAa`fjuto>$mVAm>|I&n8Pr}eUP z4JLGh=jo<@Sb*1nv`GFp$cAjw5PFrl*RV zSz&e_+f`sn57vYtCUg_pt^6@=@;C~cpG5!lN?^)yUHX!|kWTWBgvS6H`A$W_?$=r*kaTbMBS<^bks2bA(wfKF#7H zmQ(1VHfH@hlG1W)z>ay4X)R|{&3sB~Gj*`Rd<=KAj77fA7WRutlCRh)(rQtljTe+i zW@{DYlpiJS*M~^8sgAlf@1}`m4di)s6K*ujqA1Sxc;neXb*F}sL(3N0Xi`ffzSmNy z_#4`L$BJiyCeZHb`za}^k#_3jFo$!`m|sUCy{rkr@v@T$9(EFUq@rl{l=U=aWd%jM zEn%OgpJ6j+K4fBRmy%_?E_E|o^7!k{(!2k$HYM&r%aEs-SR0x#bvuiHpPb_Kp);rWMRg!+~9!5^V@(}UeHS!;^TRzrwW#~joSDiugnorWiG#lDs z=ts@#yeT5Pl9@`$livDNdhiIeCp48RW+|imj{{}7Wm3W1Upz&cN~Zj*zf6&H`$QG& zo%9D*dNPCt*elWOyHg43m2`;nV!PJ`(xXegw29xT+HQx_glvW|@+)Q7RcKrQ{bW zNlKC_k~A1HMT(FNC37M}WZ(b&b${C(htCdq-{-mSwXW+tDR-|LEVbW~ZR&J*zmJw4 zsdg5-_nQVFoC|w&!Ws7BdPtkM6%rWog!APrrnvo)N{@udOo*5^`G5`pdL<02_;%3*1Ln?Jextd(klH%Y}Yy-CASfy_KcQH^l54gKAhelIO1{B#uOc}sFL zUq~T+w$MgHV=Ue<0Mnmn;qz5LO3t;XsH7aqHTQy#>Tb03iNuuFDdazE5qbRxp>~Bi zwEj*WoiHz_Nar*uz5b7+o!SB1wx8N=Wl>eQ96l_M7nyMJY}36%zK_)@#-$_8XsQw% z==0L$+vBKgwIb{zdQ)YoAF190COIpsOc>+vQm&|--2eVea#m23lT%TV`+t5q%h|}y z^_}nSvwq>A1#{QV9b`4sYRFJnNtRZl?IgJw|9?L_`zAQri78Fc?wLAzQCkJM;`8F$7e<-QkwHl zBzg?xc6V-Lqtq4ChCQ)f67Rxs@vC{pH!J?>GzkZmuEnGKi%=u`PqrjhnY_OyP+Qw^ z(T$plu;|;U3qK_p-prBQ1jo^?&6AB9jz=OMpG%(Z`h!DCBBzGHEZTAfX4_`X;Q`MyKkDUQHUDGpwFX_$Un znO6=S$x#P-;sGVY_3#>0&uPQh<$d^ELo+;+-b*fvbEvD?Ds<|`$em;EMi=FTCUb*UMD>*Y|q{H^H98e{6oLi{`H%sRA%ZEmHaa7Zi8PIBZG zhmuj_ehXdAnh><;u&nErPqg{xRm@HJja@G)(8)^7l#>sjYJWP0KD>fO7Ej5wLwh

lxb;&mQuBiF z($a_H{BPmrb5$H$E{`kUKFRXWpA?S2>qu6*N!PX~)4-dG{B!v~{ObG+R&Pu>v8M^& z82A$xMw_sK>No5Z%mnYg?eU}UY%D7h@8LbW@V96(W~>N?Y;HgNyk&&K)L?867%A=2 z=`NktikE`_y+S+N?%dkhft5b*mFXtFldZ@Zi49t51~2#C$3q(!Ho1XQpjFsDXt)t7HWQ_zDIUS2cMs&A6kM*)9DUY zuYU}_S_CVnFu_0mhOqTd>AzXCVN|HXO+$9F`}Dr76nY69KZ_4{8pR!D-Pqx>5?4HY zkE5M)QLAxK+WO*@Y>;PzRDN(g_Wn(zEVpA)uh0w7a4E&t@JpC(O1TPasYMo;?=HQInJgXf7Q3&P>#4i;N9ok$gVg`s0g}9I zXzet?F3Sq^Pm^+1gLLDUGS{VM!12T!{3=xDipK}= zzfR|ND>nqRcZ7$07vX9U+ z)`h97CBNO)BpmgM{9E)yEVK>zfZ1Uz`d|Vjn}_)JHXAp#gu{PqSL|ArP4CiPOItFg z;Zu1u{kU;Vn*Bo;J8iy5(|aD2jjL)eoO$16c@15p>B2j-&gPDF%;iH;k+=9gTq(u+ zg~{mc*nq}+?$8T*O(nz9Y3j*9T2P}QxNwWn>+MR!T$zK-KN9dynuul7eo%nAD>`l7 zfd$zY(A`!=Dk%z;9Nd*@(hGC`T`A+b^M<@J_MOyG;j(Phu4A%+C;rm1eeQU)_b5Vr zT%dSToVgvWF>9hB-0Loo$%QR6%`Ffg7CeH#j}6rIN6Q`+-myBdGEL@f-~hP}!o3(e z8q%aed~ldL4;HzL_q%GyaH9eBzE>#C9Mg|`M1ZS2P5AN0ml(Ts9v+BX{IvVmaI!=a z9)i78Uh0myY4Vuq7)8o=yNS;d^OyC0{ORowX>Z@*viX1fq!mf)@cVrall&^p2reKT~{8Mm?uLuY_CY zP!b$(jmF5!42SSIa+oxWex&GAeV-RJM=)F+Z6rE9`VHxi2j2aBPoYgYr1w;U-=^|ldU-;LCVUg#?Lh~mv8#u0VcBIEmhVH)(vMW_|CV-n+QVme zDwN)OA#Ra5CI)oJ_Oc8LH``5)Cc=ZP+>eHT3Zy;v{7~|}H$HZ_L>{;1QtbvcdZL*w z{ct!YN%MzFMMo=S3&rDThNCIRUTA~1`&<0G{QxQ_?(n#32VOlxhqIiYW9pe0Y;|?Q z(LO=6s@0V3nu0j)^M9QEERvI)ZP>r3F8}*#qr{G$Cgc@KR~OgTVjAJ2C1XWMfgZ27P+7r)l#W0BWUnUluNT*6CjZ?XQZcKr0T34edshbLZL z$K%V~`DI*hw(tFcSN6Zg5%r0@diW2qo7xG(<==2SKA8{q-Np9Pf_X*H@nmDrk&L%) zqfH477&PlX5|59->CsL$(b*kss=Af)o114b&L&$r|{gEJBa^YNxgSe zarl2>e5HlCz%G#A%ueDVj#2!*-3siD@W;LEU6|$Y8&-W3SY?bV-~5=y%JY-B!DunN zuFmFXdB=F2?KaM?`;B>5dvfOcL3}aR4?a5QFf~qrJCEzj)f*4}=08&S z;biKZ5Qg2s1S!c7J5IfW>dxO{KYJElx3zh>IvHws+2LV3mN~s zE=6UZLGd;tcKv0_x|^p^#|lfTaNJGxpB2bo_#4($%^=54YV`1hGihz9pvJaCVy_~P z!@mEhSL=9@{Zzt{1XsLWwUE1ZUB+ET&S!&Y1)82Vnp{Ml_NwMjS?j+*X-|n6r8+Og zs#9xlLfu`k2q)9rk-KSW_boIcuuf_|qD2uWBS@z2Lnk~9VcFv-zLal;X6GwN-`;?W zP{V|%d2kjzuv-Jq(mbD+!s}^^k>3x?Ca1el^c*Mj{O2o6v5K{p_gpU(ytklTTaQuE zv)(v0;JW0b@ImVR&xh`hdyQ%}6}A}h9XJ18p*NOdpZ4x0O?rD0lmF~Nztc|e-Xy*o z6;bp^@R~n#PQ|+NQgo<)iW5&S(3%VL;C=N!oIN@c*Qf79-J$~YblyrYdT3y#zda`K z2x;ah7pXbxKiQp^g1a!y5|t`*v1DZ*!P@T!-;?*-%Z6aMt}n+U`D{3MX~(Vsow&2xW+}mYgA`FRgc6>bbIQ!W z=p<%yO|nn4q!)0>QgBiIZop^6Sl;a!z{>~BXMv66dn>g#XqhSJcaV6veOI0_=#s4e zBWu}T)jd+IaH`h(2H}IqhFsa&pU+wU6}?n3;}ab-s|p1ywCE4znuT2Scn+&N_T^K1 zZs72xFg^#E5*3}vH|50v{W5g)vjS=~gJ^S6!S z$1+>SUVUy+?ZPF|YTQS!m|o;4Vxg}o{(5BzS78!<_-;W-$V7gsP=RFS_P9~CpH_R? zO3$ueu>M@N65W3J(%HN%((`!XGK!CeLH2I!RdPqj-5J(r~7HB8I3HQIr1w zp6TYqWNgpS8q8h~jairk;NPoS(pMNl3l$&I%#_KpvuhlrAw6R#P;NNi9k7l^*o1L< z5lQZ!XIobN${q(em(iurgmK5cnq3F_(GDG+SJJi0L(sCdLJ!H01g(@7iGGuf;DI#7rb~v; z@??p7PRqI-?~l3r&d}P#Y)QrSJ=kt7oUTVx!Zi&TIM`6@n7xuy7ee=Y!cUieKx*e* zNq#=PsBrQ@Y4LU&`Y>;flx8$jmT__t!VUM3?)e&N)fabE&(p((!xLzHYb<_w_7wAj zgXHJ&iqb_kx}{{7r2V3aZb#}-&NWTx_~Hay)b5X0vMaRz=Q7OMYJlF??vdFePf{#f zN%O=G(d%JvQt-PhOMYP{lk4@J&gI=83x_u8`DzW0c8Nnr;dNa2@(ZRcJqd+B!|^G1 z8+4zFJnQ`}RP^u{O`5!qb_7nR0Oy%vhZILgrdUdsZhRHpwJkKK>a)~%DOf69piaxr z44|ebo$*6U@Ib21$kGVe~x zRrKI(3TeH3kNWXPkaVsaK1JU`-StJd@v4%16wivgnI8(4{-fjvv$0Syi@FEQhz83TfjOC#i8$A%1u0%C(wP`Ikj5 zZ+!KIUq1TERW270ys!pVb82zncnR*^{)IhHojJq%6_%}=%hP5Y=jOWOJih8TFEG%t zIo=}#XA`V>%a%Q?HeQ{>tW|ijr7m}Vt-yu7`*OFIF}$uZkr!;b#}j*8zOKQfq=h7D)EbJvksevUE)!#(BCR)lJdgWl^%+Bo?40&!}Kkj2Y6^+r3C=h2m|87y-B`udPEWN=dOB(QEk_mg)>G6LlEm$^mKU}>g zBWC_4?z{9Zb_*7iApz#}><^Z)vP z$7%v(qstR`#7PfMc=Z*NoWyf`${{vgRKlkk0{FOK@_K}OvPqY|yk_wST)(Eu+S|V1 zd-@wx)amo%xnua&h_`%J&dla^c`jZAgkj-M;Qd5YVVT6p=jo;4W>!aCAWCJ?x|12p~&0_ia4X{Xb!qcU0ICI7d zas$rOTe06iJ|deIbXBIKM-O4czGGN0(d(fM5$AaiK1c z^J$Ma zH|BnqR(RNxWVBmKS^tp6Z!sExgF@Mz~{n7_P5bJZni2#@dB;_b3SJz8bW50WIuA1?(D(URuZE~8UjL~p?>0tIHm z=X!OZG}SXprul8C%)PXd#-E9&`Pv%D-)B!f4y8->OO9G?e=wYqUWZ8vOEsj|7YE>A z(l9J->M?Jlx z5H@@|!qxzW@~s=P;Bz`BP-SO|qje2JuWibuMzMMELjDQ2Qv)`(l;( zjOZ9-zy3rkoyBfr)F;xHe9$Tw$*5e#4ofHU&}?GAv+7*?{Vukc2Ep5*CyedwaXo!0 zewiq8V_08~89S0wE{sNLtvOb09S4(`FuMIli@J%<^u$9WVfsY_UmlmyuLEBYn7#@+ zop)08sBQ?AYn3{cSxdXt5d3OnxGes!Jd%%5uYHNske4ZDN(C6;vJ}Y|hd?){hMu>W zQ`P1vQfXap4C@e1?Zz*lrxgz#SJ$Kd(OqgdK8T+6(4vq58E{hAidX0KV4nRS zkGBZmuDwTc)AI8aJYEhVwkC*Z%H^J1%p(tF@*^!5)JAm1#9=S!-`F~m7ycV%;~g|t zYXi&wmbkF9I(vVL;fH-Ua-Tjnyl9mYmt7DnpwI}~-ZF>lpFBY92X$nZ%k$jL4X8eG z3+s{|!*-TDcbKBXvFYY~ey{w`wJgZBTj1P{SmzN6?wO{4txDkNW#E;6b8(9J)d ztdx(#B3OxgHt2CiTQ-CD;4-# zZ8KW!7NTlw7*uuVVRzhm%zhtA%}oi?lDo{!Puj70QU-#})Nu0IHdxpR7PwL*NA(Qj z$k6G0Cw~=x^>g70A46XK+lHM7jOUi4<5;P1G?n~tl`S0AXzi0WnFlX@jZ1Ee&|{M+ zU-hdLoRH1vceI$Mtqu}djki*FpCkNj@=nh1oWzawi(%5OjY_+Cl5u_<&-3x%yi`k8 z8yw7oJC5Sj9e?49a4Ka+9H$Iz;R&~Q;-$ZpIY8wR@+V%BoNn)uB1T=8wv^?Q$MeUe z8__~?XI^mRm^$8eyqvo(*#+&6=kfJj84@%Wuf6QVt;cgz9ZNLdfUSey&X>8s7SM;Ip;%C=n$OY7+!QnQvCv3!_XIiX2r9pIe z(oosPP)*1{Old!!Yr9o43&@x3)I=wB!~q`qxPX`Z&gLznnj~Aphq9E?QtR4($%ym+ z3hf!)_@7{fFKkzg=YI~tqPCg(g-@fw<5rXYGBImc(dQ1y>fGG$gTCJRBaJ*HZ#}H| zrF6k6MKFMjq5thE4n9%l=hIdB7q7gEJ|HV{1KmaZ?w* z+B%+djJommlkaiaFOpImCed*5%zEs(9oxNP@ON+!ru8b4{XV%z7U?9?%&rv_-=TqK zSvJ$Y-+$?a4Q9 z8AiFwIwDPQ;x;Y$N#>%fHg1nQowZh^z?NB5xi*sWzL-rLE{e4(2lQNDY?4y?0BR z>G{%5&ptG^R*UY)O(qASf)B7AgUQa>P<1mx)3j{&Pb%zm9hG5c-VzG58ccCMo9NX#cWN(xlxnWLpv478NyX$7ObCX-n}AdQ}Q2A!K@(5&?zdVVV)?||iWriVOLjWOm0Di@GIpJ>1l$LDZg63A&ho_j9t%{$v)Lbu{Bcw`k$c_y0t`o{;f+x`rfuV$d- zR5}!6b1^(F8tR1`;685^HqFy#?}G#Qu5fp$*AL|Zi8lOvoH0MCXu`4zZJwMuj%S4C zNlTX8C-1vn=(pz;1td)n``6cSUz$qJyOX5#3j|BUxRH7 zqN!veUCVkc`Qxf&J35+r49+CW7{Tfpev$NQo5Xih50egbgtcD?)_%B%;G?gxqF8*#V@i3r=tH>2%(yb zK^#~mGP|NHXI^Olx&JpM$I8!8m47Ml<^*(kWXd zSj<<)a$8>JhXxjNuz` z|FHh$MTG6|#M`^s@V-5Pe8h4oa#jankQ9vcH)=d#T^A0tap2!`lVJPgKiD0*hI3zz zqsAlw%4ZVbFjL&;9J@m^W)gOuAuJZGh0MubqDyfOakGx#Q`ljgONxWmku;3^uMFSy z6#4Mdx6pNZApVZdu-+oN=+~}7H`+z^{on%G7kdY(_Ut|3myMwnkCxJ`j_EWuLIJg- zeMM$>0at8x;QN|oFtfOW%)ItIX?G`%8C*$~?IpAvTZfX_(bBH*{ctnrCAJ^;M&;H( zSbhzJ#}ftN#l22355~dXXeQoV3Kc!ANUVJw#NO)z_^q}Hr#$zy1t&+|OOS_bm2pg`sYku0~X15dBAV*^iD&erOJ z2<0gHVkqVXhrZH~2XoQ=+FhI)xSu+08b?n|zDcgdMs(2qB>kM$1Ks_@*>~n8J}~zh zp9w!EmF<5mt@@}7-)%B%wX1~IYEupt?ADJ*dSR7^0cMZW5OcO#lvK3i9bJt0*(YsQ z7x|R;=S3H|gFI{f>dA+{xS;y>GK_4$1Lcnnae!N1)g^N4g!C}cGR*`0AIg0K<5luZN z_(EHwC{{g^WCQk7v8yJ0)%GAO*fHso}K0=A@LmjX$98rj>PHHN$?3Vf^6$tbZH2| z6V+!})^rF9h8)15+ZhP_kN~GkJ8-SzBic1U3l|K=K&?v`cpX20a;0m~w7Cb3rPWw5 z={YL)d7#(bAUKQ9yyntTD0rg)6}>D{b~%U*12^I5Pj?{fGzLA$g6f%rC=j!vT?@M5 zS#<}DDY%8%iZ3v~q7GS7A9%g#ixIV^sP+Db_?mW{b*EXdJ&btWBtt&bPyCMEB5Ant zBRX~bGo=j?-5Hx=jL*MnsGD^r1fxhCGf|4H*kMzKkF1Ru+d#KUWn@ ze51=pWo@vve1%7o0wCWq7KIlE!vBR4kABq(Grjxh(Nif3FFyq1!tGPYa?QnuDm>iUZ!^p&p z*N3q{Z5L`L$76$gAtI{Fv1F_ozc82QVHV$!xH%Cm5BFo@+;|L8o+I;G+!<4ByYWps zGaKKmkv4ss`q<3%94d9ozCsJv??%p0IXu|=0-F8S@t*7qS;?_P>FYlo%s64pQ#Eop z|DU3bTBlfd$RA+c&mm|FlIPImBYfrBY;+aMB+mDjiyZO>K%(s)qL29cZ4VudnGo_qfkrYKJ2kb%c}t&+1$?WJH6 z8}Gx$XCm9#{2oubg>lK~>$qWBjAEOoSm`GC)-l)MeMjUaZN$AnupgE@=+23;S~TmM zCE5mc;+elydDO!l{PyAv-hJgUdWo}hvRNeWe5s52v2HMSNW|j>ei%0X6;|Y4!g9lE zu@8EH5mR4cR-+{k+Ni|GyaZDuLBjC9awyzmj11RrP@7@E>sNd5*DIZQd-^*}oL+~V zo7;KZGH!Fr9Cu}wM>rXof_b@@CNMeUc!*y*F;X`7>=C1O`AA>mZoGlGk7$08R~_fHLJk`cm4a|`W|`A z8F_|!&mKk(MkY|xt59mV8%g_4+0mrKB9pTAAJF9tK7V#*r7AQ2<@OwXzIWp9tzGz$ zvOZV;Qs>=0U!ig01zal#;tPBBuujH*96G`mIW~i#w6-@YBZu(cH;Xvu{cg5uI)Ev+ z7Gv)@7x=cNfLCN<(%@H!P}AXA8>%s`=qu*!*iULkvq)uK5G_5Qg{Tw95ji#nH_P*} z&!+@5`40k*p1`M058T|I!TPf4z(Y5y5zF5(@M6(HLcy z56iws@t`^!zQM)9vGa#EPdP!@CdY~8?fA_FkxQC(kV0IONMyH2w|fevdN0NFLDOKf zcr!M}41nIvVnNi)!HU1J*fKi<;VRL1-J}cu*7+2#XwCaW{=gykFs8gqMS{ONbiBsW zP;tK0(-@2Qt!30aDIF?CN6!w09(rk|;}nHPmiQYQMxD3ZfRO}euE>+ z8kQ$G2f&S`{a`>h=wFEf8`SHol0$#!R5ajz1-zqyuQTTM?V)Jd|!2Rbi}t z7B*}>goE2^u-HS5-?r&-q(-t})W%`>jyn4KSB-V8>#$n35XTi3QTelE+P1QUUWV+! z0@nyc^b!21O5s|1e3uSD74nO-G1**j=bN8W=?W*5tQU-5{aPw>6#V7@E7Axuq~o$= zSjVlzwU0OGio<5Qct`kihWDmL!dd$FR0W;9o*+HYNs?aAy($ZsKU>!D-$7_D@WrtZ zbM%{0M2l*CXxg(&($1J0SbXFL?p!*J9&KK*RV|`B+TBQ{c`4R!48xeshS+d(KdyvE zqsb=~=NA6K)G_rKb>{-I|J|33XxC<)9(~evYVe@ zOR6>Pd6?*AfAJMPLz4jEep&?*yt&#(jNX2xq+g~Wx$WDGdCo3~v}&Uc?ti7!VL`NH zKz}69StNB28HmASQqZG$gY4rG2U4+Hi7Vb-lDX(S{Rk1vXRW7Lsv_~PIpbOD%^KKG z{)m)$5`UF$2*=SIGM4)zINPRVdrkp|JyH<7MqOmTTw%Rt4V2gSl--Z{OYiFg(P_s{ z4n6stHAmKS4-;i6z4otkF^&0vhET3DtA*7+e#NXX91G8N36R zUbN!wf&I9)Ib6DekFwgwJ!P6@#YlY{i>;9!SpKvxWXkna{2)|PsoIay)*8&3Sq0aQ zN2Rl}3sm6P3$A(&XfO8!mDWn!ZN(LuaN86*Ll0r$vk%h99y0}3XFYYg9!Jwxg&;2K z4|D@{(KG4{LVC302w4GU%zcSg1>vt)IhM_bF6PXJAv|3*2wUSi^6RxWY`Z!aL*IYE zx$tJZy;qC}H#@V%3rGIveNdd6$56lkU1@~I6z*|;Dr12IZ}&9fvGYVO+SrCQ+nXcp z-81;+N_;nTAqS`(;3I06S@Zc76g?ZvX;)=i*E?-t?I*$Y^t{u0T~EC=Lu6CxZ2vBC)=IG3x#|xO5Mgq_1Y|# zF^AJE7P8A27mhy${$<^a&t8f6R6C9b>22WA*O%~wmQyhCv*a1?BY4`%k-ScCKI@#_ z&ZDn{Vs63!JRa6W7vl~|uY5*I17Dw(G~^$nd8RXmdM0wt_-7QT6+$U{c1j6uccG=G z#fdT294z+bq3&*Q+w%y=rdx3LUBmc9WfS@}IzS;P6+2hn05^>g9@TW=HHeo=dxzqX z<|G_;wZW#yZk+Oc8mpR29_3Zm{~^x!3%OkVg4g#w_?K5M)RO)owyqTl zW+|v{OOzx%UBN{YOr$wWdE}dLb{hYQRwQ)7^xaCV(Z!x^2AlD1mk`(uHel1T0Dis0 zmTO9HBF|5Wy%uWl#F3*h({c+|cCJBj$tLVgo{BU%amKF+NBp84DEQh4)g&?N&Pw2S zM)!Exd>@`OWfup1InC+aALGqrOR)S!2_n0-PU8wnABZ!r=^Q-VG#he)*Jc$i_}QAK z*wk2r%&#Z;g0Y27^Pzb*nVL@#(j|fR#wOD)*;mTHo~uBWCR9nAI(K@F$XEYV!DJ zX-}-km`JA#UCHF|HL9LG317S2M`*zdit4XS=UNufhX7TMzM#N;vu(M_M~~nA?7-vv zefZ{S6@EA90t%nmBGoVo&fe|NV%jDZ-Au%)0mo_l{RzlDb`&PR9^&+pOdL0iL8F>* zyr{=P=WYvaRXsqJ{zsr)ks$Js{dvnjJJyPu$*)XKp{wTvEWRHA&ujajf4-5%DL)au z6gf6?Fksgq>DaYNhYf=k^OE?EY}yjY6Ycl#*rJJCmZi*(e9|#8rU@0QcD$f?BM(0i z1v~!+DCg_(-++TSVDAcGAhK)s}KLiX&Ue&ko=Q7iu-y$Je}K>>f8m3_A4;-^BFo&bBUT7=dj8BuSmMM z4K;o4(uwU>aDQ_LC-yE9Jk?MbjSxpIt2D0PoXu@dPx7v_c{uR;4yMg|h&!vY*sWkF z&-cEL&Lf@pXMGe~UQ6eU8-B=4U5u5}1|!W^nHL=vyq(KaxaR9mI;3Jic2Oa+DJG?` zE*9**ch*?C{5)r#JIpWd1@gnMCYTyK2}`0y2WZuPK0VEwKe`OyNuTQ|V_PJhNr|jN|BZu!I$+s_|;MaUe+y?g}IQQxCgSbn2lXlKY;BwrqbH8=XiqhelGXu z$f}>JFs)-R)_5|7uZw+h7AEmrGgYp3k3v&+Bi=O(<>79YWd7(E^;18H*B?tMCuI^Y z1z&>7IVTu*c7ao>;4-EOSA@b)Ov*nfc&57Iyc~=>isywZyoNN+dq~}_yGTE7?zFDh zc3Kv@-i)3txQbPWhhgYu(Myc3rD*C7uM>Aj;lUo@WeT=h1fkdMIDB()#pdDKTRvYV-EWr?{m39@9eKV7l4?%*{Ur z2ctFU?GS*yf$=E*l7vMwVh_*IjYV`meWM8%0)N?fXuwHd=fqdb}jt6)Kpp)fhKM2`BeA!And_hDtZ#HW7Y`&8~%_TU02$ z%W0%C!H`;fEtl5>KjZ1m|G0n4cN#W!87+dT^ zc1;=fO%ZPjTk!=H)@XVF2TE`dX2&O1+g}%sm)K94u*-;}l zliBmrXOqeAoEnByx#PubE6}<;m}I}ePs@+=H_i-o?IvQyx>72?oI&1qLdhn#g+Amh zNA{TG$k})szZR+TmXFr_c+wq6>PkGcTPJ=ocq31<+RfqaaqPUq3no=z(04kIr#?kE zHuyG*!|O5U$Z*-P+z8qFuoL)u@*(?=3$;-_m|*iha+~B={gm1Yk0W$WJtYM7fVrf| zJANIPZY^rGww@Fsm0StHus=GSQF)YYW!)+F`!3mbn?wptH-mlp0oL!KZew-lHU5n* z;B_0`@N3`eQpu^yI2Cb-i_i9!<`?)%KQ>;c?>D1pm74+m^6X6|ox95vi)*Fx_vNwd z*%4@3hVsa{$~MD3wWF6ep5g89t6ZutvM_?V;(qNTsSnhZb(nBi=2qcEim$Y>s?-l3 zmm9&f*F-7DcY`d!d>fsfF8o0A^f(|_&!&IxlTuNdvg^0$WpT1J*ykRt z)4eR_GJWV&iYMiMb3w|zv%F069rsPn;I{Qc$bd}g^15Y|Df&iZ1TSk&x^U+iS<>Uw zi5SR_Ip}I24D2q#HRu%{H%3W^?A}N{gdZTq)>m5hw2>NH2lLQiW8vUchrP(R9lUW0 zg?YJrt78_gwCqV){#nu&M?b7EkCZ93yDFvKs)PDPUwXB83f~`NY9p5u%G+7MQUF5M&?^@E|psqqBZIW zb`7+J#)DE=?#V`6?GYTCR4%x;UyxO($w$xVvUcY)(7Xs#30_Ul&L$kT@hN^^T`5_) zmdfl@HDrUV#4hIjWK{dcklP45EDE1Y<4akptueukAGy@}VLXW^64mUrfX;+K@vId8 z-u6L+KakL_q~?a_e~H?*S0DFkXGtg$+FFdA!L;-vp+{JtNA z*Zp;vJKjg~{7865MoNZ%$4NZphm>q#jLS;GIaAgRY6BHGV$%tIuCF(NZ2o=te(Aw$Z?{NOYXc?!cf$8)TnK&n zSQLBP!t;WRoMuEyaVa09e_uit*p9@H7;=FOD6YbBMpSdrW>Z(cUSl~ei-=M~2m zNTa74%R;IjTRolLg54Wi;5_;b{#*G534i2y!>R)O>8^vRGa2|uhj|w1fkchaSaftt{%0o&Dar#~_Br4p*>`M)} z@ufQ(X?>(w{=3n0VtaP|y$inv-hkiN>u_Bdio6m-OfjsK!WInVdwq=|YdImwi5&Ie z?OHspTPu8WVwDRmHfvs)<8}+~sXw zytraYoRl+AxKv6P(+9A1#e%bK3BQRxYG88eeT%Eat{S$}gr($!q{KAiPmbV8Y2JvJIF=P}zh z!sT)WPGooFuEyK3RiP%rol(#hiX3uMnOsCDAFKX=H>M z9iw1)(n+{ujbWdsLDNUJ(dnoZc>h_2qlw>Xo7pqT^YuR}Slle7$o-ZL`1nDbSLFGk zz3`a%D)Zq?JIucIP4X;>p)sWorN)Xj+S--{uP+YJUvEQ&q{6qddT`yKVC?XGO0T?I z>Do>|;eV zc*7XXQW30?Z_4H^i^I)3UCK8Wgp;et&GUn-q2E5#V5a3@!HYP`Gc*<8#eFd-X`+qrklre*_h?L zVSXN$b??oW#rrp|<`h0ETk?ljdiCvNcheBb_%ZCky^@Hzm}~h`JrS%D4o-5hL@2QPYgN7X2(u&yFdSN%567( zH0co*zZ7>_?P<&pHK@E}B%P2F;FS_4SxSd-p;rjMuGeGZ%!^z!ZLIlx*I>@S*oO-` z|HX}JMLynbCcanAg7WFEd~QuH`sts>(aMKB#Wsu|i(KKq$)?Q~;Sr^u z^roE!o#@iIDrtJ}^Yr6gA~ubS!I$!NnCj~(98p7|kkMbTi~p)=lp$#>K@lKXu^7TGU>k`ITV z?A{_?SsKlgwpQ>RwQe+RNdd_fs^h`NSrois9Hlq!qwVu1<5SN=NV;AKd*LBkG3^D- zj$KC$<+tf-z7`_79K#pqd}=mV#?6ttp(VO?B?^LF{lb8|Di;tw#)%n;86W*Lni@NH zrSqSa>G`j#(v|yJ)L3r~qgUZn)pSp)U!p}x9S1{Y?_``>q=VFgZqV9b4QsvK2nw?z zL#1P+8S<3^e$J{B}<|JT+*45N+ z=LC$|`jeXbRFK;7Yn0V7h7v}Z(iOu86r^^A9E7t`vE>q_yWFIsmjoActvQYr3YMnI zU<8khLglQz7!&A>&H)jm8asm=WF};sAB=M@K6u}CJEEs#N}o;?hb{4wSv_xc5h8TXsT~J?Ec`1=@S_W7fhf#=Rd&% z&yw1%J(Bv&KSkZM^~u&(l{||_qQzzsYzLd5d)arI*tVLSq6{UE9)?)ixrmJB451!o z<3a1ppuf5wOe+1!^ZYXM=&(-kMo!V3-@@VOl|ubKM4^kwfa~?ML-x*{xRmbz*-|xh zk60o42q82#EQwyNfUHmE8=1?4<_qlx&h!WN-TyNlW&9 zCi4f8q@rX^JAJpI&M6WD<>7AaA4;A=b|E^acMtr~(nv7i3; z?NUF(?gJ8G-x>v#()*-ws(>b>T_LSeK`2w5f+2?n2`=F(sdo;s7FJ_9zpDuw4AbGaNuy8`1N=GG3ll!L@~YDV zdE5DXta>B*fq64{>m=e)TFw}}bO`pG8i@^ImfX@|43D3o%jws`>FgsFdb8SA8hs}W zff^G~^ra)5Yy5c7s7d^`p@TSMg<^H6G311wc6XJSE3fR!Hs`)$u$B^x8&=Z}vm)s> z-;}m*v?0ZoEORNO@t*J0?q z`}nEwmJ*8+$!KLL{dhG*($+a)cE~K&Od;BaChy} z!g-f$XyOQ2xK}W8UI`ydcz3+ca>hC}W&AM+r2hoJZTr_%(msQiL`q50g}1db|K<$Y zuv}NHUDZhM2QMH+9LJsQQ_(j|4mG>?L#sEC8TO6pisi6ilr5>B4wYt!zhUtExjeQ& zpS_}9!)Vx1q@D0Vs_0{3Ln)NPE~58U1)N(^FYEqCQ!*_mBgJ27)PKE1{|!EkM<1@? z(&uv2o%2BU{o{~tJdXKOl(@(0-#DRL3d1u=$WXct)y_AuAo#iTveZsCum3hlO?iwI zdi%5#WD+4wTrdy20#^vHh%1ggihjYDtli; zJ>>q;-oCbodpQ>t>jYPGX&Jc<384Nh78IqZN@o&AQ1jMSse4!f;)mZs$>$eXzbO$@ z-P2LrDELc#i=g%E9_IEG-8^H_(J{IpX?*%39qlACd~V%%lE`qLmiADL{%~p9wN|s1 z*5f$w_ZF0YT*rw1nuv*+iN`z#r5(bc+r<}q2JC|K5MR8$Zj0IX7h_GXoLP~kv6*L5 z7nwZ=vIssbr+qNYBoRj^+EOoYiJ+zOiAxx&PXF>9y4=8XY-T?5VoY zACm~0;w2p2QDxLn|jmXWAVt@7^zw9eC!$~-i zRu^IG)lPIXSa2$n{rT<01o+m@v)j`tB z);1}4NI2E78p+L_D}`PtqNJ1*{9C`1mnd~MH+?U<>8(%cO4>BipJ7bFE5tl+&qHb! zox;Y{F&z7%oD=uo#&XR|X!!9N4;mfNT(SvrnSp3ijF;I)JdhP<-jlYR+eOAfJ8@7? zg_R~e6ka3^K5IXKpZ=H$6}z!mJKY2C)+$lKv>UQ#Uy7ySuH)!Q;aU2b8$;Ry8vH60U8(iu{31sen$yMob$$Tr zx+PGj$i?{ZV=(U?{G2?t|DhemIk0?E3oF@RHo8u{BG;2IuW84VEW_Yts*1){!*GAm zBIJzd!P)gXJm>NZE*I>tIsKJ6d*Cc~@jNPKg2e02rov*xNM18y1Y6C1gX+i6P(5S; z6tBOh67x9NUD$)7r%RA$eVtV!5A)%fqQ`aO3zXJ9g!ys`|~$=*uP_RR^>dcF&d0(|0c4v!FVp6W68;L zx?}d_Tl7enzbQ|L}AdE}r>^J!_`1k)Im9D(Zze zv+G#be1*>$T;Z1M$JqSWL_8|7$KzeUsNLltbbX(S8AA@Ee&|x3Il`4EPMFE_oyW7? zPgPbRU(9%s$*+1IXN9K@{JG~?Za(S6b^qpY*!Ehf-TqEgVjo1|S*wvhH3RaZ^ZUW7 zhBLoc@}bxep00O^{~3Q_^_`D-Z^b=$i@C|qJN?-?W*i!B?<0dZ`O@4aAv`pofIkLT z@N>&TB#MsD*TY|FxSKt{^XSNPDlQ7Hy5Jqn&xP4X1EkjL^U@`!al};xwYj~JW#o^c zYcJ!^7E3%$|3C)sbx_h(i|3p&;QTW~_{1sEYuRqVWdSqzqkkvX8kUdLBO7t^+Dg`V zn80~Es`>k8MT{w>fTyw|Q zmC!XYVrU?-poVylSi;s ze2+CnkG%W8A=I|@r=;sM6c0x zPR*~WvN?nT?fY`sS{2S)sK&jk2Xd%e3)GH1L2JDi4RoJFb}MX1JE;Ll?attV_&+{Z z9>8u#pJBnO2N;v~0Z$r|(Q0xC#t-`N(a|?CWB79%S=NqM9%;{R8^yWbvKv2*?Z+WI zM)Agtl{9_&bF!E@6fKX7U}bt1ro$efy{11VeudI4qNQzW}RQ2DmFBrvjy>$4~$BtZ>b%j>_^`jz@s|xsg0=K4T zpibIEf6&w*@HiQOEbfGxT?#EiLTuNxHE-N_wJwi29`%Qhc;3<&C+Cf%2DOXMG6r zg3EaL-4axvX(fNL6VjUR1j}q?#Ck?y^W_NiFgpMR%A;$0K9btnzF1xvPR^!lXqBxK zt$p|aw)1`9uO$zwS}PtiN8}~OWnyC5HQaDMgnos-kW*SK3)BrTdpN<=Y~k6raPxkI zH^Zx-ysMwgeP&>)r5PC>oLFCkF+=Kp*^P;lV-{$+)58YO#f?0OB~w`xTOU5cpR5x3qpcRktn{a<*=MCaO5=%gx6-}BMKpKF z8(O$t3(uyT!df|!7CRNurO6lRn|3A54YDL{w{6tJ&X(Vs2Js#1e12gQK(jv!*QASJ zo^Pna8ub+P>=6mW%h6cTn2A#tP9mUN5IsJpNt66^>9b-0snwO!-ysEbe!?0&{_KZA zZX%C(XmU*y-5Ng$?(!j+V}2BqU)f@-`vfeFnt-`4t?^v2 zntEl2QtBxKO3L?=vidWfRhvZT9?Xypiw~2gb?DD^SB2MbRW44~htueLdnB#Z0n%`l zbgbJ{h0OYV?DKv_lmGic>Rk?!Bp7cRgWgk7v!3vjK7&K|0KDPV4tO2 zfp3~BkKONsTW{9Dtzi#(%qo*QU9XfnMukX+1TUrK~1`L zU#^JQW;-5YVa&B2iv0HUDQW#DRSMj&ie|r7r1W{mD7(3gk}Ny%`iKs^%(ORG?3v9| z+8J}y5mTNYp(R`L$wD%g@}(EcAL6L060ewVEc$TGI2ZU1{WkqV3khFj{%V9z4}`(w zT2z0?zzT7W+F_a`U606k4f0Eb&i2OC zn>y&UOPxL9o}gp;UHo@Q2kY;u<4tQHSX}JJO+JD@^w}K~mY=o3@J>VV^3q=#6!VDMmJR1}AHh_gkPd_W z?Kr;sTZ}&ec+Cf&UO1=R$Md>Viqg?^&b^!aj7-!jDh;V+>#X)UUyLvp%W{kt`pOauJa(?=McOvt444OZO z;7F%%{P@oUA;af`zwQz8;bg%!(&4@fe!%Z#8(ii)qcmp^-Zl9nVDc^e{UFD4&xsvQ zqyhVEHbbwlFquI}0>Vc=LBGhSSRSWIfwH<0>#|78}L2ln=6@b1vJuAL6h<=jqHz zKWWGFEmTrDgjZag%WD0W@TA@CQEzNZ#wIpWP~;Ra$SCz5_jA*~6}z=2)W=yF2%f_Hbo zl(=e~5#Q)D9p!kHoE^`5GMM|g@5jgWOWAXS@CY@R!`ep&Uhmv6V&-Wa>XnFgg1ve{ zbo8p_weT=;Jx)dK7hb5F_!h3shh=U&_MLbhX4$aYpt-EGxg*ZsE1>wyy`-F*KpQWl zO9D?qReQS9v|1-5^;Bf_Ua>fHYc5os3#IUdZ|QaIAF_|q#pRu;6n0nvs>{}*Q}Z8W zyZ*%TtW(0#dY1MFZl=v;e@VXSleDvImSjD!Og84(0la*60zp<=FyrPmSdcpJPw&IO z-road#X{%DIE-9;8a_)yaD7Rbcpj#4{0L|Ea57<6qw}!5E;7*DLJ^=73X8Fsu!u}X z)UiE~zg?4F>VYh~YeK=o`EnnS`sS8{ye1eiw^Q!eiVUQsT$ zc0P<^wSJM^k1T}#9?bLO1W)YscyisjNh)*{_maZyJiqxHM!r_1s{V(iX|#}PlE=}q zm)g{%XF#Uf^H@7o__FMBam=(DN|)52cXyezSi6Lam5Mo0F^UZ$2e2Hp@IuK-u!H8H z!l#M4ZLg!?==OKC6DCvi;o>Vk>iR5e8L=9ix(WVVy|M|=Gz&`SY%|nZ8?hnR?bH| zdmWsc-jDa@X|SI392K#ef=O>h-k#nvt>0I%yYEi&QhO?WT=NuVDIGZWp~xhEV1!Tg zhx?6WtPU5m{5{TaOK29o*&sS}agG5F(Pf$#w&m#0-2%n14=6}MMt}A-Yx*y}A`s*BKFR!8C zZiy&9Ie-H&%Dlddk-6o=CwwUUu9)RsMO4^G_Q+2XyYH4CMurY-be z4>L~c7swV-X>42A5e?r0ppz%hM&Bl}dcku3v{av)rf%hZ|E}=Z8&}!jT?uYB4&k(= zE__O`&Zg|$i^|jscro0cic0PAep@%*z38Q6=Ww2exG%>eS8d*#a+K>DzH;IPtmwWvmM+H&$8u@gfg98&WnhVq9LM#U!{gqp;cq?lu%#^+3w(T0 zeDo%g4ZCxbV5Wu{&yeEM52By(Tuxf>3@giKaS+30aTHO#crUYX&=J6YylUyBp ziMMPx$tkUYcvDnP&Q~1i<1A(VyyGCeHZQ|xmr@kkkKnaM!}&nQAMBUc29AiOz++Zw!fkh4@J~3u*BXu$7zLr8d>CDp!2P_Xm^=-uL`$&uGTyRfA+z=3gKb@rHO)e z`grk1Td>-bsCvQyI)D8TnH{l$@&gmRFB*x+{lY=vl0y^B@~O~%KCD-)gO=ojzVjU5 zZ0L%puD+;mxJap=7E)u=1iG3d1Ky2;*@G!Km=q{2+dNz{Qw@@;{JiL;hd<@)JxPZ? zX5r|MtH?0@g-MEe)JZa?S*B)G-@Y%t?zl$*Ic0QUfig{b(Vj*+hf=S9xio9?OS1X# zosPHvKoJW*QvQwG^f9)H0^@$s{Lk9>p`T0lUgc2G)i?AbVmB@vKZSvnm!LC;v7lTJ zyXF|c-5`xJjvk{|JF3N*Jc!KJE~A>w#WZdX!KeQic(;md$Hnhd5-ogBI@OdpUesL@ zR4~V-3$9H}qq$bKbfNGs4J>ycr(0{u%d3n|{3p>YlhG7*{V)yP(3#pzJ|~smXpuBe zMv;F?5UJnHqpiUUs4cF&;B5bqnky8b+3_CPUuY#wuTa{sTkL}`j;CI+c@#Z0inc5~ zL+=E?vcs?)RP2~Y3qI7--wWOGeV6Euo*yAaocJpnUg0b)w~%0QdnVov_QjyET*{tW zP8Pa~u&fas_&r|e@MJaiZVBQWEeE(`eGY4V`UQt8Q7F%r$GMon9m4tdW7*NHAKMJ< z!+oZQa$H{faF}`sM z%=8uUuco7U*QYHU=3dKj8SQx1?FM`p(TW;fOMY=ukB7vo@TGOxH0*{C_3N^P&aTM8 zON|`N+ma4Nzy91grv;(L4>3<}G7rtr=bHAaEGOp3(^vW7>!B^M|C`1mYLhrfCzUJh zH*lQoc3%5^2M4bzfn8)gn$5SODZz*z9MNOH4Z7SX?mxabcQIe;;m8sHCiC!Z(^++; z4VOnO;{Rfn@r#JvymO)}{?&?;ca;u ze_o~wpZY^&zpKF4;FE$U9)nBPrr6Wn6k6k=Q4vv%^JjbTO7*qqEq)jAqx5k`%@6Na z%*B$9^I$dpAliSKiDBDhxZxOw;MErBkkUvC)|m3IEbz;?Xbm@C8oEXnLhn+L7&IXm}H}Z@yEww=kQW$eB#5$Y|}ZT z@h}^Qn~8Jm9$plh#Ffs4WasmlP9}k-VJ8Wx}$Xb(_rK-iNG$Cr!YFH z!?nIv98&flKOMb=uXgd_o0s}xd+&Deb~;4oMb3R@z!2X3!k)8!JtVXFZBogXxsu@| zO{&a~rQ_$$l0nL5{uh6iM=yQH&Dzzn->Pnsa_mUjR6C3Xik_qQPT*l_>tXS<5KfWZ zc~$%!>Fd9@($SCvny0plcRB3g;G_lIef&vQ^p51@0Y0KfG#he?VHoZ|4^CerSoKXb zr}}&IrHcW4Z^uTSpErwtt`%pMh!@m6;kWR_81smwrmVPNAP?;{7}5TIsL_+>``hHW z+h7gu&~q@Sr6nVwZ7bG}>xyR@4<#kNV#)A>A+24o3Z^1U`qSN!>lCx4j42tkKi7^X z-tkB7*GjZT4(9Q$Hq=c$g1qnbgnv(4)MSnx8eF^H%=JVpVN;uLt)Zt+2xTXQsc}FGJbIcUT-%+zjid{X@19O zkrncC9|x7II^1cVI{$br<{U5Oxno&BS7O;1 zk?7mjk>iFv!kf4s z2#4oW(*64mFSHZI&R`yP^flw|_XI<6<6j)Nn3{ZfZoGs+%LCeHML8)ad#ojNFId;Uar_vHwl-| zo|TflN6^t#56LG)S8AtNL%qHR!zUmUH`HI_?aHTuTm67cZ5QC&{*%~M7l@$CcEY!v zE%HJ|SX`ctpg;OB&pVC0wa>9>djme-=!&?6HfjnsL#4JZ-gaGw8M%o#n3+qN?>b^n zdwmSuVvZ-*JP|mas4O8bze5P}t`Sb+r}z=;CONis zlMGiWOUB9dG#^v=x@Ok|*oQ<9d3Ydzp-ox-Xvn#%~J0jS6>6qn|h`+bj;+o?ojCI(8 z2RJkKrqbn6kZx`ofilZ|7^+@| zyQ}-~oT*B@Y-SDS(Ga?};xx%CxnY;PF>6fvfuR|N!Xs6Iz^=^*2>%Hsm*x0o>`F_# zW2H}H{!wD4C#l%Q1 zn}MWbGA_-T!!9c(aE5^=w_CEECBJ2y`7DzHtTJha-4(h$DgqmG8lefPj(3(+HT^L<$iqm(a}7_@(VwcUF8>R@6sBzzL;EPhLTkU^y9bK-+U~=?U`O2 zv~xONKVJ{^U5!%zo=hh{{PfvXu?j=( zcyN8i60RQGlcN&a!F}L5te=;OEtPf%J?4)QrGmpQ&ZY_vr=myQNjQ{fah9?LXUy!& zjfafi8g2F*@Dy7^9Y`#J43r!k8bS=l74+rkrGXd@lhoL!&m!bUPv_LZgs;U zM?Yz|brX6B#_j;^j`%#*8NrtY1M;aq+KyWX>l$_`a^OUkaQ|&Z14k$j{Lx5w^j4lIYsQFJDH^)ZP3_r0ez-f@V-11w$J(oBawxF zcp?OgZwvnNG&`({H-ovH1uCN+AShP2z(vHlZsumLZg0i6M|b6AZ&lD`L5q~vbDeai zUvIuqlqt@XfslUf+nX*=psW8#m=0y=`EHfJQB=`o$Rb4vBxG2 z?$=YW{Bmsh%do!u)MY+zon^}=+GF|5awVA53eV@F^=Rp_m;D&TmKVg%bSH z&zFsm?<4&#h#qR#HSNVcWjkI7 z-c4qgzS75A2c%x(-D&L7Kr&d+nY@PS(WZUB2#3m1-0v92-wlycsu zFrSTWv;xZ$ao=MMnyw8+OtA%I(ItOwxE^*PvfVl{Rg zJ%IUt)v3>lJCcKL9L@g}CY?CmlNMdSL$7W|)10Ob&>iWIf#avhu1<}XNqvmy!hSQ! zrZtgjmCrzZp`X;*>J%kw-o*2B@v=Le z6cT0c?wLq~J;XUpvxRIb1JV3Qk&kUvmj2x=knPlF+VU(&_Gk1=>21kKSlJvV$D4`J zJiVB|D7ncFmA0q+GcH&o@+{LrRQT+9Gp;Hei?+I>m^-o+Khr8v+$M5ZxhF|ogK+xp zOC()e#E<7B%i`3c%zP{sNPp`8;f(q>#E6`jTH_e@Z99N-9mV;n*HnCWj-%zB-RQS= zEj|T=qSug7Ff8|?kcc8mxH}W)ugggr-n=2B$>&9{y;B(m=ehk5i> zYEdc06>;|XI@220*SgTN41LPDT0?X4;!&}95pw$)K=JiLTz%9LD#Dc~H$#KhO*@MQ zwT*CE|6MXDIxe{cG)XUTfRbixqx+ikr08D~yq0x9Yxy0TGBsG_yv+Ey^=tUv@Z;KH zg7bAR0}%&4L&w_>^YpB6OLh{D4ny$sXfgSG8bc|TaUMFGLWF~DGGfXB|;Z6bbR9Lx7Gu{{<6rKcs{CBf69*xW-r{F?zv_DU& z&3(x5RX%F{)2XM@O#I5bX!c;n9Xj;7T6h*J#17%G$Q}5?Rb)1FT07udkQzq6ERy`L zmPs2YKBMP}(VUn%nQa3)^YWn8IJ9>#bdHFgrduN|ELcU`tn4VWyjb=yEJn7Y<2Blp z;=^|N8s^92Mw`0@Mw-=l>rMM4v!E-xsm$;SHXd5ef&8{xMctv%R4iCC2J3ix)bRAoBv{@-%Hw&ua77B_a9^CFv-U3XuS;C{Q_vJ zVL3`uue1B~PSpKUsFb`H>uUNrf zKJ`_av|=q*E;Hs}uVb=reW%F!2*%9Z3+tpPuQu{IYRpqj71;iV*iGO3!QFy7VCA0n zX!CA|w$FuTKjJT$X{c0?%EJ$IVN5Eu`{_l?1NOsQI|cJTUc*auFG6;1N+ zK!sqdKI@i?UlGQfbi|!+cVEu~tcFX+$^xY(?P>J>wKAQqiiMu#Kpwuwjz(1kP@jEG zl$W+k=F!AbyOv(axm+zpT^&b5t1Z#|qC(cb?U+>gy@(=S>L6I54C@?hdC_wdX`QbH zc?$2$ml?Mu^<};2ZNL_!tV^JQHzQ$Ew2{?k$H;7AI!o3I`p~$}IaKj%EfxxgThIIm zsi91PHgs7J{h${(-dqB|TmPUPctWbaHcB>o+9TPH)h66`fF3J#63&bc4b*gM7$x*5 zq$LZ1(Av%Hx%m!19<9oQ#rbMdxiS~6X%QaIpNNXN4!s*{WSP;??B^(N>DPnyd~x7= zNFuwqzjy?@yxqwM_D<$5cVA(M{XpnGACKj-{-T?^h?^%cZ){K(OuJJkC_jjh%ZvEu zr6)-D@x{qC&DnH|tHzw-y7w78FF~FHubD$f?0;?chTy@5WCX>9pyGuhMx0qi57$-@ z`FYX~@obha_CfxIQtVtl0zITydhz2N*Q8HkM?K+^7_Z8{6H92$i9J%#=b7wq+67t( z;gU=Hfo#8QEGNi~<-?}n^acM=`*jIyw>QIag&e2bMPf_HEwmQP^R8D49R05i(e8%q zzton`#9Wp3`glv@2A!mgg*s-+S3S)F4K<~A$vqL@c_c&kFn8THhSaSM%{C_6m<6q{ zCI3zV(tihZDR9s=%vG_0`7v#H{=J5ctGnT*+Cs|T<0MTjx5tN;K|J-)2~vFhK~fy> zTQ=}tB>GPyiqsPQ(#P`rzH2W&zFUbsZ+=A5dKvsqMv$}OYe}!Kguf}G2dTP{UG^FC z&moU-<*Pj_8Y}a2uS4i@b0bE|i>&(P8kqiY$JnRmWX??qX5&4c;*9fabXZt~x6U(= z@pd=7hi^iL=K*@M#RjI)*O2qBvzS4*W8aNpH_(^?hxuPHad2mji52IPSVP&c!{cP_ z^bcayV}?xSEqzNPXnwW^R@mZXZaU3qs-#6`O2RpjfxQp%P};i~u`kt-;xHIHTl6tZ z*N|@BCnm$e=v-&vc|>+{JpRIn-zZRbZ_#8@Ng_Z70mJ^S9%7OD4^T{Np(9TlZV;)vsNsJM)U%aP;U!^(>zo>_3ck{&*793&h( z+W0W?8|NH4#a8G1_()9P( z@Fb1WzptVC9zF0S-x%4p-dNzB%6S3*vBQ_H{ODC68w#FL?5zH5RV`*+2folJiyRu- ztr*H5#r?EmAI_C8}zOvL8+-6QO?0ogNLXBF%>Bm^k|&1_$^dvmpZIWGD9R)w0gt zlcoH3Z)FYT*QATiYe>mLk(0&q=XBf`ZaG(^J$Mf9$i{F~6*c(<)M3eDH$jSFG|EFMN?w0CmKFebtz1Y{EmU=2+ z^sC>b|L+}U272?tUnbJC07vP~tCf=bDS5WPXvK0lu{>hlQOfE?xb#?LfHp?KGvcx6 z*f{bQ@8h(qN_ZMorf{RVH(kCljat86LG#f%v--Xcw5Pfor=(wzoW0u9mS_j+{Y{>H zFHI+fX*N_)(Gx+**&;7u%>y5KAie7qEEXKwZ6>{B7m}CZfL$n-Mhh2hF;j`WIP*qd zl?H!4h-`yXm?1^VdUezz=de4tvBQS9oIT9*9SqGAo?n%IozSL-+ZAYPL58gBUTvxD z@G04}J`G5yHDyhCb2k1JMhh;UCW|NIFy88nv_{NXYsQS=*<16Z9a#?~ho#BH2?6x; zemCCi^NJ6jRma8)Md(Cb}Hc@S}n8t~p4u6s#Yqw+F`Qs=)lqcKQ`5xR7xJ zDEXT)-B~0u@Bu#bC{+im{d;nrO#^%rC$QhE1fE~vPc<)((dK@_h1E}u7YwlEZhIX0 z{m))lINl4l9BT0|Ph<{Od`0)u3;9an25HW*7^$f0D+RAyOI417AGGs0KaW~Sca!|# z`LGN%+3%&_!;++deF9K;z?16b3u*774s07Y2&3H0Q1Qf9^fuFRVtN;j^mOHU#bbE& zuXNUnImhN%nkWiC3?DNUKA|M?QYG~;e$kmjMHb@o&kS@He8Q`%gShLg3n;VQOVQ>D z^hxA`l0D7&)R}qw$XbVGH78La3o{!PQz&b9#eo*zaAa3gdp`O~lf%5lo_DtBC=Co3 zT$UA3?jy4CpS6@e}6voI2-pVnYlyZzGV`)*R! z(z*E2qRYh(eq*~~EEa5@N(;+9(ecv-WPUk{tU_gTdk<^#=mZ1v#2fDP`?+u(-y1@A z_w|wHca9;))KuZw5N!Fl|8e%_(Nuo#|G#+-Nh)cU%tV}fUz-eRE}}sp8l*`omF6-> z2_-@bC5?)TbMI?QvqqXwn$f6{=J|WS*804@>-~GL@Av!1@2s`gz4x{Ey^eLC|XY?s*Xq{G=}s{he_2d?nPb-wjjMWjV$U7pS6RS8_Y!n6S3Jr?4}w3zV95glp&1$?oCFB;#^-^0;W9u(V$_ z$uWu|r4Si-kW4N63r#C*+;xV`4qb3T8cP zBgPtOq|H|se#jwftCvB(;c%FEw=3u^ zJ4GxN6G`p}UGmGs5{7)&1lFgCZ0Rm}33bzfW70ju+T8xIzoCX`7kZEfT_m20(L3V2 z@HDC4v!1lftRY)gsKKLo#Ztb>mJI9lQW&-)SGFm4v|zgMkbqzNlJ13P1uw6u!hq{% z$lgU6B=}DUl345^jJgvq3pLv7442KIDzq0&8`Tx2-6#}Z>E9B5G#(Oug!duoeXPjW z*AlDsQYlgDyqtVbok@lpc#_wJe}v8n$%4(|K$1L3pG=sAg{s~kur%( zuX@sybYGVvI44boIps@W|J@1D_g)X_{X9otY=^hgwL|F9tM22Lawe+7{~{nLo~I%5*q|BX=i=PQY5RfA}~dXl^@ zfNZaCAVJ5K;AqeWI8)&TkKTMHvK+~CuHlQ&u479)x5!|@=W>FPhU9+WL|K&OU72S2 zT$#3)9a zOMqx+Z`f|XoRldWlKHh81UG|l;jdMj^D3KeGLP9qgvn>JWP2V_S;`l40%k9Tu3hYe z#I`y?oVQkx=_L#C9n{G;X9rTzp+MRbJS7|R!%{HXby(=@-z;>k(GW6TRLL?H-yO`u%HC^HcSnpg0s(t(_vwu7jKoF?O+*b)B$T!>t^LU^`mCy6LpPqc=PBjbY=;QqIIVdcq< zvh0s%$?{rVNKi8b7QB%RFYQj6)znFP?Nph{93qUJ5=%;d=|dM$S7Hd&5pTDHWO$Ei za`f~cQdR3gM$R2B4C(a1*>#vHjI!(n%bOM9mZc_iwQ47t4sS{9^eJS2+i+q%t20rR z*tijSrtory2Lyf9B$H2jlErFySQr(0vlpmi4N|hr)*Ej&4;aph1E<+fk&_Oub z+ek3I` z&5<+*b0Kv7GH86f3MRDFl1I8H$VH8Aq$qSsm&? zu9(@7XUd^Mbl^!@`I7^}n~_c=T*`gN*1M9s$u@#-eumRY6I;nc=Y;UA;j>JAbUHco zc?sD8^N391FwgRsF0m1_g}2ju;99~3;=KHiu-+}0tQO749i_U&G$K1xDY0pr$$0zTq$R|eG^<3A;aW|k z@Unu$)K!BXE`LdDy#Zt$vxQqiozQ((n=n#TAgOcjllN&qNWYiTXWO79!mRDtLYmQa zp}I+vbYPvy*99u1!aD}?=Es9d*d};6Crog;wN|LV6eyg!5kf59&n8a_my@lwvq|rQ zRPs3O5aFc>q<`ZC@=PU`sI|=`t4RR)T;ok1n6{IJx0GN_g#wJalSTHN$R-u4_tcKwHwqUN&Riu1i<`;dlTwaRo=cR|?8%d({zTb3ljQ36AnO{f z$pI;+AM(3S==t`Q5O}X6S*f8$MnBgeFHHq?|2F<7{+~_Zf6n>;#6U{v{2M?0{yQWN?EupiG@(61;-I-oo(66l`kZZL@868T zCmJqz(c&F$p3#-7>ql~<%VRxQ}JY_oFDP@=X0=!oFnT+6|JwMM)|u zY@R5L8G0JO<~Cu{h`tL3EnA>X4<3u5 z57l)^z!hJT?i#du{infdBG-BVvWu-qN$fV@-7`aas#{h{9uF%->h%}M+Lk_%c_2WW=rAb$DNWd zm=iS#Ne9QVZDe%DMRxA04Zi-;g6Dp6k`{84*to}ofnP_|)%U?~`CswSJO!HAIv>{V zXarf;93GtH#$%k0V{KbB8rfZ8-Q9Mw4QuYO?;efvfgO9IV(}qi)4GnrUAL`7HNTMj z{t+*3C_e@(&MX1L(uwf$%P#!+0UbquH8FJPaw?y}$8y~)vg(eUl+CGp~xeyF;@ z6^B$`CJvV(L>{sSJiotTxg%GwOP3VT&)0;g6wH%czuQw`I7F`*@t<<;i)v+%z@@gjB%$@>`@25&SJA0hvca{ZK zci}G2JkjBJ9(wHW#C2}2U<=*LS+RH!uWn4kOW?sHT!-+X@rv+v`5KsewwI9VAItto zyT9p5>oD^8F}zFN@as!!$nl=bQao}b2c@4v=74tGdRU9kko@PwRdrA@N{zfb8qIb! z+2Z<_{W0t9Ds0@gi$NI>vTWVn7SC}zly-}g{AD} zkHcu7Gmk%tzekqTYy+D$QeEc!kzly$JeX>{XFKKkd}eVe-aPh1NU{=Ouz3s7pTCsv z@#xCukNkn%q^*9{x>%myWW@ayKF9)Ey~M%QT}YjIpir!yj(I(-k?y==uZL zth|W~3^RwxW4vK;aV!p6S3~}4+z>0WmJsI^D_}y9H%!kiV?$eakZs;Ac+s&ZmC%#) zhM@(ue(;(!Y1Nl6o3|THI-ennhXv!t!4pJ!(+~g7w8zW!T9QU!6Roxl6I%vp@eXCC zwBhG6=>A!WyEk}n-}|{dT5q2G(T9_GKe7%_@nEXwTm&DqmZ_Yhj+9{2t_1X%-)%mbD>4wlRPU72s>;`|dLb+++ zH2%A*HUASi4aZrYgx3}N70q2#P(|l28CsSK*2mAY*@Huo?_Np_H|~dfqm*cylRc!L ztfD`AKBI5UGw6&hlq>F9&->gi;>Wso;LV=bFyQWBc!T?h?>lW2tj~i{`#czDrcJ5y zM;K_imKnCjv-SdO2<>tmXQy}LZ5oLC))m9<(tKgvqW*mKUq5op?+Pn^7DX22ydr%| zC|1dq@KX!(#h{w0#G`H!nLnWr%Jh|KbJoO#Rln$Z8QW>MF*}~b`3)$o?Nich) z1>WhcftA+D7?o0u2UcCeB|0&zA|o6weYp!eW~73}$vk26#JzU<6c+8&^{~WAFuR*wdbyo=@$~j#+&a%)q|?M_i#^) zTpP=sXiwf}oG5oG(iij>s)~x~Qm!>@1w2>rg+D6+-}f%52zDHWO4fx!-4}^vX7yN} zi~+dPK~8RX8pF@8rb1K9GnTGyM8-vqXD-P)n7=Cs7UjAMdR+=2X5eHDm2DP?!+W;g zRTW)(Xrk&{C(?DA8!p6+;C07{JEL?_pom@{H0{;^Vm$cYExq*N#Pd2x8yyELC7%{E_EQ1fg>=<)A7P}sM|JB7RL6?p;4WLR-JgDRIaZq!E;PO0G zt}soJj2QJr(vm-7rIRGxT~o2}C0PYtICjQ^qjcc#_pQJJ3h_v84NN(=i>lsM=ADf{ zGoAS#ShJ1=dF^G0kdGLp5KL(IU}h_mu0{Pv40PxN3b!ZV_I0JWdY}z{*ltW5B=0!1 zbLCYlf1!?~tGg0D053ZzL-)Dm?C!}P_$l}z%r7YwJ*Qn`%lc@rrhakkq`_Ze6x|t~ zPcWsUxA&sQtbAeQ)(|lCl6cj;09H-?DKD{7hFkhoA@=geT__UC{XA&b;7{edhujY+Y1s*SN;Ay5JUu9Il)lOU!v!6-zc@p@n73Xvm zk7jh@+h&yG-9{%dXo@|1@@gth%`t)bAV9h?^%7UKg@%<{!u`7bRcX~m?-(+x1k?u8T z#zKWjAbeY00M!B6%-%AUJ@LPZ%Z-g7*>r+Xx_y@5)S;cszPO4Q7WlBCgS>gBUmkzg zL13%rju9PN%*l)Ao#>g^ba>m-fIpV*L+^K17Utzhc|qk&{LrDE`3k$?_W~729ndW6 zb8iu|*=3Ix@4mn>bDN#VbXWw7Hq}f1n$@6xA(P0f$AP<;2os+xp#S|h%wWtU%=>bI ze7j#sRzGxx)bcUpp!zfcbdtmg3F*SlNKsyO&Wvr8-V?u;D$wB*Q-RFZr)9sJK-Fvk zgtp~KT)kQ3Lt2DfN@iN@4!QWy6l^ExkSj+e zAHA3Z?2wx&1kP6BPj$Sxm8u2bpnH@UpK6ATNtc9pT@yIG`z^Uo8X@3H1bzz~L)H&0 zg^insf?}_|as{cD$!qchvF@-PPu!mli#=WoXVYKG=A>~_dp?txCk2VSl>2h!j6jU| zG@toecR{kfAJf@%Qf6lx2C>dvAoRr&C|$ft9t>-MKi!VO@qW-V^(&UG^5AcS2>M_kzlX+X?(Sna@K4k8um1M)~3|H7q$; zg~xV22RF-FK{2%sEPTT8R#^Z#_6&wurONO+<{{2KuT)7E4d4O#Q?UDp!Q!DK#+cnL z8dR&0nfL#QH3O^I{-bH)kHSXo7SMx_8f}ZMYftbPg?)UwZUeVi_>Ny0aTOa^6_IIs zUhz-El`4N9GUkcn$AMeqo3}f?87|4V zETuhE6*Jsm7sf7KdxlqfO8!PEo7uf=Gt}vD9H+LM;-6bptk3NL-kjIM*L}`J&)2WW z-s$UE?5Kmh#_JQms$o&7``|nuvmurTuC_#-XN|CTh@@>4Lm}ssIbgpP&|13iO@XVKyq9;MvJZZn&t|u4#=`BOHzaDK898*|B0j6hCQjyi zMRh}Uey*^J{Rwf0^S?9DIXRKMb&6)1E1W>p#R@Jgj3)VhR(!K+4kn@>6ctvHz1tmN zsc;&mK6S#M%awV1=gCZ)R-(nvJ5ak=1)7Bz{J>MZ*^n-tlg)**hAB8` zbAWTb{UCPlvkly7s3d(49%3nW7x_RF8}8FP8kLqGC;Lw=B=-8xSV_7jAJ3McyjhO> zKQP>(C3%NOrNbSG2a#f`fht}7@u*H0{JG#6OIx-YY87>HW6@kmZ~u#|R&~ZLdpcu= z`erz>ekK%5`Xv~6tH8$DL**+qbmdliGUZPuZ$Uj9$%|AN0S7lFL-n2U@a@NY@pAYF z)EvA9(^ZD!)hKluDsdM*?`nci*;xKI>;!7%=AnCsw|GNs1a43M1%Flyx+ZUg@d`2G$ABiR{dol6>^sfOzMh7#h!c2X=M7wKa0&y9Hc8%C zWAKH{3|G1g0Ez1c-q-TU=aoyy`^gWjS$hM=>AMn*%p6%w*h@BHiavgl z{Op>4jV0p(m9c}}4h}AU{My$a=-a0PXQ))M4?WN0u&J|P{I=huVW$PWJ^B#4z5RiI zT9K_A`y96VAo(flflU+pq02U1vT1)1+#IOGm5t8gHHAyeeoqK1`|1T>u7wcmi4+$) z3(UuG2aaFhE6z1h!Fmrhd^Dz(ImK=fa}#Ht9_u}n%|5gdvx3-9dcAO2ZkoQZq|<1&cz}^udVU+T9596Lhtl^k!)rG+jnte@k74pQA+8 z30@4F$&&T%GL>$6G}ZqBefs+roiV>VYw5gST;U!gzTc%zy-aV?1&uB=p?gPs_hApo z88QQEx7!Fqrf(ODKJR2J%eA169br!=Ub5 z`yCrx-?a!ckLQc^GoMvVx}7QiyJZVqul0?#FOH-KH-uwdA0UQ%wBW|d77|}EgZ2!k zv|O&v4=PTCyt%rp&(a_=YsUmS!YCWSP7-BY8z&?|M z*=Nep@`pU*uO3VL zriBO2AZ;+YLhU;pr+Hac7#4pPOkb|R=&F7jCYsbtT^3Ebv`43rd2xNpEf@~mE0{AM?c8Mt@Eu!T0Xc487# zbD78^MpYPm86p2YM~iI!Fq55q6wf}koWy9?@7U{l3HI(nv9`Tdh|27YMZdGyucliS^KZUn%`*+?O+CTo?&{yPwIPjampq~UFXqw%6OH+hUPo9>MJM*q zUq`%8QOKzn-A|)Er*%d!~R`Fy#@%^eod%u3cY?ka| zHYst!?;^_S+!Cfg>Pd&K)vqkO~g+R)ogRYFJft6;ANm#Q)Q25nA5D zCA$v(^t}tCOg=?oUyQKyf>X{>w(zGlidyw?)PrkUZG=egu%ZUod?jG;+QVSItwOup216c5kZ z&%O`tAl`g-hP_hjisLdaLrT?q=o)9i^S34NX{(~S{NY}4sg(syh*ic0T@BQIzM4*= z>u9p45@|g1UE(ao$qe~%h|GzV_jjeRwl)JbXT2b~;cDE2_N2$74nw;4e$saK9(%yQ zk+!9J)bMCH_3t)_2H1L1_WL6-I_L!sZj_DiH^dbaJ5&2(Kf&C#fX@mt<{`RUQQ>$J zU1IVVT)Mm$+3Z+KpWmZqQ!i3Ub^>loB4Np3i6iaT6Laoa+N zZ(E^2XKz?QUCs@r@`7RTsO2ss`1ODhCpyV0CfYE)L!(&NuxxhcNi7a-1>VmonhyQ2 zn}(^B(Bk>ySvP$Pkt8h=P6X_Me%`;~&A|`k(Ue5`>{bC?;4gW#aWlT=d0+04An5_z z8Qc0M14rJALUz8EonO?2ZqMsMEB$P#a?Su6VOU6RcS;mp|6~g@S3hHB!{T^<&uTO^ zI4!Pt{F5XNF{PSyI?&ZQogU}zFs3Y+`mEs)s^35cYOFw&;{;NM?j{Yr63GMQ3B>Ag zF9`aq#2(&EXQyp`3oLOF3|?yoS;M^9qBpxp_M>!|Gx0vWtJ;S?TXtgyr(kx2H?c!A zHNkt|FEalj%8Tbp40}U?$d!Kx)0a#nYlbp%S@JX%!zSXN(7VLS-jvQWO%qW5lw2CS%pGlQs1UfcP zkEVOyhrg3&!`{*q(5ksl^tNi^goFN6T%t$2wa$jVktS#qe}#>ly+hn*r{vn&*?nYbcO#ttqIXBOvz*v33FJbI`Q*X&3IqsbAV_T)Bs(zs9D)uCN<>*ht{;c>H8Qls9^%*Ry<-~H3%cBlHV_QZD#l;#L~^_{?TM=m(49mmau zx8d02`@pQGk{4YQv47cgrZWC2yINGrqMsY$dEf1#z3eOXP<~2Xf}c=Z{uo;?-xj~z zu%X7qiQ>wT>0(LBP^Q~kBsmS4kS*&&-<6a@a$E#xMHRC<=Fz10a}5@@UVGI*GLzlFszJIy`ymC9d|KDX7diKqiHx5qWeX^9rb7?o}ddTzH=9o*PED zxf;=54bk**-*vQN?gqL^qZYfatjDYRqp&0IE;fXykjuS`@b9B}7}fbI8$W&zlWg6D zx{#Obh?y&}3U#)(a5Z*s-HBS?9r@l7GjMa&4piD4iyt;H4878g6>nRE3AXiEs&f=) z4Hn3_lh64URozG5>I4_^NK1nRulOs62t7h}h^r+Np=~9_o*t=5io(btT;S@&i>A)Qx z9A+=3br;@rlnQa?E)*9I>5OL+&xys_i}BmkMdVJzcJkXhfv-Ez#Ix2vn zuz&x4IKAl@ZVT#zq1&3+&9KL8%gzVn&tgkH##WCTmFV(w!+&CzIfr@8$Nl_K&?k&a z9K?U$o5OQ7Cqb)&5s3@lgtc>~v8cXRS*fq2@xNt8#-|aIFuz&!X^1Ct^oHT|VR}Tp zs3U(>x&W2SmSFY3dc5t>5A!4AQ6=Omv$!!;?A&*|7psyr*6sQkp&?V!tyeH zx?Y2a)pc?8!bWzt<50G;>;&FenTxF}%-OD+S==?{1)uZuG#~P-1p9?|!-$8&vEF<; z8y(jZV`Dk{HHP7r1A2V4ULEdm2*y(d)z~%cdwIb1_2}Kb5RZ1L$LE8UamVUZbTinC zQIE_qdfN_E^@t{KpF|3$9Zb+g%}3lSZ)Ht~tk|j7#=_wNlCgCAQe3x7j;g9Qe8?AH zPLDU>&D})^T~_f3YXx4vQVCQ)kC5uIR9I?w6;o}{5DK4V%Z%E6(9-!D(dnAWZhW09 z9=+@=-a<{u_h1yaSmeZeW~=h+ujJfd{{@u!oWdK=XUT83wt$`Q3;20!9V;+h&8kYY z;q6^lwt8PVTajc>gWyRQf1%d%}~B6o#CiFfk*VX4q*@C-cX@Cp4pZo=ZedgM+g zPf)J=M3|5zehD_m@RX65%1p)Ib`H)(E*5N4z*dOAWPlf=&2hi&Ou5%;ZSlOJHog%B z81QW$==q$36K>rxs&62y4oZbihrWuv4NpOCqz+ZojFWj?-3OogPL+Hvn=44n%Zg=* z{%lv_6Vb4evuS!CaMbwo;--zgS%6t@bl=jSzgV)DKO7#(r(aMOf3_#F@`^UxZ?Az> zgW}NgdM$P^jD^jo3*hqJHkSH$0S=RLYd^a!hmLmXJswM1}%pm@F6e(0r2r z=m4pDFdKqj7Qw*F-|)32Vsdo>+hqHglo|)(h;T>T{%t<7bT?vd=lbJ)-v|g=UrxrK z=m+y9@1414-@v%@?&#^Q&J%hWpzp>G{N0j0FtDl-CeCYtH3enR>5dv5zq}hQzkQDl zUo#&MbTJp@;u|85w8AmYIbyHI63K7A2ZlEn5aoWi*{bj%_}O{`^b?0c>{dVQ{zZ{_ zRBLmU$A~`qCvkvhJY0IBLz}nhQd_$(V$y9TsugNUqSbduUb;8MO~-<8-s&xK-}yh- zMtup!HeCy!7gU0^)+2C=f6Oa;9OS2hYxzVYXRv&-M?UhBC#>zC2QTWB=&TJY^z5U7 zQ2RrU<9B4>gs;g^{_!Rug^5sbrZ=XoUy4ub7T`6%agvXB5JZo;0MXILbkj8z`r6(Z zYp=wB&ZZ=ofxF<;Fh#of)dbKp7>re)XR$wZoq*=H2f;37s8*6Blomdj)drIq8_7U0Q&@A1HJGk41dW z%tB(^y9WR4bmqnx{dwWM4F2a~1t0H5(WJaPwVK-iH%Dm5c3$2ht6Nwh8z|k^4|LF` zb$36&rwgy)_}3t?i?GDh&C%TY(skUYaRiTge~kx~fCw)^G55D1&MaVGY&>Y&_2G1*-4d9)-<@TTPG`}j;iUE463CpAK|F3u zKub$!3Wmbr*_G z=?aI6udx|@+eQDfg*->PCT2YiaqCY-Zp?DamAdz#AC&I%$eXWu+=o2J^~^4Ety1aHqk@x9YugE_2Jq-wB4yDk2#!eH((c17lIC&Qf-( zLuWLa-a{&&R%3UsZ^IDHM)ANQJ%}HXLEd{6ppTgj@46`$H8QW`&&m0ujKN>{R~ZIki+_jLS{hZ1jkJBzY( zz`)m5{83a8FLN*AKQA}2idj~$vgHD#I9ii4J5`vXf~C{3NKdjR?G+nj@rE>fjK-9- zrD&*qLp+zZQ8fJUh)tf>U%YHFfJf=v$2DDbsOr+)2e?7SA_d#HB|O zbH_@xAEpZ8&YkVz{-J#ZJBMI4eC$P{u~1@MFBt-S<2&}GF#+cXZbhB$O1NTHG&?8k z5cTSlWlAma7-61_kL?xsh{GxT-6c!n`ZJ5AEL9Y>dX9lIlSKG(Q-gj!*B6wRdc%|p zz7XI27IATZ&Q@tY7Z1PeO?$726gBPQaAWo>T;!n0^NWC<1%U&oW#1)V0<{J7w^6Hub_L?lr3CkhK|x_W!%#&9DlqwTl!du zXGC`8I1L90-AZJ{vSLQ#IsAWwKVBMmBy-Dv&p67XJT!BsuY_`9_-zJ7Oi z)E_zy?Gzg@tL`2A*g77TOjpP1P)A`wMlRfLwk7#qx^%{gDmdZ48pe#zgFTk};p3<_ z_^vaWZZFfME_4B{c>SDANPQ$U*}fCkjwCQ8Y&OyEyAqze`ci+{6W9^H0eqG1$mTw! zWL88tSva{DEsu&I<6HD;)Q3*g#6z9Fv>rrTzqS&zkrC@tUrCS0?(E$1m!$TV5i1QZ zbUC~z+a-PCc9-s5Ea+F3OfwFzrB`VixKs{>T|)$TJ*z+K(Nl>mldk75i#~MCX$!h! zcpkJw&vkLhc5`{2JJe;p@mJCG;}gMGWfsgYtA)C%p44Z)l(*}bL0ej?$?v6yNc-4k zxbpWX3ELvFosoOt)2jwJV)YAf={a^ZKuOxi$poc`UervslFrT7q$e7y!R+#2>V0Q2 z{q-S5Sd(~=z@|=M^P?2tmo`*56_7Ox%t_iCCL{$~kkcF1(t>-z^v>uAS{-l`5{jH5 zr)?{|{?S0Ko)*%)*aF(nQ%*})7SQwAH)!b2tq@b1N~%_MgP-@E=&&{6lJ!q6UAHup z{@5{>o;%x_+MY6YnW$^yvT&WT%abN8n)LSy9X|UEz4FwP=B}O&C&M=2nYiU--L7xs zhouu7)irUMVWI7k(5m6$H*^qPTjN15AMQqVO)RMWy*e;09twv&D})C1V-ipMG$){ykh!QQM>V0lPmfiy}-5UZe9s}v* zktbp7h6U`fR##B@Vn;)*<#fgJi}d)a;dEePUzYJ{f{?L$sJL&+2wKtk0z_WTf&LHe zXo>B3TCp;VUY%zR{Ur_MSM#UhGPeh`vap7p4!K3+Y^>;p+WmCfy;)M;Z7BPF?W@o^ zyO6wctQPuj2?f=1BgpNwmj0>VM*CO#(H4&Y>L=Ubc(1@~WNz`Vod zWI)Fol2%C<`z`aX_&an4Q(gFlgxbdA+S5nKh)rn~2eacaY42y5hHs!~YvK)B@tML9 z8YGN3_Jt+QiGkE>u7W}OHrjrck?2Kc=({cn2cFqYCXU|@ZED%DT5k_~Ix|%aRI_9Q zUK-PjJ#y(s^V77;`b$FO!Zc9pu#ZH4f6Dyp!^QH!n$8Z7wqni_NAl~8J@kK~g~bjW zW1XYPUgcgi-tQJ{_}ZCvcew`sHb)ryxeDu7-4(X0NgAAYA=rQG66pMmF|Q43{4XTI z5TiA)Z=IY(Eq51(OFRPG?oVLCs$pOj{0OqYj-z{ReaN->%d`id-|oP*2Yc{aB8di{_y!Wu>AqUpxxjkG43@U2dwPL^iG^)84s?&L;f1_ z)O(@ZsZ#lnX`y1u$sO!Xb2J>AdynXa6~pC$!$|kUx#axdq0n^rn#8&c6U}w5lkl#i zuwG*;E4}@fZLfbPtDD$a-l^K1?HRs?#f5AVE?d@%$Lue&)6Ge^OO_8lcjIaOk%M&o z`DwKMQ3UKh?hm$xcVT3Y2yywU7Sh_sUG!Y^UR=>?i-{>|WZLwpFgRcfD&6mfTYLsk z7n3DY9Z(*woBb4%_MZZ4jeJ>j@C?{_FNQe1yo&ctR>6=1hhfSQeL7l}i-U`laO;6m zng5l8aJ={<2<>C2Y z!kCC+>Ns^f&emvTZKoHp(H9y>L^UBldgg+MYY=3}W5}Up{oqcJ45nJo!wOR?61P#0 zwyRaqz86P+86GV9?B+1TE4?DxbOR>!6?r<=wwV2m-%?mdp4xs?T5T3X1B=HqO| ztvr_UAyyowbBT;oaRak=9clIMROmGz4*Wj4Q;0Z5;`&d3>su6Q)%XIoe9i#qdGQ0g zbkQ90=J?QAB?yPN9EUb0<$7J`QPW&22tIRw$sER#)8>a^Vv8f)QT+oV`#pr-+R0EL z>De=<7}MIra+q;xFM0EMG}V_+quxf`Y*7yS>=v?@|tC za~o6HJ(%gK_GWn{^(PBeOw9bMlvg&J=(q(g?A4{|j;FLb(H07JCD zK`6X4EGIh*yYKTGd9 zie1{d9BsdD#UZN&8ZzW5(QFunQKd`hyoL7Ep)!yjS(L>*^pweH#}3S7O$ig4Ct*P2GnTLOkzAliEP<1 zij9xCC}>J`KwpEhK*ehYE2-NA8sT?ErN4Xdo%4QnyG#Mn3{>&KQA&BO95$Ed!jz!z zFix)=M*j7|;38*|sj~rs*1jT#@0voFo_4V0(OEKSe=v@_|CgoMHw(*^%Av{Q8g-cN zN_Thpz_u0afN0BBi2rH9>()-k`-W+_Xn_y0lyvb;hC!g9sZVm^e8sk2s_g4MKk!*? zhcDFP(F9u5$UbUmS3I9qY3s52QbOGCVd za<+5qWZ9;%&1}iduEM)cl;n8il6s9o^2OGYetMV-N)B;wZ@&>+Bg+JpLzD*Y_Jljp zYPjfD4?f70;iWVeuH5+^TiF_bR+6S-O^O~b>(#{8TykM=|A=U9CDoUYn@uM7RfeBM zv+3bElj&{Vh|>qCV(WJsbh;V~G0vYLCw~WA=SSG)lWVbWSP_<2cjxo>sq@LPm+)}y zDM(0GLGMS&Y;V*%am0|pFv4XRyz25z{IhbLuqSjlk-a~KhQoAd)vI{AXxn}^!rT_) zmd|E8la(;W^Az-78G!pNf8q|4iKwLC4d$L3&i3BhfMHMcXzBej_;GU*=u89l;_NfX zESo~hvIc_1h-B=TbQf0Y<%%Yi8_{aqDjc69sWB>xZs1` zAJ4(C&uie@h|YAT5G%;0yRsbjbnG+F7c%vpReX_reta*uVusN&7HEND@g9BHHl`m$ z84tyW+B)3g`eg|%bCkI+8%({(Q8L;q6^%AG!T@_uI?^qKUYf$N>Bld$EsDpQpNqsB zd@zhEb;JD{Aw;Ka5LTV9M_&DdB`dDLwq+a9&pV5KasPm7vO=^-Zequ3m&;zpTf?2k zAj}>;ng{0^@&}q0I8@D^)$dOg=ABaF&)q5i96gIS?A?HS9O_V3+ZoTBG~w;)Q9S5{ zIzOA!2%D33kiv9*yq!>ufeEU7%g_S$OxU=1OAs7hHM?S!NqxJdZ=x)TL)=xGd>kyG`wZvVM4v0gZPv!otE*O8PPCk95 z#CSVBj!pJU!06WZxGzbAJL)%LuprfA%erEI@fucH=qP@ZW#b9QX;`vz1n-&a$v#Xr z$3nw%g8eNsE}G83W%|RgNa%sZ%hH(b#S1W?dK;UUQouirRIM}q@-+Yb#Do8;X0TUB&Z@mq?!lOxR}NWm*JJi8PfK#oFqz~ zCd;oVfyPLGHfzONS;^pNa#Ea)l%3-wWC^hv1Oc3dCr_CHR_u8>$`q(4ckm;3Dh9g1QZav**vUcmIo_^M0iA zi{p69F0v9yl906JKIijjNh&D~6}}}x(pHI*vO<{|p~xyxRLXeH=g3H+Qj$nRJDQS) zO6mLj0e-mLdp+m8-|yFJ=tzF+)H@}CRR&ZqYdNp~YdJCUcBS6;iD2$zh0!-gVu!~) zZqM{q`m^;T&bjfOB*&!Vf?W^sAT5z>K6hR4N>jpuMm4;Z`+**`4yQ|+H0TDQl3X!9 z$JKikq0{g&D0w!6Or8=#y{{|dl#{{uVrDO7B&Ac;xNP=Pa|lkRG#BSt&F6@L#3mwkUX(B4A=hMae#H0E6 zZ!mu70NkA}$1D_w;0x7T+=Qen+_}?r+=CHD@b*7tzTdNS!Sd&sQvXF2G|%-K=NRz= z4~Cy)9ZgC!;XxanaZ(?H#GQ@TzIjAPb}Z@6YC6$?X>3%o_gXC7@G`1owWc#Lng~H$q(j=ViNs$0vQd%*%)FyrGKESIM!3cY~RYMF@Uc0Z6*nP)+Sh z-p*+?U#ZXtk48MkpW*Yls=@YwmSnaIG zz6I9c$VEHRYS>*I@GX|>Xm-Z;b=UBG+Guw5z#uj>-30IQc9^v-lDJ%tCOzkpaCUAZ z&Aq;djw>I5IucVX@`=aG&)4Jhzz}p9+)JeAHVZzIdAM}ZYIHq%lg<=f=UZOfCF?_` z3wxX5g?4jCIx*}uEj*Tm-3>N4aPT8~WRon`T-L;-W#MdPfniyRnp+wF*$p2>&%;x% zB5{eP3H0oIAuQXHPKUWjAtX#i`NA*Me9Cwlo;HLnbJ)PXm6bA+b|2<;e=5$2ix3{! zjAEO2j%3fL$m6{c6*MI(t7P3mMLKo74et1%K{kqfyaL&|Vs@b^zUJ9N);CLZllo0R zyI%pj%^T4}yqA9YA)cn255c$-=Cn^coJ#xWBU8>`iYD?*nyIo&(;9KnRY}63nve7%=lA-fn3v({zVrs57 zX&-7!UL2oAWNOoJhRu3BmeEGFW<3}0Uqr{o((kzGxY$<+$)xSvSUy7aIKBGI0X6tt z=ou3>HHu|S5^3_pgn{}fVYJ_H{BYr?4GH?U%eJ*pN)am7Dg)7H1GsPOSUPT9Q& zXTJG|MH{^6ufP71kct09KY~YQg$^4856dDRCb&1lZ$1c%m!A7X&|n8y#XcJZ&B4+5TMeJzyF9TRe=E ztX@W!+C-3-mxl1RzPrJDM?=ZgY?l)E5Ovsh^#+W0mLt7c!C-t}^n+~CgrJgiNRjf#egobgwGCyl1BHAydLh?^^?Eo*BGlXFl&zs{vsrhTuPyA(*)TFMV@=R#C#J z=n_kVw~{aQQv7+p1{fB$)Xc@B3A!%*0=r2M;o<4kIOxJLLFHOMW)K+(j%$)3C&JoH zRm^H%bVd;U#`Cav&rtr!Pa}Syyb1qU<_c9Y&Y)Az1(Sqd6G_W|U&I}iI?P}2l(f3U z!sF9#N-Hxb!`|li@H?@TRQp-snOiwzY;qL6*I$ks!9S@Cb`GCFx^Iky@l*St?(}omuPo+wmI}Z^x07Gj z^RayQS8mbkTCTXq3~y~)34^Z;gN#-O`d24|OgNNCjdxH~Rp(gnsAl|`orA4C>NsiC zQ!4*%G8*;w<%*?tLCv4XjYnT_-uF%{yj4lsV-Z%EM3P}= z48Wsa3gUKGP*t1$e5XGJAN$={G4B)(*metjA{5#30Yl+(LOHbR%knZ?OyK4!QxbN4 z3n(j=k_9FMVN3Q0e7ZuBzwY-4QrxA%&AW-l1gycKQ-*Q5&7wbgbTx_HJP33|=Fy1# zGX&YWWpGLK@9e3aM*P(xiPqBsh)z673Ko7Lv%)lCT;OadU)V<+3g6?$CwKAr_cW9l zzmK*|dQGOV%XIYe`=o8?Z|?e>dakBD2;_V8ap~Rfq}fi3+ni`ko{vrbIpHen;+B`v5_(3;6G?pmM`-6&iq}a8^ z0nA@pmzBuh!0(4+P%+*RXHE=2ON}rp)4CEj7G~mrMMYFMRpg__PoDDkw7XO1bUYy$3N3H@?t%$aDxW5pz{ENK|kvRX?5PJX8r{}GfokVoq|9`wSd z*K~l{RMhDoNbl)S$GC3+bi!a(Gee2VPP~o$ z?QVRsaW#J7HK>KJ==>G4>Ne|Syebspw!{m#!1ga4bR-HJ3T4q+dp637XM^&G&+&bc zC6^!fhGwMqBc^ow9<_WEfFYh%*o^-%?YUm8)$0=bCD$iJwVkIAz16uurAE5` z*&+PBUyWVgGgqP+n_bd0$HeSx?-(2v&`w(Bh#zvz{;dCPD>lDamJRKl#bm0DnTQ&| z{riJ)-7$)flqzxKepQ(9aWIuWWI|&zl$gY52rKe4VXuDZ;f9OX@uhw~&YQP}nk+M- z&C^d}OIsdUXp{==vvv4)wFk)DMb=z+Qjx^Xc?c{?JOUp#%>mQ$er!>^EvXe_@RxTq zpyZD((M*XU7d#bN?WW1B9}H&y6&yopr?IGAbOcAP3C7(WTQMa1K9N#<3o?Zvbmip= zbh?*7mb^Ynl)u!{mSQD*{?;DPFMTguF%~zKa1G5fMzeR`Ef6>S5=1nt1^de5m|dZX z%X2S^o}(vBD~qwGqXjmrT8Z8M@(qgyNU=iEQFg^S2=;pnq3=E@fR)uY(JT0!wr2k( z)@i1oa|VRtj`7&NHGjkdL>^+^g zE|sbrSwz!Q50fjIbA;^t2*^8R0>(EdL7eDI^(@bWf2UVq@bMLR>Rmidc>9!w#4g0H z!Q$5S>>^1eT_hlPt0PrzIwP)RUJzP1wq}T8nSrHEBa9>SFk(~B<$P$ zR2cnHnaVa27JqIO+o+|)>Xrw?io45EteliMSAF9a2~)uD;~+eGWDOm%LmnflW)g+T z&P+wxnPpe+XRH3n^CKrm;%x4t+Qe8`%G?`syqU*1NUr~YUx z!|1cSk=({f)%5qvo26auli5kJQ!wU?K8si}Qt(**qjc8ksFJ*rMhu@jjCIkw7%QvF zoxeMoQ~Q-F8J+t;P&bstXT*r=rs<;7o>Ulj>Lz?ytwpp)j48?YLQcju6n{Rb;^bO& zO8+$thqqfPbq%~iLo1~*D!C0qrMIx)Q;%?a?j0d$c`k8RjE3Qz!PuoE4+>ZIv#<9h z?C$MPSTUsqR)*NYn*$-VAV8WZ$m@}Q`y!y&-je@g`yK@G-gJt~C)U_~msR|zWiHx7 zS-SU2c=t(x*P50gjQ{&jNaqSTn~r2EyHXwTgd*y6-)1$wC9K)+0x4+GN47drFr!)I zwBB5jASZhMHW;E>pvXuFS&5F%)miRTSNP*%!<#JK%0I804kj(Se7NohSk`_KE}0(T zhh-V^JE|-2I4#7am9Ownu&mHEWh}J}4uX^Eg=`-#VFfE2FvM2ecyBWS`)QkD*nmH5 z!`1#}&%fF*$?1ty!{>)tZoa(Gd8e9YbU(xBtBF{+tA)PJF9CnoWU%@9mMjY2LO%S` z;VyQ2VNSI(NCFqZx{VcZ&;1B&K5qr;N3Hpdxnp6p^y_H~!ZaNBXf`VSmZ9N7Gwu9x zN=Que5nfrvp+Un8yj()5uw6>bSLOJU#jlBbv?ZK*-$U!OoJrPbWj8%D#Wlf)UFjudcuwFbj+&k1kZZA$HOS^jM;<)uh#VMAIs96N#R_2SW+Is%j z@G}0oB$;n=oDIk2o2WzOVVo~^jz9mcqA{W~Xq(#tklQjAmtV-`0(?4{s#*g}@)B5E z$SIh<@FPIpVE)Y~XWsJqXL!*(jDLUU1dEWmz|JMuGtHZ!y!5LtP&axP1pX3P7S9I4 zLW>E4>ZDn`(T*7Y`!z3kb*~EkWs8}}&>qsgY$uzX$1%GaOX|EWkE6F7$Yj~urTx>Q zp)dFw_0Daj#pV_4%gT86BrBB7c&g2xnIp|Vw2~uRoljD7^avf)#OT5GvuKghT3k`z zB@Fk8Wg(YquqL@2y${y00c+xzN2x3G6t_GLtD||PUnh7Iy>~EL#t)B5EYRq(5zRc3 zhZ{zP(es;zVc&c^93NLwTKcfaYR;shn`>mhUOiN!m%%Nh!Ay?)@#S%lmoJCVVpgCT=$bDoSn|y*u0hbR$WEA z(l&ZF(+2O#n9$tiQD&W|^@KI5BIE2)Etltcj}$mr^QQkjC1nji(L(Z_ZFe0~=Iduz zhC18evz0Y`nEM*i1E;ef#|L8PW&@jG;DDV!>cPQ5n_LxLid7$fkSD7Mo= zYPuqc7Epck@oS(Xzv;8gXJ29bj&6ijCyH@XCH`nKgXnqOUDAb-Xk2G&m_UNnRnp7{ysQ-crJ`xA44qiby(8BQt|Vm z$lkn+z)meAn(L(|I2Cg2$)Z(kc%v&bkPDz~8_OksHs3}Y-VhId*2I2&eQ0oUAe(0V zfvh#N;m^HUKYd|jC!P;4L*-55@Y0wFGC3>-G>cYoznx8q{5VtL{`=R&$axK~w=@>U zkLM98^zq85da`7)vIyXBrcU|UAnl!oo!WOuP`?A1XE35HZO{{@V6v3m`y}I^)=|R^eP^-n+0+gUvLFm(oub25Xc6Zfw1cmEx0Uh zrj{zRrwxPHWaDIVx!0L`dM+Y0)z3;zymzDEIT4<Gag>By}Gm|mTWTVCXoorfdw>mwI7G|~X_@>amK%6fD+E64XE6Jg!XNHS;86`{KK zh0wJs0!%mEqTk)CXkJe!y;bs;Ts!lcEMBmjxIAtr+mG9lBdnGTsN9TP_$E|%rGr1F z*pSoKe$+k84t=Y|=P^2r_;_rja)p-EZh;-W`v*z=6B+jK?-j)ILcG@lazeIpSmGyMs8IfoDbt+N#Fq791Gm)CA6e1ZJUK+k7kKFrtig@ zrIm)zcwDy408=Japnr!loAvEK%s4xl=o+p?Q!@jo&6-2A9WR&uc2LEcwN-+k)<(xi z4#OS0SE2bFA7M`TXxd{i0z*V@hVN1y;!F;ZY}(3AE&d1Q6CO%%pW= zZ&1Tmiu7~%Jg~bq47)9|v1URq>TzB*xZl4s3@9wtXZZ#kBu6+e{>HbEQ z#$LvxLBCKhWI4(WaiH&))Z@;V>+$fz(UJ+?@q$iGf6x?nG=UOFIG~>j(;jZZ4Ikd3 z`u1#$-yDS6f#Xr_j3X)sv~mYMG6ds)WdfZO2P31NLRx$zG$wR#H#(xpjr#J^uk^%h*)-0yy$}(u?s2S)NWlIOXJVzDmTOcIFm#G;HQ)T_Fn?$= z<7d8B7Xk;mK(L`LiC&Y&hq+epW=p>E=LRdF>C$+jxYL|pq^8fuPZ-M|GB@TE^f;K& zvK%tJ!Xd{wLfmyO<{DSuhN-8Q@FltZd1D^~FjlhV?(Y^iVjD~)yC&|Tqaufr*IV83 z(cc(8b!I=_r|vQIsmy|n4}UX-0wsD>%(XHfdjo^yFziP<8USF-A@~0|C4~k zd^A7gqB4GWP{uJkzKMS5%iN!Zr-=W_D9GLsNRGLz;XcTPN`&00{ zqlm)|3+SVm2vnLl8SEFNl7>$+agXTGGnsaTSl>6H=TE1RD%bTGJL@6IX*@_d-RDBR zoe3{nk;I4bPx!Y>egz=K1y}ZMaW(#u$)UGyyOkO$7_s}Y8`$nmVXUa5 zoeuZR!&6P?aOL_ccH~4P+p)QsIb5E^+TR$l4-pM`L24U2^>8jL`(w@KUp$Q_r_IoH z@h|GT@(S(1ftWKr44eI8NyE<++V8Opb6!@AN?~~zzEV8PvvcrE{LUUX*bD=F_c49B zc-H>w5KCM8hdX5)M^h88)5SMDz-7Z=c)wPQcnukfyL>j&3leFLY~q=Bz7#76IE^2? z!*Nl|T}4n(>`%t^t=xz)@Bb#M*ClUn5;yW?wQO?hi_+Nnrv8f z`5-p$R2C|w#dGx=kI`&1?K1ag|JWPN$87#Kd&Zs>qk(Z4>SopA>ozSi@6>*pyAn~~ zL>a9O(*?hVGxAuA{k^sxzdid#zXj#euG4v9|0#vqobY0cKFnkH z^5?QOQ^qo3aSeU(Cy+XbC$v+sx}c;x6L-$9$JXCGD2!PL^QN7`DTRc)9x@iaZAUPz zOG-G)N}f7~`-_aMDwt|J9{+5g!W8c~F!jV<@qTJEZ9buo+MgP5{^W5~KX@_Ws+B>- zA|AhW4x*Qnf=e&2osP;r-6Zyt2maTPgbNon;@ZJljPsnxJ}W4p^*?#gY>_Qfz7xyZ zGUKpXWeJv6eJnk_Uyh!!2qXBam-F~CjeIM-NiT*ukY+V$*rap=4eme1TdP25Trh&Q z4!MUPijrv7;?v(Gg$#Jcd{mnu5!@LG0zI%Xs;y}OfOU>2?xGJAXn=}dna0QKQ&b#&v_=hwrDhaAa3IpPM?dmI#w`p`E9Bm z=*&jMtF!8W^CWz7HWzufozgpz#HZpQcAU6Mk^*h{-$8p>wd^@`<=x@4M=zP*o5;zp zJ4p9tZlW2Fqi6+hPb;@ggZ#kDWX0TtW^u9M^eb4=v~L!|Jp2}X#@5y z>!OETSAoXMEHKNv5BKI)fb8gg{Jh`J^kLK+ta)9Jf6i9W{q735-nIgrwLW54lRuim zywaI|cWLIWZfa30IzXP~K)(NAPz}5(EZ(m}kJddQ4-7@FtodW%(ij3C{{5lRnhWrd zOa(C=C9*5N3?SVp6R}~(E0WvZLtYo3BrmNFk*|6`xeKAHk_9?mTyOMD;Zsg0JeR#i zp4h*j$^ECWSA!F&#i4$DoY!&I8+$JauX% zT8cI_>irlR84`|t#{Wn#{KBCZ2QJD%`18X4wX2?Dh^#Ou!s$zQ367 z-;fMjtRLZB#RfCEU7$)aZ+B5b7u1?Uzn6uMh?=EwGM}-u>tMD+J zJM0QuRl0=D5OkPF`g?rXW)C?(24K0{HL7NJq$K*@Nz`f`0nY;G!_j9=xGv3|HLFg- zdHehG*@4C4`SL0pmyCcp>JnnB5KWTBU6hlC1%m;nN#=+*WOw{(_I}18*3|DA|0n7c z|9a0Lep>n}W*;8O>~1RZ)!{lQ^)-<0-X)7#KjpD`<6yC86V8m&28wRgQFy_082j2g zo>hE$0I?kh_~WJlZ1=OBIQG*)a_w$7KGzf(1fOQ0OG+~yA|J6Up^04b^d&AX!6jxE z;}KikSfaiuzFijwf2(D2S<(@Fp(720)tAuJt{8Z*C6>6zx#HXX{c*+IY2@A>3Egol zoE)BJN&jsgiSaL1(v~^h^rB!wg;&e?M}iA~s-0)^mK|p`69=={@5>;lJC2YA^#V7t zlgNCT1lOrCe9`M7_bh%8sUsdR@KiTZ9vM!Y-~w1ouONyyd}00&8D7~@j$hua$}d?l zoF^F$Fvxa}$j2E+k{<*JGEF1Mfn`fzC!81P1~hw}z6MY>LFaOsZ|FG%xRC)lEE1XmRNNR0bq`n~p&pnG$x zFv{{XSFyoJ2x#-bsQoLrQ<|yd`($l!7PIcC?Y{I_vL?=KY$QvDidlhf z^^_%_Y$*l3+p|dPw?`x+MHZa?UW3ycdLek1xyXA_!F|2gN_`^Axo6QvaI~PAjC5ZK z-)EN+B|S5gDzbp-rkg=(kTqB>w}vOP0?C1t0sP@jYE3xIsESyv~A zd+Z+Jx7MGMck3pR!8?1&pvY99M?_EWY-xHy%#csIjlj{_g{blVHV#-Wi#O~a)5T}z za>Ak8+{%OwI{U5~Q$Mbc&vU|YMB`<$bl@$zFu0y7e@(#hkj1b(e;(*gETlsxJtF=y z?C|G3dCWdvO#ggSK&Nt9{544(N9-Ai^<|^kCdGyPQSUf@Oh!7%-+mFy3?_l=>N_O( z!+0(~W*#S+>&U8@7|eeaC`=ty44c0%g%>-2lDgm|rvG9u>c{)w?0b3i%z+xx72Xcj zCb#jM?KV=X??pCWH-cR{+llGg3zClLUQ9?Y!Dar}K<;^+cmIH zMS&kaAO>S59K$stpX}@}3K~;ogtT9W;P%_qr76e$2o`TU$w{w^B>D8vQl~Yy>C2Op zI^2oB&e_dE+IU0P^HsP5IugRQ9M=YMS8%I75;?8zV zAZ=|Ag)wdo+~-50*SpD)_bBUw5zkIS|E$Aw@wT&M%UFNTJW37X#$3hJ`8haWdj%fa zyM!hm7rmp^$0+wTlBQk{5+-=YnXj?BXo7m=L;ys1%y z4I`)UN3U!Yq_D9 z9S64JaC=Ww4;myy6yAc9Bg;uaj4SV_Y{mb6ln-aG1Pis|dpARE3QBo;K+QEL*p)k+ zTQ}mo;1nb>HbL zy2ffB_ol1~J+|wz=(wwRaYs3S(KLs@TV27=o}Gk)dJCzn@eSzKzJe(>VYpH62hBM@ zQBbd_DfO*f$2}c(i&TVKz`C2CNSp2$b};`lF@1K8=8B!`9Y@>AnC?PyG9`#fG%Q)U zT{viq=avp1JweN%jXrnF!0jgs=-ucMxGyvsy@F5U#_P+;n*o{hhxQ!yezqI#5suRk z8zlGFk7dy>7YJu>Bw_T^cj%opux!il1Qt8;0yzE3glmyq@cDKI__=g&*N1&48SkYf z^QR#BwBstNkP5_mIx=j>4nNu&-$RcMl43;`mCPaUI=ibWpo-`ul4K5$1of(88gpl+ z#U9LWyawCwEgmNxK7+T#Jh9X|4y|$qk~r5;JU${3pWn&FzmZYomx?;JR7I1?BcGwU zx(ze%7|W^}|DpEkK&#@6!D=}bLAU%9csZ8_PTUIgaDNC5^i{&?5*n{V;Y|)!aBERjZ zusXzl2@7?xF4@|95BDq&V%jN5 z_;8#zUd1R}nvqGhf2|h|EKDLxM|{Kdsfed9cVLOhP*#_@nQ5L=XE&NwQke#Q&*WPX_a&;M5JLS`ui9-64+{ycfZ;9BK})kgG(cH>5gHGsDl0+qAod0G9RkI z_hnf^2a+ z*G+@lt8b#=i4p8TQ6!E`dnz=$E3j|d=dj}Swv60{ZGZZ)E62VN zy+x5EW$6Vxt8p4@KMWFm6+bz}Ykt_ZY8n3hJe?T+PD1YniaCn{=|%fU$s>M}P*7Da zsA?P&F1cudUg~V3eM#Ia+C*U9h3~khVk|xCzE^VKM`CHhCIw0|WielSA0CliPye<| zLz}7nSiy#u7;joiugfN5XTC3&@*sn3&dkSFY3Y=v}FtwZ2LWlh4!ubuGOY|>#U|Do1-0Z)C>Pel%_j*d4cne9QYn|xOVhybF zOTw9Inm9yz7jZY%U`gVAY0U2x=vpqry!FPhRc8g_IoF$xee{|8d?JEdoo0uIK_ctB zUll0-8U@kTm%-ra1bop{i0@CW!@OHx1=sIKsY%KiVs&N{6fS$fDdh%Z{#sQw`9U+e zKyT1Jf(i4Fdw`E-HHuxgi(J;dx?)w^Y?}Mw2&vdVivRAhgSSx{!Hw3e7F3%XC56|Y zVclzQTs5XeNDK`UG&f(TXR=G^8^_O*7vg6Bs%jAB9=4G?d+P-Angc@Z*~7Gtu0xwx zk$!bjm96Xgh7nejg}mv<>((EGhp)t}Bd1hI6U2R~q+I0IZY0qGRi&v53#jX^5762g zL+4iB<{BE4z@lD_Z&S#GeYr*8l4*&MB%X)Tue`e8M7@RrqQQVzPpkR zZaoEV6L0b3ZjUs7(QRbz+mel?v;E=R?pLs5Lu{$}q$AUo_@2G2f1pHic*9@%LOKYy z)gFOmj8)T`9=aVSJ0I+{UlxYe$lkqqxf};OyRUPhePieVeOJU;w0xs zW4?-c)xijQIQ4YtEF0kJE-r>MdKW-HH305hOdxVc1A(}X=YvZ$`JKZ|_-o$cy}@f$ zlv!RxUZ#x#YrA_y|3)gTb+3S5Teg}Fc9q7^r6X|vyCBhf+(^bo+QH?N78rQ$DBn+O zEdQgQE^nhYktg23UvxYPHJJf?`s{Fi*taQg`o(d$y-yb=9hIe72R7qiTSaylb;;c$ zS@f0XbdVosNzGceW0_F|ZXEtkDA#pCkNhnBy`eubWqCB*w4Kg9rNB>Gyh$j0S4N9X zgyQj;Mc^DFfP?HT{-$mzf5TnI{7&x<2o>1|@gghWX_d$~xpH3O+a)^8banYhTGRPg z*RSzr!>7U6%bNU^=w$wrmjj>b8NeqSrtk_}G^m^8!JfW6xO90Xg>4C#-}VpvP3pz| zi~z^38}o{rs$@4d{UW@ODm@N z1J;55&TsH8CK>wWC2*UZTgZd9`;v<{#qW=iNf6T>NR1z_CI;G`aP)@;sN4;Oxtagq z>7eQ0y3Pg$>nD^d2j;=1Q|BZDeuoP4+&&VoYf-e*%mQ`lub8z>e84XmzLH=5a2cPq z=>uQ>`xo!n`Gfbma#--6X-Wn-jVD+Xk7vZZ=39;$)K8d=>kTu|X5uHhdsG!n-#3Nx z+qIjUZA|Aa4aC`O(BKKbfxH%&C;HW&h92g4Q zPJJOUt9C-#tO3;ic)Dck1HICjbGPHC`G;U#>m|`eo=+1Bre?pv`4Hf1W zl*;d@h5A=IeA@l1@Y`)I#=q&N1J2sQ_w5;|VHb@D7bxSBnTPPB?Foqg{R-w!?1KAD z1@2f+Cas%J@w=A1gdB&R@W!>Cu3c*m^*txSP0XrXmi{H88M zYdDK1v&n_d!*u*IEwW?P5aG%sYm^MH5LRuNMtAM-#A59}a@6mNpxPP2J$d^Z+=_mJ z&YK5d|8+QN*j`3oY_{T#jyXE3q|kTxp{)4<|vxes*7$}N*E*) zCoFhih#o1ABz-O7-s*H5eblvt29jL*-Kef~$>1B*PpX(&C5;zij+Q};dM*hIHXt8b z%6QHyl8+8lHUAhQ%SU=#1rHRTjWpX}$ni+%?)gaUyYisvOFh}-IRI6E9-#l7QiR6O zx`D$gpjeU+AO*$jvS7ATuR0_`*nwy4{qN7O38kVe?2RF<-B!c5)elZuq zKT1;Mvnw{?_+ty1Vt6`>>wg4=T~E+Ja}?XY_CvvVh__%o2r+#O0Do|WeBJ~)Os z2QrwlyL?%EPzPSvZo=9$s&N{L7V4GlaFl3RIU3$Y4xWj_*J~bO_?1g|O}uAjlV|c| zYb5?=Ls;IT!R&WNCVTcPgy%{(KVacPJEfhuwrVV`Jdw#U?Vs zd^=t%J;?u;d5q86Xu-es`o{%deM?^cy$uO429`c97xHrc5!c>jEb#Fvrc@_kmCga| zYFao8xPok~e3_Zb2piUYAccu01$xHcm#V#z$GJ*_;(PShHEt5|Q~@j)q=dtmi{Bc^2LXpC9~%8&dQ6T?*IHPj(9HJ7y@E-j|Od zy;4j+If!bEYelbU6_%!&2ur;d!;-GKEcAU0EYCefPi)}1n@;Q5g~HX;tmhv1Uk)aT z1$yNCrZE^eUmjGIFGBp^R^g1;_da6Vfc+MA<5WMf`EqJ5s^mA&?p+ozai%jYmk-2r zi&T0c=N32osSOVCe<>7j$006e48LP9$5Q8<#WRX)*xLRx%Yxq(F`K43ky$AVZ8o+< z+XE$^Gq;dAYZNfAVJW^4{c~(h1TLS^NqZi2QU8Z|Lc)Gq;bCSH=@}shGUrsV_|^}6 zH|9H%>4C@@B+z*hD`;P#1oN}@qt?h`x@xluCQV#Mr9R4`TEieL825>Ej#7l9p-HsF zZZHgM7e2HUdbb(DYH>@Zwy2>Ivvx}ILNU)4*s_;z{G*pSRG)F z8w*h#x?DaI+ZR%#dV(kXs7$0o{1WMtk7x1A4h|+LTAwbGqzthMo^8K6gdtqyS2@>$T7GDvvUz7_2H*DDp%mjSz;e$WXg_uirXLuEUqjT{ zo&(Q_uSOFW;V5f8(QR`&#*1&r(&k-O?WB9U7C`r&a;nj`alnc#YEBYAuhC5dNd#S zcMk27Gs4(s=g`5fk0@V{fF0+$p(5)aH#=36Q)%2w4oxkj=cd#UhrP>bWAh^7kQ*(A z<`$rpSr21YEacCGiYQm}6!d$hM6cBhf!Lxl-o#)S?=Wi&Sk+qL;TauNJ1vFOd~&3- z>teWwccZ|vaUkD!|16lpDUkjbL$=*`LVEuPT(uL72ezo*yhvx#wK1|(f`Uer1sl7MNS2Y(NoJ3aTL=W{e?K zZs(x-_IFNt0E3Rbhxl^kM{vCIJ>PNWD_`g~k$(_0kk6NDfCQ2KDRayW&eaCOms+0W z^2zWf>J~ZhN`>$Ha1VE?NMrboDDvXMQBuFLh*tka(mb! znCdCR`6UgZ{k(?os#_UcO}PNRMgjbqzCxN+G8C-Qvqg$oU*T8URlbQ@jvG$Tz5GcAZ=S%t zbiK%NV>7UHcL9DevqFv5!`Smf26u@B&7c*=xP9DS5JrIM9(sbR`llvOS2~yKw`V6{kY8pUfl= zMV71j#sn-hpMXpL29nV&4@#1M?4#;OMlkR5b6NFWO&o5UEevZ+K)Gv%IR3yH!y(u4w-|;i4wkX-YDF%A8#ggqmyeq`HIeI{31js z&xG&(i^${4D}=uqg}A-jhPH^(PpkD!c=~0x;5gNb?4AFQtKOzUeCNstne(HG*5#FW z)2l`>>ZuYGynb?<&-oN@Em398sS>E0b=v_*e5UH?8Fi+fM6hxL-@`|e0`mQANi z%ZKu7Zk6HP*&nd4EFFVohmnUNim>pB6ec8z?AfEu*f1f2+i^Ja# zxsq(SDs6`&POL)CRcZgr zD`}lgsxTt2OOkpsnLKVcgy^={BkkM@u; zI%SacSrK{^WzbP{mamPNK&Kh{z}2WD_+d^jzN%1Vl|f(e$z}t#Y;&m~jqMe#TIbVM z=3?gJLON+N3c)^~50E_{e3+HLBAa{G0xJXl;@XHneCVj=En)ZC+2o?lYN$CVvf5@R;!O8x*rj}*MAX5)A?ohuKP zPi=+jSJk)@-PoW%>sXOeJjQ4QuzgC0n9TWTc4+beZe42y-I?fvCM~J>3fqv4mg2&t zgNRJH6E#c@7Er2t_@kqEsgw5$FJw$QOaj>vyhQui`i6fUpTJO zeL-vb{l;@v-DvHUh9Gwcw}r33c-K!t*YUGB#q}-f|Nal-6K~RIYR4t3mlx5j8R3kr zU&sdQ?VyXVOu#Akf8pMXF2h^aQav#f^o+(H{iT3VqjoQ#Zy^k0if z(CiK_eP}%PHs=cC4^O2TJ1+~u!XKP&TD4#-mxz6`x1rLremKnEncf*PA0NvxD!uhf z7_sFOp4pj>@>4kB-v&84ruqbIowHT=WxBU))Y8G!VtHRW_}&Mi`%Z^?ZV1ESBJ=iL z&pq1vSoFI5P2`$oD(KmXVMOV16a-!A<=PClv7ep0*_b)~*zYVK+?<(%T3S4&Rl7oA zKs(vAWevDIjFGn5-r<&x`NGZpQAmH>QiG-?YVdV`F|L|*6pb~ma2r1;vZ$jAuxRWw z+UjS59-8ISHM31HYV8a*!Er2$czgwidJbh-Q%zpF;7C z*cS}VS7pAxOUQsb#-zPYpSsPfz7LYh%g4>`%n;YB63^=#MKr zg1A``D%k6a;#uv@me<%*y&bt!^B94@GP?YNVV~&i8Q17mnRlewWhfZ^T!H6I9dTp2 z0dsLoqDCzhkoQc$^3lO~O7=atZG1_dZjYgJH}J%E#6rouf>@g5>ndzn6Np7(NP2p9 zt;BHqA)4wl4Yh}65U5b#i?J`CPC}`9;4Ewxv%TSE3Vfu%!}2e(lr0YwmOR-EdDXMY zzoN(FaaRZJc3naDeHz1-)lFn4MqiS;oi!vy2Rx|8s<%>`kaeuDhBBLI{FBPo_MuKW zwwht23m3f5CGw@>r4Wk%N~WF z<#$T-Lob3`gBN%CsRBHnU`wMa@~O_Jrw}^sAHFlyVv8LWV1oHmjGdr_w`Sj^$%bEq zq*JRGF{yYXMkeV9|~M}2g)`Q|1%U{L+&{@s5h~BGJ$L_cj{yc~x7Y}DZRGfl6Fh&n zXC;RC*OAd}@8I9Uee{s+H=$5&5=Lvx=ZEgKB@+jqgXGckp)z|BANl$qU+|@fTv)h* z9;uFk6XhH^?omcAToc_A>AykkfanbK(* zrdhb&r~(rPtmbw4TJxVT7LtMnV>)Ne5ur|e*O=|qf|s5x zi4RdCtFvqjHUVX*zTqBKHeX21XTkNmckMaX^b8 z759p~NjWHNRCbVeS8_1(`OHV|i-YcziF^bX1rz<6h>K5;L=CnNVwJ4iObGdzTcQ~ura&bt9+8Vss!UO!oc6UDOjXu9h zYaPOQM|%Bqhwx_oS?>9%>ky%k4leJ5X?<)fopJOGeIqhrhq}~2sgXK=>gfd-YNbHy z`nQm&XC1H)^AtOFVX*MSQt8@{?oe&uC&=|ZOcL9~{YOV09O|6SFNk>ye^)N0XQ!Fb z%mp&|U`zxiWDkUE2NNKQdrX%%PJ$INnsB}`02XnhQTgIfd^x(BW+)W!M^i$0vvJFz zI$4f?+FA@By|mG`S`8*?N5I$uXQDoJ40R`I^yJ(BFqcfCRRx0}{CFDZ_}YW1=}p)g zGM86xUIF*qz4_r{{>(dM7iU~@2m;MoA>!M4?%d)m8qU8bKZiBYna)1k*tFjed-ge` z25NC3{_C;*o#=JAHxkxm&f^W$3~{~O3Nq#U0{Z<<9=NY3jE7+SbkXM=W1UhmQz+lKxDE~T^S2`ttam^zT z9@JrXQv|J|!Y7eqWOr_BY1pDpA^6N#;pg7*(B;1XeXqKpw5}Eu zO&dVHX+Kz4exqL|C}Wk*8$t{kiR;m;z(1srn)-@NK2t^J&EH1j11LSZpc4Fc&*ESC zrGw$|Nyu!mhpe9li~8T=T<#>&pLf3!8^^mK_!;mK zCi{6A-CQ!|c?oHYeIs3-Yegz0*bi zA^LpQzABO@*G+wRcZe+5Shk|C14%31S*E_?CO-eVKuEVT!nxte^s@Ff={${}6r&AM zwlyBMd_M-t2^N&fKBIQ~63Di3+NA4GE_Zgg*x~s&3-M!%edh1pL~F;wKRE>9xt{0Z8cHwc|}KZUtUqgh$m6VBvr9@QS441T{B z!J>Y8wLXSWYq_A3!@S5Jb(pF6}|XoC=cGE3O#o`sgjr{LZZ6S4T7D}T%EsSt9bmyC3- zp}h-~VCD23uqQ~7t$iu!y27i#i3xx!2&Qm^;>4KTgQ5{46}sD&gL1QoQxx z9qwxCkc^$TRWLN)CGpU%a8w6VI^1T7z=8 zr>j_&r$pG6A1Syf7&7(i+u4u->zI+7K3y^eP^EYvWQ31_xMT9z184EFGRG3m>;nfG zbC}f64<2W#Qd;o>t=~Kn^MrOR^r!f99cNbXJ&gDHSp)BN&PwMhSkhT>>OxBq6TTH| zm#;qcjJ;fQl=;_cvv}b>T|De4?6_8fav?^{S(45YR%+8{#}3kw+Ac(9-aX?p(Gyij$F^r4trspSaV3qDtW_<@( z&BR7J=JHsKp{jUm*k~L#;vwGcp3Pe8YgoaL(`>Cbu+Eb^vA>l(3!0ye&YsnnF}$2M zmGVUB(1umnP2g4G03X+;aBCYxZq3U*{Ce}gY{%~wEX?Gvr!j>-cNj_3hL*y+t{cMn zhm<=xZk&)+E-CjdGAcjqFrgd+!g<%D*CF388#N!BqmExBFXx}Zch4P z=KnG&KQxDX`#G13J9Dy3BKL$|e)e zr_sEgUlMQEb)0wTpTny;jl@%tk0F@N=GANTp}yffJZo3xH#d6XrNw6KPQM#iKX?Qy z>y}}f54NGXg&WQtF_bJEk_mV2=RoQ5|Da}$6+db70DQ3aDQ!8vO!)C>5dGW82@?<0 zftrd3ySn2AZhfyJJRGP6@2e-!RPo(lY%`R`JXlD-D@R~Qh#`i>#M3j6?uuUNZYUi; zl5fjez^hmU@lGpwzIci%`oA2_&zn&|R!rRvs_9c* zDoLnjGRRM?fcOKy$V%0x_Dw@WAZ^H7l!}gsKUaq1AcsWyX5)0SKOTkdd3AWAr;=uO z%c08V8Q^?ce0TR6kv&V&iGkZ%QvXC7mY$RaADc7qZ}(>yp!R}(%#eXUOZLOzi3vE5 zkAwGxn;`wsba;8^7^(Pih+MxBPL6z>LHZ5u%N~9X;*P9(Ms_%)khc94wVpl^7OfqC zhg$xDRiQE5-M5`6NF8`3+0_s@%#5G``B-EGZ~_BkIHSI zz$NS+O#Rc9;aROYIqn`mi{N?_=k~mB?suZ!z%D@au@el z+~pT#9m4GFQu|q(+!;JF!LdtH(V#W~_3Fn<-zwYkPf8T|;795-U@;>mzczFKhRk8t zua*dBxEQkUgg3F?odC|I8eq&q`OcGz_`-iYpR#ulduuibeb!ik#S5`BHsCuMqY#YI z`9siWkC!lSyEPdywM9tM6o^-0A-HUr%Lcy9$0KK*Sd(nK(C@-%j4XXg{ZF{V<|pI$ z;Lp|I?bb#*JMKwTKa531*=6Li$N+Q6*CnT%%dyIM6uV{<#>Ri24KYVd`1QZe!o0f_ zO}$sp+8Se~9H9!`+cLR>AODq29Hl6H{B>UVKCFk@^+^FGkpuUiK^C{dA_HzHXz^b) z*YQ0eMa1seUovX%9^zuFK}wB&X65TmlDOpKD<)IDBdn)2IM>R#nxSl=yux`(*hNs`mUyM>U$~W z)NVrZxCIzey^E!6vVyl8&%)FLV&CiCUVF1M%Zb;j95HKi8ct5ng;_yI!MMSZJ~CJ= zJvMnPG_QID2WQs7tF|IZa<4Z8OAbp&_c8n7i(SD-k_VYNx%BOmCG>Vr4A{(0hX)~R z$jXs9QnR90CEC@);K0@yZk;gAED2}HHPPyYtyD&?Klc6{MW;B%2-AKpf{77t1lC&2{TVO_ zTuxZC8EHdUM41e0FWtf8Z*M*<&7D7J5lzPZr$b~9HrXv&vw+-Dk=#ZKELW_$8tH_Asp3vs38UcA0zj$p3RzqH)Jx_JKP^Pso*0PKGyX0Z~! zl*I|D;~I<&OCYqH?j2)@>YNg6^3z~Lr=P(oBgeB9MJI52ps&ahGDQnL9d6XZdF0Nm zx03dm`NE-BGbC|2=JvP#4B`zHuZwr6hGfQ->tMX3Nw`$-AM}&|B<-!cO5Rl-C+U6T z@ZJnxe6XBQ@2VL3G^3D=*eS=(3_OYp@;2kg0YP+KA4*^Ci=YElo$-Z|G5>$g#rAWC z&?e6C)q(_kb~q6#bC9^d83p$iza#P2T7U6&iH-(q}YTRt0S~%F#W|KWUrN5d1R731S;>5vQjQXg|*l zEI>1w?Unh+Cf<`{@pn#O^S%^}H57M2N8G6T5Dg(#=@goU=b_8)8)DA>Fc=IM|F)i* ztTsOfgv=2r(fx(OmVV4aWiFdLRK&~1w^JR51kfrf2TQL(a8P;;{3{~hw1pL>pGZQ# z)B@0-b`1G^8P@4mf|lHCdU^6C>85o9SgPDU8aA?q8+9n0*c{o12agrvqD`||(e0sV zIZq7}w!~mr$9p01vI1wXYftW8jTGK#T_@6)kLV)D=hz|%VB^z%qI}&&T%A%52m5yE{=a3QxuF5Z_3xMQ{>RK;CH_aPftnYG9@oYT#I4*APE zj!)n#bM*MQcb_43;AFB?MTy8QMLZH#NB_lyVnVT{*b}zJ!re6x+@4R=ua3awyXRv< zv?}ZOHUXJ(Hi#d$WO`B-J}~@%nqt<{a9uwz{k{@2Wo*QJN*rtobY_0Z&+(aSitwrY z6ggNlPx7(j0oL9OV9g!FnNH{fRB+G3J#%E4meOBbuW8IIUdge*$XTTFgbKVL-%6Y1 z7Qwk`{fSos}6luBl@SldtVIrjz( zX&S&kf9Q*Me`ttxwPgE+-cs6CzZexSC!w=aIsBe;2R8f~%>P!jLSj+ zR68`un6KvWdsY!Qu&!CQ0E|iyylNQ(gVi({R`{3m*Rq1x}}dtn)SCbrSpW zm&P0d*%i4oL**2;J6J`pWaZIHwa?I0dK(^@EAX{Fr^Q~+LE1?tz!py_I(fVv|+QeDxaBn4t5>T~2Zorx*FIuIbgI({H@W!RnXm>eJII9>3V;2m7W#2;Jz{xBS_b$+*rpJ3msqop# zLtw71A9thvBJucgMp8Ye9$a)?_=|<>cs97N!-;L~4t8$49r{QuPM`gdeuk^wV(x&}iI7d||T+hqq-yYsOmk*2aV#vF{RQWadG)%{<=Q=Lj6wR0Ljg!?3=y z2>*+G1PjLxhfl_$pLm2VvwJlL<=Xu5`sHXkW6Vd=XMc&XrCA@RM3{n)*H4%vcCda& zsG&*zMc7v+L(($5aQod=Nc}X+=gyKXZ^)G||8K;3s#op}GXrMBaF>CQ{L>$|y!b&j z&(!AXOA|=ngPXv*Ri3_k{F3~NP{oI8{w!cuHulwU!aR>Ecx&*H2*W0m($;Xn(&iXO zOjl#~)-7Y2#(E%K@`7F&cLr6yWZBy{Xrfx;P{{M00#h58!fl<6>`99f+j?<l88{YM`fIEPAZn zPn33BQQ1v5Xp2G!YDQY4Pmm{OeGH`Qw;rO$->5P_tL4o9y#+37H6wID5ro;UrgM^k z&GOrU9o0)X`xG~r-Dpcyla_+H$z1C2M^<#ajis+=T_9(UWRcFnY2#JKGR~zZbz^pU;4l2_F-ho;Qy}hXJw2LK1-puk zp=zoOTEku8g5xKWqeC1DqU{~V5i^H{yLPZmA7-$(%lorcD$8K`t=n*+Et$mcn+B>2 zFM|4K6L32oi!HOJQs2m4db~AI+-s*&#V=x}`O|EW?Nb4rx}A7&fC|l9um!D_Hglg( zw$ZW_D;Vu-3sX-opiRy4RB!48dUc`{Kf4XV5xrcQtFs2L7ADUhoaDg|^3I0dV>h70 zq!lc->Ji77Bjj?oH>A0ZrGJ-=E0Z&B;c8EwfMK2|K{sAafQ{~We(O`Xp(873-`8ZN zcLSvw=ls#FvVg5y_!~*1Ipp_^rGFi+5(}HP*s|($nS=2j7}n+uXPPzP?6*uVUBMPs z&#;7-wF<(bhW~_*x0mt{xej<5JVHnqRZSOJ#6oP}1bBZ*g{~TWi0mtFC2t1}hxx9Q z%1n=h$&M0ul(|p1vDzHohdradr+sj|!X>N{ee09gE~2+oFQIktTbN`p9RkvhVB6tf zvhcmgatTsGW9=-y>1iqW$}JN;7``NVLliFOLnYVKY9$Ll$I@JfaA-F^1g&-lAqkUU zi<<%auA|Hf@5-{Qk#~5j)3SWc{$jo{Z3rHVsi%9!O1Q626F{$d5v=ds2~Dw+;ql#< zaIT;N9WD#lb$&fAPP~e$TpY^(wa1cnedcz*4jp{0Fsx`S>zcViSaLfOJH$SLL)aFw z@ymW;?;dk($jij3wZCxsm|NJW8bxHvl*u`fB_$vEn7D07Zj5bAXu!}B@z;(IQh27lGTrK@$&WKb)r=8R@mxozN@ zdI>tr+#un|T*84SKB^u^pE=!zS!5cTrc0@EW(9oCh!kC%wfNwz7JJ{AiN&qIq+Yeb z{MqFiyjXkZOTE=`s>p3qDOp4RY*$0CiKnl>+7m5HYNV27 zo54@t81LrC;qv5GY_d2D?RiObv9>yT26u6Np5)lA+2Mqw=@p$gtAid^?N1Nx9700U zAJNT);u%!7mt4FS!fA$IqFt41AozU;cR*TQ z)`)Mt=A-Y_)xz&MSGs$s*okZ`gezHFL1l6t@HU^tZ25AL?cR;MPYq|WweK-^(LL;0 z^d2AI3IZut1$$fOf!WVNtaAM!T>I}3_IssAw^Yu+2S-lRQL7b5-Lt_svtv9B{vOP> zqc3Yd=uU6;l?uNSqPSV&PTFg^Ci~H5fU(^~`c-KNSafpiN2@Mt-*FNrTaU!_Pa`BU zGnHxS`C$;-r;@nDS%c$&b2M<^f4Jj-BFiz^ju*e*K{mMsU9EelQP@oQ@}hsaRrV9M za6OEBTJ%SuHI?p z?e&{|nyOenuK#FQ=5bWGS~y2ARf)l^Fn{*x&RW>@bU)P;b2;JSSymru#T`F+ zPZGBH2HYB43Z{_@;X=h2q5b|CI5bNG67SrHyKNe5aK~u&!}9<&{5P9iu)e_cS(ZyY zj&30`GlL{CWuA12<+QSQ2Wx}{P8l$=^D6#QHNcgt=hER%qwvele0-M|4#^*ev;67# zOjX*;;>UhtV?&R#{HoccZOk~@6#Nh!rwt-KigrY&og?v|o{;vP4#Gzje@Q>ZQ7lMj z6my$UNFKLLq#2<%(AWI{`ss#Zv8@fe_hBu&velI{F&~S51*%x>=)x*wOqgQSDNLL+ z4KL}AWr=A)Y#=)Xst*#tC@BxSB%!#m{Wbo3m4()SwNUqpIsEgnr+G&A*~p`R*tOG5 z%x29K5@s()z*U1@zY>X)S*eh)b_rA$Sh7DfgPnLeiZz?al6^O2k-aG)$FGcIHn0AW zw^Me)9(P}~)!D~FekC&bZ8p#_Y5<&`Q$f}g)d)*=Po`xtv6#0oM0)LQqfp@7Y9F(8 zF&s6k65oC*aBC@HS3T?4fg^cr{>oIk|IT{smgHjmkp}Yls|?8*HI4KBR|bdnA18OB zsyJEyzHD~CA?*1{DH?ZdARD1C`kQ*$*G!3J*2bmmXoeHBT+o0AFNojL5qs!m=_(xc zX#^havXVyKWbohJ_uMv}eI)g%9i+E8!Uf+<2ow6EUi?_hw;hS=y~onKKmU@Aalz2@ z=PYJSq&U9z1e(7~2lwHBIa3=;v|2Y5;tc&@{mMgvj#3g@)*xACxD4Z(hT!k_0r<>R zbT#Z2Fn8AsvgU4%&=voY?0MN3%_!kvvR&JjFS9DE445g7Qd4(>yw0ovAGb zsz>1tt8Fx9>_GeIK_w)zbqW+755_M8RZ+)a1bg`R4^3%*L$_+WN#?P$Vy3B(n{j^y zRkmFSLTD6TmS*9Lgb4hn5Jevj9|&hQSn*eT0wG2AKjb4n!R7S%aLnrm8GhV@FUwxc zn`@@99P?;4yLTIVfBciMSo#U~*givU(NQ{gco13rObM#W8O*IbOvBC^u|e0hFn9k5 zOg-9%%?u6U`%Nt2RpgR*&jK}Ku&SXfbH0W!DlLVM-hQ4k;}_KDi4#9bI+&k5F^A7I z*)6&GB}(X6lSy_5e=Z&C_L&rp+DhIw_T!HnIYZ=Tx6-!v^ZCsOC8DqO5e|G|!0K)$ zICYm*)22_hfa5%&k(a?aoGJ7b~h{t`z1;I@b+y)A}d+$O+QZybN2o}6B= zL`cnFUy>xBkBR;^%s149ZTJ^}k9>F*G*Fv4SUaH1$;Wte!E#*R7>aK3gYe%B!kH}Wr^I|}IHvr07R}TT zgPR2d;OnEQWa`rQ_*7)cu8Xn9o!e_^%RLPotJ;GZhgETgu?^JgHnD5-ve~44pRwy} z2k}Vvf$ejDlgZV`aDD4kGUb~ozxr4aHSj)4WHv>>*XwUMNq`>NpR@zUcqu~l{0QRZ zZq2FR+(y5w`H9_rE!Z(ahVR-}44)U@mA=gK=I8ex#k+^QbCci2v1yJvlp8c5-?`$W+#n&;W~hF9!B z#(AdpFalgWpVLQm)p$r`+PH)(;{7F8$qR=I@GB?{!>4}7QiFHoRrCYl@6*3z`uhGn zebol(t9L=ggcTq?n+f~&TR`OY%i>I-FIGPfq0aq-p{d_4EE%MTKP|#(bWRNBj`L#C z$69EtsXiRYxQfO#L(AX$n3g{eqik@&DwclK4Nb+bpR)?0SKiBkf24K^E*!F=s*X08 z`n87Ue2}9Ne2HYO=z9I=C5sz09qFD^l~m`z9rQT<9*;d+&P+P}X-}j)nQ>ranZDmH z%+D}j?X~r!phcIW(J*Q|^qw&O0mTL`25!Fyhv(m7V7=UR_}8Ine?PHJ$dX#XliJbn z;bkxNR9!+=B$vPs)e}fBICJNgs>0F#wAo)V^VH^L#h&{1LQs}AUpVt1ywg_(knK-v zO?AOgDIZ5P@5SIRSLmX7`ylg^GVkMH&3hcyU~!eB*p_3H*uI0B(*3i1%K}C>OI8oP zM^l2|(k{z5G+8o@TC~h1h6?#*D@?^5V4wSvx`k^vC*5ozW3)4T|9z4y&PsxJg)!7(MC@x?7JBc1oIW{F)ag#DJ>zk)MK0&-5l92J&w@wQ7O2gi6CQl86^wJI6Msn&7~R+c zx0^PJIl^s%;r*2mvc&^5Lz1~4v%Aq~lqc)iZpq%h8Nr2)pNqlX7jV{;Zfbnc6AJII z!541L5+^LDv1*IiN84SfIO+sef9oL8wwh$S`!v$$v?X2RUWR!c%FOla33O7_r^-D3r)$JQqKh=cSDvrC?7S~YhcE{ zEc{w@imWLa2Gho*(Y@1u3y&wON^J%;l6^5pA;a~TeN4$=E_ZPTH-4@;d#CV*TSMN` zRO40*dT|_0V`ccOjpyNVU>m4?*W#1jse)}l0>04F7cQQWmD--RV0L!qEN{j-e6jNg zOc>~ns|Ocg)j?CZRq_|C`Yhv@{VSlI>*h-4M!qJOw+y5%w?*b*?!U6J%8h7{7AQn- z38yL_LZL(@6V5)*!KJOL{J)rVGP8dGjvDn@G^IZmR(a3G9GeAvinJ9j95_Y_ordAA z(jc@@%A|7^^K59hKYR7OFMBd_4)fm|&9tM^nMY+0G;D4oe%TQis=S<^?b8f3KW%Ya zfdzm3We%jipDsMCNXBz7%V=TNyi%@@J%3GpH6a_aSb}VK*}0na66GOjh zar@#NVz6#3d|vP!Lq7(?-#i0>29J@dcNUZd*R11i`GtwS--Fz(?kUvr!caO_Kat2( z-KJ|7tiY>R(y&BOAdQ3lXs`A+sq4#ST(aMvvH@3B?Y~;=0HfH2__-sRDrdNhjwL%h zD7qm&Ed_r5iGL(|MKJ1IwM3nJ?{Juh3{Jac$4=KRU?Z%2*nqBuuqTv*kc$4`R{o6^ zd*3C^tK{MB8Xg|W)RNxn8L(Gj907xL*#G)IjJ%x3pWBGNlwTf+2(Jy;i}%E+5vNbb~=8-o++p2``5#`J`ejZ?*@ zO8G(95*b9l6l>B~fBoo4sVjZ=K?@wcTcxVEzSFc(=c%blEe1KoF{L|=)Fb7F^ueIV z(gk*wuyyug^gsDT^2WoA8I;TNu6^!smOBVf=I3Hd2_Qy^~ZA_aJLT{{{fukcX(Y0f(Ap2tr%H zdS^vE?q~?tUF1ONaIM6@Yy>w)W}=;b=LkHrDoE;CVn)w*NXgSy7qE_KE1A$U826h- zk)sNy(0un8%;^Zk$=||3L%a+4WBrBsj>dRh^r=m6enlSNTurN_K{&K7kBnF--lrdt z`{eIQCw%(Dop;is&!wU0vi%3L|)dv&?~q(pnz)~EqY0cc^vXHm&Vyr;aaggZC1I8-g?b= zEN3e^({=1z+b?RQe*+%9R%GMvnzH47jr3q)vgDBS1)OrC4EsNKCH|d@>ClbVWYgF) zWP9ZlNzpJR*3%V%tA~c*jlah1+NBuq$oR#r&YFuZzQ3q(Z8UfL+%Oz@CRvb8Xe6Gi z`wG?1b%pQiR+AXF3c7gKReCmX4p=XrC+wGt6`3d(AY!UICajl2|K)ojYEcSph{&YR z7WvSVdo6LXWq;gKK8Ln<+@l$r9O(RpcbMI8CQeE+p`Sc&k%e;@RG*UKxCaxM;&=lF zuSPS^590Z{dJeXmW!uTx&F3710HmBptCO$qy>vpJ4%q z&%V<3Q>P&D`Azb@y^el4f?Utd4F_<|Xhad{ zb{c_9jyk!PKM9t_#gaO+;UGI+g}GgMf-;-Fl+900fHR{B@p!V44lHsIrfHdQMbjT~ z1_Mmkh{L7uX45`$F7+6O-uWiPsJPLFlX^^Et5#UU1_{y356)N;XwZ2|E*K|reSS1@ zQ46XdAX1{Tj z26st6SfaN-i55v`k{bpYWov{;x>tDy1_b1jD#adRSy_Wsx+`d%^E?_g>>YQ>!h}m0 zph2FDmgD~H8cY077m=|>@en;TgQz==Ad4K8VeivKRb$uI@Up#GET^-a$tk2?F4 z*Y}5#9@Agk5A97LH|rtnSrG)oC(HB7eq$h6_9B^}mm>MuV*{VdhQa7AJNm=-h0vfr z5uUr8r}195#GcT3NPBI-E32<33G)<%$Cf#qB+`Y9m~)Q2HY%m|TZ_roIV}*9F@(RH zc1UC%#BwzmO8ny9wRpGAlUg@#1f>B3?Q+fxfoV0_koHECr-vdS?#K&Lx!aZ-dTS@y zRsI&Lw+QgybQ<@^*^i#x-+^+5c~mWPJdNM=h1(ui0lS>_;L+zq!RoEJ|I9dDW_w|p zxYIZ1>q_5p_csx))gnU}ulY~nJ9ko+P(D5Jq&bh50{nnRdx?ENi^Wb)LKmliQDy!;3Vytv?JS$boFHl7uBZxPhC;&m?yAB#wy*AqoLOB3BjZ>aVJ- zS=|#0o39FO7p>?#-_a0v>JHgA^c;B@mc_=>D)#GWsyM66f=CZJ*5*`-=by!)QB6Nw zanF|y^{OEzZ+=Ml8%4DK%s_nK{}ifCxrG58W#oMQU*_RiC(|#|S>n|aTR(8`9j&nI5eRK9>UInhysKGh1H?Znw2z@>- z0@emH@G}dcnZvDEr(rxvEb-tPmurz$I|f6jr&JQIjkNO$W!=`8+2o1~<~)G2FAi~W zB^M;;buOZBpJcMSb3bU2{b0WCHSvhG1OwZv=<4OlVisKzUQE(sTKpv(Hh8+&A-*EH z7nURCBr>!c!Z`IfMfz-21Zs%C>sLI8%I$&FFj1R4TQ`I__fIB?3;p1ZZxHrPHDGc9 zMXb!OjOB{mngezM*{^biS}wFl*WruNOXN31O*Lf)E-G@mDOGfZ^*>s%LydnjatxWN zwufkM&twa#f3Rm#WUu$1qHk68`D+f*eAHH7;rW)8XrHiyo_OF)wJtBkXC5*1kmy_b_nm3brJn9q685?9MbcOOZJCExoYXt~ z0={z!#C65G7=NLL8h0z8fnv4jK-n$4Ti+kF@`h9Qck<{tqk-;wnMAl*ws3rdHvQpf zNEDy{Cbu6AWHud$%6H}Xz1B+X$)}b98LNG-LHB2Y*w)-BX8J=()i1z|q^aQPpCi7*e$h?6_Au+{GgRF*ki9?m z40qZn;+K!QbV|KC{quJ#hAN$7cgNNPM2Z}Pk)oHTa|7BKYGZ=jH+<14`#*}#JDkh! zjpH`iB@{x#C}b3$=iEnAD#>U^q=ZzO+Ed5~nW0i7r9n#KIrl>l$KSD6;-4>z!NUoHaAcyHV4Nu>)T(S}dzVTpPUjRKFk*~&*_2drZ=#AS_bAMW}{DBD9X4! z0M~pSqOOo8$y5_Re@qr52XHv(nG(A7cMzKS8o_AeA?POwfmh3qW$n@=_k2!@;WA((${ z5Pomx@#FF}g8O_kp+)(WpwJg3tV9oC&xB*Fx6l&*YjzW&{oRDb?A@p(7YUV{e~}H& zjqLq{t7Tt9RtoXw+Hq@;vA921!}W@@gsj&~1>Frdg$)yu*_6TM7~-#v|3&pDd!rLr z-1t;d(X$jv&7|4+o0K^O)#IRf#i*)#4L^Z3j!$`tTP_|ZhuT(Pzd9-0pdpLrkEXG3 zuPM0ycNjZUtqWBOI;=AKHGin)9?#tw0wcE|E0a;@hwaE_8$RqsHe;-y79}qzE$|a& z-<}~v94Z%12B!)Nwjx8QZ6t~H@L{UU3mC5S2lRD;o1N#$+?p1+2g=a*;tJIa0UxLdu(ckQZ|?#FS!rQS&VIZ(`3pJP zbeFt4r%WD<{8l!V;b@KzXEy$FUKD^J@gRHc70(>74z^roi!iI({w5tebdHVY> zrN~(5rZzO>^e<38NFisFETh8?aOJ0;6798)82@z@R&o*?CH`%v*Ix(ia}jW6eXL+| zvXwa&MSzU7oUpL^6?STgw{=1wvrb)%gA`U^c799)4hcb|1~1BQ1KZN0-S}Xj1=&55c_cJj7jhB4Z|1VVA4OSw3_Mo?g`}Z?~@! zN){Ch>6Hwk)6mhsl3duNOm8U^ z;qrysNYIlzWV@>=Jem5BO#SaFC{2{6{}ohm|7!1ni(?BscL^r@)sdTOe4V-S>)Cs^ zkL*$5Ne;!|zw+H-{7I_u@5>9=FXt~dsXfD?i|m-r%6n{P&KX=iKMtfWI^g?^15Ebu zRIK4L&@B4GK;rk5dx3OS9hK7FQ?O^2vnW zURp$*bd_ncfdk!QDn&=0xIw@*Yl=oEO;vFPX8CA+#GRL5h&ey}9;|YqqN;AEp^ntsl3*CRb#Wh#hI& zF5yJ`a3_;&5__?0wdqzlEjplLBTG?C;A-o}(AvSS6yNWmhebbRu7e3VSvVV)6^L5K z?oLiKA&(of@FPDqJ&2{tiX0@5@ck!KU^NG3@^Weqatf zIW0%)zUu=EjzpQMVN4}7hvoP-lO;Ou%Wmu_;EkUe!kBr{Sa`aFdort=$o`uM z?@XNFxc?~d+O0-|Y~PNp$+=dxtWIRX1t-!ni*)**rI-T`O{Ul8mQd9N?KH7#9hgoZ zK?gjG5`Byh$l^cUXn17^HYv)YtU?9pS~Y|%Re1s`w>Hs%L*r?gv>uI;k)ruA-(W(N z6}!&O#ihMPkX_D0XW9s%a?T*!5l}0UOg4ww56L+2vNGMEUP6*Q2T<562MtDrpk4S6 zB zQJu~nF74sVn`A(4^d88mj%CfNx`K5@3_4X=Q|GsVwCA}r?eNeP*?=naR%#+YW?T_l za%>Q}alMJ$jSCVQq&5kD4GRVSdlR{FuaXVyZsu<-IY`4)_R~+*2dG2kVf<`gA@1*u zh?I^wj6ahJFBME+tkDJd*4PF?=10hx2q~CsbXRhY+b?#D{DuI(Fj~0Q7Qc_yL76dj zrbiFhbVPFN}6M8Z&hA1$t*LFZ<7!1i?q;fIqfUF{J; z><+g|MvP8IW9P?cYM;WImu2zcT^6{rzpU8F_Kow7U(Bs=^k7S#TGK?v4zF%%w6@`i4!5jbF-oP*Nmb?E=p4+@jdvz)M0tXr}a`R9x-?Y>AGx2DjLt$M(= ziM#ZoN1VOsHvU1)ck+J3Ej0S~l1!?~;O5;mP&EWF>MgGVMMupU|Y zQUkMKzSF6m-AY-;A2fRFpf|sj+Zt&sGL01Iqg%Z++h;A@>1Zd?&(^ZL-!-^{do$S6 zs?#hkZaj&5VnK!mC_!q*MwBeO%0Bg#le^CsvdiMGG!bEfjFfmlpE3x~pkP~l%FzMn68WiDEg5%+Jy+p<^Cbv6rW z#Ui2~HJc2%l>|Qb6X1J4DH^x4o*7kELB_Rss8YNunWU1z9d~r)b4|;+{{1`4ioVT* zx+9*bdA&-U-%W%eKWEaMKvkT0c8KtK)kPe$cMDTfdBApWS3{?#1*G3Cea_Em6x*?O zjU;KO=wmun%0(W}A@_RKP-(C}n-)F|3Vm*qPJ0`0f7-|T)`&X(kJIG0mpp43wcnxc zjp%3Iu>+6&N@Z_z>QG{Ron&z_u(ng0Zh4F%bzH28`wWyD8hQ550*Uy1DWf)E&| zki*?c^5()P4PagSjNq^7aaiUVCNvEdyBD&Y;7ha$eHOF=G@OrP`4=TbPEH@+c@&$f=>E$jO@ z9rFTa+TO{&4cUn4&g1ciyEc4wH7CbA4ESMh(qZ1FXDm`EYgXHN(Nxf5=P^krOQTD&-%a%r*}HMY!&b{2Ol% z;pk#B;p0{vJp3c7Y_WO*QL)|$!Kxnk`+Piiw8jmaaJ67|Wri@UbScX3O=eex99EU$ zBboXkg^U|>q-<%#Uvg_&8(kHv;`lJg#POht27Tf_kFM|ANiPjahO>M!`+9qvgnOlp z^+k4+{7a@&0&-!lYXvvC){KOvhhpyqVqP)4AiYrz;8WWQUX1{Q$$BNYO4JKG-N)H$E9F_aTt!f+w(gqC1Az5thuR;7+%fQKXp%6`sz$SbQaXP%-D z_19wxQvEI1pW^-OzWW^fJb5$ty?qDvcXh~}4vH&Y-r9Xzu4jm=YJ0FGu%Nxd>OeezYfl? z8_a)l-XwZ^e?j!e)$lLq4gbp_hqv-8W#qqbmicotrp`PHqc@0o`62_Z?0XzWFMl!C zSLC!U7<3)_*QP_iKiS-zMQ7n)Rf@6% zzh`L+S2z`HU-FepSR99eX{+Jzg^O&DTsO;i93|+a#z4GF0_+S4C2mHQnB#Vf)%?*x z)ydOQe@Ov1cV!r>oYummzOTcFvlE5LB;~Tst0T)oH|UamifP!lB?~*Rm@~u8jj&?Y zPpEnI4=fljoPC%iEL7AlPuTef_w*ad>h|X0-}@9Nyz=C;Y!$FG=K*#e-G=#-E^=xq zNi3uDBq{GQ#N<^aLeI)ZA!@=6VR79sp=o9sUY+QTC%p36w1_G`MyZX&OuxzQmmh$A zdOGmd+#08+0FFvY;WoY=PBL%1VQl0LG#Pvh&wGEujZamD*OT&K*V?ntKR1e8d!EK^ zfAj~dO)StjeKudKG>}Unb1+En9_bw4$1>(G#9y1ofUfA{P#GvKn8}6lY|f(skYqd~K!VBuGYWYQUQ*Z79_`DUXWR%Zz3K9&i$FKy-f-_XLXv$X^rJ{p~S^spg*OIhBlqvYvaf1LN) zU%2ldB@C!^6C!`8m;X#R6_%g%6SN#Wg^=IhN$@0RPRZmT|0vF!JpWHkXy0{4sJrlo zYcy9zuOR~jwN?!w#oSWJT((cxY~je>j))+iMrg6i!%FzPB#M{*XydS1S*$L|m+1!| zW-(8+nAs(HA?s;AyY^Ph5$Z)Sqjz?!>U)$B+ZQWbA8#Qj>WiJ4Lo%>D(_C1lrpZn1 z?7`i|`ohaheJ*h2I?gle9hWtr6%{WXW}}D6$q9{&f>=Gd7?F239a{%r8!U@K*le&VR)2{T#A4P+pjDzJrCXie|k>M?w9a zXd=}*mJGFCPB(p;KyT_!qbO4aIkuG`C|gN_KKx`l%Q`sMh8PGI4r7O#5h+(xhiLx{ zwnZ@q?nkK8LDdVOAI*(ge^bFw7WiX^R@}B`ELN^ohXOlFD>9?mm{oMKALSisDc6a zbIG`DJs2@5kxD5nre;T->9-9N#BPgVJaMuRRy|3D^K>_C-Z7h#)~e?I%0!UnpA8N( z8}o=-=Qhsngc>~(+KD5o9N3{>|B;O+WPz-VyqsnI8N$yd@%^42W(oUiSjsSQmmkc- zn%6~8rRzhD&VvXMLtzYiAl1&@Dx$T?m=p*HD=!rN66?!=^-h3?_J@kPO+U^QH zz865MUp%Z8nJ+zcOQAu@f;Kx2Mx9fOFnHBn-ulXBXjZ=gIUf|M?PU{kN$Ld*ZCU|A z^)8^9=pwmn0x(?aDT`Ut#{R2#iuUU|`2(?KFkj~$1SV`2Ub>|UqxSd+W)qZ!Iig1H zeKd-+l>H@6X2vk`!jY2jP8+IOZAW9}3~|a?3zV<=vXwtWVtIidlHAw4}1n2pAfV*oQRiJ4q};Fb1~rM6n1)@9J$|gg5Io9a~u(E z>sXZX7rXiuh*`oyc4zo~cIdJrYUVTUQKjgU%L%0hlb6u!Va7uGu%l>co{j|(r5GzL z5}x)tf%Z2QYVx)}U1>Rl(t(fhvg||jE%2nES;aca<G1Q`Q*a6{fG=w_ zp|Irz>~I3wu*`~SDQi=gq|?k&-j2HMXeSP3H;G1rwh*X2fI1l+pj9kXGG8owF|TvP zAi3*h-G*W0PG2$^;7@RtvKu7oTq2Ra|3IlKhFm^k&5!QTp&!M}>$t!^-17brm=4@W zUAIKj*B3HK&6jKfdGo|w>~Yw~^y!&%Bj~GJM}_%S*9F7(B{;L%3FAkPVL5T@$TQ6X zNGr&p>g#^f^TRdp^R(HltveC+UTorb-44L&{pV2dSb`y9SMtaB3+&Z$TZsD73T2mc zsOy3)bjYuDbm_cX)OU()*`y2=R4Y^f+rBQSN#6x+ZOM4*_5zwxWZ~#Gd4%Joo6&+z zo|~Yar7!%OKbtLUB@|!2Aje(>z@Z6=L?$sCE`N@JDI4smVMr29SXaoK?idb@ziatY zmm-!b-YjvLULh2zK<3njWku9&;-aevvB3 z~^0z5qLL*mj{^49Ax@7<9`E}s2IYn*f)(Z$5^UN=yAKRf!= zqXsUV%jTxcJ5@HQFpPWOo51csoQ{Ix~E8*-@crx784J|Di*TVK?Cc~5{U%0(b7VAHaV~4yF zcvIQ8+_w5&cIuNWsy@+x({EJaZ|zrBQgMX)kYdg{uN?%-6cZ{p!-me%oI@gUQi3_sd|P5SCf!b6Xev!#=vd>RKX9)!;>Ytsv{kD+CA1U-Mvo~nGX zq%Q|MP-h)w`h5FsII~p?uFX0GTeH$(+W>XUu2*MimF8?$a1OQESwyGrYNy9cb?N0^ zIT+hyNvy0*ONS=K!;g=DAw834^Jnys>nrrQp?1ntlRrZ5m@7G^KeZ=1f_q|B|&eako{2ayo&J#bY?!#!(OJ!>9+5-?-Nj(rsS!v3XcVE#^5epA(a{?fy(F!@j)ccGz&LannJrDS&R43!q_snsn*=F-$QvoyZSNMbf&(AFIszu|Z)4e1)$plerLq+IezVR~Ch&>?#KKi$jag*V%OuP@_6P8eRse zvsZoU+>BRaIQqOD{r-qq`N$Y-J8+s63@m1~GbCh&wK6%11BlrMu`fMm6w9mFhp$OA zhQdzFH`vXed{fN#z8cF4jZ3&y*PA7Ns&S}|a>9o`P2BqUIN5LC&Yu~3jm>fp*m5yD z8a%a%x%c&OrD5Z6tX}~m2N#s4?MfrQ-($(>icDds;Vj`l{Scw`RXg$XIth1vCxOe# z$5^F*7mwGE$JnrjvI9Pe%z5?zR*@{S5lkwf_LmW~9ZJFo8IkRMR0&<*nW37xKg7H_ zN%mbi$UWFD%hd0VX9JBp$=9muq-0bAOuKfSAGhFb*`#GMET`-TH%MKE3-S=z%(KH; z_qkdwDDw=Ya>Ll$xQh&r&4jGerjWVTp9xGuke-mpJZCDP%uHFly-F3=u@z|Q z@&T(lR-jtJB&aXSW_nL_SxVbq)|YpKWt;qPsJt!@Wa>Bk`7|9LEjWr(#l5VVI1eVY zDxme3^)S!Biw*11MY}lyxAjM;LwsXfnO*x6jIO>&)J`bE&!RmT8nhoNg&Byk&~Am&;nGd_5f1&2sT)0bRQ zlCqMR&Z*@Vsa1e^hCBQ*8Ask*@4^)8LtI#$6*=G_BlInNi630YV#9{-$eo2! zlzD40^IjOH7@omt?i=8Iv=(UJR%bg^IbrG7M10j$&!Xzq!CWsH_!^=E9seDGW%Hhc zl~EW>PFCa>wf^I$i2Q}Qo8xdnO)wlx7>1q;_cObwlWgSK9^U##Eh|V|jYD*n2y5r< zl>}C3!C0}tU&nf#aIw@wSoLX`&@tg4`*dC#Dyme7Yp)i+Y33|`jdVR5EwZOg&l+H9 zUO)b<%w@JB+E4V(WTN{>Dnz<`;hZiUW* z&E!yF?$@QlkfvgM@pl=Xt=ogII{(7zhy*LSDU$p#A{_=^RCJ`1M_)CuS^V4poyw8yk47ohMk{(mi~^l}cXLWIjAT_lLFG z4;EyP+Tq7^eHQ*n4xfAK!kKT&*}17f%&mB_aDSnT5b&OoevMi%O%l&?)=c5IUemx2 zjYh(tUH#ZiO^*27T*=%Q*u%-mAtdW;3%QwgQ`Aq7;^fz+LXhTqs5qd~Rzh@YIwO~6;E#l#=p!>24th(Ii78iE z%Ig~v3u=$oWRDPG=uS+UP>r_AoX{-ez}xAj;7HlQ_${#%H+)^d_hA<6%&+Im0^-?> z+7t-iXDX1x`NEBxV}#)^S~y+DT=H<}2I!RXhwSVglCZ^sYj1I8iPv8csW0KAPp*?4 z{uxCc{rXE{R3kB2UJ5f#tC5o9QQQ>IUUt;`C~MmrLSV-gJm+(O{qY(H^7UgtdA})!97c{N5AK}B*!K=XOryMDeZm{1 zy@!KGX%rrnl|j7$^~~dRHB+^+!2;PFk~e23QJ!EAo;9vab#*TM-L#ZXXwzV4_dH{} zCcJ0zJ1F=q|A>3^`mv<2FWk8C9RH!(68_W}!~7I4c$+1Uz87-H%EIHq*|vsstd5Vxyih2U|&OJB4afNymbEO>MZ zf=Il>RqZ#lm48n~#3|yLQY+(E*o@Z7uWXWG09du z%vX)EMDxuK?8EOh!i4>H)YyAAeJa(!rg=^w1FUwEm}z2j@Ui*e9Qd3}n~;J5<$EE0 za0Q%cDP_e|GT4RK2KHq^FbFc*n5N@{ok_*)B1Ptbib3xp?4*=CzUWNYzz9#9xCj++QN4zS+n_8?Pc`5G%UL3 z%*({ACG`j6z)1B!FgPK02l+OF@0)n08t6|3+DpUgW^v}16r)MZ68`vy>)>9|D^Xp4 zjPcF4Sk(7JM5}lYsMjddpo70~yNL;%lvYcs^q278i)Nx3H$W)9{G2sCpNK!%0Kxiv zfT*|rCc__G!Pi=s*_V$G*u_t+ypFxQFx4{w_gxw&1gx6_{*zZjZ_N^rZazw8XxC!M zArD;eA7$#RACj%E3t6JwEYVk70Br@9bn^mx;lmVpa&_V>Uc)+z4@!Cn^_}jt(sDLk z?qmgF1Dna%73Cz`Xfqr3=n#97TEogT&a*r|g{dQF%E$_~f z`I2Ie@16ixin>WjrY&*kwg!u!{ZOIwoRq%uFWY5T!Z*HIjB|&Xqw=DsEIfY~tUAA) zD6iisPCDDlvJ5wpA4Y@3_vRElu`V1&p!lmorTfvo#e zBR?(T6=t2gim}yc%wo-JPB}yyS2_$}QV_-4^xNrhx5gFflhom5`xNhf`$zyrQzn#3D`X9!U=1%DW}Rqd1HlALvA)7S?f~ zzXK$zw!*6)xzNIf!ptaLv?^MSPi9A;{Z3Qx{V)XrBNIXWLIblISI%@_DH7*5-elK} zP!eacfT@{%#~^%;_65U)b)Nq4{p1K>vyX#pWtGHhPc*0U%!nMWSim0qdBkrBUqzaG z17OLG#jHE|7+d|Qof!2yPiiLwk&0Fs##=O&t@P{ST&HV7LE#V(@D>c-D@P;yp2PM( z18GFB3h1=|S9X8q1#Y9bzj4=0WcQV`nX}_dviIy2Qq%4YLyu1a&rCTkwU4GsoUD)k5f*ZI#m{@krz`x6s`6)ZK zp*O<-^!w`A#joXLT~-o*e~8$Ty*?4R5*v8f{hZx$ZItAPo@LFa)r9B_9lUVp6yDoi z&0gGEhnH@~z>9ga%L=w%!u-@#*f4*9!_@d>-h5vVQ}q3Ue?y+&-Zcl=#?(~M`OttO zcplEj#&a&)#xSD{RaRxRl33JL!=4|H;p^_bOdPUSOXM2krs}4nB#wio{65kD_ zKey9IgNBMc^93}iz=*zbo(5BIN<;QDQ&2em0%gtKv52kH!1D7bkXM?4i}U7UpuPB} zTV2ae@BN4k71F|Ej~=)`Itn%!RFF=6(N{I#C)&$wqS~bsXBhC-v9ME^ zGQkD&R%vo7t4p{n7b#)bQWZ$=xJqiys$*Tx7qk~=ggd@97&~wuK8Tfv?dy*)hpnfW z{;)I$VTKJ#CyyeZi-!rt4}X>J^*83^<(t6x+yy#iwhPVdjE26;U3{O^Z@$>`I1KGN z4r3K^@b-XC@eY~8q0C5VJ+uugOds-&np?2phMvSBybR~v=*Cqn4yP~+LDqSb(9dIr zFww(^e33N}Zd5!X%icc3x%%FC;K~suwf6@uIrWn@m5W{QK2IUQYCn}S9Zt31KSZ~9 zHKA;~oiNfy9%5!a0MDWlaNaJ*9!!m6hb#`_X)#MK8JPfb6+Q5D*G4kfCs*vU%>ZxT zar~a)3>2SxflA_QRF<0}JTiWW$F(AG*+?nj*V+X*O&92Y7cZfut&{NTuOSq~__4Ns zGtldAFdk`4#dxRL;KbLapQg*f+F>=V7O8sg@10dfN#z4^yUdHSX(Nbd2vr@)PKo!#&>~)iaWi- zJ0Mp6Dk1xm+2HhGG%51Jpu=X|^}ve~BQGy-&t6GoUyP?u9Nv;^(h=a_x{f*S?jQ+X zw@Bmi7Eb+K0=LlTBHu8xg^#7`OwG`bp`9m~i|_ol5MvzoFCTNfYl+d1!&v(21X=VT zkHb~Pk}1`>Fn$IhghO4d;PA0bw=H1MsSu4$i&DLqC&N2A)>YT*>hBXZ{ z*iNTUw`49BMcl#e^T=x_lB+G7!K>H@zMVJ4QA0U6Z+R5n40Xdb?+t{-*Eu}$wvfqb ziTdf`{`5)xVf63e*;MXvW+F0!57p#56C3Yu3 zjZK(d!d7K>vQLGJ>F~>8G(OIPKHeAtBlR}Xx0k|c@#TM9cBh8mGR9Tpy~VP=gDu?a z4OytS;WX~oG7}cc-WMcw@Nz0%PGEMgeG?)*eW>+WC?S)fN z*sqA|Hg07nk7aXyOPWC=`6ir8Z6~(VUUGM~Ci6p-YPsNV=JaRbSQ z6Vj{K!=C3)IBSDT_C`O}h}p4QQuV9YY3fblMA12{9sGyV0L}(N!s(loi?<6gN8>EzOkc?9SC1cW=B6@pWO$O zk-IYaBi0w#Gj%cl+AYl%#r=Y*ia$V6L5n`$CcYESt|BN2XUWgA$+?NC8E{|t|u}L3eTWiw=yi<%VSB%HDP7-UASo#4|nI^BA=Jr3TN*Hqh>)aajGCB zW%gK7xiy|$blk_K&U9d9^HaF|i}r#>inK7tQV@Dl&cQZ53w#SMvB4U**!^E-Y`~HG zyg%z=n-qF5|H48cb-M+wCL@I5H~+(yR2^@qY3;-I6*p7 zOZb%O$8giJqxgP;2Ke|Xf@45;qqek{;L4a-fqv}7D!>W(8ewJV9EX%XKysDZ!k z-omzrtP><%#X`>WEWt7~i=-Hpv$m8IY>SivovEixzq|~CZ{DMEZ&xB|eo(|IM)tz9 zxIOTg>5`(8r?K1P0o%NzmhIH`W{SCE@y(nRV*D~*Xl}nPLf_xd3e4xHl(ix=`` zP3mmRybJ87-d((W&j4{sk0Mr_u;+EjYRQ}FKPAH=f1l;!ws7x z@bkGtl1`}NBHbH!am5ZcpgMw>y^+WE%@fe$`D!5`xxTdf*+h{EZ$=j6ekBXXPsEZ0y;0B48G3NrGpcveOe zQ$Afqmyt(t)aOX-HXVb$-Vb0;WEUvjN#^IST*U6%oFTUx1t-`Rre=eaiWo{sw3BDZ%IL{hyd7AE2P2W`DP?7g)|0{oUrE8{0a$Rr7sspZ zVY8nF;p#`0yW4c z)9FOPcojh9DNyJK)OU=?$HP-t@v9P_zZix`u7#mqPasZDnvQM5bm*B&Z(+r`Yv9m6 z8Dt7$*nu~Fr6mVF*-HC3jw{yXimrL!ccsHU33sBZJ5ig!vB6QeK5q9*rvO#xF-~^530_l2C+QWkEm23hptNET43~sLw z7EW9)B&}@_&yK&$c+YpdDe6%5>aJvd^mM!*5W_1(jpN6?GeO<){n33`Ba{9dh_lV} zG3@vtVPks=?s-|pT**ow z&PSKru1%H{(KubaOg#yky z7KcIA&-l60ZFq4skl`~9GS$24LgsXLW=&snZ)!dfOOb!%daT&K(k`*oDKJ1dH+Pxv z>vf<|q>w}G_M~&}$?N#6KY_yXx&=b-BP~I$zZW^@>kPwtWI_F$CCWc8z_0^HSd}CN z&w1prkEhHqd;T)iY`us-rWyzzptj7c_#D$2r;1fd#S$xV{*rg}#38TS(XinRx;iKc z&OsM&^A}hA99JaN117WzFt1$OM8A>1DOFB0Lq=>oT@_`1}(r6&K zR;LRW#E!H}s?IR<|98>LJUsL42is^Iz{XCq5}KB#2*a*T6#jcw$)8pq!WNk+k!1&h zg`im%ka$X?-Gm#w8rj7s7$&po@dKg6_CWV3Id@-Mc(?yHvLiEaD))u!@k!wao{D5z!Kt|4;5)YND8WxHBg@~{ z7?v04k14-YMewQ4ceZBcKcaO`1AdJh2Nh!n(V3T~bK3&{u#-i0cQaCJfgl{R?~WXw%)dApX5 z>vx0NJ!}E1U5&7_ARgKVdBX>*7^<2k3y1qvvkec|in^k)@YC3sODRl(NhuA`@Ut7E zS3hAvxvjzz=l!Jjgcgn1mIqtae#5>8+H_LlRdVF)XsYcIPt)e4(UU7zQ?(I(398$$A)(poAtFigf&`Jv8TY5-s#PP3KFk5awA} z3XcyxM*YR(sG0kEsy=ik?cF*E9ZNKX$ulyAJBcf>LhcId^XX+z9zJJ1wn?}?>py(F zc&H$^(^0rw{a*N$qrp$hr=(ZzEYxIb3elr3vdk5GSVG)b$mi?u$<6*);dhM0Ut2Ej zi=9B}y%D)#*N0<0OEBu&WqdIuNYJsFC9HRN&8>MYO|`4!gwX@baE@~o)CFw?`@(#f z5+Wyjk5m`5?=KU)LLVS)(-Z#6_>io_%UIC1Hg;;uIifg60ySDX;9<9s4w@ZJdsU)n z)0=eiwkr{)9)Al1mi#~S7>EZpTL@m>9@OCAU^>TT2Hw5+obB2=j@8`{<6HS4zOV2l zA92%zOXNN2KASiiYpY3~-haaP_x%lzA2z`_?-D_!_p%^UdsI+O380*hHH87IXnx=! zNyEM6lG87;MEz|7%9(15|Jz_(Ia5t|8~zy$YWfQc<@!-fcO-6_uGrdm7qliEh1)%L z>|~4-tNVGL+1MPwy)q(Ka!D`F`=%^-r+br=ZZqMH)Lvlg&$G>0i$TtC2F+Ej;CAo< zr101?;#F>spN7q5ZV6-9p)s=s#SB~Q7%Fxu-_K_$BhT`Q#;I`FqYXB@ji>WJt5bvd zVK8jZCTgtsnO2*Le%@kx5(}R2%=u^8;pKWP`dk@!?MM>mA~}dv%!8wSxhE+jIw*l{ytvetzaw9hD`w19jNr+pWA+zcBu;%O;W* z6%XrM0{Qsm1>C(4Q^}e~E>QJ!1Xdk-E0OxXf(&_C$X4%lA(|uV=&Rw1F!U7%C&y~A z;M2zB%Dml@xRbst)^QYu7yjU?Zz~ar>x0CO#1*Q-hjm40y68D-&vb;z_CMG;2*&8DfmHoP9h7WP zqC?wb@cPaGymI6rF&9B5zbnQvdy9AESXmad-rfRx93sT;XEj;AaVxdj7DnTI22%t7 zWAv2o8>qi5OJ{!B!*VPt$d@8THt=yW(SI|Ad>1{){B=z{yJ$Fb_&O4t+{HOmc`{iN z8iP`mM(k(UJ~HU%N1~(^2x@=aplMbVmyt1@-OnpwHy$hlzv~xZ&b3cm;CyxTS9W5O zAF^Q5S;MxM$_X_+P2fJ*m0i$Sk6(Q)@!Q-kn3aSc2LzWg0rMU#TC z;@vd3zTcU+$KK`+StM)QkHK$2(;m=%by7(GLA1r}u3m1|FX$G{)Aso#D zIZ>#ucfuSU zQ%U&g5@xhuk7ROoHaOLWL)+{d@cYUsT=O#nx2?>^W--TqxGM*&0-f;H(^?$kAqN|j zJh=#WT^2NV2!uAeahb|Tm`T?xw)jpDQyOFmDf`F3fs3QS-s>I49G)pWnX4{Leti)3 zt@hz;*hVJ5znZxn8pgAy`X3}J*~IkMc%tt5i($(| zPPL{SuPpV%Ti*TQh>I;R*eEcg9Uf%o;;*>G?GSLO4P?di1GwhD>P#lGA8X}`#| zKZC(7V=PJLLnO8amwA_SBVmpHd`zx>2?LTM$e^%O#5`mvYme!Nx`GuBUVH&h%gjf! z(Q70N^Rq|?b>P>Y$!1L-@9~zNOEL0pBXne6aMzO?sn((K89EsIS$B#1EIC`BrwhR#OZncx# z#^S5kYjYQWPAz~CEe(vwxz5U!x*_*e1!)~*#t~;*x@6xJNceY@^!ggZ`>S)Ir@x%& z)v9A;L>j00N)sgiZ zhz4^>rv0p~3>BJ`CXrGZ3Jn@)QhxjU$NB4=>s;44&vorT_FsEl&$IVl_qy-*`}KAx z;MFu|uyzX_(F~o@w6WTXGtK@-gRRzK)NgS9D;@<{sJ`y>5z!B4nrq!n!Q ze@-*5*t4ofTUfcf74JfSl%20iBi@WIbC$~zCEAzf`CaX?GV0A91m-bj8>)-7GKPN(*vyw7*@$boH0{B2 zA~m6zfBWM$wU4(0lZ0`?AO6MU(=TPZ)zFQ)n?0q$kL1|!btf>xF_Y(ye<&TMvx>^S z5c8jIMdPZ{656oxE&Mk;629E{2lv&^3)AKRpHOTe_9jljT@K%P4XbQEUPDHtTHMd? z(9fi=_mt8N%_r$gg?nX3x2Ds$vs1Y%S7SMwi%-ct8;*J;GuqqsLNJa=vD57*>Ep9+ zSd`K;miRV;pLp^nAHQ^vCFcu7`(Nr&$$dq_sw!CHvi*tyZ25&RFc`Urntk%*muFh?PH|)Lby|Gcf<{d)GIAbw z^`f}8TM)YzM+h!oR4zO)S&dz-eFOEIuF+x;=k@5B@&WoW?1+`R=vQSK?>I%7UG}$R zjr;eosx|NV8MCgVn{5^T-7p=mWnZK+ucz>}zPc=Ypn!La(4cLNDpYxM49*$|61o42 z6M5(z5ZT*(=FP=g2=k41_?D|;7W^EJvGJq%+4ff_|v1drv zd?H<$f0g>zd5HU$rMUOfa{TB}K_|+&h*}jqL{i7pL@)GCp?=~>K09R-ulmQAkD2)u z7jF?W%`z^D+&3hO#{Ru8vOM^R$?G^Vm*+S6f!Dj)3EL);@o<=M%!)4N_o9f|h4n-B z9BDelqh9>`tm7R}u6*E_bos#t-$kYJ3+X8Ra$4ilkA3r9*}p@wtVRDdKhgOd+x^#` zP262A%#%;!cPq)T$t#Q5>f?;>T(5^S#op%PwTE$zWz;-YFCvgzV!x<|L4LFyV}M}=&j*Bw|Zjl z);qjekN`EB75LeHcJOU$2a^aC>ulIjcKY{O;nBujJk+Z3o5#I^>t50PLX$wA({^H) zMWIwu{w=q0C%-kL%12;z12B{@9NpH2Gi+X1W#e6I%V5slhTGw!Kz{-W3jKCl4DT zKT(QS`RCA)+g{NLZzXBTxSPV~gD)`u%?F`4!5iE=@@Q517;G+06!+VsaKP{pJQ2## zg^evZTv>s3r|%bTm?nWK&Eb4+;0LB~wTrJ=`GfB*IZCb^tK&7FigO48*0DqN)A*yl z>%cW^si?_!yvSYXEP8uvIpdpu@|rh&sIoy1|EH#h55KsTf4991eJY^R;9l0b7l~*kFFJKjN=s==ri~8NFL5*D;3X( z=S9)PXxSfr`FmG3S%G5aZY5#lw~+|CdziITns9%KDdHOiruaaPRn7ZF&WU-^Wv4RO z@SOeZT#yA_(L!mBOf>x_S`X)Izp>UnB|dd=5lwy_R4TVulE3g_Ia|Nxk#N|254dK( zpVrijWL>G-nfb3L^nvF_`gos+W~b<}wqU|+E91-bGM9_asx4t>zYXL6rkaYAV#FSl z*4I!Q@){(fN=TNEE1g4b!iANugu=;{n5W%$ESeQwOo?d3D+7 z^BMH@uTU)B=7@ueo2g>#3qJdO57m<%LCtPc`lHkrXXQumLlis8v~9Dfz(}mQbp0g^ z_;(b$>J8c9CUyM0RE9YpRHD--=9X#Q7{c$&!)7cDWbVv-;)c`CDd zA;Fe+$6`?37rr)7obhFt%pPpXXVcZ{S;T`~DE(pyIvtTjE6;3JnKFTOR8fB6n#Hu2 zT?5_L`+P~2CtLUPtMI#9BpmoWkvm?OO5;z2(|-G}!n*~wLL>c$WhZ__u-vK*Y;b=V zOKQ!*2;YsoZTx33*H)+OZc?T|W|)j{%iH&5n~7U#U-NP{)Y=OzCk>)};3srU>R@)W zYQ!vn<17~gOXoL@=d*fx@T;JWzS^vb{okHJl->zeR=J<1udt(%aav60=rcb3@hMc5 zF=jz8k*(geo!OqT;lkJEP`BI0INR$aEfi-=A6L$xbNdAB{oYi5r_FY%{Gy)sz6Z=9 z`2kBk+DYveo?+YmB@4^uU+1UWW(eGxqJ;0ej!_$NpMJs%C(hpXC%wl^xrU_S?Cq-Q z!uX~)7}YP%;JdSufB0WIOjwf&mbVA7$XG+1t^Jf4dyS^%?b6&Ib7k&w+6vhBBZc~= zbkkqSo4D5Ga{B#TDyN;F#P_}{M$IuQ?90wpewlAGwpkSj!$Sh-cGG=mbTqoGZGjCDz;WZB3ZOImmjiJ|%`_Of*jE_Jss=e(GtWO<)Pr9GP`7y&V>w*=v zDR{zH=*95(Ttm9QvwJj*geSLGD@JINtUB z9G00oQJC2?ncI$x8|k)@tDb&>c5lrR9NrWm*Yp%ugW9npE6z;_@{)QIJK0#nl6h) zHKEY+L6`n~dy=1gY;8%T!3f%RYzNyOB&J|}>7dsSNb=gRU8sA=PTE&<2_B}K;-q^E z(W&$&`aB*-bL&)zOWQ`#xbSMW|BQ<0!?Dkx=^e?RdAEq3I`7Bj|Cz~=FP8jCEpJZg zs{wa%ejxq5H3(-Y8qk7I^SSnUd-yJ~Chg?j;Ve5%gZC2I3GdvqBlq75F>dHN8Z0B$ zK-$g_B}i(C?5Cb!?@x3w|EU2iPux3J73Ufy>S!}d_XVujCyjreHj#xC%;cvfW@D*` zE*?Wo9A&O5yxS$NC)O!(H*fng*cs1~&0U!1tjl~|oCdwxKSESw4TMZTVa+Q~WHxVeg#w^WyUuBv&~=-=*n-*Sovf)0z3$y7&@?j;my0@1(Iq zGM!o^Pb|;hYFyqs+P%E2J%P$Dr}V|QWVWFG0v(WXq~F^s==>Lp+2Y^rZ03*C%xlRG z{;Z+_7DbEmng)*|xxujt#qs23cL%3W?dPJ zEpA@IqwRrwqLZxXW6Ln1mclV`GwH!ij;eTkX`Zk@OP*@a)}$H357OPYdQjrvG4}8A zNwnMF%Z4X~iYz{Ig zD+lUlJzX6mt9U-I<51!I@?y)pbA&hRQjK@@;Gthj`n3}#FgwT8$(Y>3~>1o?gTK2-S zTy5UWa_6Y2<=?LwiOz?d;{9d?FmNBr1Pj*E1wIG4=k;%y#Yan)cCw1~wuLhz{N3r%}ZQEGbgHZ zcYC{7mm<%adUvp#tD$9MJPYZ^uVUYpWeNZCKT}+?sE}V(+JXm8Z-h{BZ$)^13w1ou zB3vnv#N*9u95c|zKD}^oWcEYmE$}D zpYn#6$5F3kZ|Rdc?ksl3Hl{sGj@gZ|qqc-_?w{6kf1G=$uj)SPP_zM0NMEEyrIgjj zhhrUlg4Ivz@%z#yVae|M!qOkUW$(jP`0}V#eC5jP!aeL2yvT~8UG-;Zo|sOAF0xmrhl(7_7KwhGuVg#l#q#|z>MU`i3M-qqnpT@SF+(wP zSyp`h**4CMzjtpu8*mS0ThF+OnPGaE)fUXpEslft#(@y(t|2V+^rT;#&YbWGtb`sZ4< z*grZ5w^D~P4f}D#cdRM1SpAdk3>rrrj$bGXi=GCuKaZkK@+8`?qChwGBvOescP>I} zKYdjvDaziYK_nVAz)S4idpM51u%`572> z)vfGacmx|0p(A>;btBgE>hxVz8kK$A&ThXr!90{t@gWzl(q&~=sO`wFSa>R{Y-iCF zb|trqZQfDGb-lRDy%G&@gYV)>tNs-7pLfZ#>U2HPVOO{{a(Br0)nSXuIfVZ6=rgLFjXMHjdu)H1MYX0j#qw4mcI1EJL}VNaq}?N zy~>ZCPR*lH-Seok#1ua3^<8@KaJJ|nN{X^d(&^zbHo|uKRc!AXCDHvw0d!SpGMz1V zPdLn{TeveUob}u_6?M%B;>+_b_%ri+%Q|;m=Euz$!atEnXX<51%x~ok);zc${YKbPVo#lW{-b%B$N3)>(fss+O~PM9o?B4E)AYXM z!mSSZ7*Fbl!2z zE}Xfs5*@U6hom+Z>UOpqwky$F-_Or!!>0|}fwsqpl>niEqgFk5eq5bsC zsflbwvLTKAR7i_X7tsIS+^25>OK6OR7SC!E_|x_}_;}_3ZtRSpooV58pz#8idoqa& z&^V6EFC3*&R~5KVzOnSy-I4V6`@?K^<}()Pu2S|-oHMJK7r}fKenQx6c`Qm6YwJsl zY00(Exar9QoP9@~E)pE2<^mtu=Cy_MTDY0JaWsZI_9&1^iCOJNVW0TX2iLK(3oGc6 z72~+_S6`_;>!hb1=3w{Ay`0Ye6mEK9H8&|bPPkKb5!@7KU_UYsW9uGoWhKW#F;p~# zyWQ=^ZOpUdKyeZ0eqseT>7f(X9%{{9t#zOe#k?Dx`X_W?!fooqQ?99D8CB8^m`a~9%^!3Zzs{dCq{I-?Rj(G1^!E44e6ZT1#=$E60>=RiT2O?#@cNj(^KOI*tCn8Y~JQC z?3e#TT3d9He|2z-sPt(l4H^?cQ!|H9hq_VRe$xaRVqrjcYAxjdJMvx_uC$nsxX?x$ z3fFK_kE6MzwSPJD5-syg{sE`Za-E4R?+YFbj1i{U+lr=le`oK-8AO_e=IoYIHM1;8 zVh404)91d5EJ7}l9lBS>Vh)$G^@~fG<(q7@^wGuSgX{RdU&mQqvzchVr>7|RbEj}~ zy*ux5!5P0S-_L$`M6)3o=h*cfcLeE&3~7K)0$H_6obS_=&oti3ie3&?5*ei@u-C#2 z)|EDyrF<6mF^zMngWSS$?@M0gb?q+Y-KVaw>ODnbUD|0nJW9++FwN#Sx<@jtl_!O2 z2dtU@eHZpF-Wesk53_{nEzGhtNwoZNut>?zP;_&}2yUF#A^PFm5#dH*;k8gyz)`FYeXS4KJ%OX`C`a&8lKVXrSqw~9th8NUE(cl ze+qZFEoCaZ?y~pVUQ9hyf>~}H!ftr0uw>b*eBtz7a_(IUdo0cs*3cSZMtiPJB z51hfmru(uRmAWF8t~1oUv70PgHP^h^uo1gA z?HRvisVv=kTg1CvY2j~V$cm;KK4%%yM?~qj=7^*Xq(wgOjCUg2gqWzg{__BcG(j+JjrdKPLL3CBEvST#rar@Jp7{DIec=En*C*l;no25CuX zrqCicgns|kMjPLmaGR4exrh(DxFxnnxEPTv?TZeB>kiAfFGe#sa(@K3BIqzTvq+s= zdukBR%1dx}SR!Zc$vC@@qdB8LVurO&e$+%88CjftX< z?i$nCD>JFhPASgXno<9*LsT_7j{cDOjXhpHoWkgZ=CvPon7_F1#Qpwon3MLE=AJz{ z#htgb=Mq)cP(hC_7g1GCoedw-=6Bn%u;mN2bavtFKj&lgp%T3EYd)RjZbX$lTsa4) z1};@Xo!i+<=#ikU+~SN9Dr@75@%Bg1sX!Y3NKd50KYtL~J`UmM?+6s8#dz}X=Wn6X zcF#+jUTnnJ(1~2)DjjZCauPmAZx*?f=VxyYr~dtF<^$@lxn)-!x&FCvys=w3^?NCf z9}-z}8C#ce3bv!To?Wf<_p2Pbdg3(t`<4!OXsikKtrF*N>eq4On?ksul0!Jk*9!Oc zInbW=IQl2el(}bDV)LGODjNO(ABsIx>U+#l>Bs@jzrCKj^zPsQpJwi=bj*CIjX!s5jW{3kiwo+ESE5N$ zLcTP0xiCO`56$?mga;$B7UEhc`x&OhrnX(6h6cw_C^?L(Ei0wv#&5W(sqeVVIaj%W zYqs=ossb~f7R|es1O8*A&qYFSgUsFVE9=+kL?N%qx0&^*B!TQ5pP5ZpW3m0-EBN zh&C?*>8qUERM7Vr<16Lp2Gc;{^COe#r1ODLeQ1~9S>ypuZOuPA_tGbuW=&RgWfE?MXxVAL-ReGxk{Hx?)#Z<+^30yG|*)`r+E~($9T%j zKy^Sc=jJZK<;}k!P-!bOjq76yE`V_iN3FY`(`?%XHjJUl&Gx_tb>x7%; zTkwm2F2NlOZkJv(S0iIS#o?;albB5TXV#;An!f8fM}r!Yga_Y@;f1q@(pcSHVsDq6 zm_^*f*9K>^OMAD|SQ7&(_b(p(FJ7S@!=6)(qEttZ|930J=2q*Bb zMjoXjn+B-Kv(s35;IlBteJ)!Yc7!e2@R8T~J)Oo0c|5SxjUGBShA!32rVsX1(s4;` zux>>e{TVLLC44%>ZSXxr*SFX3kH^=up@xBYerhs5b||CP*OItXR@1rqF*~`BUIujK zr7U#(X2ci$xgz%KW%F7A_j!lREIM&%5IYNb_^fg>m!>_2`}*FLqeH9c!-y>UCSfPn z>Gc@kf*pQrvY?N|8BO-N^%xptLU+B$qgu~T2{k^~v;2T&9AnVJdp{|HQ#&nadF}gr)kSpPtX9leC*gxgqcf- zKV{Fr4n{DG>{*80#Z<~L^Qx1hgU8{6&)*# zZI?jPACmZJ_h*n2k0F1`#ei?+5O~_9gxdTBiwFeRIC=bH+qo~eOJJ4!e97V zFcEF8hGA*YY4nc0ggds3K*NqaV2{p#)b*Q?u=XoRX1#{C5p(got|dO&D=XHEog)KB ztB72|1(5i(M_|)8%}njbZ?l{I-Jq*h0dF4{Lay-(sI0011+P2M?3;*3^G9G^T{(pG z+JM4=4bUlC49heY(uh0ebf4dJdL+XRJh#n(odwH5uXiZC{PM(z5vfb5q}u ziBI-{!hQ)j|7#u@5Vt#vZ!nCWR)8N4robu{ZCKiNo&=}XpwXn;_`BpAezY`!%_3dM z*{lK*iKFm$@Gx}I{|;kDZo%4{hM4zx5W3}pF)Bt1|H@Xuy4UM)P$3GxTo1&Y@y6IW zb|(7DjmGH@Qo#00896od0r{L!2;mnpL3LswJPDG-xcy^rbL0}d+kXVN2YF)Q4P$(~ zXbQR9W^T6j;DG7=?JhV`c>%utHU}jyhLN}uf3k*Ek!|KqaN=127-(OFqhp=1xHSQ9 zEjlIUp!@~V_o)~;%@y}V^b!BO77|oo07}aBW^eb+5jeVZm^BTkkezd<5)bQ>K92nBLB{p5tujqFvN0+;<kh3A`=IOgIkF*o}S+X zT7_}2X~Q`|uYDNlb+CamN5t&TedqDl*yng`OCVu^EK~U9y4M`Wu;Pn2l5d5wK z!k7OAb@6?Ay(Nd-Hn>8r%GHxOM}2AjfESI+nMWV_y@JPmWnkB+hF5hBpjT4>f7iuA z@V~o)4QYo-bKOfaGsYRyzQ&8zoI7yhDv|hZPBCaW}!Q)f;N6t_qY~C27dcpJ+B(f?A#vYxMtHh5-rx zfve`A*?0$-^|f}bsk#0o!R@tA1j&Na0`Kb%SYa|BkMu9YjrF7GvmZdCXW7sJZXXtV z2jR7g$IvG7B*Z`U(q6Qn#ttKWYX(A2S1e zg4@AoZU=;BXM(PSG8L?BL?fNa_*zkuD6P)01 zaiiePkZi%aBQ>V0UDgPKUUrjq{~`Es%3Dx&xdt_9=VA7DV|Z;(K<&y9G%)@SaoWWY z+8j^2hn%E-JL74NOaAMbAH0oeHxgqjVPS57G19? zQM>z6^h@za)H|Yv8~;k+MlqjxNN5183@?Ow$)TuLlZY8C2t_?^`2JKoG{%ab+tWFC zC9)g!*$wobdK*8!-37fz&(1@4X<)(Hhc? z-Jv6Trod1}dzQ_Ss96C{F4!YlhQ~xRaJIz}vt7L~Y% zNN~JIm1qvw5CeC0TqdE0PE-M-Zi@N*`9{>=!G!9|sL|MV70TxKi)Z4&ut`b?R$lQC zT<8JcFO7$74@ba=#cA;G_d&2WG6xmX4gGSXFs*AWW*KVJ5!S<~(w&i%boUbVZVOnJ zZwG}}cY==bYFPPx2}Er(BC`cE1e+IKG#jgz580El;P;+v_@UGy_*hpY@b5n)P+IRv z=gk(-5xK)?Rm5578&(W$vx@;`7U7iTvSOB64a_TA2j@3GgBTeZJdm@4{BDyYD<&(F zE8W_7^7t6+H!(+*8%Hs8QW~E4mWh)md?PCR0-^Xs4Xkd-hqX)Mq2N#maC$j7`CKB( zzdV7qQqw^pVSq4&YEm>flxTiP7VuJ+&7yrQ@sG3;-m$m_PuJPQ&BQtIl%u|cX}0QJC{MJMjGhOUkVqa z%<%Jf@m}v7j_=PIK=cJixGl2L~-9628{S4DtclVXRm*ir| zRX+hH9#=rK`WTG4aURZnTa4DTH=tC-PL$|=0FCpnLHp(u2#5Lfnund3#Z9TT$$q@^ z(h;xQk4D3JKDgkG0XB&gF)OhS8$}=Rq|P5ab0Qdi*5tt7Rp-D?&JOP%j=|S+MREQlnJn!+Nw&RFMbEjPAo$J&cq7=33ukY`;@SYTl%0&{zx{)31J zG3!WyV1bq{I=>x-FP-$zb#)II*8K##4LM+5!Grt8R_GWTfIU*_D8Ym{`C<&dw>X8` z*CLV3{3VENv@>x`Ib*tOcM~y;UJq6ql0ofTpuoLlx|wXBoZzZ$s7al3l%V5r89AL6 zLAzcCQ=Q8$)MT*N%%(|S(8{F=?*FVb%?=@gy*8mF(%u>xrDnq@gIRFBXfHmEqv(8R z0*3pGv$fxDr<{owP24KP4}3XFuFb?}DicVz(Rkud_YkwbUHHyk8COc3g>U!Pnt3ZM z5g7g+LCB2@K$aTZa#FrEfYX4IlN4>v}oVCaKbq!+yfo1TP|=EL{M!@YL} z>eX&$t{v}8li#euy{4{cqB$Kq{wTo*>qlgP=XUa-AL)~kp7e}=5Y?E47<^$8=Iw8W zeS;h{mAOL8*J#K$bj0;Ptnut`F>}Ugw%}lpEO_r+0mqFhaPgF6JdV3iO?4t2>8eY6 z|0vVsRjt^xT7>sI&f)9J);Kuw4Q#LT1!Yf$H&*1J!}J)ei?9;hFAgJnW;w#M!4yGy zfwGyAM1w&2qj=64WQs{AmZNS~G(5);I8^2d?URO);GG#{T5m9nxcw6b7PY|VtPgPU ztvB6!#)!V%_6t{A%A@{^U$CgM8`v~E^6B>p(!62`Y;(1zEBdC;S;eENpCdCpkmD-Q zGha+9ZiV0oWnVlfw-=uuuQ5F@vyb>?9VCJ2QrP)N0?l;a!|DMC-1JKyUq-$Mf6*_j ztZqYxhp(_k+XPC%3T)&Tz@K+d=s1rudfF_FDk+I|UWN%^=d%_b zIJ)n!I&Cjeq@(rp;hX;(;`ikziNBjk4;Vz!nl1b2^07N0cI_$XTAT{2o2TQy26L=Z zp?J(%i#pAGhv%xV;b+|;v?AycK8d@4@lrLoq^TG!cUNM}GKgc-i4#co#Puw?x!ZQsv-rBeK`dSha^C*)JTw-pF#f4k|&0_(d4_a2INkj z42LQ{lZ7J%FzWASFbZ{`I=3C@$>-)&i%)~Nt=Bmc-b?TvP?v~lSRSM;b#C9B#R1OZNFDqR&xb3oW$Dq4Ludl~jW&~Z!qw?v zAX#}De(DCm$Zbbq+=&42?vkUVMuG;2f9&3(bEIPIb24Z$6s~t#krjy&aCi7PfYg&~tNpxZeXcKvn+ zRhJ%oAE!jWpH!xGeGBpX0|z{9y9WJx4biVuAD{RO@S$cgZm7A7!>l^d;Y1V>sss|e z@Qufg3EeGsB|0&dYwXW4~d7iZP(y>el;ActA^slYEY#2@$J(R z%t)xh8W$Uk{Lc@!c1Iw4r-MOViWqD%68oNC!_|M&P`#!Y{dZl41I0zKQAZL(zZ79- zLkceVcOKubI*9%;+tJGG2sV#!gbBq@NZQj^WZ^zzNZITRYFoCzqRzd7Ir<7vxYPnH z>~zT<{dLBR9dc&XMsCLBffZ<)N^ri*7zlN1Hw`p9CkXncX;PvaV01NZo!KZUj zB9Hf7B)+l>aigC&Yq+ZyS|#R+v+X9(PHkmsvONjkX(ympP$2$OPJ?j;bD-DsHhIM~ z!L7H4s9tU*W3@*^qhttK{Zoe6FPMZW4?^&4tODr54{cu**Tr;!1)-wc?^q%2+ z{eL(&MuD>bhT?oet_tCV;iC6x>KL#>nOKuq$u{ zrdFZZz7|h1$#)uD>hm<&Ci8gKk>Ky9hkZ;zi+k=AU#oh^jT_eZZKgSf{ZK@|mBR&j%O0Bg`ec}``jSdy^H;*I zv3ub7$s@$;P7Rq^b%0DA-v!Z!&2dBCO1vreo$Tn@O^kbP3d&D8(t=bgDhxNIpS_pyK;if`~gF;2C)aG^aFS@sU3$`X@MeAg0$SC;HTdPp}bVe#e@7-4)Abq#~DUNaqj+2p`z`E-~J$*7;5h>Nq2;~+N=g${;T z>!pFN`B{Rbz3N7N|Dw!P4G-X^*pv8{?8jeM1tdCVB5~WLPF|P%2I+!XXmDd0TCWJl z=><7hIqxQ#xH=;}Sc_MxrRa);t*}Sc0xEtvf%A_d!C##^vs)9uEa34mIBBCL_QRZl z4+kE>fe+zm*obt#v>NmowZVwpS$O-Gt?ATDXGvI_6848Cl2hti;FeD@9FF=+d>d|o zY1J^SDLanuF4}=hf+p$To{J6I(V($N5@er@M#u6cuvfQ$Jhk{oiq2L*bmMS*@oox< zouUPsenvo+I9oZkRTcyTVWgq96a=;9;CAREyxX^e8a$aqy;}4r-Ig!t%aS2Q4U<3^ zG7&R|=3w~2KlpF!6p&f1Oa7kN2Q4{C;;i-aC@GzR-}7D*qnH}v7TOKw!%{Har4RZq z9|rHbTX<$nJxqUjkL>K9M~Ze>!_Tspp!Bv4)Ne9ipWZ;@%$?{uG#NkG*JJ6rpX8Rt zOS14-Ezzu&g@Bkm_;Ym#c8*RYS!eCRdGjaG?EH#rMs1`Qio$7AQxB0km~Ez@u!NXx ztANsM2|TNmf@8D@+5ccRv1q(cR@;4r>Vk22_l7KX7>+`z`>SE%(@A9dvOlIvTb~JB zcIA*o`e&ioyqLI^Ye9^D3mK(fOdP|i2@9D3DjHT~y~Z`OQR<4|GxjK{nJx#Xmxp5a zg*$NK`XjRVR5T%fv!PmOjQ`zGp!>=s=_O}VYX2h-uGG%Prh`$qd*lIhec*}w*W(z) z{e}M)sNvvQdsO!fg*#7w62)ny_c|(yJtb$(*FdL-1Wx5*;n~THAeB`KnR*i-{Fiu!u~;AeO?yPnttOC| z84H1z^kH($H)0fQ1qbBzp}{X6{2J|Hfloc?cm04ATS=s#iR*0S@aua4mVFNayFqDe zk2l7&SubFyRSOI?Az0(Q4Ud(dM2QXO&}QK|xKyWzfA_7zn@OtZY&Ql6r!K;MKVTq89`wVMNrlfC=vGHO6fhQREq_C6Vk=prdY&vxxj?354*}(1IgrQ?hGMOL z7(V+4L>}pc@zyf*U7{Xs(4R@8rb-G-ab+tr!luc6DUq!CCNiRtdUqt;e@vuh4l-7CF%_ zA?Saxj4aCELLZiU(6aXRG_k`0{w4*(xV>gr_uv4@XcfBa>OWFW-=W`u2lj{MQM_^Kh!w_Ju(CEC$`@>jGd`GW!zp`d!zP%u4| z4$IVf@IMI|TG^{aH;lUhqc@0P{ZSQkPX$n(x&U^%=0Wg=NZ8(#3PavBz|97I!A)y1 z8|B|(vgcbObd6+CbYBKD3umI%0T=9?ej2Y?S>SvzYwzRPY#jM=8T@Ph52j2}fSf#M z!RIxT$tbzyI1#|3M&^1toO3fWjTejE1$#Nw;o$iFI?JLNf0juX8;GY`;&wq*QO{WIfl`n}W3Wi{Vr#WtAUU=C`3YzO{ zN!G?*^61%JSgm>oA`4%^)~&~g_mpI^QBwsrc^)^zEu+bXx|8JXUJlGcBB3eoFT8a^ zY`7GFpS)^u%k@j}KspUh`<8&l>lkrO_zrsAMq+_l8OYe*g3vX+aQuz91Xf#tqXVPy z>Z~{5ZhaSGzCQMmCNmH{y~=n&Ls+?c;~v-A>{6rmunlgCk^_*CTS*#RC3HxI&27 zA6^l&1F!!I#xEz2;E*Yq0=FNFh^5XMqJOYU5XglR$vK~h)7mL8Ns2&n*)W(H_RDPa zn@56y<`}ZyB@x`L*TVATy|7AC?5Vxc3`3n-VTFepsVRse1v6CP>vIh}RIiT{90M`V zIS+O(_zvX@0bN14m z6Ash5@ye*-u?S5Lcw$)HT4L*4MgDC2MyBaY;kP?IP@M8XToWYVkCFnkdD@C!K6Vm& zt0~~>zaE-04q&@&IPUs!0<-(XtT*pFIJ)EmKI{#Faf>pbaQ1Tu$=HjtO_!tIhWTjJ zDT(I=&bVAL0sC&cfR*kZn3@m>-O-<5+ihczHUBFx%=8hwU*Bsw%}QM`yW4`)<*5o@ zP7Wee+7C|^F2|{p*5i}IIWTRm8d$2N5i`$Qg1e6|lPOxZP*HlBZ2lNQ>~t2A@-Pux zq!}<*5Ci7+gw~!{qBrKY;s(pBn4%MgI7tDYZXAXW?lr^Pq)2e`j1y=*JW4b)Y~Zb! zn>)f`4J>HaB2O1MRo8xTLGrA z`r!W64W6BUK=z-m7EE8Z7))2|fbB~?*x`DZ*xfrvLVjlxNpFgse>IUT)y8T4z98%r zpTBA>!9YSDTP-HSB^EA7{dX8smn=cIzvFSIP9LF-y=3^OFgUs@7DtJ-`!hQd(X}E0 z!!1(r^;0{%W-EoGmnmR#oGK>x-+=V_KR*XsJ$z zCgo^BY+a;4@%9oygR=yfG+7flDIEcC9}VNv6F_lsEHt|$VtKnYdM$nphnPBC_pKyl zTa?IruaWeouR0wk*7yA!*GeqH!-@5VnzO_3^`^<8fDjk{b=!?dMk*`NeeXVZc!%k{6}K>H3>@E{9)vi8{}P7J6!(q2fTy6 zKtSmM{Q7l2HsAHY$~y}%=%Wqtn_3~aRF}q?Y0_=|iqvuHb`m%~P_SEhD2YA4i-b>k zXZm!j1o=AmB>o?Y&O02d_mAU7*?X2PWF*4#ocr@Rgm03PR8mHXQc*M{l^L>Uk(ER? z6{%22NK%Sfh6eZmO|roEj$?94c6<6VC$6tXr4X~Iro3UfcOch`{)T_ zp1sh&*8wlPX`$WhVMuuAVNOO!d*VcZ7W2dyMe!a3F)bZZYSU zZd54ziYtTvfO}~faQ)7Kp7Hb4Y58Jm_T_$(RwRpYi@7oA*LSG;X#uf$kuVv|0%_|y z>eY_T)Uudi_T5P~y@MijV)HV3?Y|Wubo>)x#S)B*OZ)u*Xz z+iz0R!c%Z=Yyd*vcfnBdN_zi)^7J9D`Sf5|32auZh7V8mA%PBsT%J~<@|A;B{T;?k zXJ$_4U^<%J;V1hiXQ=e)IZE$qHQMgU#>VeqOm9Dt_T}-UTbZo*X?h=a3v9u@QfI`y zYhmD>O*OIa3II_Ay1r*1u;lsUMkh5ns4#vpiue(3NVab89(-pD$OC(hd7@dRybw`BG@&xBELp)9^%y9mV( z<)iBr4qA&_fUY-O&7R@#qIi!~vLA8zLZ5s&Xa_VxM&TSexQZJt?cYM=yt?728l&a> z{z%m3>`2#>2-0a`a_CPWr#;M;NtV9FP0gz@i@~VZOmN+GFc^I!`c}#%E>hs!ewZ zr>83OKNIlhh%)l`sNmSP7@T|HkMS{1n9cVG%3K8T#-C;wKQ&AbHTTl$8F%O;+YV?M z+YDMmY;d{X3`${>;VehfNG>JTB$B;3{7BV@JO!FM^qv5C}R~-Z1P=1)AMnP{)dHpJCG-{Upgl4LT-*ps@8DzE=d;LDT7#UM zHAG$`3ly+fgHFmX@Jr-pT-m{N=zIF%Q~Gx(IX(?(9*QKuR*58A22w>*QPh*#`Baea zF}C+Ve{4PCjaik4QO%5xq^*jgLb&b{+1Ff{tgnmC@msJpT#dZ_{fsL7Ek&;D>%u|N z&&1=N3q0GO3N|JuA*$X5)F%D|^^%jIq&o!Vo3ydu>^IOoTMR=gfmm0t8TsE@;m1E` zaB^h=-nbN|416iJ(7#hw?^ZWYc9y?5n|+KSxlX|cZKxH zrc*g$JkXiu13Ltj$f*xLL^zqrk+hl-{xhmX+-NIVR4)&D0;^y)^*3qMJdM*Ax8aZ4 zWf&}z2$@3X;oI%Cp!bHNbt}wi;VO!5?5!pop3RgXy@2|=WF?FrPXcYLH*n1z;q28Q z^5c^?;ksoDXZ-(zH!=z!e>sd6;IX6YY+1B~M-SxZNurtz!KE}0IvjinMSJ_*5$PneMXm?c8cx9BZIqOWA$$$dpnLC_}9Rei9>eaJQZ>-XE6r$7p4IWii3*LOAT=x;t~fP=&WAaeZe-T~4UY7`p0%_HIO4PI=TIrW6?;m`D8a+> z)IfmWPgac{t zt#t!jYCwAHaedk%N|zo8{R1ueVmQz?3ugRDD1WsK1!5lIgEyMEL2fy=ifzK$7mr{@ zM-g8hSHYInYFz)J5Sx?EU}q~IUC|^-FIH8ebuRBhtCn-J{nq9MT7mvQ(#cNoi4K5L>p&2(`R^Yl1n}*WH571`*nBlC!R-ET@?I-i=K z>rYN4))V&$5tLXWj+{$hgKOe_TvPT0hs>X&r~?akL>b_g5+i&W$zvSWCS_bZv&c9j z?GnWM9s+$S3X)4U;bR>;)Ew}@mSueOo-A&fH)jTOUvg3d#wvuphw*BpR?wFxEty%e zwsf-6FEZXD4%d~KG49!#*z)Z+aLTsACV?`P6l}(hf@-wTx`!&3zp)E`VypUnT-Y** zBKvFc?!rEt7J7gScMqV?4qn=M+hUqyH$_W+w1-O~uCT4*9%Ln+ptC+i&|)VJ(;W)8 z06Z@PJrfEJxdz}Feiml)--G!(ReZ1_8a_64!+mc}DnUSO+V6 zvfysuEpRk$EvT=#T zIc$zdMfn|}c)~9dWox~#S7VVEIjxTA&@l z?s<8MeSP>5Fv=$WSf+*Nx9~tw)^T#=aS&OXs845=NYMG*oV4NJHtNMKf9kr-VIsX@ z0yld6Ms52qOs_1P{V$55<}d9a2e#Y?<*&_f`0!PD{B%3|E;)$R3>V$wvjF&W-X+2z zvn29k6%iS-0K=V05NrE}x_w|haqE9h{gik^_HTSf3V;12UMpoV?ujyvU0R47h7T~v z{T{v^tH7>%R#-K>2oD|;#-k3q=@HpIv_PF7-SS!oqwgKY6q{uH7PJPGvmC){G!+8& zMS=leBk&C)j*l_~TP9j?(6+po}C?0k9w&rP^6{TS}Iw19K8B)*E~MZ3ES zIN6IZcx)e#-xIL(-6zx(dWW7dKhV8rA->a+#mB2%@$yJ3o>(x76&JpurMEQwp;MR+ zILR=@zPf;YbRw*>;K6(aL;78j34MEL6vrDWQ;Z zY80%59B{hSmo0mMgKaapg<>~vVr!A@>;}h;u;hj~UA^@p?h1-QtE>+8H{DyLVCzaO z=X}G`&iKIARh%HDS2RiRXK(8BW-s8l?hLadzJ|Jbsal{$eY! zOkfWNJHDYp+Qo^qoEM11Z^fV!oVfSy2M~U-kKz>SBq!IWLE_0~tc^N~-gi4e)x(s= zCj;0su^J8D#NlZfTeK43K=b$jlKpQlF;I*H#dmq^1LM4uR!TWp5MYcq&r0G*z5sUr z=SNv^o}$WLu;4n!3Yt$&mAO}~Xrd+t?P7yu#qt62J@p7w^wmLWt0fLF-Xqzh1^Q|W zAg}ofak#h+yk2Wz@cdAaKEDilBM^pFzhS-26-V=psvIbphn0c`MiqI}m26xwl(WkUF z(H**L=y{b_VfnUI5YBNNX5s{K4sXCCyJ)!hNE&UJtW^4#Ev5<@gJ&G$+iYdQid`x6 zo#&aft$ilVpQi-FG1J52evhl*gMICUdo$6J3@|OWb}h+UOXOT zYvQ5R>iFuC0VY&Lp=G)(ic9d|(R~%nZ@aPXqZe+~zd;1Its<40_sGdE9@=bwCU!ZQ zV&C)&@X*)?g6qU#M&uNAc)K+_veyz8FBT`JBLQqnxvj*tWr8~9P{-D|f1lkl^o;GS z%E_)3<|I{hmvMtuBknM5!RVbq#N$*Yb-1gCaJF?A-0`e4xH~3felPg;U$YI{o;{VwIyK7n+ZLmK9t0x6_oE~RpMIhjBf*(`HEH= zc%;Dwy+amYTB$g4A7XMy8`ffbks@vz8-u>158wJFz=>06o>mj)dUBUYhS1m zp+ahLumN>QS%vN>SU{W9FQnxog30#_+QcMOhXg)}Vs|!dX0M7pO8p55M}a#LNG20d zR#BL;UhBZ*?99oc{1oDd31ms}W|Fb`6f1vrIor^xhW*R4h>B6^Wb-VEri8O6;ErW4 z<1u!_{E#)M2fSDj)B~p$=Rotv%`o3j9(Gi2qL=N}r$2-!(Ya?olDc+2SZbsW9#;&h ztHI^$(^v0OR|LIrQPn}rTO5uOmm7(K0zVlMx2JyH$)vCO#?yfeCz2lAPcBE*lJLq| zqI)MFD$O#XLpc-H*Bj!iav9uOC5q$qEAa%|9G&tH;HvY55S|_f3e6GF$$K4d<+$UV`I*7K# zduC3g7*V}%OM+JJChto`sfvAHsEPaMaJ*In6;m!iHor1hqz1vxbw}Z~b2#R81>umS zHSRuf0k&QU2h-6yka6?|KC^DBl0S&}-cARu&R7sq6~fQkFTqv*GEjG(fipG9&>A00 zVnvnVy4f<2^9_O7igWO)brwW!seskJE7Zfvdbl#h1C_htso<^mSdI^Cs1+mP75+ra^+^<&ucZe3s*eP2zUJA{(I3^Q-kMiO(TnGE!k`>@|^ zGdU`B7@UuZ;?ip$QE2=RF4vT#qZ`6p|HI-jtoBI`UqkG2MX^;~kPAlebmix*?yVix>Y z6Gh7A42hlhPPkkW2LG-34W1aeD=j_gu^pYm4VY&uLCKXhY@N}hwiFL+>p>551k zawq?yw1`U*FDcEhBF)+NsUy-cMiaIw7_eFo|C8H{WgnRwfmI6m`_hYgw}aQ9sbeWC z({nk?!9m=LYpROGC zJG_PN=PN)%s0_T)JK>q0JW?12XN6A@`&1qFyfgK%GVlgins`QW(e9la4(6MptIjbr@y|fN)KkdB&>`yEd$0s2CNJF$J6%d(y=5tc@rdKb z*TcXs)dcy1S6Swshp32*P%_A@yB&q`Mpje}t9NoS@M%0CGn>Ro(XU)Ey}Sz)Uyl;g zWqKHuZ32hf?1&Q!!Fzy!1pvj=Y&RATSnSe)9P1RH$zgZEYn)_xQrQb+GmyGE0! z>fEc4I`jZs8+u@2bt1VIyq<*inv%jA3utYVftBy)$kai7R_d!#qYtc$MnCWEgxHWc z@Nde13;)@|Za;t6ns5?2E4tueh!AtXbQr3h6?$%*hkpwqKzv4y?p-cK|7-YyZc8Ge z0)LPtrfKAWpac9WP=dyb&xj$BLzO{!+@Gb0OOj$qxqA`$Wtl~C%pT*uCNY|wyNu4K zQ%TsgAd#qRqgW1?U{kpcyjgXRtR3YC{I^TmYf_)o*=o}eF-SzH6tYWn~kI6*k@J0a3wpRcyUwyp+07BU@~6DVuXx3|wm)2v%Jn zbJ&JXb9Ydq@iyK%WQ-D`Dwy<97Tug4l9e(wWH!Bu{FNU7N4Ih~``{v2I69L~K5KHb zDTkccVTZeWy1|~~Du|turrkE)$L(3=*d`SPF;!cjDm(&2mr_t5A_lzc#z?q+871lD zPn{1|CDn4H@P4lq#Hnw9oBp0eE!~5NFO8tgUdz)Ghk0njMj<-DnhTB>^1$Vr|43YA z5u9C?0kI;v@MZ2e^tXCJW9~r^rr03mHcV>F(@4|cMX=PUfXTKdFj<*`UfQ16VZ9Bl zb=+{kuW$-MFlm#H&X?<0hF+l z06Xo8JuGGML;ek2*wOWo`o3U3DU#YlWRf*-DO(R0>nucP#V)e!Ss_u%dkX3?4;Fj>it5exIUo*8{S1AZ_0Y?Txx>D z7uG;Z&N4W^mjw^sYtu#coOH_hH&`Ef4xT3@z@pQMa7`cuv=j>=r=teeERsT-%Mv&! zCWLY4QlX?jmY7{gB`)$2ba;dt9evM>jz2I*vb99vtgbL9gayO&{q2zbX%oDy>Vf{f zjd0{g0Z8Z=V2Zgop4=dUD=v0Ii)b652|&+XXJwHY}S0Udo1Q~xPqh`c`b7^(u_ zXBn_-=Yku0mrz%u5+7?^N1=XxoQ(YnQ;WOc!~0h9>(^#-sNx${H}DOp{Kw$&tE6A8+fSE`R()T{|cQQIA9o-4$o4 zi9L+|FLs(pB<2&5^RGy3-5!~T>>%N7DA|DDtYieiOiQh$JWzmU|HQWXLsiJFdd#;Si~I$-Kop~_b3W( zD>=X&>2P`@-$A;sY&ZQ%W(*crU50gZ8i>Dm08^2^;Bz?zLgpKy73UBXmkb zJ^&rUXv7W)(37wU;$k%5RrxD)ood7v&c|`ViEOm5_Cqas0c>4*9P&dqfDsSErp!y| z|1%Tcw1i>2)_LQM8-CX>|JO> z%0JjM*~<_(%=A6mF6Thxqw_#*I|GX!B%oN2Cq{}Gpl^m7DA&0Gccv{Q@2(}jN`>Tw zy*BXM{6V9tc1-_z5x@CtVW);<5^wclk}e~T&bF@o{K7XovEr<8jPD6&;R_LrvgYnACN`VCpp}W|%{r<2%@VCK(!s9z%9qEj~S8 zj-no^xHc{VD^?&{&xoOo&}O{v$%Rph*FkSG8fF8JLivefutJdFq2Nj|t`LOQFGZ*o zbQ%?3N8xDw1=hU^4VH%K4E5=G1DqIjgcmD5ke5w8Y^C-L_UW%GgfrtIOVc!m-KPAB zt)n48KhzbVhdX}ZT0KotVtkFdr07g7?^9wmb=)@kZ~bK|Xcu#CJrqL(3=BwMdn}%8 zi^KBG{&;id6h^+5Kutagb-4~I4QA()`K%}N+^IZdh>~=ER}@W%!!GeStkev`i8y(3C%uO%xUz%l zzBtJi67697NmQ_Pb5CKxb6=bs3c~GY=cwl^E2)!judpR!7Srh=0<^BwXOt9M0yC4k z5VuJl_6_|~}?y+dEDoei+`;M-Q9;5#eK-)Aq6l1>xz2;fSI1vB}=ng+Cc_ET>72G*62F1Tp zs0H%JsR-$F>?B?@Tx_a}m!%c4H}57GEW8NyT_@l^=V8zp9bq!vNanO+Am!aU7~^7? z0~R{8!EQPF*1Bnw@<<@}AO52bxLu;u+fzx;_4Q==tybhv}*G}K`zClNe7to4-GubAZlGOH{EXqM46IRMv!aJ8l zxbP_q)dK@jIcYb3uBE{~%@8z1qTtJl7CNr9oYtF4q^*x^f_AAAvi?>Lc|BACVmThL z>fsh3kd0$L)%d=z36(WgV@%)+4 z^xSJfTD<=&Sztd+0#4Kri{JzFcCwjXkinv@t62;#>Jo6MT7aAIF(d8Ja+Y(g4ms1Q z4?BI`s0RxDMBG`3lGXS>`Zka?|6lXeAziLSW*Q;>&C$7F*m*S&jR`^!zyade20(K zxaqHb^Jz)_=Op|MfXIPZP|X#@yZ(|$yNTkLMO)#`!AKDOm<$hJXMpJU1CW-K05yqi z*zWQKZC6iW;it|jjeH2n899r)I1fcF`1n3%c(gT3LvRTlv(PxitUV?8P=u#MET{|Ap4Zs2&z zC>%EC$5p}Z$m6~ZI|sJo1z*NXv`>QRl29<^52P+eD6)RO+{4n)>43I1oOtaO2mY?q zg#N~-L`78*jwOiGHFrmG9KIsow@3(i$9RJ4Qef+JI3Ah4hz&XeDCFh`9~w?Tdsi6@ za;L-MyaYJ!7Xu>x3(+Y2H%tif;KQ;`(iicB`l(SzY&n>bQlrP=w@D5dJ~n1+UvZ&k z^<=3}E|G9F*bEXxx54wB`=KiBEU13D1G|6Lfm-Q%P+6^r=CZNit#uT5RWc!LP>g>0 zQjHc~t4+U`4a6fB+b}n0BWesNq3JDs{N!VfzrxpJz&2Z~zvqdoPCTY89@UWULv_^k zi{qFSH-Z}@-lMqi2eR{SKhaf|hhZHRbamIl*>ff+YmrIxYM+q0C=)Qgo`9~05-~C) z60?@b;i5yjm|yRO3x<#4pSnDBd|8X^VNX+4BVqg4n%M{wIP7$ZA?IiZYc6j4gK=tH(1dnH}?3bN?+2&4Cl&Ub( z2`XtNUwMR4KOhauKQv>R)fAQlw8QPw-^t+`W_=HCCx__-G8lS+#JrZIJoZLYylsh8 z(JVKdnD>XQH2OtcBkJ)#rE7RO(h9G?yafCnOs8V(8u+`MBTs%v!}GK1WVYctSauc= z15*)l@56HFGJQ{+=Y62$p1vhrca+fhz+wE#ZGdvO8gaSL04jRO!MJ=G>^8jsHLf?I zx|KQ8WY-g-GDMWra>*jmE!1;Ud)UBP0RJs0g|u7Ci2jFsa^tEdzFixLM-w`+c0D)E zv=2yNMJN$YY@!@ZIZ!og37j}M3T}*FtTWdIL8_5pKT{37t|_4Rs02lhWP(np5u|@L zC5`zbu%T85ixdyzkcSCWIo7}~xB2L@EecHjnSzD(Yx0kOK3uTA3qNlEf_X-wxbil` zVt!YNBXgR>&$5&{Q*A*61dE8u%j1N;!@P^O)+BqjkFs_sBPbY)2Rf!;=w%Z=-yuLw zXoQjB0|i9&d<-^SFh|ad-9X&FP&Yi561UA8NyoC2WEl_=Z@d{y`uT9zXM$b3RPZa? zpLHr`jJ0Z(lf7U3J~4AX3&~Q6u)$uG@L$y<9H?FZ#d$Z0>3ai`r%;Pt`T4m1(i7|-i2?!ECU{u98-(dR$nvR$ zsL6V0c%BQ|!xV7Jo52fhL(1tw3dwK(Km>b)=&(6iI@4U37G$nlX|;;cv)5z6MrF{t zk_tLyWzaX%1n=o`@J?6)N8ZJvU7!})%jDpOG7gwDHibz^Hf$<53xo2^8u@z{GsMvg z<4>-`o8K$2bo?q=aHxWu+Qo*%MmuyWK|IpMFf6itphfTiY1pciPA=^_&K9#}I_&Se;o#gCc-P5~4MA`4eaUZpDW*Vo@TQ?R)5|nG z{|lE3+LKkb*NJ>ZC0Xm-Lmm3uLY4d$COH}#n0*Yvsd6Jc@ll1A>ei(bj@i+=J-bk= z)CX(-bH|fP+_=G$g)SaEcsb`K?sI&G^(_OqcRte_uvm&zl@XrFa>p$*3OHYL2!1fL zi@XjmN4>QUcxYEF&hy@bKhoCXb)6qjCoYH%+V-m9t)E{kx z_KUxW+CB@o_&gNM4|Op4>m688tBCc+z104VyU6=2N1~SHkM7OMXzf#l*Pe{P$chX>tZ(b7lytoJ_*Y3}n@v?tdd`XbwxdYmgo zHA5lg{I?Tsm%V{L-v=OPMx(c=8ZLg1cs;3%)pkIgxT{W5LMJq#{JJw)U7ktBmoB0_ z<4lP)_f0CSubMLbZbe-$wxgCZOt*NZlO&tD8b^L~Vi$ia`szMHp4m9O+m(;mDQED# zeKwu*E{Trbo=WqmoCn{?47j!ID2*|t*(f=^J{Bfw z9ch~1iyp6vr~RJvlf)I`u>V>${2bm0x1ODcz05cF$ge75P|-(e6{=7RZ4a{7S*ub5 z%~#m#1O&-}!-huAJ2n|{NibZp-bVDTdXL`@hk*U4C+V+dwDjAF?6;ButbePT4VUgw zCi5EBlf%E0DGB$Lc&}*%PO4d;hpRUYlQ#582@{%I&mPy7twpi5>X_1%3jxL+kn?Al ztf{ZZINlm`x>tbJD}PXGjq|9dIyLN%erL$rZ|b0%cO5zfEHU)z5uDqQh@Zz6byx#e9)>fwZdTe zT@B*m&r@HY{bm)oN3$Oa%Cl7~Lyb}$d5z{JY-75dHW2qv8RQE~Ab#s97}d`K@gF=W zXRrwJx7Uz$2hykwqbu2W_AO?8^Aw`v)r9Dp%>uN_Elz-ABP7+}0+Db!K%_n}+S$&< zqJo^CM1iTjH(0lwrZwY|=&+0^TK|+U zEZ^Y+lgydvZL$%$b>|I}$rC4hwhj0-<0=*^UBR1kZB)GIWlHCf0jZri#bo9(=@Q{A zdQ*1-6OWQvo{TqcPK2P=Zy4oZz3hoZcvAPcL8}q#G;W z!X5J};Iqnw$Mb}V%Pm!-ola3k-2GdyM`{Tgv&~>#RsrRv;sI6JyNu-AiXamHnefDI z4x-;`pq`x#{c?XGt$oLpUN7{K;j~U;`ruR48*7D2+dqP``UEI#djLmNWl+db9S_$v z!{E#%D1VX*E=EVF3N0YCRbkQkqjZti#B zV4xZlIq^gM-d^I_^p+ete+a<26&5~N3oD1#kYf^G*xM{tqqfBpP#Xw@{Hy=SGcQH( z+q41>&ML2CI(@p z&r;Nj-G`@xZnAgAR+0j%8*uOz3-5eo_$U+0F|^|zc-wixm}V4kIMhJg@IvzRR}|G~ z2w>IxiqiVT$Ktsvgmkzla>NY*N7^H%W6p)Yw)6n7kbeG@n`UhY!04=5^5|AGseZDW z7=*t;g)tFk27@O3ihmh|b$=mYHYw!SPiNSnF+^UJG_k8L`9p|hC$K6c(ZEoTCG?4x z^oji;>aVpSFLs=)iTgi`ajUZ6xmtQgA$P1dqPbrRUBlf~rY8X*%miep)QU7Z)|r zcuW|%ls=PH7B;YoVfpC&*pBscoEZD=97KMKLqnBQcr4iu<b8+?ed3^!Y_uem zVIi%Xj~gAH!qRhv)LH%p%4$?ZDc{0vKyY|&}g0<8PT zjWe%b5@~~J5?iSNf*-r7$Ms@lPDYiih-n7DZ8@;kv=rVCsnc;g4Cx09kK_0}#@|v{ zitPstQHRePem1&-zJ3IpV{0R8<9b~0ZHMX`n9{jaC|s@O0u`M$%JtYgl2(}uar%odTQSOVGq|4Q}(=qTXdIT=>xfsq^!3>2(tv+`A9|7%swg0VBNo!V#^+#BsOe zA}lJ^z~X-?1oj!ieYri1CSHeL(k&R7a|zcFV{#)nga|iAk*L0CNc&?3(h`yo^~4Cw z`n92P!8AE(^b)^!+`+dNxoG)h4JCQ}27AZCSvD_6FLs-}z|Kui(DK)HYP9++H5tmL zNcnvDB*SPkvaaONOe5qgPr`$INqj<^8Ky7TNOaef6HkWun$IHv4?^=2w5ptg{d zTs_iw^fdd@`%(6|g&Fy~x&~ed$7KewPi2Ofvno0`Du5JuFE4uY)fVtC)~LUNm!uE8`Pa!Jg`7@4;e$Fnu0TTmO! zm-f+hpPti+YdDObM0A4E5pJyRX@UEn81H&*1~mSQfu!F8WM$R>S+vs&?pClNb z7En<=1C-gdEHV-+L8tBdi{*|J_?>?#bvHJQoC%pA-^SgjGlOZ=Qren4ies2lDsGhH z!gBVGfj%-zA$fsp@+(P+7LydCWnXd7AF2-^_fKyuzHf(n1gfatD!NSKSBCuCcLd8H zSm6>kS@hZ`N)&@GQ=-4ds2lIi@YZHwEN&Zzd7pk@=-KbMwPz9w$6aw>nIg9C`UK-U zT)_CDF+^RH2LHJgR3hghs;`-s{fc!SzArxj@*4uc`{!8_6|Vq^YRZuRGoGH4OQ%i# zmC;G%jS%v(3bL#UfjeV6e7$}Kj3i1x#BVcfwzPrhHfvCrbcO>B;b8c+6xNSlf$O{S zK=pP6M3)s|-i-v*+7pKxMz3P_`5@#ovBsBPE>QGv9h7WY4_}^fq4rC4?Efu?X{%B( zV0$A@4?n=Fo7dT6cP*%3-Jmf{W@aSW?lgxPmFuv<_V*Y6m>2PJp$ zKlNr5=>CND0x!{<^8u>QttG6-{lq-_IjNW7r`3M`LPs3IRx$;q)nCBwy8^0;^s!Yw zzhpm{(I#Kx&5^o#1uU7qu$*uoC8m@@=%7hbs5l=KOZTux5(?PG>8t3yLzeW7w=_Lw zrGW4DECha+gVcs^?JQMOSEEPTwMG}`uVZUUerBC8dchjaufhP+JTwV9j>X(N=^`rw zx{8aLBgSxx&i*dI^^;yG%`b+phn0|G@(qvE2QXm4ZDvNvSybvX0w26WXp;hJeQpSD z;EO>eQBSm$-9o1?wxf4m+)9f^PB1*ddh`@vW~;R)gIl*fY&9PuS%c9~oV^GfUhW}- zzbx^UjR~p+=%KrjAc>^rkxZ8~a;)?o{w#coo`%z?bgUTGAJv9gsSF~-DZs299^{*J zB>RoFA{L&JM?bYisLgktxV4&+>I><_r=UA{fb}S**KUYGcTRS`dSQ1qh$wyKjOtK$d$z&Oib+CSY z+=wiu-|=4qKkom#3_V#@AiTx^EVf_7Lu=%4osJ8f`pU(8gJ+3N(o=GCZV4T&AxGbk z{ez+ZlGyCGgY4x^rzqTgi*T=)4+g(9K#2DZ#7RDarbA0mrRxZ($Tos&Cw0I?pXrfo zH^=A#HGIgi-snb`HaoP-gB?{+PWf^6kZ1iWu(?tfujMoIl}fK+l1vo_YCXV~_B;54 z(Z5tb9m8ZBMO=RL5B+jgJ3Zo&PQT|oz?L4^&(`~`LB^Fu(co@3lNsygu$#`hezvrVX%`o$< zz3_|=jZ4}NVft-?|DQ8A;T48$*Y{xLA!eVX+K4G?@9>JmC)EEjfFi$#(IFADUd8MV_1+!9#GRUSw>|ICS<(r3e`4T zLY;6d2Ky!#2=-KgxZZv;?D~Vuk9kcLk~wIliBaS~RD;%vwookmfK2y&BFBSk$*!kE zWdDXb@`LUt#ZHe&{a`vdY*tPth1V0lP5wlG_$vk}e8L2QNtAL>fai_;&@y?RNUwJ% zCLjE$`%TZ-1;@K_ZE*|Axt8GNLraLb#e3?k0|&X!WG$Pdq;ZPjXuQ}+aJR&Av=uPN zt9IXD;OQf9(yWC%)s>9@|As{G`Aal5)Kke~W9*=9qpUg3SCAl52F^thFvhVSmi<+Q z71y7T7kUYFt7AC*?V&$yyHf$|Z`lIZGJjAXh^5rD*Hia=`6!u%e2{*iiR3Vh#W5)d zNRT)Nq%Q>;>&(#Vfe{MIQkWpN1*N$*pv|Bnen{>?&*plJtv-uqs+ifVIYQ{Nw*sD@ zdIKQ_AK{Q)JE(NX(7TH{=`Z}T?4vPSY_1JaY>t^*sIA=xdmhB1 zR-M>rD3UPEOq91S#z-lK8FgznBsZKQ!!=2mZEeA4r&IA?iY+`@*iLw#bCM}WyWAMhNsj=d zhr17v2Kxa*c1ptu6F;i?EDup@7a=WTzZx? zrFWAvZ2@H7*JOMFl{{0^nQ)t-o))g~K?hNu%6A?&V^JgR-&4{Ak{Ddi_UPSu3`B!8;D z5y_b$5?!T%uMX;BL$)>Yn{|=?q;&G9vXUH-t0lkW`-w_-H8Hi`Ppo^*$xfp{GHSGs zEb;Uu*Dg*{`l9pb%?2MZUvm_B&KyPdF~p}u^4RE=4p~Olu-fH2d9sC~x;3KM#>1hc zZ;2>OM(-r$zpE%FSP26+qhNS%ENm=m!q#nB*k%}y83WJAHk*Is1nDJPW#hq@uwm3` zj;wSYr?gJ^Q2R91i0K|i8gVTlhwaP>92p|tBBsfy-Aa(nbZ-QIK0?hO6FB$PiSf(` zRO&8*na_>HCeeXJy~!jIPdHK9{~7eme*o6up+ITwg3e9RV17gv=~)R}Wjq04J?n7) zd44Pr`3Uiz?F>gE21dd!Ksm#rUhI+#sa(knL-r3WjhTU+rpltQ*hcw zgwQf2R5$ky>U#uX>l^H4wQsYbmRl?&zA^^X%%xqF)$3_k5c3jzw!MXag>l5cq>5aL zDkTqBbz*{5J`Q&>`>EV$C^!%ZISkLthQw02_Y5hSw+ht!p-2>obHL6Y_L#@p&9bbF zrq1o|W-H+fN=H+XsASrci)yb(7RNv0%20YYJQ*P`(|(iEtwIn+$HQ2yA6yv92EH7I z9SK6%`a}$y`X-6fq3h&*ZvAlc`!R@nJ09P76H>*~;k5B~P)k??HKU&J<#G^w ze-?-W%G>aW`6_I&$|k};42VFaEirsqOy7N8O21oBLGNCvhWy_toWhm3=;bmOx2L}8gs3@WyVQ63e`s6VF38w;aR`0?4$GAoW_T8&7b2LZ8TSTqS=1_f%hiqc-13 zZiyyowNRv%8)_PzHs~a+R2oT?{zQEbmh_WVAnDN2RxgzQly8l)i^_c?bZGMZGBG-zlk($ds( zo`0V|AFr1`{KNe^xBHyy`d**U`@QVzChW*1c=MYky|8Tzz3w!H&K)qqmlw-nY(N;? zxaa`c8*7Q{pU0%-VKAP1YmJ(AU*X4@`%v?>3~KhA0!62j^yKb;d;Js zc%ucVyZGRg`q5bNcnHc&mc`F28bNnrI{aF^3A0i(G2uUX{HPd;!+yKteTPw4Wz~yI zw!FX}-fd_#Xe8a&YDecx6udz1#zB(6eo~()4_$YF{x@SAmA-j^mJ5u&%6H>X>Ch+m z?$%1~M-62Zsx+L;g>^+)Q!n_(-i7+#t6;dRE+k7|B6(M$iB!#Zk<5XcqD!+rVwS)< zh;`JZt`Q|r8}9`kBMjl3iUYAJD`fh|Ycdv|wnWGC7$aRbiy7hYgyblxKx^Df(th8U zzSeW6h39}CZBEABw=;3tiM=>^RV(?s#7SWO7=m5MBH_Nd5RAV52gfS9Nxh8Fr5Ge8 z1&dX1rm)s=XcqXaQmVAAL5X(mRH3qkRkauVTg{cCQ4g~H-sWkp;X&3&tZ{u0RCVaOy z4LWvD0nrwB(0ua~G_qfV_R`lNnZE!p8T#Umw}2mZ6yc5P5VTKSh0_KyNU_^OMsatd zD6vWnR|G3!>16{nmaikj9v72$^UTP&<{P-8rv(3=%){fp_E4-l8>Xtx1Z7PN^uM(n zwZ%dB(W-_Vej^JX&3+QEzF7E}n*}3Hp25b&?*w*_6b@f2hgk{3p|$xEnZ8X59&Wh} z4leKDdFmG!Q|-jeZ{NYJ`#6*-Wpu%Ff&v`YVO7?k@DqO4dBCtr9C38VG#;ns9KwVuZ_U+y6$ z1(WidCZy=KDHv@yjMiQC_^Rg&ekz#`EqXISahez}o5w?5QzmSFcpCpH4ugYJ zCWHL4rEsx&648pzC$sP76N|~ah0lK*^I^d@W@y*HbE8Oq(0j7Az)#i+elZ zHO++DJ?YH&l#R@wRY}ahl*3>cJOgI7c9B&#Ov#0xQe=$3n6T#KN%4|Gk`kOvPI@rp zYnT<>(GSCae|>2Gqm|U;tuGDl7BVR=_2kmxI$}M^9fm*66ZnE=O@M@qx22eDw9q>g>Fh0Z^YMb)sa=-oFPtyfs!!Sw@>Uc7*~J!&Sl(ceh0 zz!Wl=y&XS$33JU;&(Wvj9opZyjhQM8sdZB!%_W(n`>?S8+33nFb35uZ@=-n98)}0| znFcgg=+%{cO@@Q^{xIycCei!X#z-YskuMv%No>C}n4FJ9QCdAYcg7I3ldgcP!(pag zEsHUBnjp+|#Hc)?5rfRKG05cv$-!IXZ2Cp$c327DGZ`3|xdrm=ml7xU!{o5QXEN$h zg4SpSa$Tlgw5VY}jaFJnuNMivtGG(&Nm78s%O%XZ@)S~k!=1bw(#v$0XhF`iK`?)6 zH3<{gKm00TJxou6f}#uE;s~@ve;k#WeF*nSk`Nt|@cHd)%;;(t=HApGrc3WW)|l7g zl;9$C-L{Yfj2l57G#Zn8cWcqVTi`qxxnaB93A$)XHZ2^RMj63xW*p*%PeRnN*+9tI z3|SBV%}^n|I!DN4f$34RV>pr9B(TqJY{l1W=i$;LZ^^8q3UFiBZ&Lm=f?8L^QX88i z^tDbaJdEgr%X5@)=s*Gt2wtF>6Z2s}iG!(^kASCABAnFr2D@4h7@pz^KYh31h~Ji& zBd>wo$5K&yco8PJp27N{SnRUXL_1X;9G?fENO0|GeKy7y{~f_Mv(oWVWI9$PEy40C zFAQ{Cis}>ViQ2FR=D37qeB)jUtV#{cZ5Ph}I(g#TEbK44g--qpJ&gF451D3*;e%Tc z)M$pm!Jb60+clC-+$2LgtFGbtzaijq{Qy)QKLGY?48daTL^zYb14fvSq+7b=>A69V zu$nl)ugW53)coOOMx_}RpOq7wQp0e<(YH)Srq%y;JQ%&EMEDZA8#X^#4NRjx_IV}4 z$I1<$_kI)=J1fzLub*NzHHXSAhpE-M4cs&A7Vf+}2ICEUaC6yOJgX{2?@W4& z^O`%b>!3eP+RjknjT6>EH{f_e7nl?)Vv=VtT~<7cTJI3iOQ*_F^FlsO*UiBC=>~9< zyFwxs?Ib4}^We#jF(?`LP3U0hGC7KVOl+wk@mVc~LqpVYn(k#t&{~9ATH5F(l>md9 ze9^@u0dJ{J!e0+Z!n5A5WNml=S^v1t;cd%P$FI8{h#LPBi9Xq`CQDI2CO?y94sfjuKP)&~B3KMWxFatG$#lrf>N^oHOT4IR#;3hE0_oHRlDUpiyD1zs;;oMGZ3@1rA zA#gUnZ;@qcQZ|qRT^4le|HFo8J*>OYN_*~*u~w+ zr17ZZ!_X6=_VTrGP`(Z(4ABE4y>gHVk;D5DgYeM=fzPlul9+N@Bu1u`?5yh| zCv99|kX{=2?AikBrEiK}*SbKL>Sg%1MVMKa>Qepa`xvrS-J!#0FjH55h$w$xi1+ml zX!Y8O$LvxV1qW-AqgYRBU-d)zBR`lg^OhNwa8zWTVnF7fzf01l``K4WjyrNWk<8T* zvCQ$a*NNVxkBpASMOsg^u$ z6+9ZFe}eYBi%wsT6^dfVM=%?=I>B-kZBV~(51!36L95RLFwsr|#Y#(H#fECQe(WmP zZ5l2*{4Sn(pmUf=C6(j7l3mD?i}3i$V$o?XmdJR1geB75WNGvgST*Y)Jijy+7ibU0 z+uoxD=A@^_z7rO=$mEH>z@aBi-J992Z_| z!trZ`^UJW4WaHhX&{7c%@?J3{6iyR^loB$cIU9!ldhh@=*0)>5~v@en5L`_2hGrSOGONOy0*m!HOB=2dIhK0X)P z9@IeTOJJ=Twym4s}<)Fw9M zOW|>s0gCuWP;)*6ayEZqedaLqO9H3UG*=(}WlR0St$3t_HK1mp>>D*MGf@a9z!^b0WG1rDNLgM3kz!30vNO1G9cBR2RCz{-0yuY|#-&TzwfDF4Tdcsj|?ixk&DA zk%APfhk}=o6ZTjYSP)f*!F3TZbA>e|T&W{CRS$W)$uL4Li(u#%Qhxh1*)rz>$sfHK znCuB;YhZ$t%Pn^pkZ@p}Ylbe0)3CNC8X`&?nUzUKP7`wfIL5Ep=5Sn+Mc!D&kY{F5 zWTnhEXkLF8F5hZ_DeorG1Er?)2Wi3&$vbhk;L*b z`J76q+^-6Ojy9wt&Yel(7KqZ1C6Yhav|y~#M0j&+8xyjjKvdLMB2r$Kgqaoj7`W^N zo-Xf#A>K`3W^)0QcAtZubA`lxdobyGvld-STyV}3f%|$}O!AjsWHhAJ$m5yeIP}j> zwAp5cZr(!fPL;r^vt)0-;bY(j>}{+Z{=nY2Rm;JVJY_9 zGs4G-B_#fj40MexCO3vBL+i*@ptjT%o*k28e(4!9{(B-sXa4PlMt4tG?PvfWLcGX> z&sAjWJ2x=y9gHj8CgR8=JJ8-o7MHy~23@&I(AiArp?kLUmyI=L-Pe$)%RP+8@hZln zEEKdKErgTH_P`H;jXG+=MjRD85d&{)2|m0(By2z$LWC^9m&=5v>|R6{u(pst`7im` zeToEF^pj&+*+etPi{zadhV2hzg}Kf{Xy331Qg=qfO058pi&8~H-)dMZ9|D0H!?ARk z8&)3+$I&hO@M&HN4xafC5AIqHuQ?BhiZTZA_+#jQyA{8v{zi+`Nodh!iIX?9!LLp` z&>Ch+BA&%MMZA~Ar}DY5N>2o5jy@Kp`F;`Yn{=7HEz^el#qXKNmd#G1?6zU`WOq!+ zHAhjs6&`Vofy7^NR_3N>>s;+ULf)YAJ(?ih%PhOZY{AC&?DZ|Y#w_QSAHI$F@0%!2Xj z&XG~6ZB8|LADEsiHDr^c2^`*SNuD%4W==vmTI^55rg`D`<3}32es>ujFM9?9rrucC zEsM`~gbUYUHH4X>l8YV?HTZFsvH)zIAXrW zGm+y$53=#04FrCUgVAGOO7}=K0glj9Q-*xqU|;l52NEN0KUfEEtBKT|+U#FcDP77eG`< zA5`TT;)NAL-f6fd4)y8AXNzy*$eCLW)ZMd8RlN8!Y%_hi<# zrNCL*5$V%kaI}{q+J`nNL86@&NoUFbl>n{zh`X4KN;hR*rX1Gt85N zdJvEp23t2BB#*z(CiRL5MB3m437p_c-hP`YJhx&|dY&zsBpwjlj1ys+(6j1&tO_#^ z>yuk%N0@hPDUs8UB%^Iw7z@SgqW-k8Bzo|9=HuWQWP`mUHXq)Jx7*6G#bq5_+SUf) z;-%;mx}9lk>0s;@Sb!AqCk?W;jN6H9k^2}kGGaP`)^Q2ob8P`in+Z%+gKpAx`aM1~ zEWzVkG#XD>3(cw9$bY3%n3K2LnCCa=kgM}Opmw<<6C^l*F7>ReqBA75Yh6I*8 zA&vE0Ayha!&Xm7Qh8Ojaj_c86(l#v^bZ9RaX^(-ylhh!>_!6vDv4_AV8p2s2lC-lD zBC|RitR76jMpG-yG8_t|!Y_0cpY z+pmZ@Cq7E7oCjc3(>CmXQ;3EOvq<#ILF7q<6XD)$2Gzy`uv7CT)N$8{@ywgVvPfQV ze#C$qiGXTE^qh z{~e&$ySXUC--pLaU!g3PgDtTdSn_)?W~3-#y{*t0%J~GA&DyxIQDAUykp}%Kzevus z6ZlWdA1xKkgxQ1^77q%ALm#e@`%_~f)AbhkZpnjVDyQMX4+^L6+LFi-!Dt;}f&bP% z6T0#45F(ob>mJMF;l3VNEV>GD0TsX^&(E@nOb3_YvVuqVXR>G5;I$kkk7O81q( zKoSQLGgMJ&xT@eG9gahONbou*?8_(Iz{o6VI#T91=`GQO_eT#wv-u&%FEw2b^Q@Xh z%I^)p*W4Qt)~CVQW?OvdmyAQ+wcx|2T5#0iE9n`(21by%#P&-m2~+<{q-Sv zE8K*2zZR0Atv87AmCf*d>`BJ)49lc%y-4U^4R(z9 z0vdTHIBDk+^3Y3624@_C7g5C`m75A=pZOqI9Pp6jo+mIfs1U|1muEg-31?1M*OQ#> zy|6j-891H$3oT5Au%@|6b`ohgzO|FdF0CW0TQ@>j>snY#XThtpb3lHwJ~gj(qQ@V) z(i_35qGPhM4reo-ivB*6!UyvvzcCh4y#(ra09*G%bbq?7NNP{E zNHK9gnY5oLgR?#o-3L=)g1!%^7z)?si8t(>dLDFE3;p+1Fj3i! zWRuBFIKO`@^V%bo7(H4IX0vX<*nAaSykZiL{;tCKUO4B3(dU>gZN~IPtRcNzIh0v^_129FR?dE<+Wx>o38+eIw}DmG8-0t3vY8 zIE|Qm9xlxC>tUbgU@Yn$0D(@83bM6WD1VIkGwvxF=<9^+z<4mZt|FWPZ$sx>(Tu~- zYSOmqBCNd-28toapy#D2zGh#*I^$o&ve%WIH(3Oul!QEUaTd8R6AO>$H&6fGibq`6; zPy@0qWiaC^%7@d7lwi5g`8uP2lU#nU1l>Q(K{`T~E}i@WYei+azOf2Kb)m3-)Lih` zW{9~;d*R#Ftt6EWz)N2#)XT1b^Eix#PF|VYqaHG zIV}s5dWGyjOdgEBlMex9N?2k26wWLT1<&ERc=zlp(0FPLXFYvjwyrBY5SZCN+w-u{ zJ{O-23dOmf-w|ciljPc(Ii#L1Aya%u5!Hc>-cSC@*{I{ z*6jww)@4AgMIOAV+)Ad%trO|}*ze@T9|oTr&Oj=SAVM;mJYA^-3mjyCRII_Z{2V-H zwF;AO9ESx4jS!o50OBSOq5HHn==ZIf^zz4icr;xd7X9-kC%2m85Um~v_C6}y%R{hf zwzV)HcngtNUO@38ZP@ncEeTw83+)o0;<$8q`t;m7T&$vsCRgV}*tB~>_q`RG`lG-p z?FQ6r?twXsF;0;_2S*m%CRK*Apf7ZCHfgFO-PsE{i>BkZr#rD==3e|~d=5tCwt&(_ zC+zhZLNe<7$kXL9WOj=?{O2jmXJ;hgoby3A&vO;(>pXy{JP#7LP>R_Z$~%QEOoGjx z56GqeS{Y3LMT{IylB=&4Fnu{fu2NqCQiVL|qxLx5YT|=ocIG(!>Nu1)nGRV;Cy8z? ze@0r5<`6M6loacXWwvS$VOC_9F|sx#kUjq%hz|`wZvHpCHAs~{6V{BpZ3uSg9Kj`l zhp;2w37Y3lfFX@NB(vBVu5TCuUyP7^Z7(D@eD9F0nR;Y`xgus}Cc+MZMII!tN>@z( zg~#vo#HCC`HI zGYDM0d>gLz_rnz_8C)jtfHe~$$oo@KB)TC-RA-hedLVRg50ASH(dTnuSw%Utf2oF_ z=Q-#a6$3u~N3o<+853Wdz)IE_ekb*l5h;I(jOKN8UvwQMTh8I~I|4B}{|7l0_m6b- zekZNM{(XIQIx*51h0E@mqPpyO%=+?+NY^}NEW*b#wdL<&Uj2Qzp>-Qd?kU1s&sReK zrJA`j{VDO>k|tyW-;)&IEzo5d1yh@XVRDY(20f4quV!t74X+d+_rWh>|J#o|d9;8g zUYSgjM_bb)KKCJTZ3;YI?hD^mWx>>Q*Fbtu4V>GbiG8I7D3f^>GX;m$+EZ$R-!YS! z_+l59oS%jlw+Xr9y)Ph%(Z=k-4RB1j&O-`X!78~N^!;V9f4~YY?YE*sNI2S7k3u`< z9t2qagnLagXb>;#sd`MY%U~SdJv|@Y)<1`o%E91@8qjOC7|smLgCA;Pa7xUBT5=s2 znTeq|D-&`Ze!y#E6U=llflCjYnR)TY?Z-D&LsNJu{84Rz!0rc3qe=>yoS99y!PT(n z&@tEOtw-7dx_E1FTd z(hHj=>S69GQ+zT&=x7&~9xr#LVlEApFC52fA*Ycq`447$R)d|l1okT>;h779FCjDz zdEI;{c#;DLvvc8osx-va|0OV}BD68oZ^n2I9@Q0s;j+z9{W?v9GPnY0h**|P7{(>&S3bbv~XOj2X9x?@g z!#87F>Q`k?9aARJ!Ncw0_=-r_SaS?IC0USRavFY~IuGBMF2tuPL3pn>2tOTJN|v_g z5j9B9iW98d6}wng~!L2Oz)T zAeb|z-w&`};vY6ZMgLq(Wq@SxLVIvod=H!H}EjZg4~%xxTT;uOl-+=1!_GnBJV z!rvZZ98pYfZGXH*{ zH>;bBm}&$a6Ex9Kn0*|(DEQ&iq(NJhOoAiI$*mu*c*D&L`(ABFyO$#%!9$bT++o3x zOY&g(MH6;haDor>wqrva#ibi;aMG(hOcm|JUcqg`-pv)|83#N094r>OCrjhQ_qx~> zJ4s;b_A_DM&Xd{h@zA%;5pxrYu(H>hUV6vC>}D-kU}pkHw0@H}iKa|Z;A!ZvQGiop z^q}F9JrwA!2ffC4IPNZw4@!Q+^!Xd$SF{hLJh|zZu*G+Kb5j zxD4(0+wnmfk7tj{(ZEF-xa);0<{$Jz-{b4hc||0~S*Ky^=2}?t<`10gwm~fKBV}2t z%4=?8D$pr(;>WL-2A$+c zoMjb@J8q`oir-TB&C&|b2sGOtcZ;F$>3N`YPr{HCilZAuSkw`X`om_xf|tW#0y_=b z#KnR~p%fyzvjwJy;Fk#vg4KQ@@M_&M2%q%;au;s``P(b-&=raqb}E=MKZgm{S0?jj zDZ|1=*`VH$2j|`6;NUQO+PImeX=erh#q}KA_1|f%8Fv-GcIc3P8!r-W9#7urhcO@D zhBE1K#-!|;H6B4d;n`$^C4Z%9{uWIdq&$lDH+bN^`vEvGAo#?dxxquvwNR_R8{+Ru z@a(gz7#>uKOIg9wu)+&g4Ha_2&tF4;p8`5uH^6lxYVp;`8+hg3Q^B*c0f(#@i6>nTlRgtLYjMUF&IKZO|H+=17AhvBesVR%_%C0Smzo-8RSATnJ)!SIPZ7QdN|k~V$Z zIY4m8UW(E){ZT<+{U1HD9vx>a!28DD_|0%G;)5xeI!pkQ>5Ru2a)ofCI}*&6)PVCH z2K~nNfWszPjQLPaf`uOL#OG2VdFzHot0K@^u>^l!Uk2Np#gJ$siyPM2)7`cnRI6qU zy&`>|5g^*kq^s)WLUT3_-FyLU1s_ha%mLA_!bhUK15qL_ClIt>-GMS=eeCXA2tCyy zpcj)5oC+ZlZfryhJns@cH(Vk zfL=-?UE^3Yz@;z7JmM~k` zoOlNe9%X_^n}?bR0)4@%@T+kZ1Sy^oSe6Q~QE$KCLOV=*Ei>pd*Cc9j%@@)y5Ma(M zhB2%)ep@&me=eGaE+h|rkL&{dLMtp+Da7Olx6vxZh`Kj-L)Tqd%$jJ06<^Yj|9Kp* zEG@>^bS?TP(25?>qBLk)16ks10|}4)Ab+`;(2qBx3(<-`%D%`f{Fg`^_Y0r#cTMnh zeiI}tx&qQN#W3UX52CEjlXGQ%nYj-*@+9%PX!g3H0)si6dA(QH@wUTM8Z~qgRdHTN zmy5o^*>#^t(ISZ`Qay+GKMiBb|BhjzsvQZp_%~y*bOsp_k&CPAA~CmUCU&H<)WO`I z>JJgoi+TI7xG4%>k4wfIn{pVKokq|BJe@*7Qky6-LYC;!sRO%?(qakgFl;b^n;Z^GnE{f8k_8zldC~ zh=G>3tD$D)Mj(d=5#!zjrfSPy=J4}h!r7z-&EI1~pLuwbHn|IAXqi4aUYv=>Ujp!E z-FSRFO$-4iu0dqMR|rv=4A*}hC3jEjLQ#D@_LF2Bu_Qt8a;(I3;|+r0IRry~bP7z; zYxp=d4%PKzp)Se?N@gsF*g?I}r#b}nnmo~S{7jIPw-XaR2Ik%OWBS$MqQ!NnNwuU)ZR(TK$CZ{8)%AdZMfu?VO}la7={W4QJBMwnZ=%+uyZGFa zBf4i~q3S>`i8owIpS%yG?>29v1=ED{hKVkqbk!6xuHJ*r#&*bm&;}O!&SLqYWc2^A zA73lqMAxQj{3xua0tYnkmxeap+3So2d*`8c#S(nI&;yr5j;BAJhSRe(hSV|rIa%i^ z14jO#5a<+6uix5B-x_bD$~TvQ#rb7$ZP9hOyrhgscg$l9_ZTs=rz#8N5~+p%zpuoF z7IbKd7FF4p#ZT;uWq&RTr1>KTb8kF=|8(yJpIM1C@7N1m754$wF3@HJ?L6?N=z-Yk zU>5IJbCFd`DP^b4mU1?jbsxuE%i+X^y>#*Eo$ONm2bH-q?}=3gUi0RPJ^ZC83x3|P z-=uh!6W`2b@mJUH5N}_TQ@J{?;*%~n zabn9k)L>as)l8isW%l=TXrbLVE|1$q&2~m`qnJTuTUY8Um{E#X$&Gh@%|CD02w{Z)Ne6KMXNDXc-Z7k5$a ztoU|`GW()18^N}anzme|Q|2eIg?WbTP1ot{=i7%_NmmSQ(30Z)ZpE;A%VpV~FjJO| z@1cy54wgRZMenXA1`oM_=lr_VNd~B!iLcMD1KQ?O+gVS3jNl&D>1q z->a|seQzJ9yE2ro3yGmM!B5bqUXDFE>ekiYHvk|y{3qU9@_#Krjz z>=`R%wq$Dv8#A(myZ$bfD(_9D9tCpj-mYx+?VhXDCuBdj_vckEXKp~%H;?|Rq`Yga zr^^WENAVYVz0w!Fe(r7lOO^>2ueO8U5Br4a*P{6Seja>Dx(y#1|5*?(Y4CHI2L9mX zkJ#d^$;ybvaSB5Pdi9c3tfTx`c6*-=zj3A}|JYYuqV%zelY3gq`M*r&4Q@-Z^S_ou z%d@}Ym0=E&L*GYAKF>eRH>b4Fuine(-kmj^^@gGBrwg*I#-$|g!@8UN*y`c@m)uBx zM%zts{>%tk9qdTweZ9cDO;wkOI;15%!KLDXc1yaUTAhZT;b~xfYvuQ2gSlUN5jR%+OxO#kJpoUp{~gL{C-a}m(S;#$BM)u+Wy?)<1Mtd@)NxjvykrF zE8&x$M)L^^2a8utbKp*myDO-@uF|3!oeN%&Qiq7}I3WOzua^0jel;S<{l7XnB({-_UZ4#@z2=UwPc+uIqi}-bvl$ zzB;X>JFK2nz4XdO>mm0!?JLzZeAG3nWq5-3npn#(Ov~cU`Ua?7VF3;CZ)6vrEayvw z(3Fvy1v|3uBKLQ59PeK4!3_>v#(pu*<;#|)ph-+NouweSf&?DU&FbGqbn|oA=_JJ7N5St8v`-gN3j&-~ji2*9)<9mK?j`_yg=7 zHH>S{Y(_7I!<=J$0j~U$0zE~-dBH~gsu{B*xRgUy&Wjz?oyQ#yU{h`9vc8c`^t5sd zZLJLA!us~{t$Sj4ZI3Hrk*Oyy7u?Ua?wih*E=;1PcM7@RZ5uhqtj*lrtYfsZai}=3 zU5oo>vXD;nyUqRct>e~sRHK&n7&^Q7FKijPlM6LEz-_*B2+q0svoijnT>9%H;zJ38 z*+2V@c)5}HIfvda;*VEdSWmxD&L%ws*S{LfZ+KnG$xJEWhAWI<*ZqlNXJ;K^`&RT+ zbicSLzVf`Ad1TRq8+Rg#vylA6-MS6j`B#nHdee>gI_HejuKJVcxmTB7Il~}4R+(c4O_(H%HCI*=y$Qsdlj6y;Nxuem*IG-EuIT?c_=xLJm+UOAGkza^0IfW#B7cn`#y9r zl`OTE7-)Xrn}-+i3UNhz=qCs6QcpdX$BgEhuDs*EW&TI6e9)!FR+IQtdY2w%l3?v~>xHb_gtUyh{DgHIu~YNa0fVtO+|#PyGIrL&yu zdD+ru;?A^85L#H0dkayrJxW#bAon2VzTZNhgvD?*mOX;!j$%$oCFUQi=iDtjsw`X_ zFmqG{FH&&iC$)hjBybY9C`HVTJmSjw)ihS!m>33wmIi?0x@KO=Wk0{m&yAb5SImd5 zI>+<>hH+1J&)^KxPhsS*(KKp(6z|qOn!9rK4E20$EpbdVlej(Ako0~U!@ru~#V40X z@U|L_-1$P5yS2=OA2Pv)mwVyBE1V?UuBIn^#`gLA*FOivd$ezIX$$l4fcGN4yHZBd z@pwAF&i4`*zKe3FUK;Y!+amc*YeoF|McRC@;+LwP@Z)^1-z@&4;x+Nxp@;d5^G8tW zD{^z2r?9h*{iPb;R?uy02C>PLKGFYbintS55wtyU9bGdm0%LcoIxj3VaBdk&ou3>S z!p-p=BWav=QnK1*7W+qL729R>hx)xfOjmDU*kziT>=3U{95-6QIdEN!v!`DKn=yV2 ztNY(=c7s9{`=oN3^RRn0?0~Hur@kYO-n;*u-*V*;Z7VP0myf9A|IDvqbv)lv?~6}) zrtcx&l;+Kzx4O;VKF9H2LoV~~4N<(}@DwgwDwlR|?h-#K(_%9cw(++*Gx_1!JfE7| z%8#h?<~M9E=6$Zz^3SG7c!P!zR`Zkzzq;)muF$U+Up%bDZBIIfuc{n*rN^E8zNun< z#1hIqmx`l&*EBxew2z+MrtciDZs|PV!kB;IXUIz_y3ueAW9PN```KAT>si^^$MJKJ zA$Pd=5I<7-I$yh_h1a(=muz4Do^wb)&*xp5$lrOfn0Jbsz|UNy%&UeH$=xYllBN%< zCF32J@FS*P;jb;0lJxHr{vVX?@O>*dKDyUh(t7Fv-)r)W4|OX-UozbJkD3g3G^&T^ zate9JB01X1PUHhzjQF~rG5p?29d?I4(y0a|?2_`q&c(Ytou3ykcIJ~WviHnaQkU)7 z?748ktGM2mMy?BC70>fD*sy~gRX^7GQ~PMP)me+3A~vBv%wLFm{*30=2PDx)ze8yv z6m$Fbn()c0M*M~G!|2RSs`QHXUG`dKGy8<3vp0uSvGS`k1isL9TE1P1#WG(u+|QQ1 zbiAIwbW2g9F-}WT$$9g=Uy}H{{j)eT&9~f5k0`#*XE`r*uo4ZAcZhE`U8P&po^yt| zuf@yOYjGc4*6`|IOZZK{a>Thm|8N<}(VStOrm(I}5+`PwVx!GmwtZ(M@0y*!%S=t= z3l6%`JB49%(YWC>vS$`Ud1?1w*xU1#sOo8PzqI~|P0HiB6W#0i&oO^_)$|8^q>7XzV(M*RteeGy zJA=6oE02j+i5xhK_J{Z?(3-#8{E#!$6u5Xv_gT5XDAr59mwVc+&9z>1;-YToi_eX{ z%_)uD%N<>`f`1h?f?o=IdBxAmdEFhexJTAL{NZI`e7fE=US7q4Ki+qY?-_HyDs0Sq zZc&m2ziQ%N3^PlG=D)Vw?g_`l(`Hew{_I?SY>P5?aI=c|hR+SI@$f#b=luh2Ucpe- zeQOT8U0ln?2gtB1jo)*RLpN|K!ZWjSVl%SYwOp&_ZEpI;qg-R64Ide&V@054g#V=eW>Tca}>nQsWn|)fdl!l&W`@I-E;J9Y4sj znYXrV=kLN84QC^j|;zVhcZFQxhMba+x>HXyfjzD&ucQKI54>OUb6HMUveg z*GVcO_w#+)lljbwas1P)GVw%BYgTHFF@3UrsHC$)SF$a{OtP=ZM#7$};zK7{^G?}? zblx;q^pA+FSah|R-&j6`Hx1m*aV4X8kF=qDT)SX3Jhhu|e`L-tuiVYg>wd*YE(qYW zR(|BBZF8hU)kv(?RF-^x4U1&tP*L8k>wIJ1TiX>jBwjZf#!hemSwcVF@~@$>mT+6Mf) zkScEH2@ihX{3uXm)#uY~)cI-ajU^E~zR3tMDE%<<#!Ms!LM&Y?QonPtk zh+D1xq9RQr3_7;m=YDh_;v~u6#Fth%RAufM0bk!bv5U_rid8x;i%q?A#Tn=2`KOy* zx!=kY*~;mIxysiU>Ex5A=<8j-Y0l#ajyq8z-q$8gbsD~lOV69K>zbb8-syvA@#o9z z3b$qKz9na==>ovG^F|fl^L>%JsUKH-5=}CPUlZT(&1bfX4_fmf7G4- zuWBy4RLh)JRQ_T^)=4`X8HckGrTS*py^*7XhvksX5jJdkzV0^C=M7OsRI_l7V zT5p?3olf*nYk?v3+5aBB8kIs_9?YOxcXwAkyg7oFC`qVQ{b=@g0?Q^_N-$=x4s}!- z$DSLeh0+feu-9Cz*_kSf@t1ijmo_?so*Gw#eP3NzpD|uERVE!$Ul_8dUi@Zl9RpeW z4=rp-yEoez$kVmSGg$G!L0T^F&KLU#%#-D=?9nk5R5MtC*E#c2Y-X^GzMV6cx0Q6@YSY!cZoO=H=_>8-Tg-+>)^uY#U@ogs#CG29`E0uQEo zsH5D*`MV57_httg*~g0)$A{9Kl?hb0-kV+H6+z#|>JXlrAi0wkAkqDsBMGc0UEx|;q-rl2sXmO2`(Vv(lPcvN zTgyw%liz$${$z=w6!Inu3}_Nh`S4!CpR(wpCWbNMj6t8-n#GnZpEsG?_T;?t{I>_3 zzZGaZ$NJ>b{U@fdr;QW2meRFsLunoL)QaFP)Py_ZOLynxHcrl)4oOQQ2R?GEg|pk3 zFKW(76`$FNoym09;#h9h3u9QI6N%?6fc+ek&)uD~hS@T4BEL)dJzwrRN-{(KG5@?f zol_VeN9V?e)9jvbb~g$=74LL@!k9mNP2OZlWWT9oQ|S;%_Q}V5hrfz*n&lIA+`9cZ z;isJQcuqu<9m{FTgfw>h26sB0U(H=aZ&vQ=0d8NF24}f>CN0sc<~kFPQ?YpjyYJm< zcI4|b;=x-vcF9Q<_T}7{th=fbe{yOmm*i8+CHDn5H=bGM{M5_E*=X1SHsqH!pV9Gy z7ayA~@ylW)W#caKU-?viRf6O)vOalD{8=obKn;8AK}@mkudkG7HwajkHKb)%w?QNYRnRLChsOY`!0C$ zJgLF$3=QGD^e0l~m;j{H8hE2a2KD!9!T~K~ zcxg8ber}Gzp)XzOj>S22np(>aX{pgvc~2@Bzl*#-KEnm0>&r%*A4;b-FD1VqKk~Hs ziHoBSa?^*sWLB??xUhiHoQ`uS3+!LUU0D@~fwYlyzqFCwJO|n|&H}SK_ zQd^zvPw<;W2iw>q%TyB>qwQRmu6-Zkl&8se&%>< z<5zrh+MK4ISwU&q3gVLmVk(aH=1*H*!sNAuyyc!A(M&l5%IF{-Jd8Q~p2)E{uY}ZMm$< zqLjJ+KF!UUZw3?9lOSkH3RbrUV3+v`d@{6_-PzZ`A`Qy9x>f@Uo%R%U51nSs)e`N) z+`ZH`xE9Y(4`jv$j#NFwnkF`{;ce!Z;F1N+EFjp5w)NTJswZPv!1N=atuuwK8Fhgz zxgW)Ltd#hFYR|^-wew+Cb(82@tTgHuR-oInN^a)RBzE`VG3?coo-P?C;~L`0S+`_0Q};!y!W;s_H%Vup+*}cPp&w&I7GW!EAD2E_!YK!7EIk0s&=D zLG!Nz4Q{j~*^a67DJT@*YFg7T2O;T{UBssQ{&@VpLz(ei7xG`Jgo*#=vG!Fz7r;bh(*!3R{OBSM)-^>rTU+2OUqW2<2nhUql{xP>6x>|65*@b| zFym$`)*he1nuhy9`qR^(XLS&MMszacN6X;oKnu9guL9RAbz$MjO@jZH-2mIpH1^>B zYCIE_3DH4uOnHtvIY00r_4ZJK^~HOFJ0I@~zRb*Iuh<^I6Ee zI|u?I=iwRiu`oJBqKls_%f5=cS*S<0pvYMPU;UVi3#U)Vv%SyD0>8}1-4iPWjRTHj z+qaV#T{i?haUC~y)d2CDhEsGZxdp?{ALldVGWiv49-L$JG%C+tC{7Jh7H5qa4%5^o z15I1Oil2w^hqf=F%Y!t1Y5MP31uJC+?nD`6WY2-jU{TtVWWjCANOe~4!kFW`}?G6UfTuMXsrU1 z)jzV0%f9oSb&+CA)J_66kS=CE_Ut0;rt#;qwY5w*9JvgEie?0DrO zNX2)sa?@gpyqkhi8%CBLSzkfv52vy(Ub2v{97WUS#*krDG_Pi@Bz~d!p5DAkp+$~2 zI5!O?n57@V26Q+vhhleH@v)4euL(i=(^Zy~yh@Uz@CB1bN&2QfqxcnTHlR;-Bwn~w z#oc)%4gV(jQeoOLlV%B3@ z&RT-6kWp{6sC4_^Tv)D)#5?)& zSZ1#Eh_=^$CR64{BkffrJy$0zyfGNXI38bh4u$7Wlh{I|MDiQ^o&J0Af<6h&aJIfL z!$oCmXXFUsohD7;{Dg;~d}<$=Nte+g>&-O%zjh`L-^Fg;)MA!x2VvXnfp9+OB%Jej z0MFFyp!3ff*nZiU|8MXT-sR^cPOYJXYje)RlD+a=OQ{tEISpVY>pR#$*SDPi`W@gI zB@e%UDGT@d*um>*E^OJ~k)&8;Nn6}T(#;X0!RB%v&X%4|Y)CN{YXmaAqZg4*t>jYb zEUkasIM0XuwWGm>n&OfmCO($^n2Y&Y%8iCX(YJ-$Xk5M?&4}4cZ+yo>>IzSo>7EBq zuJxgW4xwrGYJ%ChFW4SM1NwWVOz>EFFwK;Himm%h`IHO)IIWOVh-u5nK|_`Fq_wdU z7LfnD%hX|?LU(RBQN3g~xKH{t>F+cW&kePJH?HsS@_@T6<@;s2V)ckTcS$k+&pFss zR>!P5;<-zUF3{qJCDhOwg+pYb_}`(|%4QY}VTFgT*iH{!!?t~x$EANh4S5?HSXq!Y zn;2^fjfT(pzl(HP->nQ5eOi~b&N{-Jc6M>w-*xiZGmc}_;YIlRDTu5`{@`zq7(!2; z1n`Bo&0wc{BIwPKhP@IWsR?IKW9P9HY;+IBodq78{V^kMX~#*1D+%5IyhS6|;rx~q zJ^B##9(VYT!1nYWR=ZFc<_yf>0()h_%sr9~+T6e!Zw_Y*Z69#AE{yXQ$$- z+32lf>BJZTJ&e@B+Vb%%L*I=1S5qh$yS|IL>kWpcVrf`#!3eM4QUQLlY8xO-73!Wp^9p~Hb z=|j*94%%8DuoOwEVEFoQ*y%K!iN?-lPKk&3UDYZ$cGEdrE_jGFR{o;N-!k~IZ}+i| zZ(G4ZRgEfwDoja{1zEFpcF9r;V%n3?zjz>brz{n}j0&d1yQ_KcWIJ|c$Xe#> zTF1oabRf^ng%%cTlhL9cthurmOy<0WxF26&srxo`4ZFdT-grfs zOo5%AzF4r``8>Z;Z#f0czKBw-Be8SmA6z%Niw&sK;cN4^<66~tuKK^5yie^z6d!P6 z_wMCz+InYk0{@DASJ4-am>LI}(*M|UMKzFqQNyRVwX^2lu`qwN8ayv?fGLahNYU2_Rj`nT)7Db8)?ArJEm9}@|hemZjsd8-L!M;cPJ133f{*H z*#L*5lAh->da7T^IxE}ZT2U8FS@x4fJW85v7^m&)Va1;M!L_B+md&tT};HJc^x zR-?(myUAWjkxqY!XKcwN_UG<3PN`)gQ~7uecRV{wj;@L{Rr(_a|1cq`>;BAi$zpK) zfFOTt0sWY5NW(V=V&U#$u66khMnkpXV@ed)-+vsGU%!F-Rv~bMX~L`1A^76QKUS)3 zPezX%$*%MkzM8%g?JIurkpbb{?pJ~Q&Y9z|ahf$n>C^R|?*aEmDv&u(NtU1#$eb%R;h z>_9HbOPXrNg;9^DF5jOTL`9=oaHyny_}#YwTz)t+FJ6Y*^<)=~*IPV9c zc>oPN;m7N!9pY*SrQ;UYCAc(6k)O6_tl(u+2p_53i^|~x_-V=S`O)3UR325v$=~WP zlVqWz%B()ln4`|qZ{g}hr`&lC9o&+IHu<~*?H`X*jvb^5-Ta>Es-x!>ee$_Bu?1xC#0 zydMO7lLz6QTyAb(GMhFl0-I-e3EGY}@}IPi!op7*U_y2PXkOcmN0>K%BTG)=ch`X> zj&E>SkQ7v%O9L6_T<{(jfPvS%1V!_=uo?w@W^!gB3$(w&+{0sVzeIGhzvfyQgj&L_ zoUd$Ze<-(2G8fTWty(ssPKg_O%>hUBPeqOAHaKOCDeJNFVS0+nxY6$(yV+_;0huV8 zp?d|ImUxkd|28anGL*(N1Y_OF#dz_EBCh`LSJ`BxA0oRkDwK7*0_SUZa^_nGVd?SD zY{_y1EPi>4^kkCQ+&g<=yObik;Pa8zTxccvtxj+t`4u*(BW~5N#9!7M+0UpqT&CX< z+$--40sG3qbde*uj6DvXQ{Tf?*)TZKxtHaJ|Az0zO5!gA#)!`X#;;y%!lvbRv+G_wzBv^`S91+$w3IS>+GWEX z_qS|S-5EY}&wXCi`2{oh^;U48;4i=D$1suZ&s09O)BvmxTG7+2Ds+=;5k1^>i!b`- zhNbHqU@G+C{_F;>ux}auT$2oTzZw5qWfeK~#4`7F_gKDc9hWCR5VfCG@Mj!a$a&Rm zjB_6fw=cVjH~LwNrR>V^l({C%_j`xtu#QG{UT0UwT`Tizt-y$QH>&sqJn5DxGEbTubf6^qL1`* zKpoZH4#zcb@3S?xLLqB@2SSKCy&El{SHd~sdKERXzHI?LJop&2Q|tJwFx zoZtU^Dq1y5?$0}=j8ob})|qE|)~`al{CR35M`AJYXQ(*T~EqR08J+{?dj zi(+o~IrtOVK&OVS6m}R~gk`USDSh_{=v~syn(vu`u2(OoJ++c;6Cr26=`3pL20`Y+ zdob814ximC5xrKt#U72dr?C@zc%Lb{;s?qDgmao2psZO|nEJ(#$Y>}t3SUj%pC_?~ zBd5V|ppI~?#WFtObsO99d_33uOPk-?|CjxcQo;v2;vwDET{zSu3D&<_$Lo^*|=?bB+pInjx@qCA81J$(olT)#RJe zAJ<@J(BT3*?&h+txLU#9|7?VUSAoKn$_Jpge-p0uUrs-qi&zMM1OI&p7MLq*}VlG zMr}bYmt*v;w-uG+gK>R{6Pafgk@vVCEL)Vt@gN^IfdarTV+^ssRHXh z+~JIjZ(#r5Lge|;+~A8=@IkbQPfV^5+@E)j<)wWUy$qkiB|Iv}R|d93PYdYt0~|N za&;*^joHVnX3H>x&G)I0T_nY8C+V--7yOp41J5o@=HfS=rj`_C8oMSLvwBAb*IRmI@%V8H^1*0H=gCz?cxU)?`I9DKGO|663$nto8U>t*l#|5l6ez0&h z8^#1z<;Y`YI`{C28*jaHA@?PG2$(+E!M2xu=Bi^(i`x(BiPoQ016X|9s1EU_G%`7 zyup>0ZyqCgBTS(7h&t{??ksLl;xhVv>o09NDW*UhX}a2^LhDCX;~=~JY=Xgj=3dal z1?co3FKq)EG3rooxSf5u?}jtRO-0w*nP_<4nnoTwC3qHU0_lHcaQ=tYd`PA(h0lq= zq}%`qD>y~{TVnWcZLgWxv;EB5palFwBOv92GOW=ur>C}CNI7pS)qlA`)!(z|R-+7g zf8WjyyXT^IqYXDmE&xt0o-2$t{|dsns^Sra7sV@O8@aMUPhf3H6HETCL0K^$F{;p# zzP0;s%GyKVtLYUe{PX}FKlOv(4iSm`7ZY23 z_swU(>hEH5zU&MKI`xEBY8&xIcricy<5rdy7|5oX<*>7>Pcr*YV`%V*x#S`n$4%Z~ z4A%>C$VerS1fF)F8+;Z%X zKwYdKY>v50lb7BjL8L7-r;FIhTL<8CN+R3*I33Nt#_|s~IHHx;Ko}Hv19U&}u=@E9 z@X0HKf6^he^&&@eD+bZ(w>Drf=oYW9Z(^gnY#Uop>A<&Z@51NHp0oJzZcwZ!hZ@fw za(Z));-20y&{{PgPOKYE7p4W0(q|hGEPsou)!vbI<_Q{{^Akr_KZBjIpWw$RY2oST zszRS;YpR>yf_>ixqt5EPoKJ|i&BQt@yk~tBVlSUTAH|VqJIDu)9GnFT$&Z+sz#Klz zSp|IhJ&3OAf&AO~%xKOM_GMTV^DU9Enq}``&0BRLKj$3W&vD@#Oy-hD{46>bJcPg4 zqy&~*U$XmV_gPX(FN&=62(~YWB^R%<%bqjD?q;<#W|}q&NKmC|8nfWW+K&<+>_Esj z%4UYYB|e?En<>fVF=Y?BD9Oplr`zrauy#T)o6!42Sj1`xM#5fHM$MNT}l4<;K>&8+uGA&y@*lb$Ce#busM&Od8^Yr znZu-};7ikZMs69%~vCGi-)pQCU{E&|? zJ%Dy~@pwF2n<_tCr;Xch(;`)Ca;)4!effd(=Z(H7;*K%N@3J6M^E`ZMnoiUF6~soW zQ!rgqM)KV+kex7re^gSBin0;niYQOBQ^7!|Yq_c`}cG4j6*#vlWnG__OOF{dFi+fi6Wo>o$*s`Y+ zxTCg*n5bwmYwXNoCpKEpx;Oq<1xL8g8H3n=a(ZZZ@h^Ot4FH%gC8d?N7KwN&huUCjd5slemQx!kcK zrR;aJvE_YJ*#X_5l`-17{q9hRU~qXAhqN{N>Z9LS|_ z&xKb*OzE$~FKz{GXJ_}##7Bwmf$tw7)Ea+R;9A+p758qkY2S1fV%?v?j?M(wG9ZTY zb6pR?M)lCp^+TX)H)MsaLucRcu-`M|f=iEMo6B-U{=05-##p;O63{~LszuTY>2QfG-z9*zZwg-;;fcAIuJI1uy-f9n8^m8O zgtVU~)N@ckqb|3wSvO?~ozJmuCvVE=-^#`vtl&p$OybPP-WTBQ1!!(DgGE*ig{yBu zVa4o3$U1osU(WCX=UYpd{;gYV-|c$XC_59vBd!Z}nM8B%a#C3L*$@a<-H(}VX`+x! z2biBVgo_WpAnMXO$gN`jARp=iK5o7+Aw$CFFI+wU;lro6-N&PMEud?4Rg`_ z&>!Y_I1s1YSkBa@ui~C-A7U9w$myP+2q{ws!`03@(TAevtny4JXEi95y$W?_C-d~U zh*CLd`O*uCUh={(vIFR6{bAgp5swPIG#uTr4py0p*xUkt!I;!_Tv^pzm}93(@6{7X zNLU``$y3dp{fHzZeQJ8zQIO7CfF)NFRS{lZHwjEts$h+M-*a$X=P;j$7l^)L~3I zxDezWgqtJKY0%rAcneF99Yo_(_cUkM%ELPNCP5*5up>&P)RCnk$ zt#8VrYnx_?gI3+Z`W6X`+5L(gcU*=id0eMfAOs zqn8SwG*jT_J!ENej_lRy@QlMT`aEo$ILoM&9s~>)FKu0gS{6O{$K)F--OZ%&ncvA< zPEVZcDI-3$ZVTN?Iz%S^uSo99O=#$|fNgauWO=oge*RfV1(q(fVUIfBbgf;Ws6T=J z-Izot^{c7V=QXWZd6tC)eW90p4_%YJM!)oS@H5Bjvf?q*#m5E(iVGtTi@$GpD0!6+ zqOL3#Xk4X1pUUb*oA#uly=Xrcef&<3>!ie+RxhUfZPRIB%xH?cG@j&YQz?E#1r5I% zD6S8U5DOQ*p}+HH)A6Rqs5rxnOv|OH+999+(0G_1c6E@rbH^4|c;yzPa>f$I*a26L zddXZp^YFZ_yGZBT4nD|R7lTbY-#4Qlcc0E4KB+hQ^L@6N>NKiMQM4QX+8|bX2r5!qt(T7 zN?Kx>rhL*&{6w8~x#ZWB!fQ+Z8^en&$@=C1u}yP5sYnvC^T)rU#k0dmGXLbJ4o{~e zy=~m?8LeEF`3c%;UJ1Faqa-Bp6>jG9YrOl+1ndoJK|{w+psl0~(ide|@QDR@b)7Sb zRwbglqYYf2xg6fTIm5mtTEYbFaAsX^&IRU#z;_dUHfyG2UwuwR&!55gC^(&W80ADR zj%R4X$hmZ1?J%Fwq(ODZi)nqwKGt$@5ss9;j9!-~3%^*Ez!-&5@aS&}TwOK}vh*jD zerpn55jEq)_I}hYO$RN%o#Z_`j=X%!P%R*sod~jIcc#mV*O`nK|2*s{HWe-x9c&yg zc=lA4J2AisCM&(e-`x?c`QTqXUf6}Z$_nXp?G*lP@=K;TVGPCl3E)7>F}RmF8l=>3 z+8zU0nj#y@xE)e3Xrh3o52?f9UAvjh%w{lU+_xn=bw%nM97h35ydd;V z5bphzO}{Qy(9v7U>_H-2GbV4uW2l#J@5pbReLGI!j?t^q>}<|!&h5IQqNb( zTFJxkZ%7m~u$_ym*VKS&4B}!{q>Q$aT*WdYNPhf+=}c|~%@rl&ese$WTyce~K5+t+ z#!rCS6SFa|IvDmA){3g5OX$jDFF1ehG`#fehny|S!if)!g?+C!2=TOn@YJx8LY zLcgKD=sfN22`58B(9k+wMOY<_Mgfw}={|R$=?T6T)ZaJB9oFRtn|2n%Qd`FX(*t z0m8kT`8gF!*dxnzY~9YCkhv%n#%3Bqd-`3lmGmdl0;*YaM+C+{Imkcvn8k67qPWZ ze{q4W48K-?;{TXWV~4)=GL7}8p#IHWHZaZsa`bJXNa_-P_%NAz?rD+7q9ZWeDhkF2 zyy3!RlAxeJ2tz`OYViq%W z4|}w28t9F0VOy&6B+Q5-^?i~8^M`vxwjt3hBsdx_zMl*GlFf1QQd!{z|9Hp<7z8gp z=D@#%7zoSJhU1Zbz%3KQ?171(+Zx7=);_})_gI6#CsM*9IDqSwOgMEUihr1VRAjl; zg6__ihl!5^V1IcKtqCyX&+p1*y7lEYDw%^QZE6)BTX&Ik0u=>!tfFmS=|1No_3TLX z-Zfs4OsQa_j_999nILIn4)bi4)W*rz@Z6I|EVjJ|d1D-eF4a+R#gy^Be=fp@I|%@T z!$fPwO0mP^^_csZI&O#f0;fKD96NDQmHV_@4X;bGLFYBv(0yp7(y1Ea4k=%LU2qYY zhD{V2Xsv=y_XW^wxsZxWMf}vCJNP62rAqX}$N9nHEo}b4KU|B(1U#aZ!t>>$`IN9? z<`c%?%ZuCKI(Ho&U(zc2@F89l@^qL(SWG8wX4J@56VN{j3t8?T}CpO@(`Q4dibp zq_KTrm)Q74I;eZsS>(Dj+$PQ90rNH$^2bdD5b98E8z^&^O^Qr}$+(vV+RCvzt9+r| zIf8clGNYwctNG;~BT40@F>V`dk1u~E($S~>_;bQ0IJMgrSJ?rMxh1Iqm-OMH#ct@Z zKY~>9N26DV4F9uxjmYr<$2*;rr5y@UwB)})jM^^CulgCq%HJM@j{(-gXE9bR>g@+U z^7(AREr&)7u-h$&cNoO?AF!jQt`M9vt&HYm4HB2Uv8OY=KWX`_9O~_B=A=y{AS_kL zMhk_wwP3zvzBC(tRPATli+h;s2@@#q;7HqB7cbp+#|7K$_?KdUZHk@{mza;!j`Xli zGIG#!>o!yEJB2P^1+=czfo8AuBp- z&<$gC>uovbIy^gN1V%_#3T!zOnts}nMISZhG8Tl}rtXjjgPVR#<%Nhnk|ySO))-Eh z72sL}1sK_(g}KqAp}$6lwf7%myIzGe$ABpkuf;H&IBpmz{!*f6?lv@KsRC}zTE;h> z2uG*wI>c>t;~tFNh*e*sNOONKHP1=I=`Yfm!{0O<9%06wTo?-1GeVhdUnt)x(OWDW zD1a{bLNG;3^wd`reQEm`Dg4!NiH@DjqGYGJe5CXsY8U9y8!PJ9xj%ua_a zV9`%^a)aYd;X|i41ZF!)xXnvk%XC$ksushH94tsMSpn{87~sUL1R9&MpUSuCp?le3 zsC%;k3PYS=XqXh7P=CavbzZaV&~Q#uLyD6A-sj3Tg_FwtO*HrlCW2 z9O|zPK+*O6)V)vtz z{F#5>coE*0$AHEC%@CBRibvmy_+bWxrEtqYG^S}Tjj@PhN0NsLs}-bRQ(Fhi{yqja zW=(^RyfoG>JBsV($HLOQhirzX6h2_#B8=*ILm)C+bdfv zn9kMPPVU;un~%NA<0~hsJ}%J*<=)3c=V7>6upTej&&7_H-F#|IHYMx3(YUracIo<0 zZd=1~$m!q1C)zo1fkWPNC%@J6+nF&Nl^n)mPPB2QZLRowgdJ@d+AZ;wxhz<|<`d5H z=)<3>11VPSH~W5mF3VhQAkr|j;)`XQu~VX%xDZzdD=LH8rqCH|>PJ&}dr}YjEXyJG zZ5z{=_(jrhx=veiGQivU4Ld9R&f<32;W0&X8Xn_Fsps|S-eY63e?MG^3Kw8_kT2VN zT%tFA>IB6X7!LX}2JF}8GLIux%y!gh7L$vDFMAKum9RH3?1QXOw6z?J+)q&Gz7|$_ zeIv}VFc9Vx{$;z2LfGfXORQ%40noF23^I*UbouvH{$cD4_!WB>O2VhXuj98^W63ef ztDePTuX#hF#&Y)KfDH`qxxyEkXyU}8d4La`Va(9kXtL3qL>>Rp$<*zvbnHkpo3W7H zd7_4s7wS>#LOtQI$Y5x64})cCA#gv-SW;1(0fXU>S-aULiq3cp+yB*&M}mnsqT&uk z*mT2{f>g*KRR@|=l9)u~$o~Fu2K&0TF!g;2z8cq$l?}dRQ-6oo+YkvVZ|dPx`x2^6 zJ1isnI8oZQ+-P)@nS+N@@P;fe@2ZE7su-20p58&$yP z{&`>^JA?GxR@1w4d9?T3Wont6BH_Muh#hQaULg#N1nh()mQdO7Tm7y*}S46$m*BfK%%7q;y(5SOQvG118qknSYbacM0q^#99p7fhjz z)_Z95Y#mbD6$K#;))2JgJmhrUW<4?PU>qx=$!B9Z^|W8My#t3c)p~^Zjj@naPz(ba zR3PTox-!#Yp|sm#9;;szg`Y3<*~DNb)C@fbMNS&Rvy&Jsgm`9|Ucsb>#WEd_bGUmn zaWVIu>4@5L)ZL{|uQ!}xf8K}Fwm=K~Xa0$E-f2WCUZoU!S0D01w7HmolTfL%kd3T$ zh4B$H&`(R7e|~HRHgqe{v33*op?7U>pl2tWW7D$s!m=nA9UWznG+^PqKs>Dx!%a($LX%6}<_UZfuDMi{dZ#NbB zGaA!1lQbqKLjBc2!kQ_Gs5fpeh8Ro1hgV8;-`xdAzH7&cyQ^)z_6m4)rBB@9>==n* z<9XI(xE2nMUdq?J8HQO`_enHGr*OigOpFpI;*Y%{ z+5zBN5P+}R&A{3?mmRU(4~5kNSTkfSH1|(|=MrXhk?KZp>srrL%!BZ!csK7DC;*3V zm)Wl~i}0;Y2=DKGoQZ~ok(eA z34ZC`g)1&qu{lpl$@8-YseLu3^-0-f_a&Um$+m%_^i|m zU(Sb0>nzs1_B5;7xrrTm?grKCgmB`M5@bvju+zb_sjTJzN%<=P$5%3+0&n*8)mf${ zkb_r$U$Iy7R9OF?LcVtFGdz3sDEd4+$g*dhWc3maYpT;PUPV5X+%N7Gl>cYI6+4yi zXB8xz&zvTjmi>^9zqZ7|HR>WW_YlGQ8LE_LZ%eZd#*;PAUEgdo)&big|PxlQ%nMFQ)S&1=dxctC>{(&qra|A#CVKmh*975eKb+{M( z*xQ1M5RzNQ*BLF5aI^<0*zO^XYTitm^Nw-vJK`l7jGlPJYp!Cgv7+z%SZW%)7?@0bm;deZi zED}9>Jr5_!{AIsx-(Y_3l8n`**ZGhCY}oGq3RuwiK=#AK6Qyi+&(@G~EgvalEb zo*POe+%5M)`84giN;TJ0A_~uaAlxkv9_90B$U=Qidxr9%1Yw_a2ffQ2Uh>AhoT&Bl;Zd2ks?uNAu z-iS2Cb#B@uz4pXxM3s>%_L$Xgm}5oY*OlJL6nd{(XqZxGsY)cCcjS=Ks;^ zZ}RNoBTWd3RfjE2l3M?K0yA9I$}60bh7&LLK-ab`j2&LX&pVJ`_Q!D`?I=;E`Lc=l zx^Wu&Pdp8;CSJmcN2Q@4Z!mORUk2jTlf2a>8{77knt~xuPUEFf0|X0HqqvNlR$#mC z04NK8L$7Wf%m}H1y-6pbC{YCbfm}GX%?J(~Sh3Wfdm+)T66EKmVDkFq%+IclIVf&o z|NcHGVOb1 z&@TItogWbkF*kp(Bc5aM`c501?-DK89u>m2`1o+uRr>JWUKbpazp;TkN6^a2^WbJ% z473kJ{)*yb7UTPwsdo*8Z&ht2*P8anK9g1c>ITFve zaPW}m7XN&?!-fhJnA(p~5LdI8t$%As;&~Dozx6=uc{B{}e$j?RQ3T`}CPDbCNfIBh zT&AEgK++#;6O}#mgTo`mFe9giDY%5N?Bao-GcpLq%9H@rNGuf2{*h2mkxV`ECdnN< zM2abiBE#7-{N&f?L^m_^XsP8~N_$0&&AiUBPqWHwGzHKdGzIQWFo&z$c;G(|f@2dL zv87#y(qg`HuDR_JPmOQbloAaa_dP(dh8EQv)y0yNX5`VVPX$MB;=c9r!m2(UvFO5a z{^y?xQQevE?C!hO>?)tgWsDnzd1*#??XU_q^ATJ?rT|{`ynw~u2E$?B^DL&BLF9sT z2=~aPvu_Rg*UEor%L*S_sxk;`yUHMAX*}NTxK9_Sgwg)mC`>%BCDA%LvHe3*G5h66 zoU%lN_1OyfM?)|0K@F9RjEdD+JX0R<_dXU-6RQ>FT3Eq=g!nFx0Ok|R<)MEQ`$?yDD zjPRTezoz-e);(3#J{)I%dsYW@wOZjqx3tX*7k)X%W zO4RsEmhv}5V$r)YESwNZTh1J#U17VaWYGptpJd2CKR<*$q?0UTz8t(X{3&{{p@GZz zRW1k>+LC^A9Os!4$A{n1!x+P2686t$)#uH~&iE6(f1OH8i|lB0?H&5iCxn7?d0b`v z{IdG8((`QfOYO@$HldNq-i=ByEPx= z4xqK<{r=|XT`vBY9cdjrLmJ!j$T-1`*opI$HmZkxxOIuGPW_Lp-wcEn&j2u+y&lFV zOn{xL_t{FrC-7&6sYLTViZ>E@FvZf5H2m33rm}dggGPQbQ}yYKIjUKItzTRU`2V>AXWV6JKGYRWv@Qee`tJBN};c z0%;s;z;PDyXoo9D)qi_XD((qR`fW^I=kDTg&qQ|UM;~g*u8=Sn)lfqILs zq1H1Q{@;<6w6D!nykx48_=M?j@tQ6~7r#W>mVK7$%>(#cga6px;wv=fQX9?vXdoU? zGgiFHN{$&jwqs&tAnx7!4-AcF3GX@T3uWKRvY+Sg@Zo1SVrXa=dw6aRe{;G4B?SC} zrtF7o<|ZxXE8WS?kDfpUwS74G!FcBJcoBLQ4y3QysGy$Bzg-DInKkgnOmZF^83B`T%|W^5MCy{Td8>{z&^e_xOkEI< zRTV+F+-ED<8l>Qd+HrLM=3Ux2K~vmW@C(-%EEX+eCZyQChe^+qp;{Oy*8ij_*5^mj z*n)>7^Z;>3lOFDu+?%qi-8lJK3SzOyoJOX~fq}9dTbpG~UXOe*KiijV4{W1^cdzhi zs5_OFRb#ixGTymLS{Pj52CItS;QPLH^rj-4RCJfn`u}@r3~;8+ZIyJyd@wuGaG5MK z6)4PX0BBB9fwfCNfUh-yMbIsn_{ajJ(<&&hWd;2^o<}h?dbFe}l}zP_iQZ2ul*3l0Enm~RE1l9|hjGeaoMP!=?H9%BtlllceV z3Pip%kqYA`lft}fge%ZZ&uLu0_QHT}D%M$x2_Y-Wh>%i;dBZznrfQ!iuWmdo7HSz@X z*uE8;eCOd~&j!#4ml4vARd6Ho?UmA!RKC1&1G*Y)fag}(Y?@jsiw~W|uNPKudi-)c zJQ&1}rpSnssys-^^a;lI$DmY`hPIVarTg` z=m%PN%wbnTA?^!B-e!d4-kaCKQb)zG2+c%LAH4~><0gQ<%~Qy7QGiYrISC`Y8~AV= zXcK?u?rwX=BCSu+ufgVUbKF32+vivG;;JDwdMFmLNwy#6T?hkpIW8Y69)mZ1tOxQNXDyFFY~522Kw$ML;N4d*lW z8q?eR6ecB9GP|LlSfZCM!0;U)leZImZxpiunPXtf{wVe=?>^>DOk+Rxg#M4U^9+h2 z`r15M1VPCl2nZ@7K{C_lvDa*N!0|TY6iy|q|GE{1i zkqm8H9BIS|dor>UnALDI;Z8*r^!_msRL2aEsok-bO)$McYmVPz-)?Ghy?G&oJ6!Kl^5@3?bzS zY{SPt74JSJNS;_Xz|V>&P(L~nsy{|D%Wk)rEFgt_IG)F@HQ(n{BJCNc@miP=RKybA zD^qFGLfWNqoA&GdV5RAZ0o__*wOs*xe)UxFR=!@oJaKWwl7R8FqjRCCrHQ6h2Myd| z+6T^ZXTAH4C)vWsLU~&9EsSflECG$?-grE5Fs)6}LjA$KB=YrRJ?dff=W*=sp{b*zuJeg3p-*dcQS`My~$InnXQN?eCky_i?50!QdXZs-izDKUv1|UF~R( z@->$K!H&&+u7eYFZ$j#|MWDNCFNVF=#92M#aAjc{u5Ap!5F>Oz+McN2?w)gpOk zKM&mB8KAdu0`u`IfGMFz;m+=Oc(dvSEbfwL_Oqt3VRRGwsz1Y^^(wr3Pd^s-tc1B- zHh~Ad&6(cT?x1IwY_K*mznj)@%kdwlaBVVlDolc@&kmt;#452ScBA4RA}Y{h z2ODvJ2~+6R$!guaaadF$oS7gdx6DnKIIHbqOXE12J?e`z_t`o3XU0vetILNjPCp}D z`Fu?XTMF24SXriY;sD*;r$!ZrKC$KPO09` z)@3v5cF&5g=x`YHxCGC3-bU3Im!ND?E`~pN0+r7;3zJrQ<1c=b@Xtk_T{8~l`bl+I zeSs0naEQm(i~pgErZIo{Qx4lOZ5h3}vJ#pNHH4F5``l;pOs-27$Mo%V*%`+&wnr=o zhy1DGzJD=-Fl!g^DVRg2eBe;*bq+Q2znjy-1Iyo>ycO{=`6d zM!sDC zyH}V_{0LZZrUzCmsg`W)xW_`iMGFysPt!}o7;=7chq1q} zfaEiQq7qRK|7ySNe ziXfll%{5ybg%O(vF|RLWOmCwz`!;1aD_uC1t@<|z56)UAe3_pn>3?Fd)K8%hs|^hK z6E>syWCyX&`>&b3AACS4u#bg+1_LS`o6a zy}g^(98IN&m&h#4jd0g=Et!_mZuZk?1ihPG%8K5WlAP5}8Zy6z29+ey)V?uvic4bd z8eiG^oPDfh!&_$kwJ$Ci_LjMQ-Gh-;GVcR583-+Ga+pBD*%i`nH61;}{Q zg9fg7F0y9kV_JF)x;n0aRl4)g|NH}R{a!*UA}{1{loPvf!(wts}`k#+?4qjPdS==z?XEGJh-s_}jrbyRo4@}RE}pnRA4d_6($RrXTvlqBKd zZVt!4Ta69l3#r~mho+W>!>eb_EJt-SuJsrW&yRVMPL&+WLObJ8+FsGoCa~M=13a-1{Z7`4x&n=sP=%j?Txjy!E(X%UJ&D^6`Ax#9ZOfr49IZ zT`kUE9fx(nX1ISui)3s1H(ar_5rgZ-qCwzH%qUghm-X+(C-2;Wwc?DA<;l3Cvy1V& z%N-oLQX9{f8H1d*3RHh`#uUT%FzSa2GkEOAu-=^iK4L0=@ZwDV(K{Ec_TLIdT?*tl zHWN27Z9Z&Q1i$U)a%gh(VqRyKNXFOX!|WS<@myj6S9wF7PiZ#cL0OkS)uziG`jLwJ ze`}%pWG&b?y$f>YPvH7rNCNH2>u_%1Wwej{2BEKuSaIL2(x56Mp-;vdnm1Wwr+L@n zE)zLkuk}414+zFHi}r&Hq`^WTPhR`U1^m(cP&!p)lRdVUVDbJ~quTNA1qI$h_mj~6;Tn9s6Au^5USW^dWjIu5h4J4c;A$JhWn1`ALfbU#uCav8 z7wgvF+-Eo!$U*d)H0X3lqsS-SWJ#Ys)09JI;H)d=+*y}^Q`2)Ud-n~fm)D`CuWYF8 zRkLtgc?u|hod(g4V==#VK4uw3}(D>n>>ow{e>)Q;WsF-K2iH|`>(h-^0JjS_5#D5ux&yOGKCdr<6P059P( zlrd4O^TQTDO;Se3aux2;&1X<6X9@Fs{=)3<)u8EiPB6=H=M=}Tp}4^TG;4`Hr?$b2 z6bg4h)G=d>G0NdKD67LwnGQ@E6pJg>CNb^%SET3lZV5WK5(J%yJgIN*f%w#JFPCq& z7fkI`P<34!yn3gJYO%Gf$EFWblUv6)J+XBKm_B%|0= zjm50&NC4b_t%`9zMl^lG8Wwa!U@`uMG-Rzdoxk@9`W1Mh@^XYRcLrmF;#@G@phW5_ zOJURC&D=oUk4&0900UG~aOksGTzJopAD5Uzch|R4+1p&s-a`f7+mFGsMQd^5;cQeM zIfyU*o-O={k%zNU4kD*Uj;|MU#d1bea4%;^pg~bBjCrk#>hFE|qTBuW%5w{FrQJ_h zz33dha?2zw_7}F~E){(}S(LP)R(h*n4B9Sh!hsKEVDWkvdfLwA=kAC>g}CY1tB(~t z>z|J+A7pIUFt?3c=(&ZOqUt=TfzfL6^=%k;`)g z`%GWJ=J=hbuX-lD#%T+z((n@Rn+=j%dyCl@qc+?+dapEi(r=6%_6zxn>-cGkC4X(R zG5=4aH@`1hlh;xX=bbLZqVFZqi@c#fUfS9V-#>VYM$ayzAl8#xmUZWH1qn*toq~YZ z!k+Jb1U@kny>YeMMC;5{&3g$Xa%^`xd)Zu))1+ne?nwm zru%^Qq;TQY`G4%5&0E$ZY8<;z)&y3C&e;7@9m`bb>ATWk8YZpaE*^}O%-DPw{I(&* zMgprW??!uO?tv4^7EG~UAhTSsSNP?)LHv&`g@LKA>_=B3wS=!DS6^Rt`i=!mNVA93 zM@Tq23O&;3BJk@xa z9O_jnaSz>|a0{Dipht2qRwT0K-gNpfi^qZ7dL?JNwP86`1{`D)``5!(k7Y1KJDgST z@@JSFNImNYau>>03qiF$Leka)5U=Y%yUdix`;|4>o$D$5FwrEFb4EMG#^!DjE)A^{o@69 z^EJ|EW1Mly{1w~-cRjWU^)Ix`ZXe5SJ+^`^?pV%zvZC3ob3JH4;sj~p>2`J>WB#t{Hnd)>$t@2(g+EWe#qv4vSiMxjyDm@Q zPr5%sMeiK!`!yCvUQ5TOCE2JkGY{Vmh{3^pJbT`5gUZf&{L=`+F?o}*?z)H{O2`95&mvg|`4;rvNXdr(f`>L?S&K2*CxdubNE|>f^JcjF2 zRC)EQ53#T9U%0Wun>`tK1=WuqMJ3LV%dhQ~N@7{y5wuize-y>8=$v~MFk3;8O#{rE>{3FZDIBd8mHqz7M2TQei`+7&dYujBcNV<$S3bUYTk`4xRR#}e3@ReR2?NIub}-GpXWd^eExrp5?4xVzK+R2^MtcPAaQ) zdCRV?y2~~?#nI5O!NP>}8PMvd4h7XcWiGvk(9}(;WNbJM2UQQI&wi=Uitm}`q_ON} zT?RxY>cH*Q2iUvuV_;6W0$CL0>K_NWo(6Qi zcQmd4{e!F|P25J+?NGhy0@ry*2?kciN>@y_V%)bVVTx(KkXhga8&$hugYtV8!n()Pb$*cWo9>p;)A`)p;45i{gwfJbdSxt6KPG#ylA3SFCo z$5Z8mnWr#k+Zapc>{VbUDv_!YQ1F+?V4vac= zn7L?|37dQG5n9LOz`D#YoZCYcw6Dv75R*1;+|g|Ct60Qd9CE~+`ii{eeFy$W-*wpI zbB16vawKkD6^lpGW>L$S_k!c58!UO;Q9Si08pEb-#2No`QCmpDX7^*brDZO1>sMmB z%3_pX{FvqM-oaW!Qv~!|43f*0k~p`?+{^L%vGl$w8lL|Sa(6U&(^WOtKCwH$ZSO+< zp4&kF#+x9tng0l$dOv}VQQMg?YBc}BJRV=%D+I?az|`JMXOGtX7B!@{k`7-rxa?WZ zZ5lHLJ9i&~=3RCesG7wr_c??96K$A3B9UFzf5PpeA$V+@IMZmnh$Xj`&~llb=$+n* zyFXi^dCDm`Rz3oHt}lSti~YjWLmYcl1Mh9!h&@9t z!pl4III=|<3&(WfaLF$8ZJCPe7WlC~H_|0eI>(@}*`?=U%c& zBjxC!WH+n|{lZEPOk{Ixy{Mr41^4-Gf{-`e0=wDR!i;aHxNAq&vRBTtAU?BD^6L6$ z?zo>4rbdn;ZSJ&WWnctF20dgWy!%j8c|YM-)id~LIgt*fsI!be>ZEPw$-=U#g{60H zvA5k#Fky|pp!@L#)4ZfXTBat%&R5aE=lAI5-+r>mDg;$iUa&@=`(nQ0d$v&Qzwmoh z*G|^3!8J;s)z&Z)Tx7TV2>Nh zWTU|#;4Nq#iGlLO(`=@q3JHtMi1QuF+?-D^lX3f)LC=xw<&?qDdyo_N_r)(b?5G4a zANpa|L@nl5eU62e6@!^HQc$$+DI`yNBAJFH}B;8+a1@CoUscMx5_HVWf@Y=!d{-{G zidlR!%o3QD(i7J9>H;h9zs}lKBy`X}f%b+rQb@!^vMHVeztgpGC=W;GObJJ*|s}1Vs64a74UFLE1j*sp$Cl@CzJez8?eq;PL}()rz|Q`Lbl!e(d^m+A@WfaGha8E zxu*ql>5mI3@ofx6WY&^u!3nr{au=ubIFL3ts?hE`derN|N}AqBQ+D3`BRv-isWJ7d zz!xXbN;d=9`hl)ut$mCNR+W&?2eH4HR)RzQn!sPaiu(LHNShLa$WY{A@44EGjMlcW zLXoqYY?Md?KAj=u7Z+&u+;)n%ogy1B>pmEKw}EcE8Eop{q$hU&l?IRWfuSQGv5aXE z?2Fkl79Xf8By>Dw^Dg}q7F8c$dpC}!s7s2py0i zv{Y8!!o7NIk0-9@OSe>8a&@LxIZNlm6#T25Rx7B{_rn{(3pPh)SvXt?M>PC#PGv>2(MXmj`N3<_SS(bR|4Aq77qs1po z-6!4-6EYX2vE+_H6ne6OMVKC8e#Vtt@Y~_=bdokQqcXNjCz1suXTk8lCn;{!Rq;RU zHhZhjAkFV8{GR;;Clxp1aG?oZ(-X*R;zh~T>@qUkAC0yL5Nj)Yo`yy zO8Fuj9;%DK^b;`1c|1(5X)l*J6hYWab-du82C|f^aMI=;{&qNuyBu%eC)IAS`TJN@ zsq2Xoc3;KFkRn`QZHYglI?1E=ManGLLkSP8W#K1OWsAS;qFWA?f_CE{LG!vCzEQY} z@4sKiLE~Cs>g)o{eYgmM-t2)_J@0UV$0l?C9(%(48^cH?x>lOpzaP$POhc0+vvKy) zM;QDx54)RvVKwCza7^(kTJB51;qUW=+9j2O+Tdhi#l1u!PMQvb15N3WQ+M=ZMr_DO zf3&|;gmE(;vc>5k(pL>fF*d52(${}xm#Vf=acKjX-|LUo=#R#ibMf~)8Dz&*!ylbk zZto@?uCALJQ`me&Fg$O_uNCu`Uc5I&?fCV$VqiYzR|yh@f7-$hGsex>q{MaiX>{*7 z#1$5gHst0%iG{JhkQMAq<~D8mz$qDBfCX25p>w?rCN7A<)WdPe8N9?fagVU-(KK$> z`DftqwTZ!>Hi=Q-PH9-k5#f>TGGIN*;oy)7a3fKVaRzFV9y`XfC7ls?_1ShjR=W-R zeYwSE?eSzobhfju>-)Jm;>_*wH9vMtzOce2Z!>HOpMq&WRfX9KR;-(oIgB2j1pij= z=0K5bm)!iWUBLSbNey?9VMmsiiqsohoNIUlHGVw%{$T)u6Lw z9xgKJ25VH?IqpqwsFt^Z;ZMzATkcuz{PtF%;e{lhRnB8tWXtvgXCH?A!(mUDQn*GDr3Qv0$npehIHyOgBh0&;!zfJmRo<5F< zzYR}*P2=LdPfFBcZ6T*%KDXwzKQlXKNx7MSnevzcEba7Qcw&1~QlhTF`p%DKS4W65 zWnCpcsO$uWnWw?#`0KEEg@Ct}m$MYlBhsK=O3b54d9uZbF1PCuL--YEzGuW~wYoYJp#U@7)OnU`eP4_1qW-z=?I=`C9f+AZL-1tZlU&3_5+0t^ zV~@gHxbWzSnEX=?7vdS|myT5|V@Cno?VLh6i@n|dZD3pg27QtJj#|FbjFUTFg7KtQsxqHZ$Pb(Ms5_{|~r1vpAP$^I7o`7YID@9B1B=KU86pRC1%&=4iSrNh0jMvsAw!mZ;yw&ifWx9+ zKRhXq{XOwanCKTTq^+Gnao#&A*}D#Ah5v;CKOe&7ta@&kX(8O%DQbE0=fkki2RWg0 z0JNVz4T=kmLALWd^m5E(aa-=O14%aIqZ7*>ZCC?KXLSmfHs}h^GR$CU$1YZ5@fhmQ z?;>M^_57L}Dnel52lghn14hO#aFY!l%o!ULu;@XKQgszGNqvBUkaaUi7!Y7B*vBX1 zwK-0bDdwBqUG9y8jW+FA)iMV=w)WxG?z-aU0Wnk=_5gc(gyA1{64wZk_{U|RbhLK? zY;)%ErPn=)QFyyV|IuZLfE?5h?}rCW>M;4F3Et_`!0o#@l7h-s!ybBNz0D4@v1n+e@FuS`I z%GR23Y2TcrvCDP}%Zs(7wlI|Q)>sWg?)Js;9UTxh*MZHyVZle;bLRb52lE--?A<-i zFD6-T4`Sz1+z(b63)|B>py0AMcVx7v%S`FP^^PsV+qTC1C$C@lw9gAp{(L-M0R;-v zo#%c=WNTcSUI@vLtHE%>986LC0)|I_Lfy+OCabujpA;WJdr#-)gYN@;)Iqvu1aF;l2}^C23AseSMvR0F+7_+ zg3B`4C{2HAL8JeS;~suV;etLM0sj!Eifv&PQc+~d>h-o_+RnasBBU5wuJ@DJ+*8G) zFXw_`w9+uV5HyKea4ZWlz@8z7z6=j8g-!x2)!z)#%9vmpls$GKF zH40d7d{t8VwHMrFjeh zX4O$n)ovV%&3T9@^27K?-_iJ2WJUVCI;m?>H#|2&8QM4Z!s5(%!mW);g6r>c$@Mv2 z?DoT_z=TpL-=c&~B9o~kT7mb>tHMs1uF$0Jgv~xhINDE@O`qM&`Yvu~WBTrfc>{H@ z@%R$h77>rFmV zRy*~>vXlD!6JK}Uc1$Lx{)o7h>z}(VIMe1{?DoR_d1haT?L81?Z8?px%M|z{`5n;y z*h=cN)l)is+6nM~JP4wLBAL@yO^RN(3>J_12}`o(;8EFd_QQNB%j}#BKmP3y!aT0v zwgOB1{pAp7?EF*l_u(NXStatrkYE!0ymau5`c<;GWLK z(U-nq!p3Gi6}}qm_NKBDqXW3)NDBDsO^1UQKhjDiXIWykz3gsFUz{7}Cd|FPSx7J* zM!l*Ai9Db!#2mhJeZKV;`aIgne6*L7tx+>+Wqv1lyENJsATMj0VI=$ObAckwHDn>v zOk^XM)eDAq>V$F4?SktJdD^Htm_~I-s7J#Zau>$YG*2TcZ!wc78#hb8ERc}>n3JR% zV@bU>ucX;-@!;b9n)#i1!rtlkWKp)(^zu{^*qNVVNqrM=r`btd5i*bMSaj3?X^TsB@+6?3>I}qC#0`>UV{tG4KO~&f&Nurpu%x&?BOesv8l3= zMkf5Cu2NfK9f?#l>@elKDbf8Q*T_65o8F3`-jRA;!plo*Scms@=2TR{^c2psai4Dq zRU=(7FYYu>iMU*`>GTs&NVdV{+4WLxel2`^;|#-HLMXC#mbjcK4BF-{KDU^u0A`nzJ0&_nJmCpD57Q_me2Ed z4Af*_t-jKcj$-QBSJZ7CY8D2Mm?Wqi@S)A0w=(dtnIA@!0nN_uO`1>#lW$$3>pw@)6H9fR zRKruCVlKqjMnKKoWAxhPKGnKC=YIUV3_ZTIaStx90-eykcxs|0&h*W~FQ+0U_Pd^f zPvb9md$|Xlf-!8HoF=Zw>SFjI68n9AfxU~PD}H&zLzwOdQIqxsT!Npngx_m1W5ipq zn(vK0=l?}?*-A(JH3$djVo;qXyq_oLpV%t!Z5n05k)HtV=asp!+qZ}rrnQimtHNE4 zQUygt9Vn8IgV$}|wAE)jEo?jk?{m$??AtPkQ<#E5J@2zf&q@}Nvlv=)W~A~?Tt5tsgmA8CRr(+ zVD}O4EhW=5<|(GrkceJgkuk){(!V!L;r>bvSDH3q&$w+YE2@EwSzXG)r3(Cxoko28>k3?_ zkbnis2^g*7A&fX51B>s@!prOf*7bpHkN<6XQiz=pMHo>c? zR8Grmt*rmSo1psR9fY3gE>qz~N{-m}KzQg6G{}VA_|iuhCH>9?^^0Pan;!~0HVkCB zqm9_f25m%#0+MX_iCK3S;lVTiSWr)0_%!{Bs6Q5aLkZ5z?b3P{G(SO7_D)^6^G7OO z{`)0oSg=mIl=sK!lZ+|Zp$|WE;%-`n9ZtRt)>bc5NwyTZ&4uHH#HQ@4YJXG%h5yzKKFY&<@VSR1MY6=7|hz6sedt@IimRtSr_%hbICk46R8$$s`+L6-W_4z=@a8yimSc%sGZS!Il`DTQ zI}pR~PGwvAsmb(8Be3G^c03S&noYMV=X&1R!j)C$v3RkbxZ-~aPIqa`ZiftH)gMDi z^1htK9cd75ywHNq1DDCG{RF$cF`pfIDrSwTu7}gS4yJXSV;!AC(P{s4GCw+o_Fj7k z#{^9b%5KHhA5s?Nu@2%cUf`B%AbDwPqEqc2Sm}BP?+w0;H*1V>p+r%py~cyom(5`t zGL!{{YD3|`(l}O?Y|5{#@5Fmu?_qt4wZy82iTkIMb&{4KpWDIHQz=-~$WWL&mDidU?}zD30% zI(%q2Ip>d|HZ zwz2zi@~ev_$YYLvYv6Idmw*2|lgia8YX{ zyLRn5L_ErnYBYaf237Of$mTZqxoIHUlx5QQawD1H%O@nC=)$98HCFbwf{!)vFfHm1 zdw4F7hBobnuAz0phgw7W_WGvaq_Lk}nU^MP44uJcZh0>}v?^rTs{8S3dkXGZXNF$8 zlkryETAbdi6Am@%@o8GAm_P9gm-fY1c=fsun_@YOJ{_*XRj7@p6y{T?wgZ?w2%uQa zbox;_9(;B8a8noc!jI=qW5&3*aCYfI_T!Emdo=S5P10T^Ym1TK{Xlc}qRxlL#*RXZ zh)SGZvKXgV>}0W{~Ce zaPyy#KXD6Wb}JCRy6?b)n;A|>55}--tLUBWGVb|plzBDHppGv)C}x~JU1@y}8i8A= z`@RF@x?nmxEo#CXE;O>EPTN_^#6xuP-x0D5l2B=DGMOKWq}u#0cF}MfeSj$9@?sEgDE~0?Bm=HAtQY`mzlkSn=wO=${c*4{_P4D zUFeEaujX-0R*#sicOmo5l?o^2YnY~qwG5@_#Cz!i?J17NtNN+f5n_NLEs0b&LRn_} zuaw%4_|W0|sWe#2ipuw=llLCcx3yv)JO4}xvc_JiaM>tG_k1qJZePE$)As7Ly}v76 z-28(cha9ApGlmOAd0ni=&xp2N%VK{<_Cn5Y6q@U4@RZSr`C2vjvv4^z?iTak?`Tl> zu9-N%DaK3ut@oAzBw-MjQJSQLkkY@acuJm^ma5StYkQlUX|i*UWY7pT5X9 z-Vz0cDSuc)R}LO(G-J<$t3czdG3T#UE5zUK#xHZX;%~0(!AstG3k9uW9j9Q<)k!bF z#KQ~mh1e0y=w-|wv^|OW4c@FRPh=8RnL%hoGW%$aG`8?1Q}LW8rIZZH>`JHL6;r6C zER`D(I!^L+ZVv0;FPS!h1v;pEl6tzA=oNd%4Fn~w@WF9m#ILu)$SrThefQNwluNHFE^0KqEfoLt&Kfs4Ml~azU+R(Nx^!d3jUfJO+&89%eqf~ zNEgQeJAdC5Je3+?jE9NLElfkE(dPrr-FinV)!5Ab1{biJrAw)|>^jZN%AuAs{*dYX zhI8R=a{GMN(yzm>!Sjp-?i=7Qv$5$Z(?6sltK64BQALkosQ!lx zEE_=m6g;Wo`d()I{Tg$LRVFwRMBO5lsYvoga(uvap>$0-D9nE*v^I=mTg)>B--sjh ztU-ycy~!bKZYI4dZKRYpE~GuDfS$j8O(8RnP*0C!A>Y?uGI?_aYwd0&6B@qJ{_OMA zJ!BtSD{L2{;xbu!LnHY{E6c7A3!wbGFO*>)38C zXxTA#xJ4beae3TA+7DXWpK&Gb#@HS_4IBn8k_@v>WyW$|EZZy+b&8+iqHhHlOU2!ePR$S?l1B?-^7$-VV#Zjc16< zJv}O6mp_)k<{?!ug+EL4rVpoO{d>}vy;o^wTr7QB;YQ;JEkn(eOK|1=JLb8$NvIz9 zn|XEC2*c99v-w{+x;V&@j(*)kb1h|@ZlRa(OJu#67H(s@n8?!DW*i`Dmz;1aHd#)B zTq`x~J^C*fInx0*JogvwJYE2LiP2ap6r+xI0`}~oO-cTm++HVJHa5zdYBw&y5BsES z?iB?rvulU+Zr(I9u7FEPHmAP_<>*n7jGMjS3F~-2m5Net33DRzVZBp-sPM>Te~tTa zLb-$@cO1u_qNY7&L@KB++fF-;ItAlLFIbA14%3YJD>z<=1kWw@T>REPuzdgzipxda z*bO~A^>7ak*}n!Zv_-K$RU=vSz5%>qRXcosd4!b)UgBn`XkvHwRoLLEC0U3=UXav~fwi%{ai`OAG!+?-?IVWs z*H*{y*Hx;y$2V@nMs6N^E|qa-Y>ipL+~aU^&So~hE0S}MsSuu}5%&9(hA%61@bksZ z=-;lwkJ_!m&q(Nw&tF~^xfzx;;p22npHU1}X2x8 zoWoI8_WX#guW@F>A?&lZ689gOfc{^!_%w@=yj!9M{IS``jG|6355qsuBQ74f)lU4q zdBfRueUWj5qUI(372Er(fz*}`r~X@9DRPj%YmKofCFxpWp^^iBoCeHHU6G0g#j^VB zdiGDZ3%+mbgp*x6Q18ufv>bU9l&BK=ZckyIxpFMv<0=Z+@J;wo;0XyD8=>~{1gt+} z3Tn@6F+nYg>E{=@$K1|hHje~Yr8JPD?>JGjaf>i&UNzY441u5j0x{L4idQ#P&ehmQZnvD=e~u|$ib`^Fs;c6%|LtKQhjQ~*2QCd-zi9Bn`D^D7 zc5!iWc68+=?oOjU+~sEf_a6Uiq5qsmQ`MF7|2+TC@qarH=l{cbxH&mEx{dZ2{9io} z`Ts0z+@gF?UR?A)$Nz0%_&+Y}>?|(q_FrB2e_pQt*Y%w_S4TI=e}Db|PuE{O4QDrp z|Hk$I|Fph~gM+)P_&xkDfBvbie`!_HZd%JL$tt(pr3-8dc|Ge*Ue*aTKF)wvD374S z_5)=)g~2p7B9PWRAgTyJ>hobd*^FmoUYkIh`XrItjH~XIy(^i@Ej6riuM<308KBFS z=PC~iW zY{ZP+gyZTb@1>K@@)MO}WznFH9e=rCR`_>HG}7GR?^1}z**xC1w4 zQDM;@YMEmNE4wryYvTdhJ>WTY7x%r53r5O5n`q0HYrmtGpy{$$4ISB~im~Dw`bHj~TKlk~*5kcR$hV^<#!6?yYw_TlF*l3OADQ8WAk8_>rXCX!eilsThw`e`ip zyNTnSj-vjEt;jYPpxJ;z3=i6mm9iChs8Yx#5>J9j54=of?L@o3K4 zJQ;hZi`j7xPBH&JSLoIjI~lkK%X&A3$@<0| z=e8W&oSTjT-cxa!K^QJs9*y}kqS5n61Wxt4f`;RdWA@k}%;?g`Me)v9_tX&mXP$#Q z<4d_K5tS1C*NQB9#wGR=hSB3yIwW8Ihuv*fr59t?P{AG*8mISwjkC{@re;~Q2SM7@ zvr&btU%h6FlZUgE8}-t7rTp@%u@?a~)I>h7yI|F=IQrhy8*i1z+5vtSoBLjqt9G7vROi?(!+0xxXRdti#I_ z)){tVJ{8@lY4|7Za7I+KIL?}p>H;fOPCj6~o2B9k_x6)5GR+rLL*ZhPV|m=4~k zGwBCxT-hJSWXWOKzy(6ms#F#}Y8NyaBnYjMNzARwp8|)BrKpE$w9Z$T%m#STm?aab zV&@;`J>wVKYrjI;@6BZv_Emu_r%nqMSwOTp!Cu$f{t6O_Jo6`WklSr4qNXshG{yG^2y%vGjM&1adOX zrFkklNyF?4tM-VcN!O>5{tQ)mo796Gd?MM@mowQzJtg?g_cyIQOTOhxkJ*$jnIsNWQv_$sP`jU&R^jdFvqVbf4y*k5`&rD=_g{E-P?-~1J zZ$>vlq^wv`o;rIBrvUdbwzb4saK2>7_C}>aVptyZ_}l<~cSnJ%@_v{zwWsI-NrbFV zPr!NPb&#HDgxSXxBr>-z-0>&6!bsglX4W!+>Wd~)V}!QYLyMxjo7Yjt``xrOJB{Y| z3Zg$t57COWF%)~%o$igYBfI3Wv|_0RO^*UvQU8nGUNeFXw_n1>u5M@E9v9i|FEX7234RjMr8)!9pSG`osU>)oV-V{^$0nrTMw zPO9zKKo3IYWTTgUr=QjbNjLos4fVK9 zk;||l?CA7bmSkSUQoYld>|vF#B+8I-xZYIcrbb&k{m40R4qZBz%4Qv2in>KSws+3Q z$?3WHal|fMn~;r5Hb}A8p*ZY_`+>Hz-iq05Ss1Fe3;UiKfTnvU!%v-FZ0Ls-Oyl5P zrV-tPVfZa}tj}8E;BqnN`kWwmOluP!#alp>O)tv2wUQn$wWO}znpCw|jve?sT$n%P zIApE+zu=c0inSLUM}_s2I;?nCt)*dx~fLvfj~ z@?$tey$Xi#Eweyj(oUAmE(W6`N}$MFNTl@v!dM#XCP!_T9HINDHH|%Ujn4XYmD+E=LW1O1GSzxfw712V_%8vxG;=L3 zez+Q4RJUXEmtfp-$sY6P&&Sqfp*Z_XG0vKDs0*}CW;6vEQvx8TI>2T2d5x5qIz}4*u(Dz4_8}Rra zJaha6o!_OonjgQzpV1~bnVARvQAgmDdlJbpjAZjT_i3->Vn$%h3y1Wt;rg75*ztnm zN_QSQSA0U%!jI_b_5xE+l;d2#a*WF8LRdeUKNlb2m`WsOT~5X;b!IqmYAPnn%|fk_ z6#Czm9&+i$WY{)5PCC~=fL9fQ+}ZzL!m|TIFe)p^4Vm2zJ{SM7@2@c zK*4^i4RrEY9{;ge5R)LGJ)XRg{G6x2d73(n+||@19iDn%tz{10!zo}CPyrim ze}&2pQ?BX1KcG@S0ojwMbJZ1Eposl{JFWTy)ZcuA__9gdP5Y&|UyC)k>eD2-rGj$Y zXVr>am)Z{y@UjC=K7J4W#V0}Sz+EEBIwC~-Lpf!AlH`{7AQgCj8lOxVp+hGo>u8Y8f4-cSxuOKGvk*C`6)94yQ0hlFW1?la=9H$H4>8oxjxG;T) z+<6!bf-?o+lPk+kTIzi4S61k5&R_Wr@EmeS>R=RK}-V=5`S_)a0*>2P9zrbBw0at(d!eOBX z7}NLyAsw$_>i0WvG`9)*y?(+BohW!Tun1aj$AN69D_mypn#+6C=%2u&%ow}VYs`E^ zZ4d3|I6g?D?eeBLaW9vno^C}O0zEJ${34BDwHSq;{$u7HcuBQuO{lOZ2a8u3^%GaSm8p;aEv|>I)^#jrFed}5=z-y zpd+u6w~qajk-7j(uCBvbTLm%S?gUPlBHE3G;o~I=Xcv4IBW54Ke1|mB=htq@^v*z$ z3$w^Ci*Uw5S_vdq-vr6395O*N!6}^e$o<^`T3_Vh&QTM#e_ICQUh%}JXPDIA{7J%I za=~9E6yAqDCqaqqecrwqR5Tdyluco01{M%yECpH4DzL&tkN>-S6<>Xm3EwBX6xRhD z#+&`BXv?$5Ib(%njdlh(p~~`>f|79#EWo{N)?2MEk$&U_Qz->L8pnB~q*xi2KaaGm z)+Z0gnjkMI1sqN$Kw_FES4+JX+(&PK*CsilxXT|G$xl$zt@3zo^K5*6yNjOfW;zfq~cr%c+yDB7lK z1+fp}pevQY^GI=y(d~ZTgS)mwNHG))^|h$ZkT|&pvKTa`PlOlt(W?xb_vS4}`&Km^ znf{H6PMwC4#{Tq||75xqqRBVgADr9`)r@4CEN-dc!p|fLIQ_&725V2ku77jkrd1aC zq$9-oDHONXYx)h9i7|nZc8qLWt zH$aVqQ|xTh7+0O|q5WT~NoTSkuiCDMT)4?*E992p)2)W+=ewI;o7qIy9=>l;qZUc? zqZlfDSrB(Q<Jf1kN<5fXSApfda=0{^0$SaHnm1qIR>)tN^YAHrx_lcv7Jr3^<^=eo{R>1^ zs&Gd->cHm2AGpYQ0-N{M!G~=;*w9-6zQ0+Iz^g(Shzx~_5p|HRmV{s8)#OIUCPw$E z3mG1d<#a+T9XY*&jO*3%HqGoNkT@bCnR&9TE)%T#J|44ByJ z|ER9?bi5U(gQ0$a(w!$6YtP*zAlHvfW?YF{Z4uF##9?#GPOM`$7aV>!vo7io*vm54 zZ3nl&pOzGOnU(?x#gXu@{sO!=2!NG$B*E*AGn~-3fF1d-h}A-0GEu&n`JQaai+izz zEF~cjxWWf=J|u89*DWO(CI4B*FN~%spKcSm>P(u>E2Pe!uHnIhY?Jn_9 zKi&dE;)M8WH%D>mmMHXq61=l13#t5Z{HgJf9*VLc)EwuW63eGPL!O98=moARQ8i0wZ(9pIRQq2J6W&NVkn4bz4EIMfb4!ZP~rrGhqN<%niK$ux3<9) zmvl&p(*ef!9Jr|V5!2XzByaai(o4e0pBIFJiLkdv<&j{$P-fc zK@>u_E(DJi`b1&JG>~2}gV;a4O;XOzgB$;e!#+G-ic zYfCwFzX=?1c}1c=vu??KzC`BRKcdaCgc)k9;kunUc$PjUM;3lBJEPY}mBm^1SHV=8 zKITWtm2PqxRAlJ>T_xmgY%u(JVgMDgK5*!GCAsP=$r+xnjXRDfk-RDEsqW6jjHmB? z_ReX78-(A|B|T?wnvVeUXv-2J9y61fU!X}mMp)H2$}Tb$zV@_f1@gnHCE4Z2*~4A_)6e0+GoF_f=newdTy-y_9NY^f0zPD{0LkdHl0EkXFuJgoe-0VS!K`?oD}& z-iFzjGI$@&*qIO%wd1L4mAJOo4gI7&5b{>iWgh>LRzYoec{hVRx0Ygaq`}bU^BBIi zWI`f)HaAI3gu`cU!28@};HA5QF^5OwoBp%RzfnsoWIBoIcpcqn+e2U0dC(2k2DGO^ z3l$fsl3t@iTCm)OJ}D`t$%iQ8aeoSErLL#T^u)pO_7do4>mNb;RKRALJUFH}(%9QB zoV%}gS@;xxBnBIkNVjMbnR0j`;Ho%M{W*uYGXu=UW+!Iyjuv7i=|HYUKP1_jp-}a> z1OhVRKu$LcUUZbekU$(beVq)cqp!%XnStb+TM3=}R~e+v$zj=k1-xq&gIeh^*ikr& z@yXRVb+$O)ZlnsMw%x|^bN(n)?}b&J&3MhC8m0Q4<87g1So?DiK6+e;2DdtJfv-7M zhde~nibePat+2Sqmr*jGLf^mD!qOQA*dH2!#~O|C)DmNS|4SA)=D^g%oW zzF7KbDf@j)#Nm=yyrJZSn~yVWX4D!rlBb~Nz(Jgs{sR9cvO7c0RQ$cSjSh>o(yRm4 zwDN&3z4)(*PP=W$ln(TgcMo23q~3g`0hvB%WM_-6`64*otQwoHhvSQ~A^P5{ot{~x zjlZtEqQ&+>xN&wqz5izo1{-X}X@P?H@G+0xQ}3epjl_u76Gx8qk2REit&`AobJ_jZ z0_w!m1v9BYC?A{){#8NbNAOJ8QPIleM2TVKa0K?3UPoo`KQvf^WhoyPs}5ehiK6+0N#Pm6(4p0OjpHCxU-_MYlMt{g3QFccrdYhQM=87n9(*H&$aeraLijQ zKUG0@d!{RCaBqu}zo6kbgs@cLXF=tg%E-MROO zRd6BmuUUpv&C!C>Zkxl>}}*poaSOoabdasHF0bDyTlUj9Rh>E()9HlamF*}PZoEKu6xt)_LnN(UFODVLYP!Q~5{@pjMQJ|`%vd)G ziw9Stux%MV>?)5TekbVBqMdl)!~(KU`Jd(8&POEBWPoFL!IwyXuV9>QgdrvTJ@xlq zLcI!t7?WA=c%Obeq?wCsa1vF)i#v;%{=eIq4^e5<<#`>szIQsN&$&Z>Ke`9JBNWDO z%5dKd%W#*o*+~QSM-biM%W_5M!Z-FA`q{aY94AX~v#}T^c7d z^#MbASn3n6(kXy+Ds7>=ztwWWBN|CU>Jnn9a|P_elwm73jVwR*fOpE`70Fyv!%18d zPIa&UCfb{?lVfT%OmwIlePiXr_RdNzCA|$uuU|0dG{*-o&WytsFZytDW;%+Q+{KE6 zCvfV39_kj)!2?dd*fi@YJ?C(U&U`Kprq3IBw^pyfZKr)n959wK&yM2A)f+U+)`$!- zI>e}=k2(g=1+zb&I6@)8aGSXY$+Ovv-_{ARTPnoufnI<-MXq+w04$by4__LjxQo;t zfsm;LH*Sg+cadc$D0E(f2YZHaPCO3>K8WzQ35Vdo!i^}MDo%Z(xnv$sn|){hA|-?D z`{GP8*|Z`Fk~S@e$hGq!8A8GSxCnu6!Ep-qBOkay#FwT~-ZCky zR1(ARZw_>W$#rUfHJ|={Rlu=$?@mv51`zZ0{&aM=4QUosft6q1k^A}zU>wy=&IX0C zERaN4rXLFjHL^&Qrz%LbJ5a+%&q?aLqs-R68%*n~Q8M?=7&*A}9Fw{@o3R%8Oe^;V zlR11z*w-Tu7qi!ZLrfr?`xgw$<&$BJ$RU`#`5ftQ9OKP@l}&!$c}H{;qDj%(t2E(M zkp*$Eqmk_AbMNUI8o1pM`!BW7R9#JMIQxlK9^8kqwVUzKcrCU{DzeYuBs{o&HcB`} zV_2IC-gEGwW4CnihgTI@U97`X3~t z)96gmm2l~sALpCc6Y?d)n+$?ERc~~l$5cx>GpkIX;*mPorfCybt5D01!YEF>OgryN z{Cei7ojf(Wc$`x{%K}R!X5f`S3{BbLf`-S9nLOhqwEB4h)e|z8~YfzGy)gs%enDvmdicR9nVZ zsGOKSSBJ^3BVo(DsbDyBD(i?kNBZWhVam=~;1eH#>!1 zyW|doj!#J6k!i4_y^u&eH6ZPZy~MMomxRi)8PIbLz;mquuQk3<=pGDfT~@Pwo?&uj zPbA5H?+s1q8emVL8^j9mch?E>uCbhGX|z1M1~QEL*>{6Qwnk?0tAMaCTmV7gujU zS(ZO+8n*%Ug=r8P_y|C%8dMCIaYNO$xL=?Dg=3y8;psph2^oteVZMeWZ9yq%j@HAK zGG^4TaXCHTZi7G$JvFo`X>jaQwXX&}j z*3T}`!x%yc%Ut1|UkmGwA>EPvj4Y^TeW9Wa;JF|MK55^CRXA1n6^+VkG#_WCvU4|hn1Lg)i%-8_R0uRXQ>KaZ~(?QPJ5@mciB^G-g{lsrq zr1)2TjQNJUEci9ofnV>jn%`AEi*FIG&d)#Ciph?X_^qS6aii!W?E9;PUI{>>9Ffzk zo=#TZPhi@fXlY^pF8oQ{wSl8hvsr+v zw&X9{H&^6dKk*!_w%>xIS|?zWw*_RB+$SO;@|@oH?VJ~)l~m+G3R-_EMCAq5ST=bO zKZNDu@FZvSO=`rwS=Fc{ScLOj+c0o#6AEzS@%TbHbaCHKHJ5XU(Ag!Bf3y$tfB2)q znKaB3GQr2nmyv(WA+J{s zkH&@J3v&U!)w3q-$z&f;+1Jsjz!?|Z62SUuU7US21nVMHK>J(@zWF$ZCS1EjBDRN6 zEB72~U@;Y!T5YFJU27oK+l(abSAZ3pwn1iJHXMl812fJ974p7Fs|6j^-rjL~_19ISRERpNUy#1f;yxW%H$FP}qG6tbDJ)(%5v?=NSRZ zcHM>jXZf%w5@4fZI4oUPL@Z+O6UX*kQuCeVAHMm@KeI6+D!xL8_UOJmL_Qq%IY^v8i@+K{+_V<+gueExo!)93Yz1D!izqf{tte;osI z>J`LFfpu(d=pdhtZiSUa(?Cbv1`_IZ;hT><^%;A@dLtX?-P<{|TK@;Ve2L9^mSp3U+K#Gg=sa`ji12fnMriv1GM)9X(*eO3YGTqBrfcb9zb z2q%1hc5l5~1$@<8n4`(_Xts+q<{k1w-|07T@`_j-dY*;hZDT0z_7U$dljSRJ7UVz3 z=|ai6E_4?ChnvDHu_a29FF!L9E4QX%UCnN^6~0P?U5uF=)~~WNPXdoe1!Ao6etc&c zgHNV4qd>e9uJF8q7rfu%ZHG{F_dACzZu9B2YPK^(*t;$3h$=bBqc>-z@c!GuA^wN{ z@XVThc?uSD>A;L#^!q~#(x=x~Zd)sjdU465s$3s9fwkZv_ZYVPU^y+rFQDpY9*E}c zhld)0>=qXv6v zsPh)Q6tWt3^WNacjZ1J@+dd4A%%yIx*W$|lC=ANWWc=E%(0jrb^yJjp=)!gy0&C=P z+t4$n?rWW;iser-eB%bqoFsqn&k%dB$l;_1Td}P{5xut8mT#?%C z@!E1ees6z;p*XCKZpC#4#TZw62sfH$q0`?u9B;A5Ps7LXayAFE zJGbMbP&*W!F+u%(+Av(g4=1^9!wcf<-2c>s`Q5XrbXM7Yx>j!qJ*C0!lHRN&g^?GC z!*zX5%kwtcnP-6&cl?+Gr&#}4#wk2{-5Nb#ZNdA8`swJ^AYAnQ5)MQM;z9jXRA^{o z&$MHx9yNfu>%L=rR~Rl^vZB)57BzF*Wpi=s10PFDy9 znpBbFp^xZ9F+1lpNu+gpNV#k_zzZfLue^lGxgAWh3oSU|l|_~vn;M9WXbf@tnhq~~ z9+2(>TbOX}E3(R0lPWw-N23ic*!*(@RW302OsEpMS8Gs8>>h3loWvL6N$?ATe&WHr zuUI_ZgFBo4ppsY!noX)k&c<`NpurKh=bJG3tJ!;R&vZ^vR~;SdHo$7(MVwF9M4`yh z1~j9mL!yZ#TsdC{fo-A;u>^HpBHosvYGe1SU54o zpZxBRAU;G4odPZKYhDH}HtWFH_DTFJdk+%0zh-oF<6oDmW)XIu{oAXP54i7d~@bDia%U|UVmcI z*~S5-Cdc40&3sJz&jnl0?IJHv1v5TcTBM^_k0c#_z!TqT3x0!ZiTv)RP^IP!Nje)K zt73w9FWC=VtCQfEe;%S8UXf~k5vvN)PzSB)SmTi*!-O_a}%M@G| z#8gRi&~;hGbc5GIT&Z}0HmJ4ja)!<2Mw<8Tbh+{?pepAdZb z=PK^Dj>nDhU35~tJ-W#R^;lXXq{kdnR{6#cfs;*5CmGxIqa>AFn4>V0V2?cE&H#u_ro zV3dM$Ha!)TiNBT~!fAF2__QbrQwJyLiLmc%Hg7W)%$b1~BBb%miwLalyNowi{ib`b zWH8s2tLW_-EBJHR15#gPgQ{)?EOS%kx<8)JRmz{sRi8DJ%N3Zz?TokN?ysNDbzZfO zt9;jnd+?Pl_k#8wZjg;RSG;jPx93(HC@m_3t~49CFMp4IGDv{1ceb!ZpaAOLM}hyc zP&j@|1U83B!*`7{WbX$7OXaTLq~Ss--O_JvDW#@_uP>X^flWKmObk#-cmpPjiDGhx z6ADGMd04X`W}fm>daEjpuKHd^E7{!A`{G@YxLl6R=~@jnPj^5_O9Bip4TtKqO4f_^ z6hif0!s_$Z~)_o1IS!WNFCu);@Ig#Ya+h8)f{XZfq9Lf0>_Jc{a>7|xp zl9;u;iiYm7#*Z&9qo-CSuKyZ?D;4U|qb>_4#O+a}<0dX+lJILm2r3_G#R{DQEODEU zm0t2F@A8JGwq-D{o<-5;^1<}N-#`j^+nJT)L3mI|h#j3@Cg;>9(IOQe>aA{vmGSA8 z^fK!+fFbf%fFBe*Ywe+Wg>XaObhoM^ui_IJE*9$AyvGWN2@p-{8)7k zx4n5z*MDQ*vn)E=a&R#w7>DD7jA%UZG6P3jT5y|X0A8~X$CCw#i0;R*FkFdTyB=q` zpwgMvnG3+FS#yb^Uo?Di4g@I?CwQkF3O<*2z>eeV!Kb`~Xq?rA3;Q0S2WK)ru09!M z(g}XB_rnL_Mzn62N5B6vz_`XFl$jERp=p8GIV%KBhHhf*-79#`U><6}%)~~2wkt8- zN_QPlML9=3x-B4+_c&z(MsJ^nVbkTPWbt&EHrPjk@;zZW+j$oFkO<4&M4;rzRY;A$ z1#VSUMAM{=tQ1&D9#_Xwxo0^{m>l~yy|4zpAE+a4=Of4zZl>j^YARi{#fbhMXZz>_ zk<8~sC@_oEV3{&9j4JMwZN}8pDh$~F4!0$?&zjX-Ji*BF~Ly> zUksX)jN)C5=;GRpRcc235>-3CdGH+mXv;9Ra9`s{`3Dr3r9hWMB@?^dn}mH?!Fj4O zi>S{pC)YQ=;@oYD1or?X*c{LRDgKJwW%bjz+7fcyKdF!5`(pw-n=QfZ@NLRlah~Nn z1XAax7jaGDVr&wQKz*YPxUPgj8Pj*D7beXg%oF5mWwqdj2re!iDWj+C2ACfjsYLzx zHawoqp(!7`!Rp^9uy0c0HvJoh?xZImy-kzrQoo9O9{K%;3(lP^f6s4>9NHXC)~`7RnfqS>dUU~D z3n8xbd2z1lyjJ+TI1J*(zd@Qx9!zKVbpP2_!{y;hh_VcVv8L1DAe01=)fr%I5d;Zg zKS}4`6QFR%hh>g+6HT@&e7?Man3wW+a*b((X%L2hGwe>zHJ1(&7pj|d6knDU;PWqf zIFa(3X5I0p)?Zi96>l!0v2qFS-18BmGXLQ+i6Q*eHl6=R->`R(W(q^a;Cp8()b?twQbB!a^ zKQx2Ap^Kqr{5$FCW$#YAevz=hI>cbfS~&5F^~-uOB;LiGoejMtH%;9wgRf7aoev^8 zP9>L_Tbbcx{;_XYZVy=jXKoBgOgO;2EdY8& z`{2~T5w?pPMYe1XWMb#OrA5D#I2!|`FhyZIJ<9fN>@2m>;ou=uoEC!FZGqT&b_?2n zx<{+HOHj}8FP$Oxns#NaKq(_n^qjH^=NNsXUo4841)&4^Zxm#@y zz3q}vu|gTN3Y1Zkw+D}J$wFbhGdOrY9YZ%IqiF1X?C$u1t=n$ldpj#^jZ(%hJAae3 znKz&^a1&g?`|!pl6vADulDZwWBtmg33eH!@cS?WgE)vY18!K@7(kN7(??`u<#$ro2 z%lUDf#&?+}!U?$Oo+(?y)sYI~F23f*4RM^o&5f>vNZVI1`@bX@3iuDM%wJ0l z+w}0zS9QM7`4a5g&E-$sr^?rllHj{djG+WK1zU}#@@IZ~iJKio_}<&nF-g}CrKYLS zmGL<=qWLG2I_M0V9_k#uf-2Y+bQTt!^8mTgcjRbLD$M>6594z7uu8-SE=R@F@FNYl zMMeWPpWa7hn+(LsjnwXWCJhE->h~>}-2L#CV7oVY(RK!u_k_R*!@A^74}r&x8*m`v z2at-HToJ>M@ZyvNck}HO=#}42rjBf8J4h^hT%Z=EwJ+hq$VvRW4c*wAsLpS8kmi5U zTFMt3-@v!&TgnejQ3RdrQqJ$xP2`e-AQ(;APinmrNK9)hvC!=%FT`1P{{uhZ$h2^L zBp;LaKD)`Qo0-hOAZ3h;oN@op894X60Pfso#!P+ioE9y5OU~@5AnWUv^L&zK!w{=Y zj=G_PgXeOn_VZDieB>ovU%nlG6i(1TRd0D+``(ey?DOKSNhl~!A0=f{FGyho7iy)% zp!dXbm|C)i9Jo46_MQ_3-L*S7Vwp=xyu==w`?r@)n{AEu{yMbho)xL(S>jPu zN)O(5d_@ujDv6QE;4+VQccSTaNSVoIZ17Rw}(*be-N5drLQ@veHzs)%b6A zELwd`#wibF;eg}$~}@mmHD@*_x29@ z@_aOTt2Yl0T)GO3=N2aD+D@oDunpSxoP(}kvCzLw16JzCK%|KyWc*5lY;=Yd_l>}b zovUBl@roSDm`R$mb?I!6HqO6!LBxRlJu^Nz7kCOg$jWLf%T3qw$dGU)>AoQYQ%XAs zFT@69SeO0E9u=q;)dAbA9WZ8d4b&6ZdvDNnlK4p$ZmyjS8;s|%XOT`K9`6T)ju9L7 zn!I(bA;g>=A|mnMIij5@*?_rMxyLWd$# z&`r_+SATNHIjf>EwP_yeO;yA%`3@v-ND5~ArNKaWE}S*r3^$M808O)RklW@DOlAde zCucx#!WY=Eq7(|B=Rt}`6uf(M3C?`m2_0oVpz!b<%V5@lj~*kW%_*LE6csQBcMNb+ zC;T`IotKb`lcpTi^hb1upb6GIIe-r&6`=5*1SqB)f$I1xARDq6re+611Ish+YZijP zeXAgCi7?DeISoQ*X2QYo8ZbKl9vZf80?Sn|p*w0Cw=GtRo87d6d*b>`ZdaTM_tKyl zw?JE&`!s3=x8tBT_r>;i(DYh}J8f$cj5NO`kKbuR?hOGL+O`-{*sk^wx$O|QE|!>W zTL$XR?0X^eI1JZ>l5(R3R3SsD-sR9DVqVWbPh>>#R>v{q$=ndZ)x?IFq@17yN@C z!Wxh!mqE@xJwnbrx<}m~G}4ax`?%pwAo}`W$G!EmeHBi9W$62tll zED#Ls`Rn7^UWlK**$Ex$&0Cf{b0Am5s(@mu_4 z_;H?c{L}0%W#&~$KIzfpYcOx|mef3SQrtikjpIPU&H+?bSHqf?N3e1V1u+9bu5NT2 z>=>^T4#75IQ1q5aHNwXcO%{cLP zkk&LG!pBD@=-rgR_42I!lizJLxn}z%_A5NZ(sAZI64wb6B zPInxCM1r@d5+AP^BK0x=&YouHiG#V2G*Ae0ekB0Eb$)4*>37pILj||LF2Djc( z5@m9Ycsz42pVIr9;X2;;C+CDgk=k;Qe7YBeHr|3|N&iSnYY6;E+X8*xSwFyyGMGMhJ&~8V zOT322%M$vR(D{J8CKtZ*8n-zDP5JPm_hE@x=NT%h!uifz>WMVf64967)6ii2Apum*vz33hQ?~pQ*IP(`#6PirkT?peG zE_G$^eTR8da=sGp0&Q5aaUc2WnoFF@AMy^3&V%nK+n9lHQ>qv{1K-A|tQ zI}dMsbi?abVc5Sk7stz6@mb$@?ECx$r@tS<4-!w%*0c~kecm9|D8@q-r%~mh6U-VD zgY&VjoIf!=q;Juv#Yi#-tXm^t_0Ch!Y-X8PTAzd)p8@-*gaX0roCUO^I@`N#}ZIH+1f?uh- zz64eAXd?M@?vZwhjnGw<2B(((BHG^;g3hN=;;1_tCe-E=;g%y{n!62V{(ViZ-tvPl zsVZ=EYXb-Vl`txIt@4%nl6ivpI< zWjId9!tm#-JE+3;Z6fAH;e$=b@uTT#yrS-aU2JE%=}iKr-cCf-e~;*;w{=u^!3O%u zr<9tVYNd|8W+3_CE_u?=1;uBP(5!rrEdCM#ADAy-w$tOv{Prm4XbU?`a}lwrmmw$Ugj~=BunX7 z*-vIfMufLiZY8<~AEVl5MNz;t2Q}nA<02V({(${9oPKpRDh@3{o0D#6ua||*%VM$Z zLMVU1LSMd4%wqnn(t2#tbimz?3dHtzIZY|KK|URK6*Zc)a$|&=0Df%I9Z3{ z-Ox$=b~eMd@}V%lV@eVpD3j zi4ffidaFXA;}+An5taK7moL5_H1mGVaz$Cv#t=!j^+0jL6SixH8@c<~8;3 z;D``6s&6WnpQOvh$?c$b`X%hOjE3o>lVBO^+)=um4#u}e$oJ#RK>v9;j1BDvd9CS0 zYx4(klVv>zDQt$1TA!h7{yPv!D}xg7fc38r!KVJhAboHVsnBAH;8ja`6IM|UlJULGJyHfIax;L3sYe9B>4TJ1$$KYcD2jt`rL&kUv82ncOk7J9V zL9+qYjh2J3uot{*ItM#7%;AY>Gud*Wf-F}Hx168;l*DV?Cel0J5U0Lzi}zvcsk>}H zea$G~pVhT=j9 zgL&_|p)K`4$hkfOW2RYfz1#%$d=doRvUnn7uEp3W@-4H%!f4vtf6Qst{jM~+4o@sl z#|I0oP;{Fz?luU)x#H(A7C+(niAFZBLh)loHvS7{z3ZDlm#e7tkoVP^uvlq?%#A$) z|D)(U{JHx6IBqtKkWEU%E@b3$&-+tGLZU*GzLlaqH8ly@D=I51A|p}qx##^s5tT?q zBuXR|4JqyNyTAXy=emz`&pGec>-k*uZ8Ls+CkWu;rT5Vcjeq#ds zIu!|g-Z^+Xz6iwof}lSy5I#q%!h^99jHI;;*)!gZKC~KxrUjEQ(a9XEx27U^3}JFk zCx(C25Js133r~A|M(s9nVc5H$806KAtgECjDX|2HrW9bcNj!FIyP}Eg7y2~j3){|h zncmDAU{e=n)0saDn2=LNME80yQ9EJ)qKEp(Hp^dRpDK`#acVUG#{#-xK{}OgnL|gP z=CJT1jdXYi2zHmgqVmhn;M(nlxIn!KwZHPw#CsV&jr769S+#h*KLA6%rP5y-l$A~@ zqnFmp!eq^Qz9u3i_{)5Gx z_VAitLre}-G1neHW1XL^f|y)UNS>$)U#{IJtG%j;{OR!mH_a(%vP;OGe-T6H7R#e_ z%1YF*nvX3?arDEbZhG1|9{1>)Vnq+&lFmWu_--a{j$VlkA6@WWr7^CHUx-Q>&nRP; zL48gw$K}VRaOATA-R6?UPGjoH;=d|TJ3fKzq$faBF%pt?mcz=BCa{~)3V*7*IUcbk zO#U7Z%=YsT{<#!3m8(Fn!wcrU>2)sOQBMXoq>!7Dk<1ugGv(E&(r+AZ@Mw<@PPnug z*Dbh)L~}kSeVmHd8uO?U&kO%d3#Oi%)N%hzCwy?{II;G8!mNDqo>ZGJCIkNrz*8$2 z*6D464PwXP&WH}2(w7BE_bzf{z!%!3_30;$(Vi`ESCtM@pm)=8YIq(1u zYc@lEQx6o&Xz_g&Rrs&Q$n&qNi1HQw>xF+WQ{hNdAd_gjgr>dZ;i1yM^zg;g*cJE) zm+lwE4+nQ5Z$$!qIsP4;le&V-QSYKbx znoeFM<)v!m&Dj}L*5EJsA-9n}&8(-1?`=t2co?~9EKZtl%pppr*ASWWPV7wGbRwL- z5&O<0qsMrCl!~dxA=dsSAxlX)&;S{!(g`yaI!_wPlaoeO_Xz=b9t@(L@ z*q7cQR|8c!mWTsfUM2^=1%^7+;P4DNC}dorqd@?l46NZ{(+Zj~ zMG;qf2jR8}1~jlU8|{_!G2_TW6uA+F#$89TakLh*rv>9CkMTG$>Iuz$_KT7kl(tUt z!ioo0Shk1b!JjN)_CCDK^(1mhb8sfOeaRuGoWIaXuXX8GT`jDM@n%o>%F~H6?lH&j zX+Y%iLLNW3ikaj@iA;tOqa5&t>3RE^F!;^a# zAy?ugtdDyRALJ@wK!$}0`gzdHp9>Q@S#bH}4_RE7dBe-aP!=&C3jSJy$+ID{ub`cL z7s6~@N&H&mb+|6n0XS(${aLH zRHo_Wb5Lk)Y_+jBkcO6wB|&B_Rl(dxQy#k-oltn(5=+m;U_|e6ToHW@vuyTbX;>V}eXGQ5-K*&IFb#b@{V;MM z8Iv@YqsYg*T)$%_t>d`E^QY+I#BqA`hG-~_$RFhS%S|WBd0_&v`BQMd;XA5dT}eBF zUeIldHIy9=!nE7m`<`Qo=Ui@~Q)4dPGTDK=6Q+1gWiwjOIe>X8-k2hC3lGn~it@cL z@ML-`9=XIZvh6Ey?ccq)I#3zwY9#PQyBnpOhL{(R(n+tADak90BrV}0yd6a$#N}Ws zIc=;-Mu(@uJKqhkCH@fHrKw!M=qwyp)&cQb@}O;vWU&>2I|qIfyU@F2Jfi|?N%+WONRzi|7p2T@;XgL@l&z4Hz&aa%*@na9%O z(o&e4ZAV{(gp){kMQ_pyva$0Kxg#PEwx#2sCjA#VTU|@KlZ{Bv^Bp9=u$tH;UFPn( zN(t+|jXb}l0+WV=$fWa&;L~kiSnm0b$fRCoYVF6;#xuXzRT?9#$kjnAUa^&$e72?6 zqORf-gC!WnJ+B) zH7^beuDD{$pIdZFkq%yyyu&)5`^^(OvKv2^ou-jXO4yM|Ip*P-G~&5$E&N`wo=i8g zBkvyOkjK3`M0C6u+%K^OuYy1*Y!86P773sm-wz$lS77VdbkOlT4d!c4L)~W^I2&XR z^B*`tkn&>q`jLUrMFDW*G=Sp^87h`?5WJN7iTA<9P&jioltlEgu5O(~cru^XJ{HC4 zQv2wxmuH!ke%fgD;|e=9QXBT38zo2hW59aX9{T?3A3FG%k^tpc+VGkq{8e#t6{GoN z@ytzRd}k~yMk8qWbCxV|K1yyKJx0#&;ZzH+ta16(X?Xsa2?{Un#AHKVbnp+OcKb}| zvZXzOeK&uQ6-MVkJ-`UwrKS+&@tR!EWisSHN}=H#%e*8+oKE#LV-tHFNqUoiXkvyZy^8 zIv|w7jt8Z@w|z#qeVRNrIcVXyk@5J%RvZIdim6=AI?4(@Q`z@&jNZZ#L46F@|J53z zyI>PDb<9DS!5;+UycXzbli^+{Q#^!OHH zs(kOK!|?O7B>(3jd48pDAsE{toSv`}%=f0j(AYJQk@=2Qx==}K%hmCYf(phQ=%dGt z`tVlQ7W8b&$ByW~cqOJ96~^^*xqOOgm8-Ep%L~IhwD8c>bo$ufH1%1&jZQGA5WIYG zhX`hgz(zMcm}`CmqLbp``nEinZ{7+y=i=bVBW;iuM1dZ)ggftN(XR?qv2v>oeW%4` z7p_VOT2wzXFN}1pbaV#UxK)Ehs_GWqvUnEi{d|P~{l0)^aReW~;Nk1X8uZeu!}L?T zC>&Ut0}&3_V9LQK;6Cpfyz421kR?qpu(brBF9(cLvLT+!{F$zg1oubnko)#GxF2Z( z+Zj!8RsI&3gg3*lat~am9fNm#`Z4rEKe{*+p^bArwq-VM@4IkGVwa1;qlnwEG>zpi_g$$ z{9;z_);GFMX_US$dr1}T?$ZQrujZR-NH-J)P{}qk?2t&tp)T(3vV9X$RWsy0jiy=I zWAI7k9Q2)<%i8MqP~AWNIFh^%pK$xJ`_%^2GJ?A=>2$+Cw&CQD=+h~KuJQPU_!p(ov&^^rtlZTA3A$2!C?+C<@ zdjWW|Ee6xtf8a}-*Qjx%5fu&z(0z9aHm|b4H;<2^%$5NB{y`eWC-E7l1E)9#peihr zy-rmIRqzn2O{Mf#us_1<*`i??Fl>zmr{5inpxh6ZzFrUaWO*b(*OGV_>XP=-XS8SA zQk>9hk1A4j*rO0fv|AoB%H{g#pWMrSKUBe3MBXDMt+8O5<`26otl;_n*Cc&U0u^Y- zlHzGb5OQo4@ZGw}CIe#{BYlQk81>>jg$JR`r7!kf$gyeXsOqn8>|b^@{#t#EZ5{i8etD+{g$q0(z^RUy@5lv_85G};RGk!rySJsHri}}6c*EU;m)Me;9DNA8JOMssJO$rqZFu3(auQHlIQCRTfp)8@(2oKiPx7{Ok~jG+y7i5SL}1rF-mN(Kz7lna{l=){d2I z(8x|A^)Hcq+_izesFfg(dOnf-9l;ne7>spW8fnh=G<;k>8$}CCF)uX>Yu&%%@xdP$ zwEQ97S#b+1aDp)AP7AtvOcdVFogvKZ7$X!`)nZY>T(sjj_R{%FiO(->*pO&J)^Hse zTh8J7zFHiH2du!na2n}#T@BhrC!tEi5q`)z z{FS+L${c*lZba+eQk=#4)ty>yqr-^^^l)8))niU#p!7Go``ihv_~wBD(k0|;;Ty)_ z`$cjlVmy8MMizSIrjo9tTOihw1;6g*Lq$|KbZifT&&Byr){qYtVsGG6T?%CDR^mv8G~#m~u}&R3m$AFbA3z%K8n7~(-mg1}Qy#d+TQumxs2n;>goiA$~> zMg1f5@$0nHxXWP~YR6Zz%Vr!aGkxYn{x-o4vwl|QBD{YiTW-0Q(VqlaKMUOKIsFwcQm14YXaPH zjDhhpT_Lf6!2A4Ms00(J_A`RLUn`mM$K1%0@syl+kw}+wd;5pSwIS-r8+O{5g>;Rv z4*l!-n)(D?sNfFvjOmkJ>g;oz9`0Gp?ugt?#aq|l+wO&!Z4-f>PtKr(MJE1ytBA|> zwXnSC04r%^!=4YF$$T}u!Gw>rkn77Nm^q!t*!R+JSV7K3X2UBB)Co>N(}v?%yTFew zun!>?w?c^i+D!UNYb+IZY^OgZ8W?}REt6ux5{V%PYIw+|Sjg`6MI zO^tCh5QEL%zY?4Au}rOkF10xtPQJX5hXdP8;ZnwZQfIus;@YlosyX*NFJ{kJTHwOD zG`a-r;Dl(pll#6-KCOqfD&h3O@=J8LR03+02I1U=_4LoiT*~Ig(~7PF+Wl`PcAsl! zL~i-v8V6te6Z4FooSZ`k%N21*^aRQL{GR;h!6)~0t>B~mcsR4>J5l!E0E4wz;D2>B ztg@aA@7^DPnjlROliWjB4p~y~Ur5rp-ky`&N9M9|36YPO#QIFqV>)tdaj*O$&Knnt z&p+B>Zn_Wl9?rrH!3VV6Sb`PJ57AX80k^)Ij~9xZXmz9<9??l9Qw;~0yhrOHThRwD zzIKEqR|%{;I}N74RU+xHL?E``5MrWYh?A;IPppb{HlIyX_ zVh&n5=HSHKU1<5k43mTW(02YctgZ3HGr9?=cSQ#;v#m7jbO6n7{zMzHYN^VH5uVmz zU6|&`gYC|tV1DEz%(69yT-OnD@NP9Jw137{KTc$nx3;juduyri+iZHM@euRwVHsVw zTNZ5&xue&p2UX&@62ZDN(Tv;gJg%OO5m%BqezTq6ze`)_AIB_4-8GV2^O{T)%MX)x z3l+)rEGbChSc&~U_E5iI1-x98#=0zQXVqfL1WpRsH1y~K!3R5AOmEtS;l_z5`^E*s zmt}Ko-w+h;3c@IVTRcE#q5Di_oSb5h1(h{asd_HzHVf#n&R2{?OCqz^Fv~Jxas*kY zY(Ng1Ix!*=?%?orH!yug@FOH0;{T+;hkNJ1+uaFTdu_qvt}8jNgO?&s1=pZ5dr(bBta(U&r{xl#rtJhVW?`mt886CymWA z#IX1r*>9-^d&9-4M|UF4o~e$QriNX2O6Uw%1^UfmDP3c2igSxM;(|hJ+$~~?7k^%) zx4+wPejfuI`&PiMi;>LFO^b+m)*bfw?tHSy*pjX67$UpR-w^Z%JYXL0y+SKKc2bd5 zwY0~ogwDHQLIcd;8pr9ZW%c~ zzl&_LZX%!9Bsvp6Gw&x>u&17cS`A0sr@PEP(lhRo*g4JvU4KoYhZhV`xsMif43Q=` zJ2_73#dM;eEKlZ}=aXGkflT?@J!G7e4#<943zs)-fOXG9U|!2pkX?NTtYa^Mt;YjM zn(+ZfHi_}?1&-lwGL++YU-=BmyWhabtUx2$~x$V8IwC-~{fV zKGI1N&Xurl)n;Q!gAsaUnIqpe75fr{ao?RFRQSu?8T1pp(zOz|aOVLI$cN|j%y6YM z_f1KDM*Yoo>8yioRt5GdlyNtN_wq*YK1YODwyMG9P#F;Ay&-e*I49uxD2T~l1~*+f zzw7sEQYtk>$4**+r!$aF(-+X626>bmoq%)m8fn>5aVpc)Dmdt3OSDdNd_S!_q>AG} z)q_1W{jsLQG>nw4$fRG))**9Gz_h(_r?z5&~If1@U%7&-oOQ{bxj{)%~O6 z4$tPgs250Hs}spCA4?ZBsbb#OKXiLAAIEiA;I&%Qh-z)%FH6sDtc zm>P;K@xc{>z4+REB}!CsbKYOC=$cJmX!dPK9Fwz=YUMv->3I z@8wS#5;PMJ-7Z6!LND~$^BiXx2S9+R4hp%rV)X#^+OIEj{1XL_fK-^ zSpwW-Rbcv<9`bsX89?t6xOTn}HvXLiKQ{-##vhBwq_Vfn6ppjE;LmNcEmjvgDvpwQ zHF4Cf>86$Kg*rC>ZVOfw6k$eh4R+Vt5YCCKXMss_}cJ8iIjhqJe;{nh_uR_y2 z&Y3v50A72=!zKS*7%|>X{N0vtyy$rl?CB5HUpD}@?1Ms~IFVg*hUDvO!w2OcxTO$G z==%yb=}#Q9b2x|w#LD2>H}*Kw?gqDyv7k?VWbxbWv3M?MC3g55p_X$iU6`mvZ-`u_ zg@cPQ>%bTszdscBk$mKFd4Gkk0_?w-i3j)X#(wTh+%>ljo#*AE0?|i3OC4Mi?LlYk zctW1(OG0kFGz?jY!k=p{VCJ7sOgy(Rd0D!!`p{XnVZtIN`K=Vr&Rl_;+Wt}%ETq9z zU95lQEK*)ABnKvVld`iSBx+qARW!LkK5jaOhVNo%1Me`kju?aYqBdR=uu1VND-gR(1)C-5-FrIldkUX~FFwe~ zUN8XVtE!-I_8ipEHu!Co0F%R_pmI$roZtNxl9m*JT07;YCqdwr=L%yQePFzX3*?QO z!@E)iVt2ii^`3YJFYPTxd+yAg9{Uu#R-M4X2N$sUZz@*w#p2gEU)=6C6(=t&#{7Mk zaenVcymEa%nl7|Jv3(s>)2u?BZY=&p z^3FAol(jkJ;QYtT{5?S=Q>v7B{VO0#9A#naLorAqa_~CE9TZ!;$e46Jn0N6A%zK!~ zt>1gV>)}4oU{=EY#o-{n!-F{LF$5j_1fA`>$$`@q)T?0*$_^+}@3UiZUwRXLyiSE> zcfY0!M^k8%o*`EJoz2XO=6-|dB3$1y2`yZeu%y!nH`S@($W9*BHH)Q_mY7i;Z8DA!j@WfJlms(D`X=$UI(xQ)f!x8E*DqdpZJtw|C>~y@B{# z`7rK!y8`c+?Q$WU#up11^ej^ zM+-Q$uZ!9Fx{v7Z6@!h_Bq2y76jbUaf}`_kIBQ-(>?YqJH#$OK`-8{iD9@P8Q<@4c zwvON+v6Xxj)o1P`$Fg6BHxq-zR`R1Ho|cE0(lHO!k$+JKbteT9#b+jLz}hV4*uEHc z@8qX+WMT~+yZt5`wC5Xr5D-T_^^4epNt_?ZIiGHnKx(x9KAnmZ__b6Xg%=#?xFw5e zkn%=SR(XfqG&3P5A|=4$$0PFX`!r}=AkL&qcoXr-6~terh6XqVu=iF|S|#K>U{w}0 zcJe$b{!WYFLK&Q<wbQy}MD0~+skg4aP`P+4XHpHkH!VWS=B>r8}t zj~ndR?B9&-#2~UZD2I+cJQjCbh~r?Z61gIu#0Gea;P7&J^oX%Xmm8KSxp+3t>9a$h zkz%@JMgY~bl0>J`nXHS+bnw`6k=(g{kyzV=lQ;9GLzCuh@?$YK2im`a)b?^73ohLj z^ZhpbE~`qIch4mif1~KRqTjU2 zCXimyyT`s@66pDRLRL1~m&{FCL*MW}vIF)k+2AL~&1E$xhRLFrgDVc6NW{e_uHla`Ry+xl;ryJ*lEi9r5Sw*#J+2sbMvLXam|1oejbrq%!N3%6 z&OJ`G-mIr?rxQrbEfHeI*fN%4b<`(E4}WM5(+CYU9Nc*x?WMzTaJnC!YZOQE6K?#=pKc+bFS0u6YJ>l3{7ggFoU{!h;i$M zEYA7oiIEb~*!*}g1}}}o^FNN`L6ry;*GR=Jxpz6{<#klw{{d%?l@Q9R)!-4&B9u^0 zz>=3m_*9*9QFIn!cE}05_O6ZI(#WtPO&rr(dMi8cNhCYAQk(d#zEAppKO(N1)akTF z1$sCqihQ2YOm^MmT*JkqWcRMc;Kccy#~yl1l7e&vI#~>HTzZLqU$_W2zfr)u8afy; zGzqKn##r%2Zt^S|%juiphpgv})nrbBGCV5{g0{$gAagwiY_i>9Ys0EkSOafayM-Z=QC*ym8IM~ONwFPBNb-xe&^*WY3cR7f&o|;oLiKW!< z0cCFwjnEbT;^??&1%0JG1$T6Qv(oaI#7M5WPM^0mlg_GXTvqcsXuVAWr!Rp}!DZ^J zw^c%a(Hp?e5#W-4A5w~%z&y_rd__2n<*dbANBA`KrPPyCd$i~p#{#k`u83$`x07tw zVw%}!PAjv@$>F;7(0BP7lxn{xQCZ8OOms0Eib{s`=BJR{&<4>xp-`;98W#B#!S19_ zAd>$ChGk!Y?E)$OBd-+j3ekf0V{uS-yN)!d9RwrKiLlWAKR6+_mCI}lkgl2-NVGfy zb22UAOHc?b*$@m$VIeT8lL()4ev{>e@+4w`2P6+|fXFCum>#bS_j%R4!htFD_@V)F z`Mww9n>-) z8#qP8qUuTezzCDS{smLV(}fdH#=z#XrSSLIad=x12$I#)L21(fy-SQBKXe`}O~`=5 zJC4JI3|q+DZUy2?n?S@i3|#-)%(lK>MJ)Hurkk`iP`z#^M7ke@!KjO1^6(lseX52% zd)4_(-DQwcK9RppFqVI#UYx)C!C%NY`~>V*eu3la=V1TjFt~oNll1)_A+@JY3o;$W zvH8_8)`sb$E=v;Tr0lgJOBGB=g_#aJH?4;{@TTIHiM`bOmKnBh{zNCGcyk=QTlAx^ z0l*=7GRH?ie)x-mdU_I68*c^!xop@xW$Nh`T^b4z!V>@iIu`t=O^p&BV;-9P|x{^6!)nLj0*X*tKvPp=P1< zS@#<@@=XrLX|KWhqrSLq^)~e9vM!@d*YQy8T6}F^M621Wn6zdpKAj_mKEwIc(pv&= zCAiQ%UUR9_oOpI@W)T!Qcfqp2DhSoy0_&~Az;1Ch=-q3BuZ6StL7$zhtld_l(k)Bu zIysAS$5yufvoZb{{6;N9Q_Ee%=L|8nP2^ zk8^}8{!4)69P1^)3#oQzDsP(7XV!WsfSs2m4V`r#h;wAL;K;M3q(<&6>gO%Sxxr`g z>VxYzUx8b*H*UeSW<`{H;f9?V!8p!;4EMLV2$ED~!2hKmV|(c&RjXJ^rWnm8jEs?B zeD4IFNkA>V@i73$u5)J(><^&t6mL+otYp@*FOvB!HU&GAqv_?@+}vGqEX;ZW&}114 zp9Pa(Yfk}8-ct*5uQ=Ayl5wE1dOGBc*MI|0gUFy!7ZH|A)9pRcFu(l_Nc3;6iesV7T5u1uXS8u{^n1Y{Nim^}Z zGJ12&zU`{dvGvtGOyY8=6-`RQ*UxjY?bajgc(EEMYZYKi$W`1vWdt`UJ;!CK$fzn4LiV{aTQVR{X`609)MlTb`(DV~lPFm)UHq^!cCF=x56TMS;W%D|3+{TO$kgsK>< z!2Fll+zbHF_n8eQ+8(1dNG;PF%uNx&0$4m9Qijd znrO{3Bu}J(7?$+Y(M8K~)2d2b7Jr?~Jyqkn-}f+lS|a9755*^2;?T#7#nwAH$c29~ z=TjcuO1y%L9|YmT(iwOptc%`EUj`dx8OLGA9%Sr9%Mb@D6 zIIl#5>z)Ux>dR7X%e(T|z*Kqzli7CzhsG@ zKapM~27e0{v1RfNWa=Fi?7OgtDYX8_s>=RH+l;~)iKR_s^-T*fi5my=hng6j=}Ekz zR%vn}a6LWpBcE>dJV>J<9CdPssZ5>)78veCE%Q)JbO}Madm-lZOJGB$HNCuW1MT^3 zh+3MGwC0N)lQn+=*%Q2kILnr@J~`{*V&!Q#bC3rr4#lAUBn95Y*?`5&i|m2}A-sNv zl_cbi6;z*h1Em5R*qP6rX-&Mz&iPvzS>G-6*29yutV#l9Dud`hjuHO4(TlyaCX2R( z5GuEGFPVD9pZxcF1-os7GM)eZKDmCpjjZO_LM8o|5GnKC-ca1FJRm53P z_lpI-K_|z+E`p&rj)N{600;AzK$;z<#G&LHsaM?z%k`!Kzv?-elWPDwddI<~n(O4cnlJ4A zkIVQg-zGCkRU!5FAGTh)inVXwLqEobu#ciIlE^4U7)~q$*gZrN)WxCH)d!Zx+QEv{ zJZNZP;MPkO_|9cTk3KMg4L-gw^&rAl^&oik&VX|j9U}*yeIZA37Qx@c3*isfmAXAL zOq%3hF+;eYd^CK@G}PFTqZ|3ug0Couo+yqAA|`lo=^1)_hYQz>EoC-@xPsmfd5GQ@ z1G7HIL#Xo>+|Z>gY)_IAb`L+qxoUqo*ZD=G}@W6|pQa>NJf2>#g!L;LWTcn(%s zJ;HIK#kglyB`zNh#qnKxu=!y)ZZ{9ax7uTn2X)y0G8{ihtjCre3-M>@H)bUABF`jf zD$MR!2VG$a;HAJ$p}nv_wO%`E;;c9L=uK*-gAQ-yzMKnlRhk z2yT6g;Iejah~7M9605kK%q_adBuv|eDPbPirCLo-_n$+B$J~9&!XccV#$xKp_h{92 z32nQ2aL|==*;ZsBS&@ww&U$0MaX5wt1mlZlXZ(4-ghII#EL?nwoOROxl>~qAv**r& z1CsnjkL37XEx91+Jf5#~O9*BSQhb|bEXbYf0ogrVj@jk`?5t~q`WcsC=VVbZTjfQU z$#VAxIY~IOrxO3A-p0Ru1E_8nf~~rRIBQ%BdZDQB?LkrD!Qqei<3bWjbtTf=v_q(H z%9hUD;Y#NhhmvP8GUQNPJz+@=bGcv)cvcRP^Xr?4#MND-HA5e~R7IhGOE>XdQ_a@> z{EvQ8(8B);PSQY$RyrwJs=ukox=DC&kYhQ{3&WWk zd1x@@CAIa=q93I5DM#%T)UO=RWrE6ynL{kx>&t*j*=TqZ6ANbgC&5`W2z2Z6R_IMTR)*UaU2pV`EhdNgv3GH4z7%G{0^Wc3Whr)Vf*bk7+s$NNkgIfqi0(DSbt}7Q#Qoz`*X}<$XWu{O z+t?(!zU%^FG_|=N`E~jqHyd2(D2pnZw=nSNZM^yEF=qDHp>pmpc5S$TXVP3SdDIoP zUobqMnO~WArH|M_jxl>B+zt2wPnbG%5PGEYAbOPmf-L>vz)632qLmA7RpxM6HW_BD zUr+9@TnyL5LZMV=ko4X8z!TZ?fJ}*=1P>Q1Wc>qL$;^P047*X6{I2!|iw*0+Bf<>M z7rL;Y%nf-pM<)~cgA1W{q>H$xsc`Rk6bUhzjjf(hxV$5q<7i&QFA3S$J@6do>wo6< z_D@iL-Y-=Abpvy@#Nc2YV!-tzTyS+Fo|yBA8UFc{thfCpXe{Fs+sjJe{YnqMhOCFS zXfHUf?GIIJlE8oV0(dH-4u(_L0;|x+>{q)-8kbixC(Kl6&bbU;cd;S6d7y<1+3X^Q zzK2Na>`z3iH5mA6A4yLBB5vkn0XujrKqWUGDxW#Q!fBfDX;~Qj!5*?@?q$+he21`R z#sn_NlEf8@*hgxM$YiN#l9-u9E9;+-!J7>1!-)_lK9(Hl`$C>>xyfu35hu4fhTr8k z*~IRr6}TL&BeU&3a!zKBH7%D*!Za_!jRFfObDsbT!W1x>9SGBcsv&F|H)CY&;bC1M z{N1hqZc&S1YC50n&oraF3A2b+-wDP%NQ8{r5XD@oTEq0kb zVX09pB>J7>oVYCLW@dn2?p-)MArJQZ$8)?$YiQEi1G_upnJI1gq~W(BFV4P>{gR$U zh71x|dEt8aZTHxJkM( zzVnPpK%EhNu=+B2em;auShNUMo#w%@Z~m~&_y9CZmJ>@SH!yhV4Ef??An!pJnO`MF zeCP+}Xq*E$=6S(4PkFHCErhx1-Vl{v!f0q`l5G#}P&VZm-3?D^W)0V)(}*S&tFMzc zV!_1lm>rcgQ^r*@&$0XVzh&Rr>yogjJ7igU3b6(;tc+E_WO+V@)I_44#X<}Vd`(M6 zlyGyZGLBz7oo@U2o0#MrBXt>4yp;Dnyh9R41(|JKq?n&cl*t3qN<+!-snJA6rksN4 zHoRD)jr5ZVg54Q<@%%=dAmWEQT?f!LAO&TJGWxFN^1=pY=qhsw*W8$ZNiOs7FFOYP z>$lPtU3vCJkuDK9szcy25BP7uAFloLg&*>=aQ(+|7@af+rq0+138{rJMm!X<1lu4X z`w#h^o=E?DDWrR>B~V^{6DHa0#S?zksJ49_8ouqPQ7P`Y>_Q0b;<5$5y2@$8twbh6 zH-!$AII$g81#D=K38d5W;3vZ|R8?yr%IG#k*49Bl#SLhYy9(;5+{|pocj(;S4wt=O zz~=)q_z}Bh`TvPNf)TZPa2ychOLMuc1*73$9_k2H-cB&8X$XE1d7v$n197qAq~qd6 zl42=AOEYE?>)jL3vRfYCaz3~VkK^(7+9Yi3&_=Zz3$SeRZ(0;Oj<{Hul80W7^m*k) zMqK;`iT`;Pw&=};pHV`1^zaNAWKh^;dLQVA7&!3j26?|I6=pg|!Lm=8!1HT?(d>PY zmL>_!43`NJpN|)BGqeZxqC?SCREQU64~xcQqEMQuhJ9ppD`&&^jn$Cs7z)e!>)?BN zI2;>%0d?t$e5p(0_*|?TR>jxBthhiJinNAv9r@71k0j4aDd~G62jAOWXwsTgCUw&@ z`aNwWW)HsSJlq{P-B3&TW%(3gsq{SI<7G31?_0HnDZzTe>aprV4+T+Sx9=c+(HJ9K zStT!Y(2U3Y;#{=Yw+Eef{h}u=C!*n|zgE>pL~veqJI^6(5u}$oLB+lAuwd&QaG3W8 zHXDxt@jsm~TB8jI?tdWzxfg)yhC=G-df0bH9s-X@gMY6a7^&OCB+)`*9;`vWIXq$a z|NTQGZIX%K3N1D7`qq8^u7NzOFI`&FxH+TM>m zaryOJukjSz&%6RNcJBtkqHx&Wm;}xN-{7N(96wJ|iT}P>i+eJSptLF!u1mB*p+Y36 z3&wC=>=5>Xum(>^7ohvcS`>YzDHK25j>T^tpz4n2*uZt@H`VrGxN|L?`cw^fGdEcnmK1p1d>gKHb**;xx5$-w55q&@pP`I{gl zEv|b&Hn5v?^%sKVG-VRF(Unxr)L6Gxw~#HUT2MD=En zf;BzFtG$+Ny}UwTJ`~QIa9teM`Yxuk7ITTk!|Q08MKKfo+(m#}wv&GSZ-TeKEuq>@ z9qebTLg~PDkQ-5e?ng(VN45jj&Ub*0ufZVqa~D)sIKg-CIMB!!gP4l>@LRNz_(xu& z@=CifLYK=I*j>luMID&qXN&qfuF{ZggtTp|XL`e2VfKI&gq{F7@P*)1z zHsP7nK%9O$9XD>>gRZ3~cu|sK!mQAn_{uR47jdbP6X{_YvwoR&!KpUuV7UN7nX3JvrL z7e)QnDIEWxg*DxnFYq(l2H9yQ@K~pcY#Qt!pC|q1ZEKbUgIQ;&&M`Y2);WRi(-d*Z zsT^)LmXGIK%h7&^5WRN_@r7?PT76uK%m4V(MY_|mxK9t~re7pshmrRqbUb{V zbBsugYy-`aMI`9*FtcWy7Z@JoJeP%>pR;Hqym`}0W*kstuLP@rGWkjDrhX&;xrn2$ zyeN)nt)Mr4ULrHz=s-Q^30jwH3^0mE5>uU+-!INk@v2h|S%d;Z8>bYFQ zPIb$CxJ^5E?V{z)wbXt0UykAMfcnlw@7W{So&vY)$tHQqa)7+2jTE2_Vc zuy7slxb}IF7R0N?7HKjWoCR1&uppjyLYk#!KC1ILtX3-AWLl z;Wf^kAu7~0=|}YoSCBj1;nMN$=%>x)i|5Y5pgLn*to@N$eY1o)kq`z0M}lDA{-v;f zs2KF`{DsujAK+RH$CY%K;XhiW!#5dH;_r+H{?}>d{16u@{<|~kd>XCCk6EM4KbNb; zcbxhHY#)6G2+n}pHD-{^|3|WHltKHh2wZZP!`l;XQjMKkG2*{Y+`Vcoj>JEs8%zA9DN=8w=vts*de-8?h(RkIVH$;0Wi{>b#nZ9}-$<`#e5A|6qu&d3tClIgy^} zxVu{Mq!Gn3^0Ra;^59Yn#i6@l(k5 zhgD3xO%F|>*(hwjg%eyk?q;{F@cW*dxaX6saMdhF;hGp{Ve`Hz!vFI5!upS2@b~x! zcuKMiC&$LHSIt#O+C^oQnfQ@bhkjr>Y6H<=@eULjT8*-|tq_)-pl?bK(Ef%%`ZaMi z+L?5re}5@9oEt&Q^cT40KncDyO2NtMnwa;@4m%#+q%n^c;FDc)cqlD|Hg!Al93_5` z1)?dOcl8?Z^J*ZrGZo@TZb4!(1MOJ05s$KyQc1bF&3IK_n?4rD9UxbCmNM-Wee84cCBN@&{7EbB0UoYiMxU z4#v^hFj8LyZmUYcY4`{1-f#{&KJS5lJ!3%e-CTHCR?PF3lV+~fU7&~C4$_gmzws)U z-M+MxkI6bQY$vfW3~va@R1-%1B0a$u)STbBHEtTHwGsNtB;G zKvy_lp@-k?!WCP*@$%0w{Aa0&xAnJCl}j^m=4C?We-C8V$lYbndcR;i;wl;06Be}I zG6MhXFUP}0_wm^Yiq?Hog^p35QNQ30CcYBji4#$nKO>8C*>1vqGM;F3dGngC$TJg` zCo|vLqlw-3Hu|74i)J@;(oN^wIi5`?M)=7KuP-za8rHJ7uCo%Y^74@MoX3`rftaQ0 zh^+!H&mO)3d&CZ5x}Ff9{5XSO3sccgtr%a4m0*O205@{E>?Va&TCHM<;Zu^Ba@Ts& zG$WRq#l}+U{t? z(9saTH*H}4SK6C^Q}upt8g_2Y< z?Pnb{h|s8kNP{$?dD8sfz3=byzTZ!Huj_mLPuISV>YUqpp1t;ZrnT;S)7CM>?1%}z z_M1z0rCnwYz28Gla=CW_m*il>xiV(4_YJzZsDt{5jK$$9-MV7)T3s z0rg2oVXojH{7~c0pSg3(#t~z{YRg*C)tF36r>5hx3?+O%^^x5q|3(W< z4oe`#v(!mQdm1@<@dc^Vd`FzG>>*KfHTYSMgw8IG19d|)Irq^O@&fcgDyf>?n-pN~1_DetwQ+0=J-lsS|Ee9*wg%?LvY6ESxiL z73PXuW}D5WFz^l!hX<&lLZ&_bu3m@w^$&5{yAU+0ZbXM1V_cfoO5MauaaGJ-JRq!u z-z#2GPiqOBnQ)bz>&U0CedX!3=DB3Yki*Pn?enxONe2x#a{Af}n#G&L(NEXd>en9rL$aDJz zT(Z(_Mha5k;NBsuW}+(#txy|Ehf^bVTWlDm#dw$;1Ur%-9?J4 zP1&iV_S4VDoKb#E0xxXQ3F>ObprYSBdis+S+5RPl?%cT=R@~!5y>%q45i91n9M6EN zls9x4Wk849WN@;Ggr0rZz>B#BcfI@}{^T`iYL0^T4U-`zX*)#6eI&2UOvozSNpA(G z(U+f%@sWZUEsIJ)lY#H7i5P*Jie?3(16+7fDN52QgbI zjQGipJFBQ7r$6k2TIoYD&SEy~NQ{QdftSerEisUC@G40AJHo8z$KdvxJ+LV<4+LX5 zNf+~x;5T#)*riv&5PL(AJ+Tg|EyW?>!+c)P>4)U5<75~%@D9V;z99ZwzFuBxEJ(c@ z1t%T-;KQo%uySlDEHIKMKMuu%{~Sjce~bqk2Un5HDbir!c7_P=6@&N_UeGnb6~q?o zf!%M5h-_vDX*s?GhH~13Nd<{;^sp5uW{wA4)d&!_N&&^~(h#GPP25lC62o%~VDA-5 zOm<(S$M{V&sWK6}#aMjOr~#&x_X%kGgIDWNPJ_6GXxLnafc7+45%CJH?>zuRn0N>p z-3*T|lt8S_X_&MBC&*-HLg)B-zz?h+q{gt_7(i3ywHpOwg=GM0K8d8CrG7K(DlmO+-%8A!R22AV5k zV32q+)I11;7t`{A7n}$sNozn)C7KkOa2&}eMd1DKTU-wOBzit?BRO(x2*{lCDobg}#X|NyP9>It4CalR24=1vhz>!9 zIPYExdNpzNN^!CH{g)#i?RLOa$w+Eqe}qXEv7@0s2NL09luUo_0q-R5lIQJukeU9N zd`)u)7sD+eZoLESHReNj=?Q3-@`6rv1IX|i079Oo@YpvGj2}mFXI<0ay=6T~QEMhM zr|5vVtP{wVTEObu6)^owIIQcOK&|)OqIXWG(s9K@aa33(RlQ_GOd`9<=9DPrh@K2H zvrCDZ-f$=PQ?ODXpfUKN@$xkr_EvikOnkb;`wt?xGp~*D}8eDwaXHmOQUc?zB4Md@z9}! ztFdTI$35roqQlb)^l}U5*3%*Q`_u_c&oIYj{H^HliQ^D`5QL}v?D5j1y_n(FKv6^l z$?0QcV>4HylV?aR+v})c#d@k=yq1mqGKA*n-KXDPU7#^_OUcgCfiU6fO?v0E38t1u zl9)~lrbscCwm21$0l$ry;zDKMcW`@$rOLeN!Z}R*x0zrgIur(Q`Z_mqCebV51l=M; zF~#B%+r3Q{QlH7g58WLwPDc$6F4O>{)FLuAQVlLH8N&;#S3nhCUEC%Zg>rn3FTg?( zT%1j~v+bM2Y2PsNdWIbAy^u-gr`ED$(?nLutAYcGtansUgziA zsJi-hsvCNqy)N{I`7yJWk+O+k4~1%AC%49a?y?*WHszvEOc2_PO~It>d6-h+N81z+ z<2S}1Pw)CgRknYjr~F?s?c=z%jkSYuT+Uwijs{MvDGq$MN5W+;Cs^TUD}0p>0TtsT zuu$^|nAq%r&w?o+FtrAA`*XZy$v=p*?_juoc?`JME&;!;;b7Pn%JUwTLI>%{!9=;G2FR%85`JON1Jow z$waxgWXy;fm~F`IKL`cG6zy6_5y^m1|66c;!d+l&0|B*FVMkCLiPHZ}zP@Z=Sh%k7x3A$yY% zmdfI|?oYIetHaqEH6Dv*JfI7$weWjEFwNK$L_}52)3J@fD%U-uua7C?$Jitq_GSu3 zyD3BBT}`Zdmqr&G%_kpk>?I|9cjnpNY+5=fh4s3>meZ_LB3F!MsLepo2b|1@mYtrf3$UxK>%;o#dFTudu=jl>IQF6PiYcd2RNw zsgQAW34FbM7Gg%FfKMKm!&yBKa!-sa8~Ra*%W2*~OkRGa4gqCEX+k4i**1}V@}r(z z_<`d>o|8`O_?=8i=L#C__mn8R)RUhn&&v00e?oFjuVgk%s)68VS@3Y(VMs1t1Y2^; z$o3a+iSFAh@URyF4fzY;_+AZ)77T^W4pSiU3OC>Jw}KgNuJCwQ4;erFMcL-gnKVqo z1DEO?rXsC&c!E`t^pbvMk%yJ&`9(?R7 zUnO>cSG40^dG?hP%Va)w;Fxz_%X*n{Ief`>%T$-B#4cB+W zoELnkJN<+FvebtsPT9ofg#~+G$(P>N{z|w0T1@H`+=!!ae!1IYQxso29f$5+i-Msl z7%+A+9ydRL16{|V%1>q7;C2!(e3C@rZ$9+hPaTX^zD~u`V)3HVY+NLfjn~#+#9)tQ zs2b8hJx*{M>SBs$$ko#f9yAzdx<=E+mLB%@RUTdKE<{bjKhw#oiWnKDMo!mlhth=2 z@NTCG{4m`E>&C=E+`{=V@8(xhdqx6cKVK&8*Ne&6yrZO2teRO@7*Bt8u3_SqJt1Ps zmkCq-fPMW!1RN)MLYU4{j@!@&K+z6br44XTFgJD!`LUt}IlL7+?y|$pDCPfLL7OMr z(}-*Bym#@kO`PCL*DA<9Vgrs1<{&;W7bZSa z1R2#i027yTc>@kG?EYp@)gB5rn_rag*8WbK)IFG|2{FvkQAuFfya-G$9)hE>n<1(A zF4?k9ftbdAD{mT-N5YK_VBsleI6H)a?5=!x_@o%#oED6FRFM^T&jKUDDPxvh2 zHiiyU7Fe{M!O0IL1!FuKv3_|gdSo5K?uTp8yXP!@s-92PZoMZny9}AIogQTQ#R4KT z!Vc_Tq=3u85%B!iOO9*&0~l{R4)1Ct_&)Ps^pDf9HeJkEr8ju~V*{eqY{#htoLkxSt&!**{qF{ zmUF4EmoUAte+Esv6Ieddi=)kMJ5DuKU$HL+$l$80qHNhg0X=Otg-W^BlYzk@bm`Z( ztk~#vko0mhFe*)j&wPws?IYto2toCbL9IZRTU8bEW+RR{=^<$pOU#@CxY zg#Xc3lD~G47~gU3L%0~0N7J3GG3%TuT@+zkUVGP{6sL+|#+!UH*Xta4=em_d9^Xxr zPw$5OMIXp~nRBG-`Ak-F+*T+UxgWH-`afRRUbrD-2LU##;jo+oL^$mMm)~>YTDUZn zdJ(uJITr$l?jyfXT_joQvZO4Nt4C0^fwG`drcBtDRgBxiaUdOAMj<8cK^K(DP)o_zM-KIq}Z(QIN^S%>}9BwRp*$6?B zs{B!Y2dKRwyS&3&Hi}975uwdRTOYG+MgR^c$QBQv%)VF#$cY1bPyEPY!>lYG0#?-oxc z;dV255ueUnIHgASgo;M`G96s^#u1xpa`C&vO7!}ufH7t!sIb)!b#_}~ z#95ACX4XNp^^3xvT_f;wz$O~BD3_KkuVg>S+Y$31A*`s|f?q^xQTy64)cN%sUu>c% zU08vV+ZLi@!y~%B^8n4&5TR?=Zoq*T)v$a?1zldwM=7SB^%9-x*w~cyT{ei%!m|X@bCgzy3GWJeByGP6x2X&{VVd& z(E}o)Qi+v?97@UBvdOnU(D{Sc(khABAgcI@vofN~X%CN7-w!F68;w$JAn~7!@t>r^`HpY4HRlviMvvacWedkF96pnW+Ny zIqIOee;w6JHIndB_K}KDH%R!ADWI__o?Jo0a=l{_e6h;y{Ew^k`LnN!@*65l zIsGp#Kj&!{t<09kC*facan5mCUD-(UJS8bnYG6Zzo-s4EIi0`yN|^Gn416`a;7q_X z`2C_D#>xrtA6^mT4{sLY56gN0TMb5oNAf%JxQHw&=%TW zU5ET}Phs(Z(fmc0a{Tq*zQVj=t+3bj4rrAL;6eK!zGLHK_}#9|&*OAU$A#3v8E*Wy z+qVK_Wn8hLv58if_uvEfW_)4bhh{gvp)I|GBERpVd8WF+<(|HvHvJQpj2$W%lsrH% zaBm|%IVB-bux&uk_$Z7{Xkw$tJajcrU<1CiQ_=1Oy7E528+CDhqu+bL5kmZIJrRCb z?5xc@IeKQ|8uLpyB5#J|o%=5NJJYz<_NZk>0Dt6}gqPSo-d#&0es2zo_r}!9e(AQA$ThLEvMoBiLi7B)I=&j3CUDFIZ~v zkUP`ejAm^mOjpQ4*mzSGiY+{0WWfj+-y+7ZK6eYe6wUa*Tra~H`A?9PvP<716cwQ#ID;f@I zb9`J^!=ccn2!&) zPR*T?>6;*Be0}^G-5b0OT?am)=3An$SUiNbf80XMrsx*ukB%c%A1{{8+#?IRVris( zY8FX~6NPPW4v|OChQT`f`>c)l0cMEiF(&WMU2^-1C)_o-Mjlq|1Q#y$EYzM26S$oB z%9+U05FXu4J%61iH`S`i z`v*(N6{QTOoi9Z5j}}udXO_HMAIyHviz6dHZDa!zbm`u?TgWbf5Yy`J%siRA2G4m- z!3~+Q_-l>?uH5v2sfvr|J(`fr+q&2mA`^a-pl@@*JzNK#T^B%7fd+&vZ~$@MV=`~N z64`k%jW~WE2CJ`0K)zovQU2;e%#((I<91ybb%@h*-Z8pdzhOU7T&E7|xogO?*PGzo zM+cD5zX-Rt#KT#RTk^T)Cuphu0w;e_2s^q6%EP#tKfOh;t7QU&OkM+d19D-tY&evh zdkJMGKS0yC3#M-smpd0 z4cTPV=!>Ms?>O^2Oqg0f(BOO}nEp8Dfq`1dsBt_56-=h#vWO*Ur!xU>cQLp+AQSmA z7HEarC;R4#R&m1oeUyki*1=Mnt3toI^ zjBUp{X#A}RF2;snsCp3I{i=pZ-H%N5$JUx6b+Z{w`$5BTZwA?y(i!YhkT z($Q5JWJAbF_W1=%y5nv(^%^p%+%Ru6-qI_fqCW1-`r2}uE&YtllH)Ys-^4K=A5W(j zr%4e5zwxADtPx$(F`ncUTw#k$d`-W;0Qh#vjhIWFA_|Yru??p8$(MGZEjG8P^?@a5 zoRW=UIf!SUmveQ=ad>inFkWO<;gTFZoYwA#qjoRB9lBd_+V=pQu+a%8PTz$u_AH=U z@8X!9GnMJ!U#X-pE|Ymvzm}xr)zXZrHKg4mj?EX2VM`3(@veRFq$|?uX-7mnJv03@ z)gCETF4!h1)s+50w1)0-&n7LV-pW2ioMBkfk(!g}3uz*UmJsjJFQn$kT5O&t{2 zPEqdtn;AAPE@G?)exVEZD>0fa_Cz76oh+U+kE-xP$T{yClI&K*OgQhwj2v%6Mc*5; zd-rL>qmla1^Kl%b>@|}_YEK5c)n(-OIeVHmC)?!SMpb%xw<>WDP^RsIIpkA91-oPW zP})?}OWM37FU)Az7^iA9WoOAYJiBS+rOp9AS) z=^*iW4R_Xfh-}pKV}^=L!Lk`vAXRaL{HR_G-<2|;?8XQ#2Yo5{oQwpmDGZ#AIsl_n zM!@4Q4&W?EVh-3|;7t#FLMM36Cb@}$?4vJAZ z!#g2_IC4DQNuU*U0Bp0k`mfT>u!}qQe5=Ks|Gl(>?HWhHD^V7vJ#!(=D_vQ$$o({W z=o@xy(g;*i7UlM_eQ>c`1)aK80~5#Cpv0mBbluuHrVZZ@k*a2J z-XsiIKLE!?PGBiK8Ad-hgC|;vU?S%Q^B1mxfL43>G1iI%Zi?s0zO854zC_bA7Tx6m z5^-!xbw0f(JpheFAJOi@9vYu!g6Wx)(Z<=Gm$T6kN8Pj`ms{M3(W(-X?Rkbs&HqNk zGxm^>VO_LJb_88@)tb#c0le~0zBJ)v8~xNX5x1DlzyZ;FIo>k~d^|>o)>oENp2R40 zSiBP-XNusaUFYbQ?5o7v;U2NCeM?Zp7G(ECz-g;+P)JmRVoPCYxm-&&#U{eb&~qVLYbXn*<`4nFsrUKkXEdPj6{UhP9V&2lrVCjE|485LZ%pVMm} z`Pj02V=l*?P#-`BCHsS7N(XcOs~)*Fx{^H|5>&qbwlj)bhM|zuK)fKytsO!y(MrEc zUe)Hsq`;z`z2;KSs6})!$AXtb=De|B@pc}-8Y>u3>I38IZxJi$aIo5!2{HSF;qbi_ z2#c>J@_QDOqQbev#w3gyKRQHg_ee6Tvvt69^bm-DGZWOZr^9nKuGY~%1|)5KK!2-% zylS{hmW5ctMNVT*bkKHKZSDi>UXFsjAKsJ0df8;YLJ$qHccO`&<1s)}0gr^Fp`H5% zEVyfrZh2=oZTIcGSF;N^4djVrm3=8$%3BXrm#sijUK(2IEa2Zc43&rKxE$kWkon{t zjp2CHG`6QvdsA+mF?%`ITj$}1nS4yU;*OWycH^)-J@i%Mm1_>IXEL%PnRx|YXyQh9 zQnA5}eBj2mne%V3o5e=M^%NCSc%qx!+!+RuaE%$6%5nG(Yh&-M7lKV|2WIbK$>d_H9YZleQ%FP}o}&N{>Jqs`>r zmxm503VOm5LXzsNF^_vC2)K7vAnZb}h&%!>^Kyn^c)~ObbDevm7s*A~dLdC68{X zKvmFND*f&Um2?ZET`^g>N&66PvOkQ2INfOd=%wttM(!Lazl?@#H>B%7#k0PLE)vyG zE-Y0FDW7X|hcS3q#g2C$PqjzKF$bS>{6&6Zpkt=S8tW;5QIRlAZl4dY)2Bm?MHx{a z{hr*rZ4G$4DOY~%RSl7?|y zSrKr;)(@(4-9e;n7Vu2RfLQo_@-vSma`Uf|D??mBKl~>7>GO?PaCKI~i?!)$wwkAt>g~TSp2@>9r#4hJik16`>k zUp9#jl4Us#a&_F@*3F1Yjw3xY7L)O(2a_Koyh(boEG!&!0FLY!2HyMKV1SSnXxndq zr^82c8qZ1)>1YXEZBhU(NhHEcg5F!3F2c7yE5uK%s{!75F1M%U9V{}q3ySK8pkn_S$Tg0E zi>t1{5mR9})niKaj+Iiu#9bI4TZkbFp19H{9>)}$lpk2V5^GBbL5AHe)^GM%;=EoH zg0xe?h6#mgPKQp~_cJ-MfXmNzNQ0=8CrHLQceo=`2UEBlq0CKdVP)qKsIPlZe=omC zBgf0p{i{DP)k}xs@Y|fO9PGmtI}W1E6fp!H9h{Oi5f7+^;g(D<)clxC!mxz6i_d}z z3kz5_oO>?b`@^-QXc$%b8YBaXz^p7BMh!R%VdGn2;j?sjJ7W*b`=JUASLZ`@52v;D zD3MTeN!+nP6<-Zcqzm6LT;6gBR>`~J0-pmocDs;3`gJE-+18?%r2rj7SE5nDBiin( zj+YWLarcRVxHsDn(=v+LX?P zl?)knkJRTR(!z6QBz)yGQh&UhaXZq)*d}z)n@^Q-e1{G-THZ`;;v{jMb1GWNGF)8v8qey}6{B-MzSkJP+DQ ze0Pdqh{y<%N91w3t{!veTsE2Qtb*e!7qCH-U(?3rED`dvV3#frA=3kF*n~5cq}pB< zzfRlD%u`Y$TW+h-2Ms66Clq*;kGvmGgAA8}c2OLda5Vy9k<%f1y%nq;e7$^Tlqpv) zbB@V6I38{czev_s`;+%32pjvmMF*UByDeMSWnZhwDIggcEQ2JtmSSM zHuDFWY1Loq+$Tir#;wA0=~C1cvESp4+;IPQlQbm5&w$9HeAd8)YJ z>_I8P))%i47>(>PNh13*qRpVfedh3!QYd zfqlF&nKC|hs6JAKhIzU2rgcoCqu#0FBHvx;Jvj&E0-Z5Fv6f2c#Zav;bID?}pz_;M z4UF*nHnRJP13hzKEnDC)hKw6lO4oL8q&ZX0l9nfNr*|jL@m*;-k)IdhNZzh{#P}MwpI+os9&&3NiFaQG$KU@X*L*IJ+ExQN zd^7>tFbqcfQW$Y-EZnnK2HV4VVE$`AG=%R1!S-SB_I?C&*Ubc`*Dzo;K*Tiu9YY@H zl+g3ZQY2;McSdrB1Eq(gaYn&G+BtqIazx~up3Y&kJ$41x6hvY*mkT&QQwzgwW3kCP z9Zhb}!&Y5yda$sY4m`;aXTLrO6sC zI>aPoC(*7iU%1+V^&~8LB$P6qP?Ty0u_FwD*6D-7=i?AJKNO1E!a!s&H%HX8gYsqj z!F-S|RIYI#8|1qg!)HNkqCyn=&?b+_EMUNV565d@v=z4ItcFF6HXuBADyDS^(EIdB ztW4$dDTRaZ$;V8TnrDX6(@x>z&37@ru?VZ;%TZ@_D1I-`#EL=j=&T%$C!F@a-zZSQM~6=5Y1KOt`Z8EBxG*#N~?Z2BS^M@ZO1wH5p-WPkJAt#@7O$ zuvC(@Nde}`{@{(!8;-f-eek}50DEv@_&fwRaAyRPtNdu--lO#4 z7f%#^z7#EY<=~86LvZ`7bb4g%eTtqc<;%F!on=R|sHeIK3i+h5%L@j=n~73H_sJk4 zWp;^}a-0Zv*N?*b`~(a;SdJEpPoX705Nn0o@NwW({C+nH`HA0Y#K54kOZFRB`c;?- zk^RVce)vwFfBwNNcV9xb>RDk$fgRP#9gjN6POwS`gTa3Xu= zmMpw%Si^Bbj)#k>TJUC9JlvRCMg|Ug0NpX{IQs7T~2NX?WDG6hls2#%7P7_&e<*D#s&+g|Efb`-?Gsu>yN*<7~VdSxe}$ zZZdx2SSTB>Ox7&|kcvbiA7%+b9ft6nH-<-Qev^yChm~2Kov zBcjS_7#8tD$sF#CuiM`lRzW^wUONU44;_Tqh1!rlWG+`jZVq=27{mGNT%FgdV2(>g zozpQM0NEBxh;ybtI9P|mx14hj%0z>tR|(L_SlE{o2?n`-V7obzx2(FEw7%X!KCEhH zZLG7|X;NCWZHPE9jq5NdU?C0_|3wXkb2@)}CgAheIrtH{{2Na9VWxK)K2KgsXM1ye z;P+zamsCC;(k{gAIhJ_c=K=klAW0XU(nRs5wKj0y+aqnDY$bvYWfv&>;ssq0et~Y(Q9#>VEtEXsM3yfZLsoeQGXtN-Fh8Y7 zFz;4wp}tC)_<0A9POX#1MyF4#$mn!Z|MddT!l#m+s<}d^wM3J_!(`wUApOl>h($}E z)A0v3W07nmaZA%B)?DuQ-aA5gd{PE|dtHb*fAl$RRk}n&w5r)B_FN5J%04{Zb`5Rk z=Ht7@4qTy>jJZKMxPsGg`F3jnE^E)h2*si3;bVa}Qj0j=KM538u%Q~y{cz#nWU8~; zoped^z+ayaJlR(0c%u%&l7aAgU=WP4(gC5DDb&Yl67yw-0iCPigI*l>)PkLeNW(Gm z;&RN=ZNtW8S5V*UII`Q7P~+k%Iw!-8&e-yR=6h`>_hcrMBR5Kjz3ov_-f0ex$BV%X zugQ>X=EUg|Jtmoo7swDv6|&r7JItFo9*jg4!F#D0q;*7svCK9|clHKjYe{HQk%xVq zUhLN=F_a&rkB7D_VtPE{sY&jAy1BxhvU8+}seCif;;j&2UAk%BK;b6MaDYMZbq`v-JDwJ`qQ1AH)%J$T0YVJ9P9A) zN_mdxREU}q47n1X;O{&Go_~#p^HZh+yT}E$E?W!juOC2)tSNX~4T0LzP7vbsnmDSD zfN8f*gUGRF&}MFeX8a>K@Q%|EIW`a2+qKO!BWAX1l(;7j5+ z4+Hl(=b#|U0Qikz==&2UxFC#8UXn!oy2FSBO6nbMr}F+X_?^^s#ugoe?_>U7N=p3^66BqNe}mAZa`sW zcRYPN5L0Azad_bj+#|jn1#x?6-2_!U$V!p2fS2qqn^=;rtOiy_9%R>>CSHb42pify z5FW1AgADW4@}YVa*vd15> z6(IYrFq`XGGI0E8s50rOh&D=*RjIWXGh@csD&2$Ln?=bPo{3s!9v)G|LNaE&Yv>LsbN8 z!juHpY_$dLrkVneR|5rl@4n+`v6JZ3ql4Qs&$C&-qN%THFl{j~Csz$+aO&(5Iv}Zl z7_N(CF6ZPDfodQ~Wq%+ofj(ret|3Y6h^AepXK7FPQTBp)1*w&ZV8-s=K_Z=_$b<

p5+-g}2I6BbcYo)vI=bfF7q+GKK2eY~AxpKdlJuA^ zvUFfNEDcS8!wQ8E9kCZa#}F{E;Br@gxTy@+Uj5G#k< z)`bxl7eMX%(v$C$ik=z3?YD_!~qFjwurXQRf z{fL-#pCV(o#W4dv=hKo&KD6}RF*>*B1-*MEjxPVcj9gaFCFa*9Ah*VcmYWYC?;;n| zfcewu#ycML#EaoLEg+=$?W-}=_DmgJ_Tf047ds84tkhw`+-)?Y<~UJ~K272{4)Ilc zLs^aO-2QvQ4=Nk+iGI1^K)0F3p~aa*u8#uH^=v#sp&VEHU(e|SavDA+E-3!;JH7q9 zjz0gEP7NQ4k~2-6gspv8KCx{cQ{n6cFBb8jJ#r=}I4^;i-(0w^^X%RB;FtBQ z(IdZ*40+Yg>UvUos(nA%uaZa3ZwVl{Vh*seD`CkKUuaml5$bp)pu8v;W-LAh#oy1t z6Ujt`)8&d|oJdl&5vlzyPH#0WV&=rHK(W*-v`*Czx6fXN z(uoX?&wWYbciEE}CQnIifelmNJe{sEn?!sKJIG^Q59mqT29wL1pfyE=|6}?$*j;=L z-Z z@^lP@E(;0$x#rl5iU|oxN^*bu>or)&QfS2ncTcZ?mAU~dH*EA;x7NmDs?fs!e=kF+ zKNj4~SnVRh{k}if9DCtEZ~w;zrG+Ki_;ZAW{#r*M=#62_m>3%IdSS#_1(XVDa2VY6Zw$G)B7e2T(3mmRXAFC_ zB_<)&=L`uByw=<|AEl8jF|pou z@C*$ZgTL)~I>}43pZ-Yf3xlWM8)M}+*Jpj-m$(w~-gYnsOm7V9*9)%;P2%ngcm2CR z&xOZFn_tWIQ zFiZ^$d;5C-{h?UchaILoJ^fzK%YU|IDA%6<`-=Y?!${AFXJ~3*Y}yOtKMnaeh{~zI z3^CF(=BzL>?ES>_4-(%Ohyl-7uXhOf&rtSnOXN@YVF}kBz23gsf1f1u0b*)wsK@l~ zlm0KYW;U?d>RSkj?w-1Nqlb*M*mNE zHS7h_e{)pIdRfBgGYr?7UXi|kkc2)!3``6RO?rp={+}Ji^4^vhm@*+fVwp zB)KmT##o;*=`|4d-?56=hb4@ODX+K3_TRAyeSjDn80s7M9=!VR6L~>zOH4RN>0`dv zKS*+4AVzwoeOS_eb5tsNTViUe$8}0CKk46+_`X26@xsu!*Hireu%xmNOL!(sA0FF3 zNPHh4oFQCmdWC}in=)+(CSv>yx?#NH?(-K}E&FOyh!~CDs(!W6x`Uc_F zT)q6HMg4HVKg*?mgUI>I_SGkS#gqJbkMI7oe*8B`+}U0rTp%#`vr_rnV?WM&BHK5J z-0i`CTk^j^OnZAQEAn&HKda%tEcqzc3q;@0z@U$%%9G0uOZ%}TOuiS0siCQnaqrn< z|9%peqtF+KiAnF-qh{#cw!Uw%d~Tm0ygpXd{RgGE#vy-MVyLHYWNKuf*J}hU@_yvp z_XDVMs~3o&p*~}z*L%9vT|c?&pEdDccPCNlZ%h6cNFN(Ov!+|M_ic&XQsq8D`q(AV z@0sG#kDq9${0(AcV8|Gm_Ks{0-a3u__{o_*L5%;aHDKJ$VSia-WW>!0OpSU^E%r@U z+}w9_R2usP;q`HM7-QA>eS^fg4e!emu2XuA#r->*+=sqF{;Qc%a-Se3eLcnfKY;Qd z`?7?am-ilv`)^KsUm%9u^s)Esv40SiPkmUzff)AQ$NA5_<$pgr$$fzs>2Z^`UU%m| zw}%J+-yIn4C%H=1|8tA~XH$BP{eS_J{^vs@MRoqq$LIdfPyf7||IgOceh%W#i~F9$ fpZCVw|NHNMcG2e8ix2qoZ*#dne?I@eKl}dyB+q)p literal 1822792 zcma%j2VB(5^FD|opr|Nd0kMIA1?ly6&w}*?^=w#?rXW%-9Cj27AXq@K0E!*Oj@aRL zV(-`mMGzHx#V*$WCU?H%5dXig@8_MJ-F+s{OfnmGlg$m5Xw}uw(9zNO>t7R%rW$@s zKyb*^pl*SFazC5M&?$X-Hqsca_Aj`smSGl-y4qMKq3`b>%NlMkilv{6r{`-BzY2 z(Ui*?43TI^G~;9qIia5z%=y?zE^926HSuX)!_jF$A(MjTzWyQdD49VpbtCu>^}vYm zN&hi7{g2tn$ zTL)JcTSu9zkjaf>a#t~Va7-d46OzlYR3`rG`b9+AD3eFl{Vx&+TPG(6XD0_Ydsz=5 zNlz~gd%3I^#}X)sm1ri&Jh_5XdMnxb$Yp({vVJuNl)d_Z>R$_?t-YO{t*fJ}Y=DqZ z!bLSu6_pnkmA8^-kX$xcDjQN`JlLy3Qs$*pMWdFN+{tsLSus>dGmN7duA&*i(Tr5m z_{e3Wq_WX9PM*C@b@Gp?jl_}WovWR*Y^;!E97i%_=j>unGd)~L z7s1g*s_0}K9iya+lFQ^$S#*s|x3{U0=~Goqf=qXDu(xw@a&mQ)O%pOr=a^=wm}YWJ zvy@D;<+2#5Y|dX7t2)!?{zc+oYin=o;^ORTD~lD9%=3zkl*{IGEE8z1YbD4Qcy-Z~ zXmP2&P)Qdjm&HqE2{lsPK|R$M{l!QJu#>&Bn~SrpEK$gq#D%q371k0itffkxWpY`v zRF+aB)g9DQeYuLJI@KL)9bFt8o$XwlWvN1%G>&G4ie@E8vr0*`S}seM$}(!4JqJOm zXVyZ(-6v>1*vi%jN!D^CSt^os9Lahm$p*P>qg1x3M(#P(n(CWXbb?fOad2?8wX>rc zu|-Ii&CzXD(QV`Ca+Gx2<+2@8+0MV{YD@K9f78)ht)qj33%v{O7SiqE==Q4Uayhzv zO1eC`Y`;`?phl`Y)J*k*DyHgG=kB+*l*idoc1Xx{m}5GkVmit(0P}_G4AvI-&HBaZH|itY(V_f$#uOfGva zmA&|juC`QvSqq(mi?gE}ccoql>0WbmZ&Y+|Il4+E-8;GLy;Sz0Myfm3O!bc{rs`Cu zw?CQ@^gnyqCn3{kj_Hev=_|)nrDXahmwlJYe*AT@s#E>vUnKlp-ObIFPI7+$d=~2Rsy3*tCC~lN2QN3ovDRRi>hUTz-ZI{ zT6~~Wd|Nu#cV_By_SLg|Mu*mExQs3>l@T*~d{Z3LfR;)J8d6Qp=u;_W8d1rgfNBQ> z6Mv>L?N;#%Muii9N$D+&X~LNc_zh^C<8Mk!6@N3nDULCurIO!>YI4SyN-5KvN}pzG zC*Z#s5YvKo{>{$)@wRoOe;SyUoUwqt6|HmZCbU$soAOO@Olw*y+0CdXXWCFHW!h5d z(^`$amZ=fbj>1%a2*(k<1-UxY2@=zuJB+}O4z$ksVNOex9~OL59Mh4ONACTFau zlrq*-*6^d2(ZzJ4uv+|ZvUQ;M5GQ-4Gj|w)A6;mj^P?**Resp;O>s;&S}OgprJ9_v zqf*M)Q_0_otFu9UGQ~JhgvtfAk>~1WZ|Caj$~baI5V+t(>zoVDv{bp^!Z*b+uC!FT z;6^n$<4&cN@u0FsLi{%#84>OLi=7*iTtYb5Gl(-5uwz>12d0=R9DR2>H0i+XPOFj> z<-qJgb)_3UsU~N7Q7L6SsZ<-7>I)5~H|_t+6DKEoy8flhU|XgSXD{%iFRgRup&u<( z=b=B}6vqsprP2Wj)#S`TDy57Um1+Z1Z3V)3({2^NaA49Iw5yvVy$*voa{>QgTIcwO z&{D-O<(uM|p|n);52Ko#8BV2?89}Ambr38~n31%z7IwPyce8hLvuAubV*&dpTIbkD z(^ADghHr{v#?n&BK8|W~W;~Tr#+OR90a5Fkg_%HMDnA4RleAw4#QA{$_2r|JG)p=ONH^niDv{X8fL^V0Hm`W+Lgi5u6skUBXmeOt&zhGeU*TI2K zB$;KLxqv^J);azZTB`V$^G$I~DlL`#X;hOlE2xw*E2&hw4uXXgvx;{9#ZDJ&4zBk0 z4i3)DYR*`|o=)o=dj>64?3sL19J7X&O7^u>lQUUVN||+3stt%*S6Iw?3RC%^J}~Kk zppyq?19upK9~)_%^J5b&Reo&do8p))v{d?$O*J{Ql}ahIjmjE+)HX14D6BR==#1ON zndaqo?l1yBcF;QK$4*+R{Mf}e#WA~Ssq|wH)#S`xDy2*=m1+aC=9-MzM-eI)ss|?j z7|e;A6*GC<5d<#mr*+PS1GH4RaFB0`V-C?$>B3>E$(bWmN|~cn)<}rzftgP`YhmY} zi8*j5pnx+Luq$YtdqBYus&Mpy{KAa^u3fcuV^~^I`f-eEa^^UdQsxAeY9mv96~~;U z{k6Md>ugK!P^UP1fh(tJojVa{XsJ38XZfZ$<{T}R4xFc&oVh@yl(|Ty+Q?K}-Z7VG zw~AkFWIEb9xj8#KGKHMEfd4YBbNp9msp7xNH^nj6XsP7CPBl4mgGwoLlS;JY9pf71tD{f z!c=|;MkakoK+l33okNy$hY|R3pVm1)9?(+d$3wm;j;Wxf(vL?}lQWO0lrm4Ktl>v( zBl9VR)$WJ0Etd_?xWfqicuwn_A1`RB^5Z4n6vw=xrP7brRFgArsFX5qsZ<-8HJ6Y~ zB}J%QP#>9gw)B>1%e>=`AaLP5t#dAXpry)%k9<=c^NE&97d}%>&U~R#%6z4=MnY7N z%qrSh3p*W|bT;ec%zWdF1?=Bx-Ah9vXMRwHqwmf?J!F1z?WzZe%r9C|`th4;bU|8& z^FxzLpT=~VGciEz!3tA{cGT*Ft&1x+pVs0G1wPcJb?yw*qowK$X!A{ROnq7^nRTc} zSD;i%89gfbHz$-?xUOtK+f~HXW0AWl(o;`wOiV-0PC%?r>l|?-TB?W}^G$I~6Iv>X z4X8$!o>WSiW>oq#REAu=pfsedf6>y(0lgc#(x+F95oaZ!HKuirwmB_Tv@Q6iIHn~n zm9(v>M%SEFN*Pls`FkLZx3-mJYuc}JKz;PNy73p?jI$Rw(1zAI2inq7_)p(+%?9V6E|IFY&lZ_w;io>-1f9oaXau$af~A^ zmE2BLqsvPwrHl)ee=d6UKy#(7wa{_{&5677x^Y$lT6bFaI@Vaucu4X6=x75;jsgSSnj>C;BYpS)Qd-X3sE&+CIgLsRM0A9^@%F+ylzQsdD7jZURc%t+qEM`+UW zUnblQg&D;gj20TyeAgc^#V;(NG6gA~RTgT=8o?Vy3XRmyF1-a;<4P)}CDvkz4+Me`<8g(m73QeeWT(=^^-y3jy9S!v9|WSzm|XA1G^ z7yb{PKl&^lJzI#bc~$>Q))*c;M~JO?B9+NHmj}iQf&Yw=PrrG*!F-`X?LPD91w4A8 z5Url94on;mi5EiD!*Zxj)&w5ANQiBsHo6^bc-llBmn6igCpCSX5y>p(QA>m<^(1nr znbb>pi)BI!^`ur=s3mnWZ~RTgT=x|=uJBQ#P!$8>$ir`}#3m@5SS z+aKOyA8(N-v`|l0!BSe4towPB140w^3n?(+)9D~@a7bvNo~$%xVX_|P@kfMs^$Y(8 z&ma9LkIomOYhKm=lC^-xDumdYCsLWLga?8U_|F*m^kaF0V?u-4edf`}dGrY(nwvQP z_v(o`$pcOa0ZoN~>O?)wBhLttT%!JWDCNEPERQ-TL~$#`Y7|Z8^E~E)5W`)jY79*n z;mr3UkG~|ubE#7Uua?M#yvb#u3AZ+{Hc_VW6&`j~2>X{Od=g*d@z;fT?rK#Z_D=%* znULP#4Q>hzxILh?81N}}i-#8p;oQZohO4D-F^|42L~{vQ6U`s_4v#GnV%4wnAJ3V) zJhD`XR6ljf#4Y1t_k^&2M8+p#IS;=tgxBUPkA1*n9}2O;q_t-%c)%keKp2c*e*2h5 zJ`o~?>B!G+pYoVzLX0p8{+io9=kYIucwsu$HnV-no4gX52+!%CdF^W+_C^T%mnVEW zzUA?iLcA~?1rt=&9rYb=@Lp&jyl4UgK50Ji@Q*^cFdhHQPgMv1#G^k8(ZWmo2hAV( z3y=LO#HwGx|I)FFM}89`)lZ!=9l!IiA41qaBI8r+ClCK6gxBUPkNwSKHM9iJ2ZZV9 zz-ZF07^aR8APhz@gVo}Zb%jXbJybP+t;eIZg(zXt^0U|aJVr-|5vIytb5~s+uP4L{ zleV^*YXjb-q0mHl=KsuF^?6t$A?#nC@JZX4$2Sq;g-I)zfU5G!fH!C=G!R}|fdQXd z&3L$>5H3vGKl4x3!Hsydu@EgxggrF#MV6AObD;dS03Aj$F>z>g-PqkwBrHog#ckNf*EQD9%(K_3X|4R`G?E#v9HT&|wenKGU?jNA?0{ALt zkNtV<03nw95Q<`z{Sv-^ps@chQa-!9c%Zis$Xz3Ppg$Mq&wO$aj~pySa*sfSNdBr1 z;c-$Sj=Ppz7{Ue6P#!i+2-6qB_+t&{K_i49)w%!U9W#=L_y{4YGoj3xQGEYsVSi1x z`2-llJ?Vf3pc#<%(-GscfUs6Um;uhIA}CV+1ZZ27+UAqqo^Wa zm(%JW_Gqdr*{4!X&P=0H%1oy+TB0Ar%%Fr862~?9ai>Q<`iBv{Q3}cuis{Ai8b33BAO>&DVUg*OTTIcs>Eu{+Q zgL3aavy5wRtK7cNB-4u0hZL&GndMYUnN%wO5i_@8pYE1*b7s;gNXWN>)_J~_R1xs4 z;@bc4t)>+vUpm$3h9xScOeU4l5|bEa4ORZnnCVM>x*N~ciC*Wm6fg84i`My#qU)%_ z`Jmdty`EP8>_^)`b)^d%sYdrIQ7L6MQ~8gOT^#7nJ8nNZvxS0$blJ4d(`}`SfNmSD z{-MjEx{_`?)#S_$Dy7U$Dx)RsW0+l3`9CA(GMj$CK)*_3c2m62g*~*+yRerk%CpXI zjm@RiKU-t>QC;am9@XT`ek!HR0V@9)vArYRSVb@LK?)Mm9inxf?l4sZbVq3Q58Y9! zE9vs7CT9w$lrjn`qb1fc457;Z5;1+R>q0-zWPsv@F0i!D=h-o;a4t+x?m0Y8tA9@V z392hyI7u}*bBanSbDGM3M@;El=ruk=K|;E-w9eC=ql$p;JgxqryFhg%-9@U&nM+hk znL;X~CAKllWvcwI5z~*^m@5=7bm1zk^LLYLRN-7uZRWbpwf`AeH)uua!%eEmnOjs! znIbCx88WwLiQDX5OhH1v+qBN}-JyzruY_y=!*`cflzgRBlQU&hN|}39MoU~`m~yK8 zFEP`hBrsSG1z!drdXELy1bdLy5|0365bZsq#O??B?vq zZBuzi@j@To(>lLn;R97TANU;$A8Caq8r`w*iE2vv&s3wklc=P-lc@Y>xO6+cn=9R} zT17!Zs&BN;Q+=n(e^ftcMM?FOYD%hKRFgBmsgyDrbY2o2&2Ph}xBUnI3BEtwP7%e_j`lWspQ=bFj-ZPdxv|ZxPld7i1fk6R&GZ-DNn?m`2{Pf5#ENT+t z7s%++oy8d7ORPr0U~4*i~>uB4ts zTT)-5Bhi)UNg7BRO7ta-B#k9aBnFbEl4cS^dTER$%_S`)EhViaCK6LgYl)epjijxl zous{_gT!27A?YZwlvqiuC7mRlC0!(4B{q_7^v{u<#9rbcag;booFy(2SBaa%UE(1T zNst5wMf*kbv&!n(qkKkW5Tj3BQ~kgI-(y4fqlD@8-~`i%_VL?v0wTg9nV=|c`w3k- z1P28&jcHd5(}V+XzjhGj7Zn=C|Jng#!1pwz^8f$4223*w@viZ$ot@eSFhd^k&oBQO zBfkHC_-LOot`%9^-lyg~QvF?!U5)R`m=?7spbuMQ_F<)uaFGD}pw)I#wNhTfSdC`WhSX z+wl)9M;7OVwdh=S5wMB2iBh{n-YpZt0S{d(|aE!b1y~22FAK`Ab;A)H4oyQ5|{cCi4)Z$Qe zydu6=7;lY{jd*}CUiEQ}|2n*IREzmuVZ61E>F%}QYKylAj}ylG*Ldz(i$m4%_Tqbm z@zxm2o;*Mpulh*t{nz1zQS5U zS7#Ur-@`pPrw`$TBX%I)!96Rd9oz#q{^w5^FTR;ObHayvf;*G95Kvn%gLr`IiDQjz z!h?mR)$R}Bnxmum$!O4YIYVcIQrZ;545iYFpXLQbGfYsp+*ckEX~zts&3(HMk2kob z`|t+8CD%WQ{@f1VH(afcdt&&<@Y z%IIjD!IJtqw_-mH=rTYUy>SI|Z_J0yN!f^eI)mQ2kHhCqKgJ&;WAS-MeHb{d z0S+CtfQ)%RA8h94!Nbv-@KwGQ@#b{Mpm2BEJDABYa(I&ezQ6qikjCzDNXleY^~$%%Y1%xJp-^7_<= zAHB7}V{#C5+UWzw2C?MGs6^;DcqOboTOWR2iNd1KpD{9igj=nm_@9mO^OUZ{vQ;74 zJSGzN{B#0;aMy6qfq^c*7(DTEI80<3;b@Z z32i+RiJxUC9`?f&U(kPnj6=1cb(eJF)ObF+yKFIPy+{+^)Vl+7KWu~J$5)~~xh)kz z`S)}Ve57Phh1bn5{dE6rY zINsEv9z^}gAQCyz2s);9$A{Y~P{prE_%&fD{;*&-I_k6rRrbet-78D% z`OXQs9N2-xH{HpN3yaaqilg}Bv_3@UriJ<|jL`y%41{+0!KgMm#Qtj)L0A1vBS|BrDRMT$ryx>XxN~ z=(IDww7x{N3*QE5sfcc>-{?}7C!E(g4?j+%E1Vn_ zLTQW~QV!QU%LQImKu{<;V)xi7`X+iz1`I-G{iH}b_BI_Kc<`paPX zwOr(9utj0jz8)0aGsRi z>{a3e&V4Mv!o>sPN~V%&!*%db-HoU@QvpNX)M;Ko z1|PnMHhNwnb?pqX#@1~p&({`=+6_{4$uh{_<{d~vdtJxI1x0A>kE3`(6Fc#ihV8^J z?oC9iy)Tfp?Qap=c6KQ9^m+2y*N*IMBUdDSzb+c)k%4fS0rp*c7)jQ7VZ%oWF#Op` zn6Uc+nc-={esa)a%iO(Ld^3kg4P2muq%JAltBp>liJ-{jBx3KjVS7(8W<9jO!p&wQ z0lDQvFSiu9_Pqfdaa;+O3O`s`rx|Xw;5li~^b~mS5V5BpTCf=lG}utHNwDIRDarkO zNTKVm3qii|u%hY+R2l6?bwA~ytR*ZeZqkb63|&v&&sa;M+vhu=Z~L3$eS7p_UVJjxY`#K{x6*-<1YOwOd<=S&X@x`Y zzYzUyqd@E39I&EUTkMsbj3lPp&_vT5cMtnfV3M{J4Hme*zzFYO<+XA7S{=JBnA{ zyHUS3V$s)1DQf5Ah!u}=75<46VA>o#xEE5SkT$La2Z8v`Y98ibU0A!0)5Qzcq>Dd?j}mWdY7ahMZQ&e@_ei;31s4a5gZ!rr z*|ME=Nm2)ML{|PnxtaHoUvwZ5^>X|fseSg zwHK+(S_YQe^&nh+j+BbpgZPp?IgxNnQFX;o6z05{+zfm~oP(^u$i)SY)T=K_CWRiK zf0*NuXMUqQXHO#YTcrvIS1rP<^j4fnm;*)PWN=!!8RmJFK$8KN$zkaO^2>2A=_J;~ z{^S>`8aIJV^XMvWV0kBP^5M!wRNQ``D62slv}|k0O2;&2$K22b z2a!Z{?d&r&8J|J}#~G1hD;naTxwlYo>@0<8%ieIpwj*5L_fa&a{w6fAWo|(q`#GrV z9(`<+-;pf}Y$cvlw-+&(xfH*CI0G8`*JJ0U>_C-kOwo)tF`(f(1X`7SBm3uBvr7lY zDSmf7LF8~JG{K9EHo_&3L+ z;vR#^oFZ^tKa}+z*$pNyIEb$wKS+LwJCMR&L9lCbU3Tek3-;;p_3U=j(Rjni-`Hs6 zcKFitiYVh(7gk;rj<*``#3SgvJ^q0idtqliJZN@P=zZ}jwi@}6taExITK?UF_+LAN z3V&`u0~ULVuPxAKPoWcp+2jNVv_06XUcCWLN`-=Td3frACTw7+CvGu17q#899z{%O z$(9cA!CMY^VC}|7!De<6>5@Kz&0kuNY-k3M_QaE|+-c3a9cd!Iz&;@2n9JB@?=fuT zbO0}vX+Zru9iZ@ighCQE9;)p0NNB%IXg6gVWauTr%0;J<=goes_H9qGY_dJLoiBu- z~-P=(R%Yqv9@7* z*8SOiq~o@X96dNf;c`8NU2t_ZnCDDnCs)YXuhX^goMvKVIkh9K-HOr0CKu2#)Iz*5 z+<`6Gu@W>|KZ8rt9>RsNMUb;@0HopF$T|4|L`zSJtZp91ZzJ!E-1-fLxG~4U{niWI zFKiHox|`tj7ELy&xGB4TdlD3M48tG%2yAQn7-x)nL%fgFg){r6iM_o#iIa?y@!?z- z5&mR=>WpfSCTdn;>rI{5C4CNvrZ0_CL~H2c4xP`TnGvyyikI%<<{QVs`_sF&W7C47Hq}Orx0*tIr(_B83Z3L z27`5&ITABvBm&o|@Y z*^+X+Zq_Z2(0$(^c58iB@v8#8UsOO0hP7rb%UiRHr&Qv*h79QwnT>BvEyq7@90yTH z2e$bI6Nqhk3Dx~^1ngt`L0sHC-22i@s2F(%9zB=`E4Lf59iR0BjfQ^sP26zon0*$e z&uEF3oLh|bFZW=VA5X(MFT1c$Z7z`I4SRt71#@s}u!Uf~y<~RS9JsDkh*w2uLR9uh zIAYuxe`u~ii#zp)27A|%ACnHi4wD0L{^xM$y0Qe%#zx}n*TyNlHm52MUu=R`xAiGl z8E=a>w=N}$9iatFJ(uCR#cP4dzDE+v^w`FZjahlrLzw*{4tcq(#(}y|N#fabyyM** zoH#N86Z{HkiO;PqgiZ2Lw$uT$4ze5Y z@_Hh%++Kn%Wokjvt0TCP_$Mwt8;AYtcEK7}%~6t3DY=97;OCQ65}W9WZ48gXt_`d4 z>;y3k6}JRO|HY_&|63@TO~fmWMq&B22+^Y(<&e?Tu3+S~$>3+zn+*=!t0?~ztBBrk z9W%Y@9BOADa47r^_vf`@x4vHqi_3P1UYDlew>n#4^niFpV$mTqHs%gmcc(4>^=S}T z=!)P`-5dz;91h}&S0amyt6*MKM2xpF;Mr;?e7IALZZ{gEI8#9iUY^P)c}FfPv@|o| z{_@edsMl?>sk9BcI==#6OY921m6ojA;qx$U?Jto}U_Ge1Ef>{KNkes7okcIKlHgNu zf9&+~6}o@M34b1E&FVgy!B(aXWcMj7;OzW(?C)3vKjSATmc8_5SNVNGS)skisl+W< zQkjoV9dl$2`lquSQ}tM1omQY1^cxSVszBE^Y=l_50XRK%BR+R$6)_s#8TT;ogdaDX zinpI8z1)!0qk2HvfmATo0ICz6OOFrbST&MKUV@3-h8)^NFu&F9Sn$1w*X zffEfC{Rz$SHdbQf6&{`%U)65jzz4wf{%tB1<8D zyH3I3=L%5+-)-0`pbj3R`9`$Ydk>y%zX;!DZ{V_|@`6nxo8u>GBd}fPR(O#-8$T~- z0Ar4}U|*UohQduuUY_F)qq)Hff!pKT6}7x%*5ek6Ew z3qFs1lbegnyY^#8O&P()>)(N!kuzb{J$<~>xEt9;<{gpFJz+MuOqUq6cVx&%@C1 zNMHQx%td_R`aSgCZV8MpuPErbB!WC$F<7KI<~DM=TL7&St}FUMnuyQ*G!Q*}(i^sD zXs{op=CE#yGdpsX2U~JOpM4d%7&#Uw;9=!bGH1>d==X7hXd7coyc>y)@PkZ`>^*1{zmnRv%&IyXO{H0Bl=w~L6Z*ONWa0a$)MW@ z6$cmTkPhCH@ww-@IA`Z-oZ;L8**1NOrlz#QHp%xD70Xk}G*fSwKYb>?*Z7vg>D%N2 ztET|-&jpE^zKVq6dzay-=s3C5@icNBQAF6Zx#Xv=1r+XUhFdxZz&*DnNM~3Pew9&Q z+}o)PODkTHo)dLp)XTf%T=(U~JV_fQ&<3@drzc+V^(9`M{Q%ETX$SM_=fjUBMQ}^2 zJ`9dK>@j2YVe%o*4JX9tiRUiP!jFtUqEDWq(AJKtp!|msG%b8X0-yFE=v;49-fIP{ zb#Xz~^#(z3KplXKJ+X0N4Do+tB?{Z7h2CD=h;8q!L(}K1A**vTK7gOPG9qc+rthjmJicJ zjlvt_&8WRcY##30vMT?3*M{C(b!cjzQO?PV$32Wi*CW%APgp?OPs^-riR5Lr}_GDW2Bu&*wWdePpet@;M5j&WavZ0 z+BM|&oetz{*+SU(%@77X9)ROMwu5HO4>auj7;Kt951Gu)1KHB~D(Rq6~6>EDM`voO)n6%8Sc{(UuQXaSk4hryTQIR)iOBhaD2 z3$aDzJ@j^>E^dJI;Bi(yKK6bEw)V(I;U0DHK50A|vhN@n^7chR zSR%rey-%X6`g`E%`cqJ``VvW-;!8>=xDeB$MI`)UCTx*CBHLdtR#>#$uSgENqcB@L z9qvsjQaChYpv3Odnw5w&0mpCSaTIWALG`8*r{+ z2wrxs4sqMRTVZ=<0Xi$&OJ3En!bj4zh}YTHurMoKaiL)Y@EvCf=eso~Ew4Ejl<$?G zO_eP9s<=)P!W%>S+Mf8}^KjIl?sR-8Y&z^Y_W}iMnE?k%+QH`73NpIhd-A$%e|)_3 zEgEvoHa123bG1C7c~I8y{Zr9A!NtWSp5d&PmBasplHt z{^m11ipzE=beboD$>yeD)yf#$U&O+hS0P}Vr3V|bz7efv$vD;_6Zaq32zRY;2g`R? z6y8o-$*^Ce$g2SF%%3k26Qe*4A1uVBBoT-Li`$)$xTEj|vq9 zp{F6{_-t7Jwh0cp^_o1hIfn5f9e|Ig$pMFJc*bH6Tsn6X>VC962G72D!z)X1vy`u* z5Bs-}$>m+hDgOzu&2b%CbTtMxeJ{cOQ)dCbGy#*yVK{DAD-z-zN38Aaz%u9vO7S*C z(`KF{9X_{*?bkXfvdT~4V`hDccAru*W8H2%*Xj)EvsD8#*7d+?nhCPjS`Xi^Jb~p+ z60oI*30QqELhkR!qQeeLMP{Kh2swTQ1!q4Z`;unDq1UNocZYXm&BSE5kUx=i@iYb< zt5CEp>^Vuk+6bP#Uxm-@dO-$ubVKG%9f+1eEbiT7A@N-F1&VjnLwC_q@}_cf=NPB==GgCA%<6g51QLEcX_5Nk9( zPku+sVbzHCc*$LF$b+?T+N&Qi8Qu)e)U#rfeD*4C>NF9{Kl(sU-6yD}Ss&andJ6=f z&BGd5cSu&|9!Qz7$HT4P3#fPa4V;NziK9n1hs=R7_-g2Uy#KltjLDPv^ zp>6?*)x1b5kqbCEuR@uRS9%;8+k@4$&A?V)Q_-~ub5wWtcIc7Q8N9nahL;vw$@I}3t;(WL$XLVk~M9&AHL}fK;hCB;x4mRpwd&x z+T9u~6#6)<|6I6kRjN2_-5tN|X@IYLCgHcH zt>Dh^jrj7mli1DY7c7eQNA2b(g3-)C(XAd@$m3=mn7U^*iq|o~QwQ(E;TPADLv15) z`MSF3!-aLQ(|k31wq85j-@pL}C=%#%qp8GyQ3n3`wYNC^yN=l9axAQ#JCBSDsUja1 ztcSM~yg89Nm3*Y1YX2WKmO_UX)i(}=?w z_X~-}gzaFwQ0_F%Yfhad$SA299ML>{+yH;=AeuhD`3UtE*_#)UZA^CiUuE8 zhwp1T<6(7{k(i9RxX$;3aKmpKnr&r=KlYxC`#;?-YHjlr(i@A}`(vG<>8!ixbY2FA z*WI!8dNVwtNpmb6w+-faUd2&A6HqseN@7y1DZ1Gw2qNAb!wpJSW9K%6bm+GTj}5iO zQ?^fnw}HKhEXEE#-`y{Adb}KFJ}4)Rurn%t?TfQ7JK=2|uA`QVo{_b0cfm~i7v$lQ zmTc26KKR!88;ZDL9mF<;o5`xBA928-YcR*=F*MtM2#(RYjla`sJUw6~DV%30UVp}~?CYVHfGN0yPxpS7Xv!%5Mso4HVNuP)woeKZa_VlMXF)*1JD9)?cl z|AfUqlVEK@0z_we!@Z8vA#&JWvf+GZD6rLH?+mfQ6|KUFx5q6s{@V}I%c290>7~IA zk2!=-nv{|Bu!drbj%B#Yq&eFwFCF{`8^Ftz?ZM=85Af(11l_IY!R8a2q5atIK!0wF zZMxK&O^7@V2CF*56(=)X&MqR4hFyU(b?O|*B|sdHNs7rj>NhLLh<6A^&rP$E3Q9jHS$|y3x2280@%eVY_nJ5 zcGm9LDOH!9G}JkNm+?W-5rq!hp`r+qwztM@Y%q47JQy@G_ra!j%aEkgXgp1O5=omi z9aeUZgT%fD;ImB+tnVzxtJ|z1Nk@#?#lD|Ce%CJr3ylWsYS;DfSid7{B<=wh=6xcY zeEsp*;I?R6b5j&T|Na;cDuU$!FN;?G7am8iS_RS`s#f-it~$Lg&3x;E-=~`u95tuDe;X z_u3Uh%GU-TINI$$#_JFdZ=(@TZtncJZ8zPn`I7=1`MUXN{~c^a;+sv~9;8X!KY z!xlGq3)(;TqsOyTaZ+X%9El2Wqm2gQ-aYK_ky{$@W@cB`aLZJP-@Xq^W@ciiI+t)j zR5Cnl;7Qu7FajO36x^{Q6USw5#paW<+4(OoK<@i0{A-T}d&EGW9k9O>_)Z&vzivB2 z-rKzZ@5!d{{BtFoeYGBSxZeZrr1f9}Z=O~J_r1s-448qhw@JnM#a8sWa3Av9rm{dQ z)ryT=?S=1jtw3&B1O^rFhD$?NDLU!j!XBNs677u1_}OG_d@?y7f4)cu?$8vZsP*`afR*%e0y6gnJ)c;ZvQ@q>-}m+ew%ec^}YM!ac@fS=fY^5 za<&zc5AO#yOWz^?xuN*>qzX|>^X@piQxuLb_JZE>@emf1iycOfVOL(v#wor{fQ-Ga z_>LCiTcMxPyhRS|=(9USyB0KnW`*0p=98BqX-`igb~8eGUCL0^_ADGfNJF$DCr@!> zX?=XXbTM(eZo~%cOhZrF^kZKo_QnC`CeU(xDXcZyjEZ6yFnI3)fu2@4_Gz>tIB|8v;#hI?FQQ2d7MaL8;`er>mmxieHOxU zvtiA)LqzLsXL08#ws^3{Ww_uGOXnncko83lsii}4gXqDEh!vAz)1HN}ZYgGamg$3! zcPt4Cd!z6lXe7=JPKDcU+EDRvI|Qs?6<>UAk=RWwU}2*;JbA%6(tEKF+blR3+dREN zQnTz7R^xN=lbAHztKI=JxeitI@ zqR|1zN=r!OtQ{zCm>3%lSVv4f%1F+x$&lN-5{6`Dv!{Z|!lyi$-_yNQ|BW!>d-o#C>M$-getagJ5$}x0Ms2=|BLT7l)wp+(NV= zD4DcM%p=U2J{aw?XJb6~!yfl$u;XGBJUnYJ7KK;gIpGJ%9hXxm7WR-+U3TM9W%;=I z+Y;>IQUQ9;9-&7qY{i;YHej+g3)u#&#_NJC#oi`!U}#Aigg2cGhcW~3>ST;}oT(J` z`*j+R?pue=&{<8_Luq8mT3v*{e?<`{Zg@*>JihkJU9@lNR>*z1h}ayN0@Gif!lCR6 z@^Q%?;(q8GuDYx*PL<9lknBO+%x)G891kR9&_0A5QWdxAlzG7X!#H`uVjQZs5yeF> z#(nim$ki9?;M}q4XyKVG9QtTID1!9hT@2kLaXk)3G&+R8U%iO~u8mZLm89cQhOxNE zw_@@tUjy5`$R_p;XG4+WesX=i7%7H~AY1#L#?zeur?lxyG@W*l`e|{%%C14G=K!#n zIGjXUJ|kz{8;Y&^Xp7&K)y3f>oIx=|6Gl`nhup2D_?+SrO47XoiR@=|+Q@f zHcux}`%XdohC1xt1rhjD$6Pdfxhq^bas%&f-UUx<9|c48t`!)Xbi}$-evxfsm%!*< zrEu-54SRCsa?<|VaB|`6Sn|R`jAZv56+6d_C0D0ag1hBdY#7jn^vhledHKz7QbD%j zP=f_Tf4wg*Ntmx#+Pfz%#>Q;NvQ*Mw>=%4kY6ksG4I#GKOWfA?Im*3S0C$^YlU0Q; z6|TVsxMki1vgq=0e88axN!dFHw`=c%Kb9UNUyT-^%n9v~$yZxEQoI)r-#Y@oUAT}8 zdoIQ=ch-fIEq;hhQta^4=}T~;bRS%d3xqvu*TWXgmh7luTKGVh9av{lKj?nu2f8vc z6Q*ZugP7RGDAAH2i)=HOnAc?jkmCa|-JPyFhSxA%xdI4QnlmNQZ?haJ*j#l3Wi4kKH@5$0LO(XIcS>c7DK9 z-`vD=LYs@`)Tu8%)yEhoK3xv0uoyZN*J07@LF`IV4CtgRPZ2 z2Mpc-W=(IPPFpd&OZ0*8fx%pG;D{o1hU5d$7hV-5=0%0v^y}GTc~s0G5C7hWfYW;()o2QL^ne zyrWGpZu&}2R%H%{U1_UCPge9GJA>)-2i+HnS9LF=L7n%Jb;c%a!Re*2?Z6ha>}Lh+ z_PPtZUgm)ty8%6kv?lM`gevm8CBm49)li(b8LYd_Q{3%5O!U6{G8lH&8}D>!L+;c2 z{F--b&|_QrjwN3oj6Lfi|L65Y{!v?rXJ>ubGG#wW{n!I~jR{ix^bCN)yj;9`QUv}T zq$M)9w}mM?AHcyuo8Z^{7P$4Yjkw1edwe1{U!;)aVLRh}_^gp3@h{Yc6XD7Dgx7N< zDzkxe^F3f@s2p46JpTRj_p<|7JhF^`e`@Ay>d1Z=Z788rbPqs#!Gkb%(rv!vG}#5b;@!`TlNWkUfo0- z+3BeJ*KKPQ4UHxeyBFQqUyduFa9t_(>NXE{4jc^4mbDg#Ot4)3e@?uzUx6lg}Zo=9@v2GKABe-xchs+X7C`p8|Vb55wN? z1JK5tp5k@xr5@A9)I*(SNhpQ z6B{jn!sDl*|Fk5yIIRJ0^hO_MMzn&NPj{d)@Ai;+<|uAn@(tPUvBgtgr^3gjXGLQN ze^NLa_Qm=Elko9&VtiTL0O>kf!)oVZm^0Y~qGyi7y{FwrzaRMF$Wgs<)t&1kaNc5) zou7rSVjXOAwi&u{#N6GWu^%~jARp2m#F5|K?ZL3Q2h z_rLqT?>YB8&-49$zGtA0br-ozZv9B7PGJs6+a-~NC7|!Cse&I~{kX{gGd{6$MSU)u zh-+J6=DI-arLGXMeG3zRqmQVqo=9%5os0gnPf;gdZO;E_8-De6z}Nys!BW*Q7-iPb z@4EwN=(JEW?%-l-RJxG}o~_1M6K}d((vI`pla80HCJ9<03NM7lks$Botd9R}SUfWh z<~=bY9pevxWbsNkqHG2oo5SdxtR`Z3hQs&O$#8bvB-nUoES3FS4;Mcygn}L7IMd7t z@n0p1v`#^5V{iDcuaYd``f1tTADB6^1}~-T!|6ieXfyJic!>NUnsX0BeEeGy6ko(_ z2{{2xhsEJoKq*#T%fPlr|LX(y&KsfTOD9q{(++WM7Wl8vhWhJY;fwWT!RJr^V9zlHD4yes3o<@4 zrr$eB8Q*^le$Aj~!j9qazE^0@`_St}e95f4L)1silupk)igCNk8}#2tphT=Y$fur% z&<&D;q(^7r$2d89@$X7%E8a^Lr){GhZ2|b-_XbQ@c968KVc8W6)f<*nY-iQ(y=k!B zp@M@m-BI+|I-47tV`2U6<#f2Fl4g6=kYuwg8fy`RJB~gl&#pV*^A}3^_nS4Yv-wJE zuAAX^IS*{!ae~C{Tn#TSmcnJ7=+BXHT+wp@3g3tWo7zKo_H``Ym3V~SVp5JXb9;$MrF4q;IlRL7$Dh6 zw6>*?mOTv7nvg-2-)un#$53)UtBzXaO~j}gVLax%6s^O?;*3dm@wLoyEIXxzuXfGG z+H>Ks{J>IXMb-t@@7H6<5FRgR&{PvF4CgD=Q*}(dm@$?s2En+-6moKrBg{?R0aGpA z;fqQK)QH??_C8)jB;zMxM@)-NPK-6#^6orVWz4~?N}>4s`~(UYUFp?x!nFI|R-^=r<*XkCOs=v*2%dMe3f%YhAi)e40K%6#)(|XTZnlN!1 z#0Bf%=H-d>)K)2MzFJOI-J(H3aD~Z=SHR$mEcT=M8e+Xg1}8D=Fg(~FhuZ7NrPzlw zaOez9KkbXlbLH{Npa_i~iNaG)kI(>>cGi8rGG5v>2Rl!xf_$JMm0vPUYD5d@!vEYc zbkkh)ixY>F`+gHKIe#encbi@x7QhpwvFPWy1v4is!izgv>5RRWY|W-&*2F81nX%y( zv1mLEqOYSsFKGwZo28Oll1G_xvlO~yMI5ZQXo1pR75O>D-__Mx2Ov`@+aJVQzhVGkLNspKuXsaKPdyJ}UIL z*^|lh3r8;z6|r8T@^>6JP~$}F9p3VFSt7%{oKFrgb1*nrk>y!-<(4u=#H> zI7XX;wdHE)47%O$!Y>NuBxaH+(fmAmG!m@P)9}tBl`18^VBTFVg-HPp)Iq?}|9@IgvZ1+Hgd(Rfc-@YxG?ny3qs^+O=fLz8~!Xd&}9TqH&p zx`?LcNp{-WcyjGW4`CC|!TO&rP>~x90@*@3^A+#eLtmnNjkr$`_>_Ct`-Ti@jW*Qh2nR>$G46=$iISpWq+S&Ud`Z!I}dXwnq)b1U2#~iWeqcSDH7$A3rUZE zCEJ^~oH$0DBg6aRVB1LvX3Q%`;9_SGx98i4zEJ=tO>Mch9qL^8BQcisIZB81i&>|m z%J71Y(#{{1H0{iL6!f3O8FEq}QdL04K3oPP5l12Sl_|Uv3M0058tAqo0rh38$@#V0 zh~I>G@G)V@u-hrdw(=MALNU>%C#()8^=FWSm-9jI$tp-0&ZB1449QP4flSSPoN|mk z*E3}^w=_={Qj!e>Z#BO`m--}Be|ZfJMl~??Qwv?3;fO!X!&tA+`x&*Pn)qj%K6cql z)Bd$5nMw^|7#QKXEu#VW(>FtK{?#MK@K**=7_lIW()hWAdRcu<>Yz=j>>>Kaz>vf$ zZy-~CiW0GpugS!~f`+OynIPnFo<=fNFzrmC&A0?ds^E8-E{Y4`4o-CEx^9Vat5&#i z@9WLE`NyTWIk(=zUj3IK^Cbcnr^(^tHpIYk6Mk-+2}Sx5VB+foHIY%=Wa7>-+9{k? zVp&7Pz+SR_?MZlVnFH%bq?mg>)4<{P5lD7e31+`XNbB-`=7^3a{+^qQmOmOuR8=DV zCo>)mMI^WrwvXsb=Z#2fy}-fBkPPE$I`x+{Y(_;`*s=#cJyWsKdHI5Ld2R){94lz= zT?8sWmoSOl&fJ%oCftud!dz+8I&xXU2HBh`=ut6^2JVYR_dhdH{M0_;8+-r)I^v+~ zO$yr|umS3WHE7StGsN0_l=_3A#{2EQ(ygr2G)^xp66w2V#b|!HqDmv^x8Tp3PnBBx>s+u6q zaw2WYNP>)j5Skg|O6R`XLa%(SB6i=G;=$2&I@F@f)ppO|8gr7-Lhynlgo?wB@*8yb z%#|=Sy#{(RRKVnn7EJRlqP5Fc(EfsI;`wtSv7GaXHRZiC*2B5Dc5gbieSA&dj7uhE zOUA%*ou^kmwq(;W2dd!EWgpC%8;66p|3QJKFqbJ`L2AT3f*gQ;t;;h!gUsJSzlUkmFK#-9Ri>}vD<+(~&nx_DDRoZyh&|1)A$A4D+jE#1s z4b|o%0(XH0Rz+Cjj)1$Q)OrXbB5nBJiq~=ft)&v8qJq2-6_h*BiTg_LLxO=6x5oW0 z-J^UO-UL$0_|+;kaSzVyO9*-rYe$PZ;o_&FcbMOUu;35ma^;i`3~Fj~%wi7`pU zm0zc^yKR3$mE|f7jM0XSX^%j0zP@0Ydn`3vF3nw^YYqRB#&O#&7tv`qE%42~P#U{x zKb&9Pj!Gwt=&sZ6hX(lLCet_TY?fAUJ93#(b;hOJu zxGZWN*h^34Gfo@vc|ipgU{Xvp_&JgT#vrhKO4nur8GP0GYOdwUpLmo{88a~3WY87A+RWYAG1 zD`aNw0iTz1P%@mwWn$8W2BO^4X3MTq* z!|Rb1cy!83SQNJlBlT`EIROfSr{CrYjy%k?G12Tqw(1~GPn`e@i*|vJ`4{>#&xPBZ z8PB+@KjbwI32x9YnQr|-aif|9=dni^+A9jM%WwfbV?JJB*3*P7tyMU^R~+;5d48hV zmfo;_1z#JtLd@ioK-Uc-tMUw2N&f)jUtfs_laGhaG*P2vuR-GOYy1~!1@|(ddCjDV z;l!8H#Il2!@aS{d($YX~>;0)~PGJ~ba z`CZ3sH5u&W&xUO`{!+6ZH`*;XS&(wP1_pP_K+IHsEH29ji^3eTvNs90edG0Zr|q~t zXRM%GW4fRua{wHoXOjJEr(x|AigQA`h?7Y*&Twv{!QPW$_Xbt8TOW&SzbT_r$wz!+ z+f8-iJ7Lk>msEI{G`7{hgq~CF)OK|+IxioB=&fN8)^3f<%l^@4Sy%BNKC>xk-2hG1 zt!N=4$?=u~qI)MEdK{0F@`;u7NFNK1uY+Jn{R2dsY{tl}Ci;8NDO`C4aa~R}PH8`i zuYx|(VBZw7kALoi`x8i|k`c;XI7a{3<y(nTs2Jwjp$}MSDh7|UH{qG#P2Bp}&t&r8ZT9LPC3^Wo zKbcszjs3oHFIKNJ;rGf4x*{zMF8}@s4}z}KbyF!$T4o4&#}7fksW?n~;ez44-*Ni* z^Z0YL1(Tik~O{ zB6u#JA+T2tVW((sr~Qq9V`c9UiK9yFL8pJ@&%qqBpfVByUgkqjT?PF9+(zDfbcZ8T zPm*1sjpS_TS*#X`#P<)!a@~4!vGZ#mDXi6l>trfuMCH><=Ss*?*Rl9#yBRAxFOs-O zo`wk)>QG+09EXNyaFyBV=&PHD2PNiVuiY2Wt0|yY95m6}x(Y@eY{(JMF&H*<6@TcT zKt(POHPv2H&&F%G%I_{3Z;Atnrwhp>V=>ei978o-vdQ{YlVFtv--Aq(25;9U)^Fn) z^31UWIy?&DVRZ}4#z@HS(uGomDo_tQ3{`Oq{qC@Zc`7)HXTA=isL)vm&aNiXZO?G> zum-%dx5_FC@a9K7aSR*7Y1mEXCKg$6g-hf>=5#ls zuk)K+D&%YCX>qh&N(N;r6F@;K8$ZvfVPAQkgE+rLyqemAQ*PIji!MiTW!M}}uwD=J z2g-@=h&ODCTLO+FsxV+R3F`M|LEr*8!AhQC9N#z_qHb>@dl%+n*5@>yXHTK=j1re1 zx(^L!y}{-g`#9b3gK)Ak(PqQ-XZSw;Dbpw1#%gWdY(q3EFd>BZU-{(X%&iWna%BVA zyh)0Cn{Fj2O%xFrJrWm~?O(>tIbz6dl32!hEwja2`J&v=LVa%jfdrbPqJyV=^f~c_ z+sJhpaxu=foc^KH>@wYQ=GU57+%Q2Jau?{3^_|GiG%i8%L(*VAXh(ksI0_Uk^aS%| zEd<^TPLP>Y1V3g=pm2r?Zu*=;Bmy_U+i&-YhsAlA^Ff);@Q^0&cVxqaM{}Xmdye2d zZGu1dCFst@8JN}Le}gq#>Qa3;-B|)_b?s2YiQybV*KncArrgKG_CKi#fP3>f{hc+Aw`pUc+o(^k#{*kd&y<~6I z3)=m_irjps!h7xR;bV2)Lm+BO#p}Ir*PT)n`)dQ4?zSY$b|;yPH|f&S6jY5)C2mpEEQRi3?Jl}-v< zjo`SxAO82~9NoK57YCh7Vdj_nFnh&VQhM5jD1J?#*Yh9S_D z=Eq^@S{aD@<;BSF?_<`wduNj%$U zi|L;qkd`C4VDxt)wxvxZpInOpwqK{Cvv;C@Tns((JdmF!*^s*GQdqKG0>^y#f$fGp zd@b07&0oss@cR-p=5@EDP6;H-YAOsIwzC%=s$56WxC7Qg&()l zj-Lq-uW}elz3RzS?bq0~ejW<{>_vrYAYV+>kl9*7=PsE7lDRQ3@WKjKOE^Q8(0+I= zCI;G)DzG9X4KxJPKxrET*1J^Tx~~&BtESVF=kH_w+a6-%?to6u^I&O6CRB|05|0z6 z*j?{{QnPO{?vB#XQ-9oMx@CqAio?Bi>!2^@9=RV}kHZfqanZvkF|+0XTjn$$qW$V2FJv7J zahK)H-gyv{Vt!84z>!i{k%pA{KBz0Slo$w&!&>S@!iuLtCAg6(F*$Vnp=NR>J%F&+ zbm)@{&D8L3o6W`b8(BMBTL>=Xx%7eHhJcmEFsbko-R`}J^hWI>U8mKca!)y4$&A4v zXK5xXu>=cm6r$m~-#Aq14qTTtsCK$SQDHT1l<&sxF{?4DI*k^uDP_)uNzvga^T1HT z4oVlB(j5yn5TV@VFugzq4@N#@ihktN3dJ&}DEKdFa7+QURjZCfi2dE3yjp<3kc#Q@?ys|cDsgJFA3 zDrS6INHq3+W3zfT)8kLt$+_Yj##v#Ii2mwgey(2y&z?I$YTPGc^#n+4QZn((JBTaa z)lz@oBy#`ZS{(TMh-&xxlP?voiQ1E9(mHd#;CSje+ACrQ&L(Z7;vzsv_!*q-RR9lz zc9L1{ER>Dv3aUkU9U^uzC$iuvl?_kC&lksVMTWb@z8CuEaPFqQD#E@&E6L-&oJsBz3#YQ}RaIeE)ab;}5Z`fkFr zY5v@%rq@v7u1131r*L7s4jY&(!-aLZaXZU%;LK4eWGtVcX>TBN^+Pr)q%@GkW!=!% zn-6bNKH$#l&-omMVN%x_MHkFU@^Em;m2GCt!MJ7_V0-F(MlxiG#vTa+Cc>#V%Y1al2UfwIc|R^jDL4 zA;}o8eH^z;-G(PBwQ%hCBs^L94&M2tk=(NR+=B=o3~5b31xpET^viry+;$RI)Xl*@ zkBQ*)oW-mfBd-6@c!AvVvzVYX1zgDnNPaKRnfTp8arIR=aa#sNOh1LH-v{V@DFeDo zi)Y|Fr9iJc3^SbFF~xNOn5sM`Em;XTJW(3DY7VkHAI4IXbxMM@)-&O9Un;I&G74S% zdEWQ95lu(3L32?WId=6ij@O%oYIkM`o~Hi9z+J-F8nzuoiaofdT`#b6Sd^3gHkW;) za2j%T#&H|m7jhp5`$>hnAHJGA4o}>l&AoV`&sE0G=kkMR;&|i zdUFeJzQ{)BNqjcUq%m9s-~XJ6n@P$Y38dvTfT&9ra~L$am6aZX`p}OgrQjp&$bC$O ztp5XXfhhOE^eDJ0Ch|4>slM?2iuUx`kenrZdHbr@`t$sO5Pj(v8yq?2c#GoEcl6Bl2torZ4Hnr>I_M#M8&u0jzU)}-o zd0+7dorP1ij-h3I4y_;a3&$!?K&?qlAW(h<`a4JI)7|B;+wv9o#%9A}E3RSe+Ocp> zr0%;z=)>T-^skD!{~0j6r~Cz2hyfMhS-Ok2clq2FyHe5!>?57N9Q6U8G=Z21r}pH+&noGNQd+zsIPN6vFGPDDJ~i?uQeQEmSo|$8L?<&p^Qn!r6}K$MUPe(K}X&(#Q95c z?zx)j*G zs|~kU*do`Vf*0?NptXWF9$1vYnke4Iq)G>Nm3J=HD7{J^l$p`ff$NxEy0@9zr;p&~ zoL!(MQDkGz%wYe!^qGqA*$v5i>p-(u7PdXUO3ZTIsC$|L+2pqfEDybc`&Wb@EA=7k zwQwos&Pt{70b<0Zg=dbcH3bcujkx*W6u2>m7odDzB)iqN5O!zAz$u+PG%N(_Vg7`E z3BN)tj$LPOs76q^dlH;oV;37VYC%Vqx^ov=_Hb*hui)EH7t!F99h^uz1qX}6xDq9A z?%1?roMyp(u(X{H?!Og?(UGs%^>QWLKOREUv_9g5HWN6ycLQ0l@D#myVJE66=F#69 zzJt!&D8Yla^@0zhM%*z@my=y3!)4FkK&2nYPNi%RV7FHm6*e2mkU2u%$ic-zIAag*37@ZrtBfYY>%{HWRRe9LZ2Bd#uJr`* zlwAb0+JemfbOxqu$%2>75mY%>8Fz=Yk!z2F@rgJ;qi#~?a*HSMJW3C7c=(hUmp3-} z@A^z%XCFfIz8a8kP==&WN63P|V(k452Fz+*Gu8p(XrZ|rUdiJO}6WrZ%WEgxYu z`bx=cjShM!^ALo~CSv>RA2uedXQPXfjG%jcv|zlotzgxIYI6CLJWl=c+~(iJzfgZN zSRkrvBe>sc2bK%;NL`CRIj;4XazTnHd?1=ubxFW;z1L)JmJXQ6J*4B+zEG#b$u@f? zC~*g$+(*^b8F+H5H`YjYP%V+J`YHQ(|6uI`W>Ui~MmQx_(Ak+FuwQH-IA}VPe1EWl z9gtDP3z=ha*3}$RJ0XQiW<93I=xl7-IYdM=kCNueS*&E01{L*}5r~g?2^3?BVWr&E z2H{9=9KErY%DinxH6Mly2t{M>ZyVM)_A2aXhyl5|TOd?(J187*1NFcU>;oU6hL6+c z;Srt5VAs}1Zmv5+N_!H?6R+9W)nP@JM_*?CXp~Twd-5oLcMN=jcxdYuK;Y|qrpY3f zynHbSzxszaI9|ENL<`xI(%*}S*1u9xpx#ThpI)QwgH}ZJ$6HbsoB_pGHt>x6KHw9j znAJ_6*>P_|nDuF9T<7*bc>G5qR^O_F+~W%ctEFmT$*Onsp0E=9GS;Um%YWjS5J@ib zfHG&g#}-wjrqchuN8`R*i_t@FF41o|Pv0DCCz6sqM88=U%+pU3&%zn-5x2oIJuUKW z<^vkPuA0TWYnWdLq+qXTKmDHZm`sfSK94-a6q=CvmtibO@eGi4c1cV=k#pf}1PfgmKExXnMiI zD~+D^RCLcgCT){5Np79Vj0rHIcfX#ex6;p9D?wAk!;I*y{^!;>T)RZdepWWh^t@D9YizG66Oe485M;OHh%h~rc->@dU z?=5stm}oy2<&tJi<~}VujFNhi*yB16(#>opBn!21g= z&#Ggb{x)bor;V4AU(z1IiuMq`j}{uqbNlM7&KzMYM% zat}_D5Q0}XtYB_-8s4ehNGhH!f2*@^4ynbC24jl<()grBUHjhx@Efc8dnAsFpHiNzLd(!xH z7usiA+1lso{60$=Wg=_vWn++{M`oT?lVNV&6 zo;d|B$W`L)Y%R1IUxY>KFK~Q+DD$lN1nYPCDQ<4Bf!MDCXwbfZZ&F*K`>ZH#mMq6X z|6*|Ui6PT=m*dQfK1`jP64nUCD*n<7v-q?=TJ zT}|e#44}b9>5M}sOH{uX;{~T#e649t-Y%@ePm>%VZ(l96j=6&N^}TStHx+DF^Zj|{ zOTw1q(;sq)s4H?1j^|0j;}muJvTXpK{E){Jv*Kz0&t*{VJPxWJU#C7ai5&3!+>quK zj%IB^(0q3fR7%dmT*F4zQKOn}y1WW!jCPX_n?Ojh<}($ps$jmgGtGJ6Pj+k$hq#M& zr2khQquSyQF@ZwxjMxK{Q* zdbv}T{nvbxpwV(_m0AjZ$7LYO*$O)gUedGae@GIHP@Qj1G-Y!k_PO`s!qPXS{-_L? zCOeR?H4Wgx`(%xI@2i%?ET(noPu>eO4TVZdz}tQq9^Ueh7Nss=-CDHqBA-9_{;(+A zs?mUzucpApka}<_R{`z+a?#Vmmo0PgqAT6qakb?faF({AXD@!iJJQAQC2+_(>~s~a z8M*`rH4td8NcdIw@a{;nQApxWRwFP6AG z$zo(i(&7C%U-VxeN%pR^2Uk}Q=y4XuFs*yA?A~)C|B%lS_@K=0GJZ&{3Z~G)m=02R z(GzQzj^(7clnJ&yNEGa@cn2QK#bCbTS5oqOGpsDq#B~zFQ0*0m$tGHYfNV2C%6%6o z5Gx06$pvc8nd5VQc5(W}D3xv$;j|Q=V!wC<-jMIdA&Vn|Id7&5dehVeDwUa3v(l5? zbf3id%oz{CWBtgG-5gl{PzUtKHWQz$G@5s~>x#$EQW$fb zg(#0vdU)84#*{>W#ls{}bUg`q50t>KW~QLS{TwVYt_6vGGwBY|e>BZZkvpI}9_4M1 zl2zlTkQGjn+~V7++>f0dIC+d5h+1Bv>nfr^jU+(zg#cW9|0gr{V;S`j6(QB}j&OgA z4nCWzgr|CiyN0mF@{IaqT0Kcw6S;(Z9HGZxa&00bzYOOb-y?^R$Ft7s2|rg z5XU_whq>|PYfzKV9GhVN9?yQX_rZ4`ceB$=2ng?*rT#!vZNI|jaG2xbVh`MOov&ePjKTQf zZW?I#nf`wL5Ny{>K#{Of{BkA(?X(`CSMhnQwY`P^jgf!}c{y-toHq*jBv6NSGe9;t zl)h)H=rJZAn&@tb-+PJOas4$RCk{dI*&Wo+te5-}6Tsx_J@m|Y-XAlg5DZ=|BeuHD z_490unfx>7pvAWZ7QbbPOzQ48*%}(+MSR zXWT(s7gj{dMe-=n~JJP{aid_L!ZP#K_hMBT>GKPZvsp<+}Me zttS<>Je3fnURlMw|F(|qw^>Nl%Tsw@^LE_P>kHd%#bSA>0-C05$4Wyf?uZ;;m*wh$ z@tz#gKT&T@E*Y zDF$67aqi1CYn0e1$CSiILFlzUFUT$j{Gb6N$FjRTgKPX-e!HWwQm|(R+xtYVa3<%C$gu%(f96WaM?=-=vA^OHX&kgnEs^oTBG!1qYk$_)DzEXwbN~9 z2$X>;mvlr5H_0bstc|LG6sSP`WdWSqUI7*xo#4LLF~+_&2-Q}GV}B&lps+{e`kBMD z)%X{f8_(i%Xm3((O)fURct_{#sl%CKvDD4t5LI8in+vp_&sE%dL(aI3$B_M!WRJ>9 z9Q13Yr=*&h=^LLw#Z6a1?WnCyV~Y}%T(%Wp-)<~9r-Luf^Lx4aIhgVH6TRq(k$xJ{afeRcz-O_9AI8ZimI=CxbBM&4X!0!n3jL*%O%}JF zLLYS*IDX~}l&yAYc=w?I0zEfF-+2*OJi!fT_4K1tsT|Y{?4&!dyPi8zJdL$=DlNskAgt!E_7J(^trDO2TmeP$89jcFoI!J|xn zhAgq)>r4La_hN&0T%doXPSbY%7xcjQEc6?bOBR(Dk`kQ)x@C_&&(58u){Z%3g;p06 z8(T}tb~NFIE6TW&-lT(Vk$7nTSWF1kCdTvE(9Ka-Nz1HRq_+Mpk@t=w-f0dr&(54g z?p;s4`j^7rqn+endjxw#dx}7>c%h)|za6k(*>u)xU>>RHwWCirCN; z%gAzGtGpyu%I-dDh8fc3bV1)S@|$a;OyNbGay5=wn7;#?Ittk?r7esB|J=jhjmYGQ z0Pf-Pf+@|mu?0Km zdb6eU_1uLplC%}tPcNdMP0O&wBbRA6PKDJE%-FcChWJpko;dWMz+d~_@JyH%Hudbm zk#WVi@I@DDSFNUdg=3g9a+4IQiI8b6QS5fnztm831=&}4o&@iFN2fk8ft^z`VYXvc z{qfjB`fXtvPva=`GFNA`R_}#I^vD#Wxn{&=^Ax%HJ~B09?_!56X}^j zd3sSp1j?3JqT}wh_<4si#`TTRzyAC2*T3V8#?d5_w<8nQH|#}5zL>U95j0hnpvU7f zz{Ifyw9{Q+0c+Jvege7CoCZLTmT) z`)|cAYI(Gmq~AG7i?U{+&(CP~=%5^<*A#@ae9z;}hh>=gPeS1H!AW4XSwi4HT?5V- z%_6BSX*A5pjQGTz!+j6jV8nkJb?S=4&m7NUE%^gB5FyYj6%jb6NkNyxH;9$Ygr0)s zs9pb@%2)O?SH>4mx1}%XVbN;#%}5cQl{y`xmWrd5e<1p`G*Ef9`%I<7Cwd}oh@B`B z$2?vt2`^VsW+>yWP4szrTp}Gxewa|=l`f3xGEz`~%aOP$yk(v{Yyf{jEGW3FWHkFE z`1kw{O04tQgo7O<%uxfbs1(!L#s0+lst#oek}%7t0yi5MQ7w-MLDq$Zg2_Tsg5NW% z;MV&-P`joFoi%jP{#@ICN8p}4yA<0U`^lT64e|-EiuES{RiHE2kgA1l?M?(ny?6zxlG^MS4D7yuh!A=qadn=?m4k z9)QR90dARb1eWg;fz$Ka$n@_9{M{@6(UrD?Y+k&CwQ0*UVD%nZZ8l8A^?ud(C4L{u z%eun;kES$a;25?`r1CY+0R6)AOKbCPQ;jQUcpbK#x%*9wn_4u4Q}1W;dFXPiocdcf z`mYzgX|G#f(`YEz-8Kk83EQa8{IRf|M3WB7*K826< zyGMK$)wk`ao1n}2dyF8Ld52YZ9U=wGi$K};5_#)!1n)GvVWlHbotEX;UcC@=EO>45 zqaJR|NGA3(IrfO2A;GKOL>&f5=zd-^x>t@@J>TP*5M>Nswz@&fOP=1kIuDMM6#~g+ z;sVXTZuOBLG-&VwM-;0S!A}oLY4N{dD(2{cwyEwIHLZtwy%{AAQMDj@BA1j&{6|M$ zXh87sJGAb(HI6;TdoUM;;=}{(SPL;&MH|@H%V)rQ_i<1V-9k=w#-jbReq23ckY$E@ zh>g1#8hUnr(xUuHp)KVPb07PlJ(C+$*NuzI9edg-P$3- zt+^MA{@f6m!0iI%K}uW>ZiQcS+`;?rc`|xh8;bApDz_;I;XhTOQsWZQ;PX37%UN_e zGXrU*3>+mHRL-uA{&8%fKW!{&|6y(Z{F{RhJxt(S7O=l87cnlTDfInE0dD(uoXYH8 zNUC%Up!ix0&)G$SVYVuoTXd6;Z;#OTRjp`HVMuSr#jx6iC&>j~%ZN70qWOHikpIaE zE}Cz{q?}`*Jvjg_SNT!Z;1YN;SxPW+;T4hHYD;H%Z6x+PrD)de8IZ;^hnd^skv%>S z4rab3?z0KrYmA{gUyjG#&*fz4m3>sAri{!w@s(Dk?!+0dccaF_hs2N7g<_SHY~3Az z(TV9K$7TZcwDbnMW%_tdIE%a){z1eS#Ij3eo8WxuEL>r=jr=|J(}s_`hJn2wKx^B5 zxG|{=)kjZa=O=wk+dB(8JnX<*3wuJ-BGr z44gIUi{{7q9J>quu_D1a^v>w)hN3`EnD|dbF#E|Jh|c3{mEqqskQ+^-R) zI0RO!|I(tuMW_@cinC{U($N!-8on;SLFeARLo)3)!Qb?ar1hCR_vDx&xAMp|Jd{63 zE=7*BF&ru;>x5QeP^=<(Hd2IcN!Q7RW0p8C>l<^=Y7SVn83?+U-vto+WaBbHi(bfI zL8rhHvhSfYDs0cDk_#UZ&x0ATE;$h(T@I(1SCEK%ZnSXr9J)1n6Rv#TgOSGR=pObB zGt+cYYn%xe9W#T|OTU0WR(_{`2|sDEyDID(5aWJb>p;6zKFHp*g(uxsg1jjfg5}4O zV6ma3z+dqJk;^|!Jf0atjk^=>yH^0>IdcWl=K6x!<6ppS+Bz#`^zA?t|svBi%~AMxSO4A1Mk>-Ic{L z&6N#zJ{~8N_S(SLXZqxXfeDO>8^akaQ{zT^+i{ocbcpY3q9fP!G3sd?2~ahKstqRa z_>wpG_e}}@^HRfD{e6u0R2lgGLJ3Nv3dq`D^57F*M;*H!f%YN+Y@c8OqtoQ5t4n0V zV`m^CPhOGhbKgMqlD(k#`~`LW?u5G(B8bFY0|x5s@ng3dNYy!UMxnDgy8u=0*QX#X zxv$PMxt>J(Ssc@<;YrHsxANY@l?~4~NML&LF37IhKt=x4P~)8z?61sc`1ja2P9^Is zm0TjiHGCK1?%y%xy!T2F(=Im>c&v~4@@_LXWe%@(`;_9^)L^V~2q4yncG>8t-a`#D z-siG54}v6rkXxVr(CXTSz^F%4(JM#Dr(0>D%?=@H73N+|(!=tp4LGZaXQh{YrB41O z^psNxlQ>=rq#uproE3iH8kOgmOJ~pB7`9 zRs(*y5DM;Nq%iEzakN~w6xprGjEv(X5K{7^uf~Ky-9ZuXou&sWyk@&@t3Fu&okMFL zq=MS{nXD!AiVmo^(5Y{8=)43`>!UqGIVk_OdWHLOKz6R{|bhr?zLRgvw6No@rURF-dLm69sKh5)Jind4 zo=Cot6r||XfKBOd=#?zMN0VLPe}>M(EvGk(;_qk7sh*E?~NkT-3$O^yn4|H|C@AW?CJoo+mem<3_SHnx4LLy$- z1&S(v;k5B8ZtzeKv-zGe&dORyK9})38P#@@J}(^Je$xY5GzyO&_<(300|g`d;fO{6 zoz!h6h*0SzPzVjQ*?0S2Pf|R-9)@y zw2-{wvrRI|hhU~Y&r+;EM1-AVU}ud2yU4VXDQ!UT`^#njS4M1O&;mAa%vZ7=a^Qw> zAn{(KO+>;Z>8tiFq)OX>-MRcfRwQsdyZXO+YID>D>r#t|Lu&zzvDiXuj$FXl&4W0k z@eHjMw5WpVWLD&n3_D&|fweks1)Hp;K%bq!C^fvMGhdhDY1avKvq%T?RI{7-C+iU% z8wVnFDOcdI{~-05A;ncJSK`!-Qt{#G6=cGCU6QC=%d{oD;QzNN{H|^mabHq}bCRyn z2lOiRU*WsOoIaIq<9WAZf(0!)7X`~wW0iKTblcUJKH0)bJUzczpg_oag(VgiZGgD9KXC3_CXx-mec9kqNYK z$y5lh>q9FW2k!ntQEsRF5&BnpJzT6j2qsVVkW66(o3Wlwkba^S&m1y?fw1SKN%a(d zwFkVmd=UVmPH^@7le?;odvXXjV~&h8Xsl$nD4yXkbVLn~#EB9@*K%EGlrca>2suQKcHdAXydpKG!%(PU9!$AKAxGCgI ztey-Iy#hUuh`dTKRO>)m$X$poh@o}c2FQij=k$I`jlk#(pT~RrjJgg*z|to?`+W5q zm{+?Ve_v4)NWU9P7k6IAx%NxB@!Dg#-%HX-!E{qdx5|MnN7CSVGN9g8H<$MC*hBsHFHXPZgCp`oqv6ZXmjoa-ivB89bdp;I?NXxarQM+gw+&L~=YkT5iT_m%pSJ zzuQtF(HkhLQH>6*yfy6|?``?rgFQNT1!0-9A$y}ZgiH&8bL9zWyDWtq+@yhd3o_x! zyarM^))h{boxq81GiqvG601|kufbZ!{kZGGDZFuY5vSeLiGjSorRH!m{jk`MQ(nHPUOBONC2D;D=rZww}zOe+K)Bo<|IBz8g!;9xY{O%st4> zSSp9t|LtUqVx3T_?J(xXhJguGFbg|hVdbDSD^?puJT2t707)ZO?L`uPYo5VvE_e^- zW>@jWRsq$X^a~18>#6TYQEKLJ3skI@VBA7?RBn{Uq9b=O%K0L3^NXyu*?AMwD=&cY zpV^Q&e5y?ph1hK+3%F~l^B_w$i*CAY%3Ymd!D(#LW4#oOxM_O|$&438#I4_>OUJ3!;KIT>SP z?!(lzP_(`VTn@z12|8PFiKqhJdNYbPn@<9*vEd9@L-?8>3;Sm+L+y24@KGg?-zz-C zzGGIj*H}hIXNO2qj{M)s!QUSj#en&v5dquTtoHL1c<7dNv6+EqfLWy?5Tkk z+&3qJJ6IvZ`He_}-6RXajjC&S!EL7?&aD~$<}AZ$*PoGpfl~#U_jBoLvJ@oeE@n7RRfs9lC@Uafe$(G-6`5>%${rmekF ze9zm6^?nwJy1g@LrGzkRmNb(Z{M*lId)6~pX;wHwquMfbxM}kr0_bP0fca>P#JmoV=!8B#c zBK&@}mMCSMr5axPXuI75Tf}9+lkc6`8a^U6PcmWjwKOAnTU*e-If*X(nE__ik+6B4 z56*e89#4P^=Qsx9SPAmt-Ax-W|yB60q_9y=J zodCzKr_-|LPExw?nBY#L3JRq?B93jtf^kP9;Hrudl+;IJu%{xlxbXK4-#PFu$%nQt zlY%lo5oW#RYMfd<1AlmPXft~b*>y1!eAQy3G}`##LAIcjV8uI^4m#1fS*=GrLwK zVkn)1OO2P2TLTVUQL+wqOur3l$MAbc#q(%=M*;&?7&^O7JELc`Uw;px#^a3>?4 zy8ozSinJ))dUqS@j%(xkJ(2XfoCbf6s#7O>6)M^m3>*Jep_}&*UHU3eF#g3Gx|xZ< zc%vx#xHY9_@ykjiAzO)_-*&WWOQsEbyGe)MUHU!!B@r@_f=02ic<04#zPGpr2jrdb z;OQ?YzFnKP42oih?L8(^=?E-5xdmseWub!KeAC`%VeZ$=)UkY1yQ|q?&Zq7mT~@O;8YCV7D-hUU0{M3f;rE3SifYP^LhUut2`!y#<(ljM}_+?c~n z88|H?4~>rY(4?XuBKFsZsWA)%X6t;xLG5zz&>h1joqdJ3EF7@r`!e{fZOd%7hL|$$HC`gFsv_0a6@Y! zu6-|zV)+1j*Zjugf0uL1JA~-;iu3gLpaPol+yJ*@6Ug}74;Y_cg9W+9G;pyad}%uY zV|V3}=8`gcrFJUwDJl^An5p=trGeRbX$0!vK6o$Ig@n`yR6O2JYFtFvJxbs3c!(8> z@_mvo8=UZ|Eui!B4baUyc+4L+lDyg&Vv{bx`N&!mC)aFvbpIbd+2TdgQu)rd^eKEP zInOFUlizQx2*!@{YpAMl4`~0Z!?6lF_j)f zE%%1^P(4D^o?g1qRg~TTr588<2xA(0F7aN&Jy`8Oo2nL{gwq!X@K5gv`ZX~F44fC> zqzl$Gq(Tzbk9ma;<~NX!qM5)ZHsa99&s4|!0;yW3gSxk($-&ZUhz;fU)O;q;Uf~C* zZo6sYbGeJ&+-<;!*r=ktI`5fY_JC-G?AMx)i3Hr-?2K#(l0xh0UL5?a((*eD8XmdA{_Kb;zh(njC&Kfav zusk9Vu9o7&6lcKV>HKqKb^%UZa~QVEU4)I<>8K_3gV8#W4h~E7$3axeO8a zt3?8(YnG54W)H7+Qmi^-)HYgqj4_YKXpj(?g`&{pPI${9W5Yk zMKKINoQaLX%gB+QA;|Np#vAdo@Yv1+(DK88d*-i6R@<(jC8DyNW&3{II+P33DtX4+ zKSfxmrNUlq?W77(1H>n)1)5gfK;x3@c=g^}5LKwc4QJx0Z-OWA$tCo-$otKIi?Y*~ zdBOT$K{ZdN?S=2}A5z^34~VAUVi>Ra41Fi`l5FD)xDXqUQ9n=CB(#nYvGQn?x4c8@ z`_GeaS#NQH*hZore;39z7-Nj%dGsD<~*xth-;s^_oZ=k8pheq0`&A9tZH_EB`VloC8G8p}ncWC)IrF9EIdLfki# zf27jz9az;LwpMKqB`U$j&?UWx|IB1)%ja3pJVA$RpxzkvwGYljF2He*!|Bcwm*6wc z%gA_dh7*0Bfa=jitjCvBcuFS7s0L%vUM1YPa|Q^mH{rjt=WMvwc@Sjoh(Dg~0H2>l zu=?X-?$eoZcywP5%+af&7r)5~zS<_%%shM;g(f|-W>+@CnEiJ#cZlbjxLbfq;{-Hm z6~jq-)q;Zx>ZtIu9QzAy;!(kWxcupPoL-%ZxOxbM9(7}@WDF{m+$T20`c%oo8^-(j zVBIZ!jEviXw+ofoM{P1}!=v9I`R^IeNEc;AT};7w-Y%wbWi9UHb1`D-9mrjNin2cy zIs4a1SW#nzZF?HfTuYDb(bi{xc#3z+tqesM7fT%ZRFAjqLuut3I~*9EimSBM*vl_i_LBYt_W5fK zcIQVUHkWIKe4(GP@RT_Fv0REZ+qD8j%j3ZOuNOQJOGEp~7jWH~-6&+C&FzYBMY$n) zgsxhAZzj%p#k!-~(OUd^@(RxD`-{_;`hlcr2|RU7heU1y*6W6mMOs35V6O=~QCNr# zKk*!HEzBnSue>4S>!skl^#pirPzy`0sbF8h0i1DY4|>TYHmNpM!`!JX7Eqkn9>$#bV&=u+8?X;-t* zkJ#e3fpoz}k7T+vaU1SRNMnv~i@@if?+KbTPLY=b3vgD@A*_9$gL}o~xTEo<;1Km3 za>Ux<*4`GH`lucA7f!*tm=Jgv5DIZoIpAIqkIKcB=(czOP5CqY+w{-4%Jv&-$X-Oz zrxG~&coMeBoWa7p1(Ul0#{v3R(zgMX3rI8y$vl{(S>>}o3tFZ+3tn#6|tCX z7=Z;>*PxflApP^#maBgoiUs>#pwQI`Fxpzp&-QHabgDRebDTCiSNkTsS1iO`U#6k* z>>%u(Qd%?01wf}|FoX+L;8VE}bgt+^%LNAPaG)d`a7vHe`bd`Leeta4X<7F6#0!x2 zUL3~x^wN?;dGwdQBeU=k&!xE-j2UH{P;pKU`lqzQ*R`)<%)&!3DprZNJTq}1dj)nM ztcBt0G2qra3_7}Nh?_$?-4=6-%FR(AS!b@%djGrRn9(b+u^h+#2pWMg@xiome-mn{ zNphd>zQ<}l&+TjfmcG6{4Qs5{aauBdoK9~N8k7d(mjzexzuCTcX6*&M!5&086&WsT zSe4{Hn2XnflZobVH(Bc1XH&@gsFp2?#e0y9zr^d9uP+ra$9Dy4jcsCj4~TN<+b{E*e(*!yE=d2Hc1hDm1+Xzwjt2`=K(GG-3ME)bb_po zFcw~YL$*8@XLGg~;EJ=Sv9IMf{jxg<`yC><%7II`K)`z#eVd3=SOLg>UO-i3Iv7^u z51c)dLe|wb(36`DIJHPU?qc;@dMuE|?&*tQrU?U~pXD)zz9QGC80W1Xgmv~UOteow zjo|r#J=?m-`YlEfxqlZtJDv{WPlF)wqXt{s`U`c-cyuUk z@a9qp?z3YKXO*FX*}d|d%Pv(ouq0?Lu%*UX(j=?;ZLJG`SVF*0|d89M!RqrLH$( z(7S9oJ;l6*H8VZHarSTe!}SH~K14y4CRiqlG(&OK1bhW6*t z2~|>HhV~}IT_J#(UrYsVHP&ontvN}!9gSTx9C+_b z6ju6xKj#TqS3I%#g(hur&c^Xe7qYtpH$tXH1q|P_o^zSJ8U3frE?xeyZp%CU^Yy{qGoC1?mt6;cjHId4zLGw$p-0DPq=Ji}jZe!mC zjQH^sJ6{>H+Tyd>L;F9$_q*p&Ex{S{ZC%*T3m>sOCk4YR+Hg)_1FRQ43+sm;!96OV zT?(_@hu|0bluYkY!w<5n!1LifQe|R^F5fOw1ER#e z^ZSWyXJW|giAc&OW|5sf2{*& z3-YQg@Q8^%JNNKrF#KUkCjAS6g=gE~y|V~r4#8*MG zTFE6KDBwL%O0($fVnE2 zBs6~ss7&VBU{;YB*Xlx#70v)<=_WclxfxFGnhOEngkXN>AWYb_0Ai=jWCjj4LwmwJ zFxWbW4xPFHs|xmmsBgaD>#A~?epZXkpZo>(8O>lzj_|W=>xWR5wuIVkh=PcW#nic7 zh_kO)0|y`6g17_`{1^6+nw2|~u>(b5J*yvf{|tkfJ}<~8p0nJPlK_XUZ$f|5b2!=i z1QdAoNyKs!D0E&7)kb!lT3#S0?!Ac9Wb?r=Q<9Ac)MSrmX+nNw4t?D*6>G)R*z89L z8DcGTwYCcTBFcdYKIe*yZg-N~kFAL8`R$O~y&1aZuOZP}e96nagZA2Zdgh54-hB}a_oPR0osl&6!eblvY;z(`H(o&dRPST9?;!0f`pNsY1L6J0 zKDxn6K%}1Pp_sopy)Ru0qUu8Il#za_`uiSza9|h>m$i{}zBfbeUBp`t;_%^}EJ4kg zW#G0XfoC||(2`%xt%-+osTDIP!v-`PtAp--8nQr)t9OJ zb3@?wv=U9`8*}Clw7Cw6pZGcLjv(#YdENna7=FHOh9xfUV3cnLtG~CBNcTKA`SKeq z6gI%gX?pPCr#<63;|IB$d6+qWK%21);kyPqca!${B{V#%95!^u&;zs1l9R?kAo6qz zXdl}Ob+~sIM9yoqVv%IaSVElb%ym;O7POB(dO^yYtnbw4lLGY!gFbnUv);Eg-;H18DN%)UjJh)6)!8AV)oUW=nmGI+x82~IhF13UW~NMZGe;Qr@h zP}pq*2HW&-x$sw;!2dRqL|;+ZoRLrOS}MZqyV0bj@P}ZB(PwJ;ZWG1~PbD{gPQbAL z@`?FK95L7?!rf6-Lz#&E);jI#gqymIaXh*mf>lp5hXyU_6BA#Em*D4IT~7oXpU%fG zbIWM_ok5x`B8SO~IpUUMNI&0RLibjFB=4`e1LyGv_Ek64Y+vdQV=Goc^7vpu0N)L0 zt5YR`Uq@lVv^FZ4*hF6_Jra0q8$-7FS&^{)dfd;gE3t0gN=`$wmgWynWHc>n7~xED zc7e%7=*bE~dBIs)CgO){AN%8&y*!8GPzIDiDaq(LfjiXP@PSGcQyTLa45L$+6$e^~ z{?r##CnJiwMf;H;^QjR1)KoBc(q($bY(MDte!|9uWpw6hYjS6pr5?{+aVF0JjLpyI z9j#hiOq4%;Iy6d(Ti=q1B1@Qa${OB_H4-77vwD)xQCD2r195tjz(cN*#6@!ui`Mf1pLQXf%D4pzvb@ZoASb$f@!`v0G#D$%O>>yXi3{Dw%T7j|TEJ;6&hX`C=2g?XK%AKzqOS$%Q;4u`yGa6n(HJ?LE5EF(~%{NI$$$L7ui$ZC+E3EHq zWK>%u_<6n~B)exo^}Q^xKN0{(Z3s?s48aLW#_)On03-A`1ZTb7As9UEfxm4(^7kHw zQ%D@gU0N-{?fk&AT4xtBKX}*4y4!H!U_L#z%IioZ`jNiGX%|>Cyh|b(qNX@(- zQ`;v^I8HYgMjmd3j;@oWFD{)1%o-2p&b%Qq;VBr)yH9@oO2LA2A;iTahAEakfz@h? zL?Vqa`;P^wxw?k8h_17ez9t`&kj}o!SkqtQnD3P9h>j27Fd=x4^u$oZ&Q_ z&_bx1JPq0nMW&zWIJ<7hz9z}Gzy8L|qqV3avmU)C>(QPqmZ&=39bQ$`f$hRHnyi;g z!|pz!FaJp5jAkWztk{nJ-KY<@cOYytm_lC9FCr&GQW^Ixa`cL?K3jFso4s>ioPDwB zCEZ+^kE)@@aFPwgr%u}3f4d?C9WEgt$?vd2y*I$~OiN5yyp^Wv-DO(*{*n!9mQ0FA zIg_zH2@(_z!m-#C=DYIonoZ6;L$z)l{4^hfT9Y|6+^q#Up%Hj}=`TUAzK39_DG5Km zZYGnQjv?hzh?D3KqB@-t%kgi})vk{G*fSZ;`1dTW9u>|#gqM@kRru998O}0_g`8%_XVUVw?b%zIQ#0o3hci3fxdFEgm-~1M0tLe z&HUy{n`td!7!}!uD-&h8FY^(7Z8e~5%|b9URKWbH-vyh#){!AoC+6Vq>)5E;g*sQu zufKY~4r7<6VEMXgboqA{RPS6MLB+r6kVzHQ*?WiRN2TD2>^IacT^DbR6~}#nGVm!w zgCswfK}~-hy6?3({wqvlj0aZ4XaT@_#1K9eiy2@?}O->9%T3VAP_cM!o01N zBKf!8F(!{<$i7Rt=$<{68T=Ge<<>o7{X3NW{+Wlf z^PJ&cRWbS7l*0stEC>IaS>&46ILH*AO>X+wkq75=Xmis7+BH|6Ioo4~t_wpT?b<6i zBi1e1(ZSi^E=RDT-2(4DoFl^wGp3?BEWN#Ywe-{A6Kpu)Gb&M3V6b-+`>+IrFRSM`-SQ zVd8VCh_>CF3#qadn9&o7^VTNe&B=dl?zF4~`R$TWYd(+bk6*%x_)BnS^QPj5Qd77T zc9afH@FF6f{PQ%f8Reoj3Q{u?tiM>_26r(H8&8cwQfsKm`vQCDnbBgBRj`H{$Xaq5 zB~qNK_XA44x{%>{pBTg8i^N;el?|;4W$m5^!@#mM*r`Eotbiv^wqH(_thU)W!f1qXk1kUJai@%O1(7%k0%(aGjumKBE+tVGdK z|3^*1re-?%_cr>^{R!;2_ZJTPH&FKG2w`gX!j7H0c<chIx>X#E=a-Ltc{qdR)C7p?x=Xu z8~2Bt5?F291I@t~Y`SI}!?9JHp=MSfF%O8Q*>7B_y&h6)kyTuZiw$>Cu^A_CS%YE2 zbEw<(sW`FVGF)6Sj`v#fXS#DfBQ`UMHVJ89JnV;(&WAQF|7{@UCsffnIzV7{^`6bX z?&IW))iTm`eKKt6{z&^bE*9kUoFzvUWiW8r5iA{T#@b1ejKuFoqORgb6^qiSAlni* zY`tMsEgMA-c3&oUXdb!v^mdKeJ!iWA<2hU^wrBNqhpc$`$O_k!5VkoH{-;t3(P!wo&Jih4e%e3-PkANL5S+v(Dfu%wOpZ z`G2mH<|7)ZU4 z!mLW21P;!#K`5Y_n9W^_jn-QDci}IRG%`f{_pG8)7wci3%XBt!4e!Box`7`TnxK)) zZ$U_6Ief$}vgWD~t=p3$P<(NZYNqVOD~5eEeB&Rw<>77IUuK7i`fl`t(^yoMlfV^Q z)OfZ@Hu>>q0bVbNq7Mp}K-qV7tY+0vvT-9B-oKmPDjtg^@>j^&%}5ll+hcKFH}rlz zPv16Pgy`(Wq%P|;*zXlZwm_RpPrJ;G?9An=DgdNp;CYW9$(S znWl+KQ_9JgWx@0oKLc=|G7V;U#bdk7ESx2Z^i+i?dv0qJ-1iq@RU`DtjpiCslIn;%! z3&#+-FU8hjKYJMoy?xa3a43AYNP;x`z2GX^LEjEYpuzkG`gTeek@%rWOPj8dxy#2v zb@@(^oOy))t|MWx;sa7M+MK3b!Wwlczk@k-kGsBSe) z5=25!btu*UTt<6u>#_?+kHLzp5R5u~k}OD(ga!L9;{yJ^p!-Ic=$9I!+c<)GBpMC= z7;yS=v2b>WA@w*g73FYMav#HLBL!TrGYW$rCcutb3FvTrOv-Ymz^rTa)W6FQ0najP zgQn1<)0Ut{9Ti-dcbibtfB5a)N#29xPW7El1mpUTf>V6GVE%nOeIU7WZ2{%SVYz&Z^pP2-sG#gC_F8zqe5Oq;IB3; zXbek79ls)`xigt4zA>S5%@dOaxgvVndxil$e_?IF(=g+Zs#1#)We7P`)R4BkE6IJ3}} zfA;chfDSieyg`(^B2z*O5+(WW?ze=12C1BM@RlMyP1=boZ#8hE7lI#THGN2BN zRm-q(K@o_p>=9hOv5f{iYA2nUD{T57)S&L=eQ+tK2b+_+P}7#5{~w-*y9N$og0B=C z+?GU+=LeBKuXy<5kW1E<-^5JyiQtqghGat%t|$zF$|cF*eF^dD^K|I0_#oIM6OLY{ zku-1bXLPpDL34*n=-j)6_h9o5H(y7aePhiaw&xyA3YB25U-(UC?Qlbt5jWB+ZHL|a zJE-WV>tJzg8%$XI0MEXyW6Dn3!;kwJs8ktklc_ls`=wNH-tYhk$mID`{|(cGNQU=E z#G*s=G`RIw4V~*jh6>a&3TOz=i=FmmGVbboPGwL0 z=JS!xL~`aF+G-(z_sc3FLpullx~5}!fE4?0emP^9`Is!LsRh&hYav_nGM*kR155cp zI4Wg?*)yYo8JvxKCdZ+O_j6{hs})LD&cVU+vEJz05&lFE@cjLNboI3DZ?#}o{}BJ7vo z?B6U}efQmrY^uVY z(CW#gi!~aa&vgPW$wkPINASpL{NI>fjstIM){@iW)&*58t~Auc2c6tUaW10to%R zPrP1*L)h`j;Fo7hM9;U7)wQL}>9Pa=J@x{Ef$8QDXxXMwi2)KAz{NQo&U}rWzh_t6)N8`2~H-26FtvmxY*zdNxj8;0ZSdB?M^1P1};Kxa)e%S zQ-Nlq9Uz-!2EX(P^*8FHy{U6?e(3?aWm~bJWc@x|5Rt_cbj#8A6FQmat|9R9NeukB z#xq%6wJ2UF7bGUu&;ysgkOy1Gz{c2MTxeurbF9UUDN;7V?IX>CyvZkRUT^2!t800G zS;9K_o19Gl-Yup&Gc=(+pD-qG7BF$UcHlXea>1c=BXnJEftQn{1><}6k?A4I;9lIo zG_E}fUbm;fcuy%DoM?`h1AOR4hZ|_vx{wJ>lgF~2$#ht3kjdNYPP=ce!_Dp^WZ0>W zB>Oawu^$w$Ky9fYV{*M<{p=xPH!z0oE`LJ18;0=omNM+flj1aoq_I^Z8by@nQpZJw z#9Bp;yu6@-2XAa5=SHpY+dnC4)jxxzFBRAf2u~yP*R90v?g}I)x1z_#nQ-H7BJ8Id zpmT*Y_Wl9f$7JA)?manzgwNccV`W#!rz-T zGx#@gnp_5PcCs{ihBlrFe=e{Bo)L5E8U2!50$rJJ`M-S|Y&;|I{R$WI$y69d3?4AI zo-T&1n%O-Grqe*V!Og zCVEz2)2B{6XdOKvJ00e%xC=!6E3?>l3U)7x6C6H%n2hn+0u^-|$(5Oz^yWYz{q^n& znU}JP8GXUCa$mI8^f}5v-6c^Hr*MnB8COpp?HeSC^IahL_&BihngDt&NhC@j4M{t4 z=%)g6f!UizVzqvhEHxAtNF-gR>B)*XTHk?sW^@l)6Lj;P#q&Lm*7%|2l#PyARfA+L%iIZsqWN7s5oCs zlJ~noQ@~V$+m#`AXbpVQ9cBuwY*Dr8H;uKE1S7**eEaG&Ir7pG<}9uvN1}b0MNYEt zFnSFTjYuF9&wb=cD^C0MV?`>!r#z+_|IbH2a;(jw>KhUR=y*>DhY+S}-b^bUG| zXg6Mc+gURJGcdb+CH^Tt2^U4bL-&}=FeukXA4$){Sr)7CQQ3QvdSqG6SXDE|e#$a@ z_-z9IJbs6E>1N_RD|w>1!P%MM6AxVR=e7OqgQbk;3 zl}H~8r_!wn>NGI?FPWE62O)YEFz< z2|M{t-UHzTd~Mf+JVb|X{n3j{sXli~;|*f-5Hoh(4IU~mO4iEzkdkH>m~PZXER(#b ze(3?`qXi3Brkaw;+onTA+iX;sc?_=9B=F*>Y?!!Hg>HFm2z`m)1bafNNQn12cyVJl zEI%T`{TKHKpDoEo$5|1W4s|ryznLyoSPUjFA3}%rXCm@Z8u3g6ng06&J*(JArvAA@ z7UeiH-?Z=2%RB?iWB*L_?h_{6k<0O%y%|(Lc4TU_4nqLHyX$q1#n@}fn3~a!T}y=U zt@8zF>bp?wViRDVh{Z;ecIYCf6NlS1?rPHQ#tRZ4G!FBLR=_q};t(5MHl*fG~nYeF{FB#0?-8gcKn3TX4`u?Uc zx&L+*z0|Xx2CO!MO@$#O`HLk~^q4?$niu`vYkWy^8>Fbi)hk5gY(4dBT1d{ZIkaKK88p0KKwifI z`dT)uCKU{@glD8(kl0AwLT{2UQHQ8q=t;&pC7e_p-AVE!#mKa}D5j@K4A1Pgrjd+Uf}JcwXd|vvoAlFUQ7FY$N=7Hi3KyQ-oQvYvHpY-&HhGgr2hr zWO79i`ga_{bM-Uv>|t?0XoT!ZmhelokkquSWOTM}z`(L0GX0Pr>@Yn_$j{4oD~oV3 zq0(GYXchURJd<@hEylv@9D&;EIrQGq3Ys^rp2Ev2ygrakTuXp0RW)Xtvnz;qgBf#B zH-=gXZ=w2Ly;S^WHcg9f7v%q04*7RgX>V2;z4fjP`mKHpGgbj zpJ|$~ke;ER)pW3Xf_ii7@n{(fz%CSsJjr;M)oaaI{6Jy^P^xeg9E^8vVcy>8g zxM~;Q(Q<&)fFP!$wSz3tNMuriE>MkGPpF3GW|-?C4oa%G=X2ENPs4 zKm|Pd%1GY^PvR0aOd4-ma+8*=<8Ih0apS*dqwGUD-WS_MH!$I_cTzEt`N@KQry7(C zhw|^`Q>gfio4kk8M{vXJjv#N`4w^qGkDttBaNYIA^jOz=s=uq0M%>RMYp+ie+_0&n zm3b-EMjP4ePqrJmyICmrBAzkDTId@hh?M_<}9?>F38;=+#guweb$ z(=k#bglBE*k%#xjafcWUPI>%zZg-?Vb!e`jo{KnWTA_UiJ8lr1%+|0H|1)$R{#dnP94AqBC@P8~ zN`s<-b6+PVN&^j*NFk|6rBX@RD?~DqQg(=JaqjCF84Z<6Tf-=ck_K<{J^z8n=XvgN z-M`=Wn>~w2t_O%#xkf%qPa@K7Hcay$EszhfAq8J5Xnbrdtt!&R7WXK4RBAzd3+{tr zP%daWK8Lh*hoQOm5@aQRAwQOjpy}WuTvl+5CTk1`=MqXMKe$5T+vLzQai+k0QpPJ! zPjL?_UXuUrXOMk&J<#CgM<&vFZMDta4^VJ^rHO9>f~dVXzfb-?_6*De-HzR)bV(50 zUNu0%LQhe7k!_5|i*NLCBEz5k;Lj`AhVjo|9j==C^cCGae+5cD*QZyq1<$?VG^{p= zBvX&xA$uQrlIKPZ%(A}qkn!<6`GzK}Pvtb$N=uvd$dDxm15e?@j4ryFjAfUtmtocI zmcV5Tj>bi9p@$0(Q}3fMITOoW^v;Bj)T!?XQNQH|IXzRTWtHGiIb#W}pF?o>2xok5 zFZj#i1P;5^dDPG=#ehBG0_V^g?t+|D@rm^JOCKn~YY!J+O1dK|XU_ zFZplMK2Y8Z#FU-``GzL=H<$*kf{)#0f(P_@nS(*9IKOIx8*g{lijNWfLWal9gxx{6 ziNzgTi06#q{PY{}wK5H`$G^I}JCBB~`9fCSu?EWoS(ww%O;kxMc(|v~okZCC`}r!v z6Dw%pG;63Wh=B%8X~y>&%l~|tz^kuV#G8dzz>~EPnRlC4!T7ftP-=IEqU;Fv^R58! z9eNEuI$CtC=RBOH`kNjb8;3fnk+^?CAV%5?4C8GM+*bQgSob&(Y;RbunXwe^?Rp5G zK4w8yk{p_fEHZiWz?L3LSk4cPZ9?S>m+61`FX`oQIb74c2k$nf@KQgo@SU3?`MxxD zMlWR*c-BsWSKfYvOK3D{WJ*ZwtWsvy+kR>_x`7Ve(xZVifnFH>?0(oq-jUy_hp%Bw^F&+mm9?|Xr(x|C+_ zJ56%`Yo<--Q=nb64DOy>Nb4WBFkSg#(EM-$sK&j)n1_OE*8LG!kN09`iToq&Qx-GE zZyLzjihM}jGa7d6sO7VQ{rCf`KJnITsz9aXBG~_!2PxT2jDC|PbJR4E(KTgZa8Doc zZ~9K7Uj-1|eMwah`~9Idc05@ckWDX$$m5;N({SK=DSWO}W&azk1Jd~qNsdwp<5j*4 zR_>ciHf?#t#jcspsS5qy(;Zpdc3Mbx_l;+*3*w={3u1`P1m0;IJ|)z$Pd7duL;Di_Z5ha6+TqZb6N%>sG56? zzIt(kc~~rX*ky`nnw2pvsP%x;Qw2AdXE^U4y@-F}Bgr@W#>3x4U-%G{M0<)<*w9sq z?6zSE@HkjaYClZC8JX%BD}R$pR*pMuh|47@55&kV11?j8SQ5)SED0*cnto`@`Qm&WKh=@(h<4LvT z--S}j+}X$i!-YkPpW#cikK?z!nq`^i0IjP(2#B9DD zKsAr7hVy;JTuj#1|Jf;>ZqUP0aw7J!R zU)A=Midr;QCtVGK75a-HZHl}J`~4doHyBIW6ze(GWg0$mXd~bCa`C^sN&K9)4sK_o z3{Cg^M%`qV5T76~a`C7svADSjcdBT@&Z_U!*}a9>dn~|@V>iOh`x|I3Z^W$V)cQ3gN6ZSG7dTSJ>Hn-wZbt6o6jV37)R&a9X67*PMj@iE6 z@cBd^HPv2#i#q4yzN@l0GJP6t3Y?9x@4X<}`X}kN(ueO26Tm!qHTnD55|3Syppps> zpgi9nQ?6=bE^CRcYqrB|jRCUyffZf%{v};wwgZySAG{$`wanq{oHrZ^^dps?f=fy|kz=J6L9Aq&NiCPm zJQFx+N19LKBC#>hDJ}+M?%Hyy*GKSUDmUWSo(23eS2w61ZOvvlJ%EG{+L*8@2g}cm zfk)??$j1M&>5;Nza!$OHc8#2j-wwOO3-erZN?OPp3Efq5!#A9%)kO?_+m2J56Cr6u zATgSsi2qzuK>lAdu(ie6NO&-)krP5qhYW*i-Yx|Y@ zTo=VxO1Zf6sgT!G$l{#arLg1vNa$Cvr{{UmxYhSAK?==IM z0Bb2Ko#}`h^`4QB&ePZmmr?Za0!{o9JB-iIhY2jmG<@K;rCMt9JNnX79Mvxtg4564 zbh4EjjuqarXG%>+x6nS!Gm6H^^T*@f(n7ktW-4W`&Bp_>I`Cw1G8{aZ#aOwjVdR4g zjGdWj_1iE>Oca=oBJ;wDzL+sAU%HDvo*2x%9Zp~pR|J!Yw$GGX5`a<$$I)~$i#tyl zP$^*!?4Vc34fM5ur}R^p_i!_FbWsc}GZTeZD%I4(eF`~~Yyi?bwK(-pi(ztPAQ`8Y zNn3?Zz1%8Qw&kA*oVFJ{>ffirrUWDSX~E)C<@+DW91>v-t#(nw7(O9KdbN_UfvUjA0n61!Rj;Uny!j9p_AyKV=PoGh$9Q9 z`7m87VR+y`JEqRiqS#7!bZR2WI*|gK|D2>9JGC)={TBQUm*8`; zBplOGgrdi1NbI|-q@mgs=f`I-{ThADLHVDI*KRfFZvI3U8m2Q>b9CYFdtnLfzf6)e zN=fd)EYO>CmV~T(Osq;}ud{)x-E}=BxZQWd=oOlv-5w4{!W)U>P!6+V{2Uso z6~%R=&x0?-mwXyt2^Ej-;T~of*&TcF=*&zLvjwu?K1UyVi*}L0OCq50W+B-gzkp6m zYJu@;XUVMh!h5RC9Wvln$0&+zCpNZ}6pv6OM}2;iOXDSA&mvi{DV)k#R2JY+Rsv4> z(+b_Hl~7pX32zo|1lg+7q~wS$`q$m03!S#3asCUcIxqv2C&mk22RAN$?Gy4Xc?rQ& zS@h2lM+E6+=K&H}jFs9PVc9<2^ugIGTB8^MbT*oXk5{zQ*U9%up?FfiRQR zBfhy(cx=oDOp6e_lr}@$i%p?b!}9ylTI4%pH8Tu?PkWI?20moc?;JXI!zEZAzkumH zKL*uLH`9&p4B>387qLBRZ}MGF9L)~Kkgs)P;nwX|VmrkJT=hiQ?Lrq*xub!bSYJ;k z+;_zW!Kv*$q6lrB}IRufau{_hp+N$IM# zx#z%S1eQY4hFFZxilX~;lyTpT2%IN$xjsBwh|7-r1lRoEaCwOuX?U7SqegtKjx3vo zBQ6}EcWPhbx+&49doPChdd&j^1^|^Z-7(nkEbcw>nWU*^!zz1+>KRg}prfq{_-+;p zSBYR(K{2u2vyKMLJcxHrrqeIdLaxN4o4T8QqD%V)&un8L%I=j!_v#+}cXB*Nm?cBr z;{!BIX(o*pvVb+JD!jS&t!i1rUN~7FijUvM;_?r5=p$x|@0SI^Jtr|%^VT@_X`Tz{ zZx_1Dr#1NAS!KAt_Y#f6x#+Kb8)ke5dXw)#AEhvIj_UA3C6Rnh#256ub&R&&Sb#NC z|B_aFAN=hfMzw7&fVBD-9DhKWzuO=wT`mNlT7kkgVYMc(;mwF}9XYUl2!MxCoy4 zEmMfj>gC{Qs*3vNI@tEfmYj-sKy_76ztV0x`C>+ExhE~!k1rxLFqbH@eTDwW22 zxSd0hJqBDLDr23isqnB?3J(5h5M#Pdxc58Z)qG$0rs4?;4el^8O)Ai*Z3TY>m!HYT zIvf=wMtyvjL)7JLc(Ta=+MPCl=T0$_7#e_D&B5&Cpjlw}Jpr$>zI5Ui;m!BVs+-ax8$6ICLmE%vS$!{TCd_WaIZfq~G(ES3l$&TBGg)1Ip^Yq}QvI8jSn0=$Wy!DL3- zZ8x*=UOV&6HHCQVT%%}8D zI&pv!uirQ{B(S_hCD=`N!VbATg_@2%MMp0DhZE;Ck*#v-@ayG2*aqfg{gD?aqaR9^ z9Uq5cW20&2GnQP$6R5bEgM%Y%Ft}(ATPNBKd5r`=ta^lh?*&5fGBJF1v>$JR#KB2eAFWyaP^TCd3@Kydw^6XJXb)V? zc7PqMCyE`tM|0oWQ5C<*41JbKUK(r#@8MK@$_8OhLIC&KX1UM>3&8JgUF1n@Hnfj> zf~BiF(fE(1kn@)3cUc<=o`?Iy#V-r>3WteI%n~w6u?A{-RN0({Jw*GD042O%0kQkK zpiOTa)O>3s;`(b*^Xqwfqehr9*(yNZ2N^zh)d%u&iWSV;>P1IB-Yu|>lCir>7IzI~ zg2}T*crSJVPUoVICkI6EELlMbeXr38LAn^ zH)&$n^=UIU-xOTqkEWxaMg!(ViSzvPJY4PEiFy6=*{5@TSRE}fcKGmW!CPuU^6Db! z1L2L}u=WqyJ9{tl$}$W+TlGoCNG+o2=7O%dY1nW72h+noP%mDFR}U(sDH|hDK~3Oa zc-*O zoXr+a(`92dl3*&YObW_^aQEGjm@;A}=DF^{9=jk6l#4@&z$iNFvnP-pPVjK!nd*2o zKX~{#S6QI(y%Yb)9fv`w~f* z!2_67x&)nGAS3A0GJ($oowp`3 z$%k=`1EC(z-SNyVf%~!S9wyJ0VU9|O^I>HfC}Amxb}CAdpAvxozC_@kOo7wB>o;Mn z+D!EMAS`W}L8RZfaNjZ`aqD(*TyHQ6VhkUUWt&E#f6}0kfjx`55(4K#+<-j0szh~d zLy5kBJC$Phkh5KZOkV1EG**$}{bT3jA{Tw!_r(x8*9F02S9212LC9veT)?{_J8|i$ zDfps9;AMHG(1E=hNb=ZF5UZ&rVwQ`@!iRrMR656DKy*EkO|BF%VA<>hS48+q1+-;%pFl5jV=c+qU%ay?D1WQo5Ckx)9iC3?B+<& zOs8mgHXl9aMC0njBx*0ZkQ5Kzg8`|Ncu#u*TNpK+_4SzpMZz9dn-K+Hj#P2S)Spw8 z+{=uGMjJ}NM>2NtRjM-G0NyBCays*t(}!x3C=$GoJ$?2RoA7BWoAphC-50%-^{n@V zHKR1>(WnGYG_#i;D9?uPo%118VGfS`x1FC~c@k5c-T1{TC-Blwu5*Cb6e7&!pbQOIkPK){jkO{^)qnJ*W&nhTJhR%ox)b-Gxz~)!6N_p9vE%1!WI@ zp>sWF@X3$5u~Ih;BRq42?~e!{q?~{w?xf+Sh7H0wd@I!R)o{dk66(F}z)ZIpH2mKP z?0)kaob7|4Xlih^))-MC+jbLsR}wz2p&sjk7m>A_%Ry2ip8J~>34dP5j7?lZW(SrFj&m`7W_JtS>9rT+$rU`Y`xibkFXq+_iL#R&blHDXjzXW+Xr_q+5n|$*^JVrvvdXWKo!g(qC6b7f7wIc zUieA7o(YZ@;X7n7!y1-d65igIjw1mZLQv0CAGHG{nHR~i*s?W7$oGtd=`U8o??z{s zaiE-J%4R`zS~%57ZieXM(=_)ci^;=N$*b>fpfX-y)^@1DvIaj;IN3x0?UO+%peC!| zPoqIgW+JJpr~M&b?Dtusl8J!>= z3%=0|F+Xw)1QV7?_=O)Z_P^sedXqN!_)ioj`oz&scdBr4pd?Ocd_*;$JjEYLeYF4a z0{+m;FzmX&6AoFo(1Ouhn9PGYV9}Nc(y(O0JWqcH>yCI?*1JQl68~t6FKjcouPMET+BRzM522FMtd81lFFxJ(FEE z=499U7$P#x0ozkm@Nvmh-l&Sf$O)df@AhBRyCcI#ZnT2V*+QSD^CGN&F@>u+R0@?s zZ)GN&g-YLtkT!oYw2yMefR$gUOT;o%9FCsW>G?f}J=%5$3 zP{*s=VLbKbmq?<3+Dw-B8s`a)UoVUSup z4=r@&VDI7cuqWg?nXU4N9@DhrtAdvB+bqI)qtD|J2g-3!I9JZM+|TH5SH_}YCwk;V z8uY~~LW!XjG}L<-wGDV-QArYgip!u}Scz#X^=#26}p< zuzunhESprzNp2F}$)D)5{_DSzDCuUR{&^m}l1v7RH4|Z~kev-)c9R);d81nX>_zfb zD*~*4B|{>&ns>F?!v~_U7e6Qk&ptk2FN%)q zAz_cAxOc~d_m}Q5=v1jiez&J`QelxaBJTlLZnO;Mr+1TmZKE*U@FKoZxdQj&|1x46VsfH>zez}UCoE@cbd)HYV70P1SXNSgdb75@|^06 z`B8Se1JoC%W43cMgvcF*HRI2K@tY(%TdJ9Slazv6hUch?gA|qjejj@~Wrh7FdC@W2PkiauI!#r2*B_*<8VUX83PQhX+! zd^Jd+9?RC!lS=QHTN8tDy~qOKrs$LOl`>>mNDmfn6S@(F(d3B02;MSd1zdb>i~Xfq z_@@3MFPdn;&)9VdJVS@)L#!2`Xx6}Svm$Q7b=0}r}yNI&O0)= zLX&D~zoK2mvqAevD96js2JBo8N)E%Yz&{j21P-jiISIC%DdC!9SC9hPGRE}7TV`nW z7U6f>1o^`*#nWc;3dxc~7S(Rz0RC!f*-3+esDyYd-bBV!Fx%b!y9 zAxHdkE*`($lHnZ<)6rbuW|@eqL;L<5BDdF@w#{0Eay}QZZhjG6?%~d4w3_3GS81Sd zs~P@WJPeO6HDJYpAE;m9g$cKB)A31#7#(PU_Z%l-$F(-*jJF|1c$_8!ku8weTLh)* z7rAFAPoe9{ZP=cx!b|@#;D1aL7^{!I(00a#=`$!I|7G~oO}#pJ%rg|9_?v>ENCnA$ z5JrQowaCgd`xwpkr`*#u?M%*SGn6b*#v}EsnSTLWu%k`~#omqPFDbmk$-V9)j19``iT*>M%rRt!<{EKlfvVMjEGmxQF|}uEmbBIAZg? zkEVFeU~e{`VJ}Y$W7Fq2!sq`KVesxoYSdIouW2ykh+GjFIBd=DzcPn+Tp+@4UB8T2 z#EV0s);JKK8$<4-ZZawRzL5N43TSNdEUFwOhtJH6$>5JL5|U{~S8Xpui5m%mKO+LC zp35b#Buv54#kWmWiNmq6+f!N~p@2TH1k4RB!YozMg%CV>*RA&9_%f zO_mk}NQ{9)ZK9aI;2L=)a*Layo>-mR{(>Cyo5z^0PR7@fJi6IQ@f}O#c~vUPr|G}L zIYTz|m_-rUExjH%MN4+yQHI?TrpoSVaO1OnTJn~!$MY{Qea4`kP)xoh%@0r6&Te-v zWw(wEV9&*^;}`k8!nWY&^x)PW;udC#b_yyet@~YIif%@^3AW_tqXRgT+(J*=2E6YT zN5Y*pL-272lUd$pj5WVhaf5CdBY1mC|M8so^HpVtYKB)NjUmsGlUO%g$bR zUB$wvmI)}DFc;tbJV!ddWe~S<`Q-HLuT;Xkk@O!wN8UyUk~>{j*pj4Nc0I0Q#}>Zi zG-}_`{2A+LrA-w%m%p8Ox~~Jxu1u1;&yv`OpM@7zNp!N%r36_`9Iui4|-`u-D0WfNT7umeRkjVF}WBPAh0R50insucK_ii>NE0?CxHZ*~f z_!i=SO$D;DM}z(rO?GMNNa7ISN=x>CClfBbfGuG|^n1H-F6h1v1|8-?H`EPmo%`XO z-e|NA3gHeoZ3h*DIk4&KbJCj-1Pew7fu;QpvP0-qyXE>zp;h)cWAg0Z;{FSEm({7bTfBT^4>!{P$ot>p$*U%HOs z0}9lkF&<=}nBe)1HLx-{j>wzw^sU5Api`t-?VFMU2kZjKs|x(yXR9El-H&-G?#>4P zGv(<`Ti&N9jBbrsj-STq61<-XXCs%|ST znAE||E>VG+2qkIR=H;4B(EB$x(xXxg|JEW4ZT3Xa zF(qQW@kvp>w7QwTReAx&W{QfM$dk7z0 z!dM#^Svx$Ptk2&KTn3W{qA+@+1sI(UrGIWez@u@0QKM6dcjzC*-m{> z^2KD-dl-rxTkj(GW)VNCB_HRDb}Qz#`kmlNU_K{(%u@2SM(Fnd7KlxiSog9KKb<3 z`7mfY9Rzj)`(X;x4x_FjDToso_Mg>=pV>5lh2qPn>vd0p?J*KiT)B zj{KFf0KT-Cn^f+@eRSLdDvsCT>)gc{D7qfXwHk?yzbtvQ!AWp*oq%~=arik|8C~+< zlj7&;4iDqu+j9VQg&AmnqZvOO6G!HyF8^@bOdk8h;8pQyKKt1$ z{*Y2S+QrYrp7Je_`z;n0^;|}AUWAVuD8?4WK)hnN45DlfflZnUn-Ovp^24q0)!RlK z7JkPn{G)G>1ukuG;v&XgrI}auVd~X7sKOff*^s_W`f8$cdKt|6)}4_@TMrLc!6C;B4V3P%N=E zi5AY`^DlbB3$ED7^H# z6+$kq$AR8*JhtXO%oEPZt~VQyjSHrR(iezszdd)COyw{3z7*a!m*Q6aW8{%l6_@f+ z1`LJ=;KF)mrg5G+#^*hyNEX4`)^Wu6pU@N5m*DFj2yd#h!@x(aked~m$b>)pU}CLr z4?D|K=$aRDczf@ECf^+O@%bAY=uzJS*}ZvWq+1R6q)r9}x)~krp5|)&*ghfk2gQT}E|1P`#h|jSk=kC~ODx$|lg6q5 z@@ArNt~u3=CS4Qp!If+hQNTjr!u`;0poza^-U~d0$xxzp0xat*@W=5PczXW`zCY^$ zLQWLvQo2tH$6qI#114Ze)|^zaEjxqb-s z+NE)LAR2v~6zTLv1zv8)Px5hm4876%ox*`YW>4%-s=e+2{pAn4{ zzIugD%8g?JL=J+%%T{>&-~pC}y))@*BDl3y2UqV^V3!1%LQ%nId@HgQ-Y6dci4BI> zb1(}-qU8A#*FAAS<0?338L@6R{&25$-z2M2cHpTWmx;urLb$COiH9S#NR1JPM>Afd zSl44Lmt07$4{srM84Pt9%7AwdErFVyghTz|Fe+Rc6hCxgWcD@8QIy~xraHnk%WT-N zW*ga6A>5fFlF7wjLwvWz1%5qIf;%@J;f2M5pNp{g>vtsXj_xAG{+5Kl6Gyub6l011 zRCZVU04JIHfz$n6PeP9lpo#)Z@2w66y>@{^5!pxoE4f8K#227%wy^g;-HT6l3Y_=m zPgvG{0zS$0_+JW;EI;&W53-N@H)5z4_gl5dY?RMHS#U7TJscFp6pv*?F2 zvO-?qJv?&xN={E137enphm-&2;+Bp1q;t+IT>Nr1|NgHVZdfV}=WElUi!&nE^1E>b z69xr)GPtwfX2YnO7;>+D4p@jj$I;>hW#CU%QD61utA-m5Q{?il##W_$5RpFPrzF_=B3|W^$9JB4ADvV!a zfrkw8a8!6N_h9@9n6}TIdPTfqc0X7NGG>3OpI;9r=W+$dXiyzG_%6W6qk52>{hR)B z5uA$kDb?$TuVPAxJx#2thq60U5Y)qQa`q5u`E?u}kGB!wjh-YIJ;RrmRPo7JSDkB(NjJ#s8RTUimJ-Lha$ZV;y31c>~0l-l@b)1$fP zgbq^~Y@Ak%lJd#q!=@1GRncB05s=3;KE6ov?y``l?FvR&Ikf8TUFsY6mbw4J4aW7$ z!O}TiV065MT5sNm&QDh1pS9`W_FaNrN-QMpo~7jP!ym|hjmF*Y@{TfJHZlxblrE6{J1eO|WEH*E`BHGsSCHZrOR37IPRa@EN%`4)@_M~F zowiu;{rFHcFStu5cmpKQmf*$3wc%jq40cep6#czBi0IclaKbAdqigc9xOWh?%ueSj zJulOyxklv4@hhlkC%}RxmBV8@!8f7&6xX*%v)S!2cy!NR98>2DD;zK5-N9dsPv|Mg zy}b~{Lgk^!+Y$=r8SuyI>w%t^XAiunq1HnGR`TH#y!ZPi@$S6CwdI~;{@xr(f4*=x z;T=@K;$zR!CYl>-c<}LDO-zvED={9+y=SSZCv;qkyw0KXHb+8?);OMvJ+x@iZ&7wq(r5HC`%LbxEr7~5e%Q0{F_Cyz z%v?C7$xq9RM(JHj?Cz@zkdwMkaHjzoHg2LSYa7rbBnG0IWH2pWizJ(`!vzDi^wa%4 zB(*;r8!zx=gL@0a_gPVA6A`lE;|1IuG!xWs1!2oqY1AH73?ZZ2P|;1BsvOIOH7C3< zqg|cbYa9kUa$(y2{)S6i*j2GyWyxfuI9q`cZj$NIsO_y1p61WVgXJ z{l`eu&tmqXQV6gZAU~20lcD$(*p+Y*f7aTf;e}ynPVc05_Xy0%lT}3b@NIP4M-s@^gC`=XD^K{iNfiD zX>hu*8fq_mH|f2>kuzQ&uKVnmL1KRFg7qp@)NJ26+?rcP-ipkmGh82%W1TEXIAG9vV1g?*L{gd%zlJM@;A^)p%t;OjcE!R zrgucG;h$PRKIZ#j^&K&c4qA;QY9gFc2%!6|OmJfD4H`S&4+|%YGd4qli=%A^jZm?L zOT&p+DeB6L?wku|;qp)(7fMZD*zzn7AinsC(8*Z@gyq7p~@DY73H;IUIjdaul3o<3bA4E>7pjV*{j5o-jQpt_jCf9+? z$ue?ql`Y&c_T(1|Ijw<)1!%v>7)8$Zb4BUB)S&nbkrY^s{+>F#^}$eP+^#h=FeD3_ zWP`D%a1^{Q=p(b|y)zbdX(NZ8`op3zWBAj;?6t-B5SWE#!V#7ujQN_Vx}f|e6Lw)z1fiv6)MYY8!NP(3fhlj3U?eT;dw)uk2R5kc-a3fsw zpUYZ{_!0Wm1Y5G?X=T439*}-RCTyL^{pTp$XS8FumepyH^E?qZmv~|Ot+T;o|{i%Zn*F5rFfzd(#nhZ?Q-98;*XF zK%ER5IHOmBKO*A-^k0<+rw55(HsDLD4lRQ5gJX&B?Zd+PLx=Zm${}_4!stlzK=geR zhGPC++{*RsIPJq}YFBa@Rw!!1(wC1J{VUTz=^^MWPk=J;rxH$}7ZC@!DUqarmU4d<> zL8P+r25%W`%^I(}1hI;P^jE$j(}tE9H^ z7hS6`C2u4!gTchm@fsH}*@Wl`=f&46@6v;^2=PnDG5dZB-T{G2nXopM9E;Hho1ZVq zsc=(%&kkE0`es0t#)?BqYg3h4yBtbAH6Z5~sqpge<#5?{;ddI7%iOdQW(j;B-U=82 zJDNklU4IFjS@!_4s@{>W-Fx^g-s&(zbqz)@9nS`gQ$fc~h49@v7TeleDL!}#cl84C zsbHbxmn*Dl8zIn8EM4;Epc-WYT%{za>gr&Ddq5wQJwG>)9&$i#?va}MdJ zxZCBgG2yQQ-gb_~-}}aZ>zOI+{i5H{ZIv$MmK1TZR}x-VYNu)g8u;NPgLjVSpu&Ys z_|)(cny&AGDTy~}nTHgkT3HTqqn<;Kcmj~cA$WaDDrBZbB(`Li~0;JZ$R^-+>w zSAV|=(XaH_oqzqv6}d~WKjAd!s^^30t#?#Ek1`kS9Fbj{hjT|I<66TSO!3!(C6}7X zBZqNt^Lh)GT5Z5K!#=?o<-pbPn`!XnJJ&y&xRcLnVK}Kk2GKU0-eI1btWb==ouiLI zCkow(h>dXg>;jxS)r@&QD9?z6{G#XM8|aj2bFlmH7&cjDD*MrOJZqk9jF62?`4`mtW7ZsifuY&4%tF8_~+->RlDhGF#g#tKMnyF+ga+})MqYlvxI zHuknSb8nmu@(FLe_|to4%GHmgPG^!`l`&F$%_THJZ%y;QA za7XC5ufT6*$7s$_IK1AF2~EONj#zn-UKTk`HhPSv7GqBHj!GW9-WhxT=e&XDN!17xfA7W5uIi)xOZ_&q@ck0~6(or?OxtTBg9^|Gfa=C$PE94-E3ojpH#jTryg z!5w=J*O=UmpA5#TN2%;A8G3czGw#g1A4Dr_3$hVroTZK@kw2b8*Zs*S&Sn5x)OD!t z(p)0ld4rCgZj35{4fM#tRR}*KI9=Jl)Xd2a%>GUXxiTA|9{-GvKP99>bQAORRyzuH z6%x`AjNGGt=$W#Zhq-+yE4>t6s!Ro^1ru?_-t%zkZ!9RJHNz223mEhcpg9$xCU2aT zAmEJ`l^eeg9RD!<&C;>_tI=zC*)ss%p9AUs3nld5-fAoit;HWzW+>-)79=_~5Zg+(ba-uJG#*kpO94de05mo<}#l;Ve#2>aX@Sx@gnV!@Ozb=X3PsNWg&M=d9JbjC| z1Lf({@F3g~8^f)7rpY%ebqwN64*N<+kJhcM2idhI;3?(;4^Am#yRrgWE4?=A zXbXqludbkxu0>TQ&quY3dc3If0c@YN5u=woqT2&`Jbbed7XCg6le;#6+m*WNE|SP( zo>~IOvtmK#g*k}s9L;Z$j;Bw>SE1TM4Rmijjziv;aoB1ss@j6!fm+Cq=}gDT?h35# z_l@x3UY0Q1l4lz%5@6Fqe_ZM!_;rRP*<0*h_%Y5HeZn+x@X;8i(jti-Of91s6*s6# zz&&QXXacG z@@;gCQv&L&a)U?O3n6@}H6+!9!$HbHO`IHm=$J5;;kzshwP6w^ZEb!KtalAt8 z3V!YA3XqC@3TZZj&^Zvx#YC+a&URNwTd)RRYrII;?lpm7NndnHR)EX0=Ja(@8k(+t zg=+67QFq(9@OPOJuCKJjB?fV*J4@Ja>}7aO4^7_x))?MLIH%(LyU0S-c2vG$6i(43@YZ}-w@gd`rrv{mwOYpV58CHl-hs+D+fZEn@ z{@D!_6|$`I`!#8=q92njl22vtPsjCt8knzAm9RNx26;Vn5}y?FsE|Juu3Y~JjYqVg zAz-J#0QwJNY8}zzZVP_VqAT1%RQqv| zUDZXVJluoJ3;K!CJXe^$uN)lreTG7w12fUH^q;^=zR>a<#2Pf&tGE7w)yEcUXCVry z8fw6q3u{Hm3VO}c6skN&^A<0=aOv?o_)Y5xDlQmh(yAIuPCDkmrLb`b3sys%{dCZ+ zj0I=^y)ZQ-6LsE&V7+Mq-rH$}TZcVJcWV$yIaLbXyclVn9*w(9E4Z|A)l^GC3BQc@ z=I6cpz^_dS(l}Nr|7)nv3lP)Ze?#0mA;6IWR;BP+}ERtQW8zk zE~%8FsbQ}oBeRUiE*c2uzK)cTBubP~TH2+et>5|m?d2bj=bZb#uFvQFzGa<{QkT9{ zWn&;Ji7(iRe9mS=l@EUZ>_>D;596M9C2*k3m_)tsrYfG=c#%|~^5IiB8y?cZf1x#B zb{~aYOEHiY8zn4cjUi~j6`qQF)9Qbx$-4KH#*ZYje=@YmDWB8GDDH;E@~_#2vlO|T z)@|rKwHj^y-iO8uuCT(A&y3V8hOTCDAkP2c6b(7y#g1cOv#A{K54MxmaTlp@OfpIh zS;L~2WyEY@1a7I>3&xwTQ@?E|(DBMka#wdML~lQenc-eU%VRr?ToGd3Qe{Y;a-D`< zmBb4jtsq)>iE(kfK-?!T1iE!I{7C2U)Z2QhHOmByGmhcc<6B`Zts!~)yUB#EXRwIx zTx@L^A>mL)?8l3tntLPK*J+SL?5#Oce(o6Jk~#ibC_o6{1k~@>-XYSAAG<`x+h4f>koQDMxHt!`9*54 z1cTG2U~~uu*d8MdvFBJ?9hd-7Pu7yajxcOX+yD~EX9P)hZMc@_WfIXSXo`7HAE}y= z3BvE>eZD82^q53Oh45$Vm;)#lc!ZAnVNTmloZfrul{w zkD*R7ZSXdFi;j?9BTYD|l!B{mlfmtz1T6mW7@ut9S!wb6QDJivwXsXCo_hQTeKk&* zyRGF1zi(flWg!bu!YCJx%zHyVhWgRCpg6cx??~S->cF+fsd$3l6)Q}o?i!@HJKsKV8HT&i)7 z{Ni1{kIYk0k?`G8NX8jIhUxB|i@`Z;2W!UPrC-1lIM~5+5H-TEQ=b2P#`y7Fu!}g9 zF%3H64v_5Gnw+BL8Mxz=PlGpAVZ*AAWTSKl@lg6ecOgs0+N)!`zAO5NxPex8Abx`> zcBtbn7L#arXc$1d-FMJ|opq#6Y%ROn^cdD}ybEU(bBO-LAR6*$H1onI zlF!)f#oF{LLDK^(H0Ya-t5ipb=}vn{3wMFKXGYLF(}k#HT*TpFXOetc27@zh;nK@W zFkK^u2C!%8+^FfW>e?>wveU%Q6+D}7RVmp#V;bCAZOu;2iKhPhcVNW(TSOzbij=yI z2jBTAHCHVv!DHTT{Iq2WJj^qIGv(E&d{h&J+ZRB|lSJ&A8-{Z>^i%W1%Yv89e$2_* zHgfVf!O?OLU{jhZmTU~iEVDQg{6>NOC` z+<0dbYwT~uSaR>EXxA=3iInO790`_k-%1?SCh8Hx5{GM?Z>lMcf=*qP-jatge%} z>PxXNcMjGr90lF{xI+4@0R+D}Ojej0aCLimpZX3{?m(n0_d1;R@zabfXhwi(Wo9Na2R!gIHYFN$4mJ>gJTG0+h@>Q zme;`FdkX0qevb`iGTff^)49-7571zB30>X%mNA|%hS&^l#AWJDRKU)}+~i;E?YI)s znkNp^Z&`x&#R{@-rYr6}agwgpvV|e>MBHlGPBYItkt5URf~m((k`XToh4u$Y!^P>O zc2qfe7gEK%|9TN;={DjMDG5%}YbMy}nc$7DGN|u!23#NH!NB%xGVeD> z@9v2>N@5R_wB`kw^2i$w8;*gp`uTW7LkFjYq~hGx_vFUPgXDOY4sI1TLBvKkVVoC2 z$B9p{>17`{^I7^W*94HZ)`=+SNQ1PE2{$Kq1^1!j0VZ4h$FpdyiB@s4;7_HtP`Q1! za948|jPDmh;3;JgX-uKV{w9(=_L|i0Q8@8({14|3OVG{fI;??vJzFp0OBj<=ilqh^S;0U{JY^CZGki;Y!f&Pci zY8UfSnEiM(ER~2L`)q=#myIJp>wDTJ;w>O&bLlaEC9rU}Cb~0>xxvXd@GMd%(_bP?wXb<& zW(Kd`)G>klook_>6wH^|!_LSNa?jeC1X)kR%k87N$h8;fUazs37^{tTmMVA%)!{EuMQl-EQid?(JDor_m0Tw$y0 zdN#%A0tRK2!TCdypz^Df{VVm2Xo{R97yChRd7W?>S7=F) zjsuNw&$Eb}NV^JE`XkIzY{s2r4YcocBPK?fjO!Y0_*T$KQu5_+lzTdO+z_X(n`LlQ zNG-W+*+rgTbEU0L5{&LcMc5r3j|+9SfnJ>fl|7t;GhdycM-)Gjv6DRMjMYy_58re3 zuv13c9>4$8#;f_^uw%m$zT@|nkS`9Xp&US+COaZKEsRbs|4j8x7T~sN zw_u7y6lg~8L^8h~Ea^v@()}4PiparM26;EKaPe;zQp@cCOaJQo#;P&N6MeAC+D-j zV#}3Glzx~)Ya4^{ax?EeXI(-`FD?m^89rz$4ESpD~p%C6}S_@uGDVs8n`rOF$Rrs#kilT zs9iOaGf)_B2{@+D zk;Hr*N9$ks;P8#}WamU}+;epw4*KXoaP94yeL8KF(VGsZkUvj!Yt}rUa|{-zWRUIm zcB0z%)99<6MA{9LF>u;i{7~_eCX49vp3&2wSuYD3UH56i#dzqso&*lPUU(eTA)-3N zrfaVW$#u1Y%V*T!tZpDCSjOV5uo+BVP(5BGO9vuobA+zSCU>Uocx-1ES#Ja~Y z+Hw?qUAz|U5{}}9$V{9zk0sM>ACb||ezD1o42sU(kLFRsRQsp|mtk}rkBE-rxkz(S zr#TK6iC&@amq|hX_@lI-sFL1J>R=8<_LETI0$P3c8A%9ULB{`N(KGE27R*uMj0zP| z{YoGl+UY5fy|xffkJ(1T?oYEl?{NS@PZ&uM+7Pgr5*$X#&y=T^nl+nh${Qgq3gxsFD4lT}HCeGg!aO!ef zR9TXNdrT_G_v{idzt9Tox>F?f%MJXp<`!k+o>5s9hTCF)5qF9`KK28k(+eK{S^GR9t zJqVl=j8&so;r~0Or>2@fFn|B`N}Wbe7JMLg1(IChiHDdF*9MaFWQC6x#6XhcMI15S zin}5uxt=W=5YliIykxqsHI83HAh*Q?3?!z(mvM_Avpf-c6HAEoU1@Hp ze+6?QGnXEE?F@^itRz=wd?1!NXP|Jm2>0Z|Ww`X$iqE5dtI=Mpgo8_6u%l@dyZYA- zqI=K-W|r;%vEoAT(+z?JqM5KV>kPHs5lE&D?V$}_JS%?ZRd{dlg8B2~A4%R)BAW>R1Y<^R$$jOe}Zxx*7j~7HkuabYpj9ylQKzq z?mIHiX9LJ9d$NmHU!wPV^!Xh6f8_5`c??U8q5-b_%y)(c^^-^hUvmO|FEwb^NCNS8 zu4Kx7o@4u-p%SVvepA5`*j!N?kQn|AqGiyCoB=$F?x|{4FjO z*TcJMgQ#mbf%h{e;+Wq~IO}c#@l`Go9GA2v4YMtXZCw@3u?)o`T@^6PY-Ts_5mJ#J z4qLC;uve!|Cf?>I@Oaa2D$(|j*&8;C{2p~1ZfyPmvu~WncN?XcanpV=9aEa<)y0;4 zNB0vkOVehPLo>l7KnPQoT@=Jtr{Y&VGZeL0LSpRClLW^ko(uGYR>(Pmtf4YV^IyxF zKX0Q|-U%R+6G+;03{ciH30i%Ze87aPK=7Uu2da-k;B?Pja?)u6?4Okk8FM=s`vpTJKHMM7jm}Yf z6F_t6T~zDc7*0G;kuxpi@Nc;`?ldFd;TZ~t#{Wk}^P^$wumY$XCzAYLd+IRD0)5(5 zpk?Pe$UZWS*nEp7M#WP>d;2z6WOxoQt1&n)Qw@)<8DZzCr_*nTt_W^@$tNF_E6A5Y zDKM1Ir*^Bv*rvGgpwYda%U3Kbe>`(j_d!-ci0Q?ez^#JkP>UV&FvSd zdLPJqzS7Iaj-E_a*2Gf5o10A31ZnW|Ne1P*eYETKr<#PKBIccP7|#dyhKAAw5c>2w z_}6Q|)Q7hPjckA&x@3?^>nB%?1R&F7in3xlc>GZnJ({Zt zT~kBBbL9h4{JE1H++hg`Ocp#l6U_Rrmc;!gtQ!LxqPsq76Oe5FWWwz4#P zerG9E)w+OAmI27VSOS|zRl?w2ak#I0kg$J3pse5$H2!jhXa6a{)=wN;aX*_3ZkUbv zmMh6=b_ZSlL>r8c5t?mrk)BPFC13uv!E=2P8n`cj2HY4=F7zwmB6oSvHi&^^EsbQk zMJ8!!`A4Kv_`VO%I82%l$i)Xwz^^Jfr1iZX_@w&as8iic@gE_~-}oJGN6T|_zUJcI zjgHKIMHlApq+rl<)&Q+kC1{zq9oBh{gVxvSFt$#PbeYLw_r^i;uh4)k@zS7A+Gf%c z^J?-SvxhY2Ujo6Zr|>+akJQebjY>ZsvoU;zVq@MJR&SFDwJN;gI+}fvqenTGUC!cLk7v5exF4r!D>3 zBO_cnL6y{72SeG8)oAV!ipsGyypPZX-mZMav#^KA@(yLNQCUv^>=hFj+W-XLjfPeG z7Q)I@b8jVqs*|eaPC#yJve!>CrYPxb;Pbvv#xQ zTpH{-&CB!Xk|mL#vFR!FY&7MX`=@hb#wc-?sf}bl%g>qeG$5uXi{{#P>%!5!*5@RyVnB{Iif_rN9|!0W?iAfiMc58RFC^HX9VX3PvpGiZ^D}YwnFw{ z4&EA0LjOI>P|y*KvAaj{HLNjQcIi^C90EAeZG5NDdjdC9Yf7Z<9tEbckbNt;A7iF@ zqFYls1`T~8gD-qwtn5l+a&8TmacUDcx7v++?l>L`>XpzeD-9jDe#2OyJ?HMK$lcGO zM78!K!3Q}Ioo30Yeiz}2r6+UZQ$0Y3wF5aUfc?SaxiP6cV{uG6c71BZ$sr2dxLulD zLfa7KWS@bTXk2-1s%wUPyhA@M|c;x%|GFADwz-moaR9SZv|@E=DD z)-BTJbY5C;qFuBmS6B}I$Hzl@Qxsjz?`O>|ByjJ}86cn8OT!OZgOWo%dp`OoUVW*= z`g967|EV9aP1lN(nw`fx5JqFf`PJa!>kgVvq=9_ZfZN*XL|HnF4Lwp!r@;qUNF$T3>Z{j)gFwN2W>b-s+{md;n= zQccElbCXtpe!)*BQB|yFS8W$j*ZE4bZUv%)^*rL{5C*h)k@W(DrFXurBH*yVEI*TID;d0`1;L&*f;-+}XRY*Yc9o4JDR&{%{XMt6yy z{6B`NKZS{g-gww*4*ocI5A~LQ#w3|BaLY>q2FIPEnJNXu-(UzDWVgYZRX{7AhSB09 zzPRi1SfRxGP5}K*_{S9yq1-q$eS4PWYER;vh0@&i`mMO{;Wx~@WQd)s{V_GOi(HYl zgfFY4>93J2^jSKN+jHFj1$Eq1OUKcPjN{4p7 zw_8_L2+XW1uopLjX*33E-~1!*J8j5<*1a%L5(163;=(%>Da>bgQBaW`h2s)bVaVZ9 zm0kA@w*E^L#=B{Av)w~5&@G!t-+M|zmE~BM)jw$AfTO@gd@2~LXHtiCc_>+xhLyB~ z-IkyX)(O$fMA34rk{iWoUwV%P4U^DecoOr}PZ9oZy-y?YJg!}|0rwm%faCGm&_8}9 zTwIxhnR$bB*}fo*2$ceI`#ssz6$)OzvM}YMJFVL-$7zS!!9CZh_)gpB(cNs(BXP3E zHv;D^wxE7$-`F;JNi;FbLfL~xu(DYJ^EguAmDDJ^;?ihI77P>1DPQD$2`gq%6S zY`wOc-Hf}jV1)03>i6PV(=_sYQzIE&ZO80h&C%+2yK(GeZ_;~SitCeQadL%_dvQAh z`c3$pM@=j2*b2ptr^vWVq2$Q&8eAc1h^vmRz_R`QC_7J!|DUO% zh7lLw$mSHNt}=!7^>Hxr-y+CR3@0Iv8?ksoDLwRy?@JUGLoUya@{;+$$R16GNq@J& zaPDl#=ynH{bZvIJgdP=X7t+}Qm#9&BGMy%21!o!>Y45hJ%yv0N?%#P)E^WCyXBHC= zf_s0794ia%aFw_0D-b=O~OC@y|!ea&uzx@z1X#JHL6Hp0GdQzzVxC%ut zyhH9!H?jY>2`8P8M6-fDU~BA&$}^wf=ro=Mwt~-s8%aXkf;;5LhIH&bW`b7h7C^XG zzd%FrHeFi!gXvkFO&_{QaMkz4m~TEDwfvP#mK@22eaq+5?zz)2DSbLbt&)ET3+Y_cX&vQn?qA+vSk3(Eil0k<5KX?15+p(D1&uh zB;cRg3TPiIhKX}ukgME!9C%&<`sxxaWCKjt9|9^9XMn||$0V5s(QzYFd9Iieoqr>O ztL$0HjTf=yhO~=u{N_2_7NeQm&YO1uuWy2^v3f+}!Z2)Rz2QCU31d#3g2k&=v1!ti zh`Oi&6zo;Ppz{x~QLBsj{K1#Lp3A@g&uGHk=@TGQR+Z{bje-dy7?NBI^eSQ zL0;b_5+gDR-2WEPsb8{a1k*-{p$?H)tHRE!8;{en!*R^}B)Y};4(-ymq3^s_V@C*O ze!5*H1AF*vWPt%Jsa*n_l+Qrx4-M%16Arb1cn8)-WAIK2M_orTyuiC86XR5gAMaEC zo9{{YJG!Hnp(d`L+e(z9ZJ=)`kS4ycpf~i3n1QFS8PRt)so1wy}7oX;LmGPuuV-$RHUI7VC_VhIC&Z=gqVxLfiQIPq?6dsQN!}C(ON8XpQ+YNZm zhXH}dOl;qIlWY#o1!He7Sj#)^rudt~h5oNp)9N$^_{(v}tCg|pMhh}u7ZOv|(I_0> z#430hah!fC>baP5Q5H9#Ao3$=Hwc53AG1*IjU<<~e+Vlact6lcZ_W9V(U>_e9Bg)& z!e#FxkjxxsMIW4op#1G{^t3Z{ewD0#vvLbQ`Z~MjW4t;H9zTtPnXPt8m>+)w z0-R^!Cd*80S+E`>V#0{@!BuEosKm)vjo=;So%opFR}~!lgagVUakbL-t6ujXb z>1hd!^NCb6Sy2pDNe1wHOFAqcdRxT%WvP5B4^EU#go#sZQdwj5?%QRKp6D#)y5|L8x*u~0O4mkv6d zqtg5La?SBxWJPW;q#wRd*F6yzK5Xp;bMePGIzl|;YJ7#? zOQK__>DPb4N!(d;JMcG*AkMM;(O8{*=>++E3~gE12^E zGF-IzW>}Y0jJqvDIX8!9c3;{mc)G)!vs}ERs{QVa>a0>Z9YpX=H{#Ur`5^3EpAI%UxhC{t~R6_Hqs`+QP39VPJgo z96rAmkL|rk%R>sdJnd{+B`*dZOJli)8)Ml!n#;gsX9KR$O$TR@4p0tmf!SJ?LQesU zrvq=3WwY;t=6OY7`kHn$=ef2%4N0KCaX#kR%*4lHN}TvUX;`%BJFJsSrK@Iyz^zYZ zxFdKf#(2%77YZK2wuM_^NNXWjmY#+w2MaOt;08Q6>Nl+J$-%HH&04?VPNaA zST5wm8#^IvbP-XDaO3V=I1Jgj+4NsNIk%d$c$LFt?%3B5Fl4W zJeH4z4G-qwQNeM%(qB%#U)jJ+8|8>uZO;VLS1XbJY6Y}fr4MGRBA`CzFj`-9q?Zx~ z83&bMTGYLLFlKGWPW}P37-^4bV%A}K5c=O$V@)bvG5?*==XYLqART0Z<>x2X%$~O!`W3c={aZ5( zm+yfSEzS@a+{l(~*hEIzmO|-@Ox9O$02WtorDI#)p z_VFm(#ovjA=AE!BR00|+LTE(8bo{7#0nO~K_>5RAojYk8zIW@kZF@XSWIQgAS1F3H zkBeoKo?pT@GAnTHo^9+{NzT^fLo?(_PGm*pL$KLyDRi`N0OKdkWXsZzj8@24nth|0 zzDtM)Lvb(iGj=9?a~nD?MBx+NQ;3Te(udzp(;&Zg7UZpA_TE(X zS-1q(X=nzA#1GPITUh#R@dHx7Cj@FDr-6MdKMVML6y2U>*96w8vS*8Bq3m2BEx2$G zpG{N|q~eaFND6 zlH70s-W<%u+`UCiheSH{yCaM8D|$)Ecrh^EbcR_swT()zXNc~|VpM)R4N`LIVD+8l zv~#SeaB;$2I7WqVvT6%V3k@Ki-TaKx?K7;uW{-czJwl;ZAWgLis`~9T1*c58OEjww zp!=FWoMa!G<7&#! z#y#U;_T}+Hk=;C_(rB{qRGTqpw%M6m1-r%ZzraM+RXUR_oD|I+T!7rYbB$P0d=}=LF9ElY z8_?&u2silHg4=xX50)s32{-T_hlrnBY66}%liPFBnDQgVuwl}CPA4dn%YN&@EioC7 zdHYki9QRr_w_2ZAB@6M&`=8e7+>OidQ5t%3~#oc#5VTJxPX!&y)lFloDg>oc# zJtv%n!8lwI@E_;V(LkE455eQ;2jI1Zcc+d&hvhqNq1c~P>{MUCeJiZspW))d)pk5L zX{#4ksm|fZI2L8*N)qYIb4mP~3XB=(K;QfXY}}qj?L`fRy;Kb^mYg78waP&Hpt4Xz zl;`NLE{2KqJjeIz2`(#rGC$W8Pz{}bn2_p(yJ`>MHlqggPM(jwakqKz!y#^!-9J2) z{g^1cm&*t%{x zc%a@w)YRgzIZ}oUXQblUfP9dhm5$YOztDYRyr1n{JO=Ea3M%&^pe5R#UaFMB9hIBV zBR3fa=kxQf_~&$GnmkwhPLj`jwV~_Zx$ryx1cK8^*eV@EhS%_pOgR%yQ^%3x#MHol zcPyLJP)lztX~3P3fX8A?Irb6Hxs5Kw4;II9^(t#F^XDVpRpk$A$zRD+!zuheV-lEW z`{AS&lIT#^fO8k#hkmItI(Ld2Tv-@NZ~S?JX897FX4h|`BsEER?2dvkLtdG>EltD{ zqhwsR(~qsjnT&)(Akf3=ut%6I=-PCFd=Z((S^dc4cCB{dRF|!!9tE=8gulmdwCr*& zXb0a}+{y1;PFG?0N{ZjpF5`3Dh-ZbOaCmbV#`!2SL+5_Oi*=naz0(!PyTxMP(-aUj z$-?SmBHV>taX8845=}P}WhQ(~1yP|X(9h|3Hhej!wfZvLDgTQxdZi#US)Ob8q(#CP zP3GTkUijeTI-H?91Eee7LgzgTu=2^JgX2T6Lp_4r6isE0g((Po+M_|*awU{xrrAE$9zx+@u4)b^72DA z$|szpDj#J=JXYg8mjK(|{R`+thxe3dnqt$U3OJZI1UGdi!?Tqt{QW%#-~EfB1tW4` z@=Oeup9`Zmw_Swuqf6j-@MZegE+6WpJ>l)?z3@}q7+V!wuxPd!cT?pU{>yeItvpwq zRZ)d@H#?}(qf~wE0{FY;A=PrNU}H=qVU@Kl`PtecaLaI_@?EWDH-OH*OUk(%J!6i&lXxfA?H^-wmc5wu2LCIoP>z3fFR9j&nXJfU5QDK>unvk*?gp z7B9}C23J?o#r1jQNzoJ7aioGAXikHHW%bl9`Wep;y-fGk_0`BZpFjn1eXQ}ZhEw$l zwCB_XYEo5Aa*}TnKbvHjTqXnOzIoE3lXr=lmI3sO9|F&dXTd&G1YR7CCV_=w2ya!n z4Lk$g%xVfZ;M<2UR;zH+@^i?X=1y|ytr$DFu^ht{w%`-zP)u_sg~BOITVTNusv(Q=iXvUUlqqu8cDFLQURpS_A@$b#pt1zQfT~TfKpBUQH0Nrw(2Xt3mQ3-f=cIk+^+ksn=U$%BHBp1AIw-COjtZdVZ2&Su2>% z&$OD)ib9rtAwBq$g>`4u=v<%Km^u&z!Jgm9E9GjczTzgV-TZ?@__~5Cz0Y19`yD+N z_tGF~Q9AHtF}cFE| zRxmW51t#f+itmU81s1{Q1vT(kCY~%?6AS_Scam*cKj@Ci#h6`YkLh1$;K`Sr*th*V+P~%} zYj#ceSg#6KPmCsztcW+8#$osHCT76uDxBT%9&UY{z-M-c;NhqsP|UfEPgAVe*aK5R zlO92Z`CYI$;2rF&F+<_SIn0&IFBvw_7C)>rLw~)`#GuF*(}cX^WJ@6E1j`XmDM$D{ zdX+FQMw{ES?E&0*G>fd__r2`?b|z5%8>?{bE}gNv8qL+kx%M?*NyEnqc)Kkgt=HG0 zFXfL=D=v z@O0R1KcO4(V7jeTul5Y2sw3`aZs4Zy}o&Z$rA|3&WtYdS=9h7@{-sqQ`@n}eL6Iz z=g^H>C$K6elq!F|3q~0~aozXLY?B<&g0h#mq<%gvcBuljJ&%~No27BX_UdZKcrh3= z^%Po%N^>R|$z0!|FYJ{|TJY+{TAu6TMYgU|5}HTt;l>Wy!=K6Fur*hiOM7{i-7arJ zEg$^E>uu5y=c`7{g?w*kS1q_pRDtJob-Xjd8(a3rz`(d}R@xwud7qcUa~fuGezTUt zP2c6f)~gbk?eTD9^&HxH&YAS=2!Y89j8XM6zY7+>#IDTQ15Xqy(5J%{mq{4HoL=vm z+U*K3?Rp(a=ly+mllv%_rT~{j#&aNM4NnE9;n$}Uyt=HW+V0y7m^wKYC&iaAXd~ z-(L#UG7g_a4KTTz8eqevM!f!aI^8t`BNHw-{y3d*EfJ=2 zO&IT`!}1_%lJY8^Xgmv}pO)vNFKdo-=G}+;;>XECsh4;qE*3(MHnQ5HX4Efa6Uvp} zr&sf0@m&6TeDiM|pTC$4flIfem6{G0v&j@_r_gi`;FFi);L#dR-3Ao8;% zEwbH+O0(j@bHBJydx{FD;ui=9<<5fawe!Sv{VlkcsUcLHkPF%0p3wQ7`|+vfRowMz zFV#>eC4YXk;d^<$b9*#_&vTh#$yRB&Vm8c7RthJk{<_@1v)_=3yG8q&$Kj<$EmHY? z3`+f;il3zHX`*E|n{}?erfs1TXkGijygjf2;F}wSkFmzEpG2? zwN0VKQLc~_Jvv2Pl|t~1;tA3+-<;{=qM_~IHXL@kRA7mrRRFYR~3iTPz{cgzF7`;4Z}qt@fYvu9X~*a`HL{6uE`l|TqT@{KeN zq@bqd2aF8f%t^{zK=IH^Y@XL_Xj!rmUwa#K8PoJRJ(m~sTlZPKYWJJej=RURBU^Y5 zx;5Es`J4Kw_~<}1j65A?UfVR0u2mD@*7FK->W@7f|80z6 z8mp}NK(B&>;PSi=R%$;2 ztCqv~Jb-}LNGZ|w(BrD|__OVOUGAKDIm*%~Qt);bjT)>Yv+lc~s`GlPb#o^z{_I4% zXO_^l3w`OdjhWQIM3v9uw^BF0(`P-#7p=ZsgSMl#!VOlhVJzEC|9w6VKR-`G?^#wv z{NOH-J}(cqoO9_oM-^`7m7nyr@-w>C(HzXbRFjeA#(c;44N01$M)WLy;lp`$=&4l5 z=)9|jGrYg`%iA+h@#i#Ry$&eonxKlIIX$lRkUTG33IW#_LatXbd2)0Xx#M^VS=A!c z3p$B%#E>lv?xUXkGdVJH0=XK+$4!+x>59Ikp&Too@c;SBZ80N*KIdj0c>8(POs{W43h-EMC(F24;m&ZWv0hCn>?WnR z9Z!5@Qprx!bjarY65dLYbZmwl+^+QoRhNEdi~B43ROv1KDW8B}duxeX@Fqg$6C4+> zMtpkbKuuT}+vC?tJU^>K^qM*7;-b$qh}~k|Z77EdlUS%r+X2mW!MsP?hjSV~fpcvX z1Md1H*q^c&E<3m|p2>?zh)yGE>8@oNw_9i*gm?t?>7zwjB+MrO1|R=dQ}4YJ>zYM` z|FU>i;jz&uI&Xk!Zp-A6(0m8%dJyyPfGK?8^Z(bsNP=IB4);(r2DP4O!N+@>apLcV zIO^|iDx7`5sahu3}ew#T>KCByug zs`z>ePE>Eejg=p9#lvF|s5*#2crDD7QFY%616|nF&EZ;C(4di zVAH~93W8t2LDOzXZWs%*6wb4OC4jqK@TLm$|_x``Tp@2Ad5x41spinXz zXoQM^XURX_IV-L+3$=>wf8cv9gS_2F~C+FRO5&HHgEJSUufl}Urq%xu0J905Z; zIaGX9pUV|ZM;dq$MzZ$7&Ywd>!EYX0tM1IGbm-&k+#s?c=?jIgT2!p*2wH2+<_wn!X$W3r7NjYhcYLU?B^f$1ae9#)^gew^ znpT~Iz78YEFTahx`6A4HuRhj3RD*omAV*bQVrY^{7FzFIhI6BC5amJ*7~Psn?C!^s zXM={|FLeV#PP~P{DZJxy?I6sziGvtpaom@BhL?yVVG7Vf;-|G zCKp3*KBZG1IpODqM3Vhq6zzJE%~bgQAletgNqh4g605clnQ$W5_Yw3@h8tm5pB01lh?xz&-S#J)bpD@rp7oZA3J? z5e2igAA__}3@rB>p%eQm*%@E=B1k7ev%CuY>ro-zFEqHR5>aT@$vY;r=AzCP6C5MR z#>D-`^i}&b>>N2mCEoOqO^;5o-8Wazr&a5r(pw+y>=I>cWu8MzWfxqywHiF%+km2n z6^-5z0ybwNATB=@n(PAM^cznSbf=kR4;{sM=f=TX6hP&F@2IF}Ay`j)2IGrgvPvai z81H@~I(FJ^l04lM7p$2K`G-!hGsEY>r?2Vk?$-?7emW69g&X4gT2K75dn1hKjFO5M z;y^`Dfa3yTytW|~jh#P3>SB8cx)lOWGggE3tz8ha*a3XYL+PMW6u#bL#4XW#Nku-Z z&;U1K&aQt7==S+CL)4K-W?JK!eR}j_pe(mgqnyrk52JN6$Dz|QE4*bki+XIyC8f)R zQ8bD2SxIF)d{{zIbIn*#krT{kxhnCe74LSlevGE}3=D|PgICRG;I@kzmbL!D6D;ra zZ7n6|2J*qhhxfiDr{HRz1?cR1np!yPbMJ2t;Htz{yvq0KizX@4{WmPoWadq7K;kT? zD5rwi^Hg!`A6e|jKe&=hCnw^?u-Zr%d>-uK3hjc?Gsc>WR0+VqrcM00!xh}tY^R6O z7mayFr0ru(+E%4W(!LjBp>QaksIsyx?=%XQ?s%7LDS+Ridc)u!GJt{?3-q%mw&+ z0L_7{vADo9lei;;HU=~RdU9pC)T`wGXdO( zVzJvNg1hiljhp!>58syO;WXzfG+|K@op!1OO00OF>%bq_Gg}t|oYkk( z%HtqbA`0U7X~VZI?$Gl%p9wm80f(g@lAUIu&>AJnrj@L>8E}&)HJ_iemybKO%rl2REl+D(6O=vvrlg& zMM>MSdq+4K&1qmaWq-wKG3IzxvkK{KJ3OVwzXw|8A}jq0Ruzc}Hp}r2;F4^toA?f- z|6Zrv)qePAnhrPjV>ymDU_iXF5q>Pt2k9PHPH{Pl<9<)%)`rKj_fo{5GF}-zw;6Nb zFclLDqq*WCo`b*+(iMvuu_Ip(msq?&AKqzsX}%76Lzbh`YXfoTXLskly=l#yWtg$xI&K@D z4w{}(n6={?t9|4FF|xMD0jIyTfLGqfKR2g&{=ocv)I;WOo(G=m8i|?s%W^CAJv3s9 zA?G!;kL&Q1=I9eQ*gnOaD)y$K@0IzOQ2!2U%jUwbMcs7K(Q5X?d}~sB;2ItJn1efX zU(mk39CQ&kgr}*hO4;X7$j} zQSO=68SaJWe6sjz60w*QfwFGXaKm{w8hbJkK1n7K)qlD4ePbbf8F-0z+pof8pBVaU z@;#7~;Jd!fnt}oQ3fycV!}Y!r;ZDVBVNum`Twx@E3dhapOKvZIo->VDnyTQD5@Bva zY!ANFmjE}DQtDaIh{?~z@lW3jC{?sXfp0U0X=wA#-7(yp(pXI0GM!uDroic1IHT~| zJ#du26aE~NNdwmfBADw#-s;Chx~YP$uB)=Wu=pmcZCpX_R`IOcXHMMDd&RifKLX{X zUa)#u0rX(bKekOQgDlWFMDO+{(2BX=P*&_aoC=;#|E<~r{%6%-wwwsIGvfHFb2csy z9)N|BA?QBsEqTQEEPkAi!>3y#$;;F7@O5biUO8w5fvFrW^Ux(n-e<#uRRVbX{34!w zzKGASsA2G{$vD<*Iau8Kh-3L~h`HrFz`3CiGFKNR4Q|uTDvr4O`c)boR>Wj(y$@fl z0?=4khAd2;#7$x!;f;Vw95P#r1C5u$ z7Y44Kq>2T6N6#K$p+zT|t0;}0o?=X1WD@gZ;WS!1i{RuUBQox!JM`LxvnrME(ag*Y zYclx!Xsrbo_*x3z6iH%CcO1%>0lOMz<~41I~pq0`Vqk!HxPIFkxf`h!@MkIVlZ% zXWa*;gt+XGuY>WOQ)I5ZtzEl?V>gS_mj#*@V|pj4^{ zi&`Y;3oe%CxJ>07L*$_Ukv6>8rv{P^#kj74C9y}GU~_C5s6T&=6@4Y(ef}wVce{n$ za+0K7$Avf_u)wI0XE^z42!>fwno1_2^ndZ^jE1{6RuawB0m8CJE zm*LO%35Ya>c@!9g0%1*9djDc6FE0hl$Dc?=mo1pT52x|#_K?^U78tBkMdu7%r2pLI zv84GJ8qWAgveu50{fjq)s8AWn^y;H+s6huMZE@;OcaX=Lp@KXUchV-n|N4GyJK;8&P2c{Lb} zle#3aY4sa2Repg@)ivr+5%5RG~i%>6r7ZcCaJlZ%!9u+kl^{0P8Xd(jUS$*O{Vu) zQ4>Y>!!Z+R=yWFgRz<@3H^EF_fD!cfD=~++hY)E#PxHO3igB|%gFg>UhMkK|FjBvd z3iX`E_Pk6weC{6vHW`A0Y$Ig9`&Ax0SONk!3kVdFr<2PF78PXCjl;W8`u!Cq=8hG+ zYR3g;?S~1_yxtlLen#Q`^E{L0=?R>*@GR!kHB!lCh*c^tFk)&GMu`>B{ZD6d9i{a+ zYO;%Vgc-t%OaGxvfheBtYM>gswIJ*LFS2S#oOY?M1kahSu-9N2mtyl1eWre-(odbh zG*FYYiweOD=?I>QHkIx+m<;#6O@?&4iREDz4Db%ma8&6$2yTic#N%Qe-mtlkOZfer zOT#++y|K8gF3O#qA610cOUm&0CSh{z^C`^G+{pAbGt9(oHTdK+&j@%r^!~^GO{ndToVs?$GFUm(@cOx>X=K#2G3n44ACkWJyUP4Wj3)bzJNR~d(qLzcZ zNzl1Tw8Z5NeWtJx{#%tp5G#~^80TtvoY z9>*yvB{VAX3bEATHgm|n zw+IB0leqY2KXK+sS@P3Di}A=TCOMj0;dkUL94_QYQP&mHHhmKmMo9~ta-->+iyKh) zFyQ7+7wX&koqiOPhhMpk%=k!e^5=XTB+kkyS84Wu;-W%mmp8Bt2>C)62)ZCy>lYYw zpTs*)9HHaxP2y*9k~QEvD~~j1)8ob!a5S$JPH&4MOILD?tCJqlJ*q?AZ8O5rZhpqH zdJa2h^Eg`vrQeKO5kOU%8Fx+nF6xLHGD?DGd{Z97sr-(|@IEt0>yd;Fsk>20F&s4= zr105o-r1}_Op`~`p(^YS#T<)C_lD6E}}wuxZ5a|;mjd=u z+o271W0u_vT;LKz@ZBV4tU@Z!BpD@6#FoueP$kRW@xS|LU(8?lh6I1ECnD*g_&mav zojdV7`*=|Z1pg2f9D1(|H#B=OyI>YqmFEHK4r{oqxP9D}t1IxsXWm1tMv)Dvq+1s> zzy+Zg!EW1Y;5qIbd;X9Wbzf}4NCt0)wD)5K>!W92r7GWpjdI01smt-A{uxXNb7pE5 zogz`ApJ;5DI}PA_wP6v_@bQ5f++5d;`I|Q32bHts7cIMK-O3VZ_@ym4d@v2h=}U6` z?<6^sCwItioqast`5`zeu0XY`!MI;MkG}f*l)7Fyj{dtQ!7)F1IIJfOXCpolJ2geR z&vX~oO|684@1?-`NAt`o1@QYJ53wsB!GwSF1WS6}lMKqA;Q~L9w5k_WLeG>-DzfKF z=B~AEUQ&qJ)8mmi$dQ4M{QUR8TwMLAul&;aScr-ZgqQsP%G(?3X?98iyzSEi-+W7& zNS1Sr4>ReJ;c4XT6npsSrO(Kz^0`c~!eK8F+x1~a+)vwyT+YG?DEaO+QKUC$&bK-Y z(oO*X&$g&~YytkY3MPY=^|a>EVNAX1%FmeJ6J|v;%BWV@@}~#>Js3lqjs6hPB@?l6 zvpV+XW`fUQp8Y-F7eD?=g}Zq>vEo8K)x9MSKXSxa$y-vOb>N7IJ)4719PZ^6@zJIUixC^H)xy`;3qP88@TKHE?SRl2|5gm`rAu9?l zGg+@Wa(0X&{@w5%Zf^G=zapFXoy$+y7MIS{vV+*ZMw(<-oS>WJBsewx`y?b)kE#>I zIdi`aVEMNZU5h`U)#yCfyzUGfTG2@#*Q_R`Q}xhdb2)BKGKP}l25j)Y)zqF71zCqN zaBF-ZarX*=$w`_p{pC(PXnG2`iRSq7T`QVCYXZZ@@vIS_hq$=dlvMqCPPYiexNC)% z!0h8`s$P2ozZtZY-K`8^Mqhl#ee%B{-gh%7tk1{vz?DR$`5=bo1Y^wKb|UQS$W-w8 z&(&YH&{PYcV~h2%F!C-$J>dE6g|28&=>(#GWy`k}Uc*7{QeZz$BNtW^3L}o7HGhas zdz*}FRwjbo>*K_ujzjNN3Xr5-Lsqm7;f(}EC~uns3G-*ri~gzLthf;tuHrknnr9(A zR|0;YYeMrM=SX+XZmM)73=fyMGvPb)aczh*`hSuF%Qz?8-YSC5SDax%kS-Rg+4K2F zQE({?#%lSksMT_r9#C_^Z^p{lJHrbm&S%J(I6XErX(xuK6_c*|OK4RbN^kCZMbd|- zVCFwVkhsD#Quy|2|u-_>9$D}D+Ztm8qfAQ2ZE3k%joSdkp_2&~rK0OZN_OnzP z-r(nxiZL(9i{>Wmc20emM?Q3)q;o>2eS72?JZ;r9Ao>f+W<56uU9qU{JA3xQctpsYS|}S=Dajf?T#yN)BTN^ zQ_q3sXchhCcpCp6Z2-$nGbu!-ppEB#2yxJY?Y=Y7EA1Mr6XX+5)l`x)Vc za&*PcPo(gu9ceeo!(%^I(f!G9>GJHmc z-Vf%lWeWcKoeLvQg<$yMJiWE28q9)J>D;_>cE^($JZt$iZP30|b}jw}DNCG)!!Gii zRl#Lo#WphspX|ls0nu=0cO$4~MB?@qJsZo-CvbjwJ^G)Mh4c|++^_c#6CS9MrSC-{ zuW<+Or4$E&h%>BRIvvfAo&?AEsc55}hdPJ<1FAX!s>ja7M>d%dP<@g4`XG`y-*pel zC1W8@*AX+~V<14Pn~u4T{NLgRQ}G-Vq$rP)o1T%N=gXEkP<%A&oiY3C8x{v@UQ$_u&@4)MhIWPlC{pT z)nGo!o3sX}Nz8!h1A*wlJ5&rFwZa&F7MK$Jo}DZ9MU--Seu#&RPP7hi(I%2mVDJ@{u_y7@|12+ z|KLr`mEQ&sAfAe$f;)JmIfodaHC8)~MNj)`XjF{`jmoKLASENviVA_@vJTqvyPPae zT8h?VuAxee4Ole1C6hunquS-u&~{D*%>5n8t?%D~yX)3~!#_zH;&`2|{t`|qwn=mB z{d8C-F&F->ETQwIPeJFlAjV+XK|HP{#8{uFHgopOXUe+r;rbYJNR=$1?Uh2{;ipRF z;uo-K9_p~iG?^s6F5%xH!XP=riQd@1itq2=WPNqF;S3_h34+}TOgutEZ%z0~a-KFiKP-g)W4z;q@8FA+#*y`V_YmpVDQs8sA~03@4IOYIp%d+*ZS*qbm`{?#JJH_7J)`l^qb3M?*ykup8$?-%bf4RzLqhS@}5_ zQzwp9>_hT-mM}`{iE>tf%LU1s3*pDa1~R7UHu#KBgjnxVDwF64OL~@2w^`xTBj*BL z_o1Cl*c6J_)Cyp27thJ{*n_n+o!_;FqJHZ`%t-(JO@6J7ol` z?Qdag!4WiHo=Ag?n^EgP6c?}A!2H<$q+B4on%$^lM<*XxLiQVm2`Op{YX7e$Bwae`{bgUt6%>u@Yt*6=AQ7DU&ESMv%G7A4YOKFm5_O zXFMy)xhb7MKc)kB4xh#Q_m^V#A49>&!^`-7t`PsJaVP3(o3Qs=5A~N7t8)j?#A-bqP@aGtEuZl^s1oDR zL)hddCNQ2L1O~~W5VPShbpI&X+{dGPupmuU0)meK6_slM(V-usXrLzvt3|Vs; zdRFx2ieQx5t0_3M1F7G|lROKy1$AyOCj(;U=oc@-efreG-qEo^C;m?UJ2L?{e5xn= zJJbamhF;*xf-}@`m$~5O?h|;U?HjT>nb0A^_r`+_(3R&b{A%b&E7cv;J3RgRt(6S!@YQc%D8Azr+yOFF#A!G_jo=ySV-DWNxU_`Mh8gq@%kyj#cC zo)Vc$LvZ_zB=>6RLS%Xzxj#=ksP{N~rtHQvf%5c^B)B#WjZYb4zyV#(rvE2;J}?)Q z;S$^+b_%m|hpEGg`}E+(O#C}*AKuI`g6m7;@QCamvdyJ|ILP~vFUCV);MxzuOexr^ znnOU13|>?@h1*tX!bR-Cd+OWK%)c9wLcHLG+6lI5wj~GqPr~f>P|k0;xxh2*I!fF> zPtQ;JfbmCPqI<_ZESbFs?RFQVW@Ig{nkX!=o%aa+Zq}j0Fz>-!_Y|`){a{npbe4Ze z=%gvOdr-nO7Jkl&!z#1k@(9)CWM;2$*;%JjEcVmoK&OmKjG4^E^(nxk$^yC}sh?SM zT@SMN8(?rvI-M)Ck}ZV3Xy=6PgUvOSDHIF^eQvE{EP1<$?j zQ%|=K^mz1L#=>+dgx0QsT$l;bTP;agYZ1b`4muk1f4^rJ#xMJfhS^d${%a1dQ5nXJ z#eb;i_D%5n$Q~$4^d*Wj=8@((S!gmcf|vG7a}lo+u;Rr}Cc-tDS>Aohbf-1p zgxf1Q%dTi5-&{{t=VU;zUJdiL!<72G7z4$o521pQ#qlMhSYw?5-}GCcT98aue*FbI zwq`-((Z#fJzbz@R+J+YTGodHq1|+Jg(N+EN{5SXrY?-P7S62wY=J#3@=~Cwo{}kp5 ze2>F_n*z9H0i~Eoia=sBww{*0cEZ+AUA%i+pX;B)XKV_~=!-rnu>12Cdej|BL*yxt?d9i*b7N7v z{~am6-3VV~iXekOn}uL2EWbR0|M6aw0?A~`{5J!+-*L>sXLU4#9I{Q{-e@b88&4y? zI%A#xIh^bBi>yhf_-3>NQ>x$5>v2hFc~OLaF1PZI5d|`4{&kom@)W-7_3#CNYfQu| z5m>!s2QJP#M}&mcVZ2)`>Z}!T9)IjnhlG(C+%_tgeV8g$+TiU;X;jR<86~au!sjua zjQgk%+`Lmw^mcWTwtGA`;$#dSTHJz;4p;D-`ZrveYsi%q?;>FzUY4gnF^1LqmeJU) z9G-u<0HTguBg>X#;=#EJHc1)zV4?k-KBFr!Ve5Q!?9nBf$;)sc956NN4YM$X@9p&3 zke|nDsL9=Bw0`?F82nI-|5XmsgjQw20heXCK2V8EpYfEW9_S#CM~(3GqB{2ZPfOB% z)DjN(O7Xm}ZnWPwn|rOqd)jt>rB{zlAuldCds3Fw+Wlt!x6T z!`<-8Yd+^1#c&h0{G{)v&EOLJG&tzA79<&nahYdSxQF+AA@+J3vzHr>zh9qb6FkP# zyAsRL@p=QXjArS=n~9{xLKlj2w3z0B2xuF8Lo)9c(3R6HxGmc%@Zng*-@eCCF-D12 znMT6noC;8JECl=UugSO(J?Q-Xh1v7pTJGI~lU$v~RBp=j*|=)CHQn8*OqZxhVdCpV z#;9sB<*Pnemj8x^cJ!f%%x#!#s}19`<4I$aHr`R)P0Lq>;0Lc0xc7887L;Y-rU|{c z_1`|yP=61fx!i;;e2=P0eGw7<=>TJxNIdrV682OdgnRD3aHQ=6d+)cqx!4<~DI6|1yx@m5Rzof?J*x6THW@5x~OH@#fbY$o(sG?Bc>73`iZ;+Pv> zgGUw}$E0;BG)dwWjlK1aAqn!t@|qU5URCCqt-ka;zmqb_Zf9iY-(&rjCBti%&EWV% zK)!xI4<>zlwtFeT(XrRbD&9x-pfv|a<`{GF2RbR;XU8nMBg*HXE0`N9r%m( zF@Yc2q5JS7nyfn(ceMSa&Gjen=?W*5I`taM?Buvtc2Zo!B7P2Xw2b1bBSgN z#?CfY!7pvw!MAuT>@m9n4_CO-)vrY%?ynitnMy&{=KzqI&w_JDEBtB9uKAC{> zp-wV*M*&J)lOS~T87sXw74K%GpsRrx76p!ngU6*{XjclnUetp2@st)M72_n`Y;5vR zrwJS5AYgL@m~72~vr{A?=WzwOl${3wms#@1!VBAsRUu~mHj;KN2gbx;k_77VPbI&vQu(EShF2k=l^27Zy=7j;*Sxc zx^U2-0d!-$K=F`1J{(biy|M?G5OxN%RH?$h;3~djyc;Js13obVPQm^iE4J_|?0xps z_U^tDI8WJ_%Zypc{m6Pl&RVoU$dVK=7vF%Y7E5TB##{DEmpHojTtn@FTue|^!HY}w zf~HymdvIBzy+(>g;MHV^$lyTtRC@L z4u94aYiC~b-l+`LENrMu2BGa|!9p<#Rox$H#lkFDT)P%dyWw>_8JI)R2U%b1JDKGR+4mfX#+a-7b$ zEBJ9sKUy|q;nP8V=9bM>D&c7Xn^fKIf9Y?LV|}60^!}RM&{BJLrAoB zAcFUB+kQSDY#|E6A~JY%?oJpxJ_Bc{-GgrRBqn zp{-0k7Ify4%E>`sr9Vs_X1ZZZbtkc`D}sP|M#QBB?a`-IiwwH!qRt!kHT7rMhAoRw5!eyCH;Yh+WSpS&!;qZ4|)0*wz%=@40s!r48 z-;a|!-W>IHME^T|OyZMgS-6LZSqG@nt*rdqGpL+r>IoK|lFo1K-pdoDaXm52ZC zjm)I6F9*m?&t#jD#OpYQ$)oPdf0%C7P+V1g9Y-F&frC09(MNCv$oReF^Ol*Ms?REJ z_kTHLc7PnMsNRk{XC0@tAJ@|5fB1fdqAHI0F&{1GhT}hV8ThGw3ncb@ru+YLbjk;H zc$IvOJk=BtNR*3k{#uSWa`FuXEHdG~cTVL_b6X&NjX!sGmIl;x%JI3UW0<~IN-*<* z2b%PMW1g5w;PU=Hc)Hn(bG4bqO>yJ-W}j3!+ciO)U-w!}3edyi3H~JS+X-;FZ4Nu1 zmBKF^BtlOVKzB?!)JrWSPFE8#SK0+>?`cNIAs4dV=|i#sOM@o5W5%aiTzupr?rt^2 zk)4TTc*-!?c^cCd(yc^tc!0iLFHP4k8zP;@(?HC3DovYWf`*H}4xP;=I4ytqFS1|7zN z?lKu}d|V2jFQ19_DNlHh?M12|{Rt%{6lnY9`Ecj8BAA@n3iEhIE?inlPklRzo6dMb z)_7fR!5bAWGwm33$MZQGcWK%=)&xc;+QR;S>*yOC1BWx~SY2;X!O*O8(7vI7XA`WX z(#6-|VN5<=oIMro@ETTi%hD%YGyT_ioQh0-4JKaeAo}N}vd#B)L$lZhsMkC|b=$V1 z`nqadyx0+re$J!8K4ZB!EsG2ARNGa81`r;A>O{tQi+i0&4om zj*qG^ed2jM@A?+Mb(Uh?8DseP%arcCbdyBt^w5P^3a-zxz_nlyhwRo-nJ!sa64Hq# z&a**oSph%SRsr8n6JXnkJ20hogk;;?B@38zxUArXTDA6=BNl;^`;X&+-=du14-@Xr zxNqp@?uB|=>`=Fy@9jv=g7o^QOiAcBj6I%)YmR7fFSfEonw{XIaH$zn1If7>$T>+)j zK~Q!mo?1zjGM+2>8K#)B;Pd(>uzXa(KFnAK^HvX&EL9nSMad87kGu&Q0}p|X2!>IK z7?Sfd9&PLDsfdIUZ8z}84W*LYA%Qe!63TV$)@FpLiriRA)M*4lv-L>kp}`@YO^Pmijsc3e|0}5)~S(bwH(HD+8{!6 zDk-m9j@Hv_nM-Xc^qO)UYnMKihK6U78xrC8hJXGYliFB4_vhGi^DL5)`$W(DIkAZO z4Uv=RBP8MEZb@b!#TZ#$0E<_fejaZ|>r)@Z)yc9#{T4AHnS#m?>JUyv< zlev|&A4JP4@%#6^So&l~xw>wqwV@8uM#&VieOwXIn|rTJY7#%!S}}-^vn07~AF8Rv zKqxnp&z)vn=HLB%-ZDVD1B2_F!9Kd4aDuViK2IaAcon|~``5;ty1o)bJECCqlofDy z@;=^0e2?cIC=khWE%dCU3cZ|VkLPS!>DinM@L#V9e4NBPY{o^QQcxUBa$W@aaUbD> z@O5~kTMn#u4~ed3(Lw1v%BY!f3&!T4dB=C`+Ckt^?-+tZXKBY{H8B2ejw6#UW8}Y2 zuuA_CJ5)QFwyjym?HuZ2^(*_Zc)bD2h_AhQXI~~Au!Ax)o zjkfJ?lYg zld-;K1M~a5;b)~W(m{E!sHnD0HlHkbPMcxPBLNAj%`Sg5^#Uvx-2pjsd8bWgA>BiE zQ5+Ws32y7i{4@oa5b=PCf5+z)?><5mz8{yo+!B4>38A|B4f?Qao%NDQwai7D2+=by z!WM-rM&_h9>Azk@UTR+=FG>}0*?=+z-m^k4IawHe9}PbyT!j~ce7N7X7H&d6d40Bp zx?Z)TUkrZIS6P0Hq{~<0GO-q1tklU;lT++)avrhuUPG!|u28q`Axzlbjoi5ioVIrn z9@d>rhKv&Fw6+AaPTEfQ3LU^3>ek_k0k zYjR+?k?pTNPyDfq1RO0Q9(%=MPQ5v-VT`~hcnxmC$5c2ul^rj03U4==;-i9FBv~cb z=IVqgc&A+qcg27nfdvdaMZVzcb+?9QBMn+8Mrek z+kcVboB(oV`wmoFX~;f_t|ZfBN~wRRInG$xNDn*ok#L_0WGL$%y*AMsdtV5X=1gZ$ zY7N3=c4x4`XePf?nv1JM=RoblpU`glmQFu019RH>_Vn8UI_*g-O6j#@^tg6nq~%Ia z+1{f21N~v$mR7RAvY*t{zo%bI_dw-`PcUxR1k?}S$;i%erv?isN>6sd(=~GFw`Mbb zKhsL;YJQKc?$)4RC~NX`ChrM=Utq zSOsp*j+OA-pp|*a^O6=HFNdXft6}5EIUHH?kP6McKwUN{30irkhSi}J@IzPzH;S8q zM86v|L9CMm2Z%xQr57k<*a(}%^*HnH3>>%g6i%I=%uenyrM=g+*m;&tkX;@Dg7gh| z*#81-U1|wiY&hQcu@!IWB$DpENq9oB6MpG`rweO0QlVa32>NgquKx(aJH36Fy6PMb zdp6>XAN%3T=>nW5w3B$QcZOd*We^drgwI_0J$;QEj=1SVe>a~)j(HB3gFNs;T`|(A zXPE4F9gTumoc!GzK}?@3TkHbL!gD!|h1S&FRuMJY(Pp5w2HgiwK<{GS-Ltg{+vd-K zYghh}9m+hnN!uM}YorpvkS9hQI9?tXxU2lh2WLz^n21*@!`SmN6bIf^Ll4h{n0Qd0 zvk4f_bqnioOLJC}xaX4uY8P!8pI#;I(6ZI6VrwtXjJ1c^Ix<{DVJ41*)M9DCIhc{{ zg#meu_${)S#P=x+jP&^X@zoe;xcvt{E^}n$$GiZBjYB~&hjp@9&{Jd&rypFwoYr7m z-sg%{yLzy9^-lPj%X5509k@sKAw(1YfW)W*cg@2e|2){r%@DbQkCm-(^Os^UkH{q{ zIwcq(mk%W?8u3M)1+~ptMXLWwgSg#05`O*;^sLXr($F~&>u?JXCe7mXcbXupx)Mtt zBBq9@!;bVw)>@j0F`pxEhx5l~ z;5s99%+!2{mCG-~zK3i1KE5&;_h_5oYlQK%zIKDr6 z6f2ysQ?6Nz0}$Z7w^Ce_%Wjx>R~p_w)PdCv0rXrC?+K9FNmb{Z$4|+Xcxn1xWG?6n zz8R*lg~BzUG`WWE{CWb7AAP{LBEe+e%0wz(c!uVC7O;^!Q!&LpgZ?;?ORw#|$4cg! z3wD}VkbsVvoW;OAI4OG%y-LhDiF5DK$Cbr+>4(^n8U%*HVZ7HTk9D5&iHz>L4To2a z<=kU3@R+kStWH?SMf~|rHIs_@yHGpsel(u#cFTpPx?D_DFT=}!j*-55>YSKv6a163 zL;Go}-08Ijf_ZuU*prowvyK}=+@d&Gz38cJiCa^7_p?9j{e%4O-F)Y}*7z8N7mG1;PLNdylXu!QQyZcs@1?#)j`Bq?OoYXxpKUeQ~+ig7M#uYcoa@4 z!CTtfA>~LadjC)$50b<1QnnxF*d<|OJ@45G1I|lfX?et%81m4?h4t{BCn(J~!4(Ud z@mt0W-01j-wDioNx+~7n_ylV@_ofu>Y~GI*Qq^Gfu#|Y4n8N+|bo3cnk9945OxUt-u(JIL znEB_Rr%D69i$BP)Ci6gCN`v;PXwWd7Bv2id5j3jT!WvD!r}Ld9rswRyKEx2mJ(-FN z&z(jEuPTu35(sAW&JuWMN7}AFP{5!}EdAaEMES>hO!=dUF;U;i-A{4&Alw-@>eo}d zaXd33WIWQt9_YQ~D`lMap;FQlXpYo9v_dzpNrCv896N)rwA*j>jHXicR{VY8`*SaU3f6c`+_r@$+p~b zzGw6w&bsiFT<}%kKED;>4zE9sCLjf7`#fkhbDMGbJ)K@SI~jNSJwnMw5w5_?1TCjl zqAfpz-q^v@LjO?P8o@Cp8{j+&+_laa!ho+XL~Zy90f~+ZBa%B^goOv{dwP+$TmN?*k1-; z`7S^$hLIory~L@J=gwW8Z)^JO1{OvI;p*&n#CqOo=6P@#^o}crIZ;_qy48Z;H;-dG zyI+yDL;ghcaS@3gJxvk+U*F}`L7~e3Ob=#Rb zjtj&Tj{t|@Pf$>m4d{Ho7wD-+GWAMo7m(gj(S>lFtRU|$h_rhpd&L4PRC6H zm4pE%W^6OvmW)(+brN$$`ZO*&@C2L3tmn@iAE5rke8$x=5LR?$lGL`9bak{nv^zh6 zH=oDCW#Mo1QP(uw61bfXzrT(9l{=_pu`tSL_N|N(Uc5b-yD_f{JFY*+Ou=q4uyixID6$(hPqx$dc^MeGHUP`tN$@$N zA0*-;-y;k)N7cxU#65Q*z~wYBXODw>zL0GnuE0&|rsx;F0A_ts;o^UEW60z?h?dXj zxuYe-pm;VIi*JY1^_vL$^Dk@3ZDZFRilL4ZGRU!%xg^hR7TK8l^3Fzy|50=vj$FN8 z95?>Mquv&W3NxB}0H+=lIq56L9M1h~GI zlKFG;sXpH$T;W(p>Sk4wZ4J^eBR?9%jnv3~eK~B7TT%TaB#I7)ZX_#e8<3L>RDoyW_V4`W4OCqJiq6cq`V(uFy@v~+aHJw&&T4Wx5s z0{=WQ)aK`VT=?P>y1Pm7+(s$*n>~k!3QdHsQ#4?AO$)uFtclrgZh-8H0&>IEmo<6D z-~6i`_%iVp8g)cLpT%*g*3KgGJTGB-%nKYzd5?n&>Y?Vt1VL3%6nra`;@)i(=OmRD zal3Z3P@xkm(ICDJjm7z$jb){@XU0wX<$@0VHl~%(x%s5@QoF`a8V__!+f8G1mhba4MD__Fj)6l z8Pwf_NokBbd_SH@lJ~5@A(a$ZvQ-&6V>bhhuECT0!%^+?csQ=d?>|WD!Jhlc@H0{Z zo`@n}4^eRb+{!$=w+`|i>Oh{Z5md&D;pvX?TtI;ySCheTkyjMZU~nq-JqyRF&uYjl zeG3o~{Z8B8>VW5g+k8gI0sNkvMa$_kvA?ziU!+e&Un5DH|Me0fT~c&m=@SyB5X;&< z84E^fn=qwqDoxXyPVyG@l7=rk=~?z3**%r--^Ev;%I;KxUoBwj&2sWHdJwhMM7c!U z$z0F}XQ;7UO_;LBaPN=-pL;n3dd_c1&d)=zYP&rg-FAc)xs_mbX9)57XO9QB%E6WZ zJ7&VPEOPYLdE(o4itgiUuJ&w2jt-b`iY80A>P1`NabPr5#a6?F!P9^o*-f9U45 zxs2fIf8=dw2-zcAL3c(+l3mXq6Zxyk++6ovT+*~-T&TdFJ3e5*ojak!J>OwzdhY#Z zqBNpJ^vQQJ{+}ar@!@35%h6`=$PEZ}7$e9(u7GNvd z&=~w1!p=Iur)vvg$6Y;mD6a>-x&E}dKpE%GNd%j#6>v`YCENK^4|<+S!nePBVZ^5i zRG!~qKJ=Pl-hZmVOyJqOJG^1??VDs(a3mU+K4Cu^UMK4v)^lmGlR3Fpw%C+=3@*Mg zg59MmAPtG!k8x3)$s7aDbN_UDZC@R}mi>n|(&eBd8C9Kgs*%}{kp;t|PH?PxAt=*sWBEQ+E@x{vmS#OA zjz?Y+G36!b_D%zhgkA8>@=oU3TV1&C;|1^S(qWckEcE~Fgu;Oh7!vgh>T4^Z*DnNq zw#Gu@uoV2+B7pw)O2nySAB4Y~%VGmf z*k;MC=X+rby$q><3+Ho?oY~f-!#M55J*yXDdWbu+4L{~mA2)Y& zxugtL*XBUS4^L1Z9wnu*$-Gl$7HnGCN7|1q28}DBWVih^YBV{VJh~c1uJFuF)t)BO zw~Wup%y8#+y%ys>g%&cp?Us=BG5`|PL&4eoB+M=A!$NT{V)ktf*}Yd3284ET;hh0o z{PD-wa)-|Ynat!CxR`U39(q7e1Az^}Sul2k4!E8#AcyarqA4Be^v9AQ-hDcri2ghe zGLhC8^KK~?#{Ngr?G|GS29tqt?j*~nf|OmBC$0S1jaDZ@fH|MRm1-or`B{tdfxA%L zR!JsV{YTy>l`=ZqWrp?+lLOnWC;B&wFo@{@$Tm zH${P9d?B~}&wlP#sb8jznK2{>O=25iy~Ji>?ChntBK`}yXC~`Q3YHj{U-isIrT;7$FX+lO=Nc6a8d*^Z0Ll5!W z9ZRqs=K=d45mpKEy||1g52#%g3ycOK&EU8Rrk(=%X`JS?t;bG{Pzle zy(z)@*9vpl4{y+wPvr2!dp_&J_u*PkIN|beYuc-D5#xS^;@6P__`&}U&9f~411kea zSgZ`wrx*!@a(=UbiEaV8-C2jr(jIm5ZUN#K|$bnzO$3f?W zFZ8-tLHf&9biO`{(d*<J!9hbjnXr#q6kUP# z{t@VM9wL((XK=fvvam#MK39>>^RRyfW0MD8=aMIMN4Nw=^Ebq@u6T0AV-5}1-i&>x zpHy~>-EOM zjz_zR%Rvs_j`qP@%{a)GD1#+sadezjHtzH(!i`Q(uyFi6`sC9!R0@{mE~$@V{N-ef z{+CUu`eqpC7Rn7j^uTwOlTblhluXu_!2Dzr&S_f}EGSB)`+kT)){qeAAfb+z_ie{Z z0gLc$UkY4@jOvfT4CNV)ES z#3$15qfwgk&(Y)-M4hK|`<+p#aRurb{71_V^4zrl+VSXN65IHL$+bsF?G1xTnjr6Eq5la0_rTv0MZ2Y#HWXqR@5ELjRsGsu(N=`OH)r0L& zeEcAK^lqS2LhsRiZ<-jD*fjJ${{=fiio5W83Rc~@O~Q`&V1T3{ZMvz+3C{5@+<|8} zQ_TrB$@tLAy3u6s-$_s^Er+JJH{hHJ+i%*wc?{n(?m}*mJ!JKnz*bBoS=GhpY+8-M ze1?9!Srl%Hj;y|8(9RC-b%M7yD4nE|NYx((!ic;t4PBs#vAaVcKq`oLT}a@VeqBs+ z?}V3Y6a{a}5G>xL!{JXo@IvW3S#jh5m@>z3`}I^xg2%!u$zd|#zg4ihG!nl2_ZZIX z&PHRYJ7{Nc5!gpjW+K5cR4&v7L*KrqLkp0$9T7sCqq7BV(w&P+99($5STx9~1;eGu zy4dBb#trQ0pxb1A(mdr_BC;ow?2t=`O)6_>Vekn3QEP`?b>|6NvW9K~OZwu{9lHMf zG-_9Ah4)m(QAoN*_O1HMj?_kT@i&)n#e4m^Wp7hZ?C=EmD8{oNrM2+st_*lN;EI(t zowR9&A^yDINBbPYNMy=CBB2yWOqM>Q(E*ioOS!((u$egZ|QG~VMWI)$) zAHNr@j@i=Uu;E-A4O~5h6=%F?$o48y^II4GoL8oPsi$eK;w@UTel70OW{H}2DbFtR zr32+6oYs%&*fQWxi$7GEeX`MoPyYG>nK_1n5@{ur=j@mtYLjq?vE+33L~*Aoe8_>& zIgly(!>o@^zym+0fYrVpI`piL5(^zHDhnd#vP$i$)~;uKa-zRXERtDqTR+Wr!5n@R8-(>R>-x{l7e@5b0x zO2O@p-Qd@m%YJq0CcDk5nU1slBriXKX0;}wQ{EK_p0Pl%diQR@mQ&$0^}t2GRy_xQ z=2~F2?ObXwZ82*8E#-Q8hPc~jJUJC64PMXifr_fXFwNyJQL=nWRSyz)UUU+F@0Dbf zGB}b{;tCcu^7M>a8vUnOLu%R$kyB2A?s^SDMY@>a1v^uaKIJ+3STDt;{qmUF6hs4a{{LDR{y>f$2e2I7?O)*5-B*_MtQ!|7y*Y-4lb>)6rzKL-DjQ#BH@{&nY%=a3lD?7s|{K|2t~!6r!n`;I;g$f3-XeW zVEn5TZkF*kF70UucQRcOjLI~LP31N?%Q#bS-+yN5&Wni{pPPz*^_v{3H|DduuBdu( zGYQ^ujO-ftNE}mnCSa8>D>ITzyDp@LyZ?lG$!-!twQaTnv-7=PY|5G#g(A{N{Q732^tV1>D}1k201g(9$#re@K~O=#O-=>CyuDukir- zmV9T%y;;Ow^pX&`TFw@jOdAC8U#CEMSt_a6uLlKMN~k%*3@;Q5Q{B85TuR<~uKbRG zv;L08& zyfwdw^vupTVdb60IpN{BzIcF zo%7~yK!%Ere7Nz7|1)sRhm$KIqqsw9g!enn+vtk~+up1|+G2cqUc+S>bsT4eZF^U># z&4N^!Wq8rb4eC~|!u5Y$@Yw!|kewoj309_fXTKG0Y;ogrtruw0-Wi;5vU{X%iW>sxf zbCU#Y{taNyFN2e3i@|t&ClPh2WrH<#z?ix=5{$pb=Z4%4a&D{&cPV!&SE;$4=KvaT5{J7m zAnXRrFlppDX@_9?uoQSf_3J+XVulAwPXc(Ys-e-F69o*rFNb#4Z9 z4qYI9_lMBj^$9z4se~CV$VIbxId*A%O)wKu<0yr zUNP^Wa(oUM&<%F^?o_Gn3}#=JwR0UmJ5F^`;g%xPJ@m&M2g! z^EZQN2nCPc2Cztf1RX<@!A?$!+dLG7hu??dAn%=s{t%Dtj3-(c&KKP1dJ7T!t|$#? zhMVE$aL}!iDQb+TY8?Z#yC)l8BvV%Hp9?&xUPO~N$db*!go)aEGl&+SCP)@8M#1;?S+*tF{X-34SM@h+IXS&b9w3UKZ9P?G!Q0dus*99plL z!v}9Eyy>(8M_wy)`P#o|{>Ku~FrI?8hY7uqDIxe*y9;ms(Wkk>f8nEq4UQ-}lBLZR zFvH-0*t@!Gsx@=)$RvI2fLV}mxv|K5oXva2|jXXf8Zrl92ff;Klhrk z%P|bkyUWpMvSH}iz6~3MQrVfF1}ItGOU>0+!q@dHcqb1Z?;Br)V`9T-xWy2pex89h z3j*-<*n{xJdpg#>(B>Jfk>rMmEZo1UjE9ajBGn(EJbtZOS2_Y#B}byD&2Db4ZYA0I z@G$f2R0QvBe@4oTIv9t9lh|r4$=Ul4!sZ=5NS?*wv8Mqj?#Xv$Z*~!>%~6mfC5gjt z%E(9O2s-xMc+9l9&#YDwf`rlOSgFv5Hs8j;_oq^rP%~im*>44SJzdX;h7{v^V^R1R zv590PAH;gQVr+<>Op`+81TPyG(#^R7?$oFm=FIj7;XBW$=Am3%WBM3d<#}h8TnLz~ zwgpF9O|X^Q3i}I%1nbAWVgCq+VTMmS?(9iI^Y3C%k@Ss77&~M4*>QAKcqupMsWRI0 zOtOJ*GFY%_Hs?J-iuVo43K|qw;w>)&NKBc`FgevUL(UlL=cbdnJm;&f_yoSY$}>QP z1L!L!eka%C3V9#!kmeRCf`hymH%%#>E)(j8h4o&TzUvph_CF4ne!K^PSSM9WyFxnB z9uTL>TKrvl8X$ zE%1pw2Qr(oi2IC#Xu({9$+8XDF{74fKNh7AC-DA@dkWPe^LUTkf;2yK>5F#)$tj0G{cIuq+}}hOoIip#A4^D)l`FJ0rGnnwGGe^a z7Y&P?;KZLvXns2c_czAi)$JZwbFU?C$e>BDM5)r?DN?#6m?{_!V%F;g++cztm^@U35{0J_=#dFiF1%v)U8+T) zCk8QxQ-3i;ob;$3H1(rVVg5S~}o^u7Y;$?U>3v2>_*z6h_#cpT*2Wg~x+G-O@S-#v`O5QM_?h7=KI30I(+CxUhp3im1<{M=voVv^VeQ*%RFkh^ zQu=vrhlC`sT}VVc^~ia#^SGPO+=M>9LM_8{*?{?W7=6-$#w;3P56)aj6_$^XBgWa- zw`C!@&u4RuItDn&!2<>_4p4~&!U7Xdo@XE(MHa1C1wXcS!Cb4w@M^p` zUFwwpUW?7(tOL)&Hk`_I=4+GYfeIWa8qdCVl7QD!9+EY>lOeOg8mAmjpnIlkpl6s6 z-r{G4uLic5d8x1Bxw6`VSz2wh;r3@DZB__L#nPzxsEILs!@tw#j}S9o0q&Kq#S^M5 z@{$)4FEbSi+P9+oXavR;mVwSXXPCS24t-`^#P-ZNPYyaE-F5pdQ}jX=TEAuBYLT6| zMN0;4CwCJYp&=MMM~Q}H%0Ng-8MXZ`3(u{uLPAct*_ffj$S9j)`{fc`D|iTzAM2s> z-X_9rNWsR_w^93tKDVgR5R7$ZK$`pwyl#7ks>(zVyD=f~V?56vzV{4oyLZr$dyyC? zYebIy$RX#;3UNwi5WH!=00(}z;^2SHY?L4chL_F;^;OGg*2z358!^ZKo?fIg9#FWS zGp2e;K@J}F69Mt|3-sD+MHqR&JN78w|~w!|`QnIA-r%0wTHX#I|)VK6AWFcDH|~^IwI-$x(`ejLC2{rxcp& zDOo$k1~T>>!{U4glsd18Ixg8{%ga>Mz5SdT1m)o?ejeU?-vJKY3?@?dd3V%jdA_$5 zPA3(yVXAH_ z&1v|o!BUk-9M}Dv-YLw*ZC!SZK=U5!v3Vmo+TLdtQMwwI`0KEKSAMZw`*o;~e+=Vt zQjLanekTbUN7=m_E9j?}2k5M8GTgorEBLl85#?qNl8b8%sJi22B9`u5!AAgg)98{(|tnXs2;sQ8y-<hY2Sy4lFvP!8$BO?`U#WuzwCLGhV?AaLAIec$s9(~XgkC)vX z$<(L0v|axfO}p_5)LvgBo+d>o)ZtH?H2tyIa0`BUca@a42+gMLk=@U zP)n%J8rUYP4#7L4VaG#pm@eZ%8eHumQvDttSk^ ze72#?7isc$)f8wvI0h4P!f{-QHki#`&+~4s;u>xP%pPCGRBY^};XI>n>&|tIU!)`C zo~Lb`&PsnCB{dq_aLuO>$9^@3WcQ`G_0n@Z*|7zBKL1C8 z6Us5jXcu2|Y$d9i)^O-n5WH1d1pmHHKlAXFVgk9;Gxste)jePgaN@D7YTmlwPmb%!yQa%}sSK-St>2~MPoVs6wp z95*A0PS`jWnI=Qn7+ywux9PzY?FJIqE5U_dDy8BxodwCOW2>w8HR70hCn^pz$c)-y zx?gNQT_$l5MUE+po2b@@BxcZ&%1a+cYvysDMn~5koAiBsubdg~1{j!KwS% zME=uL(j;_`4*XO>n<^h77hFr9Rd=%;8R8f}ZW@&xNoAXxPNDMpQTUR<-vR7;=-S1t zyvHnp958gHeQV3KJg#$(T#1uuDVIh)1^$|(+?7Pu>;TY z{bLeoO+4)95sQx>@%Gr25LTW`Dt1lezdNf*dFy<$ePaws;)QyW8piu-rG(ifJ2U9= z+I*bo*#gHu-XI@yO4#ATWV$x!HnnZ8qsnJS>5MB8Y=6%mIeD!6zYmX@>;V1!1juwHpYCkUeyn?6-<&n|{JNWMr?<1I@ji1l{!${hWAJltE zSW*mWSsp_FSv(}&`bjk7x*NCh*%}zDwFCPjP7`0law<7d8K(9xA!jM?f4pgqj-Cl5 zCb)wP-0;A>7FjC!UJ17PsBq)LHzHfu16Q$)-u2hRj7_sZJiiTX4JMNn^_ig9p@3<9 zEQ+h8fV9S3n%W{q%)ePPjKoIhvRuc09XA7lZe-%#C3$$wYZTTQy+wt|cNz=1)_1?4Z2T2!xc?Tj_)0nXZec;=9_r%mRV~!a#S9{L&%{mc zQ(;YAHQB#Fo9AwyWaWfrVbO;WeDTi~9#+RP?C)K)Tky78{fjkM7WEav1`GuMBKCmA z#0m86+er|dc#cFWuca?emJ=D%aooJYGNL70O(u(U&;?#6nUOEk&BBbI^0i_sbraf( z!A_M-dVw^~F|)&oi$CzSt`F(IV?;xGiinxXh*_r0b_`Z71*gl&5ZmSj3(H@SiofgGMY_;9mm)O0LlOqX3pJo`AK76Ue^8 zTxPSJK2B$S(c1kE%AC|d%|Hu2TT9S0-WtYPDMGX0tT>vW0nedK0fDyB?@Fuu|JQEGZxGkLN?p3!J#hE6P0_poC zy51G?MYmA4%oOqmJYd3}HS8D2$K!v4u;=Y^)ayD&?fa~F=BOG9&Yh&Gy_Lj8-<6o! zX400A8aP|!8+jwN1{Ru61`HGe@8JRb?z#~jw$0~f)$Js#qL@9pk+1WA2|=`{OFfB7zeKv_|Aju3&OG=5ijS3f%FO#NA^XKB18`*fx;}jm+xCezRn@F`+AQ3*b57!Kiru9_MFAa4jZD!<;3BjLTwX|iU191zfm(xS=$gDal#Jq%E zo94p{e(1^dK&*NVoRcLn)fzFG^L4KJO*Kl2wTln@0-mX`svb9R2Xx|XI zkh~9Ge&YG)Ci7v1@+rngN1vYg^^Q&tGXce0U&zk~Su}fJ1g*Qdj*fRqLU%(-g|16t zrJgN%2G67at#ze_%pQF0JqKXIe)7hADZ0eeF_jvf#6V#`UA5jCmip$=CH}XW58ah` ztFIN;Z7jk!GQ(8nn;x#a#P3?K(Ph)64^X#CXA<~gCVukprxE;feB@9V$+gHLOMl9e zN}tWl>kni3Nk=rgo~@!QJQ9eKYApLiXFACEt%o3CVbr*=5IiIfL;J6p#CLExy%lOe z{0775ZQ&{?A2%Hq4*Sv6kk>e7Z7g<;AHuV5cfo%nkubfu7c`O|F~NDhAkF74(sR6M z&_XNPtkYS&Ro|LTccErW&Q;UjT8flRc#4}p)u3M34zSnV1mDOj5-Qn5RzG-3Dqp6HoClvj&jf>G<`_`I_e{LCp(pDso!8mS*jLVkJ==rv z_~tb*D~SJg{`90ZhWRARr~$6|e&O8%tzZz*&v-9y!RcWEpd?lcU#8VF-}uZ(L{1P) znQ;y4Mz5iY2cPR|ctW?ir=nELQ8Mtk2_tS5vfmfB(&b-LP`a7YhKCOHT4O%-`gsyN zejTXF$_XGV^^!2&`3hZlAdlygOG2D_2>U_mFAcjh0n8$jVLm&CTQw#cboC0MiqG2U zPWwdPXUW03lV+ss^Jy|sAi;D?Yj6?Azp^V%2a)Ar%jsMjQiNk*}mY`d86RyiX zN1=HB`#02pzi#EDI2mB#Uvwk$Voav|c2P2q=pE-Ws*LWS>509TJvY>vri6MHrh zneOtkfSqqXgr=rsN zLJ00DgyHQDaO+{cfa- zn}m;$3CCnnRv(es8$(9EIso!)qvUn|MCRoKB6%>6D*bjq@#_UNP3tV(O%yTY$Tu84 zGagJ5U%-ZDFP=H2%;j1vf$Ix0$mE?~XeB-kSB)A&l4}Gm37rX(ZaI>8o~v+or#0St zAIHp{*w5O1bOOO+YcTm3fh(3ShEsl}FglgR;wO*LCMldQyIzO|rh%X(dlmdv*U)!8 zTgY{>)$oHqw{|WZWKRJ+KeYyh=u0~6z8#wPx|!W_aE0eu=@1~k8aMbGvDT})D142> zms5Yze$Ny19T$w7=Xhc)7eTL#UL}U*O~kHBo=n-Rf-5%dCif=_5xu?3a9LUgD_m5{ zTut6weWvgvOdq?Ki2HY=amRWRtI$HF4;}zj#R!->uM(u&w^7eI@*qF)3bb5mq<+4~ zNb$uYMyNv`d=hx}dUXa_dL{<@Jx z;_-VA$WHDRWi6+Y3p~Sq>9KCMTxcO9WZ%Z;1)@z)&o{##-i0_L{uCU2JwmH&T8Q&B zQ>Nzc2%TQO2+Y3K!Kwy1n9`O40gE(6KeGpxH;8cG%Q)Vl@u&JWb;P>k2Q88A=(ao#T=_n+{sBJgeexdj_^~|c zE)?L@tU~(bLl9Xv^&z#r5l5623!%e!H9FBW_Q%pB5KU;p<-AK^!{{aW;<_CR?2Xuu zLgD<~Mx9eTK%rFd3v#sA(5(7jZ@R#9-UA^- zrJHt$PX}wwz3_7GY4&I3SoofA#pb^&qIY-kz3Ny`(qAmYS!9kyw+}nPq<;=qZxn^q zwfsI)aVn++-=O24Y=$~BPn>Wlk6bfpg-8=k+_AxkmM)cLmU-u*(WFw6C4B+(tBug^ zYbTjHyM}zTorU4{S7D~N7HB+?=bBcZ!tj?@;mpyC*vQ>w|M{K)dLeX<;vGI9oO9~49XL}9K`pZ6%aoCK@A$c@6wN->yCx}HoxGU zfg`R~SFLW;mZr-c)o^E#J=R=`0sZe1c;~q^`isTD;K4$A*SVT~YkUtBU)J&rS|td4 zp#<|9GE3Bzjg~fwvJ7oCUvUD%~H-XE!e~J;$}+W!hxWS-1&p{z{M~$`(LB zji8v-Q^-B{4nptL(QcPJB)e}u{BtxzbNe0S>h(I3>nVd@rx%%Bc5VR8t}t9}uT7_? z7t+W=g=!vgSWa0z;R{@ zx{0VkbV)Nb@D4xC$7<+)qmt~6q3GG2%V#>&_&n!FoM1hJT{4BQgFQXCMQguO=h$K5 zJ#PprYbw$DnlE??yzt|L*_b#iATpQLIJPB>%n9ixY@IjS#HvAT!b*_ye@`o=UD07~ z7-ZV>uExCgv?zQ9)L!Fro8Hk(UE)Jf2z_QIZ{cG$Mr0To-+n^00^&iZFv z$33=RSO>8%GRM@ham&H-t~N4?CwWeJ20R!i%zY2AVB+<2 z;KiGD+>XcB$u|1}@ZTAS;$Lj3nL-!w7o-sluNX*<@dYK@g2 zrvJfyP?_RKd=@3b62%Pqn`cQCxYyv<^?}?%tvjHuSB~@7U#2U2zQX5WF+pN`Fy#M zx#PMEhfj%^z5lNjW?W}>g2c~a?{aRazvcsxqp<9SPQYUGlaJD1n{xz3)@MS-b{lwD!Kag~m(Z2xd5-9Rnj8vGL?&t@@zK7GJKC?Ky>APCyx|2k zM}pC2e>HPcGzXoWkJ5>+L^!J_Vi46aLS8vPg;A$ClKhXQk<;#@R96KqaXW}}PKfY% zhyiloLMi08E`!epLrnE}2I$%G2XOG!XS{FMi}_cju-hUMkM7Q4_uGh)>g4qpqTt9L zwWH*?IiIPyEG@|P(qOMyeIbVwRLJ+w8DQUZ2US16M~k8@Ro-eutY-z`{orx{Bw zPQeSt$MF5e8u;#Ljlub;sCg+3oK>~BKX;_KtnH0J?wyC(=XoaN;W(W8qz@%|e#oJ= zyVy7y2vLgX>1Ce_n7ncUPyf7$Uj|RJ8A8)Y@Jx5umoyfORCS4uaVmUBzJ$q3XL8z~ z<7lvLIe7{dq}rd7%H39Q@9a3dsI(ihw~xWG)*@(E!S8*znM3`I4oJKFh(@dkW=yBB zw0iCZ*m!yeR*mw#)vwFYIiI3x?z9fx0wm2k)NC&Vze3j1_!(lz=cpv{dH%paM@qo({ZWa42GJ;cvE z`M%iY$?8P1pnz6{%LzUes0((;&nA%`3|Tg>j`pujfcjmY=%y&jeZ9LFm$ca6-{n`x z>2u~dlw^WiR0py6emfrS7^3s_i}9UtDb;oiXXKCWhpjT{*f2?x8~JzyyC-ULl{^cr z^o}g9pDMsVRWAS0ITe!f^(w@IQLQ=b&#GZ2y@$mZPiKePvjT4YOSXB`}un4 zoUEWgLyhc?k*redzt5aE|ALJl3h>3_H+1&;dDQydA{hI+k^DY(9^XVx$FG7zsAK*Q zQy=hq!)miJG?KIoWXV_OcLson?=@ zuNK3Mr}1F4;2ze++@T{OQ(<+iCiZKdMU!i1OmF0FAcyT;ndHJ0v#e*zXjeEN3L1q3 zcuWFq9C>F{0-wpSk;UgWf60v7|7cm14re$242Fy*Vd8CZ3>);OC+Aa`)H{KTx_^$( zF7KcfEw{*0bp$1miGuh1JM`DYdHDTrC`g{D#;~i8NHp)gJDg@gx=$t0fK)$RraMZW zy9;rFBBJE(?o)JycMsL8aK(dd`gm2fl4hUY3u~`tvh1Ddbc68{8fvi|ww~S1+|^4X z!D>tKSf?dTUu_NrQ|j4urw+pQQwZUE@)+y12he(z1-*?Ys&Bc8LYKm4Isi~k}r(2TI-D3roQC~2#GZyPw_&a3mS={p09j4YLfmekg zExt2O@LT}|h0BK-!_XKqd$A3S3u`A^yE36S&p2Xe?~9*Ng(B3ok`t8K1M zr}hp~aQ^-nG;rBM&lvCJ!jp&bw|)?&%?_vHktxLMt}M}d`4^o&^ZblUx%9%H1hV?{ z>FP`0B=B;;N$ek+iHat`eUb1}` z-Qdo5mb{UaC)+F6g2bT~BCh5}AGF+{dBZ_iesC(Dee#YPKGjC8`T=rlg9?{W=7>2; z#V~<)$aPgJa|5xZxS?wWm2X}_{7cK3o%8q69`}_@gY!}{)1sdZDo_(#4hMnmL}@hb zY%+@%8N=L{s;7gMI#8W=6^{QYVO&<~kixEekWe-jbH?ohi`65bImQ%oM8mM>K`_{h z9iuDWJ|SXGjWPvaKoDuINRu3;FKf51(yj4Zd{p8Wr-|4lGume?1~ug zt&_3k-46Q6$VgCYRShmbbBK(^6p~YU4_sHCgRdH8tk?eiSU=?)=6ha-F5h5iYK@1> z5ki8Hrd)U{Cn1pCd=aC>&XZ9;E7fRHFjV8#Spk zk2DC;C^X%(UzZ{>QyG#eGLs=C(ma<^DGi7;DH00R*{_|kWGEssB|}N3%)FD|`TYG` z-|t%AyVm`?vu^k9bM}5cpO44j<#B@Fz$egdt zW~02l*proCh|1bdqBb^wE^uxnT7`n!b!UlZTu z6cY7a;j~F3fS4z0K*ES}ew0}p)p0fiw+r_;CZ`JfBi1nqkq@)Fw}LILm_#ZB=HV&B zN#IufgGkgGLiBACJejwNP=r2Rvr_ zP~IC$tx{rW)$#yRZP7?u&o`31&xGXO=%rVG{^3TSUx$V5$3;IjsDrSfCzgdmxp~2J z`edvN89wv~S>P0f9Rby18i4ioIa=>lC=x>vAbc9Ok}t zpsB+=*tuyr%*Vo!t#Bx0LFYqQeNHy(^T>nR!VbYZ)&_;#`I@%10w=I%6#OW<19w!e zkU1Jd#c6|6S!3Z`R-~E126wH5ZK@^UIHwSDFG`C4S$`+*rx(D^b(O@?W)r{EIRfFN zHynB|Fo>kvL|F@TA?lwpd{0RuiL&V+vFCwk@Jcg$tR2P=diRaqlCGlBf10t|a4v4} z@S?fX50h5ain`3e`?QO*q9qZVp=RJKXPN52{@XW-`K2sp^Z4sv_I(hj7-zs9k1DeJ zOB&Z@?}|~Pd34UgENZ=PIkh)C#uqFtC(|tskS?Mqw~{zo5p*0+jhl^Li*xB*IWbu|Dgl+$pV6H|3B*U2Q>S%fvGQyX&i-Zz z%WOtKuZ^_0V@WIC{WgfdI6WMGbPIjV{bQ&?^IgcRihvh(3-I)X%V_KUhTO06f+g)G zczJMq-SFpA=-=NH1ZLt=vL!)>$!FL@zEd)j6}owkO$^9p&)MA9W;MYn+xNlCD5ofcO*9qu*Y8+j?RuOu%c>2ox4Fvi75wBJPGcFlp&QVi7BIG&k?Xt&n z7ynS#h1Rr1?h$&d{zFrfr09&_KY6v!8CW&ElP*>tA^tEml~ZW*qmNQ9V*m54_|P_& z?k_Thir`W>*EJ2}Zh3&CeDmnAu7HryIjoa2I<4vMT#tfT}j~a|nH+=$Y zG(8WCYKM~OKo#(I>?NKnINTAqkJ?VIqf_r&^4{OIVPfN8zI?V1`P+08VsnJEwT!p$ zq5m>IkWa=@;wW0~)Qla>8WL9A#M#p#A^n*vy6h3-#?oxespQd&&mi|MX%o)#1QwsU zO77KfgBR1D;Ft9>==XFwNGx%NS3M6QophDUsCNxlQYrs*zZe=R%L-sL&7l$UjRhriG#F z&@N*MkyST>ZC6F;;<$x-t=3FN{tKXgKQAC6zw;=W(8VXN-A~UAbP!R^ZHV?b#GQZV zC*&aVAs}59uQjCcH$^4EcjX7Xa? zdxF1=(9LHtcx~vX8xLQ^n_IWh9S+*;&X_U8V(Ca?xAhQtJEDNwGt?Oi0~7G}Ap+&+ zLUB}S0{q)d=mgysh!*CRgTDSHtFpW44YzS1^+#|Mild-Qu@Y)OMq|jy7vx975|XPH zO}4zZ5Eq1Pr8h^k;gETk>FTIARCl5-+QzLxZz6*YMu*7b>Q=Cjx=rWKkdwbR=1;i9 zWl4$uz30x5mXVN)>cR&d$wc zoP_uP`%6M1eB%GEp<$50CgBfD{O|pLyM}T9hih=MbFg=GbFukYGra!K z9<(QpBiiwQ=zOOS|1#brqIa|#r*0IW7uJ4h20wm(`cbv zdBCk48U)&4*8a$&L(O+MaiBv{5E|WRxESF z&hIG@s+|jcVR9&&Qc5B#gtM{`MX>F6hNU+9sLkX|+P1%&Bx^RnxVDYZw*5Tsl~)fF z91Z}v83BFp0!}N}lCcV2=n0!R>o-22{n8vIY&Stru{MkwA#m|8oW>0yQD~}h5(l!r z;*J?{%xktfvu=*U*_lu9z*r@Aqb33Ej!k67qt;-#^I+8dHj4T%-udR0xF(FfF@s$ccyw!xnow~Vj`T`lZ#Y8dOlsqUnjx3 zRn%Q=Ect%A4%Ag+X>#`nSeXAGv2#8KU$VB)OB(mlf0!kTpV~pb??pV7DTg9xCQBBM z$Jpz_XE5aex9aXrtk;%A=Pp|)U06>=YEk6Fi6(L*d>ekda}rm3HlpU_zbI#4K>5l8 zAm_3dtSe)Myj2~#tkZ$0=WSr(m>@FRHH!8>n2kq9ULuF?YKk5HaS+|1ilVdC!n`nn zK3=Coe_!v$(bO6%67@*zUISJfyPORDzDPukOE9)mOtwrB{Jh0fnx>?$WcHRLQTx;Px;{x$k&obizug^1WxdCM%92uYm%M{;ekCyt2|#m)25fla zhyLX{@Hp-}EkASVqW-{Q~u~;!5?=P}q z-%3rHmAV8=QST>a--ogmozaX9v4D}QY$05%fU(qhzkr;Rb{!6cEo9t4dgco-}g%^aoqzOrxz{9A**60UKD&OkJ=8C zN#2RL^mj5don8rDp+k7-4Kdj5cmS9E6hRgmQ}39g@OGRM^Bx<*`fi*fo=M*XK8hxM zZMcKIa-O)`Y93^r-wuiA29w*DFVbB)b})aPJl0rQ;i!VQxb37QJniqkIMybS5}#0d z@SrlMxFH1GrrJU7@{!bS?P$EG`{^cq4P8zM;9@46+glH@}+gHKZLwc z&p!I=tu*YMrdnIDw-qz<|Im!E3pD*)7~FVNMd}?qaO31~dT{+A&R$sO)OL>*Pv887 zA60FJgQhw0b*J0u=D?x=CUht7qGI?mRXOQ!Qy-$avis`NM6nj6feu7 zVpV~8t(JsOEuRoWWyb$YwIt222)kdQ%^J4|XKnkh!z6ouc-i^GSz5>@&TL7iC(dst zekJ?h#q5pbv#})Sv;7JQ)jLYJn;1b(+6}nul1oZAp9JDQNbJm?A@MT88e7v&^&6Lx z3#TX0>rH*N?_W&9MJ0AnG_*=oJZcQN)^LU7E)>yAWAsFZ)3s3URWGi$O2@z*86;@U z8#-$9KKy5{kMF-GqK)H64*yn=KW_T9u_g*+sgrQ_Gjt`O$so8KdYgNHbR9;;&%)K3 z53tw%l@Pfwr!RZY!6Zv7{Ow#vUu4>`FJe{jEeWS#6^3}uV>7ut)SGjX9gp#IE9lNAQ@efLNzYv3}=Naa!yxo_t|sh3o_Hl>Uf$U-A&U+b}pIMPP`}#SDvi z=5Br0Er>NnK^cNV`7m6>%FKLCsyV`~w$HY!EVu*#xF^oTy zhe5v{a_8CuP-NmF@(%C-A-NypQ6M}VAZd3BjeZu z`sBnQp??{R+qb`_+i&*4VU2PyJ8~A5oH&Yq4%CqOPo>~cQ#cNNUQWVFY+-KQWjc_p z&tAUXOaea604a4zakuU!P`Ymd?fox#(}@|}^*$$fXlsHF?MYPbfIZjL{eUD$Q?gU571B*LH(tI=fiXm~pzlE^C*;+>JYxOUVGJfOYpE|Pp;5bQt)>TP2G`1!pyDt(*2|9shg%WC!mMZ><%WTxtiGU zyc#x`Xwl_+%*maMnefR$3xbI%bXo5Mp9Nn;X)-$ahbgUHZCtD z9)hE`%;tEV$+#Ev`jc+jTDKC4Y74R9f)Z185jyVB$z8UW1U*L^NGMjoJA+fHwZ9il znDms)9-IfKq@=+l+kn%47)e)3+0u#jmvQ^1kzle+6E1)7V%E7TY}Gp_8u7XY%W{v? znq|@0cWf+fO%moeLle26Pt&0Vg28&K1g?4PL}hlXKtNwGWE~N*p4DRVdb4cp$|o^& z?uMDfWrT__!(GJWyOo4qP9`~LS}M52w!(=XNjm+^3|!ujOSn5u_~b`8xY>W=W*YjV z!|M6eE2c{5l!?gV>kW|RaFH5spNl`%p2zh|*5dvJH~4_F)9CLHA862kFIX5X0okR6 zU_ZEs%I+MC6ApxMEiIv_W;X|%QXN32eH;3=1);(HQoQN45QmRn1fTEc!rJ9+MB?Em znvit_@}Z1s+|i{bW^F_})04WTW^u_@d00Ot#ktmNJ~V|CaraBM;pWXr5O%qh?6P@J zjAYDk%VKFrxNi;BNAmcb%*m9Kmx7UXmuT>Dow`AOJ@m(6E1apPN6P}^sJ!b}5|Xh4 z7Tp^P^)eFpy2h98>$0txB4bNNu5qCu$(tcJ{wMvM)k3y%sl@7&aQ=Cv2b23B)6+S> ziT#8%c-n0qT@|2>MSD+?FI9K()(IV9eq+q{=13Chu4OdAuA1BBZccKBKOogUM)Yl& zz=POoDlmG-qOVdwM@{2_0GYNxjeo;7J4Z#kM>cT{9Y{}rL+twm_%zMQQw zQ-DwVX8uijQ{;#F(%?XPj_5cGn1x; z{1zO7QbGoM740bd2LBAM!=B>;r)!roa}&CR5juj$E`B(hf03cB7QtiM^O*Jba!8Sy zfHlTx=(N|EHT0zLOJf#+Taz+dd^Q@j3WC8`#T#;JV(`cqU(C**K!1%1L%W?b@Th+k z)?|Oh&bZ9W^z+=O0JJZ8##!E(`H+W5H%KkW2mcRx0OPSX|c z4(o?+%6jB!P&9QFx};10izFUPwy}-5`E@hrjfZ`^ZxAhOCzNr|#$=aNG(4P!X{Wc)<;_iKTREM2zK_9WW(y(vpU?x+F@V3__Ozn= zIw~)lL(i02icOPlP!=L$Q=eC0Z?Mq&eEE@{`|uns&q%Vs`)|;26BD`yl5EQyeJ-ZR zfm3mQf@7|R;?YsjEcW^c_IT_S7+iJ?G^eBroF)fSDg4Gi6>5<)QN3JIp&K9dT=1ob zIgyhtC74jD!yX48CdHFWFt2V5mTUb$omYRj##`E?UP%KD&kbT;sb8puX(xSXm`3?i zQgApw2qo6mqMAkwUEwj3ybV4}RlPc}c9=Gsv(*USX3qqY6i&a``(yRH8Ze#l3kT%V z;G3q9-9C7M{TewNH(Yo}Y70$3`?xM1yERDYRFtAZ;}rN)J)cTFl@({0NsGsqIm7DF z!$DWzC{BK`l{eRwqA|6N{jdK=j3XGHyXxp>~ zP-`^0CKI&TITcO|?1_LLRdKmkmF(k{K|>-~=q@?&S)454Y{sCvswGUU>>*JjuhBZ+ zhqS48F}e*pNUQrbvF=0#U#r@J@BeD?;s1Td@uq^m(x;hJ=&EwB`jo{UR_V}`yoZ_= zDe~=y4A84rg4JC{eoR>-dV37PDrSI_8dR9=b{BTXWi$R~ponsDdFZ{&lB_o^Cq8@j z!lfB)WVzC5D%$vnyK8bAhu*q^I}1{wFINImRYmBKoyO7^sAJ=%cW5R&kfjJ`Mm`h@44Lq_*;)v+~fUNxZJ6;oz+TgW?%oz6{uy9U>89)eysg7NCA zq0l;E3hXT}1mlPy#JAxrRyBBI<+!J$we~b6rj4ll$%37le-w?M*n*6sBa!~sMJ@=l z*}>dlPW@#Z>3x>SJx)p_p>r-nf$bl%qEC&98gG%cCoEvh9$TuA77aaJ#%zx1Z}Q-^ z2>Rl?>FywkWYSz5c62nGv0j_^5jp`kMqVfJ>y^aY#yzH2<(`4zM-OnEyB>bp3jd3* zaadL02Y1&Rqg=8o6I~U4k6C{(^J@{h`i0{5W7FC0Y4)%nzyR|fN{j0pT`=WOF>PI^ z&fCrOg`^Nlnil18qgsDr&}t(lw?%?onlqLCwz>w zp8kD%47cw(i&9GAn0aLY)4uP92R4cDYHc_CTsDDi+^oumT@d=KQw1mXr)ccD)P?UO zQ^~US27KY`N?uuIqfz`@zU`AS^}n`(?U}s+MN_u1q%;!*}f#icdqSeQtb?im3CG8SkdCr`XW66uO379{oB7~FDCRh%^b8puxW zCkw?=>~m@|TQg}CD^q+-B8KGQV?_tH<^5N(UZWJ>uk^zUQ>w7NW(GPB9OZNU)llVB zCBG+m2wAS>Mbq}y@~<1M!NYPeTQ-dwmRzc{uNjYS{ek#7glpC5_pj~98FPO;+Z})O7!TEHR<~2}M*(G+LVI^+) z>q%C6B%#tp9SrptMW38eBhPhRu}m+7bj3;%)2Sk?S6dCPyPrX>PdzRzJ`aA;XYt6# zWo+t`n+VIbFg`$@4Ky7SE&gK8_L|(Je=_fJ8=F_6O{_4-n0$mB4$@>DLFsTedNMs{ z^a}bmeV_&AC*U_H2j?S}p=-rPzOEpN{Y4| zV~VIR&;geI)4*>#gNeFX3jf_;Fs}0YN|cH{XpYSf6xV(Qw{s(K_6j+c^eTnqSqSKu zfLxLB#@|9NXdWDLzJWi&Q*d^>C1&<*N5^NHb&1o?P}`B`K{NX$34Au39oNl;QbQ%; zu~Uit7O!Pl7gcE5`fGLVYh~D+JJG`aZZGsyG=O8Sl-N8YAAdYKO#Ap&DzS1Ade&5; zNN}dcU(tuqhz>NUJ&va{7m;kel{opwI;y)hQD6_-LsQrsm@fN?bn2+H?G8rb%-C4w zx^6eOs{byv6Xnrqg~0Y&R>SR^VqC16g+4kNeBMJ@cD8RJ-Q-F=k1%TuY|mw!agc?TMuaAj5h z3~=S~7}(WThEp4U@of^jKtAsm4z8@gE4wZex$}ms`k$0|Uvmo7op^-Cwv%{EI4H34 zQ{k9h8T2ogBKr&MF?KD&dr!it>y&?FhroMKmiv#TAChGSi{yz!kvpF7 zbjAF9i~8Lz`@kaJR$zfG5j!h?B&%KygVx9YuzYJCitbyp_K)-FSl==z_t$`5BSu1} z#XItCeJQ*gw1;r-_p(4W1IX4-63begXHYs1jDEhx5))ZW3jaz!|I%jdPl}k%)Dr4> zU5nUiofHMmo&l8WK z;1J~Jw$m$43Dn9UmKL~o^2_5mI-;?c{2Dg~KDAn6^!#(E;2w&b_*8fyWHi(@HDG$w zbhhaBPqgZNz}p2_(!~BulKJ`>d4BganryS9DUm%?_4W~Pvev?xXBAoH4S6WAuOtEI z6Nz_gH)f_wF|4ejQa|p)XPYwYjyTA!OdY_Gk=2;FITQUQLs6!#6}5%#?SZ5;CMR3M zEZ+yS+!;Hu-)krRV-Sl$b)`5a!iqgovtX}+vcS912vV;`)ivusfc0y0U@R92%kCZG zv)`DKZ6oC2%gZIWZ02ql@l96zaGAPz*1D5`scz!7j!3fPMLlW4#iYwUnDY>F1r@5vx3=iifpr6 z5c!;H0scY;JIHev)5tYsVYV7#`Sy?4^CVC>B35K}3A1n+>Bjt){q)%_I~*m70iVxf zN%1*7vP#t*y&Qw_PG27N9sh_wC)$q0!HAXm9*4ew)A--KXCl+WnRUOXrIK-_X>5>q zI-6A=4;ifiVDY*WtfGdn32=vxvVM!$^px|{HX^c-v!IR*=FeE#kl^7DYtJT*0PI%?*qi9Q}N#M()LJ3c`gbfzmY<$*dZ8@&aV8ZCz}*X&T+ zRe>!|w`Wh{w3SuMsM<|lDJhCq zq-wIirB7hQJbTuk)JJ7w_QNqqhg(&8AifYnVRt4h)fo~@&uAyo_J#5y%K+Y31`1f7GvJ48ZvcE z8V(OA!^x-D(7}(-k<`{DaP`zvPWpLrUBb#ISgxzgTvilP(G zvXB0Lq#`a9x+VB8=aZ4^yz$ zhR8kH!dzhuxOOa}qfT_7knYCGN56AA(+n`_eGzW*ZlyI_m!W8(2cF<#FoKf@e^Cc- zyJ!dZ->21J>eP?g=gX+_v#nrewutJ<*+AhqC43)nnr?Lc$C<|h^Lh6NmjrZz<0uU% zGTH%O|N9L+W0D~xPmMJ{9>mU#ILw8&EQTw)KhRATh0q}Tk_Od{08ia^)S6!>iY!Y6 z=UI7N#AyXKVe&ytmT5)q?it`D8{mCY5NGpKfxMl1g7@{G2WtlfB40KZ@2wahsTLjd z=dcd6+cHFmXWFyh^M7C`A3_gAvpub5U`L8bqe!aK_&bG7J6od1y{#y=~O8mp# znd5Oy!zG-4cRwx;F2~WuIoNSvG7I0(L?4-~NBxp})HErT1pJplu1x+*v)tlw>$oC1 zb#613ieHkH&Fk^U(Ym_AO%q|coIRP!y{ub2Od6zPy~*N?q2kxW5w2vtBu9)CaeG7} zpQKnNxCG;2_kSfY>B(9;>{k-{$7KN*rwIDFhaoDgN^l;`Bv!sh@Y3YB)F$m8*So%k zMmrxt3nPKCJ*_ws-t z8#S*IKMK5#<=d4>;i(k7l%mM~Y%hg_Rtdl_2nQ9JP%!IJBGb3jh*Hj5vEX`DRF{)y zzwASJKOGBhqr)3sZPFLIKW{sE+I9wytG%Z+f5yUt54T{DN)>rC))0Kcn~3z&dK#2G z0)}-+(;*elcnhN$5am1`Huvl!!$E4DE>nPoIZ344)D`nKW+wN+jQPtsbkabF)DB3(qXqz7@GN|o?yLlr4B zIzc9Lja zBFq^yuJJuLH&G2=3GM&6km;X=ZzG&HJ zExJKTiaA_Qr0N*~@N?h<)MqQh?=2pjN3#*x&x$!$@kL@OUP0p@3r=yFU{bPfDZDla zCu=7K3;X%gq`yECpFZX=phB2SUAjgN%=<%ruYL=>;4zWDy~Ryj%pi=H0X`v_KDu<162A)Kn-Rh9 z35=)7(Zw`u=34xcCczx<+cST`jk9r`KbakHoag^*B+(^GXk>ec`&Bdr4VFY=k7*!! z+}sK`Cq3jRy*!0Vhd-jF&NP-tXg ztKZSqC=vhVaW*vG`9t6SX@&=Xc?_bZ2z^1t-Qbzk8=Qp3AHnTgwd z50by;Tj=L=E4jOy_ENv|BCIVOg)8mW!(`<%WXG6wIPpdbJ+V~~amCWzmHZJUIr#oaU^GbeU_!n=^PhhcB2_%G{myUxHMbACGDpJX6YBJ3 z>qJ~D+D4iujv#F+^62DOPP2tQVV`>3*wgzM;?LYtS|<-hzk zh0&AkF-~P6f6&7lHlFgKR^P7({!|rkn6nL~d#R+!J|BuWizWH?8j{!26F)MX7=ft5d(bKV?2x%SeXf$r42T$+ei3}KJn z$MNQ8h$8S zRG6ijM?D7FFu%d)@kP6X*dhKa<|QhK4d+|nWWNx0Oy5EvEgists!QZ#^&RRP;0gU{ z!X88S68_g9@N{Rp!@w#z!dfc%zmsoM{U1_zFK9Tw^wL83A z3XGPCp-9x__ z$TLazv%JTh6QriJ9L>f*hJ!*bA>XzEE6zUUrd;-fA#j_-&CQ2R6(57tRif#z&VU*(EvzCS4rh z@|~PnQ^PH@i`3-dW1KhYppXu5a?@3W(p|hr<2JI{#0ZI9z zVQr`qRz{Cz^FQy#j;poA{O?xb*+~K3wFeP{@H9NPJX_?VVMiDE&BLXO^KfyIF~H zs+h5KBu?m`kDF(9(RH4$_^L76A$#a^65HxWlh&$Xv&R@Rx@{l#WO22#XkjdM*)>iy zb%ZpLQj5eHmDR*3WdR&hl3>pUchcpF#;~_%EvQGh!|rFf^h1d?5iL&UjT=TI9W0MA zFAq}Ro4H^gq{Vmr*+Q$jD{e4pYE7cqx6RO1`2hNy*oImgk3i)bBW{9wD(79UMz?xofYr_iwQi;had~zL zb=D|D#ho+QoY)Pb-NFpCP3D(W~sV}uqt^~wSURZGJoPh(JgHWSb77Bc(M zT|~Td1Uhs@3ip$GTy^R{=-QP5(~NIYi#vxg=)ED{c1ps1x37~y+4itw=`|wzEl;$+ z&{J$-egPzB-Q||G4`wybGwI%Y$BE--3I|CcCQWl=`s**EOl2zm_&E;`?^PkeXLjM^ z4Lampq_(*7C4-QFop7|b4Ij=bCC^)e(P2wHinD)_E7l`GVel%P?Vf{+vQOaR+(zs- zkzkz8SxzBM4>SsDaQ282ET7;E0qc|Sy#IEP4rX!#@Dh+A+%6{6vISDs7 zY5|{kowUcOVd?lxa%!v<6iY24S3@R~A+fS(cTWX%eCkN(u|9sgT@1b0As-jKS{IU0I-IZRToTO zgmsfk`KR&Et~L1h9;3%)Zj;N?2Jp=u30_y2{aWS3i(Xc&A+}0kL~GCm{$Wc$o{_Jq z9X`K}q@CC;IJ8`0^{7tXuc?|Xm;Ff{a!!(;a0m+TW#fT#Bnhb}$lLIn>xd&xiCD_`Hfoz)SJJ>HcKo-sq61dzFxVm~98ZNcLmHh|d z>~0ql%4eeP)jB$M^EQ0)yI9)=`aDclt@L0q<21@Bf%k}mND8X<6yFOHZ) z+>amO_OIYLhsVE&g+vL~U6~Fa{dF&M49Zs42QV>o>iU+$nh!KNS?aCutH zZ|^H3PXrc_MBuTy$7=R;_zWi+Hl_#s9i z!t`P&J4}hZ{|kC+YdrrVN1X(F9|yrHhQsQ0VSWBPSf%TUhmT3I>oM1GNc23?prir* zQio8XEE)@CXXD*d%UR^JA~5l7CpQ_6`RxV(_ci$oLE9=l> zfueAZnF)Stl4+TWElhAf#l3kY1L{R*_%lzm1fNVSD6Lol`&MON@Lu|pR13V#`z4aN zzyF!2GpU0ctg->;bZSH9kz?dV=^?VPVJR+)KS%H1E`pEOCD|8KJKTj6oe$i;N;~`4 zfMe!u82wO@ez@n%N2y*Sa{~`hnJ;Ff(#i}5Pzo%0N}%a=r?D&ZXL26=sl=@WYsU z9DT7B(mty|@UlwouwE|Rbu5BT`zM8GqU6xn!yh6v2f*o75w2W18h4KumSbX@LW#5YiM`bOwhrd+vM0Yi=Fzh9a5B#8 zExjXj$r6@)h4Y=C$lqK;%-&Z_cMkKXDU!;A{$%1ex8ligOch#{+3i}Eu5k9z?vsoPB6Elq;W@4`Ue;sJzv z$%*H+4Umf=Z^>RQ8Q2!^5>yp3L9f4ye}4IT?M>&`bgTV0D(Tb3e-x?VE-xiEW4>)!voY(^4*>_HE00vKJ#1h$kJeO{MG(moQHU2I9^?^!AWkL8l(%Qll|(T=pPFN=(7*~`Q_ zj;#9EYC3#C$4i8_Q2=p*mpBAS4LfYX_q`)ic=JS?Ga`W z1Lm;Mppq(EHIn8id9eQ!3|qA2*p~8i(d4M3aC3=}Gks*jrtT6v9$p$^limt6|8yLS zIcFMR>qh0d)3t!zxR*&msr5rx%erQ8dna2Q249 zg24HE@&8eD9)2~xZyaxzO4>9?Xwe{}b?)nk(m+yHkv+=FH=8orrKLfOLQ_Jep`QD? zLyAO1Ss^MBrK~c_@A>@)=e*AIdd_`c*XQ$oGOWOS(f=%`Tv?rGE3gvtiOvppFMiTFIunD=6({KgUa__EF8d-Q2H9sodGcI?zf~AXDTyojs%sixmy= z#A8vc-}JBQ%j9Bu#pofaa(YA7Z`q4u@)h9n9#@R^;@ODruY*?l1PnNK4kyw71a}1; zOul|ToAb1j)aFFS_+DkOz*-$&rc!W2@{R7qDSt;fq)#FQ^R<9#T* z&?R9P+h5XyH5a=`Z?ihkL0@Y4R38K0?-NdtZG=bB!}#=A4O{p|q`Er7mo3gbMJgH!IT%dbnX#rbl6Z&Q$q^yjoJ)!>WqX~VII48td{P$BnA_*wS+2SUhMXl$5mjkuN zN2u=g8BhPVuF^ez30y8d5S*_%Lq<0mRbQJTSFPq##RmLs*}|gxc<%as=oQXn%L`fr z>Jzp@%bY2g;W$WQR)-Mx_v;zkl*IbnIeERVUvTG}Jtl8< z=l4E+c$gc&xnokun`oX3V08xlU%jO6cC9eV&<=~V%kcVqz(jdvDw|VKrM{G7U+_0B z&~B?r^2=nBmNVHzk#L@CY6bW1#(;jh4k#qc&|7UmT-Uo%xY=(IbrZJG@%_Gd#N?n* ztNS$^p64z&H*km<>CIyQec6TH$JOz8Z!E6Zun|XTPh|aCdaP!L3JcY^&(;`QLeAKU z`0$`%^?N>B){&KuLLEn3nqx<#8#BT2!()uU*#ou3eBWzi7p>>nM^Ci|1$U=93s(&H zvy}>QaQwS^wbuMHb|k9?%~cPx4JzY=E_A$bZht-QzJHO7n!1S|nK_}lPRf@F`@M{qVehC7?S-^3WNq)*wAkX!X+gnvXtL7Zoh)wFAC_N36T&v z;Urk^oen#zMXSw!$YbrllPKG2$zG3l2IqawFy+9nsuEi*s2|XQ)gH}|yG>cB8<+*F zKDsfF`^TW+;CIq_bqjmJXOEZf?-Z;Ojba|(cF<8_i|Ev&skU9Xe6!pBt1)mtpJMOL#MAG#2Zm(YWttQ7Q2vH;&Z^ik2)R z?V(cm;LSAHHSYx0I*+3*ySC8u3VF;Mdx@mK_)V8q_tRge=Lr9nj(zGFv2D0&Rd@$I+JA=`Mhc7% znSuXtW5K^#70}?!@QZ5;o&9MPv@G6?C1;GFD=mhMnBIc_wEkMR+>s?=C+2a#9&bd& z3D@YHu=jMe)d5awWHjktG8*Pe#KSwyxT+l!r-8QRCCJ!*5?u71-Y^|L11yfLwzc!$al7-P{M-cfh$Dn@utW{urZ^k`lj zr#H2M=6Vhgty>0k(Fl1gPt4<*PK$uecqdfI4}#|sXX$?x+v&L!w+7}BnV~TLp?`gQNdL$h37ssZy zmB1N~N0B2-Nb&2R1lCJHuC6+je;0xwRuV#+lOn>$U%x?0g9Gc;o`ADnkLLo`PDDAm zbiBNK6FMK@`+a90kwybO5F5v56SdxvStV0oDE5lL!aUt(`KA`qd32jczLbrvgaJwl0&R*hoWt!i}e=p?)BRhsL$N2=wcvwk9 zeG`d>*f{dbAdH*$&lInZe8}k~?!}#QVYpCa4J}&skvjYcgIz{vL9n}*Hd$H-zI~U3 zD4lv@A?FC~U&NX90|}NrG>siM+ki$fhamTPCY<`Z2hYusB$@iVpnN;Pi2@zGT^J}> zI&%wtS?7u?^_TH(;Ve30{28oV#i5zld06nIU(kGS0hxGUklJ}aq&-qoaYeH=JUyO( z#rn}WZG8g9+?Y%j+2qqxWvzfGba9%UDsF4OFG#9z1>b;?uvFz4?4CA_zEZD8>xI%n zW;j(CAGI8Ne!n4U9};Nf%@Y{odLDnVdx9?2JV^gI4*q&wK(*VYsH_nUb8S>Hx?3B< zBLd0b5=xfGFNQhBesqJ>PIQS0qN)u``F%zaY;IC!-6?BH6eo|(y}V;IY62}gTR_iP z?-fX`^#gzZ)dKr-Dfrco=d;{>Nlg9JApB7t5x>2N#7)0Rjtz=4-KH;suqOpz71d5S z)g1cGgFgq&T>x8d?Se){UAP(eo$L1E?->W=SnH-0*z6$0(4s27>yb_6$N#0NtNHhw z^F%7ZzjC;cP1+i~%LIzzIxonVT77!K_B1l5slpf&acIAI3J7>-9H@2Sjr>lu80O9{uV=HCrrUj=7H zoN$trJR0)8Wl^7MsJ~STgGVoul(2r%6w*vOuY^Mozay1N7@~b|v_ZDCfXZr|q^CD@ z(cBghbhb=@X%hU7Bz6>$^jZuCwM)?Q_ANT|(Q^7&LY&{#xs!%AGn=LI>(KnlVyxE{ z!YJu*8k)YAoRZ#;i)?)0taUW2kZ{5`SEX4S|GjpsZx$rJ|4f%1T|~ncu7UaO!By6A z*Kl%W3_echbF>GY@PsSBGcgRu&z{f7hLuw*=k7ZJyW(fyi%bjJ+_?>eTY=i`@3j&4 zw7}-&r|HtK)4|)=pn8^PC>`3Ez`dTioSI1L;qxj7nE!nrwY!uAZ|;s1{ud`jbH143 zo8N=<>C)}6;8!42|BeCmr~M@N>sky-pN)=ZoN;rj4s4!#09IeQh}yMpY2mFm7CMGjFT%NU;KKg>eN(HP*)G3;YHx2 zc$;hvb%c8|%1~!!fG_vG$MjbsL^sNsOqd(Pg~gj=B7Zm9kS*(eQP1ab`FXq20^AqjK{rd7(<;|#80mPMw@vuI= zUvNXP>CSajcr_kZBz+?)s~P%a&f<6DSE_U!l-Ue;i3Soj^!BzjRA0*%y9dP~V3Rj7 zv~~xdW4>T3N0l=#h0=C zgA1#AEQhy-vf#_Clhk2P9~~=mm2AE}4O&8z1!?q*K<+^+Ezgw29j9xl4|<{d(E*6m zcZ0D{Ka!zVC%DsDEJ&L_06lZFz+~w|kguDADX-%MPwtrrIyStdx6OlTkfJ+iAD@9& zL+;}ERju^aXmvJj_&!=bHpfH%F5!-_{iqjWFZl9z9q;GRK(sv$1Aoqg>9)Uu;+hO% zGNzO6^E1PHCeMgQ++uDob0(%qk8tn2BEgfJBJg3*niQx!!`>4^-0m0G1phtY$Xjzh zxBj?>)GVCFDy;kHCNm*Ss!I}l&pHkN3f*bwq7Wijc?~SVj(C+_0M7-nR8le+^3Bc( zjyT>V1=ALirm;b=DZR96hRho5dk{^tvNdtUB1f=WqXFMvJf){f4w2h35iF?JorR|D z!m{)e;O)ix;^yR|s+13I{^(2VijDd1@+NAwzL_%(a~FgSyu%e~s_3=37(&u6bEZYh za4_&8wR-=Sf735tOlP)m5- zX|M2^N&&5Fh@o=9YpHc@E|c1|i*0+ZjLBbwz-^lZ4#}Y~bJBK}zDbBoBAtf#$HR}F zVc1<7%Wcrtgf{1L@-8%AP&)A!Ibt7%rlt|JTRH-EZ0V-8)gsizP)QhJ5dn8jU4b>~ zL0m!qg{pWJP2732jMzH=A?GiThhy{Aap-**duu+MO>5ABmy$;0()4(mq7Vamp2bx3 zeHliLdx%>4+u2bsV-{E8gh%byjbVCd2%c&=-PH*^X^CW2}cJMARjdUZN>K+4% zL#=rh#|o+rrM8i@7cPi%ti77)!EYpRsEh6EqnfQD~>;%AovV!0p^r^UYk zjg045+UdlWUev;~wnlKpZ4MZXSqz@jJL%)Vk#zCtCHPgx6oa!i!nHk)tk^+a*yZ+* ztd)@zMp|sfXwL^&V3Jq$B>cQ!=?6!o)=w}gw+Lq?=F#)JqhNd0ce>#G8~WK=4~o_- zg16tx=ui5Tj5%q>rNpGt3^jFnztfJK?=v9EOG@$OYX>eZEF4apkjA?<_0+c~3^<)U z{Lv{*E(^2JS^5UQ-*Drm$4a7WcL=DCS_o%{lE@rwp6zr*5`GjW!2R*9r1MBTy*V(K z#9SUtyfqT(Hoa&lm_CF4k!<8_pBd6E3Pm>WtAv8XCb`7_r93#@UkB4i-NV;Q67a?+ zF|MrrhE43jEbh+LD0&&yW@qIhx5L zj(;PWUmM7v+zNiioh8N+v|2u~-{Q4!ouA(<>@3-dT}tOzcvK7?{_cZGGhE2(tYzSm zssXK0{9fsMK9OCREwFRfLZ=oHmMXpru83R&|HxRb#W04*E?N&COwXZEzmsT|nV_lQ z3w2FP1-UgzU_2oP-mc=gE~A&=my9b^@HUHl+am^6PgY_0!7tQ)xh=R{d4N~VZSmCk z9J=#L14rMvm>;9Wzi-UUNI`KPLpKM^AOODNoVgzLUL zW3g%lwJiEb_0Bk=w4N>%2HQjKH)*C}!rz^?ejw|5eKGhc@4Q;uORnBFp#M3K;kh}A zw5EJ9*Yrsqe5UHavhnL^i&z;Q(wrwW$np0(C$RJB8Q!Cio-);N6G$pS9T|-2mMNuNZ^hkOukWt zHG%i2(W^4Hf+jHS%cW%B?j0;jP)BBA_ddz$AN3imz0szs zZbb+4txhFB1ITVIV4p{q!#(#Ic=EP%waJu3_T7IqesXw1^)5%U zgEQ}AS9h+UGsc)cw^Ly4oVoC*S36nK=YlcXON54RRfWMqMcA6I#|92YLy-A$dZV-e z#_S~kfi0IX#m&XrVYQsP@O1ibyuFcU#cDO-{uRYIM*9>uH-8eVyGn6! z&S=B)uUzd*p-zxNsZlz>U8HorhPM*cVktt zIVzXg+n)(2XnH}0zO~_qBMA_$Urd9B_}T5eov5!dse0i0QMU9(80e&)Wja}-SmL8Q zc=YN6^xWx!3jd<%HSah!{OLVfM0bK6f8RZu$bsYiMc7~~!Oq=KV}Cm3gsCMdG-ReK zHl)Wu>^}vX^5-1g^7aX}T@k|eZfM35*9_e8&zD^aJjA-X-(h2sNcE|q3oM{i6N1gk zL2}?5=2~PD^QPBqyv}B3A9M&3PZl#_-afi?Wf!cHl0xfqnq*zjL{^mlj{Uxw%*qXc zot$UHF4^X>_Xu#|Ml?Kdnnv>E1Tb$)7Q4P9i|q>>f<1j#n5(KeI}&q`-mX5zbb9uJ zOP>srb!ftkc0SZlYz#Tb&lMW&Ye1zeht{5aKu)hTq7lNKP`jZDHTbUkfZ7dM^miG1 zqGrtMRpLoSlpCbx+#)V##t==5$t*B$13u?7gxk5A=9ue2d_)47$j)ds+#Ag{ zR`C7T=05r$$rS?353vLHC75w@Ej{q|F*bcSV6~$ru)pgj2t}V>!E0)-$=b|#CgTyp z>SN-VS8Xt1Q(uzTu_{7~8F4u2WdW&J69o19rn14A70jiI)VIqjkTu$CLN#L#tbEx`O z1J)?Gj=|I3R&{<}Ns{lJLuFZC@}t5UC*1si2foSTluuOjP_Q;huj$Y zy+Ik4j|u1cbT86dmCHzkPK@ByEk{9+!!u&K&49!dgpqAszi48$g}`)`81g4dXdlud zJrDfg!VyE*@!<^Ye7GHUyuX3=M`bYhunr_>pFq*$=doh0DqGrIfbV;b)4JSP;_S#f zko6)#^pz^E&|iqzt!6}-&-NRML~+Rh{O;q@2-^5M6;_4#z{5q&RQdgEo~yJTKbVQY z^9fD@^Bt)$U9t|uH%x+Att*Lsem7lrIUG(~`BG^Xhqa-Npu#(X*Tp75NL#hwTc|4y zc&U#n5w8VeCadXWJ4doTm+@YQbyPz1J%1;=OI;2wfRW-$AY;K9vNV1qzWn|ehZ6bo zz{@j&Me81sN*f*K*UCFS`Ll7}qiDFjpo?7G#&ZcTzoE%;Px&*dEA|}zO7~aJfIl@7 z5c2aiz0#+LuiOf8Bkxo_uoUq7azz?ZW5$Xbys54mpS7bqG5y&aR6V=|dt$A?qevQF zSZl)6`DUC?(0SrykO6aKZU~(92FbGeSUBRUMR%Y5#HrXP@{NJ9p9mp!lon7o#3?aRZh?-FSKi%P8jR1U7COJLQdbaLLqpS&5jn2d0pLUdLt zL)CYFkF{!u>pdEdiJ9BDaYcow++)hm_3ZIrnIcZFlE6^+K*7V!X*P=<{lr4|`}kOE zI-Aq_2-CYY@zJRXkeK|3Jc)imuP*RHtFf`TXj34TeK-#@{ch6b3r^DbF_PRxwTIBG z-bUh|w?SiVy5M(MAvL<90?Ae@@s-AUJkM#spu`piX_>h0MF;+fJOd`PeChAgr*V4R z7dkYL@7jFfdqr>3aL2fC4C?(rJ1QHv4|UN*dBhs(C8);VCfR6yDUbK^rqN?FGXxg0 z>DYZR)yC_S75@Ahfa=d{knOi-It}{l_|<9H@o6f`6jjsQSBoH|%#sKOa%jrE^R!s* z0@=F02VUE^(L)E%5*rsyxX@IBO9xcZHeQ-}+xp{Kh3{Bg5P|pdi|OAnx2VYB4p<+P z1U%_T5TsBEpNuc!-WmVU?dDYWOYfld`qgbzeOf+>>=(r#2MJdADZi42*5Uh!GigSq z8H)dn#N(=yaowMb#9NM$Jv_k7H_{Xxqjm{4icx$c8O5@5t3cQI9h~TF}lg&KcW1`9rRY0AsfeM6n1~3I9<&O(mVX& z=09s9>ugUy7_GvUn=jcU1}lMqwKAK(M++M-s^a16nQ*!NGnrS^NXM!iB{Mhuw3)-5 zq0bh?!k;}ow_*s?lO$=f9DK6(+SL+~X}Yy*7gHlk)%D8KDrr z&z|1Yl4qAaj^ohCYHShCz}CKcE;wC+dIkjB*guPd-b*{dF|i(EI`v5Plkd2si)Y9^ zwZwjbI2yqzOh4Gf93CgL>*J+_!QwHPdfXZD6hac#A3=;{OM<-`jaEA)i za9(;aTuQBks{RF-&=XD8^XHgpRyS;J|B_{SYbLNFRR?xp)Gb&V9?TxOCD6Yz>-oLy zEb26Wh`ux6_cgngf}z+d5@vUe9LO!FKRbBN0`D4loAp4jX?6(jn0ZS*$E&b*g$=B- ze}Dwcb73{*d`@n2DKQOlWKw;nIHjgAuuFdo#f$o@w#qGr$4+fH>gE_!RySgW;%nLB z{Z?33)kroO4uak46`)=cjbU0UOvc=oJd_Rxr>#M-GbRTteHLQAK#WX0`jlL_oed^r zkmfwHpgY{&bE{2i(RSnkws~J5u?;*3Hii44z#{=YPI$0j`%O^Pdkg18UBI@L`vjH| zTBtDU5Xw(E!R?#)jPxH%C2qfMNKdu4;Ge{LSp94le6pNJ$4w3gU#k+#u$6%1@>+=U z&JaNBH5ly@h`;}OL1TC=1jk$eXX$fLdo>D@yc1z&pB$>D8-PTe7({kSf`O|$h{sZ)R;Ph@)u6X4%{jXspMJO)o0=4_GH*mW<_#FmC(EXwIn@!D%>$YkCv~U zamJNgni(7qp{5gs=J)k*r=m7VTc6E$p;SrhwoEd);Wy!U+_1A)J{)wPAUx4Ji+ox? zj%Pd7kdq33an6Yytg6EVvW9}V_8-@{pD~Mt!{7OP%9B_)ESw&Y6}I5QoD+LMHe{5iE@>r}c* z_yHWo#ejy7DEs;0DaKgy_ZJO{+kDFL4nL=2=bPz=+65SYx`@B4YGLF?A3Sue8s6;M z4GB*EsGU6)53d{rsWN&b`)_mAk{$`6QAnGWI2sb{_h9e#$F}cK# z&xdZt)`B5?S0}}OxXXZH{tp;-kwmi@x>)ZU#~B-?5zpFtsH3dOH18c`7n%&%>hyc0 zD@6r8w`#*I^<6kJE)0T}QED+oUnqOx46Ir61@^2GhZAYtc-f|coM#IlbL=~^O6MqO z!3V+TVP(OtRt}VM$1<1|DF`|34{;th1aljngY55I>%^b@9#pCd_Rd%=Oxk*m=4{o3 z2dhotc)@*oaZH!s-Xd}A+_DxQT~rV%*io;@ z;=f5T)!j>Q_UZGOH2wr$FU)}I=G#!$n+lc1HDK>)#?-c7X2BBn{9Gdo=Y;= zj6cQX+t}wQ64;5Mf;@Elbq2R>7|lLjvLy4BO?cnlSNL^AOPG5vgqtMDqhha2Y&M;W z!~69g@tJfBb@n|&(|GRL2+c82G#KHWYE1` zM{ZqP3|BgW!BHoKiY6@L@~@nNweLlQ!vSK#Sz*sfNaHxv^f`hvwi{70Cu?;t%tmPF;u-3;pGCA0O z_92TDwP8wQwZY{c|9(lGL=`p}K|=ajtd5?9a<>X;-|0|L-18MX`FyV7_Kn1#t&iu0 zZxOCd$`-u7%jbhiB|sza0`pjOi!QiO3JDVx*p3Us=&3mf$~cEQM<3);X3LR1&KX$w z?UP{l^qaUbq>hPCl);^sHltMLSJZV~e_h1kG^f-l$Cm#thSEc$nDGKtw#y*`R@z^H zeP3@dYeQq9X>=YsKRrdhszwNczVs50JyWsFIUDj`y~I-wqtUCilcvPG;^lWWP^N3i z@~awPg=r>vxlNb;xZMGz+YPt_b0ko-^Q$1>m=9X;88V~SO4z+gvAg)7@9v-~bGR8e_P8(6}O{=Ff?|L#h#V+(EYbwm@08N`6Ynai+6L6e?7DbE^S@f^pp zzvM?-0am@83$a`xDmz^xKQ@=pJ7aZNp#3eB&Ab4+efJAhMLtrsdAhXRU;zCyJfNLt zX?)xwRh^(*4Fl#3OdL73ZL~Ln`xw|C;>s?z>#$?TZ$tFqNV?J>8BD&avT@_LpoH37 za%lf-oV{IGt-kdo6^87lUDZ?AyMG5@ovt0(t0fdplk|j-Z8{*n<`S-dy%Qz+^LY>0 zUwT-O#p&&DMD3Q{&@N_6)Iq{RvcJINdUJn6kF-N4&T^D!22!9 z_mg7ROe~^kYUq9fcf9J_hiM{xr(|UY)*pkH; zTR>09BH(IM=ym>XpLXUqn|^%^PH*{1i!3uRW!xDOXV46NUp1Lt;4j?r{0DxHP37;i z7hu^7Nj9)-A-Mfh!niknu%Prf8l1_aCL>gYXXy(Fu8hKZ7Q*$oRiN>Ka+~Elq*=>@ zQOwXb0~>rONbJ}KQ%5$!)y5?>+9n96y9d*siJz@*AI!4*_fw0f zhiJR>W%fj%%~lSpfao_BxG*`Njgr@6hib<%PV*Q%sO5yulDmY)>qoGJ{AL{YI)qqg zCE<#9;kb748oJpFamLOH5;d+CE*~?6u`z0JW9>t(vr-B*E?44S<2p#ZyN_9f7r-l1 z2|WKzL^y-G!R)#sykFxiQ1ca%?_b8^rW|E<<9-cB%ogOPIN;1l`pjaNv@mD0HYT)HGMW97SkodV{9wI<`PGW?z2HFD zboK*!eV7PdbK8iC{c}urlM?FMNE5xI^7rtpviY~Meq0Z1 zKVS@zPx)DF`V-P|%2P03zmWIFibMIdCYT`UBSS8RwHX8Dr#j!n2Sr|2Z%_d#+G?tEUg$t)tgqD1lsoTw#&Mnu47LyKI zyhj$%U4VW`&#CT3~2tib#(gr}{CTO+mcUpLPr z{Mw9Tep}(KUt5@-RU3_|;Mon``RwH1BAgzPgioJ}2(`s_;#|Hr;;>x?$n`4_|LizS zdTao2>=cTZ0glq@pqD-o+!u0#t_$o(heP7RvT!xw=}{)k#px^>UHrs-(H?`ZbNPl z2x5<}mSJ`c#f3|Fmd`!-O=prV z@?gO;(&;gGrnKW2HVB5U4}KNTM?dlz zM4!)uvMH6wS!htTBi~4dl`fmMN`p;n)8Y;$C$cxwr!p4>LqYcPUbKIi$dnZMXRY6t zy*ANeO(quTbMq9toB9H0xAF6o-$Hhfe^%36y9KT$>u{286m6T_BS`rEADQ>a*v9NQ z@3x|ze5UUu|62{ve6L)3*Rq(Je~o~p)R=v}rNLt7?!vvXYf+;;ACG&r)AnZ%>B_De zILG?~_KYqha_@e?UBN9N{X-<6iYE#^{Y!flj^YBp8)QN1P7ItcM%>2@kt?I^QOWfP ze9PQVE`9gJ1I4qchu=!}ZMg$nzCH;XlI_5)BcCoFGoDks{t+ka)Weq(8_^^_2lqrG z`svQ5onOy_!ka-AL|g{U^8;F2=6@*MVLtP8e@`1WwzJ5PF=7hr4Qz zFl&Y{W;iJ_xl1be!tE|z-PlH7jaUd}Cl!gi+){`xyo}vru2|1y7htN>7QDSGhKAnU zM%LW*z{f*@c+UJ0j6D#KL96C5C1-bB6Y!YIe$;{!L!;P0{5$UT;)z&QFco@MN@JXq zqi~U!EH1rli;qKw;Jt1EXi4g@oJ^xF!%?{=2W6jmjXYZ6UB%1Qta9rQ#yW06@BfNNan@X!Aq5? z@VGykzMR!U&_999Yfr;&&9Q>%!}XZEq8yutKjHZR0dpHppitq;O!(h1anb>Jysz3O zUh@n@Cd#s}-xT@&(6&mgyM5$b);%&y&YJokmV}wT8f2E)C4L^hojsitf@8OS#BCPN zbnV<3EM-Cmb`+iCWP0Ptk{vH`cjGy@5Z*`*eSb#p4K|_6zz=ME^oQT`d?5d49nOny z<9b%L@$8Occu(AhWv;KH357){b9f3mM~2}O6^2`OwUUMb17^eX=JwllQO93>v|)b} zXbt3mXyasp&oc0C)V_};gd`_iZ}+q+U0Lh zeN;WqEG-u#@G}K1-&ss&t5PQ_dCDplKME%Chx zpXZQ02Mw!F5rukDT>0`cgnQnGNv+4}nd{4x` zLIw6usDTNaf00dFZontkahT;kiX8WLp!r&^^wXsX*rj>}3`g-yms`7RT%7D|{@Hbr zz|ljX8_MS_XF7sM-7vkY51kR|5gWtx@!iV3? zg|fBXK)&13!7GY*l&2#bpSmCjzpu>I+Fv9cS4?qDnK_PCAHeK}5L)~C0rA`#Pty74 zc*TSys#o2_d8`Z}$N!Gw?;{(~QT;b&iK|iDWvY1p<0%+_Z8x;agaW6mNM6>RrUxpI zV``5DdTu#Q6nZZbv%iyYq55Ci;g|$QiYsZzqtUczhC8WRR?Q`=^Pb0?CgQqa8aC8* zk&&6tA(rp>HuM*fPb~uI3;2NvRqn#y1>d2t@(TH~nF^Xzoxx_$O-{x+1tz~zBo7oC zFyvrrWsaRHRk-_!wB0{lSvN-o7JsoMcb~Y?Eq;9VCJL~+aV?i=U_!^Q)#H3J-rM~B zv6PFGx5ovW|KY9l4veYqrqlV(U7G%6;h6YXFzY`Bj{ChyuXBXp+FMJq%lRYuyhec* zAN*re{8x!CIadN71FVU5jXky5-a#8&X7D~)9O_%-Z@MaAqW)S+Km6XbvK%Tg(O^8{v z;M(}bxc%uecy?M3w2l1O(aHLB?2&HnW?c|n=@|yvPZOx25f6$pd0*Ck=z+tOdjG z)?q|Y6ZS(ESr<2frp%Q`vmT(2OTN=O=L8y9oW z8m4c=+p1eYptTQMEVE(o!wK%n#-}vou^OLCuB!@iSq|;*#tJ+>gwpGG9B7OA2|DVn zIw>|E$4TTVPF0mYR31yhTAb+D8;dwei#>up zkG!~t+1KdUJPq{j6$KOMq<4Or!M79>x}|Lo896Qt)I$zo-SBrvzMBghU9XZTMR688 zLEB2IUWw0XhtZ!IBRCKKT^^RAK%R*mgD8tyZcN^E%yT?Xygr6wLgq8NCSpDPD{T!9 z7m}HLWCmHY$Pb2LIQn&{)UX_`BwK%O!k<3{n1O9v;VUffZvL|srQ_{1= zp;==|ig7R2Cvt32VJI4$9?gh;0#>gI9H4ih-H0(9?;LeI>teRP1_Yl-xx=F8Yd(4$fjnjitZ{RD^Q=qoGUp4Ou;2 zimJ8r<9}ye;ogiVbn^5PI_l$dvXOTzEE=mtBQG<-lJ(W(#>`xN{LUMm2izyRCn`w4 zf)rlVRS>TDy%4AVP~`qD{Yx{CorhmqS$Mg|kBs*rg7~@FykB!9bXNt@iw}Nc?tw(~ z50Ar0j~UEI+5?heR}pXd2~1Ei8OO}@X0n%zh~n8T;5%y!yKq2>=(yge+M{Z5{>pOD z+nvgNj{8Oe9*D95J2@fm3&u+KEPvy@Jt9_Zxz~Dq? zDx!{0DpYY-w>kKo%i(4`Cb%PC0p07Kv(C4C9yV$WEUEfL`VDGf=A6acjXDXII#FEM zW)ue>z5c<@pw&F1^D;S`xrNlM5XHZ$JVWR(M?U%dvk{dU2b;^S*%1+GHo0aH&$#xI z?QiEn{IMjk(z}bM&F4YXrWqrzegO88araB`s;m=JdJ&3$k9Ohi%?;SAzZ1SvwewdZ0JdvGT9v3 zJ?^0G$)oTtJ`sNy$O!ApHSqDxgLqu2f{edzC$N?aWFH1pSTWx%|5W10oS)r>M1HpR zvi=ba%yh);sL8@`DG~ZGjH9cz8N-X@qqKheL&3=h>D;C;A9kwtGnw!516nHhZ#a|h z9NbdkHrkC~ug5mh)lC}M=N^u$#{B`~2yL==a5jwd-454NN=RtcboO9D0gNf;<3G!+ zp;$Tz-7G}0cWx?%>W&m%?{s0Z^Z5?`+jjzeu{go?{mQtZNtYb>t3oS&z2&YS7{Kih zOX`8_Z(k+ zJOME_8fY)N5#%EHJU~D1XWQmSFH7$Lv#;9FVQz^>hnLd@xBn6mTPsl9u^+aLpT%eD zJ;~u~XK-kD1Qs^kgd3jP#QR|-E&Opu@Y_F-Mb`8Fn*}v6yh0y~%{P&YKN^VG$JKC2 z!jrV{+`Mps6Z*VAN{>i85bu_3h)FRe3dbIh(8L9JJj@)k11Z*IABMZf4q=VGDZQ~_ z1jNsb<9sV7p~ID3@NmiwoA;(Kx&A+e;0(d6FbJk*t0OwxQ! zP^B(}bFf!M&uen%wR9w|KRp>ce=nkLo2M|#iYL@+`8!y!Rhm6>o59w9Fl3gvf{}S6 z*_O?^?6$!p{2{gJ`V066F(21JQ^+G5&HlYqGBJ~WbVgDU-BM))cj)Y&F8J6lf%dKS zq!!PTIsYGb1->@nJ9i#@Lt=A9G0&Xj8?3txl&oA`+*ag|AgPG%~%O1_3Lq{ zErDjK6;em9Ff!OAgE<$Ulb0FyZElWN!h2^e@J3&P;4r^Y+pDGonf2ymq*E+iGA0P! zCp;qWQdHpC?JzntDU>dhdqd8RGGclS_H6m^Q?ho$C|L8Wn`qAF?@4@i&p$n!nvb?X z>3{0zRjY#Ie@o+N=>Wlv!wr-xUPu3$PiB%obuqgw8DDMfqMsUnlGa#5+BS0|=T%ch z4is#p>xT6pLGv9gjQU9{O;bQ|e~{pj&1#GaRl~JCf5`tRIuC!YzCVr|p$J*gAdx7k zlv3RD{-{(&Mky49(v+gDY44Fyb^}o|GD6(*J`Jf=v6sqvtizp`{1vljQF)074K^@bAiF?B6TO-kn0-b-a-KW3x(j_XirEKN*07NOQ@ zCT!W?(#t`2$mQ-gwH*_#qWky(zV-NB*xwMsw!PEClUHS-q%VsUOxgt%O9}J(GmFF( zB~r~{(dg?Qh5xD+(pGCZqO`t(KG;8*u28+joe^Ka+o-0}v{+}bIARRrHVC=;DQD@o zvCnDHUxB^4yqWYXr{JE|@Ay5BM!=g%&TL6wJ3hO;flD8u1fMTmV0yh}w7&8ZmOu{qN8%>UC%rq$=(UT7A?T(Td)m(E>bB81 zdTAwI%a&msKNrE%9}np)pGcVa{3ym1t%Ad{-^tclf=g#;Ha(!Q95(wR@lRh${#I7- zgSQ{!+T9JRb0XP2g%Y`gp?&{gFN3@Qb&kF;V5yV%Y6;4v%h+hE<`}FsqE{3!6~8hVM?f=?Lcbf4OX8I$RucWB+nP&D!TbE&7}DX_2}vhBJSd+jGLPy53C_oWx}5Onb3{v`^H@}J%q8&2Bh4j0_P7G$ImT+RQ}R2D4DH@H5n;XHD4EA zK3k2lr8>CUNATepTM#RiQ0`UZ6c%cD2Ikj)B!@Swz;JKGsY`{9v%p)*JLSi!U;&dG zla0430kx82_&nJW#PwV;F-`e^H4AeHo%VI^k0)KQP+pInmGy(;OE=NsbBsZ6k_&#W zE(Tx0OJsYXiB@%tt{r+N5yhpm@phXn{20E5FG+oeqm_Q)=l{H^^@AJqU;L2TODgxt z51k*R?(TWmT5pH$9g0Y`Rl$4HCYae~g84l$&}Y_*;k!0r%aj+~-?UJ?Hhw-%o!!W- z#WB2x#xbHY#uRo$%fK;DRb;V6c*<%C>@@j7bjB{mw%9VVws02n*Z)m#L@a_qoDF5K zBVoX2E7Lc$F)vNBrflyJjCs>Rv-hWQGfpjoN9Xsk&6zD2F}{hcnH$ARdgj6p)hg1+ z`}3WL7vQMN17w)NVk|n8MfQ7?z~!P*EMnI*x_a|iJmhGG3tGha=1n_^mP84z`Bjeg zHM6LOrWxuTG#B_2lunI#U`|BJ>=A#PYdh^P%x4Mf_e^6qi$>t^x6S06e*!eF&i?k>LudM$34x_>+8sOJC&BBroK?}XmI<94dA)xhfv z8&8bu*J9Y6DX2zL@YcH!nm_*`b^o`J%Uwz!c)7q^^LL^}DNEqM4>R&KrvwZCt|YnL zhA_={7B(Et!R8;Yc{qFr96k;c)eMYbMWQ%5xojF3{~3bM2Ufw3QI$kvWd^v)2+Z}& zm$>b3C)Mjpf*rL{a3I|u{~a@@=5#4Z?lZe4v4B zG%*Kfn;Z0xoDC>Mdc$sC6Et%zLwUEM>;`>ES)c|?+W3*HS>D0VUn^v<{DV>Jld9Q? z18KObsh`X^Q_1sYLHJKno!w|QW~c0<;qBI9wm-bv+&XrZV34|kCl1K53+t0GBAw1x zv<+uF&iux+zg$VlhYf7VaxGT=?K}0lmd7O91(zX_r@KFfvyoY*sQq>o3fu0O&4evQJp78R(pDUn^Bmx9&l)eyJa?y|{kp5^_H z$HQH-Sj9&fw)?jOUU2qhV{;SO--o`~s~?Z;aqaww(j0vH#0yt?{fFN+O0%-NJK4;D zRNS4|E^s7$MJeHu)H>A}G~JR}(2p3w;W!Sjq9n1B z>hDIq=4yPf^ADZm>d7p(++ZX7W!aaExhz3r5RZrPR55lg+utp4Tql;%VAcsQz2{u^ z9V>O&+0seW6(Zzqu7={MA=k;~d@=S;)P>(CJcRCZc9&foW7w10F}SYj6SF&*%OW#% z@pxMq4yYbN8PCNm)ObEDxh8?WHi2M$Z#H>n;l`}kdC{*4!EoipI=cIm5qv2rWVS*c ze)JMaEJ>CU^1MB`vwt73&@y!Vv53_jawB5rzOZkLO4*^c$^t7x`SPa9QB27sohY9f zb~!=mI$OMC0>+=qV`GMBqqKtpvpX=1J<1dpD0>^h+VU|wy2X+?Cbg0gw+t>%@>64v zf{bDH2+GQ$jx+b?@%;W`2jM%oo2gq&CjvQ_)uaeMvzm7x7Zf0plbVh(Mpdx?el{Ca z9DeyJYlAU1N8rd01(sacjVES?GEdY&hhfcRyPY@t6`u)Y7l=}Ge^U=F75cvF7sPm` zVatL0__a5H&oSH#H43q~#wmi+J=M)_^jEQ6+bxLG=;O@Vt%99+cK}ZgwSfb}4|+-_Kv5Q>SxS>(gZ z3Z|fLhk7RtFtbiERGj=06)vfZK0J|R;%q*<6&s0%)&B5@?H)r~h;ZK%g|KPDnQ^js zC>hyQ%XSG}DI3=$k;mXZXt(@D)gLtT!{6tzrcGy9TATw*94c^&T4h+*?dC%i`(T309eUYkKis!D!#wxNh(g_82>zsPtT%fpOFFTY zcRewxfFJ7q1q3Q@(Q&HSiEfrw!aIcfA$}Nf}=j9rhN)n zNzcKY^EJd*KbmmWXUWV@7Wl-l5Ej(F!6B7_R7Q1xf7WRLwjb}(8Ao#X0rN><=jY8; ztegUM%ch`BMJaI`X$ep|o=A+l&sY3cP2$Qs$lZH++?*fs+jkabYUuan|X$g+bn~{QgO8Z=O?)MR9>{|tD{KCGmPHfSV}LfiK5OQ6i_!!Pk2t3 zqGg5Pjqmp18hh`WCq>A!rIs@6y7hH3GF_OJs|~|}1HJU2c>wO|{Kq|+Xh`jXRUzm1 zY%DCX#YuUy$PWUlD}I%m&VGS6Tc_{};=7oT_v*OgU>`^OCO`IL}PZuBSKO3!$m z!$-i?xs~kDy+O1C<>-UgPH<@MBszP)4sM*4N=6ACf{zucbmE{ZSHDagP6!>^6Iu@N zV&h7B^7A1W8~f4BQsWX`o7l=-lc=N9G$uk`9#0lZE8~QqG??9AP2}Bmu=!pndTBW=nfnHw9=eKdgz|~ zm^fFR;vQLPpxcvK=7nQr;FI?Xvc=q!=FPGqLxWTCs+k!+?B$3?kuvSo6V-fEQGl_V z{pjzOXH;jj2%;Jl;d*%~L|zqq2vwzY;62aT*6~rS?#E?ReV7d1`;O6}a&xHvfdJSsN1CLJmBb9M@$AaF_ozh?r+lTPeX}EM zelQtE$ezGr+W^c>@Io?vDjz?;jqEkp1|MpNz}{U2^xNzsx<>Cg*9Vt~hUus`ePbNJU!kOKuHMYL zG^3^Psb?$kjNU+e>p2+dyq`K;yn?wGE@FwiF6^A~otADYgK^w&>=&Nh+V3g!%$t>L z?v-~a<IQ(RMJ$7I(JlEzqR8W1&$Sk0XxnkCHIN0s@( ziGi=Q_w!GB`db+~|0*G>6aFKCzKbDkodK-fB@eTvWpHz)Kf>UeaF~;34<4U{zv+SC zSatJ;Q9D+n&+$9xF~b=B_udj1<(G)T3oV!~mCT(UD5$%ZF&&3AnZfFSDWbLuuLyly>A$zw~*B^Ng)~vt+YH_nr(kGo(=VvWdr9d>Z&qZ;j(iI z9O&*9c3cm+Lzj)F<_)Gd+V02o|vw_ZjUkr=Q2A{QwTw9$U z?$b`ey9d|8-$TIneMm@w3pGcK*??q5=bGkS?Q zZAbyMB}bBTxhbUO$RybFHkD{y^rrL6)j*|m3w{`7k2fz#i1O>bNv7KgI5#th|15o- z{2NkCfkf#Y6yeHaI8d=lZ3^Hj8$Q(pTvWE^q5 zqXI$2+U&%3Q?^x@dxU5f5f$ClF!w+rky-8o&ugD?q25Cwo|nbEC*$yOy*{eEPDanS zjyU75Tbl>tXi% zN3`59iuQc>z%fQnEa~q~_FH@|UX4h{nEUhKUBgv+=Xo1Bb9D(uRY~Infu()QqKB*k zah!AiD2%YIrn)V5^udN1&@gH~m<$}@-&MuZG969PpVdc+(%nk(sC6>+I^{x_tzQY| zPu9T(OMm!tdL+i?>^5KD69sW0dV&+qoc+A%343oGLyyfn@cZ~=A#3A9t`_JL-}q*D zYPOHMTvA2nJMLKeWC<2d6~WT#C|Z5Gh9>`%LFb{CD0Vmma)+gpYepWhNI~c+WxlWV z8?&Zv*hOz}Fj51(iM9Aa>JoV{vY%8XOvB$>Tj-DcbO@a=jx8S4#d*8zSb^>u)_J}a z?9wIS$Kp$TvDgvVmMb{EqIZ)O{1{PT#1}qy-(Kt+{SnW&_+b+}hxdoBK?BLzqC@Q> zrY9LmHy)yERwvNv7YACk0D;}wRlG03XqCDHJTlln1@^3i%4g~0;1D5i1Sv;v!lmvW5ixFoN6UYWQZr3 z_AC(YcJ-LE-^g;Sv;P@(k{ zF%eifuS~V!_xgTrplc@4KX!wtj&dSXrWk|wMhW67Esn7uOTMth?3YV4R;9L*@DUxf zWbssTKK2YPC?1K$R!wzBh4irXogTC(nT(Qih5x7P3ie@s4E4@^O;k?a=dQjwMn6ah znB;TKIM1QPT(#{6hP=pv@lA3dp|_OY4^XC#I@fUN;$^tP2n3&SA!qXl$#M%|W9O<;V%18@J_Nl&`f zfGL-QMfwgXCmD$&po=bC-;cJcQ|LwCY*=J-oqY3=g1P+9bX_oB8mf-WHFlO;CnEk3B z#*A*sv8$Jxxm5otu&Qty5m%f5rf;I^8t)uqV^R`?bInf_k0WH}8yna#>KaYUSWX_T zmY`$qc@VQdC73?HfJ)hHrrj^>k*m|e?CGEBZkB?>qAKZ&ed!SJPMxd@(`5Yu4?^UV z0(0(fz=G~+;91;2etM0=TawNA;8{7)Q9=iNnG>GQO(pyFm(b<5+wijQ3Oth5S+~~0 zfnGaUMwd;J0D8R)Jk2aHR^E%fKRtmBnf!`$?0UezUXn{4nl99q*ZGiRm)zLypy7C} zG84}|jK@!pMWpA&9>L}Fk!D%XXA2+RLTMiHSo0C8@Arxn{JIXG^~I?Cwhxp9<#1xX zu~et{HVIat^ud8G{BG}S)Niz#x!Rl#nlZ+jt9w=p7AXcKbD0PZ4Gp9NGxXubE_=@Y zvl)N;&tf=XU_^c&wW5XGG`1N9mT!a%`8YU+>iv>|`I9v0^dm{cT`7=!3z&{-RG0V| zm(k*9hHyJMfIPWgU>-HNo!b5SKpz-hA|~t#*;le114k?cOQ{(8?%f?E%XDzW$fYbb zIGv7}Jq52kJjmJCq+*w5A)MOQOf({o)eV2&MD}`k!?4ju;MCLOkUgRrmdr6HkDFZJ z_Ob1_X!A;T#rg}CR?kCj?i3h^9}jnLw%4*?>$V|qe$2ptZ$duk-F4QZFcFyzVqPAb; zeb(&dE?b+4=3VAQ8*cxDUvC46k^e*#n;FVgZwST(d^0^>+qKoqkk(JhU82;Xp%03psS!r!>?j>;H4C5_F?k3wMWnq|AK0Rgkin!i> z!&m$p0lywwz$59m{F}H&94XBHyp-GN5(8-h=ls%QYlTQ2h}{4#+$m5`%pJ&^k| z9e;=((6a{rxNiNDx`??C=%l}KeB9|3aNvt4%5Bi3-saWhOxnZdrdrLO|f)AD)H~Kr1LJ!qFEUsypJdd8~cm#thz7n|I-hBs%z-tP2P1I zLbafV4i%;JN}zak5FS~oM{+;xfhEsENR;4G+pn&G_isMr7H4kYlCEyVaq6EiPvRF^ z@}s!ZRR$z3y^(x5IYQ`Sx#7^)w)j<5josv~Vu>BP|65vxhIj3ewx6G8m>r*&uliDBb`#n==2!MrLuh%VnLkFr-w>AfW_ z-0_tCr2CpFYHs>&Zn$O;&d&c#Z@tu@-yBRx0z1nsFcv!S^&?0_^Gou$d^~fJ5NFy4 zKhj7eTL{;mLexiw(2b}AeUc3l}F0~r^L-!W>u}6mu{db#An>C2vH{Pd< z_2M{d787jQMmaX>P{- zd@?Y1htTy1qn8n&+pUEFcz$r))Ks4h@CT~$j+G`NZ6Q2`d)uIEgh;)4{nPk8e!oOWG%up zk4QM{`kq_9(GboL-HcH}uxUH@y?Q)(u>SkRL0!ygf9aBUBut7jlx z;rf)@P&I~lEg^Cvmin4xsI&lL;}zR3-=ZRe^NPQ;zjfo3oFKBhel zx;WxR9{KY68~lH8PvK7zGy*y`;BgazEdtVZHv9y&pOim**YEMZ^NheWVr$}~5{z`BfP?vnd1 zs;?>RAvT2Y(S~8v^wdZpyDUNF)1Q+5DcvMID;ga)AU(vb<|Q}B6Yg;hE?6AINqE$Q z>BAn@mw_lwU6$QkO+_x-_+}<+>pYEvVOg?kyGw&!h znv_9TB_{C3|Ln{P-mJp#JZW}mL^gKPi$v=1St@63&#oR=3}JnTNr>lJT2Z7;n}c8A zPVY6;ketW*%wGOtZxd{g%p`cHiXLftL$?()Ky-yEw70e348t<0S=&sUB62aWMoaV} z;4e37_YAhW{Xd+N-Dg{Pah83!h zeDT#_I#6{4E!t|x5!=ns(lrUP!@hF{UGeA)ZE!Mb4p*|2=TgoM5`~4%v`TFZwX>7P zS51A?t+$RyixlB$fUUVhOcZWE(MG(dmJtP=F)+{W7~L@ODfRm!119EL=znq}u2Q~- z|IIu}ZXNi@zx!m;IaB3eW%Cfx?+>bAXrBeH!S#4{{#EW>gNUp$ZNQ9M@XiJ$r(N*I1e)vj<5k1_(g=ktk&q~Jfw7{}h)OR*cjhp;;BQVh?M6|E2d z58eb`5g0~MP~B5PCS=NkO_yz*Zu>|6lam2kv3e%$9y_1ecF2LL(KcKd7ij*wiz5N?n9V?;G#hN1POCHOrkgo~)3fm^325{KR2$<3urxQ(mEX?4j+b{UY?|9;U4 z#)u1%`@ z_iY*3`6~qMpKCESAuAc1mq6Z5ieleeve}VepQ%H`ayEJ15;oso9$r0vO*>vv>b^df z=Fi>CKU40ecc+Gtj7m$$scYmmJYr-?s|yY(nSs^9j$qv7crs?Gq-fa;Es=-VCtlNJ zF=!Z=!^|9c21&zNuw5yQPf_NQho~~`Xc-*zp8(f)+rc|U8ThU!>}R2VQx?ks=4UY9CD>T|Pz9UoL`#$BJB1ToeuGhC|-hVA#G| z1I9d3p~;!s;K$Vo5abgKIVt^ogT!R;9#LPrKR^mM82+R!ng2+7n;7RWZ$+$hgq-B* ze7F$cz!ip1BGfW(_Z4}OrAMH}POggk!Lm@ylMah;=xX`K{k)hXoDEDw-`^Ao ztl3}ayIqDgdkquS1}uP{vB&G!MLFTymJI{R;;`;$1a3XlMk7xq;Kb4`;O2Ofu9#*H zVLNK!nwkcNZmFgl&rN49L)7TXTfu0JNAY|5Hd+ur6o&fHL}@5Su<6_w07U_Iw73nrC>&VbXlxljpRE zjwEjvjiftwW%D2B)xfd$GJMmx1ynac1rCKT1)WqgoZhdGvsVveb4K^ln4vM)&(-kM z0N(##h1p*MLiUI<(T;gZTsqgaW>RWD*gS2!`BGjFNe`$YP_*H~n`>S0~9^L(aq z&6f2K4TB?Lk7&_4Z{k=JhHpL+2?hS6R*%qZ&^$5K(HEcL%0#oe2 z@in$P(N;Yd#P{Xnv(b`F>Xf6G z(ZcN{mNZ@NB7g2GvcHk*$=><;>{#t{ZpF6A;5XwXoj)lL^PPL~r(Qm_G1Z{gvoi77 zjA=ytjym4D)B)sBHqGs>p;N_=qe4zN7)UL{PGO(g_n9y!C1s{1+%b;^zDM_(OpI~X zK{J72FrdAJO!?%>5?<|QhSlk~;=ov@d7vIfmUrSr|G--R9EKS`EXmyshbdF~46Rku zu|HpjJ-_}E-*;iy%`^!IOZ{Pc0?E;AliYU#5TO&8I`&XRmeT!B9&HCV;Jiy#%a zkFH|vFvY_ITepuCEz8)#QmPrz{xY1nC+jn9Ngd)dM4!9$pV0Lvmlln53x`#8VW8M3 zFG`p-0Q>J;!;D@Ve%yKq%&@pgijJ?sk%@H}J6jHS{~ZQTPM6}t8AogLR`CLRtdgs( z`GXa4TVaK(HdFufnxE%zmuv}rSsM?fG*$gO@fKSO$2VKz?er3IKKT$m^5Gf8%J$NM zgMJ`a_W}pTtQ7h;cJOEWcVac}6jbOM5Vtk?m8NL^jXPFy*;~tfI`fL1YcxJkr zS)4eDZrou^dff`nvP+njw-Z&%+=AmEHQ;hhFr_4o!A|Pl1d8IaIoqflHnR!R^q!u*K~igq@K?lFj4U2p*Mw zo#NhyPQppIreJWV;Id&C3Uqu81I6%wu_AsqPtwWE&0R|A1ToqiCz*#bPW;{wX?m6&!jmo9s32KE=4 zAhHxe<=k~5T6z^-6Fs44s|+l;zlAP1+a;V`+=V&%T`c)Li|xPv22I`Wk_(e{sDz6? zsQ9RWO8-#Udu$xj8$N($Lx!Q_$}p}yWF=AZy+9X?en+Csq~LSlMA4BKL2Plk&|NnZ z?)}pR-sHw1VE*h1pP3{E@q~Oa#B&qXZ4O|*K{(>Z%NOkRtMyS zOHzE@mB1JDGOHn5M#q5d>jyOL)NM>NVz_;JBbF|b5SdK7!R@hl0-qe$P)YkZ?#2WM z3=;OtJ2oeQ`bHs-`*#KOJQNpcpH_v8e+tDddqfNZf-^ufX%WPZ@1t+koJ4I^QJB1_8>hBWdak>gaxDwM zrPGp5^Z}A;@|j-W31>7r((M5&1V|sQ^di*9OS%5(6Iq) zadc)OY&gFVECV7**Ipwbx2cPTdEfA<>S(r4J%&GMiFD)hR*;h~1pVjIbnL(>EPwHr z>V_?1o)#L&yF`;k4_^}7Zu0N;I^w+gJUJDk1pRuob(jA;QUjO zmA$cKtyx7_A!JzhIp~AIULH1|>xaqdi6CZMg|{P2aP$6Joa>bVqnvs$d0qhyq5Wi@ zbRt=BXDNTx>`R?P!8*<%N*&#*MBw@}AIrbUv8|V^p#Q{Ab0XxQAD9)8uHq2vJXmW3v=2EAk^uq;Q4XHqysh7?xqKBD)>wn$|PaBli++RoD8>D zsgO1OlVP6PemeDyJhBMk{2bK7#VRRs>k2hkr;ivNrW;4~?+LujN0In?Zw|_iF2e2a zbn(vN5fJDXL&1D9*)m>@guD`1Be4~nrfdRiJ0F18X%=2Wf~V6kxS%Jz@LZJo1NnC(fd@rt9DTAz|Omuili-^jh47QoL7@W*`D zQ)E+G2RWA3jN8^pfiRhZETz%x*hEiUb?q(9j`+Z*^+a<;PZdB`F$&+iBx0DC8ImM> zCLSin{*7oMdp{o#oQ5gnddFe%rg}GxA6vj*vpvhF%=&~{X8{4E;( zJcp+I&gaU0WD>9IIq>y-FoqS#v-*k_T;Xd2vlfqGUjvfZ$SL#SaNBcYw=4(~Cmf?e zuP+l}e_h_SgLEbv+cQ2}vee9IhSuxTZ% zwoJgAmIa(&&~P}t3S@DcEG(MtD zv0*Tz+kzjxvx1xmmVm?!hTJ+)HF?z_B?<|hjdwjm=&FGp;;NX-KMBd8Z>185@1#$d zvhxLM4z@-wN0LN9GN=c^S8BR<5_l6)1& z+%z5bMtmgqE<0kfe-G{3RE6aznl-E&7HB20`1ZSEG;)>!;)vR zOB)XSkj!ZRq68gPh6QtR=8MN&_?H(}?Th5G+%$zX^g5zuXB&`CE>`QwUCpnz%&jHWn*#| z!mR=crof*@MURIl{;dU%mr0;azcH9VF8S{Ag!T^3!#-RLI~GZDMdluqo7RqFH*15k zvjc9`(O});A901AF?-Z3SG#%63W4{XKz!n6A(0EmeVf*x_CPdl*}s@-f#Az?mO(qq zd?dXc)K6C%QzuA6h)J~YJ|@r-)>;?-=}p~`ykJ?Na@!7JKplScWeSo+Z%KSX-6ms>_-p<6YY2>$Bu*7Ybvbuqs@AKkm7X@adg zHTm!vTN5;3ovsXyy4{DFn>_LLpc?-}O`MeXPiDbmPxEx!agk-O?_AG zM)khUuqWda3p0DbhP{4GI^W#Fr~R|AQrJOz?R`UaZVl3c_1;)D#SE_AK1dFIDB-RM zE_a2^lc{?}IMS7S@Yvj0c*~3@Qnwx9=5J$ItYHH){4db>7+L)AYXXixF$fbSuYu3K zg`g7ofmH3^KyI%ZgFEW)(q{sK;6lA5${CpAX{P|1ut*t)Msjp~=5{XGHw_#u1b=7R zD}rY6b^4c8@Wfs#HrH66$mPt$J?kYQ;CKjpGP_9&_w*CnXi=A|o+&(y;^oVmA zeg?&6EM(&^pQK+;p2uB_W;5^GBiZE+Pxzsh08KaF5Upi}Xw`6n{BAx+E`QR%KwEDX z=O_;MHw>WfxQ+N8wxeT1AI-1or*m~%NQL)hGC?N&($$nQI$UB5tCTGz0~?#jan0*= ztZX>nGf$Co3YLUymO*+O7QjE#I2>MfnXGF+1?gMU*u&ko z{j~7o-BS|D+UrN*_UsHH)4qd^D3*oXcV}tS&ZTrx&TMvVNEmDI`b{ry&%>$1axhf9 zi4KU3g_O1V5Hsiud(~oTXolcO3Me4gdQI4qp|c^$KH#nnskx%iMauIkA=^?GLrN!-rF)_|6-Qebf5jHnOXJxx$K&|r z%wPVReJUP;vf7C{S4hPLAHjEb7xPmsnEiW6w#I?tT% zCgoGP;byQgt&eopG}2!S?h?sEsoJmM%{wgVYy_7UKpQLBUvqSJ@2qT=EW{DPt` ztP~sszqS}qi)A;En=1HakH%4pBju% zxS!XAsqRmqQE-ZxzlvesHvJ%Z-5zAe3SG7*`2%Kz)MCZpCYA|)SRC~Nt?4fsQ2qc; z)hrYo!UsvPn;&!Cs>*Aasu9NrQ$hdqPx!X54MKJtfRdRJA|vfBsAwe37QlIuYQB}3 zDhq0@6I$f`V<*z{pF0aKZJ?(dwN zll+Q)&yT@719Qm6U2D-M?4aPI3BkJxku+2L3)vf!31MaD@Zz-wc;Z$BrK={e14EkW zx{p(d(&{sCBRGjH6n>{LNqtt9@dN$1LYy>pELEHDh^4i!XvtMmm=>6UyM;{l=mpxi zx%wl0pI=Y4pNB!GX$TAA7k%&WSQ=D2T*r$1mo6Cy!i2xgrq#d#&>@(ccTZD zs9Vbhn=|SzG%tbBwQtd2RSO^F;6u8cm05+*6RkJeg?m@F!RBoekSR4DU%9(u{OAZA zU)DuxD?s$&)JXUwII50zo3Oo!3}uh_qt)|s?2G<0mR=Xf2X-~`r*!KX4me@(-7ahfUgDs~@i!kFA3oWC%UK3fpPXDunEP<@ij z-}^-1xh{w9%kJz$V=khVz)jmxO+AG}QX?o4a%g!gEH2sy(a}H-&Kb3Li zk8{LiPY?4yDm-U@l-RfX(Im$x4I*Y%(!7%qFw7x}DO!J`D#AJ83OkFIQCo2j%@j3D znsX5cnn>giDfZZRB0jvR%dUJrh)z|PkW70{k2=WHGb&T?s$n+yJHT^`zJ@T@H`b_O zvW%rihv2K}QJA4zjziA3Qhm2;d~n)ZxOgE2UanmPvhtoFQ{63a<7dFL-TB~rTanGU zVGQ#WZ$iaY!L{riOvb#dpX*&*M!U2__(6p{Oqg*J_`Y*ksySCQ3iROFiDddr%?Y>l zHRHcYXNk1!bt(%w^zunDm@Dx9UhN43^`!5DTg#MnYDzH=*9IC>l8NPw`)L3qa_O;K zsLq!>YQeuC6Q{X>Qwc|0z6X(F)i=~HR-7E&CkKxo?xG&=mg3`!p>$h-20p9yHgYknlbPLJF7MbxR@rTW(OB<#%cK=|l)^yhAb`-zTlDbLemJ z8@$xUIw5oNj_iFDj|Be?Ce46ZJZ0FA`iczbyR7R^~j=bSR2hDV#pQ>R!Oe0&n#v2=jAxH&Y|Q(bvWB|sn8L1ougBYWWN=2I;22x83?hroaF}u(o=y%ouXI#~bB`TBlx0r(4QIp8 z$9ZJyme16pDw%%tKSs;FdPuod6p6kq_!4#Y(`2O|yqH)aO)Q>(YyP_=)PsTr=U_f5 zsp=r}ri);!?HZ`ETLbd9J7|DoFV_7vCWqIV(L)Pw(e$sMXlHN}eQ6s*=6tW>pE*mw zrDbWLA$UVP;FE)vxClzunxC3Y@Gw8QcV35#$eNs4+X zWz$JQqgB~SpA5*pqhGhuz#lC`5f|nOJ%uy8DE<>UR?)$&5H-+sT01~t;YiXW;&6>$3pGu&AlED1N+=oAkz&d)m}Lq4 znsIQszKBd6U&Bp*d>Gnmb#R=XGF6aW2tmsecrEoGcue|$kh#qgo(5H-6g(Sj(zP%o zK7!Ai@r-+{E%?l5`jEu+=FlsAXWQsBW;`~IbLv%rN!laH1i3~`kFpf58xC}vNiLDF zmgPkyQXG1C((?8mp_ffn#a>{^P{tPZ=0LY!JP@)Gp(s-hd7yN zKMLFz9^;FGV({b62)v&nu*zJ#Fsxme`}Q(gV78VoFMmQeX?l_HZaX@wpqO)X6@%uB z3rUCPVN$r_5na1zIE>1P!EFcHFnWEgkVBo0avdf_&)`B4A=#`|m-X2mLj=`px z5!7xJqtXwylW`*gLE+;p(()#fMCRGjE20tTGWahVONTz7%1HDY}6p_}S3#CxqEwN`-a4s$hODfm%ggK&y|X zyoI*`cGxm{ck>>)6?ZAs5AY!^#^zSts!}rGauz6m(Wh-i=4IiW3t{fj-Hr$ z8vXT_P_f#`t6ze zwmy89Rm?VRUPl(4oK0*_7vVSgcrbb|0oOgWQR~n!w#+CNeP7E%O8XhSe$*24iZj^n zkt6Z!hPl|7Iv&=3YoMag-^e#IYDJb&S+5}IT0DYX)h{C_sR}#aqrysd4M8tPcuxp=VMosdeQMHTv&rT_aXs|O>v5i95=Mf7DpjtLt#*h5(NECljwpUFm&ehBYOs$!i|Qk%7+c>b>d7a1mAP0TTc!Cg8UJ?7VzURO z1U)qpbdISE9J=X_upko`KS{v;1xaMkQL~+_!Zp#lTZf)l(yfyt1mM>Tg*43G0!Mtb3EQo`%*>+s{*;)*r#CHm>?ZDxiJtS^%IuRK2?`W^L)bm$2 zdp@*0By~FSG`r&EtChpI}Iq*d1EDjf*BA(AgQBS%a zvl`U6eQWH9!Z|r^=PW~9WG8}Ria}h2P8@!m9>&$fYgBwtjy?%KxaJM7L2^}iYVBJz zDxC(u3N_fdRrhiI0s|;qUXB9>dbm06H)Fgbp7eK*(52rTV5yL|!0m(=?s+Q7x%+0p zp>NBHMyNKH-q(Qa2Uc{PLM9w9oyFO`=KVB}Db*4wB?F&NvB|ypSo2GiJ3YgSi&@eK z{~hLg&2H%8t=d5RShI{{?iZnqzYvpeQ%IkEYQwbL^SJrd26nVn6z23dvcFQ6aI$NJ z;nGMvbNQbr%#EIdQBP!nD$iwa`PJ}#;Sw@7#TN{piQ}oIRYZ^HliBQF1|yFb5fEKS z_Z{Ff2a<$pGf61R=T?n6LNaKzPvD832jsz>@!LZaR z#_dNTu9A@ve3BI5e6s^_PlyD!d))*O@hFEiJNu~0cy%m_lE9u$Gw!-lG&w9QfNqu9 zl-EiztkIS{um~bQgR7{V^k-E2oQ)IZgZTNOl+Tgh#MJ$pL1bwt*(ALIQW_Va=4oRh z{%{Nw9GpnzZI|KFYP!i%VKEy1G!q-bWx2A7N^+*UhS)2P$Cs`SFvn&lL^KSbK<6$- z-F}M-_RC4G;0&8J;<+LNzI5F6F$zEpGPM(542jAABbv?qVc= zqC!<#e>wi$(gQm;sX%CNH|P8BGp<>%6UU!pK|j$Ao0lunaguY`Z;Kh&@bd%fZ=lSU z^@L$D7sqK>=?ebtvVbcI8Dv|q3Ae{10j7xU2QP&bG+iU$l4lOm9jk0XBu5smh%sEz zzjn5jIB;@zOi1mzQ9>r!!F1~Yw%tUF+d+$oj%^%l@{>nx{c30?NwijC3LZTFm1wt_ zF+{0R&9%@SO9N@sf}&EVQT48pX&Z1Vlp z8NBs!DVDzCSy%&8xDWp%>6K;eM0Wpuk~y}3x@9F3Xs^R&?FFRiQ!EJ{jYEYE;#`X8 zEL3ZH$9$Wx0liwxaCPNnOfB0%)tbafeQFr>SYk}~o+;s+rrfRC>%sFT1p(~3fF#bW zyPGC7X>#o2#mt54N?hr2aY5U{!}QhI3Q*6{rs>yiqyGdwQulk9DLa@;d;ZpgOO`J@ zqtm%k%`_%tz66AB@1%eH|D%2XA{m=lIan6=fegOBgabcoaC?dZddU6cyMcmX(dT+H zH$svO@8x@<`vdVBGc-SV@rFY=!<$Um4My3SXRc^^!|7`+NIe)=VaLru~e&$6P#A zK9PI-yBI3tQkll9a-4EWDGq8ElJ=XIq4kjzth6k*R@DrroiSx#tzg3}J|u@b`zz>} z3^&l*dydw&q>wK|{gmf-aHo_e3oIE0cy~V$HBG1B=Fl$kN8=onW`1J#R5aoT&1X=z zIfxpr`iOUzEn^JTjA7rqU|RA}5B#)>iT&+b47tg(QcR=3ed{J-_9qQuN@j49VhMEj z^FDU7`XXZfNQTdH>%kuW9WAxsGih0MlI|+5q{}W1^8?XRvhHON9^$*5%*|$U%cM)N z^RWT9Y0L-?sQclDXa9)z$C((DybYau6zPvt8~Vc4j+po6Gv~ri;H`lZ^iKIO(c8;G z{r&x9^TB`QK&2MAT1s;=#!pfI?@?B;T@NiRT*;9gmT>T`B77G#VdLFky!!795KSAh zH9ilY8)g%mwEfnJZ=!MJoEP|huY#7j$z=5NLrkzegpXS!1+VVU=T6)yVU2gZq$X_} zY3jD~WFRq$ikoY5y&cZ1?*uXY^rL8WZfx<}--O8&PzpCIr=lQsDO|2`=&A zV#bl1M>pcQPf}2LISmt9wcwTg8<49~XIIFK=h!KESX8gT=dgBhOkfS%o2*B2$0%|$ znvdb%=;=&QlP7wO4X2)2F?f2_GF<$+859mbWfNLgkiU(ieFdbSO}a zc&4l&D@{vquBAH^4)`FEj>50g#KCurq+s*raS*$5ADw=BK3V)}JGC7hAsM>ykh}gA zI4?0llUq_;c+GiQyxEp(nxRE@PvaS@IIRDUZo@HeOGusggid>?fuUwQiEpDa z=kr>bI^_=$r%Oj6MmQg}t<@mtp$Yf0D-!-VHKKaA6}N6$&m8@=Ca|{f5Nr>ThVBvx zu0OsPn!FPrEn1ArI(h>K$SWT#YKM+w~ZK zHg(Vo*>muQbTv~RF@*+qzrnE00xB+WX&4aXTf{@mvbY%_s4^ z>Q(gA5#vIOldHbZxsTI}wRk4tboiux7$p9lBnO{|b4#8Jldq|}>9o)7w6`!H_ddNw zr(Kig1R9_4PxF14aITNxYvGX5&c=A@X|&N{IvS`pGwZ$WAS@uAPCm--jAO@eUPcYj zoiGWuyGU^CWJ&sKazCU6*5HeRi71oa0t-4NIHwQN0!#IOY@colJo5_09iAfGtcFW` z?|>?r`6y%f(wXT0ZJ6~po{jtMUJ#jREBw%`!D&4h#?;2cY?s^*2wHZ5wRZYsEs@jC z_qi!x5vPfgFBRdxKaW|Z!`tYtzm(o9yoIKJ3PGLUQ|x>eg>qdw-0T=zC_7BiVuC2w zVkCEH^2;1m|mqRL=aV!!2kRhJe&867BCxE@LIV z?e+m%>PpdAcoR-GaK*B589c8mn-=JBU|<=6|B0!A{NAHfH>VF=$Ed^3r{~!p=VEcj zwP<+ku*kaIKnljUeZb~N4S3;+6TP=aA8mW?K@YEERf9TI@mfry`T?|fGzNrkBS_|f` zjiD3Fb9oO&3heW+5U5E_#0v2Oj0rWs>we92eB)~X>X9wdUJCf3~9k?ZWa6J2aep%Yl?mqLoHECy^n1?wym>1gvV+T9a}*Aw62 zFT2@XL+Eah(kP;v-x+X?C**Of#~V7jUzjuSx{Nh(AE{dT5x$Ej8|sG*K}0E^bd=wN z{Lem=*}H=)`oWTWPJFkTv?e~P@u8`T@vswQpr*e9P2RP!13gmcStK_qJ!XH3-HyDz(t47(Vpw_nDMEYY!X+%y^a^D>^~P;q57U|o9&8fheNpb z1ybBK=Q$jmpTr(@J6d(j=R9tTvc#s;T{z-ljq^mCVd{E!8l0X(oV-qBlan?+tK;9L zj#`2(vd1wh^91I-KS}XnA!O8Fz=n1K{+i)J8P&H~SDi~2Uns)YJGL`XtHa5;Im4u(n-0x)XH^`A6h>!c0Z2;v-l&vVfg-H(-2HsHbALk zG}gW)g8kYZ0efTbpup)Q%$A*prMx!R+7W|)-XRvb2jCR%Vcec|7UL(LfF(ATkTgyj z)+xMZ{MugQQSS-dznEt7eori8?bSf<-mV}ri>uK``5=l}{ASgB<>*1P6ewPy3JyC` zt;@x)lNm!jJwqGE9HS4obl@EB6sh*p)#MZm0l4}S_AW;Yc*3P;wJ)no7Up^${nD6DGRpj z>mWxp5JQ8VkdPVJrqhOI0UJ?UE0NBAFG|#U7lE~J0q&dJLkbUNlTu!rN%go*9G0#Y z%x4$ih4C)%BVastO>Yq2yQQ<8zR6_ZswI3(3}G6#_dwFzFkpo{;p@$Ge57z5IX82X zRB40#1NL<9TT`s9+C#a-6bO53h~201(ZFAY2&7YE5y{UkujZrk|u8YW37DM7Q?=TQVCj0Kk=Q9Zi_={ zx!MwV$A2Frh4>DX{X0?XVibs}8d1l-PPS)jH+$3Z4W9mr+~xC^SkZU(7+xZd`8f~J zRv{c#KNp6O)2-Bt&(g;3l*Y$p<@UqVS?Nz{CpfnKj0sD9H6xcJS+<$zh{WK` z;0PKOEz3?B)?f7*rI<7t_z0;m6M?cE_WHv>W#@@d-_}ipCue2hn!l z9Qg0)H`2Fx8q9uZL65xhp?uJdd6fB(7D&W_u&x{2Fs=l9om(W@l-FH@#lT!D6h2?< zBW;t9!~CaL>1=6bc={}xc&@uoT9?cx^Mwy{KWz4JYf@`bH&^0iKG}K4s9I5yXBd*=tCLZ9P*F8Hy48-2WfJ5?Mddq&Dl(M+Xt5NzEf4z=kPK% z9$PZZaoQ3CuGsh-%I=n8M>ak}F$KhLn*LzuSV`#C!{mIyS-<mjC`#r%!ZjyLi0wuxynNCZ<3=Cg+KgIs)Ql$cVwS_mlK|#L zZw%*f+K$`YG>AoQiBwbF37%$+!!f5Qv(QWwd@Y|7k-N#LU-1nS|H)wOw{)iZrX0Kr z6Q{*t6XDsgCKE1vfV|9p2qG0{h|z3Kd|HtPoAk@5`1VYqT4BwV@ILApgQav%R~Xw8 z#Ak?aRlpk(fe+`1aScTe36XRK*XfD$$mRe%BbQ7M*-Rzg)-0fbEvO#3OZFUQ$jI+m z)IL8MJ?_q+M~iF8hml$M+h-gzuKybD6mrGyH&$b&=>^O=!ut)b9K*ghb0A{dI`}bu z3N{6I5b1pz`DZj5G8~MdQ&bCmrmiO*?pvYRTOZfWbwSJUJlN2yLCoael5N{c$>u*s zq(G^PHJNt^MV;4U*$qj&)}hLk#z~`$UK`p>Xl2*e`qQ=ZGgy%Yx}^X6S6` z|8EzAW_nAovvs<_xsP`g-8cbfZtKy@nYW20bg*oX3@+Jo7|Zxf#ni%9df8ozyt~~@ z9_!TL+ZrYAYL*BJzV_mQ2jZxh$I>oN10Pq`(ZM%H+3Dj9w^RMn(74llZ^cMCV689h{j!Z;B~l zWRfoUp9_L!ub=Gv{BuNz+@#{i(wJ}yzEZ}efY=)|aLm>TJ>o=Qde3Gi^JzBSdvpq9 zd^D!79K6tVf*HH=`BwO4#`mwSH-H)XPOQ^(UuHAWLpeVOu&r(+GMm4GZ}}E*l5v8f z)zv6GZ5D1=R*YZ#k=DHNqAw=tLEq^W7_h#FzAL-S)}<%XtO;A7>+5s6ajie^VK9V* zpN7!uaAN4g>e-QukwEvf(QU z$<)KptUa(O=_vHI?IQo$E)wX;AY*C@>7{^Uus8b~HGFFa$D&&3^F(O~OiCcF67lTk zZP$rqnTw#R*d6~3s$q_wGi3hzKz45HrOWMYS(Q6sMASP&ewm5;tItAM&2!@7 zuZ5jub*Nt)3fs$Hz=e|y%x=3uDwd1*Z1HLsTGYjQ?v@ zMXq~d6-I6o!c&SDKk?=^gX^q3r-&axd?M~eM$E`#> zS_+oTZ3n%x<(1O^R=|q96Y$+YhCDrDK*jU*z({O1#1;>esHxL9Y4tEvjsC;_sEgzt z#Kv-oMmxDLTSEngyKG65)+aD6Pa?N9o#{5Ff^~Y*NQ{#t;QhOIM3(Oh-TAZ}+b|GM zYd<9O&(&bGNfMZbp_FwXdu2Jg>H6cwa4#G^y_(y!Yza5!{6_AVSR7c6N^mE(mE#m<4SA$u zO1`=Tv$JGN@R$E08h$U4Y1f^JIWP0*w5S}kjt?aZBrEy3;~bqVl*NgUW$}wk8Ep?r z!b+FB*x9fUtGinGy+Sj7(heaGvu@DprMs|Y*z+91t) zr*iYrFntj&vP_0U=ag`MTQdH8A_84DT@2}qU2AWfz_u$ zH7Dhg^lBMvk62gkCg1Paw>1kgDoWt-*-RKrdI)>;Kl0zG_h9?uA1i9BK?g5fqmLZp z@UD$6HxM-iwUg~&_4G(wH^qa?usw!BbIj1?&x zi~fJQam%9lU~i*L&YA_1TYmDOYNA8MgSz?5f;_Q<C6acvveGbi>08GZ?9p5agDg zWSo{YkjY2Rg7EY8xJ*s}X-5m;#*ew2#z}EbUws^=`?y2)q%ZLMt3T0{|H9U8EXIdQ zV(?$$R2ux4Fc0Oz*ux2J>?E0Jl&>pglSPBkW^XRuxzff{g@WZ3C>B(iwOo$ zds>m?{X9iaeo=(PCuwMZF97MuG^*|zVg2ma2(7gr;LpF3+)nXPtm%ou&6|C2>=koz z(X)bl4VUJ2PYuWN`1ABebPZ!vIYOUH`hlBSJ2CtTB;tZT2{`Y76DL@a4_Eu)+=^Us z<-cd7h`;uydXzCJ!-{8vek4E6pQtoiex0r0gxQ24T|uV(1K2h#ms)7df%`@)h}Fu8 z;3C)s6VE6T=e!`s>xKkg2=!##_luJ}(e>DDF#|i22p;4!1#pCicD9bYjw;25j2?Sc>I!OG2GTpjVtCLwo|SAr#wOb3l2Fyx z{B!k~ChXmTy-XyVcGHHKPG5vQ_q^fjo`#A!?Li=UWC0VC91Zhxn(20ZexG9_17R!A z7v-bH;hcvR|LCt znMC&HM&k9#8~&*)Vb=R#Qfu^pw*44Ga##4nB;QqJt+)zfc+RrwZFo9t5;{c-M8=?) z;X-^aca6-Yhsj&XCN?J61`immwXWVNiwQB6*6Xbj;rt0Rvf|n+_F_^rGymLjFu(Nx zx`u?fDM=y3=B^lY|2Ye$`-j=MXTR9P??v!o#w5sy7)$5ho&Yab|HT!i-8i{ShD&f; zPmMnA2G5jelDYFKL{HCw{7K3JPmAr0_bN-+vc3u4erO@{W+K)5ZI8icZV}_vcFbB? zQM|@D(TPsH2HJlY^UD2b)%rAeq_Y4#OC#A>jpt;eYBDG#Erj%XSFpeH1suLVAetNZ z!m^hc(82p`R=-PN9v9uGnlJ0woM&!O4C|@4+%#C>W<_bH6)ZSEpM?LG!MzvYmd@2# zc<(8tv;8iUjKw10mn#DLw&nQl_<1&a^$il1KTKblJx6m%Y4Wf-91i+j!MTLzVAq90 zb<#vM^j3iVKMG)inHKx0*`3zc1d^2tW^if?^Qo$_2)Q# zZYz#}pUaJz+^#UPplA>EeP#u3eLfH~IYpSH6M!$aRuN}L5!P1kgGLo^EIX50*_T+y zdR6(uSZxjX@6b2w`q6_OUp5dr&~YIqvw9{J;TrIG1|VL z7QfITFNfUW*Li?fk4x~w&v0y4nhKS>r-Fsk6#6Q-j?wc=uQKy^iQ2zfm`fpYaA)lo zI!P;+e9+RRqwD6N)0;Z7czXtE^8{F-T}9Vs^}yOi3S`Q}22wHGoSBj6Ne|`K6GQ!B z;@!WLdYyVn<$qLg) zm-iD}e;{W3E@eWwBdnDwCk|#+WP|r&D4cPY1l*B=fS~*21fQK*6{iNivDOga9!oBn z+p%qzXJGGI0~9qECdc?$qjz5jf3EvUVt%`TxbzpC@R}h>TCS*M%+inTSE13#h}0jb zLc4BPJS(P-T+dXvA9I?u-;}`=Cis#wGbTa)@y(<)WiFJuD&r$NZM0MfBQMuVp@e@3 zRkaw#B^$bE&dN17j_|z!w(3~-%b3m_{g3z`8lt7Ur0`Vhe4H%fg`uafQZm$G?QpdU z%}<4q+ufODL-R#2eIg>*{VR+c9jl2iRe1jUKo+afb_IGm$D_!{98TtIIW+fZbM2v% z1q(A-G*>v!en?pZ7cV2-a8ic5I?|2nFC;-`#C}+&uEq6dKY>V-gXB+DJiOufj9nN& z>&OZ8jccZ%VOG@sp$Qr73Wv%K8D#&bYxI@kE{vP?h&~r?B6s_4lE^4@$G3`dMt|dR#I_7Y-j>nNTa^Un6Sc8#+FJHy|7P%2`w#G<5QcjEp?AG`cD!OQ zE2Lq}oQayub0_&<@`p1d^uA^SUXMjB*9GKqgCy*c+(YA&M}T~h#(_jzZh59M?NMJ1 z7ZZ}m>Yh4s@v;cjURH*BKj%SR$!?OQ{S#AET)>*w52x~+z&C$QQU6dLWxvh@+mK!4 zwySd09m{dJRKyJDNn66@8wwCEyo+qzFMu4ilXSjN5jleE=z#V;e3j~sQ(h>;ET?fG zHQUVkal_rp^+y}o##jc^jBYWTj&iWyI=hNW?1R=F4@v7`9qZBS=8Q<-f3$j0C_S;- z9?s6ZPTtQ=hXFH9CcUa1AMNF}&?A2d{rw-A2a!1C>37n8G>09T_n!XwA&W4_99ip0 zdpwpxbw|d5hpHdEPmRZqi}X!q)$xH8(%CQ*5)jpDRAoP ziR|g+$JpL2|CqMaJkoq{1ATnGkC}URh<4@vVb}b9K;|x!;I4nIW%l2V;=GNO+457> zM5=>wUnzAq-4i zOtHrZ(EfNl{G6jspDd`vJFg>{<36FZw%{L`p_@sMk!EJH<966mmjpLewMgw~H|`GA z27l?Xs9kmvmz#}X)X%k*erb`|x_&E{Dk8*fw+~?-9ycUAOoZY3z*2C~8l~U48Q7Qj zhuYuu&hRwS{dw5xq{ei#=GmnY z0qgM3lw|rV@+!o9I7576^YHUj26ug}CNo}W{6F(%mzXGa_whV|vHqOd)JbU8%5y+b z4Yaz}l9;I`NJQgNEufdo{B{YBR^0}_C3oPKNjur-wU%6y?D|%qdm>k&1YjE0zhby)*DX-E$;rQ!@KmD~ZZ{zEySa>^^Eabb(x0xE!kIJ6#=owaaHlo#z-vK-8& z72wlb<%Dtl&K^5Cg#AWsbbVqO8a+|MiyItB_puZ@_Tf{oeHlv*&Iw~**G7RT*f3L9 zD`ES04LpCr9h^TwRokE)e78Ra&dyqbhR@bSbmbt`LJBXOk(}c@@T7OvvVGG_sQtQJ zc5dD?TDMjfjD3@Foi{Dt*XAK2M~R!$8@5}5E+sGL3PAT zVbWzqa&dbl{=IpHOmEuD436ue>wSAM$3-3`UGCF(nKitpm!;F&ZjjWCf0$aOOAs@ddbcAcTVc`jIGx`8cy%PnU%lqk%l?ud1C6v6M9gRl8Q^?VQ0wQ@akvI$5 z=u3NNQoXXAw6=)}>Ks?ou19LUMJ#4F67$4nTH+@Oi>5xt zNDvc5nk&&3^BUe~QNZk18KNz2=CHL>MNnTQ3zCsSf>1mDf3T*L=^JRInzz@}aQlfc zCP)e5NH`Vypv`;4-!ds;dm+!n3MA6c(yy~6@EL}YC!fyEEfv?Nt18BjH%mvE1=EZ0 zdLhqy`zp`9P07Vbw}-Sd&lvZ+y~Y!R@ud5YV}b zSs&U*UeA>Vjp$=oc(;$96unBUpF2Z+Qzn)79w(UhE)oo?Qn6;hTCgtq9W)HAqb+~zk&Vv)NF(ivBKR*e&H5P(Rp546nQ$gToFF~!sM%eVnr^&## zc#NMKN;by_f!nzQtjC@lh}<6=f}e_UX)-+!lu$F@QDR#CdW+W{?47D1lI zHPm>L%&nMg$?h9zqUF_y6NIMV+|q8!=`MmGryR1}DyVWv{Q+XDsL0&@&S$H?m_Z)T zV!3{Mh#BdQf~cSr9NuJxFPpRQhU-td@xw0ipCFpl4mDWswA#kJ`*MTPADWHH4=)4z zEdr|Qg4xBTuJCE~V^Sd=!K|t}gU1{+IM1$DMooDvSP)9bZSn)itEVlc01AJ-nY%G< zM0a;0*&3q8UH-HRcdfpMUJXldD9H`i-Fh+1R+Z+0NN-QY|C?vP$C~+(G3+Pmh z$!Mtak?&ccsAqbLzen9LSb74kT5Jxv5&NkZ3UEes1#MgyN{*i?Bi{@hQFl=nJ;?VL z2LE0P?iWK@TgwBOE#C*8o!4Mjo0h<>Er`|sbC4)VN@B+u5ll7-V;_eg?em>T>O}3~ zw#PWZiev`vt9pn~YL_WDEpx=wOcXAG*?`Z2HJ7{oN5hoN>U0psZbE6a-*^VkG> zwC5hxx%i5_-?sx2OefO5jx74%XB7&RY&oNN-p8Nlh%MWaaIV=^lzcHBCC^R7l1bHA zpix6M+>OS=M8IWmJK&zTEwj$R4wtR#VR|=8^51ha)SVv+^R*=i%MD@VNF1y1%oV?$ zzkw0J@5Eg9(>EKmnXR{mvHy2CjSysF(HtKle5MrN3*VzrwLkH3d6Vc%9LQ+$M-c1<*9UN3ceB}oVVs*qoj-(a$S zF(|(sq2?#flFMKAfZdAIY`lCduCddkOI32JN^5fI%(zE%7V(3|CojpEA96H)z7)Rk z&A{CUF4M<~E<|?Z49~XNiJ`xWtaWBNMFsO$~mYk!XvUHxM1KE;H~X_DvU!>&N{{T-v8!-$))(TX)Ic7*e~;Nu zV{o59N)Uc|7C3mSQYIpgY&tVawiXIu_K8l6l~trZljjk~F!YM@W~p+74<%+8oria|JbTC40);!Zv07o6 zYKuD3XH%xr=Nt%%mh%0nt9h;Y=nA~|ejRoX3vdkJHKXZ}#tj%j$4^604yu83ps2Nt? zSk4^ewZ6p9-qx;}htT6%9UZg3pZ9aTCxW{h;qc=GqVN71I{#=0Hs2p37~wNVnqOyu z`ln?;Rv2+)nk0G{wc;+RrCea4F8ABN4B20EaSFeWHT!J=YnF?W4U;|SyYaG|Ls=dO zbD9FRO@qMP9WHN_d+R?kw4)C1~Y5e3FF1Ml!1@zVZ#P& zT_uV;wvGXP=lLP8F~~C?-Wr@EErHjuZS)W6$(2M!OJgi(kHh)unK&|V26`A7nEo_@HJ>O; z7x*^Nw;rNgPd&VgK9BS|O`5*Xxvr?Dp@&_Zn^RBfCF zugdaJ_CG^7{<0O_r7XCa*MDN~$1--w=i@|cw=~TD9S+ZQ;$Y>PVl+8XOF!REz@ce! z+=k)-d@CmkR44MDN z6=W-K6TL2H5Egq$$3;iu%3dRIG|GbRoDxv^G!erLM7Vo@y*Y7JYwR#pr5{Ys(qILi zl^$XV`}?k7=*$(I&-fNL>kL|}2M>^$%Jbkz|06OkR*{oDf1f?tZUmazdRU=$mi6Gz zuWlEuID36_THFx}2`W*f_(v4IuCxzVd^HE}B z3OaY*pvy!b!-k$R@Y8VyF@-5`>wOQ@xy-HleA);u89#x4_Y4J-ZJ)w|=Ibyhu@k4| zu0T_}So|uj$2Hxrpw<&@Fi%IGi@*1lR9akS-6o%>qX%p7z^Nu|m|F)|?BnTn_YFjM zsE{rd7pHp^USZ;f(-`4-0416=S&@zVaWN@};2Zs9C*^+qB7H))tI{V>tksCyRn+dgtr^1ybXXtHVX*7QO5MK*P!u!^K zT)f7cYybBRS3Z{#7(e3q9p|sZ`REn!WQH+J>H0-4`Gg2+1!3^q+7$+cD(m*)prX7w;!UDO6aVcEn zZA+>l(gPY@dJyhrZ+%K~lrb4E!-cPIrcd)+cy`uy*gI1Qmr|BqP1UE(8hp2%bp}~o zmCP*8Nu&8T&Q$lQOx3slN?=msL-uUIFBB2aW80**K#b!WoRgnNv*ih|Z_41JTwU(& zQC$chnqlpB&5w#+)WMs+Ke1s$8t2MCL*mRObk5nx%AY)q8ouGyTlBV)y}tWlVr~RZ zcMZc;1OA|8kPJBi`+<=a!@Ds_n8+vy?$<4ZikQXZl~gKTNIM7LB%U+Hu3u?ZO*OrA zw;a~5U(Z>OnZ%6~=yIVd^9eQGOG0M{V8Ja3v3O!#A02L6n)Bk_1XsC z6Rx0ziZUluk%cKSGokZ8TSmRDh52L`K>|Eecs9mqx^ePYHbJP;x|s>3&Vg&;%j;cK z&ubAWZ?7Pdf6HmR!7nCOEQkKEU5b4!^3331KCyB(!tXIJ>Fla!q?hk0-n|{+{GlG2 za#)PR&pMd2m14tpS#H;}7_7d=G9!MWsdNPP?K7jGa@V@R zhtHb@q>wCwefX_G48MNS1jj)~95E5+g6q$&%XNm#SX@f{=;vLhV*T` zFmy=Ak*gaV>HfR|+8k8??`kyBWP(1<)7px9_vX?!{QXgWEttE}k%46VWZaRKfTzwK zCno-NtekTNE~>Ev?SKmud7gn}fC@>Sv4`HzEh2d*(x~@dj>Ko30g+l&D*wO{Hz`a5 zS*{o#ti6qwmek|%+n31I3U4wbsR>E%U*VR#Mml7oN>UQDsr#TTHRw#{-akB#Z}!^4 zqLEtAzY_{|oi^BTu8I6KjUthS`Lt)+c{o)$2bXDi@Vn1IEMT{C*FLHW+$&ddvB|@9 zgGeEBkDkCg7|o5laG7?bjf1g|W7)HhUt@uCIo>Pn06&*>Xc2qIoL+Q?oVT2Xo4+69 zlG;TU$9)?KYome-mHwmFv!}oV!4zK;5`Sm%5T*UYfs za#DWdi`d0n;JWEJ^hg@B3NDbc!{2B@i4(r9+(qwRtKmBfd7kBmW2|0<|D$LtuKdv?`%2G-Ynv^JrVVmRp4;<1Z+Iz zMqk8s(|(IPc>Sgm7ZUfHnso78HnSnDpHT&q%^vW1Y&)!FX5xqBLX1{m-s~e~U6}+~6;@~a#lyZQ+n_v$P zJQp!TVG12LqDjr>9wi{Oktn%VQin@ZxLt0k*ln}}TmG90Yw`l|7r&hAYUWsn7z zZT1O=8(vX&tK0CHaOBvYlQ7Nd7O~d|t{l-k2RpMYxQ#VUT`;O#ozp+_7Uh-3(Xv3k zOLTS~*p45>=FjS+qO>AHYLF+0)L$pPQiZGa2GDcG#ZGb97txI4|&xbyoi{2g=~J#JC5bG#*S~Wq*C&zyufuiu zxjG)!npRRg zyy>w|C-LSjqWNX=FzPj*dRt}WA-hik1jZjwn z5G;@7!+n+Y+~V+C7|ZuQ$hSR5i{(HP7c zox_bNQTAq%Ewp}gC5L_okg+e9k+)wHz&5HMyGqhAl*xxfCVT0XNs8d~A(M8j+GhPT z)C*(+ZqPGU7tyBWGVaYd$h{74LQR`UI!aQ}(My`$mw5&?C8uNYR-UtUgA%Qn9QYGl zjvMCQBR$8~FvA9oK1NIO84Kv8;@skFAdMpRljWL|0p}NaH`(051Wb* zl8`73#s-8;>wP!UoD>SBf0UHcK$=v_Jj)QtOsS}3DC2sbRjEj5QVF3+h@@GQZ+*w# z(YLS5nS*UxylXwr?|0w7bjH&v(_%>Nw#0d(*U?YMn=07)((gSlsQY$r`giIg`ZAiI zAB<1IHRJcA^)udmZ7U9E59bn*l*iSzPGyWgpXaT#%*H$Z42^EeVwUNBg&c(dlDabh zg6CvX8$Pqrt{aAlvVml=X)5{Gqe+8qogiU#rto>DJa)ccgbyD(;soo}u;ZX7e)QGE z{=H}DrIjjpC+02bDdG2f^bg5<(Tn9dTcAeL4b%my)XKXDOf5wj=R!HoNDaJ=7? z&k3Gn_hoJc{T-!{rjP?=g-g+Ao-!VcS%E|T+2q{~Ln3-g59O|wlQDuLTzpg|Y7X0Q z6Pz!=?IYnhZe|c+r5=JH_b_Mk#f&ILT|sxgn`a{Dg?-9>=y7B{te;Z}m4m|ErHe)A z-*<_bZR3VN^4@}>q&6Mn8b$RspN7&&j__ypIoO*lhtBh(NYXVQ`nP93EtpkC|CJk% z)q4q1s83|6@Gi17bO!VLp(CpI1VGY0mQ_D@8XSC7VM4Ape)-)=Zf*EWmj0cNtp9RM za+bjJ?>4~G2lJV_un*?ug%j~V!w9^yattZY=KHh?KOpf`0_Oe^(CqpNyle9Z>_?_L%qmBgjTlg#@EO{G`Zaz#CZ;p~j<4)5LJG5vc+S0QVtl-0XXJR%^ zif%D!C##?8(a*=5$l9%&;e%=$*$|r!RfdMNT5}`2iqF!gTkNDlUv=rHE*ny~dK|83 zj%KXdgdpaJG*>w{#X`JFf6+Ixm5eW3{QkIGSx8D52ZF zAHd_&g-NK*YZ9t+9ZOaogneGpsG{=&G%xWvtV1nKfxj%RZpk2%#bi;;?lIHaH3s7z zy5OQ$pWv`U5HvpMW$fPwz+EX82EC8sfoa|3SSj97G2b5X4A)HDV;U3k-VRh0it<(L5Sx*E~(wfdaFAUyG3>8R6&^c_hlfxJ_h9^ z&k?zmYMhDGLAL$dX&7p5ASI=x5cBFem|Qu4%CCf2i3>BayW$#gY#dK={7gtdS|(|; zy$)R(CL}*-D$kaqWZR`K2tNiO_vTvIboH-O?a@Yg6}UGqmA$|49wzt<%tXMVof?o zX4Zn;Ybn+(`xd#d`4rL5mc!?3Tj|ZEMr4a`5!efx)6HW(z%kVfY=6BBi+1YJfWS_4 zFpq|RGZxUpdyYY@ygSXB`;KRcoW?efeN=zfKjK~6!5IHoO-yt7Oo0@C{zo3bRq>wi z?0K=d(#d3cdE6y*ul|Y2Th-}!?-T4uTod`VT^r-?WHUF!+{syf9>4N=78G43%#$^4 z+`7D2F75p-T;r+3-5g|TYoQnP6IBrlY(77i%^yyu({e>>#@I~7 zpDnA=t|rD@TKq9H|LR^eSDZ_K5O+uqpM_NyBT0RkIaT8wjylU8LH)V{+H97JhhhVW zN}UxuQL2Ql4Ovdd$i|{;Bt!T4iLl2-ui**jY23B1&1}j3CNe#B9t0O_LH*?MWaN~F z;On#V6ywg&wbnK?#rqVIHD_?Xs3Mhc{sq%4=X3o(vLIAmfTP7rsI&1d`k%H6y*Dhw ztsOi;!I(@0>FFcZC>i`c}g;Jv7L4TOC?nl0+_p3GS=90SRyQ=o;l<_?I8aO!PZK zUGA-ZaM^Kf5krksx6!Ae}V;aKju*HL^S*A3Hn=0JAfCsG+djp@{X%4fZssQ;qd zMBd^G=Kq?_E&43R$;;kE`Gfj!eBleYNpooK;q@Ro$r*ol9VD3_^C5!&sl0!CK-o@! zyk!%i_V6ci(xQsCJ!mHlyK10)#Fm|B;zYix#)8u659IxMf;Fg)?b|QVu$F5U`+s}m z;Z0XzoYEm&aYqEw)z)FbehsQ7GhR@ibriS!9EKs0z4XYa0{q;Q28^8~H7yPz0kvjC zV^Ss=`+g~&3m?y23dzQIYPDo!(3^@Z|9}n`^T}<`*Gz63F<%iDL?sTi(mkGcs8ClM z@%Xa}J}uuzUGB`nTT1tcx)nih<~#$Auh^g^Dx6qH8hqcGMd=t8ln)-Hstp&3*MW=V zP}xdiELR8{CYqt;Gro)It-@W@_vWT#-^H9$GDP=G8HP*f(>IfB$<6FSC^VJlE+xxw z+`ULVv|)(YU+kn|b~9MPyz@}*vXhwaaG(h@BI)D4*qX}U-QX~y&B^|}fKp|~+$IHJ zuEa*r+Yw6ezT&Y({4_`MQvVDcxyirx%3Dn7c@-jhbw9+s)M9DmEeyCFz|dR|Ot9y$ zV&+A7k>mm@y~VWpo(e5AI7iY?WWw=d;lz{g);*rmOscxi;QkPPPrt1Qb(+2)q}y_j z(yYFXvva#7(6nUPH2eZoel4R@ruJGvZsmbPXiqD{5yM zw|P91SsJ&ut8q_@g{VmJL>O(p0hU2`*xH~a=)XjQ+Z0@Z>s7~6Yn5{-8aoNEyg5wv z4c)@^r)SZI{vc|VR- zkt)LwlEPTwm5Eh29C(OREwbf&Oq_|;?)lViK%6^Ja~wNQZveFdX%r4E#VYmHd>^(I zo{NdW>wi2mZC4K2>+1r`B}3T&jWBfI9Lnun(oWPSd_*JO$)Ttg!DL2`5ZBW2JZGQ> zc1VU|`s;I$Drn=crZb%7OwV{vty6EqYxStyK};YTeW*d3z*%X?Q* zSM%|hzvCpyG#@6{L*~+5{5kl3>7YeHUago?1YTe1=Lqk70L4=NYnj= z9&Z`u>oqaz!|xWpKR88KXQWf@wR16Tv=@>Jc}Ci~^RQ;M2p6v#N51j7i7y2XwD$U5 zn$WFHkKTyE76oV!d4y`nseAV-`&jD40y<_SK4UmdgrpU|lRKz3QZuYi3h9tJ9b@ zd=+LE)KcT~12|3d4;;|b#>)xY+ zw7wv5dZncjhg{ldCj2 zI3SMdturZ8VoZN8S0hEqVa%%q-c)GmL@e9>6>f;>gMDKH3~GhL`ek2gCX1!vm5}T3 z>)uS96mtpmhNs}4EvdN4_&Hu>4Drnm0XJScmh4rQqiP!O$mfc$%&E6mNYAnJ?FbThQe<6WAA})q-ImIYaH0wcwtj)8#?!@qLrIGH>i`0xwqd#M5ZRX*|gSdh8T32 zHl5qd4&%@SIquixcQAGKHduMqm+q3iOHHaO$<`A*a`Q&VJtCj$4xk@I>w$+##JqLp-DeQ{(nv@shRRafP3;O}k7(YX@*kMIIIo z?j^b|$@G_!99^UM948(N;C>wU;M|{Gp<;1oX)Rkz3i6(y@CAG9Ned%$e8N$TImKK} z*@p!?bOolyR51J85Bk<9hG!mH(EgS#u$4H0({rZLzB|$Q`s+k$D_Te_V**txaEFQB)78oBLkhfRrAB&>BS1n79elQUyT@unHXd=dZrGTjm*$`7rkLR?L8l4b2&^wkMu?`~1&wz?J*iz9FODMK5 zAq|qtP<4eh{_AT-Q7e7^`jx{;Dz2C{cOs2(vZWraO<>m9i>uDxg0;STnf$>(%-4Ta zV-j8o4$(Qd>RUM5(slxuyG|q3i@)RAU!5Rp!0$Q^yaBg6!{pYTHb!Q;E!w9EQ^83o z-g``F-hxLU`Dq0#Jg^fjm)@o}>fa#a>qcxI5T&z1FHlcA8S4FsXJ^?S#MJLu;181- z&-3xj+mGpF@mXoIHgEx5(Qpm7bS%T6xi`u0T@BDNRsqV77J%TvBL+p{q5Jv;6bs3M z!hN#1kncX*iT|b%GA9HXd|o?qXadeZ(gu4MW-yUY#zN2}LzuqpD7Ck5Cz^i~>E3gN zFnNv;e~zy}5cx~4%wLZx6F9IbKf|mMn4)HC3Bb*zcZ;aH+up@=5mr zbL8kjcs+NN{WfzoCXbH;`Ku~)Z;k^h7gxc)-!{}UvY7q6zy}R~KIeY_isKTu4PzRA z#yg)s4@u>z*>nNmI(YA z?i;a;l7m_3Km3htTo?lj5_jRTuko+8oy)Y=E08E3iK8)w|Iu-(lEk>#LXhKbO2(ySgMQr7iq@g+&k2oi$oCyy_-CZE&D}OdRN)k2|PkT?h{A+ET->PTaV_IVcsX28q1S zbCyFJeUe_!s7%~M?rrhl#!)K?&7Bw;?$_fxv#y3xpz*JW734^>~hAe*=q%H-lyR>&jqsFa|v=x!?@@M zKTaTj1&@htql11wam9W$)R1h$XSz2@W7aVk+w}`q#_Q5GuTIha$$eP2DpPRo>u$lm z4hg}Q_$<6rVu2T;PvXVfCiJ7-SURgylPmr6A6+=vNN{(qEv;1+qC(x@Y5hb+j_uq> z3~n;yOr!|6EM1zcH|!^u#Y%~)b1Z-U2Ge$J8SdY;Q11E52gE_;2)*JQ3fU^N=`Vh^ zv|>j9b|xoN*wcc)GK4va3ro?L*^b85RT$(xhALd$#Hsgf=N!H|cXTQUzF=M$k5h>U*Bt&0#e#Mc2gIvZYJ=hZ}$;m3pVk^(ntsmJ%4Q;b= zRbD<`OL3&iWg$4)-9|m`C9^w3*3*|=rSy5*Ozyf~1fdJlsA!!G892U>_;mhcD?^@; z4~M7FljolhU2QowNB9BAsY}A^xB}3VaHMg~aim9FgUlCNR{ecvA~luZ^Dk-DSml&X z1NzSME`pidK65wfeqa^(@HqmPuZ^XWE3=u{?rfr~c@z_)D$wqg0~P6Qq&H_N^L`!P z!IPB47A~2NDSplvoU15Us+L6O{_vv_?wRCS%N)#&DZ|NQW9a0SJMsRcH#FRoqY8JQ zP_EkvE9PCN-ij8O{(L=sUw@sx+!TWEpI(AJSLM)WT^6?7yN|Yi64N#XgB6@%0+wc_Lq0)m3wAC@_&pPaR{e&@d+l5h8N_6JJ zPU2~8!8NZH<#tA7)3ctpnfs!|YhlJ2USTpFPSc423lSntp{`y4zN#xhgV?d$wTrO~cW4Rh zkY7j@rU~KJpIIdEiWK#a%xBeU90vKY)LJZvWV-wYzq$Z?F^W8-`I~HYsxUOEQ8L@bSBA*QZel1Hnt{9hVK5&cNTfa zdw@wOj5s<|;pL$;F2SDY%=RMzS#NRMm;F5J(wP%>&!ng2my?1;s9{?q3(!3X1fJU=5mHy9(Rabi%K1li=~XxmaPlk2cShAeK4b$xfGE zvN?SlqgSYg#w!g;|DiQB|Hu>)*>w<~8O!3V(FU|AxlMAbV{zlUxA^PxHBM)BJaKW%lIXg6iymW1d?=Q~RF7<=hc0|4vRW}@f&5z_XZnTsVP!A}g_op;l3%CDFOTqt|}$O$rG zb0tl4FQ&Icd$BFik9PEilXnl_LUH6BvcX-2>VCUG;tq(z#&2oxtzHSk>Yfs71s7Vr zFNln3e?cxpo6$-+q?+b_TtHbGx6=DG+zm%`#<(U#MxIcVd@`@+cl5o94@2@i97@R zpD#HWyc4O{GI-nXPNl!?fFC!F$X9nQcvunE7m8#sgUz9-QAcRne>pec#^`& zYTw}eRh@YsYb^eL5{yeDf3bezYXB@(V)`T(&g#%oj;nR$9xCj?3BJ6qAe+xC?7PiE z#&tL~I!tooc}{=R7ChxO2XtIh$kRX-(re{K<}oYifzNKtykb3E&GWzsI<$)0{XDaOBq3Bh+N-8FqOee1cLN+&wg)BsDbQZ?09LOMmq?fz=_;fyEcCx$t5=AM z2?qYe6d@BKci%1By$)>E^Kkcb*E2IRHZj!KpRaFbW@<8jv9+Pl`v3pSf43)- z#%W`s&jebP;y|XJ2?hV0L&UE53KRCxkqqXHC7Y!bu;_FG6+OS0N`N}?(KEmU18-ov z$Y-=S7U7&8e#83s9N7QGou<@`vgc;ikSq4rVQBV!@Q+cWM{Jx?^0Eu>eL6t})6Za3 zZz-IN*v#ILN#Pk}p@ENkD(9&j39&yJSB_5T9sk@#+iX%WF_ zw+Ke7d<*dw-GM2V57{BYGDg$36LDKha;A4I2jRCXoxx zA7xj1hmc8Uh4HigS!@&+FtcmYSk1Kzt*TfC7E3Q;^)+G6BA&y4`~7HGs0Syhegj>3 z-oX8hLA3SeCHS+L7!=t!jkQQ_~)fG7)0`O(UVH_=j1go0cDt2 z?H{bZeGw7TL9lhrfiy)QZkImANuoS!QgkJ!G4T|KIz13n6hW8VH>D-N=2PD*7LXCY zi%QP#B)wI~=z&8gXlhadO1*t%@$p$GCuy}9Z~v%-FOfB{-p`oCj;BcNYRI_oXwv=X z1nl{l$sX}_fZX(n=$nyB1{&Uy|E`lmc7k*a<52h?E8A_2a^oDqIwYhr`cE|WIU93d|GlTa1#$wpF)NA5 zw7b~u`J#gRmWyfgd3y}`e2&QcOd^Y<)SxY`ldhZq-02zN}Y2-EJoVy}gD!ti1xx?$Q)3)i#y%-JGy*rPN7qCRET zSnS+EzKpto>(FipzcUl=bUh>g3c6v+wOJsXPVrugGMNX9a4pvX#_5OgY{P2w3}225 z6xLvp+ZKos?SVb>f?$-p!vN$g~&P8Et zNKIi%<|xA0E1~2i&u{Z6DF?Hc-DF~q4r&^k8WBp*rJ{nVn z%dl(wC|GV?NQ_0)7@>@4;=MS8-Yh_x|9XJj*>M|JAJpgEKIn5f{k*5xe?B*18^0e) z2_!Qd^a$T%X18Y^q=t1m%qCweyjXY(%y=gE-qsyt(~aPorm}rxmf3w`@2(~Y-20w( zKT4o%q#sroyntGhI51tJYvBx2Nz&vQ>=B;DP?+Wpd$k|IyUzb$*W`t;v|%1QE$9x{ivC<cQCAEy5C)EVSBgLk33)8OBUjclla!JNW>~?|4W$S4`j;2SQZg zEuVv*-d?jsO+Z$!oJOzP1dyF06Q~+3#E&Cs%$}1&7%8_wnW!CDKTL)!jYn|gO|!+)uLX?5$$EB=ycwK)^0Q`b zPcC&8v7pOWZ3JPT8W{7jo7vpEoID-Z56J~eWO3;v=J}ZnizE+8T+x|`o(H7pYYF~u z+0PQ(e<-lB;(6%vZ~*%@wW8y}C=%>(ACg34NNHvRx)sQA<2q)b(5)aQUuhew>oNv0 zC<^}e-zNoTvRJ;%kz||jZXCTV)Ogp0&pkcy41c;8d?7r8Ga5%(Kbq%JKn_();@9bs zOysk4(t3L~F>Ah$u_P~aR$%oIP~)*vFTB0zZ1||jOV6q_oICQu{7Lg4pH^jWN+Dz!u0a1 z^kVEo`rX(bUxd6Td9BOoks8nLP!%)PCc;hl}`pW*?`%ppQK@*4SeH zK1<-t5*XVM8KxsHj-9I{!E@wO!KcR=ulw&sS%vQ~xcCozto+RspG_v?cz29|&vIGF zSy^nipHkCbb(dAk3PKaMl75@@uqH&&kRD#rOP-hr(J8MqnVI5mz$keY_1v<8SnKA) zhGrY?>)0{ehus3My;qMKOwmJbpOg2 z@;*p|R{hSTA0Dn?a+ig(0={RXd&me|-;|-5V+c+OEd#UQHT2hx%NXw!gu$Ic_|Y$% zY`lApeW@OfU-fIC@?jFwFnk-{?Rtmi!TRvnR+=_@)xfn4*`&&MCv)gZ3Tn4n!|RaM zq^n?%{Qdrb)H;sir~6W5W=1kOpX7z_vc!~P3A%-rQ}1odXk2+L>o6^h z+>8lFQ9c{kY`C9syJkhcm_Nh^8B1wPUIU4}x`&q8s?$Xmn!)-(C@?b@;GlE|F`sah zU~nxHSKLRK>j#*Y#<>vXEMuYdCyXu{6sEi5TS>U!EGV4b1alq~z_%UCNtEgrA{`x% zI~F9;X*^>m<((Xt`bC*C%cSXzv$do}AWU`47r-{Z?>Jja3%>D=n#;=;P#~opDswe2~ ziv#ao7tEuPAf7Rks>)2DW0Gc~;x;K9>}#VBT`tf|mi6d0d>SP8FUG4{R#fkCEy=`< zu+7b$+I-ke*NmvrvH#hEuQq?5SAxj;9)oWsf$ZV1M}(X37=7Od;lrJ2*uy1TH1=&J zS}E;JO>PX=VKsx3xUP;uQ|Dm4MmAHV5YC$Y^~Ph_dAQw<=fmZ#qd)rN=`>p%B&GRm z44=(h`gs{MmuDPivkJ_g<;8I6W&>HtyFKph&?Cbx2ieEZyNQipm>Be4qElCkfZvgN z{B}43wf}9ym71&Z{*rr)rAaYlEXiVrLRC>`RTLOBHllls7PEo(=e&v0hhcYXYB)59 z-Wz2}N@^eB_-pD!j{&J3GakkAe!=C9^0aI5c=B`aUSy=Q!RY-H()OmFJQ`a;l2`49 z8xsDgzMu(R6*y?UErJ`%K9I+!E~8x3d6d_k$_b(taF2T<`MzZ`o}R6NdL>qYaXMawmr z5T$z#R=n6p?6XdQ>+Nf(wDl>te77N{{}__z>PsG7Q-aV{AtI%{{1$NMBAofgZ~`NvYU@%RR-Z!ZxMOF zLYZ23bdbFR*~I>JFlP7&xSAt7(Iumbe3ATH{c}nUd*-7X(H(!6G(EkEEz|9>%-0Y9 zM#jMUafC$w(PUoFRfa=<|H6f<>Fm+Cv!JqIgsIrR5|%FRW0lUWp|dg#Skdi^uxrUn zVkb}_%j9fG)1_$Ww3`LWDR;>qxm;3l<2bnxTV8FNA|OgxEsW)vDD#9Rn&@J`8afBP znWlzli@tZ=tZTXxsGYbDPsOvClR1tw(l-bvM&{yMw>T7LWpSKEK09Ij8@P4Tjc!y; zAm28K&|6wUI6s!}|M1Vcq9qT|;O=B*l4cGD#`;pP3P)1VC_(LieZp&R64Ajx1Z!v` zbKIKu{6!VRCdV8mr0}|F;fz`2!CG+sYh z3&BQ*$lgmr%wfOvtoNN2tY+3-_V5Wo&Az%Id^9wbEBugx)w7e?aPgh+;Ylu2r?dfz zq;%lHi8JhF4gRxJjDl5nrQn2g9ORyzL5dEEV2M{6T|dOpqx<&IgP(3;!>=I7PHu*< zlxCQ}>LygIW_fOt9F3m7oM;3ugI)3o@OUfVZ|wDfe&J1M@Py#}NyEr^AjWMDqnqN6 z(|3m1FztaIbZ!lVvY#&aR=mH4bT^S2R)H$}*mOpFBeDmunK28D-3WB5@dPZGq_SidV{ViT;o$87cn2gy)1(&SVqe>^6{VLW&SK zNfy;6Yv7(=vl%8b$=q$8DD!rk9}2xWhPyTXVsO<$GUKQuvtw;3seIE8Vj}mU@ZBJY z8IQp9v+W@N;y&69W|Q>o(j++U96R=oWmN-Q2KG||E1gh`V?6G$ZKFYCOM`%{Iw4P` zp8V$h^WR~{tSvYc#NztImFPCmUbFa1Hy-S5g<+};WjePZuPcGE7S1NVk+(_0DKltZ zs!DEZ)G~iQ=EKWGSvKA>A%RQaj{94TPK6WbGUbJrF?-?#W`kUHTczNf+1(1qo)3 zbrADRVLq!kt4s$UKHWjb26kYozY2zv7*=@9Gx%^^ zgzYi+WTix1Nzi@inhPsJP%HT|h+UWi;^%(=drcB;=nmGi$%*JZRwnZNZantONjA7_ z2|QEJf-L0#sxfOPtMy8e_#=#W>V5Ovr*>Btjw`uQl!=go6=^Xw)JLa z*T8+|SipUz<%K+5FWE^X4)iiJ0?SG6^YL`qjx$6vY@dbEwMe$aH;vx~s6hqUhs~=J z_`Jaw++h3JqA*E``Y%p_J@>Dp=b?8vz4;+IzbA@rKCVS49{a)CzpG^G6XeLQvzw_; z!)zKN|Hr&IzY-R9JpjAT=kR&VMAWFyVxQ!eK(^s`_Fu{_@EdyI(f`;Ft-1eefoOewDJsQ{DaIC$7t1zUHV1)Q7$YsTb~1dT)Ny3;A(GHES&`{p9^ zchHFXZ5z+c|F@G~$(VzM25K-y=mgv>`v;G&xUl=fT;S)^64)NJhU}gf4BO^K^JiBO z=AM0qes^Q3t=R<{SmlYVRs)7?(}6Cx1*rc@m$n|XgfuNL*6RK?(!60FH}v~9%58fB zm+U%-;(@2c#Hf`V`uzbyS~KC)>3O*3Mj#zZ_hx^IKf)Dn23SX})|#Uq9x>r5o-q5S zB0hGj$4cA#P?EL=l=BwDz$yzesqP5zFFrt~3@syR*-7Ne-l_DOLlnDju?{r2Yt>A< zQGrjFjlo+^+L)Q@439taIr$+Ke87LU27G_zd~74r;9HNy)tY!9a+LgMF`ZNPIl-L_ z6~!Y%Z!k-^5-+^Y#dN(m*2!llo4s;8Q538pXa5W5{7!gq3Q6(kaLAw2h`op7wzCWu zWq@hNo?+GY7_6yr#9#RxsQF?oezk0d28T;v^-Pqyj&$>#KvSA=B8vG}lnPpoXYt9o zu_Sv3p>brdyFat>TW1&As=dQFZEbv0Qb#xBEu)E|N&*Esj@}R#L6b;er;t89%(oBj@b_rzg&_(PYy-KHwJfd199`tW;048eXk)}Rl!3PC9v<*5< zS3Ro0ii&D{F>^;^DS3mRp5Qr7R8wt_c*}4y_ZOE@kBa1?mFh# zG@|C@Kzvl6j)U)J;bi7HX)QhjUk+(Nxzro1*sRVyc+>|0GgYXicQ{r`zGU2j!?0^* zBBe#k_~-7mG;Q@Rw02km8yh~c5C4u66g>&U77sa84L^%@?=zSM;csdFh#AOOEr6B= zX{vSRHFke@#g`|N@cE<+a!Ym>zWa5Ha=X`YferlsuD@H^i&}fJdfhpatg)HOwLZen ze0FfmNU+7sJ=d{d(1=^$JW7tJ8o{pSF_% z7R|s)pGrDm&Ix#~n~vubuYo7e&e#`y5HCD0N6T|a2KFm+PY*sts6>JonQI^M66%7;h zhIXj7>nDdj#L;n!2RqGqA^Feq8@bzf8(jVYZ5S~SMCcr%pKcIt6D!a4NSQFfaXNy+ z2gx*U<$w5cQYBOe@O_vh|`Nv=w{HK83 zYp;-3WePOi)tFA|PGZ9r?;uHbvAR3%1r+*dbFH^l`%CdeQ9ovl;FPacH+kG zz#ok>r(&%inF%W=-@N_p|O%EUM#K_)aAmx;y0kKH69JmDP#Ia zKF4^|mA=mCA;*lXK;_B@^1M<2pB=5?uIYTbD(XA&IYeXvlq>_NCUzlz0 zg{jPUM|xpX3wb+bGF9yAuW@ZDW3%_0Le$q-@=qgx1XaYL`6CXa1MIjDS*p-Ju^o)( z?u3m#r^(_$D>m=bT)cHU5tN<-Ogb%wA4m5y{@uCEXTKiixq=rahONe%5}QaewIyF) z$20xnA|!D(p@O>+C^{Cw#MuFF-dc#yq=Ly#Z*$UH;)MqjOR!f)hHc^Ve4R0x*nK=7 zgZ2G!eHgvloPa8kqC7q7x%(zz<;6+C$#N8?a zMUDA1E#MUCAMs;KtuE4lZ&KvP9|L+IO#v)MS|H}Ofaw1zho-?vSpD%o++;ogmA}-W zb50Q?-WXt_7tEwv+j4$g8X!)2d|je;=TB&VrWm>#!&^*%T^e*;d^kfIR+d3qyR8$7nRlWDFkME2i! zsyf!aMk?!+dDrf27#>&&v1_wI#&ibMB_)!Ie3Rp+y)%>h(m}KcoXv4-YeqB#Wq@Y$ac}(k zWI9GXn^pBGgsX`%OpBNleWT?`X9SF=@B5u-Lbw+2K_RLnB7-mVnnA;~03#+z;}+3# zC}EQaEl&Sn!=794#?BDWX~^JnZzBx3c?;-FC)B7k#7m7=aD~lsdU<3HRr+d5YsYWE zJ-c7SPHfBs&irj%JoN8deEzx|4N{#uDeAJI@P+dh6TyDVdX*u$c9c67K4e{Lz zB|IJZ7)shqh}kD|#-sljBe$0FO(|<)UmO8bcIJ|d*?WlR)Jc4AD#hFXp{V^oeq|-kX&(2&iHbf@>iQZ+{10H^m^7Fzm5XhCSLlIs zb=s6iLA)@SUUI(4iYmW@pS$FU#4KwXtTP8sroH7IB&AgPoe{ZFT1G$DZ>9y)c^8KF zOAAkK40pPygnr}M6_3UIsqumUG%I~y<$iFsMgF+!_@mdJ7T#ZnvksfW;z{;kQtv~$ zLz75TP8Msh$bxF$ZY1+|TjTgT9lTU8&eT>|V9CPM(C%o-a91^f>=x!g{~D@l%)+_> zXS}J$d$>MpqLEby)$^Nz^-5`2H+2&(w=W?~YZXp9a)F(yJL z$1xasXp}YGsL912S_$p=2ouHg>F@Rr)YtI>%okmNvyWv`1#c1f&)f#3z05GGa1~0N z+=1@p-{FDl3vlx50mWbmI^l>J#9t`EDW3Px$5xqWjb@k>Nq2jYqw zYcTE5PY8=#gLww)!E!a=Rizl*XzWPM7q(%KPAu7*d6kK;Pi3ubCL>9VAlqgs<1=p& z>Uu2}VqQ$b1y{wfU#%Tx%!);=mdW%h;ZRyM0L9|G>E@c+niIx4ICq7(cp9qLCi-^}`P9^^g;y3*q(_(CClrX|9kadF?qL&n^zZC)=*!_E-uB zonr*ztK9HN;CUk5XM{5*EkWhv27Khb2CWuEF%H#tQFgeFlv`Fqd(kx7R1iUw<`v`D zmR7Pv=nS1G$c%{!Jtm(vZ4 z#c`ra9~^nFNInOm-b*wV%?^D8ngJ0o<&OOo| zq({ZuGa$#%g&O{|p=DN=*#g4|aNy62(e_#l$&;dWjT#kpBQf;faXa?+VV1pf&7YCD znvV@jTZ!!FWiVQ?3>500unP|Iyj}N3cGO-E4$3!ygym<3>Db18sglHH5{WRfAcfrd zAPoQgKFXbm?c$1-T;a-UlVP&^TQ>0OEh1dm0v-?5!L#H!^j~zM&pa+N5#RUV*(XNa zU6V*|BJY(?Qhm<8P#??P9yFy+@%hxp#nignP{+FM?+bEmqc8JGL>X%nFN6L1H1dx&jPd=JSzPN@RT{#wht&RjYqatQ*ZkyHEj^*Ij$lvdcg{k26KccGqksQ0Vl8LMn zgSVl=Sjou{gZy|pZ(x^I$jcXO^Zf{rNC;%JFPT~WcsdqUL|1Mt*{)~6o><}n!cWv;^??Y;+x7|e z%$7!%@g3~a5031ge{;~_TsqsNW(c6Z@VvA};=u%{6iI}lj5D}1 z>KS1__rWKq zIInLl{NW_+dYVs;MWw@BE)2?N-vrklc`jZ-9e>pbQH|mS==?nb77cWf40;~5m|ZY* zNYT3D$0>I1`y@KzrAGT6J;3*eW|8uU*X1*NWKi&I47uZ_14cPg3u}G*_4Ah$DFa#L=>h@4QD@{SPWAwwsF-z zp2LljU}~Ol7V|D%U}s$OW7t|n3|&-8Bzl43_AA;>uX5ksNbeP~*fxYRmm@%sq z*UwxFUmiBFI^kKwWBfL3nCn5z0=~oK7%e;@mc%mvW1;UzieSO}Po#j|L$hrqQOAA- zT-Hu!bQeaDS;wkC>tO=SU2~GYa8)7S#Jb1=Lm9ri;7sGio|2Ln2Qm zz;?4B@Tt#+)LIQ%mD51X?*AfoPp`6;>C?%+HD`&gQV)4A(h3Td(RAn2sW`A=8aiz~ z$;=A9M~;e&p<7}P^1k*VY;Wy{z#yrg7J*>%Eggu36l2Y*ckS%r}Zm6(x+`(!(&7c!_R^uDB$#l~Z zIkMh=pP*%N31OxrGD-uQB)K7z1gx|oyG01Z`&_Abs-sBfET2lxe<>w0Til6S%y=5l zRmRI^o&a<>uw3rgTda+8V6N+B*_ES<0s(F_{B`Y*Jd)L9I_@G+ycmHJxiLl zqn(_3!gKStPe+Ll4`Ie{Z4?xV7#i<8oPqeCb+9 z=g){h%k$-!v}8WqklRPZ^zSo~Kkt#0?fg5ktbja}EknofEoi2H0R7~=NO99LW?B41 zk}tZRG-(F15ng8r_o|o7+E_?_e0oa`I8?K{qU@<roZnD|JIbOsmFqW2#&kvUb&@O2xU?1NV<4B~GfDTcsl>@O7Uy=<<2$F@IBE4s zCJk$tvqq6{LJ-27Jlq7187?4}*aqzz2ib$dis(7Z4kt_0L$#zbCmnJVgJaI&d-1QZ z=KeI4X;#H>Ei-KQR2BH#6KB_IaTTk(E|BhAXA(0-0CoB*2=m3^;O8+|BWq0lORZ;` z=}vN|&4a1C^A5JYi$Rw%BRXU`fw23nLAo`MtuG#n!+chvCjBJ!1RE%B<=seJ11fmt z!2N^E_@689__-U!jtjX0cXpcK_?S{sbz~cM+z~-tDMR#J&NH6&bg^NJCt^Ho*^A%tH{Y+K}?ZJTFR#GskLv7|>Clbkl@Jvz$ z)e8c#Ira)E(-=?Ry}C{IPoVI6!4uXzSd@>6o|K#A#<7`MKmFvqjvDYfYQZ-Tys-d$Z1#Jr}0SPCq4w z4d--7Q%4*VSz^aj%;{$wE?SaY`)>HU@H0Eiy{HgLyaERd@1t?VXQuX-7o=)MFdH{a zCU<`(kzMLR*y#5TOXU*qPO2zv>&PLkkMh}rwN2onoXnapNx>w357D4Dmm04-M{PeG zAeX0|WJ4c_WAe{&*8aDec&|_nHMr4?UDJJ;)PIZU_0D3vqZmy6gw9e`%TjL7??`Z* zae-O?J(9Eg^^#ef(9LG(tfQx{6j$^|EF~A8cVYC8Q{?025xA+opSl!lq4oJNGN7Ld zA|JM3(!EMNK4^(4MS~T|tA?4;5Hoi9w=~lGYbS^cy?{oyxCA*j%s(@>u%xG54&KuNCp@1&=D=GggL$6 z^S~hP05f5%4$*#omeIOih!c~p;?c?+qU$HkUg=5%YnK5?vAk9BMeGlirfS@R{f1n5 zJt7*IvD*$OFi{<%`Jd&_?OxX+4gpYfLThR&iU}@%) zcoU|N-;0;FrC?Ui3z(Au zSEz6;dk!WOGRd0P31s!&>tw%;7yP$u3~IiMqgOLu(dk7ssPo*PK6=4INwO36<{hDv zwSA~}^cJdHn27Slb8ySvGh{~0DeS6$i02-i#2X7+;7R;O{Hz*-{z73)+N%9{qTo91 z>@LUk_F43piyM9q(xh!F#=w2(BX_FSaE6DCxWheaRB@+29r|;cI7;|Ucxwu45*?&K`OK2+E zRDQ=@8K3C=HyPa5TWVa&CNtchG>+p2EV%0GbZR+67|ow25UHaFneRcfxrSe@DF6Hd zZt?qq^$B8hq~;`}IZ2yS6)Az1iUiu-b{sb;ncy_lU=&ewC6V1hp$}mzb`4Z=KSn=-XKaARRA8MW!gRe$M>0v)hB>SI{QzcvI?#&B` zZ)i3NDefl;4@>CS)H14*>OghY1V zR3xvX&|(v!9GgiBty8!vi6p9fIhAPYo6zJcHx$pj2V1Y|AbUL!*ZunoL)rG$f78$5 zp3idD!8~{Ud_gQZWO)pNwxx0@f;vnee2hVt_ig~94RJ%MYR0`QbZ?0-2jr#64gm=r&Vzli^ zwqd3}-m+~hx0Rnt_4z*U;0!lx++U3JmJrd>vt_hLeiGG-Mf8?&5nXV25}EJb&%gOj zanG!JiBBPCrMmDC2Vvjv^g3buK@7S0j%e!Y?M|1pk42flQDP+3N}L1l67{n|;QakD zIgZi-waQgElJtqb9Jotw2ZX_fuw3FMX-HD#lBx60ZDeZ70ql~th9Q2ZvCTh_y|1tv z4d0g1dxNKF;Vl_VE%Ah3y|E}8c^jLC$57XbR{Ge|0(OQPf!UW87;N^8zpqc^8pr3+ zUL_H#=CqS@Js!Z-%ks?h*9thQn}PlP-(cW_xb^w}e64d5jzh#$S7f%WLi2+Y$@OcZ zbl5eLev}F#2X>n=o{s&@#uYX&R^N}Vy?ua*cy@UZK?+OyNN`&Du7I#Yk+S> z#&MO`BDkHk#&BtBFoxE4!KVd!_(U)R12wWiuBi|-6$G@J QM%7Td*(p-9_9XCzz z5VaOHlFnGd^So|=mh5!yQ|oOyI`1wV)my;**A|23Bg4#z_)>he*Pc5#`zapjkfLS0 zQ(JP%Jhrvon0Nn9go(B(Kt0atZRD2eG)1QSentY(Y+*Ww-dQ zq;_A*sM5|$_$(<5yERW@(C|1;K6ef#*^k56`egiMI1Ou=ee|%fE)8vWrbpA0h}^GZ z47cYDN=l7mj}4^2P4_|m8}Nm_oF7F@e>u=6vy^Glydh@SqMH~Jn#0X}qQ&idY|5#H zS1}7;l#$Oh`eg4OO_Kj;E@bfh<~sX3#AS;!aZ}A8rG0-%x0^1_J#Qcg?FgV&2B&Dy z(GVs#We-ev^oaerelbo=` z*}H+O`wKOXYv2o)&Q9Lm&5je~GWOpy!Mc{8?>qj4ZBarvFjXHOgkFRf6IR30w8!OZ zD+#DQgo=l6Q_5Y>N|SCaV^U~;hGfwdtk&!@_SfF6Y+YD5BO&ik4z0-}3gz2Kyx6$$75khq;WR6=eiUAA2g zzrC^IxzW>T$paS*k9z>a-L3S5$vxJ@l3pecSQd9XFfM{7GA`> zhrQOvVHG=`Ya5=xk7Pmk7P7<#_Krmqy&a>LTV8gJ6d&Y}Cnw zV$;oZsA&of+Po6d;*7|wf-G`eyqFlD|3O5%XOi9Xl3|KoE0eZ^cg-(&MRKnhkmw&$ zWb&at*b?%NggUem#RXX;p{+o0=tv1vbk87`eqj*rKMAS~a@hRC&SdTTf$~`r*{oLj za~S{R1GIk<#w8CFSzp;btfNdeDKZ~JLRUnw7M-4K#1bbmdC@t((^@9f}M9sF3Djtk6(abKG& zow!nos(+TH^;=W1VMi4jXr%G%6TXubc80Yn>V&LA2G$>z{H1ywF?50cRuVSvEU0zz z-oYuNc)WpU2D|0rHm63`-9U-iJi&`N52eA^pEg)MemVB$wxYH9NA$6)Me~o%OxBA{ ztVd@WJ9b?%d|r5(`rLkuZ?o<}sLW@9M@lgLb5gIo`eh0GMQ8!x@1#_xb}V;MY={^r zDbZI}%BUv(2J=_h5O@FaOoI0-up3>2%WfH=%JXPEkvqg_Ug{uXn#;?}D`%q;KX?D( zqRSnVH{$4Feh+X$jNAHFk2C^KjRB1K?;eVFn%CYkFNWv^q*0Y3usyqz` zHRpmsWIi)yMiEB-h`_`CX}E2PEnD~_Rq(v_60!7I%I6BGHPiD9gUfF-jnPKLYLyYq zz5JNiMrV+JJ}r!|cVziUd?7h`^(9_Xi-fg5>+t1~ZCsbZE8J3j9yKkcxwvXc&L}Pi zFL|cmNMR;b%J(H6H+RwBn>VA^&SH9J>mei{N8>$2NznInY~8y7gsfM>-uH!!vd>}g zAHED(i(W$MIT`YKk}&)pQ^81!7h_(`2Z%nr1LV4bFd)GM29KRa=b%AQw%88+{2cVl znG3{igEf)a(^PTpR~!1}$)m!;nYhsEDbblB%ZhJ$0&7$&arT-BG>^1;UEuwPYcR(+keRbrK+0ncV3VjV8237$a~_}HPba(|PXp`c zwXluFEHocd;KolKLx$E?!?QpW3V&>v5vi~6pGhBZHd!z|Wd*B0K^mnd7^0%@C3ZV# z(ckkdh}p%bOs~H^3c1bWPH&1~=Q>o7#$|@^Lf!@w#n8jai|D%-ZxXzP)RmG5ssQd0@& z%o>GU`Hiqp^%Bm!*N2OX18MG7hC1K8iy!&sQSweB?Eh{G(GOp;KNg2DPc zS!pqQB=$NiX|ZQ&Qnm;bx1E6~rC0F0F%3tZ0~sevPm(tBi1alFk|{SCruNY}=&#uV zl@I(_;f;~3_k1~?xv&zxXXk@)Tb3~jV#1jm)B4w z<|!=JP(<&JcBpiE1leI@*hvF+#9YaRxH^0X<45JJmwf`cBNWK2o;8lV96UqZZ%>0% zAycyD+g-FB>xYl8yg?m!iZ$QA@*OkY=UciK;-K& zH5FMCr~SmA-@E_15y{VLuF|`F2D63lIXqJMLa=ZnjDOw77`a*i{0S!CXV#E0__h4C zp9dyRkpG$?c0@7wTafiyj)B}XSk?xgSj3&X>{ z-`LWt@5#E9M?{C81BHF>C9ewunZGAau-T!yRwl)t*haHl{1(>_Wq%ymE53Q~U5dgl zu{*SF{9l-#FJ=9EjV8^#`IIab&xWa;d)a}VcFg4!A4pHc1u!4dAUy-g=z3m;b1Qj> zI~q@Ld)~?6nGI)g@zdFy^m+~0bwC^6D~zJ_>_8eJ<4G3^FQtAxPV|W08KNg4OaJ?Q zi#;yo#a5{op!sYwT70XD^$O7>mR`-^_ahl%LcXx`AO0k!;7%7q*k8vWYFgSNCrh7bu`*PYau!!$0{cDBKu7An5y`nU;OO9?5YKHXR zCV0vnh|7(`+&$?)OtgH4EmPaUEo>Y<^?S`ISA@_cT0*?XRg!1%4)n|+b6PblO|^@b zlN2Q>_N_=6<%3$c0)fBjX?t9W$9K*j=Qz3)|79)(I0utKgy(&-hR}Pk(M%$0;1Brk_2% z={ub!s=PHAeG(oJuZ#+~U_J?FTy`RR?#_q8wj}E6zKlM-5k|M&*@~6@vh>@xD)=*ZcIq0b%&)5Q-D!qwweN9>{Moq&oJV@Yt8!S90s1j*`q4N;o5$TUzLyMJO?SU zA{R2)Hi&$iMk1D-CB6fCRH}75eZN-;3;$FJE^KIpC9=z`nr~h|-)3QJb8Z}_ zPuopIf49T4{oyqC+a4NTe}*>B(xERFG(qa#_wY_y0$E z%-~w;BNT{NPX{wXUBBp;Vp;20Q9e_#R}xR>wLo024DK`eOD1cjlES+s%d!TF znzdb005eH1j2RId{v52k-7{*fKzS+yl& zZDGJ=<_A%gUURy9*B9cxe>>eML=7%hZGG}5b4

aCr*kn2FMTQ{7m9zi@c&xe+g0hhUaB z$Hv&6B@SFQ*|#}_NtB6Zwhg*44!81%lk77xUJynbxkM~}K7|@AOr@vqmat39*G_w|!xy*_x!yMO=Vzt4JK zH_*t}5_HOnJ5&&$f@l6S!~V@X;Fp3m+4&#?irZ>fAMPl7yLKF|(`K=y<}O<@PMaw@ zy@aU`GsQz6OW<>E2YI`4I@4sY0<$8d*<%GZ5Txr>zCke$GoLv!pWa-Bvcqxc^Iih# zYA?f%R6ch-*AUMr{D(4oLh%mI&fadCjZJ6A;SblHFgCH5ZK~Xio`G>R`0Fb)3i6~; zLvye+^#OYGy#)PvozUnt6(gkPU|>F>mVADYtp7s(dF2sTFCiE)or_1E(wWI$>qvlM zJo~%iG-Fz}ggCtkArp6+va%ufn2Dyr#J)3{xqYdXY?cb6`{vJO4)b~9E@4%Y5>gE3 zw`8#G&lPCXwjSnDYycQwGka6{J>x_3;DyOKxUh8yCSA~@&wm$?8E@ahb-uUUm5~GA z-`bI6i<5ZecMXonjsd;$nGl*01~o@c3!1LZAnD1zWWHY)(THke#xPgVW93HJc5Xav zJ8_aq{#}SceHCocqg>olsKSaSJc5_*aS&*77A~7!!+1?SPICENY{awBFBL+nltYja z+{Ax;x7ETynESgw5ujJdI^;tD?g{?KUfp+=zE*SP8pJJ#&%N(#l1d6YraTF6?M>oL ztmE+Fyr%-i?*k~gs}Z{2*wHp&cRX$Dh)+T?@#jHHH1i3?WywCIWQGzB=gp=kEJfM% z61%A0hi4f0C51l&%!u9H7!q)OH#V8=#H&jRaK;7;YE=K6OrLoOXMY-p3Sx>}ovSO3 zoB9siOZdMV+b&>wDzPx5#vj@Lab`nv$Md+kWvvPkrnT?8qqnegeZ1dTir zMU>x!;){KLWV~iH-Mw!we)o6|yVbU0%-1JmFfpE58O~>NHeEpLvtFPdbCATEIpN3r zR!BL`pEH&7sl0s?U59bj zRtvOT>;luIYD5vZBh^W1LN(yhTBJo@)d*JXswqL)7*t@=G9@d`&=S9J6rNc%Vy+0UE zCGqnc&0)|Oih#-)=0xB2rodF*iC&2^CsNNd;e=NwJ7;(q-J20dgc?$bx4t1YPqb&} zUC|L7z1_lUPxrtbgZ$m=vOITGaWT=-ol0z9YLJ5$9|-MWhN`Iki7aD zPbUrErD_jK=_dZoqd!Fq@31WMTi=gty(*8BA|<#ceQ$P3;XTIDHisYZh5zoO?1XR6_Rl z$})HmURkl$!yAMxE~E5{T%xmd195pbOmu>mAXf3*y}&DQc{qp|^xKdfv$IIBd^m9! zUWTLA@!X}VCDh{* z7Dq+Pyr{O30u9P7B2x8MBzB@b%`>p72xkq&D&*iw-Bbd)6MnTqI zC(g-7jCvf2rS%FQaqj(@91|1*;teaQ4S zQ5?Q6Nxyu4P0qJZ#%n?qjAnKl({s=aCVAUnsE;gds{KF~#omN~llr_DHX2^rwGg`{ zp;V-49o~Po4{vNcKu*`KqZ)Ig;MY|bp82(w`0zW$=CXAv(KKg|4CH2i!hFR{dQCEr8f`j@Qf4>V&HLtIW}peK*Eo*` zvt=-5_j#s{pRr%hSE8wM%kWU}Qc&O313k+aqWSVGz76xj&vx>3Q%N&sW#?e8<|y%f z_g7H&aS$VWb8ywLG#bra$}Bk6gA1Cy(fRj1@KXFouBI-=Uw;9F?CW5!AQUr-g=q{Nk z*tUBg?%M1qun~%&zT+<8>E0%=t9*(b0S@$j_9puK&0-uV+Jg-`lIZHEi`Bmu($S4= zBvEV!@v4d9!#dLasOeM^(nGJH-3)!d#FIW^`f-$ zx;y0>PQwEIeQ+n=9UpIe4-02@kO{NL(4SWIgqx&K?6MnCxp_AC(rpRx9I+wkUD`M< z=O8O2-3^tYGI-HZ53CnHhd;~sY|#6Y>{`;!geYDmbIwUXl*A>b^Xn5lsj5Pq3KJ}UtCIsuTQ{#>Q_K>YzV$w zP*0l6d?AbSZ#k(6OgJ}25OHx5H&$^sp67G$n0~+0o?uKz&(8W1GjihqM zN|SN5WCA93yrD}TO|srVaEqA*scE{tTqp z#DJfIH#7TQ4_*-wAwj0rJXP*P>Q9z9 zK81bK;ZV|ggAUk=S2GSRk&^Zc2v*k zr@d;b^!4Ls^zr%*y5rniMlpN}=W*15PIWJYX1-tXUEP%Q*0+$bjAz97i9f4M!imI_ zK&tsi5mVz6u;{^clwT;!8SkPv*dNH44acHK^C#5h9X!UvrORCj9bfhrghM6|9X&FTL{S;PEd`4Dth|tT%xj8 zl9nm^VA1{ukT>T9c>VIF*^A8~=%YPI?^w?)iLdW_C!k%we0&sC9Kd{(kra;?Dxin>_%sMSIx7YfB({eFtn^z6Z~L*hh`a z9${vLEN1_A5C+QgAb4La(O#>DQyzYUc@7;g$1IJ^`zpyKt=DE*&$%eF<|!13tc2eM zJLsvKl|(CgAGMYUv1*^Dj7OTzV#d!7*n2wy*2fn^#n3LgziI_pB56xLm&QYeQ!|OG z7;AlPpaO^Z-Atj_UiMjJ2sMs*MwOLC;75eI;K5KlL+=C;<+(=qPQo8o-cN$L4zt;~ z)2o1}d?VfAL*+%12sfpbFgm$~=#Bcr@(lt@hi(&M<9LAGoT(#F{Jfdn^VNb`(!nrh zX4l9#N&YuEvX~mhc#$XP^C9r?BX~Y0k+{`}6OXMYi47>xXGZ(4*~EWJBE0-O17(FyqC})HcfaKa9+xixJ#B9q8|OuT{}7>$TN2>t z>|Zc)^&_aKz9qk34Px46b8au#;L)Co?3KL-VDG)B;IKoNTPb%Lhff?N*m{QD_jMB8 zW~TyGR3B;=1j1H34<^=rK1_e|N-!~sW$xEolV6+K$Q0!=W?Q2W*|4;mxuP$O)31u) zZ82ebWc6P1FOlcR?D<8^G!o#7cRvJu5};362w5`4nO#1-03L6a$3r7q!Bs5@>?NAn zrKP8NC+Qn1lNpZkA6wuZ5kakSuOXUuSr`291JmsesC_dZeyU27>tjN}CrbeLYCT!) z!D-NUHUv&PT_%gJmxBJC_3(iyh4|s;D5rZ4Hy@tKpEKaql#W_8dVbAP7Tl&#Cwl)yfR)xQuSRYkRy=J=y(w0Pd(6VDO7jy4>O< zEy>+L?K-+iTbU9_4^{H{MQt>&b;BwBchOyRM4(p~T)}m&q_26W)~vi9R@J!`cUi{4 zM5%4~g3qk0-_WN^LJv|?slVj@omnXS;VdR+y(HTzgUS8pd>>msgVns0M;sddd%5#$~)UdOr5*^R9!Gd)Uz_hyz)LrIKUmDJH zM#9P3ofeQk_8}FrD5Jp^M{uR@X*?SF5y#!nCkqse$cLsKbY%VkYS3$lQDQF`C6|8O zTc^prR{w&(s~_O$Gk$1$@H7tj$>VX0OPEcM<7&}RJXh?1+nY5|d)IL+jpk>S9ue%Z ztzp=7Qw9hA4v|i=z2y9mfZeOQ3hk3y1@GSJ(iZU+cC1h>`gc}<);l}U&rPN}BNgD; zt%IM&odu7pe{fb&Z`mUwAu2i79K#c$Vb@D*Ftpx@v9%HSJ+g&0JGzA&RgZ*c9&V5= zra%UU14w~}DQVd!N@F`0L%F9XuJ}BUh_01n`xh6I3z=59`QtVCZ)OJW?@VRGS3M&^ zYdz64GMiW@2~p+Q?~!5Su)uT_Hl3E@uBvBag@PJ7PZ*B}f)0R5ZVSCpAYdoWIK-04 zC8XT(E_4RoCe&>@&tQng={MB4OYP4qpoMZ(!>6%fg*C zTJ4IoR99k>cLN72%du;?13QWt^2_WE+k3QtXc(uF>gvCw;m&ii_>V4_7@dQgo25|k z63-()^^?7gj%4J|G3=@v1hwb6@Z$7MI=M2Kvawe5)1$lOhUY2f)|(*mE3Ff%(s@ta zhBFnfAEx8$lWB~Wi3#Zzxe3Wt{Eq6c2426df~Umqu}veHtkn%es8C)A|IEHIc86Us zNg#wWzD6*1RWJ#=v4-TQdsO7vp2PX?EihqDCbb$Z#jQ^{s9{4Gdvgc+`HcouR#ZiW z2YYc`v^MBap3UfRgoq`_{v^1!N=-lch%U|9jVV`c|7B_op{4$=(%$Ud7 zRGo#_ZhACZJqxOGr9k1GA$za!F~7SxTX8Y|4ao^IL0@MxcrizbJnRZ(mq$+{OBODG zLwsIlI9Q%plwv_tS7%V}^Iz)Rlfd}7E8w$d>XD)FV6f=)X zzO<*>wJddc<5;1+^A?O|xPx-9Fnz!O0=d0)E6H-{fw%9nL00b&4~%&Vo{LVg?OYoO z|NARQ+r`1Iz+g=>uA6V^*H#p_C4V9I&%H}HlCN`#HA0KLYmD#0rA;~>G%_`#H`26 zT1g08nF&e}^^kbEsv=^ZH@Wp{4Y0B+Am4m4s$H!j^B1L%*v)w`~K~!U?Ec_GXlELBvCZ*7L@A6l0Vw!n3%PO(Rh3Ygk-uvqG~?Ij@>~o6i5+I z+kYf@Y8LF!Qsfo|yksBk@4*xO5}fsuNo4V{J2X2m*=nlx1&pyhNrwba*?(12u-{}g z6>Gjsg2!vqMf1myC)VFkRA~XpnSBH9`0VD6xsAo=4rL_3!A)?x!Gx}Md_~3|sUg>< z*iuagdFVMPMD1LD(!aA9JWw$gZ!}dhf$fDDr@s~2O`>o^_yD=)u1aQz1hPZkU-+{* zgv}{W7o4=2N&jV!tw{PGWoH^r)f@F;8A7IvDMd;pLr5vkUeA%r&|r#aqDgb9R7#U6 zLW&}DXh12IqH^|HJB

8byjq{!(ck-~E1lzrE+WeBcx3I+y)CYu&%!t)akftTtfg zS-1I{AC0-FtnJrnQGQlzgY>FkJhJ5(2_i#gFY{?&_Yk@tJby1UMjp+^tl`olL6 z^7s!ekxwCm1^>v}z=F&t*3*__!vy=0xiG0$JpC!wcl9_(U~$%81R`vg^m<

I8T{y-Yh+%DzML%K06)a8PEy%2 zjNfs$FDls(a4wlNM)NdvpQ^w?K9Be*mDeS*GFMsV^$0B3u!W))8`Gtf5%lCp6RtUy zh@UnzVWH&~p&tZTT< zb=CKRW&yTEC19)9SR z2iFcX!A=!B=n{8H(9i-?tWR;DZ4Cv*fFR*WsCaJjeZ${WImw$Uo07roUlTH%j1E<1>oame`*aJKmQ7$? zbtcfY>zU-B!U|4ZbS+7yy#|k{Q(Us|5F8Wph_0Sp%iCn;qRR6}n5WQ+kvNzgs&S&T z5gC;A(*okUZeiWq3<$nbO8v&GMlv>6a1q9K95CP-W?W>ZiV?p zN*EyiUc#)icx0`(ljby7)rVE=eRnSW8uA|cIJx8Z_ct-H{4bsjTnw3imEgPUZ0u-R zh*^YA#!};Tkv; z=*u+CKVpjFK{Tp8?@|pF|UKFzO0DZ7Hg9jgklFH)^G$A#VbcUT~{U6jqn~o;SnzfZ~CB@^7`YcGW zcfxb|9A^=yk2gVz8j8oct}89LLFp=YJL(+vjW?kPMMe4%@kX*O;VHoLT8Wa69OW!) zz+vHySZ$t88ZKujWSS4io^?z9-SWiZhMQ2kER(bL-p+Y1{mhqZ-Q+CtGGKrDP)Zy2 zg_|(=80t+vOU>e0fB(L9U@DbkyDrUR^IB?9t$ipqO)NmQ=5(&a&{F6&-9jf8XpqP0 z49vK>)p1W?9=}hnj&B>{fD<#n!y_*xcCp!$QqCDr!GSHjW_tD+z)FEST?tV0K<&gS^#j?qAVKe7is$J07WW zQ^q0QW1SfGBZ8FQ4C88ErLeHISJ{wN{_y?3WEgULDEr)@L&agLr2Mp$a=t7A?f$BK zZPz)m@1lll_bG<$o|9O|*f@!z^dvvzZa589_rxB{Eb(>fMoIm}*W4P*Xm~Si2Yfr^ zMf#x|sIjb@*yv}ZC0I&xoK2)hfA*m{y)y8z+c})Jy^w2LUX2DFC+TIMZpqT133#!_ z8Mm7K<)e@4vi+qh(znMgY2h4CCJCt|KkEyy*1{t3 z{fdtrM~|5&^_tU-QT%k2UOoxmly87dt_g(iJOD+9o!L@987}sB!5`Zt&_1snTJ{-| zxtBI6pLZnL-kUhj%*QEz{X@RDo(s$tohG$&#<2GE5c+q%2aVAX`Sku1;r8Q1-eLg?gk{j(}1Wer#RC+Bgj8kpYo@MW8sV$I7-aTZCH>E zw-=nmDBCRbQ11)64jZ_gyQdMGu`NXDCSOsdUY)-4m=FC z%TsWnMhTbKs=yT960h5_2bD}K@YWo$kJUsSv#K9RUKKO!ojwe*o|S=dlQ&y5P|Pn( zc8878xlpB3%zf3DqLIj1&7Il`sookaOLPfoFWChB1wVYe)|s1>*@T}*FUJ1KlX39D z7R-|>!&;-8(D68yYnS=K=hSvMWvY&gyub7JR`%shXM~6yjF)JH!eGJ9PET<6bSGKV zLyr76;3VZ1%DOUC$ogJhWo<1Z1G?>4v#}?Q9B)h~?rGCeM`b**D;-=%D8Rk!8u(E8 zkQH2YMu(@Ddm?5H88E(@TQt+kLoa|fLAF2-Ru%J?I% zEBQTJj!=a7zB%^D!I8nc`LX%--25y@XzOPTDbl^LEU*u^N;wDbUv$8m_xI41uj7T5 zMRJ05pd$9kYQe#Er%^kAxcwbZI7Rgs8ZknH4BRJRtnCW;x!+e%y;4a92@|B*yaUVg zun-nB%n*k4n9d%lr}4c_Rj4bk75&S~F{4=rCToh_647hf{HAa=!#jhXn8@Hj(O9-* zy)T5maE9sgRQdW{Mnd4I{FaWk^)u#-eAh6t0L_rT2|Dct3a*I3xfy)a7LNxloO;S3(U5^ZYkYTpkZj!hzu?($1-i8wxsLP!m}i{_v8OqhldsR>!}oI* zo>suuTEctg0sD^?iayWVyxss$C|ekSg~KPKN%9`RFTeQI!e3NwwHI|(Z{TESe?o1= zMK0~7*kfd%ET|gWfzjn$$;j2-_$Oc}y#CqDA9mU-8__o!UHYkW*NP08m+4k`zpRiB zW>;o5Dt|8eJU?5Z71?;h(p;VpCC;7`8ew^RB2gV01? z^-&bw&dK2#!yj`)SLlPJVgxj)6o5;GB1=*%!{uLB(vPj$!i7EviI>m9*{i;=-g-Tk zU$PZyQWL=FV-_m<^mh8lUUR3uFiGBlL|QLq=L0{~WNPdTF;~Fb#RO{c$+sRGDC@|nu z3(C1L*Y9Z2yayDfBy$1F;yKOKWnl4Y0BH5r6qNS(6|reWbanCX&8OWmLaPSDIeBn`zzCVjJ$oVcG--kcG+% zaqU$kr}7*>s*Bw>yXN4th|y3V6-Z9^4486#DtmLjrxX{SW(A)|iTmMg6rEtj9?Z~& z=*S23;6GD#2}^OFi5xDDRu#VAPQcyoF0-duv3&QKsjypj2A`+94&N7#5$~We6fyoC z4pRlXKB}I22@WhUSY%P+1Z+$=D>baVL*@(8Moc#rN6A{`{XFLx6B?=ij?8# zmNpz*bCWUA{gRKl?9Pe(G^WuD!ZN4Ojk*|i^NBBu`%ppLi9XVX)}GRkkX_vRbvyYV zcb{S7#&I-Nxld3)xtDi|vi;j`A_i({w zVi9;x3IGG6K5%-)WEw$ZC@t_Ecj0GmHf+NmP&PDU2A7Q4SHl+&yr?fTo~9stymp?# zO1{(G-@TD;J z0i`b@D9j;)*n&m;=csFvu=CL*`|lwp#NB~u&%;?SZ%bm;p`@#M8&6GN$0k*Nhw@fe z_962elb^DRu^oR{o!WFUBY&2njngHoj`S4dmtCe61Cy!T<`KT|{|6dX2jNp$FIIF= ziB)LEQ=Q=~{@3~t)cyMqbLTFgeSUG6E4qVH`8L=kuY@YSR55sNH)tj((?_#K>{QYf zD!eRq_ebkM>bsN3s|^xQNIy`S0E0xPgqB6BAYWSr@opzh?RInop3jECHhS&)W!6`6olZLQ4*^N7P z(-^fJN5k_lRkkuXf}LC~Cq4XrqIA(jKk1z2P13I$N75&`2t)|2%021gBLH&!KLz`qw1JV9BiveKN{jq1&vrJ*!*b|#MAY0S2F zo@Fs+NBM#7dk{_5(Z(a+@Yvtm#7#X$rxoR4|3owCHJfOOs=)?Ye9~Qro~o<3jtb7LHHct&er9otN_5j;? zW*SozR!jW?dXRx!C<}ACi{J8W;FaM(`fav{B~No=Pf8To&+6sS6l)+X8l=wtwbaqV z+07*XMhp552oT(+TME5)^CZ!`i!sznGKw02Px5`Nb0hO z9=;J-f;L9`4BGI*je&4SoFQ(1)Wz2=7d_CI16b&$wQR-PC@N@aqouDdVTzUymVCHQ zE^>_23jAox-8BB9Qx2?~x&Xr`F>=-Imi$<8iUPztx|Mk*DC9+>kG+vlU);#TGHaOB z%24RtBZ?O8$rSlWMd_lz+c17v9T=AzP}0^X+}O%ES?$=hxbS)=J3OR;R*PiQ@)BqQRGWn zlKP5L3=<@6L7Hr)oHxBT{U}p^97EZvuW|0j>BRNjLZ!I}sC?uJ7(3=J-*@eOa{2v@ zt4vkrQ+0f(`tKLex7be>`*sMuU);yE(~rr%qNfm`JBY2W?8DlkRtphpCd0{$?2d~=6r7Y)Ny3w9>V^)s4|cCw@lUiGR0I5pqpQh z^I7D|L)Jh3&DBJj&~q`-q&;}?Z9FP}zfIw%=Ro4K(L}C$NXf>ArIvliCo9tFwqyy{ z%eW7<4G$LFe+|K2L$f*i#Be%w@h>)-9Y9?TGrr<&4$Gf5g1vLKV%u%*;FIZA?As;< zHo2sO&G3AICvz_HivNAWpr!x#pS_pUPC{hjsZTpsxCCm z38wGmQ&|tIC|24r8nku|AiHc0N?B7P_Bfjf?;Z}NipC=p^85=0#A{%u{djC$8_uuz zX34hS$bnG%VvwHxfWlD|N_#S#@+z|Nvv(XjwDUFES_Z>%dwIUCc`lUpAIh0@?u8p6 zBiXRyX}EL#K`zI?0A9xj!6j=msJLK`Jwc0>O&mer?l z58vK(BOiG!!u5UW!9$UIHqL=Or5!MCyqV*YB-F)WzHXk-mcmfb7GK9w#3oR>c z>DRTlwC?Z|94Hi_LB>RMoa2klDzQL?isU#r1W(TyfCKaINk;D$#LUMQkax=hUi4vF z4?iLL)_Ssxej;15{Vjhg;WqcXaXKcql@PCbgwFqTz?~hZxRvD*FzJXI)7xGPlh)}o z&3)PYiKoHvBJm7ANMB2`zeJvMP@MpM^OuV|(qG_L|AVHtD}uc_aWBW)Vd>_`#|=pPcnCG&}QD=hrw~*o4D=lBkC#M z{f2KSLYFdo(rYLF=Oqm?(l?{kFplco*U^V-Q^|Sg2RuD0jUMedK%>hi;)fft_`WI@ z9}UxCf$ue0ris{NuyZ30cN+>@cOJy_k{tM9wv=}h9hOs*>SeCKq@3T%c&^asDw@^( z=4DX|a3|OY{rxoQd|5gwIGn`AJNwaU&3>q~jwEx{_4!ScUUF+#Jm+UDp-Y<@q3Yr& zQr+jt9l7HJ+Xv-C3#Y@RyP9E%?lkPO+g=s^-#~DyDT9u>7`(RTAP?Uc;1s)2{2Xm( zj4O=9vgPS;HRJ)t4}OOi_s^uPm5mruU`KOx<;m^r4GfoXSg}vT$Gk@vR znQe8H`uVWn<|i7@7q!vGFM7iD@Vj_>rjk&-aRaZ|M~&9Zk{7zgS=Q&=zuZghxvZyI zE-jV;d7g>Ix)pc9YTyUHmrelfnW7=s?o|NUxr2V-$}bs`?I zy(Q7I^1@}e22;lcQ(=;tn8kRQ$~?Xo&^A|N3QpOBX0Ml!!M~AAu8D_@JudNo6q3++ z&qJ8|IG0UP?!x(aU>P`V}S!6WOmaR?L ziw)IV(WL4YtZe4VE9x)T!=eyV*SzC*t?Ma_yrqth%FeMnvnSK6{wFB?B%}Rz_E4`j z8CXgN(CRf0tKuJ>cf=ppz_Z>B_oiF06Gt-GM^`1B7dDP!o3g3Ty%>61X3O)Nac$Qt>oJa9TZ@mT$SnsT^;X_#WL0z5n>Yor$9{c&z|--ctH-$DMxO(jcdl0i=G$ zU-XgQreA)_tjqG1%vWUF6X6{*$K!1GpB?TD_-8MbEDmHHXYJU7Ad&mfdIcWsj=a(~b8x#Hh^Gv&nQVMUf7%t@I9 zHib{oWr-I#=P<7ReGYoKj9|GTt2pb6eeqq~PZaIcI6m?Mx9+hmTm9D)*}guM_8wWl zpJL`cY!HjjX4oxTjtPufANca zgZZM*c2trUfuZG!FwJQ&#mgEIJ+HI5Jx)SD?|~40ue0hMzn3r5`%mPjb5Kuz0>Z6IU_|s zxT1d>POb=ouE(}AR-gi@U$XgarboTH$xY&0xrsf=l>CeZ)oWyH*2|HrQn9g6ewGVx-P8ecv{g?3!Gl6ARTkjd1=Y~Gh^U>ng1-9Kfp zt~40jrVr=pQlC-5qan?8)vQZ-NIklTh%joQ?tTb}~UL#occ1quN9&YW$4VqQlzdOEguq-XISw=e9iWOB-KaMx%%r zG;N$qgIy2e@GgdP`uqjSA1TJ{3x_9vyC5Zf0sA;Vk!5S2fZig5);=;5QV#84s*_wL ztNd=^Y1u|R`B{VbaYyjf*Z$m$--FoH1-5KM{xh;LcPAs~5je`H6lt^?9`3e4$;6%L zBIbF{+`58`qejAnrWdHSH5L?Z8<1P|c9g5V!LP2#;0jJF;Px~JKBSc3$-omZy`L?t zc3waWsy5T55i$z$Kg}O|c9aV;cp-b1+k|7gMXyWR5om2or8ADvH0AkN48E~JXm$z{ zUcN}AZLJwNu`3^UE8F7Gd=;jm83(z?3b-Tki+QyJuCOezUd(`<#6v&dU}j?s|0(%A zH&M*G^hM=I?PzyhKjg0;r-atR`4=SCC!&|vZ z%rHKUz3?!>O5=LW4Q#?VvFpYDW~|WX^b(=h{P!g9y`7Ttv$!IkQs!s50tWxrlU-O` ziS52gEJ)IWSsZVLuC+s0qV7SlXIoL)epquji`G0RJLoZ zD?G56%l>_MCM#HLMQ`6%Qb+$U)VRh@ShrPQXqluW)n@Wcw)P)gJTwg~kCt#z7o0^l zw-je?osF|k?!touhtLw=u^8AojU{ayz__H%b0Y}k@0P_;ia(|wDi+v zev|8oib`p*T?BbHSaA!pFLPJscnRU}b_$ETk)~JnC5`4tSToR^O?B(bW^R0lznWZx z&TV1XkX%Pr!?d{Il&Qj9uR1=amjdCO^~^tFB761Gll7cX!@b^IO*(~ZgznweXneU1 z4Lx3>$BcHc3|oh1Y{m0=e|bufcV^A4A32pVOCcxT9RfVxNhYL^g$#{CDxK^{ivq`C z=c}u;*t} zojsIH=T}hAho*3S?^t%EOIP@{Tv1TDQQ=sk9>MC_aU3t$Q_iq4P~LKbzw}X`?soVJ zR~Fr%{#WNhzhy=6?q)JP?IH5h(>aJUc!u$RZE>ybdRm!u4tt~w!8jv--l0vKo0&3@ zbJ~1@{_elab^jg)$2N%$gVe|TkXF%AuGK@F;a4%6hd7K7`uCX#K%F8uor za9mOtzWV8f8CI)kUZR%pFWg1;d)sLmHSq@yXa@dabt63A45@?9vXwqMBB|I6H**kay%iiFZ8b^BedRO^T9(8_YPS-W!X5l%k`n z3}b&>q4hss(tuV^G8|?}CcD%GjqmX^>3uYJ@@RjUxW803^z|U-=)RlFYI^`CcSgh7 z@$TIA9!6Lu7SCLNtWUG6q+CBoYnD`~E-c)t4RijE;phKSWM1N~;Q7C7v}?L1GdkkW z`8JdlmTrddZZ7;C$?Y~IU{}#Ee&qVY6?xL>dYPT}6 zdE{#tcUh5z9!-P!NoQb4-`89}l_OBRn4UKk+8*XbtwYvK-xdzUA&XKI0;t`s1mhU0kMY9k*CIiZ}OF z5v~SHsQtxVZ2zi5l4pOgZccA{Y_I{ce|>=6zN+AOzgBj9d+T%YSnOpB4DZiA_=YzV2n)24Ct*|d~%U>zxX?2N4s z_urdbkpJ)$v>yJ#=@B7hFj0+$ZVlohx_xPi|2(?W843lV$EfOtrdSg#0i~F~nakzYY>2tsih3UDW8z%5*k*@DjG%(p74kI$gRZW}Wo)WF6*_tj0@{(l9#lf}k+l zPgtqApZ$I-W3lskNY9s-vxlq@qGfVy-ISiJ=e#F?CHLWtwK;y(ji>u&Zi4jcHtN*x z$9(3*&{*$ADh^6z4YqS}`|F<|sXk0oZO@Upvy{%7E@k$HS7A|JPvK18m26d4Br~qR z2{SZXsp_(X#V#|VprMCB)u#k|w7uZn&Rj%0bp<>m<{eswC(^l-x!9HT5j|>3@#^|| zmUnmr>AqS(^*w8SwNCmK<|;Ap6H}|SA!f-nXN_^I?BQZ z=K>n?DI3#7zpY*1KDMvN7tVh~CLZW$qlH17z=fF$mSO?YiK&%vSo0uxPgYjZD_rI$ zmm(+kzhB;y6%^&V!E(S9|S=A;jH&we4N;0d+!Hj4R%X_)jU8lM?I$2%{?ft*G@_km-+~1^F$B5zb0Y&8daEf^Dt(of2YCinf$sH)+{JhQtgX$-3(H6zREnU z2GPWbi6mbqrRuyWQgI$644mmj<`Kh$OT80>rUw^TmSGCbF;o(i-@Sy!Wp1#XbE0Bb zjs=t+f)|?}lH(IiuEh2(o!YXEv^^O7bI;>EpCw{WW**z1In|5}(Hu@k{>~G6u54u;`zJEV`&8I5 zdLaGWvGITTMVjWM&F;P`rR;_cg5&aJw!lCh9EZnIuN_18EHm+>w0SfhScx4#~qmq+wVqO+? zEa=NV_T<2Hg+2Y#U~I|;OBQ@CgB1NkVSs}+^;bG4=0i_nk73{W>FfdSys6BZH_oF4 zo`qOy{v7;Xf3MmaI32@QPQ#hQdSKu1%Q&8L*d5nbuxGmHtQtEBD)tSh;*B>*ssAYQ z+-ptO=KIs|ZN;$QMI~hoJ&Zlnyh(qJCf9RNFQ(bcjyy(*{L3yQc4S-@_q5ds^#9rN ziSJ)T`7#2$6uRZ6~1(I@V$D-4J?g?OtcSTkcfC2Pf!|2s{# zYP5K+|75_LU+J-WWi_hrN}?;vw&5~X2W?lrkWs`Qvi%TBCe?D(zSRMv40N$HN}b&} z634&(@e~UU&(o$R@pD*uhlV^H&6X^4C*34nJTOd$l&p42`o-S>y_|7u&89S7+SP%F zV$9eL^A^B?gZchu*P(S#IzLPnjx}%Z)B6Mm!~!Mh@=r6#IKPvB*5Znz#OK_2zhnHP z>`ye!Q-fk%dz163spMi?0Gmdsf_}sf-n@+i4ecYa+AD(PP4s}CRcEqK zRpGPAdm37^f}9#3LeSOq*fgU*SECh7_a7^PeDWoXS#81ot?UD}1+CmCy}j_G;19RC zW&|5w_f2w6FC8{_#nayUQgo5r=MTX*RLD5O8up2^PRnNWK6sz{{8~ibhs_23&K+=d zZyR@L>Kaxw|Me9}?)7xxT3_uL{?{RA|$ z4HLp`RfQuhOIg+`Gx{0Rj+R4q@r(0zVa-3D;_cRn`?ge)T=|QaCvxW>Y+Aj(bh^ z|D7UVO;cuh*b?{sy(D3yIKk+}WC9ZsQcQJzXx5Us%I^H%*5F3UAQpm>JErNy6drDq!|ViPd-YWLM?{JKEN# zal4tSpm?>CtPW`lnf?+;(A&e09@d*pUxw_<^91m76}TNUcd^*0WbTmnVJ>s{aOS+o z8Whhc!}i9<+*nSAi!4g$(SU7yfr};x`|d;LmMfU6X9$Bjrm`me?=W+)CcF4k$~muB zWMgV0_+#}C2#d2M9Ra@NnYRPyN%jzY94<3e_`x;5`iYqh+1wBRWz2KqJXF!N0iEyw z7_hGlKV;{?cWwZD(axmt(E>j{K>YbFRVEjh#A`lpfMsDy?1bp2&Ux{kv(FSeM{|Qf z-A9ELeZGL(FWOR|jW7L>e@o-fM1pjxJe$(^O_F>sUd;U8KwpI=bVb3JdK}2YqMaEm zwo{p1+gidV-+ahl@2kS~ko^ZEJ({_@?@v%p;d9c8(Wj|bf6}EjIdncs^vjyP#;KJT z*>>+*j8t5Wxm9ZHuw5zKe(DB~*Ez$q8L8mV+yMnp4#AP8T)Lk;CBAT%sOg`DO-ZGIN~ zbhX2}(W1+8$P`Q)@c>V1?-eG`D8#Oev7GV@HTXQ~5g%Jw%+97*(d76*Dlv-YbU8CF z;Ll55cV7_*Q_bnk2PE@``(Pfp5tG()cx8PUhsAYWd8R10>0xOuAml%qp`D8d(Y&++Qi1Enh2gv>Z#@dIq+Q8^RVo zwIk2pyP>wwmgafhqL0<~^kdl=I+M7MglT&CJmmsz-7I1G1)lh-{~+|ab_#~6^e4%> z5kjzbD;R8AOh271h3wn%?04>PV4h+Rb8aq+GtJ^U$~7tTHpvXGiMx#d2H|!)ReWP- z#ad1`gZjN_KC-1(Mz>&&-dbz@`Dt9Ce9OrOXvasrrXScE~j4)mX~1&uDYra=sTLzw%;Y(#0U3fu}NMO zrs2Q`3=})PWVz8QMYP^jS?D**n_uPP4%#DcGV2GLLYq+?bq_cuWF&de#`uM=M2jH9iP!;kY^t&F0hX0PyMo%L)`(-O@xfaDw zxGLy$csE>@QSuksQ91qUq0Tpz?eq%?oam$;6#`s7C;Q zE~pQm5w;1lGcAP4+w1AFgEQ4l(PZsQTH*SR;S`&yN@{b);k?2*^z~{4jJ+wc1?iiq zs@DkeT2qI(S3#(3>W%5#9o+EtAXQv0qJ9Ic$kcoWZ(k>J-DBnX2Dt%@H7L<;vCm@J zw`zXL=AG;h%z!y#C%}21IGU*#$gU0^%l515vRRw^vJv~Gv{tIZ+Ap7@`B&c3o0(b? z;nhmgDxJWVJ$^tVQX9~>(I&!DPK^L1p_y>V6ol;PWfomh3hG)VDCCw%oy*X z&?pTd!=MV7T$w^>CR^Vygasv+K;F*^@SHz@ z@aiUtiq&Mlr#P{dYBi{T-+`<6qb>~n7Yg!QvRU1KhxtyG<+L$oB%62nJZy=rmP}mT z1a+HtGcV8g5_{pixbL(Q?r2RDRQCpA@1>uqXG9Pwe2>EqiwqVLz~-z{z(`+fLTN;=;SLmDk`%6mCBe{2>T=CxSl*gmi* z%S%)~_%c4}eSw=(Y7I5vj?#yFl3AY~;~;iI3vN~!$<&^T|9*;k30>W;g2$N?tW0=; zpZn#~^^NMV-z6D#XdaVwJU>s#FB3o^y`6KPT8xsT+AibTfqQ|^A&f9^#)|(?j}U3tzx`iHbp<&OxM5kVu7AdQPDY07{BrcZrY(h`V*`v ztHTpkWi7_i`N>e`GKp;vX7HKLH12OYPo7c-d&@F8(k}exZgdg z*z^HDR;tEQ}k}xN9TQ=VT zwI`QD(EB^6nImS|5`M6f{(h`8b2Ke#JHn0!U1kL*8o2#7D>>!ygzCvd*{bh<*%QYY zHmmg|yzg1S{d&5G@|D-pus@M>Fl7Wrh`EF3!RsIa&+(U^K&&{A|ozGGJlmkoeyoDE1ufsR)lkBIl4}G3?2FHfohuGS1&PHn^ zo3(uw&Gq6jO6M%d3rkp^X=ZqQh92B8_>Ye{yO1h;Pho}^lG>d?pq+hSxeNW4<+y1;q0)k6V!dE zhLv}R!sMqLAl7Fv+k+MC)}J6eIzwJadKXWfN?qKa-%pu&%4NKBJsD5lFeA&5Wngl8 z1PuOD4o`OKW1Q`2_D{vlZXV&AG5@|j*|?XE z@a}^$oS72J9M>?4~Vu#1*P7*UFQU-mL*Jxm|tfcH1u$yNtz*xf8qHuFRdc zpHJqOt;urJR=Aq`1cGL{GPeT-=umD*72#8$dt9R9(Y=|n=+{-`AMu=f94GDwOOjx8 zya^O|iq7@gWlZazUcHn~*Zzf0 z*;G0ktRggpC*iYe<79iS6WI7?TC8c`cu4K=VwR1c`G#bB?v2%ceA#QKEZOmeKwQsM*h5o5L*S7z25VsC;eHjwj%reBbGfh zP2uw{akxuwHdfd)aK*O!nBmynv~=$-P&}6?-D)jsQAaZa%fOXZ6WDoU%`J47@>Gqp`WWaaxDUjmU%{_0?fvQCd z-dyC?s@g*2QZF2_+5-2EyTGrL+`#>Fr*p{{=V7W+7Fr#%CXe%4Lc%6fW}ESX`m8>~ zyhWPqNey5B|YTw6nzlS**R>3+fzoe>aTae}+uOZ2o%R}k)o=#$*8ALu`P5IZqs zD<{o42B$WaP`~S=D9iOdZoTso`m!&)0hf*o6nhKDOE;0y-Om*F*iAC(l>*;b9YynZ zJHtSYi+rB`8Ni+Q`K`aQ_`NCRP`4?b*$)ssSLEVY6jqED%S~8@)lPV1;wRY3Wl)-e zfso+SfR{gvr1_E+jCw8Qj3!yY-~&30kMhOGox|CY=pMqh;Xg>@m99{}xRXoS zsleVGM&7+vmv!!c38(*uqVw>l>i^?7A&EqqN|8iEkrD2Bf0UB4o0^KIrY7y|l^vnX zM3R=0-1B~)22GWbGQU(pQ>CS$-}(Iq_wl&*-1GUo->=v68P$))uF|1rJBIN$wggGT z?03;`y)1s;=Yh;HPo5sw?Gs(&!=?XLpN8a&5q!<@G-y7c#d_I)z=19LwDDR5S2Z~q z0;|GU&v{d>qSRVg5YS(6+x?kSm_Lk;L@SB2t9TYUyp(edE#%Z1!q~0KNdC?%A7QC= z7+cXilOm6beBIw2*pb!@JJ6KuG)pNg#S`6X(m_soR(jnoh_88;2Yuy^@FxO(NzW;t z!_UL7leu;#OS){#6nd$#Cj&BZzN8V(pBh4yMjyG0(>f$Q%6jaNVk$h)T1w4lBOvmd z8W*9s=~!?%E002V9X(e~?VKSUTmtd_b?izTy0aj-@k;&yxN57B1P}5_C3ZK2F$Ca@4uoX z{)f}>_SE6@&&v-#&whoD<^5?$Y7P1JHN{AU6YRHt3rt>p5dJyK)8WON81a<1wrKHkT(!RaSKPtgXdPG2MMCwA=ms~T7nKOPQl6N|Ag zE=qjteu6_|Fq>kX1qHWN+4A4c{2zr=aO_9`r}x+EdiA8?R#Rqug!Y1Jn3S9c>+|NZ zc-uN`oEC)+L*r@8&Bb7KsTu+%Zjxrteu$oLKJnKYg6V0t31khr%X|ADqo~cs7$&1c z&*Jr{*Wtdja%BU?c_~1Vnz);Ob`i!Zno-+?!E8(PF)nScIh&+X3j^X1mhoz|U49bt zdD9A4KRo0+N1n$WXJ5jq2M@7$v*_=+W(Kk;?RfJ(gT4mHM=Pib!u*N!H^Pb^_UH^- z@>YcfywZfNxhiaN#{jlu$`9Ui^czaq{|06Csv+gYaMI7WJqc6x`ipvMv1rR^8F;RZpV zFwBXSjEfYx>aWpFdm4M>v6;C{(`Jf=?NF9I5v~14frej%$Y&l#Q!k#z9e)O5!sKMy z*jdhXiTmMPO?#3r8OAvOx7_yn82;lHB_{tu9cJ%-h9Rf^L%o2xxZ~VNikRpGYxDPE zGl^DQ2;u_|{Sjn)!9jV;^h z!x(+|r#Fy=<$ME)b`<=sA5S;kn_%+Gb$p+pe&iEdO}(!AVzuZzJvA*F4?ektf#(D1 zoRTvupz9zMK43fg?i0C*Md01?9KHsbQQ+3+7(?dl&Q24SV>OqOyC*ie;n|rdF!c?Rf07%`-@gc~x89)cnX~Dxg1E2Uy+!lOVoTL%-E^-|?JdUM%YAGaFF@;_Eu#4I53nN#j zn@~B&k@5FN2?4kAF|Mr+l1-IxduAG}@(cq*^|jVxZe+37hojkgB`MZCKTf;HHSosq z;dFeg7xnrVgvxfnt;*M7%KyGVbkrsIP?FEA%U6TE|8HtEnN2k_!ysqr8>aJ3oxM($ zu-hY#O5V)Y!j9y%P~rcA4k{mHJ{?XhcgJ*gt1OM=uO(neM>R7&^d25o7qYs8D=F{r zU>bJOfN=^E3cVL3GQzL$3+hUw4WrYU-PFO>m4}|PxmkCZ#W)l8;K*fGQXRrSeyBkU zLu1J|Oi|cbb_{pkjj>*Mi6fuHB20N?E?C^%LbrSTLFT`3ksUddA~kElH#Et5T~-KW ztPt4Ev8IA;XcbaK|Pyud2z8{y6Ljvi#W9nK`(ZG%9Qe!|Jz11LPVVSW5( zvt_y4xi-=Mv3IB}`!i627K>FNuGQdtU49sO9-qfAy5 zy^&qqeE~Y9QCNRq3=N+zVO|YV?#9wz5L^EUj7%$Wv*&qwbnqm7ylBpySfov33-02! z4Tg9nDuo~XV;?w#?WCR`1NqAnZb|1n*o>8FQ{m+rH;A+S&Gp7*7!w!IU1Pj1boc6zpE5TG^i zqR7roz)NB;C|sq7AK!Nvo4G@Voums?wSuVg-6FbuUx_}{O{FPglj!nRHTpBn9nB?c znZqnyEFO^vKkr9_$N25E{_P92t}>z5I(ykUxp>K8nJX|p;h3~!cnJ3TATrqJm5UD2 zZhlL7EW0B1+Y>%$F!epb(Ej8LtyugIbu7F2^OdG}e47e>>pThhk}&u^%>~*;uIbe< zL%MJ&XqUXi@_H35|bXZG`T2w4Q@@%>IHgH!QGn5yf>7XRB!;1DPbyUqkX_21+abP3|~ zoL zr@X_DzT&Rq>ji90Z=s|}6LJb!Kqnt&(f#xW2;Kb!cJz}6m#1g>wOKP^b3i{-c(s$J zj|!k^P!3-&PvTz`1ku?mN9op{>5vmEfMLcW=&I|7<^P#cX2%11Cfz|s>pCdnb0cqM zu^OD_r$WUpdnh*O1!w1OfI0Mric$-!kI-Y!3VWn;bKl~h6FSVoXBs6J4HlkN zZJ;q@M91=QF_Rz+#tm8aZ2S$zKSVzY{9!7sHQETb>J zM#!$x#_xr?;9JxUn$6bqCcTK_U9&LrPbW3Wyo5HNMWiLXpNwwchzXK#k7Gc_W&EX#GNU7X5p^)Auu_; z7KXS@U{=$vvh`nL*lN$K@OR%ku+bXM-?fW}hle!jlBYk+xI3M{+t5mDg#NVXekUtU z)Dv89>hjBt-ob$j`zdd&I*q>VOGiuf@Iz1#PRz6tzm<1rwJie!GQ?g`v@w$_90OKg z6KKZFVy4lMMrSXilFz>F5)19!+~dLi;=J?~rk>72m3cO-@4%aU*tTWdE;s;M0~B#Y zKnSc@e2<;1P@o3wc&fE8#tjvXaQWFc?&g3oTtAs-7BEj_OE zQnemu4O>pWt3Tu0ZB=M*F_Fo2T4HItG8I&-vHk7e@k#42L1~H+RccsswcEFsxG!?y zMAIQ*Lqm)c8NSxGf}QUi$+~FVZn~` zXu-72+#8ELHXRH3zey%&dSNQFT^PwN9&aT%z2yyhiM+^(E#s*n#F9;PvlN7;G$7IsjbL#BqD6xaIJ_W9^_xxI~ z_jo>OHH&x|E-YGUE6n_A3foI9#16tunANa@>AMEg!C8)^bUTbZ_MRm9!{#*NU>xYj zOrXNo=0yFhG3Z<*^&Ge?^$+hWmc$iBCwd-J$S|PaiaF@Ie?E)6mIlsQ7VJw!2CV3- zL^Ah{$gM3AgG}bayK(*54%H~y#zTv$_o^Pai(+y%7eB+ZP__~U`_24)v zq{Wb?@-aw@yv#1J8|?Omf4HXZI@`El61@C4l*VaI6JFa5#okd)U^w56W-0U%hKxy} z0{0Ott}>O@YYb$uB8PVHXrKwdf1%R;rL3=U9_@F3fQtXZ*$+8wPKbJf6;VaBtCtk> z4~HO&GbTDR126W?=GV#~oGh46OYD8=QkyyMd%TdT=I^8TNLkeG&MknSb z^Ir?&al@JkAjS?U&v76HW{lyeXN1tD(VO*(xIxktgK>aTf7qPWi7I91a8txP+LmKN zi>ns1zk9POs=9<;DC{Q-`z-!(#wl36Rh~vw_J^HiZ}>suE$QDjC6-jX5QAPnL7xN9 z(Q5Bb+?8j@)orRgNT_LiU9QS1TeCYnMlnrwl!ZhRRXhbA|`zdA2B(w&?7U5o7ubT`#!Z%O56J)M35Eh^xIM#UGLm zzW=-9eD=ZqtZ0f4CM2Ff?7)nyc8R?2tBcqiZyngswgcwHCs26XSUOi; zM4nR|V4+kK{uag15s@d}bh#W?hymF*k5BNvQzyV4^5-8#M!?w>)8XJd4WaQ{H7M-( zCH+rJlaAFUNS4gg;deeP=*@vJ}ni#Kimvr1~_^Be9}DB{O`;jHn5Ci6d4g%uyCviKQURAKcJt>Vt& z(3{P8VSto>Huf_V7S5oHzP;(ny0e^;`VY!Dmqb@p^kIUnKQ|IuD-%h5rCIau%lMJTIrlYsW zTYfE``JyX2X<-M?^-?a9%#J$_y07+g@4}r~^6JC^sR~hU&3xi+T&HB3D*YT+N0Y`3$B_;W+KLIDdCYlJaIN*=%QT%DMOo7ca@hv*&|o zwfhrrO!9*};!M`QbS`jNGgz!R@9cF$5ATb8w9GH3V1@HFso4=RtI+00YV+l+SN+kU z-!;==wAgdIx-x*DzFL3E!w6w6fwFj(V%W zsU5r`-SKoQy~r;E+b6|nvvCW`Iv-`H@}{t|vj-@8Kr%hveUL_NF=BP!-jSrUg1OiI zWo&2(8S)=6yZsMno_hk5ZQCH~z#wM1`ZqibSPJD`GD4#MT*;~6u_E7f9O;frp}^p; z*grB8=9&){*zPVmG-Vq<<$3RB-rTLsvhpvI2);UAh6Y0Ah;BoVBw2?m!plSdQ9g;El{!P|cTE?f1i6GaY)BL^UUf?yWo_}H| zG|(Xc+EW!k}Fq1 z`1l@vZMp6w-ioy$7waG`r@winT4(J^!WV}k6&SYjekE%5jp9vpQf|Z3;NOBe#@wH zZX~sZrJ$BYID9j?NIlgzFlmD|dod>+{`P*u%dZ16jV&Pe7$E3d)`Ou50#7gMP z>^yF<_H>wSGMGY-idnapNt94KoVsJMaY(Q-`N zN|i~i3i%PYd7k)C7S-wkPsDlEpX@fScZ57naBYS&vB%iw-%8XxQcf^-*nx5$V$bD~ z5nF#JlUb+VReeP*%W;io9-$$nO_(LiGI*PAlh$MLpbs?mDMII3HpL|+Gw7D@(u1nc(&F#c{F zL@av_ZT5y3Hn)KKHf-SeAygC-%oOfdu-jeN*!tmtENHc^ zaLjZDxlb`;H|{=Ub+h-2`}DCGmX<~Kw?ELWdw;;=O&uJ*r6IJ-?I#(no6xX!n9$Q| zMo-3W6FP1o)=9egv#(dOxz}cZ&&ki2@h=R2d|xVLWG|up*KgzCvBuPR;{dYmiRffKT2!(x|hSDJz z9;(RxEYM;JMK^KI@B_5$g*`p{vW`4|I>QV7YH7a6@eX|34BEr4(IRItYteTizgjOF zM~2q(rkxVvDn-_F)CxE-a6i@7h4J>$?J%>`02Yl?A@hT4xS|Pp?7R0WYCBWJO%H35 z+Hrn@jJB$vCf|eh;^#XvE(lU4cra7H$z-C_i{Ek~8V<*cEaBnh@Vr($zkE?AkH%Y^ zLcA?|I`@-g@@!{N8f?hU?uw)O@JuooX^UeA$q6n)eo>i93LP#KNdKNO#r2$rH`~vV zoW%lARqHLaF)|Xa)}#v!v7>}bOXg74$gz}MGMl(^4OsVnF01wmVHPi}X#Vb7T&iRe z`#jr_Ee;&YW;YI^&fE6FryxDyz~_&2R{On}qrT0low$mtv>S1D`+L4;Y%IOH5`jPK zGcoD_0QbdFa5M@KQs%ye>O`^BwH7j#*l_*8z-S3A7%QjzQLzLq=kD2N?+ zX~^yv9^y|g)Z^`yXHk&G1}uF!6HPDorB7)Rb}DiiJr{1FTE`5yahPLgcgy3bIeW45 z={!ljj_6t6U`^M{Hj;7IHL@EcE8P2OEZm*Ciyau;2Q78_;v2nms9U^|8-2l8vSENP zhIGwvhQ~`;?PtD z81M2QS*&fP7lD0QoQkY<>A{6G+IbhQQ4r@GA|I~#WGLvw3}t0*gM^2x-6`jZfANp~ zA}>V4nZic(w^q&g3ey!oOLm{?!9(sx@ZFLZOs4k`;d1v>eE-1$HrAfv{nmbj9%C^p z@HGvq%i{6XQyJE36vUG6UBKg-*`iD7uJpHu0*hN02n*lKNeqYfljd z@!b%y7t!wvx%LmmGh@$UuHs>8GYtb3(>rkb$Yf^G|1`dizs`MIcm~##4nWOdUwZxM z1DV~3!1&`KcwLBvsmD*EjA0Q*8sB1-d$sroy998yjVJx-43_VTgG{FssndOD=(?TA zWgJhIuAh_xveTRS+U#I54D{ss&z%mV@^?ezk^mYqZXQYm3+li8DR*d7rBu~B0`@-= z`$d6n+`>Ik?8g98>9J$4Ay?={O`T_{cgAIO`y$T8T6L-PqcQ!NA$F!kjzd7x#u8uY zd$g6kh~sU0^Xk1$@$sG$c{ful_*QbBT^+1P+r!nwovqdM=%I3bJ!o5P_}c86spq$_=5?pe1D+}-go=5(pmD7wBL@jQf)a+w7&obh69BS z^=dS{PL95R$|1k{7DzL$h28xZi=DqsbXH1ewp}w~W9af;-tr_4Pp~7PC z_9gb;H?FyMHEG1ylhceO+I6XiYmzp>)cvkF-g_LY+#k+&{ZnC~lf!BDig@xm#lzJ_ zV&-4fUGzm?5g z*odpYZey2^CF7+n8}Ojg61Ho50Dt0UjpVh46{}AWIN$I_elh>1q^44l%GOV!>s<@b zC9e|Mn}(MPeAypJgf4Z)hi`_uJM zIj9yp75r(9>l`{D+gmt(dnxn%dI|ie7-4gwFI{-vK=fiSRUW?p=BXN#V%I`f)~=%0 z&Ar7LQ7S*o*Nhny^kvPxw!@(tdq`7#9uC(trT-SmSr;`~&=uoID$Giwg8mZDC9XeP zb!9pC#r>(IZ%-SjGf#$i4pqEA2Y>wS8>8Kgb%>4Hsrt>2VUf3_; zV%%I<&D^Ezcc&%u8gc{+U1jj6!xPlMXG6~7J-PbGIAPz`dNTe7Y?;3ywcHqBeQe$n zK10i#RXU%cp(EFj&*S^d%jY6HVHE}&-f!d#db^Ql?lA1mP^ZSpsdW6=9M#&)c4gDnxO&*UC7T=x1P|gaTes#wjJsZRN(*v1;Ln~MW07Bv8^+Dv(!nk>_4+cNS0=@;eiR%aYUPb z1h#O4$ps6BYhwDX(@-_&Hhfy?D$ZwuS>C8o!j{!-Y-qU_>%Qs3wtMbjyLUQ4=He*G zU*yU3ql{^r%@9iUwP)r7w zhs-`qA@dd=^4yx1xj5oxt1J+2(k16zT6pk&Q&KB%BMMcxE)8vFRtbVjJ^Sd`5uhjlU>y7{5(%?|q zTs{aMRotL^`)5+(-7}Ku?`8;1ff*%_Tt%;#dNw&^B;f7Ux@_dIx02Ts$|C;b;l`N& zkV_l~P4hD`Te6F5g%R|~?<>8PCPGbYI}~=_rQoVdRCZt7BiKFQYc-BgU!99M&&)}f z>e-D8CKqDZlvKKNGYo8h*^8Na!jTST7?u}{5|cD+_ZArqJ)?1XOAU2S+s|z0f8wKV zmZD6N9^2P|g!C_yt4Yhs--^)lxM!x?m9;>1{^7mZB&5 z?>zi{>^PaaUWYSNk4Vl{$58MwWosSlv&6Pupn1yzXjJ()_9QZu-Y9Hg>aqQp>_BVD zRuK8wEh>WRyExjrsSEr+0UlS1;k{E6C6#NFYLaiy>dL3;XC8d;$ft2lPOWJ$gC^C9IIy{%f9SMr`Uv>+4Wov`l;wb#c zG0Esbtq{6O)_RYrJb5*Y5Jvtf!XGb^XiA^SRGHWWf9@)CwIl z`OnZ`F0*Fy;ifNPgK!iZ9yY_e?0Ecf-wjVxRpHXCb>gh|C$#jHV-u|nsk|`&8*ilW zdrpmD1INW<-77IiIPbAAYEwGNHds;3!WeXZBJz~1i!7ZLvQTetKl0cX3Qk`cyLn6> z=F0_3hkQRlOXZfssw@xi3fxLTOT@djkPcGR(zZ^`}_oA$kIkfE@fzj7OAz-Wp=%sF=C(S>3%M+dK;oNO3 zHSZNnnq3B)EV>edxBTlQElb5k%#~&NR`>7h>i|Z)A>h(sy4mwr~#0ZPcaQ z_Aom6bsI)!UWR@}a%lf53ys7qU7+O!EP8wsqx&3TUtA7=-1)19FBwCZ%BS9+?S@fJ|vx1SwY3Zh7^6?aVxZslG%C}iNV~XyyK5SIDX~}{4hm^ z?o81TyKjrx*lHCDKQ)RxN3MX)L4M%>eF!C{)ro!32e4;u1#gyVfVbM+aLqasex{$X za9>H>LCsPpbstTpYcc|6#2zF0A*Qr4>mV)8-M}?@3=ovxjKWgZm(4XIkZhR@R~8RN zn=Qkj+w~}$ylgH(*9EqHPa-?NCKvZ#3*-hBOv3X^`cS|#6XAFHI-z-BzA(N>gY6b` zg_5_eR!vV|fl6Enw%*=DuezJSKGuQX&=M*{*Qp6hhXvA?`&De&x)QYVF~4AU4rJXUUSi1d9NfP&jW#OC3dcTI z(K5|Ewsf%-IL)3+%lmuKD*aV-Y)1_Hec&8Fe%%W0(R?w_7n#8Cs}bkb>j?aM`$>I1 zN^r@YWw`r;DT}|FFKHWSK>Z^wp{>gYF5|ft`{5o5`_|2X&4MxxxiCU-mJk^hzCkPZ zbHXQ$btE;(!{DJ)sGu$qa#FgvX|H=RLs#{cTwPf2(;m(FzNzBx3XNfK`1J`Y{8X{qCmWfXCRXM?!9Z;_a5uOL|U>d(ykXwlUvMO1Qk zEmBbiH*U^VzL7f#hcxZj>hg80%e#P^aQGHJ+WQiZoJ6!6Bl3=0RhjTY3JzC(F}I0z z;HMD|6Kxw{O|cc(XFTMs-z-GS5fv!=S)c#9G#;Yd`*1m@t{YQxO~P?!oT$o&xK)giznX8!1O+73}G4P6x# zJ?Eu~IhmKZrac#zysCu}F?z7EIiJcDPLbCd6Cpgn4?4I0ks901q`$p#NO{aNtSI5x z@~b;YL!^O(^coK(PD*UP!&`_O9!BRL7sKXOHE!4S8s@zUy?%-L+}qD7(NaP;I8_$(S(XL*&!$9=sX|3r1M3wh-t{`l zVU6@8wT{ds|7nYbjMIpDzFNXMgTBHNz0GWvBm-Yh%;rqB)KN)(fiP_OU*7(G3)`hP z$a-wjEOzl|5nFL^g`{|KE%h6^9gZ2Bk~{tr zId~s~)PmrNVF%o|if31gBYzrICP(fD zC;0(4YGJwMAez>38h^dr!nSK}VI#vA2^!9!)Mux>5O#Vzt-Y>J^UgSu+UyB}k^<7# zH}=B#KIb5M`*5oPsr~014aLdV5u9XVo@2Jq#OCyC;^H3>Vy{hC_+F`Ppd5Gpt4dmCJ7{roA zFW(Ob3$mQV1I_=-$va+!5AKiYPj{9eGfZD__VnU&tgG0|MQ2&p;iJfTlw$DL8XOUN znzl4dWxM>EsL6B_Q&?G4b|9ZL{X2R}!n6a(4%}_l=^s`5eMYH@PaOSH&pTBb!WmgTN zMI#5%AGxcj?(8GYQAme7`U%`Ak5c{}r%QUeU*PRh#HVYu=|8_CeC^35+&eP?H^~2l zVUpMIbwnI)Ht8jdk)0q^Z#HJGJq@TXW{EDW-GEmN23r?L_oW9bS}8Zbmvmj}eJLF~ z#Se1a!cC5N0V8)wAf|f{9sItYlG;l6dN)6M(kGaDm={(2cY;eF6a`_ZO-Az%3S*j8 zgk-;J*cHBo*B&jUu7`hUM9xC+`TiJ%-n%IHLxSajGut>9W(`a~8Zwrs(}U|%;B5Ro za-X_~j||()^5j3mj5&iSZ&m|dyD^C!bsE4&8u$bc4~KZf;ZGxA2WWAvj5=3vA(7P&Hi6hIyUDFvDeRvT-~#3EyDb-|G+t zCc~olulVPC!y(Ue1S|Nq8H{_62SM+LvA}(VPGU#EbN6Hx_`464@reTDb#EZ zXEXa9=biVCp>5`aq+WZ1h?YCin(82syWb(%^PnDM2AJ@^?$%=8cntW5uEgseDJ*!< zbeOI`l9#m-eG?Bn+4kcaU~Q6tu2Tc)$DTO?fAj|(ESkWA)>*MvY3D%mcLA!6n9gJ_ zS;BvJ7PI1)wp3$t7h59!QCC19tse6qE!_DGw`tjvUeX`|A`{7W(I{y`TL-u9uYkw@ z){;zj`liHMx%UZ4ZUP z-C1zCC4kk;_Q2E!pD=&X4^VNa#%?!%I67)OU6)rN?YYxHzDE~SpYmAe90pD+ztiXg zkBGBB57o-2xzmHZ@tCtc;mlyk%8k82M(Y;#+vx~vc3EM3x(qvz6^eTwUP8l83ucl0 z6Q1rLBD}k{7MPzcGO-N5x~_q8qSWV+*$?DpGV7 z+q2N8zc|HcRW>g67pS(YQhjtj9Gb8JJN@g?zHNZ8T{D&4JhF)NBWKdY1Bq_}xN zAdXx|s(_;CL5=XZOAn^Hvx}pnxgiEI!qSJ>S1} zS&O+lLqnRoTM@S_IuUmxn00qn!gl#?_^-wqEpM6P=;2G4lgMZ{_Z}%!oG&M%@>(2x zdL+}z=z{TuTe*$1ylH5`2+}tpmQ?r{$7wdB&C9 zDA*~N;I)=W@qA?_Y`)M$AAIwL$l#N(ygXhq#Ctoq$p^t2xnum~haBsBqz;o3N765y z0BX-kpy}_9qJDBFi}W7O0>GPBvbE&8Zp4zy-yN*rVl7g{Xs+84zUaaIyDS1frp~Vc?`r=@UZ%oBwy>9b2 z?zv#O%pY!~ow)Owu!42ijKyE8mekLrq@ ztK|4y(Mn=>?>sD7z7_{eRilL_W>EiRK9_X!5I%b{l6LRDOJkplK6jVZyxzbs+-TR= z;OKoDCiZOSY99Q;jj1sp9Uj1qY+sGDKBowsQwylRRsoC0OvSucDWv2T4<+Y9u~EAX zTP^L;;6nz5j5$J{_D4ahj|I0{+9UA^Er*(oA-tvUY$}-326vj(>1g95GgZi2xSqcJ&f~IMo4}@@3XHaol4NN{@lJ-GoWA%hIh#FW z<~I~+`4~B2gn=z)FSv~7v<`t>^;lto=`HYV)rNI({e|)Y3;5JLYj$*EJ{I-lL9BiZ zT-x0s=Fv5w!qtj0m)YZsKd1Sjohxz8*$*%@>m)k&T?r5B_tN}|TR33jE*O`-ok{-6 zky3XbVNu0k?&z)e(r-#$bo$n2`s_KKlhqu|Y!U)!aIyk@HH&0V6C>I71BNVib2k4f z-kes&=+RwC4T}zbhojyfMgPqQpwi_ut1eA}FMUF+p3MFaG*-RF!9!l+zR3>s-yjo+ z+BlX1G!nsgLjeEalOD`Bik0m8)r8Wg`TX2eCFXH*4*w+4m^>?!QT9PHKfvuHpD^Sy zZ+iDPC@=Xe_TxLjRANinsiPs*uaub^XK|Ap7m#V{z-(VwC>cN~7W@2`0zBK`)1FOh4I&&0pjwo`WN4xTyjE)xHjb zr#gfGz5@qyH~IR-zv<2DSHP)W$Gk|`)dYvd^(oe4T8dZ1^oA-i)xL47Lp?tF>53d$_AH5L}Mm%t>uR5-oa54B!gwK@`>22;-mb6QX4LSoJg zvVFSpx@Tbm50iJ2VN?=CtmNo~uCZjWnjW*dx0nvvO`>0KPvhL$MG%m5pSq&0NXcOi z6-Cyu6XPDi%hq>n-?mZM@55252vEYLuy-(e-ej6?8_IbFtfjxHJp5Q)hqLn=p`mpL z6q&r|`JlOU(!2%oi^ME~V>qSlXqR-{NPwV^9&BM|6PEt_ggPM;+1mqWK~NJNnMuIP z=5*kJljBIHH4|HIo)cM1+t}uZ6WK@2v-qs(lsNM#f~R3cSXOaFxcXANWaWXa(4&2a zPK6n>7ngH!MOZ0KD|iQ|+}tttPCBOBMbI+8f7~vSTQ|1Un(dU6W1K-g+<7mBo{_;g ztiqTweh$RY-MLWkydP8X+Dtw}Y%tkhk&WIog0nmy%^z7hl3L?rneJ+9*!0npj;Y;* z_PvqU>|BYwn$P4a6LvZg6ZF;MH1)smjKxvBu z?we2pF6th1sd)jLzblFRYSawMuX@vp=5$z-n#uJHehugQtf8S=#;{sfD`)Lv?X&Oe`hGv3Gbsl_Jt`LWe`%mKx3sw0 zw%73aw;0WM?#fONn?PjsSmxfCnMC#9X;!wKhtcYKGFxTppm()4=_=@itZVxrkv|(r zdWZ3!%W+t%@_=cpjf8$EAQRa@cJz`Qt}oidtg$L#$V_eI3d^Z;##TX(oef>FLdPz7 z5`Rt4NpX#9pP=cE>ujIR1TcD6gk}%spyzcdJTUzZt>^oEJBJ$}ButuE)TFU}XO&QS zM*w|(>M@n;xPTk|C*hQnX0+B|FK*9yOrv-&&r}mHI%oAKNSOx`w)l$Y5<<+@O!Iao(i50}&FzUrh==m2Ar z@eI%B67Sv)>s&;!_5C2qrfi{4KO1Awx&nIfX)WqC zanSsz7s_%@L;7I}tQY4w_UnR(4u1_~v(h0?x)8N8zF^~xIoS5%Drt9rP9yG4!At!A zjc3>vNO(3vWiTK1&Mc&B-KEJKw|0ELqlZd53=-k!VJ7$Y9_+U2$G`e5sFD(efwiw` zfcg@qzxywJ>5zpDKinW}%>?G>*niYGN1eS9$PoF#NM`x7*_^oQRJ8ah3dy$j7|&s2 z*1j!)ew{Lr`xI!vm1Zj87c&{Ma;*b#Q|)D&qw0x)`cVkZl%VT%6Y+VqJ{R|QBIj{% zl&DT9!pB}cnBtrWGP~bX&BLGBmBJ@+X&uimzMsgg_@&I5g`VIJaGj{&a}BTET#AkM z&Frp76Xwa!>#%ssYp8c>fQ?XqElZTi@bC{d%Jn5$n14pX?8TVzaok4U{jvJLAmXrD z8Xp~yCFLQbIRExptnzBab)SmqZ*7LWp86h_hF3CI9JCs5R!ZZFC!dJ-0}kH4|BTXK z8}aWB9j@=9Hz#wifM+Up(ht*TadDAfP@(M!sK0X+WSJ*nTUrPHn?yKoOP;AUe1VE5 z#*$-tSIKnw2Q;xP9`}xF(Xi#)>4C1Xw&B(f>2|AIAex(t-|fOl@M&){zN3OsgULky zmJfaKk7w--ttFGlb=vjt3ZJ9YM8j|G^g-ohZvTmVYCN|B(C7-Bt6GQdV{5Q_{xWVc zWmp5VD%fPMitZ;WiSI!vOlNO%83wfw!cK&wt+CjDqZ)hOCDY9>_d~SRc09T53>ALg zfbw6&@j;3!O*Akht4ju`T}li~$F}lw>rcdg*o355E0eAP7k($L4QdzV&{cB?q=T;G z?EYYM)>k4%>U{ru`x!`kjm+qh2)ZTw0xsHqnttB>l9^RK4@?ZllUKwG1Mg~JyT)7~86h zH?(xgosd@AlCFY>TyD~?VmCbAeFx>g#ggwHLL_S9LNai-fQ{RC)An}kQTi)|z|s_+ zWjSPsO1nj9#RV<=@3}GYQdb}+H2EBM-4`}%-FQ3bT*6vu$1odq5_0y8Fqyfw5n2jQ zLBq9hHtVYco|xYbsnbH}eQ^V_K~O`JXZBG>FIsR;DT2tW*TUM-G_v1 zfx;9wIH!1$Y*ef#;OPf_IL0@7OW0bvs>{p6ji^cK66=&|^ejU40dppR@fg-z! zvxk_M7Vb>Ej~wAviIHumX7SfqlMGH4VmyQ^aoNW(JDtx)wC*NHn~sm=jD+>b)LBYk zFZ`95RJAf8w;UMl%@(lk%K&|}zLQ4t;)j zI%qK}uHQ~HR32gZa~J%mT#Glq>60Hj?$K)-`l(sn1Ny-DDDXT#w0a^37d`k~McxE5 zw9FDVIBfy;iUx_@E=u$?YnY>F8|g%zA!Fz&PaWNSV0rm+NEWK6c14P$X@NDpctMf; z$|Eq#UX%1Ws`D}-CvtY74AE-#Wd9~4p!?2fvI0|}VA)gp<(d&x-kppO{2#;NC+2AS zr3wN{0?6xS-=Vv+M6gKhFFgNkL>g}v!n!Mk%)#xu$l6_>8SRzRiQP_noF2RqRT&J-k-Yw z0b7^Du9r9IsOkmA%TSa4o^T6WJVY_6{1Ce;vlI2g%ivV>SFGGyh}!Rh$X&i`lN{Fo z%hT>bx7-r4y5=lBz;mY9@p6*1xOG)(_1iT-}C>9UFr4AZ{=^|~(wYTsPo z(1H1M_mh6cWV;;9edCX5wLjpzHKm`XTHrTh6_O>l4;yDpB5`j})_8Qa*wLr;4jm*vyqqA2m27P889#GoGd$)$^rT8#TFeej_|KKxGzD}fM zI{(?6dB@}yx8m^l%M|N$Sm{JJre3s(y|Ba-cgRejCx^tC@!fU!@7i@Zbod8*-SGgX z#^r+ThwC`%D#00-@%L5<8ECx$c==N+%A75xPZ(2j$;p&Vn7)8kF1}7J$N}tc*~vcs zI+d({5W`*Dz-Od*f74=r4cnp(_vq2lJlcKB7@KPg=<%SdMAfj69F1t8Sxqa6bVocL zfA~7YuKPgm&eg zZKTa$F8%%fG1_g|hYf8rX>WEgS()EKU+jEJ(|v!!eRB~a?hu6Wr?&g5~kvR z24=L!Z3fl+p}{Gi*We}<|3!A;5PjoO3C*E}5IT~9lDhf0nQg++U`_5tTqn+N>4Was z8@XEJo!o!BRJiv4cB10kBetpI?-8FmLlV*$h2I$`oEYLotc(h%;G8s;1s;QDW-PT$ zyNvb5+9dSMWODm@AeUo!4~lpe^wG!$_-|SjxLquuW!v;|M{-c4Kx0pZjQOxA^0vaS12@i=2`H(0~>XR!-UC;p7bY&uJ*~)5V zAo!jQhb1vZ&^GcKsyk$Gm*Z*_*Z0SlLz`;IW3n?51>>4aTszayR`fE{`uaS)!DV7V&%kP4KFGJpM@>!;QaUg{^68P~znV zDEKlH3}vIS=ZFMq_(Zdta+1kH+c-LDZ4V7M_$#oOS4K)^$YFukb5wL!#GqSlxY6Yd z-Fmu|zVK^h<8SxFN)4U`P@KYC(cgf3`~2Xq=nmp;aDz@RnaOp%{)4*xDmW^eg4WXt z*;`~TsS&TogqO~=HtiZQ9kY$pskNYE{(kPn%?L(5RY7oe?n0cz?@~_eYNKiTDRgjy zHqH2c7*ZGcLRiHYs`Kpuee^1Y^;SIxm%QIlEuR%M_4j=o+hxI>-f|svqdQ@A<$7k| z<0SIQXckN!sfJ@UdgQ>998_x<#&PS4@Ux9N^~x%@U3%mcww?;D?dJz_VyL8k*RD$wVRgeHdOy5Td8u$Aivy4^X;*SV(%a`n+pO$!W6?^A!I_FZ@UAiZ6sOS!< zEz=@nB@dwU(bM?RDhQ8XTnyIBb~94C^Vq%9M;Mz){p?B^Q;^Z|V0;uN6WI`V;`3w{ z@15^t-KU43)x027ES`+CFAk;e8lo}YKb-Gc1dkz0M%KYyu2)d2SB>AIHnCd-Q#p^N4{+&m z5iVvV1qIp8*cq}61}(Ip%jg4bxZ+4xCeHycp536|RbVrudQu>yv6RZ@3v&mphH+0p z78xI|&i+x5qTzSP;mu9U1Ro^6L1aW0??-8au1h9d$Nj~e(Zgom?K_9uobwd6d{ZK| z`AUN2cmLt03r)0VKX9ddPG~20kPcS1!Tm;v)w+(I_i)^`cB-)GlaN9 zSJ+@@C3<|!QT*_;nJzxqfK}7SgY2&u5Yvl*Kb|UNp1>H}7fYef+6CxWV$Jy1aWwaM z26-!T6Ha|~r!tBH?Ec7igjSf6rCpYslaenEzc@iB?IqmCUuS5zv@Y+TxP+@?HE

;r;o$&>bZ03)fBc6vtzLL@DMx)SnbOTyvhmT;#~`w=QE<6R zhAb1Vf{GiGWclrX?3VLiV3W2Zl;@j5_Fqxj(oxQw;hlo2eZ81IHj%w?x` z@dF@@1=xRKH4zq>NCu@nXwQ5nG?*}nsOs~b466ee(ou^=+oxmOkPh2AIFoCvh(yEB zJiF2BGD@fq((&su;ny>sF?6X%kdbQNSev|y2_L@AbSw^F^sYRBLB)rteqa`7aAiA= z32vnxOJvCUBk!Rq?FcB=1d@dX(xl|iK74y5h6MXQ!*~UGE_mianxY|w0s8#2L_jC5 zFU}>C{~aMalUS-HYfsMcbHHW7VkB2}1SZvOX{*^uxB z{&mKa4xf0M^hRY zv3TM;s%3K%jaJm+aPbP(=0*~kYkHbw?Uf=tI2?BJyJpT-3>SR64x4U?qvw_o6csW? zci~FDe_O-)ldrU$wdSHG*)d7VIpp}}2@ocj2F2Nur2mWs7x2~>ZQa*kk?d0@RJ4no zDf*r2v7&HbUnm})?kuc*JH(x0T^$xTbr zzK_qnw@&6Zd1NwOoAYSSqbz34?`?gBhDVm*yhB3x zPcehoHMK)*gf}|!R#D7Zf@G~Um!|%f-VlUfz3LfE*{vWDd8!IstLlhx`6(Rz z*bcWQs1g+oZSL5C0y>XU5cV=cS0`(F>(oO?s^%H`0nfnd!8NM8F%IZCecP1LKKydP zi~S&eo75^KV$>2F`e=)z;5=HvPv;$&A8U#$#$<7O^d;fU?ywLkmg zg)Ke*Bn)|xJK22Y4ai2tk|LQM7_uRm%+<6b`PtIM=8h7o+@FI3u!WeW<-A1c%zKx~ zMDyPt*ixDZ6>j5b^T|_;r$ijNv)Yz-KZ-OJbXud-Ju#frElCRVnrO0EIauTkLGZvh zx>$4yxvG1aot}P-Z5*_*37q^0Ms&-eDxi&yE{P)wHi4|~cn{QN=itSmD>RGWdr58F z$GB!)f?y+k=$+?6u3v4YFKV{pwai&0HZGkeeA2{-i@`YM)*o`y_A$&pPzn>LnA=D+ z&nK6BwTYgU1B#*w*?(LC)ggfvF7udPJsf*CHjl8eRDGn_cD629}s zpIa6AJ*~|KNn#Y;Lvs{t;pEq4)Rgo=TGn4$`A?0TxpFHP(WA^!rVDb{XApU*vE=w1 z%4!ryqLlm`4D?$`o=bXw-tarPn{xwpS4A_*(^l~_>q`*%&5-F^I+r}9F4#QQnBLp2 ziV~JasI@@`HsTcWSTLTPP?L?5}ZN|S3-(kZQIj+Tn?}HZ? zkxe$rxL=WXC+kS@oilZ`FX@4Wn~pF^L7RwDiz(A2e-QmuB)P@H1ffC>Ka3ONPVA4R zHGkGI8+?w?mqVg>`}7#tmpMqCBJ`jqZyF74^2QAr9gY2dH(;>Fil~g9#_1wssg<@l z$-QyIcA4Wm=&kQ$s#QhLKi+{>X8&HV9g^lI4g zqyl{=)-%(h?=WwEMPlBvJf`P%A3N;v1~N@fW4W*t%~+U2D(;=9;-SZ3&BX+$+E~Q& z^|7$B=RcgN=)&AgT*OFz3a5W}C*hv36;zDRG;&sNvCZ%tHoVv33?>_rTV*M5=aCdB z+t1>r?;E5(AHq?Uw$qGZ9So6I!yq$an;UGw3ccS1UxTEGna3V-@T4@qgE+xCEJDFJNqxc=G*oU-)dGy~TBCGh19jo{a?v+`R zy;IWRam_`Tn=p!>%O|qRBYGI+yoKjmDZ$513B-FbJTo(#Oqd%?u9f()+S@6U`TjZR z^mOCjl7F~RDTs4YR_AtZzKAMzQ*ltFhl%zXW$QnOqp!6lDLqs`Q;Wq(ow_k;;QiuW z)ef8<3E`%+L}9={DUI|xMwIJTK>G9D_(^ycc9LUs&Yf6xuI@^5^W;0*fkvMHoWo)> zQ_iGEmSR>c?`4pDKznuW(&1x5Wcb=fvU1~QI;KGtrFZz^llCL96vG;ADBp+dHYR2R z6R{%Eg?7)C#<-yqjN3CAQoS&pE0}Vgbq`Eq8>px{UWbl*T;Zs&;9|*%6 zPj@sotA`a1g;4MIgK-UY1s8EsxYpLdCGe0F?Z2knizGL!yKfKFq>gNjbfm+#P#b*3r4PIH|K(sWS&)@5J)?}m0n4PSXR3?crn2d0% z5q4e-VJpYx;^`bG43xT$>Tfb2_!H2vKi^Rce#ST|_5^l3)uYe)B~kY+q5pcnfS21@ zEX3z@W?nd$>6xO(pKDYi%mJ@PR?@ix)^zZYI+rSX1e25^Fq4;fFQMtEak_v66l4c0@jM`;cUzv!`St~Ak{F3^L*ROI6qc`Py1fcgre*wIVX+a^&9@H64MPo*i)+?LgGSh6jHGcpCBjV`Jjq=2^ z=nd18Gs?8RoXF{xujNcWt>79`=8%@~UG&>_S=3rE5l&Sa@%cUzQk1Aj5{5fqNvIuB zE!QXN4&FFAOOp%LjHmyd7Es|{E0Ughhm1S6mpK3YLW3QypoW7E*|tfa(& zblZ^>6cuBCIN>SK@g&8mf?3jU%S>FFMs%-Hm~i9;d*t>Ult`(jdu;-U04GxWk_WU) zBNqK!#W40zDO`PZ5+mD60q{HC7Me4V1`K*=E$Fip5Q+E zXs$I3FKi@*yo>(%ra@MBiZsc5venjE?icMC=S{c!tft9N1NnYSELjqKj+&|D)59~h zi0rQe;BY4a+S$W6KVTMWxd5GMu?l@cCE)wuAuuuWz$@`zsK?`A*j;xWn#{(4+TK;1 z@1!#FO6NBGek6w)o1;kE-9l#ekQBLPd7v>fE1ayn+X?TzUom&KO+nOLftCFC{+T`L z%%ZJAq+o&tKB~C@LmfJVk|wI2dxE_ns>4Vu60r-;jmJ=fY&^$kfjNIBjloE=^^rTd zwRAtFr;~V>++?(=TSLALn@|9@<813f=SQ-DjHkb&7zx6n>&Bgk-VlbQP_bI+YA8%fU?z>L9DeqO-I$ zEH#=59hSD_zSA~Rx$Xw@nrCpD8YH3jikH+?b2{zf`SVU+1bFU#4EG}CD(UqIhqQH{ z@IbZ_J$`K6&FW#c^ z&pGJ83xl?9h{Rql5!PkJF?BH$NQ;3cJLK?GFjS(AH|k4pqt{LL)yGxbG~P=&x2>5u z={ld*u6+%6!hb~Vsa}dZ3`JpkLj(1C6F`35&Bb-g?%}zqEW7fyE^eyHWbXf( z!Sgu9$z*$u*{3HLH~)1os}`%e*7txtn(_ow6g^Lk8yvK^BqcMkeXx+&@8 zyLSJzp-|^VEb4vF?irCI@KF~Y@bk=)*qij&`dVg%n=Ird-NpQ!3sLpfOLYEhg!9MM z(3K`9anA1n_Th*$343w_+g42D_txp)ZL^4kwU5C=qNiws|4A$q%!Hyb%`n|fn+clq z3EMWlq9gHLblZz!TJzJGd+8WVAJmLvl^!S);c*MGuiuVmch?AJ#uicw?R1>KXCA7} z-ba4=jpvjv_Tl^vN2X<4jLkaN7Fy%D6QBQBPe}U-+Gp{O&l|XqetsXdbZ;ZgAO0Y? zIcmb}A9|0g9{W(&Gk3t7-|HVOk7Onq$>9;b{bY`{7Rj%Vpu)$DAw78`F*ezV#s@m! zUw9=Y&S&wfay&ij*i1RUA|_PV9L5Ru(Zp-kc+YDFw=(bqXq4NN^9jmKsr+s{=Cno7 zaNr%=J7~xf_n0H&B>ppD^Q9CD~ zlow(0T;%cKqgmu8-;qz8lt^@fq`2Glt*G06ABy`P!+_K#TvVA2ucc<<`x-?wPR%Fn zU6yn@8ILyOc<=Ml$(*O)HoWKnGO}wq7Cv&pV~BJ!=EA@|{1= zFBwOE94{vO0`Fr=sTuigGJ^K`CvbUmDfG14lKkEs7;~+gPR`%O6ln$XGl_0Uzfwdx z9gb3U@k};PW-Yg<;RFr1=Rqv^+{auaS?EeUOb+v1=9G+4D0|cg>m*0OPFaQfCYHzk zm>xrBo_8a*RE${Wd%?nAUW|UU33R0&0J(Q^^gziP99-c_-3s_Db=oVcd0C!!G4k^@ zqsRO;d4@|<^YD=QY>ZjGh!YVcN z!c?NWlf*F^r&#moOk(6XmeW}^2K(2|!BZQY;X|+p zsyaV`HTCatWM4Gx&n_kDd{(7y-y9gHGKO`#n8NyOFQ+p+CS%Lv)s5p*hiT(8A=qC# z9(K)Jh^@KjNMg?{yz!xcu`UOYG!Vtq=PQ_ibupw`SB_jhqXZFalHov!I5&;&tgX&@ zk9p>nod0Zj%-!t6gd|GBmhxr5yeJ{dr@g{|gM)bfDW8*AX$dQJS2E{CvPkK!W3Wot zk@~+VgSjs>v1_jiy))@8^%K@7Yv<_F_5D_;-SvpJ8tsC}bv3MWQY%zDoq$jMmf$YC zl{iH`r52a&Jv`NKbgUfEzIAet~?l0tQbMrBKUm|hK--T_W+32)Go+If6 zTzJz&KF@NDsw~;ZvnmDzF@cXDVUiYy($%nwX=Nm%EWkBmKG`hw3Lc1sW2){72u};f z2Yp_g^HdYEuWB;yFp%V=+=X%A?;gCBu! z?g$_o1}321Cu3HJ^MO5Q-r?7%a~vK&&DHG)#L}zh*@FFB$lmF)c8)r^nvXbc|0s6G#*_62$KX7^ci65T#+U?n z5%JuUkWeO$2h>xzbNiR#gv;XiOg@sD@Jy2Tv%T=erGL!a$TB$SkVLZcU$9@DPNC_k z#e%~Y!sO-W^Mb!?YasJW9Gw@I1lGq|nQgE5z0CgYSd$uz_m#5fQ?ERx)?UE9v`OJ? z8?NE2-{UZSq>oxF7?Xh3CvfJzD4is70#gjU(70_LbIzfF%(j#SLtPUHcGkurn1ycF z%CP%nAiXe_vT|ef!0m|!$$WkuxXU8sn_E96vxG7Hz)l*_C4z32E9uDUW=46*T{g$5 zkIqh!hdz%4%>ATB?a#ZT%#XQj-JDzOv-?Uo-TEy}4qS_`FErsHmcSLBhdXM~hEL0m zaN8`$vhs&R=<#uKD0g%SXQ|x7MS-7*!S-{YG^`6Pp4*_cl;_hc-iL?p9EivCc$hc8 z03DvE&}9Zw$b#y9(97@IMc(cO_qV=G!rq-X&o=Q>~ zb1G!cb4o7R(&Ew~`tC{&{i)K-+#upyyuw1vo)rgAyqrl`tpxgCD8&~W;_!iOG^SrN zA}73}n34(K8>4&5F(=HF?rUE_w;G=%8-KddBR>w?u6!l}N5_bfq|gI2dBPSvsx3wq zN#11-s|iq}(3RHC79oR$hNS-_NBg8KQ0l%Nyq2s-_4h)Ux{HGQtzUFFRSrEh#}O-g zNju@O-yk963YvPWa93pQ!AT~DOgXcMf&Of|tot!Yr(Gv@8j{>v2|rH$-VD5`wj9q+ zI16!IZs7WmcTXh5!suFG+_*l9(n$vB;LblYAL8dAv(sSZ{$Dh~-H@4Z%n7>YOX7`_ zEzFOB5=c1|j=wL=z!N=Exh9$(sqUu>x`j#n-plyHhw@(COc<(6Wit4z>?KKCI{j`7jUG@$v7%jq zxwd*Vb?b3xa-9L8?W=5u22Rjexgn_f5zpL_;?GH!%;3=ID_E_WNQEEfGu_(!(LMhp z{+BFZhedeyj-Uq9vYMGvwF<1U;rkl0z0_r1DtXwx5J#WKF){bENY}0zIGCNwc79#U z-V~OlA|(LAS|_PZYXNd^#^BLwLgZ_CHtsxi1v_)vG3RaqbS%*T)6{G1;{W&_f7ERh z&gi6{+FxMM)EXKQ+f53id~EZm6j|^3kB$3&4MaSp;nFt_!u@2qsw^Sun)V;5k6Dgm z)^A72>ZA1j(-)8|YD2E%9l*q$6_9#pgjGwBATM6=&y`Lb6Be_Zlw4%UpTa~qtR#cK zvhPxFy-$=)G=XO#tC)pT)7aL(BMcXv!k?R^$*rhbJobxlb;miu&gjXw+bj=fHr>OU zwRh3Wa4A=i{2%N;O6iv4jU;756&7_^a|s!fN!ct9M&G$x@L*jfjS$*S4a6;AZjCiJ zck%|D+Z|7;o>h>d3o_{CBZe;@xU<@s`S2q*o9CAgGplyU;Na5VG~7mrii}1RH9jNO z_M1iJwQF&7SsGYa_R}XLCgke7IW$GUXKL<$!9%9%L?>SdwwNH8S%jk+DB$+^06q!V z%s2-xAYz^=bgcVTy0v==JFh8&&!mZxn+Nzd+o6MaXpagf7b$`J-@ioR2~+rtYM-rM z&N&!+Qx98aePb8%JUO?7hp_9tv z%Wo;0TM6E+b=2UxC+6iyGH|9bxqvtf^St zcJPs&D3D`H;j^tUeluE#hJLniMRJg7(5#~K9=nnSyg&8t8!Kv+$-5&@?Xg|;s))MT z|D~%Rwi17TXJRz%3hcE8ZrFb|SbuAz`+gR%E=#6xi8iu0=X$>2WA<4(wQm9k%Ok08 z_d7V#SqRS+pR*AfRVe(>hn*QKO~(H>UhvV_oX=B9(CxjFc&I&@H3&{1OuJA$m@INk0e2@n<<>dg8ieW9j$-Fni6j7l!8$ zOQla#jdzMY)Lcj|{mg9q;9bF<`@w(rc{Tz5b41w$DR}fFmgjE&f#k)z&?rcUR_s}c z&+fR;jO+$lQW}6;J+kq_ZW$gumSFS!?J}~Eok}j9IZFM$eurgY=A>WV7%!_YL_G_4 zyj)8_=+SwmkGDzGYfd0sH6&RjVf;#v6=aB@~dN#cXER`iipIWw|$ zrWiT3YbVfnb-cL6lHRF)0xsuyZtNrzsJ+`n++I#Y=GzXu>az?d&q;!&Txk+~eHu|Y zHUmW$Re@J;Bh2-Sp-mnFVxqB&K9Jgmou{iIxBP?c=`VX=UCMkARx-qMd*kqNM;i6} zbDPpq{=R3CO0vBr@n6ttO1<4l_v(Ap@pL-sYp((0lg~k8|6TU0moO0$DFm|l4@9cl zz(F}*_?GBgNNbO;>GZH z_7QY)`Ht_EYsl6=cbM_L`DABx8VX4Xaoc1HI3&^7ccBfw*?17){^js)ejKC2GlbVO zesp>7bZ{t%U>%B=GEcVjV9VA^%$1fSu%f326h+R_N5{=sKObE#;EM=X>sm$c^e>G3WRkN|sj<`~%v0x?+BY&mdUgT^8?E9x zMvigCH;lNvQy19vDSGVnXAC*-TmwtxWQcFuV>->>6V)c{W>w3L;YNile3jk>VXJ@H zcJSVaamyw_na~vCpDIOmv2WS(Vr86`X+esY&cL$BW_qGOj6s)`tb^b%PSFt}TRRP5 zPO$*9FKKYpR21jG3&IrlG9*(HP^2l8R5U^`s;>YkNK1xqF9ce7}sn zov@nb#z@25lBL9$cc47;-^|=COvIz}Mro~mJuy7;kj5H0va39U$)ANkVc!-Nh@1Tw zE{y$%?0*{YWw#C~x7@<bdoE-z7Ni#5teM1!=zhitmsb!9Mf%}iJp{SqEZG?V_|`Q+WZTJj-yI`)S4 z(`NfKR5g4F*y)ym)0z~p{~gO`k3yI`XYwJ)_@{N3ZyTJtRLd0a(HZV}1Zo$s;(PnRHL;aF*Vn@qw3Jb`KxdLxwFXSTZ4-1y zdeZqO{7m#+2n0`B44d^8$>Qfm#HqX#mWlJMMZH7JE6Zl4{~YASiNJdRzF91q$<>$yrBQ_NEhW`4>orLXU`uxlH7Xn;{3Gv!{T&BG4@ zwpR2YSw812eSY#IE(tXz<17uy!SNR8-xf=z=d0n=f-+QD{R_DjXKCg8D@%%%-%$5U}7 zSE}@;n2p_`#f&+T%hblqr)K4EsmH1so}ucAn=vUPzbJkGtI*otyfadrp;M3<{=$X{l_}~{mTXOoxoS4u^$6Onv z`cm(0txqwu_uWT0-ET#&3g)v-DzDk&>t8VbqLZlivuDs9dYB$I-$xdwT_7EsoAJQP zaTv6=8cla=G9wA~@WAstWAg0`-G1B@{d2_ej^A`xXozHs&?2h7Z3YCjJ*2wT`-omi zJvCJDr2``Ch{Da&@M!WT`g`|JlJYp6lA&|{HJunq zU7`Qn$IvLvNdEbXXOv&Eqr$(GaUe92ZpzGocC!sssbGjsnXN~A<2J)fZ%gv=zZ5XB zegel^GD&XbPx{-<00keDnDT!|QM)GqwJ*xiAFsBv|1#TX`IKlTNK>D_eNiL0zTg-% zO&_LT12Vv5bS%k{xkCESN^s3%_OWhzKfvmVru2ns2{gs{GTQ$pL-NH)vf6tL*>a@o(HbcCp^$e+R4yWFsLgbp@8y#wFhpMV( zf$x+IoUgeVbXUy62M4Ethu(2``!R-2@aAxr`7-RNHHEXHSmisKrfY3aORgB~c_a12K{*B<7V{z^sbB~ z-CL{R#`DQ^UCI&mLI!1$zZjEabH(w9&JAkQmjwgUkKsMP3uyDemISV=;X=Cp;%k3l zO#F5M@v196{7^+Eg_fiHlV-ZtFP9vZs^SjC|DaZWLUzW^_rTQ1hCZ>`2AY9glv%fn z?z@t1d-5xfxmheq`@F1)>z!ZtH^Lkhyc6hI`U>xq`Qx-}V~N*XQL&E${r`o^598>S*-8eGg9;l`;P=2;sb@(_rOS zP9x_rU{V_jpP3p->_+)$HTXE zr7+v;CX!lP^w_lKeqMjD;VZl?w&WvsF0lNgC>J3uEVnXb?$XIS8^o zg}SqA(OP^a9+>h7g5xG~tuB&e=GHfCVqy*^j&xzQ=L&AE&UCUWDjWE-ctH%&S!R< z81MJGRZP-W50Y**p080FN)`lV;-zWhxX_Xa!AwaDH&7pt7P6 zANt(H?2Y$%e@-I%J@P5YRHYDIi=(zmO6cgVop@tAgR57vp{y%Smip zOQkMH5=)sddU1aQ`JClKmQDz=G0vBvUFv)Akr%^glr@0Gk7>BOOcj4tG{ISAH9<$^ zcoeqrBu4j;#LPZxE7o_Cify$b7Rpk%u~m}X(GI{#MGd%A>NlLx66XR&R&aV>CGq0H zDWuq30>7ElSm!IIk6VCSN2PENHsNCB1n@Iy z1K;RSluAh=38vQU7sZKamo$n#M=n9w%&+9Z(=K>cR!&{Yqp5ECb(mVt!uvW^+|a*` z49^;ZbhDNCc}Ff?e%=f#%p&Ph^_gT1k8eEO_a7R+GsNNGySQHD5GQq`1T9k4aJ^|U z{(klrAAdeineRb3P*N@Mb9@F9nO{(sa~B^PuymyGCPbUWlijzwY1k8AjM}Hhcwb6q z!+0(%E8kCtx<+Zo$bR%WI}@fAtI*-A44c;7L$z~M;KR)Kbb^T%mV6lj>6vSA=VKih z_+kuNkx$T8c0TIc4YAjS?~#rL`^lCyl~gVuh&<0w=QA!3c@|R}E0J3Wuey}!Tg5oi zUY1Di4y7@Ze@#J$&@23GYA41Vv_zRODHxnC40~TCfmOXGUfem3m^F9c$dhW+Nzf*F zCfCWXDazdW)*h5{m*N&Zs$unTJl&#f46dfni0Fs$q~KN{GKUp#8_$86`lWy~uDe4# z$A*BpOCnX=XHU0{`zQ!_9s~dKB(QPaEl_B3#J&$*v}DI=5_DJz7l!bhYknqfwNQv# zy5R#6v7Kc1$3&TgJgBBeR@GDI*~zvBp;q`Q%fUA5r7Mbm@PrV%40hZNL%3Zuli$N^ zByPgmc4ulP;W=jm+_>m940$%w&~3f6*07zm`LGmgdy472j8BYKsylVsuT5s9A0bPE z4r6#DOTYQML(?;Tq8u$pR_$ospUd-9^WQwNE^wbXQp z1KD=m759$}Q`?8rL1vi6r{X73eyk*3HOgnzqSY|FM1u?b_m^I;nF2$$JZn;j5c`_r z)X6`T(LZ5_?zc{}|9BrnTjO7r8n@7eM;5|$p^5mZy#hL~`$71LEVK-DMM-LoGnQGy z_HTcndXgW}8=g-j9>>tu6GLcxc`Z5T^8&WK)kb$8YgW~)n#Nyxh$Z~oblXzi->u5~ zG9%^qtcf?-9jQew3WZ^s-~#V{Y{!eEQa=P*4 z@eCTLCc=eii_lMGG4$@&=9&M+keDPOrvA@Jo7Dg_UoDB8{S$>J0{MMy>`oY5UO}ep z7a{r;6EM+MfrR?~p{GKtX+(_x|A-Z$ZNW*Pu7G#mPCeIofQR7_QFYvV<& zF`HlE)W1qxvVveMSFGk(8fP&(V*}wbn0tEhs{SYz3opPXzV4TypK^<_Z8jr;3Tn-Sxis8xkb}_ zZgBKlF#Q~2)ln6*i`gZC+^x+{UUP!d?Cq|(a?%B1`U z@A&+fLD0{Rk!r6YQ=eJkNE|%NFxd2u*tQUmNQ3&tV05*$DHi9Qb%CALkc@SDdS zR^ekjIX_Me56;uV2j2`aK*Ei1yZaIRRx;nOZqlwD;lElJHP+^@w_}|+|PA=KJT~8{LySw z)M=FDpJT_Tj^)Dnv-|a@EPRao%IA@f(y5#!=*b<#qi)Yw{jr z`vROBpu|?a?`S;0=j+|pCPDx8VJe)Oh3ZwlXqi}tJGUj$7oQezk*lVX*62;bj*^L} zH{~Q02By-*k^*#(tP|$aXqx>eAB*<1gTz5)2w(D?y?CO9j!G;Nym7nT_-%AG_buTc z`ewexnZGiKSr{bDcD&1wl!)ym<#L@8qA;oAWObLmO$*DyEND;=hrJOLXXm>>T2HFNxe}nhiH_GVY0%Mti<3 zmc7diawqMe+8zU}S?pyHzPG~AoE8Y0Ba4#$qR5#Zr_Fa%;YVg7^*CZe^76ddyrB~I zJfErAlyw#B=6J$9$tI%lT9Ywf5i9(Y>rNJZAJ0jRzd$pVUM02l@9pK0GC{fj)A&z}NVTbt`(KC7C!u;2o%T4)M?Do2yIJ(J-qXUP5T zoh8Um_y-w(9q90uG609lLo~5d|v?s@ar^53a0* zb}kKSyF$1P7yrV zkNgSy9<`9cT|zeMml`u?_zaockFJRA zfa(7}VDG0v)O0!l8{X~1v>PVydEgN~p1p|gAe_Rbc@3~MT9VH}2+^kYEeQ}WWhA=7 zQRmKa8Z+}6ITUw~RW=&M4fVwlla1r4{FYaYsiZTE6ElbW>CKnEtSYq}C zEAZ)=fnVE)h|Gge3M^s)4MhtGf8P6M0GXNGnSxlSw%y8bFMs#pfgWU0wbV*r0 zd3Ea@S()~fS?RG6uAQW0)I2@dP+<*?Z4C@#Oi;-^7r&+1!TtDtR=PY0Y72LwS$836 zyY2(xmHlLbloq%T%fQzqbGiJS1m@M-G-h$uM`p<xUI9K^oIC(Qa zr_b1c0{w%qSf+#Jt`?DBQ*0nk_9y+*Z3<0K%;1@NA3c}OJ5kyyg?&kTZ{s!SGxywJ0&hF^-q0lBx>*9`};A zzKSH<*(6%)JPxL;PlF!ULwsjiLHJ|Q0~%CQONck$N7v=uINMTC_GcHXp{xtmsvGfH zRvG4x@q-ax{+_SAMp8~*U~)z3h}TOW(mq3#%$C+6NBE|fTb zIoB{h=o^!JwwpZGH{#q$JheacmTbuvx4m_gW5theB(jHdh`1w17&%IJ&AS2@I8V4O z&5_M}D{*-HcH+Z#6>^@&f^xG4yuLmtn3?~b1Q%VP&wB%yzKMo(yI88w>y011%(+r7 zZ9iSJ`Z{gJ?`=sJ&P?EH%ETbNt&+|t45U8!671@+`XuAtHE1e9dLTg;l+xsJqS#K_>1EB_g*zDxV^ntZCR=iKeNjcrLz+*PtwJ3qc|HKH?*FelxPr9*l z3l`SD5r&owV)VZysQ7Og+OV=PsM*qJYfLdFEsB*YpM+&664()bX1amT*(ImjlNDPO zuwmj;_IS5G)YLMVpreI{`Ty0Ym3iRC=bf52`N7ZpKul3QiSaJl^hkpY&E3^Yinb*X zWiw?gnfr--c!s5?-7bQHkrNaBe27%=srnTkHDL8rQ*M*S1^gpeMZI1nqu!HsxM^4# zBerzY*IqNI+ADWhc{!f?$q8Y~r1fOS_hFhJ!h10PI#U@9O-{?<5IM&4>JDFOAUQ!* z*!pY@PR#g;dHxhXx;K(jPdDI>t_)V=k~JjBo+U-cN@&e*D>ytVS~#oU5@b~;knR(w z@R@ia`!QV{4=QN0fgWvwCwrEYNg<m~Vl`A>se`wHUBH)oVCv5v64kUHLp7gK#n?ph zH2m4nESNVaq>pAA;lz_u z(e3nJ#?l~=%TY;i>+eu~$ma=(UqeL9!%<6}kAC&S^hLm!^LrCZdI~I6R9fH&=qV#qR|h_Z%e&V_Yy(bQC$_I7H-l zR+r(Hm89dNE&g1pO^2WR!5NQb>`+%PW?g$vfArLnsn0K<=B%68xv__=cw<0I4rk&R zj}LUS-Ff=9`~vRF+e(x4+XSE0X0qayd1Qm%er%gu!;Ti4N`(6apgxd9f2?mI9xG!6 zl5bCv;(x1YohKrb>W@($@>!TGhv%A3WAe%jEF8|l3rDNzOQTVwq$>tD9sP{fe)I69 zfjHMiSPux&ZMjp*mhGl*C$+l~eRR6?GwzTp-IrhGf_Ls_W*SC)1dO8n5 z%cf4+e71wce=BCrw%(^L=l%JPuOE!vF_qoha)i|Gf6Lsum%z+97lEqpGeGKlDQ)9> z!v{Yf7pyZ*rx(SGP(xFR!#`%&rS*g^T`&<_RR>6R*n7~fQRFVhThS*B-!=LdNeyk) zsgeFW?7x19D|>l@%9N^L{hPxeXW7BLcu_-lN4Z06N)nZHv4E`dyO1oi12y)q!U3lU zykf3Ms!TjFHS8%Cmz&`tNlU6X*_nOg!;*wQkLg!&Gtf&3q3sJD;Obv_F>G5X{M+aYKs3!{1N!Kza^VpJ6MtUIFh`jjJ8(QGFyHh zZTRu08Mf|M;1nl+5ESN%a{Ut)aKYLMc>nY-T-P-YUbI;VV??Y#R^JNhj!TgLLN=3s z(hH!#T7`P=sK6q*;)0x+hIl!t%;cr@!OR+(bt|_ss25b z5& zoXkinE7XRm3onwixx52=Sd6^ZHGxF>MxYu|A#6Aq1P4xjr>5VL9rQDyV@=MGi{df( zy|;%hTCtOR`uQB@oIZ`cYCW`V+hObx@4``DGr8(nSE$rMaa-xPX=Kz8KZ}&QO~t0% zqf3Y5Y1^oO0+SC{*qK-7<3i_n;H-b63{7yF#W`}|*-|(+-ih4G@>BJLe?05S93ZdiCKHdhpw}H7P+4sxE6&e&bwfZZVAa8%%Q0P z>#%Y0F7D5O7HoO)p3gC@fP3t0>XZH-Kh>~Y6o*g7IpAq&( zX5yiamFN_sgKit&H=b^~MKZO^89g0c?ojLja<+E}|6Z9187=prJ8UcvyQSoiFa-vk zJ0SN(4el=0;CJ7~T+ZN5GE`MfMGv1xDQ#(-2hqfasB$~C0!jZwFWNWu4YT#XLUd`r zONO3yFbmL(n;AWox$DtQbf58_`O&&y+Oh>Mb^T?ACB-mnK_zZIeFN8N+@#}DEAU(% zgMU3=kolEsn7!I#xWSJruWC6^9u! zr8%#YaWr{W6M0w_POe%$q`~VC(~_G~u=Bb*Y%s1C^jcVeuh(~oY-gZ(*a^yaeu7vB zYmkmFXI*Bpc*M7xxs$(yOtP6uHsr_BvS$-8?aNR0xP}6K9PpjCWJ;i!-zD&RIRo@p z3{lxpi}4Dd{PS$D1D(XbSbqyqDG)Kx_7*l3zh@rq$YCizD_QVL4R7W$Ttra{Y~2?5418Xel4lNu#9^V3 zl)a50)hZPNuRpcmUR46M(~(Wm{fOo(j*&TwL+Ik#Oz`{qlqhD_5r+le$nHhDu;Q05 z_iRot&e!}z*DJi`8}BEWMP5d9tK(atWgeyY@F7#r9;d1A=HZwmd(dnDL=vBGg+RH( z&^#DHl4H#1&+si6H(rV7E9-H!5$b%#R-ah*>XKYZ4zdbF$OzwC7*n)`Jjk9V_?6T| zxqmmv*>OuC>AWQ|6N`m!KVzXT&K~~Vx#63uUdWslfkXYpsJ=kJg!LQH@(L?v z|GDFku;2~WZv2XahSeC~J%@9( zoNM-?my;ugEt&@5$K|kNza-bYVJ|by;04|B{6F-W&p%(KjOl?n40^?XVOPXnqmRcX zV$SFZXb{P=0V=T&xICb7s#+sRXQq*@yT)RW#0|3U$5H&ct$=F&mx!{%*GcmBa99v* z#bwL+z%<)vaB-bH2Wy2U~=Y;G7E^ z&SYZKN=tgDBoh}8ucZ?|p2H~zZ(%{61E}FE3)su&Z!qJJ?gRHR+bFc?U`7!C8^o#}az;hc8H=NEYh4%7 z(7i%BA0B19_9URF+iX&%HG+cwdeFeKpt{y4AB9o^`Mo{_iN<>Q*^ zcB#8$;Lc_G!qtoqQ;y;EbNIaPcb<#FcOM=bzCgF<7Z~D{kG886!6p5Mz;xLt&RViT zaI0Z9iNCMK@hDBS&O5_=xT1!)O!^jvN+!XW?zhDC5kJ5E2=L=0 z?-l%Q4kI(JliS`4nJlwYaDTfDr>XuAMZI$|>c3%BQ_{h}x$z`nX(Fu)+K2N}hZ{qjK}=Uv6t8-Wf0Rx-gLpE2x820s)X0i;x4`GD6-#@b9=s#G^wm3*1==I>9RDG zkcZS-d<{GC+FtIEsTU?6-iH={QXpb_8{6lw6ua);r44e$!Y8-9Y4ZUIOl9J5%XV)r z`_OZAdw-4oRVHZn=P3AnH$~G=YE-=L9ID9ipR?g9=0v%Y*1y$IAeaWPee0mJsR^|g zYQm-kjl_JfD-9R!AR}6Hz~_7fCg$st3!=ARvddx+IpRt8t+5B0bQh|uZ3=oe8mz$$ zV^(3xe(+yjkNbB`!NGXXLLwsa*(`co~E-or7|H+Ybey|>7EElYSFXMl>yXE^Tl zJAAuMmh15*@VLN-d$l$T-wOxuPT*3!xn~~TTr5M5j9o0SJg-hqB+KEr)Ew^F{ByW@ zAe-`MIdHqvh!Oz9i=p66K99zk%Xc`lzg`1ow*`VAUWn32Fr&N z*>fHAxH@p6q7R^D;t*+*yG%E{45#xZBoO=&0L$fc_`Jb%s2bS=8+5f0F1@9*9wd=; z$1*rAvz``&9>d)wz3jBP({X{C1?Bo;L3^YEW+(aM2mkpPk`+$on5y9_bywOUQA_>R zCv&gvj?i8SJv{qiE~hY1h-(hqr(5PdVdZ`WfYWdq`JN?*=4Y0}ddmUgl`9D{iQ%YP zz~`b8M99FY%Z%kZ4Vso00@n&|!-L3Wv`NPeZ}lnTxW!V)K07TKy|jt0HRflVZOx?S z^#|N)v6fVXs?lqgKI49qVEVG-Dz@>n3_srCkrA+udcRiSq!(!8*QRlZ?Xyv=)&tWD zT2Lp2&{Lsd=-k5lYWeTsLxmM)oQooH@Js6Ns~bI`y9CW-W?hb8-oXVMSDJO!r?JK z;9+7xzRN~}d}1$Evx)|-p_gE~&<`V8^5}Vqba1^bL0P@&n2{R?<2h^8J3b$mt!$1b8k|N| zsnOuK)DG3xgtFh9i!nL3iGB=H#KHxmV7cOU@N$h~k_Hz7l~w1)go|(u`^z}r_21ES zCZPc-Ghwds8)E#{1Bd8h^5gk?o_SV+Pk(zs&5cBORAE5YI(nmEQyQMj?xu#z;=#Z_ zg}sq{mQme$5rh9&;~4Qp*m*veeoxJUKBL!UL&^YaC0PlDMrNQ}o`ie;ZDN%Ic}7C^ zYV3A4=H{}S=;xMVSkUXk%-eUAn9S4S_PV9e{0CjAFvgNod^rJCR}ayiZ%=U%KW9+R zIKs)c)S+>e8Fyj#Uu+mL;ADb+k$}-AaDAE@{GGGqu`&mw}^4FSv1iu$>DupZKVxd+^itD={ftODI z6v~)cGoXH(>=aIgXR?n`;ZQcprWKL^_#F2x%BK6IrO>j>i%#lz zNUB!n(!c*%ao=TaG5*3c?2~9^qxi1-UyW|s@|EwJR$n8kGIMN00>yxeE`|(YF-&gP zr2Q-2f|KQD()m>#f4=|4UKwjn6n>b%*q4fQiMI$m&s5~>_181GI=32MtZ>H6?Tsk+VF`C# zLl;|R&eNEZD6XL`n7r6K-geSd8L(UNiFPNXL!71vH-+cRd|$?MlPp&;0>f{_M0PwV zdg;>eYvnli!vRz{>jJEDIQ4c@MoEJJh`e+GoNdyX+gIjtQro(4Ujc!jsYUG6qpol% z=P^Xw34xe_C7_`Go@dO@= zEX#~**pHK^-Npiv#s0pgPTMTS!M|oDG4lF`)AmZ^_((U5J?o9dS!=-kLKtjMt0jFt z_7E~Z7Iy0YBV~3G+e|iE#RVR_=jX!L324=xJ!z%LJR~am( zE<}e)HLgHuI(3R{B!(7wpmoO)iq}_BtLdY;*m`v=e^behV_vboTa&51{ulD{#sRcZ zkhC>$SWRC{e2vfg@?pJwGhIJv0p43}gAc!*qqleHVTGd)&AVEAW|2$KOQmCDM@*o}*OD_kI^2BOYIW z!%EX|I$$^jdQE?zd!R0s7!FaBC^P7ttAi=!!N{s>fSs6=@S8#}JK}N+zx);D{-$=& zpmniCJ7_fyKVOVN=gR156-THj(&j=II&=HIrE$gtU;Nx}0fOTZkoVC6w65f`Ya|4W zFP~GWe`$rfd21l~xh&&;vH(gZo?=^!oWbX{4F>5p3xBK0!H~KI+Jr5pGgkId=AtBL zFO`FJ0ZMS++6YG+0oG1c0`r{*vG{ErO^GvyihZ6iy+#X1KWU>!H^yN0>hatXR)c4d z2)K+rB3#CCiZ4Hja)C?#(w}LPWNT3=sUNOIccu#6K*lzBQ6O=t=lv*eYUrpnvYagc zd>{2+pi}=ot8&FZ zIae@M4oM?>(Nts*hw~j^`KZ;QWKi?QnvH+4|U==fZ{F zd`l<2G@#!9yo3>hhS2dbR(P22{3y;S7d%@s$S!mfp!q@pKHs{Q9FR$e-v|1zvQWsr z|C5Cuio9sM7|%kBwdNi?vBdB7DMUJJC1%_C!sOlOVE+1-5H`Vuc=t#OpO}OZiHLJ3 zSn3D*Rx+I5t?|U@Sp*4-7~~xuS0JOw11=7~0JjG*p!@8-@aWbSthpHlS6`=)YRNz7 z7aGSVM}4A(o#RnW&w|W5ycg7U!@=d1KJ464Cb%+j2bvEB!h|dQt~uVAGadbb*!)eR zM?dGXp(}HRxw;9se@YdNRsKUaALQ9*e3v-VDjy|Eo?+vNHwMohj}sG)F~Kf2G%{?E zoSO2AHGlpHUMJ&1%qGcyYAEUpN$P3{3i~}tP3zhm`ZoWjp3^56uFU$ zhWKyv1~7TT^SIJ(GA9ooAeOg|VeN(vy2#;SQ-67qa1dP!=cRY(v%&%Z)`du>Ee457DY*mL1K3&1zo zjxKu?L(5cDV4dnDT3_>v4RN-E=N5!Mdb|NX*10hEH-09k-8#`^mIH0_sU$?HgwZHV zqSJa}1(lBzpsF>37SFpzOqUqrpim8S76ehz+o9y6wHp)$D9h;m< zo-I9y=})%8X!$&*d3PD9OVJ5Bk(t0kx@8O-_b2_klA%OHQ5w8B)PK+$C!gViwU^qKn@Nw@` zxb`aoC;LXz%eG#)%)=8#FIOOWmeRn+Y^FVPI!MEa9Yju!6jp}~(ZjrJk5OZ2)#_t7 z)-ex522JtRE>SXT_%uDZSO;?7CqaO<51$bFu~&2X8) z8RgBU6@13jNJW?2Exz1%xXTF_?K)2n1n1NLPE<0LI?iDFE0<1LT>zcC)v^EHGV&u) zfm50ogS)Ptr=I$mP?DN~58tHI*e7Q&eTOt|c_B{tCptcVP|fHX=L(~H<)HLn11lbz zMSdGef>rSvFgfsoxp8kX+?BmgVj~mqh3!0c)LJ*}KYowA<*)7AQ|`d<*d6dG;T8X0 z*@fx${gi1~!u1?BgL4YU`CiXs)R^VK^Z8e>rf+j#hbDgnGucDF_}7!odN=9e2X{!3 zT`^vZGy<#WYeeJuL~_7biMk~4GXUq4Sa-A+hCh!6JN|t!RV4<)8(*Q<%RG9VZpS6> zn%SK9bBWj?Bch)3h25+jCFoWzflf>3##}!R7EIU)t&JXRvXK%gv52BmU;B^&wWq|Y z%N@MD=i2dly6$7J`pjyIXAs-oh(OpcM_8YJ6(_vu0;?1mazI}J zj%TFOVVk4WvvvcW5!*@DB+5XBZ7>o0t%yrQU$N6a+JK-`6vx;r!!((G@->IgfDYJU zx~DdJO>m{3U+%{RpSqcsdQ&;SQ9RROv_qL-mv-Oe^AXvLhTGbU*hKkt1_Ko z`*9h#xv2!ycN;L>GLrbaa5=YK+6i}my@2aB9l`rDR&cTT3Tbf^AqPMCg4e4KLE9a3 z;QCwOVZs`sSow)~7nj51$(3Y2XGE^fdw^Fz*5Vna8C}1hL0#3O{6YpMlG zdhcdrmTx8Fg5NU|bN%3Wy*Ktw%b}Ncd}iGKE5pm5M&V%3I;?gtCTk9B(=Y#hCwjm3 zfqE!Mo?Y}nW1oI{(ZrF8d5tC7;?5wvs|w>)*J9iRADAy}q;`An6W;-McH|AuPm=2d zW3-09F9{|RU9N=(Lo3OxH_u7`#jB_nWM^BtitlH%sn~XYsbg;YltR;{QmPfZ zihDKHm0Ot*&gFX^hKjB*tBrj>=s12Bur4K@`gog=-I5c@%2i(EylW(R`06^IN}EmZ z*dD{Zoyw#^@Ptk)DbzfjYI|fzovL13&zV2k%?4b$&-$%2BB{19 zbcg;QxP9Y3M3(UGyz&D?_tPdcib{-!4>|Y%q$Ky`m->(KJ7oCJ(ct+65<#YLbv;emSUuMMA8c9qrN25PZ;r5zK z5b@q>nzTfgdvn()p2hPs+Yg_~ zhDHxC?$W_1Uvs<=(?HGndG_4}nV6;LNy~J58JnfqG_{BK5srI??|$n*?2AfFywXXt zTW@3BUTN;QR0L;IaR&|h2FWVTuS71m3#VJGWcFKZ$C7>1xF)AHxMAl$7DTG=vP2H~!SKK`s04W3Nzg7Sf@h_iPK7Orn-Ag!6Mn$kd<ROBV(O~U1TTdx#iyb? zv$$CkHkwGcYE`zm$?M$)fYviQdA2{HRAh4*HP zaPl+YI`ro_GY4CbF?1yO!iiR6w> z;r$RzE@XEkaXFq&nzwDCyHbwgyvIwxo)*IFV0~`v1_|Q7&H)~78p}O8W&}R95g>D< zh9rIbMS~Q!(!cBUVfe!;(ztjlTskwI*%aEqzM1ZgUkwvci_hll{2M_IgcPFl{trUY zQ{%|;TVr5jc{8*5pcIWO@}p+n7Z@+$W@g2k*?6w431&nae=)sq8E1Za0=M|z1h_ro#Lpfp*~&yk7<0-~ za3Z$^ABO3}!y7ZP*ENk()NaI1mk4?^z#q>>onma1L?PAJA3a|?;@)T@v|nPv-b8J7 z#UVR7IG^V(WVbP!9bC!H7qTSRY9}-P?|V@8|BOZayklU^Moi>M+7tC&vGuzWvB3Wr zEZ880cJtcM_=yCqRoBMoY=3_CeiS2I3@}yY0v>yN28%z8=FDqYHYG8fK0Bk$?-{4U z$pR&?OdkvWzNg`%D?#I*$Ebt&S8_|P2rjorz?q}k#LBA+mpLS%esk*n3k<5%7U8VA8OjT8yDTtVnR#Qz?9EL3wyJPr=}S5 z^tdnPHo4+VSu;|Q(ZZ~~+=Ky!)zstASlF_J=Nttq2GldKjOP1AqYh2vX5F*nBz{<-Sztm95VQQv8D@{c!;wvvUQ{k#vuEKw+Cnuy0l7ovAjCBD?v zAbkNPr0LIeEK|#bs|TXUs+Ujkx95A3e9)3Ct<1vXpVlWz3oxOz-?G zBM)mLVCJzX*7$Tkga)jHm(6cMuRz82)Qe^eI$lGzk{GnKxQqHP8>zvM_mF714bG_j zM>fBaz*zlrbVXu4xg7JJifvqlo%2WGp7uwmz~`xUxs+j4hz&8Qi-E3@eRQ$X7pleI zKhmexGI=Yl;5`35UcdT@krz8ggBe*^_ihtD9w)^;eYp=0=ZE8-(f3hPn8Lg-F-2dW zyF!1_DYka;5^$GfLVM9(=6bw06faQW`nvkr)tSm{kgtoqi&pL-eA&`*A z#@Fdrg+`(N#C9@iM2FMXI*G~i_L1bT?Zl?Z9_v@7VR`*UVrgK;g>N^&!|P%&;EMwL zqu{PUOF0SLJGYV$`7q9H-xhS0eM|p&dZOr|P;z0?2s3G5DRiY@h5V(j$;3cDE9f7N zzc1b+SFcv#*E1_fxy~yluXYGSRC(Xcq~l~>c@-&}qyVcFf}wDyJjZydFwsqqgm=`> z)1;W2%p%um=(;rzZ0od%szX2ZUOGbe{IrIPO_gYWWCvGM&7X_M)noPjYHVo>0!8<1 zo?9~+ET%Nj9`_`E=6;E7UoHu4uhJl`CIi}%SR(zPf&JBX0A}2NLAtw+G48x;&?_RJ zw2xmv$L;zDW{Egh50ytL(K8~k&9+~PxsS38l8iJ%cnDr=dpSrolIA+7VZ$?`7)Ge?jA8R}$mh%P`^AO6HQHk6`@UT1=Y0l4pU< zLFGpYa z?x+}{wuMpXpdz5kuOo1uMiTmMETEP*EU8qEA)Q(EMi44ng?T%3iG<&FI=fID{8D1b z>PJ`bu+9Lx_NBXyd*gK_0e za>ODS*KF{?<|R|PDiZ}RKjH`Veiwop+Y<0e{b^8-I{{UC6X;IwC)|;v^60NJOxYKGdGu};4!2!H z{fimsY#V{8Z)ISrc@h|hE(fK3`q<%elDT?X#CE)%2e*5tELyI;$5;maMd|uhw755# z>-ji}OR^|N*O^n$zt#{Ke|;)9e+_v0*O6DV3+U!Pz*$Gq@Lq!ik-R3)aR={UVar6w zt~n0J<~Om&me(;38RI~#T#0MejmPqG8_xXFPb$614GkYH$CVMuq|{!F+8Rl6-4bc2 zLpXZm=P7#AQjMFX_#8L1%D~nt0e4d21sOXd6T- z_INxiugZDyd1*iCBEB0kg?siX6)N{16MCmCf!|73W zXYkJRgTDkVAuRbYlak|Yyl2i|ANA+gpykJ-bj4#87#$XlP0u39os8`eDICU0S43&` z@&Njn?^kRW9>e|Xwb?8ACTxqY8guq}8eBVr^y!+lB&lB%=4VSm*@3ZO`^XT=K6}98 z3)e^_KTo$diGjWeg|y<97|I3CW~S>GGb=XQqHz2>jNChpR~((D0g5ZZ;Ss-kA30AJ zc%Q>lGv5n#=G+I-5#Cjh8ct`sJJb4G0+^h%j=+z2a1!JS8`r+2m1}=eb@^B#>Rd`H zAN?YW)NApOq$JbgJptU>&0&V#IO6n6g{FOaM!Xjdv=mqr)eo)wxvF;GJ!)Gb4yv&S*xVW`+_$F> zs|`~zW!6;BVa)V7TcY|$KQJ@-CO`NSO>78l)tfr(kXgD;{_PdhY?+4%5 z(oCe4V*Hl9*pw4Fl1)7$%E`1EQ{L6kOm}XGK;L`1%t@;^g0%lG^33c&xUX>!XE}s& zv8jJianp3tS$L0lgq;c9CHd>Ji?S^MY1yTvI$!n5#A%uOPbW=1a-%_Ls4w+ju}Q`CLlT#R#kh(|w( zaDV;c$d}_ytdWT~>|u*w&9d#p<|N-eiE$-a+I2>U z!}2d=$GZ#UrsFcQuOp0%n5)xS3nzi8c9y=KCZ%qoJ3a4LSlzH9p*nqrdRY zk5K&Rzl-=ERl_au_vxGsrSxy^R8$yghg`=X%ABjG8F7l39QB0Ee^!mDc_%qD>nU94 zl1S!2)oK{$!1vH*O@XRyufeX_1=eY)!p?{zaAVXVTbs#Bwwpp7!QlNQ_NXwQs*L>) z!s{M0qi-qE?5i#Cedh=AdTBo?F3BP5OzsgmFH7cpdLR*fJVN=yDVmNx!m7#|fyPT? zfm?Pgt<+1V!BUB2{2LvJ6p;hbiiM~$;S6eJ$3y??Ac$6vrQ35ub~~Okc4s)5IDT%vy@lA=hU4dLmr3uY1o-&*2M+woCZQr~5TK=u*@l`lw(=6Y zyLThg&-*=pS^KgL-ml5hFPYTr1AiMeh}%BY*W)@46=HU8B8r?I5A#)Bz(r~YBo*El zEVus2h@5-EyfZpQGfSh$x%DsbqWCX5Q&ajq4*y@K39+7Ka{utkt{l~EeHjUflR}rV0bRCj_MArD3iUP z_j`r1c@Ia(4AEyqe{}+mQyRn}TXm$*=WPEOP( zf?1dr?Jd|sbRVrIY9{o#S;JRONpgt;( zU1ku>v)#t9F)os@s>YL&IagVxj-gwPRk`dfr%=*p90UlI+1Q89c-bzNE!1$v7Ybu= zVYCc=Dti=dot(H?N=95y*9Ek3;2CkkJrG^D1qY*}Z~>6yHZ$B!e0kj@tX?2-D+b5NY%U8Z9ZZ|^t z-IpED!{|nliobYPa~jue2gt)v**S|Qpq=G$Fn@cF%;?{S0r9tq;bndsx~N zr0>ukk8s%jCYmb5Wa23}j#H*53w#{oXa&zst{k2LVcl!!Vzodr645|Cd%SrMww!HK zTRYfG@EH*62RN6sz-1ZSUZjpWC<0&C4DtT! zH1sGNgKmA7=z6Oqc-|-oKU)eQ?8XRlceXYceS1B$OtFU>Q+%-K-x{!uWO2uV!}u)v zIpxEq5V*r1s;-)obYTyDFi97Me2!$?!~KlMkBu}>E+4;-R-<_pYsuc<}baIBa(6Ml+#htp*X=t>~mOOcTQ!wgp7I%vO9j*xh z=$Dcyv=|m}rMy$mmhTDgyBy9uHW24TcBlW3q4SQb=?~+0i-r=VMWr+ql_b?Y-{%qy zLS$=L871jw%ZN%l?LA0?reu`v`94P?D^X~mY(iu;g#6AQ{a3wub?-gre4pp@dB5>M zyFTu4+lDhz)bOhPbyR!58wbluYFmZGhwufG&77IAp9G=S;49Y8am z1Y(C#KCwt?U<@Q?fa*tY$d>rb{@de@BSTsQuP>eD^EC(YkT3)5USB8e_KGyMU4@G` z3C9N)+>y!$afTa9(Nuhf@ZWYV_@v}VSBrF+>>DQ`q|>z4*UOKb-|`;_!!7Bh2YDpg z^gp`lPo}8+tPN~zZ=t?>o>MiRcl3Gj7aUX-Bgtzou=dOhZ0gsdYN-Yo_VFFQknqGC zuW~WS?E}erx)A3Y<})(8-of#$SIO=Pio%C~+R(A`HBGWhg8qINbO_JHneYCikqM)5 z8q>q`hK0J06c@jo_?tPIM+o7Zo_fw_~J^V7q* zuXHqVvF;)bH=F51p4qh4{srGREyYcxUi5e1Lv~SPFJpgpJSQV@0rxKtAVyju>|4Bs zowIm3m~Z2qz!q%QFvc>-^ zv@P3>Uc;Mt@1p~(X2#-;!UYgYLLgtY2=6~z0|grlK=DyDd`MhNs+259_PnjsxTl7l z_(O@C;eHfW^xY#9{I_7fa0#8^REO7YC{Z7y23G!@6l-XjOk{SSp~BQ8vg^GhHT1P) z^j@pu`jMyU*)zAPrEM6c7I*QYmJJpv2BC(W9FBDRE;w8{87D6C$2%coz+-~}zUjV> zpNe(a;st)J)?Z&PZ6H*z^O!QMUJwSqN4_B;XO@G(EHf_eQ7`R2_Lx;ZXbWjm=Tl)$ z3~4NyL6=t8!hAjh?P$LM%*}2x@2hjDan)+h=5Php59ZJh{{`dWGoH9~YZ>hg{7G&v zdrfv7k>C`QPLSV=ihR6I+wXt2NT-C~%?8*})+8=L!A{G6FKRbNv-W9OJSvRWM;+tyC_DKQR3D8eqx%$v_iYq8 zvEc*w_-8&=%;$5hK`EGs{&+jujeA-N*c3K_+g8hG^l$&AADiCSIxRX3ySDp*@$^?r zZ@VQqH(QC_azqk8|5`ME(stbgq5YYS<$$9&HE(S01^_rm2KD{)2DA9`fx zRpPJzf%#IL3<;bzoD4aQ-H%VA;RXS@*!vchW(zT7`WW*0O$6vF`C!~dOXi!?aY!_* zp%)THkf+hD*fekrH>fVguZN4sysRN2|JDkYA6f?zx3bXhbQCpIt0JR}zKgalIYq>) zKfz!53}&8hCJDK-8f}XkSmOMcbbn6}9n*S>&Wo2~8UMQXlvU8l<~J$jcj4QYX~WMa zF2ug&9i1m#%9h6v^z1|Q+|o%V)@8xw zsAqVmem(uR#1mS6rb6`8B=W_Fk9}3IMsEFQ-r2AWP4<@1??+qdBI{Zdrp7~CN}4Eo z+IV`%_B7dZ`7t@qjS^mVc|ns`Jfg4Us)^>HE1lKwjMyzYg)bduVY>ZpqA#|b?C-gN zM&AR-R89bvF}Gmuh`qEX^#aWq6%1x?chh|zC5h$Q2HLLE%-*=wN?c`qa9I5(H9Fjd z8x2xOkTv3u`3Jc3qAF_nbSBBrvxAWt4qVc^a!j#G;~x2IV`(x=!;48E^UXT~;$(v4$7Vfy2Ie7*S#)%makE&uXt_xm&Ow8C#J zF}TL|BvE?cP!+q|L;^b+C*tkl%@A)gO1R?K1D+u@jXUkxh2|RqIJqC0oZ3+h{!Qyj z1nMg($8z+nY#<|h@gG@Io-UA(eocy(-z6uD<+!fR=U8pq!*J-$Je+Vbii}h0BbL5v z1bg>}VchXZuB%6r@hBgSnud~GM~yf(>g6%>UYx>=?z7+@l|-G!>Tm<6>*)SG5nZMr zA$)3M0?O}&3Cy*myjsdXRmhxzKMuq1wq;=5*_!@5rU($@Q;Z+oWc07af`BG#_6rX2X_?3F5TVT6y1|%+9 zE7I{6W91WmqHMu67@so-&YeudK95-Hmvjj?`nyqA1qu4FW;OjMSOB^}L?yR*B)O!ZTDSm`O$vto{YZqu#=yJQB zM{+7peKB%l5v-ErzlQkvit%h^cZ0{ZU$ktTH5OKk;|x|f zW0Kn!)_0~K)qVbxxXhT$ciTtsuIl?(Jx)k3Ei%LI)DB!TnP(vTULq0qjPZ((kQf>E z<6gh_Y_)JCOiU`o$cS$+`}-t}x_AuhPALFqmXAuzDlYjyGrSyY1iKsLxQn+((=~e5 zP&D%q-54K6^uO|q*2g*!ecupeiU>M8jK;EG+flE+m9^m>;$J=jk=-dmVc$O-I_!kS z4Z*biggmFm^Rn*jG^p*coeEuVj-r3#PHJo_BJP{_!52+m@_=WM*|FQ8=S0-YBDT_K;l3Yx?6uEVs!$Ecr?ft~F#U}Bc2TuzU6g30cKqqLDeTAb>?ZMaWSK)$@ zJKXXY$JPLM{5z09&TfeUuZA$piVxJyeINI*I-(# zFN~eq3e|T`&LWW*oJZ_%P+v=yhmnj_f(QgGsu(9Ud&jP zO)dVYL&)_IYMsH^#0G0V`^Ja;Qn0re| z?mUph%FG0M+eD2kFdMouUe<`>DKBtXxrGvbcc-(rfxYpkpKSQ%MkesDeel>$`us-{ zEqoxxeQ?!)O4npozebNsTbv6TkFDqw^;n#su#B!Yx)-luRl3B;)sQYcl{^kJ^bqo;r^8+8eQ?9uo2XPNf~Gx(zkaS|49=fn0w;TL z*^iFl`BH`oCbCXhWGMeiBeqs~c6&Fs2S!mtVZyVK~L z6Utb0X&U_E4r1(7kA-26z6AW$!Gas1^cJ67-oTv*lrf?=7gd+CdGkN|NxV z0pgfg$Hu2^VlFDNv~lBPs%~9CWvs#&rL_4-ls}PMTNiVW`b^2kzwvlzn=P!G;X|jX zdXk}2#ey|(k7_wOay9O|a8eF1^X?p@4!sBI#Iq$da7r)DY&lIs%+Anjho*wtO(6|m zzmv{>A!KV(Q?M`bmtF6(BAmIf03W|Pic^zKu-fhf-uaqFZ%m(s&o=x)$Z*2bJFcU^ zMGiAe^>NF$QP@!?O-dh_vno@rv36|?rgsFN%jhK!7M@(q6cR}-aHwar5QzogX&4|!;!3K zwj>yT3k7I(A{y^L6RR`@?$Fn}Y(w@TIzc~)M#@IQc!3xBX*Lu5#l#_NnTFuY?p9)# z6$+jc`5f-XQLv=(J$-JuQRHW%Ofw4&z^%22x->s0Bd^7Pbki+9ry&n>Z|RX;sjI2M z*W1|F-ps6h@sPqd0j>Kd1OGiu1)VP{=#f7Wrtjz%)R%0ghpaT|?uL92w55TZ)i66> zT8&mP?wBuU!AOAunk(F)v+ktB0=-}4#P*HwX_h=(i`h|Y99l!h%VyK)rc^robQ7qO z2>7t1k>@pCBN{)vxyp1|IC{y2sfe6IW2W^pZ-W&Pw#)>%s}b~#R}S1Zk7TAFsb~z)_mz{`ufp$TDjka91D%5V-0nHxf}=XH!|3eTZTl#V>-9A zuK~v?zreWZXSqiXRdnWk3vBY1z=d*}G_bUg4diE`O%{?waj_kZUOtMm{O>La>sinJ zO*u}&4MX@aZbc`+_tY1N7hx6S|>HZtHRH;=d}fD zb?ebE0}ISF-h~$)@w~sw0a&#;8Y{z&kP8y6usK1ATPmZ4_eP(=F8w1wpH$HRHzk}@ zl1~2K%ptc_3`pWzPaLU#L)5uH4!XYbY?6+xsO7Z`4%fAjV{6+tpOZz)<1vG0NV+!pgSS!#q&hXTdT<@g0^X49L!TvzoP=j2$q~0CN9W&wDY{+kn$}=wX^&xF$t!SPqyEsT3LD^a z-+ge5Tu3gaM?m?wZ4l6Slu3&+!I$UkAVf(CcaF27cjkvJc-6oBfs(WcimL2oc zxdyl>A7b=T69Yc5Ubtpt~K6ER2T0BWX5VPwc3 zI&Ix7`s1$?s>huK=N}H_Vn{laj#lHYDrTY#L@?!Z69ErgAUm{#Fl%-ujK~V3DGF~y zzB?+gz4o4!dUF0{Lb&KH&3mtxlIEZivciXw<$uHJ z{)h@Zu%?rSUA~SE4c+KE=Pvn~nt{{T?Zc9UvqZCgGWe-!p}GRlZIf>jf13%guy!vQ z9LOLcV+DN2yM>mC$qFlvnbH4Fe`0M@Jg{KzAeKB5N3q$bF(>~k&*|lv{oCEp|79Xh zT7Hf6zmvxV(PVrnNuc}W7J0979sy14tr@RN z$8DQ|@u{28zGo&Zy!0P5Hnc(EhcrrUr(r5TSD*5B3%EWV51YOJ@-CB0gl)V>`n|^y*knGHP6^5< zvu}*Wwt@4o_eLo>@9G2(t`&fM)Jhoc^n|8WiAc{S5V;-Q^!AIn&@;lDWIudL|J<>{ zAjJ|vUu+zzl&NwCIkRxOzldC$ZI0__c-1Cf*g|-1AHTbh0P%Z|$-KgN)}(1HjR;JE zP^EZ!!XQb{YNId_0kyeBJxaEi9EhO4NG?A)38$( zVB4t%3oGW6%L{s$u5ZJv(#2q0u$CnrlIg@sbC|e!T_$=X>u7QQe#qTeL+ZPhqG)_N zZ1Q%aTh)Dt?1yJ`KhLrC@Ycoi@=M@PS3Z$We?z41YeHA$OL|68pI)v0fD_A=XyH#; zIMwYAE&SiAHgpMf1HQq^N&^TO8Vf_V0T^T1(@e_s4O*@;2%tw@h>gl^|l?JaliIO!@~; z(v*4fc+TMwtye~r7-xf%w_YUPJG&TNvo+lKS;{zf#c|wIl}e{>{Y5v1oDlr!J&yYE zdoX=5%l9`*YJ1f5$nB~*=;KlXPt2`}L5>touADm%j+<>C9z6C`l%E{ML} zOZtC>!VjSy*XTJBhqfK0!tUR6!|^3x`}708F$(2o;wAdGYLQ@VoEGlr4JO+>zp_8~ zP2kLXt`OBq8$25F3@<8&fNnm|Q`$O&-$D?_c;&Hq{A5}yR#tfUStZdIUj|01y4df$ z8nsWWv3K17BCpr6t`8=`Sn;LEk1ic)_|xb z*Fn0-joo7QohkaZ958Mr9iD6pg?^9Qc57b;! zlG+!i;bCbNGAt;Bp`YWJdxvGn&+4PNX3zvWBC!SL-!(Xl2*j;vt^gT=l=D=^#8>~cDcFMpw{+#r>G?lE- zNk^Fx`_T*{X>|E-^2x;>H$C}-S6k+ASws9YOG+T|bfe^L)k1zPjI4x8qt*k#WOr9%H0oc6PY4og8JMmbnE`znb7=J6c@RBT4sB9GKv+Ifuq0a@4)gh=Uj0VWc2piuor+?H3>s+jQYrMP z(H3g&Fcdb{oZ)g?{cCUCQH6ot0hoSr1cY^s!ktV8Cg*KIqupci#g;StK9q%v+$F02 zm~ab+W$}&o4k}YWnXHJ;B$L;UfZ?QIm^P=4$X_-Ad-E!C^LH^j_qionG>V~e$Q`t@ zd`Lct_TtnvQ7{9}Lut}g=psAtX0R^3aG{8t`ItgnA4QPS-IJlz^fj~btj3k{kx4k= z;0cJmH5HjTMZBx}1(g>|vKy4VO9F=5sCuOny%adibMKxK{fqhZwm~B2QZRxu+pEO| zp31?Qj?r-N$TaScPdcYL=Nl6=`zBd8Wdb*ROqFM>*AOqk8h(E<8XjcXkz|jnaQi|g zc|6`8$F@%e?|30hIW>mO${mlRv&G4>z4mlU-5&buX9mvRS&3G?R=j7{74xUvr(5T_ zQ;oo6+RbOeMo;hs;fisr(z;e6R8ofrZKhy5^p1v{ROj9(WkLtPugd%UinQGMh66WW z6FG%&v=feCu8z!tBWDKb2D^2{SXvC+ZPr6R?`tohJLT2vKnRHQIA)B&WVZ zpBq$UF@1+)?9>Ua4b?x~JolTql_SIiJL| zE`;Z~CvonD7lONI?vSwKQ5fAZhg*OAGtcZ%<61%+iS!r`^zC|wUypfV$KOymRrdjp z&3T5Zf0`*w`-d`LTv5im2E{CcY1dkw9~~Qjtuw_@)oKo^ADF>CJ2yZ~-Q#N>YhJ|j zXT_M;Wyk20jlL+iMwc6t{~h-%b4JyC2E)$kf>+>glybd7=1pQm5p#<$#_0$>YOBd- z%#1MJd>h}{bDFOKaCN_e)W81WK4VMQ_eSif*8I@ZMEEPj`?Cvq;1{L}&qKLvoz@(;|G zxrlumS~1fyocMJ~fKGudM2qi)XO^1e3|xkWeL5Jh zjQG!~#VONOQ1|c`GI^_9ZK&309NxlnbkBJR)vb$2gx@LdzT-8fZekry5H3O2=d&Sz z_esesc2n&tZ!mI<$4fUoklS4k2G2!2-zu6M-;{~27OdcM(M&S1Gl^v#chQ2NLgFm- zlbPT!1GlTbp$UCUk&t7vF+qWc5##oPhC4oHr=Tp=SHM)_OIr0o8DJr?}TEqZ^0G1sV|s_ zmK);|q6TXJ0bP=Ko}XPN5cgtJypa@!_JQTt)anB7|GLp6nKz`j^bgOx{X>kdI3U?- zkF}G_@a#x`talH^6K4)!OQJd2I$y_ynbq*HN}J>_j)zx!XCd?R4^^G_6qR>v1gRf2 zBt&0Ec<#_Qs=HN%^B#PKi%p+X*Zz-mkG%x9CpZI(F9d?)!B{ZyTML7_o%CD8 z>>$I{ww=VBC3UnVK#MFic~3%wQk>W_6?$X!RqR`oMU6i=Qpbt?cE9Yk>)Unc z>D?D`TJcwi6rQK;H|GjY=|*6m#yWZ+EFN{g8*#Q8YjDG9YrJWD5Z$9Xn3r50PFMmNsgta-K@lk zTw&@#jQEvAH&m5kS)4f|xyBi8@w_zuJ2|i^d<-3tBZ2Y8@v!%tBHCS5!p?jlJL0Vd zwtl$7-dpqxg9t~$=Gu|H6$y~db08Lf>tGAwms73WWTMnlMi->aC*#aiVDbHZ5S6xx z&ifFDk$-pMw|n!kccnf~x^oVDvs-9Wa118z^y75(7C`p>Ie7B=RK7==L#$+0QiDJ% zQhf6_5q9h*nFFs##Ied+%k6WC+K~!;YPBAQzet1VQxU76Wy>9K{)tYBzqrw{uH=Z1 z3w8D}ql!;{lg-n+aOWQZ-kmpwTVTm^k-pUu>qbk+O11>C-YC5JVl6r!m&Ah3cg)MN z9aLIp1TqI_z@L2^xCP^sF*-pPMmhL%RqAKye-+Bi<`-J{o_Fba&rJfSLMzmlN+ZrK zzsOjVFVr#nDV^{u0XrZEhLSzm-r-;x7vTgpyN5-`Dr4zw-7KDKw2v<8*-ff>ul121 zeBUELmn>>&Ad;&mFdNI-udJ?}fh%|Iq6>F+&_Jag68Lx=JL0ej1_YlWAMULrd2hB6 zyYV0JZ-XcC@9rns#c8D3qX<CNzmN-hz9Q5h$ZtiNX7R7;wSZusi~Zeo96ta z=j~iDd+-r)V~(Qh3m!=|=mI{>4rauidOES=7ADke#b`c)WV`7tx%g-^{kO-I>oYAO zvXg4*%>Goe&qhFGoWG0Y{U7000TEtxRfDdnVx(U6)s>NfFG;@Ed@z*W1dE#oNqu27 zEEoS17Ui4ioTvmEiu4?_~D; zB@p70N4N28kz*4laL0}uqe&JsNyVf=QV^_;60!>T?5qxSmVIMGp1AR6ZdIsqk0jy| zU)Zf4W5B-@SgqI%*wy}&1a=%?jvSgoFHT&BqrxZRw!Q!1A=}@?VcIMF+;#>pmBd5y z}eZQ6L(`2Z0=Ery-xjVN0thTWd4V85yWKVJJtWN$T-)=mlT;_(m~&|(hv zJPk44dpT~~6T$ht=4eNDETQs3Hgc92w@t+ogiXoJbjQ{3Xj2+CA5|sk+R-$;E0d%y zJcC}-!f+J%O!hoDN)(UOv+12L$h8~B7|3%C$LYyXxl%`TIi`CTnL@up;ABuTG zN9Z6LG-uLbuHxuUuI$4tX6vj*(Ytp~;DBoj*?5=Ff2XaW^NS*}KK(5^`5K{o;C?3b z@lA1gLq1v++K1^4K$MAlUxn!2KitUFgspQPO( z{`IF}RmVvZq#DZnS!00D$JCJY5It_=ct`BJKLr!|%(-~m6kHpy9d~x@L*I|Bv_U!% zgVP77o0T;?;I$UiPa5H+pzBy4p$J`vm07|J5oi!X5%Uoz?|Gv4F9RfK$6r4GU1alfO8m0gnb|*%EKtx-%Pyid$MOFPg$M zUEK-gKY~E>$Ty;G+(eGA+l1NQC*WKDT<|LO2X7&p$GhT1rB zUQT*ki<$|RTzLwaSN_mHK;+xPae_D{w^iT*gQ`S(N|x#1+eYoN|m?t72LW+7;*a~`)J5#vs~^9=CB zO^{}6PBul{AafmgKF;V}bm!kG#K2UA+l@walb#tltCfuMtM0JY{22_7MSVj8?? z(6w#^IaVr7J?xv{#lKWGx!^KXXT{du;d|VH9fthu@j7#}TuAf_?D-7IAN2k0h@X1o zxi6w<{B6_8`%F}Dur`XGwmwWEe@-Vec+TjtMPtyn%!m4$*}?Xy3Pd@71I2@62CR#JMh(43~}+ z9)3I%@`q!{GKb(X`e6_4q=)y9m$jskFw4@R-07J?nc^*B7Oj!L}n z6^tDG3LUw+aN&{_?vrm4jgF}$%8N9~j{+wgufuzbHs7UFn+Bj{aS%{X0pCL}beqvZF+*d{E6&CNQ*Fl9E8K6@08U6%vp?z>FE#R61)X~elC7Gl_r z6p>O-9MNcK2D^Z8)UF+-2LeOs$J`G%F6ttEVPFC`PRUX&i%GC;!DvRm%K`NUE?|}Z z7&_aPLck+gxL?v^XYJ3T{aphhFEgL~O!$v(4$EasqxM5f@pqDZ`ZbZ9agcNw7;+{5 z-DQ=Fq9H~58cLn4<(U~bn14w!_@TdxMvOW~k`{&Ghzx2zF~c;Y9xvFzah29_vk*QzdG!FHQxUS1;kao!VT9Oe^1KNW}f0 z=W^;{YQmmr=_F##F=%Vv24uM`eDo+L?B*2^&39|d11>WAYU4Sfvm_U`sSRJM1(Ivu z<49tQkO-&VBtKd*$g#g-!Z_8_=yYiVczHT9nh8&F*1A_j{j)7pJksLqw0@wUn-xkW z+`xU^{vh11%FI6%g2&d5)bhNOAU{m~ak`2ML8PG`RJNIzBPt=I@Tg zH_p!3){VsVlLipGi}12cmCBwngROgmMQJ{sc;>kyJwB%ao(H{SoQ`}aeU*-)YsHCh z=~@=%9r#7%JsiPS_ba-x??r!(oTuI5x}2N19JZdSrzgvc7^AKkw0NE!yg#y+cQ)?8 zj*;7`RB;9R94TP9PsMgR4{u?O!Z+-$OQ2Otr&Fs44K#f13SvW2u<5=FguIpH9F)^Q zf9815n=}uS+6Srp?of1lZ4Rpg4B?!%K2$FbhIiK6*o?`K$gjUu*qlkYmCcLr&B-uE zHoKL3wC*mw&};^BS-Dj5(QIrUl)>y2L%3rhhFZGg$>Q(%^p&bVdQ>OUIp2BbNNEtj zl!t6W&^1O~u>x$?yd!~P`-zvbICgKkMz^{Dhc^uCFfu(BFIBqGhFja2xDkf1LGutJ zTo*+~tux1WsuHmLV*zv3E)65gD+THmCfI`$xn&_$psqKEB%6d2?m|9%IkF4ZTs?!v zG55*HZW-I@kR+mz4W$v z1-$1l>9Z7OCAi>sDLeMx6%KPG-JRD|M(C549XMsOchkJGX1A_bAfS$OKd&7?c)9OQpA!RUX}uvK3l zmpT=)n;g?nP~DE?+fvN0+d}VGZKOL@LST6F8)o{bqhRjpOuwcqC+0FGu=tQUeRym! z-8Y)RX1Q?K7qyH8FTIEk!#;4=as*sQ5v=}Tj!LoTsELm`2L9f{F*RbC@Nq1gZYqT_ z@<+k$K47(LIxfvEB%Jq3I63D56dur^a}}d8GRPUvoXSMeo-QUa?kukTevf`hRYLVJ z8w_j;p(Z@LZi!qdz07xLP&tBpthq(S1?F7se%@}!rjZjmyw_MJ3UQG-zQ5)Iy-#Ox z^8$>~Vru}qS@50a%dNzVsy^^%`zI29%aD%HT1Kz(&l4Y?DZsC_IyCKL5Yf(Xql)JG zv|SX8(f*-$t*e6eSX5MRSL@RQHwH} zUcMW!V;NTai@~zbaoG1Ogc~fe#6NnvkoOj;!is5Nt*1?QWOTz@g9hUM`Ui>V;=s~2 zA0@VwGLLV|lKE@P>6lh6C}-_o-o1*C6?7bd4OJcLhej4#TLUpE1_i z1MAP)3goP=kgXHX;k{HFL0x$h30oWm#fwTo%EcRY9+C%{6%XP0-&}IWSQ&h;H(=A+ zLRdCufL^oph8YjlApdYRj+4;`{r3&j!{IZ!PhSlQi=B!2(lBNNzb8Gh+>=y=NfDEx zYFuL%iH&Xd=$f>qYUNHHE?j_#wJMO9W({HsU(l%WX>{qTBzR-{QxrJyE%_rW0;BWgnDyi% z6Pq88j`sd|(JY?EwQr)FS21ZaWuP&U=g&3ik%V0r;Epn%{o7`NewGnr@ltEpxjPqT zxZh^BYN~*W-()UObsBfRvl%^Fx8U{B=48a0Mxt(^O}!@_qg(6Ol31R#xO{61b4!IG z-DQ8-*a;RmBFUA0Yf(bqluArXyNM|acHx6*59s(dSxzcig`2On0u%S^gYaNJOiPm? zHo<2xcfnY^HA{iwTpMn&k-%tw@PMUc6 z?jfpH$>)B^1$etDo=x37!2etsW%kuk{~Sl|M`j)QoM?&$cG<+k{3^9;lf(7S9M*S_ zr&q*wV5u$7PS=~j%{p=hqYWkK3r|;2vDkwj+xFPxoz5<7Pl za3c*9Q9Iy@z<#qeJ8z*i{Ky-PtA>8j-@4Xhg6J{+*>xU&w6B3p9w~Il@EEsXtUvzR zoI>X}pGUJfED8RV&1@KF2U~w1MUSz+>FMJMXnob3Tj1b>lfFiw&Ez(kG|rZ&U6jEQ zW41AGW(7g6=_KJZD-F6QKo6EKmBW-xx?ss!frj4%C|&Rr)jCYD?*pPcPUWWTH|6H} zd*h>S(h?5AV&7Ym$?tgHxr4ulNo(KnvxsYO*Fl`3YV0+(7`2Ybj>b9eEFQA zF8zvdKF9(R-|ok=Et#}3CkA#OIRZApCe+?;H6h^_=%m}WnT}`CkGg&KJmS+TQuRQrKADK>vzQ!z$d(HlFmchQ+6TuFge9Mnq}lju*)>vwWlmWLW(pHI9UM1qIfS`cg&!u_8dJ2IgTBzL*P4xW=>te8oc zY!{&Uw3mn`_{ym`;J%vA zT`tAUnr}F%@e;~^kHQ~EVrce=V*E!!$x=6MIDURI*J!jGXBt1I6Md_RQG*`c*trEy z{jDrJUn?MIqY}9>oDSk9OFwO{ik?h+GWVd9^7KwB?4ICzFeq+BJV$DD^NAu2GysKBZ@xnyc!99sTtrAg!{ z?#(_3idC=4rJMQ8mE9d=`0+cD=~!1hxpodF-C0c!y8a+Oe=?DZ2huqXDrouZCbqM*X*W^YJ0Cy4eS|})g(Udv z2(Ig#E3RXY@{Fqe7`erb?pWmqE8Q-@<3kt8R&NW)Au(agLiQ@4E{$-H@876GZk;xPGywyB)Jk7 zU)oz2MN=+0<8)II&NQ4$7k^5jU(Itc=tvT?!A+a)NmfPMF$PS>a$iha@B(A>Lh#|# zEKC)iquZbErbiq1k;~@zx1>>~D-i`voeT+x&_2SIaQ9a0Z+^=Mk&5F&h&u8?fR3 zGWdMlQaHHuG?X6R%{>3O75463g=;Pp(HUmxFlx6Ke(e~Gs}D(`PCzNyemIDV?^d9u zz4OQkubHBIW0D1TmfFCmhli-w3RM*4-e(hY6rnwD9L5}5OEiWak)gMD!8I|Sy%P8V zt)f?Rbs8P;x+R_rKguS~naV=JRYzFc8bO|a*ToNiDXAQ1h@W4spa~l)*?zkNR zG$A929yoA@eX=Ej(3yT1bNn=}@>j$?-5>dGtvYxvpN!9f&FPDN0yYu6)2-$*leaB^ z9U75H4v)#C3xAt);oiAy_=bgKBA1M>P3I9hc>zvp*eOceegYKcnv?lK@g#P$DQq>e zCULI@8TQ#Y@X2f;znlBW8r~nwS+5~ugR&qq>@DWf6pE8z%^b7|!CH(}>8@@rMxm7sU ztse6oexZT&3r222B;HtBL%p+O=-<385c4aM_VhcDgkV{cK6W;02tBFpM}M?|T>SC= z9#d*r&ARG%leZ7EiOJ~KbnUxE%$a$M$mfngJme(Hxn=1wb{5`jmcb`nI`=LfOZkJ= zs>>kD+oQJah`i9e`7VjDPo{%@Z^$g$UQ#L3MCDWzVR-aoK7ZOz8tfI|KR@2TGsPPB zq+OtD`h4c-nK}I>@e0=(8lghQCGt1a8_!P-p{pKRP-Ukw={9eqxo z(LP$S(};-m)x#(?6B1vZ2rpGf;nL4H=(3^=GLSX~Lwh%}jVaIBb@Vpbcu5f#HWg6Q z1~E=!n?G1gScbyIXK}@6zW*e?63>nMO(wO}lFLCSaej&jqWabf!h@W_^r089Hly(A zCowv8%Wc}Gv=Hw)Y{shX=P@TU3Ovd#5xqqPr0k9cbvJDwUK6aLqS04$`ch5pubMa< z<-7qk_DNu!Ln=8vNej!@%Wzky4V3u(CyE%;K=$#O$JW@>xZ`3YeLcGXz4F&W|8<_D z!r$*bKA)lq&sO5z?{Z9TS1sLLbRDHd$KV2=6J5Y}u4H-sC^!BxjX?$4k=UQAaRm`4K!>;6*L8r65~w9Vym0jGY<_so~UAuq^vc1Ia8pLunaZf21A0 zJMG4ctWRkjm2{>B;fh(HhOqYDcSW^5t^O(47AxG zd+Y2-qCDXted!%ZpYZ4L_oEBQua{5h=3Kz%`PQ^@^Df-Z=bmhpkMYl%!RXt`kRVJU z{!gN4rMLw5`lmQ7e|U|&``rRFY?qMP&Ex3nFRw|*p;4$QpDS3uDhYjRJAiJYWSr+clDm+qKqX3Q?2Km~B-6i+AXgj~z*1j+dcnD%Rn%EF zS9^$E7#)i}CGEVwsfDnT`Y;le$dAJ5&^Ks_Czi|70lwpKaGerE58S0Dnv1EQG)L#V z*Rwj6PAHwhphf2ys(-zk+V@G5-v^)3^gH9wzAhIcWzJx7tuzkRS@Imt*{po44ECJ6 zOr$St0@ra9uq*E?T@Y)>JX)_x<~JI`{*kueRmRT@hGn>!VP9x{b2vBRMI4UX7lSc2 zD{%RI{+<~`=(9FmsLPXpKO?4c#wtW0Ya}dw& z>$^bux1I1$^EZ;-)=Ae*w1-cl=EIM*KE(LcdVKTH2+S786S66eedXDSS2QzGu8O1m zvgLHHQ8xYXX*1evyT;twK1993*5Q6@UDz>@M$8?YA*9=iTR8h5-Enam(=Zq(lG_9{ z?+ot`zb}K~x)X@AXbPEL!n;%ohv=o62GspJ2^@>O$g_9Gw0CPXwfJuxqlbg^V(UY? zOkyprU!=#ywtnTg-Uu_qc0q_IE8RmW{syC->Je=3k5h`HV2EvepDN1 z%z_bm?L>CKOYr!ZIjvNmNnLkOhq4KoRQ$Ik_NP}czkK7UyX|?TL8_EWn{s+PC>5`{Eg%Ve9@@>Xg*s?u zlW=8@oge?4spGHnT5TsuN4OIVb*_bj&U@f*;6Fj%rR#{Qa&)efGIkjI;T6SZ=4)#k z`Eg|cFSkl^E*lfEB4j&czfWV7&E+wF)E8>cbIjho>!GhJv*1U$IoT{JjV3B1z)rlK z7G3Ax*B_#Ax^g6WXQYhJJ7!?V)kWC+@PCTVG%lyF3&YK$N-0VMN>Q3A!*lj3N>Pba zN`-_ZB@t3Wg9;5wG>52ABuez0y-pgWL6KR8@K@b%` zr*CMZf5yhclOhwC?PJO=#f4DX(nGzkb&)qWAL7sZHdJwGE4i6I3O~&$pvkXFg$iMP zWbEcNrU8s#qO&A$kHc8wRpdd20pys8fP=Rj z+)RE3Ca*HF&;J2)R$Pg((-?*AzrRqsiFe6G^F6d_vH?G*QsEqY3aD#A2CIBy0MA6c z#SXihG%$?--!`PzbFPv^mF>*TmI$1>YZ+RO?;~XSQ+Vs!#ZFoM2Qtr<&uV_Rj|6#3 zf!WX!(EsknELq3*NlbNd&uItre0dwy&y`?z+C@_3I*pT5E=L@h>^1loEL-#TON(7BCsB&{9r z?7IuWI`HPBOW4phnY?26vi&wqH0zHtn=tky zDX>~eHcm~(t0QC3RCE?IF(;6@(V;`+pKZX8$Cl#9pIsPpql3J)DH1R*rJ?PG18Mhr z!ls_=B|fTSxqkQiF!^pBu$f+*?Uf<2;lL7D&8>jg#hI8D8&3?IZ!&-GZbj!8V`=7p z)7kewPN3=V7D(UN2tc+R0SP~;V&+lsXRA$0hM@5*vynzZ1b`wnc5PPbUbLZToWS1vxlcW_?f zaELZOTb@pzIebLTE)ViKzZ;;+5*1f>5CxOXoYbhLOh>09xulW{4sI%ZcB26+3v=iM z%WUk@SOwc!k|5%#J??sU1go2?=|QhSvi*?=lqJ_<<)^J!_+tb`Hsl~g9*5`OA7b*r za!zS?9LPM4C22aESYaBCr`BC1vw}rnzGMe0GRcIoiT}xGSoh=hfP?&7c@{UV+Yu-6 z9LqnIS-3#p1%WXg=)-Aqlm4_*8|4NzsoRHZB7BFZlVd%xb(E!Kz7Q&x*H{i8$o8a#k<(;3W@j`6HfPz`|kSGM|rBpN;b zh)cew64|UvWKV7ylNKTdd*6#-%I0(`Gxe45OVb-7H+Lbl$LYb>vhluoVJ zD03TEDR8$%^0BMLmZrJiWdCGkppegjEuWtOTli^{uSqjt_qk5 zNAY0TE?CfglY~F_hPao9XBBB|1N@fAXbrw$q+BTtdZ|aE#!cbk#57SxD2bPJcGK|o zD(0C8M`P!e;G19-{821}nJXG0@`xL>+WLWMj2~TNKAvZq4fL0S z4>x=^nO3)U(bsa;+!+4n@I2KF8*KvcVzChWx+KuqwU5L<{X;KhyJCq$EWPt?Inm#8 zfpuLi2_JU1(4o*8uD9UlS6@^k$1_aQ_@mxX@yMPQx#o3Xkm527wS z40>|1;0Pqu-A;9;ze+x_u94Glik>>{Z4>4CjMM2H6D{!H*hZ9RuVv@{RH34(&9O-$!7P#o(P`rRm7nm7ipBD4fXSSNN%p< zJHAl`=u`KIXy5jrj*sutYd;wJrBfRZdgsum`{u*8cW3CC1MPUGHJ{T@mY{8J@7c|B zH*qVI0->#q&%Y!XkY5!8n9%YTeT|w>)KCZOj#_Yug)Ufc{e^05UV+CAr@-6dvsA3z z9y%Nx$-Z~JG_%u^nKgHq@6W$uDqH06)}&+bBWn-c>7W4ai(aEK-@GVYF@=7c!e^K7 zY$G#{OthIT>V}!mobbe-o2VP`m^|!c>F|UUPQ9lbbjMEO`Rcl$dgm~`ttx{PZkLd= zpKUPl*JIin#q;xK`m!c1gMwRS&xvvAX5iWUOmK@DoSAKi8t>hO*&oi6n9%LmrTdPq zTM`Qw6z#y_+$q@KB?5c?vm+BiPO`ouJ@{CBAs9|~C%+C$)0EK>=+q&=kk1eB#Eof? zzkGmct@XnfnK8#{=$H`ok%q8b8x~3lhqQ;gYVgz(^yQ(l$yXU)I%m zWzK=KxlQ!@1rbPJ@dOs1u!T?U9&CxkGb-f&-wWR#rQ<);Q`4XXo}<-7OJEw^w>cAU z7cAq%GK+ArTnlErh#@QV+v#WRSQ;$cMUjsCd2 zT@t-3Jn?zKUwVFM6Mi`whNAP*1bh45;YFV*^zDBisf&URm-IcHWKWR8|7LySof-46 zxp4srTrmbK9KW!I11)3;&n$X!V+RdLd`aE}DWUX*CSqXW%T=8lM^0Z#rb;F&(C@PW zO<0&f3Z8ZIxx@JwFl-0?5w=XGR|fuPt3|R~$8vKfRl&2zwdC=L38>wwfq{5iC|@%| zZ&a^`(Apou6Z7&>;b#pyWn&CYY(7UrqbJem)w@txP9DW?&E>40I&sUx5!=U%=Gu*m zID^DkR5)Es)grZU)MZ^fZorUT{+j4my9cwA2rRALPB!JPf)G_h-t!8=?VfX(eFyr$ zt;z{r|JCCeiI?DElp~%$6bq*x8=|dl3M$F_R+thk%Tv+;PQK#ML;`#ugo9u9-no?pZ)aTfi1 zJb`EQeB@m;9jNXxj@Yzk;*O9XD3QstGV4c2iDIYU8xg*epLM`yi{L)TCJ7`R=AdUdSADH7M|NZ41r{c$Er{Eou~U#^i2(|h5{86QqZ zX*+5Z-$mWUv(UwuVaq+wkx|L{V7bc|B$p~MwwJOoYJoU@GaUCXuXFt4PDXS6BdyAgSwA$Eh*wf>=}KbgU&BhTrK zv_5if)I4aj6A~|Jb=bLoH9Z}}@8qYvW#*+)a@Ic|g0kCyD{`V4U$W>JgD9HJyPV^V zExAY2qv3Pa024r1GBt~LbA(xA`p)s3bk8<=&5-A2&U}O(&T?39I*;t$%YaFX7OFHT zyzB+X5+3dc$EDagDrM*G^u z2-Xo$V6BQLjVE%)f;C{vsE71~{RVKk$omUA*5T(6I~0{D#Ke)4I7N!@Ph1HE#r9Fm zI9Lj7Zvk~&SBI)wLt$UvQj)Ng@0h2R5wQkCTH6|jKby;7c1O>=@jLyS-((q^8;Uz)u6rd%D?No>BYqYXkkBgUBAb#dwC+F#A#-GK;bdF+6Az zbL2xA-dw&FJLhhNWW@z@6x3>=f6gX2T`UUPb1ZG<^7H5{i&~O4GDz>;O~RiJ>!`&2 zliUoWt5nf=0AG4-z%gf(1zNvP5|L1QeBC)8TO$6j8>BtC{gZt0gts28EKb6$_eG&_ zoGe#;ashtg_lK&^L)2taAfEmigkSDB+ zSEiCseA#mIFQ!3m~ys3m-qyWVim3$4G|a(+{)YA!cK^p&mZodYR09^qXvq z5#g>y$&tE)7cux)3TBVop-TeFpuT*7-POOEv_?mh^^=lt`HE-s#HMzd(q4wvzh&u} z90j5j=mt3gX>e#}X_1j0crvn7ce@i&%UnZej2(rMZUwj{JCnJtV+h*>qMV13J2@Qc z0*#ydpr^hOB~Ils=_W0>>)ZyK_B9?VP7RXVO{2)B@(>z*)CM~nEPOltim|M{ z2aB{#;m4Y5OwFr(u-e5Q=0BbRZ&#~xfqK)Rp`!+#9STCGO&lfn)T4H|w9N#QtEBy- z3{=$wagwJ+P-|g5F}}k)g{I8G&697k=H^mpujEAp7UQsTdOU;~ZGgE?Kfu)tj3q-P+y-;I_CnPSNXS@=(gOTVvYvz!HjTwCJIS# zrw)63R0!mp4y0QbEx<{tO(d>iJyw?;z+D{<_&Y>|itDFQnPY7@S9pOYH121@ri#H% z-$0UZ*%(>>V-QfMVzcz98ssVcW!P=^nTxlE2|JiV?Ve4esrCj`_R%cRoGqNtKlOHpIp4*!GF2Y2yP z;1o1$lf|&`O;EqRgMHzef}^@p$f ziZsB1t=FiYcqBeKh?~-qrN&lOv%DFf5-N(E{@=jogd57HaxRs_&Zcz%X;8`AT z$9NIp0}Zt9eoP;muYi>8FUf1cBl@puG?*QA#3CYvYsX*YSwO8sjb}xRMqj0xUh3S6 zvpU#zq#Rpbf5F}DZ-}~O8!p;(0dnT?-i8X^n<`m^BRbwV6kEv}>K8JrRF3mIvqh*H z`V@jsGzbnw)d~;toDT(d8h9A?G0whjAU~!QovT#2!Ko#?jZ&pE6A#Reds7W z3lVn=@XMMW=45|6iOWBS9;YASI+%uvS3be=ulZzF%~oQ3Dh7U5yklPd7Z3h1i8SNU zXmAaUCvdl&TsEs?Rx%~D$XylN+s5PKUuiT|uL0DI4#I;00Zi}|<90XCz`ci-kn5dh zM6*1dwGW&_zwNt3VvWu5a&kOY>gw1`J@6A8s}V*;%X50Ooalmtjd<4Rn}zq~qsShc3H%&nA~;iJ2=IM?=UpD+kmnODLH3_hoi!ADd{5T6Z}xtltJ(Fxl}XEi+Rs?f4sM+l6mV=$uI5G%$v4PBq)QW z%a?p5k5hQoV8>Hxsq>i>Xa=B_btU7su#|k(3nxuBcNv*(Np9kqN-9a;3%+~1!rg(( z0=qGjG5^>tFg95WdFwnO?)X{UQ!l~I;orHNZ^V%Waj$9Izk2w7unG1lyOJHDuc2EY ziw~c#qY+n%!J(6P-LzjK`fpd$9oMhmXLX*1yk$20Rk_JXxvvGKVi{<8v=;~D3YaH< z-?QfmGC;&}6B*B+*);+W(D$1tRCm$lw0EE2{u?}u|9wrw9MSRIpI@UaWCH%dwwtyb?wxL}U^B9X?9^~OZ8M@u;33(pxi{c)97w?=Zh;A^);wn9e6}>0) zi#U(Q9vn*a6k|j^e;@1}A*X~RBsb{+u5n(3Oc`Y#f9);D{x-e(ocO12NF?!GvO6IPGVu!0t*7$IAzsV8L zCv$ODXbsI}CvmZ}Z(;g^LMoOu3dP^qVyAa9bucZ2o0*EB^KTMS+da?5ywL)~7R*Q6 zjC5R4VS$Htjzx{J#c(g!2FzDc?0^!xU0eGa!_o!x{0{`g0Oqpi~F;kV}!u}h?t~xpH+$0@1I=6;ByLb=w(k^E3 z#eKYd>mRd*Q|H#a(Wd-Ki#sPxp~hA-(U;F7-uqI;ceX8XwcjgbAOB%~?2;2y1>Gbi zk1xYz^%!C}<~rT+E&)D#YUa<+B6Lg+hTEg$(W~_~Hr<#4Q&(X$1#T0l3ji(v9T|6W4AkCNfK%V%fgW91tbVJHger_EIpZIUZz;hFw zeBK@FX0M^9;YZn9M~}h5rb*mQ17o^##Yi`gdML}v~L;eq@i;_y-(jbkezVB25z^TP>{G0L5_ zvMnS&^DIe_v@uRH$bc|qS>m73PPU{?73RH4rTS({HrH-w+f4a*o~-MR#V1d{qEy0B zyinqRe~j*tVpmxlziSd{(tFD+>YK~=+5VAlhJ1&2zB+CXX1KfikU89YnaC6tlY7BS zIdkDdoNlv=i2OYQYrY?26xNmC_?30wm!pCbcImh)el2MKl7eXEYMMLUg0qxu#$VTU z@s(o+DL4^9UkbaaigKZFQtEX2tns7pl#&vpZ)%|hcV>|%{f}|=hy?sCYl8h`5hOh@ zV(rh)hgS=((GW!!ns;|C3|U%1Ti+6R^lli8i7um}A`aI&PSLzuPsn%kInaFM2x~GW z1tP-=QTnzqw@YFz{pqR#^PVK4+o=RHm~@rh?750O8jyfBMk##O+5!I471Cd4MCpI8 zrHH~OYjidt>|LHGUv6%M`f(EQe7-#D>>f*E&E;@c=Q5smzXx1Lm(mKy9{PUaT~xoc zA2p7r(c3%YQABwfHYKgWqM``M>$yq$o#hEUUWE)ZPYP0L^r3Dz{6-2b)PD$$jN3N5@8otE@K3CB3H1T}a0#^e z6U11=Tt!b!JCGjdA)Mc{g5aoJd}iVmW4z9AOU9g5q-GI&Ud-$q;sd)_z>W zcP;+D_)Tl|X4oV=-A^t#m4n9SdNlLp-v>{x^Q`+!Ua&Jn4(A=hv;-?I_Du%1$WG#( zyh);ZezLT`mS=9)Tf-usDX{0;cE;Vplras-B%fjxF?qO=SQwcJYWc2Y>VvCTmmx(v zF_~RC-igmRI>NRSZZ>`c?L@~dov!k@j9=%9p?Gg2O5Tabdlg%-Xv=eU_v%EtZkZ+* zsUeOo58`O+4=eiAp^k2LR)v(h0@QB3E?hiv5WIdr$L=CcbaO03vHsO~;Dig2E4)I@ zK1&jH8J=O0{gL*&+tCRV&au9z4)3i5+|Wo33bWT>6(flsM=8^4=V!z{I1Q@bX+Tf< zd{`-42#20uB|-3njws>8Dy<_@lTK=;{i1oK?gO@-C_;nGwQ!2WOMNzhkiUqzxWgVTUIRZ0H50 zZlWgR!(7wOgzkgyiN8bu=836L|12+_eHuk>Tra2Q*6uK2MG$j2>JoQOa~N$Zc9G~$ zmS7iW$2~b2gfs05kPUne7Uv&=+q7=9vol3gb8q~!)(z}_Tchcc)l~321K#xB!h*$o zuGo{Gr{rbhX0Z(Py~8tem$~7k3k5v;!iRHpn8+zIhNNRXKjZs-mA9IT zj99`=a$%D_EDpX*-+$%v6w95l?u90|Beo1jcoylhBdOeyn>)Fv!{S`W{8waNR~&VJ zTgZe){G$=)ccD}EFPio37gIg98}-Y(Q2xjhy5Jj7PT)k&^SsCHkN1%)sh3HJU@1iK zZ+5>6$Ao5US3;NA0eJj{XIpMDizpnO`IzJoi>;?l9l%s@Hr%Dcp+MeR#Skg z8Sg{8a$~TY;fYr^+@jxJmf{Ot3&!U|GbUa*3HLr#lkqRNL-w#XJpO$juLekRLqU_# zjo-O{?Ty1qzc+Yji5I>;$>%ytb?fX0>R1^GW1{urH~pl(08gFG!(|!=FqY>kikD~! zZ@e4D?OAx14kRz*F5GFR{@cD{13%Lrcc7b^|2}}@Q@5h>w}Zq!c9_~<9wzXv6*_#) zaLVKfXg&H4`e^WbvCy{|b7cia_g!Hm$B*a!>z3lq4c?_*^#gb^=?9*cYsGt8M{y(7 zJ#?~!fVQ=1a~}RZxN~_vT3w4r&*kSy+M7G((fFBIs#j>*x8Yt< zQolBkVXm4gZ|DT)b>{G}xq@!8O@rMYX2c}Y7ADV9!HncyGDAI?4KPup9&Uxe|z-6@9nr%uUxr@45S! z{Nx>@)tEt_a1o$ndl5Dn8{_=B{5{Y^4ihBy;iN7N2>PLclb;Wgrxj{8-HOM_1{1y? zKP!^hD>cw6uQ9OJ#f6^S!qGFw!a<^76V9w2%alr92H`A6%s4ELJHN-k0gpPYjjsl~ z_v1m+WIuEDqb|8Swu)#h%fUtKljwB&?cnsSlS;R4r@GboWR!>tY;aUS*UzJ2{*auF zN>&Kd*|?0~``xM2xUWF+{MXPbeul8<#&S^2up)8a&eOBk&#|69x5@TSz->I!?ZBEH zT&AxP*V!!%sbbI3Ls^Vd3=PK>+Y-5`_aZR$`$WPo8sSr*FS~B4hv4FeLHf%%hBQC? z$pjgT(^B~Xd?P;%A1^%(x@{J4{f{fWJVV(5y=j6Bdj*`C1KK|`Lqn2KfR#JLZSY4BvlbULd~ltwt@GZWr@q2Y-$9PoUy5e8A!ZIS3i$K{HMPtqud`yFWWaUu`^F#+F~OTBreSj z@w~gwhC6U>{aldpK8v_+1v0-clB|oK_|mw&a?SFe*YsnjrPPtI+0`v_S1X2FH@V0qf8lN ziPr}nkcN#@z-M6-RX8Tf^o1(1<#?HHeJ~jhDpY}z%qKK2?_##ws1b49BC_2%o9)>! z4P(#WB3GxW!|4Ma*smpLbA+e=JY1w{6C9UJuAF>9L>}}~x`5&;-+{WxdOWKw;5+6GIx zi_}3sgZ|XHL>p#w(2LaxREg?yaRI|%o8&~a_UACctO(GC7J8!5n>G0GU9dgw9u53- zh3NZ4fiB0fL=12Z8J}zZAOXLtGFdghyBK_C z7N}M(rhi|}X5Bg?7?)i;{g_RXL2r9 z-cTd<3PhGBgS(m@{NuZUA8xHBX^(AS=Rdw%E?NR~R6X^1{enHx&u7`4`>DQ*23qjm zQwP^aFp=kem=3Do%N+vnePIZpDGXOsHI-IXvZO|a7tx&A%lAa3P>Yo%B@wQ$zV9;= zZ0ir3XZ6z~v#V%#@Cq!ddqdx>i6_coI)YU>qtGs@oxl4{#oEdQOb}JWRU%hu_o6d2 zTU3$!OOWU88#v?X2PPvN3ffiUr!FzOm06h^?X1YR>-2Nx80c@wp(_^$ zLyOHlM(4q6Hk|LW#y*?F8LJebltU&dj@=1!GP~$TzcvyQ;{=VmGN5~sXC42sWi8|1 zkOxz{sI$8=xQGFVcrqLqjySM2eFvScqM=dj z0)^01!U`@d!Shc#;a@-RH2I-!WAx-K>-zBqlOO8J_BiK|>%x1)lC8vJ8&;Fm2Tzc7 zODT1nbQ4osMeyL=G~zNm6T{UGW75h%?(X{MxY$je8=97ZFWIw-N*7z2mvaIspp5w31blemFA40{QDDG-b*XpOxP1U6W+m*QgDXweOIW z2ddnUig5ffY=U(UW6->GH2Eg$iD{LuK(gon%#dhev_4E{6BfC^^QRwp=jIYJ+T#TA z98{t=4^4r~pLMyvGppGQ-TSQZ)c|_lQwf4^og|N3rLjUTjI15r11*|A=*vWJ2)ts? zN)^2)+}7J<@PRJ1_{ZT9hu=84Wj|fxPsF}|+ail3v+$>!@*;b={*P|->X2RwGs zXL3E%Cg~dOyLAi-G;N7i{%zp{e?L$j{>Q%WzQ*sJXV6Or`-%S8SZvs3#pj8i;IVg= zoZo+L$nKs`duLSBh3BTgx;7i)T5%a(+%AC1VrOuc7quZlu>zw4IXEVApKRiDmm#y~ z(3obvEA5m3?3aAF7_Gp4ZmOnE-SK37V-8U_yUeclj%BnBX9=I%iqaVlJ=EDb7G6i2 zGetepIOfeAP^~G&czUXCZSXg8wDBmENMzHn>~S0_NpNhH68-P@V}`!d#DZ`IwCl?! z@ph-6K4T4$ux=&FZu7wNTQjOhrG{;GkidN^*OB=?5hQeA{^-msFYW+vv zD2b2>Muo6EEd+{O)95$_zzzus%zPU{#Z5c#)h7+m%M&AcdE+2Fx(yX%GwB?656Dno zi>`f2C=%Pt&PUNV^*r28^&o#{Cz6O+Lhy0+h0~@V zfP49XJW=w--OY!vDCQ~knt6|Wyq!uN_<4coKK?WRY6)E-^Wb&aUozsC$McRH;X}7C zp;KqV$LVEJ4Uexw z8N$!ByA*KgZ&gw@^Bm5P8l=T1lhN(g1n{hqrz0z+Vcx^1Y|%G;qWr!QMFO^B`payr zIT-?tMURP>bsAm{T?o=e8N8EpJPb(`u+0y5(1QS)mB) zUf&S@F~3UE7Z`IveNp71!fpD@YZN&$(+4N72}jqwqp-TolnGeyg%KE5;nweWcs{B( z7k*F9X5v&8GDYPfWV#g2@ddC%<5NW1pQh7!A05cFo<}5N>Ui=icmd+LT9~r$ zET(&>kv>g6W4BL*7VJJooj(8Kz3Fane3>EBsU(D*%LBo5Bdz;c7zwH$Mhjy1#KLZJ z7~a1MU}q?Z@$*1Ry}2?>7&ixfbR1|Ki6=gZl{7cIlic=QNq-63X{Cb-<95jvUu`L& zk1v&zM>2YFOxlT6HAx~#Jt-vEIN_-^)8IoDokhn2Wm_mf+2kCoyN}3rSgZi9Tvm zfHjd-RK5N>y}R=P>uj+C&9^eRYGM|n5$VF2@LZMkOa79ARRMUQ=`p=`D*$t*f5Ris zRuS!8FX@_SY4{S{AnbmlLrk9Cq)Xi=(6CSW#Pd-T4emP2d)3F#x=S^5D8g`4G`OlPwHi?8xrq+3c#lIixDpj&#}@k%b}WgvO=5sE~hvQ@8J;wGMGG z_e~lMm+EsFeBO3!V*}LYOvg`YDYUVrhMXJojUI}X<5mn=pfBT!E18R!tMVAD@64wf zS8Ta^1t+){YMY@cUk7UvlJM`L5y6+OL&6s_ck$KfEC}g-O(GZC2-e#b@juJucHR&~lk93!icPN zb9FAGg5MFenpO#S-j&b6_Y>gZw=k~A>k>ZR@Cvi97UHw?a{k$KAa`ANDMBz@($2>_4Hk7HEy?iL^B$VID?dTbSUOHIb1Cb2D9#?bjChR^Sgk? zE!B9yC7%u#{lg_06G8snL)I$d5S_N-lHm2>MZzGZvshc5Lmyl@Mb3#&Mz78poU1F( zy8F5e|5JX!3#we0%kH;H%#}jHqRcTgtPrsNofW!dzQh%YE4en)d-VLXpG@r!B{=-A z8uqQfPYgFc1-+6BrMCKc*SN6>?!!lq2vLc<#2BH2ch=l5EpBgWros zghB5$am|qrM07_yIcIc=D2CmFrK=1{0H*P|41Skcn8!K<$uo0u>xJA49ZKdOpsx%R z8Npgh&>E8Cx?e1ZTSZcE$+wdev$LY|B%fS8?u55YBf<8$6MLmGnQVCBzy?W5@^0xk z&e&uoU9T5MOXF5z=(KcL_Qiw_9hwOq6GXYc6*;Klw*b%1pTVhYJ4JU(q!8^GAt|i> z%4~ArzqN2>2}L;@{W1%U-7>=q1IZ^cvOk;=lMrfYvz*Tjry#YqXnvuNZDw#1j3ag z^KdA-jcQ%$W8ObV5>WrOQ1+h;%IACHLw$EV_M)B5W80yvKLR3CdYQSqOJRKR3Y6>`W8E#h34#p1%&<*4U(43jgD(%V}Oz@35wwplNljGYsOQo=wyukpq=m0;W7*zMc;RmfQD5!}*9yw%FyHmu+R?xS z&(`K9J8t7P=F4zX_8W6wPSm4Lc?K#K8)Kr1G}`!Nk)Y>1gYa1=^*A{VCM2wd%~=cS zQoYx3#z&mHH`IbM1|4`Yi z3FQfVSMD5W{d|m?7i(Bic?mQ3D`jRnS^NEbATuyyjR6urV0Z!Mwfyx+%S0Zm?Xi7FoS+I+H$9Di< z98Shn7EQd{$`sWWb&&lkQg9<(9Mmea>C=XBv_jjNnQ6-Vi6@le3VSI;xP;d0c~5rx zLPpNDnSNhy4Vu%FK{Q($8vDbUJ=%HX!rfLh|0Thdl?Br)uUg^Mab@n?xzYH!cNJ&4 z#8R-)TnU}b%3)88DtUdQ9D8#Y5>?%W#LSBKe~2gI%$bcevZ04ss?0!l6%BSQtHZco z(}z&@2qv6RA`$}!i9~x4EPJttRd5?imu4TNq1Q5Lo!v%U&}WH#TP#WZ`IWT)<5jqL z)eJtLi6J>6aby;&44E>Y(dtAUCQdm*;>HJY4|R{Sm$mMa$L{__)bBGnI8evf$XY`R zpNGtruppk1q2!pIH$7KuinC{3LraNp>dE~i{oX@l{7&Hc;|Oq> zA5MvNG_46)L_UmjhWjZd@L|$K*z!J+WEaYFcczR3%WH$=MRgiBp4EcfGn)8f<~OvS zQ$aNb+gR}nH&FMDB%j4QP83gxa!;lXk%epQNw87}D?L&J+FBEtG5$V`?0g;6dY?-t zJN59+t#O!VqQTz$T#P}3PSCf*fZk7LEvrEyH=a81D=-&xJ_8S-D9PL)4NV`gVDxCN<@8qk#JA^k#ANXv*}_%N zC?*c8(geQ!3|9Bqz}dYf7_KYL&5Jt8^X$Bdfo?X<-lzi;Hv5u5(OgIeFqyY*VQ&-Ar~|khEzWk%kYyL?I+>8MI0j zp{~S;U`o?IIPJ?3zo;xy_312`plgdmu4OpeewhFTb>^7idx1txJ=!XnQC8(3 zx%Kur-L0KTEJ{j{RojP^uZ>X5`!yC>F9HA4M~K>x6bv-C;6~RHtgyXHm-Tm&mQls* zn5Jm5{q8oheoHg!o%0Bt1{RQ4a?{8p%}h`YT}Rv>1;WeU^T?|+BHYT{25Q>BnJjOV z;pTbTV?aj&F5})(Q;~7}?xqxebav66H_wER$MfE~tn=h!-!9_(ZScDpFRi)oyL&yPPiaMi{6|DbLzY`6V~%4! z?xn@&1JFd)1bv%L=}1f(OfIX(^SLwWMVCi>J31G2r!>(Zr$$^^=M7sqdD7FY2fiQO z;K-oKKA!zT=xX^Y$Fi)-(zugY?HjmFdIk>?X@9u+8cqWv|j9>S!J`YO3=d#}+>CmN{qdn7!=ZCrxM>$+|-f- zksa#z^2%zcDD4&esohJ8qV(zHMXx|>O*Qn=aBy5Qp19@QWVFg2(aIMoB<~r+)ihLL zZD9_0tlUWxOzT*lGm6$L4xq==vD^yjC8X(o7YTcpLhrwhV~5<6SICS(gXzXG_`Smw$&}Pw&%^jRI5%l7n3%66}qi(Nx1t3H$ttsNI-s#L83L z^r>g?jP6Ueu{@bpY37nveh<&>Fu*kHQ}AV=7^Y4WvzgZA&ify-Vb%xU2~R2+J7Jg5 zA@V+5Ge!?nwd8TXpFic(!PGv~8}B7)g2k14II$;>W>pyCucTbu&v$_86Hk&T{!Nzl zXBM~mpB~1nSPs_b06EJ+G|`uXxgB!wY_FQl?6`@Ly!bj>f4CU}c#iUa&yG;P)!UKB zigPZb&*6_n6@2E{8{1#`5feVU>-o)+T2yK>?$>+Z-J`KK?TTtJ{OdM+7Pc}fs_D44 zDTY(X{6#kyDshj8thmY(NBKTrA-yXXLe{5Sz|ydHgu8wgcda}Qweycd^@+(a@TUQP z1|OtdeFkJ3f3D7!I1V(W0`CvE5{Cn-*nM{osXS9kPWMcPvavT{;m5NKJMKG)HZCCr zJZpL0Yz0mv?gg5?=4gyg6El0X3oJhNkQ(O8WAdrB_*8X1xXsIi5*ul<>fRb=t$!=}A?$9N2@1ty@7hNUNyvsu9^bB_hTTojP8q%KO7k^rmL743%o2e1& z;pGoC@D#NLi#45sr0Y#={g&A@(DNF*;_WP$@yien{#8O$3}62{~Vwb=GT?S zEr4e71Z0!r;B|~Nr@l)AJIai)zTO$TfBvQ(H>cBMwZj7Ck0K4#WD!}I$DsMo^pkHiac|NR*xX~>G&h5e7BGjYf2>%uT0M8-lW2^mw7#C!Hq zNfb(@Br1whQEAkS%tM(eNg61Xv5aT0BcYV3GL$F{reC8{()gY4A8@%|-gEX|>v`^* zzjsLPCga{cAO{MyN!qs>DjAZ72U@<-h7;$|yep2*wGD%u<^VYOD-f>RNkWD2Lm19@ z4&A|G*wyNaPG-KiSX+g=GHVz++mkWF^JL{3&I#rg@V=7r%Vb%fA9&yF26A*7e$x!W zRVS56yrDIyt8al1tR)sr&ZhojMY!+tTd39!1#XN_0*r5p0iy>yp(iC1AFI#9@owsH z^v*VtEE>l9*VQ?m69j)Z-@uRZ4mkEd7d-B9iQV8AjjIzfIr#&gOzZZ=O#b*TPH(iPksq7P(V`)cKBOje^M4ZP_`rfYL-vtLzHZ>4uo7eogQ3xVGS~L@ z9dlSZmfATw;CAPez?@EJH+W^?E$1HATx%)wK|>F&AA7<&Sm*K{hH?~3jip!Q9B`ze zfOmP`q}xkW$tmdy@=h!e4O=JDmk-xqcJ_Wk9PxJ;`crB-=d2uF|tIi%9g2)n#6lPB5-R> z6TNgN3D+8Wf$wlDwcdG|x%PQH%nbht`}XCNl8`OnVrK>3yVlTGQa_j|+iIL;G#edS z?C_jX8T9_`K<$13m-F9KcrK-Bm48)}Ug9$`RyuyPCMFMGFI+@2D+1B+eIgmW@tDx; z@$fR&ht7>6I45Bvy`HKHiyn;P7k3L(y;(w3pD3YL%rhp{R28&Kyy=g9UidJF;qFQC zfJ}bR8{qST5%nkr^Ljfj%!Qw6=bWI!m6KuZE+0~?!l7Aa8GUH-juG;DjC~Od@hsm; zbDWczm&%7pafciE`JQq;f|Ine=nB5K(F1>!1}%7=Oqa16QCmk6B(n73wB;|}<6VWxUns4$d&GYIGZws7 z<c()t)9R$`nU4zV< z-ik|vmeUjV^w4mOeeDi}yu5aBk*BjM6@AE zO+2?Y=_E{DG=XFvZlvA*;_%jVDHojAi1`-1baI<4PT)J0S=Z)petd^$sA?Yjy;+pY zS}l&>auY~rK?ZRi*o@gK+Av-+morv60PZn&NS#aooPYb2%4n(JDBmfoov{#ZGNEEgbWgKr|7vFwEm8vY(2lbt%L5Enx%rJP8>1pYX8 zt(y*Xvov6=EBQD>mb!~HQMuESTt;jc-$A-hzB{DBHCbWwz3Fd2_*@6>e{+c5u58A*1_#Wm`NZz? zd`im(G*I$a3hvHcfG!(k4{s417LL#^$s+T(9mb`oh8-x3?p!=gW=XDrbAuPa;Xo^< z4KN^j@CB(2O(TO*+i<3~E$8*P3|dCSATaC>8SRn7k?a}dr;8tynm5tI{F$soYz$e_ zIRwQDPsrOjy?CfJjW~$Shn@P-cv)YFo3`dO^%0uP&CNXvQf5|gMJ)(a<+L$tei`#o zZ8?5_l#6@Q8ySDaKu&+s8jj8HvMf?>h7-3~re0Q?_REHV!k-h^ViH8Q=D(rIO7|__ z2bM6w$G?+Wp=uH$>V->BTj8svqoCLnNfIj#p+H9l{d(f*U&nFu&b`y9=zkksNPgK___@`6T+|-3MyCg)-?e1H^Zo9N3Br@m%;ZR&LKz==9kNpb_vMyTkM$WrU(s zUlZQH-$B3B#ge6|wPfGOZ}#A3L+1B1Pw4INAyW$u!qXQ^Nbj{e=I){_sk&8ryo(@D03UV5JFGM(o0Uw~}D$=XofW z)BtN&Hjp!Fsd&C&D!F+ekq>$|(##V}P{a6wVExB9vPE_m)rKf~V)`Vug`XdlRJNn) zk0o5wwIZCpwV9Q+G()xgIH*KRdUQAzmX~R9i#iak@7iF%=j&J#x}Wbb6){h1DSJy; ziaPiX(_cQk+s6GGqt5#@WoO86$Ez&ByY@CalKz#-m2l+i9uH9Py2CCktYNGDav|h8 zDGxpMn|$|d#uvkd%$prPu)05j-(UQt%g<+3&#j(Sz$yr))m5qkF&}B zvcF_)Xd{_E@{PSnmvg3VK-VAaVLWGg!Qzm$fNxXiJo%|~Wa$uE$TyIBo}agW%5KVw zH%ZMa{@iA-Vdd8Ui`>(m%h^g^fT*rlWUJq2db#!>O8qQ{e4$wIQ2Im@{@p==KXR`B ze$xw%%EZ(67TUfuAnT@k(!}e-Jj&+SBBT zZYwQpk|4{U*@FM^Jv1k35IcG}nzU;^aWWmlE!P_*i<2HOmpZ2jl&_Vts*DNv@0<&| z7YZ;=-5$N#V-Wh!((=w#RItsT9lE#~RXe_;*X{~(a&#`)At8?opHHx2OY6z^i*sS( zo;tko??0mIK87n(Yrytp!C0~Ml3??(ndl)hiJl89tMuwN#fKA{Fs??Fs}#G8s|r$S zi+d-_Xc-_$ZK9%2x)|#k6I`_Z2X5OsL|5Nd^JO`m?A+@+aQYkz_fCAb*Y?gH0EOx%m&v=nCr` zeE1~`Q8w2oizNA_bAzu}qkMoYx>a_Of^X+x z$+HIZ{-}%lGXqIR$}vXvR4sX`pUupyGm{bh!-lF4e+sZ1_)Rm=P~w}vT@qYLL>~DOUqhVUYjC&D_u})qU6{G~784urkT^$uWi*ev!zf(~ zm#aDH^!_MZ%K%4aeB@?wryC*X#93mAL&GRf69r4y9GX#F2m=-!?P zZJ~D%Cr(DoGk(}(?niA`bkN?Bfj=@? zsc&n*e(aOVj7Pn6kzhJKdz0VkEDpuKjEyAO^$`8h9Eqnp5C z6(*t26hPLj5`S66!nEV+us8W09KT7(JB?D9vOp3am)23`yVckgB1^%c2F+!bqsXm= zIJ;yu?R_%|7wz7U+gzo|SgR~{+K>iZ+*nRWNeyH6{Rh^*vILY$f;mIe=+D_TmE+8{ z;msvIcwvWRd(T*|z(SVm(v{+R+`YKy+Cvp{#yw|;HYKph8w2Rh&`=zi$;m48=HKo7}2@_B6r7JO(TOSRvS zpRc{aWLOpb<4%p$y)ltX3fYUx#5R}vimjn7yuq-irPP(AVtjml9b2MzhT zzYvxkYnriv_mG6v5t)>~$FQ5{ zOv7C#yl~BgW$-&B5!6feaEgFr3sYp_U|o1 z`LDYxe^jj|+nwEbcGn{^?eZK#Lq3wQ!e(6e`8aE8@Vru@PlgtZKSakx8-R}4AB1_6 zFkRf5S<8ERboUlh5e1GJa5+K!7U|O*MgBPyU<XD$#C?u5a`3*}C`cZ~&zN#;`=j94 zawGb3fYSZXuKlft3?X$XYeE%7M4aIL}_3%1O#IV%$kOTpa)%L}%qJQVE`!qJDo|v4GP%w(nM+G}cMsRVZvJ(KT&c~2;G0nun+A1ZYMCOY>$I~9 zPaI(|cLA0+PN`f|{p{-n8ucNr}k#|Sp&qf|*6ZEOF-o@u^IpKe^ryO?*A9g=R0 zPX7nmpE(w9emFq;>lCnEEt&nUX-Ib-700F@@syldN_P$3C(x zEFW)`GBv$YZ;~BsZqEgu5;YP)#GvopQrs_g5|`|mPrsY}p#w}C@BJqpkPP`cX0)L(pfr9n-$R8E{@8mbgy6tJFkveavaY+XG znVJj9)9pE`ewSI`U&b5>Fl6RV&u07VB60iMv$)%~l>eTugqjT}$qoArjK$DmT&c_R zGJXqV>sWcK%wv)C!xw(1|Li6j-8zkzx5eU??XNIrwl;UMw1e19>!t_xou;}k3h01& z4YN5$3V)yYK!~RW{Bvi($~KbrzA5|r-K}m7NaqfHDT|b zHI?OMqI}tje%#BQHBg>XQX8z{C1{MAt_@mO0oL=ils>i+{Z9;nR*lHUH zmtSC=D*JxEeHomD($Z4Hbuwk7RrK{&*x$SGg$J9A+N>BsM_j~wEUori- zhh>GkR&ldc2Wg&41m|;H14oTtG4|s#A$BAZYAW`#S39)mWvOJ$)EuSeCpLiwk>pf0 zq&dr#v$)qj4cIiZ0$%nbrA#9FVRGM@8jf}8Q)_~H{p)r0hHgJL(4a?Xtv&m=)XD* znm_l@?t#zjn!+AB+`SO)FS-K_G4asxl|Sd@{-KS=a@f)Qlf1oSK;6?_D|>p5@(!5; z=%iAO%N>&8;@MRgJmwrOycmaB!l#Hy)k@6#?ZimVieui&%z|;UnUIw!2m3T%q0Xg0 z^ss=kM<3tEM2lhOkBuY@Z`erRdAQ)`Yr^>7;67NeDFD5C8fcQ%UOMtUA1AEW!6BPv zcu_e}I>d!ii#hHP}mcFcvKu zBuG>f{;0U1=iw?=__YNZmo~Ax)(?_{J`dS(Q{!pQ0!etLr~@B2%z%BX2FZQfZn_{} z)Jiqv0{p(b1xAER$*!6IsQ&jVI1*w`tAjEa59<+pY_tTj_SrJ04h)h{4W}XXQzX{> zlcaN;-r%b2b7(eI*>d;?KlM=66D&MJ$%e2k%&sGwpz-n<*i?HSvl0a)(6N$b46CqU zULd=i6M~@^`BbrW9#t6{2k*s1@UZy?GI+rXZOgi;ijD?48APE;+bT#@N+UP@z2Qxz z3in6t1G2MIQ1qw}yvdlw*$jDesc#AxzstVZm;Q)swH`+VXPz=!GQ^l?w{L(`Y9#sk zLJn@vSVxNprNy)4xyl_0khi^?{n#;u*1wFP`sY<~=&Bm$ESd)Ww=a`2ivl|9i!FL| z4A8>*UaB`dla$Ntr5QmfB;xvRIPv{7bc+vB$+#L&7wTXvPy4{>{ONcwyNZol!qT`L zWx7ASfHZ!(gpIZ?koRa3GEwt!%iB5FSa=2}nqE)g;ffIQ%W(-b6)z-d6C$YFTYvh+S%QzhU7=;mR^o;sA9VYu#_q1t zho++IjQuy>C72Tjl8tiQM+XHAcp-*EwrjW#M_bX(M~!RTd>3W7wX-BB~reij&SQ1JU(+=_>7c zq~h{%u>Zi4;^H;ro2dznySWQzid@3HXW!_JU%KS8j3Sir+|}yhAy|7;!K$tCE~@`` z3N`!hqJq>x@+nK6=TYBB7raP?Wb&|m2hS0SOh6k2MaVoe5l<=0;IC)L$mUi=^_7D} zWt<$@bi)pPPB>#M&!fnG>QBxXN26iaX%gHzEVz9+ko){P4=n~M%(Hn$r#a=3?s>)7 z7GKF$r0~1>`LR$~C}!n)a2BlnfS*o zyx2hDxEtyDaf9yyhhp)l5=a^;fofO{Yn{KIQ?>m=n=G}trZui0n{ohMZDqKYlP?&j zDYuwi&GQ(a&>~tdI3bX;xkf9jq|t86BxsK>g12p-A^Z6@Fu1({u1B=euDd6w_tgk| z6!(!8yEYELuYW|Pav$P`S02!)H31GaDAIP0(6ly5Zijsfv5)>jUVc8yN;+IXz0gdY zI&u!LcYPC-cE(X=`a7EYjeqyF`k}yl9(U?c3%*q8VZCpR@I5{aC^EG{kwcZJyKW4w z^C+d_O3rBAUj;#1HQ@Fe2Erf5v(ArZV9UUGn89qJnuq7pY1;)TVp>fu1g5YuHJcGv zykq2=__=^r44-LmhPEFOmFL}_(DKGD_~0LspONu!S~(Ik>fYm|b(R=-rWcz}Z6wXWmADDvp%r zqZXF$m7(qep8sI_0M~Ygpj$~j)rgP6>m%dof2$4P^?hM_O;aAF4j#d=Vx#nA^E@0$ z4kjIHDe!*j2j*_rJ33*`Sd_3{4NAq%;CjTF`8LG`PiTtb%ihEIgl84c<2%9%YnK2$ zBFvngPT8p|WVpYMQ?T{A1zqOnjvLo)BPwIJ!HG9XB7n0yZO~PECAN+?fQLbufBgA3#W)&DJ8YTYHkz_%}Lhxb5;IN7(<#IHr@q~f$(il6`#aKGKKNmyj zDQYmq1jqi<$4mt!y5pZO+w|=@-LN?lv)#pDFn$Ac*o5(H5mBqcv~kqLY87l5se)F0 zDY|g0EdJ`wq8&ls1PiB3qZ(5gPS#8YMMyB@b=;+|G`^Eb1rO=F$!%mxH{S_-HwNsM z%W=|2BdM9TL?m)%PUu#55=teaacY0BrAbOA+t_*#FRzuxiu1|1E+`bbOnl&K^gKFmSR8{4 zRG8+7M=<^41K6#whj{!+vb4Uwf@;Qj!dSH_Ai4W4KF>M?x@qH?_LKhT#)+WRNHyC4 zN+@|ohsi9Q1Vdd@(EQ^*;{ULj{8`%!sukKSX%qF@8o-ad#OOU{GHZEH~d ze36rWco4>kF=(`DC0%KG1BVYtLA&W2__$OXr;XJiX441hkBbv&gPtsRTu%pHT(yCW z^;yK^mJ6HL7DcZ&yK?C_V{Wp37j=j2OwDLK~^4Vt1;d5`QF8aeeT zqq88Fp8?xJbe#iP9=#fLTJ5n#@)R-pEI|`C+@M5PkJAblBYg)IKv91J4p_Uu^7L^) z3eUmvwH4(2&G~SABoUV84B&R2E$umBIkI`~^n-RGglQAFAUusLIdGS#FSmr~tOS&j zGosH|yWuG_EpS)jSht9e=rQ30By2m1zELM|`sq%3dvX-}tM3&2Sbmn@d){YA>QPkX zJ;k?&;;DUWJGE_$!mD@6QG02M<<)o(Y!*2~m)z3jW~l3vOw)0&qj@Wo>WSmMFHV(> zWx`xob{Yw}Cx_|$?}7%7Y_^m>>?>EJo;ULuv8`>yXoCba8t=eDf8WYEza~PiOa{)7 zxXB#gdqI6`q`AJZ95&h1fLw{bMa`0zQsFJb@Gm-#>DRtTG^K8V_>^#D#)o4s>xS`$ ziz@@>qv&ZR8E~gp$pIaGH#&zHG5xaGUkyS?%}h2pqwWKH=+s!M$R|ED%c%n?F|5q=RK?r zHYYj91mL~U2{!YbjLmkh$U1Qm_FvCt@{ad|Jbf;OKZ<^kxjT>3@t#+3uCoxA=J1v) zhIE{=QG}sU!7yE44R>6PVWlK@b6b9vkUO2xP@L)xUcvULbH2Ux7llRlY1@+XpO%e=Kk1-#f%tvRH7;>x22XA-zb5F8L z(N;*9_Pk^8VR-{{@zNza&21Am-P8+qR6k;mM-Nj)N1g>Ux{IDa`35(A(WW`Cqp`&B z7fCxl12oHe*actqV6;myO|Hm=hVxm_{|&(S5@q)~$&${oDR^RAG#pzWi#;1(LZ7t< zwOz+^rL4|jveb4QdMAe-%wH0&rVKY~_cDdIL)i2+`jvv%131rx--Ax~huL#}5wC|X z;5bVc4l72{l9OsMR9{W@kEqbUFO0C0OQJ@@a^#{_DXwkVj(Mk!(%7w0V9<1tKCu~O z6@B>*&^`k^b7cm-sv<;Iued^K)o-2;$+Lb`1@wdROSBu)OUL$|BX4)^$35~FS^Kr| zRBLq}>OEKjWj#??zIPf*Z$s1sX6Ur+LduBQRaiz|Q0 zX~oC%w!uBRb$SF=&L@y72!n}yc3|yVz7P0c4*SSZnOs^C#kM%+lEz!#@W6^S7u^MvQ0%g+TwsLilmuJ*ZVjRi56U4vCNZaQ$l={59o0UMe%- zvPFY2YxN{{c9|4i@hQFXj8PTcYAJ5z+EK`Cdoc}PEDWNmzrt`@r4K90z7YIecNLG8 zC8O6z7Zhn+#C@{spvDfR>>iif80Kt=>w9yEokS&04){UG|4Rg~D^Vm<+ziZ34}!yt z<4`i_1WQWm$kxl_F*GBCJQzI49O+5~xew>4|MXa7iY%aw38YcQ!ZezWhbRARVcD2_ zL?czmD*RgnbU3~h>wD?~#kB+~c0wXo1mlK);-3-(`fruSKC?$6qKgNaU?&5ikf3ZY>GgnaD4c27ro z*7PCqQj|ok!DLz;>jJv}<)A|_-vxP^~xJve!ppBlCv_OrAXPI%r8h9`?0)ObopuM6Z ztxZlM4~QzX3VcA%=NmnFV=5k6okwn$d}a1j{Uzq2a$q$kmNlHYnmpX|4JCFAk#S;Q;*Bz!$5P8mSUJ0U-Y?Q4{`jwxs&z|y0J5_@%@^%v*f~6eqPga z271O-Jye9VyOtmXZNan3Q z%(*ACq4Q2L4IjS9gm-I0Q|A@ZAf|+KBPbn>ipKwrWZ<%u*5EHRh!rc8IifoSuhMRO zA0-LSQ!7xr$_qy3cM;KnSpFVo2wx`1@-x8|oN_?{T-sGds~5O1dba(vHC%-I)~Cqj z?R$x}+J$t>0VU2*R)9r%T3F|+!Ch;Kz-foun1`obaWu~bhi#@2v-vh`QNSP(+1^U0 zXvpKhbp^=E(T7=&`Mq_$3-LNPjYR(YN6_9F%agk46Nxx-H!O$hdTXNZvmUze&jQ?# z)-1FnczxoUEU-2imL`(11_gxU@Ef zit>E7{BtE#<{qD$%=01vf5u{b$_%Ew)f8w^KAH7zgbsA+L9)L+q`!Yc)bhuJ^UKY& zX4W?TyqZVM`S+UA^#q9VDrXD3=W}MczF4{LGHvzy$iB(k0((zqK%{*?@VZg-Uz!RF z(_B#AFq_^iI!ncB%i!iWYepeY6a=<;L{DcAP9J?&dDJTyEPtgyAbXvC;B%A2od8U| zbA?2xg|Nx%mAS_ucWL4tYrIo5h1-&Pmo-(5M~^WlaKEfAE^1G~f202RMNm#c%uTuT z1Hq8(egaoJ`Eo5j#;|mu5HVA$!`a^gA>qtZHf7p2I3U(S3+Y^PZ@d8fq{W~yv_s%o z8UY67cZgCh|E_aB1)tu=LBbW@lleK0yj3R<`&S=5_5E?nxE5AVQd&r*-+kjDJm4x(DpCNcl|s0$NR$&f?aQgz zSD;$)b(#>aPuy21LHznl?A#i@=X151MA+SBy4By2&~KlJ*H%L?S^UW|Yn3*q%XisD zwa;POAqTe9jOX-mTR@e_;<`;8R5Xvpf=7}dwlgw4kxW zWUFumN4UiAk-{eygL9$^X3~BV^QRlK0`PFl(#(l{AE2c#-H4dhWUe@pzYQ)J zhtcOGP+{{jw2N@%st3n%lAJF0XH7O~Vlido(NMc=xAj+M9}#+YjQ#k7q5m@NR4%YZ-Rn|!?-HEq|!uZ8(ps(PLuBB41P{Zt3n8>zsiHbUwBI4HVV3JvSih^T=w zeUhMx^><7wCzdo&p`ci-Ji8E$v;**s&kYQdbE1cGSK#vII*i_*3QK(q>DR0YxO}rU z%(6kts?__;_7CeYc+CZ3SQSZZcYox&uXE9SZWZ5QIDogR`f+vn9j^4R4KZ7x4D0#a zXxsEqJa3mk)yN!f@$JKK(dZiSj&32jb6zv9r#X68DI7{y>7v%EZ){(=AMVIU6q=TU z;b|;qa7zR>WfoBf2Mw;+=LVmz=I1Nv_i6DnJy=?_RG?s%fKwLzV96P6GDe{h-0f{a zDB~^vjQWe~-%bK5+|POiCeQ;q9@rZlL>r3svYv9|=*y%#tlWZr;;uT3lfRxvbJ4lv z^NccTXLWrK4qaW6yKF1j4!{7cBYUw**wUBa(C>*rDu$J!lm zC0pv(!A%zja@uqUG^XbAy&Fkv4!;8l2Nwy3Zw$cA{t%pOdK_aWJHZ1(U3}q{O}w{J zP9~VQrAEvoC*+#2aq3ifQpr1mZ##jNN)yrAB7|4x$8dTpuHcIsK{Q2J4VM=EC9;jP z$n1mrx$m{Ia6Fjz`3hg4W3RYkcg8as)Oj9{k66L?ZYf|6JmkAn0&4c#nJ%raCr){0 zwBAphT22_IML*LR=~LTbn&U8b%^D_$o*rX7c3&k1ikHa8Yz5vW<{-F{l}F0dX0jx# zm*<}9LhG?RRAcd0nDcuV+_X+%S2pb6?%8+1^SpJi-6kBKMhk(sZ!I3RibUI`D`_3? z8~Sj%jU3RsPPVOkP3k@Wu}kdD;h{|>BV15Lp8Q$@eT`SC$1(%_sU?jHB1h4De=xb{(*3*qRK6btvf$6)2^H*`YW z2NXRJj#H-zp>@_xTywLG@C5?6Ec%Ha^gYfdMK_V+kquNmU=41}yG*ZtyG+|BOLJ<^ zb-3)C)BuMbh%DX-|P zn>;7%d?wkE9)st5o$zv>IlhiJ;)tah>mU>a(i?&W>+)H~YM}vlWb1x*)7DrjmShNO zQ|Hoz5lPrRT4HJ4asei9SwP1`q+#1y4{k?BIht+Ffma*MA;sMUJFS-BE`H{4NA?r5 z?gXDj-5Jd@Chb_IE7|nFTp!ju@GM(&w3?VLoPeRf>qyAv7ShP~lAnx>XVV3IXK%GB zIQAFcZ5*%k!pq!>4bZ#4zPF z?^k}oJGf_3O=$%(KeGsqWgR3R>+V#%-BOE}6&K+A`Q=1ky_h~fUc=Nr=1)!khVb8| zzqEbdQ=&0yNjB<_gQ=VRF?5Y2L={fQc`4eQGS8@x>OIA}{m1v#pKM0uj7em=`956# z%mv%-PC@$zW4SSj0r+sUF)rG3gTM1`#y9Ui@S>c9=-DuXJ6k-Fd&ibx#7HU6@tgtu zuAj)OH|CslSgWPH*?BBBHYDT4&FGbV1+*x`mxwrakwUdlzEgD+rGALP-79?lVb5%? z;Zzr1{eB;NC;ms%Z`~9emU%-a{kHA9Ro2R|B}0h zUa(&D3cptshMJD0;Fxe620Pv8k+k3F>!M4YPTnJ`d*4xRP97v*SVeAzD03S{oMB$R z6gfC~0{#$k!u^{Wj1LLIw=3e9zKI2pXcz+Jby9fzb0PV;%>!i5g);tItWZXxq;k65 zF z1Lti{<#Pl@F%n!sA~Lpi@1cofh83zhkDMh@Ki` zKMsXA>_nUyArHUA&M-yS1Im4xzM=Vp3{>AA$UNUZj+F`zzq(KVO*N{+;xdCh&!fuEtkp~ z*&14T`ImxPnIV(gXb*pDGVsl)G!;t@qZa-1F|jRxV*`b`YsP_)p>Q8>UONZ%cKO8f zRu&!iIE%#cyT>`{LZmFRovw_VL;buC({6`OhNvDWf8C%&G}@-Yo(JD~7Eu=VMtop1 z^fU42$P_mH>N;{}j~I5|4`clwXY%gM8rDfHlIWMJW81huGSML%S0#>O{N1tK7Eu*W zK{TApJf4I0`#tdBOCnI5JeM-M>mbrioLg75o8Ge@rHZ=@NbjVx?8`${WY_eAIAggk z5m&!U!W-g1;&B1%@W&YO$|ml$(<#)rbA!%P+QGV)?4X9PRH6DsJF#Cn23JJCqu>87 z=CXzLY25xPbkh8MI`DEf9*|Ok%&2l!br;7f@m$fKw^0?<$FHK%+%S4rIGwcaU4aKh z($R5C2+=8`bm5XKjOfV>^ic3dwOKt>qbr|8masg>pMj&%e;8?*JmPvYfZY1934JnE zxhbn8aJ*qLtW3TQ(?cZTVsk3ZQMo|u*Jwj$#WbG9v4z@LU%~Cue7SKcL15h7Q`zy& z2EKI3!syv@dXE%?u+41{nm7&)+&m7q_4sqkOR4coN&K_p93(W`(8z^S5Ev+i)nAU`u2*mIS?wg){3?X{>nJm)j)e-sKD{Lg z(i!Br^mrV;#L`3htLgJN7m~0{lJlvJ!#V6j!Md11>i$!nvZ|{&eb=jGOU6ZZwc`Se zlvq*uBqjzf_bq~4J!cd?9E8RThscNNr{EziVFLToa9U0uT$}a={pQAS2HLGSJ8cUD zwTI&9qHOwn;uo6Z7lbZvGN`zkD-+SPn7lLN@b4_1d10aek*W7^xcU}7V)7TK4ENyo zFcUOlmt*&+2L12L7uxqwh+DwFJ0N181=FF+1hI;o$@)U7SGS7uU%iA4)}D;(4OJlN zY&L!k;`rz12U036#6_GDLx0nwaH4H7|D3;-SQwZK_tEMn89u= z_q&2h`eyjWJ_O$3Bl>XS89dGDSmivaBo_(-K>pu({O~js5BGQD%#2Koc@PDS-6^2{ zDUr_|+~6GLpON_9=b%-#4f4I!Nb4?h?y1B9c;3K!iF>o4XI}zY3SX?IzYpJ`E7xzMLh7;XwtY{~*dmiQU*64}Qqc!%V{58S z9f@&i8+o!(m^;xtl^aOEj*qotp{h8LO3z322Gul} zh2#E5*?Wg`_5bnzlw@Y_y`#ve6zBQSkP;y+QFfuBrL95Pdyk|PDKkRFc|K0&JCdkK zduZ>ZCDreI{{CIx?{$4$*XvJSFRtU9=kxKr-)}enO6ceh#;a0#-18s%$>|9iT=2yn zX4=7Dlw5m^IeL#tjXGi1CAYzZZSiO)<<99lD&W*go|B%c3Ohq9@X*L*O4rxohjkg; z#dt#^o9PJt?>CWO@*Ba`ToH@Et>OQFl~@z6EYR$QmW)~ z!q(uoMLL|u;TPDN@e0@O*5z)Bx?yEbI{rN+gWD~QIKOk^ct*klI-fGw#OH%*j@rQG zV;7;sXBst;x?}rYWi4J;Pse*Xwh;X3GEqJBkE|1Y#@Zw=l4ht*Dvp!$; zUBU}e>-dq{M~<+YQbx%d^;ohd%ngiXy4br1ML4d&zGGKjG_w8G1=#d29Ca;IdX*`EXz!)v(eMPlNLzHgm#L%rL>AaZn^w|*;dROp* zZJ*N&*ixtm2`7Hy$R%kM3EE0GbM4qWgjAgWJ2?8A(WIvqh=aix=xzynlrxFch7Pj- zcsOLPX-3+AmaR#8fU%8>@uO-ad3j<#yV!9q+?-*6VJojv8MTwhc6OkZz(q_EnL;*8 zRFGLKgW=b_L~`L+2VS^TLHx!w5YGJ_=pUJZ`^tSt=exx;VXHNFOWlxn#+1J8eH-P{qAw(oFcHKpeCyfx`1iaQ(^xB9W>@9!1;Yk6#aQ!}%}t^aE+E zc4)%)9kWQ`(@O7GV!rI+&y6m9m63^ zm*@x#%t>N(%zv=~lFRYVofLGV88`t^z%%9N0uPgbT2J z;{d+T(`m9ey@d&w^auq97vt!{?HIAe9}ibmkz>1!G5o_qTz*Xq;~!q5cbuJguHka> z*3F6=*piJK{CUohODdy&vmJe1j^eG6FUS^1Li}zHhtzonkBk8ISyn_V<>jDaMX11N-@W`z_RjC1ZJ~{yr>O&viLv}tRF~*lVc~L-!K7x=;`6a{WdTV z@s7sk9>E2x#^VnIZTNb<6Bl>7b4$0}M}ZyP=o)(hBbS-u5WhtjZ?4ZKhwyvFn>C~$ zBLrMsWSL2OF3{pTYccgn9mx(I2PHc*;V9RL+BNNDwvsLcY%+i{eqW9mzsUX?X^e~) zAbF?C+2FV1X_Wtu#@MI?vQ^_82^38t%acFQ(_6ck9jk(gX>}r+1>2#CXVhg ze}Xyrfka{hVU@%m61AlU^zouSxM)NJRu>*8n+N@1Gp=Ir_iLAI1XT{nP6wacXv!=6q&0DD3^V7IHv6t?h&ygQ97o%X4Gdga}CTeH- z>+kAp`sVKz6jP|Nl}{4kZfF^R@T%OVx*9R+HMIhZhTb(UyTSW1mdbM9&Mt%}E#e^Z zCJ5^87jiA8d-$1N0=&FEg}qY7&k{nCK;YT|X!4a{eCb)bXyIY_;%QIr`vySoQVZI5 zLLJM$f1t+v9>F#15#1zF!zP(WgIv}LqA>P|Htfy?7K2Ge-b?a$EEXF_98e{it=o~~GLo-Bc*-q-irY%TO&KY=;D!e$_N5}Se7}BCkj_zAR4|PAs-Aa})W!)KAeC!xhYUbcXwx92?Tn1l{B-+-qgPQD-pmPc- z9-KBF?GMkxoxwkC#(Lk9^qX?@XYfTVE0KVydu8pco)^OLxpGX2bs&CeuScKt%djAP zGOlzwmd23a<@~`xItl&9KVM?wp+h`{6}FFH-5pXez~TtEA^SasLlM0x7Z3a1 zMPjbrefaWeBPt)LZfxw%vt9YD0pC4bgj4t~aK%j-=t0sq*!SQ6~V6~#TFtqYAYkMEy36hvWhz6TyRGXbr!C6qr?L{9u9H!Cxm zYAtgidM#U;T+@p&Z={o5S#Jb(Up?u#GADkfppJ_tM8Ns6KGwqLH#x=kyQ)VcLH5Xd z;{I_yxBRdaEX{jCt?V{)S=UzMD#OcotJa?L(9FWF+vDv#GM~|LR|8SNR)s5`p#Wk$ z|K{(T6?EiDJU`r zcYyv3oPt_>mt=mUCdjY4LZ8@d!KlT5snd_EEaN-G?9{0y|Fxvy>_@fO_PC$6g*#!& z1|#TgJV??%rGo!@3+8M6BVuVC4Ji$dO%TzI{fD!umv}6-TX(K;{km+Tc$W9nzI;i@ zJP+VX#@Y4z*0B41=R!-T4`j5+!_eOiur7rmpTDN@EQuT%y`A@N=R06T-y#g)+0bg9 zfn>D&0V(QM1F?nWO%?Wu7##Boox^t$AHJvj&+#%^IWD9!Yn*UF>>&AcR1FmQY`&Mwb&s@MMCmygw zZhtQ!!clKYRLXP+*|31lQnG^=wq7{JbTcto`-8YSO%$2^dZvAf8t|od11> z#KPQF{2L4W@AZHC3|9ZA&tNsv%FL8dv7A1iW29t z7HQ$ox9P-TWhB@MUchgCk(}B=U(RTnJ>TV3LWW;!?mu&l9z2jj_oT$4wVwvJt=b7U zWC@{hfiib!*c%HLMc{3@S4{ygBFVKq>j|54n(qB^6vvONhJ$l6L1;(|r!KVRmL>0p z0cfE*I}hTfXC^S>5YRMMvI$eUNY4F>BR=HRpj0Fp-1_w4O4Vti)GbTjYR)1jd}Z)M z;#bzaE)cX2@>!6a3TB5vG*(`dqd|7FumEKn}g5T?$Tmhej=UOCR<8_ z_wp>$g9or_ZVJpjmT=H78o;mU%0UW-$!A~Ze9Z4A?%}4J8U1e#vtE&LxU-*!D!fRpQ zU=h3=_nnQ%w`3fiG^5;v6L|02KeoMaC72rhq$B&i@Y0f_;IU#UEcl*9wgD z+P;yRI7i^Zhr@LC2}cq?XCu6JNn@`rOo5nSQMBC@2G@Gdv%b1*%=0h@d?&dV?$8Dt zT)qQ_Zpq>J0ULam=!j9{`J4f(+-NsWm}Dy-C&kI1>AB^?SoS)YYKL`^S7AS?RjWJ5 z@=+Jnl!;Kj@IJbY?BsGHMR9$rGd;L83-($)pmTEG!!du3X!ZKRJ)ve?w6BzDt4c?g zu1B;_E}R?HJi=7e4$~~BUqqnj1M&ZxNxSPLsgOt-abg2uTir~qZ*&Lh=PX0P{79VO z=>w){$lVIu!Ytym;@0;}k<)k%uHAhgyEOxr@1F#Jh2Hbsl^LLVg`t|uQs}MuJO_NO z47ka9lOVDCARRqOUX)4`-;mANt`rJ;{-|O}u?p-PH=kRw+!1HEm@(T7)uG_$5t{xk zj_-fe60dU{>})!zz-}$v$-bpqj_7iwKmXFrxA-&rKkd9j`V^{s z&u05{3pF2Zp!u}|xZ<}iUXObLZ(PhsZ|`Ir6gvu+?#koA#;5Gyhyh7g zs|2aeD0n1s7k;a&z&Xveur>BRez)X*FIi3I`uD9|@9<5C($Iv(qk?!iR|zEgd~rpw zF8+%a=k|FIkSjY|$qVOUTKsVkMHWbKJ1@-P88?qmvfdSN{wj*EO(6F~1TnjnNHt#1 zg`jM{gH}?7(GnCH-(ob=6vJnt?sP?DJ$Ytwm6));BqCt0oxm5fSv(ltGJ{>pR$%||5A4xiiEU;l$nA%h zSTBepcN~o&+2}TaDth4?>#%_n=r?q#|Nrd{DdUO4T8C;1lWCwh9tAE zq(@)@J2l!2^;{Xqzj_+d9=em!Tg~M5U2i&5w~Q988Df$gm1*;9o;{U%2>;7y!l|Mj zxJ+&yd=S%w`2tm>oRdbqBOy4Q&&N=YYOD_a&U~9Vjk(Itg=$CS(Q8I7gt!FYNJK6+ zo>l-$g$#J+JjQ+n~B}E_X zlrX^@Up_xei|AChK5Bh41Z74Z(^&O$B*!AM=|st6f^mz8&lutr4^hVJ@(nuY2tTJZ zRiQUKCejz)ALs#gm`w0n-lP=thG(s$QmqDMk~JDgR(QJRs+iKv`DP5ZFA&IRj>o9djE?V5s27J;DcgjVm?}|z?Y@i2rXZI7^f-T_hW(@jY zlu+>p@9wJK2Ax)XH(SD*_6BW${6lYH(zfaB$M(t4bFLmHzxKgnd-yZmUGK?5Lvga1 zpRWws7TV_Cqa=*?$ffNnqczQH^nATNEQ{owO%hL-zF}c@tV0))gA|Fln;^Z_V@fjI zlHrW#R*L$R0ht`RS7r9rn_yV)SHmK*P_YGHO91LW*~RQ>L`p!_t0$#X?$-`i%Wm)%Jo$J zpy!<0hg$bXT`U-FmJ{!gmBeXlK4~* zH=dH?WCWzhM3K#KI8WWqd37RPlEAZMXNgjgN7)!&^9)T3LpdFZrI47t3F0Ex(8ydd zVvwVZ!I!aBl1VgADY)Y;qS^mCY61`I2dKaf%;tNUED$K zT1MzC8(aG7@F1zyI?T|RrMN;No;{RyiB4I>H5#}!H)h8FB{LuBz{_(6@awApUh$O2 zT6PhC9@&q%hrgkwt_uD5Asz4M#L*L447}JI1HZBrK>eH-iatk>GX6z1+SFk{CmO=l ziohUb8ss#+00E0}KCj~u-epQmU4YXobYs&Vwa|qR)5yAIys!J0E++W@ zrdNHm$ZL-e%TQLW&^?H%*eRs%+T043<-X;NX zg3VlOh#^itQTNV6lspj)Wg^p1)v1rv_>U5si@R;zv!r0k>+>kzq6;n(1C+g+0cI`A z>?2KW*c-bZG~W2JdL?ae-!&Ijm`1T4kg1g zIHkqwP|K!(dDo#0TcqzaJuPS;yOLc&^D#B9ns(efgN}D!vBqpKJ^Htm%)776PI_~I{PtCZa`}D? z6cvTmR*v?JedJOqzM#GRRXQZgqIAe|=)JGNHKZC6&4@ch(=Z;cnK|OFNuFTndV?lC z$>baxhv?j7e;7Mg2~K1-?@!kR&7Z<#M}rIT>+Pa(c7Jg3kPv)OGo&*D)zGJZ3hy*8 z;3ii_aNGhRVzj{m+n*d|K3dE}-ILl7{FYxI1?JFo3CUEhWEQox76Fa=Es$1)c-!X^ zn|pIVx3nf5=Oq==d1ntm&cVl4hduC6VmdE=evc50l;fM~H5l19{r#1Cc9Z z$s~T>Dk7Q&tJP05ivG31XC?aFxcwpEf8!@ri%^9QNloe(@yIsAXBD;y*5m82I!yVZ z#pw>I&}03&mHudLcrm8Tyz^du;YdQF?rjILD7{b1C1I$SsByz9XX*HkixS`++ z_gc@>_4%{8=ol~h!|Ny!p4v}Cf)bma-7&|&}ndcpYfli`t~2kg=O9|KF0)c@B-#CK+3o3$Dg?sMaY)1_xe zWd5N+e@>Ev&t1qFH)--(j18r=7k-|SJtyJVAWGjZ9e40jTm82L-t`2Ig5h)!uH zTa*I9dZ#VjroNrYY3DsJ1LLTW_bxK>Lf)?P`CYo2)8yVKg}`4sYfe*#&vID*CEcMi z+?QEFO)u|gU`&TJuAQTd@0R;;Ma2$uMR*yz|Ct$gdAC1IQ_{xRKcASUOqR;Mn2F!5 z%-K)RjuAAO?ULqfbZxj@NBIn*TqKli@4(Zs zqjb_p9qwu3&jRJQHcHldqlaPubPajYopbKN%8tYI*7bA>sYh{XRwaGeAj00{bLoGj zs;Qf$Clyv7#n(F`xRtR|++y>?FsIL&TXFL)ZO`5UOw9sX@c9>eOEnWawC^_+?mI_L z9PEXCW@$9dh(*tKRVb6XO}EJPuuS46w4Z;F^XYfx*KtVpFJFZp_CCY~do3Yr#<-?w zZ(E4_sa!}ndWXFKBS*|{R#TDRY1qF!hn;0HN(P=4Fy+-js5dE|OCAnoB(|Co%Zr^@ zdaM#^Urd1;8ehPzK$9M*;F%w5&(hH(0kroiWS@8*LE~N5$?uCpG+*K%Dl~q^`|XeE zr;3Y>QsG`;jQ()ZteUQFdrLb98kqMpByshY@#K)a61nd6ljb~}hAZEBaBJHyq2q#F zx=2upIez^C?c`Kp;}4~iiQNv=Ij0w` z_wgOcsVU^j9RaTGa{!GoY$n)qmf8ufp!UBKasB*Qu4}6U_*~rryUt{irRQ5ove+Sv zOp4%6PB?hXS(OUphnYuIroxDaQa*v8F1&FhchF|)@%c; z*l>+B+v~vg#lP8$Yp$T`HD`?7C5>&PxwK_fIg=}NgIX@;J=}c_7;>)%zg3<=%l~CN z$LCP1oIr>+_&~M#o{_%4IS@JJE-v_fl&+rp5sqYkp_*6hP-IUGggt*sF8hr&nqTTA zO}%mC@$~U{eu4rnXkA1ld%RHAT9SLSR~lX&FSU&`Sx7?MUJ{4ECyapgTOwvtk2&`| zxeal;+)kSlWOemUYPg2?t5iwStII3#@US4nSandn*Qs>c%wq6y)`iRd3fypa3Qn^Y zBH{~7F;wh0j_vMZvX?$2KIe*=w>}G*p-o{_f}hRFWcj0h>U>JAH_=;f4{)a6@8MSY zsrcEe4a0Pe;gelD&{x|a^Y|rVXAugg+ns4?qN~(^oYvCT67C_gJ#du%r(c~7&5Sd8Q1d#FL(VUZnT52p2z8b zEtzDkgfsc4)hmY&ftVHDSBrMVXyO<*~Xf+Y@LT7%?ncH^it|* zjdLh>@@qN9%>GSxSh>;A%*8M%`wfX$Ekh=JgAV2Qah30w!6k8Y z{=+JKv+OS;R%;J^;fo-O&tFb>aGY*>6a$H@7{wE}Kx)ZijK1N4&&ryKRm0S!mg0oT3y9S4d3tR0H~VL`GE6E6gly?Gc%x7l?CzE^ zQ*|q$Ayyp1FRUjTcArr=rISwiPlQtx&!uPh`?TrGMR@CDHcM{Ep^J11Isd%|7H;Lw zhO<&h>AXzv)_X_7x`UwIEQHlGxKAhW9mIN>4qVG~@8m`zaYWSo=aZKXQYZHLGx1;ycDP=?*ata(`U^w+FKYR=ZQdS zx+k-MKQpxyONIGI1R!4S0W;&A7aVM4S^Kl@xP4nRNacm0VnsRT@~pD7sZ+Q=e{1<% za}Bk!Qi5w)%R!>|2N5ebgGaWC)T(+G9sFU=sK?i{+D{DFBW~-+-?}6C$J=p&@hh&`G&TNza!sL`ZCpEs0 z8TpV8xOiU(Rt%iuzI?n+FBY%m`qzEqJ*?W?kxoO*Tp^7f`^@mmA4~FB={T8YCX9Lf zbtV?rfHO7LLS?5vS>*eRE{u!BNaN+4(bH7U@0B<=r{pG8)Oy8=c>f23eV4&y+ZR0i z)ChR!JkxY^68K*qAV2PQP*cnMWL9ArvAa7;xu?qT;bj)gHBo`moYTan`zyZxD?=x@ zHejem4=NK6?zK`hw=F*pjfLMKSv`W+loBxPS_Nr1_>8IiPm0TZ_YLJLFVPtFFvzt3 zNYALfVIX-M9J!lA<{XmZ#6lABj%XW}or&UNZ|0!QDm~oGA6D!S6G8PeC+WU>qC78e zHDt|vYnw#>qv;nYsT--HtKw6MTVgOK^ftlK9$8NJ#(Or_csw+YO97=cRpMS6g=Ioh zQQdMTF1*e%7aSz9{)-6?4OOw4dl)YBg+BNN%i`WClZlsE3U^?CBks`Q-L*D<*qxb2 ziPHuiG7%U+&%z>5D(y#`aT7V=ZJ%+Qv=Z0;J%H0Z{FV9wt7h6_?4lSt`k)rS_R3&d*HYLOewTjUz6*Q$KiF)1`3I%@^RPIf7(41* zux)}I=OLR1LCaG?GT)g#3}1;?x@zEt;SyAz=g%4+m4UcfnV<(HxL8<^GZxW=ov~|4 zN99?fy7DCz6%-?>P5bes#c~wV8&6lb3X;xFDJC&58^gYAVLrMa#`|6a)FQ_fWIFP3 zOLPIQuIa!l1tEA=Oow}sP>oM95yF3-1k36a=EtKY@L)?6dtzud#~6MjB92+G?e4-EuPMEk;K1?&XAR-sz~U`1VZB*aGhEw?hYM?I)PT|zJL5(b3Apc6T~@3G{A3VD_Q9O3iqy>hQHNbqDk*665}mN-eVOV$IpfW)h5xI zetE3Le^Kb;;sT}_MRaZQDso?2j$7q=6t&C&7f!JzkDn}uXC)`7lwtxTY043__5~5U ze{}GsCvJab3x-DZtPW*}Sj}E8ewhSaR>eDdT6fSFpbxSQOJLI!J|`bn)kMxU zk#{Hf>-b+SD%Q+KqvdjB(%B{unmCbOZx-f4hj!zHP(I@|znP9q;+Zlh6l@*8kEchr zP2=ZFPs!orWn9k0hj5iGI!SBLN&hYvv=w%dimr4aw!EgOp~**_m2n1 zPL_IoZzIC5eKDSW&pQTrUX>wm2UqE0=C1{qq^b+Gb*F(#J3zki8DC#bZD#xrAvzFn zn*K0%2ANCz-s+JN%4aM>Zz(TmaTK9xHHEaWT7~WsQtF#V|#rDyM1*cG_ybR_m$AgZtsNFWFb^P332UyL87;G3wJ`M89k?Q$!Ss|K0mb->X#4Ue z{HIlpUp|g8VUaw4()}6ReBB>qPo&ervd8Sf!HZMDCGY zDqi2Pm`Z1Eq9$?&QzRYI$dvtg=0++Qpwm*r|fwrk~VS*$@(+ z?nM2(lh|3Zis|CJ1De}5(~b`p>9f-*ByaFA69=&k0kHz#sz*~dEb#7>8`hDoA@3`*cU-8y(G=Go6bNH$zNnD?-;JXXF=?DcGEN8 z9UdQrO`?yK5}DtY!SFI#k*q$k5=K9Z1E;cqcu7rUgWZ0T=-w}MO}#W1 zD9-zkGz!SJPK2<8A3Jx%yCPUgz)4=jO-}`Zhgu`B7gNbS!g)M{`6C~?Fhd@ zJ!%e+pxujz>YxGSRLOG2${DzDaxUKPR_5|*Yk5!HKJ2yN*ob*utoK%JP%;SOaC|#) zl#eCTYNb)DC7KGO1=3&UCYEE%*h*nbxwqae@1pz zA5zgMN(tDM7!(LHM%yl55MT8apBCn$Tz@uwA^8FtGN&+;@5MvNm^o+nwF{d9X5wrn z&Te9Q0aequM{@Q9?l6=B8@3b%rG-H0BA-1?RwHDcEa(12oO$EOd*DwA(tF%;T)wX# z1txpaN%v;K6=5N`?14Dfb|sig2;sfAl*}$phlsuj5bNJK}!(3h9ac#piuzso2J874gRbrOb%|}N`@2Payg;lgr-C1j*-7c7inNsCcCPhUz2D} zG<3w*E?$%KO5X|s{|8#{=5Ig8fJv%B}O;9(Nc{DvM;v@Qm;*is-gqX z<($Oz2p>gPtz48j+yWh54U_~YLd8=Fu59fETE5#1jT9Zw$+Mg8sYyZ2?iHN*8hI{! zbSfP2F2vAR6(4aO=(d{Og2=7f6-+~ zzCC9WbSi1$(o_^THiggDnSg<=+|IWfFsbJeUA{V-_dzRRMyo7-5Pd*&;&kZc{)I5A zHW^>c;Q3~H-%+|)4^Il_MD(9W^G?j>~lPjIV%s&GfEh9sd3x`z6%@Z zJDwXU*^Y|SZjtU+!|dFRNw_@A8Sej$!BNo)G#ZV@p{oP5x`%n=qE6IlZZBi; zW*SEBy+P9F_M#VmeJK=|;*$Utj*G@X2Jd!6+abLA#{)Ot-okzO#Bfmo>1?#wM7y_6 z>o|k2>tP^O0o#`Ge&y$97?Z3-8Xi@N^6rJG{$mX~=O}{s_@8uo*A-fvSk9D2jf1I4 zU&%X{K@zB9fWogrI1zC@>cHnVc39Wp^1GC}=D#D+f^KwQQ4i}s>ts_~@J%YW*B;u} zC{fK47i{)1fYz=bbeqv78qzz)_7Z;d6A#b;ciGc(#4!9xEgon%MLM7LkOcCFyj#2i zy-l+5ja4DaEN><)dD~%^#bfrZw<-#Hw3G2h6JU|l2dZ*7o>W}q=mUOc)Y|%xWN!!| zy>ZLPxEsO5dsQW!GF1XwgBa)y=*N-ylfcyF3p?ta4AU&jn|`@ZfT6dC@!Y%X=vn@R zHV)cx<2vIS7a1*t*ZWu5PEX+b;&0axs(**PJ()pzQdYv08CH08^aQez&Uv%SJWzxPxV!BS)dW4!*?9NdRda)$rDM!!9e`e9|tkZM4|8JBuw7> z29J+4;n43H=*|1r@6Yq2dBSpBoT4$uZMekL@SO9T_b-v4w-M-YgtFHoDP3?g3)Z>@ zk(Wi9XeTZV`j#f}TQP&G><}SpQNv{Jg0-NXf14D}O-Cs;KP;MWOst#biNRWLY^wjm zRGsdlV#%6dqAtQ&uHVQtyC27+wf|dxYT}aN5}F#CjAstllfuqnNO^U|UeVfIb@ARASzrT%_-^S?#F`+?JMP zh{z~|LkoXU;{$tfE{%rrE3&*7p&3SmLWo?Ds@<&Jon+KuI~E^JB0feDutDT2?~J#D ztxwZYXGRBk8ZAM3HoYhD6Zrn%gEVq!^)TvBl)`7bqR}DLkEW~JlJ4wgX02T(&3wS~ zC`SLW4QlbIxrg@}0H3W^c#MArXJEzQPmGeqMDRP%LiOXUs7{_e&(kr%bL(@^L1GOq z7@GhhpMAh>wG7{_I0@f1_VamwP2f2tnYAV5FeoO@ZTjckRAlu8pYP4bTs8y$oSw{a z8Mf$>rh{1ySGgtgcVeb;2oag+4~f=*pCxs%rm&j!eyd=fi;Cg>o6l(Gp=NlH7ld;y zAJT7rWAMh>2}Q2xbJxzl#i!#|;AVXVSbTIoMr&>+Nn8UX`^OqeZ~*wcAtXd?Fz=B_M8t)97~#ubJ-*Kg&Oz| z*mhYsQ~euGSSE%f$T1h6&pwKO>RHs0AL3_42IOx27l?GYj>pbyrByj1VDv!~jJss$ zx(OG_N%I!6Bn05*t~}!Z;4_}?D5hqwS97jr-lTDo7|I2X>fr=lvD z_(sA6wPD(IQ;}@dn?z@4#?UeC4A{Cp2!%?{;?*1GIBm=toL99__r;&-rX4M$=wU7# zs|=*`j1jUmu26?4JG@KZFvg#+*&3e~rkw2_PT+ttmvha6=d+~1k{@%y)`RCM42~mO z#pNKeB$&D8JekD0R5pDUZ@}>}!?^aX5tbzw5nBO{?hc>Bm<5c`f#x+>Wut)tLcVCT z)R%_ozeRE3GAw9Fz%T2zp+}?@m$Np7I@P=*LO-fd-!vPqGbSKBGz{dJ5WS$mbAUYW zvHv`^(f9Ln_!JTaqH68Tk!2BVc7P?E`p=ORHSzn{k9{8+WK4t{I+GxklK;v8U-%{7SNcS(+XD`MJywBc!& zWS#+B0Hq71`0I5Rx5_G;s8-exMnVjyzdVgn)jjn5YQAe=wgD$5N?~+IB4#8B;;5l9 zTC7eYc7~lK@OTg1J>CpU+`8ElxoURRLmRC3<^@vVXPi9??~XAdxp%q?`}9EH=cVJ zno4F}5GQUW_hEhO50d}nDrr8r1n(SQi#>tOaPWK`ya!)m|K<|;W~&YA!qx2R5;fNC zeLfZFe@~kgg=jRL!@KJZPs(_>-?RZfhYs z$mZgOy}_7oehYkX8|15A!8*gA=*lX>R(+mX(9=e4?GvN7F9l=3;se+cTFjbjhT~-4 ziICBH5!-uz)7o7dL8R^o$v70+WV|k&=;unH_ca|*Q)`8@QhX0QZ9mpHDiY<>HDt!r zHQ;tG2LEeK1$_f?^j!HKW9U+DwB!)voMXvT_e$J0HW@c+M!tRbMD+I+?J|zVd(XfYsH>u3mI4vA^=rMWZc}44gZ9;p@aBhw2ius z_mXw##U14&-pdP%BLnnc7QWnXb9(2TfZwH9_`+%hLStPlW^3?9|5REX)!N9v>+^@dV zs670OutSpYZo^MfUTH$EJi1039nK+9xJ_p&7t`-NLTZ}}&xX?9hmFSn(Up~v?3J7v zMo~ZqlFI-kp9b3|o#D@P_t;a1kW>hpIZ6VL1d<7*Q$YW=D#&`xgkh6ZcV6CQ7upHp3D>3Ejb8=KLE|?xlHb3_d_RJgZN;=x(t@lzF4#0(Xd(S$y$07E z?h3Q|?{P!SJ8_oEy%&JvbKb+i5qp&AR)!U2 zD!iZjK%-pnWZ3gb%kI5^5iDp5;OBP6Xg!(>WltDvuDb@3TO!Ei=F|Ab^)|Lv6cgIJ z9)3r0Xmcipj(2WBx8+-Sbij2s)@vdj%in0z@0rLLmya+{_8tcJd~y2kPz_vaMzTzQ z3Tx_qvFZ6-J*bltg>J{!*5U1wIqme;V@0;`fyH@OK7vQLqhS8~bMT`aTmwj`a_Y=V2v=hChAFQ|CbN+?-<7DDc5 zz>i*eo}m}bB{oTM(>q%5)U8=icsHNCv3SLrFAm|(8~?+Xk~}MMhB6HHPl7L!n&|%b zBzOI_KSoK_LS0ukhFpA0g<67e&YO2Qm`!P%*dSRdC&7h`CDFT+7h!Q>6d3VY*gJ%h zGWk>NVx#h=+5Vako;N}k%Da#cL!vlrOi3~KtWjXsV3U1w8E$X9Mt9D;L(Q|S;CSCC z)BJoYUaeXS`N}V7-u7qMG`|{CgBU!vWC-&oO5$LJEWQ{Xpyvzq@bcdYxO@2#viT41 z+K`Zj#tap(Ef2*L3)JweYyvkf>o^Xaeu6_22nUm#z@z^Vd~z>A$G@-0i$e+UYW)?~ z<+v_)=421M$8VTjbSMQA?LPo3Zw$+1moj6kc~6RM5vSJfh}TM_Y4+JESTSxg*DVr; zU}_Aw*c+sV_3`?#6b)yUWL6QOSqX!4xzz=5Li2D5I>d6z>o6m%fYR=Or)0tChnwk71N4t zKMld4F@niFBMpMt5?C45%y_gKayf~1*j1HQO2aq*_So z<@Dg2RU1h0tot+j4D1e=R)e%lD05(%6_@pdl8;liGl!cs;a7M9?d(2+zqG{3k}0nI zb2uMoHm2d5{Ox?-%9DmV-^OYA8_2)aUi9O-OK30H!I~Zu;^wBzfY&iXv{fRJE;XvL ztt!0<4WCCC?JILY{`Vo~W_cNL?|ezVJWD5H?N5Pp{9wMjO`t=`OCbEA4pcdHlUb?P z>D?+HEOb0hXTPW-Q%8}sle@idzu%|3Vy^48{H#k^GZosh!AvNP6plI^{m?82z;n9 zO7*uy(;@y^(JGhXo==}or|y}DJ=Z+AjUE*kxqm7eIyi7`!Mm`?+mf5Dn1;A@KL(FW zgLBCmU@xQyKI63L)qNh9Lf0P$!ShGV-E_+AuVmVGA@|BEHR?<=ERY}tC7^GNsg|rT}ICK++bZi zEBW=pMM(KKjYy@RBIhPO=Tt&k=r@NbZhCIl>@_K7Xx%Kr{f()HAI=NNgToVGQfw-r zFdsx0y(DFieYwz@5EM5wA!}a-!tw}do_iGz_jnhfkKi-%cjzkwTX%2`?zLR+iC5?r zpNxCXEya$^DEg}<3-klu(7cHkU@}$Wo~B-9^@XQ!?$wKkL}5Ey7dRaoUsg0tNS}+j zi-vIDdwH~Nen1NPXT#|?S=hJkHQhB(23u~d;#OQQz|SJW{~tr=9nV!8#$g$iy?11k zk&Ma+=ebd@q)jMF(okt=P?3hRM3I%N)_;$?&=2GWHTydZX zugM2-UvAAIM!~93{O}k%yFJ{tp6_<(cZ*_A-#D@=eTXTnPT@vF({byvVEi<|5yOqH zxa9Q}YU+1`$gRr5Q7I81VsG$m_eJhy>t^oC=40fJaVq^GWCa(BI_UBOZCX6ogsy3R zP%9<|vSst(Wn&k(+=xJut;YT972+J$NaA8ME3S1}2Nw4{rEB?l)8bp%v`VCbP1v)Q zPZ{4?*1R9F-eLK+CupcK=0& zvULP4ht_V4&s81FNQ zJUrdTj0Ey-psXx%h_=&JJ`Yg-dmNUTPRDg-7qF;T9E5mId}rezKDqsrJWD9R-#fN& z_oQcW;l>|u>-dTIp?4k3iuGnr9Qr}_z25^b6J?-pRTxx$4WldXXR>SWWzf5!O7P;{ zE21>lhU}UhgQ-;lX!^m5TOO0o=MNf5(r-`P_v-{af9H&Aw4br-2a3R>asgPJ{DdWu zVT=mz#R^vy!I9m~mqL_GG<++tcdbiVf*QPh6b$LB0!a_y;EZ@Ph!``Se^eg&~pS88n zc$_$aEG%=9=0;)N~5tRx0XCVGKSmJ;_I4LI|#LR?4qa=n%^mshKf zlG|qB#5N~dGVK+4BJ9pBbz6xAvNB}%nG`543WU-0QW`cTo|@;4$4O2%8S_#Bu4AVQ zMEx<~nQDUECqo_1b&mmxy>tcwR*irGMI?Us^95TWLkBAI3w^<9cE!@`{n!(FGq~ zUxW08S+tYq!4I@LS8c5S!R(65LiNFSxNV##`|Q0T+EN1uR|A@n3N>4u5t|=>QZG7 z?HnT)tX6|W)f`U$!$-oJc`-KqUm0zI3$`|Hv1qkP&@SJ71wMZ9fH+$v)2TXE#K8^{au9;&|R zDAVhk%0}%tOJ`nK4X4`HLcaMv_;rtGJk@1l7thuxwbv<|E9g)ROBLo{3j&>)6(B zR^Xpl!|xCO@x8)V?APct>L;PWgVNv}dASn_DIAj1%F$23OSYGM4u+%Ub<29hQsjsl?yv_j~^a2=puI* zZ>H+<9eC6m2*E4NVbbsWjJWx7(xCQ^-JOw4gd8Hd5`%KSsFK1}RPMpGK}{r}(I0HX zzr%%iCD>}Eg-29s$$s7)zsI2phh|?RZT_c0GgT9=v>U?Zk;$MKHwcZY9L!VQMU+pd zaDMCGf$Q`LGVfwE=G}3?<8z`hHgGce-O@r^y7#fqUS*R_{l}p682_FSDuPngcPJQ% zz$0@>PGO{*wLIg8f@1^}9VKYN*dPjvM{<`%f(UCg12mo`U}^F!`q1PFQx-j+wCaB# z30)FYYOg=cu^z|So>RyA`4%{ST_sVGeorrEx}e7tU*adI1qM@kY7*OW=$i+}Xi82c zK0dG*9YKeD(GDd!2VCK$@-Y1@qYkpwHW=oWhMFI$xTVgIvD&+yR6p{kevQ!>GvJPQ zPVd6IEknfF$BTxYUBVnY-o5VEJ7VUw3QATNQFDoFRA1H@Usfre@>p${976k7VExiw5cvtj!IrTL78{<~aA?JDQwQKoXbUVGK0xlh1UXP{jw`*+b%QkEPKYR=HCyVu3t;Ebc-3ArVR4R(G|yvoG@z-KV$bFvi(qYan z{=N+b=bpo(-GwA&y#bMOO2oL|%TT&tI|jPAVMvq;_hwQYI{ho9s-c%4^WqQI^PU%V zUZDoNv=UK-e`ej}R>2laPUd{}+hd$r3BEmP1MyzPXs6nR#=dl`8ZA-Zd_?1##`M`caYE5T22N3;zpPd z8Ajy46qEhyIXv+&5l8iB(40k=$>Y^U;Mh}li`OGL#h z6S#XnrjfcbReZEo3*Lxnz=sPQ{VeE#AvN(G*E#D3 z4o9SK(dr*6Fo|}vbyH>OL79Bo_+&adXFq`>4?0kA@Grrt0P0%?cr#O)rs-WL6Tc)< z-LP0(r_sv<{u-nWv$xs4v~>oNkE(d~^gmkqbUe4pub#2oK7%gnHKjWv_d-;y5jtdD zf<+HXAgVQ&oP58Av2m{?qy;kO|EInI8NzQ12qm81X1@^y6|TbF4C0YoW~X8>fm$G%|y}Nvc;H_ zyM>EYh(&d|1QL>A3{oW?(7mS3R-rZz|C4NhEsNV|P|+IbmanGkEdudp&jt`}k>_W; zw&a@HDU1|6j53Ow_`Jsm-?`Gn!NYx2wPH3rYq~^grpVIaB6X;j*24_%t*E5<7BVh5 z!in1^5c;1PF3Im}zBVEbGdA+qt0I0rWsU;(&T}dAJ`sfj zm2{VKH4S+wKvWMj(ccroP@-L!)4AG>hk1U|GRZt_EcC;PrUKmIxD@lcm(-)XJj$6xQShsxl?=abMOu^uOtlz`UeRm{lO6S#G83e7dS zPnzy3(7b$g?#!<}C7i7FAodr;K|_K*Rd0HTUh>mHLC6(*g4%HF ziJ54X(m-z?oej4i+@iw6-^d2OlhUX>f)VoLp>j_x{+C?`3bxARncpBg-)13ZRzIc< zB}eh|l2+U_as+qp7~wfzU9_>S5&!IvBWHHYbGLq_qrI#zw=VDp>%J_8rft7TrW#RN zUHg`qT-->mDA;2ZzgO2Zo&$St?}F1R>9$V~ye11BZHe~ElMwY~9_*G4ccauZhu`(>=qB^aeNg?{d73$$ zpQA3w#ORr4@zm9D!n<<7d&3C(*mD`Y`a1=6r!E42w@NIi@^?UfO~z7(3+Yn zte}1YtG@jj1$_~+zI_Moaruv^w+%yr{Z(>RI~>(cn&1xgZRCmIODYMC=uBski)H<+ zLd64mN=+7jYewV9P6uYg@|DcBd&2bGsZ_?YJQLhM#L?aE_n<$sgBrb;VI}{@p~uNd z+|bD~+?4%@$q-zHQ~Z%}cP!@firH|Uy-$uQJs^ff$KYJODf>vmfOn(Tk)NjH@Y{wm zvTOQ5bgt~-?~if}?z_UMMk|8PdvUNN6PT5gC!tw*JjUnnyR#!#>EaoW@oZLFwW)M8 znbFpShpYIEz4d+O&*Wy*t=@(GYtG_)2PIT!Q$)|npBZ)Om(b`S!!>E^aJs`exb^KR zOj>fCwnR0;4i!q2bQ?*TPzrpvKf;NJU8ILB1ffR7hD|$c3#sAPSSvoq^!3tyBBVKo ze`b1*6-z$Ts%}r~SF`uAkaZ7KyKv%Q{JhJVAVrgTTDUiiBJH_Gr z-4F5clrb7KB^^%1tOCy=0kZ1M5R7Z4dio-7)0~Av7o7wxUfP) z?)h>PF7V!J9RAXa;y)C?Ag_e^c{G5`o}h#is!ho@2@Z~#B|^x3Nju+X3E)trLe2MH zWHhrKU~qW{ov=)gs9_K zEBnr10RLoXun`|`kRqM0)Ufd+{`t=ZrQ|4zUw?Msi0+IK4L`1|Mct5&_|CDiaw`uFcwwhZnt}CQ9n4G8_)is&0|W zlS0JG_d9v=+=$)?Dn_*NBi_@#(laB$aBI3SS$Cikel>}}0y)DBklIz^*y{#NFDDJG?&*3w{>S!TABScy$z) zKH7t$B6aL=Z?`QpPvk1Pf^q7#TR8o~d@dk&Csku?(dnrY`{thuzTRd9yEjIX1CMXv z-iUhEdl{dvkCg_cq)#>0X47!QvH%=xLead`6DDcY)2#MrjuzE(YFS}imiu`0jSa<9 z4Y~MW$5HNHuq-##b1BZ*H5!mT6U|V3!3DQq-4J| zT)iO0ExsW~JPStIti}uI_3}DB?&-??-p4W{K9@lGcOIT{tzwgB{XqM(=WGL9E>WS6 zrkrJKA^S2_hO?V1i;iEHb9bE(e|xCn<6BAGyTNa$n8&l4`TU!NA)gt_Zl*`=v|vr$ z6@2Zu47=*0P@#1KH{;1QuDYccuEq<%>4sR?XCjR8265zkxFFXtA(E8FMNLioFVSzN#80HM^8|V!2}4iWs1aL~+d=v|);Wvp?dBnCfC(NE@1k zOCsCoo9GrYAxn$TjD96C>#wrrs;AL-9zXN4J_eyF#&&DoL{g1)k4eV93EVaQ^U?Kk z3Fah(z*e_pJo&T)P55kaz1|Q#swm6l=%?V%=CdGiZ!YhxRmRc%9n|5wR{W{4}72qn)chV-FH#B{H0d(9{<2eP#v0=ZB z-O9CVAnE#EnxR-j&NQi`d{P1Jj;kk~>zWDc#MYAGm(fh5p{wnnRT%ak|4uf1(W8ny^KqqBGzd$$ zQJ34@Y^%Y3{A_IiVfw|??Y|-B!^DHwDQ$wQ5B_3aZA(G{A6JN-?hNr!cj(p)$3Qvz z2%i;Ef;nS3)F9gs^*4sV>(H%uZiP0KF1SgG2lA=&P&z8gEaf)7nT?Qa15Qp0P_M6% zrU&rg5R(R|>*==@6YE0*dkOO9f+EqS!f+vzr82Vnh=FV)6TLoyoa>#1otMn)E=A~q z^scKga!wVio32vvRblkqrE2`AU+qRG-s#$^d z_ESEm5zLGA?n2|IFJ#?yKs;W*2b+EdQFDQd z_@_pX{oPW+E)RW7&sfG1Lpe#@qtd|c1SzJt?j6}C-;F_mH<-=qk3ese6pnlQ!q(Mh zR`smG92mFxDoh>!1+q-UK{5U;Gq9)v3IopJF*Bf-_3ZE@-&J;$Jwt|;f2UFjRwS=f z5!QkTvH8N!J=rYQV7C%Ey)OaBO{*hS5rX6?BT5_hSz&diEZp+Z!_rNX&|)q{C(F*p zyxUgr{ahQ%+}_89ESLemN_l@tz%lmCdnJ4%x*0VlcCoHgo$##60kXOBE;t=hf@u9j zw#|ME7tHE{2G286lZ=5Kty4Maon@FdcpkH>c|MzM5)~;Iwi{V{3M}VuM3cc1oV(#X z=%sRavv&oI%04Hy{JUV{9NroA*^t&exB-!YO;Fo?2L`x&{4a-hV2NnqT|R>rsc3=E z4ykhTBF&gF{VN+Hs)6Z&ve@xG51tY_rm)*W*Qe(!>ErDAyas0VR%;=BB>lyJM42Jh)qgvQ=> zV!m*M*$_W~S7giRUSkEc|E+~hW&H0KUW%s=->Q)bAE2ey*XjGD>yRFJo@#EM1_dvl zK~RVdY@fE3SscOpIxf8crLiCkapwJyM>rC3fKQw^)fQdz@U8yEL$RH=hZ0y7-dJd+?o(M9jrqVND|ez*Qm@vJ7r? zH`B*&M%i!w+F89bD%|G8qa>!?8b3a~iD!>HlNgzK`1|-t2=33KLZ+#(@S_Sfz8F9( z1?5RWN+X{`R01>pS#qrU8sv975^nJeHah<$eRC_jy6Hy}UUAi_{VRI}%U9*og!Tjw zE`7*cyndg&`yGR}Gv;Cb1ZR43@D!Po>H(3C7T_NE0h8iW=*G))h^PH14eQAuzxWK( zJT{-sS#+5lvGgVTzS?5z&bMs0u^oMS@HVma3W0a>TWCg&7PsfoTB0{|BltB+*){+5 zB^?f1K#=#sho)pwBmE=zc3m7b`1h_lDj^h)1ejqb&zh+G{uOQB6j~Q+D3dRScgd49 z5#rH18|LPwgXceYSo?02?6mI!Q~qx;U|37j_n463t6p$4JcG(qnxMSe9PUJz5^?IU zfJ(g$u#@KkWqw@^U$S!1{(35Vq`@BU7f0ikx2>2Uqsq;{;Rg%;3t%5CyGs=>E6}0o z0<@ql!?wfoJ$e3OBP#n9Q}=BFP&6$XKFRfwxP*S@&BZcy+U7YRY;znH(3pLHv7V@G zZzZGm%89*rB9UA!O_x4AYb$LK_>qsfk#tBj32T~9z$WygA+4@xF=Gja>&fGRa$7J8PR0k)DLAj^629vD#D?|U zW(D-uFegt3;gi@)HCxMhXWj_Gzn^u%{MH>B>ud!*zI*A&!QF7h#t@JG@rA058F10n z7S#r2px?fq8V_Bk^VrSganlFN>pGw;_ajpsgW$E}7s~wnhL?OJ*p)w|*m>7WNUuOH zaY+c^x*~h2Ia^7s+;cE_*c;Y}iDRGnGJdD5L=R35fzJKB_iycC(2=XbSl>PPeN6y+ zY@h{xZ@&u@4?LhR`{r|LOS|d0$IYZ8zSImo)t=dZEkLYrVUn02=+eB;0u4C}(-*4(rJ{LBu zhitd3$z?)BzN26A0Lrv0anBasLeBwNF7@;+5JXKN7yWQj8SiCre}$t5CAiU6brS#k zJT{o==JgKmG= zer_@3{`6xzi_>^!&UE~-JsXu9Wx3`~L+UlMjkPAaGm`M;`$91XMP| z@uwe|s$O3xTV0MKpWdMAssDf<;&Igj$vCB37f*!UC5ZxwWbYJ)ct`;&c=Iv}T`=K$ zKFQFNX9iZi5$prWHa10R0xkWcgfU+fAzt|&iCol3&x<55DzT?&cJeH|kXKKxtIEQ~ zQv=N3aV(kE>+IO4dlY$ENG%^r{+M;i7_LMyR2I{w5)+>s@L#UG8RuUIs^e5Zxw zr);=pgBldB9>7_}OTlZp9}LgY0msZ_CTBz&zb2`|Y1#99-o=-D__fGVTi%)7dZnL; z&5A;aCP9uDKM@bZWVmH5O!p6Zg5*A5P&K#$lXg*d6ITTRJ_U@~d|mqX{4HcE1u&X^ z!MxRXh|vQ(66v{}e$7^cO+WbDD&IdXHvdJNb|vHe&6D^uyox)XI);T))6w&)0^rU> z%8huSL)l5(`RNqDQ+Wg-jh+~0affBMgRJbUOREY9VhSW>#B-LvWlb#>J$8V%yG&_ZA8P?*G03E&y?@#n^I^p#y z4O9tGg_t5SZuJ3O_ITQJdTlhFxD^~H4yR1PN=A?NEE}f1)^X(Yb|oyo-AMMXe8^1n zwdIx?AEBxdBlMJA18M8H3nq_y>GiWRsB}Awvi+y&{JLB+wbT$~)~gb9xI&x5UK5cD zHL~f;C$j#_U81G=fxfVwkDCgnf@_f}3s0wzxt( zESiau=W;mVA4F7Jw$pv$R(T$_YEW(!u`^xv6(;z*)4pxbiN!G^n51b=g6@o`zlMx3J+l{=p3o)t zYx9}${JeO*u^1Vfwgk(z3K{MxgU)7*Um%3w7a zG#Ll0CTXCnLory0_~4X;9M+CoMlWW(VK%R`x4ms_Pj62eBx^rJf!=xzXwTjY3SCBM z6m}5Bm&@RIr)$K*=Oz*DmgVQ4A#iEt6as%9;>6Mc(%pZs`oJYAbd62`iMt2MliV=u z8%@T?Y6A4~%0G zJbm#Xln*UO>pstv6?uV2~b0Ula5gB~@xV=HmOA3ik9 zB^!1wg~_8%%>E!(Xoz0KY&_LS7IgBSy?K$~J$DH{eEyH#65_3k|K^b##)A$$&w}jsmvCT>IPXNtua?`Oh)Txkwo&r& zn67^izoyiH%dbwXi!;I?Uul%ex=BPX%;fwYW^?xJWJUQO~9zkoP%*eRX5)%6D7`d0Dk1sbN)3()zbC#RSE!t(p?Rag4 zdawjzDxFcoW4fL3eIxSpS@062zc26{PoSg3s4sW}aaMoRRZ}t_@rH*@+!TA7;~Hekb-MAsfqI`A{F;6_+no zO7F@pgp|x!l63w!e0w#~E^O{ezIP(ZP3l}hqYakme_ow>u3Jrog&)C>(khTT zw202`34`Lv-k`sBCt2ihg@hI}Fh!^n-#inm-QcVUA5!gbaMM3rb|?qdZ0W;5r}b#G zeJb3R*vFa#E&&xuHyFOY3`9TJlE#a#>HRB-I8#>}M^lW+nDa_@N4nG7{U(3{0`vkk7&BY7)pR(9vYvrji@SlkCPc#yyM z?@xvTdqsS=?jB^{)q}tpFNue)Cd~Gji{ERLIOh^+zBfGw#r+lW>koAFJ11vnJ37Lxv`63%TNaZzq!zP9dWsD$$!-ot-OQG%9Vn#Nr2tTJLaO3|=g;gJy;y33i zD3-m6%`YzF=CR+^_Y4=qn%inHwC6qfD5*~_%CwR*%QZo_>K|S5?TcokF~0*Ajz3G5DEY%AB$1Gb)|8=%QJE-2BROv{BLo z)8FLP?CdS03A07HTi4rg>V$h37rGppogLZQ3>~JmdNEhy91hYg#bm)sBchNN1$&Dv ziT}A8qOy?R2kcR!iRH(+H#Hr$9&bKj(lURTz~75e;{2d^G}w<+@A zn5Xy3PMgYw1TX<=}s$h;H6Ajptb- zI;}cQZn;#@9}k}s@z1B&k_+xINtQ#`C`~+m;36aQw-tZx)TXY+K777FjPo%b#JfAB z$dYCA;6T=O_|@)=rrygST*?7A%P11#UF*?VYDLYe^An-eB^J6jd}dG0lflf72e?_I zmT0M1fpu5Ek$k%m`mR}syKOWPjVBipzYzi0=A&pge4ghrkHwSqHzpJNhnI2IhhDP5 zE+2%y=rhv=8pyqwJ27JK6fCehj(>_`8T)IMVBPYT99t7WK7Oj8_5U8zMfVHvSVadr zJN7Mop0a?e6n~5>9&IDL*08W@)B)ag7SU;AMX*caF+bnXqJQN%vi55(TQEes9-BQ`@S-UTdLBN$v;WNI62&uCQC;R+embC54*>I zHu|c6U_{fzaMgsnwEf0>UM22 z+Ssmzv%r1aa<1xX8s57$0dJh~VyZ(IV}CYI=|u z!Ot!)kB7JcBToIt22#Ezj7n^JO|F~z5vObC86y)djELQf+44tW+~|L3Ijs#pL@mSz z+82nDmn7HG$=s1eE)ch6BTirf^+wZ!`!~hblcaN z@MvKuE}l7$cs~-ewO_NJ(u%ouE9xf^=>RX75!}W8IWq&ay7CC?dj*5vUqrR`3+(vV zGPHI~#=IMk+0aEZaNGSTV%-!BT`}5>Mp!NknWsa={AqRtUIlpY#4(CTf5KkbHAsI- zV^yXJ9j6<|L~3Nx;*IO5*YJ7z^m-I2F**z-CT`vxAh^AM+HMa)4%ppF>KO$NT?L$Vou(of6bL3(v|<3`UCwWkKqcN6_{8e$Qdr%h^vnMtV#1T zfrTPNbeHX8a@6S_tMK;~*%kQ#7U*}dQOrH5M*p&k)&PA&Efx_-d zux6Ve-1c#>d$e2&guktUFYybxzUTRvY@@(+{mdkiE5#wfM-|#?F4NwLcgWAK?O5az zNM?Bol8D|=Ftq%PB|n}qm;MXl_eiSPZSja12ra>a+x@gz?GEu|4Q(H{X+!J^OWZRq z7CVEK!8SOOj`!AsVFLj!;njZNlGEwd@M!Gabqb7D7{S=)1`G+BMa~9q#;48iP$s($ z&q)^J$l2BGutq8KdfREzHX{u?dvDUf?6WBB;6uh^68X-v-#re+arKYJQ41a3k!t0@ zNeG4D!e7TQsBSuLj&a0g_q@o1r5?7vn?uRc4`!UdeI9;x*QIH3zu25#_UwN*lF7Vp zyikE{q>)4OA?r*Ybxx3_N4qGwWc-_XBp68-B^=`W211A|- z-dDb6I;3|@AzO2Jm&+L$xHL@~XSAkJ^Q1mhP~_PWs+&oC%LmN)b(16}M6>IX2C!$F z03O_Rhi=Y!LLGbRaoM9^P;*9#h-_tvNx?YK3k}AK>o>tpNZjtl#KfA!fk=4#a|^vw zP>*c!6-Laxhfs$cJbV5Dipi$n%n7b&IX;TcT3ZPRg5sgGVJo~ZY(X!*3oyDQo9sCh zOnP7DVJF3WKpoKrpr+il0_lbRJEn_`6;fmyIHBph?+1+n++Z{yq6xtvVnOsX_w zz`N-zaOr&o@LnAOM;=|m56|1MSwWtT1;)`AdV{R})+qY0*9g}!sqiLX$R_)6683~B zb8h`d$f1iSpcI;ivi$uLQSpn!FSdb`dy`sq`%mk2cxN>RXNg&g_9yLN0Imy#$WbW`A^kjF_gdL?2GbBa? zLYvWN7oWBB9jrNiL4uT4Nzwl@6k*X3G2CA+L(W@^LE)zYa&>qa_}WY3&yP;TLOc;t zbe%Em$xCLON+RSumgklveZpJmCD{Ic3(j^HwVQI#mb+%qgTpQWTlciVxjHipIkyij zKQ5uCt0tgIq$y0=@RRWo6yRJg?%>o7&w%Uk3>r7Sj$6K7fV*lS0|l-6+^evo=rWK* zwCB8L-kqMr%}vZ_$69CNUL6JY*(x3CR3^uHU5Ur~UTySbBC$|-J{WDP%|U&;!IY_bh4oLfS=w|a0*d_TK~cLwGPt>6}&T8W1pa)~L==~0QBNDS6R z5@)qzK&18&IT3HV<=a}UuHL}Ss@9}}XM8|hrv>%R!bs1jDE@Qz1k7#&{A2WjuKOg8 zc9|kn_Q(?6J8a53YdeT#jxgD{u#kM-;R7Lm_?f%9JGbECKGf3O2N7R=iFxfucK5^I z^uMh)iJh`PD400lEb+bc$>2dIua?1wuT@|gA8@H#tOjvgcF|=ajm)@_XV}wSiCSOC`Ygty*Y0A%`n`AOMS^M7cozOjftog5S%m ziMsbE+ULacTK{cj&$ozky`B-^c9!ovO}>l^d=0DT#i_xZn<~U^s1#mg4v~$%S74Ry z7LfN0Mdf8rx!3PJF!~D53>oE>G(P&=_nY$ErP~v@$NNtMybb}!>TIYwewtfYb(*|9 zb(;Ou_6>je&F9{SXn^>h>D=NOH_0_d5S|1yp!4NC5;?pRT*A+gBI_M^@d?k1IzNRN z?pO$8_pZT`eFM~Xk18rYHNe^7OQB9Fk}NNkp+oOefp_B2U6#}7zHc(*P`)pkS4M(< z*aN)3I}R%LG?38PWDr`b3|n}1ti0B4d|`E!a1&L)(sTfeM<<|j{1$M~a)SMdr@&|K zcon1RbsGp#5qTrt%y&fnY0mFSP=L-YIj7Q=H+zxLlAVg7`d1 zf?oflicjvFkhG3?2uu>8PlqP)J+i6rQc#$D{wszdatb)-aRs%%Q^5Ae^Ngc}DAIoU zB(&UWAgKi$c_)7gN9>-#=B%Skrr32jR3!;YDt$~#;yILlZAJPmKjEY5=% z2?|FvVaNU`bmsf`^}OpzqV+jR_ieHj7uUfI$%&YHa53hXs!+>0F*sq$YEx@6E(C45 z#CX~$QrCfEdhaLC2Dx|*I-X30#a_1w_vtaKJS4~E@f}jNg{yEwdlD$Fn2P4xud~-3 zqiTBN*5kD6XVLgsDxA{c_ck(aG*kB*x@#W9`S;~;)b$p5(WeBP)%aN|pNCy(m58-1 z0W|r`ahu;;j}o73NcBGCp^iyEob&i^}5YN+lEs7-J3CIJ$#p z!S@e7WBv}c(q}i`U|pv>{*au8`iGa3i|GUSEA<$-ZrTKaWC{$lg@MObQPxy%Hyn0X zfyu4YkvuEM$CBUJpKk={H)~;bbZ7#BJvtz6qYDM8QMg?2BN|kyaWycz}$NBWx)j0>>Fjm3MZ)9l3e~mnx566R@`<^) zr0EXva=TBCn*&(?QXw@tk$7mh5?}910a=NYbfuW}|;C?Zin53D4vPc14R`VHd{Mw1F9wo4;JrFD`;)z~z5UGj$j82_P z(0F|qB?clq6KXacGzz5#!&|_9OEf0__C<^2Yml1OOMgX8q*61)&_k?{TJKfBZSe+J zyv!Cwx3^)0%zB6#HX~1;h|}`23?eXJ4=dkI;NB^&Bq!wM!PAIma|@inZ1=0Ay>$cJ z81FzmobQs+i0Sxm%65+Ki6++mCZHOUimrGRH#crakBC${j*3#p*hM@u&I#uDj05`z zQRGTRBDOx$$3F*VV7%01u204c+-GjU1Cy51fCW0l`@l1zll9=N{}`=PP3O5D zb4hiPF8q1df-#a|tZ#}k2zgeMTNR7&hj~7BXiDOz2_`(>F9=U>t4HtMnM|(Ebz=NE z23(H4WcpV4Lo`HTN6QGa_x1k_op(4^Zy3jIva&tOW_jJhNcu10D7wd5Vpp0g8Mh=3M3xl5!BAEBuXh5xTv?3!E$6^^M^`?-9t3-= z^y!FYI2D&xr*3Eana=U*IKg=fw`K+J!>J#G$Nok`pZ{jYrPhbOPzqzUSUFtf{eu?x zr-1pHCcOR28xsS^;0?)5z$q~~_9BlwwJSjH+snauC(mNOy9`>AD+uNNte*`Z@x7I8 zSh`>cGlc5s_NSD{t_@~hWM8MRvc%Y*{QK$VAkRQFe#f47uqJ;}zR)8fwm5HW9Q8du zKmrz*v!4E2NmcSgNd6teYLY(@3h-AskhE+b2!bh`&m?ydoGDh8IW9>bu zwYLdy8piOZCzlP_Gl!7Q$JEn|p<`E`V7BjOh~Uv&a%=w@*w-fuPoEk?)(tc48TkP6 z`^u^pSI4n|N4KFvz$AR@s*C>Q8S;_ynp5VcOoO{STs_J2nBP{@;E*uh>obA8f7?iH z4ZlKJqD9TBP5Sg&<5qkH+I-h{n9owDlhxr$?A+!k$geWM)V^Z2;F~2p`f{V@+K2?* zdQBVsg1S(nVh-O4jeyw;;~})x9={yF%lbZl%ZMFTp=k|u*nV#_F0GS?iJNkWU}Zbw zptT-*tj6G(^F4U)XaX`}1^B3ME{$^73&qB1d@nT~q(3;qov@qm&uTQ+_i;M*Tb-n_ zHk+XSfEXAHf?>tX4Cb#NlHVP2;GcRP1MaS){)r~=X7WNDB^A$%>01Yx)1=AEKbmw< z`U%|?J0F96lJJ33HtL?1=cZ~ZqSmw#yysWG6sXb$K#_1qo|*#87yBkz$}e+rH2;n;pXY`8Ccm!td)$yfI?-Me$o#*J#Q_Nw+agsEt1s zr%*fpWIQtO5zWJ&;}S@+VzHzvTLnL^IvmOVRfP;b#nwj?qUGA6%ev+}}-$1@>oGR>FPBr6hQeFbvn^&Nb;sRYxQ zPGkerh*M7>eDsilK82Mqp|p{0{_2H0AMeAh6U=B`uoPUhK1yQ}W#I3pD!Oh(Dani- zM-PeJs9BmL2KM4#=(X5&G_HF$_?dqvRuQ6jSw0m{x!aRjR#s%EbttiUq6nMxwtzc7 zJI#KrMz@b~L6+73Eh_dfmxNAOuK_l!nP+fVUkf3yw{R|<7omq?SdDl7;}+~UR1+8G8F9b3Q|3A42-WL*ebK74UEVZm9N}OjevYOx25y6NRP8!ez&uAyKCkWJe@W?)fNK zG&Blio5GpKMlX1JWj-AGBp^v~Z`hs2M=^C+R=8#DGRVJ^0v`vX@Ynr&OoK%W4UrrR zvQC;9T=$PoP*J8U=}vkeia&?-q>?2w*P?hiuQb? z?rS5d629ZTB8E6$k!6n4yXfrnf?7R!jTc925d|Y@VYH<$q{NO!_OB6;YeyNCof~lQ z{$2Xzo)h^pmv;fgc)_uT=cH|1EIAu^lsWXliu+HYfgZV%P7-fErzI0^5xuTkv_01w zhL1?$!7;1hUx*KAcz>Z2{q2~ar~Aki(1Yqrx9Q_$TKt_|iiVau(oZu_Qi*#e+=lEe zJ$q2`DItl{yrrXka*&_a}j0N*T4=-55R(>=qud<_l|rb z|0>n7GHWfp67i3&fhzj1Uebi}Icf%#J{Jxk_@4GmTPn(Wy_I&TO zcMnd=%_Z}6^k}i$H?~7DnVh=kCJ0_U2Te_NF`zB6`mj!`pn1m}IB{+}DeP{5nJdbn zuD~0IHW{Ilkq1sWR)jVd{ zhgnTu@fD(TNr7m`$zjpvLi~E&oTIXZq{eC{mT|7Mmy}YE0&OOw2VyR(3tfcqhLgN@xKI#-re=PzRMy|obSxxM^|MW@Mr6LR+JA>9v=);xw z&Xcd|#xV4ynOc>(GpqNFMm>*UIyNl;ocRWM_5 z$Bk+DC<52^cQM*`4N-l51Nk)09E&?;@YXx~YA@L_)Nb;06pea@8~Y3Khi49oWO$N3 z_f_=D0bk?tbfXY9tK zJd?$H%UXOj{v0DNwgKnKY$b9IvZVCtG<4tGN?%@7w$P4MzQ3+k)9pGNiB3(kCKrn{DOV&hp8P~PfBLio-?r&J+< z*VAB?rW`$dRg6y2`pK&0KcvEq#q@!PI)+{O1XaQ6_%N2GR^E=(pKXEhO`*iy)(6Ur z%vk%1EV?O2TWBse52SU<*{{z0Uf=H$EZ$K_HY|*S%^m%eH8KJ3+2_$sNdav>kI>w) zm+-A;96GQHoONdqE%jLraUaa#_JmM)@Gc9JEW6lMwc5l>P(mELyrAHo0huuF5&oJk z!=0||h5D~o@$jK*tbeBo6Q!0!4lm5Yy7V<%<-ipD-E@w;5732(@JF=F>@aEf3WiW) z4YF{{ZNZe2$t3B`YrJaCdq}*sn8ofbuvBdt3F0 zwQERjjRv>eVGk}lE&(qe_L8UKH)-&@NBFHf1cj1EF?`ZHI}c5@_fABsS!DI`lp|ix!lJ!*dJ?zFt8yBd=~Z*d587F=Bo_T&~gkXU1}tZ zgAlrJ&4B5_Ul^&YQfU7~fbqJq__`q%M{k(O9a|T}re*MMu*}V%719D_ZlCD1PM(2$ zT2eTeGL0;&|A6YclRzg=AD$@g0NG!kVe0#vbYYPqRckS4*IiR3`^VOkO+6RktD_aP zM=gbd1|5(vi2#w&rY zPKKx-&N!U+goI-Z=KZxmKSv*&-~@DOi59iJR7!p+^H~tu!D#3D;YkS#xTI|W@{iWT zw5#hO=HO#N_I-Xn>DA9JFllA~v?hUm!9>iyQh>pM8@Y92l}y8oA@W`^AD1lpNM8&j zp=t0%PI}Tz6pjceF}ug}lX9n@Tqzd@uU6>dlo`daTk9B@N+y7dLpgYP zzkt!R?C20bLEFzJ->-wRlQ1n z1Q^5S64r-HKDIow=XMh9l1(N1W%I}fQ#qh#HRva^ zFiu-1l)Yb0nKT++uea2$;9E zBCz!sgN~#7$ghrgGGg%#&SZb0N>eiExjE(Rg9j3zBY47IFP#m^`xju7o)(tP$s-Xh zOHp5~j_%eThim=T)8zPUdb{Q?eW2+>dkx2-n868{oZ83Q9Mc1dp2tM2_fn1X*K8){ zS3Bk;TjGXup>*>3<20#uJsh83PYsgWxCttc@suq`@A3?Gi-CKT@%&5&PXEAzne^%CAYG6kfw0mj%bpq zwU@DQ??QZ#`~VVlm2s0vB=d1+2Vv}W;I)f8EZVUWya)Mgu9G7CbF+uPiObM=#sS{R zGX_V=G?LbSQ*y>b7PZnkiTW){7QKECF+q3nWv)L?@NGeZdJTGWog-bGXw3!A&Z3{D zdg7jIJR`B!iL<-Vgw9oZ+eY(mQfd>(m1M*Ar;9*Tg<$a9)!e4?DC`PN zBI3U3MEapA?X7K~zs9SO27P(Xw{ZwBoIzS?;w!kHScoGQ2C%Z9-}$HH(K^3oYI@)| zbIUT1=sTEj{_*EP$e*Vj>c_K*EpK3Y(t4;Q66jc6N$mJe*u!dtn%0s$^4+wU{Jv1Y zM9y5#EPi^6{BEnHyla;<9F&0I!KtKP^$87XJCB~*e=iw@lB?rI_Enh*j`6 z!jfKR4mxUBxi=is9oqn|5?N$oo-`&sqI70ehalpIEeO6vF)Fqf=*vr&aB|RUCdZ?U z1W&(0=j^bny9;#gYl>~I9mH`1!#Sm`!+jo&l&`FV=u<_`_0xTxUnGiN8*|9- zm<{N*L>gaw@Ic4UUm2|$e-QVJ1=-u{A+G-)bBCXqwvQ?&>jv|&aOZYz_AMEBGP{|a zR+1IYOzviO{5(tMZxACBWz8{VxRkbp?SXO+7r1q69ke*l!3l=L_#`(1|9E%c`zT#u zi`HrSSSkriPQ~Ld3m2Tx;LAM}orY1_m#DhlcwBWcol3`FBP-Tx2^C-dW|Na!nWKR2F6=A`Z?S2HJ)b2d6)^?R*d=lr9@edBk<^ZJ>pTS207xUWO`UXt+^@7ofvx+E#Ll!zKuJGJ?}PT_&X#^ z`8~$40TG!x3Ug($P`y_KTw~ADQk5o5PuYq6tFBXT{?3xP(+h@Y)sSz$mO*K=C>}O3 zK>5U06qOLrz#KwCWz2BhPgB9!=bwmX{mPnM-J;w{J3~m_Aqt|)df3-VXUN}4b~OK0 zIPX%jM8}Wf+_u{bu-A|8ddI3j?vy;Lf33MjZ<;seNj5QiLaIpHE&$mFlhF2JIS$rJ za=mT*X+7*bsD++JU7HD9XNx4qgF;~MbuscSXIQW)hUbFnSFrjebK&wYAt)`}jc1Mf zi9%{4k#?vh4zs87@68UTp45^2ndg{M$F36bz)NI8Fz7r}j=g@19I{vnr zg9~iLF+Dnoy4Bwy2X<*a+O>ny*o%pSuUj-5!vYhPnOnS!PX{m9O`E^zgoMSfnm z3`*nlY5RQ-${l-%D&Cf)h5uVB7`0&BhGKa3!V@O*`;^7sJz%&mg8goj3E?lwplru) z_E4NCZvI*@2**~HjM0BsNQvCYJp8(DpdGc>g&<;R3j6!Q@h z>ozmviwyC|{nHp_tV~Q@G~sm9bgc4Zpe%eJP4=tC`)jW=J-UCWs-_#)F!?(hQ>9NV zZc4!N?s7E!cosKyPv`E572(B%HryB!iAPQ4Im;Y(o?E7Yb-vlSyL2J<;dnA9KcdKW zO_!mnypy0_*$Lhp{LQR=ED5(i+@XQv7H|Q3)z~wKLQt9SIJ;4PhgwukZna+_8Qppq zHMWLMXP;5)wPvt%pCS19p69dbCfswy3e3$Pz*YZga0kR@a%=OJa{_yQmc8jE`TI~B zwp*{jAg@>meB}$}7B)CAU4%QoUyL((VZu#HYvP{;H!dr|iaT=VD4t&M1;+#!aPQl@ zLHySu?v$uFw>m`zzg{gTW;?$Toy;88%;F?EtJrdDp1dN0xOjAGILp7oeleL_k_c2U zC9gd#aH!V`A4E@sn=4{)_Y2MIrDAc{hw1_$EWUtzoVA-g;qwBuF)|QZunz2du7jsn zA{TWnm77}ekY4?77a84n3DhE5U?5e7PMT>e+$L}q)}@qy+wz;-zw2tW`}J(tQk{zq zleBT|Mm6T$E@P;Yc*E>F;tB14)P>$=Q!%1s7b($6Wn03hkc-Op5b;j~{#_8II=e=r z(hezVGrI_%K4Xv{U*Y*<64YL_m#80oN+owU)99QROvcJOJl7bFw`SSm#9B$*k#7!9 zrL|yp-!&5V=NGwoQ4L;%Q!rpeasA~o(kNOCv$!J|e{v${)#Ssi`xwjjolep1qep@D z$zt4h&K6SpZql3C+wp;BFm1VVoz@5X!=jNm`t{^5!Lu1RN$Jg1Ja<(KFMYg-7tFWe z;lqbH?~ZFs?>h#ktkEF1l;`5BqH%C=jxTH%`v`9q$>HV|NifPv6K)rsCqds6h-TFq zwmv8tf2iE&9=14gA8jh=T;&+1@@fc?2)Bpss2g;d#Sd~t<{`*FY=QSJvBV=d9?vgP z!-2kfbhlI^!DB+ zCt45&GQs1p*|d?JlOrT!i<0r`pboB_coSC$Rbk9xA$??=0zMfRS<(GzbnHJVOcL&d zZQU`<-tuPb4*E+C(#5%!YAu|e8%I0%j7oykG%o*BG>X9n_KmR-oN?s6C6`{KnaoYx z(FSx*bq!VYJ4*_>9@9q01kRIZS#RHe8mF9`4OS&lwB!8}u4CgI?(Y&lGxYEY-uTka zT|8IME}!y)l)n_?Hb!Z48}6pmNZlDG(Yx0`Ou=Qgrl0p)y71WlfzjOcw&QgB@A0tj z$$GH5@q+#^im6%k;U)Wiodp&ay~5-JXK{L$F-8pX{$G`F8vdz_&nij^x0i;Y;qozD zd)fnZyB$Dk7Z}5NDK%VYvyvo>m63o#5xV)sZv5)~7hPwn;}2g~m~8zIFIJsrdJ3iZ zp1~0A&8op|ukw)d@Z(NQQib_*Qc3q@HEx{xL^568oEVVRbO9SrM0aX&LZzA9lBeBx z)aX9FHqMmOEbGLiZ~<*r+{_7gNO4w6W7!z>SP&`dBX`&E*-i~_GWbP;iJfFam0K^e z7E9FZCWd|?g9r76M?Uyrv%)*lU_J`})r_U)+*d}eFdve_FXFelEL6UI8IQd`j_-VQ zsjT)8fFKEt8V=&Rux(siiW;tz*o|M>CcuU|XPl++l|)=tB6s?e;7OeqlX2w~vDkfq zIiJFJDGo$)I*-QUHG4HwepZe*?v&CarSa6?Mig$yWpay7m~kB|$FQ-9$?Wt&L)Qp;=nLB8_^j)a6x#OA9v*%D$@)H0#9yU4u<4F1X}z%&;cXdacz;0S ziOaa_lXl}$4SwFV#TxsUCXm}VF2eqTS~_mFBt1A`F6wUYL>ps>ZX!d6^@X>Qg0 zxmFA}OMRkQJvw~PYAu--^OxDWHi2Yldok{FlP|s7meTIEHKtj2CY475L~X&UGG#dCP9;Dcr1r> zKQWN1olll#O5>&R_9S!BXe^v13M=m3qn#lIf>!sh)JHZ2MIY@zKk^Fd`;^ICPcslZ zuotzrcHrx_aL(iPEZBcC86zXUGSYB^KCirM2Mv6`;ORJYDs89CTTOVeO9xJ!2q&^U z`)nC{LHT}De16203xPH!J39hrBrM=29x5OWS9!Pm(P$hxG>#h^V}LH~xC zc42e9G&k}nmHT`<9NbRn!HMGt2Vy2w+gpr=yzjH{vC9IKT!l>Kk7<;1=%aayT5v|J z8n;W|5P!6Gp{7X<24%EkjQ=n>@zfB*vTxc>DoyJN;otQj)GcGV2QCF!+L{z!%X=~mzrE*D@M@3-10 z`ON!vo-a5SZC=$Pi$kQpu9#V0s0unA?L`L_YB3tddXz|o?rX^YgK9|d5RmwL^ zqm9hVBOhsv$QIh`%X?g%L`m&TN9;I!f^{u6govrXNkZ&=8u!Kx)@o$qviWaW72TE8 z`yGFWT=0{w{5y}eiG70+Ys+buY7kAgp2!7nn924ZQl^>x1*Bfm@3jKR@T# z;^#}x&%aKtZkOUTmW<1tOchs~G&z-0DZ zW}j^9qp{gWXcw3bVf*DFrtL48=)MT@clg36et-D#))z364abJL3#gQ)ErcJFVSShW zqPfz1|Hez28z0_)fx0rtLmurqy*=S~XI{06iwqs;oPj-lt4Zupd3+aCPp28oM8ol6 zwkHR*p-Z`f9C6bWHb~3ESHT;2RIR|BflB)HHP6g&n}O~>Rmt2b;~+2a75F(Upn^{) z$#0ibZpq$AY+iDgny9UUikPWnR?|)h_EaOg4GQqrVZQlYFdITmDyZzcFYJ(oF6 z>92TMo_VZ9ckn!#s?}d`T={m6T^WE%?xW~l*HSE5SIs^?d7Y>}sKQHpexiQK4A{H$ zI4#k!CQ2bfsG6Vyt}nL3`v^z+#a{~Wsoj&t;T6=EFn{ZpWx`dYhXR7K&6MI&_^m8 z8>0{6`q}mL$jcR&ex)4Fm1x4q_jFF#M+(P=DR9#ZX0vNwTxE9s8x4;p>vHDV_vpan zIJSbdg@80iFn==}#_0%%$0#wnRYjYmyk3Q}JX>(IRs>o4pr5_d(?oVP&cei~F~U00 ziMW4GJM^{(V#1x1RImCH+`DR3^L*ZN7;hj7x3)*&(&hZTfd?C-cA==I`Zwd`3IKzgrD%r1*2*4s~eZeHJJE z(uB_bV_0={E_`rEM(4zFT*`b+Vv!&L+r!()kIYC&{LdYVU94zESqrUOEW*9Y+XJCH z^H6VtBAnY(K{A&dBn|Qx=nZfFd3=_|?V>)yu7D8eh*~Yo46Z=ap$g(*eheq~hr;#s zNpMnh5N2mh5VDq6$uxay*kOK~u40z+`NBc=QRot(Qb!^5thfP58*Ct9n>tL=*p1~K z74&wP5i9t_b2lrhnd?XDpi=#mu*D#Z?=n=t&M#}>2^|N@g>z`!J86(|-y|HpIfXf! zY%Xj&`5BzfY((n=3*g(LZNkIDf8c>!DZBQ4l<*3O3Vpn{3K>Or5Ob*{t)?5{!j}Qq zsqIV}f_3nUd=^vk!j?=v<0c%Z87+|f5ePZbNy07V7my43Lg#N$We%&)p`q3H>0f&} zMn+^bcDx$~lhw{bBY#CseK-Xb2J`K1nGKK%l?04hXT^#vI{>vmR?wszP43mCaopD9 zm+01P^_q6M(L%coze)bxY1CzxE41j2!k^mt6={M;A zb~!o~_LA48Vz6b^Xwq{&6}pxA-OS1T_$wxaOzGxQY!M zGt7*kxqN4)jTl%SXX^Pp)|-ZdAX2uhTG_vc+zSX6tgFa@!gf;(l^G^tVlAXKu$WfP zFyy#m5pI9w0o?4PO+qG%!TO<(q~6pN8cy=N`q1a3CS*T3koygkT2+OowuA_es%8rB zuQn9sm%XK*&EuGwDFVa>En$S|N=(-_<(=^ZbWh-ad_JTEYE0Tebjbp^HMAEO#fO1y zSt_mRe+8evmSaP|I2ChJVC~FG@v~7q^GbdhxSDDSJBxBYbsO=3=wjiylw)+m zoD$e}`Ug~6tt7`@z9e~%C!oHVH;~20NJzc|0Ofa-|3C; z=$R46RcOO=o;@s{yinNpOHtS`;TVkdj;8B8Pq8DLR0S?8gHihFe3X^iffbd}U?{l& z9yW~;y6t-pQ{e@ssr_UJECNxu+yNu+@c&Z=HNa)P0qY$O0?}^(lN2tK{3p%iuek{X z>}iA-f%?p{9yi!|tdLkpuEbXBU}{!5pH}}{3Ol!nl5t-ik#iE-;1p6#-=8!E-M?wz zF{g-l?f=IN^*0cYS2J)4B0Ie6fK0HYSlV9AmDaA<`% z*`B)77H+HQlZr_6`Vzr=29oLe)_6SHIZPB9-!m<`6JhJ`iO{}E14I&YU`wme(-E~)Hn<11*?2s}^Cf=xX0q4gE|cX_)8YH^L^8lLnLVEtLRfb?&K}4?lU5UM z&?^=HsoP;wfCTr>Ta%M_V(=6r&K3LpW|#R)pmhr}nMY!?I6~w&eP6^?_cp_q{j0#l zRs~LUR+A6$mSl!TBseS}C>vl3r~gGG*Rq(lZcW1GvT&ZSAi?EkS`y|zFZ7w)gAJFH z(JV)obIWNZv*J5R->Hl6-7=jWG;N~3zV0w{rv{9FnZuL}Js>6t1Jv)$0n{PyG4?Ef z7QHP(e3JQm@uf8|I6IV?BofJ<>(68puCi>vMJZuS<4a8TdPoQPIqhau4JfZxhCKCl z_SK0JI2Iy8QjK@ey**2Djjk5)@EZpo6s2$#*pr}3wkS~4;1qZ7#y9EHxv_0ds5Q?F zo5Qwq5&NU5`QEQIX9w?m`}CYyk=8&axJE+PUrVC4Z50{cHXhUtrD7zX5U{&4z&O7; zhAaN;X9{R3Ii<4>KmM$t>9X!v@b4;Ix%Qa<9Nse%c@)b6ImVIyP7l_2g4*+Nq9a%g zWHMFYfnoDY*p$crYA!NK)TLI5 zSgVOqc4Hj#?twH{k{ySoAHZbXp#~+ArZ+8<}~8tJv_H??-|@gpJCRu`y~Cxf0$>!96fub zVc)mcr10-561`zHs94>n-GTeT<>y1}n$$s?A6ao1Z#`vhCG5oqtEO`bMq=PH&XnSV zU^H?kpz~xkyvY;?^I0RbaQ7EhpB@1LpW{)wk`H?_{Xz3xH^Sl+EY-~Bq#|l?-D)dt zzUgVMFVuw7EOo(zh9|g7eiTt(n?u=tIdqwo1ZQ?!B{L*r;Nq(cD*N{X)*Zcow=}N- z{4FPoOm>p{qZDvl(NX5c!z5A}h~nI}Mz|9nw&2ZnO0rwzguQZ7_^HpB-Vtj= zQ@&eqZfhA#mzU$Zyrf{xwDH`7 zy5OSANhUW-0yUESq4J$Hr!Sns4aT=~XHWgdJHn(fuI(wB$jS-T>mNXZd<|S$tP6+A zBuVV6-6ZnSE5>A{I6V|}mrRSBii38xMEyezb#0CW<2_SYU9&alXWCICb0i-B*jS^; zwK5cMx*>RXA(#pz7t$~H*E1_?C(*I9>#=R)3HTy+j!ABQgOyGnYsO@nz+{KlY03*BX!bl9vXg)wNmWws+QXWk;(Jbtk!ZK2h?Oe*M?acRPoN2 z6|dIfw$*=_Ga)Ql9@In=*CfEelHVkva1XgsQ31WR)8WdjRZv#(h+U>9$zA^M6utFV zlFt!NgPhNGFyrW0E?n#khLtD4+uQcE>17LH%I~pVf%T*>zNqG)=nhC4n?P4M-v#4p zFFLxd3fseNapdk?Ze*n{hV-jJ0G|~<8&wPU7pIbyx(38v&5YX4w#8Xhb~RnM#gU6l z#lv|=aK3a5jkTAf2L%;mYd|e|xke4_ACCv7W*AGp3OP@15x205&utyzGxc;D7xnT2 z>|0epWbL1`Tlt;ft@2*5X>@>135_(|`U&ZSB>W;0jlcY4>B66mXrU90sbMnQbbAi{ z8^+?p)=*N?`H}TrU_#E^eNUDjZ=f0H!??IGH%`qkp9`>A1NRemvS;&7kw$HK?!@;Q z+>EeDE_uNOr~`d)%{ojL=BCoc=af)?Zv}jNb(NepJOMqkjY08G4*B9`2BlIG?EMvc zX!8qYZ2Fi>icj8R4_=SK)mFWD*I*e&tQ3VN-o-27{}rxo8K%}Hx_J4W3`%Ln;AZtJ zD4OsE4d(X32?rz2c;_qp`B;zhT3m$7L_|2JkezU#KGy`YQ9boQpl-Avg#4edqG>YkD;x#?l)P4o7_Lnx! zKKX`z;#u8iOAk`I{3BZ2cdOC+sX<&p-E;H58QFxNNMLP@E2tim^qP-@0UK|BqmYyZ59a(7SnvBO~I&p8M0=H8~ z4Q9oJV!KNfw+E(BSC4;2p=6JzJ|MB{ReRELl>O4IR-tOrbFVo3S58aHu_tNLqtj^`Tcbo`UT&^&oP$p zCHM-?*?56l{bwy_@zRZ|#(svkns-nw-yg4I3yqNcj7cKiG~|vy&ie5bgZ?(zu5k*1 z>shhHWWE8mxqL&r$va5$(0^Fycn?Qa`oYVV+oWGkoC_B=Lu1&xhhk z;Eoc;vr>h7+B6=oXpW(omJ_)f-%@erfuHnty%e{z(2m>HwTYgpEJxk2_ky*3)2Zd1 zQ1;Oy4h_s(@uP)+`KVDt)IG|<`1c9ia8jEprmcp(olRuyL?JrxpEaju6XUHvhWnHe z%$WyIf;Jx&czC9UPBni`K67d~*sIT-a1JB~Uw*))o61o+b|E-|?F1@9~427k|je`n6>A@^~T?D8jC4 zHEu>^A${(q$rYaS;Y@?3aK+<4<7jgWPPKasnf^6Vu%U4+xQ$$!&nuRN(wqkhN4t&eYth>vKiDQre?;X?wIi8=QI%gT(EXBJAKIM?< z1r?awJ{gaUXmMM{w-d*r3N*Xx1;Kk?qJG#kZd%Yk5Zo<>@D3SH^2h=XY~zuA^PZ`F zZ%uZ6zd;6~{ZO(i9F@2_er_no1^xO=X9p=`wbKU-wynh3hH{*^;u!Ax{wti*V_Pm+ z-jp+3?Sulko#fld4hRmE67C4t4rY9|_NdeVYE}opB8e8V`*1tIH+(_=VJd3xKMR&Y zXF)M(2Bcni%li*X$-=xNL{&xzDW|sK-HVo7(ZXxI|8fahe@n*LxT##JW+!qb(KLp4 zz-HV?fbxtsrfg#r(Nyk5`S2N-ZuB3X{yB&4+P?#T^;F|U+x+!)zX2Ti@RZ-$u zag}!K{D87|sv!FUK)rW09CFB8fL$WlWITj#?;k+1`wlfVi~|kVsKOs+ zQkdQ|9qw;AO{T_=qBdtA&~cu7snonE+NXAsxNXy?i+nb)mqnc6_oIJQ5ZTR0m{k*P z@jSqc0`}E>ReIzx!9f>intNym*{gX743CNlowQ3KVwnS3vnUg3PZ94>Jj5jN?%o-L z9Q+;f$BiaWnHS1daPnPlb%J>lb-6Kt`>w}#11rpMih~DUQ+mxj6L%Ac>DLk6_*`_9 zosNdi0QJNdAGP$*zmH_uw?3!v2AxB$91dn8vUEXH<_&pUW{$^?8ll5kaj@ficY3ks zNI>sLcJ6X9Tz*p<_n(*s7PltT$jZmWA#o=TtU8Xa&+NE;F$GjZU7x9yN@R^T-z7D> zc`ohXBAhr#VF%^YWD`J*u{5_)Y9oe2qiB_Zo8LG~p=}T0J$Bi&HPR$5(EKh^hiWE@E@5Q9VCv?`s%S7Bi zo_^K|W^YK;f>lG#@2EyJ*X}U#U~6PM0bF_;+_j0C10pXLku=tIYZxF z+XMeTeqeX3m*)4>b6?ppM6HJ^q3brRUlg*=T=#jhabkgx3WX_{_TqxtnZ9xt{ z9`%Pj1#MyflC_}nz(PpmHp06&S$d$}3+JyW!SjYQd2hKfN_VZLLHnfXq0m~K<`IUf zt6mFU_|?)sFN)ZNp&Ys>Lx_@2+ws7Z$#mzYGq7oX8zdDjfRFp{z-rM3ocoC9eEj&! z&NjP2Qxy3Iv%+}zY0*iqJD1{#LTmgzDv`gnJ->}U}9N9mrdJ>$_cHIYWbZ^JotfWa2kSz(I)USehq6d?mltn z@?nnE72-e4XWRKRaKF3>>+oqJX*?N+3q^T1k#Q*gmbwkz7r&9#bZud=wMK(PDKvGXdvEJV=f!pt7IKC`xX61l6J@(I$a_O?rKcqio2 zDQ?gh6i4p#rQxi6b8K0D29-bdQ(wMs72vRucUSU#-0Lr>!MVkg)L zzelWCK|8xB+M93}E`a;pB7Tk-1H-LRG)86^H%E37df+k|@GQ;8Vya>Hbo}-bzI>y_~#Z$|tVzH+UG_>2Xa<`Y`z=ta8|HqOR zJ&}Vu%`tp;_#ph9*GWwLl<>mvVY+)t1k9V=Kr${JqF+<5(Y0U4(AhRRSk?NSi0P!U ze{$t$?tn9eYkr2LE^E<2t@{$fRXNEbkHz^ zXKndJa!E1Bh{i%8&r+g&&g$gk)9ug|(}q$>>xij>8R}@H(%XJo)YD-X%yV8( zPskTDYhzu|^J*LpEl9z-)iI>%np!@jzb#hNVX`ix9opNI!FTUZE@Lu;YLU52}UISbdR z#;~pBd=^r92XvQ8(R}MsB)uvcQlE)Hj?XO+F&N8;N}7PML&Y?Tr^r$&;)QIeo?FppeN1F#k0`&l(2bk3JRqFi;JN`wXy zqh@I|CnDT);S_Q?tB5S0Ed$M3k}z-QcG%mzi#GrI$Jp|o)=z>{FtPbOnSA*K^p$6@ zB~l$U3T*Iaw+n_kxC$q@c!ATs#Y4297)4OShlojHBnDG4$|vyp;B8Z9Zp+-8;5s5C+`@p zd81dzkQ(Vjxx^J+D;6v7LnHlwq&e*B7C2Dj%qYb;JF*#*q87G z`$}e@W>N^I*nT0C*3Sg@^v$61xtp4OoJ*$fv_-n26rC3wgMiU|61%vV=0@Z)ex5_5 zxHu5PvV`D#i~+99)#LW$6tWkJC()U~Td>P^I<2;jB`!~nbHZ;OiKp2jw*Bf8LibcM zg-hSV@Kz<*G<1#VXv{*h#vI}@ZxY73eZ}m^)i}|#g&zAllXOTOC9e$jp}f;ce7anM z$j^VpHhpDC_?=idTy~9YIO+%&`&BrnFA1o)?iHCg#)MJFt(d!YDmPVhA}qKrhZRy4 zlwCFz)z=s@>vn45{ifC6`I~3wls=`~y%P8rtsg7<*PnMVnDJS?cIsvMld)LZ$S#vG zBJIhgBvRxW`RJd`mYyF&LghultfPx)_wOPW_D>jr_Y0Ar<>#ym#r&{pd*^Qr_46+6BI?(V*4!$|c zfwspK9Ee?liW$LpbClmDI?kjvi}Wx%C>jn~U1k2QlxAv^k1;(j^2x3vv&m8JF-=%p zL*qKGvXv2um_GLwo$5Um4cK~k;q1qpi8lb*(f_dL+;q6Jw17N#QcD`-o-_IU*}X$O z6!Zf4OxU=4Y-6k(E=*(KyRIVizsSN{pZAmXQ{@pe>!{KqN30S=(CAbNy!>l5Zre>b zXQek#=+q8};#cvcH968Dd5*rC9Y>{8PZ4v^eKdES7<7zTfnUnguwKRwp2R+(@-|m! zv-S-lBz%;8s==RC=OA0D)=Z@)mf=G$OX{J`=WPQ9=!}oE@$dHUWXoDYZrw}7J#)VE zT-ImgPE$2qdcy**pPbDz;sQBoK0o#Ed#C}6R5bgBy8+ZrtKSA=r*le)H8G% zz*!HJin)w^*Q|D$w|S?8!TOJr-^^QB*CNT#e%P1wvaxYN=8ok(f2!i;MC0sbhHQn>GC@yCq^C9 zO{?gYlg7|EHyA~C{>3Rf7H~h;Y@yS)CP78lOmt^`Lf@fX&P_$5k zJ~e$tgkK6^*i0N&SNy_sem1cE{X`VsRK{-o@R@c8UB%OWP1rG7Ey%R9BHY4ia{AM8 zdbrDl#?D>O)?fTgubtMSw>r)sgxTX;Nh9upSvGgi=^iNvu499f#?e5p9C*ILo!n4* zhJG)c*l7XQ;5km6EPJ_$zFwigJKD=x{mn1v4-0vdGIP8+2#lR#*f&sGaPaJ{l|>2?s(WS;SNlG z)oOjT$&s~wx)c|#E1>T#=)sIrK~VeYDGn@jgQ-J0M9J|j-X2-XG1eXkMep%!r5;|l(kDMZ49cOUjeQ}NWdwC`~$XF4krrQiY`fAEW72Y(j$ zqf$us)&FF-Hu5_k{`+m&vjwE0)!`fObo?L_Ll#TCVVc(^6CKwUGIbO0d{&4Ba~Gb!#jj7L&_K{5H{4QB)0^%NaOZM0F&H<>*v`tM zp>H$E{lP8ZSTV%@$#5X=N^$A=NM0WljGWp~V+WhJpGxviv&e>~? z^SI8xmaHeQVJk1O^3wI*kkt@|JalV)~$osm{43kAT zQ|%k{wKgJqZ8N!kzMtm34ky}GVK~p8;l52h$~6g@aPA|~rf^3HoiY|3k^$k$#W-7 zF0D6|IF?O<*Y0}cc$y5H+_)8@?m#0_gTs1DVTr!ILZgS#A|advs${o3~L(?r~W=> zslB4j+6$G;s{k3Cdexm2F5`Q>-o9kCe+tnoQRV*HZ^0X)?R4diyQJkt1*0}7f+7Dx zN!#qpMCI}w`Y@`6OHC}r&OJq>VQv=A(zpNt@vDi32Jcx?cEK$^x5-1LF7~W@ICXs- z3$w3Y=41p@xK#3j`HyE}#Uy`ZOkCYTzVj&6{G$#v+s+DV?*C&|W!ve}YpP%xD~8On z2lQl3J&jzEkGem#@O_gqm^9gP%lMo{uZuIcm+C^SZX5Z$Aq#T193|(==g{%bbzo;f z2Z&c*VK%Ej5(vGT!38ScVwdyJLcUv>%Tm+FP5kaBy}1R<{ez%4OWsCtAq!tlMZi|2 zW^$oD5}s{tA@YgwkoV8P=ELYI*n8r;RpF)sq<)tXL_CWk5pN&Sz2zb<5}(HX zyfu#dY2*MwF^556k2p@%Hbmcwi#R%%ggt+A@R?XHcIJda`q5Hy`H~~`xyF+5J1Nx) zzl=osAM^Jo5@ajD`;!>nEqhFqCn?jU`3)!-*@?~{{-No0d+=)Vplq`$1Z=VKfrh{G{)1fr{WbS3ub>=E<1hfVccUWWpiLpFMFN8*9*Dwo=eHO zL~`yW=sWtCoR$;js$RUK(^m8Ei4bj0R8I_Tyld%)=xX+fZ#-=8jDqLcVxXHpM9fF@ z@IRk%7_<8$4n*%nlZ{(3k|?7s&k`9E$maw?RKQzWgv2SEa_80$q56hSdTVMOHP<{z zeGVO^Ut**nY+nYj5?AnK$sqmO%3zI{J~m6%lEqIsB3&f}Q6mYM>~s!)wAW%q7{zD7 z#iUQS6XPB^k<^Dj*!RKX$^0Uo4Y5fS>dt(}+N_=2jX-^_bG;~bn|vYJx7Xpr@pibi zx|B{@EzP=CF2k9BRk3KLEcYef2}?c+k@8Emp*H_2-xTCWOsr%2(^o>87> z)PmWE)kwM^0_2jf&>pcv^pBb%+}M8t$(}*h;!7Rr&$B>Vo{9DH&rbaIViav(-sL~@ zUs?0=!{ARefLnf!y7d*KW`Qd0TDz1mr<<|0$c+b9O{>njicHB!NAWT z>hVz@%)uN#$Nz+&B3<|)KLn9{zHqbZ9Dc4k46jdrgJR=r#DV^1{u@&XGu&^}k1I}s z%C9N-gwM41S8n9lb}C@ob_%cDpUfrA7>5TFhN-EYkWHDfjdiy{9@Pl6Cm|2ZA$D&i zz0vZD+EQBcn0WZ!Ah)+=lko9* zxGP(dZVKK@>7&=o+LU`_shKJWc}j7y`yXKE5)JY|0l{VAEA+3t&OA;Sk9LZ@3#{T2 zw1~@7c|K2jYR7-%Xh9p@;+2X|E((*tGv5V93JRQ*)jBr*d>>O2xRu5RcQ7jsctcdW zI%dr_XF}F;)o-mwmnH>yWdPnVI$3V1VF^=267Jai8pu3bYEK=D`Zu7r4lFRns=KCk; zgeW8K;hLjR9;E`uu7<*jwE0wFb`Sk9VL!ZjIRT2roS`r12lKw4g`T=AWaMNtpR4-A z9KE>*jmrC&+k=y#Ax1#G>l(?!tle-i?<$cA(LjB9G5p?lgJou_qMmaZSt};WjN36m zaQ1K=5k4dZPD>uc{r*E#j`stm?Gd+O+H!H1cmY-aeIL!s_&e^vQGUmy2uim+Q2f9w zF5}%|uF6Y?>>Hei`X}dbVmyz_<$@Zj%Z@;A)d@QD%{gEfCG!1WFIt)UaWErbyarV_WL-H$kCWcqK(6}>->cjUzdfrVK`z;UN zD9dsa&tzfL++J35P=kcjsX-^XMi;%fe^pRy7E?&@Pf zF9j>FB!X>|1PmPsBenVFkk_k9)HfEB#})G-Ww9C)9;d?nQ;@SU&`PI~fn(t5vhDa^ zbSts5T0~`o`TOEd{@h#di{YMu%vd2=8t89Bdu<={Yi5|%C$ z=5|`2W3z(eVPcFizFB5W4Cqf9V;jod9sUo8#s!e7J5T5oM@7=%*32_cL)jXQD8_0w zaGLs}%&`tV?t^d^m!H9BlsueZj&vln{_3Z>!b$k%^&QOgXhV|Tja3KYL8Ng35i^-d z@Qpl*412=$vy=IKWH%Agy1+K2Eh58Pr{J4yPq0gLlv(v@Iph~7QOjA z=dS2*%ai=La+A|I(LbF^I?V(b+Y<6zDOliDpG+Rb9D!R3#~`w5KWU1$Ll+%>KG5S? zbGMOq5&SzxzL_{O>#y%1#=oq|!foyNJX911ZTPNmK_PjqsQ_b?=_btpIp7EO8N8h9HaY%&AJC+E~D|NxIvX`CcaEldkeMs$Jd}Sps zFM@r_KS^1W1X&jpe|)Y!aJeV(`lSWyrwCg=+poxl ze@mxzVF`Hc=WFtM@pckCM4@j$209mN!;=6-ke56UdR1@9(%O?G-EcJoO^k=}AH6vv zt$2FMGL4<_=MTBeIddsS4qR2=0&e=*9CB-QJux&pOTuQ9Ve#-u>&J50ux8*CRf`Ki z-*8DTUm=MoXP2SY+_f~OY8hIal&}lO9)pXzb@b8Pcr=h0z*(Ye$svVUR@rjUI%|9b zqZwMs{z-g5C6g!)^4vt3AJJH!(nw+l@6lHW?LbZ38*MwTle|-fOtjkqb|z^iFUyCR zxy%5w0lgAvQncVGmA*XN1ku3Wr_cFir~dN&L9yKMgneqfV;a&NuGI@^_l@u@DGfn z0@vrXHk4q=_%p=w^c3!GqdDiJHbk1uB~Vj#oQ=flm+bTQUo_C=2*gxO!_-nIV#{5i zi~lNe{@JtPq=y5epZkH`R-}TrCVimu-pxTtCsA}X@`Aa|!OWBBHdKzi4k;T{aOLF^ zSc}uaF+>H&hdp3end@NL_8#p^g}6@+LwNHWgu)6)8(DWyA0 zEYLpJhc?K{!P1&-OlPDGE`0C|ZMR2p(Gq6-%xnmsT?~Xv508<)Id934osvxT4{r$K zyV0>zo9ImI5*X_1A&K5Pv|-Lhx~&<><=vGiwWSO+DV}gzPX)xp@O!A2X)CDx6yM`=oOe>*(MN^23k`?Wc?q$*zH&X?A z2$j`zFec$Q>*F+weZ1=q9Vsy+J2i*NOv_pH&&9vQF8&QWm^Fc1`rAM@nh%qJ@8f9H z16AytqkWK>1r!@(B}yue0Jw>Pc-&LEoP-2 zM}mUf8S=KHgWdl`6S7@rp=x3_N~$<=ufI3JwU*Q3NpJhJ4T4*mzj zHOVwX&le&l^)ll3h8g*H@$lvKMSAPdWsG;9%nj@8!ne2Naj}6m)hp_uW}}AqQdySj zT+_zsH%>F5|3wj#TRw2YZkPn#_d%HeWw!OxR;qZI-<$lI3R}ka5vN5j$b@7k@;BrW z4T?QS4y`DJ)TYPO;L$5^IiQ4g=CNeU?^`teg*f{C(xJ+oeE0Q*Je~|%flmw%;fQG& z&vWK8r{f)vjI`jst^077(M|HuP6N01PbDff$6?OF0(zM5627~(khb>Lk;1?xYv;@a zJg7K=USCUT*8NZHsVOt*`z6y+Gt-o-j_+g2Z#A)zzXxzuAn!i${Nzo|Y(=~Y+DBH=84~NcSIijbSf@-)Ki1>D@gLjp%^4 zF`aymKwPbmI_j?G_9fL}>eAnsIeLd4;h#}`UktRzSTh|f)TzRzL*&la z3tZRkb9gLmJ>Fj(2g_v*z&B|cUXJ?-hRb;N#9?PH`^QnnWt{>0X1y|a3O#4)hxH-i zz;Sv~@c=HKzX7)#xWYF7)#BVIXb_G3GDr+cL*Kk9L|y$IKNFt>g?m=wAH#pF+qV6T zkLgs%JS9Sz!e3;keK>iOtjM`po}vPA2Y8w@1<(FWBz&$4?B(C%yS7;{>?ezZpIzuU z*-%W->c`R|W$t|DC&ti00Z#7I2Em3ly6>S1tlYsnl=Q!og@tSA0}_rG7F>Ws<4s_S z*fr7^u!wvXq|srgc6OI`2Q!6h!|BqK+0+DMSSHDUsoM>3=vhVOD&Mlh+Oe2tCqw_$ zuYl2>LT0Uu42it)g~~=YvLE6txhnNR^tQW9befD{clBC&GH*F1X&7Vsiwd&inG%L} z0W42Z167}DsCDe3$_}3R_XW??KO{xxR~FKCHi!Okn@N=YXMoeHWn}6iRo-E)Ovk01 zXV&bML6hB4#Qt#vxpZVU2x)Z*Ovc#3{0lsH`#8@%R?MeG>Y7k}BW?rhZ==;+|ry6P^{)+x^yhSdmu0r4aZRl9`jXYfu2eEZ!*mM6s;&UJuripZs z(5=-_7CjY;oU6&iYv+k)+j6dQWdq3b9F=!l8g0hU$wtq56R23;P5Q=Hka3C5;N>S~ zB)XHK0WsckEmgFfEy`9X-_>1u!ac)A&C|Bk|g2d8q;FHbP80TW^L z%nPxjYunZ}JCe;o^JkAbxKRPuTx z1)e7y0ye{k9&j527cz2S^|eZL{XGl)R_GwVV+S?SO9E2s4l~XNqrG+{tv$a54~ogd z6MZpo70!S&jykldC6;=ti2(O!n51iz!QbH3Ho;p%-@R0&G z+O6gezdD3Q%4Lkb?Im)mfZ|`@i6HaG${J`VZ7Kb1HTL&8+HOC>Mz5bl7oRzWjTiF4 z;Ng5+V&(u@#X0zipGAbL>_Jz<9sFD+7OQd-sZc@XeDft&m^Y8;KmN12@ ztgOJ<Zv1ITCTK=Lx$si3>%2*O z^Lyyts1b7gvj^WjA0UtJyrZ7k9R$?Ed7pGDzNxna_Z$Ou<%)CcpAi<;tu*7_U($v= zz8}4ECLwb&1?Z~M#x9c!p_-c0aIw;Sn0`pi`co0$cMD(ORTQ*I*^&4fo}sdj%IJSp zyU82LJZe8CjhZbeB%em^!I?3Qf}~*uh`lRJJNNXHw<*QoZg9?KmTob=eX0jiD!HIk z$g{G_ACuIw9wO7A2m$|YlBr@ZiEPehU{1{-DRZV#%~f4gjPEb?M`Y1gnfIu^beQ1G zjWv(s7U z6Jxm0D_41E%0pTzOmEJ>)KvOHa$G*UwS%Typ~~`J+t= ze>@QA9Gn6cZz>_{&tx3`|se#gW-%H6JFsb5uC=4w2&j z*Bx=XG-C%+KfFexxJWqF7fWug4x_SxiTGsyTE?PTf-Hz%Oq_GVSg|B6df56Fn9IxK z?6G28o?Id7U6!N+3YBDG#dcWJV~8bty4f>#!pYK&N>p7*3yyr>z&+CzfmLg#!%&qY zUf{VeokCXJgp{Ko%(*f_i<+4=g_%rDZV=mf?*@6_j`n3G1rV6PWnzK$4AlWE^q0|-0fgw6)zC33T3hf>RF|4Pnl%nB3fhl z4Bzj_fX13(JXG1q)i^H1X=*|k!#l2?<+rjv4dG-aZFv_62e?3fs{kWhwYFHt2HhGLvo`jg6l;BD}Pf ziLUNvT>XQ%c5h>{VEQX6cVIf*w)_{hd7^}Y%h%Jbfp%=t@8__Kjxd+w3ectPGrHJM zWG+T`vq0Z-4QiI?HRc#;`5nMHe%{A5Z}-6~%SBK$%Yu7f5P%0Pw5ZrWA-b?}G7P58 zp^NNdSh>+KCX(+dW)*LSZa(KexH}nx{IyZs?k;Ihod=>EpYzn=? zI(42UOC!J1nNy=_;q?=^aY7^AIbNL1m59d9tJAqv&&pBmXAb_?Y7X5z+pNH@4MpE- zgVImF!_q!Lk9pjo%F4EozbP7OHYnl7c?0xLZwRaRGoIxUQ^d5x2JUTLM}-dXeg{!I zXx&iF>JEP zMfEXgIIau(wk;s$$}_PlEr7JjjK_l^`rH!X#W-iRJEmHt@tn_4v@TA>W0uj_djBjJ z15WKRiGZ`I^2|haqydWEPsm}3$e$Pux+&-iq`UcUop!4svpbwNo{2A z&N&NbE*mpGcNf4L!^`NmOp!6Zp^L&*iZ)`6jnsW+0+2}}+~B=TXvy!VInH@h#>R$D zN^7HrfqBHTxQ+LYmgA98{#_jV0sh-l0ruM-k)c34j50`v_3Gy!D8Y%|x<|3{$TH9q zQU~wuv5=_4yY_OI;Fx_LwnUXaQ{i*+W5#il&`P^&B6Yn4oE24k2C17@A^F7S)Q|r0v=cMQStAOJ=zo?jB{>F<~U`y*|WU$|L5w5*TRF&u)JlS98sYW6eTav9F+>)$(v*jo4y%uyc@B zHt~LokFKbGy@LtJngBMl#*^*gTc~^PE%@%j=UZ+T;D4FDbf6&_Th6QDZMj(Z<++zE z8@)@}n`We<)}8*)(M9q3@k&{-^y^C@!aWhE ziuKUj``bx{e+EuZx5V+MKQdwbXMRkC&~v+DsK~c;@EcwUIchb`@9kwUHSr0xEIB|z!gsNS zr;j0b;yB*TUd}c3R$!J{n?UJzD_k&Lj&sM3fdvoW*w|P(Npy+!d74VhO z)xV4XRaCOm4&5X@+6T+ET&MAVna-;#=+v=Wb$!WCpkrp;nrd$`YE}c>UaCox}j7&kgLNj zcsIyeSxRC+&wi8`DM8^?UbsN@4!M5J8#H1KQT0-j)%yA3sNSPS6B~Ts*0(7bk+=xH zpZY)!Ulqd3yw~eT{$+N;Vj)cAd-DSmbl`c{N)r5Z2X{VAk}XJ8hnBGIQ1k0D4L+j= zPk())2O3I1lIKa#_9$jk_X{HXtBwY_ULx~~Gi&5Rd5-nn6;SIjiIR*(ux!mRQ{wDM z^cE(=6{jH5COnl}x7e5GJ25!8dJG;i8(<_KNO4C;wYfz8;k=zB3{FGcY}`FJ-2Cku z<|o%PGxmg`zut9LyCMa$$GE`KtkukfPmb5yI_AN1lN^*fcaC}+H=(gXCNySwK#IJ$ z&3DTxk~;JbiduP=;A{zMF|UCn9^OaAUR|aSeyQR5)?UWGJdv8+3`etYHxeZxM{jJ1 zCVCBk+E@$ME8M{_UX=WtluVLb!|7*j-eIqnM-(<$lbXi?WRm$Uw(*=85tEXJZ&x;> zdT|{Qn=nXLMd{nLzkWr3ni|nT^HTCp@Q+mr-cFA7)?sW+i1qBD05VKv;ZDdZ7<2m` z{jR?U4BpLykkDi>`tLQH9&(*YcNQlC8&%}fk<@t5f)r+I!0*kgnAq=bw5%W(I;sUI z(J_g0UHg+wFwo)MM0>GkV2H-mCkuKx4fJXF$UeSqU9%^a_l3VNg|4*;WbLUCc))$a#K`u%w?_XkMlt&YmX{zrQkOCj}&HOAP5VIfvP585qC(IzD!O!K4JO zg~p19d{;sj^@Ar7OJ@n29^4AWywfb;_Bj%&H^RJGW=IrFtI7LtPq<$?9Tq1d?Uhvo z!L=5Wv9bog@_AQHAw>+&dwYFoGLi{}NM19cSa9|)Q#Epesef5WOhSLqNlpjQe0dzE zSDHGtNDAKuB;&>`t$p!@*H#sTFFfr^yJM9 zV{prH9WXq1S@0)UnmiF6A!fU!D8FGPZ_4fwPw9*7KciCE>U#pCKHMa`(!@#Y7;_BL zND3Z! zX-ja25E0$;o?*OiGhb!3u)EdFad&i^*^x+L@84tH%o^KIkBp zPy98rz};9BLUM)CO`r$wj{Ie_v|b8!-P@1tE2d)6xj{NHH5|w*bpUg0YOx z7@VB)m1+k(U>CO*(G7pANk)t&7^wp8+7>|P{1zdf`R~vtcEHlYR~iD2zhv(QpJ8wJ z%^^O0j@)deSSBEN5#DxIq=((sVb&K_?zP)~d?+|;CA`xewt5E8;#8Kb*YPI{c7J7{ zCk4LD$Ry!=SqKZfLPI{Rrqd55zz4RJh%Ny-_j?GY#tPAOD2s+~x?rul5|&o$Lw#rr zn^9p#mtWgT53QYSGh1vOZPJh-v$u)h*ySRmen}8!u5ty{40YJ}*#r{~y(hA~VvRT z3T9CPI(*K1&*Zk*jG#M1Kz#&MjsO21z#K7r}h^ zou>euLn4gH?GQ}n-*a|c1nFqG4(pbS(j~zsnN$l?EZDxE#Ca4$L3h8Q+PJQw>o)RLdll+0s0IH|m~WgZ_PwZf~eIdW29SVlY4_OAo=Z$aWIpr9pp}A@f6M z5O3@{EeOqd#B)8R!K$;BEJ^l6m3wpGSIR{oW&P~dl#gg9Zb_xhj8M!^CoPu;`}H~nFZTAFc)zndMnB?B9l$H3v93aYR%0ip~Eq-JW-;eGpweuD@e z@UbNJgE649av!(j*J6xJ)Z?l;_*u}SPV(777uL<}V!yqRfD>!x!WTtvJnOy+)!$^| z=<`Ml__YxAH;H52b$)KSF@yBkPNZGmz1ZO8b+{{6nv0H*gU{Lzh(dfbvvuZT)IP*; z(La(gj^mjO-H*waA$4r%8b>UROQ=KV14h-W9K1{7se`H%7}#pU+P#wG#Lfp~`XLqK zaH*OJKPyM>d>%vZ&C4ccs&|-|%ioe4+0`^m!3(8cyYU&Dc$$%!!uYkt+-AAoxc4o^(k~GhYB!nn zJbN2oFFnnz*`$u$!};hfX9gSYuLPxQQAmszai%|&xY9aH%u9`d4((d(oaaUEy}E=u zGB)GrIWy)+%V*NB@*l1-O-KJRQe5a@G2PKziyQPZscz;J{8l`V%-wqoJvTgMl!mft zV5bK(r6<9|GjB;?)pfdN+ztG3t{sbghViD33wNtu6|};@knOQ$33ZJwL(uPXjRf zv>IJ>sFy!(FXH4oRME92o<411>AVHUVTEcmNk4OnJus1@HqXwHuTS>E^mb>QbClo1 zPrHgQRGR22c_~;Ol!>dBsbJ$gAIQ3?1FlWV+?1zsv>`;A^Ndym5U(bWZ*K?fR53V3 zmO}mSQkc-4fIr9f60fZsk@r!~|O5mFDRQ*7CU|BDI#-2Oq@;o0HhMpg25SoQBKa2Eh88b+Cz@4Eg;0$l`#2TtD0c zpU7;`IzNjHzdlIU9eqXEzwuZ$E{|&JR*`#ifgTIvJM`c8vGV**A#LJK%Jyb33Wbp% zdPE;?FIR?X>#q{4W$RE`Y$Mazq(@)*>*Mm=D4e|g8RQwXP{r%Qm@g888u!jydo@`S z6MV_o{467TPJgA!e>*z}H}{@DfgLaiQ_MMo?>oGW~CT8f8xXqvonI%;>^2jN18` z`WNlRKBf0;?GYttzM~9V4X@Izk6byuWx!=F`45-y-R!-+A!NVB64?La0dsBrHJ(p@ znie0)V@70~ng4heL^{7yu$q@a>X!9UbQxFGNj)2E%W&HD<(xJ zRj|-vI$YCzOV^#vha?kWo0I8A+~&^>;9Wn6_df@tip@@3^wWgVXT#9`mLYUJb);{< zJSETOU1F{tltb0m)9Kyyoq~^l{b>CcBY2lbXyeB#_#mwk`n^t&v8&Z#s!}kMQ)on` zKd*vyUuL4R>OpFy^#f}}!~`Sj7Lkbu^VmaA6H!)Z5At3d{Psf*wLdo!kC}6b>DYD< zmk^>WvI1e8nT2)P;#%gd)@gKewneAy_V^=yA0}?QNRF>4B;#)xVZ~d%vu3A84ly_B zfryE4srzj8>0I8It)N9DD<*@DViXE^R^Nh#o8;gVKhSwQ%v!t5A>B``VL0s(p6}a( z@)=12iS42^Il2bFe-8mQBEc=5Qw-<)_ws$Kqtw3J6n57|QSF`1XdzjH+Dqj)BYr;B zge#TBhjZ1@H`E95xG3jn*~L@_yW@z*Lvm&7L$*dX8jNQ9 zL2+ymvCS$be=U@;tGm8t;`>g_EV+t<`KM5#!k5ebIR)2q*U4j{FxF3~hkjdkorLws zQ0b_Tkb+p66U6r z^LIcKa~Rlp5HEk3!r2c>a)R;ZT%Oew&QC1~*KMNQ(rM9T_oeSv2k$=N-tRV}|D3nu z78C;89uPq@~rJA&9wXjQKkK?b0p-}S`@Skix z$+MA%x|RHyzO0f&=&UrX5Lw@?gBDvm~bmjV#CSVe2ZKCp8i?ZA|P+x)(f z_u&-}l4u(PQ1DvD*a>$)-R>E1YEK+ap6^7DzJ5ppic0Wc#V&4b@_IDudq|ZVHE~~; zA@8lY4NuFba6kBNX@3qykt!G3`>vCQI9wqCCnv-46`$$SqI%$}J_syTcaf#La`5KA zKsY--2~D#f;G_C7B<-ijpPjp@*GV<1<9-)>qi@j;^B#Oohljf30)RhQlr0BKq+m3pQ_; zad+ge;?I*C*|=nR(z-mIm`bQ&!npHvAomsJPQ4|LJLCEJtuJjjdJOJ+)(~;E6A;l< z#*p3ywz^Q2=f?Alem6_9xM?Dc4LQu&hs0xY`)L|*Yas{54gB1RrQNCVpxV6~Hm2>x zFXsx_Fc|~3F1?OFYnRdouF9awv+L}CDM0wkF=WI=muEvNQ~UTW^xB8rnBuz!cQwqY ziE;W!!k?{zeLdeuXm1ng&sXDSdgh_Atr|Xx48qIfIrIqE| zIJf>D>N6o2HwSOWA8MoI;0HHYbha6H1|&gILlhcc>m#uPj_BY&jE)5kWPa{0TzB4r zvzwm`QeS3~7QQpLS$`}i{yCBH&pbvF>#U$`r#e^&zoE11DskRvN3v!s11}uInI)J* z?oK}lE#r2uC;0iPu@s+gPFLa{N~>a<{&DW_8cUcjyBgQ@^V`X7ra1fI4l*%c1Vj|l zhCL5u|f?A)3mz(lv8qxp6yd zaNA-no4m>>IOE*GFb~F%;iKn>r@{xK6Z>6oQeeoqXJx_re_J6VFG8?IHk~Zv{RFMX zdtnXF^Y}6=8)yH$LM*J6z(nREt{SZ-lcH|3Uem)s<90slGd3T0|Ip)7vfI&jo|n`QMNDn0{l5FN)0=0NOidxR#)XPQ&#(e$_5p1kX1(2 z>|%6(;KQvsw;RuoSHdIh#<=-%Ivn?BxbHEQWNzmR)LY(1sliWt?;wsE;!c?0Gz*6A zYH~plu2`cv9gAK)LZ?*)7*FlcO??QyyfdWwCv{2CzcNNS|2TAb+0btV5%7M;9_HQU z5UMr$g|3k0{Vb|Y?Az}O@L`KNSTq}9+@5>5nLlHCRK}9Sk!q;0BOJB-Rq4l8UASy9 zorD@F(%}`6bd6LS+kS65v*Lvg?)xNQZ!07~#myL~5KR>Hjei24+9u-C>Uw&j&JhJ; z-;-ube(#c{z=?PSg6cay7ZNuB+x8o9hV!+#4~|zcCKAs24uhzma-#?lpev|!r z<}ceOq)lxGuCR~Id&#&7ZlvyAB?CVmQJI1s^t7K!6@=DU2RgHir%WpQZlw!--Ic^W zag9aqX@A&&C0}vQ;|TtbOrV?jPM$Db%(7imXwQdzynA;I%u}(Y>z*`Wk#i?I+I$=? z%|C(1H{W4Ty&J`u9cLNOauNClBUy*qM%MM?61q;LiO4Q8z#EfF&`in&PKWb7v#o1! z?-?uD{!o-F|Epc9}GdWw;r60%^=Uc&eCn46{ym#J4EHb3;3141BXew!0q|v zRPR+d%~ENH6)S#__1%x?E6GyWek=l|#{DJi2+zmo+qV@ek>|#{WHA7W}Bu4jn5a~XHG}6$3^mO+?bkTKu^6oX&f6_xk!o|ou z+qFO=Qi%gGfT5x?^4MAw{j<;fkE8RB=jwg`xIGJHCwpflBjeoHJ1sOc6h);?dl!{5 zGRjIqRw6_qimY>ArzDl6$S7^1l%zft?eF>h{`J@U!P`0aeZ5}S^GW2a8Deidh7K>5 z!mc;Q^w^JXlvw2p3-<&uc|kY$uJBVj-YW(ziGburDZ}rY)sUMs2P~~Gg6l6Ga*OAv z1>HSIw0uSA?7j%Tqr8A=*vrq1FHeMD!@W>?asw1Zi{Lq)6}3|RGjmh*^CeNX7t}27R?K5^*v6U30?7}O{j^fsR z6L56zC%%9855;PB;L{}?bjSP>a&15uTxt&Ej^T-D-T0OGx4&dRtG40k8Q=wM@B z4J?oz*p2bEyRbv|uM7Uxt1s>G+y zicP=vo|Mikqwja*vsHm2+=HM#{B}72zfG}%PE{FznUP2vE@7I69J%z)fS=(GW8Y07 zs0%p3zdK|gX`u#Ul^<;6^XO}vc5@B6#oPtM*<8$#S}GBv58gV{F{ajtd^vZC%$>ZE zc&Mk7E5@(rsG%WjRTh)p*Q#V`dm{<{5lv>_+`^1H=aO^FT4<={Akrgexy)^U zP;sX&w@bkd?$3SaEH|Y99xP&LyWt=u~BUWh+0`frIR$SYMqAdw=>CkZXJ2^L`skt6h=Pgd4S6_f0%Bwp1hx>0a^K) z(CD=qZhGa@Lp?rh^u+~q#QrdOyfXoG54AIcE0Js(;P>-x1>kGTg`nJigG~4`3j*di z!?^{oaLc%h(4n--IU)597VfHrk`I}zMN>0YAN$EBr`qvn1V!ejW&)(m)n?y%hp>em z=P^oWf}mb$Jg#Q0Q&rv{(AV*Twc6!N)@j`#Z|0ZNvAssn?_J9t|MVXi*_5H<1vljV z9$ev;PSSO^m1=hw!f@$c5?Xc@j{KsuYx!#$Jx&oqb2p=5w*`k*OHga~O?dZi3JiEz zg4I~U1w3u)im*HnM^L!MEk zf~|ZOem`L*S3EMF8~j~IyT=uSRBAY>SY`n3lM?8fS+AYFq;Ha_MhED9ZGx9R=;B7b zY{vHCN7hZggP(_cbDMq;8nw<9pSkNKOmp}unG(0_YgW6ZU zAUR2=h}r?5>d!v1lO@&Jw;~gn^y&IM`~D<0q>Q7z8yLoME=Sy7HIdL1KI=YQ2&F@} zNb1Hix_=zu?ls*&bB$QOYd(*g6MB&bWqf0wk0il=t1|J>!+liY&ujF*Wk}rWpR+R> ztC_sZt<>Eyy)J6`1c7jMHkBFcK(MQUFy6YbYswC=e_2f$&GjH9^$;XDMbP(+vYh@; zUw)re8LmoBfK~aW)Y0J??A=P~jv$dby_4odJ*k$>Go}>CnY3WMc6S}zoz{XrB2D;Y zya?B$yqX;CmShg^P6hL>iE#DxIl?L%&;dTX{~kYx)=E4#$bTn_^FJT@a)WMh>>wLD zchwy>f9(9`=t@++wvW+tYD8D35u7R=jmwuU!*-2arZ>W@PKEEJ21v`2lT9h)n)iK- zYCb}0$QaVpEQZFsxA>dfQcxTKRBNrKzgxD}tz3-oU`8`D=fg3a^h}adPSN6b*d*~z zT{rVm*^`8=u%+YQyMc$8HJs=4$<1e1v2*%XR55E{->5I6^B2A(ePb>%4!^>oCV3@& zaO4m3&LE#K{BBCX1O=3P?2GCLjYv&MFx!580zUg94nn2+G-!7i-kxknjpO2=_|IuL zQa2OS23OU&e(s_>o~|HGBfQW0ax!y7!OqDkT@k#kH9)L23#W@mqT6T?XHslQaudGO z#IzAwj>GKrN4gj_VG2#QSEh$Qig1S4WVq{Y+p)@EkSZ+UckuZ$@XSq<@s+I&Q=Fp# zHxwqo8MBihqWy{W|CPmj&5b2%R^>DQ5?@n|Hr`cz^Au*^Ga#?-ye0>V#Hg*h6-p2D z9q=@1BHJFul=HLJLFZcXt)>R&kF3T0PRq#6{A&79Q31Yrs$z&z7q#)3i+kynUQWm+6eMOL(?!K$06;2D;#{@kM9H%usQPr zD{4?kZB(!pB2Pl4Up+OnwWYSz%g`XL7|+fdgX1d&c=+-XxU*#^y5$$}u3`o}uByQO zC=bD&qZ7!XWK*hRy@cDUX^!hXhw%ICQ#==NA2IfMPxZI{B;HS)>GrQ;u!T;BDt=Ec zxH=Q$B2u7lr3$^P9{@aT26R+Xp=GZhIPRH7@|Ir#*>7V&y`~u?&yT>-n3FUo|0V=R zRrH@u%wB>bv-s;Zo-SD$+)b|}cd|Ls}nmS03shnU^u?}3F zc!Ss!20*R2JefF#&!8nA!(Xq3aA035^nUJPwWvO0d{YjyDi~6CMjPFlr36FtGU)S?LR zTh2sGlcD88cGS>Wm+9>g0h?16pitnBxyRqr?WTKZJR5`Lm@Xs~`@*;3|5)iiO|WP4 z1EMmP_fNQw7wmf7NBlyT!njgL6#U-r{Ckr*mRRnAkW5*+ylRL!wtI+rR?Oiug*GPO z_I(_0dJK&DS%&&mc~FyoESa(MJ5dw65|j{c%Co`UJ0M(-Rhxi2Tf%@#mpBBbT5W)IzmwU z_cA=1#pg^xA%bw5NHD)=fif0JIN_=zIcT+#E;fFNs?#Fj{t(q~3Ra3UVCO0YfzEDKLHlnbS~t+av|JXT`VUFalyCvB?ceZ+ zW&++%9$=4#Zh+N#v9(}t1|F@Byqm=qGfE6u{cwJ_p{k6&4ShyM21_AdSQ9mEGU<1- zbUcz4;heGO9LbMAi)!v~Sfji|lrvg}6Fr5fi`kF5&x=aYJKYTSy;uPqmWy!Lw9`~5 zCZ8-5A7xXYtmpeAlQ4WyG)6obkB>zoVT0dnnp}E@#`h#syY;HzX||nC^DRdWmuCL& za-B+!`-KUoW9aB70sK8D2eqU3;K_7PR2tfV^<$dZuBfTZjxUW^a`+;u-aARBEU%`) zyVYT5>}L8z;Urv{pG3ya$|G~K!eOpV1=QSF1DmbigVRAVoM#_JZU?==gH4(c>#&6G z3ZDr#bi8nCNe1i0_n!xUt75+8i8|{qvCRD1ix?3pMpCL9FlSW?22OKzmZ%ZKsjhC2 zc-IDw`YZz3qN7aiA}##bd6Wrux`gn_5Z4rG;S{s`RLmh#oqckar7fRJod1*M?`?nWfi-s^{m#d+TY3wAowc7$ z-S(VVicEk5ZM*~h?0VU<_SKkk(gsI@B#y16{eTb+n|wgqq=yi4c6zeAj~Y7a)9 z-@>gry&j)_O=c$f&q9Mu<)kI!1O%#Ozyc>REPC2U=epnIJ1w{AiNge=Nm$2$gY!x@c>P`02NezVqG<7xvukv)!k zylgf0E-S&g$3;06lh5?ubSW<2lQtK@e#Tt~9+8vX6S(&`tMSJ=0q6S9ky~}{HcM{R z!OTn-j5}Qe-;ZpD?&mLIUG@StQ>c^Q4O_=W*o;!gm{aUY-c?iRq7PFQ-xAkbHthOp z1(a6X!woWPx#k{KZd~$n-XEXG$^FivL8t&(8}hm8C)V7V@DW_dpWFRJCqZ}RVkYR7 zBb=^}f&+p?WPEclIzg7-(~uENS`|Ybc4$FnnK{o?)BA1X{M{S<9)`tg@-Z18}GpVodMUYI8<9ZoWZuLbwm?^gx zXZgh9$qrF2LeUC!54m$CsKhB-$--FKwcN{Z_p$kM5E1gd0C7<=ybHAi+QR#ZbD9LF zbhLpvHC3Iw&DY>+*Q%q6)iu~5z7ue&BsRWOr@14)$hfEjIJ|f^%8&EH+dkDO_aFp= zwojm~XM^!TMj136t0mvJufX2t8Q84Dv-s#EcHuuQ-l;kthLvV>k&j%sIPX=Q`tJzb zY9_)hRvN?69bTyKI2L4T#Gv8sOL(gO6O;=xphW&EF}^l~yJ4Hi&F^ubZJJFuX5wGk zyKxoVbU6rt{eNh(Uj?a}JV+DB3@%>fJ|0j_W&afH=aNcK;7h$ec(~~zuw`R$kayfj zX^iDA?psW^K3B!jgDSL1{u*olt_U8eRpZR13>Nn5aP4mkcqV2q-R9MbzQ!S3yVMFS zT-Qr3$v(%u`n-3Y&)M#Y9O3-8YoTZd#oK%z{ED5Vpl!)Urc6B(f1PQ^3)^J5wtYri zg3A+}K06ebpMS#$MZX|w>tc8wRx5^!ZH44M16rh42~8(wLup_V?cP5R?nvb!J3E=R z3)6);axTy__5yios|=UVXJLKHSE`uj3bUE=?R+Nh zEsERZ#cBT4M%WYE&(6QQknEiE9g??PqMFa;apaggIcy<=D%!rdsw$o_YpbHi^*b?O zvxwl3Sv{5NxrM7G7l9i+2|`A)ut`W0yFa(^JqZ=TkD7K^-keQNM-=eA1z%R^VI5Pu zT)_IB2*JL-eEMnF2TjgB2P=tuHqhk?I<|-k{N$A}bb1XPdq4m_25PLBbQ!RBjsdm- z^sG;Ts{`fe;*kppW4BSEmmz}HeoHW<(-lj?p0jfq2|D)tO%iNK;jl(CDEb^E`_=Xe zx7k{YEmo- zycLV-MJwxcR+`dQ?PK_Oehyu?QyNW&6`)b?44FSx5?{Yh!f$=n7)L|!#o^b~dszWF zWt|DXec!MzjJ}eDAX9R3dkxbaR)C962jZ*T>$G`7F`XqQLSC4O3$)i}LgmC9qA`6D z^zVJje7$1JnQfVZKZUZH;ejD^Z89W_jdx;rTqCCUEynjZL$GX?51ZWelWcnBN~Jcp z!YPx-b%$rKWVP=2usL2yv^ZM~91Fc5Lj4`Q3hH5ccn?C2pFZk{G}1B2IhcH-1XMR9 z(!UETapfUlZc>Rp72BYNi_*Tb&!R&>x&A&*iqmph)_$4HnwkK`L3deDRwB)!!DOF! z4VJv!gw>7BxYSKWu*pY8;Fp>PMbA|PwP(*TKfJF{=gJ%u{+xz#r?aqSO*AAw6N3?3 zDOxgDmaf_xM&48^z}PXWuyU$0dOiC}DyHb*Hi;VgmwZ6?H~s97r53n8bA%>pjf1U0 ze7~J%P*`qI07u_+D%O3)d9ht7h#ub!Gn7+UOVg7`+CDJyAI5W5;%A_^i)9O@e!-zL zQk;##LK51=cM5;b#T5lPc(83Bu9%caCyT1rnX0|4Yp)N7iY)~0*?7ZQpKj=%u!elv z_?L{%DS{lkcG@(=;pTvkux_3ZF{oWf2RcUK&Uv1J^S1#cb;oeJ8yz6!-$fG11mU{5 z?x3^O8ZK{$CgMFJWY_T$%p2DQUCRS#wlkokOgv03Z(_6twb5{jEu2`?K#x5lkaM)3 zOgtOFdM~b~d9!**`{*c1So4C;Z?vMSjYG^hXBFqat@TXW`SDy*RST)JkB4GcIW$`f z>?D(4xMbc;9zQjXE0CARt847Zg^i7r`0pjN3p&7p`40Kke_>?n5;#8HmMR-HLqORB zSY;gxy$5B{XRHXy{b79g>4EhQ_o^hxCY)mSn~N+6}X2>ei(;ABo zRvB#5n#=5=fw{QP|1*u{zb6+)XQAcYo$U4yPcAmgnFLjhg%sBc`l+;rH5fR~`z6FV zSN$}y*kpv5Un^oA*U!LG?PbvFBMDg*9Md4&OyWa6)2W+zExrF5aw|g*y1SH7Rx6Z9 z9%Jd_3=1H0&Xe5TJ4kj+89g+%2D2PaLEOX|n*Gy^w1%=&++~0qnDH2Fi{fBO@p4wi zYc6HKx3R9qvShYJ7r7P%P`F{oR9ffo2=tcby8Ad>=S|WHPOv zah*gRO`!)bmcYTJYz(AQoS(frP=7zXs3k#V;=Zx{u1MznxXfx47;yOuuAwh3$D-2) zSk~G|*NwkOzNF?eo95?|z@K{XE4&2PXs2PlQVcqHUqz`I5_n?%8X{V%1SPgsH`dQx0ucs-TdwuZcyn}?;lfMNC?qaP}J@pu={ z;NR*^cb|Sj=bn5{+zYMffsb?G{I_JL{E9m_^z#$!bsEDn9RkSj?(byHTHbpQc$_Gi zXMmelHZ|-!4gzIS`Y~7>Nk|I`Q@dM9vo_P6MkZR^m^-Jbd6GztDj z$&u9~CE$NMiFVXVP+8sU?5lTa%ul>U0~*xPx?7nr_hsQ^Q-f1=e>=JDR?FIMU&Sfi zcVT*iUJ*})Rs1trj%+B)Bl=Miv}`DY9BR2rtKVqbG zZGYZ??T2Haxv>;F#EXdesg-bVi4@Zqv=)Nq>C*R(cDTBoXN3$lGDk&rk~@)ysHDpR zY!F|I8-LB=bO-li(n~GOE0h#8)Krsc-NSU(&Z)54dL_9ZxCRm;I#5nk9XDqndBn5e zMXe3FXU;nO?`0l@rc5EC1$L;Mql~S$Rx*v2{bWh~T%MI5NyWAe5RW7F0`H}!#BNq6 zG)0->hIZbZ=`e|GFU(;7t>D?+Y&<%H0Kh=@epH@P)a129NND(;EqHzVOOO&ioCal4}YDg*_*%kC$Afn zB`xr*vk0{QxJ;FkSL0v_uo|ayxLbzTNWjiiTD*BW<_B)T;7=9AVpbP5I4z4}^i71#@B147hPFlF>|)hi31ecrZ^n~h`ofd2@rnhB(6Ynys14wIHG|afGuW%%PkB~{JL2{UaJb%+D*e^Mn~@E; zTIe3SelX#b@9=k_z20PC#urq*Uq*g3%c5Whz{R#TFb{5Fe>LCXS6Rok>iTj$_P>~} zwdF)h$se6=c+#^o{t%U#R638!(R(2~aD{&aefKN?2X?F>fycw>?2~KZjNxbAClHA# zyB?y)f3!x_cQ-n{jIBHIEe3LOR-nP;P7=3SiRtU+9hE_p{PC9L3a(A(rr$Y-0^53A zJ6S8P2(i^?~&Z6-CVDmHTW%vBPAVMxqTA+P9}E(`v*n2oInfc5H4n; znnZb5nkF~@u`(`J7shoi37kibFa|nC5YBoLb-nir%zeXgX2)dQv#X2NUH?#<6%-Ae zoj++=mJjdtd2uSap6Gl#hTL~v1XJG?k;4<}NMx%5xnQ5`Y`A$jx^Fdsc}0&%kB2>| zPErMjp$Ix3T)2HxglH>2GZ-6l1b*aSCtH~is;gWHe`1AD{dPP(5)Pcxs$O!nLk!C$ zPQd-35O$Z18G8P`O%k4&u}a}C(A~n3$rmQ@U0olTQ|JLL!8*|OUpP^c4+EbqM@hzb z2E11>WOcx2H2N5UVwcLvTMZR7c-e&7$>B(&SAy&E5^^M7Ll8L9f%%PbB;te__vO`- zy0$_yH1pZSY_txbA}L~c^Is3Xj7}wQdA2M`_vOls?vm_#S4d4+9o1B~!>~V7X=dJR z82xkwAL>14ti{r?_xo7LpL>Vi)2M>=jmOZ^PZQ-bq&VY~YB1%J6r2#9h7&to@bhAA zEEtr6Rh~Dg$(UU7ZdnQxec@jlqLOr{jUHF>@iB8k;7w)>K0t*@f5^KaKN25jk6s?H zNZMUb82dGn7Sy}on~v+mb9-amCAS}p;=ZYDY`Ze1f4GIsZ%#A2zmJFDV`DhSv?v<> z_9b(my29BobRS4QYJ`^`ognik&&+qYN1y!U9W(z~pryYJbz0p-&IHUvk;R4NMyE1$ zsL+IjKOd>M;~sLQcQR~El0luMKlF7{EGtxW8?y{lL1~EtsY?6@w`QoK+c$kWYsM7Z zGe-}@jer(693#u@5;(EHnHXfCaD8p&K6nipRBdxRcAs9&DRo|_<)>?C0?+o1byEhJ z4rf7D@g@i#?59V4Z}ZMATMVi6p^JWQB<0Iv;PT^pJYO`4Zrya0MjX~-_0QYDg!ki7 zZ9x#WSd^es!wsS(F^%RsG*QvQ{j?$`mXxG>Va?$luE8XPn{(k1)ky1bzMPW}y;dK| z2A9d0!JmV|T3q0DYbR~8$-~b4+vw-e0{;x>1M+)ev8SrZ>V(sDc-?7u*<{T2UT@-= z{i>|wW);pPB@a?}dU9EpV_-C}lvH-=!;OM)RDL@iTGoHyxusw5py@j@cj*g!p?4Wi zj#-K3Mv_=pJ`X2qEW^Rd9@e(H#;fq#-Mti9ey_O8`{}IqHdIpn%gG zS4yh)f1&I7E^hkX-(*YhQs@oxN4u&LBK7by>6Jc9{(I{MnU8F7*!m+XE|lgr2Sw77 zoZ0!QY-y~Pqglp=}$E%yZV%F`6oY?_=x~b2OnP0P)(yef5wkiZ%F5q`M zHDQp?k;W+>g4bR5iG}k8l2hhOE0^@6uB8*(AeRF%Id0(nUmDS?oeG;wr;`1C8%WnW zXGV6w5polLFeTn5_|M&w@6AYK#1aVx`V8UBLOVu%&@KS zEcObhCo2nj=4j!xfjIj1Den!}l!js7McA@@kO}cSMxG{ppu4h*us!E0-R0IpDnk45 zaK(LS(tJj41gnzQ{I0>do!)fP<4Guc^eXNJJncZ>)8KQxnVa?Cgb*cu@?VEjRl_^=&+TWCr*8-8L@KE*?6HR)DK}E(H1T z&(=Nf(PmK!&Sf^hS*t6s$$5+*^GyYF;~gDkpLc5q6jf6>EC zktVdA0DIegFsr?dB&06EN`BA!+XH2I>M2Rqs7&O--*=&CF~y^olUZ|356bG8a0Pe2 z(4@J)Sd$5vaN-g_>z|iS^ylvjx z@CCix(M31=cC)j*PeH5kb(%P41?QE(bJ~(~alGqeG$@mRzY<1hsZC&(q63*>dz0F* zTXC_^Eyne74-rl|$aN)k(vZUGFhzA1#A{Cni!WZ_upyRuvID3#$pPym4{ED8U7 zhQ1O~6eKPAf!skcqto%uq6LCO zz9HPv82)_zN0`ZN{YNL|xKr16BXq&eP&7>+rNdG`sOr*EVwSE=Z}X0y-I9Ok#eH|! zw!1p8e_u9r{VIj&hCV@my)4G%C(rOa@1yVP|JE+KP!%ld49lsILR( zP2bUu9ikA+e|GsFnauk63S6VQUER4cZ>UFVD5t1b#k9WqSGVX=D4NWFjO)Wr)P+vz zg{~eK;%QY5OS62K1u}POX{jn1@iMd)xB{>!Ax?Zq(5`zmxZ@Rj* znW*m8K>PL=q@v0k-uK&byHg+Ft{mRCWFSpm{?wzF*3}Y&u>&|HY>n-HrkthDG48*; z-`E>Tmr4JQ5vG6KBwR3u0gk_i*H>GE;nMM_BWj8*4v%1V#vFDf@4NGAc*K11)qz(A zTJS#VC^_#Z4IisaaP^*F*nII4?zYNjowM@D>&oBG>81MM`zZx17K|m96Vy1X*Dm<8 z>~`Hexdynz^BsqNM^XJ|A;Iy6Zd|ZOA3YlLcz&ZH{9DOCQ-)Nz?(JG6`(rgZ(f&qjP zv+HnDG!SpOhm2UI7w8NMXxgkqEZDLdBA1Flz{0zz z@Q<)T&!fw5PSXXL`!5h9SKpx_JDq^(pF_{TIgcCO$w7gk9o_P}pKv`ZAbs@_Y^#5O zWhd97TjLy5+OruZ1)hNTvM$!6q@1Q*ROB4PUqH^96L3dYA2Wnk<4NaM+UJl;|Fl(* z_Y3tv*yA!hc9Ik*%zjKQ_T-_{%^niSJ0BIETH&)Ji=ixU4HkduqhD;g>n7|!4j#w9 zk*Q7NVAeHp%uo^$Jf4maHA9?gx2t1D@i-_~T1kAJ`k1nnId%2>>xib;4}wE$@aWms zq(MEJsWrV$?j;tJyd#mEZSQHeQd*v{4V&PGO9o6>ummL=)CG(4dA2D^0fnMe1m0oZFxfE)M^umD$48f8%Y&sbJQ5F^hFl@xMm4#{?+n2%ZzA~W z&D_$g1It_UL3nyF??jwWE=k`YUU%QJ7SWyPHRceREIAuJhh?~=OOeE&?HhHBoeA;3 zHrBqrF@n23Bx2aeRgCC!q>;J4c(gSJrj$)&EnM}m+R6ZY8}Cw8r)_Y>B^f?Ojv-~) z%G}e5Ja2t!02chLU}w}!CSXv*`UoeW!Du(_a^4Dd{c8Bqb`_APGhx533`rVN2Tzyz zu;H~0HqXAm=Mx@y>}`kyVOHL>woObu})}ybR*755{18U3h1Es0Qdiq zU_WoaMlNrd%E>5jAJ;MUMwMSbm}LiGLy;V?pxGHO%1yox87xyvP^IRi4wi*>&{Yhdj8R5)9#AT)C4TCpe?DL^Qqd19yF{gQN02#D9e; z2AXSg@ApiGsk8a6jPY(%Xt_pFe+#-_?He3MJ&O5E*9QJyMz?|td~K-r1@yjh9Aa(0p?t3@zPgJ*IJ%9;F@XztM7LQGE6 z0JkuHwx$$IKZ(tNnOnN)%~TKe-dPoNs`$h%ns)#~d|gQ8u{dz&doJ_%yRI|u?7HI9 zhCaRJXkt->qF2I6S7i)bxnRwldJ#vf^3Id&|9CFWv^3m*wvE~ZOvE7-NiJehDu@rd zLwkKaoqemC=sH!iRw|w#&NkB(V#1v5kuq?pLLB2co5&Ao0|ZRtdUkfe0Xa2e4jfOZQkC(Suc&C?6?Y^ z9r{Nm-cv&3Rs$IQ?`++oiUQt=N@-w<8r?hZ0wi8_fosnKppwsLjFcp>Dk}tjtQF#t zOWW{h-gl_-zo#&-^yUebJdntsuGm+q$V34p-o!|G5Rx(IPEx3TrbFx zp%frFAM2>=xb0lO@hB>Um7)H)hy43B1iyFoq1tje@HUx)A8a;)hN3O+2whKKiw;p+ zW<9!A*3s+UV^M4402(yf5zSLFBt9dZ&61D-JC@H6t$4SHuD^4a90P}a|07%f9;dTs z&1S2WqG9!4b0XcT3U*rz*gd-Uh_;#CXSiJ^h7!PW_TWMGiifOaibt<;e}yQ zu6XuG1PVsa(*T7yV&>n+yw!3{UQhkT4J9C?gZ8#O62&k!;T zo#D98y1I@XI=HuH19cQmM6Lf~aKSQnA}$&Rm~flKUdpA@OpY_%mxLfsBN(N}`+?-$ zE95|l8E)LhcL`4DGBHvG{9N%J4Gy@+%rCi41bhxqS@{FL^Zdc>f7f7Z=q55Lq=)Vm z*@L~#6R63D^~B(~0~miOA+Gm=*$*>3$gwbWd~QQvPR0Su$$v&{R-T1;sp;@U(*nCx z#t9tOzNdTCeew13gIuG-bSxV^LXOgGXmF1L`-7Iq2IO%8R`=0)<6drop8~qK$bpx# zb=}O4S|%Kl$a%pb;%ae&zPXkPBx@sal2HP);csZ0Q{tTF#NZf{BdAf3ce`ai~Gz?X!rL_ zCc{Vvq#Cu^2#m&6^8cK5Ql65HZ#IIRMFt!ne}za6#DR#sDMUxLljyWxb#FF20h{fc z@g>h;+w!&oN81nJLX}8P&PxTp9$1WJgBp0+q8uxJ^IWHshk579W4 zq*62rYOEvRg~McS8)HU}2+MGD9^9pIopPjnx;X?Wh?6^$Wbn)p5 zc22Aumtvzw_XG;4+2My!-d92w>E5Xw+o;afwhhtvQDwZo@GklOY?h$+>NHr~WGC?6 zg3O2Y1C0F*S9s{hdpFXxalys_rgZK_IPiKI`EzbF@tA4@Ch`?qUIoKY+puopYjas8PTw9 zh7@#2XhWNL4wO!5BBn_fiFctJ_#ZeH2RiwFyY&l3bn#g*HN8%IqN_2-ql=D- zl|uJZcC2iqE1PP!5cd>~7sPIe2ZVu1UmnO%@e(TtF)0*Tb7#C(uYJ7=r^s;f{hijJX!hv@6WT z!{0eFb73W;b@Kq-wX}rR^6#82Z93p+Er;_>pD-(CBbAzefziCb0@(QrsqE?RG~F!@ z3oZzwxXCQ)KfD(LzkgyMI@XiNKa!kRrkV-77iKdiiyjfZ2M=lIe*t9Do(j1B+ZSf+ z6N2p*O-RSPVPgJK49;zn61dDw!=V>eFh)ZMR(#rsicgyO&!Yli-?YGfnGg*e^N@7q zi~$qXT5Q^S73cPQa3)Xg!>{Q9D0s6LZSN1U;|k}Y?b__RP`N7Vk$wW_)$twDbQMTU zwiGm$Q+juuDHvRg0PSf!*UToCQU1t*#0-CC{);ZWFfJBXv{qsA^-t({QI@MuzYPyO z`Q7aEhPeOxP1s*#1%;NyWciL!vNyP!@uv2);Xn+SIldsiJ`17!{SZwbtV4;EaUipJ zCcEa735MoY!lpKD?tmZ*AL`B&9A2w~)rJ8e67>dT9Tl)TXg2#cWh$*7BTlW4Z^b=p z^FThdfO&h*kIMbjAy=KU>E9hKRB?Aa_o_*cC~sW_DHo5^Pxl`&A72&{hIfWtOy{|? ze?{3TgCbbGX&z|tthsw`CJ=35MqYo>;Mq!+&fwQZWG>zz!OJA@LgNTImOMOzb|j3R*TO5to)nB*55)zMJ(O zQ&apcPNI0oq#E0!j$_NKsq}C7aeQ~RpDDs)V3%J)Rjf9V3Q0ry$1H}{ zcu#|a@hYUA3ndaMdze*(?>rVCAdgLCvD~T=?)0dFD!&u*MY#z2Prt&dW2H193wK@>zY>pbR6fW|vjWSmVOHeKgGbH5eL|8^D@FRdZ%t=X`99^Vsx@Qt|G z^32yFLQ?OVl4qvASTf$0%i{A&QJVy~Cj8QSFG=X+fur5Y^f$aF!llQf3l2qU5&K($v?6`^^R(47za|G@R8%Df_u_e^nHWl@5l}CsN3# zeez&0R7N%O4#hr`4!-~2$(ur$-)578cTzfo~SWn7F*Rk<%Y>XW>?Ft8@ z;Sm(9ZKtQRw77!FX`PF?_l)t5NdoRTH-Oax_T0$DbncV0 z0iN%l2PJ%W!{zfSEDbk8<>bpGNJfD~-i@Xve^!vkXY5ek)QRg{UV%5)r$Zd5z;?w% z=bkn7R55N5@$X$n_8QyQz45K#`*N$%ZJZzZJlh%Z92h&rIrMNF+10E68 zqrUR%$)U5}r1`TpG@9NeXU4C=!M1DodEX8C@@fimMN5;`o){#4P2ZX6$)Rv!p8_m$ zZY41@&eRw`-b7zN(!*Kn6fkUQ7r7m9nuPMZacvVq9i?nN=nGjNTBhDf`u3)<@4}uE z;fakX=nLW5(%VRv)(Vnfw2j(UuE$qX7T|(~JP&2bPV6&^XKof}prGA=8xuSm(|JZh zo~AKWy>vzA?0e+G9wD9|TLBpF$Buv~inOXQJg!L0zhQY55;CBB^@QnWeT^ffW_-Y7y z!BZcDd1kEr*-COU4@upHi}*C{A>H}!DIVbSYwInunSk??uru~6k^F3dvo5I$o@FXR zw`3kwboXRJn^;HK)n_laa;{R~_C9>jR(8i%>l&2fTpYpkA8`ztjNI?ks}u(Is^9kQQazIj8+o)UKgSq_d@e64XDQSjz~<8=Lr9GLm{3P^lXk*?|E7Wqq!30o9AL5q%KK^1O4kB5yvhgsLS`~0$2Nkz zlmhC7&IXIN94Oq;My9i-Waq&I66mdhpBk=_`d#B;lTI`UNnN4Y1CsRmQ%k7s4s{$2 z;8`%NKC~D*fC#gb(fk(zRl{qb@aB6a^ma39rEP%n41N#B(3VKbf2FQj<>c?mIDD&c zp6==Cr2&2^)N;x&v$Fm->vk!fc^qv_y`~<*lm2~V?cFmZNhb;d`MJ=Q%`a)(f>>Cx zQVJwZMTuCgKJ$0}Jeseo0cpmXyn{go`^FunzsBF93$9Pbt_L-BZdPGr>;9FfS1d(8 zZV==5Dt&RYY&8u^PeqYCZ6xi##dM2wA(f8!&0NuxXH7~C@x$B{I&n=j?yKF1Tdr<` zl2w{GY2|FT;8!~H(Q$;%Ds089`}tJ*QWK4?*Fldb;+*lAP_lpaQ#fnTfI^YNOwLb>EiI&jk+ns#uqM`S=#j`8pSOI_2WB@5+Lg=Yrq~&-nS~A_UK`Tp=O%LZO#i zOg2syhRb@ds3jLy2)Wzt2PkC!xhi*!u*Q6jE?%73-y)C2~zxRV>f*2RqtqJQk%aB(*L-?cgHYf~= zL95wMm}<3hY;n1UpPr1xpU%d#@6kO>j{M7wa;2P+=XRLa`I<~wdkpL5YND>z0gQf; zLPSm-qEjR{!3T#zxXg8vq317v>V71{xf4+wxX>Lwpd~Lc*$VlLoz6 zbT~hR*iZcc$@em0*H&Gwh<6&KNt8pLXEWV+@d^lEyF=Y4u;k-^Qt;;EEY{2a5S{2( zihtJkfrpALsaZCjnr-($pUEp}U{w{J=o5i=$8N>0wzIIU`VCCCUx%mPorOG$A`p2t z2}<8)(nMD?vVP%xTyGRfxc#z%W7|UMY#U*0-6ul2hJ0|V^(g&UEfZhnLZZ`s0W zrkV*d4Rn5^AWRE%LU@*%3yf#9$SJQ||D)(U+`0VTIBv_zj_hPs8ieqiTSgh}rH}?C z>02~uqk)iY5-PK0gskV>Pe`OBsZ`pDqG%~m{GQ)G;Bs9bKIh!`dB0yT@=(SQ9LpDh zt9%Gq_FMq}30)+z^5aNxr5<_bbdl2;SBbIz8Pe08^_-n(Au9LI2al|8Q1i!?1TETy z_7Wm!R;YqI1cfkvoHRJE)`Jty_Zeju2CQQh(0)cgP3;<@OT9AK&Q0Rn=Xu|$=NA?H zZYT#cTr`mjOhfg?U6|pz4$a4zamUV$<7Sz)k`1=oph0yx>3_uM{{t8t^ESpiS~GB? zz&2X)lh08l$3cAgPsZ}13yiK8!QPT@jN$Advcvrc*&P!_Jw04mnSn~oV;%A6+aPo+ zS^|HF6h9keL&3WoO8)@~c@WQb+^;34BlOvC+lx`~#Xg!??8M~jTJWyci>%zjS-?kV z>6|lNIN`r0M&P&yaoDpC#EKk2;Zp$gZ&fGTCrZFUi+Yf(93>CiJkVCw8styhWSP9R?_rt)&MR=kun^~pR#qS=4KkPI+> z_joLj6yuVMmcvi!yJS*+Daj7<0)qpSV9|zb(z`qk58uzEnNBR$F5iR~rR~vjOEMKT zv4Cl(f{8i*nY#2BqEq}`ln8!8&*UXw^z|CbtuA3DjP4N$2Nz=WR1c22=x~bzkC34I zJE>dZO!$?39TzM=Nk&tS^Pb4v325WD*syYc5aRC{mDy<8MYXLs7;faVm) zW)D-`e}uiaHWG`C{?hgUX^3-hz(aDBE;8A`wfmd^GNu57VbAG7FC*H0-4Bk=iiCm( z#+ccsf^I{iT*vh%M0=4Rkc}KI8d!|x=09kO!7*rfWWvcC72%Q3v;@Z;YvGHXL|dhQ+SKYE$&xN?O|)HTK!^&iK)M*qtYXk)Y$H5^v9OTRu zxcb8H)ZSnc%pI1eYggQ2w(c;;Mcw~F{oE94eqRoP0}avtL^{s@Ef1}4zM_v{hm}>^ zbnbnRJ%r5Q89d`|P-D3OD*O5Z{n+`G_h?T*Lt!x#ce1p;67K=MGE!uMQW0?;Iz-Pq zTS8e@4E_0H8jOrv2=(@1D7)_}XaqQb+2Z}wcH{+ku9!`mzdhnzM$y#n-fJdy=NXV+ ze4eC69VVaD=i7wnD#Nd{ywmD$KKZaWgv(oh72ieW(ZRNxqT{OH4_OD3AnM;=; zUmwAZlZ0`Ttp;SRuqGd7Z^nrA5_I3P3-~wBl7mr2@b0N%Y`2J!(U*xfHZI;ce{dW4 zJlBRDi^NEtTMMzX7bhlpkLZVU@l2+&89caulPD)7(m<=nY+Uag%FZ-IO{XOC=#VD} zeo!J6%^a1Aw}s{1UHId9FFVU!ox$-1L~MpS_%w;pv~B<4=H?|xMC-6^JRz>vHSx#Z z!!~!cz7T_|Fm`*iCe*ehLeK09nAOn)+DEi;txXwbCwcI1UMt`}-_2j?7e&&`(rQ-U zw<8_Lo2Y_GGWd!JlN>iG91M+vUwU7dhX3|KNmT_|r?Qi*>4*g8nLY|R3vku`Wo+k> zXXw-;OiW|`ph|x;?%{&a-N_%-x@zd(E)nwkhXVU)_6=tJodPEOwE*2Xh0?DQ1=z83 z4)-dy9eU0sz`gsEZ8zQZfZthybWPnII$y<_jcQqk=Yu!X68%fmvFtVR8T5gOQQkB0 zvYKXm?!#TB33QI)H+p8N5F_xPA~aoHNdIfm!Z6R1+)K?Q9JwjZHAKv&f-fnJe0!D5 zc_3-)RX0Z5PHln(k#P{98qK?Kq9CKzize8XVPC}=Zn^3^w5+v9dyiVIS+xMaZhl9m zXG-89iPP}vcrpF+>@aa~dXImMRwA}(a0{=Vuio7yOB9BFk{MrEEE9i41MlCWZqxj* z=uaj&oaRL?t-gRY?(@hz*ClMr#R6jNYfC4%T%_W2ifMNDO4?T@2GvCkL^9$SsI6Rz z&wN^8OW;K?+&+O!Ss_d0zRHr5Tk82B(HfM>>te!7#fa2^I=R38IxdWJ=kECYqZwK= zp<^tOF83Bh^YqQ^7pwP-{Y*X&!ZQu8>z5G0s6O)hn>MN9dmhOLKVy4H5RTW~jb8&^ zlP~#NRDa9cn*WG6PP9~KmcRFeUjlVtU}FroA{%j8mJi(Hkkvc%kS;$k#^Q!tIM@7! z>|Ghac^}xtWxnO`=`l&P51IgdCNj1+-fgD)A|=2=oMlim4(wfJVUo&a3=cR>k|n0X z*sH~)Gf0B_wkLp+Q`y$Fa@lC^u^h{)TiJCJS27{`nj|hq3L`e@q1FDm^z+H_&~W}D zJ1~1awKDa9oWpC`>UrPT%LgSHqSbHa*P(wu1AUYmGnh)8NK@I zCvoX^fG%BAKz_enx>aPyRn&vX{LfUvwbUXW4OfXoq?OKmw?K0VRmAI9%Oge zkmF(itn~z7WE_}jq4!~8&>7H)NW&4ucy3AHVYL5HMYM~=P|ZPx9M0`!os+|8+#CUz zWw?@^!Mm9k-S|gs{Ks*bX=VknrO9Muul?&fnGN z`=R$po^&d?G4g>3XKK1F7Tv<-}y?7cx1Rp-CS^$XMuA{%+w0Xb?<| z@{K_1kTh9(+Lp=(Erg_=e3UqUf%JxiLk_LSVQW=7@kR#yq*4eE+gCu;{aTn+EeU&b zm+|i6b?~d_H6i3B-M!-*PJdzotq)I9=w1t^ZS$e`;5o8Mu!dMXsVBB&cA)*_J2f%u zW+w@i0oGO1wJv8cOJo-cY0Tm#sj{>|-H(ji^u+d>OK7hq%h}kZaeD$!(6T*BnC*C( zO1-ZUC zu24B~m-1uvo}A?Oy&5=UG>hB&F$90U?yt$LK4#O=P)|?VYf%GsyzO4ulkkl`2OZru zGBQP(s2qRH2EP|X+tNOK(Kdn0viZUaMCViSx2xglMq%#rwR^O<_^s97M>DyQ#LMWJ z|Aa|sO$AYBekYZD8dPVcLHwj#GO<<~8@-=mdO|!-l2awjpD&n-i_yR995On-0^Xk?cT~sd#%4qGy&Oe#EZ%sk$j!dSk{8E{=~Vft=r#2N zneb4T&Jp9^Y3!TNd)Ffh{DE?lkxJhJW`h#NTS3q!tqKA{C?M(Ima1s zi;BMC>)0^RZt$Sem;1>d>tL#AeU|1uslb&wH`yKYbE%m3TChBOkE}a<5zRP3+vh8y zNoG$g_~(yf2rEt^&VQxDeE!Jn$B<3^r8S`ApG9h{>RF%Q74XE`xaRDt5A65eG|)V% zi^b=1=nU`C>XdX{?7DaxFMN1M=VfT2$=FuRFi)jcnJyqQK>*)vE&+kKX?QuJig?sZ zprGm}Hb-|ZZV>e)RRQJb7Hk0Wvxk}0=H3i9{u5geX^9u>{sTrj5MNI#!@GaAxZuVY zX!*zmoiuv+ob7mA-H}W)UZ`_&-qnN@tHKwBnV2XP$)xtpCXHX_W7j7`?!xY)T=oH$ zTrM>QRcGE2wKolWnL>70{5%mnu0@t~)|1dAIoj^83Mqaosr`O6tm+d-_mUNKWMvL+ zX)LB`J5|A`@imkG=sdEcv(eL04!%_f-~i`K0(a+;4ZNF7!SpN^_jazO^_ifxWr6wD(dD-yF zGK4HtGk|)>0`Pc#nffVMP=`!kP}{6LSsJI!KY>KN&j_St~cc~bH7*-{jy?k_v*!{w5plv$kvdsj*XZ&XMp(> z6vHg6I!)U}IVwgqVfd?zg!^$4Wf8p(mA02m65xx ztEpm@GE8L0L(n4u?v2(sjIf%()h;Z+Q|2Ss+;jy^$I^*~JMR}e{)iTNOJkluI(sDb z8Ya9O#vA%;*s%DwWbUrLSRVJ0&iV6`ZLbi9^%@~~t~rJcHIyX7b4oD%#zq|4MlmNuhkNneiTnx` zJ#{-?`gG;tV7OS39`2Cq`plL(Pn7` z-L_~QSY!v{`Tux_V&NN-e8!y|8Z`jDvNy;JOmP3b@t|+~oi^RhBs+H*@+#Y_cpnR~ zT&SN-_;r(%t&e14p8sTiWbMYXonpAvRhkMf7UJe7U*>)lRbtBOQ@DC_CAW=T2hXQH zAdMnZFss7|G|Ot?(A%}3JVzhj&U{5O-7<+>U< zqK#`O!6d7dP+$@Ot~;MI|B1*z&)0G~cdkENn)#UMYRj`Is)+lPlW}>m4A&ZU2-oJ% z+CXHt?Az zQP?U|N`J_2q7@MzaYyiJTs+4YtRJdk@W&)>ealL0h>C*Fj9bJvSb_|thr@a24%if* zi59yTa7*Pja$#E{@SSEkyWVF3cRoJ>Lvuygn}+Rd%ZLwaaN#t6{x<_`*o4>KFC!N9 zQB>CI}Z*QGJvBw*5E_udkG`=TU!fKXg88=MqP@h`Xa~(h2-pYXtqeN0`83 zO?)vkfM#SeB}Wro)O~Z|FgOzZRhu2hBQdwB7s(?r{zyhbPElrsX^wER;sGUm~RD^;G&I^*i(O z^--d=@e*}aQo(Dl`p9EG3-rAJu~z*%?-D-BNggo6W1By+P9LPPE7OI(nY{yZJB|^@ zly=Gt)sUV#2Q;nnBlr1y08B2l>8bxhvZoMe2t0+w-kvaTNP}*UP~z-ZS$b=>7>&7K z0yU;dJQFaS*z11=k?Z@hRp%!C$=Sl7S0YxhHc;|Y9vlwH+Wd6>&dxvEOA_+M(W6k1 z1{4I*%U?I3MMfL5ZAt>q3j@p_xJ|;lH{tRH*~BQWo9v70sD7yC%N_M8#YGy1n17;x zRdzp19B!T?hTmlIKzl2l8y$`YqLSP{p#l1DN*2+`enG_c$B;n@ZP;w8Zfh!VR3Oq^LvrGzDMy43{Ok~)NwJE&fx?yX^RUOL7vMrfLKw-&3GDA(mBO zuY$uSU)Wk)MF0xbhe%lH6(Xw6+hdYZFM2G<8R@QZS_<}NRQhj z8;=o+8rbk{Hl3xo5!AR6a%e!tcCte}X)1oqZvDHE`>fW2s5J>6$Bwb(l7{q+*AP2! z6lwpUAn96ILt>`S#PcH>@Z!2IT$i{=p6@vWGAky+rh6WcVKs|qp1xvjwa;K()Oh$X z_aj?++ZuWj{c!QSF0$_UDEsB?dsf}{A3gQ$BqKV*2`5`@M=_(CfH_4xXUHo3jm1iQLJ@zFC)+&o}M zTDE^-k~=inq0u^;Tp_`QNR`0L)elH8-|@)3a+4cPT172k-%Lee;FQ z)K$YvjYcT4>lGPk)QZ-Np31i0A+!rbn<9#pRBrNIMrFz?t^ z3{tb^GmYWo#ol&WDI$Om3yeU&ZvqxHinFJW8gb|Nvt8*X!+rM+;GLVY+y#=0PL5hwJt>z?JeI(pHB<5GsT(xRxsZr9{UAP^A}LSgcbphY zy?X76HvbR4jftN1VRZ*NY>IE+Htd< zm~Wg#Z+ZLCGc}52Ti*e$%0r*KxZ8j;*e8KEmpgJ-gT1)4ZAJ9-{8v=1PM$b3+OUi6 zU>ibySQZ@GjuOfmD7>f*OGi#X_=J8E6Eg#5HmJe%fpT)>)G%W&AP{M}#EaXePC|N6XC5f2F2`XJMXXHe!)s;1`2G+7mgMA0 z&9yUduZt3s+Z6eYRAOJ~u9&lQTBHKMT4Wwfli0NIlqVc}l>XF5%ScbcsZxbXbO0mV-iiln9^s48KcwzrFx5Hc3NM^{n40l}tm|rJ%6p>7xZ9(2eu5ho zUjnFIy%~N#$fDo3X0YWcZ_%OKneN|b24PW=q*rJ==dCEuO-^6R8S`(@vjQ9OrFtH= zRF+`XPHncY;sy#5nUyz%$?(qQD6}R7C&h(xrDtvMPw#)2 zPc#{avP+;ov<6qHRpFA%LA+-fkA9&OINR<1Bws!ruRNQKEv<3zrZ5WoR3>m%g)wkF zdp-FYBuJFARY0q#n;A0Va}*Cx)5oun2HjN$hcFecZjUZ^vP+m#-kC^^=>PLW(pVd7 z!e#e+^3Lx&*c%gqY3YeXC94HBm9>f4uA4M*c?wOK|Bos=>a&qI4PeQhTKH~phIg?H z;?i^dB=pTP8@Y5r?4F(i6BLt3&&o_F@4Q7!9ipJZN`q6|P^Umk?v7eD`_fODVRgWmzTpr!>7hICsexA^%{zIP$*N6Z{%)U!MMbk|c% zlqAv>Ehm_(Puj?aXO7_S??Mh5e}+dZt~1v4f64I^$|P7el%CD@XZ-S{ZB71#;OD#3 z$v@3qe3$hyeLwCZ7UxN0jzI^ee&_$LIups=`3~F&8;OS(DPyAfJDYr`hs1cDK0Jxr zg8pxFusksYH9m8=g6~xa_!mO9wJ0uh%ciG=8sO!RSlHz>7w#;nWQ3j$)5*_A=&kf! zkiGdRj*oWW{EweR%LQNIyzdNd|L!Eb_S=Fh?~vs-or=T(F9TdtkWO@N2e3P9kHK2I z7*=cb8lt3M!Z>e_WE3}b)MO2qpsu7kWxo!Rkl{kuIN2KO-dscV5S{@TRzNZ=-r>(a z3r1(xR_w^&U1}Qb`2DOImw5C6eo|>9H#>h2hcA#z$VRv6Qq)>TYt;66&{ zy5VQFon-Wi=*^i&lsH1!{)ukhG!IYpr4jon zQ6T%glXi}KfDa1GaNX-Wxch(w99+WZ!rL8adshmn-(F!O=I@1;p?tQpRhx6Xb%ltE zIkLyR-Qat8P0fJFM>ZwqK8@V@4tt+{!sF8MbYk7 zJpnX(fT_OCkuCmnS%-;R0XtqX{ns zx=Ol_Y=l|D(y%4!EjwK`46R+mc}9YK&5asQ-tn{+a;u{lv)@nYb)|9=eMA#yca~EI zTbb)yH-Le17K(`qqCuZ98!=0kJ&}<~{Uj!1;{YXkAN?S`#1j=)yYO6<7@V&65C)73 zYh+)hlepIj=>C`Az3F+uB~BfFsFl*Ar)R_C*Al>`tKqm@p2v2o2$wbgA*1%&u{ZB3 zJ>HoKfA=V0_@QhFYFf&9?YWIBc>W6T-kT`DeQ2Sa!BxkK)328$!7L(`)%h1qUPt=i z+7;_~|HyHyT485%VzivfC^o~Qws7*}L=?&IISSiCgK>G!QvAN&pE~T_fI+uhVAaTP z=Ifpy3~p0_1jpH|H=ni0(|?Y$PvpXvP3sxby9jqC8e`eFVl-5c;a+Y?V`{E_Ci<5q z!rlmqNdjhEx}g*&)i9BDD-0zz1v9wwk7mP;r+S>9Vk#TdEew_OuaPg(li`WTO88NJ z1x`8@LFI!=Hd=I)+52}q(nw9BaNr82x(aYZ15Y61XE^ihr83xQYLn0D`f#`9ACrC7 z3dfUuP+B+#bBxnK{cbspkQ}bwZ1IGM17kcSJG;CW@f`xR%B)*bY-w(~#a+hMi`sx$UBT z#5`;wS98~dq>M%2a@GGpBQp`skFSBnCilsgcRr+)*+s(&B#F}SG~AfqjE@|v=>l~t zRz#TRLf?3yU~nkJ)O8J@d10wMU5P(EW%ljbht(5%<0Apd``At5B2q) zie|VTgDT!LsXoqRR>o`CF(3%VhVIZHT*WM#rHRr4{p^GnqaE;zWaFy?bn>p)(cw8vi)HEHVyeHAiB#;zsS(kQWyuVZHPzaxv>1O!A!uh6c`%yZS6UeKUVIRQbxjHWTK)oov9NzQ3gD zwj^EJ@rl*k*GeCq%*Hq86R~sII8H%zfL>~dz?|{*)jQK_P%Fw9vfe%;-`tEra^od3 z>RAVc0aCP7t_r9Hq<^mzH!InQd);yl z&%AZOuUbpcD!qve3CzOxhLz~MZ6`aNvIrK}7}F>1HPn1$8#doKP9kg-=$Ztl%;iTh+|9k$u^(>2 z#QWONm#+`WbFSf;1%)){r4<+XAe;!_GJx|=huGJ0^|Y+j6HC{=!F|3**)v@{C#EHd zXJa;zL;r=Mq<$1$j;u!~o{KlUV>~=qybT@xC1Tt^an9dP2P|dR;K#OXGQ+3GX4YjB zGVO8z?d18$w#`d8HFZ@^HEli?#;-;Ve%_wdZNzOgnamvxmf{rNH4u02V*bsu0v!a} zS%)_b;DRVFeWM3DfBFB-2X5p@{Y`f2>5!U%@KA1nsR7!wJtW?9 zWZ>tUn|RFoG5xDO%v5pv=t668=#_m-J;%v&D^`_~!sK@H`@J&P8?lT&@mmU?yCbon zT7mPtn2UZU5#TQ#!uT&Rq3tncII4M{Bt6!H5=}3pX16HBjN@)pnn9`3F7(&4h0M7R z8Hr6(XyK$NRIo4vHfArI{PzNG8=j1zW9MwVduDRNdf(~fZ>I3YIE0F~r1Bg2pKShG zZTKs20mTf@;j@2EoNja|d0k#igoZN6;;MY|jgg>w)0WVuQyQt0`E}yJh>|B)>q)hS zFEF-_l%IuRjI&GLO$$q%MOa2bZIk+%B@H zxswFsuEbOBd$^Zr7cg8l60Bh_UDodhDPR841A@PSosT)F&zEkJE|V{OJ(Q0GJuc4@)w8|KI!W+D8eCOn#A2j*YjFw>U5 zB~L>h*YudzqhgCZce3*jZap!LY`DCEM<;!y)0UMo-Fx`IC!X&m=O2sLJq01awTi|p z9l<>pm*MYBYi^8Ogj{O{viZh+7+3NEh50?*fwVoGR);dTz0Q!0aWKXUu~vkgIn1s) zDvB$oCFAFEA%2eeP;!u`0H4CB^t`NRYSw@W!!0~ z2kCMBgj!nD>G>gGo*|4Kk)8OA^X7Tof0&^D$sGAlAIxvZ@_c{*a{cl{`gD{((_XA6 zfyO4Bgc5R#_XTn121oJ2PgnXYNRf*xDW;K68_6u=OOW`w8!VC&K?En&H2f$4w?hmr zEr|j>S$X<*yCB^AWCBXc?%dkoAGm=&#`P%|=$7wisKED5a&vSkN#l3S$NmX&g@0Dt z#OGZkD$+UdK8$7CbRXFa2XBNY`h95r`Zx3au^4to^K*Xo80)t{ito)wkq@Vy(>;5f z(eFqB(nHGJokb1UzuFKjJ~iNnz;tH0<8`Lj%ar~6{1el@{sL~`yIk|i1R*vy89kpI zrZR`ep@gU&%zR--*4rM$tqa{avmMq*^M*-ktu5}ZO5vS7MYQgLEEr9DOuaVF1>G$Y zXx28vX1QevhM675cOIIUl;DeBd{*MeYxd}Vm*UAfP432AV=nXd87^Vre?WR(k!iD@ zkjV-LT-0hc?x!MPY6IVs-EkT5+W_tjUkc-Gw(}g7&-l>qE4eQj2ah!qsc7#MV)oM= z&U|r(WxLf#f}1jgiw}VJ_ztvKw}!jWy@1pCc^t1T9nbIJ>hOV3GT1$T%gVOq6ZKSQ zJhwcFR&AB1+PhaDU(3=kOWpt-AFcBEg+seOJTp>L2^tu z5v^t!!@mADVk4srJ`|InHzT zjZu7`5+_hOM6ct zG>^R7A`j&W=ZJ##29o^Pn@ikg&nf>W%4I1f5J&Mg8@Dybso7^kyxUqvyOs_U*Lpoz z>KsJ^Ht|`WRo3{(Cy!lNb`mel>!D>=oyoJ`?qpW9Cr-8DskP$XsA*{eg~M;jCwoII zlj6O@L;fh)9K-djx=SS2vdqNW$H=fogNanqrsY*LD|>ikes%HjCn7GiA|+& z-lT&T3n7@XS#h*QUg2tRKT|G2?98OZBk3?qpCF_1EI>33Yox$xxE1 z={2Nkl|6Ljt$d=USxM&4n26&hOo5WydtjZC04z`6h^qJI;j0DXVaBX>%t=xP`cWQ& zR(v5ZX7v%*SwlAGH_D;y=QFVXrT`r{F9x@!h+*&EF#Jj#*`Hi1JH@*hf1ezH`QAJi zZlxSfmzYBjcb3xNRh!A>0RggBb{=l72*;+kCu>sa3_4d&4krI7fpMw@ zR(U0;zH$k+i(CO0dImgG7Jy|_B9*Z1!b`E*@PJIgw@-GF3Gc2F3=oHdf<|y5`XYX9 zehDx81Ta!f16_nG&~8luw(p$6iKoP~&P#e(kB6yHczizz)?P)GU;d^OW%tq)9o?KZ6E8Mh5Pf9!;*BFn7!d5X&#~m@df}hzUEWu1ypKOe1 zq#m}`(0w_I{Zl@f&Tpgi_JAT$YMqROawQP%dx{K1I>Lt|Hn#2&TX5O(5L{uf2DV-x zL{n@X`L|dR4IMdV=#mFE`xnsB_u5?i`XJWqzkT%PDRG#35vj-;r{^KI&EMhNJzO=(00|O6{+QH}nkNaOomn zbhjWb*Twpe_QYfOId-_}fb)*GsHVrlEZ0M{COVTm;IEN6Zi~p!hjy^JH6Qv91VYxH z`$TI+4pa%2(p{hC(So_sRLj~KPUZ=dpVOLf&BnT{QH;KdV@C?;KAvgnen631*Cje<;U0yz12hYN&bYSUeq6H@;e;eTUT zx>*-Jrx=2|e=VMskKo-mhslhZ6{xag4l&enffSqw%a%3KZMqG(|78ch`6Nr*&kEy$ z=xn-KIs!-bd2#NgVI=nNG3uT&2X{s-MEmapbgJrA2oCq*L={>{-C0ZS$M_V~wP{6< z<6-#JNr+qNs)`HpCXnq0Rn*y}nOHua2U(*Hte(RJ-uLA}rw-UKTYRfXmD*hF{rm=8 z#^-}oN~Y~CD?#$giD%%SI0tH$kyhf77Uo;|$%IhE&^T^{77gCa~0dWhJSW-I;6|%Wrc?s4>ee5a5?#~~p=i0BkJh_+rn&E-_BtubufjaN6mf<2|7jw1?uhFS9s)*krN4WcI6(@N8 z945V9i|q%xadfIWdR?d>->!RbVs}irKUY?A@ivpV-uKr~!gm>%90;b(L>+#WaT zI$?Ix3|!PFKyuxpFoSC1*W;^j>dhnggWriwdRD||HauWcjW&dTIY!L4n!|!3XQI)Q zfVYp8qPwdsPH9X-+3Q=0jguq%ZS%&m%geC1y_gJ}xv`CM4Onfdi?QNrc;Y}3+5A2Q z70+BF!Fycc0q+R*pJ`4~$9kxRNhxZV2GO;HgG_p)0zD{_igk}V;hSjzYdiHmX7n#c z3!_mYv&(~#N)g};7R$1Bw-4E*}zq>L;GkyGt7+D@cNU5^BwJK}BvF z6?%LW9+f@9*E4RRlGi<|tQw6SKJLig;b);8JD4ekBJ^(O0GWT)7~-B?BvC5cnJ2Wh=* z2aiWoF~(2=IwCpJS3H$OdwIk8nQxgzVW+@$Sq)V&ol0bO9K(4DhFIq!iT-Q0!&}1- zB+2C|(PereOJo{8AG}3VLke-%fmCpemIrao4>meCEkLvRHp$H8v$@9caO2}-+BjVm zPm2Yy`(@fNM{X;eJe7!rw(h8Q*NA(?-w(t*1Z@8fHj&nBKEtW4g{rGlN%rFah--7l zPmdb-o$_q_Qe8??`aiJKXPt$&I%nD9&GBrMNCOpEs83HM39)P255Q&Cgl@<{s{i9Z za&7-D(i2@yHZG6{m7vLxdiW*JsegqB%E!_DSDWZU+YhMhY>JT%&*?J#LA+9zi){k# zAiZZcH_&jcdRclI6Z%P;te5e|`zbB#`s|}PakoB%EiuH%+yVMnuamY^*jF`PS_f+; zsX|$84){`0n9`k&PGfGI!uGY$WD-M-1>+z}crtYTJWUNA>Y~(R273*s<20vJAZ&dJ z&)Fn1zE^|Dp^6|(I3!48=iH&EhvhhF#?5A`{x359$2GQOcpiTj+=He!6ERUh8X}~m z!1U8F=1Dv8yh9DF_AkMY-Cs!VxCb=sYYnlvsf_a)<>~Wv{2utOCIa7LaBZANCmJ6_ z=`H+QoqyNK8v_F0 zKk{$|OJnT6af#%g_`?`EZh)mrVsQSHE}Enth6{Xjte3jorjw`WVdM=&`&>s1p2MF} zyDrwempg17yi5{b4eAfXD9Tptvaja-}6{_X}fk zr0G5T@2d{JeW!x5JNRGV#vrEDZlHfnEzsjvAG_+?Oys@sc*&-WHV!J$V>QW)MgB|b zV11lO@Ci$?)9plR6@Ne2ahB=?szF*?4K8tt0f)#_kQx#IT=svA-C;x0nG{0UvuDX1 zR)QNyrzmkFoA_*3;Ra;Q*y5F;^ufpxe1E(M6PiZI;|tDJ(!&d{Q-E{9M zzC$l13B?}yWMH2I&f8u~-ElocDr<0ltu@$v%@fU6UB|WM{9aXSm_7W$0JjFG66NSg zP!z2}PkygtvsGK+S;b52nYtDoy}n_&Rv9Vaa|E6TufXjzRZeBkZ5r5DLTB_{hQ4_d zAYir#sIDmD8Z3@u$R9tPoT zxL$UJDBe;+pH@E@e`NyN-@1ecNFF`CP0%(!^df&xM%2y5axXgW;C@z;PeO8 zuO1AMiHBoAa?Y>0GF9dlxb5lxYN zLtn`zkeEyv&gIZSJiI;%R~#;b@fjZ|2XAG?JE>0xsAqt8p(p8I^ znOSvXnE7RZ)m4BPRZXBzp2Ba^j_*k)m6_y^*3o;gWuy=1N3F*yH8JQfUPE^I ztJ1tbD`01EFx_o*0=@*A!jTzl&9P4{@X}-fv!O47c!^E|_-sJGl}yC|348LB_~ABx zJy@af5JPNYvAkOnEvprw>)3fJKJ6DV@sFn(%DVX6yq-?4QsJJ=v8FMmJ?PZ+n_PZ9 z6~7-9!pTc?xF=eh>D`_^oO*0Nv~Byqw*BSzQ|}r;^HCb3zvCxO=UGmhG$s=r?PBtO z*3<;caJney9`$ipN1B^UXz8LAbg$`E#@j-Xc5I~l{4K#|?$v~M-SN1$t{68o=aVC= zjB)tXUpjOBO7d@hAC0~(OY=o7nDf;N7}c!ERlRP)yt5~POUdQkAm7{A?AhhE)Ky6F5Je78y;CN;jGEl=GbQ(OVE-0H}bw+2w* z@EO0E$I*Mzwj_Gi7b@h-lG#hnGRtx|Q5iXHbUkjtNlI&=T*)1HBvV1Ag=&yvyFZcV zH$F2a&yVAOvc-@qmO*w_*3!`O`E9fxgTsu+ep#m8cQujhW$BHfHndWgB!91K(wwq|5Wn7=-QsbJ z#KoQ?s~y|GA-;zkoVy$Uo|?`v$75R^)Q(CUVWZ=dfC69+#BQ z&pdjTbWYJb`mkpsEKS)9&M!Ne51lD=@}V-`i6adaj0qj>^CH=f*NIGc0N?fh%`)LmCh63&Fte)--50pg@ZtPENefzL&W`bYj*+qN5`wiXX(pZ4T7EqlQ!@ZwAYA zX4GSAG2Z^ug_FlNbG_XkAYd`SuRJJf+vsqg{48uD-&PdU#Ukq9s%b)u2kw!Vo5Dak zAf8m%^wXu5|KVff1TwR+gBY;|#P9wHt9h-Pa?>O5Y-2p>nV3WdnlzABq~p@-z@5Ex z36FL=gW=Etn%eJUV;pA*#~il9xxg)W>^Fa>;&c2)F+6u~5Gi*y7RO{uF=kpE!$ldv z68EFz-^&9 zllbOFTh|2D6Q9RNsps1b*rOjq`YZWSf6q+57t43hZ62~Y+3b zZwc?`h{3XVQ&7#mnOHBLgoigw0_UMT+P(CmjiYB6h*{mkKW+R>EO&(7sEq~}l?u{z zVjYC=`=}9x>5wvc9h{3(=68x}wEN5@EG^v#XF!_pjRvrb-HFLtPtln>bYXr}6Dc^^ zOQlPnvF^JPLDMS+_8hwd{&JV-#Uy3A>(Mpp695?3>V_3}qhW&kbgs%}A?7^h`;7}j z=pYCS8cIotga(PUlZuAS>=7vn8D%9E5!ZQKgfwKPq*7^5RNCWny?_4x z^nKoL{sFh=^SaLSI3ADt9k_MNS8-3*{Nl4!dYu1kVXoz5<(0h5vsCD(0n~Y`qD;IH zxiUG{I`er3>e#!GQ!*EbSD6O%FArdq_zq;Mk`jp4UIfY9YxL5M+0@Rrfi;~ch&N)C zIfp(;TqGsSeKzI&a_=?i_@a6A{%tc9wwOX4qNZT}lC^Ml+BH(VYo5)x#raf4{uoC0 zT*dmP{j`ZF2$jteBjA!>Z~<(5A4=8P0Q@UE{6 zDx9lEw)rPp`d=Gk{m}yk+yXFOX+717Sph{pHE=)J69*(CQJnt`&3vH+j|*(c_th#y z+(m=VR2PRMwUM-NWvvl-TM?i>TPE)+V$b;Ri!ZQ`dO+F{syL8zsFpR5~fub z0>CG|7E|<-whYGP{99a`hF9?kXz(xjka#+A<+gftw1)Q}s@!c}jG0rZphZ~SwC@0gG^ zlEho8lU+RW8c*C5z!T2JxKUI9`mQ@*o(UBdm7C?x0wO~IW~0n%klW`-&bN5I81U#ppDtj zwb-1t6@|CFfUSA@;IMKxDHWNDA~pv|+;{%`b*IH9L#K}RztM+Ti;iJHiWq0&k-@Dx z;)G|*`Rw{NAx_R*kd;y0Ky~)*g->}c_@3Wg*q1o6&o)YN%nxNGL*HodpcsuhAB9%+ zh1^>07~B_^k5!M45$(xRIQ`FJF3O_GI!nM0&fGJBNfF9)^yWKO&C!tf-u*|O+>z$$ z8%OBkG&Sgnl0-PJ#N>MIC*D_6aogQo>T&3Q{N$@(VA=_D)Hz%^p@jCHJ%Y*yq&Ul8 zHDuo0S)eVehT*46k99t1LHG7z1}(+KKfA8~E{Uln6Ml zVB!zDp~X=Lj9HLG8$1ia$=QxQ6q^NgpYo}VdNjysIxrEvQT0pVEV#F;fOlXHiP&_D znRQ_}Boed9v~hPyw5F)-Bs}tTyPR6jwi%Ip% zv(P!6pyNw1@?>ThRV|m{eq;)BnZ>famvm2&6Muh>p|^R@Oj6=W=KT%^X8jswqV$f_k&BB#Ex-=0rtt41r7d{! zU=6(?c7>*yiQ+F8O>X~v4fOSlgG{wbcdLw4Ir_?XFnFB4%dFofwmbo-Elh?p%hN&iwi*{Q zg@3N+%!R?H%VEv46}T>@f_^_F3^HOVIOHVBY*FYSMYl_ci((h6tPsX$!BS8rXd0*% zc;STEQ$TS4QgjfnATg;z7%(cm|ZR=eduRWsprD_UpF_wjg z+3Bo~LA=eRugB=D{F(HAwiul{D1xzRQ&IQ%cGTM4%hX+}C0}+)(}clvJTPB^)=!%R z8fy#at^+HGQS*4%dbyKyo=C^(>+WFBxpHv&9Snag+#x|flN>S9#y=05hVE{HuUZFdVro)#IvmzM6{URYH118YwEsq6X_n zF~&5XiEue*bEqyDYC^9uVgl?#aPMtwI>jt!Mv)SHh~ndZKOE z%mn-uM2&YZdGDtnEv8H8(GP;0p}ZHo`nCjHd>%5d((aJA4MOzBRR>~wFaV$Kzk;(` zMrgQh94(o3pJq1Rqj~dr=F-+0xHclJ9#8?1KjgXz(r7m z)9bP2Zi!DKn>TI5g)+u;=KNI+TfIifgg+Dc8E+HT3AC|&XJ=r7x(hN78TfJa4txKw zG<+&w31tV;$-_;VaCS@u{f+K1ep`;y7k=I_L1{b{HCX^cmJPJ>J~DYj90u=y zNYq!qp<;aY?rNSO@f;U{4M#G02HsIn>guAe+tv_~x;w!hpRb97u)8@;KW9tY`^OOlJ|l7b z_aoL~ydxXWoWg9y9;_2m!u^eL_@RFezk7+~OoIpTvd%ob-8+HalyYO^`5wTQC;bfv z0`xiGmT2tZ@8~XT6u7&uM7b-$N^tx9O&W5w1XncNCB{D!NlQH?MatdOdPxy!j4y-a z?{ndJ3eO|9bR{P(RbY0O3XHeyW40{iYv8e^B<0Z^w)Nb42wlvP?>eo74QsxFBF0?D z-W)u}GkT*tl}MEKdY&`442OIN>a!{eFwg2D#y;7OTSiWkpEuVL!{6J{=od?>`Mvb_ zp;ch1bc_V&F0j%4Tnl4ImJ+O;i~~ZU+|=hCxGW%m-f8j&=~-v!Bp*RMnCAjchb7>m z%TwaI<|+|gBaA=%PouW>VJ?Ux;}2uvhy;DTHep(nXdtn{7;0f z4kH#zh48nsI&5*B1onw8L}<1@JZtbGu4YD%8M&ELnC*_Se=3=w0}UAdJO#u2*P?g% zVNSzt66e$*4J9k2anx@HsF{V(Xk$ebH~CA~XyxGOeFd`FT>)d<oTbG07CSr5%|baDY2`gk%s4(ufFNgI{Q{z^Tx5mVeR!Spq@kmMB&5>i)bwuMqd zTje{}e7q@jmy(13N;<5D1Y?&Wnneb#*K*&MiD09-68b--%vmySMR{>~!Zsq9f4)_7ZyV?h$_<%nYc1_PHS$0{O z`+jDG=Ic*KgXW{`nr>HE@1g}}HRI4>@^VgS{1Ns|4|zZ~>o>$uGwtGb5^;7Q#46_E3eMc+z^Mk&N=0_u>3BTGM(S z9{$lGD?d#on$v^e+J_aiaMA{L&f|qJ?D~n3dpJxp%5}k9RvkPtONfY{0O@{uk&dk7 zJLOwsQLkUdM!C5k)o(FVI!D2aVyOemgO@Nt#rm-bG9_QiD_NbOlCU@ zQX{)|@}I$Mw11+6*^Po2`kHsX$7d4%+wfG>k^9Sa|V-7CxJ)F`@i6Cu1 zf!ZZYqHu#7PJB6wb9m;74cB_u2Zfmp?PWI32KlFP+2yW^4$b^Qa zV}I`h_MwY1x*e2)oAUx8KBAVKG2H~h_Jd5;)9a*fgny5GrQj^H1BIW@$H#;3+2eJ) zN&Olvo@3KW1ovJg$E$1D)m3vK-j4UrjXuFbX9c{_5Dliz$2bQ%kyAVGM;>J#=JtjD z#Um$rF{$?=&!P{aRg#r7zAl1p-yMgGrl?_Ytqo^okU~u(+;}JV8Ls`!WSH3bKPH;?>3H+05UF}piPDmjIn9|;kaf+KS#HWvp_SJ;%L;Ey->HPs zdcVm_2};c^t?}o#qn!4e4eYFTI}C1|!|!oqa7JXiP34&!yq_M5&Uz8}NLv7^hBC-m z?{Zk-HHxmW+tK~A7=G@krpE(q=(LL|5ODttDIOHSkIDwz{Bhx6z~3jtYsv@-FQYvj zL1b*ge+}{Zlw^gZE(WOXJTZ)Y8gf0YVz2yh_;6Nfz_)@ zx}-mowD4T%XRFkyU5E)DS(psUw}Xk;QX$oa@K9^-RmswRfj`ny2;=BKS&r7br zF@+OkL?si9$`;_%GnF`}SgQz4?gUNK7|tg z{V)f|?~K9cJA+|T-8QEzRVwCa&uxf1gZQMA-Po%K&eNM{^w2$!kTisvap_=G zTtzZH`1_Z4EVMuL1&NKfXz5QYjLkdEM9(}5rgr1WyA|~`6rK>7;aiM)##1bG{e{a# zb}@6@D$#59e!TnZIqT)`g;SfvLH#b@Jr87=_Rb0@Sic-xbWT74zkiETEX6CGYGA*7 zFWlPl1I`+n!tU--A{nYpcX4s(Q1TnM2p#9*RrJ+&K> zgxjZn4+if{m$B7tm!?4%`VxoT~Czc9{lQY|+J5O*ZE$?}XsMu`2%ga*+gF z)8mScXVKr%!stKccfIxGR4&>^0=|EX;`uauPsmIg|1Pp6-Z3%UY^#SvT>2;!7pvI5 zwjXBHeIm$WzB?PkcW?K+Q^WQ^J?=@!c`S`~0C6EGSl;N2?z_+6 zlgYPfqN*C~tmNpdi>a{DW)w^Bt8=q=tRf3OeqergcXKgw7)~hMo1V&9g!&tGL7=Ib z$UnJ2YP>S>Clw=&yrU<0hCY|@%$@c;iY4n3lt8I~lC8!~)Yo$_w<6>K*Bl^bduK`^ zoJ_w$W(@PpoH-Sob2IPO{xYAd?TN!BWtOa>%4)Dp?IvE1>QMPa2Z~dKh>@8LeCQYC znT%_oasC#v_@)4XTN1X*`m1QBQwv6|48&^EKHJXG*QC4 zR%Wzt=`D76S2QhL`+<4A`2pVj)5zWSIDv0|*WlM#a=S{>F6Bq}NbX4A78wLu>T$oV2+)HbI5~RB2NxoCuc39p{v(nf`GYyLEz8 z_X==+XR1MA#&xFWO*J#@YKkiIZ|E#fJ?37g6bjB#r?USxp+~R+7qK}IKPwY1>`XdK zcOSubtNW<5^;O82LD8)>7lVT);pVsY&?y*83S8`9>BUae72V3Ye0hgpKbh-4Bf%ZE zNJUn21N2wT~>^ut3R0nWxu{(ZKTF8xxNuhi@ z&*EET%GC(AP}?5`M2q*C%&?T=xg*o*>(*52KR%v*)-@&B)AZqh%oftB(a5w*&LK_v z4s!D~1i0i{ioeDJ$)=-{WX`b!JXdPMRSNy48yW@4`W8_VWcrgp*HPv}z(e}DJ(*ZZ zOhK6)iI}z}gSE?ZhX;{Ai9cfurHMyj_OfP1(L0)^|20nT5bpG+j;)`PDR`I3ngv8jWM83AJA7pXKV0GU5Ox+vbVR7bT18T5eu^m$-%m=twQydj6=uz+B#sEs89`3&WZSP;$aA@S#5p+|);H%TD@mmK!T z)246f+`mfPTm3NZXsvSNk#=QNJpYYs%8sQe4!zX-fDFx7dP%IhV|Z_mEkJiGef*xm z0Mt7i%p`+rdNh%}qSPD{Z#R)UnyAb1_{_3Fi3vAl#Zg-?4`sA5J5! z=pda%fejmtkKzj#03zm3k~e*VS0-(2V^AltxWzN?HO1*IK1#B_`6PzE$ zn(lLGqVgX687ylmPPZp$`g{$}6}RG+*&B25`6{-hv$Lpe-eu~#`UzRRO&Sg=v=gh= zUi@Xf7%j@I(6UYt+-3{Xk8RV?K47QK*|i_>m2M>W#$!DRG`Gf{;3e?zS0eVkPovrM z4LL!*A;xRkZaQq3PouJRU`t>#sSX=uUtLaxt<9H6Zq01!w`mO~T}s9n2Oa7*W`UaC zhpC+l&l>!r%2^i`p@r>DyfS?v-VOQ9u3B6})rQ(x#qthfX;(mUf<_qcoL8*f?P1*i z-*n3KHqu`*emG%J7~k$$$o1-Ma6g|pQC%KP&DI$a$HU=t zkJ)XiCvcnuE~v!1YDZ#gmdC7MFEa7B6KPF!JdS^5$m!>l;D-8@dDON0#<9_X>w4-D`E6ljSGuvSL{6RBLed7;~erIU8 zC6aoXByQ5mKTPD)%gn|Vhp$L2jG&iS+{2r9+@bi36s(`T8df(>ASo*QAhF>nz9}*1 zB=@Of<0mc1lu{rW?*#C8)mi#Q-;mBIFCe7?=D59}6mEN);@Iam%s7i!IHYcfX&L!A z8ZASzc3F~6&(lQkgAJAk48qcsV7!%6z^?f#Pd4nE4)vj_*c0eTRrL9Ho23{&YrRj- zoBap&X{Ts)(=a{!c)CsZt~@fBN{JZHP1#&5!L4{NM8B$9aa}df$sL>k*ZIuXB7USV z?XrXHQjkP_VF8%_wT%o4IIx)m9`Nqf6S}7U9ufWJ1(H>-$Q>gos7O4DJ5sE`^;jA? z!3#*GE^nn-4u)_bM2cM3H^Q>nOSw^xaYWC`hdeJmivBMnxg@)Cd^Xb#>;4|Wb-51g z^r||VbJmtK_%O^WR0Q+fi?0o`uiK$-=4rm8-3D{tD$oa8`YoH(M4Wa+)cr+ayBW%NQsJMKCh%Arbj+WbVN23R6 zTVDXagnpcH{u!CLClKM;cDRJ1Oomhpx&AK&=V=bm#jAUW&-zXpo4|Kr!xAz4X&^J` z-9~i$e4x)_CF+hwK|}@e`w(Z&rr!YU*6>|dcYddD(7}q?C__q?ChXFj49WIyXo2>7 zcEPR$IAL-O4(AhAV~#f%J`w>5>u9ogKF_?$i)iSPDkmW?H^Q|jC$@I^5A2<&iZc%1 zYVbR7i56+**yPMnrn1Y&qnuG67{D#6XQ~4t*?}bx^8|E^>mjq#!{NMS75U@8354X!XuOFxnoQ+;wF_IYGq8~wo*Pfc zw~eD~j}h4OP>VbIXbrwUtOIRZ-;mGuaxkYO9gM{)FzC8A8m-i(ADgG3=`KH&BfalZAimxS7`+9w@cvW!KDB{- zyK);FZYomuP5g5}Ck^KD-hmrYCFEFe8CGRar)@W9vTJHGaLwB)0!u`ITQD0Q-B!TE zgy+?V>48RE3L4ryMCWgtLBAs&26s;cfp5B~es4c3HfuS^NY$dX)+!WwZ^->~wB;Vz z53-Ni-Z1k!{Fnp$z3T5KWA32rBQpO-DE&F@7#YpHPM*X^vSOZY#OtFlvtIQ(%bBdi zAah}I{o7&q9Q~KBUGRsM-^sF}mI+ie>lBQXh~U~jN$3_x<(d8SQN?=)EL(l=oP(|{f_TsP2iYYu z5q7q6sQoyP*{L*wC*@Az1WR=+%daP=wCdoEupKQ+C;}g;3E;O`fNK>f=KM$U@kJI( zL!ZyYnBBIpIC4B#yC*_{ln!?GexycPXYs-9KA7qFkoooVAL>Z-;gdZPxJx~i-d`67 zyQ>hXpZ}t~DA;TGFIfTFllQvR2p%Jsq=8}09>bN**Jm=5z(dH$-V1A4& zgzyD{)HCQM6P2DtW(H>A*`e`xroD@3?lgrf(^OeiVp?F?pKu;8X9 zKEWe9GVtd9Rx}=xpktqg2;DRZ_WX5*PxYx#VcCrjhbH5q=_B;UIvps;y-Lb{T_hD` z5+2Jvj`<3S;F&OkbG2Phy(|nNA$k(ukCG%oZ}oA{?j)OyGxK5fWl3($XgnByUJGZY z>oQf^vx&dlc2q8TN)-zlnfS;YI(DxF7ay7nZ38!n$Sfal%p0H=wWCPcgXvghdLO4g z`HVZng3+;YCu+ZsfE{@XxJA_uTml3chsT@Y-WoT|KA=wTDJ!IT7_jCRSF5VMs-Z6=l_n3f* zTYvL@!!evCH6P#GPJmOsp$*Pm2dI+dbYwn<(#xm!(57BLh!s+&7slPEucjV@1-*)} z_s z3=%Ng3 z_||-%#{UdQWmXleWYwwB&3T~g`I3ZB>!uG~ep8oslhE~r7*(l|=iYGsoVJ@he`kI| zQXU+ouSW)m{MroI9U%ks zyI~8f^R9$QUVKWs7As+%eiV%rkEbIG8+cdsAU#%-K;BSaI52A%+?RY{SD#a1<4&nV z;o1x`abE>V_g98jzQy$OhPm)J#Ek1WUymtgr$hLf3L>tiNwm*+kO6TyTpV-R#$GlN zBseu3D87UmPVV58eUSLNm(xa@7udfz0zV}5vU=OA@l@1I_|j(xT2DmC(Jk>1v3&{8 zfKGz(C=Z)bq~Neg3f2!>QH>88QakybXL;jzG--u z){$Ty2*`zMd^2(r%G>$4!;zMOfm^6k5~n7OQP8m9 zG|ccUh0(XOhymUw6S6;G-;I?Zd;UK8!oGyxzl$1vw`s$R^b$JT%ZQVGAcE|?Dstkg z2-IsU0ufQ;{;M1zn^G4ucP2NH%&*Dhwk}8SUy-sMpK+Y|)uE5S#L_rf*Hor+UJMk+ z=fa`QwWxP!A~l;c8T>XafEAg!HWA{MP#C8|xwOTRDxp-PBN86@A5@v z=UXd!*fbly)_So`Qx$o}xE2h~^~J=D0Dc~ygyR2%xS!{1sLppS{L`O;T^?((X>|=< z{)%@emTb3viC5S)O9)q8>y2eM8@PxR`}!^Gtw8DSWx{l-&%vCr0y z3l=O#h35BE?W8r0T0D2ry&U&0W=Wp?`^{4 zj1&!kQmT377PX4H2{ltP;ds$)7G;u{kn%tj{alKRW(c4{lK`iAzK?Es`l?0q6u z^BpG)4CCZ`E%eta5vX2$-9|?)hc=JYkZ)0OjGOpX8=KFWu*aVFz_zOJ`&tH8bjCr( zo&;(kd537{?#N1OqDvNvvZm+O^b^rxxqLjf^q!HP@MQZJwyJR-dd1EX2nx zKj@DmHspsU&-HLL1m})ch_pz;sp%)l1jwe(HtxZXE|=)k#wDiO#->{At5eow&gPYb!AcjjEVc@Zk@GNk=~mh!%{PNwZ{EB@kp1L~vY zXz@%9BBKPTT~;HT-)B!lGfvXZa6LMCzBjZs|0BoLW^(avs!&wC4P%zK(PFo+r0Iwn zYPyyaab*+gvUnX7yk1Q&<-5S`URm6;r;TitHbRr!yJ#X8j=ooQxo?V-0Hg$2-ERZv zmn{lgTDP-;T}gP;Mx474(2Jr)SwvL*1nyZ~OsCyh220EHNbx^CxV`KiKi|oq-`{7D z8?t^7+%*?|@&2pKU4rm8vzVywxxsAx>ki&;#}PH&r+f74KJGRd`@-WQl}ml&Z1*XC4ufT7ykNUd)k8l?e!Zhjed#$Cexje9t=&@0G1I7kfNXQ0-Xm)PMH#m^O8xUS=O@cG7%}y<|<*`8(ld89ieBFoT3>JYjw3 zN0NZo(-`&dYILyAgq+>i$ol80b6Kfhh)3ghGU?WMFuJAyIXh3#nQOC|nmYxsnCI|5 zbk66}&RgSTk2dz_!d`y2c8}HD7)Y{&vdLg_C|)|Q&gvFQP#@t(#MZFR-e0MO zCNI};w+_Xk-@Q}h^Qv`lW3ZS}nYxquzxzR^xQ8Ffp9AntE6}W46p&=I^IUvfFAA#tp}Fat?+ZZL;I0M+%ULLn)+h+(8KL z5yW9hQ4+g36`u0_dfk64k7c=r1}$r0ruSW*X&}ZOR9gY#gWc#eKfaq8d6;cYnL?i$ z?;#69?l7aSjx?k09(6p}yWpaYu!}O8 zSU(?Rm+(bz_h51!$e`+%K2%x01jF}Sz?NsrIp?*m@NtP4v5u}K z_Z_P6;H(>{+NsHPUCSdfk1rDg^H!>r&ihz=LPbnWGxPnee}-fZc<$J zhuFKP(*R!=82bC0^#4$1oZN1r)p>s$+g(LxTpY#$E*(X>tmwHUY0i3XCsE$2hnw~5 zP+esbi7aTOtDQwS>Cab~^OqD+J8CuxT32DSiyM4jRCT!Ptw=o zQ?clyH1%C^jZV}y3S_+BO zMg=A#f;%q3J&-=vnGI*zX9(TvJ5FWQ?FmH<=!>yZXZ0Jp4t~357N-fZa?s?1D%XzYl z)Y}dejCq9NO=KyOn@IiI2$5bUo38pFWJmgiLkj5vmWA9}UxtA_9jAzh|V@zSP z_6@Qxn}5OkDJL(;Ai zJn<+Bt=trN_(Tl@6;!~rHk1pHI){mFqp1I3kiN~&gx@bRsRi!_|D>$~ z%Y+hHkuAb>p|v(Cf1?g1_G>{n_%um;C5o0GPJ!So6KGPDM1i-S`10R2T4a(ziurtA z(B#|X?$fzw7?OZOc?8Bn*O6Oi49Mxj5&RrP7NtL=(W%PCWM7XgEZCe(6)&zt=Po%; z-sCk|eD?|7TFc-DlUTUzt_=-+V{Dk35be7gkD(`ZG1%h|?GLv=zWPXKp08joj*F+K zXRZUQDGRt9p1-I*J&lW+TZW427ihX_Dy%px!1>QR#q$PaV1xft$ZVT{P#cXRqsvjF z_9F9FZ;0x;%QDA9#F+jAd-3nqm!#;i12kr}Q^Vk&xcEH(OgHDK;a_ch=@`I;X>Ngx zCF10&^d)?FDFO#~&f~%itI-j7Hx1iLR8D^Yy%}zt-Ms-C6XwmGj-1UA>%(||^bvIr zt_EpuTXJZQ9QI!c<07x95F6JF68&tYHDG=dM` z4^qJi-SnV6B@Md;Xwq13!``|TAaSphZa83uJA>l6{+NweBdQF`e)gb1y95ZvRzXnC zW#V&F2gbKb;DPp;+<-?C?dJP_hG7M8et9yi-S(T_{$oJYhTGVbLrZZ+Oab4)xPXh_ z$Z@(%EwgiR6HOD;z*65?P{aI38WhKHOlmuQbdO~Ni>vAOs|KLsZA8t|{}H)M@~|EF zKI?W_sNVP-=M*RtJ3)TO|KD%MXZ%+h_`H&z=|)h)RU`Cd#7ApMs}GF3dK8f;KF9m9 zUt!Ra6cixS(6&JnXiE`U`%xL*?h0Zau02YE@3p|Bqx0~!(0W+*?hN_#KpoAO`ofPx z8_@6Jc1-+LfeRCE(0HEjcR<|@Pg?n-o7+P+<=tPVWA!@5Z(=zfAiQ%{LV^=XkH?>b zm&nr7Naou0GS|k*;Z;5-(d0D^Jp_-C**64uZ{q`upu3o*GDaA8Cy%TYE2G^L6}Wiy z&(w0weKHa#ggdmm(e_RS^KU~t>CSsY-^^YH8bTkb{HsD-=A=R0mEv%``3OnB?TuHFmLw&K5)G zh6$uae}&DG;$E9qU$@Z1dF8CGR}d{1T1qCH--NKIJJ{br^GVCS|3KS76Z69Fv8vJ@ zc+K`9{t^`-1+J$ty+IdD%Wp9g{yt-tY@j$KX^NZH%;Yl94RJR&-JGB(EKLJu(#IAWXLkhuEVtyE%o4F>Ml-5SoDZ9-q-e6_ed79U zjMF9x+>NLwc*#K)Qu_pn*n%y@O+ylz&nSTDI!!1)kpW$Tlkn6h3PBgYW8riIV(>E` z`(LQhDCJ3Hq~#E4pIS#|c#Ws(KMToeK~0dbsv^6xCg7@>lW?|KHWk}*3XgwYNKd4m zqlco}=zVQ_jGF8XgPLb(*@Cz9pnn=}*AV4?3NcWz`z|TB;(HnbbJ4R%nHDIg@%bs9 z|05hh_MQ>KX&ySTE9nkXdq)V(YBtax_VG4PKi#4;RchHwl_Bh->1*)(6*ayilf>-b zX$*tc7vYl!}h~hwGio@Xg`7AiDe* z3=i7d+8j%U+))+Sce9*VSqS4R2`SPp8jI~xg{Ts=lhbkWKqmtw{H{Az-?(-ejBK)_ zv+I{~QyPk>!{sFO(i6pDKFf1GBo}*!q__!32BF~39X9)7JJHB}LF*DPQLQ;Wc#mf_ zUamY%hYOVHrvToeE@HrY81~U@Z2`K~Z4%^8inR9DOC`Sy?NR2Agl)C+CzA1ne}_1W za*@%gWb*g~y0ZH$xIW6n;#VSYs9J*=w=@SYy?=>DFJ4jiTSXAQG=kRJzlD5*O_&mO zlsP-J6Xvg(gVVz8mJ6)3S=j3w$j*V2i>n*!v@zG)c+B^n*Fr-De9+_}$gDp#od;OLAPQS|C0A zNCtHZR@v^6_{l`>J!qT#-4|bJ@C>A=4qEIZNcYTn0OxXM6SbvUwyy2NMD%ApJ!Z3q z)@aW}z1^=MveA*2h3R1Mg;J)cB!sTee{NHnwF=zAr&5!W#Z2+D0xS;u!AgE@Viqb) z<@$d9qvjg2lsy;9dzB{8B)5L@NW>qePyPw5{I6Gx@Z#tytMPDtb^4H;Tz~+kNFg8hGd2T61`w@RqFX`AnAjsBr!|S4qsepTrDW zK<1`7oSgWVE>+WqI=dKxTHnZnr96Wv&8}haR69-d6T!QZ;-qcHaUF049V$+Q0s z@bke+TxQ@&EBF&mZtHTWNb12rS&rnM+lt9PV&{%Jrjh`*lRmhjSm9-lytnbdxZ3yHp8IvD%_nLaWJ*=fema|BYP9FaHJd> z0`g4oKaq_X-8Bw1M!X<+$3B?+-(^Ng;ToF!Y-3LC9-%HQ!6)k%*|M(TAT2J))p=Z@ zu74fK$Ma{&$8Z&VZI+ILJ}P9~5f_`roo0}n_Ja0AMM3nwgY{CCHBiu7$aG$uiR;%N zu&v;80I5O|tZ=^-TI|&aftmNo5vxWUH>=ao8LWZ3mt|vCXdMaVDw+3|t5JHl7HIo; zz*<=|+jj{k@cF4D$P8qo%OY_J$EIhS1+EGT{O`|Yk+6t>fP{p=|MO87uoPIg(>EY6 zWP_RCy5MzYR^}F#v!+j(;CXB=NLvKbd&1f%&u3tAr^G?Z89CdqKmi&uFORwp*ub^_ zW9Yp9vHZd~ZW9VwA*DffAt|17-BGBNvZ4|V4QZ*q8ptRkLP97cGbIuco^#!1rA3OU zG_)uqMN9I%{{laFJ?B2>x;~%xyUzx1+EUcK5y~IE-@(u50Ia{3iDvqKAkGCR+7fec z#yv+;o2Ehguj;X-;@M#NY{WeL%TE65Ec#20RW2 zjr&NfJhzc47gbS(rr_WAskkI*8svP}s;VND0BA!_PmF@9q1`C3ZXKTBb_W}SlhGjK zI&Wgyf6yN53r|yzfsgAa_*n56FB!+tERPbjlk$Str?RAI^ zbk5`WsI3NO>?nbx*~TdQ_$o*jPhySdDBzz{YvF-VHSd)+_AieY2k3yh@EuU)zY^zwppP_AF|v_F+ea2R}=)pDs+6WHV#U;2Qg#HoRhSdoqsiRK==qSMdFm>FD)C7M31phL=ZjdBz?8!31$r zc1751%&;k`{%`pfRIRngv#Oo6E7YA~YLg*T^fguDoCJ%-rdHol55hpD+02#Rd!TLH zNgL&6A(F0ob2V<6CoRU`B~5T0f#(jo}>qOXt%>9cR^U5B#r8?vr$gK zgUAUTLAT;7C>H36UN=v{ug!ZH*=ZMG@8n)G-8vmKxnW*FfF3%DY=pOs9i5Jr<)U?z|)Dl>%4q&qChEfAn;&7EFBl8BNT_XtSdb zl|QOYRGe3W#(#fc)wH$LDV?xl?w8>qc}_&KZCQnP`^lASy6oW(ztP2+o9iq$Me&|O z_&8Gq)VF;A*#&dht<#e5Ivc`1_X!7mX%h(EWWa15uH#pA_u~uq&CD5%e^iV2nA|mq zr$6%jInU!X95S&WU!E_6JIXKko4ajME8Y$s+Uo-kbV$WMDJrZKfI%Pn=nBgf)sf)? zyleA=$cC5t?A-4XoP*P!?b`4V0&UOWwhmSHwEIT(CNBg}_Z^^@JC{IM^;3GnSP5lZ z-;f<+TcN5dv07*GTu6{H!XEPha$J#fghpXh8*23FgFR)?l z$!h7B=g3{{8oqoTkaSahni0S#P^DKFGQfJ}k^7ABqcO*P3vV;PA zcix@78!`IrP7tm4#xAYJ^hu2&yZBc&$vJxt&M3RXm2xR8y`>F;ZB2M7Kb-U^DPW6> z8J&8n2A>3Z6TyQw@xj^4+;2vnXr?C-b(g*HisiU-tKQJ_Bp+qlB_V122QaQ$P^hqw zDpd>Nli~T)I@XDlaPRu9CF%5W;Xyhf&xv>GWglLCn+r|x-E{KTQAk>Ef=?gX!O$!p z@<&G!5BR53x42el;@pGrL!nT%P#s)i_Te4J8X`9IoGN!nV5fB)UH^G5ytFff#*Hhn zEKnKGY)ixaeKNQ~Bm_RL?&076rG({?(#(+;DzLETfrXi+FYo8*dUX7?fR`oOi!%mO ziOHHpXq;R`Tf$P%<-QU*Fg%^nC@kdVJ0D=a9598m`~Q-xCPj$q2!nlk*YTEjCr;da zn6Y^Ai?8rz0u*I(w~k-Y!@z$K~UAn@%0(-ic$RMSnAl{BT0U ze=V?uj$!rzNeJ>dNPdnBF@2-gLHx2jle%;<4wU^QWAnq&R;HeuH)FtWhA~J*GDKBG zk@va2o7Vp16MQ9K399~{PYnetd?Igf3g z5zd!Zyn-t)388h0Cdkbj;l~CpWe@6b-0Y1TLF~Oc@SYjcE{<`yJxGSg7;+n5ttGgt z>IrtIIl}p6m+4N+R`TXiD_nf%MXq1l%J1DF!kq1N#+VJBq?nsarl~38hn6u!n8SQ% z@Pun!JjNw#H$GGm!}n2rWU#u7UE;{B7&wjB8X`!2 zv@Ls&nT1FD{n?nxQECru91|@Aoc{}8r`{NV_mj_(^QR{==PSSSVuC{%3B7J=?xagT z%ejHJqA=^I>PTxenozs-7HmIPLW*5(llv=P;aEd6DSoU+!xTT^xo_#TB2N%l{00II zYOs3yQl@m_6Zi`;(6Klb z3>-6H_LDmhu4BmoUMo^*wd|Yl40Ka#;VLE;KGbOro{7k=B)x zSmu#JbNnB|>vk`Aom)xki~b|QF_xIO&>D~S2|=O?uoDg+g6ZjhssbmSL0@S;Zj1GR z`bX1QUgIuex7L|h4ODYoncH-5iW&-ORq!3I2Gb3ZIb2?-7nB}9gTM(&Oxpr$xFX2y zme<8W`^KlRclZan%HAAnc2fk!^K}lQ75(@&$#=On&kojyMyqE^cgC3 zScl9=l3;581=C$q`iZMfEzfGI7+y(~V|;vNnY(=rboGZ!Xe>>}-iH(5iB>4yzh}Ta z@HxnDe{IB=6-;8+Y|qD-#2>tvGH=KXoQXcA7pv`#m~eX>VHEx9!2TGyg;(bBsl4R^ zycHiq2zUR``fJFZUW%1v6b8q$DJZP)9O_m=Uqz4&wHfZ{3 zi>|e8B=B?xw^#3?D{wbabGQ!s^4sxD)(G&H88iCkE16wip2L%Y*QG#Ze> zJ0%fx+v!vAyHFaJ9Z8~DDz`0abw)7fX(Vhud8WE>w=_;$E{)||WTA8S2Z$}7K%_a= zXBEfUQ&HIjD>tmiCN9${RD6p%9?0Yy%P+^iTLIXcA_hHleHyEUA25UBj0%JP_A{P7k8R!A(cOUSUX)j|sLPH^Dbd>&? zX9;2LhEVgs3l1(WhV15bAhJw_x7|va_5S@7Q{t1!wo_+uzt1uJx7LvED)?A!vQw0O z(-=(bz8_{Y^=84GO&k{~-In-s>>!^Hhv-myJ~zAU1KXI>wANRKag&OL>`PC`sV&m* zHkTz%H*mQyR;`Zo+pkoK%SW7YXx~O$FG|Il1uf)orqDRG6{;n}{hk2Gq1Y0(A!{ zvA>$YuW`9Ro!_p8@_Fk)nU{bg>{9p@9!(PlhR~+X9%J;xabV^K+s9Z+uPQ@IE!mdjmbHOB3ksynMRamTO9j*g1G1n zLEYj>blCY6sA;%Ch+PpJ%CiN@l8exA;Skkmx(zn-k20p?+D!YjyLdMKC zv{%{zEd+F#*@1d^WFVi!Hg9JR3aK(OZu(@yKTqTEDAA!}HS+ZLUEYt? zZ^@j~mtp)$I`gjQHGa8#jQ@1bFaE=V`OMj?FKMMu0&PAm0hc74sxc1bY&1z?p7Y*CtRVB|wUXHEOj78Y zfXuRBewp)Q;D7i^PT#v*wcuzJlU`Z`x^nC2G;cHZu$~Jpp8uNmtvZW%_99rdu0qjv z9Xy;X!s>SRlO=U0X^fyV891T@M)O^1c$x&QIUtWMKGWe?SRUW9MW=eWF2W+BNfcV% zWP#wI7`wFM3|w?CBC@|u0NEwQ-d}x!s88qe+W&RYXm=f~7XN|QckQKfCp3^rJ(7&j z@q9RY#TnM8OM=vl0&c&c3afgq^R2jdO6!pkzIj0oe7Fz~%NDB9hodFDl0EJ`qv>n; z78+ro^-Ps13soYmVuH|;9}n*yg<;cSZ9G)HjP_S&&?3qCkab2HJe{vnGYxCpIBLQ4 zrI~`WKm^WxA5VCrK@j2<#de-fq-_#WR9!TTS01bd&)v_F(avR9*_{fqff1;_S`Pen zRX}m-OJd-gNvg2`mu<1v6G3sX={zLqZCD1m~aT|`@M4$t$~ zEcoozi&1AkW7uR9cFqDmFLr%9wx{Z``{xLsFW5WkOFFa{KxEh!JfogMRyqeS85^8@|DrS&g_@ZHRY& z!eX4fM+sMXz9l1H2zA1pa9lJRJDxAYyQiNKSxp5nSN4WLp*D2+yc%<^-J!mFr{G9( zK5?!WfQ@`1Jg{pn9!nv__(d30aNl5G+Y;=s_>BX>i*TtxCEN?jA##2MjHHsV^pzRa zzI%$;a|3O!!Ad$&O9k!qrckdGGjdU^o7X7e!?PZ~Mpvt>#N*1p(6Dg>qoinyEc@@EG2GU%g618nOiqG4 zGtQ{88!9D;Syl|v?Nw-_m4Ka5*NA^o4^RC;0lazT3777RqxAI)Xc%KoB?lW(sv`;2 zqomQ-WFwaoti`P+$uySZ4UDTOFyi;PT(|!T=4;GHToC$+>dJJ(6y0ge-i?89e~KNm zv@MxjJn+07Zw~jfjlLI*yoUy`41SM6H&_0**MM#P=tp-guQG-jSTn z)}{Evqz-w~Wq+2!>1Oz7UWn2EdLgx=5FfpD=Am#W&c1VtdaE~pU6>OCJEt;2N{-C+ zUoZKwZMMKS3ILvr3On!g0x~9b25K`_a2&x3;+ok3g@yGX^6?@t`xgV>)RaB?)&&=? z6To{LI9AW+99ZVYy;~#nG5!$)V&)y7<~Nzm`f80wtQFX2?~kF)of*V-?-2<0k;USQ z)$mN@3iv#oNK7Q{(eL3AFrRk}E^a8Twz5!X$^;X!Q~Lo?Yzl>v)sl?(|evOCj z)neVl$rhxph4)K8o1VEb7o*ZT^M^%#dweTR@w4-60O6=KQk5 z!i-pM6r#yRyt=;_a(?R}e~lmG>#?5wcTotEUL>RWIR&PC#UM%cvSmyoE8t^bF8;k1 zflJ=Sp{>|sQmAAPZ#0duBrb|-?@NaiMJHDC(FGC|!eeI^)L_?mF&=*T0ah*G@)tMa zEH-`Lk3Au?+0WCJ*b>!hqMQ62Zr+QA>e3WucidC>@bMp!J66VVY>n709tkL~BgM|; z95EKIr+CNj{6`N79%5_lGO+1CC3rCD1zEG|6UjR}mvLTP4*fUspk1yQO%ByVputKI zkynH~t=o9|^%?q(%emtLO>)3=GyS=ui$D7C6ZC(-Mnz_;Q*Zk*G`u)~D>M2~{l!4GbAt!ptU{K;ozlc+b2^T4D-F zz-bp+uwR62oAQK;lyF?v|G2iAnm-8bSjbBH88K?tO_}tY8aTGkhP98sgQstPuYS2{ z17rV30cyE>Xak9(%{I;)>%)ktQc2})jh_OQ3HM2^&s|#a;5GRna~#J@#hLdd8vKN! z+fY4_&C~H3BO^y-q4@p>cx3(`HmsXRQm4n@s{ljBUos#57=GrMnA@n%t$u9m4!J7C$;?<@HX0pWh5br%&@b%|U%P4_x-~3dcJ>MJKc?*B z>#SE~2R-Yd>xnzWD0yM>MhQ4!#DeIym3UQ5fDO&>rDJVBKxTCydn?cmzbkpuzGyux z7b&Jf(p+Mxq#Vy)W>7|T0@L2p1xII4D1WI%JLiAKZ)p=)&3_^!;*S#BILiqOZcJl* zR)*7n75%XDF+W+m+9rI2+$axCUdjwQU?>#631LUfG`!nSlZW>Nkq%EJi%ShMUhwug_|pLrnra zzc0uZyCjl@GdSmbPapUDp9qhab<=M=Rd_Ua6-J5N!)JG+aNF`=9FEpxCDOUR`Dc-PX`ta9;7K?CpJ@pjU| zI|FE*%E-RUW|Pb1n0>wrAW8TV{c*{HMAmI(M-uhfjp-uniy~*%$h5m!VeNf*z4baC z<#=f-O$)2?>j^I5&dW3B_7jhk4{%h?81l}XAj2|>xVG9G3Uor4Unb-H#G_|BZq1|>f-Y7i`kG@2ZZ(nzz zjzK1fUEylC&$OA$)?C~xZ9t_AK9OttZODfe--)5|3@~jSfg=y4@z&Bw?BG{l%(x^6 z$@~KNSZ_$*c}7#$)PH2k(Ns28{3L$)ehX#o|HDm7Z=>;7GvaDc&Qe(m6g0XFXE?sm z=tM_m&b|P~J75UK*i_DC6;AU5%rS4zD2XZbaCsd^w;AWvL(WR>Kbh zOGY3#OOEc+Ar(mm(9y|YK7WS4KL0a`x#&@O<-q0Pd{u7Y^H_+?P4E zQ`E!Q(m0N>9fkpR`#@ysc_J;o4Zp`E;RMA5JZM@=ocl~z^{RHBRcZ=pouLmUb1KYV zrc1C6b*JF!z$RSxJc?&&x`pqj`G)>JZUuecm$Q?9STld>w~~m!>5SXYc-WbBhQ6Kg z3oc#%LeiI((9xAwc*5o>ATmXTwVtsNt?t<4hd3XS=yekn#ez^tq?iA^S%nc>dyNRM zm`~DlQ<*h4=3uC)0VK^mjD}HO_(FdoJ8w9a{pf!l1=C$fy}=6l${-p(h3^9UeKjB) zYK79=EY@>%8!tel8`jNAq5A2j)K_FNJSx&9_HV1<``0DBunqY%;~(dJy>Nl`Tvbf% zYwnUL^V^uW_y#}SmeTBJgCrzzCCO>ECF;6eu(56*m8wzlPf;VsNc>_`vp@930Z=&J&8kFS?VNbCF5lYy>N_O2PQde!TfXd+} zu{;#=8zq0%=8>$qKlmZF`*?e`FOw-#mVgu6i2Y58d~ahNCP~PM8Zuj%r>$$q@61)W zh2!D%e(J-48@H&0Zv^Bfmq5nLI*X?KKAv!^23%F@r{#sU)H^H)PYRYpQ;sjNE%6-7 ztPni=x~b-|$I$!a02r;!0^7bl^xPee8y*{tbGYBtOJi-Ym5c(e=gaS{Igg=E@w73> zjCPIxwD9#Q#W_BGyo0L_VeIw-2$y?D(%L7&?7g<&@$(YgFq6QJtj9!Ld=~qS>tvbz zdPG+`c{1gZqHwzD1H>5%U?i89k?!Kr4IN!LG5kFJ)?9|0n%?j;TjS~J-7ApgoTgTy zpJ@=y=KGzkL3^SMrG{LlbNB(sEdD}AZ(jukt!g}L8O&CyO0l`AcaUeo^4ivn(DuuQ zC>b*ol1D=z)32IWJhzADi|8`d#jy}zz@zPAt>nfOb697+nR>OI!e5sqP}6;!EGSUL zPiuB!dTt2DZHWbX-UbZQ#4umUj97K8!s&B{(9hBc=QeyG-HWYAUGX+9=RXTvOXEn# zjpxw6_aE7RxF6*<8)4vqKcw9AJkj%yK+`1#@YGQgANi~xM znhMHH9zK+DvEsre$Z43&{@tsE8{Y%_U3VcuOC0M;+jI#ea-uoB+BlTTIs5oPy?$Y8V;i51T_Z@Yr|( zhGr^ZR)#S1P9vM|)E-L5+&j2Fz6JR;kASeaE~Bm%jXSf4V1Zoy!o5Sc0xu(~X;E*|K#a zDx~G&ORU=(z>W-dAg38(1@mjETzo{*9ve`4(ENFFa*x(#>4NxMz`*)*3tPWJ&6XNdL^`v&o zatJ&k%ZmO6re2QXN@Z=HvAR63QG6?i%_-wuN#yQC?XO7u{R+5ZF@W2bKcu$jW2s$5 z1H81lNUu&41dnBkkZ62~w=pse_MEGQ%V~wUS$~{==!+EdD=Q87A(60Vi#lVke+tdE zOQF}ub5sft1oujDFbbXzx>X8n!hC5|n{)@-w|}CqeoTb;2wmEkstRMmX>?8WFJd&^ zh0Jq*3R2}0=;v*Q@I`Pj_;H@fk1Z$RhO#k}^y@d3twX-W{6HwXkxCXvJ_J2qSF-L< z6kgoEkoRj!4j!B`ksVL|Ok+QDz4pOE%xm@`Y-2kF2Ajb|sZC(AQ4KzgwBp?eXPm7w zi?NY>2}xCMaBGJG^U1xNuQSZ~!nj@MAJ|X*$WHS2k2Ncid>56QHi6KC`HaotMEYY# z5)P;AgM8K%my68B_m6W)9_K2x4z1^AyfeYEBOP|dVB5XUsu*;deJl32VmG$tTf!s}PN`$U!G|1>TE$4}li~ z@cV@gB(1c7_q^vSIV4<4Ew|1iHQG9OLEegEGhe3(5>8MvIRQ0K89}AtCgf9suQwBj zA6Q5h?RCbk6-LmUUCJA4r+EBRDH?k*plaiC%VJI6fEu7EB|DKQO z4`RUj<~7)uvWu9>ov#|4^O_dqreWh@9dMB_M~Cj4ut<6*j$RU^6+QD=m*nrb`CK6T zz%Civdw;^FU)$l5R1C(-O9Sl^XP$4)g{_%FJdd?k$d<<|;Fa555Ro23IooBFR6at# zz%G!~17^H4g-*?=29v+{IgjQ=YFC`W6zTsV$JI_?p_vtWdMBFqLl~WDslhi}EQYH2 zns~fcgvmRyl~Hza$8{69d(0v^=8)VzbV)UUlP4SD++Z3!-uaB^cgwQ#cF%yTX-h2j zTXsU{URRV`6pT{q&SG_RDVB8S^OxvT{!jlXhC_e0*u?*kUM8xpvGMVEB1t9Vp1%A-7n0TX!MYydt&+gJOTEG zv=J|*KpJX|L`M-sL}lKHO|tW{V)~YMlV_g>q1wGzG5D z3ByYvqFA&@lihd2hqk@Spc{IUn2kq_n2%G!cxGo*n7d-(NL)`5#$Sk0)9M0|U0~66 z>oV9SNHU3y^GL9G9z3eO0D=NDiL-<|O6JSsq~|r{e#8M%U}b_CrM|fPh#uT3u7|W~ z6k{Kq!}V>cP}M33dmMJK>t#xCSojDDcwYvR^#demz>7aVBFH$}CBa)oF3Ypp3O>Ku zjuTBc;wSmb;O@K_*M4wC{f6TZJV%=?Xm>)H=6O`hf>@+9j_~K7IE|lV8p%23&2)Yw zAC62{gZJqIaHF7z%EKX|U(sCMKXeeMwJR~l`jc>%QWwk~>gK1saUc&%Co>;bs^gbQ z5_r5L2rs5AV{IGGz^^k)p|B+bp3*x zd@Lasmf_^}ALzq*{j_sy8@aqK6l|)zux_a{o-o^lF6;Zqma!suc*umy8vVy;?_G#m z>Pe){stmrfv#IJ=RY+Sp8=?oNW2IOX|5d*-44khd=~|V%78iZI{$d+=(UZ9Kwkn)6 z3&9E6Q5Yn7jp@}-L)))s_;(MK@yIl5s%13|H=n(YA5vnVc~4aJ!%lH#!9#&+PxUGE zk>)ROxWA0K5y+w`Ey0;Hw$k@8fISIPEc;puMr_sjtA5=lkB{WzaOzHIG;+qfUmQvC zPzKZ<7GVDzD<)t4ZL#8tAT!0O1v1~tAltK-kzKd~m>sTEG2V(@FV$Xc7&aG01&&mE zoS031-4}(@y*sEw^;UA^-U3`ZkV5`h-+|6@Pa3V<4C)e!)bQ|L5|e6;ClBnw#j|gt zQdtEYEM#%c(;%|L^ck)d&!I<3KjG~8I)wS^Lo{x=z_)2dL~R{`*1{xu`IHnp(D{OH z`qfTW{*+;UX6%QsyvyhwAjM`+^@Y~=#-wFg7WBxw(*cXGB=>|G_KvEvsr6#CM>z?K zj3zQKC^s2fZ}Rc~>G%l4aYn;faC2qu44uGV-(U$jQGzr~ zU=Eh2pGE&8hf)8XHYOfbhkYEMV)x)fvZj~CH=_G#^xJtbM`RwJJ|V%RstbU##W$Gp zPKMEo%cQ=c>g+>hc@Uni4noJLF;(Th;C%1`(QDRY_VmVs)^`QkV-}Cp zd=Ke-ehNHwzmZ>Ghp4j49#|?ThC3vLn2ZGg zqdEcbyfh54GZ(YcBfEK?H}<2kkpwnX6@$I9JIRzrSSD1+zqwkK(fzDP%SYdktAdJ* z;@guj!7K;NQE;-7CqKn4nMc!J^ZvLSWB$u~WbfpujO=13 z@T>I32g{FQHJ9PhyLtxW-h3mA|0^b2)4SkJvmd;dnM4+c{^U|JPNZ|9JV-ZwMy(+| zc=ALE-*mZy&c<2nl9eiO^{)!E{$@SB+arW-v#s!TKn)r;bKEaOFPv)APwL_uLE-K~ zdc)QRvNca~Ic9%UlT~5pi{FZ{H4awJ0}5tEaX^Ya=Pq6pE&4imSUZ} z)Y*;$K1d^Pfl?*+Ki{N5X+;MtD!&2|=W|G)=vJy|FN;ghuOv6)yoqE(F&OW@Nrhid zfzb|OGVuKcJtGx?pXMcFt#Q7^fwYfAi^~QsdAA2Deog`|>kr#STB<$#4H@|-;UGI# z2}eSXp^x8ha@(bh)NUQeKT{&fpJSqsY}t;B2COi6&qB1!wuj*HMq>J}hSF(1kk2_J zt{vgOwg0{`sj+inmjV`KSg$_%4yhSH4~@a4#hF~J@`S0>mEhb z(V~t#+!CS-%?IUBr)?w1somiH{uj)&>ga;+V@(wBJb-pp572G%5vou>3#+W7EQ04Q zCdUs>X5Bm@@o4iI;yGr*%I)kT*|qYpO=~Y~Fw}ufHeZRZfIg#W-Hw$_96_r$7I!F? z&=0wDsOfu#mu%{S2Gg_oS=L7Ecc(VEmaan~U5oKF@1eu*PZ5de3%t!32flJ#Po%{K z>g8AC1~qHamc`9FGYl*e%eyT$E)fO!aw|xg6#x>i!`QUqjbO4^4TC0!!}~qmV4;5$ zOAhBjP-GP<8*n^Kb_SdE;U@V}dlt?rIRNvC4_CDsNMVLI=-3><>+<5@vEdxuCcT4b zfE8HFpFk~+b2@F>1pE?LMy6efqq0?_#C6vo)IYm{2e%dQ*L>_oXSJnJe%cA~hCE_Z z9=oY^5<7EfDvOdFLuYR)2|BLICak^&N)A)$mQ}l9{XKQKb?hy9Iv5VOh6nKY*ApD$ za}Mk1z5v%9+fCejf?u_lz-P4-jKmevcO!%?T5t=G>Et0<9SJvtE5JHh z3>G&pWZDvb(o3sYvLosk=!Z!Vw&FW!_yX0>vn~<&`(xBgX#l2kIsSPYH{lQoAj`=v zcxtj5wA=j|KYw$k_dz-d-lYpqs$`kL>1!C}^wUgE`;Thd`_uRbndyx9&I#0`@ETn< z=_q7}ilD{W>%`oyl_&Kr7;i0Cr~VHXu?>%RGDg0a=sI&nC{t?z*ii)u_7+fcdLe|n z#z4j@St0WahzoP(NtIabYh}&LxSl`ohd)-!d>x z%>(CAaZr!`M9r+)Vg4sYcJ~nj44=D}+v7N)Zs1<}x#$83&o<*dTJe`gGsFBBFVq+< zFIsKJv2Gj=+Q8caVVuXy;jLO<3wmZ(&|{4&Uv#GqJSeXO;}u%KP4Vf`us`_dxef98 zQG`>U{6}KloVo1qQhM&M4z;&Ph4xviV4|85bnfy4VN+4&$O;AW$9Divy?uvb#fkKa z{w&PDt;j?@T}6Uh-oX;>5PUU*^LKS}JrzlL44!a|x{aEWZT+0P{JaZVCa`eVJQU8Y z=6EFiVGyk1M!v5)4^6|>aOz1RztC(MY0@&m(6h;Kw5yT$ob7|18&%<}lRRkDE3hx# z=rI9`1ZGEc!#?hQ{IWR{j!PKOOG|d(6TRE?!I23}@~=*w)A1|SAsU;(VD~i`57LJ6 zIho9~buWm0^at{v+&w%uF9?_TG{VHQ+GH*5CBIiqhQ@Pin0d7UFtx)F47t8SN@5y@ z+I&PSuLihU`VypFx%s`+LgtTYI>ZR`;m6lf@Upl=d|xi1x%U;&d}j&l9@T|iDOO~I z(-!{kqCs+Q3U>#wDS_&qlbCG@FkRCHo$DlN)nsuzHZ6j<@0$wsRYcmrtZmS$Mg1(({i!w z+1XVvI71WD2c^MdU^)Dka*;MAOviEl01k8nVva`@)~mL1`3Xub)?{M%%@=&X#8MjR z@RHv9sen%o+~fPxPh2+VFiBmxgRJOSL67~x>Jxs#3&n7Uz zXcndx3=wm05yoPTEuJ5p4?*0U$aED$F8fZl_)p9d80Rbeg$2(oGVgZcBD1G3>9Pkq z$CBkoMtb1BARFAGRL76I{*8XD4?>Up&wPuB82pd(L?vx0pn6G(gbrD?tQG$hE?%=v%7q+?UAYQ$E6W=6a>6X1Hvy|c+NgPW9Q|Z&#d_~vh3DK7 z(f*Vg_^p0Sl@6XkcA*W~xWkXEnKqlf@I4k1WA4+tDs#a7jwp;7pP~Q6T1cOvGHcTo z4cc!raOcfS_$}uvk6*Ez7p)f2+Z^xta_t1rH(|l1Q-ukSSO(vEOh{GBA9&p+ z!%CiO0ZFPaY zk})^8n-}((p=o6)yd6r(Jh9lkfNKwKQgtrcDXBw{k6OvD%!ev6SU+OJB=V3-pFZ_wC^Fv88y_EuXI6=|=tz zr2uQ=%k}scqvdD=xpg=mT-7Gw7q6A1OXv(Q_uNVBpSYF{nxp|MB(7QXoom1=o;DzI z_&(e!w_;-c_@dPBVt&O#N%rD`M%XY@5x;6qhp+!i;8=wLoL`p;F~cWG;8Al3(&?aU z2e>(zC4&+iyGyXL4ZIXN)s`U*hZ4Tqj^7X>c?kV7wIOrA z5;MMZ36(#7i7ZIA=W>q1xaW8zO|~AiIFlF5OWeDgC)9S5af#5NuY?ItPw)mLz3-w9 zm;ce(GhWccd1FN9;17!mrCSmIL*x#ExMG66Jasx}i`qD(dkKg79-SQmAj#j~fOM9s5>=*QjqYRn%n@dA!ZYJvWLoogFB4}818xBD)?YaGw z9Fr*JnAElyF?5vIMow|r*8pbsikrB3l>+-gT7#AUF||_Z1LyRp3uJJr2aOH;MoRuP z!YNZhoRn%qgWmpy!)3})=<83^@+UC;=C7e4+J`y*^Dhn`;uv{7;>_j|e~il2;m@n& zIDR?9Ai1Ovl0*`Trso9a<_dL?e0?6LW<)T%dP5+lNQjL%Jc>n!xHpLA9eidVg_D1) zQm?vqbS0Nfo*{7$R2uYQ!JmAR5RuHg)YM2U8f!o~<14+eOPE>g-i0Dl;~}f*KCzvv zi>sRtpz$I~Jr3m4zq34G_emqT;_?kl^X8J+!I>m-_!7{ z-8g&BCb)a!1pXY8Vw^XtGWjK&pb zkI46LGnpR=gLI=f$8>Vfq_qDG*=blt7NyK#cATHbPRfpiOWt~{FTaz_e7Bm6^sFIk zRv4oT0b=3(AMm-J-L?ND=wVz>zR7XC5idSjJ;?!|g-P)b%O@;s50sXbzsk^Tu3v|KJ1=JW`LFw^Y%XS>EIems9aKdQLwG zv{WB(3&hojvZ3vpGlWFwf%na3%yQ`A8=KeRE{#fTnCHu%ip!v6g!2?7W`o_2`;iH(}Mr1({>lc40+2n?1Uwb=7* z9>kCOV$hz&v~Bsb3Qg*|$JrL50f(ly6^6 zDlUmKPt!9&wPi0nYz$!*tQ@A^OZm7Y%n;ZO<=eT42%@t|@6&tRCFfo05n0Q24dqQcTy z$i=t-ofLRgTH)xOWDnXZpGnV*H6-C!7zjR`g^p!~m=oa5F5#~Oqmm>z`}#6`J#E6Y zpXV~VZ8^DJ#GCXv2@>~Rq@7>yf!Y@% z)?|`9=GSGDN{+>{Ca)2T>Xu>Tc1P?u*ud|3CB!;fm@%P6e^8ZYM_*5>g)rsakpAp5 zH5!aW#SOC1SD!`cjmNlOfMH+xX3-&srPT258FY=;thUP*V>k4D!n*xWIG2MP)o<#B zdSiLe4C{s+;Vf=nvz^kBVECIS4xbGU;rUy}?8CClwDG@lbW+tpy4GY7`-t6x#R6sQ zhD#5adw-6Tx|3Pphl6CL{bcah)dd+#XO2%Z#(P_EmfRYuqLXIE!k%;)WVLdqhJrcUbifL%ZAeD}mV}9OHU~J7JNaU6Xc-pMb)V6Da)1sXg z*`;Qz-Zy0ot5Sp)i}T^=25XLuFNn$9S$SE82gk)95$ypz*5$w`_8iV*HOpn0hT@4> zdn6ntPQ6bD*Y}XRQ8!psc9QGuUcocLHRN!?Nz~A{g^Vpi_)pH94UrULbplVa?}8F& z>u?kde8`0pc8PT4bQRi~L}PzvKgK#v2dfRq{53zcu|oYPxwKA-ji^YaE}VO(PRbj` zcRVILpY%}CFh#thS_4{lLeOpIDSFt@3o2_%d67|)pjBeTs^$;VdmqbSe784f@Sv+te*9Q{N#p5cK=@NatJfj4Y6<=!eQ>YzQ* zg!p`$1OBhQ$=3o$NY2`VUq7Yt<@|zZ_~|S3zFjgXY0p9X^Xl}EUpZcKTS%LPic!T{ z8079R#2=Z3AW|*N`W5Apce*p_y21cvL9-g`{9-<8vFdEV+lip_)La;TKJf@-8fH21(YCk z=olOFBL$;?QmLtE(~@)jsxfR zzhsF}ATg7)M6r3B+5Xe+_|!g$Dm(e1oA(TyM6Tmf>3bM?_a7PmwFK(IU%)>5*=X=y ziVaKL4jODPPcK@J*fu>T)7vR@i1&j3{08uMKgYZ9<_;8o>fz_tZXtGJ$`Ep*AHL^1 zlIts9QImQ#MnLHZL@QdtrthUNS#3V+Ulhb<>=y-dAB& zL)Q!in0VEQ8SaRpTlQWee%}nK*fBBO_eq>L*Wwo5R!*Vr?;rC!45S6#uDgGoEhKVohCUGEri0!7)S_uBe%#|49nCYY{K9$${K`V+x(zJ>puJ z7k|h74BRol5kv<*khGE-=v@CBTLhnRXPqcpyxxHoGFb}u-A`i9v$?EKR~%Zp&%>JB z86<7$f2cEI7nqbfLXw|7zi@j4aS?uu)prizP^kjDWeJyId~<|qHf$reg@3@UJFDTG z|0)d z`sfVIHLS<`i9>Yn$7$e1bjk4Y1&~>%$~1+{$CkSjVWr(w8X#2;%(U0!!jfT7ygLae zmG9?IePoCypB?}gpIexJ>@FT#V$HlvUrP^Zo@9=A?8YOVgVc$0a{Fyl0?)8Y2z*z< zT>M^+>FS2?Qn?YbdR3SOqi>;k|7wO-q~O=zT#qBXj*P4`WkS=Iyx6{io%z>>#`a}kiw%QzEh6lVDYy71a`&-W7ee5VwhOy`{4&4b+e%nl`ib;U z;q(7z=sf(f`rsYF9bq?8tEp|WSlXxNoRMk?I%y_G`b z*D|7{P%1@QXy|wU0k7xfxzD-he9!0eewU`7pc1m@$%g7jbbZzmPH5yb&O8-?w{1UB z`2uAJ265VLCR} z(7UMuSTWlImk#!@`L}*!{f}DQxcwBISdl`iPKdMfK9xel1u2G|e+M1D^YeG>qonZG zGvfPUDjulQW$tFagk$@6qx``Xy0j@5&;BexlS+j4p%e^W5y~bms)g*WJ26LM85!+M zqOpsu@RRRd%rou7i#$hWhsA6D-u14PQSuo4_Qr|*V2}W-KJJAXhpxf(9~EdEk%mbE zWx6!^A%Cy>i*8sU4(oEC;4cLZiY?;c-MR|F(k~m(;=LqF3~m9+clUQn}HWXts?^xoefke7mY>0-r z&w(OmKA)5ljJ39|IJn;h*rr5s&uJs^v&*N(_zRm`Wu+ARX8tB1|7Z(N*yD!jI-+~+8LZO`H3XexGc?38V_(z8UX7sB&k1e&l}>5^JG0 zB@03}|Ayt0>sgDfCzyDtA#&zyDK(RKBDwWt5NwuBO6*y0vn9|qEW&7d z=Ay2e8mC}>k(u>O60SZ~XR3=?5K1hAPj%s>U7=jie^-GMbU1P zw!FoP`Ia&T0&))EKDLOtu-A%NY!{+Aza$Z_YpJrdPesOLEREXrY@<7(WsS!tP5rI3N}&YReN#g{3N#ER1CV0CZmaZ z5+?nciK8$meNzaJM8nygG<(!%cKV=s9*8{$tss zAwk=WheT1Mh)Ry%357)yxz)GQ=^y1>5ZgsaQCAx~{5py*|0#=4PW;4(h*Z-5wiYXS zE`BIm#%60(lYF--5I!3PYc$uw#1>a*tWbw{C1If3bQW%=-ryN=s!a2%qx5y*ZF)le z2+#LXVcO<|kQIAFV5zP(*L3F-owjv8Xur!Lf!WSg_mT=RD|ID`ujl8~@8aRn%mnE7 zdPWT1i<4lRVw`@?1VwIzAz4_9^1~4%{^=?Fq&JHxh<{3s-2OuSHeCfR4I%LI^auN( z-FVTtmdqKPN3KTq!`sXv!TZ8fP&H{De2U^{a?9?LZ72DeKDP}=H@=5KcTvtnUK^LW zOK{QL26SRn(B|4UJd|}E8};6k&l*zXg#&OpRR)&*HzsgGdWpDu(q^>m4W(!H#X#*G zd9u4A3BEt%ySXo}2yz|gV#}_J@V7A?+^^-~y(5PzOPA;4m2sgMWBn5DPz`w3RYWvz zl!8mBBe`+)84Z~InjK@F4SVNLV)B$tDf>Mg8yeNnbwe_7=`jay(>`k3yAZOU?}OhN zQZUl?5iB$SLmk?z1>K`U&;!{74kVcG1MeQS^^B&!!30AdjB(u(Qh2;LUa?m=`=7>&>VLw;-d zv08W1V0_LROjO~ycETLJe$WP%@s5OJ`lWPb#~5bJu5=PU#kRtI!V;XV6~Vhj-{Px} zWvDK?5@roI6N}B8QR|~0UOC#XwfsWt-FWF#VS)t;a!aR%Rrp37=U}buVVSc)v!pF1!s+Mj8e5M z*lpWO7W@YI!>xv;ooeXlGnvaa-^zV$e1`vAR@1m`i^yW{IkeJwEWA!zNq5+vr}Hf* zqv?Y+B=KgkRltQKFv~Fsl!rgl&Y0cgn|eHKwef=IUVq`aFz;;`{78x&{_(tcJ%Rkb zuSBeWA1r=y5gxA$hmQ^b30RS-(8#r#)-5Tb9T`MT&vzNg!n!atjV9mb%#oD?J#!q3Bs5xVtw9?XK^`D%)5{^>~PV6L?01)GRugnGRkrf6^~9H#Hi@E_1;=>B*ev_#D_|{+4ub z!7!8OI;Q$gfJ3d~aQ?GBHuSxqmsf`nHK$52ZaRn;Cd9#tKk;z$-+T7kv!8-7svEHJ zR2tezULYaUq_8Paje411z!R71V9C0%Rt1;O(;dkLFt1<%J`lNq$2L{6s*ls~L)Y44s*%Mq#=5z-<_U_G(y`|J77hZBC4+pVR}bwX~k4$(4Az!`>DLBrCL3; z*s_HSR+WKxKWDaR+g@<{^Ndc~SOCopJiGPV1V{;1#_Yf#TKQitIKwDhD^#F|UfaN9 z{Uo+!oF7O&t|29xO33C>f7CHi#fs!ZY%Z+i<{54x@~TtN`r~xQ)Ib#u+8q+4T+gBh z#7ba!X{g}HHetqyXE`2Nu$!jLNyU}HvSjC$tB~;P8u@Kw3j@*OII3bzpYMvK(Q6Vg z!Tc6kZeR%G*8Q@onrMYF{CuXYhQW+n1(3RVk*sQ1fw8~7f{&CYb#n=ZmMMl@zf>su z@pUTd4{TwSax016R8PDa%y(mceTV1#svPV8o@edMp}i@#kp9pX$h>E8(YukXOne11 zGTNw?m^N$WKA#yUuEvd~D&se|<6PCzg`nG62a5W|Al5XF+df?&D0|wC6*K12zn>h* zrmnI0b@pT|f0KZQA9leGcX8&^t!wlZ>ja*8SLm@i27LP)@XU&E@ zZ7~(UjrgFq^9LAnZ6grLJbLYN40~1S4^_Bkj9uAX`0LVVNR>yj(q0M7XB)!OH*(xe z$pN^uGYoD%Hz7i*ZB+Z}08ZTW6>o1{2YV(M5eNTn6n&#X-o2HEQUxDDvd>NO>-~K; zZqag>e@mKdKdOs+xqINY^B+}MFhpLK*r9P|Gl`BYgqa%W1mfx1i~`T{80eVGTrOWm zHBIKg)g(PUF48Bs_TGua)$4-o_-Alno(y!Qk0)1tMT3c68cln08>9~=lZ@Hjd|qFS z**^RZ#dhY9^u97$=a&mdmL7n!OV=^)d*kq|8}Br7xB$sln^1QDI-YI27B3sMkkMVI z;N3L`yeS&T9@>)vLb!`;SMtTJh2wES+jTDKl{O>4Rugq^R})oRTij_eiBp>B&JFId zAPyC)QDw0WF{)byy0(Ml!2P8-&Am?W;PrGgt!ksKQ?Ed2P%+&VuX%s zk8o=uqM^=cH5PpqB3Acq(YWjWP&OC?r-J6fsn=iGJz*mx$61~`xP6qi&6|Vj7yV&q zWeTp*p2dXACJK(ev;mQ5BOJS}lI;CbhzqV%wvl5aB23wZT+nep$zHGdL%p5l$jk@Ef?GGHVOo(Uo*l5GuO9B^`-=x?a&rj9 zK{;-Qd7!{z+Bv$k#}Y*1V`<5-G-jPRje3VhsVbfTeQSB*amo-)%^#4wMgQvdbJwC)!vY_DVlY{jk+%)C{Fe6}F7Epe zRZqmDIa5m3DBT9Uy@{Tze5(_^cfoX0I*{0Bmt$JT!Y0^}% zeEtm=S1#seR%;XYqm(qPaz{({8W9>1W=4huzeBR{m4-m< zdI|95yOVjr!)QO`4_8%;xGS=2FePLiY#Ea)2%EHosb0@>E;Dv8kE6fih3Cb1`{*q6 zzcY-7{Q3ncUw)HW!~5~^P%IYnc|fON2Yha!MS9%Stehg-=xaW2r&}q++ zir#B7NWV=nwUE3>f7Aq$-`k^+E$4l;MoZ{-do^~?mQGlxZ4b3;CZgQF^8$^J2QX*p zH9Rkp21e?eV8G3bK2a&ByQcb+oi^u)%h%bse)9u7>OY(G_9mL1sM9wExtALhE`_QJ{_oHu7%sBsY4~AQS!n@Y&;&{LGL033C-k z(IP4v^_R;s=Tsh&2YTT=i(C>s-fZSMui3hQ(g4tN%Ho4IQ?2t!ip%+zgG=&e&n zoyZ{CmcF<0d1(RVAuAcH#ph5ma~;W7-_A%*&S84ICgao+T@qhNL8VQWIk4)cmDBbd zqW31ID!6eYd^WBi4yU__nnX4Ds{A2eCBtB1@1nslSLG;*Ek3qCqt#eJkZv#g7tTg@tJ%# zB6qZqjFnx=Tu&*-N6AWDz;rv#(1pW|Ym0E`LmT!i(}7_{MJV5G0;eZ=Vb?wt&Ui^E zllOCuK&FQ0d%ASc!U=z@o*5S5rur|$sa*j-^!x$e4|3eIoW!cukT8fpH6A-shp@6N zn0b)jZ8G1*ZDgM2s8=VUHw+4tiq(F&N|brIrJjmgfM z9x_2seuzf;<=xGuH?c2sP8XGKdGk+&qvEKo19TejfB#+@_XWmDvz6_rX zP9dwBqnRh_L3nJ5HaAl`k2^Hmj4FAp5D=A*IN)Q0Ia?QCcFuB)ynPCP>4XWUK3s}< z%`xzo_nS@H;zweGYjGf*&`_09MBlSyzGEB6Kh@@1w|HZ(^Fb&on8fra^J|-{7f$w~ z7^X3kn>C|`W^XwFkupIrv4V*y0g|$1EM&K;aN4401v&b)ynX)U8tt$NSD_zVUO1*v1CIs;f3wG>{52bh6QCPc(5%VBK!EZ^Qj#T6^<1)${&y`5#?9FRT}EQ~PV2R;y0)^HQT8t6$+pxKUUeHx5k^?A6*$rmfE*zkX-JPuFxb zsmO-wE6%_f$wj#5LJ2d+WFC|B@eKCWekNNEX`yw8F7M)AL$g!gk-XJv=%wWg!x%sm zCVP?Dr|02}ZCZ@6xG2xwx(4pD6S+S(S7YA75%kqpV(J5xQCKpYl`ctz;EtL2y^N(^ ze6Q!FryUrQe3jOy1yn@Y4eWW^H<^4i1!G3OA|=Zh0`=K!H-?$ zY7XgcB9N1>gEe1cVMo3`KD#1`h0_@9sO_e+w?xCfuj^pjyYo!^`yhN2wi~A@MnjxK zBxp~0K=k~6k$Lm$u^=b`{=U(INzR@iX&{W<`ZA2}f&=*XSSg+UbCd+Wy@_*fdO~lV z7H7%l$TMyw5L?~1K$i0T+QCC062f}`Ywy5|hQs934nOGjuw~wUx(ohlr9h5MgNuK* zlkuB5c6E$A3J{y{4H$}42_uRxJ-H@tBTw9;~y!qB(H zM74x>GWFi3{&pXUzVI$MW-EZ?+#N6~u^%FADE#JvShweGRh#Wt`ev{YQ?DmL@47-5 zoc4f~S|o}aKmUVsM|KFRG^1H*mjMu#j20AqkYGSelM~n^W7vQQRb4~Li3?W+C+fo4 z{x#9Cr0F`z@=syI0y8mvXcc=^EC?4So+fDeo4yO%M6zB_v=TMEN^5g3pEGhh2Jlz=Yhe9RD{Qwf)Z!$@90c zsr($S|L+bN=~6~D5pxg`Jf|z-v!Lr!Chd+FLNk@ajNXziTr>R)E_t#W7Mv=>6Y1v! zb`P|9rWwy*ev=I0I*Odxlmfb>X)NAWlHtr2CgXYMT)dpNic9=Qc$VNK^j6$Rj~MpQ zHBT2pEZ>KJVDAErEy>upSOG;MzoN(ESAsp}-spJow!lbM1Rpm4;WL2q8BNpM`0cVV zjC@}L+9I)}W|b89JhS%)FiQN-_ocr?#)7oiA zz2P+f{Q9AiCQEo|6Q9*KAjua@$iSUY2-YhE@lE?GE$)vMJY7E%y^=iWnORLFVXQ4z zJhsEC)PY~?z!<-=n=sq7jsD5eVtQ;D{FqcoHIgm4vtO0C;$uFXg`X)pOus@V^Yg4% zKTg7%v&Y#f4H4}9?Rq$~(G)-Hn$W>EU#_HdCmz@^mV4Ey$t?Zj#Iui zd^*h0DIO_gGCD%9!$tVLbtQrS&cco<3aGxu6aD^*q2D`eHd*#M-D&%_;$5!?xLlZw zS=$%EZU^Iv+biqf#`)FQv-mv;c)EbGysr$!vhzVTpqifD(?U9LS5Rk}JuuoR4kw>i z5h>-}aJE^Cd=A!uI-5@TI&j5GU6((fdzHh#LN)yRPmKAp{GSywlu0zr8({S=874_J z9M#)Zz<)Ojk~_<&?(9q)oXs=d{0jNZgfxjdn??6Fg;|}vnT(UAU71xmSE-WO40c^r z3jE6B`;n!5zvS2!bf2n0a%&#&Z=;`Nf#o4Kd{G)?dtGC9@XyDB!0)W`ND2Isf5QsF z4G=rK3x64n#nT5yY0VEU=A~UKjU0VXw*;QxmW~U7gZ;W}xa4ViMmB{`-M#?w$0@-4 zkzmV5W?}T`vpx#7kw5}gz+7?~BGnF);;ifB@x_S@TUlAPV8Us%N)Y14Ylz?*{eM<1 zT1C7;_c(4hzYnnr|B?Fhyr)<56CQeY0o3?+4_Vj^&34!6q-t|u6^mf(xT`4hMi&ys zC~&b^(eyv}c2Es5V07KibIV+U$d?^^ICIUD+_k^F>%m8@e$NqP;4^(j)5@LF;zBZ2(eQHfJmd4uT|Te4>ALR=E> zNrno9n8*P>a%W*oHoVe>1&tEuwuXP^?-(=NG$wGZYz9oQ-2{Gra-n4LGj^Yg8^lGb zFvG&3WPF1@%JIDX0Q>2%)X5)rr=@_gTn-Kg+GFj8RFqt}oERPKB!22*%>Kb8FdM77c%(*RMvyyZ^sM-ElrCKT!wE4_?NJ-m5U}sxBv3Ere|ehiP_>BVosVB0Fr)TV1+Y0X?~w zteU3V(YuKf^tiPh*K*nxL(QiNoP9I#kmwN{RM~?wy5@j~_ez@8a~+pXkYFAyeJdCU z$OG#Qw)k+@7ph(xLJz7nQk0!NRJuvhIBE=l`6oDI_h__J^H8j#IZ!*iQv!1|5~BDAiOSeO5xV!@)!6C3^rU(S*Gs%C-X z#w*H zW`UBeE!-cYjOzc%GY74=lLX_3Bww?@>S@Gs{A{}nuCtxg=Bx@i%=40i-AtHb)t_jv zuNG#0wZffx;)KQ*5Knvl`#r4A{N0tzeqObTGF#GG(P;P%Ibas@oHb#mC-F$IGo#n1a`R@a41Ss&uvYg3%j%_jAl388fIzT}F0d>!iyV*-?YR$JRs4n-gH~ zISU6OifBPC-{Czm6`ux$32xOS3tE^_Qg~I9(}?_l?PvF}%gpO>{dP^XiJwi}1GV`% zuqd%RSP3(tE|CY3O6aJoLzX1pC+)jS$lj3?5Rjn*>-IgS-*@tU>%R_|G*=Ew!g`_n zuLqvL`T)vUL#`tu3tGuNHbP+u^Xg|R6&iX$UC(5L>B;1(`)Zp>;p{wo^5;Ih-YKQ$N`AC+tjmm*mR`e4=rYGR*kohGcj-B{v)T#UHcTgNDaBU#YEs;T^^2@} zltf^c&|PSKD~UH&tC2q=eE!=xn;dreM^}zS!H*IJ%wT>9ntU$NH*2N}jwNd1BiaN- zw({J`{V%A<_RE-l@-{2wCW#Ndc5yD>J8>1iS6Tbx2Ktw^;^A%+40Ab5UmiiK6Xr|% zqxVCGYm^}1#}5*5dKz5WTMV})e~^j86Pe}!QSO%8SUh}OpP0DH;6%#{m_2fmvgRYW zdVC!GiyWmp4BmszvP(TZIwyM_50ol%PXecK5!Er&B=8s7Ev%uxi%#OF`Yqfdv#XFbjdv|pjbj#1 zQR+u9Xnsy0K^bL&k!26~e0LW#NFIRq52YEe1$r1@Axc}y zd~w9cgc7o}jpo@aDqKB0;&t$1xh zIsO~Rdj-TMK^D)E`FZgqdwMt>*+0+G*Ka00uyZlvY99dm{^j7ryRMAFlk0*hi_eh> z7bn4rd>;s`xyvSx$S^-_r7;GPygys_>o{Mu#7uF#Hv`idf4 znyptA>huWhHCo9IqeW~|a3cHO?>F7GT#MMO^8yud4tMWxhpCGTaYF1l-0dnyww2Wh z`W`7UFYfH5Q@qFEjNSF%y^|BTjvwK@!j9ZQM_a+`#!$SnjpxI!wqR8jM1qp9EVVtN z%nqGPp!35Yz$x{QRRNMR%>0)cDNS!=FL zSsmBSIfF*R6~uS>dD;>-0~z-LHuk?pt0SB^=q0A0{^&{i<%SZM(Q^vieLbPjAP3j} zeM~>y=%h#aUUpdLeClvDpDY!+M@9FQP}|#*_*RlI6JPCycY8O&>ParpCkZ&DR)#bC z#&C*XwPD8JLDngvo}RR_#zl(`;6hqQ(>wQLu)Y|mAJgOdFNsjs<~MZ9i{)_Ow_BA+ z8>Ll#$Kk`EE7aV)FZc{*utYWt_8B&A+yfA*x z?+0)Gf3$dm17^SE``5el(Y@m$9Q<|>*2qnOO>=Yc+Wy7peSHgwihfAmJGQ{9^LhAb z^Bk!24~O8$CgR|}4l+(AFoyAkaPy%N1LLDXCs6`7f2)8+Qla=ajH}8E_)1TZGf+0k zj!NM-nCfiC#Qu<@AwKoe{C;i zK2Jm~$3;;5xr$iUFwB{1ZAK~S32PNO1Vg{(lhVWD^uI_2T*;py>ymTPG_VCXl%#=n zf(tAa%z@ex4w4Gz!uI6Ac-GMm#NQBV3Jh~|PHqlnyje&ko$k|8QDDlsPYiudQdy;X) zbzLA3!#5Zc+Q@#kpUIg$UI-lAT3XxpTP{%6YU*G87WKbRo#*G1>Vg2b%Jexh=e-QgL4@J-yokzj=DH z8dZ+4cB}^u46TNrn!GPCU?LN7I*rCUNPqVzt&L;}dy1ItXa-n(_vF`39q!)-1_I>H;;fb!%!!%}VMa3ud(9OrfA#Pg z%2RY)NDulMT*5%%w>Y8lCrNA{6*R0fVLbO3LG`y5Dl?=(lH27-!ucWSR|-Y>s{K%@ z9YhV@_Xq;*33FLdoA&x|rT=xZ~V)*v^?d*^P*D8;d(>Z%sU7)YQvBws-UOaIe5|e zn}&t0#Qk^Uu-kVbYVSCM5;?r{t5cMbt?nce){D`sYccor&IOnnbcGDO_(i(^x^d@T z+=jI+@{FKGgX`Wb%B?V(&D^|ELhUEl2<*1ZMME6N@7*-;>_Q{tUIuX4FE8S)Y(up0 zdj)OFd+E|K!zipajSRgmBCDk9d4EtA=G^R|%=cJLF000Sm!ZRn;iBH1WJLoKAd-JaSB!H)*Z0B1;*uMz`tHGQ%btfdUa>gc zRUTwzf+1zrd(xkK0Uy0wff5#Wn74Kh{eW8S|st4JenfKMac(XSwsP3U3>v%4{Knu z;Y<#X@)_syK~gBC$&F>L;7sv7@@;D%)>Jw|;IZjc@Z&S&9h(6yZqI0SRx?B`zQo@t z`H-zPA26%<9@Yt+rL(2`sp)zNd~9J)mZ@aIqS|P54cj+~0zzI}@y^OdfT|?Rss2E&xiFk%!+Vr3G-BF@RN&}+Ip)Sho_8W+gl9+Ag7%X* zO#Au?O4iR|%QWi2ui2jDWrWg|)zVDf@o~(rpVlC?{~V*LHO!6~Nh4)amF&rZ!%S?? zSr8kT$Bcfw1HrCpSo~WKa_;!Sf7{Bar?wUfzxxDH!`o4{@;rXv(fKZ)rlQ*c4;W2c z$X?7lQFW$Gi?KZ&P7i&RL&;4-T)5w67`_ya;++O$^O33eQD!2zN}Yhfli_H+YYNIP z6yqe*CXpo3-R!k{b8&rp8~T0lr12^yuxPF)e2tk)d;guMA$iIex$Y4)Ucd&P>P_`jFK{dC9@F@=7D46`{w{RuUX-$AvESO6lQvW4zE62BxY68A zCi{Dl&s|B>dFN?5I4sF7nJh$3yg3N##uUNlaK2L@J()TCa|)hVdLOoD%mKjzL(Cg_ zLQS6c&`sG7p)g^!;G~Zw?vp7cw_=TPPpA$pd#;D6$A-Y_X9zR%Wfqxr)q^luGhk_C z0pYF3=;C&QJyumD=vsS+{D_f-Qwn9M*y}|Ljrr_cha{9Z`q4aF{&Q9{#J}2xZQS;m z&u%M0X8BC0t@LM{f?4{Yi|;lrjpCp0DDdsmz=4^2(Cfkn>ig&&>2TBJ`im{l|L<*q z&6sF1B$5u^{8{Iz?-cm^a4K4xgkbl@CCmfGFo@_g!9;gS?)gk}7&E4gwQL;2q^`Ea zb^lpoRplXiYxnAV1RXiSbYa#3H3P|~oG)`@3CoS?Zv^{7E>4Ukb**^>V zrq$EoOQ=x`yYr9EALECvK53^V30W&K`&he@}}`nqas|7AAWqQLwy2qEEU&dgM~r6g{2lba~+R z*~#74|Z+i7VkGhS9_QE-=~| z1cu{3lY8?`I8(Xh;JZo<)|)oNd%ioP<9-th@;$h-o%~vqO=1)I9^X=JI!cv3fj9@-r!~nvVZ{*iU9>sbYzg41HD6OqTvzM)~!S zY+3Px7+*I<8LJM^Om1b*Bm{%@;}H@rY>tg}63}fQia${j4r`R)&0ps+!ZDg$$}L2l z7uPWSh!D>8zeMxI+%c(O1zFz}j5|xy1qZlF`0s)p7s(}{HIps~Jl01Ze>x65g%b$h z$iUpGv4Z=57ZGQE-*WSX2$N^&3^B>F5TGASGA1O$*52#1`D6>xIGGRWfr6v{ftdv75IjQvx3F*p<=8C_ui_HWYvCH-zNg^MU~_p1ag_g&rDHKr9*-(_5db&`(PkS63Pw(bz>&%+p}4_s z9DQAC07F$L_-)~4ruyg!@;u~@VE=$J$676b9sP-6r?q%zr=F0jS?5xb;TjCtGUN{&Sk=kEwz$4We+{;~rHq6cgN(wqxdvyUEWYM8M6UfzIE( z84Il}aIMgI?y@jtH|%{W5K&xBmaMV_h1OW2p?RA3_9|du8_#!G;Y&8JDun*1V{HFm zG6*9-|Ij{&!k3d_rA!@U9Xv}k9?ZhX-}10y_C;DGtV)%>i^JE$hM-!=Lh}7?EP1`0 zKX*!Tx2=}KtI6IpRaAp(_`VGkj#(4QgdsR{Y&mNG;6Ja`864VFVl~4_iVp6Kq3gPO zvG21fdfzvLM@zb)qDDoK$WCHJ>e3;IFEeRU6{ap>D+*oB!P81wP}!CDt>@{+5K%^dzzKL}7*jZD3OBXm4miDXf=pWl?r+EkSeLd8 zi%0@peO@2+J!fKS_ix;}de;>YpiKz=1X_5C7vs#*%ZdL&~+XCY2= zY$X{xGRWL>7O1mlH+D25j_6M&hSd`=aQ_unEl*K4brV zHO}9)imB7B7X)k!LPahdkN(>L>anvi+O`YQj+?=KCsj=J*-pc183@hX2DyPNaK^)n zn6vUSIH-)p-S??LKt2B%_^(-%Gut8${^wX?FqjA_PcM-&{_n)3 zhJw%PYT70~7dA%Az}baKAaS;swxtZ=-hvY9JF0_{LMhN0ti|npbcrZ>@JuXz&N=gG z6DfjjkkyRGCkB2P|MnBRJx>WmmFLr(_X>>LJ{j)yDiw0~VE|_mV@=dYOtIstA-$chB;4)q0e>^wC_DbZb(evlpHp)reD4YD*tnbObcNqZe1)owijSY_gf;dYBp|I z_85<8SmBXb_V}!_ll?s71~?vfCEl;%1W#u)Qb(g>n6Xuj(O;Mh31-@8w)+sS=pt-X z?oVQ{N`V<^JAkT>XM^~kXq>D3l%9db=)(o!zsz}9wyc$iA8!D;|MGCcuHzu;vmU>) zM_657Dfj^DjA>FmT%WB4kqYA6&`J*sOw%p!q%a|M0rv9LdG2H8`6 z8DgC`LhiWf;9{Onnr$M8jOttDclGp6`8clq&^){ptd41q4N&Oi2Mii>kE(1dg(eGs z=BfX@s)7~ySS<4v?@f@x9lz4S^Hn>k>EYx4Oulx71~J&yi;SB*`(wNp--z z@t|X?W58+2QDA44QnyyV+qtzLQ;%zqdA#>`^|Pb!uh)dAM%d%P1!vOvgLkod-$WIC zQN}d&AAN3c1v@>=nY1rs;QG=uW`Cpsv#=)=Z?g7u@|>|u^pi}?i<`_AtO|iEJF8et zwMuyLk;8pGF(mY7IuU;7#Pr6wgY@jFP@nh@Emd5p>g{vnU2`@Da8<0?#OHLJ+Y)HV zdPTp3QgJ$*mFuz*$xFt{K3+m3RaQn_9DgItgCTZ4$G)RzQE=2?N=MmiWhIJ9%ll3fJi=k^$imtHhJr(TTkT z%08ZC(*$ki=hCdIp=1xbR(TV(&@tiybN15_KX=Yn*qPJHFoPF2M7RoeBG`_)Jy;GGB~PSCg%T<7XtY z30CY;T#O=b6uA9f@9?-$2Kg{D4kwIc5q%veI&dJIDy$jA--$e%Z}NWV(^(E5kNsxb zgDs)r-G8(#=_#G`Oo4Mae34wtmn56}3>YW23?8JPf&KetGQQ2@(L5*_opoN3*PiP+ zhc)x?w)Bu?<_Seczs!@q7 z2QSSW$Gki)5cpWUqtlyX1p(K@nQZSVcy1WUe|(=d{`U^N@kFgb z{%5d9N{qSxKSk%^kJbCeaa%SSX~>L15)I?IuOp+9mTxIS%V-!SDn(_MEt`yx6s07| zc<$?HN}?iZNl7Iw6k6(ce*Xc_>%7jn@9X+}-tV+uD6+njJbS0em``)Vv$K`~vi=ai zw1P(FZUg4w-~*;t7iBxPy{YkG4AYJa?s7iZbC>p!^Q(wTu&eZCilrfAU4T3KAi&2b@cD$~7Z zBbAw6j+^FX!OQ2@iTyopr_amadmx`|@DL}P*tzu0@)_`5Bp(hcF6KN6D?qqA4&=rQ z8wBbnz>CCOsGlCnUp7zxv1eA{{GB_YP+kLTF7x2Yfm*(8c_9j~^~dK|4q~cAIt~8q zLoY6uSZLDo7{h+~Fe!3E@Tpgfw_l}>pRQMj@7|t=A3KkdDiKR^a2Apx>p-}ZF^No= zv>s7&KEz!7MnuLe@Iu}}I#^MM#+hxT_=YWLb2F1y^ZcQES{bPJgImT z|9jhd5+idLddffWZna*e=3N$e{Q!5iC^Uk_c53YHd>K6Rqlf3JB>>$Tzv#u9E2Qsz z3KKlzKF0oQ=SAQBfLle&v2?l~+0|NxPo@;3E9W9EJAViAH4rThIpHj+xaQ4Kk+AT%YyNowfK2#F1tAJ z3cc%?$$N0~7Uxt}r3<<9Z|Fv6G|X_}&zP-Bj`W=4O_GYhyXxC;&;3*`56XFHqyk_i z#)H{Yq;V33s1&=CH7Gm8AGq6rbseF2_=o}SO4B;9$KH4*`ead^F@H7_T{l8@pVPa0)x=Hp1H7{1NvAgUcn z9NWl%bHFagk_T~?dE&wl{L7T_TXBk7QSTim+5M%c|nNB8`ox>RAb@;uv0p2$G;K&?7R^iYM z#z`z3$cdkHAeYNz&8&rM65;rtZ2&q-ex%lNnoxg-rEbNqsmDe|%r@|#NBoCzv&L3* z7rTYMgUXCnl_2g25@D^6Q~s(at~i6+g{yCPj+-rpse}Dvo?rGmEd3V=%Vc6{)ZQJ` z_4X^gvg0CK6FEpe`IkZX_ZVWn)|B-etRZpp!tluK*Ia%;4|^)V^S1Ar$=>HhaeXKW z_F2|7tjHdtX0I8xH}@XgIWUXXlovp=Og}Md)`Re#^>A!eGz=YD0I3!3;DF|2p`j=Z z{OC;9=bMY=X#9Y9Wd(U&{q|=BM@qEKuqHNX2X#6%|`P|wt|9�zbK-&z3#$4J54Ca%fYj| zGU3pXT)uXoIe~Ir_RG7gIMZ2;+4gS+`^Vr3?z1ps=Vr^J_@({)=0A#zvWN<6Fr-gZ ztn}dQvIpRp&_JTzH-YT-OxP%sz;&vdaltIkZIf7xT|5CMbe|N{e$5Uqxe2iI8@6+d z8zas&@QrTD_=D})^U*de3`D-H1ip|fi1o$O#&&g_eBJ`qNj2fdfa?&m&j)8|Ovh_0 z6&Q~DMV8LcWnN||GaV_zw0&DQtKrP$-F7E{(GMjUTFLF^L}Ov{;wyBi{w7{UdNR1J zD23g7W)iI>{@e`U8+u(|hu;1zq`cjf=BzEk*!58mUK_zIoFR=1969FiB4Hx>br72p zE#bh~5cqK?8o|aJne&clmz&E-EBm3$&{p)eoyQKjo7Y>8+#)VRFLAD*9JyDgz=kD{hbddkrY>H2~w5*W|zuAEd6H z<8#p$xH+;;=77)=RXD(Pfjf5to@lJ)x(FFy zB=i+dh<9T7@+K_kJ4Vv(8(~Fs96J&x#Mah?vz?+had8j8f^7w)@?7ON&<33Tj})rB=8dcjVXtB<5;&0n^v@e zI#$c!5nd%^DF`zjO$l(_=PYc!xQS&g-r!j77w}edXa1#5!k+CHu<`3#kUIL3r|&~y z)0{h~)8oK|f4>eE4xe}@AD)44rv*`M-6+UjoQyYC$+F`PZ!qbj3trZm#Av=)iQe6n zR5RrXt*Rx@Dw?ExKdkI?VpT*IWui($)AN0VNY}~wU4Y|)g#c{`7a8uqG{$1gmYyOY; zst5C6a_4rmLlMxZzd+8K&SI#_RQO(R0p)YJcbSAAB)?n?(ofq^V^cm^TD}4%>t<5s za6B&mSqEkD6Pe4^1@z0KQC_hzk9{4l!90@Phw`0s(Ed{@Nt0{Cp(Jf4B~*&4T>3}4 zUi>1>Jx18E*9 zBVn<$4HL_J1OM%3$%%KejN{F}xUFM12E2HR&BL0kXrBri7+)vn9h+&nQaQXY`F~!E z;JZ`{%>R}{-;#V-u{RUTZ#_qQMPH`Td6;TCAH$wGQm}fDB3M%&==|&4_sM{*ZSAH;LreMJn)KkSE+?Ox5)NJpPa9&mSz5AE0u4z}g^j+ZY)JQc z@~6t24y@*~iLXzhv&RI61dn3W_IT=^(E~AmDEK9o(1r4|*{)U%3K}!{>5pSkwbuoP z%YtFav`FsWa01P3v-y>)^I^ng0(-veG&3g&1Ie4AJCJnci`2xyV!rYjCVQpINfWS1phr2V%=A}G6%8@;fsI* zW*jLdav2Hj?Cz+fx??sm}wo3H+6wHZWe>3 zRxQM!=NW`3Rlvg`uz1Ox8>C0Iu)53zFQ&%e9-b~n)TPm_o5ZOG$Mvpzn?#=nxiB9u zWb%U#uOoqpnoN$4D=b}ZMkUsr0k6x8@yxSI%=FvC-mt31aARBI8m(Giy(=HO%`VWO zGv6R_vNj|%cGF#xRX`x(2Rwd$h@@JpFt^_q;Qm`kj6}E&b4MWYnWTY^Mm1Ovro_Im zn1-{K+=K835olIB8ULjrmxp}=mp1zGW|fzaRa=&^3V$ay{NBmE$Gn`N*=q_u{_KJO z1(%ZajzYfw@jBiVD>D?c`bAVV7t^oyF*wF|hO3>&am%wL2;p78tEFEc?Yj>&2{Ksr zU^jLf?VwBI6^OoB6uGwT63_F?IxMYBqAH9PX>wJCz7LW3u)qqtzqIo<2VQ}R#a&Q$ zWG#0mSqurEl2K;aVaWH@W%n<=50`~HdGGkDjGKlr(V3>jG21dY21FJ=^Sn3hF-w57 z${c!aT!(Krp_-@VFUw}fP2~U7355lgJ~X1m4~<`2GHR_eAzgA4Zkd^k{56~xa{o1& z73R(keHLV1HJj736^-ySBB}mIdphZTc#bzUah&>9a=EAM+3eQHIXpd`OXSt3dXyLH z#@g?VbpA_!@c!EfgVX(BQG^yQO)1CI8+1X+ri2|z0GI$Jyt3qYx@b{0`Rn+Y?mu!6 zw?!90RK`lnv(+Uq{N0FlWZTl2n|ZkUM=plSMPbLEV??{tjiGZh>2NF$(zKM>6FPCY zX8$R?u04Y&99IL!kviniuVmhk6Yp^e4@*lAp;Li9SZH@q`{JJ@z90_vU zS($iE*vKxwGXdT@XuzUG0cLxv8*|S2D%qx0#LTwyC7%Kvuq`PNV%l5jvc+!j=D<`| zQAnG)p*cWMy$Rl{{-Zy`uR_yo0j~Stg$^{i_!QS|fn<;9GKz+eg%M`VGSn3%GCq44PD5(en!hIQS4{6TF z#iI#2!}RDD1%~HY4eC;2tY7pQ^JLjls7<$Fo{So!;JuUJ*}oR{?P@2Jqpp#(8$Ix# zcOKiO-$H`t6x014#to!>fUNM7$J2=^;CXW^YR~x#-;_$o6dMDw?9Y0TkdqHzs0p?WEv(yStqQc@5T{fBPU3dg|b zxg>M;5N{30Gn;yjQ`Na=_&Gt>pyXBt*4DkIOV7HX=Q_fZ?QvoS-pPQtp9$xk6k@z> z!g2CYnT6K*2Ewk^0h570Q_P zv=EISu{7gJHnjEKqLqGOOqSpevTVj#GBjk*%rGp0Q<#G5+b_YnW$9$~A76N1JIX&% z6u=wv=;pmpkYt6)dQv$Q!u9;W(4^uyOw91eu=`Rq6GulC0P^QD9D%(qVzS&^UV`EA?0&R(4Yfp zHaD~L&s~8PXRFE6#JMQh;z3r(%Hq3`Y`pke2tzemc|YSD$gcQ9y`$7Nx4+7C$CnvI@!Uvj;hQs}7GqZv!ffj@U4##bMJ7U^)h z_QpP*p-=)&4A_lQ6aLagF(1jBwpsAf)0)}6F%hr5JZHJTte7SQb2;1wMRt|+1a{;y z!P#26>^9qTbW>C|Z}Xc^mMepQP>s19gXzHn*4F(4=W&?~GSWuyt6!ac=VMK`tIlQh z=FS1;l3hdD&SDVTtipUEy#y1~L4jk34oB~#D}A~k;XcPD^bEqhiXLL!^c8kv5uU!_ zN%|KKk@J6B=-fw#a7jiLnuc?^sl<5Vd`+5tlAng-7IWFayi)prm|;{@9a`vj;-cx3 z@c@?xoKhNuQ?DO|@so-8!7m!TuB6hAL1||H#$4KP#G3El4=}BJR{eeVzc^T~hLRW7 z@$HxTV_4=dbg-0T7mR0GE^SwW&aX4@RA4ABisAMemDl0+%wBrwYYTN9F306cRSnl} zPiFT|SWTYTDC6~%a4uVR3g+HhfR83?F@X~npxTXl#MwCnQnh2*`QImyAClD=mB_HP zM3e68&ZN$2Zd6=49(rmYL2&O3$hWJ315c+jSGH4j#Nii44ew%yjxS^iO!(Bd(Sq-? zD~Vasb^yOi2hjTk3t4vH5#G?1;5gZBaPh4J-4YQa>%e3@6=F}H&Y~8R%gM={SSpk51m*&XFz>_!B(?8}lZ!NSTZ8+(3yG!r%(`u`qwE<8p0OI#b>v}M!%Qaq zd;siqT|p0B(P!sBbc4yoalD9mSE-A{1X?Yz7i627xLx!uc+yu&)4M9@bRxs;9Twr0 zN!Jl1j&fYQ6yDQbKI&E25DE7I{KxSy+M=pqYtaHctcYZqnklneQw#Jr^b+S5TR3Wa zj~>3YhVc*%;vMq+PMWPZqMM;C)95*$|N7clx?8E2E{~Sx@5(vHSK=~qx%SmCBe)oU zZ;C_BLk<{x#f;G@*JWfLYtWAt=cvcq(t5;|FxqFw%#MBo=6m~V3nehqG|k8H4HyV5T~^7Cj`Q1}{wd?`#8(IJmg=dc|+o5{hla{Od|2~IAN zz_zy^V1Dsr>?n$&8J4*?XOLGuBJJm!#+`f#3f)~Nz-UJkVp~oB^DCKQG+JM8$ zmhz_7C-FJgAj^;V6696L?MZi_zsdg(vL z!@vRG8aSftJ&yIDu@N==Lx>yKX|8<7og>s`;6IWM8`h+f^DrMoYs_&-eHQb*$R8da zy+_}jdqn5B1MWH_#s*qi!LBM%Qhjs^FY0&?KHldEB1eL0`+vtl{n}M>q$Zaab-E77 zST%?}j-}bN&LHDI7vDowukfd#~Ey8$Zma`9`}B<5?h zJUc(<7O|c9o!g7gVD=YBfaH)ft5jNt1Cv{6&h&}!d)EajGNME)YwlX!UN**?y!Iw( zk*b2)5j~6xO`=6>rm^vV4goJ>71jkDhWpL#D6TCDpL=)UErG{y{9_fwFDfJP?{CA- zXHQAr_zQlA^ht&to4|-Jmth*YUPfNqH}8&N6GnvKQ4yKe&Ip6XMa-*ke+R^;Fd2Sr#}&&~Vg)??fLSA4+LCeXPRB@E_2zV2Brg@)=LKGn3C87=u~rf{dw2C{0+%pt|cY z8aoTI&E`#H^|dGx@_HIWi0jB- z{)(4ju;=Fmni^;YYo5h|;;T$pcOn+k)Wp~hwa1lP`l zy)|mwY&MMKMqUP4z9#wqvpMK-7Wq5&;RAsRSkJkc6uCo{c&Zha@BT$M%?v}$S*nbD zoH*0Map-NV7Gc$_MeKvhgZxv1UomgzL@M++7?yT&Jebedq3-EQ=NHY{(j#+RK&WQKGvaZ2SlA@0hk zcIGI_(+;PtjY9bD$6>HEeMMKTJx-47h=3#GlW=;h2~Znl@Y%A6s5q;jLaigSB{iF> zDCVNf)pN);)>)3>r}_&noI`g`C~g>N#E-!|xbfo>8T@MsJJqyc?WI<@bfEYoFo&;YHoD}`%E)9=w;26G~1AqR{Bqn;QJyUwl z3-k=m!yipg$hy`=TBl{=c86BJhP@O@JhCBwUzUO(FP!uhJcqm_9lY#j04kHz(R5S~ zPFiV`vGD==Da#J?6HP4wJZ7*udCd@hG#cceg@B@^I+?AnjOsF`@M%C1!V|mrTHU!| zrmcXQ-C-y(tOe^W9@ED>j)GHtju>3DM9()IyZNdp$O<_yANFdadC+qDJ?k`RsEL4& zRUzz(3qsRx=kQ^*6e5r7VxIFr-wk0nJf02HpB|&@l`Fx&!x{}I`~qV^MfQ)@ykbqAkRFTW|jr=BpE z@24x>3uyE7|H#waW4wc_R}pp-+aURL32F100P&TQteQdw$+8aSABdF3J8t8=ToaB( zowgRwivNV2s?TAni5^EFxy%2!W-ak*o(*nGZlX?eo4M`XAGCf?9-qH)9liHg2tSo@ z2-*c=ENfUxU+_PXuk&@N`~_V`K4KY@ctQqC^i%2hW(BsT;RSIW&cN`!&h#Mf9@mYR zX2ldTS^GQ})U&vUbXOw$uo#7#1*~Oh6YZ=x&$IfL3983Gk=dDL z)Z*4fD$%%^23K})J+~~LU+Y9>EHMuBbv8o1$zn!jO%WvjImY<_0-*Vj2+q{iB0}nt z4AD!(L(aL(vd90QgB8)Q=*g&FT+S@`?-@NWI0%7z%wfvd67aKjWme?-g24SP17p;e+fQw2Rvx zODFF^^=Hm7WD!nn7k;2c&)Q+oR*H&V<6%YlatyLL%5jZOAW5D@#Vj+R?YlHD6KZY5)kyZp9XF1I$tyrHu>JaO$O*%%p$cTmD~yVctb+K#-IbJ0@|DDoZisTlVh;%RVJgH!8D? z&E48i8{h}Z&r-ndk_rxgT*{sacY(OV-|$q?ioCCWOJbO0P}`)34!_NyW!@sVJv4zG zo*qSxhPFca_Jg>oL>&CQL>RsMnegUsB=3kyI_(MYhkVmzu(*`Z<$qP#*Cq}q_nzxf zjOXJWnO9Be_`df7ecms^R&W{3)ijAl z<3iM3zZ60RSh~yB4DC-wV2rN`sW_?4E?C_}_!ULymBk0!`d+XR6(%FYM_~QPIdq8b zBirdYSgErG=GiYtQ*~i>yub_;N>qr=l4RW6TgBT}R|gx|R{rjv3Iy~@u=eL1eD9%2 z7Fc%CWvPZR?eanr*px(|ekJo;>ojk7g(6z1KSo^|h{lfZcqYcHm}K{3kUH@&=CGHc zuk|)axcZ~d;9tm#HKckmvoLafIG9{~$U8VI8OBTqBX{2i?#;PKZL>b&`^zc3CV|`3 zbCNmys&ghIj2z^;2u$IA(DdKd~gdCcwGmCx+ zy$&z_roqh2bL5>=De-F=q8&O5@cM@R%xO~_u!?_(z3X#u-McUVZVvvXIUS!j)q&Le zG$NbV1p5m8__6kt_+NP@Z$pM3*{Ci~;U|&)93#USi{3&ae%={9|4YX@g*E&xb~m2Fy$PfpEZ&QV_ocovJx|MsKF2OfL)M;Lr|c@FWLH?tCB z+W5Br3cL#40&25%K+VL@)Ys|(TzveSSCAzFz4;1k*UDw6SUrb)sy>U`k9hLm?tRF4 zytG)8AaSzywg9_n%NLsCr~;wP3!L09%Jg>!qjf|&zWj6twypYrUdD3lG2R5Wtz#B+ zRe59ZRL(blO&wf*9ztv7B(P00<>gj5qTsW9?9tAoGmA#Sq;EbNyPIPyZv>mg^T^{3 zxe&QEi>x~}nGvv~s6HjWr-NkdidLsT+gnNZ&g zpu6Tc5ftu)pw=iF;KDh3hoqPST%Ebl$zS zZTRq3BF+E&%d(?;l$>!&Bz9Dv?1;#NbRlj}Cvu&VBnfb?P9w)Hr{otL@yV8 zxNl0K+-@fk%DhjO9tQe`d?9TMjNsC{d632$!qYciw8AFV;>7Ywd>JZ(n&uIZAxoHd zU7yIgaX~8Iw-+uray#C=hVXi&KG5W7$hap&WJV7#g+s+WuY2KmbyF)X_LO7~^ykoN z+E?+h*F%L7`kv7EBo)Y=WrM;OmXo;PM2IW6fF7+e zAod~@(#&S!IloXy9%$nmG|t4=4yyp1PLdZ&gLGJa5A>PHf{)-d=7(k{cuobSkSZ5?U!zC?I#6>!PTfvAmMq%Hyfu)j^0l|JIadR%KHZ||>1 z&1T%r*}F#B*{b17iriXeH8Ty;>p(GZk+!2J-qtiNShB#VG0_SV1cAKRond^ zE1p(LYhV1N{^tQN%w5PUmujH1W=OGO`kkb~Ap)k(Rpp%tk!Bitz7c!f9SmHJ!)d+_ zq_ABAXQY?Wxlg!R?%n5PHpj}5(_-ksHSf^qwG11 zzKucrJ0(fDx{J&3Ogl~I97*Q?>nor~@2SF;FB(`t&hz6t#(DE!SHt#B5xP_ED4b17 zK?>w*GX|DNg7vs8itP-8pM|D;ZxM5% zo#G9hSM)LEL=pBoac4053UY0p4&%Pw6<&sgS}s?!WF9}2f%bim8m?D~aOaI=dfwI< z*hk}_H?RhukMQtjMm*-`Ti}9jN31qmZ|U>jKM)m&p(THALUMW+Q7HLMJl5O?Azww9 zBs_>!BPC?)p&F5iI|b^K3G+TG3N1Hrdj!9g=ro|rEDY@f#gTQ)U9KZqlJNwBYi^Uy zjf6FPGKw2q&a%kOkId??AbC$QUF-yZ?Yx`(S3Lpj{jelP@az#Bvv23Adn@4L@3OG1 zvUdSKXm;%S>u`o4`zg* z`Bw>8_uP~p=O+YD$0bm@ycOOwhOuMr2Qf-jf;E~YgGO&oq0K%)w$8|v@89i1-~8>N zIvQ_4Lv=0=8Su&NgVChw=F$4`gJ#SxQDgpvZK@nq^`m9L*Cdis=SLp6Edw3PRg7GY z6xk6{0m}rU!G8aBI(J<#b&Ivf!td!E2eXjG|Ci5HDo^FtY*@&wnNh)S4G@7B@$1OJ z6CN-j+#H2Q>u8_xV%Gm#G;AI=!3l?Y$Uxi@{CvBO?A~q8{!x0_@bP{Cii8a?<=el2 z#;@7%@|FO5Z9Ia1afKVcw-&-h-If?8JX~+rKLG118}ZYaJS*q$0KV^vhFc}F?3eUk zaBNvCo#>fHj2~}+qVxdTn=DPFBMm`nPAVOa$bkJf&hiv)oW+-^rTn59R`}vc8C9%T z!79U(sP}Z3s@_vT{{rU0}xBE0AavVu&o4P);$mbh+Cdbm^ z)kDzs@dwC_ZUC7jTHrlUjm;0c@W%E+I9hib%JepZft3uGg|Q&5^5b+n@y1xj8Sab4 z@)kOT!Uk!A2DRE4`?HI4Q{><@z1#HjPz`;xSDfw>v;i0IKK_-DW{m%$U0fdc7D(UI zWUQoTqxrj)Ov(-;qTD)}T=q+ck?s;mJJF5L&jp~9WHo%f>WrtWY?+Pwj)LO-rR40M z2A&y%s3v{1A#9!n{_M=gA6`vl#tADnUQnHt`B_1|zIot`3A@2zYb>d_e4=4<`!Mwp zP({A`JB(VT#eON$Q)a@*4L_mQ&m&3tp*-B&vYuRdxPfmyVr7%EOvV1} zb}Ay&#rwsZj8?Z~;ZjB#9rwEpJh#KRdqD^CJs+cOxEecJewwyPUd64e%o$5J0dOt3 z3#AL>*rKrKa?fzKA?gRlJ} zSnV}~Y5vl|i;Yb}pT;|6@60-A7SG`q-?)b15`0*u|CHq3JOM^$^AMNUP#dR3tlg_C z^o&z978Z&!HiDUWbF_;e-eAb~@|!SdXes1P(PeuhRM?AZhN!b6moKUE24@~RMYCL8 zsf9om?Vb_NbB^mlk--$pT;UTW%3dF9gIW5;?k2CuG@2Z`6^6lg!2EtW4SZVu&@%b8xOd<- z{glr6v|mYrY>y04zAp<~^-s}&8;VTgTr+C#_=C*8b;j~e?Qz`Y-3N#NaGvb$d(`%k zA@fZ)8Y+I6qt}^}(BiQIR~(#(OYgqJ+|}Y7M_iF+wTXjx@m!v(P9`oM3U|om_lY6a*)ITeRk(S3$`9K?n9z6#}eiD17^9q0S!PDsG^b7`qI^kgDA|_(m zSt41KLo@@UASO&sOU%SvF$s9+rGP;^z|w<~%*hkCQDkmID z=L|k|T!Cv`!{Dkw5c(}NfcPt6gl{hmwFU-g<@*a-qjjlha3cu5&c%D1e2IkK8(fgf z$JE>msw1j^(pP-(V5B6B4XtL?Hp((3w=WZ)UD=o$QB40+)WO|r=TXU^iJ;;(LS}m0 z=dWrV0p4r@^7+ac60D)m{Pk7h4aIE%sT~h^3*QGoZ2L^6uC5fmY!-y=wy#0Af`@}{ zgEXoz5nB&NVq$q2kJ*=sBNwgV*+?Ckvn?M!UD=IcE?3CRtC37ogepHa;x=7gYRCM( zKMYpyi;4Ekl`wz5D6{eH5AydKmv5h3MciL%fv<1`FHN$9o{V3LKA*KAUn8zzs)Z@! zmj{ur)ISceHy(r`k%0Le2G!DZ(~m@hbuauO#dcbxTccEJ~kst z*_QZxz9M)oc7$Om67LoI`RqBC@cjRzmG$`&ZT7XHHmr?#gnKX%k}US zIrdchX=40c1}g7zE-*b&D0eU5r~Eai*FQ{UoxP4hla3&}usw;q-v5Z0cQvEpSt+tY zECkLKxU<&ZHgU7dG2Wv~2bg;9f39teq)DgFqxGUUBz)U$%rJFAuNS>!#eQv6GCT_3 z7G~1NMv>6HC7FJnD?@w6hxrQcv!Q?U3G}*LLh>^$=vbQ@-5E0pKW!;Oy5b3U_dY>{ zqk6%Ow~=ZtcLXcdo3tV4AdS)WlTSC{Y+Y@f#r&lY>n_8>HOFbh_yfA> zzstC4%mtk`a9%$VuDxulgatjf!8L|~92p(_79+uYvKr=3`}3B3?HZ*#!Kn-?c~P?- zQ|9)8eW*4%(okpEj2rE8aC%|9Wt4<6Ym%dh{&ME@?(7@1<7WryKcq#>d~zX0;Si?f z57M(MZ{z;yaxibY3wi8T3bNwLR4c!dPRh)OhnJSYp3tlOd8QwUgI_Gf{o6v72i}un zj^41o{V`oPMG_bBrV*9$K)O?qq0$Z$sf1-3|G0+$&uV)N>Ge8ILmNi$Lf{epO`l)1 zEx=9h7_#{-)@&Vcd19C~x2F7)h?WR4$5z=Qvulac0cFv`87Wj_6ZSS=x3 z;hqdLc28s6hD@O6;4c2gMsp_k&l*g1mWHo7lWEU47ij)r3yFK<;H|AbDS9pn(j%O6 z1yq>|Z$T_Ra1bZ?&SsW7SCV73YoN-2bKd+8!aF+sD867Bl(gQaEt|(6X8#K2{#Ygq zu)kon(m7Ojx=JJ_gb^3N$K%G1NuG2X!6TtMCQZ+gzXz~ z=Q>-wd3-m1ee{j&8=FHn(c@(0_(XEw-UK!|n$n0LdobZc4lYkR#QXWNo0eLwK`Esy zIOSG_i;eF=w{{uzGOvTEtMPRB>Ph~1=_(8u*JfE$Rd&THIrf(KD42H_z>YIQjCz6q z^X@+#*eQPi)W2`VX{L*j4o<_AXIJ<)pJ~wqF56eDAVx*+-LuS|@EOVtS7O)~9(WZK z2x^lj-9@^{d^^Upi2Wh9dFABEl6F)+8-Q1YONrByWYqCpggx_f$g1c6aQP%B)G?y` zGLJTT;4as($xx;t=lr1MnManGg%i1;%F=Y02L zuIW!_3imgV;}(Jd+br0NtCqpW%Jn>lK6Q5PPYqUWRucyH){q4AR=DBB=e!)ZV95I@ zX!Hv+x}G!GO`BTq{$fcs{zeG#m;8sryY(hw)O)f*gG!@p1wK&GM+`SNhlQl9*byL^R)oXVjSs}3+cpt!)Cnv;w{hrZ!H{s z=R}VLEhAswtwMk98JN=PixYkn423u+$V_g zOGVi)$zBa5=2P*TRwN|OEvH29A$`_<1xIX#ENA?8ho%}76Egz~oEiIwzeP5P{@wV5 z_prK%SPWMXlbQ2DGiHFwC>esf!a6Nl8`(G_>#Upfd; z(PbcUP=;+?HwN7>307IyL3@4z7!)n%4=3)#TrOY8!(-%}HDSDu+Th!tO|;WQm=T!8 zdD)s0h_!+(PT3y|mvXOBg?Z^jD9{6azCXhuXGu6*FGRFwDU-OHyWo80JSv~Fm?!W& z0BVOX)f7u0`z8>#K!&?Rd+r7sR|-zD!g$#djklZw&WFCCUd*< z5{N!E6HcC-$(MOEACqG0NtnZW+&buppN&oN?ZyN6BzY^&P=8GXgQeM1w=!YJJ(e8U z?~aQ3SJ6u=7Rk4Xu>WKW*)U>9sqs9L@lKA_yZV{>ofza8t+QdR$$FeT>p1WF-pdeK zQ$}a4nFEXWe*#iu35{2m(e;~EStaMUAgL2b4HU0aoj)6~!?7FkrXC~;TV!y*wG;ne z-3@RMo`yF#zL06pTF^7hq?GGJxC-7ytI^dkEG^8=?lr`-R&VP=f4idd%1YEdHyes& za_Pc60=`vykf(Hzc{R{U%#(vj)#6@=d(?{Gn*!m5>_SLN=k|_Ti733i2KB83STQRf z_|yH3HkdZTJ?>8(xc3aRKANy)3(9D{R1PiO%VR9J+tY^W-jHhI&*%he;7xWr@vExg zl?l(s2gP>ugF2#Wqa*guJ&t-^{;=h#JC@mU;|whE1%l|*am%7|xqJo%I30LyMpfoNHGk|eho9o?Q|>uVc&bC*3HuPGrJ za<*WvE5eqT)}Y@H5v=^Mh7F_>$-ZwxBv&j8Uuo7rxeLQNN%}~ks30S}JBK{VSPW)w z4bV5n0DmmuzJ2<)k!O7a+p4dDUBCfwJW)gfUQd8ybT2)tT1mVgl!COW1J!A)sz3C7 z71Q%njjjJLg>L=e%j(7y@h&DGgVK0k$mhE2hrD{oYJ(P-zV{%+WWFaa!=How{WTD? zobz~heBzuAj!1**rgFoPf z56SeOc^Yx|mSUb>5@jyVxe3oyOt9pRI2E{eo$NnV0lzc~`J!)gG4hBo&YdGkHoi;a zg(c+TUfWw(qf|`a*M_2vS{uau&4w4Z3US&(Kg+*XrqHl>Hqr^@P;2B0B^&MFLFNZ^ ziI0aCM{OpcPea#tZIU81opq5}TUWX4AE66UEHhK2v2*Gp8dIJJ zXRVvzM&D(+=aLh?nMqKD<7qa92r|FM7~Ja6O+pX!6LH&dpa zmo##2=e3N!fESiec87a|nJ~dEgc^<2K)Cl={Bt7$dNVul@5nZYCSU0l!~aoq9{yOp zZye7`Nw#cBQ4txD@!Z!DDV0Q#CMBhWXfI_ZLK^no3TYU{xvx{Cq@|$}O_E6B+aQhK z`TYl8kLNk(zOU=^dB4N4&VQso`Rr}zB|F$OV;_`TGdTKK4i$f@ z;6qE{OR?S5m{NH*Hs62eLM0G|urq6wjy-`2bt@U}XcK+3xq;7%m6J)myaTvh1Gi{jps~KUX+@4b zn3QBu)gN(iiRy46w(bymNVfLAdM>2zS`G1H&%sTV_gfxxguUOBNE<{j?|-g?kJ3vS zX_+dxnzxXa*@wdu@lP;OwVgE1-^?6nI}g*+S_ymn1n!z`gol4g!=yLX7;yaw>?(C5 zM}<<^`6qb}__F{Q+#SuNf8xl>^($**S5Ktb>-4yp^*Ok9_8jgr6ky=gv*0HwfgaX{ zxJ*0{&Gp;ZS9-kjCv7rSj^;U13ajb;U*^=)Cz6O>x`TVG%dzXtWwi5|2N%kD#&&uM zo;E*8x84hcQRilyXX=5Ea>B88!byluGURU7E7CayYMf4DC(&{9!lVsH(R8sG9X>FQ zzC9KQL#^_}yet=teoke=^#4GA)(JRyMhQzcNf8x)A+$J}jUhZ2eqi(;S0nKCbQA1VY9cPZ=lMR$X>8P= zga>q;*>iVtU})-E#_7`;9Mijmst@}M&XX%>Kk$L9&`=~ze#v;=s~P`p3ShEMC1B4? z2MiH6rE1mNnZf*0$a}Vq$*>KFnLE$o?ynzls5OR|itK|_hZ4>{Qk%kTY2x~?g7!Ur zi8wZr7S+d+Nxn{)a4HGj2G$do?aIvEm<#l=X#=@$vK$}uC6Faq6S!|(eD7%cUvT^F zgx`Y}GxcxZz@ix$aQLeiT0SMhvxEka90f3u#(Dzy5cc?8#LY93HGKyCv0!9 zPR0vxacMYSeaqoVnvLOy@6xL`Ti9#0p16D21zf+?k#+{9F!!#yps?C&dWn09-9H;) zy>J4NI{Jj1=s1X>U0X=Lrw`=YOef;Uy2<8GesDwb7x|Oog6an+Fmd@tOx!IefmqHQ z(4XeaJaR2YeZD*RkYbXQS29$(Zx`HN76O z5;i*fLRzAZ?0yW#6K_zB7uO+w$`t0kr#BRL zA7+BK4C3EOH!8KjjJH1Wk{`eOk^TV%TWE! ze26-|jwb9}j5$WZxGBDf9i6Mg2t-ez#KK_4Uf7-cIprZ)*LMl`?wWvV*5PF5@+R!@ z(Ew&uA&x%01NwvFjMi!!`p?^i`C#%N>^-=S&Ma?)rPlZ9pVzzKkz*5J zzVORSe&6r#kF8knodi8x&G$A6@ssxu?9iKraf)eV@6^vQP53ww#7zRb{U&t7%T-K= zYZ_Et>mk2tHJM!p6XtW>#=JYbx;ej7EXR!CD3yAEKk&Hwfmr^U@71)cVM=Iz!XhOx@-d zn9{VD5qc)imHSDdD(8Tg$`qN^HN~*9KMw5onlcAtipbkj^32IrYi4+Q3U2&Q2&e5r zk|sO_vbS0=QE7aaB)Xc2j~H^2+h37wt}khjT`KL1Q-ahXbNEy4jYUi)s?%yVIK3Gj z?YfSh0hV+`N*jd_HqzTK_v6~ZXUzHTcskERmgv_)si&b;J%| ztZ>FY@|(1`m|*iOq@lc$lGM zCsOv`pP8Jo$!!=sIS%@CMreYS8hkk=fhM!&b5X*MIHmkKDQ}7<6Yn@e!Wzo6fH=IT zpn!BtBetEiV>-T7)-5p$hd%ZJY049W6-y*p$>UP6phSz*jXVZ1i7XV?+H1Y~UK#c% z&7w=D6cMdmA>3^V7e3qh0t(-T3--m#gLR@m*#6`n;L2>E56Y~VcMIn+vxO%xHJeRf zq0d-mR%|g5zd6hnin-#;69gs?%p*^PJMr5#O6G4A$BlLEXcpYnR(Zu}nXZl6t13A;Z4mQktNTy`1q#_08+__OD@;zM<4(b~+`yJN^BGrl@ zq|1pqm-2kTgVMli-xKVQmSX>=Cz9uUKS$G3mHQI@fE686h<6NXz^_9ap8e~DJHmrx z@t+yg^Ep5J`X<2Vh0{>Y&;e%V>C<;F-05KW0bskQ!mZ5mzH zc{&mQ#$2Y2pTRWS!bDkec?EIlZX)JioZW1`I2|=l%pH%Rd*#*qz_rAjx6DG3v%RhSy37Qy+6T8!C$7wHe1i`X+#1FrvPZ5U zuH4Rj^4<;YzKIb1*MsveInJuQm&BXy-%0$rc&_H~3Y`rB zg{4U#vf(|X(^6XZeJ{UP3WNa{chvLZy)ICW2EU&}a#sMBontuvsg!hQ<1U_raSj^l1xO~4F; zcud-Ej_*gV05KROIyTj;-jv-Kt{w|^{@Ju@`Ulw8IYh|~ex@@<8(zN5CE*>F=-?TK zrAKn%lG{&uKff0SZIs9@$wNdcuSW1{wgf0hDPqBnP0ZIa5qu;&2D0T2fPL*=+-_iv zxhknB_t(cAzedM&Rp|?5Y48h91Iy6U=wr1NI{O=~ ztPCfC>B<(`c{&1rxYglj5qEBDfeCR`odRue0#}fWaO&w^q;Ir1_YV{B_w?nc*c^;n zU-<4~;(cQKx14jAafAJToau0p7|OTHf~V>!y5iG&v}mv)w?42SHf+d<=CKferWN`F zGQe+&8GdjIg{fN#>fHDF!5)@>hs~J)e?IS}j+3vTg+>U=`8pr;Gjl%0aHJ}i^y!uM_ZpLY$@X3 zfi_Q2G3*+oa5k8EM;*>iD5q~WoB_SFfuzjAjENho&Ws+O%)B8@Y8?uRvAb zFRKLqFep*#rXKNYcn?hpWQVtdLC0DY`e4i~vKK|&Q}3wgjV^3$G^L`8i|9ALr?PHW zzu@@m@2tc;CC2J@IcWbi#bZ-CFt+*wKJN=5mhPLtd3o?CYayZ= z&ya?{W!OBu4S&7!fkzR%_vn0(!10R_s0c(s?&UWWyQ#r=w1%-sS6-92CSh)PpaG3v zH(}OOO|F#B7;EW%V^@lVGkf!d;K-F&_QCWfcogyk63kklL(!ix=HH1E6FbQ5qC`v( z3dMQfWN>?Rso=NT8T@9t%vydQpRd@%b1ko4CKu(bxcf2^%|I0}bBeyyNZeh<%i&A^uIBYd~e7fOmY;xQ{_X0lWq6y$p0>P%!mmY$>= zQz~fKs3zw!?y^w|hc!+OG>Lx<=6N{6=W)n*=IzEkSV!~DKa_p?_Qq03uC?rCBI*|;|r3rl#0S>`@6;iEd!UzZQtp8cR(Rp-J} zk4#LmcnA|uuci&1N8o5@Ffo#Ch0ff+B*x?&wK5!{w|upjwj|yuA3KLsEd0UyqEEtC z30t!B>SJnC`S0$ge1#8q2&cs5ux{j;_WEW|9C`g;vE5m#NT~w^YhC@mg?ei{N-KpD9hUCn6#nE&We?g(&`bVKUDnIAL?` zS8buSb-OS(>AN^FyA%bRezw5N>uc%zy}>v=EDIHn@;mf5N|2N&%4825Cuh%Y!o@Q+ zaaqX)5TD39Yxan9=4G~UYj!388K4_(0!FecQ8LXHw%L6rj?pEkyH5oCC0grp z^As@W$7!kpr_eRwJn5{Ah6RfUU}pUUcr~Crj?!65=zHLlWdxm#|ju;b1o3mSKNxfQ`TOAwehx!(|S_o|-ZgTV};@zLPm+lE+QpJg@f)5EHO(T_!d}zQs@Xjk$M9@z}wi<8l%Yz|h!6 zPT}7+yt3^&{=RmC2|HX)JZ6+a^1w@aB-fYjRL+6sj!dF*Q31`I&9Eq7KT~|g0__GH zVY3at`}S_fB)3ysvfq1{RX#%NORI@b%Xqx)x{_yoZiO`(w(Q!)lhM{P6b~6+hihXD zh!$kRg@ZR}5N;D}bZwz&ofUOwwQGpRuEP-Gvl*GZYuGT(9bZZh5hLYSB+)S)Mt?X9 z9u;iDWn3wi>sg@g0zSL6Ql7>x(jvkCJs=iep5wG0ee8H}7BbhmF}sfnaTUc8AkT%O zuZjnRIqbWDJ8wvG6ohBe(bb{a~lg}|7-V`0tf^LS_Y z06k-EO3beqp=`uz?&jA{WH(3#Tgf-5Gr5sV;fd=mxMThU-}?r_{Z$cou`(V)+LJIfP#aq{ z;^>2^T5v{XK=5}|61Kz*kyT&MK!|e}@!~cSS^kXT^+=oll81owkr4q ze&qYPRa$UvRuj0UOBg*YbOWZQKMqfPSpmZ5GNGyeILLe|gW$pIBxJyZzTGw6+Hr>%=XbNWE~#!k z?c9_M*Q0L1vtS))juzok#7|O@t&%8Su1c>8cqfTQ25HOsjN`xjB-0M(*A-;_A=j3b z1Myi}_h96u;I&FDW}L3Thjocqy;Yqt+%-hp?9&AEPaS8QbK0SE<2%^Assk3~jL_VR zjn;bvUr{;l3%wLBO0N05AU`kusf)TMAoFb#`R|1$?wmG@({WZ|3?)?&#m>NG5or+V z2xDzhj6qNyL$8GTLl;@c$Q}F6mc*A*NxP3g=5>$zI7in_W78Nl5FV`?py1?}C zJ;v5~>rmmmHRGQa!nH6O82mmHyd6N}i7f zaB<=$-23vW+%1#Se3sW3KDpe0Hws~#rneK<(x%6KbCl=&d?tX#{ygGy>oS(EHH4)m zXK}iR9M>Jbh<*6b6vI*+8P$}Bq(pN+d9~;y)Fvvh!pw9=s_Y)t#;LQ46?-|0nA>=G z3eR2^E`fZD%jDZx~jNC}r3z3^{5Lwp*c>6O7erLBsOD%$Ai7a_qAPtxOrLmYC z0#8j1?&mdOW=v>9?fW8C9QZk#<*rtN*P6BX_}Falo^=^i46jp(ZC{AS^KH01`U{!! zu!^?v{TKPS;pnsI9$9ep3>1{ThP16Gxka{z)n#{_Ju0< z>F$HNvvxdZu!}!u$SIJk=ib7|GbuXW$6JtQdJ-!9DxmF$9~7TWh5wo@;dIR(lBP8s z)0vl;JrXNWTAhN2fAjZYez%vcZ^X>`>`Y^$a-r?pACS(lUwJWyw(3HvS_z&#@b1?HmMg64^M z(xIF@nXZS;&P}`%;WwKwG=V$y!VFUVZVR->ZRIp4R{&!mRhz(7^2w zw0R9UjBaBJD@}3yMnBxTvJcko+amaJ@fh4J5y5~~pqm4)vzf&i++)7W@ZX+sFxIL> zkQbzi!}_rx`RfxsUUC|?IXPLs`TL#tAE>WO-miy$oDyN)tvTG8$X9e>l|1))nk02z zQw6$eC4lD^;A^8}Wb8uz+2L6SZ|tJ^_gW~}+S_pd{W8FUo6@K~^BTrS8sQ~38K_E? z;6_HS3PeQraBmCMiGyM<&8v~(ey8P;Hl;X}J?I0kR5mg;K0Ar>D1lvqx2)ivB_no} zXJ^WdWWvwCaJGie#NW~+zjY)a_2yN9?j%Pr%P*%pc1XbWX9gJedNSxu ztj4H06Bxf;#W2RBjkHE*9jh zZ|`9t>BFZEHG;z8D8jDrXS#wJ^6O4F{@5VH*wh@ssB#g`kUt~j${d6GySi|CO#;vD zJ%R@lyP-0Gv3@w(3RUB?utVc3P7;1bHiU7wLQ;r3IxYsLmfxjE-<31pypJ-qZ{{%f zUc4r!yUcM_Z3`STkmADPx`+YFaEEix^E0JXB5xs07teh`4$liEgC8yE^uI2cA!!T~ z#p|f^I$iv>B!`|9zm41D%1B1(YGVHJH`wvK3yo0&80Q$wdl)vMXh}7|H+_1zUI#1! z>XB6OUaLdjNYo8QL45rUln+zK-~7(VL`(sWTpy3CEIeRrcLAQX7h^i47BQ_u`moe% zD(9sm3%|!3W1Qr3ICSh0sWJ}dS*;gvG~5Ut-PPm*9;MN~*&f)FFV9Kbvj-cQ)8Hc` zi4j55QCUS7x))Aj4$LrPWwT=i_n(Igiguo5&+Thq*WA1V z!R8AK?ZXREdrdKEU;Y%0Zo2`|mBsrhKdHflCV^PI1uS`(3bnb4pbOG4t@t+l$MfVi zO_>Y*hBqNAJ&s%VWeEPg8N;dG0FtF=fY$u~`-QF$Q=Jk7HYr9hJyQcge~4&oFoO9j zWk5AyG49mlUEr@x2%9HQy4;01t<=R_mxUc3JD|atIqEUeyk~3UpSd`BX$rX1HM8X+ z1$g}Kcar5WOK{Ba0-m2djT!Yf0NW-xTrD{rDypZ$@H|asu0$7poRQ&)fI(X|y5t%fSOArzPW9*+VpFwkHtP zW}2AyLog=rG<@otL#poA(;K38bgqvrjAWF-!0+qe>#!5X=>#xxKX215O6H(BW0-~x zoWSI39VJxa@JXgL;U}W>AKxYU@a+Uuk(kJ-{um(_T)vX{ zghVXlo}+VJ2kt4jPU>QW8OtRz@vGed*1z^UT$pqajFZ|RW$8xTTkH6DKEr?!l7Uk}pK&A6)DA+Zfa(i1@vwR&Uly@m* z?^h=;%R@sp!VSmf_e?L3h%%JlU=FH>P3)hC51P?Im7*< z60mXHe?*vP!}i#B!mX=HOqwDGD=UXdUH1d>5)1Lj)&t~d+-!Ui)B_8DGemU7BZ%%e zg(o(hhm;8xbad}tm^#f6s-vx8`@Y%m>}@}+@H1kD%I3rVNL|=+YZqK8Ed={@i;0cM zGgwlv04JTjMO;7i5n)8>*>ptj^X zZD%c!cApdcjeQ3NnTueMKd)K;44~l+I`HHwQrp9R@S$rgEcm>+$PotHb83}KGfV%1ppnF3FO7(uyOy2Ku@#!!2prrv5d~OKl z<{#u);tyduzq{FUD2!}*^8>z*DS#$TNrpc$VCmRD`QFI zmo!|l^AsuCtjLwrJtv*geBbiOJ9zIY!PrD}!RBBWx_T%D4>Zi7@ddj{`mQ8;Di6^2 zXFUWje-GkbootZpOR)D}hTXSq@WneV-kl^285bDAOC>qbXdHx&*UoTyu{P1qSc&XU zK3iPSO#?%dh}GN@*fwq!2Nj}BtJ#p?hQ1q*#CD2W7T^W zg38`e-GDJDeef|Xop&5U{l3w`nZ4vA-x-RjSw|P_|BR1kEX8G;?I1_!JIxYv=8|$K zsD!@3Q_%%5BBz5Sq?Xk3XJ8j$WsDk;0D0dt*x$brovYmN_V}&X(I|3tIiqI!3RHeKX+{mh$^C13iD%PNWTmMvOqXAQB&HJXsB+h(WpG@D3=-}09 zI-q_)hr7}*fKYWlKM|P(7B#|5(Aiu%FW-dQ;+~84M@#TRRvvs>@Cd}S-lAxzDNLUi z0Pd_P(>(Dqyk4hGbEht6NQxr<$Fp}nl;qO+;m_&i^1~oDbsGFtKE`&?i}dNGaga47 z4ilCrqLkwk_K|NYbj!(-KX1f1l~)1GsU=5PM}u3S-@O`!hkQ|DisDaB*Lw+2)OxHfKhpLXl zG;QHCOyn8Ihc4U}n49Gzn|hbOWSilIDN>B}Wi>|FQ5fX&{K>z|cj2hgAQgJ2#uyD9 zWZl)W>D|cjprgun7JjWIQu}$Q*vKN@ubW9$EI1}mi`b4md^gnSG;j`|RhhMo%h1rX z1T1Fq{kI`G?mOKMOZ0e$-{O;aQfW5GID7{0tj%!QKm#5ptl+lHu|fICc3eeY1Bx}x z!Yetpf)h7+zHw0w^^?!V{$^detgL|@$$!Nzw=<*8n?8`lhP&t!dz4;2WC6GK55xF@ z9iVKw0sh@S1|z4AW29D#b&Etk+s3m1Q$sf4&@D}H=zfo*Rb!cz8T0W>^=T+LA^{`4 z_fdC(5%=(J6N$Mz57+Tbp;paLbgI1^^T_cSG}|0P=SlmZVd-xoCbt_byAMNTRXDu3 z<%+V)-7)pWO!&Uz0^41c0n!&|(tqmbaoMwCtem1pBtKNrreCMf{@o;Iv%U=}aKGz*lYy*b-aj)wTb?EXkJ{7b#6| zuXkWSd=5i}3XZO}`U1LjD%=aH$*_;_>Azer#OS&i(tWY-$Q-9_$gGY4Ta7k>$fgl^ zoOMXxxV3_Ywn@MpO=0GM^(s;|V*~8F8iujYk7M0EZQR}%funrM*mC_;^uKQ{*z~cT zw5agE`|v|d|1WW9G4~)I#^rD;i1(6h>BmjOQaG#-2>$}E;{YAcT{&EjDeXdx#~lMk zZT1ZEcy7|X57Q^pU4{TP59JB7Sylr!jn^z@VM7F zoYnS+Ubt9luBUlo%z;Vb8GS4t~EILVIG|^e?A+RDh%(v zW69YmEC$0}Ea{X4tI>ZryD0$oeV@iYsd)+Q`+6Y!o(A#MT?waU;;j3>CXw|o_Q9<& zeAYsL2^4R1fv+J8xZgiTapwm&TzcU(X*tZ#2~TK4q@4oZPFzZEE7@a)vpLEOTcS*U z9yyq|gi~;H2kWF?xWZ^6wN*EPf?^@m750VD7g4O{#7Vd$>6`WGowhW^(3>=5r3&jU-KhxM>=Usz9vFCB=qYdn&Sufa!u7pPW2#E4*E4abGkLCnPk%8(c z7M>jL?*n`On{8|f3Gn?tXh!*_B|Gum*QOq9LO;7Dg z1hv0u_@5KO)#I~q|7Z&S1sOQ3VSw&8mJqF%J+x=54;>iYizkMYU~XC#8>V#;W(2Rm zFXCs(+L2jYWPdz9d>;+P5-&(=W;gV_@n#PQPlOva*KwDlF0q|ghF{ir(KF+eaFyZ@ z*tIl(i8~~Pd;XeOuW4+A6PM!2>eq^39@a}H9LQ&5>znXRRT|!{FJw&w4VY_p0n#PJ z*-$14T>p#3u~~7rJD?R~G{~#yG0|5%4 zU(femdV&NYnRB?@h-;`bL5Zdu8bW4XG5AX=;RnrkWX-Mf;1b!$R@%zK_-iu>^cUdN zJ?ikjSe5&EJr&HuXXE^tg*a({9Nbc_Mwe)F4C9|!{jFaKR;}h1)-I=&Z6$27Ofp>i zxRA6zsUQ}cp3*?kQ}kT6B^mQ|8h6`rCVZ(a!pEWaAWC`xKC*s8%-ed%r`sivs+)_S z`XU6+a(n2_q7zWnEeo5P)2YK#fNhs1avCCfnEZpntY#(dqe%+d3?*Yw@&_{M<2*=M zm4uRzL#;Fyl6i8WU@sHMGy6yJ?uqf-vhmv(t)M-uhE_5WnVLprvx}fSYz3AENHYq- zdfd)_^P;CCP-h}Yr_VHK>)lg8ChqgN!1jJt-VJ7k*qT(;DbUO+v=#!?252w?%vx%;teiAN^u0)Kh7TPt6FJ< zlQZqze~gHE&f@uW_vx*^@7hO z6p!NH`ewm0Cl{2-^~YUC4RG*B1nF$eA#Li(sNOD36;i@6pnfbfu_v0HxjBX%ijg2u z20$)-7DRfb@)^`WWSZL@(1~3Ohec!H>y-+8KBu1y?(i0D zc31@${#Ah`KhrJy9g6Fcy78})0pu)8hUK1T1^;AY1P4B@WR_hFgc;e7z*P1+?eu#I zRw7F#l}`laT=eYkoAQ41Ahp*hf`+C~taqDAqgB6&)uWDfIO$ivc8%TjF3s>4)L9yKmG>#{72amtPww7KPc=D6fiLC%1 zeh2pL-cr)!p@(XdJRo4bG0fgCPlQk%^VRw8Qj8J4vi^lq^M0d~({q7Xnh~aL;@us4 zQhDaWd$y?35@yZJ7FY}%VYyD80Uo9Xhew}d&(iw>y^3UT_-`F_2VaII%i_WN`84LV z)nOQ&l>lp&donkS*I-D6Fb$k##H`A6VZzoPg}z-!=(TQb8gxvP8`~txZM!&&+O4`r zd-Au|eV>^NPT_|I+uKCpu}l%xzS&OAep@rd`aEp6=RGi`eD_vX3F2*a`A(~ez|}$& zE*sb4KU)``DH;J?f4a$@cTP}d_Ew;Nv1SbU88=*5ybraYkdA zXBriB>$Mue$~RYN{(N=p{iKB__I<(Wcif=xK_Sl`aw8rWoWNhDnmz0Bm_$b_^I6~r z^v97l@V(1>11B$ns5k{q_U9U|&`^&*f92wuomHsa_k#>>^uqT&icDbAL`LkcGUs(- zjNn@5AF7`u!(5=x`5oIPnEW=5*)t;w{=QiXU(Sbtl-*cn{e#uK?<#%xcEIK{ZC&Xc7Hww}ncj_F&)14wLqsM~{a%2i@kubvC@AEm;TGnTMO-3dDa7>LJpFED)*Sm-dyrT!3MK^PbN7JZz+gMySBM9rxWjeAalUFJV@Lb26WM-*wF{f4V%G}k=r7V_=TW(6dCPzYK zb|{^$CP>y3lW%GqWYfh( zlgMbS&}7(2*Pt$m&mc6$LFBB5l&e??v5#v&H>(jm^g{%PAuKqa22mHi7_XEB{8D?E>^FXdPyPBp z_?j;`wLQU@!LOvB=VeW8xr9nKK^SbB!e(wQrKgtVQJdIExV@403}%``!>U=7`A?78 zEeL}dQB818(VV#{UW#IUNAX+kdyMRIpj-Z~#@b$ez_#5)VSFP=d31+VM~`5eXdCYS z7eMwb@5Lj|MX+F&7G~L{;X*xWEL&TF)f=MFwyhA;EVr`hn-Vd5o;Y_w`zTx6Gy_8A z+Q`4gK*(95$ORQwQd6fd7;ZI|sx2-d&Fw{yU&Q+poD)I(>t<^~;{~jX{s&#d!OWod zdAM^x0Q+nwLGwTyF`w2*pLU01nt~>`+mruW3^KX!d265`d4N`CNOFIK>gXL`UHYDJ zWnKnn(6S~S8t*kseH;s!4YsDtAsHQ})UXSdjcpeUuOEU7{2pM#w^-^lKZAGV?8J|Y zLa@3-1I&-+q0VR~_I$5_*Xu7~)*5T*SR97(7K?GFwjR-GmVk3<;!x+&2stK05_aOv7A@%hIcy^5e6dka_ z=__NH3Azs;V~-h=mYxeUexHPwnRfJ%+6j1iQOxRg_Hy#-$z8H|v>v8)T%cJBD_~rI zAWoY1mYN2gK`%_fRUZyRKc>Re&Bd6U|C-$N{!7+H$}p?Kl0o*THn+Q470v{eV*M&9 z7W#L^fW5GzHJ~{V~dDvTvV>Wo4C4}si93wr51&rxSi zgYaN0Ovy0C(X<(;WcZF=yv4?qMuM+PV<7maEo1)p2KKJip>H>}3Bs?Jz_YI9bf@nu#!9tH zkePgz44cNnx~4Tm((NBBs5uFEC?8zDn!utUb4K#?FeJV^iyC)7!K_`z#C9-N;B6_z zd6a0wto2)QU*1f3mF$D2Bd^H&PBUEb{WH(|JWKSZ#M4z{_Q0Z4A#~DmCtL31qTxL$ z2y6F7i;y>%y5JevY43*$#!Uj*eLrabM-S|9)*u_r)9}%uFW`L2gp**ssc26G`TXHE zdBEJFd&fT`eK8dC{d_Qh&tXC1M25Ucvu^t;%5zpiu{U@Xcw}sXSxe$-+fIg}{Q*0e zeE$=9CKwEQtphSqb!H}-uh8Zi*M_jwLpF$+a-ct;2A_Xr zQQ?JIP}bH!4WC4T&aE7<%>PN4A|(j)eTVZsbz%E}2F$ipFW`}%$M{E8(YIc^JdOp`{r>Zwqr7X&BdW8k8p9~)L@jqVQ%VNYEd^*IVK zf9iNfpq`ED|8cM?6ia8=1{eD1@{ORZxdymH78oyaABP7%lkZw0n_ zyR3#U@8qCng9y=$S0m@+zoPR&BXqPL5xnANK!Beta$PPM;UbQ0ax1aP>^N?nzMXku z(Z;?E`3}3q12B5K5-xbWl>QdgX1tEA2V;pW@T`dkQummi(is#~&m50$ze#ZZGN-vz z?`7x^5`i&VRwV0u2Aq-4 zaz4MhjIHLoM_Qk+lH!M5xZMU|Xl^rEQR+(U}FCP%6{f1mgMLb>SI|&w!U&{Uc z`jK|clHjU%r%I*09^^$2!oiq3{8=3ZM-4o%Kg%AJTwf75@s=ojiKEdeWpo-pt5(oa zgV&;q*xt?^b@@wfL5Q3?_9i(&fXo`+S=~TJXSmVh8FG+RJW4OsZ6X$jw}bpp49Vm7 zRCk?J7~v3WlzuP;ZxlRc8>h{s30@NSAD+weyK{nGM`(nc{> zp{f|3@^jASOQIy@whrsAr;h(A8}gm+8BDCrVzhkwoKVLETKAdHny31}rJW1?Vg^CS>@69Go`!*JOy47uif_#b6w8cx+0wsG^EF_~p5LuNvp zz1B)88Jaa|K>jEdlF}rRS%##LnbJTdlsJ3sbxcW-R3wGaoF=4Dz5D(AetP$Hov-Jd zz4yBIUeEK~_x+3Q?EyV2ox$EQa}Zr!3i@qRsm%VBxX$PS_N?CmqTWYy)gBb$cRpu0 zO^Qv_{e@io=2t77i(b;gTyuQ(j;+HjaGQHOX93PU+f5s5DRH$d%jr|YNn9qeAGqTY zAkwanS8(e*)Za0d^Kbl~&%tw39)8-(Gk}`zCpw|So({Wt15^YnGsX(} z99}5PXitAZ8Q2Ga_FPGjQqlvqDF<@P9ae)2dU?1HSJ1r)H>t^qD!lM%EH%$|DPFSp zGB9JxXce^#{MJT+Ns<-f)_&^2l~sduewZp~&oiJ_Y_$gYd&~h|ZqJDb5oLB4By!vM zgfM6I2nc>I0m7LCT4wMiFh6k`I9^{2Ja(ti$=&>5+rym>79&E;VPt`QwwnQ)2yvJ= z`GGpmvg>Fjl1eRjMe|H6@RH}ctafiPNDg(OA`U;qc(Z1`+umgQdTkCi`m>q(!7Qa+ z%J^B1<0<@W$dDO)w1Vm!k))3M-^PAF%&@Pv1EtsOeMExk`$=QbGmTSxe(1>*AP_Iuk%%xG;{J<$$Zd9Bkn^F_!lJAn1TM=lzOII2>0G zmf!tBUA~b6Pu&lsGVkuA9S^18hFJz+zpe%|aOMaos!syH7GdPqU4?Z7R%h|Q181SZ^f#v_4aP_lX0C%nhH4~qxh<6|HE4HTk z&n})WZoLaskLbbiSuS8$Uja%O%hB7VK2aUvG7Mf;M%(DlP*VeK;BrhVzUXcOR=pNw zW`!z27dJngc}xU!y!gT0Z}69UT+|5{-O9&}?U}gS<1b!tQUso=ilAY)HDm&%fOC2@ zIPUWbA3h!nJOv)nUJ=g#?{x@w#_%@<_+{}Xa|2x7DGg#5OxH{GZ2?-%q98K718B5# zQP(Ga)9)H*Q#HBPptMg8%nE14w;o1VZ1r5qi#(({))_;ytNwJdu`ON^p9XeW)Y1D+ z&jRu?kEkEXQ`CyTAFvI}lYC&=Px}l10NJi`thQvF!!h)LWk3=hwAW-gyHW7dr!SoK zt-|bDFk}{q-T_eJBoL9X05~)ryzlOz^lr@n$0HkUriIDm_{pA%qqZaIzBkD36B zZvXnE(F>IOx&v4}P6J>6ya?n>Hi4%bMVK4KTfmCvdQ5w_Iv(4In7l*70v z-8#353fcCJ=1uj}`djbeJqzyxi;Pe#y81AP3pE4xy$<17Ro8&zTp>K@){aB@S2MGm zAK;!9Pw8{%Rh*!S5BR9jR(w_OJb1U)42Bf(z`eC@u;5D!9d!N{j?wX8(u>xCe)mt* zF4ETMl*6qAN90aKcdyN_&qUm2rGMRj*^m9Ih*&K05d?ld&K18v8)ngq@KY z3VOpJr!rcrs|kd~oxwlvtiv5E4&#Zf@$@sNv(z_TMST0>X38!k66oD_-dzgaIe|--pD65mpU+{yozOkKpux%f`0x@;9AA9H5P?f`K2fh6 zKU2cP<6yV`LKvhOz+AcoDLLKyAS!_$C@wSy3a0K@ODB=#o_SM;Hzoj1<5H}BcO0Z9 zKcb8imxA6p0VsA-i(2`&AJly40#zg5XyY*rxXj!Wn@aj(*TrK@IIGo8oQ~nj_vYh| zn?``ckNe>2`Vycft;yu<_>L=5r5XD+_S_w_xSNh;;?u=b)OhlK9DmOWTQ!^oy{VIw z^ODuH;cEfrlCL0m8qKBJy5g}kt3`UQ<^dw6_OZO9LHf(vG_0`dB?z$QVE6r)RtUcg zGQVZhn?1rXcfuO*fBObjILlBqR;JXGv)T149n0}_y8u)4!-^_uUJMR2W`h*b*PQ-8 zSLodFWX{qqE$~+(nhD#xl8I-vYyZwSV3FqnU~|tq>clf?04-hdNB3PIZ~H)f{2Oy% z*y>LU=*+_MRfZ7kVOVduG0qaU$8i@_abwYEygJ;884bHYe;3=yI7F$y;8P6sJID|; z?3choU-BrXWhH(d@z|jPrqhc9cQMUkmhk$bBCM}q2}jcdVc)z~AR#jW0^|H?%@3Y< zmSHnxX=?(`&o==Df78I<*P&dEEiKf6;yL*3g)?AXNfJ2cE(2#i&ZU&LFf`1k>=aX=sLH;Tod%K{nMQ5~@8@qIe> zz8B>#HH$G0k)UR?`nCsB0&uU6C|H;_0ydQ{#k!7W;QOy$%BLX*+&vjjFP^!|9MsTe zroW_foQ=KlB{pyT!O0lP)?+m@KA?>IN-RO2s2m-??h`QC&!hhtU8Em1j?v0Kb3n?c zXx!gm3s&Y-QvNmb=pH^H=JeI0Sl9j<4i)qVPXazrIt425dFJ5X0?GsFN9(2RvN6;696G|J@KibDe8Oi zQy{i~hBhoa3XCUzVR6}d{E^k4ex=OdEYIt}B;N-F2Wi8%S3gsin_gpXY#d(If0Ej< z*cFR$8mX;2E`YXMLqL5r2VArc0~4>Cutb6&lyUcCb#r`p4qK}Pl?v96sENYPHD%!M zMop~AYIdbFn(*U!R$$)ZINb48k@IQ=ptmIi;8k010e+Q!YLl)NKKLw+eppg*_e}I! zyw*mW)wi9)TKFPpJYxcy(*+=tq0Cr_mtw2Wh?;L{1D@F4WF{0^v1ByMrFX8v#`;n? zeX$fRo3IBf7Y)$ot_T47@+R=1cq0{Xb~`XPS->n#OvHWSsW5U>kD1UD1hQEP_|h|P z(EUye$8(nAs)PRA-vh^iE<4-KeVYJItE|S?G&4Z>t`6EDyb(_ts>2nXW+1t_06W)< z!;mF606T^R3+KJZPxfbVQfizi9?ydMRy zEThazwej7OY@jAAK=s5K0=0Q@Aba=%celVaSRDV95@oqcJ##Iw-K^u>eQ8O+`SK4; zJ>v&2g+5{lCq3?~mJ#~Y5fNCy`m?rN2?TeJT0tvpLVZ|x1y_<$YTqn%e2uCHW$`z` zqKW`q|0oUEjywdzd%9`Ef2YCNwmHn$*_C)LdO^9?H-MFA4}g&G@%Vm$3^2LVM4K<& z1B_X3&Of(Nt^l7sH>D&7EHF)WxaTd#=zSDtr0;tIP5(Jqa`p>+)05>Kdk0`@-a6{6 zs{t@djpv-D{&E&k*J%yx2EOyzg5<;oYRp>&KDUg;b7h9`>@Zt;JT4Ff1g3DiK1fk- znwoL>?nksBTm@oJr-9=;5@?S-zSzv;QN7gPF?vg4GPrx79OQ@;&`l{2+XO~H5qT5r zee@dDA{#?n)E)yrru6at*oBlzc`uG#A&sxhzKFy2<%$SZ9pH$>F>Ezg0q)TbpwASV0Odq6cz!?=Eb)n_FMq6|71j>{ z5Gg_BpJdlJZ9->d`h)xyWsr$XDRE~}`lP!wWn7zu``cqF#~+`7l*1F8wZ;cLb8*=#8T*nb}5Q4(5MZj_&4LHY|&1VXB$GaX0GPiGi2YW49iqaux_}F5C!iQL` zq{b3>_`g2vBwqkt@f`ql_G{_xM|BRx+nnfmeb2z*ra;bbqcb3cH$%UhHUP(-b#SB{ zcvST|X|TxF4A-uer(<_2gH0XAcwplVx;ESah)nk2rlrN8{Ou%H`^ZTg5jsP=E{en1 zOW7J8I@y(~%_|SO+=;is^VdVG}Xcj5Nf1o6GneheNp$S|W6NSqT5pduY!HaiyP|WdW ztl-ZB^A7(>cx}it`C?;70&gG5V)#G*4n!aS3*XMvWO>OEGj(ww9{|%iiDv1U4YdO{#bD0o< zrQqunkItx+gEE^GxOJ70^{gU~V(Wgap3+XmG4DXf7ghvC`t^tPa3 zdXm46x?sHroY?6KN?(@JBk8Bvh}kvF`oA%t_vcKj;m`g1FM zbz~{t=xqm!V?Tj(e+O_;ZzZl_-g5K*#Ikc3O>Q3t1G+l~xDHt}S$th^AnuEh#=ncVfiGVz zS#51J+!A&G*ldvo*Cmp<;RYq(`uuL%JWvye{b2Lc*jx{n!5NN14?|~%H*!n71n?4F zFRbu}GF&?0oVbT#ZqMwY-pK%xJSP}pkfRn@%>#2jEP7Lb@xgbh;T#TQ?Mg-)0|6*upHkL4+&U& z-i#{o$fb>4?Kp0gc9he)GR^?KlY)Y|csN2C{0cClZ|T$En>D-E3InL#x!1tQb~8X0 z9|a)`t^oY>3m*UA$kY$jGJ<30=|e%fV2-%~ef`Ni%-PYyUFj{$v0**FF{UG27wZ#UFxO@Q3(3;97D4|F<;_FV_rdn+b@;?*|#yY4pdLZB6oKUc?f4GBQUZ;ICOnoaknjML-ow)lLbH|`J4 z$1B~%n7dJz!Sn5R0k2XDCn>(D^Dr;Q+<TC1CMQoICssM|+zP&bF?qJ>+sb14P6t8|>oGx}){pw3K9(Y?*n+@S5( zfTItgGc@culOAGV*)}`ii=@D(=q&u~^&$tztEY^k5@>~>5IQ!!4Oj*X0{s{%yh)4? z@XN`8L#YFRzHu95-d~2jw(IbemYSyiC$}&1P^S0&o)ku2P<#fS4^?$=+F zsKuY=9DpW{iGZAJ08gUgpwXB=$7<#rX#SPJ&U|KQ6|Ek6|D$)5a#jsTVnGxbmXiUY zGk$apWlhg6Yyt0|{-7GyoTCFyCIKn7mbl)$jdFkQh|TVpf`fjcB`FEvK6OdJQu_4vvVk`iUE9Z9RUtC*QnhM)@)mDGvM3JP*)@>saX?` zKzi~wYF)q@N}cUpoNzh|p61PmF!9&*NSHhrZPMrJ!$mltrb#w*ac;3ju0 z_!ls2rzw-eY>rf_B8b13iD&H%$2zWSfsn)tph`WW(n?Y(gH)P2@m_@4X#NaXmGxlN zL^I%Ksl;Udw+P!^yvI4}6F}h0~_A=Hgy_X4@$KT+1?cTbDC?`c4CZO`6zh&J)gP^*+48>^0T1Z9Q(^egu#2 zc>}Zul{qnOc3{C(K^$f8K&LyTV?PCBY&1uRIX2QpT{#p<%MZ!HdRC(ueJlpM(>6@G zc`5ek4CB7*3B+3Gw%|pNnyG0uapr;kA}sTzj#KXujcw%SQELxUR6^g{`WK5rp`c4R zPmIeEoijn+4ExNQ3Z0z_zDNi`kO-8>d*Vfl`{+kJ~m;Zu8P>dOvb{(MKuJZ%< zCO>0!{~FHM6u>(_%Q4DxIlv`8kzW4B6CByo296ER$B&iw;byO8VD;a(ROJ<6ywX4f zU$xQ(cl(6F<S0@_Su z@u9aDuobI=+CH<6))q8{Z@$WN3xh4dZ*@(eCYjFNzq=X7B>cx5Uo-)d3XD0Gz5`Tm zR4sR|UK2gpycsqs?+4G++kl{F7w!8@3Jj~s)bC;$tfP;&0D*7T@U5jI>z`+t>L(JI zh$=~@o~pt=Pnz+v5l8T``E7CF|Sw(UV|SJPD!-n$L6RHA%t|yMF5D=o9+Ujo;v$!5kQJoy}jrBY;hk zcj1q*A)L@LV;IHGO?P}u!b^^Pp?CZ~4W3kL120xXDD9JnQ)Y93Y(x(AFd-L&_QZo9 zS_Mp}=PY{D(PX^lPA2e7+Ql?1Rf5|{9x%=tpgwh-2A8|9gZ3&@D*nZLeE(=KHOa^3 z>+lMI+52$jWnD1vJsyvDxb}ha; zi8YK4faRs%!DiXFl#kpl{GjFtGd__7vX-;Hj+b8@9=&ekOn>IDbx9ipcQ{SlBg?G8 z4&7+HJnXDG>ab?vN0yt@kK|2p{{}N|Ue`IWO>r0Ku+o8cX#%kK zgCyf@;#1FOl}F8vmE<;mLM!fxv5Z~ z7j_I^yQ6?J9l1byycYY8u{!F$9aLFm7ac0J5a0egZ&BsLP59%dGS1JfA)xza8`#WW z1@1o1r_Ua7fPSs#uv^oEyL=}TVU&Y${nN}{+@zW>pxck1+5UJR;MAIdt@2@%uWv2g zj#_YpdFARiOY zc?rAl;pN`G&~(Ed@UN)4eq-5X@aNKFkh+T1 ze(W)E5cpieHI<%?n`2^`vt6&i#ify;eYyq?G|s{n8TJnRIX^k4Le7Ej>I&Gvjsu@W zECPaS^*MRwDwxx{2fWiPpzKyogWiB#eDr`Ub-1ShxaG~qjddnm?%7Ce?(i0uV=4T% zoex(kvAxo#0LvPc;m&Kzz>BnZT+i(#^moO4(38uLpQuG*mAu_-W^fr*OLal-Awzh0 zvk)-;Ef0cq6rosnCWz-}^Hqg>!B5peVAYGDtoSl+Lcm-$-|RG?PZQv^BaPJ`Wq^o) zg#b0Tf!TjB4wd)=f&>>ap=w`fj{O+kr?ic_#LhunVyN|kIb3Y@ff|l>1_~|%^-ZcB)boSp%px%x$b9}w-4m$c)K;iK z6K4ZlD>F`S%zkHo&q*44KN4d0Z{EmkcfEk`{rHC6cGPoD$vvQ+TXulR)+%Z#A4R|Y zc#WRpJiz_CKpj*px5xiAOyT&pLT`#!wZ9OlUw!RbhsA%>!meB;mKU(bUtlWpu`H>D{ice<`Wni;T!f4cmLz0LIq| z!gh-}aJbzb&MY#63IgxwobWgFkyn4Pe76|wcv1x{d$k8!^Ek|^ODa@nrUK)ja1@Ij ze~a7S^MNPxFMt9a6Fgq)Ob0#uLd}yZqQ1T_1P63m>04VHsf4Th0KMfGU02|NT||el ziyQ`@qFAon-6_zu&d;HqW0f;)O!;KPIaG`cBc(a8c zCzr+n)wTstNL~Znt<%Sp#aZ0`rJR=7EDp=}_JGfeU9fs|CkT`E0||O}Kx^d*O1?Q9 zc+0C%+*ChcEY=0GX72`vSP{xSw$ARoYz^#}HPSXqatE+WL*z^Poku zC}*Oo9KS43ua{x-r@sW0f-43q=vHJwsqQVtgF^^!~G*ty5Nj zkfc*!AMZCd-;xEquJq&8*?i0m-v&^Fw$dJK_SpWh<6!pjBD}Gf^}1L;2W>B$;6AH0 zY)(-vu;p*!YSb2j!^fY|tB%fqidq$H9$`bt9_$3A)}2&}{a!F7U<4nO$v}Jk1pH1} z6qo09jE(D3#*LW2 z25OXaz9ycpEQot%2CzWOD4rI)izO=GfSc>gV4OH+yuXPvF5PPwzAM`BmcdbafZeyo zB#JN}FR^uFqAV@IG8CTWa=0ZTX?WSvM3{9x3dl>XfU3>@_zlZnoqMVboJ#x;qg+9z zNIC~(DY3Z&XI9|kAZc9cy_FJ5wWeS0C}ZC-qGB}_???EI`T73$v)WZqh>uT9jPL&( zYJ9eQ-urh3>^k_LnXmT&Z?i?#3vI2;%`Gh~7uqheT;^!Sx8?tTY%5Mj{`2M{YqxT! zv058hSx16TowFHarpYt)|OU z31qP@8#YTVMw-0uq+Ps_h_9|A=ajh+6){lYln4@0L1^!0Nu+g382!#ph2qzTV4rFn zTyjH#x80H^U)q0@lJXi@RH%+-_FqStkLrkeZzb^(lHjS-yAUTjk^H`Y3ZV*K8Hhdv|M{0;+=6!O&NoH~1kx6LNVMW$= zU!LWc>jINx2PV@^HGV{}jH{;NN#p`nL-I^av@e3hjDW417U)^H8YR{oovH<)s zErH}-_b|%chWO-Y98+es9M#$%h2LLqLNzr}WP9c)6KQvdsDHgeA_uQB#uLf#zki>h zjqq34t167vm8tNAwzZHs0`_DqC%Lqm{2M(YW z1)-o~Q8+QYnoZtRB@&_MnI!C}6nV6Z!_Gbv(5|P-s5$I8#15yS*~eC*d}9mIc~eB@ zX!oJtWifDhG$dKa2cT!!9|$GZq8-DZNm4)wDUWX^>2)?leRe+Pd#Hj~Rw>$iJIUcc zVOzp|_?&q7l#}R)&*b&{vryO248`+&(XL4|<}&{(C^YgDhP_rpdl&zOGr4b|(VpAP z)jKO0pR;jHv28Jwe%1;P$O$2*WPr4H48VH#8QLZ@ksMfgi&PBD@VuQm$dTkTeatL@2`C^e#e!H4v)o^okmMp8Y;$=n5V(Z!RJ$lQD{ za&7O$Lvsrl|8ZI3$$H|o%tvt1%{|1zpb?F-u@##QT42PDdFW!xU3mMBD4KenfJDQp z;lHCYNa=hHQ+2e1k`W7GELQ20Ekj{U;g&GCTsj`j^43O=&&J@-d9~!v&k)kJJe+ur zui~j$i12>pXOf;HIqCTDyq*ia5{8<@86du22qyth= z#9co$*zAEsOM8iLiVewzkgROrAm;!bM3i5mjEwUK(fi@6WUu|I7@C#j5x zY)eJVA`?`3>J1uJDT7QutW# z3RLNthn{x+f&O;7XqBNaMCD)MA6*I+n?~aeuh_MbuZyf-6~o{IYUtskQ)p{pJvw$K z0vTU?2;Z%#C;8f!$sgY`a{2oV2(y!5P{(;DX}FwJ^yQJ$ZFdO&bt{zGY0Z2a9$@BX zU?Lj*oowA>$b0rkllLd*HhGhGk(|i9P0BV!lk=q#B$^f&x+tD!?)Z(9+Vv4kaQT+@* z;&gVw-D%P!n5o2DM3>@K(em(3#BXTxK^u9d97lhr{n%&5F|<|DfvjEl6Hi}KMq@U+ z$!h=mjI`qp$UlDoCUr$X4Xr2;AM_Fi+3$tRYnH%%tibEwW{_PK=Sj?AYZP$)E?j=s z7#SbgM1<|_$+aq3(lpot0}4)2osZVS?)8`8v&~v?-mnQ8m~R1(ybFVB%U%J_Xe{1J zB+(m5U-akuQADR)Vul9#iF%nWGR=}e!VTMyE#h3ViapmQ^$KWBS30zeZN;V4Zbadm zDsSdtKD-e=4jFS96!X{+&5~&Za~FG(BfDylO++5z|CNmDJz9ti`=9OimJylP!*I9L zI1KDFLy;wCK&ssz=FEd9q&7jftC3x2U9^5AK0WWsT z!rvVRal7{*3A-r5n~D+UwV2A0SRqR?6u*VM|7=IhYwCdRxpKt0VMt8Vbjdy2Rt4&G7t66B<4@GWEN-yA-qR=_nVaEDz z6?rCd4c`9UkADbKq?tW`P5Dry@qQPY?XwA;Zqh@aFT1esbboSd=r5B$mdr$C#gGL@ z3DGuJ;weVCqoUh~&^>YpIqBDvN6isrS6>dX_fCS*g&A(tOBdpl_<(B@?>Rp!09J3#suM3AlL*Aj*343b+?gNKVP z+1}}Ocu6gbbPPRXn*3{-#LE%nsahAKUoDB3)iuIxX$;J~t&L`-U4;rNZ4fplz%OI( z;g!ZC@WhA#+BFbMr0i2jm8mGN?tvf5tQ>~T+7+O6HM@>)hLS@?QanS8KQP^;6z-FI z0%Lo_k&B`+`uocjh1c{DyWD)zP#;6K4qit4WRsB(yNJ7POYlZ2#d#;&r%ADb3W`&7 zL{*)oD9P?O5G#*_RvPn=ukJM_|dL zIb>aH8FNvqk}-5MLFX*q_P@n`aZzY7@-+SPM^GPVD>P}Ksv~bV1ad>Q-3a|JJA8+>NZzQPu z9qCD5OY+y50`-0Db?!(+!g;<(J;Q{zO-Y95Dr?5OAiEyP9xsDVEJKrD>oAmO;Xh>w z-7scn7Ez2WXKvfILZ$n*B&u==F-mztZtI^X`hBB}My3U0C{vBLX*r;HV^{R8JP56b z*^JJR2qdc+j+EDGqS;5f;Vb_PbW=|P?UHB*Lfq$&^z^_Gt9r(2-#Hj`Q4f(JO(c}F zk2!of1-FxCkiU^bst;!}zoj~uH~m*x?kgXPU#f=`cPEkzP|Q5|xeRt4B}|1z54nGI zAus9@MzJ@RAwDHjB&od~g@02;nk7D{+w~RG`&k%f3t6!HXj5|3@+9NVdS|Ch*Q1d` zL-5zk1e_J@!E@Wi$18BmCy|^`-|C$)D z@o5~E)QG#6>_V9;h}NmSTBVtRis(N(*_%m|31^gEFFr3KR;i3XzI zzadKG9@!I6PM+S;<0U+4Bq{Re$g#o_qAE?8#eBb@iN!D>3AQ9hT8>ce*-0m>X32&;(VtNgz(?DAZ~%f}REEq432A zP`{!UR?z~ysFV+6;Ij-bb^A_KZd1atJ9C-u@!eR_%p6Kfj=@5y0%A5lisTOMAWu&{ zfpHImDAjYUKhJkNlt@1ZldnjiXw@cScVRiHNnQe0N%*0b9Xd$c;5c+!qlg^qQt;)V zbwt{A7V>iFfx5>4gmnh!rc)cdEwlg`6*{A_A3Ct%fe|#k0FXiA7dTc_h$=UnLz9P> zBL5JK`Yx7|2-k8Vd2o;fmC?eA})lAV9ZQ^mIi#Y181ug9^xVqDcFkWW7 zkD5iK`d=xr&)tWF^(W!0rHVlIj3^QHo<)xJGsH^Ehq0Zv6*nu-M%Ny{CMT{`kjxn_ z;VrO41IuG65*9!a3#^`_)d`fGLm0uMyql{y0r_K2Y~g|aB^KLv8nV=WBtdIl>jud!L$58>bO z<0!$(2_>Wiz+bPQz%`92OyemptQnoiN&PzxA8MGROZ^GRdF4MczhV;^ev?5|U4=>W z%@jEO;{tM?twA=^_d(IJXz(wq04`~L1r@s}M2YT3!V}$4qd^i)44Lr)`(la4#ynzT zbPUOl#KYe=TjBX|F6m6uA=hk{W6Omt(5#E)UeqfSs|yp*Oe+ei*~FmP1^lS@0S_em zKBVR>Z)H4>e`gGfo{#{)2NSYzCF%FlQ-nhds{`+eXon`@Ub9Eat<&L2l;uG!>5VQr3R88RShTBE5dN` z1o%zgg@Ki`!RY2jB;A+}U%Z?GKf}$S4QR1!k5323( zLa%nlpozb$kjdr(92)INTD^B8)fK`B^;d(~r!CCc$@@h7v^KB4iJ#XrZxbo2@g(Bw z{zE?pJLa}fr)fu2>cU4B1$w7wNK?D+---c)hR3y(v1c{8$JvVwP1>>-i+ z-|yU)4>jIhheex*U~t$lS^X!Be4d#@__E)jjubU?qALUMdGdxq&vnUCsdYqbKmfUc zK$shJ3C6c4Kx@Gb_A!wbc>RC`lG>t=N`{irkZ=q+SbdNAEVhIk@^&W4+y&6~y)s%e zd>nDq0exgzn8<|OgIi3j(A|-EB)21jTn>n3#9co@p!l1JKky<}RZYzG`Ce%6A~Ez{ z^F90(p@^a_G|`n&Pqt1lBSs6P7>lHpDCv$bK?g#}9P_IrYia@Sje;oeD{nS$OL!gH zXL1mkl**t_$5t?+Hox)fSL<0{+sQ8MxCl6+|byA${cl$hYAq zd~mpl&1flMmiwfTLXS11+bfN1y(i0)GP8y^7ZYG7*9;qL=OZNxN9N2*MH1k90s0#C zKy3+SG$`{GuH5+?HWP@}{;7lU`k~M(`71oQ8}JtVzD=}OoFuo3WQl>;Y7%n!H*pvD zXHqM=flb^>6cV6`F70c8FS1Qx=shL0f8Hx7tS^jKhW0?m4LQ)ZFbsvr3!=8MOYrcm zX&C?L2N9@kBHG4piQWAxWW-I8^i+R>68Ce6v&nU)-$S11%lSqM!D+HIwv+7N-UFka z^6)N)0la4{o}A#1Bi0uw61sj1+VWsF8ml)zhwAH4eq}5adlUkG!V@U{MGEwn(#BzJ z17v@CCwVFUh?pO2A}_)=k!-d$cN3z@&LLW|CQ#ZMP)CDFPUh*#cDk}xZU{Ft?m z*c{F!nKm<|GHVyA8%zQHJsViIyaf50`IZ?ywVp`nWAan%6VX~I%d;pzNVnPxHj7xp z$wkA2Ua4F0BK(1BMF}MYAIgzWH6R-muH%4BbW!(kKy|7 zTxj~{I-I*Ek7QXaCsX&+2%_|n%fS#3ZluS|DjDT>d1Xc5WOAihJmSOyohD| zJkdAuyoNF!X|L8NbrCO^!O;}r;19_7Vr!D@+DS^f;)qx1H*(FvjTu+|4Fc9qVO{Sm zsAByA>a1yiSAMQU0)b~?oa$zH?gtlze95P0zsw;fnR008L>G2^p+baKs-o}~Cp1?5 z2sWo*Bn!>nFgx)Y=BCGexcCYO2}wDi{}kkSb`Cj2!bP07myHH3Rr4j+{Em^OI|PX4 zpgfsO(<2QvlTbA#5$<>0hxYZCKpXXJI3N%UL&|etOI9cx7Rg{1*<_LD5m`h>q?HV3 zFGO*-zrvwcpP@hJKgR!g4X};)1y_WH6Zr>w7|)V9sG1wi{Emuc=08}-R43j+lk%=sH0TbNcfj^PD`st zl7zZ?vg&y@)F~FnxfOM=)>;>aPP~L4a{PdZ-!1$yT@b4O*?{s6K7eh`O)&k+Qz&Y$ zjx;AtkeY5ha`|%w?e$kh(8?M5otNjW+Kx%df450ZJwP`2DctmIGXkUUnRCmv$opoZy&ZE7~IeV}D`Mf{xL5lU7C*r82 zSip_LGhvYWX*gcE3D@!;;iLd_p?T$H?0+|!J#MlK%&&8WX!|oTpgkS(TBR}b&!Too z$wQzcEh}uCwh{m948_-U4}_gpR{-KVRt_l+4acDfOmtWV)J#-!sfvG%yYE)!Sp*JNWX%UCrp0w0bSVG74>;q`$O z*t*;w%u-RP+fc3!GS->n2lMBGo@jsgXqhR@j|qn34-4R}Wi42#KOTop&B4!n)ZjMJ z*`r_o3l=UI1^d>Q06XdySR!8l{mte8wdOed;Ak2CSm23G_tMy1+6`V_Zs3p#H}iIA zJOTsu@4);;GvQdsQD6cRY6YuAvmJ91IxO~J{eztF_+dXNHoYEPU$qam=d6Pl&k_(S ztb~?=Ed2MyVO+g3AOFh#f>l4{;g3}}aH+g6j&6#;@qN#*;vqR>5a zE~p}P6U;WWfYz%Ng~QnrLUfoC&Iww`#srA478wRh%<6#m7XJjfoA$8{QQz6?;yu{l zr>FDC`LaXGlLR zKBR{a&J4hJU+5T3=QfaY%f84l?#ci0=2_ z;QG}nuWc0dQ4py?0N;8FWZg78m0)JslCCb zai6hq-equON&%=YGJp~CcCfHs0lJ$zz-3k&VYB;Nz~80$j4oVS0k%Hx8h>3 zX4$;j?FTl3$1hFcsuQFC5|c!E1Q-(rXz1 zGlBKTox#12^6XOyKiF5Z2|kWl4Xf_C^BO*I*%8H5;G}tucj2%h7S$ zOO7Y}oZ1JTjJSw+?1y-hb9``~&t}$nmIT~2?hZf1cZ+hSo`d1!r$8q_n)L_)*kbES zaP;6^;QJ*CzN93D)1trOq%cdNOw}v4yrzM@OUmN}U?YtK)P;@V zB96}S1t5Xjj)yi$32obU3rP<<*mCGE_y?!Jhq(gmx%3(vbT$Tx=jDUNnBgs{a01Ic z6d*E}fc6PL0X%yH{56sQg&A7xC#P3fY3(aKFG54uZ+{SeT-gfdkI90EQY&ylOCx*g za3!V^rSXq6%rmrE2%2sCaLU>jctTEJczc;5Z0fuUn!2jN@;Nhk77CkqjEN48aMuFW z73M(Cw-Z#l89_-AjxclUZdj_h4nEGA!K(V%z_)HyB+zf2T-T*3dD-1~1hNER!?Dt!l_~CnVoGsdu>Fabn>yJ2_aR%4T z`TVDvldtNR1}i9`^)S{*)`;Vbmi&=s=Z=LyGu8?bE=u0ZZpG`#&y zlm+|#BoG-_!IsEC{6$g{EBrf&weNq$gGu7T2R8_zgw7|{BFGY_4sr3$=YM#G+r+^` zJ$0yQ+zs|iZUC8g=E6|3Bpfcg74O*2;=v;u;b?at+_wZpE!Kxt`2oh#K6_t9lXbM1NTLBP1nb0d`YPa zH$C18G&KrqZ~W83Cz37TlX`w_;=#GNx&LMDrfd2j^YbcjLBundXE@U#e`_iG_IE39 zww5MtSW}BDpRNQM6SctjsRImL8P4WKrLh^A1K^DRUQoU%5j@HM45ldk#?@RJ-`(OU z)C`0;w(%k6oL?)%$2H+JNfYSfz6$D*Rx)B{vjhC{4Uuzsp5&YZmf>KDsE)t5SO#WQJGcL{;%MNY6O{X1)% zV#1I>HZU424HNjR2FAlHh1rINN`-0|aXS0*jxlgU8=S)U8zss0|yg z;?4gOgTp_>;-j)I_&{?Yw5r<%tJJ)qtknnXaIX)`gbm<%zpESsclQ9j`%bVmt`$GC zK8VeHEphtwWSGR?0{7NA!rn86ct?UhE_BGo)mtrK{X9uH!~GCUdGG^#m{|;#&DVlH zI;O%e-cyA4G?ayRUiaXojq!M5L_$cfI04!&H-gluv*9M)r{MODF7WweAe=j72YYnK zLAS>Z-pths(D2br_-Zr(eu?t}BVuYGFk%jGr}RIpx3~ct<)6plxteg76IJKiU&Li`WGAkF4ro+ z^6iOrf-Y}hAuTW5;4z5XpJd|zK{ZSME~!i1G6|e>q~V7aOA&7R0>(2=@Yx+rcsL~< zMl~XA79xfh*X)8fe|F;EZ_{zZ%m?_$%MW-+ z@@$S(1NfL&0qU=dLy;jd7Jq}_wZ&igo zZc|t{elgqn{t%8>d<4RWaIY*-E8qkdiC z8IGbbBHBS%aJHG%RaN4p<)5e{tv=ybtyi$vvk9#GssLY-(!hS_5nFwR!n#(Cu=OhC-Q4O2zl%#jM`?W#cDMuAu7UWdL@vHVPJuzMb)oAfXBa%_31#vg zfD_%KOtSYr@bvm@Q7>~*q=8Pu9{a!4-Lg#uf2a9CXFnk@dZ!6BUZ{$=C&z%<5j%&O z#}?u_3YPfXAs>A2P#|{Ne^R6rNygvr3&Dv>C8*iy3@1!yU@6UVd?9@RJGribRAeLQ z75)YfZI;1B*>>gO3+uARYC)j=%djm@7UFZXEB& zrop9Hysr>jc&rc>{CtV6{(i<*PwFrVeT1ius0%GCr^0(TazVxm0Mq6zg(sMQ?1$J@`8R-1p9GxhKLg&BC<2ooUBF$t z9oS=vJl3DDEUbK0#QwJw*BUvxg9oXz;L>v=wXPF2xWD5xe5)A^jq6?D9HI>$)cS&> z*S*A*k;`FFygKwvj({InY0y~s7|1johWq&|L3-shI8&}1mQMA9j&sk!FTbuqpC{^Y z(m)9|yDG!+%Wdoz3445hZ3MO+FXeg4-2j$11K`O&+wqPi=df(u2Yh9R0e(Acltq3k zu<4es?Dv-c*p8um@P5rH82)1eJUZ7I9=KZu&agQ^>9RKXI}r)q8cD-~^D*$crVccH z(+!TBB|toR5el8Q0fmrmcFWLr+%ou(qI-qXZhbefAp1>*2!<|3HdBRbzgeq*~N2b{n2D0_PI3Fs1s7^kac_`JA2 z-VL09ak(6{8@~>Roc&OU2Lq+L*?Qut`Cowcz%2MyYzRDVaw<9*zb8d{_`sqFQ|IU+E|~)lRk3z!97o$ zusR;^c)A*XN*V$48^42hi$-w1dImnz^8#=Aq$_NDUyNfnxZ>NO2B_}ggVF?Fs2MgN zo;Eatm&aA$xy5N<)tYmlE!Yx%&0GM3emTNZr_#Qx@e9Z|4uTKi8occ7 zqMYXy1lTUKURWrvBAoKpRJdAm54^o39`1Y?1M`SeSd@i|>pIS0+B5^+IJOapJ1Jw? zt9*RmlO4X48_w<$c@=o1WrM2S;c)4?1>oZ|d6>4)8_wTb42}Fnctej4%-Hmy#;TU=*6pcnS!j#c|iKq8u;Ld9awTCf>lx2ge9t^ zV8)~p+$OgZmKQ`qH|;Ip{uw*0HfI5_Jaho)1*yX5JSS)rW()iJ*TJ0^<7x}r?*RGP z^PtO;+1RSAg*W?mF__G*#dE#Z;-DWlvA5qV+)mfvY1dSRn+gv@v*00MrL-S7|I>#t zCuYF-3SD?$OD8CZ&E#bUy7APi&%^uVF_`qY7Cv*f!UtFem#nZ6%6|;USsSn8+t~|* z2X<&-4<`2Z?&A03vXlkFyPXu&N_qpzzn|m191Vtc^&jh=N6TPMJX<*NBnTJCo8T87 ze{uPu7r61|EBsA<6|4AeDXVYr3WW8R;2mn6_`vG`VddS8V840|IDRA&e8B-2jyvNs z8|L6j>6P$=pS=hZ-w5aAOA2lBbMcmy$#`GC5xg5Q3Z8tO4vn6;!7$@S-h!ej(DvDK zsMkh=h|nbC@S zz6o5m{^dC?|H(M@Yl<+;&lu+U^2Y0zvtv(c!Lu{3z~7SDK<}&_SnhceF6dYe<9oNm zTOF67o4*27D%FKcS6jeEW(M$hia69#=mKfosi5k1Je*Qv0`Z6`bc-;7&S_(;|4|Qz zSTYs{LjWaS-v?>?U0~{(2?zDH+ratuCNObiy3q7s4VG&i$9B8^f!!+=p?Xjh>}@*% zSJa!pqbEcIGra_;GUqN(I=TTacx5FlvHK8Rd?>vOz=_3MEG|8V=VG;#&a~fK-`Y&V6o#)Xi{xaC%LJM zU6!;Ce=9u$J1(g>yk6^s`EO@}8C6$7$+F|{>CYQrD^FVFe-#Wq_r}1=$9izF{URvc zBag=y8RLf3N*vl21Z6#RV1w2;&?Us-XK63yzdND~D@AD-mBRxnJ*e(ze=a-Rx`5sE ztQ#pMe*p()yMYo3Apb^QCLS~2#3zOjeBdp;FA%b*&{t19P+syua>`1TTm^@=9g-Sco0dTI7APK(wcyk zyCKL&=o93W+Vk~J60W#>t3$1+yWS#0dfZZ-MAeh{SN#%n(K$@mq#N$N8E)5?LN z&Hg|*H3r`)S6oi~VgT;?8Zng%x((!RsFf;HixQEU$l&1taF5Yc7DA&kW&=-cpdiQw{Vw z&K3SAaSXrEJb+_XUISq_qj=LK{;|2wM|iK?GeFM*2CV6wiw$4y!G}+^<7$VuKu{?J zT>T7i_Q5RJIHnEte|-VoWk&G$$4?;9dIoT8oGW}{6M@73US#KWs>00+W565#)H>-+ zbAV2&DC{zJH5_v{g~RTLK+u;naEea=6ega8^k^0|H?oFy2Rs04a||pW?d0t$TFt9< z$>lkHpfRVb51S<@3wtN02&W93#pm*0u^uH^I5*`pPtN)spbM_SGvaaJ*2ge*mbel8 za%lkEjyVff?&D%NIuff)=VP1Xk9f1*32gM%2iH`*13ly`Q1n{?nl76U9~}3Fw{ufr zHPB%vVwSVniXZDLPKe<&Hm>f`2SeOkXebQ)tssp1tRc*mSpiEVO2OW(dqKzaUBFSR z4{Tat3nOCV;J_m;lw6ksA}Vj;f~Noz3fo08sV}_uZ?wPzLpN~c%1yl2aV_@vSAuUT zgyYHi{dIxA|MGNx&4(QiE5OznUqG^gCU0Og3nW`U0O@Bw0{1=L$fyuR zbVJyf>Q)zfG!VNdEx{&UxA59ZPds6nf>)96DAQFxSmmxStd-|(j2B+QhsN&X?1%N( zurwNH%pAmvJT}1Nk_Tbbr*@}C*{Z$COQB4DHxAK9L)=zwTmx$XwycsXu)WN1lIAX!+N4V#II*mti z@J_)yz>M6$=QpKcX7Dk-0{XDeq8!Y1&SAG&XTmEYT;YAC9L!ML3&lgFO4puRfG$ihC%$E zGH|oQ4S&*8=gmmoSQo{PfE>XacKYffY%n?t>ZseoyN{+q!|H3`L#b-rlZyGQ?xW8j z+4d7~jyMSi!7JRez#jvnwJe@DAtG04;9jHaSWQL`Yb>)BX#=ic?~zixSG5rr{d|E} zu)RQY^&N2bz!l&+PaZPmFWFOJt8q|`m{2?|23HL0V59z@0Iku2E_p?8^O0Hjvh7IS z_Xnx0O@J}3zN3SmzZ}G8CiQtr(`y}cMg4lt$z|+H@6%ZJ`(Mkxe zbQ1ajsIDZ$NBj9oV)=T&S{$!FRNt;>4pi@Y>T{ zpg4C9^t`zr@0Pw=Cpr58>!Q%fu2)*iMt%EOXX;phZQot3HPAiEE)JRtH;WmAM~y|b z@b+Qcr`xNun%wevj@u<&tRk1=)g64*)TJ1 z5B_y7NyO{S!cE~)P;r(TYz03->_>layrz_QJw6;nbug@ppp*HUc@MOET!EMGErZ%| zgW%}*yLHp?5;j~KvlkN}ta2y=uC+P9+VmH=`HzApg{R>cc@^GBvPk!*H5KX!)z`hb17U_72I94;*+SzA$RMmZ8f<^PH&n3K_D-Jq{wax=X zL>Bw$T@!Cgxg!kROY`ar)!+)_t#J3na`0as2i~1k1Z&3Pz`pG*K-zpMk= z4ZEK3UQ;GmWAGB|<&w+Za)9_!&LQ^b$~fMuU>j^|a|?JRZGj$V%<&(8OMLXJ3qJa1 zGYrg`0^grM2j&e!*2M9AUBHkV@Qu%f!MP%ybekw|J?b92-q#Fo{=61zyWVFfHecZJ zq>0sc?0)C8DURH8Bev_9Xx&-RD0hc2Ak(q>-dN|Jc>+K1GsRoum*U>T-+1kS<;^52 z*%Mg~4nH+Bz$8h64=@!R&>jR|oRy%A$(q_L`Qmly%yZVFQVhNxmSq0xcE$P|E7=SD z;@Vvi;_SAzSujt02&^U+!qBgMY{TaZ;9i6~oFNs#Gng~%u%a>r)J1M(1Aq6iS^qw< z>_{lSYzpx3X^2Bk48pb2+3>sSO{nTx%WK|v2wW>Bz^Va*p3PBd-1Tk)z9hmq{-?iZ z-|EVU0g>QdwEmCv_ty4Yu9eL~yZ?{+`~O?$|G#}azt>p(zt8_$|8M)S_<;NA9^P0-I{OuV3KD5G6=PcAt{Z*z$_0K(o1Y-wI*)VvRUY^G>nQrJCdt)XvWr$J zN}w~IH4>_l*XSkhr=a&w8p-tNrD%nHDtYX8JU>EbF2Bl9i#t~NnRvG640-&yJUQLu z5j|*nmzLk&!hPzvhxu9Gj=ugybmB)_JAwEZhhCh`{XKI#dif+7_0Rr+RKzM#aIrr# z{(BLV-xomc-kC=pNm)ey>o+Aj`Ibm`yn{|Wcb&7kNXPyV4eiN@Uu5^crPS%2FUTRA z6#Dn2&2-a>_x$1e5@_$DQ{;aYiOAdjja}`7bA*+P3UN9*`Bqa#LRH-6b%K}-%j)wj ze12Q1Mhz!MKwdJHqZq?cGK##y52sCu1amne(y|n(T9ne~%X!F6rIL_Pj^kP=?xIiJ zwyz0MSGG@hKg3}yUyviRTQbm z+jfyLM)CZ&S2K_xE3W#Wd^@LNr?x;``w73uQcG}NEs6+Oz@WbOTFjv_D>7V9X?1d zTGLAob}wSqCw@eUWfkP*Lw)3ft;flbV3tHzWXWB;K4MnsV>)%~Jv5lph3>n(q6d7I zF+;jj7_0j2WZ*GHbmZ1AG<{ktT9tSkX(z{{T!lG;k_Aur2JL3ZfpD(@%XN|b)w3vi z_cW^POb2Nkmr17HXe6B`TgVOPu954Q3FI@ROn%k&L}8A%$ZtO7=T&d0z3He?4WPEUqvZ8Te`NgjIZ|toM|nFH(e0X8)a$86-mzClcb`ks_G&x0 ztIkeG)eAzAz}15CUq_Q)WuwTDE;U+V^&wO-GLuO?FhF0>j6^k_cM&PGgqmBbP34)m zQ8OZTksq%#kS&KBNR$7(nMy?shBs));KM~+h5Ko!gEnH0uOW!Q!^!l%kdJ~M>3`$~ zg#cvKoPicS4CgM~sDzf$HMBx@2>saWITF4#Vm5tYL>bftM7{1eVo8-edh2>uP%L|a z1KV8qt4s`Re`my1m#L&OWkK0Y&x&xAKpT)BL&gb-mS?n0_XsV}>l8?etwfWXG7$B0vUZ6)9beP&B`|FB%^o76c zn47<2m}$K`xovrpCC>Eexl?j zMQZw|p^3nGjKZ7-5=dS`!EZCrKjmT+KJckV(#e9+Y>6QoHY`L_B4(q|1Lox0zG|Y; zaVyFlI7)aw4JGZaloC}&`ebHU2c7E<$(M(`5k}oa{&otvHI5N>>+jPNy$MJtd6y{t z{gklh1(2p~_eq87E5y;_i|BnSpMZS{=-#CeB>zYiiT$jiSC7!#@vgnp5Qs;iO5)6B zSwH%}L3#Sl@pi$7+Utzb$r7f=c{Ssx6+tI?h42HGP4bgQO!EWx_nsltBN6Js95w18DiunJBDDfyq8+h+?wt69&C8f-HY++AB27{>@i$X4c%Z zC`bArH*^0Fw17@UGu}N%$>HkA3^*~1H$^fhqTdnQqT9HUnXzc!P!ebVEH(Q5*qoXs zM+s&J6^g9m3&?}U&S-ebTy!+3n2Sy{F`o~PGL>zQnW{u7bp1gYcibYL+duw`DCmdw znpsEe%{DaJ7fp6llm4x?X&+c-sUVwyF;|$AZwrW){S2Ba|DM}YtHp@v?qh6ntQap* ze$^AFGxS0J8$wNcgI(pOsa%sfeg2ECVI(=WoT;@oVggw?^zPIqRA99msYFH*Cqteg z#xs~0tm86Yo&RGlua9Ps?mtpuP>t$-luhtC8BFufV@zsaB9oBxh`ycsgR>wymV2w- z4(&NF!yF&eVRn7E$T)STG839Bm^RfE!NUB{0+Xpb1?q#Q9E;r&=)Hdkku0N!Zmb&P zx`b<^*t-(~jqaJ;jvtG;Rm~Ss=yyZrdP5~~y=E>Gyto=g4ZIQzNd2f@cv*$c31$Rj z_Ect}(N?rZrWh%-=Tp(=O{fILBjomr9fEIZD`<26dE5AXl1wD?2sQgjphNhfeHqn( zvfiwx@6ld#6EY!P<|!lH55vf2NRfF^9?hK0I6@YtyeE#ytwhhht8$_pFkLkCjt=4$ zlhfZ{;!MiMa;)Sg$oz^yGy0S;VJ<)IX{;9baXw_ zFMW#9e4)fu(pp41AB-VQXD(n~INf3RH+7l3*Ia>|-fVj`CPjy~4|AvIn4sdDf9dG% za)dSm#>66o2|Rj{-cenIyxyhEPe6*PJfVlrd zkCR#@;H)aRXP@hM7hN^FiM|%CKx0m)(2tU{NNZ;S>X%bsq|??i$fl1;EEwi*oaRye zQO}0|x>rN+X^8^8HYb5dTD6HkW}zt%?@;4thZZ93d78|Y$Hy4oPILO8Tmmh#UY|Z6 zphRrS|eVDaike*dmVENv@Y~CU35`{FnMLhvnOm0F)FObCJp;i2EHU7v+LJ!?KorrEfi$+$TE)%(bu99U(<2iLp zo9HzUd9?436ymFNA?Mq-d;DoC9|c)bXNiLy8QjEz0-E=0F&$AIM~ClRhqC3oiKgmt z!M*Wo^ep@W^%RZK9@ZDpqiy&2+|_N>nd2GNRaOh>DhCU6tXY?F+8BsR+GEfR;W1`! ztpYjX^M~#?btYDgwb8qdq#)gR114#(gRWYTMgO^8Ocyweab+at5fw|s=(qNXOybw0 z%x?NJqbIU-w^qN3OqBVf%D)A#<9DK_)2);^F_VF}c&`N(h zY=JKGf6~HyKa}-Xk5F!1f&O#3CXn;<7PP?@fwl2V?sEMJ0YCeWAgA>bF=%m|1D_iU z+ICcN9`$t+yL6?=!%h3>@oAsvA9v}RB;OY9Ia;1r*x<;x246tgKOPHQIi7@PVL0lY zVU31$#^@aVVESg8Ci}!u#A; z8O2<`_)I=uN{4<_?9UCx%hMe1ej$jTymJ6G7Yw7PeO}B4tuf@Frh!(UTS2bYn1;TOuRu?ZTM_%VLKJeTgn9Y0g8BRX zDU(0G482m%pl=rS5^k5Tqn6T6wC_tGQs~%4CU{(C5|TriPTd~V6DosV$tR*U;<8lB z^_{eKqAsI|Hu`#$QEiGgv%wdqG_kAr%I&B;PEwyu+ArTmcojcas})N#`EZ6>`j)tr7dSM+}s zE+g;nnnP@My+D}jo~7&l?Bjm(I!AUo^pdV#1f`T*z`RM?&Fsn?WRwyl=;@xy+_t6W z^sk>yrxL4&HbmwJbpjvJxG z2jy4f*6C(smcljUJG};F^%T(`d(-Iy;YqY{S1#Y^Zx*v5Ih^@Dw3%6V#R{DdT*c7D zB?eBhWkLre7=__-6qTIBXs^y<>P$W|ayCmS9aT4KsuM*$UH+ZK|3*pa-Zb(*snbX{ zz_8CBHVJ`m2edS(Q|zQ6iO?@{OFavzqjcUP@oHxIvijTglnac_(ms zAj5a9N}Rr^jfD|Q~EogRsj(@m&$dl{zB zlaHpvnWIKmB~<$T2zr0mj1K>`7u`-6uXWf+f%L ziRuMaqB?sSX%k$*Pk+|TCKdONhtk!Sw5Nh%bnilxS5PA+C|iETZ^uAA0(pEo$22TWov%kt0uSFTqc9} z$|2!xDXO(QPhewEMFvJDqr!;=%$5Vsn2;)C=1;By<1g>R5Zh7-uAM6RYW{U1C-yY) z;u&OA?HKx?>r~Dp$VFS*mecpmT_|6(U!=RlOLCd;JT)t}nOs=6lzJy2&-nklNCv&# zL00XXAT>;TslQgD% zs&s5B?N?ktZwQehXP*M3LCZSwO3f+cy3>-L^Wg#^zGg9TMDjNokPu6nepaR8HA9#+ zQAtSeehzJVERR^Er9?@G6p;Z6q2xsIHIl2^Mn8HNTWvn&kM-dv9FEvZNvhT0q^M>+ zNrf{50!^P&gx3lMq+gMgxJ_9GV605^4yt0%UH*ucDse-8=)S> z`!SQwZEhrv{KzJC7pYMCE2F7T?`^2pH>JuN zrciBxfJ*Fb}Y1hazf(3+DGsLFMet2erlbl;*? z{kv=y@w?+JH{WRiU8>@Q&PIJC!?Wv&vP}sn%3}qY9wg8GG&7#N$#55OI-Vp4{t5_w ze*_AHPK=_-6=rDMhPhG@K%U)@N*T1pO_uoOyaqGBmv`Y9le?2E6Ru~h(9}NUb zZ4&~&heCAg#W0GMZ$;OcN;=MT1@di8CN;PHMf>4Yw7lmgx#Q6s>Qd=53hcW@4No~q z`}IcBWjL}XF8MmC6KqS!^p_CQs~hRL`8@*ndy(AgWjtmi(v!Kh+?uXA=|-76UrMQp zJ5m*Qr&E`J0aYUtK*?!WllOmVQcIT#$&;{`b1!4R{f|L?JNu~Bv>?<{@HN(h7OvPJ zSolTCzPKyezG;RD$H7wpsm&Ro-G)}9ETc9g`&I?@J)1-B4VI%f2b>{Sq?QQMtFq~p zfmQs2e)ou@d-`eZbUQ)w2Q$)CO2~csG?ZJ8Mg^2t2$I`xJ**JfTw z27SGyhx9fpM=y5&M7sSlWTd4YnZ3J#uJTGFziWP=p1$Au^T;qbq=@l?P7B0<2-Wg&=5J0G_B%9;@8`^+GGX0 z^`*%R8@pZ|@-2yV@Yb0sY zxRQPx9mVH+8!>~;B$MQRgc)2kjhW`6PMvzWjGPvyOj)O#Lm}^%FyiwC=*O(IYO|Yr zt55asu(@xyhB^~sM!JUZ>Ap@q`Zc|oDE={k8Y~M09|8~4Iys)C-P5@0<=6Xf^kjAL zqrPMi-Jf(h55m@?qL15c&yL+ITh~98)AMa+O~|T&YQs2nLAlTnm2E5$SRZ@GajPq$ zUBa{MORS#?=<^HcM<*-@{rSZLS@U>OIZ2M{6u*cR7XLDaf|N|op}}f8j+gnXA;Q#$>B&2DV1o zc17B8R!H8dw(xmlySq=`N9qiGq?zQUWk9ulxyXw>gyA&^QO3mW9uh(W< zyXu{^wd?y~e`@wcJ|(hguDub#H&D=JWn0tgG`ki0Nw#-5SM0WUso4E{l*IY*`vGTVu1oa}W#anv!}6S@t81!@ zN8eVR*mKzSSfkJFJ&A*rs4dQ3PwrTa8r~_`x=WMxStZG@y2@1E-P&RAlD>=Y?fCq* zM|$+_RU17VeA0p(mM*e&cwbY8K3}2ee@b_V*@+jZ_OpdlKEHwba#w+yQ5aOCz3mjC zhnLVjD+}q=;>R?cw*VFP7gev@K8JLwQsZu#EGOz;w-Dc#OCg&GcN7pcF4%l{koxB^ zK^>^ON)2h()5Vwgv@=*sSR3%kS44E<*A2% z^z0J~BI#A}=Jep1V!`(OJW~11dvYMZ233#}OzP5&NQ2qskbK$GK`i!|!!EHP^x9UO z;TNAorEQ05^t7ez9!nhHKB$~t+b3B2{$2ogejzs%XIQ-o+7%*Gb5Mo8ljUfD)N~>9`=$m=97odT%jiTvDBT8 zUg~W?6J@rujJj|CfQm@oOJzDGQd{^ER8amQ%ETdss?@nlrHpuxdNxl&p%#|HPR=|AjvT{GZd&x;kQQ%MQh@`3h6?3U*8&P4d1IgBT;_?7d7|GF%Mduyv6aw&15BXV}oS0hWvLxJ_=i|7_Is^5j=rAJW< z(n2XJ-Ji}GwnQGd% z&I>BY8@1|m;Nu(Q0S83Lrm7+T83V+RVkvU*?_=c3jidDLT6y&2&JD6bWe@YQC5EZe z@n-hjBMECG2SM7YY-0T*H`M1PPkvZ2gVZxVNC$+Z(1goU!TN=V2^F!s#8okK?v(G@ zq;~!^D%Q@DdKbI`wa`jvsHOt(q+by(^D4Re>ub5DH(rvF&MMT;`G3gZ=U>t7S`9{6 zT0;Mgw_#q#*`paDmq~lG#f00bRe~JnM53_SimBp!K`t_y%pXrB@# zDcpaF-c^1Jee>RjLIIb^m?}?Jw3%(E=$4#ikf$GFG%~y1$sy2GBdPT6PU6=Uu zI=&|Thz%qC%7ZBi@n&wnR~3l$ry>aWB` z5}#N?)#`7h%vZ%TzwRqDvzAq(t`#l(f*nCb#RYL9_|P|^%KRr9Tl0&yTYiqwuFYb6 zx8*W%vG(Mb{t~ip>=apX=p-^Xn~o~?m2)TJ0JC@t$=Gg@XZ{J|=)X_Zi15R?+?79A zG-}m|6uzw|h{jIZ%Txh{Ua}Mv;Cew+Lkh9Od^z{_lOR-X0S>D||((Uy>H^g{uU+*xQyU-`c%x(dC|U89s5E}h??1S&^SZBd zKJWQa)Bx*$Yf=U4{HYsWl2*NHx_3LhpRlJk$FqMH%*Q)Byp?0V_P+3DN)Lhq9)Z;}5 zs68d^)Ww86!U)Nrb}=>d`jt1naW@frZ9a_;S{@+9)yHYup=Qw0F^`QorN&y#7FPWE8~7WkW4w zu;9Kh?Jz+bPlO`h@w?!JiaK9%-w$H_cO}3T(Ei-x!$^=3$LTs9mV@=dmP@STd0Pgr zlQ;YHiAEtrdc#G;FN;+SdM-vR{*Xibdp*T08ieSSdjVZIAV$XOaY&a?C6Z&+MVM*7 zBfry9q*(eU(5N?qJpVqQ5LNI)^BTiY^&8B6=4e4|ELq2#m`LEajNW2`NrEh|^Whzp z;+sqUen*F04WXYV+w=beCfw;)wdB)`V9;rMm!#bnptz;sxWC*6kCxeTb2Inzj`nA8 znx&nP^xv6i_RJ?_QJyp3aKxTFq{?#MR)`}7^;qO9d5)=_Q*^jV zCZl&;5#@YHAi|cqfFlHlln%YkOUnx~>)GGMsmUJZOVOWsbNd~cx_JqNV|X7OeJGXr zbb+qwe_~c`^Jpo#FQtpLwp8b~yfZ^<_MK-+&-`M#uM{xaYSY}dJ4=C8eF8c!*H;}F zuffg9*vGhSI7J`7XiCJ^FwDQg?R*N%Lg;UBgT8DW;p)prS` zK$Rr#swM*qp$W43nKJK0h&}z=UW_<+&y0Tgs)&hJXhWAet;tWLVyLn$3^gv+BJZD` zB#v)NVg8sGkm+Ubh|;(+6t4V_%dzO?ep^uv%KaMnYb#2~Kc{oZdABpcSoj<;Z{jth z7Wto1ZocoTKNn-;|vd8>2_so)2S@k@XO8C^LP@)I+-DcL@vf$ZG;5jQOK6|J_(sQHK1uyo0z~6GR422N5I7 z4{~2Rz2JP?)k1|Huco5;Rd6_>2bF!;NIsZ0#yXS{wdPq4CG1;8y_^+Jh+JPpN7lmIh8dA%D zAE)lg-Gc#t=2AjBv*7}wRlp~o!d8O!cD<)DYuo8bW!%c3RvfgTXvGU`wu%93VxPp0 z8s*T3o_R2@QbicoL*~ff(jjKCv@^b+Z_9EoS+S4SFJy*AEO)nP`dmF0dw0 z2w1Io!^9Vb0=@K?aOdq&$Xh8**}YI;|HimO-tb4<_PGnb?q7;m=@iq{@ill-G9U6^ z`(UMBJIbn38fwm3jx1;acN8s28NWzC_Ywub?Vu{Tg1&~X4)wzg0iDcqmA7z+Ge&~! z6QKXWeCB{dFN~K920{1$2>+1>BYBtM_v>%Li`?C>#`x~vroOL&3MZL)>`={Ai@S6Pua4mZI=-EVP^ z=~pzgHi1-@pNmyDO5)aK6?n$gOq}nZkKJZuF-Da!U?5@$YnHC37bL3CbBdK2%h%HE z)0=ZxwLQ}8?cX#?-t?0ExAimlI`I!|9E}AcdRh=Q4dKs8?u@Wo8}x})XLR~+5+a=` zAgMzc8a#LlM0f9n)-iwaf|Re|XiXz*pOC`PvlkzKI0ydxScq4atN~kupW?^wr19>o zkEqe#5{l1nfH$ugQGup+!0z%!Jh5a1S8|uL@1NzBa^37+wEuoIxpb1tNvbWWVqAx4otjkCY^aJyLPOD;x9Wmj)&BUo*H5s1 z+7H|e)n^{fNWoj)7ed=gNh-@J6IV|>;;vcc&QI>HCLUdP!1bD!aD}=sd)&eaE|2qv zAEKTEoykkRtAvZ^5X=~p0JH+CGBhC9I>;$N}1^wS93upRSrP#ZQmG=t@dzhHOPH8}f=4s>SEBZoIt_-bG!D%pAqc1tJWx`&x? zs>BoJ8B}BSyu;|!@<=E<+>94@EywR7Q(>^+J}lf&52O1Iqvxg4DA2qQjS|V^py)7} zUHpJ>wS5ijE`_80P03(<=CPTteQBFjdbh8 z=Q@PgO)EXwp8*ZHNZ}`V{`Mt0I5~u6HZB%;kCfQEyMsVr!w&Gd;Ra*Ws|F@}!hn;I z9k}7UiOo8-l`ZbO#to>s&AzZ@})xgLvg73D!g@2MzaJhDDX@!1-mRU{hKI&kGx zK)*7yBewKzK+N(itZbmjE({iD(aCk>YSDP$ymAaU{yfV19!kIinnLiyzX$<;B!ovV zSKxCu*8=A~6SN;UlSsbymniXA#J6q8rF%->u)}uw?B$^pdMxF{&!eHeW4 zM-bLr1MT;I;nh4?}GFfRhjn%>@eNV&c2t{NP_!zdF&t^RR z?TDQBlgLal4{R7#VJ$r$Vs$@Pw8l>!J{j~wac<{9m)!-RQ7J-6&M0E~LnD}1q#6#+ zEW-bd{D;WwNZk5sBW+=(hVolqq065PP+b8RJ~8bE_9wD|H~kX+uKxxP*7Ly%K?Pr^ z;S~D4c@XlX_fWeZ-l6mw459IpOHd2w0Q2%|7=Aa$Vbm^ced-0S?y`b*9w9KS+>m`S z*?><^S!0*W*6hUb7ueJ(9PUvuVs$&r*=e2y8@$ej{Vewl-xV*!Pb}q`?Zs!Ar=$Ov zi2roqwF`?tzSuDW3SVOWliY~(mMcP@bp&j`(+#5h6wqW{D(lY{u(a=XNSwA~Tbz{H zgthPSxj%=Y>&H|QKGXxY7T54uy_;mtk!SGyDl?MYnMI!GUd5D`3vM{}4#*i6!c~1u zRPS%WZf`-O;)}pO>(zs>Qly5jWEwE6o#&(;PD3)o`Y@D)sk!2i46jqptcYQ}=ffxN~I_&dQboy0_x^4~4IR zS-)br&rcxg?T%4sFD*jNRSbi_LykeQ&@M9SgpJkyu>Yu`4?%tE95rsQ&VHTgg9F9$ z(Vo8V`0Mxo@YoS=tb8vOCd5dSE@%CyTizj5o|qd|eshYtB_2WL3=8z`yKCrN0j`@| z_?)BSa*CQK&_rhMkip$fhoNp?CGtJ4kMCzAG^o1*P3g5VbEj)y&#K4p*Mn^^&_kaZ zk8}j;><+kRg*;R?%p*NcrGo!SWnGY)3rCP8Q^cTjVLAHm3Z7vSz+ zHiSZ<6X7!8N!T2=K(&qo+}7Wgt0zt-*L3}6?$#OpkcF0V&GRK#lZp|I7& zzMa&CP$e*ZT>;P`Z-{4e3#l*VvQ(G|jgAg)0pO`D^G?eN`_A4$+Q5sr-AM#5+awP^ z@yeNaZVw>#jx#5S9QgBUD?oa3RBfvgrS0iO8Fb3w6-B*_Ao>etJl5l=`p+is#YCW^ zW@7|Oy+h1?6#~=O?&Zd!FveBh7@W=8$5(kV1-j8~=kVKK>Xluj_N^yZ3Z}kn0U(kfrIx=Zs2FzYrOzgYr z42+;He6Sf<`dv9zm^a0s^EqD zl(E^~Ylu>K%VL$%6dsHJl+&T{aJzFiAV=2_qWP(8txi;6mUI_+U>U`p~x!wLd)q9v{4oR~(Q+`8|G6 zi#Q8ia$=!v$_luwyum_l(-&g7Y%p}6enuQ<5CvFT9)%7!Lmk~hxJ*nIh~2wx@r+nc z_@1>wJ4KR+fMwD^^MxK>?-NC6b^Rt?gspJtNIJpVE~Y;&isS2vX@J;>Iq>~=6R0^a z1Z?!>07K1Kxb>bHn19)gvAtf&$kzS?F|Bgoc80j%`=oi>kNH(6B*#_RQ1#V%VmoLT z?Q;CQ<`gsNbqN2heuhMp7_#J<1pZOI2aX;JB^1}&l7Z(pp)2cHU}I2$t~Rbhj^iw; zda@0S<3_?MHxsyBxB~t(--8qUYm86fF!}kxA#|FyLrKF@h%f5~bsdG7ZGL5zd!=-l z9F64r9k%Q?Rl|nwhhF6YrF-Jm+6lD*0{O z4EX7z6a4Kr%3q*^36p#=dj8v76n9|+pS;XRg5oGfVn`Wz_kAQjk5qEkXrJTz?w!xw zM%^KtB|ngfT}HTCu9xxA?6I_5d)4fw@Jh5!Y@9IN+iQ7wUI%%sc{@R;iGydShAjU$ zDAGs#zth*&wvfisWq>E04qRty&`NVmxNC3vG5p8w%LzlvOzEiGh4x=T?%0srmB3_ z?4C-e7-#JB*@YjLHiPduYQc{|`pl-GRb-WiHCeLp1`kU5pyT8E#OsSWeD^!SfPEr? z9i&sp^&djX+VN%R#YQ7A-q*rxubu-d9_HeN>iQ82OZ}o8d-0DOy5a zJ8ey-4>j>6-52wvunP(sH6zQUkC8j5DVV3#3L3Rn0=0Y@T&}4Mp2jP{ub*Tg(*6eS zha_@p8cqX#)*$3Mq=Ck>zvzx1iEyW>5B%fu4V*q|$_x0B%JrEy7YsOCGY2fCI46Yu z1BVQ*f@vudXbqQ8RyLFH>Dl=d>N-z7+80Z?_k>adcbJxM=FQ%0* ze*!NTOESO6axgE;hk1pg%BM+XLdVOAs4tj9ap(So`$~1FdPgOkVK)Rgt9L^OjxyAL zF$i1iR4JjZOe!ktDJ4<)no62sgVb}@5{G96fi%-7N~icXwfOKw>Oi^|DKeACkIm+Q z)h8UG&ukq^g>{5^dn{q{U?gyy9D{F8loD6e*3qNmrOe74S?05bBVW=_io_KSq(op2 z>h@B?JFG3t)tJTuU8Vh@9PqO8R4AXx78C|3^ zj_SL9fUB`P@%S+n>d~2ZP`p%(T7D=N@{kU-sF$MRpTto&1H&nI?RY9pCl_QD@?l|` z0p-^B6jm>~g7#gy07BwA;gm!KG?9G|HDc$&S|Wo{ALGKj!{5lBiH~5nsSpey#+cph z1;p4KCGxq0IQ{ZzCi6jmDY^56I}$%SO>Ybif<6&m)Qelbl-#{-l;od6O3C#qb!+Ka zYBgVjS@JE6ynEvav(m_uN)B?Mu1Yyj@AsR6!k5Ru($XmMyKWOvdu@^&Gx*7T@4AF< znLD#l=hv}*PCvoRXdSRIMFIPJr{cyMdv?aWTClT{z?%(DptBdBF`JW%h;|r6*yI9w zZDBOSFZCgn7bn5Sv1}3rH=!NJ$V-#&`7{&^#!}CoptOJ>dq{sO8!2jmJLo`ZC6qDW&QnP0 zpD#W0vM5#0(ZZDGMbOzP)SU}YJ_ zZi*Ygch-vF4A%-)@kRnmFMos;1{cFDSvSyHf0YcpeSvirX=HuY^s#>x*1>nC!fcCx z8=;p*q9EQG__Z*C+0?7be(P{yKWs~aF{>PK>S-ai88-;H^Eddo%_EfPRE#VBMgyr? z4v;_f667u!M$Mm1;cOHQBOXuVUbCMlUg{s--QxS=Re^G?r9qi=<4p!!=MCx1j56-)P$kA{S0mcyrUS}D?~`3G zt&yMKZIrb15BaZj5uv#82AM~efLQ5bsQ!2YIt*3APpfAzk&on%Y}G~5_R@YPOy(Q? z@Ao5~)!+HhMp1?m6l=h=hFYMdPlL0q=2-DvC^hx!6^tumVQH%cXr8GBN+LSIi!xDa z*^Y3kW}hmh)n`cU_d7R{aN;bF-txxm!N{15<${ zN+0?|yq-OSE{c>U=iaR4|19T~?^Ad}70zv?cHRu2&i*Z9vUKB_`@iI{!>e*?*$X#H z+%_L3RqRC9C5OO@)?3VQjE*JIB1d0-n{a2@mCsc42&RHu-%)L`)lhM0JzTkV4ZLcw zk}`$X)WlH(sv=E;IvG2M@5$AbupFJX+vJI|AUOv z6KMU14bF* zW@N*UKXj=`H$}oFM3?+r`hrfoB4t42>W@27Cpy`PIOn+l|z`%0IB9D+W^13#gG42T05OqNd1Z ziXUD{tx%o|Rw$87O3?`T@BC}vP_qY$`&wZiMK!YEl{RX1P($ZGuBF_2|3IA^AK~+P zQ!wCKFy1Zl8KgGqp&GkMMkeGRLckqY$_adyYg}RDCSMR%bc|f3GM`L!i2$*NUy!MZ zJ<9rzLR%6@Cb`ic1=q|64jq-m#?x8Mk8m4klrRP^e^bY2?yZLwyZX@hm;Z<#A52ka zTt4FxxQ?Xc_1Lq^GBBh092_v)hleXQ@Y2p2;C=WKYWdp&U4Ew^eLHhbK-)0AXzUzP z%Q0iUC$sSLh4Sp{w*zQzlP`QEj-lH`GL@PvSnZ^n1rM|!WMN@S&RF;zRT~+h0E2Ai z-mMknpLeoUdT$lZc)A2Wk{v_3Rpn^tKqILdrH9{@f5(RI#_Wc)ShSx0P0k#C48qiz zDoJ?}CUK>})TQZRNyyybS&6(e9}5b?T6&>yo?<@K;ng4^&O%%@VF~_BI>VmPnQ+?? zXK>Y96i#)#0lq(m(Pj5G#yvZouG;^Vbht0hp1P~ThMyK?y-K~QE$Inxyy-c-lW`J1 z*!YBbp{oLCDLLS~*);woTZwr+VU$RSJ!~?V3saujv)(Napw(+j_$#-VX;c*9S=b0_ z*M(NIhP|`cc`4~QVf-6B***eieI==~^!w<%Wj*@)yqy@&Tm*%#gn;}Q0H1s1;5m!@ zaD2clR{N+3nqg{zK3DG|Zkgu-oClz6{VlX}YaC-1)JOVk(q=?LCz!xlTN&LM^XZ2_ zb8+;(v*6;Taq`8Z8i8+IlFI-33arVAMz{ZR@pynaju#}|KD;)=#PbV4Ycd<|&6Gt& zft^6q%^cRW^x(tdD(t5aXSgLqjJ;{u3^iN!!{@t2@XT^~s?xp*xY}8vxW6LM<=7s^ z+EkkK>2@LbA!Qh!yao3>ZGsJ}c7a*m2k-*r6?nPgTd+1zliZoqL+n_!5B*sz%sMvd z2z*1g@nHTrd_#39HZ{v;eokk@hfxD??vVq`N$)_EPVsfuU%ltoenb!RuG9Ou{4H&2YwKj{NxOUPT$T{%F8pWIDK zi&%jpWtqekq5VjAdl(vOGoBFm04m) zGl_$9#s0#>zL`)`LIDm(dO)jF4^c5&NQdq6rPQnqsjp#YpzHn|FhEJef+Zm!@W)qV z*Ab0H{STvn2V&HPmsv2ypdT7#Z-p{jW>bW|7GC+4`x!vQfH~6X9Ym}<#xV=ZVj!}mqt2g&LbwQ zAJ8?+fANpU$P=n|i9ENH+Vpa>t<;#L8}+s(kMj6(3^^XIL!`Ygn1E?8$N4KHk~QJ^ z6&qm-Du53`CQxXuK>&19XOZR!yGdPppt$K|lO2df=?F_uZ*9Sk* zuSEZT%aWfPUz4>nGr3XuOTnHffiC^n8H%>W!m=qX@TG4b(!bwCipVUa)fBvtr(Gu~ zHT+KaImct$<8K9->`MGQ>I7_c)nt?W9N4q0JiGbj1Dq0RkBx_P*vM2t1}xM8*H#T- zZ1^63ds~Ai{>V}{{$#*(%O#ZmvH3{L`V+iVd;<=-#lW07$zb_f6JW8M12+Gd%~v)_ z<;%N@@}^C;<80n8f+Nt%66hTGO=uEIdC#J(K7M1KTVGGoamlMhF-}bd#x6eS4$nS$hLWy(2hg4apR1HATJtp zeg^)MYKB9OT*CUdcK{8gUV7_MNqkK<6Z_R!5N0o0xMPi%>8mzl=;6O|gyLd9ejLZ(CNPT>?ttKm82IYZFEViQ65RjAi&JBq#sNnVUuPf|!-cK32F$T~?!zN)$3}CiKXwxP-_9hh>hBXH zYjepQrz70k?mFDd2ARaVQwpr|^a2cv{(~>bH1KvT0)^^`!oK-SfN+f-*nRjqUiaSy zaQkZ|%4qrwcHa5O;~t6TB<1hqt6MEc@{BUFl|F+9&zM7Jy(tiHBbe2Xenmf%;!*Ho zS8T6%3ojIE!xz;HusgAg9NZueMdldbvumZOSs~HX)g9i{QeiI;P+m^f?VxDaxR1=K zrd?oWnK}F$BaOcvJBS+7>QL|-TkKf8A4b?1K#yQEc+rXGob<7WMRx{)*W3;;WZw+V zAp=gzI$T{mv6*Y?Ek_TCb0L3oI82Q-hN%Z~p~;^yIQr@aSbg1xyDrcVT`zEg>l%|` z-HtWX?H@IeIcG`rXqiyN&o(4MNCDA15#%}XC6t@TWvc%}9L0_4#X;AMuzk!Pcxx3v zM9c$5>E%}>Fup)5xR5&XA_wY>#DXXPslvTF^6>s89n^5Bf;?%_M|0l_u<+A!D5J;} zbiMz{Yxul`T2y!*z*HI2`cQl4VCnjP7k`$4p%)}mI98& z7J*N}-l$n+iY&Vzz=Ln%VOOLz`$lIyTTtAJolzn;Es@Ldzn{c^_^XLb)OH6;Lrx>x z4GEl}DT+V3>k55*QxPuBH^cuZVW_xWo=Mz%1&riHBJ)elxIan<=kpihUANUi!018p z${$JWmJOKGqs|;h#VgeYc3&7>UxdOYWvN3oEHV*1>*O#`|puR@`_e=~5=J1zG0*+pzjsSBIF-h=HYzL5)q>w(ZO3cXo=1O7V^1@?RW zAcag$q6CM0bYf-)vBhK!_Au8%_w&2K`=7VKi-vxp@^B$}Vf`33`*Iv8c5i|rgW3F_ zzPxHX9|Mb;tFdTWLKvT`_>Gn;sbQX)2U0Bdf^{7mL0F?GR1Atlxjb|1v{>-Jt=$f9 z#N@*#`Pf={x*LXciWcxW`ZVYegZoGt~6&N$5g};R~MK=-(MvD7tGF%q;(h zm#>y$y)!CU*RZE-Vn;YzuO>|G68i(!+%m?7K~cDF)<(e@m~dgkPcZwRJ5U)l;fApX zp>L-U>|MDRHni(QqZ>j*!a*m-N%tahJ9Q0S($~R1?)EZT=ldC%s(Nfoo8Se|9~-Rm zLJRgOv%0~NV0Kmt&iwp`eo()kGdEiZ9^M)S{w;}st@n#y(T)#w*8^2gqO9*`+y@Vaj;&ocP|ItU%mzEgCd#3%0e@s4` z9^_3Mi@~&xaJVTY1vjm(z+^xsUKd{kJ%(kV)YfR|6u;GckfXpa-1CFTU7=&ft=(ek z?(o$zY4ZWH<3c0%n5P`4^OrMt*#8|kT27HdnJ>7hZPLJKDTIy>3iwq9_H>6)6knyt z7F^v!lN)5ai2vulUQO&*Km4CM2=6+}X^DG497X8LXMG&Qer2d^ zFcVm(Eh@i(E6JhPyNIoF_SJ26uH>&Z^|UVc8dt$Jj#y_i4|Z--B0YSsRm+AP;Y6E$ zWfZk;pmx3jv#{kgAKp9zzVniRQ_L6s;-6CdzC4y6Yo!DArBewHAz@g)J(7NM*_pFK zq?>u(>`j_ldXojd@nlBhBG8pwP4L}p$iT;U04vyXW#yy>9nuuRpS*auC8r0?xL;fe z^b?E^Htshunqge!TqoYEyQl3&3~IeCqD2EZnss9w#&`WlQWr;KiN@l4rFH87W;M_nq3sd|IGM zHobTWK5>N670H>6XVag5I}9S}|nx3SInE;Fm8>_>3FWqoCGMJlxyig)(F1 z!ROpi)cEo`n{*+ZJ-wC%HTxgK{Zu%vkc~u-zBK~j??ROS=3h|naV{g+tBuG1?PrYU zuBVD$M^TWooqF$>%&N84u!_#E?0B;@<+6*SstZbhXIm@&wqJ)meK#0+w3LENgclic zQx4es&7cg9>_BDj@38J-Gx5!Q5(^gu0f{U-)+c!xYn<%Pu9%U57*kjHn9aa9aV=W- zCXpE#4Iw3myok`!)qHPzN6W((m#|L+9Srv8*R$;w*}(rr1bW*03fz84BlU<>c&PCS z_^309XZYmdd7dG_MVSY*&+P%SwIU$q&35D(DB!lwE@uZ%p2BzEXkxa`o_#jD8tz^v zkNu-ovD=qzrpEKN;irpylr7$cmHc$sklFTZ=iwW$On(JDaOnn?FK9&e^i?o9sLouF zs(?%8zXT_|_1TB%v#9u)+SJB{uc2|=PVD(%0L}G}#%cGRV5h*>S-1NXbQ*|*UCK^y zz_AHx8OJiZA1p}aFK>}pbR0bAdJt2F^YE%hCp=5j9BZoA5Hd2#{5e{`$%qFfpoKRV zKb8u`zipy$@)cijapzrFvnw9{l&wHouP>n!cdwu)YfNCtw#VStcqzOdu#^2-W{ZOx z3h`O%5wy!thD!MpOcB3L*@Gh~EN=$QiZ9ZEMVBAKkj;XOi~A4w(9$3DtDQxMl2qB| z-~)L5nvGcRq##dwD;wT=mIJ~3&FEpGz#}tq9j;yb08Hgkc+LbL4m8YQJEolR)&42= zj%^G$yRnU_Es#gvszI!VQ498bJdfS7(x=5#$%(y}_>Y&DGjHzc zL96bMjP}KCc;4?sV185ry32PV@5%s>zi1Mk$!~#@3pN8XZ7Yb}aEY^UI)o7tcIE86 zXpb~(bAi;Cw;-lU9G|r_WfoQ(0V>0>`1s@bXjSoI6xf>t%6jd|0ws5J`@hx5=G96# zR|aWqKh8ten{BxYx0~pVTjzph zvbN~bn>11;>kbiq{W^Wr{2G0(QVW_F&VVsXNA#>@g=2hd=SUkQ(6l%_5CHKS%3aM4&~616eiK87(zT6A%aQNZ+dM;Ih0Y zM&eaaY<^ovDX0`>7KbvN=`1s8+e;mwS`GaPKkhJGcYCSXoKGe=rAEsS0x0+NS8}J$c%Frz`EKl1Smp_`=4I94!06f!vg=jADF-PvbcMZaQt zu7xTreEyR!Gq4ZT5!dOBWmD)x=NwerEsMFUG+?%!E&DVK@?W0y=N8sRk~2(Zp^2p` zXw#P<)Olz>ts`h}Jg_u_z8aZ@&o?cGEHng*ei)!I`;BlQ$qf(5JR`bk5pZRdA#ERd z9sD=p39?Gsh{bhJd6$Gt=vhA%nBt3$m^fpJ`_DDe`WiRD<)^`5xIPb@%9cbCn-l4& zEv57V^G(P|x(sCwe8kDu2<%kfM@HHY@!r)RffM3Z$YXUX__N_LU@T>c)qhrk@c(wB zpi&?7X|*=i*AGJ-m&H)&j?YZP9Gd^FF9I2U55*trSY}+r9~`-`2~1?aMCbn7f$pB! zkI&AOBiG}fT$Z~NG^Ti|s>ugFu`DqaSogs@YLJtvM!j??hhXwFpY6jX$ zdXkFFI6C;t5itK6(P_O>Xkwhk=(=u1NdaYOmxmqvpzQ&>x7|i^G7#10%K^<|2F$(Y z!$cohP1pG{)yIfw_h$py5CO(mSn=ZRC?^=SR83^983tv5OE>bomkUhD*VXN?(!wt#gdT zZCwzYqKWM5ML}w>C0=%JCSET=g720!AT22a_04^QH1z+HUlV>aDdw`MY9RqH_Zz_4 zciYfld@W-vEDSFPg;2OfkK#@!QFk|laDOv#r4<~5{`xOq zcZnwT{J-T?n71C)>{W=j_l%--MjCKWl@-3zI~#E%HJSTyi-^n9LE!g7PozC_4W;-< zh)Ql*%A7f`g!#GG@XGoE{AQU6^?J`9YOp_s(rhRPC6#-S;_fa~YkCEpnIA{!>|;T! z_!Kuf)PQeWd!0WLR!4b^UZX-{BB-DJeo(l!9(9jOqZjE8=*!Uza8e=-JPR_wrJsbs zr=zw+?vq-^r>C2we`KJ5Z;jNI`8m`}eGh6!{X%HG^b*|rQJ>fnITvb*)c`hgG3HCn zXB2d8!L)fcOgp@vDORW@7dH=)><%S7-!2q@{5trKvBUSzD1xCc0&mfsX1ICr783lO z3v=a3YKMv~byoKUJTx%CSf8;2++lTS2NPjtpCS2;k-<^Bl}2=LP) zk`&9>SeODSh(k_1+j z^aKX-J?O^HJ8+e*2{`clD|kZ1A=S_}dYjD$eoTuPrL-*{cIb@2;n5$+#loJMUVn-4 z=_JwG8>zyI2es_>q@dBbF!nQy$a29M-vV^g7UIE?rN^mG^H(0af z3d2XLsPDjIG`N#OsohUN`BqC19vRJ&Kks54UuRZ^=|ocrZA`}VGtBu5i`d@EG^nFa zle6+yp!0sW7;8yu9QwWhehHmxEgG<>)<2}AUP;az z{Ro66>agkUXKcrvLDuq-)Ls8pHz=Dj2e0xq#$*0H;G{|~^Wfhe=FvH4_WGSmtV8`7 z_SK>rsQLhIcJN$`6L$E}aR!EfG! z=;!uOeq@SqA?|Us+=AJ3cX4);tp+P2BE}wjwT>;wKgVj{(Pw=yWUKT;@yd=rcy!iQ z_Dp6DUX~|=RxKD1ylKiXZ2n)o^X)iudvGqBwRthQs^~5=$5e!Q{OJnnw%Ev??z&6w z8GXg9Y8^mZ{1w1N{nx#lI_^cQ-fVSaSuB#OT}g0bwPg${^=gy~IjqkE7kaiS%akU@-IHU*_0AFYiW52E#Sk{B#_~X}#rvC)OdR(Xe6hL&u4LykHXZhJa(yqpyM}@ zM1~Q*KpCPs2~~V#w3i8U?_toGmLL@yi_1NZR8^M5aJJ1nX)Y(DO&U0jfjNR)o77T$ zQg^zLcCP9tY^QnzzL6;OS!)oDI0!V+D{Ba|O%OeJS}-|rWp-+b!le99xq3-b)m7*B*4?4z@_6r$SN4{!9QT)})TN z^=$;BdgLZ?@?0|KL*`aGUCEe~w?0kgwQlAe6WzHeQc|EZ&TOE2xt-i6 z{;!Cddo4t0XEmcG?1%QPPX_@x)xd_4Aie8^TvoQFR}vO4>6$ieLf-a{`PAC<-bnK_v@FR8X=3P(+xS-n+X;5W#?;fM5WWq@scf zV!(9fz4&gvI(4evQ~#>dRejM_UA1BB*~_)D{$VDjB7jMGz$?XQb~PH)e_f=yHKlCdXRvnLw!e{~;;+c*c^$lH#p zifTJi0k^2r>m#vb<~QnK5r*>OWO3^w5xBQj1|~ICgGwz)Lc1#V@NJKI!q;ip7+R&q zf47EpNZ$-NSri7Y$km{lA>r5~IbGaUP6{{b%EWA&{n33bj#%;DTJs&kR2(hwp`>Z33!t!nzN)lEkqx}E zQ6GKjsveQtS(K9$tL?8kOs8QULXo$Iv`rJw=q(~MYj zzH21+IOhRp@NfXq{ca^4LZs0@fci0^Nm+zwQq4E?__w77d1sbvzb*2pSU!O9OH<)f9}jpW_5(8TCyc95_?zoJe-LWj zZWNLA+n~f9X|&L)1T%fvfoV-T&`aKl*dTq!;Pis;h0rE#*0-}uFhORn2S zy?-U31K<&QdC+1y+4l$Ns(uHYn7Q!%u0r_xS{}7(+7pb?kE4v9{G(R-41m?UAJ`u} za>1_Ftx90&H|OTOT?03z8%%Fa5SkEbrDNL52hdj&UqD=1K9wvsohlc*i?R<+qaZ^Q zwXtjg7HBnhW%nZF^|$MSuF_I&y?-4QXO&Cqh6T~B9*wZt*M<(deGjg9rX)yFJp>1i zCcWinQ_?w8x7C zxJ)A+4xA_xSPR5~&(}%8>)55}-Q$^pc`Fh@e47A?Pq9O~3P-83ZI|kPn^X&!EuNIk zN)xcB_W^7UDS%ONO=wrlQ}o8=4Nyg5FBO%8^Rs<1$kmq)wCRbPFm0MWUHjV?K6^>S_3G`= zW1C33PCn$PcqoAQiDX1cq@xH=VZlu&S5!YWTM)jRlq4oXyFx17znNhPXM8{FI+tH z6r30N8O3+pM#WS~>h$Rz&e&K1*m>Z9J&C=oXwlX-Lo4c?EWFQKzf!YG7`MI8Ob; zsWl{_#b0|Q0&Z6xhrJKH;qo7uK*YD_Mnf$$Cj2+;=ifpfc@|F_KYvx{^XC+Q=TS)* zc%T)9+HL5**DDdf*`swALQCtY7s8sGjnC;e&PCeBP>eb9>jL_%TL#;A*92VcxdA`) zO~TfwPH69S6b2sggf`b}k@D>4&RhkhTiWJ39ICtq;g^>?=&CKPbkKk-U0m}RV!V@Z zaB(zT=$`@=mvln^klBn}jRli*N0w<%e}d+P+0z>QVHl33(O3MQ(A~EMv?{iRn-K7u zW3;Y`JM*gpTBvA%+Kd%)FY8aIRkpL($0eyC`G*n8a^q09VOy|A-xkgPw?Lq_xCwdv z=O7)p*&e=C2|@g@JkFE!JN)%kExeo$_s~U@7WQbNfrxvafy&RZgKL9?pz?qQ8npT- zC@Aq17>ii+dD*S#S>s&P3Ym)^%s+u8p3}f?{n4VAZIy(Z@_(a4MmJHaWdU}td_S6M zuorqZ=L1^A=pEkUK!27kqVLV;)5TA`;51Qxy|}?N`kROwrh5A~1e=<$0fP?e;ok^k zZ9oaQ9xe-hNb5kmv^1!b)PwbGOTx02|HOnN<{&gv73d9Wf{)i`&_#L&=)gppcAS5Y zZk_gwmfl!T2hE5^WGyzMQ7`^*?IdHcL^oO7*zp2}9(F~~cX|PxomB{P&IIi>{{Y(~ zzryo^B${`3A>DO5gr;P5VY25YSpCrry>FgP7x+D;n-j|DDNQ@)MNaO>wvAQP05S%q zh1tStlLR1ra4!g%a|e8p_=)xvc%T|N28hX4S(rHa4*Ms02Ky}Cf=&4fUyfg{qH#*WmEiVAyUm7i#vs_fmFNcl6R`klbUi7_R<>2m7 zT|w!fI4b2%;6X=qaSwVawkM_tY*(`bmsV)t!zU~0XB}bm??x^9p#dK}*R}@+BjUhp z`j@)yd(lW=)@kJDF@5aCI#v8HI4tUU7^YUE+o-L}Y7v7I@937R&Gd~MZ|MH;hg4_b zPh^$JEa+GAmiw}6JLsROj~12OsNFcng=2T7)p=Th81wkuR{GYNX8PX#E->T4RY65w z6gqJ<32R*~>itrUf%!H9aMXYR=lUzqnyv-Bmu)EO^4l2gSm;HsOytokE;>LW%v;3l zc7Xn~E2w*0m8sQ*2o?3F0h$lJp)W62Ve%h*r>i}FQv=jm5#OZ^E^SPwKeory?pxw$ z*9iyE=lQkvd6yO#tldDLIiUehUN?mIH8b90N{fa7|}pug`PD4X5IlR5WL&^GLeHfLlB@_Qrz8~+MbYUTm?`P)(U-%PM} zbT8Fttt6;ZIz zV0n5ex_Vo)ApA-r*4qJZ6ndg z9n(3xlt!u3UQ58k<&S}7;8y-J^IX)oU=6HqTm+u>&jm)sA}&ft9yUAHhK_sgOM4zU zjb|DJ;y3R(k9SCm z;bs26>L$xh@}`%Z1bSLIsO+UaZW}Ji^DWC0sAoOkA2rqCHG~uNNJ}w&+DH!$Cuf2n zWe0e3^Aqq;e3-lIh?b}abU!9mWr|7fMM3peF}Q|puiLrZf^Un-a+LKRVg(ZkxV?ln zZW^hIXBS_z9R zk&Vn072#9ew7~hHL9EMWAG+1a~OVu5j6z?6y$y?M0*ZP@Sw7KPix z%p1!1y$K2Y%S$o((Rd5olAnv$5t-O4Oor|nD8w3G`qM?$PhhL^bo%8+4QzbK2Ky5= z4+hE~z#Ol|Q&;c0fxMUx*hbT@n7>0Omi%rtD8Ao@I$bD06*<9J-R+M+%FhxeWLIMj z(Tj1Hv4_|bG!@w?=*7^vskqLjNVM)=FES|6hF&;39bT~2pqIZ?#dDpyz^&TdNPggR z!P~$45Y5LC$nwWef$0}DjC;wK-rk-^$1yw6)0GOS3vmlw-a3uIh*DhO=#NczS_6T~ zCLnh}18Q~2W4U3DV8`mqATJJrNWQ-1r`pJeGmR*kfp3ya9T>*oajGsNg3)_u)Sq)NsApw|J6#0p7Ub z3w*QQj6UKcfUO(uQ|X0kz^`;0s72w(e8jL@BBiWzr(!iKwtu+f1GykPt( z9%Jf_*I6|q(lH5OvF=qt^!Iv9>^g;?)&GDPukXO;%HPC!H)Qb5V@7anJPv_l_8|9! z5{SELfFXCz!k3K-_?q%WJj=ik?}+mUr^F@TjP@=tdR!4LJpKpWZgK-2O1=T@mI&dZ z{RQwdxe2i$FTMqZFK7vs{ z`>>jgT9`-XCTJdc2eh0uhFUkCeTmRrA%W9Z#$s0c1?}tC`{sJX)=g~L^d8i-C57wMXHyW>nQ}ar>za~?u z6>F4W;&eqgvw9v^YPOW~8^vnpFQEfmn`2g}s^Ap5P{{@=X^86nYbL0Lv7adMs6I~Z z=ZnanLrTymc@@m5YzE?^w-A#+eQL*@3c;*^cCcEaM^HFmgdP&jH3zc}bK|B>gO#T- zy3AUM9u|Ekg>{rTZ@C@AS$tau(L1u4{@ApImOddzx7%f-ntrX|zG57*a#cIrl-vlV z(A}_IvIiB9IfPL58n8P<8R?q6LEu*8Pc>y+faMC+uvUH-Y$7#j%{m!+u<8x`-6O}W zivCKUE51qp7~GGXYrG2nEJy~mca9=!5Arz0aFClbd(7EILZk)W6Uz@=oJAY_l%qup zyHHs;7hQ911bo8HfWskMD5;+aRXzCNNSPYi#qmKDeos-#ub)$;A5J2cR`03So-8`< z!hCxFV_o`|`Udzgii9VQECQ;{kKnQmrSSV$ASiaqr8^?C=vCp`^s>wPsD01`s=6nQ zd*PiIJuLL57yJBxT@S;+`=f8T$~IfTmdzLG-MiiCC@4#tC+|RZ_Osx&>mZQ3G{qeh zYvH3wd~RZ#1u7|GSV_(1fw?a|=rdoI(ZhFr=uZ(mWbf5>YN}!`h?18O7$s}Ms>%iE zsR;!%kV>HRZ%lApq-{`j8+Gbf^pvxM$T!ZUHxFJC)#$aaoPx(BqhOt{9qgRULONgi zBKLBSaROyjV9DnF(96ag7A16YcdO6iH2wX^qb&=ezkMvc6m=QaXDp-(-b+Khm|@r$ z^%JC5ErNMYIUwS6IQr7?3JBxN(8mxvgcs-_^0<8KjHmZ;I~}>abs?(g+e#f;JK_yY zR9OILmBhg7Et&9`ZwhtO8Hb-6wgJZvm!MLVIOJJ91)BF}f}E9cRGe!pdgkUY;Mcef ze%N6&URa_+Vd3di}gBa<=3N}@v|$C z4XM)+SiB5zW7l$Z>W_0~N!0NTQVKbGk8+(ezFTtdbaYUa0UWg{KZ#vJ~A6+wJUyK{WM1i#nWld@KnKpuKv|G$VM;=Y^6Dzs=2OG?56`Cu!ZU)zpe^Q@N`BDK#%_f)o z?~om`F5iqlZPi`QziyV_p_k47E8We%CZodz*D5$aLJTghoU6171m74PVB5Dvr!#>XWuxifheRugWll^t~FK@5QTJ)VW z_IC;Qs-+np;RYk8o9+}X`0|Wza3Q^mP9p9?cp0l zp5-53vaGIXotFK9%wrBavpBryU)vBhyEY{8+BhQ9{TInS<={aBf{E=swK7c$wf^aXLSb#LxRrC5nWN$W7PJEvoVcb~`wGLfX z+JdiFXVzW)n_)Mosa@An^p?Nix&l>wwhE2SkLMT$ALp8v#8Go=Um`b<5B8712_8c^ zQqi>^xdCnq_z}u^$jXRGs`Oqf;;lFv;!o_*W%I8K2LHKZ!OE+E<&JQ~C^|<3qRGhw6NJy_#f1O>6{;jaa&;N}ofJw$pVbbK;K%}9Pk3D254 zk9+2Tbz2<4>ZZUt5#pb_?R+Gq|1q3je!L&OA}B}C=Nv?-D}mVS*iNKw-VOA9bO3tJ z=`dFr!MOoXrencFzo-k#-XIUh6*1GF4uC957hv*E&2C*Y%sgeCa8A9DfEVz3`%TJcCxm26#O3BNXhNGAn<}01WLCA=>60Rs{Yky zZfDOKY(F>*J4(FaE#ygrQjFbfH&*8U4Mt*M%7FhquT>_ zl;`o5S1BU;ISJI_#fwnmu&>-rT2dT0uQ{mVt3$wYEDwBoQi2A2O$U7w7b%Rd0KMrR zSR1|`K69Nw)?u~W@2j(%+dcDu`03T)L2j&(}xpJILa?`$A#z3VE!(>H!p9?|}4;Kfpsq7OW>X&@*?s!{A#*sP|Yp+I8|Y zH)72>P~evX2b1rhF>)%nWy~m8HrWDaJhaBk=Z>S_KgeQ_g0ivsr$*pl&ui@20UPW{ zQ5a}7fatbrb!pj#OO_BYJ}dH>@1U)*MZj}OxvcX%rC#*|jw!U8YguO=SiFRsWDimEfPcyV7y zqD}(b;|=y>f1W_s>^mlKOvMTkcVdBuU!%)RlK58>Y{A@mBS8vB4xbgCi1!|Og{O*z zLO&53s6^v7_G<54G=`pqFILaMY~@Vx>8m<$NBd9sm&_a_E9emjpI?S7Nxp{9Noc_1 zKYbACFCF;R9jT~MTr!0XHRB(rDdAP00Dj%k3)`&F4F^S@EZ)vrXrG=x*zs8{D0SCJ zaEu=VTi;1Si+j67`qwthXrTe_S{Uzay~db-XyFnr5q$;wCE);#7VLw1a>?+c)ok21 zFdFNAIVAY?{UdTuEQ5-8orb9GF~AB|zCxzB4LD*Y`m@U#ioDo_*uH&f_`LjVoU3Dn zt0?s0d-goW>)o2nO~eOa*&bo{EiDQkV%h z8h1G|3ZzuKsN3IXfdb_Tbi-_Sh&zs;GwNs{5IxUURo77S+qU7aJ2delr=0NCice@1 zaT7mjT!o_wZg}G~KRkTwGCmNR3QXr66`UNp2r!>b^z~kAy7E~hz9-8TOKx9)Ry6NM zU4)O}x8K|8Lz#N?p42*S+ze&PjYvnP8+@kFEpNfh#lNV%PiIn?kt4`4J%%0+G2{O` zk-?3N=oCCC-N>ytP{fy1Z066l_eA5@o5P4_52(45=S3crQdoanGKiRwg5>Re$4Rs1 z1J?!=%dmTluCzOY22BtY7+NCGn}m-Vgt3$-W5o&_y6@*v(L>eC9!fNfrSwBU6UG%xu9%Fdqx>a7ie zyuW|IqZbPB$EpPCbBQKjdHF4F!NTI&{UZO*P-iDN=fZZF@uvd)z8S&7RiZGT*d&bE zcNIIMPGdD{XZZ&49|X6+6W)!}>6Cx&OKQ*c6M{EydEDa_snp7=3P`U>4m{Be;MtyH z)Z%RoBCcykHB8zdwMC}rLXSjBGBihEoj(eWc%_0DLt-%QRRG_3{qwRVXCm$2qZ_cR z=X246UjUW+{x_;_hQe)mBsTu$F1qv46C{W&0IDH+*pWAhnB`1UOx61{*nJ67hk{+e zR@av(N2I%6sFsS=mHQy$y>YzcC7sCSxGd!2B zDljRQ18848xBu@GAa&0amGzZ`E2C~lJTj(yKCu;21PMPK- z;G=aZvcs@b&^&E9Jlv#4D=aA$Twwaajdz3K$-yxAeV!(EOT<|E;@65DyO@r>O`y~=3q461TJ2F zPSD#rhnw%Riud4jJ<44#!fPtEZ(@ZA1~#7#*F8Qw6FHdu*WV1FjD<7?o9gPM?WmZ(qwG$#N``cb&neDaqSG2 z8DKz#XnJ6wUWIhU26g(nh_ksWy@twM^NX9W^-yq>`T@3h&7gm!<XFD6H$_Ks-_& zSI)DCjYTiOypk&PS#~=ddU6%BNjAe5hu?>`3u2Je?=jdPUoR*Xv;kAtZH-T}x(?kV zrcn)^shC=}2P}>>!}Hob@r#^dY+$)RHeXQ}U39mQ>-NS0%>BL``fEvIvy?S48v}WG zDexP%B6Bh9eh~zf1N@I2qg|&)SpdjSoqL4;Ly^5Xck+b;buBiQI-((lPtH@qiHqZy&pB@>=j8ych6=#NwF{y9<;(E5 z)prnGl0r8twbP&KSJU~gROvF|7s&PQhv6oga2G!VePhdk#-#z!N&PM;i_D};IA7tp zX+Fs0X-)dfZXFBv)jMT6Z zV|-`}{rn(CN8h|kpGy_#NA6d^{fFbADdA3~9*;(|!eS}+gl1SgDMi<&MAP%ib)ep> zBk-;ERcLc8ggI|e$XwP3%*m`dSdx{(SHSyoZ}$JTxr1%_&Xk zN4~bnP|nSUwEp`*T4p?!-fY%Qw^`)Si$4LHe)$kyecb@5;x71k2I z&eA`%K4S}ptl-+W`GB?dq!k9^X&K4)FyZ%i_|i!o{`&X|BEG-jAG=z(Z~bfVRU@6g z{O}*t`DsMQ-kL=}=S9%tuo<>ERHA)DSHR5#cWmfZ3W%+~B+@Y^!CQJdFSsS^}?q@^8hAHRtsXT|zJ0)uWJ+u+Ly5U5%T$v$wut5c8 zv@L~WR66`gx*$Ov#a!zq6J+?zN5na%5?EmdoMXagJj3Zv>h^eM@KAjz1g<=RvK6Sm@L?A03#NE0|^B#7m-#5ub8xq+6fEaWAeA%)Y#o zyT?QdsjN5y_B>k+Ee-k5b^d%rYEwMQ^lgFLs0YZ{Wy(?hk0hU)9Ylp68>Z~HWm07a zo>$*rsp$Mf{RVR9{AYA}9*)M``OZ~q9D+;5tAVD$a%gYCg6EcJk+ShT!QxkS@YF?1 zsNem8mk^@IeIrz&4lO#!Qxh?m|AlHmAM7A3B;5sHS!K?{D=e?C`3Sr{cL^u{&swO} zuFSn7@~XC&eT5U7AP_hSodjObazL)Ekc(Bzqoc=Lz{czgKsCD;6#csmGd;wS6B754 z^(_wshfMFFk|Ajn!b}Ito-RiczvuwhA#a%P0jVIfekvKBrT$nhg1UBT@JqpFc*Rl? zO@IDTkaqSFrLuAvH!p6#{jg0F71c(=1v5Hfe$s3@@6S^1J@?;KuF4dMUYP-0^?wQ8 zj_Odmp*`B8c?wvW5HQ1C7F=-a7Yvx35e(NcN0&Vmwy%7E#ooZsFW2@9nAA8C+jNS5s^3WD7cDL@I_`uxttf^2 z=uoVI|Bsrr4Vx@(M0$s88kj*X`FWyK>2H@4H; zd=AlOtCrJ7X=+$d!xAhwcRAMKJDt(_G)^aOH)Lk|j)Fn|J?NU%3VcE3Jj&tjYo2n9 z8MP=Yng8zTKVF()Cx6dhMMm<#H(IBniT*pU1$cbF1rB{sgS++pz_x|q^xMV|dc9RV zT=dlv*ab^NiS#ifR7(Qe-29iKHrCO0;w|)NX-(!6dLG)JIt%d$W6)ljf*y16##|E> zxVqQnnK)|==8MZW`tVv8IB;_XxN}HLfJPuEP|dz>*(N)0pzQw^G9;9no@Oq&&r3OCTuxv z{H=ouXUi!we~}O1ZdJ^0Z!cPzrU#Xyj*G0Y+rXh?u~g~iEI~5TCOA;<3)-jIqf@5R zSiYG)s(9KGKN28f=X>;G2Ttw9n$Ca5&S>AjqJO!;1Lk?$zSc%?S=8gHof-u$NSvnD zvl=w70YcZNlknl))%0!a1p4&x5$JZ|BZ~9Ou*r&Ebl9XF_4s>Cbm%i`bL~q|` zPVu>1^m}g>I;(6K^d*wuQ?D~nQ6`5%bkD-yN+KTNRa5+E^JOSJIi2ol5n?g#DzMD{ zI_yEwLjH8QB5LHwEOe%L5ti7mjsO1Hi=8afK;;akp>w59Qt#i^3(g&jrpi?ND8oe+ z)Q`MlpipwNAaP&_Jk>c&tqa@(j(&HfM#?uqu?%xKv)BmRy{8NlpOK4|l@AJ>Pf4Q* zm!i<%kxyuwzBo$!`wbSZ`i>?=z2-hiJ_%QST!@8d+ygpOm#Dv5fBAYjnPAn}W^Sv< zjz}rm8m4%?0W(RFeq3J_eOB}W^LkT_epD?5mYprw&*(K6Kcq~=uhzs*JaTcq(tgq@ zVBH^H@v5Kb23`*K=+YTXx@?FWpSS}&?#TxJm5b*=t}C7S=HAc$AW>F0}d zjG*3vQ-;q@yd(m@H_HsXeQBS?IcEl5&*t%T!V6Q#>Uhfa?ROHnn&T=3@@fIXiH{SVTE2sf`7Pvi1RM zjw@KT_BvD@C*jw!8#~IuwIB5YC(3|( zW5#*RaIYEGUz~#8jPQUrb(PS0L*iJg*$`!NP8vP(ZYOFuTn@bt@K8;Q+4PU(8f1F9 zDp2pA5cPxD(;y+jX!o zFc$sXT!g+kWQ6FP3_!T7FI4;F3cFNRVmCv3z{(SWg8K&~u*2KWgJ7@Yuz*R2cN^aU zBa2w{yK#qze|Zr|jbu@)I#-GM8Cy^x+l?kjE(YffgMh~Vc7P?uqJH_laHpRcT;VAr z8y1}~Pb^IY>k#cLJS9cg4p~{z(f@DL7vqR+4coqR*P#HL9a|4>wLu)6I1aXs90x~d zJC3uvtAp6O|9<=b0{v$kJ&n!0{~Z5k|38hx=6^R1CnsADV(;we?rQPhj6-Bk?ed@4 zyLT5={)bocpZ)(7+u?s7+s;;$30rqpoBt;Ef3o#I7vIi7bPxXP`2U0Xw{x;b?79DM z^6%h;IQ%E}e|`P`L-}`bwzYHqzq$YagZX!Gc5*=M|I7T_gdSqslkPCFDpu_F4i01e zqJWSDJR+0xkJ#y7N?N3_~`)4 zwkA7bfpf30*1LbPn_5Mdd0c(r zGaq^3;%7qk_%KS?1`m>c`z2jUxoR%KTDVKx>QS;t*Pd((WEn@pYL;qjVM`;8g+3kc zSz%opJ8Egc>il^}Txv)rJstZ=2B^45(-+B+BR*u}0uQFFD28e1=d#gL1?;tJIjm)3 z9~&c_%WBCfuuj@SBH&pq*)5hz+N?|<&4zPH$=^AI-^3HjCD(7Hvj-7$s+=7L)byUkJgYdSbfN9okh*m&qqwn9El8nTeX;j7r39=C@lP zU7n!Dv~4mbljY_z3*5gG&OUdU6}@waOm}}mr(KSAi$Bi1w_8UX3Ydk*yw0HmcR3Ot z4*Y_n#m4jjt5po|b|!HxD~2fTInVf>JV5m38WGo?-e=lv%JH-}0eDGCF*ApkN{62L z%9J%rkuToHl9_M4nK|_j*oyNvSmV$g1gICVzwQG}ty-DPci$92W*JRnN+e+}bmWza@9KwuNAgp|pA1fHzOKQ4i zkUmkeWI*a0hQ22u9Mv>tsgqBMuXdNnaP@VhqSy>pcD5mV=^~`n6W`F!esEc(RESnx z;giQIddL%3LWpj=7S`44JNtIC6{B=6nIy9xldm>GLT;}=yDy8w8b9@6Hy)nJgp`yJ zS{IAS{Ojdp_{wL@^8H<`Q~wD2_^BE}&vAAsRrpKRTuf(E=NJm#e!tFYMz1G>Ps_Rd zbXZDyXy~(>&e5#*;djiX{tn`{Wg_`@)@h=7`7HL$lmVl#-;phA?<2mSKEs^)ro&P) ztJoECR)lV6C~@pS6>&H68WEAPo_WXJ&AzEJVP+ruNuN2Ci)SV%lb-KWh=)2d?A#_> z;&*5n9^Mki>^;zf-dooO&q-+!qbuau!^|S)zNQ%4`%{Kd_H-iIvu|j@YzgAojzZ?T zsR5gGB!Cs4vL*#juG2x@=H%&}R?NC8EmqOhiQznwV0CRz(#)nkysW?h2hlzSoy!5}#%<$GSf-!8(=XKz*);&b=@0;w#i|g2& z=ez0biVtw1!d7yY)e|PT)sVqHC$UB6w20^x3FO$)2%;Z*#$0oDVAUqL>@E2W++6ko zX}VjEY(I7x>Ryp$=j0_3+LG#|=U{**?03WR; zPx))PypJYH{6PSbVSJJ;+_;u_kiDLa7`GrF9|>lC2klwKmRanE$dhFC)^1{TW*q6Y z<0fnKjA4hpud$0>ZX)yi?hweEmyFojH1_7N&Fqqs0ffQ5F!G<>d^~MaJ>eRZO4MkU zlV>jFlcD)KH$kE6kRK(Pn+eI6gW0nN@5z`{v7M&u(?{0 zKyDeV#ux7vXLMTq@!gVf%#z*StmlvwQ~NxgIbfp9&fK`46ZOKm(~*nb<%9@ z@m%)HL5h5OLYJ5>y+m{`$_V|LXXxwNa-`-8H!{?*0~(kvrfZx^i5th<$-oXvHiZkB z#cK}|FQ-O`(_MjV)8BNaL{ptPq_c>b~*zD(%(mb zUtH5es8)uOtrv{RX(sOEsr`_?_+$%ljbp+rOxZ~;OPR?OMK30~yKa$(eo3)qD;$OI zQk&RoKD)_n!=^5un;^L_cMZGbeIe@+3@-E=+_s9$!W=QE_|L9`}u4wv)nX}&1^WuCVsxn2rM3vS|8)c>B`Ei z&*izS?cYJ-+Px%t$HGwhd6N$l=>3bvl1Uc2$P*pwDoFag583%Qlofdkv(MJ{F$)P@ zQo(aA`TBw`(Ialf{)-5pd5x=yc$E$K!|WsU{bduxOy^~6L8ch%A~;J7L_Z?Gn*Jkn zb1$-w?&%82n^R1m^D}Z*&N3IVmFnce+gifl`wGH9=@fSQ*&^~^o3_jG$A&Ijul6uW z8)b!4U7Lhr*8c4I6wJjTX}614MLbzOOF?L9dqPOLQ|ysXGhEyeTNj^|qr|`bOtwJh z1N#UhGd45UzyR?L{#!8*>)AaCaC z3EemMG6B}oE~^#NS$&sTjOHn2Vs1qREAB4i;(OQ+Ut+2wpk8R z%NKrP&c@dgdpjf8$8NFYH{>xfF|I50I-gJKHye?POXP&kvaiS-&NL$V_9v!9B8uVZ z_>j%HZSYljH~ksmpN*2s7d^)J zzse>*tScf1^?ZrN502r_cHARQn+e!)8Gq&usAm57-eCN?ce26-kg3jfVC7dYA@8?L zCyzB@WWjy|#_!5o;>B|v(l+h{xi?`GS#MH?=S)=*Q6j!k2qGkIo2s(=-b6E7-W0Me zXUo~#${2RwFo|zj=fJKpI?n`KE4$F+r%B=6Fm~%Tf6`=X8|l9C4s%yZiHJR4OiE-e zX0JugXT|h6O!tmutoD^?R?W_m-5oEvBS~|gekd7vTzno`y2TjQl>Eq0^^0|b=)*b9gjU0U4%53KGl`pha-5{MXqmeKq zQs@nM2>yDGDXy_+A&hc3O9YlUkyUrJ$&3vZL~nW_(MxMH@r_0VFWr+!y}6sQUPZFU zZq+k2NjZ#uh$eedR-XybR%fqFUxR&nD9KiD`AUcHz017Z7f+mb4oMZt;pYUM_>st^J<08L=g5k)IC*o86gloQm)vl9DQook5}~}$g=u<_$*4Iz zV4Oy+*@I2#^v;FmEV9{@jr5gZQ|0Eexz``io4ytk-@o~iI%ez0o|%nAs+kOF_Q096 zPj)6ECN2^jox99ftQM(s>mTvK=qU5qIhtK&wT4a5I>q+aTCh&RN7;q8itL2KV%A!u zv#HOHVp8p8NwF*rnL4tOGz)4Z?cL(Y3%ZxcwwpJ}S>q){`itkx-sNh#HC(xviWHT(Oz$pC3n_TsXwcFKl2ZF6?3-q}!3Y zqWRMJvO@A*MK8UzT$w%R!DD)tB|>kj0$RvAo18mjtr|y$R&%HJxOQvj^#Ts)DVt?O_{E=dugsGRSKW zWL=OAN6ECe{_Maznl%qt%7(kiFq!`@5&KFfh%xsDV!sZ;6wC35foHxfecYF^>7D@I zs+IV(2o18V_AAjK{}i71HXr|46~Mk}GGV*#Q%ta!26^T%?|(6L9)30cQ5^Zu%vl%|{b{VL6yrThB?qm(Nu0Af`{qqE3 zUU*+Pb8^(Ex~rq5GlNt(f%3_Dik`6E^Z@PIaE|UBM+jbiQ*4h20pnMOoIUBSsF?q` ztgK)gw0%kgGce(0%`ae`nG*g;xdc6jmcfJ4XTseRwXjcZ6NJAT3HO4BV9bzz;>X1b zT&g=8PwSZCz6(WUVK|Pw2j^4&g$v;O<5%M1usE*2EzoN7R!aJFpA3C6XzfxJx<2l# zsD6Ds+n1~pI~MeSFQvyt$G&Ma_IwqMopM+7p7BX|uM-OtvleT^zl4= zEH_1++g9jeHVuv1Y*4#S3wv7Rk=9BjvX1;l*+-v3XXSaQTard!6=OJU3R7fv30Wtr z(`54&N*{HMN-}zJKQ|jLN*u|yCFdba$<1-MX$(!>&=X#C+AsZN?@b$hr;->k4G-Fc z+Ick(Hp4TCjr zI~w{u5$!^|idPD?ctGeL&gxSPnG4z=ea~z#dXKzs>`7Yp+mqqRy=VvXW` zJec-^kE~I~ZzhfmM!opc&~jWI9mfX8uR(p~9R6*)3R623(znM)@nnP@Z}EO5`qp)o z3^}z#*0uSdsM6nG^3AoH(W6oA!!0%U7O5S`qubQ zeoo7L-@*->Qg#)$V|LecPPv`Nlm4@j*jioX$l0^uTft!*owk{~?cc#3{l?>v>+w){ zJD)W(KI2d83v_1gY2H8K2@ahp;XgNy@`p7_lBf(dS@ltSG}SwVMN3Lp^VL(-YWc{~ ze~yal?~NRRn@1YTiYlWpC@U20vrllr z5)s3;wAcm}A^@V><2fn}&A+TyeR=;pLPNp0S{ZWI~OEPd(C;WsZ7*#|PQ)*h+i+nNnN! z&0#-PMC}rvhAH#0ks~0?y|Xyf{3E#d+3~=QIbhhYOx)rWPeaSjP))8g`9CL!XdHMa9iZ5v2tHqI6D5~MP}_f|RJz9!{~BoH?^ScD z{?s1Qc{PeX3>Q(L))}}RZvzM4_N9M+vcWn)ixO2Y!+%ZQv}Jq`adGci(y499wB^_g z$WDzA%kE{+;h9QslS*l^FcqwQ#tNwq+;CswVOpvsmF6_xp!{%sw7e5S_1$LE4md9j z*S`why@TNPz*ADMf`{~}bPvSNuM*M+&!?qB`eNe}JxH7Rfm9?jg_mbOI|jcW$`-%Y z;=OZI;bi_1m@=$RY#h9clofRGSf?m7lh1SU`(!KJxwa5A;$Ml+wr}JpQv-|)9tY(= zR>A&;!%S>Q{fylNG&e|-D#qCYkp{o{ z8o+M4*?dCHgDN4NUNrV$rPb{;{*elA9TiUFbE4_=bxk_+&<0(#hl^GQdr3v%yzt-d zJ9PffF1mk3gXUkzrR+}%aL#ZDSnoIE1>!-mL=^B+=q>o@v;gI@ny}~61x(rOj$Jnu zWAvmLaqz-CD7|uzvP`cFE*6oXZI%r$wKtLYVjN{(`vX2+zHsl5vltn#hCkL_p!!KW z;mn0n@z45&u!v0H?&{ zuqNva`I`)N4BnkB{@vR|6<^NM)`45aYbCG5R{J_~Q#uJvYm+Hz%}QF35-*O|>JVnf z?t(cJJwRi5n;0H31-JA$L`Orjp;2W9hE+7eTAS4{aB&5sd0mAc8$8fLAsUANJ1rex z?+rb-^#;f6F$DIxuzP7=c(T8YoPz#PS2GvrwNaI)jW(hAZBJqRj~?u5m(Mr#3|RC0 z5V0n31*_?qh;BE>k#SU@xTL!#t4=GW5if{zoi<5Le6+-(!5N_A>_wwyCPT)(FEDm; zkq~h)0>V0X!@!H$;)NP@{86if&v?1`N!SXzXb=>xnt~1TZ^X)WBOQ+FjE(=5fG{Ey z9(#GPv;G0`?6lF)HYgRMRw$BUj|p6%t}BjRJ`fr&D{@hDEq!otCW>e#Gu1Fbb>;`r z*HBfubANZjepjN5R@WIsAZ z>_pwfE4cZ|UA%74f$dN7(NgCy`dLiDd*!`^@sBd7>|QL#R>;rJgmZkSvYo%LP?eQF zKF0l5&SSTd&3s_5BOjLcq)uNZVcVvO_%-$n4({|AJw0kL-R}vSS*=AMmDBjRXPBrm z$P!k+HURilO#9tZ`1_C&_8hsAUmktJCt)LZUn}7itz)@wO+Sirzacu^9gHuk_rb@C zEjaJW3Jma>h`YZfVux}aDwW6Ln2Uj8lgAP9a*rm!bPwM3EuGuc_mIJ}U^dy7D7w7c z!}*sMknSN5zUZaOSAA1xz_^hxB-9p{f4C$nu9=UCLZ$F-!D2KXnS;}O>R`@zMZ7lr zD-=~_)nWCV8nW|ygiVDR9(3BeIEP#wq{*J2mWicOxXYUEcG-P1Vxoq&^pKu z78>u8XTJ&7TX>+fB_AgFcE<9j=Ry0}5g4tZ@7TGM5q7#9!Ep_xVn&q(o0eKbw1yI` zGrum>ryirfmz<%~OI5V2)5o-yjga=yTPSx`#CoTz(A7qnJ+)TB%dtPfWI!?YS~o+i z%8>j+ykq?H#GM$!pRd%lX)LHU8@(rLJ=v z`LWw1K5(u^oHStuKG^aPcDJevr}tD4Qhfm^L!zy zcLyC)==_>HH1W(Z9;aMJmu4;CHzSg{bwIrIE1V2bfoL;3Q$n{@VPKK(xUkd_o0((~d!;+E>8H0Q)Z>TG%#>_S7R_p+h9 zVrv5IZgPVq&&R;bTt7!Kuo~c>E6sm45aumOhSuVRpgC9{kLDH$17d!|iYJ3nVf7DE zFbII?$#39Lwq3 z-67b9ULf=OW}52P2vu{wirt&TNq9L7ZuafYZL2@gta-oaoaG^SXA~`@+}p^#wv|!$ zHf5}P+!xNL6jJ(qHI9v6A>9!p;O)LUM5)Fp`P|)EBwe-&$7oNd=Afw#Py3GqPnA(v zjbqV4S_q4l$@la$IzZ2UN_fU=F5b+1C0x3uLpJRxG(GV->|WVXHr836u7o#$@{fZg zc3#POyLQvRPu4VTM~7$8BSsn9m}H*LO{&bps$*v_gG-~aLr5-uF1 zuE#3Kb-Fq`_?g1dtWl0u*S*m0fG>Vo(*^hB4w8o69zY9Le1n>OyV2>O2Q2hUqmZ|I zKwj<&h20-ek7uf=A-{`U)_aS_3)}mv>Xtu$v5Hybr_YPZ!8{@&jpMX#<&8 z#EQv7yMwS<2}Y&fq>bIq3Nsw@;C|sfA@lVbuvTjk%$>JUlPp>s5V97=o&GLb?bgQa z{{{-3#p{AzL>;B59)`7QkE!rbvyfUCB7D2Ag{85DR9fCjwKfrS*g%n2X(fp7#V%-- zo{7eaZU|2ofq!C?aAlek3lk2oyK*Ssx~{{q-wWu@-yOJh!*TQ;8IG?m>_DYYSLs}_ zp>W~P1NJ_Ail4Nc;f~Cev|{5&ST%SO?%cE&`-Z%fYddmH;?DvhXiWsmKCI?}vm01( zcLICCRT`|Vj{bf*xT>!VOXtb8%#-QpvA%-LY>AzpRC1K%K0X__fVW9*K=Fg2XqR~f z`TM+4K)7neMe=%8%?A?gWG-b+ zvbbNWvbA@6^Tmmgxbu~{B;0ePBv)sI#AwG;957f$T@v(UmlSnn3ltn=mTOdGbkvwk zUr)yYuHSK~M>om&0xd~T&)pcGdqNB^wBUI=JIk&t{lbfd&%9^oGCov&SMVJyUrMv* z70wyifcaytpxrzlQ1ir|!6B%~H6LltTl{+AzlF1>~L5 zm6c*QNxNh=Pd>MR^_2W^?fI#)VV?_e`ImQ?R8hqP?rkO`_fP0MWd>LGvJndddq`f! zW^hc64BHKxUcWJZJ1q4{74XMW#JF2icY8^Q;il4m4a&|^D3Uetn@ptH;^f0(4k zHjd8?D5oxIt+1fvo=`gE0Hl-)ved0sIJfT+=#g=P`v&CT5lusO2|R#TAIw8EY~`2w z*D$Z*9OmpYkWK2RBvH|H<^j{M@`gTNaMIGpyiv&wrDc(HPq$1cW@S`u+d&4AMX0st z8*gZfL4|Gx)V5BIn-0r88BHabWM3s7sFh&W>x0~XmIoTHi=~_oGui0OCA=G$&kEnt zaK_h8sODSB8XL+nW&H?#aa@g4hD^bTAwALPfIX*9%z>Q0sa&P2ivxz3;m_=S{O8_R zaY5AwY{lKw<*=G;y=gSASue%$G7aW8W~{mQ3xI5f+}rSmAouxnux~m~`*|PdJX^+n zwKwvHD}T{wXF7WaTj0yvE?{@ajwi(^;_U9Oavo5Do(91~vuX5x4J?}Yj_ywwjg}tGZ~{wt%H9anHnt@-^E-6b#)eiIbm3cr zYw2d6fw-h1nvYhz!M^KH$e&LG45Z^FSDxp|dadgrxgXOTb}s56GyB?8a>cBN%tb!* z(IEFdR_Z12Q@!D=JI@JyKS#rZvYk9|QZsJ(afrLTUk&R!&k@%J%l~VyN<5Xfi=FEy zQ&#n8SQYzEdZhmc@y1#OhQ3ykXH%nj+vnl3g|7Aznd}0qn_i?9r%JJCX(X?TH>LsC zx5Le0yM-xT5E{HAczJgt>bCDTh)E?vlYyO>x$P1xi!Bo_zE48^mY%fKJc%Q1-SO(w zBF7PjCX)XyCocHco0aFS!p#$1*!_S^T0oD4Dci1tdDAZR�s8%ldLxx!0J~VT+B~ z%=L?AKxgl0Y#H-{uVilLc)NXQ8~8`GTW-c#k4B>P$7XQKix)zak8#24-$>8<@i}u- znf<{QoM6xaamSOn%4;*f{Q4DVMdgeA2W0Y@ABVBm#3(e_ti_+B`^ZMW>n(A)Ys~hs zCET?86zEs#$vro7F?je}arO9a^r6>FJpT>2sw0J?=8TtI+Z8Iazg&#Y!P#g!OHUT7 zcM%tF`Y6nQufo}`Wm0pq2N>IG!`Z|7aACK>c%!O2p8u~qn^_f6vHk&iKd!&n*ir#+ z2lwFEt-oOnyYk%B5*Rh@G3G`LlbzYV7}XRran;RgHeC^e4}6BRTiPT%K421F(>%z= zDqXNs`W@`;Ge%Z_yd1*AOeHgG?PZU@6oJ~a=Qt-njoUP%xu^YJTy^#)1dM#j#S_9X zeUv>uYL4Nb+o~bQYYc2kY=Dc&_2RR+kA-o4o>Skj>G)ko8M})5vb5xjc%Y^qzTI?| z|GoC3t##W_srOdySL({%og5_@v2|=QXg`N-vX>m3_JI4f2GNU+YhYt)GoBrkz`rN& z=RRIf@NMxG4*$CYmOOicA)2*3?Q93Ve?1;MH6P-vm@`nJrbPkn!|>7F4sqT2+dNdQ z7?b4hu;0DG{4i=M+b+(*w-0n_q4Xh_4mp9mE(2X`nOB>ZVWEox&$BAx8J*1~Cs-PB2RkqO!)CY6;<%$$ ze0^g!=TEzU?z5J$;;9PS+gJdj`+S04 zU2%qY#S`AldjKA`4>?qOAv*tj2d7pX#hWtMIx(@Yq{u4)? zIt`8GPJ%_!AGqo20dX4wF@Eo$kU^Dm6!3Vv4n7ZAB7RIX z6svPbg3s%B!p&iSgSsdY(ro;a|W@W%|c8c;odBc&Pgw zOn>!7-=|w~QAr@^1Xh8>wOA7H__nuOggt27aEAdv{lUQ^yuf zwwk8IuWB}nIsabJsMXn^lO2gZK3CDpgNFO~^2?rBOc zed{JUOYxX7dk>pzQId5}*OW~=tHV1rw_*>~U-*4REUp_IE3eV4*}CN|>uUbwpvILv zcU?Sw_3kCP6KNqClywZZ$ZN}etITCTj+)5E4lCxaKChus=P!2tR)gI}inw=%nJyic z8u7{@8GK~%Gx~8%{`>4r#f)b@IIy8s>R7#*9Sq*FtJ)WO-O&KMo?OE*IvKbnbTOQL zF@us?TX5nyv<(Ikix{dus~>`Hi9vo>|cG;00LzU5jo3y>aBhq1a$!1-0JWFx0&( zx+P47IzvsgI(1O!x?2nUqoPP*?g)PLwg=zpZ^HkEq;kvcExa{r2-j6=^1kP0EV-`2 z9;T^mHCLC9m1=OP-bP5u8i=~t3$SqRK-})6hCypo6giiRO{s_q`rqY+r4dDB!km3)I;M@jJzF;3s1`dDdWpe`e(LKc@_674WYc)(u zR_2)YLQ4K=$ARj-`1Os?LXuZ1EgYkS>BXJ!oSiyAK9r?pU?5D8TmhKTI1j1jZ_ zVd=<1x*(e1_`_yc;N2h^1YMV|@OUXoy;cZ{<0hBMt8nm`t|V0K)fJ?3G;zkB8)Dj) zW%OT7G3CU)mI|v$?7c5c+|jWEhCS2d_}86zg{ukdh?^s=of8JP-YkLvYj%)%;|Gd9 z_m|_>YIBx$1+^?cOSixzXq+ zU4=KZUxMY9k$myPE?ytJoRU?%aO#l>D45vsRQH!WY?lkG1-yimc}LKRVzB?PDxq8S zMC$ze6+G)*4$e0WIMDZtw9KKq+^5QrzaPhiwjZvnoO^})tvx~gdySUAV}m*Wjs_Mg zmZR>2ava>$1yWb0^4>^g*}``Ta-TgCb%M{McIYwen)8EhHP`V~0)F|fUKsJ_A9%hz z0##|U#D?-JMEP8Abx zG*j-&zJl|X1{iqgIV7)L3be?Y`aLq?{Wp(@^Y30nWhwbOO;p{LM==O94n;-c=M+)}}miyblxvw&+-tXI5URMk+cd==>sUjtSRcU&}6F8Fd^uOGhZ_p@`~!DmZ;+vAB6q zjI^XkgzFpniI>u>aF9Zw81QvA#OZtx?X?#3pq(CNUKU|udD|JW>(gOuTc1o1C*)Ac z=Siiv_Zz~BTcc3bNF7fq_rRLIebA%dK`5UP0QtfD>78j4r1zY{mp`Y#M#bKIZ;(`+ ztvruCN^_-FV}|gUiKpRkzyfc6S z$EODn*V#j`xjvAe6nEytch#)-Czf3c4{(Zy7N4md%X+!Rw5IkusO1jD>`XrBUfL?OPykA;PN~+ zwXWo5rvp6Xzg#|Nxr0xKe4s1;o>RP>r+)Bc3~rpU04C0h!fzOe(rGq0?~fh~bbmB@ z-|&;Hvh4xYS~|13`z3ffvyQ@tMT+IpccMa>gZNU@4$6D?hK+s7;rW7z_+p|id;z@9>x8|{b@B6&(~kB-d%=jl z>gfHu3!c%{#@2v)@SxKb_@29+GS$Aq^85Q>_~o5&qtTxBTWZibi6{KHmM6SW@fIC^ z)jO!pNQ18ls^HSv6StJb!HNZ8^shLO1`R25SR55gzvjkJQhEtZoT|dzjE|6Jx*vVs z^^vk8YrraHi?ljq6weOS<=ap9(0@}qXj-5ZKW7u}e5@;9zW791gmMz3X3 ztr!33>fdJJQqQZjPPYpUw!7^xE@8CLMb69iDsB_j=)8dQ9chj+T0`LY5I1n{_-K?Qj(peF=$F~b`gQaJT- z7R+(%hIh9P!M4eL`0)zOSZTtgSA>7iZ;cR!|DKwI@^3>+U@1pC^9eGW@-(2o*PP=NQ$6bm?mZRVW{! zhDCV}A4B}H%+L}vW29$zN zZrtejx2miVuo3FzzLDbDKeTCdBUCI{LdIDa;ntEq(8-{bvKLs8cB=-CI$cY+Nl8?D z-$c3gGa==05&iz&Ss?#z~&4t3d|jJ%BsjWl+3tBMnI~DBYQqjbF|caI5b@ zEMFc24=kMc*XQB*aat4%Qk05odJg8k)wdw4WGcVxGMc+yyMu53%i*0?7Wm9k8{gFL z;-HNmas8n(3htIdKj;2&DD1I|)-@$#n}eG3 zQz>TIPPS%E$&18UeEQcne#iB=N#z@rJG=nG%C|*TWvuZkQIdfxk%KQT7v&>md}a*IpA2 zwr8`?i3ND9=Agxpx$Jc<3Ef^^;l1xtc>mQ{+%iW^cHiGX z(yO(QO?J147QLJB*vZlSC3FbyZ4N+u^qOis(|N0${kY_r$hV|QZ27|%56T8pN~r?- z`2NPrdB4SwITn1OxfX7`Dd+drt1vrNk!Q&F@)!kPL=UN=EM09nI&2&z_A>C{yf+4t ztE-30E`L{&WJKhl-VQ-Fy;egKxY>xEBacd#ByALIBf8LCt2FwMzD8I!QAhX+uCz~O z0lDw{Ao!^Fq|Q@}P~+(VsI@H@x2zAN%i2wFY)wk(JCk^^y>0}b4p&5bt3vVCIel9F zO%YD&Od;hGpa&;9n^;d>Vm|TtdZ(!SZ|3@hOz`J|c9CJVbY| zjO9_A3^2moh-aE@2Ql;--8^9Hpq+Ocrj9A1=;exhvU?J=%{-3JExNH%@7`z;bcp;( z0q=}@OhLPcvyUtcSk4{H2!BJ9Bd>#ZQYV6zHFV?TUTQY$FBtY#<99|T()+KjO7D&d zz?g-%V7}jWQQu=J)|w9x&sm)&Z{KRsZqi)Z<-LfD9v-CLuPoq1k4Vap_to=B3hx~Y zX$)r2@!JvLnlcEy-*<)1TCO;M)l^LO$%or6I_O-|P3}cIfXee)>Aja>;4{&RPg%S1 zhm>5>cAm=-EdncDTg5MvugblhBJpot0c>!V*ODf>7$_{ot|lI6KtDydLy@=+M+h$q zXXBPGzvL78?!i{?yCl0GPPU50JotAE*|})&{fVRa`j~+ny4aS--1dUXAV2tZRv8P% z2a4L&KDb5jg)P~=;rwJhvEPMk_^DKdo$C?=r6lH<%X5V3i-W1S#TPB|5-22j58=Z} z5L9#o`&J#Njmc)%&9$!Vu9rWr?v7-7dMfA%Pr+L$oQBgOQr8QC7s^Yx{|*+CN~Gf1 z^%a0We$(M;MdF0a%i^8Ef!Ob`3fBd$qsRr7uzl4n*!8j-R-F1EydALzdYJ0s`;|%1 z|N1!C({&##$jfrvI^U1J&a~j3n@6J4!Xn}^7385em#s4Tu=$EFv}wV4vOJhZOJ6&| z^}~UX_p2Q~?C6g|wt!+_I^4MGgIN|U@yMe-pgzL^;?|ADE)h1oK3oZ3eJiA+SNn7O z*_*6s9nA)tJ$duifnx26B~W3R#`Qlf*z9r?Ec(y|Bc^9!nA==(7FZ%Nuvx7U#Ji^?tVkTJwoWjiJs!VMO84p&wjyrvMNu?$s~8%0M!uL1WVR@YRCm$Q+cu0cnG=RkAdkK!o>NNc>4EVG^}D= z@8phtnwx0H`F~`+{~$L`uH|=b*LnH9bF2|;&krY`rFpskz%^hBt}MTZa==USd*?44 z-+c*YUfqIu$9sq~o($wUqd)S$$a?Nur7dHN&)iXF!K>SuYSw$g)BsNzK#om!uLG ztNz7BI)WrdPft?x?Fc4x{DxCoe0ktV86Qc|md%vwFymEqW%JiHaMkKg9Qx{yV6uNV zZX8~TtJa!GhM}BY`EL(a7-Zpx39VH9(vkC;uW-g#=0gIpt6L6#joT~N`J-vb*N4#e zcL9!il8TwX!f>bCR=k$q4?T?4v3r!jT4skuy9Yb?--sZ-fc~6N>cY;ohSI+C_X!0F zy`X+XB91?9gwvH@!PEQ4gs|WLpd&;kjxL@q@7ql1*=1cm;vYs<3h!ymyc3}P-H?Y& zeGG5nhsfc)=ipy~s(w1d^Ee2*8eO*CAe`Pf{VkhtJA?|||IoF4 zkKwnYA54+wt0!#uDYj;+a>`SEa=mFnD8Cz4#%~ALc6Y>)O~Un_LwI|=H=lg(4{n$I zMa{QnbZzzu>9wKL!QS{M7;Pt@&F^SrlAJ9`y$?SMBFKf#fNzYdoB?=2eRd4PaZ}b( zke4T}i!7m8oA=ZEcsJa$IRezrf1ursKAd_j4k`bED0mw3(>5=hTRahe?F^=rJ1=Oj zMl1wh3Sg_+7MgLsH|)_rLwcU6@J4ACWs7y7cvQ~Aj;bQ3HW!+vJCWliM2YzwlN{f= zP2tA}l=yc+lQ^hkFqCvTO2HQHl-MzvmMlqtB{O2#aJMH!)ePh1Mk9ETdn|^fv_R-; z8wjdUqzcEYG*&5B>|zY~Zs2HfabZsyAou-`TiSBb3OkG+F$C+949nVdr{Ix09`IzY z0IF$A<#k@ZaBr*+T&tc3dSS^psBtaW>JEozH@)#e<6DS2RSM=hi-lobM^MJLK;imC zh8;iGk!Dzt@W639{`K?++s4hJ()l()<6$g5H;AE8!(T{G1ns9<$#vQp5d)XEsIle_ zJy<)h1D@KP!BI*J(O*LVEE0=Zdn;b%SK_sWvjb&Pp;h?KqnHFbKS=?qWl;ndJO}pP2nqNAmBH zGyYbNgbbe{oWA=LznS%nW2PC%_IWmOPKh4J*an00nw|KpzpiA?sz(?(ybf=KMPbug z1-ho5N1r#8@{x3A=e7$Rp6Sp3<|IK+b!#khIUv|vnvWOvY{ua^t8e99k920X&PNpY>;QTT6pni(#DKX)JtRgr;y3qkR{n z(l0YWhqpnm;-A?5?-GvI%fi}}-Pq9ho|F&ia?Xh< zYz z$+mtL_&28>3ewu>_^)Sh-rY#}&-1Ti{}o1f$Y2)a)-D%ZPIriU1A>JXMR`r~U%cau z_%JcONJpH!@v%63C-U*{f2gG31{}|Hk>^R=+3#5rP4C=BjWcq^P3g(v%(~qW8z2+w z6km$E9)7rB&umKDwhC@dMHsaw1{`+zVSZ{2Tq1KH-)UNLc-zw}#uWq<7ZB&t{*ryYJ%=F=N)mwS# zRwHrkDkBc;^_HH_doI-UG?wOTeGodo{UCnXeoGp&twEgo!y3!&N8?>N^K|p#Y%E`~ z6dy12#GZx6VaBc?R8?sJzrY0ew#tnL+&mAf+E#+=w7t;pi&U;<`f?AC!E9wZpFXOX zvbWPF?)q*khuTf#3v;tsQvI4D+b@dO?we6*vNbyTnmOaj^Q$1@O#!4`m;R zVE)JbIQ3Q#TK3SxjCqk*ly?mpqjK?gPCUlPd?`!*nhTGXgweJ%2Y69o&rc^s(B}sY z!cmW&oYLnFxgYP&znzaTZHVEDagiMQrzbbg7U1D!Qw$oqo<16?bCu~n!E?z88t)rH zMuD@SZvnJS3zf`iL8=n#eXGgXS)|MINr+j;fD3(Veyy z$6txvx$JNsA+4+YLEsnk3)He~u-tElP;S`f zu((A7g&J2fOFAK#9DFa-CO@K-(1Db)WfT}5u@z_jC=jdVtoNS-XK7u24Vm6LOovks z(1ub)jy76NOYaYa(ZO=P{@zGFJGeyb^KlkMu296F+k2t!xBw{atbhT!KJpyWMmYY_ z1V4Yh5Bi@xX@O=AGRu1HZoqUd$T8qniHuC-Y|)@Y=F;vxH;Ya;zCyZ+1MWNGi@ViVvtsKX#fRKP?7((jkYbw8TtaKZRIwV5b;P zY#YsU^YewrL)FMCHV(%goCocCL$F-30y}+(6%t`K994 zzDi{A!U*Gr4S>(5m2l~cM=(D^5v4in9b0EsinJ|5IIoj}ZHmp1aUom)XB*hP+8955 zS4Yc${ZJib&X=^hV#Vke(7EC+>{-w%q%|ml$^A2sdAtnPelSC=Z5d^;B^ea($_=j2 zC0KbsK+do(C6!NQRJG+Hyt!}>ZlCZJyw$ggi=21RPkH>k_LM4yug?||^b*9c|4HR{ zsye^<)&Xm_DAK37?R3Juof5`&5&J4+2(#8r;Jfm?V`xpYSe~$-CYhweli*dHpNG*)p*G;2_u=8|ALy!fmW*6#0k@@*GxvriEgJNDyE^2}+4VU)wQy{8?W z5{}XBXZGAj_Kb{r?~t0F4yULo-vqm&iw^551=_rDy*wl4L=9aQ;EIbcC~)Fnu9D0a zdiQFCJErEO{XEq1!vGWbv^1K$1N_P4l^?2gbVq~9J?L`(MQCW3M}O95kyh8KXzIKM zRt`+2dj{(qm#U^YMuv1l*C}zrtA3VXI6;xNei#qFGX~JRKMk;C_Bp7i_#>2*zi^Ce zJx3*>Rq(QBecATuO4Qn8FSukq5Pp2U3RzDRNE%rUpN=F8OX3<*5PlwZJ z*I#ppesc@1_^Z?Vx~r7`MVrcNN!X?mfLHwcap?#(mc1(!X2|Qs`|GozChxku{=E-5 z69@v{n&3f2MQ-{s3~d+2VcPJw@Z&h(+^=@<{uN}gf-$vu3y_wW$b2BLC<_h0G7~;FWNpz8)k(bP$#n@`l z_%s1>yLRRI*W$&6J6(9v#YHsX%PZ+8%Scp~d=x^0`-vL@11LD$j(7BQ$I8$)@CzRd zp8uvoYLPeS1}4I;T16ZqeGRuBx?^#DjboqjiyfaGyGh}dsl5EeU&{QNDjJ`^EH*6I ziKUMku)}ITPL53z{$6tDLSubdjnyw+HvIxe^hm`Myo?wmX>)JDe_cB9_g8wd>SSA)=jRc!wHuf7h@(4T`FsmW_7_2t`r$S{ zY+Z@d%|~;Lqoz#TN?8_h>mK{*zL)1IN8_A7=doj!qC{ufe!RSO3Jx%6rzkwilgwZ7 z9mC5!$>1h=YfnL)oqKWG`U}`U(E``2DRR!dK9oIH-ltjyfwp=KHD1^!Xw~eb$Gy#{ zM^vp4xp*t(I>*7fvo~Om(1|>&a^4wh5c;!XV>u0v?KL!rcZ1nDtLV zGH_v6iKogHyx1cH%eQx+^GFF6IBALV@r<|#SM#~A{+!aUt4x^ukFEYQk}bSa!ws!Q zvX(!<%?mH{uoa^qQLat5yi~$%&KEK4iMqu7>`Al>G?Q#sGLwWYy@_`W=HcFgOE7-9 ziV%5C!ZtJK^N(ArxZ~<5-hKH3uhN&YiT^Er-Wr};*58@(YJLl9oipIpvr^&E0XyN(l`}ANsukaO=8L9={e{!2 zhaK%)Lohrp1xw|e+~vJ>kaDXR?ccEy4laEm>gCtc^15DZb*%?Kij?rL%x2P=yq&+c zzMvDg_i@IVNjTlj7?YQ3!#Dp(Ix$cOUkp47vek!h#{d`XBCi>KuFt0Ny=};2P!dWx z4BIVcL&D%i{7QP3-mcVR^TgilGEMFUU5}*aD-AL1*B{v7HkIu+IrEunTfS2Am!fBn zhVr2&XhTtz7{BhLqw?$t7*KlvYjsEA&|L%RLs1IMulfK(gFNwg$4_|ebp+I#7eH#{ z59<9=hXOxN z#Av0SIQhA`czm894nI>2+EumCoR|dTe=VoiSLO?&kH)}N9SwdEvY3`@+$s~^+kxAm zLOTD#kv9)JBXs6)a#g=CLXr`m3i{*dntdP4oa;gLSS(21s$%5AMDC}h%5RIZ;jOp) zdu@q@+ZT6;+Cy8&D`&j?*;k4;3g^OB!$2`iMG433vxWa|7lMkh2Zy~krtgFG#V)cC z+;cCC3^R>cp_?)1EwQBvm#4z_+7~d-u0|ZvO`BiZ1>v|zEL-_-u{_6>B`oe6O%uV4 zXZkx}m2qz#|3%>7vOvC?Jr^ZmK~%kaBesvcA`BkjEi@PR$2~8v27nuOI;^B~%!o@zcih}~;^*_x~=VAx0cwpp2M z`q}dyX&U{Sd4zrwqj1ev1N2zphu_qWKh3!*T8>r?DDY8)1&DdF{=;k07+9&Y{U#5a~d=Z#fWoNjuIe_S)-UulCjFoHtZ9bXGar40Pfv|M~DsLw|hV z?=hBbW-wAnd|dSwG=IK`5V>j*lu zU>xfnlF}b(3a#@_qknRqICSqDQB`fI_#$kY`1$Wh+8L4`D@jizAo#{>V z{}i3~Kb7wv$Bm4vY(-Q?QHesvxv%$0MN6cCjFxt2e4?VH5;8KfXI4Z>L*(4o`$Q^K zWY08cYwx7K_xA^X0LSB8*Yz6Dml9a2Pk@DP31nSPG2x4CBUJ~#GfN&CL5bE*hU86! zo-=ObYF8|2)a5cMS_$;*v`%UR6KK=mAax2l$6Cc&(h<4Oq()H|6{!KF=RcJ&L*e9}ZCA30%zeI-px z4@K#WL$rQeJA2Nhi(c3^kNinmLuZ?WQ?B>{bwosU4z#R}BofE}vS+n|$>Svi zM#oCf(Gq9I(M%Z*WOHZI8v%vsy2LkQ3h9m*X5QQP61^+RaI@?ZepGYEu4!WU>GDPt zzx9K;@bnC`r^=JgQj3A7-6C{wPB*h-633^zng)B-F2mxQ(_FCd5TqQIW(W4JhCk|G z7;V*0)Kkp^P4nE)t==9rgCF1$g*;R}Gy#toRG}ra4HYLu(BU=d_~gk%p4)*Oa(L?% z^0d+srf-*kgNkRNpqT5lKMa8liY4&T`5IK`#KEQ~vp79%1vJQ?BV#LKXrkUry0_F7 z7ur=}_`-{r=5r29x9`H8wx{sNTfoWxrsLF39qJ)l#9ZHI%bL}zlD3pUa#=|h&RfZn zcz+9a#lju%!MGl(qRnB`+#uS#L4dd$bj&V2Y>wSe zt_!q?ZkQQX?7oO=k6PiwGqcF8!zS2dG=tHeREWxU33NwTH1m&_LksdhvW|~R$eo@l zBGtMGetFwaH?F@h8h3^)d$$UXow&f<&{Bq(8CWzI_Q`iJLu8Fp_m}w2peP- zz$ZKr4!E3vw9_4Us5F``7fy$G%^vn$Q3S?v{_>fVm#M}oLt+#W2EB=A@Ww%JbUX5Z zOp9|NIbV0{W#mZz0@6~U3Gb!EaKq>EP^&aV*m=w7 z%F<&HTTn&h-NiAwWHsz+@nFTD^O*bffaa_3k|PFX@OtA;8n{CTdK=Q2Lz5=bf%%^W zTc<7r7l(ej!*@6C`IAd*4!U4y;%Ow!Q+W*{BkTgtGLCtriXQ&2=-^sYwEo%%iKa)8 z_qCp$)^%Xx*J{w#Zh72HjxpbcCxdq3X-1e|f=P?3$iUoC@*{35^$a+Lp2dl9^tdEH z%)F4}VQPSq)MR2{G9CAo{-C%ajD92ag73$7_@sA`f{G!=NAfg1cyF*; zv-J`YUzZOTj`+i?um9LPbyrB%t!IL_sk;~#7fv^l`$1|4yg|3&A!+C>XZ*FC>9Vhu zI3d&oGpF9A{g=5u#?NCY95KZO_EETN{$8a09P{aRFTEr>2X{DR(CICQPSTxMY@dBBejo|b26!f&mld51L(b|~D zYNqccuE#owTHi)`PtOu}EbzoZ-EwptI|W}jW@3c4EZ*29#g6=*OBFRw{vvBfnKh$;c`-CWH;%H|3bFz&;-M~W69EH zBN|iWL)}k}CyQTxA_tF5MAc^>>1vCwOj3>nmH2(0?QP+D)@_?f>a$uJ93ll-*B0Q- zd!tlwgC>oB8bG7)Dv|e>B$7%owBq&)64pepoO>s9Z~MgPtM}4lPXr`$eiNxNnMuRF z#L(~lcE|~hBvYQGfy6d*_B88&zVh2p*ZU0GzHr4Xn_SFW*GrN&83B)D)7_6K1@)a@ zVQTVTaC@{2GT%qhu8RvW_&^J;(e4 zA$R;t2yx_mSEp;Rf`o$EKn)Rm-T=3ks-pD)ZIboA0`1nW#_h-7;>8gIJR5!v#lsw^ zsdoa|eP13vMtMWBW(dUfhu8a;^6ZOynVDsq=mM)KT#=NF)lDuq=}kT! zJgI}5IL)SZzZ6N@;smQ^D8Qt6js+Zh4XAr49LNILDk=}Zy%@|K>x!YXR^!>ft$2}N ziZQqI&@!9L=&`rR@XN_m<(&t7Tl592e*3_!S=S+QzZlG?cBHNEAl>+DI@aoJMO*Vb zD9%5Q`U9CL;ywvwr!63P|LH(>=QapC90JEzZHKlW$D!A89IW`FLVHv{Q>Cl_=%V~I z{61a`H@|D4_KYYDES*IkxoM%_&^VglvyS#AzoSh_e0UoaOXO`wVBQw4k39AY@s=>* z_?gq`YunBEeccHANYVxWcttaXBYOl1G4aIKRgN5P3neoHq`}Wti%53&(|?ohQ(j&b z6Y*s^v(_<>*42zB-;*Dc{+He4vzY_T56EW3E{!mU4}PFM?<44;&jRiqE`=Ee3vjP_ zCIqfu!@T$RfLZG~eP1h?JlGULx80V<*Y7!|)JZp-Q1qA@4wuon7Iwrhfs#2l)nT`% z9xU6RPacyUFga`f`O{5Vhnt`B>0i2Msajt|UC zvHVvFYK|PE&s0*#G9MwFc)A{BB^E)wTsJFzqMDIt@unBTLa=&uKI#-};kdpHME9fx zWElCtwF#GqZEy>n{!bgb)?8*E3pxqkRTICNE1`?H6;UXg44M1agXe8emQ$2aEx{f0 z4`wn`yG^iR@p?AnVGD6s@r|@QZlhiy3!yPdNUo@B<6hG|`a+OU-PTR0TyZU3mFdrZ z9kFDrPyfgKtS68hv;{<+lSq<+9!_X9MvwPV_$+@V=B<9gYIXV%m#Q$R$8Hc2kATrb zVc@n_j{0VQM2GlFR5HoHn&1G)-!BITUKEnX<(l|s0fTJH6HNRVMR$d;%-q}yloMC6 zW1JV|@t2Z~-ZK2X7Hi?Cp&OB3IZ2>8DhJA{X92h~Uj5Z7nm3YW&!R4|@@C^~?7lq7&mFK46wwZwRh67|T zUj{R_sxVRYTQR^;4_~W?GiBXHbmx`??2Oxu`zpm)lLwk$B;o`gHukegIwxo)KBivR zE|EQwxp4Dx3_O1Eg1r1X8{f2=63M_|YWQCfc#dVeXT*D2MUD+O}#Hb7>GY+Pjv0->wd=yPb%=i!zR!a6;9_4r=>Pl|C>rhTzeA zbcMYp*4aFvnlXE!et0c(gj^(7l3viEY8#yAHi>?UcLh&5b@0)YV>mt%&TsfY6+Eue z69MWF`{V=1e0xa0YP?`wGaBg*zhmT}c|3food;~B6p6gQfW<8t_;LSRydmsI+o)S; z-hG3Nt?31$@>ig4Zp-&#e?iC!Gmx69hCf!$5+-C@3a=OI2+a>XMeop3@=EuNO||kEN3!WOg$7Hz5vx#eK&H=l6Ku=@b4*%EMDd9MgMH7d&!v!C&t@Xed;} zhQbhdq%sZE;_p)6E`dJnoAA|!Y<%%@7Cx?Qpw*miQdk~N&%|4>eqkSo&6ex%b;>@t zom>o&+h@VZ3}a}}(nG!VQ7Cz)0C!){$L6FiR8Ss|i&pZ<*=tF#dDBg(9;?bXn9>Y) zlXt_w-Ie%v(Fu%u`W@FPiwH9xM`KJ_GCrTC&fQIC!ZoPjywoHZ*Bl9&g|lGHSw+~^ z_?b;uwg{UJuEJ7rCu}Z`!9E!gd{eUoc?0rvmt+bYYh4E8mKDLWPuJn*Egg70U7szT zJBGY4ZKIx%Jt#AJ6{p%KU|rKXoHr~Yh&^-&Jnol*^281p{MZ8a?ZqVRq$)Y(_?jIs z`hcr)Z{pygARHf)hacw#(gWJD@a5?R_+r`);_r{a)GM#Z#_1>Ny>V8!xibLG3bx`O z)dW1JoJbPDlc;I#hcht-BOnBCjg_heh zAge{4X^W^wTctwiTI_~9UWwq;H_i}hGlcuLY=&E(`e=i<6E~s_cL8)ufiJ-_?R^?Xsb|zZd(Q#$kb-KbvyO9>xxT zCu>dGh?M;{nEEUpdd_u{zPu|mEnJD$np8_qp1vfo_W#PJ?J=YN8fWN{m~kjun#NkK zi)Z4xIfmAyUkrWag3HF=r2n~HW2dc*r)B)vTt8PpzH07*_r>YtNTLo%DJ@|~_m7fI ze(BUB-;y*m&mjN3TSEPxoM62zYS@;#km~hGYEzy z^K#L6te00~Z=LC;@xIA4C|()W&Q64sH}hfL8$RR}&m~7X-?H~clps{HpUn04r~hr5 zLI=f`Kw@hb`6%v0hWVkiW}*~HP}RqkLE-4gp3C|PL zMZyS8=+j4^MHU#`ejHT=zBpfg7hV-B#NIWHvU^v}WO#-5$ylQSp65JCoE zU1EvNshCP8eO3vPJUX6c8wJo46OXc=&Wm8$w`nxnB7+{eGM7zuNT)#yR$-#=LrOnR zpu>TusLy3NT>JS8{SciaSfb#HYcD9!ztoO7$Yrdb{@%fitDMJXdsz_G*RC{Cw3F1n zSp|V;hw3plumEGBRofcY*VnTP=7_-BZB0aTpAr0V83%hMxp#e?1W@$@%+M4iGO((O z9Is~CEfQQ0@%~rgQpAAGdd4Ntn%hS;5=80-^RXxD(9- z84DG>Q&mCybdH1W@HcWT${wDUG?JV#hD=arCTUQf2xbSviQ}%LR7voP^mrr?N8R6~ z$89BDSNDK=|7;{WmD1$fSQi?lER7dUAJAFlm89ahG41s-N9iYv>Fyv;n&w|e7e0`{ z|G0>}c5OHrv-cBQVo*u=0aHlk6GJlRI}c|^-lh#aX^w}rl(n=~p?B@C5{qAX?D?fy zAT?$(w(%V(Z>=-PEpB3$Dg0)ew>48gmZ9H%>QLJe4UlMTsV@ASMcfY_rZdl)k-biN zTqgDg6WNnb0@wLLPth8ieb4KekAGC~i^g%fGhd92_fQ5Aj>j_d<}~Vm@fy{rtfeXm z_n0n`Sxj?P8}sV+T{31z139onide=x=9RTRXAgaH!I;trD)+gOcI|h7nDx^k>t!wt zQY)i@uNGqRyszx#6-QZJ?Qdjs>sII<-^Aqk#IR3#O2|Fc@pv@y1l?nqMs^uV!lVh! zW&r0>$;{sL^Y zoQeh>KWP1>2ee{!73-s92s#r31^P#h;jrd$1e0!}np+7amDW%$U|`X69WvpJ3qAC4 z9!ReYg{j{1@HEK*qE`lzFJk`8&Hjh%q4O(fPa4;08mME3w;mIu))})_8~UmBlQ?=y zjfYhW4&rL1YMi(51a=-2kQo-eTt+&DdQ|JdIl(U0sNEPHcu&Z7XMMO|y9&6#P(K5b$8)S#t<~UfoJvnz_(#p=H8T66Mu|f{ zg0HnSu3xSI+6I~sdMf}YIZHvVvM&j3QR42F$H3Q@WUW;VD3)Oc!p}02J2*;3>P(sEscZ20&1#s|cbx`=RFIX?1cS;8iTSAr>SiTJUHEm>BPE1m ze7caeC%VZPIjYP?^@HTnhym>MXYFf z*;TSc)0LR{0ke71YGPYHi|m!rqhkNsStT7+nta59UNNp<=l#~Pc8y$kpG?kWe7p*(^%8jLAd>` z3+QJ$Q~laNoUhaan%8XTcKdr|+4XT?E-0hI(X&u{vkYzSYoeo}9Bgt>6yCoU4IA`2 z$ttN*)LuLt47W~&xg?4GFF1g#-mnX+4+KMjf-R(FhoJSW6YzO!9Z7k05%(xgBoiAi z(&nD^g2eDe_-w~E*lE2E zTb3U+v6(o1&t*Ma#_>P&g~FD33(eJ|Ln7StvLPFjj7fvbDv%M(z*ULd+b`7~WH0BEhGQc{Sz-oETz-`t zy%h;w7fd1lx;Pl0up**!PSROim%?wB4Uv(w<(=ACNIbX8<6Vg;+PTRdC)shD)}7th zlO={{{vF5Y6({k2RT1u(Y0u^16HsTt6_oFc$CgNKELbN-{>n&UQ@byv-My?jv#)aPfAfgf44 zZ}|n9XOzu0-&sIymO7Io_X>$|X&5WW_9Y1_iq+$zlc}cE2IlrUIe@}6@?pjwYO(P= zrkpTE+h_fR#K+PL|C})YrWuAjnu4xtj8J~cY0TeKiqBJnxW4!dQg5Y?XBMwS=_T{A zV3~j&vFIXW1|nG#qf$D9$s)Wz_ONZmHR9DNM>St?+@;Mop??VAy=p9YaT_iBIPU+0 zrDl87D#_^6_V^97nUV+Nhz z*GQkp&!eLo7IB|rJg)w!0v_Rd#5?i?c9>7Y2g{z3^qdGLN-L5^%@d_IN{Zm`)W<04 zonwm^Ci8Z5M$+}(&xlih7@67qmUaXt(8`EUuQQmaYJ4OP2|0ot{r>D)sYs$XRKThyN0RrOr!lh56N$g8J`5F(vcY@zu)SJp zCCY{#JzQjj==#9P$2caZD&VgeUIq z!`peTR4}6ej(SMJtHe?w*uNHSUP6MO zF_N{;kjiFpt zhUla9g}N`&hHpk%B=)5Q{^VzZXw4Lk(IgJTrPt_w_CGwG^O*Cq#9<2g%$oO=Fxyi* z$?wtj781yQ7I$zxp|o;(nXSEUKU{Z)~U8GGl0gm@EY6R0zhPAzinhS-+!mqKHXv<1ReJurD_M7Z8&%%_ zhs-!2g*(#6prle1)Wk-hj3*Cv)IB90x-A$qKLTL!5_o%y;yA_G%v$cu?J2#0!HO~1 zpQ*!tbNU{pKW$~wRh0Q6%aZW@tao%ziwPW_QNk!KE5TKAZ>j(79B_4rM!AG~+8IBE zAACO%b2mr`qo%6ypB5N`(# ?_5`SuREy^8BVFzyM zLGYqZRLFV@fg6U|@kc}Obmt#%ADPA~%zTH%#e>ksvB%@xb8+2^W3X!cAG$`)27=q- zS^sy1xM-FP|8&R@^^?)$?t4|ZRsTD<7951h$M50^k%M&QCuzvou?Y8l-OK(dT>?)o zTtw@?^N3DiFzm>ji%%uKa1gYutmrOL;&-Qx+rx818_S9O+xt0YcVNN3t*LUp!wXLrB=4UD% z467t9_3EH^X&GBut|ij{ z6FhPFN6UAOWuJ`nkaF2c%#ABv;98f8o}F>{F7Xq@DqJTTuZ@KnhvGq4xdE<^&ma%u zig1;fCS#VsCSZ(Q((xqdbwOkrZ)0+_v`a7dEP_e zO?i;|IR`vD^iU;7fmoONfX&uqiY+(riMkh%3zKk8>=TlskprYmU#K$f1o&+%1`R7A zZupS{g;y=mmm2dFJ)SWl^&p=JnXARod zpuBpv@N4uWzWqPW*ZvcN54|Fw?9mYJpHT$2P9{LbDKX*74sP=<=>T~(-vc$TYVjNY zP7_AxHM85UkHDtmB0?v_?_e*I90X4EElGHdK%K)h;#=#^O<)9%(pOiik% zmL4@s?uaJsce%hEoZ&&ntXfY)V5vY+=PIi@DUqhUyU$KKc7Z%uUB&4t*>r8kchnj*B~XATtpdqG#vNS38d4%iJ14F z$G8YJ7?{5v-_Bf$r$6oh(J`C{+NXzMx7^^L-T?hn8%o`6{*a0X|8d(tv*GFzAKKp; zNtV7VC2Q`5kqu=P#6cn+oc>E^9%y|e&r&$A@82*z=g>)(N7}(o2YZ?_D~(;@TB-?Ov0J=FCyN@*FodT{UWCKcEycN+~= zEm6uYkzDslroLNL>A9*qn&qp$_>C{;=XId5+Vh+F~e> zwzRIdn0|Y`3C!KZ;Bd@$8oQd~m}RIysG$L7^e3aARz$UVr~+{>DP>hRuV??>TTI*B zr%|YDq<7XXW&TWvCE*LxiAdd7B5NcLAqPLvt#uPIOrwa~!nMNfvQi|xyoDT8I0bzh z9zcV~9{9EDCAk`Chpv};F-uKaI63tiZds^@!P~fOU-|$Xl>81VyDj-v6ImE)oX2rI z)rH!&=E8j)E<*QoA-5&vOFqO-<6m)F$Pd$B#h1xb;Tw#~(8-eqP?R4cJRQABm?$SJ zT=UnAS}YOe-`=~FZ?VOjKjDTHKf0)uOlK<5-f5|DjGm6LLqbHT>RLlj%SFN!b4h;i z2MK<`?;|ju{3T8I^6AXAmvG&)Qgo~_$8|o0xjn22s?XIS|IJA(dX3)DUCD7jcF)rKk3>(f}#Yp*I?7LT<@WY}S2BtrQu}g-)cuG9knpaH^avKN{ z?!Qqap$Ro|*5N*h-DHD7EHtl_;Xk}P0yU00;Kotm#_@}=C$$r2wLZep+uqt=&mf1bJvze1XN&`T@s$%@B*DgN^wd^GA=2! z=Z*Vo1sAH0L;SA{I8t&4G(MyO`*|lE^&ZF5?}vALsEt0*Y9>H{Vd9FkkaM{?SBDjdoO~fTn^RhKZd=g+~#G0i16u( z*Z4}f8yA$FWksfmlPQHwFvR)8nx2=zv_}j^jATID$O?_@)le#Z0V)}B$e<}+^uJh5 zI61ipTgNX1nX&UBFZmu0U-}5)<7EKLUZS_QB-uMQ9e$*BG4mVw$Zs`5Ig&*swwv;ie0Pr<_e22A5i@ntWs0Y5t@;rVCRV4&tv^=JL#=$9(O z-!`)m*PeaH*o(Swe}g|%u1<hNwsL zg@60^f$Y&B7^gZ+H7`10+ifoM+OeM8O1g`ejtHQ8$vM_fM^r^z=y`vY;u17+RgBGsm`r&qrM$Km9!(e5S(BXv%_Ic`v$X#uaQmtb(;g z0W?zPALtzLB=k}esa3e ziQi0l@kO}*c^;KA4MlBd2b7K6Oy_kj;ywHg(W*mK<@S;p=}x0 zjrscrgvZ5&(dNI%v3D*ImG_a1okuut?Hzi!aC!C0N-7jR2@-3>`CsaK;jUK|QI-ir zD_$XXeW}5nmE0!l;1s&TU@p{Hjp1MPk>YRMr3m@nskq~!g0PWM5e7XxL+6O)!r?dr zeyPL}II`T27Jh5SqwY0WKI<|W``8&~&DjQyuPn*#|8IP(S%^OitTDiJfYe+_0pmOI zaP6!XSZ`g4Ik!u&d+050RGEg&z0TY@{{<2~N5SRuS;%hQgeCrygnHLCgi%cosYYlx z+^fD0>#DQB>*G`sJ6D%B8@S=hfL}~-%r7=!uQy7BW}v#60%maIG{{KF;NXr9bi@*j3fPZr#YQNi983?4z6P~R zO8nnu4P0+(8@H)Df_|tfJet#s(OxHr_44)bLPm;juuuS3Cnk~>B`>_Z^$jjvn18{x^l_BY6-s4GGCm1$^kz1 zTxFw;T2MZ_fbCg-6I#o~+43m`Oh=>?PINg*pU=5U6D9hHKj*!6c;07gH1@+;Gfn*2 z>5c9De6qUv137W@8IvV88v?g^qqL;B@Du$EO;``t#WV5mras(kahaEUXENXSxHR-u zD+`A@Pw}2>J)~o7-f&w#wwU~<29@W1h8wS2>DpbT9MhSPk3SZmp@<3J)o~b_%&ml1 zlQ}NLW-Y$yCrdDSmyS15>p}Nc9p-3vQ?ZCuT-JI#mXDR;J64-R44Elh$l-J?Rqgp+ zGk>vtyMnRfsS@GIn1C>JIyQPqqBr+H;bL+b%b95S?|nM#vM?5Yzcvi-OQv9Qa4rTl zi1DrR4RFB15=UGpef)VbFr~LaQ#_sXd%`fMtp=JJHR<}d3o-uLC#FvrN^(d2Fn8!Q zoOdXItS%d2-dH!9t*^_spEC~b9d5*BGM`X(LK}2{Z-DCkU-0rhP1032mIf(xaa#xq zShlB+`#hCEFRc=^J&YNhb&|rk1)1=?<`nSeWa6Zg@l;my3OwGXAne(s&E+7+@xMKO zgZJCCs4CZUOmdOH^`>ro&vRX9b#Xs6&i+%LUCN8h#z&To|9_?O<)aJ>j??jaz2y$N(ljS5*f z?>?*MBL{Vhw(#)3HZtE)5BTz)WJg>qv$=<3n17VU)GHjP{L2dbSj6Z9~cPg;sJ4!9!@6Ql| z?|Dg%@t`hbY~6%`A0`Xs{;=d>9~uVJT7YyY(xiBoV%Ytt8*JZwHh!65tW^1@9=&6goAT z3QHn?;Z9{a-2dAWypPE6djnGFhJetH+m7J#ingcUVPkToN^UQEl`xGp5I*ga+ePOLvR@3AMx9PX# zpQwGM7o9t^k`A?vrDpjVJfYfCYtsj6Xc{VqZ8a6dsd+c8EhuDdbAQs6i|XhW!An-{ z=|*zkUI#PV;|R$uyibp0kJ8(#1Bv3kd@^kEjfjjHWY5SBvzJbAx?ETk{e5-<^oOQU zBGOTP$lU~d+B?aHmKn@6)wyhaZ z@=aY83W7{Xt={55aZOLj)9yNmR-vX6m>Sq8=&F8p|&MpX`&QL-sP2 zk6cgQe9I^DJN3zXb#Vyxk)%cTI^=oMQV?jqtv1kCB!3QHAR2BjD4Mj=!>aYnrtJqf z&OBxKrsBDlV&=$|F8bHK zhaRX+#x)nlpzH1mdgXXHEt1-ZJGv*+6#o=PO23(Yb#A7vi)^ZgPZeT(bPH`(zeNPB zHw>@;gSoLHe6inBtm~ICLgo2}pwPP>0uC9W-LHG6m^+qTYb1-n#CO2 ze2-@5MS^JhE!^bwi)se%;@{k9BrG&ugfn8-@?9KFghMBb>0#+eF#T14udZf*Ahm%E zUx>rAQK9%{^B}kiY|yXZJ@NGwhXd1wFs^P84)#bAF`_B7i^zuhsb|1KTtb-MY=%FY z%=nAFrSGvYWRT;9rQ-AW z9I|k{IR8(JmQdM$30zY=3dFFNZhfMK*-1CZKbtYQ$8Q^C?r?;K4h0+sR3As3Mfvjs ztC?UgTMVBf&UfiG5_Zm=%GyW|LLx<6q$vuYqsCyO{aIM#rO#98I)T5q{$=H!II?A1 zE;>0z!AE;>BaoxP%0<%jY?29xvT+@4o?pnz4o0U_@&Eydpe}}SF zHPhKe?b2BDM;>!uUgPN*uVikT){={I*~DoQ$6$5T#TDGPq~K;WHCo+Sy)z+<&ikPb z8u7n*y;|#-^WNuq{w6NeN9rfDeX<#;J8_KuR6IjP_t%nnay+u~ygPgPawZ+K^8%IE zsHVdwO4-9&_PobSB576UVG{Z{h)$i>LH;b9$^AaHHbURqHqSpD>OS`3YD5&#Aw#*GrJb5 zVOi!wnw6o!u$Nv@gYHvQAp3|un74&4wtPloI_*$;>MRWS$B=)EE6JSRN!WO;jz+G$ zOSPqqnWMJ0w14qtvNAQ8QTHKq<3yp&fLa5i5%!!+K6aN}|7s4aH8Pl@=qX@W&hk>X z&jp=3>LhmQVwnGXFA3?m0`olH6NyR2tTj8xY1y{q-#vMd%}*n(=Xt#UxK6P^rBL8e zHx)lf_|m~KB8+CkerC7dQTnqmmfBdP(eRqT^kdRmEFYSV#-})j$Paz&ee$%rwNncZ zA6ttRx?BcxlK^=)Z__i0<&3DxbBfLkP9DtT;qwS~TtF2`E1QqC(VyAJwHu)NV-Iib z_bw*L=sdG+Z6pkBlOgMqI+)M9w}D+9mn(OT1JeLQZvXi_c{ZR!hTlFXN8`<)D4URq zEA71RtL8BlQ4Ab%+6>80BdFu!Q;gltqwGwXOm5Rn67O?77%_}v*9KeTlSfzSi%KI{ z%yEqa25m5_DUeyBvxtscauKDghpEPeDC$MsNMQXDdV#NszNep&RillxM|~Zy zK*fpP!R^F{6tlI5<7sHzLt^YL4L!?_F#O+xWM_K}bD!xWpOd4=mNneAkL63!Qq<1$ z{$0a*h~H+Xt4hOd`4(2r`3&u_@gf>d9c*-Y5?xbaLglQFV&kh5)Z>`0Af;Q8`b%*c z(hL81{|=AvmIV${lb1=fb*U`wj2=tUSG!@Vu#&gEsE3Jb*ClQb_c9e9w-J|#!Q^6v z6-jVhP8=mFh_(MVcy;nC`J%IvWE9MX`JxSE`P^VK;(D5R3ImworsG7oGL-n}C6K0Z ziEOR@HtM%o9L>M{W~+PL>HcdovA^ArYFjy>xluU2xYENmxqqQ%>XG=qQkq_8ba0|m zH~o<)jYdyoNd7`Cwmi;IkUV=1cs)m^@~SetsZoMnrL{y;;stSZj|bB~t6={AWN@4{ z5wv^F;l-OWa-lq%v~*^ZEwK_1e*Y>t`NNqmG^?kN_T8;du^Af5c^@ZuX*U2&52xzrlA#)f7 zA|dGnHBmNj{>?}7uAhMklQLoYTnVTNnh0WT#x&s98j@aqpXKk&5`geB<7gctxVicq z8Jd(#mYOMIO~)3zq_6>JxEd3~23<%w(+)pg<$`AM6ufI8D$Hr?M{`bV`LHC|<-AYHb>!oz#;v{(VZ!_GeVu_ylC&3-r<5WV6+dznIrWP;H zldPSsOwE8b8fw&5-H9wAOIu~YtxFR|g4JQ)hekT)n;VY%^#N7o7GbR1F(Ucu3v682 z2^&ppq0Bp$?)1BX;q)G-ZQVf?@ndXf$7!fK(*vKxdcisMK8abMMz4IVLF*4Y-pChG|G{+>b+rw3$p$Ou&0=<<)cu7%ZGPT=Ei zIpG9l387f=b2@?Jmi)W$8Ol~j@<$yMVFlMitEe#N;C>T?F;2=T!Ty28)93KT%(VI0 zT1#NGUrng!ZZ4cVTT+-d^pa+OSPatBZTM0a%KX)13*kWBR*V>M5=Mxt33EcC(Rn!o zeQi_tTYswXQy%5Pu7JrH6x4^y>~w`t>5PU~TiMAPR`7P;U3fda4zB&WLoVN2g_UDE zP%4|-9lNc7KH3&=Qcwe*9Th=fHbRy?KZOTt?s6T;zpS#ub)wvsj&#vMoX0zb>0Pqu zJjawo8k}K2Fvn2E+6)^v+fetuW#rgHA2N2kfaoNN(@-1^A zcHb_j{*S;XejT)PUVtx`S>63;E_bJYL$CC{W>$!b!o^iL$j1vx94jaU7fs8c2DVdh zrt3jmGI|K#&m4!np91ksT_yfFvk9-{exmoyN2t7+1^&Ca9D7;`)Q)pt)3z^WH|TKN zoF6TqX~+`%@7sf(>=9BpyNtK<H2M0$@1`8mgv zUOEy%bYEQ(ymjT#GRIL`KP8Hmr}VQzpEvZ?EIzm4XNn4+PqMSL%85jg7OQ(JoYu%Yw1V5{?2ve7i_|SfG&qHv+@en& z>qpV0a<>_yh$eQ__YA$tWu!iz{m2F%`$8kjg^vR*;537f?Z0T=Vm+KA{gsZm-=XS*7pbvr zA=lv=q!%O_n1BIF&Srce8mq3ulISah&8Z}c0TIMtfZNBOX$JN?A5n$qHnv+b3(Mu! z;1lT`c*sN!!vucZj@d7pp>=2Je%=6e$ZaE4JIvvm^c7I5T(pQ;heMgJ0dE{kfBm(mWrYQNybFxgrrh3W|nZy+52sXQb>cQ4@!e5 z4K%9gp6|bRt$Ww{-TPkay%y{J$2sdg?z3Od=i|{Ge*qMS)xsKbFEm$gf+54n@W2-_ z_;8{CoH*cz@($M_X;~@UQ|gJ`)1P5`;X1rnLIA7gtiY4UGqCcv3Dj%99Ee;Hf{sF| z@V;C%+{PE;wdm)=S=EJbOkfV2S2zHal~`20<1yjac?7qm4B^g{8+fOU9Dj{+71kRI z!4fr>nS{r6V6Eahcy!M>xceKIeUNR1O;2t?_IDB7XM8Ow11mjkz%GVf*!RCV;FPHitlkv@mD_(q8s35LAICw7T}NO}{|VUAuR>n@ zCdP684&#UBXE0w*9k+%><9EO7ar~td2o!In{(OBvikGheuZ*(b4CiRrx%nV+zMO$t zH~WGb?=0$`^ftQSstq?!J&bmpXV8#aGuxQ33dP!4VTD<1@oD2?*1h36Dsf$ilT4Fw zWRDfmtvwfC7Y#rq&OeasHBaPSt43-jsNx4&@_3>}lkRTf`ZPTns42$>XG_##A<7uH z{LDm-(I=q#EdgloejJ6zIk2M3Pm=SNRDzNA9%S_A8d?y&3^vLRQ(7CU*yoGK!9{)x zVGy8>_P^If#fv#c;5!fCeS9NZV;zQrIOla{;XBa&IT8IUtO90BZhpXR z2JGLagxVZtK%Xp6R+n>^I(~0M3)~Hn4t~b@7IfgFlVWIURsv{Ihz6(Ph+W;r@p9Xdf z2hgm4=_u~uQnc<-8*sn(j~xP#z1Ve$4GA6rPh>_=3$={B7ycS$mhY#4^<~uM6Hkh9 zIUc1;PeIu4AoRl}6S%i8L0ZWxfu>jvy2-5#9_b$fJ`x$AK(7rIU3LUZQd2;rMzr<5 zg;~@{S}W2wO9gGajM15=iHK`HqktoF;8S@NnCGv7o~ZVK?EkoVa*_k{wLh5><+|Z1 zf#;}1+d^cZw;K%{=|s!QS?b=|G^9P4NzLgA0hHh;pq+OF$k`v|90QNo`+3j60bM=( z@6`}16w-!7PO3q#z~yLv*BwSsHWx@8-wxzmIHobj+AB#cLMOKLGDW)D@Oi2TSpO!L zECkW$>x;S6SVBB2J=_YiE@puzepA3}NEog>5W}8I8yCuIUrJOKt0^3cZMGaLhIA&e-VWNZ7g zP=L^t^Hr--PLc@)rn->I&d1zUD%6`d&g16$KpZYa0+-q^n$ljGO&2^ z8OXi-9p!Tz6#vfzI-7otMRf;2;;0_1hn%+}a1)*+z9J&Q4|P1{Izb^l;Bj3Al{|y% z&Gp;DXKOE^-`B39-h*+JaLiTEeyASI7l>i|m2I$%`XK5Op^(F-cEk*?z)6ilSagXQ zIPXyiojbDOLTCsNUt5QhrLw4ieS@59=MohB+yVlJ9s_9u2mGlZ8DGjv!nHqsay>Oo zEc0+1zIq`B-drQXQ=J$`YwT5k#q+sPX(a*{b2>nw%OX^eTFZ>cFT!#HJ1|yN!F!FH zDF0PqFsCvOR(UUh#v5xvP8WktMHS+ICqwbLu@3$(tsb4=GDl{+Ghl}2b@+CT9t>u5 z;nnta`0%>VSUvg_erxa_dc9B_o#{LU8>+8ChnmgM>7XRsJy!`o&)SW<@^|31^AF1k1b@Q@b;trcuznDI)C#Q*`@UuI3}dP zs)P4|I`Nw-8W}}Ra?cTa`Xp$w>p^v@1?;lp3e>~!J{03Lm*w4#L>sOyggxTQP&V8h z%E}qSx6g7=+onMj@}d_1>_^ybFa=NTc?{z1IhWqpH*S3=!qfNVzBjj?!mk4)_`jW( z@!wd;^KTif#xs6TLM1JCo`W30+qmxvgw6%%vfU3X)^E$N<1gkb1m43b+`Q|lt{zV* z%8s|#5b*RGufk*HZ*jmr3w}l&YEXfI4_8+5S)?jhR%wG;kWZna9M!UXu?w!hFSaqGEescHzySwU(0oETh#Fx$uacf zxF#mt6tKFJIUqWIfIt`ph+FE|cX27mzdaxI?J9xm-|vM=18tye@e~#F>H{E}w7}2b zyQ#GYGFh^A1z@))a1PZXuJ7T%{!kbOwsy1d(m+R?Vxogz9T`Kbujiq!Z=0zPmt{bU zu^vpzIR--x%!Mab-T?hh{$SDXLLf3{9Qfbl9Pk&EVeHy>w7fKhK5>0!@5VadZ@wNM zFF`E+Y=whm6!5F3(I|YjD6)9CA4``q=(tTdFpQ5!O}l+be_>N_Bhw!4akYeL(HG!z z6%)kMTnPFXcL6hx^KdLf4pvq0ft1Z`yz=8&qIFI_GVlI^PIBHADXk#fVN!v=>Br)O z2QE@(+BVoi>n}PR83b(CT|wtnHdDi`>0oBL7Hpr|1^?|2gEu(7n&gcx2qcHWXNMB7 zJpVHw;T^Cr&;&g`o`)|lp5z#=*|=1U!s++IQT5RX+|OkmWOtrIM-@-u+`b)PZ}BQ% zddM5DsJRES+qOg9>H=8vxB_MlhC<_<1n~2}4?tzZ17=REB)+^uoZ|16!b?(quzkoA zpZpwxjr6!ZXt5d&sXxN{=v1>RW?z{=MQ>#AgUfc!R)fnA7sCZrccG44GdvwR3+nHM z@LIY%UOiNR`(w25EYySQ(?pU!%l}+1RPg2rY5g0u4ig;rCs&?C`WD zTs==5N*{HEp69DTweCG|COQwuByNFv_qq4kXb=pMDCd5U7GlFpj@@GT8x6Lf#IE^? zEVFnCtjfF(H1}2lFH0pV+I%~zUKByK4Gp5cqNT8xHHD&@W{?>ZA@4qQ1OKM~ptpm2 zfXq5yJbh>xz7mC~O+HEB-$F$!*&Bv>#iX$60x4jyK^wO!^imqbMnE)K0(J-M!zb>X zd%h>0D*hFZFA5rhP47}+IKK~s>{9@L+|Q%vYjLQ;qXX$)yTdAqZbZ}7Ja%7gAlMsz z1so&Y$y1$=kb18hD1A|b%6Oy1F1i9J2t7lk7T#Fvycty`?+gpg3_;fFaBc>xj$?8% z!M$Z$u;|Nnia%o&64X+GM|>6V_Bv;V!?KSW@6LN1C~kwf)| z#~~-{<3Oo#2x)#PL5dc4IKI(ul<4D+^fo>Q^LKLWwW&NHvFJNk-P46EAKXJy0SBP) zeSI)b#{>I3`~c>|+f<;n2w42P5d__x0DZ$5tn)*N*G#D4Rez>X)*}OWJRlk@JCF+n zwRb~yjx!tym<#_)&jWf3d8m16IS}OhQNz@1us!q-=x|krY2i}b>pc?+{k34yVi6c2 zQ;_>r531Zf0R%K(r^CGBsAQvV)^VZ}MK;Yv>{|yQJJH1&n*Kqm=I*GjP724DU1F_1 z>!Rfs-=X0&3oP8Y0jcEHfCT0xy6P2+Gj|4JBVHal;;|BSMTc=N-WB-jM@Qt=TZm0s zBavBc3*xu_VBcD4;i8Cz_|=;Mw$mU1W$s8qd_4i|sBMo{C8eU95x#8f0#P=@-uX8<3S`+CmZsR>+1oqdB;**oGgdStDO*UWftoEJOg(nHK5QlQK&=J zp4ERJg3Ru1M*sa{ko6lw?4LP`+(}k)cn?+~=yyGus-#IuLN+@ppP^mn4kZ`2tARzXDc6 zBA_wg6gsKx0<|x7QO_+&APqfH@P-pu(9Qz6dVXfbM3nIMJ!$yf3`txUIs%H#k09>? zjz@J#2&oL;K}q>esAfe9=sA`M4$X~ZV@IYy@AM(K^Ufr-Kn;T_*D~r}UIWNor2%Dq zLcvQx^HmGZ0Cqp;%d*2jt7`aKm8+ODthb*yLcplR* zOBu$+DdUgl6L6&6BKF6TO|ZtLnX2yb$9YF&ko9W`>z4%^LBXfr)W76|;6Y#s+BW3? znya0mdulKQ-8t-IB}3NZ@e&Ny@z}7tKSA8HB93!=n;i<8U^NbTvDa2g;Y|*A(aRST zp#GE(Jk9q<*lQ3)r<5X5@zbay*_@QCSpxJQeF5V~6!GxO02F7im`&}eL_#y~vNBVP zpvb!+Aochm$X-#1{MLFRjm?>O`_w z8+2514w%Ek)EqNU&>SO)9aCN)$#)8DzS}2sr469#ITt96Qd4B=TaL!#evyZl@zHu6 z4e*;J8g0NK}EP9;*{%@0dSYAUG2@F5{T7H*(R~%mHxl$x>iZw+(c9<*{r5 zqK-9Cuyp<)xURbyu8cZ~cw&oz0e4*@A+`aSjH<%T+e4v7`d^UsqZJHm3BzwHXIQGi z87Hf(!I4aJauPP83PS2hV$A`x8CfehHV=oV^ThR<=f^w=bY)xPiqL(IE1qEPHjf zHEMX@3Hovs;i$A7_%Y82Tw0k5E62{lobzMgX6rE^KjS>yG!O;fZs&5e56s~2es}m* z1Hqcp->8FgmZE@}L!e@PGWf-HHojk6356ERpj*N3*ekW8$Ubf#`m!|#YvQO7l zAKig-e}&@UJ!ZlJcT(`;Iu#goMg-o=x(3_|CE>9pg&?Bc8)AK3I8#ypj`^N}%i~WX%R0`@ z;$Z--SH*!pN4t=rsxhP=WCPJR`*2F_d(?cNPf8OJpkM~aISFcG#jPXZ=Wj>hbK)Y5 z(+`H$K4-z*pAqc+t{8Yr_bb@rrjBwR?t#C0Mo7B(60kOvM(I%y@AcEb?Yi$!kE{(g z-*F1BG7rOl_r>9JG9SR;!6|etxGKi=oTuaQLEkCObXai85_(u$9fC z5MPi3+JY7Aak<%O7B`QdtKA6dyf2_zAQkT!{f$b5&M-liWyr%{4Ep}m;$|rq5%{5t zrTB`V*z`R(x8N$M+j$=)z7_y?^uK^UJ2U(z+Ya~{=TbYB7lU8$IlJ0%A>00RicI#r z1?mqj1fIqz$hPFVCSb^Z_t5 zaSLhR7y`G|Ccs!h8v5FJ9r;?f0j28;uxo7+7_Je5go_N^lq-y98^wZC|1C##(@UsN z+k`h0xq$DV`3dqq$U~dpSme4u7!M@^Hi(;B_t$cqcJB`K?#pYCSCNGF z&b~o1Tz>A##$vGdataW*)&?GxYQv}7q_Id*6Pi`54Q9_Y#8F(1^kK?UP_m^VuS z-=}BtP0woLalbSZo!tnN*0XTNC6CMqkh5~OUKv(ijY;jZuuSZGs)53IwuW%Lui(>ohHyG4WXf%B-I z&*O>i%z$y?Mquk#QNDW3Fm8`Vd+Pd_zMqS5ev~Bm@LeD7AGC!I zy^3((SPVQtMZ<&31mFPYJs#aL2R5dpqaej1F#b;pKbOnK?CW@ZL-_*s_q~Ae{Tq0r z!%X~Y>uGrKDAy-g(F5}t9M109hD?Sd@W(m-VXd4}oFgI2 zpZI+qkNvoeC)$h9yT%rjRgw?6DFHNaz7Mh3eJH3Z&I{bW0@|johc-UPK+j$?Jgb|J z-JGJadDJZadujsnAO6DK`%mD8_sP_Ia2VEJbA;EA-GSw`+;wt=BG2l6Ewn#T3Z32; zfH_=#ajZEL#~=NPMO-khvnj!QF5Sci4azuqI*CozJ%pN_S3t2x{xFi`4DMb25Y`;Z zg&u2-;cklaPtTu^%GJG5m{=f==Q7aNku7*XZv$Q(dKJ6<=|Fv}*CDTfGnCN!L$KfX zEQ~nB@uDZfpzC1>*ZU3bS^g|)Ah!a=~8&d@%wC8tS1^+5QrKqo>FcT zm1rn95MB@S1mC+D@H(;&4Ah@S)1wcO{@E_ddd?}7qrw1-aXX~sc^R!%`$}6A7tq?S zcJ|$i0VZq9Ry^)njh>mVL&lAKWV$m3?S4{`H_fu<2LfGzg zf5BOL(EX0PAcX67KfUIJUR>@#^2vvx^%q4n=W`GkSR{=Tsb7oRL`6=(#lr+G-`Cma9ClX`74W)-8ux*9^hP+MRHSmrfRi*ubY| z#&CE^El@hK2+Bk%v-^#20y=^R&xRj^`$P2Lto5g%>RJMAJaiVYBl)n{@&Kar4no^U z;o$1vQ8>x*VzY!L;0_=FbF8JH{8l~;OkE82G|z_a^=>fz{37r(We0Ton8Laaod&Va zDR9YiC0L&x5C1fM1=)$CjQ?yBCe?Ysh2rJR(IP9T>YxnvysiP)FByPIE?Z>yO%dJT z9(TVoLWzsq0N*3bpqFzluBbCY17XgTZQ&D2=}juGTHXjMc`?)txosf6w}f4?LJxel z`9$r_wx666+H{a9TFfdQpyK zVg-=Vhh$dt%q;LsI|y5;akFYIaSVo*VeuwIw1UfL*4$iAja}aC z!`KHZ*EF*ys`EkHrZVuX>?Er%&1H4gegW3$$!M(@mpSM!2Xnn-;F8g6z$(QT z)FSH)bjV{iE{JGB`Q8G!T;mQZoHIa-996(|U4EcpUndYLdW8Hgc2Z74UU>P2NKg=; zfn!gbV5(>c{P^^jlnFx+CV1kD!TeJ7Th@S65J6rhf3>J;mse*IPa4!_GW{*Ug;pR410nctCz!T zT9I(?>>4=#N-cc6&lPHTJK`$`nsJ1RJYN|gyj=1mJ|&p|E-wVo#z>7PEb#@twUgtS znYciy1JZawY#)w#F3n%LN}50LL6N`h@N2Yej|7aHrO11ct;3VpA(@Pq5I~eE!JSzc~C(H&z&y#cMQKi4CW;;A@&fmOfz;qwRD;6n-z&leEjZ`iR&r3MIcn9>&`OADyw?UVdu0m&&cjL_E4%jAnGj_5~LYY|~$qDHm@IZ~r zQ`N435wki#9_JsbH-8SMDL(r2YcsXzxdVFhMFY*}m>H4Zo};9hHaLN3VotAL4@L|l zsT4~tr;(%tO_pzj-!}gM^ES!h8NX)Wokn5UTKOVg<}!^o%iTdYZ>mCTsc4u7l3?zj z09>o)3hl*3&?)~D_}9HYbn;0)_V`f+0uo=MI}uA^x@+a=8~qcxqJ?^y!{WIldEELj;p}Ox5D7p91plg zlXFRQ*CB_l(@>YFwcfnl6+e=11pb)}TVdpj71TM`%_nDY@(F<_6@!om_gr+H$io@I z&NzOsAB{V#Cbw*SgR(U?A_hC7oh#Fkf8HvpKbm7H@@vulytkBn+BPcA-3@y-H!%J( zg3zI|0<1UaMms~i!Fq>&uwP0Go(bWuC1d}R=X_ky>+f|yrl65l8(9pdwk*MYBVT~= z;BI#C$^x{>Hwiq8k^>sOTs9`4o*HR!fa=Cp=;F6X@Fl7ZjeSwZmK9e(=))0U6?v1* zF)+Ye6rND8ug1czb|EU{a5%ns2UFcylAOz+4VeWslRI~gQm;4rQvq`_P*Rs1yyaPm z%4$WiUf~gHDQN;@RIhS-H3D4bA&q=;PLk{R{zH4C{FwYnIpEuzh}tGZfk5>P&<<~* zNXMm6fy*dnZTgHtvbn5@s&qsnE8TNza5PCUuggp?vkTN_d02Ufof#_qR zxZHa=xV_E?Xm3o#b3)=EZs4-0#&H;3Th5c|w!>xOq4+4r*v`Ch5Bld;!1TNNd}sIu zELT$mMqMZHMAigcTz?jybxXn8uPk{gp$e>6XaR2YT#9aO{t33%J)_t%Ke)br_gX{H{*iEDjEBP(p!ENVZyZ$X4QZ<1Z6*St;VGW;Y^BkSHj<(AVnD9LgPI5Dh>C#$!qSFyS zX&(Y3ig(bq@mYN0o-j`dIKW9M9zJ=9>jl}U@&3v)IKjdnH5tr+^UihQ)V+0}UrL*= zK<7io=^+|cxrp_j%k$px_kn~Ln!Lz=QP}F~M*MO^1sr)3k7rlC$JupDp;+rRY_^Jz zW8aAI94dyvw%RjTnyQDiMKjln6T!2tJY!Wq3PDSSI6Nh}6Hlf4!@gX=m;NZrTle`2 z#YT1D+&XJs-}4^4#76<9S>HzX5t~?{3ki&}bq5H$x)$u7E5$EW6U0^n*|7TPTUnm@QPZD`fuQhZa7fGxeahMj5~L;21m_Ow>?i;@GzmT1YRPU=4hQQV z?m#P-_EJHv$0#M^ILcNs0zKa3XSG=H2zv2e3SXH}gF`VgSnOB|dZ#%PemI&+>6g6$ zXYQ%7U*vm{;0ptox4aSr@0vmOaE`j@8K0@yvUV`r*c07QjYYo`V!;k)Ki0!f5E}?f zpyNC_w!)wsBs#xhUjOE5et!`$x+hzuCdymoip#%7KniJl<`~bG+Nn)L_S8V9qL=?T|0-L7djH*SX zaQ#>zI2i2CId8nc$QvCXcW)3(hH?zlS7s2rtOgml3tZl^0B%|@1aIxz2|pgK0Ie&( z0r|4IFyQe^%6mPxM&Q;wnmSu?#F#xcbe%(G|91#&oP7{?n9s-39u!`6i|hKW62PLu zt+c{EOXREB1fmuEfvML|N>P6n8hjxEU0llHc$xsD%K@~#$K_If4g%?V44C`gz(I{$ zXTFPugS+n{bszvwU#2j<>_tH>{H9-b|O(2XMm=Niw) zDaLwOLU zKyQqOw=O)Od{yGn)sv?|VO14Kx4nnBcR#XubqKH8t%=TFH^swD5zf~a#CK;N!+~r5 zLvKb-V^eN#w_QP(S{S(=j8z^6)JJD7FKG$;zqvs-jd&b zS;OHgAy8OP6aYJGsD0or3XR!_3w}4F+j_ZpZuVYWwlo>@i%;S=fqZ;lZ42(6A&+(! z+(HN0OYH6i5J@c$g(dy#;9Zw{;K}p*z~KrXl-eOEe3D~(JM7(wB>-V(Id#FNc5euDmn{03!t zQ9xtlE2xOdfubejVDx|!^e@zdAALu_+4@XS;@O3^V`Gpb;{gv}REEhx3#jM=F*v>W zI@YYp!oxSeqFaxWxa$}-9?ty?RX0b$g~40^#$YzTK3kce=Fx`DB_yD1*c*6jN}YG@ z#y6PwK?E#PI*AQhZ{uLKFL>3aQmpdp9C|NZ4a>f%^WIB{@$6+S;IrS!-2S*EKl+O- zKPbEyZzz8W{(TqWz3^7&>8_6CxN3#y3c&bfU^ljXHi*X)J~vj@-*aXLe@3|^UNz|jRp)8&B*ka)s_!j=Fa0=ARrhmTHEqdPevpqF zcX8~`S%nbG8uF&T@u4B$x@jh5IQ5PJ-%UrHzjK=;)-SjRSAG}dJ+~F%g|4gwQ@*`; zb34~wiLAxR8K;0m@ieq-c>@;)EP=XPtgxrRBtEm!l7D%tB>pU#$?>g3dCRL4QOapljk; zct}x!8*3i{%XQ-6hK?Ulc=QOATRj_Q?#;z#CGTMMn}ayH+YpQU41tp8EigVf4VDz# z25L>lVDn5@99aDb?dnUyqhpyU-0vS%tDX+mHQxhePKR0k#?#zhqAHd#fMjTjBXYdh zfPNT{14@YFHUEo*TQ_fpHt)NsFoR?8ptu|%q3Eb-*74{uT;;|2c&8;{^7TQmq-!7k++)d@F06q`I!08mpC)#B$g*dzUV`dj z+AyG60PPLcLPxf)10z$%;YgW1RxMKk-`19J9j(hCnVapbf8B)S%ib}MH0AN|{Jk)| zO%R8DpNCDx-lO$5UIT*yIl$W_g7se(;oEPuL3LpaB{Q!Ed_L!k7Hz$Nh!380(DgBJ zbfFyX&g3%E<-a+9xFdYfybj&Ul!8k1NthC>h7RSdManYjFi_|{(8@}t-hg)iz12m| zsc+cE>~hv>@mnNWR2eT$NP=5o0(CHbvtQoZph%4OW{ODG27F~%f zyZmwN#XMM`F%x}mZ=qUUcj0sAG+_?Md%S+`6pH!J9E~L0qV(G};H1+yOUEt4i5#E$ zAafH<$&Q0Tt9R(kkw%bwRt5IWUyJW3L72E@81xf!vBH;I$oIbqk{p@_eShzh0;yNQ zgVT{{YJ)dk{wf={+>oIx>z43l)M)Zrf9%4OnfClKKP~<;k^&#*So5OyYV+p2;H1|X z%6#}%fnV=eO1e%zhH7vA!X29*fsIeL;=XgqI78+hDl+Ovl`m}B`x8Yl^;|d19X5d% z?7Z;20A2q0?_F3WE*YK?)8bj53xq{}C$ZJOKR7Vj8MFPjfK>wnH81snT^74h#e)y1 zF*p$^Y7QdB-U+ZwYbkPG(*>N?XM!&Ax$sKHAlPJh2G1L7!b^WmqhRS6=pOzDitXXt z#`h27>{w}jIU&TimezxPhZK2>KQH1faIItwa~1iM`&9VB-Yohw|08_==RI7PD9?G~ zj^T)Z-T2+~PQ3W-Dmd@;armBd%ciJFgFdTKr0M<-oJ;t|1`YQi=_3v}hqn|DUzUNu zsSdV$v4DdT7m;Q_2R>nR9Iu_d2$mf`0|VPO!{Z5Y;InKpE?sJg?Tx?6T1W+T_XcHf|b) zQEQvgA4f5Me6uuvSOBBHQV(I(pEe8O zLzz#kuzrRv(Cz(2{wF2{{8XipPp23xB=u23$s)?OP8^$?-KFGKxxlLZ@j&aC0Z_Xx z1+%@Pz_-uW(Rt=0W#?E(-j4qTGQv2gqb=vi-XjP79u!lfg*MdHq!rYfA{jXJ{V=*8 zQi0ZRj%b^yX!OB<2AID?4b4_e15)Ok|GRMk{x`|8i}=lGWB+P=ib(-0A1?tN)!BfV z--6t}9Y!)4QJllr8jO9+MfZeisHa}@C=;D#)Ux;#m3qqqyZzwiXy+ZE(_T52OekYV zdno8t^_SZ7Hj_R1)dbXBl7X-0a^9fUayE3)7C3uE70r`e0G5bx9g0mmK~CRC@UZ9| z2pX;hK>~@$J^TZ87zjf7HAU#R^BzEz&xePc*217)oXh>GI`rIp3A85&s-wXVC?IP@c{wUTtu^ew-lfmY_ zHHYol`=RxEYZxxb^(i&0;rPH)XoJ?lv8|nOwSWk`a#RkIcfX_3uy}k$^An!8U>tjB z-or8_!u(vHDy(!M2ybTB;v%C&%4*Uc?zGZ`)_1PK^-77bqhuUfgdH7b^Sx(A;YF$f{4%l_zc{rYzt+>oFY7H(W1I+# zF)V|#7MH?x$F9Ra?HTZkPB%Qi{2a_`zYOhVN~y<$Er{F}iax$f!3|q_vEu4QSa@R# ze%cK1rm;QveUAX{uk*xNr;WLc$ZP+)Nqb!^onNRE4k&aQQVizIe2i%iB=Yl+{R2Jp#9bHW4aCCd zXB6S;50Y@+_hgQxEedv6V5L6B;^T@6NLKPc^u+rH zcHL({y=0t_x`iRM&p(T;`$7Q)vo&mb14=!vfRlRnV8?dvDQKE`$1n9>Ac^FuTR`ZCg;MD z>t7=_p|=J+evk*3NicB!rw#B>Ul!B|i-3=%bD-+#cVKj74EyTrDqx&qiYMc&(bi+i zSS{unR?Swxr$cBoT7D9D>1yIYp-dc?&M_2QE5UB%ChDL?HSN%E$4;#kg|u4(?0+l< zCU7zoijaXqwOLU5XDa$7G6fP&`~`X|eNC^!dnk^mL4mocPXY*V{_; zQLYwTdDevf!Zx&H%R1!0TM~}CtKsm77$oXNVQC+(7d;?_=NCT((V2y)L9!5RmM=xt zn;+voM>TY_?GH%%H3SAf&w%b?4^d~(ERg-^Fpia!!^6}JY%a6_Ykd*L%+Dw6Chj`j zy`qivIy@%ImAY`J79$_c2Z=9QQu zp_~oW(1B3QuI*&%MXq3-z*(SK?gESAWl7$=G0Iubj0HAdkoOcf_uPM*bmG-gRnNDh z2mF_mqFX6CSh^aUry3%W;mhb(<7~V&eVkG$;arhF7I563Kcw}dXXrcU`gPp73oo#! zM_cv@qQW{^kj0Bc53#p;Z zIrx6}EcQ-E8w#Coj;X-QDEE^F-2KrL-9NMqdu{fEtx-z2Yl9J1y_H9m$j3pYQ8A!e zg{h^TTnC=(M@vcRAaCY0_G_C0l|{8EoC(6Wt!;7h*BC5*opW_{CSjSak3e>?3;5%E zhIzub!i}reBbU`;;w?=wz8oV_;D2|Q)xt9b1jNO;|Nd_`bpcC(twG!U_8i%1vSaI^ zttQqMODxSzmo8as%Kfo)g{`^3hX3;$caUZOJBOr;IbPg>1^##ce>(^B|A%u}y4cFh zVuh{A|LHjh2t@M!cYJ^U%jN%nj&HTZ)Y9z#!}0s7BN%e|7v`uz6SJWHCsV!$F;y04 znUnsQNxjg)EZHf-qToU1{k}3rzwtTK(*1@haFk=(!6jyV!8hiE;!5W0A#El>>Npb> zb&UC}y@xq!Gew_izeun7Hb#6JQ70E}TTUkbtt8%r%qI56jS<@>-Vl#YITFgUVnnx# zKJkuw{D$UAdNf6ZS#?R0>9$^lGpRi$O%O`@^$|UVu-9Ko+U}s2SU~{B`?p=$J;XK zO_4w8t4>pNfSVNaX|pK#Lvj~sIbu(a&D%t7_<4ccKb%EAXpkgT*IALh%cHG37p)>^ zZFD2Aw;d;Gp;$8VjR3hU$CE62#PR5)lt}dyK~lkS5B=3umMhU+w+d)GNgUIm%lcd+i6mtIU zL^51ElkB`GL0&v@mJ~ZaN<43hr&lX!ThAJABsL&TM%$33-HfI=d=Q`cz55hn-jmAM zz1hON*j7sC?$sgBPd=ba%q=+~PocH65TbSS4VXc&tZBU7nas0NA%i;?lke2cny619 z#O!w~Nrhu^#FFML#;)Zslkz~8sdSjn2$=k6`h6vZ+5bJBxxGT3u^kIxa_?Eu9*uyp zIkJ&(*Pcyg-zCYfGG~b5>N7Bdq#7~IVM`Cgw9f(#XN}*r8A(p;kjLely2}R+osdW(G%;*WA%?nY@p8MQU^#u`F`TVv@B^`aK?Ix z;SA>6?^BG|Da-_~zrqwpKW6MBxpwXKG^WO2F=K1+n*Ql~&-&oqaJnmJ0qG}QK5rxjbnG`*+MX{-7uRMJde`i%2PAIO)N>CyQ<$OY-hE8kvNKGl?Jg#f z8fF$51vBL)JDKVS8<~p-9GFKcCQO{nEJo10v(fvL8mV!-gSaJCOWF!oll1CTvikld zvUZ?=+(XPE+1!w(tzNH*IxzewUx&`NSsqb>RLOd^@iWRcxQY2@EVb8?%T3>h`%PsAjq5<0Uo zh)9Wx^qaMj%(kRhX1kUOGrwArDcs@1%yizun6*hTRm-F31UW@!`j-jQ+H|(*(}`}v zp)i8Tvo;~8)2+!kw@i{gZb7bdog^lM-N^%q6{MY{1o^8*o8cAxW}7y%XuS@z0AOmbwQ8WqNK&RH`G|i zz;^3j11|KPRt2m3-&}|iKO3^*LIkH{%ZqC?(YahzNsznuK{FOL+=V#x8HCbFHb zB2S5ylCw>}5@k!K39+rS8>7FM(x&C#=_hd=wBB$RZRT#vyj5LFPd>dz=g;{^i#&}Y zDm_MsP=71ZQ%RD1?J7g|$tRM54KE1esi%bY(`4dgF`wKWZbPP~C0ei3cwlwQ{aVxY z0VA5{t~|ztWf@QfiAE+eYRQetJcOnY=bJPbr(&eezmUeSWs}Ta#P# z3Ej<2?FaADp1OdUK3&K78cs4dG#@b+51(XC)XZc)u1TlkS3M?934I^}97daBRn8MJ zScn`k`9s)9-*2k$-e%=)caBhQ9w$yDDUbqZvxtGSapWxj%jDee5oG1-OtOH?A-8+z zlQu4?)&;38^nUib^`7rSv>@*cp~(&qTHCHJ6_ei3v?1GuNdKx!c3G#Asbvu)FUyo1 zUH6C>zxtLKy;N=OmN}D<5jslCRhZDLy`MKNmQZL45t}56ExU*$8G*){vB)NBY{*(? z?f~&>sVAAYKbpARszS6|i;}LzfQ)hyCYvWpi6u9-(!W2t5}TYc@t>_I{pbLoyJa*u z@KHZ;X7CBUUm%EftyQI0#|Sb-v!Bqz11V;JU~A!B@c2VL@C7k$5? zonCtV1zjZ6Pv`guGwP3KG7sY`n|h6Ptg>wN=y@82MDY!I63r1Kg(PDvZ~ogyyf>Ob zUM)IBpUO?4tz~A@PR1K)H?y6zsu7Q_(pX2|LM=_($LBKSk-zk(7!SIqV=Ecj8%y?? z=QV^XDUmmJRFT5hHc;Erl&FqL3##$0H`V!k9yLBFPwD7OQ_M06D(Q<1we@8@nKlqh zH2zmlaQkZHjFg-7KiM6O@J3yxrAL|JJzGY9{wz(0d`YH#-aoLe?_R;|k=@6Po(GNh zw;3{WS}L@<(+sAlPm|g8=nd`brN{)+Q}oOE63hjMK&waLD$MnN1&o6C0mfv|nF-?W zVZIsp5YI3KYTg4cUqU+Uc8(bmdv9)#Uq&g=_+j7%`s-v zf(^{KVQJQdpx9>~((GIV1=cx`W!_oLVjo&bvVYfSFkst1=IO(H=CYLw(+5kL)1F5e zwet>)es3frp}vuEd%lFh3w)TB6Q7&DjqA{s_ljxH7v;45^Gv47){+@}*u)r=D6wT9 zh1rJ7y6mR7{jAUFX!gj&9`@k^SI(UNx=H5yA5uPOjIK+1)%_H6xGKf^Q;Hj-OTY$I7aJ94hWF6;3H z*5v#1sf1~26n&)Goi+;;B}5h(H_e){mDut40O_t_NLun*KO~lcVy^W z+OAo^)c-!l9O0|6ikAi0#l82L>6quti@jRR_?QQ?weKo3^{Jyt<NS2{pcAel4-ibbp*D`G6Q4^Q|>a(1eOmm=hh`M>1tP*FLz^^BWoI%M_s;5*6vPv zV$*MWG+TrG?sbKjGoVB~znM;?tyU%FmiZ87muJ&17Rdz3X3(}v7IAA46GpNTGCucX z8GolvM#bnSW8XZFWn_Oa)GuN7Kc|mO{&XE*%vo7xTNLh%r8Ojp6NjNbBUR zpnt9kqlZ)OSSy~FAqyk6ki8~XNlUY0a+P5wiI3hV_ufk;|Ac%X!{)yuTh2TtGvjN> zVBK+|u`rLkz;h*+cD4~BO5&u4f+;CKw%CfkB*Vy-)HMmr%ciXuw&?{S#{3(oY7GCF zM}IarK;+JrCJm2-la{9w$!o!0r1HjP-o3og_sx|b4lL(_@ zQA0a>doWS93d}ZM8J)MEww4`ivz}gXjdn@W6O|f9!N zw8obBCQHJIIqa*)m}f;Ywe)2s^ko}^D{eA#6IvMUMVFYAoFeAWWC5f9fb~8PJcWJAoo9WfhAT!y4o4~%`#FF$8y2Lbz z7Fa?MYAf_eFV{QPM%#bV=DV&idq0*jR^36&6O~S8sxXUL{NM(2?`}KuvuunxeWHZP z{*cOO7^g66TJ4zHdUd9{etnbh+FRD*=Qa?BybMXn%!Txfiih<2>XppjBdh4-u~Ve{ z_&)MO#{tqaGldN6Od=tbNv;VBAnyj9CflZ5$krR%Ne6Xda+SQW^;hv1v`cwO)3a%Y zQ3J1H{$E6oXwOdccUD@EUXyoy#7Dk9=~Pm`+hx5yi_ zUy=N)-Q0n$mdhbKZ&~rD0TPQAAOcneF@j1NZ*$+~?eLKI>@s!oIc;PrUMHg`_Vc(!TufGvq?ujD_)5(ZcgX%?4z}N;_svs?=dx^u z|JahqUo{4umlg1$elNJQ39wc;PH>Mm!J&m;z}}%47+x~S&*6es$}q7G(13?Us;IMj z5pK6!!{m4Ol1w#2>Tky>MM?q%Vy@;^bm3)w}Rch$c4?g zF-pfa-)B;us8W}GbKv6W6;iu>0eo%ngm?7@z?~`v6Gsh6cX%dMId_^wUK(RM7)845 z?h87~^&kVSKS`lX6w9m1qKey%iR-ifm&n^WM)y@_ zp`j^eIhx|nSTU5(9isBrccaPQh1eECk<^~S17;V|I?5f-?2$(=dp-?1_lBmhTk+p* z5qdPBgv2XOCk3^+w3r)(O6w1ym{v8u58&aisjKjwSrs*F>m~vAVKmybl)Weam|1Pr zN*vWp>Hbwg%)WDzFugmCKHOl9cQOmm;{F7FK%g*x?9MPoRI2djei_7E#S(n;&>2gT zqtK&E8I>14qgyod>BdMyWE(@6Hy>V-w{geFk%`j4&g~%mM*E4-{4}U%JmK+wTEJ|2 z1O2cY9J8a~)7Kp&73`VZOJO+O*d5ah!f`>4FD5_9!p3tg*ykp}{~T6`W5REcb87&T z#Om>5>IWRtYe&=WQ&{J=2cMPW7oyj%cpv?6=Rt@8vL}+B@F%knOV9vZNM6k1ztQubj0Yg*4Cpw?x z%*TS&{=f~ZtDzzT1z)P z%DW8_Rc#>pqYhlZbP9G!OyE9V(gYgESP1p+f>#e>;pG=0c)V~GT!@o_IgJ;HLD5t= z>?VV4fu~4w`FRYSzZTVZv{7BDbQ-5~oi?S$Gwqvl$&b&$(Da6dJG&IPB=Rvd`<8%Z zc^>fZ%z+M74|qQqO13JFkq*U6WZ}VBLX%zK&eI6U3Mhs}9c|>J&|x~hXAhp$XvTnl zS1|k5O;o<|5VIEZ@bsA`T$3J;cH7$M%E`Ns*WF0H?8VtLC9`No3d1fsy9^qvo{&$K zyTNB(0Z6Eo5EK0>_S%_NW_EQcCo(veUQS&^l%Jj^U!J?ucd21)*o7g|0BPjFp1fay8el&I@3((u&p9WNCwH%t>9*SN;O$xRC(j?63u1xO6#_+$b);G16s+84 z1vy^b`_DI9;{Opf;_k(|hMm>R)? z`T7*NF*gscmqo+g7cr3OV+Ovw5)!w<6W*WP2LTQ)oh{+XO|PNvQ28hErT9xqKw!9$0OaFkt%LkBph5I4eD76Qb)JV3f$ z$biM`*~DS~53+94C^fA##XH9m@rUX|+!kVst}iNamg!_PQm(?fo|CxpmEaCdsKA;# zHdxhRj%hZjc+2HJ=6G$u8yl+V$DL}RGFnA=`FZ5jdsTc8{h02}N@iq4rNQ<>EIad1 z9+eiZqxZJ9GV955>RH{#2Je5tUe&O}7T1|nDE}QbdA=P*l=xI?qb8mD`xCh^-I@6> zv6i^ltcSHZSuphqff_*{?nSP{CyOJz8#h;y-652gISOO@kx|+Z+t0R#h_%sy5gzn+6aT)%;9Ek7r06k?|(Qw~wTykO= zo)SNcdHP2%RcjtiQguS@MYVK5EtkzXY6cC*9+F>=#o2=yGf=8w68=3Pih>7xJf(;~OKoX*uNdl|`$Olgc|wbR<9?E>%-d-Rq)&DN%x@>o9MweuC%U_!`wbH5jGz@PZSlN zlI(fP5Wsy#w9UH6>kH#Vxb_OD{M3V+K|yfnA3$MNEb&u23Ypuw$qh$Yp6~rGvVLhQ zeK!`)oaO65qwPwfbYvdge!!P}?#f{cRBti8=t3s1{Y7M*zLROE!)RLmT{?cek>-x< zC5vLNv){IiP@k^`$r?Lxe-&PTUQ zM~qZ2pwX4WeET0Gm?bB{->9E}(jUfg^jS7;{~M0dGdob`ObAA=sKu7E;mlLP-j-t_ z;W!|B5KqW1rI%HQc&Z8CY0Kvb`rQ2q75WlM!`;6yW?!xGWA|R18rw_%sa;3&FdcLc zNul|(Z_!O5b{O_onD1f#6K52(VN>GAf0LdS9n2fn|6i5sF|anm7SqO#$7XEHn`Z;8QS6L5WBOFW-{p^+N*>F-DZ zCZebaE7vs;tFyCV#kBL#?z08j{HB7@>N40C)XZ$Y7!844(jd}$k8E>qWF>c|5S3d$ ziN&ERk}4$NE(T=a)A9m(MEfnPUr`BjEC1IczO^3&*$K zh2EqyFk9{mv^WWKzm8?V(i&lS&Xa(ZHCiAuH-olww&24l&(T2RD7uXs;I>?Sy#1Q; zrcfokqm)D*z8WKAS~BF0kt0ccbDgZ?mw@!ti%@ZKb!DEu9n$p+XqZ+O?RcEeS^Ulj zH$~Uc14NPec=jW&nkmGBN4@x6u$yh|vhlbw!By39?acrCo$*pNH_~9Z*{n7(dRbluW`WI>vZh=~rH!S{~ z4|{AM!TCK$!PG4ptdx>Kcc>8D{PbayUIIskZKZNwCt~NFy*Q`Z6@Qt{#KX^(F!shN zntpcz9zQV`-_9_gca0B_pu64VWAsA7b51ce@9&U5if>4n(;E0Ye=P`mT0!P~Kk!R4 z0KeN2aDy`i;)5=L{6HEr@L`CzQ&|D7FHnSY#=qDncUtLQD^(nx7ls|X^XZzOqPV}V znwZA~v7QnFTv#cG?z(6Szudo3DT%%8{Q8Z=vG6dgTqA`JGV3`t7kx?A`T_RQ>9tsQ zKa45(tB8(lDXrX_K~L;k31a4Jptp7rEPbd>VwVVJ?F?sji={ojTo{QC&+{}2r6G8fc8cnVEL=i?3z@io<2%K#Js%k?-;#nZ=41>7!5BfS1)YXx>@k;q-?3dic;&dLuOL>J#$k^jw@%6i41KV@Obn5Z$(LH?!-k zB<#=~r5jAnlftGH*yqmy{Y@s&_WTAY58r?-6Q%gl*?%#8!4ce8)kF=FJZ2ltHuJeOq(_BF9xVj*#Iulrz8L$BKiPE21P9WzdZ9Q@ZMGO*= zlOKaORm||U%X_Ms5rU$32|nMj3FVX=aB%fsx@E|oQ3$rD6H_MP%o;H~ytsmQ%Ks$^ zSS<-tHz~ox%ov!F?g$M+XF%J@1j;5ngiP&8+`qB~;1>vRQM(80|w>3en>9hTmPxsMyE@`icn{M?d071@X$A#wOA|14GAwh-H=_@d{A z5KQ;hqwg-fW0rfLBL~@3;;|r5hl8l@OTBHo^vj9VmUd9$(xK z!7J~4FirIg#-$tMrm!qrA;6l{e~4nt83kCI84EpggrMs9CiY!o6v?@&0m?(ggvKu* zzs48S_<5hmstvaA=Vd4qtUC=-uYI8K!442=REK&8LH}Pr8ABu9u%{Ek(7EIwmJL3{ zyY{8{BA^AO6PhtnD;@8}iSgCOjQCk+<@loo)A*smbNF!~GW<6=!}!ACHjWIQ!pTQ& zp^Z)n4o4?o?`Sj@9{JAX%yz>q4c&CUi6;H;nkX5z+J?;JDZn{y32u&&WR;0Dyejz0 z9PgEaZ|A4MXojJnR{tUN?M2WL1lC$O9_6V?T!Qq~^5mJ8B>Jp9hHlG~@#{hXG1wMAgR<@65UF||es5aH)lV_z{yCv5Q1qHOu5Spodc?uughymz7y<22 zFMu7t$$y7>$%D|N)bsFbD&@Zq16AH*$EnGJj9^86`mITPj=2iI{`?>8*{aBYqan^e zIJyblJ&KqHIeg<{)l)*X5Z|Ix*Mi>~NiK^0iR9dKn z&T+^huCWZvQ3(Uz;cO^xS_8)}bkjYa;wX`8OKaI(w8c7{75_4uY;N};51l7K)D8jX zbLj?s-Ry&p=SJgOjVN@C_=0b}tI@Kz3q1v%uH7FB(M-k$C+;{$=ZBwV&uzVG`f_+X zdw;ho4O0t7y{pcwu5B@WbK)WSxM(vO6mVYs`jA;G)nJ1`@Z!QasJuJoLKOXxD48YzPbba z{Yb&AFBY;RhCuX4HCzk21Hx}kf>dS}H2K|u>Qj|)DBvH_P^~B3PS@z!4U6&D z(h_5inc%wvc{EPuE8VSRg?BV|p-=d93=$K@E9?PWea{cO3)j*Vm0DV%{D+tw&jmZD zaJV2R_+LH=hBapL@WFKhG_)TFxmpjnpFRhCZ`>s5Cxk$6{3TOUkj@ECi=v8k(Rl3R zVO;Cfh$*t)Q0$fxR_t*@t=ZeK`Q9VSv{>QR!k<*7)B?sFJ3ygL6@D(b2`9N1V7o#s z>{QBunOEMx?uKnJHrEy+o+$!{m8LJMgXxa0(}cS-hN`&4;MoT+krt$4T+Rd({u@G< zx4W`=%S13^VGKE@(GExUzlBZS-{EAq2=|ty7}s`JGCX|t4t}{VhT`Z@$UGkl9rEtr zTbD@cqLz`pax)=s=LvywW&`~^V;(M{k+}MV6bbAZV{0xCQ|IrwROkH%s`%emDt+nz z&9Dii^$x3Xz0o4P9=#B^H{POuN_SxYlM?z+&KwVmw&8@rAk-_%!=0B3(XY20S9f^g zorzKSF+vpI*-yjosc}4?$>}6kdXPQWokVlgqUp@)dfM^hF1_+No6MW|jZ7I;C0ihz zIJ%Uup+pCQ{3@Wj^CZM^E`fao4Cdk7;Xn6B)6*3ZEZl!b{I`sK4Y4yNx;^`FlL1 z^i{wZw+>>KdP2|v8wg+04juv=N9NIa*sPlkrQMgn>B%DaX}TURZpeiuP=;U`4`!27 z2zXbklNNhlxU=OVtX^ydmxgLUB(oGw=-q}ji?)H)*%U}AFN3-lA@KUO5wV*y6WrwM znX&i1`j6E@g>q+&Z&aJ_dA4bfbO zDGJjC@9ATHs$V8fPd1V*sW*9N%A}zG@ekg$%L0s2;v=K;i{SN-uc&&_er)qgNAJre z=&f`K+jecn>N%!3UQkE1jpt&JMhO+hMfmUTLiBg(BzNL!8IyM`Ets~Hoih+ZGyW|_ zpS{A^Hy(jEJ~hyWK_{HDWInoz@1=Hg>xu49Q!1KeO49e7F$+Cp5A7#g$PSb9M*8`nuTW9z3V+McwCUHNt)HFN6anXc?2 zHV3D`}NvXE0F-x-LR_Dfpo?X3AAwv*4R^WMD6425oU#`89Fa2&a ze@(47zvPM=-!ngi@AGy)Uu+tOZ)vK=*PSqpzrp7&nmT-^BLNfhiR!&9T6Hf6-F7~}eeu)zLv%hr z;-C~?W@IkEWL6RCeVT>#n|EQX_X<4GpaT(`D81jLgkQ2J;@uP}x>>Z4*j<-|xBU!g z)s{hV?H*Y2DilKZq(PvzHPm)`!XmCC^e82Ru-`b5S*!><7sry$OP*9=%XwC4k}ik0 z*^?ODFd?5`WW!baW5g$16N2xpfx(N}yn_vS&L^cLcOH;G+;QVL%M++qx7^ibE6Z|K;hE?O_Fg)TZq^y#x>bPX$q|8{xNJ#>($ zh3JF(_&T;|bsDYy@sC*Kn9(Y2Bb-t(9e2E4j_>9Kp+&C+ic}py8Se(%DVd7T$`dg% zGD^@lD`B3VVAE%<9ln~JL)r0(wD#$7-jl{XjFa<7rN7@y*dg*C#I806&qXJpG^Gl} z!Y6Xm>aW78-}gXY{x?jl4T7qzMbJ@n6Dl^0g38_~&~Kd#GWt)U;#@Hp9J>W3+n$2h z(keKX+Xul#kHM=n0%TVR28Ad^Zlj(occ8WurW|?#4faahjz49v%wjM6?J)**L62G< zmk9964mMOqLuL0pM#20OC#tlQm`ERkCpwnk?S6o0*MFv__hzC{Y#RP!l<@WF1=N-c z$Gp=X^kGOf+3~5B6P?OI%O(2E5d#_Ar{INkS8ZvB^(2fPs-r#$qipT=0QRi(I+AB) z!aFIHPiMr=Vvb};k*it@X;+pibu;gvH~KQ@`1nTJT`Sb& z_7HRA!e&(|P)R!jhqE)G#iRl@Pss*_mReZ6tO8amF9y+tQ-F8z2B|udL>K6^(5$>} zcD1GhFuO;)5{`3-r#1I?ppF)2z|GzydGpuR^k)i`s_z=(U63pMOm> zYgQC>KEH(c{!oGykAum^pU-(SRDaS@lli#!?@IJ7EXLnAA0cn33+d@TJi;5qhFeF` zW0flw<=QZ_oh#_xAu;^sKF-K!?-TGZ_vqV-7|tZ)3navWR;~|u$F6mbXNuET2xgmL z+QYp|tMgUi)7VRLF>nezelY{C6;1}}r)IF^%S<@<>jzQWY6G%!C&G*3B+`;AMYnDS z=4sD%`smmsV*6U4OLLrp!BtyPf1*F$nZn?S7#$o7_)6PyPO&S$KjED8SwU|ezet9A zRGP<@ z$POC!AeGv^SH=O~7M7=5$0QG^lT<}77~Ja$@9Q#P#s+6NbGU{6SvrjFx#^hp8gZST z5^5Yw!6O6bv0~Cw#CxwWIb;XMjilqV30o;oBbKh0m`P)I9>)pG>;(GVNc5Su90P8z z!2wXgLmndZ*0O%qF?}Vik@&=3YZ)g|jpEokFblsf$fYNspsiDmFD7+_9 zhW7F4*gpCY?fm$YT|SsJ{gGY07RF|=1T6C6Wz7f@ohu~=i1#Ax|6v^ zjgt(gk>6Q&Z6^R7+MlPWDGpvk#esFOwe zNwS>gi*xY6LxwGKh$7B1$H3|RBjT0{@ZUpCkUpRS7Ji0kuD1X!9ZWE0!Wmp{B8&$X zyr5Q(ks`qW-7o8#%QJ0O|y@tRkVEcGDexC<2j410^L~zYWE}Vd~+9%DwpH2x=~tr zU>de0=HW=J9xjvkLfrpUGcVRIq&|8!SQ{vZrDY|QPMS-@=66%auqr%}FVBzDZ^U3h zjo4W~pE<94jA#{e!TjGJvPQ6Y4jexXiGRNkWp98B1ux*eZ~?R{l)`NP zASf(Xg208XL(y!>vK91bj0l~%DH#htWMi24b{zS>6CLHB zqTS)i{ONgbus^UHC*Mp!TmNx<-`|XT`#<2O53yLbA{meBoxn-s19*SQ5LU;PVoO*H z{<_+XA(P6`;p5=#>P3joNuI=bQuW zCmSIzHjzAbR>9?NH`(qPb%aW45a|!YWQ=?z?dj9uhgm#$i!OoF#!3)ZK8tvUb69y! zBb~x<(edAUyks>GwHY1!{!$*V6kFobQ$s}TmOnK1wL{|J9*8Y!f#pl%LCPzMWL|s7 z8t*EhV%q!gkcB-;&Z;9umqkcRcN}Qxm%y=8;@qM)Ezmbjf%~gLj%)op0z55G!=)3N z@T$%YjD&ZA_tI>L7_b1B*~-9Oum=iNk|As8B7AeR1ykj0XuXpRRpp7)O1zC~O_0YU zX(wonLmKPQw2ZcHZ)fJ;xJLRvnUg`KeDdMV1W4)LNUyIsiO~=8u>4X8rnd4>h4Yj) zUrxXgbr&38se|%M_s~+WZMa6;nAR)xkTi*-ih2zMh!9n%?; zBVWuarRquSfIRWgSD+n6YsnWBhcWAkL~fA=b76V|*=-g|W~zD+bKPxV{n`_@sawI| z?Ngwyod&lH_>euf2>h}(LW%%CFp)Y9y5IFdy{(7*lz+|{=Ec&&v_V#-*oW!VyieXH ztOUyyrev(AgD`R7~N;arjET8hJB(Fw=b-Mm?hF+7p0BMegBky-d9O+ZU~+B5~CdAM}&6!`<`h@uGD! zZq$sSu7Z_zy$?GG`H;x8J71w;-tNqjWl@~u3&pfXcL(LkXOWY(>GYqX1#jSC3FrBm zA#!Wi9(W{t8j`*{!PO6EU}OJr@Ht-x&qS8MKhql^+#TX948S{7b7v ztg(5nK))>Sj$P{lu*uBlbOcQXmFK3Qyt|A4s`P~#sD znJ0ay^`ATP?uFy&nXy=znM1FAl?40ZEA*{RD06J_2cq2)MLd_{Y&mC0j4i&9)&8IXgdZ~X79;(p76h#*rGFP0we!m(*tK{H8bCTKSKxwMEZVEG%Xws8i^Qj%z znhA3+;qdS2vKFd)scT>Zb^QbQ%3~{LehS3DhSzbvV>%Lvw>TF00-Gf-;1FB z2%A*PSl^Bb0!&NUtYRpJa2#uyh^>o3BRhiJQe6ddp=(IT#ZWlytqt}rJK(;A9juhK zf#EAlVCS*JU~V7{8XZgFM^+H%y6y!fPa)`;F`fKb=?o9@v_aes$Yb+gM0>gt5I1Rf zy8bh{(vZqTuK!ql+pz#MZ=;}G4 zWVwGEYp9z}$DKs6w8EXU=I1Ob*}sg=tr{Txes8Ht#Q^!q?;?5^qQROU3mb%tK|wr& z*oun6zgIWNl@-&;n$ya#ao$(@Rx5{mG__`u`fTAv%nHz28b^Zuj1y1a>&%l!^GUcs z8~MC0mlPkAfQIGyL^J&it^AlkCwwZz#=HoeAafLHo)S*a-i9Qw0gtPPp{r#gHcKSq ze!=;;&5y%JO0t;tbRN3bCS%z5S5#BHhqe?Oqw(u8I^&WgMtT|GH6<~ukGH`XjrCX) zFq>k>EoOPLpx$V#14)lbu>Y?%>}imLhFeGA{D=o^eWC-i&&Wb~x+d7mUxNE+1!=9j z;8=wY_-|Va7P&oS=lBGAL+BG36P&G2bpdhTZv%Cf?yymyXImf_1{);PpkL|+*u2Sr z{E5R5bWxvpwLYUx+7n>&-pA~IejnX$sg7rSTu`%h9Zu4_hBAJu@ZZ%ElpS=T?xDBo z{O$eZK0+nZ*VQ-QhlvYdw_FMuqJvISuO**^u^l|*r!ogSP_tSND)^uL@X&e$cjBChJ zI^l8^uHfIHC(ccyMY>65HglZ8+4~}6eP9by;-?0(!e!vSA{Fwrih%x&0Fg&WAo-6d zH{rCv|Eb?gzTJu;dsZ$crG4AU=(}swzek8lExbopv|VH>+1;GcIrmuq{6J#=ItM(5 zib2m(;1ALSQ1yQ$z<&h$t;3bLOJ_~sUOA@2jXl`|D<@rpEf!L6U2-Qhq|JkB)(wIW ze<0R3ci`8puh2@r4xO4M`T6_4V&u`M`0qqpP0oa*ej~`TwvD~POv%h{7 z89(Vy#V%-q;_vGuzh0m#D=a29PWOr2;f3UIWCInp4Q4;2H`Dd)hpCO*1y@b zjNobE#6nZ_@>IaU)+ihm@IKW>|DoGSX}a;~JUkM;7C&4Jp`vnv*|~5wqc?YiQ*zvy zOzyu!*!5YQ!MpeAL1SeY(OXU*BnbSs^7)wJ8%1sPheNwZ?W;HHch}r^<-N(VhDwl-)=gTtFzOe&|Pgz(3_$ccf+b=dBt%BmI z?v8Lg^)&q8BV4i{mi|v$(KRe=)3# z-`P!^FejTzvF&^4nUfu6-JDP=lcqsMI)fS4g;AvD;uOK`JqyZ>(?DXUBsVv#8`gJ} zf~7?+JXMwEn&>HTZ%-_Q&PQ({&)%Fnb%_=??vn(!du#?*>c#1pdIY71-4n5!`{!8u;BJp@V*)gcNf29`q`y$ z$ownY5Sv0hMdWcf&={MSoT47{^=a?}UE1IkM#W2~(%2YR=4*0aWotn($vS_Jr2Oh6 z!ecLZY0E6=CGin*Z+sHzangW=F?pzBlZ;d2?&J5hPjFhz4O9r4$xnPB&%gF>D!=H; z7yK|@kB$9~7@Fvd*G{D1H<=V{2y4LqJe;uJvYa;b8B>|X*^GU+1QzXUqw!%|G5?7& zZj<{>hhJyVcxy4tezF8_%UwfJ`zo9#orJO~nZ(>)7Ht;qL*Ji$q*$Eb$_@AFrTFJ` zpW0l|`4<7ljtqi`Q8(OnYKQre|Dd_D0mRKFa7E%8;CiA5NCZ!Z3mW2ZYexl#3U+H3K$wve$k=ZJ-MCViJKYU#-4G|opJ{;2 zrCg?4SrHFCmL<`!m`mOYR2@kngYJ#t$_pChJjELlL6-6+F^Z&JlEod zDR+ge0hcSG$DLen#4Z16!1c8f<7PN$a5Jw?6XYc<avKYUeDMU%F+vvP(1c#pW;fk}u{C29scmHm{?|$LP zA9}ThpLAnAfB7Ha`}{EB7k^#A@0+g6uQZv+FMVN2yo!-N{y#?+jlI%R$BZE7_|SO2GUu)3$jTC}0E0KG#cn?497v$#ybKt6;7C zBe=jj1S#&PVa|)o&>7qYBNM)mO)?9pwVf2i+M1#Dx zos5ePEXMxF$7y%5K3Sq=iWbF_ut++Sr~G|B8sxYkz#^R3rh`#Fgb3H4i7R(Fa83NlgYrLw>1@Zllev{K-|4t<3}HY4a~5`^#bw&Kaf30PWmMsPoa z@sdX;)tO$6eW!P!md0XQ*0T#74=F+IiTk8z(>mCFvkD1|3x1>st5Jl1dK;@H7(6nSwD zL;8hqe%BK6$!iHL$bU&w{J!%XL}TfOh#{(|uoN4W1-hCu5%8oX7Y?6!2m8lk!FpvS zjQ2UfHy3TVtl$MXp4woVVE|(B+E8owmx%S;gfy8=upq$_dS$jj`Dy`2bNMm-J!y>I z5&uSe%2hFMmN2U11)_iUDm=P-3QDgl#`O^%82oP`@;+(sr4^&E1iec-Osi^wg#D2)|35`g%<_h}vknlCy=ml{ z_7P%fCyHJRQEnz#>8k1<VI&h zaR!wV3TOHT1_a)hc#OPnBGA4VgxyUW*Gaf=E-`Not9}gJOm_-kd z&cx2>DHt%r0!4p|;six;jB4JF>o@|QWmge>wYG?rK692fPg?=uHIw1%=5=sbTpunh zng}~ulHi2x2I#3Z2i1)m$-VfCRH6r&MGwxC^Zz{{UBUo-%g^+?SrfIBHb+DIL3&b< z_rmC3Uu#`U=Qda&+60aJx?o$)aS(46 zhv*eL>_E9MDIgV$lh{gXxARw}%B3%)e6b1ydF1jA9}S148T-MuEew8X50N?Mze$HYIZ7Vxjh};tk?~06O+lk|3tvFMvCd7 zwxsU`8024l09K#<>#~goC{Igdb7p&C8?{jjnIruD!EAPYZNwL%*`5R^L*3p}5J9xh| zKC)rmhiLH4N{+4d+e-J3ouuxAKkw6dS)5;;LuH&RX=;5WF+HP1*T1!45_i33+k_kG z2Ip7Q^q4+Y8MWioOSc66oJ1_pjlqus-;uYRFUC2!V%v0YPPlw5S^R-kYTF9@h7!i$ zuDc7g=sWBZc$8GLrr_^;2hnOt9=1h4$JFO#0&Q3ZE)(!N>MvdBeydn&oxK3RZ9b3R z(+;7DNF$B4*oPvMkiKs{iv7oapxE^&bXdL;pKAG{=9)5k^GyK`TZ!@w6MC`ckSza5 zej;AzlI2fv6yyIq)`PFxB=}yx&H2;n4fs*A2K?nRQ~8EQTKpf?)A@70HRHX1%KZA> zgZQ~xmj5vGC(f+;h#8MAphN0U9DgLtKfkCGosaP`e&RmVKYfpV?apVv*dC&X4Q@c9 zeGJ5166k4dB4GDb0OjgBVB@2~(94S0nbpw%&x5q$elZxDEC!_(1AHd6kj@?!!jOst zBq4k8MUK}EP_!@6HzNFgHyK61@@%9hds?OU9CwJDv?C{xK-;GLK*4CNYcN_>2-){%gVg>pI{fY$?pUx67lnD@`Lls2aiR$O1^y29aWamNB(TXh+i{tC zJgv9!#yNW%vAJjtUi~S9o-TPM()*d<8|UJb^OGIWo;E?s@(%>#!e-bd-CXtCj^Le!ltJo5HX=0@-x4} znt6)c14nPdd4FB7cb^O2V^6?AxeS=N-46C+D4cDY1)X#neE6UU--2wwO=&ayJMIFv zHBZ57;c}wotOqLX{XAjr84~@gidk$Ti88JXJ_(;lT_Pvpci-8V@YVx8Yg{q7S{)a$ zi%`8k9TW0mal~pns%G1uam@i7<;38tyfd^Z=qxt5mf}c`F)kADa>pgE;S9YDO#WSr zaeg{zbKxkCbOq9u&_L30NEbH0{zl5rS&_ldvmtkaH>9lb16JTGxD>^QHCh6k;X@nQ z%D)Ut^|D}&P79ekGY@>bi(q>8O`<*PDD>LJz|XP)_@X25SY5piEz*y{n`;N1!aQi- z{g8zC+$K+2YUso>FXp55VNwxz2_m%v8DMut*PlL0rG5nTJ1}UAba5uvY4t{b3y>1zSr?U=TZP-ZtX6_>f z33+74@n`G;H(PwQ)E+49hx%QsaQnlVcrN!Gp0rlRsWrXy^4@Xgu&p<> ze-VtYzFQ*mQwd$vmSJo4PpU4l9<<{X;kj8bv`_H|6~P(*`qT{HUYrA?&#mzLC0F46 z$R)=kCouebnHZk51j}a2;FojpjKri3G~Y~&&iHMLa#NOK$LmR$^j92x8V-``P8GJ| z=6iOjQ4ViaUKIVHlR(lBi^GPxD6&=JFPSz}MxrkYa@hVF!J*a_;C0Oa?%S(_>VNHI z1#5us1-|)|#Xo`|3j7~V}`aeVG;m*|?hH-~<*OuI?Eop$?n5 z^uu-_p)L(Shz7I^-trA52_!yrqxH^|F?95OLv&1zLJzSiIQn9tK=3Xn{jCQ{=gr;$pI(Io4kj{|qUj?v)tKP3@sdO9a-R3ya4$+@R^F6!7Llfm44Qsi_5VYMf49HH#8=?HpKeHVSlhU514+o$&Gz4FI|I(37Z*kj15{R3COUxLJne^6om6>^LVK&M0jr0y8NnEd@vP;UtZ zk_$=9IV|yz#EG3(@xP1|EZK4sPkDXCT-g+i zHGPb|Q<8AZm0GlavljEuIiqpseq{Ge$G1d=`Wl(heA8OS?tv0B@>x9H+VGb7D$z;b zJxHh8x_Wf=Mnmi~O`y@aJ1{J;j+=3NHr_f(@X-Q-im#oScl9P*?W=L5P(1{j#_gi} zJ1gO`<3JYK`|T!4U1Lr* zsvm^B5y`~Wj)yCUVnIta8)mvJg3;cJpt#ix?#4BcyN6an=Z&>+J$EfgsYnnzW(ql9 zX%3w&o-naE7uJlN0J{1L@F~5Ak-ayOTd#JWGhDl#?y(9d-WBUf@8}U!EIo~RbT5+z z+w$D=r4P8ykm#ns>xzQc;XPexFq7&9#pCZ3cT`Z&#;WsL$c38X)v-;K>ffVfuC-Q+ zKo$G1sF1LKr|FgvadeDCDVbr|L-bV^fS6$tNY+-uve0%QNj&uBcFkNNU3MA*IfU^emTw z$;qX#RoIzK@JxXAw6iep<7_Zp@|)Cs_{1|BVKl`*Puy@) zvO2nvHFQSm7_^=yOCK$ALpOmciZRVx>6^TEQ@`K1#$L7aM%)@yOyNSyx071ry?pe*)tQ*z9V9% zONrHzqeN1|oc8b3#_3&asorWSf}0m`r|*5YzHr?Xe3j-wVsAAybl(cT0-JF3loa^> zrJLLuxXI)kea-CsUC6X$RCD`6ii!M}K~gaBDD!^29ZgaV<^t}W<4#VhXwKL*i>N)Z zfkiVMA;m5iV)muOt)z7VXLJ`)b1a0N!)?O3l!ZC7`bY(dp-B^snECBd)~2J^Ky6$S z9QiHmU%WCvFQ6OReg?qT375dH^B#QpE5$}7+!X4m9Qbfhia0+NeA3E>&=Psj+NSIl zb^G>_cq;#B%CV@{~S*ZH*tE zhT<+qL25c85~ut*PJdh+OAP*8pi>4Vai)Yk4vbJCwd*)W%E*%0d4yqhY*BzvE)m|9 z*ua9DrEu{_9{3$E1q+8`aN+7t*m3y`Z0qj?Y99`E?n}XY-dY$nH3mGdroquY3*cGE zBIt4!^pv9tVT}7l$g*pMMN+}AB4#n1RgHn98}VfHe;=E@oUxhBe$uE`4-~!xp|-Wa z`Pr(4VW+f-Zr49%L;77hBRrSh94O_oZzM1mY_C(9(v@g2PMk^HIKa$kQXz|tWoY=^ z;pXtElz5p{GotR>m^pKssk!-m8oB5Z_aLT&Dn40+r${!9OHjg(o%`_pN*{a|5reH( zwe;gNGgR<1#M4v0)7Hs=lin4Pt~sXklFJl&^7<`OvT_rwK9LTKr+1TWNn#M`f0Nve zs3y5xN}%X?ne09~8o&80K+B3UT=^ye>z8HWc)3Uvd3FM&KODxZrVH?Q!BI5-+KyE< zi70vZ7@ePENAJ5U(|21&k+|RlnDE#RhN5_KF*uGq{qmij`fduGWG4v{-v-(5-!T)b zCqh+52D~_Pj!b;@hsJHyL+^~mF!a8e_-SWZNAx&RYZXm$t?F3AOQlJ$A!r zcGaiTA3F$lur&81*c}`qz2FYF4PIR8hQMEO5c}U$xUL!x4sj{)j2R(tH^d;UuaR_( zmmzQbW#QMiNia9cj2znSO-wdg6AgVM^lePSFZ-8aqI)o^{wc>-x2IvNMlDtbAHk3% zdr;Qo2e-e)j{aWON?Z3wSbt0Y$}PTHOeS6tBf&)_+~MeK$}dQx%f>IE{=rhD*Wn+n zP;+Bc?B5g1pQWI$6#_n!4wAlR3&`rMg8SL~V9d@YAREPCi^^%J(?1Odeq9EkB?c{p zGT5zH0Oxz3!2$7N$gdQ7ix0!(FV=Ig=MkAcd<5TKDa7*mm+?#W0W6t(2{RYwqf6aN z4D=eJ`ieVffo=fCA5Eg^x7G3SrCR*HuLL)?pTtA|4r1J`R=ny`g%UHnF_)gj1=WMN z#Q7o)t2dza#6r|EwZRGLJJ4s3u&a&IMUB=Ibo;}LWY4E<=o1rdZK|31hy45Weo0}HaLyo#{(L*!+s5I?(bAD5B47$tHS;@Fe}TupZx zS-E%~&0RSj-ijEIJ1I%9=kPP;nzk`H={O3;s3pKz(F|yHvmwd8Swz#CfvTwma7Q5% zHqE*XMn`f%Y+N&JTQLWGDyv{;LkJACo5MHl{V?caMrvf8KqGmGsB}CfKX2Tyt`Xfz z3qDHVyh3YGQ`dsq-v{jDj=r~x3mEK#>#r`B_j~k}4+uKmy zQ}DF3&%{>G`6#udS=c|>qk-RdDwu-lh52PD;i!sl9IY`>R2t>hCE~)xfky@E}yG|CpHl&$8NiZyMjSKUGAfQD<&=($r$g?KkHZGFb zFE@q}v5LT0z2qYHWDv*Q(y$Jqfic$!4S~v;*XAh9f zNP+O+`*3RMLogou9)fh6U~Xj-WCUolOICb`C*Qh2w_1wb_fU-8lk^fs)lk?wZzQYZ zbPb$xhas?BlkJn2Wr@T;P+w?8izfC68n26_BIZ8Hu`4IH@5`dqmdd*>Dg!IR6}wT{z-Bc&gz5ac)3P8nDUlRduxr8zBSU1MH?6! z?{!rD>VEQ>e?cOzj3LqkqR=|c1|AEHz<(uiAU(kV)b0L|)_yyV4@1O<`Uzj3&or^bb%cc{!{;nMoAQ@ytbYgO@Gf^l8mCOR}V91 zZHj2=mTc-1o<`5!JitwTZG}M{Bk+>G0ele{a_&1fLH?9?q~{R}gGI66x#$}+>Gx#D zpudM+cz2L$6|_*Do;x(|h%Ak0QlJw=JE`{TCB#g{ofM_kkZ~`?Al)PiyxR7_*!d~I zi!u=Q0^mX{3$rZ+%_X~?#NYVA+0ND>H7k~qz2~)Y_cJN#(LYUKfw&>LXh2+yYH9eN z16C7H9;I8{3yC#%g-qI!-aKhN(r?o@(sy!WX}0l0>fgN^->vh&aUIX8g;ON`ym1Vg z#s8#+69dVI_z}YVwVxcn^^925Z-e@uA+UaZIB2-Il6Wmgcy~mYpDkSj8#-&@)Ic1} zttl25F5k)Q>@L!sHxhEn^2nA8QZU^shnTJr>h>F3sk+=8Y83sQwEfbC{U&$FJpENr zYG4n+TXNu<%XNS+$-qt$c7B!P;FR19Xg=meii(sVCvzTz*Orm>)83LvHpfXBlrR=y zI@ICxLE08_5QpswurjL_dowTKFJb?c`&1guf?TnBuLC~x55gw}y`1y9JXbkK!=8&iZAg zFK3ke%=(CyNYgUoH2U1zK$z#6V^+#Z+_dRCtreYvpA2quGryIR{k`wFt_2?WR@?>$ zm!1;#W2I-CA69t2Lx#(fiY;(R6|Px^UqsGAA;Lb3A>9R5j}pnRYRF z{icHWMoW{H4spRRk%AMx=A)h=QmyZ;G&6ZDHr<^<#fmifZE_4hu3LMq-rPK=Zysl3;af6IrqJC1aDaf;d0l1aG8fz@(`onRaav zM!7YbStm6O=8af|jlC(jexEVb*}GTp_(w2SrqN9C&1h2ozKDrPc*2~RqW}+%&8Ajf zQlKNc8f?x@Cl5w>!Lz&_Wc$Q@L|wfEew(a?)1QigXbUX-i7g~@zYi&H3Z|Cg>0B@? zk712*cyDb7KAW$>FErNUErRZ$!|9Q{ucQ=ztautfBe4WG%^HO%RsnQ>W*odqPK9pO zUtr65@1w{(H%9WZA+h-o zP1w+L#J*_(c%JHop?*VF-Pee%RT$4!?Dk|Gw3o5IYZY0)gYj@vwhWZ4?4a-lrH)ar znUUKoIrHxcT-vS;Xe6=&vlrRn#5Jk-KB^YSo$IE9zkBGGX|bq#X$^*IXOkc$hE}~D zF@wAm=5bxg$T>>li<8+XUz3K)g7;nGi3T{VctpmBNic(W_kzI5MZ%8EL)#~o%uO%H+bf*Zu zG8X0=125@{rzJGR_yUQT<4dGmV!^050B&qPK-Rp;BprwQ$jd>32WBrvp+~}lze}<8 z*;CxvDzKGT_0Rxe$0_f9g3{pprd@r>^hMkOK{vbzd`2a~1ZEV(Up^0`l)6B)_!A6X z{{_1j4#T~h$KZqKUsz!$2hGBr5$e@NBGfX;#<~BR|Grq<6np3-KX0G zf+uk|cNH_+_N;<5$5?W>H-&ns5u6ICTkI`5+uGiz==SOi>zx>*+> z>eCnqo0CYcZ_q?r<90e-dKN~1bweBTaEhgNI9R?8FTXUxj4yw|G4X?mmNU z?g!EPc01~bSK`qR&bWMz4QjmLX+XO?-U!)-W3Rbm?v^x4SLSdpKC02qw%PQ-xCr=T zyA=ZB2bmfLeWIh=#rlU=)k=d|U!;-_c2zW+`BfaWd|( z1oB$@933lti7wT1=3ZZaNqV-sJGTSy!zM= zXONlX{0a-EV{jF{Hck(hYf0hVZ?nz;f@cxh^k|{H( z_oR>1F})biPo9LwFPEeD2NRs~(t^gDeVXy%PA((*bObfkiX`9f%MnIf8%8{z0(EU0 z>ErF2xuSqZL5(@U#uRlEJ3@#of*Csz8Uk67(oojt_&K`ja>a`dgl62sf*-26z+dw`J z9s{@NL~vqns{~usrv6bsUCJD)1VbRf26>~*Xto??@AD{NG8R*OGx)! zLlB)k5wuTlhC6+caA4ma7@Bm@>ds0DX6_sh?#@#QqR~4K&4hk4EwvqAjdH*_VyAGm ze+tg<&cZ_o=)QrX73+*sBaSG0OpidsT zDdXjYLMoG;O)B^uB&cHnc-60j&0Hy6^>~16I_V7B%Y|>1yA>wSwPH7>DYHLL>$1v@ z%h)5S25hmE5xdH!7Shh^!p>XEU_p8wO^h2PQQP%l+}ZuGa^@*`-suVPTN)swe1a+-rOrDt@IGfr-%{paTsqH|$yc9q~(DFov-@`8}5N&Vv z;q?<0c$L|QORgl)H;(<(ZgehPV<3kocKo5s*!BbX;j9n+%t)ZGDaEflIV^u5gVJs{=mEzxZi4ehQu#v)Q%mnMuLBO#HHAj> z>YY-uSzLx*Gf={n5s{d@)P$d!D9Mkj)8eQ4+VMrG&wEO(=4Ys?^S#|;dF}Ag{MS{w zyxD9Oe%|-Vyq(5+zB?fdnT%Z2%UXgKkKWRDA0D-yuffV|GN`?E2^QG)(u>i*X_tEz z_@f6L&6j~W<_JE54|Q-s6e!OA3iAffz()5tQ26H#T>UYSmh^(U;m4#pLILJ(&marR z&osyXH$X4c?#H2v!kok?7R|?X)7xi%laQrGu)%T`Ty|Xuexn=UbJuJ5E->7D3mm~L zbSxb86BsZ1oym@cWwgB55Q^uUz`ZsT5~>nIrW~FEYmU5Z@(-z}J*VzqVD2|G@f^#` zi7WG;B7b4wcEpf^c2tv&!n&vb=({L^7w~GB_6eNC;MTS6onaT_t79&$a+K4I!91trh#AB zQ#w=dh(_w<(ygKq^vlh3nsiMEr|Zb0#a{)KJ-H1>DguTlykZuwi>3Mj2RPgAmyArq zPtuV*i}^3rl~#E^V|v<;!y-E)5--+88(fh(PtT=Gp2;Hegfn|G*bP`tT{4@%b6Dzn+YjtqR zlGD^;bu~9{Wdu2^W+>=ZMexsq-KbyM#!WBEq8@eL+@ZReuqvvS8An9OnJne;s%RpcjG6QwabXOg^4xWPyfi}sw?4xLL0nge}tO; z3BpaqeOR+J1b0#djcRP3l*y3gEZO|I zl*|g)1zWehBCSa^r2YX%S}MeVip?hlAAXWW$qK-Touyp^M;XR8mkOFq#_LNR_cX;r za88Z_XL#=gNjJ$OD?Z7=%#{>chF3u6m=5T7%ZD8{%}_ck07z>d+&(%>@b=onE<{47 zoFI~=@pS)YaTwiwnoM45NsG#3iA5ZNAyF-m{+LbFH-?bXqYBv7mSp{abEf%lo$8G} z*8DT0hh!gdBu3|Cz*X4k$qJby^RmRTOZyIuoVkuZKeQU#Ci!B237YoYztTb{W=hj=?-olD7>J>AT{PbyIKcM*9z`ysi%H;xv4 zTZkbyg?yGP-grOYj=*R?k4hiXQ8D&D_Kp+dgPb|MDAI;9=1urcNshmhFZhu26wz2$ z=&zOJ!A?m5&aYYyS4K{SvnvEoo<|O(JIBE3+1c>-#TBCPzy!kDvWRogO0vgkGRDOx zq57j}?9eR6=Gsb(ksXIRJ-PHtW({3DG>@p?P(_XA>!hINyY-z9$=n$|MQV~fi<{x} zii(sU<2J2OVD44yrK3CUk%KG$Bh9xXxs$=wG|@&4cLc7(KmNj;@Lf9YSB*ro2!H(B z8jfoROVLMH1hsqPQPtF!JF=&o*lyemXC+vW8Fm)L^-TJrbwuufn&l~IGN`6g?-$=L<1_m^#n~MB5=1Up*d!^ zEDaccm1xT7!01sM;GXgg5^9$&%yUmemsulZJiZR&HKJjgGfyx6c*x}J?V<0loglkk z-JmD#_i;R=zRMiQTOs;-PtwL>DA5m4Bn5bN+l{ zH?@pdn@ok$SJlur>JTZo+Cm&JE~e6kmiTXeGV$;HP9`~b(kYJ{sB5|ou9Nn~(?Qp% z{@Lx^w4G0ghP4ue2Ivcktl_ZZ(=6yd(hIg`d7x2>!kdu@tCAOx-tR8ld9?+ws5^w= zJtv_?PBrH;^EmpRScH}1wehp$P_vO`26t}e2kKuYLf@`DLF2=8afHufTsnF`c4&=8 z)7#PH#*iF%nf@Ua$132qum>mU*|2Q7B=^9hr1_asC;c_Cm)veixBfLK16CqNV5~C* zqRw@2P17fl32nQf?$%r|Ub9o+ye+}RmvuB?mNr)0_|8fF?c~P9rh#iiFbsbSgIznj z!5|Vrt|Ap&^CCf|q?|S;3;8S(=6E_K3Ty5p;P?z48Jp8W4B356YHac)CH7KS zA6QoDg3^UTnwLEfWyjUw_^L{LwYn0!7F1%(_-Im*&tTPSHRqO`2^2HF__ktXgafO6Z1!x(lBMp(= z1P!dwzw|o&`tvV&b10D*yNQr}?Ym*e!mVV?4+AhRb%u_>V7Ra)8N~0m!Fc5iu-z~K zi$3PTA+22?exwk#7~CQg4CR@^#X7fNR*;F;J5C|ctKy}~;c#)}iL zTj~(<J2U;-tf`M0K$n5XfCn^ZN?ZlgFq5I`vjHQmrC-(XVKGjAn}g&fEdqg^Tt$t<;F#chtg|>|A>Bizn?gT!9DNY^bOEIwGoYm>LD|Y?(Gij-}lizn_t>5vH+PE6UL~YCAf3pTw75 zy)-diV#GwN4i!udR&7_K4teH8v+_w#Szi`L3Z8%O}w} znwgyLyz7i>n-`t>V=T-XbB=TfY_U^Ut>NQ&U(Q6*2Bbr$kgesTA@-3oNh+BQ-&*d0 zslx);cxNLV%Kk;dSB!&>{3sF_r$jw$bFHnPHIiuU2c$3UD5onW0>)Y$q+B`~A|Dh% z#MtW~*4GLpRrg_FbUlPUzb)*6pFr?F5!RF_vzLe8!?d5DAyw7~^!g6L?wMC%SGg`c zlk_E=nFUx(@qp;{LGal!nk3l?8j+i6RA0=A+q(A-HL~@27Hl@phv=j(gqnxP|X&zc0c3e!MU?j>|w$c6R03*h*KdhmTC z&K}=n#eVVWhMM-X5SNz$zpkDZJoimRH;_S_Vr}#}7=eQcJ1};62D;o$$39_RHmhR` z&P$3Rr&%*xw_qx8R`c z#e@4i@KM-a%&BxhvWB4ajITn?%2DUP`gGTxlekjvAE|4w2&qGQ0 zX{Zj@E8n2DS4D6tFZe;$tQPchbEuuTJ_dWQM6232^u^9T;+T<3_Ow{Sr2FF_et89~ z-lxr$y&TE9DvV@9dXK^Glp|oK@R`hQ%L7l%-Oy&41%AtC!d~TZquuY4OHR*p;ig$|BjM=e^6`ZX#RefI)6x><-cV9Kq=8**yOLxlNx~s``wBh zl+O~hm92z7T>>BOw?eq|Log_H13ke{*+1I>;*AA%@Ut3H+!{@vXn2vfC59MwX#u|7 zbA(=Bq=4_HMKY|%3bHg=hU5z0U8VRWr1af~P2?yP{rnEyg9fY`$gyOa8tbDrmR;cM z$^K3dW#7Gc1S@K@;Bose;wnEEh8yqGQ3Cw(a=Hlie^SD;6Rc3WN7$i8-@^`z>-aUV z4V%P0@q9!y1`1h8SHDbU-jw$@&p4qB=Rf@;KD(k|?#mqLAMq7D=T$?usVXbpe-#!F z$HByfSzsc2iA)c>N`BaNQ_tT9=r3^;6N@MCk{@Jv|4DQCcN4Vv;d}$We87N@=$*jd z*Vf>Bl?(B8{&oB^vJxA<-^1m1OHiT08<*J?(UJ{WWYo&BxcX-z8GJfGVg}>Mpye-) zjL&3@&P8w~(erVSA4lg>cbt&B2D7Ax@vEaIuRf^8tJ^iBkjHV2(~aLHF-gjL9}P9m4!lhQv3PHTowYfppIhJ09*FVs=?fkanYus|+-O=H`p zpk2MIu=5$DYtFu)i#j!Mywg`Kc1XZpVdiVKEe^e9>@ikX5AUYM!HV`uNVMM#Sx2fM z@PiavX`#jXwTxiDuUBUuF6#olHVWo;84%QU4DP!Ny}I#daG7heKPPfJGsW=H(@8k(PbN*;dfm#>w}_fm88WY& zy_l$j`yjAxDFk#1JaEIUu*CH&c-aR5b4d&??Nou#s(7MS8wlHXi$K%AOy=i-W#o>i z;FJB3-jwVZLDqAk*rE~Eyen}f#y)FAi=_hV=1mQC2$_nnBFnHYA`eg2ccMYZdz{x$ zi>iJlG}OlmEsdAp$%|XDBkKYaI_(T8Tz-X|f2Iw4T*SzhhuzHUTqXLrWgLEXb)&65 z$4QlGHhtIshSNLPz$Edec=rvEGDkD3kZx~GTzwRKpGITToH)8yp#&#IHekJ^1!goM zRr3?WiY8^0YaDDIdTj;nGAF6pfxlF3f&{MiYa;uzvT^Rn2s~;e$Llpt;otbp;s3j5 z%cE-muWh!JZyuP-*Z581CC88ERVt_P9zK0|s<>(CaSK!$J2bS{~#^yoD@~AJ;{BZp07Vc}kxDtJ#H%9*pO6?6vs&AIq>+-<&V| zF@is>Ii4>W8qHt2Y0mo&sqy-vO8hQgP5x}#SU%iSo7bB;mfz}DhvDxlapvWAocvFe znYB=ne3U&36RV_IzfW~ww^@sQ*FTlD)o^BO-!SZAvENYsdJ6ltq#qV|+=Q*;`{8Po zz#BF373zqWj6#Dgei_WiQ+f@UAuWr~UrFJr;_v!*4=9Y zt7PuTDz1!RA8#*U>#|~49nE5PWAI+~OKc!(`CkEhSKW)vcx}Kg=~rU289CPf;bkbg zQVqeKui)5EV^nPHBBE*+p=&S@ZjOwGud44s|I#RS=7?u7zR-(pnq|+PNZHQDT;In& z)zf7i7F>XDrdF(Spx~zp>i~Cqf2a}k*XOk7;^I%y=yJFk_g58T!}d@tcay|lRj;U4 zn*{8V8P#O{DFzHJGC}cfp}?X0O|_WPAB#GfZm1~uV#a3wA;l?|$_!DFof@)w3 z^w&>@Pok+XbNMc~v~6nh@}==)#qt9ry?GqyC`-fDfD^RSIfrDn=~1UgS4m*QZ1~#0 z7(SmZhLPTz!B3CEdgl`O&wnjw4%$MZi{PF2*#OJcw?f?EjqokCtND!G6?)-88_jJ^ zqt6^Paqb*vK?iyUYd+3K1Cx&^FDuHo9Be?-x%Y6__(zzdbRRd=i1E9hHsVDG!Q)<8 zi97%9MCnr@+%QPuNaJVZK+9Cg;(-?nH1KU;A!@3i(cKJ%`?TXk-@d_yDk zjUz||?W^>iW+>Qb2W|CDu-dv1)HRF2?sy~_Y%~L7@j6Hv$cJj%5@>(07o5cJ!Z`jF zxZrx2_F4;4<~UNA7C^?D0DKFVfMeHE$Uaqpm9u&xxR#8EFdKmp6MBt?=li1bl`B-{ zw-K7pT!!;kmZRG=5p1}34*xzoi&A@BXhP>{^7`dOJoj@yQGF9bZ)TsP3pbc!pX3)V zW_C9tJVG?crIyq5quiSdo}BjjUJ|@g$f5dok^HxG8$?Zb2%54~V4VE|xuRky5EnEl z5mVV2J>MYo+zI&Nng(Zfr-OI*St37L5ym-8fD5Q6aC$D$J*QHe=e-yUsre&dl+Hx( zez^*SDL#yUdKE4eiLs7x&p^TZ545#Ov-9Lfvrc~cti?G|_R_S25M%RHm`yH$R+qih z+oGDz-(ySXZg9o}pDS_m_NS;n@B!ydxr__he7rj#h1G-e=yO|Vod2So)QosYM(yb( z@lz_ug|~~ShKU;XzT1Pp991w=yr0T?FT+>qV=;bDlE9!mDQL0EFxLGoHm&-AuT^TX zTwt9h|7k$^jx;>2Q%v28o$#1}B2xQ#IF%ToyTDS?uQWs_pJ6;UvkaScXJO{;gMzMG zg8pkNCZ&x}Ikn(?s3$Tz0aaSdVZcw}%&z_tMSe9(#w$ac`p+o@!yDMH)WT zbHQJ#X_&cQ;9G50;%i5I$LfuxXl=6%(?TcXiO?tv;6`(e0t@8)oo+INekarRFCzhS z#V|LiguYPjrOe7b_^HkW<$VsKUcVGh&wftZmb0+=gBdIxuzu| zQ1IVPFPjWsbTi@oH!<*QeM7}=-l3{vYtjGB5Z+eR;)i8b_~2kw{%tkzWk&OO?amo| zcCIl$v_O_unEn+NEpsu#-4OSVU5IlP;{>0_W0>^5GYsqn80PVlQWxAA24R9>}f z5L>&?;YNo5tiEDQD{k$is`uxE`eR`)vnYwY?72m@%Sd43U>Wu1^&n-&chO_s`_Z!hGVX8@ z<7cgMM!8mj5w>d^48&%G{CA=MDk+4qe;$JC-vE&LGYKk(9)r}Vw zFJ)K2yWA|;`_6<-wU=WWS9m~EQ^8uc{`AjTtmoZ9a3WVBg zLdFG4>}{|#Q!MJP#9!wx;)H+|xYbz-W1hXBZE!+hUzL#4 z2{&lCzavwb9snVi13(t;GFc~tY%jKjyDHO0>Ml=%mCJeXv%CgBBP&6rB^?e6{&;`= zc9@mf0A^1=fp>{1dn)o4l!wfN+`Bs<<@8a|8@gdV>0laJ)wPhw23V8P!>g>-F4)0% z1APeKY{^C4`LN)G8mKM~Yi>ChP5K~<9M(%GW5?-`W*Y?v@oFMfIeVESxwZ_&#e;^7tS88=_=8(Q7GmdzIXFw# zN??rU(VLZW82h!5&VilCI*sKk7L4Ju53c0P9Bp`mA2Iw{vpl}VG@bAA@!;bNt$E$` z0le7D?Yu#SCvUM-iyv&2FjewH2kAeBA+ zWf!}{X%l-g$(3C)PL!2Ce+q0q`@@~TqabP8G+bxhgLh|+;LSu{VWW_RBC_0&FRTr~ z)r&n)`{f<$H|Kt5}TbL$lg%UW}7@?K~_zMExYs> z=DNhvs?XiTs_ij)K6sB&Iud;Dj1m0IeFOOL4`ReDXC4%m@59{1+yKj^b}D9KpvtXu{>` zqWra|-!WJ0Ep9RSigP3$WBS4p#M{%EZdn%3H)&#L+cRdKlm}grTuMJLH^SwbVvwM} zluDrBKiT$@b~wb4g0E|^zGn^Q+^ZGdr;nVJ-h48)a3rZ{5yKOvKWS6Akk>Hg40Yr_ z(@}lx+!QfyvhJWWSe1p5b3X!sKez$xU%t0~KGPhQzn>!Hr8JZ1*ShfcOb2Xna3Ocg zY+-8h72GMTaXaAu;=C#vDM<727vJ#WzYaguxPx`AGM z%Ry7d0Vuqi3?b`Ag5STbq;*OL_N4~nWZ{etS69)yE%iicO(CW)$fn+kBaxR*!Jirv zv0g3-^#>&|YjZbdW*rjVl|p>me1_G)np8*@hG*PV z!#wOzIZ8jMug3Jr3(#6C9@n%zX0{n94JNiR7MD_p;#VJ@GJ* zp+#GtQU|(|Fb%KC`niR|j%5d|iFJdUu4l=E+1bo(6FK}Xn#0U!m<-ORSApWba!B^7 zhASOsVUu04z+yWH#$R8-g~^TZ$bLJp?INswXEfYz&x4BzIUs4g3mV>EA!8m?kweR_ zFu7}vaTnXS(W+O|@a=<1w7sE~7%fZXTBp9F5wl$|NbtzoI@RLGuyP#zZW5->+>O)B zr*Pl%voJRDGAFXmO30;5qqd1=xGQdpkZIP4y*pNL+Mb8G{sCuvn;nmOD)(`XOC#>T z+=1s0i}7M+2Qa6r2q#?YK?}EXlek$@LEn?4zDV=}+CQ}Hzi z=-tQ1>FK+NaObazrt>#lx#c5Eh--TUIrc>YMe5c<)LJpFPLn5#wQJ!~zz10N{ByAyt}(m5t%Ja6z8G%S|3dju&2-`P{nknQ%}}(VliU)RW&2cS!;--olDWbXM%kn6Mw=JB!k_v1I5t}D*3d{c@U z-KKn|i6sAOSs#j6p20+4Z92nDjHF*F@X%_36V+G81@&8UZ{t0h83j$Kk^KnS1$}V-ojkj`!iYWW&A}Sa zZm=2p54xGHAW1{bv{=l8x)t`Z2d56+@0>pvbp-Xv|3Rrz=;Y?bc`-T`-f{s7}DP z)}!bXribfZcTw-#vV2Ex2`c3*#ETQQ;(*35&Nn}Z>n00#;7b*L!__qW`ur(&hPGqU zWm$gh=QU__(w*Ke(B-6^PT>^oIQ(zmHXY6o<()Ftk)u0f$Vb7Op)yOIPVJw9-@oeP z8tc<^%m+#QW1B==Y!lF4G>@(oSHsx;sc3OW8%H>fz;N>wR61YCsPCCh=LKuhnFUX1 ze{>I{CC7rzpLNtYq=RgGD+m8p`@rP%bFgjvM|kz17#uh0z#m_scCR`LofC73z0)HO z?oGi>@6Bj?zTifv0qnDFKx*qU&jQI zp_rT2M^;Ee{ms45qSaxws8;8x%pt)J3pVionK~>o^nd5S+7QA3h1c zj71_I0>i))6$4qSJG~U&oOw>GQXQ~e@-jW~XPDd5s!Cnvl+#sz6!AxF6ka#EA~4|< z@dyb-#Z7lG=fo#`5d8pWJ|Nfu=kb8R%|4kb%Qr;EVdU#Wm?$#@>udU9{+3&yZe_^Y zYKyV@InM0y=oRd-JW+N}zZiSunyrz4vzAT#SR>$bU6uIil#8expGlUPn!*xx55#s|01a;gHqYIFP0a!JLx(wA z;v&Q5ypv(w{f@#^P=N3ombe%n;bVW`!|Uhss}F_oqsx=|ahG%Xh1Ny<@)=EhcVICu zn9%r-TVi{ElEHh9 z%;Ehnh4AH3di?XbYW%IPQy4T<&rIJt2NdN)p_7wfo607$XWLC!&c}u=q?RoAcsiS~ zL7UyM`7JQ~2?!AKVdKjhS;zIH!u7Vd{) z9Y6dl>VZRoZfTsQ9r^xZ0J6Fr*sAK)>|}NcD?`_?QdutlQ+DRjRK3yvHxGpzq*GnMpxJa-~Yq#LkxF~_*SR$x2Nr6Xgbx<6#2BHGBKx(!n z)ElJ}yQZ_u`wE{yta*x|JB=}WNGz%>_QuY5Ig~Gu#))t5quUTZ13GpncJFb((vl!{ zSmeOXG03E6)ihzZogvstB}0@y0eg`yV{6{h+PNm2`NSJ^)4bNI*b*Jq>|lg)&Ifot z`*;lM4Z&uqc=Y)3g8lTTs9iT4FFYus58_QhGqez-wl1OhW=1gOP7xehX$FxitN-tu z=f}#eta1pW$3Dp7o}z6So6OIvY(0-lKb^%ro6GRm!m+4uOa5kd05%Uj!crr9m`Ddo|_oX?R`(_v0R#AenL3nMiXtw z-beK3+@!ar_OK!5Z`i!rG`i4Doc+gD8}?!gllOY!l@0j6F3 zj-C4$hW9>}v;9`$v}5%&Sj}g&lX`Z7Lsu`+xZlZ|V<+Ld$~gQWJqj)CgK;w)zzmCa zw&LzyR6Og2F7^dD=HwL29lXz)M`p1tp~h$ud7X3^n~~-lwnE=5eU@sc32N&uu(3B{ zSlPmM)~zrOrz}6q26uOprQA`v;mkr@`+P4R*nf#h$rZAeR(~9_auYgs-6FQ(W*Gc2 zzUt?eXlnU2kbop<1(=NnLhk@=?uRmKhc$aVT=u(%!Y1V!Myv&Gr}_))N(7( z&q@xJ%KXsUU=0!ZgrNTkPZn6}Nvqq;nBJWL_O2&^#ePrWX5wS|LO&Shj#~>?KE0(q z7yQX>vkxHQvY|)21lMRgZ#s53)dl zxI#?03Amihp~qLM;lZvB_Ia_i(EnQjv;KRFx-OlFiN^Z4vvU`w+n&WCMn|x@iRZz{ zx(H)eE?~cSCsgf#uL&7y$-+rr0(&4Au|R z0KGIRFqN&O(_8(7nW@njtTqJu%H)~b)){O{oiTSR@1fCJ#rM4P z_?o>R)A$*JcQuq~)XmxC^!k;we$pTEey=Q+bvLksHVT-qdIoB0_~3duh6()rPwmMI z%pcN#Q-4T{&QyNK&(WVTEW8Tux=qGU26NE&&Lb9E8_4@PSF_@@4(6*lpF3xAoSNRR zp{lDwY1V^%SP|@uQMdSu2lcv(wRfYPC>IQjxY-6wt{dt=Po) z*WAU|Z+S;>A$vF?lV*!wVZBXxRmKCF)bq{P<91_kV`y(9~pcbTqreoO|kM5Oxa9H=uY23Y0+pH!A{?xGB9S?aA*jaL5-FT7tTGOQ8i0a zfFm}I#lqc}*_SQD5x0j>eT{J1^Du$_kv>i*)pn2*Vaup^y^wAnqXDOC3TedT^Hg$P z8n;JL3nH{0&_=UEG%w*8-4#5W&iz|Wc1>8!4vJUN-r{POY3IuPTYRg+{|K4y+w*k4 z*m#H&MUwj;#K}0`ZFTVL0R6REKn=#67n*vS(E_PgY_k4IV)(|3H3xaKpMoy7I6?vk zgEt7j6(+I+JO2`!FLm^wZ=^84_7Az@bWEr){{y?o@8y3gCzIf-8dX(44e8i%4rEsA zQzkz+m1n7iV8^{!6h~kDQ!xn#)JV8qOBR&H~jzn2|h zw|?wp_c}KUU*z$w+}F2+jyBm$P|cD3v(K<+x8hknosVa^iRf7yh_3Ppc>T&LWO})n z>z#v+1DPnVnu^7fbMf<-RCJEr!_SBr1u$COx(oE9MhF&19EE(zF1lUy61?(UK#SXVvA^T=n6v&tJRGVeYPa_i zt^DmQI(25D=)metqPzJEL~{0lB0rO5qR$fZMfxXXM6}&ZRH+t)U#}EmAkRItrLWkN z_zg^3xfmi&GnltRUeLT(T;{k+vz{j!pS4dUHfg)4_L9HoF(eKBLM6ULmL|x=ACeD!;)qHu4!hj@P^e!DA*{{M0ugqcXr;!4K#nyty zm*xtdxw{BFbb&L!VC6_ zmO(_nEwn$LM;?L)c#nTi<4Tvq%nNU+d80dQQR01y9-pX*8t?dh(L((Wgi^l|Ike{e zCFUy3dAbyT*P-&$~+RI_xcvpJj=oWg;ek4EFu^jhqyTIJD`RjGd8+^0s zI*Pj;#lJe8xT)11-#AajNJ5Vj;jz>eai!)r^dlFvX-Oe)-^3}y@CXm8*fF4Vb?j=R2_cy2ME zA?ZWmW?UYvdViV_^LW~PBbumI7}4QRZc>wqW;#&zi^gr80H$MR0UY&%1$=INOxt0Q z4RMBKs~~vvb1&!(i-W3HB5+mohGp+KnEWjj+#k<@ckd^`of9|cnPqpWQRaSXFjTN!GKm|{@S$bNm1}tqTc9sSWu*!uD@0UQqKNhdwa|6n zPm$()KN>bNmB#yb&;>(mAy#23uby8FX@~T9$K@EBm@E&^KdqsQ7Iq3BY&3#hs|?|w zk_8+^D>%%5rgee$>94d3s;;w^X%|UiqK*@qc0}XtPj~wprEp6$Pu0k$I+EDZm=aW zKDcJ-Mclbe9}{EM@TRH^E_oDEb^p#y*6tR@wr!L}QwK3D<|wYy9#&=Ya}TXLVTJW8 ztkGWD5Z@I^!W6{_Vb{+jw#xJ!``O&jLOv+Kj#=jHmDo%C8J>lkmvpe>pB2#P`dnD; z^PHtmY-h7YFN7luOUb5p320rv1^rg&aWi&$F_X}{Y{?ikJSHfiV;bhdi_)9KXNocX z{d@^wvq9}g;gKPWF^Dib#wk88r0V*<~gZJwHlv(G$bW+NV>*km^hcJanr zrUT@WM>$QOsE^MohCuL**;HP(hsoz(Wh)-^mT7fQ-P=~p*?4r)poq|dbh+w>GZcYlOFo3VlJ@O{r+Gn@x& z+qB_yem^yO5Xc~U2}S(Wyl`z)95|2jEbs|rsY&CX|OK(f*tuzBeZzot)xb#6Iy z-Zd3&V($aG_f8n}d!C|M=St{S;U2nXi4PkStIUkPZlNE7O_^Y@S$Mm;m>LcY!zLwH zc5m@A)-iaWM2=m-&V<9r5kveLvrOQgD zQJ3vHc;-haULNg())D6TNGB3+7x-X8yeCc>HWF{xUtvzihcg9@I@a{-K64GIVLeVN zxTI1Hua*e#o!C~i%+$s=cvW~`_q|YhZvZ=eNQM4g{+u32@L=zg>~MTZGz+Y*BgO16I2oDl5w zdjq^MBA|M;ja-=|1Fd0ygeJKqXsxV|XVz(;!pujN=BssJ*$fFP3O0tF$qRUt0T1iH zQ3A~u%Hfzr2zZ7Kg;y&tRJA*tr<1z}$n!M=q&fB<3ql7y%7@cO?4uawta6WxDZU0(IAGHL-46k^g8`jcsinDNi=z11E zbOdHw&tOGQQ^&pc>5r8O^z)UiFuNfW>ea;{Zln=azP1xCtV)Gr zck)2`%sG%6u!jfcK@fjDftvBT;NROfL&3w-WM}LjVP6wtd+eRCZI6ub{=o;-F+QEj ztR2l~j_0wb)3xA5xgMLN?+T%@{d6!(15573u<{@AtSSiEW&YNjv19kvB{NU$8^yZ_?mwTRjc2QWcRiMqc0%sRD<*_aKh zh0pG;WAD30&?1`&!VrNU4OnS~*8EJ?Xr5bSd6lnuTZiBl&pPIOS`jPj!f}UT4_oZ_ z7>{^QM4T`h$5o7?>_9ZxpwnFSW%ov0z|F>*fe2JeI)tHfk73)=A4sM?zy;xxn4a`` zCeJ_j{r=sIDMM8;#cU$}GX6tHmONzX&weubqa#smxfuS_S9Okn01L>EuB++8wRl|?wQrWemgl~-c z$-W`ATxXAT)i7Hr9I@ycTYM&p?z=t%_9bqm&69?}wkm!2y}z7J`4UbRgxD~#3?p0- zCX0R9)@dDuU|p<-RL(AP$Rj_MK8eN%PFp1}dSY6rmbU87)1{8Li3;V&6CBwhI9 zju<(0s+m5qI13{WUgX&rez16W04zNm3>s~GEvYF5HVl-3N!Vl9_&pPb9?XDQemWo< zoXwJUdSLRM`55}g3BAkY(PVx-3)|cv)DeB6F}|_zlFy!BpD2RVWpALSNL{d0VYFb~ zweu8sTX6xB z<@bXIq`?U9d{|}}3j>4ow52+m{l3%6^rT#IieWWsdZuHFc|I<;KZ`1+LAai081)+w zl=&Mc6n`)jZ!X!*48Kz0pvM@N_OMC#GxB=XsW+}{R5G8LY-(Z08?3OjIRpQkXIM9) z8Ru`-7mY126&afiV9;L^QP__`tnAYj&AP9|^L+kcWb0)lH;ORvL<=q}dV}>%Qn8m|}JG>w9cm*{~-A9LCmWCJeJ2s=P@|OriB}CxskdQE@e{> zpP-f#FVL-9%joe=Ip(-NkAAzQ0eNv(*s4Az+;`Q7tf~Iar77p}94jyM$0^j)Vvq}o z>ZMh2{JMf4+$^4j_w1THn_n>$V?VxP+T{Z*$aEy$P5;N#giY++pAcMseIcDWxsg4L zP{j`4`)pvED*-!2jQjP^e)R9(^dVgU3-fpd>S#1&~=d8p&8HU!+)iI~b6#wq|Mp=FrbKsv5P8{rG zFRG<+7|O8B2A<`in#?W^9iUy#E9mmrHcF@iyuK#~cllgP#mH;$B_aWmolZi^mNs~% zCn?DG9ss!wE#N&%TF{&_2muWbV8q@BaL_9oqC2m^RsCw%Id&v0IeWEArlyW2q?lua zs}63hPe6$#7xXl;$FMcYm~1?V7Ts0Y@#Y?`JN1T{UH-wUO?I zCOn_MmR@_}4$2qhU{}i)7#$?dt}MCAek80Q*(+1o;z#FM-30?^w+LrX#68%b=tV5# zj3Y5_n9J7oOh(RRChnDwW43#)GedJ_aFqK7p&Ccp!vu(Na0 zJRy_UVr<_sceZM?7XD}>G_Wa}X4s!6vWl|A?8XyvN!K3oydToGTS+v3dMfcKNEJ1 zI-YBwi$1zTQCl3iCP~8H-?m(sawi*l$e38kp5P?L-C?syKAY}8olVl46#K+`A1j~wh|LG-7#zLb$ zWczmRz$Rs7Ou8tfrF{~xxc)cwNM8mUb@{{pIo+2z0n<;6V>^GVqxZe(XcV*pUk}Sg z_d7Bo9eZWbNeNAnXt#=}>U=leZjl$Yy6wdFzRCFcPAE>+e9L}sze-IcJE+m)p->zZ z4POthqDB#U5EmE$`W`zWr2QH-wjN4x;37Lu8(G1sD^%(APwM&9m7e2=I>Mwk)3aug zWL3C7hz9#;{<1T|+5_(RFp1xz&3D6yw2wl0uUhi!Wronc!3j#|UxWm=tKi~u9?sp# z0Z(rk!QKfea6&EvKD%6@CRwhq_}Ef-pmB?Nxu&7qmRqz;<+_I6IG7Z~DG} zR{Gja5Z3k_phbPxsP~dnbWZC>x@1fwO`S5Hj+^s}ezp5fV|bUAM9x0a_^E>FYbxW- z87mko>SePe)$w`sT}JgfnJ{1sKJt;kjDo_#Jy^ykJ?`^sm+Mb(LH)t}ESQM-;! zI+}@+8b;`}e*yZ{?#KUX86UGF@T>VvwxD4?+0*R{154}R*5hngbwylo`ow7f{bY7N zsaLrFeKwFG!)VdWQhKL2lpUH>3_qS6rxWzF;ZS-h?Ce|(y=QO2kG$g`S9%nj!^@!2 zu?w2wG&cdR^c6^13B7x2bQN6%Obo;fGNdMsjEKQddsT90Lc;k2-CfvM_8E2Al=^t5#$?ZHf&Oj!X4`p(Cqyo{1e948=CUmtmy}4GV%9=}XdmdP zw!w{@CeZJF2eC#}P~&AS(0%7B&>KHe;OJv47#7_L9~JZkCSopvsA*kvc#$DUzuUup z4?oX5idE@)Hd^@Yau8Krmjoqv7Jdb6g&vDT;9(&Iou7YU+m3k1wYmr*X=C_OpbRJZ zoIvr?4r;wBaz2)#{@KwguufSjZzv(RrN6@KPgZ`Xl z0QSkyVp}zSr;7xU z)LL;Wo#zj9PQi0*y&^9%`YR)PG@=+Ur#0aCZ{=uV)r&21BAgjpfkvLE@GRe>a*Qd! z3(bjmYU)KicGQLUQ5zUJ@LatUy1xrUN7XZ!H^dvIlkTC;gijdW z_MGR}{K7?BRYmCQ}$C9xHxV-lZ+!FCRg|3)j}8f`=y$0Bbw6CfDKjNRJNGM#X7Kgw4et`MI_VtNq>sQ4 z=cnQG>z&Ni^Sy9#K`I>;Sx$WKCDZ7_CD7+L3`XsMeI$`n(SAt?>Yd~n0wD%i@u3*kb(Wy;LLr`O zZp3S)_i$C=X!Nv>$Ar&9>{$B>HO*@EPUQI{1R*3uFw_nyGe!V5UnQwN&_S?H8` z7>Dt^rf%~Y7@;Ge5j^PKkRuXK~NDpb@P0OOfE-*Ob$vepyCV3P){ zxq1Z5r9y%En!#t4;q;rc8~s!_7LL8!2G3a@DcnB{^0pm;@q4bp+dH>mok1x?>Lo** z+gbYBlz-;#cLk%nPuPW1x$HrgKTD6BjEb-0`MPp6?!Uhtb9^r`odY)ol-rqa@Z6b4R z7T|SR-VLBHg*^$!d|X^`a(D|^%}57>?%$w#n&*1iorU}JR=|n=BQT?3FFXxx z1?fq3@S`Ri=C0@Wz19)1RCgr=6{W!jXWn7U_dQq5hy!K&IgoDim;QM39S%ONfrTCC zAR3mTi>~+&o&tGXs1<&N3fTE**imr3{{k-q0ypM&O=f0RG|mRg;8_4Tc_I zomW%IHmY1{&*#Kuj}o%HvA(Dm?T2$pBk{#LM~rhwL6w;waJN}3ezD)oqDVdccTkpf z-kJ>0MCZZ6tw=at$+og=$7ya+=15wZSV)I1O{HSViL4(A%YOQ$B-D^OjTbRF~`P7xOuvw$fH|JH0x&>u3X)NHNn!Nw#nsqab7r< zyn2VfWJ1t=LLn*-u0_d?8g#fXAzG(th>P#;qobyV(r7^f49-{#o0F%({1e8oHFZAJ z76QF&_ll*TWSG=DRV3KAQFPX&>S&N02wO--#jQDZ~rqxcN zH7$W`fA1|i)^i=JjxNQ7i~zjBj$*%_jL6;K9tD|rv zr52Rk-$2(wdqL=ZfnfaAU_oP%vmnXTUJ!Jn42qI|fVa6KpE1pbtWj3z8P|>Kbx$zP z&Qf&O$4w+Yna}6WUm?0L3J_UX1&QQ7q={_W#*6AlH(|m3m)L9`iLJ?tFmGKD!t6QB?BBX`R8ttrJVYnhTJ;HNuH{`-W|hp|xn>IcqYg1u z2|fcv|FOxFe3Jq!avA1vEU%39BoV-K?F@RhIVot7g!OV^raH<_X5`Cm+a>2q>= zwk;DI9m%;g$8Lu8tZ@2OY%=8A1k!gdJ7DdeE$~G0E!8;tlG(QZb1)R^q0;6(+#Sxwhbu_MGNeWRjNS(YSAZKEC+FuwiZ+KH1BO zKDDce>a>(aS+Cz=hozcGvFAOGNa@F~h9Z)(;GByEh@01|g z)BaHSwqrRwi#P`bzKuf5PZ~IO@FxownP8DJ|65tK8JimJq3>G-5%xSsizZ1?TuBQm zWQXCUzgN)iTrtkPc$THU2OPF9hvvStgX|CQVcL?%WWOEnnn?|YH{Kgyti*e0d|U#K zF{*-d(sBZuY$050vJkpR#KNelCN#*@S$Oq*67Oe~6>W3u!20SRXy%iSf^=;>=&_#l zjyuS*)6NT(hQ*-O%?upam5Ij~A2hbN;`zgB{Qpbhqsl#~s2qomJfpFk{b1jetXYvi zkQqBSP_k_mTsU$S-UZ%=%0+z;GE_=%_NcNTIqNXIG|z?>ol9W1%9|#4pJ43D9vW&@ zNIwqduRkINsHjngf&o8SqVMHX|Ut^ z5eV^%hLdYo!Y?TYcvl@n8;_1*Zbx0&?48L%w}8>y%sWD1kWoCX7*b5to|V#06H0q# zJ)rysmEIk-moH@NFL!)Dt+xGa+ZjjhMvXlM)F@nIiHQ1L?r_JpobKFtEQY@`*f zlNsmtoQ--{M9XjfV%x5)gfrboE`5`!q=#x;z&m6UY;m-Kw9hk{)t3$yP>{srR%XGd zlo7D8aw{l01i`nyhhVC-9r8xrf!p_5K{riLaOUYdNSjp!@tYrjCToO%l;3coI15&Z zn&8>GG2k>d4&;WOhtCS~Fw?w+hTe^!$*yrkJ+6uw1i0f)hvBGV>B;vaqp_r9BWitl z!1mmhWV?;#apFq~sh>+CbCh1fb>5gnw|-TE55Hnz!FE50ZSjEu<4Q;{JOFa?mGnxk zC9I#72Q~FupmfJiIx}5ah+j9ru_rq8z03{rU9^jtPN`+*uXM3K_npk6Djui3e!)Ih z)w0~N@hEok0y-&-7Cj5?#UUOG@ix!IN_c$~L9-H%I94JH)5nVik+{d$8JFDHhT~`b zmj`oZ%k~)KG`BqA?rjy!{M>5xhIi={b(~>(j{4|rbcCsV9nB2ytik~6Im}GTlId3- zr&6P{i2U#8wBytb`amrjl8#$|aS-1loY4;F^D=o~nj0jvzM#*Rm%#4@7hqmj28@nt zgC}kg(9=;0V{;EfvffBRRfvJ$jgz?`y`}?POwK?oKg(?EGC=zczR|BizVv==5L25_ zNq)}TgA+u{Siq@^Oz!Vw)}fP6{1ivSTrqbl=c7zT&5wmDjj8nUd~cXOQ5I_Un894L zv%*@5iJ*F5Emi16Wgb+mcG#`S)Y^c3Gftb&xh`0_@k-1NVeo;GbGgSDRd> zBs-X%v@E9|f95m2wPT2sW+GkMbeHToFpSJPy8x4IG*RM=5a+0VVnN$y;H`kG!p1ji zdH(8a7Pxu_doyJ-`(p8z%{U&xcKh24Zw31czYlF>RuNX5@l+<;hULN z)Gp^XeehixR=qwU3|DlbV=ilwV=ZZDJ9Ha9m?Vyy7kUZ5)m)@&t1Vz$VLp2_?*j=8 zl7wMjWk5wu7kfP|*?6gJBoB(v{9g|CO*n;G&uZ}6h6v1mbsoL^R-uM(9L|s4i+8fu zGW{E4X#TrKved{BQ-?(eUxXc^=BLKe2v-#-n#yyac0`drc@x-Ea0njn=>w^TGVl(n zfo+EAuxR!;_E`5$A3yVcY7c@Vr|O}m{t--?y9iy=dhlFr zJKChSV!puxp4o95N4VX`n6Vb9jZ`LhBshPNp!FU8G_NO9XePUp z`EJ)mk9USRd1D@qv#dcc5^S2a3!&FVseqbUO+ZI`N6FfX^^3kBT2CAaIR;~5w7UcM4~Hqj~tGEKxX+gl9!5g+&S-2 zl(6-<(CtN3P|EJTZA~L=?rx6Y_i_iSl&>lSK<) zhk+MpJ?STZhHFxr`8IUh!$?vi${-!%bBX%garDGFXPQ04pC0|&=Ai9l$XO4{b6s1O zlMk`+q~L2fnSM)?*4^}>x9xXP@7tTm%5IJneU&E%$|pP6&Q@{=ck8wfE6wMaN;=nm z`5YIgQOmK<$y|!%Th6L=KWF*s3O82wC#O-L$33h6!Ts&t#LuW_aC&p@a3^*s5W__c zbUL3u*82E?2G!&epVe*T%mQh8v-Uda7kfsWy9UY9gNwQIno-g1Zxz97&xcUo@)cU~zx_J|yyH+S7jwOdkS!X%f7avT* zK5Qi`rJ~8ETw}VT!Ih>j+)r;n0(WwC3>W!kALo2vF*iN>3}-jCfeYK0On$}OC2A?x z$yh^O67Whuez%#Dmmh|aQP!@+=5r#MB$3CZzRcm0hZS>57OrG>;1W`EGM4yKARpoq z$m`ZS#6NQdvC8o#4bLLU@x)5fb-9x)$sR(_Iu9X@Ikx2I^BA(!YZG}taV7cs&Xasx zl|g=2Um?@hw~*&o9+R`=F1a%0HtCC8X!knmpMC6w0}dDxN}_n!<}%?mq8NJ3Ub;Bm zey0&}2n7pjVr4>oTC90KgCc1>5=5jw9U%KAHIsxlpUB1+GIYZAP-5tRhKwAiNVTq) za&yML3U@FD&K>7YtChKgmL4pedTg?{dE|8H<`P-;ShH);|DjhqL1^*lp*gz zA2?{9oXv@QZFJCD?M>2VA12u;m88DpAi1TKMr0iilhVm5q^Qi9q=xR})|Str!#d{E z^#k5CeugaRn6rmmlQ=~b?mr~jd$edapI5h;F`QHDn9hx9Dd2QmGRT+GheUPeP&#&M zGTGLbOoqqBlY4u2kT8RIV$qdDKB=E0q7gOZU)~pzwN;8pjQYf7Bs}1pS~u8DU*l#s z!T712n435;b(bL1>tAzyCq~o4BeUtzUA{Et_d{;*hBdi$%$K~W)a5RnTFj+3dhv5> zCJxJkHMsQYbGh$UZ@5uow8-d@R>W5$pDWz;kF!iSB|Eo?$ja7xWLIMsvGMF6Z`!12 z?S5UF-6qSuuKVh6Z{bXCjJ5$8xVw;;Hu@9EKMCA*vwseihLYUIhszwgIDdyw^+bn| z+J)qmR}}fudy;HvKTGCK5TlWbE;Ld63<>)Fg#7;Yn&_=>aZwj(Y!-e!BIKx|Z>oXuC=DRNOooc=QlSe_u+TNwkuQ?#IYt+a}^mevl*0 zH4f_D=G=$VBe|mZ)ub{tf(-15BAYhoP~Bi>D($zC8tW@LRP3GLATkbj*f~edex0I> zeb_E_`$(A;+=b!mI2X??oOF7+!|$fC4k?c>I8=;k;U1pUAfh1# zd07_7)#q1oLkcRmj)IZ&{Y_7*`yiB_$qu4aI+|*PAEG`om1KIx3leD{OMitY5(WKa zGX2>RvU#I6d80p!oSLOYp1NC5r}rLo>pEXrbUv381{84j$Qy3Ryc2Y|*G>A|_XqtE z_knw~+lmz0a70wQic_ho;gXz`$WXS5w6w*Ls*|Psb>8H#cU-H(EjtD7_N)V(_|mi7 ztK;>Y{ckaPyi1NA`XNST-fMD-se#-iy#TJyeJr`3vY%Yfctv*k*mKXP26ApL&fFOB z@0`*b{<{!sL@Z`s;Yjc|u6%bNcd_^%=l{-;*kvpsn$w<;=T9W*v1A3hEo&XoJsL$k zujG-S5Ib&K^D?f-J(a7dKFqZ(>*XHJ&>#!f9d@|2Ji%f9()SJ(fo=A+DGwa_73v+j zovesv;(DUKB$f0IezCW#lyo?J-PK{1ygga?H<0`|7)g$fs^yj}8RYB~Rmf$RpTvKm zGPNqvpy{PIIO8{uIae=9vMO2HVdP|MhvbA>`&*ei90t#Sa(J_)g1hW6p4@dALwvSC-vr%#)~$j^Q9XN5N)S%eEntG_F^*eUJV%&KS0(>iBYL4b$aWb3QcO#b$C2z z=8){Y$D!AI7#Yz&mYh@FO4j@mk$CgRWVweVog#LCtdOc8qxFxHFUB#{b#wxCK9)iK z?R4PTd>zmlr43wSseM+pyu;OyqxSEz#YxiaF{Jjf5s8~Pme#wBr>E_wQpepUWc(;g z60Gh_Mpz5Dk>y*tKD|(``b;{v(Itz^c8=yGw~yhC(gwtHu@+hWMx0EWIf7i>wVq&tmpT0xyMS5k4!28X2J zA>34(aIS9JU;7cOQthv|0G;`PXykVG4D z+B2W*P)R4;kP~F`xzHo7NXx{pTJ=>=Z-sYw9?E zt0}&nCX16Fs|oVkel!*f0%M4T=-nyct&oO85?MGqD??Bpdac(2kJ zA8ZkvhYEE=bz5JoSsIK}4=e`xTl?Tm`C+hBGhqH7wb@5E19r5)hzSQxVRPrWv1vk4 zbh%w4z2JJCzRSIdjnALs(u2Y*eDw}=m>MkTQIFuqpYzz&@R@9J>m2q*E)zS;QZeLC z0tSXQ(A{>@5H)W*Snp1Svw??U+}}EQlMxLm*H44#wz(krCk;1NpGPh2lSr;C6U=`$ z;6PL;wjSr<7Y`TwE2W9ThrHQlEnnt8aUa{iTAzKpW5!bNPiD?Jzp?Lx1S_=|#r676 z!BDOSmUyRy(@xO3?#R^pWz7ahj5#Lnp!8xhPXumrX?ceIs>ijZpO6dyH zqkNr?(_KfJ1AK|TwiE4p6Gm!W?70u-pJ`l{J}vw0OFp|K!TvAnY1$hV;yzxNoZddn zTlwoE_iHC|m2K%ZdG~<(@zWNZ(niUw+}YfzjF-?cTMn9D43e}Rdx+iWAxUTer8N7s39nNz%}N#%B!nL2|;&R69IpcdLYm%ywA zh9DjkMI`+7In$eooMpB@nMY3ZN>500os$Zo;;SajIPXiJ?p}^d*Pe%ouQaLH!dlu~ zoJVKfEayh1uB7SiHAH=X33td}1aE#g2&pD%6XW^&)yDp7ay|sp_ zv?maujD_5$P!-%XxE;9AJLKEAB_NvZ-)p*L>NgcG z_w7Nt-8PHp>4?+qPhOA$M|PsO0ONyyCyUw0?On*^yPwdH&wQ-HuYiZL`)wIv+1V9^8B-;&MqM5n#MctjPqDVsl`ZI;P(#vI(_p*n%59!@i{Q7zrd+I- zJkRBUExltnmNfgD(W7^U$hPJ*u6xo%&NJpdX^`*bjt8W18^jKC)<;@+(gw|(&Ovn; zm64{#JA=9Zn)TuI%hz!{GPyQFn(F;_QTj2G^$K#nc8;#p+f z<^2%!_0AV2fYR@SRNrBcCUh#!5!7cqP@2c!eN`H4*?*os zYuARgaSFg>?WxiFjdYpRVcOE5PUlaWL7Tc#Y0msZ(B2XTSr`0Z%&CJQnbZdVVnuNz zDiJcZe#7Fk#weF43B8IhC^;q!2`$nvpc4!w{|-V@Gld0ZpNcyo=L?e0) zu@2b-^&4uz|3@d-8Fk>&4Q9+db{EUM6is*SXrr%kjA7wEQQJAc%WO=HscrMoO!`tB z;B~<%Izf=n?}h8)JgXS|JV6^Rf4bqPJRiK&B2RkWGp^V15c%6GPH(UI$yqw461O)M zoW55kH)tKtEjerrWwVkX_(u!OcvFJ8*@HOra4eG(Ue2jcUB|t@@R`g0Y)?XrZxQ8n z-^h2pNb0ddmj*P4b3L{Fbl5`#X2dLm3H2>h@%eOU8=nq4-%4T6?^Q_bqVUt&`P>=a z7OrWME$Q=Bh53nPur-0q-IuP6R zhorh$)6?6(l3v3$^8V9Gdbund>=*ab?9DSkp-vIz9{GpK8P4q8A%7M)vx$hPt*5dH ztyD$R7e-ZEVIa;7yVBls*|#>5EorZ~cEx0>!aqXyd3(_~`!xFdffhCD>ZSMMw6P}E z0rx*#jDJNss916pO&J#nE9)Zhp22;r{wU6VzG>pl?TRHC*Z*+OL$kRbZL>+uZFAc1 zQpY`6aGq=*GNZ3%>yU&dcQW%*ANeZp3bwoBz*O`y2+gYnllA?uJVXOKk6hvfd^NQ- z`qRvt_vCu6W%uW%`@oc4d1EAGD}eJ zv4#xM4A|FR4T)B=gkJo^#g|>6Z&wV{PuE}3cr5v$(iuyzA*ZQ ztE0N|FfW;V&-r}MCWHCqQ1(?1ZQCin?I?sREqCGA>?YVW>Pv!8+fZYLv(&%QpN8%& zphY@kV8ii0-0aX+uGp`PMCOFhvHNZ4Z50tZaI}>iyZnYYe;-e)uTO$EqIc<~;wG9G zzXz@de1@uf26*D$WH^2}8KnHh@L+TWw>nChn_=ZojzvEp@po?#gXG(!uq=&t@;^nc zAj^~2ubTu94~4*l6Q@99nh1O?+7A+)O0&K7DL5C2J&xWxvK1Age&l+ zN?+sX@UxrLZ{Y+uJih?0ZrhAUH_Ng!3+ym5&lp{Ka!e*Afi8T-qjw)U(lhT4l89j? zVq|y{csqOP!NC+ReYy=~_qCI4|14>+dKeUJ=(S~PepL8|237?6LgK0-F3Y@@?Drdi ztPQC|$Yea&_@|TR2a(Y8XAOS7GY-f~QE0r*P%mygy-*TBeI1TLLy-~5PBO&NiKp?& zY=OS;&4SZ68b@t>6Y;^EO~`mp@$fHonlGUbo3`x$_j5xwUo4C$f6p)sp6vo&OB!sT zcTgyID`qV&C1!=LFn_~1+8uO_JMr9&?s}|1;$mXRwIm$~`Vj$FLM$k{df~SCLnQ6b zM_M)tc&0Ir{IoWQ?MWQAxJA(S{w;8HXaQEMHB+G~ ziIWe|cgY#Ht=Lo$cav$_*Gag`ZxXp%TuWu!)pdE=y^6 z)Z&T*k3y(u|808k{uzipn8Cey6-wTQ7{kkIBQp8oSmG(U1-dm{;PetN`sryeJlc1L z>{?|82jisCHiJX9DFCmy%w)HzGTn0Kp{+xG7b!jSAGnq&kcFy+^iqi@s4g{yM@~H4 zuu=jP@=b9#`v}@EmWCJWg}CkKf@y+ABlj-A3?&r#Y?Z+%@pNBLx)nCTOQj0>vS*ZT zbr4vBoDL@q^A_gSyIdw@=MBk;KprPp;uW!(X9MCp| zxex8h8@qhCs&WJB!#O-=cN))jH4zxjAwk0laM-hq>S;`%I*Aj&t5X!!?MLbJgOg#( zm=B=-NtEZhN0N>vZbGM$PCCyZ50n)P>D>+wa0(v)XFq3%a@t=Bu)(~z- zpb{+je3A^TW5hZ_7}68HXxR7=E-0>w_TFEM&pZ-vL4Q968QQ_DX^%J~(Npx+Tp{jn zTL(EOxfpip#ZvK16*@s?4`*c}sHjd?_zdBKc z+$e6=gk&(T9Yu}9dPrSfL22VT%=9v!|j?)2XiZ1e)9QnoYZS*}% z%AU%qC+`6MrQ^`D>?@pIQ41w58YrGM4Gkq15cB_D(2dQ>Q18=+`|HJ+$b4fK*WE@| zG#b&D$xjzI-ogz zGTc3I2%ZX`!L<|0(W?DA#=UUIlL0|^NBAfzN2x%RVhjYX>jFI=1^lV!i|$u%qTZ$X zbT2+8R&~|%ksxQcuO+B*<}_A$Tn4hpADg`%;qc!<^1VNhG>!I<-ukn2x79~FLG(CC z$&8~HCmT`+=^wZj9Iv@i-wzPY;(-@p~s=pxPVP{)tAT>$T{$XE#(nKL!D{ zmqEOFJl%{5)WbLkp8Y)mA`=|o@7MrHun*>j$9Ry$HX#~Uu?8A4vmm#;9Uj`82GxcC z!1lihxL8{dOLx7a%U}$wRqz0GF9$n^51?(TMRz2wqlGVL(H(lLF~{o^`dm^UWos|RdtkqDo+7353#iC*T z68fA~(I*D(;1DMYUz9E2Yr8PqOjo1^-*?epFHTe2wDk}weFst(WkcfGRp{}t7VoUr zW(y1Vg8iizkawEHA+xz8>v}DDx9JL5JT0B*E>fnwd(~;do6R(|sg(X_ltG0@FXLj% zCY(7W&hifc&1#zihq?^$JzY)Tm}r7)_*KA|4bZzZ1}?5X26@Yt(?b!9XrMczs=TrA z4>;KD>HrF5YBZvCGL2e2N_kqXP-?D+AQID^2g^oEv6m7o^8n?5amLPG_)IN4YUlP-Kq|F zQeyz3^&PhbhpxAwyqxRxsqTm)NI7m<}G3Yc}nQE#~8a) z-0+!VJeUighU?e^w==z{OkpN1m@5rQf4)L)$2b(O$V8E4$yKbx$3mRkw;uPu%)msK3OqCX3XZsB!mGY3P+7j3 zY8-Z@8v>M|Tqzv~C;URyd7A8+&S|)ivl|xO7sUmaE%Bak3U1BpL+OqzvhavBVF9bD zT*aH zs5=@SIf8y?55qT&v(SI<5xjo29fbLbu<*GSMx1Pf?NPF*=%9fIH=4t}*P-ybNDMc* zpThS_x3D8jnDJJsg2~Q3P^SDADqpL?=YTJ;QGFp6Hy?(t_axD9=OQ$FyA9UQ3W9kX z4RB!i0`#g!z={zS^q-nEl4sSPQmR$Irw>e0KU8z#@1hBaaP4TwB8$wIfCyzW$s@tWYHY5xI>@p@7|4r zb2xNbIRfIpmDuP~L9F_~u_Ice^mU~oga*e!Tf8-d{kcZ-{VS<Sfiu#YCJ2LU!0IMz@}%4C^2vaG<4ph=Vh(fQ=;J%)5ELg!*z~4+_b*1^VE^B}~9nYiJ#--vw^L{MOASgQS z3|`I&!OR>t9I;Bo7==K*%(>yRE5UR`dNG}1zMM?Bc%DvB>80@Qe-zkEf* zvP$UkyvMZSJcBtqu7T54d-VQRf=V)%aL21a>(QWX0{1@6OV}L8Z|9~igevuq`3|<{R2Kl#2 zV76l}w4^=-`6*qnZ+H$aPzpqidm-4@tpbm9T|uKF3J&yTVyjdsHg{gbaNaRE_(%y? z@7{vrccc)hdhG9Y(DW`9ZdXB_!8&MS$Hfi9B*uKMbT^*%p2Z= z;;!5A@rz+B_iy^xmOI zIO{(VwlsMd&2t`O$lqpk{#S}zxd^>{LxEHb##5y7sG)u7Z4}F>$MYL5pm(b%7OzdA zD%&E+`DaU*PL>ZV|Lnr*^P{QXcs`Y$mr5%-(s5Va44k{A2$q`F(dgkC8ajB9o|;|^ zX@_or_U10wTJaL6JnO-M3wN+UAq4$x!g2D(1K4?eHk$VgVV;g9gKuv!9Dd=pY6~_e zdopeO5=Pz}xyViOxk4|>CBr|7lXTzePh`5|KQNss%1}Xsrl^Qu#2-IwP~J}U_6nh) zfhb;hElGt3rqU%pmEeQ&08N{Eik@G@ht=Bi;M##0j4axZ{j;tS*N}Ixj9-b;MgI}6 z@h%Y9yAcb-Oug-_f|(wBX+B9N^`ED4VBJd?m z*b3fi7T9?E5FXdIVlBKra_h7)9;q{AdyDRq1;1>-K}Z2>-*G(Ite5oL9UFo7-*NJv zUJ)368Heiw1bu^-8h*HO9>p#kp|R@sL5GdPnxa{#@9WLn?Iii><7a3f*L*V7_8sjfj&iv5TJ1v&X+I8@t68ZvFR=1zcZJo>R3d- zZ;hgMmP;XeQx<3!&!O4t?$iBu$C8-P=R`N7fS&XE0hD?cCjkP-tL z8b(ku97ZQxxj?UuO{P`9H{;EiARJVCLH$Ky@mf(T-m%+*ofot)X?r~0aVn*Ohjzg8 z*l>uy{FN5B+2L5(i`X=63`h-Gp`CXG-dLsvb6uuD)zn+Cq+AWZU&sT!KT0@KbBxqC zJ)%EsBe<7(GvM8?-RN4VjGmhw(3|hYux@k_uCZJQV(}(ec=9St>nx)OV^iVGU>IKT zs{(^@?bs&%4)xaVAkRZ?(^QRF)MMcUn)gQ>zh8fW%RfG$vl@rULm7Fx>d_d?iOGP{ z#Wz4!sf-ky@}s<@t6;glmiv{n4xUY)h%M6chb4PcqrBl;oJ9qi0RSG zkX@E2WqsWxRLyuItq|2>2Jy4v8kScP$tLnj*pk%4^pSuqxgUq%d-FKjP0hf<&lPC*`8|ty{*+0m zzhj#ORy#W*9+0W~R*|9Un&g7A0k`CJCHXbA23!;U;lx8t7_1IKDT9yjOnfZV{}N#m z_oUgBtrM8aS%G)DO){I=oXL(IFvHtpK}Y&Bk8(h*G6h zxTOCoMhspS_yUyjJrhd#8=^MgB-Jbot9p(Rd(Y#}@1-c2+ky8FWV43LH`tuPZk8mt zt(ID?4K=ew;ZvXs7CS|t=_vtQzA2i0+!)H9MMg3IyYsPi?i75QV}*AX`mknxB2&G0 zjNNyfakXQSoJChzn=)h_P%qtn@eG9><+P0CHvXhl@sxU-V*fqRDt?Z3VfkF z1AgpDOTI&86wB(0Vt(60*h8IJ^u@-tbnd}+>KnC%MX3j}-UA`b!@mkdCH_!{PaJ*l zBZ#cuuS`0(EvDsv-I&T#FV@r;$dRcMSY9mbooyoMbz|}R(m?#t zkcs`}d8oPn17@c%rq&e0;zR_j-6K!-wD&u#o%9xbi*-?z$jn!ZY^XqP6p$^J)&nH=`SeiqZ zErWB#k*T<;f z{~o(U?!wd#Rm`5k;R5plNL_guWCn+!>A+T0O-;b(5SiKEKpl$? zF-wJ|49>4%zUk+2kM(Q37oxy~&Lm@c>kuy3ZOC#*uA%;{2bgeoJR4T>U^8mgvR2I; zc5Sebs@bJcLlu3JYwCnS7b`G-?F)QScMJ5YPobNn1G{;qkFFgR%*f@{Fl)|4AXCd> zp_n;Zo{|Ug4IG`5>q7hO#PFPppw_eN2jSO?L2ft%{`th=irKXwxbH$2HwsvojzjqE zmj@ab-^9M^Phb-I9~@`;=<3`KLQUzQclH$+O)zJH6Xlty`U5n+@Qfz9%%=0d7SPd- z{m_Fv9L{-yH}1$V*}M~IoTS34_OHfJ=S(Q2AyC^Vi=+2-aCx&8E;(QfV}o+xb7%~< z|IK83JU_Ep`&aY*zikDvpKCCRIWn=R6x@)07SmN;p(~2BF`D{p>+RJ{GWQ@n3&^1+ zn?~ua0Rc<)I|t8JOk!t$y+nDHY0SXj8~Zl653hHqW7!EY99)@AFD3mTlQIQ7_;O!- zGENAkS}xJ7i3hOVMijjwzEPFG6TsBh1XOmEK=sng5VT?fAioVv24M>)e7*rDU1!#A$5p9X<5E5(RE$ZuHMqE!ZqsOaHp8gYw~Myq0Ohwm+?Azni4- z(`8*`*&ETeTL-Tl423Cqe?cZOfn-g7Omi-rhK5Cvc<6^c?h7)*L#E4d#BnUHk6wg! zXTL#v{3Y0Pl#i;>>G0U@5On#Z;GalMoIW=ZtYrY=r`^G2*E>t#q&JbFzD?5#w#0 z&(7WbkGZT7M~lrIJ}o+jM+T4M&s%0pUM`XuS~lYB%c1z=Cl6g_Z-QfO$Dqc&64FIv zQMY&)v^H)P?3%>G@moF6QFj!1`B`WeDTcOGv3Cf6h;!8IVAf319;fg}C`^$cE zu4*Z6y>bziQJDR7j|VMI5ettO;ro5Np<#|2xJ(>}cSVeFlD8ZTLLJThEW&=wpT&f% z^Vr&QFYwK~);YiD-BTZ@H06JwU~z%I5a;nKX_kb5`;AA43{U<+d#hu$%} zNB!{jaWYJQ7YAp~x}vq>Dr_^eVis=}pm?+g&UQP71KT4Z`Gz&x8ug+6nGkl}`4DS# zjAZ4xzt}F>cGjxe$Yv~SpzF$Z)A|ReXpGz~Vgetz_Z2_LnoBp)DLD@-=O*Lvv8!;U z`~pnKS%$g7I?xpnN9}ASVX3GT-n}I7lvlg~5=|5{J%vz`>w$z~U0ifv7^W(Z$Guyk z*wQ2WS?yhKrY^e~BV)bsi-Q*ieW?OoLHFadv9rC_{R5V~$%j`2_DuzBwGtk9IO zs23tIuIL19Axhv+$)ofU+*-dV>&`oZ#NDS#q z)}?#x<*AZt0*%-bMZfeNr|LUK>At*PR1(BC{l-tYe9J;ykSbt~R0MNrYfaX5L6;SD z8?lnv$++>!Iy6NNlXm4YpM+GFX&B4g{7y2*;(RtI`6&D6c@g-X+3>yN6TH{I1lq!v zz+>PB=PFEbx%tV{)FPqH{$TN^(_%b@RbIgF$x7$Yn2yGsn_6M29Bm8Zt2 z8#M~24ZeWO=RYX6^dYuSy^k{XqaY*P2=7jtqg;(VC_B8PyG8p93j6zmdaOPM&TpDbQhj~WW^}Y<$w=DtDx}Jzx_Z(_%qz_)(69r{luH^N{hYH+1g35END$2 z3%GI*UVbZw@6vDJVP+$~(K><3{snl|K)^VyUXL0z@ff;g2;bIx!pZ8V(c}hWa-qiT z-8p6U<6#r&daO>Lgo(n-WulNzlutReb0M=1H5_ zQyUjXu=T0L{Bpj9jMyLCRpVY6F#T zjG`y)4!mzUkO5$Q{Dec>#0H8jP3#Xq>Iq5i4Dr25iF0*yw+(`Q;Dc)en zINJdmEAD-^5m%w54{JC71h)q(d4+Esi2kR^U^MWYyWE{ZS1+r9n!CQ7bNV}a@6k8d z<*CBcnIlEU`*1Y%@iL<0xtQn(U#C0%%ixt%za*!3mD3ih?_8;640K*th4H}=#OmV| zfkv?$_l#HJ^+Xxl-iVIk7A`(+`)6Ssw|?ecs&ZO^V*!-{EiML(?gZ28)f5BwsWQ(r z%B2eE@Mu9i{ak;O^9#_110|AZ{-~APX2(HCj}i9Ri}GT2t>SX7b&~KKJUS-&1H5-%k2B>v!aDa3NR z77^U(AR^f5iRBa`wCDoG8Stzon>TVyoaSHH3nMRGcx``o*;>r&wEb|^ha7AZd>fGq zuz!vnPxP(|`My2`oC*SXi!~+bo6GUA&7s_8MZF#O-6)I3&8y<+YluQ#_z}47JCDqi z-2mG!@4=_0-ke`d4Anbs14qBvlDK+PNXSS)*Il|?wAyZ3?SCEoErGjriAS%GIDvgn zEVsboIlcAtE9|#=#T%WbLF_9|(&nO{HU+*PZS{ZIkvjK&YIP(6C)z2qVRundW79?D zzxaUA!S}qlS$oy1m0h#|42(rF#OnT z&uf`=#&)w7p+P=MJo_o7M5jGLz;ng%{F9z>Gr}UMp7I9HG{BvTsyM;~XH9OIi5+$P z>korTk2&A-vqd;a@O0^P~}aVq>tOay6&$Mujsk}`cC*+ZNI{C0gdLGUieZ#wYsDzrXya3zV$B+l&Vvuih2HwQX9d~kEVE}ML zjc&AcyaZ<>7e%&jD4~T@S97z+Po-Y)oL)OJgU8)mLA;7LQ!!r^`nw_zW{vpucg4pXbEl{dkN2^UgMr8Ind!tyWp+lSd!-}2hFFo zFhur=fKP})J$)yp+#t)nZZ0BX*EiEiY9>U(>nc3A%0=(qF}!bL3@aSb18_9XZF6X`zQRCg48z?@%}T6auY6ai&%X)=3G@ z@DB1Z=EDdG4v|36xHOg%lgQjhVpxYtCmyS~j8j^Y@T<2jo|L!*2W6t*nT`;9*YFGj zbIWjIhAH1q#GEf5WzElu?WW@sV(GM^2)YIep>@1Il;_Dosk=4*j;}snI8~8-+`(Adw8Z_Sffa0(c9MVW;ClA;&4}BH3{eBY)-QSEUA+xdk zrW?+9C1C1a>)_!EL-yZS6ZUJ5J~KPM3)B;BVC_O7@Q6nAz5Nd)zrO)Dp(E_Cy(f!| zpUToFM_`##5NfRoMZIq$GRAa`TZT*kM~I z4{x1J$l{D2fU8>}6x`kjlT6RxW6ltJj%(tj$$@x2eJ0+2IuV<#HnN@LXS32(JXZaz zoY-#dBUKJ1#=f4|`7#er;hqaQmSwTlq}&oHUp z4&1{9yAUohY_(h{n|&dO)w9iLdV2?!ORYk&1YveTQ0M(RH;RTDg3+?11kV5R9O|p5 z!1K$esoSaoYVcN^@7*ZG=LY{Ux1~Plw#FTEKU<;bn0D;Fco&Z)ox-M_S1~j;9J|Cr zuqX?dkIyIk;8}x8#o5e=3uo6&BiLEH3i`}%6IuV{2gx-mfS*#k;NZkiShhJA!mruF znCCo5`t}Jfw>^X2M^&)xV>ruji(r}gG3<%v0J==Pjp^yvarL-#I*SXXrGoj#(#i9{ zIm8G)$B4nhO-i_!$fHJr4D#c}`H}geeDMSc{`auAcw4s}pMJfHbqbOA#>fW`2%d4) z{J*e+U4@HXr64>-3RW+_M|I=g(nbqIl=>i!o|WoYYa@s?-GY8hB@8EhdJl&o6k^Bw zg7M;E_FcY@DSqr_O;f^H`q2m`d0;2oK6?PyDJ7zxRxm#4aYk{;1^Dg)#VQ+ZHe@%7 zNymC{suICnN96EXU>|IFkOslc2;O%^fqp4r_rfl-tG~+Fi-hrf;b{~3%Hz!WUw zpW+1JJJ5Kx3_1k6StA>+V_QfuF56m-Su5|bOr?kHq~|A=e=VE7DbJys6NTZ>e^21n zu5t+cr-o~MHTgQ<CXiib%UwBr^glzA~>pA{@XCy2>ZKcskL41Cfx zg-}IZeA};xlfK5HrIrI5eKw6P-<-(aWP9M&1L|0wr;WR|?x#8HV(AvKZaVAGc9!!p zmJRGlWO?^yvcIeC+2#3mOsgUQi-V0z3uJerJ*} z7VMhEwzpd`xknyM&=BLVAHo>kDut@c>-bl#&+uDcr1Pa0%)^|MCMZ(AP;f`253}X& zAQKfZ9lDXMR3??ZU7OEdKDdbo*A3#Uo5D= z@xNL%R1R>)zccLEkDJq&!U1!(<7FAuoGDFXpUtI9{1+lC_e6CCf&a#-WsG0p&LUc8 zGB@6LShcMHv;;fnejofS-YR5tlY)|Ri?HSxI{OIv zpG^skHR%HT{CiNcU@HDm5<{)@ztAITg_H0(1bnE0o5P*3q-`1-F_mL~AHGJLUa=QZ)(_2_#eG~IJ-@r~MPG=crCT!nlEq2@TBK}_V z6sw|d}F)_*cb7B7rTeMM0JiYmIT)W#nt zr{H~o)~+=z7wc}eV{35{{%Svi2Tq`a`x;R9K-Nt51-N~JC3W|-hT3D1iQ7^})N_%p5o6<>eE!2(-+>tre5Tj!wf%~3q) z`~p`Qy+_Hk0=yNo8qGzF@#FzTzQ^$~{8QZ${3)C&v#l~?^v!Ih@0*4?A2QHG;;Mkv zRDhk+te~ws5eym|;Jtt?sIB=4KXfx#uICZPuTEpXYh>}n?r~T`rwIHaX7XFht@-uU zgulo;3wdi}vA8A|D?dI(sl_#DU~v@(*Ue(hE2gq=!Whz}Y&_v_*RRBauVBYUL zkf>A#M+?s3r-_5OY58j;Z}uSHW+h4(EWsD6kKhiCW2hzHf+?w$(9!e|Dt26lklvdp zTo{8#MmOTd$ARp7{$>^|xtTqcx{SYNf-$IfJ8CGdU<13ISm&~NtT8Sa2fv5l%=;C1 zMfepIRW`%s5g~lsBE@EQO0fVBan>SqA0NFH=+Z?G@Sd|LG{7Ayw(T;V7V{fDihJ5&a#U1k6GaD()yo9hpjuy7(WeXolkapD7spQ?UETzlQZ4(P+@J zfW2)uVb=~wu?_X+%nqz^OHUky!bb{y)e_zO6w^g`hkNfa%UW&vuN?0<^R!>^|QkK;*u zp*@sn8`ZsMeBSTRjY=BGUJ)WYtBgYUYVVY$m6ny1k)miANs+QrBqJe7R>S;cV-ez z6(vy{mk)Gw_cSuYN0ZbroqVfC7L={N3kTbRu#ld_wzku#9MgvW^RK{_U7C0_)*QDP ztRYXXT_x?eXT$n;Cx}wrVX|_nB`ANhgv|@b!_3k75Z&+?oD5j>Hk^n~*{_Mahrl@> zw#1f`dN}9(F1VrY1A)=9CvF0c8Eg3;+7qJdV3WLS7=h>X_j=!UT->u*}`|M zQ2>*cv9Ngj0*p74!rj9LVa3W-^j6DOx+rcr9a$a-kQEGDGR*Ks^cP~_Kabp4&_&{N zeuBE(aEu!_61#;yJkuG8bRSyKc7)euT*mI&O@DSE&96IeWv#^7jIoOAyf$$8sA z)_t>wId*?B)7y|<@{rK@ejl7_7LQwMvhd-lSZMwE2ITyW@%Qp{VTLWjq=H!d^Nody z{nNp6-UIM48BVp_g&9XYncC(qh2c%ASZjO@>$Q-4CcaQp=7_~L*`)u;Io@ehIe(@7 z0LU)SfJ8HSY;RPfH7~zlfw4N3D_;uZHg&**al&sPdo*O+E07yBQeb<5j`9{ zFm~v9RN4`bQML1N^0Q?ydE9JFzN$i3JsFAHGm@}J*!#wpS%CY@ncxs{0sPg?u<=bU zew@FUZj(7Ju1#y)DaB~A<^a}yPa~!CrV)vH2r)Qx7EYy~gBKkp7^7@`gCEomSDwbAMViht^A#|1N=Va9&iqq$CmR!*t5wG^mmoO z_|=hMkd=z9_kwU*Wdsf!Qv?0p@%Y3)1)BzsM89oz7`gurrY=grI!PP$xeTETZ_EJu z)bAkPB*8ghPqA;BkoDVm5RJVafN=_gJz~Krw)P;@1Z4vw-3ckbCgAO6Cp>ZJ2(GI9 z!0V_Klfx5Cz`Xe-SX9cQ>~e3s*Dwc*6c=H9{6Q>SCI@#M4?((+vvnPr3zoG4yVA7* zf6Pk5ou%>Uw4e_|jpa#J^n89|WfRf(vJR_a=A*6cLG%DQ0Zo>|*YC!o+0UyurA~&T zqnH-eZG$i$PtflPg%cLV@HJNpz{HZ2OUc3^7dhB3RR$vhzw&cDD@pw(E37=PgS~5A z&{Ou6a4xTin?H2|TKB+$s(IkK=r%lh_!2w*4yRqN3e?|z2z^(+8dsgTi`!fNV4vWS zd-rt&eO@0&Tf*Kzo%SHSSJVrY*=I3jt2Vt6y@+~bN5YI3GGG^G3$O3hLzL}5XnDF4 zgXYBX>h{05`e)JXd&jwu|2UNSvUwZ(VVDZEfBYpL?Wx1Wzuv};83fXyeS06%96gQnAgrqFnIK9mGuR zPh=ixU&7OV#*C=`7JIt;D5*K^EzS$-W`@62apsE`;?`AG zFivueT{rzK_b+9BfN zw=2c7_pZUrllyJw?Vc?9A~>gaD8-31f2)b7$E%YgCG&Wf>@t3%mOY=OH=1+U@5rYf zc}MOPKCzuN=DGNe+6|aH%9LGFRKcEHnS`HQ>P18X%)P!nID6t-=I*((>}l0>eq4Db zcW8bJcWj?CIF-#{YaTeV17Ay6_b?Owr|EC@vu`a^_Q!%7h~39dukC^vorbKGb1%D1 zFNh2LPfrx}_LJ>{&-NVe7ARh8aa>%&IP+R>)7XXsaZIIu3;Otu6~B`o!ED%2jj~4# zSQ_w_i%WVbK0JLQSE8fOX_zK6S3OHa%_qEUl%(^yExE;_v!&{yOyOKZ`Kb}Ns^BSC z@U;~F^9d4v&s{2hSw5fKusR@~e(naN7rmKj@wODHt}+$3+ewj43oqE5wWi`bU+0o3 z+Dn+VlOD2h1}acgdxHOwbAlI5ih^-|4@A!K`eL7#!0r6&B{~^!T-4@`L}A=kZpA)z z-f#Xspj%#wJQXaNgC#=VVfbk=OqLTb8u=f#nd^zy9I{~k+|Lwy)oo}0es5&d)3RZ? z;uY@a9}`kOrGm3oPvEYo?oI zbvD9Fm^d|@`Bylj~a~SC5WZ-^xUtD%=F2#=Z}Apm$9zB+ zc!Xp8%aQSQr(igV=1UpT3srf7ITdYu^{8RwP->7dh{jKr zqCL5-I4S29=1#u?bNzLozd4v>2p+S+f$~)KY!~uH7s#bMC&|8DWn{gVBOQKa44u_v zN*hlpV6I99sENYh=R;MxZ`4nGq525jJdvUw&^NJTXyu+knChv3&vZV(zPUT$rLzlI z@y@U$Q`o6=N$}rSLp*1e1*Lsk6ApRE5#(DXux$czP{n+ZneGITfU(BtEW(!)J+ z@u&d$+40@Uk15xo% ze&Z{ja^n}@xLlWfk8O-tFbg&2JBxN!1JzfC~$uVL&w~O)P{q&aLzL5D3gHQ%q1|txrccA4T7t+ z|A=DlAKq}*L4NtJSkN|&AUQSbhQS(IO#cak3YZ#_ptP!sNSA2PP1f3Rn1Q!m< zMQ!&koE3GKFQiM!Rz*cpb=~uE6MvBFN$q5cixr&ta{+c#Z-?La zC&K=|Rj}LgEaU~h0`*%LaD3@3G(8i5;y(-UVTCKo8C&73xFpydbsj#PmdEK<4@j2z zPSU@B1L-_73l1zbg@db>!b}~W*3IJR5f5Po9(0YcpL|K6fejgPz7H?H5*Q=fjhN_P zOYF`GdG=on{7KS+rYrYI<83ooZxILS1}tnms)Qx0-RPoaD`}HtF+K7sig1}iUhVli z!3M5~V|YD$cu0d@KF|R*2cLsRk}{?QUq#oJO4NH_iZk6c$fc?yywMQK$L_udwQD0zkqG>6RtjpPjz3qxBs!8p&|uv~b^*WZv(7i*RponDLi z?!vxb{SodjZ2+BKV^CRZ2Ro9CsjgV~teiEeRzL(Vz5W7k5ne*C`*;W$TjB3?Z%jg8Z_ z&@VELx6Ui&oNmdI9^C+V{cJ7RR)#}nvpQT4oB(xii{V#s8O*A7g)tseLDXUVM>a$ixV_{%Isk`!T=&^DuNWFhS?_E?78q2-&l; znBQNs6YWP?!y!Lw7(cNP_nXwi(`bPUFnk40UlPGgn5Cr5I`P7{KXih$w#39!U1BnA zKK^>V0*z)GWBj!PICe`GR&5c-_iRs`z0U)+wKrh-<#V`daSYyX3&yS-ACj%;b{DCEA6zR%q%jpCP0?j}#3nvuZ)=lI3n zcSGb0BUm;;8!VeHk)K)VBsW3^ivElOrLtI(8Z?s}eyavI-~J`GqggNr&m(~g>PYy6 z<&bG)3AL3YNO))+f4K8C`tSaL7jr)geE4<#w89-eq-itxxoIMo)1uCeWybPzR{3MZ zPAO!JpTgm^q1c$c8nQQtq2FW?srWsa_`RA;+|G_8jT;n+=?o+xE>d{5Z7K8)Uk45E zZV}6zy+kVQ71LCGr%nYtq6GHwhtIHl-1( zp4Y^$DG8poUW2t=OYz#v-7tK677Vj^3@MYZ@Qq`B@ek@k1XouG{)Ihw^>QNiFI$5l zx-)Qhu#>>M+7Xesh2LE(D1AGvV2R%;VxzNxjE!l4lSvc+hr^p$>Nwft8Z?yULB)%; z@U-It88+b?nO_h|k z-Z_rElK;wYGoD1WvZZlDvI+#M9U{9|hhfr8isvhbqmRixKA8+6bw92Wr7!yAr}9SP z=TJ`8YF6_xHXByG`0$)FXnB!H^H*TGl# zjhl3dg3}bzy1S9jY;)m%#;@cH1Ek?%Sq~Xh^_7@TIw5$4ve9-)DyF>}1b)pE1^?hz zzQTJqpJAlKuYTam>%HtGH>-Qe`_+a(PwJ4n;!eIdAfLGWmVx%|Kgep`*>HL7Hdr$I z9`Sm0oJ_TGB0-1d5{HBoaFUC|Z(I~6?nuL4#f`9DW;;wvJ_nB@-5?=)J)9}AN25Qt ziB3TnX}U6+Warw$Aki|AGQ0^5!={kWy_3ndx$fli%m^}N<7iT6$Pmj}1>nB)B7xdQ za(74*%B)w$d8dB!OFa_#=$)&`c%No|gy1i^r8$UmKU>Ph*NsIJn+dRb$z)Dcc7!ZC z{fF~=xPj{~isy^e53m=P&S83fbiukk)8McFB+h`&L0Rh|IAL?HaNYNl5Qk>|O5p^y zsjC4z3d6|HX)Jg4WISYzI6=nzlV&5AM3R*YB1xW8Cbw^BG|sxBjqxduIdvaxynW{m z&_~&j+NH(ETqxi#ue`t+ecDd)J6^Nn7wR*qYG-g>S1lYEcZh#zqCiUvN>J}p9tqgK zpYWT3*pzWxhF2@$I-c;uZjR+5jt-#)>U!9%6~(taKL%2p6v;TB3!LhsAaY;pId^2& z5w2)pFXVB5_)?WnKFqut47d-3RZnM+40}(`ly$HsbxHi+%ssH6`6>UxR*$?gJVsnM zrSdoKR`Moy;<+!k9oe-(`P^p}3Q1S8A;!J}I@PYixE;sADdz*ZVEl_baASyR>IN?P zStg`cSdk508`z~QnlaL_6ho@zgbAS!RlWl{toV=r&?30VY<>~j9|qjY!8_oto(JjK zroru0_akSl^Z3iTvLgO!1{t_ApU-%@nU|=o!gI6Z!Dg8fk+=UyW=@hK*%Mc?vD=Jj z=)UoIbGIKaeeWP_o}WmX4!jbdj#CjbV{_sDmIGXa!Y5dyIvCQ@^!alELY^^wBAMYS z56q8p*t5V6>ePL?wqZ8#(EAa$>gEsLd3*!e@a`#pW=}8|^7b4U+592DRkO&E z`9QY+ab)%ioW)(;RMfp#Lc%_+!}oI{;L=W zq-z7Kte%mk|N2SOsQ<`xuXmhvY&x$prvaAdkAmZYlUchMQ@q{r5DMi}h_YKO=H!pU z@=+B;TcsAwc5XrK5EZcf?>e!Sgp%#{OZkTpub@cF3!Zm|klo|vlMtPqe37{UzyIM2 zGH8MvS2aYJ+Z(wNi;fS*qULT=;row7Jx%AgG`$lC`Y!{Q=0sj2s)`FcG7VJdWq!%3 zH>@bkkjfr3IjEQ zZ6hlFJu~^!1>U~Lms_73#H9^?22Z4!-yFNxg%=ve6VC3oc(P|=C~&?D}OJAQ5%)WA3bl@_fs-IKQ9Pegmb7M%emab zx|t;Br5c98&u%5J~jbt}J+r)NNi)P5$X2*OY}XHxQjvA{IM0P z0(X9y+i6!yoRZz)OO6)1qbHnSsk(ujd~e2E#?+H}L2Ds-dp}$5Y{LfBIg@|Kni=MO z8XrG-6FK2f&!A2PfA!S>ueHUQSKX1oH~!Ruoq4US>CnsktavSOCxzS$=@N2gY$V(% zS==#1C*uMfXY7sRZcunaH=sDZ`>E z58!RkBBpA^UH;allcY~Jh-sKy$%^YHkXbdeIf+&|d$i7iG@fc>b#F{3d503oT+deynSH;eoSSH|m_OHbg%O&`#INXl&QY(Hbu_!ghmA@W z&R-3Ak6HEP*4R|I^ld3~_}pD?sFW?CDPtQb=W$zgCiDG9d&x`R<;=DH+}rE zVK(bywv1Q0CJh=(I=Hk!U8FXv2YTp5CRX`BK0A0TnRw?0+dNT$9Qlw-Ca*ok`gd!S zb7wA+WWkYb;w{G>k(DK>%Wb%Me`l}{14ocoi$1XqtFMy_3zmZ!>&@x9>XUQpzY(WJ zqnPoQ^1R#LZRGOGsa%6qI?0MOg0t-$cPCpyE@!L4y2=P4%QJ?5u(Fu62VP)Q;$E_H z!xH#`>fn6m9Xy-+m5I2S#?SNokI0#8 zu?izLk-0mip>gtMwnAzP@7TMufa<_yj z;g9kWt9kBmv?@7saV9XE^SIp)X7l>aPe^O%ORnqWBFOvE3u^lx^L0-OV5{s_p?fHc zz5L`N=QDi>@px^Q^`7R9nt*C%kl zTZKiz6&dcD(^1w(UzV$UtHAE$ayXyKGkH0YCzm+gn4h)DgP*CW%*@DXVcItiFpW3d z*uQ-<*rR`~*|#4)iyWfN#YbQEvSa(4#R0MVn5mCXF($upDhrMYRDOjdJI*E7Mum$=BI?iqYK1 zZ0Y7c)G`=fTmB?rZ zliE-@e(RsVWSdPW{VMH?w2N{k2-;>V<1WeUcaCbnb@n^SrnX8czJoP1pFn zV_aBQF~j;D2xYtLyjbP<7Ou}*kuUmui`l*NHB+)Ko{yh@iuq>pT9o8Ggw3ki#-(hP z=f|JE#-&bu$X6J!&I2$Sa#$Y`=nAYIK)< z+T_FD-51J9A~%yMDmt+5odU>RtPzuE&%~V)9j-`cKEHi_4gcHY3O7wvgHswcoUgih zgxeuJS<|O`3cHeNFl5+X+vgLhXi+QAw3HQz=UaA)@1N1yvls+T<~5B~OkFfsrrV#P6C0F*_ZS zn0xy5%;`ZQHtpOauJY+)Mst$|r{;QyAM^AnWAor1+dCqGf7lwvTDqih@4X$l=_;+P zc3VG}XL_FR9Q%SxTv5p{7$@cfq~o~5y;^+k8y~J#$)5E+m(C2B9B1u6trB;v&ScJ? z$P&MdFy^!`WpQES=5nf?25i9hBKE55U3Pt}39~rzps1mGF_+z9$!|B7BOTlCk$Vz5 zzG6B{e%cS_4r;Cz*Y3z<3}0l3Wpl2Je_e58?CkQGpYlU_P35s`` z3ZK8goVY0Y^7S)RNzt2hGV4esKV;ydxQHoY$n|8lOLGKSs-I5!&dQKcAsIX;__!B& zC6K@OlUSv-irk3oC@#4%fEGRqq*flG^z2PF37Ib^xw%G4B3}?pepGSfY)deae)APl zcN<`+rYp+KO~Ql|t5M_00bI-$!CKq`y@u)VyLmiBu3819dlMjfTO`Qs*bm2>B0)Mr z13w7oJCbKYu6>8V?F)QFce?@F{WpQO&9k7lCYsYK(P1H9l7+rW!hh)jh}u>L{V(Ow z=WRCT@Ig4|v%uOMzCi=6uTiBrm#I(51oA*VfV`~^C!c+Eah`{;GiEq+wd$aj-=5If z?GI?l`&XFmb_P$6jmP6NszB^g1GybF@F;gV#5g;E?9+{qW@91AS2mY?I6YRUgt zu2|VDaQQp&@0LaAH^LvEJG>_HYrIJ2OK{{h8~8T4<=DLTLRG=0D4 z6Fiq~htVb9Vcplcl552dl9E^|@s^Fj>^E`Ps2hS=QBTph|1R3O-^RT+IVvS%OcyN> z(N}NB;-n2D(IG<#H`|Pds(}%ZW@HL-xr=eWUjqKHP^K9R=HjyNGtr@c1Afnai68b> zW2Sj6&a4k5ezMDn()2XqcP1G-mL}rMfe<|M#)0mQwWY4iXd39b40n4k#RIEd@!{bE zXgUENRO_79_XwyM+hqZH_IeRKRdL>U!RK7iR*Yg)6ynd+W&qDyO9;qK+% z5WaC7UJzKg!Jp3KbE8)nbF_$#jyyrtZ)DTpYl17gjbQ8-#IB6JxaFz9Cv}EmqTLLf zQzXV4MiPMy7T%ZNy+8;38uV>6rpYr#QjKP#jLhN0w$ z(rxm~a4x?8ITI`IL}As+l|nb*e|X<&2cGf@po&QgsnW^$G+|>6dS#zOtHyjBkyQle zh0cmDg+nl=q8=)Q-oE?(O%P>n3me}~f^#1mp+3nEpZV>>TE$9qc$1Hp94`sIXZbk2 zVmOX=W^t8)0I>d_`o<@GM1szV!Q07F}NBUGKDgn!~> z@ONG{N_$nH{OTXrHn3Dszb;0r!nG(W*Tyj+Z(-LzHSF{_2CM%3Cf-F7JYKpCn@@d$ z+uye1=baqUYHj0nCRy?O>~)B{ei+%UD2wLzj*z{hzwvnUHVnIMK<0ayb90uaV%qQK`Ii@<@Khoj~Y_P=K2dhVAoOQ-Mi2Ruji_-6qldUPd_T$f>oO}pgZ`FpEHIoQk zB``J9q+pM?IzMsNBlha@C6H^d6P5+6Ar8xnh~gl`*6cJ0*C^n89ruxYV>QUaiyL{D z#e%4q9+w2V^x6(dRXMm^UR3 zXY;sp!+rduss>OWw+T|i{7HaiB&4@W!Mkudvizbm*<8GVKkm4mtdj8{e$@rs3X|pB z4HGNWICTIHeGlSvj$VXz?jMmi-47b2bK!ug1^E|l4*g#=!C`bh@wF|3-0&R!mSP(_ z*SUayLt}Zxg;D%&#}o*gDR|mO?;uZQE&|#NA~OwBI1L&~mibRWsX4aL$lnFlK$ZkM z)N!i%F63@D3wtK_@D9@``~G`7(4TRzDQ`JW4Ud7_N`d6%xDb+6)=!qytRzSOH1o4^ z3yEP-I&}K$(*oDi{G(kn_=bW?q7XM9vR`Z>c{#?s)x#&;hfJZv=501v9IcN#7e>RF zBFb+GOb4s2>15aKv&4SX39xfLOS~RW$!a4 z9zXIxE0zB=D42VCdjuqkP7-DFHRQ*7J*fQrmUkcQg4}U$-obn|`BUMA(yL`LId>Nr zJ<@`Hrwinals3unFoomqx_L0w<-0QE;B=aV=ogLQ52<>>&V^&~!2Uf@+nLJNPhW_o zcVEIC%NRbgVk+@kHi?Z3%plEeGxnDZgq?6S%&hBzcb=Uld{p zSAKfJ$o>gr-Ko`lri~hzA}Sz7F)G+E8NwNA`*QI)u|#i=h&L-)!bf~DK-pD!q-ufS z%>SH7DtI}Z6EY9nb&7~xO%GWSE<+~TDf27)>$ueVLyT2cKAU()1&-F|lJ=u}$sWf@ za;VV&zDM`MgLHXfYO4cF!4iI4zf?n=V;8e(Tc(nB} z`M6LE;yteNZnvg^_3RZeOX3Y~f^Q%+w1{*?DwDX~2l=tK`1qnRjg;&D2lVQte z@@<2L&_N&X(!?tVsp@xKdU-&TUKBq<^?4^?jcEfZJbay;n5K#F-5LwC zr(kO##r}1RaOP(Ts)*lV=hHAO4j6_m`#ouqkS!V=dzM}|v=yF3NN@jVMPJWr0hJ;R z$SWNS?kfh-#aUK#n6j8U47fp><{S)^bYkDvePrbM1b+R(H+(5+#;^(gSnM$vXYJ~t zT$`Gta>^)4N$CX|)Nzl7->s#WsWmZ1OiVD;W>t_$_q^!N{bn>3KjE*^9&*1)2jU;gKuh)otiCOU z*zf^7rhTLlRR`(hm0mRYw=rL0?MhzUQHAz@O1#=VDQ^4ueZ0FzCULEQL@tl1C)J*( zVS2wk-t?@*cLHB~>!LJWnDGO@HB6_iuY&2yxiPf+vl%8WSb$4$7oKV#Df!mNN&4ea zvTDUN$)!{eNwPNsExa7-P88zv>ua(2)hRHLIR|@PW9f>T zAR68=n;tVOM!&*~xO&ks^uB&p^6mId$**PAl2y&m(KM(L8K=YecULmz9qGXDTSrg> zvtzWoBSYwwOQh>xOeZHk?dFH9jUzJ2Lf(B!B&tPDK=$)$S~HM~Gt?QJ-k^*xk_>Pw z6*9>-OX#tH5ZW;(m|BF?K~PdGOwAQI_!n#Nvha6!O#v>wb($_(TSIMr+@V4NA2ykG zU}0VfcFKt*diO0P-DqN4#KY@xhlBsrBG#z~E z5&tUJ0W^%{Fvi!DMCjEL_YLY`BxHxLQZHJew1zHgi6!Ox^`UU%69_#v0Y)vz0@FY> z%*13I6;loCYDW>p&EK(U{zk+wZG4z%O8Z~~b^V<|)8y1~qQIvs-VtHI-zwa6IRif) zs>CDDCee%reLDEQ4xJ+}^jCeV$K~z+aMR~P{GO^qC5sulIp8k$A!i16>v{oSTF2tH z0%a1iAcgEbF`Opwl{jUXH%@$50%bNY_z7jZh?)2zCPt2?J7fIm@5cw|)v8_eueA?- zyL}r8txo2iUOmLG^D&}!OV49^o{%Nj*b53h%@DNuHfT93)2iv)d|gK=FP~(=uByp zdZ>yXV}$Ol^DIb>oB`h%E22MA0`1v(FeRrA*6(R28*U84tP|@of6^F?(^AH}j+yZ6 zYB(0VY{c(z3ozV66Njx1z*?QFICdq6o4N<1>&(Say=^=!FZc}E!&l?%8V#D3XGs@u zgo?!9$UtHPahwp(Hwuh>W0O4o(ccHs=ND4n5fRk=jz6W5j{y%I#RHpd>5$pq1Ya~E zZ$I0D-^h3HD+Tbv$`0t8=T1k=O`$Sg33LjZg0Z)yY1*$WG^+j$nocq3Y> z{Q}(WmX9fyuanl%5?ET=Mt(YR_#)j3jZ=KVbV4xoOA4hoTrXnD=)JHgHW3vH)M(&; zV~H1DC!vd8k*LQ9aGAnTA#>_WtA|Elo$+YgxA!RSihK?0Ol;8r%R%kmW1Lmgg6muA zP+jPFnSVlmZ^A7NuCaVAx||N=D}kpzQ_mFl|u*RF>C6XH+q0xmKfhjwxNfWfFCEjOI`MmrPQoJR+i9!4Tv>9dFkj z!hM-e_<7W39J=~Az8z4gQQ62KgEJ z(9kwk;CBk(Ub8xWOaL6QTX;t}@D6lM12O1FI5*z!Ea^-6LrUC&1(r_W8BvpZDlLPr z9dTr2ZX4;|R1Gf!6mZ+7?RazYd|Xf%g}sNoA;ZlQoj)!>?fP7Byq1rN6HBn=L<#DK z^b##)Q_u?OBJRSoGgit5MSG;t?D;sfwqJvn?keNZ&&MFIdoQ|eZWK7vRVa1$Bc6Lc zkM7WYM&=jjz|nROa(8P#gnfSrHg-SA@B^vv?137hTQT_S-vsZejX-;jLe0v(=(OJn z$+VS1_qx!HaPSij+jtCJ-uMw4%NNAZ*O2rW9e^dZR>C|o6mKR>B~Jz<;H)K!exXt{ zC8-zfL?(1=%Vbd4;tN57A7bO5n{brgf!%vE48F={=3jQrZd!;ByyuKdX&A#&w zKF+~I*QcPbbR|(U%7%bPBXPLa2DIB5fy2MKqNeT!qCImsJ+nQLcBI{=BUbideUm1& zxm}7DU1IvLXeQmEJrQ*kkC5NG2Z`i-9@(_%8rGP+#=?q7%y9F7D}(A_=y!drWRF0i z^>EDU)4*Wm%Y0n+T&`C$m}?(Uz~gt!S?LFU+~nz2aO6>^c<*ReX0Uz)7k--;&%c*0 zUSGKc(l5WWJy)+K-uUVO62QZ1~Ul*SH(7$b-?%gLa}N19kG;)4cV@GQ#9fqCrS({<44*zaGFnE zFcPP==+t;nWZ+dOUe>i0T1HIZA}R>y;(wd`+ct>{jEWI2863?#l!>!>H+87ZFoAX1 zC5~n;+r~3LDh6=D;akj#)$hf(2djePYgJMGV9wS(_ay(!a;lhFGMDiv%Y-QhwlIZU zD|03wn17*ghFLdPnq@;~bAe?)nD;7K;tMC;h+E)6HY`Yv<&v1Jfq{)a^t@w{Zlc?!Jn-=36Z$9znGu8@P*C$K+ZmSo=pQg!)E!%4DRLC!QvsG zjKnetR*<@K90Rjmm@7#d#6?SoH4LU=lU#zueqrqG?P4w<*_%MKmrZ`(ESopClrXcL z+G^RIv|Z5c4_gMR(tzvk8$`DszYu?)Pz{GblTC;m!`2l%Moo`cai6$Bq*8VUGfSJ9o?UagGB4YL^Q0!YV*4|Gi24Tjn(*^`)4d zHIM{^x$mNQrp@VCsDnOE|a;X zm3<+Up09BdJP?vZ+xr)kZI`BAAn#x87hA?(6?I4s(axs_=>~-a8rwZtvi&AV^65Cq zYdlNRt+tVS>t+*&p|_ylt_Ko?+~UHJa%$gJLW5jNY4vqaJfrA?aXG6n{>w}#`Zou3 z=nA;j@D7d_w8H+Z8_;xXA9kjP;EwKa+|)h^rzlD=$;brvxQwRxCL`&Y@S)T*&V(kn zjG&*5Evd=}7M4gmgJbJV7@hJAuSPYY@|a%$)wJ5Wok*3zDVrPI>I%a77^bkHfl7L6m zQ?QsjD>-JKE$pAOB(2Yn(8H3WbXey>Is)_Y`Yz!;_;w~vu-ik8zpbQR7W3#3fv2Ef zN5R}{_E73ujEZmwuYEg)%0ab_m9Palb)(pI>9kS$)bn8nKlteK}mdk4OuV@nVESWm+D9b0gDr#pTh1k}paf|?86^;+-qaF$9WYCYYJ zv#sYz?8i=(tTb|yT>Fwtr=3WqkLRUO+jM%YCnGJ(xW z&DNuLTMen=FdbUEk;fW%A z>$V3!w;aTt$wByP%{KHNg|tTAo(f0t^r5yMeu7ogVwC#C3wYX_Px8)epH8<7iAO}xedSMkd1nj}Q z2io-JCspd>q)4T^OsSEQCe@y2OgBiKAWQyBAnW!&Cssm+{#dGn&iX(o@lZqWj=^a3 zT^-vawo>)|E2)&?3L)RFijD)fLG1kqPBmK4?j~bO{Y`10O(Vs@CG^QaCf!``h;nZg z@Ic2~ zk=C+&;+qCOG-nw0Pu9Ui952i>JvegVW0V?MiMb^Pq*v%l8eL%p14493r{9J)ofaB( z%+qMawRmdXnM~sojLD@iQ?h<4OWp{J#imPBX#HFdioL?`V@L(&Tqs7GIEgy+NGK$j z&^_J-_YN40()v0{FId%IzoX!xu z8fI_z;G4Ihn4mcUXO%kPjBXAW<*L!y|2|>w+X~e0Yr^S1wRmjtEqwg=1=jEUf+D*{ zY_8R#>!Y>lmfPy|%}WnlKE_aRfX_frO$E|9(9K7d{NN2GR%o!;1myyZP{lWtR@HB( zGy7N2pZ|3Wd7?|O&hH}#i(4@A)(5G@UBvXdkn4;X2eMQC&seU*6KB1!xOfIWz4#u~ z?p_AjxN>;*#1LhZc`SZD9up=>(`dKfI3;ip%@0z8Nnfs#GNtpx^+gWKeGkRnH3xBH zMG1NqR$*se1wPB2N0ts~63e|BWaJxRPjbuz2cor6$G8NVzubnf7kOY9ZAE80DpH3p zLuh)ZG%ox25n^YnVzNGm!cd6$6NPSqk5eU{NmC?499<->b0^cW2gcK2)|gJ}hr+_i%vs+gy(0m_{6S;0!M31ZP~DIbJXsiGR#TfffBnj^64c zo2I;k;ty>QqIwL*WjWv!%`tfNBf*E?&(dXF9^LZh99=OpoCbXHp{L7c(6L{J)5b^* z8a>B|{`=4fEwA#yC@m6FdjG)Qc`rZ-J0NAl6*}f&1zm8hfL@zaOzsb=REX+QMT55@)ZLI<#(DHT^K(I?w1>0hN4*m`-s&@W#c(wi-)4| zQd!jQIE-#h$+%T(@pAyX;_YiZ1 zYo`@-85`r{@St7 z33#YB9rgZxpt&OX)J`{?{zz%ST^al{w|Ig=M;Ds=){{o-jHCaYP1)NS9IGE?&2%OB zocFIsFl}!y93GHo^IPSa%u0DCoR^PRgb8R|U5WJ(mub(JKze284cfSCG@KV7rsqEQ z(qOaWFvHjf?iMh(mvaPTVm9E)>_{wpvj_LcZNsD)j=1l}aq?Q{GMAorms{n-&(bRC z;q6Bgah4&$oPVZhu?ld?7iYL;XapPbWk7aBJlu`+gni@OA^Q|z1sen`LukyHaRY@O zF$k4>O~-VJuwBI?+5D>CD77vJ*C!_9>#s@J`}r$MJsieg?*GvHcLkkdkwKr9PJ-R< zo>GB_0z}TJq-txI;Pd}h;rHoFahaGrI5@OX=Z*?0GiDy&kMzc;LG~D>vkxt{Zoo)O zA54<|2J24TfDrRS@LH3`4o7A)wdxc$zvdLaZwSMc8|I^Dd;&HP9>=oJ%klZ*H2QEl z@0t}^N)-~@AmjQjko;u|P4OS_w{;6H+t9)Hziwk6&tKB_%SOf9DR{ME8b0GQ_X^Ij zsJJfx%ksBiW}P|7+g3w9{~As7oFBmMS?!>A;u$3W4Q3A$eVBjdT6SYW8=PHp6AZaH zcs6Ae)b@8#naoTYm1+ikIpblRKoN>_e0k2vBK$LLHmXe-52`0RsOYEPH0E?EM(`pD zb<5>ApLed#I%CXwtTcHJ*(c8I&l+;8HJN0bOh)Nr9yn}H&^^Nwb@zFoqvd>bm9Zf* zNyA)CLMC@BA_2MDWK=vEi!1lUvaxer8Mj4^xtpsp-MRr(yYvjD&H(niO~p38N8ixu zjMrn#P(EflYE@bax89s4l<%7=KU__Ma=!n4VGk5%XTIog=uPLkZn8`-gE!hkmOZ%_-!ov{x|_Q7p7v2 zc?wpHslwK2skkFO0keW~alM-h`Yw@2JM|NUZWNNGO>en(Dfgja)j41*6JfAh1)G;t z!N>5;Aj9#F=Vj8^{`C|5>S@EZ$)~Ysr6(?J8i}0&BXFHm3zS!vVY+)hj@g=yQO4i+ zK7K1yHob$Og;%gw`ZCXLd&JizhOkxg8gBUzk4>K9I5^uMvJNNHBjXfM_NXy#;%hzu z>&$r09>0%2p*Vl|Gz?F^3DFLpU}=yn*hrVs_Bl~>d*WH#=W+mJuDM|CY-Nx=dX|3s zq(*ZJrFjmDCQIfS&Bd9&>5d&y^mo*5+T<_CRwXI3uZtAeEW-D$XD-7nydUwb?*i(j zuR^1w2Z-+kE&ON0*EKhvgPW$BSmz{#HCtan_4Z5nyyYrp%2nX(Pe)KWb{FP6Ux4;& z4qJ;&d|=HjQRjx0o8bvJ1p8j`JoLl!F|d%}J~#Si<`X6xOML z{~cEQuQ9YdFc}?b_*Wmq~V^H)^4rhA~g1$%>=;xKuSn(YC-{vx6;xQJ( zXKG??$yaE+$lEQZ<>Jcid>$)zHjMk)N$=iXO~qZ)+3nMDY}=V|_Kw%^g%x(d?=K6% zW3dX>>Ry2Q!^hx#nh#8o-wRQ{MuD-760EKmrZ*(Spr`R4?%gno*@%o_Du2#k>Hau8 zc>e^Rs9eRSWe2f6El1d#DN@4iyT7xr9a6#`eLot%D1@H6bd|2Y%J9DLY8*Ly0PkHr z!cJ=*Wer7%i`_Wc291m_iiT7V5V{1b?{_H5hfufCIVY>$EZ>YlO_;Oqk`~=@th%?jWL2SZ{ zI5zJ7CARTCLA{+*@J#P)tlU&XXDoY4ErxGV2i`}&=+t4PjSF$|V+|%!s=>TG#;}2y ziD|>B`-1?jDs8b|&IfPVzF@N2FPXj64>n_tEH*S6 z;NLO}JUkR5v~7tNme@xNMJwK5kb)?G59&kr$_7*~zJd$#%W;adINlnN!$o!`XgG5i zn)thm-;ys7>M4e^o+zT+@?Q{5L|Efu6*fz02J_Id#7|GW(D>+DGz?G1yyhhQaOOOE zf(ROI;N4>CqtIFW0iM{BhpPF7*t64@<=tG&Hg@|m>DV_I+wlm!E$VUTvkS{jTh2`4 zJ(#2QSe90)&6HbInO6;ma(N4I`Zp(Z=Vx>tSwF^8Gpg{IN+X7+Tz3+YV6Lz@ytHgf{8sn1Ccu(fXb_S&`#82 zCDf9IOU`G+cL&S2+{#X-?_=fbqu{I36)3a$47Ydg#ND~R*jS&9q$3ayX&gs?xon&| zOPW0?`h%mNzC(@OJlnI(h)q2_o|SK#OhdxGsMbt(I{l>^mep^>)wL-&%-?t9N)oYe zVIih0%!X>EW=OJVfQc;&nV@tk%inL$&c0j9wtZW|d@_$Q)6?E`>(m08;y+Ax{!nGY z_quG7jXqnE5Q;wTCva*;DxTfU^Sid0GSx~mW->1cH@`cN$NtN}wz0hDV&WN||8fQB zi$?GlALRR4-SAZ;0M`CXfjZa!;E3yK%;-d(i`a&$x^FGObuC^_&H0_p)gGk7;=83atjF3LD??&FiJrgj7?Pv_ZpGM}(qRR+t&5S z5+qM9gOFc;;KWgFJTDcFx2kN>Njwx6xeQ_7jfXgSg)d%;D#Q_w?b#-eEEZ|kgn#?~ z!(Y08G3bpy-ubx#$>t9D5&a#v{ZeC_O&GH_I*pV0Z{pl^U8W_m0S=vRq>=h5)M4~9 z40Ak#k#A+tElO2rt8OnWSnVQQ9B0gyC)%=$r-PZiXa@YgKw$^521%JwOxd*$-yaFa zgenK2UEEk<)7cj0KU+#DVj>}2qx67n{QCq$G7~WR#e5v`Dui9VzMEB-IWyU?72vE@ zK=lqJ)9p$U%r@{C?zLA!v+r`kiNZnl;d%iR&7KHui%jX(#|!;x$7*yA1?lbV7d3e{L~>Lq^IN7#n@L2QQY8Rn%HfwPA_P*>=N z!S8>tRgDtDDl-Y8?Y9zUQY$VrlNl=wZ2dp>;i(-P%stNLW|T8p=LwPFKr4KGNvv`Y zqP;B3;u#uM4x@4K#|Nl#RKzu%?!3pLpB_*YgW0uuEbhxZ<~u!*=@)#avd06VKvEYI z6i?xI9z3(oZaf60HsSmQ*+`ua_eG;(SjSGt$eWxX;coV^aIRA+Zw2X)jga{PBFc@A_73lGR`fn?l$N z!Lz7<9D-gsbb}#pVh;aO(x|EJ>QVK84k-DP+>k1#Dz$2wOf-&wjPNWLrNM zu(+MenWo}J{Q2Yu*rRDi{|Bm;4q5pFU0$8 z`Dl)NU=8d8?YzBEGF}p&o|eKXAq-6xnF{Z((-W4g=7g^Qcm|Q<7yLzy*r{K-xb4<* z^s-9Fk40rjw4UPck6P@=IaMH&pV1iEZ}bPxelfY`jE^^8;Q5>U-q&T#&ecb;zXry* zK~o3Wxe2IdZ9qqkm85A6gubrgy>-5UsMfj;A5DFMw_I+Z?B)ji{=EuD29!de<9nFo z=Z(%wXW{!%HmLV70H0pg#XKvXm+e^3>Rc}~|Fjz{;#ne=zCDkbo?r3vFS=xhLM@q*o=l8 zR+(MJ_igrKLHlgfySx~;rv_sEw0U^W(FJcDUxR~=s|VwJREf|1D#%9 zz#{&fIPItvUbwIXeKX53O{;*f`(44Z9UpP@v6t8t^Oo;HEP#TK&D3a39j#vY2U}X7 z;55F@yEiQXHoS@k>G3fzUr~>Z;@|s;6u%a%y|5nN`tVGj6@GZI zS&jYu^#=PMHR4<+BRW1zohF_nw5LZGXX{&{*xkKYc1Ib<=}y2PBOg5TGa7t;hQgH0 zB2Y6sjn>~!V5@vI=I>Ev^{3U@Po5>67IlJJs~6L??W4g`G!9d5oyL!XSGZdC2!1f$ zhaYc8V8JCZ*0l092K;VA0|^PV?G-^M4`nRwTL%ta>mj5l1_oY*zzyX!aB9(Ehz+#j zS!XfeWB3)WqZh``3&Xq4udwoaC7Lda#s1B~xZY3%=b9}>4cT%ubO>Q_UWeH7@S{xh zMgh8X^SzwTdl=_bh;eGecu{;AlXj|x%$p^!sIvx?^AoZC+(nG3lwi`u9;i2BF*->F zU=Ht8wJ}(VYOYBr9w1H!9JYg@{bGpRs|ZK$%fh{kc;| zT8tcU9YJ;dR3TEhNw6S5lEhBmN&4qE39Rz$$(W)`+{K$UoX>W3ICNhU@+=dHq^%hC zrOlx`e18g<)JPH@yOe(a`JDdeOJUgS0K9o8M=nRtfVEGYh-+4c)ii+$=OWz0+57ua zs|R)TuhSW-T@_BO#@CSQXmeQhyH-$TvVn|ve2jkInJCzHCW^$@8`GhnaN^!wLZl0S z(8#mnxn;Z0lY!AlhxRWZxw;_``ENbE6kaCW!2lQ?;R6c`l*y5~ZM1e=Ijt^m6UZ%4 zrsEXc=nfru+Hr6;76hk(cS{~=%-Rn+$6tZ=pZ^4h_GQwS(M7%F0x8f6tY@j$5+8iTJIaf)BX$0x7nF?b)kCFX#Bj~Q> zm2`Vp88@oZg8Ipiq{c_WxO&}K>b2+xz2l;6ZDFEIjOD|LxpCacoUaRw9B z$lZ$jWQtG@cJt?VPoWBJ+4~+I=tRQCnsUz9>zAOkx``-EQzny}H-L`%RBiOw( zh@?J#LcC)%NHw<-Bu>Qwjelw#A@|$*Z&kHG5Sl?=R)i8SF$ybsX3<+o&Lp>wBko_r zh{=;l#3aO%yngRY^5Vldr}X1wKa-}`1#V!x>jynBVIFaFD5ctJ09y~Ha-WCylJ8+a zO#a*?JGNAiwU|ULWiUqzZ*DN8gTtpjii2(fF{o0O?`GJ(>jlFU?`l*nQW7!hm95!k>E^j zrMG|@2v-xkiHcm32JZm+Fi89&c9UR(4J39&8?C(ffX>>qf&L142G8dI;_liUfWN1W zh@CA8_AODM_PUeE!4qZNgLqN0R@RR!%qZvDQ{`aELORPESp;`_L{!fYzs54jpu$RP;TFfW_qSdid=sG zmfQYcCB3`gFE?sG?c|x5+J^UY2g7pPGWGrb03(Mf`%`6?SCBw`4fPXCKD6 z-yt$H#`3eT7vT8J{bWH(B>zr)B%e+$C%~0JiG! zF?f!vHdnyv!AlS|rxn8I9>Z&wHsXaJfJ7rp=<+gKC^a=fcqjb;>y6vQhSc@gVV8ZF zZ@V13EH>lp>jPN&{VBG&x1h1621p;&hcdZlI^t3V-n7ld+4b=lY|@S#zYpKy9jh-J z2f)Ow1x!AV#6N+5$qV-c@?G$cJYFV?h8{zZJ|ux*O{K6o={VGm35EQF#XL7<1HAfg zfIf0NiYbE=P|$uCBE!~W+n@_J_3=K;O~06H)eq*d_$xEJ*NtuMndoD)9gpWa;O7Mj zDDi0s>K~-Ti`c7Rw)+;W3Gl$bZYweKi9Whd9xH5A8!sF$sVOuJUjr%Ie$p8EOsb)t z#3t99u*B{!sHLUG_S%nQUAr~ex8fFTpVy4r4DO>UDzcKJHQ2i`8e#ig-b)@$IRhma zcqESTDk`|tt_x;ejlib@O$@%K1{VIlD9+Eh=k^kmF0aFLDQEGCWfb18Gsl~Yc+S3$ z8g~3^$4>sv>}6bwUn}3CX5&L_T3v#XyA;@VNmZ7%UxmHsU&E#xGG$8(e_}cBd9-ai z1!_E-tYv6Ej@q;mg(J7(T)SAjvur!AzHN(&sdnhRXf^s@3&81-*J#HoF|ZR4r#oVg zqsq)+Jk;sHzgrXNlc8qPq+vr;oEEY!@pa59bt?W8_>p14Z=Bet*Y#rX6G7*!ZZ zb>27eNyk3o65j^SI~Ib)X-%-d>BTNcdoeTKW3<#SlG^Car-q+`=&Dzn+4Yv4j80v} z5>?u9jY}_vH+;Yhy)V4mN)5jJY@ix>A&_FW6{1gWhZm9Qu#_Bx_^iDS}o+H7}(po^tzio7QAdi`? z$zkg@<+1d%SD^EH3CL!M!kgFGs6E{eqjL(uEQNnoyaP~L`3!U>L}60r8l2&8iCfOZ zqVj*qc;Q(p&S@OYe!h@p?dN{r?0#>ykvOpTnfY6}=mhileMNt++S6-C-oW{tym`O19)jaxHu7(tq z_n~iI7ar}rg_n*l!foppp>Sw6isVVM7T(#B5H7}^f6C#x^NOf;M+uL)cCqlvUM4d8 z7fb3q&8CH^98`X`^-P3 zD`v264ZEMAi?KX+u-JmGYvje{VPajXUPT<#?2^eu%1)BmU z;%>S9ShYJ8Pq~~x@x6^CV?_>;eWOM-48#$=-FYYO3ar#x!7hX>XF2<9Sm@ypFg@7{ zuWl&e<;CL6)=!kV9_qtqb<%9D;61*|YQ=2{1eycSLEj7|d^YbXo@(#K(|Lb!$=roF zWO5i&Cir3FznwHHMF&C~yXl@@F*dnNhV8K!#C|IX9I4!Ol0=X)p7%Kfu_LBpkJ$-^S0b2F%N|zPqmMF&WKlBp3kYNeq3gqUkQ*U~ zH$_+BCmAZ&K}*H%X%FrvFbK)Cas{tswYlhmWw2b zTjvwrr<0fa(@i{ID#XeZWaW-Qx5`DZ(CUOwc_z3rE)W$r z#p9wIxv<#y3^aDu!Rmk3Y{rm~B`PqMGGPuApRdo3cdD~Re9mKU#T__U(FMoTByiKk zc9?mx4&I49!07H&Ogj^aj>~_-qatOTesK|&&zKGyA1;O(=gY9Na}3J=h{rP-hV0lg zJ$&~1EQX1=v6^GQ$cnUcMDD;aIXN(nR{YH**=9SS_M8JbY_3V~f4d0wj@`5^W*?Z& z5JkDn2C{7E3FTP{*mud0KAFI?TYhN4^2%y*_*EM*vk{@=f>PnsfC z{88LPpNSd6N2$l;ycdNjqH_4!FAStV1!Bq!9neJaC~|TCouUb5Zvmp zuGG|MursI0FZo3T~5Zwct`P}%CD89G#*qBYqeapV|gwSaxzL81SyGWMQ zB_iDHK|y;o$Y*?n!?W_hZnY#>FFOY0RRt`H?SUk38~QFs7Q$VwfzuXUXdnLsmYmvx z7O%I^jr$F0Vw@EvMeQ`=?Kk?uT^T1OiQ+##R}dU3$5OY{Q~6>mSmn#-MAysHK|^(T z=u!YXPBd}tBaac6rxxI}<_P_D^d3oTctsCPk%c!Umtm%Z9+odG!9#!CSaQxg)*^C} zi?rFyT~1A=zhf7Waw0-c|MQ0`J1rR36#@3?d+~K+5J)`t23>zHAPJd%Kk<|oNlC|wexP`o{STt!ad6Su1KvT37bmk>F5-0i_>q?Iywb@FEFB!dRZj# zr5l-bY8&d3Gw6|d6YDCLlCXDU>CJ@*p}d|&?~VYTS@uBfEO%P|VFoqKiJ+hEX%YFX zY$|q81&>x)(%dc~sJ}Xfzq?M7u4*M>hsxy9k{Wuws*TUe_u&<5-t9NG99rsl*WRll zFc}re`|os_o4ygY@A`u-)y2%jVKdoB_zbj>2t1uL8e6h*vDeigQ8h>mr(S@*G*=9YAcV zld1B>B5<85Mz5Tz=bnC3rH>b1=JKQ~>6R~D-0u=U?nZ|sjBL(;veIO#o|?r;9bAMD zzpQ|nnfX+4BS*8D5lL8+4EnEbLP3xc9mnTlO%IKLOFsvRzWFYC)^-eS9K1qJ7fi(z ziB%1HQQA9@mgS6vbDsa<4PMhQfHByK)2n6to#hn>@!Eyn1(-wkSRY#i7v<>pT+n zKjyh~=}~0g6FwUAQH3S!UIEFU2EaEthN>^W#Z5BHx9-VG3eBcrc)y~GawM#*aU7}C!Ea0MMI#CTYgH_MZV5fWtWwi$6WG=O58H9dHL1O^%%p@~N<$lTHEaOnhZShXRS zyED27+@Fikt@C7wU;F|{nxO)}^iPBG6bt$<&W~isSklpP6wIU)=vvJaG?~9&Sy_nV z0}p<9zpo!WatKuB7}8wYh+9XSV#-NU^d(c0M=JA8VhF^kldoNY=YeyyeT;jx}s79@~MqPC6Eo{(8JW8ui}5on|N0!~dj&+}`WiL9*z><>9X#|Xn<$o7CB zQdmR`B=*z40rGI);W5`bW(r;9B@VLvJYQagcOLuC#5@B5ZofPkjpyfri+&h&o)pSm zKK7P;7j6{XTXKeU+o}aTiLtFhU(RH&S+ka&n|+_k$TO@1-c?-kocpjYi8$T(KnqhpaQW?NM8aNCXd@_Nt|cQ#XInj8Q(vSNkq4dHa6gWm`U!@VLn-H__5Fh z!;NCmcF$J$w915L#eRp~lSZ=I^3UjdGYQS-II-{h^cbjZ!nfHmc+~9z{@N(R)YviM z%xWWHPJKT=TVBZmjP|pOGv%0r|3f&o=o=hv&qd!)AJ`B1d^Q=RS@G2fp|E9>FmTsm z;k))qT4ona<(I66^U^c%h1fYL_Fa$H#u!bRYWPJc_*yEwyeV4vJ9(opNc$G+amc`# z+r3bHmptCm$;Gn+lTqCy4hLVh3B5fcgq@nULX*#5n5lWgsr`9ljht7pL(VyquPetb5fPe0NezxOl}TmbloF ziOkhxgGpW>J!=D;Sso1!A0{y;ld@T^K_k|__>QN# z$FWbQZ7AiFi^D-bAnw_7G~CeyQvI3i^U05RlN`Z$`yDZV(M+M;=+Q!rB~9#Ei3?sV z8Gsuhr=a9%HxrTPXPEBx;m-Yz&viuN7g-14uyC~SLUtod z-K8OX-`B+A=YL_f_w`xFA!!!i_yX^aZefcwbJ*!AvsjtbE$06985@|Nz$OZ9VBPZ9 z5MSpA+r^B9nHEaIk5M6PF*k$Fd!)hI=FMiiM4z$I^RpS|EMvd+T!g=?&Vt{Qr}#Zx ziA@|jgn3W?U}n}8;hc!0LXt6C_`RqNAN&Zxf`&j8UptS_UurYeoGGlc_AwS_``|kL z6x{6YAPjtAB>a8Wn=Ki2N@)2uR9MILu}5j?)X~?8ZVvPUla5Vjzs&&8A3TBIvrF;4 zf+MEc9Kww4lDKY0FFXz2f!Arc&?97yFhf^LSYMFG(g-r=DRY=zUKd?BK8ntG?F?^w z8GC;s8^wAuuxY|lCiR4`o2@*D^F)^mwRCNSykfNpJp(dbHA7c%HP9YZ5cebuZDf!GnJLT9nGZf_TY~m zWOHM5*ip??Hs=^+WXuRQtR9cvQJ?X4UkHxZI)pz&=fSNrBk0GRN90ZS2-NK>p<+wr zsLECYQt431&9TlCxXPpuXRCTn>QfYV$@e7Om#Ly9@(%^e)~%;nHhx@#?J9xD-9~Ov zU8-PT@l~rks{H)dnnW6HsV~qOJpy-4w1HRrzOtKlP+q&yAjpsR5O_u(rLME|h}7K_ zYfjalKBzM#OM1fuXS7dqv)uf+ND8hcMw{ObQ@2#w&_H0HI@vvGiC~9h7=Bp&4Ypj~M%eRvRNLo1*`Tc@ zSb8{vJX|cv=^6<*|Dn(D{_9P;aLY5!f2lgp>%KzPZgwHz`IBh=pKNZZvYqP@*$&Iv+@%)A7EL0)%@(Nj#{rVhij(V_VyM-8j=pGiC8@(z5Yusle!mD@Py0=Q z>s2|yc(o!6naX1%>Ek`?X#TYRRw|jU4CxUGahoa z<>`#}U&OuUses1@TV|bnO+GFVrSXN;ma3y2@Up=l>hCv=eEG45>`fMuGdx4GPihNr z*B$6>*F^%!)z?U|$t1ysNn^-~Yvy$DY`oyz^v42bd7cF2G;)4rz1*9;9%A~{h(v{- zv>Lj*630GDh3En^vi|!^8ocus(f@T{aK=&$J3SVHqgj=})9w)ny_4x3ZwDg$T2J+T z=TixDQJN`ILMEPCMh>b(b0Y-|0+#(EjlIc&4G&7_o6ku!E=-5_Qgy?tqXHPX=qU(% zZi=&O!+35~F-cw!z&-mH!5y5_BM=;UPo>71(9C6}+y~Bx{+=w&y*Sj!$!QtV4&{>q z-y8D;r%#CCgJEsBJ@-6G4NM}JyA=ejwH;Q97Y=X*iuKk}?sGWFUTvyWlO&ju9WL1G z?uRQ6^?<~Jcifhy^ONF9XT&jqD{-Ma?v0@Xe`N*zYdT3b2%oMDCcN=x)xZ=6a4BW3# z!(@z(FpvArOv*oq+KmmN;a*F~b)hFcz3?6U_I!gC|7M}%FFv;ta}iJEUxJapGwDum zbx3xRp{=G`@X~ZPZWy-^K24nfU(V|h!;cFw=;j1GanTY!z8FUrUY$u~?Gs^bmOr}P zJPkKTcF+e zU>8k@R;N4z0Ti14Bb5A^PE`{5NW%019 z6rJ@alB8aYq+ZWMu&;r`u&Ov{oq7bP+G?Wr^7F9!Oex6Dwt;aeKJZ8P4Rqd_0fCX< zsruMdOmaMl`BzVZ^;>6_<3E{AZe79bgT+~O`y`xeuLLR?$1!E>eK^Qx)~5_A(}OIK zDA!M>ZSJL1Puv_Trt`Jo*9PcW(t-L(!OYV*7!<^VssFpp@ML^Db3fq5e!skg)n-CC zKH>o!zh#GQ!|iZ~_cY7SAB_Wl=CU3>BWNi84(AMBg;Min5GXGNA(zX*Hk<-~Y@#ok zJYo9&Vq7QrjBP#O3&vAx;ZP>xt`*zCIrAs=Osl7@15c?9f4?j`f18e+bsOgdHDOY8 zABZRYMVaTpIJ&+aLK_s({n#uN{(S}04Zfl;&-Xj8c@$<$@JG8lSK!K$sq|x{kRI8! z09MKy<6xU7nk`VpMQuY+*JgpG&qkyBQx&WZ41v8?_Hby96gA*I&iJ6EmM+m8hhwhIV?U?5vTmnFd|~kc zT;kq>`Aj?B@vsEOX;;HXtBtUq_W_^K{{#i5-XJ+s0nf?Y#tVX86r27UjOIwQjYVQ? z%{66|*wancEeRsqI~G$9uR@Y|X&y}L&&AFIHgNcC4_$Rl0Z*KaAmfxSQse%S=oozj zUH$rC;qMqY9X*=wO`Jt^+{8M5gmIT%gi#5F066|N5H9^b3 z`XL^l&ES~Y9xS`HgDHG-#M+ct?2~sw)tV*>`{PmR>LGUCAP&*&X&EY4d177fC$votW^XR-z&A}#ah=IIb|+^Yy~qyJrqM=3_Q42fo~w&-HZSnD z=tU}ZJRFu^zK8AK4e2mh3Sai=VP~EknbErpyq1r^-C>3FM{@;VufE9VrBdMU6#*#M zD}!^+RA36mIK80`|F&+SZ&8|OP`cq8^KXz+eg=N7v&NgJw!;se4dLxuiQ_CqAVgys zUbl-aDuiB|_iJ{-8a|4lw4HEY+J`1ZxZ3$ZXWUj9Gzms4>jNX0NrlIR85Q9diU@i}-ww?tQSA=^<_g zVj#V}AJ!|Mp`q?iY43MOToG^=;vLoC(vMu)HYxnO}ry#japo}-rc zb|_w73orL4;O*ak$tsBmlI1oHZo8Y)d5$|_@{4Kw+=DZ|+QGY(cTQ!>wVHG=k)LZ? zI13H_7-9PEJGi>(2HP_%53_&E!`g9UaFKQc)$;j3n}_a#ndxUx3|NkH3yheY;zsld z8HZbzi10PJKx*C^OKZy%z~RVz@b{R?3{L!F1wzX+e=EtCxU#MC{XS7;QO@@ zM|v~ddch1^M@8ei-FM)_<_5^Iu}1N}0%(Yz1=^O=a9Hm$2Ctuo`XgS$)9|-coRgv? z@GrOZdjftLVSvw%-3B_VjFxTGh5XeQ!FZDfuCmj_8ge+BeQ z$960}Fqx@_&ciPSMp)!=0ekZQl5F1Vv-aX!2;csYI_X8g6xRqmKH?1B&{P4wG#}fP ze!%J#cW8ynFwZy%L8o3(G#upsEym*1ukZ{FHnIZ012aLRZ90ydRDltJ&Qzly1d`RG z(REW0>}?r|TW%X-)}{;8b=_O)&@dK{`Z_?SPZ(NxPh#e>_h?3fEDX0zgR`ZZ>9{$G z(6stGYNYK1=@<4;pvceRk9mq4yDuP{d;%Z)T%#FwQk30U1k);{3Dwy}cg>84JuT;8 z%o{8Cax5K)U^d;jWCRLgHeuQ;4u0^NnAS%QEZ^e-teh?ZAq!4J>(zQH^Q{3N87ybi z;3X!nPD7)j$7pAF3LX@@;cb!0%)9b7IJa(OavJ)=!`VRT%^<4V6${U=`+=(8S-c?; z#Qb7yVUY;`o*UnWeJ9nR?PWYHt&YHDwbGD$Nfd{~bg=BcHvIZH9+zj|$M(Dp@9ujxcx0L&A(1(SvwNdP!oFlgA zJ5srFs2z!sJ}s!?z+9i#iBbvM0Ai`2^^yRadRPYNgjQn^d9Qsze4&bb1rk&R|0#- zfKI|I?y<#ws$n`AlqZDJ@{BZGxTXPXKkTImyF%!so%~w(MvuE75=Z+B2I;cd9Wc#V z37y}?;OCDHF#dTyOo}Xmh?Z~MaDEOE3w}$J2jt0B#edvRHEEiil}fKl#BN@vw{fn914`jp7S$BuHiGD{LYJIsmAkr~8BA%bj8>*rMYGu$$33*4lp zNPqfU5HG(>+8(Dx&%HcAhmMxO_dy%-&SC*+cqc~G#%Iv-q~D}!VLS}q6Qj>IZKUV< z`Qj~Jr|GhPJBVQ49U7u4#e2X;(0YktTDCXsRl9zeNWX5QCl49J z<`)ArN_-pDpyNU1kSxzne*kZ}L5SUHOy)n1rMglh;M&c7P%l;jcDI+{?7Ck9xA-=` z-hL6>F1#c6mHwk!wrt})tDf9Fey!9VD5W*Vv2@&;o8+oT5KKD#6jDC$b?)Y$FnZZq zu6&y;G{hZ-<`0f^`I_+{b)}n5-noxtt(ix^Nt;2BQY29q9SNoDHDHf$F3G!-N#CW| zLsDcaHI4S9A$%?U&d7s;SNhT1e`VTa$ED|_q{oUXs&~;Zw_g+4xCnZ+C>8Wvb0Jtk z5tSl8LC3e7wsf{ldxcY6$$oAg4Gj0 z!wts{GBnl&S5H2R?$5@Oi(!`t(=da8%6ar!dOO$*ti+yGk;J6%756!P40RbBy=(L${(m@54TlEwJU%JF`%ez+1tg-h7qOx~@wEQ5Hd=#sy;`XF6&9e3iAu(-Z34c>B0-C-&7i|r9Lx6_q0ij==ouq|TR+I*WBm)LkQ4!0 z*H6N$cXc4Lsv0E2GC^<71vv3Q4t|QOL%e(dny-py*XYT`f7UvWGSqa&BI%c4Z zZmfL`=el?v#TH}mObGp-qw|dC^8fz0jEw9=D4WPie6II7=Tk;$2q6`vz70*4(GU$G zTT&=Rh!V+)B&7(Uk_Jr;m4-GIP5rOm|5o?h=)rZJ>zvo?`GmqNa^QW%2qL3ZU@&GG zEfou=QSzr~i>M7W^xA=Mh84OVn+o=x_V6GFQ5ccKG}PjmWAp^rom|hbixJh1;HD>|F=gu*T)1C= z{q8Q{b|w>EjA+H=xtjQ8%}7+*D1zPw?o6*jhi$#m!K|_b=hsnH!p zg#mcG;}b+|egjQYRq>--3Af3=gG+CE!%jYJfC&q(!_s5N;iSJbNcs9fVURVNjD8B; z@B>O3V7Hi{~ZSZH}!20K`-1(v_-(M%*5byu9nO-ptYNz3yu@)&lsbUdtc?&i)$Sh28(JecI&%YL_o zz@)tjcz=};T$-;y$B2#iplmp{$c;rrMcmUSi$1sH`1$B7e_Nc!4pfw~fo5-b!(X%) z%}T(?{?X*9;>j9+TxN5AMdR2}Z7>wu1;;fkG5>%h#9gt1xJidVt67|kRV|3^I!py= zqv5!UHO$QT0?d6s`{mNY-nO4Y^)br0bEhu8Qrdu9M<`*iQZ3Y%je@a$BY;jVg~^)E z{B80I_n7yWl|*b|!%Gf8+suP7=y?OGFB(CE|4-1~--7Oaa+vk?2sS7N&?B8BEOX_V z8Z)y{C*A=6&Z|YvXd^XDHlTLN2^49CzmNmqm4OZ9Pt6aE0L<7MjLjT zQ6A?&xlzAS#Jn3-!d~H;vCm*(yCS}>FYBN)m*tD&mF8M?;1CI@KYmY7#7pxdVEtn^3VQ5M6cuL(#9M^ktqk zS&BPQ^)zc(Q9i_yj2A+GS}pno6`-EfakT0VrB_dO(vMq*>0zWPj@QvgpE+vS@oE(_ z1OE2>Igc&ib0UM8{g5_U8pkfvreYIydNx^~#7oY@@tud@tHW2&zscX0)^}pV#!I+- zVJZ&yjpZHTp}3`fvQTaPBw_kf5V9gOiu$5JM$^X9`h9g+?|&WBa_aHHdY+xAEe|oR z9PE1i9Nro}1dYozFr`&WxU5G^INtUHnOr*$l9yh?!s7}kTpJ<0>M~FGZOdF?kYYEA z+TOt2`E^+I+nMJ%kD>oU#i)K?IHlQarz|HB|n;Y~oAyNkhjaI|wz%gL4 zgU^ed&%qT|_4vp<1xGCkN4dXS@%*pPsHsQ60|Yb_a?lvoQ)BBRd_e-3WkL}!_7mx(IIXN zrWVb{#LBOj_`DadFZ_!dco96uMQEYMk3}hiVOpt~yHdMe=Cm z!%z&m6pej6_qrmoh>xdaVeZ^Q43Xh5?2SIQ?6X2S>nSAPK9bJJij(>?CwjBNfcmTd zAg{4x0WEx1#H))%C?7=6tNXCD!4Iof%A<|eNbEM$!Szi?z+bTt_I^GG9nP!h>18W2 zyFZC8f7wGTc;AHh-EG9J8VAY7O0ab1Ft{Xl5VbAJ@Q+z3E{RFN6LaHn(}I=w`iT;q zlH~bSyG`l$&`I1sDhelyCt_lMEUa+l!2Wdzx7QXysTI7N0RBxnuqZJ54K zoy0y5A+{;d=OV@;1M;Rk4@mO*h%C0(wy;oOZuNADI(aB|W{g509@+#B)LU~=Rn4ErdCkz*@y zeeD}`Zqp{8^(C;{{xu{jI%3h>Q!Lc)Dm&C*1uKK5vZ>Z5*g%3jNab$=3*99!{_6qg zja&pinniG1W&wU&poBx;G*NVu4}Olz!K%JOoS0RJ(qZ9Dy?;1NNRz?^viIPxS2w6G zkVXy3Parv~6&iBt!SZLbAjU44i}u*d`zqJqJ@q*J9aw@jWy{FpdjjQsDJRFt*I@j( ztKhgqAJ26$-dnhZ`)FOsS*uH^r_3$ntmOO!R3mF@!pPiSUCS8 z+hTZQ{xE@+PhK}6lW$t_;+!B z6tbbU-M+Lm`6PuUh~l`O(RfF34i@oql^QL@Z#DZcc+*vGU+NFeIX{zUsAaNo?)y2Z z-hAe=OciFGwFbMKD5&@MLg~@XsMs$-VauoR{c;aj(CZEVWY6+D!Ke5*MV_YT&u8s6 z2iT`?OTaJ%V1>O4OsPHuT!$U3*c}T8+P9&scrMgRmB7E{{a_qe54V1q;ES7?*dH7L zFsT%JR_fyQb;!0)^JX%$Rty=?kgU$F3^(tJcmw*Z-Q6!j-#Ar=^G`M6tT$md+eWyBy7C54DY!{;a~4V zSQoyI65@hr>9fNmF=-D*>n7sTkEJ+!%O{W>_72=s2cSc_628vq0=u>9crIK`IC83q zu)|s)%zZf(AIZhzSI5ifVvvn1fAh0Vr8W(aKlZ*jRTA!!aF;QqQp8W#`$S!GG*j<5r4p+ex!l@22NP z9;Bb(MeDx1)4~{ks6P_{FNJ4d)~|kOUoVQMTP<;uUo6fE=6QOhdr`7lM7VS1AFBQQ zgYve{#+OkJ=svI?pUD^Fwb%-ry0s3k8Vf+>a|px_7DM^SaTsGc6HS z-AcboCA<$eWXeY9ypRpcU**ueBRN!M&_qw}@4=16$1rIfKYvJ+qvnDlTr=zpPHo&s zZ?asd0NhB{yOO_$UB^{&ydNxOG2ETJ0PG}A!-V2}IO||JUcD|td5%-C@-ok`x-|iZ z>{9XN(xX^qUW1kVH{*G`a0sjkgC|q2W4v(%--&;SL%%99QZFAvztmtsLo_aWZBWU45E!&y{Kn8pFQum2$iFH;JmRX=~v7k`$bNqU7?Q~ zzS!ZcE1`JP;~QrE9!6v8#mLDqjieXtqJ0Yz>0hKO9ufS3XD0XH=^Zy5ZR3mM=cJ(2 zOG%!OppFIOb+LCvE{JYh#i_Jm}BWT zA?AG;}F9?KQ-O1>$(R>I3ZbeGSvBW>T4!EA@6e(@{GOfSZ-f zYE(Nj`*DTewRB{NQ3sJl34E~BbME)a=ku=Xsi3>AF|Ak^0VJn95?b;C0 z6#z3*E1=9|G|#xvK?D9)IJqSOwur0;6`5VIRkZ-6hiBtdVIi(r*MwImR^v^hHdN`d zLPZ%9)Oy78bsgGhc}xfGP<})vOX6|5Rw75p-&XGg;ee^WUJ~;f2W{ zob)@169X^c`=ZBaa(+8TYHh(?$78Yf-CR6BxEq@X()ixc7wntL(cu#T)OsnE4r2xR z$o`?h@#b8lg$Xw?=BJ=rKLs5QL?Okj#)q1TbhNjed|I#4fy(W$t=twjh9%$xmi`_DsUpBc~K;`3;$_M&`p0s3kV;>ls3 z(DS1#-LjLV`?AF_W%VV<50b+wkKdr8?1X*!hfuFmpFXz_qrC6+Xn8@IHTZ63hxN`e zRjI!yGhCcL%SzH1A7_eD+d-cOj?&LGF$~nO#es^4n2_-n%v!6V?kVp!oXFwja|iHk zTO}SZbraS%t{3tSU*XTrVWfI*4L#j=g38p7@l4Tj{@v7ywb?3kpvi{(9X%+o@FAO& z(7}3s*ny*gC^>)bLzk*Sl;0Ubn)@=zWJfD0em?`L#S@`iR)!Pq3Smtz6xr?6Xx4OS z0K$4s!FFeWoDf5_+)FHj6m*;cB$abopN^Vy)%~GJB|`rEWLq+yu*#w=CDIYsPmo0? zZVG0o=i!U@wzwpEH=p}SMek3RymPW0+EUtKPx~-zn2`?Fo2>C;&vf={o({`2y~3V5 zH=(Fc4(i)Q;mgaeAisCfBq7*I z0k1`&_WUV!__j4$abYPf-kd-IZP~P-_y}71WMlW_RJ5)2K+_{GD5mIu-y&N^TKy{4<|r%)QFo?Ym&#*cz~SRsnxTJD`-xL{tv2LgxeHaa6u5?!3V>a8;h+ zu-E^vc$7BXiqW9kFXc$tP=fX?Nn>^I)nJ;|W43nSIEG0VVMh|5FA2K?%Lea4TMfgN z$s4$AuO|YtR#PtSXDm)AT8k&IyQ1}@+mxbsnM4n^&{=mO?b>Qc-WdXZZn1zv5}%ns zng4b&%;%XhnfPLUJa!lc3T-^yg~Kh)g=5d|C(EmG^n7bDJ=AqX`K{(S_rGD7eW@8+ zG+VJU^C&)%{>mh5dYP+r6Ki}If^Kh@qTfL+JRoX8CuD8tjHE6p-Z6#KIb$JsoB*Z@ z4hY^0Bn4dyiv?19XTi)XX>gPq?`;QOF^NFKfzc$_C`}pV zijnSIR*O)jCB=GwSJkaU32xun+5ACg3n@cWnDO z3rAm?jSC#dV(HKu_`IzNR{y(5J;I}W7iAya+9ipC3DWo-jWM9kT&Ow=g*%te74BUB z68pk0p^9A+X6=23PogR@foDm%-`Rr3zmxDL>_C5|c+POhnN8H`U35D-E=pZjRNTL{OQ(CGER6j_zw}(M4M=3cN8JN3~2rdpmO!<{U-SvIzVh zv;*%&K0_CWYyA7C3tt$$$JgI4?+e^1@(Pci7{J9! zu`JkiKO5$i%jR~};Qf?T)bR+$J6ALyE-(PrxqfE56dz;SuQGgjU>EB0?>&EhihWj9 zh#KQF*u@btpiv_OiISUfvea@kHJF3_H2Dy?z&Nde|}2{CR(H;C*%2_i)yFmyte4O?}E zyYQQ*3A^dy!u)1%x-tdoxjZb~%0F*vH=})v5=a;wVW0QqGF!d#)SXjKp;_g0_}6Gb=`e%|u@j8Dfv-1TlxW_YC9TVA`^Wm)j%{BI1&K^s&9r#>O z3dRNTeplT!R3X?!ac5@XZBHwd;s2dAFB-=B1cM>p>+~w>fh+DsFhREn#0>P%JYoc% zxNnceW?GoHOCNQ!m*eE4Cj6{3mOiwa(5#MB?3A&?$ce@{{Y4YZd;AVwCF_>IFgp@l|X=*3*4;Ngs#mD<8Q3Ry2u0A?7mi5y>OB6x#fDH;bWkzDt8K> zu#iHwu0YG*Yw^WjA52z8Iz9$yRWL{L$9}Wqp|Q|sZ2*BbYtbt-8H;m1;ESzmY3sZk zO6l;hc!?d|Zf#+S93fgTq<_MlcoAG7 zs*5!eVZwQjR|=PYTOmwf<*2m(2(ElzfG3alqIAef^me;|?=DA^{OY|Fkda8~EvC3@ z!hHU&=Y@#|ui@&8yKv&vU0Ck<5@(DWMad)9QP%@6cj9)SiL}cJ2g-)zG&v$P0>#=+%-ziZwrE&E@ z!!xv~BG-b%6->x}N;EEC?~k%8mZEp-S@gYLjt}e0ahdcm`d0WC@%7khB~2(!s-Nbgruu)Z_Us*!B|;+4~>9w)aLEetxr_I3KTBMq=#5%{bC> z4AsB6jVXg&80f1;Pt$(lt|{syF?SA|H^PzG=-pw^pV_CvQFqm0 zyok~8#OWyP?t26KZAMWBze_3dmL|Kc(`i?l2k)K^qg{sS?2B48bJ}1Hn`KmJ^Zij2 z;$%RI>W{Hs`YJwn{0iUtx)V%ZP5zp`6wUWLPA5d+qNpew-6f3^M_hzqf9Aj@*DD|w zvfq6}(=aU~59BXb;2F_Ul%Mz; z=kQ$1pZlconEeMhb+!smeLRm>3##yV%n7808_=_74+edXrBy}IbmxgL&&>7a~l;~c8w z#=@0u74W?84`UAo;?hDVoKk4cKg(Ni?zijcS6+pkbxmCLd?nuF`XAHTpNLP>7Gi|r zRID>_rx7=7NqwdXjp14QlN=jxf@~2Qv^@fYgxrt=zeAHo2UozQd=^U7|0U#4u7(3I4x_(i1b+U!2-k0Ngw=^JSnIMaY~zVA zoRMUM2Rl_z&p?ixrQ7k7Y#aXPB8G+4XJJiM6!;w#WA}z{;|g|WbJgD)pdt1cxE;&` znYVc`eo`e^g%!Zn)_8zaAiFbcAovMX1MNMGCGZ3cPsbSTecCdc*3bcN8(^l_h z62Env3KZku&%Na!d}_`+&&wz@>LMwmT_FE&S-8kQl6OV;q3^m(EX<0>)q3ktlCp7d zSqu)gcwo?AER)rZU_W}*VDXkHh~t?LIRo2aM*z=PD6qqn93e(GYNMj=S13}w38{@E zcqTv%+;eY$bEB)-sEvnNh<_y8^zIsFIagtH_ipT}NWlr4_G3LiXMBEi6|Y;?px00m zzK!_|uLmE&v+B_pVOh+!el2I0_OAwo$Gm^w+IalP->i18G{nuRIvApl$KAGm?qe4g9QE&=B&Ct&gS^>}cu9UQZ>0`b}xY|F6geOC<;{ zru_mt{+<|n;}ncvB}oQ*KHzr#cX)MJGPdwc_>X(}?Ell%oL1R2!3XmV>}P~GzBd_* zg$W|KJ)#DTd$++&-%vRG+5oDZ=do(P66R@L!`L86T#^y_-iuYJ~}V5>$`9rM}R z5Ay6fXfbcQc03qy5oht-+<{rn=zP))--&;NuYC8pxwM59+D?PmvNN#VBoz!MXM)(# zMz%SA2D>vZo>g0?L1I@9XejN4`xEDaS&}o8S-gjJm_%XwWet2%B91F#Z=pHQeR3-2 zZ<|*&P-JHuta!Hvdb(n%w0aA@v+*bGq1Q$q z#47q&6}JE{UsJF7XN-ng*2vzjl4TjO-?_7^7IKMOMY#Dl z*Kr|dNbuf$sbEIhZvA(D4(CZ{Nnd6AjqJ zanHC^(^T$na4YBac9!6K-*@X6e^VSK&9h<~tDv!bEbiQ;iAN_ahxg}xvs1O_nY&*C z<5Yw!`NB_5ZABt?kKe1wTsaOsA&1!K@D{c-nz8nilUUEN&+xK43qnupfugAws-GW@ zVyhp6#`g}^mw14!5kJa}z1qemUUp}ftR#gj=aJIs~Dj^xfzL1JgQmWps36UgxGI&sXi-huB7Rq<$ZJ8XXYh?TD68En5q z*`UlDw(n9F*Llc;{d<(e1bkO(Q2Pm|nWTBrtmlRJC)DbDfKK?`*$9ux4NP9pc+~Z=WonBW3?$In`PbJhLczBz&bAj z9Id+>%u83oqb0`BlRX9>|N9NUgozMU=K?0SIm~aQ0TVHP#+sxivV@DrxT!o(Zsq(O zyzAhH!;(kC*3Va1=f?szZfq3XZZ?IOr{CBD+x5(|?izQ{+LxQNNeSPme}j}0I*=k$ zi5^AA@b-HrG<|rU3+Z_;&}?)PBqoeux3rA8SxUB?YO4rjD`v8GdpjmCtqO9tJ=hFa zeYT@-3!JFahHC9(RtoxHFB8wgdX}@PZVNcYklli;PZwdg0!zgQ@y+_nIN?(&#*r;FJB6i&P+<#!Mgo|l(*VKqF2F@qS#W4_VXli^IeaI_(px^`-|8Yf6UOg` zI@6)~S|o_#bjp=5L!?|-NQpdke+)y5isjK<@UeOu&?`Akw zsQ1H}HK}k+(FaS6Mv>B9do1efhS_6}V$=ILPPW>T?TV7(Jl*A~X#PZ+Y8#HbpGlL} z&e?S8dLapt^x((jH!x9Gjk;>H;Xq9!n^1g;d*}8KMoSN%hVFITbv96Fb$^X8-ds<( zT;vnpH!r7k*h#J6YowQzLzN6Jf5#lG1CtW?FDE9Udjb$=pRUQId|ems^{~b?~-XCcw=>yOD9!6fu3n`#j0mcM<1}#3PbIaTl*IzBez+NF5n}4BQ6ONMH zlW}C`UIAw>&ceNP`~za7K#mm~1q z32z!GzeK$Cf?KPQ!s>k{Gx646tmbhAaLry=dQ_4OTQX=t>`int(#LD!S0SaK9)f0l z2PLUU^t8H)LOvtj_DGAe$KoNmnGkRn zVeay3>%2Feg38VjuyWII$hzbVa=(vZsgEQVbhDN#;<*{KMdr}^@29Y>BOanHBl&)v z8hzQ;jJno-q{VINy)1$`N$_0f8KXh=$Z|fby$mNC$aB}uZDnIhMVR>$MUokALJI>= z@E)AkD19-Ini{RhWc(@ipui5M{Ng*h5mT61;{(`PP=M!v#cT6>on~2z68U1w!quka8NaphMVxiwJi zJ{j3hTimXq3~zYnVPu^hrCE>TYaP0vw)`i0o%W-C;~;7sR1to%bS2{rHjr1zfk8+% ztUIScT_3vX`kod##D8;@#%x9p#m#(YWB^V56sTz0OML#K3g3&C3GyH~ZY{fme_DpmgDS9R5z9eKY#Z)i$a#KTjjl%Nj$?i)G0(q!ZLEu0Z?M zRDKq?3mYwpAXI%N+&Mpzrl~8@^(F&a*tea{9=V<=U?($|LiSz3k6oRY&Mf8bvi)bK zu+~9CmghVVD*s7A4bug&W)2>S)UaaLjcit^CoEDmhpaR?$X%TQ)f)W}{I?HW6_25l zK{4)lkc%$Q{NZhH929+(4d^?Ht*<-s~t zN@4HWJ1}_fJ(Qkz$ADZ8=kcsx4Tu)3nwQCj9^0^ge}k~zCmgq#@cvx`o|k`QGCaIA z4R(xeh1fs$z-mJ$EIBF)-xS)|t2@KssZ%BB%|8cn;rF2Rx+y(9F`8NgYV`4+2@d^d zh{id_sJ7DxPv!SQ#_U=!IGoO2oJ(S1uN~Nl?V0d%)E^K(#(N`p&#v8$_Bi+eFkN3A zU-CTVcSrN#+VBh*AMGtr_v*EA-*Hzk-q{w@N;s_lWsk-R(Xd38?{Zl-!2S?lY*`V6 zQXwm`iIc;b4!mn!_a98!Qo-GQH-oF)blzsVSGYi}vx1Ax3gs*fMVV5|E3SXR2=?~K zb7)=K1P71Y0nLNn{GOr^P6r%-jmkDGEv}vG8K2BWm{&l8PA06_o&d%b-#}~I8PIGz z3GX`#;J*X%5a?0Cw)45Z8*bmB=IV1;eAWevR3iM`Z~hn&yR zFxh?uv?b)z-$|Af^iYcS7G(>4y_O42W*7^lzOH6R{(3R<;9&NNzujwP>EefW6TDZ| zPwVF0Cz+mn>XbW)-?X&RC`=5Muls>(!4P{tUjPsI{&Q0G8oa;w9&A1}OUlhFc<$(Yif$ai0$&Ysd4a9muY-9Q9~zDB(cx%x!czF9c!cozocomjv4z6C zduXFwGX0&l4=qpR;{=W4s1{U(0o-cz*gOSyy^Rre7e)zR{@f-kQ_ka^P>X5aM@8B` zrWgB$H(>w%PMoTAgmlt((=4e-$}{U@s);5b%iU*=^|`nu&KoC>Ou~k(+O&ZZpkU2! zK1Y%ZPG1zzSjHZ|-c#girtRe(MES7F&_wW?^c9pVrlUc(C=FZk7=sMIw~8UMjz~h5I`D z5oZ*V3u05&V6&P(%HJ?X-E}Q^(=if6H<)12Di@p}6@phsyJ33CLaGVld#h81@KEIx zOfz)H)yeu;aBVaF=f9s$9q^?(FbY}I?P*njfNp_>t z-rnT8X%L@^l;VxB<5>K)2~9mgww)9{D(9iFq~k4wYNaCq<@ z`jTHoZzXopq^~E*X=fE}>0Lxo4kL}F8@x``95%}+~Dg|GBi%*)% zP_5}FXRp)7Ee#w6g1i_?+Bc7WJR40rJ~dD$Kg%zQJWkqji`j((Hf)>~@4#$ZMy-kV zwC%JW4eCrJ(K7}VG((McdfJk8(n#7;sKn2W>a-}Wm+uT`;OT>_@sCgp9|qllD((6B zP{|*+CB&hg^kPu{F~FjO@36aRH}LzrINbC!6`Nh0>1J>!)f`wzW5)dgmkH94a5n-r zuX;)M4PTRL&t-aGH-ln*6VbaQ4%^!E@UNse^jw&Wjl-8<%?HHSxg&9(oi;6da1LKz zX~*VKM)bzYh((!$YBz683?&z<=&mC#hVHrca4MS<6##stVYM~i&cW8{>2&RGh ztYb_Kr*K{db{=2JJh}|vMwJ}xuhpUzOLa(tPYca@8jbn}tI??SFszK@_g*heLACNe z7`&E1DE|UlA|0?+poUvEj>c>9?$|Ku1DMKOhSrU>vp>tFY6Wp;=Vam33q=^m z=h$}3MGD7z2MhJrE*4&PQY3Aw$#h6|A`R!iacVo%*tcQQEJ|qwR(bj3nBi_17(QG$ zAT?50c~VbUcFmnCB4<+U_NjDVtBzf8JH#rF0ob17VEy76wnjl4W~^*yw<63zTWT1b zdC`gL6HeiXgfg^nZ6ZUj4AN2wq1cXGh`AXBySh$;LsBxE`|l}dyyUAOW9~)fQE-n* z{3v3f7XQ%2M}oE(i%>!T84NV5!sp33_;^GWeVlcU@>B9DncW}0ZdhkQ~F;%2+`IMg`>r`MDVH>s2ge-&m6H3Bza{P9hAn&16~2oBH}_f!&b zPNGR(NgzuPU`X@{bg7uarTh8p%GAr8PtO3{QuqT~8~Lo2CwAQtl9!xqyS4T?0iCNlUIDQ<21zMv-L^965whrws`qSu!MDogurk*w* z3^H3(f&ekY=TT<=ykKqCXB_ z8?ME>o5Ij#U>kPd5r?)1(@pd1_v0=%*7`xLHWa17!1pK^KZ;s9Wk}D?fxd??q07S;Q(@m4=6C3tV70xcph_?e zzs?iGX|EJsh;mE#to#0FxNpfV z+?7{|ma2(ZrjUnwLXY9n<9$q1e=MBZIt&Jyb-+)ujqQ4p4POr$7ei7M-(mcknZ~2A+3Ao$mV7xz3IuIFHJlfM)n-uEjx;B z;y(C-&*c2%GnUi3AL05{M=HOTg_pJPQ!E%_U3+nD%>&f1dV=27Nk z6L)X!UvP_i4>GojILo4s6zyM7#njiNX3Xy|8t-9?>nDsVse#>|$Kko`9?(Cu2ltro zK#9#kIH>cE9ZRidi!;x#n*tHED3d{%OkGUaB}qM=BPnx`pI5f=Yz9qz5SeTaGXj)Y ze8YEc*Wp_3P(vV}5txXP#_Bj#_8%^t-OOp#Q+x+W}VDM1psVnb537NjSwpO6aL-g`aJ8 zvF(TvPI$8jm)B3jpEI0MZN?jvF1e30+iqjpx2Mp1xCU^TsSWq20sL0FmLBH z6#2dzpXeUO&C(-r-91T6Z|{fF9lT3wML61WQ8;ne52pF7h5d7U%QB+UvB6^{j!^=9 zpwfT=eZ@G`e*u@WKotH!yxm}mt3N--|4OUzSHn#{4}1r=|EtDZn;I}NM_v;Fg16o*xbpcJ5g6dL71V@O{d`ZKO8Oo4oqmDCm(S zX0MdQA5LGO{;&(as&>c4lf!W>cMiU-D1p?8_d({wdmMKw8<)Qh!8v_mq^Wx!SBaKk zq>3G@cu~nIdx>&;^y^{S;zywP-zi}0Mx&XZA*TEDjE{YKuszcWJf@F_``2IK&mR>y zjGaVX#Vh!!zXro=8!=}cvD4Oz*s48hZ0o~Fd}kDfl__!fJYNG-oqogAbDyEIM4av( z9m4wdFSxe;5FYrl5>xoU^Rc!&!Dg%*tQ=YdvBFw*sPi%N^KWKZlcnIXra4%BQ-!-P z%n@Tkm=#NmM_5&84aHI)oLPFO`H04GdZg~)o!SPur z_Am)&*VW+WAGdLNt}#y3(d7NPhB&=876T=-`5a3LHpcOcs;nLmTi*x2qaE=tpXre< zv_;d&c4&RW8fP5S#EbGpaQ(bArWfyn%W=hASn?0<_Bm;mS;fCc`F^f_-wN1WBnH1u zi}^hDW_MQ}giO&o?sT{<2jZ68sib-A@;o(WG`D~)3o(KmdTkxM0qzHjL@p z1hebU7hvFpg{X1a2|ZkHaZ}c9;96^o1cq#Y%YVFr9T=m{3UWN)Lcl(#ca4XI$>YI$ zx)Np$e1!51;jB)fk3CmE$+~AKv+KfBT+W(t>@rS9v4(hD%=h3He9dLvx6;{-@)-7e zG4BZucmrcL-T_cl!35dA5L6_Er85_^fV-wFe$8F>`S26|J=_J-dNFV&&lC?Y7=yFp z3G3XAp^5KU{kgXb986WX5ZM}mdc882_+1@7@f_)FqZD@SnJRj2cEpd97T~@=Q(@h- zB+xW|!>tIDX0__07@InueM=i+`o>3K%G<+mUP%Y*w;sR^sc9(l&;_yt(s1va3<&a5 zSbn?|&!67MDlnRDZP~?6rkv%B0?OHum9=or>KHSgY=prBC-BwW3{-gXlk@bmX6`qp zv+&exP;pKHtKhRRN5Yhq35$6~`$yLP_#&s0X9{K`o-_T?^{imEF1*X&_ns=6*z`>r zwU#KOu1zG{3FYjm&0Z!O%4e5uAA-k?qhQ<%dB`Y#2Z2RnAyVxH8<%;3T@2XG{_B#o zIXIOG&Ma0Ju-rsgT-^>9v+jW7fe9@4W&sTG-Ac29bhcsFM}F^`2J3Q;2-@D42}ET~ z1W(2s2c6g<0xjnOo z*uT^+W_8jDOI=oDTh?;Sd8@_VD&=xzywfbf_Xf{{c0%o1YxD@JV$Qpj*|~ZDII39< zrr#sMFy<`0xDvsR@r(+k|1z1uMqPo&i0y)vwrPU47mv8ZohGo?>>}GY-GwDRY2@~1 zX~N>`#cZONG%g+23%B(b!sCbSpv%vnX+zmuXXkvb*ScO%WMvD_qTL|Vb~m(T7<1$Q zJQi4~-WDWZEP`#k8_~cx6KtQ`GgqN2i}{ep&V5nD7o#O`Ro)kPxbZI=vG*sdYSjcT zyO6#77r=86rm}-e->{n#w!l+qQ`jaRg1Q^M@Wc6;*!)41wK&a(NohQ@-TI55H(Hz% zA#cv-kplbnaU}~WUo~ZvV+CzVT-L8;F6v2v0jB|b&bF$lU_ly+bC>KkH*m_8u5Q#EvY$f zh+2mbE?vUKnN2X=sSb=bM#Du-7nqz-2hwwAV2P6pU2$APDbIuG-l-Yv{JI$S{eBHQ zHD3jl_vvERpgne#xsj{?{}K1*(O7h*dWhz94(mk~FI*B^qc{X+k;A_dV8!?_4~cwb>7F?e>`hh>$Q@e806o z-OwTEud0o51ot5?8oR&~9TC_g=sP*u?h2-FP5`+AET~jh0|-Gs!Q6+5OonGC^Rifu z4b64}pI7^UvyZZY{?E1Wg!@itb~XxrqhHydWbo+EQ66^WbA#&O z%F}Jo>rpoRauWfK~2u`B9|*x<*V z>>BPXD;4Gf%O*y_OtAtO9rc*?cqIy4K1qW1^sDgF`j=2CT@p=-)CNiW7J#jN2f*Wk ziD1m%A1qJF0aA$vz}v-Ffc!Xd*xEb=6mFOWXtgrXidx2g+hogz7!OGoIpDIi6C4&9VEp{tS$oMqcCkw}`vu=(7bI>5)|(jf2}>3&XSuQRrkiD(JzS4lANpv6)uc>}7!_OYfYg0uG2TU|aM%@Ogt7 zFj=}D6sv3p*-STk{p}pkG9?5U-YI0iiI=k*4m@OD_hhif9a5mFw1qu-tb^yYd6>5Z z$uQkpE;2!x7a5JD9n2x$INmMWea!kr+Kh0>GMor@*+vtCjaq|~&ugu9{o6VoG#{59`(8}XXcZLZwpgx|L ztW0>FhKjtTm%Uas%>#_SgdIEk^aLm)gG`d}Nw!?$5-Tm& z&r}Nf=-y3LWhIkxczR#Dm=lwOnE>T>-u1faK%uEg(3|4IMAdy@*9ov^Z$GbKxwsmp z-Rv?$mt*VaTl!e{qz~+-se&`Vb#qu#`6B{-X)I@nps8(I5>^6a&Pf;F18h4ou%#%#=A!f5*k@$#ORF|yif>=Jip=J8S= zp1*HBui$5{wbjvaJjr8yy!i*`@Ul`Duv>Q}vs0cKJAfprD7!0ckK#0@$Ncy#4n8zd9#2`ZvMfXw#nj^l^6?t7XsY+Amtsh&SHwy zr!ft+p{#A#WY+7=A|_EWfX#3AWWNraWL6jWGJj9sV&+^t&J0AbY-;#jHgH!W8(R#S z7nK{?-tanRlYKW6{z9MyIbU#Yus?%|%SvZ*Twn8UTMMx(XB=WDw2QMx-t1&0Htl2M z9CdkHT?-iAxY|$p~-x$`p0yGN5nDbp!C@$WeS=b0kgnNiLAKDvrY zv-`#Sygh|ynzD+wcVwgWdsPkgcJwxO=_fu@tZL6xg+wqN$7-0pq?B17c8qzrp3ht~ zjAHi83lwnFh_j}j3-Js^|GuHcR_#{^t-N!cmPtM|=3J)KoNiP&+>@iAB4 z5~*@1;E;paFY^WGE7oX>qb=h7a76nKA7@;gwli<1&MG9C|@yk{J{e?lO;z9ySp(1XBRy(aeUZ80z? z^@y$SsA8Y(^kto^&%h?<6EN*vhCqWd3)uf$51Iw{CyH&$*r`XA*|w`&nXmc+Zd3uz zvv5QdI+UG&*Yfh=JeSkZ+Bt>UrFfb*X{rcg6eI@>#Lo!c=i}IZZ8J1{aVj##lM!D; zA8LIQhU_;PsO^-)w)ShW84BMSMZq)ZXG914W~h~&`R)-EesCTNrKiJ@CVhA@NFByV z3PGc-s$k0IDwfka#9EK`gMgD)L8nL==$PTd(}Kfln($(v z0kp6XXQS30WD6?A*_Hrh*fL!K?l)I}*+q#UDEu$`xlIE&RCL1kL-*mbgeTDT{aq+x z(Fh&bI=KDrHz0$afxLy+!4eUDFkGY#e)UO!hCKpaRyvWL@Vbd{<28f){BW@O4g-1@ z#xb)4yO>gs6sBSEITX616lE_sg*s3d+bNX8diGypU-{|7%z8Drc&R$v>=F;ZlTa8c zv>iImnZYcKn#4?*v!792ew)=?b%C8#UBE{EJqkvCxB<<)alm=y2e!rhCcC$=hjo24 z4dyybfuE=ALU;FTY}d6cw*34~Hd61DfMiJAC)2j2Cz|&Dvmhj`bOjXrL{71PoNCfknY` z!0Wa;NWPAMg18?mdrgvAH`u_)yqyI@LZ?IVrde>Zc{*75G7Ct(P6Z7QZm`pYblHM? zlh{-05@5(bpOrj5jZNal1ACh+c6*r{8!=A^rQ3gl0b~T8+ZzwO;u688rev@n$G|B;@P7|7f%hc@rzr?<6z@o`?-mRr0X0`B9=1Nl&6O%B{QG6-g8 z3A80@8-eYy_tpv3FRZ-hAGTU_Dg!77uLHj6reMOm9Ci&6XSXRlWR7|VK)QM&Pe0wOzL43 z4boVz?}_YMjs<#xdH6WHojrX!n$_YzW2U#=V`lp9hSAHyp_$WKxYRHdo?hY)kBF>< zCz30GrsFDLf4+!qYB>!GC!PY^w&sH?*JR)-Dho{>Jq61JoVU7b2G}>d1^Lk_2Hrmw z1$VxcvWpi;O1Zd!{X)9JLjR1#3q-|)gruZ|{$C$WAx6l5%bE>AJJ!uu>%Yx^2C%Yb ztgS69t*xxBtjy;*SPJ?4=N}=V1GE3}Avn13yuh%~KjVMpgXRCs2TO)$B|zh_^BiXU zr#=Y(V_frv_!&X?LjR2am2uwxr*SI_t9cF<|EY1obp1aQ{f`^Znp$f9`To!NU%4^k zf98gjg}EimmL81VkP z_5U~QTUlB%z~cXyeM@Ve<^L!E|4sh?3-JqVo3k_j%lZHJ^6wuvz)Zn1{QsSQR@Q0+@|1ZWbaKoA>sKNhH{QsVPYXP$f!~0Kv|AJC$_fPFj&d>J#ub)3wJRs=s`^WZw zT>t)s{AXrb3jY4f_5XM3KVtzbtp6!N|8e~P-ukn^VkXP}r}_7<#BXhGZViC-e-r=z z{`w1y^UTey1cCpbslV3dGkJpb|1azBzY+gT7FhkW@c&Kx|62Z8n_F6%TmF*+|55$> zH{xf(Ompl1Q}fSQ&g5BGn*X=&zk>4nKk9Gdy_Y5bZTz8GQc zDkQ`6L*YHYk&)*sGMj!q1? z2K*u_(OWM^Ty)zQy$I5zyR4PjT?u0F@RP^5Y?T9?yrX?? z^Gjg5SR6N$REeYYii!4teByLr42hmIv-$GAA6dh6i|W%m`4Jkp*LG7LwIL&Zu(t z8(8t^EL{335~=v*!;kOg10^9jRN`HQZb=WJ_Z{DneaRUlzWyVbqdI|PC=S4gs3-w9 zUj`5w8N++w39@YeLHN300R8<2(Am$Nkx~^UGF!maYrnk;Oz9Y@ z7JeSRXB7U)#c(I)1m2j8hCHTc}FD+{? zL=`T#Vf%heg9Pmd(VTjCvu>IG!H&% z>xTEv=%LPA*_i(Ih9)jz)i4A z-INTUS<7AP%B`i>4zZ65<2YOSa=1QD1u83CMm3oMsEK@n_e7e=mpnb1-rGbfV=73Q z`79#0)e&TwG=Yq5eR$A62GL+&it3uk2X}e2zIz23$dh8iy7{CR#ln_7ru5*&Wz;k2 z0UX$KrRHf$E@6#2$j(#)5@h~~OsZSV`Pw#+NfX16Mb3O0>o-U=*3Kur{)NQ;>t_6D z_!ka3%fol=zmR2afM%+NkS#jP=*#MOdZ8^C&i~nlWX6pmw{SHoCGV>+l@9{JW_PSdr;=!;j8ble<8{()P4lht(R|IJ*r^62xbMdh`hJ3Nm34DoCZ+`9#d3aUO zv+8glnml!=1S_=@pvOfE$hF9jBXUXXri))ldHWHvE82tTUYJBKCB1^S?$gkU?;F8U z-7-o##rV&rIq+{d9U)Jrn9<#l3rPQ}P||Z{6iVQmWKvrlylLDF+sYQgug~w|#}1OP zqec=2|Co$~9iF4z+Z^Fm?FslqTr0j*6NbLu`HLQ28Aah}5IU`jM6aJ@pe5szxVoZp z{MBPR=XJsh+q^A8J~IWJaT*MD4)h_D7q{V^hClFNW&&6W>tMl30bXKJ5frk?tgS!F zhlh`f(%c`<@K<@+O?9Y=sxnjAf4K~s7v1*3ZXz!e)$nRXb$0_uWMOLkBzTzJ>Yy+Vppo6iKNoCS=I3viF?1*-!^S3C`3p1Bf{kjjBA^{LcoRtuH1?Im@J?`iN3bzg&p5aWHMhofyY<0JkGz`P56IvqWR-@OOP@#8&a&e ztJY;3fJ>nTTKxAmd7UVZJQi4>+ht-f@=+-fsq;tS?N>pTK!adlp()v=Ezp5nsm3jF zJVVu19|etqe7_m`gsk{J8$O)ZCa5jBgl?D_o?~Inf$V1H@-fzkH4ne4@}oiWo_S-;(&Jv@Iq$_ zG0)kEpU9*@gE48Ec<~OMb!jcFRBof`v&~THcx!|Q{=x&Fe-fjLi!eI5kDRTsq-Rb# z((ze^RL@_Mza~76l-KH^o%ORov983dU*);(Sdt5&n+Yuk@6xBhWXmNA^FSpf`%Us7LD-a<5Y#oEVg*2U9lDhi}Kg z&V&IZbD@-T%^cxPx@ntHTn3F4xf3Mn>}kL}gd9V5^{}&K)>N?i@~nliLH( zw9|ZcmRq!WF6K#OT3)G#^?`!mbg==Xd# z`gtTDS7&*UleL5#l&OGyGeV&Kh(Fm{t4#Gt3D+r^K^AoM1Jk0lB+E>kK92Y+sDH71 z%dFY_i&H!3JvlZ0OSqY;N&RMd&MUBsjTtt$HVUWzN&vzErSRcSb+CEF91YjH0khgI z{Hrtl`E8xM_;ov)=%9ir;~5Q%MSBH0f*E+ay(o=ud`#Zw zU&aEaK>S@VnluR+P*3(Da5=Ju?u}bO{_-}Gj2FIC;*tWD>h(hVY<3gjbGmeAYa6A* zQvA??NWS`Ch#y|bCX(?>;oVXtYImUsj5WQ6P4cqO5Y0#Z)qC2w*FAbL_HzOTLzh@GdF)QY7Uz%`LVj{ z*DPwW*8vtMM{}q2pK&?gL}|+G1`;}9FX^4Rm@AkX!LZmGcZ51J7e){Brp`UZDUPb~ zccM~U?(u`{Gfg5TK`FFg?kQpxmV>#iVm0UUUsm&UQ(<(FA$Ho9f$x@XW6i&6(%#oj zOym4G(A_hItv%F(zl<*+SJIA=)}M`--0vri;m)|vza2Zi%fm~R9@aLip62$XY2mrS z3i#QEQ~2G*D%O4OKK58_9a-htONtvWkcrTeU8>d&GWOqMr(E&DW1BaT|Hu;*ii`ikg%di|wxmXI!3(nX6h=}4&P5TH=X5GhsMHJ`{twz$idWe1H(h0&o zt|I-J8LZ#cTV&OR=cH+#4pkiz#+9j?h}Nvtthmc2`m6}hdBKTTXpa@M)T@C}{N+pVP4%kkzlNHS@+wwYkAH`@RsA=H8RdsT)(sexU}qkt@BkqygmK{>2tbBd|rLVm&MMJ z!un;TG&qf%dw+=iWqyJcDmlqmCzi6qcjqvQ#^trH&Qr+Bcc%#mRwX+(5n`qC2zLZb zAWBImIM-=+fV)jQ4i!!#mv+yjhO3tnJ<~+8V4?^6>zO$5T6u%SKk%S7$F#@}%@%Ml z^gZLhdIM?Hlmx5FkKkIn-$eZS0ce}&$ML@^FrT;Qf&A!6+{%eTWMEj6ggNhFraTB^ z6~=<#ay2oi_h1d)QAF^_#j8*uNF0@3JINkuv4!KgVVwTag6l~8$ozS6itUlkV& zf}i??k?EVu4evY-y-M%Fc8^ND>9QK=R=xpjWfef^;XGoc;R3uj@4!j|et-%ASNWi~ zD!a?+AbxyMzz00m0%qK{#BPSyVYoy-r!Zzk>?aIBB}Zdo4n<&Xy#k7rtOXYQ%V2Q- zE4Xk?Fix#%#&+}l;j5%e@bo*~+LH&CK&>m1_?EaCez)TQdrags-dB1B*Rqw^uBDIM zn<3y&vx`Jgo9(ewiVV?PWQ*?{>F3SWTgiLfYF}M{Oacy!GlEzm1RQ?18a$o}fZk{w zY0AAs)W{0FIDZB_{;UJc4~qfE-JWp@(MfFS!~h`ussZa%?c!SO#?@*Z^Rlv0-Ol#d zy#zIJim+UueUs?mi!8-E(Dv?GwJo1#;;9>YnCIJ^YC13HgVXamLCwQ3_^#>x`PBbQBW$<)+NqH|iBd~G|yU8!;8mM;2%PaiNR$^$FOSMyAe_g271JFpc5cidt_ z9HeoW+b~KAoew<$=4__KRf>jZY=st2`1qlZ$ z{4TIY;RFv3kAokCMsT2I7S?p13S+&M+2A+AcyRt?oE^;KjB%%cQ+6v}5-9_CbBx$H ze|ZqHrWkl?E`i#bgSe(Hn@P1f>70E4Af2eD$ ziUuye1InHu=#6?eRG+O%!XKI9{Mq{ie7=&f`QtCh-b8SMX$|A=;e?O)%HXD{8*uBn zpIG#(BUJNEA;U-1>CXBtJpOtKPG`Gu=|~NfR*`{KRrm3Y=oCoP;%YA(JcE^Gp1|)H zWle-Gd%e(Q3m46uB=X%hU{Di}A0zo~L zBR`{k!7WK?80wmYTz35g(Na@k-0i1urpz7Ie|i}k6xImN?^onpwn7Hv(y-wDHXQe7A6|cJJKorIoveO5h&-lD zVrD9T#z)Rn*j#@4qbBIfI!42B5+Z({Ag_NbmaQKTKTKoMjgEZy^}HQE|*mm`(waNfr~@Ky3e z`0|DgEG=&Zuamr?_Wo4Hc*`};a)U6OEb{=*OVNa zV-0)viXr@aEes^_@`3A{@ypRDwn0}>1XRZ8N(+W}u1c}<=ysy0IT z+AdhZRtmg+b`S-ZWP{&w+4!Z>RHSv~0(>#}y*kfRmcD;0P6F<+%srKdB(Ossw{L0U zR%s_7I^xaV6~xw4_XKgrZ(xZ>f7xi!0C?-78oICF3_3>?$a6{G8o$pyyg66BP+DC# zw_}qFO}tLfyF15##ku=<$^b+g&hH?3D_+4|xC0p~JR!E@E%4OW{m|pN3Gq>wj#qaU zpn_v!z9lpr4i29B+S$CWmr^}CPMg6|>3 zJ7I^-;XR7r>|9OkqZ?1oq6#3{_Xpbh+Sv@OPbQzYzUCrc3lp{d(cE+!UtGtPAwB$= zgf(-tL8F;ozJ86K&W@pj?xCplcrn_(HbbF6C$f=QK}#PD6I%Lw+O)} z*Q!agk>FW5{yA>^;)9LN91lrntXCzinM>vB)!u-iT4^SLVb%c zwDll+(hrctK_PN?$46kWwE};gBuCo6=m9a4ad5yc1zu`u6JRl;uwvsn+SOhFQzECJ zXyGW}C9a4+ESDt{oZpjAvhsAtN@v>CHUd}O)<@HuH$z2RLBDl!0&%TXqZb#N(q7-W zH1FXXJip{HZ35+FN-2aBKp@zz^9+o5jNvaML9|LfkzB%&sB9WXWFmi(n$qpm@!)Ih zRxJiK_kG3l+h;&|(_FH&I-P8lHV1BIhB!%nGiPIKL7vR<0EWc};E&`fxIQBeRPB3C zG|$vPOS+kcj{8B+bWI0K%FonJU0n}-9m>cw_vb{&xDeC)8}MSEBm7z+57t&j!?#Zw zVQrW+-C|TTeDU#tfhAGkP7*sBv1P4ZHdCh7wD7;f((>yL=+j2rj>M%i~qNt$-gN|Q4OvH@yu)D8{40r8;^BXUK zYq6=d<4zlcj@38WNzdZpGmV8X?}X$z>##p=Bdvcb z@l(kMJpDPJ;icM0V)peDejg}_nRAuoa?eJze^x&=pJ4~%mkz^GgG8vn>|!0(zF}Ro z;;=^d7xsI-5f1r1j3+V&$&0KgT)oW=JTq8;?pnm-m{C=@Q(GGj$c;yQWpVO4wh%1d z6i2+D&!OL1*1)>L1T6H$g1jbq@Y1p#ZuYkUwx^>CSoX=2rMiY_TE`Ktr~W(e|DZya zztl$?Q$D~Gs-dWE!)&A(w-?$4?V??BTSy__2j`jE(2>U&7UQE-Sm-hIlQkw0W13_^ z=yiBva|GSf=YwuOd;>>EA7Zxw5yE?vjvZC%I0qvWIHw^M2n*)FN(XcM3?--~|0|HU zdq>_sROTPi{J=>#JAiTbEAWA}qhR8~Qs6f`9AAu42eO;WxN{f~3Aa!5@;V!SEO(qv z>w8BAg}#&Q%dd&2=5r$F5lj_qqe#huZlGvoM1J;NAxB5=0ZGB$xGLL-P5|Thm$xIn zs8<=S?I@y3k_YK;;bf}P5>BG4?C8e1E6`yPAzI{piPRO|CG)FVk?pvN$av}^c=?Mj zD_=Iqen?V)l7<*=u}UHCgQ+lax(VMXZYE#&hymY6rG%Bwok95lSyU$MGqY~I3eb?c z2xlgWz`HJf^mp$e^kv~)SUd4JD7Tt~u6%8y?LS7TQ9v_YNTPtdK(}GfqK#<4z5?ug zag-FSy@;RQI*6}3=5obd_Ryd`9?Ehrz?x1qq_*W5{c+u%FDm1Jy3!s1{U%Rvg}9-~ zijTR!Wldb;iHT&K7eu)Umte@jBdMgBBdAHL9A75+3IRqTE6HFGZ~5N@1&l^SQL@^^@j<11=?14FDX zdsny=JgpBG@XX!fZfU(IC!D3I@@jva^D32&FW3pQm5xELw=t|~hY1l+T1kDDN%6_h z1L)IjMEjn7Bwhl()a)`{9QQM(`VcVX^M>sB6?U2YN2_Ge*lAxRl;wmzUAj}NdsB|p zlM^B~-PLf=d>WNrBt=z#CCU3B4|f^~`0G^Ex$#dR8Y$n5q>uLy6Ll|IDe(@*|JqNo zbMBLZ?w_1ba5D8;6+?s0SI~X&m+``=czAV2J~&!@QJ~?JMb;~?hDochb8TY3Xv+a@ zzLjeQ{Hak$qN8JI&t+Y-^~`T5sWF1jdkbNu9TQ2sPAc3uUjsbVbA>H#AD~S6MACNG zpJ+(OVZ8~N@XoRuP*KJcjdgCNgWhRW$@2*@zU2*453PdPr8|-3f@ZL(H4jz_-k}e= zJ!LH_oQaF(EAln<1lWfgfbX_wFuJXWtb3eHAO9+$4QbP<)FfAWE#?H)JZQ%M6D^>f zG*;vvY4v8Gx`>cSj}flW%LyOQ+6~LSZ-E=nqre{9ZE&r40ZdpF3`)JD;Nvb0v=Hlp zuZLaP^A} za`tC1jP7ouH|AzBMQ|QmPb_{c6o;3-TMX7jnc+Q$iph4#6lghr zDOhc2gN%a?fZdCi!>Yn0(9zotTRb(fMz1#Qe0T#r;Vav`+VqNS+#@gDeVWd{^~K=&TDb- zp&o87n~L@c-a{9ghu}vJXTVs|F>qG>FL0SHg0QgyN)(64VKf<7-MIk$1zMAeAFks8 z?*=e|e}tQVEdvd`y8@i{5_rz~GW^_l9P9s_0T!jp;+zb3q#ksTO-MKi$8u|kdORh& zk~PS;&EasquNG3Mb+a*x(6*6Vp=x8M>56{qw4j{L6pfUwM0XS8(AwS^#PjoUB=#)^ zwVz%j(2~3ex1N@UUINTwkH-x3xAk@Hu20jTuZ}59Set{^$4EnKF(G(?-3X?~5)zS> z0ZXh+xWRqxAb)~3TJc5>75+*FGov|zf@hNSIkjBc?*wcfngFGZj=)X<$D)Pu60j!k zC^^~l3TEqHgNx8OAU(+dZIlQ=KefaJ0!rhc|9&s1bs!U(XqH2z74482*ALYm3pBaj zug2ZqV%U(wgD@gTjmBS|LC9tun;SRB+ibY}9jVWoM7QHoGWBvJAXD9dVd^VbbKMV` zDC|HHa)HRq62osxKf(61>iGAl82K4=363dGM>~7gp|3COpw~z{T&tW6^EI}@zh_^= zNpUy9)g{$1=gktBuudHQ9M=OcH18tootokNS%=XkXK$qLp@bw_$Dp!$0x~zZLzA-h zqmWE(v|MQsG&y$-228pMk5o3pO(HkZz=R6auuusNqy{37rqB&URx~%?Kb|uR zcY>+*^I-Du4Ei_Z5Phj(Orwrk6SO3StZKPO)KCcZ(49&<62d`=!5z4(Q4IFgnZomt zt?c-27p&E)%RKy=4DvSIV;Z0L!S}r7_8*x%@|z9-Y1(trYaLJ^8d61(l5F z`iJvzpMY0ItY01H;&qVMSq#3vE<(n=F(A$5Hwk?5nsKP*XanP`X#`8EkjsuKOb(xf4Ey=pFqnjake}34&yGi0)Gn*M@g;a z=-JnE5bvnwV&+Q0^81!>Zdn|u4HgWy3k1v#gXv3aVNm=4zTs?vuO^Gah9mC8vPIO! zU`jB~O_|Tz*Cx ztzH;~ywhi(A0iI`blr!ZD~~|U-QvjKO{{w8k~&n3GeVy#gOPN!JaQEoPlUgBa%MtN z`1-8%OzXF7$lnj3+qyn*`O;x7r%3|b*X@HMy#lVa z56E&tH7nnzB?o<@u3orDWj0T(*-kXw6Bo-uvDobmnhkj**iR{kf?us={Qg5I`qio3rKb5%hF4K!BK(!#nL@5+4{g#plVf)4ZfOSC6f%u3eGKj zn)jlIC03;Vz&mJ>JYAp{{gs>ffYy`*PC!#kQfe*VZe_L0Av~zp3qQD};1cyFE^dQ> z_f+v3$uuj%rf473YiP#nue5NHFEI2_-Gmo6ZlpTy`=M;H8eFs}51u|B3Ji)f!8;j_ zrzGkCqGnBF=lM55Ro$z=<;e;-GvzFh4MW@m`MlagS0RWCxD3<$?O@QsRQO5sEm&pr zgw;M$0Avb>;YXplPj%~?63kUb|T#%bnHe912OkO;dBLQM@@LOIHX%+uX z-t5!?TMw3D_bJ!8Uz3jzed%26RNn{5pa-y8=gn3n*O9KziJYyaJx15c+3}xc;NI*(9#wJXVgFBI-iBWKTmt$$gXIxU(Jwgb+5(^f<8g{Ju9HH z)_SC(R1K4b6i}Fv3*w*J1dl!|#y+e50Lkz;$PL{=l@I+<$6|r@m}4|lrWPnc#+@jb z93c$~6KJTv0c;SQVJJ-*MPgRY#3pGwnJ#XL>(BfGZHKeTwc5$3^0^C0`Z5Y;CA7jc zqJ>A6)H1Tu?vp zc>FmEduzv|!6U^$;vOZUYcWbN$l8U;-E?<`aQurmgq`HDH?=+DwvGoI89U{y> zt{=-ceH+Su7JiO0a|7t9o6pJD-+bDuB+hrWnhkba#?mrui+%K;kXA9lI|O$CdQPyW z8Cq@R$kYTXcd{DD8ZD-#byu;1Lni38I|U?rZ*$oXT-jEYDPS&DrWyCvQSIMD+)?@TCTO(td(LTcIPnsz8mi<3ni_Oi^%rZh_tDfkf5J_Xm&s~?m0dmwbdA* ziB%T#C7*4RI*%{YA8m+2SZXQLQhQ4r>2L8JueV1 z0T+@SL&56>tUAZFDS zv^aeui7&hY_9dC(lP}+pF;Yq_uMO3>h)ts2hqLH`qE|%X?l~?dQbx(tMNVP+^5axqx&}WwIB(G<>%qO zA4-7g=v(%1{4OG27)m}^i<4~RKs49a(n6WTq;NP4x93>H{AI>aHhK}%S(}gM9S);o zPWm+cMgf_1j77$;szGi1UD*0^9@-sT2CM&E28SyhVe>!__i|Yu)c0|)8J3%6bJl&D zO{|*>I(%jU;m)l?%Z+1E>Q5z`{I;Rx3H9t6k15=y3waMm56VsGg4*Nd(A*>!n+FZ! zZ9YmG+1!_NMs=rTz(k97RIs!V2L0RxUgKaW|0f+}?*GJk*+&a-EFHun*A1Th6^{yI z%>OTn&OENB=MUp0NzzIi(MpT7-+Ru?+)5>7X+w4;gbJk+vb3+36iIt&(~4B*oSBPC zB@soGlqFlrPL_P}yTAYL>vjLS^O|$+%;$NY_am^%(-rvspaqRSfW6w4j~l$|v0u}5 z^xw8h^w?KRP#Y10HgMUV#ljZ&ruAZRMT!(Vi;>m=iktg!4?10a6$f63fM0Znz=T~zP$dqtBOBzw zc(fTPNgjr6yw7mYeLoy=wH02Q^%&k=x((^oT?MmV^We?cRP?LA40dX*!AJJ3L7#6^ zut_}|d|B@V18#JpHG$TG{ulCsqh3o9_OL}`T2;Vy_B@#V;0kEv(!s(B07)aI}d5}%~uxL&)y1uR1YTNj8H4qgakN1oQt+uI+=U$E@0x+E>~fK=aRPqn zmdafZXyT$Ovyt1qL{vI^A4+T>T;Vwb?D?ny%Ph%&2@`{`FS`bXE%^jowPWGyjWUF9Fd3!>wW7qD@5sq&5n6rs2s#zr4UCt~!+G12k-48Qy(dUZD5y^(%-_92I}=mk zbjDctZ15?&(5oFBY!&0OHpPOdOea>xM;+Qf(IH=lm*bUQ6NH6@g-bVy_5Yv{g`y%<>`G;D3%~I<3*FKsRE!alR1e9l$DI;&98~Wq7>w71(=x z3Cd{tjz{F&adE{wQh#uajntlktRsE!-rdW&sNiGBuW3EJI%Nl3tv#LC1V6&lm)+;q z{Mdm9rzf)y-)_Oy3Qf4n@;*LjbR3SJiy^y?Tarw5ePT*wk_RsRM7)j91Z!P-)n8jK3_+CbZ9eqX5K`m>Iaft>eZz5sg=+xBU0$tn+R`)RDtTiEwHdLgNXGe zW1;IbB*iMCn(ov1!0YRr%xpb$dTbuK z-#Z`2cHPBYjn>e`j1L2DCt>B*Y}~Qw8dO?djCUS!CAadsaYgcPGH2Qfu&AjVI%-cB zYIHboX+xT%-#V3a+P4Ew7aeqLlQYs51`?0^k|g?x=znsl9tI9kTv7TuZdboGxpuaI zeYcGx$sdRVnsbV{pBlxz?%w43$XeugatQm)u^}^#F2};gCa^1dH4g1L zPjZF9$V;Y+yjW%@)ZJ$!%#v{z&KQay{*IBbF@Z-S8{5g%r!UBuBTslEJ%fBGiXwfq zG8ui=OKl$H(UZ58({l@s<5O}u;NhWa9PloQP7m3Hj9s6C;TKvY;@>!&7&ntttZgQV zCi-N8+&z-|SReM?UMSj=y69TyZ2aow7F4Ai4J|6KVm0*--0N2ZAS1GiPHg)_E2!@Q zUw-65=f~IJ`Y+WaX!$yrvPY6kKOp*j>k6UG8yk|zG-KY_e5%ST9XIWPWFVU*3-anj zo$g+0G`T3q;HHG> z#FmZ)R-qSRQH_Lfxp6+(l$}og`q&a5MvXl8*Ci6OirDtCqpVc+FG^>P193NS#md&{ zT>EB!?u&~!=Y8Ze=O3U7y%n|~;QIz3OWaj_RZkl|n;A%kp4#ENHu)lN{9N2p zt4-=uyWwAXD{L_38cg({i24Q-)X<|%oO=w}y>>m^f9?eGFV^A+@mM(3e9XrX66k6{z1jc||ST%nuyO!{EQQ%`E`YR&8fEr%9lJkA6QoB;Q4be}^XoQE#&#O@j4pS9tt$1!H~dahs%0nW=>7!m zTnNp|&zk_JRG){YR_;LN=OUmytsN##e?a~I+uhKnw~875vbH`Yb0+O%n+*>1ucTgy zo>SrcOR)aC57vDg!-CbExo14i7qjN9mY8!mqz*q(9>cP= zDinMr(eO@k3()rTIie&w^r_D@|xVWfK)*PJBnaVkTnaRR7J$~gEf-g-Wj`~Fy*+1LJrx_#K3WIqvOvJ7Tm(~FmJRo073w-#Qb zt55cFioY~z>sjNR-r;ZH%jJ6RY2rAjENzCSy=pZ4mxu>fhGYF&Z?I$EE_lO3Ola0r zPvP$U+=LxJX@?u_Y|9!myt02Doc^_hle=>s`zdDOpihqU3PWWsC@C8L)OT-i=zK}{ zDK4T~)C}=7#qG4uwhWl;6h&E?O{P=Kj)6d1IVRCMn`>&Gg@Q)1>L0b2fz|RW=^5%D zDZ7+Ql)A1T9hc=sZ(z~^H(tlSp4Ux{ud8QLq|`uh*E)7I{yNySY&*AHqo0D|mB74Y z4-B7plG`Ws4CHP+2JcT;O3%4xhpWr-@ClK|y}f@EJ@j{kj%fOi%5XbaSLU;j`OiIz ze)69-{piCd)+}7q6+1EsQVWNHx@QBoK*S&WODaTNP9e-pKPPT}+dQx+z!49hs=&HW zj>1{*?(lb6T5*=|o4}^J*Zj_51L$>J4Ve8{O-pHt!4U2&{3U;jzMQ3vi`TElIecT< z!&Zh$MHA`vtl4y^V3^vwQ;lAayI5k@-2cDM;+)pld`ECX)Ua{_!*~tdJ6q@#u>Qto*o^Re*UOOe4na7*C!vikD3p-lfNfnQTHSDThS4MJ^w)5$t-4U>rRla^axOIlCjCE zkK8HOQ|yOSN!CI75my&nhKH9x^OHuUbeyXk>pdXSq+*C>MJFB;Fy)SruD&LHqq7#rebN8C!hYSJ(+l)Zy@Xx@f?ZD-J#PZI?z&aM$@$L^#!PMgcmn1IB`Cu3({ zic8M-;EX@wTM~&bZjmDD&d1pF&RHP)V-WnU z69d-XE z^@gN?*FC`fv8sh@EN^qoUjp&h6^ht-r8^#UKMxIhebCnN>&RUZLemAmpc^$&aPHq$ z@Vvo-TQ$o8o38Lh1H(S3H8Tu%J(EN_EAmmV)=t6I_i-frbTdgaiY4)m&Vul>_u$$? zQ+l z{C(+2GGYgo-BSnbWhh)RMHhy+y@S5K2)Rf|31)1Zi)NZ@;p;co!-cVbiHv+I&H|*lRTKhuJ+XnchK*pZj1E-I(?P$P_h`HRFfzYrCvYpbK)>zWQG?Dsv_p9sNxl{W zD_=^oT5Uh@-8;#|$f*#A<~oS((J}(Tj0%*-14!#ZFPe7vHd@%$%LVjp<>YndVSbo7 zp%>00#iJHjw>bu9BOx9S6DKd%=D>NWE6{005tp*)FLF^GggycjG=px2Rtkz}c#W4ky5eA-6IauA1}8ENoCS^n)9QdA z+(kh!d|Wi&m_7hAWkR6Z-#S=5J0829&qf!qCJInEkG99pMA}ap;q4uw+;a&)!7rE5 zhW&9{db z!+9t0`lR`&E?$7fyDq_R-J_^l7>8z6HzT$27!>KIh<5K*L!J)|QSlc^EPf>#E*|^M zi7#%&nt}(w#tzd{zf6F0J{occ&pM$>_$!!sdjYwuf0lb_t;x-4i6RX^o$C~3LAz&8 zK?g0rp@V5RQG4%Iq%Nd#IU#EVk6(Ws=s|Mt>rXHp{_B zuFH}cL-M#j<*@L`*Hgl#{BU8$;$#@>Bbt|P7~sr5cH?M)1Ku{k;oH_TCprHwgIu!hW8?KFlNly0WN?Qu-djBv zt~QTEXXH-_(|_0tPZvHTot7oUw%eI}v1lRrOJvFNl1TE;qyuN$T)}DQrjxJ`Z8EfZ z3%RO(hg6_V@ zLz~f&FCWnK`xB7j{_`|%LmtVJn00+0u zz+1}1NboZ+;e*Ov^264c)CYUPEyFRez%2!@D{QB4#Jf?l#pgJ4e*?7XWfD25*-G{g zKF9vOnYbo~CjR%95YL>MY`t_O`Rr0dmgLWYY2Mf1;XDJ;oizbQ_eP^>^S)s`K8(Md zjYV#@q6}HYC$6$l3VH9T!xNM0usS}2b#9BJuTS?A^W+S&`S?kaALop{?VQQWtuxT2 z#KZXgvr)LS&=pp6YQTpk*^uV~G4(SDju%uzYe@l&e{_rc&v}G3e{=;K{^{ieM`h75 zkvA>;$~uv@TnocK%_k$vLdkViF<7~53myJqEx>nb!M@Z(z_da@7AER*fl*no<@_Z$ z%U~ip=J*WN`Ep3|Yz(?KYa?8-)rPFO9!bWwi3<QmlhP`v7lSc`S$mwgHHDK8lX74~j2^hupn^O-JcQHVSP2bX zjfCfNcH+|glZCSf8cBm_F5FhN5nj-u1P{?MvOp)A$S!vwGO{}G&+R<8>$jxflF*Nw zy|n`#tSQGI&1_K04jr`U7(rnJX7I_Ud-Q{^ZX~B|yI>!cDJtPjfE&W?Kwg3YjNP~o zo2>W(7Yg>E!S`#>g{WQVn%P`5S!+8=o&SRqzIDP(ix){mkudANK6ur!94d%XcxHx& zpnmuVko2Mp(pFSpX^u|wx&=`v zOOU>D(Hi%G3AhIEd?J> zgrmQ|4UzPxB$P{1;H~a5?$!-Sl6Bn``fOE2*UYDqNyB>J-sZVAw{aWLGYil7?k|EfH;R|!5KmR=zYH$x_|W{dZJ;8yrpKLo%UA* z*%uEAbYIR9T-Gs$KrRihm)Ss^ehAT9=>uqKX922IIR~%EXEU;wTv5A2FkXCLitH}U zh9D_g5G%i3u>XmNAlju8w%(P+xxQ=QxK)y1|9`b;g_a&VEN3AoKiz@!?!H4gM}MJ< zgc0TNx=`-BlPEY|tl?fO$)T?7Ddk8<_whW2gV2t$mODi@cB0zEVa!Kt8ZI^(ime@{Aw|(*mM8~1#iL+ur4}% zc^y%RuY}ExrATl0JH$%}L3elP!TsDb>~}1d`yeNS!Yxe&EmwNckwe<3+<6YaIdU=F z7ti1ee{N#UDP@G876cO&V!FD5WHMB7<8r2Nz{4TH&nlE%fArji$p4xbjvUCj$UmXsvq?^KiiyoSEd}9hMb?Gd#1@M^6 zWOB_&30rqfC(G}xLU|++mT%KSMoC*ita~!tcSJOMe)tuwYpg`$cRoPzTl$D+!XPR@ zR$y-T0?+xi2{r{z$M$_gTw2{Xt}8nO8?4g6&t@M+5;q8xIuea0$UaBUL^|>Kja%4k zb}m})?+#b`=fP}^5PW+40&Y!;#&t$}iCTl0@VuiXXD)686#Lz<(W^WpdQqYdpK$ac z{Tzz?vkcAKt|Ty02Dpm4!u8HN0~UBKfw|(>p~;yNkXE@8%?^4GO!Z2*&D(Nul;bau zC3F^PQ3~*}A)u2={{yPrc5eRd-%vi~DiW()h`WyJi}(;q#4|Aw^~YzTm%2|;uc|!x zzWFjv60Q=l^W}issUn;fFoaWbX2Acxv{JN^3An3%8EFa(;NHQtXulv)xfTo7@03Mh<0(jfqZGQWJq;~?Zh|~a_M*QUEGz2b z!}DkNlF;W?;DVnYYOXYZpOf~1R?ThD;fMh;+&2xqjL3!->DA~=*(}b)buB(>r;bcC z8R$%GE{gmtJ1k*nJjCO{ zg=>0SSe6ZUHZ=!#4mbXhq`}+m38*>D=9e2WtzWGSTwj5FU@lfoo zh@tXx7#<5L2QCNy;;1kGz^D%o@m-@(v->-w@?;&V`96*xh91Ss7N?+j+cMB4K>}*m z!)W2`udwFdFeo?ifToY!QQ!_SflZ8|V6;99$Ng^R{Li1@{_9pptGvz7lT2S!abOu; zJNW@4FF&6(-w^ck4XjRR_kM4t{=93NQ3(|Qi zS1&WUZ9?X!MlOhS`vt^$#5kvsH?-p4g|JlTa>JuDdSEq&z){CWpnBY;USs+{P`a}O zUfKDtA^-O?_EWGdegln}vBb%&v)v6&13OsJhYr@AD}RY(gs&Nq(8-gTFbj>pUdoeR z_M7>!-wL_x+ekZwHBr}V^ugD?>8!=6kNijXtf}I!wU&_~U+cM9hv?~|3CzckY$o>n zWp0`0$%dCd*Mm2|mAHrB7V(psN-Yht9nk3!b2dj|9@|&50+c!STP~QMY@FIL2sZ_n z*KU!ot!h7uMW8qsSFG>k4nwtgG zj!NS5kBe!42OoaDP@FoPHkmQ)Xa?$uW^mD*g!&GZYRgF~PmwH}#4HNgSKm1EJ#_2- zV0m@FazlYR51lsgx12fIu6D_~X6W>Kd&BB2ip<_Q(rC#VS+45adSIaw%`ekeVEX2j zg3DQn(CJVBqpV(8=U?y({{DD|aXSd=%__1HIO|m3AthJ;9G*d^?OwAN{s>s6SQ@Y% zVN}4QA^t~OUD!QX#=N2H*=pSTyU7{JGQk@k)9rF%vdxxAv zJd-r3xvbfPMCy}#Fo?1LXu0z7KHkSy0n|BdGt-&)asAteqfA-V*yTabV^qNHVMhN+ z8_o97=*HPM)O~(AZ<6j05X{eJmIs~URgM`qWL5@Hyb~kT=-xyw_-Kh`PgHg7#&ro$ zBc+Tz5i$qphaZ49{BE(D@B2Yf{1C7j>9z1y!I$?f)Ml=Bc3YM;Tk^CvNQ3!(lc|Wu z-N0$9Cps3irNPlvoL~7Yf_AyVW9VYx<*mCi(VzCwhLp-B{0{lx`u?b|jG|EteP9<4 z{XQhgU7Ya~WT(|K*F6B(A;AZC9m*OSzbpU?HmO5Xix$cyva-HuyL5e2SX;yPkQgAR zDG8gVEx?AK32(lh4EVGEIx{XASN}Xxk{2jz&Naj~;#s?o6X}=LM58chjZ zFFJcGqbRdj?$3uCjCSO79OG$=TJo*Y*|urG;ouP>TbfSX7R1B+EElX+r3z%5Irf0& zI81yYK@u%CH%PmjL$Fc-Ie7-*TR&5fMZq8Bwy=zAD^26PJtpBnvue1uNRlHDltmmm zX{4+!FK9d}D|lKZV!^(dfZB@p!!?X0DpnbVeikQj`ko=KXY@E$Qa8hQ>vo{?8j|Ff z$}HltISP9jI-<(G$C;EeWBgm}Ih5v9p{bh^s+FAxHM@IZV2&nSeZC(I)fm%nmfhjr z{8AxQo)VdEy8^Eru_Q&OHHl@<6jbUvh?O?A;vX4dc#^Cd{?;wxXxhH#+>FzpL+o|f zb?Fiwv%H4DOEZD@7G;5R&O8A`=lVp-ptsL3M&wRB9zN`KSQ_UKY1wR8)&A&kRhMmFic;E(f){Lmz2H`IUc z2bglRgcvQUMsL!Oq9-j7$0TYa36*KY^MVe#dTTY8^e36+rscv*(+at{OFm+anjtV+ zs|`L9{R+8(q793LiL;z{uLHV)x`tN5w-m==rP(=wy*Oe12`3aPkvhuvWBA zySm~Vn3N3U=yS7h)J&n?JQ*9+4d zb*VAQFZlH4bmF7oN{WV#;(KpuVBnrgFgK?fa*tVvauvG)+Tve!1>~Ka@z~)-rLDCemtiK@FkZ=Mn{X zSM*23JGyjbDYCGs!TPU^v9pdjdgBqyS;hPSy_f}3KYHNlJvvBk)CJj8Y~|ujF2LJ1 z2T>q(5u)*Eu;NJ~Zaz$tSVe1MWaUUUjK{#q+s1+12GRNkT7XAeLty4+7aZZ2idW}# zW7m=ne84gU`wpn$$>L#nT(sZp@~%*7rm>utWEnTZl$LFt%p{s-Ui367!lTVJ8K-T0Lr73s5@aTxVkTn6c(ityS^jr zfrSm=z2qgF)pHO?ZPkW9P6l8b8$~WDx>nRF6%7(+dvRT(GB_%w3kuD3xrTZP@LA-a z`4KsRs673O*ZPOz3Qi7{4pabVbx|H^X%DEcUdELir-6OkByn_&7cQy*blVGvKiaaQA9sfY0GeKL{a8L1mT;OxFvHyJ(h-pL@-@BdoGGqp1j87D`y?bMexRK-_l%>DVbw!hf74u~C0uW%FwEH6aKovUZ?$)jD|hR}Y-2 z+KP86oPp20bz!)oFW$3!6aBb7lDlb?LCTGy$jlkGWP7hNOtyvao2e(9znPDRz+=wq z>nQd7?d!H|-1>~X*soXUvgw4zQj*M-{L z&kfJG+Tm#Y{XbRkOTU}bpWcWU7=&Q64@IJPERPJ^BpmOg5nR9*W`ijM;Mj7mh&9 zoEJdrUi1t7}{B=OUkY#fH$r>DE*W&nzr`?e(=v4i?ui5 z0IdyN$#p;O#dUkmrfDyr#Eyf~nN2`%N(5f&)QYVR7UI4fQTOki3%J(z7@4zOm)mpY zC@0^XM)qXY@C%E~5#M4Oy7!@uGh3bkTSjb1vXQFruzef3F;`A_zxFeBdE5>&+B?DL zK0cwO6Y%qu(d1&!M{d*deW17DI=H`L5`LbL4JIlFa^DuIqSc1G;iPp<@WX=BSUIK# z-`T4Q69$z~%d5+*<&il=U3NB9nyV?|zl!eSuYPiG%q_T_tq1WdC=XXjWI%(dt#I++ z8N63shszwU0r%ntxdiJBu-#$?^uFr?2JShqiBV(RdH+}V9`7JNtK^DH7fI6+F_Yl? zPzmg!;2`QxRscxrnZa*5o6?%3I&i=6=K=hwoCa$4Z?m8s;Jw>;V$v>Z9_?gP!|v*48O zT-@|t4JL`{fyiS4ROh-uY%C#;!@FW}udWo7J%RDlr-kgP7Cm@2UmHn1T!AuHt3g51 z1-Nlh0*%_#P@cOAG@7CYZJri!+Igpdew-8>nmrc>?G+Josco>QPYFzk{}0O64#1R! zH}K9SGq98IVb0^b5Y@Mqp&j$fMa z`rZpyOgR9np#@{5dzQVMF_AN>5pt1+SK<2LyCS4l3VypcUDOY$0B@#Q;i>&G;8vIz z>o>;)yNpP}$c{)ZqVpEiIBEfX%l>fRl9fTObqd^OErE`GQs>%}6S1O(B0T*=6`s&r z#+gV;LfPG)xh)QpV2e;1=8hSone9W+{`E#6m@JD&kClPPk2hk)Q&V7QzBJO)kV3H@8L%O)5`BHL}rg@N5(~hre>F8;^4lE3IkE zq0`uwDBuwb8?LeSF|*S15V&+Zh*>sX0IOaXbM6M4aD=5b7}@Jd>!(FS`qyEoda?|9 zrp<(Bc!jj%g-9TeV{mNmD80EL3Un4{Lendk;f4PVLbKu&A4QG#GGke(KcB_xC4!wf@8Eg%~#-lopIJzPgWSpEGyH`B7)mvU-F-gNJjBU6pRaC(PFKBxcCEg^^+6bUo-T`u zCKtmw=fdIZ*S+-Kr_ZRIUz5SdZyn6ar?}zE)=%I{m=MZGXtKPAWl#@EfaKLD={~mr zSp0pEew>;}$M|}~eHVT+bjcw~BF_h^)m;F0BVX{vi^|}j;#vBM(N5~V`3?A4RvKJj zE7-lAa`Z=_NLMQ=!v=+9N-?8_QTmq1j_6LuW$I&4*ZnX2?5l_R6(Zp|!!~%}{zW`6 zngFw3`_Y+IS+LFj5i=s6&WsK#ahfTE(C~T{oT+;i?pPm>ZtYG5Cjz>`!B45saj^o; zM(l@o*I(q;E-nQw?tj@GFKyYGv5R=8dmn>2XT-n~v!$$B`(AhotN^M-_TaVNc6?l1 znydPqPq{z#h6g5hz(U*&4$mEeyq)*h{CYXw;(}G!GxrbNuW1Jzhh)&d=5%(+lN{FL zYYsME_6!7WErx3UZK4xY=E0VhQ|#LX*XdM01^R-p5LEbohcoA11Dk)|p=XyYfT!;y zK)-Y`gjq-nsdMHcr0*0Oi7||91iCJ(`Ycts6nOcRjP> z%phy=vxY5Qy%P2wis7$_^kyoG0dkrfhn62}gGC2vxGiJ}`0;Wr^RZ6~r96wGhbHhq zoOC%%)hz`1D)CIF%t?GBb0VDDuo+s-j=~MCzu<6#99p+Iq8gy*5|;0l3Q?J z^JB`hevr{0VlHpcy3lZ?VFG9IO`i%-j}hfB8o47HqW$&aE_dfzG^~0w4_@6A$sM-J zg9leypq>&2zT0sch;4q(gf(Pxi-XRyZKHg2AXucumlC)|`wUc%a|ABJjj&(d5Nud; zoNf8}8`kn}!Vha1`t7<_@UlP+%=@sD4YarqGKYrPu&Fa>tLg3dKw2($`zD{saJt8m z9#cwYe?FEAol0$ABr9lxcu4QxXTR&E3R z=QnMsd%S!)Y*iOGWOR(R{2R};-@1(--mj!qbxBiS`liEMAFj|~p%-^p#ODfgif4x% zXx1a}FkXLX8}s_oRF0i|n!96NKz-e`ke*`8gQPr&{hRL1rfw$iicUD2X;nvGi&cXE zz!lcxj~Iq8e$c9N6t`9^8A=7k!dLG-Y00P8*j;l6AQ=Bk-Hh1JxW2pw%Bp{|Kc8#j z$byHYerg)MD9D@%Ff;|~r)B8B9jU-!8-#+w8cK8j9BzDHJrMcnX=-L2ed=}|v!P0t z>G%b>kVti~zu*HkA;Js4c{hh_KT9ZT%$~U$+z-_jcd)NRZ%~0p)A(zww5h4n4MG12 z1EMIJyW5D5*w4}2!cH~z zuJJiKYR5!4{qPCq+EPvC{Rb)TLWd%A(s+c?3Y`Zf-Coi|Inm&?wG`=CHkW;>UCG)| z(_sJK@9ff(s&H9NlYe}H{)D= zd&v%ZqqGD(^fjK|qeA283vt}7%0xE8e2l5okAq|2IyEWFo_Rkl8cKVdqf-joVc^d7 z;B&+h?!T`?V0oJfmhLWLx|dvK23=G^2TEn5THkT}3!z-Z=wE7oaUyFvUW?al$>(Qw zsE9anij?S?;7@eXW+p$51j}_q9{8q*^r3_i?#N?f=JC$kAnJuZOgUP@>WW#wfyY|p z&*|yxK|_`v_c@2fw`)+g+I&3Hb_mx`{ayEYX|iZEc9ZKk_l$CC`^}WAi1OjDX3)|e zp{!3d%bq=y*|0I;ClgImT+YN@^!XQFSbL%yyXH?BwTJ7WT6sS?v!9WCRm*c+fcdjz8-fpCfQ~9i~b~LqdUKVZVW=cE%;!~~Z_t|$x zzVi;bhA|VW6d6_5p}LqqGK|CSQN~$Sfo~N3i>DqGTg%I_s<+pbqba(s9Y0%~l`Fc>JET!+X}{5tQneD$T~BsW|K+LD4+`ah#+-V3)6z)(^9p0I^{zkD zFF=8A)jBhQ$08%`^P5BUSgIyO<+yyN`SVPG_9V}L8tx(OjeO|!zqs@+A}?m&GMC^%L5@0CJ5>LdD(P+yC2B=vXu6} z&|QC0ZUY@%xQsR$olU7crLTfiDCY4jw{aNhsW#{;(_tmodHm+mC(ymiFO_AV1`&91AecSqJ_XZm-#lePk zVxElN(bs&pf5+J=@-@`)Aq~d1xrdU{ux;q9lc9qurZh~H{$6{tD$;8Dl3p=m-9HK>m0f@2g2B|5Zd*!yU)V_4kkNZ(zsn^VdkE z)W^*X<6pXeq(086n!mEctYPa|2(^COlRCp+t0`ZpM|_oUJK5(`)WN4GU)dX}PSlJQ z1ys?0Cfw!(&xUFTdFI1s7e=gSF{AkS1oPm|ZK{9U1nR6tFl8ea1Dao5U>@VnI>q!N zdWo@a1NHGty~f=!`i|mr=JGE`W}&M&_hnp#-j#TSxp7&AGg)?=N^hS?$4y?qbe%uk zaB_hX9Z>JraPi?PUTxKC{+-4$s=U*Z_ickd{klV`G+SfE^iAIG+6xzD_!yy=e1ARGg5_+w1O6=`7pLABvht zS7ap80TYG%hLteLU=_rE*p!>pZ{!bWM@YNVFBJNAShsX2u={PIQZ z4rUIoCF)gO!UW-AWEx8D2`C}zqja8~fIq5jAr=(2>)>;lR%2$U8k4Mr^Z# zVwY^_!o>F=#_9^W5^pO6la~mmeolvF6}srh%Vcb|cM6i_{e)=-$@oD|3fy#X77Eo_ zL-NZHa0YiT!<0vI@WZ~B^rd?X$Y~QXI8-Nto-QmVO+ydKk~t!d@Q^OL={FHguABx( zB`sm`^E8|gSdJFkZ5L$(43Wk!ef;@VD7xA+fvioOijQv=alLMe!Onl_IPk7I^#107 zEh>u9xR8f(>?R`4ybND3=8^R?_Hc&+BhanlQZ!&>jZVSSoYHwW)L6ZTs~6pgO&%q| z5Jw)g@l{7}%sBWVG94d0afp=EOOrcJhe*HULNsrAH_gMJyf=m9EqsiVoAzRxWs}kR#ZHKq@rN4{d24_FJ_O4; zO`y`EBskxy9Sb^~3BA~tL>-zTVzG#3u~m7bB_n|B(tSjR`sQ)J4xa;UDh*t@Xil%? zcL_y5aAmc!GtqCUF!Urx1sntMANom`rQieCG`eLfhG#P`|&tesKwSwed@4+##@9@~w z6Oeg(2=~)>@Y}+4m|u|uRM**&EQLF`I_!TGorhmd{~yO&no4PJk*H8w_k2E|tD(rs zh$y3wRcMgNOnazkXi`x!($Ki)eBLJ_8QH6>FCvnaY@&YW_b1%Pz4x5Y=ly=Yp3h(0 z(1BApvUdvia%cm6RaTcbCJ#kG1e@+e`2jenT&Ma_OYl7r5C`iX>dtfFVZe ze0}e5YHjZY(`$9$MOZPpIs6lTHC#dHRVb4ljm>1B`W9Xhp@8)gCz8akOJSq?5GaZ~ z$F}{Jh10nV$Pax)6=F`)_jVa{P~)|Hjam*t6UZ=*BkeZ?6%zRjU}(nqPJbUQtN zTZ5ZC`-f!XDmi$kqs~4;yOBelExYNkDs%4s8E&}4WXa$7M10|9C`~;2fn8m8Rx~{1 z6S6+!!KK^!VtXoot0a3?EMI zMeZMuA+%H?ie@mC9AS z#Bw7|$Iu7Hi=fN@9d|iqIz6UuDvAGjK{BAENMn7+;p)M|+~e5{yE}dvn&p?uCCBPg zYx$XUu~reDIcSAVEOubmh?m@;LLweEbqaU+6$VJQ;WOo`kiEHxdwTUOt@YK$g%Q!L z&Ag+?#H5AoS>eKb9aun-XAAA9Z>7&I?U3uZdua6c!>HbQGRWdgt~m4sJ+1bVTQj+s zDav5Exf}1HJ#*gC1uM0oq5Ul9{0-rAosSWF;1!BX%crLvb)YRTj-%xfx!j(C9#n3w zindm~MfRkGVB^xD!oVGCM4o2RuP}Gk)>J>j$!F=E9|JVm##2Sr?1x5QK|0zu)cCD zllMuONw3P`jF6mUcgP{mWLPNl&KrhBsj<{sP8~h&*iXe?o%p+LsA!CO8cloL!)1={ zVp0MxGd^p|B@;@Xva?o?ZxaMsj}`>ht&p4(~jVhGsM|wZA^Gl>qeCrS1It zOD^o1@HcFRa2Gp0D-Sh!f8_iQroe%2eKaY!0(~7KWH2hfGX`mvsB=gYJ9hUTe%!K? zH2jDO$Zpi+EPW1w4rb$-2WT}tDEE@)bSfyVEh5rn6Y4@ zbrC*`njyG;AB_~U_!{r*`24aszUN;wUt0SNzuvu++nHkocJp4~X<8Nd2xA4CFC9TL z!VKxpfxq;(d_9)llYycQO;Av;Iq$a8hp+yx0xnNY<5g}q@ZapR`5-aO3LT`4v z6;#X#q(6xUah9Kj{+3olUUVLMu67bqy~W`0VI=Qg9R(FfuG8nE_VY@id-ybLu_>xM4-7S&K!+TWrZ!5g3-;WQxUI_}W-{HWH z^?Zx-Vt#!?BhQ8v@F692LUy{8%gve#E@3zMi-$k+Hj|2B>w^|QN3j6*(A z_OWi#nH<9(?Qr6KT;%u_>09{L-SIq&lKJ`lalHGMG+x3hQYZTz~TCS z*eztGU%_&iSl5OBgw8-cMzP%C=BJV=!6*hd;#YTRYln$5O)x_;S_nFQfR*7b* zALP~sH=`G~9)jT|dA_W682=T&5G1j&*gZ;-{3mLoo6Ty#esMLbu|#kFcyd7^%n+TENT=Ky65n|g%;zcc(f7^x6b~JK-@i7< zZ$Ck{O^yZsxk`L}oIaoa`6~_gyaJu4&%m!c6WEFmO`==PLy`MXEl5w#M*r$ZXQn+Mhh%+g$-!Yf?{yqf0!8Tvl1^J8^GVVpT;Lmi03W4+UU@5DR$hQ zLW!?;1>}(z(7#PZ@b~0$MaCV--rTE52lN94=WP`P_6`Lt$uZHCHy1#+cm@Aopeeuq z(|wqIa5Fw!GLFoce3@7Jdx~#5d5GUSO`5-9r^yd?P^vi6osSrIj=ZhVA;WeM&aLh* z`|a{qq(5LoHawP)Qp54&-X;mGI5&k?+~>{f6xhSez#OprM`-V{d-O(95LiCS#4pW) zi0-fn#9@6AyTGSbB>r>}-FZBL3wTtrMs!XLIWItMDlMc+S8ikL-5}!(s2ILs`N#E>JfU?eV!JFirxA>4jh-_0^yB zOC`{k<}GMxfD^tq#tMJjHiujBIhiVySJB;rCD0i&jIW+_6hDu?N}Sv#pivQ9(UAwA z&>S6KZqL3f%(YCVuCdSPapFb}_SJ#cygzV!qcnEEuK<=)Z{kU_tjN?NLu{*RPUdxv zq^=ioh>=|iwYE#*kiZ6=XqSZx)rWC^-fBWuf;XKIFrA2=DMDo285*3YK>EjRC)Mxe z$oc8n*j@W8?zZS;6Y{U&r%fk_v;RrrIB7JjfAWp>tkxmMH!7H{xBqCtcnk8k(-rnf z8$h+TE=;ggqx(A#l9{n#q}%#6o-!Qpg#0h`du|*&+!G5=w9Lt{SySPm!xEYz{=?~q zPejY^J*IQ^m_TlO19dPep)ne1C_`k8b9`gCds}1Rz#A|8r1%*U6(>@+E31gqBnh{F znJuXVDM`|?6gVZb2eRj_poiIHSl(9(ir=aTH^hY8kE+I7H-wN0g$bnAr5R^GTt|;6 zD?? z$X&%Vk37ydMwO#&AGolv6`{2TGeNI7#xIXj*ItdH?eueSc7&sO${K z!&9H&xTI*@GO7VfDY@bQ;vCVA-Opr6>`_axBY z{+ZhD);!xa`UehP`kO;Y1-9Qq^0X6J{spndo;qZ zUX3+bn*ExsTK<}=Z}>{2;vW+Oi;Y;Rb_K4Awm?nAJJHvw5?W@*)BcWK_*CaX)OLfR zJ)S}r=+#xcG%_16HMfQVVb?fgxEFRC1S)k%6Y|dqK5mUdYR+BdJ}rL0eTT^DP%{CXyE4$*tf$G9E|*M|56)W*c|Qjs@DTCr9}?whg@}>z1;;ZvD{aG z`+{tP>P;!McA(j*Wt+d_)Q$(DJ=dj}Z5MKwtzCNLc6>;qNPnY4SIvBKtF;t;{}G5R z`wCGJO%Y}OT-uK~c zeXAWGN?F+b{Z9k!DUqqmwu%vT-I|3w#KRR$hU?vx5c7cJ^iAe3f7`um#>L# zr>|5_%I3u+^|b4{FVM>&Y52p!Ztxx<*#XbRJ^{TWtwc(pU{nar=oXiXq{#O^msoQB=N^^YsCr1-j#i_~v9cvQ0$B`NAXZ)@1 zayqJ)oEp;dot|y1CcBrap?ax}%$EJD8Yi|VIb>XDnz($03vsyK=(uQcX8qLdJ>1M^ zZW5P0ZyZBD6xy}gy%BvHmoEzQoJYDAjdU9Qx!);UN}2hzPDvsfk;U|XPh)le<~E*| zNYxaTXW?H%n;p;1uxdOSmw>H%JsZ0_C)diZy-&uj2oQb1mqc?~uMp|o`3|19Y8rm6 zJW9^SEv{c@r`vEpZY5bIk!gJGHMa5A^$t8Y=y9Wuv>a2Ap2VK&iEH%hOn1_ESVna6 zq#K?G$Tk)WIlj%hXPJdpKQqgW-AH%eBd3h)Jq?rn50Ir}?lT7i^cZHXFXy|75$U3s z#+iP}*zHo6C|)@N!L1Y~a?LN%l5Q2zNxS)yYiM2LzA@KDEtx4KcY`zPYIw<5#Ae{` zh1&H?^F;P9ous+b?xu~0!Z?Y_XGK&T98qtVn(J_);)(D+xGS31cCw*Md8Byb_C3s^ zW_9M3-{Bw5)eBoa*J|S~8|Fp?bXNMU6_mUx;^IwY`qSO#g19iKtCv zxhX6NA##(9)pd}>RqT?SIX|yn>Uy73$s%JeUiyNay}c|`F)j@C zjZ|Y^P-bEZT?&kL=1HeV#$n%ns7$XD{d7TdKAnd&rrAvldOKP6Eqws+e4 zTaM@}eHT4_yvug>A!8UjwwZT?6h2lQSHe(6Zt)oET5YR!^fQ`_tcX~QFSv(-P=t% zAH3kc>v|!@6YBW2_YC;AQ^?A$Y6P7DmfRijkkqf4OoPszrcv}Dwi_W!erfL~q3wR; z%Ul!k<9-^wxOWLeR_4&okkdFIVjH<&y^cJ)5lt+n2cn60hmbqF9K^NjjKp^`W|1}B z9{5dY2r*F$qpq`tW6ho8k+$}oX(a;;! zobDn^eC}&JdHURgXgMw@nzt+L8z}phFFa5-pDWM3>6Y`wL;Og)O zF6@05sEzFh-=CvkNaGhWXWwGF&7&G8T>6HPdM+1gu|i-e+M->7A}T#$3cR0cLqG1@ zMJyXV;nwF-(5d9TC@lAmE(KuCc?W0$?TO$oWf3q2X@RNsACEti# zZy!1^N`<(mO6aez%CKeia~k(tgm=6(rVE;C={uL(@NvBkZ&oTr!m^u8QFQSm#i%s&Vplb?cPlPOrXoPwGqiO{jg8v6_W7x_is7?-j|BvJVY`EYC@ zG1dK!U00-#*DqDXt}jG<`3X<{h4u*E+J8F^D*DQeK5UO0|GPw;4kf|H?YZ1`V@0ut zlcCt@nxnY-bQ9T|JcJxvSA-&F6w*W6^J&HMYvjsteTeyW7=|5aC+<LXz3$WT64OITHL%z zrkZsVub^^bE<6iD%mT=!tR$LuH5G5p+5+*b?$K@es?<_Nmqd+-!E@ZNfk#XNWapRT zBkt*3g>)G>4Y>tcZ?4g}`8<4|NZ{I14Wij13zI%Mz_jsQxFo&?-)ZTl^Sndw5;Kf- zhrMC_b{mm$GZ9*`q=en&+z3avUxJA%*MJ6chRo$dAn5%tI&0Y{dj4Agb?yVstt=Hp z+zW1A>^_-j0G%vVtEsHVG2EB0-#$K|i+YfwH8N>X)W)qcVpu#@p05)slV0JXD8F z%b!clZYfiP?wxQ%NnoC~2hjMnmJp*mjsClG6jYAs^L>w_neg+OINEYFiCGncXVNvW z)4GuAeA|z2+&T}d_Wpvgt~&h8KRysKc{IP!D-V+V_hG}7(_GsAFs4OtxHJ@OM75Pa z@cT>4NL7(G8DG(f^#kA1^Fuq(x}uwGv}-vmuD0ZdbA%7rF64ojP@M6ggcir8<0n49 zXorj^7XQ(PD^piNr^-bbFdfBhX=LcF=z7R|qz=P1e}jR(0YpSKQNPw~ST`{Up@Rlw zW6KRXc8wnF8{7)9ha9<+rYVs9b$~OD8c(;byF(vOI*!iW)d9P|fa4~s!=q(KVN}{r zdNt)ZJhMrFnPyX=W&9UdB7838c|Yjt2!W|2`<>2z#|Sft2(oy68>>-PL7z*t&{eN8 zpu_bkQk~I<7B(w^hQOgVDf}U1S+nqvaU=0A6&~#JWnfp8J{*{!$t`U6LUjs_lC-80 zva9<9sVX!j>pDC|4Yx6j$}xf0U%a?a8lz#%px_&!9DbduM-H7Fr0%Qy=(eA#f{*Jc zw|A5}^Y^thCwD^@M{Y0W0!I|k2mBiv-d{vB6n5a(bEm^O-Rt-Xc7@Q3$H2*dGrPJn zl@^-laZ{4b=%uOG1Xg?${adYuPu$d`g)17+;dyzqbo5%P@v57S`Iv&m@|HOAnlv* zmGGi*W0%1U!NV}v5r|s~PcX969-#2k=jfm71f-ukbLIWA@H0sbhM&#FTY@&@xa|}1 zu_;ny_ni+yPt%bskhzHm6B$@2GmDtoxsfmB-RzaPVy+=)B%4(1PqoQTE^i==K1??! z4_rqQr==6HYP}%|Sr^FNz5E%=@6N-kz3Y*|yYZh#V|Ts zjeOO%AU)=N=$AzcwU6szF6!(MSy#$2yQXbpT=lE)@$BEU=fWL|#yY_5TP*H$Uxev2 zS;UR3_!W@0IFffMF+r~m1(=e z__SI=&cqD%-NHclE&){fE}zZ`356)_bYN$_q|E^{hulBnZUjI4@npcN`QPnIu z?A3kTzPONx`j3&OIA1c^Y7U&;JV0mag~6QgIFQ*dxIfGMfe3wy=L2WxmpAj#+shJU zuxy#=`P~bW!D}B_Tfrs1z2g+J+N?_kJnESn`O3tj+<~l{F$pi9;6`FLXp;V=uEaOo zfXLko#`iv4XI>d=;oE~V;F{)pTzg^(8+3Poz6cB^epLnZ@}j3?Qp zSV?}IYY~1+)A3;BJ-jtF2F0{ohqN=VK(###l0Qxan`T>RuKvrMJ^hQ0GCoeTkKg33 zEIG`jkIKW@wtBR3K85CeoD{Ly&ag`dthA$K8?)oI`up^uGZx_*uA7!AbbPc{Ygn@ympC%}`ahlB4WpD$j{XUTwudkt?O(?5;^L z)oUZ%K9#k&Ky){3G}Xv465ay=pjNI1ntplszmsDHmOQ2YnHOo9(0gn(y-IU7uZB)t2M)@v zaMP!MqbZ#yK$0rM4C$`NBXec=4_^a$rCv?`o}M{BCwCOz{K$&8ifM#@t|bB&uNl;| zTfq3u5MCuihObJP2q&9j*n#EOILp=BXv2tfsurU{n!nb;%~l1zf7TB;TOG>^Zzbw} zxq_P=AI}{~h^5!|T%!d+-#D$_31FFP3oC;KpT~?PFe^_Nbb2?yc5ORw`#2g_w0)-$ zYt6WLZ*wyCvNBkx-sP6P)*!q z?eLz1^0*QJf%*)lzfYdE@u}SDE z;vumky~6iw6|4*LkKVyYdpWY?&t}k3^Mr_qa+3Oc0c*5)IGn$Cm3C?CVFTl0ly#+$ zIsSA6eg0xG)j!*fM{H>$y*s1mT8qUHyC4-(mW}2A8a=@kfls+<+*_3Ra3+6fgBJf; zBFno+=!z#AZX!!;&oTd{%8@!VCE`6v8R|sGaaZeh@=#_U=HDNo6Yrd->y}2q%#4|2 zRY^OZ>nbQ8HLfsGSL#V&U;^1w8P4g~AE35Nhf%k2uj#&xuds);558N}M5k>vgO;}0 zRLu|x}GWiAWwk)6<@;`z{?KaMJssd3PGbPmYdv)4XyZ7qldund&CWO5bnO2-CXj=C3wYxA+V`MlNyk zw5Kg1e&c8IX1~k8LGvl(7lgsSZ5ZZRZ-c)|44Ayx0Tbrt!uz~*val)+*NNoFlbN>U z&(c0Bm3j&e_8kS^?gbFLZ3lhu{yR+SWPykPdN=5UPRm%baz`hr?O#hyYRbWsH4mZK zGnu=aqRcsoqL^$K1IT!FgPLyCqI##dkW)s*WZQI(wCBasX-8Vg*xy>>0r%a+_ux7D z{lN_GiEIkh(iqJ)Sw|DwLzBSYeI-41TZU}3Z^Cz^Msg7Y^XTdk!^FlViA2V`j(8nR z#vZNO)NtmQ-eBVTQn4ZU08)9j300@=C-8u_bzgzEs7d^nT+3xbh!&t zOVQAl<5=%-61{QY7p+xth=0=8oJPHxi=}TmLV6j z%LFFeBu2OM2eH2xNp9csrJD1miJv|AP8NvwGpfyt>Fed+sNRTQ)NaKWbZnYA4f2l! z*Vsh*dF5d?WaBX6pSuxE+cY5MR|T^Eau%Pjy92kb6mUweC$Y)87Sw(wfqhq<#J1O; z;i?vy;DCKqc%7mmsucQME8h6xf#t2(+Uyglzt@ACx2$QrvfmSm^&p3gryM)sdRwNM4Ur1~0%N0STmHb00Zn z{+=9l8x8&fOS9%_F5b}Di#1aQsCvHzUS_lmwTucSlb)ZVBi*NqCwOvLM@jJ7u*K+L zjvw$ffmn6+W}-_<-eQE?VgcesSO@CzAA1n>(NI(B}v7L4oU>BpyLBwUi-kAy2p7 zoG5 zryL$5y2F-F)1@E5g#0bc#sA9PNlMvO?46-Y+&2GWbdfg>D;-0&R=46ScT36mr2cdBWBO+})8?w*O z#Cie$s7Grb4J))on}p8Co}otY=cx>tiibh&y4zH)ya?uoh`|sh($!{TNmr{EY^il7 z7pFZYdh)ScaOy2~oyjHS^Y{W+rCLetQ%BJl?;xV}c?z}j&&D}>R?$yM%Jjusj!S(z zmRq=DFQBP6$jR(Ve1GkE8Y`Q}O(>GXgNnhduDdVGr5KQrbBEFV@k8(k!&Y?H<0yUH z?Z$@ut$;~6DU3n&6I^331idU6MP5l)(pq^BZi?$$YGF9Q%^CQOwC6Pn=i5r*-RKNS znkAg{ML!Oo^k&lh`shG*E^g1fMYXh!pp${UI5=k$oLer2yyE4sr8}Rf98MRWN1w^1 zc_n0 z@a4N6be-E5Zby9~J)->y^?y^Om75Q9m;6iVoR>G9B-4M>hYMs#!CzyvAzKgJPCdaV z-d@61rsp`R!9E;&FrD6yJ;{{qQv#Pf8GiMKQ=nSf!Yp9e}q9`z8b%Cv=xm}z6|zDFTyVK0U8{ji|+ok#yf|7 zMQJisoK|%_dU>QBOrE}i4bK^vAas)5zl_DTm$IntfpTila)G(WQ^rZf5l`7~ORw(p zr2ReD1>4FKiU;mOi*_*TYd=NZ%SLm%ciiNLo`25GX5y*I*8-ewt4lnif`sSZv~`UVNi?%0ZGG)1S$YADnc)v-rfH)r{qx+PvSB#mo)UR2cyJY;cd_3l zs*=2EC5%%*0oX+3&`+}u(sOZb^uta~GH$?&YVMefmU>tavj$l>nz#brSyDu5?>xjc z4Gnmc7R6pU1~3cGLXD3s#F*crclQLs6;V2KE9*RNi2h4I@v-2xIhsop_5g>Ts)5(H z`}F+o3c8^y8r)9o1=nf4G^4_jP3%2@_bt#R3wBN-WBppNRt^uKe2uyW=~2^Z3vuAP zIpmRo1Klnk24>A_Zl`%|gHW1$ z9V#EWKzF7&EL#5;G;g)jvbJ^fO6eAAU9=JE4ZY|vFB$yv!9IF}Q30hxGdOks68QMw z0+=>YNW1=yYCJH)8do&9(6ZT(Rd52V4JzPgjSHF9u>xNF_yK>9IMX$c@?qLXFR;cB zsl4(Ln(*I!I(fbn=~?Cl%D4UKeQPW7GJ6KuWYx}jY6%RL`dThK>m-5;dug`HHEz@C zWcp~rdg`#Eo4YBnIz9%5qwk$rRMRC3q;6NEm?O>f0#S#>0h&V6RRto(xkB;89_(nF zNdku5hH6JMsyrf@7AsQpWA7v~`Hvn1UA7=IwK5?=qYfP0NeuM&dVh{dCG4clNT$5^j#uQkW~f8sbz( zA*r(yN#mz6q*y&1es4*n)62e5+4|GebPq!J9)yvp>?CmeZxZOXHB#fn(;(D48jfpK z!XNu>+~k(eNZEBO^X~ap{O-qkm|^$<3LU#ZGip4Mk|}0Ri}m5>r0aOuo`oQ_{Smjd zeZOSw+xOfomlEvr;R%Ts=C`rfh{{i|mps(0Mx*1-bC09d@MF^mR+$*liNB3-7)^Ha zXh}djHK!r}{k_}@m(LCVHEXg%y3U~S`6+C7mA+)x*?r7ggCN$uuY@)2Ux7xcuEnP= zzi|8_*Ure-K4GJ0u8`PNq@jjMKbUukcO^2@)-vY|?=YHG;he2*3SDj2!JL29z%Iy; zdAFZd%_Y< zYC|>qvp^qOdy`R-`YKe@JP%dmWzomZdT3A4Wsc1oO2aPg<&Kz5q*X_fXurD``!=iz z9VG4FRYvX7>?NrnWOvh7aqn9DPa*>hj_ zb5bpX?4f&;MfSpfVSRD7=+0Jo#yzhPPp$olyf)e60nZ1b-KGZ{zB-CY#G52s5hcg; z`Y3SOOD}U?GCw(=eZxhM`rYv4r@Oe{w=Uv_RxMG&hD}1=IEqb(HK2zymADzN6h#NF zxv_Ho(Ws!Jku|J&ho&so;jZ3KYoG%g8&Q?3#BGrau?kPbLmY#!)5AjVU z+U8BN8cm4IN^^YrhZAuaGm8Cte;3ELy+l_PDdiwA6$%|MZLv(Ey`s2*BrcicODxhaIv;0 zx}p4#9eDbY!J?nY)|?wnJ|>()U00*H=_AgG(l-Qg z5j9@uv(+8EwEDPk#y!Y=ON&G&H+Zqbj~a>6GhEpZH}XV1ZSBZ$;2w#S)*)ne4HNN& zVV_MBp)G3t>{)pyk$hnRyDRR1B-K^uO(<_>?z-G%-aNi6(X>fxY+ZI&^2XEzy^l&0 zc#L=0LqnrQiv5A?jsRD*GF45I;o;u6%1&S8&=rH$EFI5`5{cNS|2cCv-o0bCHXPz4 z2S>2RtF*E1g{e$rkqXw+c0B-C?cOf3hRJ z5o_}6AG26PikMXvvws*zbbOQ<{?!_d7ZnHNzo9DZ`9A_@$#NCDLAQpvzp)q{S1rN! zR;=XIHz_w-XRSuN7W`vxcMK3?-u$$^VB-+_oOu^3u@K@derzwU|{c;E1q3Q{6gHAkzJZA#L^4!bk6^wRB z9#xf02AdNppdGss65h50EY#!oj2oot2UMYP-X$z|q72(@IF5G}dEpuQ)0ir~1iIJv zy3^)I$!O=0Y&tsNAJSLZ2g6ReLg%gk$nZFeO;b6vtML;zUvdVf1pbEATb@BA9L8^x z?dX3lUlpcZYhli>-@;C&k!CmlLHAk?(ynv+X|jSzNEzssS-MV7almBR=A z%jG}*eur=C4$u;%ujp1_EX3se2jv^?g6D}Gc$vKcvIO4V@qKDU%$;@$Rai<-ert!- zXNsWf$|dMlu0c)P^{IE!5V$zd!<`WBd3$t2;n}8dT;&3W`!~J>nY~dX@e*C*-$U)nDqxyNHi+XF!S_mS@^`2PReV*?l}YsZ^N;rO zwuTY>x!ZmOFS{VPj27bwFXs`55MkHD>(QVnW&E_=obrhaSX8D!ieG%gWAscQp~M;b zla2Z6?jnAbtrp~LQo;XqC1EE<=w+#ifZ37_#aj(ItAFF*YI`^QXb6WjTPO^Ve*xK> zq_9fAum>r5gnN#kMso2glvbX?W*IN=;_@uX#K+Tci~9gQSuSMuZRb`3;7Ba;$A*V7dq-y06jlQTJC%ZE}t6yScxWq4kf0h7(^zz2P&Z-TDV+tHi2 zuMuLDpnRQ+xl~Di*u~LK=Wb@$GG(x={mVt-3atG-3x(Y6lN@T^$vhbM0QzcVcrEJ1 z8@b+RA4|{1O?4m8X~%w=HvTvqIcmahU9^$}hAEOuemcaYRh3+<-p!3PJb`vvoq*k* z`{B;#I2vGJMD6#Dgam_=!d~MuCYCFR?I?K??LCKV`FELq@jFiyl+#f~t`|GIbqw># ztAHAHji5zU{-AQr7;Js}DSj!qyIZ! zn(@?wQq;LQMBGdBkw)W_1}hJCd7 zd<)DyRSfr3TVMr<=#o8S=DvwRsOUo>HtWrT!5DSOWxMD#`=R`9>GCY#PK5^0`ja#9WLXVHD^MFEo_S)+pD}bvW+c7R zv7ChW&cUurJjwotzbUyY1ukc%VQ$?6a#Lp>y)*F`o?hcmC+%s3{*3qFd$@18I33iI5eKTPi;o1_i)Y59LA8A>Sa0fr!4xwzQSLBW zX`(@L=I<2dI$6;1^9Itps)WzKE~gHkYmijPdo-}XoNAByESeM?LV|)%=O< z?8Je-A4vU$N2H~roBW+7a7qM*xcA*F*kw}&d1Jp?Tyw%l-0WaQ(#FT|S zhzDVYF{+9#>QjQI<;C>5fvosng{)XcX{fmNYB}}}T2FJ-|a-@VY--qSq!^t@{~l!VbZgzL(U0nQ(ta_vmjw4?g)*3h#J# zJ>T(FgMW4FBbYzvfRQ-|aEri6Z(3bPvR*EL9M?($E9J!Z({#B;)08KcS<4o*3&`OrX8XNzlL#L@6MoE ztxq}o+i$VwHXmYmN}l)hjs^J;dyuYoWg3sfb7Rl!AjPlxv10L1GCRzN_@~xFSX2$% zzVFLLc&SSI%0;kbmI86GZbhq@9Zi+j6r*C%^ z8#f)p5yEUWVYMq0>rswI_rE}QurwdNq!%tm%kY`c>TrB?6Pdnx0b%*g%u}23^svV& zY;yU4{*CA)>fACL0DEK&?<;>cj(6T8n@a~N==$P4WQTuca`g86?!hi3?*MC0{I;yIkTLgYs>anO``ASiUC!e5EF}A55B>4)8yY-y7>(Xoj766h6ZP}A@w{eX z2Q+38u6a_9RbF=Bt5eV6WBQSJW2!XPUoeh)^lJx>IFc*z&hsV*PV14*$Dhclfu*?4 z>J>^Fj3JXBP9j5ZVKOk|B`&`(O88zs;FX=m_#PU^$vhs${8VGPj@0eA{&PR}Tf2tL zSK7}09#EzB)-9Yu{7u&Ifec#jAx~C)UO=|3>>?A#j}m9ExPo_pCDGo{h8y)AnCFvN zlv;k?X>n&k+QE2jJx!+h|~lBd}#oB`k&#M$FlMJ!DHmt(^gVxVzmry0s4eWIV zff=2jLi^4rz}rMKj$OC_z1E2)_k~&1lrx{v>SGhg2V))bQGO3Hd@FQ>V2;>kX0Dh? z|0{NQCWqygpGG4xb%gu;R8;Q2oih>_A@w7}NPk!s9ut#}r@Rfp9|weU*CPisY_2!X zY$15oh)SlmY7_}Hzer7MYDiF(6EnW79!=>H0Km>%d`6;|DnYL?`ZR{)8ecJ@L^68#t@7xp-`-F}Zc*F)emU}^?`$=_O2<%9`M1TE4@&yd_CIpN)H#U*FkeT zjIg$3h~QzMc=$+Q&-N@Qj#np;mwK6~a!fn7e{>XQ;C>nFDm$SyYj@Ge4i~Jdb(wm+ z3?|cT7vacZFS)(np5iN)=HNbEj1K8H(YZTsBTGXKWb*qEnjA5Ubb9V35f`4|+_vFZ zJIDsV93I8g`5r?4otU^ger6?wg^bdsF7DM^f4ccbIJ0!>R6J)0CVyw_Kts+4H!gd# zk2)Rgz*bS(dxyQqO;k>fjZqVupZtzqpHw3Msppb$>;^J**;dl|HjTW! z_>|LS{iuwEHqAZUz-H!~AmcJ~vU2Ma*8JWhcDj=;sa@ZTY_%P5qRkyfS$Qq?Y%?KC z^7_a%_ffu@%S3W=+uJ}Ciau)wsV%ObKC~_w|xg%nxBPRq>;}cQz-qKR;Sjj|A zRxepFYHk8vK?)wJsY8a%Mf`<#B1* z^D^GJH(@|<{0V0_-!Sa>_ZsS&6pGb9XD}I~!@$382i-X)jvFK+q4n?`R#R)C(^JD( ziR-U8Zbacy_GI~L_*0Wdhm`W1=!QLv>E6r@7g!iccT%|oBUNhrdn~TZTSd>M*KzrE zc5Gc_M*W88AGu>fH;ZucX#1Pg2D@+@q-~PO{*R*b@aO9Lkl!5`C$eJU`im%y@7P9a5F!!{%)u!(J}sU0150rB|+yKbk+8&AyhRcmYmIuXIf_~;7PIus4P1S$GkYle^*Hc z|7Lw~{2PdiPS~NYFZ-aWPXpyO3<6h06&#yBlMK1@lDq%Rgvd#G5$mRG?6T`KxwvT? zpZ=hLP7t$uKYFwH-dke!$tnojOf=+Y9yh>mTj!83b01djZZu+IU6jc4#u!e$X+9#e zRuau)C%BDI4QSH08T5OZIt_cS$ouXLWg7Eh$&j5&I6K^v+kYYjH|lD@@5yrfKZz-S z@y&h|kh+DRHqWW@ul-b1x=arH)J8(m{1;>kqfX~(m@#Ruym-$;{Y>uE6m%vihN)C| z$Bk2Y%#3*x%1`?vPYxEI<-J49c=9y`hd*{ko3y;R52EKbLq7#qtX%+7tFMv?!9m=H z3KxkH+Rcx9isylkuj%X1LKX5_RDt+xMVLb^JQa`i8!1txp@NNs) z)Z_wXS!RGgjE2+MXXyF~Iy84e137z3kL9;zv%ZdrtjS9i+CR9SIehp#Ejr~2`y&!) z_0^RoSp!N2pL*2dzCx(Eng5^?z>xemZAK zu6rTP3))D^Tb`2}r|$4Uqs6}B@eQDUX(UxVXhYuA-=?A4x~b1n6T$rZQEFA%LF{Vc zsnw`yLdI})Va+xj;gMq~?6a3+=BH(tRUlKBww_q zkZMPBRIN=4&L_Kj3|29^SN1h8Eap3NZ8LB_&5#cW}uI}va3cEYc!Y|7nrYJzwy^V-{SC58z&J7K*by zU`k1V=;$kAf?k*4HD;#B$bK)s!Zk_gEJ+tC_8k`{jC6wXJvV{8lm)?8AMYBMhotIk ziS%A$Hn(joJG^Qv)a8DI@7G%)`cP0SBR3x*D4u8brV3pP=Qky7biU zC|;{6gSzZr0GhHnkgn7K*F2ZQZ^awrUAsWkq^`o!Q{_0dsEO>`b_D-kWDg-%?C1*d ztvc2}iF6OkAOo|+z-_}9N`wDWhtrBe_$OJ|^+mj!_vVwLcQxFTf7hsGk|pXnT1{`< zGoZ63wh+@9LDa9J8}q3faj;c1ES+vc)?Rr+-M2;IuOIXv-(VJWU%ie3b|vxUKi9(X z6io=6T*TeW(ZdD~6{t=6BiGfFgiZxjai+s&a7S{@!F;qd*u)rez4~YH7dkE2M?)7D)FtEjO3vJP^L4O!Qz%64F5|kL^tnrp3o4z?KCJBE2BBc` zP9k4YE~#S|GsClbm?ah)D$8nyGt-|@bok0uX01{iUZ!gd_p4jRdRafmizc7K{(BDK z>3UDOncqjCBfl?8WNtil=I6|~FyK?^fKyFS?%N*TRD_20Lp$=MP5rekp zJz|virf{F4jh!Ru=5uf z_YN1x9vqK?H}y$kKODt!8b5LM-a!)PqN*e;>Z4QO$#`y5_72$2Kx4G4>lL` zt-d0EPUA4|+{H$6ABb&6x(T3muk4M%wYo}$ycSMXU@ zck#p4p)k^09u&@)L;0N-6@Q+k^GZSK5)ZF@@fj=Nw;aU|%{0H&n->&^3_& zADp0K{3hq^V{#z+@n0M>5P*-otmi!24x=p#jM3@ueK>lD6FSlNgLgH4%14zd@M$IT zd`4U%y4WNeobO7R~Adm=Pn<+s=!CGcCroHC|8B9|GvW9==;X6E6U`yO;-c{iWjy!m*M-r8$~^QN(~8UOTx~WdF8vgIzY^qB66w@b@JOKJ3Y<&K-=b zUaLS;l@+wFlyKc=qS2)tKPsEb6wu~*g;@A(0M8az;=x{5_!;SPsOGadvb)~LygILj zw} zl$Zvq>a@woF>1U+d;<1-yABmy$ip%{RruD>>qwoP#_(=b<%x5HN&8%BdZmmbkJaR{ zM`=sN!gOQ0>-k)eTHB3?tt$USZ3q*tEGlvjoW~1qnd1Ws{xV@bzBF>KCTVlcLk8Pq zB?DgStj^m%c-FfY*h6kKIkrJ$rda=n?dP~L%YO%9FDXJc_O+0Gf5rX*bFuRJ>rdkB zN*6UdyVIc$^TAe2Z4}J5WJ%XO(k^`}V0`0Y4uswrj~Zj!EqA+RrdHZJv= zfj*`U#(@^^nK?QlM|tGW%6EDnas8#$I4R{RdXX2*^OeQqPVpt^tVtJlyN|KCnj@2_ zG*0p@#ueXW0`ON~XKboL`1JZBBtPj3R?nBm-zzLg_ANU&J+}z98x+HX-2hX(I?+{M zcQ_(CxzEHWgWt*w98sKyCYeZM4fmlWp(&kY!%+ey^7LrC2ei-A#Gzx&;cRR?_r)my zPnaFc^z5)EvwqHIy6zWZN5?a`JLC~;iBM!qQ8V9{t4)9Tw()n3|HCEWm1ttTJbsiO zPDYpIhz`gd+|S}U)NO(T);n>GS86Y$`mQtSRH<`RPwEj#jW|jqi5A#rvFHMxnm2Y_ z{vX`?`U9T;2XRWrVyy2Mj=M*`LMAnDahv>S+NEazE9Gmj-jAbb(Y7h@pY9Xf_BTh& zq-l{cw{=PKkL4sl|2BRVe}xR5V2wWoXW~TlCAdE1Hp!Yd7i5k0!;IoVd>d zk6)O8E=Qk+qS*xI2{ka_n~zLIA8l1=i7DYuLC{;{A3?@h=;$B zwX`BX%5w;e$$ukeZGWJLf32|At~RV3sKfuw8;9*G7NQm1xuD%O5>DTKj`94v=(^Q9 zZr_mamAxC1(b(OC@Vm+Pa5NS8s|Arz`eq)P>Ny`RjeQ5l`YhP50a-{eufWDLM$?A$ zVo8X?95Cr$3I|$-!}!~gV6plx7~NvI?=?rz>jfUT!)2u7(=K)P)V;?btExv7M$W}L zCY^W;UqL+VwxYw0p>%(hB2M+ILudQmqmK{Y@l}qiutrD~6EVUESDxPj?{^xr*7J?v z&nRUC;90mArWwWQvEJx{p1}bFpSw3EtQ%p5@?Q%)qNs zJQ7#J#Cd`In^V>}?_CQ?Yxsr(l44<=)EZEG^%Xs1TfWN~}!g^_S z;<>mGSB^*%^Kxf6_cyNKt5+t@h?~#_FozQZQ&1USiOu zr3#zky9FsXnvtffG2r0D-+2*3%>}db#IQY2~TE012A}fEL z*{%5)Js+7wF6;*y`=1wG3AR)w;t<{#Y6ed|AL0EH4*!~oNpYw$G|E;XDH9|y^}mKL z#wgLEmZ*!z3}+C(6G<>-83#X}e1(U%)rn9qX80qt*g^B-_%8tx=#w7;M`RXBbQ+cM zr7v#qJ>G$-xH%UPIg8!zB!Gzm1&t&55E}9UUH&Z%t=F!Kj?id)tZofR;Y!xxnF6-o zdYRWu=*1ypmhn-=<59`NW~`(-j*hnK!P4uD**+C%uut)W@L7X!f#PUv0c-HXvf85^3q-i4We>?tyN(u@vPEv4nR`pDU@kH`Y8{JJoXTsWTK0Mj+7#s?;+suU^rO+Sw~uu87{udl&%op z+ecJ1iS)3qk?|45$J zQM{%#63w@p#Q$vQK$p*_!BsEHlqr5^YD3rKi8(fWSnMb~U&;kmGKFM^$nza*af2Lj zKEk!?q~KX9HN4NTI_81#A87Qt1|MdP1Gmq+v0r;DDZiV@CuxW-?EJsz%-DxziV>8)kV9hi&q#)(d-4)jyWVyIP zHMI9VkH)`{!)n-y8)H?4im!D`X5vH0cX9<8{_hMka*{S$^R}EH`R)KWP`DUvR?OyB zX}sXAX5=CcO8Lge)45Qe$6P?~f7r$BBU*65iPX#1Bk$I`mA%Rwlg8PiO9o9`-1mB} zZ>=`zym5a68`ws za(-QJ6OR8?&E2fqUwN>#7JXb?#Vb8H$;3SOM1FxY_-{4Ak|!-^aK>gs+~}6dNHwVP zGtE1Zt?fxJWXvz_(#ut3hNra1in@aB{4FHs6^adp2$(%Gl(_NM=sMBBat#hl_Wd;l|yV!#y7(#m6cFpI$u#B`(cGk@8vuPHXdu zYS-}Awqi8mWfnTtVTThd^+k8)XU68$YP|Ky3jWp2G~|BXiN7Rw>(X=QDcs#G~g39HgV}4n$Asovk|_cjN4YMC6mJ}(5%`ve(RJm{6JDA zik|I==h=i2jZ9taqxzix@OlQnaj6V>lK)7uc#s!a6aHoF?<-Xj*SV%>o4zV>>llWd z5-#ED(X*KWP7klHe}tp6M)D6fM>%Kq@VNh;BXV825N&>Xzw(<^Iv+kviX3vgfL?Ni zXwvIDNbA8?Mow}O{hqainXNDq2el6;l~&SZ-?Kb4d_fSms!X&gggdYVrJI|Q3v$8HM##vh!L4FL| zfo#WAker7_IOFdzZi}ltDjjhZeJLQ7U+T`F8RO${)|8RlZSTWq-;P5#-lvS)Tgo#^ z#mW#YZ;3{KQzBpIUBOJg19NTHINm+wHn;YW0dF~8j=%ix8rN}i3a|Vpjoayz$cOh# z=cN~ZMFIXQ?*s+45Wq7}aAx|f**Uo~S6J>Sc}ytcEV z`mGgTVQ7U52MP8AfShCo})JIyn_) zPlj50GOvsbnO5CvsHfJIt2&m#o#{|PTQYk2?PrFN_%vlEK4A#|>EQ_6Z1@=sJbHw( zCZzLiW-HO_bVYpQ<2-aGtq_&!YLUmkt&sT-72fd2F+6reK1mTfv*sf6R;M1}0%Lb{ zJg^Agf2}}PubzOIiw_wai%Z-e^FFQ=GV%1$ve-2B8D8=(8LqV$faWJ_WVFi;a(C_o zTpxkUCOYxMw%Vcxa)>I5`SJ*xtd78w{ux$}!#Ng(XUPPRQ-jQWDZ%&P=j6$kzr2K-gJX@IASSZyIqK z&8TJR+0+C&V$c!3V9ZGN&XR3>zLwasefJwm)I_FH+C?b0{|rgD60yGJewdrq2kNS# z`}sd5BAww%ZU?>LrS)WCpGOj2-S<6O8&FHE(++E!`bnf+4 z-ZSWOHgOXA5%On=Qlkw4MSwsZl^4i_m zXg(K?Z=c0D?)fovcY-Q^?Q;(K_T(Zah54M}+A_Y!^*#=Mp$sQ93}C4^x6t*h$IH%6 zfaw~OK~B~Rs)qc=cfYK}-#UNb_1{mxu8&XnHO>DR``?ND-p6W0r)VSe>T5#h;X8cX zR}+y3A+qa-%JJjnpA(-1RcL9P26pwb*u@{C2%Qz=>jx>a|4TefdwdO0>{vKqKNyGl z4KM~-)x0nL!T+*rtaR5Ib!YaFz@DA3u#*MFg(l?P z#}GWh(h3zNMUd+^#TyhzwR2Skqoaln$9PeI|6kLLg5g)K)hXGtk9*pUXMxqmN4Cb^3gO}YkobNUp%r0qzX45+C zJt_+uY&gISnRY>Zw~PVZf(2~GO2^83WCv;)wgy)eoyN}wIrEWq648ye5)Krff#W9S zu;R)-axd#K^YBX&9(p$vJo?f_Z`mhOd7y>r*4hi!3KF5gx}2F@9)SLsE?_nes-cIB zqy(eapXf7BcOm=vRN>!j5OQn|p!Ky^@u93Yq$}}18d?}n&%ZcCA};meGaX-$N@+P? zpMqe{pwGm{Vu09-ySinhoa7Z$Q|+8ixW2d(uYFRE|Gf}9$KM+0N@q2}bett&=h{&7 zx}!uH?O1(PTFOrrH$K1p~^n#b7)SNAy!B~M2QN2Jn) zg(Ei$`R7IpLo&^Ah;cvdZBL_TrvD}C(iYV9jVay8_z;_p*Tnir13qlzRQ0fBMpfdy z5mhajl|);)jo8O*rW+1c)9q8W>FbymH0JSlTH+@wyh)Z7dZx_dw8bu7%^x??c3fNd z&qG!C+w+7r4Y3lUFSXH;PY39j`;_Xg=^{t27*hSGfrwcV!!MkXN&Z?<5Pxk5Nnvw0p%eF?;JtAC1i37s)JrOUifU&|$aU(s*NSq4CiJTK&LQ zXx^hIRP|35W(k49yZi{@ad#PMOc_Hr2d^jAnLEggPp14sdu^)P#tLDtzR?eSKb;t( zFOSY^gb6PG{=*Xcftc$7Iz8*U7_Hi6uTus}=-R~g_hTarga|b4S z(=AWpu&iDZIpglZV*_d7dgUM?v|L}fWxtoG*D))>{2j&vz%c34GNsH7`27r6_?7K4Rp*h-@_H)$D-a%;Ce}%ycv! zKGomF+6#0cPUZ>34SfM03j*5bONzBrqnwhVnyp8lrtLAH5Ed zCL;z`@$>U1qq3R@*zK1l*p5hmq{0Gxy|4or6~~boH9W1la1G3?zQEx!Q#QAd zP3jc~Dp&0H)(w_JcpXQy=O3QjU zI&LqFS8axC+2-spYYMZQ)cA|nUZUZBb!do)nbuD^Cs`gB#hmI^Bnk2PNAFc3|WBZ?a zurl@Ltg_5FSS)V=$KI)dU;7)ZGIBMKm&c-p-no#m=PW+uI19ayat5!Lagc494p;VJ z_PHvsuPiLtzu_+UO1l~#b8S2y9;U!rwkfl_TvX}jb`}!6-{67d6R@M$fE{`E5uEC- zg5k;XsO+d2^R;0cZmLxwr8Y15Gv6$TXY*Ygqc##0cXi_y=$GpqthN%$c+GvC+{o2Z+^`0?g-=z>)xUBV-0cU zR0C|-^9e@%kztp-y#oHz&cmq`S&-V0iK|R*nR98vQMAHVh#E9qK-@#EW- zFWp-Bg9}}`4Gu%$&8weyXQ2)h%e#U~eI<_lq6BYhZ}8WU7XBezAa2!{m~sC1&?|!! zT08qEILX|A!fA?Z_W}RP(v=oC=(q!FYgqvwLQ9dsq#B4k=ZEW-Xk+bWUEH)Rn%6kh z!{Hv5eo9y)@~94zZ7VMF?ZHbxT0If=`D}r@u};`#stxXLeSz+6sYAiBs_<#qdDQh_ z4tQ?-g^hd;z`-UPs83%3J~#Z~XUR3(bK4E{a%_2YcRqXkr6Zf{_W~5(Wbj`e4T7k~ zTF!!W!0BN%knBf-B+A{F>*Hmn&oQY?WBQU+` z5*+>)3lrWb@Gr+JIoJ0d;SUu$@S7Ct@W9MzpgCB9%p1w^f!^*Iv_k0B5ss9|WYCk{ zC+W&fd+Em1L&R0Vizx6hWXkem{E}@C_%$9rSS`+(jQ=irc(g@7{GLo)?A=2y26dx< zzL~tg-z;Jt$Y5vfO>ljd4N+lH)m+ex)>uYk z5*!JtYd`Sq7lt!FYf43?XB@p#9Zd~W&FGf?+*n~2 z{u&lW56(`bYLT;P_w-cSm0>|Q-|OHrPi*F&PI*W2zT}c$7Z1{9&v(+x-R{u*asuAG zX)uX#zk)Akt;7?w@8f`#cI=%uI&8hS8EaK%OonRM(xorUkjJi(FxveTHVd}G{gVrD zPH=* z&qhYj@-9c}xPc`aHg!aIlsUO-7C;xpMbYNh^7xr}2fnsinN=?+h29})c(9)t)}7@G z)>E{I-JoH#$!7!8U%=5n;d=DLd%JJ5ct4HjqG@ZZKw`1!8s zQ2am@@_j{*f^H%|r+7D>dOerPGIHa$jGHZTSkB==G@6%MWK5geKO&3%FzPl$mwE*> zGR8h4zx1O4F}7|a|2>nXrWPFi8FPak%lb@j*ZrV1{rl)LZ3ntu*^?Hv9m580w{WWA zVHnSR!9^gCUkz&`QCIW0jAbGNCt@wE>9|8)&KUxU7b=MEwm>pVH54rj2_e~bJl`)m z!0KfNQ_X`SMNV6h`bpiVYrWLLTxLJ7f5wO|!NF86p@epo&-f)zTsf?vxu05ig=20r^JB8LxIYw%t4ZGXtOxOQUnJCgQn87g{qroz-mZq#Ac zbnMeR4?3#{(-D^@(GbINRBrbdx`ZvH!)8Z_wjE8p#P}Vrqfp4Jp9rJzk;&9C>oB#R z_<;Od6+wTT522$UtOxsjqv+5do-~S0!@`YH+-u&3mz)=Ql#85+*Jd|*B1M6!cn%l; z-*wpc^KEeJZHM|^Ett|T$49Hg^NKn_&PHpl!1#4Bu)=8uE_m1gzl%)SyFp59oBB}< zIq9%&@e>Fu+{oRVZ%rp&If&mr8i$KNCt~UH)kME)2EMu2O59sV6I&@&y1rkE=r&t` zn&Bt-o6?HAlA?&m*Gzc)`4k9$hO(;5Z!^rp{Ztq^oT#_NVmB%)51G7q&`%c%w6 z{ab_1ia*F(7$%Z}@s2nkF&%8j$*{!%6WJG2z1h$s^>F#^e{BD%eSpn0z$GLFUTi&u zt9B~VRJ)1vx2XhQb5*2D=KaukBagK>`ju7Ws@da9&cQVKM=-jNWwn+_!`uQn7)X4< zkNv$LKWj*$)Bc-JHCx0bpHqgv`0`EF*fcyR$_`IkF8yJnJiPzfbXuW^LXWeY7`0Wu{ zop1qs&wj-kzDq%8VL6QH@_;Bb4{Ai7OQw4i4mjxn;T1>0u<0B)hz^8Rmd0G%iCnyD zkTvfy(-)Sdj{~_?Q{ZD)8=QNh!M@kL4Hah{V6gRRXwJ$8!CD1|8vVuI%g#f%hJe*x zW%12kO{ClJ8n%@aGp94&;z3tU;KR+GV3Ax4{nEws^;aX}?W@GEv3$TNWshJMd9C7w z^MAQ-Q#$aX;;%$y=QLc@V2;!)=9A06N3o||r?MCPVjwBXn68sOKurv1(GTK`Z`05z+W_Y!KyjSJUW4GHmx$8`e72oQ0qalFledtd`j^6&8 z1#A7-s3UXOeyM2o*-3e^b#55*?94G1&D}}1`)yoLiK6T zIQ9W1qEOhDcnTJ4tOw5*{xB>^oxS6)#;VUAF64Mi3pbaD4eYmw{%}8xZ|yrlCTz3B zo932P_O_fs|8iT2)8SWqe#kA}f6{cJ(8FGMC2J(au5Ti}!NV#CrmyCA<*IOfFAezD z3r>=co<*d~UHoRqeFNN$p8^qqP(?-@!t*T-$V zx=Nea-}_A$tWO}tnyJE&@U24Vfy;u1^k#ZVSO zO8hly$b(hq(cu-PbmEawRFWxr>erbFOAIZAnFE=^qD||G(c=$%=<`u@`_F!6;m4<3 z%7%Mfdc;riF+YU3u6jypa|Th_AwUoB&Y>?_dMVPmP1NL6sNbMnf>mII(Br&Dc=mG> zr@z90_Qm?`&)Y^+t$2S+h?Q5Z(yC4) z_n!@?>zcxFPD7-y=c%Spvfv55cH5X-vdV?GBDdWwCYwLKW~NMjk^nx^Z@(RVkL z$x{9e{ndSu-h5Ur&O4;Y=3$yJbe}i5Hq=-+Y-=K%J2gZw`(=o#TvZ_;-wc9(ou(}@ zO;jc471@5x6t9lE%Xz#rrz5Xu(W!B(=^A5I`ut)d@;TxF+lIe}AMFwF(Ct0e6*%(9 zFpqD%rHMYeYLoBoC8T3hJo&NUD>Ysq&F!h4L@o`Rh{M&_W3%MPD6ysm4PF#Utvz@$ zV%0G+xO*>|obrg7uug_%MAZ=E>^NHZq=z0~`-CW0Ugt8qmY{$-RpH#aS7g%Enbbc$ zlcvn}qd%?Vsom?<{L|&p$aeV;F<)v+EWLcF+=5~%OkRoKw%bx^`Fq5f%|x%yoOQ zx#KX=FV|Eym47;88n@HhpO1Vfdd)ObNcEHFYp$$jKZf_|#NK@)Y% ziEn4c!XmmjUrV@Ym`Dd{?WMti@5rKUej)?ak^N?~i(S}W&i1Azv9D%0vnffRpm0q# z-huiUj}a5eXzvWv8aod=E{Udf?gs2AbtBf`x)yuol{<9oxB!f95e)75LT0zFkmJ#6bFHQ`lky~b>;sX2KD>##`+OU@=O1EtP`xcYc1 zlpf#5Zn+V~PQ9>&4I7)x&e$Br%J93`a06+ua=U@&NBZ+A(^6ns`vS--Jp@ViA!5c8 zVC_Oh*re9atjY+058C_iZEJ6K5MZ`p7{w`VgHD=b zb{zC8?I3hh7d%>@3ljp;L_~z>25E`p@>`W4w0Q;uE-r(>5Fe;_bB4)AZ=vt1KI`wu zK;Jcd9$nL*TLOQ73550HJNKP zhQ1q<3c4xp!9yVhhM(IH_`+7OsEmW4x+PGN_yEs&6HF~XII{Q7`m<9%uV!y>r}1n=hLLVQ{=>^Worvwdsu+NN|? z^LHFub!Z1`I5dzf{`{R-Z#cf}ausH#DS`}I8`eE4V=QY%t@Q(W~IwxHUu9CkV zKmzV8pr-?B@TiLE_(1YG{Ql`H^g_mfti3oC*2QyZ(1Br$yHphR_s_(IVvjlh*CwiT zbp|mV?L@ysc|p-&TgZGZ!4b!Dm~G-VY!Fw8sz2Q!fyYX4z^xiI^iMI)QclF-_mlCk z`muC<&o87^rvq7@uW8G{6EyXk9&Hl$@^RAENPV@)Y0}0p%0M34w?;v$-CnZe;};a* zH<&KD^?(jc7(>t2o+D=O9}<)C4@gMT5Ono#J-*QGg}3wK%rIu1km|TpSh-6}^sf#B z#Y6tES+W2$j+b!;BQl7NWfd9kI)tpYJ&Zk`)Z*Szk=W6ov@+zL0vUGbJz9Ue173noz0QKQ5zXtKOuvk^SDzI{8>_|boVNmWqy&w$Jvs{A5LRW z3v)>Bv4t1k&2XiHJO1%Jh~)KUadj?t@!_;1BFD*tZm*T5g>5Y)-2V)nra(C#e39!k z*hdy#%;du#r^ESkRyZsAEIyJy30gCfNsEsaKC1Nz2hG`vZ7wwNGv}nCV_(jTZlxD= z$kpR?L!>W$^{9?te7cP{UQvgY#J%zOD}O=OKN)9GQ_{KA3~L{oO&^5Gla<~{WcNm8 z_>%Gz`??H*6N~iF+TLbn=jwi}Ij)erY*M2I;kIOH!3U_jbQh*x5Sc13UEr;=C$s8y zDITa5?|X|+kbld2`6KdsxRQ*c{0&KpWcF=6T6!lI-+F$XT)K4`e_=(E_)r^I7qpb# z(2gUW^JUOYw^w}En^U+KMSzy637-_}M~rJ^Eb~#Z%X&0#)w$1J2B5RLN?rr6do+ zp03d@pa`)su6y9+i^@aUqEG)+2 z1wpDX-EA-L{5=B&dR)N1Z@VNBzB=*^e%NG13P!Ain!hrzOyLDN7qW|b**24G19P}U+wg&HCZJVU429}1VP@tb z7?JseD@aixh1uKiw0-8pF2V@(<}x5(k%lc)L+HV)xun(AnZGZt^?R4rlJRBhiN}d3 z($9Ai=Y&(F=VTduIdMOov(%TyB{|S}R~_iATeFGQ%K*YPJS9e58_6~|1-$R32R2YP ztGwNpivPHu;MK(5W5d1C{5AD#$=?EbuHDQLGe@@Ipa<1VxJewjUfW3C#W~QyaT$!z zxv4U9)>Sn9&U&=b4x@h8ue{N2Z@TG25&4-fGMY7J^In6(2xHvO741pDQ4^P8DUTrj zy}P0iTBRZ|gFC5n&Z$GD^b{tv1B6c2WGEAt0G~ZUo&BV2uCl z_vP+uB-7%QIC`VKnqFU(M|8%?lV1bYm;r425zfJ?!;X1rOhCLi!hqIs3v5{6>2l{GW4~B;I;7 zpB+>~ba#}H)k-o{zFm))7b}sZ#wq;W;wiZ4-b&I{k;8?^8jDVYMl!}@JdHixLmoWL zLsz%0LqA0pTJ_jA%yVb(uMh)~1#{dB@V14f{&e#P{{g<`sMq%y7R z@}(0B1Ocy4h>@@c%Ck$Jk&1M!X`H$XsYeDFUvG_PXK<)?RVvs+Bd7JTEz>hr` z{oRL?+>VpQ%o#djdI@b+TSRp}%F#Q3XxG_cG#Xj*>yusay0#)-=ow42F3!TK2M>vF zv=7*8Ne!O4Z4}wkt4+@-7m%SFEa?3=8_8A~mW&Wn1Z!Vbk!Wu_8oe}{j)SXck>U|@ zGkT;U&70R;s#d;bh^XeF4;zo)-I!kdXMSoKi;HQLz(Q_UC%#|Sm25ld;VMQ z4dk&{{JRj}OM1`$%>2qE z?pERlx}R4LJI3)Ir|bEV)5RI#(G+st;|lo|XhUa2Ore7wtI-JSy)>{{hqvg;;&+Tt zBQaMk`LWTaV`I9?D64Yef;mlTj0 zzhq(Nh0CNYXfZlJRe|UfJVUYv^#rwWEur`FKdN<3j{drkPHtpxCOd7#>bM;wPQ%P{A9A+_(bK`E!uFf z@g&%kG{faJ`Sj12bmEPHzCQhi)4q6t3oZS`IB2GmxpDgme?EzKlPFVZ`%PG9L4D=1 zkSsi0GaEd^f}rDD6Zb#L-aHzs?~VU9l$p#j6&W&=IcHyc-%=!%AxiV8G|)(;G@3Fb zQ!f*ZX?C zG}7)YQ<#J2l~&VR|EZBBSBnJ+7Aa)#bP@>bGr}Git88Ibh{BsxU00a;5tY4-iMwN zcRascM38oGAD1L(!*bjpn!i)B&hFGluEF&Hchv2>;Gz2fl@4p6RY4Ury=4PAS);_e zv{s&5Alu45w3-91Iyv-d{a1E!o~*$9n+7fVJDXj$JDr=kwp}nm$jC~VBZ326KMMXC z)xeQ3TX^fOg_Un-k;u)9BF(8ZXsP^_wr%wj9v5w|IMx_0zt_ z_WxW*mv=43DTxh8a3GuAQvQWCPhvUc{juEQ`x-cMq#IkrEh3jc*K;$k8q)D9LPxt* zAor-b4qg9ZMIs*G#zk{e=sMFey2n$E+c3ibbmpsLm7TJOK#$HhsK$LS<1zDBn(N=(LLV&NMS7pF zoxTd4RFA~DMu!sG{OSvi`!iecv~?re*E53# zx*R31MiRIHVNLw6iVIhElVugo%|&Mag;VP_Ih=~}c`RgMvRf|ZfL**H8cF<#Z&)0L zpHu7Uq_5WY?SaqeV#Sk4IaG%zcxefiXP4nL5kIn~*B|nU61w1w(a znsF<0YjJY@NmzJCmcw&)(ah@ubeJ#2i5vG&+5N|6yw>vLPJNH1M~X~1)sh|fqg*<- z;z|@Z-6{j!ZI2fe)Yh_3H}}8?*`w59a{_h3KkR=xm{AYKKH4;+nN@w9Oc#0*cAt6? zV*Szug3XC+L=6kW%U;8lPm*AoGYbFBDy6mUQM9uAnjoWWE_=MtOptP zlXCA1=3@2QDd(Ed0*@FP=@*CEX#ty4CQ0nayveaA+v(XCgLFYuKIfn=Qr8*o3!hS# zg1~$sl1OjDL+hFme~J$sr&P`5>`-mx%4g4Kd+Z|I=gE)LYJDeFb^ zN)z_@)KJ0g7U6risxe(>?gqyy^tg`eLa;Ck{Y2=Y zbGuZ6xvxId-$0zy9_*w$235I!HkP~lr3()ZYh(Fiw`f>l3>}#_j3+#f#Bv?8`K!Z6 z?EkIY%d8GP4tK7L^Jm7j(GMBEB)?UTl%*Td;)y?D>E5OA_S<~c=Q5xE{Z~#KN_TNp z?T@kRi|06N?F>#u=qt)xFLX=xnsRgNwh)D$6}au_QmRt?7V=Ubuz_#9kVwmT8nyc_ zTk-xPk|}u1UQjqmwU;T-7kMAp#m@1xJ<^Oe;^}l<@EDA|S&pyGiNk2zQQYz|O^|om z4{Jdio8CG|Q~iW4#oqCC^0%LI|B9bt?%ZnJVyS}X-T21d&#&|YyNVVk zIMU#tG`{}JJ%a1Op(wAlEqb0k?bcanc6^EMb2yeC-%{10Tbfk4Ee{Nq1BotuxLPm zitLNv0!3uFB8?w3F97V%#0NYi}jp@905mg>|k^!W^oq%Ya~Vq>UhSLKxR# zC!DQXUqYQ9`*9s7j&T)aBKI*Iw9a#QIH17jh~$(9hSk1r|bQW$`^3 zPWxt{;B|rsJG3|)9!~m7a|XgVnTNjI+dOTuG$w&fPR(SCFI3fPyu89}tQ}x>EZ5~q zR*cb}p5>hOhX+VY+m^Ok@af|zVfgp{)p+rmZhXf>Ozn$#fKs&CaegjuGy@GnLjOWH}pTXG-AK(fV zQ#toF(p<=%Lb_n_Q{rr}QD2_XjyNm1)p7H<6&c*h3ajBIb3mc8{(aII%OJ*{KOlQZ_@wMjibU730>tc}LxU z#|x|jG+?D%37oJQ#k$OXTyV~lykaEUv6M*?2|wyb0SoZ?DYmK)BI5}<@rus603 zKTtY{-R;Nm9!Fd7lv4Y-4?D8BB9CEC;$Ap8jU!0G$*H7Ic>?D@c7l_=Jc8ee^wL6) zsoW=vpEU577f~*5#b@md=4d83=`coH3AyorOS$catGOvBe;*5KAtxD=V_IYt3!#H z4e0>WX@;!XBjL=L+$Ywv@;9v=qxa_AQw1VXeE_Y_LnPEqG zBr2TuXssx3OQ;6#o0lA~Ll8`!xCq&_c@d=5F_TPGc}~K}NphM=$|dG-Ek|WZNW&OA zw{sT!o76;Or{2J^!u$zS)eg}#hU+?hm1s{3$C^Th#Cg|N{INO>uSq&h((-PQF4-fb zb;}LzNLdQr`DZZ~6V=6;NThIR%3He9BbKCVE+g48WiZSx{Z2M_exf|5zf=Wo)tX?%L)oOlH;g=;xPsjMYtFlQ?IYRr+>j?< z5zfjz4uT8UC-A?vCUT>b|mFcQmux{|Ng6gOAA3yi>H(Ko-x-lf(_x zBVhI579RPdh5ueUPFzj+Y{eyGv?cWolO3Z)dYA1WHY*FrBiAu(@_PZ<`9>9Z!BuR= z$E$eU?|tkC%hNDD&ljyRm4*AV*0^i57JP5@;8Bg_pd)pA2kI#>do>aJAno8nt8tJ(?Iw7mN%s zTPv62AUAn3`Fb<%T5yGnoY2M%6kO)^tL5TdXY=qZek0zr+>+OLKZ57fr^(y&uVF^} zT3!4jKobudIgqM_4~f@`LejMH8HkVUCp{5U$Rx!ml5-C5wXJ1j0aTKML7%YF`w~2I zZ5B^|jRx;imLl2yD@WjaV?S2Uy2n0!^M$+CoC4!MI@K*-eU_M6oh6Uc(n*&CA9wO5 z5q4I+pxZo&ckh;f1ouqk{rk6tJR6%vR{E-8krPu$sdP5hwTh*|4noGUY!s_@^&f-{ zT9ElI6N%53V|X-f0k4SZB+A}nWaGVbTvQanroCaQMP>w9WIm2}`D!fr`K%F3$~s}i z`i)%H<+;2R-C=THznDzIro8fJN<5<@x;%V1jQw==J}f)|ID$JXc%S+mx9o|-+eDl2 zf^Wt`CS3-(?V?0Z`>(_5A0Dtjs)YT(bq`41ZDU^Gx(IxBb3S=+G=eM-mcrqWT-cm+ z4T1mm$@ol{4OwD-9~-nvQaj6NHq=(YuFYIaCckA!!_TwW{K_w)H~NQss*>X6-9O1$ z#zoQmLkXOeY!F)&QX`OwG$Bqiv(VH$HC)jq4^2lp@in)Tt&Ox+ISI&x$Y_ocfKAA1%`ty;72gZON6Up~kg z$@B?st-s4Q$rrJsmw(d9E_XqlH6S=Vas2UE~@(3la1T;h}Ozw@+-drm1+^;#=8$w2c;QY?%*oYSJsCQ$z8<;`5(#N zyKCr#JJqzX=__cA9U$?dm00V5v(PD-P4}q3;bOX^xS4CNl5%#h)EO-LY62MCQ6_)r81BAa1jp-&XxiW$V!x;mH#^kOrU*nvmBy13N9Ad~ zi)7u=chyvB=o}~YM~oc!Y){x}$$02+Cj{=w;w;Mla^HMYg)WbLeCnC4z+yo!TzU~j zF0CI=)(3mwW2-NNYScJ#lW8Vi5hK)0?+aD8iKQ2pZs6{Zs8G#!Wi(uF3wC!_z}{64 zkmj*;_Q*$Zvb5fXyI?eh)cZ(sbtfispFYLn4~Jq%+xqSJv#`Z*s*w z$DYuy%cZz%S4-0Q-)17)&5%2a19&Ydh3d09I$>CV{f&Vb;MybYK9+nDDQbB8nwBuJr639F~=%yr2R;MsGI4g^4rV%!0%cadR=jr zMm}0ZLRMWTQfptb^Xr_k-!vhYq_~-TY9`Dpi&avag=_I{Z3#|y7#vIWkdrW51H#K{NxtRBOaNZVTb}F2mqjZBh_f5w6C!%q^=|Ox+aTj?t zvjt>gHOU#-d&JMth}ZMEmn2xLW6=(A?)`CwJG!@#zC68vdu@?VUFAQ~mopr>#;Eb! zq~b8nMCfy2Z609_QjBHmv`K-`Ep~FChX$9uV%^VK5q3o*o&I_S{cV$slP5f+zi;=# z_YYlIZz`YNZCzV^>%~NTKykM~NTQ_|6>4aI)I@9(fv|Fe4&=G#v0n9YZ1dfTG#8uF zz57AX;r4>5Qq5#j%0|(O#m4NHn{g0wI0G}XYPkD;0{i#qeww&+51R(f56}19vJ+2 z&VJJtfk2?|s;F1bU!gfIahPHeK6u5+^Ma#4I&2$HRrb!J+H>;R!Urkv(flO;MBzdDYi=~;Zk!KH`wk&n$CA8sxsM&vR^vIx zcCrJVCUo_FRaWKeBWm;d7S7z6j%LNJqUS3S^FCIderSo>m@32*$Hi{ z4#KQEwV+cF0b0Gi)ZwHiXX+|O*Iyi9>pq)MYrzio?b$?j@kx%FmRZ8CfcJtq9^tHQ zQxXgnE++Rl5u9GLluZoI<=e+h#Yakz zi5J*_jz2IqXDTfhbzvAp<_nb5+mDIvH>w;lVmKM5PeHZ24 zutIxu3b5tiHZtd_7>upaMEN@}0qXz41U4(Py_S^L@7X~Wc6s7cYrEN(XY$~}<0aU1 zt2{oZbOzF^H`70do7ltS&d{eTeW;p-FLoV_pj*8W^JldLJ<+$1st-?P_pTRmvYsWt z=70+L=F)`x^sllKw2oy@WV1%z(RhcRknwh8GJX&^jCFV3p@~k7H1E4Aw|8Lx9WRkg zE7c~^=a1WP@ZD5={^t~&-Z;e6t&&Erc<~l1XM&Rc@mOrGc%{kcg;wx8nWtp3D}i{;hv zdS=(n%@{|oy*3kwy-Q**U7jy&v&4`!jnA25-&}G3@JjZ#&~Y&CyDMjZIRVNb~wv*?@lLQBruf%54?Z{I8vG z*>$H3@mFmRND()td0TIg(Cr#LMwm@MQ2*@i2#;_#ozgE+#cgS|VU z2K%YSay43?kUF{!6@Gv4Wu?zJM9i7gBs^yC8)N*n{T>x9o=oz6ZsJ%fNjA+gAf6-3 zY0G{M`1o80Z{8~OUFnE%FRU)HqKcnbKcfs5cb4F+_TQL(TSt1c2g$5|S>%&K3%TA> zPn6pBvwa#Z#t&fmF`CJh<0I!3d3`&GSo;q$|I z{HD!RK{Z_9l)s&uJFAfV{@6l{X66$Ig(0eQI+WfrY7`94D8*mr8DdN2CahsM7fbcK z(2u?TWajM>vdD6nT;27UB)9*=bmtsywfSUj1>;G&cgyiaroALTE8J-MOELOI<0aiZ zJQ2{UYl5j;jY*tDKDWlIo^IJwOFx>=VwY}f2X=iJxezBwvo?Ik*ZxZ-n==!*lXo|8 z6Ly~AwtdPbv+nH@zK8E3^KVLVC1;Ju#n;hf$AU7lc%Xp1i{3=aYVR<&O$|_`sUA*C z>7ca*vh-+C6wvx1T(fm1K9sJ8@5hSq#(wT5+PmXOp+zf|y0?m||8&P+B@ zeCWEeg~%>cCk9{T$)2nnBywXmS7~aEnK*`|s0lq4-o->xu#=p6Uq#Ao;>h*Ti{!Or zBVO<$l5RZy3(UO2@V$q}!9pRMj^8ZGJ0s>pa;i*mY5NM&wj`7sYQ4w4IN?Uj7RKQD zZz^%Qek|D^K908}aR%>r^M6?R#TKmm`349!=nAt@I|+IqWY-jBlXGciWYETqY|MN~ z`Koid=?5c8hW<{P+?+_ZmhKgDB?*x!+D%N1qRA_LS)3i=KpfuQ!!h61;R}&l$k%`W z;ggcjg-+DF^uSUj)`fF{;7`lpL$fpvGJg#h`aDH{J3^0jw&IFiAO|eDt+jM=j{kar{a)3H7sZ2jpN!3$$SfOQaFDx?)}|Oy$Z~@DeEP<$AY6QyHAVC zHCNU0L(b6`QRBE7w?}Di+yR=n;UWEzDNdOU_fYsb8GP-Xk)V4GM|$si<549eCA%wGaNl-mMI~ae5l0^sR;WvnSB&9&_>D-xf$#NrRexQK6Pk zcC(*aB-n}Gp2ErCMCPjfV#>^4K{q{1L66c5*qr!}?5SW`0j?55`;$Cze=7yYK}A+( z(NeIyDarcY`z%;GIR*mXT>_bezpxzZk&^fN!upIk_h^zmt^P%}GCpdYZ97Mzk5 zvMeTm%bgrNe^ipZJ^TdE^^g-3iOR7Va}Q#x2W`xY<^p(Sz7S8@RK~u{&taX1lmyG_ z4za6rwdvna5^PysEG>1Zh7Vs~2{qQLRPwGCUH9@A%6gu}y8Zl4O`pxCcizg=UOO== zp&^C8Sx$gc`^u>4E@{XO5YA%xyr;b%Zh-VSF)ZUfkrWACa(5fkaT~3N8q))?MPWKc zXN04;|NVsbi>3Gxc9*cM0uN7SHo@5`PFSbWTrhEuoS-aIgI(Wg$o{!ApB4u<(I3-= zXN7+&scyS2*Q>Oi=Iu>k_lO0urV)$T#s~Ycs!uOc8M)6_-B|$JbfdB5hFiFNB!)^0 zJq7XaHi7D#5Ae7<4qa$^0aJ<@DDudH_4kXh!IcXTe6t)WJncn|nsu-ORih8rAH(sF zr&-gR{Y-@u4@GRWWxEv21<^O8Vey5(topP#D43_hhB{h7hFdK(joV`{*ZvS*4Ggk> zht9F4e{X>9>2>t6OC@^qWELBWOR$-uJu$97NFPvonfq;e`B|Uf+`kB%G}{_#M3-Qyo`Qd`Gr^4>caY-qNW7BN;!!C_ob)IOTOPWM zn=4KU?z@W9hd7T#Gi_k4Hv!ca6#@HJnRhJa37OInOD@qOd$RE z7&q4TobCAS&5lIrQjLB|YB|Uk&dL|jV>{Y`o{hzHJ!c`T@2zeC2;qXDlh$p22Z`SpIlN?q4P}Su%ZMj5bL~- zb;X|G)EF!5;kKNYdi)qqUEYL8TNR0>#Z#_5DVkG#Voi2FACGkx)Z>D7f83s-MH!WB zyx~YZZ`ytb-pLm;c!967aC24@wjGvaopUA-rN4`aQ=bFO&UudugL|<=l^)Sf+=l1OLnz!~L8ksg6n|mS^8$Ws932`{5GI9or_z?9=7vsPNe{j4e9{jpx7J zbDB6;Z@_PFq+tEDJTm6R;XUKNfzR7lSjpo&W4I+#$UHkr(iHB(A+i$(E}lo;03l~q zC6WcE(!8Q^e~H&l%ySSACUPlq9Ix7z{qE_rY+M({~PM zZji&*T{KCOzy%zer?UJ@vZQPAQ|yopP{HTei6xs^zRN^9>BCMER2YeSr#!(=LTq_P zzSDRUWp?vkPP5`ZEYhLA3X|}#uOHEt%qE4K9uh@i|7Je-k=knb(h6bS@JELx_)A}b zxhvk1m^Y1h`3d2SWNZw{|2dW1Amho)N*>C|zQk{ReF1m3Z|7`2Z03eTe&DV9g)G)D z?qq4R7jX;mB4uAgakaxPm{+Nf#)<$NIbEdlT^xzq)jsTddjg)B62@!0CdqScFyI{< z8YYInQi=8BG*WllhO9JRMQ(roh~GwV*zL9pz9wYOn=P&*ZxjCz3C9v*knn(Be{M?* zAN1pkOGF^q?|W@(TbsS~<``@s(*vI~p4!CXgV3#kxhJ8su16BXQJ@)0;X=K zutNAv%vpqSkDo?xUH#vQ>9kXX`Ik<_zr4lk-Irtc(zAH+rllC1#^I(9)${;Uz;1Fs z3!`VYV!0hV$x+P-jYt103-eg))BUnrhnMatF_r5ox3>%DC;0w^*eoI~^_67xv_)k2Ln6t@o=BD%8xYHj z_tD8#Ph1u2L^D+1ATKvbsUuIbQ$+McMgI5oauXjXA|fRv^8fm1i!dU-K^wRF2XC_U z^WE!f33fA>8H}xswUwQfwf$TtTalIj_n-gg5B+Z)FPd-E{_p4iegFT~Vfp{H4tpCL zySYy0|2OM+GrtV1f|}t`HXm$6f!U?+@JSaQG&bD)fHm!A9E@$owM# zHge*yM@IrWww{C8C55o7@(TRZ(u9cpws1rw7Q%o3K_SJHA#U$1Q1{mdH7{*oQ?21t z_Yj&GD+BV?6TqX>1J)ZYhEioqh^uS__n|77@URtj{g;pa?rue|^lmYyntZUk>eC=0lh1dbn7h1~<-~ z0#_yxG!FcNJ6fjLLtG1cI9Y&~&sG>e)fM+k73#w16bR{gIyhKf%_8$?FE@I`n?rOWr_ek6+mZIIV{-u4GEHZ(ZnZh=yiiW ztWcN==IW|2UnCnuKPE#zJ_nL~Yv_#hfx_$=aK*$B6dN$q%m+}~tp-!gF=YBhfyxvm zP>NOnZJ9+-vrq|4oIW85yH|+D>;+lFAjpi}31CnHMm=KqbAlXxvT`}xNzH`GeSPp! zX&P|83gDrt0^%DjVPEY^h(F{2t12eL7~TMS>n$N9ehf*MO@+;6Zcw-}6l!k#AGn7I0^hBY1ngj0VLm=3i?ig5IN!wgkO}!~ zS-9bP4(5q%f%UWYL&D!EI65s6CS+ECZErF>F)jf6qE3)nQVVA!|3hD+C82($0n`fa zAt}8EB$av)j`!>tVv`~4p%jqd~<@nDDxjDtyWCxB?lfzx|kFqNJP2YzpY zKC2)IOT0x#@P#D@d_U(TDJur{)mNj3GSfKsR!FWt%Zr*JoqXJQ2x^k zvg?%LOREz2H5tN!4r|cgt`2fl^1#}(!sE$#Ac!ag%hNO9aexDOHEx4(LYKzS)E3Ax zsDWQF8AhuWK{R+g{C2Jf%hda@BD4vdVxEKa%jfXaxE1yaXVVWnSHrEVweYso<@j5- zFJ3+Big)ix2YdG>u<>hv(e;s#lAQ)wmqS4@SOc0hjN!#(Ef^Zo1K6kujpHYQ=!Unb z$XOh0=O{zoOJ;g^k8(B2t+T^ zf@gF zb=$F*5y86`8eyO4=~#Ts79UBRhYkA}{POc*;eMF`uP>Lt$EoX~biom@u}Ofp7J)b+ z%LdOA*@;j4r-1d&AgBoO0k75hpt|8AnDNtu{i<1*mui7$|7VF)hlDe`j~#JP(GsjC z(F&`w?txU#e-JQL9v`$e!8?PN;68^Wu+@o!?iI=K(R~E9myH8$uh;1OeibOnFab*& zJJ4x708cdcgTk{YxWCyDUMc)UxfKY)S^}UnXEi8X%!d0xBQUx}2H!HX#fK9v!{oRw zusVDT)b0jCa#jlTD5pSr5d(*FF&G-mhIi~`9Cqv;zGf@*4hJ~lkb553a9${Om0kp% zM^1sGVme%#ED4op3f%mv4vo@^u)O3m8j&`DA205}%_;Yx)~yQo!u$JUNeCKBEQorP4~GJ*!Ng4rrZVF}dVL1eOKyeXwh*{stP7hz>4FV^96S*^6?blp1!tEi zh}F}_7Y7XRH_>@GH(Un3oz;RPj*_si-waPa`3;z@x|q{4!uPkjV>ZzhAI|i~7C*f( zli`gq{}AxXQXpyEN?3Gc4Gvl8jiU#rV_%OuaId@+!mKH5URw<8t32@BSp-_AAEJ=# zJ{0WpA2QND361&*Fth&{TpuZd>9_eXx8f13JrD#Xd{4Nk$b&mZ6NFixc^DBxd~|y` z6n6&0ZjS)));WklBNAU3h{t$I1<2<=fPMd-!ts3orA@jJ+NJ~jL5UFiS?J;_$b~c& zXFSu^5w}^cz__~;j9V1(+4~aMp?L>9Opk=OA@MM=B?k`cJOqcLGq6fu7lMUA7t&(` zq8~oN8GIAAEPoDT^QU3{4g+CbQVZv*Mhmm`Y49*02^KHg1P|ti!FS<|{U?u&kSuNk zZJth0dHWcIb!38?-8Ja_Dv5nJjKetwqS(`?6NX-o3ir!;NIw4=Tn|maU+Y!y^rBi6 z;&2+h_RdAowojRHPCaOQiUsPF;4=q*0;BY07IS!G3FB+i&3x8h#dOr4L^F#gqB_wd zX!|V&wfbnFtS5P=3HTtF}%AjY!T)!yGBq`?Gn|j}|pOGDHFR1&2TIy~va_{z| zRe!^nXr#_ub6&=1m?$u266N;ysygjQioT)!)^#XR_`KzwxF9+mA^ZJyD9vjQI%T&F zsbxo@7dPjkicC=yJ;@e@R^DKm3SF63YbG)&@f(oR-(b|gek(c>&ocZO3z&l4PR!z~ zQpnZXjk$NYh?yb(k2#sX5%nPp1?nRoHN%#DZl?4J}D*w1S2MrTra z2rYWd$m<_xtc7QpMr~K-pNBaz3lL|fu9IdW=RHO%_Cz9i(HF>UejHQi&MX{ z4|yGVgsv6*L>~`TGvgBx(;Sh>BzZbAvUyg_x`efi+SgFDC!&zC?vF)IK50zJNEfr` zDPhbs-lLrg+VH2w2^G99L6)L6Xj{W-tjEVuMEib}-3Vx0$USZA@NQJ7c*mj5)sY0u!1Z!0c_(W9(XFA^L0#s{T6< zS)5fxQr>eIp(~GBQgn^Uc3}y_N7(z%=YWy%=?kqjJA-Lh8NF9>!w^l z8?;#FQdd1QdqN%a_vl|{%cV3X@?$3R`%Nk{>5Cy#-fh7|T+2d_tRm2_EHlIo*D~7E zvyofG4AfyP$^7?OoY9|W##CKPK#88hcNLoqq_a{IX6`tN9DNedzn%_cv_}q#Wc<;` zxrflm7Y%gx(G2wMbpmS1zsG2puRtnGZX@y`1??YcK~K|+konSTwCQ3S6TK(}IY!-O z#MDoq!T?9qRZ(yMQhXnyGck`j)S}CTIk+=x79AQdiDYhb$-7~bCXjM471%(fs;=5?JvBbhA%7Xs7J zM4uDLH8+8o9o4`DyO%H(ac7w}(c{d^07K^e9Rt)9p~`5EZevKA0rM;IioHy^8WXnH zmsxM(!|?RvnRknRFw>rVwEy_cg}LZ0inJ#@X1dP(#|&qLAsNdEi~( zrZbneKeWHoZ_HSol1JkX#4=+`&oRSY*N||~1u=`9km2SYWZT5 zqeGvW{pF4f@!Z1%_((B@t&ON5x|TWOwFnuxK4kb$S*GeWuGfTZVU5~y7gb=xKam*eG6)u_7dH*ngL0$7DZe)MdBO%kZ$l% zC=kw>mt4333V#yNk{S1r%$M;{*{uQ+ZXZ#J@EQABBMdoi4?}4#v8dy}GpOgLB`Vn3 zf=-=rMxUd@(amE;sQI)LNUg|+waaRtWo;2sY&wZ5th-S4d_Fq#+yUJ+(t>gWgz5}e zql@>Zqor?VBGYe9NO@=mN)_1+E41U`mt-{<=xjyC?Y$^FQv!-ywxZ=;TakC*G9-52 zj*&6?eci!3XHvh9k&Z z{y*eVbQ(EsQ--NOvY~oHB#q~g6`Bo` z?W9p(kRlr3-$r%9-zhKChK?Z%c!Rc~4pVb@@p&s8(u+d>e3FrZ_6F1)KNp2Q{>pr- z)I)8zelusPMi{-J38-LJ8&aDk3#Q5HAb4~bEu3$R@U8%KaLXwsdL)j~*e!}0)=mK4 zW-n-(k_ew18ql$o_t9#9CD1dxf<8a*LQRV$;7w)&B9pG5w&?9BwB-`a|Mv>K3ytvA zM;0(}tq`_usDxe@p-sBHoS{O0Xzb^!XpdbSsi|iGYs%OZ=iY&FYr|ppyN*jsNYu&?u;a(f69$WORp6DkgYK5Yk zQ_&NtYiRh(2juE6Tz8}woH5M>v50QSdiEZ|M9!k48jVQu@F`R}+`vepxu{mt9hJRZ z3s1lW7WYy(o+S+{#lvB|4hi%UJQ0k$CRz4Yz#10%m zhb>D{^wt67ws!)I%xYyi*M&0f5A@Ma{zPcBQh{Hw6XE>vM8=?Jj47y8MXs(5j8Qg0 z?je)V+^B=#opTn>DR#kEI|Zb*(+W|EP_%c2FKquQ0M!-!U>`|Q|JY1qv7#5f9u7kz zPl}LTNE4byY*0o^49b_12JBD?{drG7kgt!QnSVq%vll~V&3h5+k7M;U#b@ON5j?hk(C%3pC4=!;+-U zpveG;XE}hpi_jUjdnYUlbO*P4EF$4ru-vf+)z*AMPRWyC{sv9Bxi$~R?i9m|%qx&5 z2!&rs(?I3?VaPX(gR~!MFvlhzZl4neqmdzWTPO>08S9o@SVE_9zmbH$;N?>;^bu%tFt~rJy`l4YcO{ zL3(=1@YHuI5Vg;!e6tyxy55g|HOhjzm=(-g_Z{qfMKCA#0gR@_Lh+hNXuRYOA66@4 zA2}J^&r`+Mdexx*&M1o3kb!%xC2%zT04TQ|hMcX^c!!cAez;AT+x&Y1YEQ;O-^?1w z;|5Uo1r=c*_c?lW@dav}@fx+RoeDD#MnLKl4`{SZfEu}a$b6mxlh>8P@48GV>k5aZ z8_t94`((&>t$^|Bxxg8l;#cTiW%eUp$;sNfF*-&0u`~D%b^0 zz&{>HV5?O#ar6UMoZIb+!_18Fl=xCuDclphmX^cx=1I6|kr{TorHGSK4ntc>K8(MS z2nK$};JZ*AM&I5?CAW=0tjG`^2=}#)v}DLIPlSnI)8JR60!UwYiV~ljK$&I;r19s& zu8#>Iu5le&4t@jq_{Z>WARd04xeku?$?zzh2Lb)Qup)Q~7~PkH2@wb$$`OdSPsZ<* zkK?d4QJ74V!vUYH@ln1JKI|OGFF8r1aiKY-o=71j{`Q@;eSd#`~O zwuMfYmGF1^JlGR61&6*;!nfWFbG@4r;H=tq7<-xz)pgVH#{qXt?(%SQT0LZxXM@Xs zv0$vW7p_bJkXgMDCLAe(=&{pKotyx_)OX;#3EslY=qfB%C(K%l@`3!W2F*AMsozgS znAmmjH$DWO^AAFYo)1(g*ux4RLvWLs4hP;<*|4#hZLM zUmpu%#&Iw(oCtZ7H-Y1bBeKJJh4mDlK~NiQ-Fe~9cV(~+a%3$#q+4$87vhZrwu$W|&u zUm^p9vx0>vFDe#!&Iw1+QVEF5OF>m%#-ovGlaXhbFZ#GV1l>M<5(P|4K=I52q$RNl zG;7C@c4j#$o_P~>s9!}@K8lcNZjN@WKZYuoFGgD#Lv;DY0n{o}idrukfNkm{q|KLs zi&m2GTrVFf{yBAU{Dt_LX%MWs|`^dp5bq7ad^G)Y^bWTc(Zpd^)SNeU?{ zyAndSe$PKp@8@}+_rA}$&iA@bhIJIh?mR&`zOQNHo;>okdP#H3e)9FHAtH`eAjdxv z;cGonX`9QiemB&uuEEH@7@A52IR2oV9^JBr&8q7pX>pn^2jG7olPs+ixJ z49SOXl*#|!{|-s;&+!($9UO%%f0VFk^LXrR+=hkwc3}EXYj_u*!_&sMZ{-xNy{7fz82UsE2te4Lb0hS8W^iZT3r!=>aV%}Y>*T2T|!or`F==`nShrLcA@ zC2|tjOCr_hY2F_Xx|ric5flG1XMRtlFV9Z=opzd}n#7=grmaj2V@W_$!kGh=;qQ8@Z>A% zdEH*a{d?(E4NE9oAeyj zl0~#|)Hgb9-a}#DHN5HW@`nBrY)sY zcB)jL5>BzDm)MT5PWCb|jA|mXn2(+Xtz7w&eST)j_X9HY`RjTz9#UiPy24oGlB29e z^B3E=>?Z3En?+{_ZOHf4KkAj%hSG@z)aA^_MHbBE;Xz)u7{r4~v|B8%B?#VnPcsQd#`Mqh} z!2%ZjC6T`V9%epwCCMTsgoZ9Tv1LKK+1k?&*^s0t`&p~OM2`fswt(#{>vuTonAyOD zj11}8W&z9`mPL)RJZ-b{r^M9#Wcyp1tpEE;j`|u9cH6_Ys=6}+hdK0dc?M+KNMUWPI{*>78_7wJ^{5mVFyu7_gGfnHEWxrfeh!mryn( zVvwztY-G~O{Fy&vG(831(@CkK_VhKXmrTcGor#n&*^@u-P2mv|%=+zKv(oY-G%j@) zOVssed$I~xOu7$qx0uQNi?f(P?@~5Yd5XQeHJ=nlNl{>}1?Mv;!3Ga0^!rW+k)bnCS~7ES$+1q<>S z#iup#DnXHLZY!rO{*FC)wv!pkmogRUWi&4R4vSm7pENSfDZp+!OUv2D+z%A6N(~2k zJS&R6m()<{_zdzfRK(0+YfLKFr@fY|$YE^>&5hg19&PJlW=V|hnSEo=L&fOLiPbdY z$_sWiW{5p#xJAwpopi+L3*}@>W9RZMq+%@q;du-in8}hx9%M(250Stm0}RRt;h6Vt zy19)%vm3*yYpMc`R5zfdFEVJ`XEW+cGNq8Gk65{NE%W4eR6G}<`gi$nV(@^HMp- zuIGm{;eiw?d9jwTY9u=!%yOmxOEtTD{x-w_||e#UFgg^NhwM*uC3 zQKQNJ8l+iU#8!V@PMQXz>5?>&zyw2zj25Lm{>w;AeFWJ)ddl;hMa=E*AUl5d2yEWw0@~jvbP!yS=k^ki}!Yjk3m?xBl!fX(b39Q=Cp&u7pWvF zpVmVGZ3^)JVt^i@shBJ^n~Y3-NqGDSQeQHfR!ay{zRWWg`O%MTRvFUn^w+ zN)#1u57U#F6bRURAm2m|FP_Shd{hVfBzcqtS1iJgBFew$y8S%Kcbt~T9IYaM#cM`u0iKJIYl61}il8N!8PUT^EhE-tn7{c8b z-%0cQG@R-fqJL#m(44&j4Smz#amWDI@?O!bKyO<9y%rzGtwYPXG5A~Rxq=~j#?l0j~|A-t!}rL~HkOz=n?+fa9cw3jX+ZzfK+SA8N+{c|Mn z{37L3N?wInu;QKN<9FVQ%@?oaz@MCSq} zDPgh-8D1}Aul3I{=9$D)b#F4whgIY;(uN%0wJ@tu(KJ8Yg1o*?r(MNk5nTO{><`A0 zXYN_{6cOyvs0&Os!kP97>d=8}`^jnYMbH>EX8E#BE?=>#0lU^eHyuR#?w{bz0~*?n8 zv$Sr>XJ(&lP9G*bW8Y`+@8arnH0psHEfF=K6!9Z$#HcQoTJBFKr%Y(}C_|D+bw;C9 zH#II^Mg6C@;Le$c_@n91@0~hEv0)`N=TRMfsWqoKZ#nuPRK@(=H}P?8KPoRCB5#J#q$xmFwPc3$)O(dxb~h} zqVs6lML}pjXrYICd=0zs9jVe)DxV)j`|BzxC-4eCv*i1>7q>{twS^)PMWV8*_^SMl z3e1P8=6)}I-rh^OvDMV$Z;9SvZ=`kFgR)x?Sjp zPwu0!$99;mpL7;#~h^n`U|P2 zSqsIRHDJx_=0lHVF#muC?4Iy*je<$2HQkBRzowu&?GkD5u;D zGwwK|I#Lu7ZO>>}%$HV^DLuS5mIlY4pdF17G{Ieuvc`TU+neQ-osmG{4SJHY2maMs@|rAAyZkb@Ht9>+D$Z9;TW}u@Ia;;+m&W{%wbpkG4or&_uBQChWG)BCDCaR(dv$3>#k1ukXn;H1{OsJi1Bu z&sUPYY$9#mq=`#|G8jF$j+)c!=x$R!Y5e_8)+;7Jd($92iM&b+48M`x(L!=HJxbBr zP2jpz0p4~$XzUeP#O2>0=Se3h&Q=tTYL{r1raM^%zovuzRiyvM5$guS$gbltMJc4w z^Y~pnLz76mr%OV_NC6f9_)O|YpXl01OBCj|$+O6p<$Oh`dHTC7pziciTGZbfeCeU;p+Q{8b7|H3C&MQ?!`kg zuh~Qwd`{8#v@6tBQAQ;>YV=wvkJhhFC*AYm^kJ?XB9{K5$j+1WLxE?u>dg>y_#7=C zrH{$uMO)myIchdapX0jwX+rOyM0D9QXFx%z3Kgy;FTHLm7ojkBrrPaO4Ju%oL}-;l^QLAW^;(8Gg5 znB{+#9F=?Mv(qQq60;0p#wmC{RR>qj&&CGpWvB^nCjanuQXZuW$7R{{YC(2K2r)Qns3>v}}%DtM{p zs5DAX>(HjJHgr1Eo?3*a(zlb-DOva|%{UTIpWkFsrhyv2k|l}8!3?GA7C`b;A6;9g zh_0wEN)6V+)^mDLNOMKN)Xi9_kc8v+`*>acFgD*ij?~e&VR=T56Id$DZPNUWVC~bS zbxQ+Z^R{AHTsv8*31j!*0DUrYfR(^;Ji02$N%`{U;PoO(o;4Zo1jiz7{R)V`pM!Vv z6R_5&06!YHWA8mHh}YzTA6#+GFU0XRq>rkN1}I34_b7y(LhsZI5ZQf;_!XQe-l~D8 zL+W&Bye^G?l|kL7Cenost|X$IO4VvYNad>OLB%v&Z+EAc4)ZB|xJf zGGC}&r#IEc{Jg>f6B@)Z#b_E{U*_i{-$uauVK04u$-%FrfIcMepuEsJnjP2;+r(&W zvU-NMouRn3pbVZ*qEKjYlML#sY0r`c*l>0$iHmL_yXhP0{K(N%GtQpucN``|#nE_f z)laeahUs_AebTyjoJ{{4f#!H4C|K(tL~jRvs(91VHy$Jz>_~cIpQLjOh9teR5d88k|N97M?$c zP@dPx{`rLR+LyxnN*KD7qcJlj2`w$Rsb^suj^E9RFaZ9M~Tr$~5<&oUN50uIGDiSlKvCcmNENKq?P&rQ7 zcTSW4-!@$Pnu)O1yQt=WJ1_mmU|hcpjCQ;5eP9|Q6?UPeemv~s74cHW3l^u-=-tn7 zYAoAAbb->u7G_I2@WO0Dq}M>I{*@VW{ZrhOiJg#x$P<&wqb}qe%%d>OPT6+lw{2lVnewdCk zX$+g|!urQZ1oqye!3b~CP>QA;(LCzW{7&5ZcT{)d6>W9BOk#&FliM*-h}(_gQnW{K z2bY?2Q+-^ZdpZvP#mquOFuWv+o*K?&svElA>s%?&(4U!b7174~%+2_FP7ahgeAbSIB z*Bjm|_l=_TpVJ|?Nq8yhfMm;E*gu7z`|UW2GZv}1B76uH=blrp>O=bB^PXP2O-FqE z1c*Ph#PZ!M;GVezWnbKJeUv0>Mtq~Glf}_5$=9lTmOv%K4?26q;h-xAweDZ!_EQCK zG%T@j&j5{m(n1~nHz~O46a8F14mDei;1Rk6mzzrAS$qS=D~sXSwi>f5#)7jA!SSq{ zbm5UEGA_wrUljiie2k`_GaJcsLl!OHc!9KY8tLF68LT-v2A>+9QOF28TzGyM55_OU z9m`F;k9a?toHEf9e~RqXGpL|Hl_XLuAd@p5yMGB`%<44qlRHfxca_kjnNE1+>i|8| zb=c{l4*&MKupPM)gA!+Ht$Yi;?Z}`Ve~ZAeJVdM9gMoz(M3*~ah@VFf@+^bg#jP+^ zE{0o<1Z1u$z|TV)W6me>K9FR*+064QS`C=SGm+QdHNopjJ}tj=g?jjm+=Z7f@tnW| zx;?3vdWR=rc%~Nab&p5#zf5RWx8Yw>1G=WlVE5O>=CNwMn&+I2xQp4?yY(?*J1tQf zod`|;lc<_F8l~Ickp2%Xh`9$MX!8LCK2N~knQ>&^E`Y1mt+ZRNiO$3bqHoP*GJdcK zGhU~ouQeXSBClYscnR^I&B&6GN5QZGW-poz*)2Wvd|Ep--_=K1ZVmaL{zb8u#~`-R znv~_%)4s{Q6gc%e#2Xbk*G&_+#&su<__zm3>Rp&+TS+$WW$@r>KkaWE3+GZT@Z&H% z*fL024}Z}b5mEfUWQXT2`e+(I0iqunPESxrTA(Uy!>j4o!*_Hsx1UViGH680by}Hr zj%0ZTU@`AuQP{77zMnaG=oOD|9Y=6(-yiz3r;rW`zoAX>n^7DahN^vD=sc1{`&-MX z_hQxCW2h{*n9G#FXP!Ozn@d*pJMnqjF)UdTf`yt1G<9Vr&H8YkCb<>TQjrI=ETN0%o|^eu zxQ3P=yFjqKLjh_D6q(scPdv&=xT%QT`C9MgyFgm)kxEs;`{~`A)!^iNF=740Rq4}QZ_uQbAti_Et zBCdYAG-oP(2ptRIn&vf z@a+6WQZ6{1nuWp7dw;QNMyOZ*2uMsWX3vH_vu*l$j zWloY@2|i-YhE8PdzJoVi1?UnwiCmF-C^bJpsm6PdodtrXIH9t=7UO(gQw4XEwxl~C z@slZCoRdPLz5}$zLlDvJO;p*Wi|CMSDi|SyGkaFRW_H@37Gy%;=lJ6`E*qf$jxm(tqoxA!Jr9ZC-na z*NcR4|F8~4Snz$D)K^;9KMIvay|iH0eng!;fr!ywaNHe(vTb8%)*^K}dMlBlW%;}0 zstwOK6=L12Qt}NsNFfjU$hs_yjs(A=2TFfv+8Lh5%*iB=qi-lDe-oyjGlcF^hDG5g z$S?9At#zJFO8HsIcAne68$m6f^2yPygPv~D!g$_W!|ztXZ0A_WX{Moc z#s(UdB|@v`q?7Hk1RC7Q>lyqE*g4?^3!1M+Z}t>X+HV^=-rmI&Y};vLi8ylX_mXxN=+4qelHF^A16%m# z6C{I6TkT-=$(lmSmeAqF{!AwIE#$Xrh?LEvNWT9$^;@;c@ZZs4;I1Tg#g3>aFAld|(9TD3ldI;1X9vg|Q>I;4aZkMro`!4hh_ ztOCK;)9I<+Cej}LkMuhSs9Sy&!i^tc^?x_0eLO#lczzoGb!8-8oKLQ`yeHz`FON`gQh~b=UUV9oQk=RGHjGy;e z-e+XPbP#)MEXFpq(yxPyp>Xv$@29vA6cparKI5%^ws%*MGI)7PdXva0zSDUseB~jwQ7n%{o(aOB=P%uPwF$QAjg;q~Pm8>} z=uUMSo#Ayxn|y85Ul@gVg6=Oy?Z!MLDOqH>CF9E}cE!MxW=tp*i&$ zD9Stp{qoP0w&*6E?0H5TJs!}^<>j=yK?s|4HA!KwFU9K;{+()|t_^1>uzCw#7I)LL zzk={fi9l!hI1(LHrLFwTM8fAGRgJkx(&i2bJ^!0LZ24^Uj%%bEUq>Tvy1;s61|$z{ zBEh)(q_%81&o?i{yyh_M@SXu@ziQn0orhmT$8ogm3K=i>KnAwaw1w9#kI4z~{e(E? zRDY(b;mfqdHH(hy?x7`1F3`Vdes-Ax+BH*@EPl(A(7-&Lw=u_v+lDw`bqmLy-@^Nw zL#RFUliGEAX(#`A{Y5O=VF?b@n&8fECDnB6_kBy zK9%zf!G?4r8ZR`DL}P7e=&LHVxK1WhdpTN_os1W0k+>DiGvf`y7!rFxBTGg?UG6-7 z)Rf~#$XkTi*VD*m{`*~WpFBO<(bV+nIUyem+AXOCMYdU!GTjFQf1 z(?*^LO7PVrnVu=g-8l{0UB{py!VzDB6OkGj1+$oJlwCQFXTpVuwNgZquMi$@?4Zpd zi_q!ihD!Z4FjAG^3UyAv=s+HNo%q%Om7B4NIPAOb$N(XTLGi&wc$-&{pd(^Y4QqTG(S z`!G%8Yp`BROstf`(V&q?$=9aoujbIOw-i0^Frkt?BWcybZnkN;3d|)4{ijuVrr8P( zjk-uNc7@DCJM15rgzn%O7@M$#HvJ7Ce;1x7>z<0kx})JLDTO;)k~HtYZ)R03LO-S| z&{#Mrjsk34(H1ui`S*s%L`e=kyZ_N$*$4C>VH~#CYvI{#L9AVFNnJMc z$s|FKjBRsp`p{fy&uW7H?Q zlvqXY_kE!Do-0(fD1)M-BKgl_%)MT#$?a@b<5neh!gEnM-Y@7!cEBkxha>2zNyFp9 z2Pi3PLx@~Ie*Vis@rZN?m#)X{WCIi(P(}AIe!j3$0qqwzfE%lgt-PkDBz+Y9S5uKS zYZ?mo1N9Me;gMww+qy+?u@1s2twMMoevC;vg=n`;!G?Xw=pPw@DQB7})AeRWG3zFQSCKK-O?;_k>(vG&{ z0-XJ;*%+KH2iHSgbT_^WRtMT}>fjCB(e=aZQoZxNn9P9 zcrQe}YA9k>hM_X!JSLUKW3KNOoRT_*gXv)~Kq!tkEym&BQ_y2R5zDr5eBU9)g-W#H zWQZ77fA%oW1T|qD*NA02A222kn7U^gt{U-tb-NsnJoiQJ6nRt~QAO)P4V*Jr1zMSe zP{DN=d$|_B#QNd4stq31w(ulx>V_x%`SCBpUd_qg9DYkpU!34+?IaXI4_*8hBl-;c*~sYS%uPLSm8Uzm&e zjb3O9^?J#UAuD=Z40}as1>VhGEMkl`6fjzCxP?g(&ZIc!v z<6$VW$85li6ZfDx^C+}Unqae421|lBqVIPh}$TOK|kar;YNHuB-uENu&2}i#! zhe4Y=R(fwk@L4-No2>vJe{ai}GYQL@8UN zOVbC}Z@8h1l(FiW80y|BU_Ys#(yRuL$eUG6idlLpb_+7l7Yw-K!1H_DuL%}1yui-uEb&q!9O>ZIu z^-A%kS%C9gE5|t)3UL>8cVJ}1JREoyfvld-q!purx=9i^5_1_kH}0aM^8*%MJPX<5 z^;oE$0A)TG#qhT`s;vXyZl3_v_9~P%e1NjREfRm-Oi8y#;9mB9%&NMIir~9g_TCih zW5OYD-3gyF=i}|Pr4TS&itjIKk$!goxvUB92SvEK%Ea-S4>y1NQY2il#-`Lz6p!We zIo#`!`SC5zwR|G^R#yy)xMO0s7c{d{aAu|tw8iDnGHWTM2N&bvo??8yTZ->##aM9d zG~|!x!+AnJj#P;vph*k-Fa}??hr#d1Mo9bTVb)@K)Ib!K%cEe-Yc!S@G&tdKHO}Nr z88*(miU-EeaCc@pI=&wm9xnq?kT(X-k8LmnbEC4RhnO3O#iCPn8ob^wmD0c zj2nWO(^C^>yds0q?wLGSVoeG=Ptr#5BIYTj$>I**V>2u7Go|@;rgM33YP?Q1dlayd zSEBSt^*meEQpr|q zb0+h%$xOAklr>BeCnZq{YTOygT=OcJU3m_B{W+dxRur?TPJc}s{qLG42)$%W=O|LT z{|(ktn?>En$I}y@1N99HBMqK0wSQ>JR((?@i3xYvd+Bl1|N10(kN;+x(DljGS3Qx% zRLfG(sW5g%V>Gq29$=g1Ml-Lq7g=l5f6U@P5mxRN#fEa^S!doq(+;1}l+{~IA;)g8 z`nhfNP_f+9+-*YMoY!x?Of#Q?J`(1vqv3pE$nk;>o&TBLHy z)Lcu9<(GvqiBL`Uhkrg!e|Zn-4hww0tKUPs(@Ihqoc z$>yxSPH*oQkyPd?@>+a^EI(9{g#1DFcR~jJo&L;}E3`0m){Cy2xQvUFj zC3QbCHP6x_+bKb;VZt@jZmBdDXW&Z{a__S~eJ47kXvwtnLQJ1)IkL&J=h=*ng6wsu z64j0M<2CGRrhUqTsuLV2u!G1#>jwFx@*MPL36k3Rm^N7pvOOn0naY=`)3p!dXr1pb z=BA%d-(#XpKTi%bEt9ff>q=jeW7H4k`)U_SZ6_+=%$QSL7~Kk4z#i93=h-VwdiS@A z)&&mHsuf?UPjMW}e>IW0>{`zzt(B(--p-`lSVjhV>hz>%JN3wfklPX+_NPOeC46*c z@rB#jA1P~gwCfZrNmZg#o7U4<)j0YV>`iXH9TdM;1@pCJ*}a{5EMMH6CEv`ak~RJ0 zmi>pEg=RCE1Rv&Kv6wBMITriQib89M=d;|$V3FEb+)@S4MXg7$PCFAkq zGR)M-L~8s+YR~H?!6yO;JaU2+`wp;gdgDkiemkoX_GCg~H(1T9cy=Zzl~r~4v4cv1 z?3vnOHaZ}m)w`KN@})h5y~D6_nG&*_(x1(utw!DsgfCe_7zw#B4r z(jWf!eEmMIyCS&RqT5Q7ul3M^DIC37XiFAz|I*WC8c@&O zgg&8CEOb)FD0^v4%=`~G&zoZ~eG493*oO%V%}Do69`B2JKt6vGnCjj*<{9*XX+4%k zO`;v_@+#3hsK|Tg=g@-j!IW8agt|pqDfq55hWZX6z$Xf_3!-3C*GePL&WB@KEFPZ` zfycRin#tGw!tI2?h5q>A!|(qnzf3=MPLjmI*@%5&&2FE(#>A&RV%?#)1u07H4@V`rf(vZe1vH9z|AB8!WPUDBwZx-Gf zOWx`{KV<67y7n$)%^7Ez!4D^P`jQ`8J2!=;XUSosi3Frv1ToIw5VJEZXTeS8H0}E@ z<|@3Eg1hEqbD+nd;&KUwU^$Z5>YR-aAd&&OPk6>Q|s(d1oW zMJ7#Y;D)zj>yt3d;cVD0BY);CzKq>a?V~9dzmsX5HI{#OM*E`!e70yj-)CfF313sx zCl*1?SD&q0Ig^zxO=6d8Hz8)(2D~sxK?U!>NnR=nmr;hO(`cdlUPWZwQAjNoXP8p? zXBK9zLTMVSSYKBv(+PRSioG||Ubl<%x@!V%Tvepj{VS;Y(tg_Jv<{8|e$XpSz|t!z zw7`5Ty?{{Xwy}qG_4Hfbj zXkLNdCTINH5{3N-mC^E813Q;3K=_z!*6{oWlTK8lwLO*m>{Jx-GvCmD*%eGAzL0q| zHL+{Ie$uIab!_8vw+#6R>v#Mme7<4%&72Gx_TYqh*C66#V4j_aq1=B`0u1>nIf8YrryBjGsaDlRf#OWzSL6rtO08 z&Y!fj>mxmPlR;xjIJmS!xb&$IIqqqwl|F*mY9lum8|3+f+as|j}zah;xlGv|4 z8WY!cBW=M|jEm6YR(YvFSX&fLX#>R7zJW@mGB=Ljiyn5knY7u40d%rB44zL#?feq>{L96%?v2=6_#a|E@E-f*NLc<<rLgp5Id9 z8i<29HwjOa*P@8mz5YDffd#=)xTUYbO^ny(f}$QmZ?GS|J}O+_Uoq~hhc7k;9Yi6& zn>oE{ELR)#p5Ny>feXsxy*=W*XY#u@iU#gc-He+QIGN9|$rQtu`8v4uk7wpYjL#6(;wO;q&`K2<(o= zi5XjP@(!PmTrdKS1FE=uzz_xTo4J-FC7kFubI#p^->;P;%KbI=;x?54oAv}?u+ss3 zd-`bH%1%n0B8IPxDmeL85UF#e5ns6hV?XA@Gcph>3Rhs0mlL0PwGw~H3wPEY!fLB% z=;^&D*U&$-#-@|XTIH~Ifjqt(l!1uQ7y8~H1DC>wJWsj}>fP&cL2NB%U$f>sM%i#5 zqn301_HSvl;$M1oMGjA0ZXl7*Sd8(_=lk(F+_eW5TzXI+cI`Ccrf(g`J=-V4-CiVy z^F`8#k&wa`)9IXtvj#Wwx&(KL4nl+Xt-H)9#AWm2xOwvktlYzqE}coH4OR54m*;$@ z%HyY+Efz#`c-NDMpAV1YAkUcW`N3fzP8BoaUC@=Q$7MJvaYsjbb8$n~*eYRyHKNln zcbPtSHP?o-;rFfWP&|%gqiZl#I)ZzuLu6P~OB?efa8xP{L5hd)&mj_A@D(m~N-noG zwTi3MF95rqjJ%~|xnH%R=nnUW;QPJ!iCo;$K8K_Idr{7x5kW4a_&tJR=)IAP%(wdx zarG$nZvT#-Vjp=9N1Ah%)4_^Gnn)DX#+OkukmE~eU*?D8r_H3P`=WWdWY?(y295%J0`&{qz^&`H&Gv6 zjCqZRarj{{JWd9{HA#!}(iY_!i|_ItU@d%kHxVr*z4SZx0k$P2V)NB1%#`_pyIcj_ zsfIt}(r~ouFdpV4fw%O*biO?P`AfoFcO3R^nT4j@u~&$$j*W+KiRiHdVBkNUA}t{!J;`Wo-Mq`Bw>dG5}y zv7BXqG&g=tA_N}q!oLuA82@d-)0_GD#_z+?FS6w>Y}?KSG)?BtSSLV=)b z2J_ZNAocx1+?}%w(lVilPEAJWp_3?8DZ)f43vSU-9Zp+Viwm)zhr8hpxb$2PP5Bbs z6+0oWcv>H{Itt*3V$AMI$BKG&PN-oN_u{$$r+CN@EA7Iu?a)fhH`V4kE{Jf4uZnYa zKbtXq%~kx8&&Q+wbi^wi!#tZc2zL5P7Zx^9O-wFni{_!#q70Rv5-=R}6MijqxIMWB z{@SrPf7lP*iyh%7^B1FyYT+VL5BCk>7+vp(EuZ62@qQNP`C5Vd8qfp#rRfNBJc)b$ zTlhMCFaG1Rd_S(61FfE`c+`9d{{E+6kQTx{`L&aCad+UxOJ0UF&lrB$7!75o49wS# zLa0XoPUkM?Ip3+sNYTTYS-vO{b->X(^AXY@#FZ_)h0Iny+a*wq`;(!}`RxaEdG(uO7!W%?MduWKoXtc3By@yPW z_gTGGZlc*T>*<`}C3>inNqHGk_`E-pT!g);W`Z8wTV+X&Dn77j-Gvk3hSFsB}R#v)d8L^y;$N@|8l|(~+-ma^h$%qc#vfXN*Fz zo*Q zufn;WXk>)8u-^PrY}CflWHN#fu!HA5=9}=*jMeOI!KGr+?FmrPsWaj^RSd9BOJ-%-g*WTM`Xi$;LOg-0mRHz6kNrg%xloHW0 zQfW#VrM3A_bJkVtU8c7-4q^w z7GeEj!bg4-dh9r|=hW$h?Et6#I*1Q@lEKdsn#?=qcEfJU4>U-O<=fA%A+2S>^zfw= zwcEuqM!`brvp!4{t!B{nDF?{DK$!$ZRFR;y4DJu4(0WUgT#r?9N|9H&vBOfNrudTU z`TU&IoVK05TvMS?p;M%|VlS=nSWH(l7Sc<_d@`KuL3dszlKzw$ZlI`=lhUu`K59wx zo0b{!PDk(J&qrqpGSZ?PPLC!9)l*ZR6s>#hNA>qL`OuCpc+2hvMjyW6$=*oV`ek5A zYZ3i@c!-)$t)i&U?_u8L4~1*zuwv?e+?M25oTz6m%W6krx#w;eEe}BWqcSdG(Q!`Z zV>-9fT$gTJe&=kO1V{st6gTM*H#+$S_dwB{`uYt?pI?6oH!lMfszMalc7*%ldWOqTj>RgU<+$a_v>XplDidEs2lTaQ z;NN~~2^6Er4&HQqY947R?5E(j9`wvJkaW%T>2k$%l6+%J6B!>xSk8nbHkQzd9&ZvW zh#{jzez28bUdFjbNE;eSZ$&O~`EK{PExZ5){Sc(SD-~SYvmV^&uY_sk3vBN^LBn|u zXtMo9+M1S30eiAYZeJm}zW&DT(cQ)Q3EZbU`H`gZI-PnhdC<2pF0^e|A$^*#fJS`F zr@QZRDbKf_lxyAS{Pq|M-LZ$fn#$;LvLMP+gy3?PWqZPW$oEtPy{)}X2UvC{d|CtMV-M!t6_ zd~>SlNvkbw=o?S*%P!E(CINg7XPNG~r|CQMX-W&7rf;e#R96>7!fOhtzFt#C6v98wO}6l*d-Hl=oS-Jk7KMyuoU0&?&W%f5v`R5zdJ|M(=&f*wW{cSQ|-vHLS z`fwi0<~}Y4j~|rc-+v8A46em+QX&}@gpug}nT$)EN_Rw>C}Cp}ZLa6A!NdXaO+L_6 z+D6BYg-}P;SIR1`rSSG=)UbLSA|w(J`X-ur;tnD#M;?D`h0(X~Af*~LP)vUS)n0d{ z3*Pd0|M~&BFqT+KOAXytsG|BS&*fBqm18rRaC zwILLGYY~Y|-^}t@d#P`Q$(=l;4lfz%ck{T z`{}rWFt!REru6u1I#c-mnNo+Vp&qiOT1=4vmGPQ%! z#z)exxKav!nnAjCx$OL{q9-Q%8Q<|I)tGIhUEEQcG4mG9Tm6cvpFW{V`D>6o*9%>b zN9dcDMV|N1vGe*kl_`o~i}h5bCrIP)kTF6VKGR>BFsz)VfX`y3nB{#g&JYpmT!E<4b8@sS)3AT;K#Rj6pnfYwd;FvquKlztES)ja zZy@~5QXFqNMG?-c=;^wCE@}O45_y?NKN;J^_+l01+A6}vKotds!+2`fj1L0}yy?Ch zv`E?j#|0!ITf|rzA8wG=%X{Q}X)n#cTuimql~f**OM@m&^t|T=eHiYgU6Na{!=r(? zwpP+R*F(R)SHd;o5sh>%pqROfW zj{RZSa`G@lW8=|zhVB03auF@`A4^6G!>LysmkQO8rMwQis(f(bS_RvG?Ly6xApF}| zj?1;ziF=z&^F#9K<$*8Mct=Y$CZpg`6#V?8(Q!-@#ZNR4c7K@0hcW)7 z$O8B!%3{?b5p1(E$6sR!9NHuV&oy&VwN4BPiSNmLvKC&kZ1s7T7uzAPi?3a+Sl4n3 z+a7+v&UK;qV;2IeykwLg{X*N$zodI-bYL-Q5%fptB2{+*tVkJme5YWTn~G}1rD4%+EEj{)q?ZQtZRv<;f@uw!+!>lVJK`59}B3ftheK7ECq9 zb$$}w&hmh2eJExH?7@T`r_tr>1l5ihn8@41Eo&|FWyV8dR}O|p>;bJg45{bGA>gPA z+j*)uHBAj-s{82i*gdqq_&g12Z-l^&cs%frh38#k= z83_Cp53|_g$Pc;ASab>~3YW)xibUJu-SE>Wz|}4dC|(kVk%~6nM9xIkX-(EAAcRlD ztUD&%8cGW6*@~RSU*4(AubMcY7cvvUyzgBku}c|&mb=iM;)_os#QQEQtiCAyWo&irP%%|etMKZaz@UiNvez}O`r7}3wNX>0ai-?=1g z{2Pwn3eC8%?iDWFeTRdqm^Vot$T5k<=>RkM2c=@KQxqP{i!(kg+bjIghV_Pci2BS& zP0}P(9BQMTzWp@L#tWX>pU5!cCv_OPp~j7{);bKdtr^+wX2^100e)>DPHkL>?I-u( zo$WrXY9E8aHXd`zrlR$P2Bzz)AToU)c9>0rvt=OS+XLaf>?~$AwL!;Tlvf>Ei4A|5 zuF@*X=bmSN?c_PQVc-Oj2iB-gI*3fq3Y2~MPVc*=5V2(iq?(^o#JDAJ+GvUy?&|n{ zPanf;XQNl|0K7VL;OQ6zOHa0sIkghfDGqE7yn?Sz;(V2j0Dr=vg`FuQk)@)BrfYjK zTvv{a+nq49Oh?kL%TP3`#NOjJSa!_@kP5(SArV}hp@LZ30qU;1PM6mHrGszPFve9I zPcm61c0n@SEh=#!p&h&SYw^35h$FFZEmXukFk);m9KIF6^`$tU7WWn&W2O0~56b+! zm|pbz3-Iqh>F|G648l$6D|Y40NhcMK>40#nTJ-eI}vV z&;n^TrkGYF$=*}r;WOU^!dJ8)rR$0|qXbCz_#?Rd6l!ObAgQ1f28T-VWwQ{UxKa`+ zvQOA~^o7({C1Hbf0!D}KLywIuuBlJN0*xt{8F>r4y^|pM^cfaDmgAj&+{UJ>Mtt7O z4*JV$Y>S8(|D2Smon@{8HMFw-y+uF{*( z=Ekyr>Y{u?&_8513-%Prmq8^e=0T5=6vi)Zo5htKk>Z-FnaV@zKE zX;{9|AAeLL@aM%5_|)9S%dbphd%PHR6J}%c`0e;LTLbuOgP3+-bR9YgSLS#A^7<_f zoKRr;gdBf4B_4N2EWvL1c&w|?!LuMOjJdrLi=; ze26Z<>8U?()7=$uqEQ%6B7Cf6J;EkD!!uhM{*mkgvS7@Q5qe>CebZyA)EWhE#nD(k zzlyTCPTE)ZAFcZDE-lGpy3mJ9WaIyx9=}z9$$J?D-D#!`uWD)D_$F#Rypm!cETNBIG!&?*Zto1KNhWwD`#mbaH72#oU|;v0(yfl;Qp{Yr38Qa$g-ms!xA%+lm;2?zjVEN$+ME@+7qR)#9|Z z5l-K*#t3Z_%wkOH);|%nY-<{=b8h5b2tDD9nikUVo_Acyr*_VvX%VH4TSJdDwo~nl zNb0<4!FHr>%!HH*{rQ%Z*D;&!CT5bDs4ZSB%wV3!3pnd*$aE89Y`Rhmi8QvG)xLy^ zGw*TfNIVVZJ)%d8KG3smXX%FaMM_M0L>iGoD1{1~HKNh-#egF7v?#kLhFm{HQ?YCe zIZNe|S?U#3d@MzW(HpE`S$;G9nK-&P2FnK&p>)~~lY^LtVQ@8qTaQ4q?iePXUrDQW zn$okXV5(R#o#qv)(DdfH^xsbsS%%SX34^qz6 zpzExU;iG0D>DXkzSLg~3M4UmIwH_8OHpB|^c>Fv2m8P(>-Xcg2>os?fQrJGqugfCc zo!Y!!y(9mA-ckNVTRHs0Za`g0h*!RoO7<3KDeR^=%r92b<*&IEHPk?}zFs4xwkG0h zk5iN4WLR?M(1?kGo?;PYg+$R$uRo-yl0$*)&!=erqM-NR=+OO}WYgbIVh2-b+(Fid z^hp-?s){I6c#lkCU(nl=q#?Rhyq`FaHr^LSj2dO=%ygfQ{a zDBSz81|_}%yxa!~Uf{GmFPifJHPwTtvsd6>)-+I)!4Pen{*z<|pWw`?J9xd|4O&IV z(fvAI>iNEhMu_de=w3feW}eV7t9!ZWWA1SiUXLQV1&!35P)_~x1fdz}Ko<++sqft_ z$~9a`w|re$o+p;Zy*rBS4+gO3&TB}Ht0B8BFKDOaNNhhO!T%Ec47(Mo{J$x$p%!E_z7u?VKkU%#D%&zV-ZS13a_pPkd$8ju)12ECz6kpJ8T`Qc&|X_dh> zRgh22al@6MWOO~ein+C$cI)>lmVMKO;l)!;WNHH+Zp0y(^X!F0#HW*F0J$e&JH$shM|;*;m6qqO1@62H`7 zWAz^rvmcFf26Awg+)RI1eruZ0OIp08m8w=6!O?9m8Vtu_#T6Y;zB;wU;f|mL|LxNx{)mz>U*4jCj|GF|U}%oT)0;_t z?pLzktA+EY{v(63C&Zs)xiVW-9AbAI%hBs#YyF)33!hNk+Ywka{{>BXZ-StyQIHWH z#A`K8K69}&uisdO#oSew|;^dh28Kl#2a6Gp5xbOi{Iwq@{XMxPRD1y21fo$b5{LA!t6^;64VAt$Bs!_UcD*vxzN;X! zY$BhuV;zCM3 z3&RHH1sXU_=kFJj|6k?>JrPG4^<}i$&6pONO{Wb$+W(c?g-mqGB$33&2$Q7jAraF=FEpP>DEbQ)b~cxo3ELq_4F|*SF^KL z%9utjvZ3v#qc|z%V}3T#j*{9wbAI(xsIRt%2JM7sFe8tyChet$&=n+f`2x-D_Cjh+ z4)q9+z`4M4=+V}p87n-f?)Plk@vDr>Y+OT{m)=q&>+T9o5+J!*kresSf#OcQ=UV0_ zQgzQoE;P7<>+}pFQK|1->m31n6ECCJkAKkBguS%DzJ@-=PNO^P-qY3ps>tK221U*p zK^g1E)BU&@q4owjtAb5-Z>GG5~*vilOl z{a9c^3%)DUUoB;B`@Y-U?-rJWKik69Deb0*R_(;x&9rZFJ$+V=q@v;gF81+xDsw(b z2TthIG_i|RuvG!Sr<|p2ekW;dDa)@7$dEzl3>147V@mlEvi{b|ZPMOEPo5~yFUAOZ zvhXg^-Fl5IN5);X+DBP~N>cl;ysfYn9kXeh&q3 z=Nt1~fB3<@R`8{HQIpBCdo+E2>r2_zdzcSOo@DyhP@HrlH!Q$%EN_!=WoQnrd=$cH znP1c==1#BE4Jjma5@ieLb7JW?xJhjrNbIXWS7{``K4Wk46&Xd#CG049!ZF%2bcAb8 zc}IEddwuBcK+iPeXv~%py7R-9))X(L=%;5XQ%IGL%=IND^UrjyB$;-|OeC-QY4p-U zl_Ccm=zI~|7av$cT3;I}cJW`Dow^)t&1!V|av@y}7o(NMVf1nQFY;Q=QP-4SdK9FJ z(S@7I+EyB~>}+se@COx6h^DYxN6C5y3j+&k;hLj1(zXZC1?L`$og^0z!-?Lz zu|2nu4)$B5(Tv6xJPB^1C%Qkl;K!RtK~@}tSBy;&|5XME@ z+|3z?Z=js*&Gb-*CvBfe)FkM1MLt=^mj9)3-D3l>{TVaWB%YugOi@$hk!zeuS!QpD*ObGio`vX*3x$ZLCLNkIgNi=sQ$oKS zZf<=<9UC-I*nN=o)QzXe+HI8c_c;ytF|K8uHvYB#q@DjpKqgoP|Ar=@t~m%dx9?{@ z(`ei_4Z-rzENl|_O!?>k(QI~39xr`JV=q6Ujwvd*&)6)ZHCEHdiK*1|f^pnlS=0Kl zX=L}?o|dy-zWDehq&T35<<9$&Qp@;!DgtyaNR7%v$I%%7EhyMB4~-#-n0RtIRas~< zUZx8b9v&uPD;unRzX@S}dN`xDfbF?AgSUyK0KaBZY3ifV2W{yDZ%kW!R#A6JG|Aud zr62b9$;@~UwdvocA@gDSE**&r-eW=gjzj*!OX}LwL0eTvunx;vaQSTnvrcym7!NRx z>w30t@khL}8J3(Jk1ad)L2cU%JYM4jyR-0YHZVcZ3(_IgAv zXBjunRuBI7C!z6CAia5bgnCaD(^_-JaHC=0=*4~&jzu10PA?W=yvElbsX1K} z>o!KyEZZvts~)m?EsV0+FX`qGV_T}q@+KM|VI47&9}%nrH-R+FDM`kO{v;BrPNEwI z7l{k?q1TLko7EUk$FI#Gp}vJwPS#YOI1=L+OZmz;J$Q0SlrXl4b$qpuo^~A8O3p!j zMi4>=77zs)lFrQ-YO)%|&fU3WKf{OiUs_9o-Rmg#T0Er}{3f9(ifHj11;2Pz+zN|E zKJz+zu&hb_oh`J;VK!N8YNJu2Z0DV`i1{aV5ooQCC92{0)@Kc;vS_3v7C?9N82Z7w z3sQp}DTV!hW~M7PGG@|!-9>o#U@uO2?ZwA`g0T4|hc%G~_&lgh0>06-$9Yr5$ey}Se`g+nQQnc~=; z@ix1goDw?)h4@c&QR5P=Tp$nU@P009k_>6SlBf2YRtV~LK_^XOyVIl{bWv^@4Mh0XpzVoX2! zqBRGZ$sW+}EXK<5Gbnu}>o7>jreBPmTsQp?nYl2YddFhw(UGIu`PQ^w>%)fTvt z?T+co3!wM)B>rBxg56)kaD6LdJ*^8yd}A2vXFrHw!90w*p#j}U#%`0|fLfs{SS{U+ zU0G?6Q%gi2e-y$Z#nAC(^WkGn{`&zf-sJYhv#oTf9qk#kUWW@x*yH$`X9=!c7n-YgyNrvn8f15#k5uO7KcqBl&kb zS7SGurPR+yvM!|$XxcB%-;x#PgAPlh?kel`9$-A7EA@Ch;RR+IK1S=S@%&>8GyZ0z z3BOV$5fK)wU!CMo_rq*X8Z^ZMwdwfzh0VI}#xoD@a=elZMQ2Ss?B6zH?}Rt_ zx=V~#Fbsz0>$4DlbP5S4k}!3+9I}J=;d!T<5qW%FB1C{v~r`Gcu z_XhJzOt$c+dzPWl@DO|qf}ruY02e~@Fu^(p*JNe+%Q545p*|P>-l*MB8jgUBhd;KQ zt3|cM4cwHIKG8&_iM z21n@m7~z-D9IUz!f~y0!@b}3GKHFV_f4n0G+AK5g_lM=MPoBWopM`jkk&oVlZFuD# zg&F$`anIV5S2j4rkDY#ye||%N*U-}7vsKLb^MVWUGR+H*3d1ma-6F<^m<_cnVYu^B z8){Xvpqb~)bfjQBcT0xj$2{b24Z!W{(@5}n1i=$k2;TVwJBNQknq_B@#J$3aQw*vCw@)9VjN#;g)WUb z2;OuGhHQ57ar^@F>%XzmW-KpsWF)_`M1^0csL0C+>+p@K-f-Q26v6juF`v!Ef4bE8 zsA>(KQ!Pi=?H>5Zf5Cm$rzLeQ7EUkEpyo(DJ&gH8$0DZU$xt-PZ|9*uw*}{W)1euk z53cF~>fbNJUFMe!<5uI+rSZI_u_hlfMVHU7GlfF!avbdrhx%RcXY~#E@Lp$LD<&Q% znXkG`_A)dA#vm$dJ}!USfbYrI5R=e`sI~!|v-RS4-U{dE)g|%Da<|}6SdCp%yVx#w zCoM=x4t6uybKnF^wYgaarnMqKD$rzP<@CXo3oKrzp)*t zeH!&@exQt2#+sfj0B>^x%*a`YP=RTXE{TNPcW=B~wio3w7f9nrAysTkru3PI=xWLi zT5~jt8WhA4cXSqlC#o|{#Z9jGmjK;18BLk{Trl3B-6gykFUW2b(ic1=@pUud8n}?= zn@yxJCoTFn>Ml)DZz1W8iKOXv3(A#D^Z0xNrRV=~?K`K?P@_A^owmd#Cw=HT5OPu+ z;j}CSeH$9LQioz^2t(Oc1Nzl1bo8t-8a`X# z`(B7J5| zLp6D3wo$P>>o}cmfo5+zEZiuL$GM~NmUTS&{*$GsOJ-zMGK!KfNnzuvxAZ}-pKSg< zN3#d>=evsWH}#6ixjvHhB-T@3djlyuwbAQ~pGn%1aa%uS(EjVp-_*eJn_q9!X`ySZ z3yH{QX*kQFSx`UAwSLbCqQ*zNsZdlMbJr^4Y>WijCyj!#~2S$*m>YIz{;zTb$Nj}eHQCM1z~dz;w05K~ z?OL;sY%S;F{1|KKq-}=!-~>d6Fn-&3br@86L+EZCz8v(ybCy$WoUM-P4Ll}EVMhfx2rm6C`%$DPQ={|@RBp~y91eKO=rmB~Q6!(#J z*Lg0)ESae=_%uX24#!d2y+Zn3s)x8V6J(7x!1l5-N_I`4b*jnq)%OB@`TdqMY~Rp_ z6+w6)=7uN7V<-_8)@biy}Aw87s7mPz#XXIprY?Xj%sT90td}LfL zVf@_e0@?bcfV^EHHGUy$nzRE(Q!JJoVKUf5mwx~wMpDqsMF621zg6v&z!xc zKHbkXq18RJX_?_PdXnx)x{>YlfbB2ps<(3aUbnc;wNA7|JD$_aktFi?PSbvmATDGC zsaQ;>m-#s+)A@9hBe}(#)a*o(eC9}RS}xOjom9@l`3radqaz)z@#cO^7U%X%+0V6J zspHgZ)#;n{Qt~`9m1ZaWM;eOd)VO0Z_ekc6Ny&?Bt~hZCnKbSo`HSt;&*ryvvwVm| z_tBFHF=WB+n*ob{(ysD5+^pSiD88tK?8i$ZQ?ZWDRJ767NAcYFTLv_Hp&zMUQ{>!C z1G&!Zcihf@ew@I2Juc66h?{pskj}C!?q27&v{+*eWd#|NTZ9RPeO6n%=Gzzr|!<%HI==lhC0jkc9$ zIr80fIqNKKyuOEZsJ&$V$|)R9&*jQI^~pZA%*3c{Iw$X-O{$VCbDCAaJu_#$-Rq3G zmDjYmCr>@NZ@*(`%&KiPu2cZ05|X&$-`_c_VR^*==r$3J7vjEc%I60Dj$|E+t&}rC z9vNe%QKCjDUF|qaQ43#ka+0x}DdUBOS;>>=&>61xb0FoPP3N@!o6Q9Y)NyGEYv|n5 zgXGAVA1@t!xCLhvNm6?x`RN;Qf-g>UhAOu>$&XU>^?ERAk2J?FD`Q%er$YZW-y`W_ z87{8Nja$|uKx+<+<5KrM+Y8|iSFvO7i~uH1U&qLL z)n{&E@K)Mx6iLys-6Z(5kk<6|l80>)=jro{xb$~)Z$u@l--+eUy9<)scN;2m$fanj zSc-V{*F@}}2X|x2cW!c5IG1ItPxVJP(!yU?x$SCY+}eNay=pH@x1vCs`>jdhv?EuM zf0f%+n#O6aDC3^{nzPLNSae^ly1X* z-5Sm9UNwpHm^z@luRWzy6NPrU-Y5vfr;L}(VU1&Ah+hyEAGE< zv&rC576~-ias2|Z+~_08oa?eX+}xe|q!(#VC4SdA&8XL$iD?A6Re$4ds*a?vb`L7f zW8I9kIn;LkBz^H;Ko$R{lG=O1*M%mWsH_AH4YI$Nx0;gFPgCly0n&^mu4&^|Zg}n_ z@~Bbdwks8K?TXP9r#_Y@1|MdA-eZ(^eJ+<39mUlub#lK)b#luC8abV|Mf7OQIMzdS zoAVq5(XJw{D^h?Q^H`VL{3NcA*E)sXu6AMGO`SuwGfsEDWwzg-1#ssXh5jG|~X#wkr^yWP!Wx z*Qm_$82PszV&|p?PCNFJNX-e-9%#lrvuy~---J62V;~wYh(|rr5Diqt@e*0)&trGP ziH9)SWG_sDijnYfA~wxlf~HnO3^;k9_Kzu;br+tgEdR;6h#j>!RMw=BXhk@!nzfdc z4YotvIufJ9ZekzvWsTn&hi-WfHedDA_Qj2)(4z<;`@Q(lvWaDwwJMIXa zd(z&7y4wpTKaLMWo67_#;1iR;O3YNV^cQ&p7=l>BT7lU%^Afn z<6taShLiEWG{z7- zHhTeXy!64e&V2TcDn+)40f{fkCRT8S3_pv}zRE~`~cucl|wkj!pq&YFYESD7DM za3;U{gb4qs?OGS*ssAQ~+fBG5_}9}4BLcil@0K4gxewYFGW%eYsMosn8%g4-JA zaDFI`7ws}gT_Om9kFIDKs)Fv72e9d8J>5f@xIUr?3%uR2G|m$O-fW)~ScXZ==c8PC z0$BxK_@(3l(;;urp1a6C+JiI9D=S*Khu&&A((MN!)FwLxZg$pKyleq-cOQh^GD#TD z`bU1x)?wqeJ=p3w0lQg;beG6ZR9<18n_2Og$z`DOUnu$;b?j^4Q)rFa9=hR z->xsgmlaIAmzqglj0aMwewe)aW9Z)VeWWpvh<8mlA)a&?{Xj*D9o|Pckv`xZn1?-Zzh(Gy-Jm4XQrsp(=94~mQOo9bl{}11jpl6qts3ZW`;j$=soL4%{_>u%tAbh(S`2SgK$+lg)br} zptMI7NomYC{j`OWf`YKVB_1P>ox>N^ESmdnF+E>tO5T?$uq3hy>n|l^VTc1J91exn zf*CM#uO;K{rIazggre(nU}~0+uA1A}?Ryw!v^Qg>6>H$KZlez($_Vcsg?Vq;o=!!I zdDJ-c=(9}hA=a6<#txc&no#W2hj4`euKF8dsks^^r3k~+;V^2pXD~isBJ7{B%=5(v zT+ms{I#^~>N|ry>Fn!48)>KHGk;QETSl)5IN%&qm5r$Uf)Z=eXqCGNn>9GzSk4-_&&b@HRJb_y&JX|aU;lX^oIk8F@ z%+*G2i8M-TCE@dD6h!v^p)HeZNJ9TBjbE&Za())lFD=K3@&F9|Vi}PsGZC|39^!sK zr>HOi=)LYI)4CXFvAlLGUkB~03!vp|2EEQe?04Bm8JUO3;pTpdd#ZxfvcrrIEP?7P zf9UDE$&eV3#D9ZEnCE1IHwjbW{izZnI#qbI`3&2yOXJnCQxrH^ob`O?W4!2L^f(&w(A|HQLrYA~KQv z-#$Jg0fBllsJlY4{xWPY(@%Rpw9=Ggte5EK0W|I@!FJYTW5Rrv8 zj82hUYBv>W4^ZRp*VI&Kijwy$A!uNajsipETkGJwpC*1~jX-Mk4f_0C65__cXztWI zG&%o0&H6JLUpLJ_cZ)5IezH9NTnX%9?}yR-xfH&gb%$S!qe6ilD3?2lN5N1{>@$8jwVPDj$6?yRLE3WS9;x|Gg_n^r=Jrm-&B6?v7RbSt-WsIrGv;Oglj2`* zAH$zKaDe_YCXMoX#+zmx8E0 zi~741ew$_OtIH6v~s{0Q!ApTdoVX69Wxf`5U6yye0dsIw`?_1y+2 z>aarfumF~a#A1Z=SxigxLw3h|QcuxCyY(Pl?|w_s!=~_i?+2gk49Z_~g3hyUl4ZJU z@keeE0(_Z&D_otQG&GL)UHt)WO#iCt7>yrKN8_VRExexBv(ABIu+Pt<4sU<@8+DKF z}=9>~m&SE>Xm3<`H)I`7RV`xP49n!hBpVDt8Kqq!L zUf(*5;UjOzAk_{ox1@2TMw;KYMuX3M_ZuE8d!)MTHK{-NPG9cVq9uGR@233>)fd;| z-|R!!;hv9c!KRGeI{^vnr?Tfhk$F=H84@@ z#DSHskkR!6w>0n5ai$S(bA3pkryYiL!$ao7s>5)MDZediKJSt?fxo+V3Gb}8oPVgR z%eP$gW*Rul-1t->G(Q!I4e8kas2HV8OW4OaC1%#dTVcB(Pm*B zzNmq!qHc0zS(x^Yb4U#@!qTC9d{)f|b;rSHc`>HVPk{#G0gc@HfNBr_AU;1T+4wqcQ(VR+o?3SBspgO>pl&iDQzQXqH=y%`eQb@@g#!|CYs-?kZ}@ zEM=WFHFV1=gSu8FL*AWve)jK0$F&tG-P;7Ki{VvGbGy|Ko5Z0)JHDVt63-CjUXPvnF5R^$GT#YFPSO5MF-`@j;&TDA>Hm zY2n*gHD?UJ^M)edp7s|N8?RE1h61jiJ4edCBcc6Q4wHu%r+*O#u?v9+y&sMa$v0@p zn9i48Z^p^DF7PvTz+9&7eAR76MRO|alIeriPNoCY5S+)X#f_*@yy+W3{+Wpq-*mtg zJ0Gt>+w#2#w`(WGbra$BRst_gMR`M6Wqu0FApCTGhm4X=EYp$Uo%^leeQ-K#zrn5to0~9@SG(7 z;FB`{*lRRDt<-=|H`e04mug|F%3LfxGannae1rIfkGSqwfuiYMY*#lL4vf=~c`h2y z9yr2J-Ujt`v#~~)@rPfC;y&(?Vx=GszL3Y^(tY%k&m*-R7bxbT1#src?N4-w2;kg_W|paYO|lQCU~ZZp4l*BOh5AYoZ%m~BE=5p|C-aK2cV(pVWc>Wq zMzYoTdeq;^)w&@^nn-=c87sV=lag10ciTF0g zdYzd^gJVi*NnHhn+^L~`7tTSj1?j{k>z0Qt1&lw3Li32k>`fRdGqekyzDHVZ%vuPr?>0!6lBFWC+hMt zo*4+e8-c*;FpT?eADs<2OrZtGi4Gs&xB8^;F=>Uo<)LacuRDgsRY#zl^BgthM{wZDf4y@2EYv2sQz35ZV$BeLjuy*SpZ)zE;|KVJqeIU7)|W_t5sk zX6XHFhBMyMESp+MCXDsus@Y7Z1apwM^C&bFjzDPpL5fwVrL-4GjFDGJ4+DBSJ#&Y{ImJ^J?EzeAhU78Kc z92=TAZzZ{{Fr>qRz zW8KUT^fuc8k%roMZ6|<{+k08w;0rxE{0r`*x{wn02FYXH_>duIo}apc=Vi}P#+M`X z{#G%y4F=TDP=k7?x3XLPyi6*(u#V^gjUWR4C}Xo@Q`&Rg;Uibnj2J+t{^Dbx5H z-zM{8l_v9#p8~h)E$}PZ2dN*QvF>G-vrXTD+ciA0{(0l@UR|7xH%0Cz5gaI1!jym= zxWDNaSvu&VW@{=MA3H(Pcr4=TWg*G-?TIq4D8OADCgpXcc_*FJG>>7JAB%mV!RX8K zN6?S|adh5cJ-%%iZ>f|bB}GN4l!lZ%*LkNzQku$0l1d^HQ9?pfLrF^mrHK$_g-S`X z_m=umNcKz;?|r=g@sHyu&-eS>_jR4;=Q9+sOI7e`oH3eoe^ZEJ8lBzWN|tuqqh)o7 zekC~}dVwhBN%I}#YHPgL?xCvoHhS*#1U~|9B3G&usWXPKhZR!nTI~T0=HKlR{okm+ z{4F*3<-#gq5$~)Lw*MG`Idx*Vo!Cx(BSvHTolnG;JtdpUX3{!TO)sUoX@2%IY*w#; zVNVejsq?dTcMzV}D`D%pb|mdBfZQo1e3>>9cB@)xo847|aBb7RS?xfW!Jm0WP8nFvU!}!`u zDBM4cx8j>{W@bF@NZKIqvO9XEkK@mrbY!*eMEc4${@$&E%IjoUDAwb^m`vR6or2{~ zI}!S}5`TVb!y!8gx&usbicHb@DgzE@!yp|#7SCTWcCY6-7X1@r&VA01JzyHFo}V3!<|7fCFM`WIpJQaKom%B$m?*5f)*TY8NQefX@AXK{x` zh(m7VM7G>(02|c#0GBUF^Rr$J_d#66?7xm^br}iG{grfgvnHZhI|a>@z^CV8d{7v@i-7P16CY?tR<>nnfhUg=kg9Lqhf z_w$i9HwWX!bmGK>F6@-#GaJV)=x^VNJG?h%uN;Ss37c@WRTbTJJ*d3Od)#%)kTLuu z4tg!X?&*H`v(FCB&7z1qppIhuFJ%3*j1HWypc>wl{#}xR2yqEWo^9fL&8;|oN*$l? zxTAcFIkq%3QTL74@S5@ib4Hd!wOkLcM({me)dfUk z?&YGGsy`ATOH{CWSSrT+7lZ!}sUo?^o^`yKz-FYHvq8y~=rhfPNS-gedDmNGe;R^W zDh7u|;BI^zb|#3yCz#>G&6!v^l{+{lmt)Z>XRI>kS-JW@c+qhffv-2Ab3+0y;5UWs zn2guA)Udlr12b42m3u#fv$d`vnF zH5CQCJ!FQC4ZI(;=`!6qI*5JS`W4Yfze26?22w{}K;Ohd#BlFkoK`em=odjF(g{6B z{4sJt=X}FF}maYR)(dKs)bsV8s*s(-YzDIG&?Y zI?P>F2k~#sDtN7$g~s+Zn5rC)zAx!m`(qCJ73N^P_bTr9SO;T8eHeeR$i&^nVe2+bQuIOd&t=g6u^L&?Hh5v-43E2E zoSkQZFjLNTv9QAd-p6y7Nr1IuE&puDv&MUpEUe-ktRv<^&&3ao^8;~W^-C!2=|Q&E z2blQt{MF#!Sa3;#)rw2A-giUThnSxjWH5%=f7D`H|MXe9@>)bhC*t?F7~t3~G++6O zRO2>Cx~^tL--6kZ5(ifCm%GKAc48Lqx~CM$F@O4t%%j8Dwrf{8)3gJHH{W8+a9tKN z4Qyw;I=lZQ7;X2mP;0&j$45Vftqtd|X*@vLvCsTF^$|2aDKL&KLgE~4%!<^-r%Ar3 z^4N;x73*=W{3m9gyNH{UZbPzS3;rfX<5Xo9@}_J|)dnErFOuIxIQI zcDe61%+n3zJOKyTZ}LFNtED)*q7oja=dg`uaV+vYndCz&rtn}6Gm`AY)r}&|M~Y{s zelNqxAl?m=j>NnX6QH$-cWr)|VoG!}*r&xve4B->w>U?1rU#Z^UWwE0qRi6iHbxBo zj^5L}+cc>Ib)A2Z`Y;xOvHmdMl?mmDy*zVu7lnUM;cDhe)N`hYjMj1tOYK8WsR(1O ztr*jx%r3-?VlBBk%<)4kvPav)#&Huioa@065fPRa{g&@4>fzbOUAXcWp!)1KfBq%e zym3R=&C`|WGO31|WCL!RcJa(u9}pusG+-mYf9-&`(GWIMm?43j;ZUNpq}#_(|4xe0C(b6+#8GI zK{HU&zJPnMy?9P?6JE%zJPegS-}OoD+|) zw>T@&F&^{dN}-#dgWe0LcrREQ-);%$zF>|6=Y}JEuO_mVIpEe46-1|7q1}Vue(TJ=B>Hz1V5>$Z#9E_ZZLW-oN}PMkvuuV7+|l3bgqGi4XtvG9+onw9z0QX2 z_5J90+zQR6!??fE7#A*Dpy$6th)Wft<9ZT8?GIz6+;HYUYZ$X!XU4jFhO+2WW0=RZ zQLOp>Pux4Az3kbqZG^=x4?%^OGBLlDGsPx&|U(c^Ip_l9Jk*vJJ( z3zovU!xOb1w_w-IG;H@y;r-BaP&|GP2czpaFEtC6BZ{%OZ6EK4`6JzNA^Q1UV8Tjw ze7QOYyG}-;yULUuuUO9ZRt2+jqx&(0^DCu3iLjX+;Rw%9h3neweBWaWB?Zn2339@* zU<0)IjDgx~FRUAzf#22q|0W;@+Bb_h>$Vi%XI#P_$2j=yKZ@-;bI`HE5q)L$*nDLg z4zI32E!&QN`=in1F$$()d|$R;EK&@zFm`w;swIzN@AwYPKg9i=b&s%{`=A_-xM0Ie zQEWIY%DnjdVEGPFW*}FIA^+8(V%KeK3Le73!;DzBt}65X9f!}cDR7ySg3mHi?L(yIKN{R9hme2?oauZVH%7d~>IKy}srVM} z7xlt@-hEt?O~MurU--}1f(yP^(D>^;!aFWO#+7qd{%*$!&m>s&X|g>pf8o{pUT6-A zz)gO(3?6*|rYlxKw0H)d>bfEF`vWBB-hh(&P3+jAz$|h`Fr$+B?8t!>Ucaf<11wVGHLtv-mgmY$)faO%B>i%X<6i!tjxhUYh~e=5KuTx_IW#%Izlpoe|2X4#k0fk4`}mvmL$YmoM^j=Ck&mM?rYAq71@H6m zUhf=A=bXXxi3@R=A`rA=G3GabeWD@sujeOkUI7a%GsF+=;h*@2e0|fl&E9 zaNsk5F=z7e?(JE=r<7;Q#u&2FiGQKFdl(z3uf{Z1#9;jD94wr&1Ks`K>1Bc>Lg$ab zzJ%BGa+f%kPrpeoR=uL`@JBS{$3sf|(?`4Q|4_`*&*aCS{n^w1K*U0v6;4@?3kRq2 zgjoXi2PraB)yZtz?jh{+IcY56-rxI$LlHQ08TT@}V7l=R*zL1q8#RpCRGT5}Yv2y- znZ~o}qE%>B;90K~m#NwK3=NyX8TlqdP?q{$1 z3*$$Xpj_4hxApW;xcxPmD2;*5L}O+lX~&8tSh3Hl3T#lHI$N_~IBS|{%dU?LV($(z zRw=(5E+3+Br@9*hr7btSLeRWGaPr z#ak5Mno7>Uj#ACkdvxrf7$y!h$GV{g*qA1cWh=kZN{dn0n|PI+6PjozXQJquu7qO7 zEbM%+4Sta_Y*LLf3tc*l-FSKoy7tW&XT|v}7F&@Q6bbRp0&Fb!N@HVQ)30|5_+VVf zY{uua(Wz_M#7kgXOZ8aQNj;XiWCh&$E^Hs~+TT={LT70wE&l$LJ6R{9|G<5k=Kq8) zyxxPg1&vtpWjplCW@B~TEZCV(LZ0h$n%jAcq~&#S?XD*5OLVb3YAj^)7NMwhH%jWZ z!}n|pg&A7oK$JQ%ILk_l-$PDqnTqeXT-liAasQ6Q&|1lrD*yR3Jv2G+(BT%yvEtFv%}n3)Tdo=8_tz<-xdGiDeYoH%%?gy}utT~!>~{EWxXvhr zmO&bFcJs5(Y)gJNiGtCA7{~`N!k`jo{@ljHnman~K262wzx-$U`vU8EH~UA}bDrfD zWtXo#!7a-x*!Q)SyA%^)kX=M_wzDX@?HP@IvPqs$}Vil|H33?xl$0D!Rp)Sf`fqJ#ma3;CxqX&!VVEwq*(za3mnM2PiPP|VFNKu_W5}T45XoEA(9O>6v{m>*m-%kw z2hRw<<{Yk3L5FeTdj%%)_iu}zI%N4;^`Yq#nvb94H}y01lus4r%~KU-M+69V;bOGj zbRfCJijzpvB%1FhLm#tz>Gr#^WbQPHa)&FDnoc=qa>`-E&J*WDbkNoN}V;8 ztY1%RD^3ai=Zb}2B8LRUOPe`+I}_b{M-jEP4l~XbVd~%#tf@;!)B8jyY8Bwtj12PF z@RbG*lfc7G5Y`@x5L1~F$&LMP$~7Kd>cX2UQ(H}9m4J67UiTpFJAbCy%Wak5jdpwFki z(>dL{f~WBz;r6PL)VFRw*|e3A)q|b%Y`m?|)RZa&@GPWge+qdO#L?1(d>Xw;5+~=} zp$ny5)V1Ce^(MO5=V}d`LOJBD7Qw7-2GIVtgpS(JC-eA;v@*qwzNhm(;JrvvQvL+# z6a0Ss{vGEjIw9{!81gnm;BQ(rxy0WiRfX?lD^e?%MOjgprYt3wHwxW4_l20dLumo` zWyggN#b>1f$h*yiYwmLeX1Y*V8zfD>{l|shT@y%ZSFh0b$CN}PN6=;sZ6P*5Tevh^ zTTuEVPkp#4R9?R(_~vN}bMl;oS<@y6*TbIDtab7@x%d_xS2-wHef=d&X)q+sLm7~9 zkAh)P0uuWO`Hf<5m9@d>oG8R_Mw*>s9%p*%2tm4&g%LyT1W#53C7Bv%Z%IXjdmuFP z!|-W>8=Q6+A)9Y^_0mnS)Tvoe7FQ>g<2LloezcIfcADV$!&1-~<&0=qcl^0(iuX2q z=xE#}(#^a__G={Id}T2Hd}$=j3hv83JP<7U9R;;VbMJjFG8&iR@_o_&=X|#GDB@dm zieS5BwJ;=Pm*7^aiNY}x@vS%ln@@4abiXR^sVZUarUK4F7v;NyofKgnAjAd-2_c;A zZ9KYzW?d_w*X!!(?Ii+RHN9*?WUh-1*LZ`)Hnfbr%Yv*9uXN_EaRlmvk1}(`&vf4}N+OcPvjshGzzB z_`7K)e~%PQc3`~|X0WdNGui%g;wW11o|Xq_W757>vbcDg7L9yI+c?WwX54c!GOD2G zt{LQ67DF@c-=m3rbtGGv%lWAPsOsq&^4hCMr^0h+4rd<~`3;~I=eJWF?_MV!4X5Xc zJTsX1k~-V}(s}Jm^wl(yKAFYSCsBEv5LrXK8b%WcBzF?$)05D-P=o+-xSd>b39fVZKQ;) zH6*pvn}plxv^r}YO{q_(w6n%^^l37!=Ssrg_Xo+epVI3ibJ+oJ!A{cPh5t|*<}>IB!{-iIMZp) z5ZY;CO&cZ`lU&VIs{S>J+FRVIJIaKN&BE!K3jf_j+hWSK0murzMkiY?(-L!@S$lPu z3cFgVU=n9TOccSq?E-ff*P&cYfj=wP=yCjAQg1VY!s)GWO7X$y_!12JR87zM?sJ_V z&*lo>C_0V1#T67_$IrTYTGQ~)#|qj<^=ae!`LxQzifRJIvAQjbR!vpnKAOpR{?m~A zrJqn?kR~keN#Mj!4^&>_Y+K(0WNuSPt!l%tVVOT7=Y-NO?*X_T`*u}#3#;2Y9n2#HIzyFNyBM9U2J6J@gk3WE<4f1K;D^t zl7dxc*Xi=ze3Ge@gs$rq62HzJ@|OSTx8G=7(%VCtlS?UP(jlt;;RDGkbtDD%;=dL# z{AcZkb3c`Z1K)~Bcf=XG{pk}u+SSaxhOso%hx3lRm(#ziX7qb>G{rp2B8hjolazVK zt6-7|?uqkr+jn(XMR6AFd`HYW2Q)es(m3lK6k!!jQs2Mw><`c6a-P{%#|%1jqmr`k zI?-D9ZxsGT0~^1M#UV}Z)Cnww9M3 z$QW+Kj=w>euP%#p&0_THz2MLOMLPJXlN1-t73PYQa{Fb z?%L5P;xn>y|2b1=?+u~6Ig$E&$ML(tZrXoL3C3S}{^IxzdQfP9wNk0{E&3R({(ZPoVw&KBT2?QSQUt)D(V=!i>LCz}$n>8}WylUpCT`i|@(m zjUDz|bC%VVeA1Z5nN{NXbYaswK_}%Z=a5Sxx5E-aaKO_Ecd6r$B~BgKP4mysCeyaN z`8CA?#H zRoNd}FO>K@&KZ;K6L6%9=Z5%s?coAjL>8`s?6Elz?Q%h0kq$zCj>q0zZrDC;1~z2O zMdsxV2wob1#u`P0ez3=qPgAiiOBOP{vfP>Vjn*V4!E;_Twl*z=XLKOiifr*#-5Ir} z`{2(#Hg`%qP`%m%QQsWUJ3YZF&7;gJiD@mXNRY@Xf#kdZS!n ztYQdjgUNih#Cb!KH3)Y(gp#p`VEWtwPXg_c6X6UzQD#S`cVe#DVK`d8pdlJkJm1wt zrEv++D|UyQkc=Hu+CX_7SbFLvOs(cYDa8e?doplY7zK?b`WWzYINE+cr!nhAuy9}< z8K+J{sEi_7bVlRc<`}-?3c%8`Whg8T#8|DBm?gaq3g2UJp?W2fToU*UPlh>Gi!pbf zuW%dq9xwM4qUg#i+`BY{ITT5;w`Yg2_$%vSK7;#l@44enQ6U08r(%tp7ZwfghGYB{ z$R!M7N5)Tps;e#%5>;WofzM7iN5J;ub_^-$hFMe>o~Q5^g~v-6y!wa>-sd=HJRA=v z_(Lr*3LytBWAcedD1Y=5Yk$RIN}(OXqdbw3T1wP=pXB22&|jVfeXy2$-AD~do3e0k zStv(`2k;pY-+{>fq|@)fnW9RNvieEaD@NkQ>^%G}D8h(G7qP@Q1}+z(V76}~BDc8W zfVc?;)R|*ox)0w?=p%IUY%DepVv=GW%tFnU<;k`pJM<}PJ$@i|`+Vp}&V!lraNft$ zL-3v{2ynN+y1+?LaWzA6g$BfztUPv3$9eWMpr=LaR_j}lX{x?$EUZCDboGobBh#|5|a5r%+ zVs8&*-_BpgP4vN8@&Gp1MPRb(UZBhm?jsjq=9L)eyvw22zxPnSL?yk-u0p|p?I`w1 z#;EO_$92mZ&kir=zCbO=Br9SM@A8z@e!+tizwx>M7o5on-oLcr@h%h_cStiI2Qk*< zI*2JA^g%~mBI;HJ!oq(vHq00a*K}*ln7oDO4h!*BBpx#LPTU3O#Eu90FzK^PP;!HF zw?=Nk#{4)G$p>;SjT_(Z%|Y2PA5_S@Vb~!Tyjed3ITlX%@BK>nd9H)p!Q&X9%)Q(@ zpI~}tB~CXbq5E?oOSGF1dMpZuK9(aS{v?*I^1xj6f#{y=fU-F; zSbxbM`CGUnU1JQI4@+Xo_D&L4zD+R)#L%A9O$PH~XwV^NGV~9j%|{fmXnrpVd6Ti< z;|(SpzkvRw+!b+k?-Ws-c;E z3+(Z^@q`p-E(Pw73+%G@Ok z?xlpv0X(xjfGIa#g*x|{SW7&{(=EHPC+HDwSoCpja3u_f_F!XcIxdV?{I;Erok^Epp55zG4xnWE=Vw(N8at3A9D4y{KJFpKv>`2A2dI~%?`&cS<< z7gUV)0?P`pY5t9&%oGu z8gN_ShT^zu*qWJ&347(3J9k;`G$_GZk823@ZlsFYEu`_MiYyk{BDLKFBa&Sau<9&* zQ+q-;46^7R-)TM!D8~=|Ug!;1X0u0*V;ejtv1vQPsBh?c?k^6b*nzElpK=0S-rDT5 z>sFX1aOag5=O(y6g!+ubSn@dy4OPmRHgzaAx(-Lh;xycsJdPO4_sH+Nfx^(M@M(FD z8#lzzo9vDrQ%_XjBb}4JOwZ~r^N#5@Os?9AOK(;pvGFn<$~WS0vpkyqf=aDEER?sv zkPX8iECu5DbJSi=*rXSXpguXIrjJM3J`2qAy-iIm71V5ZmfA$VQscXuG(blJ+jdVw zaPJ`GA5}v6%y*O-{Fs(K5&^Eq)8RooXr^rz$=Y<1^PYB6c*&ip|4i`q$|P7{nT}uF zw|Ro!UpAM1qq~;JX^=?=ZGCi!q@pFTPelT|r#vEgk;n8~{Wa}sEucY@eUSUX4&MrS zN6e-Y0Vj4rwqZLSHXfqyKCKj+|CIF1=i|D;d|aFB2RXZ+q|9d~MYGRQGe2*io>f50 zmffcKCN^)P2Wz~NAt9*gfF5e-u2y5Cr zcoyxI>4o zq=@E=vZ(m4oi=>AOWTgTrrV|W>CK}PbX)ZjrfaltpX4Dv3txZ}bGPE{q6`FvRZ^Gt zLDI=SNE3#hp#!ST6xGH%3UbwSvaO!BbX}s;4G+osFyE_)v{Iz;IE*j#!SnZ=61NW zhWiVx%klV30rvayJ5kbSTA1~SGt4xRK0yy(+{VH!%oaaS>*DeGxmcI94R5t$uM$y~F}7Qv>kjJnz)DaQ9454}DsmL%F|F=-8h;dY{^Y1APaeP#cXg_LZdd>?m3A zzVAL&Kh*W?#Cy}-XdWPiV-*s3tow?tn!lut>>U;R+@!E>OP*aGjnLvLxTK`cw#W5A zS8E`1&>M%7#)_~K9foq%zZerHhT-2VLqtxfp=;ZZ(bIEhNuh}|FH^)Ya09J2SpmUiI*xE|)t{S#aN^%U#I>m*arhDpI2{5q^1w$GJhEoX__H8Jen-jxlR;xizf>Dg_gpF zq@hB|)HWf&ezcc1J4^Wy zq)bCdkx&p?|1{FunUXM@)k3x}@6)fXe<`p)5jA%>lQ`Rt3@)Zqu22Qp@Qp~@&a*Sp zm&nPihQ^yz(0#{Ta+Eqk|LusQ^HEC#AF)k>p~^-fw05>|;VlWF*3QEF^QwZ;>`B63 z5p$vNpFSM;`PGS^fqh1w5+1)Q7bKtU7j&PC;zjuc$ZU1N*|PmY`s5hFPj9}kevu?e zoKvHx+E&yzuN3}!mSN4rT!=jBqW!yX(x2$FG`iYP&>uEgxIM%|7`#PCSX1UL7>N`M zgRXxTeoI#i2V2IFZfy(IE_y&dHMunM@pB5ecb0ONcaTDV0`JQ2rdPjC(ndaON*Xy1 zbvz#;?_&ph?thv;SsyP>)k5KLI^z2>d2hyrKASHj+cWxfV*Lb4)}2oJcNA$)pa(36 z58&S~f9cK4ZsFBCNjm&yG4*{Yq>Jk_NU=PFCRE2_8|Siw9^^YY(TVU((}G9T89M*9 zg5K+kLXhBInZhY3<1>QZut5m_;74mE*U)0;5E}9Kim+jQvJiVHUx?GG6>8p`5@y-# z5vK8OZ26E-Drra}_r6iWF$aCl7nmgMpTMZ=>r3Hr>kr}DH5-&YVRhIJ#Ljd;iJ={- z)Va`qzo*a-I}72@(G7y>__>1WDK)|7`B*{Z*7r$4SuZ$8@;3Q6)syibH@d93m26W} zNqfy%vV0d#vSU+8RDZniL|s96^4C;YoO@Mxa^jBQXZ%#qI&C4Gij@}PM(GKUwN6lO z%_fqbFbYZ&65rky{_J0gjwN0gejyW5hwTX&#4&KcY{rbz2``-C`616s*>x@B7aNa63% zn}*34dp?V%1Sin`z*=(cEP?m^QfxI|it{!D5PPz~Rqv1_KwAI6o{EDaY zGp&>0y)sf*KRaC5WS>GCUwoi(XC&b1v6`&a^GG@*mjaiB&=dax+O~T(eVF1+)mgi! zDI|}k`$uEhovjFdy@q?~Rp^1Qlq>Pn&4m+ZQxBNt9}n?g(?%zZ6>U7-EdcFf^*U;?iOboL#1diTVRjqxJ~hhE@1C z{2;#|4X3M!_~BWkPdSH%0Q<31Koa{3GEj#kT8$tXwGSnImwxJF^-rh7lXareZ7)1 zu=blLqUP#Ww2S%Uvv&?^gY2ML?}0I%*(lz%3kp{oV0@>DWVgPiwxTdC}VA=E=H25md71ak|!FqmoKm#m4S0hLBHbkf9K}94D z8_Ji$R#gikY5Mrqp#*E5%Mk(-Vaq+n+FtSaUFU)8)*)~Y45A0C-D#};G@7Gzi8i^t zqkN@(wCE9Gm6tX~-8VyqZz}a(;++=%U3B!V7}QJ7Q8;zbJ)Yqy4&9BEib7mkZUe(d zVz50q7CGZ?P;jvlWZFfbH18GfdNfmt3_s5oPGti_#aYlNMOHev0qLVJK|A6sk{3Lp zt7kGOPiH5|{vJeU)s?6|)t0miKMFO+q-c!P17T0E5$|vxBfC9H_}QRd*|NKu{(e|DW zdP+m0^Dg(-R$zBP1E1;j3i6{K3$`LMG(R&6mX?cfI>H3)-KDr6e-5!98ekum1i$5n zVJQ~R-+@nr^fNz&xU&j0dSfoG%5j%bR1Wsf&cdS8IVe@FSF$P_3E<-^cd2qjit2I7y-K zNJXKM#P%k0{;VuZ{N9W;iG!JMrx~PY48}}%U5G7d74-Zr2rW9Iv}8jps=bo%b$TR} zqldD+6%$yVH+ShSUy94SdC%P-n`c8GQAU|4Vv;8#`j##Lp+BeED=WZ5?@9->i%3wMzT`%08CP$C03n`8j((kAg`cNuKm3O2m zqizz7U!w&f%nVn)D4?|MG0xu2LGtZFxSMW<+M6`=?y{!I3`;1nbUNhGb<{l z!3sKTwII&g)+jTB{n_|4exG(fo=8ZJ@FJ)e0DNw*;(qaxka8|24H)u3zkkiNfWp1!g`5# zgl|rU=Iqrd@C}7;wGRxY#^Bn`qqx^qhNtTXb4T3@^ki+s;Y?3vTxP=-U(#jAw&}26 zr=6JGPp}xx=P0Iu%=Ee((;Kq~&-@b+``=-l92kvW(+jZXF8D1^otbn(CTo7C#603p zW8|Y;%roTos!U7h^6w)p{{494;y-wXfH{cTF=MeAEPLf_7NJXQt&i|Cc8lK@B{hD(>PUwRk*JJtyRhIf%iv{Ku^8HFDk`Hww;PyiNlM05} z@pN?EZKuCwyyx*yf%|ALGm|st*{J+8?6r|Gi|r0%8{V6;e$KJD{xlb>mn1<&g=aPj zualTX8D;P6p>1+vkZ$=&6XJRy`QSB_=87?c@bjEKP=dSLj^S|I9E2>2#og@X$jbaf zn?6bL@4z1F*&he{x&v_Fe3PxSAJAG$NthZHBltyPZA?OIg-{#>4?=JNQZeiD6 zuVdeSMzW*xAK{VcE7y4V7={@3ZTadG0k87YguEwt+|gQXEQ5#kJ*?$l-s^l=i@%hhf7*E1 zn_NKkcgutw3hlxMo~;?g-*00+y5I-T+ixDelD-C8lHm{uc1}Dj5x3dmK8BHKc9?H8?+&iscM>X!_FhuQ?Rol4;EeOK5Ilef#ECU0B zOSn6D1Af+W29nuU*zldP&59UYE#@qf@3(M6Dhg9PYhhbugGqlqq5doof2>1D-F6`* zM+x|t=}KSJXHxF(!H}sHC8uy7l1X_?lheEqW|4{7{bk7Tw5J98{OHDiDU_HY36B$w za2ZmKqM;K|bdLM5G>0Q?@JQO>96=_rXJ}BwSf#U)QiO9ft8)dep7)~> z|2)a|%12r=^*o73Z=x|?;+R#eM*XME=u%T01!&Kpz}>a9yJ-UF5XsY{d78BSSr@gR zF^8eE0yHw$VUwR6e%1quGO4&fdncw^X|U>W-Z97?K~IlOA-i42DEHYa98dH@dT$ak zzK)_W)ge?@u$&HL&Em85oxI1lo->vO>Qrzcw!s9t-*%Ahy3eF+z&-1nIsLCT9SdrU zU@bL~PN=nbzD>=Z`{fp^~u!*}XZb%YcR zn>m>By*nvPO$x8PVkt=L83`YwD0k2v`p?#oTpsz*>_^Xq3fFpyw#lc6(2+Dk^MfE} zI)HW!PZS2fP@zhG4>VnqD^N#<5XsU68_`R`_z=z={IHGeAEuM4mLv6hs8QF?H^R-` zCJ0f{LifFWlv=e{DEeL`%!^A9PCPwGO=G=jfsGz5Yg$SlpAO`#x^Ce{;4pgP**i(p zRECVbmI>{iQNpmdn}ufCCjcUFO!h~xLbh$l+nl7xN?OI=jMSo6F{V?u<|8ib1t2`lS*}kWRZ3=Wh za}{;I&!uheeQEKEAABh8eR5WQ+NR$2Fxk7y7R^eUE9SV8dL6%G2(e2@1g^C{&grO6Qg`}}XwB(a7|Ll&Z zwd?u6FIPnwd)HIgJSo}|=ql*UUnq>^EV$&b9%1AIH>!MUf`CC4xV7;O(H~8kdh@i9 zq&$Z<1l7_USwC|7wtzo5gA(4S7|zTTwz5YFK}@VaA0~anSnl)*%u!_+R!ld-#FI(*wWJme z5t2-@w+*T;493~>IpZ=J`2%g?QSFD6WE2q?4ePf3w_?N+ES2@?xGi{>IsK? zdOqF^D*!wF0O}*T6Tw@H1>YV9|3EDqjj>1jnq6>NE5WoiH5u%VF=SpML}ItY&hipGFJFPdK3VoYU=S-DtIYey zGa>$JHoix1Kz>pUY-fGMuASn{YVld5-ce-xw~uC%=AQB|hw; z!J@tGd0#aOCi6BUYUp`fQmDbYtM_nkK^&`^7Q?=s$zkD~m*70H1zyg_5bP$-9@NM( zjj-=Fn#q=|_Qz<>_9`^Gf`IxuM2=Em(=+8+u=!4G zx^M<#u71I3@n(|9GKa`1ZM+ROhGz@!&en1cn9~VZ&i;%mnh&_+d?A#5=Ap7N9n#rW zxR?{k_tiHb-<$~bFPt}6+Ky*!JO{0wiN6!-@b}Jm#LaVrv3@4ZlXB4$%e$&=2jINO zgk90GXB(7@S*F@OEPd037S-`gE7=p%xX=5kL;|ul>f?gbS`__G#9AEz9o!w%tgsPZ z`iElPBz{l!I1BN}-8h)|29FJ%LPzL8_w-gcUmnPEhs8jz-3woaZpR^>KN+6KyT3Pg zVCw6o*tIAU2`-zUFy}K|27N_(-(#N9n#vTeJF*K7o7u%3=15xNhC3_*4Yv;QZs`xC z#F(*XRpodV+=GIf5=`Sj34W#yWE_~cjv7wJk2n7RjAN^)$>3JWIkQGw}P zmSfk%xlcuXBj*2&<8ISJ{L$+qkJSq3!e$VL;KY|HCl}Pyb1AG2Q(RqjU__kqO zG*lYeq|%a-6zaatM^;l=2~CQO_=)V@q?AZYrL?z_2o19$WTnVf5@jSKBI~{1|8*R` zN8g_3zOU;%KcD5=!rU*$!fdk`Y;5v{2+wtVNWTU@k9K5^7$Xdv`__c_5mXG++GD6_a0mAG-ql5BVP%^tIx9qZKGL&MvkUXrD7|jO&G;FImxK~P>HUmwfMdFD!q!7!#Cqi)YAWg z-E5Sg4o7R6Tk?+WUSLcgRy}30-3s)v`6vt7qeA-&6={jwC_2*ZN`Z3+keswC&2mS32DDV=7WLoezKOUx8Z`G9nT;EQe>rLNw)r%> zc{!SSv*YaTxLnGxi=|eJjpT8!k!@Z&iA)m4lGk2a?q_x-`H!Mhz9*O2T>8L9yizCc z?j01AAx^9NYH9Gedn|3fE)6|uL=xN5DE{7dsw~VU6S*>GKKKTE`syu{e`8926KqNA z%X4}wDG$>}#Z3R(INVgzpp8LyS-~lRCWLgb4~sh4#4K6TUsu8wCyP_c!aC*{w1D>7 z520rTW9jPs5>o8t{z3aUWTCd4ye?c|`cE|JXj=s-GtO3(zs4EbJd-w`dz!}E(nO;l zB;AolwQu|A&4Sx>fBtoPdv^nV5B$qqFZ^L&^Ttt2`z&ft8&Buo9;OvWQ;?H191RbZ zuqJd7wI|z;tj@-ONp-*I9jeT{oX z)+>?Mcte_OT13Ipq15;|j&fYwsKezJskRogKVvSkh8u}&uIO!+9Ql?Fh_s>mRl6un z9O>hOaOanDwGksG-@s-OX-$Ju+WA8g~j zudM&~YnG_?m9_e_>H zWISgNn}^Y``r~vp#E*gxaQFJ<5%gxe1+6Rk$%cMwXX9dt#EzG-3k$B%@AKUxW>!Zg z0g;&WRuR8n)w6dI2DEXcD1B&+CfVoH>2z))m1Mc`9kK>Jyx>MLx@TDbbw;(T$C9?2 zAvM03PF|8*DQaafNjAJArLGQ2yjn%K>g{P=g93@}TuhoOr`YRZAJ|;eV{E;OERB#+ zq({poXzZ$BmVL{IjK`>uQQr@Gl_`QRooC2DBb2(1A0|ELm%@FtJ{E0Sf`3CeXa4aGvRyU|U(-dQIZXyx4^`khPz;Xcrns)&MSnkkrRV29 z(7D6i6zUR$jnX?IWX?y}gOl`XWDRxqR}nrB6AFHQ!L~rEOM6j4une{gu=N|c)@dUew)l7^wp58*E9N{aEv@xNTT$v9zO5-L-u~I zXgFz(_Lp|Bv^b3=GRJW~>;gt^GJ=lOXYyJ-6}fVWnCZC_DxWj?%x!=${PSo0z5W-= zK8gzmqJCgcoxITH+ykanfyUATtZ7=p9bSf|T&E+2(}neI4#H{g zF~SRHS3H{Nflc0FQ1mLm-=cI#PdN;UXXo%?a}=(=KY=k9^o3$$W#Q54A%gy(oiLx{ zf|Jtx?7Xxcv9X=d*wl!)pgio?uSP}SY2@Eeqg=6k+L0bd$rD5oR@X{vKX8s)pb-Z6 zM8L>uGiP6LH+$s(y!xe#Dz6FHlJ14}AZM`rSVS*4g^G2lP`nxopGlq&xTE^OdS%Sk zRmPY(FX{KhrRaJx3B~T~&~Y(XSg1Qp2u}4Aa&li{a`ZFA47dlcdCQQq(G;~i7Glr9 zc3S=O12rls;GOeB+I#yE4P7dSUN3)GZ(9tR%WF~lzJhZ6dB*YGTWY%!fWIM52sHP_ zfdCcU%xt6Ek4Iy~){S)aYCKsFy-5FF-6e+*39OzYh0R@hSR8MNy#^}C=zoO;W^dv7 zwi1WCFVH?CS*+Ooie#GaS7;I zq{LZrGwPz(UosKteHNEqUP4!nA;wjXMAsULNsL@Lt^?3e&-n=*!s%}6HUho zQ}aKNx?u$V7^`5LyfVT+ex`RVO5Dln%Fonl&@z~V`E#XkO>roOYVrMUWHFkOj$u9L zAr7Ts!joHrg&Mzc!h)~WbYHcTiaYXYug4@zFyXAIcN#F`bG6-1Ij2?W2E{l{ftaTh z9(4}F?-TFgag@8j52^`zHyU7htrXEW1__elEvR@~i#fM$LdS&rQ0Hx;=GqW?w@;pO zJ-@Nw*Apma*=#c5{lJ6r1=R9y0Ok8cFctYv?BMiyWMQX6RS$xRxaWr6)G(C$vx{XL z+4!w}?1BFo=6EZMTFdn*`&A@4jscBWQBQ-O$k7}BNpSUb1G_aIK5u5RnJ2fi*)LsK zTi9b7@iK{~4=*N#ue`5zE`#KGrtIplvm`krm*THKqn*_UQ0uY<{%vOX{{1=G;4O_= z(nMDa3+RYyGCh`j%G#|g*y062tXndaX_xx4H|>FJ^35>TJTRPjj8bMNgaxc;ktc>biV;1DqKj8&Q0wO1lw{UH zD%oAU)5Lj!YX+gavzugUhvMR{P_{(3m`Oh0#>D*(ve$;YSaDV}8x{S5otoRj{Gz_H zdy=ClX~;p!-r`R~RX0(LMqs>_ZsLb8cRh zc=$L1rdOs@`xMZu^}9(nrH~mqJ!HVQh+o6r*+O~j#_NP$A zyj$$km0xU!#6@-?xe>AY=OOZp-_dmguv70Krms;!_onBx=Is?~-XH<@D+1FU9LH+= z6WC^{7^W`g$FAKCVMc!puzcevRQm7^<_Q;CxN8RebUZ=JCMMC!OeZ?~!j6tA*;Ds| ze3miTiEW!$K$fXX=;K9C+N^kqZCoG8W;C2n%1e5;$)A30=kwFi^GI!x8_!;t(>%pJO!MM;wx=za?ffA|TP9l3 zgN3W8Y6_q6%yTEFbXzLfZBCnd?WmCFBO@kk#Y?^ux>2l)*6qK^W|k5IljU||} zcOASlhoa{6ecBP1Lp{ZpNKb4b8PIq3*Y`GiqqKuoJIy5H+^HmOIh7VQA7Jz)H=6Q`biT? z(y*X(zX-Z2%KO?S+}FM+gHp?4sqjxKnR{-cPoXENNY$EJ>Yt%yPbv&~r}nVNQOs|P z#{H85W`}RbV~@EwJI#=@xh8?@>3FZ_7u7q2AdF{qgAK+aSwR~q+ehKHx;*E~Xu@jD zSX@k#$83)p*sqMjf4p}w>{ui$!)!R?E*jUJqagd=NoaG%@=V^b5AZZX%-jLo@m_+A z`b4~z%|m(U8C;ll2Fv3ju&;TfuvGUx#Hu4Oe3rei?xMb6<~mMz(sK}}kEh|6c`%<* zeZ$K51dM3$N8BwFbQxJd;?)hhtS}FI$Mn%J2{kBm{UQS?b=c}R(0$_p!jRrU!ZrSE zl6K;3d@0Td?evCzy0~DfttXtYxrmmJ(zs=>fWV@MH0jYRq;PJ{#~M8>)esXlNr(v( z9PYrurV@gD5N>8{!IC639LQ8dUF8#!;XjuxH3GM%G*Im3BN$t6fw0@Zs4M#o!lZAY z&>$1F2lv2;&r2+a_(SJn61J%DoX0t5gqx@dGRssJ;qLY^u$Gp@!W2E>cHK=he1DGis5}_;X2a&^7Vg4HM)#LsJa=%w68YIM{?8I+ zjstMcKU64IEE9ema}mM|&4mv8-)Kvd70j2dM$(-P@IR*r^+a2+C(`KXAAKu;>tj;!FS8i8v!gNp?p4%$w&0cgCCwCP-q0pY=CfJi6(Ln{-xU%#;}VK z7sBq8<63Ykx{9X?d-yEt!L@H?)$#S<<-?NTet@nI&VOoFezW9>PF`;G2XMM)%x=$7(6z?TUL? zGyVkJ#R{=)?-sPq2^I9~T?GC7{{-WTLag|(711mQHodQDFJ~2Wa}UPU9ZB5VyB_u< zr{duq1w2|QVB3m`kez*nIua}B$l+R=V|ti^GY-?*hmUD>+C*%w8i{Ec^T40qDC~ZR zY2(}AHNKO|2D~PTznsf%kc0mMc@}GXEoWiHk;LXy>VJKY>{gvbe?lq>76jpL;}uLw z;N6>jUfcu1grNn~1wU0iVcA7JVblEKLZGdskn>*@o?Pa4kp47We9h-}{EqQ2{wVBj zG?P;CO}hH4mInUmpyewb(MVaof17fSO8B1VTS^I~W?2blDI)~sSLMiCCx!(bT~r+P zl7c443C=5&g>Cuzg6N)d+F8;_uCtoRG`J4?uGjGI`Y{N1iT@wQ2rN!iP*E&K-MbV# zzPJfJ;YpljaEuNFRMFx9Cww2Q1;bZPDEN07e~;v%D1pyEee$8$odMZgKI0Wng;v*U zsCy+L!8spiG}AHFGy~_aO%nRd^o9NeX(k?~N6!VTt{%plpG)!Y)DG^*h!yTmUnzXp9Vz5LFG8n!G;-?>;@AB!Bs@O_*MI+i zHtPuWCq;$fA0&lW<)ejl_0EEIgQ3u~vJ$WE9mkpP>oDPvs*rm{MX1(q3 zw_(4{HYD+kZ{4Jy$W&;COTbAuita^Ez*GFOJc2|i-d$^30`2eGczanJyP7qy(mGAJ zbS6tE{v0g~ct2OjjFl5SdNqY@yzAVbyB+c60a!eT&&(R?DIwxAt;1lZ!$M0rC|6mrjj(bXb=AEYWw0?@K;M~16@9Fkc zIea;hK_C4&3z>Hh{GOy@cTWtSe@jAs=@Q}7c5A_Fl&Y|FZY8;XI!Zn#O36-BNAT+S zz*$-UU|nM_?5t4}+V$U}lQUk|+D;NA3g6&MT?I{?{goD+yG6Dld=)!oy0C$>V+&Wl z$5yvKbe((%qn67Ux1tD1>J@OhUP~<>I%wgx9D3a_AFgx$gZM=Ue3<m6Ni@ImE4t$JXdvq(0jzJiZYi9Q7*^)ZTQUR{1G5O)AB;v5{z< zngsFRDJU#TL5f2%rt^Ns-I*$IOO(Q&-k0R-c>_zs9zxx>kUvxB35E?*g}Korthg+O zCJPHRZ5xia=UlMKeHA7wnT_pEcj!iP1!a87plt?cv0mi}CP-Sr;;Xu_`M0JZd1Qvr zIsgmc{wjuI3~!=d`07Pg8j!(+*DDtC>f)W7buE&4IVEU%%A z=LwWJLk73^JfKyBK2u!STe_>BO_O+rKz8RmiV;_!@sH+_;rc{!-ZY1fcaI_ancZ}$ z<}R(wDW%wehjjmMHRVk^Nas7Ou`X*Ej^ENj`9x1@-C{^Cd={!{xRTE{M&nhQ8G4I0 zB6_AHUR#XCT-!KUUY`ZMY;E3oQx_K96B8^;x-l}?4#x3joC~3cm& z&w{99MhFEmJ?>|9Be{HU8nM=dRHQv=(Tcf1E(+&zer$|lnLX`-n6TR@Tpv2=J!4~n=S>PBKN(r6l-i>D&% z$}oKDQl{HeJ~H#+lH{RiMDS;vC5}dG`Qu#I zTwLEVi_Zd-@%Pdae9Drgj~l-{Amfn>#l)VJS>imS52lpLK~gXmPaw~Sg)+9 zKymXiH0`Ybw^GA=a2oG}B|z=>M$9SQf=zK#FwW)zDtXQ{CM^vqZ6y>kypH~rX4Cmj zZ~DA6j3PHqBhT6uR4idYL;c6mZV0yZX-zbRsfAG(V{NCieLM98B%W; zW4#feM( zKh6JcOUH)xQIbLh89jeb`L$2+{mVeX?fySJEAXZGt&FnuW{~u&Korl3#B$@&81v&9 zUYOQG_jnQV2Jns9sR7CJ2Q&B+h}=u6D|6#fl~>;XwRQkdU3yy3ilnRw&~t< zLSvBdDCt|aAOqQXzr$MC8A2FJ!ocqgvs+(swZIL)FrllRf~ zN#m&|QXdaW6w&7Ng?81g6dsB=3sa;g2Ww8C>k4=;-AgSQD`-rfKN<6WW&foNiVi(SdxAgEm*IQSuDuLCOX9Hqq#XBMexrf& zzR^YNJCNg=)%I=&GeG!l1KH_TX zxp{&LZ{49%Z8xO3IACAW0&Muc4XS76p>u{E>{FU4+e;f7W>=^^cPoCx@g8$%IE;># z!u;xMwA>UCKK4F=%m~ibsNo$X^-d}(6~Ty*`{ZGuhZ9#PKp1WhQ~oS-`1K#&G+N`^ zxp4v|{)5J%LBhACgeeiW{H|mLIr(4|xbWY_nKH<%oQXsc?&0%Zio$EZ;i%LJyV+8L z(<(6xP8^1vxj^}#Qj}k~gp~nZ=nn*L8*RZuM}B^v_=t8azeI}KN1pSz!N!H%K-A!^1!`1}_L!+D%x z>(3e51BEx6jeB#X(X&($Q+dC8OWkMQ8U0E!@jCFD8ctIN zbN{Pq5`F7FhTlUvt-_D%XT_#@9tHu1dMB@Ay#;q3bB2z{@E zw)Z|r-RyxAHQaqMQ3uwFA(--81UC=9p^-o3k=*aYd2Y5?8N30{IPa;yZ7}BY*}=&R%@eI2)OkyD=fD1U4CW@HR?9IO!^al!-&3?Kly~11q68H5HHMT)^f_`)K%! zV^m;05GtExFiA`Y>Q81vYsxbk;jV#R>n=K~riG%vybIQEg7{Zf*xwui*O9w0S@$xE z>l?5E?@|3@89gsvN2j-(|2Y`F4Y(nIZQY-sSNQFz5ID4Czu+? zQIzX48cU62Co%b9R4^7%I2NR1M#bAM4rrZNmZGmQBfg^0F9=xevb zCYd{WD&paua}oz$4#54tvM?RbcR7ouAn1-2vfAe3A@7P*ybQqY1!dUWHUfJaxR2tc zBA#z`#uXoM^bcO>?k!DPS=+;w;j>4(l3l4`~MIa~4cxGqZ5M<99IVEn9?#_8#* zs9LVdXT4(ZlQo6g`qv;pVJr3~M~b9SOR~=bqk%-^TF&&mXD< zTVc!j^B?VoV``Hs2DfU#g#Y_kjUt2wwIb7?n4->C(0A)|^kDXHI{iZdkLw4*wD=H8 zH10xaQ6FB3%HyfC7&d=3gs(*y9@^GHTv0^u@d?4HH)}A-EgF~mQy{yz85iEXhuxb~ zSbV(|d9!b$Byulzzx&avwnS>Ht);)K2BIWI28X0mFk5~*bR|V^s_~FinTc!Nk+2O&LaSIap3Ha#&q1O>(Rn3-Bpr~^k%p81 zCBQqu2-BA6;Oh=sy!IW9d%YW=R$&j3^wIS4x)0U1<3PA-r;57ld9X0JLW9dH=`FITRCsG2S)*$_{rOW%3QOX-3u*`0uI(6kkTXCyw?U-i zHd$VOK$MNHZ)7$ ziQcxZq66Dq>F!_`y8px!N}FqG#-8V-!uyj~zr~TJ>=|;O@`=sucvr40sM&d9Kd~}z_@y#+qM4=$u|7tCa+s6GP z+W9d4Re;t#W9W+7bb38wE(JgMLb+R2pmRkX!wyVA&y)9Lu~``wZn=bWFR55t8S36j z7_!?HW(9^=oY+m{7K$Kv9%uM?@@}^9mxgz2M|QP7y128+fBQ-tGSR}mn+7QJms2PfC)?c`oYF>Z$bSkTzxb31m0Pgc82!2zsMG z;Lgp_!uf&Ysl|5~@9ikl>6two7^bNM2hKf2T|1W0DX3 zzHoqEipP*r<559OH$#}eEJbjcvK}*QcHyF74Ng|d;^$QnXhewMzNZ_khmOQ>?ZsHm zy`OV#kH_a<+F*lKNP7ck7tFDu)vw2s`>ECRQgat2>UqI?D(|0m%*Sk1Q?gg%`)YM9 zy6(!E2L3y6$H0I)09-NXP9WxYM_}dl-857>ijrD6Z@hawy|1_9EY&z#)5Gwgd=z|` zGNdGX$)={0^V=_yGw&^H2iKF#ePx&$ex<1gev`z=R$`y^u`yE=U1!BPFX#&m4@)2) z*IN|ySCc{>jUiKO8S;z@;QZE98t5BH^IAsXjGO^vi|o<0dKvj8+@Y4<21;^dpyEMz zlO>A+Q47lI(WPZ0rqiZz@90Hc6VH4;rn!6HQ`wVXvVD4uHr+agSl3Vt(%*@7dqS{F z`v!NA=i@`|Gs+yn`$IN0RBjZ2t=tniL2e;xhHa(VGc#z$uMN~{XF+QZDU-9J83l}& zqjTeoIQLkNKK|tFI;HiTli`LQ(+l*HdoRCL)se@XX2?~?;9YSlbmiUY;W`&uUOs~~ zPIIq%e+`9omy>ht8`!?R21DUA8kO&nXZIi|sGlaeBQE5tJ(0wJ%F@`u5oF7Kt1a$_ zX~8RT?BzYmhrYk)m2x`irvIaRjlI;Wrj3gR`cPG~fXCaRm}#VnUv-mE;1!AAM^58z zW+A%sxVPzW0o-23VTi*3p~P^Y5b69J2P15-c#s8*f@Y&_%@_n(jznm>1N3>f{Zf~r z5Y*C%C0--Yx7m+0 z&RL%*it2`Da$P!$(t2HKS!yVqIsF4KzQ@4l<_~1X1tRZWA$B=0LeE2lXWkrj6eqn$gF* z#+@iD5F@?GdSos=n;t51erfw5%q{+fT1zP{7@jTbP4H><;N zbm=Z!;IkoLGi}V*7{%v1^RTkDkRpu3sq0lXeSO}{-)m3F;$AE@1kM!9jTZ@K;{L)N z8Fju(R>hud{th@e7rPt{ad_r*?Av;UBt$;amdgWh#=(f}PHm&J6Mf0TdOq)SyP#TU z5w`Aa$B-({EZCWYEy+*m+}R;m=%a%E>Jr)=-a`jOztdNq|J^H5Al)W&5+&PtFwfTv5n z(KKf>tiqEpFm(y4IS+h6W;}K9O!?q5Vo*5$i7r10x-;XiRzFfaxRiVK^ZX&P553dS6XXM$~d1%{3a9${C1>=V{-TDyn=eBMdq6 z6VmrZgq$b*e0x#?@d-TVGNTlCpGCuCcp?f4t{@hUsWrG_&g za+2}&MLaz6V-R#E1p2PW;m|8!Q2!hH_okN=1)!$lJp&Ny8KO|j*iW1LM|4%PHvUKvTM9Y+)G!`aC}h_N04`-AOt{7@a8uoS`m zRoS$G?^NXS&tuE6T7*~}f@7~Y%^JUn!ndrU3pL?**IBNj6@aIPF@$utge5fZg*1f3&R@Zw$QWd~YmQPN2)3duv0-%ikp z2c)JVg0*+A(L+sB&K;SIg!!BKbE%5H1@LbEO5Rg2mBKJS4^3)+LVChJwC>2mKc3~g z>bjrK8)nhD^bnd+8xE;o9>{mvhxnmtNc?7kZB{z)36DX?;0Sc^-G9_K1Ng4z8HAY) zWORS9pjJ3kF!K=Sj`wP^ahgpLb`q>wb_NxNPooDeimd&EI7NvZVykkKE#}m=vi}B- zVR}v`7Iz2B(^<1XayyvHeoxy*5xx54^y06DR-7jNOwMInKd-Pbh}%t8uWqu+-+L^+ zo(iNImrk}yCY^Qa$OIN0IjRhp^D%L*~;qoTc{-Ds!wqre0&ylz2JP}D{p_6F;gQd*#q$l3bJwruX44CnN$@HJw9rmcH+q`97 z1Pz#bg*CqQV*|p^a*pB^=Ko;0#hi80nEO7Rek~rt&fm7D^KSRq#NI%Q9`}55d;Nn& zwGOv9SeZkbyL;K6)o~W;foECd6C;)_ZE5k`W(9!L zkj~@`Hu$+0D{Naz_WO-kaZ|7P-_iiuqjQM;Nm913U){s@O>MKtslH?0ds>YU8pXCK z%2=58`O=IT?M&^59~0Z6k74_Qsqxbliy3t%$*)?IXE~Y07hYYTToKJ|k|$ZLa2-wy zTSk*^b|(A$ZyI$6WHKew3}%x%n3`AmurKWv7OT4t)5JAal(&M|gn%Pdbo~#Tw8xko zmv^KZg)?lzv)!y{&?ss)u4Y?a$}lBg(5F`tWZL@0Y#xA`K=NP6Ou&VKx67L$7t`Lno^dFU8g1iCrX zs5SDm(r!Onlr4i+cX#^tx6S<4(;|91OqmAHE3&9RqDzrBA#9CFqxpRGJhe zW0BudLBU;TS^Fb1W)i=PqKlL$({HZDx_L6xw5^P_{;D;vevwPh=MSct_oOT)Uo<2| zwOVFb`M~_5&RwdU{Eej+nOL;BJJIdCg-l`DU-P5ihSIM_2Uh!ChIM^COov|v)ARAF z>~idH8oOsZIh3BZD12){TNDp6*-f|1?_AnIlTr?oaIW5BDF0rCY|LXTq!cVn<(5!& z;xNh-k7C!Q)!e` zE*tP{sKqMFXV8nPg#Y>)1UH4E^S3*~zf8ca8qTYUn2%>&CXm`~gWK*tD9G@Ha??t< zxUPr%xNUqNvH`7f=4f*;#=^e2oG;`9`JaUJb=Amv(0~;E4D?w?K<(CMESklAkgmBi60PrUDZTo*9eRs$rOMV?MTC;e8vm>lhYPisd>=;JJj8doRD5WW7VJ$L@X0la zdq^+w9Z&--rvc}uRdr-oi4x#rK;X&JU zXk;zKF+*Lvev<%Y?E|P-76#)}zcDUDQFxj%P*`(}v$DdgFv+G2K2grNl(ZD>m8RI# zWQ0)~<1v27B<{8wj<%N!vg1afO>YbipX5#o&em!-xqwgavJw4a4`jR3v3$fiC~T`l zO3GT!Mw@}N5*9o+R1CkU4D9HOhRIPUOepri`ff*9{&<9a{lz%0mj?4&YQj;0`_Jds zUG-#Iw32(?Lr(KMkAE31 z{f@xch$`qEsDxElDo!iS;Z7E3Oq%13u-W_Zcghj81bU*7_t_4I=p*X29qb-dLCO9) zqEpJz(-+A*4Do2(6AI^6?lun6!Q2pM)O3G^SyD5q{H~#QGUrV_TOj;9I$!viWrFpb z$v1I=2BLCK;Zblh4%P2R+! z?;b#NMHURS6*v9vC9T8SIlx&JAJJ;csasbSVr$ge0CAPZ`=U0hVN-z-< zmg|cPoy$LA-rQywWEJySeg-5n@8kLJT+Uv*f+4rO(U7RT`NM5#umfOt3Or;}qbB!2^w9={HXgT}iZN;_(WwY^;OGw`#hx^TD(5-=Gtm1nN zo7OvrB6=*Czk?2ID%->|FV2IVzXN0=FJp}S9=0pIfGL&sGOt{DmUwYEyD(w}dp&6o z&)WZ>Tl0sYa#cFzTNaYmrEL0mDwd8|MbIPnFd8_F&ldZ2g~?>|*jVI`dP548@`;%teuIygx(>qaRZ1PkXv#JcTOs2B7S08g5xG z$C|jg=osV3rYUY^W%Y;HhC@@KV8)s5!TaGJ(ZX(d{9%8WouYMpQ>j(Tf#MdmQEbK= zCfAz5lA>an_0N+W$9Rj4v$7+V?O!QC^dyy^*8jq884saS z(*@M}K^Ynv`X~;|!I*M&db>oIjCZ)vsGu!W&HZJmsVAuE9-q}m@}5aF&HR)vy~<9c-On4O7j@VYBmgu;QEI6zQUZ zl!OK<>iJ1&+B)Q{;1Qs`^hEy|JI@o|+xrf1b*FR3>N|KaD=zb0$^A zWC|9oB(tl@BoiA)Ed!z{vq_g{b~0!!?|Z!&Gn(f`_;axMHgofiXGaU~v-$!1$yYvt zy5pylW2OvMthAuN5;fG=8^b=`jbWChFWJ_@EPRSEZ42=6{U&9I;z-(c!{=1YxPW6>nWecB=$;63yctDO6r?M@1!pu9uu?_qaV z6mf-_j2VF^JkRGO@NC7yL(Kg3K4zC}$!do+&}osKw5977^@z$*CGXXU-`qg+-o=t~ zLNSRdPQXMH1I!hzC-L@1|MymHh16 zLx;!xpprllNFU~1ZSJpe4$Z_3jSw6i^qvNKzN7_4lGu3A6ZCK*n$GdLgw+~I4qJz; zcTUinGaIe}F4&p35GAw2U=ckVb6#=oy?z&7^7p~}C!)f>G))|Mu7}FShG^b<3H|#qmkjCQ zD`78$fSz)X{l~G0?iva45KVx0@$bAI0(s7bJ9CD?X}2MI?!Q6{_hmiNkPt5An!-0j z6T7c+-<78<>K|Ic`1&y1a|@#=V-K?HiKI-k1d_=BxsnjQtUfvsWThW&$X9dj(!XU}*uV>?^@M|T2`l8;mY;|wQ2?NvOg-Rj}uVZwG~Q{ZUH`cvhzu=U4I zbj7$pwSEl@uFpU^yJyF*5b$Xe=kkFzPJH<`<|un~2{~H*xWE4*wfCK%g;$H24>TG# z+@?V-)fwx&+%eV59$)X@Kc6;pU;6D{6q%w*mm zd^%Pg^v3llGuF*#`^>Zs3Yztqidd%P)%9wm==8z!^?fvSUqjINYdF9%Gw;k`ZqtjA9Y8_W@1gh?gyu>M#C9FRz9z9Cc|vW1-frQph@8caGJ zgBugyl8|yOr8VB9{K=OowX%ukxK@zJQzd@eVg* z&a3TJ;>BWTvivbYuyZ1!Onf1vt&3^KGvF+;5}Z% zM}3#U_*MxVnR^ToUk*aPG#$qj86V**b7z*^rv;-6@l-e*XiEGOG+hkJxr-HjTuFkO7)+A%o=;CX)NY2znF1 zvh4kK)M6`*{b$^8eugrucIn}UBFkn=*^`Xjea6H#!N{hQG>|KaoxN{p=gxTY={-X> zH|{gOP(PV5;NaK>DCM*kP_raVjxTw8>w)7pU44jSo9}a_a9F2e*7I>lJ z2!)@fv`eXr?keh_R^J7#d-bt)dlrOuw9wM95VU#U#E_*v?U=fYwtKvzWuhan>z4po zo-5$bz9%~UG+BRIi}fEo$WDJ9Z0j4*+x>~=x%{LYH*17R_z<79pLsrO(WJDO{$@&` zYNIW7DU7B8t2E{VDWxBV&h()to#wOWN4-KAe}|Zt?A->8iG4!Xw>6P9`)=I3XgvL7 zj-zfi&o6bXW&Ge6IG3hkL`g9@qzJ=y3J1-J+o{F)Ci(nqB;{E(41_vBuGcj2VDlx& z1$APNq71J!K@Mgu9+3Z*i)W8@XqQtyy(`mz(%@a@YyU)^5say%{E)n~MPa?n2oCm* zj8{BBW4E(e`tgxirx1eI@)5ZI=m*WToDZ>AtMK=2Bkii!LsPLF#wGQTi?2Ush#rC1 z%4fufkHZuFsf>dq4ldgRS;OLBeohN@{}d4Yel%{LuA`Vkb)>v{I7WUa8D-erZ!$ThF@6 zAs6Z9d6s_+lSSIB2{@86os>jRlVj!!vXNel0xkk}XPa@Nbs6n4Y@jQtx>$B+0%R3T zu>ILs_|&)1OxC;T%g{l=*bK%+_J(axC9YOg(j`wZ`2LcF^k_3k>CJin~4;U*}X9gDZgy57rbH;?7rY)5@$H!J~z~InSZ6p<^`q**>yT6Q}?5wW(s+Dw?R1LkoMa(CGadtQ($5i;th8 zInpYaVBAJyEmh%uP#aeVo#^=03DhN=Ny#DKsb5+dDeOCc(>yV}V%|&Lu1UDystd)l zQ(*UfAzarcz%1h~>)>`m@p1>vX%I(i0?V&HZzW0nGbAJ_z}FIWl3l!L$suq1-xttWw?4katY1lI^6p20OU}$!Q^mFc$qsL7uIHV5;xt%yF9s?&U zcbe19m_Z|)83$6AG)f(5S9dOzy}SmwNf}r#Ux5{$e$u9kV#wcPiLu8;=qERkYC31p zzE97%=5v3!_s5q}r}#EfJ{d&5lu9@H|vfXdWQlfg!@xObUEUp@uU-f2Qq0|EV5C!nB01JXB_ zqZw{1Nu)A^CLc+r2IeBUa!ZwxXXsIc@l)=M)B>t~;!YtS%P8j)%huEcW0`3twhnxv zQbxgf{9`h7m$cIM6R+v|%@?HfJ&MMatB~@R09tC(i{~_+KX=BEA3Y_P?E#NrRPi;m z%dMfffkd+ScZJ5ZWm2~LMl$v}#r!vuNP2iW)eM_NEr;IHf=re{4!1!{o;;%aU$PyV zFn0HN(X8}bnlkb-QN1S}yM2I$FF#EpE6OPR-%a}TZy4NOCen(Fek6o!^4XJ5hsU3% zj(;k!-;<7-k7sajObuKbEGVHVjM{G|l3Wza|ILwS_mc?hsEfgw4M(vyzZfTKB3ZU5 zAIlYA!ScscOtWetg*H)S{M>{E_DOIwIE&}|PSVReYbk7NBT1caATMWmD0h#>-V?Iq z{j7$w4Ax>NULB1H)<&;<3GJI!N$T@jsZPWM=C!^QH{?TM-ARku;X20ENF z7zgwctyuPxwiZ`ld-N?N`+Np};t`FyGYWI3X`)?>owa%ylp=MR6e8Brr5|HxC?|}9 z%M|%@M&tPHjx+gN={%gzaj@0*gnm~!9sim_^NoV(zF`-Azodu40bPt}Uq@a83#mn> zjuZlK(T@#cND@7O+dU&dN-L0Lb_n`wo{{ymQTX(ey$^IIQGI(Vi6(T@tiAwhd5}!P z$!BRKgwVk9%UAx$WAF4&B&-8G9N`1OlRnb59EK}jR-yWE4Mjb?NhYx`X^>^ZZ!q5G z3rE&td=|&%oa|ZN9)@(u;aIg<1d?^qSfW-@4h1M;adE+{Xv@V0W zzB7_!?BMvZgywC%L{_F*v~SdDN?8(0av$R9!ntRNI@SOu8OFf+rwyg`lYyX_I9+y$ z?&O}P=zHO`$vp+mZ;rrQg7F75oRNP}7peO^km{rgwOkI{Z@S_AvhmOi6yS;CVmMZe zhwI%JBs}*ijqr_v(%BF!<(<&I`XKJfctdknIR1R;rsre(DY9=Yo~WfE;mKOOc*{J_ z^LR9Ty1-_5GWay`j)I38DEPMsoK}uS#~m~5-=+qkQw<~>G(aC+?-A9yAbS1;2%b;I z>yC7cdt8P^Yl{%?SVgP*H_*<+DpFsX1C7`utgNj-{rP1Gt2qG0?`v?kB>|RYPB7TZ zvhf%1(k($ESwCh@y-OXWQ}%`GnqO0u>>-@^asbvz@t9R0%JWUb`3R+coHrSZ6|Wh$ z>68-wo<52Y`&3-%X@vsgQ!RBGq5vaXG&eF==Bg6J_?*M!4dPgdKQv$d9?g*-kCv8Q z@KpDQ6=E>u2lL3ktb|P0EqXQW8tFdnr5p!yyfE)2!?R-{tP=#|;R)E1kcg^3>>ih+ zkHp8uI1CXONc2y18d`Opc6+8`$gc>q3d_;;Djmiw=YG;Z8W&DEbP)jPc*+{)zWgEqli73&)%04a8k$v``rBS!7C2d zQ$_KuS{|Y!74U7{1Sl`E!OIanR1oo+4o99OS-M4j%NOGN{ngm_Xc0tAMq|25u}H42 z1h%6pB8~Z{80RvSrn$M|fY)-Y-RcB7Cy#bhXIwB{2DNdgkP=*gzwxKBErrcn3Uh&* zhq2J(KT`kLOFg-z^j2~X9{9N6)yObR{2s_=%t7oN@+CVh#)m2CW4yF(6ehT#<$fY| zW|W}lM>Ua6)qz)3xSRwE5Qp`nOt|%3i3F z>2-ZNDXdK;<;C0x=9asxHo$G(ew>@tk;Zl1mZXGk6Vl$gigG3%qc<-Uu%Nq-Cao8N z`4vaHQDRSa+kNN|<2V}TMUhB{4fWL~a&MkQbJO)&xCXmSGV*2~^Lwl4{^nNha(5bS zUwDyJE5~8is}wipLw-R^!UU7?t zb#V=hW3j)=;2ua&qWPk|bP}AJMN*BKT&??gNSW zR3@i@tGh+8ZQMjG&;Cdz4vzTqSPI2Av~h#^PJO~Q;l%hs+BH6GF@hUssZ-is+aD8VMkcTXxlwfba+jp zVysB|?s5_hSx?5vzT}v`mC96IDdIEpHmVPjN0u;r_YbGq=Tc-4IEk`%IgqO3aT>Q- zhVh*`>ConEa-H2pZH2y6><~%qwy9LFJB8M6R3w8%o0{yjq!p zhbr+nAjbBs6BXc5qKd8IM&uhXisJtBCd0@Qt|+gLtI5C16)aBRejg~{Yz{_o?nMSP zW`8MZ>K>*uc}7@QVh(a-*=0c!tyHnb{KuL&(3H+iyPe53)t~3;wx6MD>whFuc!$i- ztmCE}JINX3Cv(a3bUY^%qs3J)X|XL1JUf8(s;5ZfZUNcP`9Ob09HhEmMmWzpTm@dwspQ`u z`ZJq(@GkXpl25Yf$fR)Azis49cF9tq?;O%QUqa$Djq%}_2o7kMkmnYAw5xulcYkM6 zuS7omd3S`O9xubA?r*eI=OOjjUf_NXvD_QWK}nzDvCY#IheuY@S9Zsbjfj;mkez_FGrqsX_QgdPqvKDv+6jx^;+FNJY|V-3;=m_oq+NVYJd$fiC|$NwdUS$-VXoS*`A;RhwSWj`Beg z?>j!wmM%I?xZzYh0dMe#kw}Z{0Mu=}e2BSM`XrA#avVNIN(rmx$p_T`& zZS7dOe=3!qzeoy8QpxVO4%SWD3aLkW=nVIQO!Osm?f-*<4Ho#}Hwrhc88iNoCpGql zkkHx^(z)YLsZWnHUg{PaR&#)|quISNtAXx~I>y*K7qNMJA?iX4$V^9?`I~v%>(fWd zr&t(rM-af=P2I!Wv3u-QG^})I&Z{)^UbzOmsK#6`#xE+W!_7bu#59b-#3y5bkh3&V zl6i=fSXdG7rTeUmrYzoKH z)7#MWHy?r5hoig72|pB-AbGA13oo`H)3+Sg9x->Z!+A(HdokaWB`KEYk~8z&hJ47T zR-J?7{CGJft8Ktg&uo}TY+<~b?eyB@EG6nS)74Y#9y)ZI)|rZ-vhFb6b_SraISdbs zy2yjg4DUAa*tBUF6f+f=A4U=tIbui=ABQpTCbEol6Lvk^$ZKbL@;PiDv_SF){6mWo zD|3ZRga0GdFLEee970Co<<#*ij)qtV$)J$+hDr}o^H|nfmSOjejec-9X1(8<5g1my z1a39!5tO?O)7ygat9L&w8O3_b8@y@#%NA0#PM|jTC!{?!k}}8YkVAeFU9x0(nQtbD zN?n4tQ?9~)`5mmcDuSrRGLi~Trnb60WWLl6g@dv9rxgzK@o_LNn~O`0saSir6B1!H zxZW}duk7=r=y8RT#}Coc;v3Nal!47Y!u*)p$!M%Kz~v9g_{{he$+LvM$X{-D-Q8N9hN7HbwWj>muOUfE=bCaFqVH#VOd3$N46miOqg{eVoy+9*rD zPw#ihVTPe8j&8ff`m|j*B{PH#lRgJsTm6r2RJ zPa}AppNjnCL3RE_%^C!An;6G*FPv@&!Ax5K-x*RsxDoW3hkcQO4Bp3Upjhl4Jx?8m zkHaj`AMXa2d*Rsdgt@lb0@0q4gaWfi=i{XFh%M)$*N-0B=_4cI4Wz!(@ zfqam?LfZ0rux0-B%B^-d%XWU#U9(|&Qxrxk#=y&b2d+L~oq6{mVjN|}l#SpIm{{;M z-HGQ;211GtV@`lV+XFC+1BiM4{IwYS+;a1&9 zoVdp3*Q@OCFv1VXong4vISZ?jc+}-3Vdnb){`~P0KKDWdzqzLuGj3}0k^l7hpjuzF z#)V?o?ZbFe@R2#yf8yVZVZ7XBN8E~@i*Rvo+!Ospk3wfbeUv_G9{!^4nO|sCoB_7) zD2C5K1A;8Sp+t@K2gmG$|GG5xp0}KLo4$${{m+|^o8*ia+d@#aaubA|Z(xty2;REH zjKBP(AJ=cage??#>%`NnlP|%S_SItgTzOP_$wNlk7WC4%$<-X*d`A^;Du0`g z9PEQ_(lB1?@f&D#OY%oeBzdRThP-!Y3ls%PeAZJrUb|r>A3G+450E;-KWQk1Bn$33 zXI5fv$5_7RfhRvO&Xo^|??UGoZ9YxXjL*@t#6O;$O*=Ngt2!SKHq_v^;VrDW@B;3y z6nL#<@S6qx2q_D~SgUPl7Cb~_^fL^u6y}|UX0ZEB4o*Fuwu3MqpA)4C>mWu-q${w=Iw2b7q(EwvnTG{;MayINg+Qn0^lb{So4KE3`u^ zMg%odBTz$QAn!ebaXV~K^;8-aG5gTE!VVvkk7M4;9z5+<s!^B!thx|Addz|N`w&zeW+LaUF@)0QLH$TD zd_KOzoVTCx+sTL@7Brn-Vra{6OS+Al{~lqO({YsTY(sF|XH2yf=HG-}L8^BL+%7X` zh*>@JMt+7>mOQ_3+e-+S2=Pu1pW#_H1CxF(#|+*9$0m;CA7)$d8?LV7xjG;Iw^%5j zvc{7So@9%dnJ2N~Z828FNb!=>O!)^xllf)lC-JwY2!pGWu{3om>wK<;!6!H5)&9V| zSHt-c`&9X$kJI?pb}xQ*`~iMU$06L$O2UtI*Wt#V|ILY|7&a~yv1j+g;#U<;%uPe- zdPCSJNI~+JD|Y9n@Xq^^_`#%uyq2mZKePw@=y8_(s~ci4(wqYB#42oHXV9WtH|S5E zhHo)bVI}jDYD^ze4)cWhE%-&nCRJo3rGnpSUG&7^1}V;w#J;^9R5s-mJq<~vIs5&o zTV@y4Trt&{Eh2iIYYOc{hrng3<*C{)9LV{xn~pXn2a2>}~<*Gm@sbN`9_%Q6mm z?B1B3-$TYxo`|h?L7dWfgtGgMsYn6osxk-i+6emd{v@r9SV8ml=hB*n{WLeqfigc2 z;|q>`Mw?p`Y&T28XpbE1yMNG>kC_SKA`E93Z9?GgMtmN~HUS_qwABVZkSh-NG8qgp?u z-XH4CHvIPRdHjqgCj8gbXO#E*0|{?s-KK(rSp7E<30nh@62#oHR&Q}oN|bMUtc&Ms zCgGREQ*sx?!l}|7P-lDA_ojT>GXq{RU6Q{a){5MU1PmFk!nFLu=(rMw5D{0L8LY*` zh(oM0TGYcj>pA=b&N3B&keV|AVH7KWqze)jOnh{5Cy0(M`fLjw7if3Go?h9<30}I~dI8hiN%&`mBGp6pOwZGiDpz!hU#H{=-K0}~6{|N~ z$K)B2D7|+Ov;FKC=WQGm7Pr$F>t;G9Ye`9Q+W4IRf+qcIqT|Qo;Ktrx;J+F#dSrRI zzF+tg{0Q^B%Mmj_4{h$g?D@Wp*sOe5E2X16RG#;LC(GNhGuu*$-3!G|A|@je_ZDx) zrtN-MU1^U}d2ehAW;u;x3Ron~_~t5kh>lCcWQU43 zY^45)_0X~;5W93NKPQ{x$9R9o3iCdg^^WFW=%ztkbS5^8`$g8pdhqS&hTPl_Y(IMi zmH`oXmllAjd=e(LZ9!bz9>nB3GcR=z-fXeK4VFijT+)cNIcfM(b%wE0(y^*|6GD!z zLH5uK6zQ;jV8%OK8=Q*1QWMPECxcTg^YvQ)7wo70hc^e_VfsAwdu|nBU1}sM3|HgE z%3q|rg>lyA`{9bB6*k+?LX46b2Es>U*PCQC{78i6wFOYRrOn^Sk>~4jAEWi!36$=( z!l}hP0^1dM2bGV^$vKjr9^1oy?g!@Ry$)lu3A{^FJC^RaiU%iO!gTo`c)5!61*4+S zDY^%nT-@-~@)~TN&tvnN6Y%~R#J}dI^Ch0r{L+uv=z0>3{#DG|>p7OcRoaKQ*?;kU zqzoT-svX@sub@C>BHvUuinqG*0%hH7&N|9>-1?M~) z5Z7k}22eykn`3Occ@QDFQxQ9yF$=3@_@Fu8Sl;A1oYu#n{9Y)w4J~DzZ06jqK7bu- zrXe{X3H}pLVSMjY46|o_y*G>Se3}K`dyRtM4>#;zHV;w>XE43F09ii^ab}$z&wISY ztOpbMYyQ&wF`0Zs_@$xlz-P`xnx0bRcf&A-oIH;KKyJ@x-J9 zGwfmzecc{P362o^Ad49`vIrIbN4vAHupCDv+U?FE|LZ3dT&=@wvohox6f<|}90=+8 z;G)z==-9VHw4L=+s)KQ5(2VhG#1VXKh;hIh(Ino8g>Nz-@hBLrFUR9ulQ#yh2B2Qj z0OkeKyhhU_99_hiRa^)^P&K|83__7G z^C14)vzeVALxM7?>0Dn!HqEw+=A5h)IpJfLG$FTC0IQ1vO=%&je;~x^8Y^<+3ktYQ zS8J}vRlxP0@u97Ky4-@rYE){yk^&Y42;7|i334aA;Cjjn1eW(Fa_2t&<}CI;;a>ka z&Gk1akiF$1f#)160Rj}b*_TFh^0tMX+DtoAwy+aSGW{l4cg>kJ#nlAe7jgwc@pCzE z{((TwuwG#N?-&I*9pv0&2e<&1FX=E376eCB2=b0qYiaP}Y@!o%e&rylo z>yculuQq{#!ri!bu8gzua-)hgE5Y%5bz`<1M-47HPD;Kc4Hii00~y7n8w=1g`#3DmV4&R|;eQUt=7F z=EeLQUwZb(5@#*UXlgwXpYQepnw^JE6dSnf^;?G=xaEu10(qT=z zqXs$m1rIr|$ThSu_o$$^r9x0#b&`8NLX?x5qreRq>?XedEoZE+Kw}M!Y1Bh?!C#9v zf=y2oI9vB^Tu$;=u5R{yax*g(R2=9P)XS!D>FHjA5Av0QVb1HgQNz6G-&1vxU0Oq{ zKMxahNUs&N{@csl@)hPjgkIvLmK-1%@oC()p;W=IIUZd6i+Ik+^ai(0cQj3@U%>?= zm2)k#Pq9o&J7Zp6g~r1Ib|1YX$kK@ytgV)z#1pf*8jVxj$m`TK&yuVuK_@ssFzuM5|ua)_J0_Y5i82n2;! ztOQzSx|~I?5Xavf;=CvS=H^xGkG#o+Z zUB0AnD2aNc=8;_17aCtD3nRxoPR>J-cCLNIr97Vj`#Fv*YtEQE+vVZQatnX%`Qk}- zEX7UPL9=oc;PPIA|4}iD@4Pvd58c~Jdz**j$tp=qP-e`EZdvT~P{CcJX%uYuf$P)M zAp3wiuHU|itDQBHavv0M_iEMYilh~7b1H?cdOfz;yhh&NFj9G8Myls*NWU|gIf4xt z>pudc2hVVWZld({!cLlSavyD6Y)VB3O>p`&>oI1(M{d|~{-<6mCH#F(T1l(%eCmH( z{fJSNzeb(r$q3Pm6XjfUaR^0e4kyXf0j}LUfqEVtC)s`9ncLbN1&+$t{oDl`g#7Tm zwSpwBvK~sC96Tabu-3&I@`kzCXEu%!JMU46Q8;2VM)EVxOYkrBM)TqZcWL-9_O7Wb zk3M@D_}l)b$+HbHbHJW{NZh3vdD~DOJAzKebaBPkHc`sb|H!QKG2K4244Iz)!7r&D z0cpKZ>E>x(U<3^_=wE?*iB*W=b ze|83qmA58=syHS8+Dp9~7Soa!VU#l^nRrKY-1pMk=h=@a4I4dnckvQY<$@g-SJMHM7{sfBl?7Zad;Zsr5d+<6}HhmX?krA_Ehwazr=n2R>pbXl%VFBjMw$_5`|Qc#xLntoZXb~ zT>FeK+}`&;sQW47i_Dybc%P4GF8hKr2Zi|5E!Vg@zpGrB_+ze6Pnk6K%%L+?<~Zb* z0lnk(Fjakv&iaG6AC!k@VD}7jP3q4KqeZdODC$^;JNo|E8S)k%W8LZJ*wgf?<|_3J z*um@EQjBFB+if_?eN*V>UO#i9tTr`@I(eQRJqILOB~@X zx)DJ_Avou1KqmimX!wO_I>(Dp`P$ed+$|O0BeQ@^4kpl$zYbPA zXFIzKoBzQS%UB24U>lDv|CxAlU!LWn=D@kHnOrje z(yhy*aKa`6ViRAXJ~RzqkI2Jh4I#085`3QOz)o`x?rmOxz28TXvKC{9f7hZ9t7NGo zQ-u0O1mxOfOtV|X$*E)-^%hOUi(3KM|NJyI3=gM-=9iR*_arBqhd({XF=5XDYEF!z zSxyENUuH+MrW8?*+$=v{?s<0!8WW-R%K{7MYA~0u zBTfynUH781IAq<9#RW~IZ`Deh*}mOp+6x-!T?i+q9dQ0x4A&%M*qo1pz%&G(H4~uk z^N#tGjBzp22_}+@@HTfhnySm0qxvzPwvB*8Da#q&WPJkzOM0|7m!uOSDNrz+-nL28 z!~O%*r>BoJ(|PzVvkQY!;^b3qO3B+N)A^B+bhx~ef_7^l`Sx@~dBkC?5p&H{DdH!a zYov?|!L_xS6hASWyo051mN9H5eaN6!dk>M@D>=SpsXV_)Q;k>lO~Az~4G{hBI^Lwd zB>M+1X>Tm^IY0iwoyv2fYm?TJu9E_3l-F{$Az!(kj}BP9bPFzQU4mytKS}e#Y;>_+ z@SbSa(d6eb2VV*b$3NnJR_oKKS)%0f@-ZEsr;H9MA&gzy%iYp^$;JJ>&uPpY59ec{ z>~ji%@%%Z^OIrn_dog&UCPq(hiPA~IckY|%BIdHbg@*MNxV>o$^JUDzPG#nF-_;1K zZ+Vzu+QBk=%-y%9neIf2;Pl_yU=vih*T018uCtV<@}9nSPGt_nee+d!$~0L^Edq0OoHNIFUu9pcR>Vm^}; z*4c0vGYJdXyy?W2Q;=^N$w&S-g#P0aP}80P|4s+oeSDqH?>RwFdmm8n#$fVaCywwF zM%cl2l@*u6urJmVIR{_UJu%iJiVenwkJfE3rpH^VX?v>^H$SJBi>t09<748Sgx&^Dq-Z@AFWg1TYY)?tip{teC_&NN zOXx+;aFh=$;tHhgxF+Th4$$?dBBM}xo6ekQ<4)2uwIZ6*nnt5Go0EM(CT-}=B~KWU z%E@DN)FFv>3eJ(`s%TQ_ZzazS8%g2AA@Wl!q9Ua_PFw6gcU9{X<;H*H?#_SCx#U^X zsJmOK-)Je#SvDK*7<)LTHl1Fq;qZCoVeVE#ELWo%M8gD|$%I^}${AP$m9yASNxH;BcoJ^DpeO|kh8U@3!g|UL8`v{D!A$1kGVbNljzmZX>LKB zJo!~dQMr601$$)^XY-ae>#U~lW)tX<^$m(+6OsJ(2JZd#6I8f&I_J6M5I4)Zh^{;3#>Enz1?lU(Q)#7pK$k zjryz;xc4vfIKy(H!s%%g7#d2w*{5jt9AEO`i=dwe;c)c|ZE7!Kef~fi(NxHJPZ>@B#KjP~WdoNU^Nq6}*+$O! zN!*HP=9zwHfmLgdk&(@M3iR%yvV|wP%)V;w!en9Wozg~H!_JeCg(pUAwIZp$SX!JR zjrgR+T*~rBZffHcY8wcnlgk#-jdvB~@%uStSB&eF^N`ZP*;x| z#r>RwvBK&|yRwbXoT1F0-n)+9f8ss2qE>+>dPdV_{SXpOji%X2KS;Y!pH47Wf$bY7 zT0B34#-*R75i5>RO+X0udj#W5=?&6=1L87dg=1)98 z*4JLp*7c(Bo9;^erWMSmc!9bUhp2K%7a2sVBjN9Q+&bfj6`8S+OKK;>!CI25sHNVI z(G;vbgC4{;)tk$&>8YskAzBDo(s#gBCLzly0b{(lMuLX7PD)e;G^bUE^tY z)kQjIV8JqvJL!63B4u57!`({OxOfjNV=UEFiFPX4J4l1xu2{|bDQ=azXlpNF=VmVb zmdd3JtAl7$j>Jj*6BuAQ4*B}>3P*n^OTm1xfSOxe>uQZ(y2tZs>h#JX_SKMBV8eVPc= zSBAbR>vnzJipIRzn7)s7-`s1t%uNpDE>!%ec0ZKbk1DuC=i@-#7pjJznH z@Zv9q-iz{Y{?o;T7G;=E(?j>gjkvQa3GaKO@qO|E#={Q9!yYf@)u~_(f;{+WW@3hV zEM-e)QBCn}@*O9Gz-mP#w~ax1*KU0H8HyuiY1q_PPiGq|XyT`^LV+P}w63~?E`9PPG2JEPK6fsa^*LefgLTMb;*5O%5KMc# z8)o~8Fq9fclY3XwqhEP6;@&4(^KulnlnG&uv)m2~n<7aotB?enlIbHYMn{+z=1*LL+3P#WZ$mlx`Q4-r#waXg_qL4E2C_(L zrM|%|axYbarGy5a)IO$&iFZi2F%IiPj-qg56`Iq#$)0s2T#O|#;(QLh%!sE2ZI{S? zoC%C-hr!2a2D;gu{y|F^QVmX^W^oruj~2(T@d}8$ZG>lSK@chkN3k*M#wG0|4Wso` zf1`;4?D{Ds_9jKo7>QhSDTHWBVoFmfRZd8zgvUooY+MemdIz(JA+Fm$tAnd7vPsGM3wr?OdprzwTPJQFA=x{&0S9HacPrIeZein*)3=~MVh z(z7WdkEnXA{&WVdwp|EqPQj0mBM5Ce4xg>bkoRKy)8VJk_#zt%Cbhz3{&Vz|WZ~M@ zXo#}2QPFP=ntVdw{OJ_)MDM_dTpy&Z_r*uc-RNnXkMzdB;;SR~LoG zbTh=uO~Dh3LL{5LLX3Ai+UAGg>5B-Q+p-mgcXvV5XB9*v1A#U%EDab3*`H(K8XFGz zF}`@WE(H_+^TqNVtFTeo4eK61z|0?NeCSBF(>5B(|GhDe&zNPvKXRXmjlE{b*L6o) zcsg_A9mByBb_l*c3=S* zhVS{a2?6u`A*GxM^ToYXE73=L_X^|DTDE)sCBUaPYrMX@0l!aY<3MgKPW*g}2`iq! zV^JF%HZo4P%N|TMh{Zd`e10Zg0ilpOc(it7&pIJ~cD^|;=Q*6;_oox-PGY>sCPjXd ziyZ%Y`6#}bG47w`LX6R6_7^5^ zlzEMJXAxP&m=x7182Wmf_Pk^niSTYJ;-{c!=^990bHrybCpM#F_hz~8dvSK%PPldGBKx}y zE_cLXYr!(!I(|B@BFFwV#Tv$aAHo3hfzCg@6O!#U8T}_)QGP6+CV=%w~LH z5xg!VC?DUB(AkGEkh2CRxdDg^PQuNxr;zvcIu7h8N5+yz7#TGi4%W6f$vjLGjOM|j zHw^dELh!0Z5B~%USO;(%YKqhGgfV!yriTzYpvH%&YW%89(%)_mCwuk-t84p`U`D`!dzOB(^pW9CCyzYa@@ZDIpbp#)7CCmR5(&o*& za}lMIgYTD`VfA|zTIyG0ZbCF3>)&EI_{Zez`kB(#=c2pN6<%qK_j~yRT)!1yS;&uOkMSgScw07|+@V{LLo@e9Yh`9R25k>|MUdNbH7r)))LyRp+l(vCPuOOL&o< zjd~po{Q1osI+qzIY-AO>UCVL1_8o30vEJaT4Sa`I7BAUx0*@0@@HD#y8*ebzg?lLa zSQcmeN;_1241;y;0qAaCg`*2xq3OE>`<$8M@y`?7Uho(b0`>Wk!8-iWmP!09I)SYl zPO)tG0c3of2$j?93>mo*pWH(5U`#eP1hBrd+Y9TwvIw(+#JOQoVxAu>yKU*}U+lq6&&C6dvQJt9RzduZ=c zR5T<~NeKzrq3j|d+2czn{Ooarnj70Xr-;li@ ztHRbR3Cv@#DYJ7CWAWC{*_^89%-SWCJq|fZ3rG9XgpLgJL9h3+=*Qcbic2EP%b3mF zP3xJ&{xrKN5{vpwg*u(@5CZ1D}Hb8FfWfY42MwE201o+%sHmF{}&T;wcA18h^=r%}^;eW^al4 zv82=Hdo|tJZWVnhIA=uH)SJkobw9CL+APUy0F8bAl-=;pG4D;D$5P&pWS6_6%kOu?XMW)Ha3(Ww4YO50⪙KbCkKkR=;>pI$@JA%GaKy zzmH&2uBTa}=}`7;{Rp;f^G3Qk{}aocJB1nhUZoQYzM9W0-^Q+;Jk8oxsIVaESLXA( zg4vJliRR`<6xb72#^!#HH@E9EW45KP%=GhE)@u~cdX@z-@0)MU*Qq(NZ9o68<0Z1R z-Y}1ORPesB`YHDI?QWK9K9!!tOVN24FxQkA3LGlOw72=PzP{1Slno)tO9eEwZv>uL z$*>Ru?mo>DoiDK5HMdyyqfh4Qbq;Kwu_~K5O^%(_du%QeS~T1?Lms&`G<_%#bCO9hIxzVD)ZN7ZRRIu%QE?t@8;nzDw*swbMvl{1yPRwzdA#9Ga_uiZsz z5^`+HUo{poIFi)`EoGTI7e_kia9$vf3qq~hwI~{{R#4)-A(^OdTIXpLQ3t6psa*-^eFNxy}4lw zNmW(MYIQ}zp;%mM%|U(3CtAN}Ac`KTV*Q&OjF#XG!Y$Qsyu6&G4lgCQp~<8@Xd@&l zgW=l~3j@P&46Ld|MQSe87Cl0`OczR8424~PpW)Jk^AOwCfuEwmNaFq4u8c}NnX>}E z+{<=&XE;p%J4!|~8#vRonuKRLWF72I61=Z8x*(WBtwU&0^ak3pp#YtuTj7&`pR+og z;C*E}jPvaICg&x#j~gIl{wE_ml&Pd=tuh!<^pBitbCJmz;*O#(QKu7wB;^L~sx8Im z@&mYe@c{DwD~0UcbV~Y{OAkd7$XDV6$wiAHB|ria(Ym;D)koUxi5wH&h=BG5Z# zCsd!!Lg=nQ{FO|_ow)so2?*jJHy4+eorLFyqJvIQ2wB%b*C|OV6Uw>;*cfDG6fR1mShMD@vNw zxPdGO**{vSKja>n9Ov$Y;d(GC)WxD##&9>;h-fwg|5aav%Vo~IXK(5C5AIGe6~X=J z?__#b6PF*I#pS8>P!9M7`DxrqwK<&k3r^xy@hB`lNjRm>&p8Y45?9DU`|)6i&h`~_ zG_wU`7A}n3`VEV>Y6$DQtOetZgo#f=(dp}mUAI=!%(hJGnzoj-#?9w@&^&Vb#2Em~ zveBrUkHb#IcoQ0mD}UlJjrVBnNAG}dNIc4|58|Ep33~E2kkYL=Pw(6!EIBIRmP06? zQ)koiGH`#hH$X|~o^31)Jvdv~8OPma=lK3x*%>ky2Mf2a4Hi`HF=1b2DKvNA zhTi1QaQHd_o7%a@lKZVo18>mLQD?}6ciW3HJur0H0cZ~NLWhe31{~TC`B`BIu(8D5 z4l8sPdvYE^0}eNr^4VG*^q232j=c|ZH}kV$lb#^cs3|BIO%bBsJ7D+Gz1Y8=d(j#V zIOBB)WRJT*%QOeQCxT#oyBsqoG{e<09i`j>@a%06LcZj3e^V~}GLw;-nE)&2HQ;F& zy!h#gu>JGUb?71u@V!jw37r&=3>@503z>=zoN_pV_Qgj~y+leFyv850B`%ouCxqXr zC*q3dOk|%(fwe>*1tt$e=Y0v>A>g;QBV?D(M}Ph~h(&Z_jzbSZ_?cJYXf}SzRzucg zIx@@0VOgsce3A~+_%Tg1e{C@hyl8_H;WOd4Mh$NcKBH%e5oFQ+g5Fm5QH{tgqN4*4 z<7|cRq06AV*d0q;`K(7t8}S$9(0kI2zEng~cTN}$KevI*yIm>T!Gk`|8jo2^E%3^I z2{J?;(UIK@-|Uak`pQ1azt=+RRHg9Z?Ft;XwnR;w2b5E`VL^QRq`Z5Q}rR6^N zM}^Vr69a^k?HYpNG7%yF>qPF9;GO+5y;Kpi1J5iuJF7wkY98 zvJ#F=7RT)-p2zn3N6GebsJb^m@Q*CT=JZ-j**XPo!M1p_a2C4yd5-uPcZ;_FqFYyr zX_-U`Ww&OMM^rwI|7l8HBdzJ&>h~}TlM%|TzCvP?9q#LXqsx5XDLeNQCFgymA#cRd z;@8bqtgd8#?H;g$Crs#fq6P^v8YJmhMGGh1CtJRA9qt*9Max1k)H)k+M@Qk-aPElx zDi4u=hf%1p5I?55q1wNXYIk!cV5l_GtH%izatj25`{RU#D_)V?JZ~DC9!XENv+-eZ z2xmZqVc%;tEd4P6TNn3H`We2%eJe(Fjsjh}c%LHXN~6tJ1j6F)m=Px}oOV+Xf-MYi z-TEVKuumo5Kbd$LGZmG+7MRa!5M#o#=X(?2=GQ_SlZ(jfhd*_0ilaks4$}>bHKbh} zMR#V-Cx2tmx=m&%%;4R&Wg3`a^Oi;rI6-|<{CR(A8nrM3iZD_m8|h=XY@LFwX$=s! z=J$WjtWGoF?vD)?s2wgKt8pyE`J6Q@a3C~4D#LHi8lin=kkDN`QCMj92ga-xTlmgY zm3IXy=_z;(m<&g7&NZR@-oQ7I+7p!hLVkX64&1h8` zecq5fR*I5H@itr*vT)wd6JL{j$)J5H$@xyBxFLx&XqX2z$fVKdJb7Vy)=wy}c*nhJ zH8h7iYcD6fBl)Wmf~+t|`0-6rP&sIfqQs5R8`w|zb)ztFr52J~6p+cYCf*$>l)OHV z9y=To^uO#99%dB^zhsZXHZ=r$4#y+1&;n~chrr^gCX%MFN5R7aDF0r9XIsHxfwLieF2gzLhr%Fz=m{OP+eTu;qv*HHaEKRA z#9L*;w(?-Su$+g->l|?GT0TviS3o8o57Ka*ZL~TimS*lvttM znRysikDH)d?uv%S_2_7dK-0G$^hkOlVs{#H9-#?mmx&_iwJc}!EJSL4BzLyX#-rSc z$dl7U(`OyX^vFa1fEsGv4TH_%B;K`hhi%OP6dJC>&G1D?%=JU=+jNrhN~VVVOfny# ziPC;kr0t%7U31-N$Y|c_@d~FclT^scb{oClwvb*$e?nPd4~A(f2;y7AC{i?os*jh` z6^D&<-;2?#H&bb4Z#~YZCL;cEHcqz!dQZ7mbt>=Eel$d_+bCSmn1bW8wGheovMbw+ zcz=2yT|DJRenTDTLdR5O@%(;d`FecU-Gc<~Nn5YNxgBATX=BwN%A6zzx0rP9YCD3d zZZ+si&!=blk<@5dLW;-lQPQ$xntJUrjpjYbWo|m~HI&BoqCQgbl*PkSvZyL@f^Ohs z#ubx|0W8HYb9*_>A+EjV2qA}lAwA>_D9xao z6g1d}){M5tBg+LaT;z?&QzMYFRuhx%nWONk1WpdRNdt34akW?mCu&4-Q>T|cYh0Pr}7f zMhNRPgV?L}IOpO54gCmQ95#vjbEiNpc_Q@wM(|un0?Ixl;;Y^}@_&Dwd^*n3c+T^) z=V$M+Z4)u~Tn6S2Ou(U{5~NM%9=lcC<##uca)O>xmr*yVj9Gvum0O@(#?OV%7o)a~ zp^5KmwY#Qc{OVHNPmjX%R~2X&dyNWf9?@?l-UW9$MZW)zQgF!`x+T2@@}nY=@-i0R z+*438GY2o)%25Ao2#$SHMQzw9I0O`vd}lEg)Wwpj^iAsiR7o1gyJ=j_a0D~?uJi!199-(1?us7P2)=D@&4Ef6wh3X?AdB`GIBrZeIHMkzVKP? zjdol$|BS|rlbHFt1UXNxA%)+|OvBE>EwCQ5ox-_y*$mZDd+_4h5^9~jhAe+gC$Xni z^!)yCire8%$w_fEV;6T_cvRD}?;iA@buyhwSwU+*DwCdy7&V1xQN_~{Xi)8;f%kcK zGxh-5Jr_fkvtSo%W>b#OS~A@fiFL^Nc#z-94Sjr?kP}}?IR}2eNsCP8FKe9XSkUU zz<>p3sWbc)wI;~p_X>UnU+YYbPH~jNGxZA=gj3#_M4nfQ=N+zG7+9!rCQKOEtfkm? zh-a+E$&rSn3mqFNPp(}W)O%Z;rpUF>ubByafAf@*EAP|L&wLjc_Lp8;9K^wuMX+$# z3+I7Ds7!GL`As#T>#)X_T3uMEo8i&0k2H)kH2eZi(Pl$Y29QO@%}G zbhzA4go9Etjnnd>i#@@d1$dTxE2K~vb%*}rtjeXX0} z|1Iu-ozz8p+>cR^j{=6oN+YW&hV+zm$;a^_i*8XSw-f_9rD#nNDrGEn?R!?&B1y*+ zVrZ42DV?wwKr5xDVb66X81awwQ%)0XqBSbz`JwQPB|V<`p4rIj(uD0_NMq3(>h;lu z+i_Xi{6&;Vdjf^@sY1P*^Bq(S5x+-FIHvOyGpes)VL$_$IPM2?{Vh#ncig8<+>=mh zI1*{}oF14Y(ce`^X!rC%7#Pt^CHbdl`NUMtHE*JFqlZ+Z)k99ZE|HkgdCD4Gj1Z;c zFy9jhtJz;ERCX6V)`+Dolf6jv%4s^kJB{=%Fy!*Pn??9!^k~dR|5adA-#k7sFF=TcHS-tpd@TIS_gB{JtHPM78V|Es86p11%~rTlbH4a;N30?O9~} z_$NykH;rcfvZ1MFD`~u}DH&el{yXK%Bx}X@iajnc=^l>SJA)yiTSYyw@+hp7#F_@4 zT`ncM8R<=u%G=1mZ#Au0x14m#j?n7HizM>yFrO!V#Hl6GSQS@+ElV;m&wCY~qzuNB z8Dr4?{1J)j^;46vCThm=cXnzFLd%V4hk_lwp0=Kjb568_IQOOD1jUHB(;DY+x-h4Z z%x@~u2XhNrZOwWAnM;}f8ZS2cp%UHpOdzpU8Pu9{m-go9!$~p_aeDjlci}9aq4>p= z{k2G)`^zph|DzvI(kXIX2HSBgfj!l}%Yt8K)4Jonbm86wYIuBrLeB)zjpG@d#d8cX zemT%Gc?`=(7Vs6pI~h^-`KccahMvz*@iW&1vIIykKv(xFVK) zZdI|<#-@0D7+B z=>K^MS4?^lEgwdont4v<1n1FS+XVk>%VDWni)nTO?>)}O>W$fWs#?X2fA+9Y?(!Tq zA(9+YE|QxmXFECnV{i8GO!ZrP+Bs(<883IEbkPSCZ+4%~ta(7bgY{9f{5E-hdP`cK z`mnolj*3?HP#R|xJvWS?;_0)gbY}sT#&f?4?!~YoAxml;b}!y@mVfoKiVWUUHU}wR*CB zozbjMyMetrzn#?1oFt{jfAs7AU-s?Wb>@+xM?)@@Fw;|)S)b}t_GtV!HhQ!?O*%Y+ z`ppWMm0C9YvQ3nd5+-8V4GU;x^7BGO3cK>OfO!wQ#e}YnFxtNl6AIc;G~z3Dau@8n zAw%$;=OwT6z1~L049tIWl<6KTXZp4GS>VP@I7S8{@xMffo=d_YZb{TBKE-q7MijL< zh7??*Da+G@I!;ZYamoI4OC*7rcQ`Tm(}!62i<#(f^2G9#-Z0Z1f?0QB;51!H7HckA!Ac8skS=}yXfH!oO+~gF|;Hr!=+Q1ZFM>0RdE9};* zm9YQs2p-F4!dB-mOlJ1Nc;r1u%soM0U-(nXrzW~$vYB-!H?wGQ2?}UCLazOX=*Uja zD43kix*KEI!)tA9;#^xS3bDhfgcG>1;v$&)2L#8QLEx=w)^e+YnI)WHjWMTb=DSn0 zVDV4V7H8biH;Oc`IMC$Ec~nJf$oGo}iJvJzLT(B)uJz*8k5C-1Jb-0iYwDSQxWH!{4Mx^&s zANLm9UowE>;#}Iy_aSzYW%PDg1Z936NjFMus5*Ha-CEa8A10lkcbsQ6ca|99!xb>e zXc7)s_fv-<@8Ua2BD=YOw(`8+$>kr&-Tyd!bLk-Ef3HaVXc_IAaEH9cTjODVuRBTRNxV;4UjvWGN;t*rqTB^e zq(68gXRasG-7ST5&?=Lzj_aZ!<5V#SJYyMhg!X*-Kvs)vaEZ_US_j7PcfB92cxyu* z0g|+!>k3nleNG#Tlc+mk5~+W?L)#{LQSHJyVmvpq=gLJoU@4E4_UmZnu2pn!YYZh% z@I~6@G)ScH!M*=3p#EPp&mI>eV^kXT{QX9I-*KPK-AsDdGl)9~lu^~PjiTFDlDW-c zN;vcimI|&oVdjRjF*nJ^uaDl8)sUB`FMa1eyg?gxQJ;1?THlSwrY4?sUl;@%WFzGJ zX;`alLXX{H>U*MytdcpnTC@}o%(7tZ>Pv@4)NwygHk~=CM>jY#a8tAdO<4^pAEZyV z8H;#cG>tMn=TJ(2IPG=!qnd>~`A#g1=6mVm?6MrX_FoyLF6^aWpVq>Sb4O=17{XCs z1+t+-@VapwpHbA&=oQJdsH2-USLXBiIcHer&JaeJUclTTuP|>?K5B+Jfd?iPU&=DIFhqhJMPs!$0u=b~l^Bj7ZCE&%HN=%w#fkv1A=(qn#l3v8m$MVfICN-R%-U_BW-1S_v!kaV( zNWv`XGF^7QMZ@#-F#e`3XP>#Cy7w?WKN3Qt!dKJj1qyh#c?`_r(rMO;2lQ#(ep+X7 zo{nBL!14Ckh#timP&e{Xw($lgsT{4yZh;hp9tc3#=2y~ zch*@QL}Mm~P}Q^P^vYup+1sQr=RGA%+9aG@Dn;mMtQ^g4pF}Qxxiom*5SspCAh`?{ zrL6(?*)6B|iZnDaYK5W$yX-eaF&TOj^)-z`?ol`C(jqoU%dfc8)FCUAT!`W0cmOsO| z|H?v@_pDdDv!wDQ3gG9}f(0`8%rmLeE*sMI4Y!$J)-ASk=Uo=2@PHkC^@zoGj-fe2 zud{-IHO$eVk~MlNQ{neEs&PL`@9SkKwrL?9ntp}Z+t|{Rt7bSdww+CGj9~eB`E1rP zbG{q+!ImE^qwK34tVhg`b}v%F_TO5xey=!{3|FAgF*n$m5Y8}-tY*iKnz5ktWh|(w zh3!hd&SWDNvbI=j+H(se|KA!KwUhsU`uV=LlJrVKoLvDb-X4&atzoNaPspcyOIY z%&eoNAWLkqN?>)fi&@f+m+VA$7}dQZs^xnXJBJ3=^2LNE9s|^G9-ya6{PX|025EBK z9eVO4n`Kl;5v7MIrEVHk@R?}ks~o!2vWJGe7(_D$abMZ;I5w>5J?qMEW!dkIS(E)I zwoL63-FF#BK{DL)6(UEQxtH^t&kL3|M&R>Eg8UKgU+ z&0=>pQ1bvO{W#CoNEn;q}?X12ffZg{Lr@DccXyPJS99XlT)}Q)K zC(cKZz4lbj)qG0#=m#tDxy+QaOWD7H!L%wcfMz|vMw+L@Y4BPJ@*T#{TA^j+l=ql) zUguGPoev#ac9f*E9r5P$Ww!9cT(;$BD|@sfi-vC>OD@-PT8oReA$y(Svi;!2{Q#O&r z_egfq^r!i_0q@wYP8Cc!Ih`^t_tIC=qV~xx>>YQ8Zx~$7a`zfAv2S(Egx^gD(+j$| z#~No+D%k3aZLDQ~FAMKiA>SeUNm<$z*_Aq^uHD3z-z}vVgEmrIia9+vIEUZG!l~rb zeO4xUme$#i!%_KsumkOwb0?YYN%+Rz`X8Xq{?#<+=v4Cesi$oV+F1EcNqUkPMUOlt z(3cVNB(?4byEAYlop1S%HB0_Sc4Crr$Y}#PG+1DG+!1OX@`3(Sm`tmm|6^70;cRBz zdOEZ75{^bVOpMAZud_|h24g6jNA-uT zsdZ*GX*6cx<)tEQ+JT>VYaleVUIc!?&IiWnPokgEw^S;zznf(>NE@4WjxJP;+M_B6H4D zBmX{5=6A7xE%lVD+eLo|X3(l%!P+l2u!a|z}Ii%u`^~YR@wa}XMWE(m9r7&=JP)JhYGSB zkVT66+_Q9H6Ox!8^iAes?2BOB$l?7v#qXq1Ylw{5)r{~vGEob@1 zJh&~5=-(&;6a5o&Lwb0 z`Lw^>9WnrVdZQum9HwZ!D*PlHAQ>ep7*CQDw!Yni8OIWk*Y^&#YZJ(?%at-kZqUSa z_sJpuDt)OELx)=nS!dm=2XVP51UFSGvEt!b3{p9S zw3}trIH{T5UaO`&z6Z+D4Ml*+9R7VjOz)$PQSJ3e`l@~av-~r;*K9d<)fQs#xGR|H z-HY%IPbf<%H>2yw5brI+te)u)H#*|ruJWLDy|)qeaM+el^RQaCxLoBGG) zBS7jJ!fQJrbEt|+E!)Us`WtejEA-~B5(3M`aV=>TKAEIIjrSftT`>}VAG?6-4HJZ1 zo7)6~p{TmLKfZdi8m5GjA<{h~92V3lr;pa>c09$~n+^gKzoKBkC! zLlN7zn&+vs@NxMXoY8tmiaAz~=}BE11y*0BI+<(q!YqQyM%&_se;4IkGlHK=HBCPBgIqkqXyNsaP2*DeM;q)yLXr1byQ~mvvJ@*RTVG4Mp6$0Zm zTQRbED6W)Pb6)cx*hGjULGm6nRrs0v(L!8Y^c=w}6$Iw|00X>6L1~vNmT9Wtr>iF% zWOm@i3QMSed`p2pWsuHU6MCt#xVy8CRL-?ixBd+(UvisT@1@g)_cxJus{)EIT2P^{ z4d-$*y!4hv`Lkr4O%6gyn=KylyQnUo-&q|vPBWbo@y#<8;abWF(|reR(a#V#N9m-r z39@GmhPulevY%Z>8+YBHtQFjs`tlTc%C4a_#jRALt&QR(;+WZWlpa1ljlE|J5x6o8 zkqSGoJ!SzWs>JeVLpcs_jl!GRLon7W5lhb z`#X5f$e>vx>&eEpjADouybVU);RDExi@@3m zMd&@k^ZLR0^tR&=1w2|v@#V^BpD7JL(;NzW5DJl}%i(j(mNUcnysm8^&(Iv8%<01r zv_FDuI=7L+pQX4lYcuY&reJc8AvOk&#-QVzDL9eAq-!*E9~$CuzB^p5^3OZyJ%#sL z2%E-#!;ew_Q24D2XL4(K&f`8R{l8Mi$VM9Z>N+V6sylRw z(k*OpaH~FidY|yQLmFIH2f|K$GGgNr;XZC7_tq_d`1#*d^!^+<|K;vqzcnZbFvL>6 zH`~4a7ky6br`Y};T9TNHhy@#9?XU*{NAv~PEES>dw488iqXH!2c;+SW4`nqTgITv3 zJ{j>l-IN>{`VGFu+4O_h?dgf~tO;0bp&=|$GZF^XSPO|#fe3Tvu8}3naiBU7tCXgreHs56 z4l_Z%;&S|WIu5D*0+i!_(gukiX2ov~nsCb~1Z2i>R??IvExx-b`e^=xpfgm>g- zYGB_#Mb6H;M74pUP@H@pPV)*eZHbsL_li9#jqPw-l%I(|X@2<1y&js)Sdg`L0x-yIl7Q*sjgJ+6%m~?$dpQJrWb{-TI1&GjtLD z*_mhjj$_1=PVNuA$!ESEaIMlr{UQx??RCKB-Un1=oGL?T~ z%QhXs=8uo?W4tZawN)ab%v9L3=PYeJAdV+?^N`b!inw3Rs4jYox}n_bnYk33#O6cq zvICTc>2@3qfJMM6wNMmVZBO_<}>j(_FSg8zo;!u2iJscrXja=2vASof0r7>L4Ee(HHdR3=(cT%M0^P4;Mr#UQxY~0Qb%W?x8P- zL3bR=2cN{IniWEIl(#TvpQF$op@XHLr10rhDO@|R;ZU)M%V6OsPoASq!QdndJd-HEafuG(f4+yR(k?!W`G?KB`Ha^}1Qw+x z=pO%r0`)AQ*}fX@>lK7q0TJAPbRUC~EpXFi3f7+3gv}E)5GZ1c(^b2%{3m}l4M>8a zQXTZlG_YdoR9sa%f(O~k=UJGL6qY>fX?TmBX;*Lohbf_(3O( z=gI2Yur(kOo*xTxMQmf#kDHv8gdj?9+_Bhq#bjAPC`3(uMQ}BiP3wW z(cU|CaB|TRj*KgZ&8j?97H+`Fn3vScolv8VWrZA#SC9`<5G>~#3nAVANTqQIcVYO# zfP1Jc_`4?c?m#rnZ9)fs2Z>i+!i0%dDakn;<8ZnX&pvJ6ya*h zEa6UAJpCHhOeRy8BDcK&uWz+NGtER;`tJj+dng4bjTl^i!}mvecBm0uk6AL0abfWz zG=Cl|>>VW`nC(&#L@yl?`tuBsRv(0fx=Zj>JV&mz@hpzvaU2|4tpk>%9@T)!y5 z@!edk?%RWK?{Uc7?vB1e58*z!0rt7iARBBaj1xleeO?jnIyU0Q=QE@-v7NpaSz*9> z7X+8ALe#_%4By*JVI%ipU;lBe9A+Z8Xm#-&L71R#QHkiwXJGers<2P|7ya6^4Uc3a zkhJP5Uh+R-{#Rj}9D^Otc3?{3est?>#F^+IsFrQPgo?d5FKomMCv#k;AxK1jB4#^Lz+c$K#h3jP)l+r0qs zaicM5_Y#ck{|`aky_h=v0|r{}hxCu#=<{BQbK<`to@5~O9TpMp|C@$h3vIlJwZIzj zXFRVrK&biGgQ`_3g3rKygiB}&C)V$P#q%IM&e@Bj7e>Kao&Wrtu;iUsXSCkoEU$C( zF=lB3x;CW1Du{D)tHp$lm%JNj`4f-8&n>&6A$mU&js};|VRZrJHN0;ySVst`RTPYF z$O$&9rG#baIzmqUV4+LTTnKO15gvSE?nCZWu`$k-tW~C`q-c=O}c9;uSqzr`Kt;T|m zeGKok#z6fu&vc*B5^_HF!Do+*P_p(X*0x^6zMI$3u}TS{Zha)}F$m)OT`+%H1PtxA z!oE9)yXaEz`3ZlgF0@Cpmj|wlbwq~fUW7+RU=3$^KD%)r2Y#K$fLm=asgC0N`y)KF zor;;!emKdU7tQUvkgv(}MPCmK=;Nso9R&=Z$`%tfo>D#-2+hL9YJ!>hTsz|#@7GLCqw>W06yi3r+t7$F7~ke_TQ z1SCumZWxUfd_|{Vr|t}xoABJ8@^`%a_Z8pjcr#(K|lU zsQITU|2jXrZ+=MuTCYiMs5>5itszCuFVVdLlx;qX8>$!ZuvtVXUt$jgZbDr{KJU$R zkgCfS`c?jqZX9GV;_tW#@*(&t!i*@ zKMN33F$X18AyABaMqSa9Va7ezn$CN%eD`UHSAWO7r%&k2<{r{WctM;XgNd&TG5=dU z7V@sRW5!-gf0&6b{*K-stA{yfZBfT}Kp8yGwW?DdLdr85HA4|~PD3FK;OuZ8J3KSm z0qLjV7?2!~BU^H@)2oZ}KXlMKkFT_}ehu%hM)GG`1(N=5hWW`1{M?_1Ti24%J@6NT zDm!2lm4)V2C!zH32JVht$vK1E-Dlto8{X;8Fy6!c!Qm)%`T+MD9btdfXd$A=0s{AL zh?;K2+nbI!|MV0lOf1Fk17lGB)dM||LC9Kq3*j3dV(*FWI5^uZ!pC zLzMB{LV*9gL0ETZ0nQiA#P`{9*zUde+}ct^)*ntJsL~HUX%R?RVdf-?77|y z82TCGvJvMbPd$j77gfluloH0*M?mhzGBinAz>j<4^oz$JZRAqUfj&>!nIm9&>m4mg z7KP8KC-gYrA4M}M*xfcjhK~#Sy^K*EunA`0li;M~i}-KZ`1K+M^1pu2@kVV}h8Uu< zQv;uQ)_zULa_kfTLPJk!Lv8ACWFFbcyY&88Baw;z=bLbM_65wUY=LCuW70hPoTA>z z;fHA{u%aH9nmDt1kqqul@Bl;vc-`f!1eJKq8227SFVDubtkLLrHxJ(jaL>q!W{7Hi z!coaIG>e_&-pKQ~{;h&0xo`(o`Wp&!ETT1>q4z5xkITDFGKm<(5Vu!gVYc-<~7aEv_&yzTTisI zjzaE0i>=tAxY6(-@reL^< zITnO?V5gcP-ek2?!cQqaqgIFa=h3JXY_M%XA-!FohO^TYaMisHK9x`4SN;*+>l^TH z_C=)H*JIPR2b9#i3%7%eap^9fcXS*_GXI)SoEU?9K5LQDa|L%V4#Bm7w)l`X4<8q+ zLU+#;oYE#3rS3*7e-;hcpNuHo`A}Zvh=$?u5Z^2buRsCQ6c@t5X$EZ1#^Ik#A+qO3 zVaYD;(kv~;Zp-Ji`|fI3+k_!Wu!ZghFKk>i1E1#v;Fw}O{&DBTf1`rX;KfiN(t5iLCDLh-YdeOT^FI+>7XV4iOg!l-!ICxU_z_cv zZ(cXBBdi?B%P-;jy6@1+RmY!2B`CCbh}p-bg;@=9!s4~M!iuGxG)tfF^B>N|hqfr( zZ%#+|>wbJ6dk$(28p66vLqR{Rf)38&vs`ZhVq0e8?6Y$iHuEu5PuxSxve)>1>>pkl zEk%fLG0%{`#k2uoSY@0IO|6G0{lxR?pL20B`WAezL}G`0CEu|;!1b;>{vpMbt@4}(NKqPzYL8bQ?xbL2Y+n@I0^i2~C(n`eX=XIF=;uy5H zorg_h2Og`Y;rq)iqf8{j?ej|BJ=iVD6{yI*jMPV{v)@Y801kL}^1NR`WBr z!i`3Jc-nwR60*XoOa-A*P!xLZuY#MWJ0zrIVPmI@wk3WrS1EycR1YN`y38Wr%E!o(@9qbn&Rc1k8r7fhkKKh z1u3bcbkkoN1Goq3lJDK-b&&he&R|%F59S>>g*uHJID07q!$0un#g-hLpToaa zCLstsoPys$;>fd`ji6t=x0&UNx6S!5P3yp_d0t?v4#TA82+9m4k!z)gxe;Tr{`P!+ z*N(!pmSn!~+JLa5Ti}uufBQc<;x4)SE9P8AnWW-Gt zUp9>QXpXTYn@h}wL)2V$=kopL9(vY)f#SFWHr3}CDJk`_s1^TrKJDMu9K#e0luh?M8P&STuW2L(ND!R zCvg&tgn0VO^DBM3>zP%+N@VuyBW$n%n$$Vt?AH{s*|nM+lqBeU-cXV|(ZOEr{*O)F z{*s+9sA2MU52(#hmFGl)_zrt9`JZxra`KX&pw+c=ulBhZgU7E~mN)7&B* zjFB0}8OhVJc99OYZ(y*Uah85XMo^ByJo?p;K~+v`iAq#x#H|II|CWEy{67jubEd?i@r!h}Q z!>RNTJ+BBRyE$v<#jI&apLCIIc+U9hk5ndaYD{0|^fM-LhI(XsDND+fzvBc9PaFh8 z12t^rEY+s#L#Ww7hrV&I;G3GM$dPwKzkdt8dQwP}B?cmDsV|ku^1gcY7qZfKq&ub` zDbMm4P5rWkt~p+zy4j`}qGf@dYZhbm2XkZZkV-D!?c}{PkrB`@H%8v~r&Q{3oQXd!U|#R; zvGq@0vHxa>kXXI}HA+8VVG}f{qO6I{D@&vs1B2*_+6R`>sDasIBvJa?7-2U?V?vH8 z-c}C6)Mdxn|0p~2XsX`$-y>w6OXdtwrVMeO;TTdWWGHD+XdWmfnkn;4luSt}Q3*vj z`*{utg+@vuz8WNzxj}R8{{3n;V*kx_TJCu{dvD%@4tUqIg=NgOHgRN)$qS{{ZnH*{e5866n%^AI?1vmcbXD?_*Ds~qVtQ=s^j z;T&#{2i!hCpm#hGRI}dh7M~Vi(pACk_)0*i&^r*mBpE*1W(`cQ$$|;1bs)UH4HPV5 zy}!d-piZAEEI-@>L<*O|*8FKe_Rw556Bz=&u$(;8;$@(>L>zPt*MW&Uji5_T68@#z zfz_sXP%TG6v+s0Z>!ifqMhFzXIt_YwgF$XL)H-UExec_pyU2w*+7BG6m8wNKR!H(K*;JP0l zOu4B5jFv+4)P?YzgAy!VI)~+h$pL5U7XW8)!QF?dphcPkn4=in77b@v!9k$-a30`x z27;a!)u3QO448a(0AS4|Ah<#RY@H23zw>O+zWYCp=KfjWs^WK!BAtY(+_AHTm~icQb0{-1qi9x0E#48KMcD^+-PPE)%p&?bLp&K%BcZ} zbdQ5dxwk;WW;VPy>Ia)7e*hyfNhtGm2K@Y>6bx8g2Rl}b06M!!m+|S^Z62#ug?tZ-pHHjj&>RGRvY3gwOK(;EHk?^x>m6`pe^>sxh_$4k^K1 zZCUUwtDULZ9flhoMZy7E3V3D|EdHR$_6@axG${q>aa4-RcBa4dh_h{`>5F z)d1%N38IMGzksXq6v#Feg_c{+!42Xp>uIq(YMGk^;xseByZM=5=n3T!@n$z!t2j_agqTw$< z`wJIVtH{Bu7rJm;g(=*5BnnR5$bkz617Qnsg)1#r!d*V`P{!&7T-)#)j^^ni`+-)t zb>suPpL@j4d8X*mhZ!j6o(|FsG=lBsme9a^D;)pk08f6ig?@*^;k`3#-yBB;^#xzs(_Wv4t)b>)_VYGU)5B*=SeED2#}!fTr4ip<%BYGMJ|gYvOjp zUoXO-gvTR@a>S5_!6?*~`44=Zp$}cm*yr7YTu|!}1B6bz0G~DnuzsAgU_7Z3oZSbZ zV2201BXR_mJkJIZ#SegC(pNyP^|RbfGpM4e2T!@x!%(*8#bM2P*x$4tW=rYA{u>#v z)P4c<`#S>`xgj`Z@)$heu{o0$(s1*x8EBQXH~Lw79Nn8Pg!U_ZXS>}+5vQ^P94ML% zRlNM!{^0?32RjWev>pUYvmb+`jh{itN)h$p zXgakSMvzqa?!|wg>V6-XeODD~vgg2QY)<{sjt#J6xPs+p3&KLqKhXVt2`pr`z>y2F zaB!|Xl3S&RR=CYaqOx5;E8qvvF6;xR`^w;#PxY*R)&QlmhM|(eZ}^2gfzdvqsNky* ziYOFEzO98IE>ZwWXN$tmdSYm%n+S4WGy*%yr$H8g%4Yh5p-@;Rv`j2DTq@w*0$pg@8w9-sj>7mtR)0F{3#W^nfiH(e;L_$l zKr(A4Y}L8|XLjYlf0;L+7Kh~_PPW6k3yLVt+5j%@k%TwIpeDq789Ep-Y*WmFUkPC-QB>M&0)4z@PSb6Z7_G+0iY{#9_Yj*fPtSu z;NTrgpq3^J7QQnF@#*^D^fgb=&hpVuKUfI7o!PD`Z#5`UIUSZsq_Zr^2yk&lGT-rW#HuaE$~d)Tas+ZH&vZ7Y1ZBpiC1ZGwN}B4EY0EwE4`9dd<| z;OZr8Pt{&A^m^wPn89*POqJF_p0Wq5hmSb*7kW5v*FWR1Wd@)=O9nJwTL9)si-A-* zBhc_+1(+{94a{?$4n&e>gV^jMIP$0rx@mR5Q^`xg@i&e@!gLvMbxej=KA(qf-)=(N z(<&gRRRDx&>x1ln_gF?lKBx_U3eK+^Vm%&hY)`Bh>cK0a&^DHXqUr(HE&Rl(yeA5B zb+rJ)@}UQ-G{FvJ2l6Cuf%x*jK(XZx5Iy&kQ#&W0v*FNtj-@~V^uM+Nxa1h6N8_!wlhb&{VS?URCpgr=Ivi=WB;pMngXgaA}78;b!>0sh@N8!&^?l7dhY+ zHNcr^)Xq^~*~6j!>ADh1s?szToi3>=yPJioMZL}f&P&{!?!l36<^>{tya_lgw={tpAmL^dZF;|qM&Jd=yt9=@?T{RB=}+=4*G8{WX6yYkSJWucx-`wLE1D5BL2cXB{K3K|4 z0cy>Kpv}b?)SrF=cC3m50}fq)EGYuNW)A_?SEVcuyc0ao&j!mIUILTyN^m^nD8N!v zoU!gTAj&fYJS~p^&bKasbJIqF0`C%7cP9rpxNip@Hs1{xZwz=HMB!1WOy9Qh;(!hc@@Dj#fs z_}kMUc%vNbI(7xL1fK`o$QWR)+XL+04}o6YWS}S|4@EUcIjdS{fnOgjz{Rf@LGp=4 zmOK6txR|{H3Xc?^*QZW!F4h`!>Rkt(@u3{QK3R~Pwi+DoivfcrrJSN4OTk#wY!JO) z8_bo|19Ef2z*YG;uu-WDm}XLN_t0H%t?oFe^SJ~fZt;QgSUr%>uLA9cZvf9^H4K#9 z32!cG0#B|S1UbItz;?n7-Zi@ld8Rec>p>mxHyZ{ey~^2l8&%wS| zEZ=lA2>#HS&1MRmSwCb8=i@0pXbEfpPksr&pz9j&+aYo2dv_UB;eG|Dge0K)#xw9~ z{1Gtx;2MalW%~dhRRRqywg+;SG92sK1f5;K4$ z%a>1&-vW->mVnVSwO}aT2#D720k!6(;9$)tCyeTYhL)p1?%pk6v{DqVei{aT)m#VM zk+YmBnd{(d%4}fxn-Bi`?F;60hl1OADZu4>F__$+3@fr{ie zHEjgB2UCD@b|#pz9s);C1pwU(%Ygc^OfdAW6#TVJ1HBK<0xgGZP!$jk+>8rB=%Q4X z+rkI8$OG^_D-j0&yb3Gkq{9TSL@>tAeYaCwz_E`9!FWnOC^cvWmMw!IC}I|L3A+#M z1}O+nPk>T(%5Y^#2`Jg~5d4wP12e2{0B3f-5GkDGxEp2wrIY}0TJ0@ZqM`uJZ9jsQ zqBG(9*f$)B=w?nX`yPv`ZvcgV#=%%;BUtMi1TKHi1_Ql3VB@C_77N<28KP;xL{l8T z5;B4jCq!V_c}=kM^G+~n8qN9gl4Vt8gaG=~1I{~m1nf?d2f?#c;P|V9z)8puXf*kO zfzlFCWK#lGE4~MR0%xH3BWIx4&RW>Zde0LwszB7yXi%{rh-KcC0#6ff5MSm87UsBu zXYAj3ct;etb8|P)`EM7HUzr3%vsm3|WD;TJZA#7-_2az3l;Mwv@VBlHDa@7(3r|`MPGbhal<6I67Dgc>n)@{`W8XpX=ytYOemDpa19k|GJL(|Cj5q;8=kL zjt)luKi3hdE5Tj6;WBjowHB$c`KZmlw~)!3IG*n7c&y%J$=zGggDYG#N!r@wM7Qt~ zW7Bn$bhR!(%cT%n{BRjj>pqRLIqu97lPmDC{VsCsLpMnkZ-#y%$*9se8V?Eia<6YV z&fSZKu-dbOTsK=WE*|N^TWvR>`QG)!|EeV3cSDW=`4UX@@x6>}i#%7jQk`3s=?LDa z88OMZuAoR=1fFI)Fhdg5&`&jG)D);HuEo(V z?c_73kZ1(75UHY#bg5`Ded<3(ic@T;W+>3)GX!%ZR-oBAK-@`R)wi@!}OEXEb z-Z|o2S3^Il{-lgTd$pKiFHwuUNkp~dXj}Pxl9&FNkZaS?9h)ogzU5p3^)=8r(nQiv7?V!4~A7+RE`(8$vsGtfV_rU*Ma!PT^;(0?{F-Ar!Jd7H`mafo%pX zh|%+Q=xOYO0P7u(%9cjEVizE#eLkdm1*Cn>9KxS@7p**`mM!HNin6BHa%1w7u z;r?^fAO-eW#CZE`^dl{Td3$6YcioL`+^~gi+?mrOVe|eUP~Bz)qa#-dPdygm2KkC} zjX%{f^MCGSI;yLACtg}W-P1RjcjaC<;A|7#!_UVrl5+qTHUlrJs3h-Vy@_F3 z1d;A)Ly+xh; zQY__x(=3y~HigsGrA#C;6VQPJOOT^YAMdwjJI*y1O>QvrSd-9`OlcEsWXlKPRoZzAS<1C0;tg!@80iLzxl+MB1s==%EcqPNvhnE)B$ zq-d?y6W&X&A|%g}tm5=l8G(y)sm%ORv{UXYP7)G823?1V!69~!mF0m>#FY@4xiL(L z^ef(Cv2#=_LLE8D3Ul`+X3>xL2+h%ppoxUtE4kuR^R|_f|Dbikekj)6ny^$p3VD0X4R?DzSwve zd{vAxi{CIp-eZi&IZv7*ZB1`h8q=rE2kHKo67=8ED`dUlSLnO03DwnIrH@}l(ZQr- za&qrI9`}9-%~MpM=7GXAb~FZ$#I)nY&bv7E+X241zdrxNqVM$T7^Te$pXprYEUkGq zh}TA4;sbiS{t47U29!kro_!$rG}aMhVJB>a~?Ti2ya{xn8$r@ecF zi}_C6>D$xktL#`hxpOU9Ke>SsTGol}*2{vI`(nsBS26O`dwB!0k(wZ=co~wO9fm}X%{PrJK3$)-WU-iJF=_SlJ z^Jr$=wtdErBGSbfqQ z3+sQx&oDts>;PmMw93jd>BU+%=opF~{lQZ!Kim({Q3;5kYJ`_t4hY#e*y zEqbB)i+LXwLsti#C#6!Skn=O|)nr{h2KS8;O#>e1L?SFTP|F;NzH1z*|jLzhR&$Cf=|Tl8I3Yq;sETpBe!RgJmMzR^I}7+u*}~hmTn_fUP)5t-wsCxB zhSu!8)(_?H8>76(=a^a2ADHD01g^Q}NA)kyp#KcrX@7zeIUk#jpR5VMR)@3j2LT= z@Zb6U_+BQ5d%gV?)41&dHuT(%KS$T`-Z*WC;gNah(3APtn~BF>8T#Bm#Sd}lBY9kE zXhDUoA+cortpa|Yc;Tys*y{Rk6xAw?#kakMHx<2z_^D#tnx2SS50>Lvam@xc4&Wn2LA=*&bLG_Ii;ah*t|k>ZaHwN+NT#qk|8Lo7J&GhDKJ zDauj=TEr&2o_ifX6;4E7S_Sal zO@e5fb0A*zawcBue-7JvL?VsS3Ub$W5Vif$t5G=a0}rkHh_9$9VL_pv)>HgyBD>cW zdr#d!JKA5N7J>IDBJLc1VXBH3E9ha%KpP|?_LH-fQKhe3ok`r$0(5&4;LloWxbbo) zUNc#SHI)LSzk&wvwCo!_2{$y{oQ3l5&cKf!cr&Y~!q6p`_oyU;Wp!L*J12Jf;D1TVSn4x7LpZ={ zix_?EQZ0`yQVWoaSt)*|p+vvD;NiAo7jd2KAw0cX53g~SMnYDl;P!AW{HJ-C5xC$) z4tnPxr}t%KedJg4*>MN5eH)4L4k5ht?q(8kvkmo*i!evooVLs}GqNKu9^G5S@TUIC zlFav~=u(!4_&r()>%^wwvi1g8`==Oj-rR%=<`Go~X19~?_s!{GUmLM{CO{+u)MyiD zHL5p@MB78-N!rKfoF!HXP(3(?_s^Qo^9pbywLT_DcK8>3wosNjKh~o&4kj@}x5S8T zUJhKiu9cmqK%tFQaJA|zdUrsM<_dNpg$w^s7dF9(TqkIp zy{mfd$t)0F@*UpRc#3kyexM*xA842y$@71!OeLQhF@qmJF+a-k;AG25Cc1VyY?PFN zE+?OYd0R2DP3mI)ep99;N=9_cfC}{YC?uohkgi*L0sWfDL<8!gg(V-|hnv_m%kESE{HQlgxeiXaEoMcx0^1=gQUA%9RLrk~z8^Zm} zgFlv9kcP$Oa47yVT)0*k{=8R5ckO@5b5xTA(mP8D^j%IDJr^SrxpLey;X)Kw?XPoo?$mOxCIjvR9LbR;T4m&uw9 z{$$RR7}OFhflTe}(B%ckaOUyd^j;Yz>cUBIGK<|$y{sWSS~rnhOF9VFl;O?E7h;|m zi81A~XYvGZYU7w)=|rdh6oYS_!$UITjKCCT1V!_hm0^$IqXtNK$C=Tp8|MHqF(U!b zz0kcA>Tu=!3m_uu2I`U@Wb|hCve~j}a9_|gw9jpfh?~s7jOsn>hI~I>*XjsTc`A)5 z{63Qys|Le4Cl@kFBQt0c{}FjEqDW%397&+I485eL!Fmo0tRv^B)6C6ZIM<>L7~N0l z%%$ZV-c@;1DkPtbs@Z4C&o{kH7|U%rwOpP1G5!Xgv`)nn@l!SGvTNuZjhE=Z4k_4@ zdI0vSS&=)7Z<4%TMQZz-L%se}1wB}V@%sA|zT^BNVA&^zn=4L_9TwstRZ%!})}7h5 zcnzc0cbOOSTaEgdEvB!pT9aq2zweO#Ua%}|HTtmtQU^9Sm1sTyUG>_L@p?z(nW=!^ z)V87TmIbI#eT>P^$suJc_YmVnon&6-Ok5RKNQ5#HYE)jO)>v>fKnRuDGJPnClgBM)f zMVxo*prq;+Mm}W|v8^~t&$7>$6@4F=_GZ1e*GP~1q9TnIfS9SEW%7RNl>YJ zmj5aJ0A+^FAiHg5(vYQMD7HF^6onM~&_|8nfq9y9a0v+-+-1qiP9 zA+!GD6J9FIeU4Nk$^)NZ@=X)o#x+}^*oYhM`?v}@w3-vIf!CzqS{70Bd4%@;p26Jx zxDhRz9S8Ndx`so%ss<&=}M!| zVFFBc-A1y-U_K!y?$cGD&eOeLZRxH$UGn5-9@=fX5pYLqNZmvsInv}z7Br;6;h)p! z!iSdp()M?FJi`Dt-8aB_b|;ac7Dh)NyP&tfI!X5jQyNsuHf8wPLW|?4kgU;vyv@Sr zu)eGowZ5xasdAA_0r=<`Q8X=)Mk5 z?_WnZN6OOUcTLE`N(t=WriFzCPoZ4X?R4aEAr)@p(xP}Cx--uM>Rmj?ys+ye3Cd~^ z8Cmk4Z607=?b?DJN+_dr{~>vx_?T#lq|n5a0^(I0OKb|P$)9~9@P@81((cY8fi^Di zqNOgHmA4V;ZUxk_e3S&Q7v%a>t>K@LA0vZ$a{Z zfZVMV_*x;5eE;oBR%}v3OV@?s@(BZC**1rxTf4Pdzq1FHua=^(3f2;(BZnw>I)m_J zJgq&CS&^Lg&eSgM8$B94pL%JoY0_O7XM~D z6mKz38!O4z6>3z!_bCuK6J=%gv>aV{n@?PWKM~H*1T(EiftKZD(E5G8%-M

Y4jg zIkQ^L`G(AXI_qpKDNl<<_eS$;eB%GBUT!2wzHZq?Z25<(D#f>vI!SfXWdDp@KQuw= z!rbVNqjGfH_ek1vJCQlCU?VzO{F&zN{Xj+a?~V1xnNMP<<_Pk7B!jKjd@MaaCZL*mv(RmU6uOS8g434%Krus{GnmH(ZA^cX4p)W}=?#f?VT5G+N)uY{VKcWWRKg?iw2{Y*S zhGoO0fgz)%Or)+TUrw)(W^ca1ZUn*YU~xWG+$M`jpt)Ia3|gJ{loBmqfHpp`mmUTDAWXEMfDA&B-z_^zH$)s3ITP zm*0m^Xa7g9Wm&3?v7*hKl zO1yO^kFu@l^AdMX;<^tt3c9ikN9zl-OnL}tUp1%N#g*ts@NMv^M1}9vJey_~w~-V{ z59)dP5Dae;B1iv3k(0mgFp@KM=y(4eR9)~RUGr)w+4wID)fP+Bv#c(;ep4w4I5`)6 z^sIPM*WhVNL#`t zVy2@>zqdVRnzpbkQ|CpnA|{CpQcpBJ_aX0ANEUh$8c0y<1T)9PgxzJ>k*^`&h@s$H zl%^(${Zvipm1}X-)m4mtxKWG--J3%NRI)SbTjspRNoI@Xa%8o% z6x3a^LK`gJlTzJ6a`1gO@A9qyrtW;DwdDOE;^OlfJ?09K|*^F$@FcEOtkFI~XM{~N*GL!NHWZDWFD!hUNc8okHPAwOS^>G3C z>E|b=dW#a(UYkI8jqJT3%0mnK?Xk|X!^F(Ihx3J2lm6*|Nr{d||J++^1Pk-YAr*Dr z8J;bbFPIgB?Z;-3%3Uptn3f|+>K#~HkGIqvJM6E!AT$#_sZtGKVlj9La zBTSGJr#}rjwsX$1Z^Ij!hG@CvGNM?&n-O_vP4)XUskFj7Xqe_mHY9RT z>=QR;zNQDWe@qCSeR6_U{%V5d4L6{(|1b>BL1-i`7for|vMl@`WGqaF>G`TjZLggn zd6VlQv!M#M)GVYN;bgjX_7#%jI?4Ra^CnyiYb4V74^0MlBZa<$j1Bh?ZN*RMBetV^ zrp^MorgIMIPm3jIcI#3XXCY>Ddllt==khP)u}gslb@H}6p4@qtOWti?k8ZjK@t%j} z^0IRriOo_e`hG-fpd46<}H)tyZSM9 z{yV}eh!LXkRcFzE3%qEq?P~N15Ab$(o(5)>A3&YFF`b=!7p_=w9d*X9B2M1bDAscp zuTZm`yjF-qeMjurJyJaB*>;Hf2&EEEd?KE$H51P{t53Bo5&wt6Q5rZY!JoPGHrcfP z6#aeEoNkRWM*fM1$E9H4kEB4o!g+M~+Fhph zy#RlmDxu$l-co_RbyULJi&yHdMdH?rlFKG8SbZWJ)jntU&Tr$$n#Vs$Rq}6==@3HC z4w_LhK^G_VvdacbKPDumnwQS4bB>Xt*HZnM^O}>>r zAY-C@uxYUlGqFqr-8FsAE9=}x_W#sj>;_z!cfO10mA$SsPDl(-)#;IQ%r=_R^_Xs| zJxAi>OK7r^Kixk$jN&Fr=wq7-TGsxN(-BpK-W+#e?n;iKFX!b6akqj=$4=0S|3;|f znF!iax{mBXLrjBI9{nao@zL@zYwSv`~?19#g`eRchQ^M>A|KvkE;onT9#; z;^<=2dU&F`og5kQr}0P2>9r|Hmwt65|FUan)bs1a!YC2ff4GjS4c?PSI!8$HuQsG& zf1Bn$PNNBHXY+j;g~{i=bLbt>{UmUiDs!O66h0TrBr<2&^y1wagzKY4IWK$|71^s) zNBsy@{PmRhco)&6h-`8`WC8sY9Z$W(rSU!=9|G>rp!0|(_PHcT+P1~P9HVJi{+<*X z>A!-yrKizF1?BX$q#fU8^^ z{8xT8zyFsX|L)L5zMR8yrfSlMtoIA0%a=Qn`~?qLjgkGw$>H;wIPFBMXD6wgzn@+# zkE3?W6!{~ST8y)1A$jH4!IXcw&8u_@BYvqCJX?{4==wfqv~11?=A!ozdXsCyU+XbW zSAWhSt5kIO8AOP$yL2HAY7-#z>^5@yfGpmfE`?#OF?tia1C8yDLBR!$Jk@u9@ZYiT z_}O3-hJQYgcRwcRw^DO@J>C*pyf}l8R$$^!mHDer6j1+F8vJu*v#ElBJ9S?90U?hG zs=e5WrcB$+=i194E5B;A)5Zy$8a$eJVlND!;VPCYiG8&V6BF&}$%V0Y|XydaM4;Cc6hM22S2hUFir(DM?iVH8Du zAIS0L-TFz*EJTVAEM`=BkIBNS8uUirlc^B6iwXq2N? ze@dl6A>$YQTqMd41zf^FtpYA7HSkIG)=PXc}yN@IqgyZ)E!FZ)( zBL1dxn>4LDK1h5&XHm@RB9$ZY`@1~4&!%?DAHkTuFK93x(5}=PSiBqvl z!%*N}B&u|uM|IyhlS_vwG(U8fM%j1J2a8Q0v~VuY-DH8g{@uj+>znaRXA#=B zEsTuWu(@M3cV5RuF}iifQM$J!kv2Q{;^J-3afY=J*(mPHT$28c;tMul`z5bQY1nUa zMn8bTOBds-m*la}-UleD_b}SHR2>;^s%PIlS@gj9>GZ*|GbH=$Jc>dVl7QAnM0LYm z-0c@hO#UdNF592zu(1c45_csF2EOrL&c8vPjwFLFw*+d-xlga_$IyoRn zLKz!u>F?|)O58n23rI(@YDakXvTK=t{;s^aW_x&Yy$VS4axjfsp+(p7BXJz zelsF+EvWp(eQ4iPRufrmfp;9*NB?6r582ZEBf|e?I8|p1o4o zb?hzDph7&|<7bhWmbj$=_dOjVMEiXm_ipq@gW({I7QAW&-V@a4LhwWtFiY6{(lhU=vpwpTt zQnz_Mwz+zflwCC=>zy!T_T@aZFq}g;WzR^-@{e#R#+8}4`V9wG>A-m{cgWzEaVnhU z&D-)`h<~kWF0K+@fUek8l6zu0xC5-?Ws8_GA`5qdS;txL^W!@7VyPUhJHqN4Wuin` zsfT)b2$Lv#Q$}@cbZF(`X*5%ZRd$XmOp{VXKK>SGpBdBib+<{BPYkSz8bBMO1&Q@|eN8-? zK~>AFP`@1H7rB=*PLn)T<$VHuf1XG}`-kYVpjrGz+Xl{Hzc;8Em!X4Kav5i(Z$wQP z(swVU(ERTc%uh#2RJdyt{uOCJ<_+DnU_C|=t0Kv=6Gdd>@)z*vuQ-IYQjtENi+uGk z!I{xic}Y7_uUx?BShAT9DFOOLN0mH%^$ZJN;IWxpHx{<84dolBp@6y4+&@tY+=>?n zZPOyGhpd_WE8GuHHk^lZzMUbn*)v-3QGZgUZANzHdK1;=My6xOKk_H78~OLGK-}{m zd6LT*X33$`%<40-wEajeiFlVt_a}Ccz?f|WWa-l*%{ky1+xD03leoQ;doS8<8cLK-DN^6*Gl^k)Z;iyASu}-ZMhom1BPteqd39Z4 z69rF%TL5FE*(Hl8V>3AZeyS2Z%1 z)VoMUlP=yRa1k1Kf$|(l+7e@i|E@}BjLkXpmCbk9HTxvm zf1(*ZD1vlxX)SI3o<>g_d-7wu4)g6UaQJT?WYgZLA|?wZz;_D|&|G&3lK%QSdiTc| zujPsW>7e&;dQk|7}%JjCsa;HuAFz zMEIE>=ksrD5~b-&jOg-Z>GVb0YhWgRfqAk$pUhQ0MUPEBrx$KS(OrM}1 zKGLFtH9O4F^`qmArHCD!Gh~JzJHAFGegBvO-+cP3DsO@f*2Hk2^l1<@2RzJ3Y(udDck|yxc+?^%IH3seWkH z{0E9%utD+i|54B3R+4y1i4MQP^yPziD8FDkb5<^jw_4&N9M!VJ8)}Vk?b7FPulHG$ zFVKM0yCrCBbU#_(^ciJs9w7_rPH}vpJ@crm3XQxHJuZ}U?Ww%Zr_ zT|*@L5|62FlSVo%!u*ZGD*W_g%6y@<%izJABfw~u5_AhGh3V5`nQU1eLc3qwf8{Z-Zj^~Ml)M{k$Mw^VD$)P66Ni}*Oo>14h z->HPX9!2Vdde)e!YonG%iU)dD_rxRz$r`I!RyyPw1x8Ofi^YaIp;;pA!Zk(Zm zOV`jrkCm`4P=Trj`qDQG)KF$!8n2}I7EDN)LtZz>;Ns*2^rdM(Q!G9gl}}E?y~|rj z=KLqr^wf50m~oI%yHtdF6w~mED}}^d{WkTJh)4SFUP$mVLC?<3!?9?5sO>A@dNsndE@tS6pc%FDY^4d`qR6m;J|;3jsL(2x(fVBh z^zIdt(UQ+>ulseVwUO{1sR`1hpKl@up9Ic#>crfbn~tpHrHHMa2{t-Uz8Y00VTNVX&-s0^q4vgZG;D7Rcf9UxzcD2J(@1(2u(WzYUEO^z?fJd zBll+y(p}*M-v+qyd__t?kT)OZ;t14U6h>49g^0G~+5g4RdH-|yg<+hsWh<*9LPX2> ze9nD7>RTa_ik4)gB`w-%$Vg=+DVs<{l971MIT}U@m6TFSlZ>>rM83~I@ceRK&wb8) z-`DlNQ1v>0_;Y41bc(9rW8#0<@sBSNZrTF=yF20A$1&)Rn;R}mI|60@rQm%#j|x=% zcUrby^|9(mm<37u7U1~IV)*qg3q(9x$x7db#Z{;X>*9s^^7YQw(B}fG&l?w6e8f*+fjH@UnkABE`#o{ z3`lWxpx`piT>2^BZ&XC>cD4-gpBWdNca^zfcjkBysMW^2m!VJfEpi!+1*RS;;+XM@!b<-zr zW*6Y0b6)tT@nw)z)TD)L^Eo4BGu*T{m|L?!M_6E5PeT3wp$iXfDs&A-Qc=k;JMAuu zWfZURI>JfxLcmD;W~~Q4Buc}U&bBmH^E~!mEr!(_u3%Z29jG(124zi9p(Br*;=h3& zbiNkxU!+Z-%J6t0=XIk+Ln# z_(uICyu-F0q{fHQt);j5e864si)R8C9!lUk@>h^aw^HEU2s`+mavZ*LKWNEjAxN~% zgV+9liSk?xZ1i@J+|#SYdC@XBnT!^_ne_n%%r&SW^C{hOT?31m4dM>HQ;;exFOt`N z3Aw$z#_YoiLDV-5p=s?FZbi>%K+ru?&0${7M-_#kfs7RR{)GN zGN7qfD)8mGMs(hrFQluz0RANn5cynjm@0XXWF_$XX_FQ3VoNqUx5pAK$@&fHH}~Pz zCn}hHIfyT}ErHDig~Z9fK^UGd#!< z(w0t?zE;!5&)GCubwApB&YRn3S}b^_be67o@rGW@P^JT)2a(!^k+9!KKocy=g>HW* zf#S?8l7kIkuE|nT^s}BgywGK)S59KvE0?iVMhvR=wnKBH;sky!7BseN6wgOm3tlDd zaD8qJ-czm*b{p-{1mPMO6&yuw=V1bwvUGdEpPu=$wZQy5ZEz!q7J!?1lS$&)H<9Q`UC&zO?w{1g{ zer0e@eWfJdG7w$JHzD_P##YGdyryAA3S^MibMA88hKu#9AiVq+%A36n4J^##bXIwD z&ez}JoR}0M^ZgO7Pfh@7Wlcy+RuMh_v4H;TSdDN0GeP#rw(xU>0WNx=3r3o<8kBvjoHg_vt+-g84j%*VO(vOmpypHcscmiOrJ=pbtuE^S@ z7fu~8!uJNGM7d8pQEZAl4cZyWw)KX9*%d|ZcGFR)Tx|@|>QB+iJ(tl1_c-v83qiqC zN3z|g&eMw;o;c3qIGbpkjVgMYudaQunR;0k3mrw-WNpTBkSOQT`Wdlw?ICG6KE__C z-Kd4zH%`Huc&2;_ucL`ikEMxc83}ytAslWoqRvI{1ZUMXIk8PAq3hc!?0?@21u=a% zaA5}b$87}$}BAiSy( z%$qp)O@sKHXiut}HxtP9Ftq2Y3@Qk}ilo)5arEC>e0^B~UY5EX`v&j9mUbTa=AX&1 z&iV*`aqc)wa#O}(x8m^Pb=kOUNia@uT1+G+jub5(a~19foW|Y z_#Ia%#tW|F@=$5f(!lZ1SCECyOFs}+5izXtb_tHszKz!6ls!+Sd!J;Hc>#}6RnhH=b2%{;V{R{pNnUFB#`|`m!I_^z z&GvM#Op?S8|1r3|ZxoJ8`zxHsb2;|!dP@B~o$iFuZ1%l^@|K~;gJ2^vyeRzQL zmX}B2?6y1QN(K{2?++h5G-!wUq6NFQ4#V5MyP>!SaNxuo>~&5H+x(pfyLa@!q$i1_ zh3BtY1@4B))&6wJ1VpW0X~X4)V)9P)7=GTe2*%VXV%9#2-utngZfh8U6@P537-JTN zPCwgDEAOu*J>`=@{+KwecO8d|o11WlsjBF@_#Qlck_+&J6&CW)5uG*8?uO^}#=ik7FW(SdohMW*^b%A1=*%iK%vi+FEGoU&7Uta#W9oD7 zkp5a%LF$B$^i1MWw0M#|nPKpc{B_h2xg~y~WZ@lHs#t>egfwzy`NvqtnOT+1mlZ0_ zO`g!w>m97;=XG{(rV`pIHVt(jE)(qG`7ePZ)H$7%)lAR7jGF%WLg|)Tbh_V??n;?S zzx?M*!W;%j+xn~2T8<%uf$y;6dM9o8;mGRN+py5&LE0?dBHVN_mp$EfhWaUwrcYY| zS3Or1>6o5{D@y~&nWsmffATa?{ujgTzZ*=RD!vx}TcA!{Vv-@$O9N*f?j}7`UFn_} zC$`ta4~8F^!whe2rk9tFB`mUNi&H5p=Chd2AC49{oehE&vaW24^8oX{@5lz9nbXz} z-$?xKV)XHX81womWW`@)*oKG>rv2g`bB>5&J;aTj9GXWn{+coH4`6Qp@>xT{dn$4r zWai3Vv?D@`4r(R9%4Zij58G4LqaQHTmKskLZ04i+t>x&@c??t{99Fns9OX3=evy2b zxMz@CQ(O!|O|k3}TTeeopQ4wS45RYJt#EFK0$dJW3FqEVhLEHKbcc0kMN-pURylb$ zo9-veh`uFG5YH7%XfcJIpO3N@%u)BTGnt;8A!8EuRAp5nRzKXuE&lQqZV9%~rv*i5 zQ@|(|TyPl0w@rhn3(Z6(KN9V^E6a3$onbB}rEFG02Guh@L?b5!(nC2j}@qbSj zdex!`^J<8St%j?wG(}1ujM!g?+4PldF*m6%9x5jHaN33w=umwylYZ&PBz9$yf$}D( z*FFrHPN8(qw*^e(Zp-$cvZtG@##Oome`X4%pV>ccLw4-@5qc@e5pEC8U_+{=2Y+c^#;76bvNDe#hZHH*J9<*+=a7VzoDA%rhs}T--r9=(Em0G z>83I*bgfU9bq+4)wZ@BZN|z(ei&y6T;u7pXQ5<|;q0PQ0O5)D2PHWxfd{S%~LK8xc za`yS5f-e#Jl~zsmtlerUQ}6I&g_E6VZF#j|x$`1+;e|R|7_yhnnkYnlFJ6(B)Xliu zX%Wtya}n>e*)7zXqDXO=F_Ty+yX7&0;DvIj~}SQ?hqQ9NQL4X z2goitP5jQg0EETSOnB!i8}HsvvbJ{6$Ae?QAgdC8;JtGzV&d?fe_}X1FOl1vAf%6P zjU<=xML0RK0t}~Tb1D<}K!1WMc{(MGWUYBZs@sRTm3s1=XUYLuU3wWGS7^h2!I^^d z%k{Z+S)rWx2X)r}U5TvfI)bI=+d}xMQm*2?5!?1Jom2QK0~IZf6;WNb+>8#M&l+4S zz>7>-&BX~2rf?Px6(_P2`}gwxf?--PZ34>tlYwru#Np4gQ(-uA62>y|R6^J%Y^u4# zE%KPfC2rnGlKsQ3E>E@+PV(3a3*USrDqTw8=Y9|bLC!+4Mtg7@8(k51_q%Yb46pmX z*nw`UWYX{h;q>477JB%#Bx$=K$F}74(eQXZmibDVjZA*V#p{MbmRlLEOSwXGo{V6j zJ&V`^sfSc{)p)l*pVrH^izv~sAt$if%_OX z>3}VI^ym$tjw49O;7Iy4ln+Mn7{Ns(7||^A5A@q@b^7pc4Y_p36-bRH&9Rjdb@$xG z_ulDP*LJBhHGbZ|zCBa;{q+YLeb=3N9hF1x)^}9cYI{{kIjPe5x1X~`3wJS%a3vZa z_>ERMl+nY!ztR1+AUN)944==nf|Qgm?nMbu8AjvLRm$-8YcO8u^%!4?m&bnb@uVbch}_}ll5e z`KJ8QoKJ}yt?sxW3_q*NGA8Y!^Y6=3XY)kVyzCd6Flq)VeLNAQuAaiyK@OtqBzKYh zuUVYrglZD=@hH4Zj~6KEo+762bI8UBDbX*dU7{z_aaeV#4!aa%B1|%M70SH9BG0L2 zqNi)#;ejSCI;v8Smb;g6jklhVgBQP)`~DrvMF+P+VR0wEc;A&O_^9B}(```BeIb|B zY+=rQBhhvp4Uthl=GhQCU~cYNs#SPMDEB0d^lYx7?k_^<;O0WkwC^uzm>)yl9h<-@ zN_q)JX_mNQha9nPQbPL!45+t_Bb_9?$t8Z?N=7<}(EAl}B*;vcn$BxL>Si37?*58; zDIOL`&fWrN_}pe2MRnr+=?Q5$u?`_bbO!s(_+6&k9iLqGM%?V3VUMwX3zmbE_*M@6T!XjDe*pV3{;#&Fon-<_*tifP>(%G zxx7TXU5~GWTeK@ZxLvEFSLhI%v(&^{^67fBg zN#%zDSZytVGt*X)!_WSqKN}vP2NAlc`7+Pt7;6fX#;HS@ZzYTw>xkV$;ycR8h8=s}f z^vDSKcx?vGow^cN*VIGCmGxwtS~qRd*#&P65ghOF1;2I~l=f*Zn;6>xF}|bGCuE3i zN;|EUe#U`nP8$B)Pz=*@+Cj?v4LIDj!@blB|Fdfb7r9KxG)l$6C6V}H;XbAq{={n9 z@Ga6JeF2?R`zh2iRHS1jwWAZ-eBPDkJgmE48Z9u*rxGz+K`^NptE}6F*Qn*gp-m20 z&Mh1MhgPSdy62a)IK^ z-!RFd4o#l09Ga<-$v=Was;uzs@LOnS4Tjb`)u6Oc0WNsfNFx!UcHA zq(+#cy@ji(RwTLUcBp3UZdz>@j`z2lpQN4d+cc*i0VgOX`tD`(-IU7e^V zBN=@6s^aFJQAi?40Q=vpfqIpVeBQeqaF4fw_sQ=lt~QxxwKPEWg#$$WnTRfI9wyp6 zzhkpn8ax{1FB~;Kf)cFN^O1Dr5_Jf)I(kL|fKb#%LXHOM)LuKa~f#Okl6yRk+ zgU*i??Tk8%Qev-AhoEMBSy=yWZ*F4o~=`+vrsp-kWW658zTxt}a+Bb!|%^AX% zW=Y~j&gxu})u+-uXLAKE89N|oxRZqpB(sclr)ckn0W_=GUErxZ2ZfYh5sG1X&O!Y& zr>#{@KV@ts*~(*t;#EyJQ?v}%Z8N7!!X(k{HTiIW@298i@y6#f%|&t2KB8YCLhP{1 zfv6e8qwfzbh=So2;dZrL6lh;rF-Ph(+CJ+(@s6BD&pWu$SF{e#HW(su7fvx*M-j@h zj$`Mmw!kRSSeDdpS}7%{VaDG)*oNRTQsZ}r+6KvyTV49(ZRIw>s(a1!%=l;&`y?7~ z$#{c4SG|QbG1;(u(RblnF9X~uAql|(MXd8@HQl|U3ssNQW=o?JMDj}->8LmI?8%FF zoKe&^w14O(`uj?jjpunYt((Lu0+P?5`k?#pBz6kiX_<&4tEz<;S3l&&ojifmeaC{4 zy%bK~YQttOzCkO_S+l?!N0`vsiH0uTPY=v~#(3Z^D!n-c>&<-#cXeG@=I}Ds@OBQp z;j@K)3A+#R30DY2Rv=t9osTUj#_#`T!$tWi%=n}Z>WErIgFU9B!Cjp+(zH!@Ip8na zF0&r)N)EuMH%F;0H>-_gB)R~LBJo2|n7d}+FMf9IZ;-_XZqN?^*e8OriZM~*LRm=@w z?cIgc_mKe?+W|-8sYSh@wmnTKuTr+ z8Frn79AO96Ykoq5b5@bSJUv#*9f6VNZ4fMcXuTtMELa)O;S}OVk(kv6$l#(WG0S-_ z__J;+4e~sK*1niQMbF~UJ>N^HB!dGhEh94N*9Af7Z%xv#nFnZKAI_LNgPr~@#A_ui~u}*v;opVK%spiJB;RDI+PS6gLvtlLLcjPX4{C+f+u`vWidLA#` zJq8>e&)^OTTIh)>Hr!EvC%EPv!Tmm|$8?+w*`I;se2zl2GJX7!D=LY z7^%9%U(SB|k?1eYeNoFo{D{23S%R-vY#GC%qckKqe;4V!_w1LY`78y6)@B zc_b{vPs+}6Ytwms&JgdNUa}FMcKE^bx1N|CX=KagBw$`IWofzBNsISuwAHbcgf*6f z=b|ZOyA2|Ll0Dc^QY&?gvO|Vbrqll}6<0K?J_99J0r{$&Q*nCcGc+gJllk1=$nFK+ zr$(*WBr9nq(;B!+7rBn8T&=#C85N%-y0tsl`{i4SboM z#heLK+i2^%_54oY34K|iL7Stp1ie@1!|T8nD6z{SIpL?+#VP*m==bR)>e&(^zcmfk z&Tpg*F8v_ADOmV^)KpN{-Gq-fg_xR8|r4nyq_ zcuR&oxS`8t+8ku}UWKqIgD_^dJfB2sM$lcGkJ8b6Up=Az05J;B$Hh)F=<1#>TJ&Z! z7vd4X4*dB>maURxV>xlyH{m>;(;;E>`2K#OHM9Jzd|^rni;{CbnqT6>wqjafo09_Yc$!&gDOBAiOToC^Dv zL__nR7;2{ch(@o!PHl@y@br5}aaq0`j(&3(S3UbdvnyS|z;`noiTfxRy7CSh0-Ipz z0S%@ZEbasr+uy5L_P7Gwi=7OF?yg(pU%@#^a9Fhde!Wwp^zb|_T1P*BIQZ_kK- zYAX6S+Ysxnzm2MEBGAyh6bP7659w1bBGZm|xIIlA53i#F&ZLK%4eybKzZA(GH=pcH z$fy7M3aMCDp73Lw0yb+7W${wWscXk&IF$DVwH_}A{`B!I`DF!*^FbRSb@jrwQNxGCZ|Kjsr@Gm51Xf6qaGJL@aXI;e@- zCLNB%Dge)>#uydsJnp%160Y{UjCp}-q51lAJ9P+77Y=8cNLiHGAcQWmH8 zro1HeCPg4?Wh8X_{TM}-1jDtsNNo6;=lRbtWMT0^sC&N|UKD>G>{Y*mQRH0sGa-p4 zs>|Y$BZBe1caOkay@JR#{RH|Y8VB<}F#U$(BHSqI__A ziw(W!dyafGJBH$$^h9}PXW${Hga>*wL>{-qMDKpxB!=!9xTz=^)otbbJH5&1!EIeE za1X+2$uF>^sTI_0*+c)WT?Ln3g@S0(3y2;)x*}j=5iF9Qh-+`g!hziq(0Hke_B9~# zDCn->(V1?t^h72^71eO_2dm&$)JxL0sDZ5O3Z!FyouGeetUzmcJRXedLaiEVWL4oi zlIM3FtN-^1WP6S1UX5j_a87rH@%gPdrLYWt8F&HjPW}MrHhHROa+Qs)2(7BUrz_8e#De&1x@x~q^_jVGS%%E((y3tH6`U)eKRwi0x# zD+}xPlW0;`F+MFoaM$((eDBl+H_aPF?DKA1p&E#5TNRm6(p+S|xn3C18BM{~5Yodp zK$zNGdM6?Yp61OU&uy-8kCt6Q@jdEf?;?5nv&9hX@o_vPn1UB2U!>-zTj-jjb#(XV z6)?tUA?f+BgnW0f5?)!NL!Rx<0QV=RmH&;etMC}>3!zaNIP;e~9*WkdNN z?Uqc#gi9+#wtL-QZmG09}xM4>j4L+-PwT zJd%h{KRfB!z)Dz z0;fN_1V%lhz^w5$`Z|4_FlLd7NP0;TowcD7>6|EGk5}!7DdHt4zHTGS@6uvlKOdye zmkBEW4ocywsm1JGRVMoHPaE2>FPO^>Du4xvOX)1BZWPkLKs2jxgs3w3I9BQo1y_$k zsFT#gWBpaY^@TKk(v%A=rl;xqKSeluRy`_^Jcw=I{RYElF(}Yt9Zan`3)@z?fZMPR zb~bf^HE}%4_n3%2EUKjo=T2f5Tl>+5pmI1@*+SLyO>u|y8LCO~MB0zgK$0;ME05tb(3-79ijtK(OW!i(Amd7t0(0c38X-78zSsIwRT$pCIvZB& ze?Z&oO|jhDSS%aihaJNi{Nez<-v0y=?X~0|(WlG#{6kGX^Z5CrwNP`}7(%1x;FdxM z6uC1P<%IRahY_jZEf8Z9?l|H4ulZmzLml5t7>f<#?D6Gu>*&GAw`hWHNX5g^G61Pn zS5M#ZLJzd&z>eA)QhB+VEAKZFeEgIljQ@d!t!Lh$NfX|Y^eaN#c>g*2NO|4bC2z1a zmZ#@*w&O8k2tDZjNy^r!u~oe$Y{%Be^x6D!?7Aiq@&hE`U&LbQa@d4NG?t=c+eJEpU6P4b7k;#8{Wb$&OsNKhMFg+T= zmH4Z(E{l=ufUGQ?Y`YxCEk8*X85v^xISO>uc0|7#KOskcZ>Hn*qp+1&EPXJliIZ!6 zQSqiimMc!XC0u)C5?%l!Xj{NtG933C?A!^kwkAKwjr5SsTJE= zx=0JpXfA#B8inN=abjU7NmZQ*^V@fp?pL#8+hn4IZ)B6{h=w|{N&O%`=PpUBdT*dz z7W*k_4`q7dF05sGBR!wi0Ijya(c2sT#IjeDd5buB_#m4q8EfL>OWg4>O%JqKX9OSi zmIXmM=RkXE7CE!Ooj7MD!IU3fbl#TR=+1=*+?o-=j#{s0**tq{b8P}-e)D7NTfMrEd$zs(3XL0j3^r4NI2QU5(dszX(OX9i6f9VO;dz#+JeEbR ziH1JwJaBiviS%U5;ri(?x;)?^%-DGfs}1Sk{#RmS zkcNFtKdGsO5sUhH5{}%>!-K1r!JUlj#ADSjm~=f9JlB^p=WWJJajQS`4wJzP$4T9UOOTzcL& zmFALNG-II!Q~CafhW9-t56-0scE7p>8DVVE(dFc+xjeN#(uc?oHk32}P);Tiq9op11lwgY|LaM)SHWrv`L($V@ax|x( zlfO3thwkNX^d*Pbg*aPi+BiaBA{RlQOy5~ib!a0#_j)UE~u#bIOfouw48z(yZ}eXsDi>rIA{md?>5M7@jRvrWH$zsqP;+(c9Pq*wAV*_Eg9tO$n_?^2~L?OJ1{mOzoF;Py%P8T<_*A^44ac2dZ(o+HF!)B6GOUvl(_8b#6bJ^lHN%Gx03TMA^N3UMKsCfBYrLqJ}D~-cHF^ksQw9ZC}73G{|5@!u@ zg?%pXp^^lR^s%hRU92+jjUMJEy~IDaK7l=BFqg5>1cF?SfN{!uf$U;6(aeD3ux-{? zv;y(`kIt>kd)`?#&SxXf+q#IKPBDgoRuTO=U<>wR3g|seejhX02cKY)_~lC_oS?o2 zzW5x44bKEf+D(=nlx?66qcwO&ZUo()7z}U5E@YOKSxojA-~9bjMv4cDT)pmFomE54pMQdr36HJm;isa657o`+`!k2sUUL0s0S=$?Nx_bi@fE z+)S0hZ{we$^*22s(<77Bw*J6Tnljif;~Io!dsErmI__`nVzA$sOWF6I5V1s+e)-GK zevj*8LHt}c)pk02l|C9eTukZ82Hs0{NrE(fNG>m?Q4b!_?@y1?oXWk_uKzI%ni;}~`5~O! zyRwR*EjQ6?ZY5rMX%DpTOecX6wd9`s4E}w|NZYYqY_lMgS?wXPS~mx7t|=fVUe9F3 z1#i%V8BWApAORb9JraulkrM6f4P*Y-dxdMv97urkVshp8Y9j4zCrl2R2dmdJj`sfm z9r+Mkw9yHlZC;G)uaCn9pPX>~i!?>Q@EqSfCbL0-eViqN-q`_kU&|?c!S@za-kOO~@m*G7KORCw z;>?Q|QTYw1GHtoP*X4t*Muyg>6{hxF35mA^>x%y%9A>b^`^$kMf<@H`%Cp7512fejhU{cpEm@bnK9iK)+?4Tlc z&zo9tMlp__csvVD%)5myDfZG`O@-h$=s;bUNYaUwvbeZsKR%$PhSmJ;lX+P}G}>th zwi_LXX{!TK>YDxNw!$QY@gR~_vmz(@`qAqpSF9`7e4_4w{v=>k9dW!Ejr?8T(YRk% z>EDBH zG(cv}bNcB1XYwL4mFIwpz*4!7<1+rhmzfyfK2QoX9;wpTBY&e(`)c~seH_a)MfBs! zSST=jk9wXv6VV?_G-u>0zPF4=k6#agfB$a$daWei8x+9CDNC_-Y#kjVnhd=a;;iHS zZ`hil04vsL)045Yz^OR~{tInGeimwwlxNP#%#&k}Urr>yeh7uRX(qJm&`e(Ppb2WD zXXC4l5@cYf2DKY)!fu4il8^c=u=sZvt!=dueqWjo4f2kdK5(MTl8>Sl)ic0k+CO2S z!XS~`c$+?X?9I;C^B3~tV{CKIZWtmf&{gsl{=R7E#-{OJoxk&F<{m%xZ*LdPC}~4> zohD!?jYx~!ZrbH>3ZynCA+5SGpwVjrS(-}do`HzhO-bYJCu8BqGfaD~4C?b%s*M9|-dF94j!?Np~s!q4yo*)9Ow@Tu<4<3P{F`}7ASD-0xf06403+Ns@ zJHlM!;hlpc(KK)cb5@Qj22#0?6MhoMwym(qvXB1jxkfciMzKTlF*=nv9S<)%0UeKw zpy z9wkA>r6|&A&6jBW=dpNGQZKX^JL2u<16l4%Csuu0g$^aPP%YP)bc|CntqoRWN{yjd zqgt05R4in{v&Z4_hb(YT_2h-Bn?Z3gy=Bs9< zU#}gAt&L+@xv4lOOof#4bM*=TG}xQkIP$c&fUaAXfSiNwBCSP6OeX#UIksdpv9)%f zn(DQn>^=i#zOSU3O5#K*J()i0Zlee$vb%cU>CwzZEZjhoPo}si)XxYf?W3EK+L<6C zTXUW`UPvYFH%!o}r9~+H?*L(5d%2#IfyDCDN_cxX1eG)!Kwo4QgpckPba|wc5X~6! zaE%K)H{k??8D~&7bPiP>FGJZM<>}ew^3-U$0~J*|vO_WM_+aC08kgT!er2l}dwFUp zi_-i5&XkLqnEU(*E&* z?2NrB)AIa4|JxNnha|n&>B1r!ei8}ij(o(e4~i8|y*vybe^_!;4{FlL_0iOMmjW%> z4peSwEZum1BtF3JwfDvra(Op4KxJVCY9A9U)Uj@$YYz59JpMr!Gk+LOkD|l06B+cV z(8Q*%G$%WXJ=37U%bO`wm@Wg~o@PV1-+fqMJqdsFd@h{wD2?o%7n`=@o(J7O|+Hd7b;mwAXT zJpY22HV3oFz8&~T-Wxs_B8%2s>!J_y-jhpnUD(yKIOa3v4fEaAg^rG0LHB0@9lFbi z+M*gJxYt3gmj0tIr{=M`C3Be5`kUOk)pxnk6BNt$%~{JNo+J=kO(&R8B*OxD-D9wp z9rt$gdv1Pe12?PlWW}4B7+#y$Li-IyvCAdu@Ytb^G#TxqWo3%Az&44>H;U0SCac*$ zD;DMzy+<-%?*(vZ-$SFv>WUcuv#9p$qb9X_*Q(ueJ+}EUJ>7DaTxqz>P0|+9e6Q0Gx=ov{=@^1isrPX9 z=O1!);6K_{6-K3)0=v~Vfqkk|!_j*y=<^zBx->3_DQ!34u5HL-tuDWqg*>MA{u%6F z1`tolNyscwhDmq6Cw`r4Xr#|A>X7x7t{ik>PIoq=-OBIDlcGMZ;CCc@J~NqV#!Y8S zF3S_U?Mbj;`hWCqX*ARJu!Qn~7M?M%4#_8+p>sa9$g-usy=v4=)3JI2P|{{g0QnKZ@9gw!NG1Swe~EHBec18!}=lgwKP@Mpep z)Dliwvlj*TiShZnQ{hpZh;8^N&g|BIC3yyMSS>(A%Ox5~eONUa2o9iJxeAN5uwwlu z&cfs?<=l@y$*AKHLk+Jdg6)G-Y;#dAdn8xMrm#SAI=&dvTPKr(^=TmKdXYOkjz^&B z{Xxy^?t{Kk8u(7$ zvASYq=yL-eSX@IcmqtT$q$L;rcm&>Z;3_1ua@c{3mBYh_*j88_Y+qlkQUq>!~#B^d%%EQwpm# zOvh3=Lekyqfk))#l0)NqxlG?7&`TKR&TT^M!8INHc5)~596AWEsyMKXx(N5~Xo6RC z1eQCj&SwF>AZnSn@QiFRQSX=_d@#lsUwtmX$1k14&tqOdZ(Rf)BN>I$g+ovrz;PFj zR^nOng+G}CuS|eT2 zZ!vq`4|WWz_bXxj#%!vx_zY9MDvqB&89@Ra!;nGA0z4~VNo8uz#LE0aif^QaVAC`^ z9Cg49>zqE1{2CL{lfmzJbEJi6ZBH1s;OAx02b!qggF`rH#%c6U=Q{Gpn*pOw?7?q7 zO7nhgw4fvsWluGxHIIoaDVo z&Aai?6TbiX^9WY{L3DJ+I=1h(7$i1Nr|S-i02dC?6$5tEY(N!{ENOw^^k|&(Lmyrb z-yl6DJ!qDtILr*_K&eZf!T%JUc|c8H7lu{Rpi)YvQc;PhkUD$qo1uh|DMTS5^OTe+ z6b%$jqCsgQA(GO)d+ke_^fODO3`J4NJVYqI^ZnbOw|marYwh=ao_pp&SaxzKNwM!_ zCaU4^Uz#hqoPAZAs(gbzrd!z2Ngu&DB?cxOt%dzxA$4keN?&*V zplf2jld&K51vRlVx$X{tL3J7oi`S-abd!YHZS(jIwB^TyD*8C(6g?bjM(s{6P3Eu0v@?zM1!qsKEU%Wa@2IZTJO%%ZDbT!POBMJ~ef7mz(56xV)O zj16%LblE9YTB5&`H5Ts1;vtTLg1S3aWUD|A=EifIp2LjXA#`Y@8J5=9vD~!d!lDsx zp^vpb_@~W;=5G1q+oD^7RbncXo%X_Wv8sG(R}l?8ZY1*2ezUP3{K@Z(bF$8pQ^5J! z3|j4IMR(X3f}J?GTg}vi@(e3jAdN!31%rjUmhOB>Wf5~L=tT`C>C=h4lg!L(BMp&S zWVL$}F^w2upECM{a4m8@7zNB@i?dG1j zd@Mbyd6OPA5gkYlJs{Gh4|QF50*ifa3zK#{fr(Gn@qo%|4D3?jA67;2hR}X^Nn05P zhIfO|1s~w&xB%J|aSfl0+6+22Eifyt1(%xt#P=?`JU~Ke>aJY)6L%O#DJIf`o(|Ns zcLSltFJ-dm9emNLemrbR0IdBhgE#LK!A@|Qn_B_ zR2b03i~7*A_y%a6p90!LtoS`W@qM{T#eVbb-xze{B7WQy%f^d)J@ZL>*n)a-KcnqR zKm5|6UFBh5EOrFDe?2dloh}CT74ew&!y4b;Ph%(72a`U|jx6B55?%3J6>fGtV4aco z#hbQixK`-`w6j%0i(5Vf+MB}A`5&><^&-qIi)7PxuEKRKTgB(FRkr@J69)h4Mg402 zkhXV;L?b+m6;0^{yYh$Omr**jf1aVpK)%S#U-aP@8n0sIX$v0y;X2O!JeCe}T}7gn z_2fqkLdcQ3-q8QJ4Q-q89Bj=CWlnSA@Za|{fXxRX(o|J`Xv|w7uilLA%bJK@WwCVh zzkM*ItOwz(&jiP}3e?whf?RReBDgd8HWn0S;F8m``Mgt#B5!OL8TX_V!iE+>H^?RW z%D&L|!7u1GZUJqG5bwvG-jV8pnGkdKFSLh*!!S8`Yrfidd;~L9g{GttXr$H^0|9eDUYbEm^ zW*Z6Z;xv1>F8?+$1{R$vhJ{XL(n2j0y5QYhoM<7kldrf^v-#C%VJz|vgU3O~jxTs) zz*BJRE>O46X7tMDbM$N3Y1qH*3ciV*%=;YP1QPjEzSh$Mn!g3Z;H%zbq}?>Sf3^Z` zvnj)$=URx>tLyj!@}Sls9WD+`hNE?e21hz&%Kcu#W20;6X*3iz#$IL_r$Xp{`!F){ z>v)#i%^2>M8Nl+C1K2zI2#y*OFRWYs5;nF3U|+Qd`1HSX_~WTJo#zxG{CU>_Hi7q0 zy~;oyH}x^kI+q6q-!Id}(`qP=x8w)2tmvW5YScJ?v+RwEI@#=anFn9i$Exq?QV)?` z*=49tymD(;Zc-@bj4-6veXMD!YCRm0tVW0FHt^?y1`RNXfot}*P;Fqt$J182qi8UG z-;_;H23)1dwiBt5zd7H3sT-XfY(krz55uDlKM20*P6uSaAosV{un%|C+04{5+0KGH ztdE$}*i;Y+Kg+J;VOvAG{Za*4s`f^(#@{g1LjeO$6=9v@bbcmxApg(%06Xp(PIZ4| z^PLg%A!)ic9o4jr^%fm|<0k}CGi7huw(uhfTKgRW4?Kg(b;fu&F&Z)rdvK53$1<-F z1^6;75uEgn(X5BvY3?@@KA^iJ_x!e)CyT*6X|G9OkoyygCq#hz7zJ7zI1g^Db)$(D z_fYXfjQyF(r`g%ufvnaihaEnsLA_h{umcI_1Y40ow^{T-EKp_Wzg&ZycP?dZ-^MXF zV z*igwga_-!6rtzkdJzv|OeNZwG`d?18Z}(RrpAD|DiUNCH^5CAZeo_cbDJ+GBnfruQ zcH`-c^6B={%LAdwGlXt)?oONP+S!{oF2teA2D0`Az?U{gioE_~(($~|Z-W&$?mbNA z+&)5Us&A8d9R}!En+n4ZkCu$wagSD|novot6f9?l&^_tF4y4V zrmg(CSx=rk_bFS~c?0sdEr6gM??{GI8ce-g@^W_k9LEJ>e@Xx9&p?-Zk*mX8t&B_9N-0Pub$#Y9j1;xEH$r zI>PFkkD%iAH1Kq75t!>TB71unEcfZrcC$$OUfg|FFaHj6#tndRBWJ^rvah6h*nL?2 zrx(uc_(Z&pS%P8EC)8aKiKDiDg|}7b;HPf}j6U=TMonEv|Mc*Mq3P2>p)m(9zUoHX z*G%QneJZJ!?`PUG^B?scRza_LQR*P>v=2<#jms^{v8gZ}v)5?S^$L$rd5&mE#1*tz_EW;ryy*6S2ZkvIJcRNRjxlneTMyw}vO6 zF}E9T?z2_67Qa44>kBeSn--?}ZMrz61=?Q#x`Xh81h}^*2Bj}6FVfb$M zaQ>~&6!yNKHm}|u27TJENzaGeg&S{r5xq@Dbo`)xG;jSNx;*W$?4g=7ZP@dOu-+n@ zWSlWQm6ZVtkMH4Y^vm!{*ezjO-@~#}lVHg9vxj!i^Td3~VsLamOJp&Jp@nJC%*CWC9}z@b435%Ar=8HjO&z*F_(J*=A4F3} zio5Q%$o5f7$?Mq~l8{%g;HBpmZ2UBq+lueR6)u->PR%C1!haS^OkPZn&wj|B!z zZwJ#p5Dq(va_PCsMABst#jFN&lgv!Dr7cG_Smx;sJi;fDre@W^OABXOWzvnW@yHR} zCwxb*rclaN$K#r4HR{z_gFVyYU`VqBcDeadhx|f}@h(RHef^+y^L+?hHWN)oji8T? z#qs%jpTeq-FNDG$WB9kN=6u7a zGJvLye@=};q^J{Y4jXir@jjLslDf??__@na;yF|YoSQ{&!S8NzNBi!4_^}C+Gv)FBd|6s@!LrQfDyQH{=gawJOZ zzPtP&e!Y8BZ(B9VgdS&T=*o>$f&r|ne@_s$Z=k9w&$waFeZ*&V1DM}EKxW>Zh~XX~ z)Hgo~j9eDcqjN@bKP?gZ>u{)F~1drqHSQIVuho67g6&!Bb5J*mgh9CUD*$KOv+JBpo<{JVUNPM^Jue)6_}ocA+vLtl6F8TF933jHN%M`P&7*TJyr z@l#oJzyVq?>Ku6Zd*H3_n)LJeL|VARfVLq4`R#SPlMVwVR%dA5-+lwM}L0lprM&_=*#n|%r;>R z&X*<=Y34fm>x~=#GU5WLO`b#zR*$DozxM)bOTsm=-hM^|cz4_e(j+|T#VN)96P2@J{-EvFD zjwTWB!KVmW@s<6lS7*~wPT;zqy=h|fX_%vS9v-?WftNuUv>gz$CG*R$_SFe=Y3WW2 zxA)|abpu)9%?mJLULj?vy0j8Sc28X_l|P7qRqfw|^$KZRQ!$TUv@$W`ON z&Q0cFcTPfX0)>$qyJKvL5B_aU1qBOxW>%rd!-_}KDc##hg8MY#?)l}s_0DEt#MbLj zJbD?br{+NGngO(SE1`E!>A;VBLm;a@9XGY@A_`CDz#>mgVQpR-L}lfYsD#~6<7o*~ zro17Nznjs_ZWQk8yurq9N`iU^C))7#4xTePk9#8X$?tu$Xycp^`sl@En)p14-nbb` z_H+~5w)3NLtjQeOr5X(F_TiYASBvu#@_E3l_2APqR>ott;PbpX;yL;V|MGSsw|*PN z{WlM_5At~^tSI)R5pKoMbEY+X8GjC|XI)3fTw~Z{l*L{zUrufQo&%PW1-gwN$ZhL4 zvIBwFX>)-xeOEJ_Uh3^a`-fkofwCg%6Yoc79^MDtn%=Aa-hU45%=1g@}Z^@TDPDd zU3)ni7K>TCL-TCtx>+gUTAM*9hyuoK$%`Q7UnY)i`VX`w8seEfMHp@s3{g=Q%+0ru zEzs&@KU;#}=buG%*7{Y{#&!_#Xs%>C_ksQ4RwdZ`u7OqC{)Z0g>fpSxFD$q?3r^^i z5gp|;-Ybl z3nte8Ms0qHO{inw)on7i#$97&*PB2w_Xb(CuMg@BRAVhIQYN+j%T7EzKpyAL!1E8v ziLa{*9*`H~gOmBp;;cT#y|&@zbHeOZt#rA=Gr(nAE{N~tyLhZ+B#OLvzS8v?_VW2n z&K2yCzX)-VFLM^q|F9}nHWiSiMhaL{#ij3&+U0-Yzg`=`2}LdlmhnsURu*E~938$yn1^ zLi%Z`qQ>jXXrSB4rmpqI8J~yIoVY3Q=*w;}@h8kF;Hh-z$SXW+V;zPrGJ_5_PFNl} z8%t7l;99>aIA>Hn)et`U?ow{&H;XU+?arBU1h(FK07oK2K`}v} z|0@5?Lav0vvDgQ=dbBO`H=YR_K6>D&DUOh9a370l7$okz#_|_xGrPtO5fV;$(2#=!2H zanOG12IQ^OMx$OW>}K2(c29K*=1eu>;~&?-(?&}u_C7rJuH-4oem7&t-d*_JA{OUv z8;B+sSCApUwvfmR6@rb^MKVA|i(2p9L83cVaNx9y>}yZaEA6ldg>o?;Sn!$sSn?bk zr!3`>JwHO6%_VXtV>Ubft{roSZ^zUDX6SN!hWtszQBosj0#_O}k%QfI`JERR__E`w z@+nu%$o=&V%xbj`Zz$g-4BxRHJ#X3aPjXAXuF{mN{(g;BqPI6{_7j;(+E}RiBVdeE z2JGJ&fo}KW=)tPdl>4^9wULwPkF)#fyNs#y+Qkv{_J20yVfr#|y1W~1^K4)pi7R04 zIyuk2oX(qP7pAE9$rS+;*YnB@M z@LrXB&DO^&&c#^iW&{n>hoYZB6xY0-C0}sQ1tWCKVCCM)D62Qa<3B(0zhV}wDDf7} zpY@4e|9=O{^0lBjvJ#gp8wuu9y~Hd<3coPl2J+er$l2Z&n22SOTpF}s(w`ly6n1dLm1TjnI;7rrWyvHyplei7zJKB14qs)bPDpsA`)eYn2 zMS83<>^oL!{a`jDmdcF>_UB=~i?~MSXLf)8bHRGXHL1h(75JuqCRUsp4c~Gbh@n;* zhK1~9@p)%>wqy@_N6h1M$Ct5ryS|Ek3xeD0l`(3;2VrqdUo4@mGFkaow#Bp``7+27 zA09BF7o5-0#ff|AhLBM7A3C2j)=q%7vGd?1t0#@|?y?6zHLh;UD3D*;`+}eTyG&U7 z`lDU!r!#`NQZeAtk0`f44yJ3{$qd|#>pG8ulH)-h<~)MGnf?uy9qNuticX31h7lh& zqn;LsbB5@slGlH_BQ$rvhdT<+;K{*#X~4Kw{KxzQeCqr(3JPcW*RM1AX6r(tSm#A% zEq=~ME=c4hy2*6ETzr1dT;|Whn((+rw*7;^zA&<8l~nDKT2WV99AtgJ%i@%jn5X&& z{vXYdr$_3`?YD1+#044Pl5iGkOw^%=hc#@y*eKJ~zCqthHeuf95AvDa@4&(%I$Y0q zF!x*?#nrRb1?PwAY?H}%l66f$)sshYm8$~}=xT2V_~G`A3bubz4hv{w#t8*GA!P(?a^*ViQd` zag3J!h@%~YYam8;hK4<~r=ufZ)4qQ5=^eMD&@ncdSZocZy(e7;zxsVJt?j(r^6 zerwLyQr%lHTQW^zIz(M!GfhQO)A5tGb{wXMMm5PNjQb$}y-!bm)n@?Bd1NAadTj+A zr}RdC*E~cXe&Z2N91_aA?fVF0&;5pm&-dlk2baq`t~@8w#r8DqXabSl?@!%V|Hi*6 z%JNFjWWJ{>9HM%e(AASiP)Ucb

    PNq6}bs<%IgHrP(Vq?*B8Sdz#0T|FbaY#0sk z&evrVMtI}#UAy4H%^a%i)Sa5Q#gHS`kB1KhnrhNR za^}-NTJlCmQjz4xi@(0%6NY%n4=1F{Qx~Yn-#*J=qY_nQs;V|JyT%{fXr&=9DR=<+ zRxja@*#=2hxkRE>TLbKNJ*?iH1h-2qi$3aklD>~iG0t!&eZ~W^q?Z+qTlE~)oCzZB z-3p0v{5{q(q)U)_d6Oii)%0NcXgc-Hd$^GG9S$v-K?na{LNOqPmt_By4c1Yj+fKdY zYeN0`{3S{}DXNQ_T)GVZdFt}=kYsw|@=|KDQj>t5r#yU8j%;zvHJCSS13j?hIgbsQ zg3NM0d3*U8v5cKbHf+q2D?a=qKh?un{%M7t*k{sWN9v}rcr6FH?qiwU&&^QoJx=7b z{B3|L?LMM&%!E#>mQeqKEV4bC(SaiK!1dM(l4~I0q2jsM2a@r2 z=Wqj#syyZvNW7XP$c_{gX0pvGoeU&JG~H__Y;G=rw?log zd9D_@?DIzUbE7S-zON#=tzj+swbd9?i4FvsPJ|u9v{85XR6JvP5}T&zbBC)V>EISk z+BvP4BywV@Wc2|pNoS=$WR8-^E@Tzi$!0#MEq{%{A!HqtCqz)U&_lGqd_7H-R!9{; zO_YvUIa3IWSEN0jHIu z*h?6rQlPG8rgOaJwyckjiaai$)* zWS0$Mh!cO^4?>5=OcL#rg$wqHyXMw>SmY-1Z1pc-+>K!RZ^Ut2q@aWiHCfj|- zp8@gI=e7rR@yHT8-=nCp)HG zEP23pO9wZjqy9K-v!5t_&g$gd4~pyUb7Tu#9#Vem zD6Kf;2tAz7gSUwsji(L4m8DUF&vUWQ|8X=d*;9$iBa9?ZVxH4&Aw)8Gt-7Sm&ja1l zu8{CpE0{NJ3aB@o5LW#OBn_&6giaS7<`h;SggF?oO*NNEx#41VZ_h>MnwW*wJx;RN zAAb0^*LfCWG8$KhM^ilgN7y@|tf=$$2H}lW4qX0Wz$dS>K`dr{wao^EB(3 zJKu_?aQnj?VtD1ZeaA?DUa|cbD1QioNot9_xo#z{2~*_WbJq(K&2reo%dhS3#m>R1 z?saV3qQ|7)x^qm`s#$6ttswm)vIks-l}g(^!t8stCd*9gu0Ubv3fOTp3Le$yV|;ui z99z%<3A>6}$aOcORX3a$96HJBhNua%rcRS&Wlg1je68t?KbdrZPa+)OG6hUal*zZ_ zQS5cFl~henMG|u8DZO?h2pUbV({9bjV3)>7aC2G9s70C3J9@eFX;=t7=4KAnYm;UE zo|{2_S{1_9EMO{PzII`LIs4X=vc_x+GH&`2O#EvN1M+*))lL>P(jpF8f)2p)5=FY{ zqa*Gq>O(ZXh@Ov=ig0PDB76Bv8^W@F;z2r|hF&(IIX-o0HD@0YoKvObtD(RyZRKYh z*Hgb~^Qd;xY`#Ukm7TT86nqD5f!<-u@K0rL9P{r7jh}m$J~wtm<(O<~Z2S~_Wiy=2 zyw;sf*wTC@;_^~i*7p#0*`SokH1h;4WlLfE@^EaJ*MoLNo4?`#w?1;qmC^j5ivo4;W`qYm z&t*eKETY6una>{<#1wsc@$jYymi9JU78)x<=g*bcwPpfoZaqX>wc6<xG5Yjf-c>kaBm?)r?tHF@a=v#ULs)xH?6jne65cEv@GKMm`Ka*nZ=!?x%tmZIslisYT!Z0p zo9N>7EUM&h!w>9!iTUYrvOubjS!0aYgsbbo+SQRbE+0n}ZNA8Q-$~`alJ`o_RnEd+~K#Xt-)$pQZi+NA%B-up2FpACwpqo=#WHai2-{9_0?zilbo z`!+{5I&+dVBhpd!u_M9Wqrky_?N4K&q1P7Kw$>mtnU&8p!(Yi(w4KG3=SunML5Wzc z(4X(nvxQEPi556mnL1boP=lu_O$*G*sCLE{MiYTr4ICT zY#jtlNg;+qYV9<~E5OCU8L)o#TV^=wDU+O>D{Gq_1Kz=F;Dk~iN|aj#7TZkLmfoWZ zVV|+-5}z6c5S#4GbVt#z1!L8p+vqm@oQd1RJp$D*ac2 z_Tsk?)RscC8|1D#B5bO zC27HxV>dvvO;5Tk$dOu{F+e+F!J11RksR@N+h@MEB+KF=J@jrNTgH2%$H8{^sH;b3 z4S5TLKOdxHn+HIGt>_2^XD~eJ4sP{Qku@A9oLG91Md>Qa%0IOc$(&L+?sSZ-i(iRK zPTlyHnn-B>o=wI&CL&kqf#sS~m~5yD`4jBPp@41d$E#W7OO^-zDp!W@FKWs2kePIE zy%pVha2DNE(ceC4doQ+kw;GEa_=m8D0uW{-@-dwYz~PGp)sdc~mxFBR6ICxLzcGWl zDs%_U-A%Hz#R5jy6~OD+BkBHk`^dzW9-w$-BCVdaiuQ}wq-X9c(r=sc>Fv;+SiWu; z(zCIoZ)tZZu9!=lEIZ-D!}lodHUN??DbY{;3B>xt{ z-ThK{+pm-C@kwUdlY-#ajcKTyo5mZ(SwT~ED6Y;6#At<&7<*_H+^IICNmmz8n{D3E zs+UI1-|N%*;$vXFXbcSg@ex+cGp29eEQPLG6~6BKSytZ_L2J~y#5?%0bXV&Tw(sI# zR2`VYb~G=c>rKKL^!Wh#(Z)h1{VhEr}Y4%3@13B6@ zk=0G4VuncB{^f!^^134&UK{$NW=H{<;nOa1N4k@9w%HJ>94_1K-wq};TX1vCmK6*? zPX2kdL)_cZXffs^wp`i;L6?Ha-TEfk$%m6Le)MR1bzL*qiZX+H%Sx!>z3&itcNczX zl<>b5g+ixpH1@c<0ZN}eAhUy8u=)d~N^}=(u&#plX`hI{DQ#F zR4}|Sj|Iua97v=O`kmU1~s<{v|4c1E}mPvH1U-@8}6s9!iD}ddz%q-^tD>3SeA=-*XGd=Q^N47cNr9|n2U}xYcZ<#QXJ#= z2DeXAr#_>T*`FvcwBLLVyX725vn^pzHs(9~xbro(PM9j5Um5{)C-Q2sU$B14B5X80 zh)>5pAcp4=tE2tdgw4OusK%13lup6wM+mBW^tkOPGrmx(5$bOkL(1M0+F1PWQ$oeZs-<)e4NN41tKp+C<-2Cb$&n3X?)gptv=<=+cd2 z==v`L#`~TDwWcJREgu7Yo@(;<+XnLe$|YoE#zFgt;eB{=Ou8U>FS<>d&&fuepAIiN zAELC+VQ@e5lY4wj=hIya;qcTpCOq5-(hm>Foxme9w^5_`_}=0UpzaoGi&^Rmt9p~* zK4qYhvjnDp$Y2c%KC8u(cJB* zWt_w&nY6P4w;}ATivrk5TS&i#TnzRd!B?iA!0=kYRST5)f5A#TKlT>i)S@Jxlys0S z{Mt+kGsDU4UsYs#SqRkYYf#&WGW&;*3dxsj1+I}|!qUWE_z-_rcs(PPU!S5%dY?Zd z)B4gyipShV?|u(>$dH%9lH)-*qCgeT)#uo&C7R=ZN?Xwa>xib+E$07WHO5T5C|loW zGaIWdW0yh>llqka*Th}q^rv50xYi-=cR!TxAD++O)GWl^LKB;`FbI`R(|KonF+ZTJ z%+qB%>Gz~QG)GuQtyb5AWpWB1*mMGCxOS88&rjyx)O`7vNZQJ(h32 zY|MwP97C?iH`92}0{Y3git3wArq707WZg8b2oc7^sqeRcFk(#*&QGbus^P==R8LEM zr8@|J_5Oz1%Xh-*d8T~6stWJybj9kmOUM_!0G7FBprDeKh(;D#pgbi7wI+#uyTi3i zqa%?k#U^w8+NFHav=-7WJ%FtdQ}aEpkH+DHZbI@lU3zTjxv{NBFJr>qQK&v}KA5aq zBl1`EP|3nT8b5rzJ&n}h{rC9rOM4_}M9WC@c4t90>o?>#JJ7s$CH9N{%;vva`(kwu zJ?WvuvrzG92A#dP3-;nI7fUwnu$nm2 zdvl|1XFxAXLMK`sz$NABC~889;ziMKvSBGru=s@Om6`l+xcGgx+Cck;G_nQSr=eTl zd>p7f7T<;hVWg=JKc1})-wzGor*9nNr6_k>s6lesmIcl{d5)7z60gUzWfj;zcOPE4 zwG&-d)r-!v18DP8f$Tl86+$cu`A^NunB3xK-&)nc&OHvn&84IHrP5$taP|y_45*`D z&ws<2xAic$x{$3-{xvpjEAWQ3k-Ree4qATMLY*%khMV;sd{Da^|M8-Ry}xh(N4+?J z^oTZJ2PIW}#J|_+F|9%Xw zmiqG<#xJn5V>>&rs1ZyiIP-C?zgTb~qS9%`h7a>*FM@juvj;mtY@V2F3RnW~b3-BS zjp#Y+Ziu(n^rEZG4sr8|rMy7@HXNC^RGz)0gUjb6OOpx*(BAJZ;qDPf$qH|O;hW1m z*tnFz4*5|09X)`jWZClP!P(I0nhxd8=D4zZA%?VT^X7y0`0vSA{E;vWjhyE~O3?L~$+zsf zOPs#{VD8OVgfeH*>!mmt{fr!hK{^SfzciJs)Bh?vD)P4mDd*y`>=76~yTD%O-8J0# z@-J*mJtSuH;z^&)nN0fcw_yD9p#7@6BxxL)3je$U*a@p@S)1}al#2j}wwfqV&Gkjq zfLi=-3S|?Mr}JrGLWZZf;EDgTWw~ab@aa}Z5}3V;_B=isGQ7T^e@rua8TaSvYu*t5 z7ir|^i68c(E7j~P_!fGj;{pUcdjw$(OZkQYBlz@OLnQIN$hFK2VmFS$iJ`GV><1mT zal~oafG#!Z`VGnU*Mg?ViY492$o{p0#(`M-FAl%R+vZ5bxyp3ZEGrzgEefkw4-vT? zCOGV!I@$a3p$vnyWQqTVk<~U~tXH3*IOOvt%ucglW3$7V;)~O$K1YiW_4`lInw)7L zeHVqCZVUM3nb~OIYz~DECqV6-JNY-Fzi^;x9iCsEA$;^5!MfX4qt2c%+TLvhjGZx? zZE>826Kmx%N#I(PhqbZA2YiJ3%4`xc!Wq&oYBI<0E!1Dtleo8f@~W{vFm(Du9K1V9 zc7Ip}PFb@M@y>AiF60P2PSwEXX*u?H_r;KygFQ*D-)3f~;>W!GqS<=qPb_4)k>px< zBb~fuAg{V}pB}H62sYjG!DGG;=B_V=EFU*;@$1RAM!tiOLEB`rHNMCcBfgTu*TMzZ z+ySbu62zW(oKVcBLS=0vwpA0GTpsrAw$;2s-1@O)6=ix!AmCa zC1V#qsURZ;N$zanu*r=5qz zeeq0ieW3`l_qC8*JBEA~S8)9A0M&-Ojkc zisObbMQ9-2Ym>p(+5zW|nTA_V%F*0$8d?9Rkkyplgj=S{%y@zfpDLBJu|FdB^4!(hwT6U6eIFL53qBZ(Q4Kx^0|*}d@>P&Y>h z%+tMz;z4(85_y*DyFan2hnt{vuQ+4mdCQKb){*xo(ul)}5wOeoIeeIUT;#h;;pSX7 z>K=Ovc9#p}`rl{LKX;GF6bAoiuWB1iG_=)acg~dI@lpYsjEh-;*9jr8Z4jJuzlk>@ zeuHlH1`PapkkuVMhq`TVF#GLP{=P+@?g+4lz^^8tkoH&BfBj={A21%Yw@#MT>lG9B zH5DF~-y-p~i(%r z^zD-Vt`n>@cVT;mDtp>%9^d3QnAwX=hq&j}ETbwMe`+`3oYEtBa?ogAROrXg?6`pw z+MkeLxx<))o;?ZDJIHR724e8gYFXO30sPqLAWRthjve*NhtJPT$=Vr?sQ=lLnN@9I z+uR4yQRQ}mt7{*$Ub+ekM^xgttBbi&lMb&fQ)1&kCa|#!9tl5ZKgXD}PuaB0PI6_m zCfC$;#tlPe@#_zlF?08^7^~i!H#eTJcbr)$opi(-T~BJ0)<9)_0L! zzi5b04P))|9Ix8HC<+iBWXy(P%kK*>{v86V7FC!S_nxhA(`6sWJu7;%Fqfqy`^gl? zErXp!cZ9(ovt{KDkz~M4U7_TOl~DHcz2KUqCmX-{?3Gm;hlA$Njbzx+Dh8K#%X(yP zvp<_XoPAR7F1!A8xV_K8WsI!dBhA>kr6_F($Vyjq(Z_{A)X7cloJw`CdEwPoqAz>>#msP}te7Bc*2$?xFrybK=?>cC*4i4V6Fv0F{u;pe8atfKdQX}vaK?Ot2(;F&FK zd*635|2ye+$_s{#?KXWS2@}05vGyc{0Q)n&iSw&0HtDK5dA07c{Q(Ou?RPs!SiLre%-UWh_+;LZ zmd6+ndZoxNbone{TI3TpbbS@p7WYRMpe1*k@QmZK)l}n7Px$!Mk#%j}gBR^p>AO!Y zvi~%c=;y|(LgJ85?CG&=D0nmr%qAs4%m{tFyG#kBd8vY_G?BKC6vL8hFXFqZk@EC* z7dG_Pa=d!~5p(R`2#TinL1N#X`pnj(VNQK{U#TLgy)u(^xLgz+z8Ur{Q-70^ExtJP z?gKWxn>F7Rdy7oai^MIhRyhCs3*4v}A$oz2+fU3&g16%e@%YJwEHmjjpVCJeUp|$x zNf*_jvGM@<*KZuC-LRzU>5BA5!$t_(p3fI&3i4|++Tp9mA?CAmG`V@gL*|)h!OFI| z!i2a2c4F9Uw)kH^-rraQZi}1(+35RhTE$-&llcJhqNihJ5y36_(fmh{lH7aUDDGV{ z0-a8~!JLl=h;UC0Rb?CPq+6`8@Kqj2Zma_wzLU<+ZK8w3eEWvN2w8=Z$N`lap@O#y z-CeqgAI;CZ=dfBHjuAblKtuzv|!+1}wZub7kQf%9-f%^!B+ zY!@E=5snhtTYg5j2y#{9Xpp8UK6(~F*IYY9^VUYvlDKs2>=^?oS*}1QPo^60chab& zFL<=EAAGVZ0{@M#$ar5LZqqu2zv`L?{j0Y#Rl7tsEagAkU346En=UZ>m*0i!&tu{7 z%-&F(Tp^qF_&!O~T1Hp?Hypd9raVn^Gq~%R(k1N)@JA_%5J@gRZo9%=Ms=Z?#a144 z@&~W<3?s)9q%xg~MEiy3EHHvD=Pi#-=&vz-XyEp%r02+cY{lYCHYlRW7pW@3Uga7y%~c${KkWCyo5iSa2#Jv(&C}&467#}W2wIO^zboznrGNZywme= z(cwPW^XDHPKgt&xcXsC?KW~8arv;8^F2N5K8*qjq$JSxVY)8H|myFVr$F;}vruY)* zRB*z`=7DH^Xdo}@xqu%WbCDMxk+T&uV&G-14ow))1pa&Ok~M>GkfeX%VAiI`r*~&) zIai=wUk`wxw-PnE_z?qpSh0-ugP1Y#78)_^3Kl(T%e=?z&-hFW4kUY{|GKtPA zzlCc1%2`@v9L(Lm48wl<(FIGC*rq4h{Ds#~zPVFd-ZDNNpNO*+TB`78U+)Ru3cFZh z!F@rv?IIj$UWEgCKgRlhZ)JbeQ^*RYD|Pt{B>u&0YS}QEq)xoRtZ$n`U7{MT>f8@I z=PcoWPpk8jf9p*pA&V9Y0)M4+kJF6{oQFFkhk9h#~elGEu6VSalyEUeIH zo7|hQaBWY%VYer%NjS%%t!Faf>rH&WBoQOi?%=Ss_Hv)i*8>TNMS@30ex6$hg8fiV8}mnJR-6%5bn z3Gdf$$K9c$z|CJ9i*9(Z7}^H=Cr#snw@2~y8t$~>;4nIG^;|ku_KbB*T@U|b=)B{x z?7}!sSrIa_M}?L`i|1U&Xds0`QE5ktcC?p~oeCi%GE%l=Jm)$Z%4m^>N&}VZZE0)D zd;jl`=i_rf&wb8y{eItX=6k_CJy4YkcUN(2iUVH~F}YT%?4&q&qhyWy%vHGdemZIy zji)+6X=GSnZuJkd`YAegqxp*M5$wyF3+!})15WFFhzq6Oh?gx=5dk2>oX2;K+2& z7fun;;kUJGL!gx7^g5J1IU8cz>gcO8RWy6%ZvH*95YFhWr$v*0a{BoZeBZ8&T4!3t zKO(~EV#!XrE#WGTemVyn7uJaG7c10S(t)+1AHtw=uCL&YnPVCANSA7So>4pK)dUDx zk-!HWcB-0ObA>pTX2HFPIkjp617h+^LPxAGXU#K5vRPSQM2SHQNK{(}pBGg?f(;`@ zX-BWKd8lL%P;>i?lu04T9UM_|ukM6GL=lMe8uN?SrdkT$x@d7rT@rJSSVSH?$9j|@fkLAu5D8KME zPC7nK&`bJ|oKM}G>g+eh~cq3GpDEN^Tei{}`?pb-;l%Z+;3#m?6_ z&&`tRIwsMnIXS|-^%lB1yvBW`gz6=`HS z_oB(_=%C&3$!{h2w$+HDhWO&sP-!xE$S(4CYp_UmTBK$zl$}6)w3Y;Q?&cd z1M>c2z9uC2za^z|uZh@pEMK7g19M)SWm-N`bo;6k_+>;8p6Qqh7D5j4n6;5e z&)Jjs^?8fWYY)J+Gg84qqX2X^C((V4m(U1Dg01I7!T-1c_Fvczedn{qqxW=TMO^|$ z_6=t~N#(bERIrHQ`{$bS?2%G% zeY~;N?fwMbCn~Y>ruW&AgdWlH=pB5={Qbmt_9GVUU&?nG*z)z7E78nA%pX|x^VZ@z zes0hpesN<5-ji>ETPIB6zmiF8wp}indixDf<^ZaB=2 z-xo4r!Vc$wYy$>1mO`EVYJ4ipPRfK`d1(JXcKGZHP&-yvGw@S~cvQOwA0aP?_Sbar zW$zf=u)~0#xpsv-eCmRCX9=!i7b~irlt~v1$RbBof1=9RTr%kAKH09UUMM?3z zcTO)kr*)F9*se`8ZrV~usZ{t)?YYz~9R}^bFsL|yFLfz^vhg9ff8==_^(9pJ&gbCW z934EOpQan~lD5#_ZLaRbzmCqx5p=CIvGYjDF8q1U}{B;Pw{DA!1n zsf~>&MMq@%iu-vbU=1a@*z!in0 zYVtswA8E&Dc2D7MWG~CC2xs>fTXE;k&k&OwPuj>TNc9V1l7}*&YLOWpV^uh8$$HQd zN5YPF85|+pFO=E>;7z|W9atpnd=_-$9GQXiP{ST#BQ%?{#oO`GfKrlkYbO6LEz7l} z?BK=HOm^g=IqBN;PE@?=AX~1z3$0#%LD}B zh|ZGx0E-_u(EED6L`B_~Pl-{&MRrHn^5$b~=Z<1LQ{u&2woau#2QT0YE~kK7YCg;< zEMwa)jD^0rz9O&2OKh2szu?He!{ld0(dWZU(V4H}iPM^J^TtLTTrN1LAF8pp*3LvH z@+2miUJyS%F##W1oW(H3nb<02h|}NfgPdwtx?c7wxvJyM(_i#a4ynT(UjdN@t6%I?Td*L;5jcQ@xNE z&%iEa2bBAHi?%KU=jvKC=O4uqV+PZGr-DJwVsy{`->HCc=v-%S;v73sv>cMb#gfg9Y-I$Lt zXy9v0#E_XBMyAg7guow0{AIfg*a=+b)E$yFclAB-Ol=V=CXV9SB|-eWWhJV1%dwRM zCBb0NO>~%A#+)vAQ%~_T(qnZUn<~5T{I=uVDYK3bHgV<~>=n3Y%unoiXvIo52jUg^ zJ3vQHz}L0spkjgvqYf)TIr|Olj5OjW1`J`IbGD;rzu-e0oQY$0KO#>~<}&d$IX3mv zT&`mKi>&pY$?IpH;XiE+`7M(pR-dwC@Z5oW>`+1#vwoKVn_`to9F>J}4Px+gjv#J> z7uO_SSPOLT1E|=jh&d+3V4o9CZCK!UP`elo_%|>9qKih^14If2jJ_lL(&J^sq zdJyhS5g2-Z3eYG18WLrlh(24NK=?9CYa*P|l-=5rUQgdOGnig!`zNHo_x)&qa+AE8s3 z5&W9=2M5Vc=I175lciE0Ab#o*7`*=i4?J1GCqER}rs3yc`<#{hYfAth9&}bb&_0uN z?+(V7hc%c)O&eScSi&<*{CTs?aXw&k4Ug>c0)J)%Z9UQ~i`ug_mrpP~`V(D0){=Vs0t(wH}1B{T^(dml1nBK9I)M=hKGv z6k=0d0MkEabEECDye*;{ErN#dj)u!vaw}RqJ?RUpGkae>xde&r#%E;ywh2fJRhX~j zYc$;&j&sK!W^rG|?DY%{7B;h;{c=*k&PUVuD7Su`J7K<^7dd zsrm5epz++R>ZjGG;SIQJj4G+Lc_12R-pyvFAIIi^o8-&W*XM^)-KQn+c4ro|+w=_Vl;XkugA_OaSjclU_CvJgQ8YL?fXA3mz!$F_ zgk5+&DG88e=X1~FGi$*Q@j{yho{oVf^r7HbRVQCk1&{Dp86N16LWk-W(H&j0p>vB0 zHB~dFuDW3$v1b*$?KCH|AD7?&kL9p0>L}Y&vYXqDJcaPO0axf8YyKB&`J9W1d$k%HK{CPMjDJy3jwROT{!I0Z8$Z!3>y!h#yqtd zczVNLKKJhdoMf_E@DK|<{1w7k*(wBIrTl`5s7`X`aVMw^3853?+gRCSOZ;+Q0>fhp zu=lC~exEa+=T44*uI4vHy1O0>HcDfQy9~Gp*_D3{yU^YG7;Yq=MQ#IfFnU~*@HYea zBf)3?_}5f^)p|MC+4g{)yb+3Lrqp8K?@k(0CESzp%z0SJ9DYB18La6|hn|jsXz_3* zy|0!5laIY+U%k!wZ{MFhynG`Zjm{ATE;@$4{k+Jh{|dez6*@ zySRVyHz?ZI17CLz=0|k*qtO^wd?37k{|j(|OST%2Sn`Y;?wCp=8is?nj}yQ0iEtr@ zfCcxQ&|OWA79Kc_8U0TIguZ>nm-85KH;YtU7hF3f;h>&dz(ECPiR^>&9g$gFG$^+3p?|ji18?=`mt8jcWp3}p*E(0;l;&pXq~`(q7Q&;+?!r&}Ai$C?8u z>JwZAd?xoi_5s_LHNr<}1E~3F1|Q0E;ihUWdoSb?+Vbf-bmiHn9q`hyiLm8qFln^TWDze)L?54O(=t!t z8F8c-;XR1?_YQ<|P~?tI)45NQ6G-ouMK7PZ{MgC{UQ~04&U*Nl9#ltAm-V5~UWU+g z&oA83eJS`hx)aB`UJ~7$3o_QTQO%tb!v!bU7tvR?^!`Jr={SmO4h-XZ?_%+=!gqE< zSqAEg9NB{Olbla~!_|N0^MM=FP|d-aOc1y=^W#60Np{)rVS@vz)Q;rF-7Wl)$u%BP zoWOtle!@bpjK}+b)N0puHG_ zef<`%w6P&dX0J)gY!m1R$c77c0jzbQDg85aIqct(2y4pZn9s7!0yq3IqzEAT`^OWQ z*TQfzeD4C>^xl(RFF!~Z{*~tug>fhpNugC>=8JwYG52DB*{;sdHOfeXf=>5t<7mYqhc>c*?IHSo~L$ zuUwwS74FFL{<;mMYF?E%+B%3-_zdN?U6n9z3GjLyXRfzJi>Jwss9lrrA6gF{h7}X0 z67^O&a?$-2x@+XIf#ZjB9mztgXA?$|-ATgtzNVMyY<`I3zAU%@n1MZ)j)^=@NAMX- zUSPGZ6&aiJi)DN{%N*tk*@mvexGYxqEvk#he1U!6DcQ^mi#GGUpC;lzXPIrqz5L&)X}B+a#unx3Aoa+3`7etI>i_Z<&=lYI~+gnFQE%Bhm zj|WqqPl2!{V>FH$;|j{NALCP`qws#NaMv5XlXq8-^i<^PYKuUIfE6q&G{fI&x!UK`eAG&8KF=?Z3A^_^T`PE z-B*)8xB1O356WXXgZ_wX>xWR^weq!|3$LK3;Zd~uW{clir$F-AQ(*bj94vyXxxKJg zJSI2|lC{+NTb;M~-}~oyNSMo=TEBtKcQ@j02O{}{1%Aw?se2f=JFtnXn^JgMbh8>O$8vwq zOGkuN(hI>?>5H`=s9uVV;8^Lz@YTsQvObOP z|K`9I(j@q{7oxGx1u}$)X_!hJ#4RNqZT)I!G6gi92TbppWRGhl*L<` zHddS3d2B;x0TDfa!8r{%qR>bj(q zlG8}9hh;#5$QF03%Eg{jL*e3Mb2_tX3w0W1SNkH~h8K+w=eJBl=$^m9w16(7Ioyou zpS56xrOtFj$rIM1yabP3I|vCfPpI0WGAi%rM;9k)&;=gDpnhlo3=qyhHAxr61x%6d zY_p+GTcznX*%RV*LJz-=Hi^awJf}XvE0yrMiK!_r5Z{*TATcTi0uQ2h~ zYP3btL49i0xktl=S;#BiSXxVER_D?qwR2!qOf)0Q-5@iynbrST%~cvYp)}V3($8L} zyB`0d*CT4^{#8Q0wn~ye*;39n7ufR+^M%gbK|M0<)FF~17e#f=)ah*b!!*M)fyNh4 zqR(ch(ru$sh}ShqdV7Z@uT!iiQ@mT@)35QcZps9@zQT+C)re**HjD<_#Jk{O;70$8 zDkgmq_aW-k32OIyIjx!S71}(5DRW#wyPxUNZ3CJGe)}kDB_l-@t979B&mpGv+n@Wt zT@3rgqiEYGP3qRa2QSBtg&Q+3!;iC%V2b%wOxq-)KBLsF8|M024|||(ZPnGv^{3B< zho8>F?j0xT-iP{pd)j~4pQ?;H4@S~&Q{w2w1TDH{dJi1T9ZzA+ZhB>xeeH*{ef;C} z49FVOOSYeg#>VgrY$Oh_`CS@4GTfGC*zSPqrC;EP-eNK-djq`r?-ZE&?!mg^b>wJU zHLZ~T!z9hPm7-m~mCJ!}>iwdNELt;|o+uR4M#Ts0>m!jkw?vI^4pRipp=&_q%wak$ z&yjx4JxMhjlIS7zrSzowMLPM)GP+{bK>eT^fls>)#fx0mpstq#VOv_jRFEGo8qNFsaY(n61k^p=4v zO_i*J20F5~d(mCGwDhP|#f=PhSWM)HRDpZ-(+&K9Z`InO`{}S=+yRje&Dn$v)y=4wCd1At~ACNhJ{J9 zLre7_P@urM7o3KyaDC_+CQoZlJQA3_@wB4t1X;LuJRKV|j_xj=hfKu?{2QXE+Uabz zPVio)uMOrE%cC(SunWHyDpR+o`7HS36fl)f7ubS^qQWUZ$?%&-)I4MeOfj0z4?dA# zo>S5xePI#YI~T^rEW9S(GHM8}4v=GYTI2W-$x;#&qCveCJ;m)6^V#^@mcq_hmW2Fy z3u7^bE|ETowpUNkPis{v4)KN|)idbbkHOS@^9k}x^M18r=y0^T6i!rLgrno0V$rLB zeQfLTqadYJ39YWuJnGCmt{<_RCr>k@ajQxo)+q-T2W@1f)7J{{i*j;dygz+9F^M{e z|5)yu@sk9O2h3>_;klmaZ11Z;(e}ZIVd5xLte{%t?wnY%cKKwe^0|&*+k!D^?RI+p z#d&DiGy%g^Y>7%#EM7f38;zqS=*|{Rbk%pD3awMPugzD|OVotMDACr`#X@^kR7 zvnDD>Thffw0BWDNl8!AEvKwk?{B+MPCUH0vmTetCwJd>4N)cL_tjcs)1X=q-o(%l* zh&k2Wpj&=ZerkgPZ4Vy~zwb4$*Na@Ky~HY*vuZmm2%Jn)3_g;)&lyzZUIUejFy{l| zFkfMkh{qHzpx3kKWJ&4(JojfPw052#uNrsZ#+9e=X6+S``eYTTcv?hEC3X0XMbVHr zb__4`zK_>}WAU?OJKDV#;d7nCFed5>9jpM<=yC;+$kQ+lPl|iI<>{O`!v3*!D=kn7 zpzb*q)cCLi_nNqqj~N$7H$V5r?4ubp3ZuX|T#@dVct=}LjH7?O#Gvq_nA&C7@Ha6P zJno%xZ5{6=JB7WU%5W`|mi~u_-^5s!?e>J{lSQ;|MI%&aOs1Jhf}^%<9`pZOitlUJ zvUXK7)JQX?cm9oKgSM~3g_rzskd3tU%45U1(V`A~UNnYp`)-1s9#iRX-D_mBs_=i` za}V!iYtWv#PldVJRWjb|29>zBk5;u5gHKa5+AMSvA4$2-Jk`&Fe)1&zDe;?SW~q@s zyBG0Z!7-(1TuZ&KKBAk7BB17>3Q(bI{W7fq>(dbDdAVZGoOozX^2Vr$6BwY@%on9y zgjcr-Ja;ODsGA)BZvBCuyYleNixX7W<^i;(j1@c-0w3_uNWN~3C#Fg-V!9Pg?DU_# zpt04A>g7Ho76+R|ZR)$R-gvNhc%=(|llg=;repZR*k^e9L@ve_Nbr3VucNhaZXK+d zjFa9rh=*Q{tC_#!896Jsz26%oQ-7UU>@E5Tn?Ju3%_~1CULSWFPyefjS%vk`SzbrZ zxftU5yGm@#jmJcNP6*44GJx65apdI)7p@}cywqoF@GtNH2fKt5^*fhIcVZe=Nw{#u zx;ZGjvK+T(zQzy!^}Kh_9=he&DR|7)d6a?**?nv@%fs1dY<8jp{)^K74T1PALbTt)v z@&0UagSI9nzjXr7z6|`6BuN*3+l1*mf59w$IoO|kohQL53ldTr4t`Q*tqFnTOvj!=`BJ>FCTUC*Ya%vVf^dP zh5VVW8iMjLJm6YNJb!iZ-LdIhS2T{#)0)U{KfQ%Br(XqMR2RkTOy?JmM}zmZ6sF`Z zgNIsX@zq8>pf~)5z>YnQwG(FXGv{COv~iBF{S=f%ZhKJPcmKTDCnF61Zkmfa z#X5Y`;{YBp&X8YS_K&r%H01ide_^VbEgS3_NlYZaLaJ^qNZbzt4RslQzPlVt923b} zo#WszvK_^x1Nj|?C45#!q_{F~I^8g4J(;>CoR{XE)%_L2U)#**P8mmd*QistDswy^bSn=v zWqycT4mjY5Bz4|CAqu5~!|;_x7w(>Smh22-q6r?`#3NZ`o}~hd=j)r?Jv( zPg^@PE*k`pD9J~kSKEP zsw=x;W3>dgdGMC(+cAxwyaPDUpiy)!bu%bEctDzOPQr7E`_Ocb0}4Pp_~>CG>~%(o z3pyHzN!AzOvR^<|trOCo__9GgwO}z|lW6g^an|nTYv}!6S)!(J9#r)zAx=jTW=Y4< zDn&UyYMeDsTrd#lg&w!mexraNY8KG>LnhJCNKLS=o=;b4C(_QDD%9BS5Iq#&4bzG> z#Zy~#$#V}0dgAIH)+Mr_AwnNK{A(Co?)$^%H?^Sm_$P3ru$QzB$%b?X;8Kqr(bw^J zmDPdS;;VJ4*3z{nahdf#s?=3PGBjer!}Kjg7LB+5`y!1rhmHW%(VJ*j^BGb- zWNge{belOJB*x#MJL9S_SG65yITfO`$jG`?ALt3yk4&t!M07%Osio=BIilx_H;^Xf zU!q###%I1XfZpA)xNYiZae@^RgP2y<@o6nQo#D%W{&x|Jc3j0FIvqHC>`^=;Q$SNB zQ%H`zA*)}nCEm-7$%}^*ar^mrX5y+xuU{!v|MuXE@8F*L2NK?lL5Sjb&<%3o;&(^JMMdl8&&G@i*mxcz5mH~3su|za< z^iR>Q?{kRdnQExGe2SK>6k)>1H~1=R9{h;ifWFVh;>?i==$6l@^F|$d#m$Y{JRWT= zdwQx!Jn1J(DUn9gO`*^%R-veAZ*4Q%z zUzp%{8{6+Rh`*->kpeAbZC1J9GBgkk{z!pTQ7JPL6<|r7;Chis!&;NsH1===s7_So z2M=k}_B}%(`MU*1h)kLJMFnV6I7w$;G!)pCGvL|$oubgLr5IkKEN~;t>4G<@LT9yu zt`NG*Up>XdSELKuHci88&ojx5nlB>xkp0#*c3pH*+dycWR)bzim&Nig1KGp(f?H`u z9ZGs9)cpFW%r!etk>&a|HJ!)z;YZtb{2TM9TRDKW(=_hdcah5nxmky! zEP5wOTN}-JEbu3f(pNirSXhoaH#+kM24A^D_$>#Lr(VLY7XjYc9>;C>0o!7r#wVTk zCjTn7&`rsoAS+8!bSme$NUJo8ucFC z?W%30GiCt{D$0T-_b$=Cf@B=rUQt0?Hn|f;L(#_cc?EU&gw(~|jdOo+H zSN;@G({4AeV?7iyGNkzqmj!%Uyag5=|0Xh8d0uRNz!2&>n_*9;9rZOhOic6i;Qq*U zAUE+Gj;LOMuxABdJ=lt9ugV0Agb$Fr%9afqzmGfZ`-5L6-x4bnheAivNV@b|82H;P z0I5<}`sAoKhTYU5@7{=Uw311U;?7XZ*P+e$WlA_mt1ZB=q0aof%TIhYQJo*o9fO1I zQiz|Y;QX{4!=qi_fU1yNyWHr31$KW#vA#<9#X$|e$y=h4RxDa)3BQ}!09y|d==z~9 zo>5Uw%Wmuv?huOf=DGsWybOIB-7oYTdwp=DtdJ{v`I+1g^hfiARLaz*!<}Q3>Gr~> zWPi+A93@u=Wvl1o#emDA>EDL%7oREyTmx z>e*_iKUG5x$z#VmNB;Fhtl&Meh4iJb#ozWWrj_R*oj&k(4kb!aDbp=wZ>P1l-2E)jiRpLoAUB#udhT+rW zKhSzf9`0G2#1#E}K&Lmw>TG@r&+YB!QS<;ee-#O@lGO2p>n;e~2mI$YZLEA_%;W3p zQCjFSHd;r+FT+$8dbAX*Q;V=s`aV{Cm!>P66vRDiQ$cw~16cPuvUIkPZCW8u)qZaT zhuftvHZTSZO0S~RvpI0EOc{o(nFHV5Qo%o{l#kt1h0{E?QLTu7@TEMOj(?{P!wZD> zd*%aC0e?#lj`U$2f3}j*0}oa^YTD9^2b3`~dkpqi94F>K|1dSb`80F#I2xV4gPdA3 zhPw~1!f^8xW|yMFmTL|qCW-RM^j%4g(EHI6pe!eDS%OC3KRjR>1Qm(#a4%Wg+T-PR zVptZ>vYkWN=a^|Eb)^x@9#ew@U#qd}o2KBvfs?WIgS*Hh*aUa>QSf6L3Q>va72iEbGtSloN zG`#)jobS4H#lKR%tU3@+9azsF7cJvyhS|(?g$@*sxLJW!Yig%-~uvqdX;`-w1ac_lk15V+$D) zCGbzX%XzzY7?lXPMh*X&P!)ayRGQ1+(8k{|SZM(BmY9TV(t6S2mnT2>P@j1ATT+$I z3#4e{C(xDFrB7`1!OW?h8626%SC_r2p~js!YUMAqFigR5Y0+G{d?2qHkj>v0>2n43 zTwc;1ZMAEeKm2&x1FGh$g)D?7H)>6U`1m1YVO}a}yk>4Ku_T=joM403-nEf|GxF)% zwgJ|bBivb%;vu@Yay}h7*?|t47|AZJqj)Gwft6=+qB%yN24;+Q&#ey{?~B-r3f-# zN{*+j8wblILu%TV%!Qjf*7EzG6gZ6aq_6Kf)AA9$R=3lH-BSHfnr;7^{`5IUmnoR> zd}A4maWlmaZub2By5Z=P?8x^Yd@MTi{TH4uNn!PK!yzJAR$Ml5INhncoA(=XbX+_D zmhrb}n-qydt{36k`6n?p*PAa<^`S*Hiz<%00Y(k3TwSz{#?HS*)7t#_Kk;_4{GV)= zG}MS+&KpIyOPg}dhE2Sw*B`ttePW5ivmsUb7){JnNu3!I|K=j}Uv`oWeCUduwaz?u z>QkIMuo?Sx8H6r~=i^(Z@C6GG!uJR{ICVjl{s;GH$J$KXfBqN}yAi1QempOC*oaP2 z$LXgFcj>-+f}`-^4Los94#q65BvEPktY1?M`?tC9H#es8Kedx;A9Z}@|E_%GZv(d1 z#@PMg)!jMRB%97xC{5)npEQDf!vmZl|3~zcg!7%Xvb8Atio41lW7}@JvYqW;aJS`=ZrhGkC0Ri*^( zl07hHY%(KKQAG247i`ItfFDYF*2>BicqB0fGt1h@8DlBFQY+UgYPSK5^_HQDjrT;c zYffNYg&A5UwLs{K67uVJBW?fJMf(!6py2mIa((^~diJv-&suj3hd8#N+U*89b;5hh zY#c|;uN-6_BtB8;mmb!Snx4>))#c>Zk#MV3gF?vcC0lTCXBu(o+{C?lp3qdI;i&k0 z0KfJt0#4x_Qruc38tR!PcKG=di;v2}$!tlW`;5uV-%Bhfw~WMG{T1Tn0`q*``T$rk zM2YHc(4#wiCc%_t^U(82h*;ZqBAS$xu~nNkVN99Odr;qkEwdgF{mBbxucH@D``rUe z7t9BZd=vaPA|4NITZ!H0V#FP_CoFQ}_i~mXJ`bBJr_m8o_SWZ@T3K6H4Htcg@uMSV3huBBdv0S(cu1oR zzh;vwO57U@a`#N&nZX#S42uUdl~|mVuFp$Ul*P9;X2bK)A{OJ?K=LCMal*U1SdqPe z#*i3lJJOKW8HLl4+L170&;;w(#*Dr&Hlr)L<3;MetKrAeX7TA+7PvsC6_YPCvL}9r z=mcF^erUwynubb_zsikCWzPiB;kQxnV8~?JY4!>l=Z-~vxAWM#SDx=o6c}6U#Q;K& z^ooTI*9jVrqkZFr?A=0kJU|O_jO2K;;UkPVr;CT()8KRaA4J)6xaWu#+I%yEkyC0! zBsx^=aW@s-HrwO#CKGNj`aY5shB&c#I2Zdy@dAYjeAn2W6aJcq(M6vRz=iK+Vl(L* z;$sSBBz)pRE4ndHG~M(hkvzMaey)PK zv`nFzgVuo46J4y2E5ovRck%jwne5b|N;0@PQaHn2B0;N_AgND@j$Gl&TRQDfa$#%D z-NmJ_sWTGV%aviQL6pcjX&Bikt1aH}sL5)7m^Pc2bc$R(I8!vodMm1K-B0dj)DSD# zVMt^qz`yJFQDpv%JT6PmmNZyz^RoZ@V9CkTN0j`U=tM zrW;x3sKS5WAHb z)w;BO?G-X?W)!z=ND}6*ThKOP9DV&ESMaCn)zsF$!gV9obLE=@xcpBA=3){~Uh3r& zgU)lE{;Wb3UQ5zHaxPT0zMAR%H;(GtoDubY1=#;p!rE-wEov8{Pq((K!{ySw zqDuvn;0F7^Y~CKnElUpK{S8;y^fx~6^^maVu=IoW6K~+L#a~dG^OSTupMvfYvQWRG zfX?$9L6==iLa&iAWS)?L>IiYAYi!Nv_?K&`~saE)Ovhn zJW*7c-iQa(A0w;^2QL>3o_uwv;QT6Mr|&3|uMw7XxyTd_Ek6YhLL&K+tpc;Np_OhZ zPNq+Lb0DBDoxb}bh3U!e;*Xk%L=)f&Fxm;{OdOF>|DUT4{U}-*hb1t?vYa-+BLUk zR9MMvbD{G$8Q$rc`SCBCuF7PnbOXG&Q#uT7!{;3^n{@dm)!UbXL!%%)>kVqOXD;qDXtUOqZvvN%>9@QgtV;WQBy~Y^uN3W-Nah9Kure+ZcV|!7ag>4gbQ6|n+s7h zi@9IS3zXK#gS#+}W>+1+4Wy67uzN6Vpe|jt$bfEo#^IOXbr(m9aQdSctVQ(# zaZpP~f1f7N1;zVfyl7LiD>OsARH;^cP4X7O$~f|%C03+ZE5`C`>3GJqlL^jrEGT?N zTGeid9t&;&nf^x**tSNzTAap{2JIjklT-25gNfX~`!E(C8qTVueux8hR-v}FD#xGQ ztoH6j*eqmrK7=@6VdNV4cEO$;aB~KOOi2hH*2NSDAHhpb!9-52ojWg$V6RQ4bF~{y z*u8KN?y>n!KB_A4xjI`g&hC~tt1AkR8dbAMeu2d!iY5>4Bro)a!MdS4sQ-UY+5O3v z(Y@VL{OraM68H+x;`J3#@JS*tgUoo=zbL+bK>(jV$elan8{zfXkt{hiK@?Q(2~Oc- znZlVF^jymdIQKx4YgC%lC@4#cRu4VFgS0JqZb1+(8$3s}|J!`gklkhikF%4s9aa-7 z9?c_FFEr>#p?j*KwF^7ko7gE=gc}JC;i^U*a_{E*52RU7g-QBT?V zPa-m-r;X$uQYCqzpTysun=yW0iNif*+`8>E>9Q7B#((p~7TIU<{m+g3k@BEcC z59nZesfVyUEsNb+(m`C6)9?$~g_5VLxKCg)e=Bi>-THEsZ5n@zpG@%L^JNdR3C|s{ z_=z)7?!3d__?Gf$4}E?*U+9a4ePb`Jf8aN>I4*I7@Cvm=TJ^P#+1%a_$t4-^>-Zu5 zZruca=2AJnym%1z{4vC@TY@oXeTtR%sn9vM$b^Z739$ZPH(7YC0&e~1BYvioifLV{ zyqZmgB1~f*?q(QrD30y(+`_0zF*zA3I2q-&aKibGIO@?2Ecvw?gKfh2jiyfG*7(Zm z(V3Z05Twqx{T$8wwlpyHqeHMy$&R@R=zbMnxPBB3y$LzLDwgLzX2 zg6YgqnE0WZHCgI0$q{uJbLR?VY2V{PM9gg7O~;j&%>l!jQFFNux1$+W&)@y~7oOMaIrll&_4&NtCzea<=zjh| z*vhqAZ{dPDcO*04oo7}v?_$=yEBtHRan`wJ950(ICzfT);pFiy(B3BSht0?M0f$k1 zkMB$NK%A$pbE<@!@~R+np&u-obe4gocqd=^4kx}+p!fehz`5To_!bjI3LEBPj#?~U ztIL8H^TKfdoO49*`XqAa$CK5@>xGa}X5?{WtOWJPb2LV5eshxKc z-QS?hn#W`@vrd9}`4UoX=8c=38u^S{pKJErSHRG#!!SM4f|>Zrz}-v-diC=N>ilOi z$j;cqQ@kRb<)M{W3-nsgn z#b*r0eyi@0mTePx!tGU(%x%Z`DjyrvR(VVQCBK31hZ<{gr2zg6c!SfH`N2|q1&C1l z#*S5-V(RD0VZ~acff)nQ;dUaEl{Mt^))bQwzH5bPt9yvfs2`Gldflvbm=6E_B9iRd zfl&EqKOC6YMvBE9hg4-5dk}pRwj8}Ebi{k2)XpWOWuO`c%6}p|o+{G9-Kn%UIf(8n z>Hx2~lit|8VfmC>+c7ishP0RVj-Oqc85CAyYz6z`L4%+t4wz+TM`^5wug()W=+To8Ad zo-e%M_V$Ou=fGCxbKou5eVGn_Z};cpeIKF{)Zp4-iu8k93$2fz1nZ)r=#+d{S_T^d zIwwJR$WDpM_N6Fr0IWVlDsO77ovOL?+ysNkdnc(nT-r=)dXhWH}SNKI7K2 zHzTC^3B@I3Oyg*J=ZX@QJ?Ti#Js)5@+$bL2{L`T3?GyN#@Hz0b%n6r2v4&qBJ*0kC z7<1}(7*Yp%VaVHb_Bkeibgo&9v%OMz0+p(bwwpkgbuB8sXKv|rO1#JJBU9f*3fePwqj~xV{9IzjjJ!HvlSvqrFL^?9W2fOs4}msU zKBRlRhYM1&p2F()8~7YmC1w$w2ZF0I_1!QLE-!vW`uHCv3SZa2@x#k#s!A-Kqp=r0 zW*y+GH^lOVFK6+O6AaN(d$58o}9ZC&$;?bF+1 zZA*-zNZfcGTG3dCuV-AAZI~!kxkmATxV6xQ{+YWbH!forfyy*k32A)%2bA$d$s)Prpgu zEtdRoSQkpmM&PDAML4E#l{6N6vq=s=h}+*Ph(7%mg2ETlYX2TkxbTThPMkt3L^3wq zYGO;1im2_pC&ViIHk^NV1d`-SaipgucX1p*H|EL`zwB~WdUXWImuWD!7#}iGk>ZEj zJ#5uhCt+ZA6sX1|LsH>Wh`3({CO@L+q7|C-RQCot@l>}k_#PwmU$m&{qe^mZ+IX1v zsz3W-Ex{U}3+&EAGp^?Q9O;+8SU*G_&b&*7!B-bk2Mvj^a{gF&dD)GYA9sVbg{jQh zsGUfP&amyr-=S_}4kmi4(P2Xmvp!4u;jgYybk?#E{$#~5_O|{gm8tm4@+~uo^tvc< z)*X#~#lBIamjc|(e#b_@SgJX9fbFwOm9W~yfE-j3{loLh$;$d)AYFKvDhEknF#vo2 zYcg%Na+etGkOEn|k@VHo2uPDz#cZF*kTcSa@}bclEc_X&qA zmlw2XBMeyIGJ5Xn736;*~J z-|Q^<@n;9^*ZWZHf7J`GM=WCNEaYtak1nMHT%~N!1f0SB-@B+O3%BjZKC+(|6D9hU zx4^4Pn_k^e3@7Fu#LAsV*^eqMNuwl_nr}>iWqhj2|Kc@`;B)KjcV@C1PbH;X@9m$^^t6+6$MX>luAG#|}ow_9rwJm+GWNVV^1YT42vdd-9 z_{Bl3bm`7v1T!U!l~%tCLRmXs4z9$J5f)%GB_3B+YBk1ZndjFi$;(op08Y=*D}K z$)aPSz5hE@8n42?j!S{)du=dwK|PT_tU=|q&2Z1gv3x;*3<-?y&;Mwz;W7`Ri0aM? z48C%QEgqH4eA?u>)waVsPfgBBRItNh4@?()X zzTB-Wnaw>%(VyuH_WH8u_#hqC7uKNCrCk!Ye=@v2^aS^Qxe2GqYvTmHEI6>L5RdH? z^Lu^b@MrYF8olbr=zlX8ee??00g*2eYqy%VOmIc#@^-a1vt|<0lY;Ku z7>SM66&r1LL+)5Syf)o=8c!>`LpoMAfOu?5UMG2g# zN#xc4QTJ>nqSR&9)j zWHu7z))BnU@IF3U=EKvEl;P|9fjH1Ai2d$AL1O3RW#e;b6n|;dhHDMdNTxv|zMkkN zI`>NP%&JMKc=84p-kjp+diHZg87Wjge~itYceSQV$%3nUoe=EhrwTbmm$7F}6z;le z!@mwkv~@hk=GBiNzq2aQz&HuB^B$xmfeB{r>~QQX;+!X?U7{SA;PBl*rCh!4))r; zmmPgOkZ0KY4K;nH z?uQmJ+ud+aniv1KSe!Ae;fpS5aQWBcU}4ZLXe+j3Irl_pzOxKJ)U^xWmIbl;jwBg&SXbB71&?xndHfWXTk~Z`x4sH4`=l}$I{&?=r{HW zu?xM&{+YC}x>@$({yUx)PWHsN<3Er|v*l^Ws+XWRY&!SuEW=nw1FX4pl`UFWijyx+ zrPJ2>)4ArVw7+T!L@BhecPpO48RZ~IJ+O?*NoGne#@qz^zcV3qe;HnCE#tLUn#7&Z zes*;aVx3c%=mSyWFCz8}jYC>kZAJjhdt*;$wFXj_76&egFIdKo0XU|h1f-OM*vXp% zpliJj{kMApjhXjNGJi`7WWSA-9KsZsIolY26kWor>E}t$U|H0XwZKu!UXz)5Dm;qy zLgCNpbhey$Mr)MP4<4V`Z#yrpvbz_p_pBppWHVv5^B!o3Nn{Vx_DD=ZSCgkE z!7wHa@%aA3SUb-OF8py4HtHrx?stf@es&SsgdZ>s2r|Mh27S`HNZOs3g?hVaXd zN@VobqwLSOcw*lk51;QUf=X>GJkIPTU3eMGb@cK3KndH=#66Xj9_x`0jnv$sh&Uw@s6;lAd24;9q7tNF?`y$FdO#C=+-aG0D$>&R z{MpcDJal>sbjgOX!OmmpkOp@gvSBI~^=(J9CT+BwJQ-EK^b(`|At5! zlLHwmxN()NxXX)Si!NlqvwhcL{p4@3Ebkt;2|r1Io;CJAZiZEvviNI!FnuL=3EY)d z(}3M4=#*=HSwX*4^sef{q8D~N=3FwaoAp5QM3TXzH;E!Sd0a&cy@{OI{YMB7f-W{o~>BFL7m_Bbcd~H zYw3!N6EuT_(+|H}@R?A~&lQN@wnj%<_WKCUp7{n;YY+0HnXCHp9^=miQuNh2y2bN;!8*^w7=O|Mm z(s94o=^__vaFpj0wEtv8??gCb3~z;gjs>K8@@+`ys$rY@n&JwnyX@0D=OT2Gb%W7w(;#TsMi8r7B-dmjzT4tq zbJunp@jNh&m`;>qG99|0w@%K+VS%EstVCoARgECo*Xl|Ct-UaP<7miV@kxX@|A84la$$^?930M`jVaDWP!wKAd{<8*t@?YbHhIiGd76q4kgG*Mxn>ImEmK&lw)l3re$txxoU3wBm!{HXIdvwU-xaAUU|nW3pB zaamX*vfob$-KX2x>L0gl78$k?$J8anRXPD(Gz_VKM|ib`R0JG<;8YXhbDDy-pV!)zaOx4d#t` z%~TAc$Su`q$|q;N?Wi}_r}JSQ31G>CnmIGXF850dN1ubPv=Dw3G>8ALV4j_q)5lU%qmh|XLfPr>d82q8Jtx~mitV?)9Cd>&MoHi6e4 zU17hU6mQa9j0Iv=IeNwtda%`wRPDY-p6GN{tD1g-pIWuVKBx_ouAhPIl6-O|WIh}7 zDwQeyo@Ujkzfw@yRY12c%cc?oc`o0v3RY<-2v?7lg5$1D^xXIunztqw+(&#Q)@_b> zyP=-!oM+D7W=;_%+Yxekxgsn)>`+ZR`(V{eRmpzk-_^zjcgd$IN66)P21bAGk|&CT z2x`07G<(D{w^w6HXsEY!!s{w{xY(1<8EZ`&V%4Z1?M$!4r;2CiJ+^0}4$SL4Dh!yq zjxKe`5{A8ANJ34Y5c%0J$(EwUETrWaJE|0YCiv9Un%{j)1j&xIC_q6o>D^g}^t0UB^G0dpwYabS|U&Pu-%@pTr)o#tm3Ce+(3>yOId& zZW5>d)_TRGvBK`Z4`A`|eEL*DlRtSq8#WJWB1Pv^an8Cn;r7pul8H=%oO^W%dZ*ln z5UCV8!XXis<|mNH{n8|3GwsPsj|7R=wum!K%TP$Zz$IjaD%>cOhsdPKHI3Ugp3ylw zRC3N>q)->a{=Z8d;6hXQy$ZzR#utHJ8%BGP?kFf0l<0}Wlz zxqbcyZvSyQ-@y{OrZAizsgPhxk__~g3w*|+5a_g3kLrt*9c0ZT8d4m0W8~}Gt4}}S)bLpb3r|H?{yXe{YbhP+*0p0AXVb0ZT;ZRXJ zN*xL0f{qGZ*?%z@Px2?XkH(?TDRT^x?~Cna*KzC4>rf)?2#)ev)F*5K|8Pr_s*f$e zLY<|2cYF!1i_^s%`4jl-?<_uX`Uk=J#c)ac4t=gtZppuX{U_$1cVU2T6fxYL!~&29<55V&3Cx5P4`Rv0YlnMvOg!zt;`mkJ?rE^2kBZV4Q;5#l1LZ zhYL!jpBI9B8p)wm*HQCS2XoV#i5J7~p~|7rT#}Rm*ZTB8cCisETxi46D+{>GkNLdf zatY2cOT(!X4+)LwgYfcL4_*>5oyS#bLixvoxZ<=cU8ht`hacaAMJa%+suukk2lEQ~ zz5IcqA**Qh-Jao%Py!fsdhcz)_!o^Ii?6%@t`hUj7u_uH# zVs0&SOtSE>NP)j}&jD@yz2w`iCt`m4lR&uwF!930L2g!z-m z>L~I(&Xd;P$)YjM(ze5ooWz)0#W2lx4@^>u;=xbV`PZTAxp&qnyc_fpnoRDI#ZR0e zKH)JQjrD{#t0fqyXvW7!PUX@HU-60T5{!}FheK0+FgkJye{t>-IC}`_@xdADz!vp~ zKZl-E6|AQ2r?91RH5+%ci8c3{;-rNB++g1=B6p{rWPBS!lxG{De#B1lXZ=s(8M3`sj8MrT4_xoB<3q|<3!nRYVdy$z>?_V@ja2&Z2!n}|2VyR4ZqiIn9=-w> z%RkJ?{UkA6vK;hX`*NwcF??=r0X{kK26hE57BcL%flOsM{O!C-?nr8CvTgd;G_P>R z1qVaeXs?N|-9+RiwH3mDTRyp7h$~3Ao^-4YB^E$O6Q@VMV?`|ZO9~%7@*-L%9mYkxguE+!N8=W>(zo6J zNY`CYmOfe=On*!Ox&8g=y_sX_9LF(CWBhNjqh%nyoVFXKmgW&1egb5bRWW>Oxv(JZ z3f1ydpk;|8sjH>}O`MlgW2WJOTRKle#>5>^?eH1XPhKKD;{#}jm8qb9{uq&K9>O0_ z?n6y=7DJ$81%~;4kjyccm5}R;(1#4fjuHpS)Syw^>bZnzZ#+vcjy*(Y|B`{{!)6Ju zr3Ucm;kD3UaU70qJZ@ubdK0Rxbm-#&BP10gUFq1Zv2=cTES&K8P1+_7mAqSfl9xTz zj-nOWlZ?Emg2fH7X7&;qEM&)?Wx2ZDp%-$%N=~>8+1?{7S zw~(~Tmc!b$+oAv2beg>~41%I<*tUIYbeg3#HNWIbx9kh0DSFT0)#4=fXJI~*s|!cB zF`00>eH!#ypCmi!ezEVMN{9a9Tx!EZ`sHx|s?8cH9M;SRZ@B>azA&CYRXGm5t3vSW z@+ai!Np*7CD}ye%QjAqw#m$f zwg!raF9e5iEp%nfc`9`woHp1#pq(jgz{4I9k2wv@AlhGix1R!Ctw8V;XV0sp0>DCN z8+CYHfrFhFkj}pyWQgKkfT1lk#Vi3V_Y4J=Fc-}_f05qtE2wv$uR`E3X)d*+3Jhky zK(Flv?EI2CDE{Pw(^}T!L+Zts)fB;UD-9xlRJB)>rUz9-$QlSNzd z7gMKH1?qco1dabTjt;w(0ACdh=pwbzv}{cDTsnAEi&~w7T zA@xvkPwafWUQb8;i=u~>2ch-119Y;j1{kESM>X-SRqnsgHs#1^8*$#1bPlJ*^8%35+G8h{(IJ+i?TCV8D{JDLx;kOa}yKXIN&pHD8 zPKrE2rQ2kln+biH6)9P*I0O{G4y8-Ij3mhh79`fUL1ec!i5!d(bnW~M7FhcT?FYYy z`*UVvscAXJzR;x$<3$+5>I%9pXa!emG_8@1)*)|2kFMkRDx@?WR%A5r)MRP;=*E5W z6iY?_Y$-XCs?2vDik2K$`~-$=-$nvgC_>~WH(a>tCrDhjz_M$bP@>YzZom9%z1Uy@ z{+iyzOZF6Ui$fK>O>rbO^HLXcKdrcN+Ax~1K$gCVSVs20nacI=DsZ#AR#*|ET>DkK znx+k?;LYd0V&eKYINaO7a)9z~G*eTj1Mxi0%#3DbkENlS4hNe8Wj5E=irj9oOHmo1Nc;jrJb+%THK9pY$eGKVoB3tLq4XCLMXMx9?@bD%-vRU6i@IU#3ObH0W^?i>>F4dPYU7saP zs{9dL%{_+){r&j+l`HtY8gIB!bRENnv=F_kvm{3bI`W_i4S2uy2)lPj>`lsVsBdTtc^;G^G8M3R&j#(`cV$Dir>i2g!pA$<)0o_yu4?V)mP$T7sFeMzv{kRWL8;{~R8BR7{A`m+v6rQ<$WjRmfA>~XJo6xMt)z6K<&DDqb zwYQpl*6z7{@-i(+Nk;^nj9$&$3(D~e@G?BaSt>bpd>8$d#$TNEKib{`} z0e9rB_-6Yi7@Ag38oZCzd=v8=MU!=?>b!F}$Hz}{Y+8RzRk;eYR~z6ZnMkJhx|b*Q zHgerVkC~a=A=t3o0Y(j7FX_EDkf(fphnI&MksDb_!mE+j$>SDxVcV2z*D4J{yGQbdv^DI=g6=i4MA=!S*!IfVq7_{gL$UrVX@CeL1{-kow4o*o&BgUOz%;| zi>GD5D=;7P0^YD~cE4}~MA3f#EucP^%0xF`6sEP$WsT=I(+YEW+su`%sFPGA?CO*j z6#nbO3*XE_|K;sJY`yG*TD{ro}>erQ4c6N+4E|3LBphM-DN z7G6|;Ei9a!$6lqp6S`G4+Wb=)kLz>e(0;r(vp+GO=e_G9LAo!nWKlM*->-u;*|UiL zyg6|0sVsjoU*y)vMu~fY`!qJ~BW*h$jX@4x?5ERko~yco?09aDi#mEOi!^6cW5i& z*<&{Iz@Ve3mQ68dovNT~5+PaoFqZlqywCpoFB$irddKI;Me}omM(}AFcbJTK7EjqE z@F=n0DI2a&2Wt(*s#}5h?!_hMN=&V7`kWHI*Y||XZwJ}QgctPWyLr57h(m2`j1$zR z%_jXfjf3ughPE^*o-SD80wa<`z&!mDmAep+vIp{L^u`kWHQ+2;*SrzEM%R!Yx81mE z)j^E%QA3^FO~h%4DgA4CNiwzi412yXg2_!18SVdCvCHVQus-r8R!zIPFg2F=CftCMwgEdIjQ}(@F>4TVlJcKuj%7 z%CR{X(8g9BaKYcd)X*zq3Hp4Q!dmR4c$WJxe%{rJU)H(KG`_^MwQF7pA3+P03AsxDQ)FMiWA1Mf9Et;nE%-=Em?R{@v<#&s1GD_#GMRw$l zQfE|Y7{Ciue@S`^;^{TnA+)YI9iwlx;n%YNg!Rpzg`jhk>y5~SuQ3)-8(WRNp;7#3 zpJjaQlnT0Z;X>M@*TCeV*c;h_?;ahq z(Q{W}hW)Ox#uc)Z1mr`^(IRy1@!{1L8+g^w@Y>rKQ)>(4?!lmq?YJXI1kHZP=W-%5 ztKVV+$aR^8)BlN`n3|I;C|Dny0;Dk9JqtQk7^73|C){#ap0Ap!Or>vA8vMY4^r)`o zizoi%8MA=zeAXAfuRO_OZ6r3$OI(C^f1lv~3r|Rg#tHO=rKofEHamXd3A>_X$^F|G zFngQ6SaNg@ywYufqOWmS(RCm0kBKJ%J{xGs4#E=-hVZSqdR%^sDX%b+=Z?)XutHu3 zZY9nKQOOP3Q=STy0nf<5$LE>a=bO-0?M8Owu7@ex&a+bs7GRc5P0h96?Ig!FkDNNy z#Ha0z_%=P_B~O3zZ}W|7<&PVP zto5z<>VY=Bw*EWx{cK0hn&?TUB-^tS2P(<7o)R=T?*vo)rD@lqM{wuo6G(iw4ODVJ zkRkCuF~~cN)S;<&@Pg@MqBrPNALgHkK!LVWu5wW6*nDa=o7$PiYcNI{S+KRhV7tReYAKO;zGqf823{o;D{rX0?es1YR0`1$Szyqxna7 zJdzd;eP}j$UOrl)6f09nY``{sBUw#Y59W|kSW6yHm`-_RFRfAm{f8nuxXVFQenBRZ(5PPyTvEutua`j~joAK}+ zdNB>Y;(^CpXD>p(!hDk$ z_I@1l*;{sS&2f)$VwV%X!@eZ{(*n@CFvcXA}J^LutJGLvqu%7+##x;VaZ% z;H->ptk95whVCz9S63Lg{7~d&H5S}@!zOVjwgz6urHa||g*JK5-QlPAvFfNyGfDc_ zfy{ZGGLwF)B5BxhpVgilAbN1G6Ge=~U7z~c95p`on&=_kJe*R{Lqcv!8 z#ccZXl@qF+jpW|{j*>@ukYjWlcrOGFF(K*Z*BIlf==IaO1 zgf~}Vs=XWOUbmFj^p&ffl<0`Fk6Z&!rIGZbW0It6gFwH?PoOim9pL|^{o|eEYcW2` z0G|z?Osb!^N+!R@nt6S!Ala}MUK&o4OrDcOvp){vt^Ym=Bl;|)Ue6S0j;#gw8~$QO zbJwEWnUPdtwgMZbSmED8KVbNr{&ah>G?jB01A@qZnRjdtEEO}g+oNTW?>H=(IIbUW z@ir7JYp2pj0aNMIy|Or3@ez^tRfDC+Kyu}}DGdHN5dGSQ3a1xqp-%KOh-r494?Z8F z{sXUp9Q%R2ZYN-CwH>`!?nsx;m_SWirqb})ui=pU8|bl<$5De^*#J`ooHRn8_48=M zsx=Ql!)q{gEnNt)m+kmsTLmsTWlSx#m+^yvE4co>OkQzZfxC2N(Qt9x9jbN8Y_!g0OtR1_C{xbbe43?DQ`|+0~GqvZ_sO0^S{ zhletLx$8q0MWloM>x-a!bT2b2P^SZKXVL!#d=d^W-9Tji&yGAh9*a+y@{h8=@MB>V zDD_`Mo=v?*9X@*EPLFv!d3g>x{~bV+><(b+fdqQtVqaTvbxJdC4YaL4J`G1|$)R3d zs-$l1DO4qml5sV_o{k;}^8*G`>va|)&o>xM`vpi85~OgZnl+9yuO%Ob9hQ9Q+)rC` ze5svH5#3oU3qvOv@zwt+h^+WKIAE4K&rDCHGBFyqi8*q%9mBP4^>#Y*ePJ(ob%Y7e zUny%_Qt3x`>=dZ?LxPJ+ijTziy(vsd_YJd@Lm1-i>Yl`O&QZ;($*($vTw(VO?!J-M#1@ z`?5%$$~8~6y;@{I&n?b^yd#m&Se0gLHciG>HEa*vZTOtb^0%O?DvRm5rQvkbDjOQ; zb{wqgt`qAN9eO80mbwP4rB+`{L1Dm4R`uxw_&qbCdAdPZ8+#h;Ub)kZrITU!pCl|+ zx5pn9hw#6Z)9BJK&E!t@7(9kHwkVelriKpmw@)5Pzxhe1IQ9=7Hi&+ePbnn$`*+gZ zl}ChaSx~-cFiX?*qZf`QL-}-|jf1CE=j!dmJDPupQ(qJQH2po1)3}OKQmMk{q`|P) zR*J7WyaeB!tYllu8dzlJPcrY^L}-+00m~c%NZfOpoL^bXo^1{UE7ua?+mNHgX#5Kr zJS&l&u2#1Ft>*^n_axL{?@1{C>qghS(xTqGJE$jFPM>K!rcV3JY1s9XLhzy&r2o+8 zkd(Dha&GY;dgStB`uW2o`g9W`_m9NmAo~rlWWN>M&*?*NoRdPi-h4bZES6G5q2`M1 z3798W#=gZ*75mMN%>8*2o@>*OrHzCl6&}Vf*(Us9REw1wqNAE-e-}uG+`0WNA?k z**3OK!-diZ$<$!iLi)zi1Rj+qLwk-DpIEpK#`nA=o6^qJ6xuuTTb^;$`{yEC;fyJ6 zO}3qM`Z>n?KnMV4%(xrn{Y)c}~k&eMSbdl|F`g7W3oHHd$BEM@q zX=Tfv;%BSbvA{O-Wj5U;=IM{0 zJPktZZk z`r1Nja8R0-nafh2=qwVV>>={8V(1w2v2;O97i`JU=d(js@_CxtbXm<|c-wmnRtH#9 zmF`sRtDpy)vSp#X=_r|XPePgta>%^|UAPe-1@~6GAPeR$1=E?bu>QS)?Vbw*ZDj}b zkQw9pAsn!zy-`nXzDt|Hsum-eR-$ z4>?mIC3YdgL02;aPq_qvl3^u2bkOFXwXDf>zZ)37WfB_i5%ZHR0cbn=G!&~0gWFe9 z$)gLJBI{L)yN`LnsOK|yHoTt9p0S1giJvCQ?3kxt+{SphDvIz7(7Q)(B8IW6J&vsiB@ajGZ zcsSgE6!+OdBHtSGtbO9nXz~a69Ft9-7o<{$!Xol*rxu;$QbhEw$;0pC$M8mo9!BWZ zlSMy&K+}*}G`Gfq`gj`hOt~A(DkvLEQvA4C!5?T98KxJ8RAatgutYL$CsT-3r>ciX z@Qu6HbDM5$UhHTO&s!hh@&DRzjmbZv{a`qZF^pu!@2VlB*c*n1OrwjFM)2CdbGdY~ zHy@yw4vwqF(<9p#NW34vrtl+wW@R>i36a+m1unwAYvTJ=)CQXQbii zv}UMF_F)63rm*FMj**U5hRN%aiA#GVU#&9@+{?mXze_UN?mHP*m`dZGhCodJl~Oa~ zP$8<_v%w^3Lq66;WY`CsVgAdbNR8QORw~y)sub<%l=cSFkhoE}{oeo{dO4ZR?{>8r zzqyX=lP)0dA1($Gz6XaLuta-peHL@W6b$YbVCn4y*qnZVh7LVqegD#TwtML?{>w{) zUK($N9~uU8w_~5-&D=d?{FdWZE9}M+8pu3c^2^&yX9#SK+snlW5W@ zDbOoU<^7i|<6nF>@*r{>e(okRJh zfH=6YZV)fb@PvTT77)2wnQxs`DTxm}1HpUD>E!rfT&`*>{JJ%bMs4~Iv;SS9gU+wy zpWK!B_svuJWi3nkr zw|x+P=s%m=zSu>dlzH>S@^v_E{(3$`e+pjwmXH6ndh>Pm*XV}hnP9gq9Smegz>Urk zw#hqkYMf8F(2{Y_@OR8kD%)3=-ks%#a)uo>Hycq%7;y_)wT{q_mTj0e|FGz({y^GV zrq`HO?t#|OggQmY|}Kq>b6eeyVtD5qAYc2pA<}k z!ygcb%f__8RODXk4W)C2o?`(`>B7F8>o{YND1}b#OB4M10(*ItvqE+JcwrS&A0y41 zBu~gLg;Q`O`Z%gjuEq2nOIXmSd#D+%ggU;1@OMKR|57`Mzr6n(RR8!&ob_+g$tnx! zvyv+QpmZa=iZVeDZxf`W!^xlx5_dRJk&Rt}g1#jmi59W(+2RC8( zCo}kHZAL3bYU7Jd%|zbp7<3CnWXs*ePwSMJ^oC^XJI$*kYTjKmp(Y#?<|M(hs)g`h z^kr1a8p}racCfZpoi*tJvfNyKEXsxaLE}Yd_^8ptaO->{9`xx7f3SNrS6Lx``}b$F z*_);c+AZ$TQ!pCWw8~KpxrHJx=_6yF|Ijbwt8mszgKI>ywNfufw#_>v1B|d!cEuMNRn15e;63n(6 z!mG{N}kwUhL-;o!QW*ypA6% z7+D)K-Msdq4&|j+&TSHHNI;O%G415a*Zb{LLPopGa zevZ{|>m}>%>)V`K*Vk6-m4&T$Q8AqrIh6XwFiKx$LRhvDIF6U15q`(u_L34d{L@ob zTWN?XuXQ1E(I4DTZqeXr=jhCCHQO&+_S31Slxl^w7W~UTYqG9uH#^@o3p6fE(c2m7 zT*>{U5M62kLEj?D={{|2tDYq)+h$U`#8L2hl^JPY?jUl%@8Qc?_t~@I{UPjxB@Fqy zlJ%>-MS4ZX^@~*{e9>`lJkj`=Mft7A>=`Ch%Uxu_I*oy?V``zXVl-0@m9JTS+6K)e zVMyQ0qVC#NsChqw98~GC3A_~xsi$|qwb0XK-P4yW-bM@m83h z#vg3_c>(Vj9OUKhO(@f^03U_h@%ia>$bwhGpBoN*glrl!T=*HoMvD2n&uM&pRSzBz zrtmXjMylk|CGb!(LhUQfVBGgH#Ae>drZ4#zS0B!A42|Gpr+k8~(tDDDxv81FcG#{{3)1Sj=|qd9{fcwXb5!bZUfkhoy*b&j6Gzxo@K> znTfROGt8geg#MG>;L*h``0AY;_t%nO=!FAZS5+Io;!FJ4Pf_q3FR~&%>uAm>OKvl| z2JH`tyUSbK*$4N0JU%ED2j2|GEkE-awp)ttQF*9!TZwmtFbooR6NcQ1zS!jpL1)gv zlI4SVz|Rb^C$1toyyj!z(W~TeLNc>pB5QT)E?n^_l>2Pk$2;_{VCTjV=;?NWnUmA; zOQ{O|(d5fK*L}cL2NnKd;2yHccQ>XR0e`0Jh}K~jFur~_k*lA{`xw^ZZa-k>MIK@1 za$QI;QR0gpKNP!_`uw0c-wqvS%r|8(<=^}+;?ke7VEDaUvf|Krth?aLy~R#(WK#$Y zS*gp-xAlWA(ce7n?LTt%@=@|kbdcRWqs+z~9EaNia)iL>7Z4Gm%^wUj;eYldKwDoU zEIc^_KQC>OWR;8O!I;g&c}};m<3%*T@;`>o!!O4-3gcy!hK7_x3Kc3c(tDrtqLiX+ zWkfPUlC2@3rLhO*+(pHWYqiHR?FG|WlwLXIjlFqz?@^4L(w1GnF@sn4 zt!KCUXs{}kH)P!HDDarnhq-+?#{*SLpE_V(ui~%Yu}bd}Mr0J^Qi2iRJ%IXZ3w|vB+_Yxn+$K zd%6EJ%Zz?PHfto4x7#&%{ChQib@^MKrEANLUH34u$$)RF@kQrEai5!2%@Z^Bv6O0& z2k6ksmwqVV({9yRxopT4+|{;2LO}wTnP;<(jh~t0ibl57%AK6aRO0;xooBOFhcXYX zOl}itz^zxSu}5{Wq}RV05b|UvC|-QU*3bXU!e@SFo4UOC(U`gPxlcdXd)$$cvF<$g zzn?7ekp+9Hw}Pc9?B}JL>)6%93)!}_hG^ROl@8foOEvC)=goMA8)urch_7<|P5OE^ zTl8aj{s`q^<~G#lRe#jG`kjXSIRN{|p8)ywE<9qXSm5IeYDxn@1X%*$;Zs z(G;4p2fzhg7pyRFz_8H{a925$ZJ3h+)1SG(+CFNq(j{K(GU~(UPIq)jni=GPV zo_bjO<}Xe#%mS@vdSIC+50~rrKymYWVp3ttGX0!cx~Cj(x)u-iiJw4I(6ic~p8>-@ zrJ&`Tf$*zx1e$6M{UQ7VVm<6iDPlBWW=)==G8Y#cyQkkG|fiC z{*qpJ*XJwbS1GU?Dq76b?jD4@6@mP%b3&{7ONkoo4{8zK?4$Y+rmNH&T^$_3a;q_p zS@{vz=dhv$rspMn*XY5b&+FmB*=p$Df4s=onl33^l?B(w)UquBBk7$%2Zf(~PQ%?+ zHCStC0cE|OlRK}gakk@J?D!A{>+JK``K(kDZa9wC8tj+8{diF7IN3~E0tqPn_*Rm% z^AMY7bC6xhb`xIMPh*m{X~a5kIppCy<+Cyy*GRh9WE%*Wfr1%FJABsV0X|iF=QHobtgy>bkgt?VlL}OQ(&) z?c+bdj-`RJe?OC9X~hgwI@1s81Ir*O z!K=%2$iW?hQQfQ_aF(DK27HOdDGDwB})t24!#zKckjdn8?&$w-En0klcAlMD;&4JvP| zh4x$C!o)>i$mu^fp{-6C*Vbpio9?Ul3q zd~A~|ketgj#pVgymw1b;_LRZ{62W?*5*d@UAGpY?yyvRTr?f6oLcZB( zX=$*b^qBi@ywkk^#@nPpd(Q`$Jjx5!q^64d{~hoeB7?LoR-i?9vKcqcF@2I5k0~f% zk1~4T;Dwuo!1~Q>$cdG-e!D>Cg*AixPiKgo7bd*jrNS(>d&AYY7ErC*SM(7l;L?vxI_e^2PZlqMLw%F^!38has&_ZAJylLR=QaWL*{vk4^D50SRiFn7I;pj8kuZ8% z9Jn?Ak!+hKu%NK9AoD#BM|)I}Ze2CZa@)>!l|F!bhi_3w4>j&`>l$0U{W3k-9FFBj z;%VsY2t4f43}?2VW~EB$Y>9CW9n&@dD-=}tHFF=nsIZ0FY|CRymln}#*A(V8^eCSY z_JGX@YGPw+kK(x9L!^7oC`dou%*IuOwnv_TnvFGg#V5 zBS;)Dj^9n%M73oB5L>a58{OPMpB&HS#BViQqLoHu}}EQ68`rKCo4{A!`J^(=+#Y)?2tz;Jr%16ibtB^j>!}9Z=4p% z=&!_Iv-JCNGkWcl3g8@D40=sZdCUK{{+5Zdvx!^zi+Bs zURn!gOuW5b2O3XApiIe7 z*;hxhrn4C^@29f#o!4;bOt~}ULzD$wGPaRa4K9YESxIdEg^e(v^(!6SmTq-%Kch3K zHa^MLW?#S9lX}r#wYsyK_KwqKRoS`3&BPouy@t>+p+nj8qzA&#SO;3S)dtL5$I#bn z=d%uv$x_=zhSDb|GO;Q>k)}^o$9Z)<_`P&7&p)k=Hhh^*oBvbdTkff0EvtvyA~GK2 z{jo7imwdfpMG9UdLQB+15`1tJkY@zly<1_1W+&O`))zbGr(@TpT;?&hnY|QeqRP)z zRCjj_Zn$M3br|l2t3`g*or(IOZe%ANs9KNxcLiZb(LZ6ll02&^{zzJa$H7az(a<>} ziw^CVgF@O5vMg4f?aHkuo<)i9%tl}4{OhyCqR^X^BwCZguG>PE-DYm*V}REJH$cbr z0FZQkg>Ml`IANRx*w>H7>xR*&VYvnsBi2FE97p$?OriL5hFybviz3Jvqk^Y%)sUKhg#0h(z&=6_F5W^YiAv=C ztMbs=TS2OFP95r0Pk_tXjiB72gSga1_O+o*GHLiHiSgdmjBQ&>K0EB9-wcb2S`N*? zm5aIHY8Xk3&K#4hxz|FPOx&6Cqoc$&RE?EfPh#Gq2JxWTw*;-|hcI!>5bRfWf~i(E zvHQ{KV&~RzaevpBhK9yS*#;Flv!Ys}z56`*^O}*T*Y*p`p3En@h9lwZYkBSwq6^#J zKV^>H*Kxge17=j+gr@iR$k!BS?h9Eoqv!#7cdP+yLkrm?lYAVyHvtWxo9tEAgcEN2 zF>9O!KC=zSQ-ih8a?%xS=}`lJe{SMK-~OckI-jtl-`lA8ILF}KrFd!aNjyAh5u2<2 zA8RsiVL!wieFi&BTg(UJhxY#LpIZdUKWhfN47))^N$gwEOCh~=T=AA&8(gu@0;@-H z7@oKv2U!ip*8MMK2fh7q#fJ_&muoD!W9Y&@7fprp!*p@t({{Mwrj6NC#`E~@JoubH zA16w+SxrD5pZ9W9$S&#I~wz_-;+pucX}Y21xMozqp1+yu-0nJx~pKG zxXx<0Z9EwnZy+6TLr!|xQsfr3`(f(Ud)R0Fa(uC%8T($ng2UhRmiAiL2S+s9V17Vv zKD~b^hQ@uvu>l>}I{&8-X}^{JdV`Xvnkx2jwJ($0>n$~nc}lgm^dQ%rHo>cphG=IO zA=^0R5~ya>(DqXu^iS$2+_8NCR>m~&{LgJXs{R$(<+={Q$_j2zc#UhWeZieS08exl zutpWi@b7W{GU|%xiL@ctqWi#DXI06SJ{he2PXf)*FT}-~h3K|guOwwsHXJ|okyiR0 zrmr$akcj$RG6L%96Zv%T^gRQ`<@4cq;xJLqwTFQrJHfJU3BEaLDeXC9m$Ympa@&4) zxJq6yjy)rTt-t#4*5n|3mAe3rSB1~KUOOXH8 z8z!vL!YvW0usAD=Oy8#q>%^|ax(k%_k&0ZREgewj+$EH&+oEF60@!u=4=g_Rndxmd z=7|sAv0EO|EXuntIkfc<&3@5?IUI@uv(*1!NNq2Yl>MGQ(;voKH!I@iccQ+cH=QlL zx|&@dK9OxYV#&wxm9(gGBQti*hndR11R-`4ee~3n`u9`^y_{E|{w)srJeI9d^VmH#^ zFt8ezO_j_1+1|_160{3up|@;EK<#$$ex``c`WKjUfIs(CQ|0q#jHP}uXO_Q7lZ{hQ z;yK@DvA^?DVO!xRvfAyhaO!TiFre53P6jTbUOt1+T&_weSJfi@SO0)h7vk6xQy*3r zuYu&&XOh77Ni;@Bu=g9XVL;6($Z=7`!nW`5epf2F+q#*IHq9eujl)Ty`~p~^5(F~` zD3hFmKkU-BOqgd(aEhBU+gv>v=Cm85Pe3qsZrmZcl4=6a$DU{NkC>9RYzh10=)u%x z-2$>#iQS&o#!`EIpjYRs@DQUkxDsc^EK+;(#S#NJbtY2wcGG_>;lz1z*<&#e`>`Fe z(gQ#-{isY>=)v4}<+F!dzsWpP%3++`Pts%FG6=jB$YM8D$8^C_WPenT%EYA0$}F+-J2lEYO{~w~Vf8gC{GtCa z{yH!QCw+CtjiV0Y>WWN};V+NpHpJ7-=M1^e_$nG0yTNYFG!Eea=>z#1>k~Y%rqtI*hO|o_BbSgJG z1TwDQ0lhgXY|Ocfpst((Tl3Da%4g4+>x+-DKyIsK*YhTDeqK#3E!M>KXHUz_m(~e# z*XNVl8~@yhtb{RidfXEG0fX&1Ke&q4FRK4sRmJmA8#{ANSg&elq!;# zbzGT0bWt2Lg8Xp{l62H2zzhWi=8$rpCh(y!TA`4as^!tdQ(wTpKA0a)7|DI^C9`M+ zCyDQ_IOaX)3K^-=LhL=B3y|=No;-bCk`p0==L4R=i}r<({%#AL9u)+tNmIee`KV;> zp6rS7W2`r3Shj_80 z8b{&Gw<%2JxD!`+vyUcyi;|>Va%B|;kLlcuDjaysR@Rg1;RlN+bWpCDP*9`_Gb#_U zvD+f~m`5%2r-Ly(PFgDK{98?aeZPr@p@vuUt~h6Z4KD? zdcYac%e+fuu7>CUzdlD_ira|Ti}40NB@HG4-P+6?l!&vFhTs(?fzvxIB{mCs5^(DY z^M`C>(b5<=dv-YWm}tS2`?;}yB1?O9uYMw{%?=HFE0@f3jNy*w>u^d!7`ySLP4IF) zNoRgG!!4~#=)z%U5}9ELKJ3v>{d_7UD+^ciza#GOqSK>!Wuh0F<*me)pw$xhFIlYj zKndiY+=@#{oWx>wA?rPF5`G)w!jzTNsp^zn!kI$}!l<5anQU)8^B-y~*s3mK7!fO& z^JKg{#*n=ISAlvt(Iju(HkjbH6Jo=kvD5|fqTl;7&5qAvI{l8Z>B;Yz{gDU!(UJ^) zeq97VW1N7VHwR&@qY|m#@(LE}R}yorPgEsfI#^7fg~_cliAugBocuY8F7~qp$&gr9 zICnTdtay>1T~bUl^cGRAwyRiBd=jtw?IX)$6uFPp6dn}5f$>?3xx~^%T6km`sJ1!E zE$kV#{n?)Q^3h)M zeAOdTc0Y*au5qM2Ty@#I0e+xR<^u6zr|ONJ2Ka3Bdr5ggJ5@EC$d>BNV&P*b4K3S; z4ss#5C{Mii&97u*h$-9|cnU6-cJW#HD4CIaQufy*ipZ|32|sVe%if-vYPCPzUAAM( zLRs}~RW{>Q7i&BcLDmjp@c!y=Rws*x2U+<9wMNlo4Oh7JXP0!~@IaAgFDLq9QYGDY zwlfo*El@SW7Ijr3*rAjG-1ujw=m)xpleeD2xt1*!sXtHb^&QSn*+y(^m6%oN9ZYFmfzMWa0K-Y{Z2W8r zO^LsuF?lZT?ieODn)@0LsRlrFqaikXSFtN+t=LW@W4?S%6jKbcBj*0IaKnxvID45L z9i&+X(~2&l-qiK@_+tW+89Q*%s%)!Xog+x~PHT9raa5Qo@!(QZEr32IIv6X(A zw}7wMGajttLRiHZRs3&LE!>}aop&j@am{#V7ILlv$hv;)Os%T))3Wh=%13i<(`e6Q zf;^}!tOg42Plvi`%UIesHyGzI4@+vVu|o<|_>vks=J2hRDepSO@9o>n@0>ml7sWn_ z=&H?Fm%SUav^Vn?bK7JcwG6G5oIyT*5-x3&@e^zF*zpoYuGA39=b5&$Q+_6_EGCE? z8+eSZ*OlWRC(p*h$~cazfG?YDKz`h60`F!0r9whEQ_?R5^}d94&1gYu{sq^ymEgx2 z>+tj7Wz60s5>oP}(V{3@dZ%$1j(zb2*U8)C19$*&cgDh%`D5XLlN!6N7KOHdyW#rO z0z5m*h!3q!gekQuymZiM92D>x+m3wX*Ouxt%@!Tdm6(z(dG@?vulTJU97xkXzk{zK zSu8wLhs#7Zv(%t~Kl*7^;-P+(IK~}h9XS_RPs14SbnSuV_4*>~tB7gMj1j#-AiQbT z!THnDaDJyEG^}@J?G;JzO6+0C(|v+F=M6;l*W+-LM;X4I8U}qG^&}sX0_s-s5W0nf3c{r@aL#CoQL=pBJJ1EftzQrI{>J)Wt0`W69l|LWq78NhdAVWA1C* zxZbAQm^aEo68moy_WFE>ywd8X$H`w<==YGMyD8y+pU+LwdeIvk5{}~T_8>M(zKkh8 zF=g738-)a~x3IopGo`at!5})3t~sSF(>yVoX)J3Xzc1AZRl}?3m5mbiuDO9FspgTX zfj6xd^{N&#>?eu5V^5-TThwWuI!s#hs=w8L4@|9I%uL0wFL6+{_?V=@-<&_1o+#pd3RWKXMAxTf4oB3BX!U4290ze_{#{ODOY!^Rd3@~`0M zDaSFvs4x0=4~1Nt!-7NULb~8~2#@pZ#SiAxz=(D`7$7n>-ijZ+po0+)h|#~k8j zxJuHzB|vCUFSKf{d=5tIH?qFDu|(;{P&_uE7+$WCmyGuEg|;70C6=wel8%)W`t(X- z6<+7r?=(BItw++N{W4YZ;eH6rter%y-zSpgTI2Y#%rPu&g)xow{*7)0C*f*ZI=l#p z!IoX;(KYcX4z7#Ctm*-n9B_T zLQ3mGY4?gB_~hvW>{T3&TU(;=*ZW=gW86;GemqXHt}}v8d!&GNzk*vNOL3CCA7#Q^Cq{Kh3C#i4WA{dm4_m>Nf#QbGGCBZTx>AX(-s<@Cg z?h9m0Sb#0hmg8F0yZr9)b3Ayo2KY~WN+tI@S?t>ZETHBx?7Hs&4>}J?7Oyja7YP#K z%hHvwymwq(?{qV-%Qqkdrw7@KJFV0E)AEQdlSrEHyLsJ&H=k8w2$&_$^{qvx@B9$GuXGb5otY9fq zC$rOzCW2=NieBko?1j8M5%$>e{&Rxpms~AAaNkz2*fw6GP*yCt(lrp4{p^CSv_@u? zufQ8_Pk?vXKPBeFV`%St2J})yC*(SHlZ8=d;nuq0_R+A4Go5VGb~8Y zE3K^7^(7RqTusjCFXYaVOY!HLMhJW}7B2+3kO`6Y;3D2U78vD|^AVZw`{5S&Q?EcH zPmN_}4u9bB_w&SP>IxEC)sJ7SX`<~W8RWgg0iwH2oW0I7Vf^AYGX9&qY@>0bB>Z<6 zlSb*2DKwqT#b@xS-&=TiYabbu>;N?(ZIZgOYtUR|iH%PWl2H5pVDD2xD^0Q_^X2Z) z)?m@!|7Q)!zcWUN)$C7BsoKka>^7C`=yK_a$bA{iwH|@{ z7M~@(y+h&C%1y%77;Wed)yD~z9kAYTIw9}tg>s=U$}KJ;Gp_o;yl=ng65T`OgtK<8EYkR z3hYN3lN!mjlHCw>k;09q_3T~A8M-<;f;<-cHgcbBg^40};*2HI95uY?+D^ZJ~WLQ=LepuW^tjffh zsuv53-|JaT^0lM8V-rcx*b_1rvon&N(RvcYIsE|MUB|jO6>#{kR?PV1uy^Te1^xd* zFz=xaMvQfrw1k@jd`-rphiegcpM(J00%H7S5bUTyczG@vdR7d@(gp9pe{wCk*kitx zqP91g>plcRw8wCR{y}tni9O5sbDP=Bjw8!V`huZW48GOSrg4fr*wqs!SW=Z1pOC8w zaiPA9Dk#;lAY3RVY+jr~S4u(AYx z1+4|=E9!_lf5WzxCnC`piC3Dx0=D zjTc#(EShY5D$jJ};@G{cDyvh^jbO+oU0BfGOn1Gqqb(<=5Us7fM9z>hzP%z&$pHkb z^Xu@&TX|`c!C?GPDF?6aoQ8_Nt?`mR6)BwOo9_0xD*lW~CV`?3rQ zH?4tpzqQcGSqnRZXD}b}zF?T8kA{Qy^17BQEd9$g>@_X~PB<8fJE6$BV_xvjBnsAe zOve{%YMB4DL9o>&j_*7;9Ex1ku&Ky^iOC*PVrxfk4VK3z!?sK+t_cOZr%AA8z&mOo z*$Y=Q>!5n~XjD9R4o>;K1jTp+3;vrZy;vn!oWDZ;b`J)X7guq2xe8OZSWS+foPw*K zIzqQ|p)=+VIsW22QJi%~ux=U$-M>v>&z^dCzs3<>YWuRTjtJs1 z_&?HSBqw(5IkV#B=cr@01G!>hNEZDOdBx^2j15&|39ED`jVd)}<3xR|I^rkux{ypt zipAVgwlSz|Kgx<$bXpzWkx#sGVtM6qYyQu7CzxgSlP=YX!IX~&IDLY|qDJfnUZ4W= zT$6?FYz=&L{2}>Z6ba9d#6tJPL)dkp3^r_z!~u7EU}xMOdg|(57~HH2g$*J2_EIbu%g(ODu60;0S#XI{x7(XtL z&FLD6uEWzo^=%sL{Fn-RezZfm$}T?Fq>IIT9U?N%ce4@ZanSHC1Uw&G!}Qr4+-f=K ze8?rQQt!c;KH{csaSq;IcOwfTJ0QM6M%Oy!L1f20XbY|(f2A34_kBK;d@&(cFOG!6 zK0Cli&4T4@J;t_gZDk&Giw!kVNLY=O!;G`{2w>fx&7fY|Ax`Yut{ zdCU-~y*b_$5Q=n8fQippNO+yX=l9p|IEMI5e<&GU zAIC~&IKWuPKxl}z!CB5`cy{D!OjmIezn4w)Wzs~j7IPqa1!{}=Y2aXb`5Wk3nZt*U zu_Soob%>w3oxHnT$e?37?U!_y?#N$HbbCH#VhwCc5x!fe!xpQVx^+ z)rj`?9&mD>G5#odir+(9F{;f(`p0ky8vBYjlXbsjvaoHk(*>4L6K9Eaw_H(aw>ta! zPAc9G>uCQQhcNNnKyX+!2%C}`$R@LOK>Njj zU4@;Iq2zPZFJb(LXdoJWaQ6`xTx<6P?G4eZDuyh|Sl@Qm0$Q zeU%Q_4b>*|f5>1~wj3_AcmvfB*JDDK21JOuR9R%BEH}AO2!4Bo``XEulsnzv!!F+^ zbZ?hX?D)@e^^&L9tH6c4|8IOr+d%_5$UvWUuGz_~gQiq-g(E8mqc>URfHD|Sg&fa9G-&V{X-#13<7x&1`V}1G1ZdXE#!Z?}Ko>&t)s^S*bEr zjluB#VG!J1HvlRuM-patQm9wWqxv@%GWvD8pxYMBS~i{~4>nw&pk~VB(u26;*N!6N zm1=y*K=E%|vzHr9^x-><6Zy2Mifr2mC(hePa=m{l;_S;3F4ro8ps*U;bq?TQy$iUp ztBqL5-xPHz3)~eW#kAP1tV_0sg~M99N^2-JwYZ2mQ8)Onou{$fwE}Z|&*9tiaa`+d zHvhf0RhGH&5LD)Bqtcu4P^6d&PLGwzwetsr*P3DXd{um{+Vn(4ybTMd%N*}28pq4Vg=Jb(C{pn!ghhN54Fru1~n2Am=P zPJB!JW<_i`3;r;LUYj;R+HfEm7sk9}bN8tbzoVjFXC5f2FIjpGgbiU$dLNj7sb;T9mAqaS3|+AEEb-MiI85fdx${;I7s0x$WR>Y*kc~ z5OYS{p=8_OQL`_u+w_Nh^qGOrMns~g)h?dB`30}LG=eV(4Pm!7=rD~JW4QgFF3MC? z`JKW6Y^ncC0Sy!JkJdmi{Oc&B7R+YZtNNg2ZZ)gTN@f-*cddf$l{^)VZ)vKZ%^K2$5XaZH_4w z=6G}R57`r`1tuJDh65o6?D*jQ?E8nloa{`1z$Fvt#2+qrxbIf>`^*>E`uYxAUpk&C zd3__zwwh$c7ft4KZYkJIO@jEx%5ePmbNa!~hWNPlU{1$>T|ZeL1mRor}90o$+&{6W78B(o_^h zE;VGr-E#-AS7sCGdYs8t4jO{?(MypC8 zz#BcA@u=fkJT*R#S#B^BX4#6jiw8p>!}Tw$*{+XIv}$2e^kj%CXocj(4J@3rvEZyx zTr%(@DHnG~!vJp*={*u()_J0LLo`~134E!XKVRoY)?D#|9KR(N9r`q!)$pPHkD_dA?a{vvto_1d}EF`PB0zySoUPrJ+gM&U1qYhA3rwD6Gz|9VdWE^vNy{^ zz^G;zey&o$B$Wf0F{lszx|EDo-KnUfFdAiN3{f?29FJ(X;ySJMMm44ri==*xH5=vUza^ zdzn?s%EfN9$mEZZmNF6D(+5gx&BJk)zC0*j2GK{Q0!IR0Na_yQpv#3PaN6{wn6q={ z9qGaRRE!z;OsSK2X=dP;lspW`ti<_cv*4M5E9@B6gB!U-k`KNwh1gqS528&2(f=(E zMSKL$4q5=)^v0ll&T;lAKa&5s*2I`rgKS;#AQ(6L5F7UT8yiwgnS>{Td#D=#z7yW% zguP%)yEQt>9EN8h4q;!|WkX;82VSuJzQ$ zw2zy}nq49gCFlqwsGfmqs>3m2MJd>|H<9_C=HPvk(yMaknC6~2Fz}TOYpB@SZ80fs8WaR7VY^QgTi5oLlq~rT7tTDgG7&w~Ki6!;&tX;c z?mR2@pu-nk?`z`s>jKD)RD!G1Mvwy|!^m&3|6uu|UMy^IEKEJ{m-g;eg29sjYso^W z?H>u229MYPWka&(qz-KCs-sTzj-ZsIPA>u!Ng!JdT%0)FWw-XE|U8v zW*EVyzB3`={66-f!xV+&7cg)0Ir!dRhaL8&a8SMyQwC|^RsB?$@CPwJQ4?GH&Bm}5 zbMVde-%x)k7JK=RmA=3E7T!GF%jyPca<#xxj2}5vDqG)E`aR|+?M;$lYI!@+DZM~W zr6l3Zl>tO0+7fHOHlxA#0r`JAjmJ5=ngW*=ky@%`E7x#UY#Oae3!9H zAw3}Hnge7kYyzdS2hi@Aj*F&vQQ`bL_WE`W9#~&ZwF1ki{3Vfh8ZnV(oPUhlW%AM} zHx;Sw<9;afc9E2>JPT7Dy5WfZ7SOzV3nJdw(>S?gXshmrJbpOrj1svx4*l69&!5D5 zZJX7VB|C|7VF(6}evA`VUO>xXrdZj1n{3a!3n6V{UV}o0#zsv{zn-4@X16@`fj> z+*SmA-q(2bnUMUjsf5h=-3{wgD1=lY2z9=JGHsj_5~y+XF!PtpcQ~xS)sERTvvlhqHd&gd={L_&f3;M%a%5-KcWd z)w`GY)~kTZBc`|{+6C}(GRz!XfP?O4Vvpmd(j!aqaQw+-;N0d+>enxZmdWFJ(B_f& zW%zNNcx^lz+Ov~%j*SDgxvxPZeYwOZ=Be1{qb^-gFrG2@``{4dfxAYAK&EC6Qyu#r zCMN|_7(1Be9BPIu%j<>8l^XCY>N5V@8IJ1F1b^R97GlIZlI3h42+m4m^X>Yx%*COw z_o?U+Yu9E^*L%P~7qPb|bO&o$x0{wo?!yU0Zmg%5R&^7pJmPx}4ySV!D(|qLYAEGbp1hg!=DeL>^y@cus@U6px+%!<*pY8RK zeVIvH{X^*qPb+roZyS4R-6$wLK7^Z=$D@I85hc?CNCtXZyWKM^fX&9)RIWu*H(Rc9hCUQy=H$NY2b82 z1#EiU8)4or;p8_>e&$3r?0h;#a6haK{kPl)c}XIX8fp_C>ri%}ZUx=*_#n&X!{MpZ zHG0%LlubJ|pXsQMp(pg>_>YI(Jf?am{=3>hlM8MF8PSI)+SJkDllt&i^l#Y>60;aD zzJg+qC&UkT#%2A*@7*}$Q(cXjFjSETxGpB+FEy}z%MQ`K4y%dn2>}jm)WOGYcI@E& zEHGVc%#sWzLcyU_<}kVo%-wrIblo2|KV=f!T=SW9uiuR?Pfo_ha!cms{em5QJreao z&H2o6mh8d(+hkv{6tm9xV_m}otiCamP4sr=a}_@Gd6^ox=-yc{n>-9>KEK9x)kTmI zyo1Lth9oyoj0C)v%N`yp@G4IZHU8=B`OVL@vx{T?@fkMPunGQT)pS~Lug z#|Pt>#__1SNDj{x&4e{OZn1&hPVnG+Dx^NO#kz-Y_|wLfe7L?Vn~)Pij9V|kKFyCf zalRkxBhz6Sy)HBJ`pT-K5@F}tTwY*5vcxz!h))}q04neHQR5qoow^&ummeu- zpMRO4o-CNT?~EfAJ>_Hgk*wc&Q55UjM8+)MO?;!KL)8QiD*Uzt?=!1# z{d+Hw5z~Zo%^$M|wH_?XeL=~=?)R+4DT6&-nt;}S2jNI}HL2dF9@2~niL@j64vDNf zhC6;Yv&65A-ECRUQuqjR;+zHBufLK#`#QR0)Q?2bgZ7ZwurN|l7|DL$=!CNe!|8!e z7yeN;fg9J95-;x-Qq*eCKHIJ0wJeJt|LMo)F4%_Ab-H*gNA%w)>?}UmwL&HF|l!8$UCn{3$14$WLZz8+A+!9SJQ z)g<3q==FmEn*Jvj+IMkb@;lmPsm zkbw<+8OaWJCw+iuvM#PDSt~PQ?$C- z<*c)U`)o68eb^hwmS9|bY(4K^y9q{mzGYqiw&KeNjW9cBDef(d1Z|zuFjYkzhs=r= z9@V^tsKXuTXQ+WET|BX`;dXGHq7E0E28r3vvpAz)PhziO#ex(O{JUr?<7Mqr$~%;7(i0e3>dXwDmI}U>#r`b zv!ji;_osLCX3j&{6{Uys8b839smj>7{sQiIkz<;hovC)8r&OFy*I zHaypu>2PT(}04tc>8pPBHRGdL>=^N{=u%OvM7K4I49=CkmN%lPEzQf3+@vQn~S zSUzkbZk#^_=bnsVRg<;YBInmE$~>P{^!MUXqGr|{SPk!gL~w8yy9X}|?0L&j{$kKQ zWC02|%%TMT4DFA7e#K#|yFX#CZj!_M3W?vjUR>ENo!sq@s9Wg-l2tX({nbF^e^f9R zhXJhUwHoxTxkl%G*5-yCUF;NFO%GZs!v&jopB(x;K&%aZl>BKjd zS@V;x|FecQdJJU0q6e@M*)c43?Ory*a3Fi>Y=RR)z3{W8m_G@Nz@anRAa8mpWYib2 z*@sNo*b;kw%yFY6$8$8?tuBH7qHg4k{UP^ZPgp;Gh)`#&OxDV}#2&;4fcFfDx9NVI zeeoi5`@WZ1955yS`Zhs=OBozk9gS7r)WP&t6xG>~D*Eq-k?P5r{Ozi1ymsL@w%4?r zHq7hKr<^>^+jY*<==%aU(A$Q+CF`Zy;|lP(Uk3Mjb%}qByw8k!i~6QbDl4&$q}N&( z@}pt$Tx|6q>P~sYC~y`8m3gu@HEr&9>^uEDm%*+V@)C=4vG9dw;%+o#)+Wz|*>i3& znmw3l|0a02+<~>%e4wu5Rzj`HGK}@hhYL&lvSnEzeB<^iJfc2H8ZajZC+d8{>;26z zYCtj8Op`Fjwn~X-K`uQvA{EUf&+zY@L#P^<&U!u$BblRhv8(C=vu<+W!!PCHe0N*0 zJcO87IEDuYXtV8emT^^;Xg)rB60dmogt?tl1G_~c<2+#n+^e*MbsBe3eQz&3R;vQT z{Fk#`-BY>mowKaeN1m@0v-suR(@49=Q{thM46E*>!vNyTj)@$IhO2jJ$m<)hbFLz$ zBoF29u0N15yU~1gte5Qh;z)Ya!xf%Sxki|d9xF${#p=ewb7_L>lCIx7*DcN-(SN@Lmdhf&HJv zlcwEQ4czpPtKND8wKY59 zFuWUDWBO2vawMq`e%9F5h5W)L$xwW4J$U#f(Hj}HXrXgC*jbiSYt;!LFTb0FZKxqz zl}hlQjzf4+mmW@X4#Ir{5|aBn9YuaH#vd1^k|kem^V|DQ!vfbnH0S3CY)$=I*MON7tN37s>K2DKTN;A`f%(*{Mdy}uk zu#T@?t6)TnCYDm!^ayP7OpF&RAHcKp5}@}(Iczw0$8PeT6|_fSo$MIZK;rW%Vb{VC zrd_O!+t$ayy_jLNzgFNe5`7fg^-$nT*3(}FyJ>m)sagl80fJZe!Zy_|I8*i>ZZeJ@ zzh%oiH^`%c32NxRhb>wgl*(PZd6uiZUx>$y3dPX_gJ8%sU$Cwpjc=aGApzkr=(M># z=>9!JO1}|Wc=jWG(%L5hE?#6pUGS{z9G++M52W9LYV5FiPo*P z!BNsn={A!Xe#7%3AU7*uu+c<3>6IyKo7%>&6<5<~H)QGi>`O4+>>Q39SBd}BmcUyP zhi3~8CjaLYuD?4)?yaqW(y~FdBW*lrbYpN7ck?!M)vv=4 z@`Uc6kuChL_T!dmckr|4p7?^!877e_BQ+1xC<$vOCIOK&v0*SSS{sl5l~>W{W{`x6jJb>=T3YX%)*I2So7?SVg+(@Fi;y>RK79BTYI1$!%w#l{UUK(c;3++Mr|f4`y3 zcHj{D`fn7pUoXSG>gVCj_It?CqKN9J`D3w6HC);pfi1&F0i8DujADL}(LaasBaQ}v z)<_N3;vG%58Qdb3q9!W4btV4Va|&Mzc?4lfGN?5)5Ii<1;sH_zdfBbql@t@UjvLQ5 zt&hg@hWV40S+Vpts{oWW5w~BOhV{xcLGgtJSTA3MV|1+WuOG|c&o&Eu^7=vAefBWg zkg<~;mDa)LQ5fe0bm4d7e?dowEe_GT22$QCXx}z}&h(fLH}760D#rPMH?%;qLlhp| zxECL)4aEBvbM3c8DW!Q9P)EL{!;*Y}Bcm?#tk%HOr;rLnYW_;$S1BdF@J18NUX<~p{x$HY5AQ>s4+!Ua;SsE>>VQYJo8WKiURe+*!XKYorLP%BS?t2k&B z9^~{C=IzyhTcsnwX5Irl`psIr^?W)uu#6{Ba|zr|y9`aj)bxU}A$y_!8-XWw?v4-?nIV$&l?+3_E73fF=yCkv63hXv`(m%;YB zhC()*QO-SGl#yphFZn-*{!eCXX^}SlWxtRV3q1|oCIu_MNs~Kf`mDSnoO)lf0Q-AM z#uY^Y-84}e#;i4^Q)SBdxGj@w-_OV-aoj61Z|XRF3x9_1U*q7QV=dwL z*um$62{gZa2pjk+XwQUq$?h;)_+sUO3QTiq;}`nS;$H%Tzh@|w&q$!kcTO>Z4>Qiz1hc9C;f!mp4ju5M5SY%4^RjwR86hE96N0ZmT;Ua5e(74bXBs;R4IJd25w|y^gsv~2_!)F$vr;kQeEC0Ae(hZ_n(w|}) zy0MvL&(vkf2lk_bJq}Bd$mZ=Fx z;wTw9KV&U`vNeFOTVhA5zWzZkOT$FzRx(I_d>UP=R{=A^$8(OS!ii|XKGykUFy>d3 zQsbo)$**i>sG435%I8MH=?(F8?aYKq&t&eqsu}oJ zOohH(E`;9^%ME(<4owZ10~(7Ok%iE64RE+d3hT0wSnif+URPMH`5Pso;eChPEN`~^ zfk=&f^k+V3X0_;Yz5+kYONyPaISz-%Fr+z1gxYpz(a6AmY_?>H_h$D69t z#_NM<+oS8mMpA_S#LCmHdMn$8WQuRIr1 z9Z6v=e`G&#eb7T(C4*por_h&dnTircqhV$NM%Sl4;I7zc^F8njqHX{dFH_>B+_%BR zd26BZO$>AMn!;*o4zO^uCsd(w7=5J|M~yWU@FB}ZprXAS#_Ikdzenriyz~oTKS&;_ z)b~MS>U+=%VASvKRQgHoF<-psF4<%3!$m0c zirgSMIdbRWWV|zEDAvY8Py5+PYExWDlP*Q0Rd3Bmrs^r$@wbU|+MJ-phaN%E-s`kJ zUFe;x8OA@l)C6NU+OUK>DX7@V4vpNHh0_#&peIc-sH$IzMMXO>XWT*aPsmVN?TLhf z9zAAofpk6dXK?|M}8`8U*Sx-!%M8BKdeq|m;S zQV3YI8;-jr!E5`G*zw4FxaWF;PBA|X*W2ZALHrBa)488^&6~uyC!gr^FEUKM%z_`@$$7c{Eo(hNewO-J^ zu9Wm^NwKZ9r|G=5WhB4wJ%oCm!73(Yc%q{YU~esIpy|YR8()Pn`8Qy=^($0xXEvLj z>_?AfT!ZNkF42A~;f#F$mxX=U%$}|WCK{IslfF36Z{wAzOMX0Z5ZGP&t0%)-Pa#{M zD@_|eSu-Q+Yxw0sVgB+m3Lngy2?{pTS)6|kTmI5gWOrgAx-xw|J!&(asxFpeGPMyz zy?!2rjd9GaBZE1fI>3DY#pBvSJ#adFl$Hv4?hSifp+7^GPq&O_IW6CSn9RkY2g1n& z!)TK9>Ic1Fo&c2&toDh(;`+0-8w&SOcm!66ll~~GI4ov^rzk<$lq?+d(F;HL{eqN^ zyF-F2L-EDo611h-5(ZN2@zve$pykq63_@nk>!1}9@A9E`a8h)p?f?<_4zf6T3c;nuL~O(FaFJDDoD|iGcwGwvz(^C6FL$e zWoSk65wt#XE_e(}6`k!(AP4qrC-*v6q9wu7@Zir2I{3IT`=q6ey7qV1JaTrXiwtJK z%X?*9n~pnDQojm6+;7sC&L^NtVF-*h&ZhT5hC|Z#e)49_d+L1r5Zt2P&_;eyk6+s; zJ*bM8UT|i;$84$gGN&sV-`7d<(S=x=WP} z^r>yJH@{!*E;T=Rk7{(6(^KWg;gD!O6;JYFnr)dVDd`ht19wVAHu zC_9gi#a)2>#ybAb9A5C0kD|G7kW)4pO*Q|G!O|XzxLNNvs#uUkGxqOChpRBHAuqt( zK^4cFrD2!V2ib^13wGnEEJSY~P45rA#2;AGO>nUZ{XXb930oD7c3<_tTAS2_vtMvk zdMpGt_an$ZkfX{u_Vh>p65jLWC8VU1X1ldf3xB*Q4>PaikcAJu!KmMYTogV-MR|9q zi&=|b#(#sMrgc$R6C+AP<`Mpue^k1qWy`E&LDD#(Zh&_x%X{);^PMUs}XzRbM6De;=S3si{nd zw_zr_59#!8aV+@OY1%MR8`nzQsL$OUbjv(7+7sc4Mqe)BCT!cyFA8qpcE7fwnRD0B ziX;q6Rvv<$)Kf$y?H}>mypq2Dm-5u-%s_fQRVd)g!ymeW@0*dXo%1{AGj3ZA3Iye*xQ4 zKa=jC`-5^`AJHb~(YX6XQq4sFy=dWyt7OKRrCgwYB00I>7YVbqWGnRtp+g3TaCYD* zcwMs;IgA-VzH>N~IC)>q-+^EBvDls+(5#`>?HlO*{e5)*oGkLsK%Gp!a*Vqw)<)sC z{E%whcs@pD0d2^wB^tT)eE0M=%C9&Bd~b_}Wz+;c88Se1bq>H!HxHQq;W*lCc9y2D zDk6`jw}H3d(g?BjL0jDaqF3E(@R!Cxc={(}uv=GwyNy<2t@BwV|Hfj05gbahYh-cN z*Q0cKm^WM52l$_Q8g2^kz&`V(;KM3SI`Tjolicyw{I z7y7kPggiA~k>%V_tW;x&!^7WD=a)n9hOf)%x=N0TKB=>(BShfhdmVLJ1)_@$O5lD+ z7Stb>vuk@l(oy4gu)RCXVd?pmta9)WYPI4y${C%3!Y<{JC#jR!4(Xk2=>9d#uI3JH z8}XV-=ANamwQM;R+|7S4v!W5)R-C$=uN6yGLc4_`J@hh%at8~+EPEflAoGm&$Rxu( zb9eA{l_m!lf2*1LV=TVie-_rO7NYN$QlTQ%311j74Oe{#78mN|i=S0Y6R*zvK~1)_ z&^-qp(7w0F{y*PS_~9b{X4l8fA9azlOZ&xXIu)ai>>7M2-vt{P4i)znUO;VSS~$Wj z9$l#Egn^R`Z(FvAJ`6s_BD^|auc{9cGdaB9{1<01rigr-wVA}exd0($-qc=G6`O4i zqM+4RL{8q=~X_kurmNAS^@IB{|f&ptH-qOGB%eDkkrOzFwcuAuyd;cR0{pT z2I2+VbceR82FKRSr{1gLscDJ>jA|Vq zVab(jvbq%*FDzyT4vXm)!P#o?Ya_jNH2};blW~P+E7yN&8JMg8;5#-N*WTHigMxnU zhqYlNA<{JszZi>T`%`Gk9WOp69#wyV}v1@RFOER60;<+)W`q9;L4Qyy*Mh$MwhDq`2Abyt% z`8yNo{1!#NI_5PDd!C3pjupYInm909;s(i;5wlDn z&Q}?vuk%*TzYZ60>sCW=>@FZz76MheTWMgnKCZcX2CT2$B)&ub@^|rJdd{$v^N7+E zl^iW1X$HH<$CAMy)hXouVjI{Z%RdvY7+Vt7zrr!&4Sw}*KhYMX>p9M^P<29x)eF2=e=fb2L@$ltX0ooB^hyA{f zWoO*-VQRw!x-BQ2Ja}~otNi)`3mwzw+dYjS`$Pk#U_%t2!O?BzPOL^%=*Fy!q2KOR zLsPO57_um?@Rla|xyqG}d*KadCdCr4GNKPBOeXFp574#2Z>hn;t5rYyFQ94ec5iN`0LaFbS+jo9^7B?kT6}Wy^B>pHCQ^{-P&z zx>jMIwRvQyz`EaC^^i4BOJ_SCO{H&-hgRRXFj(a8UI}L_5SAPK9Z$0QPFJ4J0UK>O zcJG%qoa3!wD6eN{`z((E zQrv7!j_eEVqN-B|h?YwoHJqG@KJS=Jk7pJTx!M2tg;#e`H?0uSFv%305R{F7F@LdL zf*kBgu)!wDZg_?*;eU5;BCEW|@RH*qoZGh)*SkFBLaiL&@6P8CEO0aK@4v~to-L*( z`q}v1<$p-~G=qKbR>L@9p3@MKO|yn85c$Xl^wmTc_EjZ_8BPg+F*(o4#G@8m@i}{D z5t~VSju_&3&(<(yrzZkGCXeYC&SX2spJ&Od)FobFLnXbgi)h2Tll0{a3%+w)8XbGK zh0iT2;3Iz7h^83-AU3N#1q)V z`VKqrXDxhljvi2IcI!;%siXReNhqErB_to?)_vkdu$@@HZZ0= z4N@#)?*t*U;m=DsuVkLrRx|Z;!|2gEQ;F%Pi;^LGW=QgGMIn118}$8b9>20oFk_z> z%$5w9#BRSWM;{XIkgs{YoJoxvTgMZ2?9x-ZKue9R`d-ENgjRDq$4Y%ruH$Qu5lsUndfw{<@#yKNy@Nzd zHA|7M9f+mNckHDa4{syKo6#skI}sFrD#7j=NAho7Cwb;8MfWYgOv8)~;cmM;KIN1P zwJS`JfEj@?0| zg)Q06UW%->dx`#LE3&EK2zltgh0uSAP}(pBO7>Jy{~_a9_cl3pJhb>}+0)0+_{J#I^Wh19=e~ z*U=3!w@BPx;CQo0EjjI=LCnZ?vUyldqzzMNWLOQ!l)KIt>q3J_Xeavqa;9Etw`E+v@v&4xFI9s8S+k>#y+W$nL ztcxeSHN_PXQm{AH7urkuK}*LQGA&jJ{K7tD8TA$%ZpA?IsQ~i9(F-XZsl}F8~KGtfD*1 zqT%T35L#%Xj(*0cz?Y;9w0dwU?y`FceW4TJ>1$W8SgOQyWOTXDJ4|7F${r|2dG5%HN@I@?ks;mCXbbnWK=I zs0fy&sAlJhG(6g2y5P2)PBkYX$dPPDTU#%{@oOUmu1qBxY~_S!zpX+qG@{s==p>kw z?@RpC-fYS%V_pWV+jr&d_Mf;U7dY58td1^T-iFru=9a#Z6$~Vxi zAF{MdI)e_19E%ih&H|+!TCmsS3>ka>GCUG^&&@#|Xy-{e2u-wu>UT0AId%|b(o3S= z4-5EDUkV}9RGYW4aez?KGM4aLN)oa_8FxLoiiS2CqPwnX=t098z-WEg{!-!u+StBL~O$St_OI8wMTzeS0aJ-KUJ$s3>^R+=aM2$+CFVMhvF^jq~5)4({A#$xEi8&p`DtF0AUj8`EhBimh zi+OJN-qdY)t63wPloG*Sf6c$3=d1W&5jlK%S&*y`lhSAZYxt}K6jm{h|_@azy^UjlKnwl||AJ+j6RWCqe*H*A?{056vj4{!( zq+g|Hu;tZ@QFX#xmK1lC9lSA^mF1oU9Zf6zX^R*uEmz=`)(pq5mG|Qh=mT1i8;U(n zl#v0|aX9u|9G^RKI9AEy$-zg*No4SO7`#xAy;8hOrwMb@0|N8%`TFPJ6tD-|t8d4d z^VR6-X=`cT4RzQPEqF+0r_=K`>0I}@}0B|0`D6x_^KL;R2`+?u=s^@J^_EA#fy zT&s`7&&7g0o@vZ>I#Qgwt&_i5c17gxrj1AMyDoB?`GPiH61q!;_e8l^1)W(wn~;RB zXm#&WJZJt7_?VgnA&C`~#?OGQ%QryF-hNaU>48rkyoVbGit+52!*KVP34Qa>ieFkN zjTf~&=fkJ3;M?Xgc;+07Hg~3y*5k3DChWlzWNxvN^%vpE+e5WGXY%x*hcsi6vXZU; zJ!ffdoun^00=L_Hhrc54^>KSv&P{nLoYmEY9?yn-htxdLRchy z7Im)vL-d}up+5DO$nuXmLxa`f*5M=YqRJ0su7_YXNiJ_~bOJ9MzJT5As3+4^c2l`l zS*EX;!H4U=#S-m zv}T+wzF~O_%P+hRQ+H~?(AyF|qHkK&+4y(F_OB~Hkh=#ujSkXtTe?}qzBs%iavhAa z|H2i0Od$4C9`K95nZX0|RG2$x3|@cc16uC&5^CBsAZlYL`tjJBtUPm&8?&jl)^~_P zZNi)kkuzBGvmXvYIft){6eTBX4?j6fuSqmHqdgPBBXRT^sK4|InF){P5YWr>(2Ac!yuGx9PpI4Y#NyW)DB;L>iH) zTg;3qqiJ?}3>Qfr)6XS8Nq6-a9JqWYWachFjiZF_QSU<`x4xA%&7OuL&YecHJEuUG zw0dp1;Dn33@rsX0*Rt!+S%^?VG&RgfqKnR^unL!Bq<#8#vhe*5YF;&lUHdFgyT|6i zoFGd!yf2tdi#ouf7mt?QUKGS~*ABy8qnF_u-mdIfdM{Tw@(j{$nn})!&(w~)SVKLt zPO_!ResI6x6l~k12Yv?I*wsnrXsUJ|eVI_i+?6h}AzsJPAg=>pdC!5ACxjubGYV`= zY#g`Fl@yDyR-LdSv;aIt26uWiQ5Sm@~;B6Jh!ArRRX$yVfn|bfiVIi;eH~k7N3F9P9 z=V#FGc@x=#k_|9lV2WGx9IcJIKMQsZ)+EO*XOr~;Z#Ua(2JMb&17GW9bjVX{*1q#K zADMoOUcvd=07zs8Ja_Se$Ize`!-;Y}>cOBAxHC*2`%(eru~HiTB(oQnb&dyVLnFrTFh>`+j=+AWN3dHev)ROj z>p<^HCLN}=knk?Y@qg}DvE!6Zcrw$Se%rntjSl+3%P;9>DguMXYyD`JulLKI(o3>upmk>&d@LNtTErjusU(XI{iQ>cWJhAf z{MDfIT8{k)GhG z%_pR>_v8UJ?bt;dylcq5zfr4mC4-RRudr__9~2&8T*(r-@x$XT5Nbej&N zKd;)N*(R;X>~$2o(eaAO%=t?FDr6+aQg@K%u?AEyS(~%^S4(b;x=+RkSy%O;j$}jS z4pf+$O~)om@xQ13C0m{w!P~P2RBw&1(EpV~-7EDW*lIQSOyTL-x`Wgt%9%}Ue+QPy z6X|ztz{Xxa@MzL(Q2jBFt_X}^zvddS+eKsX*SeMZ6anNfZ5&KKvQqfFS z-8>xBR$axNv*g5oP25PmcNXf{?ghQt#&rI!nKa2`00Kj1qWumrSZ-Y$4!pV&4A404 z(tjD~-Vurt7EpS{-3aGx6TV$qmb4<_0}b*?;(|`j!uDzc_kmZ$_r`Ca?N-Cct3C&m z(6SDq%{_2S?=0@ehKK0e@8RT{y1*1Vu?We?_i}1-n5`RrmS&G>Ag9tDp;+fO>1YqK z>x{D%RWC_J$#fr4Jf=mT{7R*&sl_O8^GMvj{W`XE?Zv-81<>z@qRE|6-6X8#9$L1; zizvNlW)psHQseeBs5#IdH+;NWA*UQ|z!Z z4$lf7Mb?Z7;3}#lSp8Zjo_C@d*C|fLwo&!y%i%8&QmBAGq#eN~eNDLbZXQ+?IApI@ zEhEJTFVG3D$57s||45K$FwDLlgZ@>h;m$aJJW|=7Q|g`0+}B6a=2_#(n*-ixoOcmC z+&Yp?Xg>&357^N@)=J3WEIQV}ld4=-1eZ_0_>8aTX#U1By7k%zBB^l0SJVAz{!bu{ z_4(-Fq9y2==>}TcCHTh-HL&UR6g*{)44Lx&A5?erU_1~Y_Nmf^jma0`sFodWQHkQ> zIXmVkctCm-g4pT;b5`UZf?o%8(BtF{#GcHezc#t^NA{`Ho#VyS$#@TS-}w+#E+3Dd zN9$qr_sLK?GzCxI)k@{puZ4K$U$ofI9XsK@km`8`9v#|8Eib<2CM{EkL+XaqM143+ zpBjtP&s@b0mV5B`W=rhQEKNsERv^V`J@lEL5`N&~i7&mZL2YAo@fF$YuvVu5#OC%G zMsRq<6)PNE@SQHtR$|lt2+q8PS@>ChEar(D?s!zf)}JtxY&??4&I;UY$+WB3snHk* z6;C7m_rK7#e`bumkw(Qv^SIA$-jFBT2S24EKw8I+ChZbp>n}91?I$VXJd}g%%dbJr z;0wsjo`j?4Cga|p%W=}rAUs!K_8oJdk1xDap~Y|4lII55Sj##M9J^()ukK8oXS^9| z!tT_Lm~eyGy%LL~_Nt1nW!Ylo-ao(|Dd3+sF)S#-pG zBFPHj=%eZCJRMO4`XiLN?TeIP((=KyORF^ri>GJdFY56->RSl*Te8< zHQ?jE*Wn(&8~!%LgMa=d+IMI@tPT3iduvksrmGz5yq5-tgbsfFXK&b3vH%`u_LG5D zedNtqEmH)8NGf zffZ=l4#Um~4m*cIXlqv~WX=})S9kkFJ`D{hY12y($!>vN9*e>$z;J_OCSpN$sWJBH@@AI1(g ze?Ye>1t~mOLbrh?Owta-zk2Oq^0W;Wr?=qIt2DF2b2Y56-jDq_>p?o~a`^-;Q!L?Z zSRpY9|HIiCO4DWIC(+Sy>5%jF9he))aZ`OFsLT^>vihnwnO$9rd(8e~tFdz8C)bon zk6#TRyebmpr(ETyPnkfEL>KX;cX~+H&aRquribZ&vMaIP#S{4Ntt{L#O_wj#_z%A+ z*^dugmxrHmf04!JZdBG~OO=OSf>=FmI%jl z61?NHJHGJ#7_2tuQFHn$RCz)dm+f%Gy-jQIieyg+du>5p6}mvNb~#b~aUJAWt-}uu zGr0V?4}5Fs1-#{YEIJtpI8;)GFFw@}AM`R4t54As&r=!1ZvW~g8>L*yvUU#$-akO* z+;ZWT&Q!-rss6xK>EK5O0T8J?h>jaO8QH(Ff~8C9kwNcbdTxgoZd!8{V})0^=UxFS z8S(+%8He)0rNdDB;x!<75KLRAbGT;taPg{*a^k_~&q4j>S$H*%vEpln*S^?~C+ypR z6`OLvJ7y#dG-%iSExA-X_)Ij{BYg$N?rr4??m253u`rAM)xxLOTs-q7F?ltTnt&XJu}qLmrD!vcR(6{%J43KRMimR~$vJY$LsdL#XwxL+QEGEhH%E7BTw zy_zMiE^ELYHW$H%Q8F@W9$Lmt#0PaWvE@8-Z241~rJ3xe)8~BR{~gF<%dhH-muuUK zySM>zpxKcV{7YzvN!sp<0h3Xg1c z!TydFHF;|W<*Sx}-1?hvtW`nqLtTN3n=Ngv!d;-WuLLxE^`NO-lkO@zM1PG8C!5xd z#%uO30Pl)8R2FZ6kN!@D@uhyqEHa6$8h#hG#NR~r>37k-NIx>C|0sW%DR4fiS)y&L z>p2Uv9KPX|n9hng2bJwR$Z6evGV*Bzx_0P2?D>&_9>vCzPH{GzzCHl@Hy_}&xPlDT zkjHZye?#Hj40<4EBz-4$fnv8v#1>y7GR=9YB>E{)`Y%JIVA4px)n(%`ovHZkl-+RD z@eZ@^Qe&^4yrf~_rNrRe5Y(ksg<~Al#UE{lVQnLQa$n&hj@p#VZH!yXz59|uU7zW( zpR+#$bhyyPB?(lwCK4=#j!tm=b$l!98eKT83_HF{#DgD{;qhLH@Z9$+eNgPf)feSJ z&EEh}RdC^0N*X;bdz}8vSAu{&(YUj9DL$zsgSPAu-fp2Y$hG;uNsV(4xVc(`BzgtX znE9Ptv0wcAFRp%FKKqNenII5K!B z4qcp$i-xA61x7|VZj~(!{xq1&-J}8k9p<`p|JNq=?)v`04+A1$<8~f!yAWHaLPs#s~tRxQ|~Wh9^aQhu;pnQH0TBzsd5?h zpE-m~{~I7N@7-X=zk^gty$R%PPJ;4kU6%hZoVGNKveP&&Frm+BaArqe)wW&nf(>;e z(Wpaac>TN;FlUc1I=QzMT`XD;CL6Zm4grbPez64n|Lx@UBDbMIx5J^QF^N1FYsXFr zbIks~d2nQp6GUAZi=X$c0S{ zm!NQoE)0n-#xo{e#hNGWfk~vfIXxO^Qs5zC@0wg|I_D+~6(R8mEb!$tOQGpQ3M8(y zV@t+9;<&M+=?-~D9cNvVTPdk^EUQqfd3C5Ec~35M}i44en4!N)JJ zxwJ!OATEqTxXGCxk`hyUI(VT?k?BzMMS7W?#itT*_qt8icSnP(hcq@$;gQv$8?~Fh zYZFkQSKe1qUT8vTDq3QU<%}3mdA;$Fc z*;qDXw-!;~FC&RGJjq7U92Qbzgrzq;hw~xtSYO&AoFmnb-cMM_2EFgI`(ojYPVuA3 zs>!xY+;|%`U8^Tk+$TzcA75p;FMQzbEq9s*C&+_sgRzW)i{!;RJEpmH2@^#gflsZC zkm${$?xH&C;c=II{?o>HeiKv2dsnI7ry`>4(a-ADHcQ^Nm`gevf??gKIDUrP8RRi- z5PEe+8!rBl)*0+);W~`xevRDN7R)9}Dixa%#LIj`>%Qz*Ez|5~mlN@j~4y5*Vb@XxNNm^yJYKMBBErIE_}iz$a@e=WkHIC?R6=IG16Rdh{#x?{ z`u_73oN+@ z5g4a0i`ADzp}&grS#rT}c6n|AdLeLHlG<*db&6Zqu=F_g_4{9%;GqMvW{ro5bEU}k zJ5mzksUo_}+k%BKHA(QMlk`EyeiF6oJfe#S!}sHv{E!A?l9jK*t~7en{F3o_g?K!h zKQ=APw^Tq;9cBJD=^ZAyY zDR5lirO^syZ1Yl&{<+ys)@+l3fSyiB?zbiTMkZpt_G}is(g)9S-$8D~Ek`p1=g9i> zqufthXYAXcB=nwd;!f}J;IhetJ0KcPf1L4RYrcD86ANi-_<25P3Q-QJ*n3oQ`+2Ho zgs9Y^<9tebF5bQdGq>N4k}{XSWZUs8G?3M?$Zhk$UB?&6+#H|*q>pd@qJ{c@%42Pf z6P&#MbP}CBiKd|eex3a@P@HIuHZDt}Vflaf&jHHh>b$3P!(amYm74J=*E=j2wo3L) zOhsA#m8|;wOv!(u5t6c1bu4w)3_Q%R2qilyvXH^UB+F&9BwKHPX8vuVtTAa1+gjYo zJfD81cCiyB?UtE*e5?{1alZ$)XUCJxuIK2(6FKaJsuG*i{+E8#7hJ3I`jWpPoa9O9 zKiUu|%_=|Vv;4!?*~X=Hta9-)`gN9_Na^K!=AnO-KAw0T&US8+d}~P-edY7$@6q3A zmL_3Kxfo*M{e)AWwJOuFjB#>tMLC*FM|$JDbxVa|CbyX7Q%6#tl6N^_Dda+d`xHf6>i z^N94QZKT@e3SDv`ovgQ+Lv~fYfl9?dCN=W_o2V+5OlSqRCSn3zHT^#Qqke=fSf|Uz z1^r}i%VJr8N-^7&q9h5Lvk8^di6jetTw$k=c92kkEj?>j8?!3KtT|-`JE1;{75#k5 zB$J0qzV=E<8LZ4r$J~-d@{9Iq9BPsTn}#H|B(?+#SqnPCb_Vrlg+%T zCZTCIEaLkIGU8@6pd`-991lc8g&S4sB_R&^R7a2XGmisi*ls&w?Qu5uzoGo7L z2a7H^qm6rgS!>@U_V&9clgl%M{*vAFPeuW3F&~CZ?=+KZK6dn6&jd-^gV)SE>jcsN zx{G(KGNirdJz;~<0%q&66Cdhcg+~|PWKie9KKA}%fx~N9eq#+CKdcf;9{-QB>yD@L z4f|$fhlWIxriAm{-|HNTl#miqDGkxk&`vwDLfIoLt0JSJ%yUrsC8a1LTC|kXK+)EF z-oM|!-skf<|DVru-_JF^;k>}ghBB|x&qA!;m%V*Bk7uqa6E1Eeur2NO`gmPtt3+2N%Yl>XFQ$-z8_e6zK zUBZc9mX<9WEhAuZ$AwR`=859Y+!j@wn_PrV_$^-Zt*@DX15)2hI7*A+qKV~yO(S&635zDP@$ zbab__I`T)k+xT(iirW{9T$g5vJg&=>?@zH28EdZ)vn#6_GrlP8@d&Y2Kv7gace1F7#Y3^af$xwNRA&9nzf@Ljcxi@0KW&<( zC0I5%kuQ*y6ex`!ZfYInD%Ng47hX15Bs#LrUbMsitgu~r1HHP*2{kW^b^YsNMYqa_ zip=h>7H*V^7I+jS2o1MS6OLIWCmLZ`EA*x`;q2271t(Wllv&wT3f;TIh0XR=!b8RF zWmc~Yg_`9z`PMTh3UVFO%d$>M2#ZZCO@EKuZrW84YASeiPxyXntZ?&^kwTLj&xQFe zgG7tnWJJ0n-wDnHgqeQUuoifnPvYk(Y!u$&ohF(ngqxbG#lFK8qN4(J(es&K%kms#gasdxg&lVah4v8>gzsgK zivDWEi0rUN*xz1TW+v9xeQFi&%257Y_IcA>k@ndS!r`XpgttRfMJ}h83nY8*3+=N+ zr4N{R&g|iZ!reh3Wot*Pnw{xCcemaoA8vvO5vRB=Xfi#sf-)=EsS}b zB%HN`Cn%g?D-69jmH$NSt$M#wS0KISW!c-n6GDO1PodYDa^cu@A*I9pQfRgc1R7&X zOcQszoAxi)6XjP&3#ZpN2_^2Ff_=gy)A;Qt%aTIY2rSfXh3^y&6z>f@%g?;jBn-|C z6nzl&@}FIrC2*a4ou4=B4gce!IfCSMRf5!Gc`#+mL}TVhh{Bi66gW*(6a47VE!$C5 zTQ)hvTqyT?qfoB0qReHSvRGGSDw61%BGT1~6>>=@OTUSIYtA-P`FfuuM3zn``BgM4DCl3AO`fI7qB~jZ#h%JZ!m^)j!hhGsh-@A|5q@_| z5{@0XQ|elIv~1_i7s4d_p(4HRN>Rhn_55dh77NPc9`eV`z9rnBc~hve!nSO@3oEov$S_bBnB6zj6w#16}R zX6awHu!WY7*o()R`1N%aGG9ic!XXb`Ig$|XEvBUffpoqnnX}7}WWP7Gv)1Zu+^nQB zo_gnTuGqr~6JF(@uIL@IM$M)Bvs`JL&3DReY+d;{DBmjLmKGVa$}6uq zQENZt*7;=Y)}`{4~d=bJ;Sa6Ad- zM$(1KAXw^_;Maycq#ygv*uRx{R2z%>!AxU#;MOW_shuoRgW1v-(pwQ3>Go0f*Wo5gBvgz zU?YC`VzJy~)>S(Os$c9dK|dZX1!=hQ} zye4s5_MGC5&Xr^zc2)9<`kpiAg^DCf&1SLBIJ^!s#KqO?$;LN~Hq1Fj?j{#`JgbLX z^&2gA@}Cv6%-@3fV+WIupCTF?&SG1<0=2xgM|i_ABnQP{^0)$=OT2=d&UBOw%4Jvk z&$Gs}yIGFD4Yz+@3_C8_%U&)Ujp*ekFn`S@n5$f)NZU>tX34d{SymEbq4c>oX z7V1bgFQ$`iNf&zhSHR)Ud`M(WLg0Hlx_ZZ&-iz;%&+)UQRFzDI)_!Ck^O_|sh-6D` z=koZ5;(4D-;&Cx87A;w?+4py{;PytbXA`fXuI&{HI`81+gA8Qb7vS@+^VqyK5=(Ut zB2dK>W&soU2hN!AH!m2@e{3p=)Kx#(bSWtyA%~QTi^QB&6aAj#3%V-8x?RPv;N8Hy znUArm@C}Zgui|`<$}o-Gf1JK<8vVIYPvKVw=;qxEq}o|bk0eDD(I7zsgIe%%aw4|h zapsb$d^rQle73*mJAd32DZ%pBg9P!_?$8j#VB1F_>{Lz2K4lAyH4CQfQ*kV_br>^E zEa2XKH-oHCJPy69#F1y0xp#$%to7JfF2iLsT`^Xng_G~#zF;)j{FEX6fi!%Ij$`i% zPO=95uiRL(b#yn@mGpdE$^LCH^vz}=D&HIGc8l=M&lP*bJCK!k#G_>iq|lK=RmZN=-OKHV5wJ+Ie*5oeil7UqJg-Gy5d+XD>a5viTpi;8HydGxu0PAlZ+$EqzF^ufj27 z9f-8&V<2fIW*oO+qnwt({G=|excUvj9_}>$-9x%L-~{dUJ7JL1jDy18uq$0j{Y8cJ z+;0_G$omstXEimLedc6d`mp?(PWHB>lzmU}MCzeoj<+v3x zgz6kdQLT?9t&$!`Ma?5=#e7XtRn|iFbVHo^_np0clLwn)S26px1eG|pGM^ux*<)`N ztP8D%_l1e1ChbNa4iAFGrWb75g(jxBc^4#1@1i5*2i(G%*~6w$II0?onqFnzrhBR! z*L#7@d0oOJ<`AAKp2MPhDcENyPY10mX&GMz!xP8iT2CZo_ZnfUL^CdjCy( z{eyAQ_yE`h8~1x{Giwr0?kHA zY#J+ox1>L=1^2L_gN%^7cs~YZI$?ST2i>`;_+osPyYu8PC$~+S%{)7b0?U2LFS3bv znf@@4>P3K!30(|6#p{~t$+jjsfZMtWYRCAn=Ic#{q_3Tw+9xZmv zq2ngY`!oy=4kzLB=q!G2zXzw=8qlpBi_q(q*x?{eR`>c){_O|+AK&K9=@<&Xj&5k8 zgVz_khMmz^gODHR*f}c?q_qT~?x_pS3*1a|V#8?Tk2u(lmZqX(hV(U#&;RK?j{o)T zZ@Q*Xhsim2(4)8zA))!4jQR|2ePJqZ-}KR_RrY{eXbM)(*hrokj&!nT5j`}srXk-I zY0=-Ss0<&0QT+#a<3~^DoxK~txX{s@%e@KA=Aa3Gpl=r6X|ECA`zNu72d1*tdB)tA znWJf?_+H=nU_+{Eku1CB5i=aGf>AMHoD?sHyY-j2suiuwOWzvFBf|0E;bN@b5WocQ zy*S;JTvD(}reU!sX%H8V!F91{y}TK#qq1pySt!lCv64Q0NM(-;0+@v2d^YmD9z`jN z6CIApUDAq)Lly3Yc@6yqnImaxY1CUFI*> zxp+O3bPZ-J)l}ICI0#D0mI-bQvJogOH6_=P$-od<)CZ_ir2Gg9ERZJ5QKh}@7qB#D zB&LKbKsCIb4Y@9X!HfEh@;(O`t%|y1=5(1ZRX_^gxY`F8ipf%!u%MXF(&Ec3b_9RxPo`z|=Kh8%jq%j9dn~fOnsp6vV&O_!2S&*i9_O=+wUPoogc>kAv4bJU4p+CN0jlIT6fE zNtRV;H6d2a+eVvx#yXd|Y;EHl=AZ4OTvVY;kpj{@?MpAJwCS@%BI@!!;`Y#E zSob@Cb&h&O56WIsv)OmbQR8qTwwM|8l(T2bxis;wBZ#( zhA3i?sRGWdQpPduaja#qJ~wBTD|RJ1W8~~V?C!eP__b{;jUTy~etL#tQ}-Mg^bW^; zg#>C{UQAD9i|M=7aK^dRaI-e@xV0;zX?c_njeRM1`=Al538VhAEG26ZJMApNOqoAhJS>K<^Dvdq+i;mb{OKrC zovlDJOG_~1$7E_2;j|1L(|(4Z1GUD~{UG9}G=h|~9< z;_$53u#{Vf?ORLGQt<@`k4?vny_Ib8+pjEnoH~^}|11War;)GXZrnXT6`KD>;oV_P z+7U3BK93nqq2BVeLF*=>t*@hT;UrwHw?KBv0!$8DgtaBtQ06a9w_jP)?8PB;^zLb@ z{-;bQKWfl|$sW}3cP_Y%^RcqW9C@ihkZ1^iR<8^l+9Dv2e0fUpyNpjyf8&_>Fp}$; zMz!U$*dQG(D7mdcRc#XbAFP74vn9I5t%cDML)b0)#v)ZU$z${=`dhC@J)efs_03}a z*&z#Rj|ruL69Sqgbq#NhpP;N|Mfh%}gi~Hk*ef!j=|=)cdg&u}VqhcwNFK-jd29HJ zDU12zYV`OXVj1w5^y%<>5{W5&g)sK8gV)RoCfiqoB@O>rYsF5s{m@QKy5oR-WQD@X zd+4A|I1Su5K?^Gk*wUIo*kUmbbEB+i)^>Y(_{E-PYzsqM)a4!rt9Uq{d5pBj0Tzue~qd~bZh&W?CYuKUQ@*34xc?`GiMnyJWLKNq7#o~U^u zLAi@M(BOC#C2ixGz|eykZ}wq-VxOYOL7BRARY>@xnakZ4&Z2hzVrD*tuv-5d8Po6L zj&?lflG?~3cADYozCwCXe}eqq7*YS57|hw`h21u4I3;-n$9&Jj({=^=4tlclrgAtb zorJUea5hV29~<{JknK_!MVG>+P-lh-S)Y-im$8T(f0}9Z+OcuRlUVSd%dj-8fuol)C2v|n zhb>%5bCw$kOzL?%CJ(SNz59{9UJWN5R3LJj4x40knxZk0l2)707O7YUsWcWYuA`ge z?y>oiCEU{xCw5l<1CJY5&n>vHNL>FH;nV7FJQ_BH);y1aZMQh)+a}@FsN3v@(qxwZ zJS>eJzjV%arMvj5X8P5yUnf)4;W{AJP6h z1i$^qfpa#Z(<(zL>gRAuS4gIdRVT=MVFWq;dVy;)cc8JT8T+gn+0kBYjODaYyuATc z`BoS&xXa#sz5tH{mvOkF6_3>1D5}bxL^-ikomWXmonO$oxN=h0SWN998g#Ly1yTnp z$Qe(_&P9%&@VFdJ?(tajeK%gky6|0MSMVo}HRG@N{uBq4Qjl!ni+m4rvidcV>>oD5 z^GY+ug=8RYK$cl7dd8kEvWLfSF@AAMhD@JLr|QM~Xt`rDTjM(+2`nkWelawi>1YF6B*yZjG2CG<3a|j!?rmBjkdM;uv(S! z+Xb}R!LKnxZs@mHINvd1d_HA8Pnz=5P5WdlSIrbRhkf9{pMP zi3MD2X1zI|*s+DPNOH9@btXyDESX9K)$WAnf|&@YIn2WKz!mg`b4j(O7Pw&V$1bP7-k-#d9Mv3(Gyegk{YYd-_{K>65P)gFOwg^- z#s=;v;fTEz#foEdb3{FY55K|K3qz^SWHRw4ROR=_XeO!x?VypWLw&aW_^*03^6u?syw$mEN3~(0wa&BcwZax^&=GI~UH0M@d~*+~3AnkYU7QqDA=>_;()# zXQ&}q>Lc^&s$;gh#*mb=GC62E!da!7Y&|bfv(yz73;39_ChtTK@ zEZI5~MFsBSbK8i5k>yC@E+I?h1lIl4f%?e`roPmbSIybW=HkH`g37{jmg9MYB^NcYwwVg3-rtNBr*31LY9U_zBJS@jGd5FW9UIBB zHF2CXm^X2e0&m2vwJfr}p52O&!|t3`$c!n%Vx?nv?X`hwHki@UzBzPl_&h2Il%t)l zB8-p9MV3??mgUaD$s@8jUGsrS9GHS-w;w_6ZXte~1;Ds#0rgF@A;%~;GI*{5bHAr7 z_j*5bvekiW-bpO~dmqNSQdl-)BKCY!!+XImI6RS|qb?s|`>qO!zm_8?TL+1Ye^R$q z7e!rYB7t!?7cfMHb?ZcPqbe`)zTXVtZMfpZyU?q_7m4u;lQUmPzD0-6Q&r`wi~iCz zKSTOj}7(>Si|G!<(M7*tl@Z3172~kY-a$;&X)*b%mZMGoIpN6c&8YOv#&xF~ zh~IYR+S9J0hJTh>`8k*znezaz&U)dn#4damNRp3cA>dHQGT2($lB+?lC5!R<%?3Jm z7<6EW1ofO~WkyB%SRu|w#-Rf=&0qw7^iFlY{{5kNnyZFOe}A)F=P&4fw1Zy1Os4}I zq-eX14Efu9L;Jf6^pU$rKW!J$Ke;|Oy`qcDe&l8n#gpgE23=!MH6KFo(HZUHnq{N; zMcCyIquAJDbe66X<1U5VrB~azwiSKcumjI9`$ZaEub;smkQ+%03}#ZS4NnC{2D&AV%IrYsn?`&^K3e2i&&Rr417yk+=td!tF0=QvvMIu}3U9ALRS7Upjg z;eEamSC?n;ssaXMb3-$9TR3W7GLx2mnnqUwlvrCT!;VQch|8XWm+Ey@96PI#l27Ndw+nvis8o_hE%>!|svXiPTW;$blM8Q)4dE9~8yD zxol&j&Sj(eUMOaq{>zTqm$5mb0A@F$or|dc$pXv|W9#-i`1UNH3z|0+|H`Wo7Fo?p zwpb(n#@1q&jktC?Fp>_h^`+lWPEu3^qlRCv@bB;j?xX7zwyywK6w-mMTYsS7=uhms zvyWYOwTF0_1MPVh4g2C{cwF|1xht*1mrvnHA9E4|!2+a*Er88gA+(ohlKJ-$wB;L* zL_I3_ooS4`m=g%wXGvyRs+3?8fZtMv6uNj5xuzecm1mc+^KqZJ>OeJ=__D#TqXuuv5B1dcEyD zIZr!5|IDVs=&dG7o8p*p=}sCRVo6qAlPJUS31&TM#KL}YPFUTAjSc>IKXwbG-Aj2+ zWB2ye?5z`N!OE02;U6Z>$qJo&_0?}xGdKde#rF%k(E zuJV=9!Ot7~hkvK;I&QD`M2PBh7PMyts`u_iZIT-@=aq0qkuqY;C!WdWiF?PLn^3^^ z;Msm9Zi2==lj8MWyaP{WlJ{35n*C}5ZN6#+(``$!-hC#fJKRNFObKGE9wF)XRj73- z;drhI+d6nKSvYG^f%!5D75C1Q=85r;gPuroI>5fpI?6o@_G3oZHeux{Q#=^Vpqr4$ z>)UdkTl=3X+iDjG$J^HEzU=_LHDSd3v&hLYi{?E)Oj+L!Qsz+ydMw6*pQ=uTcZ@F! zJ|RgX>^$hbb1K<1h9J=(mz_L&f_r>v3!XT(u$dbVGwZ%IGMxIHcG~xn+R!C5B{_mz z7o?G1+F@FMcN1~*X4Ckq73@x!KdWiZWY6`NP(k7`x||(F*QDNZuPjwrN?Ht?`Ta8Y zYojvn(EU9ow@>+VDc|mK5)*i=qq7IE#n{-S4jE$Jy=>umC*-wk#k1EYxZ{$I)Mx2P zSP+NVnrC3RD-3bDcX4>)LHZe7M|YHXWTenQi{2aay&YrVGOQ4X8g4`XYd9udUJAL% ziumqcgYR)~5N4u7u7UA15G3~Rgr1_t*tbZZcO7FFCSiJs8~ZozDVMLbfD5dh#`?Px z*r?K9?4i~<@^-&Q&2^6{YvLjrSjMB}x${VAu@f>ApTH|}5asIqhkvuY4qk)VQ5Ej%+DVCfb9 z%tI;#J$Ky@v7&%g{g{okWvR^QzC3cW+$q1xjC{w-QkDBBmg1^|pbw+*$m{@`oA09K z!C-oM*PLzdo55y2de0?jO3k&$LKYU#zbPUzvGb&f z7dBC9zZbs6!gZA6Ab09;??~m7{16x zO4}L?yHO0|bDtm>OBi|69_G#S5i=s6%bSqM8Lly5&T*@_VG$E}c4ke+?m5M{G~XB9 zn%20X8;L}>xj#2enbj64I&0VmIXPo04Dn`;zdTsEcPM*We~Il`+s}S|9tE9e zkD1S5Gfc3Lz^H&_2pYB-(g$KO?fGBa&$)@K$>~^}w2wArdr-ve6;z==0;WmJ*~G*& zc4?gySAF_7&vi{8*H&;DH3h$6@P0haGYTT#$_UyL5K7;N+fk$4Y^vu#zY>g91* zi))qbofZ^QYD=d-&7#4c!E9^mYHWV82ETlVplsX7|DEf^f?Kq3*Xs_8v}j#&+%3VGRL^voTB{=$jqcQipy;B|Ob=i=q^0V=h@KtIa8_cp%~khJcEqJkr>(;%y?4LOm6x!?#|&G?1=9Vra4X@ zo2-Ovi;4`}{5h60Ja?NpC7p+5krLGqLua$LxEEc9%y&ezdmmCkYzG!;Y16K$M=7l#fcx5!%U-K0z*#bk zJ^NFHfBAvrG*tyV?A4*iEyZlti!AM`GRpT|fzsWjobAONu44Ej7F4N=Grd7*PSB-| zvQn6Ga2$%gj==G<9yc*lniqK{noGJa!J2v#*{3oSOfUL@_4(6i>!T$!rp*TWCxWoo z`!W6w_*2>=cUrq-HEmkHgW20zLoKEqZ5b3AN9O&^F*#LO&Yu1 zMj%@`82yVE0OE7MR96D!*BQ*XUk3|h&qDS5bC_+_pwoYski^wiIJTTc@wzNrt50Wx z#u>oJK+NBUByzt`{l|XZQp6&O7Pi145ob=dqH&2iKE~}q(X=q!3SY@I|HiP9Q5=g8 zzreyT&SgrXAQgM^&Vr+V(>{C{lDtURcO zYEbLpAH2p51-zykjl9`IbkTA=07=F*u&OFz-+pE0iBzt?4BzpQ1ImqS=W+X2=x zppB2^Zt!ZbXPwnMxP@QxSkZMkb}V5WyWi5!^!*QV@%&-j+#)aL{dPS{8_Z#F*8ojg z(c~lbkna2*DxQ~L%RT)f%VbUjveyqcap_Z)*yN|_?EXe4+)(a9f*qe04AH>o;cIYA zEge~|FPNJYn1Nm>H*takr|vS7JNmHe5^pY4Xn6!kJC@RuD zqcBJmo}%n@cN&^1Pji<#Va6s;tbJ3)#(y17PT#6=ar+8LcfY~%^$#s1&eV-SK)M{Hy=~#gU_k{R)=Pa%+X~6XSk$9_a%}Nfn@YcTDh}h;< z+}$ltNB0LJP0+y{*Q{Ytv$F9m(E>WB7qOWp6_{yt2HRo}LsKq-=C7Vdf3!M~6Dz{& zK0_2vFu{WKJ8W>gAA8-A$%4+0ro0#CWct*K#*Y&+)h|a`_s=ElXVqd_Fh!60B}dS< z3$FBJ^KmjZJx}s`pR$p5K4@|h&vS5EiNbhq{P;QtYdw$Q;y@|(E&G8LCtk5^+bUMt zqrn;xLYQLY{m$Gw$*eYlbL#s`22q?Z0;YHy(SatrYm?B zU#9cEEi>Z{K#T2r=gjVKerWtVfp&Iprr8%GY3hBB8+ZN}7yP(^?e;#--UXC!--IXF zgFXpbGT)Ju(#lAqT>`=DZn5Nkf2Mpqj~jValbyS=3NtHPp{Hg*YnJ&_^4@36=4&2X z*_+Lc*O`SARl^bSVly_cD&R?5&0u{&IqXC3MAjrfhh^-##Xaol<<7>w;~KMNSZ%i~ zPL7#`_MBO)BsxP(de71H=G~3}B z0?V{NCbvkITfKM;`>Pa>Sp=U=N#C%wrXrPVAglBKvhfkv`zKn_DKy+&j+V zn{*L7^5;E!I8GJ14R$EEUjT`^QFy#Qm!}<<$LU;p!UEr%;IcRwRwUQUe(>#h)=6#L z8+B*=3s9phnaSi=_#K8Dl1+-~KORY2amu#$SfKh~yxw&YL&AGFSqU|^beJUU1~hoq z#!EQK;3V#rSV!?nI2eyuJ)2)Jl-Hg3n!~ME>|~yphpAnUg&j^f_a>8VOAKN9#m8A? zbqKrizL{Nfk;kRS5!_fmA66sA)jvOCybO!ItUN%hAG>#&32V#Qugzs_t7M^xe_IUq zwnv9uS~ZiSeTJM|dlIwUyq&3(S+L?-Kd6S?;bNlX*zb#FTy&c;Tddv5riA=pGNoI1 zSA!;SAAJ4T%<^1TmNW*14PtIWU6Inm%;~_Tb+jn9hKs8`!!no4#mTeZ+c~fYV^Zr_jS=sc3sEi?v6-aDqE~Sm!rBJ7-hJp8Oew1=~h5 zr3)KPJnSVYb?|rSJvfUS)p~GH9E3pWcg)Cy!}-VBxHnYXOTU~5iyKqfx#Xwp*U%AI zBhiDj`9k_0F3&$`Wx`kE8S_8&CPS^MpQTLt!<)W85e6F%W7Gm3d{@a+$J|y7nq^IW z?Y($kp+HwY%FrqNLRa)F*y$?J-ZVFwn^;P|P5DsSn7L*OgXmRK9~X*^42xO+(km3VbGW@m9uEahz%d zq;@LPwHF~Io#p@wgNJOhxdPUCredD7 z#lsWWRz)W^;bc6XMFbPyV+Xuur(?TCD~7xjYu$EPux$Alrfn)m5n5tCO;(rWpIzi+ z&Uf*~s1+LD*fo*lZuj8UrL&Ni_k`5M^Z4{xoAjr?VQbECNDtnJsp>mmw7?JBT^8f{ zlmooKyK{IC`#J8Z)o%8v>=DcPVSxE6UtyDX4e1`cp&N1za|VwkX6#Cr4_swBSB^7l z2{mYGoZ)G#DB(TGk~N8INa0kAmvSYe9e5AkF&0sIgGD{KhO-K)EWOu;Wv*<-+u$xt zkNkrAXDT$UvBPAmN`Y~b@UTf*+h6X<26tAOuoad6N#fU}UUph94O^-Pi~9v{%uKC> zjKUxG>WVBbsvl3;&)%*L>5H)MsD#n5aN#BOrmn1N;C*6BpIQpL)N9(T`#EHMN&pT$L)8H_yE__GY zrs1?BR*wd~Nnt}p)~sU2M`mGnRIC@1LX4+2#=P}M`qOLppsYsHcUR%(t|B~b{)eMd zuGG@ji0d!R&~^Vfg8b^R=dUrj*stUN`Rd7^oOg)7?`9h2kIu&oyXPoqNW%B!BKW*C zq-B=d+4S68w%$+x`30uzs`~_PqFOZXukTBysq>zNdB$S9!bObMVQ5>C$By@g(Do;L ziHlrDheE{pXoD#WZY$zCg<@@7%VBt)@5KCDBk{v@7rYjh!X_YwZa*7OW3P_E)s}x4 z92109pYOANufrI0W+(PEJ%Lft3b-A}#He2bVtwpa{_C4n{Bp$*e*ZlNr@b4Q8Ry6a z^!&hr&o9t3eGP{DE1+X_30o~S3Oz=fkbBz~50{qWlSvKteVaPd*wQZM(Pksbd;_)z zg@YveuyVdU6(t;{#ry=Em|u*xBtNXXlfsl9>|=SAHB3vz7oOR(5PK>VtDd;xl{+7s zdrC1}dAvB@t59I8KE2dmf)=TRD7rfc^C$Ijp{*tWFrvr_C*bEDOM+AKPgbxSyVnt&+7igP~)lW`fe#`NrGcQ(&(YU+%3O z|7D#rKV9_(zMo1!aIGz#z0pBfbt?AMxzMLuVWcr+9UU3Bhh}CdQDWl{yuOwUe{F`j zIh*M_~kFHB?jMP{H9!4!2j(wHh!s=GaedhEL)sa%HKfB?MyszOK1n{ee_0uEvb z*+regrG*Z-AFTxcl1XgA?>M%)?KvHu*iB6%2l2OW&0+a>POz=BrNsGHjW73eJYVDP z9R8JqC%KiCS2(Ss*33xU=l%ETG;`ANg2Ka6R_XSPlb&^nOU?X%_Lz9QnY0!^{#j6L z@LL%i;y&Q0xpgn#vgdD&0lAs$ghwvM$n^dHgftWZX)fYNwUh+FSv|> z<MB6m*!jqI~CP?BCc5t#7qR?fC*p)_`ZW-FUo2n;edR zq-QTH$ZAV64RRkvL5FkkDRd(y?7GTaWMt5&ArI?siP&;goL?Qq7>T|U^xUPe?u5AK z?It!wyPr!Hc5yGJO{C23RkV4$829W;p$ngjDNoFQ4p}8hN$JyQPtq7lNgPV65`H4L z_$3bYpMsk2YxH@WQDpr>>TDT9b8>zm@1`Q`oJKPV_kM2laYZ`rJe`gw*-^B7GVL{r zrL~Jrknx{TQf(@rov&}wr~P~?$N-IynM36_!szd}b7Z3)O)02gD?R?>qFki7wwf9o zZ>ogV=bvc#rh*oOKg|Bu9k#D*CY@U&M;WOnps1!r&Lm3-LwcaRRv8*q0jM2u3%6&6 z(#q2ZXyo7J^uqH2XK<6HF z&^BZa9&uRY7=rIpJK@&9l)O&tr1^D0B)oZ_6bg!I;=drOovp{;5hu^T8=XjzZbxXs z(~Gn@?KWu~^g&F(Rd!^xDqH@=opcq~(uLD2X`hNexZRp4l5A%I>W4@{tRc+iUgA;K zQgTTZP{brDdVORItah!1Ugs=4RJ~7MwoCGx>y-HW1+K{Ws|!y1AA41_k7jp8(8r-! zwC#B&7N`wCB6|QykF)5}xg>I<Cr%ulZg;Qz}I@kdS@Pv*&bbYZv#eH*k9 z&uy2YqxFMW+xr=bXYL`tuMS63_o2f4Cf=t#g;7NWr@c#$3!Kw#lJQ+YZF0Nl(%)mG z;T?pe1z(xQ`c~E}_F)8c6(Kn;AG0g;(P_O1fy?*8ai<$CdA)_ya!*lY_aAH%3}C^W z3)p3`5X#Zv>}zf-FRtq}H0y)mP*s5JYaZOfTft1r;yOFqqE4H~G@`I06#7=?H}9KHz@OaKtudVPE_h`uunyuEfh=NpKZ&@VW_q z<7$N52*LT;V$DLDD}9%Zr*)r?vFcV+B)8?@`jB(1e3d1F+XJv8$6Z|4xM9(oxrp7P zOE0HRp`))Isj~bVtLaW-7ZzP+r{^r9w&9;3Dc!@2^5(HE{KL$6Rw~P1ZHti2SMcD! zLG;qR664-yB3$1R)t2kAT15dR>jqe|d>!}l+-+{KzcNcc>?e+?ZdkFe3*Iv{Q8({6 z_RXk)Z^T7b($vZFnys;cA4U&StVphQ1oe-<$Oh$nW1fXJNZwosIV%mi>%Elx+(o>h zQ+?TspJs^oYf7Paqp>H}mo1MM>tdJ*t!_}DSEpa1RIIb8k2^^G=N%Yd>_qD{BE@>U zE*f7FO;aUPu*OG{U;WviHrN5JUcpO>h2Og{BW zrP0{MauoX68DCYNu#ex($Vk@&S0xg-@Oh4OSy+r$+f^aeSdQ;?4=~?64K6xav`YOs z&8cjnuL@jZUj zKXE6;TS&6$H^2=ozQjBHsS&sLd(gh!p_DL81;;BtFhesSSK|i?v&1pcasvij4Y1~; z6|6D?&^9`ohV);jsu@Wnce5OQFR!x%I~f+8S;2z7%CH!PXqGrzj5SSD!sMUl*z3=F zCj5hUj8pb!@kT5QqmDj35*n0a;}S7vy<-|#Hn`Glk$6H}uOqbw6;Q?QLyW&h54E>W zq4ecW-bR%xY!o#C`^6ZQ!6VGJutEG{UuxMioknflhpmz8XyeCtD!N%thJu$EGI0he zn?=zNOH-!TI|TnkQ7Dhg!nw$5*m|@h^@s$O6>p)vo@c1(&{6DZ>0_of&P+|J8(STh zu=}SrnZz$X&sOm6GuCi{RX9t7AK%4Rcxu7-d?1EIl(NJ?O*T60Hn+u!n9gcfHe~5T zL@ZlIO|d7)bXzRu9h^d@^J8e1%U51^=Nk4s@GAPtBk}vtR|MW4OUYAZsoX9MYm3_1 zjiXj{VYe5peD6Z07rN<7>`Rh*QATeI&QQE_9tB#KQc!gQ*KlbSPhoiiuk*b-f+m}y zKI=1^dM}6KJeN~Y%WovKZK0Ycw$yJrjm351L zuKh@@q0)Sx<~STZRR9aWdQ1;`Nk26P@l!R&^Pl_ApzY=(sdxQ#IBa^$E_eN4%idSB z-Y@k`w?vaYNKNII$@)XMP!pS9&SxD)_H^5!1`RtkVEtPY?jMw}y}p8_m&KC4Mm~K# zbdJ988kltb2hRLs6mRRnIW*C~6Sp_o!X?~__u<`gUQLG@H-1kho!lNu%XA#*_JN0( zFg6{#2hB#;Ja@#&HL$;YYvwz-8l6{a&?3=<<_R`}*3)JJ3Y{jfQ_-e7%Ku^S;W0Su z;0a9w7l=0jLfY*o&Pn}5{FRruAT<>Q1+E$v{A>E+|&)o_Fw?jD43Ays6)sbO%aw7tS#ZCy`{NM8^1n6fxGC-(XBnZcd<^ zPydIa^A5}L|Khl!p%g7iq>WTk&wbwKNungHqRbR!Br;2b%!ZbPv=bsCWX4yNQ3xp` zBNZ9h31yV=2#%7>9nj!OTVMu_V3Z zr+$SzBqE+KJx-$4As>Wgl-04XtXdc;|?EOzJniIIdZo}8(Fp6G;TIh z;b;vbcKD{tM~3&|HIFVJ{K^M-yp!c`e>9~lU3*F={+5@H44i@a#pdW)c9R^W4KyM- zh0Hgm(5z~)Z_s9iFbzL!QrRTFgME0Tg(gdOU%=nMyU129fEZIkqq!IUJ3AP82hUU0 zm?$dLx_+*@c~CS{`}4zOM><2)}6yb znV`xue zBZuRkzYdnARif-n9{zkSMyG-z>vyWN{-j=lm68njj3w-zG_Wr%Uq{uhE8!y70J}2KAV; z9C3*_vExBdbP+?wGTpi;WAYHK8L#U zn_@rMLh$XEEc_R;29xewr=RVk>3HpD-26HYC-aP{VP6wX{T_?0vT68tKa!QD=KSaL zQEccPEXX$o(FyfgknM2Ayj~@kTcFP>{dZtrwm*__mIhQ!=OM4{Ijr~zhJ8&FK0cZ! z30SBgxn#3LSXWR=4MV;|u=l3jEsx15J{_@7zhL?5JJh%{)+TIPq@=pX8p$mOSD{sO zHV*sQOIm(rv~+^OhpQ%WlS__1b&3A<4xn`K6 zcm%VL2h!KyrttQPLC&2R;a{DP&^ozH;w%$RIbNp(U!6I^+9$^lGQbgE?X#)-k6t)D zBM*kDwRk9_Av7QDN%e3-%OF{aLC@)Aych7B(VgdCjo^Y;$5|5Xh2^K_Vd~`uO5GVm zMxPxd4%y>uYR(T5KAbNW4otr!G%ix){L8rQ92qQ5|@_5#J;ow>cwe)#QDmfWJVDJSWrc%2V8{tEs5 z^sy}0Tez1ol@B&4ik^mC=v6%DfVXMl^WKqt93QgDm>q1>DYAOrmDu3)9+~<|tf+Pp zv!Y~qyjU04FAc%EWs8uJYK^bkeQ|k+6yvU>qSu9bd>ih@aRU;#>lGXRw=9Lntj%SE zy?78}B!-NQL_Z5+G& zj%J4@aN{dG9^=xD_ZYnt7F*Vn+I>|l?U79`aZ)IU9>VHvNqDB0g-^y?uySsi`G(kAOtU%+OOG15@}J1Py*eQ6)xKOBzSj;%&z*qV#ba3P zmW995UqQ!7o}Z5Siy9YM9<1{Yh0mv;;Kw?Q5;JX|{kkJtX*oUi{3;ZP=l_`6OTvW1 z?bLWX9lLgyVcw2X{AxbRYl0TAZkI0N`DGr~%K#W-jgfbIAk*g{>i#@Nx5@Vr7
    zJL>TF_bIe3e2$%G_h6~I4XbPR;`M7j!TrK{cooWuy^Im`;rc0-h_y`9yA50@vS0Qy z=i}+Y?@-E5qOLrfN+OohAe#XwcsC51rYA|>WFB{hPsNuM18Pu?qW+$D>A)6ybb0Rt zyS1`h-)6-#B(a=sHi>0}ba~fwS$39wL|?&x|9!ARPV*XiXs1ncn@-cUphfi1Wi0k+ znsSfhCD?qW6tg{|5HlzoBOV)~ub2mj6*B;zGOYRe?;v>W8-tFf;mAn2jat#OU|iIQ zp$;Xmt`&O`wPx(rT!5&u%`nQ>#H4$t=;Nb0fwF?gGEnqqs->~~knw!!*%qR#<0dV=ukI!e1wlAw(vMpn_# zaVqp&d$gcXdsEmEV@GPiN;u%6#M|B{A~akEQI7Ai{N-;%eR_m=U3+Qof5WA->c&g& zR6Qi+iK%!z%Zz`DXNLsqLi+Mn7AIqz=VIu~;gEj~< zPqKxkG9P-dc`})J4x!TT9<1bf3p!VZ;qzDn$-6ySlGq>SLa5Gd;o{eAg8kj!!i_(f zRNw6iIb_JeSG^U>u3W|~vtS&a`$TYDeNE8YZAW?qpGf_+yvW(l!zo;aQTa9`#qPoc z#Y*Y>ilfq%e}bjL8x7LvuoKR!|Fn5B>Zvf;+nO@EZlfQS&CtF2AN*t!u|1;+I?MC0 z&ea|Nx#m!*`%yZN?er$f1`lU;$MVKeIP+8m3CeC*sqz4klV3nKr50&Q7#KHbeBsC8P{HKZ zZ@j-c22u7c!n-aZ^l+p&qZ8Gg2i1dZV#_w!Y^jcr99td6`&$jTd&Yklj$od>!;TZr z`k*jc8*AJSQQAf=y1eE#mW2oi*jY_MHEKLmu^90SMEqif8HLD! z*Zzm3TzQ)$s69;>^tU@Noh{*^UB6J{W8tD@0-kM~Cd`;sWPKgy zXqBoahYsJ*Z%g!ToMjb-smDyHbjC=&BQuUIJ)L;z{lP-4GnZb&c^bC)M5;;RLAn|z>GBQ)3` zbpd~S;lm9_cOqQ)Ma!!!$zQ#K2ez)}ue)3z@B7VWc-~Z-DUTLOa{WNthlZg#L7s2^ z^23ZBa;!Ry_{|q*967Fyfot1nb<1JN%+vs3f7T{id*lf}b(|%=^1e`-TREHey&T5! zTPI@du&MN7iv|6vKSU{MF$laWeorHlI=#1}`0hc-HYDOeeW^rUUROA`&sKOlZaS*7 z#Jqa^L>{j&fVPC4qFO&!_H)mKS9Ur+nSWI`cGjch0lX-en<#gh=T9bSL`H$dB|k zHPX80PZ1z#ej zT#zZURa?1o>=+&)zF!V%s&nzpN{pF$T~avyg>ZMtb<(r@DfmqH5bUS^7OD%6lj^fU zaHuQAwClfxwzgbhyhRFJZoU^L+H?xJ(n3La7LDUxmn9dgoFvO5MkCrb1zjs%VUMmq zua%k2G}497MTyFsnSEFyp5-r>s0a=v&BB~cqI!3ITB9(Gykq83<-(sBop1x256;C` zbq($ocoH*Kd!hfbIl{Qt9+FXhJ*Y%m8)2ST5#uW1#qR5Px#~FndO({k(nnyef#@Px zI}z(0U*PF?b9O(L$NyQ#(9w?{>F1dWk}khS&xcyW$tza8X5y$WGX@*$KOpu`S85De zDKXv|Be|@k$|Jh^qx56DV0-@+srkL8EtYjOdc|@<^P87Y{ADoh>E9t}^zM|5Txu-H zd~2XLM_&u&7yAlIuTR<(7d;Wiiuoz`ES5t?~2%R6KXtbN0XvjXh@bW20XB) zsao!oBqn(lT)S?wXGnzLweu5MMz%{9RaFRa$3BwMgf@hZTY=6;{c+Ax0So%MVY}NV z#EATH;p1~q@Kj>`wW@ry>JbL67tcGlP6?M^9v3R=dQ#v0YstU&9`ZWS4GIOk|GQn$E<8;U;>a=QyhSp1|Et9P_-ku{chjHg=Feel%tVVjHo=#{JvjkAh~t}DWg z7P0Q}TST5`>%<=TVGL6ng0=QyuTJSLH#|Mh&pueuSwCNz?G{8q$s=g;*et<*_-h+4 zb3eA&=*KhS#<8>LChDD%3YpquBzhXNcjy;f@y^8#Y@nHSiS+60d$O6Rz+GE=^W_vp z4ndX8+=<_9!ft(${LD!g{&f37Piwcs@vjr^#_8d2*RF_IFLH~1FR?7L0%pC|QpLb1 zTI1qP-S)d7=X)X!8Al>|YlEi zcNB=M$7L;c%s)agviqQCC(G{|u2box!+1PToF^HmO7~_MNM)8-NzF4yLkw0zX1E&4 zhd;%(?oM!a*(3bi@mJ{Qy-e^qHk1~|>PrXiFqPgOp&~ssWC>Gp0_**WV(X7-Jj8mF z=wA4Woi@3=Q)0vKrt85xRdlfB$@223MXaY&L9)J)g5OYoK^kktQ;j}Dv+^u5`@~>V z-8R}BcS0iDSc2mz0T{6-fxNywgwSOi;$CRehest)GF0TSX?HQDYk+i3#W3lY;s@Md z{FcJH6`{;WtR-*Ghv_LDT*;n8KhLLQ?usm=Z5xK7`yreVb%6go+0DUD`I4O#q;>$o4jncru4j_6~`LS;*ZDHb4L(yZ1Zn?5HtK;4X+3lcX|k4+`kAXD(^@a zn02xJVrXjnrPYapk8fe4T~XXKL!dJ%^T^)5lKK=4<%S${Ht(y$>HgUQI*tmRMSTRL zad{M&m?Lq}K47i2cPD-AHUifciSx$52qcLN2FXu&GvFe_0m%8E>-6xfBT9#7BiOXrNvq@ui)$E z#A7T%`1rre?EBb+cm1{DH}kvls0SVN(yt#}!w=KRzoP#tx|a8)YDt4reK;s@J?Bcr z*@~(>M|K;=8?&eIf_c&8{X^_Usdhs}Z7Hk7eP*RVIq9p5lg0CVCk5;Xr;fpMdEOUW z?D7wjq2J4^^`g+_IFB8vkwLTQc`K@NWxGU6jm+#Q9gG z;eXVUoru+9WpHq@yoFcXSjUgfZs5oVQQU4C!d(-FvbNqtw7*=2 z`{BLNTvacGEB_M~2WL~olT#?ZDRPlU2VoocTX3>pEF@iamna=l7ZgPp|UGMK; z*SH3#AD7`MYdg4Vc*A1fIoJ-56kS@;C{8Sfw3`9<`Kriy^1m=@W+Fap?Tr$*3gOw; z8tBh6L2oC~y>K{()oluS+{+7m@Z2W0rz1Rb?;ExlbODm4PRMxo=lbhwF(p=(6WnKr z+56$F(Hg_*ZUx+-I}O=Esu)!@irT(i6~@%3+MM@$Y%||x23qfgLE17IB`4xJ^kfe2 ztlP=KzlQSMl>@nNpf0C<>I(N&`Q*NI7sU=$<+DB9xh82VKl!juoa4P<&)(`%y)F~#(>~%XW$s3rU!=rm>~e5ERTr;{ z=3_uoKe{fvTVn1bj~>_NQpuJWVZL2Y{_U{?c1kCxZMiWYvEGmBfV0?35?TZEY9 zO@j8iQZ{|@hSfg&W>5V>TrSi=vUj`i;*bW9t}{ksz!AZEybPAk>OslJEy-e`C12Sy zkQ?iB^H+1LReQZ+l1t)mW z? zCM+eFliTrlWiE`?&qH3sdBN&+v&3{vFQL!jwTLcnqi?l|g3gZso?8~pF5)wzy;6Y_ zuL^8rv5w=0=^*aa6FOeekCKw6VPI+rIh)r=REM--_tyPzRdz(rf7>y%=ma(OG@w;6 zrlOy&TykRKIZ5^3JfYewUveke#HPDJS5z#0OYtTpbXn~nU3U8*#EnlD48uQCw(Tfb z`i?;G>$~XRPvrc!`Jk_LE1sm^gR0nXi?uMou`L&1JZCufm@2Z6S8Z5vWpDPI@t<(S zsZ0oYY(7oG&(55LmhWYh4IR(BM^54Us(x%Yed7O}sic3?1pe#a z;mYU=kn342H^`&KMt!g-^(%#&^x#P6R%od0M)&3NsO!k2^2_fi+Uzl@ewc$`$CXG} zQ^iNBtNGrKN;WUi;j$P-w%n$|hA+~jdacn?3;B)GJL8l%c(9oJy4!;0LmxO8G;4 ze!B;gidI9=|3`1xKtJK5{E&&lU48f!~qXLxKJ5EYOi`4X<8-B3O^m6|qg zp*tsA$vLBrI*q1ty!Ak~_V2`#A7(rp9Evp;71Y$T1IT z@xytMr9Ph@)&gJABh=t}f?{4}kkh-;W*n55H z&Xk_g`A^%0&%s@&6K{m%F)Jm%_Fbk;@p_ov9*M=(QAm(k4fWzAY!~_5`5kAV<5njO zeJ)3*wz8nEV~+k2>+x;$LS#)(#1-Rxa7gOFw9N~#(tRPeuAhNN$>*?nWdTnANJ5@h z4lRuOBDz3Bsaf+pR`~8lb;kj0Xw4Qm)f{~4Jd8s9$#nWa8VyO?N_slidGD6{JZh`B zxlpVJZwHM*d{P4TACbp-OBr~Xt3&VC57e})@S9H-{IC3|FeSs093#v~8Y{lb?=R!l z2q*6LX$_gmD59&?n?(=e}nTTh$Y9&5gdX4NgG%)OCH#n{}LFT3eOkG-y zmJdg0ciwv%IA0cHL{2TO|TnFWt>3E=pX@oc%d9Zks(k$<8b`_rfK!oTB0AHp!6w{!^BC=WtV0U-N% z3k_*uwvgrAsS__K;jiyXIl?&|A2-MwUOQ^uv+v z&S+dY0ak}{k-q$-$lDcT-Mv1@nA-~%^UN^)X946#{ldZ7W-Mzp75%Ne&~i!!;fw6C z>qZ8CZ@dHNk4Aj>sWms7*>lz&v1ZU7gZ&;0VO%P@^?S|6)WhODeePvAb+n_w?IS$z z%!6(65Ilc55VN-~V7JIkY-g0u8&j*vTT6Uq*c(AUqZqSxEJx4#QgjhNGc}**aHuN6 z+wnK3t)eU5j<&b^?&G*Id@g28$w%(r zitpX2iNxR39DcwOEXmWikF zF?TD9ie6E%!#%9ol!}V#=6)(?EtN>5;I+vZ|H_hK6UzO+kSV~u)Q>0-0^Aq4sD~Z z;ltADuo&2jqn_yV;_=3ux@rtkO1j{8{zy9a&jr~-gRsv&1M!{m6tYcUGI7DXVb|y0 z!tBv4xK#fOo67x=)6^A?-KJLB4WV(r{G5FeY$3Skq(y0BRbWRouBJ)^SFVWl)ai#PM;R$`j-lK1`Xyl1*+U$ z{};yZSFzgBICh`EiN7S7F?xjalBk`$?R*$o=1qrsk`+RqF}FlG@VmPc_`-&~mGdom1fW~qSo*5>xm~wNtc2NXx-MEU6i+%ff zwat)M7TsxABz$6~8lT!x30L2Cs&nfJgCXNdcB?<;bo$ZRav4chVBPZtaL{{I@4!%RjVbS?X$X;wJ zjr5-&jq~%8UU*PMf8NVN`^+!enSG3RtlGyHUS;sQ=ewYuyc@2Y9pL5@&DAqKd67e3 zF3h`4^44Rh%j|6Hl!8FMZd$@$FSl{`(owuPZv_9GXTi#!E6`qgZR~EdH@d%owDehugfn@Rx3fH;XQfyV{tP zqRR6hHDR9lRoL|H$8P_xi7kxyv+5G8s_KvC={xA!)>?XAKZZU?lm%@UC7NOUP;gJ1 zLsJr-NM?0AVO_tomvvQS8=iF})&iCU*syBs zB<|rE!Ph%KBjTy(uzujejo;ldO009KZY5^D3dd(<9TcrKLdG9wzTCWt;|xx*u8uLU z8nb|pF5b$$y#Ao`Uk7c;p#lMv+|ii$(l&|0&LZ+e|(N&O?9vUj&|=owM}e74+Bw@_gG`eL-%&#s@XLB{-c8h)5My>x;qER z_23DuKe24saP$>(g8A`_Q4+KX2S$BDOQ3>e-?=Evk?p_|BCJ5F`i|Kx$F;osd8ffS(R|G)BuPnXGo%0%L+T>3- z?dQx|xkfa8(>?mqoQ)3#E9k*hHF~M1E1ayCkPMCqi(dZ~R#@I7{dHF;u6Q_&_%WCC zlh@N`pAnExTtH8KF7F^zE!AolkEQKK&TV4<#v_qmafig9k`Hq^3(6Q({Rg={452 z_h6NV9(*(7BgGCrgSb4AQ(fvNp>J6d%YeD0YI}gH?&j0UHNWXh%sL@7?Un9=W}f=}48BjQllbvPVc<$N zdOmHZ&?>r3injcqE@CdbbId1Xt{BK0vj5Wgnxm8_+l_)O&QRF5D$?$!g2=w>gatF> zC4FU|3u~SYqcuSbg(G)|3sybmQ{W5+29!F|-&YCb zZ8?#j%@`nlb>Xp5J&?M87u`vpOd5T!jPV<*C-R}5+WxE zLr+P@KGBv;2{ppS)yGKjx1TUCtS4?O_NVs-?}Wcb*Tk7r0lzsM!pRe?DREMba7T5! zF#Nv*F!Wght*i_xRTF4g`69vE`=5~RaT6*>b+EAOK)8whIj>Qs$j{wJacSx(H0=)g z71zjE%$?cp4ulYA3AGj@+dW|#w(aZWY_&@TD$3EPAnR?y@3yblesci=l5diprYkbP zh9YU@LPRVmgWsORu+S_(K0-g`B1m&j;_*9uq}UjU&gQF>@Hqo153(U6));3E zW%%>f8<;nv0EI(8ia^6bIvgQ7@hs1>^Wj8Z?z@ni$d5<5*s=W$Wgc|aj#a*xvvS!_ z41SbENBfFyzO+#^sc9_RJT&F&t}RI2`;C@9o82{0P z?UJt~=7z|j;}#Bu79efmIefGmhNW(UvARI4AI+!W<;hp{>--I}l^+D-ls?4XZL^7VNPH{Cgr2~@ ztSQ);s*Ew)yJLZmsdzrxg89)k)a&g5$s%#)xlp3W3+n1cFVr`jt5S!R-Ey+XuM+O^O;^^WP|W)#v_&=Gow$Rb?KFO#C*BB0Kt!#+Ao(bm%VD=y#MOw}~FQ&P1A!d65EK zd*IWTdEBd6lkY416nCwv;E&urnm^z^^}l0+*5Ck?8Gpl_KIY=Cu>Hcm@HXLnZajaG zEaj~}i&(iso;jccG9|Au(z6M7mtKX#y<+@%I|%dAf~n%me3H#jx1HIfXnXM1Z)v&P zAWH8aM8QsWIHz-*yq?adTA4nity2hXkD<6dEpZ9+-eUqPpS3mv_j#>WDKc$joNJ4luJ@yah)x1|l?A%n2=_ZZZQ&IHB$ zi)5R;OK6{=O%v9h;X1uzoaA$atCjj7Ts+5x-)N<#DUEn)IEvSW#j&hi9{sv#N^Mr| zG-sOw*H4_r!z%o_W2<-;?-PQN*+mHN?tq|^);N*v4w|@v%64g!^)?e)(OsSA&n!Yl zL>BV1PVm!hV_ELy8>n9L=In&7ywz+!JiND1d210GvPbdZH(!L2ukMobpJeDNT*i%R zEtb!7Wh=L1GRy5~W;Gtxu%2H_Q^>e=BA!QPp!uQ5UuCOb_O z0Y$0;tu&=aaZhP_m;E#W33PUUtq>3pi=q?9P{0)!6?+rM4{Sz$Z6Z=c&bL(TtIrCv z;E>4mJS{t(*JQ+sxl%9MEB;q5v zWZZk$NH4-?U}gG8G^&4)lve2|UR{NXTuIuacFAk99Dy*;;bwCKO>p-?9DPBf)Fe^(0Ir3~eS7MkW? zPpzLbs7m*rVAn8_*b`6xJg4v_htV9oa5mpL>Wpji`l4X@TJk$XocU+8 z=+OygBah93;lz=2>hny}_0}fec@~1)Z&5dO#qz&tG_d{a>D}dY31A z*|4vk5r^k!u~Pj^3X}aut1r6Z`7pqvO~7UIk>YvF*;dZW!nSCEhV8(lWmx#g0_Hm} zl8Kmom=yXPKQDg6kprFxEa)ZXx7(>R(v%{eTqLDC8hB;YeOS*teVc9RH*9?F{(`ma zVNC0_8@mn&Pz(-1YEBZ48RtXYH3o7&qcCf(0)L!$0bfSABGt$hRtjPsWu7OHasXvY zqL2Ub7PRO`;a7{k2(kX5+{3c`>DqUk>nMV&(gj*EHH8$Xj3YDA4Ktu^ERNZZ6#a(N z`CP1&yYNtsu?uIpl}>D;NUYoO9VZu^gsYtBtm$!%7roieGaoHrou|izE`5GUp0BEu ztlqjnvOM~%u;SfP+S2_Y2OlZtrn6_cMdTkQa0K68GDBpKMV4~16UTb%bH;OV|Bw7u zbYzOIgel>?MrH599ki|BD7oKX z&bNAbbBt+kcJZshIR8_~aY)AXE6JF$No4W+?8Jd5-EerA0y<>$#2tlK`M}sS9D1;b z{r2lgbJuj29?kj5=Y2&^Cx1Q9TtA!dotYu%MSqlJjM1aY`yV6+n%&65!vVj<8Q%{5 zM|fgs!;(J9l6_l10Uf>Zti1p?n@+=Z>H)D1h!K2K4htqLwCSd2Z~mvajFl=fc~oL7 z-B_Os*E3+#(M$QshAliHC68k#T;q(%7x_iNHf|hOESyU2psF8lV7_Vyp7=z-W9)2% zo!P*v?nm&Yfze!+>WoU21+bbR?wY%^3ysA)@LaJJbKYOV^wme9bZQSm`d-DL>}MFh z^gCkP{_zRTzS1Dwq0)P8d+15gUs{-Xold+<7v%owklmVrq%t zc_!yR&td!Ddxc&#DdKt7koMkqYI7*}7Rzqj%EA^oMEE{Rf|_2H0Q`;7(b1pd7Exn%nLVU$~1N&lwE(24em zH0RzM8tHgS60=PZMjqNPIW^al6o&;;bD9|)Y%#>spC@6y_bt*smm>C96HFrqarImk zK0abFS7q3;?z_2MrnZtNrzf*n?O8aoCGYAh;qjWoVJ7-svg7BY{!a^SOz1A|BvpV~ z_hNcIDW9ZTS}1+m$GZH|6G_P0$HMdEe6qhO`fESUr)SheT4U{?o}P^?gU$5rLK%5! ziyqRe?%cXioSO?%*?!S>lGlvIt>cP3_`EkCOjynr$(I{+&e4bdgJ6Ha6}wE|kYD^# z=stglF}jhIATt(8HKMDr#vgxdF5}e8PH6ZH#f$~Luu`TAo|c^w4lNOD7yTCa96m>f zUlvf;(KkuSF`w;%GWpcC68^np51k6UB+k>%(($V8cy;I_W`~P=nxc2o#WP*8|5Fa+ z{`AAUx6=`SD-9Z1pQwKLCHlFRC^tXL=6&4-p^xuJs=ji>MlP?6uF9Q2;(w#Xe#m}t zKB&mIKR)KJ0rkAb_!D8H~!_TZ0T83(T^yV8< z{n8=CPm2?(zO+d~^@}AF3}y;1=hyMX_4)ktfhSw9i(|!#Fut>P0C!J(%?*Ric>MFj z^!92g(v$Kjao;%M^o%K7{NOI@X_-jJmCdJ!Kbb;B{||!Rg40YNQn@*Q9oJ4*!`-(x zglAVh>2v#WT6{_L^8R$A%%cTV8#+evz|&FE*=)%HC)D`1I0Lsg4dz{&ZeaV#{%}0^ zp1!O&2HVr({#(N}(2>i)5t+UCt(;8`;gQ4zw}>u-43(Oe$y`VFg8LWO0s6@>A~ zk@z|-qpYI4=Fro;!ayF{&QQ7Um^}Nfns*Htv7u)^b*~7 zLr81#eu?gsT{bc*iXwC0lLsr;qqanajE!c~?+vFY#AyMq{pG`B{x+buVo!F&;)2wBT??4{y8gCk@LTDA&9SyJJtV zGT=A-1O0hK_;G%|+>H|p^w^`z2{h#_p>2+X$l#hB)s5Fc?s$K!-Mb(6w_n2IYc)8( z^9Gd0o5QBO5#gbt_h-o!8hc}%=(RqGrpI&G@!d4Ou*H;XwxyyzbUlT;th8|$cMyx; zno!TUo)VK61EAX`?)mL;0?%wF^TiRlyy(Rpf%rGH^Sl3T+z| z#e3HYp}Vs10K5#>hg1ghrJr=IpY}9QSla!FWh8;zSRP=~O?&XdZ=yLM=G6S+p-c z#R|IvQ0iC=H0y~Mtm zs{CNI5o?7hu+eTAeywT7cP3f#&PGx8k+qbE$EVToZWW|6`vM6gCQ)wuHL7^~m4s?b zWW67UOVN|iDl0x~)d!0F$SE?JltpiIO6c1p8Ih?!$%6KNo^aqcmn1(GEP8FGw#)%& z75)23?c#oc&XJfKWxzi)hj3AvIRCgMONX}V!qf8#>gV5vTGLr*U;YWF;SqGT)SL8+ zCQz2aSK2+eEB0n-<3_s!xqQAuInD-XxfX}Wsks>1v=3Jl#q8-AP5iz$6r+zZv{P$Q zsjA9nzKi=16dmF7aV6}%V%g(GHh+9{jlI{br(x$K=;f?E)OEHN3YN)2*H9H-jor!M zXe>>8nNH5)uJ&8^|Dz>2>R4EHg#xQ9$m;ucy6x&qW6u7MqVsUX@@>Pg3Lzvrlp+yI z<$bR6dL@+-(pK7pYJ` zg|3~rB%QDdrdj4uG*Rgnou8x5H60B(`}|;zm2G6D3$c80LB7a{bw|bBc{Jl%wQT*< zI-FCh!g19~On7<;Eg8FDyEzv}KmL^*6$+)Yc~(@V+DPdM4zzK3tP~bsOIN4n(ae7* zsC0|qa-^>3>pw?x;RSP)t(`=H>0QWT_jT#Q%>>Eh(?Ys_%Y&zB1&H(dD!yvb8}80d z_@J7Mg{H6hVB|dxcqz`C+5uSN6ot=APU50}uyl0BODQ-zfs*F;pq+NhDA&D_K2A=g zg9d9!X{0foS~HaI=6Q2s`3yE6(2+*BQ=)DSAEcJQA{W$uFqiD_$m^Q-(eThsG-u@u zYMB2M&jPk#uBkqJ2glHkrYLGH7fjT43jA}p3hTS-u3Pi-(soLv9NTdupSzWQv_#O};YV=V z&IR>hifHgl#K$+&5M*zQp93$`uOni2-meeRhIV7c+g7}+w-L`wyo?Pl7jbU>JtQA= z#)$y~Fi8HJM)j$}Tf<`5{@aM;pY1s8RvhAnhF^ZcJX<1t}v6v z4Jf60&sj7+^Q6?i%$2IA_|mr07`pn=jPI=P!C#ck_}6%AE*IXnTl;>Yu3{*6-s{Rw zo{4K}n=^F&KPTuNK71ckTgXN-Q7IkcoU@m82s=^lNd|Qu4u)TSz%uJ0R4yB9>my}b>tG=3yY11nZu(TP7yHsNC>Cul+@OC^?~iBlLdm)rcV^`a@i?&FlpEu^u+D8~gEtu>NI*l)thL#GjHq`sHJY3iY z_V>-vx9bet_!`g2&QV-5ehsT$kx3h8ezHH4yu*IVst}&D{4cidQRThUc1wSQw@4iy z$4Tk)BJn1(6i)63@hayj^qmT@C80O&e7D2#Il^`Pr~;W9U#Ugj3DRl7L|E1$k6o(m z@b8SaL)=C^&T}~h9h=Q4F>#>{YqUt^pdCGEI3>JEJ}4HuIK%pPfwf+` zU0pGe)K+$qj0Mwgxse_Z2~=U_L*eimE4)dtz&ovbFzomdR{uo*Cfu7ZJ{in&AGc>T z>T&;Dc?DQ`z z;=#V_xc`A}+*$J=Qna6qorNQ2h%bvH2C5f6qVFpOE9r_kPubC%FZZl~N5BZQ z4xNkRYuBP*EAuW=nx*Tx zaunjL4}5&g1toYIZzHc{3YA5>0;@rw)8~apA%f{Ns=jU3>qbhCWvmSMU zb&&}s`9@RWYzw(lQ3LCD+RXwl&AB=<4c#`g9DxH7E^-=C(! zBZez-a=9yP1oLgc(vHHDF_iIVG6$8JC5d&I@7gbQt*2C1LZx!`b zKTV_FkE9a|Lf{;A0xN~vd79?D}3Z($0Aw2uaGCKJIP%z4kqgraq4RUogUelx;FQqT>C@x z&ME`TeKh&Q;!w6(a)sYdPT_S93wc+r6)(wi=Fbb=L_Q)4oiC@M>CqV!-&x11t2c1k z86U2y`9sN?_oV|9_e+Ip?$p=&o;1X$Olt8wM;nZuQM;F~NNKS%AGo`MHL@f5fz~ao zZ+wnnVQQ?g;|euAucDzp`=NRK2`bRNM=naLNYxm?)zTm~={SQQm3i@f;Q`dJH)0Qi z4j77DY_5KZj{ADhtL4k2)gR)eJB_3G$_I6Bcj}vXPZ{$~<7T*YZObkX3Se&@hX<8= za3x+&kGtQbw}vHTbhZF?dm^z*winr_dy3xnWlWmrgznex;pN4{*nDgg!Z$gfeqL`B zY^@@XD-E)>PSeQOR|}T{y?Erto_sFUgqJJ~#x>!^e(IV7{la$g6C)JmjrLk{!$-nD z=(i8Sd=EPMZE+?22km!kp+kK?(U|T1(K50J>}%&sV^TJezr#qA*|2yuuO3BRCMbfmY;4(~{@Tba(hW92i}O&dO=% zqPmK{CS9d>b3cj>eVyd=xh;M8=1p%uXYx4Dt^8%|Og2^B#B(D9xPJO@F8XbVnAs0$ z)WdejSBQZ@R2)JUpA}9wCw?&1iL*2N^M6YAkZf~l=67#8tiv=RA_*$ghKHoa(!c#N z_2tx^pmR zX>JwVmfdVIXe&?m31=z_=SOwcY`#m24RW7ix@kU5FYbx2!^2_q%na8$rXy+pci}Gm zz_GZ+Pfx#KnPU+ShJ@hElWcr$6)ukiBRn;2z{r*whe18tl^6+5mq%5R1ik*ex z7RZGY?V|{{6d9f6!tqmi>i0IRHSQ!DN2u|imX|0uK7*ioEqw^3dSV@)NON<+nCmgX8IHOw;=XtD`SyrSmEpyna758L8l8iUN+9 zq~lgPLGO_=-j-Ql@?{0MOi!fGGG#0X_rcYJ87SGR#S!^0@iplp9Cu_P)W=>pa%ORJ zO(4%tdjX?Mfta^10{IcQ@$*=3{&s8$TX`VF!Y#_Wf8^daO=P~&VJarQ76 z$Hn1^T)o2@tNk2c_@62MbQ{Sr&u#eMD=Xf(t_#9%EXVBC`>?TpHo7%=qjjq%k~?+g zkg#A(FfoT|p0!-R%2pomL|NWzp&Rub7b5b+O=w=WisZ_6w7W8!tPY*xs7nvot?D|b z<|M;3z6y_nT2c8y;@yisz}0Og4z*WAudWKHb^An)KV&>IwH4Fu2)_H4I|%ua0sHJj zsLG1q4mM%DXpau>v{t09sTMT0cMknAU(MH6&gBGsV=hy9Ks!GVL74Uj3J9#lgrLss zHMa|o&0;(b=_UFQM`==|8c+L|%!`)RvZ+-d&;HoV4Fh_~&5Fa=@%~yqIC?4f zqWSz)HBofGO8KpkIcA&RqZ=k&XlECDzVY6hPizh1R0YAaob^#MYVSgeg?l1#wh@o} zw-*Z>1)DwXA5upGi95XNGD8$CvDzXHnv>PzpD~4 zbI=1A&C+D8{wBQnuO`R3CE-bBF&%h1m`az-kd{0z7+Vc(*AwuOkAj$ zi?OxqkoC|3GCw&^%VH~t);EI83mG5SGYb`7>?pW(i|yQu(y z;$GZ%cOXXU=)y{|nEpl#lrGu2;&pd9HVNjJV_k1)<GHKza?UX@5T zyooyJx0t%LD}R1Ak+SY*kVb3~+1Mpwqg^qYU9aGr$O~NC7mk9&w=}V*B1&BLk!IUn zQt)(Kqqf~Yxg>;(2sbXcR@R@(7(6`R?ChicGYMl~pygjtuBazJy zrEuh*C{A1W9j3Q!`FYooZ1HXYHmOw5bk*N9JyKH|UH?KdoW7U>{5!z(q&8m93PNz+ zV(zucgon@(yl*^3W0JO$fnqh~T(#%YQ&%zmmOc*ToxqdeDB15nQIchOfuyp`n(jGzQgncCkAQt$$6jILcmZmqgx=l_KUTpfdSUJ7pv$C3?#VRWq>?K(FAci;O{ zru#+uP$((fv*WY89W5nV&1v=GD_J#i7(%8c%t>H}h}j z5_Vqvg=c;;fW7@hTF_hY2g{YknLHNVSL)$~Uq`IgoZ6~sPolI2&`;wQ{I;{HAh7-OT zbL&qt8t_mV177Vx|JyQZ89j$gC2gptDdTB78P;}>fPG#v;tImC(kv93w=(QM&+kH2 zF{kL-*BmYl$!DdX3i3YpB&hFQfwyNBFkOEWYL{t2Oa7TQmp+%CEY6jjUQVKY8ANGWaTA?>a*hsk z(#QUUT`*gI2vI++&>9c?8{&rAiLI0~pC~Evtt9u|O1B!0N!?rw>9NHW9A5NP#u*)C z_eyQ0V%b*vmWNmDf1J*M&V#OOd1@2yKQRvrHXlQ2kvbQNe1zQ=HQ^)JBDszeUP#wF zpjV#-wRR?EfjZn7m7Htk~aPwjsY4DbR=1<$8SW^JCj{d z_;3uaui{|Yei1+Ln!#tvg6imO%F_*|{WLp~U_E6HE zwwd)dd$YnjR~|V2Gd?y-+%Q^~U$)ob!*-8w{L>-y{%@hQUoic8mYUJJ{tb{N8bW=3 z3n>iVhcB6-Xd31N*P|;iDa9Wrd#r~;#ykEz>o>a@D$B3k-bMK~_o?}v2bv$Mk&WD+ ziarjev@ey^Gvp(k_dY@e8)Kx1=lf-bIhm5etOUO4wSwnmx$?p7;bib#mCBYY)81MI zxkk%bp4dB@_g43X$}xM`o_I#XwiZ!C*eVL|x{Y!@j44PVgX+_D@I>&k$6pC!&ndGw z&1e*FSlo@DXura1pMSJM{UXh%Tuy-t+_57=?0KX!@b$eS51V#UzSccozF_yWa?^hV2mQ%{omf#H09l)XtaHhf|MWIx?_Fc1*43+MnpQ92@39cB z$3OTNeGA`vm*aGe@NajiM8udha+v)88P|3Q!;dt0pBa5``aq`lgctU|%Xro!3Vkjq zqolko{0m=@{JQ9Ry{f_GrB`8U!)Skd6k4w=hH_ex?EC5Q(rA$ z7h8Gr$59-;t`qwf+~?c#PV?#;yLo*6IV?&j#?%85n2_MY*7G9R|3n-&c}7T+FZIR! znTaA-Bp5!z1!#0nOS*ZuyL|n1Q+er*&hp;oyLnCf!(4s%A#YDz4ez5$xI9iy`7=B5 zM28Pp$`4^X+J_hWyRhFXYkuCKOGYDWrJE%UQdChJDS3k@#0CS~lJxC2L_U{&fA&IJ z;CF}gp7x+SnXWWODUe>Dwx+y4N2PA}y5mf3H|!X-gN7f7=27nw`RL?KKDMN%yyB{* zd}dFUcWi26ynW6I?pFmn;kr38YW1w8+z#W7TxHpxl@(mbqYJ&vEF#A_bdl^fAAz01Zus{P??I zea+enk5QH!TeOj{o!iG<9u;u+aYy(_T@63_`Ii5+D&SK7AS4_(heCB_jyl+d|0_L- z(IZ8Nboh2|Z9kMJ{&wW1FJt&#q9s>u6FI`CFVI~s_~sqI05hJU_vq(HNN=DX1#+DE z&l^^OZak-cHzzh%bJON7xTC57e+6L86$NCLm15BjU7iw`fM8-tUBa}1YIw1iuXY$ZeU<19C&baHWhwc#(&={J5*{4;x z{(&uJF6kh+*sZc9{+omwWf%vn_yL2~>sWG1d?Pyw55)1l*u@hNVJUi>zi$(@wS~3t zHODn7@YTv%ynl8CuO64;-_Uayk#-9%pOR4D^As{W=RpKN`QuhAp3~EutDOH*_-0Ei zS+Ey3;vY&`6^YVqnJry-X(BcHERa1+F`)S2MyM<4jSiv5kl=NRzV1+h=`mHDp0u6A zdZw{jTr~g4>P=Q!N)$Tvv9$i}K;Gr!#eUz`@}k%yK=5G%S}XIIIeqb^gE=g1I^e_p zblT-JpLQMGPR0dM@b;)d&HzO=_;L%j{hRT!-!XL4=`8#L%Hnq|day$@xWkYIY&pdP-cabIz9xI$Dx_m-43jKx%clXV%usBeO&6dW{nQD)T zi$h`A{}}8NU%=g>6p>}8(WF<3Nwb5HJv33$lI@^hx|-Ns5&@l`>ipu~e%5{_nD$HB z@U_QVk@)B*%5;qRb;kw->g&s|X@6nss}=mx`v**;ZP}~Fi-T+WBGhCz0w;Iik`;}( z6fuwsOp{r@Ct5b#Vy~=+Q=M$>o9-0vnnBJMK7w^`D%{jV=u2`8MRtn8hi)oJmifT+ ztPP!XdrlQU0IuyKZAdAR2B&J0^_Oh&>6=5|zn{_5xqC3pWHH`~+(=~W13K?wkAg4j zvC3d8Db>w_X7(ejrUP)DB1ip%O6n1iDGxgDDi6x8W~Y>m(vco=I-+PJc3yshU0Dfl z!7Io}UWjYITS&9>APW7v9Y?e`!D`t-Y?BOd%&(EA80@Ff-$T%+-vcr;cOl>2p42s^ z6Xe519{B79IArff<>v!9r7aWuhFa1bw~qEozoy% zf{)d6DK@U3q&Mjh=?wlsKKFZxv&;o7xu=L-cHgDPK5HrHZ7+&A{zjUS*`CIZuB5O& z?&v(HEjI{PYuNK-#5mqV^S~UDB^V;k#viGEq6yhOOO)Q-H-PK4S)@C=Jzd%Nntsgj zq9sdqz>k0E+Ai#v<(3GLMuGiZZsZl$h#I420JFLadO3XbN_u2J|e z-36mU+M#&OR=T~sl1^KvB6FB7w;cP%ljfe{XSvC2={JPetdjWqx1L-*y##;i7%}Fi zn3Wen%c2%iqkV>VGfV%QY9+P(4e{?iYN zu6-t&#T_nrhsaUxF2<2mOOeqS$<-^B`K?s9mKx7#_<{# zHL3fs&SbhxnVK`T`L3BAk98iz&x^-G@0ADYZ!-2)+TvGkTW+gjz+2}yO1tdEe|Z-> z>EqcvjQBSV`RO6}`J*oudYfQ@XE2nCi?IH1FKiC~M;lM>p;Z%i(CBP`ino0zceFVw ze;QCIZ_-P`HS6E7R4BvDTwj<)dBMI9!~60X!Efn|=0iG|a5WT1Ber8)`3mfab&`Eu z*ikyr8Y;6(569%N>!@08$I4eSpnWR`eZ<-Pi(M^BGF927!^(N~eB!w} z&u^Fn%Qn$CvU3_brL>`8p%-b$szcPN$2RtwW6gQLpTga>o4oWwJGpeAi4*VSN`E$I z)4moD1V0`C@0EsF^wt#B$+hC=@4+LzZ1~%O1$ehW4%0`!NtzvpDpem$uDe4P=g-PY zenv_4J)TJukL-o{;C%G5jmOV`G3eAK1g;u0Q8#TV{O!d&t$hk@|0p0OL)WnXk}<4X zsw(f?TE%U&V!63VUA|&S75fKg^MF7Ho)~S!6PEr5FD-qpRLRD2>(8{H(U4qvs$+fY zI2heePJJpC2dn zV$VV+-rZZ1&vOfwcD;f3b4=*N^*U*f{&?*EyNm|i{wA3P2zP`>cYYE+jg?e`SQ@Vk zjiotQA7I1bm(TO@Q3v>TCz1bXpNF+8-s7Un2)0_TE;VpIwMZ4{ssD`p{0tENGYrG; zbin7|f27rA#?narcPI|thdi^6oEdnJ+Fo*nzq2=LRCM6$-b!a$YiOXjv+e9Yf&#w9 zNea=XyefY+%v?{C`om=EygZklZP10uo?@74sq?H78Ar&5@vu#n{O7Y8Cm8P}na?7c zzFvj|!>OpznT4qOSXf@Hl#f1DE%)zrM_wiPS1w20v8X|O$2+P^lQIe={gh%F{V^C) z>RM!9nt}`6bTM(84&oX*!b@g>q*hzxuULb?NRfeU7=rUUCc;w>>{}R$yRIGC?`~&a zTcgLTdx*T^{0*YTkW1=_GBR_bLuTpGTwEhWl&(T$6lwEbs>q`qq#IXjI- z$9^sd7?OwJ%$KN}*%$k_ErsLY)nfN*!DTBY-W58RHMf;X&qDjr@?Lf@*msw%p@u9pFUX{KR3GWOLV!Rk9jaNw9JS}A!M^@SwCSIb= z3xJds)KT-uLO5(S`F_J=n5cVU^W1h&4nIxX|LL>odVAg#W5`#m zRHScx&q~fyo{{`P43>}n4W-vb9AFoWr|%cyWB^NS)7l@CjVS|@Txm|(C&D@&6ZUMdcr~e%!IA6odfkkApM^U1d2%K1d zmwcN0NSBUH=dDvS_}kuM{`J*>@&+uY9~X@=cEB4HRvU6g%v{0!aKZN5sAQL|UkAvoL6p(LuWL-$%Q z{yoit`#wB}PcME`aBdk|6q}L!I~i%=B8Rv|19wDT<5k2H8mZ_TW~L>r-g;*^98n$#75cX6c44uZU>FlSe4XmHM3H>tEyee6s;tzgvna z$-}6f$gr4QD5CNN35BL%v^hkD@-=K|+?uEO`BFG8D@-}2m}!A)F=dNx=e@>(RITYq zm%sJGY^McKe)}9<%DOYVtcLmLB={HPV}5UI_DI&{0|A%OQSS)1EPTiz{aXYFRfX^0 z6P(beKPc|xY^{wgC_j;7@*p3$5+3h!WuWQe(^?@7|6v3VmJ2|_w zih9SpVeO&$;(T^qOXik+sj|a$ewi7j%fPyZBJAm< z%WuOZR^8l8SMS7NVtFUt)c2Hd56;K^%XwrTK820HZRMv{VXTv7&VFAO^O?NE9G@D1 z<2s!4u4Dr7i#HG zrzmo>67I;e**sP29vk-1MEkGDvBGjH_gFes8lN0V^K25JW!^{J1FvA%{(d}Sq1d6X zdJ8-Em(W~m%i5VCxaPY;%J}BR-|wvC8jldxO7p?Yu^aJV^GaNEyabbh2QbTLkC==5 zBjBtBdd^FtX5~CqdbEujcO>%?{alf$P2$;=k*wk=GC>BOyewca2MnFd6zdY*=Ff%-@@ z9D;+k>g?9J3YV`3V$=9th}0EK_wkF+Bc_2Z=^M|_)RB+#lt^{VUfL0(gRSBnAlpBS zcjOG=PL|2=dvgROb@$<1HwF~6 zD}$=}L`+oJK*68mrKUUuwiM62@$R{(^FM|K?IJPWdIFxfMoVwb{*>vw@sT3`_2kP6 z);!S7nu8j(STRvUoa1cyLE$@C^gE0W^;_|4Qb)F^^5MdOe7317}&d zh<^%d?2aM8ArWT_2jbJM>6qnv0NTdhg4fp@_ouv*4aiTI#jC~3;vfHj_uKyBH86|X zdK>b29d&L}`is|hEqT%L%2NG4(rAXj&74Gd%gcbPff4W z!8@;nb7B{-bMj>fx7LuX&bo?^!go-X_QI}y(*29ldK!={N`h5Uu zcGkSLLnyDm$gte(giGc};h|JTrwdAG;k;AypzUqkKQjzJwmQ)<4R^R#8DMv$F%6i$ z9mig0!bW*NPCRJBsjLeac(4j{zBBK?TnR7X~=SEHoTFS%g`-whMU%now z#CLObxnpQFroT9b`8PZ`r&CY9cVZcrE&79Z*R1*8qY<3nYdx-cIN;}eUpV$Jz?XHDr1pwtJWk@y zH#6B`PX$*ev{H1~ZyIxWJTwernP-xE@cchHw+_#`ne~04kmVx|9?-Oq9n}8K| z-()#&O=-)tW7K834zJxH2pvXicudehOg5X1gf1u0=GqX=tK^}rCN757kow3)yv4CL(r99Cq?}$ zbYkIo8nZuJO7eIwy?oo2Cbg`joOg#P$Vt2dITf#bpJ3fe6~28zpFKMXAN=NP_$|5! zIp5p!Sc?`^CYbQeidy78UW4{c=gCjo6S=R_5E>hfKI#SdRr?q}#%)DrzrN6OYYTT< zH`+G4QaXR2S*qRk3R!EfV$m*Du3}@F=8%=q`5#LltDa*~EJIqq}f-Je-eH8Cq-AZvfDwyRm z1yf6pVfKeR*t%PTQ-x2+OXMwI4*iF@MrG_0U&^1{3VGY3Np!*6hy17OQty=AsMe{& zQLR4Q)+V3c2koX$5i@D9;X}Abx8v^-5Ab$&Fzzfmjqciz-fRZ{3f>49s}^ z+$oZIM^kA*!+LwwiCTg;;)Ulni*WUtxirM3S`xk!+WqkX&Fi2<8B=W~vmUmz)Wn7^ zkH0CsQ|wR5V~XgMnjI=OEyf>@Xe_Hdhpc~tam9Zt-SyKT?(~XcGS*1T?)|i%nyQFp z@ji$S%0koAZPKj35NSZteQAWLSPzR{!^7Q# zP2%nO@DLk*eYQ9Jc9}x=mywv^b(YB%$I9+5J1to@=gMoGPRW1#cUAsy?jxEvG*5C` zV=S$9z9IbxekX0(lO~1eS<<0X^XOIPMq0FEtu*!1Vd>270!cA^n)F+Fs_ekB&a$4R zfvE4X0?`}-^~e$|seBI41D0I+HI};MsZ#m9XsLgkxZ4-p#;@uUF{c?!YG*vD|7oVn zcRZwHjo)N<;w+??ErGI^oA%1C>^?53ynII!TJDlnN)5eHjk4c7bCPcxfLEk?F89(=;m+whh^u`fQL83>zxhE7hCew~R%c*f)7rJmGmqIS-qI7Z= zbRUX6X8wK3|8|xJr!SWJOmvq8W|olFZDsTtu8KWDD`~s&cGBo|lwNEQGbG-Ii+h4G zP;iX*j*O#m8Lre}!C@@xCK%h53M}hwE6degA^Q-XAd6EmrqaiGQulY8$?CyZN=Q0Q z(!y;NBs(s=RKllSas>apc7W%|p=P521I0!hssi56FIy$2ngN2#sM$Q2& zN&G@LANEJ-iRtL(WDRvETX@Zhgw=%z9M9>1(kMgJHOs_z(w=o^+VP3-jUs1u4vW8i z#*qCV(P&qI<7X4l;@zK9{NCY`5qNa-2==`S#g?}ck01RV*BheXQ{4{R zdT3*e>ow~3PYZe-`y=!FP{d^!Vcap%HK}c+IWKSVpf~kAFGo>6Ys*%=@32{T-W2d= znLZaCzJmh;PL!|hZsrK1P`XE=;w>U3H0kC^RTzDaNLEYW??JhU3`gu%IVxQ-ly zPn|_J+N~cR&d8_M`b;XibDVA!R`77cOZ@q1HEX=qnhm{Y+?d>sGT=@cj z+@mNyu9M(l8B6X{3S@&H--F893;ooj7){zGF{gBovX?!~1endi6)KAyr5 zEV(-X_x_%S&x8Ufd!ECXb#;<@xTct4*Gm~+G?ChRhamE%J63(ik(>FrxkAGE_nA~S z>@^(;2U1p*phv(CDAB!2cq(?VbaEHJKd|7Y|x(g$5Qe;jXZtTFr zhu!(0;A2__&*!l%HaKS-PlFd%(c)*X;AJri(Zk(f@U<@{8%NUb<74PcCwC5e)Bv{& znaCYxCx2KwSU%TDlADfe!?$y8;Pm#hFpcq-pEviBKkniozpzG?7wdH9V{$L<7$)3j zyFSp5cw4>o&pkCpv2Cs3!p7-xLued=2kN5r zTNNcawBT>sG$g#Sg0b6Y%vBi5{qy&5&vjmWy4y;w^W4Yftj7V@ocR8tVeIjKE`Db` z#jigyZZffE|8>7`H@gyAp%v2Z*p<@aFM;Ieu8X*}!|-qRc4UhEbZh+@v1@-yFVaqO z)xinuP`?{VB{^)M8Od6weR*rqY~1{@4`;>S$Ea&JMCl&Fs^1RmJK`ZPPiQB<;Umd^ zcDl^+VHy0~Y9B9-78zd}fG)n(2rrrwBJT9^LUN+l{e^Sj|A>xLwx<-DX7jb<7uYvIPCaa*3>*fiw=vN^0%(2jC{r` zK9uvA3+Y@FzZRdC@4(_QW07eX2m6=C5S>J5Ea*l3#+OK$t7b?cSql7T+f@vg4}xOz zInfip%DwiUB&gC>ts?OvZaR;dJ>W0QsUWhSXF5K-mcADVU&RILC=m>B0iev<(kHLQ#!;rT=4yzOXATH=B>F2+uy_bCGkl@rM zZL-IN$J0@pAfA7JQ|fv{Y^Mu0@wx~5xyy}A{&%tfK|L+d#dSMgzPtgWfGV_=b|bOc zi3eFkV|ZE{!HOwEty2v?h328H+j2Z_%|p_l*VwePnNDhNC#7vFQ1npc|H5YQGr{!p z9j`8ZT-PL(SYDC3Js-rDbKavh-yc%_5w_32$YtyHvDuVwXyWDE`&J=;=`sd^7M@7D z6N*AF6;4l@%%;L6q8Y2elQyNJG5I9@*Y65kew5*7LK2SVZsO#bwcKu{o;>F0NXj_Y z53@p4S^B$%bHA=)$6hnJ(eb_LhCIZ>m7y5m=+0)z<5?Ew$o58DiR$gCcmKW{QG z-*A&$PxL^c&K_*jV;sK9hFD=kLIS^ag?}d7DJn2M_ zmDDVAscJ)n)Huw9T-^h(rh7lp`RWhb9DR7MOs5$SJBk@e0_OBky7v^! zboXQEKB5V-U#~GG;14c)HDHvfjLn~lcfipEzTp$Za^(qZv{0277%b+Sfu z${f=+l63^D=(D$&6<_qg6m3O$^|TYrUXFaP+gWsLF=5YJ9oekA0>>*z+~!Iscf04# zN}@aQUx!|NW0bg8?P)-1`AJl*%izSvZ@7QDy4<#+Trvq3x&Nd<%zZMAOYYC-qYY6U z9pQ`T$9JMc@aBG=ZN!BQmW);hUiirg7XNgh8FzsO2Q8xk#ZzeYn4S1|!vpPir{SPY zSKhRzD=SR4;-(9iFs9E~O#j&)`J>IqdYY%a;lUEQYNr@^zb0io+B6@vv);qIp`5fH z3={kAH}D_Q7Avoxr`69-l6m$sT&*g`w1x9JfVqd`84>kwpt*^h@H8=-VAms7|9{C z#Lm>(gva_D@*OREHm#S#GfRLDix=qz-J###lTf0W0fkL2;%tyW3r%)NZDZe2 zx#f5Cne4%jEIoLK!gMx#9MAjea@Zv}jr-)>qRHn@P>FmqtVh(+HXkEAY@35EK6N;; zq>GrBdh_eH0;8~419OUWF|cAhmx*3v@xn>`D18UD*VU%r?yu|z^@xP`q!LWL7z-!s z+mNce@sE2`dD+mF!fzmaQ7zqJTU>^$TXS$%yju?THQ>z4AK=do$m#3Fm-mStMS~&_ zxT%D82dqVh8%C69@}%wy_}=6-{Po-l)*cYSI*a@93l~>z z*L@30VBMik=oO@gVv!SQcl0BTb^1qdFW&=yeuc%xgzIgm$UWu^lBcwo z$uk~BkiouO8u{rTIk!va!|g1%&CE2&#dBZkeH-Y1O7PRS@t6u@PTA^*ed119yLN)4 zZ7^E)`LpQl*Cuc$!zWy=l>;5*Jsk+6$*S$!{_dUp!R(~O2<7QEPr>=92*()YRRrRPtTXzRrlU<=JB}DWLhy1y*+=e( zEY4sIQSgVsPA_y0y-Ocd=Cj*xJzlIAfeusa`02eXd@HtulU`5Yul1q_c_EAqm+j*7 z`Kt1oc#^++EjS+C!$D?wv_EG9E=LE$ska>hou2a8G413j+TG>HI&a{uBeVJM;xqj5 z%qU)(vVxbHZQ~9>e=tVz4xWpy!tLQH9Mqc4Dpg5*-1R&KT(qO68O1WKP0?7S=1giE z57}F3sUx@jQ%b)XLna9}s9qO?gAWp6KH7==U3QTB@NLxhwgr^h4#E}Nu`pj~%#w0% zHty@f-=pSoc=`@L*;LN+MKA5_?X!I7bt^w`Q^3aIAJMctmRl8-Io}`?4bA$Pv}H8a z%ev#%kxaZy_{*;|)#aX>HRK_+IhcL91%1jE@vW{($kS9HpKL{GnS+Y-Y4}EIZ0`!G zW7n_jSbdeZB(CPrnR}SZayY4HIPY`4!X@jb3&+|_y!)lbwF~}|s%N%+m~A*0Ry<;_ zuBLLAlv}d(*DUGEVIMgB7JMbk!E79u$|rgX&)ii{+55GjlKHzZB(&B_3i>Na;gbT? zM~|0{&hY4OocazQ)(C!IZab{$^oDi!jb@L=?Wm84MEQUu+#Z~YM+ZYNBJ(_Y#D7Op z@et_Oo}(kf`cg54a$jS?nezo(THTXnKT5P8djF@a_tW|8_bigLjz_aknK6GK)`7Py z*I>n>bFgh51Fm^Z3$3@ywT4*Cl{_{6*sX5u)&j0ke z8pRvDB>Q_j^+j8`WfOPfI>}aCy{C&YT~?vyI4AU1P{$eCmTYvkgThwsAh)flq*Lfm z`AZJaYVHZ$;3t#ULUl@zm%_kL8Jy|mouGO66=x^FSc^v1Fw?6EG@|_Q!}%e6ne~h} z%Z~L3-poODwb@+2_3RsFp6qtm)9^?&+WK9?d%qIVb*vrkUHkV#g34OIo1emOOl9VuXEf+d$gbo_as{ zQj&ZE%XCMQkc|tyd7we5v(0Hm>1QnZpoAN%PvYU%0&w7%f#HMZ@ie2b4gFkG&?o;o zJqj73rXC@_?%QBUnpTxZx^KobUM0C$DhD^$WdW|ux zS160G{+i%nac6wu6^E6RtKqJk*{Lk38R(T+63hJTL7jb6qXRk&(n) zjMV7jE_aW^S0sw*`Yj;zRvUk8JP)2h^&pw4LgQX)QOE*$Dl4!iqvCw5cl`zvZzV&Z z0AtmD{sg|EgS?rtL7a;776|XWiR+5Lpx%Zal;0@Ni_DSbq}=3S=;e5pQ8A^N+CHQ| zqZX?wzoK~Y2fQv2g>ie?-eUfB3|SROYkw`E<1<9a&Z?F6WVe%j>`7YlX*0|pis5z? z&EidssYH*kNDTdIg;531h#vhR`wSueSD6g5?`)xozEXVE(g*l4wgD}=p5pzpR(RL> z8@7J2rH$ojBpFgi3MHkC%^?DfLwk7_=34QhQbxEV-I`qaxb?g%aiK7$CJlC_-U8b{ zi4<`xg!o^b$?>s2X~u+;Ge45_Y;S^ll?)D>Y{dnpQe<+;ic}9okyvmV&AX(;PhYm1 zpQe5q^A58f=F}6E+W&xx|0(ciwDb9!&nw~1i@D&x<~jF5IFt&7yD*jg+vle+=IRb- zT2Yrt#&?TQrfWTJzGjKX)M`=hLpwGa{6+CA>UdFl7U(4#b2qB*&?)viY5vhpiNUO^ z`g#S;t6WDn8Qc8e{WsXLPs!`KRYZ@Ht;&=DC{za=8jHKCBn-dVIzk2R>o0=pM#d@y*cGy46L7lNHcYih!F5haSkM`XTUPjCmb)+=xN{lb zs-)r6aUNtbe28xMrI34%G!6WjPW6YaNH+5pa;X*Q_4qPcUy8!UP12|(a0>o4e8!<8 z%wx@%9)X2IXdNpGZl*VQ7H$tP$7eOVvHskPD-j@W6b2uUHFD(v*RlP1AEwNlOsO-s zu#RvjJ-0tY22RQ}=`r&$X#}u&Kod?BufT`qO~|(G7_=oBv+JXAub(+JbvRLHZvq(& z=FuO`WYW_-LA|n1arx~b3@M(#x=m8}A|o0K4g^7nnjHB(v7*JVqKL2HfcvftLP1A5 z{Fv)SVuJ3(FW5u;$Kg;Yu@2_$V*8`T>(HW@L!Q$ekk`?`laY+=#d`wJRXCLVEQ$Jk zx1r7=0jHJ)390I3o|AI<``o#|@fYoq5)`9a>;)Wn!%s?%wfnhrhLx{coXo6?C17IedYEljy_gflkZ3xmfWaeDsa0Rrl| z%NH|AV%7=TCbyepHizSr+|M9Vo(>JE`}vdKE$1ghnebg~M2MKbT=HZqcGnZ+s|s^g zLDrnDzymlKtcQOW??K(>2*#tR;7&+R=e{Hd)1jqFBwU?IXTx&2;LkI`B-$W47{wU!7blbQ)Ov1D&k>IP)^fxBiKJ`fPqpLh zY1DZ!{b>}TYeQ!c6V6j{(o1@}{XNMe&eS+4|y zjdXE+BVS?fy2~)O+YH1;D#^>Njug|ns3ECzWeySAnt{T5vr%*60C$>~)jOKj!-rfUtFXb^Q*br;e#PH)9b&QbxfjcJ)kjufZ zXduI*pSxaSuYLoD#xQR;=|M|HD>we>JI0!M3umWBfr{f{%ouk8U!9M^x{^bvw!a>o z>%__RnG^G+$kH>d4Va*H9>+^+u^d_k%t%N<-vwp(h3)do%Y9HL15q_F97STE( z`2@7kfscRi)#oqxIsY&6X1L)hmZ3iseg_6;38Uy!MabQ;5(I24Fn8f90A-)UCh21hZ}Dt{& zod~;(ng5h!57HN{1+B%<>1zJ#j_K2#zss!pCVxm>*Dt z)?a7Ya^{A0)+@#|rGC^B-^%jMkuch`hg-gXgxl4;4ouobAe}j4drcy6QjZB1|Bw4sa~8g=eFV9^F>tDU zCY+Z1jWs4c*kde4nI_Cz*0l|-br+*>mOqFdcZP9~O}L7_5y;zT3io&HhZ=XrTualz zt^y-e)?%}@L37+Lt;J@(`Y8Xv3~$8o@sy+nN=|mfL(`)1R>}fAvA7Kqf(}BNG-I3f z_~SFicW!4F{3}Wem|uxw{)ZHt5|W7yo3rq@F!Pga$OV4zO}HDf8A>KaLd@;wuxcOx zW}AjV#y)@W?W)GCppz)RECLs7+f4d$YV=KPJZ;^431l)W;91cWFrBj*Z+-29Ty?@V zb0WBiFE-p!_XN&oyEu-s&cxr5e^F%rHQveSbDR%vDHQdZ;qKMV&^a6et1^5cCXgU` zeIw`EcL6JQHez#>C{5SBh8L0=Pk+h#06WXVD1_-Xw$ik?h}IWRQxj-$l8w2 zA4uT|Hp{j$VC+&eH|8x0z{D7TdLbB3)!IpP>(Dh2m>Ud(iT~ivp4T|ddnR4qHJ?J` z8sW;4R0Fr^nLb+#`#x&G)`BjM7qcBkoUg;&>G4ebX@t*JkK*G0 z8n9*iU)&S<5G!xnVnDqmXHad;o0{a#OR?!93B8H@%)rh3$SI7Al7AbAy7O>qNF_K) zTEeHssbG*22tDWiz~Dx86#5kij;6ODX?ZCu3=9E>W@$+Nl*W~b1>!BaT#Poog_}i7 z(QWc$<{Zt!VO@Jn@!E`cV%Foq2sex>ZNa{uT6EI+Jsw?Nhd+KiKpm--SX?WCnOA1R zTH9=Z*N4GHVu(9tq5+04BRQX0`$1Q%%rH|>Y3}xIC5Fak#n|2$j~5yiV1-IF?%!IB zHANp#+dKd*J{jSXfjE5Ft&N2?&de`(0M}Oe!^o~S@UykS8L$1=UZo7@n6=|a#US)^ zw8q>&!EkAH0Iocwj~=rEFxG4luD{`rbyY_&{eTI#=xhTBE)T;BgO#x2vpDK~*~qwl zQRw^oCsu11QFmI||G&dM6OVzOFQ43ewCT9(PrSV5G#B(r1u|rUK+E$4R>yTLBOL>Fd3W2-8KEod#-K=^3!Eu8KgttVJj3bIt^Xh z`C$7q8OHX@;(Ad_Y`>~c-0RsS@IZ<--hNWSn&juWSxngukn!-`xb_M^uX5fU2Y)6SPS z$a8lLxh+y4h5q{#yfumLFkjLHfirNq@E0;hHEBM2g*%UmG1prNw|7H4sZR2{Ck@dSkwKE4 z|Bw<@%i+#7K6cMwI}Q(bG_e|D9WrYQcSwQGdxD7PUcu17N1C|#HK~~0r%Pp8(3^ab z*Hb^1_p0*%&)#u09`;fvA>DjbdM-|*>Gm|{v;-MGJ%gz!hNv_(8}vRbfsDh&AYpfw zo*69UYh)Dj&x{YHb3+j%n{teP%UcYwZMI%rT(&81#2 zyj%|A>!5BcDK&5Kh8`&IB}SXy8V-#lk{eYdo^kh4!eslf%GTT5z7Y zbwBNKqP--orc;zWelsPtj-~md4=~m#7ne?rNBa$V+}jvk^nP+1M>ZAU;mDu(oOw$; z3XkEmPmNgAtU@C$%s1|T9aWxopxe7*@Qp0MeR@24eNh>IHRnK6u^xCF{X*{PTgYbKynnvV1G$B?4*a&mXbprsKVBrkrDN>8-o94F?A zU$q+Vd>DXQNCeM4Z@6!2S={stqQcP_{47xkhPKM&y55duD7KK^gwL!?JdD*{lc?>E zIp=Y#l=Hc|8^))5U`u5i-2T?e862O4_D30yuj&@d$%RstU=JB&X!1Ln&Cx@m4NCQs zxdojY$VYo3rPhDL^G>6bYI>9w@BD#tZWod3?o(8`KA8AALZt9=6i-i^PPu}5c+A2d zKM1nSURxrJ&M<@vGtYvT-YeXGFN`i{ex{<{MQnb^N0sV1cvav54l6TW&@>e?m(YNY z?~1r5yAZejSB;tuu_%r%Sok0aqgN{66axc@{d|Gz;W9yYgA8sP7l1F{-Nl#IpD@Sz zC-W_<(oJDE`p@S*G_{LEv(Iau?6&jt{=in^DbA;!ckyr_LjlEG*zSz&V3tZhM3YUE zNpDCVAFZ4R;rT&a+2a$q{oO9CZ(M=(fBCpFLRsQdc%`TTdmyY~P(YAtxQJcjsd9${}&3NnAcaf|Gbi&PaiynB2f&u&>Q_dzcUDh~h0 z0FxfLeq}oT`EUgS4#=a4l0C*e`T+}j4RDS|2BuxS%Vq#yA$jgJOnwnb2Jf@b&btuo z6!7zZ%TxZEY*_nF}H2Mbk620@^XHmNaygFn5{|3Tl0UFSGYh{DMg|EA11uM<2my zeJQX;xQ1&#<%bvk{D&WFZo#Q-PFUAd2m>7}xc|iCaNG9?bY2pP-`2L!Lx*7lX^|R{Xrs;Vo6-p)ZCEm38q$6o=$>Y`BP~uHU z{5|(+krDQ=EZ=+P=I$P3-!oqenpk2=3$-dSTuLAF)z*V+u?=V4kPmWtpJA5mUA}AQ zcYcP9f|0zI6g|?>A|Y|+ef=~YN``~DS4mPZ;1P;rs$H;x$2b|A9l3&SKO&$EPfd^eT&F zR>k6yu!k%QQ-QojVRTg=hZ5puAa45)>YC&s=7JAzbM#W&7r6q*Q!0kcvIoCK@8ETa z26B5pf?ll}R}xt|H~e%U*7a}4-Sgf+u5u@q$vnhI!s+;KtpYdI_!!5jFEdE*siz3Z9AnRel)= z=Kh6c)zeX7G#)mor^6Wck8oOHH*R`RjEe`EOChj~INn{7%qb+p(O#}Z<2+1zriZ!v z%h9$o2-gYDhx(g0=$zA2THpSOq(5z^5+CNgGJlJeej|9hRgD^gs_>&;2E8#nO19~< z$Sb-U%B;GfZrM9He|I)5GPWT%nVA%wGZ!0Bjfa3>Yr9Z2IH!5BtnLK z`!y+b-(LK>$^zd#i$b#@9eg7xk4q=)f?Yml_+?`lZY_C^O)NLq@JN^@b5lv~{u1iZ zTSpDcO(|)|4|x7?07eDnpd%uY`T`!2T%0_Aw(%&`!spP& zHWdrEC1D8bjhweC!O32kP{{88Zrdxlea>yv=bb~A9Z~cuU>Y70-+>RzJaEscpICES znrvBz^W_%{ES(sN3t|=Up7;kCy03%-f2QE2PeM5Fk1go;MRAk5WT{nCfJ)y_BK>*Y zgST+0`RAMx_1iN6UOx%NeevBpI`%@`h)Rk0ZH^367c#sag0R|#z zm~8KlPmLn+-{gy!_hc;T2ovePmZfR?v}n!@Rhn~b78(Su!bwF@_@F|C-i{d1%FW}5 z7oSQM%(eM%u{N7MrJiD3csW6d6AIU<63tXrD=+b0krFhc|}Z z3C8W>jg*Xfm&p}>w5{89Fu}tGKs%rX)vgby?d&Myv zU0Q&ruN2VbL}C7xy)*dxb=>fD{|S_3ZaSeCOGv0`4T)Z`A^Uq?SkC;n=D8Y}aVi|e zA8AnyFPI!Uudv=~H;w)nC9k}#2s`)Uswa%MRN~3ld8wSu>1^&!@hhsA)!?5hu;O3m z2GAfuiWHLn;uM#Sbh18xjJ@(n_T5!*Dp5s83m^17%jUgjI#DuhCY_1Q1Ie|^;cbjG zw{~V1XRlVkY3-@x%2`Hyub2z!qY7#~W|`)q&yauX49Ik*kY->uxl{#Ek-I%k&5Xu> zD=Kkh?@SykO6%1GXW$qqaNixN7p z=nJih7T_L)~_54RYLOl8}Qx#XXp2yGvWwJ16 z#BcU#_-dORuC|y?4aqF$nm2$^dm>TDUxh4mf~fOF2&7DB`9J$^EcUpA@lQYC)Fwr8 zyDv@pYV#@9ZwvMAsKcbj>GacL4Y@zd=8_DAA#D5r&qjCyo!k2z=Nnm|MY1t$zqb(L zL1ezy=UZcO|3H8PD+J^&JpwY{dgb#2sCew#j*Nzzb2D+ZXw;q~=R zXzQAWYO9Z+ZOJjvNmhfG0{ghkj(+&9RDpL|?Z{@#Y+5;O9yOi<^4`_MTvmzLYQGfq z{0z3Zv<;i^S>v#D=jIr~Poznw;PsYfHb4XHZ5Z7KW$ATT>acRMP zREd$s!HzAY-?5C;gLNqI&4nAQ<~xM^EGfIR1U%~Y{Kx>8*rT0Z#d!ifR6lr zNVaSqHYuT-7OMRvb5{X=d-*>oFx-y0%~x=C_E9w3k%brBcj1F)hlzpBXmEW5S?eFc z596odsf2u3FysMG)`fD)mj(?Vp1o(dwlIL7zYK6Z~@R};sbFc^c5 zXAYxeLkY{?9;3m18_D_yV9ZW&l>O|8hUt??Ogej%soi`wH^>65{m&fD^#=E+o zjRN9-Fv$HBzM5Eqr|LTJ##T3M3_OJmH|kL9Gjn)Uv|+$H#wT%#;IAq^$~QRe%da1{ z#KLP*IR08cEKe(C4xtis)M-H%cPYx~9L4tDleqlgL|URZiDDgf$lBcvPeg^{#l07> zYq%aq*Cyi4h@;pv!4JRGrl8WJCZvE)Jf72!CpNX?9oqGtBxs5ly*mQ7ovBd9D_9-@mD!@#y&n9dxQ3+$`$&TO`8U#bchYJ=f0(g%HuONH8VCP2L)%~aWH^Ca$XK%%$NfEGl`xlTJ z@W9uSK~(WIhra!)q)B$cct~s`Dn6FNOvTGAmtI9b{*}KZC)@&lZt`a2M|Me&sQ|q&bCB=7+r}N|Pp`5-v{@0|( zX=6O6syGW=H?HH)UU7t9HzSrm?LZvf+PDjYbMsJaM4j#*(IAgIEQmfMgxkOTD0fG! zo%3Kkl3T?RB#}0b);!3ACz;HjnYf+vV~&ySF~6{8lPLKoWTVKt$0(5f13egzW5h`n zyK<+Zi&Qu)o~ei76JMkCfn&HYoMk1=W7)hpjqRDmQ29SKyklApedEU9@PBD( z@cq|`QJCgJx>%4%LN8y!_3L*bzBUC6>LVHZsFG#hUZd5UyLhhoIi6T^71iTw;h;`4 zMEGRGoU&i2vS1A9>dqjE*eu?deRVwLo|nA-VFA)#)qyP=TCt3G8>Mo6(fYg>T1{iw z0ho(x8g}52&qmU_noSMEwY2E_HCo&u$Y%j3eqH1YG&<3P7ay%7`)>u9_w*-vxKE)m z0yl8j{vFHdY0&q#hv9nNdC-m?g6{=}^mY46%JW`JoeK-Fc+vnWu9v2%OHM=Wnf1`6 zx&pMFw!qL+MQ-~iHEyJZPj`G>SqJ?s4{ptIZA(l_q_NwzQ vJw=3zYtOvz> ziNJ62D>&E1d+GG>6VjTi!1s-jgqzo&ak7jZn6PFJU*o9;zh(SS+MQ7j|I!|S{wZbL zdjCI6v0seOeviSj9vjk*?ZCO(FJaYUC(03;K`w_i$p3IL%{6Tz(Z7YX=pjJyy$m?5 zTLshiETeI2fX246yTv09%xsdtj15AV_o9+aOT(z}+-scE;trEj-$Cwx6g%5EgRw_2 zUMaAr3;I6PcrJ%xPc&0*W+8@eGs5pI-ukjj)^J7ER4zio3S!q=@h?Xd^Jg^+7^(fR zpzc6@`gBr^^bQ)bv)&F+)?0;RC!J;-^9*bZW&F5IH|7PNijm2)u~lIy_c~LT8yghL zMc!_tcbOgJAe>KiLxc+tX)`9w6m;qz!r%@Ud^QPCdW!>jzdT3|2HBJv^@5s{%PCgn zH2u~60||fUqeXcOOpXwP5(9JSc-GDp+^Yrg)F6=XJ_i!RPKHaoZW`Xyo63dXtH<6~ zEZZb0O)5j`B)vj~j%QrP10t;`RwP3@^7^FR;*FCs=TJh%R{DJ^21Tq*Y05a;khnbgDh}04l=!4lFn#D3V+gYvyPrB06 zel@(M^%Z#YzJU0dyJ&Y?pS;I7)9z2PY(5-{6N_~a*Gu8ZjCcrM9s&{-nkeY^2ekeB zK)m4~>MhDZo6~Rb+zn^kHvKQM04+`N?qb=^YagXV0 zawu(UC)zX3i?$>i(?^3TH1p6z(3SK7jRZloS($^mU(+$=%UKNBA4ET==#%PKmUmk) zf@>b#$9~1D*l2Q?+n%@-4ByE^tkYk-qOW(qUD6HEXJAY@PgymW^l$6GwPp6>vF#G@6Zh?g= zH}OTmEsQTL$KC!N{2h{Ie5>^deCHM&Gz_`|qnov0%e<55HTgPTcwL3TE&H+ex;$Qd zc@~yD+kg3~4b}`*qHx{5-6E;sn)cU2x-oDfUOX;Nlf?nbT_{3iLZ*ii$7Y zQC>{PA1$D`W2Lz0Ll!z$XJS$adwv9R=~z}N-Kdv?A{i0*kpGP34&j`B=f3#$H%I- z@g3tNk32P@O|xuBxu^hR*bXvr6%Pj-_A{<{AP%y-^60Wpn7ut1kEpl7zURzY`B5D= ze-p*(R8ewil^_EVQQG7kjk6E!#+s20nEmw)%C|+Lf^!qB59$YzHWi$}PTG#!*jzL- z0i{CO_kP?T{5B{=*=gFeE{i!F9=gzQ4ckY2=p)Ii&qy}*Ch0B~Bx8r;s3>BM@6*z- zr-se?Ol;9}=mGaC(+C>G7UKBz=P*@g5;;7Gp{Et8hFf<1}R@G3Ffq)Zyg7r z2C|e9ww-p=9wVP59hx03Pe=cbrKZY2Jna90?J|b&W}i4`UfshjpUbitD`O4kd51tw z+8R_@5s1g<_TsO#vuV@!*Z9+}0RaZf_!nlDMW7<%XeKO;!v^0v@cpv^pYu8~%d81c1`c2Y%SRm%?%;BM$3u#O4)a;K z;k_gU+VoU|>`mjKvYx%4QdiQZ?Q?K^)-gOjxds(C+yG5QM_i&PM4J!y(BFxVDNo=E zB@UcHulVCwm2?L4#olvvcIKc5=4ia_1Fn5|94Ck*;KDb7ytdYTT!mvGTzVFXEhU3^ zVSXBUoE-wUNo(-I-T}O1a}J8w%+t@V6*q;qWAmUC@t6nOSbsdU>>P{sA&)S}SMRc=p6R+)7K8PiXVWr6) zIIk>=Ew7j1zZga8*7K!S!E{=@Km`|geh2?SIrQ6Agbz)1=;gCAR@>)uQdPn*Z~sEb zG5!J-)*6@}?1%qk1h8+F1ip{-!BVy9*e&6NOXE)Cw3Ze;pCG|9(BkyG@hq&>T8WbL z-(l_{O-^&@9vH$o40^v2J0^?Lm#QUH?o&cNQJu8YZz})uX({se*@%4_J)Cua99r5a zVbf$iwgX9_@2|2kMyZ^ec`pl_BC^q_{sKC}v7q z;{vMweVt^M-Jn68e^^v4!+!5NlrUuqDZiE?Q)gK^R#lI_mY>k+kT4~Wia~%%2K;*d z1U4_yA@Q1h^xE?*HFi3J=<0e{K2H#rnMT4n$FXSG?S=W-RUi~S4@>)2qSFa=xO?&t zCW=at=;&1tO}PuA$MOI)T2V%5E4EDi2tvmi;m5pe=wo5!x2v9^G|TXQylqZ7SEDiM zQz^*5+5$4XP<~313cqLCMSA5D0#_GG;nl7P+_O6cHZ(=T@9Ygw3zV^d(Q$rJdn zBMB}qWVwm3aG=|zU=bS#Usb0uuR;RL_SL{4o)PLYFW>2@`M9j_C$7)R#|~ows&Ub# zk18%uweBHQtQSHy&W8jaN8J2WkUp2j<8{;hIQ%-AIhq-J!!!VURMz5Lo5#4-%^zbc zY;euZKk)pj7>Gs<@g(<~Q&{6BDso*(Cqy>l_9x46n&|;N9_|f>YA4~x*Wd8gIRQtr z7=L0-JWif)8z&9FhYJf=!AK$Nq#R}Z&*ck|i@6M0Jv;F8iydglX5dG@R^un%7QCIQ zfiWj%W3ibLHdtQ<;T)Fle=-x^fBDTdidaFyhj~DDjDg$!396TDLfbYc`10Z!6g?8e zo-iVf`ztAG;XZn*!Z`V}1!04hGI#Cd8Az1S0*56d5cE+T>&#o=#jje}khlfMOY_j< znlZYim%)0^YACpN5WF|7p$MJ5SQ{S5J^U@oUw6QgJ{cdyu!DVQ+E#1E{e^B*DG*-{N0O1-h z=zL8bNw%J$?x8$5EoBX7&1<+7dneO>-wV;yc`3A&Z-tX@+6{Ng{hC|q|D8^ps31X) zy~N)YO#38QpF&H3-m~t@0bVMc?|shwSoIOqR-2#;>qd7AkMN%g|Ke90cJoJGZ9v;& zy$q-`9lX5E$RJvuC_{+W8bsl~G;NeD>4T^xV_7b|1Mlv=j{S|suy6DoZ)eZ4xep-( zx!cF_T-aXR;v|Xwt|c(ra1Q7hFQdNVw|K(u42n)Mquloi=uue-u3-wye-#8ry+kl= zPcxK+%V39(H1b8PkZWq$d7~+{0O?E_E5HTneYmU7^&b$#GKcp9Z_4>> zN?-T0Ztm1oY{vZ__3ifH(Yw`PTYe2L-^hl}F)tz2p7EOuUvcN{RB?XQ1CUPigmW{u zlIM$N^fiyf$*YB^MVT?WH!X({t^!+R&time3Mw1Q((6P^ik7t^@$3d#87If*9hk%K zNi3matwB=NmgBF`il;@4Rd_yXC7bt80G)pkyuD^!yvq*8%;O?UBVD5y^U)Tbz0|^A zq2F-+#Tnc`w-;QvTQw}zIgL3wi5RNejCEno`1V#9{+jHC!lqLwly%h$t;W)c)e~^+ z_+Io;GG^zWaO(S)MDx`Cqt81MC@LX_+A)&6S`ifI-2`d-V&oS~;iZ#^B}(zwx_T}@ zT66|KW%)#Ypv*=#1Ixj1PX#*i;1Alc9);h)U{lI8ZQSuW^0)y#= z;L!34lGJxm`N0{~SDQ^=d=TK*9N2A*u#i&O}OfEHfjiG;^VWzd*lq$W&_n3m1yxV603btXQ-XD%w+_`o}6v3=PET z9_47%eVn_lBf*7tj|HvH^H^xf=2V@m=eCMD)_1%CWwUNL*Kro?)&1c9kxuY0SxevA z^%(z>xdY!MavyF6gGCg3rk7W+jL|3LdA$x~>XoC1Xl47tA6Hr|y#7a32W6Vc-=^ZaH}SW-@GlB8FbK4&(E@9^CSNA}O6+j1gtbp*G_gb{A+@~b!{&H?r6#^cFmNt8D^3z44Xoc-o!@XWdw%=tzb zw$lMMe_p_jjr*`cO9Jn-Xksn9kDl>}LUD<1RDGpRQM$w688ZXpG@jsH#-*N}m4FiC z_A}o08QdPS4AUFLz|Q?7?qVEZt)QvsYq66H*!as(YjDJHkrHDFGbh_vp~)!L&lnZs zJ3w#mAF%Yy#B%16TpXy3t^=VE_f-U!-dTn>iun+hJdMkZd&+HZi@`ldA7DwNEPb36 z0i8orQ7?Qiwk#P#|3t;8OR5@sBMxD*tqB%=jmF^(QQU+pBbo8)g039p+j(5^CaLN&7FnJ~d9#;&ZzoeJ9d_yP) zQw&gd?mGAxZ3i$@7Mn~R;8cGsMD^U{&1X5yvnpa(BX}9MtYlo0b?$J~>?oJZ_9tqd zx>%Lz3+h3ez(#xzHxO3IHUIht^$S44xN!`J1&)J#teKyOvULg;SW^!X}W|z`fYCL*^ z_RnLOy36upl6DKNe<_jG1PgRhRRr+=g*vA#Y0t)3QoOtl5{G&?`#?Qz(&sBQVNV)) z$9b}|l?wW6=fI8GHz9UgI23jhAm*X+x(07%--DRH(kR~&MJvYJP>t6Ygl#u)^i~{>`Ng=-w1^Hb(WUy;!i3L5 zxhDzVdCg))Jb~+0)T!x8Yl4~C&pC+l+z*k`hC>v%G*s&-cdAhGo5|_;5aYEJ}n;H`=*4Egh5#R>3{J({bCc@zj4@ zh^`b=U}AhWXJYo6GY&k)O;njf^M^N)`l$$7bM76ATmHeMds^hY^CG&h?nKMBadddv zEfW94GH~y!$&43D9#zpazR#VeJ1OFnAZ>h6+z4_#dtpZBdC(k8hV>wXNkavoJEN9$ z)n@V*>hIz?$%ON4FdsJ!zJjtx<6z6CG33=FKtI_Y%`z{DbY7$rcd(L6;6Xbdof!jwnJ~wZ|~1DhQX{i)p`C5!EGBlaPxyeP-cq ztI{r#W4^Of?QUfL`Z(o@XJd*c^Ooef;>bDX>YA5NhF42z<0T_pi-4DchhWu}T%;ul z`0RN<`__+P9@uYCytW!7G48lXBllJ6 z0=M>t1=?qi;?lp%DA7fa){J5E{p2i?Z>c7mqbF&viVJO%y+C_DMbLl^Qt%5sxOI0v zezACs#|0#4J5-~bg(uc#q?195Cq+2g5~u$Qxy%MsX0DSx2M00e7h_pTXyLY^myl;O z9VQ1ImG7guPH=6l{? z%#q1-)8;739!P_~OGV&WDhJQx2S7vq7x7B$_!`+6{Q9`rboJs|+EnRAy2j1Cu{j&L zzL0XBUU&vxytfLK+1ztu@kA_na29x7aZspaz}UJ(_jOxn^spN>ZxW#; zMPKmoc^?c&Hl@OYCS+u^mcqUYF$jL(B-JT6q7WPMD&cyu2`t$m zM=vZ_;LuucxYke%ZsFEY`LzU&Y%S+4I2^#+u_KjRm}g6q&l=G4d^Pe>Rb<^}X?m&t zh`sAuaQn`K%n=%cy@P@@f0ZshlblV}I(?jqLL{zXXN95f3#rVTPoH-!C8=R=biEvj zR}vSYads$H3C7@Ye=$<;U5RC5_ri&|n~>Gg50|{Hu(fVChF1@P4BL+m%)J8P5|TLe z_#iGDvZF-5pHQe)$?|Rh8#b3yOl1WL8=N6;Qz=~9tc*nsGf_Y*9-B z>w;78n4J>xctx;pychW8Si&r++t7K;0>0;lLij!poO6)vhClruW#=8w)%*W{BP%kJ zNEuPeig=xC9hD+Ng_4wpwg#G_D0{CELLzxfQ;OF)*Gth(R?<|77VS-x`kv3z#D?|K}iw$G$x=?9dYP_S3DYd zZV_ZZ3og=E4x*3qqEPqav1qw<6uff!Fbv_$;HB~u=F##Pcwem=hPRDG+m>E|_6l#H z*U4$k4A&Oaqz?!J;9Ua0`3ONllaLpWaEL?*rUfB?Hw~h$lypJ zSI-4-a+)eSqFRdu=yHG%Elb>T?A5;%U` zE6|xI2X4PHhH~op)s%G(0837DDSiqe%w?J&Mt3* zNsCh9=*E*U`t}9b=z9$2l?pSax(+D4au?Ei8;h!csG@(X&C#$iE+|J=6%9SsMaw)e zvba|W%Pwq!4o)dhet10`sttpiD@$O_Mj3GLdl4g*vkc5BJO`DYeS;gu}ygZ)Y?;bLhHo(Wz9=Y7qD#jBQ~m6l#`g3)sr zW1@{ts>CDy)@md#jL*Ni*$?x6Ya{Vf12j+_EDA1BX6#ixfpnb}oL|BU@2yFYZFYtm z3Uc7GrGKE!bxWb1sDuLCWst1lW1u!yU-UxVNp!vDgQ)zOBJi+^gKKKq(4D+`RA>7i z`cjk+@?O}$aqog*WnmopHY^bZzl}ue{S&~$95*=pNCxB{zF{7RxC85w(;(ET(so41 zeSwEJ6Q&E-{p64J!1Yea{%bWi}{IU|o!PbdbrFH4qifeJav8cTH5gBL|hWrlAnEG_=LU78M^c zLAu7KXm`+S5O!%S95@jH*S&IwPhJXV)r14^{9`Gye!U{OmNkU8zWE35KA#5@2QP!E zqI|gMpTI@X3xun4lOY|P1sjP4Ok4gNY@NLanC#sECcU>6&8MBBC&&6l9-DWg?v0Dk zL6IC<^-C7ccnF~S1P@4xDj^fJ6Z+LJhK3ba;aHbMn0|^CIB36t>8xd7d~PKuf9D65 z9heDY)oeY#I0uD39o8K7a5B`|FRu+d1X%&bC^(t#$;+p8f{JR4BqAs{}M( z?l@X?>MVk4r_lFSX}qFR18+?-LaC{?C~&eRihaMFX$UH0Ce_^n53;M7*C)q8_iySb z!OI5)Wo$r|Yr;^qyb+9hu^JZVe}I=K)WE$5CO~P;GLZ5G!~UrqFzJ@E;K-9g)jo${ zUGZkPCZhwsU-}TrcQnJV+up!;MRss^%{NAI|1lKt$OTQh9xL?P4x!f4WYm`yir#&= z1`3Ah!_52+IKr+Fi37XP3_D|d-!K&x4DN(c!|w?>g)V{NGzRWECY?~i;lV@M@8IV)TWDma1ApaZfDR2aXtHQ5 zm@mBtY}1H`mZb%7+?^D-t3CogKT`=FRoFq9(iEtV%+ZluYtW`eg=pmxN4WU=arne$ z3@WMF4OcB~gylVxQ2E%s=%czNVlTacFSSz9B9A(>R}kCAvdA3Y9%xk`eRa6 zO32C90EQiuM?=#SQB(MR)L-lhSJxQ|`*a?bY&?Ua_Z&lcvRTOg`Uo_XZzynutkIr6 zT{IidL7d5A)U2Wl?p$;g{Z)S>(ka(LirQzOWVZqO^Pk{Cn;{KX_Iv_nO?K$ddluP$ zRzn8wv{0YseYpEvEX+n8Fnm@IoIHCN!spz^aOk-)n&F@d<#Y|9#*K5(=jtLb@`oH4 z(h6geG(zFl-YU4RjzL|wn^EyyA3QSkAg-co0kXftte6_YcnNj;QR}|LO*-FUL$e}I z%^ij1SIc8R2VZ2gZ6$iKdNW#4_7*N&q=`ar%s~IvsiN-uEOc~bpO6V235>jo!G9-f z!PQg_u5k854_DPA<2S3|j_n6wZ^B{dBh&+(56YmVWeUhbuY&2CaTDzR^&SRg#6#yD z=i&Wx{cwh11~hwE2u(`ep{RugZ`G2S_8SSJ)c0!W`&2V@*~k;+lv$u2f1z$Nwgm-< z1)g}m8VqPW58|!wA{p%zl;aS9#_NSK+w|Xny`Is~j@=DYE}nwHv%BEOo=WD*>~9QN zC}E0p*TG@gHL&=y3`$OC;XWaQZdou1$}Y_W>Z!Ky(EfY~538Y1mS^F|=@Vf_dKuh5 zqZXo{4}e9p1FVl52B*{}0{iDz7-x%_AYaI!t7)o2yTHRRZkQ69I@1Fxx0^umL@k)G zY!#}MOGIzm%gNmpWkL9Y@# zJ1+%3`m_WVde%VctakX+QIq*<)*-4LJ_*=%pMc|>PC>i)lpwYI|w|dTm`qE$D-?UhtPrr6)4o!ggI}bCo)*CYxLV8Xc~r#6`@CixtkM`$sX{O~Z514~X9v)cSb@qB0nF#i z&l%kSKjz<|eDKdk0a$JxA@K1D8t2l>cqRHUy3ef{{}yMau5UW57Vgy-0e9iOj0dpl zR~yva>c~tSJPWRlN`cp3SBYZJ`+)ab?!Y_^Z>A?^2y9b*2=|7kh;Cet6^#${7C3yC z0N(k6o+iP2zGG37%5)~A zUy*roJ(c;Kwv4&eaFF>bcw$@*PXw}a?=$*^@ys~)os80CT~IgO7e;M)D*EYv7%UBv zg#jB*0J{-RP{scztXMeAR?okFDsg>bQ&G^!Jm5crSKbo1{hl4r+j}3J4_plIzpVgE z2IE1i!ao7qT+38zo(ElH&%kfD*D_n+DCUBvF0;y7O>}+#a-j4)9R4_^0G_}`Mrg$e z{Fo@P;7KRAcETM7o;3ygQ%jiG)5n?B_!48fD_!LC{e&oMmNIj%bUG7wD}cGQWHU%^ zo&@Bb4Vk8VT_!;0sc2-<9p?SecIIpv1FL@QKppKbMNd0&Md?etL^pFvzV zLlF#GagjN^ZX7e_{yEXw!THoHP0ozCQt{g-60`;RNP&dH^$-dtmm`iSXE78F1sN z7PucC3gk;)LX%NiD6(q?;)L0xOr7)4L}e~?l@Avsrt~sZpOj!?hyzq{s|Wo>F3h;| zThWD@Y?P0n`&z{_de+P8ns}tq$*EhjGP_72&xxZlcnMHsp zZ8C7Y#1Qj1>JSenVWhyVW9eAl;?B=1#epq{*IS`~ zLZ}POXc>uCy*&i(PP71j_l1M*%)Y7N*?UCun;e

    nFmK(n+x1?hUjJD}k-~Ny0wf z57I5=wgV{IHtWK1+r&@FKx?umJTE^U#fY=P^IHSp$KTsvvEMIPODMcpxe_k1bOVlG zBEUU03D7`pV8v~rW~KO0$mFg9{@SgAgF97}ccx2pT`z!%ORxeTb@G9F>m2yy?`G(~ zE(pedZUWtn!@z@jYZx@Y45o|JP_wf?NXycO(o-|wKH+yepA9iG_OjscKpUeqM-eD5 z-^{E{)`Tr5D}asNFJ?^fY}+Ln^)_2xj~1QTm4se%(+ zE^imDmC<0144MMlzH-p^922*r7^=Fu1sUGW6;ZTtrV?L0yOi z9JQ+g<`5xYWUhc_&5MBw^`Bs-y&ei^osQC~Yk;w7J}_GK8w?g)hT6jUT5kUYG-b{w z)Th>nreBXmYNOUOi_6{NsY8OJI_omz#y^D>(Pc2>?0>M#dOuWc6zV`G6G7753USgi7`@*d6+2u8k^tX*A?r1ow=>*j6)NseW3prHxI$kLjvLQ69O~*R55V+W&t{n zr-Gud8c@Y{KfI?^ExagZf!GCd@X}6qcwcbTsc+c`og53{+F|kVwJ_73X>u9l%$^9h z2A_dZM#tbf?;g>6Ssn0Y?;7x~MopLjR7cYn9E9EvzQX}64vx!?hOhFbA`KgX-Ly$y zpU2okExkSPW|b>!H7NlV%aefrw|d~Wj6uLV6irGFKtV?1P}#J6_^w{aVc0(f>OS2} z;FW_Qby*?UzGV*F?d}MlB~F3gTtf$iiBkK}NdmH&EOW443%c zhyN|-pqKa;EO=f3r`)&;@7_p(uU1p$%!1*}h_>}0`J4}s>M#b24I_cdW*Jz%dpa1h zm=4hBMbPP+HE?zI1rN?#WlX*jc>j7ps-GNxhTOazq-N8(c56x zzvD1CCKoOj96IO0YoMwA3Y=T&2lE@&LK*2=7&|Ky0fAu5k43jj4wfX)aICnP_F8yx=ygcGOa9(Ubjf9sL<^rdedN6hAURbcP4=nkV1RABTGtUo&fF$Q)AhExTM(NcE zS=15~ZeGOXDH{njYgyo5X@PppyCCi@gJ|qAXpxi(N3QmPcjXJ1B!S_!>G6Jn>BWd9 zI7Ns8`{F>|$Nh+zWrmpfPvE?1x$q~EgJrVEfYs$MaP4(Dl+*VFT4iN|qW@f>)SIg? zolC;@+`3vrO%)z2B&hVHd7sh2uq5Z3-f-81%@Irb#JQ8dL zGTr3B(xD09RC+49U|1xaWTVjLgfzJE`5{4!lr~Mwc!nq4cV?LN2dVwC`7r==pnpFhaSTIVp3QS#0=+neaseo6H|X(H|Zl zDTAj--|+*IJ9Zr%%~gVX8Y95vn~GpEO#^QS7%)+175Emq5J~@xKurtJqL`%PU{}CV z5Z-GHRHuy-H3=Gc8_RFt;-{6OPZ{!}Vr6~CcidU9otq2yFTD(F`bL1D;Y#3i+zC)t z^${#wd=B_b*a}jn9YV7@=Aea#hlG1q3Xfj?6Wu=7hN3>FAftH=h|_w8K1fPXl)z9Z znV*hSWM;!4do1{iB%FFko3)ip1!Hd<*-iwQv zhoNzRxf2OXlOv$ah~1!RoxmP*{0wg_TL?{ZD&S{zS2Q8x2^h7;m$`Q5Cc|ec!dV@% z$a=^b+?Cq~L;eduUkZ)Ux3a^)%o;Ow()-YNM{{&F;5OVjK?T>S8sS_eKP-tJ2ls^< zL)Sgc@caj9)ICoTd0$OLg*VHX2Sq{#(CIT|cdFU4WwuP}Ivem~bu{zIS`H@Ox(Acz z7mDUI$uU1WjG)D2E%avaHQ1~7gE@7wL}c~DN7NH~hKZ~m!<;=cTI3s~45Mxc8RwI+ z@WL%^^fvY!{9e@oPR9~3(XE~-3KW5B>x^L z$4b|t@w=~D*umoplh-qe(TW!{(ca#0;mUL-f0wSnh>}J{7pKGBPv?UvK?h-X_*{5q ze+Zm#y-D;|m>J6dwHH2gQ-RAYFM+$tHc;AM4t=kVL=9WdqrpKN$H0;TaF!b_%S_zcF~#{p@{K3nPHZ;3(l9E@g(A zc@Py92HB=qbXrLo7a0<05#s|=^Bh25gAeQyrY3^r&HyM51Jjq!gq>Q#daL>dm{^&^ zh=dt%-1r(;$ION~oulApMMeB_lO0NRXobF$<8aLITWCe|KR7;aG!p9?!JShdgKx3{ zaPt)(r086QJlqr%7s>LCQijq}|NHT9lN}}{rKl+N|9r+uL8(>S*Z6M?UT3v-)y`E` zSR~*Xp{>nS8;C^KPP1&Kmj3^Lid+o&suDNu$-`2*4u|m5kJNFs8w|*yNDKa2XCM5u z<2bfh`v-yzac5Ncf%m^r)z($K+j?{Uj;9{FZW>0?!|L(%#CLd(%n{yR za)F!tIE;bqCbo1$B=k-SHMp9EUz@7)2ZRiv%qBp0T--pb9QW(%a{qxZCF z`1lq&bXu0qtjMBKkFW6#*H_cYDv@;K?Q+^1)4=;}u;S02VEA{>9Qh>YbK*7AzM!u3 zbks8Rk#p@DiN}y%cY^J|<`@RX?4$02ao&9I1 zS8oJ5l~pgCXB(l_-X1g@Q?f3~m2@w=g$oQz@%R=;vU}$Q`olMguJj#6Wec8iH**CR z%He}#oqrHJGJ6->xiXR*`Jzuf!Y4^CRaie&%LfSqmh}%=whkU6inGllR7@Y*B)XN_;)fKE&Bn^`q{(J z+7pg_ca5SKQ7=D#hYvUV&oVy##SA=q#%i)9=s2+{F2&h?ru?7R8WcUwBXU-giR|oD zY6i6=8J?PwzYZfLE0ac%H}kI&Z#7-^?CM}T>_j;(NRKDeH*6#FvjPYw-H+^H2|q1H zg9wdc@@BA**d>RM!U-j0!xDY+`{^3e?z)ktJJq2MZC?_h=0`sL{7tx(47>2?7jh*6 z^4r^zdAa{;`FYB>1OeAr+PJ|Jhjy&R-j#Yp_1tm1vE>V|^dk%}2~Xi3TO`nddG7q} zRsZp^=peuF$}aM-WHsRfui?4N2k`fkS|qdd6La2^=XVzzB#{HRxsT38m_KuvOba_s z#?8qh$)#sW*3)`=c|$X`FxW;f_RG*k-T~shdw9Iq<{X|iW-CeS_)gAzU(XqQYC(2a z2L$D-ESo>Jl|0X}qm6S9;DEaxBwS`9yZwzm5yd6o&E3=J@!CaP!i-~B+4~Cg4gSop zUH<^zTi`_-O%Qvb2C}o7e-f2S4L;{N&n;C|60h4_NS++kr4IE$G>$x{Pd6W?;$CU8 zPkb7eB1Lk;%7y5ro+fUhJ^1&tXLQ*71XAAHiFagu!3+L0@zK@);XqqYGV{a?_O&FM z{BsDQ%ApUqK&(o{0SRRP^0~bC0Z-z7_!1GLbNH6aZ`|URK+4kY^6f5P@a2O#?1`-g zBw|c4ueN3xEdFi-o{Sp z$>i>flhh?tn^vj(6c6-@@#d`qL^OFmYgb;&-&<%wPR>6^cK^Bv@7NwfAx({FrCS#1 z+?7MJv&1B>?J+K0_yn(v`-j^esd0BKmhsL{49VJ@GV){jEqastNfq~<;5g~yd@d0O0$<@b%C6My)kO<&^3-i?)b@w;5IG&+i0{(6wE7@J5mhmKNG%<}E!YjD)t zx7fI_6=}Cq=Z5AAda?|uxQ zNy9SkR=m#HIcQs6H6LbPMVC6Cqz}WslZkC!q}9fVmzk4`-xMX1_NXJoBDji9&yl0+ zU!AJC^fHuMgiYb*(;HZ2%O$$~zK-NzWHfOc^Mt>0GLJvC?jN@~BaqfAzohBemifu?% z&Roe@0{;z9d=HS;QIy?o+>EMl3tTRihd zJ(Hr6BToD{Tw-f&Lf_B+O}TBw*riK{KDm-iM%!=W;yu$*Peu$jzGce)cqzyIzTAxx z(n{zsnHt*mYZps>Z_)H+B5b?Bgm}tNmRJX^m;Bh$Pn*7sqYX@;ETm3<_sT|aNh%%a_VpPO-_D10 zyUr(U=%PnI502*N9s0z-8fi$yD2&PF}0MKUaulodXur`;Wg;_|1#XO&vR!BZv zT_RB)dP&zzRiM4y)zr}6hb&0(V$najw%1v3qna|8~wP7n-cM6-?99E@|v{x${8-yNF9fqdq7=pQ)>0kob+&NlIBh$I@>gm z|J;iCx*A8~qWhA6J)h9k&XgV+`G-E%_mDiu{6V#cR3trT6X=+=>cr2N5Pw5ID&Mz; zn7)`q1{;o%oQd_=SN$=E3;t2_R$XeHp~ugCDn)x{*wL!dGW6vsAPH~*^!}n&`di>p zSSPpQgRAp+mD$Bqt6og~Za<^%hs$$iO7ZBWayfr2?>#Rkt;4tP^x_?g!bzlQv~Z5n zCr9p$rJW{MG5lo%JRgQ5zhMqUt5=_H9=JglKdRtQEX?HfJ9Y_aXK%@;oYS<|;RCh) ze4aj&bP9VSBRUv)85NhR(T4s!RK{Hp@3fr7RYa+<|2}#Umjy#O0>9-S8;ki&=``}A zKwiS#yH7o`(rNX#v6S=_(IY^W&X)D(-QN9yZw6|`G6i1LV{R2Ly_bd4ZhYi?@=Q@# zhcCDK#ChI!FpO3f$Vv`7IrDmvUc~XHh`$_I%s)GljSo!^AfW08e{;<@en;gn@nqK) z$$+tegf)(m)XQArEslQUO^f@u%1&xk*+}G~qp9+bOd2G& zm1>=O&b<11oNGpvXsu5>S-i%Jobi(()05-y`$|i)#UP!mESoFIZ~I15I$~+)_YYL# z;X?X;atr-qVZk3_9BV$U*@=4-D*0HCd3e&tjePOZU9@!0J>Ku{aauOJke=RGz<*iw z55GDYNFSU}=cT-^^1~c&Q;V$YR8du0;x#6lbR3gq(_%`=s+uNbTJn#q>sF@j9WHcf zUopqeT|omXS5pHuTRdW^4awN?2FEWs#@mF{iT^GLp@j#}b0aPsMseSM@t)_MvCJ|f zY7j7%i_DwFd(QjJ_m6F*#N#v#6j{?SFhC{!U#XUVH%&R0z_&iSih8qKaKGYinm;B{ ztUSq!4j13#KTUSyC#KA!I7wDg(p^R?F6^WqKbz8)??>si=g;Yvz$x5_x{17(%PoHY zpL~2|iUHldVuSd^xeweuzhQj)lpHvVSEaY(6o_l|SAKz|vE*&Qb{d@n=;-FfwE1ro zU3J)ujy^ITZF<+iOy5@!MP~GxiE?RenuZB^FYHucbAi&z9k{=da)q zk4o?g)|Ef7^*C3RyoJWy%Aww;4SBgst~`{>=10D;=CSz#TDY`_pQC<+Z#iZuF|$}o z&z{yJZ?>-^&Ua3b!87N%Irp^a)(_6K6HlQv9(r`sz4tsj$B)JtB}i68te4bDZIi6u zJ|M7PRq56RiTq(+5Mmvz6W@8_jU6UFB3t+UMM8rQg0>_+ zGWN;G7VX@i1z-JALH`g7RtbTqs&aTGo=>LUG-nMhCFY@;Uq`jW3>ZquOD=QMauDaD#k>9A$J zv{knNyE>0Vl^P59YF`EJ-dj5y=v~RLm~e_l%zI6%1|QLThd6HRWQ_ZFUf@#;Yx&P@ zb)fs$PhREr9{SJbFYmEjn?&2ZcneU_Hv|6S&2$59nJki4E6 zJ+PqqHileYNjhm*5<=G)N%)t}+2Y>WC1`qH8OP<=^MSp4=y{7{)Tv-1{$;06AD#Wg zPgj14+BP-f`gQueRaF^po_m_^n4nJoWHj+QM_y3RhC1r8Dvq8lkD=$&gfmjp8Tuj0 zg(|MN$;)~QOu$(h1e~U1%}RYzIZA^qy8Va5samt_fhi%;|lrc7EOxRLgBR2V&COe`T%RBU`I#tCR_LPwEc@5-}1)-l+)ae7IF*H(LkH#0D$1h(?lMgLRiF4X`lK85W zkGpn&20hQ=lDziPE@yvgyy_TPV11KFe-6W`E|<9foUM7Ox${uU)-*he-A;bieI_=C z-q#p;$CJ6OEMGpgj;!pdAZJueh|{(~oO*UN8L%|wcP&dHc6J{yRsF`R!F6Q)m@cw$ z!$dw!rJ7y8sF~HQ`OS{ke1aQ73y45cCU%_;WTD+^!q_yUOx%Z?uIQ3qC!5H|-*#l& zGkv1$XHKT9pM&lbE&ls>SFS$f6mAd7rN^%9qraBwu@&M=;>3*0yxx5mGV^RCx%on! zeRXGf!P9WfzLcs1TBuKZ2{BS%^a7r$JWJrss@KOBum}`7j zMZCD?unStZ(i+LWT|rFLvx&N58*if;g|`d@k*ZY_$*f!daLFXfAKGZkSx0}ui5(i` z!~|h~5hceq#>%nEe?_cc;R{aIGlO3n+>fhzI*7tQB{FaQXZ&QwA^bROAs?<-hR+Q~ z5l!tNBIPEq4fyyowK7JpkbcBgHBoe{s_^Jc;;D zB0o~+5V`;F5&x!i8^0={0s3cV^4iO%pkq4QSkpWVKPoT_b%ogSNv(GH zu=xf2+c^ub>@DJJMqc8cC!In&M*30A$VdX(lPLV`jm$<3@m_vEkw#ddWcZ^ziF;C^ z>MK5 zG4zoBG>N0PuL;*Opo=c1(h;i-`D`*geQCV=pa!pVxsc}Zzrnw@^N0jVsKbv# zlyS_VUhZRSSFA9rUDVzpS&}C$ksawq?@y`Yr9N0vt!ZPqgm-I5%#%o}bm%JkvwX&FXFAbh0v%XogML37PEsxz z&?iqH)4GktH1vrE9eU=@y9F9cdi-Bg?d3hRPE?4GXWru%a58-E^ZRt}`(o~G;T^uz z{5K!^RhMVaR#PzdBxb^Q;a@cl$WQog>hfCpZfFi!cIG$@SRzlq&a}Yg`xi=bjb}+* zh>7ImLIcUT5?lHu_%RL+w3bZWYa%Hi_L8uv-JA}iftyTs*VHZ9!r4ph;m`cg;<_x? zOGX7&NODi*N+ue6NfI+|N-C4DNuC{_Dbbw2nJ%{XfoqO<(?w~XD5l<%_fjmPx|L7q z)S}aLxMLKjt-OfuI$p^2OUgL6(RKJrZW4K95k(HZizn}E@`>`C@odZE0Ya)SlA*s! z-9D?fGLPW&h=oofnFV8+BY;Jc&8vxYgJ$=wDI5)>gI z>^#fh=Ry(R?_4*q6c{ntOnKcw3RRBKN?HI$ykQnhd*f*Cyh3?F=^^pBDdJz6@9V zxQcO^Ir$c@PyQ)a;F5iX+>V3lEa`qrsy%(lD$lcIYY+&U(eE(1$8MaTCG{oehfANuG zX`k)lf)^CK#pdH4^$mEi65yY2Bl!0XNg(UP5#C5mg^ZK4#yYV9c-!|VJZ!WB-*Dw6 zezBv3*Us7}Zuj$GjZQkTezDqYR#X>$Sa6!eDC7~Rk}r6Z`FqSBO~D6R({bGy6Kt%e zjkaeKblSIq|NeUlzfMzxO18eSvdz9%^LoP}fgpY!5BD8Lq@~o!pNZp$73WT}zBKZ4 zXUpQ%3)Yjj(RsuoeGw~CN?^msUK4jLN+8dkh}evpH1;bjA{o2l;f^I4#FGC(6s0c^ z=ig^>jhhac-<(T2KPdBaGmc`H(Qk;ympQDP>Pj{}pqQQapE6oK|0*h!xrkRgPsJI# zU<7*-yO%t1UvHTPa{yv z-X!9^Er67tn@HZe&momk3Hb8WPJF`&Vbc{SNhF_5lDAiLpN~Dq^MC7;#LX5ct?(J% zFioBaP8H&%aF1UWU4o4}Ca|P@6np7d3%U3G2hMNZLLADkk{GifzPekD)+E_Wl*{x{ z-}Y>N@$2_=*8x4MUmbuePR=5)u6-j1yMoE2zvET6HBy#&t+;!iK)m`Swu2veuZrTmPINzo#jkNhhZH`t+O_P}N7x-xzOK^Sdagtak%gVl8L_f+L zplj0?;df$|lleXf=bhLlFl3u(qj@9!^=ysgRC^mP{n!V+>NeAqzCNCBc!ode%96Lb zOGs6hm`46ll7z=*NDS`1khB{q)piDtlnirhrROi1@S|U?!@7?o#8&AF4k?(y8%KxX z{te-DYQQ8m!t4_9&^<=nq~l@q%ZD`s2NGe&(ss-p$w7Z_PABi(#)|vzRx+OtFUIrY zw)2kk3^`bJfIQdn$1`&F;OO-mxphfV^g*jM+gN16{<H`xtaHJZ^%bUsG_snO9d0eF-qFEA6gZngWj5p@=2lyZCI=->U zSh8}#7D;f!G>PSmFWANM8uIcv4JWNRFaGHFuV$#om-DlHgA4rEk{17W_}Y!fXwpqB z8gxZlGQ9peuD)eRs#Oa(jZf3b=G7bMtFePLCeofY6B)A;Wz|`ou2V!RwwsJx5XB$m zLvh^ETw0>yD*-WWI87#j(COJE-o1dAR{oA}Z=6BCevpyiRp04WYh%gM#%$`ZBx2cT z8`$J!+T{AODa3ww4xV0UOrME$_`RmX@VM8O6i-`ETiYhkQGv)}r=wd$CKn!2MSdUa(|Kii47@j%r%Fk-^AeyWcJIyJH z@JAPlQw;-1(X9=%ut}f3Y*yiXcj}X_hc`&`4P~;NxRDYg7ji)0r|*yK=lynGWKSav#D&fA;aOQw?BCu_@fr^p!t4)q@(!ZlJSn3uytLZ&*B2 zmu%{J#GAcrL!SqDd`e>{o!=-+e*`;X>B{${(_kb?e4WZ0r~l%*Qs45}9q~gB@%+Ix zk^Jbo$DGv}b4f^*nq)#;HmRu0q0SMRG~A<;u1wbBE4r8A+QJ%q@NNb-@~WB8J2zsb zZ|7i7lRi9=t--#!5#-35cyv;JIzO|$od0d_!)rCl;Cp%|)a$1Nn{Js-%snz|2Jg6W z(ZVz5d+`@mo2<&FzDi_Ef7HBf>Ff=oZEUn|INo>av3L#Bz|$wbaO=1X^wC0|JX>r}`xZ{24STidE+d{eRNo;- z6;6j8)F6lQ*)5Q0DX9L^CWInLa%a|IVmIcXQ0JeR?C7$%(~3FEwz9?M?V) zm>%&CoJN`#IFZo0b7YBOI~i?vfIQ$z`8git=&NT7N`CA_xXPL2!0tlw+jkEUbPv3) zbEq(Cr-F`B6_Q)6K-?#V5l$wFWb^6d>|hRg(pZN7`Om;pz6h=^krDr@_9CgeHjn*r zuaHc5l1(B zOIAm0!Y`NKKoh3z=C`LS^X|u_=<8NDKCNdb=1(Z$<&FAyrcevNG$I_QSc_{ydN-oC zT}#lkg9_XPNhPk?YevkqAUQo(xR)Dr*fO(Hc7;9Book*GEFmx*=}lb$m8h8vm5)!j7TdoTID>@9eMwH})yj zlt(DyjA>bXv+QQvAF>|@?wLWVS9=k+?~~XOIuP7qm!ZYSJV}YtMG{=BFaGR%j#nVr zWO3F-VzeloT=p#>x>5^J>Xo0o_TJ}2Y&4D?|H7PYnP5aT&Xf_iQ^V-3`yX*(^e9^E zTZ`W;Fr(#bhtaB?sr<1-GuFF)B-`{zpLHl6Mi0JM<&9j+sQj1x{-h7c zh&?l?W1uBbwY6Y}52~`O&|Z?vKgDZG_mcM&Hf+6qgs@)tvL7le*_NnQdZgkV4q7&o zOuXYv%qKm?d-U9itnX#AqIea4F3}@Fzn)TSBaGeW1(OAN8tjtON~}RbFF~!+qfLN~s~rn8O^zxhvys?ZyLwPp1q zrBo8|v5>f47yTba=i!jk_s8+}($F575=|oY+;eV8Nl8Ql*Z*SA{F$ zM^#&^D%U~L&SoV_-a ze3zXjS7(}0^ICEU^>ii2$eKXWt~!#L`-nI9sz%?ho8k4FMw##!>90}SJ!Z?<< z+1<1<4$iMQW6}5}iEnXOcsQ^to?a(H8Vadx$bIp9)>b`+Wg@@yk1^|@#h>TgI;aq@%uQe zQ!P)Y1xVr|U3=#GR3DnXV1V2bI?udFzJ&T}tI)=RXy)TkEhGPa7M%B~=A^Fa!xy(( z(!V|XnVw8x(m%Qt^?u9c91V-5mV0E%!Rf+*3gcz1^YprU)AmvxM5M ztBh_>{I)nBb%AU@Wl36F^dRW*UyCIR`#DFwyEvEb)G&I3uc)K7*4TNkDQwn~h3zh{ z!6fxOw|AE{GfUS5g4MOT`Be!_YxzbBU6Y3?PgSBU-hua>7sb!I?0{{*iR|J-SmP)~ z@YD99s{Zg}Gocp5NA4i5brNX|{my(^)J4VWo(&xvPY5_&nZg=j9YM_H=&uye5} zdq40I1Vnd$yUk`+J>(<|ty==hzLC(7CeDs7HDoUuZ(={VN>Zi61Nijb`E1-nS(2X0 zQb)&A&_~${6s=~*MCiRFA3K(Vt<6eMO{v0{OGNQm;VS;R8ur_d02t$TIj-@zIp~0hx&k^5{DUP zA|0LxjH-({1>J1olg^Begl??e|xl_`Vvn=h!2Q=B9E4%Z95%*Tci#^P!W8oP8J`zM`%SsvfeTDPx6xs?x;`YwT{ z!$sr}=SY41T?g}}_{o*aULbPUh^G>jnE9fCZ0_K# zt`}K=p2kJy=Mx3?dUy+?(2$8DMWSF6>E>*Dv4A`(G=!J%jD#%_XFr|Z&E9xy5Bf!B zpdYJ&`gL3(aWawomt~F=#|xp4cNdv5Pz0-`LrlZ?Zz}0wIi>%{gnjAzl$_Xw@qaFz zl)!gDSufS`06!q^A-p8YuEtORc#8CbW?zJLn+X3U8-e$Q3) z?xTY4N-(zhZs7N_f|Ocmfdx~HK9($nVg9=yXB>{^_i4iG2T!1^fiZ5#Lp)9xYCR45U50A zupf<{W5}*Y(qwhydLqB&Fu3kH!{`}PDA0<5C$`4yw^`TV*_&>(+NhX>Y&6CnL)Nmh z3LN09`bX5EydIi={Q!qCWA^VIb9AkL28=9MW-MoaB2#w{kmb{h;B<>L`Qkp0lu52Y zpI@gicW$`B)nBzT3U&v%lGsvbUK>B76rswTP^278uV>Bz1jK6O}o5Kby?u>k5 z3XES+Z36}z?Qjq58U7ec^oL`&7hB-bYdvUf)W-`Kl0Zx*!UlZ|@|>)q|<{^GfV==Nwi!?*UY?3t6kzrO>%PpI9sxg9Ps- zq;&TVyvwf|{ssipEvOksLhGNi_t(8)_Y2mrJKn$L@pvE5XU|pC=Xzz-d65hGKl7NV zxMQ&V^)D3MT8YOh@4{afOX#kY?(%TLPtsX#{%ZS zPhosaa3QPxQ;)qVeV>fDMMHW#g3d4gRP!4l{KY0--e)BjR)o><`7`G=se zQ6K+J?8eV$cmbCU1AY5&Si{TiW*v2h`Q_wGs6($ma%IJH{y~<$IknI1CvmLVO$44uvb9U@K#Nl^aol5fv+v8(#y0 zld`Zt^Cm;iB~hiF_T=3ZA5q?Af-=rtWaM;H@%tO$%#~#uIPZC0_lqU6Z0FH$5Rq6+ zN-Gn{JHm(3c~T>*P$5>fbvt_Eznk2i8-_QvFDF{TEu_{ro9HAoXI;aZuz#dkly|Y z)4mx$X^)z5`r!L1`n<>qFmmW6XXM4PV&Ei{Exih59~l_A9Yv1+lOTzYnvt>4FfA@S z!_tyDMa%jsBbRzbk`uO+Vv6Uoi9>p9%E>w4v>=XNV)22lk$X&&nH>D)Ukmm3&<+&q z&W|iC_^Dy-<#0!6X8nDQZb-;kLsyT6;43vr^t!57v=h?>tkWPk?Cd3MU^~9OYp@MP#)DVJKZ3e$4A87ksWW>mELUUyiq+r@aMM zOT<8v;Bk0xIe}Pa)xg@~4RFc8iR`ZwW5plKv&Vg;S>*~bq|# z5WO~7*kj4|-P_B;^LTcB<6$=X$^s%gkwIVnF-{-wQDt|!$+6~FWm)+fb6`R^3cl{H z1SO*}YJI;6XE)Qw=3;A3a(+yc!W{;*oj ziOi4>22{HSyw&31``roh(f$g%4{}(EN7n2&nF3Z_?ZY<~%Z0|DbO4l@&Je-%N7E z79-yoYe-rHLlkxTk%-4xbTlN3Id3-;PF21q6=m7?c!`& zJOdTk7tprAnfS|)rp4~rYPfH0lI9{5ENH*aMopz;ZVbLMz^2_KQt zX<uQS9Em_<;5zk~B}0H2=*4*Lq9B!wW$(@fBxl)_qbpk%HdKQ=lO( zz>Io2qYwOFkl3vX@*{$m9m@10uO9CwuAxm>9Fvus!58`NH?j!LaBAKie}a4yBT- zqk|HP^!XRQ13+^vF$v`v^!AtUBv4kH> z=U!({|Ih&Y2{mNuJcvdW`ccTHnIy+p6NzuKf(wOBWLfP?M&HYnyX|;95?9l}=W2$K ztk*pvJv1AA*~xQC`gl{alp&NA1VD|KDEs}S0@ZJ}2;DZTAaXYq*e*S7wswaxblzXd z{^Q<+qa8oNmdBA?bKJ>VbxE-+I-E#6bq`)_y3WSeKV|JD>)9u#C0YN){_Hj7YY^rd z0#W7RT&GqkA~<;r@KFWO5w&Cr9~V%WIWO6RJ}=oP#z)x^IbldQ(Sc1JVc>p$2WxlV zkeJzD<+|+VW3&B(@gj-GXmx}&Tyk?{HyQ@9(jvRy*Rx@`vCoKg-;x1xXP=Yn>sjXY zyAW)1MU~yqAi#=5TCk$G4pVMhFLB;Hc?wi?1X~ml!m=(p(C)2DihnI){;OWYxCkeq z9EmJk`Kq^Wu)~o3nU>F%>fM6yfF>CERtvY+Di9~0)BCtoiG^%)8T{MH$5!4~VI@{c zvnwV6)Q#T5f8W=^;h$3|*r$rqdt8Z#j7dR|^KLkv76q@o6xk*<7j}t}7}sXC9I4uC zxcJikTG08f#XcS=XZbXiv(ims?8$sNHuZfOteli%pTt(fD)UCTRH(%^U0cGgipYj1 zM$6dN1(=nx2!)hcqa?RHADlPmfa$;yVihR^d&g+dGwdPBOWoj!iE!OvI-^!nth83m z_6~@MY=NuU!cbGTk`+!_%-;Gfz`j&JiuX7Ff~w94R`i1}wL?jq_}T7amM)p*2x%N7 zUu~N~cwr1%OlPsrd(N^gy9_v&2R@N{<4|bYnudn-%Rp(K8avx(DV%ON%tjn@X1CuG zgshN)Z;B_TGJ0Jv^OE6qfS(P2N{OC5;CJLE+p3vS+~sYQ$fpA2|7Q*(Z`1;u zS{^Ux`W*RX9)n&I2h&CmVf)rk-1!al!0-->+{fu8d8Z(9kSpLE(SHdWv%eA#i9bZ( zfF^xl-z$6{Kf}$siL{M=6y5jn8`j^LNB>dwra$t%z?W*KIb%U0%@o|2UO`8*o}|$g%kg@a%7P1i}kC&;&xpjL=~Mu)3s&z zq7Xk>dQ1hq01bW8Lp`0_bskZ(>Z1^np`iw;CS9+dLNIFINxJhMhfRW#<^|)0#mI+#`;}%M4bjA(+u5{Os0{!!b0nSKv!=KI?U`r`0tbZ9JDe+Wl=J7k!-Fx*! zUiCLwaQYGPIU*{#T<7%8RdWZTsKLS5I zSb_X?{t)$xCm^##4h)^^A$R>b>iN+Y;_>|-vwx@sdrv+j3dv^V;5}iyY)L6|eeN&j z`l%17VAf`AT&ae4blgUqyRpm%#tZ#f7D&%@+)H~8>C-nh?V`U~uHvy#A;^mBg$xYT z*ysZa?Ct<}4ri>A*{^jQnZ|G+y2%4BXTOAFkr|-9J_F93TYxsc5vDryFHo~(dWq0B zK3w`t1xMO$A;I6`!RDF^FuEh;WJ4v~H#TPHfHW&t>yF=5e@6>=yRRMV7LsL_`;oMZ z0uDMXK#wfnhD8d!IHGbjbYZtP{Wl8zcHW&z`x`Sq&*2k0<}R32{gKTygCVF}fnC3*T^X z#y&0++|Tj~#56bq?QgFqJK}4}#?mpedT<8m_sC?nU-!X|>mN}m{PDyt%mFu zH6gs#K^#Ah(x9??HQ?l#5NOc(N(A5OllxhkWcJ77@NdIDw&?I6e01IkFH?@OleZPv z+XL}X`|K5|*HcHLv3rTPQUUT5TaG*P19*4sCU&LJ8MY7?vKAf2%+@`a%$bO!mRly` z4+lC)$S!-5@9`AQlqm-OU15+7b4qE5W61i`Q1(K6+!tXw9 zz|WuCqW~oymvbqGy?;E5y_i*p{K{0Rv==`t#-8euCq~+6L53R2AKJnbT-Kq~@CX$4 z`V4yT`4LJjJA^`e{(zr=8T`F+nKbtNLeVoZmc1H9{ri@NxV}xOC}I=R*HM8WhkQ7^ zh6O7LF{-v@9)2W!hn%aBfo1C>!ANY3)7&AA&ig#3Og0pt5Q(*Qyr?cZs{IR1*IhxP z?kZTR&;s(dUgT`OhtXYs6SVhj7e_fmfUKP%f<*7dqhIH5QX$;09CsIgvXr(%uRLo= z(_jT&Zu}M96#_JnxgIXhxDQ@`{UEn}F=)HpqwcJ0M6@FHvxC{1D z^WK_Kll*(YkvIR`t+k1Ff0blZ`G+Zw&q*8w!P|8kVlR{Nb1TtXvs|biafiPv9MB71 zm*4&TTV%zOz3|FEmNYrWll0yOrX}J6!&8qjar4FDllaJ6%HguXAC) zuMjNXb;H72eArC);Wz42$Qr79<1|*UoI}4!d4sq1CL@FK3*4^(gdvDzYp}wOA9D!#uoK!<7{#=*tot+$^C5t-l0WC0c}?cCshjFEkv?aX`Wu zUNtRI80j~xKo=eDD9KVW-i7Ej$5bGtVtG_%)v1*u7AF?(WfXhV5#9R&8P70Bdk$_Q zb)xsUy7(2slq3w^ zMUhJFT;-TJ=Fda|2o1g_X-E#)uXE%KoLa%L9-j*zd2WFb|7LXDxez_$#c$G!ia4vD zucDlOj8cgQuW_yGk3q+Sld!4tJ}fca3epQT$w@g>>+w3IZsg8c?zo#WQ{=Le*%Pgg zPwiR<;>-3xjodK0{mBIH-uj)CoxmidBc6M2ks|sRQi1ZnG;+RF3)Jyn=wf!PxQUXV z$l*D|ACPg(8dTCzfskq*!*_&_tZi#Yv5Smg{gfd&!}IMv&Q@jLj+e0Z^K)6_P{(@F zWtR0)cly|N`xz{|Pm7I>+01&L3rC|72a%RcGLLT)q!my1VVNgB_;T(7cqX6?V?mW< zS$h<^V`xegm9^l3^ft!pOBQZ_dJ``TETl|-&mrJ?4p!e2W#?oFvC${h*xs&4a_L|> z-Wn3d>TomJmYJ{F94CqT>{oX58NLsAGyhHqGP%dg`*q+|dk*7Q&t}jXJ=%0LI)_$C zOCkBdFC?V56B!t8CXd?-$?(2qB)<42vj1@r47U=JXOWJQ-q?|8<5Ilk)<5bw|0XQ( z+JdolE2h%6twDz;vQW_YS$rU6FTQMhg$NuJ1Ht5PaNPd`%+~8+?sr}yUjCVA7oJ5% zczoBxSIZy;-)|6pyL1j=6RlIEI|)d0^q%B!jsZ`JrBF2_$G` zA}ib9%!6PvR_0(h!_=j~9TNg6IX0-$E`vBdA|N|qM_<~YLTmIROjfCIZA~v@$H-Yk z`PUQTeAXG>%#w$P@$JY@aU;oRq@Xs)i*2MO*s?_dAl|1+YDR?NFTWW~4}U~vwx^Is znKGW&zne`{@nShbs;p^^2%F+3z}ogEvnna=FdaXG*gZ2L=iQEw!T}lLH=K$3Zr*3} zYqqjS)RGr0W7?5Y(hZ>fO|iw{7|-y)cdSB6VGRynNbJ=_e{We0CKLcB{9%Rgfw z+hp|)v=Xhs<)l2Tb-onps&skYi36Zwssd?89LOr~SLEaAt=wJ@0j}zvbtqncBlr0q zb#y~$B{}FAi`+I|Bu_;cV!?AjT;65}8@wZ7x4Q;(HrQ~sJ_}*AfAOKiI(#Vlb}uu_ ze?Pgiq@Dzw9fG}Y??Gmg8rUDZN3?W4pv^tY$+Ji!kg!NZo|TbgX7vQ&c3IVhx8%ab zsBBQWatJQlWA?%Cz3h=K?yMJo6S*V0mMXoqhy=~hnPG zoy|~#mpYO2)K6wk8U=q;bKzgrMOZNVAo(yJNlfdzsEcm;$n3oVCHNx=UQNidho0U6 zmj_&!e0-Z!Z`p*-n{h}bPR3!qkr+pq;niP06BD&!#yZZJQ|q8fWD3>bRg?~Ndb~yR zUY{kaHfNH=#n)l;93SZ0HcEx$Nw7Bu`Pq4V=IrZjrvOgoGTc=k7||0=xJOHZ!evJ%j9|A1hMhaiAhd}L{+`N)~bt}G>*PJ~Zic&4- z;jQPh(VdPEa_wCZb}2W(eqZECUSTu2ue^`>Ehz-G$7P6F-fb$@y^=JvwU8a1V<>7< zCyEljK=f|BCTsm4!pQY0xba(oJvG@(22=WYe!Ov1_4y(>-uMP3{!FMfN}Y+_{`sQ$ z=Xt#pNSFGvt(6>U;Q4jd%|%Ne>EXx9mRLzFm{fneh=qf#aJv&?H_L8h?WjVC8+%TP zKeB_r!Np8Ks|(4Fe9g$;U@bO3JIO42c7)-NVVQs(C7|+Z1FN5{!2%^tInC#>6zV# z!Q*Z$cYo4FINnuGHV^(XS2DPZruIBxKBh*q=M}BFE%&vUSho%oy-pYQCi;`P#ypO% zRe~9G5W`J=t8mt4DROtFG>LR0wJVBM@Z4=K>=-c zz)Od{uCNV!FY>XqCT&psxeuK&y^7X#h%ryoG}*-Avv7Tv4|U+3OugNB2W#~u5kCLx zq~7rvP@2wR=8y9i*3@3mV}I;Dn=uBv9- zqtC*sf>~@(hBO=SEsOmgV#^9HSqyh8KcFjCF5INsecZ`s=a@pzc=LNIvk*S)1W(jX zK-z8rxaaFt*K9Zgog840?d(0AytQhmyi5qme(r#TQ3t5lwhLxjzDCxc>xsmVTkvL1 zK50#O53T{Ctl#MiOjKnLnQy_zc6(-z5mCAP0pj-3uKT~tTb}Cl8c4{KOt*LH9RwB1LMvPqAQ+@u>R6)GG6(V z2$Y1AjH2t%+r>}5w5(*VSA8K_6)Ldo@Ip8`zZa;@-DEkd!X(Z%W$z!30l&g$XmWi4 zms*rez^0oUjqf6tyA3EUz#kn*F@S+lob{F7o>LM2Q{eYl=IPmel2Hg@lRPXaPqF#4~%t9rQx$-!Ba8WLa7Zd|M z(>b8JG#9pQ{{_nFQjq`V8j;NL0l(O-P?8|ShP^oM zuT@x1Ce>qc72Kr|2ROsK9q*Sw+mx0fzuY}p@@28y#$GT z)@K6S{E4CC6xF|_gS^?;OnjSVd3)Ji(BJeQ6l*0BJ1Kdvy)p(h@ABZTcN8Vcyd(;) z1w{PX8BXh2GgQ&M0FuKx;c>(cSeC{`u3{GCdq*$Qm24+fY5&2}m#Sc*Tu0g{3!D?P z2A`EfSZql?X;{-mj&EonxAU&SXn8$6|5^bD1b2dZ-(!+#`KI;WQ^yi)Y0lkBi65wS5_0gs+5XXHEzbs>J+IJTOv8821kg|q7bS~ z%>(ROc0?ml zz~OFRJ7O+_ry}}L>5pwVG2Mx*U;7N+ z>vJLS^CVd|xeRV94-;QS1>C)C9{t676bGU}ymHlIxHvHz1dn9m{TWNKtz-dnXoWqa z;3r3JR*#@Lo6kdhawQdG%|-pE!=ZIlmR+bZi`~M_1s9_t@cNg?sp!fEGPH(C&b>uk zwr)cii~lhPH149o^J|!gp-SYC=UZ;UWDh({{#y5`WdKv~)(#7B$9@5SXy zG{DhjG46B`AlfJGQEQACnh<`%SR5`zc6ll=x!Vfgj2ht7=8afrr1)Ha~RMO7%xPaeHLBL*6Bf8fa-GpryKM_2e4(L-@=wAW4%b`-0z z+B}Z^ScnRa(aWYGgb8Lb3+CuUT3Ya%M<&(^|UUHwBq;~!dVGfQy;w(&!pr#+^ zFigBXS^Lk7Beky+t!dQ6%uZiy_9BuZflcV!emA`KQ5PO@xq&x6i^BSjqsa4tA{C}C zO0P=nA{oqas{e@=J*0CUnT)(Ztac9ha@vL*+4%QL$V1ulo5;nwVT1E${Ez0 zo$oor(J3hSQVc1Zoe7g7J`g-8$%^r}pl*+CxL|`8jGJYV#vCs4NSuLN;0}5+L!TPG z;6)@mN$&R!OMPf_Ll5KPL(H_FAF8{ zL+%g~S+an(J@SEk3@b+sW@~We%z7$O`~~{`^#SVm9L99;^^)mA8ImELgr5xfU<2M9 z@3yxp*3+L$d*&wLtfLD^s(=m%H#Fe5tauz%8_ILj&H%ZB92Brsl9`;tWj^+Wkh4QJ z*cD0B9t{d)G`X4KM2Rprdlpi^mZxAJr2)L@(O>dNsFEb*t25J`S}5aWA4#_AASNdE zl(xY+bZuD{HL@fYpSr=tJHA9CrTqz%Z0lK6(r$tJj?N&?>wlm;CJ3uJY$BY)A>>n> zDg-VSq?=AW!p2t<@NJP4H2JZEq}iwwJqZg=@X9u>olU{weOHU&`bOUGa-6sC@SlrS zDW0#YRGzL$@WH7`F{CxRkrMCO!L+nBnmfgPt_|bmutlD1<&{S)pe@!dK(Ee(;`e*v z(V|(wBvmmNg}iLStuB?sFK0Uvw3Wvmv0>b<{eYgW_ebvg5u*?pTho0cWpVe*E%@et zkvQw?G90jKE*26VL`e!3SRqLnYP(F)gF9Q1st-#U>)BwwfqA$pl!qaw+QOMPvmmZV zgLClROi(MWV4RE&klU9+s5?28DCGA>Tv0ecX-$SA#}$N_hDM>%751q7iamZ?KMzZ) z0M-!pgqB^#Y9q67`nNx5dP)*!EPsmB7V=>xx*B_Tr{V*v;;^g6H1e`F z$7++ik-z&hRL<&g#CLI-#}T|N{U<$o{iRj3gW+4OJCuQ|V#@L1ZLg{GyEAakC27!& zaU-t9z34~24s-4MLfGgVPoChrkn=W?JD#NhZ!%O#aF994`%jRY8n~QljF)0Ge>O5< zD)uDqLjqoZV-&kN$k3*qcgU@Xza+|{6z!0gA%ZpQnfh1lh`U-Fy9ov0zq?ZL{V_Vg@#+5x$vGjL^k{$je&>+}Vr7 zC4B(x%$7oxp84n`^$^(zd`7ao&ae(&J(OMNiM+qJP=5qg(Hr;#=t#;Lcg+%l${$Ij zS4a=Tixs&1f)<`3oQO}ajzA~P=OXf16!;x3pcg%j9NDQt=FPt_s(p%gDfuuT<=>s; zh{_J3Nq=el``i^2D4_{aCSgQeAcyRe{lu)u86k6CMxs@ATNw%4a3uZyFIuwY9rEwk zLsHGophm}Q)c1-Ph)MOq4-EwAFX3lx&#+>`eOwk4fSVO( zpra4#7`1_wczx6rF7rVIz1ST?QUtPT^|yJno!&LtY+nSvrT!3E2d<{1B|NZ-SqW-7 zm`#O+jZzaC(!{k~k4QQ=(=C;AX_1zhwE1XUUH9w#RFKOuN^Qy&Uw$J_%lRnLI@$}c zZZ@E4stF$n^hX!;ub`t&fa>4Q!S^=Aq0%#HgeShio+{Yy~a* zIgh#>mqf-l-sT33-okDEuK0rI0Cn&9MHJ^DLkXy)as1E5qY`&sM#LnM$C>fduLQQz z=as%=QT4BA_XwAM={ZDqM&#fwhfR&x>vld)zH%82x(+k4Tji0$ z#^)T(sRZnlZU|%Bq_L7<1iHkZgG^1P8J#n=OyXcDcf0OPN+Vks3r||0?;m71S1keC zK3s`?btJK5h6OG@HH%q*bx>A@H6FWpA36ShhfdVo$43ge_`#k6tXif_TfSUIr>ZK@ zHa)xkwtO)(6A`>;QA7kzri9!690_?US5o?Neqw>)y;#&Wd zQO>U+*XZp)k4}^G`SSQTw&0z)G(clcA=Pi!PGVdmA@iCzyRB}Jk!+|ZmHumq)BO~Z z(L4t>oi#?4N({4ALxk58=}VG!L_zrL|KM`mPMH1hI-Xn`k7G5Z@r50!V6(>_)-CZQ zuMP&Y?its?`u2S|wabp}7dB+i*j@r@^#$O2Xg3+CSqK55d}L%-6L+t-F;QKXi2V9< zsKk@+nNs5o=<0cY(t{eQPttS2a%(s&Y}1CJ&=4q}=?k;V#KC_=2-UCBMdhb{QLlfN zl2`kL!D8-R(ojARsJD|$k5qcy>K9QYNc-#Rw z@`Y2RL{<{ichOLxffhQi?upOIkd4jdfY}qeeeC$1(1Bb@ngK2Ix6qV;NNxtUT z;o2<(pQot934Ubr(wmfLgamoM(-0 zm&T$yW$&4HhdIomkHQQ-uL~{dRiyl26tu3+0FloW-f*oLJ2^Rnosk7OF;z&MgfpSy zayn@&=0kt8+sWlIK_+>HJEMMG9d7*m1b%NTK-y$8^!o}TUbq(;m9Id`@D{4{?NX+; zrkE>O(8>&Z=b+Hd*U+ojBpCdaO|Ha?L%(h#d475y9DMMYRO$0_4c(z|YFAqpB!*GUpj(mwpZ6Wgf!*qp`4B zwhcAgN8GwTFo-R}zdT+)0t@Wg=~*g;o?C zqKdvRVs^@vf`eHc=!P8upL<+5yzwh!d})Dq!%Ogjr#$sV@Cfzgc_fL5uOxd;ETUNT zo2>e+6qd#1?Cud&_>;T=^lKTWb;BFDTk-~Wx~7Bi@i=mCofa7$o2(06yB=?s+=

    FB&_2^$_GkPI=8vUH{mOJ~%T=afW5KyHnY(JILbNg8j+r=r$&H z^9APZk9p+GL=v%jWCFzDIJmjnJ{@*bVV zY<;#B>zR9mby#;DieKizz*`S62z*8C6CRV`srLk_7J=imHw0NB=<0h~EM-{9oMry%HL4)MI%&VU$ zY@4dkzO{XTu6BW!Zv*T*xrD9d<72~vrCEOdU9hG)3J#%1%!P&-><42m?>{65nU8)k zI~pnEFHGZ+$8mVcy{ov@y9=o&yyOhdng>hCT4Fv_h?-U0@K*&1r0FO@R+PmtD)N2k zQw&B*o#r^aVlGy%?*{?iJFfmHrn?{PB*!ktp`3v(%C;$%G^=FdsiiXXuGI)jE@99N z13x6VssQ(Jc^Q7kLWsF?9r1}T#5I;zEt256_V?vr=X#q2yl6u*+Z5nD?@a6x zL*o$cJ{*AWlKfb4l(4yunxWtUV{HqWZv#hp`-Jl#l9Y?82kv6$(Lro|;te`0){kr> z#;HHAN=b864a6SlhM(KJh})?YCcnfP&+a^guCH50{f}2U~ zo)b`-a|F^vA^!rbgZ@&?mqOF3R=)bjCeD{`Fz@}($PGS`eOhSMLR$t zv;)-@{U+DB_ekko1JGt9@XYp6^jz%{`k*HcyJUD-%VJLuKVX1Up6H->*O@Sz=i!=n zq7utHy3lf4cax*5FQKyr7bvwi+Soh70zcd|%@IxTN5LO&!GenlY~k$;(&6kzzDOK^ zcaoBtR*Zp{;J%Is4CFbFHseKNuc^dUqnx_^ znv{c7DV(paz&Gb_!MkRRp!hFIQ2RKNboyBXbcVo3{?}yvf+4bS{#7t2H)9I!3((vJ zmaw`b6PEmK20K-2NV3v^C=W-<=cxpg1@@3Vv+~gFqry1i?_ad$!ZGxCz7gy-7KG=G z7m??9CN@&x&HitNq1;`%xb5~B+MT``tBBcR8|Oc0&7BWi1xIm+H++a*{kwu^uULwQ zZrkCbpC4kA_A4lpmp@aN`N`y(O7NJ%v%+%*EE4vbbDK9v?Um zi|e_eI3sfxUNp$AcCYboUwZTRoTHe$VZ zK0X+%hgKAdVdH=8IPdlwG%lfsKgE6^roVNuru=^NH|7c>B=eE%R=y9>UrLG5*ErOF2@Nc>QJX#5fP{qA-qZxoatwUrT8!5TqOZ4y7xX5U}}ORaF`#-AvR>xvr_i*apz24csqpe_3xI9X#lxNt@@2`$=yi;wYUVx0oC zV^}txVe*W)#reYV=>wd?zlLyuc0&4{Ysk?hYk4lBFzjv@jSJEz(7HQYanPw`s$SU{ zM;LG8+7>I|4_<26TsMhY+58dL9dpB_cWvk;5lUFJArv=mip4U|&%yj_iNxaR8t6R} z$dn8HBYguafn(o`?zde=2UPWOquvyC;r4qT3nYv69v#7fYmLyxnlA3myDMRtCqt?d z_aGaaGC1S9gPp%&i0AjNg11-I!1g51Dg5;v6Z&Bx7Ia8}ITz&cj>L8h$`W+jDkIuf zJ`bn8KZfV=Jh6Fu1W@A+p4axu6j>T|iXh)lXj#T4B>&$Wa&*r_inBSE3>|okHioy7 z6MyEA!y=_Ht-1*|?m9!*5^2&I&{7v6=Eo_1?~029-l3PpS&B(f1?oF$R< z5he8RoDa%WD`ol~M>GHWx`{&RD0^!8G#hzVv0l;nD@5|#Mi;C9gTt|N;Or|EJl=hV zH1je)i*(PCi41=dl^#s4A2vsYhx-`K=YPDn4a7fishowOVuEF5q52>E7Z{J!CJga!#VTa|<2X(uGWhsfY4q-LCDu^A zNE6C6miV=|lbjdUAn;9`eRQXm5#EpweAF{gYxxb^!H&H%zJc}8UqmWjoPrl0)F883 zft_5pjg6n_#~zE=&UwE_2bVh}qk#FbxMqy|0l-oNOEJP40PG$yu7K^++=K7Y)N0ZcbhWaC(Oy7BTL1qEJb5QE_m^P zJI>>ADde#qeX zA{||NoJ0LIwnN&>I?z4ZhCI#Y$I!+_?;3@XP4z2I*pm0u+@*`Shdv6BFh@)3@_-C_ zVs#QJD|+FIr}?b#<;e)g*o%ZCShstIyWPS6rv? zo_1YYvt5x^kkqC*Oc*`ts6^R~k63X0gitm$kLVQ~=6SRhKz->o^ApyX^WaB0Vj^#o z;R|Nu=t+0%{BH*Ge{+(g9ML7*fKHB}hAt}I7|9flSmKcg5&E=Z0DaTFkm*v}iOsMm zj-Ol0@lLo%ZCR&<@=KPYvam(iCXMI8*dhdH-BjSinmn||;{rOkdKyKJ`Jl3GxoAr% zA5J~{+v3ydTqgR}6tnb=5Myxg;Qtsp4}Y${FpkSABQ2srWn@+&zV|$rq@f{_mQX4A zH6`sGvNKC&Dp6!q8ou{Dmx@$MX%LYLMOj6Ii2Uw9fS1>~=iK|8&*%N7QTOH$)B0jY z;$Z>d=yLM*MmD*#wE)xua^OFM3i3=Xj?CPhMJCv~Li{5w)>{}{q- zw-LUJr>d|qIw#p_J8rVmDt^$iQ?^vz(m&s1(t7cUQNx%}`Og;H7MqdX5g$3aW($>w zS;NThI|Pfk&rlw(L+@Q?=vE;wd3%`(k^8(tbbhlG4nHPhysIS9?)X%0_=XIS$qa)> z^cws&p2ca6_uyJjScq5dw`W%ET+S?PQ)MJ_9q5rbHGEMw0<((inI~gE5vi+^+{vK3 zq_}@0c!f>@J5Ot7=k+>KnsFkRniZID5?mvCBe9sVOw}as%Fk0{$(J--*M^lZ3xRK% zK{Ql(BDyH9;O8{y@cy+%yyu)O{MsIb>co`4ylfsHGen!8J^mwUZm=V%1%u49mM?Im z=qDYgyN4Po)Drod&UA@=4Pz($kScti$X{u-;+d{ErgXY4j(#t&%GbIvqsv1`hJgW* zcDez(YF44ozIFKM$Xl`^VJN(A_`(f@2k}YsBZPfInxCsYo5T)ELvxQgsuhI*Kew1^ z@;Ze{ok&!}d|``=B+Th^7CqQHfnHNPAdZ~ojs_DC;`H7X#64myL8;-S!gUbmabS|Z()5ZNXP<(!}n44=s!Oc z5B)j|8^_$Gr)mwb>9GWEyto;wCWVXMZ?DE7kI$gJ)7o)k32}XTYvNRp=^HCDZ-d@lpL}bUyA!8{?Nl;g6+In6Mcpws(o$UOr-W=iT7e zmOUn~*4mPE5>I-%hY*(qhIG<_57erw8~=(*a8k=5K53mDdS{r^oz3Mm!|njl=)a9| zD<0wa-@4E#+elO;e8^6VVL0sJTTB$Xg1gQoL*+qzT6Luh*Tf6^Wk)?d6Y-hyLzk1C zx(Dgc^hJO-r9h&}2XZ#JVMNGTdK^Z9@AXMw8@mLJW6p|ytDWaY2TAZhzuWRc?;5kJ zg6Or|bC~n0Bgv`!k*Kof5bLCLktzhN4%*;_ zhfB~cb2OgW)ec*#Y>`f)PA%SyS~B=22+yA zu=R=pQ>FlPEA!yZ3opKTQV@SO>@mJATSVk{zT=uAw&9=hUEF^KGWgu!3D>)7J@@qK zQrtg7f^G{kVxKoR!+(KE&{AT6svi$vIxmerGUoJrhabLsq>LuTYPkMsHvRD>ob=yF zg4-`2iyQ-Yk=b)x;Etm|9Q^bfp08BFPJu_;L_VTQ%~P&RS)Qgm`$Rq!<-pa4--xo* zPvQN#k;ojJz|^4=rvG^fZ3ebPq3OH8@>c+@maDMVGZ{*X2Wazr2ij&P$xiE1K~npU zepcqF()P`?H1zi^cqceDPYVEg)uo@1|pl0MgRVnPq}+fIk%%z=*XE`ma%ynmV%qx#g*>ILpOhd}Wo4%f0X$ugzm$}T7qh{3n z`gz96Qv@2*jL5~-#Y{n~is_tc%$u<;iB z{ZtcZid~09?l(5I7QunamoUrr7*%q*NDL!uL1O9>U>!EW^A9B;A+w$MuMiwV?z&hg zzn)A}`U6+oHnRSVB`c%i%hcve-~@N!{5yRW>J#sPa!U=T|LY;Vb8G^fc$-){HKXs{ zOzI{MgzZ&3_!Bo?@(ZNO_{zG&I4!WAzU>Pq|7n?`p37OTsq;CMl@y{`ojC?G2dUbk zXdV`YAl^HQIc1Sd-u15*CkxG(ZSyXZ^`1*{Xs!=`v&a=QT3?d7)RnkikHsnKW2jT1 zH>xb}6BVu8%#A-h1Tmu$C!Sx&|7s9^o?Q>lSZG3;-^uWeHUEg)fr}vXA`2XBvPt~p ztHfhlDyJlvFYBIIldzn*aM8{gO_xRDfSwefdp;k2&r~ z8e{9!4wl7*#5`{=O|N^(to{6eT*g;Sz|s&%I-&r|pA_I=z*pw(n?w?)FCx}w;y4rI zv*@-{mcO=iHeYkio9C`Q=l`ilLG8W@F(+{t+l1L#F3pC!aZhMgAcK!e{o%{DmqgY$ zm0a1n8*@ygdF@PVx;d?w`aki(_N~9>?Va zKgq1(NqBVHB0PWb2hCa*MGwW5wHFY=4MaW(+^1Y7PISYyryd`$ALfD)G2MDZ>R0q0$N;X?Wo>a6YcY zCQO%SuP!&^Vardfn(T)zmxr*w*dau7MGezSrcs*H;w>p>v zJNLwLssAK#ta1g2&V-=;)Mt3GNE*}s8PSEWygALaU$_ZWUtnX#FgC+)kj>*BfK}rz zIF_)Jk8hjFUzgFt`95!8Y5X#54ByY)Tx>-@U0#pHZjm@S{|Kw?@*jJrUxoFpIYSeJ zTg+}PiZ#!3_+l|@vJ&&3q7-SpLun-0!i28yBDdAH@v#}J_~P(>{GIKF+E>S5^zJlz zc3C8i7~x8lrnll|t*g{z{UEK2YvXQxkVV=d@LBsMdBtOE5$wjY2fc5Dc6J^dS){}! z`r5IXhc$^maTcHcI)t8@9nTcEo&v)w5wPrPz2Ky|1hu(&uqEJ3e&{ECQnPCpgzv8g zor-6WVf2fP+%}0Gcz=^PuT-Uz=iDIvr!r}B?r413c}6Vxy$sHqw!z_wAa>oGdOUBQCgn07B)u5A*K+LR9kzeJ3*f%)^j-66qKbuO!%1L@8Q0POi zebU2K{wK^{-?Y=(nm#Vh@2|xy1$CzGcq_Ht{ETLMCsWt1H!#7TNsX;bAQmDv}5#vS_4z}SqcWmPlPe@ zUumH826p}XUiLSgkGrF6N&S=OaCb&FC*dQ*%sUx^J+3K`I9LR;f&xG{Y$)^g{A}@# z*6#e(L>|5i9ox0>*Jxb(GUzZ+VRw!F1TFghkmhp`u3xF7`v$&)dyF(_21khhG)`iT zdKR&g?e45rrvvzZ?gFiSN5R*zlv8_iiX4|+Nu=*RCGC5(;KME>kSk20MG-QvbVME4 zOv}36?`%#whh8C*B3dj$f`^7RcrTnu(Rfv32G%s3 zA?npSurtR5w%L6YQ#FCh{VkEt4aw>9 z6^kE9vgggDVN8;c$$o1DNh>zPX5%YxMp=SvT%AQ0f4?J|5w)7m)w|1-$PJRFZg=Q> zSj;Ri8-qtBy@amwb>{Y$IPN87L`9Ewz*!>)V5h~4CM}o_bqVpn|8~I4`We`)I37Jb zi|FW%1a7;}Fr3waqG7Af!W!NR+yn-+QJ^PplC8jNJ)Q}H^P=Edk-gyA@jHVUVWT}b{Vt(b6<<)v z5kh`)i5m5kRbuQDrmWoe2tmtJaQ18mV&NSF z&wN7!CXzKtaU2dZJN)3?wMUH1*8QCR*}pV3w-TLlx8NG1dc55l%G4zU3tfsf^X1Gk zI^)<`vEe0K_O(kH#%_Cvk2pC@eUDsOkp=1;s;3GS#$5E@6!FFf2kFj7Bl&>A0U{MQ z9FshA#Ls;{kbm{TXr^vr?m9S~^B7_(L@280X5J5TKTgM~BaOuFFSpXcEN|}K?g6gm zG2~xy8Ik{|;WcO!UxZz2gG7JSPSLGdEVs|kpFXW$Ln7XN#;hw*SD+hujmS{PUxNnN!Rc+vumStzw_?0DYF9rez7-n8YW!*#-{AuZ$M1@6P8ZHv1x36x-jF=H(@z7|mWfu4 zH~|4lI#jp9lx)`UM>TgXTzX*%@q4R?PyOz|oiPh=Poq9G6vqgjlMebt(hj|5x=@3K z6JXYoR{Gkbma}#0AxYQA!MWG7(B|`Jy4%zqm#SFs@4F}PQ~z6n{#{`fG1bP@&bWuc zTM;y4^BLrR2;T|KiCoK2X>3yeO^vv1RC0qNpDyEtrd#!i(Y-5lW^5~I*w{#BHp9}XoEp>%TD6*_M1DxyE*9mkFAB9&pe%(PYC=pDPq78}mi&}XfG=~?AH zv~Z6qNKX!>~SuqHln zDj!0b=y|EsB`JU&x}pcg$>n6ajVaA7%%&HMOzGgr?H1d2_z@YCncNIjMLhemhpDq5 zI4tTFCy^gd9{gv>J=a`?&&uLCOXZ7<$Cp2(sX82wc1PghK{@`h@mPMiXFjeL^XMx* zlZ>4HfP`GuXLf8k4GRRwWyEPI2yu8xCj?)kA03TQT|)vx*|9{`A%xL*I3C7xenf4f zAql%K4GX=NlkE%k5a)tcV!P@g7xw%$ecyW$7l$9UP*QAXDm+@bv0s03`a&O2y)Opb zYRkwI&u`@Ayl`eTS42JDF2OeoEqX?RN`j@O%ndz*P9CHv=K+>xUi@6 zw1*_gYb)TkwErbPVnUhmhZXR@K}C{NHjb@vw_r8f`k*ejH9tR2mooaAR7I(U(_bj? zfHV<;?FV*GQFx!peu+;&EQ;VdN>ruCE# z&y2!TX_UL=VnUK+-;b1HMWjKu~OBYeGCjVt?bmb6WD;8L|l;`du( z(iWw{tX@YM%YF|CS@nUK4^rm0&Ks^sbviw#v6DQAx*!_V7{P7brAJdMN0AEocq(<{ zES@;=f`*kjVT5ibK9ZGS!)z~swtoW{N)LmB=L3OR?F4IQUJ>tn`H;*VcnfnCgQ#Uy zDfM&AqvZHhn$mMtG&p-X@f>mycHepmquxug@7kKg7KP8rqsCceUzQ9!n;1vGj=4tH zEhuO5>=S9tirYlp(uMFV+-c=OWhmI6O?K_;VI-?7xEQ|+Ozi26Wb+h()fj7r@`j^P zm(`)3Z#D5zxgvL$0LpOh}|qQ#rHmw#O#7RFTpK8IaU;DF%L96 ztUfBaX>zeUwaUPzInS)k0F4H+T z10kz-ANL?so_2mrB!^48=rYeWyuG#!pXj}rpS~iJ?q#L17bJ1^gCJtpEJ5sU*wH`I zhP1b;jh_5AhRf7`!DZfALaFyfang8qn(CxV%Ooa=XQc{0lFWs$N;!_yF878Y*;CN? zYzf|}SSHRqB#HP-o9fh397uVLb6Sm%8=ni__w*p~*%p|XQN-kH0EDle!}MfKhw}k@ z!GF$nyxA}t99|zIv!1sTqH_hBgJq#UFb+IVdqJ!2d{IxVG2Ol=iafMF1t0ZFF!c6t z;%y)c)zgoIaYZO~N{(YJ`uuRnlF`@`Fb_jkPeyI~0DL-J8W;B8r`5lAlgP|QPIf)y z%Zlg1aT{}(pLA68+Fb9Qz-`1fOsxkYCp^&y%Dr^)nfAMi`KgW$FO!5y6?@Hi}mUjNHdJkM>X zzaG02Cc}VSN;lz_X^!Kjc*WA;GKcwTQtteM6-vBT^&YzDuos@OU4d__SxB2Ok;OT( zY(q~ecii(BZ7xY6a%P$2-TMq|Qw+ctg$e|c=5r`8`KM*8AV~I;ikSg0J@7SZ%b&t< zn+gn8QKg1)%c(o<bOWbIEAbsq+OT-*Zvjx62YI3Afi)h=svUg{OJDv{*+9%&tzoepNPf=XQJD`ao`d30<^xmuz#xllFq?jT=IbtXr8|wx7!3`@slM` zA#`z;<#fYfOA~$eQii#oTE~=fmM}|uB(4lPif5K|ku=j0{QXy&+##iEoUZsC7df96 z@jB6XFe(h!of(cL`u910b~N@m3ETpYpLFnyBAofp9*2B912VP&)QcluC% zRr?W4cG!rSQSmUgQG>myzZSx!3t++HQ0O?H%@`aU#r=?TCyRwn%!r>a$pgRpq@~24 zuhO;SwU@oZ2}|QqU1>jB9}?J@pI_j^ncMJgNj@&${uejcPsPJ4^@yCY8pOA!ah3M# zm?diuLDh;XQupC0Hw8z*mQ|C5S%3+QNVzU#$!}rJ(RrBBa)_RJ?2nrho#=xjXK?%W zv*fkz1biz^Y3js2`o^|`IRCzcg=rD^H9!vE1?li(LTB>b2Oi*>ppUd@XgaFWoAklh z6I|)N<;2K-IOmXM15Y~KV84egnU&DOJT;t0hxPo$-u?!v?XHLWRdVo%PXrPBd(%}W zzbNGlK&*>-Va88AnxRk>CRVDF0HD`ExDPmE>A2N!F>#GSK9 zZ{JKhVaI)H{d+s$i3B*KnF|qHHZsdA8@Q1@T56`sAby20)uHJKldaIZr+5^ zT1W7LVI3{ozm`hdbaCZrOHutuF8aQ?g+Kg`LDB6Y$P+%-~7=f!TOfAiwF z^1f^upcRkZ&BOSiPE(0WZXtPneg-Lx-UOHS=;5#-3ov3rAe=K=2+u7Y#TN0Y_-n^B zD(Cl*37m5o-^&~&?K(iC3Z8MvTmMk!6T#qmHG|mZtC8Zp2QYkCz3^SC$UFaC&F6ky zK(B;9Gv6SJg`1nkk;5l!L4RN)Xb2Aahe|2n)0PJ|{c^bVjW%r1jH0()t-=397i~5w zqzk?tq;q>p#PmS7;Enqt*}Z@)bz_%DK)ca$*2)fl@YchS3!iTI;OosSo0oU3lW zf;kIEupPT+u-!}&EAe>)d**uw=}r*g1Jyhuu)b!oG)r`NFi+%(yB*YweTu_PPKhC#Md zDwv4{c5QnBdLQz{q>fV9<0Jg;3BhbmcRl|2B`|ohqi{L(gB8DrkWZEUXW;;GThhaiIaoNP;6~M2Rdv;+XHS8x61Vps&)v! zEf|I~Vv@+JV}_s~euJ3p8U<>rhgcZbjKoh$o^023w*{~ zfQOen_~(5THC|4{4&95`J6I%iUM7?IxgKCteUa#WxC4D=lZok?Sc(tMk<#L1vd%J} z37Bk*_gtbmtAi1Yx$kr8TX+^j&t#*dwl31}aBNfAhKA}RuteG&mNfpOm6PYe(L{#d z^Roim9lz7;6#@8H*sl!cvsjxj9`8oHq_e~wbj)?3+iO_{Lr(;F7W*?r;$Niu_G# zct`S}1!-YLJS*d0%VvHkW?N%!!aJ=yqW&-Q;PAKMv^BqwCh8}W17wiU+O?p(b2SYV zJf6t|r{Uj-7+}sPQfs+7AeWZHhb21POG_hOYWxLM__C1rz;Q-%r!^_NxEjsYTqQro zRuQ?jJ#dor5b=*Van~MyFb_Nny5aXp`#}%zn#X{}ZZkA)9nIF3tYP(SfL;1qM*L#; zFFNd4HgrVXVDHJKvD;#Yuz3PQ*vU_VzwjyoYf5ILyO}Qr3Ue5<>6>w?<76zF?7$T# z4Ck#3D$%axIzD~0kdbw)C055ikvqG^V)1Gn+^YD6PR~nW*qJB9_LkaYUhz2ofKUsO zexHEcsrxi0vlg!YmSzizCR4T8VUS^a0D3m6@xP9k@#pka_}-`n(tkpkn#nE|w>m#U zr|4iZZ|FRD{8pU=gk7?zmh)sbHrvtEqA@5FHy-KEsUWkxn0PIEZ!tr594(R{T-Crg z##yFcU_51FO4ev$%?!o7USo)Tkwf;~Q0C8me~nE?rTOOAY-(S6lS~+L7?QR%V70tC z{xmOuoTZ&)b?FeYBRqi7FzUfoZzTn;yBg0GJEO>(LkGt?)JcB{s&9nz#U zSjBZR*ToOMcJb3>WedI=4Dv1YL;1%KMx%~<0J;CXl0JNO8ecqHk8cv=xo3?Hq}pUA zjF@+fvzbPSZK4r$`$x_UTx5{1)Z$}tsN*-DFwR47!DS-R&_%vbRqd3vfGZ=S# zfEGx8Cn~i*m>6q_L((wfFTz{`p6;|3?&hDt$s9$4Ydm71Poad+CgA z2kHFI9DH?U87f~6LVR!#3!8_b{YeShAvmTFg!D2o8(qoF7xqk!V<7eVY7JSx&pR z6yVAGQQ)7vflS+P3hsuI@H+82x4JWg+cqhWKDaTG%C(i?A@NHb9wW^s$2>=6hcJA1 z%*;Z)TMFr%JZgRA8I5WgVe#b;C&W>ok`cKCJBydlZT32dE|Ppn<2jtNYzCg2zlT1Q zxJW$>kR~uW5V!9FJm_>HlX^tl4M)QCrCD!fm|0SR05 zFMrMB%hY$GBUhQ4%h`_8CJV>@WbEHeAfnQ9w9NYy6qX&N6Q2Iy>|5JJu?=y=NGTc@ z%v9qOr<99(%R|vEHX8p^JIN(!b&=$^6Nf z(3LlY+2zYY(%7L4ZIaUH?=ur0T;EGQ4#v`&uB-W{3gVfmb!zlBoq)ksNwg|i6{D)M z(Bq_*c%Hhsz{3c~yccG~;K)SC$PXmCF||m4PUZWo3-QlC1^7>-z^Qk-V|Ckm`bTO7 zm3y2_-zJ4|;i)DV{Z1RN7Urb6DxXgB!goVwP{SXGFqEl?3vjGc&| za5h>UAIUfTd4auH`jF=4+x zXR=69WLr2Iy{EWREfqt2(q~G{cBFHL8+_;tTFJe)C}6zCJt4PuB~eqM^Sb7IqIh9z zIp%qN#?!%hSm$&a=6QzT@6QQzeE0%PH8~@6;_YFMMHEaq!OGNX?Z ziK%cFm#w@(o(XwPEv4Cb#Lo@*DpUB8{gzCeJCUU|>)_2g2RvRoOJIe)q0SmjT+)Mb z(!j~nkh`ErI3+$K?OzjMO5P5v^Iwrj7YgG#4>E5Q5Kc zq6Mw_Xt&E96)R8EmGj;*lUBW>VTmWjiwgH}TVF(rt@D()oAPf3SDQIFt_UR}t(&k- zGKv^$e4s(iF7#cACb&-~&}BQ6MlU-|p6vFeA1*mEt|vcJr9VnIy1S0D9&5M)om|>} zQWdrDYTV#BK}5^OJwk zvF8i8MH5C5{qZZ|=@lO)fvXXp5ufJH#8xwQE>UQstcdfb1d*)~jj+Ec4Z6es5dXV# zxzX6KF&`0?v8B3y?vN+iWAulA`G~(H1eaNTgW@S21tH!kAg}Mnm~K zQ~EMYhY8&lc;|u91U$9(IPRT&h8&bX#sqH=m@RX5ksWH0d08U9%B-+un0qg{vtLf=LEZ0&w}-vvYKsE8pcizQ6z=T zS(16?I@fl#o0MJ)!Vf98>9V+3DEplUBj(R!t2$?}vBIoh!u}}=Q$+6Ssl$}*mE4^B*F=Mdchbc{VNe57a4x_bH+DROUmq{S-m6a`ea&*R zc!4Zr1)rr3AJ?(hBKNRIOpMqOivqmzx-zOqG~DX*M+8R|&Wo1w{R~9ypzfU{~!q&z7h^U^lxJft#Bi zyY;W|vD&3U){fL5mL?HUps`LAJJ*G-t-VBwxG(TgLz*=i$$`&iWoUSN6*fB7f}${| zurFK%hve@(*Gl0oyf{Q73?m_0>M#6zJe8GuS0s3USvW9vJ-H8tU?BGgZmyMJ zXIu^=QP<YV2NSiCBH+b1KR;1C7nYNLhRY zH!Rs0<(^1zO>rb!Ak; zJx!iOca3Zpsh*z8KJDJhPEb;Zv30qk-?qz%OT+>AoYKW~NIoX}681xq@(WJ&(@}VR zQ5_Z(pNA6%nqihOPmlPP3NE_0z`ge(cz;)gH`nKY+|pN~CwAN6z@RMX9@1oL+%Djd zHo<9GWehrpBdF2OhjdcmK^k{dm;Ke#0EwRjuFWcKu({Gq@2`adInJU&v^^+{H=QoviGyn2~sr=ISbOz62r#OsC#Vr&$S#O54bt&N2rwo}Iag@Lg~ZLG8vKXLIU+PgoI&AF!VU}8R5VC+tHrk|!?yL_qj z+<(+^@j|lFXuhcVgBonKttRDHMI>{!AMp#(XS*!NurL43g0!kQ?m2TCI?R)xGj$?J z)jkv(C%Mv)i^E8Tr9AcYt`;)ZKyxh8$f&t(P-!7aCB=axX?#2hdF}%7UmlYO%u#yo zO9x}rbB(-f=p{9W2kFb-WjNTUfK*3WU|bs0z^w6PVH3riSr<9!t_?)=Hjmzm%%Sb; zb}|;VY3R5353|`Tn+zSklHL8!ndsWwho`BtAU@v=)U={F2gVJ2x1_=8R}?H3*x{Iv zR9ZjhD%rSwEbVwxL1uI+k*kxh&_1_Tap^}R{G0upcuwvm?K9;`!jA#Y<=v}0wg1jj zJh+278ebYd7;Ak&%@r5j_3s}Plx%WOfIC6Fg36DHU{hGgvdwp)fjbG+a zen5t3TI&+k%mL1`cMdo6zssEZ-!Ph&_=#@5^^9gReRS$~U;J=)7#H}8kdm}@7KMHJ zq{{?3XCq@cp_|NP&6~h&+Q^bwy=y={RE2)%HKhuSCGFIlLQGHPlIDwhz~xpPT>CQ# zT<^A!kEOfGo<$iXKs(A7-y>m$o)Hbf-$pkTXyO$i&K-Or78}QP;H_C*SY-c_Tj#!&sPX^E z+deU}_UBMX;T9Sk{Z3~_#?xV=;)%oGvApY;EX*8If|(kE$3+d1E*y^qdU?!;nn|EJ zq7!s2rTML`2QlcP9y&|Oi60eBgxwvZAtmc9R~)?xH|lHRrxC&&_fIVPb(i9nb>sLf z>52Se8D}m{eFBNz(#7@eIEBH7Q*p_@N&K}#4t)8^6tQ(jfoor-Y>H0lJ9@XigPA3FZn!uTBphB+o|BzEg^Vm^=P;(uyprHuN6MOGT`>g z1V}ICnH76k6dS3b)u&YabW`YV|6|F{N(Uh@8XbK!n zpGaK7^HJOHAzk~ZL~QZ(3iWySoy*=YaQ(_9aF6diwq=eRxOm(k9TyYfM?@-wwp`)t zBTq9gQyVPIEQX^@rWA&^zvUFOK61CQn`X72;(Vg7Gt~<;@oLymf$j1REoxiw>E9fx zWaxtCcJr~}*&;g5MXg*3W^m7Or_(S#YdGl+*m7)~5%gzt9ELfM7suw|brPFvs3 z)J8s{ifR(5Bycq>hl-)PQ=c91Qf6w^4?@E&2!2&etJ8=R`>YfFYhE=5h#flOSe#~7BwQ(_*h)Rt|H51hcP93zquW~L#S=t z8ff-744c;=U7mjwB+Cbg{JJmXrg{eR>V^jXDm#RAFXHnPv9RPG`&c~n z#Xj=XY9UijmNlK zcj6jva__z;)BgEM=pwTTQ$B0(kHu#E6(4W@XZueoq4SB_Z63}K-f1U^3vW=5A9F;L zrRSr~g-i_pq9xpK_k*d=Sy&t~j(l7k$XV?kPM*0OrTvA;^ud&Lw5*on)6D(BSD~JJ zwk?2^RI8Jm^|y$|IveiXXG7dsT0zItY^H3;2ENQohTlKigBD6(=k-R}@eflM^CQ?v zBy#O38oN)AT*&JH8L5>=9v)51 z+HY}xT2ANBHhCv>gpI&9vj~O--X{}MX2QkiYcZ@{ShGL&fXC$w{FZhfmyQ@&&=xqX zfU9rjukK$%-n!r73Y<*^MsFz>9G^gB;FbGLht#HK4s@BymRO=AOT#?nb_cAYkxqpHLn zdvXKb{FQ=R`&mBLFO!eBQ_8>E8BH4!9q`-~#4`@ZAe){d{>vut(KUG_(rJ*+lU1N+ zb-hI$uP)PuikA$@T*>r}ETnr}{qdE(ku}o?nWLhdf zscQl~Vp~NIEt*BEeJt_g>y=ROBMdgI^MKWxd}-i9B@~^>z*nIo$=tnVI6`|VFSoxL z#}(N^o#Q5Y^jQKYWncyoIx%qgPa`pkYbLud9fpK1b$a#FOwlL#7hKl(x8PQ`1Yn$) zP9C#|wGbSs7jNmXm-RwvmibT8bmBVX{3ijgy>AlZ#|p$SsgP6reUlmYTw9#HB$l-i z-bWkX{)O;>i`LFg5 zBzU3&?lIpk>9W_}-vgU#-lR~vpQz>a(t+hxT&b6}-~xQb>Hp<8SG(Am^~e5@$x=Vz zMT#?9+_RD$^Fa9D*Un`pJ|4o_ol1i)zZ_A^#M7eOHhJM07{Oi%v|_Wn7qHhh)RE3- zCL}lgF0q*54l{k~XxxxJbh^ln-5y-oCR|9=#FV5ooS=z#4kzQ?<*pm{e7eW7P2neZ^MFQd-NFZODYR@ zk<^X>TAz^v>EYYJDqaMJ%d*Zj{^0t0JsTyE zR7RTFi{d2pLBdy;79LXNMySrh8qSyo`S z%i$7VAx2fdG9l0o|vqPNbM ze0w6tC$eX8*}PArsWXnbqqLgSmpn!1y^+QVi#~ysj|AKK=r7Z^VG_O=StT-dy~S1Z zUMDB-g~GIlhoSoRYj|UJ86w=&`FH}RZs4;dDol<=s zD~r77shEjGd+ZWOxc3_{qYe%Xmu9VM3aMi6FHB37boHO!uMv)N)=UP1z*m zsv_qwCibhK_3d>$cix`g_oM|3Hf)BoeF`97Ax+nO%cM@@w8SnGN3iK{4B&r?&ch$8 z_l@KBNEsn}Wj4r&=iJw6DJ?}2ijvk>8ls_+lo=XCq$o4Agp}vp*D0frq9szIv{bZ* zc7Esg4|u)saNqZJeLnAZLXtB7^}YnWT2fEGMJwPGe-UcMi>bj<2WHDnCxP4llRmRQ zLUJW{l9Q5i$vJTpx#H~2ge&f4eqFF8bq5o1L}WZJIOYUJ=Z>SYWD&YL7sKy_BM~uR`_v$y?Kn{t_U~6?^dP500_utn(gPJUaAf#8w0{2> zubC%dgev0Y$&2~c)_?fB-wKW~@npxl5^{Iq1F$Li4SA*2uu{1kB1`mOz;ziE*>JGm z{t7eINgX?`-DO-4&EeEs(x~aELRvUa21i`~%OvVwsJc*4$QW*aM~~-96R#;I7(aJ1 z_vw)dJf9KlpTy z9;|3}f@P;)2`GhJ&gEGQzTfK0~x$8eews!0#@*D#;aj9HeZLW|b~(1|+7@XM27{AB?la{h=cPI_Dc=8++wb~6gN zC5uR(#eOFF@mkRMXANVy%b>kYR^UnOW@i5BV^YTWpcLJQ-~ZH-y}phR|h_bz@v< z>V-2*dRQGC`*8^9k~|XG*+NZUZl)X9GNy4w0?F%KDzFG#$DgtKk1N%k4-P%g!28ZK zZdjot1Z!L(e>Z1>W$#4pv`I3IZY&2yftA2?%0Sz7b@qwSUrSGk1;^N0vS;cCYImcI zKI-ZgRjc<>(V`WwYIP{t;<=LS5u1`T7vbaK_p9qvV_^ig>>4J1`#VhNPif@zI<7$eEtfX-6XovolGUR(l5Yb5*B?COg}8H_U_aR-dSvuzz@bnPXHRHxfF(ft;}Q6`S4v#BKVaPTG|s ziEerM#j4CFB9|~-(K4TzP7RapUrc*kFK%hJf|`jN$@MKS#MhDtjh>N3jmtlCQyWH5 zslZ0YHN%qY^fkteqcL>XVHu{&;*Ln7)Rop2Euh`KyUEz3Rxt5d!2G*2k{dZPlQAed z$DDpY0zc*0;B#RQ=;IQMr*}Bux>;#-ZTnbaY5kIU(Rz^F92LzR81KmZHu=asm^h1H zBQQhVcV(jeDPOL8QxkXaV*}^^eJcIdDnq+6HgF}8d&vvyF*NYA0=j(Z7Eier!YOXQ z#P~OlVw6T`Ukp4NDLOkd0ME}lO65#K>7}0)c>HP{E|3#pTm3<@>S!X-5q{lijUf|^FWIzx1(R6%oAdfQ4$iH<34UX%K=VIgP8S+O|JEgmo!w`M-z7h0qMxYZm#MaR ztnDh7uKJxCY8&IZeezVsX#*(F;^F9=R`Ge)S|YLkFPXAfxLBGfjotsGoX&fY|Egd4YO$` zf52(di$DsbgOmIePi+c-&zjVT_P-A@GK<4$sP8h=JLrc}t-Qd3w4^&G#Zqe-8F=_> zGD!X~guf=1baQbbX*zob?yIlI%6)}o<6C*Yc(D}U9F>4Of0$#qr5^u1;yS$&rVqv3 zNYu)g6!?F&i74OioZF67ugyd_bwr8jCRoDcXb^*BW(y1@E?ree@7DFbcH%DPx z^lN6<)EcgOP#G)bpE{*XW3WTuJ)fCoh~8q-sIGeU#zf zpS8vB3-tNvU6;vpIcbpTmxat#!LYd4fCS{RXtm7)J$|;*qt8_6#t(DwWXxsi7V(jW zKaIv|KQeICjU+n$*;)E^!8-IA)`MP;cueUF#URT|)cfjT9Q*D#f~$n^HW-Tj$thHD`BGhkNd`nwuN` zifP%~DDtVTWok#f=avL^aIU-DF>%jZde6L2B;$;-@Jeb8S?TNn6N@rAX6{*X=aCfA zp3Kudj|<4NC)to#Z39IcJHY$?WpEuc4$Z{TT$X)GRr#k_d?9Iy{SO+swPE}5?>A!{ zuDps-+LZ;)t7@6)>(a=rfX(31q=U0xMWK4G6_tKC7QT=5f`Uu?uws1;=co3Mti4%* z|7>E3+VVr_aNvz-9tNu$xTnJqrsw_hOplE?7Sh zg+adtF*ZZs{f=g^U49Wf7;+jqPdL#rHWA_%VjZq+s4Tc{yUJV;7=_d7ZjfLfA6QhW z#BRTA!k|+090fRW(52u@jv%+=ONYe5B?6Ml?d@4cc8c0h@X`INzsBR4$#Q zXa5VLzQ2<&x3v@{JcM4CYYeqC&4iCh!_hs+mBf$Q$CdU@fPV=|qHu@gw1XMut-^nqW6DPrZs5y%!uZvK|3Uuu2jc4W zmbAy_W2pIGOtATf*E;0cF&CvFx7Y-pDOidkoXeTaWwLN5a3P6KvB%Mix8Pdl8S{Pi zL%M5PICCsmkL?=c0n;Nc(iz?3aFm-9GOfNoJv;!GW8Ghy66Lh+b6Neb?iz@s$*|AUQKRGK< zv(d-rS^>DDz8edxTv6Zg44Q1)Odj~J0v~l#m>b)SA%7O5Z1!8`OUDF$sMwL0{Gh|n z-!hSUDwWaEDgwK6`30OCdj-#E*kaIr%60qL;`rF%v?rOzrK+ds;oEQMy{AKP&tEUP zF!doPVW`0Gd3J*QsLCN8UhcejUkU}iDA2Df!8}Q4a%I~PsQ&L14!hcl8~e`S`K0mi zbwn|VT)u@^vI>c$XeX_|+De0_rK4(bCVKr&743_>Dq`19rO%TzG5^P5(qK6rtBv2H z@3>ifo55s0Q|UQQ8#9tmG(SW;sK1@y90?)RzfZV+K zpwhbo*M?5UmFI16+6hJeS3?VF;wji$=x^%{-~WI@>>o06aOCIx%Ha7G&RQ6ZU z^5s1jZ^EDE!}#X=2K+*Ag2vq%=Qane4_>4Kvq{_YZr)|HW&VVHpoo-o{{{!7$!v zYysqY>p)xbZ@O*d0_^p);jf#0z|^BYsG{KuOa4mH1M`Du?a;YouZtWpiBd)9Lqo_9 zn+tS~r$5|%^#o&*^o2~e3?995orHhO<-Ch$k!wsU$&PtMw%D8V)Akl~r=Cl&7w-HA zO6y-x&7jfzH)&zd2i=^U=?M1J7fZ6*IUYKzmH-KtW7SNI*x$Rq!apYscKv$`2${bX znqPO*8y_3s;nI7MCG^P`ROXO^qa$!t_<(qOO9Ze_mz<`l%DHsXVjEPWrDSc22}Dk|gx=BsBKd4DywMqkQnf+!_xdSt z=7loeTi6J`JJt$WuybsnasfLzfv_7#>l5wWtHJN8;NCM>3L7`w5qJB2BxhrwWVtLhbU=d;q1)Dla z+QFf$l9(qY!E-T4s!rtf$PrS+%|vpp1gmo4EX?`S0fS0vY=(;^dv@X`wqRc@IQ>&* zYc|QVs{;f!e5M;%3M^|&B}@2Xw}UzaH86{I39P^^HW%+@u9tm)uaTtDz{101MfMPoGG54y&J3c4vlfx_5gxqC zumRk?KNs#TDq_ku8WVoaRr+#YCG{uTV44v{d|ecHi^dYQcnGWe4%EY5g3YT7*1yMr3ziZAETGOdco~C~n-j4}Te&5a?1LB*Sa?F|AKOwO)42+|X!u?Hta1(s#_Rs`cb^AFKohQU~zl->Fc`S^c z7@Czpx@H65acJLjD&d?Ny(MR^MOanou}=j`^H2#FJp_dYtwN@CZM_ELpb>-UEph0;A1t6<0r}0P#ZRFiy0gI zAPMZaG+Mn-4i9?Of_6q2>6Y+9voRxafIUiXClLVE)jUwFrL3^~m@O zT;(d5{^SlhQZiqB&_ou?5{tzbMoQpkR)Z!-nz7>_*T4vwIFNOnN~rN_GPB?qY20Q; z+sAI4AR?)>4$9oW9H2^|&2^K!HW?|qA- zs&D4uv|L9zwasX}M5sKl<4$3Kk`n)NQWH~f$CyTpH3H?8aj0ohhPG@1b6}ko_4T?yPjBAFjn4YzSasnlE@>N&MVmt@qbk9ExZp$P=U&Bj zZCSSRMguGWZ)UYx027BsI2`+t$*p_;OS-5+GHqNPxvBwb>s-+ zt+s`$>vU&2{7BWEeUdbB-gI1Rn~Rs1U!=_u#~AOEmuau;C771u4{|B8EU$kOG=zD~ z%IaV|Wo^tye%XzAF}qPR&79f{I5T0XlH5<1;q-%fKYhC(USxJL8jR)dl4bV4#M>V{ z6@M~UrvI#LMcpla%#%NMVvV#X+{^1#Xkw$yzgcI&U$!~|3CHh}mS0l5$+VjwO8v($ zqD)ey7lX&!G}&$0YgzYBAIwiF#1psfab3)Q+Omys2d8UuJ2aM&orN0UE$IxevvlBb zdm>DpBK*HQ7tCh;8p(bv_X5>iPa3mD2BH-TNytiXzU|so&M3zglS7_R@-CVBqcv@@ zeMmC`v^o92_f?G+Lm?}(iO5b{gIoF@k`0$`!MeFcAp5Er-S>#l{-Z3G8;UtYhd1QW z(fv@Jq{6P>Zo=9`0wl@}l1_EOTd-Z7Z6BH|ZgRW<@6t=y}M@dykG*i{{%DPr#8a)T2-8^A4Bi8|0OY{QE=|?C73<* z7|H!bK)+lb;#W)&xT*kw?pG*InaNIhJc@lWU`LkMxv)~>q}h^z+Ylf-j!~$n6u59M z>;O$9)3(>bweObfr*%Q>ocXia{B2X&jMi>e*3O!>Fp>o~|NG3bGiA)iZ@`?oY{N?> zhCz%;F8d&SAK3350F?t4?37$pHc7IYtiZF}UWsI<{Wo@T{BkMcJMlK_S3udf(cSE7 zE5aJaXR)0o?rdfCHR95g29~2|vpJ2iG>EFvfJf3)ZnPCKw0kTj`S<9NkR`;lOPw?= zExUMOeF+@hc#W|WW`mI@4RL&2{e|sXKbU)t)~v(s7hL6@uiQeNy8@G26~{CiVVu$h zft5Fm-Cd_gb*pv2M1DG$7LB5{6Qf9UMhiXkuaa_eV}W0}5bjR<0y4V6qDAUO%#;)E z)Nf}7j9C4FY|1f)k>lOjj`u$7oxjh(;YA!KwA`UFhAdvY^AF@By1}5`iq+WmAM7$w zWd}y+u}$lwAn?&sX6c!^%y&d`WJ(66kK3& z<}Vm;HkEyDn@n3yyMS?^61%uS2X1+k12|NGk&6~)e3OMSyE zn{X>Xk`&$YVwP4cgwS=~xKToI3*R(@Kgr`^`6zXq`{ZuB9`5i9ux*m-I`Mo;I-9GkuP&2}nNayNz>+GPhRaf)CtyNyZ= z5g;Im#h`myo~=^U=gX@1^6h@Whk6zBfu<3>hTKSg$#4md9n!&hT4>=WhXi8(v=Y>v z??SF*I3;dx=qQ0dvduZ2&KaS?-?!R{nLj=0s8S93PSQbKIQak=K5F9z@}u#Rh8}bj zH`A4Q&vDi9gV=KUEbiXXi{l*jGWLsg=##%E>EbUIY@eYnByTF>?;B~cUn(-eziK1h zeWL<)HAKRySJk9cc|ZAiOB(!CG@<%yD{Xc%MX$-JqA!`dxht!U=+8n!QG|IEW>86f z?I96ALh=gz8SqSGo!@jKK@1jKO5bXKs1`8G|@XHnC`5zi%Q0LJp zYJJ0oULa-SniN?S#kyelfj&`T=QuWE*f92PQX*JnH$ZI9O2~M1obgY$#xWs!T+-_E zB8%%WqTI4)_1E`y)jswB!@nKkuPG)dJMZ2FZpCE52TcXS17ewj&f5iS%K1rEZK-=FG&>go& z(ot@?w0U9*o#$dDy6PA%_IY`a@$Apy65TIx5qd4mi2W>BJlKrYeL@~#qZtO@oy_H% z-r#=T{lv-nuLWXup(-n-hDe$zk&mC=k_~-hnTKbjSOc|Ea1(fd2G0+0EB_tEphI`b z=xTc|Z<+!GTjapmODZtzK@X9Kvvgy~9^j21g7P9$R%bFt)bg*3j+SPKUk!E=gZGZK zE%+n(k~@tl?F0DsK=AP@FQsEY+hhOZzs$zqmt4n$b7YnfN!+~h84WAEhO9(9M&^AX z;iKj;=`jni+Fughc#63F`jeT<#w$tJ);5NDKMn$;oS^scRFQS+L5fd%$g450$%XdG zXdv8Q8W(Q>DV-_Mm$scOX&j42mRGrb?McKUFqNGDPX^w^ZimZRj*z>ffSTfWl$KHD z6{q&%Z3{_$!7DBPQp#4&|7rkiEbkRuWk+D5`(-K_x0<{TI8|lU8pZwXoGu>u^fnF3 z{K~9b?t#YzcJ&8=+2&vH7th(2piNXBPTn|)?@Bi1l|Sdh^tqidM(H4|ay|e7RmJ4; zAz>bGpH1RU%CJr+<=8@rE~1pyNp9qwrlXH}iIeUWiGuQKx%GIKC<|{c9j85vM@by{ ze4~Vj0#8Eb_Y9}?ABTjd#I)CttI{(9bJbz{7D^zT{i7P96 zvC%-0wJCf6zTYU!u(AZ%^f7R&KNPLChr$SXcjmmOEzM1vOY=|sB-6(%fYgVD#5Vjh z853U!iq0u;EN>BP7_JDmF_a|Fs^e_qtf{+`F@CO@LkHyKXjr2+=tsVW4)+(ZOEn)} z{n23WJ&k3HlGd^5L#D8j)AxWxLAk5lnwQ`Rp5~_tqlFKS(?gh87Fq_ zzXQiVsj+5Ks$WU<>VEowFW_Ph?PG3#l0@yFL-3|aA61iC4!3G$*feK7R@!S4xzxRmW-sst zr&mTGn_2>vafP(Y;Q-G66NJ-`E=H9}x~$Q3U-teo!M&r+lY4muoW~j9jNMMrx_WtB zv5#{qJQD^v!D8|!*bUBx&WA6(2w6G;FIc;dn{xk&*#G?_k=|A_Fs>>m+g>cjhp8;H zc^d&GiP6ygdnk#X?g=uDkBFa;vv5B%8yY;DVdhpH8eM&Y{3~`LB`u2FfUGw?^=}BV zX}AoDqYc3?rk4vJP-inkX0T7Q9N6%|TafM;011E8VB6`A!{-a#$xb7!{*>{UEKIxCAELLQlKT}@Ly|8QP*}c|x>#w@ z5qSa^bLVUB_}3?7aNB=ackPOJ=l*5XwQ@f!6dcvIFV@2_=DyhfO&|>mkVdPA*|2r( zenys?g|k$PFvTnf*BdUSw>5x{b67*vmg)ezs|2IejV1VB(nXxNRRi^|R6)}`Np@mE21FsYc{1amS3HGh;2E`mBCWK0I znhjza5pjc#_Pj{l@@vG_KaJ>tyQj(ezv)nRs~9^Any_e0C6##RR&}vIUc6(^C&xQu z$MR<<{lX04eiWbZno*pj1s`u@Lcx=IIwxHLzfGITSzgkk-wqw7ProM9hnb$-$<2kF z@59-&$V7=6hbW1sm1xk=3&u?L9ewhk#svAB@-Qab0HYKi)3EPjN%*^Za>QsA#_c2&(dxqGjhKCA#*7--?8Vi z4|N(@Ba&K}b1_%*MAggP)pU!kEwz8~PW1JW6+D?Tf?PT3fh|ue@I0n;P&Z_8_P0U@oc2=W-5}l!-(c5lzV#*drGosbG{Gv zd((Onu98W&&nl+(Z#&Upi+#DCymCf2CY7##A&=P_GI&h7oOz_3ESk|+kG~~`;7Ro& z+;Rc%UKoYyWh22@sSWfsCFq9i6fUmL2@|^eX?nd5wu_oj%J(o>R;YlJPagG@D@NV4 zTKdBnm**Oi&PIy4Z(=o6~{v}4I z^)Px>b>yVgd@|0xmAovtBlICx5e?UmWU7}oH!5)&zeC%V*KYRW&3~;Ud#*czM;k*< z8>SGM%*)LCQ^8O_shttmCXmNp4v1Za4(f!#CeSk)OU{qqr9mvnh z9aK8QgWh-hy8Utl$eTy zj=u?E<5jk?dsIfUxr=$mXu9C{N%au8(Wcz3FM?xCD;94IpCfSSkWSW&!918=iql-ys%WQfzv~cW=muZ3dbxUKUsN+lBcGNefTb%7 z!YG9~5}nPdeKw+%D~EHkhZt_7rUp0q+ajoHYOC_$4v=Arv5f7-b2O}bG)i3VqZzt$ zAuu_Q)Yt!IisXl&Mywn@2yP}je5Wu5CYj8wmc3ATuZzg4N|I}Z*HP}32_NBV#eZsg z&V(nPA`%uI%;wC0_*QV6m#u!r#akW0xounU^~g~O@5k{T6Rr9CBb@NmnhiMMV9Y!J zd5MoB-SMa7CM;luIl@DGdUv!HLdz|@KU9wI4GG|%*UsnDMnuu){$=!%n-rDznn~@~ zUS=*Wn22=;L#fMChNe9>r|JzQbj^dyv}uhU{;WPEelS55j;$LY2{w<#a~cJ{Nb^Hl z?RH{(s*xG zJ=v^h2W=S^I77-Ew=Fornfm;PpQkyYUGQCjFaDLf^!n0bvrsxXwwt@*SI4!T8K4rA zoyqFKSQ>Kq82)!sfy%G7!~9W_G<#AO$u@eyTu@g+E3JAay~%_-B)(Twr{;%iQ$+OC zOjhW8$AS8=6>!<8ghaL&GNXlkV`oYXRll+iSEPIL9(y+OT8={)n;8qm+=3pev-Se5 z9r;s~%Bp>itoSkl`J*;;1uCOBEROTI-^`7CrbDbP z74X-hr;NYnLe6i4Hq}qMROL906+dn8XAE@aa?^(o!@ODZ_=rY*ew$1J+WCIsZd(^| zpF_@7S!~c3{py*|?Kq|{`u=egJ`G64HixySkgx}R9!b!Y75?OL;viMhUQWNNOOc0l zQ^+MgpDAKOh_}!cX;2wM4q+WnlS zO?>{zpPXzufh9}FQIAVTlq}8xlNv|Jj=TzP&exgWaTCa3Nf9lYtIL(KE>f+dy~p zVG@zsOvy8EEJU-R$ZsxcU+>YUDfgB zhGH`NqL?fX>R^mldBC1e%OG#s6mI+CBFYMW@)|#B2xxtS3P&|j%`^c+?^lQ-5*r+Y z9-EOq-H+VVTzOizYlOLS{Ra>J&PYmvhO405isV z>^3^|?oyCdyCq)z#R|f?;jmGrgB!O`9`Y4R$f&Lynks(J^agJrD{enxhW*z~?YsAI zXPlOk?iT`!$Dp3<==P?kjTS=E3vXN|I53s;1pif86#m zKxDE)$PhV({m>_3yN?*NXYa7IuiF}bce~)+i`IPT(+iku>yO6nCa5g2hhF`U;*3>K ziNvEX#2`unrYwC%)~Ymdi>w$LW+!+F--luU)vt{A`U7O>;#^c2Yl=T&Ha$eo^xMMX}I_ro?e|;Rb zo|8r;>r+H>=^v3eX;8dHaI}0A%hS+WYb=sGh}A(GP3!>ZoOs$t;26boO%>I&__7EUK3_(!rNR8_ ze-H74K?2Dd(n_ZO%^^zfuW=c(hTu%Wm-1-iKV1Azl^)*zK_nx8SJc;Oj_n?|xw*4% z;h%rzG*{68w+tfm_Kf9s=GWu59~F>Y_n4e2z0RF`Xh^S>Iq>59*YK2cHYZk0L&Iw) zv44&h9&)<}SBLwKwF>Jb@%?~=hO zf%mKh;yFtCbgXO-H_WnH#Lm1T4qK>7#va$g!lgga@L?U6o_dI?+F9g><5INTB{+u# zFUaQtC7fO|gZVKxlMHN{3wN1&Bs*jT3`qPX$2-%jUT(Nd$GHa3xBX?}SZ`zes49uk zVMn+vtE4bhDH-3dYof-9<1Sw1qMc4f?PX>u?_lohogf+U8;SCtP};HBfkc{x{6D*f z%fpQ^TiE%VH2x%&%LloV5tmTax0g8l+DE1@^&&rZ8u7y-{$i`)VVv7bscuLvUApoR zZawji8#(YB0a>Tx26w^5Lx#*wk5=Kt&p0pcC zk=1&_d&9?&pRwQvTH`D<+&W2M#4hJgZn(qkUt9*4B}dYOCa*;*OYbw^Gyagh(qdZbztBpgyjK{(yUF7|OXd?OD zlsPr4U$lI&z?EMXPk)t1WAFL_I;ryq-TGP^Cp;S<7v0uE4z7#e8tNtqcL-Sg1``yq9d;MppuY(-J{S!jwWc)V7obZ=-Dok^Xe}1 z>%&j7!ZMs1&GFjM;c>2Ey z{0{$deAURexcDuD<5sMq_KCrCyzdn*8jj$4(uMEcr8(ubJ&r%xUUO;-BKfOYnY{M< zdHkWwYG%-`hUV0K$5j)rVuxipHoxqqJD>XT*~ytWim9e=(hT{hy3+hJg@bswBpJ*k zhe7Z$D>CYSBrWNA<czZsPPQV6^Bo{=6*55AC-X{a_jy{o+cpYMe3bs41Wk`|jfo zFEtFzji*YMqnLMphS5!TD(IEhZ7^P@9Mj_N<4?OYSn;5UoDfgDO(6 zUi$=&$STFQ8ZCV5GzCZ3?WH|W2#z7Axi4$gF>0JTIoQt7iCv43{@p7!R~d@7+rQCP zt!hT~k})*iDkAgV9;TxrXY_;Jo<_!||*?e0l{UmI4zO2sTPi_77@4L78V)Z3^?;U>}jmd|uF%fN(=cchQJ zAx*+wtiiEC$d*p#b0o*(_&goFvH3Qsari*luszf@Fq976rAVU-EZ~BW=WcR7k0I=5 z8lRs)qfMIWq9bwCVRJVJ=CNedvEO3x`am*zswzx*e~p?v|0(*OYKy_^%IL+<){KvT z2d&xpkeORMlah;v(O>O0E}r)c_wLf?JyQ4Lm@|#^v)@5R>t_o$XkJD39dm%CpN2t~ zO0tk2IDoAGOkC}0$xV9R2vZ^&NU%*Z1`ip-F50t(?cUV_&nq0@zXcr3GL~c${*9(F zOZ@16UGwO*C>Q+17SN*a2eD%KHCQT1u~OZUSKi~n?pV4S(wv5o{nx6A?phTxDSEDO zM_r1;vb zGZ$+XiR3snMnkTW{IzQ(jz!u~r8!z~ImwWYZgo0bO$RPkx00}NgVZGNwn+BCFmifR zG(GZ28#((Gn4O#<9=V%hlR^cK*8RXSH6uV<_7(A)k;5GIFDL3rV)zAbMRHCjVEny9 z@I-wayD2;v^a=qwS=^RNie>}-Y2xyK=SQUWwf`Ossh+Q_rjx5Y_q z{^-m0!LU(#pxi`(RV#fio*6v?4&7P@m#>_5l3Hj_mlW;h{Jzb_XsL9xkBX+!ZHMX4 zws@-Y_A*`4wg6h}0%5q_45B#ep}2jyIuouLOpiX%#J67nv>)b>B~jiuY>MFQRB6XQ zd&24W=Q5D|TM{1J6p_SWOX6GQOdEBYVd{N1c0^zb8|R}9QOyw$Hm;VWzZiom7v^Kx zg2S|`=`&Ti7E3Gav#|a0K04v32dr^VgumYidu^x@+uP>{s|1!@Y+wvZZtJFhCeOw~ zuPhQUD-fGUIO0RE1;Sa;ig(@QXvxens{M60ycJvp*WM>`^*c?u9Q2_N&l@tPzr#WC zm=7Mwal%#JW1#$+CU1DD6s@EyXoz#5`26!q`ik!&n!eWTB*#sx>(Ke^7v?Y?emme0hNA!R6`-i*I*PhcX~X+Ueaz#^S`06QY&QLb_gN{_H7PkWsh^9vFb7VCiu z;|@(;$kprF!jcE(0MwhZ`4gbraSc3cNFmGLz95q6`p|Lx9K1jD4U)eJ9P-6$G2^fw zM!qe^yw*VMy01X00am8SqxN3bLg0$i5p#VD`CmIJCoKF$tgF5(SMeM(NBG zvUpeyW9T^qb{szrlH>JYa8?dk-ds(5ZikT9C&$ySleKVl@lmMWwuf|AW{}X~!^pWw zd5pKgBYHZsoA5JLsm0mpD2`L$Y}cog?qRcpH-k5Ik+@CTpTx1N*G*lG_abhzJ~p?}Q6Ye@OME>(u__SgadslB84$ATEyhR_?s7Ll;+p3F4iei+#_offO9 z&_ny>`Qcmb_yJiLUbEkwzq7=j??=FpU8elN`}1`6feuF5RuR2`flH-AMsTDrO{cH=a4jo18A>y;3{GT@Na`(ZLP0CC3cku6#;Hm3c7h9Vei%-+sJ$ zGZlM67V#TDPT*ss9^2oX|bXjGy>X;Ntr%_JJmUh7aYL=-7wBy)ul!pGR%-@os@zsKX= z^LU)UTj$X^d#}CTuh;XrLq{8HHJXcl^v)91j$0!N`w>f1GU`Y>i6#FU$^|o03x$6- zEJK?r-_z}h*F^jU|Kef^T~VX9wJ7ZA4AFAgljQZ%Aoc6rarwIr1 zIccdoc!np>uQ$j-vu2$n=Xc*Ema3j2{6|_8(32#x=x?KE-*j{@EYPnn-$}_7Dg2(F zp>A?YMCo(9aZ&IhrZ6>H_-r6a_#!L@tb;4Tqe%~<*Qkm%oRt<$+V_$A;23&)uY#!0 zbdt!_E}Gh$zr_7=y+Gr(@VqGAHPZR_BE8Y7DO!}zh>Y9k(;K0>bl4P#=by{i#6Ogb zG^AsX)FyKOtSMa*bVoS(_A~r5XfHicTuL`ReMU7BKH}MTa^bo6N_2_I$ARJV=(-n` zq*mXBPMt7@u8;G^squ&D^ux;;ck|=eKy^PZyx79M-E)-tJHCnpww%`tlo6nUm24Lewt(Pnyo;Iz?{_1%3>zI?ISH%mijn3i#D5F zrXkyBhz^|oO5fl8M5V38Xl26{;qRc=$j_PuyyF057kk06*(KDLh2PO{pM5x-$$_VbW5A|2mK@z0 zjgLQ&gHO?gxcQ41Xoshgpj0F3XcCVO&wGL=D}KSUju^5ucfm(5gg;KVV)eZuz+Gx1 zi&cx@&_W?(&vb%1-ka3PbJm*otBTI`&LxiBvvH#TQBJNljyz1$#W%m+N6bJFT$Q*4 zmo*vKK1YY%y-4ws|9-$$lt=cKIMD{xC*;NO0&HL6kKGLA@rlyglnXsat1fDaCM{ou zZ}MKFhH)2R?yF+zx>1sM6x+awWv8&$#Uxy$sY$1F28u@Kn~1`+`pE8`y*TE_Y%=L` zIVl`QL}d%*q)eakT2ko@otYpLq6z1eC0OjBj4hB{n_?-rm#$ zebGYB^ZP{TNazHqH+!H-Wr64jQz*KYDlH0eDiE3E7l`zSB1ALe|I(K652@*|KpOlY zfKz*`Mi<1_qKk4{g|-t@@J9t%dgEjY+({dz+10y5^2#wNw`54zyHQUtQEL}If1#aI zlG+bNM|R>-1v`oCd|7jnMuMN(uRI!4r}Jcr>e3FO^2hhLlZ<5?j|A?Q(Ms%K0|IPo zJ}_B)PlA4`;6`GrX{kyxWDc}|M6f2?99B)U+s29J4cw)69iQn`ehzl7^*Pl^{v=G6 z8$>5ek~!=4KsxB2K=(;kacBRv!KaIQ@a}l(x9K)r9J}>)wH60 z^EL3e+0*gN)XlIV^%R=Gc~TFuOti>)GHqHQMZFh|kg31JP{&V-?_Sg)g^5!5Zow(I zXVHQRRve~tSGUrZ3FnA%)El%vWFt&}I~f?Mc(MpL(I%Ncv^KDah%NQe^dOp7MpNct4H(whZs*`uyghaz9BMs_^{h*U_3o*7qCyl^JK#zkUEH|HjC-_lza!)|oxraBCL*mk zeNm%(7B%(nMO)4Jjoiy>ylveGIqbl9(V0@L?ev8homs&-tU8Q4HhtkfTXNh1el9mt zIEpIEDpUVG-lS{oN}{n)iBdB@%Ve|=4OHvmoG+uuwYUPj`$QYcpY;Z@|%y}%ydPZ*f23l{)h#6Rdn3UB3eSEsn`1x(4@KqrdM2r;|&rZyL&Xf9nuXC zBVR&iaV*@mlVa~^p5ys;<6v{NA>VUUr5C^Wk&B~?(VTJnXry%jSF2i#BUCQKn-3~v z4JQtk4%bO4?@Kxqtw~2lkC5fAie%U9hgdzu8TLLqj%#1#VAqJ%ka0$mo?bYazR_?X zYaZ5N_p{$|#z_pOpS}@Vei(Kebs$AE{J-Y{?&h^+_&bgld0qmM?MiXcXQ_>7+#KF{ zvx6fBmqugP1OE79b}bId4?>yp9yt8^2se9I3is%61{MqN!VV8jAiOmZzevtTQP0Eq z^A1PeMcv2fd@vW*#GHpG54XX|f1!BJ3-lllQEIY|ciQUKETYv^1#iTV-g!77cY_HpFdo1|3u=BYTfe5#(&S zhh<(w!;p*_)RbmovG>(nTdfvkUL7S;{fc>uhkeIje~~2J_-PECf6t1NQ~D&gSp!^e*@6A9BJeSCWS!Js!CI+b%sRbR zE^-`)rm9)cA^S9xP*IP|`Lm2HXKkEwsS1tsL_%d7e~+Bnh^AKE#(B?7@%Z1oYhHb# zDD#_)=x6LZh7QjV<*pi_E8lC=4Qr0msFVWYTdGgz*c^ca9+TNL-p}3PGhWm)K9zc0 zN~bEvShnv$3Y$dEur~q@(4$F1WVFS1dXRTbW>m~ZeQ{~bgex|**3yjYuMmgCsXN(e z{B`D|_%V`weI|B&F9DaYxYD9TAzsh>5${$+a~)gMp@^UP)@L8*)J{9HYFnyc;SE*j zuRcsx@r;Ib{fFt)7$;HmIeF1lb%IMLNr{HGR8rs37WCXX6DVTI5t`OeKzf zAkWzt;`CmLxXGqLt6@642)DoFI`=O7*%rVn&p z3(nhCMqVkkaD$q^NXZoy=EBQ8q78#cWOU>kRqK&to?eMzYJR^I=*<91sWB4)| z*ZKo1**4;9Jd4EDG!~RPGvEO~zk6vX4k3rq;K=K#(ERBTdC_Kowi>nK@t*`Z`k4kd z!R{}~(aEN@jeF>dH-+@-k|?~pMhEST>!69sm&i1&NbE0G#Mx}g6`3n;6qT2oi#kVS zpkz!fvNFGe!z5nQtwD;S!Qo80!aNmtxhrw3x1vKU^1xhX582%Gm(+^*tV~fd=mp8c z3Ns`0ao7zxhE$;+$u%e{xg8!2%dy?Z0_iS;`&9Vh0fB!kEfKnb-2Hj5Y2N@nwn|JC z8&yJ>^B8NU*FuIw2RO~R&NW1)^PE0;VrLYD*C2Nio|4F2X`KT75C7n}__5fN)glJh zEXaZTYFt=BIQgnJi|5d0#+>IBRtQT%X=*squG1-Sz-4q{NSfE{5g7x2}~Ua z3c*S6@Kqt%qkkCvT0Ra6@*|1nwFJE4%6U|A%9j~ax`_(s+3?Q9>D-I8P9X7=chGN~ zgtRA$;c3d=Fnq=amj3<%H~aD--u*ZC40Omse?$86uro1}5kRBYI?`aX1SHN|acbMr zad~PbT$`K(bg83fCL^Xdyd!J50UV?Cs2E&78hMvN7}_zY1ZBo zM3id+VVP5*?MpgQ&Ces!{bMP2QJgmRN76L^XK0wMARY%7!66$LqCKzzZ3&yK}|zq+t&YaZToVK(WP!1z{;7oNAHn`fFClYJ#dL`DBNzIk;$9h@Oc z;|xksN?|67toh8dAGgqRNvEkG#fI)#brpXexJR~kRsv%fOA4A^;3;0`aP3%4ZjR$h z>`)&IZ!S6#gJ!q-+z~0F+1ZTMZ<~YG6)(Us>6mmxVtuJIIOy^h6jU;k+FJY~cG555 zg;E=6s@Fm|e{L!Nf~5>@QM3OU6FOfnApEqWoziVt8atEc@};1i zc7!a?5GVa67Sz(*h#uPhfuKTVGRt`~`DeR^PRrbmLqC^sGEW%}uiSxaty3_1T8rNq zET*%^q`-eeN)XRK=k_~pqb6QLnw310oXIyPXXL7hpVvaN<(@HJuXKP`B_72wvClA3 zb;eDXrD50a4su^-P0ueC(0CV+g-#j4ejvs zKUeX;r)%-yuynYZ?u)eZHHd0@FY&}sG)i?{Bzvp)?KOSPpR~{ox)X0tL4ueEQ@HJm*gfR{N{Tnsc+* z@pC4#CzN=$%l6;g&xa~#w$m59M6#pqv$zHsBT|B|(U;$9r&3JR)IegfufHUIvT=Nn>a+|tNK`rl|sZY@pnfIJxYNsB8+I>!}(h+fXf8lT5 z=XDbd&L6`U4Qlc9Y&lM`E{Oj7X+-Pz*>|;g6ZdPW3Ar+BG?no)1uf+k?6lIA4G7c+ z*@M4P(u_(j#@CIzecytUjFZQWS_eqO$UU4HHv~-9Pkg`f6n>!%(6!K%)P-x3?9UlO zOSMDrwro8Vp!ab2<4^AT=tJDhzD|zOSHN4>^R99&AyrfO4)k<$s#pgo2@xv7I>cSaZ|EV4dGsM|ktvam8dZB;I^dV5}9tIPD|*zva=9VBQV3$PdoOPX@=(d_4BP09(=`8vV@y zsr1=$RswsDSsGb)a#b#OQm8?f{+mTtZx7;%j>i$!S}_mS)ymFT-)k3gp7H;!Jb zgU2f#0*yL%y!w+0W>RjWo?GhdJ#%2knQvnM>yU(&#di3T5C5I_ECloDG@_VUP3@N# z(Nk-lla$wJsrvB8?OY$F>}bX<7)6`K_tu< zyan|dFLvI$7Fe*3cUtez=I%Z0N0pEQOX(ZN^v^I;;x`S4N4ntQt|sn|RRw;rV-WS} zm7%GR#*;0-W-!}$u1TBC2|vHKuyimD=6kE8LppB|zMF}Wh+c_kwJ)qvHyF2!nnWI*`$3oVba6Xh}Y!FFB)!~xho13cMNTFec+U1DqTaWT&tTrcRA>kf zzum)SL&hRg>lakb`X=S)12i)40m<~3LNopj(1Tu1^zO(<`ZrVu-PT0J;T#_XL)TNbUOp7B6uF26)~t?H%4kB`zg&y!sL z0TI~~d5N?y3}jc~BsS=78oR@{gUWn;faT`u;loYtG;WbIJw4Y^6kf4al$uZ|a`&zk z_3ewM>yF2YE=-y$V#|}sKGnS>;@MU5ptcb)i(Yb5v;Gs79m)sC^qWv_Y6>;WY>@2K zV$NsJPeGaIMA|f2k#u`YO(qvsMi<5Ze&B^Tzk>$QSa#Zo^B$p zq)ew8?1tx899ds)S-Q;VF6}<(B>KGJh|p~18?>P4Ea|B_h1tJpc!Ty2c)hHbCa8X( zYHnJh8?8^EXfMxXFw|fVrhcT$$DSb9l?u@5s%j{`B+DLEHvnNrIW;;lLN7-S(B7Ox z%E@qK{^thDsyagO`2u(-7X=FcblG*&J0V(UKW#mj0$yu6x@VsJcpsTw+VeqPr!9eQs8S^NUlEJN(=Tzll8GW zLX|%W*vrZkd$&dLePUf0`%9kPC%X*pJT_q+d?&GYeVy68Q^ncNl_QWKCCz8n73tub zi8N5^4OWhb#2b|PS?|1RSYN*vj z-1VI2m8@jGC|VV;>TE5mxH8?mjv1F+lC8gza?gO<)(o)d8jo~&QWDm&M*Hm@E+?w?dh zU3wZ`Lq4B#qe@#7)^eq;ft^379b`I)Jt+giYV zvaliCf10E${1sF@H(_^1OklM{671Z=TcK;#W{|IrfSq@mVbU!=>Qz^a)aQP|4>IpV z(8xs?|M4{Z^1q7@tXvKg)+LZppBK{g=g%`*);DnFfdrG<$McoAFeg;5x`0?#R&iXPEZgTck5yaU4CDG# z(eS1u9Ideq)K^Kc#-tS*?sS99q-Aj6(KHB{R}XK}AL8mO+XT02HlVO9M_9Gt0)8To z!Ax!?NNAM`hfQXag_*P1Jt`?MB|Q!EBTb?V9(ejZT~6wL_&4_Bzx_gRNaOvO625e3e4nsWivgBra=$%%o+M2S z#&ZHD@Qw*XXe4}9=f@dho+k}1+M{6ZqN}LJtpqC_$fcS2M`*#Ai*y%@@qM2|+-FH! z2+hyMx~Jws!Gi8+khGP1bH%Tq*)6<*TsZ>>Zx`|W7rn9rpBQ8w8fwS!xB zu!-}V^+4;ml)vQ zLtk*Yl@>0O*hoCzjE64^YO&V}6|VdHFjapuEKK_rh%Zdm#tt=qaG~FHEPJ#T>lQE5c*M za@l!0X>K|Sm%m5ukQ!!NWj6A@#PhngT*MMx+rZto5d7p10XM~#YK1lNtVJtsjU1x) z^vBYM4K>vCO#)p|WKZwO?;slti-ePWci`uzrT~w6hSX*D(82TyukBd`yAC+P)6cH( zbMbjRQ++xdnI}VTty)W+mm5);EkooVe>M@P&_dYaeQ2P!1QyP9V2ABn@se$JY-2$? ztg4KF(%mV%n>dDLwVm`yf&$Soj>o6f9r5m~YnZcdACaOzEwr~~1zc#k zin;g8g!5(U$>LIZPVJm79T^fMabC6DwLgXQdBl0R<=RXN8>OiHnp46V_rtMj^&i6J zOoS^_O0lz&6kfJ;I`i<7JoocC;o`LRh#p3a6MdW5M?VbjqGMjD(U1Ex=#tekjQ{9W z^w+CRV%dpKn#(*ZcT}s zDJ#P&u?vtcE8n(+jM_PaxLDVc3~h5nXFp5%tpAtsBz~DSZMs`VM)S@bd2J#0{@Qz?Qt~(9(~?ajNj_3I@#<-A zY;QEI8+C~1SEZ7@JLO5?;WApl&lo!^F5!bmv+ydX$J`zXIXXUTF-n;&#TFjh42!pA zVs%ytQg$TqjG4dizORPH1!{_>+FqmU{brEmLU&qXUQOG3d48nl7s6CrBr6Ick!n~Y z=eK%T5d7PQb978EvrejDUYFJIT!be#Blh!cEjX2t)n$bc?aRUy zN66iK-jo}fMqdYQCVApF=! zd+`pc6d6Z4pBd9bA9m2!QO2}TyO&$_R+bvKXVCSB4pOjC0o6H|$gu5F@~&zFJ)zr0 zcMF%0?JsVU<2CtY+a-0H75RYvh#n+6s#K|g%qj9^j2T`du>um#T2ZgGQe-SGjhd5R z@cm*n()j)>{hoGI)UC2w zu^*}F)m;2-)NxWOB|=;5OF8XDCS1Zu0R8dfCC=Tx5NitO(6=L!^!nBWD!Y9f9sQ<; z$oA+E_TL5?Ql0?oH{XPz1#)cLg4B?rd3klAY|QZ4=&$Vh_$v)pS~=Eg+4=GPkfBqo_|_K>AT;dzSo3n8h(28?64c0wqE~cb>61Rf@c+!wv=Cn+nYJD`+Q~w<{#|H(>i`$5 zm(rg)o|G4&k)hf-uymy>`ErD#;^)HX=2<*<+HjUA=z0@1y&cI7B$^S!Pr-D;yL<3` zyCVA{C;+UiGsqDg5gGmyN3{nsaPF2%E)RY65AgCnmQ=V~(TV@W@NheS-Q;t7IpU7o1Yd3NcXI@7zMtq> zCItZ&fBKzG9K;5xxNP#hSBnkm<~yKVxc!)IJZX=Y-%-Ik>RCL&$o1B#BWEr8==U$qp`E(P2jb%P#@a^{S-*x;ugTHIhx4eodMK|>7-h>y7{F1=w7Z%!Qo zTf6(*z0yvcP&EMG4pngvCA>J1ZA{&nrmI9@;dGMaxgTXsF-0@F8|ppy4rp#bhrs*6 zaq_qL9L;W+&fN@4AdN1W+}*4Y642dEj)^JZ`}U_eYmfV!c=sG+v#b^A=lT-y>G`DY zUJ^03C?;Q*WzzXYr?LIZfA!1eHlib=&4}r;0&4uJp3ZtKB6HvKuGJfZxWRu0mzeU5 z?D$hn1NVlYhbnK$k5^AgR$?31vg9XHY<-8|QZDLCZW0U?bTNi^Duv-2{b>7l4N>g< zIz~KI&tAdlDff8dUi9PZF5<$(3PovOg%V5Sg|;b@csOVPr8UOmsJhF@No#~@i(5{v z*L=e+N-Ma{js>JOqLZV(O+4fD8~$+QkYFm$&z|HgFRIix63tAKK+f0rYdFj#7M3m0 zx_JkzIQ18b+xIg|bTb*nXG8WcKb_&iO3doVRV#qnrz7ME^GJ_S$}7G0T9*m# zc{7L;{cdB875Tz}4L`XqtU>O^^XL8tehCw*F4KO$dd_(p&$!a^nCEs7 z?kMLA-o_@^i(c9>&3=+}viArnl2OI|$1924@&FPr{%XD9=WfE@>EVuzPr)-bOcO5t z@t9;JYSOlzh4uDP%Te#24#pXall12USX-uoQ}#n!n_vXE|D z{+f&QN)pJQDCE3vz7w|bXzjf7b=*~}cBJZ-ESP5*#(Dd7a(Dg-u=BSk!WRxU-07pd zC-5lWod{Wkl$w>PyiF^OD@a0@9oz*ycNU}6g%zm5M;dnYH*>=Wbn%u|v4nY?gx9Xx zNqTzu&NYAbvvtf895OZ^sk`ZbkKj0YpA1Up7w(vGY7icQbf+2?&2(;XkqIu4m9GZAB`Ka#}00faq2vbZhW7Bf2!@L_qH@s zvoASxVz~k}+$2jUl_d+KDx8T`fOY-d)f4gHrx|3>W+MLIS);Gl5FIqyDB5-1ka)(0 zP!0Flr1-WxH4MB*dRP4*>1|hNXjLI?yrW5tmm}QZp#XL%3V7)_Z)0d12RM|Mj;A@&xHl!g@w53l>LPHKvOeZ=qw zGReqLG-tay{f+f-mf%AD?rm}0<%1Q}h0&#fk(22CK5J}$LlSn*5^>XaS(8P<$@E}N z7oDSXj8K*(p<|7>l8{`kxHv@+y7(HmZek2QoD)vpewPy^_j~Z}l%3eWWhvG9Bu1Ym z5^~qFlN9W7S1tkR?Puw{z)Q3;?K=JM@IyMT(1dPvIz;D>xD$u+1R9ZYolZZ$ zjCbQGk(MQ?RIAaIbF4ha7)3}w zNFN_VC!hI3Q)ZV@na3A76Av?dllQTw9qmRlvIp^ulQnop#s#kF}k$5f|SWg zQt_|2o-OBj4&RpH*@x56AJ^%^b%^(=X_eEKO$P=284DT9xu>|*(=tiuU2k&rQ7Q>} zJpnB~$*eb~)N4H9xZne>4n2z{3`-FY{lUxU7NEo? zLz>+YM;GKTpzOEPBz$xryqD&)QK$3pLT(ovENcgYpjjZDB+gz4JP+;jEa-g&Yy9(9 zKOQI=MdDn$xLa=w!6Pvm4t_I0a}Ki9{{ASDkDMKRzp|Rd&HfJ!o8Duhl{M)*K_1x} za||90$V1EfKg2{&AAHwxJO$}SmK16;voFL~FG|8>?E6DCIW;E2IlGY8k5uT_{5?&7yr)yd8D^CnEA`S3fz_D^26hoWTaU&w&e3r;4X* zIO(1ME+SYAcAjx1q7g*4GqZ_S+zy;|avN%LI)X1;odqhq!?Daj4r~6EBlG{ZkjT_| zv^hRcVBhX7z=7j%Ns=b~9w^64b_L)aTYd=(!JeUcoR`;%<^=uUU)MR9El zrO@A&1mc?IPSjdzi6tjXw>@$L;j_zl%crZ{TIFyOad97ZSUZG+otj|xJTdgAri$n) zNQgS;OF_UEWp>s6O>C9YaW=d49IF^HguT42q4zYO7YiD~wJogu%G;urX4E?i#DI5)-`G@G2eBINq4l6okR904Lm)`-soh4d8^B zXn6DTGp5OLI59k_UbL(e>sFgW?&kCCnsLWiX5R|-Rrv{=vE&wNxcLaLuD*iLySPAe z<5Ex;y#aZr<6!lz4wPfHi28wS?(Fs|Wbxo8(<6TZ$?~o;yeN)U?V86HykS|VfSI6N zww8M=WdqUGQUHJZ(XoyQh%{|LCmwWz{%iww@qn4b#|MlL9v4nf}Tf$RpdiqPTyJKi%uDQdDmSwWB4F@ z_{AhPk`9le#2vamOL0$^ z$di4mm1yAWQdGS`Ul5#IhaNAV;_xk6*FmCA$>I9;HEhPu6YSKBr&!D7<-iP=Kwkb3 z@>{h8E#f9%6EzKT+AN8csj6Z7rYvT4K@WC+^8nES4{Rq?Lv~Jcg<*VNX{zxJEZFrM zJyMDhu9cnwd-N#W-?Rmu^BniWqi<2*T^FqCu?+mC9fdf*S?*x$aAKa2_3Hb`mgVrwv?*g)`3?50gyLRY?5adZ$J)}w*G)KQ<`J?m zR37WBkfbT%6?D}@KC@Buoo@QEkh;A3%UREmz=5BS;%cpNXzVI)z@-w)__9PoW7nO6T zPA%qYI!`kLDj(|7SIi{)zG`yDJ9&r1v?%8I93R@NVJlj8T>(eDk*0qPmm=xk7zgz2 z!>$IJFheVavnrTLX4G7RrkrGIc60_wJ)2Alrx}Yj1?J#CeKUWMsmHqNUd=Q$$fPim5)yXs}d8O>|0N|mcJ$5 zm0j3!g+3O!uLa|(N_=`>8volJ1EVeF>G0fO?)EQVgALnxJNH!nXYR4Wl|ALDCd8a$|8c5exr;#Fgzv9PoZv?(B z-?(HZ1hdP}P+gxrRKr>f9AA}?q<{T%)Nfzn-)v5Ex0!>#+GWUJu>v^nD99hB%`OU( zV3lo~Vdj(s`sm#vT5G=-{{4%E*OojBH+(ug_*9CwUiPM|R~`kgd9re+N@Jdi8H$Vj z@9i~DQc6rrPEPFq_RpFG2VaIycr|4&P> z{eND99cBbr;4;hd|LYQz2Twt}x-Bfav7pd^q zig}%b=y>pHs3G|-_UtJ}mk!1;5|PaH?jrd?)A-TavV3ne7yy^)zY=NL1!tCZR1?#sl)7zq-W zU4WJ6wnD@PX*eIJgbR5l)v2jD%z?z4(A#ynZUGTW2(58O7} zzBl8LX}lheG#!Nxjdj4Og&WcF_urZCshuc7?k}@0VG(2Z-#f@Tasj4CghIlv`|x#e zG2FP{fs?9gQS3t(6!fBjS#f1MyiqH}lS*#$_wsCroZt-c(`8}8EGGzwFoi9$HgJA) z2W%*CfU6Ux!@?+cFshIQ&nvyyBK<6CGyKOi^c` zC0(q_WM@{b?=RR~<9Lsl7JD)K8pMUHgyUWow%I)|Z3S-?Y??b=;NXZnmJw;Al$Da~ z)TY-Enb!%I;?JQE7alXsA#sf9=On&+wVS=rxrcp^BSF{y3Xbagg4DHj+4j#IIO{Eh zjt+mgRXGm4n*U+(-QqZ{HUq70nu%V&C`Ox_G#KUQ>CBLGG-KYB#zZul^LdVM(7K@) z6m+hD#=}YwR`O07y&;&tDViuT>;Gh$zv?;=Pno5D)ImS?rHuEUwscJ#rTLl3h0(UM>X=C$fE=B<<$ zy1>NY58icnW0pABr(Oo1ZRcQ$pdNasO%^P%eI_tp6~`*IRcY;2*}fIq+eZg=>CnxHs{*qxS&W5%RtBdGfhf4~N z@y8NNFnO5^NBMrQf9FI9_3OfG4a(4Q>l?`L6wHv(Zp{n@w1m zn$0SgQMUhuOlQto#|aMR4BFRyi2_G6M+h$$ht`OV0xw5XK~6xW;MaH`_S57s?ER2i z(2?^Fu||ey;9@o7wBs{2ah{JmlDiO!*92wFa;)@jHvTWo7Mn+@;ntE?bV6^~o~=|C z7|-t%nEd0%_1!~^Xh|z0m6rmeW#&NCh%6j7eu_uMMdI+$n)rHE2ONzKhndNPIKu7{ zs@Z0W7FbF#TeGbMJxg8+3Y$Wht)s6oPd%nEV^+5^TNJYy746H+m6IGJ#k)sXi$MI? zp&zYkltp5b?=jtT_^rPWM5=hfc%7L*#9gsd1#Mt)sq7C)g%%|oCW`~BJV0ciRiUou`8R`sqBwLIMn)-Nv(} z4%?5pci--H9AdPe`oq(zWw1Zdjo-)H!wkzooINTJ?-s<`Unsq0=lV6rZdbrXl)WPY zmEAW-=yL@Gx*UP&Q+;_4i6b^>X7E}Y3*0DqjrkOv%uK3tVB+tUfb?P)xNut>vP@L* zO;E%h9ddXd-phWN=ENS<(Py`I`M{T}b0CiQk(5s<#QnR+;1L^)lC0LSucZ#L=OuTs zIpbK|{_8#N-M$o-6hFjV+yv0k8wGWzUSO^z6z@K0#b=^FqMIpK(c@SJcq>k^61fCU zV)m@T51v(%<_X2!I;(*oQqj>F#N?_u$TcDUqn1!qjQ1>Lc$VAJCl=v(VY9J=`g z+zHdf?bf@I4~{}wkM*$0fB;J;7~;9>au`6#>PIu z_VpYSJzR*~joX-g={of9KL(%CgE^P+ zj_a=QU-|&Fbx(&srM#cDR=~z@&}NnRXQaiY2iVu?+3>=<8eGk4VdXx3Y{b5@Uv&Ho zgTD;h|K6o#Zx+{NZ|kax{iBkY$z7>}xz96jkC+3zIer16WA31x4HJ?4x7C6T-`;`j zz&Y5ulmodj$FNK8bijK%c|Wl!vR3e9UT6sfG8G@06%o$`k1u9pk5jsk=v|1f+^s-0 zZCW_`Lj|66<2+7r%7ya9W#B(>29~H!0-@4m@Nkv}P3aBza-}<7d{qHIJ(`Cer|}MX zrw=Hr?;#4SmB1yzHK-*w3+As3hQbGyFn4b_9?^V*vhGQstJ&Y}u7t>4zH82Ldw2Svfdc}saOy8@V( zoji=LU9CiSj&@+X(c`h7SUobnlg^CZQq0Ic-NQu1ra;Z;RKPEef^v*2UcLSxQb;Zfv|dTDV35I0c@5}Uk|bJl!wH!7Zk+%3DE=@@ z8^YeJzyc};&jOxf&+OkgSW^Xln&=9wX1du8v@H_6x*>&knVv)b7CZymO$J-TYP8FC z3v+O^88*>6iFSwXV7Tsg@X7QI^c1(jaE2aC7Mj7>Ev8^Tts8BqEkLdfmdL!>4@qkr zL4KJ{DDl%0SaoXx#0BSJzxyxHI_sxs&Ra7yYg#4*S9rkzp2lPxaSobgQ=kKXpi;jFgfxJ*eLplT{kc|?bINaJ(r8$tv!SvI~Fp68OJ+V(82)cf|{K}gWpGZmhN zj)9ps6d_wTnpuDJ2Kv7^Iun1YzVD5bGLOlSsWPTgxc8pD*E%Xn6cr6L7*a_zP$~^1 zvxG_+qD)ajl9EtSiONujiZu9ykY*(*e&_r93%KWX@7ZfV>v^8{b?wt|L}(LLIwDQq zNJzmJt_$0@T@DI2EQ8zni}6~MGWUEhgXI@hzQDbJud+OqsjsZ0d)3EivDSI0h;62o zJX`ul>H*;p@2C{%gUZ%BadK%DUhHkgUwXqh;mBW@RJ;IASD&LXch_T#;uaj&wiE9a z$>X2ZO>k0d2&`6g@n5-l2+q1KVWt?jQq=|#>i<5I4y%Qr@{3~_c`*(D^B5F3{t;uw zNAs8u>NBz7-g!_|1USEU30li<#DC-buzpm799f>sbS9-T``SZsRpvS5-6_Y%jDQBX z9i#1MuhE|eAH&TgBXD`qOHCsd-~da!?$25IuZSs6hO&_&EQ23(}IA5#976WF?+L|%rN~R2-dM; zVw;Ld^ePz=?4VEl*QTIfn+L8h*Tu%%+hDCAf@jS=amChm*w!t=w$=W{!ur#6{6!C{ z;3EuX+nhngkULN1c2kc%-S9@Vgljz2d+w|*(yHTD+Z)nw{m z7Q!eyMl&Ds%EqM0PSyN&T`y~?cd zRYbu|Bk*l}NX|L5(AA>abiK0+IV8A&v$E6h+QCH_IeHi0OIPF5>Lf6m_F0aY13t1TK zo{EByWZX4%B0M+mq}6+G(i%TkaCR@GQDe=tyV3&XW6jWc+X~EE^oq`zok$%&I8iT6 zN!GP#I$I}X!A>Q!(f{{XPLsa`$#K(xZ!ra~v~gX5qb98GE=@L7wh`~DoQ7$en?b7T z2b^vx2j6D{pi%Z528I{Iq^d`tWb1&2p-AnQ=)>4?Tm0;?0(Uml;xB0__N{diY(H;- zRSIdi-qc5+=0yehEA#j!w`Srq-ynSB5rgNZ^?;vu2&%3r#?`L5#5GZlN*{0|leg(% z$ph|p7rGwZ#u89Jx)}Fq97ea#J~))0iYKPxBP4CH6a^Un_>R(}kfr-aah^fJ4EB_-+k)N5{LwD(JDOnKnX@_H@ z(KxSd7W`Y7LN}>elEb0r@Yd{LwAGfw8QGIqhY@jB^Kuowo>tGits5secuZ9gGD`|A zIp6lSi0!zfP7H5L|AVo|&!FJ+Iq=z63}yc^!EMqx`uLA7of&BeJI4S$B{k3_xf^Q! z=+jvjx6^}T8|n3cmw0aa2v(QpPXAp--yGmF&lvN8skVx&SXNAg9XZ6`NLPQG7~)6ogr`}XW&A~0-Q6^15FLgp`lwBc}+*LYLYKC z6x&7*0pBByKt^v7Xtbq(&vr$4vv(Z$<=>|JE)mc+`~g36SL4Xiee`r)5Uo69 z4L>Kx64l8Ep=0?0EDq?S`cn3A!|*d02_M4RPf=Lsn~l$yO!V7Q%5hwB&}-E<#`IP- zvwg}sX6X!ZXmb|9J0_bjys#HuOQu5l(`axS$8|=W*$EG{=YzzgG;rkdt1;yfFg|M^ zZCiH-PASbpS?(UU!Pk-$dr)G0c{8?N3BlpdTd{424QCLO20OD@*m$LpI_CvJ5_dkz z^V{K}<5rBnlaAA&4vl4-vH!?7ZpLo`QB!T4nD2>`JnZ2GmsvbFCkq6Vb(nm4b!H%? zQ}A~C0oq0~=}oIb`X%riefu~ap8HGUCmnw}=Xw%7E8a;b?IGwVlZ-W;wRoV%3>LoX z0IhWkasK8JW}EjPCOs;SXm|t>=Ga5h>Uaz^1S8z@FM$W1@4?4t6x@z!pjEgC{+E)8 z)u~(XV8B+q>Ae?Mo@CJUpFZ*r*wD}mze(S(?Nqd83j1H-OjdW&Y_`jTVZ#>9XGf%Y ztd}{m7Uy-?pzo7dr<1XG*J?VB{_O$n`V!O5;bh*7f1##-zsup=YIFMHV-%gN_Z$ai z)#9Glj}WqU@Q0(OGI(VNU$S)uTldg_9i2Iqt^ScfFWGFOzApo)dG#CIqn(TgPKIE) zxgL7i8{sh@FN|^xhYpvuu>XMre3c93U9&L~><{M)EF6;wDL+dxwd%>wF-3L{Ux+>Z z=>x_u&%oMeoVMN8jmZi@toPbr_UGtEwrx}ceSGauYqcNl>gr{3vriEeA5WfgGrOfq zBwaURNZ(W3MXejR#--}b1Gt-V^r>yRd~+b;#rwSG|3yESxd zFa$Hj{IOwI92VZvz{(NA-7Or^%qE}dc7DTbDM}=YVUNJbvf$H8jHEH`f(}H7G%nVdQ~u%JEU23n!v_i`iY&QR@mXC zgJ$}=*i*8JpSU%Vf9|sj@5iw``ghhDTBh4Z_vT5mtLFX0EY%vkuT@4?)yC2BoR`U` zaua^C_eJM>C-AYb2)n|f7sZWQ@Kp5|YB+F)8dV>sRx7Tse*?={y>A!T-h@6dSI|TZ zUx|Acxnr%_L@W+$p%aVF;jLTDF`>f`obL>f&b3c}2vI(GH+7(=yxS3Pv9g2Dv zaNPmZ@z=UbII&O)C3aO)Z<*uJT(AP>ab2`8>ZRz=1_^RGFpn61lNL;hl;Y3L(c_8Dy1UC#X+qhP?Y5!QzE8G|l+Hd~?ZV=5O>SaS7|#tr}m@4|v$A@{bDV9f44T zg}BMk2ePyt(XL(naMwQ;#H0_=k5LMAaSlSi(s_FPqzA2a3t^gBK z(-%i`>AMFS*gu>Pce}&jQqC=St5r_R21^Nw97k8(kB}5sPS0QLq?!B;l;MZbZK5vZ zjioi6mS(r<(~5t_JZxLs-WgMAxMZ z8q~<*Aa#UqHGBj~$USU_ZUkVRsa6#XGYRuUyr~MF#2Mt`!NJ zAPf?|Hlr0j!N&@1*z{`^7Q~3-ni(w+tk#7|FW%sc(H`VaJrAiSCm{DrAQX9z^8dz9 z5LkRwKyUUDpM(cg?`>ndh;wAp!4jn&sz%+kx-95+>@;1BE4}^i0$gT6k>< zNwW^8lOtm3^7Z9ZlzZ2YwEVzZ8h`PN#Y9|Kcmw|aSq%5i|HjJV?|8)RIW`#!V_f7b zNVeSv6}Bv1=S$(Xtu+8^jv_f3jCGP>_%g(qS&*zj3c7U35s?P8>%EKr^wRLR`f0FP z83Hb1&Y=EdHz@ye28$*h$4Q!wE%UXpbx<8U-{s&%t86s(kHr30ZA8_xlN`txBUook~p)qTQD%moYC>i6i8@C(Q4Z!dU``2UBB-jsOz#& z@?tf7-#0+bl)upgZZCCe&qoY@@)Wy*F5_?BNo?$L#ahRic*x9#TB~iPEt=NU>)S4T zaLECyg#2+?k`pqotnl7{0?fAgLKhgnq~|~WqjyARf$7x(+wf3w_f;KOhF?W^UV%! z)d}cuji%1_6JbH#F;L5yNE2;1cH#2P_+_U(I2Qea89S`;QLQ%<`=mi&@idg#ll6<5 z6ik53Ybj73kwd#RtLWNevtSRI4b0AEFto!A9Gp0w9&-M6z=y)5xqpWO>IB=^L{~VIfX$%De>&7c57zLkhnsQkc~@h5=4d6B1_pJrK{EOr{uV(Zn$gy7d?0S@-Lh)cKOWd-E0oXQW_U zy*&6{lmmy4$6#QoFNCrlu=h|5+K7dmdJ#o7d#(8yQ z$$|0Y)|O28v1bm-t-FhjwW~nvS0em4)(*7bFU{p<)v}#c^l1A8l=pGqv@CJB7WWBs z-7N7=Wia{-=8y)@LGtLa6>W}u0Nz)R;+ytvwC1v*H+!NWep)ODIsInbmVBcoN?sV+ zFTnH3=2+_#guj`)Se~oR9^PfZ*0A+ZwL1|mpLT=3oYS~z#~947XFxGIU67$Xk=c_I z&BR*H;<65Xlxw-86`j(!YpFAsK3xG7Rgn;MQUbmRL|`E91lFtP;wR%mT(0l}hL^~o zZGR#v<-5VEEs^xuH8hp?R>3is)Fck4|!kx3^9Ft1=8PZoMIb>Q;XEZFNZM%|5sz^Q)@yxpM)37Ywke1ngEyfwJA zx*2T_>a$H2hp2k)IJmvZ1N`Jd;alTzus)Utf0ctFxdycLg>~??l_j zt+fB)J9=xLG{mfnfwZrRDD^E4*G7ZkI_Lye`|BC1>7PX1*=%^P@ESKx ze1O?fy4dn^1$(mMA!aEzg6pK+Jj)%;{F0(%W`=J7`n0E^%j(NG_Kc6(Z)@Rmb2;5- z6$J5WJh)J+LjUWvCdwN^NOP4%Bjo_2;~3P$E8 zpj>$dis{~9RDbd4nt|JN_0fGy(c&n<{4X-RA>EgxT~C$e3eRqs_x(Z`jKI4m7YOJ>%W&cD<(Uz&RX>n{Ub(>{_&(*e~U3nDp zt+@^XL#{uH+b@5pPDf?t5=t=-(Q5KDJS^0P-%kFAGW(EfU0zD>EjmMW+@$|s+mF0u z3%OUksrKVJlp)Kgmgh3uyY&s6aqs|!+oT0KtwuwKL|iy-CcF3TO18Xu7wdo07~d8x z!r3}ov2M%*7k)j24H02zoN@%sRU$AW=p62em*K0=3gw^A{K4Nl7Kx|NMWbQ~Hxtj{ zEc#~<91dN85?OAaE2RR<8>FDN;wH@~H>KZ87LpMEEL=L3>qroJhmu*I%nXkj!J1V+ z1v7Yyu(E1C-lAL3xcN3zy-x?RQ-L6t*@G&i2A3z@M~@fU?C2?1R{vQn>wCTsU88bv z19J~mU)N#rtS5*!>(IMnfbnIm=(GCMuqfmKU;ThGGf))DcupG?B&WXO=hc5DHx5^0 zUr!(BMG|Kh{x^<@T+C&3d1K6Y$$kuq{EEj#zT>s%tN70P0j4m07*KjtVB=ZJAATgu zoQySREhimew}qTydq)1FVb6C{35gw4w_BL*2|P%Jro4uD^A|9*wG%#Z%gm)D=fnc0JABRPNr)AiUN$EDf#$t@UW&vkw^9>a?8JPb5b zMHv%4=z1Lv33-wH&s%>Q?~0tpV`2}(zxkKps@8Z+cAt*fcarh)?B{rOTqb?(a+DOL zW%C=gAHkCkM)*j+17(%g;2Fuc@U!I{y=yItm!d|1F1!y9dQ$11zI=FU9*j!djJ_!$ zjJQfy&_}8-=#ii5be7j*f$H%{x+3iaz4hTg(k^?PF1WoCu4c@@SeF^#`E@S_x;(^9 z;u6?!-vw9j#<35cxX|PIxzNsOG<~73;n!+g*xS%ZY`DK`x4j0O(r?j*@7Cnmog5;nFfues_1d31!C*V=+76nf>yyNs-w1$HYXn>N^^#3WU(Vmm1=~bM~mnk zmHTvBRy};LI71xfmXf}RBIf#c1$=$-IYe?E{!L43n3DMc%(SzM=mqarxYy+!-13rz zscl)9xuza3=Qm;34;5T17s&NRet~}((P(MLrRn?=Fy==+z1^+{tv;EM?3gEz9J;_$ zIIPA?s_mtbg%@dAL@Es0}$zF(!I>Y97J825u9=|1dPPF`QE*eDze=9oR@(2RBs}fJO+t`V%W($1EceR zm5{Y$cUv>8g}(Q%=~I^aP4p6L8s<^Y}MB z0m}}&XTG|xV01f`nfDKx(D+X~#`d@4S4@SEvNCY%STR*vd)1$|O7vy(YFa|i{tQtZ7R2{yq_ z2<1d{xbDX_xNh=)oR+A>wj1zS13MX9_H-)lD%HWwa*2=<@E`5fFQF}k$1roneHivs zf!ui`M8E$NV=!Js;DO!9kNk+PejhP#y(yKhiYKbc0pxc_CvADHPPZNkC)2(N@F#LP z8*@=yzC8$cee}g*8+XL6Tr%m&Ek@+MEE9e4C>H+`LL<}vXk|tMR#M4UT$#fOJ_hM-9 zUoCK#y};j>;LG>Fa+&sQF{g8$eIds;*OT1t_e@9P5i)i8eyV3GO16~v5jEv)K^K8m@wIPZl$&B+rI_8)>#i$n(SdFuq7;|$cxGh}`*R&td zag*h!kewy{O@=YgydC8O%P?A+W3`RM!sqNz*t|0sq6d$|%=hMS@A_n)MV z&FiViiJw$&RXhfq5@yrJF>E_|P2Wuj=U6&YSQ<7=Epy@+-<@|%-8b06*$*mEm_D8Q z)#_p9_C&4^?=_AV1<~2sJ=CT=5bj!j1cxif@wMz9jQyyFb32aVwkzD{A4x}xXTqEw z`-J*cCE-AoGp;D9g!{_Tf{yrBW`CiBVE)?OxZD0AUVku#@6C+i%LXwzXRJ-|v)UJC zw|CMtk;h52XgV4Bx|{AY9i-K6+pyu~Lv%D0VdsxrB6b!MWGF_9I7m7`&O@%$;j)pP4_yrnJMQ8~huMFhI;272O!B z%mCR)LX#p&LP=vn#;29(F6M%XyKG4BmDOgIDA z9i%{E(I&_k6vuP9F{m?IL&w$rhV2VuaH7o*Tsx-18gs_8CryVi=IC)uo4*H*s=D}B z58IpUYah=OetQ>3%a7r`N3USz%5b#$kcq-=MRKVAm<>L<|n$p>7o_m6pZ zVkQl__?q6>%t+s`?4(QS^+ z#hZ)oqYq({VhjfUI0NS^2Pl8tK2rEo2+WKUs9UubEdQZ{Rqvhg4S1up=>~`jdPyh7 zq!Oq3d*PX8BzmVQu{$`8QC@Bl7MTQsrZDFRN;yWB^xY*yI1FUl(!j=V7EW6rPf9P` zqQBN&fz^|>AhD|rUhh}Ob8p|%*&8$H!&L{Ueg8z-ZIMM*cy}6}Tb zB8?O(yTRuP9@J^B9CRzp1LL4hxZK_d|NLxW`O-{iY2vt=Ppj#x&M^ALK^u~MJ19-5 zp@)CuQwcX`>YfxrH`F@N5aSWzwyT-0iTOl5rpALs-a~MV(!w}55wd&n5yt()a>nTN z4QkzCL^ZO@sHv*+&te+96d*6mRB^hvS$0Djr_z+p#20yZ6=?9Ug zB>!gwv0uV*5LRELB2GQ@&8tXydz%~AThIkA(_`t{!*1kR;coIx<|K7kmqKT_`GAQ? zIq{M$A{+EY=tApUvUOAxm0qTy;_plFwBiOVPc4L_1&0Y85`rGb46qrnWjq~SnNGVi z%I6}Q-8_bz6vaV&Ha||oxw^cJ9f>C+3W=4 z$!!07eRk~?UH0;*6zjJ-4-3*fF?I7S6t0e$lk}s-dGgz7AxpE|0)_5ooCvvIcPem?wmj^tdA*g_a}OGNhIU~*MqR_ zI4XWi#<-cQ(C5QhT#%KGQ8{94YNHO@nhb0s$4fNXGY!3Na$4Qj`E+4p53|SS6;U+` zqX&hA=zi;#0=1azBfpzx*>&qq9+XJ3@bd++&>G*=#?Khr~~Zp{!vNCl?7PsRT8ImFzy zk??DN5Ziwq@F=|yy57vD_`l(~f4hY2w<0W#%?fMo6pXRW92V>c$wXV*!r zg{1>g5Pm<6W0R+l1$MVcn$I0_WR?tyXLLZ`Jgyh6V#io(AhEeQO68$=hpg2m2X zV9)({tVvGgxPDn!;ZTe!GT-q2W-ZokWHLKosn2?h(_=3wabB-j2J@x>u|a=a@;V(274PHTj7FL*VoS^U?nKN(3_e81(vIu@QM*N{ zIAt3TUAoFaurvbs-2SUz{7THK+J-YurK9h%&zQ62FD;oO1W`WA!7FGIoLqU8Cf`Y- z61UgUpu8r?BmD_Q1w*~ zBmF}0mYyG8AHM(%rcJ|b^EY9_e>|LhP#C|f+B4Zw*Z2=x?wcIZEyj|lX0(?TXXQ1r^@aeK#Q<5Pf9O@x_J^|1X=7t9czfzJ1Xa9>I%*(5!i{w$24 zzvjzePU{Ak`Z9(dn`=sy$J*$@AO4^`p$9FvyL9NBNc6b=1r>`V*|EQ>tY)h+YK)i0 zj)kJQO5-&i*glLiVLYpIB$KKvy+sAKp#%)Hm|8tWMq~rmFZt4h>`o~ouj}WN>rpM( zFl7imw0@)P3`L^U6-I`mC8&PF8gz}w!IO(6*-sj`@QcVJ%o<-1D$ zh`bOxYNW-ExXon6eaB({qR+5Cy&t4xy};+kV<@exgr}4KqIiH9JJhGa+CSmrgw2cb z(sF{CCQ>N*sUNN`{{@1TzA$y08l;Pyr1t|GV8winH_2sKP2MT9Tb<{#hk|#pi!(++ zZN?t_QS}IqDT@n!ci9LePafo(@0*DB*LLE@=sGOrdJiXNx1&!sVg2q2(Kj{9^r3<{ zxm=ivWA^db+v|@PJzP+?dnX!*uEBYmwxX5AVsyN?5>>ZnVc^~;ct}x%-7Y2s)~{Yt zk$V~R(M@;o((s3kGb6!ghYmPcJ8@jE!yFSSALTALpsL+#yn6m2ePlfwtbBdp_HGp} z_pgorl^(~-2HT;FyC?WcrooeewV0W64%21sV0BpuJ`s^+bxq7z-er4Iwa}d?S}$T^ z_C@2nPqCP*mXE~`OX0|_YxuJt^Q zOUs(*q@k;{W^E9yX|(|D`JV8@CmF)ZuA=XQ%Q)*!2>QnPf#WAv>?$or?4D05W|+1(oGw001#7^t&{x!nDwgbW;S*94_U z%ILhYGc>Ar3z&W0O55ucNU_9q{&v?e!Lzq>nZ$}pqLwy|S@iug^ZfWyj1#K__P!pN ztktIt!fN#TD|4!t^b+-ht=LL23HJQ=$?!N@1IljLgI}s9_9nGL?wJ(GJYENAa}~;6 zX2YzK5bDmcNLv!)sgd^zGJ1M4eO1osgj1I=vm5U-=9BH|m5Uo7I$Rp+e%8~89KY~& zN&(!N(m*N>_~T3IRrs+t8M@T2fy$f^XgZrnKP})`$QR3L{8}gYbFhb)e-fiI2am(& zua<&Ip2~uxZTaMrXgbU~pa$Q}&Qi6QWV(JWa{p%{Je&KNnH08?=sr13@6~OED?$K6 z>khy>pLfvK^c0>qZinXfLZ;bm4e9b9N5|#|(ClhTMTUc*cSRT-U)xDP6vWZmRg>WF zxFNFrV>x*|`3BZ1&c|}@`P1-fuqH#OOrhiB>xP}gD)zq|#W^ zHB{!;bu9TDimzN`vEZ;ReN$UR)n5$Lf43iEwbn6A`7r@)g04XFbWQvT@d15FnM*V(mLq|)%x(!;mq&g5{(#;`sVi+hV=EDou1n^s5 z1c{A0Xm6s7w~gg6XLB4iP`ODZV>_wWhAoV6Yc#nyR!Js4E+Nvv$GALd5Lr}-xYdS_ z?F;8%+`=hvZm9t+Y+&xYfkDQu-{as7RV+}o0wtFS%tUeBr=k#If#VF9@-GwNb%V2JG z2Ieh%2k0P*0ZCH0K0zHH6i>oKR1HPG4AYsX2~~I>MEiU7YVVbDCgH)Kfv6n+#nNkVjf#6lk`>PVhY#4o|jbK*!)W${3f^sQp!RU#>dx zMnti0%@43@n1&zsi{ZCT%mrhroA>#^(I4NUP9V&Ar{q>TG(W>m^aU}siV- zEy5XLWmspIg9q~kz}OjM^55II)Tx$)`5qvx5pQYVnoD4mv6~(p^ z03!2S7G86E<>*cJc&4I)sJ11+)cO-RVM-xT)d(y}*Jf7+*1(R3Mz|RwxlGz0Jas1= z0|M+}z3V^xJtl`rw_|BRX(R~^krlkZywT*!{##(Jj2Ms?kM=4LsOX|dBDt)Hf5Gr5 zUAf~fvtqiJ;K~o;j#{-HkyW>@_2AHH+Pd(GpzVvAy^smbMtOfg9Iyo$Wy&QQr=HAAJnu-J{ng zk8(b`t;DhF1;^faK|R0c(b-UgCk#YbzgNm^Q+E&^6v@I}og6nhnVUPuX0b~DELinp z+&jCi6fcdd!j0cA!a(>>D7}*g_e=f3M{_6#C{Tt+>g$PekGiPJ(9R?AOK+78z%2(3lk9*i7o=JOQ=Q|r@jJ= zR|)=KrxA**8G$3MoJZ7UFId)W#F-;vxV3c~_)jhr6n~n^FH;fbdwVItQ{4%$ztoWa zr@RW5QkaQI;=49aud zxcDzHp4(k4ymJdSJsqWiQE5~$cp4hb$U?1N88&2+HR~%H#%}VCVaNSF#JXRf#;*PQ z6186OSUfAjPLVi;hoirnPRq(NmHT>>@9`%XC0B80*n(xWp*Ire4C!K&3OA=`nvrmc zr_75)ak6F=!`|=cN4f48^zgLBk2#Sre@F+sY=NCoK7fPk4x+W$CETT{jc*1c!E9qH zc*iI}MdL}Dpp!$lo>gOdbJGM@qy0(#>6gO1P4n!p(7?1UaJg( z)S({C)_;r>#S>AWeUUpGkHfbm>p-_|6^^YH#^Nvui2Sr0Yo!X{%TZT|`yRyapU!6n zLQRviGVfq7e{vX*(%;FtU z%#=z8xrC^I0L_TP%QQ+A~1@ znlgJ|`UQsl4nZk02gg)2amz{-tnHeB15($(;?zcv@C_nk^Sg-Vm9KRE(0jqxN;eWw zXhbJ)-A=QWse*F8EQr&`Bbr)trlAAv_7$Qs77O9ky8w9o=MvQ2 z)ngaFabgYIJ=yAj%kZYR8JcAK!AzW?)~~mc2Rvc&Z?!h;FBAg(=FjwK(?s^dR1>zD z&*_bFB{5^L&x9d*CpwLPf7hgyRUu~s^d#0nc z#5ZsrB(Odym?*pFad+XvWJx^dH)vL4^InRuZ;YepBYrnsek6njSOrn9A`iNjo}r(j zYK6HB5#1Z~iZV{ zS6vd3kDN=aH)=4MJps&?x?;NE%|y`PSZcTZXEN8jGX(YHr0M<~4aRom6+t=oZueR% zPeaZ{lE$c;M5$PgY89%{pqJJ3`G**C+H^cE5Q(8XpQyk>p?bP(jV8>vYX{$)rs43b z0{qz{0VB4q@T_?diYEzC5!3S|xjBfGX}Hjg;1DX6Q$?Ke5=qlMVYuiv2KU4D=&U9K zsx1hm2NT9aknLa2Umi*C|IB1E-r@flQJLF2XeVY2uLymQmWqd$%KRkN%ZwU-IRwEPFLOua+0CG3cfb16hgwZh@% zSa@7!4IjT3!hc)0LZL50!7^uPFFN_`^9-Y|gRm*H45?-8B7;TRO~E@K1F z2&xCPVX0OZ{oq>(ihqLiz@r;Vqs5l?eIl^cEz z-u+IHurd{Pnr48iQVjhrtxf0EU!?K*MP%dQ7V429i-B$-6fYIi&%ejPXV*hyp2Qnw z(}{zOtNkMUbKoqF{Y>S$a9ZHL!ChFlYYtvn@D{W3R%7s08Fbit0xe}C@QjZe7A1$m zxI5u=YB@s)S;s!Hj$n;4;@Fl2mFTu&HrpY5ihT|1*hN?8vWB$^?7o|gXs%U;hTUnL z-_?ZwMm@!3V$@Hb^!-6J(T>7%*E_-T-Zu8bkpy;ocpRG_{*?63yvf+uHZwc*TJX8* zB(`OaH9KVc0DT5T@MgjoeSRntkB!quA@>$|v@8*)E1X63B?;JAD;r$XvvQ65X8dA%fMLTEd>_ zu4cc>&EoD0!5lYN3$N!V!i$7*x@h%!TEd^oZdzi83!jF;j-BDWXLm*T^1mdA$JNPr zB{Bq5y$Gy6_8o5Q7@!gjNY}^K!>fn`AgrkaS`)KC`S=;gsc3{lFK)n;6VtF}Pbz|Q z1-Z9{Pk4)V6Hl!#xS(btJ{bw5TlTxd4Xpt9a5n=qcimyPn(`~ozvPTM^vLzi=Z2qrJkbbugiZ_a(6<NaT!}H2KoU9v-7KOnW8+DSt+xM5oTlT>_ zvuDh@AR+1|(?Ub9RI`0&Wz5{8waw1gb->3wVdOC7XzW)DJ8G6drQH_?e`LvW;S7F;oU0-1x|q~?1X)v!tii?1qZY&!^U38FYY(guHW`2nS;_u4lQs8c?J>oi>PO3qr`H)TI97@R`l|A~=jGT}^c@oX5lt&ORNC&2C-1fw41qwdR_ zq*{*0@Xe<)xyQ7yRqZECvZ{rYvIC5&_Qc^^0&$> zG4kiOFv+dIsQ;oU+CTj>T~=5J^X}Dyxbzs9&pS)r6>vT4O-DgUPX)B&xQxw!In-Px}DRu@5#p*TkLGvS=wCjvs~^U_TzLi0Z zd$Vb>!UOu9(*yd|+~`XCGtfP+0Q{bd;U>FO`tjrvsuhiN(0LyACQBm5#=<1G#rQB) z7!OSqgTF34Wcd!95w=RX&&Qq|;@GbrIgWH!1>|x@Hr?h5V zE39}_0d|vaz!@h_Cp*6xe)cbhXVG`a(34QId4V=rve6OGT%U`h2b#gS{3iW5kV8k~ zljzQ~9Q&>DGD_}>MHh=qjB8E8=aEUc)=wBuUb+FSYb%YfnufxC5*R%}8nuZ44|s^8 z#J#TnW9`j@YW(B>?}|3<`$n5&2`%TG>-~Pct|%>*NJwN!o25cRN>OQ-RxMhjMT?43 zQBo2qOUROstt<%@l3hZ6*Z1Gw%ze+?-?`7sndwY_oNLbZel5@E;|WD7`plKDRf3l? z`hpSV@v!Dr7j-O2qF(J+QCp!01!ry}KkN|5CsWwnIUfG4b^!9SmX13&k=`0wf|;Jm z*s}L6418D$4qMDX*Gvdlt|zZ9slg8-pKwTKJN~@92nTj(aW#@IS#qm@4B{%9vFasU zX+8!np?~QkcPDBQ5J!=+A?_5FaO2IJXY z%GC+de<0k&1}oNAfQITxSmAIQey{t+H#asHxLF($oZ&R#odW@kUgdQ1^w}V-5Wh(4 zpZubex2>n5rn{+g;C^~2AsfW&DnKiy2g*vgJ*w+=EdA?=?ObR_A6J89mQBO?nKopu zjWsj;+MAf4jiG+!`Lw38o(k5SqD)8*t$C457s)sfuQ`K^QB@L|wDBG|rpVywDdt#k zW+HuJl}L?3Rzbo81Txzl?y6^lN>3{Z*;+;__gmAeZ}c$DO&ld{J0a*}c66ysQiH5k)WU#49gauC1!AVjbO58NyYljrz1px$e zoa5{uM_Q|$Jt{PLp$`{4VSpOcu#a96W$BhzWeU43oHQ*OWPja7pYA_&2!%+9Th> z?C9x?O!hct-1K6>wY;fhUvv;%vv-hI_0?nHu@>~-cMJPJ?;}tCOQe^d{-M`9#uzK_ zDa36;0Ll0pgTbu_IUZ#^)*fEYti3;pyuTz&{o>wWMzRzusiwlpW$=NWcL3TymxAH@ zC=lIQ22V#~;PZeUTgvHp>I)9y=cEa^X7C3*wt5D>9A`mvdK8>4*92pMG^hwfv7_K9 z>a^5AbuBjwxVi_AwTz-6QDc*PCbNsyH=*>iZhYO*iklus^8W5#Z|G^hFHzEXBEQ>D2AtdK!CNky=>| z)AhGg$-U#_=n#%4*VKNKYo?!Q(7W0-B#s~Ni;quA zQ=wGLI4h_9z zSrgZ7U>;Kms{YTQc1;9!H?P2Zr#QcfRTqv+oWuq_GiO~S4`cJ6Gf>t?aZ`3YW%`p~ zwa_Iv-M1J&2l+uu!6BMzw*a+w5%Z z+Ee(i57d}C6>z&YIXm2%Bg(d7EZ5peX}0^^p{{~To~dsQJVzYL!^jJehSf2KY|bMrlara9?TG= z!V}P*%ndYu6T_%#W*B#wcT-p$xlg z(3G_?D@0d%9*r&@!~Ygc1LZw+Bt|Wro`oiXXXz}#gVSqy!37G8q2EPvmHA56@(;o^ zj$tt(BaVrRb74D|gAuE20a|raS@bnw*O!^FCKl82v7`!0hwMg0l?M9s)n_{8zj5&8 zk1rZ121ET#P3YpZ&V{>-aX9Tg$F46#ckX?U;yJvd;uHSS&pX@2jQ@# z5Z>IFMK>1PfZ}OgPBX8FyI+c+YUX|Du6M;KiwjV#*h!teQ{mUT&CvEW65Hn;BmQa| zh~aig*fmw3u`pWC_$@reFi*Mf^pWL|!tLg7>MLNoyB!Rdi{XP#S(=npMGtXap_AiJ zv7w)R*jznf_Q$zRwBxh~=D3=&?q=NQ`-(cMaIBM^eZLU3-7u{49Hehhz=#c+c=C7` z+}&UgsI&p@7qmg|)KGc=ztFb^t#GTc7$_3NGplhJHamV>i_J zHxI6zvBlrM*6?w<8IAC}h3eZZk(y^=qh||syz~NbZcZ$6nCqOv9t18(cM!U%?kTS zz}24w)~unzFV@ga{m!WBuz>wy@C%DaIPQ<>IE;B}NxrxnLe9N;sIjmCVomBm&UyvJ$`;#zj%n^f+ME5c{VrI>r}ImU^UVE0HQy4Nc+?Po4=1Bg3}`O>v;?5YJM zg!@CW^f;I;ISDp93Xsox2FpLTK;)xNIH5{~s0(mSd>4!Bf zS@7P>11f0(JYE}3wZ2Xw)xRdN^9G)y=6@g1$5k1>n-9XD|gwsKuzmIQcD6p>H z7U+E65&eBwR8E-y;(S>UdzVLz)zsNo5m8p>-BsMlf<7 zVa-fBt16Ke_Fg96{hx5>f+~m_O`*I;b~G$2f{Oi!Mhgi?H0QKK?$Jl__s$}`Vz3{( z9Uo!M=`M7c`57-Z8sf%NUIyu$$3U7Zd}QI7cbC)ZQH_=fuybN;DbA@=%xbu8YU0-}RyP#u0;Q2T8r|I+EB zW}`+5uwlarD&ankF0WjP?C=n%yp0E`^H=G|cz3!pek0wjHV4zroyF38LpEWR3VKwy zK$h5dn*C0mDokBQq&H}jQPB}J+Lnks1sSY#x5xMtjz4!X6IAL?K^eaq+>DmOKB`1} z%QT>5ODwpm$H1+5E1}u{4mC8WVb+#wG4EewGZSbIe!jjJe~*_$b=ec>_;wz)L^Z(X zDVp@pI|-sGa+2}?ago$1JQu`^2noJkY%$4sG0xOT;yJHjw~!z}B!*9JYw~YNo*)IH zHcZRkd`2a;nz10;z; zNdws^c7tK1-!cYf+i3jpDt@@Qn5*fF8LKQkluE9}K zrS%*lV&A~P!}IX^`%jwQcY((B=g@t6q2w2{m3UnrVV;Y7kxv)INl&X2Srr;f53(01 z6)xrW;u-XH_5||wemEntX)APW5`p)H8T8o!KEtj~6ny9q6SM`Zpq)uQsLh=VZ~t7R zPrj{(qUDF6BuG-An>2x6H*Cdwoj*Y4R|#nL!bp0H)3vp}enC}~#?ux%OQ?*u2FEk= zV5sUE)p(Ib*IZ@jl&&23=Jmk!dj?>du`XkMyzPEpTxX-b=Sc-50xwbdfBvE6m|K4iRu#<_Mpm z2V_=G#-;9xsQ*w5-6vh83zgHTfw7aL0BKAxfqzfD2LxE5AKe&I3)gYcxRHe^@p zz!tgy4!<0PM&}aXOS*G^h1io5+8r+X+m_A-}bD z({iT~VqWl$wDsg5E1QO8o&_lEWR1DHQ*m>`By8zbq(SfYlZ;_)vTgDc%43(%RkjyM z-wj2SaQO}LJt7$Hdm0)~pM@OHlkk;qiu_hS4(A19_z1@~*VIMNbuJk9Lk;Gi5Wwa| zjqqT*0zQ&4#F^PnxOcTORctq2vKBOu?Kx>R7#05)+0-!DILs);&CekD3#)o`jI@JreYmw;p{y{UFYkNJR6XO6>V* z1wv}i;bfdXu6^u@jwB45|KrYl-GiLJBpS7vcjIvdd(K6_8rrhggGze>=cg3{59urr z2~HqKkEzm`g$L=T#kW!NMi+X!Ji#h8ZpH_k&nt!Vcx=;WSN@vGhVf^zF&`b^xsfwC zwN=4rIu9x?Dud)BSuEWmi7x&&*p?EBQe~OEb+*@d_rHA=gnZ)ms+>{7V3#A7yBLE1 zP$g_Vc^e+6h0>oh#2|ULI@oax*ouBJ*!SfC6gLmk3m+)_ouvq>w{{bkhL>dCylJ#_ zqd0E+ZGl1DT{R;oQOQj zTs*GpioK&Dc;sR@Zi!8xttsMgszd~48Po`d=G!xhS_>GBI5FnTkvAmsRScC4nm{vm zPoQqrt7+t_NT#I5fPU$FOZUsJ$D`90p?tI_-ab?fe)SwTB;x`+coiu4q~*wT$ZC@= z_6%-{;&LHB9m7e#3aQYF1CUaE4-&uD!N}!GI2h~?;n`F0o7{Iuo7xU=hB@Br7l@aad_RQjmJ@jc>pf0p~3DHdp^{I;GS?aqb45oGV(vj&x^xxMI%q!iC zgJTi6eC$7g6ekc)ESoJDKeCl3x#f_%>UXIB$T`jjcnBtk9)P7sJF&JY5U)J4K*vQj zT#wa;E)O5!;oHV^XQ3|f%IYH@orT%hACl}mJ2BQzd@&t1m83@kglNcO4}Sfv54>TW zgM#Iv!-6k`t%7NP76{_YZBQm55+x@U<4+kLasOM!_}kbLBaJg$ACQDb(g!ilK?C39 zc7fpuj?r~Jg85EW7~2F1!Ndwz+G=yg)7Jo$p7zG!m&?Rula- zbpxJP+kr!xo_OL_3k@iUqFGQ1Gd!z^wyZdO`qKpKR*0|-@}qcrb{lT=5@+{VQg)Aq z2RrUh9y;cUu%iCf?0K~Svbco*ovZq({Sp&WSseD5Y_$j z>D_x#+^l{N`ZV?s3pG3XDNT*j+^U1Zbt~B4wi9?uhRN!$+H~XLFdF^f0~9HbK*`uj z3>lq`MgCbh=YAh|j^yCjhB&;j;y5w1)2xkCKO#(OXEx>XvlIR2fCLIWz0yidWfTAupm;N-43Ts5e z0kdmxDPl5;Jnx6?<9H}}h4bYtdnq`Vwt|t`7){P5df{Z90MzXWz$LXRB(Li|<2LOH z*_3k|`&aD84X0*eN3sqZw|octCoGdqnFOKb6QrH{IG3%^m;)y#ea5cBLuxcp+(ti$YTFUr~rIh$<=c+Oj?AmbM zyStcG_5kaG*RzHTJh@ZRh7AzQg?Jwm_~q>Za+I5q*_)%p%&$=MN6EyF_rPREe}dp! zQ3kA0mxKqW^Xba23s^8N9)0gJ_*1u+30m-q5#4x*(P)%ruVjd_i#95=vH@Pusc@cd zJ9nPeuX_x=v#PbT6Nx4`rh1e;@;_nud|~M1?EoCD%l!Nj%odcrx3VA*#tyoQs|P%g>qzm@IgU>_KNl6cq?rQq3}%kr0cOvG z73`9a^Vt9qeYWTPd}`m+N=NbsX{7maI`7v2nR6O! zW!W2-hlO0_P)D%#fg8)8b)5D&nkyEGlOo083VVhQ*^gRJ}Bv z9?E@0ZO!Kj{&qA9PTrfuZ1hnAhq;?*RKyoDyzUTb+v4RaKyF|Z%2QE@Dnn4fAtiM{OC*+gL4_f{wlI0?>7DU<`Qjx zr4QZH=fH%Oe>vZ2H4HEACO4h*X~w`wy6#65)I_vGdf@{&B0I!5U#ub%|1*cZI+0`v zPKTfvj+gLqjOht1BJN%ikTaX}<}HZFH&Vge+|D15-sLpsWlQk6=>+E4_+ml(ln`dj zsE=;wRe*&V;V}NkIDD2n0>x8|Fg5l7Op)}3U;i#b-ewKjyi?cNd^KVP3&ti6XBso2Q~ zw_hTG;(An4W)>}(n|B2UlG_=O64 zUZ9tIIzCODgn=j3@Y}Ilu>OKJp7yx~?%aH^;pufEenXhZGB+3*`LB#+!dWV3#IgHK zn{eJqj#;zt4&GH5p({h*!l4CQG0@`=@49&{FJ`Tj>1CbIuz!X(rdbE!(7;nzn!XL| zd8g4wYadhEc~Br#St-u#B9 z=kKVGY$(^ZkvCf-lbh!{IT$T+R z^mjp1?{27nkOYE8XQUT;@$kBtv}a2wX-p0$hPDgQ#hBw2Uu~e{FC3))3s%B=n>(Od zu#-H`a-{COVs77c8aBtz#?BvWaL@zUdk-hDh5J=G9c~YuKRXYKDztI#T3?trQG)ah zQX0YQ;=gGcF#Y^7ho>_6DwTT_P1Fy&ksV9IX|d-nDCM68pRCi=vK&Z=3}saQi$>$~ zV(gZ^rYye?u--L{(;y$k;bI3Wbs&W7s{O)y_IaIP@=+-&Iw%9DIPOxgatzGi;v2rl zPQa+#F?jx225Vka<4M1K9BwGZtCCi@zf&ILbh_bgpfYUxu!pYc)1oBy0>muh{Bng? z;nXQr_EX11_IaEq;;;Y8DF;f0t2m{utFY z@ugimJaA7lg9!mTn5gW8uVpuLI_h<3Jj+nUEfC%c?PnzeCGHZ1B+hbp1OM&zB>ygHu|COsZd%F^np|usW34)GVG+WVN9M= zjV4dkuwd*s+~9mpO25qTvc@wwJfRh2MikkjRy}xV-cc^=G!FiGrGwV4954ymLHSKF zbnaX;Y7Z-LvI(G(j})qMSu0mwX5vDna@Y(3Pk71c z(!7sG9D03y2F^A2#~1Zd91D9P<|6m^=UDE1_+@r|M+E=xg`0f2P)XKZ^*AIw!E{LqC|q@d;?ysgCVmfnTCd?RmlJ}h@H_m& z0ZIJ!-Rp>Hs})sZ7E{R;v(f&$3U>dM!cX)D@N~sddfOfeK`5u85~!sj=vM$!xOtWFmPWZ@SUZ@_mz+5>$#WH*=fmCM=X%K z_sHRcP(`%S*Tl~9c5-YajQBldn96Kb9?4uY`ZT^#qk$!~%NA=m#Z-iCGQ}noTB{`S$ zh(FaO6I;K3#e;fIthU&3EV@<;?=~KzPv>&}Wxt(x_sS;p*cS~u?fyb+YZT9*$&OfHZ{taFnZ|HE_6c3ZPNJ)1Wnnn>1O&e9 zMMG;&qaOYOhd7)yjcpe_I+i0v!u)egvOhT7D(W^InD5OT{P$Y8l@qLHlV%K z4<3Btd;sS+al7@S@acONNQ5j0zke6tuZJF}FRi7a%V$$Z&TBSlo&{cwu*A=8_INwX z8>`JduoS~_Nv;cyeN)HD`-ea^B_E#eoe!N2_xS&AJm7_=F62jSji(X6!-)*%(XOeH zgEM~1;h9G~jITLNv3)JQ7STz5NKM2ck9n9{8;1>?hCbx}d2p8%fj!Hw!rm($;N~d@ z{oWK?+XUzv$YAM^2N+ATV0zFLo|`DqD{Tp+)kT4{3679oomoV*!klDCC1F?CE}U`I z5@qI{0cXw=ar*0coN@j?IIcAehdyax-nv4tuDAgc-t|J?h&K#r8$#))e(FN};e2K~ zHFLbooZZnOXjh%W#BRU9jJ-OEb;mNWTbax7IBW^_pVi@(=t4Sk_F=kyG17%gD@ZxN z19q(JfQxtJFy?L>Sy=r0h?Fv5@~5bq!T| zb)gfij+6UwYrlZWt<@j)S4fH&40daCzkjoPTN0?J8YhlSC`+%+jT^&l3!LS%|?_ zz37x5#vXYez%~l4XMJG@d-=fzwlapYgWN1SA)^y{pF>gZWwWXOkRC7TNwSH;7VeCQ zI}dXji@DxxG8>ww$&P&<$L>v!7yLL^#yjj=%~NRHjb2g@Fl)qwJ<;-;p1f^D^Tk+l zKz0D1&uGSfGt$s&`5>;(?M7dE2`6ip(^CFaa90S2eq~{^dk5Z|o$6>bP1Mz%_l97Bspd+Nj zghu+%yXI4=^v+1q@-zWvWs2Z$YR1(Pjo5u61$&0vaIe@i67|lCM3xLNS-+iNHE$A4 z3(93|mFtLp-Vo<~-b}Nn>aZ10$FoDXx%y^oD=YEGhFu}hVh2*rK$xZ@l_}F;YP14S zE6NwKY&*`s+62aSm7K=qIykFU^K&kE@hx;Rc|k{`sc%6h-IlwDjv1S>JMM9Olz@9U zcHEf!X`V;Ed9)Cp0*Y~mgz$OY9Z*bI#@hJJVQ04KuzuVQGg8D$5Za?ea(Z{OO}6Q* zwZReAaV!t2@*`pK$T4Udw-MgPorNrEOPU}11+}OpyR7;;4o}zvn%mvrKa=|q?X?3o zm8<~=k5DMz-9{HxmvTOp7I4{HPEWQRpevdWK;bG)&WB@*@9uS=Jp!)n=)_&mYj8AZ z2k*qoQ2vwK76Qu$Wf;#Gvsx2h;raPI94VcHI#XAp#YZI)=oie?%TA>igX?L{#_3Sb zJObmUdb-U2FvtveAZJRW@3!mWl&n-7$+kfy$susN^#>}AXx>{D z>b&d#HSjlpWnX*fi1j1p^j{&Id&2=A^k2mJwVB-huz=hOSwk=V3xyfmt)Saw3hKYw z0vjx&iH}1riBAu~{=y1KvyOy)Qm66i=B4brhLcz;)kl=96lf27fSD<8gU4LCEP77Y zsoUp8UyZkg5{1i9B-aYd|FeK1>z9;2V2(-T38&|>hYhA@k@TI!lGo!s`xmmqRM2$uLeS@!2Fn6uNx-gu=(NQf2YTHg)afW>DP}<1^M~+a7$LKq2o;-R82JW1Lji;ufy!1pa2T2|KoP??S>O}Icuas&h>%h0rTX6KnUfflA zoqU;ih`9CnLFUPEOdY2!oP4oe@JVPE&3u|bpR6i@)Bi=%q2PaX!sBN!ahVeRkIVWU z+Ajk4yZ6w+m0H}+x(5y#od9``v6G#igQ2FDuw~{LtoF0OaBmU%B*~oE&F!JM{VhHH z_B_3#%k3KUPEnOZ3qT7}U}(UezLcz_0nQKT(;O~uSu+6MFKUI*+jgY&+hHcHrHncOGNUy-}wkB1T zzd+(_W$Dk)L7X?!8MZ!Z0iP%Tp#7*ku2YJHJ>8wqx&0hWyds1L?*4>-KR>`%msI*) z!-Znc1gPd8<5>UkRB=)n{UEXzOmDA8@thj8U7bwSPAQNrTF;ofiZ&o+R{{K})hP9E z7L^Jp0Y=jnmsOd-j{Xq*Tk{6pRyDwwQVm?3`Uw6+*@OOq$*^7~5qf;LFw0j*F^BYu zn4pY}bgkME8ZA;m53F8E7s<_o)(6tK&^8~&Z0lefeut;8r_l2!ISq04I~tzI(BVfq zAkXpV%Wv2K^spdt$qHWe+mnv1FPQs*N%Ul5Ee^$>LuJ z-9D<)*GbiP_`uRa0o=Ef2HO-XYB|QS>wS2{`0-6tIXi$0YdJkgYdsc+)Z)~s)tJ&O z27ep~Brwb2lgnqqmn`Ji0Ph*k@1wl(hbFuS0mf!K`+hKefZKnM&U7;t*?Yh)+%v~ovMRJIR^Jq zT}nlEM^KB3uXNbomAX`_(J)k_yFz`KyYqK4Z!$_qxcoz)tsJLd`BHKvXEkg)F3o9{ zoXLYNvv@(ahWxhV0}Q=>2tfqdM2hRojiA2@%!Ot0l_A(?j+1jajS_*MtI zcqg^~L-Wlg*k-W|@rN1wm=aHivNUMqXEQouVGMJB=^e&otPAgN9Ky7z!t9sp2WW)k zL2CF>5*ETb!DdkhzKOLE|EWp`QyBY>xehOxTh$vedXp;lo~-AXmN8WF(8`v68g%+1@>AKgfn@e@cPv{l)sxLc1J*l>Wi{;9u-o)S*xGkp*cIo7U6VKQxI{VQ9LG1N zrz|RQA?`INZ=kSkw=60f&MA&C4d{#kzHFo{= zhbaqnX|y5W{7O+Y^AyLi2d4O2VLcwU$-s>tN|_}$JOyW*zY!`aOpoWB=S7yL^Qx66 z&MKO(gqJy_!9z`Aq=!z_7>)xCUd3^jpRV?Ix%IK%9#=N+Em++~m|h2UZ!k6j94pc=%@ zne#2l1^0K9*6gD8y|MJt-d7Ay`~W#7$%9+hyz$v3eLTn4!N;eTnVpcD%zt8CAxNK- z!ttf6z%=_4d2u0xW-h))4c{Mt2T8$fck5!dd{~@)V!MfzjD;H%DWZ{9KEv`u9)XNf5e9G^Q3p;}bNo^sH9S0vv3{z~XxiLj z#^lGDHw0^%zaN}zp6XCR&(a4pv04HYUNk}Jn|`QE;4+3{f8giUk8$OxTd3UNi3=Rh z;f5hm*7`#TM8Bv6*ZSXZ(6{g@29Q|=SvwON=* z9KFgEZo7ywvfprgnGrj$>o43fl0iq5!#*x=$MAD6{ys8+tsc67?iv~_8ClD2{V|ua zI>t~;#(=NQ(BPg1`tZ*zT4B*k4Whg0ovvH-vMXU-a%0$*DBAoQ}P3B-r6i z%h^Rwjnn=~a_d+HrFfPfe@Fc~d-4czy`u(%--w zX?x5_DFQl*{ldf-d)OlDdRVru4%YnM1^vhFK#6@7%v3+f@f#1()4YCw=6swWbr0O1 zUxzKfbuoOBIL&>nO_hTTDeDu0dY?UTd*ePV|ItV{T=IpEpOQG&IGioEChV8xLpbwX z1kU6bOTo)NVtLdk)^Hw*1M81sltVR}c`1VJbX8%egfubD+0t+abKs>FU+~U4RHeiRshPAj{^QSpjUD~uxwA234A#+Z&IjMj=@`V0zQv74)!1#%%Bgc% z2gKQ)K!-yG^uQ{XzTN$ht3^(*ua35}VGGW(uj7sJh5k#pc;p5w+hGZPV=<^Lbq%Fe z%3=I61^j4q1rxj1;q|s5tUK$0Qa-b3`r2hgN!A#|jmmMw<0IHq_yv;X{-C&o4ExpU zKhznS$X2E2V)-;K3$OPo49bqsrn+`8%GcyMEKK1ayLO$QYAXsBL%XSD&0`pG5@!#X zim~E<%kbIDOL1^@5C_d#b5dONo69tXJ8j=lI+ZG&~w1*C;iAI-Np8x0SHjA!6&~HP_0r8ZfCoJ z`IZtq&Tx97_Ih%wSeIYmRE1vWYcasT0skFIBFuj?>CPj^U|irNx^A`@X&Z^AQ#>7U znP4pj8()Ukyllv`aKz@tUD)~lJ@cu0rl57ZIq8lyfDIq*z;nnMyDl^^i!9ecOXXb5 ze^&`_dzX{-|NheBMn>!;*Pl32(t#e`#n5i}ACx#JVRX<|RIf|IyET{b;o34N7oUdl zd19#QO?Z2&xAMKmO%s$ZZ6I^s&Z6SYd+9w(A@^#M7Jf>p!6r@< zeLdhA?dfQsJ0HEJ-xfVVkNl^2PWTnx55C3C;}W<`o&)eSbQ;W_zlENwdQUQbq`*G zqIOpLFvac$HQtsZfA)99QXD0Qu`yPA4Ua zQoqUz3b=hs^f;Lo1=04+d!PONrKByGK7QLiwOO1lbYJUZ;gW%-H6w`ui<`e4?Gj)fGwrq4JYwACX zH8$QvHs2g2+Tq&t!KDrO)X55;Hcmj3dN+2X-7@xEoF!|g9S7oB832YS;em4>uiTs` zm~*5_U}Wyk<I@38Kx?%Nwciczc>KZVZbST*jGfKJpeD z!uY)rbcwJ(-De}gb{_hOKe(OI%V|$(D)WXems;_4KNc3tGRgwyTet73v;D7Qu~i|y$bG^b zW8RVWf5p`Q$wPb@8;OO<8t7f8j2F&2hcdcSFOGEx*vpze@MeQ{@25V&(l7uVS9)O%I#un*ILiR!<&2@B%4PIR&cqFz zcUY!76W&K=LH^D({-R42#LvEy;;M^e>cTZtXk9Li)w_goRjHU_6Np%5%S}4lQj=N3-(_YiqaBC`WZ%GqU&d|dvx!K5m5fuDvgf$D4QUCS`G?Yo< zkxYN~>udr0<%$SwUEfIsuy18)G`y4*pm5QquZRGY=FCkBCAx^7Wi~3!OJXxH= zyjM3RkGvO&+Dmme2<;(bL@>#I^R0zWf zk)ZZ9AM>Q0(Vb&yR(l?XldE>Z$@mv^tnn_FYrGd`O?pq?Z(ISM?J;mFAQ`H~Z{oxi z1DMbz%|1U(x&Eg{%8l8i2FK+U`WLfhj?yKd;uWW-SE?oYoeLCzZ zb-)$xSk&z1VRBguz4={&>dEQROQz*mAz6lN_Z&po75Vsg#s>(xJ_Q`>S}@5snd6nY z03NSiOF+IN9K~xzmUd9(_Cw_7 z@GS_P_#4849zlbADW(=yVOaEe+>)d~M^|T&@){NLb|{6*ZTk*4LY`C6Zw{atmIwv+ z%7L1w;B!Ysl=|BVSM7o2BG1`X!qeF@TNbZ(F`&oP5`7_A5Did*mLp5y+Pqc^~Q0u zdy;{c-3M^bOAFTj6S8@wCaiv`2EE>OlLLMA$w)%JT_sP2rU0LK$XWIfZ@NZsd(-)j!94s>F&4S`?)gs zrceP2itFh`pAuMN%yAgLA3&8?HCV9nF$Qhx2e;20C0$1ED*e_s~m| zMBvop@3gLz^Nakq18UQ=>9y`c8hg8pF6K$V-Y@nr;a?nll$V4BRWTe}^AGGAD?&R? zSJE^7IWm*Ee3Pl?ahqE!2H)K#Sb6k_pjys@xjQP4LPr4&!?vODGIjb+YYBuE*28s~ zskD}Lqbe8d!A;{lT-0`f$R0O;mI!lC_doalw+mqYA1{DMnE6&_|BDOw z|Ka(0rltZOWAeXyew*tz8`Cy#MN5%dwxXhs)mmBr=kyD4qGDuY)AdPBUw<-qyHC7n zQr`~zl_J(;Vx@IHLNyXL%yMbGU3aj_|DSwQ`t0?M+wNtsF?A75E4-XojmhJh;#&T~ zr8_P(s;aojEmn<9uYB9+et$a0kH})3Mj19N>s+JsVsAFF!lf~MFn|P`G&atk{(mSs z??5WwH;j`lTah9}BvB#G`#h9rXs4tt(jZOWb}CyUTe7m15h>$5_kC27(li>BN+lXX z(oh<|_xIQN@BDGz_dW0ZT=#W-u5rxg@_s4`JIgNV^m36KE%Ct5o78spHTqcM6gs)) zz^kqY%|}=Ar66dPs#bKcH}HGq&7pc z#BJ7Tx_GGzb%@9pPFb8oB%7RQTG~Bg()*Iz<*h*<9Ek++ZQ(H5#hV)bNrP{ORf*MOWc0}_n>6-!gB1Sii6%d;_mI?wIYcqEiB#-8Osnp8fRpnMI_1@N zJSjbsu348zFaK9e<71?##h0TbuceU7xtGd4>IYym^?DIYR2>x|1_Ie#mO>?@)Q`-I%lpZh_j?+v@i=Mh6Je{zD% z68Hcr^j_pK^7+Rp`gd|A`Fk%AU!1Uo1wl)w-&=pUHu8<|%oIl=@#{M77}Y_{w8CgJ zKbPF+85FULb?Baf?Oe>m2=e3VV>WBA5gln^Lp{EXrE1^I>YsdZg)Oai7z0N z^DdDeldQPuvD5Hb$41(Ebu?8?=pYNVuacKh1fOl32<|6hAa0C;?eFCMM1OH9yy(lJ zQ=0jApLiUmwb|i;!%xBa#1{dYZlm$5HRy)5Yhn35E4s-*jJn2)9rG{fT> z$xt|mD+87gvAfTO>kF<4>oV?gW^uZBW8n~)SH6K*@;XLewJf!&&!oroI*D%KOx*O@ zff(LBLjG$FC(02v^h^lA(9$e2Y_2Zdd~Xa*|FVXZWN(Db8;|Hki5}velT|&ErbYL1j8sQ|y97l^rPNR0NmrhY*>s--vyoA|_ru3#}cO@RTSQ zW_ShRyQFZU#75zTkSK~1H_;oWQMhVpA=OiSDg1quGux)qSL;2jAlHlQ}EeaV536uasIaJ-WNq(94m z+i~OYi~kA~PPc}#Dod(=Y#Pcb%VU?n0`8nR4-0=xA}3V?z$NbjSU)}lr3+WV&js(H za$ZZ7|Lthb|63>ND((U)_Xs9b3}tm*qapT?6}|hqkZx_RCok3gP-%^r$nRn+x@iVe ziJ)Ore2OD;2us7jqY-rMq)1BVkB1N2;$Y~zE($%pAg1gudGDJ+|Fx^zM&Gl-k|UYm zW@kV>_+HCd_Pg+xHDX9h4V1T8VC{-Ltk!FySFVX@X>1Nvf5=}a9Su}YQn>l`Aaozn#^l-7RDRZW+9%D?gyO}t2wKVdh=o|J)<-rRTLA~|m2=r0 z2K3L9%dlyc4|GgQgRD`%c;Dq!`fktw?d_@{)jF7RZl7^uj3qTz&L9_(HjsMl&mheI z&d#$ua`#EE&|b`yTo}}>3eCF7hx&m3rVwQpX?L@v~YJkyq8C?NWTdMM4zqiN67X<3p)U zoD`!bew$92FVF6(JkvNh$Jd>R+#}FE)M2S9rFa0ws!K`${C#pCgYXsRxoU1Dvh1; zf#A+j^j-8_u?} z<~BfPUkKZ5@gACE17Y9d37oI;Q{kt7&Jdo|R&~)*nN3NkVpqDPaMO~FWH-(O$HdR@ zt!O+OuRO?n;)lbtg;SttVJ&-P;s7qw%GkN92~fTZiK5^VJf7FZrk_;?zjaZZ*1iDt zchg&VXcY{3drH{^iwQ#4@E@FL_8np2OpZHim?2!Tsf3B2c}Eo9n8Cg3W~LvjA}~Te zi*2zHx=L4YuNRD{dih-etapiFw}%9r@^Xde``k4!HQp0~+D`4-R2w0mn z#5VCW;{{nO5WJ8UOcKV!?$PHAr?sHM&wehfqHmrZBC^UJwhx=>p z$?`09!sw0i*vTWC&8Y9ZKP;R z6Q{6g1N@5~4Y^hmh{s_YvJa1PKOE*VmBXDxbZ;@7(kks~l>sd}UsG7f6J_mT11P;C!}SCRxW8ac5=Ip~|J0doy(`e42RyEW*FQ z?!DjPRjCY|-1veVdUlIvjhjPK!(|w^C6)ENePwiOH47i~!j~@ZA#2VEvPEHl&H4A8 zc^rQ!=v%Ordr}(>ei<$-dyziekT?#V=Ehjp69F#zmzl}$dQvz%hA}}hIaH7Vt~>Ri z_m;09W(u-ir%G<u-YbeW!69#33vu4-P+>6`e zp+96I1gj{)2j81)uzEQZKHfuoO{Q`G%?xGUN;{dsuNlndvmtrmx|TRMx^sp16Syq~ zA;jXsIdJbo;Z~n7rt-ka%qw=RD7t-PmD^*DhUVg=;tP%uzG;`*%Cpg!*9o+8<604#TF; zN8s9a5nShc!t0xRnUP^KgPq=Fgw;QA==CRkUE`s+P!{Gb-vjm=CBd=TpT+wv0fWdS zcF9VDcsx7`?fqHcAoYTcWdqFj_9Sw<_#+v{CV=|O2c+WLE*Q!7km_~boK80(vOx;e z>hWf#$2f4&ybrO4N^Ibp3_a#Jov2wBz=s*?B<1rHn71~9 zXM0Lg`v3TMR0!J*0Z+*p_U zxN>>|G4Q;{Jk>UVe~UD|l>CT%Yt01WF_qj)lAx)QUBav1?{Ej!o6&`x+sNiPCy>-G zCojFkP}2J)9GllAh`tg5O6O`}*!wnEwsZ@KbQi(<$1QwU#BEM5bpZU*>|y`Dh4i$F z7X77}0rhjQ3g1uA0*MvXka?mU2Di&WO4UqQdRLS97ZeE^jJ)97xwUX$^&76E`Xjur z&4tDb3rHPbtg`)8Dg=vENbhm}J$J;7{8^qxZoSo^zoeb0;ZbS0Y?sL*URHt}+mGR& zLt(cV&mrm_C)8f*2bo!8K*n(y8}eI*=OQb}xgXi2)KynFqvbVcvSS=e(f`c^%lu$r zU4UTCh~v=TXhFr4C)3A~IXI+5ac9k1va;Kgn|jHg#7WH}yXPII`X6S|;I9< z{jMpW8ykW>J3Wcgqgr_EwwE>MuLAOVI=KIAffU8zoN3BQn2{L@(i3JANz<*cBr=vM zRPx?ZTOBa#v?t=Sku19QJPA^0r#Am|!P7gO{M~9soj8L zv6oCz(`Jd<>U2$15^yRT>AhHgnC!lR%2?G42d6tO=qt`xnCB25fM%AZ1AFiXD^Y|<@%&=eh9g1WP~QY1fQ<7 z$G*qbG`8gzFIuxp$kJ>!{59-o?Tj2`TRCi!sF9!k%va zn+k8EPm=|$N7&KsSiXx(iuWl$76@ZRf+bop!uqSXNLgqj%{vzdXMX7OSxjKY%}?N7 z&TDeuFwbu(d&6_vY9McVJ?|5c<5K1p!MV5@=sL^+3xCQ$*?JQw|5->vi%wT5YRnYG zn_R%oI8AtZz=A4|Sq3AgoPc7LZ0K2IOkciV&ihB^(1KyN$>72J!WrBd787 z=N!lNlK$s)#B4OruS(LH=dfXSdy@J z7QXnroldv0LD%dQ@JXStaQIYwl(h@Zr(UM7Q-e4;4Mm#!QtK+oK(J*pB z7`QLd#%bc$psz%Wa=N_F<>(d2DOX|(QjZhul^V9*CG)UfE+0k4dvLHJ7v`jUP#3!z zxVzPr?ml!tuvTOTcS@vbs;e2!3ED(8ov+}PMXRafu#04~<#u$Cm4ZcUD#4(&l%8Dn zhaK#FN^M7chh0CNG2H71*WY9smM9oAwdV6FCa^q@?4nJQ>@LkXB}W_z%KJvCv6c^hle;Jkj@X6%y~WV}{XLPg%O-y; z47i|yWcu)vBA)KQ46}NlkdwWkByHbeXmZHJSz|I$*GZEK57dC-;5nRMtB5BfzQYw^ zyztV(ZSYY!ladQMbi%$PFe3UO`2SjmWn-`4=CT_4&awsO)TEI)(=O23H*0atLna)~ z($V^65>-*zN>1$^q5{3u@VoXrjNh(7qb5t?e@Uq*zg!1jol--``Oi>lcLJTPv!2(m zev;cEOX>8iRp3@D#l8Qjj|X*>ZNFai;r>di6Ujr?X!Cm(6fLPHtW1I4^xKGYZw`=r zi4}P7@&hQiHJSW5-N4eH+hRMl6pWHqL<>(2y(}i-utWt44QGV2GBV-0IiuxE++m!+ zfjaG3N_Qxp5=h;TL*J&^G+%NRh4gr0dO!jNHnUlCt{)N&9lRKxMZdmo27k$5Diih) zJ4b#bJ6398bEYIc^GTiD+xDF-xo{m?xHh)?pf%1~UIEt9y#A7(1~z-jVQR8BR*mrk z<41BB>^zK~{+B^KHp*a1nSh=?)k``PW>am8WoyO9Jsec~V$(6G5Xv2{jYs@b{M?GHWP_8V4SrTZ#-}$hcq{*O~;gMlOZ1 zzP!G8Dx994jPRW#!q4xcF>RX(X1d;_%;6tCn&C={tCZ-CP1i`d*(DhNtPqp~Oi}y3 z2)L8!!~x&3DD@^Hrlbj8lA(ls<qz1Td~mpw2a|5dY7ZHXkzth3`^i|NBU) zxHXx`9-0QX>IwGzEFv@Zo6*AFK_d1b6)i`4WE6O+Vy~&3J`r6JU$k&+7{L#dw8Ozd^Uk_O4 z&3QE9nF%=(e~~N|`^~)`sYG6ZKg$gBA%;If$?x{<+~bG@Sbky!{Z|zYnTCHzp~ijk zc!L_+9uTLg%D!Y|b2d@EbxKD?7OA@-xh2?WJJJg?`z+cXK!@!Ce2 z5x-mT9R)WJwUVLnr`Z=@9jKp-+&060Qla_~>V3b#kzp!A+s)3{}SVOJ`MZ%JaW7+bN_PG8;jKEZaXUl(yBt3b2hRt#j5|G1Z)e?elW1m3z z{b#Inu@^|lZGcsYLCE{5c%A4s`$1E}vBQ)+pm3B>iX$)~5Xuuf|NRdMR&PS31^y>Hu?d!`F97f(cq{a;|k5_4Yr zV4Ux}%h1ufjHUc7Cx;u1h_tgIo!q#cdr{noVPUnXR+tWDl5r$q@k>GI?#*mxkR+qy z_}@V`pN!HdCOt1+5x2LSY3d~fx@>(%RnD!M+)(pyv^ymtcUJ3yOr#FJ^OYe7Rr5&U zi#_y{`9WeScF1Ps4_!7Sqk!>I=U`jkaMGZb%oKQDv+(@*XY54FTo(6t8#$!>nk}MF z_#)}sWbTdmWHD%Q8(#lmkLLa)CEqPDz1R)I+cw}IojQ6nPR3Sgm@e72LWbDxaDbCe zDcs%lUS#cBFSvj444GlgXI0(=(Rt>DG`!6a^;@jy&maGx!PgFG&HD*?hlI?g|1_KK z?M8L^GxL%!Z-pZp_Y1ccnA29Le>@|#kL2`Il2w^bPVjr!s9ZT1|4vg zItEQtbHV4+AxLCq*=8DIO5W&j9JN3*~8#c;qqrcVE*wtHr><}>?lKN0PzcGPvrzCg~ZzPhc=`f{oUj3<>V9$mtPs)IxSMXq~a3;*IZ! z_Ur=AXk|U~nSF^fpTAJhR3=8Qj|!m1odiAlV!_(wEs-*s1i#m3vk(3pbXY3mD(*P< z&)JcLuIdo}{?;b64b>+bDnnt_w@5alc?#q_SBD+tNie@>p>Tn7G1vc}HTV{W3U{WC z;v9xFNc(OT@Z5Bs*qM%@(!)8rYDzvUQQ1st2aDmf=^$6B|B#)T=fwr|zGO97njAf3 z#I8#IWzxFYz{H$jLZ}YT+3NGm6;cVMJruBxv zD%B?8#l}vke-{iKpEX#j=|D{CGdPD==LHX@SK5Sj&m_+p8+aD#Th6BMKBw$rNtz52 z*tL&xIA2neXm6a0nP+da|Ei3SK5H?N=er75w{_Skx@dHqhI#IHB`!qJ8A!UTesW}vBlwu!=J^&2pnG8gO!@JO zYrpM@TW+1>rg>O^V3ZAPrqX2h#OaW{K7qZKf6KP5nhW2vi(w0~Cg$^Qlf_%^!HJmz zP#>MkSuz`py0l#=^8N(Di*AF_m_OitQVgvY6~U_wb*y?_F)WwW2GyZPVX{mt8MpZ^ z_s}9on4~2SJYrbr8MPnMvM;eX-D-9`Ar7RXH^4pBg|PfwF{$p-!bRKGLZ)sZCt1=C zsihgDf4K>_b?F$8OpmNO6*Ua%+7rpW0R`Ns`WoJvza<4pk~nW$2_!C+VRdzg8lx+? zkmhvm$j%IbL!3KYmE6sHkV44PmrFUpx@j<(C~=3aG~tM3BE)c|Y3@ZWTBPgCsv`r8S_lq-jUm^EHt)gK4-hI5WJNhvAK&TL)Eu9u<;mxeXD~odg3E?P=6wrWQd@5-Wu?9Sq~CrqxpUS z3tC|&i?8D9n4O|3Tu>=x-tD@ysxSeLCR;)BQ7haL`9-+R>lq_|#}j|$C~ogwMclYy z9gA|@z*d?@3zt~$hIMl$*pyHB0iPCyv-iI>sabpt(Rz+>=G1D)UDE~zPyFHTe+kSf z;~rGbwtHoiD^|;Zn)lnWQN9=~_xwoM4i4c->XAzOEDUldl z!^NLTAP%A+`0wZ!2;%u(gM2661J9zvom}&EweeG$6 zJiGW7*ME5;W|bO)(pLp?&b5HG`%Z$3BmS_aOENGMl7+{V)Yxbx7JHMEJ_>kif-@WHC}OSuD#%s2IS}VNh3eTF;K25W0y;7Vm-Qm4Uo1{E z*4A+!Ggt5;%XFdF=@SrhS`k-UjmPmg!1^?1;wRR^jc-$>+sPODdFgvPZTcG;D*FUJ z-aq)a{66qQM!>;UL z3&k%%}K7`TcLg4 zb}Y(y%`N56jT`$+sA-@Z`(qVCuB>;&3UZ5U`}>N`uihirIAsQO6l6nnnF;ic?E&l5 z65{yQ2Gbk(=i#Y9uc$WiqiEKElS=l5zR0HY_Lz(-oY!-X_JLrVZ zhXYzg+(;Q?GN9p24n>}3qW7Pe+T~9qcaj2G(kBT&e5b*Q${Hs9%nY>mm67ooYhk=d znwHG+A?MeJ)Ay#;&^kUAP8SV9Zmce-ef}R*0`O-z;(}! zIJ!><@7{DLUBW`Pda)K9mB{DXJaTP}0(QVX!vxs1$_m`}Jpjc$ufRY;1*AH8t>_(4CXu z`T!ax783M1({#gEN zBwnoy1MQ1#+|#`Q_`#`1@Z+313mF^_*<%_>*spIKm2-kG)i=3_7{24YWG$?VOy{1N z+F^dhGSCw@BXup7xJAkl%ZhpS)!9c}@u76SYxoa)u!Bz`1Nl-dhu+8$wyuipZR_u*KUl@ z_-?z)yMv)LMIE5_K9oqsk?Lvs=%;%QrhY19uRkm#PHpo^Nb7m->UtOUxiJX;3#{cP zS7(q+w==QcW#a zs3{Es6M-xIJKV^wzMBrUatZKDcMnNUKE*X8-xBo7m%vkxab!V*H{`Dl1!3zbysEyK zrv5z-Mi=&Qw;n5kP3LNsx}*(S(ywvjGlgV!#XXX~Zw|NYi8X(I2x0f9&7lgd`s~DS zLptWHHJ$lq7q@=hbU36eNp3~Dan~|FGly+n!iG`Z#H!&JOFt}4D&HHpj%e4eGAl5w5Q5N9%)N?mCHRsB%%UBwm+zC;MC z!V+N8>}m9#$5%3dlIU#M!8NG{;b(&%q=N1Shq+e-68aCx#>JZ0w&6R>;%C~%(QRae z#RWR3dYOFPJ)7$|a}EBze@l{IOc69pc+O^Dv1eLRrgWLlD2&N(C-+N-2|C~IB-7v%cwH9Jd>u%N*a*-$Sx+o>rxEuuCHjij;A3iwgoiTqKw4`Abs5ry#+mnt z_?c(GDhoKzN7Fd>(Hr4h!z7H~|Cg<}`i|3GEyXsEuHup=or9Rl;|%=& zGe5892z+{6E(lw=4xTQ!#aZ+FU!}6ycy&@Lo>3`>mSOy!x89%l#eD}mtyu2RZ@!nJ z&5w=#B9D5~|48K4AXaqy5J(>MhOG2;P?_dwvphhWUHYKR@@o?D&+F@KU(W)ZF>Q!^ z={!#kOEz*BboRnXZC&DT-UenCp9OE~YsijatI=<-Bx@8zlUviZU>P?ICl*YgR_ijz z`%7+g<1J%$(6EHN^OYmseiLBJh67xvR20cN*F{dQ19D*bEN)L{0TGroa$k;|f#2hl zAkFn4Vb zIXY|OPQ!F3$PjaBwzDi|0yQ) zGU7fok0k~VeK~`LdOVY|1S-BRC5=7>WXGTa-V-R$q?sdF`D|@kV090M4!9C^=QOZ8 zv;_thZeey#S%9)5@NaezS68-!%l}itW}Nz3Wg6H4Ek_9b9hgV$3hmKvml8d2WjmcS za|>wR7RML0li7u>@4&R%hsL_)adp%8*k}$tA-QS^?87{s=T+$nF6RP?s6U=9R=1)j zhU|#k-YELO`GVj}*=h2qQXf+sFj?9QVbaHyA1l`TmmX6tLvx$~R*eoU4Y zFLVOs!ryRr@?~<{=mCy+s0V*%Sb@a_e@@|P1d;3)=XLzCSQcOg^|fVWO3W>|{%az! z-=0%7$Gicu_%lHF%p*j9lr0$QqzjJny4Q}-=>!J~1ma^&IpHXK>fx*fiAsC9qwlh8 zB%aukic+1b$T9OU^j#rQeX)9n5@@pB~3(?X!I%2Bb|eA{GA7WGq_BD{_KXrSuVmt=_njo_lHgW z%rh06y-A}>Hm!W}6vPdu;Lf)pWcZO5C|taUPV!NNlRS^w_ehSgq|O@;PXH$B?j<1` zeVOVquy`> z2KUMHkWqB}O(&?BHWJdh4Dj7YCkQziO=@@k;Sz34K!Yh)VgBcS*2<}q>zl9gH|i;l zP9Dvqjbu@+8#%3)mdwszEU0+uGM$q%@$z#uw6~I_UvJ8xsk1(uzWtD`;Nsv#K_a`e z(23YIy9!l~cCl)nCzWI<0~Y<0k#jRu`Ic7F4JN;4~nprMEiZg;2md4|&yKR+8eB}+Ev<{Mbl zuEAQ;?vNm{Eb@NkSFq_nPkz0hgYoXg!kN8GAZqXpSGOgfy*$qM&AqjUk>!Wr6;}*h zV?4MIE9X<$GD&15|DoQtVPJfBm_YZ22Yje+;0CY1WEls~Q{9^@$db2@$c)3j#ZUlF_$Wt(|zmZeku!q~XwL$P-+Xyzv zq6N+$kLFU83~08lI+1cckM8rtLT(u%;7hK}m2d2U?^S+bfNg+7% z)jYOg{s7NI{hR;*s?@4Ern(0TqPKFD#Ha>$F zWyNqeP>Z!pya1h*2BhZnKbE)Ew@P{5Ij$(DkJwMW%n2@R;kL}?Y$K0WlO=1MvGc(= z^5m>G>g)>Q)SDg>71bAl?x00*LvAsxe>{xt4wWR@{}iZ#-CPn;+{JgI-C-vUSJ4+1 z;}~1cvnp{6@5yt=jaj@F^J68hY<QmF%cpRMPk*6nV%L$4qJJR&UlLV)@QV~| zC?N6Y*Mpn133c1G20tY(q-zqsv9HP|j?Kj5i4JTRWqIbCNnGkPFS7FM- zc-r%{n1)V$3E4k%sjqc32|Sxf2ZWPoikU7I7|z@hZpWWcb0dv&oPhX@w2p=170= z*xet-#Ac-gJ`q{L2}T&P6DX*03!-+8jG$&@^)$q_tXdmgU;=qFv9r_!25>$#+vvtf3? zE#hr1Mb};~!j?OBR5d%5XjP8l>RkZhA1l*{eL+N{Mgos7E+uupYT4 zuc3c?D5|fBgN+k9FyVDR8iaMhz^Se(qf8NWm}hgFCDu@di6_X1;nLLZ*Lxy(6hkI@ zXYyTuyQ%H87z~j%;oj9d(UnIkV1%p<*|L5E$&R3CG;#$Oa@h+zj%d^3#}CP|+-pR< zHIw_XT|`E#Uj}t(PSc;L(t}#piEES_u@-8fbJ{7ouK5yO;v!8x7-rKT|9CubWjbyA zoKm%RT_xo0^%k5>Qlk5#khJo?g{v`xEbo>dUR0MvhxYMw7wF+p)!$^{NuFOfBLTO4 zj-v-_Q_$shHrTB$B}{P>6-`v5M-N>lFXLB|7{l*)qvs@a?z@Ys{+;CD=$lM>lR6kb znN0LNR-?L=KC$;+NY5lG!Uw7)a=K`W$H`Z+NG5?aCGx(z3#>|d+ACHf_8S%quf|pC z^{{j5d$M`zO_<=&%wIAi-_AR1M5$EW#r>rO58W-;oG5P@|E9%9bYOzh1+M?hC7Zy*>oY*IvPzM z9y?0^WB&Ac-yV9kN0YSesj;zJ5Dv>%%c1qiEHK{cjv*fJSoTOs3@km%Ev~%`zueNv zmuao+lk^R^QvVEeFHXRF%?bFrrJD;-l);l426(;v7u=W7#_~D!T+))SaQqP?=I$A2 zbZjiH(;ZK4zsm&^6Fra`&9hS9lyY5NZE!(W4;{~M#TEW35VkvpJAHgLmKu1W=;dbg zYWxjheJNmY@i2NH%z?c;4|w?T>$rYc1HNgwjSCMxgWr;NEYlp< zSWU$>;UUbI_cSe%4FjEu9WcvU4Uaw2$H(s^;jwEgT({p25ky+F>y|CPEBXKzR*T{D zmlv7N=NS6ca~h5 zjEeBW#6?(9dyOnx9wm_8Q^!Tt^}*y#x1e@jE_cc22AR1=fSuzWKUG^fsJW*~2cp>}PrTMa<`DBA8V#;pQxFgCWy(+zhHi>IWMz(PjkaU^M}U zTO|_1*)}LK!1t876%ePT1Khmj31ob3E$4M@qNvYLQe;{jEgI0ALg$rD1r$cX?hQb8 zW$-!Z4=FUjY_)LC$ryUW{66R#B*Jc&46Z-Oo3=kofnl)~+zqt{(2z1lwB}bc_VL#y z$fg&TIFFkrKXPvO0MO(&rO9yDsv`OA6^AodO;|4&KYD5 z=7C*~J*IUJ!n2r*xCu|7P2*GCbXJJz0(&eGw?nB-r*UCM6O&T##3mOtm@RFHRd>VT zp-mrj%bN?$PNp-(##`tgdIqj+R}*#b8;u4=o0z#yEvZcr!`QGnEIdDn{aEQiH0Iub zxuX??yQOu}xlkOxR``(gc|}k(wg`4j8-*7uAHbB>-%zdT1#@myv-GFGS*P7LHh*&j zI6rO&!+1q{_?a5+x{{7IFTR0UMH#lw9)ZEbW)LNlTrx33UDR5wCK@JOA?hd#hrr82 z?1h&D{4&piLv)Z7i6`NPy78jxa%G?xW`m{&!f;GgB-XV0(;Gj{SWDv-@c#3H{WstX zA76a~ov2lUM8#WG~9z_cps}0R$^gOJL-Qo1*6*0Q2Ja%K6B^b#=K~zsT&0M zZRX)Fux z5`K9-KrcwhiLQKmhHoeHptb{(aLbE5@V<%9@334!(eV>3`|lZ8*ESm8yjNnXg?Shg ztA-MWiJ0wUAR3_%g%YRbMJF=$h^kH*iLN{Q;@Q9!KF_s^&nWB?dYW`V&*)2df8-cZ zufGdE+Li~x4proNHSpb^&v4&L1>=zyIN~xAQ@(^irte7-9Uv<-Xg|*PE?09?d_Tg{ z+)nm&$zd*V={7jCOo{AmpUd*Z7qImm+fn%023D=gBjL8USo!@Xm^reXQ}Dcu-MZO~ z8P3{FsSuVVFrH_h0Vjp;@7Nmxa8thmKZY!6!~6< zCD&5eheOVSaT|}LFKVNN*&k>!oQ^Ka12Av>51SMGS*}pJ1Dc#4ftS8*<(sTHP*NQZ z8drY7W8X!hzU5z0@~ay@Eb{}KEl!BXlc4FuOwe1&Phj$IFw3(Po`5MX^*xBw6Bl4? zUp!piIt8Dm64B(R6GVo097QYTW}~ER23)Rr3GQR1K=hQ5S>9#PIYS*4G%k=2R`ukQ zvobTh3)(dFk-144qrM9%GJhT?UYjJk26G@1G3CZ+ZNvCiP2;7 z5*Vr91VNqSv2i{>*ZsYLHWxHSW5-=b$2Tw8ow8RDUU{LabmwON{KNZPCFAM4hZYzg zszj4wLO>YO$K72nNi3(f!^(ukWMA|ITh@62HIHt z%mXT1T$!I|DH{kr4f3t;nENC#`gBP;IDX!Sb~cl7YE>cIlq3f>ucxy5P+JUWeF(DW zM-onL1}5!0gS{O^_&v;=go*j!aXx=pmlDeQqB6+1{yik@NHF&8N`td4aWL+)Dvt12 z457cyqR|b+Los=HbX^CyHmn216pH7Mp2n?y>u?|Ulk2@DDPrxHaiG!`&;3$?ki%+h zgW+nv6KXsz-#^IO=0%aI+fu6*YK+9Wt>HLydMe&?bmX?Zb%y?bGg0nL3zR4{!=~%Y z$>~7>uKEx~bx!UgN1I~NCO?&=g;#NRO6+lTYcw?&^;ft=ej`j77=tfN<2lLeUO4N^ zG<4YY5SOe-z)SK+QU5~^M7ytqRU&!rugyMYnI;bsi8D#|J4rf;^F?#jdiZwJ9rtaX z0kc<5$F^y!IG52+!Ik&gj@56%C%Qe5D;a?b*&e{tAMn7)K$x!^gWE4n#g|#RAQX2a zBT9mWr?>P&_f={1N%P_+UONXrj^{yYi5A)nw*+rb2h8143g-{xL&L3`Ofz3rq_I?) zu5;ApD$;L&+KduV5`KcmE*S_Nb8#o{H`(GJ&svj)Q-?4$9A_8Fl_f~f8mTGJo;`y7 zJvSdbEc;27OcQDh9U?Cyjfh#BifEX$5T$EJVbl_79P#-IY5(U)hiqOzQ(zDI&0Exi zSFgp2|Ln<3C!3>i3yamr*2q#fTpK@C3)2eE3RKk*VY4|ZD1JCTKN2BK1Fjgi3 zME1tGia)PJZ&jo<>y|*>W=FPnQ5qy~&c`ozEOEi}qj=`cBy1StkFy6y&`nmGVcVQk z5auO7|NIK_MJk=mY@LQCMoZzr^Dy`^F_@@N2uJ&4n^7U2{|u1E`L>pD-`pKSj+oQj z8>`XNJsv!!WQwMD8HgSTv{9vhG}#fb1?6X`;7^e`rZg&o+UxZgzE2m29$&{!k4{+9 zI{@D%#K5NMF>p!#A!on>Is0>o;8+`oig87_%+F2)L!P+H`Zz>so8y*1EwsP=1ix1` zV=~Vxo6;PJ)^k?i@Mn(r{^B^Od2LTD_P>W0lWyVtK_iiMWe6%R{VWt2{srBFEZAHx z1N(YHVEVJ|ShZA0o`(Jd4?3GBP2gy7K?t|_*(e-QJ4TdWe;qs~yd@v&_h422RBV1Q zgyZjBM(3R!;CTBLY^|xm4PCuZvfK|XhPzRlm0>vWARm{ED8pK>YV?@A4BO{QiXxsE zVa>G$>=))BdBKf-Z(1S41d0%f16?5ar< zZjrxYKF9f+QV^?G zu8)`1+WaECDcP+Iyp&L%Bu0-BG= zfjNzZU^PJp?RCEq&ERKnMN1y9jnl{PxePxBilN6%CB(h@U^eMIRBz>fXSQoaIoR7gLyQ-kMRA01tav3%+`2{;q7^1Pi3N3PXrlXw8@Nd9sa%e&n+CRrccTb1!^_(YcC$g}v?_@eUca(sXG3Vv)$#9(G7lGHqpBUGNS zqa}o1%>yXV3}8>gzM!A&5?Ze-FFN=Nz;|S#(omMfjD!Caww! z@F!+A8RcJzYfZHA`PUsV^_>c|YN+GLr<(XNIva8%ZJ|=(0B-M`A$pe>gTBknM8)UZ zQO|LX=zYpOyms;k9Q|Vn+|l#QtG@>Xjm^~mN&t28A*dgI4h&DufhPqgxKUe*sq*tI z;&M5kUQ{-~n;~zwRyQdS&r%_7W7c3=$6xsJDHqh+lxd}0Dfcbj3@xAY9T0ajs2K*a zMe{aMBVTzO!+t_RR}0aTj{}FIc-Rrp#XVB{BP_-Z#E#!ngnOoe*{iDmQFPvqTz+pD zH?zyg3K5b~2=PAm(Y6|tM16`>N>UP~AthxKEs|tpRLCgte(r-blt@cOgJ?Nl>vh3G7(>$nbugye7tW0_r~8PEeUP4y{hHKJ`yCISP_JdxuywD2 zebo~WHle%><83rhAejZN!YJl^?J*ZqsYK`a%Fqe14OsuE0Fqt!z1A9cIx#X2DoYp8 z+1=q(_k23ryyXfWRxM%rd8W8@!B`x;#&e=(B=R#uFI^eX0Cl=%^zoufe7bZinNq(A zp1dYh!X=s3bojzRNils^{1dl{EC4KZ!)K$c*-w2FT4kY!^ht`~?VP=Ij637n7mdY$ z1#(oW)sgy-N}~;WC3Jq`TcMlo7}_&Rm!_|tLyvltqpeOHmC3t>ue~hkxR@RY&KyU7 z@;$F_f+b)K-)V1tb`mu!Xu=~?)7UG6<5aMsh&X3{hv2X#bl3jG4ZdzckCthWd`HZF z^XU<_A0~OgmX&RICPRWY&v2z(>;-Bl#ek`;BK_6W0PD`JqjKS@_MDU)jqw;oE+2HE zv0s-n)iNh0a?FpZ3?Jcay)W`Tw=MYO*ke=jjehM*|bOOJn0N0(I_KuV$wU8YoU zQ=PsG_d0bmgA7j^qj!T#AH58?9Us6wq#ImEc7tQhDx5SF3P1Lpp}%vwLA6hUiJ44> z7>O0o!DpA7Z#y#LDGN3?Y?yg<1@uXJQ+#s*=ERk-VS{b#X>}ppeEU8sOv+(bA~kTn zZ!e5`Gmq-@TT{!g;@p>^SGc*M8CLS1`B@#OsN#QfsJ-DSE_Jyk`#mxY+9vg&wFyAd4^uo+fNNPG_aV<36_Z@YweQ$M6=nZ{%@_xm_Bol~Vf=;-_rpPByb9G&zKr9C z&9LIc5>9KKDJ!m7OH{Wv*OeM=pY3P-M_&vg`*34xh>PR}V2Mw3l=Hx2Q@|PZ!DUo8u{kFIl z=b(<4qJ2z2Cs>&_Lg1^_Ao(L6*@oAGy06QzAZ7#&q}lYVaRwMy%%{XZhn60hf#u;! z^wpD}=zl&3OgE}f8Wjt<^ObO>iUWIP9FIDp7g?6+4Euher2Wt0LsWxj>Fq#g`(iI$ zn)5;p?&?pZv$I85YuI0bPqY$!k~)TI1ZrX7;V1AV{X5^mX3S;lZQ+hin!!{Q3%CHS zR9Lf27B5Xa&ds!NWEOv?(!qx_(LHA?oO#+sey4SE+pjs|4{2qvdUlp`EGs7;7POQ0 zIr(_+pb2eGyhMzap2qhk$6(*IvG~BypeFQn558YhhjKZ|Xr6zFtS>TVf5&@s&RD5 z^nV$s`6RE+d1(%F zU(;Tblp+amwrIvtHxjUlzpF3wQN~3(&exb!{So%mz2<%>x}xU^o7DtzyQRq3BYX!ln~5Iv zHw4p{NaL01l)N>rX3p@t*hUDSLOcG(}qbCHfvBBl)cEmW{+Q+!Q_ zc0F{guOT>Q1B!;Mh7h(4`xZK*to;pQBNByGC9Wh*Fa(O*GVto-4A>C6ijJK(2fBXS zfu(H$nfF>6%oZ*pDkTRXqg@04`j`_x>GKf%S_D;%qfm5%HVg$jli@Tq?tQ8n+>84} zKAel?wmgV}T(M`u0|zHk<4vOY^xk$v=E~g;MA%qU#~n0kAj_AH;>w<+b5tV=MJk$j zzvgKyEfb)~+ZOKGzhtb69|5oZr+~PEG4|IvV2rggyZB!vmOS=@F!4sNZBz&=+kFT; z?#k8(w<*}U)$PF7>JM@B_y)50ybK7VZllFS30Qq_5iPkgPv|#d59czl7~h}k<#*Zs zoR-#1BA#sw@o^bMSXRTy|D8;oi)V6QJ5CVQeHLi?d$Hhxeib}_w~Or(@My{vLW169^`i3 zHsPugok1!(1=s$1X!}Pr2EXpoL{T|e&Z##Sa;q}AWzt8u7mVZI&2QY))_ry+=X`|^ zzTL&1;YDz7nGVs&pToUh?GNYYcL|q!DN#@BKf+yU)9ABfyryM7o31O{M&FGO<0KB= zAiG0rVENTq++P!XPiBeMpEQuwUgDkJ>rnn>0N%eh$_9$+Y@K+k9|Y^n&t zd$p;UTHi>f%bmp7m0zH(_zT*&93`F>;~>WEIE3CfEK~}2P_o zLU_htq}hiW*TZW$)u}7+bM((fa7F*Eo6#pY2wt5o zB+F$*!51zQ?}RC6(=wO4Zgiit3A6!tuR(v4D%cq*Q_B@Kr&(a}YacgDAs+=f^SLsO0G`X(NqUS5`7_3cq#ex#f!k{g{AbR0 z#K#GGPRYZT#qF@W^pPOMp#TK_?XZoBFxNt!$v$f_N2OEX+tsn`gvhI;*RdgCtoE_aHfVSpyZm*tvp@{i_>n9a=vqA z@*7E7dH)2w+B*U_mzIN7#4u+#K|md~4Nz|QJ<&92fqyPDg1dJhJH5I5I(U4O8U}KYJ)Fs8N6k8 zq^$|y%{4C5V?bD{c~}@(R0Lw%OJL5rI6<@YA!t4RThI}*A9n1TLKn5@(3UagxUb!m z&wAO=m{SR;@*)L19YqCE-Vf~BzCDGb@R(@rp8^lF^dR?PF0TcBLx;w6bc(Q|U)~<) zuB@C%PsAN32kK?HdkfTXKv76SNjsdem8J#B6Ugrm`yjR^2d-L`L-AHhHdUANtn%+Q z>JSAjz*Cbzt}0Ch<_2bvq~6=aWJwI{9)E-gi+u?j z9SSH?&n?ax%X67NlD)mN=%m1KGN;y*w7)ncn0Ig$)DBjn@q-)iNv4}@s#yctwr2QXlT* zm*?_a*GQBLoC?d;4e2EL=cG?+9(nt37CY(E#AlAKamg-|Y3-lABuXX?>J46VyJLn0 z?spj~oYEy9?kJOg{lUVP?Y&@pEgx_1m`9V&zrooHuh`8QVa*hMMWM;?IdQu=2|63y z;N?udUrB2be4nKS!7Gw*_4xuWQONVE#wQaGnYYCLp*mM1!JzuK29_MVPXZ(oK$7R` zeW>HLm>ddrccide`!~2>&txYpcd;zq*GMEB1leubxTD|_+(-zg;z?6Lak3iCWH(Ug zvW$%q$zTOXhM?unI_y_}2`8?);rG$=SfSP^cp(#vS0YYQyM+f?quL;*D9Mu7uTytU1db*mO{a6P+ zGxC6)QMUil>db0e@4?%}MbMZ$4lXss)Be&JIxxn7jD%}Gw6pjU;uSC`PcasNnyr7n$Wy3BT1o~M~=NCi#r z@W7b;+&isV^nMEOTX>d=-u_is(mKqX9~9G5y%)F!dr!D3S^~~Z1nydv3(aO6z;AB@ zslB`z^tt~5(M{ekxxW`Vw+Jj=7|o2-HnNJyJoFy%kG`lp2g)rX?CY7M#80~(EpJ=W zJr@novuqoAN8fVuRO@M|t~MM3LA7Uf0(|X{Ts-t5C^mW7`C>{cGu&C_~tAo#T>QC!ld;9af}v;T3fQdb92k zu5{8tk;pA<)n+wTCHI&b`euQIP7xkYSOS};reW`^2gFt=3T4*XH1FtqxU)15;fx8s zjr>6@-eu59Jdd~~%ntsjRFKt^oY?qLt(g1u9E?7p&@{JRp@J)TeI z!wo3HI8M)HgJ4K?JeaHKi zP^MpIA7ODWLvZm-4((dmfUfh;(@z#*biK=1+8!YVIue0I(>9GOH9C#~8pBYd?G1R9 z(1i6m+_ghXSVQs&cyT$GDz5pD%1t=HlxEGPDaS_EoC*gJUpLeKtVohPe{ zy~1~q$)k9`xc)qS^1zlhU#{d9Yxj_W zea6(%;U?WEAB!882Q&H5^VqvjmtNIMMFC_mPrkp`WR4EyeoC@oekLfLmkO<08~84X zUf8cWi~0}WWEFaubkWfxkRj%1|9-O+ciC$M>zd&K$GpEYyGdO*{7k^2|Jt&qm>LWV zDP=ZQpV85H2)kw{gR^1=w=FH1%V(om5wG{Hn^VMk_c*Yf8y(oNX#>-hK7}(R@8HsX z_b_tvahxsN%q|(rQ>B2ztaHO7*4b>tW);{$Lf%1Kr*#l~5;LG7cOjb|^8;tU-NQ~^ zwr6FxWLf5;|JW<1Td-L%kI@It_&(?Yrfx50In&m#cTY~>uSYU$sqr2*aC{i2Zfxbg zor%GJ!j+)c_8)83=U`m!Gj6SfOl?N53p?-S$gZ|j`bSKAnaq4giZI`{dKNNG4}Y7@Wp>}qn1q=PLnCpvN!*WW`0r(F4UXWi z-bI*~S&Xy170IgIsx;uJ4r9Eh$amRn5Qu$1-!ucZ^qd+_s(pcLd)DE_SPiP`kw@mt z8O;t(Ob2VnY?hsRih1V6v7I66$cHX5wc<8w%qoXjLq+W9zEiAyNQkEu)-aFUbh>t= zCA(@ql|EmT%BtuKw%Rz7eHhJqway0N_MBJ-momtUhy|GSc`_T0o<-l}C9;&N9}rY6 z&OYBNz$oq)_xpP`IvSj2Y4Zdu#vq@C|NVlW)U|Jk`hdqWHMa7P|49=nz}RLf{P-WU_i?Cy+a;=JZH;R@rLV;8b`Ya=$Vd<6t=4`mm>EoCQP zJjK+i6#m_q&xB5wS$on!Hs^E?>&Q6Fer>=!2GK*ps#k^Fo<1?$)KtL^e7S+` z$*b9lNn^;mx-9m5@)>qRTbhl%62zjW+(L=TjFniHv9Aw!-r>X#%%iA*8DyPi{}L0J zw~rBfFm!?MkZ30+wftV#mIF1Lxpe%QM=&pd*NNxk@%YNTR>BK#fuOlULNbEyO( zT3&J{o!7vu+zFd5ShgfV^?MuYupyswx4R8%MrG4xxuZg} z%Q_HmSOniBZ^B7a2^JXxw4j2)sbW1U5!(k+t+Tn;%Lb!zBmK)h-g)Y^em>h5DHN*MT zZqUj58)^FpMY3S2HkkRGgi%ha)M26#{Z_Y^NuddZ-%AChrlN)6aMb?r1?`UVff7zSZd`?QdtyP^nEXAWct&w zPPggbC|P?AuXDI|$~AiToEv?nbQjH(3!$#h9UARCK<>ONuKXp+y~;ktJ>HxEzIl~` z=HE+bM#?eBPh3csZ`C1|UT;aPx9%^XHeY>d zTzCOzHhmcO`e%|apW|jTQfw^3MwDKI zSxeq=&dm(m3-&_R6$x5+XaXJkoX@>niiXnYmBjGcEpWS`B^b%3VCSeiWN2U&u=L*` zb#zc*<@^SIpeKqe;Lakuj5u5E`n>D^2x3ACEQn8 zZCq#mNAO8qojEG!!0FXMcK=$A&4&nX?;9Xyrvr^Oi~y3yx=;dII>)<*-y* zl$A~|LO;o35YhS%o15mr2Jw&F4Nn~`*(yV)2OmT|lYQ_xX#=c0ssnY$iePl6Jneq8 z6E0~;z=>Oh;CAjk7qDKLt*-);0ll-@Nmf(+v)^{r)Dm-`O9W7oNi;T7%tpUPE=UHLMV;lOFRR@MYoi$YzYA4=;_VgEU`;Fy)Ys6FUH?!^al1Ly1b4slgJntC5Y z-b!Iv*K6!*6k$D|DOk@s0=HZaajVNjK$m}x>w+VpbXOJ0ti6kA8On64pcvXbEpW^hU;beAe?p&Jk_(|y`K-{O^YXeHZF8c@jWiFM;n`KF2UPZ0knO* znNUfe_e!sw#4cJi!`afgotSeC&+-cv#18OxnxB%??a>tUef1wcKO+xS z2Lp)R!&EGin9IU18MCv|6n}=R;E3N}h+h7C!EtqOT(jvoNF3Y*7q}eQHr9k4Y}aA$ zy&9RU`Yd*0>syx4qlrE@?Vw3>B`nU6=Oq8wf!1tAR-ZM28mEWhtx8!~SzZoWx6Ros zCj-`;b_JW;f1&p3Ie2^IBa(DuvanCElv$1{f`x025DViO^y8Qj*pWX1`+e^UxVle# z23UuZ$rP4KU6+`j2rY`3AAcTnOZIe)}F?FG2t-3r0n z2i>H0{7QPFX$NNAFkptArqtYFG1MkTqgdV~7TLItRz`GkKB8+_Xvj_cfRkg~j})9( zF@icYck$YF72LJfw==B#K~B^?0NYkK9@MBp@(SCC9T^6{o1NIxaTkR4woKPSA4Rzz zFy&%ARoU?X_GjpT_(@)WNszMLlKE)n2M+%P`2P zIVndu+NQL=dfg^3RC>CF(YqU9g2)=0@NN?;Xj5hGX1yp|tA=wf=+g}q#~>-88#=dW z;zk#~yFo*oj+(QWh#gj@7i08j#lC#uGliRUfMvl(r#|js`C2+n_5gvG>C{*3Ax&;} zgg1-RG50OUb6l3=0)EdPF~Nb}uC0NAv3+>uR1%Bw5Uc(AVk=7BiQsnwoygs>V&~-( zX}E$r%QfyLpH6qsdkr(;qtAB|bHW?fZ&IgnS)1txWo5=qPo)QND{YE#zv|EAMLqvte;J9-@1?RShg7Ke%}`ho<^$_)i$nc>2I)2@A?tM;xtLeV zJt;ZQ1>9Ui0vA6gU2nW;f-J?C|B~=z|6IH`m_8X)g!$3(6SgAIRz3sSEcQm%!$mOF$rShW=|l+@q5=nAA80(o7#h z>m?B?UwEEG^y{->>(TW20uNj`Pn|B`vK!)VkD$hmzq$T|$7nfmD!4XJhZ*O_LfyH^ zurE&m8+=}pG}$4nESN;;1R-6>_o0nh?!?v|>*0#TwZPI1;o{jo-iwa3H|#kCdnJLX zd?1wnD!`1x>*$t#8G2`e58GrO%sqZFn@zjiNx)r|h9=VF5XH!1d9Cf9a^G?!1 z{R%;6#tKw6I|K$<<@kA=1^40gYBW!HfDeU6tga`V?dQM$Zj03DmJ{MEp3g+8WxqyJ z6bIG^dch;Jz9vT{iW?cBCkWa)fgQ;@0553(+o3Uv*^Er1VyBm}Pp%eJpe4jttX(=Er&`ROA#Kq_n!E@Mtru*pNiO3L$FyKN~yOC8m-sUVah+_zzojr zb|tINabPHEfzO2>QUBdU&VH*I)%vN;JmZ|G+l_PZZLS*GWSL1{Z=1}fiRt5po7aeI z`%@hIrIkEUPsSH3SF-H`Jbz@D79P-wr|11+>B<);goEo|;MNPV>~WYQ?HcJyJBF9T z(sq3?n&1YeGA`_Ws1=*%CPs}$$3y1kDoB3#6g5AOp>@$hV$cV8=$$Fv5F8UkEgMD6 z4a2E|NF026D9NrX7}EU{FOyGFRvjb_ou-AQ6)=8jGr5G@- z38;y8I0P0&3ypiC=$hR9SRF4!*$Dn=`>4PvkT03O~o$Pme+efDMaRgg|HqkB3f*~i7JXpzn|DpdG{acK#pLu?4-zb<0qbE;@@=Y34H zoQoNAo!Nl{n(U!k6c~Jn=2kk`(ek3{jNW^OJNIRP;)+ysIlG>2FbSn!gDR(dutB#ayQq`qN~-hWJOt`Ap`lR-1kPPYpYG>*!5^h?^MEM1IDCT$dR07}CP|K`@qIjFwC(~JBrIe1wp@fe2UgG_olLsIdLNv2sU)7hky!tq zkOp7Vpx4)B(tCGfIp@Gq`cU~ZjeVgN)d3n~S&4Hi8EsurCTOhUyI$Ysuw%Qo^Y3Fbc_wthhnux9a_)IFk~|+e|2t`q(OP&Q;c0%8cedJl+Zcg=!7%dDsk5O~N*yQPJIa%Q|JeJ-7x9~T--)KP> z)M=utjVFy=`;Ux_NTFA5rs0@Ao=dc8FKjyN0$yeFL2ZFJ)gB#-dF~ek(Sj);oooy9 zuQ@^9B@22W-VQX+Mhl;>aHH3jxzeOfnk;mg21f0^0Wyvw*$l&Y@R!a7)yu}zD=L&m zY&rnHWC+?St|pUDzJv&iK{6s`0G2GAO!tUTV$q}(2dT6>l?M&yHN$FPTZRp0% zd|yZ9`D}M_VL0u$8p+FO~JE?ZN z!p+_l%7v$BqVfq*);3lFmux&kJl$kyclJ=#9i5f5brS!+5qwKLx$LRjEtZA(*FEP2SC{gQ_`d zG-v#2tSo;*7QDCyZ37b2;Qk=SD;rUt{e7g&<|*u!cV`nC2^9Y>gDYF6;@3GoT#ZwM z@b8)|GUxe7INU!ED~fLsjmW*&le-G$kNAyCjmt28_j-9b%)Ot0+z4&lkE7ukO_a= z;cM$-HEVw!gQh_tvDh08kNqlmuUQ849J>mO7h6!b4evl{G4IWoco_QsRtq-olO>zq zn}hc=N%}YXhrlta0k6N;hsU2jfqd_Ga!PCkzIRT^YYY1_Ee%AT#|b5Dt? zlgWu+1Ndx5I`;dwb6RR8sAL!h9zLhQOIn*wyAwsc_;cso_cfsY*cvkAcXP94uj1;K z>GU7}|2LJ-WB*13@?MIKq&Pu=9pd*JQTIi$LtzQHST|$P?hcgfs)8ep66{cMui%IK zJ;Cp;y8^`wEp%CPT&Sfoie9DCwg0U8P=4BCIvg?$URg`v-%ZtcE$okQPf{8){dfhU zlInC~_nFmQ*(fDWRLC2H*WK1}?45gR$Qx;NbOPT+%7Q zx|T>%*_jK_kiYvR-`GN*w~Nz9uIuOxyG{rTiGWpmEZDxAQ^>C65_Em&E>6);8uWyd z=<+@_*6!8|2Q!@L{XuD1wBjgU_;C(@wP>>JwhUJD%aqybZ)BjF1|snb$*&_zn4b{wW_?YB*$uOe3%u_<^P?NjmuVEfDhZX z8LOX$TkBhJXw)_Q*;&u5lB8?%E?VNso7uQWej(f7QICz6EtwI&llz`<5_S3f?&@#m z*sec<$=W!w&f*1Z*YHvn7X1nTdt!!B-XW;{yaSC{GvugzV%+)hwW9-0U=5$~UORXO zv#hG1&HNtx8}ks;)u&;1geE(f6M;!d8LaP+40RuU0ln5d1aF;c_GG?7Z8y0J)l1K! z_@4?iw=}`b{KA$x58yeKSV7s$7S8C!by)4Q zh25INdp-}FFqvN)nY!l@fr`yB_Vf&|IbF1Y%**Pv9>>q2)TR-FaVJe-z)6;!=4{xa zeM(GQbThAk++hn!ma?jug{=6?J!}r$%iK&Un-KRAw=`?9EX6T6&1e^T-x}n?j9f|W zX9Kopts4tVP-gNX-B`R+0{X0_c{ZpyGqtv1yY_M@vfv&Z%u;6sPb=A&?td7&`voR{ z8_j<0cE*4Fj_Kl7ea?QmBYl@QnT8b3hl@r3@Ze5sd^2E3Z<#;AvIFPXnxC;Or1d&B z)ooyExoS+y`8?d_eda$tKVxf0@;#J12hZa}7U~_?fo0`pxHM!tJIS*@qaW?(rc~}H z6Zr1AfhFN=e!mopt(e26<^`di&V1(SF_QU>G=cpgrQG!0(fD*sHol#+mfe+=U>+xj zNJ2~|6UauhfHyjjx$q*Y|C!C^=ni9SK{(UucV|jXBPm}r!!kBY)M{V2z(O@gGM#o~ zHg$^!Q@dr$45pvL2@U(1xBEXbs>F_N-B1ha(q~cPz)YI7d=twxJ;yzUE2yzu|`y1Ms4mD@DcFbeOowb})=ma*IwxFlPNIEgf znk_WhMji+Ae4DT*H9Ja&$U$Pows)rpQ+OGTkAA_i6FNU@0K zB=*~E7p6Zi#mu^M_{XeJ_%rPoi)g!s^Sg_<{c*)iWcD!ru4q91>dVfoRH4@28`odpK1t_F{rLoiS}b!F=y9U79K@# z@0vX9sO#Vs?a;)p$46q)5O7bDH!=5XGH}vjCeFU=!d8x)h#QvG2<)E?6Vto8e@ogPk`q38B!p|bpjpIqY(I3naU&S;cj^aBvV>~$e2TIr~ z!oSh&c0WIz=Z>;8wPg4+%`?z z`=_f6zKBV&8ySXagk*l7PXufPS%=&D~l7saKR;fX_SGdi_r+u}`WKkv6OFwrFTE8-@1>Eh@)BWhToO)g zF=7|~8ws&e9%RQwdGdW+A=mUx9wVRm;p{t)@OGFOyZ?L=_d8)KgWpq`n#BDYI#rRG zA1cNB!3VH)cT4r3jbg0uUOGyT6!1CFP^S5EKRTwa1+%H$u=zz3?rBwLSFg@uOP86! z!Ro_!ahwpp&-l&l)|X{1<`3~~YaDlNe>8hA_Xh^M4w0B!C$LW80(QNP0j0_!&T_3Z zn`ho{*DN6U!~Z)16^F^=B0RIU@We-H4?avxr(7Ti}B0CF}Tq( z5gNKBxc@q~fVSFUjOUJ_p`9nRuUA6j zVsB77Hgs)+6IF+}x^Hb zlsWwqnauI9G0K=tqt8oTK-iRL(ER9bC2!{hbBD1 zSNkp_t?tB?^C;W4>l5*6`Od8nX+nkiZmdj^B`Pi&Ok=z`XJn8?=9Q|l3;MR~g@HC$ zZ{J3?!V<)T%3Q#QTj;mT2B#!xaNlf{*z=ipIC6eIdy5d{=V{YQ2`X)Tu z+JN1G3%PaMS2MADi`fF-4#a(LxpFOQe4a1F#QZW`a@&p^-ZKqC>Sr_e#UAK!w~L#` z=Lb^99wB#fzu}S5l_ZN9?qdA1H?)t9q;c1ND=y1W?d z{vE(mrh4qIN*@vZ#CLC$U&LUP=O+1eld($`S%&>OcwKvvk=4VA&Smn99g?R2xVFa`0DR^ICFCb z31o_B5~5DB(sj9%fAJ{umCs?Q+~%zJjKu%)6LFjHvEVy#rJGitr^yovusEv%Wk+o1 zRN8nCh2bHrJX%bvM0uXtuRe6$Qvk#LV?eg23>GPFgS6M8Y-yJvov&+4W&Y~mmgK!~ zrb~fc6%EDvUH^o`JAGNurCb={#9@Fj2S){Yq|sM{WjOHPp3wQkGC{lMApXF$vrcj| z7FBaAFO6h-b0vX#YjAA~CP96^Hr0IQ#oIeBxam3jqqVQPs-Y?v0)V%lPXFB~y&(rmL4R>TWC8 zC0o$X83(cTM;Yg{XHXFF`U&T;_BJk(zYhx@YtfvClh_fhRT#(b=D$Q&gI9hV5i=bP ziqGXt8HC<2f7*|HLT=IuTGkt=t+feuPv$i(%xh}a=fI?J|!X!NVF z8T*tmZX4lR z6MZ-nE%+|YT(t*yGBO(KmmkC7ZD#C3&^B(3RRL)Yb--w*#6E0p#*zUO?oD7mtSV06 zY!>H0g;oaY&G`adi@TV1-gk7}kd85mSy+`@jZ!?rv`rY!nu3>r=%|sXUD5^jqnfex zW(Mb=5(y&5GC@VZh{nFLN6Y1j__U*#zfVM=^0t$lv7RJ6*Z;yLNKLJ&d6LB4*ir(j zBQl}nNCOmI>;%aJk|5+c2UnmHQs0fFB4^5|@5}_K4=Tf!rK53!Y8tAY5Ym8wxirr0 z4Jhzl>>FQ2xMK^JGPjKjiBGL4`y{G|x8gq|ibycsO?#-ZR15C#v%u@t72IY%Uv#H0 zhemcEq4L^+ILRio=G}`$Yy}38Rc3BHOL!-?TaRM}0qSrmgl7wTh_lVVE$NcP39L}g zjO(|tXB)$p(TaCHxGBpYsxIVWc=4J%W z-;=~gw&3T>&%S%cv1PZKApKG|crL6(sT6g%xZD(e2(II)tuoBZSDTFEXP70IrqcJ( z^#Yg1HZJaz1h-S!jcMFgg~;1MEKy5~?z)*lFRUL05%wue$wQC6y7Q2zrDf7cV}!0) z6G(jWi)y)?gDq#hkaLZ}7MDDeYX57$qXx?k5szx(^zZLnPW*y#=WG4#%}QKYb%Y7qjb=54s}sT$6uE^ z$!lRWUCSBS-}AAf+~6rrKJP2}ys?{A)?LHQ<)uu~aSZF^IX9C%M$x;2^3?AUuQ|Et zlX3rUP=)U^;ljk}_CkX=dRLWaCQb69X6t#rt8@h&-2NMyV+SyPdLW#RT}VPMzXpwm z6Y1B3qiNKQQIIfYDs2qqeG`1PeELOcDiqJ9vkY|w_r(n8+3=I_UFi)*#r_m%h8}=; zu~aBg;+ZJHsdVjfE&K3I9+V{?qSw_mn7#aHW@xO(_PrE=m*t6g{JcJxzu8QWysaX~ zmZf2sngabDD8g*-InelBNNbjju+OQ923yq=bco+|`l}sc-@DUT=mj}Q4qa6HQfwyO z=;}|$uRVi&bsFDupuqA^Nw6miN3hO)Ep#`p3C-z^L5H1V$nBH4Sh?pnpXL8bR5*2J zSyCj74ZDZSDo$|)F8is&-FWEelEaGiv#4d>8QfmvL3hMTLQ>v;)F4vS-qgFC%xF9f zPil`*h4+g=tAoHg;2GoVX0g?LH_X+!-$BND4lRl~M&r-)qD%P%A-wu0c=!D~)g5kO z=iXm}3G?!q>YE2xYIF!fqI0QS@)A}u=+Awb91BhDO;mQiHr?@22g@9;6Vc3{u+lAt zt&Y40RX&Tc$s~_ldc|S3?_tb!Bkazb0H(J!fkg@X@m~8eddv0_5l>fP6)n>KO9*yM1U%$RJvr zJSu#zZ%p;({)VP^J8-Lv9i2HMiH%Z^q+Mbs@v`J-QefTy(o>z;4Z}?A*gJx1Pn%*7 zx8&)-JvaJCK9xlhgg+X5-l31sgMVG&vC4NY%KI>#+N?-+{TxSAgHLn&wDz*Wjald; zu?zZT`0rF`EuH962vLD`H1wY)9izCF-o3DlW@y;6pBeddXHl(COEVQUKBmzyy>Dbo zO%#2xARgkc?WUXK94X~90EZ2OaD{q5dt8%=+B0U_O|@FZ(u}0Iv5vrVtvp+R9S)9ZJ}EUq5!4jSR~IdW{M zqXk_Yg7E#YkVOQJV(ZhoVMxV;#W@`zwtbbTeB~)O!^VzT9+hHS9y}I=>~O-d_KTS5 zu@u45nGeV+E`qDSEz1s#n873(Cj5`0GY_Zg@1nR!`XNIiQzRv%MB$$OEe)bXXryS= zNYR8+${bS0B2$SpQBiTv-Z!OCDwUE-N{LDuG)Sp<&-$526*)&QT z_MA2;Wl-NNQ?~Y|6c((~qRJJKc++YH7Fx(r$gXMBX`G5zgZt6a?kwIl&XCTEZ}6_E zAFya<7IyliBfU4K&8s-_jrHKeT7Bv8>!G+zZX8+c?Bf>?al)rvXSo;SrefGUQ$EQ! z1#c9}qtD<06nMH5zudWmH_RX4Le+n~=?FP8^ff0L2AIz}&f~^n5XfCTr&5AMjG-IsT zO?v%h`MG0IO(qPtcF)DcBVjnH@Cm;5b3@rZ12H-O7Qg3t1l~BTz)g48N)N4;)^Xq1&+;T{Brg- zj&1cPmD&H%rXw?G`NtFt%vpfV+b`oOpJv`K(+sc6N0F6ZIY>%pFMa$y0Tc)yX3+ti@$?BygkE&^9S(S`53rlV7^I_bOqLisPUbnHYVSGZXU zx;kG9`3s(JF}j3J7B{f8Z$ z_o5Fpso*_L`#hQMJ-S0vxCc0PSq46zFHHs?I`~oVUU7>;cF_}uB+hey9K9W{gP~`u zaZbx{ihn+bTNic;U;Pfn5r0*v#3KwB^?ktdY7#vXx@x*NxAQ{+EKoC5jO*hMpo6s{ z?W~N)R;yatSD`>z%V*KTLmFsVevMyWa2Bsb#tJ=s{rQgp>bNET9{o$umK^zUgB~47 zL5EFQ*fK5_Gg{A6O6fuTx;BpfOP+vU!9!_{(P0RjCd&+bRq)svVIF^a7qzVh(3em7 z_-b$v&9ZA2dmREY7F{P7ty?5Ikxs{Q7ZMEBK=;A-vEsoP{>nTBs&STK?AZW#F!Pl?xsI0t<`RlRoC*>eqI3{_1(mQmN-#f+Km)dl{tB zK7pp5v?2LnL$N*UFzHV_LqFBC(XA*$T)AinJqkR7ZNBI6?)6g4sLH{@mxOP1H}J`C zlOR(65wsLfz=)hmvFFX(=zGAO`|kTjUYqdf@Q*MyW zvsTVHx7UdOo^dN1I6gu&e?t|2^KKlsLeR4^wzh~q3Oh%2 z?HdEpEKb0W(!38uH#}u?YYjKwfs># zMY?!A0_P-qa%0P-`JIElao@Bnxx^{A`9MMQyPEl3bhcszXZq+it8z)=h7_5>z1o9t zLoWb)ZX03cho^8t$(?)uRSVVx%Y%xuNL+p4)n)mL2{`y|fw;&iT6AN%h z7yic4Jg(oI$*}2(2DnPK@T11WLDBu~aCb-pd|p>0iq4AX($lZ-nNt&CkFyu+S@00T zWY6-mw#z_K%zJLZYC9OBaUL`rr?HV*b3rR}K40}J3>LOWb1F6axuX?g@H`lS_cB#j z{J9V`J+A@n&gUU-_8`biSq7y~^}w>{Czmj!uuShq2I$I-N5|etqL+J5ahubGexn)= z?i8lOrSV=cSzDJ67;eq4S*^>3Z*k?MJ>0oFoH4XE1&c;X_VALzuRQg>2Zhs5IEDV> z*kgs&(0d^i-udgp%55W{XH7rO_lzcNI#XSyUatTN_VZzf(p`Ac8G67*{HImw+}8h2h^uwn;dsZsGTrxwSn|Qu+$-}r@MiRG=98%d*3I)Ubk`zg z^UR8W>}`oxEal-w^B4?2ki%ay(!rI%bGf}wP1vmUV{!iPLVn;On{lSgJ8H-8*d_LOJ$0uxb{E0xB}sG&Qx5=DgU}9-Y;XqTSVt` z8s+i)EzepoXxYR4Ws|t43qEi?uZ_WR{CQY@t(_YdaSE1n3}#cRcf&o!fhhN|6540? zfos@4a9)3{Y@u>ED0Ry+TyvNi9Q^`YXPY3qcAQI^et?(qG9r9t1r5*B1?KTdQBGF} zXR>oG{OqXWhJR5Pg$o?%p3=4O$HNEMRF0LI^n=5G`=RlxET#y zIkU}``1<=K*dL(B9@OPPrcyfZ|1}R^y*dhek4CUWloLN&6HO;SC~$4}uHpI9#7o88 zgCNr^co1brMh9fc=)E02E;)`e-SPBg-X_Wx|H5_NLJsgt9{j4+p}Gzye(YQm`aMO+ zN=M|PX4oR3r}zh2DJil2jvrA);9KR~6c~d2jM?s4_UL_kB_CUQ9+PYZjiz=4z4;JI zMzyD~V)kxv>f8wwzE`++bgGH^TdtsU#-D8_8Xo0#e^w=PyIPw831PgB$4=DWDaV52 zq-nKjgE+xM7XIA5B2GAFixv8Pyw(1Pkn~3^^!EY%&D9nBA*cCkn~!3(n8(SwN%*5M zA5)AY`M}d(FfJeqqkQaP;ybdwa`g@bOj5?cg3)}Hpk*Y?3u95M6_{P)KWzLbh8d0} z81)dyq1OdxtH$7DjY6E&q{Ho}XSnZsm$*Djne%-3fD7_`$9X!Q$C7o2!0(eD^LS^> zHZ2rDVsb zMuz94UH9PX;uQR266|1F zCgc9ZtDx49g}6uc3w}E!O}}hRs7*8#*O)(mjtn)4m4>#&SNQ;4G}UD?a+T~8W`E~!qE|Q=WX;2q@AjjOj|a0?D!}D?cVfRzBl_5U2mdCS(LM`9 zYV7kzof<f_WG%hTI2F`(XG$goLo;5=sueB!B3xM5BExr!V)(%80(j2=k~ z|Cc%Zrj5Nqe@j00wMemzo)YL#FXWrfPaylssr1rt6m43)hWwTWvX75U>4&2uF1$Vx zmiSoOl+QSXsormK@W%hxj8pqaG`mme>6N8$kEht%HyUO>$`Hs`J~02|Al$Off`yn! z;nk*5aN{8-&dm&@wW$M`sKuDAA6r0dWHRZyNaJ-mb*`{_Gq!!t=0~+R@EJ>bVZGdZ z7F2J=+mE@&YaLOhPp!iHc-026_8MDskPU^AH=WU0$S-u5A4i7+!n@x_j-JOaCO8p< zO=IJ+UL#u^-2WM`G~*{%c?Hr47+|LF7KI4SuX2%8r@6!+VbvNnIlVdfmKO zU6&JEK5jbK81+_sWS_EFTsw#ladfZ|&kDptK8B($sVMqYD{xXz&&M4{?n02cFl*aS zAWOFlcx%@QvF-!u-Hgj9eSQbA*wH z4#9f!C#aH5{HIwuAb(&4ss(?C6CFQML0?Z`iN|s8oBnWVmQ&c50D%*}>lm~D?Ti&2 zy2$cp^4&@%xNwFhd3r~3?nm_`3;zpaKjPF`+LlM$vCrN%)^)M`t_WvV@OvXYNjged z0V~-2Z|!iRdBCKzM@w+Alm+eRSj41)DuJsoVYO}9{DDtlqUt>Y7w4*)II(mdewmdZ zmdOoZ_oSj>&Kg_JSUyH1=Em@DyVbx#QIVQl3rXg_9(TO)G-R!@pd6`UI5p4{PH8$& zUcoV3=w{2ZjW3Abt!czDcV6Pz&NJ|}%aH$GEOY@HdBXMuM{swfF{xT6c)Iiy+ImVC*zj@|ttv-XK6}E3>mrEh=BQfcq4G;n$Dr=$nwe%?+!iL#Z~*^uc-1+;N7yB$ez5kuMYt-#)wyTkIR1k6>91>I{)XoXpSs=1_(x&JEo za+x?5qWcDX;{wciIfEI-{;GwQbQVkKP^-)MD z+{%W~%?&8Jn@zuyThQq25csSl_#c<_vvH*+?4Hp5b0+y4-nV`U0iOrZa@{~S zW$sNUAo6WiE zEEKeuLfXCMG|Fkj65sj<`xFP@%+&&~)j0tc_Bqmxhm}~>kdLYRwJ^0;303_SklVe2 zto970pVmW2MdqpCdlGuiI_Hy#I#Fh@j%4eK&tNst9i@VWJ}2WRFu%DH_45@-YlAge zn17<$XGXM4JC7Uuz!QGMNq*y?H1r+%6+=fjaNC{s<5KfXx__Vp*A4T5JL+vhzo#01 zBy|m~nI-UD<2GP&;6vu0q%XPu&<>+QgXnzB33_4Ij{4pGpvPB9vSYG7DU6u|9iN7g z)y|)MN52`|KRJD>TAM-I;;GPRwT>2hjDfanpqK?d)G}udeNeT9g>zp(T*F=ZIPeB# z%J)*ph_@tO`yIt9Z}H@~zhz3cAMx#rM_{u|mdch!Lce>hxFS|UwvbZ6s5GL;U(UCx65 zymxbz;zV9A(}n7!QfSWBJaTF?VqZ&ML1J7t2EDW)Rm*JrEFH_4w!|~rYgMABC$#B9 z{AFsJ3 zLp$KvugfqnZ=sMwT0^p5Cy@JgAt!#C@OUam>f0~FN%MY^q~tzQ+)x1zW1^XS{6e}I z?MQAXJ29$42K&Cvpf{VO*h$$dbmx~Z8(c67-dd#Lpf9b|vfQ0o?mb4ap$DB?VMpeQ zLZ?ol!0MUjOzQ)-P{We`lCYJUa3Moca((s@lz%BO`xSTLwym-tt7nUy%Y@st(P|p{ zaSYW>9001rSJAM6ev-U}S!9363XMJ_^SuSFSUEsKALduWpwXcen-PUl&BMT_=pq(n zJSRo@Xj1iS!0w5;lrl$(c4owIIs;9(rltw#=dA=WAsOJXR$DTs{u^oD0LiuPfAr(# zA6%n47gy@#lA8KKD*9qdUnZsTuWdJyMMZGgsx5h_5IvgSOwPa;yC;*(_(Q_!S6PzL z*FYxg(#dN}4%F+U(I*O|sDWcKbDcSTYmUaHMyE+yb}RZTSMb?wooTVh_EQ@rA@Tzyy|dCOZQE~}Ua^a2b<0XNg_J?fS#2tPnnIuMxpU!7liBa@ z^O#|d3bnnO%-#upt*>uhk@exhGzm}RioR;PF~6D$&vc-L<63sAF^b8K4P$bngni53 zl(#(60b@k!`&FvP39<+s^mO8lsEu zXq@S74=vwZC1txp>0hXZmA$7%`G{JGs>oo~dei5p1k?kc>$ z@)6&5KEm1@qe zr>A70`UHuY`(}R1njwjF;sHq1@uZlB~jcG)uIEUS1m@ z;Ur1)Qn)qLEp4HbpS9^k;$TVr7NRCwHGFc?4xj4AQvS!C?Ea!Y8n^uwt@=7u^2q)S z8I6=s@lGR&nU@>eHqMSl#Ke(#P&XzY?ZKbR6X5zJJ$87L2AlgW2|}k{qsD-44Bjx2 zW`z{u$;)RTvcv*iLyptukX^X1{0+trl_QTW=Gby^I4uet&3ZCNaYN&R#A>ga@Rw7W z$lWU!{FV!hlf{Y5WA}I}-c?KK24PhE)_`}NX2iCKG>JBy3Bw^@L=x{(DT#OaFo|rP zF`1cbNz8O>>169liTV1!)FJbP3d;I;h4mU(C$Pu=1}>Kb45*-Q!;P4>>UvsJH=mlk z03>s^;gzT(l=NdJ70)jr`SVr+Q2R9gtO|#)xfl7M7t*BE;tQ9Y7fF7`3jSJpjdKK^ z{EUYc6g!@iY;F2Wer`?@xrd`9k)}c}UwbB5pSn*+#vZ2`f)_V#%L}T@vy#Ny4wlTj z{*0I0J%uS*`naPwi3;?5n9ia&O7>YGxtf$AxxMR@r10e}G7K^%{&WQWr)NPD-SZa{aC;$(>pz(cPIZ-j=b0!8#>+ zyFi}Zxwx7+-LoRskJl-!r;(gZGeCOlE{Nq~VOMz}-7q^OGM#8faRIkP^VX-)5?mme zkampCU1iUlEjWo%!YT?{7fgenI0$$9htPfE9X!`dr=dk&Hrronaz6xMd0VY3ICUqH zVMi3o?~!AtPhRG4@3P{jc_dKHonD;(>?T+7Egr2FMdR|<6Tq|fJW6dhr^VY#`MVcH zP_jFRCaoMpwOKv<)PvFJus#d3uu)VR-Jg_Pbg0_RpE`e4V8hQtyl;*+tp5EH@;ijt z^)b&UFa66Eudoz(&Hsf4-}8Cz0mKg}JWL@q18BjNg{YQUfr{%TxL~d~DjwKLJ-Mku zzFdmLzb4SSR86{b!5=Lh2TNQtztDdTYzVdVmE()dL|H?$0 zAKZj}K8kc{t^>?AJb;aA{}6Yz;#fgb|8(Pt=-^&o{LjvXR{MR1XyI0~v9O2>i;AHS zIas#bMwOj9dkx)gv~wwk&9P>17p^pa3VZZ!W1USpUaZxk@edB;N99@cFYYO+_&AWN ziwnip7h#xA3(mbdhNaC}N&7-pQ_4q6?#GB&T>CkP;(ZhGctA0E+4{qQ+{Lu4E|HwF zb=U+eV;T{E7f+rUNg?ab!JwU;Q17nIPB%|vzNX_PH@#j{*~@;C7h*Luxo-m51H35D zel2-4#FLDS8#M85Wc7VHJrZulzJi~(=Gr}~W#y=mzL9Qj8DKNiv`+lecNB@`5^#T> zECRhHPu$t2cB}sp(5YAkX|X=8rP-q9z_NuKei8f*;4NF znEup%UmgVhlA%!RbXqm+7<^b00n=7*pjy>GSd?-e{n~3vneZH|*YqX*Nmp%_Ew>|M zRU?YjbRxeQt>|*|75+V-ga^QsKC7FNRrp?vOh`nT+AhBK)e*dPT*#VOF`D$PkTyP) zXKRF6Y~i*%&CDwXF(U}m`>hPY|*Il2x;%rVn$i_#oZ;Xyp!8< z_QP-@bC2{FIAAwPK1Ge}ZHJKJm2)_>UmzTrZb+-@CG;n<97lXzMT_kliHQS*EDDDM z+pocr$3hpRjs~vY{Ro3!Xy8n>3QTN2PaXbu_`~o6{m;Au{j(-?>q$J<@~RsbnT{v_ zWhLy?(@p%o+oiPqkv!dAQ7H(3fg}ysTGwwk)$-dK~ zXZgvPIY|z^v=njSusAv=ttwg9t1NMq@_^FomvLG|3Yi>WT?5KqAJSYb%YJbwjxC}ZrYYH3w zW--X^vSrTs@o@C}dGgi#gF7B5OE$+eiT&0#LVR=K$2zGE`nLlvd41!}C{u;KZ9dsY-D>^*6R)1B4QeeO(u@i9b(=dyheD zPc+?oJDXPh6Z)>zC(@z`Y49*X=&pJb4Qi4B68+39miR&}y8AgCW;TXV`>lF9v}PWy zlABMjzZuiwohz}(PGHD%o~0X0o!BDX^ZXpAa{g$TtfcUH5nd9y_7->glc`%ReI*&m zv9%G*BO;$lg?;zfOMOx~^&5&l-QWbCEZ-$Cfb*7L#gc+nzOF*>+F~J&Z#^aKp}8~% z8tD3$2e7eJhOL)tq%T+PvG{>3E>&8JkEWld5w)^(GIcM>8iiwxv>n}zAS-cA}`a}-78@4n{i|)rOy_> z@S;JYX8bcK3?H@Fa-+=ea|gTjkn+F+Y`C9GEyXu+wA(+S(_Mp{*R7?ha8vTXu>q8l zcGGMbIhsAs2ToV_r%hWv(DC2PC}4IoITS|-c;`-ZyWJu5haM2TXqk|F{}LRMQe{TH znWP%hk2YitmvrYf3L4THx@4-1JKQePM-Ab7Z>7<(!-ZVxx*9&b)r0O1;c)PV@nk6Q z`hJWLpk!gj9&q*=oiENojqwfC)0#>{Rjtu@QVeC4`Ln5lxp|weyrijr3eGBNXU zz>jN6G1Gkm*}6|9la&+jNKmCKLk0V7#=P9v#O)IuEBa+ov z3VQmmjnLRC#~#Y_wjh@e|j~o9~wd%`pJuxMp&WTNe^7+ zJ0CvDEP?1IGs@06f##WiVDQjf*0U-aOFEZ8qKPsa+~&qgtb{$gpC=#Ylqi0hxsA$( z+2UW7MfBmyIJR$*E^oE+7&|s;I_@g)ha077ux7&_dg`Y24Po14)uC^mCXgL7c1}-Q4jMPkRq!jV7CEcV8~#pZx{T zg>Gb#Vq}@Du>bRldw8SjD!k--1QMTr$2SuvvS&5zB9&joG^D~ocpvY@$uqQto5Vdl zUOj>K{}DVh2j$q%lksq5)g!!a=m%XoCD3o=LiXERl~hk}Vd@`y!7Iy_wS@`Y!G-(5 zkbeb(EaaJUc^GB)baVEO$ylYEfDeq?OkBo)n z^p$u!bPuzVIg4(uP6-W`E$o$Q8%j@=#-zh**;M%;7G4`I_(mPT>_8aB8!e$rV1zLp zYA}7wN+>uG!@_!E;MLsKHs0$Fuzi{uw50nez8v}xMyd5IqhSsB9Kx@l0H zkj#v{x3gU7ZPaj4nkK&N$Hr+{!?a5a@zI3Q%(_pXP8FD8&aj8jV6VuJ`A>yklAT4P z4Anubdm9>@J(#kFIWg2?87oFm^v~J!*{6#iG@u^ddulQiCBo3#(){9WyZEvXq3|Wx z2rry03M`*Zdp|aTQMv-B|9K_u7)jXJ{1|PfKL-a7H}+w66Vi@(bnMUv zd`h#KtvCVJrOo1c+>@EsE@_rxdJAVH>oYeYQ$4+;fPJ1$q6=$#p}2B7Mu6Z+6gXnL z-w$F(cepeAUkR|v4$04(@#l)uaMk(^=dp7ym^~dxW~+zM!Ote3p|yY( zS8au97fxeerW?8JjT5w*Ui5qafG@vUg(GaV1^?oFdK`OAG+)J$w9UWb%^?DF;-(9; z(J5k6m3Cst#z{0vW+eM?Oq(4`E{9~PKR7w+DENn!^De$`U{0z5Zf)}+)ud6R9{e9X zbJ<60eFcp#HUuLzzkzdVAdJuGfOl(C;GX?hHpDgze_A%8@9Ga+`#op+6qHV>JvyXi zdcnr{(0m%9xQ6ad+lWIJT!#Ae!EEUPJ?1q0HLlv%iEEqxpy2=`sJ(PdH0tVQ8lry_ z7P$sf3^dXF5Bu=ohy~ooO*(Aa>I{DTv~E~>YXbupU3R5ZfmO=b;qITFe4Bd@_ibey z^}pXouk+8-9ZNOw*G@N<`EwfmT@Q32VlCIBwvfC2HW$LC&tX3%jesd%_T#Z-LVo7_ zBY3efh`BEwBGH&sN|rw&+3Sz4Bz9U!tMi94k27EJeNjAVa+>ViFjb0c*2jR>0M7n< z5MJ1mM~78)nA_e0s90^!7DUIPq0}MvaKTQnSF3~SmSA{3<~2Euc>qNg`DE^;O>#Jd z#6zZnT&5gMS<#E5976DKr8T)74TVsroAA^!oeNb?$GO)|V8ebTvM5jG-ImUv=c#`g z_XAj7+YhwN8%e=(!$|$>EXs89rhx|9w8O?5%1S4a^0~##C1^LyRX3B!ue-?HOOjdO z@n`I{UlU{>y2DaKHED(aZN7HH8ymS*{psR|VCs3pu>92tv{&hZ!$$k4VD?cel1~sX z%l4#kj@`I_$amJJC1UOY+u52y`m{`IC%Wt%MdwG$(cY}5^xsEsA=ePc?)Buef<-S$ z{Hm1sc)nqEJx(lkpcGwJJWR4{ZejY=pSE7A*P|iJm^X4~4-a*x3iNtTgYs`0}H*w0~U+D?4w==5LE; zUN1D*%=8HU-OL)$3pK)8+XM*o(HDH+I?Q&HBfCB2E-p$Q&v+Ze^}nR4P*6ul3)9u>~U*sN%zO>Pv{?<>}qm2-34 zUVx(QKJHBE7}~Y6KQ%-jge_;Qz|8qLyqI7I<1WOoCZW&rMp6RF%S2#8)f+xxy)!1w zZx;nnqTbab4(Pdd7a*Kc?oL!uwjLlSMwC1#Djd*jd-%iv^`1aW4>z zn=)+7*<6lWB5-CV?HnFdc>H5~Q3+E%*;>b#v z_%crDG8@TODOi(N_-|S@r3qFF+-DyL1y-N^iM#UpF&8eU2nGG!>Gi8k;QQ+zb*Hy7 zhYO=8Js1I#sw?{AT;7m;Iu7B;JL4(kzzb|1Us`&^Y#dwNUBgOdOcH%{s|K%U#!$?< zct`)G45c#Q`i?qqe=-BFZZntsjQda0E%>Zw-M+}T4en1X#zf+9m7U;mKM3-+SAjvb z2|VZG%Z9s#LDdp}Orml&E=)nP`bQ%Bs~?Ufnl1F7+iT9T=MJ3^on;T2QaJTpDa?Lb zJpP@&A0P1(P%*WI*Xo;$35~17)#o1bTjHddPFD$}2yC-DWpf;;tHQ3hui{KyG6c(~ zKge856^HFw!QHp>;a*S6=Pdp7Sb?Y!zZZ`K~x*c2k zIalBVFX61Rav|%rIj5WCi*eHu#HBU6S@pIu$m%QL>M~N_+zjotM<^+^?6h_I!ao*V`eEZ2@_HySWSbjr=$vze`(#OjSJdbAl|JCmYzJ$ zM-Gs{NT*X=y1EYgK1i8{q)cVYOcb#7NC4by3<1TZBjEd?FPOaTI9HI-33jQLY(VQZ zFe{FPyVir)<;hWCC|?0jw5r&uffu;H2Zf^f;K}e#?ExC|Pau3$68s990p8M$pjmsI zpIq@7UTCVbz(2}#a{M7KQ^}qW8|}@lGp!N*r1dtgoz48i-g@Z&Lz(&fnhQ(x1fAn{ zDf{*RUa)toS!K~lHXuO`6SuU(>NgVDz4`)|FeidDTwlUp-Z>dmH_79eHZLZvC_gD+ zbUJJmMS**UIb+d-Av*9TIBb~6e)bQ7ng4Qtd(Z(9_g=#2MatlLZYx+u>w|jLZ7?(J z%3%~x-++KPY;U;PvE>3*0RVG`=R6bVP1NL8n^v$F()M{VG5IHv$MZ**^T=X zS@p4z>>@kL6+iZ2J%!dRaA;iFlE(|6{@zjC+?~hggw&N)oi&2hZ?pJezRBE^H=~&_ zS_MK z=sNKS{XRG_^CU=zN~6J$dOm4`Q z0q}6VEUVG5C;splzHwE5l+$a#gA?UBv(%5`DN&X(j5-2#cg}!ua}-!E z&4(ekE`ZH3O{P;+!{3cOQflmG&v_n{rVH;{=z_qojyRpp_O2X>&CyqxX+H2H5;+L! zljAqn$dKB~P(Gl|kBPlb!`EUPnisFeNA#VCrM6;xm3awNCiG|bE@-gv!fqK|mJhDO zG$xJk9!;As$*|&ee_)EGJ9td4SQu*57Z57;r2Sd;+Bs4i-AQmakQxxm%d^GN(+4g%cqpVlI0#0xHAe5UR8l%{vy^L zTF!s2=tOoV7~a_Iqy2L)!{k{*@ZnNf_Pb;v9r6le#ZPW?*N2w#Aupu3bMo;beH)JD z&l(ReQ&P}*c%b;k!Mm70a}24EF~aNz=LG)E2O;axj5ACpird6hykX~kd|#q0POOaJ zxw|A*D2>Oae=?}p)drT|1fKk+xftp`2|r1W^AnOyxX`D|ar5X}aEz(v>rPList_Lt z3eQ4`<|AI!LV?uomBLCs0jvWR*cdG}=3z35DP72C)8a8=e#WbpP4>IKh%v8`|5>wwAh$!;U72p6Hr>oain zrNHY*|I7PMxh(jij&e0e3NXmSj9j&s^Hy8Vp!pXgXzicNzrOMxJN0@48o^w4FYO|? zclCL`@9+(jDK`Y?AD*-154e4E4xkzoeA~OUQvy@^Mv=2t~*4=$V1dmL)PY)0_XJpqN3Xc&?@TW zlQzn-htbk(SYRgPeCdRpA@=N-Mjed(yo(Hu4r3!Y4hyEg!w*B%(L7)o9=s?E@9KVI z!o98h30GH&30VPQyDKOio@qz81aMAQ(+$BFRQq8-L zJvN?LcuSGZu33p5>(Akh$&+dAFa>&EJRT;8&qWQFB5sqjBd7(b(Ebe~=3gX5L04V5 zDaB2^^`$LzR+;Cfxjp2xQ-K<9GWNs23-d&FR5*DoU0jsIp+W_OYN~5T!smu>InfVIO@u6w%~|g25oLC%Dv7?1 zQKHoPkNBVVKK7)r5AS{qheekJf0%|YsO^zv@Agk)tIVGAZhBU9VDxNuyGexF@fA4V zQVnBLK;Qs%^0rn_K=a)^YIgAx*}d+^64dvTPQR5bH}(oHUhxLfU)%ClN2KW8nL=<0 z*QL{${jkFtXl70vo_Zxm6Fif}F~5gX_4OazuChohJrT}%{qu!xm!Y&h`6a@)K;|3o zL&G{a7_wD?rLA=!l}qz5An6LX)p9boH8lV$4eBsyM=^e?Eae8BNP}q8ofLoR5N?(- zVD1hN$YF*iu42V7(P=$*s->OJk5Q%Qp!~A!p=#oS&F*wRZ3(q!8?XRYia%R|Y5M09 zu4aD%t0eb z!+o30cv7u~yEVR4LXn6Yx%)dI44O4E!#2-_^Q`onzP0^sw zTKmzrDjQbYSdm*#w5Y+Z0B=Peq^87VdU!e(=WJR=r!78W(3niN@c1;?^>0ntO?w;s zJnAV<{&-V-vD=Qe3{GVkAs^^i{tbB1*2|Y~Yo{gkb$BXGL-Of?15SP!ffqKNqJL{5 zp*AO!`tP~KYy#`}KO@7~sw71kH#1qt97p1h9nEm&q7jCEHeknHo#Afp0V?bnPS%5R z*cG2x+NLp&x(qcadgB)eX-J~6Q|Hr@;H7x;#5vK3ta#oaxPY_{RdRRlPGgZD_6T>+ zaik~gX*o6yBGiXqhG{s1Bu!$wFVj%HN>*u$?~iXyI%h6>ig( z7JT9Q$40F|8Mlv$MaM@=$%?*%W?@?DD*>bKLx)=w1Iaf!y< zFR^i4U{Cpv8$owU3r;sTW8s2__>1o}x+gbKVlw(GX+BP*;vccJZ^vn27Cy{g+I_=* zB1^Pha|VC()ninp8(S^hp)WpNOotAcVe4;s5}Q=Q_=zK_v}hO&|2~=3IZY>@6@}RT zY7iT(sYOQryhwNTXT*Qkz~|IP{Jt=Zc60~9wb)cV)>w^bwtz+^reNkWj?(0c$igp* zk_CsyxT|C7cGOehZ~7i}N`~Nqmu>==V=_hCT2L38MandmB3ym>*#jP-lCvfa9Mi<9 zee8rozKuLBOs4Fw`}vwNYpF?UKaFITv}a%+J-Yr4TNVi(?Fx?YeMaz%dd z#dtK;lw6~K1EoOrZvne^@B)3FQpqp*c!1p~vZI9?)X2}nRdja03&ec)qrP2+>HmdpfOl z9Me8zlKUNjZyF-x)tjr)+Up>6#P^e|8`g}s3zrH0qY?BlXP>o|`)mAqJWkAYJceYa z0Jt~os`y@q0*uq%#3HncDCp%B8Xmrbat56yr7D{l-tb2!iy(qYf%@4s0?TR;hXV&!gdqi{Ee=;<*2-jl?e!Uu_j5RV@o6p_*S(q_P&=3c-vy$7MljBv zUM9ADcLxoY4xtS$|50yx9()mI#r;A5&`-7p4mAJ=jmpBFP=LO@OL70KG4$z4Egs%G z9|FSOiN{(Ugqb;aagwQ$c#_a3r&$}zitME*{+2a-m&jrGCku3rFU6C+V(4GFgzijF z!!6IM_~miUHaUF~&i#cNrcB6&Ef?I;KU$Gp*etMwCd*Qa=sb$fck)l_zu~>HGG#sA za(RomL9pen+z+Sp z+~j8c1=!^v#b&lPiu1b`;@eG0s2q3}KN~Ef$}np_%a`MJgGF%5LzlJ6=_I=q+x_)L`E_aN+G3$=YG#ar6@{PLwi|iXz^?EyZ^!Cy03fh z_ngn?{niA0-NtB?0vn}tm^tbm;Rn`4O8%^P2Y>Zy`T6Dps9Wf>t5htdRrT5oNA>V> zPi{(fYK_L$(p~Vw=_EJA>5VA7tBbwcbebKoF(ga34QS<82fJq9#mc5rOzm_VzdhN8 zdQR%&>HN*uVJt^klf0;LV^fJ{T`C&4_`;e0yujvyJhBOGtO=&Fr=z_Y*~G%7#bYGF zo2&t7Rh9$(DVl--1&y{+z$nLR`Gxfd>1Q!gkeC zFy_~>%2(Q;bIu7<9*v^dpibB!_mbW6J;>kTouJrk6#P=p=8u}*U?vsg*c}&l=Ktj% zdl@QE7C})ZCy!p@-#6bT&5#6k`;Z9*)(jQc!6zso>piP@;?ACQzkx5yhQq6ssSxKA z2XWi4vX?n$xVGMX;1ZhxE)!Fk`f|ZDQgR=z%v#AbzF6VyL+e03+??KA?T-%&qseAP z6}NTBb_|Ot;ruvaAt^6lZtGQ%`_QwNBE4i$1%&V!zZD3C`OMXNd#Y>LOS6(J*^#DS zT%xHMYjvjc_t&UGzs z?fY;CHw+YZgh|t2>g`|rOz{BPu8_oBmiV&avr1T7!7Hw3-f8%n`w7YvY?x_;55N0W z2PF6o;(mUdiFJIkX!U=3tjSmnisj>3|6vD3`8{ErQ?D~#=?+Bae!o~i%R2DdI+X3H z3X;@2?xuc)7WCiYY3$7~M;w{g0s7t|Nb4B@9ohR?$jmZWUTP2OT0F!mo3h?J)~NR7 zBFi29hSr}s%Z_(zbMbyTT(7`Wy={H7M0I5sr)D@0_ODgNy-OKea@v|JxgpOyUM{7? z3B$Oy+9W346$Vr0$I!VG!R(yCtDP^sEE#k69>l3euyOosuz9fA7e>O>ru z+SbT#PYPoDT!P{7pJ*XNRS$b2_iz)I-eq(|pJexaV1J?%NaWg|-m0B~vjdBVkO|GLX9z*bYhGo{4luE8u@p z4e`(grr5YL0e6%>fMyFze7CuQBR+`zcygQ#F{|SD4mDwi59F|!=eO`4`6gU?%`fi3 zB}IyIJ;f8c@mqhYLFVu*IQ**)l-AYYQ?tWNU#h{ufF~?FGKT*c`%Yj85Kvqid|T1N zzZ-aze`aCAq+>KRzz)oNss<+f|*VLDBA}cxeBXcfuSZ~a7&kZNL!DHa*rEz5cFO%~+)4;2= z3tp=2S2#C~G_Zfr4(56QWp-hiucio|XHMuPjB)F_&1k$wyXL&@ zootBEkKW{<33u#Ya<_ACFuCxPmOaSE%{XL)JC2k z6?0bDf&66-&5QA{Pl@E`tyNUynkvlxg5VRz!n{Qv_;W1*tk1Q{bbX~2H567shh8lF z;*tOO_`T@$S8dB}8GX=M*b5)Od4OBIXd`pXQUyzsG;ZEdH)a-;&wue-K=R&+@VfpM zclEm*EBH8-m&?$VoS5y4k%|ViEHMy2ano4Ky?(T&OyE5~oFP`(Jr4U{e+023L~O2h zJX>jO3ge6$Ie(#Ne(ujCPDkJy`Oh%nHE$e(PRso)QrVBzoLNO-#s|>pb1R$+euU0# zuUX$XRc_Y4O^|2W0TY)@VmJRT!BJn@_=c}Nl1JOZx!SHftU>7Cn5LUj(c|^}4GqDk zJw=UOZ`s9WsLI0h_QmW?nkLjV>q5wb$q*dh$vbzHvirR)%uzW14CyjJ&4pgvvO_cZ zWPjnC(aW2OUMEP#<~ft$G$XFh=S^@)q=nUs`hw;yD_Z2f77dMhSn51=TAVYTm49}n z5u?gjy})KLxgtv?ky(`MT*H1CgplLL&oDO}+{!H>J_2loHLq&dB ztaTR8Or0%-o@F8hJRU(Fq1Slj+9npeI*|?S717yOTWS4`9qf3S8ohQ?AQ!o3G^}*E zbiL<0+V*}pGniB;u$L7mXR0eL@UNnOZA70B>M-jqRoqR7tt_E_IL%HNM_a>G$$yE8 z^vMODsuG0Rjyq3ea*1Ac6_Mcp9||W+8lg0WDm&LmZhw5w{yqK3oRfN4lZPg)$aA3E zABE2BGbhM4xX7*4my_NZsVub%k(bIX{y^BagA&(_AvFU>fiJBoP4qG66(*?AvgosEH1BDbE7Hjk3(_4Y8eu$OdRbrii$2;!^y^Q=udh%U@IP9IiO zlFr{idOG?f&A84B8AE-h;5tso>S{@<3W}M`dtzj z(t|YA;O2J26h<`}b zciv9Q7%Fh?-p(cI$4|`Bvze8BoJ%ol{Fz*@22J*_XDbICXLqJsV4uJ1$!cN?Ybeu` zDjRf=N%t)};WSK`5lx`zZ{Jx~(Or@q!_nwt^JvZOUvzMdx^(We*L2UooM!PCsrh?< zDPQ^@Ggx3Ga6jr;nCns+{4{{lD%GS{RDEgxH7VT^=P~oLM%ubAnKo23QlGc6?6+QD z3LNs8&Ab}QGR^FTe1jZ4bkUS9|Ip24+Wc2?d#Avk$OxsLF%!w@TNq_I4HPo6*T`U; zDV;qpcn}n~)0^bk((n0M6c%!UEQh_IDe#VV%^pcM17EX)glD0XXBJFGUsZ*t&bn@7ngiYQ2hVhvcky9?bi5Tt9SWBG;hOvhS zjHqbgENV_TLqX~1nD-bB`Xm!4vfk82c5yyrrr<}ip_%M)SOKf}CUlcO++mIj<*2FA zpZWh5Q~!0o^m52y>NaU7nG4F&>31cN_0^6oao51t)xxaw^%nZ)lT4>HezOgg@g(#1 z02O@cpr}dfseEV{ombMP16Q(W|Gy?Gyz9x{t;?p;r3qx?U%?Di>$qzRMi8HBE!FoO zBK3PSRH|6qha7&zz#^`SO}I6SqUZOKJ}nZKMV=37xZDadU({dP-EIXb$8ssba}<`s zSbP-fNNK{(;N(dcwsT0aFbfUl~s57LIA0WH&CfviTnGY22kEu)nM*9rccrJ}=OxV@iw2-6xw34j3eLPH3d; z&S+{%$mi)^Dckcao%%mfqor3WXx{W>$%VWm60Lj%@8++B>DJ|(ugpF+&|nCyj~`Bp z*X7Xf(-ve`u!m+>N-61|;9gm@lWwg{pI-H&uUxob^L7InamN z?O%~e`4if(xq@aKd`_i_UgV@T5hst;;1q;eOTYFYURB$Yy|wF*eEVS{weP1W9kaqv z`uo97SUGKk^pyKwR<-ysJk2~pR)InM7E>K*QuIvecr|@##mv3XYLh_cKNM5>ssy_D zu!R*3dO&6M8Pt=NNkhEcXu9en&S7i@R{Hm4Gmhy?t?#unKf`@AVfYeq`K~Cv^RBCm4ebR}GQO=A6Z`etP3viX+K$UraS&q*`Rx9je4&FLXmBP&H zh52uKn6E^Bn}_h-nQKUYRUrG5D)>$R>tNFI+uJ*N8XHhER(NdulfW_Lc0kc!Sq(CqLm6lE<$mG^SeL&CG{P1Ir^vUnEcyJN93cItbGhefeuT^Z4b_*}+N@3njh1@~^EmXPD0F?$N(tfl1 zqADr^h1S)NqDX5 zFq++Q!U>_*VW8Up?oDAO+j(&w%H-6+ZLPK3_w&PX`s~qo?Y%DRC>@WvZO7odcRc@W ziNM%SNQT`y$N3dzvN-SAL_GLjk8e;^r_-CRf|JKr_!wkOAAY95{li5p%jPLtIVzTa zY&jElOc{$WLKU%NlaNo$&K3=H_LA&hHy8u5&0+gWdHi-`FurKM2HBS_1a9v){`h7` zXc|3-y`GuF#I@b{*iDl=aqtjlVtt$~xMRu^w59xhRatKKz`=ahv86aBGMXJYv=noK z3|Q3sYJP8}nP|^l1)M#1A@3W)5QQ<)fF&i zLnMyso6oY3*0TLB(eV3}FMQE-=Z(sOnUl>4Hsy04jA*Q2ra7`~TVe(K5E%vIgf7nP zp*51r&PzaUuRNAJw9+)$lvv6|bFe2Q(+lzz>C6*?0@#ym9+1uC42ffx2aIwBJc?*AiRU%0K6xam0DO zzYLaFl_*7RK7MKa%NtCRf}*)T+C^@FSyoo8_~am*X{jP<|4_*u&P(OBr|Mwo!%vd4 zQ~KhIKn`PvMwDF@b?x;I9EQG?CKo{)qcZahIAw*_c}mT z-|yTDA*;N%?K$_TER^4E9fdu=?%^zc1o|uMu-$c;=;HGq*va^hw=M$HBYNw@+oiz*{Fac(L(7>8+XYp-C2|Pa8how!kJXprm2p zcu}7nf$aGE1b)|3IdO!h4vq{8gjwA}w{&PIcT3<#CR|Y|xe~J+zgHqQEqDQsx^nQ{ zR(sqYqlr^w7-#d`3hX1cVu3~ww)OGG#KmK9?8Q>>P_jVNnRD@1)(x&F)w=@Et6u3zGD%O&79A7oZ$Q;HBCVT<4B2+(yAEE;^)wx!y9k{e=PE z>^_XgS~O^HY&tyt5y57JF5<%f+~kICFc)%@N*H!#BH)22)XCT&dF49;;u_8PhNHvq zX+;ybFIC4)!N;M<_8=^i`wQa=wNWQr5!v)AEQvde*X8r@(9pijRUwk&1I5hz;dT6X zvJPKv)Wyfqm0)!;3*z+`LDyIbbWa$G$FBYra*6BEE8_rys4pt$PQ)2<19|4BgyCt! z(ZRP4?k0}r&Lk*GF4ld50v#2I9i)S|Cb`pyv_sG;FXg#uX&_!X4m_QPAe#(0K~EM8 zO0S~FmVE3StpXLxm*b9W=Wz7RO8oO@Gghtqz@1q#pS((RQ1gZ=-YL<*$L7L)9KRbw zADzPe#T9r6W?-Vha;$3#;A<1Va!(|U>{C%Dm~Okxa^B1Uuase2eyRbVYdR6OoLvQd zRW8ET?n?Of@&?!EpN_c9pU1ObCJNf4I2hkI3B7kEOKyiRM>T_oxb@0e{@1Kz+|gSG zt6g3|xoj|(=}`ua9dBUr;3{xiu?wQ|i-h+{Jj!Ps#ewcGxOrM9IHTw~IJ@8~_^B6g zC9lfa>@*{E5WnIqvbS-&qC4S!6wj44MT@4?{f2cua(F&$07^Xcv2w(Hu=IMy*ByC| z&Lh5I#aB<<>g}S_&w`)CSU4ZlsZ7B&?)O2~Mhr(9Zi;4pAB(H@ zzGM5hA4fl}hY;-Uh^^s1Am3dDE=h_ww@P08e?dv%FSLK^ zfc>{MB&q!e;eVqWL1$zhtXy3JyD#m8Ujs_HC;T$lw)rw_iq%Gig?TXjoh3=fSm1-! zEp+dqlz;tY3HR?zG8DYsj`#ZQ#y$pH&^1&Zoh~H^Iaos|HFU&4g%~bamJ=K+vY?r< z7+tQ5;B#;e_@v9>hwUJ3v!@sCu3C#*4i#X1+eQqYB6wr#f?=IkIoqe=2jez|lfvOF*6-a=!Vwx6 ztepmj3rskx_gY-5X=F?Zp%VG^wGkgx=6Yc!A4fiYppke_}XYKzXJthBH6 z{qvLf=9wjISTc-#)fr0pFN+}i%4C*!b_y%}EK8Mre)4x+8w3{HRSKyy5>0gu2FuxQ zczN|G3^z%ndNVs5T5*!h*6aeGv!y6=#+*LQ{=m!FoML-7j1YRUvxNKM3x6e27sQR` zEUKmq^4Dr&>XZbS?BvDN?FNB=%WW=Zb&I4eAP_5)t?~W*NmQwQ7V7Kj(fC_2zRq&T z{F1r!Xq^VRk8$OkeHO!%n7@*Iha5Um_=W9YM%1xKV0b!qvI$d_#0vC>`)QbuYeql9 zb^8X1Z#rFp(AUFId?E*r9O@_J5d$!E!f?EPH41(R{0qsqeq#F!3-P113S#ZD82)5~ z;3ui*=1$rFD2}!K3)Q;w*yr^nAlx&@x(Jf@KH`)j zo}*Y*a3MBL!K+*ADLTO$ajT7lX0pC| z)A*G5E|?~C95%$whlUIG_U))TU+Vb$VfNQ>7vCLy3}<9`Tr^P&Ys*&RcNJ@#Hr<^Sw0>cC zW-i05%7v(V-~st9KMOyDT=1Ifd(cSpX7UT}pq7pbQ`sQl*D@bM~ZJ5{Kw4I9`KL;^W~H#oFiMY5%GR=NoLwTHvU=m-xO9oBTa@vK z9O9{DdKb7E-+;B7QqaF50%r%>VyuQ5DL<{m{X2BU^;PTftM6#>_G?w_rezRqTkHXb zwwhG4%as?MvZem(3&8fFkeB1Tc=6Bg{N9cE;B4l>|GL7Mx^*5HZtGx^4K$eVz!fn2 zXat)-qg(VhY8r3xsE65<*unV?=S3%EH}R|e?-cvi+~h)gRGHqw6j;>RpRUdobA}Hp z_~rZa;Fp{>C(-m_ws97a(YuE;?LG`Y7tLj!%bl4*aTHvB~YH>WM_lRF|!v)y^%NFbkAUhrv~sTaYqWn7tPa!UtB- ztXEYQ6_n3%8VwqhI3=9-^L)x1+w;sk{wFj4xESgT6mdseGRt?+hN1(BpgglzqG%q) zKRP&%J-1)Z*P49c#}?=Fv%l7J9!}wq;-`TDht>-{-)Qz$bc?(8^$mOMV*;z@4r1Hn z2E%=W9JZ*thVM*10>%|iyp`2(cFt)y1*d#r;(Q-e$Pw7#Cnxb9&eL$N+Xk3X=*(7{ z1+o5_Eo`~^2u|sdGcUtGu`qw2$*psi#VJm{tYbbD4XN&go_8UX2{M404cPfbEkSk~han^Cu%e~@s3kI?zaO}aRN+6HZ7Va1=|uG$PX zAS#H}>6Npr@+kIwrVGE;%8vypm2;tkquH!))$Br*GDS3umn8Q~;O>3B&buBXHb#35 z_wAh?>`3WP6N_&1Up^_ZkLkod2VP=ZQ%Av}ADJTOh;a~~-^=-p9ET>tY#zj3=lNr9#2r3kvnG>!pu%31HH!wVj$l`R91%FM?%dVpIDyM`kJY`@q6Hz- zORn$y3fCrA^4fwYZM)%kEQs~znv;2cZ@4OJJ|s9Jzbo==lLBn6wr06ie0>M8^pv*Yup3#RvLq7{mt6qIEfVm>z2F#D261mg@YCuyFrw!opZz(C&6d}J z!Gm)6iec`|sQxI};3xm0WbBj40#R%S^eK2cY zVS?R`ia2xs96Ykz8UDO`2AvI!%z3oH+iom}rJs(&zb(D2*;kjEhjen=Jon@M-e_9EwbHD{J^IxVi>u;jQFHyxKQUZkKo!#Cj~G@HHxI zP4F&=9#O`J9-f3v_4nBP5%ZzK8-`vGl%NhCCKv$OBFa+OYHstD+gLj)Q8oct6?0n}6Yi2h=RIMZX zdU%s4IUtcQc{UgtSIJP?rXbexRTJ8BA4^O5%E){s$_Ah9WD!Qoj1pi zf%D^id{>M$THX~nzq>t4X1|%s{NuBEnQy0IP1I)o^V3o6a@sAvT<$DnA5Rh4-_zxa zXMN@?^_O5`rwo4Xtbx1UcR7}wTg@pm}*OxA(V zZu&59s0#iSzvZkRrD4W36MPVPjLU7-=c{jj;Zo~jShrLS=LUa=v1cB^Z5ui6&WtMP zm5&4UuqJ-hpha-pt(pt4_{14)Sq7K+(ULCfd@%lNfnS*xx>>Y=e$XVm^Z5^#V0RWg zK5Rn;VSZ^6QpR~0tbo(I?~ArvJ;m?nHxNBl3i)J#DzS6MKrFey!^NeE81^R$i-*L* zxdXva^?4zvUy^{=(G2d${%kHc?zL#at4ydKGz=q0_A8N!_cP1YY0&4Z5t}UYkS&q* z6#5tTBFVFdd{r=&w9mH{b-hxiOUHgP?9(I@#KMij0|HEw5; zC5PL6_aQ&itCLIU>VvVvg#-D@g=}`q6dd(ruqbNQb-rOslOzfcpgDTMVbd;nIOrz- z-XvG@YH=(})$Wgt7xY07P5@>$m&`Gn7}Kz zrI8BPO)^;A*A+0JQeE(%3jW`OREYPl;qvE)gM0CO95y@@U3WI{ew8h}QTt}jOH1Bj z`{qE9zKVwl3J1AezphF`gnss!^7Z`n0qbE~)Dd*w*kP&S+YNVqF2Gj@9I&@nmQ5YL zkz06k8Q*W8G4=V=pRx=$(LPQEjLf?Eq2Gge{nPy@L}>Le%~->Zm@F)H*`x5lv^iBjG}Tu{e%+k3}=K+rjokIX}=pLt^+n35K;ITOM;1N$$Oq0$93}cEzYbjQnsR>&FG|Lg zJck?l34BcHINo#Ha%LI)NaCWN!SRz{a{BYWa)Fx?aMaNEF#6sgTsnc^)~+&qWT=EK zj~1bKNGHn9*v%g_s)0hIME+i4Jd+bM%pr(o+v}X&vsDV(sj)|gXFN0+19yl+Nf$a+&cy88kSnk&dV3ZEJCxt%8 zS5u+haYN$vaU4X+reezJ_4sSdN#4D461aq=&mf@QEd6M^=rF6+} z0}G2r2n$lAA0IAJxcyrpWa~*q)_tfxY&{-swWL9E9BV2zWm|SBO7rbJ=%890?H||* zBeI1}r{@?tEOZBuPiGG`?C9EOJ6bxvo2Gi-q&ay4Yr}mrQ}{8IJi9{a>%XJ)X+Se) zTOs&Gbz51sLjr9$(L=k3n9bhFbadH$a!6T4&#ydWCq9oOlhJxK zYU>s%AGew!($eACpfDOUMvi<&YqFp9^C((14Nhj+v4@JndH%y0+C1NZo4hEQJ2LSu zd);4^*4-{*>Fcy$4N)x4Zk4{rVD*_)0ECMU6{fOQ|3r0fi* z8S$K!sQ1O>T~6e9MUygyq)_RNRxWu&IAztD)1Ke9B)JM{h*LO4s?OD#wEUr8cxG>QZn9j$E@X!1zyI`b4mAcB*JfV)g>$b&@HfB;|W=m$-&h%XE2gNaE zI$^$^v`>wtT_<0%9hpMsW7Q(sW`36`A1xCAH4?T4_Xxc96goKNF5AAalpQ`Ezz#M% zA^3fR8rw8tAtzZor<00qbgtzq84n)7zB(NuH?3+mr^x|* zUah6y3*NKXL%wvl$tUhC31aO^jV9+-@V z?6XPiNOZi=5ssz9<~Ql8kTH7q*OXi{|Ix_}6Ub`lTY7x&7ugrCrSDmNq_iiM4yDAg z?`fJ*-ZijQ^<)!9#PUu*IztwA`v-_a+^ddgn14XI6$4FAkt} z{tD6}%x$nYhY8j1h-J(8BgR zG*0yf-BWcCojs9XeArgpR5BL-bZLucY@8$Bam`4K8$3}x5wP?}40yT5vZjL|3!4;iXvC>oTtyQ|Ibx8x1J#ITJn81U-awe)IEEU{Anb<=L;`Wj6P$f?Rjr*o!=mtalaaLFC@}w9SFI2!2 zqpzaZ(`x9QG#*u6*zkIf_HqB@X7b^@;Arlhi7o$5q0&t)@sQU?#almmiHo<{W9q47 zP|HvR?e-9UX{5ldx@dyhGOjqmQE-10UPV97o;F_)fvK(qlbd>v`+Gu8ym`xd>~qgk z9J9I^XImtR>()GGJuhV7WrZsaE{eyVfM;-2sT+P6r*M%M>tW#ASCH{17Z>?oLvMBz zf898OE_<}Zyw40YcySo7R<991^Y6l4=W6im+X^rm7=yzd8btiES_t^On%dSl5=L~g z(AHp1uD>^HZW8>sdMSM6lO;6x$W1OF#ETgXso>fd_ON!RCYI}bi%Wdg%pyNcqv^4Mxx)XA#BWy z^}O~dj=3ajQTuBxT)a~bcW+rOI&ky|owrqx=5IlAHL0V)^KMi7(8p|gn-+X9*hl@n zr%_m>Hu%-|@|n|x`TQ^Byq84sqRTv!4;@bP{~43M`bGBmq6sb2dcqp*H7RkYz-rmE zoqUHj5sML%S=V#+DX1U!u@!$$vUQ_nTP&R4VSg>{7i|xZkvEMtpd0wrDJEJIOr@ZCR?)pQTvn`UX z80$m&juR=fbQqua%96gqTymSWg>+^bGT(~*tk$+a>fLE)!7<(-o0two)A|XI{CqHf zdx~n!#dGq``XLRiVWt!zkHdG3pkvM;nonHnlK9G`q z^yp=(3$xxF%)UuCz>5LG^?Y&%Og&0ie~l>7$uuvv_Tfl+;dqWa{a%(#Up7fz4bFfK z>BZd5ZL_iJK@0UYl9$H)JjR&XM{Z3`FiYxto94-v(C2OIDM7lKEQB5FJ+pf>r1w2& z{&S+T`{U^74sEdZG@)y?m)T&0b5vDS4<&KqsQzUse==q>o35F`LaO`1E&d?sDes}b zyZ7_CgZs0vHL8@Z;=nx#ybABHBuKtK8pj<_z9>qLIwjiYvx|M&9?zm&Z6N=^BJ%P& zLm&J<(o0(n`ZHdgNre64MEP;jmn9LDN%q*a`#gCj#4y*0!*roJnv3XH#by*t7EOUN zCJnf2Iom_p9pDz!sf8r@ovi&rH_sjz z^Y1pjgI(9=QB1KWu3GNRM#ny2hD%rT`Q-(0cqiiq$_WqHb)y7C%a4j|L_D`5&|6ZjGZtDxZn7=sdq8cy8Tt5JfxfNU02kW0xpz4V z|1X=@mN)0F7}h{mk^=6G7oo)tGmN;93H}HAkjDZ?gbT|t@INJzb8i*dR7J6S1HQBB zFJ+wV6MfpBF$p9CjNxUl7Qn*C0;3|0`+YJDl=6;>{x%FH&HPLlzRZcKUs#FKq{ncx z>O9*Z(&7f}(xll1b}anUC2p|SDR52qW6OI~@&3L&P<4>lnh;y4KM(@9g?ZMZL3tpo zT7*cT4E#6pJXcwI9wMvnT;{|pwPvbY4&-iW}dJ^)TQ zWl`qVP{?x|fj;V+`CfxU{%dC~PQO8pWAJi*@Snd^tIO82$<5w;h29{rw6wsP zieqrftu&Hm2BONSHWt_shYPmsWA}bM%ZlNY`FOTr*i9z4E{>Inve~+gJJ9#}HSVjT2Gm6h z-qzv}p|@|(YzlWs&SxKiz-g=L{h4f5-Q1s7byXvmRs~u-G8$ea2%Xr|7g@ySQBul!wmYLN26DhslN`usr;b*KD zD8K&1%0uq+{jz4VF-2o&Z9+d#ZmWm$0@rhfR|2l#*5G%!u}GdnV3yy0_{N5kH03j| z`tAvn>DkQM6&O3XB?jhi&EgEJZ7BHJCGhhK!_t?B`4zXWa_Skyd{tRF94-v!gM^&R z)R2vAy-_tQeYuhP)GOi^KXZux+=Y?vHQ2SP*?4k&5_VQ5O5Td<+5D!_mYvhGB;Ho{ zVZY-9d^y1y8&*z2&HDm>#&;U@?fk&3z8la?U*Y@RVWi*?D~FTf{ZN{jPDL|p2;Ppz zN&gwp!oFLf_FWNIH_DjbW|t4OTSwr3?=+xKe*;|C?!{RP92^@6VRl#CKx;-AW_&4t z+M&r%H>8;B`ErRR2{WH^>3i1Jbc=CecR0nbtxU#f7LICC!O9Kicqb<>7~CLcsVmO$ z*zg0Uc_xCDc_9o8I}Y-TX5gR6U2yPW7c4n;94?#@_%N<2@H|JIZhq=Zry?IS$14Nb z6J0}0U$Ov8v`>pt!jAJl`VEGGpSHl86g9T9_X;a*Y-9(>7Z%0nV@Hk*25g%PCi$&k z5HlOU9UTR`UQNciCk0=O-VyY29)zm;vG5^!CLRBr2(gRa@EQ9x(E>P1E|-kI#)sXAKgp-azJ4rG1StfZyeH9<>o%DL=Tr!cQDGW8h&?iXILp~Bv0 zMrH~-W4#`&mza$R2kNriC%ioDP4HYu#8an%ysxO_3N; z-<9i`N#;G$Sg@adejLF9A8h1frVo_XpMOQB>SxJNc_u}Uor$ZAhtj-rt)z2eIrH7G zPNOrv^G3?!_#q=gS>RA3QW>y|uFks7mR#^+BPJPAnrS0@t1yU3r;h~9m{y73wQFoj zVI5nxa|Gp9u42DmETjSH&8$Z{LZnpa#_RSx=I6}oU`kODqBD!Tpt{hD0)*b(&i_*A z`I;kw=SuiH?{A?SVF%d|ui3PxpU}6B{ms_odeG?PTy~(3z))#PgHFMP$5wT~_SitG zz1=~NJkAQ8!w>91&_db~zmU7spYN%9fQxR2nK-pL-5c zNkgHbp)?e}DeclG5}7TDq(Vj-DEZuTJ`I$TG_?1Sw5Lj_e)snuczAs7z3=xquh;X* zCHcBMKD`fjymC_dA1$Hz&##s;Bduecrf<{ zeX_I%m9Oh5d#*Ma`68v1B~t&xo)jZ_>ha=P{;tiHncY$>qIR`9+FoJOnuMrwQxH&+^Q7 zF+6721jUx;G1Q*H;#(KLYHfHQ|(Z^vsM`fz+zH)F=Yu(x&pAfA6?F^)C0g`h}-fMf2pHcC0k1h5wcr@adB| zIA--9@NTyabI-)f?o7KxuPa=**Zu>v?}U{g9y|-m?xRp1vWGXRhqTUT>>%syVTSygfSmIg10E+pd~U8P*x zQF^c{nwKTk(^2W1yU*et=X8(Y(c}N}2C*|AcRIlv$K9o-^SZR9QwLh7qbt_At*2%K ziPa)?tcFQ_0-Fs_$UW!^?YLe}83A&h^q`sIX8`%P^u^07#&i2p6~&t94F0j?2s>Qv z!mGAyCD|~E;T+wad%9hwH^&FFVOURS+Umv*SM+Ga;$3X(v!D9Ohl#;g*0Fl;F$#wO zk*^f%i63enbHa=;9&oQ5GWVWh@#zOzP;A3BrU4jrE>Cz@&>Nax+2DqvchK}?8BRD> z0(MQ6Lhhd{;F%qW(GA1#&4NGBdB-O(ewcu!@?H2~c#)9PX90Gw@RDu+o&~c4L!r%f z3+&&O0DJcbqRo&L@RW9GbJM-hB=Re)_0I*bf4`ymaSG@k>x{uy`eD19OQhXeS5W`! zg+s;`qeWE`nq0dKdvgoWX!~3E7(E0{+|uA^^gUeUInjAir$e|y^ua%lLolO59)9jS z7x&Jdh@XC#VC67>{8Dd6`(4XuP5LUFcpw>b7o3ykBrULg(>kn>*oF14RGg!#M0}g< z|iB?<;Q||n;p*lp(liy4RhX} zuod4|1md3)%kgK+KJ;l>i}t2ba8TOk_uJhb^fO)IwbmNg)JF+V2*==pRX3QO6AmtN z4ZOBg%Jv%xaSFt2+W*sc30-{@B$^xh}p_tZ#SWo(T7tCxf3uyXj% z$sb3I@kRTdPMFfX3}ZgUfP6-naP6|S#L{uWk8gsY{^UUPF4&1d^A;1MUGs=SM+l(5*uTs-8_iU4@&>d)IsD z*8ec59(I(M%LP1oR}GIndbS@nxPKpxpNF!oE?pdu{R<1;Ulr`+iVp3909f6Gw`C?T1-5-9Czo>#)`ck zXvq>ehIFgP#@Is0YS*4sy^Zk9A#40FBNHg(6qH?`=NxUb&N~tKiG>ez6Rpkztmi0e3PmekWL-70RC z%dyi8OC0_FFuajZhhv`8;OgnSFh6_(whlGN`r?)7;kOL~Trv=P+=qpMJ>kT^7STJk z4#tj%M!WY?4q%_exARygF5j){y!^l!P!WP4ddFMH$?FJ#jutTf!VW=I`-V^yAjAF^ zQm%WbvO=@V23FOsWvk-5+|}qEO!}K?#FU>4YkKVy^_IYsEZ%=r8#B;7|4&VcKo>Io^-&`fl zRUC|V@}|fZr^yp0@>lD3#F3KI>F7FsVW7fS8g8>o@B`MlQbuqU1iyJ$ z!^@v;y^aFMic!%FK&YWn+8=rab z`_3O}aYp=qvBNUCZPBdh$27*|a770XaQi%QH9kQq5K!My5vFt?LOvdZg4Y&fTibL{jL z8y-Dpy^Bu#((yNMNIcHRJ-R54nN1_>A#+&2vx*{cZF|L5zafhK>sRy7L8n>qrh}sH z@hiS&uAtPw6#hNSM)5S@IDh~0owFC7;nSDr@IN7foAP@=?vO+tn%RYAdJTl6t|)Es zS9oPx^QQMUip})(u~ho(S-~6d4)Y2OPHT?!-sK)rT!deR!&nl zq_KDG0EJARMftspST@C2QMI~(f)DKAIH_CTKW4h*&Ju)}=v6!>XA7n78p7(q^*m4U zga?iO%rD-bWmYTa(^_$~=3y>>q;-5kOY+^1xJI?fqEkGTK)CSDM(%b#XkgmL;d>|s;LGq2qxb6Fv4R(_Jt z_c=<}240}!j{1tE9x3#4+<9KVPr;X$4^-rhF;QGD?#6xG`>;c;1yK4(!Jn+Qg&!rtcfM%^2b66GuY7*N&27#lZP^;Szi-^*hF`-o=@P)uj2zlmgCf5g!zernz=? zg80~#=GSO&MfaU}NtLked={J+D#$(}kyce0(Y@$}WuG_p6Zfa)QhV$B)Vyt;+^$)Z zI;(`r?#w?z(=WzK{G2B6HOqn}!y~cf>}~N#_c3y_m=OGX=Y%+9XdRRfFcMz$e+tdx z7qfj@Fy&IROjq zB=fgf=SgLM7~0@Cyk;b^?%=3PAnB`DA=b-%rhhTq)DK5shmrIDfN(NZWhKXie_=vQLKYBPTMCP(!C z5eY40GlZmuE*!V!DYdWM#A>TX;S5X3uad8h?Rq?w9eI1)Vfe2tH2$TRq~Cc8_xqou zUx~(W?@bg1u6QZ#zb)~X)!p#5tp;{llT4=jn_=j|+hsG>{}tcFya8d7kzl!Aozq~N za4h_%@Z`u|q1Sr{n(XO;2`fzL$GaXJ{~?3(cSlg~D~m8jG~-75y|6UM29E78Ayr>< zmN&-GfL@v8GJ6_i4oC&-y{m+xmm>u<^1_KDI-qrJUs26jzzdp@rR4=~5cIcA zruwvsK5M6P)%7Z2;_@`Kts? zavHoglBZ%Zt$+W4qC4r~%M01?`;i@#xf)_nS52NB=@ALlWp25PfTXRw4(+2!>p&I^<+$fG))diod7=&GN zS4!UJQ*59<7*ro#q9KhYe08}kEZ=AaU->L}=GnmNxKF}mhY zVt4R4&<_TF+Cjr`9vi%udKiXk!t9$?Y%k@oyv|!U7jZ~qdSQwW&{dj;-#7Ju4xoxcv7@IF@pL`&k&}W zo50_M;j$&qFFWa34-)P+pBE~Fu84y(bVMoBBo2GqPMjaQ7iwQdh%sG0%kQ>U3&RZK zWTwW0<$K2uq?zl#if`LW#TyGHKE;vGvRQw1sIK<~vD;ZCxKwQ~S;_N_Qzlad(sOq)+OWKl>}jbrkXcsBYt zl18n6Cw9Fy1j;=_VQ*(`a_g8SRPSmPZnzH+qaGN;k24PBnN%cycdQq5FCQ#>RXbJm z4{Dab^)I1h*eg1G`zr>S)QHDM>xwNpPsBUtH;6<2R?4m3YtgTHV}*!Ad*#~>Rmq*! zXVdOk8^w$3*Fn>f0C{r_issJ8=thsJ;tfkDA!yVpskrOZS7tnc}1VUxaV{ z`p9f;!pVBG9b|Y$2=mgBZe1QpgO7|5x_O1lC#`xdzR&B%w;Nv3sWC$!GUtg<7utcY z-y03VNj=FURSnF;oH@y3l(2mKbJ63-Qd;p}n(*5!UcPX0H8oWH6V4pfqT<6%@Uhxd zyyHGF4@gPpre<5V0RtsJu3&b4-cA{rL73w=cU&P)? zAkXY8%5qHUuUeHDduA7n?_cbMh7L5SbO$9ZI4fpjl9OB8G%{Uo;`C=wvpD2A3#ogy zi+@|9Njn#lp?XM}pe3LQ(5Mw)i5@9r{}<(Zs}k9R1@T`E1n@ zOE0;SU!+`6-#%AJoWECo=0J+PgN>T7?`5r6uGEXWFHRAv^UGv**eK0|uZl`@jj5Bw z3*Y3aNm*@;!W`L7(e5V8O1%28UV}4DiXbP8w+`^=r3=j)XDzN2w}aZbv9Pnazi4@S z5*hS;Egv)~j66PQK~vyaxda_9`<_@X*e%;aXO@f*rD>M1r0*5^e*OBg$Nz50J`H{+ zth>BN+%K=jn{x-?&X571R~rs%lZ}PG4Ni`$FWGZLvMM%LZXumERdjf$fvEEX&b<9W zHT_?a)u`Q~|GG7NF3(7q9=nFTeptgArL@L7#v2Y0rT!anL%)nl33{A&0{cAiFV# zel+aCoeej6h(@W?p^&~b(ccb^n7<^I>=9J4V*tdj`$_#Tt>kY0dQcWyBz%+dmE6w> z9JlP3_Ubi)=ars<@6|Oz(}5GBbMppVbYlol0u`VOyjnu zeso>gjVAcVV{k(~y$t*)td6rJhi`spZqN}Iixzz3$x%M@PC=%cY8+@BO_y)ggX^su zTy)71LnsWs2YSK3ww+{EWKFJD_VU(YUTCXcEyTs&L(jDQvSsEsC~0dhx_Vd$Q;Nr; zk;ek=GJhXj^4SVCz6Uu{J(g_;tYq7sdR#lF3|zbF!Dzpa6sn!ct*Tl)uR({u96K#6 z|GOJ*2M$2}f2Mp_>NURZbeqoZIzr(A^M%HhUGcz{wS1|w~lvQ-9jILs%ZipEQ>>@voW&bBkE)xV2i1APUv0NgVUd^03jol6w2ysRpd>c>!#zg zzsDt(`xUsDGLAZPJw3nh8G0U-c6urm;%uFb^vte}GCw7vftLeaJ-8AN7#^UDimBwX z$&1cj97abbyra8W|H%2L56*D;C03M$3kRM@(dL9B5aBSN{W~|3mR5tH+@eN{zMIhf zz>lP>lnArV`I5`zO*D6jCn;|oN``0($FF{+*Pjkxk4=HlU}?(_Ga5mkXYl>}vyhZy z&-){TICO9lALGH?_#>WA&5B_0%u$Xinj{RgkL71iG;zZdSDtb0u$c5@9k)6~6BKFD zLhoEIby@@`(>rp5@o{`$*+}hLw@|WS0-X4KS>E|kEu9~d!l%bZV*MLuJ~APWdTNUJ zV*OWIU$KBMw%1|x8_A@TJqc!26vEu1K4OqnIQyLy>F0+(GDII~C}2RJ+6Xx&*n#^o?Cv`A#9(4IJB&s;w1*@Ye!SMfai?@}); zfv(N(!1n)C#oF%~v{dfES5{=uRn2B`@zTk-WNZvyp6ZS%TL$y>5ec{@a~z%2V|LR` zhjTIuZZ@9_GZX_@MUMD$r5{_ISVx-;j5z(2l5lvC7H@o$PwQ<4u+Ip6g<;8ZuJO#{ zk00+t%AxHN+v`7Wvd+gZj}1BB-+-;dPKn)C>(SqoNE+GFAPe6igEhM*3Imjvkz$h$ zrFaaYQ2{f=F1Iv9{VxXWmDQD_wrI2OK~Jt8)seT9MnkTCF0Clkr6Hr5#9w7!WM>?b zVR!van0(Zdb}E3LA4;TgeG4i0QK4X(AeQxb;=rpyP5G3i5r^NmAsIKzovS*DlUF&? znnU_@;(3!erCygOG#j%+jsq8&#L-_1JJH{|O8T8{mp|XwmCipB2c#NW&t=93!<*uGcsSywANc0_O`0B^Pk_O-*Ap%2hJ6Pn;!`W8j{F1 zsu$g>&7`t?w`^JOB6ULlT7tRk%T&TJ9igKGsFc9zVG-Nagr1GM{6F^ovvz&%GtQHa%OvBvQV3_fqg zZ(jM+pLrc!>6_WZf1c(q10Rm!`Ve|bd1R84t6|IS=_{IYE0lz#B?;ZHJ- z_#;RAc^n?{#wS=eT?KsS+j5fv36co}fNEgls-Z^W;f>RI0KKZIt7B*GZb^b{5 zI1yt%U@#%E* zU$q###X{~jV*-6VkPY$UehUWQlIWKDbD?5DGR=9kOYBkC2}}!)P&DX4yEzST#I`@> zAI+qp#V5oIU50}=rVoc$eSx4kyXiHmQjQusy&5MNXC2!zQbc%4ILzt{i$Pv8mKOps43NbO@F-@I_ zBu+4db(8byn#o|{`p0)-VeC=T?A4KX4vD98;a}m^Ko81?AELdAI{$3GK@+1RsL*%< zY3Vl7wyKlz)U|(P!N=5u>*{00Nf{@}@bCt5mUgKF(`P}4YgW9k%O-xGG@aZNi)hrE zIb_gj9zD3tf~zo)Ehjo>sl5>%u+yF{nlitUD(E(qSSTbTFXt z%WufvsScqDACrZg6Ykt*APU<@NX7Ex+3cX?&I#RhdHsQ9;tm%}IIb~VwtJT}cL>#| zIOAQS@BK(w_oscxw|*8G?#Q6mb{+6boeTb!&WZ;^8wA%^PN@GD>EFp-jwAl*i`DDO z*jC4t6C2y%u7ByGRR?R3+33N#Q=cK%^g34ee+gAHkII7W2g8dA0oW};8~1j(BKEic zK$X_LS@V4W{}~$2&}Ay@IA?*ccB{j9X`VD|@MdAlpm}2Bob#NkvyglewnI;Y7V?PP zE6lxR%1dtNi}QZe($z(M1Uq~OXTr_+sfrQCS@fmpFZFPD`#+Ex{Fdw{>w)Rqp;#$( z&Ukhm9x~|XJk9M5$ZqT5!{r^}Xzg&mbNM@u?eaijGM|wDbKasE9ZVb{AqXK}f+Vnoj)eUIaLlj;9mDQi&&2kn0l35h?8ruAEGQ z2k{Yfqf{4a_ix06wI5;YRZHxvJOB+#_T!pDW6Z4>C_LDei+473#+>96;=?iNuqz}9 zPM?*SQD6504>~39u305S&(wxfytQX)C5XP*1^c%pCNx_JyqP@#9lit!p#-Y z@T$)dG#$Miqms?AA-*&F1$4vqJEy>cnN!&&@*=eNH$&9{1GrJIO}MSO75wISbJ=HO zG3LM(x;SYVdZncSdjljF@K_C~x|hPQ z)7-!|xF^|Ph{1D<&&URtnW0B?B4&CGL(_+rs2F{Ur^-iQd4G5Mxve|rY>XBwruaes zM^EWq(qI-_VyQNFs$gpU8#;}S6|SyZ&T`!%F>UZ6+18*Fbb9ItEE=jQhJ-)GlF<1W z);J4oMrdRIPb1M-SC!VMc+qn0C$wsc0hrp)b1a&43EE8v61_{Csq@w6IBjK52B|Ay zyRQNFeS8eNb8|sv+zFJWJHx9*GhouqTBo243t5$K7tyLgz{Pd3eezx<_fD^6s9_+NQ3&z8YBHb2P3r zvk_AKlJWhIDb&?xgZ#*Y4KzdA`FM}rCVQPwOfT!^3qIq9;ka09@Hv+Vg%)E`b9g_F znUeu}S}FLkpcU?{y@Hb*^MsK0lBa6#RY+95h3~s3z>TA3SRG)+JQV*3)8>7p4&n^#>d~HO{0v8Dmh!gm zN`&DbGH~2TEez7of{uOTC~#f?e>m+=MzcfUL)SFeH!Mj|`S(i98XSz4uX|vVtqpEb zUrZ(&RCrLR1%^m>lk|0JI6622rZoK}g}pkC$=wTDfs?^Tw;d#@Cqk0SZSW4%MfYwS zxVNtw7~frr6}oQt)iwiKCv?U9UT>)-s~d!8R+C}ZSQbypg^Tj8bkR`C4wg^m!Pz%O zXv(4!$9_}l7#kkFX$&jJ7V(XjA4t_!g+7grA*67KALx$@})abQ;lv9k| zZt7JX1)67W!Ru)n{PQ5PvtESgyk-PzY_*`|cVDSDwovD#Nn|lFfU`zlp-=Yl)FGn_ zcde`?>J>_bhRgZVZ)?8NyOzFlB06R-8q1UC-J&L|Lk^86pVHki`$Rv@!*qIecj*flz@9^b8Fo9c%DPlCh_#_L zd8({m;LpMJ)5$LLHr!NJCf^5=C+uBk9@O27`&EwPbG}=6lF2|eymFr=jLhN-KmUl1 zMM-@1RuF%lEIFiFQc3@=hnSbOjL%)LwnFK>R=WJeV)ZF zM}7E3KNp_b#hrh;RM4A3D^?yliFJ>d$q)5K7I%%|?46hC)u$+k$89Wk&!tloAmlYw z$ge*Qq!DqyY2by)w8B4-sy=x_mzG}gSKmS?ZqQ+&Vs)IDe=&fAlDF`vs1V6V^@`fJ zZKj`JU08FK2_O0!%_uM6+avdi8|%tBY0DA5JWHRGJ}sBdHc>q4>p)h0(JHzP%_Ebk z#xzBBnH1aTyeR#WnGG}W_$6v!ueB;dv9yhUq-OERDVxljNI@6PNWlyN~ z=`QecyhkOuJ6KycmE&3xIZ{LNq`LeEPYq7;S@p@h;-wK^O5e_lDkNV~XsXy}{#1_1 z7I^2gk^JG>c{sFcJR8_r@)4gQd`P33$~^43(&q?$?Q6}KP2*WLG>y0VbY#lv!*8WM zpP!E^9hqW9<@;=?pY=vwz>>==V=|@XM^a#UnrNpop3YsJ!yVsx@R{x>ST&~yzhZOuGT9M?W!xv%SmKQXoGLA#G zP89-Ewo$jx0*+fWroH$j zru#jIp!7|+_l?vW`@LG2zhO9jo#=!|mz&^PsY`b6unbT1`~*7F9r5(m31a&h$fl=lroc%tHzDznTF?;WTNlGmKTB=Qtt7pq6NdA)wCmocheg^gSi*UnT13D@B z+y-_C!&#wu@FS)LGOl(J_pH)|f&IP<8{9gA>fHip8)q#H6ZQ$-|MhU*)x{0fs!hu} zd>A3k_4Uvy@2c2Vz63{^1>y&rBhcY#8^n%ukQw>5!9$xy@z+Xy`eVCW?EL*aCeJ7A zTfLw_QAu?XW-)QWwLKc3ivfH5mwGg!&f>}pu+ku z^Z+LudMQt=KYvabeCe^gQT`r&dG5!f)3;!Xa~fQyHdvFEh>zbaf$Uv<(7AXYEDQY$ z>5@y@ebZlD;r>^4Y4Y~do4v2Wy_;`@O_z5I$GzO4sCqml2OI-~fze{Gen{v4u9r9| z_rP@hV06?hlWA^`2E&2wFv(S#>m;2O61>J?vr?ACBrnA4ajRgkRwRrH$wsa05S$wp zK=&^962B+(0^2+T7(RNpY}oNyaQ7TjIm?q1a$Si z0B*8n;Bc`od{4KBTiNsB^``YyyY>=nJK%+<8zU&Yoh9l>&V^~iJ)qH}51ya4UtC;s z5JOZJkV``X#z<%9mE6{ei42e&+Oo% z=hH;N;L#Ov&M<|jv#l#0{m&bmKjsVbyh34tjT?OHy$3!Ub#T`G^$~8o-pT4%3fAi% zKx1`Z!Qr=naoWEHmo{zZ=#3)%@&5{u9iIr`E1g}66X3%5*=TJPORG#S!=V3u!@rUY zR7=rt&PtM4$cyvPd`T!BKYEawo76G-_D-kJ@<vRUfCWnK8dp1mXG?e3pS>P?5e=u`Pwmer}1*>95ika_}v8a%M@V3rz8Wwj(EztD?u0fcoiNPR1xLBA z#_T;NIBt0n9uW_M{*_6x%$tkob@g&QSaz5kPGrHfw}B8j#|VR`xZ#Y2@wfp_W8Q@t z_>o>97$0+jOQ(jw{pCAx(eIU5Ji0-wa@{TDWHy8Cj3_Al=8E(-3o7HMiwhD4fyL9& zkX!y6_h~CTk4<8JJVEh^BioI_~ zAekJ)S+)0tPA^?aySq0Xz9tGLQl9gx;W#|IQo1kIZx5D5jO@}4|E=iA@lqCcti(HYdfyD))QiDNwi|wWsIX;78tN5n#hKR8I9N|r7CY`8 z?4A|{Ui;t54F1y+y8Uf|%He+aP}>Y=R^12FbEZ-c&j11!jl=U5>bUNfF@G!Z!g=xE zp~?3>nD=qSA?EAwd-YrjoctB+3_gKFZWLO6n2CzCwb*Zok#v~Nz>?Emc<{C}tUcfl zWNRzZz^0%!1%W+Uz>p789m*g(bt(aO<5exZvwC$cQ~9 zJ7SjsYbv9J*4Mjn{H(8#wD1%h?i5P<7siP zE7$jTlDKGjV$tX)w5)ESywhL@G5(!-RovzW#J?=51b-gTr~?%Z445}ItrMVYv^p%bR28ndSd#m zt@zNTqw_SW1GIF^NStDCPU|IB!%n-&@L@|00xN^ zi*O-(izq)9E`;^VgoA~f!T<3E@t(T@45%kz^T@t1cucl1?Z*LH)fx+z1PF~z{POriV2>K6Xcu87-iE+LFlzt zNNDd2Y9n?D!zHE#wRSlepVpX<*EU(yD|_(~p79k761Z)(S0iXMq3QbKrYUiFV3I2|g=o z#SvS^fNf?l%`=nS5H5xEC`AFmnla);=T}13Q^dxiSJeNF474oM$;I85Dym-!LmC~i zbNF7#fo_PUuQy;1zcjFH+z-P?H3}-zSBP3^EXaTNrki`ypzW74EjY13=F>C?It2a{ zu3sQ1mM@iA_3^^?>y8TzY3Idea&blSODYBrFJUgMioqd`1jz*YW zdsOskQW952Fa?Ym3!B3G3Aqd3$P2nmpc%jHL4UU;6vSx4*?lHZ{?rtQY8(?@oR1`R zm3{DA$sEt#?t;U1_+ZL{Jy5y3Qr_%k;p7^B08AHToUSYW3OBhsPamH`8>$<~ZKS5BRE}hK^Ja}CBK3J%%2jnd|j4$O1>oQ!hufsg>?+`2Yu=y>l zhP4dY5v^Zb&2TYw&K$pLs1LLEX;uGn2 zC$t|es+}AR#e=p%O>7Fq>1=_}qEq7dY41t*ZaSqP5y-kj)8!^x@K7p>QqC{PR?t@z7md zW?n}J!wy5I2bUmcP6d4JQU{X@heN>u4Q%W74U)}_&_-bnOHpE9JWGac>+6MkRRyAB z`~~4lMQ@3Bt0npy03~T<;r$ysF1GOB;ItL*z7V^({u$9_r=>Cky?zT*BBCM*9(^`IY7=C$MPPCFG!vmUi$^w&4TfmR^ z+fY>;<8dY(cuC&PrpP2&=e&JK4L7k1rEnlDnhcMm(Zy}O#F+k~t33bc*U948WFcp5mPGGiVO=gx5>|QR}@S(oHH~essSY^;)Tf``qTj zHE|K-RGQ(xuo)b-I2T-&{s7fU`=L6wftpv|6Ac@l)3MWmJg`GGMeRBztp9XYzTA5! znK?nDD%M$);oC#y@8uK!OE>8d8$lHcH;%vKjl=-_q_Nwzkx38^2 zWx_ddr2CX+7cIM5O_a7qgDs4Ikoj|O`nZT`VK_KF>2HM3Q%7^VrV{uyAQ#elgp*$2 zE}=w4$$9RcFidp(c&t*PB#f27{(Mbz@It&u0n3%Iy?`2SE;w740 z=*KPhlyFz*Oc;9Y0OTL-1b+ATqQ5MMc1pRC)_Ypyd!~c}B^Ose^*%f^yHr$3j)hZW zyU0#Sd4Q<($*8p~30s^l!ycb>asSD!I5LSOtkeX~JUyFh`h~#mamo&!elo{(rfw`(jPe$tMuz5c<873$9Rj#s6f zku0xsr>j*ffaZpD5a)U^~- zYKP;Zs!;MdZSU;-DG`#dxzNw`L-B1N6<&0Z=+w2%^hMNi{xLlr8-!#`J9rZHJ?7!& z(%!f+vI+~kuH~XrI?gtC2BKY{G=J*i!D9S(^nEo)+%`^^zpgewznn099xp@JjXU|o zrC${BWEz^ymSIfp3%vU@kCalqc)X7j7`kfXv-7cnmgJtROON7aiQ7LnwTr~-n#Njz z$E7)qCOQv#C#=^s#)_;vVAG`#^VZygNB4%|*|H2CF!&GmZTTV(8ovN^bWPD{mE@|3 znvY8V=PF75t8PoDpnv0QiD`6_LZ4c5<8(cKR^H25JKdTy4;O=1m%Z#iVK+t?x})Zy zK-5`oMhg$*)3yt$&N=HU;i{Gf9*NX*{+?iu`m2=1=iY|+@MurybG#nB0ybg&@3H7P z(j33{=>K-^4Of9{deg$2eiWl75)7KLke+>v#n(M|;Go}8Fuvy! z%-vMRPm65PEW0m;Saju?|2@PG(MIy6ThC(uXIhFr2J_!C4#+rwJ`F|a`2H?kcHiQ3da%Cu z`*(lxm~|d^-@SwXbt|Vo%hXA5Q$bDBDC~8rJ6)3xz@G7EIrHT~ky1C{%6D5iLe&KZ z|Jx%x9ex#pZl2~Zz9D${-znZ1J`ziuI^h_nLpbN_NqPP9rne7_>9DZ486`aT@){R?CJ5t2{BFb)g7j)9Fu7gm40 z0&nEi%8I8bVTsX1+-)}v=dE-_y%=5Izp<-xLy;MVC1=WZ-Ft(7yvK9VtZW{9elf1^ zGl+^J7h{Tn0nVRN30pGSIom8*!-qaa;*}l?ant%%SaZ$R`KxmuuJh4w9(Ak)9iLTT z#Vr}06DHvV{YE_SP~X{2`y;l6e-Os*G!z;{5!XGt4>l8JDAVYVg&tj;2X@^`-!f}( zu%-L4_yG=;fTQ?kPO&s)Y0B1)%Ip93E=hj0-;b^Cucf*S;w`_lr=% z?)67;c~(c+Yww%l3+-l}n5eF>_sysGw@jQ@=a0vP4au0qRqCZaievP9z|x^z#U%$LxluNktH0~AO^q7<9$rp;3-(F5fR%hCuY~tF z4&?~H6EyMoBT@VNe6k$j$^ZRc&HoObVtc~^F4m9aIULB>?)6fHSW4XMT!pAH#6hv` zg_%Ohu8@~#_vUT8syV=YGv5tW5oJ4m6Rz6C;eO^E5;TKM>-LcHPw5@C^`>y;=ZY8xvYB!ej<~YCemCadUqoaOzx~0ke9YyJ+hZg%Cx>(=6h zXlwDt!_E9h%5apeOynJP%2MCno!?n_@a%AJ8ar(fUzA7FqDoE9{nVSzj>!^VEox(@ zQ^R=9mNnez=n}rY>Lj0i{eKjlcU+F&8^_x!(o!mvmPC@Io^zc-LZoDr(b6{~iAdQ>dn*k|B5kE1r02d) zNka?Si9#8X8IjHJ{{7oOdgXcU`<(0geBSRrY=UPH`?q8oyA_bfShNE3b6<`}Ulp-d zyBu`d+K1nrOlzD)_thActf;wYIi9)dd}5&nxol~M4ig=*fm#2Y%`7C+1)`%D@malI zJ};HRt-Ujg-5C)Jvm?UU@48Ipbf$Fij%1Iv?8pc{wu zScz=(B^|k&B(3WOmdjRfS2X6bD`7@By0roYmv&&KVF_DWEzhFg6(Mo?#zwEwuK77V z9bG?(vYY!faGG8zb24mV{RTE{I`3?K6I;q!{O`iP+%gv8t;wwOW7#jQBKl@e2Y2Gk zLhSxE#Cuo5*xgnmrg-@^u9c2udcPA`X7>SBd|iRb`yFP^b2OMku9v_%*cjXDpwP^T(+jL^9;MaV+cpS^MzA9kHC|wLeYtR*zmoPGk4Y{f3KL=_>Whu ziLT*4YqRTF40o1^(?%xmnFeB>ft;7=cC6SWTJz*ZIv>;d%!!CrVrbD+_WYPRr>4@t z@*n~Gy5(5Jj0dcB!zpIzQNuRBQ)441?gd}xo^}lCWFn{;bOODxhR@#~e1n-^t0CpfSuWIJ z2M&v;pq0%j_`H_)enyXH5?XudxRWENEi;>lO|}KYf=y_5@HF_kvD`1C`AFW+Do%8v4orVoBq?J?O@{VjuZ*IKE&zhHDvp&xS z9^pyTuh+vr(P>Qi1BWJOl)Z@%6`pNDz-&F{p z`EzpU-bFzm`9TwV0xNJCe~%w|H4v8Gzev9xxy6jPEv9WEsdP=;GMe}~2|iCq#m7>& z;aZpx*}nWGCQwOHl)$&abKvR>vk#?X$B$Bm2vL# z5b|jDVwNgyO(%;n)PB!EblWs`ajRHOdwwo8|5wCj+*t!}RVLEjd)_qQF%lz@ryNsw zK!W#lL$SwRdS3eyyLd~NDWp~6hpeUasPqw(urOg>J)_ya@sp@$cN31DItwUC?*?3N>rU4&qKu3QO&Hv zN?F{zLUiT#!EMJUqUHt}mRcAHzXp`)lh+x7MdLlW!qd-SdO|%I2X4W&Oc8bZi>aZ* z7YI@v$NJ2oAXk`y-D?f$DUG>grtxuH)chQ_uBTMaQ32nJo`Q$`ej+d8Dvq!#u3m3z zME9NA$u+JS#&s>Bw7NBg<^><2X*<;|#|KSfhCXAN-hO{7|F9WeCR)2|O7u`z0Eh>%N&Mm~KEM3+-V;-j2hr*W^p)_dR99r%+B>cW1 zj?PzkO}*!5ew_`Q#70y|Sj zYrowjU&3r4DJPi3Ygt0N>oT}sED0yS>QpI)d?Mbr30O~`0XoAy5Ga3&WD6I=at}qi zTvFcRNcA z$-7Dr6^$g{<}vXx8X!u(IpqGI)!?MM9;V%GA(!r^!^p1=WOt4=mvt`$RA;FQ$L(|j zlg144O=XfW-pGKTiGL@%?%fbZAIyezc1rNt!5k)74std5YoJPG5_sKc;?#AM;C3d@ z&*<6By`Sd=Wruf?I6o_peBlm$)4oF8cUzio(L}Qb!$ehBR%uDTk0^kg0w)@OR2CfnEJdFWQZ&jmN#K)W3VT*Y zfXLuyvgM!%y>A%Jxz+?)Gm5Tb)Gtux|a$$=Ssox+n-7HRR@@}HXSr)9uxN5 zQY0!mhUD+|)o?ldvf$Q=(-vJNm&v}gd*H3q$FY!mWUY&klzoka>q~>d$t)3iyE#c?huLNbwOrT4>j#L(JBmrnj1{#(K+Nu@EEgd&lg<)iciwIoY zpaQ$sycbONnN9vSEA#v|CwL}!Nlc%rQJL5_C^6Yfy7T+NvC5T%*$xSGj;g}g7j~S& z!6?u#cmikP76*d*1&u!Mpv#_0h+T@pos5) zkDZ`U-O!}~?h4{WaoZn@gX))rJ(}7uMfM(1*ZM;^`&elFFNicqH3>$Tod%V2&p~kM zD)*cY!fIu4&f-oec^RYz0e@16*V0a+wfGloEu2BlslFu%PdCB6pJPd)&>8}i*TRDo ze+aL93E3JKV9Ja(?%9th9LxP8Ns2#&^?z#xcqxkLCndwW+$_j=v=zL5N(wc%4Q-2RD|1KsXKbjW$vj z*Dchfei_}g%NlOR8A5pY9MW*%BcvI>BvzBH=$1HLa@U+^(Kzmc(`u>^^j;nc^!s3C zqCQkVY7!2883mUW&Ju6OQBc#~EPS$d44558h87*bIn$9uou6^9r!v8{?5x5Zt# z`deUZTQJ2N7q~FBAE3W)BD(SnBSWhX@bZfZ{nukd!_SVOvrdhsO$jUMU`{Ui(LI+; zxY`Ln-Y9_o&JWyywbum2y+QDJW~1=zk9rvG_7?uL{0s9v7lY;27ebY=G4yc4BU1X? zi6%un0)=DtP@a#lYOn>4Yfhvlor!c?VF;a26%LcmMN|7SV`^n{89qL1hpNB2u%d}t zE^QLGtZH~czl_kJ2Ep0X65sOApfDP6pqA*FcL<$VUxg8SOW{aBuF&gy1RVX<3R0K4 zh|iNqa9`2@Wee>|&B-KM**C(nc+@bR@K270{9r`ZOad#mZX)d(Npz}Y9XHiN5#JSA zP~r5iLRl|c`b}yjYz|yZW1gAPUGdE{NFjo@y)mWh__O1S_Z_tSkOf>mE&>DMf#4Ty ziaGq*(@5np89hG^W|soc=gVk~p9IV*-AV<8V%$(`3dEZ5UG7)oX_rSc_N-H&@jrv< zBI`8zq+A|kY}4uDmPo<;o@5C8Buj%n_R&q(OK8R6Nb2+cF@&U7L(DgCXzaEkB0_Um zAeK)5T@axS19QN{Mvp2_%7tdf%TVF8gr1aetOJWCt7>_6jd($N{&jsriStxV7}%<*gU5LWCuIo-0WN6`R5Xx zk(Z@@f3o3H!*Q73dk$2*enYc~5iR6Bds?F9g4C{OVC$v>8r^falZIw^@K6u+pF5H! zE}LmN@Rny%eRrUutr65{)C_8@@fptbbwNahfEal#qT+RJuy@31PJhlqOB?qSbZyTO zn$q_UX70AOY&05+pB&TR#N1DqQeS4y5myHMC4sY=U`RPjaQAF2WUXrin6;C;?GEJX$x^snkV!7RJk4#m*#QolQb<@| zghhWH2RRV}sw$=r8LPAC@zwHlQvO8E6*^$SP8K-i7qf(L2Libqg}c6 z(3vR>SNtsL(2Ftj%(5zYxC&q}-5Ms0wSe(sC&8p&3RLa&9QC+-Ry}DRZNNlC|=dZi#0o1|A^^3kP6) zVmz1ms+f1)&7o%TW9Z6zQ)$%keWc2D7IjpJh8X^Ptp9304Ue8iEi4f#zV4&XDptdr zCp_ZlQstyEmDoP>rhFT3f zuxA|iL!%8#SPR)%u^uK*>&5UM?Mu2{_qkCOqnU=FhnNp+ql0>{1dBFordth~iA-D! zHtqX^w)6BjO#Y6o=W+$9J_7`W)m+n`BY5;kI6iHQ<@4xA;g*{wiugC9U8y=1)6~OI zaVd7lHv#qNi z>~BKjA10`i&*v3q4iU8rJd4O(iuD@)f@>azlzK~|^ler4_5B(a=&8b_{d7UoWe0mT ze1vh2FW@DWwRE^Q9^5+Sle+6aIR(WS_QF$%O&?f^zsAHfKRGcNh9nGJ)g*Y`NvPot z5%$M2mwZuLPM4l~hX0ke!}t1CJbN;PR?qF{2AY1tH;>OeUnBryGAc08#0soVcz|=- zBT{}%gq`r0!?Ni&F*e2yuidi+i>D^S1^p%5mgY(b4jabd0AJkqTaRvZxr;BxC*y{N zcW~;GZ0=4)8(wzWz>e?A5j+jHM0o`Pn>Kwn8g*;2DaJ#@y7K|9F z?<4Fvd;&x~F3^B|S;QkHoRyKu^rQA^Fm!tYxBILG(N@PVIiG#T8Go+l zuc80o>HTxyduTq!CH~=N4Cj-8soMqaQnB2em>W3LQ=EBk;q!&lZ~hp#v9p&I!MX@DTFz^eNBJs@$f6p-(rneG2CI zdv*n#Qf*Aqt{O9kd;I%t>^H%;^ZVd+&3Fr(SB?&H&qZ$N`mZ_{WUr0lqW&&K@lVUJcycI6YHz`lW?uZBjpNqd^~E5gk=(jxTHy733aJ(w z4~^%Y=zeYvsC^0GfcGHJjDHGJUzMp_zagi}=Zjwt3$et#oXhKcjz3qaGYk7TSlAN{ z9tJtkF{BI${TI2l)f1qH&nmwNo=dx3x1n5%3f*{1n;!3s!80yjAwI1K>MX~A;hh}J zoPAK3bj2SgN90ojnkruabTKs!%TATnI4MpgZMJCjS@4Y<~4;7qk zHGr8*=7MqCH*ij#1z+N(;N${TD4E+LD9u*DJ=XmFboF+OxOEIBJ=cQx2Tve+<$DtT z#hkYGOY=R*5Q~HT2{3PH5v=6gF=^@;NKsP7soQz8{kT8F~Vw;Mpou#S5Wu!E?6 zA0nqxdA}&1d3LJg8L>AQ7_n`5NjMh21^EdNh~6Y0LidA^6q0Dy5QvJ35j1U0;SwWk z>9pxHVfWdIaQFTxGMf{U1)h2|ZG{kK9z2Et#%H-d10!hO(bDRd&z10p@)*2b5lQx! z&4bU86KU;MMR+}?8czIcC*EdpoJ7=e+-A*7!KDt8QRd%;E3K3uGP?w}4XKi(4}B0k z^aE_#Rk7@02Mo?wMHX+Xr;AtwUD4178$VJmXT~+*(K;`PzuXGHL@RLASZnZnyh5Om zwUfvxUadBeYyy!FVk{$Z547i;CI@E7(`gkHj6HKn!M9%+thP{a$KpM>X*%%dx6c-p z|EbYSZbzu(ua7YHq9Se*Hgn5lqKVR{6yiIx4Hnn0gz>&+aID-AzQ`>gn)Q4J{>WAc zoFqk~pIoNWY7X?q=2Y@sN*QjaUcG(k87-8c5_Xb!vC?r(p4Dc`Co; zI~c|4Q1bRI z$;dSBo=YT7G5svO?wd(+J7-d_SQ%PaIFhzz<#9(QZ-k{P>*!1|117oDmU5iGNA!PJ^Iz!Wvx~os2MbFmK^mDuE=0y&4M~N}F_FD|Kw?6}^Rmya^j|cUX-AWbW z5`}i#^M&r;=FnyBt7u35TzYZpJ2F<`6eu*L!N0eA=}=Jtb&xcmEoHs1&X@N<>(`SE zi^VWM)EnfSd5_C_zNgD)21Gv(;K{vlg3snJQ0HAXs(F2aLE(666zK@l=cvJw?gqM( z_bd0eF#a=FN;ixzp>Ovchxa`1PIb5$4n{qK)yvlijua&c)?M1b@_s7du8{fMtmDq~ zKOJ?5T@X((ZUwbyIt+CtSzvrfhFZ<#ASHP#y*7nEcjw=^f#sXmiE#`E}Yr)X&3_LfIg=S^}Kg7-HlOLyOhT|H#yZJxr_NJJ0K7I)LvTk%w z?Ood0SPS9WxzJ*;fhOtjdjs<@`eBAMefZl6j?Z00x3(71#KBE;!KbTq|7`x;!*{Q? z{EEidBY7axCk=b=%F=F;Fxb5`ksiEO43+k_w8|)w<^_pZ&VHh0+3da!?p8}tQ|TD~ z-Ysc)EO$LEkG+o*%|odMf3K@w#y=yp0;tF}SGqR)8Xc=8NoD=^QIkXZ^tsLSA8P==8%IKPyA+ka*o+E`YiO3l6gV6bPfzLnL?3MeTQ#mxjpDPQ ztThCOyWi6J&1>nuOy0S}cbBEFPoct_4`A}13{W;zCa&T&a6)ng{VCx^bzcTq>@^J+ z+}0@PKJj@U+Y_2}YcbFB{?}uA^XTe-uV_BIG74hm`bmu2_FL?Odh81ti^u zz`mJu-196@OZA2iYuboMNrjMI)Tie6{mG}DPUN_(vEbS^QSwB|j26XMLFnZrc=1mo ztZT}or<6i4XW}tBPkbH=b#x@-xhL^LXdX3VP!;;$e@emy&@Z%=JYOSne)AJtR+dm+L%m45xTgvK7x2H&_0 zXgfdKB2lOZ1$z9fq(YYdNQ|ZDL;k_MJv))tO@Ke&GYMWJ{Bsi=QDu6!RERaO_Ukt#OuMxcL*nX@;aVx6WGD6d)baxJflp7zeAsv zOyG=c{&7g(fNLo@592_06ejU=7Y|2 z55iyBREX0uKy{xp+_u&v78|EcuWjZ#e!e3?^1di%DtiI%nf=BYv%;xRH5E6M?6Ka6nhB-nS+$Kj4Jez-tQ7jV)Z9`qh;2u^ z>GACOK|cGc@tMyzxv+e-e6H+<1iSj^3@mRMgY~kdpy?z_x2tx;zGDS!o304Y;NbZ_ zRa@AlHBvRT3+JG_@B>$%*M-lu?tuLR9VV^h#G=d(GOv}QaHd*=t}YWaykV1z6Rz0d#juu+zGWh=<)`p_*wK={On%U6YoxR0TOaaxRA6 zf1gatenvrwrYbWktm97K3FFzyNZW$9k52U^Q5n=9$9p_Is~~UDad=%YgzJQlgiE_5d6td|mbXo0Uv!t^aC<0; zyrqJTrR#Cug;;8rVFGoh-NCMP0W*kPO6#?yX?c4c98_^;hi4h#q~LB)%WlF+Hy`ti zOvKSVYozn0KXuYegAJz==;@8)nfZIZTXuLS_$zFLeD`-yRWh2DnZ$9M4h*sJ_e{A} zz02rnnJ&y-?8z)F&$4>wKw5C?Hn_d)g14y?z~{zk_EuH{oAN{HjN{{(-_LOTe*Y5A zi|vOSt4FdAl{!qh_c@L)+QWs9&9U%#oP}q9IWiTtgUW@Er24yRz_dRejbHJ8cDegd zJ$E^s85&5>iEW?~XN%Yir%BA%_A@tOjSTg@Rf?57W9oD43vl>PjCBg$!2ONgEN|9! z`hMX75Dnr4pC8-M>6^9aph^I#-(HK+$AM0HdzDSuBFnDWtRxfvxzNYUqS=$~#gIS! zA-2dRaIam$=s=YX8@?7wLu)6o+|{ytr^b>MhTeo{wf`~IsB_%T=hbXx=?v6HEN8#05 zugEHsAujT9#;9)tJuJm-(Z~>2A2N zyKAcJkIuVa=x-$Lzh14+%H7Gjen{S z{h3V8JmVQa>AA!o+o50}tU99YiolZ3k_K!6=9_<)V9hs+f|MT8eY=voG~uB|`-m=> zp=d>#zS$EmHxI$CY4&jCwLI5!e>=FW*hf~F#*!aDCJSv3*pY^f5rT3NCwvo{D7?Ev zQIKU7jUg$239U=ubOm3?3Y@{M5-M|(Z9}kUQ!Pw(t_I&wAYP^J;N@vb9wcSp%Ya`* zeYG-3`g+4cr;o%yvsvJ4yoeJUKL(HUnfL|Az3F_rcf#fV_OPG#q}=g4P25AJgoE{k zTy1?6*|af)EB^M1_{c|~QO|uM{(1zK{7vPuA6$azSN3v&I!at|v^qCn=0j#(*le!b zD2;E>3m#E&f{vqu2i{-x)nOOfXoD2JR znah4Z3tm5|6!Zj~;NFZJA{{|eU{JFN?mc@<%I=&aQ;OGc@4qIZ@o{PRa`TG7s_-=q z)F_aEjWXorh)Y6|w(le*=K^VTnZ}j$y9>#-Q>0KJ&bm(&a<`_X36lCB@n?ZvvM}>2 zKK*WlA0l+g7T+-8jrb%IwTNf;ZuXqN z`2Bq|`F&{{7T-{!4i6cr# z`7>&5Sk4*!xJk-0T)ClFt%60SCrIFZJ5ELTLty=28@JR!gNjC$3Ps{yb6br9xmQ6Z zpzr&hcsV^3TwfsrnmL(d>!B!a*}w)E*z<;aq$5I0%6|&3kp}|N)#LElJY~@RJO}Rj zrwK=`*a$!LMssJIHG~Fjzc6CkPq?*t7u(?^h4)-l*!!E~*pY52W;jWax}1DKMc+xm zg$q{PA5R;s8ZX8+-@67uK1X1=`)%->UBM-cxeVShA*3#I6dV{v)LD~5^drM*^_c~n zt2pmT{dk-^v929`&C;mevnDE|!Z$abp2ZHIskmd(V@OSyMdk0015wFOaPzzm{<*M) zNhG}i3ww7MENB6R8Ivq0ec!Ql?5j~SWQ3(zR3rqCHpYNy zG1Tt%Qp@Cti!HTcqAc?^tK+8dP-Iz`=-7Nk+Ulr6-zspl;Jg?$xyA3DsElRieIxp> zat%1S{pME7=yCF1_fg@x5}hh44>eP!pxewpFec$7K9N+Q*dYMjg8#sJ9M2y8UW^kw z;!qa82xg@;K+3q6T=!rC%e$FT{XF&;HZ82A_rp~!cUyJR1x|hX2C~(>CnXooB1s&CdSQhDu&qn*etzRQ(Sm$GM{~6Cjjw;4K zKEZ7AB_)~|`B&(?XEdGTHwe1Y`*By^3p$E-4%fXZ;v~N%q2upURF!|f?w_7QCKw?^ zPtl_ng)U_8lUVM=1j59==Rx1`i`Z3Mg(l}rY5kuxs^4t`yDnU&x8?S*-GZZz?;fkma5~4_lmX&@0DFU{u~}sCktym&WXI!7X`&9=cS49#<-0Yq>v;@~Fa^0c(2i`FcENY{s>YK7)^j4u?qKK^grHN@Mk2gdvSxG!_2@ru4{;K&0(~u>_(^EwcxEbk1l*=OK+F$qdR{A zs5IZAQm+m|rgt9=9LQup(mD8`HU_u-F=7piqiK4oB%3%_177(~!Uwa?!O9tF*z(Vk zo)1j`xessYG0%teVERkieMKJgBMg}7^y9>KZW7*j*bnZJ-(gR&EQFyceJ_#8Ex2+G zM4lah)#EIwOtmVNe)EhdI4!5Ox#qOuiUK|Mh0(Ov&$u0*Hq-rYo9W=A<1psuQQ`5) z)m3IbvFPRBgnu&2F_(9w+9#=SE)E5pMQbQF@43hAwiQEZU3Hw-xEn8;Bw(fFTDEJi z8(W#A%ier+WJ>K5(R#!Xx@?eUe?M-w;LSViz?E)RyMwS`r~7P%d=K8)@fwRKO=6Go zR-o*MOtjG(%ianQOrDo8+5Tt37428(mgSk$?LZ6+%Wp&_g7 zS>mrrbY8z03m4=zH!!Vm`O7PPn6S+|Rppya(bPTHMQJvkNLwP~3rc z`OjC5lL27A}@xxz9(jiYi5fd(Ak1aw&COn1Nm1#~|_V6P&TDkt(T3v&hl+vFfNb%QNi9 zg4P%4u6+}o$K@cY+s1a!I!ncFK4BiaVrfEaFb4na2i?n=e23vDHdjr;u;^zfqVs~L z2Kllo$2=Sr@f%ljn^~^-d0f`F0H^Tx<7Ep{@W5mR);8b(s#7{(Zs%xb;&lSAsAb}o z4f15x_Hcat^*3x-$$zdZHnHGrV_J9AU(m;A6U~~g;3OYDQ+}V{4LU~Pq*<$2QRsBe zYQ#U>AFYKgekVD#CL7q)af|)=9e`6Tf3O|vjUo3z_hd z%<7H>9JWzLpJZ3&pcKX`)UOEKdiCL-lMTksR}&moa6-QmvI1@Y(@eBF4gY$CvxdXD zXzt#^PNZ$3*Xs;0K3an@Ut* z{p4{>y~cI!-@;@jdN8YpE7|1WQEb~S3zQyp2$sK;=h;1&&CX}RVQ?R=XUfSm4YxoZ1{fv#0h{_i6+xfF4u2>`P-l_u^ge zB7ApL5*5=#*~yS&c@cOd5`*Qp*?aZ8-SCf`)UsU=}v?g#X#n)H5v!^I-48kARos5i#){QCibIP=y9~d3mDwL ziQBs37Mkv#2;-ElF%d5%I=WqvDe5QT{)>8S?VCilDxU8NuNP;tWJ6iH2_a#MI;3yL zej@6yiTtt1!WTN#c(=umuFgHf1lNo(`U&6vsVx#_zl|h<(YbIcPa0qDj$tV|QJ|dE zj$>=w*vu!_*{jJ7ShS@LB`Ut5N~|24+tMKD8BAiuUwSe3OESCGW{6z|AJC-Mh|}wQ zFXX0ba${_&amlKD9PR)t$VN*(Q}S^{uQRRSdqsO9WoYlWSk&AW#Eor~;7T6$bLXFW^Q>@NHhdw54NMz_ z|Bg?E4QDp9omcy~Q-O<^!H8vSxMCgKliA3Qug_ymTeqW_w>bN_v6Ly9d-HCS>w+P@ zVL^K)We%PZ<~QF}!}p>dob+*pU|IHIeDTwcUC;eQ>a*wO{IumZo>RsgZ_gdm}s|8Ze zd2sa#r%1R%GI!%qNA;-9=|Y{(NbY7`K51U9VfpCB4=UhybJE>;$OLCaWUVpKnFYMH4?q@5`ttFCkHogi`vCUpWewi)zW9 z&@g(}Bb5do|BB+Tce3p(C(>128@GS)WdM+G2$_j(@Cqku^Ee!aGQPm%kkdyTk<5X7A8#4`A zUZ0XB8ax6m$89u0r5!F5>rlnw6gqB99~^U;2OE-uNy$}X+@|jcw&%>KtcIu{(r5zD zp_##%cg3OEfpPSWvko3tX(RCxJPUh7I#geN4}aT@vBa|wxTh+VUv!cEwo^fNr6s@f z(8SD^wOr!y1DM}9121$2a>>IRVA=vVw2r7id(UiI{NOAXv~(`B-&l#lRu5?Wz&NXr zpWG6QTr%SY|4de~0Z~&wD*ad%zv_lTLB|dB-~1aRn(m`o{5{TIP>C=kSp_@(^%dJ%2jf(xH}Ns_eh|}x^pD#n*}C%j%T|+7}7&8 zBxswrEXz7GM0|u^U~`M(!bhvY8IwAgd2|rwpGXl1=PrWqvS57P= z4|&$yWv5Ecr%;`iDe-J8K`p|mO{>E6u#l%=!W09mqm!Ey$@&cYL$j=-Z8X6#wAE}T5E7jHK) zqCyV|yu#`&&VL*UBfS<-{_7yDeqYD6yIiGiF?`NogapF&9C*eRK}~i78bA%XSTK!$ z-G}gA%^&d2-G%(=Rs<9Mtr##R4df3#q+v1*kdxhpE6kVB*P1exriSBKOX{HD!%;ip z`ZR)z3fYfiBa$&|avq*L62&FUNucn%7LpxP;arytPUf8qe)*!@tCA~NlD40{n;(w- z&sCX}gFOsyI)$o1aagbW5W`e#neU6+9J&4XjnVaa^l95S0tz5bW8_;4T78%EX4 zJ6*+cOp+mbRT!DNZ#*`3iZk)v%UF3Wp8Yr%fU~SxvEW!0Mm~N(mV2~vC9c&VmwFHj zc1W>d-{@I`eQ!z3$T6_+;!fOJ#y^)-b8)Hbhw83`EIw0}NOS`%1tt!>e{gLmTURcQ z}%s8Nk&eHDerXShib5P0ED^JbAhv%S7yP{yfU|M(<_nD&jTfGAG&S z*zs)1UVh)At;q^cYSi54N}$}slo>tAWV_!vGVyU4Y|Gm&Hd0%U?JJ&3BV***hp`Q4 zsOk({;|BaIh{EZ|?_t6o5E<%zX4llZVN5zk-^>!qhUBniRF5^}rLnYoJYTZ? zJ^rYTW+&IDvm*XyS~{26(Bw>xnL4rx(fK&KKb-hDNHDVjx9WWF^H?c+1{a>nwuleR zV!o$cxP-Xz*lXI2;geQk(R^w4?7IkKO=5)qvydHsFQO=)qf|K?hsQi!aCTNKW}Mpv z)6UxPo`DjS2|b5hum73@JG3-ET^;bg~ZK&B^mT&V+*%kD0gXC-yZVlKs=KWh0k7XUmy)jcl+Qq#gT;f~kbf zQ8wc`uA2%cSe0-NEBL%!uq^siCviGfqU_eSn~lP z%zfhq+g665dB+ZLvU-i@ql<7@YAzboeG&{j(uM~H<5`#AQtoBrESBKBirtYRcwn#r ze_sw_>w0CG{dwN&YCMZAoIRg@x9&pm{WnO2N0eaxyEr_Q;fn{&<1qYb0^HD3f}(l1 zFiu01Z4-L+Xcwebg>r@3|FK-l33xSgEVtY82K%VFIG)%Vsj20BL?T9gq!qj zsF~(&PB9>iHNphOMo(o^5-xG-?+JBHE+P|_tb(Oa&$11n4eUe>??T#i@&B_Z?7muv zgQE_!aV5i~WJ!_mzWYoeY0qJ20^;!FMIZdQK3+hbZ$ZM=`RLl`EZ|Q|kZf6x8PnoX z{Y4PlZd!ysr^B!#Rf#z+Ph>?KXTzj}^@8R_LdacGhGM~^z_G;^U;p4;^<(zJ{kTgU zJh_1VzR5T?Opfh6`wmWB&Bl)EMKnS*8%<8Eq^(QmF#onUf-!@@G%twZmUYgo#q1@N z8d#(5=L-|4X>pbYu*|kpBzV_#O zzh5u*43@$#`wUE(=SdPySrCt)Lv+S-0X)hbOCHJHAaZY-L7vN#b{&wSD`ysvM{GQ~ zvzI01*?u&8xiUuPMU%Thv5=DFKtE_Nz;1*8X#U+u@T}NILM~;K6(jQ8Txudc`r8a! zKK+E#zct7~dbU(z8Y@x5cyTQ#g z1h1IQAzlr;s9i}6^CA~8qi;G<9A(R_*BDLToN47B2;0Hk_k`@|bz`Y@fCO#eIL%5& z8c9v~HwbQCLH;W|NLDQv3yF_+lauCS*?pDcX=P_FsNQO4?TTHA-%mMW$<35YE!9zS zlQYxP^Ay)jlc#g5$I%jwt2#-$h73(HCb7E{>2>KCvVQO|nLq0_I7HWzyXg{CP5KCV z@H3gLT>6!~=Bdz@`hUc?!iD7fiAbW2E`2%nG0fc7LQWX&VD%M3;gm%aF`p_=`du8T zwMGIZ??#c^o@<%j#+77y+A(CrihmttT~#)-MV+^qc@kG(@jx%(n&sqc*mri)lo|+kILKaC zlcDcM)se*WYTPr#bz<@;f^3VAB|j8*kV{tLq;bz>(lsE)-uykDh>07LG)!jqzRD!s z((3f|SU1jPl*B)J#-8wuqNwY>Q2KP`4x$(1z;2M>&aX12lkrlEi0>68VpyjD!HN@T zedSbQEes@{6J*HNb$=ivaS`qbzRYy$g%jaaGkWX~=UQ=zhBk?pq`>Pi2p=nuvj=h* z=PRp;oNW=KeNdM6Cmm*1q&#JH2j+s;S(crCA&hnVY|h$o%#k%)WvK_($@rS5Ld2>R znO(=dLHBVB`?735v%flp-N6^5wgDR%zexsk{NZ}yx9kaPd#{Ep%IEO|z3qv>c_x*2 z$s%Y<(!1zdw}m?3hJf1bdMeM3%{O%7Y>; zOZs|y5o@Yu#$?~`g3X@4NoV;!;cassbIzrkWv26K{hd*O#%suER~zEK=QruNYm6TX zLrJLCW%6K;6direi9GZ#CUTlL*>zXcIQGYWa`Aj8$>ZiDspjGQ`U`Dj-q#X#Fd`WA z=6+>^TCXrI&#$v>SsVjwNgqS+R)T6{$G^=UdC!| z(q}Ll(~X3TU9PBxW5DAMy##phPMX%FPhwrUqYQ;M6LRO4EFA-1iJ$xtcGHZ@ zqQ*7|C+*|qZ2#)WC5r_+B+J=0%Q}h?ex{sW%fPs8}3y>)Lk9KyLv%~l1k-9Aoa98IQ z5s&>t(iIz_{kAoY>bgv%4?7dG3fGbxL4Gnqnd5$I|S8)8++CSuJda|&D#%qEY=4zl;Jwvh)P z-H26vop7&LDp~zn5nap&*sr>|DR%;KbZ4E zYfzh(rF3)fDpIstjANE7(NojHiFe-^;n_qBX!JP5%o#VC?0Z-WnX3Q6nAm)Jr(!ML zD#IfMfq&TShD_%E<9kfgHCGVYd|<-{7ZC^d2y(6~3cDj$;@aLf!k?QXabkWdDi3D! zowqO;{cH~|Yf{8{e@&R3vpbk`$7U0ea{*>t%*GzU6yc|yV~mf7CVhzpL}i68EcdU( zVV*Qxxfu^z)sNt5s|38;v<7(hR$*WEaWuZ8jl0&Kz>U76NT#X`eyG((g_t@vFnBvk zPI!#lemmeQ4Ld$h_As(X^U_bef`TR z7|*dU=N$Ed1(V)ky^9&%(LF+>X0KrMj94(Jmcaa>PS`pn1S{|YhiMa}^l*T1?=+uPcQgVYo+0u$*s)_WmJEq1#zB zn^i(bN464W$?=$0{6e_pTO5u)^9}{OBk<{y^Emd|2K+VJ1RFQ4!H-6AFsY5>g0Mnt zS{{h+--Y1^c`ZCJss#SaZvvXc$-y=r1&57Xce=+IG;K>crhq%<+vZ@O(rZ);IE9BK zAA?wpB96JAhxWx+K=bh~oW7(Kt$rOqUG-k9{oI3=YvBu~KP(UIdc{*%83Z&hD#ZVoAdw=c)n)_e*U|Ggw$IzqDR(R`e|u>|#l#(?I=22?Pf zhyTo;nD_4| zq2jO+&gi$qo!2Gs&CnhG5!FX9m{y2CK28zrE|`m--|oVNp0TJ|pTELoh;X}tC7%|_9<3x4B28$Fl(i4y@r)l_HdH}Xg--t!JAy`yC8RO2pfu>h6 zn3ZCR%OV@0VEZ%NRkjWnp7I9QeGf69a{+dp&O?JQ3*p69bqqE=ig#wjg8PbWeEn2W zFrzmfZ~K)JL+vk&&$UcY%-cAc3@|_6rQ`doW4GAikP@_{`%%j z7{8uj4>rvsSC5V+v!=SCYWfV|pLQp@<6ez@#FJ!zUhx{ov`IugIalknhpD?E^qq$i}uiga_=;KqgC zwDqVJOh6}g+#zckeK(r^((|H+M^B-W(*}q{(+P6GL7isQ*pd&0GBozrZ1OE{EuC|~ znocfuq2FD0vk#*6MJK-OA~J?Q8S%*Tj=C!a(X(@eX0x-2U8^Nnthqwc zn@5Cur(b6!2A0vfws{!Madrlfv{<}9I>tRrW%fV)QKlWcki8w z#>%zKK!Ys|TV5aqnG4v6ktVwB+9;81S1gU}pG$Jbn$>7JnbD|-vG{s#Fy8i!fd{iz zf$5%Gbna*--^cN)7joBPSZ43dO$NCfnq)8M%?k5ad1$c{(>CY{OG#pCEylz4E1H zaa$-kDNb|4D(HqzANJmYuXKymS@>5PME1SNBu>AUu&=m%*1z>BtTMX=4nApSO+3xO zv3L}<5KANrPo9LLo>6eW%92i*97YRNr_k!Xp-`P8MN20akZj=?`jCA>4VGAoPFD!= z*(A=19d?|Ve}>CJr2CVhN=tG@rkOY0^9s38=S3@5kEhuij+2JE>9lyHoy^@%NlV*b zF#q1fFbB=Zzzu=ucV^?heOtIq6xKaD)&FQZ*2f6NX?d)ED#7F{*nfhvsqMGh!Gr}~vonJFrJNXbPR zy7pua$HUd2?!^yD@WC!3YokP)9>+5aGn3I|yEax^aQ=b}JKWS&happE<5O!HVj}Q@ zF;1KDkAeiM?X3Xc+7LQ1a2BrJ-;L+gCV;uEChm!N3(D!asJ!R~+5dGX=G9c;;R`Xi zFtQpAel;U6Sei-Rql*DO)|fQP4!xttkz=Ea!0m?|Zit?bD#QwJa_f)o?hw-AeVM2= zpC;l5cA{;j3hGWwp(F2Q;QHFLn7zjZS0~NGsU1tP^i3qle0vDE^DRFbWWaCVE?m7u zOt9U45sAit+_BJkXTuXy{_pt4WvE--VJDZ?V3e$R@VeIWB>h~)aHMqHi zrwz;QR=Gwa4Uf^s&-T)pA`jHaH>T#z=dd=v%UpKt52e6hR0da#1+ ztNIJ?Jnyj$&4y?xSHZV`9K|^)zT@L&2cpJt??QG8d5X*l)JdI+PMwMZtw~Ps%jGI& z^X%}ZRyHn6Ru?Q?ybD6z@-e8;7lZsZ!?clJe7qqWWAC42vcKDd`ObE<`jG&ssZrEt zku{#U8Va-a9ix^id)eK`4kB-lF8g9DpYC$*LMT3g=O-6puADrswR1*%sErrzdeU8U z!trxz7=8Km1jjD&APb2ScD8yz`wKJ73T%W(w*oW>b^>*8HO>L9FK~IkUXW3uDmVr* z^y$_B_Q~mSH1}cwId=UD#taNYQcx-mxt(PB%Qf(-?{rMst%VQ0=i}|aL5y;K9%?t3 zF<(Z$!4!25oMDoL4jPRpKK?ngTiFL~R+JJwtL5w<&A^=#qpT1&6Aa%+E%Bkiz{r>d3x zFma48n!|Q9a#n}!>ig-?pFcwOLlWkAZKvCGI^h02A39;vQTXFGj?Rb?G6A&}r2UXO z&N%jlb+vy*$13^3dh2gsInZ9C?k9w~fg0E>=p$F$kK%rpP6%=o$0e_C37b-OQt3?@ zRDJt)HsRJ>>h{%<$xdhlHL1h&;!Xn^uuYl+XW=xGa+dRaoLGYm`^$*4_8M4Z8cU8JXQ<MX@W3y#4Z)@WL|ik^`Sg2am()R*x`r&2%u~rNX!`y5Q}=Wl%0vqx_qx z%ndbpdhF9XVYk6o;=R+4dC}rYCrjTYGvBR*^J;tW&2z4};6INpqa!fw$PF?zc|WOF ziibAIM7lPj1SU&gfQ<)axqadxT34?}&q@}NZnbi>Zmy;~4o$_Y;ScDdC$aSWfe>c1 zyF3=$s0Ph+&h2!zhpsbehNfGS=|I^u;rMYrw06!aaz1Ajrio=>@~9#_GIa;6)|o5v z_<0KKre`o)<(A^_?8i`H!{xwVx08t(n;^jc8SpMG#-xlAD!gcl>~0BEy`4^P3@Xvj zIWn}scr)pUNfh3?wUj!4uwvP>UrE!*HWK=)hp2wGCvLqn@XudoZdO`FK4$ErBO(=8 z! zo*A=u-KUsKpW+=V;Vjqr8HKR|Z$Mdr@Y_MUaH)%X`m6g^VO{^|k zpmWbp;x<Csv+RvP6=aC^8xNYxF`Z`#~h|l8deD< z7oCQiDc767 zRJacQNeqI6!VkDV%arcWb;lmdVRB_b5vdgWPLiJ*f!pdhTu|=I)GN#5nl%=f^ezjs zMe2;sj8~2MIxGDfaiZrM4`(@QsQI;ip!S4#fy)b%{?=y`^l;3K5?G#RE92y zV%O3oX#=!z>=ruz@eMXGat=*Rx(@#LSHg>sWhB0?8t3~a(V6RylEXJ`$q3II?|*;H z6y$CM``F*0Cu4%n%jL2ETLBE-ip3&@4zlJ!0mnX8W!v|Uu*ovYSW&Eo4?;dc@SR;? zW+6cq>_*(rsWY3trofS_NW699K-cdFKRb8>)taQqEJ-BzO=<-wy;Y!Ri{6j}ao!}a z$`W)VxNvGz3j4@KhM0Rav9p|maN?>4Y;C;Xh8l-39{pB0o)?aR+}{4`zy~;;)j-x; zYthsbWpGeSiuS(vz+Y%>1S;?Burs+A7RiP|mikwic}@Z?7cRo6o>Vq1btbFnvKF-` zY(@DIZ~V&n3uHnwNE+t{I$i6C`GaG~i3%(Hc4#GCJaGm(GC(JXOu{r;0xOUIhqg8qAnFXI zO1d%-`f>`{S1^MIg8oKJFeenG|?b-Gyd0qK_!kOy@l&Q%-2*sMxp z8xvza`8Senz4IQl`YXsE9arW- zWD)URcn9R_H=?a=B3~4_73U4^h4#dy!2fd<=C8R0qa5_0JvA2Eo{Z!7Jvl>S<%?mI zi6Q2BJ_H*lcV_Y9`)tQnS7c9!v+-xRgFgdj#;^Dd$X|{4jtzlIs2V%EShjNd|bpp4llW>JI!c1FhYGfvJ0rvy!0RdX9hUzsYrkWf#k ztq0rhH4IUja?G-PHM;h3EBnuWF&am2hab0+;l1fT{OaEgb0sH`b*t_Z&x%6KZ;7UF zqkBoxlVY0l&m8ycZ{nA0M_>$ZE>3IYa?@=`v83k;Gd3@Uc+D|p;y35Pnphv~{ae5o zZp`QB2ReCkIV;dYI0R2pw_?Jc zCYXHZ0Q!6^h524G;90g6)kocDbBoVn^8;G*g_Xb;^%v2051>?$x;6|2RH)-VbA)RF2ID-_O7xylNltB5!vgmotoh0*=o}i1n|~c9Q+s=0 zzkd=_p}LS<+cO`x8TsIj+CAh#O9&I?^O|Wu4zS&;`u_>jnpfM4?q-=yf zS6}pDG8nJ78|iY1SiG#a3HfhVqSxYo;Cty3?yn3)vE_<_n$zdt;XYGA>-<#ATXGni z|C>S=_~$`#Knc_GCkmvlro!Z-r=YGr3E!EuKxbGHQaSKp0i~2vJ6a!>1KD_R1?1P z?Zho*7r~zP;rv`_ylb9>QGG=)cBv~79Xtj~zva<5wh&5xZH7~S&cW!)Vh}q%8}IjP z5tYIosJlE1e;@M2Y8U~t-ShB;Z9l2tuI=A{Gf>Bva}lq-!=3d!z{`D|7?*wy#yk^4 zd7Tu9=#Ir*ACY-d+Cbj>sUbT)i}JVFxPn++L-DH-qy7$++`$cL`RM5+ z5rZ%^xhN$F_4$dlF*U^QlOOiVM3C{Lu8>1-2(Yov%-$uf+}Ua>&Q(Z23BI{NO;G32(Jxhk;C7VuqtX1E?ZoS&s=rD&v+0IM|^@e z>yu%dMiF!DQ5T%_Ga@}#YVhl?&k(;d4_B2v#Vh8|=eHWQU0|M@E^mJrpkhUwLj_HW=C9GBxrPcEOQl`m^)gbHYa{Wl7R86 znU2;LwtJ|J0fHC%~*$5;uMH2o%X?>%98%OaUcLvF<5sRtcMKX!r=s1Q*0>wWpWG(KUec+ z+~a}k6$Nr-?FchYV;SAV+dwS-WHOQqzA+b1h7-?k-TX=U*Wm2q0)B2p8>`xT6V~UJ z!Fj7xY^l=(xBc(HDsr1pBWM&+5(bc$_F{}{qX`+C+|KxY;W3LpBPs20CXYgFg;9;Z zkT6q@xxBlYS#T&5lK1t)$4x2VdrAep22{}Lq6%3@q-ojKIN{+8hHO^0#xrpN(B&S^ z-y_=yd5iXfHa{7ToV*Mrdbi2oE-w2zbril=$|7kUi8aSA++<(th{4Mgix8eQY zzf9KqC}xf0Fk|}dD>1WfXWSZ|F|Yf-v(a0)XU|Wb*rDsf&Vg@cB$jw5~D|W zX4=s0+{*vD=sh$0cNp{M>Lfgs^At{`a|};jBJ^HUs`PEEBryhL%eu^k4l=HRB%V9(|1RlEdGhOY|NMlSs#Hu8Nn1nZL z`EfExYZ=r$oaGLC%LvrHtKmGxlC)>&xKQolCoiJ&a$l_c%#T{|Uv$HmtUZFWK3y4WX0e;Az?@qHAe{2kQ)J zu6a1;)Bes*p6>^(v(MK&DRY7+vouKqH+BQ6|L#x)X& z;(f4iV=Q~>lrvG`)|D|gkHWS$&1Cn>ELazPj_X&SB|Bv1a83+OQu9xhOw>7n1zg7C znL`rDX|CniF`TqP!QFOL>nUCSgMf4~e|oWg@W z7a{FRBpiI<4f_||fg9#tu!z+LlldaBKe7h&zZ+r2QzO*uCk^WgZxL9c0!{9fIG&c+mKw zi>i!o zvjm$_G&l`D)raGrrGBt~+!Ty02nORhx%h9*X6O_pF=qK6Sv@5~LAy!>3P<;1FxM5? z+%AV&lN?Y(@hr9okE6Z$9PE(j0Ck@v*lBPEBtabqb0#r@AZ^sTdL1uyM_?mchgsW| z@czYl`0e#Q&h?^)|6P-$na(<7|FkgBZJ&;*?>YYcrH9bI>c9^qE91r>{q44-L z9QbTS-O&U0ZB)ZIEsY?$>5T_3%*G*68VVe|@W(4Dfn~3hKq_J(`dt`>PuEKc(n@Z@ zo48|8mb(mArE~cK&EuFpHXl=V2=Sw0G{?b@#(PrQxcib9UUo=>OT+p&p|2H$^;^+t zN*vzTpNKZIjxN>*lW-{32?NVDP-A5db~{SZ7bmkZI(!?TjTOqzf=DoLsE=V5+?d%NF zHFk(G>6XKBW&d#g#4|X0QXjuSBASTyDGPRL&w%xPOF-N>gnh_;_rnv!aiQBq2(y36 zh%Fv~d!wpxR9Cj5?l=%tw_&Cvf@sH;@pT2A(TeIO(v8sH$6`-8L7F zznz52bUCKR=Y#P$K)wAta321Fmy7nG_<>n??&S|W_MKxij&{RismW+1c!4Kn>oNaz zB8kjZg35CopV592dT;y(rKVE&x;qx;M;f7As}KG+b~N|5785kpm!adfDT3OGU3jN1 z2g%(YUgIAVD8Kj`uCmT(``C#1G9osQb3KR|>Ito4Z&u z_4b=Freg{Aa_-*gDH-&lAPaA98;{<%PBGUWY-cx(_5|HBOT1)hkBg(zpwjg{tb1<= zQ>H%xr=3Nx>p%iIlokq=h1KwV!%Cics5`l`HxX;|`|+ib7q$L;f$q9(NjacA&3t(g z`b;D+X=5j^LFNp3Dwava?G@&Rm4=n=ymmWntk@jtR6>K@2#DkYA0w~#~Es)(Aw zHsRL*5y~Dsg++&wg>H{1$^DnbINXjx@ia|RTin3k;xs7SWnB;ar8O{QWWo8aEwHa9 zjI8j9!$q?3CS9)4RI4N7D4Ap8+`e)iJ{BA5JzoyyXJShvS;tHe$a6&izj^2UM<~7-s#h{pPxm7a94YyJ&uoFpuU@BAO&cUhgbqls2gllLechVXY(ni}}Uu4~T=s z+|0k!+XP%&%9#3&R8nlDLjQV+nD_qLIA*RKN;U6>%WEBJSKL{uA$}LE(^|-(RoR%e z^91%qynv3#xtJv4Dv2Xc$+o{g;Ct^|X7smOa4(NSKaPo?$FXl#$Yfx5RWa6DCZXj7 z?pafF0>)@@9kKVv`O_7L*)7A}U>@5C%hHyU5~CD0Y2Iucc)gmOyRw+2b|ovAk#2vI>=y|a zA4%eo75eBF-iGDd@`2tOqSsUa`%tz|f^o4JWP-~CA}PVGW< zH|{J2ydL)+>0opq1&Rt7xF< z3AW#!4{X7Ac=|hr%(_*>?)bcu*~fnlO*ut)H&c>Ngi~uBIvA#LX; z2*=+44Es-{;8D|XloCj=^KKj@1;0PQ($z6wQt3n&&I!ZNIlqL5zz5`1ZCs>Bq;Cr&5X zU+1GzII!PU5_r}@KbTRst9YJL`ml3v5H>5`rPqfF@$yeAY-8_me&8Hjcs&aL{^9mG z<2{(w1)=PvU^%o%I>Y`NdyHGR7vYPFhtv=AGH50RiL!H^mq$CjrYA*%P@!%}5ubo?_4m7hllryaa6 zv=_ey^*a(t7Pr=Hla7GNUsPdG>Kh5Zod`ZJ+wuP5YNCH(7OGzjMLoqcaI(*gEqibm z&YrHt%7uOKZrM2o_n(Hw6&k4jc>uK4FT;=$x1Xyz0j86<*{a+^68u7zap-Shf6wrM zX(<)3LFFd|pQ?j=XT2OTbS6}&tdBRSFBM+H>}m#jPpmiK=l`0^3h`x z`R13v`Nf{IleEWDha?-4r}zj0Wq*Qa|5x_eA$@-EaZmQu-Z^CZ)=291Q&wbu<1TBr zu!v1a7^s$l&-6 zZm%JGf>&RPu^RaXMDuZEJX<4xg$qas~gIR=eiG9Vcw2h(=k zgC8T(%n^kE@as{*M>mqV^QBAV_smA{oz1blz9*5I9Spg>egZid$gxOz1;ji@7R)bi zfIq=fux~>2!X$$k9dm}YezYApzcu^9rwj5&j%<%%JdlX*jE;yb1no;PS0OPT+00nu@OfhC1_9(i>P=W8vt4 zeI?tO&0)8}v{s%-8Xtt%o-5$VabA26j6ua4cgXbE3fMTm9bEdCqqxc^R_RDG8ruv& z{d*_8Q+yM|I+meHX(jBqv6KwFmmu{rkAQ#TJXrdv(BIoFg`2GwV64ZUnuOU_=wYV- zjnPqXqJJ4qS2+a{^0G)C*~1asfRD_gsQQnIWW26AU9wO{wrtG_N3nWwvHy85Stuzb zCMGK@_WyZ}7vqU}@AmQE9I$b!ulHW>sf>*+&z3iRI**erSn-`4r;Dxn-@pFf5Bi_$ zc-(NK`hR}@pZEXiI;Q^rT?cm=R!)wS|2Nn1fA{nGR{z`2@9R_*T>sPwH~wVM&O@_B zEmT${e|nZEQF$^(pYbGX*Uuuev*W4tr3R{@`_YP3QaY>9n#270b%yO# z)}|l7q>6eRj6^?+&QpsX6Vdaf$#hIu0ljUvhxFc;jy0O5Jee%^z zbS}G}Zd{g4uhm}S{o=Z^iSOr;CxNeN%0>&(mY6A`Ji{ zbn&Zj>Jc~*l520%ExT8Uat?ISoR9jVfW9@N3BIYKe?^bTE~~%n%CARx{!?y{3-jy9 zP;VKF!FNST7B@w!@q#F3{9bO}Dgyy2W5~JmE4aSeg}!!PNQ>f!nf9VQlC367{(g)Q zImdd7t~J<;?%te&_p=&o>-SEv&$1-izEtei(a;*ic043knp52UT~C1wEQGv*JqGf3l7;0Hd+XL z%zg2X{~GdGdA#V0`2u`UslsuKdHA#P7M;6Gh3;!9#%3<_uba3NUnYdnD(<{=f070! z7Vz--WiP1CQ-H4=+v>GIh;P4c$M`?lRArXA=s<*nsP|2RphbJ8z-#|`Jas;arhIe5 zJz?j_degCXOMjMDxvqBS`Sa(}NK;F)o#d10&hzk8e@5{V zj)S)JfXP|@(hfOj|L}vox5-5y6}=p*6sM6TJ@ary@n}eJoXTt&)r4J7Ckp)g>S(Rp zXVLe@OQLx%szo*zR?s!^szN`n`KUd9E_HHcseMm99hMFU>lX(|qp^rA?G3A`&&=en zT%yT_4p4SyNegQbtB6vGzv-7cFH!xjeIoPEo-n3K4dz?k1N(q0ti4hLHdrqb3@XHm zq{D+n6GLM~YR_aK&0!H#t&aisiD6WsLLeGGDuSaE@8G(44-!$JORKM`Q)h`J2;-}v z-c@mW>G(B{)BAwv{4%Cu<(x_lg= zwtfmC7weO3=n_jVyW&N%o!m%}QWl&#@srh#6V~KxGpSj7`A3cT6gPI-q4PDrj|?-r z#_N;P&ADiqwg5%VL&CK4BfPM0o&3H%&%vp&7jHi`z_3XVi0j29lHsI6kLp%Zh2zGQ zOIOq9+8U6*coa&T6@tHJJn(igWb8goyl={5r2HkB*EcHJ!cVDe)5SbSs(cSDo+&4o zwDT-_sg#qbZvq;ncbpcrR*=#DXKCO&Za>1F6}l++qPSrY6C6+l8C0AJ^QmJG9#$f< zkF!8!T@nCuk|&ihjj7vd0Uiw}*upd|!5sNJcz#tBIb$A42Chbt$;}=_uvr32FS(+^ z{+*EOwp^eweVX7;{X|rbo=$bI%!QG+0q|wg5{SV(nCZL@ylXd+Y%4`(ZC4&UDH5es|@(+iO#VJ*{cPBCrbkh0%1+~>^*M4UYlO}q#VWG%9 zYmsQz5=DWtJLSBa3YZ=}j~%@=m2a+jOsJ@vh1Bl|O3aAkU-9Jpxk^ej_V#G1JxvZZ z9ywq|e=(eXILr>&ij(rbWHy+b#M_-FjEPzooO$w;%k zCf3UZN#1FA_WW4EBa{}nZv4kC%ex6~O*hE>FPTJ5FX#F!#%&v5-9Oj3i67_u?uz;F)`5z*!bUbVz(lWzEcfnzHF3pShAdhp{nw zMGU4sgN{rm63Ksn#b;K*#ab<)>`F-#vz8L5n?@pAokE(zCoogTXweDswInwUga?a1 z!Z-t6vSBw5$~{g}l~Jcq+F5}ovBO(LO-JzJFB^fiv%SDH=3h;!N+|DX&~ElqdlkE6 zdJMB<^+61qT|%sCrxC9AiLdS!Vx`vy@=ME%J3F61hmYv8H@D~sSA4I>&QrPA_E`&^ z{^}5qulLwv9%C4fLIpbEhAsv9(iIXwMWhW%#|2bOY9!rLorgzM+!2`^mZ zu~!eDM6|30_Ib?`g%1&bNWEXnoPZH$(KE~w?Gl(a1fc|mw zrGXx2`Q_t-SX1i;HuSSNomg;}r|eTlR!$zK6WWwGzHL6Kz0?AM{+a^y?FG0^JBtn{ zE@r%T)WhvzDFJB+z|W@^l0B0);yNx{e34@ueYHWDvwRg_!)6hF9W;jjG;OH4>tXtB zR|Zz)zhQzW{A7%DjOfWa3k(p(u(5p;gvXEFt9fp+M0oM-6->|d!HUHfVTaTgHl}3> zQ8e8KpNk`fH+48?cAF#r-s7M6!N6G%CD$(;d}Bnm&RhVJE*vlW-$lHazZb=a*D&`= zzu|(^GF*RuDJ}}(X2D`lR4~0!i#`xIt z9DH~v3+iujNW=hvcPh{DhU_Ce{$U(mR67g?d$pOO8`o;&R_H>b;#_d|9Z%Mt3u8z2 zg`Di$^$)&bo@@Lu5zrs z2xqz_DwgR>3m}KqL^FSH$Md3z3eij3ORpc;4HLFM1D($=iTn3%a_oK-F>YT4*{9C1 zE%!MNSj%;kSNlyoLSsSxZ7IwSQo+ro#&Er4BWXMOiCp-dNwjs8@ZQcbIKeg&Z%K|v z+|h*3O}JI`qY%E8ZJ=cT5@@WtLM1Qz(dn+yT*gz3snJv;N2b;@846PL_QEdMYZAy- zZx|xN_AQ`%u7uqGZ#n6We@NZC=7_G|OQKhX>ZrrMYQq0}j5gm|Pea)!B*5z-&rJ3X z+v_U{{Q3~~u!klMv;9XT?)cF|8RzNDzq3f{FL4?cC{B3~jlg`O2g#L=rRrUlDC!W1 zTrCcYOkahITqM;+SDUp&;d&3~$KCSq@`O3YY?y#^eu&X$vrcn-P8HyB8PqY0jp+hz zWs6m7H>c#`rH|U1}qF_U4jguBD9Vq0TUs+W8;- z%H3mArb|jL5p@^FAb_NN7nrl_E-|FKH+%qBO_~ zMPw5h_nyz^ND(TjP*fyTv`a&~e&_eUf6jI8<9ji3i zEdpCpVzJ@&B5d*Y9(u9mDGqrnE?7pw1b*#({4b7GWZzQ>wz<)kC>meJuYZ{04-KD~ zw#;+vk+l`9>wRer`?Fq?je5{d6?O(Qm%HaNzb;v@y(!|XaIrU=Q#S`SAJ-s* zLYo==fM-mAv<)?-AegF75J7o8fV?j3#~TiEzgxX8(VQ?jSRJ&9ky(9{dM&NV^W4-$ z_Lf=^vUUqmZL^0z(OkbNV+zp>QlTTx=)vNo448Q)gjqk#aYK`3Nm7Ch|KO20zTe-G z2H{8z=#)?ag$1rI4FpIf#Uz@TzpJBJj$%h@y*Aje8?SY4Oe8yDp zj=8G%osn94n`GZ}LOHyxXs(G81oy21;~_n;%E@HhJDjnH>pxz4zz<-}3rICeT6Rs+#-aChB zI3P*Qu00BM&qSd3vA1BFiw8OEJ(~=6He+^|I*Ftwk;_raM0!~(bLl$8Xm?Rm^vg2* zA8@YDpG)!E+<08FZv*q?of9KkyabXWMacX|AMo_$DljK@862&Qr|irNIiQ0Dqwe9s z{<_r*LG8Cn}Qq0hc60Tqo?zY_-_PK>eH}>t$>Lr)~2O* zDA0-T;wVvfHsd!>}>||*5@1pyC^LF*Mvag5v*{>O|WG_ zG%1UUAmg_SN#?El#GmV3u&o|=x2rY$wyr1Dnn#K0h$fpIag9`pDiFSB6rMh~jMQ%E1-r>`&sXMnS~WM@+{qYLnd0jb zX?VXy3dZStRGMu-ISpx}&Xm*Gp&*usBU@7ATScV$CKJh={}@l{hs=pgA9N!{gsfG) zO19QpvV*g(M_uVFYHpEa``A2z*37_oZ zE+aBUdM$D>)d&7yE6TjjC!&$Fh^j~|IX_93{qsVFeRSsoYWSRnx7Ui0$jM7dj`?Sj z7#>58jeligx%{-A^fuh^ecw1qJ>WsNEx=S8D_*Ko4EB*M-hw2(J^pb%Ad#*x-k9TIOWMVJVL2b6`N z-}^pj)OiYMkz#6N%3nd#woTM||6Y=%?}%TfjNyN!zsbpW??`{(ac&3gLn)ippbc*O z>9m7uXf7^Ji{6kRVf$}Ehjux*uA2Soq#9!PhtwpcircS?P^5D_eDm?N223p&+ z9JDg_a2drIa1Tp|z}-Svr6(EtL>J->ZVk+fyr+WX327?5GLuRBHkqcZEa`=s0yGr1 z0kX>PA^Vmi$RR%fTHRKoUthUh;ORRMjD%^+`%(0+Nr$1<<2Ix_oPr@P`<$fp8^Y4F zsr>7gz(=Vclulnkej2^#Saln^M9+e0<&)95Szln*gD`Yq=1#OgvJSC5-_VpnaTxa= zhC}Yf_$%)imKZi9{r+ARr^Gc}otwaz6$+rAK?IK+k(KPE42X2AD5iPW@E4N7#w4C-!%1&f|0 zuocbz?0Lu~xwfXP=7Yto?dsb^pzTgb5v+h&t%{stubbymY664Qob4`^Wc79f9XE6V4;A{_ zeKK;!lU^QT9m`XYxs^7_3;8IR$-6-2-6_QfZB`P6l)cz#%WE`Ge|>{dWvbw)M;pKD z?ilaf!cofA`MtnX@;VU?5@pS|6EbhGicDU)d8&?T}#duIb?6;x@Qo6M!OAotZQ_8e48yi=7S}n@ki3C5`DlT>k{pP0ewYgWUs5CMNgi*}Vw2`%kz=nG^FD00q_&8= z@S|vP((y2kc~lULR(w$p6p-#6>y@5vN-zAL-VkEP= zj2GP^L;Y-g$l9zeU>9!HXKx7E5p}~DcIonWYi~08+t!aB^!+9aiu^COs*cs*yLOJD01V#)Yc*bVfU_y3d_=cWMcCyXnEwy&sw1 zdkk5L3zB5OY#-H@vyNJ_s+Y%}{=#2mtH=lxrN}^t3zfQ0o+uj`(}7{PU}}H?bttoe z>U(X+SY#~~)cqZVbdIlSbXJUfDitO953Ql{RwL6X^bdXedY%#9(*bfL2T}OfBIMql zfKOOOLq=l)vU8Vh*m&~`FYAd7$+l4ey7~|D<}&f!t@lub=U!@{C6pM{Eye2gcVP8v zJIH*|fNpCBZKO z(f;U6G-J>ZU0oMX#7dqqoylf+i|jZ~jGTuT8rR^zd#4bcg^Fm2i6XY0<3oETPNq$2 zqp(*^5m}{Gj8$qSDgW-n=+e4eG%soom*LDsv3D-Q`p;_g>?N(Jr@4ul$y748J6dtA zi8^t(nE?AI9mMafU!t(~b0EFq4`Y}%grg*n2xet+{-S7MyeF=X$$QP^b=oqR@nx;3 zF=!*zH4!biulq~T`MD0gtg%O*3N8rzKJUQKlk1pmCZD?VsJuO+V3_8=FCE&m&-lyinD~QDUQbhHX#tVI?ox^U5}fqW0TScuj2u z@ip`$#Vd_adEi6*;_Nt?waS#8*?f!*^D<+Z@1-PGJ(JzN;}J2M5H5-t~tsm~=P+uZivAvR>=3$T}&UQl28%fQyL7 z^I7Cx$N<{YegU`2Ti}vK4U~n8D4V6H#9kP_NdimSSQTa;yLPcEt1ong+IC_V@-y;9 z`QXCtoV0~abx&aDtDVDV_9o$VoU>xp8;Z<6*U5YP#uaU_t|O{mH;GAaFBV;zK-^MX z$p!O0q((!6NSrC8q@r(upW1b5y~l2{gG;D7;z9Hp9mg-0_~9(szu4xK4qoZB6N~Lu z!Wk{X%$J-OC~44&HxX;ki%!ZAgkRZ#4s`h=aUni6yki_m-`t2~JYOND6(6Z{e<$qj z^&F)_YKKTCwVpKYT#kN)3Iy|SMc_DoJiZZl1CMeXht&3`P&#Qj^TLp(9)5UF{b8b* zLSJ$C+E9;l?z#$oxAiiuUH7TWLHm*CqD1s+&QY}VO#~tt( zn(L)V$c%mD+2IEyd4fk2PoG8CQf1(K-5#zRu?mJY#mQ9tE7-Zn5*@ahKz;@zxXwk3 z+PGK)W$2dk+*8c>$E)Mef*n4LrMCoXdjEvkD|QUOkywemdN&~DsXM9UwV~9v7AHK} zED>K~-s8c}{dh}kA9q$r#%(hk@YjNT{IjYGzwi+z1#8b!qBi*?(sT&>JWHoq!?oEJ zOZtdGMxy|~n}Sv64I-D!1o(So2v2z$PplpYprUdO(yyGs&-nD5bgGK7nHu5bv`heL zHVS4BxT~|@&RLW1RgpL>_$azEUz|iQ*0xeFM{I zC_71u?ODN|*!P)iwkjjXMt(Ctf15H#-%Q6RM$WR!Z!KoSgHlM@=}+jaUpX-{+r?^$ zYLYdd9+3GeacoAF2HBeV8D_*SrGG@ML~}gvp!B#*aBoi`XBXt+;>{_f*H#({jg3+o zk3;bSwI?uOEK3_J8`1~jZjxQ+wvkC+3(^M{QZ3^CBvWsK) z_7ID4a}Ee{1&oCMLZPJ|EjN+F_-$Q-w*A@8#!V>Uj-Wnl^P`&-n-0UuHIwO`yPA-- zN(bq@tVEP^j*+Xox>#S89+Hygh=w96?XsrY;1}s1c)8!X_s_|Y(Y~LC8xH*$ zHg7iBX{iXay*J`G-zI*+?F9Py^lS7|hjx0;t8bE#70=e%9q zgM_WqVC~-;;yA;na3{NzEsayCZBuNBSWh_IdU}+mn-T3qqUoahPhk7TROnu+ zPe1c-0jokC_@3>G-v}hQd2b9QBGQZpC)&_MiC1XWxddwCoI)Jh>j>gvy`UPt28D#( zK>t9I_OQ01t|CuC$|7@KS;l(E(=(v=j9;cB@FCc2EkpZ#kfi-$&Y_T$V}c8Blxd)oc?hhwtYU2IHac%YlT!;>lZ_(%)3TsR`(;xHVxRh z$%a11_26{NbMf0{<=Cj|9IPJthF%%GK%*{c9c%-ETa7;8sU!AN04ohqBs7_ggF5-$fk8GkW%wulzTxE%)U;b zx9z@kqWDVs@Tt3W)7>Ok$-9d_alV#>w6hRlvlbfDIET>l8{9o|C9DnjjNXOo(!XXM zp>L>3+W%LrjFb%$(8~S z2X$L+t>888J%u(r_lDT0DWLqAMYT~6@FiCv`r3&wkViuBr_+J1^r@j6#EtEz&AtpP zCN;p@RC(IrILB!7cn>`bAHeF6ZM>-a#k}K7T6t-72L1QQcDkKbpj~_hDfL+=uvU#2 zYP<55h*TAkkHHzJH?$Y!ZC;O^+0Pu))`N_`JWlc|c*NpI1m_bGrN6p7f(xz|cwX27 zB5_}iEjoXmm}#o8o4Ov8Utjgf`Kn)}ROB=qH74EV-~phKI7k?CqD`YT>Yzg-YRJGkD%hmX&| z^Q{8l^`f(Q?%_s)j!a^+_HL&p&!zZ!FX*_+ePlt`B#*Fg!k<@R)R&nt(S>$!=1^8G|@T7Md9 z`d@*ilND=caE<9DU!kCDx^0i_X}j}WCNs;w0;P1WrxvSu;F}+EVEE)&Sn$IV?;qL< z=iVQOtmT$yG`GgiC1D^0xQ+@Qe7(wjnUgyV1Un`X{Y~``jN94OX7#K%Z(P7{9A3T{&gv=wDGEeq znxnA)6+3YCKSa{jI%A<`gms@Q(>JGdfeDp`#aDPEqx*VPyX66tyWuw3t7pbKB+q88 zTt5rs&yEQ~e0-SBnRD5VGCxQH$7GAxa+T<%DzSCXPq2k^yP1HvODNwm0ZlgS#Eo_0 zD9N&bJoL+EOZ~cuW70Y3`92HY+&m4F|CGUP?#{g3<0t!)ldAYWGG+N1Vmr@_BuQGZ0h{y)Rke+|S+2(# z?7ho+Km3PJ`E4VKCzM%-tG?`dx`8|leFUZw>sVbdXCJTdCABjr+sr-w&W=B~%a%X> zoV->q)1c3 zdF*j?6{Y56E^zES!fOnbWbX!qk+8dasQ~fUL|&dD<^CaL{ePY0z;>kaXcmcUy{Jgpb zURjuA4Gt!Lod0&gmd4=w`d2>qmyn;hEZ0mm=toK4o2tLK-{F|bwRj#v9gG&e6>V6zI zTLai8&jP7)3&4JB3gsWU1BPUa8QLL;^CWJ@kF;}`{t5|N-76g$mWQH@;a$x8j5gF% zvIHL(mi1^kqAK(6*Wps}wV5Bt`D zl;Q_8Xrl{%ekg&Xfeu=BDhL@^3?uJf9-uJe0InAZ^P86?<04rpG&Astty=`{!k{1QB7GB2z-JF1x4bfD&J};C$h-E`u2C-)+|F&nJF{`OE{^QSMzBf82&;;l2Y#y|t7huM<9w(H3+)jQ~d-%P&V zfU}@z-)Q~&vnRpWOc^?4k3kgs3!S$*2vV^lg3X~fc`wOXrqpzU;I_*?T(hHt*eST; zvh&8+^G^YObzcp;Tw}ZB$$v0a0EtQUJ?M6qoPWv^f^Hd^som1han*-4}w}3jA`xh-bmkLT> zvxxMLmx3t`Ib6Q>7k;~2kvV&oV`LZoLko-lpr9!& zFJ)7J&DZ49yqrBw$g-O2Nc1%@PfK3mxnn8Bl*`#qwr)i>56g)iT8TUiT#3~46WH1> zh9?q83p`Eo(Sg7iB$oOXbw2)$PVPC6bQ79edH0y$W^oi{lwLz@ ze+@8ZYMQ7|lA8xUt)k8`SD4_Rv*FLbYkYs-GCRZQ1~Ox=3^RRC7#hnfXU6&-GA*Qq zI=nLwI95FDc)4G2;r$v!o=TG+5-j!OwciQhXQvjI6d8v7WrsJ zb2}2)a8Ls_?YYbOLxQP8YMs>hUkOb?|28Rq*_(PR}}M zNl$k#1=$A-DD}F~?w`}?V^Ig`3}2CYFS)a}0l)sUGyN}|_S~3Ek4_DugE~{_B}F^w z_|>cExF#JkYmB8*m$mShSYJk}4p-4-f3Dx?-GDUboMcKKX9%V!oxpp$8~IinyO8%X z6S(?s4}@w};~%QS)T-M$jGUq)YzRFdWpyqXInI0-1zHvXYy*tCqS8_#3m%pNW6Q|Hm`-{w)Q4c}DFAtoMnT$|_^=@k?EuS}*W%%U-Q1Zp ztBetk|4CLi?jmNVa+vAY_Yvi_I_%UHcfiBp8Q8zK0J{QfaQ<}{?_-YPuoF3gqc$h% zGU8cV^{PnPVc8(qOUc6JSK4qp(hxhWJ56q9U*$`-u3!=zeW`m>&w!$T1dQ~gLYLn{ z_R%IyvZJ<<`bMv(3%WU4*ZTReOjML8-Tlktm44#(kIOKQpU>sMo?-W^vl+?Ae(L3< zHPqhy+}SwH3C;hog(_XTm0G%X1f?blleKSlAQVNV*`tmG~cj zV`n)kDR@U387)?KRFZ8SXV6BUi&SITEZdn^!_m*HFYwmoxj1EWIa#Owkw~4L#5+jmgsYuCOeixR5Q@IGyrNE3rXxdH z3D|j4iw^El{fa(8W1M0tIsl;RI3>MC^P1(8uml|>IfH@L$&(Abo)Gu%@2V;eSuzE#JbVua zzg0uk@PCZqRULH8UzU0|S&>v*U&Fi@T`ZEUKz4nK#}2qqAm`YEx>A3lGBwNu9p_&JU2y- z#JWyFr9P*r^Da@au3{s#UdNuPIUvo9cV?lSf02+SB|(327Ka%YS5d0yW#r|iCRlJk z7R_@ug2J2_yNgwoyvV2q>W$NP#&lIG_7amJ+S~hZeXADv!cW9&uL*FmXdt#)G?xgb zG~w9L4IInrD!F(|YDmJ0|p|_+YITzbEOb9+j-QmBQ91NoO(~-i||4p+g^glLQA7@*(x`8O*obj1Ok;(YGZA%!3vK+?X4IKi!S3Kd5JBJKI22 zU@?7^I-e6mop91b9^ARurh6Btik?TQWBKs@X)!ugq(HB#>VPCMC%A0Yf_m5SP(Y0_ zV>161m_M5a7b#5rSNNTGEcPMw-%eGSet^%&89(8@m03Z(d%P0An*SUt8Qwx#zw<#n z>k3me{+=&qYbU6k-9uN>^uuR);V(=XBo&3Y-IYHg%@%6Wx$W_-)Y@a?H#Z-}O#_H(y@W4D7#VmHp@iUsXk#xzLx6 z=O`7Sr`}4mdZ++mMtb2n`AYfHfep3)RUw}jPbjxw-}v7rjHF__zEp}eNKw#NS9#;p`$3t_dE3;HiY7m zMKC3#12@Ln2#VLw0F$^J=(G2SLo#Mi@B9viy18uY?|gdZ`(fg7?-@}l*}}>N9inyB zROsYbX*#M>1n>UJ;A@kd*zIML1Q#wRFbAbJ^U^*!kx-6zXS2kXRQXEM|HbV@2ezt1 zP0}{k;>deI(6W;aG)y%ZsVQu^-8CWI>qzzySlLH6=*AzwxYq)mNs1;1rd@=IthZ2oUzENx;0tG}Bx!x>4|HjUgH^-| zNdCK)G5mLtWY0cD8qB5H!pbQ0anm?z*NH*X=fx94OAUN`Q!#oJ6oHR>HR6sJGGuH= zEqJY}0g=9)*g&t8T%4aq{1&ZZw;lX}4YvkUBTq%>{^zaGQF|MCkF|mpy9^3u#{h9s zg7W>ZP}-lflyB%as>eQ!x~3h@1Rc37$h{ZJ<{VqXcI{uw{yu8N=r;sV8|<#4Yg0Kd zo!A+Ys>*QhssyQKy0ObZ2&r2<3+cYEqax$;1^SED(V9Dq=*55M(TzJqX}>`|`pk<5 z;OXH;y=oS~{cRt?JwQ4M=y$KnWe!}T5h^1YGAXZ|6T-}C4PT|=my6 zPhblgfx*l6usw7HA`WzdMcH$>yGIBf+J42;T!XM((_Un|`UE*3NW}I6FZ7?>8uUpm z5`RCd$}yj_@k8O=(L-51R4K*0D z=%TL?*d@gvw|5Wlp7}G0)~Xni<03<~+iinDwV8Cj$QXPWC>F%NG)G7B-=pp0)1X+b zk35mvhc8_ehd?9Hw%>tdTJYxagQr1sxLxW=4;T`ayz(d zSPWBFMZpHS95QK9sNKu=bLo%o3c+RWboyzwGu#+$r^4lX$-nIDT(09H%n%r&nXT1= z0{yu-=ddmFmbUbe%&?$D$pi&gNoR33~hDuqQ`YdN-xE&6jsSul0Q20Z^= zFulFOnlA2|!JS#;pk+-hc5nEDT5qp{4=HDLS#!7Nr0&xKd@6P>;j{W9%yLuPnEAtdTmGy*68Ai(2X z;afP9p8zhqohXa0?-VxFwfiuv#QCk1XyXkN==5$yeBCgCv3W;{Z9mG>EFEU9rW?@BcRqrYMH{LRDn!4x$-*Nf4yP|Cl9+f0(vY8wCkk@m z+(14Y9LwQC7UmFCy#UTx=s;?Z3jN_L27z5S*uWk5NN+{Lsavsmp*vYIxe!T(UqC`H zZ_#1XPtoT`DB5SiRnWP*irQn=jO0DNV2R}%kQD4iI{Yj+)3*%jt-R@t-om6V=o8BN zcL3CUb3s9S6$Drp!#QJRx+LHicppha2FIHlj$a%S{PK(uytwN{8|ZN%ERDO!X_Y?h zaB%~@tl=|EQwzg>Uw#S-tkS5uv{+g`I)YZxu%h(`2BGGt6AWFi1$uHXPE@&r?xyj; zGmqnzyfDX~Du>XY=p!gm#S@)OvO+#FVo-NOoibez!Yr<|1Ffh$^u#ronfzc9{moFA zruNFxZ@inR^TvNrjd%@Aclm?Dk7Pl_gILI{ng{WhwlNB}?YPTYjP9&YfklF;aD9^* zU3+c=Z0GJ31)TqBVUP@?YBmYk>4`#supEeL&BQ0asDaGRxzKUA7W&scLJ=MzVEXh6 zjM|COZao)4ecEr34pgEorC-8>#J_O(vp98SRWx(XFPhpGPlN8vO3H8UJL*rIF6SN{mWIXn>YH*Htye~=vW?W+j&6|7o=v}up=r@F zacta@K=83Fp4sGn{-6O(W%tja1RW2lplLSvjTh(QTq{ls|DHpC-(Wzyq&-3oHM!K) z1$LZ+z6+N2M#9SDrGRP`P`lVHru)fd>~qT+52S8n9?3mrAcpGzRJ?)uDs4J;(@MI= z-IEsXQlooAUxMb9bShq|0?8HhftITWeY;MBChxBb>Z__4_V)-=TV@9$&)N~_?4lNR zOoD`d6{s%E$6g@AUZ|PN>i&^oXRe&Z$~h;IE^SldvN4ynYN=&MGKE-)>nm(SJSW+0 zsgrGZo#)QpFL_NC?zE&m53OPwHtl7bCi}3pqtkF+)flre*`)4yyBaOqR>?G7%#GDIa+E zZolo9xTR!IpMdgvzMN_7J3;lfW!wGrc3_U{O*QX$06W10~4s8#(Z%*0_qNP=%;f7>F-91dBw$1 zyiZYYZJkf*(uc}eYvBgcDWXt+4P;vbumIwI#X?fHMR(N2WHUW?WgE9!x405y(6o*$cZ)a>mf7L z=cBXFqIf&rZLsZ_P-h39q_Xm}`q)bgUc;W-S5Xi`c$rH)|L@3Ss?9=-s!-ZMJTJZ> zMtwr0bz==ft;`{_0#B1~Wk>MxH*3hIOa&s<pEH}-Vn0@yKYIG4<*W#L~B(Vv)Lb zCS6IPEahK%J$tC zBfa0FNpt2uY=6TVf;HlB)sUVhpb_KZ`e!$%KDl21$RrlL_6S z3BzZeG4-wzxY#F&v{%@YfQ4rXHaka(UaPUTnZ=}2Y6*Osum^MHPiWT;JA$TML$#v5 z;BoIVzv%oi-n$B7Czn}EG$&-q6wEo#i~pix#WlR3fYoizq)>@a3+2H%2E@M=0jVF}&!DxN-aT$C31r$|3YNrk$&3$SL-eQW;n!*1D~NQ@RNc;C$*zwe2rwF>ZsZwyaFA)S$Y zBZqe?zTz14Tz2QRBKxNN8#$_cfGnI?O%DA?BEO}kuu-l@$)*=cgi3C}<%XhgkMrAH zj=O?a#icV&ZYP=FT|-18OO(CkdYHu3<<Qlb3oBgkdSpX8j+7F=E8PW}Y_ z!GoW38I)VY+qFiL2yZGU0g1JY-~B{9A|68g*U7Uo3h#-ubp?5s9fS`AU1Sob-(+g# zBS_ma9b#IWK#A%z$hokE5^L)f{1VK79M@1-A?*!4p}Ei{Hwwz5nzW`!8vZGCgiH(2 zV0Pn=WKW6>JJ;YdoB3#z$jg=zn?<8!UsEP^irZ}oktvMO(;i$>oriBPR3jtf7=uqc z@{rRY2X~g^`TM40;Ui+q?LbEoTk{Mr2oA)bTc5*L7Z+OeY5;9~Z@1tHmj@~{v8Sj7 zo#gJ9^-$OH3H$vqgQ6vttmM%S@~?I`*|KjdW&?8> zcC#zT?K#PG&)Q%kS$>4?a6+mf<+d#y&YYqbXgSjk|1OX-L+#{}&1<4*Xi5nm(x>#Z zo9x;OWZ5mUuL$$5m@F~PvWbaxVP5Q5gXTm95{ns<_=WE`D#GtM9!lPe0v2Ucmb*^j z*&*}ct?EgrTyvWFUkmdrH-nsiLyqvskNraT*+i(7-4<~gvT@eus_lYyqw$+&mOnAEy;l8U9` z-2cRirb?56)I4Nr4&GswX=dSTg3FBc5zT+oyar(^5a5=*iJmF?giZXb-^({VH`vM=+najG36p`dC7c?PV z!K?_1rvDVH*HCMqP3xfRw=^(u(L zf$h$yM7|acYhOWk?HDA`Kh2~J+fnIb3e3T6N0r z1vX3LHjh?H+2tV~TCK z7{@A2mx;p*30Pk!Nj5st%+T0*!8g?m%0<%_iOJhz`ESoqsc8zD?O%;O>yt?9Ujwph z^#H!;HjhXfULa4~PNSQhdeFDn2s+OIGk^F8hG#$|4$;9bHKK}=o}r7^Y?#G)k_uSON#WcxBfXmACJ zw3>>l=IArURj0`C$64(4+FIguRE}s08G$@EAFEBVyiF2nrs=R!T0=vhV?| zQ=LQ)dW+B_PkX>ahwFn(X+cxN))MBM2sIvv1nH zTFVzOqy7KqO_D=(cG&T1I_tRdARF=Cf%{uAeEgOOF&7vTlg_=gr^iL;9f37K zzYS>3`6~4E`#XU7F$)*(qR93?Q%T;k2aNp724thK1HZG&fvWO;Sg0QiC_fcn*&#tz zDa)YCk6S=eQHY)scLSEFwF!<-*F;QqJ{}OR;6JL)rb=SX6 zzlz5CFj`<@+Vow&^O}n=O*O=`j&14LH5Q$~R6Ko*;wIf_rOeMOO0hD25 z7YuWpYCi*2^4}3g z(kTKo*-3(Rou|TDCDn*t-O%`F?8O4HT`iMZ||tI_f{zl)xGEQIiXT0N|Ml5!(UD(!nd?}G-3C?T?ELLu2QzUPPg58Qhn_rA~j{dzs01-Ve)s0KeD8?gWN-2}_D zQ)K7cgH(CLkJc4a+Q?5|5BMN0&iTpbQrA|-ldSXlFjY4ZCe4q6KT)yZ5q1|HSfY!G z&{N0mA!$}2WfT%lWP|nJJuu>x04cf}V9NCxi`JF^bL1`bulfv6zuQ6o?O6CDGoKn7 zAn0$aT1)r!oDE8Gkzh70$hDE1+Zp`5+bS3XeI?cZzyB7B@=A&Xa5K z$+cNnO<0dD&v-$ETy1f|?j$@v+!~Lbl0jK@ef$~!90gk}?lA?03{~NqDhTWLC2b#U znRmwV$am~7^4gtB<_t9oV#;-qeuEFXcT0qJJbM_M&kllG*DEB$um-zwL6yMO_p#EI zDWLPH2RR>;g1XO(`DTGqNP0pNVh4_slA*0s@BMHhE)oMNQavPPK?YUj&ttCMp~xmx zJIo*kN8FR4!`vIFajrY`pZr5qpuGhrw(Vxv=ux7sv7Cw=dQN>`znJm4EXPs?RwPom zo{VkIqm=e-#%4Qo=zbMSz?)dd+>)J(4y3i9b(Sk}pUh&S=OKyyT(=;Lde7nD&O5~X zg%nzOU59?V$(GBFoq(XltI3qT>A=Ezrc(O|UoN$R-?Bc1baA)#9!F`Mb&-OY_)w5{ z7ZYfY&*%MU2x%<|*^YM{rs&NvbmpkEw%fTzTtLx+P+@UB-cNr99W z?7pf5cNR#(GNVjliU`@K?}^zp^F+Mks*2Bqc}5%NUmGDy6z=TkIcc(-*X&@vrFhLA;)N6m#3{Z&W`+O=d-P~-bvWj zbp^e&dl<`}KfwGQ-vYVY;>pS_wrs+>J~;E|5e!(D(rQgB>85GiE0<#^?Kj4)h7IJg zrx5rzEpLf_oBBCkIK{ESJh$hn7?>NyCU4}awf~E)y%D-Bc!|K5vu%mk-Bj_u%*suucN({3EzrM zr(LtmXq^~@J)i9nT$Hw_f~1b3PhkhqFIp5GlAVSwiDyx--}#X0w-p4-6(LZ4g?k{H zUMJ5L2bR}TkJi)t*;U7&LFY7SJ2xM_5}JZHt;Z06Jc-Ai`4Gl+$UQ&3hj|B_v6+V` z-Ya_?e?7M!ht4d)t+5R_uaL!^;^L@pdk%ADvotkJ!>(no$qW9HYmUsG zH?sx2<_@Z7&mO8tWC&e4{|Z+)D$x-|Gm+z$lJ zK-`-GmGTTha-2VZNpc1?ddUITRQSTOjrG8$j&R)V+xT0-GAwBN4@XYKlxz>${f8QD8+8fyC@Rv6W;oE&&xY{tteIGA=NQga6-QTt zxeP^y0-eJ3F?4n|^R`|~=C9u$gW|??X{qTp^!}s*`r775&V$>HZl-YFok>U0rUyrv z^om*JPW~=D`?LY3_3O|z?uIc%vu z<2#K2W*+f^8#?uj+!__KZNqweG=3`gz1P66Kb=D9gAcG@=VsblT%V>kRKX$xnl+lT z1(FgJvG$>#%=&*5=!*4O#{WAoA@+f-E(iPtWG>Cf$)9Y!vYulrUmrwbZ&u=6CC6|` zS~9*+KO1WwXK;u6K{|hb3!T4~b1>fz#6SHbG2vMG1}ncISrIvWsip(fec*h!pM}u$ z7n=0kh;P`LnBo>uIo30A4_vuF5Afi8_Uw0kw#rqO&GB<5$qS>Yfpavex;6#$kIS&m zR!MMF=Q$S6Mfm&fuOQ#1$L{^E#9mrw1G|^b0;A%s5E`RO_GTlLE$Yjd@_NZlNdwXw z(@7=&br9G%))TF^>yT$6#7=kgVvEHqLGi;mRGc-R=pH*ktUt&?*R)jj)QJ?ftGXR%ZX#2b}Na0?c}Jj9hd{huoN4QnsZUJUFJ9;bJX<>ihnUh_u9B;+!Cm=Ji+hieg)F$J@D&rIwn>}nINSYs^6OP(cGI)%=)xp#oZ=& zm7YL$EVBUSq#926asUq(h|p2$r;c2^(hKK5S+WuOM%3{oW2hl*HE{?`VFMO9vffz=toP?JYGg*8 zedR)yEK2;u{84nrqLKFW6bn7}wSoaFv-kp37ye`3wWSjhvxU&PiFmu}1td4M746KU zK?YXB^|}|R+HDX@jvKI7jIG$;F`+P{QIQRl385;*_>>q3gJ0}0xfD8#$`79+n^k?G zXp%DM)$Sk)*&$p%VjFbtcZb~6!_e{cALS{Ln|Rx zcq%-*umSvvjG@=kmL#vR!*l<;39I@P*hnc+Rxjo+-2Qq84S%+yjXT%k36}z*^yxMT zWD-eIsutEvRY!{kCxeI&VW`Mo_^rP_U0zd!BTfzA(eoV`mq%covSaY?3dc)HPC<3q zH<<&bMPwfO%JoVQwO;7@(emHI0LLcv<4od#t$6niIplQG7-!wr=5nklDEMxapwgh5 zG4k2RFD+XLvGfYcb9^sYtrJd;b#yYKyFB4(jT{yUoJ2HV&JtKTY2i(OPvhXh z>oEP0ECZL{G0nw7X!z&~h_VSIY1@+VCtiwRoxcpmBK4qkf(IiNYJ{L5eIT<>0#11lXb*0-NsBkoM>awYSa! z9)}l_n66i3or1t|k7^J<;?6#*E9o>*U9yp6jYyCwE1gh_?n6|%*%|%bn}jW5rsDXE z_K5d&3DPgi1Uc1A$ode6z9ve6zC}B%C@v(0Iqvw^nOLlO*#)9}VnM&=5oykf5m277 zt-98)nvaLXqHFW(iM!f0lv_K9zx0XF7XOK0#}sc$dlg|GZ}#K6e(k1e4Bnye+r`*G z|k>5ZYa-jl0e~#NUK3V6lull=6~`WMQ^|wAXc^b-Azb zm$mYAyQ)65!*eTh#{3Bix!#6VjJdqfnP$${a)A=xw2tW-OG78xLg+#RQF^>N2!}f7 z;$N}Lah?=m3Xkr_LKE}x<2gN4>`F?oKEas(By$?Cx7OT2?7+(zsgYGJq|=z!tI*DS ze&ZPLX9-+6XU!_eZ^5`VzHv$_)QA)T*f+;hO z!h!2@&_VO9h_@vRX~?FcO}iS@k5vNP5o0Rz4Q)k|JDFM+~4T&#%xkCN8Bqx&eQuX~uz zt(NpxuHQPKD2J9BTt^C}@np!=0wf-NYwbVsqE&I7Xv^24GmQP-5~{VlmB|%~r=C^2 zQadS2@_T18Zmvth`86SU{T(sf_g4%za=Ai(&aU+Ojtt)P_7jrjzd_tQjIDcm(Qz&_ zoAvTDf8*D1wDSVsnUY<2gPa20Si2JM%~*n?3^ajRw19k6h=P1mF?ROVE?jcL90jm9 zz{_hgI&#Vyg0nJ72|PkxMfNb7F9oMu+KCB66ZEW?3RMMiWLyFApGLrsxoeqEzY3UQ zu2Vbq;tW6K@-ga&^e(DG#sQtu8ABK6R-pn}Q7mzO8h*roi|qF(Qp)<%nX(rx{G`B# z*jP0ld3(3Pt%&JxFSicFc61Yk6&7Um?M`Cv84877`|-3-&xuH8C!s|ilhmi$xUtur z98l^M?CrLOGj~#9`B4^atCPp6%IA>RK`~VMX#uQz@r1PBRA}aZE~cK;?WIzt=@BSZ zCYl!Ia7)C1Iqn8zl2IE0)mXHAmK}(dtFjHVlu1)v7Lh;IgL|*qZ0?b-a~anz$hNZ%#4N`Z$?P5t~z*kEK63HMB9%;9`?F}dsyjw; zg?O_2Ur;)I5)0p7MaC7^!U6sQFrAkM|NLrU|KH8Xed;}Ocf~k*^ID(!l=&L18$Lnp z3~a<1j&tz2;3njByAqj)-lo!RB;jV5J#_+(Q`%N)Oq=yJY&z(QeY4D&GSe`8zJDgI zv&Wp)+BF&VaQu&%9=`=T&iiS-9%1_PSy_78&RR76SRk64I8Hu$m?2(r5}3Y`VDAoW zr`=3=^p)kpv{m0+>S_&xRD&X6&fKf$mhoazd_sg>9khl_9sf+6V)&#%Hik^$GzGqvAP2L0|uY=h6&0)|m zy#_2Q^;y*rrB*2^gH}^&u=VZlczk|G8m9IlTK4HuJxqgz#9%T4Zmd1F?F&A7r*MZpO=)l2|{p{3b#PS;!H$O3JWp*&Z=fyDI6io~_iz1`Db12Im4XV>V0L^Hg zjNdrM;>}*ynes*Q?AQH0DA1*l^h_QmiaDnRH=asS%CD4Lzbki8bH(FOv$_$UCRm4@ zJmvAF;{o{7vu2>ALhz9Bdu|sj!w!r-LzDYk(T(bMuFsHwL$;s6wHGakwR9x8l{%lg zA5lb1MrN_H)B)Iib{Aq?IKKL}ZTR5!cvR{B4yj+-O3D)@X#*({`fpPge$=keYQ0CD z-{V;*X#VKS7E9{Gf{ml(1IIDWon4IuTfb88hsDX?07vROjBEpF#WY)TUqF z<8EJ%%TT4_X*{?z1G|}4;k$kZxy)t*Rvyj;HP>VxTsFcd7p+6J8Q`Q z@rE4EYy#Im8>j&OPv;6PB?<8WLuQq7=lvgEx^kjyeGWm&8VQWo(Vem4dj37K}+B*^!s%XwX@`z zgQ`w0Kh#)>7&UA}e{@Eun1M+2@!Cp!xX}Q2ZqWwaijSbBG6&>mSivtpMev?B9oil5 zkb%=Gl+vwWCQxS%`$CwEP_Y!14{;oD@cDaf?>(z;x2Vw^HjR~I2tY>QBqu{8A zO3Ud23=emHVBnk%HV@s)=wJNDKW~46$kfjQVT-S@{!27l@H3XUH!dPyUmGGR3nfDS z`7%$2RruW-tDt;t5?o9*U{yLJaD&=yq(1h8bib=Y5PYBBm1E=?;a& zyron?Lom4*w+l9_*pi=4iGpeIOPMo!I^lP8CE@2@2aW^?Vx?2qGMiJd|8qLne?+jo zU5-7)e8d{TA5lbmAhOgx$WwEzW0uLxqGD7O(VLKFqRxDwN)9f8?e+_(81pM|=;H?R zZuS*)U7&}DzguD#8+Y=j=MC)ZsAlRSa==VI79w%0*g9J<{5r2<(B*KWo&%_9{;ovzUqO8Cc8BM@r z`WNiU;1L_ci*TjYl3H=(5G+5&vF5h%*y!U~xG;J$w!g6m?c-bwF(uDItFRECf3=Fw z<5ZUUf1O&J?X~##BP$@Z)Q_0{deJeP3p*!D@6q1Bx4q!6y4TP>60kuO5Ex7cGdZol8_hJXYp*W1K zkzLB(={-yQwwhsu0Uvn1_cW|kGi1&GL?eOp21xL)CpX{j#_M9Lktf%G=u|VnLd-K> zE_oXm+!f`qi#DsU*C$UQ7x@~L+P)vWlwEgLEOF{)GyKmrIQJ4i{a*s zMTrak%rH-7c8lb zL-8<4}Cro0j&x4sbn@>eIRj3!!eYE_F};kVX3>&(!4O-barM+<8?zQJLs3Ie;J0fEa# z7k&(3m~(EyO!{PTw4ckZ2mAU_i(a3hqWJmvBprvJi<#osKF#L#HxFhw^V6Emo_UhO z_gc(e2NgkMK{!&BUV#f;ZjzC--Kg${G?p#;2cljotm>-CY^>KoC|WTSf6J7hv~}_s zH>{272a{lCODW~*9E9Ai|Kr;;)5!LLpRFq|jJAm0k`SCel_%Jf;6v42xX(;GQGlPQ z9z!~jV(fn}MNqO|YU`cwI%fGKj;+yXh{Kb8!TP)cmhJSQM#m1r(JnW^`6cR1$b2Je zt$8dfyS)f@1jK@j6_=quxs!b1nApMn3-Pb2R1#aVl@u9<(qU^9=_%vu(9DsAXlR)s zPR^01YSwpP-a#?WS1^pVHB_j1Ef;WCyA!RoR0*r+oWb{1oN07V5ijeZaqot7ys7Rx zb?T-$4;Al1x`Hp1j@e&E=S>@P=SDVukZFYGhRQ)nqcvG-p8$b&kI3=|W5_jm4_RXn z4|P_>NW6YIV>yWMxpGq{-{UuE6A4RP@^(EfxSmZ*Mp@7eBJl{N2vHv*Z!vb|UO4hx zJk+LV!Qba|;F)VZBQ4_&f4h#5>W3CYhiPKgWe#JcEJLq0xPhx)zF@-6SfV=B0#v=} z3Kcy+nXy=XiNCP8fC)PyM8X&SM2ZUIbnm(I^d^rQJh~x?{^G4e`^_uI$4@0Qe*QYV zn$${W+p}r(_MN8mgHt(pHmyq!9@nC;c}s#^!CI^`n2HTG@1uy3HRSUxEo5sWiUmQB z@fLCd-BTDK4qaE7w|CpnhT$Tts+&VEF}#d(qn+7qhx?@SP%Yz}?~kmONIFePQE@7= zDx*7iEWLNDq?6q#Q>V}H81MX;!My*s3C`p$ro)GW=|;aMT4wa0pl+QJ`FJJ(o7ev3 zFS;g*d|EkX<(;);?2Hy3g~M36Xga;P-I?AyyNYhyn?l}fO2Ve8HuT#par`i*9J%~< z$J4Zq(hrR*X$i>$`t_+E!9fjC;<7sm331(xu1nB6cL%RQT_G7>}BHv~(OVg*44uwQ$?|4RTttZOTSwp}6u z*>Brnb>$-P9{7iN`=|1L_w7UhKgK9FK$q{k^AK-GYZ^+eoq`?-onS`3A4N(tqS5Kx z4b=S&gVglOLcwurKmTt=6?CnUXQwoqvsXVUfo<0<((fV;E#G4q*_IX5AgjvLcK^*= zAa6n4;hyz$B6}fc-!^0$Y(O0L>0|Q?H}G85apw1f3^wZaG}iP@C4|?eVR`;+yd%XM zb%bLayV#X#7>r`p*j0kiZDIDF?q6cLv<0>_ErG~;-^sLNy7<235nOkH!W~L4k%fCQ zsq2hHvxz5RZN>mSvt4ouvzmNI6qUT^8l!cbmNYls6lmr(vv;^)l6o1Wv zb0q9Jix!H!hYh7ANa2nr%H{e8Gfw@$4-=N-QbiXy88H`cymSi5u6Dx8vQLq-F@=L) zuE+h>$4TnM3&D>IUd&Q<7G}3+quTj2>3HdinOc|eytNuCTdX&hBw=G7K`-afl5DRP@6%P$o-(Z97 zMaa~QKz(0RDBB@devFtdHttM7hhOWGgyW~d{NxR4El&my(=~Y4+hSDKbeQ?@sT(SW z-td2^hvC^f&7kV9I4pmi51yCRS?w2wY;MVX_IcDYh|7OP4(RZ)wvsb5^>PW-v`Lh| zC2T8hs&pbuY77otCdMAr&xH5*vOrpMI%zQ!C$CF}d3(I=NLs-RHtMZATlfAwzI7`X z+xmJ5TEs=D(M)eNt(4C{WU!d^>E8`|D)u1_RSMAUOgJ@gnuPUvAZhD1wB&(5wpro> z(U%{P6ro1=eAF6KejBkc<$$*179{aR7agaZ@U+isafbH`SUIZ^Vjj6Nrve=zZDA)U zT|AA$-k!w|eJ^2A9Tg^QeJKW`G@Li4)O`N!nU>{3s*aB$MsZ-2IK3@A18;KD#;I3- z<7C32c9%3eByGr=p(@kpSS3G+ow#-rzJ|YnKgxTk&8J-m+IpF<^P3N&?;kRM|4oC< zN?q7IL7cv4hGcW<$! zpDZ<{)5Cc{y%WZ!d3Tu1`y&vzDGCkWe1ty03>M?rN2VcO zjVa^ql1&24!kBpH30!hf1COvJ= zHDw0#TGEY*1Ww(bNKC#d`n7ivYvPu|HU($08_(+?S?>~_X~k%(vyu=C9ieQA&}BAm z-wb;E-898pM-d3qnZsk3{&GS_Y?fCql}kPA(HS2#*dtg>c4}eY7Ky9Xip(W-Fd! z(90`uQVGCaWg+GVrr^FeUg)dnRqLlh?@gJUWWgal&WR^CgTeIYK- zex6>4SDjnTuSk;@eBWtHt#(vFsiex$X~!u1XGF1`ks_?hIVH0FSPn|>o{pqNsz`y= zHRcpAikQ#KfO+}H$jY(RM5#ZL`QgaTwfki8@Psux|7t%BPVT^q3_{uU)wdwb7SRXI zelt@~|8043dV@e}(?R@S$o|$f|9KJDcLikT`zO@LH=FQbE|WRJn({@`#9OsFe(Cw@ zFw#~XMcSNF+5IoHSec$WGHVV6@?%`rZE!kC_&tT_n4N~v`Ux0O)nStpuECqI1-N5F z9Z6qo4S}`7@Z5`KKc}B&FVEf0Mym$H-w#~=>?;jFY$h1WZz;qjTEU#1JP7(|502mF z!r9%2psimIhIOULl6T_l)9DxC>!azQKIuJjz8sG}(#xS<^c|AjWP$41~wRf>EsIR-joRL`zt4aVC$vvayR^33JNf}6KlLV@bJx=~A`@)^tWT?bP zq2DYSat2ocTUkavb#yRRlEtm&rAf^6%2i~Mv^Xf*og|{yGa*t+ibdVs;OolAVH%cr zgESx8FVF(M(-gR)eVyofWI(=?Bg|IY52BNNAe)zrEL=Y$)Az}+_G}yFKEBFv<6U3o zv$!}@sGvy3^zK2gPzSUNzlD_?pV#;KJc#*iL)G`>5i`mku=FTXzDbqBE!;k(>;m-c znhRH(myvfeswCm|dSdkTCHZ*1n*5s@B?wxm%*)HcBt1VE9A0&hJ}y6-GZ+D(Ez7A3 zdWF1sKege%4`c9LZV=vxn?ooyPR3MS1$Nn`aMSJv_&(kM_f`8_--U^CbL9~@R&Nvg>C*)6eusFeS);td$YsU zISki_Zb2T$_Y@RS>=tKRb`m-ZR>N=M)Xo$z`ZfbX`?rA2YHci9RDjh3UZCKUS-j15 z$C>0q(h%{&iwJ8AnA(^ff^(;4FkM+iTJ^SI?mSiyv$KrGdxbAx|w?l7Fd2a7HGCFX6op zKb&KW7hSl>@rjOO+0tM!7TMso?zHoz=WQmz@;7ixqaxWvh8Y=MO1f zVI@n7TW5UNqmDMTQ7g;@tk(?1^E0i#v5xl|*py?(+4ZSP?3Lx}{3+`~C~*``4Zdw4 z+p}Z9%tDeCwbX|^oxQktpcq26Xb1_=WaUZ{*~?0LxO&PQ!CKp$utaa6PUS(T8JB3a*uZ}w=Y!vBxR6eRIFOLsPIPy3yu6z~_+GjKa>f26 zwa`+AePQyB*(&>$k=#C&`FtXq75WeW8J};Eg%L(%PJ0oVa>k9R zzZC;zW0#@4t_8AclLfQai{M_9W9Y*l8O}*PjErYY#<~j;_0CKZZg9LR-UToA`uV3^ zzbO(AD43A9-LqROxmoRILAQK+6XOARnVf<<@nN@9O9H7O+wQ%*rI= z`%0|}E^Il(Xudkj#%N};^H+U=j<7^FjqGKwzS0AW{!n;%%nv`?B9CWpD?zkKCht{# z1+(LV9udmsp);W~aP&S))}cR`6}_v;=6WtA(yoo<@NgGj>}Do=Mb?^i{-#VBeGtdz zuN@^XZT6FdMK{TM#i@{dKaMO9N}%?wyv0wm+{9-8NXFg%@{HLf5lHv^PI+~_rripIjg@ z2E~jL$9O3fX+*EIE%0)V1)28iA2HHbK|@Bd9P3&Sc{ZNq6*x3f<)_M-+M!mo{QO=t z(9wqG21ZdrZZRMwvlK1VZA6P*bg|PGL-^%08Tnn7qL==cfvvUN>D9iQ>B;pv^uM|s zn&}Orr=KZgp3T=rS2X6~arZN*$|9Zmy7v>=5kC#@bNY$xWBy?kMF-d~sD?=?W5ir% zDNawU!{@ns&5vCi|Gmou>FPRKOg11B2~aM_zg#-!K|6^|mUTYeiKc7BX9{)*8h^55{B!p`k58F)|SJ+ZYPT-f@Cmr`6_nW7)n>#AEL$n z%_6yUyKP7b4Jj~UdnGw(^&)rVC7@BMhE_$a>o%nOv#lVMw+1XQ~JqIR9R$q(D$ zNmd2r0saQ~*V`#1BjptY;V7_ZO9hkZuHdn?jLHjsPLz);LHT`-PjVC>ekr#bDn{sI z={zEHH;H|$f16z-Gt9oKoWs^tXu+t^UGmkc7fn?;$gCMu#tkh+C~(bD)U&98()>87 zSyR5wVR&C9|8KN4O8Kb9Tl9O9U~FIy)E?bHc5hxpfwC%Ee9sWI+qdwtZe_3`mNQuU zPoC_9e;e5xX~aJN_=r(numZMHGN@x?A}N&2WbkuY###+kwAKWH5h|_HvAWLL%OMAuiKWs3ks0 zjFS6N#=%{iWJ%vg=`Z;B#0if1k(NLSg%0rxGlXGH<6_DtU!1paS1tdD{`BTE(o@mF z#Xr$u(+aHcFbQ6v zo1|*^ks1q`C;x5;o(8^RT2+_f6>5vnuBvBzb8A$rYQeZQWwT_~@yj@rCv@9*f3pMdF|nBP>{22dm+oqvR_j4zhsLS>N{0ngldMsB_BAju z-U$%{^31IH?&!X6G1BlbK*zpSvy(a^+3l~L*;$28;n&`0kZ+j;uh;&E*X!*>^wdqD z!m)?n?^i*O`s2wJD--N|w+#jdRzdu6H%9Joj$n;^B4w76503>d><`f-kk2i}&!3N? zZMyPoeTW@S_jyM=wEw`;D;rtwpF`02@gW>tVBf6pKBKm`#H3}-AyxLaz61DiK88F0 z=D^fjlUYr}aQ30~Bf|9pm?h?Ne4))nI7`V9zdpDZn*Nl7qM9w-*CveIP#Me}y?{;} zOouysdCas=P@c72IADxIQa++o*-L3igc|0}e+OIDi>C0?mYiXhBUvV4tXg2=;z4yN z^^c?B-i2_UA9Y zzpW_y@({)`{?6!EW>15z>kXb|7vispXh*~57zdXYFqRMGQ2o|w()0oZS~I_r-|_!Q zaFGv*f58|0xpayt59>mgljboKs%=DKrvUX%No<;7UD-N)ii$w!qd05D{oj_IZ9(dW zTS!AgJy{8Ns6=};)~oLb+ahhr&eRSDg(eL)+|rx%-!_#kSUidOV0)ZFJK9*&^e{F+ zdk)*X$PkXM?4q2OxiHO1rM8P-ENivcC@Sb(El=*;EN6PQ9KkZ-n`zha1$0ARJyS71 zmmmA}El+1G0V#gFz}Sc_AhL~a^w9fduw&Sq@wuOiHV&Ed?;LY%Rl8ly8`_uSe011x&8#VYKb97!)t5hjaRu zkXw^5tZOZTE4UI*%74q)JQQNfZXv8k72GhIP#$@^AY?`9@S!;zt8ET=2^mKT)8m|7d$Q^mvt2JiL8To z^Y%b8+Ao7j4ZLO15N!N+7gll}9AkGryXVU)T(vU` z|L!}8uiI3U6{3s5*Si3IIM%V}9=EXG4vX1n6LXTcq?~xl)!^k;!c6W?QRe5SEas|H z6tnpFI&#v^3!fTRq360(;Aq_c|<5V27l{GhlC~ZY-~;rOs%_ubOTBRjuJFZY&OUGMfG@J-WxoW zBtx&tj)&V53y_lOPc&HI2RcVCAw0MnHAGG@&w>M?t1*p;1PhRg{tA3`i8<~Xh{eKt z=Aeabifp}=3$8-k=j}!y_KgZ_b=D5+^uCAoS%r{jRu8W}g&@__`FQt;2OM$=Aa_x) z<|iLtu-k`Q4PrspgiqDz%i=el>%cFwmGM}5fNV2ANOg|;vEuI+urDM+S-XNz#H5tHO{B*-Mj$bRh=OFuPQ`G$)oYPtI5$`^Hw+e zP|8S(BF|4fB&R7EHm6UO{p5gHy;-s#B$%MI__An=hYk`mK8e(Jog(|P=JP*{Jr+Ej zc+;e|SPq<&{PB|!BmB;E5*yoK$3_R`z$1sp%%|;3h}F`~{0HYa5A<1S*y~ye7cw2; z*AX??*Z%;0v-UuZayQ8$=S5J{X2fc@PG|3tGaOIN-yNt48 zpR`M`OXZVcV)r>_;deexSG`2_q|YUaSA7tj_=W#@*HNa|FO%~0>VhJz^<-WKw`+9% zK^@D>fo_l0WNmQ)did-jnQEO*#9VfOE+q^x2F^tAdLyWLJF=%gC$iV>o@10dm6)qv z?U`B!y@R$vz|12A&bBu6$C@%R+=14E!zjdQym$L=_1Su z$EEz&o64wa%k6?kT(`)zSC4XfD1_Cv8j!l9kzzQNes=>r>% zCG=p@FaG1s1Nhgfg#xUQ$1px;d5Ov8jER^Cs?2&vcEA4%9%WAGbGHIrkaC;sFFZ;9 zJw1n>G)BV7JFih}2}AT6IoHI2YZvsY-&s0o{BcnkF_O3?EH^dVM_>mh7UgC(J|MBzp=nWJI|HO_i4&Fd)XZ;OCz=%Utg zw+hQMrPwgsm^_%(K~=jY5vOETvi;`>b7f48J)Gjh3e&r}Ysf#ei|nMN{pwKnQEBRm z z-!&q8(;GKPNt|p{L$RTAiSG+rwe1;S|>I zn<~4UeghlZrP*&~aZsVAM3CP?csk%iOg_Bije1XaTvfea(El!#@n1ayYko08mB%mO zr`ite(8L(*-WUK2R`&{AkH|0*MQ$V{rU5MdiddJyM7D;FBUV)bl#TTrf{mSFpM)wt zXY7G8IRC-;rrWqXTbs`8G^9<*HXMB{9?7fjXgxObCgb@xm~Ut@m2&$eN5zI5pms;Q z(7X29(a7lqK3`UVYp4NIk?IE1BHHnU64zxc1NkH-eM)U)NJmQ>gg7@_nkOybWN$*Ss+OaH}ET)Ev z3>PuCd*sp7nnO(I@I~gsk~ln9>;?13P@jF~*wbQNrOVh(dq$RQJjCeZv()}TX=vA* zL}IU|lVOi|vTmvm9t^#XpLxgPvcVqG>?}sl)}2Y4vjHfHwjmQ&&r^nVLbTjSBeuRH zL*GAZ%C3`8WrYlW@VnINTcWwuz{$#JqWNkqw3<91=0{Y~$B(P2T_^TqsZBxPwm}W@ zyHlY*trFJldye(+b!<4MOt%+2$1%cf_-5xu+Gki9E}WYIv(vgs?h?e_IjhSqnS2QB zA9J@t-}@lGXFuC7vdCcR?Q94y|eNxtjL-K*{eHTU#j-kj6Kl-u`A^t0%Gx;NF(xE8Yt$FbFOZnhV1 zz>go5qmgOLkm+)kQmk`*bOw9=<~bdiTiJg)OlCk|UIcr4YexLVZm5&TRhffOIs3aC=ZX)5_(+ zE@l@a(Rn%iJpN7e%4Qnkm*|kYB09LmAPYAyQ=>&KP3U~Ra5(Tf0gPP=p=nw*l6D!Q#K4ne>DP>QApKYw({G`ur@HdfIdgzKL>Plbe^g%$Ecj z&^|?lIu*1!K0JsX_gli)m+44IYzFK8b(omlGe=y@(hIISpQXK4g1Z7+k!71hp z{_E4mFB5%*d>kvtzjrGjMPeIdsArPC5A`rdsF%cFoQI1IRB?i;Brd8(tW=F-gWP9t?ym;to8hH{yBS}=Xv&fzu&K6znjyNE&hDUv2d;v zoAG9eg8AcOF>~hs3hP{I%yaI$SyX3tUfD8QvsL%{YOB+ z-|Bznuyt!zAvdtV|6Kp4Ihg*x%>iU;GJBTM|8@?7|1-GCE73E8+`#{H{htP>|L4Kk zbYo*?HhY%I|8{WhcK!b{zL_cS|Lgb$A{9`7bPDzo$MEqEpOL{5Hdka8d-b#khvQD; z)qyNnY$(Qxub!A`WrbGRdI;SUWsdsqv2$z>&++~n-lgj2yrmICxc0db4zpWfrBsN! z$w|1odo83IA~E)U8g|IkksR7U#ORme=;fx6=z zm=*2^isfU&C41a6w#419G7z4BkItG@Qg=lPMMtFLx5h!NE{Mh~LsfY6O#(kq4=%x~ z?6mc0*0T5m;ulBam;Pa7iiY6((`7hY?uGK3-0#cJV9IkMnXqp>vp-P{q1#osT%3ag zvmK$DXogk0w;;}58_VC0gG}sF#5h)=*}D#wLJ4Tl)xnEib(C$%eW6smO(O%+)lBeb{E!~6Yiej`p zybInB7wlZU2^IQtU^ur1yH&?C`D|5|SR=yW)~|7UTNt`O?c~R)9pF3Jhw@|f{n0RP z83GsSLScD3a(mlRCHe%NPPsHY-5B#W+v8H20E(o9aK}~*ZSMy#b6^mI+xLKz_o8AT z3*I?l_-UmK`{$!k?bbt+3-wr#@&L?gZb7JWHy-{xh{f^iQ1rzPLjk%tC_El^kIun5 zz8ukZzIX(A7Ahdi);L~;=7ngs{ZBB+5$k@y`Po+3Uv`DAk||U?&mwE?1?>M&f{l}p!p=Ysmo*iz?uQ)vS9>2nXDo-} z$8~JKhbfyPCC${f#?ri-MWkJqNn-nQm~&ty^7M7let#HPpdC+$OOs5Si4#gT4Qg!If5d zbQ3=CVjxhF4iDRUdVKE*T~=66p9SBedgL9h`ZaThor$45h>+4|jdaRGnq)p4_Q%8L z)I_XJYR8D0Jj-}K21(WO_-7D8k#{=DOR$q}wO^;DZGI>Z%*Ev6-q5pefKPD|#nJbufo4QhQf+k^@vINiia&?(w`l=BrVGLIBKQKo&H4Lm z_4%$6S(sp$4bgYbU`Ej_s^&2BJiV1g)X!iC;*{AH-#&aiZv!KFdvv^gO(){DSc{Yd ztFr8YIA5DKPnb!i)`L6+mr&NXI)GhuT*@X!&w_%pHF6ZkVYOQ>3fC5bUtNmGGdHlr zwhG6Zyl}E{9FBU*VdgG1I43LO>Bw)|zDf!es>N_NDu#7#9OhlQ2nF*T1k5Q$no}5N zi1{LHk`HuNeaGqW>zKMQA5KHMK#KrOdq2_Kr}AuK9}|CV$J2t9xYN4~Yo8TkxZ*OTs(rBMX&!uXt|7`Z85{Q-vn4jBtnueK zrZkqrA(}Or-)m*2dL$P|kOC#;b#S%uL`w92sQ8B<)<*-AKPthsOdRJ9|H9Npd6rwO z!hSDYfK6H};N7|q-Wf0X`?Rj}XHp?QQtBG)cxkX-vJE^bQ#R}KB)00c6jS~(3;q5p zaP46*$~DJ9W0D*mrHI4m<~=M8e}Wj#t5Dckg1Y@&jJ+Qr){ayrudcEYSzksDY zbYKJZlh~oX{b*nI6{|DDal-KozV>8rXN-6tv$!7FvX!_l`xstk{rFn=7lYHU!Mje0 zFRs(aHm7B-+!k^OFk->yNU8+Kp`Ez!X-aah#Ryi)0O!;p~F|!G7{YhK1*rG|4cB+DPcaQI97x^@fBSv*#LPMm#zbPu5%+cb}bW5T!% zm=1!vVThtLdvtn@l(G>p3;R$?g4-tNeiWDm!7H2X6%IvIQ7507# zfu+hgG|V+(V!1{vzSV@?6O+eausr07vnh;Q0^8+{*yBm&jHk}EEe%B$k1}F+q67?MdwOFg^Vw zkH+V_S>mB3%-qO~rA~8T;hm;zhN~<)CU_S9qhs;8I}-i-=CklpC$`yl85927g_U!- z-+mJayY1WIzcLsHK8ImQ=Qb*1FT-(r66U}9g>8Z$oeiCH3b-e`4KUC~znC`WkDkM(HCeN9 zyXLTJmW9_=F}UZw2fCG_P!Af7*Aa(kO}QL84a6YlFdpeA2y!lfOwKBd+OCdWCIVQU z+edEPGbyqQN7H#%oR{7Ur&ob!bv!yOq_IV80-Rj6@I+M^{{nx~+oD7a-$=ksyBwTP)501jLp**q1vk}?^3y_m z_=oOqN3P&vKq9uyG@!*w`RH)UOY~WQ!2& z_!Er{(aD&5=rjWSJ<+qz9#x4-_$wO<^VKKNZI+C;(&yPDg)DZIm&|_66@zS-3{GEt zO#=Rtn7F?d`xmaq<_%3`PUSl6s);W9uxKB~ygZ7mT}zRGTPSn9g$ZL{p-k^F4LY^c znhGH}{SCtb(I5!&c0gTGl(m_6W5PLKH157a2h+1Dgr);>;#u0wQKBq9}BL zFk=sH?c4(6F{{z-kpkuG0r==O8TwV%u;6AX`jC(M%{la6Y#yn8OQ53HGAv&}nsv>W zWQWvGV#dE1#O{m2*FzncD%XK$Ms-+x^(^{oE}`pi0p9V);*jto@~RD{-cgHDQEr7( z-X^GdSx+-Ai(yi!JX(dh=f5EUJC-fP-+>_vO&P@Owo1rdRL1x`4LAv!qHlQ(mL9D` z-`F#VtUUy&Cr4m8Z7~v39FW>!38g(6=$PY02Srp!Pfw6k$LO&RQ#CfqT8j1eox)&M z5{fnk;+=XvWr@8dgP<1_a6y@wjgw@#gObc(SQv2qLHk#TKv^XK{r$lhRThIe2i9Ta zD-RvJO_BU}KXO{vVc*@2=(_Qi9%X8wb;>BtnE#Z5{p!dqA)30rWzlQI5U*f4DOlbk z>ve)itN6kYsAMT z!YF$tjt>+gd5$8BnytV(o_|G$*%aDo^NUwL`6sVkp%hyLLUANF8b9VH;osIHkQ&s4 zX2NX9*)M~~A_x3ldY5*n#L^$$=b_qW z5q=I|#6O(^T-@b_Emj@0drB1@Hs4F~^Ecy)pDk2Urs2g(9qj*Xfx+JvIA)^CVtYoj z_dfSv#ND^xgMqNn4Z;}x7#N$Upf^1pYd1x+%@vspfr+f!pb|Qf&#|uMHclKF&(&fh zWH$aG-PF8Ek$>)yR#O^@F1(HR0<|bQ=MD2>9ztBTk>fQ6-ES_y#WM-Rhj+txl>zHZ z5oOAMGoT+R&5qh}>;7I3%v*XPf8`TieEbTJ=^?Pa{2v-y+i8Ym5fpeeDEVH1j2C>Y zoV69VecaIGBE>RG6`9RW1=cb_fz{7>i{%=baCQ#iFPR;{SEB%aY^M+WPOV4pEMVa~ z3HGH?fH@6+h0&=dDzB77rE4ZOy3B?6x#dWgwnn$I0JDg!#*FS`TospynWv89%BLV$ z_1+}~+Z*&bIg6-eG8@0-F7y}Vp?Kr~-d%A-9&au>OfsPK$qUwPpXjP_Jp@iXL7Q$F z-ldLazkJ?erJ-kTMhF94j#$Ef2@j9QN5n+cVHqsaxxJhv<#n=<81n8S?fVZs)Tm5ew z8(*)>?nfn1;D%@tRyjm*yBpYp=Orx2^cXvwI|bFMi?DZAJQPL6`1@yw^0jI{us|$xKT}haJ~df|J^J@*9^&i!DuA#{^6E>X|At zxb6xacMBx6&&8pkc=rCmQD*9#z}jOCX`}BplKVHG?!7n;-P=)kp?8F8;m*0C-6P{7``gpLrP{f zUT!~za~au~Goc2?!&fmQX#s{3=fT70B+j`hus=ZxEKgdJX$Ebx(H>xsE3yFg9u82c$6 zPoIV3h{HBa{~Asw&S%nt@+YL}tir0dDKX32|1e5pJd8q?z|hl;-MKBm}Mb+`y@#T?5Nm`dB>q=EysmCizeraG<#N#MV|2B?#0q{6Nx zP>T!2Ug?uO#m{}b*W-mLx8MX`b%bL5P#}i47t)1MHDv7gkTh3}q2C5gG#oMkW3q10 z%Fb)l@HdpyT)Rk4b0YRB0_VqR^CtY$ruhHr=)?X0&~scLZ;ZHn$S{V3!(Y(0!~dwj zN{h19Jn7$=`E<}o8Sz|vk%&--@I_M^kzPfHhr;RTHXXDmuE6KH&RA4>jC}U{(YS|h z6wo@zb3ea;#D>bK?vfN{OPxZ!kpN43n@O|X-;(b3iT~4Zj8+B}k=E85yJs-MD z|8_p1BVQiSD|;m@zp8@rmZLAt_3@>@fjb+vbu%bwAg6&m3#Nt)!n(apdlPnO@X$ z&#iV9qR@Q*I^$4@>Rz)bh@VLMRY~-(cPkolgJ77O1-%OeB(7aY9gcORc>Ndd z7YyO-76oRKcAwY!D1!E!y+|&x+`9<{JV~7*iWw6Lm9tN4?anCFgm&3zx zLoy!su1!LwJ|EM}jG=nU454Y(__Syu{FEc;pJ)Twa``O1zD|3Vye8#s@~{+lV;MLdXRr|t>jeZD!2 zKN&)#X(j@WT)??AJutYg4)L}C2(Hz(@jKWY>Ov3blvX{tet$`0+f88g zWGnUtxZ`HpWojL&C3)XonqoH{Q)Fl2FNYnAc}jveYAcGYc0p=XG1<+!K^E0{WV-k* z^>+8uq*xJ5?JnTu9^A_d&{ZLa1_tFDLD1i{3yXhs5dTRp{qDL$8W&emZKEAkW!zKG`99FU6*#nw-?TMox2JiUm_5r5CIhrN3eJ4xD$09$-Kv;W+4IH zu$AzxJw;OG^S-YwR{!ys(*-HEK3=lX!9wvba zc<2~~+cz`W%H^L~b*CKv_^gE#=JJKAL&VTv5W zfWl6xAtzfAx$}9L_Hi4QzWfjGxty}YVKTO=t6=YXb2!?ZKw0Qch!h#H!kopd!DAuQ zs6RjnTTb*^ zcL!7Hj-!#a2GqC87ro{j=4r3WKAiPp(UPmlA@dq#`J3QEs1+R*Nu@Yuy5fxV&{*ENH)+HVu)x0Qw|TtxI^?m`gxXaX z!*G2Fgw=y#w_l&+x=rBgI~Ok;BhdIb1sQqAajY}mY|Em>)SBVJb`*N#i3R& zf#j0aXcUjaE)D|@Y&%ZM(?zg3SPXF;CrD_@7wSkGgE*_Dyh}mZya(@Z@x-GKquTE# zjL`+fsT+BxrH+z*&@cKBHB9+M{Zx2R7^&9XW=Vg%$Yi@U>#U_LUnz=x5(~mr!Fbg0GN2eXk)>`^XCM77 znP+(s_J~}8p=cXgFU4TiK7Tx}It`c7a{S6`NBNQ_EXWh0slPpoHd+Cm}lK0^AnghNL0KIh_}TM&cJr z&{N0f(>t*#Y9>U2XK?%d2XgVz#xCIr=nAi+%D&5V=!_yf#wS7ZC6|MA72#Fs4=RBj zomLMZ_nsF}lGSJ4hcwv9PiJvvQVX2ZCD}uBS?oP50KL&VIQ~tS9lmSCa{C;Z#Gk#` z#o=m$D@$-R!vemwGvMui4Cb%yk^Hb6)@+=EOPOD2%bFz^m23>Gn~E7rwQzaueEgP4 z#jaD2@mS_2O4LU&Grj=VL-i3Vcf5g0jVv3^;;?_$a}eJ72peAy!#*nj_0~p9%mZ2iB0D)WGIpybOen3M6 z|JqU?jJF8G_tr2(AG(4p-zwxMJcav?9FqKaj_w@JLv@WTtd;b!lv{rjt8O8_|0eGG z9>?GQaC~e@L+#uY*j=xuJk2h8`|<)wZ4_cDrK6dk{baVgI|CElixD=Z65ISw!T4G{ zhmS;HtV9xAr!^p?xe1bw-atC%J=~p&pq-G!b`EASsb@oM>3LAOTQF_g=SRCoKgaLA zO(^_tCbJN##jnjn9RE3p{C*i68dk@-P;J~1+>dQuCD?0x4|)59ux^b6UadJtJJ;UD zsfcUfWxq#iu`Ba_wSXOUieSF_XDR8`3-T6~!dTrT=GdCfzQtv+=I`HmyS6y;#906t zKaOX$26aqB>ph33MdQr7IEc9xpto)#rl0qMFIS^TpXd0pD^HR0O@JA=yr2+{qq64e zjQ+9$uzzn5#usL@f4F0^lq;Sa#Gy6rG`s)wEc-pGhs_hNr~0mL8n>Lws}#cVfs5Nt zf~g3-HIDsSG=nV=^khzmz`4fb*nA`ly-^=%k(d0Gdg?rcbG58fP$VpO3!pbc3HLWLJPzRSVuQy}t9=V|BY!k$?Z(#?5$K&W zLXF)6WTY|)MSbl!9NmahH#)GV%mATl91(in94FNZNOn;x31t4Ei&nXqRM-oh(cf@5 z>p6lJJcM6+5|Vm?NqVS~`X64PF=NXp;8{1_>8PWdrcE^W*$};BH%M+qKUFJFf_U5q z3gZdzFLf&Ms}oK6&)yzG=Da{;DJ@5G@n$yQvyhc@S_Au=wNM!yfr;xwq15OJpJ(<+ zTx^KR?G*@HaSA)!Q*dTPf@&g-saC9?H&0ZH1qwXH%@YmSw^kb#yG<~^O$#Q=c7Tm> zM%8CC%x=C$zwC;M-|R+t)?N6|?IP+N(vT`01<8_ykk~vC$-n-?x>jqL=uboY#!a~0 zyBztioe{9}C9U*Ip%+TYlv}j`etUTc?Jx&zw1v2)8sx0LQKRK21Shm3)bR$c?{h-B z$0971*@-7};_+h@$1jK5VA{n5ChdBFHTSx(TV1WVKlKVc4RUZydOQ@43*+!RAvC|z zMrEEfq$Z7p;kRtG{yB(N<0Z&cpM`06r{hVe41T+DhC;sQgzW%%7V4`(9QV{$4Vy+va& z99;~fEp?c_{Q}k*#9+T)5X^qA#lezre$-Td{yNQZ z$1{xnDaD=9WQ`1V5_h_#;j72}o}UDk?BsI3o@lH*u^BNTWt^5E8#-K!)@(UVg3BwY zuq=d(QUzGp$=4X)SPj)fksMBZ6!Q-3!#3O9*k|B`XY*WeGCu_QUH({B?+%4U8Mr+w z0ZPk*;VBx&9*qxRIl2xkYf=+!`cXtJN7E_lcpJVhEkdM58VV+xGI13V_Gw!ev|3`Z z>drsMfAj5y!q;A{3;tC9`02{QeT&ls6p ziHe!JxVB^h^4~~c{P{`DV6!}%)*{ZPYL3OvP+4U54$;IUS!Suj@nE*1Y@x0NY|b#~ zHXFk!ERZP&A7vL;E@dGSUJz0dfu3O`m(OKm&GUWu7kdOVI4;s6PnuP%7GmaIr*Yw7 z5<+H0kk`WV!|EJn4(T==eD0U_75_^B2TF^2$b|Fs+&XPw7N z!F234NQUU(e~7f&jXMIyunB6w+JD8I-sd4U+i76dq{SSDv;aGI971aPR(!FXjFapO zIvU_?(k$emkayhZ^31|)ZBhl#?&K7Y!C7kkczAX}3i;rSx=5~~)Z^4}2 z7D_xTh8w1p^!h(B_EA`n9q_yZkA=14&^e{EWHPYd4 z5j5epBB}-+(A^2w=|-UdQi3{Znb}tAUax~!t;T4~9E)xD=ks@qGQR&rHGajvXw1w> z!PcX3DE(*1^f!C3m)};iTS>zhb6t?#U(*KO^E6O%BBoUOBQ)bItQXcI+w?B3+i^9z zMid2!E~3)A@=VKAke$%%LeMffn9Hdk`>iVe+OI`X&SvO5+ySkn#;~p$gNYSVIN_Je z^^-n>>0|3T{f#(WCya+?=M=EJb^1Tjdi9ZnLlhg_>29KxgFHlrL4yKZ4a(}hA-$|Eny0#b=_in*v;YV_uuL#SYOth#PV$laZ zs194hGb|T2W_PgiObKo_$D^>n0>-y4;Pt7?q~MiLSGFIg!IKl2mXr?r7BQM-7x-g* zz87M6({RZu9p821K$`9l8B+jrfjqAED8$>2Qk-7=4a)9)P%K0fC&c^hS ztB91`}_==rvtGdAOKnGz482f4c*>RM@1?(>BOjiNDh`^kzT4y zP|_ZOrz6nebP~dAGEq<%&hhqvFj8>CDV;p7ws?sd@&8EIQXIuTgY-dB7qvnE!K`;Y zO1ndGNh=f;{l`#uAsA*O`(R^p7&?xxY2V0OnshjihHjiC{}t6F>m5szIqud&>I2ny z-y{>Q#fW#`3{Bn?%)746-*ZocZ`(PMe?T`7I~E>+h|4iJ$Zuwohj+5qZuacdls7OR z{|3cl??N^w4_mcjaNqAJ+>_3OYpsm$vv0#K;y4vIt)w{vbLdXH3=7ndXD4ULFoBvH zDn9y)W{!GC4L>eHC+!3#_M|}U-wH@fTEOwE#+WMOgTI9lux-yrYpx1{dKDnuu8rr@ zS7FNa-Ea+BkD!xEEbI=aVcGQuUN4p-W#LBbshoz7ads$Zo`tEe$3Xwe1rl=1Bfp+Z z>XeVi9Q!0Zp3P}AKFmP=EpK!^b;K$eJ!Fnv4o~4(*xIpxy|&uU_D|f#)IME;jQBIG z-E|SIcQsJ(PYv4k>R3=PNFt|&k?AK2olB#TzEuhavjw0mu8Qi(3o&6PheJNS%EgL$ zSd40fMN<>3Zhk}Fs$s~6j%O;1l-bw)Qmm>j4tL!yp;4&<2F<}x=?cY+$H^Fq8U;ap z8LWvgz)9l@Qpx&Ab2`3~@IqHK44#7P^NX0)T7bi=9&$2_R+z~|@w?W9^Y5Gv<|kQN z!OMRoq`%q1etsg_lCENheHBLM_0#+;K`3*2kvp|12=+UP1G570IiVQie!qlXPcsz# z{oxRM1}^jbv3~w3N^`qNboM;W_g};|)z4+_W2Ug~nxpXC7JyMB5g2qBM5^)!Bu72P z_ccwpyyh}^Aw>whnt_$QweU!2g831ym0-mgHa6i56E!o#>W_N3y-5nSeKXkAVnZgZ zqr&9>=&|dYbeV1GSSF$D1T`$;G)#KX?cvVP|18FK$$mwsO*4Isxj`mRs;GI|L;URM zK||aFC{)d4OU10&iSwpx?0=VN=k;P*n|y+9CEZ~YO6%FE{u;KqIiHT1pP`}NpA@yf zij8~qfZe)qm(6Ot43BZoA#HIVucEHfs8elppt*&r_c<_;=0!}vY!2%`r3lxh$|!#) z4JdA7Yppz(yUuEMAl4RV7jJ}B%S6;)zsMdbWx+swu0cJHMuu*;d{FVki2+$WbIO`ZU~u8$<JwK;)rQ1KH7(KdLBz|`y#t{0VRM1!ci(bwZ!bc8g zOZ%yVl-E)?q{|Gn_@8JlZ1ydTut-KpT4F=(r(`h1X`WLvdz($ z?`noNPM>9M?TVl0M?t&g7v(m*BZn$$*09KpIS)EA)7Tj3vNZIWg(7Q23lBN1T%^Ss zT+&lOD$_)*+*HWLXrlAR6zn)4f$r_nNR}F)Vh+2oDLR0${a$Ep_Q8*}I*<#ogqeUj zJUCsW5XVnHmAD7jD}7MoeSwzt14PxSVsyo1u3nqP>Ae=fndbU3x{m&N}RNx2zM7*JmITiFfjlkKcXQlbO1v;w!-e> z4)pF?2Ze1X*xa5VmhT(O%%|0n)w28a+BcO-F1>(m-a{;_ZNvBZgSgxy#2n6i!@*x_ z$hPFPX6j9}@lX&RnC-))Ywl1juBP9d?)i#cB)zX(0!5W2$dR3kOO6dt@cxJkt2s?9 z_ii^FOsDVO$)u4om;DSgWbr!sEWMpYxeo1Vtbs$^F#oYfSkDzDJ z2WL0BK~S5+u*(#&o{2KK!XaF)Y{Wjv5xO(wA%~INrtyA%(As9XDTsi=O@h16Ji!FNtj+- zia%GYu`!^4jvx9?haR7%%lVnyUJ(i9*fbP%twR%;qmI*XBwdNZ(Yx~?JxP(vrxUPN z_8z{^xB>lF(_r{;HA*+}QK`8M>gSyBA!jB&dmhJ9*(Eq|P9K>DT>YGIjn-G5B$=>6 z6e9+QWfI|Bx&Sd{`j~u59cNzpL&1XcH(2e!7}XrK#U(-Qel9k+o5MH88Xs+Ju+CvV z&HmUoh^;Z5@)6_T!TGMTk9$!RK**N%(Isu`*#CyOIX~xq|G+QE_%! z^Z-J>FMxWIG@f&suu^3OIIH(l@t?Uk=(`T*`YhnoZ~=9L{;B2OfaOlG*h?Pkqy+svyb%HoBAdLa_Y2XMCyx-&q&7~$- zI#CGo)P!+-=P)^pti!DZ-tf!!Lw9;7p4mLY`emmuS(ek-{%*kXrMFN=arkui3?`(P zBR|XnUn&{W$NYyhaZZQ5B?9TqL9o#qjmdt3U?ML_V%kZZ*?j@KTR1$7yT@XRMcBW2 z8(jAr@lEBe`Loj9__{W2wEKMz<(ak7B-3i7gcreW>s6f1YbBFPbu6hkfp773^kdvh z8oBbFO1XIqExUvPUJzW*$Yb3f9>QbwVSTxd9IJ0r`N*h)@7leZ4E6{q74$K_=YugB~1op2ixU=I7_ipCi}LlhOR0WQR%#Mq~Wxdbt*1pa(x!; zNm)Og)>c8gmK$P^m9yf>*O=(0a^}``0NcG+KsV7Bhws+VwL6vcNALl)MTX&vPd4s9 zD8rcS3bMWXjH=x4(4Oh#DB5rlj}A3(7$V2rJ{qBcSz{5?WzF)^SF=ikSoV0I7CvYh z!}MvYm|o_LE!|)7 zqw5tiW_MuRJ0l!yn+?)2hyRykTv1q$nCCO#=`O}H``#nt+jopi(L(5N9h`9I;h6ae zIA`xdF1NRZ%dW?(GH)msZi9!fCuV=$fJ>h?qw%2)=_GF_Q7*T+*cpbK9HwlzARZp6 z#Wd^ZFxdBm% z@`ud&tYE5Si43ws_Vsc6^!)LB*%Na7^=ASxsVNku_CXjunX$rss%(tB23zz_o|XI* zVpE+Indj1lIAX;4sB-MEap(iY(|TZ-^%!E;WND_iF74MEN83{6n9aRm1T5-=N4N=u zIc@H!N25`-nZvFV3~%l3x-!~LGrf~R^A9kh~@%plX1hWscW%d+fsaXX~wia4XEb) z1~T_UaQ*FmycFF5=`o|(g<5ImlrGN-H%w!SAB|b#TW$8N&;thpy&$VM7orLEIGIt3 z${fyvXxquH&wRSxQbnU?E@aO_X0c?Q$*g&%AZ*GM;GZRnH{&z;ceIoEhh5|OWm17? zmDmkIw{=_&_!7%mDe$re0_(4m*Qn=I)zeC^tmuu?3?KbY zoLy(bz629neoUQdIZJRo5&n>USR)1Lm?C2a4+0Zh@L5I`m#apjJ69Au-|evexdIdI zHekmKRhDo=k7)!fW?yaE^*!&J9}0qznZ?Qq)^}FQquPL&306L zWZvOV*d^DCP%BA=*6CafaL;i10Us2_`r^d&@yxbEf|&=5Vh7JzVY#I?7S+8Y(d+#f zwW||BUwW{_dIAE9rsK_Qc@$ZmW5sy|Orx`r=^5$cTMomS(JENqk_exONHkCJ$6=dE z%-2?nEm_0Zr1q~E(0&E6{)?QJXDnR9Rd7W1E6qDtft4Iy`r}O{Vs`L3PDl^6pNw!i zR|_S-4B=a(g^!9Y@N{m)clic5-C2du;W)_PILgD0VpG-u#HdX~lGbe$h?k()=`8Ad zbg1fz4?WkIPipqHuxYG>OIJN&7uDc!*;yelRfxfYweUzT|*2En6s7Y>~?#f-lz=vV6{TIh3yu_SV3u94zvIofX_PuE(+NOGi(_Q*fyd|9{YOKcTtJo53-If{!ZJpS%56TbWr z1Afx{O%h=UXl|ZYjr)D_n2Y)n;)2u^v5k za?s3|f|qs|4P>lhxi@yOay4i6vL*t?D&Fvo+JlwxlMuqaKk_c~P!jr`W~ixQ6X)Tm zU8RXR`xT({gY#%*&A~bQm3Shi;;3q~Bw1 zNjo;&cSX6X7w2nOip(ji`QOVu`8M4f_)`KEA^Up{6b@{LKve<)yi3rcm4}DAezd_b zgnSd-DR$C$_#B)F*JwTDi5vyoWEc@8`umEojDDQb>g0)J6H35~i%nTf5W6fMLw zON7~q=>ymsv5`Wv{5wP6qnV+wo89C8rI%gba^y?9U`ARv~7=tB5pqGuR(M6Es9wEh5pR@NRnTy5p{PlLK3D!@SqH9*{;GeR5jT1=~=X6PZH&F zdskjb7%SNw#onzs&g8C5qPn6ao4s#7yUxz1p#)06`<}-Fgk5xaC2=w zebXtT%rA9x?_eEvbzDKlg#bLVoQAuBZdmwgJ)HYe@Wv(?)9aG3h~qT$)&-y}BM^!{ z3~wLI!T^_V+59zxfr}kJ%rwP^J*_x(HXGyjW?+==8czGX98H5p`21`XbG$f)@#hP% zjAdsLrg{T2?A0B@v(Zxu2pbMT3Y^Dt!y>y-7x8WS4A0%1A1+lqN-~ zkT%lP5(+8mcVE9hT^BCb<9w zsB7lC=F1JVyyhdFx%G|$y+pWZ2L$v?$4B*p6xk6%fd#4LxB4oQO?ei~CI-uWKC^Va zuPk1-gB41q(x9g`WUtRPfej0hlCcmoOsBy6zbtGy8I5jvSNv{0PVA5+Rg}w<@slP} z+4r3eXZ)ZV=O|&W%Vt6M#4JJ0NQwQn{fW^UC0H)Goa$tk(e2U&zDI9*HCdZEbOJfP0INCO#yvkVwe?YjO1EFNH3Sd;GD%! zN^r+1_m#MDPeA>N#dtQw8S}33Ed2&c%%7(Z{o`t=Uo{rim1E!$=S15gji|7936(xf z!0eE4JT;EPt{Y=1e1;}HOpOz?_o$14ja5aSE|Q{((ACs8Z9A1scBZACYmrzq7lT~9 z5E${Dl-FOPsZYmapT<7gT(W>pL{22X*b3@z{g%GDUZknz->AEfduaM!Cr5>G^!1 z0gq40;qvDF?5vCpd*CyKX^np`B+nD0sTG%mN7B42yX_iQGz4L(q`>A?3}?fXq}X4p z-Lx#viR!lt6eweiG}j>*y92Yvb78ekstmUIC5eER6T+H{td#4tpcrNOi|O+T!wsgsEX{a^o@96tb1s{XUJV)EMm2 zi^2!XC~_a1LRJ1@bj@fh{_t7z+0do1(SJ(1bMMjks%-K;l#LfWyY+&6k-2>}{Wi5B zU90(Y)Q|6hOHwF=jP)&;rw1>U&*FhZP@?ecKka2rE(R>xg@m^<;Og zhg5Ax;`QmDLU)Y|efG4Zi*E`cBNqp9af4N}J1OREBb#V<+W2EMrn(wq=sTZ|u7S2@tVLyojclArAz)A6iW6$f7CGvjL)*!|3GwyL-d z7;zqwFYkahhOs}3lGx}+Va(5ABwPJXhTW1G%tDL)L;MHM5$TG>pc_p{woqoupOsmK z*&^z=xR^|wMRer)9JbGW8dD2)WG4-iA$={bTFb%l5l#qV$qM=-%C3|FJd zu>A35`20PG-2D*2cr=6jW~U-`@^OPaOL&|^K$Dr|c70o1qn!6Z2hVbhLdx^ftTydn|N#<8pm zk`bI`i|VC(?w@=JH>`M{sEinM-`@sLp&2WSq*>>aZYY@5qVU9g<`7IQ>)2T4Qayng zq%UL$Uch_FTnj9MS*&c6nDO?o&=4`e*uj)myTdqRI=N`UqE`($GcU&K-!QA!?WeTzS=qN~p-QHvP zZ-Ecy4qA?x$^O6}Sr##%371#hz^y)UcHmAQo~1RQ^Op|%6!`sIaTs(bEMXJs*GpCQE@ExR$F7cqb<{oUBD8r9AQlt#3>)rdLu*hETPq%r*pyu~Ci#S?1=UEK$vg z`^J3uy?+th?2TE+r(w)wlrhWgy${J%5t#HR0$Vb^BVp@f-2d_#whAq%+&Gv`g9XR!w_=QNtr(m*s}R&D$}$-V5W5&6(!w$!ygc zYc|cK27eDkVcWYvC7gM49*(h6cq{XR zZryrC+EdrFSB48&_X&HJJXMAbov*_#J=A9>L&l?f(k5hH-icdTudrl+87uwl%CsyR zkW+dWVTBhEzr-D1wy!|XmmR1QdyC`~k1;;|6ijESG0}sM2nc=)h2Y_s9iRXe-yYK2 zMlg#s#XZu)9VrpqTC89nt%Yxrl@Q&C#DemT$lWXw^&jsjO6>Fy6>N`%C@=kpXuS(5!KNX5PLpTSQ5!E!~=+*vcINpb=$ zl9ppFyL%zMm!IW+ZqORQvlKpD?;30m1+J;@nwf}!)(laixkxz>f)P21XcM);JyU_* z9^gd!TQh08bszb8NiaQ!UPR?QK-LN|Jb7t}yPAd=w_O5D0xtM!T8jF(6u&Lihj2bA%{#3iN zo05l^L3cPjj>e&7L2Vl6fDW>X~V|>~u_`E!h2C1J=KhugP2Sv81egO9o z%@X~aw?ZUp+$X9o<$kh~-W;u?Pm1VG6+e9X*eU0tYE}>x- z_o?jF6*}Z7&W^m7VIz3nAmqOvsCAZRYqMq9{$;l)j?YP*|2jy)a1yTEIgHA`S(xFZ z0-v3Rke{Ou|A{@wJu-xiYnjZv#x7?co-APzi}o|u-?`*8ubg%dPayO9P?qa>ihVB4 zXBX_bf2~sw{VxSzy-f;})Ox^V4Vu}V1shPfF#_geGN2zPj}<>!XvVC3>IL^)6ei=& zp*yHBc~7Y+Kk10&2Qp1MjHpeo@%;KLlw~Z!yr3CK2{?>1dg1Kq(EUv9c07yzI~m?P zMHu~#5O+Hc9>cS6-8=;2>sxUovKrBU#n_q>FZ9iIzW zAkm2<5+AUurks|p<2r%81Ge_#dYhy)YmgqtcAuKe;ymXuqm}E}Y876%nCOb6uliUO zbBFu6?%-eW8(42Vg4%Uk;B4%QsoFEJ_vB8Laqn~P+{1Ld+KR@XKSdGmAL6;(8=T)- zkLYXjF{W!VMhPO^ZCr_K0qYQUG7Nq4JZn5>6D0XQb1mPiMfCnA{JTeg?-Y>U#GmB) zw}-C2I0g%L8RN#}ppfUC@76rP3b8^A*qBUtPNg)uwU9h_4<^6lHT37+d`esVjW(Ay zP~&w8nEp2bmlC;;Uh_A3{FN42%or#N;AMVG&IjV#hhXGh-Vb4y8Y{0=XX_UJL2yzj zo`_{*bJ7i*@HmVC(R?1N8UVX9r3g+tjTG(^YV1FOS=ZV0f^9+fS6+_+Rc1WHFOa~N^>C_;DTGm3b;DB5^sJz!tI(Y{CjkP?xd^Z=imO= zd*cu!dt$Kc{$dP2`tN5XYJ*A6RC*a-5xfpTWlKJZIVZN8=GwF6;*e>T@ zvc5D}X31dtPkGFh84Rr_E##8(n_7=(U`$p9j%&stH^C6wJWDBg<{!%P*TJJwKUmz3 zg;q`|G-F=kto=LDabtFji{mx2xzy5(uor= z&`HBN&Tl+**%gjjpD=rPGafzqgHzw)*`BH>Cb@J6`!PF+lwWTn!$+n4^lz*)k~mDUF`CQWReQ zinV*L<3NKx`&~v%M|Ko5Iv|ahjk2hJ*G@XIVXWtLAT!jBWdHpeC?t#sAkCibNS7;N zFJ6CQ7nANY zpyO~EQi)&T`}Y%ehYi8AX=-S?AmX0~1~klJ3cZ^&k;LsJnQ6u+a1$7&TNTpySx@N9 ze?c_B-2x?I5*YhQ1~EStp)P$A=Pla8xy=k_TRai^KmuNu*5Je3Ww6+01I7PjaC?&u zZni35i$OZp|EPv|T0HhXlw={^{n<}TzUOr{0()kR$mJV&rYsAqszOn@Fcssru0ykO z5K4}up#8*FG>$m}k?vvOZ6)2!{>FEDX(aZi8nG)I&|+Tz*MXK8yJ8AJIPr+S>+`D#|XXeMqubdeGD1) znT7uwB65=+BFa5<8s&w-kS*ANf8y4x=&>vNWIcyzpW2L6>($ucw;nHUy~lOl^|VFu zIi`z?v--d4Om(3Go3-Jua6xe(X9p=#<(P|T$`9mu3O^{nsz$@Hrx?+ZiearI;e4Gl zIaW`Fa$q7w&Wxc`FS5zL&zXf)d9yhg3)l-24Fv9Y@HMK#R#a2Z0n6ne17VP}?eyqrO z2+aK_qP@Tp%io9M<*(!DJItRQzAKnn`%zYJ9>=1`f5XUPMRsPvb3FV{6%XRxlj+4_ z*is%x5i5d7Jivj2nclfNtUm9|T*ardZ59sfg~xTOUiOfBE5c~FW-1h9PC~&j4wIK8;=VuE0ptTQ z{uk$CDXC(+g%oC2WZ=o3GThcKhs)&SsI1t99{mv*1}|3WV#40HYp~K&3z=$%5v#Ws zF}=Uhw6!aoCLRnRBa=ENm-3Lc9Y4c@QUfXD;tNv#sfyTz-OOsenCL}fKhf2^bU4jC zjk486xYQ_#2QE^0q4bl^^9<|&%de0(?tzC#97+5aO9f_W^zURp*7sSFjX18z(nlC0 zbjB$7ui|XHKl|8>1KZeyDu1SVx`4mNS4rG5pC~mDDfc#GPs0vO&3guW?T@(DEyspw z=;K$~Mx3n*!ri$C$xPCdJU<_zF1;}RwGaOA`D^d)5s*GN8CNe%z|TUS)o}91hcVWO zz2S!JJDg?vFbg}Qbx~6`1{dd;V@33N=on?-((GvbXtThzyd`)!Z2{howqxn@beMbZ zPY3Ez&R)0#9mVKdV66F+@9FL!NXAz=v964O4*z_(AQyd-9Y_y*? zXdI?`Ra3Im-psuP+2s2@o^IOIQT(1Kl$HO7{##%G`eA@sMb9XA)jZMAtqVn?BP>Na zxly=}bbOwlgoxWCnUTp7)@k`4lZ(<~H#y5md|xL@6b|CK>j8{YJ%!YOYP1dMMx5b0 z+?8HH^J=U}b*~Yv)fvhpV~4Q>BQ4fgvmDc^gYbPl&&xz`efqZ;jL!32sf0fMjgZIk zW;vvN@1j@}HFypk1r4=QdVC~_mT}KinzIA<3E1F8iyJInhe1#XgkRKZY(JDs{U+TZ zg$3#4ll&XypXyMyv$su3m)dI z2g48;yx)dHRg*X~XbZZB9ma^(FL2!b2Z@irL(AtZ3{0+J)znh7NvX1P@3dI+I4u@p zUj@(Ws?5CFfJKzYVYpi?>NX}IEp{yaPS}c7K?%5CBZfIkWKr^)Gi9s7pjpz&c_4aB zarY-WldS-UUz(_28zTC8KSoq`{(`7(;S+os{}`9j!r-6Pg3xi{rnu3hY#s zIO{*+3w&<0<3>^^TF&-klg`IsLS{777xGMnUI8ApKE<`^_i^lQJvw<#?TULHKIydc z=fx8?Xu)^ZWTuaPDVor5l!B)Y_jL41GTm8H>_ESXtooJ(Q+hLz9ga0b?=2CG`wLKX zKMcE0e|*~#0nt+K5%@M7`?Ut)*fE~<-k%Gjr@a+Qy;4U57<}nZ}AV*D|&2Q0U%&2RHZ8?0a_~ z4bL;=Sx`9$RbP0P;t{448Dxy^GAH7Bremn|w?5>R}w2yfrNCyjxhsXHN?F7En{fd@Z8Wz!2} zy}g9&H|KaR;SBVpYpM*vbEoC7cUzC3MZ<9_qyi&?^x2;41sJmJ zHlzmmVEKaQv@^VsroV5a0X0`J^8Ewk@?D>fsVp0&(uB0h6VFCPN{dB$XRAQ z?U`>&Cc~_#NuhN_L}NZI6HN+r5!Hk*#EMhXu;9Lk zb0Gv~GHC*PtF6s7gSAjz7>UuhwnBHbJIH1y^ut`Su{{r?E}z4v%O$X0Zby_*E7-T> z36^66k@PqeVQs77*Zr4fMJr(YZVgN^*o?zs7LZl7Md>{YT2r@-tS-M55?-ysT$O29 zCvSsaD^^2g>u@LrYrs;%30*V2kau+!wyq4u{yI+tecg^*H&dzF+LJ18j-$N=&d4b8 zhMdPDy#7>Ac{MdOTD_9=)UUu_CJ~k2QqhyHg!(o^yyF^fjf@UH$s54utrk)ihO?vz zfvli*8OQF2!)3n@ye{s=m`d()IV_7mQ$Lb6XIB5!YNfLM3i$m-A1Yni__EOywzH?e zk28~ZUgkc$4QH_|D-olQL}J!6Zw%zmw|=P+Xnyz!2i|^wc;5#+aUICEooPa-OCvvb zbC4$~4YPH+$cc|dS0C5*bC<$?uM}<-e5Q*=Pw8c}0vqMgiMCbEu&J?zEZ5sQq!-|A zy`xBXwu4Bw!BV8}J^{5+CY+z7g4^dZpecSI%7gQuw(SB9-J^u4jMcC@V2nqShA8

    Uy z&Mkw&kmcw*59GvlQlrUQ&Kq{&e6cm?HzW-zN(JaUQw}r9UF`VVwXE=lGHVDcro!fW zvi@C8!*|;u`Jg|R{_3UWe!S$WItS|`<9P0BEB58hLed-HAJ+sfsg@#8_YCI`{-U0W z{ya0O0o!hC*wwfr_BdzNH;A#2P$_0x{0fQMkLkjWblRg6Om0?NnMuYnc28>>(|OfL zYNvFd$h9|NRWS?C%wyX=#j_*5$!ME&1||E0xyC3CBc4;Apso*y>9fgw1K8Os%Iu(u zCf02EM`KrMa^HwFd%i`S?H$<5xv{gbKS>RXHp!todJEh4)RQF)w`X$ocj@8l^Q3++ zjEYMGkruZLTg>?#yq#wfx!=qtybW#@@tn(c5VvXr@F+it^ev;QVAK`5Xko(la@-GZ zeu?tzRw9>ctd*WT@4O)to%MmJbzKX2?u{R;a2%yqBcX5ll^TLR)8~^ic%aHZ%g%ES zZt^-zmqe>5T|AIE*fp zKzxXuk8_J$S=cT;AB`qqJlkCf@Mmt$D>E{a2&3poCwgG|k%r8yrLM)bba0V3RSt=w9IseH zj2muxx#FnDQrvzfA?lr{B&vy*6}5I9!!~a}?&VpIQ>(h+k)p;1a7Loo(XUu@z8x)+ zkFY(Ivye|}pndjo&b>K=;OQ|C=PcJh3;zf^9R^UJz7owG(*XIH8?bqH59btlj-y5y z(SdE0)@qAv(`A^!8T@Oj2cwZG!mwd9=E&WopI4vKoEvdurhbp!eW|9n;*Zo-4ybi% z!rWyJX5}ozM~fMlKLc10Bo3J~Z8Wc&vBW6r1y$P9Ui^Tob&?A}aTuM#Z zyK(I>zZXkIVb$JK=uGe?n_25=!&rChEiIwHLtoH<#256`>mVlQtbnfnk-x~PyGON)gp0l>~4JHUc;jgGC;$*hGhE`F1Jh|xt_DatJE>l>Nvh>Wn#*l zd?XZGA;Mxc=dGB)zAlZet&d`gwrT9*+evsn!~rbI0tSi!P&f6!6z<)<%DwxhB_B}} z{{*E0=g~g%CicF1$eFr(u-thA3|~5;_RuLzpLztFY&UW}SQpiHX?%AbgUvC4khDvu z=?fAldE#Mux5JR0H7%kWeG_TJo#)h=f0rhFzfFmq{%i_ld@>+XPY2`ghvpJBRPY>p zeT@!Qsl398{qoHF{!hI0NXD|^+(Ry$#Hx4&R_FE%N3%}z9&V3W&9?i$w ztt)W8JOs)+IM-G=g&EC?XG;10*gy6-X8@#Qr>jp5I8Rp(M| z@cBXnbnYJFoYw)kZ7Gj11rv1Xo<{bZGl;9ZjFhf6NKVsY!F4Y&%}$)p8;VRfPl64) z#CxfBai8}nc{VY87u3~{!qYGX#!0?-+FF5(jnzQPShz>e#1`=t_<-jq-0_<;gmu_n z?sMlOF_Zq*hy7=AL{oPqi1PZ!i-v}H;M_bPbnkLN?TS0tx92OIe176^hb)rhzmc7H z0Bol7th|*bA|3Qml_}1i4wYgJv#w#1{&8fgCBi}X7*>q=LTN|q$atR^^h%^zxw;}# zuTx>|p90}ha|UzHmZ7lVIEMRXLHRn@Ack;`U2-n|P3C-%E1GOnmIm`(@*X!*PO{P7 zIjrSm2|MwjKeYB}qbt9WMz9XvIna+q&uhem#DBnfF*f3iDzkoZ8V7&-;>A6GjJ+Vu z#;Gf?Gl$gJE7jr1D>uL!TUC^c4P*%szpyGa3kxmwvBZ7$?Bt&Ltia$1wQNtPsyhiZ z5&6ubUp_k#7|T54uG6;Tk0|xu3DUIu$=0fgi+ZO0V~Ha#pjaUX^ZwpM*Azq8<@|?; z$2MGhv|-M5wk$$pK3m^k6J6Szr=M<#7S2dE5{y{GZN>tZZN#mwjyTL``9(X6S@x%F z7VDYEtnvb=-=Qk1cw9t12QHw-I|(r{DOfB13ipbd@h!L&$0i@em3NVR=Uj*d!69_1 z{TVs=OQXa4GS0rbfmd`7jfXw3DPb*kI{M;5-b?)It%JjdBAmJs4K2RUYI3;*wGB2H z_<0*rB38oGvkFIMmSgPT3XIEGf&Q=7p{phk4jO~lmSJs}w6zf5)?2fD>rw2$Mop&s zi|^2;2jSA~NcavEV;9uw@#F&63QesbIBdmD?jvm;z|cF|7i;d6!exXDEt36@T(*hG zyq^j?n61smyANaV;JkP4b#D93dq)b^z%0%RBfQ;Ed2I~)Z@me#3^CxlViD&ry2IOI zIrecL!Qjo)k$Kwzdu_v!U&C{$SN(C}z6zUfJ%FwEZowwoFuJfIm-mWXB2)KysI8rV z74r?C?D>dv7eAr-&M)bbvp>?>{UOVFA^Q&v7yWG-EgIHpEXw1%v4m|Wu-Hh1N%uCQdSA+eB3o-be8je=j()-#%gmZ7loNae`HtVTi^m`E981__1 zQuu-qEj0-L$TeBjTy%t%Bj-m1&a|mx$KE-}{4p8o$_6OUoC`%m52Sm1!GnOoj1B9> zV1Bn5EB>3*8};y)&z|iYduW;7St<+V-t@M3@Z=HhCrBY(f%BvpRwu5u4#X+ZP{V4zb0GhnTBl z5UVNrg;C%Bp?ysc?nh|Dr#NB2ja-cX#ER)l|$9(v;ok)rqo`sdFe#diWcW@y6e zp$01 z`acI@@;(U0n#uSuItkwb7)%e_A%Bt&rr2n+aqS9x2Q`R2uRV<|OJX@^>^ZIWb``Dv zvOrXKe2(aBnGbG{ip2J+02J_CQk!E5&!Y3rnxJUfmiU2ojShrKqXZ`I;u&}`0~mhE zPz_U#UvRT{rT3Y!+1#ND; zn`wAEEaU&;MvgX9y4k}sub;8bL!9+a7=pQWhPeB95I*poj(u_ne_fIgXYB>Uyf7%Y z$nm^8*CY%iSgJ%ID)Tg0{L$|`)3XB4;Hh&eHqv5$ByimX9#4+aDd@eefo3KU7~!vT|23cA!k1xo}oy z5YLC`$ciCtj4Bh>3}B}!1~LhYxj4Jc826iX;4Klw#(wo3rct-wrZc=BY5-w{P zV`y(Q7AOVct9<}`mA}DV?*vS&6*w=-kojqMWATDi&Rg4ro$U*-DO?fr!h?`puo#hP0QWG|#-*ouQ{?8+xU98wEHpJp&RUJpRYYI_WNITZ`TMX+gi z!H3m$(4KdLGk=TFyZs2BjQv8UU%NR=_6-f$HjQ+echZ;@-n7<19M9z?@%-i~+A(S~ zM(r}kEuNv|er-|IR5g(??>rJ?*P&us2KAjc(71_bDu$e39_u35X=&~udZfmBT7RL) z;WhU;J;pU_-qlz75CLs6EU|;{s#P-R@Y;MjC!R`;>BN*$$Fnj&GnV(O5RJWeaQ?+v z*u-x_Mb~<0haQDvxE3DPj7I;pQ&7I563zF2AWK7*bq^i}xkGl?>gj>oMM(&i&A{5R zxwsvc2>U0o|IcbZYbvG70XIl$N)~-CRb`LtyRl5a4bdIEr(wwu{5d}n^QH*w-qq>M z@}Duw3LgCj#0V{zn6=9adD?NBn>ivJ`dAjMI*ke`BqUGvc{T@KHQub`!U580!7@n_C& z4E(Pd`hK;vKK~Q#Svd?A8TQPra0zRjHJLR${fp=;(#&wYBI};`5qFIq;mwfi(C1mn zH%DJ18xLW9xeu%4RoVMN8f@=EIZ=yeALFWv=z@zHc;7i4Nz5cskceITV#c1ln9ITz zyD-_UOW1#JZCQ|B24?eYSlYo1>@H~Fth^q4bQfo@o}{AVcO1el2BPeS3=4az!%m)f zi;+Kr*zzlZ%uLdY>1^Ve$M?T!pL{dD>gF2wxJEWd|2>=U8iK@ODQNvYnn^wQ&F&Qp z5MB1r7ulA6MpFA1NOnlFYco?(`kDLJxo6=D|NXrQW10C9uulJ(kV~_|-*{M@6*%Pbl^yc{un)%}aSw6gl-S2L~vi=pc zUzjrHY05TPjb>3HV$7l6cZmL|F!No~yw`9TmK>gnWv0VePY$0YS1GV_2lgV+ve!iJ4phWVVE zH0MMx++}Mpy+7kBG%9_vi=Ip=%DfaWU9GhwQl6PwMVRYs%D4%_g2p4H~ zx^)nXc2sBb>DgHB9>Y14T>oGEhHAEdCe?A3^vGxso7l_eCf?nrW><~n#ZR!cp&q|X z-ecRc2HvAp!+A*jx%hqrJDjwV9k^x`wr1rx|Aj6!ddD)x=)kE@Hc@kz3Uimr5!y-zQtNM+HTSE)EAn}#Kg z+i`4*g-H4&XZ9U;5Y^Y8g_BM=Hdq}-v9`eA=E%M-pTRcvmO^4!2JHBJ^KQXX%uv|F znGn9{kgvk?VU;*k>U^cF@(hV7`lVC7A+mEZypjKy@z|$M(a8w@NTh>5t!)nfQ4v z7=s!))1s(^#zv*m7I$|#`!N=YSE3Mc?jS~Z&O$|qAYY>SrMR#t6s6c^6y3WH4JxHc;M+RqR;L{`^hEy4+%Zmd?WeqzPuNae%qwGKg)7 zCcQvk%8U1fYJ=J3pZE+OdJ4CvxL2A=0$(D4-*sg#D%n@;0uO%hz2 z#;~}ZYOH^c9evmdrm~%ZQHWWKG5|R8m7CPaVQMNG))knf%bF~2D6FIkU&N*C-uA&(!m2{^! znsSm=+1-7gVUtye@6(oJf3Pv<%Ie|zs|vV{yM_Xx0KR*&;c_?|O$+1EvLFxl)}4cu zNd^?R`?58TzAWus5R>D3*drlQ0myl9O-n)o(7vs5C+!t+ULgDZx9A}k} zzT)X|jz$!^LW_uUG?zEbvLO(A>Uc~S^i?j<<-gJwS|X4j8hVqRUDEVeBIY7)5^+;Ryj zE7dSzoGRxQ^E}d0{yrXK&8^hEgzR1KOvfN+}yN4Rx_;g%4UKA%mra#nXQmI<%*yj7b57-{Vv#y_MzuzylCK;*|f!N znc2>h_sq19PBpt;KbB&4z7Pg0rU_HmY!M#z%@y7ySPO;s2h;MC!$~3NnK1OrN@1Ck zli3O9O4D=3ijePAY(~Tj%HcbO zX~Qjq;~J8JMDk6^|E`Yc)y#q6Sse{t$G3pAa^c z91tq|4hU?;3a5U8k;GlfSA7j5vMUv;TTkRAA=Ul7o_wa z2pQ(t!k6(!g$I3Mg2cr-VOUG0S?2@=VN*bqu=&Yp;f-sd&`?q+bd9j)S;L`}wkJkd zkgrUQZ89`#Zj*52W0>$|;y58CKiX{DQYZTPVHAmq9|$HXb%NXVBSOxIX~HwF%Yuhw zs^IjX*{nJ%S=co@Q_$&+6O?#EB0=U)ZMQ*zW`GlaB%cnev-EQRP@^Mt>tOX!2(PbJ}j zlpnKKFv#5@44OYc*yF87o???JY_SfhX`~CQEDs543ckYJw>jMV5lCYer;`7LokFom zkZ?OnTR3#efc(8BNNjS6pmO4?AZuJGd`W2)++6$9)pO#cCetl^GoLNw-7*&*8cPd7 z5}iUW>)J*k=iY5W^O894i~1~_-26hVW?aexZ9RBlmk z?3tgwS#*t?us6^{cOh@5FI*xKlzVxm6kbWM;V zHW=&HSm4AWOXx)oU~i1)um%%LCKiwa#k=|NEbPEL1rNMjmyQu}Z%Af#CTb;$;5hCs z-YUKp?oY2428+pYCW#6&JUxoptg>fkOBCSmWrZ0Z9k9sD8J>%4V`@F`J<}CliB5UW9pr4319Z*%ae(IHxiU-?vPGn&Eal zZr6oM(@3;4EokK5Aj_iHv}Aw;zE7ErlKIQIS7{%j)@9(f;RD`}@f+Pg&mv=JF;vEN z;hOmwObN?@*B@_N_mGF=Ox$ zLwCmGp3wx{TQvnACfMN+#o&rmJl5^Jg45Hs+UcZXAMLOFS$eanI+iEWFsBiq^$^<J``nzKB-Yu7% zjsM2%f>lL*VTtrmHJwU5F4Bct(#%118XLOfAbT0`1O0ejOZSr=OW0LS8~3%-jod}>Y5zeTW-n@@@~^PF*($^u8__I z)RWx0rR>LNFV0GiW~JLxg*kJt2*Fln0IL}5=6 zKYKH3QL(T9>RTTmi1QM@$wYB2tb}e%sv!A?7cgSzb@&@L!6>1b`iXz1?Oey7HYSYC zvboOEGICk5WhcoS4THnbNf?mNJr5gaA>iUZbjsi4J(iNJ^RX;*U%>AxqIgu6T}S5C z=QOdfg_I6T!9AfC>(#{ByIy0K7{Hll`IB+7-U36e%|xZGFP8E=(DW!@oaXl`FMcod z-ns|9;U_Uo;~JhX7=wO)mf-KP{m63crf3^2?r+q@iZA~m*5w$6R?Nh2KF2>QRcCj% zeuB+iHAu#3qf^I``|&&x!*`#0E*aQ2+!uz^Ly=Qdh7<8J_}VuX-f!38=*(z35u8U; zGu~1%*YrazK4G=4I$Qb54;$8GW1Vpkn!?06BmW60SPtR6AVoN6(hBlNFR+0L>rkH@Oqxez{1;`d5T z=BhH1Z4-1^WbtDpWSqu~<)o{pVEfX_|l2xK<+?h=fM{(*QoX*SJFwj`CTQ6!@} zg0&j-!E;>)%wE4E{qbMPWtIlQ|nv7Y`#2_dsmJ zn<{B!d28TP&3H(jGQo~;UC!cm1Bp$6)bW*QJU14HZ%H9nYA`Zp|DpwNT+rZXhMKi} z=RZUr57*wI4Gq_5LP;7rmgk~9{}?9S6obQxpVT_~1_kV%4T)_dF{D5p!C}17yw8>m z;e64E0U3Cz77pp&qxf@O5v!I|(Ap_!c?U=jUc|X=@|ZJF z6DgKI>0eI@n&$^0-rEX&Gq=F}?|w|G-HvlnPjITPiT8xuLh7#wbQWL4fY+z6y2k-| zp*9%$VkEx&Q^Ygn>D=418utfqe(6X#_`d6p&RuCpKU{>8*ka_a9e}BA6G2Z*pr*W4 zbn=R$Xyv}OBI}#FaN<7`b^RGNgcM_AVmfq{OR!3Gmt1%Mpr<$Ep!+tLrvAJ@gVtUn zE&T{Ax%WSc&O09K?+xR!M@A^CK}0DOp8GnF(xkE~AxfmasVFU^(y&R&2$iH1g%n8< zkxFG0g@#Znk%oq**6;lO=s&Ms4?fQ4ocmnY`+XQvp0OBhwx4&*xpVFNG8k#!BAcvq zy8AJen#QQGi_PM!;8+h{^)JIH%X!%0Y>%ZDfhgc?1e3w1Nao8lJltuBy?o7Y3_AzS zSkC+3OyPvkdD5Vz*eWA9*8=vFc;2>-p=&4B zY0Dx?{0FW3G(ZBSAf_`mi23YZ#xm}f(Ad!(-ytIp@yC;fajW7Jt_^qqxr zXfYZ_%fZZB0$*LKD0XT)j{El`>AM7*Qmc+H??lihdX+r2d0}2liM==|%bZ_mV&rko zoZa=06x{5XY5f#7Za}~mR@&igPY@*i9Z}C20GEm#;bg<_dV-EaB`Xb+J}1EZ&_VpV z#=XyGYw$xMk!*GxrH>h*9TfZ86R_hZ=QSCWvH%Oj(L*x61)8TajV=5rVmzOtpj26`w3Wj;wv&8p5m`v z4`SCnpsy!d=~h$$4ZHY?w9gL4=C%Q1r6D*nlY4@Bm*{xXG|X0+gIV6%s9B*8^%^ah zZ#+R!3pf|Uqn$EO-6ZLtGJ1Bbo&Nb&)0Gcv5mjY}>!n}m-H3kXQPRp(wzhL_z!a!P z3GrWmB3{3Ef$urp(7xJ?QN~u78$2E1hMMrax*8+YR^xt;9a2j+VBHbUn6wMy*-%M3 z@Z+^We6}QQn+l*AFqN{$O@${^FZP5U=DZ=1b*gytNCfW7#h|@z8U0pWNPQz;2+C(h z;i!2G2G= zJ)aXt)G9IdU7YFuD$)Vv{ZX+37>=hGLkrcO>2f z4A4~0Puf285#9dRLTfs=qd$`~jWwNM!ZkovrSpB z?h4Wv5R&nBGxD+w!22p+R4Rnv?#%$sqzuNOejki0H$~Ek@AR?v3;mHl1V!F~SiFa` zFqD5$$@V)`WnM_)EC$P$?ZfZ;E8sh`4MhrF=vdi@w1mO5;P@CC{e#H*f-O5R*O8?r z%wxI-&XDoCezLWYz@eH*W?b&eO50Yj|KFF$F1CRKKRY6Nx zBnqj+L7f^3WytAo$)|eB%t)kbz+6rZy96{Y_i8h+((%4Ej)ENeMY)DI#YH z4sMx_m=)u3ly|$2-qwUxsvLGoaNnJ$3;LaBqUMy3dw_ zZgDKww{39WxdnTZWwF6g1To5*82;ov|7?=ndo05p3Nz^NhFw&-W;Ff&dlZS|52K6E zoHp!ACtbZR@;rQl3}tO#RV~DnQvy`o<{1Y2aI88Og(Y&6$>6IA4YBwj(6YTurhX;V z!5v19H#IPcd)Mt-Rp7;E&i^is$DEBOnD=xZO?bJQ-Z*H`*vK=q_zQO;mL$@>c~i+s z#hUWR{S)Mk8iVxqf7ElXl}4<(&C+I_XM@7BS?^;-yc{utRvv3S% zI*x(8=Sa+QiJ{YPqv?BlDAmg3QmNcgGGl3UGf)n*PH=JiI zjz@<+)0f#V=+E;voXJ^0<6IAr-tBb=NDRecUmqB#Nuv9!7#8q#)?m#omQnnW9uk5KqCY;K0C!+d67 zD21qJ_K@*?V_IAniS%_L zJm(#YkuI^A%4e_J^#i#d>>yPclu^}%d^#w!#Fvu}+z~Sk<5I3t$ggs`do+$-RxN^{ z+YC!nEzq|s3DOS&c+NTsJBLWqW*c>i9V1DejpNxbA03waLYMh}+eQ~(XH(*Saqece zXMZGJSi6liYg{`6BR)GrizXtqYbmL>?WSkneW;~!55HI5i|Cwm=qaU=+van0O(TwU zlCB}Ttr}%7u3`0^H{^8eJz1WX$3Wt1^eyPZD|b9435+;eS zsY&r4{am4l?PIj@>~1H`b{~bOJQwXOKb%f&vLfyEGf1-_hWowU@#L>3bhYl&;IenL zC{qQt?qjg2#S}}lB+y#WO;3Lh#m-Y>5F~MnVr*k6Z~q;dX<&f*X(M4fR~F~{xR2aS z1u3To!&;T+S*P@q1g6=2B1Wpy3U@;*YX0vyoW<2-T?EXd%w+xWI zV-xjey(QnmHj;M`K$P!Y>^&Fb2KW7v=^nEG97Um=!!O$-4)GJ$C^h6EZEAQ)_pF-8 ztWys0Eqwjnz_>Aah z8)sydg|GzBQ z=A)PJ$!Z!a=KEvwMF#BH#Gito+VRx!PLb{@H(^@cX%vbVVbGjCs2!HdoqAC?J1h*5 z*?VzqN(g7@-6ZqMT%M!tBPrQ1_Qf=kJ&c{nGPJhiU3M}&{-z;TwFL-?#L0u!nEG%N zF1Uqbn0z1xc#r+#-cagYP)JdF7ZAdG7VD=R#?h;$xY{!w_Jbzi%&>7xHC~(F&xo)l ziSw{bFF>tV3C}a^geK2~kJU~@U{wMOdzZ&{dgn5qSx@oycQ=M|o|H%3BE)JC{9bXt zbGan4MoZ(fmlR5N6+mf!DgM4lL3p1XcxVxgTDtHYRt~pGg)rJ%joa4Y%t~Y>`xqn1 zE_G_LI1fcOC$SGz2U<}*ONxDL6K8sNx##hkKR)~WV5@Tjrs{lz@Z|?Qlk~%JsZi_* z-GNN+KiKkAlli|G!JIPwW8!)IerDEvH1tOZ|K>*s>y@L0lVh`ycAA%D{9>^4@&|Yi z6=!*4MOaJaGwSQFq_c;ear(F%_lv9HeXSg}?V8Rm^lLMbM^|wtIf?J_6Y+^VaHfR5 zq6yxg=;Nt+ihgU&zTJ>zhuRu(^~^>@uM9w|<6=l$osR|j)A0GU4Ln3e*rgf>G>vJR@EOp!YWlEtFSXhCKz>sP0@tdsgS{_c98nAN z+^d-EWsiV~n;@QP54ok^Aor5ar0ZCbF+h<5~CI*MyzD^rP@EX=ktRPIWWL;sIzbd%zmH?7gY;N< zuQ7|h?Tf9PdvI-eJO-cP{eZ(J$mP5s6NzyASnLnahIq`(=M3iOb78;S4N4O(BAH4s z(<%`eLAPN&{3a?6agRbH|9`FHpz|{gBX2CkflZ!;VF#Ft{v_mEdZ8RcD;WFM1s*z|IFwz=*&r2I;;sVN_6{SgrB=d9q%yvJ?q zim5A~;7-{WEN>TLv6Us}W~!t3{u#O+T~9TEog}v4BJLl&h@A?BcyA`hlD7Rp&((_% zeS3_Crtc$1y%37rGYdz$XT$K3II8NNQPP+?(&?(>U13kCrvum43`SSfN}>8~Z=u`+ zSK)I{GT{meocnCula)Kk00a8nNtYZF(1?0T=A!7JkG{$!l^kM z(Ac&TbC(p5)aZxwI^+=P-tNYhf)2Q6wBZgMNP>5mYd+ng>077bl!6(UyEZ?jl)%UxN84;Paa&yEL$2?J!vR@(!6?3)y%mVeH@tyg0WXt~b`hqP&EL zX?4-CiXQsH=k7fZwJ~{%2s~{b(LaNGbSUm24d$$x7lr%baM1}lxBt?5e+gvok;l-0 zdfLD0JvAR6paYQ;S;YYdmK`>ol_|^O$?Y*%&(G6N0fX^&@i&?_tdAc4ZKXx;nn?S2 zAKmm+gxAT5NNMjO!`;B2SA0G?XEKHeIooN>KMdl(`&253-Npbbc?s*1f>P ztEG4(UWkEzB{(pcXMc`lppdVjx_7w4<8B3+G=T|>G{6E1W=_&#{XF>AjdhXL0%RK}Bf70coKtcRBKy@)(I|!FfQRIvJ{z~^xx?k}45;fXGTksSrZT7(F{gsaNj{uD z>TjbDcG2urNDLbuvxRwIlB2J`Bk8jJOKR~v%OakavaUtNY-v{|L~h-KeNj7Z3~{2a zsj_4y8!T8mTa77=)aGlJ0?+FiU^HifFY6hO&dRB**TRwAG`3^&=Lj)a&Ip4p%cHAt zCENJihHW}IiG>*%(A?>+)XMKL?OmNP=cqll+@FirQ3p_}e;iHUcy8QIkuydNp&h{> zo+?UF;(rDC?k@#zc1W`3$1+S`=?TKutU$N5D`)n2A#(B>OdmcE^CPTqcEL2BPg#I; zSvMHjS)uo_FFwqXf%dnVm}51GJ196mq>Aq$)-_;?btPn8&w}9)d+4fiyxIQ* zb5}jWUV~34_3Px`m3-Wv{{b?Vt*Cfa21=QNSA`ScvXt)!XcctEBq2F!1rGbj5#9>v zO0pKI<{!bL|8fvpkc!<(A4%O#6^(Z^(JECyXT`cHKKv=2i1C4;Z9a?#cOln&KAEz$ zB)(3EJ`YGEqkl9c?+m951&^0BUMX(t7>Vh(G(IsFsHHg6#}<2^9g6BQFr)yk&D{1i8zuYjzMbuWPjuyNnhn%`Wt+PKcRyHMt`DO^H+?PNp~} zH3uJ(cz@|zB4V}%;}oAGpBx(p&Gpv1;}__n6vj5T0l{uz%K{ZC@V+oK2@xJ?lc zZ%`hedqnd4;5&;Z<2K(j58cZ>d{>@emDFQ=i1eaKPM-A7X$$=r?ZM*jEo6_28S}AR zPx@JH)GLgjhW0(IeAqtr)-!_5HQ{~1&>%e43PI|4A?_-T!j+TU{op9UT01^)$6qC8 za38(K>(BJQ`ZETLJ3D}lxhHNKQW~ML28kYUVdX2}h&o97UkuWGdjiVXGr-(+EQCCnjd{m;b zah@r{R&KX-S&=0*> z;x(KBC8{`7@6iI9o%5uN?5Be7ShVIRJN^#Be zaP{YA+`2iJJK;TGuD6Qw#{=z)=KQ3w8$Z!k??y6GlEB3xM;LYghp~rj`1dgjOFqeC;}(5X$0ehLXP*~l zMd7Q3F54BN%)%Nnp&)3%1HC4Eo|nzDpO-On!6k?*SK<0#JJ@6}45}K6y8G$ej}nWp z-Jw|hNDM}6WFelU$onlKtn<%344M80KL!tCHx|@Dq-`En4lxsE8)*qQ+ZqT{+BV>2 z^L|KoI-rr~jt#%(^E2~#e5m2v#Lmy;d3qgoulP;Ryaw@mnV<9}s~QV0<-%8u!A=;2 zADIr^cVo$OCfwuqDTltF&Lund7l=6>kD}f4ar%fQytS<{W9()q{UQ7q4?G{=Nx6w< z;2M{NTQUhy*ZPE}yH7ExEDng9kWPee@#V~ zQyQeVrNVYd5?K$eQALS=ZN$wH5 z|FweMUvms)yE$u0Vi#v_kH_K_)6gqFi~AxU;>o)UFpRl^_3g!E|Kbb{b9hEl#zokN zOw2xZ7HtbhAuCsebEnhj#Gx&0p2aNowr2xN85ux1{5;X>d4e981Y_u~S@^uf3Io;8 zv8Vk!3Zm{HbM{{DYH&h!`!0M^bRgNgMN|`dn*zHQ;fMVcgvAZUP$>_J{CALo(kf`z z1!*MaDnV1wMQ{&X<3`oxgK_jzVdf7Jr77F}rFZjFt-EHz2^= z{*RcV7Y8YxpC8;egiVn>i^zmfIIr`9-r?04`9lwqu3^}*Xcj((w^MsO@2Q>T?&8bI zxDd1o-xJ*soZ*7}z1*{)97at^$I0MG1b0;PpZAvY&>j|nPyek)n`ID8?Vi*6@o!0; z&q|`--9Rw+ldYliu;vcyn~n;|$Qc1u~kMBPnQS`bJX3oX9 zeJz7$K$W=nWdY^6_)^M^L|Wu}nk-&s(AmKmwDNu;6%M~j1@ESDzKamkuB$;TW0=s> zM?$D8{Kc$-AL8%GA{>rb$(dg6>_;JEr~ED1%*INH7xJ7+@j+aFb{fIMjqvbQFrseU z#d(vzaO4@>=@iJxkSAVQ*c0jC2U0g zgY)Jlav9x1H|G_TbXq*d)%ZYV+z$Ng4M*!SCCuxb1mhgeLDrIBAxBD~B)t!<=35}N z0j9^t;qobd*8k5R#lGIG%X$?%@!gjVvp5N>Hzg=GEx;awIe7m~pF0Q?!K~WJu#(Rh zXFjG&(^BDn(n&2lq8f1qx4+D`jLx@4&b8xBtDArb<#P3Jk zy7wm+@{Dh7v<5NXS7xhHP^=@a;wm_+s3sZ?Jc zO;nuAVzqKv{+l2sw(lS1Zq$e2>@667?g^VatdlKJd&G)-vmtryAnxcEBQ@Lvqts_$ zGIzIZDCtDd$`9~-rp%5W7UhgT?y$Pd?}#cznCQkKZ1PrJmg~3yci%6=;5TEDcqN`? zuk~Qdt<2cMPZ4Cjx`C2KF4LICOdJl5!&(u4tejlK8Qt-Sxv&oIa(SHLz=tosY4CgA zLm{`m(ptSt9>yfF%ol+@#n6j1Vvl;9<{@ojc}1&_P2*82@%Yb=kR+-0X0=;46o* zk2K-%dpJxchmm5`5z?DdOoLzSr)8pn_9?r7;9N{{r2rZ`TcK#y9%nPr~^(n}`O7vUc2T2V)G zwwK8A+yOyoeuF?KT9qWDEU5gJ4V~3+qLgJ0bfwvkJhz6Dk7co7{`z8pu|lK3Ys@~9 zo|Q=h)3V6XWwhW`f2qKG##_NEsS0`*(?*96)R9TDIZ6K8Oa~TmH=Aa@;KMsj@;Nt> zJ7h-SqtZkas9VC~RS1=w-Ag{|Y2+oQLJj-%NzVC?;Om`wL51cZGVD;Mue0sw%{?a) zTkA#>c9jb}Q)DP*o)|3};z>KF#FB&?=LKYNH&H#$JqYiS$M6Db9;l(ZU)6LeAcwBT zU8X3B+cZJWl=dE8_GR?n z%u=#9ETuiu_tS?1Bbc&|4nvR$)10SG_E*j5_8U*?l<}ksDWP;~LNYxWTqbz6pi$tV z7(+^b#K<9PG#y)EKo*lvk^JFY@}@&1s%ArJg|=j2?m%Bg=L)tgz9aZ{uR}2A{2%&e z(@*VJRS-Hcl%DIw(tKB*88USt^|&qc-F`78*k+PT#ARyjNT-}*1++Z9hP=y<(3%(C zFba%7wefN2jW{E4HNPux{Sz+`_bR3Se|IVIt2E|JJ3&h6g_IFqO@@kDf~9W;1a*TX z$oQ}hRj>1>%&U88SClNBzoSVj&x@0t{VUQhy+%jnZ_vWa!y$TpI1)5QAzO5_z+rHp z;6K@HL9qB@)PC5BWO0A=wHy@a)U*pOZ^#o=iL64;{ABzmSBeQ@^7O;Zn|4i)rdN8p z^xI$#X${&%&XGm@8EYgZ`Pbx?a7U0z4+L-8%LEoV@#H4fK#}QBsVP&1w*>MW> z{&t1u{+&=;8;|hTb%OuQP6?!XDe&JKdZOu09~&0YrB9Cq^6T~Kz9Hi-sbVtPeVDYg@=58S z0)kwmusA{$yNxYrPRATVKiT?xLVJB%MH{y>jkw1M$nK+E2$zdno{5X5;P7LqZbw`RR4qbky`4= zE%F|jOr1@a*GZ_!v4~YfO)xRbx%_LrmV@ zi)}ooR?p8S8X>{lKlmDTQw^DN1Y-%?hO_minygSql#TeC#l52Yk!6~RXQQeRX7U0x zXYRvaDxS<^nknb91N~9?hTq@*VpX#!8yqiy{W@n%$Pr@hn`N9Gy#j4L-niu_!Mlh` z$R08lwK5U7cbL0p4jkeBq&A9`lfv9?pvm7Am-wEeBSaegoN<+Mu?Q=coW$hXyJX|X z*P-rm8ouu-XNNT*t@;&4x<3;LrIV>E?I$5cjENZjfU#dWa{Dr%F(wkPcg17p982;H z8iXQU&WE+F!9Q;YmSew%`3-SFs!cU_2UOzy!EjVq*9Xn zE+F*085)WN$h$d_Gd9lQcyI|`o;`=X(;_h7pNbHksa5R%PD5_|qNLnP`o(>Fi~Mw0 zz>B%a()fk1VSn-JIp;Nue#X?3 zZ?K-f4eV$~H7z@zO*^}fQ(m+&+hnH33V4?L?BS7YhV>*ic8QRs>QCX$kxAFvsoA_c@!TvY=vd0Gv{>*m_n@u^PE>t=d^q1{$O#m zdz@lXtB$b|F?-qeGD*P~6Hzja-Hkk*IJQ`-ll?i`$Bdr+AUC6IJT0t2*Y@2=z8DKR zmmmaAE`gVoCfl=5hZ%~A0vmtxtF7vAx`t;;@*oet$>UG8uh4b?$cSKta|WTdjn6 zCL#f?f_J39p@n++bM&$13My2NfRr!6=35C?tl@4X&Y+dvK^VBNi5*&XCY6NW6F3#t|KMbK+-T) zxlWgP)_=fb2kx5PbQYTTACcV6>m*YiMGZ^}zda`5!`H<)<~^R?j~_zfx5R1GvAJyP zI>wfF$*^5bsdSig+cxdYrUcjd==cBb1PW~V)3EQZ5I zrkZ{~*-uZKqiLbY7F^-`=zY3;7HzJC@>(rKDZV6i?s|8WuBCOA1!UeL$wc27vuB@$ zOrmapZqFNpmf4yZ!k*Ioza?bzCXbRg?Zm05{~))A`>?a6(dRP;Y?d^N4H-7oPQ&*K zWlXi-hPZLNa3{qc4_@qtvCcw; zd8>qd8J@z{6~%PvQWK4B=6}09n>(5_*_Lvi@9*AmQ0uV4Q|ljOX2IugocGvzub1y{ zR-@_nVr;MxV3xE2-b!l2MM({(@9R^ittv^|GRi$ptj=r#o6sW5^41l=EA||m9OKda zvJ(Sl7vLnGh}wP1NPVIK{goP6-WCbb*bQj!or}slw^&GJ4tsSmj2#onpqDluDaxyc z-d&o(!ki767H{;p1p2eS?W@?6Az)_T#^6JO6S`lRVrBPz*tR{y+M+gOIq#rnn)_(B zcQ`ey4MpY%XU=rjLguj(aK9Lb^M=bIwl9;eeTyWMt+5nX7{W5DcCwV)KI~zXH(f1> zrN}5N_>`V!CO0mzZkKG+CDQ3ZfUzSto#*TlyO%3B($t3bINtuji5)%zr<6IRs$m<84;ThioyrpZb6Yl1l7pmI$-WeL?fJ z^pRD#2x-}qA^Dikg!kw|GA zW4r`ufBZ41-XD!p`JAJgi)QWs^E_ffta>f|;F-IV2c(&Mj|fxflw{}6C?h6!D2A^8 zM9QD0L3QL*I4oHV^(FGy96KC6;c95faYLG>H!{Dt!?{8cu2IABuucbYn=No5z=3DF zw_{z-u+@A;gexjx(ny5%yZE#*vi?oiT|%D^AZ4K#Cq3t29yC2Ij+izgn% z8KZOr&hI7zB}t5`)`L9H*nC{5&S$HlSaXnbMI(IKv{g&k_ABdQ6y=TXrE9TVO&R|c zsUa?ba5bIZzeO3K_wHnTn_|uH_f61vT@`EG_!(Vm0v?1-!mSq>coT9O+aH#oY+fF= zCvYzH?=mQSIFB!jE}%!5KZgddFvY$LpMJLDZ(cM4(n9bpD-z4co98H(`TFEPbo|UOx_@vb^42Jz_kjXtbMAnZPZ(4$9>l{#dFbB18m|v*L{eM; zsVh09yOT{C;Sy|3%wTqGzARfZg3sB5x8mg1Q2gS#c&UEg=jZ$Db346o@z_oX zvcfRiI}4gG&!TMG4P5$rgx&m`!~0v+Od^c$A)GR(`EoX8`t$vl+hBH2RD-SlPRzuA zJTrFV?mKUwz>P4n*Z?E9$HTQL4v8JfuD041Y@c}Ful!%6?KI~Lg zWIJAtVE4|}Q`Xi-S}pUKc1-eSss(G<$U|Ywv-2XA*gd2ZEehE8wS^jiGw%IU}d1e9L6X!>(MGq zt5O{Y`?T>r6Bzc#nW@fQ$9A>GGKs&h$^G{$+AJ!AqfaN`;NG?P6}*!7MoN(wSq+a3 zb%?a)-kBFWa4dBX{`u~t1-nyeu=z1sySfzbGHxTdqaIdEEz!2u9Oo;{vFa-K&cAg- z9l7HO4r1l;MEG2fgZS#>#9Ij$39#vPe9Pd6nvi*3BB+H zgc?NP%eZghfv_EGuzm7>sN>%Cw2ku-;yWLKxjP}Y zv=oICUqL+oCv|R@#bxz=8ejFD1`dCveD18Dnd6Qi&m}N_z7h5lqwsT349;onL;edn z?7uVw$IRs6we>cgxL83P%0&zN50ZUkJdIY4Cd1QZ*f2Q@%IT+J(<&z194Ic-$^6Pz z4L<{Eor}n;Oo7iQH+J^qS{4}S#(EM=n6%YY7PWFR>%29H1$K4edG=Zum?*NbpLAKz zr)li1_XB<&e@@ef4Z zCZVxK532K?VCCo+=#%~k`J4C2aG@?PYb{3gtH0P?KZqSGzlBQE?RY(eb1d)fKu>BF zdCD1M%!f_*em;sxRur*y+itPryyv-RRVaME%|liJcaALx!I+jqoWEiQscD2mm+Vo! zGX^8P55eq62&{AM5FNe|Rc=v`*-y-Dvl*LuVIq^%Fk&Z<&tltLz=ZPCn3C&qc4o>F z=68A!%h%Lo^G1PX%)5p^k1G&7eGnVqcWn-yBCM2+U;%pLz^>R~qJ$epj}l|=D<-p? zcbnOx8)|HFml1n&OO{Pex+IL=b6HqCr%c$>y#SXtY{Y@)XebU_hrJ0rW2DKM)lM&{ z|AZ>vAOZ)tikOJx;B<#tiny$SzDsrin9WL7$W0n z0P_vo#nvWTvEnF4-T~%riqa!^-#d&2U6E&3DkYg+SUv78tikl{Dfl>U5u4@`!n~I9 z@AHI+uyVyOcK6y~VbakF__dw;9`BmNu*jRWxOg!cwJ6qIy_J358N#&scJTRj8rEj> zUUy~~e&xxtdw(UElgc1=YvdHH-)#$hdG0LMlwujrHQ1J2JkR^zF?RY`Dtq&O4U-SK zNw*>|lKVg*Melyj>c)LylTtcZta1fy-dsahYF*)d`x{f=rzLdG7%OD|IIpFHvzt~v zLsf|wj>k{I*zO4Y;a=3Cyz|&SBZvi_na7z&bMZ8OBYdMavTBVz?2^h}_WSLAT+d2| z&4Xl~$!ucTtE$wQUY0IwFvt9fHc{g_vYs2J024;Ma2zvdgbPs_8d6{9i*cSDXdV5T;gGf(ifj zu)>LCOce{IZo1Iy{Cn2_Et;Qs5^$mn;*f-_ClDX6~TB_9s>J#mel<- za*r&dR*z`z@)=5-t{JkL9$hB;a{`OViN})h`|u>;Ach6vkm~xl*pHF~u@t@)El5Z%Q-Hbl&?RCr$XP?!EF}ZNw4g9(s zmZGz<$48g;1gCf6Kb{kCFZ+h=ZC@~H*h5IR zThgkmr95}~l_0pPR^i*Kz!e%VLT?s!Bs ze0Ce8k%B6IH=-BrkN!|qEYp^ToTdmaii@!{e)jH7Z9@Co1sM5Z5+p_#;zbq^5IPe9 zfws{462TPXlUeljgRDO_3)dc};ACqa=ZAY?mQ4T-+u5SW#}Nhg8*n1kiMu&g^RwqV zBzy+0PM?X9zim+1G7YUf^S5pM6Mja#1GS`kxK&pN|M`WuF;|0`aNb$yJtemOpA27* zil%9XW2+lw?mhm0CtU&IWPYN=GF`#dYL#2ugt>O zyhx#o;``V#SYn5 zvGti+_|T$?xGy4@{y~j}C#bQIVLD8=O^>ZuL(Kou1g11O7Jp<8;L-LNL>osyV{kg$ zt|r6rnjyZO8HOND8QhOKgh#wTF1{unC00Y(-jPbIbE`JzU8>>nsA0IaeK;hHbJ&GX zhuDR4huQVK*EFCiiwRaQ=xFU{rugy~yQue-ZT(pa-79rSu(%G#SSuXu(t**tkysP+ z7OTw0Fgns$X#Ht~9H8I*6$cTZ?xVtD$V-0I#Mbma=3Y`+ho! zZRMGwffMU+(Zn9hpVs5^sVfjvoy5dBJ@|3y1EdQ-LZ$gR=Jz~>+oxCf-P1*l&;O8( zYYuG){)lT44S2k_61x+uQ54|=*I*aOBNy{|NBWr4cI^IFjhp)};$+)tcxrP_>@qoM z92$adiFCAFtHyz`$FMYKK6hF0eiQGq=Z%?-ke%H5erh(F7mQ-3i*?zI!8)w}WgPT9 zqi~Ueuw>Iyh~?depUN%H3CV-Z&ME{gF2?BEQ*eA=1l`Fw$i5#yaysFp8nTUam894N zE6z8+Cdmw2$~lMN1peBl?Aa?UCmT48^Kj$3W*wxJQ7M%4lq7z=ZGf8G(F)27Xk)NwNlebV~ z4<-Z$T5AMX@oI30=b4KOv z7v|*p{honQgQgWcn9R{gCc8e436wS> z_k29O-^8Mp=eJdNaA)ood(523(9*&8;SZN!{kSAZJmuLs<2_ihmSbM5HNd6MxF&K7 zXU;{3axb8xI}E}|OdDzk;3giQiZ49 zToz`2(}1Y&M10|$@3{0gXw$BP->>^{)Zu5ujP;nR{t`7BziEKq(W`O)RoNFYRv7#j zAN2aMvgsC{2`b>-d4k^!h{MfY90Ts6DCdm1f>wRDX@MLo=l9+&h0)k7S_n=B!nKkx zD9Q%mp<@7!G$-TS*#nSH;^)WOFE}Un4|DI;veX@|?2eeUkZsq10iW&lo{_~IITQ9= zc`Ca<(}z9P9?$$2^E|2L6gDQn726gCBmBS`Q#E-UPrw2K*&}m!Pw8J+^boxzl_e!H#3o^Jf@i;S1PUoIN8Fc;GgFUBe zFxCGv3)=FV9r5_X%FFUFq%{j_OHX0&S)QBARzb{0YjkQnfReHX6I16-*IYeZU9N^@ zK}uLu^b60OWY{#T5iHu(4Y1{TXfkbID(XBM)9H8pIRp&QO0_ds5#D?TX~;Iqg@ zgqu`gsznFyfj@v#)BtpHx-N5FjG6!KF}Anid0=W<6uZO%%(p20nZ(k6IQyAr>b1Yn;tzZ*C^lnuHi&*qQb zfu!VUxVMHwV);WX=qm-!vEpp5E%W62^NdNnN3i+`wol;-Bm*=N+Sz_pbyc;>CiZzdr<}8|Ne6-xtYM zr*UZ;chR@S(Sd=pWcfiF4xutw_+7wRQEnu@Z#RtzIY!6wcL>IW`wE7ntQFKfl*P;h zDZH2=z~NoWu+8Zqv!ry|Q*s4KR*!IfL^(bi?}lqQcNcBWhd3Pr}yRAgRB;Czi31id|r-zU@2qO}y;S`I>1#0U3d zB`~E^1;dQAv3#aHUj3&5SBnL3HOiy;)31~2$E#$_y~<FYKTR6aCT@{3p$&#@38i&;vbkOxa5;Hr)P|ta}uXCly>xDX9xKdBPpCT#cb}V)D zCs2zhp8;3RK-iRVkO`WE3EiV`CPkAwTISI2;RmVl^mh7tG!Z2~6S0i%x`s>)r{2O~ zN6Z$A+xe@Mv)Im)H|92}HV*h3%iu^p07QImA|yE78m7kMUa z=nwK*tApa7pQs>WBUXC(WBOfRh%~CRnXLva_Np$sVXsBWdBZ8|gE9$>Et!dd9XsLV z%(@3WNGm#kT&~5F{`RfRc#AjF{OHS$#;D^WalfUc8df`Hlkv6(^r7b~ISkWA{9XR0 z$>gkx3}ZU(xrzp_noVctZ-DilASBN`hgpNUBla!O^o;YoH@dNhuDh9`-a$5WsS&-C zG9n94adL~&gn1h8b7cq+e&!8&w!Og{ou_#4(uZzwu2x7#E+r-T)6vW`IIhGERq zca&Q5nbJ-SMSlNFvNnB5zw5*>p?D{sC+`5Rh2y1BF3MH%@b}*XWKCO(Lrq)o@G0kW z%?Czb)kds>A-W|TvFOTwXpRcUr-26gbU+H---hzNbs%@M15?n>?lJb1&(nwE|A`9wZ*Dp$ZoX%(}OiGecI=55xWRdY`8t+AB=p z_imkFqfP*pPs>K-*dWY%WC-4BhnN$vme2I(O-~R+z04P6^XyHn|4YG{57z}>okA!l zGm4fL@1Y47MNmFh99ojEsLl8nYpm^O)2sfnKJ)d^Gu(D5p0SZgM9mnuW(J`{6572tGpF{VcIcW;>@oj4{=FWm;wyOeO~nnuI;u@B#! ztKhfUNMz3IpaH8T5R0>g`K#&pb*hKFjY}!{k}Vy06^_S6M{&`91@!j|(dDa;CnXH! z`m^xqi74W?^ib=hR77vO29Y5tsCI6q>u!Vb<5oBwC<}nZgAiye+lJXHk4a`j2kq>8 zK({X6g@kP*ett?qaO+OY>-5LO@xjnJ`I*!gjK#~?aomM9gT)ElnB$QxY~Xz>t>k^k zEs_D8m#>dU+~v9I!(;?|f2Rf3dN^m!f7j~FkkDlcb#Enfz4=H^_c*_*$qbvWUc#Og z{7tt#5u1-@<6KlC#Fltrt6Mv?UUb80!gpMDtH8&k?}!@5JJt5%_VzBG@P4jIe*_Aoc?C<0ryj;WY77q_#Ogndn^;ijeLlz2ezH<;R z8$1$^p3KDc4NLI;=^`wM-HP^@Kz!FNrDYc_VdQ%f;%mjQPgIw4@XhdKN-F2XCF51s zdf4o7fmDMtMh9?DdU!dh#U+tM$}Un_C(R7H+p$+E8{tM$42q;Fb;r{3Q*kI4^K#C_IF)^IAyZf`Blh&}*MbGKZw{G%SRZH1xzN0DS z1!nEffu&d#fAKuPd(%?18fhWs^%#6g8ppX^Wq4k489uu!(WKB!he8WU>hKXND)+_C zTOOGGoO8nVs<7+6;>?40cEt*# zUEueZ^%~5)WHcMxxg7n0>)>c;jCFm%EN8+Ic26gO{gjKNM`sGDe_@I#-Q#B(I}%XYQdpz6E(* z<4~_*1~n-)%r4%63wJMKH@_eJe0`t9cy6G_Y#2Q3G$8tK7B+7(!`YMn@o)buG`GB? zM4dC}k>OsAbv%!gT|!a4m87~fk!&*K$kHQ)PP<1?Z*3%55_im<;#o%n9|+5rV&LCI z9CdDEf4}@;%8Ng+)n02ba;Q7Tj2eMwQcBD~d^EeeLX!P+S7e=YIuJ9x6a!;)Q6Q&> z$m~EeYxh9FxE(xGzY|IORq3N(5MBB2kHB2YhF$8M#?nM4vN;VyFl&rFf2X&T+GpNn z`Og-9ytAgda3hX)uZJ7$M%$oS*t?UveA?zgN-T%e(!Y}e|Bf8avqb1hg1fpT;zhQ@ z#c2&r>lj1oK{tiJ>!yv@B%#@zjF}poH*dHcCt`n6)uC&gA74uwb-#1`{B2yY&%!wF zv^a6z403jg5c#kbCJ|P6^q1$7GPkkR=TYp&s6cjQ#Umu9#X*GcV4v9lgMRVsx=ue` zP}`0;9Y1Ud^1|40-1{*$80)g^AduXND$X|8c6Z7? z1)t3i@Tz_lxF!z)+s8tqL>7BpIw>o?gsj|?aL@EGbbfpD-e?pSybQznZYP9v=2Pn) zKXT<=RkM}4EL}#KEjjWQZ({gf;BE$B~wci-U+U@<=2n9d(e5*Yh z*X739wr1!Uqkw~{Az1t*A0n#hSo)p&i7irSgh?W?cNq{kC*vx=104AN7Ei6qv1Qq9 z+}Ub}Zu2?V{Z9e`ej8cv%X#eO46tlp1$15YMc8N$SpVy0H@5V!FAH0kw3RG7aYKX! z9Qc5&yupya@|$+`e51E@y*NEVk?k)QVdmf6AZ9chyE(TnUU?MzbytBEG|90CGCY6f zV}Lhre$&GBPRw)qEY`&tHWdnou~oo(3WWj$=Ol1$mp8)lUEyo?2@R&7a3SY8*5yRw z>;5!^4UI?tBqdY{+9`uG==59V**k~td@j+0Qn$%i`BMdX<%-xh&Ke3&#-ZiTJ2Fl< z4~4)>cpH|E1?ShIB#L{hN;I*~s0n^QALHBD8*u7OM0Q Fua;Q8xv*9vP$mUnTD# z=V8!{3n(}f3b)2coacKPeSIPCyk0}ap-|XItVWaM4_vcshamhJ?zI&{c`(mkS@Z5b)Ak z68DH)@-9%&e?IiP(-?10&w!k+9xQUQU?h3~chpwH^Y#{0iJIY6oh~xA%|r}$sD?gJ z!EgR{G+8o;?Crjg%iuy96LXd-uV2F6{tPI}`9o`$im*3QSva*wT6oXEA1@+! zr)cpC{L>{CeOH~iEEQ!RUk9N5>LKX8<^BHb9Qd0BAYs@|Ocl9_$9o>b?$9-?&h`>) zO)C@}*#25jddUVpsmpnf=_o%B7vcGnAkZ6Uh@^9u6kC%Agha8a!IteTkN^k+KJV5Agh7;uyp(*2BINlW@?g3H~phV(rWh2(EFD zdctGGPkDk`ktxE16-L6Xi_L`s{>>dWi?jEh%EGKK1NnaOxcpx-3@&l5SwS=9@%+Vv z;S#u}_k}*E4bW_*i&$QH7ZVDv@$QT!=A>(3i?S*_#ltA=Q38FM=SS<@CD@e9Vk|@Q z7b+HVW`@*8X=h!d}{$v4S)lG|BnPZX^^Q!6^M=Ow+eS&Sz^(^;(Aq!9S_?VJmGPTS@BY)$#ek zEGX7{;qyEj1P!-AOTicF`X$HOvIj9!tB**FQe`Qeg<+Cz$kw-RAxW1AIvaV4X5a8( z5jMveTN1+VsVT#F-ZJ#)*+QwQm~FGX!jh6MvAS2BBT{Vxo@;?K3nt*4|1&Z> zTY^txaxr7yJ)ZqKOyT{}G&jbNhVgGsQt~w<&3TM1u?h%0JqB&A<003%nrX@{U_CXf z*!T`5O7_`KN6DS;4wyqVM}V5MgjqaWQ?@S?ar(T|y~i0AL%m>i+ZRJmE~kB?F4Cu2 z;gtLK3*~=N#0UM6$oc+|rW_E1PS;RuE$2+)N8Ib6X^tJ_fsVXY$lALDO~LYr-qlX? zx?a;=>3l5LFTn4HY<%3OkB7;_k?3rR1Bss@+trTfv(oIJ@^JRlUYa#_X|MqeEgX;( z;L>q91P;uB!LtQ2ikTw(RzGcJ0_Ai|8RHJg?37m26qru`s zaHnqwdZr$P>!cujJzIl+VSK*sG@NII4KOFAn+pAU>D!I55b@(@C4E3V&kNsFm4kJK zF04%q@Z*pHldx4}gVq0`cXJXH0}J5DXN>lap2&^Zi_W`qaY-VM&9%*Ci`Pf6d!IUS zPxA$5rJP1gB%cR8odPF*e;86WAKTrgVZ|_SXx{vSf4{Pj?>YmUZ;s~uR||B;G8i-{;U>9O`!s(Me=AGXI zk4v0~Ft{2~<5RJ3z7CpIt)RVM35&f{*|8@kZ2Z6|7H8drzxOhb`=9`|p5a1~%G1K5 z-i5-FqZ=VAJ`Xz@J#b^LEQ_pHW7V1>On%FAx?u^NyH;B~YW&}0x zjC;`&?CLnm->^GiFCT#N#dS1LQbQGHoph#JiTk#Cu|7X+XjNy&hv)z+ zN&xd%HFuxuC&Yf6klExV+<;?z&v`~EFMT&VBPa$?!NJ&zZ zB@g4d1kDv}%q0=_C94x<{?hE}xl+2dIgOlflHLqi!g)vsF!$OnNW9#Lqla`bZu=ZG zu78G`!DrF4{xNQHM_g*E1xrl}WfP9KlTs%?&tAVKgM>u(F(impwVz-+{>IXY^`kIh zt_IJ=3=__sBP!he=r5}-kAeIC0(2-`#AM4ze3lBvZ=NZ7lyMh&d&(ety8wA(lyJ16 zomLL0!oceRUJ7oa@XteB`{)4WqAd_V7KXbG+3fMiDpt`TCLEU%g5w7~aaVge*!WNo z=Ld9bip4X|Vd@*uV^1Zfu@>oD$Q1d8i)GEY7vTn8e}=wbI?7($f-|2xd!+Yb;_h6^ z)o-Hos{_1$I&{)#Xtb&D1^hJwSc=dUGQv?KX(kCL};G{ zJpI=5dE#QM5edc$vy~{E8Vx5qWu`Y)g0*n=&A!Vah2wxpzOWdflNm3vsv~dqdJK8Z5?Fk==(RqwX}=yX0fd%}Dfr zdP`-pYH(X=f*Glua5B|q{)wv0^=|>nd$!}io}-A-dr1-3KU1vR9Xg$M4Hwp&!x<^g zc$WHu(4mcZdQp)9yQt^rHmV8mr90MoB$A>^7CXgg#5{KhE?V=xtpWn}nBw|*Q+PJ; z>{+dY(8bSAxL$Ia@DiW%Zz(N8g>o#eu41guz=}QnPHavF=i-EAqbd464%L)E?MDf2 z94v)bs}=jPhS-*inoP~SN-)J!f)oxakhV=Je81;HLc9tlMr+{uY7P<|O(CY#2%oy= z*fs4YhVa>ikQu{QbpuXH&Sp^_+U%|?&p_}u%Lsn=tk@Whjb?pt?7WFijqP}7?u1zf zcOqRo2rgoSu{l#5BY$Z^PTmcxJvYKFJ^=YN#j#)3oViwjVY4hP$)i&tA*#F@@Wjakgm#Vl}52V z?q>Y%tjRv{OqS)$r+m-#8ScUlDC1eX2VS-)cH98@&>AHCR|eL00|z2v5MLIAQ7Ky? zCEtnom_I1uj_XxFt@(D(l>IYRVj9Z3h5cuC3D*T47pg>=fZ5G}?X3S08OQra=)>!& z{TTY@7~RkKOg{oR(}!m`;_q0ZbJjeB&y{24txD`jHlG3Ak3sf?D2x`}ht)~SNOS!~ zCv!51evDyUN5UdiX0VVjQLr5d$SV8(s$rAO`YUI z_tJ3B6goCX65B=%!q31D)F}Cbk3D3t=%OlW&-9Xh=m056w2^znB4PYCN8zK%ON7af zGEr!kf+JFaFcm8Uka>E6>=7iUy7I-kJa#4~u*9e~j(`>=M|E)4f?LGY+Y2vo?z zjbn189VbU0Hi^;w{1jwOJPo$;1SY*;cs6A!h8%Rpp^-^gG4~i8xN9N)#7L-z_0upT z-X9RF!RVwYJUASM=JHsa3;z#qI@FQ7W;u+cCSgpmC^BB3f@#hPOm3Y9_uA1=8#+Lr z9~Dw&+gxA+`x%hEdGju0yPDr4Oi?5M*h8Ld2 zF5xc3BZ$w`N73Zb*c>R1)?KMszBLM_#+fh}eiYkgBq7Km8H(Kf9nU-R+U@n25ITr0 zz4Zy!2WxTGUlHv~Msp6ADyE-GgUY%+D8K88t&@(SDOHGhrIXPzb3F6k^%0x4mZ3^^ z8*+_&c}B?uXLg+yico}5_rfuuNhwefZ-Xr}%n<)Wi;0hGL+R~SjJVuMCGC~8ew8=h zUyp-Yy&krlR6yvHa%6Y@J< zG0fmNuRX}IGb$2Ha7>c9*E_@DoB>qkY2hzB0u~#Ft<%R~n<#gj-jHNTxu>zv-4WmJ zIzz?L3U;~qXghKqm)9T1vR4K?_sl&?^F5%au>rDA^^g`f24DR`=xM_#I`KD|TL0;> zg-djp#Ph*SRmC6Mbp4PLw-?i8KB6GH9Dk&*BXENuD-@_P1s`eFFir}8=1ah+P=bGR zW@72@W!!@^6t3P;=&VbH^@;@OMjc1ae<6Sj_mqZ4)5?>4w?aEful)nDB@J{pJAj^b z)zXu3KWNfH5lnd1NF$#&Q@d^>2J<_kFg)In^Fk)G^lS^}QQ?Rs zYyF}0V>xVp55}qgWZ?UUXWp8}VQ8f`p6^xUp66)@s^qh|2lnWl%D+E#pQ-F^H@&!U z98BsR*gmLfpMX*o>@bX5<{n7Q1|+@#eLhvm8Qdq2F=X z<}L16i!hUhFA(Fry$u&0;JT+R4)6^0(Y4NS%;x?1Wsf+o_8NR9Pr#Jl!#E>d4X29o zd6y>vkHpTx@YHzrb*DCq96g%-JsCs$gxrtGy@BIbMzh`BQS9o)XYq#BJ^u&;FXq`i@4h{Z;?5&tbnP@W~)VS$(2=&bv|F zzYd1(d+_$j6J%*ULBaI`>riC;nl%#lI})2pL^EW>#$w%F*s>;iSq`N;r_0T|3Bh+oq|6!wSFTU3O(>3 zgAmckdog*EEcRdnyxkJHkLCm|8s$j3P+q>k2oBJh%v+V--e(M0AJBHyun!_$8OlH$ok7S>RT*KXxo4nig1SU-j zVI9eL;0jw{t-cl0Dvol0$T1kb(xD?m&M%!R~l~II z21KxJx(NC&U*!I%dl)t)T|=$A|J+ zA)lq6x=d}>X%xTd2fm;C2(A21nDBS+N97Q#P4dKbCwI8kIC4*=HDq?}gXQHY?5#cn zUE5-0$`wOBh0n!064{dG6!ti(h;^>3r^o5{=+!aqe)u5CE-oF)GSBL**DVa!VK#6IELeyz8G!hgz9f~_%!kx zJPwXw56u{iUkPx1^)ygeB(u)i!=8K!WJZNCRDFCt*?pWv|JlA_7v<`i=*0U>dDBB| zFMq^$^LLS2se>F@dG1pAN`m`hZ2eeKHe&fhyil9~>+iBix;+###nhSI&wmi#H-uda zT7|<6YcV#$8)ZB*;T6A!-BSo;%4@wjQ`QFQc6QidU4&D&VxX!U%N>m$G3u}iYdk!h zo!(Uo>64sE>|YDBI6f0<`Al0+KBCzsFEQ`)UAXZ-e>Eut4SaXL*>Wmsbfcm2Z6DIF z?n8OO707BoMg5%9DAMA-_MLNJm1K(z#+=Ds{snV*ucX#)Bg&W9;%e9o=mdG=tA-bB zyEj4mQXi}&ThX#hiW&TvhQ+_eLbG%vykl>}^YJNE@LWrB?N4M)`H8DfnlX{{OMZ@y zhh1nslAc%6@qr;&WhjLK110us%s2d6_zgjevY}#I$nSaaSk8GOcCU{@KZG-|H0t2n z)BuU57U=Nv$=0!z?^<|1=A|XR#;BmqN*N;~Gx1k68GUu<@UqDR>&Nhpzurt_S!?6| z4N+w8lEO)Y6?EdyO8W3bqKV8XV*<^qRB}1^Fcbu@~-2|b!oVM`l zwL%DG>v6-Y8s~73=}q3qvQw5Y`7C2rGHMZDPo1RR!FN}3I&&0h|!I;eNQ1ZS3Q0xds@$Cc*8tH_+?@z!% zI|&>3F0Fv)sh2M{WMf&U=}^kDhMb@uY06^kJZJw;T(5zcbZrb*wZ>V`VDz-@ zgMHf}JbQeZm3+9y%2IOKw-Y>jKc@;)C>ulScjCET7?j4wb8h-9NX&CX-BEr=oRI;K zfZK2vd_b+!DX8%dt!`-y4(>5yW(%xY^Kv`(Gs%RFRd!;IM-Q-jH4~Yu*fO?%-d=V? zWf*%pRD-R#H;q+=K8M#SG4|DFBs*}Y5z9W_!j^CE@WsLy3U*cyYXiD|Td?Q5ESa<4 zB(_`YFE*Z5V&Z=d*odJwg|79LLYTZ4epzgdM@OvT&-*les^ZM%>`&0KpU7RY2D>x^ z@TVpLPjVJuALm24dpW~vX(g`jR$@`z1}y(mCe{s3LC*2(Ft^~${C-W&CUM4i+x;wk zQyl9}JHon`9zyU^-rM(n1POlEN;foOB^p+&Xw(zV|9_6rZF~mnHi5s*maxyyL)oB_ z;zH>YTEZn0rU}j8x6z*sZ%K2WB;54(u=Nv;^UgyWOUn#kBQ?)3gM=&0sM8yUvXfEg z&CohQf+;3eW8H>dSiEra|M&fW3}hiFSBaI(GG?Kf!&$(+Fg{#3$0{dgu&`tOWF_56 z;mT6Td0fL9L;Bd_hjPM82h1R0!*`E2)A0V9itv%;C}HIWBcanx6}Eh@1QX%Ee>+@1 zlEJpGq%&KCJ1pE-q{VD@n9m&-i5@^YpRLR6jlt90?d*zTAiJ^a6x+$!m^-3k!9r5_ zIdYE;deh80UVLM>Of|7pZ5pQO*ki-HD>x&34%;Dv*>%g|Y)tqVX7Xz^?_OQT2#52? z<9+1Iy=!T(;~Cmp8&A%c`To7ug-sdf!%9vhK#{+*?dQ!xf8hz_+&G0TCCQld;T5C@ zKfuqn6#h-PfV}Q}T-s2DZ!5LgFGVF*Cn~`VOuyib^$%3?u9bV90+W2C&Acs|pEsm>wfJ=^jK?@^4s$9J4bIL09WZyglxV3&ws}80G_?i=kNAtjk*3#;`%{ z%Iw_Kbh`L5jlb_+(p+74cyvxf|7y;JTz(n(p7-Em)r{<#WV~i2xV%P${k`phGi$fv zd;M{IX!}Qw#-B+0LoaFZJjJN)9tsMUgHipi4#=zZb5^AmYC{iRLQdDm-*KPqf}u;2eP6eozVg+oQz@gvRfS@aP1CY57~(?ytB zrD6KF(|o2n1p#qJP`}kgZS6UD^FAJa-(nzhbRDK{62SD|JRH3w&X)DP#=T9&7?sY? zA^)hp6jnh!j~uk>}y_HWMCyF-KYWLVSGv0V==Wa|QKNe7?GzovB>K z+8j@^)Q|qOzU&aawy>k|r}H8ErV5(H(oAOjd%O}fV%Fs{lt<@MSm9IB+VGXL9c9pQ zX9~_gvB93ZJR4U$0vjKwBSrcv{`3yw<*-4_;Hw6!Ptju1k<;1Yw~I;Jayf0BagglW zHnRY2Ul#kphw0nsa*sef9TH8Z)Ps%8S-gVT|EOTsBQwyw;3&#AoW$%_c{*35O{OPB z=`4Q}*_&5DIq4d{D96z5W%1ORo=V+n4`B153Afex8Gl#_zNO#j+JbUw(VxsFdsr~{ zpf#*|$scm~DGK$M!?0o_XLs{CX?D&bY@U1+c(`BLm`VZ~t{7PPVYNU7s zbhgr-q^h}-a%dRLKb(d@B$smLzo8@Vn#k_PPm+>VhpBo8wO^J;aFzrt90y~&sB^JTDH(E!W;9%A?g88)I!k%{@Lu=VDf zATf0+epafYUBMe;>yBf~uj4S8_=S@9aW0sy7)#q0^Hgry32vA;%$WF^uH(waHyp zj=Bc>()j8m8gN`gVNVuu@9hdcPaX^HO?F6#*@P2Ic?NVwEIs^@Kp|5PbB<*Qop_c; zzasd&ep4!~aLFTYqg(WE_G*m$G#B4C@Y(e~E#ZlNd135VDd7;87&4g4|NGTc>WZwv zpMm#?uxN+PY6CP~GvGN3161A8#B;OhSQ@kq$Eq`-+A%>x{b51i&!d7K zyHac~Yl4FAP*$91PCc&=)0LxfwAI-JuZK>?j9;c`JtTqc=k;JT>mYW9Ou?juCfK>y z9UFXq(y@?5hx z@l>3xOAPa90zF za%0gi>VSqPoPmB&gcerpCD$+KXoO5O<=2PPh2#QKivEv&1^ptg*OM^nrU@or=QEy{ zaVXobAZ(M@5MHUV5?*|gLbvMkDEUYtIf&KMmMibcFGU1vMzqk?dRf$bJ%ho;C#iey zCHkeBOpn`lqTpW`_Uw)Z$vvUTDdKo&I26C+$J4L*V)X2{76n&TVCc&)@E9CeSB0yuM9p?OYCS?wNJp#SuA9ocZ6_y#dD4(yHSx)O0k?JXxw|2a$*a3pX4FR zpFT&$EV1gHyPK=XTy4YouG8+vG4 z3PzXjK=rr^+PVx8z`5dk66IhT#@}c0;+X1w2EK0!F|+;wCa?TLBc8pWliU$nG1!HB zO%`DKriG{#cs0!l_K&y+I4g!Ug7|$*DvDNpx=4D5GwGu3S&&89Yp{ym(pZO20 zW+OEoBl=ze8fM&vzP2OIJ>H5_F9F$93AQNj8-5>og{xQo2-bWbMiAAZ=jZ1#<&0Tu zUeZ*i`b7a}P1NA>M;7yj=Aqp-9Wp=8^FGpUT$<#Mv=e7wZxoESd5d(a7EPgMJp~-UCVqgMn z3qp^EFXEeadIH#o%SAIblxLyoIgeni42EBq;klenEVbct)z+Vg+?$9SoI$U1n`bnS-4bq4EEZZs zUK5rb--O)Wowyk9j)Gekal|VZu4{83H)=NbrFmhGVH?yh&V|ep&d*Vrh_}Aips$q+ z$HB1(QMiH*WWi%h3NF8mrOGP|*g8XVNw^{FFR_6VM<|L_bevl_7QN-v_?`QFs} zG=f&h*W259*<-R--j~Obv3lR`ZHBpf2Z54W7yGw2-f2e%f1>7$An$8vAv52 zolbSK@!=oY#}Qvys(mxYW;G$q@gp?bxqYT~Jr=nKqJOjt+cRqsyDo3R95(Tc${rD( zj~&eUZrC6 z$9${XjW-K^VeGd*5c%;1g6*R0obNBFKKl;u%y-nhGmrF|uF|aMN=)5To(-xT#a22N zp?rQEvK_bsB()iV_by{P=Q6mae8%hx-|=n#Ys`DG8du*0a>l|j1osW#guDpTI@5{W zb`!8?q6GxGGtpA9lJhiuAy&)zS?8tL`cYDBcaJPHx=c7M;Mrjd{%>}B3Qvy`-eoI7 z?d3C=W|xRjaH?Qom#3RTKalD={VDXiwZtHtBk-+Xl@93{tYu5QN=$!8d1 zmy2&br(x861c}#9B7Kx2vpwR$+y}d|f*l9YZ?zJ;RMw!Y?>G*h^F#d7E%=<_3+;|uz7*ld%N+MJH&l;mtmRu z8w%MKM#(4bsY_QIr)6y5FR=hddxw#Cz$6;oq(^S6&cbO~D9`%)fh3E_Yf(0}Ov$DL z-(Qo9su*OhzNLW3WU5#5p|Y110+T%n^g84onN7b$GL0<)yK)Y>e49a6LUSp~bqM87 zIwClz)=$d{b>PR}2>}Vq>Eo$wls&MNbQ<-rUPBFQIQ#8q?_I$*Q6rjt^C112x&zO% zgW#_m2$2c}ta+!*yL|fCyz{!CuY5iol1n44yacF!?nP;cD*F?Dlrlpk@ZL!W1wT|U z_+SS$UF)Rdmt|2}Ur*N8<#Fm{4L$X%qq6MBG-%Qj`WjYEHXb6V|GEHitIBYr+gs{nBu>5_4R**9-yAvw#bbHMF(wdk@=UZl*rG|-Lc>j2_ zF0yRqlkna@8qT{~QtH~Yy}*$Sp10B&y&-rkr+_)P)L`W?2`jtwIIHbFEzg;U^L1Qj^K>U?`0ZSxh=c$Fbl+5Yng5z~mk)gtyt^b-fQ#8U+Y4pNJhPgCOM~3dKiC zC>->K499JRgyDZ!am)l`um8s^+XYN#-b7}+Lx)E6jv>oJS(=#=3`6@psO>dFb^ynU&j-qd@q1hR{kLS|2oLaJ%Mx{ z@T|ydL!7NYNo({|>B-J4(w)~x&xUHV<XY z1T(z@R-i%+2sJUokh=H~DtCpT zhv(70+p(k!8-4?AiSDIw%M?K+%JUuRh2ki!zV{Nov>ZUJtV z3m|U35Z+M>@wRIX+#bgue_taK zY}HUi@;q!3#Zn>XHx@n7hVhLKl1_h4{{`KprrM=YoIVGSYgVE*ZZq0amSCd4Ic7@p zxtNkYz4h)840DX+97fKbJg169J57+9s0TlhS2W^>6LPXA<8x#U^{lkVmcgEw!x{NBL!yna|G zyn}cCFWjStP}zNp2G-Y8iSY%JNPmh%^(H>w=|bc3YqaJ~70o^Rf?CG=poQ<1Cae3R z)47Y5mwchjPf}P`^abv6BJ7&@Ti$Kv%FN&+kQ}@Niej8|Ua=98uTprwKOUd{1;cKp z2Tb?okiy7NisU@ZaMdNa{eCfLb-SX1b17O5meNRei&{M5aP4(0=v5{bJ>%YV{dRJx zeM$6O1J73}Ky|Me{+t)G8Vfg8d0{FOaZp2#|73Vi)r9L3Q5420@bwjEn#fm>`;4;` zm6%L(XS|}*bKldk#s<1mv;ZR>*kIxoE9^YaS-WH2q29C_xi%(j4$lLRdq0MMn;W5T z!=07V<+!ssjQ8x%b60&V8ZPkOmp12dzbSyGydNBm0?=T%0g09>kmj9?=_h-+pSuVS zS#@~9U1^5h@A=PA1!qXNAYa*9h!h*4Qp76Zp~s5wt`tWQcgJ2l?t`iK!=Zke@4Y`< zqu{bKnyM;;*k=luG)@N}j1*CKL4+;aUyET+TQGZ@FHY?C=RNN2C`e1B1s+GKp&*^= z&uOv~Y9rXYA(pJFN0sv*`)H(Z2PurM#YjH?KB1X{j;k>Uu}^|h82>$Q`USVFX7sdl zVn_aAW+<|cO>#?RCQiMy{LoV>^ev_c7cnOHsvmt;BiZ#2V4L@AF*S*iY;g8E$an;x zP=6_groKWa--$aIwIC+2o<=R@&c}IeRG#-1lf8JR^3GW1#+js@H4|85g9r22Vn;jH zEg_Y?Gig-u71p&gm9+_y+3v*6oQqaOyBFEgG0!q~srU&Cj4R{Z_hN zJQ>Q`ldc-Wyv{z%oVXg|$?o_+%Fa8k z$2aQZM2Vu(&d4ew>la<;{#p!ApOVNyTye0>oJMEk z=hNj}MXH`7LwokCQNOE}v{5mN3?DzCbx!$gukBryzrc?jP?MqY^@`NH_8Hq=eVM8> ze8gQiA3B(nLV=NDZ&B(L%^9gfrGq~(>zjYr6McEwSfNWh?K;@{?jek~tzvxBWcKP> zDJ_Y17vlZo?a%T-y5*fm@@Fky<&Vx=uaB{zwFgOMcl z_$O2Qw1oD4=wNPpl-Nh70J2_ipY9C4BXU`~^<--F*-@W5y+QLv3TZ)qNw~R)wAL!p zB=P^>SJgDOA?F%3&Q2je#Q-rjZ>0W@%2?_rJH1U?9O>!(-SlDKI@)S9kER)op|<5q z#T{5Tnm$#V4k&0aHCqGH?2Y`-wAGqv@3URLVFwkjnGbD7Vo_=yiIHaC)AN5OyJm zCa8x~%GGRIDm9Ew=M}RNbGPV4j*%gh)-tExBbk1hE)A-kLun^T^c);WDF%IMMp7pW zvGb*SheU2|=L~UHIhU$k}jgBw}~@-o3fBJ zJg>8gv6|HMP@mp@SxM2Aedw{y0E&3gm-McmAZ7KFR1Tp_>c3tQ_Z;BwdUz0``_Y=7$ z@}w!7!?gXxexy$p4U>7wn$3m_KdO|4@8(gYeZzpR{M4+V8PRKI`{e=4@VzGZeY+qO z4QGPfp_EBYz&8=qv+*bid<8$6VZ%63z8`zhM5oF=k`YUys~aS9oGh9YkGlbve{85tN* z`pma1Sh|3z$@Qj7WtpsYuMA7>wqx;4AK9%%&)DHPru43EGFiB~vT8jUp>=Hq7)jCGF``$M zVE{|F9w4;586|{>jBM?D*XhO1U@FlJpgDg;hSwHfmgMTfh& z4-tI}>Y3rzX=E$EliDtd>>7p@=yCH-w=uwyQH zw;9vQG2#wXZ)xH1Dn+5YUpUp5#M6#p4=6C=Hu1|jG}HPj{aGAIKPKkTmW}zO`g0h& zb8tJeZaTz%q-K(tdp7xRE~3rFy~)@@i)_J_Q~0w7H6O_KzV4QF~YWxs6)Bw=tQl zw@l$-e_A|xiXhoJS14>*Bn*;LrM>bx^g7Ix+!JM~K~>}nigCp9aSpqEzLqTxf6RL6 z%%QM7=<8 zuPOSm$Sdmv9Tw;FB?r>|Om$LKQlS&a!>CX% zgYst=QCnsDqB&+ly$GMq0PZ8 zKuJYd*ECANP(?wbemo@y{ztP%gI@OXr)A?V(YAYmB13r_z4QyB{8_g|j?izadG(u` zDG}dcmF_nji9!4%rR6MJltn1&9)f;J{-{DWJ z&ofnWS*1=QS%S2}FOwKCND`W~E=Ge!il5oVgi6|w^@k3;Eu~k#MDDwkzTg$HTi7+% zhE2V2g#{(vV$FAt2}+hHgE53@NDh1+yK${8)#A0S!tu78$VnNwzVBGS-#T`D4<; zhSRgj@}ua#-n5-uugqu1nkO;$>H(~~?hef!6;C%~^U36OIQxQn79iWr_((;{d@_*2 z<@(am_i8k;QBvF~8Be1FmWe!!vy}JfHf>NZrDU%Xa-N(*i*{_JOKCf4slg%o68(_s zZoQ=gGonRSt-MfFKSmf}qafI>JVn!!FOhocHR>1=Lz?3LNz~yiniOI{*)1Yt?dUO4 z+ZAV*2RV_~18>^qxqvCk#<27HNsQ~+(%91{sYv9`yeUg#jK@XJE#{J|(* zb_E^$dU0M%l&FeknaaZ)bJ-nFwKWr1aUwa7`$LtV1c7CA&5i?5$fAO?)IWqra<*9GCty^pG~q-(g6dv2mmkn@*{sXSY|c z!GhOOY2j?6qA=y3q!9Z^S{UN|i$V@Op{3$(T+oMRu@7~SmOpT$GxnZz`k4i}eX*yX z9;URgY!_Wvwuz*hsavcF`(Hd@KDVLgLsgOdu4oi>4_d;dolMJCe+p9IpdY9%Hn ziK^ERka^(|TJ$lE1^DK&Wjun}e3&Gd#!nWGewr*CR5?f2_nxN8t*59@L>g%gizOlQ zGP%dUrH_+~X~+9>qOW-&<@Jf8uLDcz&BU8RSnz4VRPM51)=!sW!?ft1%THDuD<{l8 z(J9XE7K(eiU#aJKKfze^0XsKcjn24Dr{QboQbKGb@i%$2d`mW^O`1t_Cd*Q<<3q^2 zT3X1-{6e)NmtoX|`NHtti-bze)xz~>zRX1Z1WVg0&NIw>C@fP-75tL2gzAxR^)d#< z=v6E()YIGWRv3J=M<_q=KuAmrqBUxzbah;jxMOvIyvMrI&r7~!G)i7@8loxeO&l$p zyY-i8`t&8MK7*)V-87+k);z&4jfA~ZMLwg*`IXK*LE-6Hf{9U#P%9lIquJg{GmG==<$M;yHSV zS`U^GF5V@@F(MPWe{dk^4Te*jq<%^S#Cfit*}$-xX%1G)$QE zRa@{?kQ8T~f3iHcesr)@n)Gp#w(d`(>2o@$c+@g6PxPU~*(!7>vWs-Y_`H0cj4;I9 zlZE0j=H-&so&!e)UwPJtDUgWa5kPUj# z;?|E$!6S=Z`cT3qhgl0)}$#i!%@nx&Y$a@Q!#`sa+|{X+o>aLxhA8MaqDQw zEYa)zR3E{0v5FuoGHOgTQ)%1|MIO?EcaJy8 zWCe&jAc16T@`85VsG+0A59reC+a&Ry$p4;mo}91C3J-q`6MBiWE0zlR!Z7DJp}5RQ zIK8@!Jzv?6o>$4x3 z%xxvXeAow;)~l3lI`xJfcGnPQ&6XB=E%GNXt7C#fpsUan)J0KOFSADx@obH17W3Vj zFRb4oa#ybw2%8l(nTO>Mru$i-ileQiFnm*6&BM+mgg7dB2*BUT{h)SotzoRP9{el&!3AAMnhxaZw+#*I3DFQS?45khr- zoG^Z)rQmbq6SFZ@r2JJ2L>)L(^q0Lx<)bnw!siEFeA+~l<3q@#$c~oT`B5*WIGSM+ z#~c@nJ*`9q8hBlteZQ`!W-yfq}U8_v`Btm|ZPqLP-Qi}pK}Qv7avy7|Lr$AAwM4!f&Qq=408Cu!VOwI{n z4SF(%&FW}lZ^a%-){5j$5==hGsWFhv?;v6HUOpGSXYiA?23RjFmD8l^{i}Gn@Z+5HFFz>!}^{q1HJCCI? zvpJn~K1iM~&(eZBn`y?6eoW@~V>XpUKIPxhv_+>c<(ti%BSPO;3#>uJbxf!^mElHrYydVF;NTPrE*B**U1?Ff-S)-5t*J0G(dAw|qrp%1OO zw3}sI8Ae`l%G6+)Ci-0Eh?i`4)HgO! z+%t#_Hlh=L2WeN?bcz_J%^IpZ*>bZ|R<@6eJqm5H*E*3t)g+LCxVK;|bCRxHJwcLP zH^`fxrUN?{l4gM^z1}6}ck&Bpit$YP=e&{zxb&t-g{d?*K#}ZR8dyi}ANKFtH&#a|_`|NgX=a9JB}r=DDOxXi zhrW&LL+UkiXyxBqG}bVLWy{1eE8RjCFwv6|=Ix?pB^^pQE^;C(bSY94_u2oFIvjVlCS+jQ7+QQ;*j*UPoP*+6bHEzI{lrY%J(5X*D#Xb6I zbS(D-=}RWi+C8$pVOwN50^gZY! zvlua*zK>HQ#Q`^%*`dBPR%bQk7g$oZyp-rGp+>#>O(sh-H!8icnQH8uC``YD4Rd|X z*2aHk6`dl__tRYR_t{MP4@>CKe6cre7)7z?U$fj(@0j%8k8Fbe4%()%kj|YJSpu0h z^!I})U39Uf+%F2$?LLm)^k*bITFcxu0+>;JD9avnl5VWOMrQL4)7*IrsCUUBDl`+l zvE;w7oCI-h?#Br7b`X1zB_ZUxB%2!am{|x6+wKr&k^+4L2-;0cOey}pSZ z=5E0!w)*HJ_Tj`+X8Cv|$%tMKS2r!A6$4WBWQO|c?LXtD_d}euop5TX$a;{btqD2o zqI(aUA~Kk71o*J7kg9Fv*xSLX0VNfhL=o z+@a_8@}FpEY)jLfqKStV(#^0~mgV1v^n&-1^sf-6uI0(DD}^&Z!{;4TbYr;DvyC3RR$Zn#E6Bz*Q789F{j(Gnii~#Wqaz*vc6vC%)e3% z^M?&V$m@wv`qjf$?NOx_b}cMt$`qU>6bxL$}h7>JcqbslOuD6NBkBxx^lxGk8#Cb+D@wk}o-EN9=38wM z`d&FLthj3?`g$H_D-Wr&GczT~cg7zUc)2gt4ZKHl=ZZX<91JA59^ zcN6o{DZwmKDU8ki8ONLkXX$wxj$>Nce%dtlpbMzF>U(wrOeL^qz*kq>OZqL??yu^}c#IQdqA?!l;4wf$=?yk>h5Q2mL z2}aBMNl&~uO10O--o5_!kI}L|eS7timhSa`eung-UfcFt?6f|xeTwBahiy|ZU61r= z+B8i4Pj4Dqq(80K#{d7%<0NHnyG(`aDr;~{sUduwj2a)`G?<%n72af|#Jif5c)wHy zuCaI^KX_1vM_e7m_iU2pBYXAZ`{(rM2A`$*H6IziBz6$DZyvP6dCv|luDer{hbAuT zUMTW$+JpHJWhE|Eti-MAl=zVtH9lmv0vCObd2{hVexN~tW26$_vr>t_UZc#F=BjeP zAw&4+e;Pd8W(b%5t;wtUjpm0HM(}2fQM_NtNFFaUf-lq@$*UAJd40(UK15+SUz?)B z_w-lisryy|*~DRBLzgZa$_1#VcT$otGx z2J`NX zs{Ek85)YlM${!f0b4mB1{9xoz&O?TAzm$=Ddbk#s_8G+o+G=v+Q{vxShx2|fH2HMv zF?>hYD89RO1piwLT zC5G^yzm@p9Qz|^~m@2R9tH|#*D)6X0B|e=g^OH)Byrk)rX0|2+6CeK&sp>McIWj@hk4?AM?PKp7@yna!YB9I%N39B z;hv$pxqs_k{(FEUm)P&bt%o>s`}|{k`$bp2ufH2V=jYC2q&@j$9S?50(Vh2Q@6Lz2 z9pld14)M@}{d}v_ejXFNo0rts@TToM`MQlZoNiihzgpJsndk6m0 zU>|>G9^eZ+5bHG+;v`jWuphz zSnbBW<&W{-(MR|W9}m8@!Gn9Yc=E6P-T9to4_@kjoJ)%t`adZj{~+T4rB%<$qf(~t9%AP+vG+m)|b;l@AycI7V;-FZZ$2hUvO&bzN4 zeKN5G4znkvLonu{j)lConHQj?NzwqLtH=N*g8+>@5yPkZ{Qcpg@+>LMl z4izFb=Q6u3|5Cs z&|A?M_caVK`}$%?bQ!^{$Oz}Ox8TDbQ)rH`M4uuH^!>9QLraa(-+4XGoi~NXi>>&u zcRTDGEU;B;2WBPjfP~j}r2Vu*LZA)Ccf?jcBl!l#@*hQXwBb&;H_J6wqOI^weCQlUYpQ4-4f;ACXhR5 z4EI@Up+0LJ3`VWT<*v=3|JGyk5=+E9w?*;eoiJErgH^5jkd<$T^DB2@!%`Rgo$mnqU)vGBeLIqhOd&aJ3zk0I1&J(k z49>AYdXOcQPS{|!pAAfh?8AhQjyN>S61vy7UA&^^8f#{$feerqd?dzoR}+#M+X zXa(oXTVOnM7g}#yq2t{y9BZ~ffBS7{ZMTB#T`PW1CR~dj*i~jF=>=3j&3wX zY(G<&ecg)fzs)hLdne?5Z4pK02==$f8d(Qa&p(JGS9jt1gT3gPV24MY`@tO@p}f-` zVJ^;?Cw&kLyZ3>Out#033*t8&#HeHYV4mfGKerFVxo9w0s61-c48-;aGN@c94^0UL z+$|l5#!N*_AT?;2DZ}};GSr?8#uIlH9J{K3ifO7SOIE_Y=?d8GFbr?A2E$8E4R3!c zV7~Y}l2nG#a1H$4qXE5tT6nNY6V@3TkabtZ{6Wev{xlR_#;Q2^Oap1FhT$_Gjct=g zV5g=wMvYN{=lCI*T`~qMG=}2nC=F=Fs^Rb>Wqi?8#?jHkFh@!a32&9rzj_4Xdkum7 zb2Xe@sR1n;bvzk01g1X*!(LeqvD4%clsE*5(^SxOZWwx*s^j4nC0yE|j?ykw)XY$W zO0ha#w#nh#8+qt<$>EE#5;Dfi!^cDp7p@ORw}Ap8UF9$?P#$6124jeY3=}2~!teBf z$ox+M)vpwxv{M;NRmJz{j~brp4uO7xI`rQvpsQ03*KZF(c#j-599Dv*gE~5*m9Q*+ z2zu<)QU6C3zZ+FCV1g!6S7>5RDqd}C(MMej5pyozi;MICMG;P#E?JuY{>hhujH#I8S{DmoA{_W z6CUes&fl#ui%9>!cv$B>^jH{y}{EBRV~L!SJ68HcMO*E+eHlhb-`GGP_B)>+S0 z+cxpuN5p%KHsixXH}M%4w)14uoqXZav$AhYU358WLN% z)_8M%HgGHdw=KQ6MIoGbT;w#>4 z=VL8*@TtzWod4Rvg z-nqk!k1Vs`+5u*K>Z1+()HYMT)Nm~?Pu$2|hHT`sMr`5V53J$sZR`1NT_fITx{`;_ zG32xF8FGb=O}yrk5$_eSj(0n(bNdM*k^Mjc3dU8zeoGOa&o04axhJSt zm=9N80KFT92>wt0+8_4lFDcn?n%(-3(&3oq~7hvCr-44U*1yJH@} z*d+&BrWL@k=^@6YJ;K_*SxCH5gguUVnE&koY#s8j*s>7Q5{nVH<}nstE5R@Oax_*I zL;YwCR_>}s#i<5tcvS;O{Z{C-wZU*e8yfyLz4~C#yom}-8b^_<-sGU z-6_QI{~luV>>{)#72&;cF@mLv5oBA8_sgH)%e6|B_o+juT^+O~Dp8?Yik*hlaJOnk z<%cF{9IJ$9eFGNhm7?Ht36gE9u_C+*cPvWKU|52k>&vj@RWVkjmf*nU$Cz-V79*aO zz_Gj$Nq=gvbY~4zf@rgdUyw}}zaEY(Ph%1%2H?a%_#ubpLF2z~T5}Z~l z!7eJtL*WSw>z<%{vMFlEm?24wSGH;!aNA@(rad>q#%r6g?W!rF|METh@LYhSgyGiQr7FR!($EX0ym*p%sn!f?8WEco$$G23;!H3 zmdiO{k+MA!2RdNHzXKRF#Tn`|_G7cd0W6!b2ahLNBXr_!^c>oObDuY3lD7pejM|8j z8|zVc&J?b9Eih%O1N>VcG&#Y92K8! z!KUuP`oA{#b!HdlT-}R{CiWiY+CsW`AJ+Lh;-!-Xc3-xDoRc|>j5k4B#tiDN+n}2p0<~o~ zG06Hlg0Eb`eZK&t4Y&lU=*w8PHw3=f0a%%J19~Gv5Iph*J|B$2uTf#BZ3u^+bOJ_w z55ujZa3pTLiM4(=;ZS!Q_YPdg{C(m0{wWf_PDUcI@dh;B+y=jM2l{p?XtYkim)?nR z+K~qNr}scM_c6RB2|jvvkyVtAxf9ZmW_1s`4tF81mWjE#S@?S{1*L3O%}U1BiE(&mk&N5b_wdRw9jjgLW5R&ryrtTMJ9~O?D-T!hyYn#r&JXdF z=Z^g8#e@8+o;!ae6h;@;}3BGKm4|hK2&d+~${;q zHsBmTvG@${3ctY3KAz)83(oPmYtQoJ^ZvZd=PXy$IKz+h^W*<+p5XO=ym{+p@%oY8 ze6WQVzh~ghqpZBSo6!kgy6hCU&Gg}_QQrLcKX2|-S4333C$KQaNg36xh0Jlo!f!$&nwZswF0BR6~pb;1I$1F2=}fRpf#@u2No2dZhZv? z##iH5XDQ%NhHZfzaQah?p`$yHJGBBIl*?eUs09zY9$`duJtF%Tpm|k3K6^bu*@R-0 zL{?zKktgVS{|JLZv+z{40t1#k#_HlMOmlsVm|g|gW>bujVmu2S{Sc2ND=_|LG1f=i zgUy;k%pG5Z$fz28`qzx#tQrc-8nJPBJ+5A_!S9jHIMe+ME0#7we^o1fm9!%6Y!yBS zKEp`w)}}60n^&VgtsbNO+aal4g1g3LDBIEs>lf{C zwrT-iSq`_o-H6xthUTO1aky6}B$vI$m?3SDZF&ml=Z$FE(GHoMYVfUX$UjyO6^Umk zl5E8I1NB(-0G6cxY5G}C8LJ^cZ5dUfMp_Dh^u-U{O%FT~HR z8MQezxN*K2nz~Q0?od589jL-OyK3~lQwGi31<(|86z{?^(33*6BtC>rM-kphX+g_8;qEU(o=aC5^c7 zy8*QsjSwsv;bqj0cOTlY3 zhaLF(s2!VUHsjj9R+yb=!M5CPY@U<*bLw}njQUPQ&(A;ZlP%ZKkrNZ?*bp0~rFhCTQ*Z9lqYZD6v(8PCLj zM{>eWo%Xm@;D`l1j`(%L6Jzdq!(^Sv?NvRF<V;W)-q_}R1hL1DA~VPV-Mj5jx6}cCPj|!M{(kIw>j;TDSLn<=fMdN5qdC$UiwzD# z!rV>FDGuY!B{w){xxs#(E2{Io(B$TZTPY`?x4;cS^E~j%@EAJtTyg7z2Na}F!lKL< zEc!TXPI^J&v!@KaXtOXzMRbb)iTKH*+=bGhHq#kR-6rmX%?H$lhZo~cdRwzkzV~ABZ z_MUo;%AilsuKWnEj!!5X{1Fc$U*M5=rcH_I#LmP~HGcv`q;B06cl9WE-K+YX>^9NtTXx>w_Ty8^0VmC~rJF&8~ z8>ch6(Z8Y-38Pz4F4cniDIGYc(u_eR?HC{4fN7)K@WY`8cJE%HY)Ct9t8_tQPAfEW z-(vpbXUG`Z0du3b`1<-8c9^|};m!~6S<-{%X|EtfFR(+d2^Uv(z@VuPL49h_%c>nU z22XJzx(1QMI&o5~1CG6)VxNw9ZXA1w;!&L#Y21R^-<>cB6Q5_*3U&1sOfGwg-tArZ zq4x&f$3Mbs&I=sze}jxeZ!ksoFEZu6Bf8=pCja}2usgr7D(f$Pb$rA6_y0u8-yd+; z`VTAZ-XZ7rJLE2V3)}f`ur=xfoWFd;l!E>|XhmOsm-pgtcSvxR^4@&-h2Gr0q!-V1 z?89?C`fyK|zv#O08G$qjRuQ(j~eg4SXJ!VD05lGLEI}z zo`0LA#0Slg;R+81@Dz6`zF?vx|8H+!u@CYaG53CB(AYn4Nce#Q>(5v%@dbS%zT(@i zuP`0-70ZYIf%J~Qkka`A6VD!4eESTk*B`NO;zu0t5`U-l1q1GV#o(jgaJt_IEdTNx zpK9LW>azFPc=0p7&UueF$?wtC@dgtezM|i@FSzpR9~#&H!tZf?xXH*ru$Uyl=MDRZ z5i=zCrNt6_d&pn(dmzCFH1*-VH6*z1=|6Zn;TMJ^{DOL^5;;D}UXK6UCc~=(2J(^P`}5&jrTNs-{=7{` zivK;-k7sB1<33~*)nhRsI0nPs#bWovBzy_H z58-(NlDp$@qA~;W|FWT_l8*iM$>4iZ@FX(_q2;Mqyek!~_O4i8rowVu3S>8>Bh}$P zz9>9E?X7!w(wqXxcNtjm@-F^+5)V)N`&etAjctvY(Ak}UJ!Ux={xBD-gL3i4@jmK} zv+$-M8{@?sHi&S>D5oluVQ+rl4p{EK0W}<6lA=uBP9I!_st&Nz8@T)dxsZe+0*MS@`9Wj=p*K z@#<^}p1Iz``C+lp4^M&)OT&`&_u%q44F=tr*fuK@|1D3)==3b~E6%}tw<2`Dx`#RK zY4F!d#k`VqadYE7K7UDp`s5U34NJ$&$cKok$VQ%WCf-fRhmL$MM&HlCnUQf=<(`Cj z(OIyv&w!IjF4m>qgKTO#4&>!PsLaMhQD2+0HV<23it)}o2Y+Ve;i^m#Cao)lqiH#e zRu-TpBM*6PGxQw}^-4gk+=yC7`)51@Gt{{KCY0u+E0h z@Kj8+O~*aE9MrwZz);tRa8J#IoJS`7y0UO3GXdXJvhX7%3(FqlVq5?#N#=s_@$hL*^-Hvrx1%h)8fFsijSWXhq~1BAV`{;#N{7 zMvLFyN1+MmSbiIOieizIb_ceTqS4b750pgX__JipT%HK+zwuCKDcHO(6ms7}v1nKT z`aTSY`iCIAcZkBEnUT;pOTduuXoOYWM~6f*>>sD%_042RK8eD=qBwLH#-pE03@+=6 zkMD^GyLJm&15!}>Ivvk%-p2^L49I$?W4?PL4wj1g#*F)j@VTcHcs?!%nGeqLPdEMf@SOAfRoO-U>e>aKap*Gtz2*|%9(9@9YzpM}Edsgxs>|FiCy*yy zzQ#`vzRG13u5%g9D}1fwHJ+||lRq~K;j}iGn|`{&<#L1gmsc10$E|_9a_wclfA*!!^ z*Ac?KCWY`{7q0S`FZ}UQ={y?K0`Xe@8hRGqK-$?L*svh{&A)<=LxS=7{!J|Kh{KNP zo8tF90w+sv;{75~tLlk_ntV8JmxiKwX(;+7+=TL4duC2D4UZbIck1d4GR%LazQIO00iHr+s@5Q*<=uHo;#Ynad#if6hJ z*fHZO4s8p>Pn!Tt*?tWP143Xk^#am2UqE(}KR$iG1nEK7uu1wllIaFC4I;3*<0`H{ zio|xeJNPy>77lT@5mOO^C&o#5U=@is;yHD?^A5f%Mk9XK9lZJxi-p1Qh}o8a#Uiky zRMa77&J*vg@D5g4$0BWGB>H_7^*i?%NSUPI%I9?K_>&5!wEHM`NWtL?DbQUXhifn6 z;9MC4kw}K(h0*w(7YFU=Na$=2$I|;z&<>77{fSuAExCoQyFzj5cqq~f!Vu69hTPuK zs226h4Oc?Y=XWF&mPTQ=YZ&Zo@1RE1EFU-C!V|fhSgLXhH)Nxse&G&2sfNMH^Dzd` zDinL*Vl2ohgUi@_jM(-Vn`MhJe`G1rHLD;J2H>O91b?v>y|wQ-dOkN`MMWhJX}6#z zxB;b8>X0e+U7~9$aaOwm-DWi?>ny{XH-(6|D#en2wK(>v4w`$avDKm;4a#Dl%H`H2S)2l_)jv64bv;lgC#i&ZlLVy_SYV!&p zqhBTV^^5UY)R>|Ro8Va5hW(X|P!ntY2NmUbTwjH4y0uX5UxVJ;YazF_6qa6%2=Fe& zn3!-)wU9Vm1vXjKjWSwreP|0xr`99)VKY81X@OsJ8&=F|$I5=SxE)vv z!{hZ3Pg+=-S7HCA3XECa4EKkk)*@OjQGmk4~_4P{?ZUs&6M^VywHOMZ**Q@de#wg(5?USZ() z4*b@8iO(l zW%fT5op(Hz?-$3@)-H)?r&3A?`5cs$WR#3l+Iw0hWN(kXAA6L&vJ=T(A)~%Z3oR8X zZHj)^?+-5(J@<9r&vReroX>fG&ixYGX7<3aj=fJl9dK0t5A$cY!bqImY5yY^bb6KuQ_zCs!j_z8sIdt6=uL5;6Q0?A*%nL8cDsV-3iR zs>YgCb;#eq;!9N(+!D&zXQKpr{VFk3P=%M?ET40!gnLI5cw?Gyic=5k#m)Hfwh51> zHzK>P4%*inu;y(ICVZ>Gl-1RcGP{S>)wS67xE1fa>tMH|7EY#BOqZ{R246iE%G5w! zvK|JL>rv@ZheK0p(ZjT-t{*Mfv!tHYKebq4TZKDKcd_6P%e!2vQMRcHnQ<-HzPJYQ z4K-LLREOlq`w;nkAIdZCBW`;gLOoh=$fyx&nXHGl`T+*#HsSHlHZ%>iLT60_;M9zt zCLP%0-GNm<9$?yk56}?ZjC|L6c8yC}Px2o6C)eWG{(H!HT!+1*Rq+4#41NnAKqsRG ze=Iw3o>~$6UkiTT>4xyG{~&VjA^Nve{V~jYs@!b

    x9G5x{y+o`|sIAhE#>&(?7e zSQG@ePaa6`4`=sU7|Y9pASmO9D@nc(nePnfy5QRz52pKiL-m0Nw*2$PPdyJT@AO8@ zI#*mV_dpgO7yGI`aJ0Y+Q%8Itd)^o4%{?)1vOBW7T`~Ee8!DJavRr}ru3GHib=C#u z*)9-mb;F5l50qxP;Hr}^&P{bk(>j0LWNT;;al>DZCp>?8Lw_(B$~6Iymf+z21P)rm zy`VJ38yB|)VEnE((`ntXLCFi1Mjl8QbA+512U0@;_<7qO)ge(>vC$V6+5QO1350!y zH(UqYSdTLhW7R(JnB^+%2=8uQ{KIjYc$99)MRLtliqCQ(P-HRvep(2+l6-L3As8ZJF&Giz;-y3+j%r0=f>%7;|3=}c&Qg+j zDoB>vOUS!cfaZM>q*SrRRKhPnKAQz;AZ`J@**T94KF_8Fs*Hc+olB!v_~^jb`E)E| z0o8gfr0)`oNjH2EUA-(o{KbOg_CS#4Utdb&cEYr6jS#8)6rjEhLi8zkA-S(uNYSf= zh@Wwgw!;E+ZMPu#sqoY3NAoFa#sZoiJBN5U{N!ZDNBVykkaig3Js&Kf*k=M%og+Y6 zjD_S+Sw%b-)(}tsYP$4jHT_}Sx6>((B~Kv|yzm z9hKv!bBXNOd_p9WB0zdoi|L`J5OIS9NVkZslWzevU7ANfEaubovUzlUY&yNXFqcmM z$Ic~SE@{kQ46OcQTJIxB5)+n?R4hOFA`SJ1*qB2>0rgl=-y(EBVA`gnCM z1=_78Bh_`Z(PRU8T--zovQqT!-v(0CUrR;T*3tgAqBN^v9SzHg(1a)v+BvZnHmMCb z!q|=u{kwQ7RgTcSd#LADpT@Qu~x$Li5sSpzvAFIHv(K^h1)qrO0O6)MG#uBL#{7bBWwA?+&$yQ=D>sxBNS3z6k zF8DmFz+X^E7ecy^!%UW1!HZ%RI z0UrHz$U4hnPD};-SzUJcCF|>mRHC`=KBCH-v1wfk#F}bxs<9SpFE!!%O2(;NXvRO$ z`&jU-9=fa#Vp7%wxl@hcFX@EQ^7}B})qpos?_>R7BZ{MHvG{Hi&hpnli~V~ErwUW| z*C9Tx2JON1SaqNY;fD8Fe$k3i*$!+{Zbr`ORy^c9K$7}nc!o7$CSz`9M%3aAi-oyp z!KAbn7+ZB9`E3jOL?0kL`YCz>8Q-G+46Ci4;^Cv`*ngk{IZV6MU~y`W`U5zYx8P(( zJv#3+p!7&BELeP%nbm-KkLqx2RyFpNZAIxj!jbn|p>|M~F=|Q>lHShvNCl+sQb4!E zHkdsWM-O{6x@_rY%wDh+T{&A&*ei)Ul3TEWjV{ldK)Cr+7N@^%2YmvHHRNHoPZ?Yd z1LB;HHlkjrV--ggRku~4wN@34(kf6K z0glbsid);I(U32W=$&F{j$Vfsa+}cmZVh-GC1EwR0VOXbP}i{qnl=A_UIIZxLJr+F zQfOE$i|!(M>=FXTf5{_stsG)*%iy@D1XL8+_mgs%8zP4bjs(p}C1?yPVR5|z+G#6} zy_3Osj0|{X2shR$;7HC6^e@$b{Ik7i5H-N@0bSU6Xk+z572My>p4$?nihDc+Jdwwf zRjd`{FAM(iEx4t<2?2XHpcW)dyPibu9sdZoWy6RV_y}q9A!twhgh{i%Ks{s# zNxO$2=P-mwtIr5N@*49ue1J4tv#8q;R=D=V&1o3dZHHj~WC*+8eL!RUcMM6&~(Tk=O}@%Lkr=10tu z7{;o@pCB>*8{0nrMOVUK+^ihI(<6Uz>F5|FHjP7D?*}{=ePayX7_LN5BIW2mFuD2% zG1g-c*8Yci%l@EU;|Cf?M(~$@LUm{e7f$fdsM8M!y_-OLv&YeMoBe#_Z#3Bqqrq_k zsm+;06Ss_DbYKh)1EU!F^9zn0UvR(S6C(8gU{la{6mMg>#$QjIUhIeDsR8iQ^Tr`> z4$gk`#FP&ld}jTuzKCFq9{0oc%rJya#YRz!h1RE)Z&QN5yP!Xu5dgc2gi@zk)D7&j%kT2Vz&g zKZ4(eK%v(k-WmSL=m}&!q)-$djE1pS08}jk5!w_A`E&jlYzxBJ&QLs;55VV_At+4s zhw>#~d|VfU`W7D)^LgXWd_Tw!`{O&u2QdjAu<7=|VpVrEUiHUB4=y~|7oS!DI(0Zu zUd?i>{h^G7<6`!gF!((3!UC2%IXnBHq16j1TAnalqiHw1Cq(8JdFYSaT(FFlca#19J- zLm)2c58dccTxaVNOAW`lRl!)U7Y32%;q3Q9vClpL7fyykz0Mz9Yl6|RA_QwchT>^i zD2BI&Lp3cJhd)H&NlOGGErVcrHx9F0Z{lNT08B2$qH#C|A6rAQH#rQvf#DdHio)s> zT+AqpMvspZbne+hHH@+OA5HM2@H!s+w1drJdvxjB;?V*p@cOZy<2x64-EhVUODC)l zamLGurZ~`OiarLShs=;kY%hNewyRwd^0Sc zXAc!E8!U`<#zh@h%xZQ*1k3erD>x`i~TYN2Y1^-c7^gOV}nm7lX?Q(?mMk~nom?Fj199LM6_5C9&?2~Z7=}0%$ zCw4`Yvo(yfufy5P20>HpuqMq4;YY6Hj*$x_+ucFiU2t-?3qp+@kha|(>wMf`{oV;1 zZrJ0Tz*Ve#VuDV4GwgMCM4h!WT7S8q;F|;Dnq0uE<$@>P7ASNz!?8gVd~+~EqQx~_ zSj3KJ?1s`89}&`Q7pGm0J!s_@cLmea`Hp)LL-9h{~?Gm499&=7~Wlo#Q791PR4UFvpI}06;aq(AA_Cd zq4+bM3#kd=m^?KKDWhQ+TNVV~u>h>|4?ssKyCVC?05u)L?R?E0$cYcA@_7FP9KOu^0_FqO$&uMmy1Y| zU<|hg;dyO1yp+Qs@5<_^EwS)POhV|d6ns%k#-d-b@N$Sk>nL02$w;6s5=+}-5OjxKI?1#H!;=W1O3S zp`X#LK8?X=y&x!N2ch#^BnFpp(UBAlC)*HQdlZ3fY`?Kq3BmqXA-GD>xak~)b0g8% zkr<2AHDOpL7>#3h(jb-`h1>q&5NL^lS#l^Y*hWFoHUgRQq4-ZB93^70sA2Wr-i2Yz z?-7BOH^NcQ_R=GhSgq;FLDqb>cR%pK6+aK0KIe;`1Ycm(8!7g#@ZoY06y^w%1UJ0j z?1jPUp5S@qh2A(@IrxNQug$0J~%8H0U`xu`qL#jh{1*r(6p%KZ@h_#23Sb%F3d z5C&0`P$aSSrj3O$Mm+*`dBG433CF}qp*SHI0Jk}zNV^;a$?5F#qZEQ1@kmSz4`=x_ z7jLhHz|fAxE9r1t-V_18zDR6sfVO1A2f!&>M>*;G=|;4UR>(chvweB5Y#lpG))7RSMR~{AbnJP)`8)A zb>;z6MX3HRc0Xu9%2g9@bW=>!Oo*AbJ4ymB+F2 zsv$)4kK$;CHmk{YVJeR%GG85pVYC6JY3t$kum-voYC-b5Dk>-Lf_0KM7R}Mai&0%H zE!~B~9a_*x(FJd=I*MX6@Oz0Gwo7V5!fy{&sHh@v>mDec*o7++J1{mRk7P4d1V2$k zX`m)%oYcV!ZEZ~3Zj7L;gYeyP2#(hdAiV52R1@|=vFSL*KkbK>nh|~)9EGfm9u6PT zgZmu=)(_v0+ZB3@?K6h1jTV00*oQ%Z-Pj;x0Bwr{_@ZD4OWULHXgG)ksTZ)d!U(-K zhFJ9Y0F2uXz~JRx6rIq<=6O5V-l>f#&bly=-Hniq`?0Y0Aj*x6@logy4xibJMZ2`2 zS9%aD&+G?($Z?$iX@pld53}6s1lscsq42p8?p`{Kj#&qB)bt<@#TcV3^$?UpPD4B1 z2v4L=;rD?PnC5s2k1n17ck@pArLRIe#FdDXszfa#Dzy8WDouK-Ocs7S==E)7I>Hdx zom~o)@koh0H9>pNZ=tRp8EVuQCy6)Gq&7p5!l%lUn6)x_&sU+PUc1PI$AE4tYSTZz zeYDQrfM#hhf6zZYniywDw^EF#?5sY?b{-%bH+|Asc9@cW?4uKIhbU(2ewycLNVUfe zNMXGJEgauXIn48;@JffqGgayNI&~5lXKOvGLGpd-r0J(Yi~s77_~yMdcjFK2;F)b4Q ztWIHTROp(KJaHy$rOg^dQx@(dL4F;&GF^?%NV4C(r$MeN+7uAJhrF1_Na&;vi45-{ zk?h@M)v%jpRqD}mjYH&ob{~oRAEqsZ2gpvzh%$8!Qn=E7iX1kgqH!HsG}({>PwuAf zu{}hU`jp3;D(lHxf^0@nMfWBe!uJv~Z&KTeRM^>OmkI>&hF^JF`8hysu3 z(N0q%>T=sdX6!mAS?ZANaSf7tyo1(ts8XPq2JH>mO%?s>q#>+BDR=ZJ^t(Rk3mcHG z&Ta~JW?n2aJz7}KL2U~MgGN3u%nAU{T3@^o^@C20KQ?oMkrl_{qJ}^IE(m}jPZ0j8 z2Qx2Y2+o=YLBh}*Cm#8NccL#i1;JRxV%QRm5SZ;@y5wPhcvkvB?u$PHUNUWSp9e-V zJ+aWy8*>(TA;!cPTfBmBUY_a6EH~Z8_Ra3L2rOJ4h3TQ;h@R|^$e$iKFw+<5LSA^- z?uIjJ?l@oShQ4S|oOkxaOL$@Cpf|J{yij542bTnY$T7Wi%b+JJzk4An(jAdO{)|fs zhU+_aj?w|xbjTl?f&LJ14#jtcFpMA^Kc5GoP=e{)$z1T-MZ#(@6th^a9K_aQSs#vM zwvRrp4uRz6DCCp|A&@s3Ra*j4%nicE%ONbTu{>oB%g-N%Kr4~$|K35!TOWc~yuq+A z48$7te9(s>fVPU#UvQ>S4*j0ttQL zm|ql(@slAi4GYKH*-@CG$Aw$@O?;9{hEO<*AIDf6Ee^*Dqgc$}!Pb7yA8Y>$z!1MT z(nmQsx0-`V)xo$uD*y#`A=n@jf|D$^*d7RExuY*q46akoq3iVHfE~S9WlOp%ooHi& zJ-u9goeuRfUcShYw%&H7*cq;*mT{dmOOs?9(Q6-QWs| z&op5^HVgW5-jxJ99Vo@|8cj>GC6A|8R4n8~ljb_n**jLWKEjDE&37W6HP$4~yl}mr zuaMFtQ+i=?nFb#oqh__s)Z%`Y8m^nt^2{@>)qBpIMUH2Q#wVV@{&OSIMQ;j40Ba zKJs6oEx+yPwb3==W9ysq(2|T-+mU6W8x+EeaV zTN=In*WZL9)5tv_`>;mZ-RsFkGoq z*qJt$IFcRnzURi-Q4E)zkAow1dt9f?zxIrau%}dSN4m7+Iz9PdM?xE&DPgt)nY?zP z8`UoKEX9>Zd|YVkg9{yLaH9A&Px@@myy`<9^k|kVz4mY+D=$y_Y2Zo*M?L62Lk_KS z@uZtyys4nhlb$Z~BK<0N;9!T zXhZ0-hxo{PxbaKc;5yiXnJ=5MMz;-TzdXW0;Ro2rd5H8!Ehw~TL#13h8dGZ!TwaC2 z^K}@X^8k0K0b1$}$kFdaBexo#wC-b$a2rO{8Zac&2=#xJ@c&zd8J-PTQgR>8rEOU6 z^%U~+AL9>W`2Sk=;^?g|Y(4P=7aBXzp4$bVTb-;|+W~p0r+BT`$JoklI0V0f;=I=| z$#@2t6;H6?;6sR5b>gGwBfM&8!}ML9sD0f4XU4fbVNCqAxovP@%z9c{H}=LnL3>aO za#uIPUA+xQy4&#mT^GugTJU}1BdnL~#Amf0yl@!6t{eSW^L+@LSwDVk;1%=}-@vD$ z51W|YqvG@ddjme8F6=#CjeJ6Wa6f#vu^u|#53If0i(tDRbSxjlEymb>WSToium!8t zI&rZz3%(|qkbaYmojsYDVV925Q08-*&iqf_nFySgjb6JHxSdP~Pi{7p5>uH!A_I4> zWkcjjI_!$lS+6Y}KPqma)F=b5@1){KStf2X-k_ZE5vTIgaKtzX2Lp1E^eF|a_)?M2 z%|H!rF1p9FaC%`5ZZbAxXK5Dm`sYFPY&P7yb5K4#7ip9U-AkG9s!D-|Qz~W|#pA?4 z0>apUoB5Yh!RbrI{mLx#jNgQkWjfscq+v8R6^mD8;=w5MX)R7?JwA)dG0g1JvI-UQ^sIHd@#0{Jj_T>`5G#n?Zw z7*YlKc;%7{Q;$rHT4W%gG#8gR>EPOBqfROxnk|eM+msJ&n_J-T$-~!UdDwR&2UB0? zW2N0KSUBe((KHJ}J*gPjl?rZiE`Dw-z^2PNh-H04K~{h16b6I8iuF-vg|TZD3U`wT zJT;AiX>%;Zn10lsdlR{taWKtEz~T=vDA9_6@s}9fQjb8e3acxg1tVKL97;#S(YS^6 zdw+(b-z5Tn8=|0-&1x7$*0a1H0R`5#{LWr)VOb0+Toag|KOP4cL|}$=4F0*s;xJD- z)49@+YLyC;k_;?QU<}T;SbU3&fke(t$oIzMzg0IFN1Fm3tvL8jO+tY~0%reC#o*@* zHftdTT0?2L7;_V=6j<#q6^EHd(HI?#gyO4koaBo~>qtB%4##|ae?46SJ)f6!ReGU*jyCsT*0)~Tn~6`cE|Q&S5)h{L(|6{ z*Do@C(8C2ArMjK!9#G+6T))LG!N`#{rNm459kbg;6FYNrgVFvbge5E zMS39lumgr&OmHXI0aaU>?<3U~-xu2Apr^DKQGZ59(vOKI5h*T} zf94X;u3%d8JCrWm4@La0tb!h0_4z zNk2c0rv0Nabg4U*bPKpN-5`Pn8E@)Z9YNE&i(#}L5`f%Eh%GCVHNti<_ zZ#Z<@*Ppr-gQ%k}gjP0%Qr`Or%3T&iO=_{!VS0mfAH`FOQW6C^B+<*w$<(EtOm|vu z(%#Gj`s^M@f?n~I`8EEB^^ayABWS! z;z%;8i=!~_`+D zxx`R#Yz!UXi=e0XgQ;mj2&vxe(HEg7kXiTL^?A488=-0WkFWG?eij}}7x`(m)bD;$4M1R`q&>lI1IqJe3B_R%rWVDUpiI|}-gJ#AqO$x{ETTxKj6o|*7On00a0bSc5OlLaeQ>N=)nGyl9_ymMZjYFk- zEaG|-(E2VNnjxuJeefo1z9vI;Mk-QfrQp>-3WjRqv5WcQmmZD|h+Kk#RD%OuGS|6B&4=n1-Zm#-#4Z#w^D8iz=qTmF@RJ zU!0*??T%vhJyOsG;x&GFddVGT|2aeZq!YHkWii6Y3xECGv7+A@oo>!B<8_72J0~2s z@Iu;3C-nBY;v3tih4k2-p5=-qxl9L6a)u3Ky!sduHj(MW)4f<qR)aZ(2VzIH*$usil!u|8G< zJEsya96#!b7}gWq&1Jp0eH<7wer-j#FG`d>A!Ox&54x_z=s#pBt*%82iSq!;GgM z_%zuAPRo4|(B}y;7U!~O`C{<64>BfugImQk^;2Hx^aw=G&!Bz3cb;BUEj0Zr4&kskE0x^&hfGb+Q2tVkD zdLdtyV|&B=Fys8j{UOR1h@T4ns2|x(8ZI)_`9YGRKWwH&pEgnaUr9s>03Ce{D zG~4x*+qHp8pKqYPjhm@5P?B^zB`CFb3tiv2h3tK{67T#i#IstSl)o#`fT#i)s1W&S z%aRe3Q(wevrCMcK64I8YvKd?GYtcp;Rog^y)f;H4t|UD@BtzP|vUIY4E6vZ8Be!2f zN8hT^10z+Mt*%VFJa*Etud1~1%Pu zDNQu(96NT{HYzueC4DIwin5X><=Zm!gd;X zIYpHIm8>PbJ)-pc-5QcDTu-(;)>DP@dRpr!PJwea(O3Qrw8>43=IBXM-%D|lnJh&D z{W4^JM4FTgtgwr%g-bsmNZPLPJ$(ldBr3mMK%U%XaENqC#Bp9rSdyDjg|T zqVOjQG_OgK*l1=t?m=`UTan~1DA2393S>C8o$c@P#As?#Y*C`0rAn0bP?c8T|cPg91k_peZw_uo&4*BW%khRIcuA&%-@7E5q+q8fqqIq5XUgLe%o1cOswprf*}cHXZfNS>W@?z^9@t`23fK z6&G?L+MR$K@@Z%+&cu5WRu>hNqG5J9R2=T$@8(Jv1r_3TUm+|9(x7oB6H{{2q4S^+ z2g>vDw~=|^w%o;@TvoFks=`N`B8;ECi-H8^1$a^hDZvU%nSKY&lQVF(G#_$hS(xmW ziP_vtyo`*;Y?)ioy_?DQv0Lc$W*(!hX|Vl~ht;OV7>&FQzv^NZZ_2R4xCpOZOE6|$ zhU>oyv1Vo_%tWf-%3pwjX$6q2zm2QixvUpe1mU7$1iF-AT&@TS#kaw)$^1mU<;*W% z4ZD@MG56#h+@4d7Z*|PGCtASrzaoft05>$%x~sneH+{dxhRY+!vfWER6S<7gLfHH zn{Q*wq;e#RmLXEC2$ub<7JgTVBcJlIi(P{X^9lqqwp{xS^Z(r|!_2;7O&2< zZdr-<*+n?>p%jN-mm_54HnN?Wm+y5kX4_VvT>K8^mu2I%Mkz+c?!swWC04AdgqcJs zW7JA9%oxkQu}YRlRUtU57Bb$=7))$}m23@Od6nbq*=Af{P>ZU#3Vdgbp4%D5>8-0{ zUaoss%I-ZEmN&_Kug12iwOF{Y8uM?}K}og>@uqb+Rau2;Y_?YF+-j^Axra-J%n7uU!lSe&aA?$|)=4U@;>I&Xi zN8Dhszi6>1YHe&W&gu$YL05?KI)l&38d@pNcrWRKFR#6EA;t?`Omki$!}RUnJ~(mC zmA$8)n9F*=ZqDxTo_rl=JKT{t;10F@PRLiX#bA*IBs^W9c+MVko;$&VSJ}&s26JB4f$^#=weK^#cdAqG+fZ>&OG!z&d5)8f!1#~gtPkc^dRFP zezW5r^TC^_Al#|+hKNrPvY3AF#~*>Pg#nNm3PLN(XCzZNNY)O6PJlnIt_y%V*Bd71 zJs~5^fw?9JzW2Pruj32Rlb(3+!wvE7zUb@XV3E2nI;V%icZCmQ-U3;D?}w1te)z+B z*Sd1NnBO!UYsBJkwk-hplOm9(>x-qFAoR$1V;Ordj+$@~T9ANW}_D!qC)YVp(;hZ51*ngn*k;6d6=ylIbP zOl1uAa2Z2!CUq)hapuzm`DdA zc3&VaP7VWyaTTqFVc0t{9+H(oOaqUD)ib91w8g__UkdiAN5f_oNp@=V{U*13VBN#TeIjX?X_8!&ZGz^Fn7+Qj2g z^gI?SH{&rqAr;>D(@=6e1010&<`;{D3A^^ZZ5i+wipOo$IQV3yV>pH}B$HC{aeoS| zucc%Cfi%1~y$QR9G+1WeM2K<kIbC|N zoH8D+qJmQ^Xj1QTntyr~ndB}fn_*#EwP_jM7Fb5Xi-hQH@nR~Uu#ocDdNf5AP-N;n zig#T|U-H?kYvQLgI|2ISyntphzrYh$LDJP0q%|gjH0~};(c_D$I$DtC4G7ZU@+EZi z^fJmfUQSWW7my{ooL=b(k>KOS zI8D8#Xyu6MvBrrpEWdzD@s*OYbhmWBTdNPNV_XH(iI~y zx*Q`;QNa@Q*TMnB{ms zz7~hRt%CAJ5yte1V%?VwXj78H%(>DKE!hd92})3(riR!y!kJJxq*klqeZV%H@Z84o z26bqNDdU5+Izkw8(o~_r*0&1-n|EM;t`dBUc44l!8e@SKadAB$X03pw6BH1 zp^?2t7e^9H)8z0>eH*)Gs^Hg9g|d_~+~bt7vvuIxBcizlOlc_OLtZf}@{p*&HuN{Lr>Rd9F2jXEFVF z8q><{?Lk}z*6*{2w?ET_`L5yj1Z(ttzk-zQSD~9=gR@Iau-x=2e#qNkbEpL_30}v> zlUH$J!WGXUlumKN>^5h(-giSnswa}SIb;1YPpI{H;YWrCDjkFH zjA^{1Cp>WGvk&O0KW^vx;3m_)#V)ch#t5#e^Fi-jcYM?L#oQJT2*)s8ndvmYOxB|>s?V_`!)_0lK{X9+IE}7GR91B`- z*P0?rOew|Jls1K#Qf8|eZN6qslV4n*i!NvBxzibv`+kN#=^Uj+;wR~-_-Q)y;S?>m zxIq7AAE!TSk5TcOLv*tHAZf@N)2kJSsEm)j55A0ZdVZR|4xgcY-6qs|<}7*MIZN&_ z=gEql2hXXC^t|#Cosu`FbYnBp+1m!W>UMCyKE@N9=h)7?+AU|EKym01L}qlO+qw&u z*B;|k;S*e1(vD^Ox*-(bhOya?;A-80U5!s!-<{0|ob?O=OPPnNu?Ew2)*y=Y(#_bM z+E;h#A<)>2+sxy-mG=?KJfDL1U@v%%_o2$^DMGsY5#-tj$CCrl68n6&K~=4!R$TTBtTg(E7-}E4JfZQZpVOWV7rHUP4N$7r(X- zBBgu~=}%r`l^OGK?_jf=Ua+}y7k)r+!B=D_3}A85J8Tqwh69t{Kuxn7=W?E)sjHXu z!#-l@Y8U$VzQA9%r%-BojQaNv;klX37Z!hw^rY8NiF^+AivQ66xeF8gdf;^RDT)l9 zLdM}Gvet0$hV{}Ov6?F(*$0cYeW7v86Uz$%*bIaqygTIspU=M7Gn28nZp?!r8;rI{ zKa{rlA+XmQrv-i4zUqk`RnBZ*cSYO^SLm@m-wH2RrqQ^uYsY$D8Sda6VLp;=tOnW0 z`c?G-;1%^osDL~CPBQ+C@xt@S3$4xGP}25+6|3YEw$Iw8_x5+;PuNF4;iDLX&HcsF+Xr0`e0g!AJPqiA*1aL$#GU6CUfxZ5C=bn zJ#ff}gS+cEa4-(QPR4K98ArmdJCt$BT-dYTK_P`A?@KUvXEF{wKM*yHeQW#Thbc+y z*aF`0_~L^Ow#F_C4*Z_69%LgsuB8u_%lSjgkRAUSW8aN@v9Kfn<^6#eSsQ@kwNVQ(7%y|W=mNn!8Bzfj!N z@yE-+5D0u_p5AvM;7*N3{Qwu&-muzijCpy7!kG@inEFZ4(2a}0*M0F!e~&}Rk7&%Y zh=U(v+W0f#u=Z*c-mHi~y+%C#98AEp;ds2>!$ZDTm`CX8KP>&i9*(v77f6}Sth)9C zj}k|b(lvt9(qlMbHHmE9f8plxDP-Y2kzDsrBL1i`{C)Ea?OwlNb>c5}G7r5-9s5Ln;4FTi>knJI*AMgsP9g4t3H1HIFX#&W#1yXy zR3ZEW*EEI^z3&@Vwg1K1&0q2B^<;8>!9(k7r_oEFne=SeRJva>nWCBJYPRA8YFavh z*1qAPh>9^R**SrBoEnEJJIA~-9`buKjnXtH(~-(QaN9ABjNVP8{@FYn(`9Q zAP8O_x z?Pvqp_vazcax?RP*C7_>g!(fUK%y<~-E&2ow;MDaT;M+Jg!YFlms;b7*K;nRKFbWn zisnpF3|bsg0k!Eo@TQb3*2njTnIb7@@9SZr*_~Tuw&Yi4fqW$ zaE5aoQ>2|RKiCS_C9WdwpB0-+bs6e!k3&%G2qd4K!G%c|P_xn!yLOz$66p(={)^3x zTxN!BEqlD(eGZGCT|zZGkLno~>~nH~?V(o@`uqa=4bG$Cs5KU_`Ie_F*}QFbeBtpc zs2nwAPgb@?KFeP-pV^~%j~$N7TOmM*&6C_|iMJQ+Aztr>BE|wpg?XaD)*0`XUdOT3 z*YKkEI>yZ$p>f+5@&|0umSu|>U!9Pm?|^;9Y|a8%!*stbj(W1a)zunCj~w8)@fy4` zED?I_3Y0b3?8=#pldxmDoUjXfeyuC2n0IN)Hz%~&*+S#*b^IYq#5>sHi@7aAMNM#O ztv#L{BsPPS;Cn@m&EXeE`kRedqa%e-?=2{K#cHXi(m4BfE9N-KV^5tt4m?(2+^i~O zleMtp)h_&KCT!cc9pW=pkl{essI&#&SuIyNqQrcLo7t?HP4JVJz%(H-?0dO|afISH zb3~NQEM{!hvxlSH4x}?~iDv97na>&qB!s;YRIM3Y(`($yfoFfYr+QMdWiKFS11giWt!q!9# zMywtTntD??M>Uw}5IXggU{<~gMr|r^yrj;W6is|NvI+Y?svuZa19oO= zsNy}JQZ;*%I~0){&Og=~*eSCXmjc(qxOEK%wN~SP?J6w3p$xMV>!6Ud0dK}{N3!i! z9FE@zx8qwdd;WH;3f&4F*$qg%yBTWdwxD5!Cf1+P#N-fl92eP&K)-D`@@+E=@2^Mu zpI3-o^%A$*s*tp~61}4<;b>HauNIZidC&ZxVXrW!^8<|Xs`#3Jhj{f@sGV7g6~Zsz zk^72upwCcc`xJ&d-@sD22J2O7Av3KNQ#I<)nNkN|iLZG3p%I>Q>VX5z_u~I23&x+W$DQvrxG?M!T>2|8`e+5FzkCBYz2oP9i+%ky>}RRNZiTOy zV$%TqSY~r_Cd4bPI=ou;0fq{nm?QTA?J|{EbLT1kD84~>>}yzPGvBkD*~!P=LuJYb z$e6xIc*%3zvtrHY?q~QG{~AwXEAirD1;&@Z#;*6(7+LZGqB_<1GO3O;0GiSIuO26Q zzansD9fmBgg`E96%uTGq$s=#D^;;$E!mH4A;|(6(`hXm{8ub3GLEnKd=#ypE=9NuE!Y~`j4M035v=(EIue^;^E9eo_+-4z}UXjdreMKVubhnoRzD zf&bahNbUQ~Ue`}JGyF5YjAE^6>POV{R$_Qi4P-XHM)44T^eywj-R0h>8{q-xp?kQd%zkie9^Np1es&~@McK>MtgeUV1zpc*!!##o#x2AY3s5(~k$?!qPBA z^Z7PJFqUSAVa`fhIL;`i*hy{#CraaE?C* zU6iz*!%)dHSmAO8ZP9x%NB$zt)av0Ndv?$8b4ES0#_osq2$HeJv?X@vNU_CXH>4 zYlDxh5me~1!I|SWSYhXZF{ANU=2fGTP)D9$4FjZ3zwV1)|>sx{r0$^ zVTFAz*YW72ITWgxqdbK*As$vxt>DKRx}kiJ9rI*ek!`^H2==ZTB)IUkvuAKyZ66!1PRP{|d?C$fLA z)*Dal{2?aci-}eNSQ+69%^R$lx*mueM_-f%c;cC+D=z4|A@)A&TG;!%u!?n;2kp>M zX^kDO?61zU#UFWd=n#LVgSL=nolJ>_9SWW>r>NBl|DE>2INk>dNwM#~)eXKuUdUEfifDJ)q3c;~APvQwlTHaIn&SX#O0()p~u|P;4`xAV^ z_L#EI8GU&?Lpn{KtBwb{G7Y)tb^ zj;J2#iO2~a2;b`k6(b*vzT<(WXm3Afw*hYBSHlcGw z2e#&ZhWgGIh?-r4#U{tY)wq*g1rvjh*mLR)p7p+C5A_?2^Q__- z-+PR7tVV}hEjDldioB|Nyfayk=fS ztJFK}UseIPu3F4({)k0?nd{e3hq326AkfhPwG}N0KHr0;>Rv1nW5&q3Cg{Fw#I&W& zI5M^YYyW)4-??pAHlr5ZBR`>>GYG~%{D3<%t57ib0WOl92NBAg{8p}C(mz9ZR~w>) z+p*$72ekCs@w2!E`%K%|6W)O=e*TN!8WH}d4Q1cjVKS=~O5ycr===<$&9#_S*~H$l zZ`eMKS&7U#QPgRN#-b)nyxfh$!#m;nvIR%-e?X_H9ZH*i)3uRq`DsHtlE|%Z!t#E-YBq z1n06Js2JIU2?gIVy0?et>pko*=Q=|78^+6ZVa?(_g=5D1gf4JJ>8kCyNle3k1(z9J`B~F$DDH;g8uiQ!*t4ux`%iY_5fX-4-t@6i~}vq z+0MTMp>_E%W^KxfzC3(W$b;#wJcwJfzO^d?BZ?!j_j26-w4?=Lu?UQg#Ey;#-2N8^ z(LY=>sf41khCPil!|;1b7y=I_W5}LlR9;NLCW#p2-eTQuIde~A!mwL92C)j!%(ah2 z{XiV5e69HiYMk$F!MS&1BR?s_&6z<^E8tnKadRJka!$DkU*!- z$I`#}MA{RaM%9i96!GYmoPsKBt@`@7a zX+a`+%uOaIk63!OD2{UPB#`(06p}4YqYqos$=W=HT;^v|_T_Af_D!W?#|%1XkV%re zGU7_zJ!y2?D24V{XV9^tbUK#mOzP}2-6iNmS6Uot zjHxR<-{3;Ghd5KsUI#ky(}`qXIFnbk6Lnj1ZkLWD4Lxm39mDNNzXc z(Uhj?7}B4>>vUYog7TEj=)W*4wzk_)=pt*Hbj^l>r&-aKQMTl-Y)4n`Skb#~bDF!$ zjG9(iP{uGzYOywJlAXSEE6|^0CF=1nuo-t{8{oUP75f}Jv1Voq z`{?VjXL}Qd?fQz-iJUQ1)s9R1IlpLWD-zTjAh7ilKFWWE+@2QXMKiB0p%x!inG0v# z0f)|7>{Dw+sc;A42RoUq)`;m6E$k6#hLc}Ao-sczEv^Mgip?n0Y=PuXUvxe39p-MG8583LEv zQF^Bjr^S8)QeDjB?Szm?JB}}FKqF@=>23Uq(&Zgk@VpJXmwiQbNh@}Ac0uQAH!k1$ zj*3y;*e3cNORU>j$JNbTopvnzz{i@?h>FBUEPlXQVOKd%rnwEPnYA1{x(`REf5S}I zKBUz5V}*V%;!C+lli!Uon_JN{u?4G~+gQu`1Ka$$9#t7c$_)`J@n!~_<`0a0J&5T& ze|f(8gEeD+u}7vG)ss50x%&@J=KX*G&qbWiwW04pFU|#ZAu{9#<_>VZ7yB39c|Dw! z)QP`3-%#rQ9k={Dq0TeHcPhU*7w9jpjs1-l#~)aET8L)wyrw=%h{Wp#VU+lf*Q&oT zo%Uj0G3I@1qiPSV!52kANI;l|tVr;q)6Y1B4tGP2xCB1<<>W{5WJ_TNl<=4_=S zqT5JqzBak-)uy_bO?1vslL8)WB(d~$Bv`0Ug28HZQCF4bYOJNSo$F}RR%JR_p+;)0 zYUDq51G$H5QcwFfdiZcFHAZeBEzQkzWtukGJlCR_l#SHxsX^uuYUG`}juv<_Z^LaX z?Q_~dl`(3xphcH7Y4&byiqzXpYoqs3P|j|e z;Qfp1f%7GZ6YJEcV{*KQ{1 z(A_lI>mcn`IzUhJ_R?S$Yl3H(sX8BL*;hb@M0ewUABkbyJ^w; zZ(1ZRxRHK~tS5__^>lXSdQxF8!gf71QfyMC!42wk_1Gpd+_Rl3e{Umc$6fUC_D;It zzl$a@8|3M-10>vafEq3yq#j~zMcjT;ox78i)i;x6mpbiE)1(Y_E$V&A?{oZi(!Zxo zmp|^HmWJ)*_AjB(~1r2Hzt{}T?pYs6XJhlF$gQSebeG9UEt z-s2Z!@BF~Bv~QSi*bm?3z3_1Sj>_kpkv6**I~smMA)p&CxBZ8XUq3PKVn5CYHR0pM zUr>wqhIx|x_-ytQqYwN-MME!s2MbW{QbCH|CrldtqV%*>f*Q_<(5o%tH0UKx*Jnx4 z*CFDR6DLUOVZUHMt`9L=e_(+n&*pS_jtBv2sTxGtC{eoDEJl^}5|q#>Okn{+)Ol8f z=a>>yRmd4a-h#Bpfaicig=m$y0EKQ4AeBo4cq%PKZ-4#cKA8~BpZ*uc_XJ5WU4S+X z3X!gbAiYf#p_h+@C`afI_kI7t@#p|XZ5n{&m_amf#!^nKAT93uhl?72VBgh`{LC)2 z>Ga^mj2>h^?Stc}A6V1#6VrwJaYk(r2j&XU1{-1WJt9b>7K_uet>P4FE=J)hV&r*J zgdR_mpwPX-lvpiH;vpgw-Y-Js3x#Q-i73637bCC3q68;l%Jm&W`VWTCxaB+_9QOyG z%Kqcb=09*P_=6jp1j*ckXQ--!NNX3O3A=?!fAUcJ@JftQ8-(bf>>%!a9>8jczc{ni zh6>JFQ=pLzb%xv1HE}yyX<|duuG&$xlRZ5;Ye92YS<%rPbDAh%MyiXM7c$}Wl!erw)8s4hO8p2>FXprIz7UHQgv-fOUH&*O54z~7BkZ5v7rmMY^XQUmbNBX zQ~D7b8k=lK-nzU-``Xc&<4%z>rt1!F^wH6sy5UO&J3Z;hcrPkC<3Zb{T}XbRD_uC| zOov6eX7c3PO4WfT$J)}};SMxnyFE$Yu_fC*4wU`Zj#56@(X98jH2kd%O*zPQ!WMfH z|7T5y3LUt{wWpu_+Qmm)=z@+jCG|RyM2ao7bM5k>z=3kZ9O*nCO9XpkpJ}<#@h(sL z8_iiR&fe5_m%X{|z7##gmm0eLslnEV&d%g)7tRB{%f~%(z?~$zL*S$kgt+e!_{Dzw zyyX$N^*IunH9?p%G#paPW3kqO`J3etIJzmLtBqcGtr-T3Qz4vB%)U;iKzMqDa$k$})gqy={t=0ev2mPd$X?OdIBZ&yi3J~{ z(0D5sA6_SMrU(0m*!RBRXE4SOV-J3PFhqR3m|+;o8I^wc4h%<`T=0bCK2FlWIKze>QR!F5|#Oj%eKa&iio37+mm(!>{B7 zEZdra2-ke<3uMn#Mi%6b#=u500q%zJ*saH0jUP$ekBx?taSV3v4ToYppZksw4Djo} z(aFG^XX*H~BNgQ{lVBW^i`5qCuw0vicCj=R_oZR;ycG7XXF_L4CQ`n#M|FA}2Cd^b zXDJS0SD3^2IuuF!qH*Fu0)F%R3wjgHIc~|A7n+K_yx+DslZ0O8h|CH~LCunQ)*-|| zR5OmXPMk&kJq#byVsLM93?iAiFlHjNMG(6*;fzkkH*~H{P+aed`N7E6fnhLu}zP^^HFqbkB zvp%IDJT)CxS~H>XEf9|*(h2RMpc;BB1>wg&taL+(#Y6eL_+3D7#c=JV(5`bRBwo226#A5 zdB5us>%bE*`f3*L zY|MbRMmmyHn1`ht&wGs+;B_p@M#Mr{F$pIDMSaT~1@6Sg=vzNbfT?z14 zi-xRtHb#$1M(pJzU~Dp~XQd+3CmGko(lJ*j3#)aq@%()rnt!H2{$?g-Z_dI$ZSEZ@ zCEzGuKdYZ|RuOwPq*^nO`!*h?X`GLwl#ZRnd=1!V!iM`Z2@47^F(4C@GqPbYJ{wlf znTXw$jn=YcwB%;;yfzQgTQV`xIuof4nRp+Zh1;9+@%q{goZFs{{jam&6q3#QuN$n* zDdOuh8>Wxau>V3llx1#kehK?6v$OGGT{4DQreeG@d&apxH|#xUx{0xW@nagc?BU+h z`V7nv%w(Tb4l1g$@S-vc`OB&?UGoiEc{Xl!p$fJym|JwG5$bjCVWrdx*^%v>#ngp& z6S|;~QP2G3cC4M%jxjxLFw^hAvc@jdPj2VlY$t>db~78a2Z?qAQ1MTPQBA!{f&_*x13EN8c|vQ~4Ro{(eTW=2!eu zX+Xu9W<+RkR;KGuc=t7Ojo1n02c6(lYWz6dgOu149-b|BvfuhERD8d}>U0}SezhQOely%{e`E5#PBerzVe7VfTt8Nal@*+K?Ny7@ zi@#w1%tqv#;97S~H}~7yv8J#Cb98&~HKr3LW_{32>cX+%Z8#k#K(j@Aad7<)oIBS8 z@&CRfMTvEBv%BFv@hcjie}oUu9TXxOdG_%i=LvsDXxbk*PV0uhW-s=x_>OI=eYm%s z_pcs27x3w1Ey@qnaXp!wBt#nX{$cQ$0A1WDNI8m}-)SL0IYk2GJncK=mb7DMcQ?YC zZeytPL-YieL+8a~<`Lb;^3$Aq6Z{A{Ey&+y*;B{nb{EO^@!gv#85o(g** zYf3QIgY$oGK7z;gn+Uq}2!qXe_$gKj(lNSn|WQyjgL`W@*2wBkFoE2IhM(mNVajeG0$jRZ!Sgj*h=I5Saf8H4QJA4gLg~Tb|%7uN!}AIFCZ; zHiA{2An;}xcFEMBd(8{X-dl#X?(Fq6dxIO3KcITB99pu~tTQOb+>8pm{QeS`Y+k~; zqY5h&9z(J96IyrHqeSsJ-dDbWkVYl@{NLbT?HgRmE5pqRFEJs#3VY?taW9%PADEq{ znN-2qiRGyK_z+7Xo*>}i3wU4ucwXh>@gnN!+lE4oo+ zP2J`;tlzYv3>j;hkZDI1vR3rI)Pi!2EJ&@$n9g|WQ?|7(MV!Qz#1e~SCu{tER=p_I52`cM6OBRL)$$H%hipxDoamwdNwfQt@IGv*fD*E*A zkpao-7&9}|oZfN|?9De@l8>^XkAG}v(>Vt!`DsJvT`g(H4jU>_${o5WhXL_{->ON|#Md0z+1Y}BN;=r^dXms#v9Lj|4$0StpOlQ^I zRP_JO!qInm$eWapBggNeme~eT{n>cQHM=!42`>AjgVOVGbaFNZ;tKHmdOFm8^6Q@c zid@~#IPl~>rgm3Du<-*VCskv`=qbDwnGT_p zj{~gH7ofz+LZtO&5Vv_9b2RCKc11r-#oLiA`4jf}U6}uZ^>AgaxcIOh5`9hRFZqVZ zORZRG{tZUgdobo{2R5#2!ZV`|Ol#r#Be@F_Yrk<;b`N$XwLsjz1^Ig$@i(~%&24R% z`-W?me=YbpyAS^sv@vh%JN6s!dkg7-#?c-u;Kv);wcx%^BW$`Fv1d~|)-C*kf#YA0 zAk_$gqOXuS*@njMMwoB=h*k3SIDWVmQQh?j4Ec&{8nw{Os%NI$SFFDL5ypNEa0+U} z7okS@EN@~>3ulwa{Ng@s58`J1Mx765o9OkTIq4^=bbeu5&>%M2{9(RYFXSu+aBtK< z&Z-uqsY`}X-?IVKP7$EDDgyLLREVCN{Dsi-zi=KVNW}e~^NR;DJBqW{1q4W7;{ZB; z_M+j+09+LXNl#0V(&hd^d$}OFY#u`9>jX)Kbsfc_MmQm6f!F%i(af6V6Bn%U?u{Ad zUT4mZojLS^&9K+Q0juAzH$UAR5$P7V$bPj?(wDK|xH0?G^dYlS7f;LeQPXb#n+505 zqsQ9rs#7qLIF7wN2ib$Y4+$-M;j-ieI%+R+&N^$#pPYqMogRk1zl4Tk1`t_!6^bLz zL3_Os0v}z1{!jzdjMYWb6ZX&DWWCv96C53R4PlDLobzFYw$auw(y+$%pV!bhg7w+| z4DeIL2+J~zQIv0t2_ub}5n+l6``J5ab&BsVXK;4*Y1HgLjp1GwvD@!5_Q>nusLy#w zK;qx`Jz;F2W@CDx$yZLB#M1GTbbXsCx|t(&pIeWQq%d2AF=z0=qX~ zgQeLOOcS^Oo7Rg+JE)7+b{)uSnd7LHHO{g2Ey}_X5A2-a?PG^??1_z{A)k9#c84XMd}2=bUt3X)ss#zD zTT_1(*H-te=#PO7&6l&Hgj38?{&<~kjK4f8SYNbf^Kx?h$~(G&xsP5(>MK^D@_!4Bd<7jD$DU8<>9XM zUx^#Nnd3+;Hue-AZA*t2S<%&fRunSLjyY3yw1NB0&oo@<*kLcyKH)~oF1gVU4;R{O z?nq0nIZ|hn3;9~x(z!-U3cX}adl%V~LZUq#*y=#`Ld;Gc>Oi~XooUiI2Pzh{B>`JI z%4@cwI$tw#ce5oWJ|{EMtw`XR75z-KB3E5Yj*T&<I%lbOIpH?TlDAJ;t6@rP@$%=52sEf3{~rL3bT2B#Wtq_2mW!n)^RT`m6*H7_kknm(C2Cn%#O&fyjRM@!xrIj7r1@Phf;BUZ zs+;cN_reDluThMknGZ29<}McPD`sBIBka$42$TQHVJ}&Ndk-H&tnv}s1&Z+bWibZd zKVin;6P#LbANPVDL)p0uf1P? z=!YQsP75JWN0{^fek+Y#`A?~a`@w*Xd%p8dte`#3SNMOlH32Z7G zj)pQR2+B#}?GhIPB%-#rL{M;#uU_Q=YV}62x_-WpvzGhIlclIxFe1v;bAyDP8wNqqw(s8H1}F$ zaNlA)LUhOC$I{WLsvCx#wo<6Klz{(uDM)+}#t7SC7;%0iDy>IBdq5IeqT>)0JO;nV zjb!iFDD1m40yA|+BG7m^XIhVhx#}ovc9eo$^=Rb%md2@ZV=!UwIArIHMA;=-yxlPx ztt-YM&}JNJdeTt#ArBkBB;%_69jx1wi;~;iixy2njz}^dvsO>@UW$XsimViZN>!#OPt zQ({v%S0*3J_+D}+JpxLD$vBsv0*#N!NKL=N*TyZJV4Y(|aXzw!F!QD~A8V^_;3$8O zSNasd-u4deEKP$^N)o1ar~&@%1U` zh^o1b>rd|D&2VO;9pZcR{ri~ISBRZg9wPlo2`+MPNdT2t-&_d)rN!)fdWfpQn^-UO z0EI0#nS1dFg;$Djv-~FJf4_yY2Y0#epNm@6EEIh&!rF);h-tAu=cfboS9+ndJrspJ z|N2l7jrYt(isxLV_#;tBOYn#3x(Imr$1%Gi82e5|W9*^e|FNd0&hmqRu0N)EaUNZj zFP0_oJ@;AwVsqngZhjbSvm((jG8>s@;n@E#0X6#3SU)Tfr!<2xf^&&_ANg|DnK!KN z@sEZt8m>m6|64FLr8z$=n|YS&U6GjPhI8oXec= zheL%vSnw(k--EodY+)F3w3x}^6~;N4{umYJiNbARyw?xMT)vL(iTL5Uvlq0usD@qFvgUY-+*g`=*4GYx;|;JH6DKZ|0ZEt`nAgct-ic%%1B z2>R!w;Zg)=pJ~K%&oKpo;n7&k*ZUk*zIJ+2F?VbfMPj+ z4pW$+HrE{|e4J6qUU<_C_NVjQRcbMFyh=DPP23x!ZQ1{R*o!&roJ&^`jG&3ZnC!$p z8lD~NwfkeapfCDXd2@Z@4S8j6EFdqu@D0MAJ%RZ9&mWtdLV11`0EtOKc*&p5{zZXs zX0Oh-QXi}x;){yC9&nf&gx{ClarmVxdS)@}_PY!I$#_6ogV_P=JdkPX2j{Vzlf^k> zRj~nx77fG*_I44mZzj+a@A&=@?#_B-Vs6^VNGz)M$A}pLn9QE7FA3~7qY&Jw@WmAS zD4g9Dh0)zHoGTuUX;XO4J(W3X{C(eaApxb_b5fX=j<1`Nc;>~K1q)O0YgZy`wKpf1Zy+yXN9) z(@Z?ongh2@yw_BpjZnqeXdjq|2ScV|<;uBe?pc6&59Z+1;5^KmyZ}Yz^B{F_1uSP0 zXN#!g}nlIDI*CT~}eII^ZUyh^KdfC+n0jD64|o%uQNzW&`@KszTxOdPFnZ zN$tPoXem{~zbEVmUbP%Ue*uBk%OID{&y%8vtEf zzzTDQ+v=$!xKD$9e;e?2uL|t8DPyBP;h(MwHosWMOzd^|bwq=G)vIv!9&@0UE90j5 z8iX8L$5{s|eBG{LE#O+nimrxk<0=euS`VWu{F;@jm{(7@9lDY4lN<2M2)L!b9`{$T z#{n4?2y`kVrI*=HX3Ow`RzYujBE(t45IdD?_rpB%a zFQ0%j{P&IE%n>Bu-{V-=$3&rdT^uUaqfm7`3<~WagcRvGzhZ9rAjqn9ViS zl-dLYa7M+IlhK&JhchKQV=!quYf3~C@zk4ZE#EX8{gr{r>o;+GB{Nk^Z*f*R@4*({ zL3Q|Dq#et{hoSk1|4@MZZCQ9Yg=>?z45-dZM#GnQ?)@erxjhk`>S;K4fHeTo>A1t^ zZesvz0P0e3J~Rb;pQNGjLk^PDbD-Oth0~MMalwY?#I>ntJeP?MCtfQIr0G-fQ2JOj zjJlr;qXJ1u)+$NTc`GRjtCAqy)si%0j~KP&52biDak_JR24ZB&D!c z&pAe#;?$)m=JiO{EJ>5zN-6reOOhrvNz=gzQq)<_{~kD!HoY81GZ#uxnyVDO(HKs} zLXvdRZa9f=8p>KTNiyz`B>#iMsZCIlESsh1RlrEfu8^WzH>IgsQi@`)NRx8zNQxRK zP1j|lDMwU>e$O9Hk8>nx^BE~pFCR`L__)tKWt|=KGUm#QQkwKonsZm2MEQOFwiBkK zrDCKdDnhH*Ns#D6QMwu=NQT*>(MX#09m+G8(9Cyz5jWr+k$csGoiE{&uSd>lpmqx5AMIggX1jJrc=ll2Iy3m2o{ z?xEE8K%CmLB*?E(oQ{@>lh`UTdVWEaj*k?lAU+Qkx+5vZd^oMH7)~Ki#7Mhw7@ZN7 zrvACo6#h|~b*;n6Zle^f${9}KLq^iu04bXLbtK&wDnm*B#>_o3L4U6~9>=k_cF-IZ zW34b-)D%_jCeZW0g!C)d@LI+QaamUo;$#JjLsl48Z3iEJYgEeWqe;yKgR}T|EX|Kqk8sz8d&1XYja-PP4Ote99xuZQ5a~2=wBC@t#k$3 zJTBvqvkpqG7$Pb1GOD*<$M8c&aML!$iW4R{&K}0;3=VyOv@0~4t&g0jF41$1z%P5FgWZA+LR0+ z_eBShB?j1F)(lq zCe!p;hj9T(;|%d?(sjJ8vw%&kF%nl9!K3yx9vGj+h@Y3y^!Fkn{#?QCmP^oibq%SF zHW)2$idbVC#2&ZCoi+yq%y3{Hh%Kz#EpV^UlzSq2obh@UlHQl0cJLBZvkb7K%!oNc zMi_0W50f$j)c$uB;+4J_zS|S#XMK=0!4qQaF+DcH4|Ru`>vY`<7I2>Fg&=0hixEI9(m%JiXWHqloZxei&R*Tb+rF@1CRGU6`i-g{_}Jx(R&igP;(C1DpHRHp9E*QKq4>x(-|&u5 zl>Z3F75PAvvCijJ$h0nXvG_EC&gdjpb1^1W7Bj8pTR&0;K_T!Oo;JPpN7T1SM*sJ$24&H&By|o|} zQVOgUdzB7H-oKBNNI~Yt6nGCLAXq;QcC7V^`W%hT7lUy`hc$T1LP_&XDD@qSqDRe9 zG-6K##oUgjsZS!wCq10J!ei-ZU=&^189~J&!Q`kLNIpg0^kJ_T&HLg*k%jIwBEXy8 zMte}u0Z)o#&8*#bcRFO^O<%sUU#8rH`Yw1g57v*48u-!ltpQ|P0=xRKlh4{CRHp>^k6>Ass2O_}aYTSFYFdZa6L zcX`m7x$bmY#hY?7JZbKBA3E^PoBsBAQSU(?5?Avh`zUW(AL38anL)HVC5C2Y$Iys( zv1BC`N7`};WLg$Y-*?B-w(D`Ua5d-O$i!0H^eA#@iKR~dJGDko?2!oiPc(#vd=H@0 z&i*ttIFLNN0_dn$F#XuUc`rv}=*i`1+P)~7-tUbdV~sFs7YnE3ejy~>A4)4zf+_!; zKP@QrAQLTLQmpc&ukL~LdvYj^E({@4yHM&Z4J3sZL1Zo$P78;IQT#4{Ix-?;!MMIF5PymTaAPg!*pjz(NCis0MN- z=t4*A`0W5k*h7-DKF+$jU~H8$?x(t9+g%SN*D((*nKLQRxT52ZH>k}6zpGr4mh6ai zaSl)scEQz^&bXQG$((LxMQexREo*Y^M)7wmkZ1en_+C>HgpYhLYUjQGf&v$mPj$t? z%O1?rI}-^ML$SAM9%K#7#*rJkj@JJ{V`GFAYG;XwD=1$T{R1 zJ_wNxK=ME^PUdpfx^MtG?0xWSs~>Ao{h)K6f3L+AbvuI5p%Dy?gb3LFA7x)24(0!a zO~@`0B_&%?Qp9tP^Vp(7i_%_}+ocRMmex~PdQW-D6 zhkqNzd*%M5Ab}C4<^DmlyMNHjwh=ny|C1cm1$p?w<~eEvc$e%^{Je^Pbe8$X$uhq- zr>{TAH2Vhyp83HtVTZ}Jx}U0k4U%l?A2yXdLaWXWk>KM&TA%-y9M=Az{`nHOL7uMrAj?|Yk;AU`AT zC+SWX=Kr%5GV>YPBV>HVZII<-_whjMiSQSrAImJy1AF^q?o@Q(|X)JJxo0#!xUNF zLp#=YlYajvQV;k>4rkv|hyPdd6Ke@dvN_UW| zV!qPardG0e)Ufp=7~qnT1On@D81 zo%)%+q-|pt29a43g%gu*G6Ze-&3MTD{YQxBcCTvnIFMRlAha67uQx%*Vz`* zXn98}mY>Kepq1)!8cE?xBlY^eqY&M0y3DkqCq8!4tmrRvVinW1j_IT_hmT~wsf)ZX z{Uh<0uW9S%c9J>&h2b^tX>WZ8<^O0Q&7^;H`%Mqor~IK$rv&&7-@Z}G4K|(N@`t2G zM)9_}Li{`aCuz0~kTLYr?N!2j-&*#&I?aAh7X|nPfnT)d$3Gg3XW66Tk0>GVDXHFV zAdQ9>={7v3{%I8q+k3<` zJk{*`!=7hZ4e73ULCvq4XxZ9&O21r9rVA?QzVAcwZL6RY8V|{;xQs3=zfVP5p3p*_ zVlq2=i>5N~gKCL8bY9^utG`MZ=Tb)b|B8uwe~VlWmQwD?68bJ(O7ZJ&k^0Gdl(Xpu z8GGJjo=L^@fa#?EUcN=+DsGU3T`_H7{6PDfd*pEO0ekJcG^4JB49}NR?d!+1D6fiM z46$R|QA35526D zg(I`s?EV~nWan(YNoFR0qDY%>-Y}itwqOSDq%?ycoHLV`J*>}1<{I*63{82puRQ;A zyCwfsZ6R-;ypZ1>Ys=>LZTR48^ZEJbjCs4=mb_!K2|udKgpW}$;=Nprc_CFpKISU& z-+s^G@154+AFY_f>+hSxzl+x6(_{_!+9rK|ldU1&`qzlJd}YeJZ8c&3)8>46uo)lK zJC_$5XTm31nDUc!4f#!`hP?W61HQc0fVZoh%@;Yc|DOr`s3s%6PR@iEy>85}@*;k1 zsR93Ai~%q3k>h=u4ETo==kPBt>hcZ3`n;*C0Uut+^VTuMca1gU``k_Ww^~Mgxugl7 z^@HR8!YuZ=%;D{tb@?;Id5uOB{#3FB?}c-D z&Fk~|AF7u8i*x3DY3w}y-2|3-zsicQnY4hvQ)|tCuCwApq;2?%8*KO&we$GGVhg@u znFarm9iv|?_g(S4CI7~5K0ibYdEXuL`GxZHd0RFw`?hZ$A8Ih4SGF?ZxosxAN*eG# zQ;^?&V-s~Yd9sXJH`>{JNjARF_(|6RPDzN}bM(-&;>Hv8&mPjy>}Ybf(8ej#MtViXH|!lX>@lG+)b=wkbK&6fXys ztGR+KySKHTeOR&&e=_6)59tJ!A_>V z-9eX5?w~nGcT;$KBzd$&Q#*!};T2DEc)Nw=wQM29$L>`4cPkCBHPo=7;U9nB#ot>L%yKOJ`7@^B{Aren?yt}C-4g=% zBZEGC+0FnyY98}R@AKg;blmwXdTzY$(hYoKy*)oG&57S>vyR_$WgYJkzK#$0;=q@W zU(3HYcIM@7Zs0+3BY#}Up0_t$&2PB1mjC)_9p93;k=I}z@#Xrf`RH&*KDKW?uf5_w ze(ioYUTl;Xe_`8Z{?Io!eu6Puqp~O8bHJZ3RNca754rLauesDU2&V%oE z@!<=;`tThy{CSHIe>St~%ZDoa@V^v;_(HXCUSpgefB#e<|3BUOjh!KUjCuf{T)Ble zSNGt%&U*42eZBeCm0S4aO)h-HtTnuD_G(^Pa}{rt=fXcyVBYvEJ$NON&3sj{Gw-A6 z!8a+o@>_Pf^Xq$E`T3_^`38@z{Ny@!zR_?qzxLW@K114*f8Mf{&vf$Q)9S)_^OPt) zTz4lwxyFy*eQZ0=d4%yYNuhjheF&dxu!r}t4&e_-gz+82yLk7l5&XVY5&XUNp}eo} zZvLce2%j0TgEvy%&d;9j&-0Q#yd%#%xgRngI@ZVinx959cO9X=(`-iTRt8xvW1iA2 zN2yBl2xa@GQ#GWq`9anrG0a7GVj2lDu1qHO0P*`%NpvU6_bEuHac7Rv@Ut|Egk#iZ z$Go(QSO$+~3iy39C=3f{*ua^hL`QYPKmaD+xr%_jM4nRLF8VKH+U=ckqae?7-{X{OR7k;B9n zr_=dO>GUi-jgAMUkRjuodsxrv8+?S0Pfln1;Xx8(>(8;xpjEpM)32mdrUy@pMa#E00nX z<3bH|vgn)cVVYH%MSiC;$a4YH@rNXnWH!q(9nPZXcaD%MTSG46Bxm-f(Gvy6zg|8- zA5JlU!2@xuhe{?Ju8kzZUr|<43+Y(Z)9BDfdNHPfUKBlHnZ@;_r&&#_9V*Cf+6&4M zeM$Bz^$csRCEfjvB$3@lamKB*py314gm*H`w~HJv_ECe{7aDpyLWTYu0n(^fCaV*Lm6y$@K~MV-r-zlMA}MciTj{EwTd?`I?9(V9tL zjCp6BYNFpOp3_)1`}6NoHA$U)P6@G%EE|l?e=TUC{L9Vce5#HL%A07gyP3@Q*Hg^T zm$c;YOOp88LMIm1lTBzd-R*lte_8#uRQ4%-c~D1fmn&$X`YD^)s-`0kE6DN_%dva> zl$H-ZB$J$HB&PO^W@S7j_2McLaB3o*ea&R!*+i8)Tgc^UEgc$uNwS)agk{w<;#@~P zKVOjR_eRETHBsBiMyg{yOzy!RvR?C^R$%kM&W^r#^c4;xm<3^-nkcov| zWYWbv*(HbQ$`RI!H;z!ux{p+N=>u6S43e1`^HK@vBFV*HNM3@?jF~Wh66GK4cz&S| zQJ+bBQ6I~8D5L__D>QB>pH8w_&#QxMj`{FK(zUrr`G=SmP&J>zyf2YeU_SZ8UZn4@ z^6A<^rujR3mF&J1P>oYD<&_pv($<^Qf95*npDG~F-kUU#bC~xSgkI9wTM2vW_f?+<;-*L78O+8Axop%Y)-R; zru;1>H>2xRTzicwpWk9U#BCaJV4iP>%W10jeR})rKJlp!$lJSwH0y6tW&btumn@)L z!}+u^u#n(FG1VB{Wd2(>sdVZ!N?w13EE9_8M$&c0!Cobw&DZGg5~j7-S3sA?Fz(#y z3fwzJ&40MO5Q=l>#m@ea;D{ zkqlx!dKYf6J?}QHce_S!6>l+5x9hZHq?G&~7gNt-cH8wn)vDhitKSt2=Y2wMSMO2n zA=Vpxy~*&G2Q*ptA>+7jQ^AT7inhK(X@~AnpIsT9>AX!+7B}fZ`8}FhbB|IkSJArQ zXY}HD70vamqH66by4qVwwMsAOq+}y4!lxuLntiu_R#3#RS}J{4L#M;)s5I?4U2c9# zzsxG>!9G8FxHN!qt6S(3%jz*52&Uwdel#a{8>#H^po7g!-%jCF7_yVPcz>$*4JEH9 z!6d#soa%qwCf}gmFn?2@zDuvfX)^P}HSY{h(JaeP`Bc4nf8Aw5y zJ88Lh2>opfBb_$pX`UKE-M@E`wc2)4wc1Ioe?n;f+YoxD8${Y0{YZ*=)V2+WQ*(wt zoeKA%H5g1+&+Z`G5ca&8Tj+tgCtcRqMn`>j(ZlROnsUOA4m^#Z!BAF{?GK<4ALb)J z5=zh4GES_9WzT<$AjJuOEK_JZ(S3i~AH0Kks0P#KIom01j0fEk4Q3fWFXd117yOvUgMFU){#2-U?B4%#Q5HY}7ynJ}TZ1;wc6AAKnN5Zqa%f=+zNAY5K24)1pkJheQ!3!Bg{JJ^i*ZEP#@t8r$|F*trTS1 zG{fJ$Eg+Ec36}iGM&~)Fup{j#-dMa7-@S;$iH)1ktKb}d^u2&rxeK`FswR}TC`0Hs zb+|8a8VY8X!Oid8;3ler`{G2f{@f7EofpL=sR0_-3n2c?!ognn{l;q z5C*;o#NolUxN)*C>bv@2?!kU|{#poq!xZuMcn#cG%i)v(4wc>Ha6`s8ymfvYUVL{A zj9bq_{iOtW5tfP$#%XviJqfQ=zvg}g{o(3wG0~1nBkb0#fnN{0LA*f-k8Bu$uZwEn zyqFdqcASiHJ7qEMX)gZBJBjs^GtuwRQT+5I9jkLQ(A%z@yBF5V6)*V0W!UV;QBZSSHk3!~ZsCQxv=37N#tgS10_t;~qPCfk7 z`Uu@VebBl25wwnKfO8;*&QkMGaNY`(b6bY%wKd?vvT^XCZX7JIjKhIL2T&n)AGU`s zK}E4mm@{P^M*Vbz14DjrZLvL^@*RSQhlXJ|PZEW!}@qE+6rp|Ofg7sBEDWa5#1BUp|jE%EVjRZZ?&)D3>|CSyK52lX`7&zL@j4y z`<3&1HUVr+lyTBYL$uH^!<21_P`xo5#-rZlG1zW79`6`V#Kv2l@Tpn^Jzo3+7hzv~vD*!wEZ&M!dqwcjQ6=;_ zDua$64Z!-r9GKIr1l}iwabDMGG!T%%4Izdo+hUBT-4WLq=Yrb#V{lqC6>>%MIZ=sI zT;+{tT*qU5OjxXoFM9Q{e#%(zlQo;M}9k|{ni2;Ecs9roB zk0i~;m(}y|_C+&X-c$imF4ZvIG8ghEOTqP{U%3BjzH=K!^w4swKGuj$!=OC|7^$Lz zW()PudGHH-TKxw;ZyW)$v?f>vm2gG=7KE9Hf^N7E)Liz14u4Y&kX?dDt=3}ynsTU> zu7sa^AA!6UV#6v9uOHFCMr$MNITZA*=i5;8+l;gB-@%m& z=5YFMA7SFvKCr0of#)w?!#xjSl-@Q0zgn+_AHfm8eF%ijy0dZgs0FD0zyv!ySK|El zj;N`<4LJ=!$IE~}CL3U^UKqUV@&{V*4Irz}C`_|>SK;0GW4 z*TRL;#gMmhJ_Kd21e4b3;1qlr{%RLPqWdlQ9R7>Dq$CfsM=yi7(;vZz;0L&3@*TQA zj>B$+@fdh$5^fnXh3ZBtXnhWF`r-!&<_BRj3_^sz3idZo!KaaGm|5P&jYx?@%+fJ1 zJ#7IV4qk}$a*o)rb~pSv9S4u%_d>bqG0XsKGtH1(c$|1x z5ALRF!FTg{kdUWzKnJyUQ15>SL5h7) z=(-rsw=Kn|%1d$BPaWI-8;8h#KW3ACA*9{~6iYlH9wkveXdIf%7>_3oYvE~gUF^6s z2d}oQ#_U(C(X7-FEf1G*TTQCC(f#+iCQB(a5|l^xDJr;ZzyNpMGQ#4O^N?Gf36qZ< z0m)nW5Hzb0k`5Ju@|cIf#180mLkdg1g;D?cGSGpoP!hKWNNW*X`RoXL=4^m3fpzdt zt{QZ*8sW?1v3TY2I24OjMd@b?F!iP_#zZ@!;D-h{6VwJ58b5=2$$jow%n)~LL#;PJ=-`Yt-5Q=cbJm*0UWO2qI+-8c*%D}&3YN}%mb8O$nE#M++@pwqq` z^5TwypWIZmx?qeZo6RwGuP%gl%0lk1S6r7{33v+L2J?4k;L*i}h}#|TV>`ua>&2qRybPDd>CWF3xuVG0`D|9cb z0j-U4*zY?Hqmr~SO1l>>*$#ke+z3>3>tKqR0gf4?jY>^Fxv5(-ib}tu;Ih*$2tO7MpJoA`7`p^RlI(GpMh~Rg|AN2sMX^!* z7bveC0ZofP@OOeX)>+NLWSto(xY);Ngi;P_2F#=Gn%Af+Rw7 z{x~pn8Rk^+7<^2~0hPv7*lN2T1QmTiBR?80R~!b{6`AntMm9KPEWmkN@xHSOq8es+l`{T>xy#wP4;-bI35Xf$p?^P!|3I z;(vd^s59(2B@RK2&`}un>j1|yeXz{`3sltR!Bo#um_4B!UT1W2lUDxY&L+q}#@j?V z@gf}dr9?x^LqRA~lZLyh^6=8Y5th_Bf~wtG5I3lYDQDh*M{ot)pW4Z-i5?B21G7M) zE(2X zoM)TEC0}N#@o(k9a3u$erY!-3I}WhzTr@mZih**ka40l*$({Kn0@q)?KhI4*u4fiC)9$m?Jn+PZ8R78=pc8#(;eIkJz%401eBkB3>DmSn04kQB#x1Q z+v8;+JX{fW9ZQ6Xn`7W%_C8o?GYH!&2jCGq$IV)P6wEH31v9|{m_2n9{9Y9S_YcOv z(yUmB{GI~uAEiUd<`QVX{}8y&2cT)t0W;t|WCZuXG56x~b zorkGWnILVI4f|dvK`E}S6bZ?6c{ zE*OH>uZAF0y&JB!3_(#`FU&BihNl=NO4(A$UM)Api5ZU|n^3&&6MeR0;$aMatn52v>8$63xjFv)8e z+Pp{M)+!ZD%~QjAOC|hi6NgK>$OC${BCU{9qK@{a}3@9Ppcx5f-IrbvK@l`-mW zSb}xu%?0tR^FzApCUX`}Lf zS*#o{gI=J>yGBYjuZ1B-ES7$n?4=_9N8Xw$R6iMFT{s87Gu%a1^B*L z2iJMHp`G7$jCJ)z(OhAS6q|sb`)A^Z*p(>!bt5J_ufxi;hY7y7N*fMlM+aK<_`IA(`Fx2*B7@eh#i7=&5ue0bC;55kpi!^t&wA)@jy z_|@%)-VIye^HE=jy5j{KQ#|48@D}`G6o5Kb5m=R>i&2vev0{e~n&z*^)q^WBsTpyB zW&r9o_~Yu9-MCE51##t0Y!Hn^5z|Oq_uoFWoV^#P+zdyp#u)VVj6}Q7erU8j3=OA5 zqL(C(8v^ao@`59Z&rF0ok!(0+e+CxulTiNQEG+L-!@7l8STJxF>&z~ohvHUHJRbzU zF;P&p@(AqsdmS`c|0ZP~fM>a=$f#zZuFL-^=RF7%Y+vM`NW-ROrt_$=%2CdU`V$o@s~86IlJtdWA)g z{o%v%SlH_DmfNfMgfrpg!S%E>uKY0;#n~~r(teMVbh*nN3p&R61WE$mAPG_re{hqx ze&*^H7I8N-&T@&bcEH>=Kj>YS0jb-Zz~)de^dm`iYJG;H#vW}`zkRoP1O$uzXifp*C<$) zxeG21?uJm=r(j_75*n^OgR#{gxHC_BxKZzZatRk-g2liGxSZ4qC*=fTr?3pnW;KfR znrJRGHj0~YE`?jR{wie07s1Ae+n}V+!oe^&+_8%m(k z;67|G7sQ75zu=YlU-;*l0!}>#K;}m>$gX<>2iP@z*#{`OAP2R3^daHdeAwg3_R@(- zkS=%vZZ}L~d(SjjA7lyLN;hEP-D{xkb003J2;osJLHtg_$dn$Ox|H z#kY#cb-d@fzM5H_>1PYD)td@h{!^hgs}A&@vEI|<9auZtLw&+VXjmTxqdFrY_+U7E z=VCzZWEbqs`vl^RU!k>oJT&~1gy3*lxbSZy=qfpbr-LI@%hf}$P#q)*J%;?M0nm8% z6&iOB!5}Av-BqHvp^RZ;D=xzltvfJmR0dN%r*Ox(tK3n|Y|gv58l1A8KveugsCS43 z_l`6;9v=?@!x1oJ`9APmoe0DC6}hQPHMxa9)HtVw=Rj?7E_>bUP+WNkZkDHjN83>- z4IYifZ$vR%N)G>}ti$I&ZLvev3YA6-An%hUobTQWBX3>ta!)AgOo>3NZ{u-h&jgHP zeUrrv0nFad>QBQq5H=Hql-sYk9ep=Bt5;5tspbXa7REqjgd)DORl`3n+Sqy54`quZ za7Zx(TU$1wuF-CMmg|oN>gsseLKR=yO+qEbwJ3ke3nRwt#Hp%t7)EM|zf>&nZ1!0A zFH;XL>&=Atdt*^8RvPVmrl8N7TM#Y!0PMo9K#l7L=v(FvZ>?>ip-C6*MkC&FFhO6j zFq}Ry3=7x!V(&%?oW58Jmu89K?B0W13Ekqh9ska~f8vQYTH8=6+YK8senYA4ACMFI z1%;tP=w~I4|JKUljJ{7$Tha}=8{WWh{~mmCBOFE7Y($m68}RKyXH5OQ2$k2^VV>a{ zd@ZyREyC}?Ip^1~=C2sO71codoilKj$3!&T{twDpMX)7h9IiRH88;QUqo(O9tYS6G zW*>FrRn*Y(>Ur=jy$gp{HABvqrMT(wX8bnV9}fmcoF7G*~6mGCm?b{7F={sglnzdXz${MuPef^ziA4#?3{^mU-eK!a3+d4siCx- zG7hFzfqz#iD873DOJAj8^M)L}(R31RRuEiVVF|;p*1-Gxr(hO(1Io6Xf&008amYIf zt==W!kv<2MJn4?6jXQA4lRc;`5rfueV{x@b9PSTJ#nQ|~oYv-ydp)+HiIh9WmEVT) zfO_~dc?9evGQeqJ9(e7}2WRm+AY*wSau(c!$hk& z(LnqoCsi^LWanA{8n?hgvyTvaZxmWy3I(a7n?W&YDa36Ehwg#|c$$y^YiE9jhH3pE z?KcEz%YE_X%P5o{iAHLefU?a~(bZ%+p3{wlso`lb+V~hG9*n_tYYrkNC1L(nY5eOzA(19qGBKykjHo8;ZbfVScFatq@(;-@+vuZsPK*<(#wbInMH98dr5L z1a25JJYq^T+&CtTM{MMA^UzpKR9}kbs+-Y8%LCsihNF5#EH-Tl$4vvKDB7_Bk1KA$ z{Birx_E0LG{FQ;f)x06NBLQCfrNV=dX>jw08F1ycuxZ{uknIydd{_-O&ZD5ZLmd7| z|Ki?$nvLP2v+(Bkndog(49@Gy+4sH|jyL>=S_OOYZu>!8vvV@4l09E&0u z945NzVTW2YJUJK(?Vd*=+p_?48j2uZ_7d!6*x}cP9oT$666byW2Fn7+;S*s6yf@#M z;i-Eel<^}nM`wY4@nSgnnuqz<{P5QF-S}Xp4?2xc!;>jlcyrldOw{&7tN0+an#=He zO%cr0lfyC<1$-c6fsux5aYu+98m&u!NhXode=#21ZiQfzQ5??CNWlE0ZD?&3gW9vA z@WLfCEb%wT?F%>zEMjXJ)Wd%@gcJTv!#GC<-x^t>@U;o(cWo9Rd!%!oA<)#llfSaUWGqZo?v|6HSV?cZ_eSxLU3LD7v2Sr!DG6bXqzsNk#g*u{pl6h zEVYAo!}B3*ffX3=U${Nu9h~pk5pHwgV|=@*5$(sdW5xKZ=o4FpHytamcR!$NIbnW+ z0X{5xk3+q^m^b}BF4hczPi#*Zf8s8@+}@4*WWV5#XZ`p5W-h8`!sO06X%! zvGDvS^bA+v+)l`Inuq1MW)ycvFj{TaNSZb>P3|@0fpf5KAlnay8{*uryH#gg?%O(DFDq(R>}I z`KThjmqSIZFL3$cA1->yJ8qhEHz#*w2Cf)A8@=Alz)81F)Ujo|~$MnOs$Ucbqa0~itPvG^LXYr_EDvo#g zgx*>|uspmAzaP4Y{qBW0F8vP9ZE*d+d&4YB0khY~P-W3|R1P_b6Iy2A+xZ+GRX4(| z=?)m_xdkhRHsIM0htWFrF#6|2v2#cQ{?kgtN2Aytzi1T-KHGuer@|2o24J{Z9GAt< z!4m>z82Wq#ZkoOpld5aE-rJqr#Mqx4XIhI{wQumO%qNB$Uc(D-D)CGDQw-id4o9?S zVetWNTw?PW3fn6o(YOFUMpk1;L_0dL$A9!ZjsBDFpwi?boc-TPt}EyrCoA8_Nd+fi z+lEYxynO(>7ocd8 zI|XtJq|o*Wa~x_>z#k`t;8kQN*CbWS4aYQa&eMg!JhGh=uK5Hju5`kBc5Lz+6d~?| zB#cdB^=9EgNGeN#N%g5PeNPrBEzO4WXU@W#`Tg)_{Qoc+0gT}EQFMsyll_+X@cDR5 zTd9u4hbN&(eK99gQpq`#RdFf~_V{6oHNG5Uh51KL!%?ez7?^SzCYp6~8nVM&W>^b1 z_vcM0vwsK%^U5H=UKICKN#K$WDYOd}K>rCNpntFzymE%QArl!e)tm|i2gFhDi2(k3 zG8%tJN}$Nru^5t~fRFbmKvvXbINLoA-UmqtS@R8~MOxw4?N6ZJ`VpKCi{OHI zaeQ1b8YLRspsKrp`W6N(rd^8(7NOQ3#bGmMOG1mPF201Dqiq)#}U zHi&^OTVtV9M*_2&rLjOt5;NxBg|4V$Fej%FuG}Anp3Kph?kkU_PCSgYGKS8(%V4@z z7zigG0Ph`p;gg^swg(C065Ro?R4~AH#!KB2R7Qp0zd_P@5RC1Az^#TaaG~%Qteho+ z+inEH@1=}qmI#Glizet$st3hAt>CoR6a~H#7JRYBtT8u0FYOX+++75o*AGJC(-?67 z>k2!1S)ICZ3TkfEM9qDZaYL^P&goV~`TVzV$hrf>wtt0)wp?hLcn&U)xe5Vdf8d33 z59F@+0v&^+ap87BY)lr%k-&A(Il~X`#_a>q9)OO?nxL(s1-S<*L123|Ov@aA6Ag1= zd+i!HcW4((@!rA3&pFIZH`&WA{cjswDtCdx8P;&+_!&45cLLg1rh%!n2#z`;hpLm6 z@X4LK&~JAG`fogj!(S~SD8dxPo>+rO>NHr8J{O*~o5FI>xtJ3-4;NQ1MDNT|aP)x~ z$XbqY-Hundn_ey4zxBVlp2F|2##;=-YZXvfTo55n7H(F%|Zv2Ger{C2Lqr-S2*D{T=A`R7Js34UC_qhqKIof#Bv*c=w$kehtcko-rrjlVlQ%J2V6a3M1g4 z`4&z&Ohm7WaTwO2jMp|u;B_%kOrJjnV+@XPtNX5S4-{T<#?IoXHctU_kISP)Fo!{# z0Hc=7!{Lu(aL-3^ED#)p$t$&S!m()>yg>v17QE(k-Uxxa{3Nj5Vh{Ay7Q}x?LZuX7 zyR!k7N-V*kbyHBfS{G-V&qB4lr?ATJIRqT2hpVr$V3~6f2y488>7z8z{Dl_Ai)o{z z>2h?mUV#~l*WmW^^4PF*4qEQC!70@@LFC?L7+L=a26O^IZAUbGa)^ZN9n&%LsSXz1 zQ^(weaS)Vp2s965!xsg>>brCB+Btn(cj*qS4sM302cLn`M@e+_n~2w>CbRle0w0$s zVp84&d^xxec3NM61LMo!ib^xrakY_~{q!w2D_9uTIjcf-mNm$~&x5XIw_uulFSy&; z;~0Ned}QN`|8bAN$F&@~he{xp)!Uiv(l~CB81}A^#+cpm=)%~JM(uI9V@L-h1sPvh z{~HeP`UA$z0yt7^f?p*~uxZ^Q6uQ3`ehwT4nFD9xx%duL{<)piWN~O=GadA2&H@`p z9k_5?0M*STanesAe6uqgUAetzSDJ_>R}O-$_bIUGJP&d$e?f>n?uKR$wD+m7`dSyY zu9={l&Hz-%e}^4xFI>O79kOaa!voPSFpwI9$ER!JypQ_0()unpMdCOYu|JYqxla!- zh8W>`)#W%K9}NxP5 z2j@fqw&zWQ&**Gqa)!Kz7=snRIReDBYQ|o9P zlOT)JJzJsR{YQAL_Z9xK*AqO!;hC*7aoVma7;$g{4op@9C7BuO?bNYFKA7W#@7l>$Y|PBnW?buesBOOqqPxo>`|4tJ++u>eqvxW-9fvnG%)Otf#57`~p|FhU zpL#^`Ql=;-_i?bQ#tP)xB0*`T7JgJS!kG8gINtO*w0cy)jFubl-~K3&2uOyiefuG$ zXf|~B&V^3_F3`GQ0Mhm`U5u18hL!$--xuVuqHqeP1k2!LcLkIRk;4ZY6tF0J5~{bV zVUJM@1UYwt{~B>jxx%5X73)!D%yC)aH^xEtfn-J-80!0Ar3Axi5;kGO$%9-$LLL`7 zbth*kQN^vgI|?4&YT!gi4MAOlIG(&LhwqYxU?^q;Hm>;qs|2L*`_!?xDN`ErJbR!p zRSaGJiR1lxQ4EchK(pT?&|cCAxefiWD)JAs#ctuU$0u;T_0PGF)hY18E&+-k?}es~ zMc_K+HVB=*0pafLaLl|F9$LMG2>FS4<-I!E&(*{|c5*ndbt-o3<&b~)2gEnyz zxK-sPC_x8&bY|=MXA6pn&T!cy3@r9c#ZP^jSo2g1mrm5h{-PPU&ydHdjvL|9*m%af z+<~bt?sCe{k8sA_MO=pQ3o!fg6~^X&fZNGx=(toDTQpQrUuZ4PPj*7?+G=b_JPX0m zNucuR92~w`0b5GmLr+6DeE!(O)u@ezgfUX^UZ5FXXupNDD{ny0OA=j-giu9&G*&I2 zji233vD4WUPcbc!mB}^O<#-(QtR|z0>kQV9ahM|~jVH|&ap@~fOsKMf_gk04J+&Ap zE-(g@E-(0S*c)6JPA8qf@IN7AZ275$M~pa(tG2@Lg{lCu9Ra zFxVHoz148%FKXpiGSM_d@vju_i8xP{(Es0}QiMgi%Xo zLPgsISZO*H+pnskvd;tzs~F*YQ%1quoS&Sj+-JzU(+*aABLqGf3&R?skQ6Kp@^h2G z@EXGlea}OB_c^$}`4SvUISuN|`r*TQhEF{gMX8yB7=K3yHH5`6TlNa<6F&>_U(bT| zcva+>*2KhV7WT)D!M;#sjPugNkq<+l`}sFK)DT7ix4kfU=NNGQhd`xX31_-Xqve1! z&Ux;IPg6Fb@&hN_eY6i!|GtD}thYPv>H)u!_k!K@7i$$AQyIDED(BhAg@R z64VYS1fRo@#e6WiJ`2h}*g)*WZJ7Vr6|V{H#7Xz`am_*=%a$9V`;$4G?#6XoU;S3j z^qn4No3py&&m_Dk5&(z9JwfA_E3DqCg;vEg(cvPZP`xl-ijl-9V@VV}{|s(Ts)Lqw zjWEM+Jk+ERNM_A|nY$O^@zu+5*+~a9xMB)}j~9XQrd8lJdNI6gp92u54?{n_I2YHY z+^f}vocV7-EY<3VftWEU!D{Inoej|ZpdQ@6p9W9UQaEi=56%^dFt2$Z)M{tKqK)U_ z>9|q|ef|deE6v&c1#oq&GjvOzhnwe$K=1lX7%eXYW4}y*?Oy6|CF=_8SIz|6(o^tg zoIjZQ#lU-sgW!~t0&4$VfWk{%d=eh!I#{*SKVAd_+^+~S^-ksZJ@ZajGI63 zn!EF+m&?g`1%Cox!M3SgU|}22{oI|*9T|UtTmNt`ygD8YKDx;;PjM$)SRDi3t#aV# zG=@7b7Xa5mDNsL<0NJv~Krk#9SP}&1y}geM(NKZaZDueq%?Of4Erve|apEbvrZ z0MR=ZL(4`}aC>bCjaOg5X_Ydt+w&gcPiBFm-5FTLcz}PiBw@|L{{O?-n@3ai|M9h;InJNkuj&t^Vc)hn0g*2CvQkv65qf#nE5fYg~88d~V6h+1~Yf>uBMLrRwQM0?h zcdh&1U3abf`{VrKEX&&K9DBdt@8|3Jd_K5iHbc<)=CP=|-U785bLf85Ms!gs7T;S7oZbR7Q#ruP}gF%_gX%nFs{hcLYH2+0ngU*08i0&{Pe=RoR-k^jKb(@W9BP`bkM1;o1V9YL+j%F*8YcaW^>Bb1>yog3~S z!iBF$;Et)i<>DSFqVH$qk>zC>Jnn@oezIK;FCC$bk1kQg@22YE-<~@7TD<{w3p2wr zhTlWo#hplFh#ao2*v{RZ7QjuY%;d@gZz9!q&rs3P|Bz$oEcD;_ImkfP4~+>tgnrIQ zMFU@Nqx-7sxiU#CH$JnL`@R8j`3*)aKD)#}^ zC4WZ~XPiRZl!r)b^%GS6{1^&6or-n{@4eIYM)$u*B2Vum^x&lh%D-ubehiF8_k(65 zxkCZyM`bvw9&rj;Pp?7Q*|(5_d@WKwQ-=QRt3~tNZ=;icULiH%Oh0H(Jj!oPL#J!X zP+AJclP^xhTOuc6@#$O0Tm3b%mivfib%vsqE%E4^XD({;%R=AkPNQ|B@1jzf;Ydnz z68hOb40&1Gpo<0s9apwRd((@Ne$`bp+w~rLm{^M1A}^!PA&=0jYv0fe2Wh-CLj^Cs zF(2uEScNhyA`#Oci;kH&q0;MsYMh+W@Bl&UwevuWUiza~tx1C4HVWk&ynxa}8<3Xw zBcwC<5jW}fKkmp*L$u<;85DU~&_eQVqiAjvTJX4x2<7MRL9=us(1q?4B$EHiCA}Do zT$7a0iSHj#ir|mikt~Jx%xgil*B+n?YJF&@S_Jy3a{%pqmWG~33jV;<9`xhwcjQ@U zh1~QBsyAPVy!$VsJtplaZ{-gZ|FIkyMqNcwCGCje_ktE9gS$!vt;tmj4I9a$d-B$3 zbMPeer*I8=u*wfb|JQ|TTx9Tu19CXXaS)FGB8|VA3H##6(Rjp8As?YK4#!&7Am7|( zBt5+oomEcbCdZxRQiBCQRF4!UZx!&*HWj?jb04QRHVGIM(5mhQErMbqn)dT_%eSKwI&AXsqH~_j}wvi@*FhNyZ}uY_L#eC@Qo{} z@8wP{4M&Scg`lBdqR>c}KF+tHkCV)^Mr$Xl;pKk@W2;Zf_*|YX3h#G9Ic}OL=~xQ7 zQjvykl^jMd9v2~T;z<;Iq5uWQ4WKy_WbmIgO8D-Oa&$NP3|by?2AQZz;ox%vsNLZw z`fsjqUvgonMA&=Xe`<&~4->re>x^)=WjM;v3Pex$g`ve#S5UHbJsRfKg0dS%;ZN}+ zFkK?Tb!W=B#qTb2*HcQkk_jy+OPFP<7c`-g$Xe88eGhrqK1J`!PoUv{icnPD33M|+ z8p&>uMw(Tf+|h`8Xo+xtd>Qi)>8z^cyra^&PQy$t!}<`ar!i>Z+{5UC;M|SUw?{vv zZ4p@3qOZdDvH4LsO0myF1EKk7jCm{yj5>j=MVaW|{u5}$!alBI@JBA~%PVfs`z&NN z=?v=JP>nXve#t$)b(?#$`2qLkmJ7N(XcB6E=72slb#dpfKj9YKkU=jLt|CX_92a0v zg-Rlf(Ch86K-a&eiqE>h*1Od2bOqFscmc(@J#lQWiSDeRY zcoaETIG;qw;H)}n9BCno4Xl2lF4rD3^XhMu`_};}Bv_*6g|?{CJ{&oXk43M)q@#Y< zNHp0r7In29M1Gz<$Vt?P-WPmFPY;BlHo?F5?=_ zeYjk~+u%3Z5XClGBG;0?==r!o__K`?=A#^uNq__Tbl(B}a#_Ts51q+9(pkY--4HaR zcf-*2a~RpI`GE3dy3x+wE;M83ARN&wk4G7+;s8}e?A0xeNrATD#XpPQU%H5bZ`Ywo zDQTQ+c0AYH8_b;*_T-#$2J#sbhIYP)MP~VeuJke$T^L=3tf!TuRpdCD=o!y>Wuv`&Iutu*9SZuBh;}J;p~X$T=ygCR(zohBo9o`8`ovz;=$eC`t;<6G z*|o^8%M}m*<&NtOeep6ih8^a(;>TW&IN@my*Q5803!3$w+mPpuAJw|y|9WQOF;}fKjo~jYnJ#T_blZRpfF^^6A9dM+gJ#OpiLIxw>qP3^qAy*GaY})65 zCC5f$^>|x6ZoUW)6Y>d%e~iX4TWxXs;gJ}XOhBxBBpRNG(A)M+!rmwbeg2Yy;sPI| z{d?L`PxO8C26v;Qd*!g1vnqDDVu&42jlr{r48xG{OgI(4LK)+QJ>|jC=wb1akKXM&WkrP;ugR zd~ZcCZjppzGT91Qt`nhqWuuW?m*B_QHyE$F^br{cO~cPkT)8FZ1Det*HejTIJ}~xC*of33~jGCCI}q7$w9-p~gEY==S8} zC}jIpq@-noB2Eky_8A;Hc|Zog-YAh&K>1XsIKn2^KmB-G%HF2}KF5aObgWa76;T0Q{@Nge>^jJ|D&Ct+Bo6|3%$Gs&e zrRW-3FIGpB4#=YFBZJYET_I>pMl^E#up4=|enfQLBb2T41QpIIKvo~lqM9AWDC2bm z>OB^R*15%_$s5|y`pdn@(nAX8-4pJ$@x7>T^lLPwZ8*BPY$!UXpn-0UKZ?w4icvjR zhAthsDtL?TAeYJ}W-GlN1!Jw1JOyfV6lUH`4R3AZe5d1#7f2m=Ca)(mFmm$xpL&(x?BPvc?gLePfik5%c zjodCIqX(bjkoFG~^h3%N#VU?Mm$#{5pM~mp^)nreg=bU%zKC8`Uq*u$TA?jUlhMoZ zp(yi}EWY?z4adIF#A*AoQO(p`w0ma?(&q-FRW33pdGt`Uu3Z`1XQ|`n90hFBT7vv< zA4E&_)6krTYP6uY4VjidMR#xm_txYUH|tR+C%;q|U#&2}?pIClm$QaA>&bBJ!Z_R} z8HuiR8lsOzHYlodHhOS*JPN2Eg?cN_px;BUAcN>uWE|d#a+;r^Vu4hXY}SipD`oM) zg_`(-knxIIHWX{w8DrO|2f~{57Uf;}5A91Bg)}Q9=$_&vB;WfDZ3!5Rdt_wsaBnkg zTW5hy@<(FMZ4^!~kl?}03FpSfqc+WCBvn<0e70{!1ATEQ+~gSAvP1*l?jDK*ErfiE zB$`WV+{5J-?dK|=XyEx~TDW?sF8<~&Mmd$PXq4O>w7onJO${nQJA|27r(Ge^m~EOn2BYeDSI9@q^C|(;7heR$FXx6C~v~6t^vN6j-ak2F%GfvR$ z|4hWgF3!dtg)fn4c@^s4b_7XJJ%?x1m4r*kUy`|5`C*)04d zZwB7ANd+sPFvXkZXyJ*|V({)Av3O_RL9F0qhD)BB;je3@ZiUGLQb(0Y3Iw~*e(s+D&+P&up1s~=7iTs zJ7Cu@4p{&51nm3qC0Z2x0o8o!K}J7q@C9W{Y&+H%Z)~D?gA#|4kWIVv(+yYM$M}@n zFnoN3Ild(5cOI%@9K7KK`UAr3WnT|EA!rd7Jmm4K7hll@b$z@9We7yzb3+qUo zL?6%JIt-8T(Z+-8;?b8ahmdLgI&>sH44WMlJWe6I@v^3;$R_d*in}i8KU-vQud+5~ zO2Qsf$qIkB#`yfKF}OKX9hdA?!)xmXW4)v;xMWN)p5M6#?_O~mWq-Yjrt+nN{{!JQ z8W>l$o8qPEV!Xa<0`|6;h*w;jg}-?%!4=F0V`Pr!P91}z+yD>RqlkmfD&vkOZ9M&D zA9^b=cJ>btQO)`e>ZdfbkzEM|?Mh$3@kfvHAF|xMl7h zY+v*kO5GM@R z7d+dlxZ#ys0DLs$Tq0IqE9UA=}m$8{qp;67JpjArYVkP&g2>zm2QGV5_jPi z|9v=fxCv@pGJtiWI=P*Kr|sA21YCKm7>DeSCFN?L$?#p{Y1;9*q$OLPT8tP4=l@pH zyd_oC=;j98JN+p#{$+!&zPL#@t{e7$^JHfV{j^=hg*4)ZK%W0m47%n%Fzt^fWFE1? zjy|PGUGNIOy{5wwf+@Vp;1pz`jnYZmqG@d1a{kF)XMXUx(fspaXK~XL zb27ptk(6Z*NM_{i6H8@UqYS#8YmqIKjGOb3sgHm6j@O1njhpC9d{mZ(dDft`o*^ zfsMt;x@8P@iTMP6@ADvCv0_Nx5?b_O*ARh98R1rBM>B z_Q-%VED-oNw1XCfzoKPN^_beo;cSntG<&o3fWUHjPd%=yLq$a`)CGKj71n>?k>^)Z z_;3mo^3@O$*b41^3anzD6#KVtJX@6_VYi#-uo)RAAVfa|vab+GR?H)1gYrq!5ic?% z`Ie-iA)kyfxJB99EIL1D9KBGaE3jL}vJ6oL#Qo`_`Ewe;;re2j?`j5H75A}ZwZ+VA zyf2$N{Q~w>xJ(Y2nNtN>YiRNv1P8RPL4rhuTd<-=-0CwJ)oqZ2ds~fQhtXXR z{h$y+-E@%|UOmB5 z%Ma~kcMD?K9q)YRxgr7f@eiP1ts7!88<}R%ZFZ{V2KzEpMBE34;@K1S($&2$*@A_mSp_W>*QSYi`!fCN^AVfUIi@ck1BucpU< z;jm<8eLs$!x|_gYxeXl7GKbaw5Ny5no|*lA%zTDEXP22Gi;_`dQny~gk2`I!Deo;L zk$+HGR>0Ez&ap!~ni$um%rT%;b{oU6|bBp)T!jVBVWP*rhC#)nJ z#OCytUnn(rtpYSb9d^oC!MCIcSZ#j{lGqa%$tSV5?@JkVsADS@9|Gk~>ChjP0bVWv zOieR{skN_Ree+eRY>6UOzSTffoRr{zQYifXcNPwRm12718S84YVpqSOqW-sw=$LFz zdU9qqtCuQc`-(CdeR~_b#>Qdm5%PE*KM6sf5$;TlMGFFck=z!r;)pb>!Qbl_Wkw$W3Ks0`pN9x-~b@@g=8W zmaP$FA4GJO(_y%@-FY#CNoAR^iY+7A z@f}JmdE*r3Tk6N!CyU@}{Bn9osgTHPt4rRz(!(28d5d#8zCy(TEtWafg8eIu7y6_O zsm7TjczBUBA1REW9zVC_r^PQLNZ5;*)hVNKZDVoN@RJg}=YiNVuUZn}ERy&eJ{0>- zZe>q2?lQGsMXctiI(_NUND9od@QZ3U&{!1=O_QR*Qi+EVnnS>Cj1Wt+$(76$ydp>K z_K0sX7iRHnF?+AIlvytihB3obrJtk{{bIWl5)cwIXFWZ@zCakZMOA4VbGD;>NNQ((!Vs}S((7>K_*zyW?A z{pg=X<_CSnkKrQT6rM<6$wjQ6BXCS|7D9J#0F2&p8fU)@$FXaTu=N5BxEPs2b5^?3 z%H0RyD9NLCCmV^2?GfVDeg{uK8HMiDT&E5~AL-~j8njd3x=IPpr-o2@Hoq^7$=>@W=j-! z(fF6Bc5s$td2cYb`7T2$57lG;{wBQE`8yXsqLgzjuHvqM68k&VnfX?Sv5DKFh>1rH z8Rhbq9MP=B559zxZmT6U=J|5k@*tLaZ`wjltj)!VX3@wcR951C-h(E7uAnU^RRM;@ z!A3pKy(s$lSns^$!Z!efKr%~eBa33S9Vw+9nXA0x81Sed#1F=A%| z9hl;Y)o?#I4jkrZ!F2SV%(fawZOxWZ>60s9SHU?LaY2<`>J-!C@df11=0P-mMlP}B zQb-KCsW5)};?uilk9=!-z`OU&RCYg|1 z$sE;wxt|`f7r03g{qaGm{@ zez4dGr)F=2A+vYFrsS(kb$l7SCY8kk{g!}R$0dOQtWC8X2k~2R+XgO{nm+)g5osR>x7TcZ7YU z6*(gGD{dOyNIka>1B2vFcHie7n_X~;X{=NuznyiEhh{yNT+d~{w;gBETmcKb{gv`# zSJ4WM5HfxWhHDwm=}(nm@ZWtq-ery{A2>^o?<-#pO9%VGqNzd{Cn2!q!X1)#cOLC()ug)jpM%@KYcSWo0^n~x`0t4Y z^NBNHKuZ_C`k$dA!iwmZx5trgn=*y!jk?UH{HbI6GRj%HhYSw?NfuoL>8?ArxfYPM@OHMY>9(VqWs=i66!Dl5d@ z4-)B%s#dyN^#xtJJC=klTTY}te8N79TKJi(8hL*1Io{m7m-v_!kl&N4$mU`1(bEUf zl8LK)#Dg~1*wlsZ5YNBxMzmqJJbxp*hef-*W>j9Dc{dm-stZ`CxygN_u?J&bKgz;UqUzv-!l|HstQ8$+DBv0{%kI9$vj$8 zkV$VDq|kW}&ohOt2BxjOy+N5-ZEh@9HrVsfDzvXf*lzXc##c zXTe_I*JO=teekGLkM9|)%)d1m#B2XK1}lD^g8dPBa4FQ0hA$jVol1|GWRP!Y!dIpCm*%rM+BJi2hVkq zr~T=)@ALn^jDk$?_jfWEMel**Vo^EMWd} z)@hIj<}dO~4L|b6r*x)338!g>wiZ{NqgjG7S;D5pw;#S_J!zVXTsilXgq;3p7T&+Xr4OgI# zyFY^~>xIIaK3Lte9t!j$pz>~ja84I7n{|i<^S}l?9LRB>B(m?R6#Jumf98VoR!8ul-|4x@VWPJy{)!rYhv6Xoq3oHH zm_;0LX7`%xNvYlh@?X#u;%ohYe>uI4Z(sa|zj$+jeSoVyX_Qx?tY{9F6U7P{XrWKB zIM;sW?#K2jUYjHdA7Jjkk+9>S zD`<2%&Qg}GC zjNP$2%>r8PFtNbEa!4!$!*3NJS9p=VkU7J=ZOfU})M#wmd>H-bI~vOkCa|S&J}68- z393cztRd2i>Ft=u4qVw!ZB>@io6?2!v4b7fU6z6u%^OAkYiAHz<^T_tZiQ9<=}Q&| z#|u3Ha^%UG270-%gNE4urXy}H2JN?NVd;zj=xun1i>AKAWRp5kUh{-%7Vo3OmR%CM zV^=d}>2<93trzolw1HYv1|j>WLf`qN?AtegmbQBxk^-kBDd2jh90uAalid#*@kjsXBw^q-3EZJWH=f>00<215 z=#X@v(?X$1OJH2eI>VRFW1yMa4Y&RcVj=gt!RUuEudu_EztUpF-%JmrdsE}-l+l?~ z#ds0J*4`}f$b2SFUqbvg>5-b}wfOW>B^DPg%TTB^tM18VOPvzf)iVoOf$e?>+qebR zJQIV@&BgaLaQJ*S%c@+j=~p;;RAt zF-f4h^IeICkri%9{0!ZZM(mv!vxc)hc!;qtTBlVkPShC$N`Iojulpe^^GRXWU-MYf zo?>=Ru8ft`=f?0>Ct~>K@)%xuKxC)tCMD@lM`Tm&4I~6drFq2Z&3in!9Akfct67Br%e`wfA4@M_8jS}yeZhkA#?t(*+F|5%zg{-?;d+N<%p zvG?HC%TCy}q7fdrpJVQOOPQ(v3Fcn$OFaI4kU0CJrO4n{KIo4>2iAeLP-aSCzbFcv zJo4dyyDOP@d%zn;sr(znUD5`bBQ4FQg8m0z?nTH`rY}n}ym| z?!;XUb@ql=&AFmV9olrboLlHnfKIPhz!JeLzT(PTE|r}ohR+}1osXWP#g37H#^%8| zCBw-7$+@bPr{>OD-|~ zP%9f<_J$=HtcCSoz2RZm6nOmbs3d9d5lLJ(kH`Jiq^Aci5Z48nBv9ypHklPeIg2uy znHf({2^^owandl*qQi7gTCfe{$1t0*O8k>_4gS<@4PIsTX56x6C-T=+Mu+q)nCqHi zknzxkV{b#nDj^T-$81$X(;wS1={gT)y}*yv51t^JA1p6v);l1H@?2$q?&weP%IGhg z|L(J*lbh#B_Kw(rK4oTdYO2=aH?y6%s(ZU^{l-SwKQssNubCU||9$oo?g=ZR+|gp^ zes`(l@}x_+<3xj~f2kAq^njQ+btQ^ihn%zl>6gsYlnXyvvYC`liVx5EjGFJU2LJfP!d*YE;*^BCz)&X$$oCj zSW(1-y`r$yZ^Uc#JjH`DyTr3@=h|yEGW$zo`o&V(mEs695D&x{qr|LS(ZZQ>_Py$( zk&?SLS7ChAZiq#&UE-Q}8+Y3gw&Pp&i374@MMrvuN+MqVLZMg3i?v1G+^A!w7d`GN-aj+&0JuSC-_6mY5CY zCM6h>m=tgOH&iJyq38y*uJU9Hi2+#U@ z$!@ECm-v0{X}sw961$hXKH48vx+UIoMRYaOg6r<< z729`T6?5&)wjV2+m|6QnM)uV+)q=GGk!}&p+9ZMgo~#5PeZ?_9p)mu zCgWv$V~FzoYed#4fb3{jBnkMiq%Sc9PtJBA_m;Qd9kn)OQC<{vJuRYsU(b_yYjU7| z_G#$!D1fcmpGbGIBdG{mhZkz>gm2msSY##xvV1-~k0CIx@icwr(nCu=9feI{Kf!MJ zF)~-zk$h`VCl;be)N30`_itPPsdZ!7^LLY3P>KheQeMJtoNHp#>KPj}5Ce00HyHLq z50>TL5qGR}k%V?}NMrbC_|MLWRd1NZ4s2^C<6Iia#TkfvbX~{)O$*@Z>0Nwc@pOr( zB^Ards?x-dC-5?VWzJqkEOFITlw@pN%USOWlw4Am;xA-r@*}1i@rUKBNqaFPF8}_+ zCa1G$#7Q6UeN_gRN;LrsT&Cw`DYSS~zU1d(CyAfME<7sX2E5deW<7T%vf;Q)C_1hM z?r|fy?A3)K*Z>q#G+_R~GG-gSoK^1J$W~b%f~ymHKyUjuNYO22Z=BAsMyr!-UDh)k zGxIX))jWW_mc~Kk-!vFTt3f<3m_86X=39>`LHNW*;_j-92I95w#y88c-7*_QT2_$= zaS&;HHj2a-9HrXnk=#c6D`*AjzzIQ>bm2h}ot&3KjIVoxaYnVknCmB_MvWxf_g+A@ zo+>2dF$ba+d4TX@n)Edcl$IxeP?iW8@6+Hx+5*^pLmAvkx3V^FJUh2slMUNzz}y2K zK-}q6uz!CU?rrRa&zg!X(<_g)pE<$eR-a;1^SkJjD z5iJhYqI-K)Xj1x1`t6Y#WL6tN(X)p%dWbfRo-_#*1nqE~_7b21>n47496Q`lz^WIN zvGOI;F%tX`?swFX$AVSQZrTp321%2r52MM-(ec6_<2H2>SYooNn_%LcN$g?a zcvf15SY_rZDrHwq{X2ST@V;7>c(<6{UslYGF`#QQ zF{$bJjjs!P>e-iK$gdAxB*?o8p9p32OhgXZwD=U+4w-bBK_Jy$S4S(&?ZbIciP6MBZt&`4tWq1#!g)B^Hz-dxQvbTM(GS9G3r~29^EvpZp+5#Ru?*Yp>8b>&3X! zOM{J?YrtY>=`fSLP9dkCilzN>v9V1lzqlfwU-c=I&%^no&G#^AeXc~eJyw%kyB5N| zeSdT=>QMDYBPGR!BZ)Q}%?!KC?fDQOru3_tE3gZVC}K`qD* z9y4ih={QZd+2x=Q(;Selq!DlW^_Oj{_{(}Oyk-W6Wg)^!31I0E=zbRmu5o$b^7jJx zKN`dPs+scP2Q_)8`suLEF&G@Cr^D#6QoPZZZ>+fbEj#_AhTHPs3pahnVieH-4w^>{ zW_Obf*@gd1*n!?=crY;y66QZ5KhC}mW<)2g^KEnCmHZ zrub+Wvn)~Om1jxwd|D?{(jK50P1;~Ko57o@1Ndw4EFuw?6P>MTWah_itZ`;LiP&(8 zP9g8m z7VKcT0(b5*Mo|Wf@z&cGWcXBTNlcxP{YIxHM$5a=s2@*odL$tRllPIE#bKmw$~dw( z`Zn2q|0v!6Hk+!~50KNj$Eok)O1iQBj(F~$c=0TwYRSB>CM5FH4f6246&ZWq4DU2+ z#J|Qe!n?Mj*Rj&X+2{d#Yhy+I&v3Y^^W)_vWJb zfuoza?r|V_aP2l3K8B|*?(wu)(Bluqd%#TGE|&C5+qZ-;u6JZRI%Aw7bP?~vH#>Go zMr&{5%)B3PBaiMTA-hhJ3#W}~%#zQN-$MU@T<;R`lRIBgcFaTE98o9us^U@5Ru23|Hb=WN!Vj=c;P>S~bS%VC(%3~X6!Gqng0DW{GgFoCpU{8AFB*)Gv;}DfN zvf!-^z4JPS{#&pf)lHTnnz66(Ba%%1ZdRprx*KTdh)7Aq$|NpxY7$N^_GadY1)$cHfdzzR;(8p8g}b*=6L}+FPV&NE?#PPs3A{Oi4uQD{S^>F4+>#@Lg z2*KTR?6GFmE0X=!j26wAPvsU=a$j!Pq3zdyp}*m>IDPef^kRl0@_I@+H{%9erDKNn z+YXa>T^J?ONG6h_tt#9PCd0YT67pDkhN1km9r#r21ESDJNSwn^D)RE7+ZCtaxskEx zsN^o!>k&nKkB8BJI|qYNzCSW=S<4+N`-F`3N1+IxTWE0ERh)Dvf|V_w$Feq8fn{qf z?3P{)Da-YtJ?}nS(sQ4EymgWpn($a;O|eBzGInXprVbwx>C7$e^sLr!a&5(PEXnu4 z6QlB}_~>|QJw-&f%r6C(J~3P?e@>l_#z7$zK&!|TzMXzR*SP7>lDqCWFCiHW_Wq+Q zA%pgZTS4*yRjU81ibm}WWx*Sw*rUsF%xQrw;e00Jw&^dp*-|H2_`(_%9D0{+RhHs4 ztGn3E!|hD{dMh(Zxyc@rde$hf0MAF1Q~$$Jbnm}v$rFo2`{J0-;w;AzY@aq`;P1{( z6y?%wFKX%1=vw;Z@K64fR5!oz;WOS=Il90_{IVn${?`Fherx z_(U>t$2)QV;8e-M!$bM@F^2s2K11GgRW5WHXM(kD4t$wd3$BG_P%pNQ?u7Ib_bw|4cd8MrWt(qsK>;6Tc&Rgp5*xlG*a$sjUvR zgHvJO)bo(Z)k1WHH6wbX*_n_b?3{*}UmZD|zd2izuNMc=vO(*qnSwLDdoGR5uN%ZZ zPF@MKMT{8N_T!f6diYZIb=cI|1-`bL>{xRzT`)}wlqPM1D9aR@U{Xor^FPzqSB}CO z{t;|lG5`;9^$_P*CuW`-xuZ8j>HD&|RIg_qeV3<5_i9uTqkv_!t=kb^Z0VylZ|>9f z$z>3boeIw<cmX9#%cE1nXu^2s9rHm8lBcG1X(@4a_AT~SH$?1(ZBHYZASd@PS>o%*BvCp%`Ba{`m-*7=;F z_j}~o^%^l=bJ3kw6UkHKkyP7%7QMAa30BHaftTvvsm`s2Iq`%0SP zt_r%n`f$WIP-wo*CVg(vBx_M7k=wnNPO9i3C%q2PO9o9a_E5dRy1fGjhYKyG83EMd z%uzbI^9xM8sK)q>LmB^j4@_9M1sYTeV2*@RO~|KKa)V&}%9%`alsAjb3Sspr@5ns* zmsrazqRWNP+)>D`c$B8%$y?u&wTWrOFS(Sc?usWb&Oauz4|Ec}p1Wuy7eiLfQK6GR zj)ITtqu`omCP@FKOh&ws_3S#rmOMF4o24aLIv1vLcODCWw^O*UV&TB1Di}YtjG0-rG5zQLEGj7#?cY+! zRUQq-mwyE=80nL+iff8*~BeFVbB(ODKV;p4>}AoVXAQY?qj zG%cReTR!y94;}hx(g=Fz@I?Az{u}6w9K@=^q?y*>U?@@u1Bcn`K*l=*(2v`2Xz~-7 z@9~cGJ{m;}6Q@ywu79AmP@a{}FlIyl9%9u^CG4<56U*|S&D>6`VjW3QOm1%oIOk@A z|Dg5IciK#{YFaAy@Pa3%O<&-6p(*dLJigY3LDWOghB{j5p#VyTj%<89vbIOeGQds#wvo9s9*>rw2}@(Yu;^X!p&X5P?#_zd9Vcw=N{X z>cwPO%SPf*y9ScyxIy)<;b1nd4ESZwLG5ZkEVz1}+M^+$R4WhSOg;9yQa z6xjPxUDh=G2UtG84C{}+f~$!Gu)%UG`xbnRHBBgH&P{S~amHxyIyMJHP5~IL%S3mN zA0e7^Mbs{9KXne6Mz5fYxYuPVc~;y?%%d8pjp7*ao4Wu4+~2^y-foEa+XII7=UC8* zR(3c1J=%e9&i*nJI;#<+2MWM@VlGtos`H*hHTWTI3j9!gLpZsuj*j+NK>O0OvHoK* z8M1aUK6AH}Id+sV4XYxSaADsJoB#_Q!Q;tv?R@xH!kWa!~q^7_;jd^WKHPE^#x znh6b{d&~?fIuhtu{x&%n6;AqVqsW_vDjc_0lh4mme%(*nsNUlqsj##Ihu|eR{YgRs9;;w>^@> zIxAS(@C|Huyc>)0+yM&$jX`wn5|!PML{2%1blyZiYAg9kZpt-6-;3Ko zcU6MIpA^z;?@OizDv)hsL+GJPDMWRGCkc3ek_wS}5_b6jdU>`6kGZo8`&o&xVcH?8 z`1%cjhI_=*cn~{euEJg$OS5)aQ+W1V8k&MNAWn5FlW01zl&Qd4`DdVXCIMD0bp>nr zVXW2zvoUN6`z`Z`{oH+*P4%i`k2YJdHT}Bm>mE7Q7+p;d4oap6l&4aSucz_xZ^?K| zWhmwrS;B%eOOPbX!Bz8@&`zI#^3`+T@B2`A)Lw=Tnhd)K9M&FuT-DyFnqoe-&0 zBqU=PO+M_x{KowS564un7$Zj_y3ZlqdmjX$=MlV|)CR9Z8X(2*B>Xoi4%&*ALQUfl zDrGDrxDH*Tyov&oEpC8KQ(~d5K8UP~mL>MH5mvfCOl*}Tg<8%X$L=Y8X#6*0n(lU) z`Zs)%%uWa=nX}c5om@|+bp(=wP75uq`g#B zlu@Lq@jU0;_jTP*G7=gJ(Xg@=k(tc4(ni{m5v4&XMKllz85u=Md994d$bQd%a6X^s zJm;MIx_-a!R}8MfUbHi+*GBZ)&fra#x18LAFs1BbU@^xp!E z=ZbI|NN~>jG1zg>7~_XWfS*$=j7u(q^L{08xuXO=8-~KpR|D{n#V|}a5~F3%UeRbD z6E0(hIW1QS#W5#2^xdw2(~qTs!jUc-WImmS%X@OWS_XmB+a8Fi-b_j7Yufto6YXi6 zgL&~Wc+qh-N>}RAkB^7Yt5!Y4s#6aBOFlxsR#wvS8m7eJ0R{azF1Y^NEYT?Qi)5g_ zCTzPE2c2mb!LHQ+`?k-*XEYjZN{{^!HqMnle)j4KU{>Rb1y!1%Dv7 ztU}25m67z2vnaY*7jO4Win9XX-2r_HNY1)#>f%U{$ z5H>KAo^MFwvmP|lO5GiBGTRns8n|(b90W(3=NJgtn?>q8M#IvZZy+gP4VSXQLzH^H zM`RQg!@*A%Zr99A@#vMOLC^9fm^@TQr1ivu!wf}zHy?|)1|H)b zh6?+)!{)SUVyWoyp4;M;#?f5RfkUL*Z5!M)nu8l8$-Lz{6&f(3g&w5?M5}Uj#Sho- z=Eh6~?(NF4ynNAuT*fW` z{Etp^)I|LOIpPefXJYwb3NvoP=An-$9&oQzAZo zK^q5+nSe#RAK-;Rw8Cscp48jY)6&OJ3<4(pi=FzHrOC~^dsiOS$7c)i6}45(0@ zpmsW1#I623RgcJl?c3hNara_~32{N^%gI=R6#mxXue9ShzF~)iS=~Wz!g(68y3*omW|w?K?k|P ziB;6i(~>4nJwPA3sao9(Q{$Gojp6U7AA!Sb_QU-@QBZp2BTT!Xhu(7*;Mf=dd;cJa zG|GkG40ANN=!feD#^SNr8RST13fH2hOSS6fVTk1}oHIHO{ZudW(a)xERqgs*mUI}6 znsI^dv$qC?dSAM6XBJ&CseuLrULn^*vv{w>^<<<&4)=HCbn*VbU$}+iQXtx_iAEc| zr(x=fw!1pli*J986QB93OS`vr)2%N}~o$Zv`KNf`bk;II)}fIWOWzt{REs zKMFqgp=Y4}=Lmf5a2(2Ws^Rw1`!q}^8s4^DhN-FfxS{t8mTVZsbfuyozen)9C(J@2 zf&wxFbEv{xb>X9JGlGe>edKxLDPA`A zD7i8EC9jyI$nOS$lksO9?I;^fmO6~4D`a!PD6bhdt#}KOA{{(nW{VH(obhLuwZy=uC4Fz(KSfb5KhE_k>=E7ZcA00z!n>E=hI3o{hYLX4xm5A}YVw#vU89Ltf z(4qDFh3E8XI^>)h+?}%!Ke^1pBV~4|5L}I&rpIvaiw)TJrHam+nM=Pk1X8K29ax+d ziswhJ#L@1t$Uhu}{^uiM)7HtjUIB4?aSkrJf04dB5l-EfZ>Otv#Iu6=iR|2p6xMa2 zot@T|lI&4Zm1KloWiDoQY{H)Nta^A8>UUnj)fuT+zr~iE>QbQHH%-U`63WWTBiZ7- zz05?tjqGkUq_P{=(aQ7rk}p-qBuoA~CYjni6~=Q<=v$ZTR86?Ahi(~4>$T_6lU|y( z-D`tvPbRDpFKg75%Cxja=z)JN#Pn|g+jl6R8T+1O)+_5V*Z&+we5pnGa(!gOr7`8hSd@xtpeL3j z(&%NMpey_tPJ90Yr#<|NeWNqjv{^|E`r;W~n9ZFqRwi%T;>kBj8U8k^!IeLAF}+%u zIZM=7rt5IF;b;^*P>F-crC~53x0ROP$c7o^tu*aJIePA?LusL(W_8?{&JUhKTbiZm z*z7*YCF|h(ge3Uh5{5h0E2GB4VlentMSY7MskfUUyc?W_)4H8dv%3}S3@g|!{VMkD z$VnC)<_;?HJ=Duv7lx0DU}+P*SYhQnCTl%{^(tyI_jfg@Zxx5SlOs@h`BJP{FDo&k znv!P@VcGj|#>rYtJmt?ZtMvW?_7SS>Gi`eM}lbKqn z1}nI_oNo2Vrw>l{kbxgn*owP@*@2HE*!JIP7&z$w4&slXg~k~ukUj{0y3w$3-5KVo zpUbMU!kN3zan9r3ApT_4FtImHq#ATBja%hOTQjxj>wR5Bw)+ilFL;2}kJOQgMZ!Fx zw;p~;m4W8=9JnaeMjjv8LPyUQf%m-`{PFc?`TNr>i9vEW>K$2w9VaK@qxGxkz<J!{ znXdf~_W5`W>)5h~-kak{%SxoF>8o#|*VDJz-q;s!dFP8Bn|)K8g%_$bBQ*nRyni*7 z`(s7LOOqszMx;nar>0BFy!*uF1!i1UKo8;X3?pL_4she;^Z61#YU}jlD5rGG!}{=C z4N0c(`wg8aWT2b-c=;7e`6*s@rml~imL3T{i?YsfvSu?R^L>5+1Iojl@NQ+99Xkz?9{MH(WdENU^bJq(NJF1b5%stJN ze#%0K@+P=@yB738m1^7+yfzwgRPm^x_()fixGFu4OE$U<7DG$G=t>664hY4s4ySMj zEyVo7cKmtc4o`$Kodn9`o&dD8wnw`N6LTGu7e}MW<(Us zs!oP6A6MY`8_sB00hp*Uf&8wqBCBn0lQRW&$T=G!@}}54b1LW?-lrYg^QqkS5T-d| zt+1Dv!|u#Y5UpRck9R8!w*im>|N0gMYe3pN!aV3Oe^kWTcXz0SU*@pC)5=%FvM+aN2E z9@od}L@UuwAHifXx5)t zDhgNzVwXW+e6gJxSiZpTAr0utAHywvk4aEjFq!FCLtbA@gVUkTkd`clUH@`;^-prd z!b_Q)rl$M^os(qBWx^5SGJ%5FG}y}qd!aqly_E-%L)Zb>+<^b}T}Ey3jU zO1ySnTe5Jts$}qPc}Ym81dg&jzzuOlUh6Z=DgO_)rI+y{PbIp_u!uM=5;9Xu zQ}Aj2A#C*>#l+vuSnO<$B?(#8$=d-VCH`12KZq^+wUnv4FJ)o(GI<}3RKDoUQ%*lG zhn%Z+=0)%C^H$q$^Yw8v$*~=eN!0F7q|-@EY|TTSS|s$$w856TsW^LI z688N|q|QPvC28p`vQsuxnC-2_g4jtIb#5+RerAWdj?3`N`px*wK?^07N#LV3hq}u9 zP{)ZC^g>r1T1HGl^V`vo6IhG8z0>gbz&)6DDV$yKPi3uZjxp22R#16F3Y^V8)Bhf- z*$lY)z-C5Im&nI!FzXRnuzwB*FBs{YNqJvm~b1ZgGub>^~TJ>My!4e}*}a&SSu|9JFr>!?#{v$+4l4q^J$Zyy?4nm&ZfNYfBx@ z>xnW(0)BY*9`-m-Vb{kvvlZtYSh)R1 zxM$uETKO*^>vk-<#w4Ns*Ryy9mD!xH@32k36ZdURMVZnQn6v8~zSb+EV^bCB-?=+z z-S3y8EpKM=;^?=WQ@Af2{bdNY)-td@wiaLgK7t=SQqUC?SgZFBp=WXt7yQ}9PJLO) z4(3f~{pV)#3vL0)n&d@fl#iq3=Yx22%M~oI8BZ@Pok9O*jp3Wh7Q&2$J7IA|3QX&G z4b>W6sCjuSCJxt!hf^Xzr~C>;^(g>_beJBih!u+rNQq}7H?_Az{Pb!uz14PxUi7<5 z{}#Gay1Scx%POOJ&sD)W&IfjtEQH|2!!TpRWHe2(!c7K8NtL-0G%j5N_0PwGo|6ik zULp-o_GCle-5SW-`U&_Mw_vSo55&Lt2RBY+&;V;&7z9}$yLk<*aLAyte_E;OF$vl@ znxd+?A#T*l!@crvQB_HSH9b5I=66P+QCc8cygx~6XHSB<#1fc!^1W?wL4@sqoLJE( z&+DX(Q=ls+3)lT=0e#w$NT&seU|B%{+hp^OMcver>^k;{clxKpg-p829hZ$DVNXj) z&7m6d_Wph`$w?O<{c@T6u>L5H!ACgNT8Swqd?JA@l#AlA8;q%+5xbz$TT$4#gg}aew<3`hw4i)rj zjxl`wDu(X9p|GXE3_4`zL&?}-pi*Hf>;YVXOFvI&Oc5z<@h80U0C?fqLN(u-!nNr+ zV$R7-lwS2tbm~O_%rNr@eZ5VfSS0jeS4v~XM^!A+J4D)lDpLjZNNTz;5h`yzgrM3# z08t0QtI_P3Stk0M{RAB8!ymV0F_m(9xbwV}?!U_kQJxXq6H@QFZ{F4nBuizTs_FlJvL>KD%gixw-$dpeVb zw%4G0{b#g){R^iwC{PDeW$Ly+luip*pk!ksRo~%huf?zGEkJJS;kj>zu|6=9a{J3@sxT{R#^X3DRST_FaX58{8#@(}&lbkE zu+Rl*%vUpm9f?h6p&|twn&gKD^GdKOM^-GKEfFuiIGFR>xeNzfQo|+MouGMWfN)=S z(wtk1AN8>?zEKZNZ3NEx?e78( zzSFjN;sZ2Y8;J7O$M9T|GOp5n4`+62fK5qvuP#%58&%<#x<7mmUP^#^Ao9-@~glfNrqx+o~VB|np>>+0svP{GZT?iT; zTZ6}X*Wq1_g)G_EhwUi#W9tVV#mCuq@amX1c>G=>)mVRM3qI4wP zd3#V<(-}$ILX0S^qCp-9NXS@MVps8i&gs$xh23L>JogT;-hCd1e7O#1-=za3bwoi< zkEHtBVbe2zY~CA%E9RVLlV9IurMJJbyc>nEbHgMUx~QCv4A5mKy(MhZkSL~ew3Ihp z?@snxy&zX_{RiWA_JZPUPsF|x+?s?aE_<%vQreG)E%|-OsFdqjT{Gys(dNx4Wp~-DA}QWli!r~ zhF^bLmTVp8O#C-mb1QPR>9!@+wxjcMtn-sUa7)Io;|-fP6Z^tPg8R)H?=9sf~NdS*7nn55ffEL+{dsIr_XrQpc(Pp2t6blSu~zh0ul~pdDUm9IMe4 zc;HFAZ%hFBlI%^FAC-mr&${Rtz%@Y;GP8TiRMFDOm4=wG5(CSO`qROZ0l zDD|O@%OqSy&R!C}hC-i2kV47bwp~7=PW)?p8|~Qpg=)`lpbDvvIlUE8B;(##T68#! z`&)96+ZA8M4NzzjI)InqM|uxT7#qaZ_WE-f+Z5?nCwo5m+9tj)+lKe|iRDW_9_P-a zh4EeQ()d&B5=iIGf%M6WQ#8cn6qwgV0v?F5wOn-3#?ms@rm}Xfc%-|n_>Jod5%=DQ zE4W7Z#}^qtFH@R~^7+iWzp4kQxJ1%3 zVX6e1gCNJFpUR2`13B>vYv@N*IPes;5B~+Faf49`_P`_W>-3cQ7TWk&lfKsb%;rq` z#4I)pkl1aj!N>jg@Vn$4zH=+Z!RdE#IrLzw@@jTuLKu^K7tJ2=ZP;}>Qxh8F8sr|%i^-h3RY17(=>J{B*`x5f?cwzIoyBbo2kJ*@x9W{8X(4*RH(NAC}1 zo!{oOZ!e~^1AC;HQvDMg`l<#C5(l#&)uHUk>=A78(g-%qGK`f4#QD@(LYCU}tq<-lIuPxd9uy7|nx8<=| zN$V*;@t+>myAw$K7Hh%VPttI6jRJ_Q_rTy&x8c&!E=V$S0e$}qAouYvEUBLjPu=&> zBV`rbtY0y7M{5yP`&2^5sNW$wKfI+1c7ySL!fXCemJhwzJQ;JBFxr}Rld6vOfaAfB zh~?j5H1O;b`rXtNXR!fjdwvvtZ%V<9rRr=)IAgP2N8^O7>)Co(JNb-bBglZ+HFQYI zJ95Oho@`urfT*{rQx6S2P^ioV^|@okpTAbxhN%qT|2EGAl_l>;_mE@c$d@v@zI#4( z3@@f;^NPuYH)&!$iyTs^I1y%jxdN-NJEDbq5tR$8r>CPe=}{#s%x;^28Y`mk$e0Oy z>mz0QK>9tg>%U89-E`p3?U+oy$cEC80>;gsS3#d_v4z%mi{aalg`nu5L`B#`^v?Lx zX^QnAHSIiH8@C6*qk!97H-(?$uPSaB`j#JYJdF>oiW2?3gziesvP4L>-1Z>3>1fHX8jthG0Y$!F75;)YdzLPBwc_jf6ae?d;2_a%UJTYU6=V zEd{NfAqW$m(@l?xsO#-W(i-f9HmSv(p+;{!(jf!JYtzVeg&;e59X{J4!~PvC;G25$$q%*lw0(ph4JmCScXxQvf#VQN zSAM3I^DH_}hqhSY`0aq<&!z&yCcemHjXZMbhfz8+W%9;sOc9|8; z{N+ryE1xIsleFN0_dCcGJ5f%x5RTDM3^(i`aR;Z-JgYOpbaOfog6QwyxY#7!6cmi?j?aT9n;Ce0hI zlqZwgPs2Z{pOEOzp_0KjvV5)Ib~CV^69 zo;Yu6K38}_im%x4hdiI~jZRNYhb7nUQ}fm!(Ahj13w*Na*X`bP(?na1d?=G8@O>8j)FT7Io9@zi3t}Mhy#=oH^q|W%#?gZ0 zK{RyP5twy=$Kdu_%#U!O))j|nx90+S*^<&jjX-uLULf_?UeKDY8 zr~9fps9MGwJg0dUA0{MXl9L{dv;RuoSwAK>J;rgy9}KwhHFo^(;zjhtvq3c8{Us?E zW&|_VxzKZ$lF8|6Ef{^Wg)TS!K~s0$A+1j@3NEeD)Z0Ls`01*Uz@hWWo(?at-XO3M zXO_TMUoE&%TunVMkA^ItD7tK%8Z|EYOP-pVi?7{I@@xM8-xK(c>eWiSvnzyllFNck3>|5xp6>J~|fd zf0*MPdp-17FO5mQ8j}CUeP`G1ykWlKYV@jaHt~HGM{)xv^CK!Ja`vr{IX|Vx+%u^f zaaB*Rc-4;uY`WhZ7GXS%4cZY;N31ocPcCny_UtLO(Ogdt&kUi{wbSX*krMi&$AW$y zTu$~ak0!RSUhoHNOxUYvZKfiUWf96h;mM!B5b{bJwTGO8WgT~5TfH=bl?>jIF~m26 zL|EFKK=uEG(ixSN#CNR$nx6a!qf7NrKRH?8=fCDxo=MV$EDrr9txBU~L)UgPpVt#ScKQ@Zz{!J6T9jJif>~c899fZua zjc}}>3AWvBfQ;$>V7faNyw#JTc=i-*P|?6oKl>qg{}w9AenFSph(I`&!{^LeYTgk_ zuVgNzI}5F8cjR!o&dmdd-POnKYA;|-+Xhyb5X9=#=dgY+Uo@8S#-08%aUTw0&#vjQ zxiUjp6J}9M3wkNX7C?DY$h*0)G8F7$hfVk)3~4$!%dr?7l1%`fo3XC$o(3;mc?MArDf& z)g3QPdH~^n4}yP%h~Cz$5qr0paITK(?E8b4_@pAmhf*VDLuzNyg9+f z73HviCx!GxkpyPGt^}RMvniQSNLKEdM|P_J!-5%g$n_o;&KTP8GQ|bZc`v+lJ}!w_ zv|keE6D@I>q6$I2>uAr84jNb)EHc#8#w+ z(KjfAr8&c(dqyWMExL@h0*`oOW)OBi{7z)A8}VCQ%=mvz0YE+0((jLrg+SgEX!|&w zIPB4~glz^GGw)I{?Q?Gw7SMZc9Hr)DPj&e%td=U%0sf_;SVq20J(8p9>(6Y%`;T)Z>+ zB9>mikLs%;nOx9Tw(svwmb=Lcr|cKhqm`4;f43qlUek!bo+RVCrFz6_>M*i>Q#Cg( z-xxQy8e`q5W%$snkv?d7McHN*$TM6{FK2C}??h{8fZ+$4drTFEzUrgfEVRUX{;m^i zw5=5P|G8r8x5!UiHnm1vApe2i8yD z)h4vlNA$}{%QkLn92vP!mHMg6(>M1u__G&6c=>-T_?5=VwEo^rYP|0Uo!OAYeGO42 z2lM}MPWP(BieBr)re$OJk);N-U$u#Z^IFu%b-wMg+N)O1y-!6y?Ul&3+r|96tAaN? za}?w!uNU|M`4IRciFaJ&EcnE(@}~+Fc$1MkJUGkeBv||~WZCgaDCXeMBGVh6v5~lDY6LrKr3*U+|R%YAUe{ts>$E5IuSB~>0 z{ey|y+f1VIrk|g$szpovJm@C(D7ySyfM}P+ZQD5me~X^qROg02h_tnimE&Dhqe<3( zl-Mq+A-78}k!Gt5a@2h!9es2O^_-kWqa#A;=arl3L9UqwYY%|(T;ZG1_Z_z_LYD8+ zegmFKLy@IU#4+;+(_x>rs7rteHJ*~m{d}~EJ2<|Xb9Wg4#XA!rf0s0_-kQp(+_}tq z-|yo(y+`u1>gJM3nW5A-$5(KgyYg|zy7|wy)({`l$z(*5KY2Cn5VuI+2VR-{gwy!F zT@+?6WAjR}&i48tU2(_KI8HLifUC`_Bln9BlizCw(C0maARt)-#w{ES8H)|MkqN@= zZ2cNCu{55OzuM0wot?^Gd8^HgOq^KNOkcJ`SoFn455TU2I(WnD9gbe&!Pc0iF{x`4 zh~JZQMEbZrc|r%me?KI!M%5lR<~qX7)h^Kfd=hLQu8O74m2h#FDsHUELZuOf*pqb) zuMV3XOv38)I#N_XIiBje= z$*-H9+`H>e{0H69)clSzjs8BEj!3#lw)KAI)V(tJig%@)%-AZnL|aA@t)MO0xIz^? zA`a7cQpyzflK)(qC9%1Z( zl)(7v&m)g4htb0`IP%i7p0u{7QooN!VS7$4Rg89rL;F9$$p~2tiG2h%`{gjibpkC5 zLwf(y9IAHS7`H?!;bZ5FnaZ3-2W~L({}B{jNlrK(EbcEZd%w(t+40UVrI=rnEufr z?D9^+mJ$>8?YIV;KT8WDlgeSs7r<|AtC+UuVdgQifz5NPf-TiDsNVh_Y?r%&U(hUg zc6$~yTuFvVdH^0xTnPKdOL)5!b1vZKQ|juNK+J!PpbMHGQbCa^83rR)|SWaQ)m8oAPlZU;9SbY?T1Y+M5xm%Z@0 zYchg$65ijt6<;iUMMt+J(A#Z6R7J%G*SFWgx#iB#_ngsRrSr($QJLUQWwy4uLNTDxqvV6REF!{45;>;ZZgDe8_G`&!7g z{_xvKvHyrLZq#%joR7YE=yG)uBefk4#bG!3)etb;*+*ef-nF14MrrrPX_V zZ5~c2z1HFb?Tv}kPZ6$( zVrYKX3U{p+MN3kbg5c-j?j(UZzPo7|L-V^*YT!tQ4&_l8W29&}e z4cr`22wO@_!6VEGOST(f=jY+rkX(!nRqOGgg9Bc=tN_l%LgwSBG|2Zm^JX7P`JUI+ ze5KT2QW~y7^sOc2%FRhA3VjHx3NFFAm8yad%Z${N$MYr~uVCmk25;-%Q{`Xs)N;Tw z`f=w@dfiDE6j_wo9zS+o2FOP@yA&VevUIM6lyr4l+R$v#N z2Fm-Jz`ye{3{QItVY6>w>Wfm$j(>n%4r4(+`89voC|dk?a5y{)QGy?4xKeZmqp!D$zs8np}8ZPJ!Pk|?lOo0|;SomJZG!pB1b8#(-B zfOqxUO^&flzRaYXm%Cm|@2`If^X8p^f}}1uqM%Oez8|2aW6J2Ss0cXpub3XSEfplN zjw~Q|BO6^1%EFeZl7`vm?LqM7az zOvp~hq2|ZPrrf)TUKu6@r~bP{ z<6ouI>*iGwpT`#@15B$W)2epS8_!bYSOu zY}v(?X&||n0F`_9LqO{*o6{qqX>_Fu)u8%+PHUVF4NqU%1+EV$fgJA z;K&Feld{4IL*Iy*{EDH>beS599hQOpnwc1ylY%D|hp|{|;ah0D9kc0p#X9Rd+04O@ zSk1!kxbsIhCJt)DPxBX23-wkyU!jg}99v84PoJYJ8VYE}6&?arztP}FQS{YMBm8IO zg7%vh;P%BrpLDapcIcbPmL2CLEplQ>LcK`xV30Kw^!CxqZr3UIc`-|RJC#MQ5i@-i zC&35un5>_YL{@*kh!%}EP&KO)HOt)Dg*PcIQ|&Oj^=Pl~UaCTg?PF~5(`S9n&MYh| zlG)AIqM@mTM42UWW0zWz`$f?-Qfnr7SdW0#UsBZa&?X{fJ{{DjiG=5P6^Q?tU}@8K zJe-)1<@%NwoxTbS4sFB~lsh3*7c z{AM+bojMKWe+`3~5vS;z-nPDuG4SG2rg?n;z}5mV90A zE*XStBr$#W#E*vQa*rN=pv&*-3uiWYUPUWWTxPq6>sqPK$W|~m+-0g5KN5Njk>qy;ZH{=k?f@N;#Zq)iJvc=g6sp2 zc15@F!_37fe{wUf%uc|NC#L~T&Vv3aRg_a2%{G(^Jnp=Bmg!gzyOPi1rLZ5UU$l+g z{2sATgNO7(SlRLwx=x&e zLBc!L$k33UvmHt8_m0A}zfth%*a~O}%jAAfeI%Z-`WE+oP%gAf96;q%9q4>GftyJ( zRwpdQLya9Yx8(sDVn3YP8BWFIy0JK*NfXDvU55@jn=vsa8a)jZ@n(%RlBshs+hi}R zJ+_RI?T&2O3k!B@FV9>S%dyO2Av!JW|J{Or;Hs<{?7o*Dd$DORdvV4N<=jVO{MH57 z|85$@46ml6e%R0}GUK2>!5``u?uG6jDbzyP8)nA+BtcR8;n;qAkiJ|8uWyXPQX)o$ znp9jr&7H+a9b{kIZ?X^S52(5E0D5j)B`=#4$8@IWvEJtyEbzo4+MBYO2I||=OYX1e zOvNwIZg&u)lO8kv<{c)!b(-D2ai12~+0*WKGpS`!y1;ibZ}a&ru1$&Z8NzRi#~U?gsvF&--#$Krdf1iV_i2oplRaidZ`E{?Cl zIFm>CVPF#EzL$aC7x6U4c`BxaT?F%o!vwC|clO7k#mL5o8*;&g# zCGIqRQgD@adOcyX-!-@dCr-r=7wj=HO36+vRo-@>l z_2B1#99TWnf)oraAos6y@@Wmd;v>DY_>ps#@edZxmc*`flN`M0A~9|Um}L_|zq}j+ z|4C0knb>IT*_?y57MeITWhCnM55$kIU+7huOYhGrq^h4rN}f+pmSj0AOJoKf0NI-T z@Xq8goZZul(rd0^*T^iK)HYh;HD#EjW8NUi>G&>zX=TX1PcUKfscR*L16N3rT$fAS zM*rlufGl~sp_8-U5y8Ujw=v^c%h^kQH(Rzhj5#>RFiG7OST|`Iyx4gNI#cSP^*2D> z`eYEzZ^JXU?_tcS7X10_3-2lT@s>7iCNn14V{p{~tl4US_kX$LpQc&RX4(xGg?yvm z-EY*QFPGk!^#EsVzJ%p#t|GSlW6uI*Bw>w85^ewEII%Yf{S!9g{S9~7i5=Hi+T+VY z@5-93|E4X$0Gy=Rq07s5_|_`ranfUyk2~^U!8eJOS7rvZ?V&xV>T7vF`?u!-n5-9 z>1miQi4S&{oGvPc8jS*=`_n<(dYI{0rLyATk!*ODKGi!{Pc3fM&~+LUScTzKcC*5s zy&ZCmguiX%MQ0fC-anZJ2IP_Rqm9UohAud^a}`{FRYrA{|B-;`17uQ)Jl&-m&h2nL zEUr~Z7C+bMCx=9lWTO8vy64ezVNMW4&&xTI{jNFWvp9x-5cr&|Z#+p44GpGI^J2-S z&ux5mQ8};nu$*sgET#rt{zUE7K5|6_*3sg0D38i~CGxDNraAL@qsL$xZtgLk9 zwdJZaQSv=Z)UR?_&JBd%NGVWvQp0XbU!z@nA0me(-q=!pF&>wgha8MUAn@Qs2w9tm zE9#n1C!2wUrQ;BhTM2n{R6sTE6;66{mdU9+!L!>pN#wCsg?UT6u*Mn!8C`e^W#x$SnI%lFvH1o?;@8k$;I?yd};2D-4I>?6J^ zmW5Y8;xZrB&E+`+u48^ZUk-X+24L=>3NBUS8L#0&-t^{dCVSQye931c?olzszXgx* ze`jT|LDd92Dk%*n;&<`7i{aSAE&fj zAbOt_gNVshq_Wl>^%qzpx2q0lp=S_M7GI0Jk2`ZNUWw)WsvG6(*%xZr5Pgv|d_S3M z<9FF|agHTdCTT5Cu$aq8Ki$AxKiQv~>~`1EJKm7{ZgMDB{GEuCurq>jZ{?!{p0{}1 z5=oJxDFE%U5^$^UtzbeX-De^NQk>~umT`WSm6^F}id%3e3e4;ilz5FpZI)vrZCtZx z9#0?;=Y1MjWElm)9F@Imxcfgv@)TNHx$E^RI74+=yx&f0JY(}WT+f?5oa9g5+^l4A zCZS0it!n;?grYvV#a3OeqVi?V(6U!NJjn>%{vnGiMY+HVw-NIQiD2HBi%vY9Bes^7 zYa_XW)lLkqd%-j|?L#l$H1pCPykou_??!%V@!YFRV>ru}Trow9r}B=Ab^ug`%Tb3^ z8s|(;1c%dT#oK)^luvtDS*EjX9PXzq;VSa*KlxXr%AyS&bbo24Yi{aoh4olxRuVQZ{m zuC(?m*KBz|m#PHv;tOS&-p+E)#$94OJ9EklQS#&RUmxU6JQB%sUO1IkqtqnYQ_Ns2 zxdxoZu$kQK$WNv_Y9i+DeEofHux~BrNzWzTpky;IY0D|*bU-t2%fbsx$(d|sz&e_j zwdo;GU-uZ-({Z82%`4X|68Il1rGKxn2=ZNJS=u*XSz$SaJ1W)6o&QgU*Yq@m*|Oab zS!s;te(k=-U1QqGOBlVyl~}Wa(JRtJ(RwSnt@brsoz<^6l|8OJ=jvBH=ZYsx@8cxP zV_n<0M_m1R`BAf(aGx}8WdX}Q*ENo@P}W9S#XTr@j~!E$_ljF+CBuvDt>8J#n9O_s z?ldoEtid9G(?W|oQvTej?rG-nJE|=l4(#PbUPJ%=jYbXjw9@x$1)32JRr;&%4l6K0={ejhMCLtC$eI zT;48!FJ2f~zw&Rtc%m-}h^d*=4tF=o@Gap>sxak%7jJIY*OfD^jaaqJd3G(IJi zK}M#$nDz^ta4%VNUQRdO(-)6-T+PKFw5PzXlLPq4`4iZ}#1daMwn6VNpXPlJ^T*9j z`N;RQE;G5y4mC`A#;7biX{mn42^StR#~SmU@SjOh;9F4wbJRj0Eg>I@i$cJx-U~8B zyD08fk|GB1eaoC>n&^Vf0!Hn{6HDLow|JdRiaePwemuv;LfipQamNG+@K9Tbzg@YC zoQ%woL}4%^rY@G13$~N?< z^a2`lT7lM-o1oXL<#}JO*W;*?IrzoxU?#F*IeJD?Iere4xQQPQF~&AByk=o4chKZ9 zmdY}~sZ$PeXBhv%Eqmg400F=Q~-&B)EV&$Owe@M@E9al?*fbGWO0 z!B=eyRJ84d{yrDvx#Jd-t9pSsr{0dbB0e+opIdUzyLjT+=8JK-<|&l;W+pDYlZl+) zJYy7&KEZR|i1#M(-9W`9XcB7}@cfAP4M^F*BGmOo67sx(Li(}O!E-~93 zwUFWWMa88pT3})K%;y3U}W>dWT&o z+s1AQUWD(~Tj23W4KNp7VP}M1VHHeD*pj%Z{N-0I`8m0KeuRAtTetcQ`?%{I8@5&h z?k9W0=B<~(IH8a+>-IstyW$ugHHH;QiA<(RfRonoiF--N(5Fb6=v-}#EW?>>S2iTT6WqJe7OU%{G_8Q z{F8<}e5uhEZII1JE{iTe%G4iF@n<5NTepMnJ>bu8yAjGC$6SC_T5o`T)&#;YI{YQK z4EfO<3%-u=3SRi3Waf8w40pWjJ;-zTEAnPav9_zGu(Ma`v-6MZvH!3oG~_IXm75nq zRb&A={wM@{+pFP+j`1)YdL9gSC&1M?dwD$7AtY- zCk9l8OOaPdDykQjB3*nQ_o;G38L4Q@-5ZWwY}euUa&mx|O@PZUHiOAP9E1l(v%5u| z+exnkY>kEk|8&kK{=WW${4I~N@y!l({65PXBZG77exoe*rbj&cTEPiPJ(os&6?s%N zCdPiL>Vu93Mb@`MA0O1S#g@Uzc-N{CMrw;cDjQWm10SwpZ|sX3y2_Ah?;!WRAYa6^ z`i~j#ZR7^LP+*=t2xQb2Zbgiv4vL-kkU4+I4$tTk@jwoo#B9a|w0lwt%J?`N5A86< zQ5jO$_jEB@))>P3a50H1>3Wvi1xd^%Pj{4@lY^6NWJLG4pHZ0JG_X(CLT{yxFsmFR zcz%wFyuByXkl*)grgKFkMBht*#UqE{(ww(26W77dXQ^=VhbIe)2g#>Z!rSEX|Vmu!r}Z96w%&Zi;sAaes#m}+f0XQn-*&CkSK$;W6> zh$5B`Jd3}c{(!x+Owjw&UdSRv1^q$mATMh8~f0?#2hqE!c*j>?!aSe&+&~EbG)rN3E{J1sMtym z&68Hg?q|aAzMLIQ=Y1Je_D1AxGl)e^_eH(X*h|PbR{^iOd5!Vyc)`4UumfG5H5Ey4 z`jFc*Kb$2Ijcb0K$7_A!G5@tOHeJKRZrA2PLYp0YNw^4{7uL-7%Gt~g2^Z#F@-)1r zd;=O3G50S`31U-M#IaJUi7d_1hlH&hI9F~5O`oFKjm0r+;lC*MNJI!zH#LEm?Onu_ znTPUpS12;J(fYir0ZM3CAnL7+-D1xC6*(7VMQ%}@6l|4p1<&smhn)Dy;CW{`1YPlh z;8;)iy;={IR|cS7gYD?pe+A&Q{xGO3+rbIdc(B|n^1+E*sCPvU=LzEDA-g&WTRxeB zU%Xq5RrWgK@)Zqm)NtB!6jrA{iM~??GxH^)%`QIGNBmQDs36YfVwRwl~3cg9LED>j%@1A7yoquVnMqAa=h)21=Dic;tZwEUmr9p6_mD?TjC? z#>`IMlb9Ob!f!uCb&)Zvx8A2td}ovDzZwjad5Iz#-#I7Vl^EHl(9$H-pW$204Iky#M$=m2Zq*ucoa2w1S@ z2pUhK&;;ucwCL|KX#Y?IDK*V7Z*FG$}EQtLqdX0Kl zv)QiWt3Y?+At)aZ@m)p~AzarBE}OlErI~T;MZGxIsx6&0e=;3T9oz?2OK!kbtr^IA zAP;T&tbrYZE10fgQzWxE3%4H>&5VvG;635HaP5O?@Ha|=7vWyeR&EW)Tos}B{y1ox z8^$&)*v#fX@?z}@SUgfT9&@WsBAIUt+c z+_DFgeW7^#iV3*TYY*P~E(-6`eTy`N^YGjIVfaV;C0L{T0Lond!pcX-L^FjT=xhxG zmjW-mG$IO*D4fTM((>%cj4|+>_yoG1WWvjm^KjzQQs^zktet`ZJ6>lpTjcl?tE?iA6`q?t-fk_)7@I^_C+2Hi?8v6R$b$BGRyc~(#J6s?XCqD zY2(^oB95J-I^yX3WP<$_xra-tImSy_p3PQkezo3GzN)M>|3>I6Y#{PUpG+G?lfA}3 z*YOV&=ng@(%PM$ZZV6^u9&pF@CvUcGI};kX0!g1z{z2~lI3fD%VJ$QE zn;mqV%wAMiX1i+qc!Oz~ zJVh#CWPaU+dd0_3?kt2!I%h#^ujqGZm>Xl%LjTgL6wP(*bd$23_?qxf5Zi1y# zJsgg!0_DAH*xg5M*oV(pcHO^I5T57^^)4Rp@6A2@T@9Zn71dc!;b+>8N*#B7JD2l) zwlk-+Q;kvI?$6xq*v>@y9K-r6)3CnIIjokc!mN&+#_(!aF%xDdF)|`&w(oW|o|^ay z%L~#moc%TNywLk+nDT>7%-EI?JWtXay%YI)o{l}n@k7OkJ@|*2=01@bJUi9Wm3%fo z={t)nG2tp4pP2!3+YUp6r~@u-l8hFfm?Lt*M4=CTRm|AM5nk4%R3z?y9=*3ej#c7~ zEoP^m;e9d`@q_lXp$@fNv^8Xexg=K!N1g6~xbP;Jj?4hT=PazZNCraWXD|u3OnK`2 zOSnceX7UEleBk=t8?|&fe21(5-#zZ~g`0U{3ZYp0ssldh^%?J|31gD$dKu#fA~ol~ zcOp+}1vl-_PEOZl9(p8r%jD(Dqwc&_=HbF89R0=5dD@d}@%|7)+#FHE7}ba2e~~h{ zKT!oglTyY3_FIwB9WiFn-eMeU;Dn#J^Ks$Q5{7SN$g}aP;YvqIV%ZsU@#d6bEbKAj z#fJxQquOM6Q&We~C+Yj>PCIq*65jLdFHgjGE*K*I=owzv14;BgAfM5< z*u~U;C}0*CJ?D804>O6=8t_L^mT!wYgUA2=g5s*~q00*oW9QYPYZXpK#vj#?tW^YZ z>h5DQ$9D1#tdnN9$k@P|c9DbWM;Y#4QHiHtx`bPIPT}7;BFDe|<`-M`YAKWdP7-B2 zO=iBt``~&1)bTU#DilAMgLX|&#v40_k%s+gys!Qzn*U=X3a|18yG#*d%=QmXKNgMZ zh^UrO^yN8Sn1Z4wZ08xDtYEgaCqboMF?eW~K}@40*7-CIKXBr1{n>A)LS+}B0mZ?Y*c?614wVx&W z?bVM4LVhx>tNNMdS^3P;vo5UaE)SM|Tg(c&Q*cpfHh%rR2)`VD&0jaAm4DOXCSNIS zKQ1m9K~-LV(c#c>OoPHV%W)okO!}H!%hII%7TJRi+~kR#{A@o5evsijzS=X9*T7yD z-%5Fp&dFldaD^Pp=-daM^iBvhyoO^v1o-LGPfUHsLex0r0aG?@4*ypRmv3BZ#!r!s z!9C7b@cp7r+;i(9h$T@ta`pkNx~Rp^{jA2HT`J8#_n-q_AGUzY_ZqQf@>@1#<|Ed% z^$xqY;Wsn?hAl_c`aEYwmKZC$REB-nFUIbTw-tHZ*0XYk4lIAF81m}4#c=$z(V6p^ zxK%72r!ClonTKB?M0NmHW_^YNQNF-b7=$8r{9+PhkK-p7Ph$Vt1Y8?v!ppq;$?I>IWzs4g(ZpZR5ZCP?x|E;Bj(kaE+sb2E*={BHd21Y`FtTu>G!ebo zB=TsBW+bkrW?0g*6qWVK;En5SMV`?GtlAG9R_BdA-z{F9zd%WuUlSeM2ZDGCKlOq<4#k^dUnUpK~TZ{vlgAx$Yp2w6an4m4vHyJ(u zU93{42fOjI1M7IK62A!x#?#iN;{|tjB6CR#bh}^$I^Sah4(C2$<7xM?g^xEQigYu! z`AZpP(Jb^+X$SAWes|v1H-*gUD_SURTo)6sVt}T^h9kL&mFP#BHi}d6Lb-0zSoon2 znJcxSYwQqWmTK}G`%QVgMkhv}ZbO^1mC&rdDExEV1$6CBIa)k15AQrT8;7qJ?bMjf zM{zggMc$2k?nq(;Z?yawEcfVu?w}Un%Vm#ky&$GjnPw-Kr<^botc4w1Q_zGG6L_)%!v(jE zaQTf4lca9VJk!l(YNsEoA!+m=)Fw$7XYpIIh0>6}dSZkQ~=kR={w7wNxUti0t`^chuj`?`a zbHYUY=Ad5R3(S|(lW@urHT?PGe!OZz864#9g%SgO;Cy|}TwP>>CY;)Xe(Pt!$#=C- zq@}>F9Lhy6UVEW1`E~fdbRC|*P6?kqsDp%8wb+~1TI>gJBlh1bMd&o~!7|G@SW3Mb z?Gi$*k!Femd#2(u7qw73M+=dO{b<{uDJtXM zK#!(6!flC(uyjoyR=DDh%KdT>_t-t;zHk_S%Ds&%JT#zl_zL6I8O!7il`*@-Menl% zg}6D(5j)z+$i5$Zs@iwb{82+^-XI#KQ-Q7qg* zRa8&G)3?UpetQGB4mi*+&-7`tL^^)(R7jS1k0&cPIisPrZ=|_M9TLLtlKfB=>Sz=M z$98wX=IxUO9XbBwaP&iZb5n~T+dh|8HS@9m{PWcBtUGo%(b)-cH&(rV) z_EbXU8(CtiO5&C;ryEtAsOwx?xU)+gAF~w0a_bplo;8iILOXJ<+5-Y^DbfqNdd%N{ zZN%_LAgQXp3?&zmAm^(rVfEGNZ4*c87N|_D-zN}FBj*Th+J0a8Wj{7 zQiobaR{0s1`Y!oHe)?Y}o^FZsr>rOWtFswb0U_7ZiJ;do1~z?cp$)yiNiy#g`lV7w zhjL7)+1Kgh`L77-Az?=cN+i${BQ@dn+C+M3_&8~Zdrm_h?W0RIQqgzS%BqLouaLFb zk4TF`gFqe0()IeWG*haJnESX>JF&;4eL#oCb}pgecO7X=(?&X9?iubq6$qfEE_^?K zjqu&XO!BVoF}b(Hf}VMnLSH5R##3L%!h%LZI2V#h%8r$&OT7&ICrzfU$UJ3Eg~KPv?mLv~sHxG-yJpj` z*6uWSjRo}#HKB|B4${Cw^Ju%-dT5(E6}B3vQZ=8oP{*Ev!bk;SWVIcglsAQ#PVu1k z1>NMepomDc=g^<`Z_x#s31B#T4&5<+0d=<#66;w3AjPqu?XK(x$Z%@4-!`y*p#DAjNYXtPhv07S}qbgvx$YU85NxY=Wo-z)F2zkn~ zv#zV7#dX(-I%y#zQq`2KYeT0JUXl)L3<^7EVn4HcV18VM&cNe|y9A;$wR6d`p@)KO z%V@MI`wUuqSc_Wfa7kzikJLnPXhFzz`com2_B$8Re=)Z7zjzpQ8j5nb( z=Ec%=i>G5h$z{Ye{1cthQ$VjS|3?h0j*@}5reI=a0`n$kQUm!!>h>gs{xFFYJowp3 z{wg^NUe>A6to^M-;<5(W(5p;YjdHRjtXME5LlfQ|HpdIB9N?>)Sk=B{2?%{Xp4Pi5 z;gcPPf)5@p#76tQz^TuUxGHJW%?7z>uw(~Gbvp_Hmk-m@nV0E_(t~70du$b0M+>Bv zIN|4GnMm#NTFPwA5j4I@fGJ+r;Irtv-l!yfDIG!0N{ne4=>qFI2i8K|gH@HkEzk~? zfS9sV@JB2i!Un3rT5A=1t)?4Di8(v+R1Jm&Mxc8775X!F1>5UhA&2@N!NsyJIPyUk z->mzIdjqSn_E|U9Qz4NZUA2(i_|pcg#V^9)>8Bu4Arcm=odxT6ZSZxSI%{_C73Ac% zRN>fXD0U@}+P>)_ttqjj`*RP#7A3#pmCL z<%!4PpNA29EZ{m+6r~}*Dfy73o=z@)tj2f!vcUJwLe@FWn4ReLka%EgvP^9N&;C&h zB}!kbqU2U#m-7p8@v8(}@-&9E)pa0$e5s(SbOEce_7_|})rdDZyu_b*+H9=1C@bn} z%DR`2!_R#}!J;x1XIP(upu#s~W_j&Ac%Ye0&7H6k59*3g}2K0I4 z1Hl&67#Npdjv#Uz`o84|G`4iWJfG?8wY_O@TA`KnE30G0=m%K(=@Z;g7h84CRU0g} zxC##M5|A0Hqj1HdhXfkUqP$Vj*_|WFFVOnPe(&vKzik-@Ln2P7AMYYeemN1k4i}Ta zH>+WaX$H&-d`-fVG|0oR`vv+w5xDpG&&ublAqDg_&wGND>H9G&-u>+ zmuo;4S-*zRv#~IH!4Mm**u*xU)DrD6AI6{NjA422jS!va$J&mFvl@ka!F%=$cI)A6 zmVY~r{oA^ZZdBOHIOi#V(pCu&UlIXm_X`+pI97GBMISXSxPZka5W9FEmpwQ4C3en` zW+U5Pf?7fpWGi6MjYVHDi~an2ANxR*$^RV`jMr{nM|?gUWs5{}&d<6r ztV*L6Y)YI3iGSQ#hhHuD?z=FUbIuYkH0&Y5_HU?q`*nD_&Iu0crh=8nMwntSAOBtX z1w0HFvEl`a!0S%MHNzZ`x;%lM>%V{vRpGEJPe@(&oYe*`8AqXItN??kXE}NOWoRD1 z5oUj{#19ldg5|;zaIM-2_rkj2zlcw`BtsDn3|8QC$-7zJZ#T9~^x3Cd7fl3FbFPmQ z?OJQ-yd)}Tbck&E1vutA471f!z(=hBiX$air>(CcuA)Pf4GLp}D|?_UzD~p+Y81S0 z2!gv29ne%5h{pTf$ASENkZ7?&u&7!YCfXUn;{O^^w!jtC4=#lnUyj1~u^fn5{RwBh z;^0Wf$FRC^KZLIsgJI!Pwk)O`nz;kGMJoyq@>+4c=_f41Z6;^aVxVclMzE?71NUT8 zFx~zYR&D7=C@SLJSWrPzHCyU(i|E1;Z2P5Q{b4#Og{v`gSLmQ4x{rH|x8CB1Z+5 zuKx>un@7O=^EtLkv==dU>;~H<=_auMvj*LtGELx@v;cG^CyJODQ$Wo22%7Xa6gx{z zqnzqgtoq*>SpFp&z5W?SPa0kon65~qAF9QP=iLkF?SWi;JS735Z`{QDR;UWg&ucLW z&v%g4S3ik@Yy?)P#@J|AA!(N9ki`bpg0I`2(uWCBv|cI+#h30Te+Pe%wu;Gu=lUJ^ zj{YrOL_QhBhrbM7U=E5nHo#YA{qsyzgBQ z7(Q8nHT(MU$otjwQRystHSH`}B0UqgoR=j{HHyUO*A<+|)Z#sviy{8nK`_ddfqjaB z=$7w6>>hPkboPt`-m(nz$=!)r5!xVfe7gzUo&?c7(`LZ3#S~vyEe=x)W$~$pZt!N` zeY6MGQHKxvAnR=b*qqa&GZPM=-tL2f4rNUd^hXmOjO~FtmEKTrQW2XjPsJV4Z*bp- zbdv8jM2g2GY02OWDD*UgB#ksu^k_PocPNAC)MVl>G0jz<-}hjJZXWsX@m>7SLmzIh zMKJe^sAe-6rIWK+fm-ie2+#=@#C49Rab*Xvz7~dk6ZR2}nl$Pn(M^pOX5#Rzyeh@! zFr03*M$o>-mAYTiU>>9{!eU`Z;jd`1a^SWJ&DazoFh8*rJ+^5flSDu}`>&^~+8-d4 z!u=a-M1YY9>^riX{Cs<#nCuZEm!I$G^U`xT(5{uJOm8&9mqTL(SA!?EoO_iMq zANrDEVPigdU9SudNo`f9bzRBJ&TG6|O@sK$^D4OTb}e3Udka`H=}3Gy1q@I->Ms4s z%<#P@NROC{H%v6E@|dK6U8~MO^W+E;YJ8E@gdQbpU8G>|ZE>pB(}a!Q{zJw=NmzcZ zJ~}jWk6`zcH-d_RJGk6UmU<4--qZy+p5tJ1#7eUF=3_zd*I{yEtuO9_ z8i7Hr8Ge7r2VLe^62%KU>7;}Y0_j^dXs2N_el;ZlYsx2}1$9G$7ISfWD<@gt6{;hs zYMUhR*)=M=sR(!S~S_`>?t4DRI!(<&C-u4SA$qZuX3&VHnRH0?=a**|&Aas~~ zoi-OJVc*JF!9c%gKJe-~F%Axd+#&(C^-ZDo*U8YQ26A9`%T&;zf`x-l_K>x`5s&+J zQ^ekm2I+M^!mZw!IQ~W*X`LTR-fa6qO~W_QnCF_1>N}OLOAny4$sFM$PBPiQNr8kr zeFs&KrS#~CpPZ8M|$0sz?qvO2H<#(5VeW3nlOTA$!C%) zqB;D zt}3Nfk5xt4r%RwQE|qHc*?}zF7R21RO8c(YfyRHvLeHQPDyG^*>+(P13WH))(RG1d zJ~M~Dwe%ERZE=FHs$uY-YK>@ym_yc^TT)cq$?FwMBxABdLDb`yMBPkWC{J^R#os=X zdMh28*Yb;8yRj2yuG~m>w>yJyxs;GoxI~C{ZxEg;i3I19If5;M)%3$oE4*LvG)i*% zOZM;7L^I;V*?lEi!T^0g;r2N*g*z+{3D>;V6F!+xOzXZkV2|bNsfyNo^4||JGP5!a zEZgqkxq7Fms-TBXTgs5WQaM`e)k7|BKS5kpGn6d1r^)8J(Da`!loiwoGX3MJiDD#u zxNspvg_{ajsc#e3Bn${XE=^(9$5zlA&V^7N#w9CeilN1~*3na!)`Mg#2&42`sQ73e z)z`R87g@Q{yO80}vmFrOAgSD(E~<93hzlrqH^e#M5_8*o%B7uJE5>S=KbY=BD`U!VY$x>4s>xRJd z=vOA=!Vpz^b(-Ga(L|%~ek3NJe$%LUbBcCr3hM@Cg|9CPVTXRA@Lqco&6-dO<3F)G13h)I6dYt1C4oG-7}J?w|=b^yz=ajmSQ-LQqgJ zg+7~e5e}JLr9F@Skz-k%cxJ{4`Y))H-adGOWZK8j9G;3Wch@x0{quU5e`XZUkG57N zNoflrPaY?i+Zek3&IHG# z7DCzrDv{2^L4nENPQ>j0h{gA9!xyb3$mIn|`KGi_WPE^qDK_%R5(}6rXd$4AI1D;cO1}tNQaN_QMGE?j(zSfb6M;|#M z$5J^sqNochQN{vm5f3scFAW!*Q5VLh?IBv9#|gCxRtY`NTp;h7tZ>F?JiReli(c84 zf^YKfi}tLdK+9tZY%tXz(Z9aZCEH?!P92`Y_!nuo(%($*s{9|h4Tbcsn5uABpr~e9 z5lwF2il#F^nF`H16OfsQF@$fGM<*twGLrf4nBXv1vYA@IMdD0COwviDdm!BVv7L@< z`b-t|tVmWw2EFa6Axt${K|@2wk&TDLnXAWcK=ABY-uYU{b`WHi1|L(=lORwT{^Q2IbwIseDr-P44XOKTK`Sg`v6}^>}ON-)%N$R6u z2p6j3mRG9qQ^X2RI$naVR05Ql1)_`Ogy=I-8}_yPqTP9`+J^_c`fQ`(4#k5yV(;K1n_uZN|ob!eOnl zE_GS{7Fku_W=gX{tBy>n6?BETz!N=rXxLPSHcCs9{8?+D$oaY8xBv@=z52*^^EB)# zlTNzHQEHtYNHp74;VIWNLG#882;aM%sVxblX+DwUv$jYDDtC(LWJ!^GP77e(fdX7p zZc3j0I*hbL`NNF?O;}DWkg7j;OBGiIv&+yqaLEyu$i zcVXz{PnfcC5<6}46RcP@jm}}FQQ0rsn3IwIuxL;fbWYcRU}p?8{^tkg=H4`1cMtus zzn1b6DEX^-foP}BgCkY_*rRCznrl9fMD2eh*f8xNY~}2StNZWb)hD{}*Ys;(*EP}cMW)OmkOS}y(w^4vm&F~vQRaaif8{Sz*9vv$cySnwP?4CNiblb+ zD{vW?hqrzFilgrS!Tjwt(EsBI?96aQvf2}H!464ytJ_~SH|-Jjl@OAVOEV{nb)&~-iQY1_y7(3>qqQq~TVsZkNcuuPw7M@xYAf|=lzsD;loxMQ9BQ)%(k zUUGyh34h1xz^w2iUUAR?+AHFrIm3*!ylMwSS7~O%m4X3fE_;hxO;1A#<%J6fy0In>U!8W#RWxsWgqtEYn zVUxmO%y}A3t+#y^Y}g>qTAS%mGd*3pZ!m-w%3UYC9h;~yB0;nZJAwixC(=&XBUtR- zjWz9fu-lq}P4#gwx2%*1-9(P+YEx)8HbdtAs^C>IBDzP=hgGi4@H4-fq}IAYTTvx! zDa#Y|4@i=iNxya5Su+4AaUa{sN65&b)8nHQ)vs? z8?K3U;$EP4Sc}e3%%T_VHo?8M8L;s}1XS_#$gF8gu)D7Zp1wGhhBUq)(;X|&%g%Eo z_1R>6uXvP+h&qk8dGz3Achg0)$ItL&=RHVHxCGrlR)d5Rr z-JS-G6+@!lgE~#07f7?LoaoM!=|mH(nA*Kj3Sx7mUU_?$QUrk+fpNWSr9?IyW6}k;2u{ zENqIc`d~DV7PZVrkLm|V%uFXb{OvBv`I(9zl>J8nT^5qd?<&a`DkMu!#nNlC?Sh5* z_o+wtQvB8|8IKeNQN!7BI4g-WPRJ^7A1X2yZ8~ zN1l^g0ZCP|slSP#k%~aQ{Q-4x521~%3n0TG6l$e|@km7-N_QAX$A+F0vHPN#Vr4VA z3ID2`7p*4guV&G;S~6t%*8>ovHyhWLt|sS0B52ckCBb5eU{aXAM0oGs4Z3@SxX?;} zAI@WfXuHErx>B?sKUs=mk2^B3^P3Z`<^_`9C>MO5ZV-6M=~J&{57KA*lRSQX0LNR4 ze0@SqI<)tCm1cr7Z_(IuGIAKB{N`umzWxzX8|+Bu^0s5Io*0}t;W}A8!7p@%>w5c-?{QhzJlI z*w;pvZ%!Z^L^J#^BU9*xJtfqrXAV96Vm{>hw)4qJu$m=%w={`@_{LybrXEHM4jYingTdrb z^FsRNb}032%bukc7F0-PkP>Yv;^laj_~-o*OsW(0Cu`RV zcErkp=T~|BdU%XTEGxn>`f&XfY>y_Vc6UWjOAdfcCt-)P_-M-u^Kfa9M>4GkD`%pBk@0l%pUQvg| z2Y%r_|22@plq*o2`vP9Aeh)uFpCG5>{kT_EjD010NW_4j0#P9*5a}KcoJr|0&)Wez z<%*cR_io}7?j_etNZDOh2Zt%%pP z1^C8qL120d)V>X4k3*KAB#{F);bMZ-sb}f8?;+TA_B5EFR}TE77?>y`U|hbXi{FUZ z!J-xVcwD9(H2w>$a&N@=K(R7h7rTHj+g^eP4gqY}SOD8mnLu)PMvD5M2=3}j(EGbT zW6e|!Ui2t|cGXS9chAki?H~2Au&NZS_ILxY?;Z?^^2;x*Yd~V|bkd$z1iB^~c+&M+ z^49S`t>mPIl<;yYBX*`iJ(qhSd{8ZuPqllm{0cSHn4iaODGC0@ZK^NUPWw2`xsgH(wxDc znUm1TqK1SXh(`ONAE+E0hi5-kKvbAC*g5T{*(YtmSMUNCy-ddIQR|VpRaF_P5=63v~#AK8K{CCwKKhJuIon1|IM0g#;LKiSI=a<>%e_g8aq-i@WS@OKz~%RPui+%)_a zZHwiiH(A+-)}S5ol_CDFwEMj}&gAA`#J)ZnyJrHPe{_uI?2ck4OnU-m&+ibI9k%qF z1iy1iE@I{8Bw%%56Zm;+BX(#6%WxKpo;t{whP=B`Pjw;i`or8)VeEozWnNgKu4tVzTfxzfU2 zYkX1fVlE!q)J+$yeMN6yDM8`IaFY4-BI#PZimq4B#9^6adSH(-{`Hdqr@LOT-Fq%O zzS9=X4we$TkJsqp6(w}vkqAyME0Me4?T;H}Rq)%xA0+MHUsC?km0nj~#`}gQ;Lh-s z%=BH?sAW(Z6^&krB^O>$?Rr#>am())h~txBnnI_-Nu?`OiGG`zT%fvL3H&*^4#@+_3)k z(aOS7NoJL80d1Pvht(>#Vc|D3p{H*+-f_yszGI0vZk!+Q8M}qACKmA>o_>_QDo(^K z2L#qt{={vuCe5s{Bswui*gwC-(N=LB>MF=_5mD8GjFmsA-PXb~I!7Sam$^ zZiWWS#ZlXVWdt{8`Q!J`R?y!a=?`Dc-xwVr~DV9+sJF8HSaRYUUx++cU?5w?~DqfewcDW zowj)fpw#wubkENuc}+5SGq{iprsdG2pF%V@e?wQuwL;OIm&D^p4Bb2xkpz5WH#TNr z{?5aw{a-9Zs;{HXGVwTay^79DxkFXoY%gyBWjv978s*QKt@nVe)ZhcD3bUB0+hb37_fmOwYH z?*eC+7<^K%#wCec0u^irF{4)&r)8_*q`iUg<@JS-I$;xRIrR}*dbNdjcb0r)29V&r`sHnhD@MV zH$F(@cTa{^g+?}YsW%SE&p>~JqquOlE(H8;0RNH&xcu^0s{|$r1^D_TIBwbm1>w@~hRJ zrpB*49gHFS z-8%SCsDlkF2if>iO-M1KFoEaat!VYfk?sWAniK{4^G`v8Llb<|kQ1(y-vXw_HpFMz zdEnd{z->c2b#%#~m(!k+zKtGe{~;4^UbV*Xv~NUl);H2ByhlS$%m;>lK8rw9=sG@donofF*{sl_96N3J1np;286D%&zB0 zaO~b7H2lus_c?{kxgR@e(C?Y>T$g8i^_k&+Ng@?86RPpbxvRv2f44q)qQXC9Gufp` zpw0aTp18b;U0pF1*9MAlt87+MAMG)qYsz!gKAgu{H@}gnMIG?6wVO;lok_?;V-WKV zhmJY+OzbsLeBCA^)SEj|SRxXSw!G&%WM32csj-D!v7gdlSH9b`C_uPz@EaC?BKXF@ z6jmwT!UZE=v2<)9gukNrXp$2SZ?>R521n_fx`|XNOA}h$=Hu^=MV3jwFTjP9b<{L@ z6n?P2OqK-2fI_A(gtu}~_U#)H`13vCdoPf8J=U;Lb_ibWH-}_a7rsxvMP^K|C8MmZ z$<@h|siV|+!Y22T^qgnRjU(r=TDS#fd>%y}^ZA2BjXb*acMrX~_#N$`PL-q<)(n;7iIynra^b9mk6x^jTU(TWcS63p2+*$3@}p%Ow17oDO$X!vIZ; zKGLyuj+C=BfF0o#Z2p})#9{g&ShikC81kV8W}W%Ov(BP~*NO~;;v3`W8UI0YSxSXe z4P*h1j|G|MyvFWvgv?1_NQRr5$oN!Cl#&{v>h}Whlc&vblmzSho$+z8ACiya8Zpz!+UR;H3M5|PERmh%Z;bz4|p%N z&jS+Q*g|qvPNG3?d)UFBqIf?{jcX6wP0D`P!sfFQ!njdAm8bsQB|3HOP+HT1y3x|FxoRDLw}7TsIMi7(_1FPp)5zd(=;7#{&$AFXu1fFE0m$=sT!y~I!}H}#8VxnCRF4A}ZohBZlrq~>)hHE30V0fw$ z*A+g2n>XKxYIAXDFr$evdoL?YXV;R~9^$BAc$S3b{^4h$>s0xcHt0u0p{wg+T3(q3 zr%&9#nKMh!V{z9&hhULfXAug8Ni(wvK513f?RiA^!ngYnxupwRs`)-M(1h~6LM7JJc~ zy>qCX%{^Q8&+Ua1HgzT42X zedif9yG(L;&nRxosr&eupZ(?kO@oF+OMKaGixpP&WCL3d2`(|vB)CH#n3!Rb%{7Yy zzL5f4%g=twvcj4A4YZrjG1bj+A{eL6ik2P$k+xd!tti4+-QSom76c=McaaVFf>%6m z;x~rDsENCAXPOgs{wad%Cb1B3RvI^ZPDkG7%s%K&Wcs&9K=?IBs6F`sPK766f3q>y z9drdJc^syFvd4+Bn*{Q$GovZ zoi)`oEFg%6S`?55?zF(hDFUP;tYDm176z@Di2i0hmAC#?vMrf0D3@hPuj-zoaw`0c zd8mXMrs`v`?@@u+av<-ARoE<}B(kbr1pY1OnTT$m+0tzda85~55S(+EXyx@1(IE|N zI$=fPA4)^G<7oEg8#{W&tA_1um;~7g+aW!WLgN=>yn10fQCnn9S5*aLp!;PskI9B1 zvL0a$C86p2(B+X8 zfkC~w?3b92?Dz0YzSZ@>`erR3}Q{czx4fu(KR zK8XL=Nh(6$L6h})C_0i$H|oF1iyvX+D7{K^DO-v;?T3}(Zm%#Iw0xS>d zr=GXvpv8lqUrz?XicCGU&Wp#?sp6bv`g*+XpNt8fV^D3K7R~g$k7h@_>4~LU94RSA z^;-jYHvS&@5Sm5mla4dtZ{2Xx*ktr06}Y(2kA^N;f^BteYgQuK zMg`(o!E`EA9F4M{x8PS*A8gVeq^T7J^d&zVw-^M{L8CKNK79_IwPQUIA5q2GM{9{& zM+6+7bd)Z;BZ^!8t|R?Bq=Y-i8gjAsqjB{0MzT}%EcS*uVeG#wR=%8Puy36~CA@ay z-II#+T+%tri~BAW!Wz6c`TZwATBfu0W53ch=j!pr zNp-G~j^*YYHpktTO&E81Dz+T)L_NPY3{I8AMX8?D^k^!S7+(VUTXQknwTxN#!j3*} zkERLp?P%14r6j{M3ElSCa0MUxu=!Oaa+wzB^YlEeIjo3JzxI+T)C^^U>`0m8Tiko7 z5&zA$g?ei}qVZrR8f6dhoaEIoZ%aO_`0_Y4c%p|3)85h?r%S|bi!=SL@RcNgJBNN7 z@^Ij49hwg`uyK1t=!CdssN?sFE`4-{eKvC%o9$zT=Z@cJ_q-2-O*NW0ZH6PJWgMeM z>$S+yE8k8y5v7{%p{*g zFZ19%Ch4TC!yXHq+VReK~ziFbU}zaH7vG;;Ouw&Onn@u znHZzVP5~-Zc%oeG4vbh|kDHJAGlRMvbi+6?R1_=1H)B%yAWa@Eap&k}9hSM_8bLoL ztwNFAPw<=ZAl^?4q>nldgM|n2vO+5D;op07!B{M)>LFE+kTx{jqis69boLYmD(2bYlLO<$++SsRqS2B51S(-=J&pkSLdx9XNA$eImH2H{>~^!sC_i&TXVyziz;w z$UAtiOo3^x;q+HZ>`GGN^+MN3NyO_7YkyNYC#nYj`(Jsl9dGzuWea_@zw;PKa zUv=TGeYI4}a4&xL6h((w3OH|G9J6@aFWNb*i;-_ep%Pf*W$jU1#gC2Xq9MU~)qNl# zC(px-;so+1asl2A^??t9uazAi?eKK;GR6oT@cZ~>Ox4*K9Gf(V6Z}#z+lawqtInds zT{%uYy9&eduAk;_~Jt7~FghhmGTKh|9y9>Z7@wY@{yfTX6c9C_4Ld8h*X| z6u%v6!JYRcaY*?tjyPxF%CR!IY;Xf~#F&#WvDx%tB*9GOL+F|rhSO%gps^LF=*@lq z1R}gYQn`5=DrTJ_V{^9St;5&QXn`^s?pc6Fp&2NrBZ|cjCGqVQQQR@gj&>h8L}m?} zlB9}bG*OqJ;*3S`bJKfjU|K>~Sv|yUGrp1?BGO!!;%|(5e4ly`_|u$j8Js*L3Yzkr zsl)gmU=r3tXJ{p`t*b|vVqW)Jx=8}&+gOptFfEc96hX%Zo6#?x3+S!gTGV6CQRZIz z0o?n$hJ2mFvru1Ulf={GLH~sZq1BVY?~gC|&BB5iV z5k%@8sT>u>vn#{}V5#wl)%%r86D58!)i!<11#xkF`X~!`PgkZN?WNhEMtw5x+*h8L z;Yp=DBAH)})r|6nC1`juot)?&!*joi=WYeakB* zzXTUT3aG*3=~Qq=ljK^*lalo}@mVO3Xp?criFR7F?e%%MMzK6`zb z*Jbb0;r*5D@y2OH%FBTKeYTCfFpGzi>w|>T7sau+*RO<#g{?$s<7A9|qKF$MWcfa1 zf6y(sPRFJ_L0@0Ci)pn4Vw&7TT7Ty zhr`gBhyvrB9wPWYk9Pk(3mzLXU{{*4P^R(>ZCaT}7VuoVN%P;(pBtCLT@N+5WtIkF zCljH1jI40ReN|ZbsfK)VXccg4>ZpGSM~*W9Ga^DOmFMb!+G<%M>KTi|<#zlYs-E=4 z9;R;B=Hk^SKvirG(Y*oV;p!)C+Gsn5a`sB}U+gP#Z-Wbc6=KTO29*-t6-H@Pdd+wLg9&Il9oA_9DckWhi^@%eTpwg zd(C}j$)}T~Z+-*)75Rx-x8Mfj=XeVy<((6F`t{SNHX-nkN%q!wFdeoO6qHeQ^oCnaVPYrY~Z{^eliAnlmvI zDoCm*?*Xl>CIRky@J2=)_Wd=5!*l(J$Z$18c`gL?-WN=FloHt{{hF~3&;n0g5pvx{ z4JPHDW;YsnlDdu*y1Zc(Z9ZjB;2!oCxDmI+@oDFJg2KcL-Km#M7E2w_JGNJxVs z6|5Lf4n%(?yB;ah9rI_?hl!JMYqvigJ^ducm-Bm9$>-#9NIC5~rc3jmE+AFQzmuAB zS#mwb28ZTKVN>oAAc3)spWLjDxN<&lTpxvf#5XnG8v)(9xQe zRBGA>BH1;OT+FFMZ4FN%W@M zw*Fze27a@}2T#&|-A4lLogaw8mXjn-Zay48wi;9=yhujqDze$h8)}~@)8RS`;YYGygAI&J?^Z}LtRP*b?hu`~khrh(L7`?OUd>nl z-z(%{k4zv`xQLMW?0IzB2i|vDf0e%bHu$EUM6*JSJ_!kV`=}1kBnTo8Iezypo+48*+&H> zq_#z#L{3;v5)^rU=I0iOKl%!$FH^$cz8j=NatslH=;0*IeR$$WEe4ybF!3L>NZc*HODpj|QXOtU zUzsP+&}(^A!8``%tXz!`R+Q0|;peHwt}JFm)|i>}vYY5^sAF0T`R_cF!!DyDni-xX zm_o*ptK3RleRly#{+Ui68Sq|dsT!)ElSNJ>Rg!(vZWF_ALR{ZBzw&fKG(CGo1rqve zi9ENRyq%&2HqNR7>*KQ2sfXtU_36@WRT*T?m!lA_%-1I!e?-#;&fzmogL>SE#p>)A zq~S{&G_I8sVss8<9^1lqkB>lv&m`drS$UzQ&0Z>gP?D|+TMY`*e*{|Fm$Qm@GwJ4E zypLhjWPIzd4_ez|gwI4d;dsRYVfd1*1iIE>#($zb=Wrt#9=Zc3!t{i-u?E7l1WDnq zaz&vuU#W{;Rxn%ZZNA%sUi&lEi;ooKonzB#n-YqKBLna2C{6&Qe;v|H7 zwmA!53q*w~hwt&-F8#cQvq^Yu`w8~MJTY7|e2N^NI7S#xdVyZC5sJR4BaGX6oX!-} zE=yOoI`8JyC2G0gm7C@HKY1lsl;ijq-+(~i<#TUg8Qa)8U zKTJ)y*w)OekwgwGCy{#Zji z0xF2ez-!1DP=TI{7a)GJF0^@Ef^3g+xKcD)7+GycWWw*0A&DK(TzG;!9{FimH9iNM zPFKN}Z6Y9|&(DT?qEOFW5eq90VRp=aLdENL$-9%_`lOq5k&PC)x<82|`WnKe zLPt{W_K&_x@qleBoj_sWCCG-hz)zJyXkYvlE>3zvLTB=f-SYcT>fZ^QE26>c^H!+a zkxjZzFJqHD?*rG9Av~g5FI@1M=Wv!Rqk0NkA!(K*IF8yTba3;Ro%VHh>rADdN@i91w~B0`5Kcz&muj zaP`S-c(M5-8D4AvV~^}2lfKoGlkA>6WgqfqbgDq-{56iBvI6$W=o3m?7kgGF||AkrX4 zrVbP{%c3X2W{3i{-YqcE_X+Q(lYo&;%H->k9Z+M_K(r6~2zv_l2pvR)!tRt+2T1Wf(LLL8;MojWq17Mb z$`iRJRH|>U?``f9~!8~F&Q{VHcxY!5)Y6&l>L+*w%AR6->rmXmj;QB-@oG+UD6 z2~WJF!CU179Q8T@g4Xx&c8a&KCQ(dyX$O$K$K)WpXFp-bn}WNYyzoK4p-`p5m>Az# z0kTcqpwQCGm_B_B%|1IJ)_x+)J!VY(4hqSQ!&+4RO$YP#5l7wf<)PrLt#IaR{_HrB z4FOVOP`%3mymn28`3_f^=XOUR^21~ZKVJ^VuPDLYM>l9EsiSM&9Kl7)2rK>M2Ar@` z5^l@Whr^Czg{A@xVTbKfq4SRI!fk#vLdnEB^1uVYwe$>=(KrDuvhvB|_vPT0>R+ zVqj)d47Tg|f2d zY$`c0y9~xp&IX_cs1&rB>I7y}kGKaUVD&$C%;_ZdLQ5#`qqQc@uV2E$lvJkftvo*N zoP%5JqR8Z&F|hi{U8*ZF0k*}TC;Mw&5?}M@%$fV6>0s>-R_Xa|82`DO6nNgl71pD0 z_lBqRc9;})y|@T3G{uA;`>oh@DjQ(yXTHqpVX-3rEL%z|8@kUp(6MyyHVJ-xVZh?xIH6{uxKMPeEPm>ILD!~lr4O%6f%0uF5ZCk%G)OEA zu4{uA8dD+v+6%1abVQZO2n7$Po7SG(rAS6RxlR z5EVDyL67$zBq22jjj0`$BInQHi%b5b@^1rd5e*>X?IO4*vXacUxytX=AJE1>rexdI zQ&6vej8v^~g{M9vWV>Gq=+?}Do+c5@<{7>6F43SdEX|b^%EI{v+7Pr}4@EjQ0P7nr zEQ{DF+_!=OgNa#Wk9QHQEZT}cJTFt*sSTiAC5==09`w4*8N$u0lwoolZ}2G=XMi(xKDL91*oz7i4gSDDsS#t5R0tEc$AZ;#6`@PtYN%}< z!6}udB;?IH=CgGKQSz=~KiCVHQI!IG9x)qwCA!FPR{@>AYyoPQrNGiZjj&d?l9@Wr z6#@zl!o|iI==x*>>qlAPzSrx?n(huJ?U*%8Sw9YxtaV7e)gcI}Yt7bAEdeGoE#JG1Cd2b9L7=)SNmtz9PX>%bj{|Xt|WdSN1B4Nsq zEW(R4(wTRc?A-l@S!*&#pOkpe$LW1I$NE%7@wXY!@zD%MGV{=1V?vX zWu~lIhc&K4urJ~}w0mcOhT$A|^3njVwk(7@<;yEiKrvO_{E*62R>L9DQABjVB*<^s z%w{bJhRDPM5>@zv>|3)R9p(tIXreeQi8x0xXXKEg3^6kC&0PNM6REU1&`qA+mw-Zt zDUd%}LKt&*DWt1g!3K}dtY*q65ZiZ*gzD+ieSvE+p;-$TRm{aJU8@E8ok_%E@(6i( zu8ubNZ)VmWK1xRGjR&I^fR+4gy-Ixz=y&`jo4oIU{i7c8u~HrmeT@O%ydOk(a3_aj%j+4kEV-J4ktp_f3TCii$ zG|GDPLM5+&zOui_=-xXDPCWv6N$LbT#uunm%6#754d-_K@Z)YQ8q`@!3f7BDATbXDKvAm0HewtQ;UHP=6Q=q<)(yjV6?)MG5hoj;>Szl(>9>% z1NTr*V-2z;z#Ik_xx$s847fFXpHBRGm`a2^GS?@Zvx)0lkt>jH0>fq zZ-&WasmB%?6YT}l=E@SkR1sli>{sGuYef}js-R3^E&D@R4z9e2fL-whpnv}kBat>8 z(&OD=@N)xuKunsMczX`(nU&0(Zh8cL>RQ4-I}PBWDOCJu38qix4Kk5gLonf*gYa*} zDxrJEC~~M}E9hTL1*>F9a$?U9fpB7En7y+i7#@c7kvB_qv0U*IIdQq3q$`MHxewnt?8$q~^Pdv0-?mg)T9c0Q z-OpyvUdP_coWxw7Wy5^?&NDB)OCox}bjg9y0zyit-5qzd%h)6Ep0$Ov4*z2xk1L}sOIvBx zy?x}Lx;SLbyFf46x55qSqk!s?Wa-&bW@nuyt-Z60-V8fcx!}As4XN>f#O3x-(pk-v z59X4#R5_A(-;p|bw9+&Bo^ZZt6Xb49XVzP)kQA@(`(5HxCrfj3oh?Ce%fJ26=6(inH8pXlGj_IZ!JGFWx_6zKJ`6 zo`BMTp>Fcw%`6Z&3xyRs=Lz4yMqzF35-Lj05LGgfF0@j@p2Od$t40XdHMbOByqk&p zf=|=)Q&c&RSF5N1SAhPCCLHHd%0>Ua#vNCxA|+;t%sRPZ#-`{r_wDFd?q1+!_R1qE z%=?na1x_^QCQN_C*a+p&`DYvqmzbmfl4qoL*AK?9DwK=)HozH3=5cLLny6EXkZu|$ z1G7@4xEQZ{q*gSP=lp&lOYRL~am_DGOI^n%K_Ts~b)q6~RZ*^WgjUC=kh6I`N(cqpce&wF{NLK~A zBQYHG;>uVT?!?tu?5zWG@J=&=-1F^bYWO_Pw)BfsEIyvZge7p{qxHE9W6W@%N(~>* zFhb?O#>lePsQc>-Zr$*h>#rZ?dX!gjzS};ay5(C;T;)q2<{!he`rFX_zz?*qA>6?@ zPdxBw8FTNEDc3Iy=M*NT>Ltg2)kGhjc?e@v9ZjOev=o;OfCj=QKo#7$is$+<1epp|`@ z%`{aaoqkZP z>0Qrtpz3~*yX8K>xf_mflUYSFU`23I+(AyU*_nPlZ9w$8 zDBH*)dTu^}DmyF%;bYoKYtg8{Y>d6Jnjb2$87Z$mM<+0_oM}R8QLg>7WcexcY>^+uc;ut^xgrqWFwa3a#c|azkgmxn)niA@-q6)ueoF z?%#Q#} zm655Qw(RvbMTTb2r8YsA>7T|gG@^G9{~UN*lyaGkaLcn-f7PT}gk=5UF8L7mq@W6&pu$%4l3T-_I|s@P|yRo~y3 zlNnOQ^u+niAX{ICAswf%G437eSe?L^58|<3ArJ~{tnhuMEM&D)7OwlEoU9bLG&~5; z)#TCLQMc(7KCd(Icr5qw*;@>AUCNy@VQBTJ-RREP&AEKW1V3_D%S_%f@vT+(ZKmCa9_xS9j%5|(zbK$rQTfD1~%N|~@PwBgT}KneFkTwQStWxUh4RVi<=Yk3{M?rg2R-<-*wRZ)h;#UuFYxe?G{0g~c{oS@y6 zJ6T}KIkvB~#H&(hZe~l@iw$CS>0|ZhL-SHR;e(Ccb-dG$Aw!wjhal)s*XYuvAX4a*A0Zf@~!Hur;=5n)) zI7L%KT&y%g3*$SmVbL+15ZH(=f8xnhZxdXYcaV79IF6Ac>f}>dBFbeBV1;}x7At?j z+h&p2b7L19=(vu(ShJiQUtz?#_PK(}szTbf{R~Yy+<{;IiBZ{x1~fcBmi|)baAM?i zv{)PsZRMg|hhH?Z@%&job_9R7U!%)dSA2Lu68%Gpa9UM4rn@~S|4sHJJGD(nQ1l;m z$G888eS;Y$8mXcE+xb}2l8D|x+1MGxXZt){fjx7I&K%u~Q?st1`0O`Ss!RovGUe#v z7Y<0$vRFgw^|Uf+Nbu_^zgs-7DqI#I5LzcxVC-QI6J;inS96MR%K2eZF#b5oI`5zw zB2id;g6C(}N8qxmgZSpU0;l+RCY7;$&jhb~NOK>yQT2cLHzl{OI%!Eqs_Ng_^woEYiRVUrm=Ly16$fJvfb=8@R$cUMi(y)=#CYMlVNp zfj<6hwkxx6%; zXQW59baaTVY%V@%zQQc+Il%ffT%tCf7ufY75*Qh`1h|h5Fg?=%TU5nh)B9hf$!t2L zu}>`iyV%5k7w>6sUo81s*|+Vq{@P~nDv~3?msgUZBm2R$v64*wlSs)e zQ9*;jXmWdUCylk$CL&Z9ns+XRTwWtNkv9&vJ+Q}t^~R{e{Umk z0ZTtrvFU|>1T&3)l3N!=X}~8pIwN=AOr_dwQS{U5a{6?dC>i*mg=WqQWQwE(9T^^>e+#VGxk?YHdqpKp78=1% z!xVvJ!AEvpp#gsSD~is+ige$k_3+ol7OikGbw0WRPrcYc_J%~li-1&AD-wYM=@wEt zV+WDZJVGB-P9#x(l`35&qRG3yk93*WS6UyShF=}`;5-XIYRvn>pKQ27EVC6L;pA=0 zJgwto>gRc2pD+hhhg@mbAn&oWH%IX)f7sz)3wa;lWSsR!9d>uFu565%MP1*Suz#9Q z(ekJS=9c~;xbv4~DjRuCM6e2W?H>ygWye@qpLP=H(k{?dE)&H0iKC6>9D4SkKaTCb zh}&H`*4Fe8UA_DF0HmAtZrTBWFiL)1p*a=an*zg|wndlJdd&SVFbwCM(R{x09jvI`Q`v$80#1Q5=w32^@ zg~TEk;DLzR4L(?Rp}t?cW7Qu^F_Cz;*3 zg&B7>kaf8(#2IVeQ++2%V!gx-&a4(KJ zM3}S@9XkE&erBIbCdD~j?Ca=|+y%_eLF4=gDSiCgFf2H^7MUPo*_TJB| zUrH&nDdGp8qcewv$sV9+T~8Y}kH&?eF0|re6r_1MQNz%5GQYnV?a%vg5t;6s({rQ) z73k=*Kj3%po%*D&kF^mI@(J|E0Rn=~o37XF6^ ze_f=__40VZE*ZOB^sr;kTc*;K&*O^h$2ISMU|nh&RWuD|cYal44m6jN6FV;8?_(8o zw~ZuT7rTOO8jfUc=Sdp&{0z*B)*(5;1?)EG)i@ISkIFsX%Ckd`v++401OzeP?`-<%iE0+bUU(PK@T7dHqbpdp#rfMI8M&Y1}Vd0LS|d zV6?kFI@gS*bqa}e{=G-=_PJf<4F4JI`?3mlW?}_~PB~6JzD~t)3R~IOdnQz=cnaF> zX{W2I8!U(S*0SAkP2|JLHlB@T$;JOifL*r{#=pzO582gpmY+7hyOxf65A0#}_|fdD zmCH#~&l&u(Yb-i83Yp2*i?B7!fV&+2n>jdk92LF1fd)kMpb4KBmy&bha{|_Q;q!K8 zT9^UWOPs*l&s&K@z%zPNt`0x?T%kjY*5RgaIXGwhUe?Vik-XbcOY0x+VW;xFbw?K~ z5MR%3L=Q3U+@uuP)+);OKQmz?a&MC}vZCBzlnrMa9nZ~OIzWX7LtZSh?EIdZ$Hf%)S8k`cSbR^H9^ z$4}L1Ox@O*$nNVB1n^m=cE5*2>{~giwn%e6k9SZ>kqJ0Za~WS>jmD&|M!GEApB~F9 zK~R*1?w>|D`t36)d>BKb_xBOqivKA(@4uejH;$7gNn0stPYt8q=f2NJMyM#Mgt8M7 zMIx(6JDM64sf1LDqTc7aPEllHVr(C0_Kk_+#~ixZ-Y@b>6J>{QJm z31@agaQs7Vb$1Zw`t>a%JE;hJoL)e#?KgODtVC8!N+rRrV~K90J9z7yqIGR!=(a<1 z=p{u9SpH6#4k&!#T24?%t@on`^WI# z_)PfMJQ|InT41Px0(_%zSdKN@pI0=n%kXl zkDrVUL0jld3t61q*+^YKUL=uIwy}v3u53bMB|TUhNgus2rVSrh>Tu~Dy=5tI3wEax zvdoSK#}8#i0UTDHKTJiDC&`^F2#Y}zyXK#uC+|q0(Iy646jiCb(Dn8DXFAVoqb^sA=-VrD>_WLkZ1%2SSdnW5Z!@J?J*hzM_tshTRMzy1_&GZc~}y;b^COzqY|`#V4HFBS-T2#bQn)ZmQr^89~RtM93}Jiw3=SFq73q zy?SHX|E`az)2N3o`|*78)NTxyN~fOd?8&54sU&N=H9L2~RW3BP5X=9_v$B?X5dYGi z?Rqf^FYV&#C%r>-fsH?{A7RIqIqhWEbkAZ}1{gzd!33h~CeV5bdE$Rb`EU8@~`KyzsC%&-#H4VzvrW`ZxsX%IC9=Bg{6VDz}1D*2L3h; zQc|am1?H6e^CeL55{3$Q%=<$BWkkaoWJ67OwYi&luJSaqy{rj%;hQ;ck_Ij4j%4r7 zo4|T?JqF+EW?Xoo2&TjrQm66StgO>p`l$5^%HNR0kTEiRM@1lUd1!?SAKnn_d+yMi z^aS_bIfu(cBe1wO6CH6sz4$zYbx;jtFWD);9pV1^YRD+6HGe36u31l4X$yBXfrIi^ zLz0G#uB0CZ)9Bm6gH*dxhL~=NBu1Gk)SVPi?ZNkyU6YKKr#Dlj{urvbD4nRamBNx{ zV>B73gs0hJ%ypLMl|y$>pUW%g#<@$V<^f^WH)gCLHO-|@R*uCtN3V*UZKcT*<6ubJ z90h*EXQM%zaKC>xm$yA-$PfAS7Mm8%;OAvFpo8iMxLEoYD$Y%yW}-@P8ZhE&mn;V* zLp3_;zh|^&nJR1LdHrb1$WI! z20foB8X|Rq))$XtjRa=Nqe)Bb_MM0W{iT=0g_9=IbvohLx7Z2S)fZsG`a$p-w~Lzj zuC1=vy_0TE8A)&c`9%vh%90-|$FiFOeA!KzGDO<)I+3q0gVvyN_@zFb4wpTJ`{g^} zy;>%8o2u}y&Od}asT4Fho5_$(UUYJFEIhfg9(!|zOu@qjPU+$>-0o?LwuRI9Z?D(z zk%z-LjpctpX6;obxL^`2PCCzJ#2yEYAzRS;=2MIqVbAZ9e?bPlOsTn&H>tI{4bvKR zs0KTo^g#{0v#rA7t(90FEz1=8ipZoZ38LM<+95$R4xLY(z>;WdoGRIlKK_z?{nm%* zTR)tP+@p%+muHB-SqGraA2Ynod?M?D2BECWkcg%~6wa@njLm8({5t*z2D`@4py(E| z>q{P|zgsxJ)-MEcbs(3M--8SMF5|SvIyfhDII1g1@RJXpC(p7L;kFY=(AfME_x{L& zOSRK6T}P322xYecj~ZMv0CZ;(RU2EH=@S$VjM~TF~Ds#5;$ukQgQRP`7}+)&z+z`fb7Xcbaqwa%|A}V1cxQG zCAXZmu025RKb{1deOEa54aJ=L3m$yeG{StVQ{bRAN#If@z;r)<8d5kK4V_&1lCq~z zqZ5Wt$9~47cM{~vDMOrJD9Il#rc|LX2Ui*;Krea{a(y_?(>sQ~`d#>2Czrn;Iif~_ zyUgI3A*9POLtJ+>0o2r-`O9aK_wd_Box-hg`OyFPmm$k&Ys^s^S)dDgD;EnMWE*<6 z^AvFz@(9!n)0tOarl6ve22Q+Lj%O65`1^jhIfXG}`Mjoeyq}H^w{*l8a5~lpty|pL zEB}3_LmcDz>i&^5%5#I=itmcl=Y12sx7-;DA1m`?JB{)D)g;<7{){+(U5v3HGbeVQP5p7xTEPio7xr?h?aWsllru7_;FHMzcd`VxrJP4o|{ZZYqB3 znZfVA8_#>*n?spxkGS?R<@mYuE)ET?g)KXKNWl3;xTk9?jyn|(k=eHVi-d1Xkh=la z-u%Rj3XK8V+>6k9Erv$7s_~9Kd-+Ga5lR2kN^7l8+dW?}j4iM1;rcelkBrsG<5DE>%8V^}q$CkBJeNe?zPB#AX$w3sbDLcnSKLHb+j8eQ3bh}qIti-Kj4JQ=x_ zY6@rLUTq7K_vspj=nkQ}$9>t`IdkyuufI4mJqw$QM6CSPv9#C0f_im6B$_j`!EDbq zxBP8MVm}qr)(PFCssgJ=$FY&uO@j5!1o8k)P;Y zjdiTD!~or_uZiL6cVX4>>$G94;AIwA7z@HHFv8>=PM9doy9;ypV9&{T(PRT_{|ldU>#Ft&~A~)*=ZiY^;6@q zu6Qm6hiPC_Y#7ye8qYdS(PPc$52Xngwz9TgUD-l)C!*WY4xfC#aD#@=VByFcaPM;u zWTXaAtG`Fcz`!6axm-l}6;Eas06l`=*;|+x5&O)!l!mH>6 zNFO~zJM<3;%wY>OF{N~abrgPLW%-!`&-ClI2qLYiO-t@*qwG9;@R%W@<0~&yJ=YeT zvDF4&My#eq7ITU7hs#jMr_rIpuCilhEfw@J;OQrY>u0*5XZ<8h={!!pJ)cBx{E}p~ zr#j)0&R&kSd(MnF_7d;Z?4z%IO4$>(Q=qgz95Uh}!M^7e^U!WCWiJjv|C52^ptBO3|NMUKkFye`ksLR}R#Y=S1GAyl`Hn(0-?ztGu)n zll3j|z>X2L`|vQLVxxf=rj0M#J;~RmcqmqTB$_=`3oJub;n25hxU)GJ9cv;( zx!LD&+f_f#D{vH!6M8dEmggvs*J0R}x%9APE_0--j^^D?LhoC`taobytyqvqT7P+Q z<|_R#zWo^atWZHJ&Oat)zeR9Jbcos5XO44EEd;G-1B@Ga6t+*AN4s_e(G>?0VdnK` zn3HuMChk2(WE+0K#i5bZL_Gla1@1!n+YWbD+tF)F3rOS|Q||Y*7}Bv?omd+zqF2jD z(_vDxX^E~Yb#E=;ylWHir|Jj<#YoV6kxGu#o6rN}E)x6oi@}`tgR_$IILx4)iTGU$ z`v)h|55l{O-98$9bJpT(F;5IM)#+E+5~{Z}g_4H-G=8xmwK?p=gv+N=oByWanPE*_ z@zNbI-PDnW+-fBICe+Z5i{I>4m4AfHDYCSnC=RsZ^RRIL28^C?1UJ_^V}?k`0aV7* zSqgb1AmA`W%^ZTmr5mW1_B1-N@;BoUn@rbb8?!}Et>{5LIr3`caX8RyLSA2Tg!ah` zxcIaj@}ToRxcyoUCR-j7h4%zQ+!b-v%`_@2?TxyL&iGKOgt*APB%5BOljc!VK<-LD zK8UOZ)szxlka8Fv^5Q)7WB*{>R`FXYxg6=7!wfrfr`*#Ece;EVU zJm$i-CPUnD*_G~gT}GCzUxQV>!|-RtXj}h#D^Yb^D11C9gA&y*Va-Psd>*zRt1Nk@ zv1}TB{O}HWqxFQj|IvYx6jez4^$nUcieTwvJðp`+Rx=q z$AvQeE8St=cN1bL541ef+C*0-Zsk_!*di~8b_%WTp`ag-}{;Hc| z&yg2E(JP0f9u<7c0;^c}0EmA~ZDUR*b#a~sVvd})<95||VX;Ak_)OIaPU?go#uQD@St3Zn0?opfr3RJTIO%%n_VEJAvbm z2xeJ82lG&-T68#Z7B{!~4^te(GuA#moNH+;DgS3qgrg>t9dAgcL^qS5V;tGPPnC2! z@#5WIZW8U0dffRKj?7l9X2KQCVS2JOr(-$<<77^frR|G}vgTxnT?W&M##?oqxuSD@-Lx zv%}y~v^QC};U<~#3!o`-H2yb1aDeIVgAbq6#eH)kIr9=DFlemiq$c{2oLfJcPDhbw z`r)VCsUPcL?j08L$K2#9G%Lu1+)Q%G#gumJIf9}MXVL$_J2>if161dhg1-4qZmii7 z((-sV7qnqGD4D1bO`9fWTT+3YYi~Nt9*6+B_p;p95yi0ag22q}`by$LPr%a)Ga)c= zIg@le2;S~5hFY&j;tztGs>ijT@Y?3oKE4~y*0*!Em(<0cOKh3HE2f68Kz}WGo+Du~jYb2em^R#D`={HIxCMtcE+*ZH)4_R49IWi9CUI_VqB)MWB-eK(A?M>s<}Xc7 z^H?ki5MV;vuJ>7<%{*%B(e+}W6%o^g;B_ee$a^$J^9MCW;g368> zsII&YhJpv{Se^qpsHQ|$eYT-{qztH;V;(V(lBP-0@$^NyDww29AkWqhhxt**!JPp#9 z7!T@NrQGYiYQPjV*&fio1)d9wL{Z&Oz#z{M)8~2Pr_>?fRQ!%>lTGG=C1k*KmJ*TE zJWf*Dgj|E90^!akkjrC=KxN7>8fEa6)A}V1zIVTVw@p4c- zJ&Rbp&4rd=Nj#x+ff*&0Ko&ONf}1x>Ve+OCxXJK6Su*!C*fyO&-?NF#{g!NzO1lbw zXLB!+@_&iTer>|#X@5vdvzT1>tsK2TlpI?rk!_!8A zOlut3mBvG{jwGq@yoI<7P`!KufLs8PpZVNTVI zZ7q_R5^PG6rUo)YwQiAZGs5t?PcQTEzeHN9ABg8^7G|`~$5yQo+|ljsaOOcIJu*^* zeX^ENDgUi>;Q9`FF$2(F$hRLcI0^dm%;_WDy%-rcjBi>02Cv=Bhg+qM#Q*jr2=Ju$3bBbAxAG6Qug&yVq0;Yezmoy9a&l6xxJ6b z{>$UOL`UGkqm|GmznDL>DpBBG*x<+VCwS0eh;T<5q=l!exgP>oZ z@r5^z@uc|xxh^iwm^it^+ZdG-}oaAR>r^Whwht))W$&Qiy4$HZ?t1S}S zeQ#i?!95C%v-rW}Sl(mc1E&?;z%6;d3)H4xpzr(Bgt=V=U#uy^&dKt_JiQtG*l}`f zx2Xm3(KQwrVXx@AOj9;++Ht&+Rl~*Tnq!8aH!435LDhPRRA3mcVe-_Z}R7p7rnnxOs6&816#d=M9bWr z{r37jUF%msdy5!)KkzY=Z?u(fUTV#IXxHFS<1#t4l~azJp%xQv~KqVg>UC)bH&lj)f{%XZ0Q< z3r}(uu}T8(_!LS^Uc-oFav=MI8CGwr$E3-DI40v6EC_#2lJ;Jqmi<@A)LRDp)iLIL z+lOp%ZD=xdu6PEI?3dw}kB8`={3gLaJOhvMd+5TaR&1ckS8}YYky-X+4;>PBis=2$ zqsnul*!=VRS>CFX{#XXoq^%mt?`-6@$(isPoAr3Bijs3DuyIUg=~`>^<&ZiU+QJtHYlWn8tG#I?5fhq83o6Mvz$IjGd>a3XX*B zkiFIcAJkc6t=3P@EHDhrW^BREKd&Jm@(9NN_y>Wjs%&L9lt6;P6QWqU3hg%#w90!5 z8AeY*VaXe4YhR9UjumseuE@ck$_)^3JOYw}+F`tv22Qb+gW!jI@M8F7#$Rm&bxdi% zJL8kE#ZVm=uX9ADHV;%SYJw-%e6e@JLL6xkhriRj@zx$K3d$MqOkT*c`Ie&3n`k0C zVkT+zHvw+JI6PAs0FbJJV--rV{p(G9a8-iOX~~0OkH3PIaXkKaO<<#&on!E1F}Te> zgfUY4(Wv<-+*$Dlre2K!%?JPC%UMDnTcaL#tjvIjO&{R=k(K0iTsb~mn1q!Ak0(mz zEqrd1!XpA}W1r0ij2g0Da3t=-;r64^h}XwvU?-aFjDxJq#TTzr&v8 zx;X5@MNzPV;9X4PaBR{l+{+s?+I>biy+;9#*=O;ma#r)VW-j2*MlS%LLziI7gg()b zb{)m&ZAHGR6TOd>?=} z&GFoo##rbq{SM;oyBX2gDp-(n0`!&i@$c(CruAVcluvCXwcG5NhmXIa+uaD9Y?%tr zc^UjF%O{NvKl7Z-$zdJ1YVs`jsAaY$jX*E&A5?0+@iSBt>P}QnT4=WeK)ZPP}S%NQ9FTDeEmRv;DpMVX+QiXZR zH?rV#D4NaNOLLAL#c!$wXg+-?R=a4SG4&%cQMPcnWFPc9_HsWnY6Lc96Le;$aox8c zkvY2_WBMjfT-BdVh8cIkkDb||c3%gB&ZzPq{FmYI3Kfz(P!3*Os+olDhoY*q4cJ@L z37f}HWyk~_{Lnidm2SHue?jnln8|b36W=olldHJK3kx_qB|i{9O=U#GPvXCG$Dn3r z2;4Qh4|x{bP;bL_&>d)oV#kAcvS0)RtNrD2H&$>hdY_oD1|}l;7yZQUtQP+Fb|!uw zHXG&|o8v@{33z;EBez)KT>N@p3wuYtV|JG1prJ-6myw}MVHCmr9@mJi{y4fY%oG2m z*TD*jt>AA&@TzSooZEQ}b9a>D-zFD+%@23F_k9MP`fel*n597tlP=M1?p;*hAdTTt z1#gk*D5}sFNyi2HQb$*H`m%dEogHLE!>3mX9a$j18ZL{MwmFd1okg_d?bWJJD>8_> zuuq&-dV(}96I1O!4E+&vgVeZX(wArR8ADH3?u+~)_$^;Zt42;_=TCKEqZuW7vr~eO zJhF=HdRj*h8xlI_+5?fTK`$|{`9UP3r_!rGBZ-vL2IAJjP!IEX(Y3y-)fe1%(nmT9 zcq-eBkvySCmpoicw{08FmXroi-?i_^o#d5t+o@`9qlCaL+^R#mr<|ZkR{8kz_cTzO zb(&T_c7cYCl&&|wM;}UTBzfI4Y3Y(WE^Fx}MtR~Q*m9rURt-Jk5gntc(Ur*w;P+@+g zA5T>t-z1jZ>GZyRJFPp^PL?ZQAZO2e(Wma=MBi75Ov-T~^H)`qF8v|wnJtM_SEYhJ zaCt$TQu9TlFK;2Em)6jfXK$F%@&@p&=LFd`D9K7%$+0elk7(rCCVF3%(xuz?Q~7a( zR_RHwXLiYxzsAX=&P1AAHW(zQe|giYx^tACTS;c@5!?_P)oJ>fK2AAP$RkCkf}y7~ zb+{sQ%e4i^_R2mIW8zNVR>?xyW^2}cl?7{GWJwpXZPe%I09|1t_+xLpCHaEy>*)nG zSn1YB<$i8sVV6I9$l8f+km89n6HVW?RB-NRG}z5q9_)}y7WBYVZR$am5byh@r2D}k zC{dn3R!Y2uowAnnplu^j&9tMxy(`3v`gg!hk5QN_bD2J~-%k&Ot*2YB6p)@ZGvO_> zgzB#fqyMf8zM%JKNF6IduUZ-sE3FFpeT64gK0cAoTDP6qTKGp`-8)j{mC`6NR}wd^ zdb>=W-2e6^ZfZmt@9` zNan(DXBy(>O18JwGB=73bAwB4sDwl&-F|B%dqCLzH#QX!N3ZduY#F691Ye56*8`$& zqlEA1&sZkAMj!R;=Yh2Kb&;;K4s?W0qQ!CJ=zxn9?fzm(4|}E2YMZO1edTJp)sDDzaKZ*ZiGR*6lQs31j#?8iPtiGeTv1^txIs%d0EU^ zdtLmM<7zi~FUJCI(z_uzzeo=M zTBI|>^6bcxFW;EIvG(}<+HhFkYsjRn*2HVNyh!8Y1Uz}pT`aFWfDbZ9!V32?;$`HG$G!6LNBJNwQ|-dV^YljJQVO`n-MMKUwfn%e}`AZB71{dO239d0NjOhVqpq75q?K_nOF0eQ zi5-T&`v=1oM54nYcYGx$fuTZYV%BC;PQy16)_-w8<&S}QJFEz8wkP4r8U^$YPUC(} z6P|I;IqvMowbCIpqgT&T z=)ZRe##}Im-ZgGua4-d(&hEye1@WLQY5<40w;zChyi=XW=XtOt-b%_!8PLaVz z*UNA?Ap@lQ!f}|u4&I$&#<>m7#0`1?gAe8KPUI^Hteyc!f=k5)Vj&AHu&%Rvh5XOM zGHk!M9PQ1Iq0J3N{`g}pe)5M*GA*f&_y$DNld?!l49;<>2MF3}?FWh6qhjAZg74BY z50j-m>5?7$nEu&?G;m}WSu4!XmF@+Tsu7*sylGQ|yO2BY6n-yTN1tEXPma4tuttMRsKak> zI@3*+mCRFS6;iUvpukMCZ|mY*pX(9E_9`{i|4S<43P}6fk!)kuIQGRhWqSM5H#)U| z7kWU_>~HI-^yu73YBjW_z_LlHe1lt?9*Iz=`p$}NEXxDwqr{1P*f0{REY;x(!M+g zEcKD&ohNoeY)BDv@fGF>BzOoHqLOghEKmGrPcx*U7CA1(H8M zjT={TALo3Z4K)h|C+j8|RG793Z_<5mD5wY~*&BlT^+2dj~kT^`K^72v=9> z;@!V~==M*R~kfh%x?1f2hO2|;w3ws7a8LEDcHtyIB7xv9W z!`Hjui^dcPvJl?KosURB+#ldb8SJTxfiP1Wydrl({NeLnY|^wt>*a0ah^x>uTAd&+ zIjs*GJDun<^_gVc4p%fAt3w7|^KkU+O0M+ZH@f?PA`E~2m(hHaOFVB3GJ#7A@zc4X zZ2p3NSoZ4>IXij;NtrR6{gl?j98}1o$rDc#ZSU(4b|FW^JgefCS6mmB&X=W*=^5h&dR_dC?7SCC(%&A3kb|S} zUW+D^Hh&DbxT>S-YhC7bm>L+{^@x1*84PZcpwCv1!-%cJ@c4>aI<6^>HWhZ0LnfcO z_cLa3t4w6++HBw!O&N+!a=zkr`Ey+G0|wr1oB&lL?69HB2agPiL)O+8lfR6j{q8Gh z#J5MB)Z$Q*_cn!2uN@Tan0^pA^A~oGy56Yp+nFg8OVLrn`N4fA3!`!jvFQ3;^6ZKc zm7JwZ$5|F~hdcLzl|(r6cT6I^Incyd54RGxa4)FMKKOsfgUmluD*UhM*sIQ=lY?8PEhZ+8lNb#p8n zm_C)tJwE^sbbeKfjwjQ$7#a5GIDy$S)R>kOJb;^J_@aoU6={9a~6b&PQ(b z7i$QFHSVcw`3&wJso1?p@^i;hUn?TuTz#b{l0DJ%W{v!=Y=t3Y`$z0Q>hN z?%!Gq1*>gITCp2yI~8+Pt9`hN%OkN*KLl!@#S)iEQM76C94fX5SSbc>{`PVp6O@wQ)TG$6(w*lyA6VudPC&n zbh_eZ4H(w{}V#v6*#F9PsxbExwvZ8fS%b?}ZK|9{B;U`%Uq3=nL%KB-{luYlwyYcVd5g z4b}0FB7;^>N%H0y;AEapFHHMLLn}@(<9x5v!(mZ$)^u0YRW+tXk9&zhcQh7t+meMD zeh?B=jHxTAd7f)1Qp(wi!H)6pVdVDeq@ek5e((eQZQhLX+pX!p@p*K4 zLqB<#W(+I*E)bcbskBfpgvQiYkyZyyoU^JORiCQTUy)|yxARn{K|h-eyj162Z6~zB zeh7ZevPTI^$m$41I<)F1@#2P)J%2}`sa6fS#aF`}84>+#)LyqWP0(69bT)UpET;vp|!Z`?mHi`3+w7{20sPioSW zqHwx1;-y_b>w?eswh0_ZV`}fCjr-RQL*>#QX2gsd7+=vqOrxHXTS5+Od1xf3)ggG~ zV|CH|(+jfs9cwrK_cyrMD2uJ3f_qKKkzL7_X1WgX%$Vm_$hLl7=uW8%`@>lJEo&i6 ze|m!)y{Si+mfs{3;>Xg_E@H?k9!0G}Tp+#gK2+!L2UjOsuu{50evCgTp7&Hx&-gsX zNmC}1!GQ0g7r$1*ru3>0^*Pzgxi{g z-lmHioiyn!v1u*gUQZdyj6HgmY{ViscqL(nOTeqTA z!W4=tTZjeoQ~WUO5*3+8(ap-Pn3=ShNZtse`4_aAOrJ#hMRXk|Ej)&|T@FL+FbBNW zzZn^GKl*HoDKjU3IGjHt%uOm^GrtYu2;Kgdkx4DLTXHJ~{FaTxl4U;`<0bYO^=>h} zFL3HBr@m#XYUlEbqxbQrJ{t0|E$_Kk<1FbZp&wbZAqqAu2;qV^?_>V;_zEn)dg!y_ zLE1C`rpEkW7A=j35!c7lrebRT2354T0JBfL0(QjK$T~2q*_&r=}_(lQJ9th7X;_0)3ME1eKr(_j}{ZJ+3Dm#av|`Gav*PF zJha89Ku5|&I^|O*%_z#Hvim>7HH}@^vsMAu+T4XYiHTHl^#L;6^`ZEqXf*Q!9l&_c zf3RBZC>HEi-~)}8;iKdP+HqQ*Ej(|=@yoxH#NUE%&3HS}m@5XCCC-HumIaj6Y1o_F*;(7MM8KtF?TbGee-ucPW^;$s?e98yc3Uib( z)!8stG6JXcJR=ir3`lRvV)WhMNblqnbGKIyqq#A9urPfs-Rby>nq@Af|7WgO#TL=7 zunL$NpT7OhQ6xZMs-87&?D1`C(;e90D%_h|hf4qxM2+^6_#wlR zmJNT3;hTJTTj5=P_PjBMHm!i0`XV~~lq%Kwa+zl9Nb>Cx=h1PO8#(bVl!PsgfDpL{ zB(7{9-d72yumAfO|!HX_7qH_?W@vB?}&5g8TSmNT|NUBC+rSQq^WG` zYIOZDo`zkJK*RDkT+xB2@N$X@Txc7@4`ivroAYzH+nO7pqf{Rg&Fbk~zfkJEHJ@2F zPyk`ew6JOQ174=S=*cSGp(BPB$p zG!u(5rlT>;Ay2}8K$x@~yR2g*C%@B`?a@di0p$;v?EAw1afGSh6V)K4hwi}3CHtYZ zS_UsHVp)>`Eq2U}%XD$&E1LgLnXOM6f^k(Qd}C}rZZ{t$ni$A((WWEO^1KYbxv31Z zx4O~>oiDUpTb~aJs>F15e;; z!veOmK3}w2dJU};>9Gqw*MLOzXu9jcXx#CkjO$)8NDiqBXE;q8cKE+o8d@lYov)0! z-;+Nw*;N|6p8pXrjeAWO)Ci76LsOLAEP;+wMYtyIEaV&_xcY4?j2!a|$J|Ln?>r0q zUA-ChkDEgC7isa0lTM(<`$W{Saf4}A7jW%8e`4u$UEuA zmWknZ-w0gMSA}as{P6tvT3qu&8e@IR?C!lD%6F(t!c`_(uq)9A#k*?3H_ip``VWN{ zqT9q=QbS;|t>6r&e1fPj3({QIz%6q<0r?stqAKRGD(oZ-Igx~px0J9}B!>oN*=Y1C z4xi5dUR_l62f%n7dV4nD@kjgEtK*l``_54`{iihBz9$$H>ssl{{9)F`m6{i($Ev$jzMuXHJ{Lh~T~C zSila3Gp$D57wd&*=ZtyPdr*kCfwILe=9<$!GF3kk&Q_kbTMZ9j*UG1`GVlV4(99J) z2zB&W<9h0|RUUttr(sxZGWtkuz{e-!q3(w_ow$8GGs(0MerI)pVoHe6hn43Qows6} zVI{q9F_t>SJs@KaY{xHlD@f;)W&(@nW67>z{FR`Gc(dayyie`nWL9uw+OjqDv%V|U zO|rs|87n!TvKZVFDUXt=9~n`L5s8`Tgo;s>7&FlZcqJ)MT9V!|HVvQP>+1wKaAu15?CGDdcCZ=# zFa=PsPssA?y5YJhQ|Wt$P#kFg%=}(=g!3_dD|($ii?APtQ8|YQnC(9nhCg^mp5HhM z>g7UL^`ts#s}DnswN)4$?FcW6)oG8f1U387MRMoQr^AwWlE(Z1q9ss&-L)+6trqK^9g(Vk!G+#G4Z1(a5(BH71V_Ih z)$MOc$0aj*G4u#I(R+(LFk{Hxh3+7hJdHYd70VywpmAIRy|X=^Y*>DpbUpim*WCeE zJs5=5ZJ(LA)-zb%au_Wv=0d@X!%)?1f)4(TiC|J{}&V$g>I28}I#i7}) z3@lZDhqDi~kacnkQRnVEE+f%~ACbeOrT8+I2A#*2A4)hgHMtC3Viy;A^f_%VL~Ts2sZjQqtB)1pdyZ#|* zoAik5Wp#2h(GVZn%3!ZnJ9(7jLErVQAZrguGt-4x{kf%SC^IDz4`~`vzol`Ey7zW0 zN;-sEHSttESDVHxsv;eOGsR2&lJJh2Hy(8?B8ty8liiUuB)V%MnYBiR^6#x^--@S1 za?BGl)9?~mX6ispLd;-Ns!A2gbMSfWA;5<5P;!p4%tnB#j1=Bg)CYmdb= zyx5g?DwM#DpE9VnYA5-qatRZT{6^kH0|&O-q1)6q0>@KCBhTpbhezB&kK|mC75Y|9 z1{El2D;b1R>NX(wlnR}F*a5!lqsy&W6s~+7YDTtW|`lfW2mO+3tt}rW6}<= zyJyxh&ALY{Q1&)%)qh_cexx1G@9D)XQ;)*W!G%Jni3u%wIGc_quRzUgO}1(ANjAe- zLm1pX9<@Kn!tuNtn7`AO;^%B*mv7oI$L9y}KxmF2SG@;ojAl}Z^g4IbESedVSL?PWi zt{zIi?;oL>K^$znwv)}y_MwS4mXJm92`GMZ9X_9aTpV2Kfy)Dng{)@)RkxQ=`g$e$ zx;~0?d~g?!aOLAe@}hp4w3rY zxj?A~GjK*f7i^6;C;uS_N#~jdt@^x@ElhT$&l}Sz*D`nNlY`+&Q=&}!G`#Fgmkkzc|gn-jrZ)4=t7*-MRp0iQg z1vkUC!_`Fx=^MnN@}gV9D4o}W|Go;G{d6saM6?y(_~27?K|BxHPE;r5jd#JN6>LrME@)O>i?1~VUH$NnLB&tWn%IUd7pZxl=ImmO$N zNtckOV8)fLdy4}_rjLuR3w_NpAk|4Xh4EqMQD)&rdKf;2hOK{u{{(Mp_6lJ?Hx+XI zzvZ*LhO0^acN?BJiWXvpGkEN^q3}XQ6WWw)So72i?2qMZObZ>#+t1g9FZ(R0_16$y zD?g6qe-0(h`66<^GeYR~{SK}`!tHfpkHBp@t5I$jCRupX+xgYFy~~A8@3tiCyT=9o z-VPzW1BG5*ZXov|1Al8s$mF4z7yN7?jH?)eB?eE?SZHA1-oLQfk)6SMChcPpk9AL?@5W0G$K3Vkf+fd0D65OeJyv^6{^j-3hQV)O;a|LKI0?FxLr(h|6; za8Td|8NkbZPoRoF0*f{Ufw#v3mJocDdIF+ZXV+EuZMjT%wM%s6+Mj2zIuHz{Hqag7 z#YJ7$pz@vB+~HyUAl<4?7@ICnr`H*hs)`~#DRd&AS5nfl>R{~^cK+&hNj&Wh9$`XRSAzbp0DHERCft(?60`x7b-rlofgF zC9r7ES8hz^E9Mw}5u3bTajkYD({FZvslv`scG+@s@sTN+R8mmJo+Mkb6H`Txw)p>9 zd-@~piaLWgW-CM9@v#(?n@5fr`)Ex-1ZlJZnTnaJh=1RdwnUj_@p3yOcRv-nDp8da$c4xeqS``X{W%?k5Rm$mo{vz|H_=p#m`<=V}oZ? z{?sK;$WCM_{ta3t)L(sV^Y_{c?sCUqC{b^-Df*>B3d_8tC&T_w>Ci6J^bJMhMLY4? z!L4MXG8yarE+I4DOznS^rM;dkWqM=HDNn0Fa^RB{mSS|Di#@%kZC8OC$kmJJl*Ji_h5-+qcW2htl4$FwQzBb9vEoI zN^57!*hT|XLrhh#$MARq9H?IrF4m(#)2I%XrUiJkJEVvoE5 z4gC#-iK-0uF*SPMb-q};!B~7((&i7J^x}8wMc_ug;WSJ3KGv5mhHPmY^l12Svgk<% z-Y#bIntrm)!%vc9Qwkeo6+{7M&vEX|TGVNpi{BJ3veeNM5UvNq7?nB9f2u9q$os|K z?mG<+wHP*B3S&R!AH-*?PGU};*a0~f2;=mEp?`igT>IxnCr3yGqj@i&@4ep8@-K-# zzBxu!k>dpYc4MhZb1qHts>bp{S?L73Wth3WABH&pf9I40+ivs%wK;zQYyx0_+f2H+ zQ8WOHg zxh-y#=4;EI9f&Oc+Hr~XDjE%E_xREGN%GVdITO21s)0|VE`%!2VMV)M=C{U=rj3tZ z;mYw1I9I&i^{*&o3*`ObRKOv&xV;DkLszg_b_0Gp&qM3T{kX+{KgtdWrp@tT)MtPd zX<7~^*Gcy=)7hA|UVKY$?Vr*>Lr-CU<8OBKiaxZsPlmh8hQaf4kx$SWLumJDt;PE z&s^@({K;Phn{Q`f`RCcNt9?Hd9Dgr(g&u)@YB^{(u?cTE9|5hu$z=2W3AOi8gtyn; z;LSxLucy38Pys~g4v*-HQzEF`@8EVC^wMvBb!C|cL+~+K$C!eto z4Y=c7qQv5j9faE&;DL*6T&}S%d|<Gz}pDtzpJx&AeR0dq!Yy~%0`+)gI zCqC?Jh&ZP-fSZacu=0clKG}a61|CX-kFIOjzJfi%d80KT`%#={1!VDxuh;Q1t1R$C z{Tobe^=1y$+1zWd+5E2Uf56==fP$;#QS;t4$bA|F!A)KuQ>emkldXi>;3BArxC2AX z6rg#V8uU+BV&ls8iI2sruuyysgyhQeGtWE*AMMK!5NT9p zw&DBr(}OK>&LER=3w$QNEmYs(1(w6S`TMhi;gz;4bo;dzovWFOKWj#aUE@38xX%Mh z^p(KbH43Kk2gz9M7!LcBiW(bsfLgpRAJp+1N>%$&oxeA(=x~DYm5mGYADQTL(R?PeeMs7N&;P-%qj8X)%Bpoz96G~IWdw-R;0^H60 z3u*>xyv@lbtPD|rs&R3+y>>kd``8QRW{5i$dLkUyeGmeEhp{QWZh+&)XV5TB0uQH! zLEX*2MIREP#aUkmd%qzAqN2Oum6;+=YMEKQ)3O=LGUijvZasGCsty$UiEQp(Gg;f& zO7=uXkH!bZz)yoN2+aS7kH*ZztHKG4+1rj6M6S~DJP$e*SHrDTilK`~j?&CuONDR7 z+u?IxS^n+sV>ne}NPl$A*ufXA>~i*0x+3OlL)TBoAMgHxbJQN#qIrz9s=efN+HbM0 z#(ubcK?khLb>RXT!hxoZXJ2@UH9> zE*`v>RuAZ81NiBrT7Q>1?q$%JoG(Ii)oa{yY7}~2$QJhg*Nj2(hiKo)>0+M9l9VrQ z5OzKN!KsRTngy}(;PgsB`wKhp)YuBa5}iFSaP)F`n*TNFSTVQhtvc#~0TWJM>goAHMQlI2hwf z|HQrb^Gl()(8iw5R|ayI#sF2_nJk2Z$UyNRZtyy!OQTqu_A`4u8H+7c&yZBSqIwEr97&RreqbE{d% zaaE;ar+nHfx1VO77K3y5#?zgxvMjW1kHm>rqA}cKK~03<1zfn#ewmn&eP3C6Uu{Zb zCOM&8d=)C+w1XEvw2;659JBwHur=oU*%FTfs5CE=OtN24v*3X9ZECo2-wx99yPBl= zxCEw#PeaR%j>VFp6WP|N_0s>W>u4NQ389M&=(5K|3K`jq=IJKVPwf+E(#t{c^z2`< z+-5C3c_)viuWX~EXG>@^6w<1yEV@>jLfd^y(Q@}7iYo>Tch(m+Z`lD6KV3mHdJO#L zx6#~$g)qV47!2}B#^qU7)HvH6gIxUS*~^*K^-hgn>^+hF9y*xGie1!C|CP|W>I!PD zFp{>V=HrhA2JGp<5xDf^ROX$(mu?IgN;4A>6((L_+S9IyxmjzhxO9sam8a6sFeKxQ z4m$5Ti=0Kr&_ng!aK_e~GR_T`M)>AZl6Y?4)KyV9_tgn%;+j$KcRZC0*p6q8h;H;X zVWhv+8BfO-;>WeU>EpKC@kyn!QukGf+&r@){53L--c)3e(?ds^J9!FwH#3d9`~5t+ zS4L9Oq|vmZ-&blrZ-(D>pX0Utf$W}A%(1j%S ze9S58dn1+x2)!xgPhY{*=?uuP-%Y)B3c-ZShm1>N_9AsH4c_Vnu|u+H-RONZc3~8o zT)Inm78OHs!#nAW$SN_HSxB_xf6j>_2Uvw@|5Kd(<}t$5 z9cYt~$(?w1kM;klFY;V$=xum7x9h=93OajBcrz#f6Y_0oquO1rccnXRb(~6jZKmMj z+o{y*)*s!~HLE3D7i!Bz(*gw1!NaMfLIH^e&6Fg-pB3hTN^$TLIakD7-#Aiv= zf|Jw>UgKP?dM2qTBlBPH=|?Ylsb|gzi0Wqzj)#s5iB?ZZ^YB~xn&%{uvHnR{eutBp zzc2o}J(&Ly-&(w?O`AOPBJt$-bhzc53`@4o#lKgl6g#WSqonR3jR|fS*E4p1bRcUN4ytg#U}X!|`YH&%j<%-zL-Z*@?g0*} z{=_~cm@%_xkD~MWw`tZJFLYfoWIh>7vmP{)UV#zp^X^A( zUiZ*6CskOrXc_c)=d-FaNrDi;!JCAaLeKE*;`D*$H1T~b>b#oAU3{QUgHDBT)obEd zK+bBaD8GjNW_5sVksP$!sqvd0^cJl*R-`<}4IWHL0Gp<(6yD|tw~Qj7(A*hjUzCM` zau3+Nig!#e=B03bLIk_4SHrX_SJAwmFNDjb5!54k23#(UqbJRKU|=+Xb+$Xx|KuwD zqnuBnt-na;*k?BS(|9HoulKoCyXnd`Mt9^QvB~NsE;>1vtypjqZ;V*LX*r2({Skd& z$C?L1?(}##*{MqQH}_DllSzW*_=oth>^M_(3uQxV3)t1UrL^sSEnXh|SojK3A#s@p z`)2YVe&3Wsb&X2AkNzASWw#59jvo^}@~>INim6bvAqyPULs?*G6}vd?2U}utg1q04 zrs0hav?O&p?sET0=7llDaYtCAshrf#*n&39{fJTTG^H~$+Nnh`mX7^8LSy&7q7kyb zu(#|X?K|EGSK9*cjI%#9O`Z(-Q^&$8(J!m|wjE&EbM$d)!0%57Q0nv}Y~=EB)c;l# zn0Ac?$@n647`c(rZ-`X~Ak+&*5#Oz~QBH9vyXF;E)mU5el22Xd>*x?#3lKIAB#K-0S$SX5>fTs$mx zRnLx~7>`VD!UZ5rBYE!V%4(MTWI0P%mkxT5!ttW!aUpp_BpuytMhib4CuM^s7FR0^ zm)fth`A&H@WwYgJ{*6<(E$5055ne;3S;^$Ss+9)4$)qIto9H*#k-`=Z6T-yZy!zOQ z;PL$&b4~BXdL{d^X2Xr_>rq#hoTNqztjCk>nM2ff-EJ<+&XgS}9SZOEN@!KqIf(D$ z#Ed<%X-vmuc=&iC6{g=~(~9Eo*41&O`}4H$t{|3G{+6R|R*M?V%JlKN7tF0P;rzLC zxH|uwuv7D*=tj6qI}^tWrB;Wjdq65i`kfF|97XTa;Cl4$Xh!?1DpK9;`Q)7bib_}C zLVZhyO~;Fho}XDG%&IyS8$4-KZhTF)$jU-w9AaTa99(qC#(r)1erzCGg zN0V#OFf*F9murA;zoSCGrCwa}Uh%zF)q%Xs99x?_fabJT|5@nDz8Y#L2!WEX`1$ z0QC_xV9Hd=(NU-PdsjPBZ*MXCvLdR`RLsDu|L|uEQp4#?vjQwWJ4;w}$pki^&@UE` zd5pL2UO`mBGxbCK&b`TGP+6;S^-?AH7<&2X~Ij5Y9}}h0Sr9cxFu&Hp{N3d82O8 z)1-socE*)8jt-$yj-K>7FrN0RPo!nX>(HvA9S1(LAjRNxI=V0kFI-%Vqjm)_>xHf? z{#Y4>h|ZE3Q*($t@q&oJrvt2M3v&r34$CHdpD_$F7 zfImi$XG^^^SaftJmUoWfs;*bzU3-G3XTG9lsTwRONaQwG-WB|BPmoSon?eKBp0YPv z-O2Z>JtmFaN#n|lNqgWL7QfyHT}+ioAxvPie~YY-CU4r?9z~;qC$jiq2kG+Y<+R+o zLfHR%8cq#2Vi%42(uV&#jHfIGnMu9a9^oLK&il)$XKO?L#yvF1{Q?_1Z~?8Z)q=*7 zgIL>f2F{wL;550P^F+Mju9%5JEx59Y)Qtgw?;5Kk0|zV(;_TB|Ci0Vx&@0CCPSQ=&G% zGI--^hLhPw)=y5IW-@15o){>3T^B?7R=J#$peGq=ycNAhoniXiVZrv>E?k&l&bhn^ z6c+TdB=4$7GL+v>1G|i%?TIB@-d&AtS8_Ouh5OlgV=XpwqzVSA+<+A;9|?})eD~B@L^53kg2IMXw6e2+Bnf?`ETFHnqEr`@3(pA#RROR}bvHYx@tS?u zX34jxoP&q-1d0M?QP*BgP`-2k=SiNhaMKqkTb2c?HHr{YrU%~FGdZI!dyv38n({FX z!rrRVanB+AkH=x)n|PHuto4RhpJVYzwlTcQ7|3LURQadxWl?#7m|HA+#Kva+rs-Da zs9&ccA6WGZ7I9TvXsiK0?4JVvfpdplR*~>~_kDWv&svBK*^ZxV2k{zk#AReh3_?@Ds8pkAbzv4W)aGlc88C6jE$$`KoEdLB_92_-#6sF6X`= zg-nh*+QVVmGZ&byUBQh0n*bs5R&dI+mOc!7$`(}WOF__v7rAGFg2ld)+(WS`Fokg& zU&6Gna{PrpS)$Kcpz-~uOXpZ?vwjPQ22qk1b^|yCIKmAhU^;GU_<3Y9jhs42+S_F` zALno%<8s_-(t#Bqdt8H`F~F7VCRO6*g1M-$I|4d;|AoAa1z_P~iNAzXf=co!Hhg(D z9L9D z@R(6NPJK{I-}@xO1AY`6WjTpEpArM}W2GM{W##F?^yXTPz}*(3CB{s44tm`o*Gd-MM$*ubniyTESk zH27gO9JO`a*qcK)K{7+0O*TId592u)uzNlq@9r#h+1879`D(}~d&o&w6*Yp!X*;k{ zzsHU@8S%-dqPfZcdc!L3-mEY$OIY#z8N?{xfx!pDSam;he)!$7e46W7Sn}dIJNNdA z;2g7<27gFrrR5KpN4qNYeSVlz{dIzSBijom9_UXdx)ywjQ4AUw#=^S<(doYB8-&e~ zvf_E^?2ub77xZ^8^N3Zz<_cmL#Y}tBoi(Hu9$57GxdhZ(j&LeYkqlm&i;inU*z{s0 z_(;BD%3B>A;^xTuWNoF=NpjMo9}m!^&$*yKeh8np){$5F|v|X6ukOB#a`CJV{E9=Y`*g21|{N!rAD`SV4PpB+c(uM!Q`t*mh__@rS>f?0(I9 z$=1GxtSQvASj8tC=O`;mPtP4e{JVQBHPw*&VD$!{_GC&NME}cz85*=9vX}I!W-U$H zZNP5#*@n?d9k|O?jkbRMz!nX?j}A{<$>W?N?Fke;S))0Osy)oT@YbZG2Wwd5Mgyj- zaGH!tmFfKCY1qf+8gnXXVe9{UL&MVF(!tG($za$yuI-jRg;|Y9qpvg3=Tsf*{mPgu z`u`^+4e=q@5o=lVkG-B%-X2C>*ovBrC{xDZHzgi34kEgwmRNR-v*2I-Ndq8I`95yVx5=S5LPW+-Wt0 zksJ1~w@0IJ7?({`U-zS>9(i2;K5<`FT7h9gC`u;op|aUp#qJB|3*XmB1(UBH_{QFv z+XV^CBu9rz)^-#Z_wK~8|F)r>?{2p0Kp0;670Si0=mQ?L3QWgj0P(V?as71#YEgD# z>0hkS@$NP5#e}OxM>kXn9<+nQ6XsOSU z!}~Kn3mf9)X~ZKP%5Ui}y*OA&dT6gUZMx|ooqBYbv?IiqKr)a@*EV3u<+b>8X=-7{ zRwuF^Eumh&+?lQU3l?q}&gM8DMw^Yp#vff6!Y+$!>FN8#S+CnSHu<1E8?@DvhK1Ur zo2nGo@)E%|KUmbd&P{X0uO0Gis!2P$T`-rev#At* zOxVCeQk}p`PK!+#5kOzJOu+kJ=Ll*UdP2+qb2cFSo=tv*4=3Y1irxPkEAcHY7N&01 z!f6X9k{h2&MpXga)A7zY=;~IME~_pbx@wx#>EkTv-4E?7bIWS>G*An+POjlr@0OKz z-B>`8Oyo@VzeeG2bft2tgG4T`JDnI?ht;pt;LjQnhb#LAW)AsW>^1u`+dt{4Wc3bn z$ZVMdkp~*7%ekE0O^W4KX{{2o9uY!Aiyk?(9^sT4CJF6IG1&bq5WDTp6;Dah;(t2U z!?<_T!T0JT)ZOpGAVZ1QE*^#5N-H_VstPV}+7rotw>_wP#1=_&qXzVQJsj(=oM(Y@ z2T8Bql=d%Q%4Ny6pr-O+v|S{+tQQ;xqv&nWEPciXc*-;DiL=P!tqVBSUFK9)cCeA3 z#0=8h5UQDO&Fo$)f^5YXVdumK?trW+t@gL4E=|!LJ1>#zH);%3YU>IWuO)QC?pBf8 zIA@R>)CXiw9>dp<`(fGjW{miJ9!q@ov-$Is`J&o;U_56rd^i}4o6McL4G*W_z}Xjy zSG-K2d3p6zzdRKdOnk-`t#GCpX?8SfNH3V?90@AkrEKk@`|Ngn0jpA|qtLS>qkNVPIb$h?VXEmlG*)*6$w{#2;n3v*y5`6aGwFe=5tmc9N?r8p`HZ zaj?DDLKvL9h6OsQaMOl}&V(_k6m;<$+oZA^8k>UQ%lHv+yYM`Oev5}EgD*qn;$HCc z{Xwi$D`jc>ecTyHpvf zaobPEqihL=r-EA47}{yHi|Sjp7qz{2U^~-};Tm-<$lmJ-Ygfz^Im<6m!TT_2S7~wg z8$SqU6*AN)9HOwL>B8oYZmhP9#u0Pg3YlMibCc2?@MOw5__Nairb|9B*$6q#w4WDv zNoRtguWHdImH9$fwhX-=JP_}M8;c&#>5w&i3q(I~hg02uSg%uB1XVKNx+)J3?L=6W z6al#w4Z?xqRH#=Bq_yrUQioYCuqFQ)H%_fTb$NefR~{#WlIu*+e;SAGOAn%slMj7% z5JZ-8G(1c&fX7;XEJovyP1k}RroXKOZI#Su`{NYy`8o~H|4d}N>Sw?+R!C}wj@19n zWNQBU0OzUO(TV-vkf{x0Tc-GNT)#0k$F%+N!lqk0#GOTr zJ;GM`_eU44CHO-&9#6jR%l5kb=B(OJ7WS=QB)FLE5)AT+S+Vg;ZdZskScn{fC*RXq zhv60SI{%$L6lcbhr+mSeXVWn->~|}$i^FyJ zTRHWdtkw;j|1FW6lB#h+y{;gb{1x6hyprs4I|=#aR(#{fFy7$yVj3N!MCVIR$+?^+VrS+ooK(qZD>dW zR-k;0JUw@!vdbAs>a<~~yLK1b5FQD&GnL4GRu{|p+{ILvDpH1JIT~Dz!Ohhd#H{!c zI??kKw#r|?smuOj3e$gaW(`l+mDNiy@2v}JjZR{EfmP^KpG;p=Wd+mdER5VUgiIq0 z$z-_+x4$Nql^1qnx!ld7vUO`=hf)()VE+?8E}X%hKZ>SwgCaI^>@DGx$hX<4G@13Z zWpbC&n(*HtOVFzd;z#aEp}o_5=|e*;0)Q&ckW#QkDXwLk`PIwwMhGV}kiN4pJwDRRRkzLvezC*`CoX%9fxK@|1 z)lb2kP_ZO%_b3azw4W@m%%%O^Tj+dY7Z{%150hLE!Lr2pk|2XYMYj&_Wn(-pVYB8o zn6zRhUu;x@lY0iE*QHb3q5e(Wy*^@pyLk|I*}DM$z3}F>_)LD-%D3FJO#?*Vv?_n^ zOF!7UrHSPmsPhL$?kLV{Mf5K-<|pjjLrZKf!HfV4UNvVEsDGZoo7V^NA@RdSSM3c9 z9~OzR2VFsJVqbpa-mUziWp~+&+Ou%o{tbkkOklU??-wStXv0sVSmu25372ZUfcbh5 z@}_B$^6qu$zV8_MYb@l??GEOrn|BJzcfLVz_XK|4oLsinKoh#_4zW?MdrQ+ve1Dx0 z&Rd%Qfy0SGurWwrr}ULWjDNBTghtwJ!kG)o^c6VKeL1* z9{4X{3GO>>0g8Rv+1pXi*eA~v=J6&Rd_8Z##<%el`Rp+lv7);;JT03x-)&$?$NNC6 zM;7x`+mFthw4mkS2V8Y|nz-}mjh7z)IJ_GH-8&BB+;ovIw`Vmu2Ct%PxpAo7IS>bV zTB7{)GjKC<1b=eLJpSvGEN;KQ7SmhR#Qen>e6Nm+s5LH|wZ#mA+o6N7C9FuWPV5c4 zM&5x1v_KX@Q+PR}F9er0LwP?LQa<+;ZG*Ps(^v-Xp*cZ-%;?@x6Dm8^uOVi-j%xm=8q=t+FWrimTKvoN790O$0_9AImY$iVEX}N(Ps)D7Da5|NbnVzplHREs$)x81ciF35xbrE1 zy*E{6X>rS-TkNYl&&rYdwN7Cj%4$$2ead2&2BTEiMLUe+@%i#!OzBn*8p~Xw$+!fi z?*m1NUV%_E^gg#?f-l)F`HPyJVbEuDDy*H80c(9`1J;d{va{LJ^%ml8m{X^)kBcz$ zjU4ZH<4SRZ?sqm|vOYij=xsE+D=&RB)B|F@cY~u%6isOx&OKm#F;aYw;^njX&I1|wEy!NaGD z6u!cgTff#3uGg1?<;g_&v$v8GQ;ManyFprGc3f&U?k9O@%M0O_`%&%n5lkE14D;9g zVz<2?vy8TA(hUs5P~#KDA%psJE@{1K;}fD*H5aP+JWD$0V2U*7>JBdHt1HL{|@zC zoG0DfwTnK4Z-4DTR8T<&K&=<<2-h+oJc{1%PDHiZMxa{oOV5!FMK@ouBdF!9m$H|KQzZBL)t%l zi}bqLG2!pbfl{TVy`@*)A1Q9zXGEYrwiCk(_Rd60161gt9O9(af+7q|x(*@|QV^S)U9vO7KE` zw@lLR_mifiY^PEC)WkP*PkQNZN3nOU@N!-vo6}m(Rx4UkM1~`IYu&{=d@77o{Dh0Y zXQApn9s01=i{w=g*-RdJjCCFUBawbDWZvKP@XMilob~n)9ARQjD&gwXJ8p|G;Fb@U z*S!Rfr`Xf&iV(mNGIY7O2Hw>=3Us8Boo_x{93A9MqdppvanN4D@mV8xbiNn9yevM-Qvo>C5zbok3vq20+qkpfT|5~csHhoqk9=RYm*Cp>Fq?$pA4xo z+Jqb|;+biDF4;~U0qSka5IOQA2Dg@@-@qbNSsa6=mgZ<;I>?fGQ;wXqy=ga@7^FizKVBe7pSTfQc)GRJSnRzI$cr=(#o~uaf z%-#sAEObce(jRs-Pm%xYqQfV6O=l(V*8^6@vkh5`*`sj@xZtokq+FSX8-A>XvQBmQ z=5dFLJnzx+dr#PP-#E!+pK|Wti!SD1V+|2H*FkrqDvw`kxn_k(toS^PKlY)PTkF=B z8n>M0R91B|_mfj#Iu7NRapzMjzo1LtD)SpO>9H@8 z-*njgWC`CGJcnE5cLbRJ3z#Z08CK`tgXd3-_$Y@{V3TqfqLvKdt+jpmb)_%(Ki@v` zYCr8*?aV_^=bX&+kGzJ3+MD@gRcHRJtOm?(H|1l$h4cS%x>3%Pr}F)YLW!7T9i5QP zZY?)})fcld)=myiNcEwY&v9rRs}5H;Ux5dSUtor+96yp1v#Y5-T;t!*IK9ty(3_Ue zbZ^-5MT@5LJ~2i3y)Fe6Kct}0dU*(<`sFFbsV90fDC^`-@L1901hD(K&)hwB@8RyH~V0-ovdO^#Ri z@}z})XJii447vk*&;N#mpm=8SDH`N=Jiw&)Ls+uW6Uo8=MLK-JlHR!{kbhtzXPg)# z!T0yDW=}O3#Fl~Gx?a3V(KFB+c$(@>_u!RY8(~zQH3jcF4e#8J!mo=}6nCPYF3gi) zonbNeU@i~V$idlztDyCMCM*#%9B2IRvqhET@$-ok{RZ7_1k4_IZBzzw+?2}SRh^0mkA!;h+0V1MHu92Xg! z_twXw`Q746R4MxV$;HeWi0egC$Fv_#g` zsh(GOU-=#$)r^BDT|3!i#Yf;ee=iJDiDf!H2~d<3xx)9EjS%}U4wPH+xqGv_SuaJ=%iE9V{i6`75*66BiV5WWzM3X( ze+5r=b_&O`H5hYnX1z5gFv-STaG0aPA1-#HCra8-EIkco4g1-w`6IyLVHhhbdQ|K& z=o3rGSr2y-kW{mEc#}`5yl^fKtu40Gj@%?Hh;ako-qWGma6Q}c?J0ha4`OTg90vcG zD3+pL!%Soz!TGDBF?Cle+<)x^4Sqp%TDq4m7tALU^YN7JHjXdzInG<0G8S{wp{#n& zYIt(OlNR6aPt|W0(CMgk@)%$t1_7H8DL+FO%g-Gd}tHE#!7)*&VRbfiUk zmF(}CWsto50rnAhgBIeswe7YIuMwHZhsGa=0Uyoy%EwmxloUM*09mGI(HkaC72|x4 zD$+)`ATocPJYGX~78!Y6Lj7%hF}lf>Qx!HrY_%SrdNYV``LL6JGHE>+rQF4jh09TM zY!YYr>=u_hLzO=#8_m~Pr}IBvGycezYi0F=2Aj`2Q*tvz5B&uFU+$OatXn1Z8AMKOI z7ijnAgH>uE`s`&sJ=T?f_W1_B=`GJrbh^U3HyIeAy_}~#rF`ziFxDDr08M+Rvq4Kr zDa2crpHjD%-{06Nd0gmE^*eh}-e6v^8RUZ7+v`|lj|D{PzGREimcZSe*Mwa6+pOW< zQReb#6?vVgWmnAbXrHRUe=>Jhy)izL8}YlI=^U$K2RZMbEPqrkTB4LtH+ z9fpJ=bJ*fpyl}f3>D^2b?k0(|^0D%`{+JK;I(<{va&H)R%el1-;=?c&tztb{wicdZM|94R?6tIyxV3M-tgcSoiq=yne7! zn7L&g`=qY{+c%d%`Tk|B^GhN8%kG6||Bi-rEoP9l(;k)P7+^=V0aIF&$xIeJVKM4* zT+QdntnsGky|y)jL#;qXgMP6ip$<^BNX*OG$x+pt6z2UhS5UoZ!fvcM$4>g@qt31i z?CP!*8T#XgZ&k}uCYHf`W1pG?DL1&5K%3deZUb+pGvL#p20Uz&B> zi~b71%x!E2GZ1rlBUdhj2a;pZCnc6rw%n$+*Mre*tUW!@If^RZv@xhImMjAMP_%9Y zt~OXnXG>yed0#smoM}yCkF_D^^%CdyUPwn>-=R#U5#_kdNBLfUbS&;6$(oF2*90x{ zkTIfOSLJA#{TEhuzZci#HHvPYpGA_ZzI5rK4>pV|z`;Wu3hVcJp_v!qmYp-WH-{FX z#<6>{&}X zF}W{DF7>5WOrXvB7r4B_?`)vc9JHOQPTLdWsW!!u_!wK5Hme?EF14cSn0OkW@Rj-c z&Shtg_N3~=3jmhS6SlF6rK^eQ+N^R$klsvTpFzvHDEAUwvSzx)ckqr!9$XY923VYMm()58gFfY&udp%U8(PtHLWy1*;r)k4} zRMgV?z!CJSTUqLqX9>#``(n020m|D<0dKWBHmbA^56Xd>s?AUIy`Y(}Y zn2r#4L^{l2fyhUB8%3VR38_XBRvJ=*hPUz=($-E&DlLhmrTe_EgQPM-i%LWp8HHqz z@IAl3e?1=k-QB&;b*}g8^?Zi>mffKvIYZLgM(9;qs z-FC<_Ok5AWOTzdW#-Gvuu!PL!Io{I_jHYh?T50~SCyZUQE)I&Cq2<}Gek%*KV~+4yF-lU9v?&h)3LSdMsPk(N)+c%@?}ajJ-hS*!2B zaq~QEoTq`qRq^Domm64{XOqq&AIZ+yy|nOm4pA!FL)|U1tR_iD(!lRp!ULlu zu~q)Du(Nv)SdDL^kBnx4NsL>~O|ML%H0Lhaa(X%F_dh2w*(H$Ku8!YtyJB1ZAG&J7 z81m3U0`JF3K*+?YSYX@%6YQe^r)Cj{G)c(o(!*WLGe}cq0LeXi2R}7_VN)NDWsOfy z#{1JpF=a}dV8#ZX2UKhTD@PHq3Uzv4s~Gy>Lm6G^bOet6oI-~b!$~2z zCCpxsN!FTvAi+cC{B_S5C%BkW8#~@tt$B~?c^xAGsg>X}S%8_13}lwbv%h~Ulf$!Z zK+E#=W)}VS?do+ zaQ-h7LHxE~P)rQr*A8>If7OKa%e`j^YsA#;8zHw1P3UZ%&2*B_XR5x{h6mpwNkUx( zImrbOsW$^;`Ko^;qAY`Se0%_JfGgHHHamB6v4K z9kKALB%4ZJurhhiiTiLeEL9Jo#n~Teyw7{a+4Ug)9JqjO+3(3|r}ywv>>B7Q^pcUe zUF@im$*fzcA9LjUEHLlg4oNSbk;5yp>D!ajAlT*|IA|I$GX09+IqV5(`m4y-$Gm%_ z>>&8P>Jhe``^4znO(p-fKP2VN&X8s{2F|G{Hus@++?MEe*}WR%M1P*=QG<>*U^U1WAxn7vt+^IH^K%DV_5ao zjr?l6M~42aCR=Rsh2vfeXi{$ryS=&$UI_=;Q9H-bJ1x67c6i5$#d&0-4TbWK;c3GQ(RDB#-$L_s|n`Vsiy_NPGw5 ziId=dfHkz_oTQ`wo?#MNQ^2NA11@>B&^JO)_UV`@^thG{kgJN2ZBS0~!}a4E+Lzq&Smy)xXi># z{DKt=y+LuTnpIhT0U1~|jdcI#LMmF+V8cfnSUPf;cDsi{+JP3bkIzPqoudp}M6)5- z;{-KYQcxqkgLkovji(EEMtSBpd4XH{T{cPWB8=JB0q=$F%-0!uL^CLtT>m;9-SF+byEog3LEUTn8p0RD;4$Ds! zLqO#;X2UT*sMJaW5gGnFTV)!$E;D5wecORb)3x!q95A@`I+5$R&x&oZW91%72)nOs z#LpF9sBVNB%m*EImYW81&c2^2l}S-yBG7MuRrs+XsTP?hcQUl3TbhJ`5U3cC9I>nZNFtZPH6q?($YyiVr|oJh3e8_PD{rF>Af z1Wumm#*O^^E_g{7R=Uv)0K+?0FcC4oDK`e?7l1}fGt zpN6{EQPTq#;J0jAAPZ8K@ zEeZ?L3TeyDVpe1S06U@^0*)t*VAr)H#BR$@@;TBU+Pkgkg260!^7kP4zv6qwuCK_^ z-d}VkDW&e!XL0%YFyY{$RXEO8Mlk>KVIo#K3;S*_!okY1n4zFY`_`KQ(QhY-oso>h z)gkmt*hyv_;+geYPAL0hH8EGXKq5bmhp0DFc+=So1TDEVSNjmTUnUOwvW$hnf%$Zu z+z)biw?154y%S%Qcd+c|Xp|dCWFJq6B4qX|lHWK4&HD0!ou8kubsG-SKEZVM%`*?& z^edjd%9jzxq}-$FQ5sk_IR|oT8Tvj$p9xktNrS87aAfB&#?IiM%LPs3Qf;zu#C{Rp zw=~06m3y>H#*1j&C}X2%D&gIYKggF81@P&b9&7Mp8hv)>DO2t%i??0BFw^SA&}lf3 zd}@-0qdOBwb$Tcs5#ND!hQoNJ{ty~#b2Q{aBCYb5!~WeI8CEZbOE#0pdzT$x^D2VG zKe)?eKRrZkq$#v(>w(FvFj8|zgp;z?pgZQrQKrieOdDrH;T~xaHQq)g(wwP#-8J^( z1Zl473-1mKNEE(ZA5QKTTq0YGCV+O#D5ih9CAfr$!`+ZOY+GX(}>5092@B^#9pokdINipl#}^C$UiwDC1Q``=H#rS}TsK8Qi~-IFBsBk*jw zJktF~9=e6~c*AuZ7wy3zt{uVl8`to@7@tX2*vUNXX%arqJxh8HHWOZsNX|tXka-dL zknX<_r1mAiTj}RCdtf{%3AoN)i^-+ekH5EyI>)<>T|DSdlk0q^ItH%$rodzLhAPo~ z@}GJ%T9&2qjFLpMChs#ANF<_MfgXN|8fI+QcG3%#vmh|63MWZ^#?bRMm}X32jd43j zriyV+TNZNvjSuDSzy65h-fH6n#||d9QlGP(8&6kF&!jSuX_Pb2M~NG0czm}2N{aeO z$?s5H`O<;=x)(qqkb@)@ZCD&83g1L#!h)0=psBP2)EZS zg(EXX`Xu_6e5ImW`8~-uZRmBsgry#lm>XVzIpZS0!ne}$n3f18 z=A3|kQ;mrqKab46zn8Z9dVt+`3#i(@5qE!BNG;`OgCJl8#x0)-e}Bqy>(c7EimX_; zNW}#E(v}KjPLwmOuL3^t{Eh|U?{M}OCt+9VL#jS4ki7Rzg_>($ST|JxwK#R>(&L1A zIC}3e{o%in)7g>2S-dxe-6~&*uR<=2_xVar+fBhw8w@z-sp_1=u6VB6{RG~xyG2%f zsi7xs%)Ip1ArA}#KEPGAa?n_k3J&{jqpngf7iIJc6V4swGR5Cf0W_)#E(S@Fka1Py)O87Rbi6jikClYU zYkT2{))FR7=*j!&i!t=`Vh&#AaiNnXx%UH?;dVk6Cm)u_jbGm+bf2O^G=5dkXE{H} znlZs#Xl@IqUhtRv^xtL`!0!c*$#uZi*Vo`ck^?v1HI@{+++)YCya`>p%V-5jW$!pD z(QRwD!f4$FxR4tPcNCE%#2SFRNhT!BZ-7U0dSUWL7B*kE1KBq+%!M3#U|;7$&WC^a zgVp0)m&B1>Q(rUR791hVL>9pEoGtJteJd?lRfyI<$8#%6gm|+12Te*JAhQ~}NccnE z^V#T2_qmBKu+OEB0r8L^);1-Rp;^d1wIdOL}&SB{% zu5acwF6&kWt0H}hDvcST6*tC%(&v*TeakIORn0}2O?f!sSvc`E3WW7f zO$48wnbSKzmFcw;eAb{|gFkOS11VzfNL;BEM)Fx2>5^icdPoA!y%_`kcT;f7c}Jq_ z!@`X|KgbFj&39aLNJoV$tlOJLm+-KzLLW&kqyG#{ynhr5u3d%sFGg_SeHPce^)DB0 zuZKr3UZSQ>PjUX7M`TM>Jbmk(jdQCV*taQKyhC|0#=o(kz3XIgWOXQhIeQw?pH_ig zy#!2~Ax}0M9izry0{J}oQQT@g4f~JRF=`g;V7ccnvi=@F^9ss@#hsdT!;25Bq4gYG zFE574*Vp2?jLCvNq>|FQy#?9Z9aLd&-S94^RqhX!K%H~ zW3mhCl)j}!g@N=g-|ahJp#@iFB!aooZZ_&}8eR9q+saC1A}qC^$?R?5KPNM?X`Pi8 zM(&&zC-;YMCshun_dAtft&Qw8@n>{xElB1sIOPPfPal$xS-;36!&NXa6c4M8jTXEq_)7kk37DAQ z0ib**jCL%D;^*>svhltK+0&f}&0_~i+o~8$otX_=_vlh7g8(WKZ%1A@=-`iZ4y*Mo znLJ?%6sGKgo>MB29+*b@=JS30i>c&4{{8(>WH+9FA|;rf8bcCdzR(6eFIvGQVWG@0 zDXLo!WSK3~xLF$0&t%X5eQ_K+{yVA$&j9oG0}#1hNLQ(q(icNMIA3!tWd0mthYll& zw=blc&kn%#R!4YMFas_7Bk`<5E{?bNMEd~U0M&%F5s`bhg9H$LMUAwl?@u}6uB{vBDo>QMjH1f%Q7N$9zHdE1FA~3 zka#H=Z?1GDU(CLe@n6cJ!7Ekxyv>2mON(KDjbuW$t%|@}C5A+c#A4et9V$0kNJj4N z0RNU=5HAn|E$JMzDFu{y8$q@j5tNWx%!#hpi31Z%@Tu!e=$74p3rw$3`=x(TB7mPc z=d{r5UC-$Y*PrBt=|!3`J%ZlXP9W3kFFl=Z!Rk2yj z`nRj0f<-%xSP_qoFV^9tFSeUe!}i%wGEf7CC+1lxwg{QoUc+REl_-cw7m=3rVmRA_cLFrJ;Z&VE zD>@^9Sh?F_RrhzOdpiR?84auYC;sd>Up_M#Q^icANuG1Q)^X_AN5_LxW6H@2=StA_)gPBGJXc5d-F(wBOj>#6$RA@btvD3F-3 zpXh{NBsYXPL|Cy5V-sazh1fn=)Hg)7H9utC4)oH+-G_+&OmlMTcP1sQQuNp$NL^S;c|WB^oA=`Wd#Bhh z<6Ch5;%sWRVo(@ysGOO)CYHVV1+Z^oIp(}RN&TMnvQwPHaG$~!dMjU&gk@Y|R!by7 z®_@M=A6(eM8(jB%V1T0bUC?+o)MAN9tsE;4Xb)LDtML8SCWXDqd%5l7#bpj?< zhLgv6Myy<02pabDj-BnYpqT6p?)Sxkoo@*S?+jq%m#t-2*aMokb{|cfX+f>{*IM>d z1i16}Ie+wv$nOEZ_rT|cvbT*v@2}Ho`tf=oBYAuVWf2|D5+yT_wUYD;5uo^>k?5uM zkd9GZR7UKgP`|~LIxo&4W}a(FjQRw+;HDdN%IcE>iv-Jaj1>JHFcDNwe_@=OcEep; z4`L&^NVs~31H>6P@qPVu_;2k|I%Yut-F8h1^+#Q#+1r7Y@Vd&Fms*3(8Un+|ZZf+J zBcXBWFctBYLtB$vDr$QiQeHMg$D>fIt#|6FxcqLKBC!HL@s1rA4+E4wF`0Wi^o}hK zj>OlcEKWJ{j|44nrM_zw;A_3f4fvjE`Sk7^Vzfe;+*=gFXSk2D?AiPD(HR9yyL^hooLkB>ND^`Oi5Y^e8|Di1 zRP+S9`1x?xcqw4*he$~MN_bo{g?KuwBIBBxpvE=^?5<2D@QVMwS-J+LtLHI2AA4a* z!wndpbOAQ*sS=8JC_-r7EqY_*68x?|B3Lff3wDRs5Q(qyFf#TijQjJIP1&4AcTYS{ z5+;fXJl>SSws8f_hbd2Js_0m(b03Awi>=i4%M>fMD-!V8Ne^~~KVn6WcG3Z!fgX1^ zi>BUofEOzxi1k>OXbk9)HK*@@Cw~vwcW98TeP~C&D@>P2Xx%Wf?`%Y zPK@h=MFWEnlq)LeEShg+I(Y#WIK&ah?hoXK_#8N+Ist`)C&*#5V#rhVr2Y5$pyi4X zq9y<;&Xa-IeWT#(L=p1w@Oe1%XaFXf>j*|PK7)aD19&XtySB}7xM7_FIoRONNZb;V zh5mVT%J+0M8eRlbBpqPuu}vVyJCgoPP{&s@x~O_rmY^vxU6Ao=jXEWg+mMdJ*!cp9E$%pUFFs z09!uRLMPwJRC9ENL0vy0Q!_+5x3Pkcb2bXbtD6ao`@~3mR4rR@c`sYiJx`D&RtWIV z22S=}0+m%u1@i-z3hK+v===RTyd$H5*o(EJ)PZe0=co+!Yq;?9Wty<#vMa`~krBA? zpM&D@{!}yg3h1og1L99osO~Yo3;IY7ZU=h7!3UoKg-2lO{L6H#|9kqoTmdGHIZu{M zvJuE=h6p8jdgc%LaPJrS zA~sBuV=L&Je0}iX8PHO-VUYfC23dA6i*ELl6}*|KE=V3Q7R;U}O7tH1lB#_($d3iy z@R@7|n=Wf6pj`usA3ua&=MM-*1dB-9O%+J$Wa-Ar3DAF85=z9E@%a#a!DoLdfkBoH z{JAUzS|;&CD?bjVZT@MM7|4=HCJCh8n!y$RdUdalpx$vVoPA4) z^jJCaYAWv-o6}wP z@0zc~?_E9^dKy5xMF#W7bT-tDvxHGmnRs=!J*`&Jz#x4Qm|@@q*ZW0b{jn;V5-S1Q z!bU(4wi^<>s$jE83FMqJgqJ&Opb!Fa`-5AIG@om-*9apWFTS($4xXpu>`N%MK1BAm z`$B%$O2>6I^s8Bud(m^51;J754+D7`p>~?x z^QcwG9SR|q{Mnbpi~_6Bz?PWCQb1iq9&vN zk>;WvHh-Bb>!h!Q9==CW-*6{cF*Sw0nwU%s<+SJ*1zC8jwgn~xY$Wn2JDFR*qS&20 z6X#j(FDB0J8DYK0kR9tbpkJd1&a$s%LiVJvt9J&oGUbzTgV=7YgTI-fb6Zk!uJpLt1CkD{;YFn(3 zT8^=Nc7&6-L%q(X(mj)3Qf_t=Z7o=cBmDWc!P8OD)G9;z^!Bp%2TQE_dM2TS?In8Z zM>-s--^En*XOXQ+L*RFDkWSI^#jKAaAezIUe~u|-j(wX!u0Qj}W+xF8O9+5n|K(z6 z-(k?SDTJ9!9XanT&b-|79~lnuqT%!RGnwNpp<$vvR7_k0Vr%*P@O6iHj^aYP%I-YX zR_FV)>f^{mDc)CdqnS>(t*7}dVf0b{4A!^PlJwb6fU%{j*fntsW8LJ0p7m_9 zdfeAewX&M=##|x3o$3LCs9Q9-To%XOSH&+XBlOA2PI|jyChOui3s={u(iY`Ta5)=n=PZVjbV-NBET2Jp@)*}H%$4T0C z3wZqKA%6a(gP+oKm^Y=PQLWG#E)-9L+DkJ~&Ap%Q^KPX3-$l~*KX{({YAa$dbC+!X zB1KkzUkpKBb|l^3jt%sB&D`o*iNyjshmgJ?wkUf;al|YlBd+eR}9_Kq){b)6vWKjK@J7oB%N_{QDVtMI=Ox& z$i{Nq5;hOD2*h1&vI-ybOGk*^ciooTT3lO4JzK_*BD7;x6y?y%z5MgzG)xWr9?&^yNsPYX0Ik0? z1s#qc9Xec0?=QMwxleH>#Kg|S_jg9{zvu?E82yP%6q|!`c}M8aCtb9a=ig~*=)%I{ zU+^dAEZqN~OMK$S(USQ}_&qaY{p=Fd7SQh39cvikQ3ug(Q1Vhj&qv@pQ4(`iUTr` z);AfRZ+t~{qL*UA*q79AK!7ppN8|K^eqg*^$-aPDeGt4DBCm zr+5XL4NGttfnuC>VH`-bt%gal(m;A%LS?oyyIGXar&bhVPyJln5u%UM?Mis!cpUy6 z>ZB&SzcA_bXULUS39v0m1=DTY;iU^hc6z);?b2|3aK#FlvIq3>$H`=lNgQsK(IL$S z5}>+ZJ-|ynbeYfd*JPd$S^szJ*MGe8ZomeVkIrEi9FHVh@+I-K%O$E+oXuZzD~OVR zC|r4NML*?-!SUbs*zJXysO6^%hFeV_^2&d(e#Z^894&#zQk2m#F$<^Lor5RI7U()| z0a3`zCP^V4xV8Q|uAgy%w)gC({xxq%R%s=T{zZ6~-C-QP*&05&Zia4yRNur`9YK^zj|U=#A|#cSj7pcX1c$norPsR z-Hvm6PGOC3RI{uq03?*VX3a>TCmCh8q>oSNl&5z_^`NPc=LNN6dOwcWe$r`-nTy7fDH^6+db zs+SKdbCT(4yG8Ww19IovV8G-Fb53+I*(H6@YW1FVbXUX? zDy8N``>tIgzt_Hj59=?3yTb<%eu;q~1Iosonoe$=^(9}9j}h3e(ZY(%ZxkCk@v!GO ze5JSvi(TH34Bm3-QwF{-ILSsG>c)@?8Mqk~M}MET##Nu6LjL!?{0{LK zn-?lW?;3t3o#%qdFEIu3XsIl@ESk>BWcCxP$KN~ZzD1PJpl?;o=?3Y1x<|H#50wT( zxw0;%Xz=sa?FsO9%v$=Q$dPoHse!kwusW;B?P^1m5m{wsB!q2x%;c@;6u^foQ}q)W*5 z2s*3sJ8Nryl486+Fqi4W*5(I+ekMU-=|a6#<40-AF|xhauizoW9|^gCS}Y51(m6!=#)C{ zR*0t$BnxT2{dhdBD-OM1-KeXPHa+5~LCL*ZRCfEvbN5bBr`rn5m!~4w_k0@4r=4`qW%s ztBBWRDBZ`bfq3UJ=wot}m0ur4jJ5ahtji;GRClA*bb}AnPFoTWzUd-g2RGmaX*;z1 zHbTX}q+mqmbQphH1A9pm`yw?Igf1bpxBT@H%& zIO2SzOsXg7p{sL}v4`q$KeOAgTY8T0pxsw?xcoNJys;i`#jM0V%U{sFXRcAT*PX&H zVotKwWfLo}M5=ExAAPhh(BSN`aD#Vc?0DV6P79F2pv%t8$8$VCXIds*bmlL8lNSwl z7C#_5Lp=9#u!kAUD4|EUMAOZR$G{<24*nYF(MJ9`jY5|0`*ejATvI{mo!7`$7YTG~ zUJH*MtZ>cH58Ajqi;j!9Z#7M`n`TX|CJTQ2qvg)WsrI6|s69#vI)j6$h?_GxI7$m= zJ@NsMc?zU^_Zn{9;Ba^u(SB!me^?J4hR!JmaRCfynDj~rv>H|)0Z zbuz$1N(*TF{pEC%L?i(>e+qjXto`eIzjN zFCXZZo=H!&?WGONHxQ@a*;HBm71?vl$!dwK7FHfeV?Dn}u`hX!{KTdtQu@z~TpF0l zuI`GV%6nJx8Dd)$3#_2lz7awt-8R}E^qHD9NkM%lpFOlvz~%#~ur%gH&8B6sWK)C_ zU05cCi;ec9hMpI$Jo1ibX5D7K{)*=_<3HJbhh|f=6H2gGxt#W?pJFsKUexsSdtN=} z5Bb;EMPn8j&|Ha$7!;xm%LUf-fp{{x$)6n^7dX(o4;A!Dxh{Gpe!j&sVI7<(#qfx56m+S?o}kBcMGZe ztphaX>IlplV~w|m9$?Yt7`)|ZL>~Lq(XqNiFmBmr=;3E#syLasw^kc}M%3UL`B3s=bJZ^JjN1Ia}-cF?* z$A(qbFnBctB;3C>gmERWxGvFU3tE;1G*P8AmzjtQaf4=y1N7nVSfa~iCvNVhz< z;R+Se(FQ;7Ov=3s!ed6k%%#z>sOI2D-L8f~vgcG#+s^L`H{2x3Z7Iaa@c`Yv+#CAF zYQs|pLNzH6(P=FJA=Gf5kCN-x}jFeLoVm*$QWFynrL?t`d{TwN$Bf zJh|VsRPexGMetMeGC|iU5Oo_v-aj%zk8N_y_0~<8Z9WApqwVNy>mz7<*PH%p@#NVC zGhkDP1i07RGo8Jf_;79#wb_|Pe4EcRN6%~p`yw;JS9?>z@z!qeJuy=-F6JiFCDM+% zbz`wz*@d|MIz%sNuVl7J=abtPo{-}I4DpVQ5Gy+Q8ScTcn7A@6YIDHXLWhxDmc&}C6)~R1`fx*SGdVtEIu+|)gBNA2X+iTHdiZBDE%R83 z&(0%YEdmT;8;t`&^&xOt7=E0iRZn&mao?HLB7pHA2B(b}-Nx`LX zqIl^76M93F861koDVp<{x}Xbm!|OtnodU#nXf<%(^@;4JKf>$fIn05gZZ=0S&9u2oH_g9;9l?^^v+a3UtX9A}-0h|NllYuS6oaNaTk*m< z7m~L>i8L;Xpo?$P_9;!1T>c{C_69DJEOX+Ri$M5+*7M95FfKjt#;Ln|v z3>%DO@84b2;aNM=XcSI9O`3vJy^oUY3;oQa#wB=D^(nmkZ?s@fcN(#&SxDb+-;6n1 z*W=)^DMX;O1)7>w&^~5A+1C(36&&Yc;G31yAtFLpeZqr=ybR)<5iYPII*c(2JOJYk zq%*$*#KGvp76?|}!5-7nqO_|J3V#ff+%Knb;HV$OnmW_CB{77o`wE$te?Z8#Fc5w# zq*q+0Laq5;dVis`V5PALh(w7&*d`5-ynBo;;kkQj#%0iH*B*fOTs276`$+OPOk-nn z)9IZ8bHU*gqXZUwhk0ez7#cfCMlkU57R=)vW8%r0pfEn4ZhJbHq-jO)`_G?r@2Txj zl_Wv4*+AmKvwYTsc{AfZRWYu5F4(P)f#JsMpyAjG<88+Y;+9;1TseDa5s4$(1vTXF zHe0fW$$;XjHkebsnQFx~!Nl)t1;HLw5Ug_xWL<_qX5l2rh;pHE$JC(SE}H)z0+ihK~Pk6hx{Er;x0{8pk59n`xu0NWSOuolH$Xzjds6kBA=0#( z=ba~jYh4Qz=PePGA6d+Iffj-H14Dt$&O)+1=pT7~Bbc#jd_^up02b2@SQ#=w(E345 z@cMZ_n^*gdS$n31ZEXr61_q|g1J!I`oK<19rX=AR#Mdx;1v~*wXS{e`}#JB+rJTRsm&JX-4G*{ zr?1lTo9ZM%Hiy*YXOjDDu0XhPzhL|4RiOMgocxts#ZqTCMyy0eASS#guLY8URY@)`W~pn8^IdRr#6?d0bTyF4qCm{5q0OoOJs zPvGvWevmM)XVT9J(RZo{&rqDq-;X^d!7AfGb6!62OO%1a>R3|C=aO}A-J)rA$ARI! z3*D<9LiFBNP&yiY9JSCHUy7J|8E4REGc4?b0nA!{zI z5bpX!+|NMZpfecxzwFOL-yGn!w@WnQ+uYOfbLxBYa`j zf<@#EsQ+^To~*RRS)yAZaJ@YYc^xIU?mV?jRGvuu1%HXR*H&`(QZOtDO@*DM{9mxP zhToIEljm3a;DBzlz|o;Xu(Dd2J`s_`3+x%(?^lE|^1HCe;~y4}O2tmAQCJeZPPht> zVVdS|ref6*ynlQy=N-0%>)6XW;r{pUK(!S=^p)`P|oh6AVAL3AHjp z@qkkR4)FcOREE!*%jBSv^b0f{cM@%$MDUJ*NTT$E_jd(t$KE*yupuE63#*b*`uj#& zF{T+)+oo`Z2~k{Xiz~OnUyNhU{6l7o6*YM^OlOWLbIX!|`+aT`Vp0|U=5svDCyWw) z{wm24zK2;_Q%b#VsxUr*`ml3s9PIaLLiszBxTguW+=9dAoaN|z)Ia!zi!A@hofLn` zb!gu}o0q?+)}!mx-S909JiiP*yD9JCe2A8<=g{ZF8uG4mKi=-Eq<+J>)Q4>5G1tdGiDBj`*=ipJ8D8MiwdaLcGcoN%-N&q{nmh1BPGCA5d! znW4pHp$?Z?D$ec8h~b|4D{||9nsKwvJ90fETHK+i5^nGA72N&&8Js}PjmorBI=FQd zB>5Mk!=y)8Jc)OAuE@bz$uIEW-346wwngl`C;B+j+{4HLu5PPoi1EAWMR61hTMJMI)bi3haxbVegDr4+G zlU`;}Q&}D4v_d(&CUUv6`2^P}>w&CYE^fayj|;2!!uF@H=-UQMy0+spI*~1$&R1n_ z$Egq0jHFQa!;!obonXro{_|vJ&&5Zj(twZ#d+Gz-U` z@5Hl_<(TRifvF#Ash4*z>HM_-J@<}>xwEW5Yqu;mhn&EOZ`o*ev>Ck`W^-}phVjXA zO>SFC6{P}wJe695i8?75GioNTxR#8UL-teEo)p}^s05cD$zkvEJk*=|me8^12>zaZ z1Bb3hpv(6*^o!;&*<={E?Wb7RxQ%qU#$b!DE$5pQ#qFPIh1I44?0F`JBeA_0exQ_! zTyBNc9jWZV_8GYAVH$D?kBQFPxwJ?!oZfN>f&KFDnb)fK)8iZ}Tn*)B9gXK4 z&rReKUiMw8263Y5t3fX{ZO0cLcDUHi12^yf!*(XISi>yDsO9pc@oyS^aO60xuzQL> z+z;X3oR|20XO(cdXss~jiXX@9kK=qFmU7>(-$u_A6)MKFFQW1X>7>SOC?^a7cb!`# z*NMXqJptTdVLn$a)W+0nrd;`oGMu(28MjA0p?^N)GQz6SSlh*C(F>yR!J|dwzt4L( z;oBtK`$9;+XbiE;z;wDGd>#gTI*4fl{?*lxIosX*U`Kb&H zl3Imb{(9W$xfG2wML3OrYnXxX+vpw;hd(xIqVfA0q4a}d+@x&6We3IMpNJ#4XyiLq zOC9E9jq;Ua;~2=vl$&#<$3 zdIXX+=>UHYyJT4m+|O2{C2)~->-t1PMds1BO=mbSqbZ!zq)l{|a}=i>F`82n*?|*X zv)TO00_I{d&vx#Q60yrWSp-K46GVQY)V~GX3SVz zKJbe^JGq7UmAavMbs}1n$8Z)7Qv7+;OR`^QG*vF_r0R-(v~P0+30!N!pG)#ST7!Hl zdbNz5es==Ri>M&0!@sf1n~cy_!;iZ9PK8*hb9l`$17ppNm?oV<^40bMji~D&J_295 zB&nIM`hA6V4zw|=FI3^ENdrO=#Sge=YZ!XWE+jK9rsL)n6M27lGCEXPW4mQ2il!dt z%FiFg&qd~R+2<@Sji2dHI{u_Z7V7vc;xEsC5!QU4CJz7BE8!EHeEQq0l}f6};R&+@ z8lq{7c~ACZK<#*Z4JT^`JeRZaGkbZzdn&QNSqi#!YV?W8Su&8Q3|`k`X^@j8l-SIm z>(*={5BJX}rDE!^sq6@GetyBq`JpBqV|bk!-;O1jbJB?WGlB^hZWFVVi@05P4R)-L zr+%{jDE!X_!wwMIB$Y&6Go9(A$VkRxSsalGI7>&ChG9ng4_r4gio1FvoJI{2)Zwqs zd$o>pt+MC1eQPdra>cpKxg(<3Ebjt4)Clf-tH+cV7~((gH*AU5RjTqgjchr#k-fXn z5Y|caEVQx`nwxik>d(4IS?77o=kQnLUwtNdIVX+U{jsOe_LgnuY?zRY@nmSCB#5dn zVg4y9klgE)DfL;qphsV${4}raVT1m#*EB3NyUA#u(UFc%yV-hO%b#4 zm`@hv&d1Y&Yp>Wzs|$sPgK9{qe>eTSE|EySFhesrZF1T#lN~yBz-oi2BQ=Z@fw{ZS zlQ-vT=#yDdz(tNHZr%(u8*HX#zVq0M(;*nOw3GcZRtYzSePu;@E|BI|`s6^@6LyZK zCX>5jCu^bEP7`ZHAgY&V>sp6X<%I(@`c*Zl7mlK@lfTn-Gq>Z{59jHu2Q@gM#~Ge& zn<*H6vxGRuC$k={1ElNYBwDvAjk)=IkRCOiM0$JY6Z5zmjCr^zsjrwv73wu=YCbj) zIi`6v#ptxrB`@3wu=XfF7T%hkN#?r2R8)%SN4jp`!Av6{-C1r(`^vck3 zCi<`od()P8btvePzj%t|X~)puYa1X}I>jpU=yKAUCdYFUlkwi0P@Hy6QxK%6Lr)3J z=uDhX<3DB5*xxHr>U$-*+Wa4pK4wSGSlE$fV=dCzv5cOcU&Ofi$IysPLu`TXDavVz zko_w0Oabq{nl|{Eo=MwB7IllUIz6rIwgLh09rUNR-JmHKPDQ@xrQ7EST$FuisT@qK-R1izQ1zhB!k6_qC}UBz^m zEk=#ZSrb^zq*u;Uhj(A3;2$(i#j2xri@ta zD$+kNjmQov!&>r{*xzp@9@bL){w9)mIQNmt5D7N+nj>tiVsd$ zkU?{R%?GWR`q|^aV9!rBcF74EeEkw>iA-T8KMW>&o!#NRQY^W+eJyMASCgGE{Rnew z>V0-+uo5+xaFtvf(8Iu!38<~u#q4t#W)0`6AR39{(D+#TZks20Jm)q!96n5GrW+e^ ztd8l_i=Z2ZzOr$RBJ}i?iD)zFIGy*4V}2Q(r&^LjG%37|UE*$mcAs_mEbnQ{jaK8W zMEms_8N*~@^TQz4Z$SrhDACcX$;6okUH(Sj*~yWX!1p`{+nXry`=qCSZ-~X@B5+w90< z{m1>!eV^;R?sLxb^_T%R{)^Wywh{E;OvFZW#EoWTJH=tNn9u+F>;M0u{~gEUhAUP7 z`~1K6|Lr(z{*T82V2Z=A|Ci&CxG7H4b^l;xs{og?T!un*>STM;qM9lm+dh{|+i7w$>oI(HiSZZA2}a9J$Y zIEwQ{`eeLs6ZsE!r08@1B`m4&U$8c{1I2#G zh;hUQ-&ANk$i>H)cNnt%E!?BB*oG}}xVGgRj0^uU(eb@V+oV9(ipQ~$rPJB9A9YMt zd@HQ8ym6yc9;?UhMCasj*j_3@t4~ERT`g;7rsKedJ$b~`KFQJ{<)xG$oy=~#%0a&2 zD2hk+VrktdIJ_T>fpsEZ-F&Tc`J zb_vc(O4E|=dISX;()0ER(vHl=JI%jf59X4MZX7;oJ5qM_4l22k$8PjWu=sFQSYDB* zkG64CJfnbqd$wWJk``o1%TvnJC``+(gpBM}ENl{hbunKkc1S?~jy;Pz{}*xNpHg}8 z4wl&em}xsN!K0uf==>gnfH(5ks4am#KT??V+s|xDTp?%nF$mHPzj5mDX*5i0!6wC6 zJaP_)juwOJglm}FGKVF5rm?a0%b-*sP7AUP>9(w^kfC=76AeRf^IRadg-xQZYu#wY ztWNwfk)(P0J!l+lL6_8bVej3OxND(J_g55Q&i36?E*Som*eZ4!&q8y4t|SYL(-y-g&NPp^$Zu>+WMEZ zCB0>5bOPCxU7rv!!kY@OET#~b9JYC)H@+>(KzP^^a{T!TR!?@*uh&|n?d3x%Z-3Dz zYoRwovI1hBnN&M|5!Uo9$0q-a?19xVviPJ=k**WS^k6L6AB(0hRu8bcp5UJQm3?d< z!>(8v&|iZ|RK8!2N=7b({TEd{+VuoA+ebpxU>Yr%7D&YxUD)cJ!(5-*H?Gxn333|h z*huGEIA0BgpH(IP>0C$KyCt;0wVTF%NT5O9(d_Z4OYE`JDb^V-PQzn=p!MrMs@EkE7h<8Ti{_Y}r3uTyFG1=0G)CrxZdjN3@|Q(vkdc$4+J0{ViTS^o^DuyJUxKd4)hVUJd_*r@o8+JOw zaG)t2ZTpBiKU1>TKFcb9Zez<=O5xZpXKYAIWj>x^?0?)*!cwBe6D+7ADwU>=%B1h7 zVrhiOC@im%hGwNH_NRDb#(YJ*TWbtEyJ9jp9z$QQZKT{^t10=XCymJ;MYD#^f~H&x zD}6hVIXh~RMRo}u8KkryptlxR<_gKzh~%MZ!!Hzy@-Uu z8A#FEg{mWl6c*frvWtM(UoCub-OC#HWiT7Tv)=CakZOA$Q0T@}G$3Uk<%n&ecH>0) z-6F;-c%Nzp`(_m4SDTSu-WzlphY-Zuhn+YB&x z%_DZ+<^p9Wv{Q&d64mxg&>^e8m?64`98(K2Uu#MEPI}a5D?|8j1B*YDBFEWyjO&Ew z{N?&$p|>msFOKX*+>gbWV{bs~WUfN~`(KO^|IBTNY1$Y)?MNsz;`j#Iy@Eq& zo(PuFq0OtJP-`g1ruJ)-euR;rV@-m&o;B6{9z};Nr%`ojG@XkJre~)U=t#pEdi41Y z6(234OI^I6&DNutNfzYXX#ur8{cOzlF!*(52;NgqT07T{w(6cIgD0`HHM@$&hH8@H z>;kMQs6fe+Y_eS*L#_L41ieWJP48VrU*%lr)Q3&9U3C$?u`;5}6d$}$Q^$qVH<-pr zcX|^zh{om#yvni~WI8wDxwJg4G)$tP+n30?u#k*a9;60SEh^kKlkP5;pvg}U(rQ6h zAC+;EafRuuCI16ED|MTUg0_VfmkrVfHn8j;M?2%2wT1nI)XX|9OYnc;2NBjzf4qSvGEQ{6zA{T+w&&Dqia7 zQAURH*HO1Zg>{lJtwgq%NmNCx#72a!4@F#=V2ya2YC?+sOVF z44|=1XXxRiMf6K=Ib}Db(#zTE)HqGZ^mtZ4C)Yh@+qbD;i(xIqe(T_J!fJdQdIC>0 z_90X{3S(olvA6jowB>%u4aJ;sgv57eS>ksS73xX3({He+viHKvy*vY4WD#tT{g z=Qj<;kl!1SADF~cYBcEC4346QN?p-SUAdUDqe`S)}&z|8^6D?TuN z^^44)u@AmU!MHlC4N9p6*jBd@<41nOrk1~WT>hP#F{l@xBKFY-XB+6I*`z@V zJ+j=?KqXu>JV9lnJ`4O}5{|ckH;^Us@IM=-giv z+ey+n-K8|9BY|c`ZKU-_BkAs#t<-iTkIY}a!I@!1D{n2OiWz3?VqOZ2X}Gu%$;|iYPF4 z7fdczB2Bjv*ZRY;N4FbZ^=_o_*^+#IhEesqczXOjn_bcAVG|b1BYE{58m1>?b@s2M z3scniXFuHer(=EiAFowN>*Qo=h{&Xoj%RTq%7h*a&coEnyRf8Ihwc=3P?}FNnIHR1 zLu$481#)-Tr2pcuxu6{#cb}p#X9uPGJg1;Fy@HRV1=zKLKGwP-{K+i(QN5Smri{cK zRdMQx8Af$eMPzU>9czF7#sn*0dKuhEgMXhUv7JJu$?ws8LCjG)_AY@td^8$8Tm0y& z%6SrTd&tPxl$I1XauT`Ha7bRqk`AwiteFEDHzknurJwlsLYW%*2{e7bAt~>eOY0`c zkn*H7HuFLviY-DYerY4*rpzKA_h=gGmqe!X{Al;jY*ej#2=~uvI5P2`NV01mbi=r5TJdF38vexQvL|HX5!ey?P5e-_XPyC2N(tr0R;D$zY79wB>vF$L*E zETUQ;)t~HWc~T`ktr>*A@22n#+zT@)bIRJ9Mi$OzNx`6#DfYaF-NiIjM~hIu^dyjBLQT&q*;vJ`_|HC@M!GfPzZWsM{YsI| zNx8!H`v1fmJ!M|b^Mok$Z-<>hbRygIbP$CPZKrWg-u#A^rDWpfPSefAC|5I|4c3~3 z;0Pu1w%U%Xv-?@u0BJVwkvV^NbvPY*&%-h76w4c_iC=AZprRGTB0MFrWd8*W`cTN) zwrk)(mnUA--+-ZNKbtl^mjy3zW6C36!#2N~E{v+Bz=^`_nI%W7eRslk!d@h-n@kC% z=H$QChc0L{+S)0{PG=c#!O>%xwEZ85kDf;77e~@ar}HfE>=#UxJWr8+!{C}(jkN1J zB(wK6E|?Cc>et6mk(fY!Vw;X=*6akITs*tB;_RI4k?-8qZgW$zJE_*#t% zWB$>y5v{yJ?oD>|(?izovW)dE$)XL@Z_&4~6}-9TeQt1I6Z3QQh0&ric7N7&$X$ve z|AR8j(DV_rKV?RbW<;QOm^RszPNT{BW)wHV7M7Dnuytx(xb-p@4%+TWKW$FdkwtW1 z{!Q|me*kxL?I_tHjPkyei)J_-5gj`Jm>c-t59TDwV*8e@$GYO@?DzH8Olp1_?u43P zr-=diZOtXM)55dk!*yo3GL~6g*iYF?s$^-C0=J9P=<87j3jTK(_Vrh(HmQ|-v5%Z* zsFIFwj%M{sk>XGV7VRFz7B$+y%w`N8Z&MUM}#diCD_SRc4UydldkQQp%gN}j2YTuzLJArDn^GJVxG=;agBDpsQkG5!G%l&$0 z^mQ;a6KmMHq&p{``*9-bZXScd zB|S{7%*5K+zfi3|i2X4!WwKt{oSD-f3=j7umu1VTWY=m^nmw4-p#?8Q>*l?YVOm_QkUx`?8%)SgAp!r4NN1zDfLwd~ceiC_!&Wn$gzq z<21C-jOMHrL!MC_C5`_;HwC@Ni(6CZKX~IuS}luk67ZAp^+?=c3I$yu)8R=myyb_( z%ujR1(Xp+WJxKV1 zlWH4K<@SrM)NQ8T%0${RZym+|siZ5WOQ`>fFa5bPAMe;_?40xjvXcVomA#M=ykv%e zwbf%`j0t_!Y{CT(MatavAGUp2iP*CtkgHz-wI@YPF4cl!TCHe~cOI%W*HiFUV-j;! zB3bhWT5ae~y>D#DsWXY1vgT948lh*gZYqgq^g#nTxV)eqeVMaSY2t^3<#{x1 zS=0DzHY+cdNe^zLc&F`D+$K$HFQ26=W)JAi`2uoC1sRSqqp4?Jp?&iL#&jjwgFI8V zPWuo|?HWwWyTHss5_ zz`)Qy7_%%3KP`>GG@RfZ5<(4Hv7)j3&Cz{hG+FyIGg4Zl!Z&ZKr}7_{F?wRAs4=1( zb4ym!_uf`oTcO6U4)3A1y-pPBX36iJbAy&z7K1mtLNhnMr{3T=!CO(!1x%`7YDZ*{ zbgm!H>Bm{WMUmZIQ9k>ieuai>JEKFvWqvd#-bX#>9^0?i4L~WwIn;* z{v!}Ja(|iE%*#yKBV8nRWIL1W$>L_K7>KrAaD`IGC?>MWXSb$`puMjJx2p!yo-2EJ zcHWgA7wMo=+2#N6YlzTw%GKNM=zkr`t4(%`#Vo zW=9J1zWE5BQ-;#=HC?z9k_7Z8;q=~xY^z2!T;rQ!0&hqQv`;@pR-^A-hwb z8p7tV(t}61oS@$#(R86t;Pya1ZPghvcVOgHEJg>%4~4{m5Y&(QirOGwXlYuJ;dKd8 z;Hp^sJ&q0V1Xjhy;e~Mw9oizyMHODyQD8;AS9z#?Il|Nm9%BDJWj?(tl9!7#BWX=3 z3Nsmv#|h5N=W;ev9V$k0X^z-7zyK!CS}^fxJ4`!rC@^3Vy&JE|d+ZVV2ClD%_aON( zc0;G~nZtKNbl?=J$ym~T-D2*fLJ%izV$ z%VQvwRjz@5PXa6MFvKI*Lr`c)!RtM*5l}5de{OF^&+iJv{&zg29(^3~iDbxTp5`ZtmPi340sv@wQ^-^4og z6+{Vnx)>C=f=jBrU>EVyl1*88g$vN1%`RsjWvi5=M4W~^w|H{`wF>=!g|oZpkC+eK z19fP!SAc+z9!2Te$#lRxk%IgBIlTpjxKfdaQNtS8()ut+U6G~R8yC^WE5m7n&L-+S zGanO&Ttbf8cf5k+X@lHP=Q zP*b@rTITLV)5>~0iCK>s|Ef{mGl-_h2f!h;5HB~JN5JGD?E04pjXlX|o5e$4;~sWX zGYoX{a4cQe*ThKC`UkOaIweboqXX&Pz7fRf>5@~O9j)GWlpPlMnUvcH(AKaVNg6|t z=zE`Cl-tK#lH}l0rVSIlAIzm&9;RpZpmj0uFZ?LpM`|Nz#3W2OSkL~Bw8f|UQ?SO| z55q48V1vyb_GtGI6o_5N!t$NS8dMLt-zt=F-kZsCPRw5H5Nl1AA={20IB2yQwmXcV z^dkz6-jZatv=d$YZSXgxh?{DKk;`^q!Q`(9>@8v7PQo=kkkk}QSXaj)B%gYZB1>U^ z+vr8zC#11c2)STVMNhd;bePH z9-C!!p)kK04-Ji=y=4N4e?Lh{zoHPmGZRib3J`^%XxrE8)VmLDXEUOS2U8 z=xcx&))*du(*kMI8u1J+pPTV%)*5WszlH|KPNctS_T;X*n=Kp{g&Dm^;d#S=-I?u( z7Zeg`ZmuI+eC~g^ zh6D|0c!dPzcWhz!BHTE>3vzB>*#*ezjAI0{Plr_FM6 z*z1=zSh6_@(e)45-_Qiy(tC*RIyJgzpF~&mtElbOC8l@D0)DE)5g%-WLg!XEev+UU z*h0zLx5?Gln(sf*!%foJhcctJbmW;7a_{)#G?ej0BTC2#d4t~odHh{UGJQmTP62fK z{R#F&+ZI)yYH_SCgbY3VG4I+8V%-Jsuq;Qz*Vo8#9zx&i0cPC|D{Cib{3z-7hrT+k1W3z43}t$3w^ ziq$WfOqw1hJ}N+9MJ`T^*n^w#hFHA)xxg1o(26EC+U{*g`W5ytkpGX=C4wo&!Hn!* z{iRZ;yEG@Jo#wfOVpee@0>x6%rC>@!Y&_`Y=4o_g_j|VV-&eN7_CC|Ucaf!aJ7V}1 zXY5U`;uh~NN3BT6&k#65Q>kbYzx9*`PSGV>?OjynkxcSaQdzZE1=3X4Q1#?XD13Gw zSD$`C#PK2c5t#y?_-M=-90?ztJ_M%j!hq@DSW&zRV&6=MZxc_;r(MLF(S{Ut;4Gc) z&LpdbWZL=Ml+{>Ovy%HctS`=+avRjic)lw^C|{BB=|q^N7k%hRfY!Z4G)^tTFm?dp^Gn@2JQ2HK2qA6$T-;Quv|92;a zg-@jCUk>m-nFD#5*gpCfdw`C$RZ)3F3mJI1(y{5|NbH$}(ARB9x_bEt*AHZB|9Oz5 z)>XPaZ8*PcTRdY%RuC^;fj8MB_^5Y-`3mP;vVA%MclJ&ot!H;>$SpB`sDBbUX}?BB zW*Hl?_yoIAn#XnC8ZCH2E|BNmKlHj;nc}TxQ@8PPXdf`9tD{w^cKbg_2UH?H>o_R! z8on3)M{dV>z9}(|-+ul8OIu(I@0$OhS#|%VR^|=l0pJuaHkB`D{S{V7bWRTY1 zZ=Cw)Z*1R|TTC(~m6>I(!yngYXvr6zohznebEOXkY*eKKW-3(uNs^+^tl<;m!}!;( zEBK1Gd|F?38R=eHOy}}qnlGg$aKb&z*trG~jw@+sl{(%2af+H%9#Ct1DqYYu<7JG% zuh25%=ZpjmY5RbSwSE|{wwcbZ-$xzaj?>0$Y2pQ~bI7M&+vKBaLw5pKlhJ&Q~CxF`gUC7~? zFV8{yx(?6d>e#A*rNu1b+VzN($_1H7?3!woBArLbSXi}^XX;Us;X zIjXfH$MZj0Cq9R)74Fi{$XE1gz;&`4w~Ul8q!91di+4>t{G~MUe8MnxLb8A@^T=W| z#D*gNo(ngtXE)PJo6C;fH$#Q-PfR~Mg!(*2(3y_uP+@av_f8FZ&?Sqm6+5{0Ko6+) z>__>`5Iog325EG2QAT3i0@uCt+$4>%=OP`-gbW$nkr;{oicRdz_i3m({GHuh6M+YoZcu2N$+Y!b zaXK%CBC?F>zlorNnRlR@RgE)u*U`TzpOGqeAB}^)utzyr?4IIQc1oLvsiPTwDW{xkT$BV?sys-|3g&n9H_7lcqjbLK=i~Ia-9^E{9jS5cRA{TQ*s-B*T zrz%glDq9swJ(P~eH^T(ZBNAOnCODZ;!+F(QVxJ}thRvr+mVGiFKW~nLn?Wa+VpxVu z6A3!1Bujtp)lqe;8UIr)k{7wvWAcMcOno^I-fGLK_trJE--^YFd=r$+j%ClMz2iE? zjS!Z%3n}yqx9`S4=jbr3e0`1W8KOsa_lDBbORki6?Gojk-9?IbPLkaeP4cWZp&RA{ zDSD0(KYQ34Dk(il3x178z5GQs_huEBvTPFtTbq*Y{&2Y2KA`iPmyn0*P?}w;3`dKB zSXvd1!<{Cm)X;-iq$ZvuDA3gs9XhVo5Bntt5Hw{jv=a5P?CTx$H^^eiuYQJ}wK(u` zHdYwwV*jLdR4U{V**W$>`E@)##rEOT=nsfKFV3$yxRoC6euvE}rX;sVpAtVC(52?gy;CSUX4uv%e4rw`3SP4G6< z@Mn=DF$3=753?zj`?zIIaTMR+LW_4 zbsh0)vuSa>1$(v2iJjXq6cgWyv-jhsbDwXwa+MkDD0YG~MHCjoRmBZw21T%uA40j4 z;yM^`%mmx@ydbk889|XwG@!_yn#CH>W>p5AKS~H~Y-jU^G;&i0zToX&2sikxO$`Wm9dxu%1C@k!gDsE05j8E%SUX+(Qhfl#HN0wV&Rf z74j0-$c@=Q>*vt8>TXI2${<6%a5lhg8m-&$ zNzgDIK*x!#NS-x}LLCb*Vf|+;Ov}WL^p9+f^dY$1T}4iTo9Wn)N3=dRnFd>ppy7cz z)bU)%fj^{$bMDj8+MdU5AHOa#Sa((w=`<1jQeH@TGmh#9^wBlDg^VqXMgNE#8nI~& zdt#r>+9Z$QgP8-D9|f*Y!;PKwZWetkzQAHe>C&N^^DNXL2}QGNp_C^9r+XvW(w!o% zS8g;H{V|POHK74j>2ipl)x#d_tYDpq$I+qDkFB>$>GfMNHpq7y4va3OyFVAOCC(C9 za`PZ|&wk1^4U%WQyDOPX6S7{SYcE^eFTo!|xczQYjDc`T$Rj+W!v;>F8SVQu+ z7eXG4W0n6>pl>>XEJp+*;++&iV^p#F-g?e(Y6*L2kcH7XHO%b5Bc%6@Af>yiX!R4L zBgG|T)jE&OlI&so=a!=CfILU{let5BZfumE4pkp}O15bx{LO<3l)?|ER8MtkKOw{H zU5oK>usa|BTEIFI{Gl!AH73-}fyPNaI{rt8J`K|#?J|3c+qD_WO-ryYB80jMP5E=J zn=m?2lIr9~Qo_^}Oxvx8k-aJi8ZrxF@1wDL5YhU%n^av;#Gq(`!ESr{x!Db^-%H8TpqnZC$?~foDvqgmaS8@o)aa7xGuk)vIm$Qv z!MKRiFpt;7z1OB_b1tC6t0gI`eG9GHGL*JwaM*cJ2G_bJAhoiaOK=v}#Ns<$D%-<^ zG7BMlxt*3Q_zz!)E78`j$uRA)e9x32C9lUzI%+r*CHkJ$^H zt2|Ga$IHP#Ga4_7ylLC{I#`%qR0pC z8pS*Acj7%)dsCz0Tv~P3l_KJ@Mb=||*i>D4ys$6gqP-K?z#xWCC0cBHnShYis8jFR zckHyJK3Z4rXJbqB5F-DNc@Jv7y@ zkY*g&Kysy{DD27rO3|`q$;aifrfmWKYc+$rW|5#}>xSrF0KXt0jDN?k<%4Qx(QZef zSVzp3J^QFZF5RC|yZs%stS9n&+)VfsdvV@jofeKwuH?qQOtn+__8AUd(iG)?3AMkj zFfYqec4P5r&cD2mtMVDdnU(2qAA_>-DysxlZw})2!Nt_?Ye|cwo=O#OrD6{+v@Q9Y6BjRo52j^6PeyHQ+CW;(679f zq&CMQXeN{+(tH42&ly5q81Vs+4T zx`>B}^4e5y<49>(A}#Anr-D;%)U!yR>SXsrxoSRF{Hy{-?{jG1d~5#k$8wypx(UOX zKQZ)tAN6loNpmh;z>kQvys3*jZ*p0dpZ7JL(sTxpN$z0mjJS`r|6RnQA{neQI72Z5 z^hoZ@N7iX<2$|$Uw7b_}@}Wz}^)43j{imZos0!;JJwyE0y@Doc79alR1ph6hj#sTF zQd|=bt;PA!y*Z35cL{rhoKSrDcAr@+k;2wt_DoJ}Kie#`2j{#l!YXYK>UYn@rv0Pv zWJnSvO+H0(H$zBG&=Bct+fCPB2>9>X*L14>Kbl$QMH0t-_@ldJ_!s|uqpv@5F?onF zmWOC!T}2CStjfX$=YQ<t32TD&l$yDe80?|pO>fOp`Qhu zd=XyGaNyn#EN0{P)!=bKFb+pO$F=)XPPI19IB>8;v zbM$spn}FR1gRkgk2lYLe+vYAxE6|{rFdNjCHq*04q4V^`Qfi7n4*ek!aR0Lcv16Rr z=noNG*lI&=x7J>2j+;xT6>p+Zyc(sC7a^c*EBiQqKVBBk#>Q0^SQt5)6c2lobc8!C zPb?zyRa+^54a3NIc~)2W$fCHnWAN}&7$1Wqg;&C0#0d*ji}7PofT?F zv5^ZxnNEKdYe_sSWU<^sZCEXHvi88e{lammAeNueHhA^^;yDmDj{pDz%u3@&tqN(ukID{c%Qkd1w&HS3~DSSwuDW8>OD*7Aa zExPbjTvV;?PY)NUlKHe^d~wlW(mi_xUCkd+PPYX&zei1!?Rmy-)H!ct_&K5OstM*R z7?8{>js_%Z(DT}-Fy3*8>D^z#{MFAe*FYI=hMJlvXpyGS#hyZ6-TlbK_#U1g8jF!O z`%(CKFlN2&;mWrU;MT2tgc_&u_|bNOsT^vBZlWiQZ#pCUN)A#o+*sqEbnf;~L8JLC zjQm|EQTFNG)N&vPEn}9j{i72If89vu`AK}~u4ZPwam?WU0WR*vOR_NRpn#q#(g~g^ z;9lbN$0r?~{x1Av^K5=f_9gy_L=*M7oTjj(bV|{8r(t(9kk&N}f!VHXQHLexcr#qI z8prwW96urFb{4NX*`2y3Nzb%;XZAbXP6 z1_zw4;;2PGoWfM8Z;=6M5A9|%6ilHXXN{HRFS){zH$)d|r;Bc^c_5O!79a|XyoVoS zB(P?TE}Im-mc?Cv%l^#(+x~1G?oRp44V0}GO)(wCOh$}H?iovz9c-q-+YM=i<2j%y z1fdgG2zww8aLesDjcuyjQ-g(^*n129)z!l!9(|YwUJk*`?K`+B?i}}itP&1=jN_al zi@5jSPEcM<9>so5rbRRUv6W50gg5o9YFaZVlGw>Te0GYP)ay>liwxlHKZ)(l8O{pz z7jhToYKm?IzhUG5nc#cZ4Hn|BNrDuUBDR^4vqKleAm|3=rI8-1!VUi@#p;s$IG>ZI ztZv+4ZgaI3ckNFgw?m4^tt^hyCdJg$scazFtZpY9h1OU>Oiby4HPO)vs7=6 zIdoN!a@=n!`<}pyx12)%jzJW#qkxRerqUzd|L9lM5c();XRex?aBZb0Dv~xZEo()g zL&*VkMM8IV$TX~YxDV>mk<8K0kUi~x#fgob#~yky;Q1(8m(W1Ld6(%|ayLz=%7DVw zpZLV3(MRnvbn`YOdn||CD{BQk;8$$bQlW99>jmBP45*KhghTHt6wW$=;+6XlwJ{9! z>+4V}QQ_+!8J) z*R$?Pn?;%m@?3?w7Poc45SCbVp9N$R<|%0~^K5%`7EGasuhrO$<;t-6B6t=GuXEOi z53$A-``J>1bIjnbJUh3zmP;J`3U?)ieMEHOSUlTYs*N_J1>-b;(IiSCvxdFX61Uh|P8tNr=)8sBjF zT@9X(Jb@3RPP41ScCfOL1Xj09;NTaC)65f}*s5eNc2)f|R_@Fqm3ndZPf?!9Bz|Ci zx4$skbK#ittq$EvDQwc#1?=AjTXyh!EW2222Kz(b@h0*qrtTU*MWr&-+x>%C&F{uj z3lqAmD1+pNT-XmD9og1flY7i)92xNj z555a~@VEiU3GG4hlyG`>p3(a3&xji$gL502+4ajwc-ifQ!3jmktV%<(b~HYV8Q{T@ zA!z?N4)ODJXmPDQDLpTz<-X6@PM?9enY{rZg$Q5YWQT|^WE){{QMp9oNkGZ;H#U7+2_~Lfx~ktxIl$`4)-#X^;YbSdNm7( ztA&}`96C4cBONTdirMN{aNxjTQmvJtISn%8czOc87;%ZIEjERU;u$v2TO_Q9Ic?jb zLnGo;s3UYgl^WL3`p8OHyjn~TR)o{m_raX^ibdR%PhOnHtPw1(D~$Q>u4aw1?s0>d zCyu7{fGu7Lm(^yl*j6Fvx(-rv^GAf01+argX##&Nbm_<+qwe8qLhm}^ap6QH$FHaN ziY27~a4(H3JWRh6JxD2RD7`-WosQjDNw-dH!2VIv^ijY@3>S%vqgw*Bqdt zJ;!Nac@}AGoI%k=i)i7+FiOi5c&nYde6QtX{?2Wl?^Tv$yF3qZZf)|S+Mm%h`N}am z(h@_O`Cd@pBnBHv7gn%$7k(bT#tb5ta-r7@@aoS&d~E-Wuy2CSWXuKfNWLm~d;XBW zz2JABn8}RW+UVJ1eKK&Df$r%oXqhVLRF)himlLb7*83e&2bs~E4hd>ppN_G6AF>W9 z4?&Y}h@|ReZ0$K2R+(bK9x2ztVf<%A)hj@KNH6#D^BRa~67}p4q$^6psMY={(k`sP zfiu;V7apLE4vjPLe*L0@Lh!JNc zJ=t?`;>~T*PRvlb ziRBAsA!DV`FBOtX!~W)BZ2tjrdKpJg2Sn5NpP|?j^Mg&|ZLKEZF%0VwL0!`|P6 zTC7s2#9%2oCWn&cw?Bd|@j2@p??KsieYjxzon<>#(c=lv=AKAMzg6J1u{U8Dw+P;A}-`rnn)q$Kh`<)1DDt$!R-wD$M!-B z9um7)@QLBrFxCOG1-CG)Tb5mUyI)lIX8`<1jpTDBjrgZQ`{`Y^JoYV7Ma4K1N()e? zA5YA&CgwSdKBmT+)ep0)A<#nCcJ@z5zxd327mTJ|zpVH%_42zz@om#-6w}UNnmSoe_9 z`REuwmkviPqSSfT^!Hv1dwRy0d5%rNu??^AZv09L+m%KqkV{0;Q`P9r$k^MLN{g7!0%2!Uh*j?_l%V|K6Nb@Fe1wB)p$Z*#5DL>6%zap{@PfT z`x}M4)98LfHFyhZU?1}h>Uj_1Zsy+;Ze5GD84I{?Yh0n}m>casm4WGY6VcsR7G%Cj zl6K>A+NE@ceAkMRw;a*5Wulb!h>zkQKHye1Q+aM*H2_!1oHEjb-O(i+K^b&8q;LHVxQt&Yeq38zu)ADSq0ECp2mG zJCfe+PT@5&IH=smbvTUYIQCr}diDTMv79S+(*h6e4218Cdf;BmDez$TY|W4yvfO-( zUc*Ctto033*H+>I-x?Tgy94irCQxjF3*47I1#d&gz*p0A&~s1(T8+oyMB7MetIxy- zU9sG}ecrfJco5ptFLHbHs$s^OWH_p;3GOZ-aA%qT2s$hStMfsa^nNBO_CLT^y5G17 zR$lPR$q417kCE`vAu3*Yl7CC*D(18M?|X@7^lNbheO+~iY!>PA2R?iR|MKf7;U+-O z`)x7e*L9ZFKaLLjMd;Mug_PLOk<{l&==4q$y;eAYQMwKW&8@5uTei!9YY^`nHy zHp9?^>Y$7CBp5Xp@hP5S5j16JX8Sh!ll=((xLw7#)Bf}xE@S%WL|U}h zmS)V4L*CDRe79;jjl5zzwajCbTB1x(yoB+!NfbU}o!|u~OGsvo4h@AH(3_%byD3r@DqF z>+GW4%tvz9IFOs28VeoMSAg(a9sKAu3JNX@Fl=cS#5|Qii^|FPTUi(~MMC+p zNZ^*f^x!PkKLq8e-ca!@oMqE`xec#YO1Bpx6W7^aiRQb<{az2UEMZwHm-_n8kdx$1 z%aoZ#4j~1NuC6^i1FA0m9OlIF*`q!^F_oPBUf{7S zYbb2}EqasDM2TYs_+s;p^YsLpxrAXsOmOW%R^Z2PZ4~1-M5^)gZq>rRv$o&>gOG9k zGm1Q2LJ`dCb}+jUZ_Vn#?e|V&MokHNZrMccOG4?dm>TSCaE3sS3-~EqodnG^=z#b} z+T3!Oq)QU1Ekl5x<1mK4FdzKOb9xlI=_Uv9%Rs+oJ-ppk4%f_nTb%xkpE3yltzKxOwV?(MJ&T%A=1 z+c!q@JO^iRuN=hi?z^*`@0A8F=U)&{<>F)PRSiO;BYiMrRtN8HNW^N#@RK! z0?$?BsLp;!&(Cfn99Qba1KXYX{%?Zm?&R6%B}H^xvkiyO+2Vtq487Om?(3bxP~NSI zX0i;5rYp`baj}gzSu^c2HdYhmG^_NER_f?mX`F%Owsm}r6y(=Lj$p?&;*J=rz|f;y z5V`2Y)gHXU)1U6daXWTHwCfX|Z(fCd$Q&MDeP9a}&w3578aND0y$(aCG$C%y7<#+$ zA&Gpo=JW0(Rp3FI1=f(?Z=IAw4GhJTt(hidBi>uw10ug(60w$Brwrbn8l*GE#PdI(Qu!YEJS zpd>EH9>OAz07{T-qMe(zqp7nZYF}%B+;umR_sA1%s}UojqG5xMB=}8|1n)J52K6_S z_@(#5Xk2>^wocxJ5`QOSd`t&wI`i<_l^)=UN}z#!0*vEr2k*M2G&mfH8T zy)wh>ROTO_mBc^50>|S7*U+8n7K(^1!^I=xPasdjTrnQ$@*X zg^*un1=!6pLdAEXbiNdn9g#xm&7%BwF6nG1Qm3ySH+ThQ+Hh*H4W`ankNH)$@UihK z_lTJeUqso1zU*1fYUN+Z;VZ%ksdU^GCJSpb++ew59w&HU5~ev;b1Q=H!J~6|@HA)& z2wBgB?@7x+_+Pxfo{$fZ^I^9=6CvES+W~IzCnQxhV@rdiP+-@}m^fdmIF6<=_%d=K8En=-9mg^lRTuuU3o&%|Kbs^t~Q$ zn)qSfgn}}zc;yh6n;roTKF4?qY_xdJtiQuSDhXBg^owF6Nj021) zk*0$;&y((Z#u^wK#T~Mk4--5YCoAYFH1_r4(^mHRtkXv=S4BD@;Gwr{hCC`t0$Ot5)~Aj$*IwkU+am|xf9ApKX>nZo)hUqHXARRL0(hs#R&&Mj zceuR{H(+2kyUPvC^6Ui*Pj%&Ze6%f;U@HDfwyvkp+j$tpP6v=k-uF$$^;(9ka@MRsU`+NCkNB(uHmQ7KU~$e@!(||3x3lV;_a3Uw6}bVLA?)9 z$af5_s2D`)$RPe1P^ub~W}%eb`5I&UhfrquILflOWLXYtd^7DfhG)LPoZ;y- zy5=q#=rp18+7BqX9BI0ACtRNN5Tp+oVb50!Tq~4_ajSNtNyQx~489CUt5@Qig628|}S*Q@=0{Xpc_5X)3Nt$FE;21kZ=EeY>Vp0=8*e1 zm~Me(d70>XU7vcq=V9rPCCiDZ;f3L3e8lJDll^8GaLF0lw3z4VT{UB9)!;v)X8d}7 zBMN_u$B%cI@6ap<6Vg(kv3)LXoy9a-4>_61O6 zGZMx!fm(Qe!;#KjTu3`QZOL@S5q$F}49DIIKsWQn^kMWorZMlfH9r)aBe!Avd-j%# z9HV=}!lV#F$qAc{aG{LVz(vWJ%N1=feICRS$ z!ycQ#fWkXimJy3*y1(Pcw_+s!aTyu1-GATV@#x)e4xd%!F+`JtEt?c@L!1;iE_ubR zNtyyKrITT19`iKXCxiJ5TU2}g0^H;i@Rh9~ZhH^_4f>2-23OcktsQ;hzmV)$Km2a; z9XqrZk&{9wZO=SGnrqHciQ6^E=?lkmjK5=YD4BNM-ATrwpHcbdkUn6rJA7})6dUQcx~rQ64SP!DV|rM`>Y}vi&v1x zZE3n}XHGU(Hj#Yw1WMX7gk>^bG}~kmEAt^vBv=CP|0ji$x^vLtat6Nb7NZHu za@4N68fO+XVopOStxF3dmHc}cnp%zhoe3oUT%IoZx8tQlcj1uH2x?73I-#tGlhih% zp}PUbWktiHtC!(O*nH$gX5cKdXjE#^qU*^U=qU4r{(d1%RXt^>yEzd5+7KkY6r^<# z1*FgJ#};c(!7mAU{H>XV3PR=(E%g*VRnJg&%OvP>`wm_GyCC86oy;cs(6*C8mRZ55)iP3He6;Xq*KmP=5^YEm zB@Ll$49M=pFG15t+a#7gW;WA5s(+AEI{4JXDT-gbAapd-2p-bkcK0^Vle)l>#vM zMgfB-y~9$O=lD@>E;%o#V1AJl{Qa7-^F}t(`l}!C(W+vsYs|-m-y*5K^9vdJv)jht zBM=UmOeccv^X6eVY3{N7pUiftOCeF!DBTtz{l|sHVO4l;~%nl#yq#zXl=0! z?P|%ODGPqnrAtAmbC6{>Rx8n+H6bXs?H*pADN12d%kk&yRIFH^kGrz|;##x0WR#ak zHZ3AlcQggxWlh2Dc4nmZcr_i3)THILld0PNJZ2WVW4o0r%{iM%5&O37-9A{y4Sm=$3RIcg?OwK?7qN&Uw$PKG?;jzE0(8 zSdZ3~$N&D%vkb~xgc%w}@Z7bM%RRM^%&a5n(`*A;XZC@{%2ZQ+^%*+ja~t;_yp54H zXJGH6YU;kVgvy@;pwgZyzL1SSKehikxt|gN$+VmL-ort9*CUN7&~+I;d25bG#9jCg z7DV%X7ufQ>LT15(eSX~jqCD=}`K4GE9)k+<#+dn(F@y$7VaI+s`1sq73o^UU8}BN~ zTdE6G`dWduGoP@kOag7b)J8u{y27koa38&~eZyVF|4&FYBY2OG-r z_Z3{CNk`asUB-w)QJ(27r7-;xKt{wN?q%98I#y&zT<9B|^KFEV8kJDR=dEP)pFC=c z`Gb9+FL&Tn1mD_=F}Z&UFh6%9)1lJPTksnVqn@6#79i2x(++gv_UP(fX2~WYOG9^E?J{%L^Y|oJ9C` zgB~iJ8s#!xY~iw}7V_t<3+Bh>D)Sr0+R#X<8Tox|#`cm-+8WN-X2RWIIf#cjM{a6zDCQJ5X0UH>1ZtD|CtjSn`F9QFdFUvnn91@~8pFZQP z|1LoMjaP7c#wASgT|*@a>uAu;#~}NSkb%ef_57kGiqsHefRS64ft+79tq8bA5!WVg%UCpD)O+tK?)=#9)ekWQ!=i>|qU)W}I z9+Xa`LS@Su*u%2%`3|b^?|C^VbijcEr!Ar&`*KY69Kqsuj4kjdSvZ5eF3@ zb>%AZ=?tM6QLAY|#}I95D-45enkpD$Z@qLhWNLn<_=mr7A1Hv(t!U7=ngsl& zP-y9_CC@=`#_16UZP@^gS&Jq*BJ?$gS%y%H; zxCpHI_YAkGnZVi0AK{N{HOCKX;CB6J<^HUhz`r5)o@%+f)cPVYY~9S12}C$m7PF5im>Y zJD%!E!PMPspSJIyg(~qB_2UJasAe&MV*=chiH4b5rqI#h7A!pH2i>~joZ8MP-bLq+ zycn-o?#Y|kpkkcFVO%B_v$y)q6j90>u%YM?YkHp0iP_~BaM$k|@D&e&c>ipcxp4&1 zDKc>LtS4C&uch=NYcehoqpaHx@Za|`?5m%I!zWarX3JXeS?YnYmv`Y5Lo1XMm!z(X zW2wQ!iDoblh2=>f_^&LUOUgdXxQ(%tGU+hopUTBBP7`fE41fWfS9(U~;NLf!Kw*sw zxEJ_?s9*?O8k0iLvty~=*q^T2xsaf|6j@ArjXrxDQEOf<{S@q_$>;Wxj;beJ@_B^Z za!WEu9Kwa>tMItW5FE;#1#$LAI7`M~{Wrmpubi`(-=}*GWh0_6CO#2eC;z~^%YK5$ zJGP4QM(L&e2ikuxmtI)y#HQ(GT(`t!-i~7|cf7d;&IQY|`7H^5MDIh5{iS%tfbBlV z*zBu(7PWRxqQ}x%Y-U`H_nT$N>{KYypg*Lpf5jijKf}Kn6vbD$w*gn|55<{f_rcul zCnh(nL#^c^Xa?87`L8sJ=_sPQPBI;+1lqJng(6&K@Pq0R5GcIQ_(kE=BfE-RWu0hi zzvKKz8;6!4)?)0M4RO#KA=I=5^U`Jh(V{*a8ZjU-WD1{x17s3<|&V5iWV>r ztODIFUWT69KkECQ9JMWlum$U6yj@1fp?wD>xS~ z2AH#o-74Eaxl)(hON&uV=_d$|&Ln%?3#jD-_#^K!c4v5kT>m@HImeWu2Y#W?*>fPb zsfu*AO{a+q*z9)iEp9hRK`RRhTC+(200GAcW$AShn^WkNVhLyo^N1+442s(-ei{ zuba7ji34DMOrIN=H+ih6?O!c1 zB;W`a7U)jKnc}oBq8%?(J;Wnym$7fu!yN(QdeJcOC? zT`<~}&UsH$p=U)O@l{7Gz6>&?2?H}oR8F7fZ&->u4sC=Z^DlArr#Dk;=W)E%=mht; z6`;BKA~#i4o79k{dGWAC|(8i~X?<2sphq+@}=Vvl@Kl_Sl z<&6}exR<(pBCvk>3!HBgfM5FGa^IHh$3cs!FvYJ*e_Y;gijbSZ|0rb2Z#7rLyK8JQ zbm}TByH!H_lI7`&1=CW*C-Mbe3-TWiB+?xB07}1^j5YGd*x&mM@}C?6HGwU>@=8b2 zjbNFd=9Abz^aO_wKV`G(Rb0&8fX_7}Xclr2D`_fzRVIIV zz6Ls-nu2p~T>wvKDatJiMKhBU*l=q-WAJ=}5qk$vm)K5+y^?9d=rL-3I0gl7od+Mj zGvp8J(}9L?x^Xy)zOdYEQz=7TnC8bAQ|hxZqXbQ6r3n#aQ+%2R#b?p;iM4ROD3a&0 zN}s-bUx21w!Eigq4fLl!;Y2U{!XKkZl2@HipR~$w=DT#3_uoUxmSgB->K!zZU5JZ> zUvVcJ^-1iwJ>9$e4y(u7(#}0w$)?$kHh3%|$P#4vkKM@g6u>R=vmyBKZ{GY)b^Kmg zjVI^u$Wqu7Zx{UKOcdHV^-X|%#*g9dcm>Rq4kh~PN|OpcV8a?Ca%J5E4#SK!m#4{H z9c|*4Y++t!ubq??FGri_X=2XCY)Hy@%Khi32%c&-SWqpBH*_9zJM$_?NbxisNJ7R@ zOQ$!7>}YrSLoBP93`%bAxl8VkdDWea_1dn$GhQ~I`)XqV-J^9}&z(fxi|!1lnwkK~ zg=@jXLmdN8FhgP9b z>w^VwlIbbbxh8;(i!JEGw8|Hy>AV|R?>0iW$4V6Tlx0v-RcY=8HHxe9qLh~%=)7_+UFexc`|^s>uJ|5ucebJL zktbL?SCF)NH_?O#A$aoWKU}>sgjC)YVsC^L?XJ-!mxJkeZ^1&8efb=m?0^znCy~p2 zVTucR3lohIM=r-;uzV|iD|&@yLg%nrYXy2%THu5>#>eT1U>zb4QCMmT9@7hiJFgl+ z{;3o?bi4(jGez+1${_6cynx=zS(3K4INzb=BtG9SO3#9HXxKiJ?#tBB0mk$gTEJKc z8U?5@@d>T|8%OIya&e#QLsWaz4l}m&L$Anl6u*>&zvQH__~lZ3B|o34;;T`4-98#~ zuMUMTSz~vT1RlMhK_c%a(u%uQwB*tUyz%G&`4;RUoybPUA;(k>F6o(@9Q>vr500&Ck#YjdeI^H ze;qZh4}@Na&A88e7&kEn=8ZX>)ZvyzEvXMt`Cu^h3SMRm z$k}xNwj+&Gc#2XvuR!fiDcaY~pxFy%qfy~WaJ+XK_I)p)jLI^)`1cE4TGRqJ)>~tj zYYfU&)pEZti=kAwG?~e=H^$&THqMw!K0bTsmy#m6=ZKN+l|T5`suQ<9et_35`yyBm z!}{a>+{nl`>|9=s9cAaS*V>M{3~R7qG!CVDt5HY(J5FkT3ejV;VYJ+gad9`Wtbq|c zkY>Az+b-OcnvH7D%8++ZhvIXN()|FY(Y*zN^9mkUHyVH>1lIf#bAU}XK6ds=vr`zcn9cbA0VcE?ssGM$S56?Ndb6Dkxp zWhc+l9m$Ub=@Wv1PuDyWuEwo}4?4_{#0o1FI50BGhUil*X+W*&p){;^CwE_`9wP7e0Li?Msoqw=gcO++4gGRE_rfN~E>b1?mo@ zg4mQ4u%FQhFLu;Gl}Qtn92O;MZ+$9xz>$OcSQP4+!@PlFcqYe+^u9)sqHzknJii1c zJ)A`C9aqS2+)^M}6%6v9ht{5q>Ep8<9|%7}>ra)i`m-cXX1Suzire8tb2|RKYC?7S zoA6kA2>wzPC0FSac(vm`_V~nM-{eO$tooDYuN)$;!XwbqKMw|^I=DG`e|SGUXG6BE zEIwfj;G_=;IBRAa)|sB5H!qVYBEFF}SRMx5l+&m+=Lt?fP|7WN63GSSCvXy(8%XoZ zWb&Ez4qJ5Fxj#CGVVQb06jX|$QPMKzJIF@|=RM^5V*yEjji$cVWIDUGoXkrF`TCzj zL8MOs&pv(t`W?r?m;L{b?tF_%v&K-yETV?kB~-Fe5WA1$;a}5s)IXs{w{BITu6hve z{;!j4-?dW6_uI5;j14#pmT;>IBp}f zq`XOq>Xz8jIq#P^_9>$0vts&Y8bc?2cF;xZXbhKKM6=&T(e*Y}F#F+$KQI?RZ#{vt z7AR4zggPyLqyiQzAxI``#3HNJXQ#JG+5J z8RL~|Ns6vQTJ`=eSX>!~<=Wb)d21}S8>-Q&Uw-6%tOHa&#bNDY=EDfx4fc07!^0(O zVW`IdmSQt!bwC5v{`0}V^3nL6^^;uh?SU1F`!G*j17?T4!DNFITxNzRPjK}~s&sjW zF~tY)`B7WC)6s$xI=i{3Q_T03W={PdCel|!Q|j9yLP;wV@$>E;vh`R-aaq?O`ZJ(y zg#>v?t{|N&Pg$qHcPcsW#J?ODOV`$pr-bF{kW|2O2kz_Wi5gOS?kl=s+d!wrd(cW_ zRWN-h!X>wq=$TipW10}_k6L>HH2nZ=HqD3d;|V;I&QLg>kj1%duHtpPQ^Nd`5ZwRU z6GPjM(cGR5gfd+?+%%J>F20NF{U_iLn>DmrL6}acu3cL_whoMYL{SQK)5%AMWb1r_i-SJ24QRrrbv=j#87fM;7XXhsXc$Zy8rh)kzviQ;@& z-`Al3GaWTVzoJZ#I2G&r(B<0T|NS z%_XP<$i-UG?ess?wO|HLdUXU(ok+%I^UV3?6EaED%^4MVhE$i@OsAVx^X>a2`J&<-IvEb7#$`l?Lx*#oW!LDv+gWaAu3~R=m!NGDR6WjEq0ngX*H@$ z>rRg^}?aFiHgM9yqBi@lr% zrY43wk#X!^XV%Wx5bS-uyaByeUV=4$MNwmby|4abs8`?+I_HgNydR7S4m@$@p$6c7|Yl|m$B%y@- zS-R0|L)CvD;6A^3{8O$oDJ$s-H+zNaVUAF$99%^r1Y$Sx@U({k-}Lt-svaUe z=i>&SFr59m6ql)4Qn{-xTE>szoZ_3Hmlw);9NfYsyV>v;e*Z{=1$xvpMVdZ~hv3;W z(%}4Q6|OxQ49jxT;9z3|Zd7Zg?dC%?@VSA6H(sF{AuSTLTZi2$GvM`xVZDpD!ni~C z6!g0bMD<>-J;K$g1o3WMIT4fs4$ z_|X`~33)7mEeh7m*_FiIO!~m7Z&bz0#)>GCI3BGsJV?8`7%LB};2fK1%F@*$4`0T9 zvB@UIS?AHW%Z}#cI8fyrTkMKAfj5Q%IC$^`NWNXfHMa;-OviRHmY**#r9m(WLwKWwYX_iB=1{VGeyOkyqZZ;6mQ zn>iP!sZ;ltC|Lir3KV3fp{DhE{?z%Mbb5akIT?nq8SWdNbK8v>iJ!3TsR4Pq^T@WU zm6D7sm=Cs?B7Qca^cO@&r$KNSh{7(T>#&!Z(BwisV!zY`nvvR#1IM3{!6y%zrZSn{ zEN;aob)i_R5C%VjSe~`Doz}}RHiN?uKUH3w-~VceRHXjEQP_Z|!>{9<8@<@gzTdL* zQ{eCHCeB4N02;F#!A$Zr>0Zb{&Rd+gPbYCqk0m$|?mn3CB(s=Va z>{8%x7R&SZOo_qe$v5%2y(CSwy@YlP6wxjCH2n;&q}XH${>`6#@b<-T@S=Xmo)V1X zg0k=mUj%#hrlO<7Nenbpq01Md$-|-tJU+y7Gt5NLv~e|9aFUSqJexw=+DT$%HEH+C zQ>SwrE@)j1<~1r*k<3`s=lVIxh6du5=+PG^Yw|AY!&-kgoT>f_m)!XS&+G2MX_Eq& zDYlU-dr;3c?-b(v=N;iLOxwpL?>@(!T6T^xx#B_FRg;siQ=r?ehp|Vy4wbA49}Jw} zaz=REjLbcBcDoFJv1%tZdK0O9xk?ApGpJcCi%v4_G{cQYcLGi6LVY|Xe>p{ek5*Ic zs{d&Ef14;zMV<2emNFl9HoNPrW_?>6^(2^(z-t~#y@;nL-t|;E_B4@H8&)gspvBDF zGGwkraXVtI9bD3&rq*jj=FJ!26SeX7;G8W-xFH+p%NI~xpk>Or>dK{KT?~PBBUcU{! zNQ$S3DaH2(~>2NofX~}=K=QpkMU-Eqn@~bF2@mjqq!m8gP!Gv0Q&o3Vh1_f2qHA3+ z>^aB7$0L*Rle7rMY}-wO58UY&FA_RtnUNr7(%42_+NC8!EsHyFr_Ca28;PJBbr%+E`O|v0`IH^*je|q&ct^>e zjz!whbG>8q?!Y1%mD^0m8f9p_@qRMsSx)W8-r_aevHYFZ%lI?m{4h853J%quVm>~` zQueoG8nhR^5cNd9feAVp8)1`<0UnlfAXM2;TFa-=^~nS1K7TxoOo?MV+Xm`<$$TU_ zAD|TlNH4{U)_#0RZ&vxT&OQVB7qyS=$uX2}7)GwDCbXc2@XBUBUb*5*FPcwNL{l7< z?y@1p$bHn~A3`JBm*97n_Yr;~i;ZtjQJUfbDww1}FU#X_)n=wI8GpjJNt$%rViw(% z)u*>JDsaO_JF3(OqJbqfOv|`UYk%IQwIK_z%WWlYE>%b4szUTsTu!S$*wg9X`>3@c zk){bplk?51D8aR&-oeTA^KAvw*KFyov?mHeviOOvH|Yd^Vd^!soM#Kx>K)L^vA3`Iu6;9$<)m!%JZPmnv+w)TEVX zPU3$&K?>?!i)PpO8KcIo%sko z#|q(QiHD$&AB+9pKjNU)Sc=g3hhgPGG~uu|j{cUy(=qFyNILMO#Eu zQ1}Jwn+g}FEip@3x8M;BRTiSm@h)%_-@%86(&*q6z{NUi!ZDBg+!;G-NMyYE#xI8W zTkkgPH%P)Kdm`|H94Gnxu=O zO;&h%nK1mGngLD;X&{*B57YK0qQu*$XuWth_Bc6dPDvQw*&c%IM%2 zL(PsMq%jgmx`B&vXIw4&$(}^lo2@8H#hi?{&O0J}nK@$ovOh_z+x#r^ zT%1Dde3r~p>w}jg6Yy@NGQJM9Ly^WQ*vfupKkTq2*CHi~DO`xpx9Q@km+pAzgeKlG zH^89)9lV`0K<5^pr#H{fQ}p^-bl|8OB@PC%PCg!uE#6I=?W3vPb_OoIu^W#a(ZDXr zM`*N6l{}tnQuR4w3`}8nAJZY+=G_QZh5oolgK*1KX);*1oNlfWA;ZeM+&GVOaO(R% zcqn@g(?@zRDudmnpI*SY&UD-md=+ddWl zgn3|TK@91xji>u_kc#tn;7idEd>8x)-yJZg`+;ol`Ik-6qK|0Q=^X_;c}dgrg6KnO zAx*E_Mn(mjaY^-O@RNNAQA!*ojJZJBCrZiLyqxsE9;1r3MRaFJJ1#d&#?!eU(9(J$ z{Z&~@XM^`p6U`+dy@hn*nKB((IET1aS^9b5AbRQAq4p|cpPGj8L^^tV_R60Fe?@8asHAC6WOOPUY z0nBd9Ci_ip^yH~GT@?Ed)vfQN66-j-qk0bZoY)6hxtrkOYDLJkn1C0be?o_4*N{JX z9Le@F{>SZW_%(PHequUgQOGsWO^OD!#g^Pa_F}mGY9FkqT?H}uSl`Xp{jDWeUCw~|uj2jC@ew9M%eJ=XN2g}0;WBg^F%iO!(D zl-Kx;?@eY?{b~D@Ii&t~4P7yJCh4v$YBpa+!n%s|qECn9XHTS&&fjR~F&A~sSJA}V zC72~=(5An#ecP@qej3aMsB(bN*FzsLk_}tRQ{uBR^%bE!^ z=@HNmaTj!v7>iG9ci^u4NR~HWkE=A!ftZmHdGEiCo}0E{W|S_i{ANp&-Ud@}-(l9J zVodVJ-!WmXFZt#yq)&gPnYJ%M_dQ)nJS&}oT`Z~DP>i;Yq+-~(T^N;Xho5D3VC>Ck zTz239nt7bS_emGEt` zzfYcq5*E>Tr%kkfzbQN$^@SxJLGV;&H62mdNF(mzG=&BIfPt$3P}BGB6(~}rPhTP$>CZKI$CAmyqT$>eCIVXRx9NfcvGWMHlCSY zgcFAPuzGnMEmZcUf7?n>T-$&akGV*LySmBq>RFooZ6~ez1GFS>ELrKN;myQ6oHk(_ z{fPO5wV{QW^S56}w6@Ok5-j_Ug&kQz%cacluS7?(?nu4huh0#W{xkV>Z4rBSA4 zBsq4PlT=Y6E+{s__DS{_!fD~4j2Y>UaiQ7^?Z{~w(p{A$v?kb}{(F25zm*N6l3EIC z%p1l>ePSdzw2jt;x{#N@5qTIMrt{yPq17!J`e?j^l%hMi&tI=Xj#n0R=;Y$wDN`uy zKbFy*%aN1w5Vj5)qBMP{|Jozz5bN#w*HVW~T{gsbKThvACes)1V6rrwPAgoGf$3^p z7?;?|?zMJgG8#z(g)Za~Jcw#hFR^RseVlQn4CUjGqvAF8y>4MMsYx^)Pje)R7e~pg zz=_;GFC+KJKe*U&2$#QXXZJ2ud{%1??U#Bv(a)uv+g=Aw-G4dQvYhk<10me-`z1_Y zwgoQ-UuW9ZDC_Uc#UrXt*y1&}-MNnYj%;JPsvl-GWW&E44nC;~ld4rW z^32mo3n)-->=-(i z#&luBG8|sdBMmXun_ZTL@4T6QDW;S?t<*8!QWdLT&h5 zQq;}E8#WyJvTVcV&pc}HcB8!qZgPuTlEMC}JvcXBpq4^)YH3g-y?c@P=2s~5gsny0 z4Si@|6M_dHe8U6lCZOd<=8@UD8)n;@QN{*cy7N_wOv;n#BJQV;L+fe&Kojn{8A-N2 z8Pp{bK$W+rP{Q>F6#1MT<2-oA}ZL#(d-V$T!JUr{F)o zFl=={k{#p0+v}nLe+%?D)>6qwS2D2X(ZiV`m?1NhiUrG~|Ju_iHthe@U*)+O6qDSthtZ9u}DC(}c27e>?lp?X3-0yJYtKv_RUDL_7Mv2CLhxa1lA%yK zqoz%+Qp)5K?~kcZqVQB@6|S|ONePQuaaS$-evGUmuFIaHP7kB)4hzz}I-hQWGO6UR zBE!aTy3?3OGlvgS_rf#^X7@VT0b|l*86k0BA(S2yf=09Vf=i|gxQAw9_2zawtF#@T z3$@Y=oz;}s@C2JG?3^j~}j`r7oQ$x+mg9r^a2NijI|Z!+Serf8RzoT-dX- zJr|et|Kn~gsfP~BFD&=)8ovaxy@h}PX`~*5{3~m*EO-T$=Io>1;B2h1alpM7RA@{5 z3|jN07h9JGP?7vBD&v;1T}2N()`)@1I5l{%U7Z%$NRt#}0xz%=rLx=2=(b&poq6-o zF5(XOZPdd34N~;OQ=Z1xzQVZ6pV74FD?Zv5hCes2!ke+Wpl*GR`=oM$6STMEHjMS) z7UX%rPA75bK3&LNN;k!jr+2~9;|&7Dx7b|y$y8~-d#-{ z1{LJ#JC7EKhtQo?7m~E|r=4Tdu;j=@)baPGV8w^X=*ZY%FrQe2iq35Fr(?Z`sB&&8 zS^WM^9e1;7-Ls=K@8SX4Cc=ZaCbN(SzBG1*C{^z{h7bSEg8*N5swk_X)Y5h2UA+k( z7bEG#=TO6sLB>na#huaT@aa}l^2=z)YhwsUmjA?W4*ziVn_4_9TZuRMZ*k+nCm6D* z27`qB(Ot_0+YgVUU2yr7F8K5Jl5@CnTJ(#H12%dmClCn(j{Clc6A zMf_+g@6n={8zmWk^fanyrjf3WJ2i>0Jy7dm`Wljt&vh!e>E>UtG%t$gAM~LQ6|7s6 z&S72VVUS)TLuO`nZ11W-rB9#Zi_HVL#L9r&&o|NSFPG`RIpe9@^%LeC8B6bPThkxs zB6{C*geu4Fpm#0jLFKzV{+{v&!?Y_XbIMV67D%P={v0y9bDFj!SCQEqO)`JxM%FB+ zZ5vmDY}-Rpt9_|>t^r@-sxDu`Nr>P5+YZx@twA%{5j;>wIQHOd3`@1d&L}13ZBB(O zFLN#|!jv@XglX2DVyv6sjS05@;i7_-mj{$(uFgNtdVZfW z*gbf4H+9ZQ!-4NzDAyBBA6BW+^#|f~_ml;SF1boa`=8R92a^1`L;fVRYA#J|H=tJw z7QkzNanzTXf-8B2^tmdOREw=iGgJltoUP|tWJF+@yClVx-9m|)o%lOChSteh(rMmY z`a33q`WAko7u&wme7|`xxI781O2jB9GSc32Vp9tSUSA%~cgU`2Zx8X17v>}fbS~UMrH&$Fwz+(9@I8ZbWW1eJD z?C5uT<0!;mc_*6=EviJJXFFl~auK#$`+@Pjk*I#~6#djVMMv$UsdIS=DflLmOZ66l zV@R_{GO=TR3Oo)@#Opn5?!!}|sn@5I#m@#@^3#coWS22dcm`t%BvZ)346@hMrtFRT znSa)Y0^hLfWcn|dI(&w*pZbuKg*4r$`vJv)Qut`*caX{*rvF}Q@OvJP>B*G?t7lb`U~K6O-UP{yTI z{g4yi1%>zxFl#eh*m8<<${&XdGd{p817Tb>@iO{fXWXTQ4(xiLig#{h!|G`(pd&>9 zAF7ma-?w}*DjV~h6#7)@n&b{Jm!FNTQYUz$y$YOGO)dT2+(w$28K^pmd2yFs!Mfxs zG`lWB*I)?6`(#mGW+MglMUY@Q^SIA{#viXcz}I)b&F}w`gU5~*a#8CNjdBn7(FI!{ zlF2h8sbo9;r9fZ4wyObuj;#qNn;6L(EOF;9ODrJS=V~-DZ7xRM`oy&v3BwexiCnMy zDahcJ@RXlR@{0TQaH)+Gw$5&Zt{^e8s2NLj$1_k=jm^X^AH&njr{I~iCVU$1Z;+zJC_t%;JotUMy+ncLPSnT3 z6~F1pkQx<>PQzcE9vPX1!^q1>BO4)M{>87MeDi}b{BJLgkohk!Iy~S(XU5Fr%k&k} z5-u1|o?pfHJCMmgI$*)SpZSl-qi#H z2H)eKKvC*u-oA^)(VU`07I)*tM_9o)m+l_R=!)qcRCthxC0`Dr+tnEut2c)Hjex(_ zZUH~#`6B+!{Js3$(?a>7cfI+)GW7Txc8cO!&R;_L<}4q zQfaTFAe)JM;=T7%`TNZ#@{@K}(e`bdP*Y$fO0zwf+WWD1!T2tm2#p5!#M3zanJj(w z)FA$mI<{wUKxO%<5PCKkN8j4Ptqb*>jX@t}v9sXJNOzo8GC+#X?Ad!^4y~UPMjay3 zwDUv^YOWiiHQhGkF4d1Uo}Dyp%T_WGUQSU-`>EGBm^3ahM)rwaF*lUSmc1UFvIM<-QhmWP++zA0X(SjO^6 z%CY8e*WQDC=5fD2MT=7|w8Uv!g>kMc;*x?4)OJ%Np6?&jP$-2d`x3ar5B#|!cf`SR z?PpH*R}J^}cPRhEl1?geT}WELBbkG*jY}^C&@wB81M4T@tCzh{F8-5#NL;5eFC57u z;28<)?xcq^HL0(I1646ABg1Jb-00K;T<}mUuf5op=a%@#=!#mL(Zc>D-huYj&~)cK zd^w*E{j+Q7$nPN9t~!Wg$E(2av&Z4Ye=O%1cpo%m`?;HvLfp~2S5Ut72H0+m0JlFq z=;CkAHGX@?3EC)AO^r9*WR5kHEOi>x+K4j8>Y(0ZJUs~)!^uUVQ2)@29Iqdu&jnJn zq_d8`-=4|8uL%6^UOiNezGxJ!T{CB8@oBov=01}G0+2h+xDi5f{MB#W`1`}^Uz-3VzZg z$9z|ncb34OOX4uQD;S@(Xj9(HpU5D+D9idK(Q0w{Gjk9UZ(iWu^!0ESC8m7y!CCxu zHHt3mjq~L{GZ)~SDkanX#f-1_ zL5pWQvl3E9&Cvg(GTMc&qNwBrSUgb#OkI)9ujW&m`+YQCdW1TE?!f|+)%ZCv3Dplz zW=_bLsEgiwQMcRt?~(cZ6t_BD#Cn_SQ$p}nKAU+BJ;T^7$8p@aK=}dY`&7_jqz^s}Zz9PJ5`5$N=jgMl9aX$}i;vetQv282RBE3~Ro|rP z$hH^~-&aag#p`HdeK_P@(B{nxPT^UQDtS^8PI@#4BhD4mka-=2S(edfUmEy$U-UFxJL`3478BH&H?XKBoR;QCVp#s2ky8Ur@R@(>huq|d(0s! z9Ld7xvJYUXygChwt;Ze4YuKD}7ws9_2pe`!=T7~~=SEKEaQV0H@lF?hC6S@Al+_mx zBibflTeuDSJS*9mbQ+13GtZgKT2LvNh9VM{us(Ab_M0psf2|a1YI}}fteUYh-3P;& zBiBAG1BUKw;65C`#7h){!f!LF-LUnfHiizx5I8Y)jF3#v(n%`v!K~rUTena3zWJ zQT_OJbQ$>ob2>%&o}cd1pCSbs2o&NURLsDQF!dfvrbm${F& zf}FaQDUWBN&U4qv=Y26Y0_W{FVf*rWs1)kL0^JK(Har=lcQE(!3nlbUkH-YwdG1K7 zAh&O$4yQbU^}Q$P(*moTICXd~{?*$8eM)ZJ-g-NF>NA&)?wn0(M?=x*Mim4{Z{(&X z2-Bk1>oHe%0QoNzsQQFFNvN0MHkSncXd3Xv17-OY{*%aZ{|t66xn(45XF!whe#Vn4 z#ZfP z;m6H4p?sS;_qti07bO3U9z+MyUe_KJ2{;bA*?Mr_Kmekmr(tB!O<1wi9||Nt!{?w{ z(7M&jO*(jtt^_Gi%l90VvwupxeZ6F`t%Yp=*`m%WOFB|mL=u*1ls(Co6g(%=FNuDV zABv?;Un$Z_OM%vhv$&LeOP>6ev-|_Q&hkrzPxI#{eL+S0A(Y?05gkLE(I!9yw+VK@ z?#&kbt(PBBTTnEq9b+Dsm}C$LwcwV1$>kRI+3>D?y=Ig=WC??DMex%ypYu>;9ZutR z><*rUZ%VbX_<|W!o+;!esm><%t$mn%Hyx+XT{!RQ;KX@GHa+|_Q9tDGn~42(A@F5f z1A0!$fQesxV8`nq*#9;JJe?D{mp*#@Z+H5rlJR`n=S*NTCQaD*p@4f~FrKfw$%;}! z<`8elfPb&tjqly;&e#54j$Y}raXMcZQ?d^bcTIw}dt~9HpM_ZaH3nqMUhuxUPp114 z_V|9%L=d@~L#byNPkVJC2Hg41RjoS#JC_OL>Y8VY#$bYN8$;1xhxkpmrUad-lx;T_dt3>$FYF1S3b(=z;{C_uB1kc8r$xn<9)uDr@Q-`IZh2n9H-Ca*kA zJoe`u=d-JTOSQa@U%5VbI%hh__=d7M2K)JqhxkBmJ>ycvL!#Di2$og`n{yFdnW#T+ zQBySKRZpU7p?K8pF+)r3ShzQCFVBMKjC&MofVQ~7qx>3@KABFJVuk2zxiZ;aJBi+{ zQlN5B7sx7s_tm7HTj3kW^FA5DUAD^Q9!zoLx?i>#NeQhq8tR#Zhb$*C=Y|CIU1L6r zvNbgCtqf&5>tj%97?jm%LfgUnX#R&yb_Cy}#)fn>syd0`(j&~Xb`!6NOdMA?%#)d!#cT!3>};$ zB}Tv;I<8L zx4bfj_MN!O948UDn!5)6>keVtDQQ}w7)|d>ir{VaP52ZmjN*)MvpR;k^W>AzO^Cq% z-48H4MQA#B1~WT{@Evnch4fm2t-^UYc4-Het4d(}zgwtWHIL5Je8i+2ReE~Vp3c0TK7z763? zwV&AAkwLQ`zL>##uB|?QO<>ztwZ@ zZSd#QVsto{j^#;-(2*36rR$gz$~_H&G6&%El0s1O)I!x7_P;CEM5p)v@W75)^lf-6 z`4|>*F4kQzcJ~)3Ep3Oz_J^>stc$(RY$CVpeI$Igj2x~?kfGT+diMM>_0+cFP2;ba ze%+MZ%D$n3%W`@eHj7^C^kc>Mc{FUcfl8iCMq@Dp^g4VV@2l|O;KE9{W)lmR|F%NB z=2Q%lTuVPIe!=1yj<|iO66>3#@tyB${1@Jat&Gp!Tg^kq1*)jEPzuNU|H85D4$PNq z0@v*(qdm)k#!3&;o!5O7-TjbpO|PSvNj~oVT8NWA#M8}&-(-{dikkb*(}YzTd?kJv zd97mq2g7xAp!5kz1XqDY+9T-mipJ>*cJ#QufTY)6qLlsP(DZs33^4jjb$>b@?6=2C z|HJs-zBD)+J^~u8a=0({9I9ICP{jL>SlKw9ZdRPckZp3fKhlhwv?Uch|GC22r_5JA zW&({_tx7^ml2KuhaR4vQWImV{x-dzYKT}MUKO=iQYzvjchpl_?ked~~@(U!l`E2&4 zatwB!ehrBQ>tU(nHsY?Q(@UwX=~vDJ;N*w;kBFUWuxpj=NG(40a1e zaPLGNnxMxxYc_tAuvH#K91>y8lXR?BwqW@iS6uyrIci)Bxbywe;Jz`AYmt`caz!j5 z^tT$MKG=dASJq&6c0GPJ6~eq3f8ftiVa`15K9?+0$=QhB0*OO8=-QHx5_y)m!%qbD z_cg#i#RsvoPnT`Kus9?Lf z2|daZ87sfrR=0XbwAp5xsRLZtDZ* z9lTeMqj)kIhv862JWRft53-5J(7S9qPWgQc>n(XOPv|4peY=oLMnk;zpakpAHsiXs zA`t!H2IoB!VE*j;u>EQ-6!o76m$fO}ne}(M75&pW$w}grVp@d8^3C}6c?Y&IhiaDc zNtA;Ysu_2S{KCJ|o(Z8CG`s_2*RR1qp*oCKT#0s0vvIQ1SvdMN6xOpD4z(R4kF9Rx zWw?_DA6THu-RbBcItupBLvYgeQk<}=4CAwB(MY@`UGV6@Vlit}J{5(DiFNoiU>k62 z7QyTRSJ-WT7qjf1Acg_y6jk6H$wX9o7KCdy&B3U7O8CQB6BYlSM9rvBbQ@WUp@Z+J z`fD2blxLF4tW+3Z8wF|$PO^^57{;ML4t^d7;feQIcpR1l;{O#xtoUloxTA%S8mD8J z${FDODh9ugqIl?>93`17!5?a@hqPINu8vmWsyD;<>Oc_|ez=6)w}oiN*^^i+m5wzt zquGAy4p=GOgqe#}QKK#l&+VCu`<=w_UrQ6V8|+0ZCnu~edk&R9^)Wud1fyLS;@6Mn z*uH85p1v!AADhOYyUHI}JL@&vb7c(wRV^?s@+Bucn#TpLvxGP6j^ORMNIWJz3WMQS zz&JP$nma1tq5pfv#*xRT$(=xk>3FWh1@*ln@WF+<*u+altDFk7Z~TN7x(lfzFND-( zYq72`3kUq4qd3bVrQ6%$Yc&IWGEWXZ_Kt8WwtR4Y#(MXiM;H$y76XeO!gZZ=IH7$3 zxP)7PG0FK6m4CG-+wtE`R1YL^9^`%)kaKTWQcX1UobzX z7dO0^M8QS(xes$|IDIiym~!I=zR0LVO>G8j6)OS#gd^ZSa2Q0MCVDRELX7?e~K=|&sUOAX5w+&Qt1tTI~?KfWd)qRb}{71n#1oqlIUu&f#sePIcbA> zu0Lv!yErA8TjG$#Eq1yD;?N3FojFjj!PR^C%WdSW-v) zVtTban$8>e!risLcxst6V=<)C5?SVaxOD|T6er;U=f~L3W(VKO?_;r?0)=m>z;6eU zKFm_0Uj1vR=u|*<2TRF4;1n%RT!z_^DtKx0ayZ1rL;w4SaQwt3NFHj(-CI7OP&WJh zfr0e+b~?R3dzxl;*Wm2Ud8p3ZdSXX*;*RmBG4D8sdBt&XHNzRlDwW`m$nhjoV^4as zS5SX}IjrBb9xmzp;BJ?S($xA#=u;p}+Py^Lhd*O)m;`ZVJ#g%wB?g)9gv6M$ut@#} z2;7Rphh_gjL&g(FA6J1`R4m%Con?ONSTvKI2QTf@K}=;lc64dcs!i)i`o$ngG`Q27 zyg=4n$zx|gV>(`*j6d59aai{XrArC$H4MH`SX?4W-Uugba;36#w(Ce)$GU7gFh;5s z_HNfC-Pa@JUZX}Q{T|}L!SA?e4a@zST9BplTU>aPJ-5t4+53AvdF07c+2-4rAJz;( zJ@wFZm2o4DTewq4HA#poMM;Z~+(_VL`T?x-9`+LDUrZsb5B4PXZW&!~QemDbIVvp@ zqqG=l)<0I{$8;X%FAKc^!fY-Uwki=z;Wh4k*vXjT%W>u1QG8{Vji!Ik;;8U0_WW8G9K6$&qn955Wq@ zXez*i^of6-PBlhgqVG>Uu2+wnug|9KsX!lo>C^L9<`ln2ovaH+FhIxz=jW8;lY`8U zT>K0a946wC4RgWi@^2_vT8ED;jxnZ%JTJ=&CpYrHxbZ-2_4MdGmq>mYgKN ze8$QDF3jJ{_!MuwGD+iT0tp!cHOy(ozi*CXh_xq9X?gP{y+L z^Y+k>zOQ)X>28cWtc*fm5XNthftj0yNp_$5f8MdmROny{VJHVmV=**`E3$fNKpqBQrzW8AVc88>EM1HJWc zAnR==joe>Dv(Ej-HD32A;lq6@c1)v}h5fwUnUXv)=Ltsl^FCmUbTiJr9)^VvqDb0M zgVr$jSfs}fIK4+3^$X5{i_|3A)0YY_dc!!^ayOn(ZXPeot%f(%J&l|5;|DKumL4~J zCz`ev{X$f}i7kICNQ`;xo*Z>VBU%RC&Xw;%KW)=^q;HU-qBQPvFR3pjb2 z$A8(zyPnlbu6vVd@)s|&hzD_KeFOkPEE$;25E4|C!AZV_6^=fxWiW!Px#z0 zmjB#hm~js>NzZ94?Y+mcZ^sgFa=|rvrme_doi4@iiF-)G(|qXop?#Rh+~(OoJ3;sO z4=&)+C*H;{_jx*FJ9$~sT6|sQW&8(i3;A14JF{n5C&r99fW>7-IDZ9m*nTO2Uk%gv zQxk>wAwm&!#e;(+$q8JC@DX0lvLcpOm%Yj*zNr=X zvdx`hJ$f0-EE?l?M4)Ls>z&9)vG4XZe!OHme{FUo-){Ly%o(eI)*Eg@*1=kQsa6Mh z?T6u`s2y7V+6fJ>*TSSlTggv8h?cNyin3}HxR_jm+D%Jfz4{sQ+{XM)hRJlW!v$q> zf8qEpJ}ok|q{r`&tVaIg-UX*-$BF_EtVwIeF4YK(dYKPf zzwuy)wA2t^3S{j+){j++cQTF#!DauWQoGI-J7}WBUPZw z<^!r?7vNB^2*o^T!k~*{XszOa+x|sk%z0f5PY;0mp1ZjBybfNZdNbILnFsa9j)JF} z6;&SfqY#)yYtFyLqqU*2pzfsrPH`WI{7qdF0e-!0AEJ}Azam0#oi_Phvujb8SQ4}`64N71gd7BBU! zz-L)qc<5+4da?|G>A6XuZR3cS8duYno?Q5;Rs}P3I5^pJ43>o_K+eofP~P+$Jc`$1 zq#oOi*jd`v~8Ct|DC@ccO`5E=j!Li?eSy18GXjB`A;5Q zuB|~o#yfss$<7uT30&&pJh*nh9`{}iO&=0_h^Vy&R66*D@Bb`9XHRZ$p#Foy~kY=EQIw*2|UfGn|Qn8 zOL?-JyE#dZ|2U0OSy)!{#>j5{6<*4kP~Q6of^;S;8$)VkvGXK{`%IJ>|7032H=DwD zO;F|!-MveCei~?S^eB$iPDSge0-U>F2mdC_f!9;7^GaMF@CvGDb7sM*)IKOqw4xcW zFCUMg(T$w_`J24N+TEN@O&^%G#G-?@F3xpQK(BWm=&SsR_if@co_5+FZk6g=-q6Cw zymF5fypgb@a4!EXSIe8m?cZX6b}k3OqVX~pr5A*TL9ZbDX9O3a-Hcbv!cfh29JU1b z(aS16ec571o$1hc!>vIFIWzRzHW~vRGK9~!4 zXEa0cw}0ICRk>VaavxXg--W-HN26<-C`~u8!?C_`c;tXJExrS2>-`T>uH=EmfsIgN zNpPm{A`UW7mg2K~Zf&wPK&UJH zit2^!N2XwufUoUb2;bSw4PNQw?Q7p*@81z|K<_({YN1+_FRVP2ktaq3jaC`L-dsb zyc&KMcTf9GGsAY#Es0tbJUGDn8QR2!loqh(Q6~yr+K#pVEkOR&X713)SkAkE0|nn4 zZu^X*T;|6?F5)T2@~_Ol`-5fBCF)tGd;=8=&!ow$4imGZ9UfXM!9Ax1IBSI(rW|HI z`0cTH+>gzQf9@dzt@rF2AI7NPZ%|G<5msr&bB^JOTxmcVY??2K+naWwQ%@YKaMS2_ z&Uz}<%S2_~A5ht8!gecraL9ccE!h-E9deVYbWAx;=S@Xjdkrek<73h3C!lQ91h?hY zK;9?{(u)*n<$(<98NNs^HLRDlrUi?%0VfQy9@=g8vo!+LxiOkf&sa;&Xhm+fRcZa) zJD98d5@izOA?27h@JmbJcV08|9~_}_s;B-EN!okn6Hd#mz@S@Y71afF5Qwk ziYmLWW5j`Ybk`4}8U=mIkGY1kuAQN0Q|I%&`d0Dd?F!MhXa_n7uEFHSPMnhBjQuf+ z`0Knl`pL9`=+jmZV)@OgoJr)eYYg)`EyXgWWK{e99t#c1;Tk4l?I{&X7n@F6gZ~hD zwP<*<1DA8Z@$uRkRNU4>vtK8WW1|k`Zf)ioOnf2p);sL%7Ni7s#@oF<0^@r4ls{UI z)^}JBtziZ>zpjAA!MC{OuE%g$1$)jfo! zwbcrTd+*^6HFbI&AWpfm|FA28=!@-LJiH?x1vMLS)m%3em}h|UJfr~g8kAwZ$}Zcp zm~XWd>*vZ~wBUava_2ZHvfk3`oj`}XbtuRy5szM~fZ=Wr=+j$=AJeNK_Vg#P%Q}JM z+Knmx!*9IvUlDG*FN?{Q_W0;$r z`2ofR??SgQTcc3Y?kraPn1YCcKvd?QG?%MB7`3^jK89t59s^u`B zXAu|&hGX}%eB9bpj^CVwsIEJT2H!?gW>x|f|C&wB>p0>C`qJT!aM=ItHh5MV;uyh; z+ziJfob|#?ShY?PmJCRORf8!A2i$|69W!w86A$#77lr#(%kZFWDOQzAqjE(KzL8@) zm!nP0_q>yqq`pVTL)LU;V+rvqOQ~Rr7rp1gaICx!L@$_x4)PMD7gR(?yyLLwWyhg^-k3HaR*akofcqFdBIlJ{HbL1?AHqmm^Nnf9h!5;WKpdk25(l?%=Eyf3a!C zVhR}6Cz*HAnBI1ds`g3n6L0BY+&Dk@JxPZ5d%H3U1}~=@zRvmj6Ozz ztQ7dQqzTR%7a$N=kB({NKM>;ef{KDzE`g7YK zh~tmqnHX*tfj_PS#v%IxEKM|Cr~Ji3hkvEF0YsmT;&Cqpf>GpKVr>qTY7ajn{` zp!%}`-1yBpCHXtx!srWHRZl@7^ce-qTF0My}GKo>gSgJRL<{iZN|b zF_urvhk;fv)+w8hX9MloKJEb?aa=-$EA7G1Mv**q75Fmg7N~e*HEzkgiV}-{u(^O5 zyUrQUZtq5vP3i;>uM#fPMF5L;YmkIkKIvEcQq&ev(kpUCg{|=*HQ^xaST+_Xgq%me zMju-HDUWjGE7330f(mEGQro5<=yx`pTuNK$s+=0dvvcF&>~7}XV(xjCt3R=BKUK^c zMA!9;YvWQc#B{RHo?BJ`iKBK?;#f%rxiByal_pYFPV zr+O4g&&H5$pWlS7#dFBb{UFt=uHQP6Br|QkqS-RI*M_*k?14)A3aemqAAaU zQH%AD`==F96oMD%cxM25cs) zn>CJ~u2@1|qc`z_hX$6hZglTm#=^BIqeCwJsNm~}8>1dW#DVwF+Uy8Q%TH41K~er~ zv0wBo*N*hMM9AsuZEQ0hz!tYF_|u>hu5Fme1?ciIo3ST8v$^7%UR#=;wUQ=Rts#H8 zV%)bZm|J>{kIN6ehW9AKZ8{kV7ybyspZY4^!s!!nL2oYOOKgF5i#ibS3dDe)y6fQ@3;mAJ5RT5) zqqr;=d|zD!dN2CmL2M?yyc0|A)5>Urvk4vD{T($}7h8(4mE2BFq+M1@^h)L|bswmr z67xc;zGh92d}ou&R@P&~Jm_cKyscLvFnDGcoox9;H~0(r5~Chuwc;~cWgNp687FKS zm`oYVX3}T(i4?WVh4oHb=z*sJ#)Mf?s-6?IuwC|>ZEw(NJ>zGpMN;s&AnFLegZ9&A zl7w9Yx}}-nCx0o@cqd08Ycwe4>pi^j?Ko=K6kw$O5q|HLNWRkJqx{_y{^2pV8g}L> z#~I!&AoC&-eUw-~chdmuyW_$bMA2vb~`3fVD6xzCrU?0gfBEr-mgRVEK# zT@$7ik7Ot*sfE_5Pvg7w86zxc#`9jnB%E$XN#~PjN?{4v>dc@Y|5#shbR|9zDx!v3 z#!a8LkPhGR;u5tk(V`*aBY1pJ6Auk0;GyJ6G;7@i zTAtU1J>}kI&RF?H6dPko{KJf~mArti9ej$-V&%AFY8qOFti)s2D&XbMTnJ$O zltz(xdP5Itc9)`_(6>c71-otPV&~#)G=clX$P60iK+lC>KnzA ziZv|Xu8QJ1)zGOw9#xeFV7J9HUfazxQ2lEcBzJ|wqF>{%YhxX(W*wku@6_3Twhzy0 zjiSbp8Q56d1RKJ_QND94Rh{UeA+L#idL~Vu2E8fuKLcth5~e!oi)wXf0yvctG#yaK9g_+es06DqM;f$~>7 zSQL2-{)yZbnNvV^UX2hVM;PT+P*~5LkAL7v%d6cC!wAe&#u}b!tQP zr?FV{>K(>;4?|TeV=y}1#stWOW!?RI|L|J#Q-dBLpAd zGL986B**nps2yvAp;60GKvj<1i%+5H`cjDhunhfu$J2braLRYrAom<6yjP`)!41=C zYWp5?Ft?_q`o{FcFP9vK?^CEq3l5Yr7enVG=z5}qnZ?G?^IHHeu4zPb#|!v=-Bn!B zcAWn1SWM%(qtKnFLuXlLZQI})`a3XzF6St){EsHRt=9pkqb2b1j|XFyh7{j>z9DW3zL&Rg#EMFHF$GY4-gjz@|5S!8>14jr+!B<{!seAtnP zZu55G?af_Kop~8Mx7@?j#ecZWrCH*x z@mQCBD33IC<@hrrjQQVq2XV~vR7`YiLD|QVbXKAqt)6b;uGSEoj-CeVBc?#Q+7#5)UoENBic>pUGD!@%!TlAFq+L+^|Bs6EWU->ybk;vbrh$1JCNFz znWS?~h0X~YP>K5woSAqJrK=rq;PwXGzFQXEUnIe;t9!w5^a7-enSvL;O5&jpz3{qB z7e|-0V@2(MbZgBCc>0!Q^+Y|O|78LF@k*w~1IsBWKN$lh4e3VqBC>XwPTMZdrL96% z^l@k`t}-6k4YLz6fu7p`JCH{XYyBJ_*qwIoZ7-z zbT9Fkaw~J_J;i?$m8d;69v|z+VAtqTTyt9!UEfW?LtzW?y{Hd%Fdn>@^E1r6xRk!X zAzE8Ak2cIyC#mIA=+84g1^Z8->@lej*j)o&bDWUVwuf=c(qV4B5>7a%fdhZ9fzbxU zU%|g2SNSjeluLr$)8#nP?8ChD+Ei|Si!n|)v>vP3d(qri;arLx!eCJ)j*O_I!`}dG zdU+FtE}llQpeF3CeUEviZ+RhpdTbv#6{=MP;qSjQu%3B=L|K2NFR>iw%1ohW0m6K< z>mR8gLz*AlR0nhCJ;yAUl{C{P6?N-@roGOkEgkllq{i;|l}Avr`cKKpRMV#wGm0x?xL*lQi@&diftz%a87YO zz7_?#JdLp}t12l{#Rg?6M(|ykIwcImV!Xg@bUYwIM*9n}dZj+~Z(TrP0%BZD_6DfC zZOA+={h-O5JJx45vGN)EM z{~edY_RVErp}Ul7rmm)5Db|Ek&qkl*COmgrg=&I-qf6x}wzIs8GRlv^rp*O6Il7?a zxvj8XAO<}@UBSikqp)Ch51WnsLm^)=>X|4=&&5SZx4^y^ zp%@n14q9(wkay)0KDzV=y^MEI(Bs3T4{LCxKJp;EKEXQ+fe|Tf`07;J*;rFc; z=4+qT=G&_9@VcELRmCLJSh-|~IUE4O8P~Wm2f}d$;{$KVVDE8SOEIxM0>y3gaF^L7 zsJd;2AJ{oywCgSYJu60r>*vs_Z9Ei`(ZdlpHw;%liw`^>V}d7BS*e9H4|+9vHhe@) z(@W5p@Eksy+=n=6Big=YI{m2dBJ~AT6f60F7P`jMrsZr9l%-wx47%?qTcc@Ie28S~hE zE_hWwH#cpVi`;LDpO;3`s;wdHd9{x86i(ym8B<8S14+Ur9yc0jB3C9_viT_BT3|(H=wHqA9RfdOb+QJ%*18=6KcB6OZSA1YM&# zcu?z#nerC6V`4pyK953W*Z25tvm4n+#Zut=L2P`dfxBJ3@Uh1Y+_6Fq%TI_frk@Tj z4f_a&E2Pmh*b=XPlBB$SQlzvmg6+4RLDr%cO7BUa+&oo0wEq-0USD`iUkxw80cvn`r_cf@Hcy|J0O zkT2ZWNz<7#8V)li@#I4Ebd{t}0~5(d3F!Q8Tlitic6^m1(9fyS(w~;}BVspAd|m|^ z^Q_RVw-eumGyZ?jV)|MdgJ(tW6kpP}>c z%ef1~c!QRimM9TfMMKZ;T=$a-4c@Y{A`K%WBO^OXlSypDXDKgqzmdGaZrP-dIyq% z&JptTSV}P-rR<+Z7!3N)rK6U9Ow0E>>t8k&Kl2lDvmzc|?Hcs@LnQXT8AQp8OZj)8 zAA7p;4Ep_LWHPUjDzD3lbt^y7yOmxv#X~}DxHN)R6jZXw+rE`Bm|2eaAs^6dv6Z$?971|8M$`WINYddgL z0#{rnzXT6*DmaX|JRS0i;_vhJ@APU(Ke3N@Egf~XXUr;))wd4A#d+2|C-)mGEq737 z=V8|7Vu-?jrr58}y)O~kv}*iR%21GFr|svk1r>eZD!GRKOt7GzF~?xr<4bLFiz#VC z78y1AK^%V_o_NPRqRunL+U=|?{xNff3=;Q9poi}Z9k)FcG+SmNR_z@wb-co-j{c;y zA&kv8A@Nc7M*8owBpU1jm}~n%I4CcpQ*E=zIiGvi%3iXAztw=*1^78RmUa!4;kiIt zYFd>;H*<<;uzY{<(sN&EeR48FJ`AHtgML7RcTM(6`5@lwK4hM1(c>=A595tgJ^5ua^0&orOQ@$9l5+h9zv{gbz$&ax4Z{w!kQ<2pj%-;&O~TYW}H1R%tVWV>+2s zvL$c$eC*mHO4_-Xp0DLhr}7N+$_7w{Z9Bb6)Mab@ z`tdtK9#cJO!DP?(ux6DY-2EHL2J@V8#fMR}MuxL2hd7b3ET4n@D-|Bf_CfG)Yf3$p zg@`MQd0efPP7ZLxUprfL{3l|=jis>DdWh%bR>Py2xM%P1b<6}(fPQO?AQWdcEGEM{ph*_C5IFg%uhhbu6$t?Lz1!W zSqQywt)WcW5Gv*z$e%(XvFTsP$kLp<0~*OoW~_MdgDf0Yn1;1s={Wy4iuAto-re;Z zl>Vg&;g5r9^ih8@JhXsiJX664HV4Uy?YOq07wSG@EE>I@y*s3iF1e$W<{E(E?Kyas zIEXGvtCH|~BT2xTCQ2$Gc4dEBH2D;5DO95C#XBkJ(tJw4b53+)i7dJcMiY{XsrgF- zr8!-t?$ThEJ%BS>3pj)DW;F%fts;+lX|YkyA@T3UIbs`iYq7s@&F!SUdQa$d z=YDeEs3W$N`#`mi#!$(XG~sK$3@d+`56e?~=tzY-ebuugh4QCtSK&W)zi%Vk9z2IU zRnO3@s%t!ZGn6hjjHMmz9yH|6DO{Npfs{nfJ(Uj>?u+7>nrt~zZyv#u(}Qs&N1H-B z^l9wD#r)krm{r~lMD&C?tl)?OQ|p+@I)*M|`wT>^@>?*wzw#RERZC&tO2(n#%nbOp z0;f&9@GHI_cDD=@I!ihGBVS4!9@j$Vk{5a3#fq-D@1c3ug9(E|$zSCS?P<@Y=1*gJ zH!=`^J~@!ytxyV(K1hu#vJk&Kk{Um^)0gNP=Eq$YM)CD5Cwqx3s^8;_7*AvGIs zOvxF6WdyrPj%;_>hlqf8la=d*L-GayFi6!$BIo{wOWo>_-vp-Sk1BgR-S-X{FH* zIvn2D&e%i`Q*U6~7;lB9^)ti3X226n~B5 zGxK)ZxAz2Hle400caBrXvnsNkPn6QcxgMf}H1Kv6UPia#@ApHHR5#&{j+0zT>q);G z&QRFeb@);o3pIB`N=#>1bAo#ahm_)qOAdC$g+fF799FOL6kE=EB33SH7gxOIv(i5+ zz$;eJ?Zb1n(|3{BkMqEJSHVHxysGE*u&FMk?xwd?tD!8Gx~D{U9yY?VZz7Iy58KYS ziWDu)KiqIR@vg^q;z@ZL;-h1|>Do^}a#9Ya(2{OC^)G{plU->?$zy&84TbKdG)yg) zr@TTZdbQ#>{Ze+Lk(n#$xq1RU_3Fg2&8g_)q|9dV``U&39`sV{4q1*kMWq{T>1X|R zx_B%Sb@QiS;pS7Uwb6=-YfPx$?o2YuGGO=hH?XG-zu3rNMbh&cPru#f(b+{_R5HYr zw*8wyeWy6k2bo^zng`3~BR;Yp7g<7l+UJAQ+g_tb1r|vOMJddD4-NmNjUp69f%J&80q4Rf>aTA+D;!wh4Q&+9?K2BR>d=FMQa>NekKQtUat>OgS^M8Ucr? zcW`pvTB`cwM(a*Za` zi0u9!|noemXDHjPqqm^l#lvvgR|NUP)QH-fKz&b`BuBRd=!R>p9eqYe1g} z8EQWM5mnpHvM&einQ@)HaI$VC_NM-3HtiGHA=7oZ<)Q%9cJ7)u?FYvlYoPV5iAje@ zk?Ys7^yb1?%4li8jkzkE7wJKglQc0(tpa7Et;oD-2o^jwf5?$H(1(4Tb_Tc2NAa1R8JWFZMSsU6(_+`b zbe4|c=C?zL9RD3b@fDa{EutwE8|d@uRJxVZfcJIRu)QrE7HczvB^h9n`HGPIu?sy9 z%P=aP@Ay96pt#;zlG>S18|&u^h7pSpaKwPh=qdFJc|?aZXHrh0KDLiD#5IG<%)$8s z4c^~C;fF5M0{*x4vs{xbzYnD&5kt_W6Ud#kUs2sz!2*2GV)`e3NB#JWrCqs$bKVgo zHLox9f5_wfrfpE($M34%e6KY402Ck0!dp)R9KYv0~o+<%Ud?uzt`^F{~k z3B!j?N1$cq2MN1qJRfitFX!;NYT$1Cm6}7#L!&6pG=q-io<&E3DQ)$#rsj&}6tvNg zp14TT-ndQl;KOE$zPyj-f8lq4hw|hcZ%9%D*W&IYE#%GT_v}t(^5V?b{h?_*r&7n1 zGDg8;^K@2tg6C=!Bq(uXE1upJ=v2ixTFBX$ubkI#-`RWah+0opa<+(0JIwDK!7$3a z!=~%3g%eu}iPyp8dO3uY7DUpr0|L&!+Y0?D@+3dtI=z-G<@d=X8kzE!G{i&2rph^V zcE25!Ee)gAxyMNAix=$vr6W$@yk3V5R4#j#LTjE<%3YqX>k~!I@^eV9u>p2`zWMml zUtCxzM{51lsQVrNZm(#^qBZH5n_h{4*$R|>W;i)?Ytjr!8OS&*;epvV*6kmT0rUOv zLGwAj4iRYTa5=hkG#J`m@5z2a9-Zay{FCLmLie{~c31y7dpW6qD(~o%b!#?S7NoEy z*V8OE{SGs@afNz&>bSR{hRR-)B4OD>9JP9kI|iA!v^*WbrZr48&PDieVIVRp9wO`V zfAq%7oh;_fr|dXM`myF1d}2D-tV%mn8%@QF8M9eR&nsbqOFD}RC}3wptys5tG&{#V zDbY^z(Ow&j4ay(zif0JYblTYWrW3-4C5m)4*B#3)8e>OC9;?Wzgk`D@xkh|~-Q-5} zK9;8y$>x|>z6g=7oL^@=4;i<3Ze-dsoc!8{l1_9%;Qqb4C62f;@dQNj7qGZf0{;7* z5Pbe1*w$41jLpVD-7|>PbAs^x4EnTYU|Z@roG5yPEzaDF6sv^v=*_sWC?1#8Jy0K8 ziO>UzG`~a&qDQA#*GqYLpZ&`W9rRe=o^9-s&ko)hTE_4EC8*>a$T@8Z{NLlO?x?en zt3D4=BWEOh^5S{hGK}CIlVPqSFh@@bnrqbYY3U3s_!)_%Jt=s5CWc+vU&Chj_^_HA ze2*~16y{%paFn0p8{H!@wt5NXEL((1cV&e7)Ufs&o7sZ=7&cDU24mA>Q1LDi&e`dx z_X$Dj&;01q?&;0^p zufAkOgC{{dFB@9B;y_y>;m-cDxlyX<{h^Ao^AA|_+*>rh|VcPH^(K?zD0 zJwe_10X#GF9phir!rS;NKK5(Gs|j`Z_4Ouh8gIl>mXDot?!eScg7$yW#YKKL}4^R-?Bl! z)A4%H5Snc@kQQHYBTLN#c%XP1-!x(nwqicbY4)crb5~Q_vncX-kwx>Log%f2$C#oc zNmC?LXqMMN=x^fOy05LgPoadDhVP-ArAM6|N0Dh%4-ubrzT=J@+o`Isby^J9n0Uxt z&A~RFZ@6pkise&|z&4`_pO-MS`{Y2SuR2w4zQOia$>SKQ!L!#NS6b#Gy!94q9>x8` zpH4vgb}$ZGZXsz)13IK~lw_aev6KjV<}l0&?vbPDUd%{RS#3lwP0us)!|^QcTLxP= zZY_+gR7vSSZMw>L>-M4hapUHGT=(RikD_|yeEy6eqrc3))d5i&-|;Op33+8va5k5u zG3nlrc2dG8?HFX9R6wE5UFPt|3Ae{vu_Nz43eP>~v*1YyY_!OaZ66TJT&4{{b*(!p zciJH0)Hpf>>UVFF?At~tczi(O-KX3eUWdts8Tj_pmvq*(P^9Y@ za$j0Ud&8$v+s>V&GUYEux!z^P1rju-O-fwy!CGt}d_jbvBJLIa5!NY(QOd1>;@Z$| zQrD~#R%ZI*+nNZZJvjxLj!f3D#!YaZ;g6lRav0FR5>Gb!W29yxdEe9%ml{;lRh>Dc z!uv^b#_H7KCQEhw)`A`SfW`U)X=6hToSO^SlIE*S=3E)GO1CC0o=4i%ZzufEufh1y z8qj%Z$X;gO6&DQZOW@(M+vB5=VFLx5Ap@aEdI*63(uF_vO4f3IK*q`14 zMSm$8RP>uNkNb-c)vlv0|4Qk^Ku?m0vZV2A%dw|djD0%Oh3W~5*@>1w`gb=E$?vZs zR_PFq&P^gIoiIA9s7oeC9fcWu#?kit72STAOM?rf>3rA$ma}3n)k<8UI!5&4mJJP@ zF2?nbzl6ze)VT-P0)02u2-}@!phxl;N)GL2dqcHZ>%xQ339aK!YY%4cYe<_Kc8QHw zhT(`uFRFhl(EMcuG-vZ;`Z$!k<|jr${&O5AwNB;z<@-#U&r`Bv`-+E1xlvk#ERFtZ zjjEYfpt2zs%_;97TBSuFxs&8(k3UVjmInQ%(NWPH>HIrB>R1w1Jot?6U+3{#Gy+PSHBgrB%dCR~#k5;RYKj^d5&j~ywT$%M-h%E@LWBgnQBUi*@gAuf&2G~J47SsnY<^} zR;{7192HWX8%CQv?vbBF9LcRrC5aV3X?Sf5{tU^bo8X@%^#yBkEMu2gs!+Tsp2Zw0 zz?+io6mmdXytpZhiYI&H*4fEoN8V>M{n#Yz5x23Q57yAwE;F&(Q+d))S`HoCo6KGD zmMBMRvhcG&LS)$?OAGm)@?M7oEc^$OUsVVi`Z}|{!M736(#D;DpRqA7NtF0GhMlMs zsAQZ#4KhBoRf6B&&z;B74}52l9>8)YW@3(ACUg1}%*L6d3tbbh^IeoG+w9rM9JIc& z*{kO>ts@06NUFdA*=m+{IvIKEcB6CDVAMT$NP#(bY3kw8Vw+c*)IR7bEX8-Qp0m2v zG}Mq!R5y1^ide}6UD!tQ9C*N8w%~^hiDvnc_4yxc=`v{=(PtiIcpqhIQC_&PLzP00 z4u#^lV(2`}L)+I^Q28>QvTbdMeLslg*_SbE22o$@p_F_+ie3)CO42RYxU2COc9usX zZ246-+HDd|-X20DI@h9JzW~Qd)TpL^1a1yo2{}H`*qXWo>~j>fa29`-J3{}$V}7^V zPAgLKk?5>WC$9KX{>LuX5!rze+qTl0(hhcEzC9jr>5b#D2ci>NovfPA#vOKyWVbt% z5s@|q=3&k#-(HV0$24TjYhdj*yIEaw5W6k?Ll}Mh9y1=encubaux9-L+{@E|c|$D9 zgLqC$+K6%$5;2$imuAWyqnVeF3Na_9rfd?e_Y@CiuU>+um|4yH|+eri_yM zj}b>+_)1?@WW-%na@ZT628x)J`8!}h{u*bdbBaA4Xt#CO<4xUI6BjjL>qjG(xu=r-_W!*m+6dkgqN2G z+Uoyeg8MzxfBJ$@w?5RMw}9@s`H*?v*VLAwAZ}bWLp*qSf2Q$#E(=wtL|=V9ifLX! zYI4_TO8*gT(a&j2b;cs*eMKN?{jtOj$kGj`MrN25jGXkx7wnBJP)< zQ0Bzaru*alMnkyvzQtaJoA|eH31P=%n(!c+wsghg>6|X?(XPY<@0-HX`T|xXE@ydR zW10T1%`8_<27P?avUAZ7*anv%(7gp)D-A= z?O*Ckb&d%*H!hg^xV2Ml-69%)IhKOvb01H@LGpzkT`G&A*Z;J{Ggs=0Gp`I3my62JV$kg`t`PcL{M%qO&Iworb{QWjRC%npCD4Oml5Q zY3|=qq&&0>4=(p&$!dk%d$_oz%!o2qPon7`va#g$ z0h+Mt8s5rPz_O~7jqy%_#VTL!U)WBo{#aA7_9Ds;I6;bSW|U{8MN_xM(&w)yNk1T; zZf2|^VWWk3&w&*2rq45wU04Q{ghteCy+s{{d&ysI1ug#D2d%NIu<>yi&ao|6xK577 z80?`(al`1{R3$o5#ph)_4{#%C9vMp}($^IJbAkiM>zt=+-L05YN7S~&k>`e!skHPF zbqB@LJ2nbVceC+W`Uk?3uao+|qZHM&gx0&3(eE`Obp6bF69425`(u76iL}I&T3zZn z#^_o8VN$FLrg1(6WH@RPIc`ot_(KiSzQH|^HY@0vqzQG}s?sB!BECb?C4bR2@_*$) z9^on!6(UW~Hy@yNY%YEB)T5l*Xd2GlKf2!n$-A}^bs8CZ!@NE^QKbuJvR!adz3rnKFx5Gr>YhM$_%(dd(A@WM?xGW zZ9U9$7x5&~J%dfC>tdIa`mpQUmeR10Knm66JD&p`$o}$`=XaE8GbE_!yAn>?=4IYYO! z5~)!;ga&?mKv(9zr11X=$jfXRUDdjVdlpv(B~PAl>!ZZEiB7aAX_QzcxSx1T$a9*( zb1Iv1b1))rk!bi{8L@r{_c|}!LcXtE**+J0VcUu^tg}=JQx$w!h*1htJROf~3xi?$ ztN{HIUO>N}Ki>rga93_PX*n~x6>o@z&-7@#gD%ozHRhSk$mlnvSfa zcMrcup2xuY6ZD^z4sC6ih{g9O(%I6rH2)rF){l(9=4O2assCXwlR!Dyfuy#po#u#j zc^5UD=IGp@cYZpokni<&%=ySdEL6l}9}W{yN}S%ODCx~;1DSd*vC>2x(kEc z#iIFX#jLkjlJxu3()%YnXgEJ(g)xdW@AqKRJ=4VQ*iUB5E|t=fGkwJRelp^s#%u7o zRVIwQI7FB+QGx6tG-$D=9zRFxkQW11eSDX&j~SExNeK!|8%N6veQ4uqH=3n5hV+|t zi2S%`W0fog$(|#bT34~(@Fa1%!Y^2>b7!hW5F*30XnX4@(r>Lqnn5~N?y^GKC!X=t zy~=`jXF<1q0C$p@(209#ur*MEV*Nri{`$pi9xXs&w=L!`w4rof(Be9lw_6I8y#MoeWC7oYNx|V_1Kl|uM`zcLqgeYEcD;?C zc!r$$`Ta2+=RB8%HJ@PNt4P}G#?YOx-Q;-CnOy75=*w;mYSFW%fQRYmE)GPmE=dR5@MjRgHo`@#G9ilijrh>M-!EZt}%O-&j~V%d{e6B0w8 z%5_M0@pg>6Gg`cJa3k42E5+YO;e5{g3B!-6v0J%;>|n|OR^t9hs9a%0@3#!V{>}+( zwv;_(ZoZ0?E+0te44@M}S6SitQv&2Zu;XWnFhP9^xu0o9%iHnXo7RL0{BzvtvzKl+ zZlP(Xb6EJ=)$H!T`@*D&RVZ3sgt9he(z#v2*?^-FBdfsHEuT;Rdg>6ZOk(3o!fEL6 zcv|o7PVJnDuwL;IQ~apNYPJkuS~XMHd!D7AnD88%J7Tfs#vV zj#Q!d{B4EY((1K=fmORH?kl!uEhCfKfe`E8o zH*X}9U%}ZU2{UNws-KuC5y5U9Z|A(S>%NOSqTEnM>mEr<^!ZGMI~%Dh9w z)bEH%k)wrw{vxB2pVia)(YMuV;u0Ns@v#0K-1Vcx7F=B-dN@K}IQhc|`$v{Sb*n$t zyO@iUmrIM~U-cK~l_d$L*)2lGpIl+kxJHD0ZbXnv83sNvg6~;7Hu+_ZXtbt2eC=C= z)d!{u?}v#|u3iQ|Ya_n%olI9hThrvbR@_+(Ds_4e<-8;ue>{j}R+n=AY@25JMe_atu7QaEmm>6uZu&44SRT|Rz0vF%N z!F^XJ8-606ZKwNW>HUfKauaD;$6)#`8H*j9|C(c}L461R#`Ma5+-ZJ>&1i{a&Y$+N zxJv?k%<`qThaE|0bPpV($CAV+Co*k3PLAr{w76>}{rVh8ug~YwGVM4DN{pv*R)%z8 zO%0}9+s;b%-V+{wJ0h5=+Tr9sH`1NaMxV6PU?W9d9M`T% z(aI<}s#12v_^)L&rPG}%B{tHCZ#}fgZGrgkyFl?OO zU3hU=Q1+2UNW@=Z@7TWZBJ7f%uL+XRyrZGx4&Aw4lI>CN{WIphp?q z^Lk08;~vqE@0@k>Z|Z>x(O8>?ZWK$+hO}=I$heRMyp*uVDNvs)P1>$`%$=u zrg9H`?Gb3&>?ySWWIH)tZzrWod315j1nRXk<}8f9WNJB?W^Nk`*V`K~>(~+DPpA)5 ziPL6o1{t*4!6%hkFv=_zJ)g~C5iBp>p!1Vf$~ThD;b>HannOm{eTvwoI6UIBZGf^q)8mn`+ZE3F=Og8Mt>(5aFT%!po0lJ2UM zbH|>goZgSvo?(C&w1u{sUge+Ip?SJ8(P44EY{Y-iINHm7Ky_~|L`BfR?!G7kwF>o%b` zo%ez>-SH*430>y>=wHYv+B)VMqTW9eI(SF5__U<>M>>C(hJQx;EmcfDc7o+*awpOq zPG^V@r_}}R*iu!-Y-738qN9bnOq!|Z1$PPbJY$>v^^vLep7pbNKt5M?)5EAll+3nb zV>67n7mNG%F04oHR%x=*7tx1J&jnAjS{9{m2Zeck#I15;NjqQ;6|Mt5$$nr4+(TQt zU@{_jmp~=?3LCK294^MY+27~2f*bGY>`nd1288P4xAZOiC*6q0qB+pzUO>^jiA-bq zSN5Lg5^I8EiJCrOQ~7i>>=;W6>gUnM?~7r#<0t-^_oe-{73|T-7YI*$50|H9kUSzm z!G}*`IPb1>M!#m7&s{OIID@sFF~WjPqoF=>G9=^|!q0pm`u?e6y{%>}|MfB^5p@vr zR)Y){*wM#qU%2m{gv#4$EG%gV9!51|XrKv|zIG9YtVm$PF7W&|=cz9)60@G8<1lUf z8e%%G)bz3ei8T{p6yL*|!V_8G1P|J`^bIxs^&!Rodg#{4S~_&~4SbK!CF%BkWE%b+ zUH7xuAS)ASsmwq_$pUI${*X4;=F>vC-E4}P1CC7P4)Y_CFdbC}BQ-15FThwh*Eo-z zo}>!%J_P~|J;zG>4nd>abu>TqQZT!jrOMj3&cSPyzDWJ~4>vAv#?l>puQVZ? zj$e$!6S*#I(=>tp?nX8@DN3le9xMvn8ZPKnxe0l**Rn|4Yb-eAI6I!RP$<#loq%O( z!t;W&>|NV78kGH+mGwU_tPJa>QI+!IpezlsyWV)xX$_=lwb$vx?lKx8d7g6r-J=Dr zYcQH~&HqerA(sgT$X+BvM<3c#*6Q)NU&ChzuOexU+<4e+w}GDAU^LADJE zG`)lL@jpN6%0EWtPj;eptRk(SZbPR`t&wQpiTktkY2$@7{7mMY3CSUp^s*W6&W)he zsyZZdJquEq*J1r1Kff)x*WLaL>}+fBVdHEpSXmFZTjvlTT!0mw9wdyarX_XLU>IVL z0}*9x)}<><|KKz1FB?K0QWh-X^dPqKWGb>!t_UorfNeYz!Mv?^urZ4+u z!+ZMT(SJT;j^_~yvnru7%UJjPi6fKZ@#^<27XD;AN)zSi_2?*!x99G()zVb-cZ=|LcM{C= z+bGp6m`To=g+Gm(@cEnsy)5D^_vp!V>ZdIB`*h(_bv!o3Z(_lZDzVQlmu7G#j#|bBsWTz8&pOS_lQ} zk&tLI!F0(NIG^rA$sYnJY;pwl6db|if2VNpQ6b*0vc+A?a|p@sgXQfHC_dYtUVl7= z6TU6j7#s}GEw0eGaRN7wO`yp3ZhVQ^N(c9I3Q|fxX7$exA%|DfS-mV_?(K=(A0>;I zrB>M5=Or9(^7r}dC|cjrN``(Psh4w}n?EIE2G3?BFUrNDJJwV#l};gtPLu3HU9m*R zdhs-c859y(N)5hWsqSkc-{}=%!KhT+kT*xeM@Q-s)e)<@Kyv^6qVf_qakgVM{`6|m zxZM$ylJ1Pna&|PcsEa%vq_Mf%zOyl=pRwqM8HL(%FUG-Qn$U2Y#mioV{T?wz^W4_u znJ$Q(5`b^&>nOM`jikNm==D8W@uhqv@x|nabUAtjR$crHcdfnTJb5n7IOIdmD$J-# z&Wx?ORl{y&)w0(e-Ix^@L~E*A>9T1n3v<2zky9(0#}r}v_um+B;65x<{8)R$XXd{l z5mK+-Vp)_Exr`5|(mqS*c)EyI-10+=u{1?Ed_b;A4s5Mo30az%Y+avZru;mSRb*>H z_PPdcndC96>_Ip?hWFaa%dq7BPqfM|Cx_k>6fk*!c<3HY@sty_h?h&toFr!ZF9#DRai5{a6*TrL(eA;=Q5I)JYhS03{sVuOWoZhtpxd}S z{5>xGgQ_J&BvB&%6txL2EEHHIp+wlIZ3A z65KSMkG?7!&@agk?nma4^Wh7`0-ETSUZgo?cN)BaP zTuRulh8ygMRuda;XNkJ37#y4F&qnlFii(~pI1K5BXeSA5^YMnds=(x}Z?eFq3bx$v zAk2El(6BTQQq7Ksq=htXUm1u8f(A{uT0{q@ET-r^<7n^@dz!TRHq}-nuyUDmNIqvm z9u;z=bv%Hi?T_+IzBJL;c@%IZmS)?Qq0qDyb~$32z1Im}9);lf`9>s+?uB3bToM;V zkYsu>OXwMiTW${g-r&ykHpL^-ybG6C^`)QprqcVKU3AUchmB}iiEG~ekh-&lNtFF! z8+R3;Qn^P+dOwfl4SK>xIPt8=q7(G4?*lSRYeVl7?wU%FC0RvH@rXwj;(b3?igmZ9 z^4#HT+H^xCt~^u;g`sQ7@SY!Cc%lUfl~~-~Qvmn=mPliZ$WAMQ&i%>54U_q#wb+bm zM~1?5wG_#JzR!0l7g5(xfqxYV=M+IDO<-l{>RO0jUzWC{*MJz%_65U=7Oy%lx`ZxSB zbq-Y%|FDi^1FkA!{R$~`eOIM%+Y)H%&J-G+7{X3o_kfyfD|*%!!RBlM{Ik?4Zl*HJ zo~gi6JC`x}f*&Z{r$gp;0o)0sMC%$HsJd65s;aZ${v!t!oQKo8xKXfPD#@b14P&|c zJA}tJ*(}CrC{{!%!sPyBe7?63)fSm(ds%{8#)edrn@`r-QevHH%3{^m3RKn;j-O&f z7QO5@vpYBJmYu3>F@DTdFwwUrhFDBC`B{((0hi&gQV5fw5 zVeaVj!oGK^v@gSgMBBnh%;$6KPtJyK)hNvU_z++IYS8|Red)&(EqoeRf$a_ERC5b- z@0&I6Mvo<({1N1ttx7FLUFhgiBPE{2OI9nxU`HuDF8GJ$h%ib|lAw_bchR)EItn%U z#XM1=?V15z)lrS20eiBKp4kHD5!&Veoalic`8lM$RgH*R*+Rigjh`7zXWZRg7 zc@#^!auBXf?d-|AX5m=!8{yi)ChoT_!<%qLddp|ov#bY_W2`ng%r~V`Z;GJw=o6d3 z{o>&&^<=hvjo9BNmo1fE0(kl&qeX${t&ycaCOXulx0f9|@|G?8mCoFvz;z++f1OhcwXi=r8eeX?W1?d!GFLvqlUubYT<1ty|4pJ72i!=b=MPLJj>5x09ZByj z;A?UXXVbc|x6%|jzTNEI@~LcVvpQ)TT*b`2kC@%1)wp{64jaEjit`Zdm!P8zg(0fjsnNDc+}pd7o6Vyd%ng3r!>YQDwz1)SyR z+(El z0}ijXhPG}368nTOi@VC~aqcbV#aSD^3fqu5#fZtL-e;QCD$MuaD8V!_8=bFLL1I!m z3Z_25!FQ=NTjwcBZB0O=gf_jK-How}8Cmv^pw&+n(v&@k_*j&Iv(qhb<_$6RN4JHl zT>+mcx1z-TV{NeqE8;AajSJm+GQW@ zB5nF^QG?<~Ge~8NE}iSEMt8U~?vGM%Jm+!P{W_{ zUL1u*l`q0xyW+s_M>rhXpWLd{>D}l+SlkZ5wCAhXOwRF}x5x=64~H_{kw@UfbN-Vu zPGZ8GdYt6>!q_CvRA?SSJ8ZORpQZuL>NA48f6I})csSh(48udt^;&0bL%W))F;}~j zW6Svaf2{;sTaU0yebQOO{C(J#b_CP^)ZwmTG}fIC#iacG82jEH39mHad{PhQJzbdP z7eK*kwp@v$gbfpb_h!>@AUFfbwq6*Wa0J_}Vp&bh8up`kCiA`a9&-J4$ZWnXZ3}AR zZwnc^XCy~kN1Ks`*)J42??cCj?>IEumSz>Z)9yVpC}@@n8K`jw!Zj%za?r+U*DUss zdzim&@x+Bl3u;?&2ET@$!drt%)GxmYGw#Hae$Vje&Z6#~Q*?KD29gx6qD@?m z68;|62#sSE$wyegb0u13+lXOneX;AR8TpnxM$dp)lpd0z;Hlk68qkW*-GMY|-+Y=? zG=^>rt%Z4Z3Q1+o61Pb%Ld#J8|8oBnHdj_*OGha5&nH9T#9tN^Erlu01Cajj37dCn zJe2m9LenV^^LJOF&hr_v@3`T=%qYlB(56(*nYGF%dh|M!3I=||^+gX5zA_1NBWsYF zGlZOsM$x*QlTdQ4!uI|L(e5lF=TH;gv(+NU)rHtqBt=E#ULAJcot{-yXhp+dK+K0CZYD}9E^<3L$l3xx^=^d?(0t@ z+nB4gyQ7q>CN_|Q!F?P)#koL+#Skuz#=Xo52uh=uNuel(j@C4kfqj4aq3UlWbVO;54c&{>{ zH>OKy{?^4b>DVVMnWapU)wOtAdmnmV5^+1GghfxBgxPA@?BJuNf_~s2c4zQf);d>U zTJi7LckYSJlb1*MXa4&my9N$-CZhL>2PC_+(KD@{E&nizrc_T#k%Z~Uz4cw#T(<9O#|7(p(&!0)Mv&af4td? zqiM|T#B6rqT(odJ)u$`16`XEGK`gqa&d@nl&LzF4}zX3bdM8d4G)l*)-8IDQiLH>8T{ z7alX-wcA9vvo}n*m~SI0vzZ`hPKy_80yRXQ{%_c#^1E!AV*+ba)kMvh7tDF;JtneH z#i76u<~{zhuz%Pg#6|~j?oB>>Y*fZ9O8c=pPpyPMHlvvG<)0$k{hgxU%?8Y@PlGU{ z?31v~Eelqo^Kf8`FLYd-1;@%qBJbYEqSAL=u@Hu+Tz~h>{o0 z2)+0}Md#ty)BA?;wg#nP6-h=#h3a#j>pl^YjF3V`WH*Q?6d5T|8k&+aDkLP4P^qMl z)K?@5p=p(}XMWG`FQ8thbDrnEuj~EB7vmUySMy-y{9G~drY6ax*I~V1JZ@+Y!G~c% zc)|JqMsF)v9{;{bJ=SJ@#N86zpJ8*&PAcq~evWGoPN@8z4moU51z3fd>09)gAT*$sxEg37Zk=#5x&>}NuvbmA& zZ6^5dV#=z^*tqR$gt(EWl7BlT7JH8<3Ds`HE$$t$vT)PY#`UPxxF(n&<)|?Y=un}A ze!3VWOLJGi-13avGMaHf4SQm-K(jv@%*>#+ zp^`7Zvgm*Jti``gaptZPHA#B=HFF=Y73O0M?yx&j z0m8N6%EIdT8-$uA=|Y)Lny@=+GJVk>LKFHfCKa3gC?2;1xgiU1QMw8Ja+l z0v^`pATH7si#jTi@Gcl-vKFY!eT>@L@0k0!C!M%|1?t9!;9PtXi{_cqcs*%a+)I&6 zdmTY-zjH|7>>%lqJ=pgt7=x7KalY1!78WY5HQgCKC^{$RvcTK457+sZ4&zPBw1l1sbk!hYT`vX~cMQ*goljUVHSB zQ=N&=>7j6HG=keJJ>H{}p$~08ux3FCCZw;zRFx_G-2a{xhe_edy(3K5vK^`CrefNk zLe7FBj4NM(z^Xvl6(DmwR#?)$V-fm#mBJieGU zKgwexisMi_ocCt=a>?&+Pb}QCQ3<$| z!(GslL)gRlxZbBi;5|GOmX6l!N5NV)>hvMz`KT3-=Y7Ot3=Zs#ao>6N`DYyB{72G7w;QN#KLG>xQB<8{L|qd#=vQ0}v=NT?`!wJ> zQ56TXQ`z}nCCrY+<71@_lKm@LN45-=TYbk`e&)SfG7@h%1I2E=7QWqDhh&pRsI2Kp z&wCuiw)rn$H1a#5rH`S;CK_Eg7bDV%?>x=b;NCd~2i0O}-eOtGj*#YEms>DMt;IVB z&YG+qNQ=j&!|lE;x^{nraZ)~-+xNm+c!?MGN8qNLfepQ-Sl{Sr!a%oQf}VFS?~X;H z$~ztxXT0S8-$QJB$2sBgEE^bJ;Ca*j!_j7w%X=1aXzn=1?{WulVNEDnvXO#zO zE`s><5?-6#WTn$m+46y#S?BK}Vd&4-LQZ?IaAEQWrgd@-8#QGKTXXz7Q?~Dqs$EVP zd+jMZ5q+LHA1-FYDH2azW-Ker~b>AXgR zP8HF!?Fr<}Jwn-0Aru%~M1|eH)ZM-bsF|p*JYM@&Il8VGUh%1^PK-6Se$E zB$?7|`W5qv?zvs0BPZi&)UZ&Jo2||~pPvv?_05F+bK+@0*e`119k-f)Q|Q@?S@bw% z4_xym!8q_eGnw4Pbbl=t&&^N~7n;7Lfuj`Ad8Lqjpv4$FL!R8$Jwxkz2Uw<*vawkU zQCySFJxT}YmPs=QqLX7pmayvQI{e{-TH+o zJ8ZF|cNo5_?WQ4KBAvVyjLk-4(P1EmH?^tkt?Ol6sx_e#GGpjw$9r4}$$*AcFvKyx zNMYDXdZ5pDH(Oe1ZQ@IETJax+_cozf(}$9Ny&A1F&7w~(mua-6Ck33D#9S-%;Wy+c zB9&HA;yOFdxXQxMcZt;SW;I>ac#4D=HB?#_u&Sw-*|kq@boY?~=}uB170$zc#WOeM zCi}_L*OYiF98c46a$+=9TCBMhhyi3<1kqso%!?q;fHh|7nohLiauHU}v3SDJyNiZi`dv!G~<+Zc$ z%TnmSP6vaAdmy`FI{Gf;4Dk{-ns==?>AY)3f;!*LZm-6l5@p)tTEi^6LYUH~SXS1u znXV<8)4TEak)*qaOrCC`Y$TH}5Xo_9ck(7{`O;^`H#GaHO zJU1RfH>b^KBX;!USzb5Pyl)c}e$QZERt|vq`U}FGONxTSwT;4uNB&^b?6F(}O#hQS zb#*3TZ?XsOsLn;G;&h}mjD_~k|LDf}9D4D39qk$+Kw@+sJW+^%=%w zV-ZvqK^K;uqqsxw$vC(fHzJhjApfSE-R@@(PG-F7y~XO8gxqdS~kcf$N&1P+dDg3rJ_Y@3{jgrFCQG>M@rm-DD_(rO(3 za2Ybze_0FvS@xZKh3!>CXhmHP3tQ@fs5UE@xn4lIw=Q+*3v}yxIeWIUH(m}p#%2s3 zPkD!pDYu2^MP}}&0V|~GqI)9$W*0KQ#zqzxWyErnUD>QnMuMvKD?z_t9}CLWW4q@n zVbc%(|Gc0@ote6{_xV$ybQ{lPJpL+-8~0VvIerXJY>g*>IGA&8c3)Ys&%{Ks?+!y#tskA`JLJnr zCXhbW0i^~Vno`q=5|>LrupG7cE77fG?*-8$i#b^yVu_wA;+7+Z;_Zr^V&@z?92{^N z39ZAh$7vSrEk8u(*YBnM-M_ImR)&09i>OyzGu*-X^LAA^5 zBwDlbE8F<42Q?phgq22%aHQ}d6-a4_h50FTu45FM25QovRfA}N<{JtvK1maemFTsG zG@I#Qz@|j+WUo6WvYF|Ma7p0ZnYAq}(kli#zg>h}wJCi$Kb^LZFGIya3)ZV(KgP+8 zCfA$*`ulVgeR}Fl8|viv9+lC2-plp95CMy%WDGA`!jALo?e?e;*4A0de&zZz;dcw0 z^`{TcOkU0&_w4~T-3kh3=TQ4%2h=MKkYpjt)+|0QEU>e(c*!JGe=7*Ro%XSwGW-s4 zZUHN5Y7{0v2%(K&wBvAZD*QBxUWX=d9>+Q)T1R1K%4}Sn6pB~RVyWcIFJv23;EdiT zv0vCKQTDKzsH9=b`)?h>mUlNTW?~FwoO#B3!INO2_Lg3+tD~QV8>ngG3k#1v<0SXY zuUbs)SBnO#Z`j$(7&BFdv88Lc=Y48C^WQ$0eRF#*nXU;*!tEqX|LcPM^ZJPFv?t|p zB4su9p<73%Ksn?T+nltRSx+!ych=2h<~{6%@cskD4o4O7^x7uM%fE_kO>fZr=_Wou zFJKOJ<}B#uR>}R7tB@?thnD6>>^jBgs+=_)lKu!AqNpl18#niL0a?4rssSCoH((j~inM2=`x!3CFP0GDB znkICo;bw*g#wn?jmdqCJch0A#`l;L_qCs_TdnjM%2CL0u8BwR$w;wt}q*SP+%zB~d zTD(S-ygMM?a_vi6R!MNY`h@)*a}zUDw_s%K2sl!ZxKv73%;%ZJrv6U6b8~~;_i$wO zFLMx6DG%1xhw~O}v9NbO8?#%MnNPTl9cMCF>#I#nBX~I!`pw4ai+l0PRfn$6^5UFN z71S#Yu?%@5Yw7r^QLOE{1~1D4cvxc!ubax$WxEnfM{H&(HTT$M6V7H09Kz!3uhD0Z z5PBMHOHSXW!;Ie(vvguuO}&h0w`L9PJh6ppnwn{;g}QhnOd=ju^xzzsSBSc~5K3QV zL^tbm)cL+QbxgV{q}J*)-#-z;pqu`fLvCMm%H#(NhIB!8aB#Wq_NR?Fnw9EX z&MU~XvFp@?BPw%+sEMa2d+r7D*mIL^yI*JNE40{}x>rJ#yb*?Lyb|&(PYWkz#nbK2 z(`iz+7Fmxu%$(yVuqba;j2Q0Bj{hgHX>~P%ci&BvcIzYVdPrkh&#%no`B)Yek}Mdm zzW_VKos88w3#DVe(8VVcsByhD&2iS|InZ|)x6=Tg4^Iik$sZ+4JB~;=bX4fyc0^J- zGmWQwjU@vf?Gm=lj1jg7{jp=|AvU+vmYsR7K}KQi`1k1=uBRokNeyAleCAo!%l9_f z{p*18Ymxc#d-@`qy>zSU2Hoi6OLw#8(V~pQP+iU004K`ngJB^RiOJL*>OB~qee*u-h+zOx09}@isPf0N-md@T-k5QRR5#w!&lu1o&Rrx8DRGH({ zG(8r}^TKih_vHMUg-VlXy2S5;0k#piQx!u0|1d1J2*b(jASxVwgrHXevNwV}r27ehZ7J)uGNfT}tOY;p>Bx*|q5>N$0*inT-l#A1u6C+xiyvV~-1V z>=@4678sx-T#rnLtC4xia(bDOM%ABx;aS~z*p)AbM7&CIE?tzipoDyVf=OAajMlID zK;4}AWH|CPd+wGmtjf42JWAGwUgdA3DDasCpZlPgnZt>rpMCyz^^jfO780g76@%HCYF>XHZl8vV~K@Z{Y1bP&=mPVSIQ}pRAobMiw z{}OC?9?62=yJyqvRh^JMw*XI#T*Z--dx=d_epJ1EJx$qs54BsDv+IsLdtbxzk7xWb zSvmu6Kh`5};Y{k9lS?zxR?&y4vZShtGu~@vraFC!c{-J4lwQR;o~T#b8J@4>Oe;*=)zuLi*-!Niy!KIQ?-qnaVuH zknYcHYV>M4{;-td9k)=rwi1b}%5dCM8D>R?*tuzonZ~%Dc(z`aF7)3^uDM_6)5!$f z9lakzID56~^j9?U+2)@=7m->c3#}RJNb>v->6!WBc>57_{!Bxe?MJ-7{0+rc+wsWl zB9on>$?WFIu>XSpvh*?cS^Jb`;evxF8~K5;4L_H#!_A9vMwo)>DP!sPQO*}XrNk7> z-V6V2^QR5tqS3IOaC6uwdSai4QF9C6*`9?QuX9+MC5Hid0VLbWcgE_AD8v@DBBT~F zZhn|~FOY84$B?Jx1d2`K3@lway0YLirnhs(^Vu|#-5VGfAy(E!j>E;rG{A^0-_>l`6kT&1*khm?)Ab|HfR%nMwDb z4kN|!T^MgI5A*VTVX*BaVY+7!+5EgiHch7|HW~B5BF)p-*Sz#PT2}W|ujM4fb=;IdeQDzfB-HZb;LT%3-C$ z^GCI(Sk2Gt?48auHlDNIj@;9toW7iA*^q^cl~Z7t@Q^KRTR{C2{vf5YH}y*`rSr0C z;)Yxuv0kSyHAONC*fNM`D`hBdwFA9wVs=I}dv z@+Ita2#3|={+Km>ATFfUvd_kCta@88x-N}E-@uR9GW{{!QrqEJcpCwyBJgj2FEowJ z#>5UcluOOP(QHS$xN<0kYYn92#xN{RdxBcyEto2+Mk8F(;W&loDtmD!lp6Oij6aJQ z)fQOte&LIVDdf;ShTPo@DRbL(jOzWFUYy-e`aQeg(aGn~Ss9phCrLO{R4Akz3m4MU zoP@`<`{`(mEv-9x5%MqP;Fb^t8_wmcJT?s3p`52;Ihj>Nf5g$Sm83p-D~)v;k179T zDNO4HFEdsm!n7Av@!U%IjbHfbU4=eo*RXi@IO^{=gRbv4qK)5IP|wUIWKoobjdSH_ z!o6q0io{0dmNJ#xZ>y54sUnp+HsbM5Kiu)Xg_&WaSeX7d!K><{ut9$hXC(&F;hPib zi_3A8eIA36c3py$oEioizhpjZw;);w<=l=LwDFb!H3UnOyUq)wZ0(Edb3CYUJZ-ZQ)JOMC2aQUEaAWT$-PBIZt;P>3k(m~`k%bzq2%JHat1b4qpA-^7D=%cwdnVnvSwuZ&f&2B}~T<&ju z(wFwyeM8wU4;m@c0jX{^(hnR(dU|tcjp=gwelm!RT>HZHPBQkm_a{w-Foc+haP5}h zi?a`v+*?dNzwkM?)CnYAFd}&iUy7H=P_=A3R-bmq#Jc(XuBu5#@;B4uxToxHq%TVh zkb`sWeKu=J53G!Q#TM}y=!N4~(d^kDzqKszZQNFjH$8;k!EyL{UW#rS?IE@8^XXo# zIfcIH!)K5ZOg2(L$NT^oX=cFxnI%px&_zRgC>x_$$nv#}VC~+B=HV5v7?y;N>zDBG zZ7&R65GXj?_P~l0`Rs!G8q`Fp;Cq)m%nZ4URjOJTF?ThV93Dwe&4*%fWdKxe$x>dj zCEo#Y5BuIDEW~&|ZOxrbKkazu_trle@w%H1*1n*CrTXITf?6KbGNwpBC-T+tpz8Dm zw8!Ek4O@^!VXv3a>2N1uu9GjDRP6($WCe2JjE&6I53zQP6ZNaL#(BO+z`$3eaFG-R4SHD z;T)j5L($2(bFcOrO_?GNkEaKpwCQCZ?vnQy&U0)h zSwmMlrYB9H7L6F%ZzDsICI3)}0TgPqP0)NhnVtUEgLHIk=-%>T{L_BP9v=0iQ^6;x z@2y0Ny>$hFy!(xfN|bH$3x~GWA(wM_Vlxj?w37jS+I$H;D|+Jg0TmQY0QT~E?1Os> z^w6OWS01a8y1Eg4ytjy2W{)N1s|jRXolb7kTv^NgQP5qP3FtVonOpjx@qsIHNAdIa z#BtQz>PDv>BI(9{FH*a1LYnc{Xvo_+lzs3dwJ!~)X|Hs|-!4I-IN}8ko61x8)fZ&u z*AG^bG&b(|K7sVs^Pa~bDv!&j-JE-VV)8MlMZFTVk9G;Uy+Q=@U#19YilPtBD{0U2 zAvD}O14_*$nAR=LnX4;k%Nb`npC?e2Nd(*BG8u7Sy^(YC4&J>~;w<<-Y;D|Vn0$7^ z__fRMRACUExch~c2c4sFiws!p#I4LDY$e8rDp6fOc?y`6paQ zoo%P6RC+BvF4>HfX&*2`Rga9PY15Dw|1kf;YpiqJMD^UQP?cy;-R0l$W;yrCn@`8p zMine6m4kup0pX+t=g=p(peC&!w#^kdbBxb_MjmGYWif1ORt7WlE@tt?t$1QHjygNPzVme0uD93dh2MYF??(fbwPoSK&5Kanp-I!98}R2-njY{zwwZ*p$Y;hO zD@v7iYRsh2JWX;)zveLq(bHtQ}78%%9PwTRZUMaSyCcT8tj4MQC#Shywl%wmsR3 zEmP;gdC(r1sq~`8f^`Gn_FZMT=Am z7Se_dQ`uLot*qry6-)eblz!E?lU}?U{b$#L=2IWBVUs+aTVpEfrE7>A=RjP>{kV&d z8PShBBE3ns#ui1+g}p9M`PE&_Bx@gp?yX3&?I~X0FE0*ydzJnz-wCaF0g*EY@OiNw zjXbGGKMvoI!GCtES-X~IsvVgaT zA+-M;L(JXJ*ttN5w4CN(<@nQV*B*atYh6y}spjOV*qd5LT!-(qEL2B7#&yw&wkHfG zIsUG2zo}3CluOBcY8m~t;k)C2GR)pMlIoVP=giU~Sg9J)TBln)Kll*eI=^CD!XcQJ z44{!dM4Bh6@Ol4D$OMg~vClrUjd?B1x=;q%GMdyVa<=xO6Ex~XD?FpivFdeSx;yhM z?5zm84#~J0aufZVb?9;PKpNfWFIyLHgC@^Fo*}7bFV6y17mBd)Q#hJ;Z-J4NtQaKGr@DnE^y-`m zU0m%5tv3^qw_*$D#{CER0n!wrszPh?=8^bm3DphvqONyp+(~v2&5uej@{S))aAe@S zUXf&GVoyg0&!z&e2=a~WK`&F!BK^Q)Tq^3z4w_}-VLb21Z#SW4&s}tJzyX?>(~I7A zucb^aPtvu`B?tdQbfc@9x|fY1=V3$1@9q@J*1bkHqYhC*z?0pFU(KyuikL6@uJhEB1`^b9pZ)SC^+ zUp@`Xgc@wy6NSmDSK;2VQtW?JNvyWi=YHE&`0J;PmIfI{GuG4H3!m{hR0W0JI+T`k zfTE@Z(Wr_TINu#a!G~I)5jRF$^4|o}s$_(Cr`3=&dV5j)%`?<8zJ%iUbJjkoh?_c+ z>13@mxh^PR$B$m4slSiWg63d)-=#;P->*}9@;p(~^%gt*_n@TnOtK`&=o#hi`9t=8 zPf5?q22u7R>pAp4;h@=G%6Mf)xmOa{(B~VNOHw;KuksQjo^HdIQ-^52>wlzEbCH&8 z3Z;8BAK?GMmeNw}DO@KXGQ0Lumq$CTn`tDbzEY!pg9GV$!(-}N)SDvSYLi2P30Yb? zK)=%x#R(d4pY)%gTyc>dss9eGuYJjFx+7(NwWP6yyV(0LxsXZ{>C(TyBxA$7x$`~g z*!M*GwW64UiyP^jrifI#$E@s$E4v-HiRJfrgZfliv3NlNZC`nsewAH?rhO>A;@#u< z!Ztb+E`tj_d(e14TXL^e6eAy{Q-1kx*ma9Mr=-qIEb}FgC-q=sQy-y1`ZcZo$-6|a z{3%3w7G>7)nXBGeW{|6ZO&Nvk&8Zgnj4Y!Iey7MwJVsgV5*o37K27B_45f$r=+g04 zT%VQ2diK{4j$O(IlX{M}RUJsQy30LZ{Jr$%GX0w{8A%h~z$!|LI$Af9v*iSF@Qn&F z|MDp~&z7gT-%ldf?Gz= zSC(?j15jAd%!a60)9n~0%0=6VtG>#MvM*#Ym=%DwZ;3-rNzMO2^5ebFCL6n7N5=gLYD^^qC(9R5z##7>YYqyys~Jd7)<-j zdZA0^9X=HZq zagbIA+pu95{A|V|_R=`CA7u!Gg5eG#dpynlwwDlY4q-6}7erf@*i{DLh%qT+R0zTgBgMj0y zETe}GOLNy`-%n)-QzoyX`s%01?tho9j@^YNnw*)Em&3M~YG9%2XIA}X4>R#zDbxhX z(tsNZVrXg#%&oJLEOi|d=K9g`?i7@^Mhoq?Eui|PPWTddPEzX`#B^3hFn`tkLSeuA zEL19;UDi9o>W=!77cC(7zlHeaYtQd8MoeR|F3WOXPP2}E$7lx;GFrBHnBom*?yY?C zR+r|iQ6xQg88S68qGc+l5wboCdOTm_l=~G={i@-Wxq*)L6sfaz7{%7w(Au`Ia9fgy z!OlrE;DJ7UJl>DgzmKEbjf|qIXH(_hT)s!n!icF&Xzy)Cc17c;)b~2sxq0J!zq^>G z9gCg&xwFkng*FU$i|!rC>{5{q3w%}~y3oJ|UT$OGCyv0>!gNZz-j{55T!dH5BRm~pP7ag3 z$>~!ImT78GOZ0t|?};XR`;m0JtQ+$~WoU@qWU@DZNOw1yqW)7~JUsM|4Y0UR{xXp? zJoPK2e@M~L$9Hk;nlY($h2h-4tuT^)44oS-wD$*ReU_WjCD}Y2wRT43jBeIzHs8zH zoq+*oyg#_ojwu?wXqqQy$@r<0DU4|12YYIr9gc$C=Fku7WYWh^F}oG3;r{J1Jmw$a zJTp!4+NqON)Y^iF1DB)eoh9nJWXOhR+D4a_L)~)*qTZRK=Z`Qp@m(&m4RcVvwG7`A zit&NZ5MLI)hOcQG$)A}+%@G4gj_OGViBx%Z40YMxK-`xRw6>ov{pUD^My&+p?F*q` zKJS_0VNBPCd6TdC08;%lk*4PwlJ+MLl2{Wk@=Ss%bXsR1}&F@m;DmnZdphEybd7uU{x#gxoH$mgCF-AUc} zHmU_1v@bAsPj9$euEC9sAygf`mBy?NqG6@$(3^XNs(CM}swji%-W8Bk@9m^;H;?y@ zqG)e&7kUqvNJ{T!k*l*6eS8~&sR_f8|2h%gtNM~l%~$+NzYOQC+-t8oiJT5>CEwI< zY|)FcnDNONXWNe4g zo_iYLjm&QruXG7Fi$eJ<{u@8vPeHk{36|eq#CreT29GmoRCZQVoK)GuQZ*xmD2rVT zcQWzxraIlYmP~HyR(P5GP_Q3ZDfmy|clC;aI97TBi{5Z2w!sy=h<2ue$&+Ygcp+j> zUB*+Lt%wdifWYU=*{PVDY{YvL6!o1$X~Rp{WVwDwlh5Y66CHZ`{2kl3zJx`oePtbo z4`AXGMS5lYAJec;W%VD*5T&Pxl!`r&N&AT!?&~l=&-0_9(VStXK|T009x|^d9aQnB zGd;3!<^>34E^I!S}WmW@oF?@x^HK#?_`|G`9;t(! zOa3qfz7lRsZ(!2`mczrkH`A@UFWln%sk!p!SmerNX0T`vySX!sI$}PcrTP$3`ai(Q z-525hcpUr}>WOxB_TsDmriro3TR!q0UqA4+gJ>O~f+H1f( z&GW_9b}un`j;!c0X)T_Xsp0B$8yq)X&WCI-#UwO~FYuz7ze^IEh5Yof=Z zalRDnGZ_1`y;i*cV}p35tB``*2e4g-QdsjCOMHIwR~WqaH4BSxMO(fi66ScLwby)f z$JJ46!a?diTZMAwe_{X1V)3>lkQ@$jS8SVf_+m^6b@=T%puVwO?{sJvT>jkhaNnruWj5}&fg#hGZ9btFB73+Q*-6ZZN#!5Kh84m;~)f=AIO*}u~V*yKgr^O+*WGX>+|RHI3C<9qWQ zw|s*4vGVs?RWuchfgt+(XU9W2yMRaJDXHFRHF2 zz%Ok)7sTyjh`zukcZ9LwHZg3gBlnCR&0k|%2L1t}Lko$HPSx@w-k5d0y5j)aSX7Ba;j7S(pEoM8m*k~qa9;CZ`n>D} zUJP3UJw7+wSMrh#^MA&?n#-82(-tZ@B14DO6sfGNL>M@91T%bP&wMs3psJ@STk^?| z>Gv=rpZVwT@^ffv_`>Er8Hr)4+;bXVz_M>-An=JlR0@LQhjt^VV)BNJpaufQ>T zO-z40iERjQMbFpE(IuUZ6D`wO&ek;6e`^(!GSVf1J;JEh6`16-h`W+iF)?93o3-Ns zwGB)o!GfF5wC?b;4|%3ZHX zSN*e@POlm4$rB%z(%v9yExjsg`+JJI3kH(5>msU~YfRr`a+#D%08^~9fx^jR`k=Uu zj(nD(i8=fXo5gyKrRk6f&Rt08~vZt0u`?Ufz`>jy20;Hx2yR0cR&>wL*?otc*l)umRjR zXK)9lGHTkhn0oabaaXyDxN>X`S&n`TCfbR#Fv6P=2K;#N=(UKCPuqzi|;J@ ziZjeqg}d^jSzj8#evEpKoE~;~^!yCI&Pzl6R6i^~B#Sm5?oBC~Prg$$$o|A%#44Rb zqr3$guNt%X8LwHeLJ=E2q=YJDCeq%JEUfz}A$Q()`9VieF<+aUc3pz)qV-tWVNA|n z73i1EGbk1?8ssBVy>x$C`0XpZ9XJYu@7v=)&yn;o^9J4&zd}>WQH-AY6Xm?y^FYO! zR-f5M{{F+rUvdh%!*?OXUlmfv_i!F&FS2hMP6ly&U*xioJ-_*l`KcG-6)KQ$Z67Wx zk76H=jbJ|4o-xx)Q$@q>*JO14IVtU2hV(Z-+0otlcrtplxa@NW&HfihuU7ODukdqB z#+GVI+I@?B+(Ic0;bdoC#Aa(L-P6sxq~ zFl*!M5?5HW0qf@p3r@{pwGU0%l}c4+u`~kWz<$_iCWWbvr)agaGmXleOny_mFmR9t z>ifvD%ps4VyQl;@<4&{w4jsr@#r^Zrhp^dXm}r%_oAT{bP<20^^{ZZoxE<5rHb;pz z7W3Wqf;{-@?!k?vbue!3N1h$)vC6=SLObn9Q-^1~ug#_-BX5%TUB>Oq+ zm+)j`2+TC@vR`$zLjSjMu#0QM>rn2eJh7d;wkA@)p3)iXQI5xxUx1_Gczp?C<70_{TWLIp522PLTJKSo)HlKpv4& z7B1?Q+>tFid*#|&|MViBH%aA|f$0%iEEp}f?o zun3zje0}g;uo*a?JDK)k#Wo`tyCk#T<8?8_W1ldp&WzoVOJXzAIw7xq9`{2JqBC_b z3U8z198p4mR)l1DBBEX1fCHPUBYI%N#HY`YE6cK}CdKp-)U@yAq{-#~_sr1&CBxaQ1{Z)_wSj zLCLGohs84wk6=0>KZx(UK4Fl~TQn|PM}@b&Xn@KHmXdl9jfJzRDLNMEJ8H21=5B12 z<$J96#gzQ_1YNnYkJLhc)931sv~rvkJ=3b?d;Tb@|MVPN2ll6pIT7@FOcABd6lk?v z8FQ!~N}5)j0r~QRprOKZ$UJi#KDSl)7#S(<*fd-W`@4s}3>!>EomXL+c$2O7eMzm~ zCW!~Shl`g-$x*lNb9%V3k)A85iSd&jkhRWnvaSnd^_E%eV)i0d+Bpf{-oxNBPZuE# z$`ox_1%ogf9Qz(hN%ahOt8Xy%hhaDz-CBppwW93eLb?5`CH0DxSh(5 z-#tdxT1!c9Wfgr3szP^?Cro=DLT`S5m~7p_T6e3k^Ld`o8sCflbjwj-LJHlI)ucuD zi;=JJ2J@a@V@K={2!-BO%q>V!DBc?^Bs}jA>B}#eA559j(MQZe>nhWJc7*+GQ6RnMr$db zMKAh+WAW>#IMan@6>KNP=hrEG=o6UkFr~xfPLsl8@KQY=pS}8%sh5GU?P?4Alcz_j z4{u`iY~Ba@bCA0D`#1c(0{4q0P_B0nyVqnvhH=+vjh8@+dfC&RKz{$7aS02xw8@II zaAfYkhWbo%%1P6w%{jT~pML|Y4(G9_hbw=droe6v|M&NIF?as^c=W@7ejLyvm(Fo? z{_R1U;j2Iw`l`~AK1=9wo+o>sxf9Fx`SKZ_GBav9&;G;)!rJf$3!EcCGiRPfyT@bj zq)YfE-HpK>oL3OV;9=*=djx6C?$nW220nn2&b_ZA)B1rjr?c zxfad7^0PuO?%!Ff|C_}W#NykGXK*MOg24_-2)N2SLakxQgdgmF-Gh_sLxfotV20*( z*ow{+)$b(j*t3*EU9BYj+vTAUvy-~ET%ygiIF14o+UfR224MF!%XYym~LZ{2I-@aGUrX*kgZ5e9P_oN=B*;MMx)A5I0$lN52#+-Gdm)Q$xhM_=h zjbC83)r!8?h0sz1&bl>ffJ@G3`t2aiRx6t`4Q?6jU$cRR8gh33f0JqPvZJipFb_}7 z+L2bChdH)#^t<0b_yz8VBw#YnX3nG4GZ;K02GRaDJ9;+56w(DJLB1on&!I_ZdAc3B zxrb4xX-pei{HdS%ZF>1?D_dg{%5HAo$KJ@=h<2`uqD6cg6`qiYu2)BhX1S4~?u+Zx z`bS>u>?k4sw{B=_+=5R}|5nbshwt&0BzOG@N^H{Lxat&g`HV=yGk3Y0v}33_R|o3Nm=Z0nJ2moD`)N>&9ESgyLtT1&_snp^i%mXmllVz8DI9Y z2q|y2)bRs0a@XvtO$TUYu{ovu7U*!~PVU{=gWwd-8h+~l6S)cOTVGSewsN1|vtVr4 z??8XUw$OyvD{08X>Gan&0~enrV12oWD=T;&b?io=;q*6QywM)E;`kF`^6n*UlTRd@ zGF6rBJSGb(tCff;pNvJ0S5P^v6*rt-uyI-;5;bcLi#{C>$+0Lr zHVeBl>S@8Pa#AtsPYrtUl(trfE*(zBYxAE-Z7*P1`>k-Jv>B&0N0P-IfkOBhc95JF zd8|55iG#WCC}=rdnK}@zS$6nm9**3v3C&;nQf=BnDhkYn+Ast1^D&~|my;1~Yr$R| zY_`Zd6aWLoC|1FJ?~B(MvKI>5cwfws?nP+hvE6wgvE`b@7uAbcygQi2t;nTAx!ch` z;is@fDpTTdFpYc^if}NYC)UUx6*gyI5Vodd26f1Z`Mqdim)DTscjlwTqedy364Of@ zyW^1f^ZXc?H~(Qxc6ta*{f!-c?w~a!h<&~^Q9M;^DY{MnLN_-=<9PLQ?y1%h9t`_{ z;9G};tu~7+p1KTT5zER2+pGudB4?+cSXYlS=W?7(e#e<+JMnYS0I2?nrNQrc&NwX` zgO0oP{AyU$t-k(wz zGcMYTmrtwEm$M7mGMyaZNe?}W+n))!{h$xM&T_WhH}2E?O#0`ZpvTg^EO6LXcJY1z zo%#0?-a*W(B9YYdKG*%Gq#;6NN0O{4 zDx-|h-sBS^Dk>Q%DwWW(lbt<6h!mAd8O8T}|A2m}r{|n=-`DkeUFF`#GE<-|i6Wbc-ur}@zoXp$D`iDl* z%bE}bZ>tv?eVl1ulPqb=^he|OF=DX7RB>@+FY#IYO;ov@!MMyCc5MAHx-Nf?&Uq?} zC#U4Wl5^#!WiqH*0X4hPC=)6BjzmKiD4Ogko~n@Sgh7Z zco1ADS(Ft3&v(tj^U@crFr<`>o!jYTQ5ik4*uWjOb6I)BRRr1Ouw}ceC2#)QCMeWc z&;pamG}WONcY9c4j>Z>ZlBWZ6-QAx$w@-)6DQRZDZ8O;wsfvw?mGsMVHQmnIOlQ_; z(6g>Q*d+Q0KaVO2E{^K_9V{ngw}(i^J&j@Z+T3dqv)$s`7&RF2*@~TJE4yAD!B*Y8 z0gE(qGJ0z+6egCjjurJdv;PfEoW6>lulGcV>OUyo_N8|T(NsHeKdJ5xrWE^^SoJuT zzjdSe|BtNbHgyP9*+&Jj)h?n-VOpYFzd7Oz%U5_6Y=}!;j>3{7{61w;NJ_8Y zQq;lyc<{F`_YHC{bkGkRyATNPl96nkp)YcCKCv%*eGriO5S=69>FDe0)H!)M9s3bX zht*zVhOq+eF_I-i?uEDeF%Lg)XJXBrwOBc20vlZYlWjG3M|o~JVt5}513vP3c|6Z= zu{@Z*l97Fvm=DX$doXB$5WVFU`0I|rKqsQp0+#p>7@x%UkecO!xJD@`E%Is?iVchHw!TH?0E z2>R1E0t)9iD@E!m$}Nvz4riljvypUfkqTXlx1>1vXgdESpMGA56I{GUVp-Kks=XM4 zn;Vl*yW%$NOF5gjVi(DW@23M(HlwNUQK$v;xur~%@cHdD#GRc%3y1Ba=IM**mcmpj z$X|`dgajPDy%sy~?1dh@ykh8hx@m|;mTUh7cXz*qRZxD-o&IpX`@W7xuEa3n;Jwk9s8lV@Wop(zXd z*MCO!1imSLewTjj_M!uev~hgsO1LVVN6ir>N_3JEbKJ`42KO!9G71xq-CHF}-o#Q= z-CO#l-9f2A{0($qAdUUcm~!%OU~}AcN-b-oq8M}( z{OBz8P_6&=8cq=6A8vjx?*ff!)8Nfkm&)Nsi|*610*ja(+KLYvasf6((Vw zQ9J8UImtreD(Ft*HR^tP9AEW@aHmo)I${-y?He^I@KrCm8EZzD1ImRtQ3qJ+H5n2o zzCxbnDl#7CLtny|Q%Y7ZO3?~~L(3H08N~hVH(G^*nkPB0<~{4WV8D70j$&RL&ae%` zq!6`f6uv0S;GerW`n68_13Qz2zB@}LtNVPnIQVxP)$4r4#Ha;WJF)?* zitKqmp@Z4~3Z;qD#!@(EzO)@WNq4V5MQ`~W$kuAop6564V@EaOeCp`Iv!$eVp^399 z`qG!^3W`1(MmEPXVf$s3;4;QgQfoDl1{cXu{q+5KS!6~%J{?4ZOeK15`w!{2Wsush zjI}BCp{1>3uz$y5w#R<0P#(WVh|kqy>kgk{(>BZ!wthGw*>rF`y$qM5SLYk?&i4(S ztlLQa547UWwvDV}_Y5{!PhIe6XhYbvp;+bV!Io)=)V8-D?cfeyn~F6wZuV@tyZ0kb zIckMpEt-N|d~K6)_v4y~@%WveZv*v4s*64}AOnEa+- zJbio;OH98A+kFkc^S`!?x+xgR)xSv>Kv;*GKu|393mchR8H1+{Aj|V zvy>H_ONBo|NGfLz4c_>Urtcmvrma>N?}*9N@=2W%R?Z`z4MS7sB>ox)6e(Y%2<7+;?PS=k)6`ScEECb>nX_rs{DryBA7vE<9+ z6v>tLzmk%}TZHMm@+~S;iY$_4Rmp!q34Un}rh=HAbS=-3=FQfpt4SFMT$#&Mwo9?2 zu94*0K9*ce{$Z%a$Z-lxeO&mUE8?snyc0j6oBh4v%i@=|u;-!G!n7wzLQ^+soo+S>3VyqEG7HcB#@*(u&ez6!&Gen2zqj~0Z+GYC^lZt+` zbMQlF!3?QqgXwhElOc)SRx&b ziKL%{ylG)s$^SiK^h~9hw6e`Gi+8Y63O)-#-O5-x_^8m^e53HJUpzAUrLv=e{n_^! z{h9rfPr{CK{=x=REwTOIep}ff(5&`@8s4SQftDpS`SeG6pOsAM zO>tEFErWhbS<@V+CR|+KA#|%x5qzErg7t-Bnyi^d(dQD#Q$vb1*Gy(R2UiQjKlMib zlY?;TrH*5x<-`Loy=Z)&VbnH>|IbYuPx`Z!A z^_5&(yKylti)0&<$mV+>rY|y}pG}F>JG>9J9Lm7fa)At|hYAxuj(|eYV!CKLQV4(6 zAbfhyofPWfLZ#eN)~RfZEK7Zy{SW9?{4?yx9WKY)!b>5ci~8o^|lE-aUtk=NY6 zV72?HU<|)oloknb{-=ap4^QJ~+92%62!mUA3wkZL5Sn_W3Y)ada7V96Xt-7`H1&3Z zZQcNh`M?vBVU6X&IQKkO?Qf5UX<;zAmWAsP6;M21%blqi;-o3X;&RrBpLzvbZ_=kK{zWxe)o8XT|t=5HLnbjBK(I zg?Ecl>+lIu?*`J8&03syRg3n3QKZ?~7aDz9*_p3;oH6h~@@#1ki(KY{OOA}qK46c% zJo9WZdkwpAX&7#h6*YL=V~a+Q!NDFzblB~wq-XEDLc8fF!DrP`MoB&Bxn(~2ZZn2S zzc`q`P$QWVO*T){mPWclwEOTL14Aw7qe3)i0^JtguW@2)8(Y}P$)A{NOcY8zyy$qJ zST@ih6AhN5Xlw0xwpMLEu78lDBm4BodCU%s_G)H^%Pujq)>#-cGmcW$j%RZRYT`!V z9&Bv>z zSTM@g66WD46K{Qi&HUr&UE_>8^Fz>=+Jx^OkEqmKNjz?oBSu@LV28Pa*n4lKSkx~? zaFp+X2_}2UE6|fkg$PLezLR@VqA}QY9`wt%)A7{B-1U_VpOgL+D`U=fwvETcl+92Y zs3L3~6ULS%a^~dj59}n4LN0bLh1@B?n{Zv4WjTwqrh2fNZXjKs=}WmzNtB!2 zTb#S37gffdrQv@J#dj-S;N>k7@t0J*xb8q8J2+UDncsNG-j4~QAvJx)h?%)!($_5E zkV8GobG^cX%HN}zPz^fB zJ3}wz0^m0ygl%0~&OBRgL2l|Y#1#*Q`Ohx4ne!6$s!I`g_Zxo6CcyU66S&kogwxbn zSg`E`YG)SW(3f(I-e`wP+hTlMP>#qRE$|rYPCMobSTOMvs;zmJ+_?`j^t#!6-ga9& z(i{J=B1lxFNkhpLD_x)P6Zi(c<;XyJjw7v|EQ+^84p-jEWkF?BpAq52*)v zcV_LuS13W_RA>FSpnU1;7KKk`Wuha5Ka3=Jaorn0mZeX%() z0t?+AVYgl#d+|CCX3f9&Oz1Lm=qzAC+v?cX#|F4#@Q>95JYve>xitCJ85-Dlg^ukj z76Y6wiphzQqMhnCoLy0Z4DEb$JsL(Y?--I!_j1apNx-;C6EXTnJClhzM;%k5=$OY5 z$entL<*CY)eQ5+4`e@SqTvzxv)-%nn9QMxfCwu#F0U~Cd#@A}@PrufaOqzWqlkO~I zWp!uR3TY?0=sSd5?u9_=;!0Z0`HK$&L&(4F9n)|(MEUk%a8MabBhU4vai{o>FfbJ1 z@xgd>yUJ>K7CKw_!@}NoP7lRu$p~%cxpGo*RDl4XBG{&A3}Zj-MMSqQ^uDw_&Q-2&EB$y>17zB z0T*BxcwDghafTgnHbBi?3GE0DrnyZ)+zAj)^N#wGW^xe4Uv;1deE0mKU@%Z>OO?5L zlx`M8jScDieNXV7(-VpluCl!j0ygnIn58@d`?o1fWF09C;7kl+7>0TDL zJ@IE*--ojAx@XuBo>|e!HKG~69jN(oKYVn}U_Ns+aQ$=y3IlFowt5k)oNlqu*8vb; z#o(pY7JQo1#a5ODvOq)bw$<#-`(*p+Y@s{M{&6=2_Z95FG=RE51DH2cp2X{X*uu#Vlxg7R}f8rBd#*bxi(1 z!<()X+vCgU%le3(d7G6tN3yGX_L5WbJMwprp+mbqQk`ZgJzp}1O1@g-!d3(J^4=G7 z^SswI{mcZ>ENhc^^vZkwna;!IE$J}n1|4j?jyt`2AoabjxW#9Ixax_ixUaJ}9UAP4 z@5}6k{=K>5ueO2}g~bbQCp|d(X+4e>)(bxi=g?&Pv*>Qo##Va7bIZE$R?|Rbh=QoD z8cRpiPovm&lelGmrD%D&O?2F3jL6p~AY+jN2dz*X?|BGWQhM-DA0;VqSC=$a9Tryq z;U1>-0W{LdmuOah$Tkc^opc>@dn%9ASFUurB!Xn8< zzD96v+^0D5vx=u4Mn({qeGypA8Hv}s{ZwMc&&1X-?8As&__5#=dX_fh;kK)K8pLCHPCyb235>H!SmUAsF|inpF8teVCP|p?Zi8@ zzH%aMzIFleGTrRa#)X)EbtL*t8Bh1(jCn`s8(aropb5|Bb2euUG#lbj=sg}g$|hmW zlwP!D@>jNOyc`;WqDc2@Co`L@l zBwj@G7g@aX@M13BuLS3l8o1DZG%lt5#~%E>ME7d@(H7G9 zrXjt3ugl-bUGUw}3;j}`33kugh4|*noa;M}hGaa2e@H3Qid!d{m;V<#m?pF|irC~e z&6sf}ioUSzw1@XYxAd_Qu36Oz(k9`o*v?*P+Uq76Hlvy@Puog6c|X^_PM-8q-moXG zFC_sVPeb>%BWBNYV`<;p$SH|)BGsbF9wrnYuTKkKmm_eAhTyVypT*5l$1RL(Btl2C zhLAJUPco-1)Z)JrGc1m{JTbpG{3;vjZ_kY5$FYYyJvq~i`vCV_(5*iIgy$}*%&qY# zvl73vU1me@r9Kw2x^-|Ycczt^hv<)Q7|smf9a1Yhti668C%NNmSBbh9)7^xXqvz10 zkQ1bG;s{1~C1GW`h+`(w2rRUMluRg&>R(51&c}IhY&DG$TUk@h304*80NtIrY*TV5 zn=dtx@9m4}`Smek{EzkG#r#b6qf>yp{tg(PXk~V7_hD*1n7k~rS;O{cEL?PFH$^3=-(6)!Q+Ke)K{imSyoV64NwjipB&BAoqkQS?!US|s=9A;Ph zm@f93O{d6#f_->z<8dfr%ra2Fuma<+OVNh$zE~d48I($z^mp8Ns(7GFK73yzdhMny zl_hlH>pI@+Hz3uYuV{BpF4bh5qT8doDCc`YRc8J)%`Ypmf zHDReyI{q!_4}FDMD7o+!7mlqW?NKM`+c$pZEa_we&CQuiC3hQlOoL%^Gq&Vv(LI|K z-1=`6iJ7Zt$mZ!RpyQuVuu+qBIS6c=eVdS?^p_RgQYYDcL%7e|oodP@Le{c0VbNRz zRx`IB?P|z?{}k?p?FghB{rP@-N(*iD4yClZcAE5ifcWd40iH*Th}TrcuU^h{_Va2o zE$46bX@xNMRis6`hf<4hi_gGv$)NWflKZa4_DywTUzLI-?95hN=(Odm&y!^NC=FNt zq#>P>v1qj(6d!OWMXy)v$JQ9QkK2dVA6|Ig=#LP!VKnN8D;54(gf8_CI7^$8Oy&V< zXuV5CPwFY(f_p=@6P=|9^w#@CVJpLE8s|&N^pq3h%D>Q&JqyI1=~t+(v4m1xo>Ho8 zuyA--DARCKWYJyORb-c7Lfp35pmFJkh|C*aE&{PnjR5MFSdH67i=e!c~h zeHy{U9TQ=>z#i-JE4hCq3ml8ab7MiU%+!OKmm=mnn&SBqLnKcMMSwUD#(fXszm|ON z$ciCn&)f7(94$VQ+0A@HI}p3)6a`&cOL2#W@Og0CB&$C%{hVZ)&K=F~ zRIX7UTkZv2IgsLeisad`36s9`rCnOOoYDSA81sDqo0V^kv(~3^bkH{}nKyzq-TTQL zD-N;OX<6)abPF3dvkDuo)6$l z%uI@AtS-f`)%{Qtn1-grNO*VkryCQWlIh)8D9WtCtlU|ceozNHZ`H!S&4fn2oPk<1 zJ*-*SpS^J!eESg#{DYr?UfC`k*Q{bcF@t|J#O#Www~iXJ2<-7$7e< z3xCS4vHXkqLNCW&Of-sQZ4VVN_Lu;*=v4Og#$5K<<0yZNnPbPZWf)pD2ot8h#R0pq z^fF)^o2~s#_)(F}J_ki%{QKp2OcQW8@)TIZGelH$gN+#~CX|d3|9A`)i}l>ue-=CO zWu^nAN*{*Nn+s@+SE0)64inQp?Vr-Hi*B55Kdzr1p3)Rb1OhzBbJ` z6W@Yy^(WzECV`&$JhXh=kDuaa7L#-s=RArL`RfvW{T>CWW8<-JdMXPw=w#oN1qf+7 zVdtws4z`xG#`6l(3tfgFm42wdUIX)0Uzm499h)-ZAlB;kpi#RQQbD^t)ci$Mvga)2 z^&^D-dx_$wVb~TfL;p3V!J~^PdnBQJ(j{!`H-z>+k>I}PSv-1r7vG~6P>)I*(mB1F zbY|a!@yJfl5?Klv+68&>BsNKTqM{`MGpF#ko3Rbgv(I9)6qB*8gNUp&uxU;N5=SUd zzw%3{J@yxO&t5}+Zx@WO$;Gjyo``v*fXq>xi`1Qk4_&cLYW-w%C|$?lB{KA}K$Awa z#nQi~QWQ<%9=hlKh zw5$Da;b8&-Tsx6HbuIr~C*x5Q2b zY`PP)1{7n0Oa1R)e!&2?y$LL`VjdgYR>(9@pJF>a)0p!8Vk}-U8(9_-e0bf3*5FiZ z&`w}yrahyM1#KuSaAv{YoYi$fLRw#TQD;sHne}$00}n62xW*Khzum{}E@yh_-i|GX zf01jTNtS)~(eDo@sbS&>3XPu1S!zZ!r$UCzDklg9yMtNNn;$6VGyK^e0xcEjq&~uelEXA4X%q8)arWT$Xr8j!I)W=lJeABp=pc&%a5t znf~!u>G6u?m26<+L-ydaeGJRl86>zaxx-#^)}5o#Vdil=g0A%4L7k_$ziM|X4fvZ* zF2@pSl5vF)C6PsHNHNqu4a2{0Q=ntDkwtqYFz41R)??gMcK&h^6gf{(U8gUr3ckX^ zGY<+bFOCR{YxH1aw;#pH>!1-aQIL2lLb9U=mB$ZcYgKeG*e3-V0Wxg#_gU_yqCCYEv78q$wpi?WlbTLZ20e&STRzC zB8tADr`;5K^HrLZbhravg>&{&%PFd7BbD}E$1~k+%w+L=o`K%OHtsY;QSfWDg=o;I zv2i3t*J!O&8;l3YQ|tmqiZtC$Ltbv7vR6-RonA>TC!3IS+mb#hEh3Ys zeDu)tgwMf1X12c`Me8?`sb&G~uuGyF3O;1>QJNNC*^5rY<%n|C#KE;G(B%7+;z29u zos}29_gAEK2{UN2(tFHmu%wXeBed!2JIvpyKv_rM;m@c_c6I0vW;-d0pQwGQO5qzO zB{}))HKwNj5gNc+3pxtI;4yhTpGx(>{O)I zxDfKyeNN^dui?Siuh=)S2VIyQf~}|eqJHsV)=^M`s|m7XXT=?LUy4xCyB!IWPGdL? zqOkIf$eWzQde)4_geN^&EuYEQA3wt8lsthRahP z;-=dn{Ost*rmEElZH0GepRKuA@#_oC{OLtGn(cO<7hUe1Jqk3ms)R0)8Qp& zG3x3q%#cx~NqPP0f?+JJOw32&vxOwC<6Mm`+3b*{2@lr~B;89_*k=AWI2~n7a*~ml zzx^|ECZ47B#`Bqa!X9kNlBEaLwxn9~1icRgkjWO#rR&>_jt5H+mCac^yt^}UkspPB z_>9JTDx}78lH1racrrl7MSBc}MVJkT|>2OPqgYzF3ho zjN;E1_=vE!f0x1Zy4B?uuXm+(CV-O$FJRnhD;g;@cVprWiVa+8%60c9n4o} zB&P8D+oX>3Y}jN&a`?K3`kCHB)Phg=_&kg*O#6eaZ*?hYWd?N`cCz8>hIpePpd{!$ z>U(XY)ABbdkKec3jX3|>^FLY=ahdH39mA}@ti`{Q9_*Qy3A7uf>7Gnaij+5oVwwXK zhigEkFb4a>|KMP1G_DRXB8&Q9-nVbV>Q_NXHyB5)DsPZBNR3Lij;9F?S(tU4fB$V= zjF0_-iEXVSrQ2%a8HpP^Y*)bMZkmFTVP)8zkxds2dW$iB8Qf{RO6>GZ5i8$l)4nk& zbkjyoY+iDbcMLXW4~o2Dr#wL_608qscZAE07}2*Ajo83>Nv}31(htoa8^aYsDQ0!79nNNN-B=DqQf#1Ncx;5xp%DL9Lv5m?>A@C6-=jdA9*HJalatbB!dxS zmZI+INSd#A4NE5~k;kWr^!f2Xn!itna>rW<;qD>KXy-!69(#+IsSW7l9j$`=1+*&M zhPpX+ex0xnZR+am+LG^rgNqJ69F1Voj5`L$ze2*yeHc&}gdiPVy21ShwQ8N1`pATK zai->z+wWn~$B_JV{~&+ZTEex_cFTB$p zE0{;!!^Ln96eMZm&@Wwh6ntQRexG2=ou%pFEPvYk*Pqty7)q`0b;#YL2hFeTD}LXV zNv)@M(~H$oG*Ryj68344N>vf+_|H_C&1%zhr3M}Fh(P}jiAROl3p zNeySAzhytBL|FInmdcBYPv@K?k^4&qSV%`MqGBKlfQDf;-oicfQSg?tf+Nf$c0%fm*)JFkg z!7Nh|>aW-wi!!F=?aN9`7hquS6I2+@fl=l`B=~N{^wx;r(1}lPqJV%Gm`nI{bp!N zX8}p8V0-Q^^Em0qUR}#!%XhUf!-;3u2*Zc$gi8@dENVtesGm4)pM@B^;UR_kog)7; zxwLEPF1n>O86hkLfd#J6Jra((z)akkI|rAwFR;MZ$5`IKGWO=)ZkoL!hQ1CvMKcc> z;@W~qcr{=yrfAQm12Hp6>wYlFUK>alm$tM1>>J7vJGK?;6T2doP4yPCa0rp^J*&uUQ4QAsPwzXYYjC2@Oby97~Jq zU8rip7>XR24u#Sf%+-E`jYVnbe~S0e!h%^#^9jLaewt)wdp1*iJBekcA7^sE^Ej)$ z4;DU7U{bmXP-{JnoY|S2Uvv#?j3!`jKt0o07D2gBPtX^StyEZd3{wJ*V{O|tJbcoG z3P)Aylcz*mq|*`Y;D9x8nXK>4Flc|e#Ipg3IP=JjuD{wxFM_j4cl|+xmGaE|Fz&4n z8Y`Yp_(ndrQ>f>ebJ*tI$nU>>==rEver|4}o~b3|abq{FS6W6{&4alwit{{zM?$TQ zvlsWIorXHgAgs|n$wr$bVbYY#xY*AUIvRYRes(lM zD+XfNp~sl`eGnyDjix&Ladc!x1g$!3LK){Y#Pd6<=={}c$~>Y?zZO|h+;}HaObB8g zId569v^-98=8ESD31`pRQiE&~-T#wI=VA(Js^Ea1YEDpiC;9_*|MXCFCqIX_ z(`%l!>q+iBUH+nOC^@a4MX&27BkfQCbG)V|N^eva?K!8bukaolkLlt4X)jFpmCfwz zy9CFE;jG~22P_m0v3Q}`QG%eXNL59`Bm{Af9P zv~Gn`#0fmH4?|^iU-3h76WLupPuTS(mnQGcU3`RI0aD^v?;vB~y1v<9nA2t-V(qx|&YMI_c z{PpWU+8cA8`d6KxAAU=5&N4@+U6mzVd^;S2x&@r*Tg^In-t=XZ9wj*Rd4MCFo3|o% z-eVMUufXywX;IbYCiPKlr6;c#_1W%0NpE*huYFB$U2jAg1FqrF_BagFNkky$X6-BU zrrNzzY2=%gWW}DakhW0PRCk-X^gKs-^?uZ4=SQPstY9-N29E=m!|GEavS$~-_Ol{2 zuM}zT)WP)pdnY!PX3=W@5u~x;2SPLY(X`|LD6D|3&6Z~i8oHt09s!vHeW|;spRh)H zl@QiTP8e)m&A!Z@3DezEsD1HWCKo;c_Fge8Q&kl~j#*4qqKqxhQ>pc9FPgn76Uy;z z6mT*Xie{h(w8H((ZkqQ*ERrkWJS8l zd*nI=k+DtcUIk*%?8iJ}fA1a3$BlRxY|XOi~4`x;%pN^m9oAYC1O zfF3J{a)#hAR#%}b#5ed_d~tluLIcBCX~!hCuRWCS%^y=w%^~8HRNhbGxjq>)bv9+9 zC$rLbMrcY1s&4<4%w2t25|y215j6HBo0PtQz08}=+(z@c(vq!`)6*9SiCHzYIWmvr z3}(#)`IJUab~aogSuDNEi9yQ^O_$H^C2(Sjhjp1Q;I!#$a= zVvuk@w3PMQyb#B$OqhDBGF#s_OV}C7`6iYp;A7)W_pJly&L($CDM%zEw--3jrA#x^ zZ(u+7AJ}n*!~AEQLoE4E>@OoNW^3-C$J_P^_b;>vk9YdBr2!?ZqDhCv@75OrGxxI- za4Uyt6`zmX zJe3x-#Yk#^t)I@kTw}92hPQcwsZ&<8y9!sq^g!j!9{3)$t zub2O2-Pa{h*}IS3*`vriHq)5r#5N&u>ltB?oQcFDV64P#P`bo&Qz`x#YBPJtU`W6q z>b317*5wqS?_e$39nemW#fhTp;em8}K@Du8h9EXsNB9vnMsnkMlW?r;i{SO|rf|Ze zfd!n6k{n2UEwNd-PndLf8j8+nv6(u?+ymUe%4!YR;_Ih{e>M*Eb72mgl=3k4VmW#F zO(u7JPr7LIjm=nZfKA%#+0gShSjY^mhY5s=7|#J@4Q1U8Pmw}z;N<&XvK<% zH|$fJ5{_>^4+Xm}Bzaq~gDonYlTm`xO7Wz=7~+@rokF3N6gwUDOESaI7_UcmGnd{x z_dMh@8;v}s@^BrzJLj@v=jO1w<2-wLHHUJmnqaW~DZEciLU>RHo|pR5{P<_^jk<|! z){eK;rWoq#%K9yR0U%R&tzjWpT=tg4k1AuKc`w+bab;}4@240N)f=|Onk>|$6?5Oe z!*As{JT$eS$r&=36<*Aq{xzT>3ezdB*qZ7W71NF&1JYB-#KR;@>hHIeWHyFUsLoRQ zv3>$Y1xV4hM^!ZB&PdWfSBf1G!>Q1I5k4MK2GPHZZ^BQXSy;B)1=`@YLx19Q?E0FxVjR+{LW^IG) z+2?U#ID7F4iLn!b{NW8hZ1`Cr3z*p?-7#f##Uy1 zWk#)Kj6LG}_M6p|Hsv`PS`H?ct;@)F;31y%oG-3hxQ&#;CQ{eieH8nngjU6u)4|%_ zG`rHD%>Cz6@c{>iZRB;Mz&DOP5eA|Bj#U-=EE#$eG?LPFPTB&3BIylChtS1(Ka?SRW~hcx^*D z7gZ@_NE~Ifji==wzT%2}1l>E3Naa&aX-w>4vex$#&#gZ%%9M{mbh$EIr(I-8EBjF1 zuOc*!H~^_1L0EZA3e`JaF_Zc?D7W&bKXN7Hr#^^Er5jPo_j%{`7NLFEE_$>(j+Pm{ zr2*}SNPqcz8gO$Yx#t*@LU$0>%6ifX&URS)ryQNFiS$M}fG$1OqmQGs=<9#8_&E{; z=|76J*Q*x3;pbrRYaQy`dSkoNZy|dAaY}i6kbD*g&~QPHCNJ#7hN-d?pQ=TlXBv{e zS^;9XqrzP-2Zv@#)BTfOtZGLG`-PINH%Oat$~Am35kSMitbhYqwC5 z-!S@JJD=X<5-$62&cZ>HL97k?PUrou`yRmnb zEK1aD$Y9%atkRI8U;4{PEz*@zdB-qya3kf`e#5pXee`*Bh~?NkU>(WXY_{G|T9+hJ zjk6qe863gc&5JQ~;Rq;tA3%ZcS3HvHf03}-A}XHtHZY^ZK0EBVkO{0^1Fl2K1F zf2J2x{qa_4lkJaW?H~BoZG!Y>Z5%z7f>?!9`158CCRL5$JP2K~dR>5|@Hudh|B77u z=h*z)mfqWguI_Or9kn=G&7&(pD^$tyBIjAXx2FjnGwA%u0hs?J9%>)XV#X2`avW@k z&5p|KapyC%M;RbuoIDnf&cf_BjHI%^6Y?DAcD!i$9#}y3weFB#g z85W-XEp#vL5<}`GXy9hE70h z&?I`cU?9c*a;3O~+U%6=Vb;`Z5A*nu%+5y0;dY{j?|#DGc@T7 zcRG%?He)U3Lsqb#cZvN2KC*nH8s@M> z7r{$a=u~G3e)Nik-PoNttTGs9Hx35zEYDpdTQc1?k7~QFLua`X{H&ib)ic+jI_U^v zM)sok$XF`A)j;c#YiO!s5BPrL&YnLnVZG@P2CfUk5VKvFz_Y;p*2O~QFXvzV3!-dM ziPrud1>08>=xCZ99nesrX~wJ}e^Lhtp_(>{C>J)*P;oB+qZ(;@&-`ihIEuzPQz8xyKK;@T)J{H zhEyb06yE0|U3n8qjpi$8%UMgh+{Sy3fpO54sb~M@-DVMyFND30_9#~1J~_@6*t|}Y z>^r?N?{*rS&vzRUfwB1H>x)@(O7!iBF8$P=M`}!ohDsXI|9BJNQ^KO7hp~4b=dz+J z4p6$svlFg^v5a?Nl(c5>&WkeE_X~sYehNySXd|Lt71L+7qhiED-l1-Qu|h9u3v5O9 zdR2bDO=Pkf--I4D>jXJRWnsz7ACgRu=PY^2F`;8YmqdP{8=`bOSg)lD@N(0pzM+rN z!Lv7WMl3;31r>Cz7=>kvE(r4sa`832C;2(2;jYsFr{{reOIv?){GX9Id% z8G&_$&!KmBGO8^rSb6+zEH?;2SWyweeq`d!wlG>-Y()2ivT(+5z1Tj!jD9a)O;Zij zh22GO*up3IP;5T{_0UInJx7`{Q?$jgZoNf`yocBsSH=$PC`Li@|Mw1D%y5Wzt4CWmkMppByF}hU_S~s6b8jF=ia6@CO{$8;I-r^I80(Wq2#N3WX2$p{vIWY?bX6Sj=i>-*XhK zmL7)G`b%gXAVq^dzM}4fFKE>0Qrc(L%%b`C@Nd)c*x%y~ZL+VTwz(bjGvS&jJ?;u^ z*y{IThp#QR1Y_OJx#lLa5^Cu5ZDo%u7PhcEyf<@@}LHU`d z*kt$(r+hWVgwIYq|38k_Ew7{{+so*u-bu1I%%F8Gel+6XE%Jjq$v(Y>Q$BxDxWfN` zl%01p*YE%T%}_Rxl?GBs$?JJN9v2NnR4UO>8boQ4qQtwhva-ouSw%u2ujeaT8WJj_ zqBN+`)Kav*&(Gh#zkaWCUdKPqah&rU*Y&vW_uGxkOJJdQ+cZ3|Z46FsI*gCA63~`= z&wO&8hwtBUd>;O9d>O4mXDrsHpRy-lz^o*=5--Vl0NSDL`45o2t&dHf*`W2!9$s(b zgRoF9dEm(TTM8!Oi@jmM=W=~hiicrYx*oXv$|M^KlgJG#- zrM+OjdIEx>d;Tg(PsNN?aJDjv(mPQHlH{}L1Hg_iJ+LDgL52ir? z$Jsn4C5lHCf?&&^?U3952+LM8HC_F=n3unT<&_Lwf|SvvWT9po zk*ISbs~)J4x=VUYb6^5d-yFzW#H%+Ae03hwuG_%nd?!-r7K#gR$zaC;S^N>U7KWPn zr264MrZA%gGE@JMtmG6XeP%HFaryDH{+xqSToRus-h=h~jzi;u^<><(1Eke>0r8sh ziu9$*L7S%wN#Hnnr}zAY^-)*h5bFaUTJ7Lscm#Ru6oMIl)}mW}7pTXkkt^u|%*EBc zyjR2i5VpY9i@2IEx1Df+LllI;E zK;k3!L2S}j((9hd9DRgDdUGquUB8erTD`bEQy0zZxD0kxCGTNqD6@3CH@xP~AdHh< zk;duLJV}!YMAzmuV=cTN&M9*&?eB|Vwa7zQYaaqLH5oJ*Os(!o@yBFe#KdG#CX2p1vTuWHZ5Ws>vp_?L^c`08$s(}nb`z>~W^~_nTxBSke>=sVo~rT;sYZRl*wyKSOG~ob79%S3gW%%E<9_f03{`W`=wt`qav7KxEwN)Vj-jP44nI#1V68rkl4|scqnfwu3VD^E^~Uo&VlPj9c+Po z(KT3T&V$0?D^}$Mc&&zFqsmsPB5tnDqrR{@%qT{M;b9 zmVJ@ro#;U6i`&s}c`p1;^oA*E%Set+H>2Rw%k-+w<5ecV9IXP?=@ z)2lV$O>T|l8ECyHY6s7d8D7?8jR)t8pZ6O&bP`~A(OGiCW5s0{G)7Ouy_2rUfFFG-)&$f8RZbb>jr69hi>)dauJ^ z)m5zEo6vpjyKtjl8^r%!4gU;p?%d4q zDJ+!?hfeJcP~LwB=iGjQ`(=b^N!uIPxT%iI5=W9{N-Y@B^$eS$&SJ@txg%c#PwTb zm~g{0rtjrBZoQ2JK31B6-dU^BOlc3j+NW;zxMLpC3EM*!u&hjgn@lb5rUpYUF39o8lFqq zM+2T8Cua*JNUcH>@8`xgIBfC{h6iS$R>Lfs`BRK;j^p9{Fjc5}-2=su-1*==6}TH9 z3pFC+!Otp_e6SiO)b|6E`1B8+?N^}Ak4iDwWDjOKM`D$IBHAsg!L|0=P(LmUbae{Q z@&0Udc^wGexpQ$;i}PM@YeA7?@=We#eS$Tbq-RPPRzEn0ilL8D+tm!6mnP#UpZTab z)e7QYTA;6LE=D}K%ef?;K*P>+ zX&BpcikR&cgS5Bm#Ct{sy>6RA?Z0@@dciL2v$I7r8zubalZ4jtd#U-!DC+a*0&H|H z#{gqtYU{?Mj@6d*iQH29w6BlY>+(s-6=$OGHvv*dYCz_1Ebdv)!?#Ticu>(2Mva8& z=5l*{y>A_Sn;(rCp8crwLyRigPoRdW^RQ)rCF{8BQPg;Zbm*=oyDYL%C~hK6+W!~p zuJwUfyAZxy^p%{SvKKc7Dr4P-0_bN)N%(Ir=lpI1rd_#%&$nh{W3C=%gk8jCbg?uRNT%c_`o??wKhYEQGK1 zt;pFrYoZbrg-2g)CR;|wnT}nLz(nbCR95Lj%WQ90XFVNcpYfQ>e(@OLIDljM!Zf%( z2@OZ$aVxh!dTQ-SE6(@e7`GRTo=^>B=stX})WR1DZ7^867xN4YP|QsPlww$-^3j7? zyh;I&=&Rtqh!TkL-%f+mqG*$Mg=kM!fn@!AZ-x=yh5dzh|Yvw1(wm-6j>NDN=`l{F!j~ z(GucXG(f7jb@;dYF5G-;1SZLhCy$jH;pRkdtW?b9X3}w+T!Us9{bBTGY-NSKK?cn$}nT17Li=DJ6|<%X#45s5m!AEExnIq=vA5NOWe-jJ6l z_$Na})&}C3^-Wwj_aXj$vhon`AGhx29GF0LcgIfR*D zwE~8c)bY2NHKr{##y?Us&}i35Fd8?6Kd~O43|-||-P6hQ%&XXaD~0A;6v4*Bd*PhT zPjYSZH4NPwius!)FwSc;ydOod{%k}lx5^T9*#y6JZ1K&KuOO=J3ac#7kTd^6V8-1K zAhTjRMtE7$w?>J$K43d0am?fc?dqtRc#QKF&7zkMPNa{vvG_|mh%9RACpTKBF;Cls zs847)ifGBA`*A0H+`0dn1~!_VKM^X)inx^*8Klun_UMif3s zKZ`S!`|u#Y2#4DYP;qt*iQ3A|wbjPqH_lf)>yZR%9=i{()WYfBC3)!lUW}Nu?#7o1 zIq>aSH*@vIbY9~mduGnmR3=3|2s>1bFlFa`xOu*sy#IO({yljF)uk;U(EJRcoA<#$ z$tX(5l%V|wO^k__+IzX*Celbg`kEg;_I-9hY-(+@w`~yCb7qC-bh~C=P40*@9h?JimekI6vhBDul5d7Jh0>7-fi1wGIsQ6GB z&1W0{YRQ0q^d*j2rir!3*Pur<5ATL9#k`CdbW@4N+45H*X59~%X+H&5aQBmO!!xk| zn}Pgg8uYx;9MpP$7hErg5gkbtJX*XNxymnUeicU9hkal(hwDm2^mA*>5n{PS0nF=X z;jSNgILMs?caunBc0rBlJe|bc@cP8%x-*&5&M2Ok{~55#I}QG|h43;q5*BmW=abvQ zVO&8t{Gxp5?NRCG| zlTFt5nA|nN@X8<-x;SUT%H%GvUgVEj&b=IO&mBINSwP|&HPD!L9NX7SrX^dv=%wZx zpwiF)LfeDkp!W=z^~?dhrn|$5H6u*Oye~}V#`}!rgO9L!*G1g+-vAyu^oB%?B|_xD zEY6#=6&u;3_R;^F0BK0XpAP13}%@wKq`b2!m(7sB9*On6(#GBzq)5D#3j?2^+ z96O1E@?22-TLL#Y28D^l5$v(wiwmru!5fe194FEf3prk!aAgS0a_k_#;>F?PeQC5` zqk!jkIpDfXC+0qvp>xeGhnowzjPa`yFgG~^7kLV}QQQl5#-}iqH@Pg%z;ZO?G3e0n z0jdQiIImv}vkm?*KWjwcz%c~~u)2bIqZYW@EdvtlCV<-SD2|6FismuTQ9@@bwSIPh z{(MR3fm|W_#b*q(imPyqUlEpF)x_n-yNT|^9gJQ4ad4B1;GiKNtoapiLNjKrbqd3GuwXhL*)e2 zNhZL4Q3~t0E}PX(Gt^q%irHK2@yL3vB@%BAX2Q4e^(0|z(NPAm9TMc#)^X50i(~D~ z48aQrD{$c;$C@~&OqM;h+(2RYF72;XAYk3>F8Ze!xZ`USH@SCViXlAlKWSLVwgUntN zb@D}!&Ajx@AZ0G+$m}1E#Gsf$#r-vq)|v(uZ#gGPWiW&^NW-(d8@TCF22M9l21k#L zc;Z?pI^~z(>;+Hok-8L>%Wp)t$>rSq_%hs)No{OXCv1`x&h0orO>`V4u2<0W630A+-^|~YJcVM@jZ@Lr=Ez7=%E@C@g2hN%Y?{dyx!%UkPJ(bd6H-0rioS_5otfRrS=p&^5lA}d>6X~G$EzBRCN8D|Bkf}Bc zkH&q%MUI`ASt*0tWi&HWAw&O~v3&V}fZF#{7f^3X*IF+bM|_bfHR z>0HL5Lrscf<%ZyvbMG1Fw+lh2K@Shz{XpJ@^};)Gb4(3u!~GjXX|tjN{q#=`Htq|8 zItvFhX_Td!l2=eIxd*J_JRV+o6N>^A(Pw%VmXI&FHRl`#-jT*NAB53J_9u+v_BhY| zn$c>tE$4u`gE7|HG-i;}_!=cNe-eNL2e;y^Gnzzu8CY8o8KgN?Hh8!qn|8IJ57oc1_|tbMvDF>F`B-ANW~3VdG6Z$Oc0v`Tugcr9RLDBl-_^kRf<7{0=eppPv*Jq#Mj(QjTGS`Z9c$K3D z_b$s1JPpAwxz6Xy@i=b&44Sh~4DM;`GpiEk5znzqJoM-px8I9E<$)!z><6DY9^^yl z*E5jzWfC!s(}p4EV90UyhF*ta_;qm}*=ySX`+SaL)$6ag;Bq&YS+T?0kLxic@fptN zx+V8{f$+9qDvE4~#{N&*=+E_Z>?&`e=m{6@f0D)rH>03`4!1^YnF*mQ6k&0|a*#bJ zhv7dH@NtYG_c!`sGH)A*hnq8Zc6}q9#TYVVq~ZAXBG8?~hil6#q5mky8aWz`{Og&_ z_|uZaz@mU>U$zU%xlEPYlB1aYd@JOwNQMFtH?-x}ehb1Lz_~RXyS=l4nf#-LwEHbV zsn~zuS09Zptx~a+eL!rU$C7m>Axyi`X-53lc$nXv0OA{$5w2wpJ4F(}U9pu2>QcxJ z&(*N4WjvF(`ylC?I~!9s`eCvG=V%j>V&a`7pucx3{-?57a1-;TcnQ?XhvA6P2(d`jCb>o0vznJrE`=Ahwfn_f)!6aWhOcqYUff8pBA3qUqPz`({Gr(BK9RSt*O_;{F zM6z}Tx>j|-*P=TlZa5$2rs?3gn`LNZ@C2{Cx{T?PX>{|KBh;w<7+R_=#CghqwVM0M zeC{6faFb=0jEFG_t~2nz1-sBFk6Z7L@$kuCu0!dt4TKH|z+P!4+*4i)(Y#e$R;mg= znEK-Q#LG|`5r%WxCF#$^sxb#Z6}FenK+~6@S21y$RQn**OA#5-I&YQv|-iYH*!i^0tT&4K&(bJ zROLp(&d@X>z3C^jyday&=-Pp*UOFhb(Vx_b2hx10N?f!l5A<7&h}=9EvgGMza{ljM zeAgX7Lu35uO$`$qr@k43#?Qj5J&3otK46vbV*Hn&gRl1tLuDD)kA0*@*(1?3WK$c> zxp14~y!t~r@mEtO zb{0)Vr#1=PmMDgMQ#E0I@^0um>VV$yL6~qd2NmAM5`V{WztKXAMVjPVqRZnDHF- z)`Z}HYzJ0e=;V0sb@(AmmptlHf}>gM$kiWw@^a~C2rLRj1CLZ(dL;xU{_e&-2{Un3 zL<;x>g~L476dP;A>F$%3boC-xS}WCzM|C7HQt>tt!-?Q1tj25J^qR5da$H;PF2aL2 z5#5gQaP6i~FvFpO{_C4ek0_)<`l?QRS?WqRj)hSB^#|zzmh173GZ(vHi;H056k9d~aYbxvuet3M9Xw$^LZ zExd<0U-pqXrapz|I6;9PO?Sn;T$Vs%!8H^zJ&WHCb8})pH9X&Z9r%S>;4Uu3obPJp zRm+d@ULFB-sBgfwl`punqwC=R?A%)pDrJ(4!I;p+Y&U8OAqU)cgV;9F2x#jvA zz0!neweVkTdTE3`bBef`Y#iG61yh$MWeR`xAicC5k7zB!lqq(oyF?q~xaVbVCIfHJ zmEh#8ID92H8K>-ALyz%V@z_QaF!^psUU^R9b-P`g=dtoXdcaSDem+(RR}-EyyFDfl zJ11u*Z{h-)D*6unE#sl2LJzic~Bxa&96S_N`NqkvK4<6O0YB$yJFW15T!m;*Xl@J;FcmzZwF2Rwr zn_%7J2;68r2e1KkYc+EUh1K9 zydRCXO{>}O%(O?2%%5AIm{}7zhplu3tl4S?Zg!$f4sSPaZGtn|bU=;nYnDKb(_SQH zK_HxB?cqg@Ex8mP0>_NaIG0p1TJ{TZ-M3q$dSpD2$gP6Y96vja>q)LQ=t89@%!qZ6Hz2`n9MHHBT840S=S|p=~GmQ;@>F7cj^X6 zlGVfW*NgG|f8rRlKNfqwm*Z65`S?;e2JE)&C(qy3GM#EK$GD(QuTlNWmIJ@I2Z9Ck1R}X9p7eSlR8nDRIhSP3txGMStHr>C7YxhfIevJyg z=GbL9A6H@Vcu|y=a>FYf!r1BH!gc5;GJo@mQFmt|DxUleCmxqF-@;}xM{YGSOW9>~ zEOi_`#LYQ^6kd`PsZ%8Sx+6^N*nyud`ti~uH##ox7nzu*3c98Dh`;wZ$ozW%UV09I z_T&Qi>NgWU)J2i_+AF9s%^kgUF2hK4=?7sk3FE^QHxy&TtOp+efgY z@g!rwCSxS%}ST1hOr#C#Nsx2xLF3Kiax~$^DD55%Q(N< z|B0ynE`R{PE?8+X9poRS5ewxLM2*WjB$U;F<;O9y`-KR7FD*(J8(Cu?=Z%Wk5s&+Q zD!8p@JT%`E#v|48_;9^B{_vTDHbXU#{5^?OOV*P4?%YjyCmD_VxLhx99V~Q@LT9l6 zjJ-VrJMJAN-OCI>WA`_5q=W0T&p40gKHkEeZl}S_r3n&j?!jr!1><>80d}Z&aG3*Z zjvL0H?Buz?+GUZ{3FaiebuV$+U55YSPvg67a=6k(97aBrk%s$Qh}*?Sa4%K}Yq(sb z+IB1ORUUx%_0ga=GYw6jiBk0QrMI>QQpXm1dT8lf%pc?2gD$Fg|Kt(axkjECXwD{? zWlu;;^)SdJXkuUGSs3Q`ki`vQup%H8QoO@JQLP@om!NQf9N3aiJX(Gcnc`Xy3B3m&rPbkU*hCaRAcS^ah~7_=P=C1{4xi-CgnL5K zx~K(0%`5Tw9a|D-l1i?g^dcAHHh|BF9UgqXml`|r8JVN67>O_C%v}C=NGTkEgJ}s+ z{YC}?@Q;@;%Qa zsSuupm=lB9Q%J&}AhK3q%*?%Wl3DMS!z6zkB`2~5i1z&_j9uI;u-u_YJch)G^9(WY z*44$v;!+Ifx&R6_Vfe}}gITa8361rhV6ntMT+UC%X^tA)p3w{(HGjewtwg0#DZ2PO z*Ks#mfpJa*SA9A`ho0}Gwh8f{!Xaq(wa+%w4z)6dEg z17i^;*ykPCd3`1m4=*AbLy6#Zbaic=04x? z{JjQ@J6nit$Kr|ej~lS#_z4t?ng@P1%fXJB4-)Jya9tb%y4;-9;}Ca_dczbfTL;Lz zv|i@!!L7{F#+{_Y+#8*K9>bIm&mbZG7P)#!9Y3Bqg)d&ZfYio5nDgQx<3DL0gZuF=_4lJY-v9~B0`U{-NSzim7aBw0Xm4Cr?3#!0L#1Xo` z+oC_m%uS44hcjI~;S874FJ56vf7u^Em&R-i^m_=o*+GzID}q~&hJZ#wCGHzrgkJ}4 zz!g1jJTSyyAJ?~?eB&6}WbMb(|KkBqe1{v_Cet}pljx}AB#7tOp);4yph6jc!SK>+ zDDZ581#`cEY~y5RSK@ z-_EDci_|C>w4qn23eMx6Aww}X=(OtrGhS*93HV8g!4DsV+^7!8d5b2!qI6$V2}vG{f#1z;Fn8j8vUv7* ze8%Mz_H|CDQIfOhnFH$dTGT_}$6SGvufKrz=`eIi_r$ns9bC6kKy5xp(Tt~$(fW-8 zuB`cn{gbQlVu(DHsf)m|3p{cq@(kJq4`Dm!e1#RcpyJvAmU2PBd&>DVdpM8vuo!-J z-AR9T^XMeIGV~iV#6405D7sJ;-yD1kq049C9`3t)5|5U7j$p&iy_lsGLjsTGll1Ng zsO&#NU%vRk%^9A;_`BBha+wP4GVsOQb2%pP79pIxJP>ta@1VbiKkgrG1#@9LlG8Jl zsON_>=XB)gw%C98aOgdkk=DogOBQ1(HwQDU&*b`mMS$8_P|N=bN5WU2i*GK5b#M1H6dvz`SoLtYq8uUSGg zeuTlfX-%-H>Ltj1*Z@L{MaV35Y3A))lg7vV!x~?8zoIi>8C`E+=@Q z5f0;5t^vrGBV!thB<1rMQF%~IOz#~h^SCTM!(RX{5wh^++b(!ya39VM*F$Gh0tCMG z1HakPVD)t+{Ju9E#+u^6w`wA^xtl_)gFY;<3-~yn#XKx%pJrNFt*#sq3bMpyc}|qNIw`O5#_%K zG)%<`D`~v@S_vm`=hcE^VK8-1D6I3O0c^Wz3budkK@t9u#W_)AvG*HNRR0!Sg0F)_!GBP}b>_Bm z>xLuh0Hv-;@Wp8d1Re;2FGcz|J9Z8x4^HNK`mey~R3AWv4&LJYwrlKrIY(+6EO1hV zfOW#4V^B*pLaSl?Spl^Ck6|>-%{Izi!Tm)s?D@A1mTf46laEirD|=g57A6ldFCAga zKs5B~?}x~RwWPaP0(w?i!?sUy@PjM>p2I@$3}_=&vHx*>qutO_&vm@k4HMxN`VgtQ z3f8zSfqGeMxaH?S%6cBchQ>z7mH;^8I|5RE^{`j%G)&l_!DaU~v0PppbE=o1Sm<0- z)7HbwR@vapv4DFyMqHIeF==tHBz57|MEy-QTz7pAk0*Ww{nS3-F<+sI>*&wCJ_EkC zBJ>tSkan*sSj4?=Y8_93ZF&PYGrj~rIEQ<@c@U0Lck)el5tP5MhjfE6GAPZ5u%pLd zTWkU(v~PtEO;h0AnM}BKLKUp-=Yg`h9he?@Le?5{{`r=P@Oa}(CL%wNnG){7$W03) zb-}Ah=IYPPgF92fgyUR9bDWa(h5v}pv{F(eJq}halaQQecW9%~bz!0ZeOE0L6%!JY zloa}ZJ{m$q$aViV&z%Q7jNM&*U5(L{M|fl&gN!L*z{YZ((1!p2Pq_GP_M~nl%~9^8 z?;s3?(~79^jPvwD(p@}y;}$q7DZ_5}5w^>|72S`FGyC8k#u`Ve2$U2u>7RYmxg4Q3 z`)c`;Sd&S88~kbO`$D>EYC4SxSxU_hy3m?i1+0bV32MFR72Q&X z^jgIp8ff89PwnbtQ)Y>q1sD8>8^?~&E7ki2lpjJbid9n0Ct38r{*$!ieFWu|ZJ}u= z-_sfSYB+V>K^!#tj)!aVsPUq|wAN9`?AKGyt!LiBhF|-_PB!{S^-SV1Mtd>mFZz$J zUJ^$uzL(N7dZpBAN;*{)zDfB8$#jLC2lWuzOXGJ)v4zTR{KrqcagLjiV9l$qd~qo~ z)?-OLJN@i<`uO2@sx-qH*?dj3`aT!wRcW)>1WB{}f6@Y@16j0e)*Cu0`zxYkA=N)> zFHoL-oXUQ<%)aiNB3NHDNH+;|AYWLDx|Xb?1))z^iN%s;*SDQ!U0zFF+!xVJvu4_{ z>$F0ticK^vY}OE*U0KQwEFOgQnugTk+#JFC6Q9^vs>P4woNu(n=_t@VC%c;f5 zgS2>xfIhmpof_1yrgkDpv|FkX*MEG)xB41Pb?VD7;>Juib?R1jp~g#kZbu+}bhv^( zu%9lt;Bkp&a{1GQakps0uSIlpLos{&bOQ}-3#G!-PO*kx4^z$NF?LX7oSAK(9t0f@ zr+T4M7~(O)UjH{qFn{efzG|O7Yw_^~Rd#HqQ)V^5_5J^$R8J}z$UdQ4l0s>Z!z6l6 z&_FL}$4~{0cPwv$D(Aj!s2gH7u}|8Y*~<8nRK_|6TFo4&X>S5uI<<_}w4I^)x&P7R zk6BbunZ@}k7GP6nBF;5+qWwWK80^zPeKr>f`d%KQn=M@F*;`N8E&iHRdcQj!+B6=U zz?g3kKOcmK9R)$B^67)|KD4iSoLSpGS;`5D@KLrJHSx@*q1|oth+rcf8g`h6| zdP8C$JZZ!;BFF)@7ucw2m}GN_`N`|Eg}bv(cK)xUdWj0|h9&_6df#bKb2L zbLmONHMD8vYlMR-SpItyMW;Kno}J(ET*+7Jv~IFs>p^aQa3~h%uZqK}Y-!xmdV%i! zFp58RcVYWL5uI;lLg%%K)7U4YIQ;As3~a8)9EIgn!>j@s{Tiy1aEGdkTF}$n9%&I+ z)79R3_-N82EZ%W|zh-F|tO*LKZEt8 zRO}gCL+R>7>UE}>uB(cm_F|`KU-~h6;Cw6%Oddd;vF-HNi~y=XZYfStJI@Zhi3df` zFPN{FN5>^RLzb`IugrT=K&{YTHs8Z}8eX)H6?Vg6w#m^SeFc(8wJt|LUJ6@(m_bx%hOIzxoJD#d^ zAjkSyhUySTwOySBS3hrIgYzCTqMjl6&qtXKJv+~Fa7^gRSvAyoRUMV&<{{^u5~)L1 zA^r5p5S5RY;B&!Zs@EA##g2#3rhT{QmzmFS{Z<#cN>qpL-=u==X}u8h*NN_)&NbDGOp zaY=L<9h@$pbG!_xyvz|AHs(oxhJT=4k2ecG?})3DdZQ{>>3JTdheFWF+5iJ=!l31I zF40_I%RXOG%`S`(XI(DjvEPrrWW7vd_+Qem@i+De3nGe2SnvM#Y~z8|f)oQu!4&ra zmP^#32X{X2`aBY@$pynrgCPFbj2rB*T_1^(Da1?FL-;z#06e&Jl9J$cEDT&>e@v@j zmv3%{|3no9_jL3G`A20185J9GoOA^!t7TmD*_zp}yA8myF|fc>fciY+WW7%Wu{P>6ep>`a^%QdjcE2AM=YzqILYa?dAJINBA8~pN( z*RcKjJ$BE&^@6n)>a{su1JHDY@Gbo+*n1141kqV)g7-?B*{^|rutL!RCbrh_XMK&} zf3v^KUY}CME;@FceW7rR-R3-qCYO{29;#Ge`(G1(!caau{M^g$OYdjJH*6C8ndmAQ z(O?8}`euT@ye-Z{3R&jXJH0EN);`!`bWmVLjmch3QacAd;RPm4RS7k467sa4SMMUg# z6|1mv44yT`ve&&71WLlfY+J4Xt)EA;{sn4+9{w~z(~g@MQ!7j(U2@pOu`t&5?-zEt zPKn?m5f%j2K47m)O%P1jxP%h_R5m{BM_paseRi#U9e;i2DZIMyCp#rglO0y-;Ezha z#`k58f?5Y9L2dUHAoKU*?8Gdb=jzIe=knNPNgnL?;>EOULIO=V^$_P(d?N$PjS;>* zMcob+bbQS*kJMf8xv?m{a&i+jRGm(53+G`9=OFAk9)#=be9)#vk8;m_ieW2g_MPqY zH^(9!`6fYoyCX=qvjnaV5MbO*W9%?VpgKLmczUE2r@rB1V$&(Kzx@kkOh?fya09jd zb_pN%f57oi4XHt9F%_B=NL^gb(L6PrALv!fuaXL)Hc=&1e1Ou)Xn%PtrMC+qrNp^IQO3XkvD+B|BzOA`_NCa(KPxHANA+& zr)MlC(wPeq>Cu94DrK)ovm?J@!<-KO3y}=0cshkYJ4qX_n>Ul*A0~8Hl0Oae(?{6_ zdNlB&InCan%`rNsQuVLFxcNaIHur9()-~?Ee^j7!!6f$KB_39+xPt1J z45?P83H393&EL|slg`P$h!$b3$WhTyW)6pLiy6ZsSGsXo`XEYvilK8QW$Dy}9GcS@ zN8`9Ghn=-9=ek*p<7Az2r+go*xEM*dswdF5^9OM7)J=SIRhY(kY0$POC+Yp0@pMy8 z7aq}@OU5_Z06npQLY*~T5>SLXUwiQMnIUwYs7NIZ32tg}q0c@C(!X^UwEvGZmR8TC z`?S3AvW7gCAG=0+Pt~J#Vj@P=3_{cGrO57i4jIxl{NssHm@{NuCx26t`s)SLC%O{2 zN-_)kk1EmFcdKYRETG#jJJEI0z=vHfRIx7^o0VqbI`vY?-2jr zTRDE*fGS^ef;df3SW7?1D$%)S*GQ0uK4xBt#y9U8aooIr`0cGEz4u@({^_60GL!V+ zxMe3=wq#HXT}SFP?JJkw`zp80o;i+}0P<378g0DFWs^-#;JVgs5L6bClY6@$^Q;8MT5x&FXT~H_ z&4QgHuZ0exvUF#gEG0_)L@{Z&&X=1z#!WxWDjH3@c>eJf_HyoJfCyoDm*va$9q-2@ z_vWG@mo-+l9)L|Bmg2IxwV3v-1C$5mv(4h(>~haBh}F_z%LNgvYf~YHE&f7qr<@?e z+m%~aUqZp&a@11$&6l2X9h|Rekk7S<7141R^XMcdzgdlkWV86U%wM79FijK<j6a7q{W(W6}JT?{{HUN*-!-3DLH-3jDt=6?JXfo58F968l+aL)|^^An>bQ zfWtl7e5wC7u;;$(vGv<;vj^nP;+_0a{*Zku-~HcD#%lE*++Dn#H2i&ro^c^{h5CwE z1-ZENMk&)5_>#D$utiQ!(6!z)E70cUj z{ceBsu&ZYGUS34?r^?d7ysvfn@?q?xxDYI^;j)n2{D}X$8tOL7(xc}-vD3{f*w^_s z>>c}3+?=HX`Qg&EyVjP9rxel?2lrFqD>G>Ohw*gYZ5`^C#>XUqA$?9agLs>-pu6iD z3x*viukTDBaeSrDOWfN3z6ez~YfYz~JwfrL7~R&$J%8U?(07Y}0p zH94V3i$EMa#~n?k)(tYWaMKa=e!m)n&kokfnD}7EKrG+p@&U?sF`~Pc zD^lr=FX7Ex4|?fdF7>v|qnVvTIR0t~P0VtoBR2!6Lty}G^ym^w`B(Bc&)H9g9+}g> z51!+~|DHo~|5bMIAfI1QB*D%M+wqqu$8k$sk)gj$}9QP)})|GD2p(c`mmiTGgMrQ@ZnU*#pX()=oW zU~f0B@pPfCZ=Ym`-0!?`Cev8M#)^qp7@+!#5|HvD%H zw?DRJ&%T)=u>5@iZ%?gfT`$=4OAJ-n{rkHhLOPsYSy%|I4^H6w`N3@dt`I6Z?>fT8 z$@I}@b*i<}j%xMHsZ)770TcB9urJc{!Sk~Lx&g;(C`jjvEcT$^O)GJf^TKTj2uAj7 z5IY=K&mK8rjJlco__xm}vHE5m>`3aui`|@K55;1M)(Rn3SmX;h#XE9tG>ZEwGBLjL zFuYCMjlT1~!Y*+yj8zxopJngYs%Lca`}Y8QxShiY{hACtv#-LIoFjNnX$91}rLvzK zI9F-TMD%uN>W)d9vWeW>S1~Mto#)6yH9H>uY7E2)67Kw&o66y1&u=jMqRQVOXNC>M zx_G2q8NZDhV#ll@V!kYq@0%CFF>X2`_DUfC(!SUHrII1Y{7JzxwmlbJO;urH#btiv zP$9fNGY%wHO~q*)zxn(%U2t#fXGs2%$G_DT!Pcz2#J^^vj!_q5*o^2-{&+=-Nsm-v zb&D_Hzu6d7Ii2k<+|KV6&*b{3d1S2b8vpk3RcIG5i&Y=*$qs5GsHMHf3xd#$>#Ks_Rxbe@qP3Evw0zC8BJh>T3McMRCvnb#^A; zRCe9}H%l@nQzT_7$#~Aac7p7~$vl!Eq{7L>PjHkufZL*3InTFQ5+!GFXiv{*11Z8&z^U z1~m=(luh~0D5P8mt;e%z&b}mu?KfXALFF(TYs1;p&i8zKS4-G0r$BF5lvboPDR zBq&ih$L^No_Ek%(!}EGssN>4_WK|*)ohl70#f_<=##zBrS#|g_?iyqHNerH;xWMRW zcjhU{WTJ*$g|MV`H2AO@^?f-4TAh8tBFBOG^wXI}`cSc+ zdJWo%P0NnsMbmr?^1hGrTiY@Dn}RS%+z&%@WQ7sKs?dJ`A3av)U{ri5jL=#=~EF2TgTVPJ)75dG56W*PB3@57Z#UHAj7WEFI-Xr5_;B^6&9udqAms8+ke1<}!<~@v``xpZU zD+;#@!*Sf!yEv;xQh4@Z6wdWHiiejpPWEiXx&htz z+`Swtm#PSDr~ZWlcoBU0QhI`#v^-Enq{!w9!b#9ktd{95UYsO~b}e89_gK zc%&gUpIeV}J|^Jf)R}nm(khG#9Ebq{6|`Hqh`NlPP7A~enM=o7Fy84d{?N}vi!)jD zg6%1&a9xb&lON#?y^|QLTY~0;O!3*o&A2Zs67@^RW3FvG9cfZQ4F;!S+DjRs?~X)# z6*W{CCsY@9PSX?yt$D{->DUpOi7%M>+j%63^8>$)Yaj$ToMQltXSF5wgJJE7F({V13@GMawB%C!YHsjzdK776rnB}b&n;D}}#IVO7o zyq9qPt)E6DRFP}fy-hH*R)H*SQ6}n%spS0PeI!6j8cPjF!i+@|iTsHgxVd9LX>~3j zzv^1Rr0^Os-IGBk4*dZRK3Zg<%W=UZrF3Sm!&qV#_nBJxL4)_86n3&CQ?W`Mn=QInM$PKQUVRSa*daLOdj@W1i;C}eUNc% z6cIQbVo6&NyK`3(QyHsH1{V7X$`^D)_>U3TswhSR#~&wi4hO@&@3utc&O`Du+ZGP( zI0la|S(2Kub72m5SIO{-B*7lJ7tB~p0K2gUWbu9vYI#eZOq0Ueivvg5hwfnBS`jV&PS0I0`-Qgp?i}#S>zl*=)z!7 zGL|P%vTF!&I!fl8Y~W&BAWh6_rGW)~X~@S_r2pz6MBaD~*|^w_Y-l)7r1naXlruUI zmRv%vSIr|b8wWCza^(bG2Ein1?{T=aU?a)S=ud*H)|1gVo+Le8MmoB_@eR%gLPv-O z+^q;BR&9$&YiRn$ewCDLD6~fSt#;Vyae2 zle0T#K=L;-;ig(!;rKa0=rE!g^-n*? zs6G3UZ-Qt&L`%5QwEt7@gejlYgi|7=ghmfMK;!T?nt%KN>>bvQL!lXZ`&gjMBX z;B%@C(&ut#Foee;P^^Hp4p#uLuuB-TM~ke<MhA@IBCp3%8geCDt!X6^7AWXc^pqaq>Px4fFk?A8=cj?bm0 zFHWH6$TVd3H-fC$Srq8@6QX?N??dyPKT4P6UtNOT1G?cv^er$-97vXA`q6~>r_kqS z8bt9s&}g}tFn?7IW)%rAaQ9^liQ)b@;h=D7GAa zi3=yI3ik)5(t?!-@ZD8y=Fk`&VVz$Lrnm2>&#!#K#i@OTgYHJbW``i0UQ|FOw`F2= z;2eltW`Gt!hj8dHAd40~fqC0+L7a;&81;FExfgOU?#2Qdb;}nD?`lCpyqs`i|7z3; zD#G9?`Ium%DO4PuMAaKEQgh8{8f&{8eWjN%&WD50?vOYcP@RIMTjH?obs834NJmyY z9nFqw32&ZOhOHgBP&vE_P@t0ltvn@2seL05_bFpbz8wJ1om_0LmB*N!LmC*z-D>dZ z$uuZBH3A;B`!Lz9E7=Xx{-R+e8<<#jDBN4<1n-NlFiOJ)vkSwM^5V`fhZ!A`bQo5$ z)hU-)hYMkJqtGfdTqn@Tt?2l`g!+JTfm}x?T)tkGiKaRNjhe^qo#i zT=d}8*gU#u@4*0h1z_W0k0Y~1m;Ww<8`5)m+~kF zuIGs{X}9h%MYD&~{L?`5n)<*F2W0o{E@n(0=)ktJPPXqt8M^yQ1>^jBDx>q|1Ssx` z5zJdEpb_hwsEYq^aQwy6v{*~p5hMc-`n+Q_t_9Kj@n}%6c21bf5ukX%o=Kz~gk_jyO8G z%!-aNi=e;0j$!lpoX_R^L>RqF7it$C8GO;Nbl*xF3-Lufvn!^tzEy z)_4JUvj;F;QUy%EVOoq+r4;1coCzUwi1*EKjTMJS>Zv4H^{#|4*jFY3ZHxx7b;w6 zp}e(Ws8`chc(P<5(z*zkm#aVr-cLlERnsx=jjGUpO*aPKrO#F8DB(tDnII<5Mwn&n2AXR*k9>ebDX7QPgg5 zMXzCz_%2>dD6Od}49Joa&e4<*wt42@&&iAE_f0+!e}6oF)I3C~OAK6z+)10JUBi)Q z#Do;V!m|Lp zR~e3}MPKm6^F*wQd5?MDxU{}oiAO>iT;So1??X%(tKcHRsZpn>;YnRclY7b^{V);6 z?pXmDe0geV*G>C|RxuBSoX_fFC+xYUNjH_gfOD#o$=&n)X>imC#ypOYkP zn=MAJ2~SYB!IWCGUf`>`7;t$X#BVmTBo<*!@LlsNvuyPhSle)!PW!CM%GTY0ZELSm z4Z|ChwM+*U-YwX;^#jZqQw@BREY{8KHt#HV1N~jK`5>dKg{M9>fZ4P~)Twm}do#Bl zs&YeF$rBG5G1L8E_Hdj)-+Cuyx9^4Obbp%hY&@uyr9#cg&GhNSG4w@e07z{1qgGxg z>Dsjc@bRo0oP96>miQ{dal=96fu<5YcKaLS<29Xq@1O?BSKHa}1Kb*N^--qlp&|@y z^n&IS6>QUy@6dPG9GD<`jbi(9`gZg@{@&^0q{~B|b}wAQOkJ;!=K|l*oSQj}vmSxg zw=6s{UO>C@e$w&_xlHp4XIL~LlWA7r!PPB)!G|&p{_gdoiP#lgxT11_Ii{e<)6z|* zx{1+LbN3Y{TCI(49z7U6pH{M!UPmc?+>iIrWFm14UI`ixx&75j!SGJIKrp4P4`I9g z$^8L3f>HIWY0;fxR(@$Q9b`NgH`azx4K;td(lC}?v9=G13SE>}^KKS-o^lu@>b5}V zmU^nRd=I@gFOv@8O{C7@HPGBJgjwqpOYIH&kwx{(f!96=rtVAQZ;el6_2vh&@Kg)r z*QL^#k6Nfhy&rcD>ie_Fe9!4L&bEL<3T#RQOPr54Cf$se9it41tGUYRWvQL9f zvhL1C%-(dkLamutC=vNKddTM zgmTHnV7i$*2T;A1O1o6hnA$n8>%0Y7W_*#cbG2tPT+y~oYB{Y-x1g5w&GePD8EoiR z0p0s3l{(o0qvYb@>e1V9i@WE-^=%-1b9*s!F)<1fx=z4{v4couhaE}woJ_uLdd|9M z2GfkCGGwIwTV}E9ML|w#22;GRhFLn%pWPMU&x~8H2Q!1NvHRMO^JZlju>mj&s;Xqj zz;07=SNsB;a^42+S$^<(cndIBg)me-32J?lpjz%ZHCEeBCUA3?=1U1o&7`5sS3^EL zcvZocM2#mBAHCtxjJcqz-j}qI1X!cE0Y+R3BHC68WaNz+@b!sfCoX)*q<>ul{RYUv zf_^JtlvfxyrG&tZyM)xge*=Zm#$?47b@EL4I{5#ZK?0KbL}B-5*b^56awCS4n)-Y= zxKDwIbv3{@J%-E^n+-vm35;>zgQltp#P@Mxoo0>(iR5-zws$4G^kA6;uW&}$^BG7c z*RWAT$CH4YI;12*kq$^RBGNy1L#Qt2x4y>3Nq^V|>wR8Bw4DQtd+JVxYIj3l?`_a< zVHI&&VMvm+$CBh9!%18AT=0J>MNT!Xq<7a|5v0w13*)oeU|yCI*>vh2e|4P+!_FjR z+nl4!&e=&IKKUV=f9w~tpo?28cw`1*^)pE)kBhlAeJ-@<%^@D9(nM_=cW7Y4?v0QH@;*) zX>eK)&+amv#;_A7Fw1c$G0-i8wezCEuy_JF5X>i0sdZp2#v^mby?_(B7UV~sKWyKX zK`j$=*mIv8AllCm?oSy*TECmpkQ*{&@Av@nQujMsI6se7^Ii|rUUB!+#)rZxQ%~-k zcZJ}XPcAsBO=f<+d&exS-9p@~*HP`Od9Y%*Ho0D{MMI<&$aC2>Fl}WqQyf2-v|Lss zzusS^KS!T|@^_P9PQ^h`y5<3uTde7+M@#q<`o-n>7A?f`WdS(ofHswT;lXcJjpp{6 zHPSDhQpgrNfo*mg4O)JiZ}w^e+yieNQo_g1 z9jtet8y<0x!!@lh=q(LTYW^`9a)Go!I?nPtw?*rPTXEDpP4_ zLN&7zX(ZSG7(yzWnK1~M&~i#kYAHG0L96cfLrb}h=v7)p%S)!>{J>QR3wGcg%e&Nm z3x%2YCE@G2KB(7sCro-R4@XCSV(#_1KsVnIu&dU&V3Ty7K>zz`I6T{fJry{gUdkGb zF<L1sow+^Qd<&q8Uv}WJBj2dXc{_~FQNn{=DuRcT zis-W-0GGzexL|fLmT>#Vz8>^oJ{~NiNuSqX&cs~$VXXpQy01^urg`F8GdpVcgV6QO zf6c!i7=G_6n|VInX2*PSH1TYuIoGdHljmZ%n-|KYMwWoW zeM#cqG8oGrMbfer$DlCX8}-zz&_VYOm@Sv01DWkqeQLcR6XZD$xWaMJ7?jv zm9w$%(MG(=lfhv__TUUeWH{T3#T_M7GjOr!4!#y`yx ztglUB>$vq&-t-k1wM-136^x_Zw~jN#*&1M*@RK%$1<{OjCoJo@&5kHjr1SZDn5O25 zc4c*RY5!+TuETuKGqm-+63IR}op{y=!7{Ck9^AB*xg2glrwh~wub_(A z`e1|L4!@T9RAvpjj3KN2DHuNZe1N*=TVddwudwB8Cb$=GfKg{h!I-##L^I*KV9U%y zFmPT5^Xvs>jKnfwj*A#cQoRhS+-3&xx=Q#yMTv~8D`t2qsSxo#04iP%AOmM6gZf)L za-EeVAqU!Foa+H*wO1>5rg1RY#oZHWH#Zh)$G+uHtGvM8+p9;a%iLkms>9UNmy4I% z_aQ7^It2=oQW+Q5+fY3|k)=OPxxJDx@GCwAqBg82eJx4cs0!HHKNmu0-UW<02p@JlWd`nC0vhQCBq&je z1Vog<82iB>C3%$%=rD(?vlcTBIfJRytiia2TP!@^kildvOlHo%>V^V-BkUb>7Q_Rc zU{wAkCSYej7<%Xlj2V;<%`XRlAEQmq_?eK!f$0!DMVx#xZ-DIW@@#B`H#y#DPWBti zk;-e_a&)sK5j>ttXMML6#3f#3rY*0LjDbboOoQRc-q*2$5><37RV$4h`aEk|4e z@4)M+QGz9RAG7=RA7Gz-{*?FYj52BGNt5jGE3jO79623fPNJunllZYUAn9Vqm#8d( z634NmbXYkxTh<3gs%pUn-(z5KZYJ0>Ct*Y98Mu_=M0W2EqC;HHQq8@$A!*hMP_{Os z2VU>TGH+LCQ~6FC1Fm4i);hXC%nC;ze9D}E-;c=GsgPeSH^HF04`?~Xuwo0g(Nt*) z=eV`jP5W=NKb+5l?T=y9@uVS|walcS4Ju%eu?~EbxodM)t1rs!4y4<-PklmSn zKJUg~N8#DIKz7zICBfS-_dq!FDSd@+X>RZ?X0}cq9a#TRV6#Pu>P2yTBkB{FTUJQ}XL1l3&s#s9I?lDg)2EMNqW>gZw{k!38Ze%w>2d3$FOI^V+CJnnd6lP= zu7!TN65J|-GRYnIm2th>L0#4O5P8p)e&J&A?iu}(`Rbp>G}X#u#M&dQP408L^7T0E zy1tYN_sfJaQkk$@ERvC5r_SxS^+CEWlO47)onG%6g6{_FkfN9pC>n8^xj3c{4rLD_ z15cb{%ZGQsx@;HV6EjeB>4vwb#*zyb<3VZtQJQ$)k-jHK=+SGDf(oZcG~-bcot?Xo zIvYFDpZ98+&u9YApY&zMZB%0BDaTQ%e8Ag%oUnBB7^rU8Adn8`&aCHOgi*Q2Y19C3 zI;?&JEF5(n<|+Nf-Z}Y%{nGgYR)uqStS(e0BO{t&`1TJ}>V7G+JX8ZcoWrs3Wje@~ zj)Eu8_%No;m4C?X5d=o;r!6;hi1Psz67DBYntWBriI}aV@vsz`JnS-b9(fFJTJ7NC z$yZF>t_uE|h;qo9bCD*i1%pp}K1=RAv@Muk`M?<59YXhyryN?r<$`*jlzH_oJs)84Y%3V>L7 znUNuVkAq_6OL%7bSa9}z9p8-GPo(bk4aV#YC#F-4NH(tv*rOe=CPbQCWv4*xno3q7 z+YBskgu^1L2=arg+L8vF=k9=(Di9@k+}^#R`2$|!iZaV*t( zF_`RcKFzdVTM6P(io|~MBJ$yiBXQ}|1P<;ga4T#ZsP^ekbj0(Z%zZUzTrmOvv{}$@ zAWpoKi{O=dA5edAhj|#ki41ntB*V3(h+M@OqQ=GDoE#YgPCITf?rNu)>?^Zj#SH^; zd2<_R4jy6mL zuPj;WZ#4=Yl}v?%xfh^lumV|Kst0yeFW_pHDWM??$ZmT(qFS|tTC6f+@6Sx3W4*sq zALq4rWb{Eym$ziAvShI>^F8f1eTKl*>j2fZPi0h&Euj-HrGWbTAvC7{S#D2b6jkKr z*;Fr`&NAwU*IziGx{5x!r}af~*RO1B^M3kQ(tc_fp8)xPHQEMM_`|L$3G%~n1PT?T z=-sjxPAcW*)jvZLF5llaZ(J-PGe4D4RsLdTgS*=LgXl#s+9$$-wpN z&NPslFWgo4rObmq=&hMRvmPCzMN+o~_WQ@t&Y@=bzNn7V!#ZVnxV1LPzyt+0Y!H{AyuUKc{DvnEEor}R+QT%O0} zh4iT46_x)GfwJEe@IX`%wNo5`SI3URma3^#X{#y{HjA}ieTJUdmcrLsH6H}~Uod9^ zKEs&4EWIu{2t)Y^w9zOSohLk{+7pd&hExiqt+F7txs`mrlma}Nu@Eg?2GRJ=YmnEj zjctXB^y<0@)R!KJhjRdZ{5z@NrZ|3cUQ-Z`}@&2JsOP+w9)0@A-d!8PrB2}5|cs~V9}SY z$i_6#Wp8Y0to$VWF)@p=E;~#y+yTW-*3lhHRl>~W1H^5-^?A^o|v=V!KHmV$7v zPc%6g{ey%ay-kX0%gLf=FNtknJrNop@%dCt9$zXVI}XYa!*R38gZ0zM>M6H~>IDa~ zaqwrd{&oetF3E!CqAxI>_XCzz29pg6fn@VZ?mmhpWs;e3mV|mb63;~!h+f2aB2~Eq zlJ5i(f|Q-hPLC=`JCMtNL^2=_V3goec1|S`H_s4kwWpok?t( zkThupk|E!M$$kDcke-xIjNvdTQxcN;P0l1UxCRRDJcH9;QXsKgfyfp;fi&)%;g&I1 z$>g0uWRYqq5&wLPd_IyzJZldUJDm=qyS|t(veU_%3v#3>TSyf42QsPSjN$sDY4B-R zH28k?Ckei12ot1BoOfM^HgXxlhJGetg^Hv48#a)eW($aT{|F+pG@rbkb(#ce=yNgZ z^~t$&)no^RkSnJviSs#MGUCELc(*@}Y&a@LMyDE)iq&r5J;a&VPkl#TQhQP&`2fZj zSrFNdlO)bVfu!nM61}d6H-Pe!A<@F_UlfES{Ha;SwB!7_b z>k|lnw>ilyJ4W_>7)AE!?;szSMG<;)2s!0)nAA*PPgVqElC62Uq_d@x*w3pY{DB2z zT~!S6-Waj>zq|D+HNoaK=>UpOK zTOu|J10R#x(=wyju&=~@?<=CIIgfNj+mnw%d$RcCcSecx9mT#+1?3|z$wqI5Q4_wl z6YY`_U_Rs%la{s~eCL%A%i>OAFaCq9u`?pkLv)Gnunk07DUEF0_JDY7FDCDJg`{6* z9;xmZLN4BjAYZm;5i1)L;$M0IZ(crvgKVtO@vt zDo`AWY;Y-c7q`GQ=R2v)6BGK#K9Aa0grn)i6ZqDD9nLs-3H|z{Vp7ODTvX0rUh`xc z;W!OLUv$x^w{j@EZxoVTAH4fK0`;n9;}6mPu|hpTL>DN z*yDKTW!MA<@bMi_bkI^zVoiIGg))1|m3u7>jgjK)au zd<^MRgd-bW@W*{wRH-+|?VQj2vbF&(i;>2JvCHuJv$KeP*?4k878>|&#QMJasQb8? zPDz`FWrb&P$I-c1mZF7Yj<8@aaTsLhZ9)q_4V+P@L1*=KrPd?QfZqU19O9*k=U#lE z-1;1Dn&*R^!UW8TT7$F3DxeGdlP0^I#oTw=m}u=tyErpyAFPN*D`hY&_7YvHtb$d$ zBWZ#Ee%v8&p?pW*vu7uU83+^l?$_R=Mma{Ssa%&9HrgV8nvFx z#Ol$}*p5wotui?% zSW{L?2OJ5(#w#l^=lOb^(|!QWoKK=T<@R09?T-$D{#X?sjbFP@E+51n%VMn17-XGrw zSNQCpr_ODmEA9JXevvjF&6LLYvavWX=!glr+{eAxap?aFQ zm_2vko_j=}zgH4-`rq^S+m>zGylTVx36A#vKWFa${gl3S-`@%<5>|LhZ~}fmZqDMr zfBkO^fNH<5Q_QA_iT!?*fB)fcKmQ&Pz}#oe+CRVja9FqB|HKy1`P(;lbG(25_J73_ zu(q(Uvfy*!1T6jp70c+XPqj~{_YQKqE&^m}Vbe>}!V+RGvkV%d;TajD_IeWtSl{Td4CrFjkmh8|5I!K4su)2M-<4G4?T+iSwV5$?|2)( zpTC336M6+%^8u`EEPE&t4f0Ct6=Y-0SXo(H_W*734;j|GK|U4%vbD12LaX$sw4xQc zS;+O4M>c$G8=fVgb*^he%+1he(g{*isf0mkPkcVZjifnByEv&iP_9sYm z(yIOeWZ5&bMJw{M5~ax23aolmTG5~?Q6OtuusIXua@2w{RV+$kEH#BB2ge)u9Ea@Mnr?W zK8R4nfoyw@YDKq4k41oNt$5rrS&wE!bRK!M_O6JFB*rz_JwT%CT7?J@*FJKu{~)tH zKK8DN&-JsmJq=Q{qQ@dY3|Lt*TyX6_=2riZN1u9CWNTqz3t-zLX`(?DB0)ThUPk_+ zLGEoL6j}1PZ0gbYi#Es#kstZo|Uyl&t7EDiIToGSZv7Q{>1p}a+?2+K8?yarkjfrgvwELq zJr)h(S@lfXKPY;nC_<4X)5|Qo{V4u@ z{&7I>#Sa$04e3ui0=*}XMEH+bjb1^V%MY0U#E!~4MS*OA*V8xh`eTK+2s`TP9mJJ= zZv(CKx8;g@&+PJG&0dqnfq2~O&pZ-cMci_=dItfow<*NGhq(Qt7x@QJFC(^p_tpO$ ze`({sC;R)I*Jc0u|8L)bo3pgc@AqbKuiu~l*JuA93j9sS From fc9ef3802ec8ac439c2d50716297088740e87fd6 Mon Sep 17 00:00:00 2001 From: Alican Gok Date: Thu, 5 Oct 2023 23:37:45 +0300 Subject: [PATCH 2/2] Keeping Codespell happy --- izer/assets.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/izer/assets.py b/izer/assets.py index 47328c67..86db3723 100644 --- a/izer/assets.py +++ b/izer/assets.py @@ -278,7 +278,7 @@ class MakefileMapping(MutableMapping): This class is a modified dictionary that maps template strings (keys) to values while enforcing template pattern matching. It also handles special cases where a value is not a 1:1 match to what should be written to - the template Makefile. For example, pre-pending source files with + the template Makefile. For example, prepending source files with 'SRCS +=', etc. is done "on the fly" through this mapping object. The key:value rules are as follows:

Sd?FII=E8)& zN$9Y#HuhC|Axe?FKx6XLsQPdfEDO$(x+g|crEibvqMT4V@ie8Q^!!9sg6`~it|be& zvY$>rr$$d*c3@NT9N2b6jmII~Mc!r^bJQ)b)J)Kj)^yRKdX#W3}1_`)b6C!qGURK&SmL`$4WAl$MK@8 zhYrwBkzZ)TgDun|M{)mcD5D+&q_T&iC}#IOo$4KHCM`*asm020rgdz9h#@n)Kzl4b z^zAG-icOQzEe_n=pq=jE~<)bhq8~ix83YiR5?4jW{_^!S+b}4xxeJDRc3q9=U zi*p}o^w1MD!Lgnur&}_+rYE#4xs+}`_LPdA>9U6S4w|UGpKdd7qkYZ3(M$dd(M97t zQDZ_kszpzWdj8lUO1#n!BlpE9s$UT8T8HpPH^u&3H5N@Oszw4^9_<`+QnaUSAD$69 z3I}IglbOML^u=YKuJI61*Ckk1IyMD4ycKafVr`JFqotkKoXRXE;hb5U(1{nD+8cHg+Ip!uN;p&D9-v z;)mX-VqFre3_Hd+c@O$3dpXV#2yy4B*L2THceZ`jE?IZ+C@hWaKX- zY=plLa@e{R7c2BVGffFM>Th9QZ+%#wGvTZaZbN$&wYJsiO?dByI&`zLk9=xg8Ou6$ zkv7)f!B$CX;?#w;=!*VsHY@W18#6qajd|gN)rE>0b>H3x{l?oPrweyZ|UGwtL74;Hd4)t~ss3th2I{xbZ`QNnZ_A2D!l0Fw>#$Kuf! z@v~#rqK)!KwrbcArnEN&pD8_!rGf)=_T?}pJa5Mg3zd-C8H)1{DtHk~H?xjCmzmZb zBWApB8J4x?VEqBcSpCXry6JNWTd*vU^}5|m$>6J1ibyqa!Z)C1=7)Yx%ytJ>}!F?uzQVzAAW7MvG0#-lL~?zq8fZedU+* zZnN0e-gw`*KH`SB?&9|3*P>qfO>B#z_B#0WDXOM>9p5TYynD11oNmJovkevR6nrz2 z=^7g!yn9GHYW34Xrmj(F^zURk=%zQCyO@e}JWkL<>IzPm9jUZmW&z4_6w9uHPxR>Z z6UZf^JH5El9|tIMoi`|Wr#F2op`2N{Xs`1lp~vTgC^z~L3cPiS-rG<@8_SbXzbx=7)|p6O3}o& z44O(6ePhqBBRP&|(ZKCHh4X)mqbfb7Ax~Xb(YGco1y3(V#{zR{cVi2*_Dh+pT4s(z zBO2-VkZ{q?99?8pGeM*waF#h5AEl^P2eqDihkh5ui3WG{WO)sZD5I@EYwh}s+K*YW z@k-`M)ufO*p0`Cio9t!hA2`zKezs_h<|BHmTG3k}vLAkLzLqvl50@nhmZ0VQW_oFT zPg!rtQoMgsBMovi!%;ehvdjBM;mSVBGJ*PeS`>3rbVVGAzsJ|gMqNLOrdCFv`)3?b z@>+^YpfVnKzFanU#v(Lk<5@KRqXF7t=|M;C(i3Th9+BBE9EPKmGN@ZpE85m;1oGT& zhe+B5+2Hdo=(@>Oy4)?9&a*LOPnUPmyPmgX+P|JC@;N(cu}DRM1wYZT##HvZ`5^63 z(GsOKb;&M%o6kZ&XwkSl8NH!fNY^JgF`H@1^r4c5q9@X1v^ygZ`_2;KwT16da_Tj- zZO|;b(BvR{cjG9V9lVEyE8@Y|&d+2GeVeG-%6K%TSb|?aKZwV?=_Ss*eGHGjsNmi& zia|ij0Jq+%V_D~3FzE|j`Gp5hnEEeOdAofKGjTX5+vK|fYe_ES+r4{>VWf-L|6x16 zwDK_K?alC+gbD0Ia|-L#?IAm^-AkSvt0Xr*-^zY+iEQ4cRO(y&Ml{3dAYNk8f!}TK zF7B;%6_@tY73V~!;Y}+}qVZEEu<3o1nElGPUhQy;sw)s z2Bt4Zt+Ol9)4n4mMURTB)ucA@N<^t-r4l?fv_Z4U6rOc{d65AI% zg)I}!qEim%k}LF@F!VbuJQ(aLb`l(SbD6 zI!@LZoKE{yPG&jsUEG9NmwuQ$Co zY#Y7)C76}j__E~Wb0~bZg7a_H6jnB@8O5Ep#XDWggsNk1iB@?Iz-I>~QSWU9vf)j& z*g{dI9i~sDXU2Q5K{qEbtIJ930eDaK8-rNTHZQibHJF~M-ALaq>Og0cHSnv`naK8X zHqJ;Mhg)VSa*OLT@zcs}$nJhP+IP^Aep_isg$D!CCxbmSE~AQ8Uk#x}16`Q+NOxvu z86{)(*J$4bA7vp%huPh;vCPv%!D}0lOxG#;QFPaE!$&*;?%S~*7Y5D8x4sNg+&|93 zGp{D0RbQ;|$4E__qNuY(m(x&5kO?|_dYdYqB3pn8mU3hyog6Lz&*uNR}QrhK=3i&xWiy zBHa;WO`FQJ&~4M5SU)iqzY4R*&MiA}yW3#oIBglO3)+R&%6##dx{-M2)=4OuIOCNK zeUXou727gJh3cs-V!TT@(>%C}4N&_|6OJl+oiAC;_MV+2{rK@b{WPmPd$714F0@Ib zPU&YvkJL6`QR^1G&&&horVYjZemYof^)=bA8(E@BSJ#Q!r@fLYSzBWL)dy(e))d)^ z1y<~ki#xmNHj{na*PHeU-ocC$%4CB!ZN>QuyWuM{O3@pI53Q}}30`NBPY2EJK%J+I zahrucPPDm0V;|TspS!o|$zi(8RIJFII9^D{*e+uc$qDSaYImmI`!PD(!-AIo8i_~S zBK)Bc;|;eCNPqN+!#f)N(W31wxN_V@R5P*#z3r#qj96nLTd-43uO4irSMrV7jlNYh z%j*z}{AI*`rpK{X$wL~nrkKXM9TM3PM^S?0GJ3B&lg|D88g)yVjlE1ukWbMh?5n7I z##hDo)97eyD7lVohAu{9YB|(N%jx*I9rR{TOE&(FJ`1`WA^CpSnBKe4N&^OkQ2UY2 zOy6q~d*f`%hH3<`#+5cSa-9de8`X#o`un0&wWF|Tnh$pW=8NM$#Np0^+pw@mhRZij z!gD*^@z91OYtY?v-O4Rbi z1sPmWgbqEo$Gy64IDpI zJYern|6xhGR(=%$<5u}&E0X7C#N*^Z+`wm(4TQMT)JBR=lh@M z|1=KQ|9u<+XBQXuQJ#)}c^nE+9)DyfAIO=an7GoP=l>*|`!8g3j!p{Mu76qfpCwcN zBfQ^JwZqB^!GE6rZ{dbY@c%8`)zOu65jeXE6w=-Qmh`Dp|N8nHD!Kfx!*Ek9fU67V zI?8j{UmeCjo4=c*Gw0^WaesgQ|5Uz00gld!iT~a8S5i{MkrgJecyZpb|9|^4R1*Ae zYcFtfaRh%%z~!&5{U7`L_vl}sut7H`=f9!-kI^KY>~-~_-Q82TIf|Ht?PoZX#W++F|f{{J_~7XTpm+wWh6$M+Zf&;J7d zfs3=No15d`$NzsU-`!Pl4EVe9|DpXWye}uGKaS_$wf}#ve+BltIR9}0{`=lPz!kVT z0e8h7_{;eFZ*<1OIQ&Uom~fx+`A1f7ky0 z(fTWV4}p{7{P_>gKmQK5r}+2($aeZ;f`1Et{^yxL`p@6Nxx4wh zI)IlBkn*p-h)DLe-~5T3I0@SC2~Fn(fMZG4&?vi@_waE5^JBvxcJYyT+)WXz7@P|o zI-ulTr9J=lcr-aZ%}Ve)nQ>dzT7kmtPGDM8Ccn(OuH2S|mY3NK23|w-$bFw{k}G?+ zgXKD<6*hW9`PJ9A^ZObPfK0c&0-rZ`dEWdUFR;DLh1OWZ@Ce)B;P+qxP&elq~Z-RlNMeae7tj{4x;Jzb)wstmv7og_;#21+_2OJMzse12J$ z1<^ZV&Q&N)2G%onfx<5)yn3G^e!=)pWRJ=klGnqOTkX+ZI>KZC9Mh5wlxA5;O0N9k zwnBY=PVF(?VB_B+C78MIM+lj+s6~wwT&M?U^4MJoC{YeA0*paRHV7`{a{a@ z^T6uh7yem6x?8E89~gG>8DDj~6HKbV4WAYI3gi7O!M#WJqw;=X;UauWU)=iE-r7p?o2t z#zM1@Zg52S8vb*PGiaG|gYPXWASJQB!j=1;gPvqB7-#HFgvm~lnU5vBB>y`UF#p9Sk5jF34BiDVZ~N9k=OJ9=~tSHh$@VO#W8NelAki+1=@jji7$QJAs|nKp@Q@ z$Zu+0EZL~3C8@t+Abf9SBe5}6A=lMZ$yk@3FktKe5Efv|&&t|ZakiyP(q;A#M7?+_ z8PzgNaNfHXbR1|0!{2O&w$BZOGeS~9=k}A5`r$SF(~evGrotI;;>#3%vd~KTqb3O)Po0 zzMtI#3e)%%+%Djs{+l1-yMb4}=))IO^dSb;GkAkx*1YC>P2smvZCI%NgAZ)emR_%o zmo6W_M_Qotf|$IFhEsFb3ie-*ERXEdPr7~C4ryJ~UGCR}E!??qDgmFY;i$Y#dG9lxuxNeKEBq*U{5`Vb=agv`OA8Go`*ol29^%8?<|A@)ET9_5C(B7w z&23^CcZw8R-v=5ClwptREO=+l@d`=$a>)zhB$(M_8fYGQntUHJlAQiAwEXbeSW>qm z7(QEZ0%$I@BM)Eh;5NHVk)9azieI)y1-@-@lD_FNL+Z9}inQ)Zxny!&Ip{T~h+DAk zM#baMop6Vi9#P=Vzp&>I{wR;OC0U|7ye+>=vaqrp4A`7ZEc3(5hgU6vn_G`c9!P!3 zqT24{{!d6OnqTp%-Svb{X+lu6%8V?2q$G7|l!Im`RbhDLJnsDERp3|h2Tq)=MXq}o z6F-(K(8wP_imsHAC-tVprzM7@U(e<9n>R}iPj-bD&-4N7$M}%z7ZTxglOw#>v;gwW zaU1WNb5(G;b^&p1pUrR5tmb=f?^*G4qP#rH;jQGFc0)=Ux`}F%^&q+naEJ#J=LQi%PjKQQ?wXwsHyW$KbSY-9V@|q}^w55b{6c=_DGTuMi-1d$eS1 z86!RmKg(qhyzcys%r1OJ?0siL zm&jmZI$#fJJk$oSUpItFLt6N!u^MF1`{V9jcD;EUy*i-ttS6k>J()B+Ch);lOF<2b zk?b2DL)voHfm!D-aC~P62)myQr`xXQ7w;Mbe?E@lHVoMg+hx%(<6F8Vhp&KN#-_we z_mW`BlmJQMfg1iMKM~B|-z6DvXurg~-w-l6#EldL4~0A2he@+~wvv<$AuzTvK_FRv zkZ(FIm*~~ZBedocDQ{08y~0xY+%@sSwrPIC>S+^sw_*gH8;ZdiJxj@psRK!LGa&xt zC7AzXD0x&bBjXR$llt)0@O(ru_@DUmFAP#!aDzYpGDGsz>Ipx0g$1Y{dyg<<9ckdP z5z@0q21()QUBttQ^8E(w;VXiF0Ef96#8ZU1X%n@f>Ci&b(6foJ@ccoJsC*Rs^qt9< z))(KFM*fhWm|0}sihlsDy%R=DtCcFP58UY;Pe^#T04U2`RY(}RKLwp@PV%6NX( z3KL$m`3!I@{UK35nF+@6hQOe8FZbq>g3-3+I&U7bkW{{vz@DhB;;`3Fu%;nf0ygAG zpx-%(&0{NKeWri}2kaxWS|*YUUBkGvrJKo}7J0>y=v*-R!a9jIw~RZrQ~)+V77G&| z_901yVSMnZN-}R`iDXpm55E4&Juad60vQ{0gR~^R=i_#@@)w0OiT9m4!OYGa-hb3| z_-$JXA3r9UTf3xz_edTE!cnA$&t5QP&L!}(JdKFX z#*!ta`-#JVWn|^-1hVH-u_SGPm89nIE#5LkjUO|(O=5C)HqWfA$nym`60wsfh*wF2 zc6YauBGVy~=()v2d68H$zxffLYxV>9JURtRZ>vJXu4-`ETN_SqRg+dkxbZ`$rjjF$ zNkDd5LgIsDz;;;+MDUzCJstfnLpy2V;bbh?ve*Qu8Mc&V03mn(9loU)p&Od*8 z667s@!M9(_Abo%hzvA^O^4!0a^tDPRv(+xb@)dLVn3+4t^G&WqdtE#++8agkdRFnX z21k-*pLUXuIs+t|#1e@`*<#*tZ-KO~N0xMPzY)@l6kDYu@dR3bzBP^##YdQ*v%y%m(KM_WRQ>T*P zDbdnL4T;i{1rGee2U&cUvNdct@DuQ@v9NSmKjDA!f1vGz!ab{d10@wL>E$L9=$&T;V>T`4k9@EpCl06a zx2#r4H(g5yrZO)WP+|e~XU6j><88>T7go^jQ977uQeH9Fz>4G_y9$hgEx8RI!w6UA zPF}7#!G-2PGBm&p>ibRRr|AZhT{T)#_u}Wg`tBca*4ZiWS@1^5!#fvww=@S~hxL1~ zHN}{WJlYpLn4CrYH|CJ?&OorpFbW!Xropu8r+`NO=6v3pL7O;h;SiS+*ztZOm{%W9 z*1Giv_7-Evi#3;En9&9>!&igg6ESevyfCP~ZvtGp$pEB$RhKk9Zh=$6JjlEoCeoW# zW2CQc_JMbL%oMI&mI~Pu9#)&%keFZtxWKHeqAh$Xq*muZ=CLDO-j5TYZsAb=`J5sW zn!BGg7xb6qr5!=r$&X)l(q3T>;NTK9qGYrdD!okPr;3e;Vf%MJxKS6r`P8!_5)9`Zde%UT z5hGyg0#z9GG`E5~R}4)ie}|x;ne3kP9vn>%g9-A!WKqyiX>u>I)M@DzQatQF(d=`Q z+-@a&=&%Ix;nYEX>Hg#7h^L11sH0eV?t=jtYwii78fu`X$xBIQHv{k}c_3*(*U9Xx z6tY+Hft>F(gs5JrsK_hygyGxX@cLGxh->K39Cgn@}=J-GcYE%7l-TlqKE=Hj*ok$MK6w?g`qzF6T!oe3f^%j)9^tbNR!g zV&T=KBoe!C8{Zt#4mBL2N#~yRa5FO}Q&bt*pLSGoePbSX{A7Pnc>5>!yhL&KE72z3 zhSu?cvK8c6*BsK_?J)n-WjefVa*BWTK$plvYWan0U+`ap62ZEX9q^5(h#1s8g!=wx z;m~BjqXJAe8e9P-ef6ZFMvhY3Ip)%1D^G$V1#a|PWhR_oQpZ1(|Ac0z!ib}{h?I&i z5x(3}8m4W-7X?iOhNr5)M*TG;%{_<9)ZR{%PH9Q!h>uHx?!6@2XWir*$JB98cXdhn z4%LKfJ-Wl?zCDGvd)HTNTc6Eqlw^=-lPE&_Cy@mWeWi2FoS{?VHrOzSLWg^Ogd1Jv zfbAza{;kmg5?i1uHGg%I9BS@PK6!tG-O4#Q^5rCcj=d_sUHut(secF%=S+~c!j^NM z{g&)#9V7KFE|o4ewUV6ZwpB98P9(Xit3z&wIFf=B;s0^|`j`JR{o;A>+2}l2@4I15C<2oAvFlzQ;G&1OY69W=4*j-?KT*aITpUFu&?laxj^C?CI;;- z2MC+4C=~W>;o45FfwGeh;OG|tIXm$bXoA}~n^!xzesewHdhdl~V){|o@GzObPP%~u zGj?%X4i1MzatDsWhA`SZQn>O?C^Su}BG*p%@uMf@xo5B03d)aqKquY+x*DGV&z*-0 z$t@?iGbai5qbr4HtuK&i5miJ{U1-jdGOhouQ0mxF6UWQ3YU8L3pZX? z5qc)3!(#pS;8XW}U{Sjpb`Ir%_@oJ^qY@ygw7n)c5Vsdhm~rpW(k2>cG+EhS2SvGpDAl3h!@~2p)42Igg07lDAjoK*O+>tN&R9F6xwkF>%LX z*@YOg?Dcvu#dR^LyE29kICn}SsT#&Ll`MeGHI1+?%m+@Lx0`p)x>%v?62ilsA;PSX zED|-SKXkOq5zi=T_M-b218n9n=b@`6OCjPspD-67oKr#nBuDJT!4`{E>;~oYT z6DOy|M7i}bjMIoCXo53{kEj4N(_T2XOb!FSnM0$wia27KF)->Z13$dygQHuw!rLcz z@Fm`X#A2j|@YmT*LI>x;(vFBKeoeyza+gWS#VfDQ?su99o3fN7gJxQCT{Ys0!!j9| zmpomf52|4F2?JvFBOH$1F-$mrR38baH3#e(5CSV6ua^cIRB;a{-GlRoz5xb^6Hc?8 zBeZFFS+T0cjLSN)oHq%XD;e~{Mc~mpw_?(;M_jSlB9iYT1t$iW!+1>_c%+XHwBM5> zJofShyt}SmGO%GcEccEee5wJNrg%5h4PDGjn|6X}spi4}1(v?<|AZ9I+QM6J=mOu= zcSFu*qr!i&0bZdpxU{SkjF@x)t_FK zuNB_Hr4!7B3pF-_D%f)yDsDy^6oyn0TeUJsY7 zh`9tx$43KhO*LmF>CY$I50>=Ve_JxbR1N~0Jith6H*l#z3-0PrCHqZuNXP{{-pp=x z#roXC{AL3e{#!^PaGs$K{RcY2wvgFipYd49a?5ov5qy;#Y4(El`wxJ@TT4NvN*1@! zsDy;r#d8Tivn4T8iX{1u-tt^k!Sfhb!xmEWkaU-b`poQs1ZwOZ+ZK!h=aeBS7#_5+=DyAt$}ir642fGxa6tVCBAi?NYXb}O={oJ#7{IE zEUC&{FG%*P2c2e)Jlt2o5AeCqcXy}(dPT~d?W%XY9-axEH^jjW=pCGtsxBO$f2Lfj zXn($G@fyj=H7B9U;&0%6inAnU!9i$ld<`!7eHuE{Fz6g|3NAk{fctly0atP@;3T@3 z6xHm3x7}Ugo>60A}Q z1K;!Jg&k z!SBQh*mB_wDEFET?bZgs_%lVkGECt0rbLq5CAQMwx!(M@({n+Q@HKXAUQBM{`I- z;4eNxc@4haES)rUg48l5Ryy*`5VBRJK%#$R6ZkNs z5gb-9kMoxo!k;Y$kj>r=etH@4C7N0A>9Bq9X`CPN`#cN|8Mu?|%YDl4K4nF2og6}X zd^QD|`>*jEO5VdIlQwW8d{g+Q(cXM-(@Xpt(@JnqHw0QKkN?m36RmS+GK8;A!o%yl zgedYZG}H0tnb`xrrs+0^zWIYECL8&z_8t7nhwc2>h&HadBN)cGOeV<(V<6oa1V_Km zk?bxV3CmaZCOK}c5N-9mEWPKhFBg_qj{a@UIJ+*X%?LGx? zp^2?9;;1itwM`xNUQ{B{cCzOqQhE`cR|26VLP?_sV!cRYVOI}PZImEr4G!}#4V zYQWWob+B+|Z|E1k1zJzl6fP>ZfENwo_+rxs+{lYoaEiJYh%r8Uwyb)apzK&Z*O{>l z)@h_c5`P>hAAHN*)aw9i$Mq1Vqkan1IeU^GLzIeG$ksXKzo>tT{fQ`E{+ZfcUo>38^k1q~86 z!};9hcb#0$nxpW-hd|i;wgx8A6xerCEZF+14d_it2IoN^((zG^Kj8D8Gq`#MES{1; zMs@Qb2|C66p5>P$rhCVM-OWvkwlj)`XN~iqb>vSF5NiX!Z|M&WFK+`&8rtC0*~g)m z#FG?1iX>wnFM$Sod%@W_lrMbtoi80=K#p`>fN%8b;U}dPz+uH)uG8wWB+~ks1n!K3 zAHV*T{4&-7%d(|Vxz}Z=X82vwmC+COoxK*QN;-J5KwaUUmV%i$jvqVwJ^%dNZcw~) zG1#$T38*PNSmAyplJA_BPHbPL@=Dj#$j#fCP$ba+$&IUowX1@lmxrZLskwkOIE{k7 zv))0Ujvc&srM~3FBv*1ttCV*uz~I)YYH;Sp13q21H|)RcJJ9l21wVE<^Ij@(QJJ02nvOjh$(6#L8iegv;N$Xn7Y%>cC4 z^d=2(H{aHK8=130pFExUfQ!$zgUNj`bm%nYGqS9bEiIPWgM@Wyv&zpNN``WL`W zpS0m2d;=oCw{ZBN#bjpQRq}kMF6pJALEa`_0_{U~z(Z$K;p1tC{xkk4Enw;3Uf3Q< z#)R?2&uSl3|M&zx-!Xz1J-h`~+~;%SWTtSn$eeet^8=RKH8_j0*^=J7H}QMNVE&Ey z5^_I*<2&Cqf~ighd_d4E!9$aUeA+=@5I0-}+MLz`q6{6f`rusNIl^4hTXzNcA5^{f zUr+!4Ki=M@h?bU=7DeiDzu!(Yjg%xK2^rahLbkS2qN0gprmQp`=YBgS6;vLnH-!xT4J_86n!3FG&UaPB@jVeE*zOI^bcr5E$0rq1w=ZJa;d7X6 z*Eq>1M5eKbYrFZ*t#4RaTs4y$T0}|`u8TYhlSM1%dC+sC-Ox161}@Ii2N|WW{GN?o z?27&=dgjGWTz%IXt~VdX{kg?#_BR8XInf>MdsKu|tJ`p8WhGzYkW0(kCeiODMf8rR z3_q946t`_M1Bb&#@Dof}-D+ce^5_&B(tL@uS8kpM?ZQFV;Y^WodzwFX5;mN4v_q3Iw^cT3Jil1nZjdvbh32pMi3$)_OA6J$fl!;r|@IRRVOcpH8dl6R2gtT^2sH05)zo4ZdpzarYZfVaJ>- zs9!ISrxqs@wGY)yHjIK!pep}!Y6!Y+SHh`hrr;{$aQfb;k)2W~b?VDWU=@}>`8Wp^ zT-kpsON@*2WK$d!Tt{|X zold5;EN9|475Gq=3uIjdjjnpf$89#3@Y}NJ7k-i%{PJfmH3IqOGyns8)0onjW1vzs z2S3IuWAghZzNW*KgajADq)C@p;@UOLp>8tl-K$1d{9DXVp8N``^0n!_4^m+Fse#H* z{LL>?G=kxcE1bIL>C*#yPt&Hmci7aKsiNdtBS7s*0K{Ba!`;|0mec#~&h0&$0b5fy zg5|NXVBKF|q!)9FANBhVUskO~TUyLGegj{0U z-N>vimJ#D)+u-%97p%qi7^c794AYzL@v1p-M6GoURUF?;#?QYYDwruvrbj%7eD&S% zruZ}Pe;+dStE15PLJpo2-(_Waq3rTgRd!=b1hso+&EGj9xh%rvJp46op_ku>;9dT`W0^cb9u^>-cc zmG?e&^`Rc~d+o!*ugD7(I}Wn6yuE0yW{tC2pEBpM_xa^n0ld3^5S!F_iRC<*PBbnZ zWkw%2vH5ZbSVH1=K6AeoVt}gfwMsIJj}$T@9Zv$ACh|ok``8(Gk@%*`WB$a36n5hx zWhc~Mutzo7)V9M(aJ%V+mw)B3ajtFb;oL(w?tux?y;PWOp(`Z+aYFk&jMdwivl|{C zdGYL1*`VMN;B(mC_TI^1AIgS`>APlRoWEz1>iB?ZGTlC;A({)ebL(Z%5ul$y> zAe#=p>x4C~QnALRwRXt14P;iCJMho$7vgVeL1c1fGtY&WvpmVSTONFhc{yeB@&;md zLOz2So;=1{S8AiT;ZG)}Lxhnzw*0b19Z*s>tb&xP;ui0knSnlXAacX>3#;h#fQ6N*}>LDI5>9| z`loEd#Fb0fJ=0b)uj)Q)6rEzlDS@ngpf+0ly3WGq2jL2(%Vo0G@$BE{0F+yE0S8>Q zV0#KBoRL8=B;K$f!;-Vbqs8i|r8oeXn)9nx z@8gM1f4r(#B++j5X_G$1?!+xvUo=$I*uRj?YKmk(Zs}uGs0NCMCX33a9A!0Y4zc8p z2XNy+1>L;YgUwQQ#(}f*afQ1jej)QPFXI>f-ZBW^4pL!b50&yC{v2Q{C%A#%jVpAM zcQzFs#^4A0d|cVnh);LjVqLQBtR+?%zrK-y8~5^gtLdrqX4fh4`v^T=t0#!rnNQ&V z?xGm=OpS?tOZKJaNUU`+=jb|1a?C=Bzxp$u+>G2RgbbX9rNt?*^zt0GRd1-|%ne4T zaoEALisLg#60E zVae-pidi+zRqf(y17g`tQ4I^bq7Bu}1E@{11**IHv()(aEcbOFrc`L+!(baM5AqT+ z?b}%W*x`8Nb}0;BX9$&_9x>@^DU^M402X`L|EGWbkN=500Bb?%^g-}yQRBW=da&X} zS-9uXFbvbvrCIaO;GMh6=*aJ>tmmx-3%Qj_-f0}<8&|5soNpqQQ!U|}4=rQ6H#@PB zrg!-}N0q39N)fRdR3vjK;3z1rNxA>QQJbEVC@kh^Q5}nB=w9O?D|7^73 zjs8;j5&Dex{$N1zrH$e68#R$uV<_{+>y*$kX&Gl|EF%*3Z6%?=1W18p{dH<>_E%v)lo9qloNz|CTNy z4P;ipM*i%_-A>Q8MlqYCXK8DFA`!VRrom4gr~xNSFKvzzh28xwR@C|;n)f0V^ICJT z_hANEJ@N&}dw7!fO4Eo<)EycmTg!4z=)qCHulyM&RoGDfj0X61(uS?0*%*B_6rK&m z;HW~j-tsJ2Db~f1pHpaNYYNM-??_fcq2i!XX=R?Vp80@51kNTka!jt9qTqT{IEi-`5isu6QZgssGYAzqjyQ zxPj8of8c}VvQT>UxMSmoZWg6?f!eywCuSkr=;Lkvu;I~sYEm8r$@4yg%#dbL$AviB zu>TvMyx|Dx3QZRON$JOqCcT0UqEB?q#S+^0Ll(=2oy1}8Qv|(D>TF)*0rquU6>c8& zoQ>8VCb)Sk3&O5ivF4%Uyp@ul(CXTWj}4{?J9=`szY0gFZ&EEEZz0iOLN4RRgmY|7 zqy;AH{G#FZui3#q33rCW=y<2Iuq7gri$C(R{8Phq{$}zQy6xu+@z&<|WbLOlbi4Nf zlD|7gRK^{ljZ5x}ZbrQ1FQ&ALlNMVOgCHZPnwBA=2`0I0>n?kGz%d4S_wV$$brK&S zWeL9%Z-`1`-iiZ4Qdq*G(Y&!jE!D{$0^y6M!Q^5BD|&DWn?|eEY+YEoL zf640NXQAuTUvMwr6$}`)kxOlR0&AYOLfD&6VDmXww0~hK{oCnHvn4x+me~$|ho28y z<~SBF*2@STI}hQ)$m68qtQ{oFj)L=DTd1f-jny|Rv3ZN!C7R|-436+(y@xy4{j`l# zdgexS^_a-LI~T>RnVP}fHnIlm>*wjQTYJFO#~ysla&WZEAB|fAebh1}r zKmT=7OUpy-RHFji3|C>J1|CBB>>&O+?xQb~cEjrM2B!Ejp5^z7XzHUFnqnLR6URS+ z=fxTDa7Q9h%@4#+eiN|X!JiiD+hS=(9BOU&!&K@{(`e{{U_~>w&*CT@o)gQ`E6Z5! z+{2{3?F+9HbDg&>si!Lq(}?Kp4YH&9B+hg_&AgJz$mRq^&F!bp63Cmr376tV5iW;%x1UjAGe2nJquWvF zQaThC?0(E{=+49CJ6EHtk2Ksq@|D&rGh*6sf{m}41Xm~Bq>oz#`mFpF#EtF%i(U`D zqt+41RBQMvgY&7$m3`c%?636eC^cb@uLZ`*JHxfp7PL0?JPge{#KwyYd2ikXM(-X6 zePQ|R_LyLNU}wvZ zkd-;dB)-*&v}}qu%N(joA5dR5DS8K4XBfcZH4jorHx>J{tAYJ4UhMRIojEf-d=sop zYx&LYSMVzQ^+g6`Q`P`)Fikoxv*0ObcB@MlW~&HR0}{PuYf^o{N%t=3O@&G*t+ z>T;7U|Jlu?rOL?@Wk3GW?7N}~NB5BM>>AP$eh)4@zQm-zZV*`6)X4f`9bet_?il4qTov9)9q>f`@eMgo|%N3CxhAho4{`M z8W@ypMMXO&!}XeYwswj%=dD#u?G_kA;i%aVuJ;PpmnY%iB7L-+Kat+GO<_-^ti;Qg z4CV@k_2bO0Uv_jV{vbN1WLMtXZAn*M+0VRQH)75=bMY4=#Q|MUMu{x^TRPi*20L;ixExxSc8+CD?+>z{u61xR+z|bB6WP{SSBMVW!>L_T1r>DX7q4{UmTGC!F*B}{^4-$( z$VDwEXhaew(J{uRc#!$+Kj}xAvBX0A4BhL~F4i+%D0)ORAa;yoW>p@7)psUve>R>6 z-zT+n>2O0x>DB@F&N%*zK@!urs>8RICBe7&UEu5D4ygkpK<`!>tSQQ(vrk7bm?c7= zq8M7)--=V1YtFTI+rxq@k3nav4)^KfGiaaFMYHbjAr0>r(T&c>n6zpvNeHN9OGN|0 z@3ba7J}U7Q$?bsy-&5gs(lc_bSe^e8wubNg6bT>RD-ppgjiu}Efb_+_EGyzJ4af=R zSNdOIkFQm*PehMQN$pQJG)+RglTZ0h&2DNsEfSnpRMTlKo5Ahp47P~5P>rsU?9#{v z-oM_18=4pk{*#A;nTZF6o!te78U!xI1k;&M{*cGrXQ;hzAx&NUm3Nsj4y1Q8(Y8uw z;8`xX9~i|h@2Z7ytrx(j|5jM1=7k48M8mEa3eSJ8VJkj*VE+?(+~0w5_)t!XhI=cq z&sHySM?V+rJyb*=mie*b)qe2gO#(A7=!M&Fc_xgV=-=UUAZC|0T$VXPrJT}e zNU9l~d|Qt>n>}#4e_}7x43nc_X^Q0ET8`1_0obH-lzq#(z%Tu#$Q3;ipl^^Pyx5`0 zZ7perlyHLTV>a^A>RZKq>aJwjlat(CyJ4tPGZMpcK9EK&75db%mZr+Pu?D9%xV%(~ zoLlvpWtv2g(Yyx>QIl9Re!J7HA1&E(-!D-6cL9uZ2%-aiUFPXM2N=6MhAr*3!MleK z(K{Q>_`}Ml;bdSEIG@mfxsprG-s&S$n$`~@Zc9_02NJJOgmwAx%z^N7%rubn)ZnCP z?wBxjCb+C9r)PI8#Ls7vaZHa5|9Gn@q%ZwUNu?oNd0s&Ca&AyeOyFmayD#$lW(lLz zD1G&$TJlbc;Ej3(jZd8e>)l7wsIs$AbyI+lG0NP)0jk(-ehzLtw4zs|lHp7Ltu)1Y z9Lx1lr$L{S=(9vmnC@Z29f?~Ygj}=2<%hR{si_9%ZG1xf%%GPZ-CRNUOzAd?lN>l^&qae)NL+>>S{`W^;XBx8YuQE9>v$ z!r8k8ah_ZE;O>Ia9H)JYg)i%ckfJ}+i|`h zx88X&XSV7&{B*2^K?`1B!G}3yf&W2z-#Ua&ubfLt3tsa*C(km?-|evePy#(MvrKH4 z+5j=%rozyu@!TRACldcxmv$VyL(6nN!QO-F+`z+sz-eYW>05OV)H{>8qF6b)Vo(m) z%)S8!PGoW5HI!2|zfRZ7t8xi*4#JF4+O#K7g>2hm27k1w$z#-_w)=~~ce)Yh{P`I_ z`IQ|Es=Y-^Y>%`4!*;^*vwB>O;~;9Haf}ts>*emcA7CS0&N#ZCv_hog%O;YVwa=xwDce{*X!d9`9U zr{md3lbw#yMe%1L<&+XPRkZ^Mo5vZx+7B5qBVhZKVQiGu4Tw@VBNH`lLHO;9+=#!a zMB8~Rtg&o>QO&3%FMUVmQ5*GF*PmFhE#OPgFNTsC*Z!yuS-4 zTZM8{rucDkUmnpBC7!TZBNnJk6TP0H2Ddh-!iU7w)Zbo-dp%I>J8U}19P54X>*;=Ot7;46;l&5l>MnW{HdSFRE)wYXIuP$H>WDYFWGlt0{oWS6g7m58_NaYKiAfsR~ z(`?dZIdK!=V(DF2xhomo%`wG4DSPot?nTk8!_)HH%&YtaA_`7sAtOon66 z#4$Mc(rmF09mqu-S;Boa8pnArI1ZLaHN<_lH*hyRv^d4jySZBvr8tYCVptyY8O99? z=DaS*a7BTdobR-wkiJ(84YvN^2Rd9=y)EineZk}FoT$`cd02XJ3tW?Ad~|mzb6qt_ z;P`PpC>TjobCD8#JXetmA!#u7-v<2Wmqf1}KErkX+05DaPv!EPmALP>C>tKNnk#T} z=R90D3u`xw1;-p!zD{{Q9pycii}hH@jk^^6-~9Ox`$+xw9DG=x&r#XQLftGdLH#tE z&MFY@jFZPpVXv@lAwk!iedyXNS*d!~V#QM>p;vPmOK5f=Rf|{R#1C#@^tFq8Tw4Y; z0coi5wTgGrq{2$qCBi>nVVM4BD=N0jD7x?Ax3z)HIx1L98dGDrLeF! z9bKDsutVkv|8t%>-d(W=m9JO|&K*u@s-{IA9UhKK5mvaQJpg+&wo>_>oAF%7cbt8F zJ1(LW>+am=A3WTQj)3&F!ZEeLM;!u^ZKgw0QLd5=tW zX4LS2e;KTRZ+?#AhisN(VJAn5zFzj{1Fxit_+th5d~y{Vb^90{CH!U5KP8$J{;N?ZPEzJM33HjqxMn*?zSF!rF@GI5e@2 zH6(Mw78Uu5rBf~opA$|Cv}-T6Ui8MRJ7wVPbBSgn>nA#YWTTLmI9k~AuLz6Ie1jpn zX>@t}Lpm(Jg6|YnFqv>AL7dk?8y_tIi$!gs)6v&3{pv%!bbA#2XI}`-O@H~1cG>uE z%Mo^JOb1myY{15(RIx=W0ob1O9`Bi!GQEs5xbbrdJKA%NALkLwq(^4+MjHwGE?-HP zrpIEznn>JfxeZ(+gJ{n0dI(w-3!m&tvH9d#;rqBe{CVshRb5|+g?pCaW!F?3*<;B< zzRJR+(QzQ9eHW*$=ke~JIyUaAIeT#J6yY5z+1HFT8a3)6wHfh?uBxiWMUynp-|iwV zQ;dZg;T$vEMsU@hc-}@doIm&`m>vCMFHHPeCh_y^;k{N|hZFAa;IPaqjI7iU{+PH4 zV;-nL|7UH?TdIt8rM+R7t8|&@UNRrF?56m|gb$dsMZjft+0aLb3}9$`LYMx1e-Zw@Zw2Y>>D0#~f$pDV@o*o`91z7U{KsQ~=K`qT;e}hIQ8G@>9Fz z*`vMg%zws0W~iefBcDqJtPOuyJ+K@-%h-(%4PC3%agdr*3;ACB>HyQ+w%B4PZ;|#frX~(Lg2q^{8J4J zBF_>DwAk^5T9ar-uLARS9SkGyPUlz6D%Lt3{kW^Ct#u{Rl~)-8pK1op-Pp#(v#f4?&M+!vnt}=nU-# zUmNFO@HGv7(SbiTRLi%jo=9Q$-qsyGhUs6SPmw;J>KdJ|NSca^5EHA4DZCfeXGxn^Sn-fpIAVjw>vP^RqA}_ zeml0{{^N41Z&ftG@dm%nAdl`KH`vL|m)XDUBP^hADScehEy-nDz>2p6aN5xla?WoJ zYppE9&CCxv-&s(rl3J`67n1``W66F_8~&{=XF>KBROvX+EJj$NLAe_}E3LsTxMd2p zJ2mNz=wGz(Z#;E85eZZHb(nW#ySTCbBJaWxXn&DQk7uipqGRj1zQ|B6WB*ZjD%ytE zwM)s;;GblFLlUUG^nwNw1Qj!_ptiz;j$dm;xBXE-19y8W*7c^5Z=Q-GS5BrY4n?zy zxK94_AS)_%ol8PT%|zXO8=(HvNUEUP#a(=(;5_aJ<0dt(5%o$H!?sjK*lTi($_DPH zOTToA-urxVT)Fe=e?C}PcyOU`v2-)?VSkxR%W-x##Gicji>Ac}6L6`0DqWaW!{X2G zg8-Fr)Vn)}_vfM!A1Xk>ll?@~Vm|x+uAJ0KN7JmahXBKOz-Km?`yHgr=|4<@eG!!) z{ox=jGHSq`A6rUh;lB2>9%g5 zVO)J~xDJME}?=Dlmo6c*y?GbY_lJ*!@^2|8V@Wuy$2bw8xmBMPadcn}--Gmu=; ze9l}x59VKaSV{7l1^h3a&-ixva$&$AWg&V04>s^^I)?SE6)kZ?J|!ZH{TytKL&j>c z_n{k@IIjijhbQ5%!KQ5VkF9*F_ZVEe(g&-XF0%zopR;h!Y)N*m3YR@C<7emu@KQn^ zx{RNR4WEoa%5w#~{^crZFa654?m0)(g0pC(gk!grt>bSTjRu1YyCCngAD&&BB`mGi z5f;sOiurnVh`RoCal{+iW3e1oj!~wb=K46v#~a7JxJDxT`|=|l0G7Bvqu##)@JPuE zD%-4u*K{V~ge7I9caa@ya9Kd6DvTG-u3am%+_7Ljw%VRl6&YL;`jyANN;r(c-9;9d7&eP)~6~$O?E#leES5I z_ZHxp*LBo1^#cZ;G!YU$#^A!nc;Xr&i}B?n@w{reAoF{-kbP2J81?%GRED@i;-2+n zS;InHJTVyU629;mlDt@+*=WdU_2-VDBE2o~WV@b|5qkVBki|wDKz(d{hr(mX;ESh%@YNrNle&LjkTE ziQsb1aG2InhF65?NE3rmZ;hXL)YKcaVCW30uyYn_eze8?mmabAJtZhf3cCbh2O zxAcEV%EOL9^{!oX!SIpT>$HcZ7Dmu0V_&vv@)>+@ATM}D)Z=~4lMv%x&CK>hurDR= z+3O#5Y>i_Yn^0B3VsG&*Z`VQ$kKIl~jaBjPyLgPdu^WaRG3HL{xS{NeYFw+{g&O_~ zS#7~+8gX+jblp(u}qv@vsB6@7%S@tDZM~K=og84SPz}hwa$$ras)bY5S$SXF7#Ys;< zEhvQT5+AIZM?IT1V;Mbr%96ebH)UNV`#5J^1LvIqW!zL_HwcY(VHsnG@W#$_Bzf6= zL|62Urv8a!O=dFeSWF2^OS@0$(-pL6(nk1I@q?M{mV*^PR?*iqmEF%H!jA4zu}OnD zTo_*pXY2gHBFCF9`g@k%axtW}8-Fkx+bX;xHWPeD8bU<(3)Z$X0*0(sz@^UeZ1u5J zaoM}CEG%mW9#(gAn)$1qwdsej*Y(fAFykum%Ue#|yEEy7>@}?IZyY2wJf=&c`lzs6 zll%K_8MS&l3iF8xn0`)WkDGZ|GvEoFni~PvWQ^gS{7J}I69UWa6}bHPa9(|jim%W{tU9n;xb3AZP)n6Ji@+LU= zcEJJ1SyaC+mYTir;u8HYOL9OlsQ1|nH|@SfJs(FxY2poP{OJ>EZ!zXq+#SrC+$S>= z)hNtPE@!coFUf#TS-f@O5zc^LOh=->$3>i`#d0Hr<9CM&w~ka}!17vVZ6HuD)l&9p z>jL<2HyN^3zQ7^NQs(?N4~%j}u*7&LuuW^ZepZjTp3G!^&8KhFpgNPrNVw1de+8~w z?+oY_>$6^iXi!PM3szT8Lg^m`E^EhQIQAtL)2$`lEj8ot7t@BK6L;A9`MPx2D{CPp ze>L_U+{bCoKFDY9U%?)9n8C5Rj6l+-Z*|VIHOmr+zImhA;L$&R_O{~; zr|uMYcE014mRmw(Q7HfMxHS7?q(Y+Qv-lpLC%o7GmtsZpx1!s@+G3WvKNL*xcpZJhK$ie?NC3=&pC|!3QSL6{yJnJk^MEC)-0+) zj~9p8eXBAO_$f{(i=Rz~bT1_GQp1SRPkH`O?H#gb*g3YYuvDxc6T+{SAB44bpUJbh zdSaQACM100F~QdXzuajS#q8QaHYlIr4K}CZ-tghHk0HFg zpDW+kB}Zny))X%MG9~ZEofdnITSf$#Wut!>1@MMNKSi#N$}IKo9eU;FE2eibTHG5D zL>|VwkdNjipHdZ~htb6=74rdv7gX9|q`KZOA`7ypUMW zt|zT4`kdaX+LJS*j?%=RJ+xI#o^L60CO_t6le|^3crdAs%^I_eKlnI`xa_!9o|n0s z_q2S0)rzI~^f)8(v#wC}s`)gyY?$c6tgEE6#F4rUb|km{1k!}Hd-;?(2bfYQC+JG| z!;_l3*#mE7vJ9Dix@$we&X1O3b zSFmtG2mFhEX{34nFMfV%t<&pmhWzZew?qn; z2QdFTJX^AB9GNjOm+JW~XO08($fE=qTDnl+Z{#%5w~t(@_Yx)kLdSH0T4)K!mH!dh zfrrKKrss+~3Wt;2KqG2g5-0X*F{IU3`jJ^iv*_3_7x`eF<>h_}dgT`$c@slk4sVC= z7XR?oh9*`@GD1V5Jd%YP$moC9@gZeB;}YM+U?s%=C9KLQ*wYDovDE7VUzIk$Z2#?q89zv z-&}OjRv_aoE5&+;L)jNyNq+PDRT6CwL=A?9ig}MaV%tlN^y-K+;?+&N_Z>d&4u^tujj(%?!LSx94z#%TOK{GT}9=ObC8*eL2=PyX}&{5O04 z?{`Qi4W0~b!^5fLG1PVtJUhM&%3rJ|jhB<~V`w|dgjx#gud9RgpH*~GnF$(=@IkFb z`FMPI7w=hTgsY=GVXlS&SG)KcdHO{O7Dp;T?$w7ZdT%~^t#w4?HGLQNSdvMcrfrGB zJyj;X`8u0@Dw1`t7RbGNA39{|V>X~MlV{Te`s7!hs58x-S%tjfhv>$Txb8r>?0cVe zGD&wpcn53q@}?)M-iRFXC0M<&kv^E`PI&(p@GH;`{wk`%$PLSBl=(A^oukCc`=`Ts znhBF%r_d`IvRu{z0dBo_hHp{B=#Ro;e!4~grWDV_7WIkHET>A1dtbA5R~L5b)l+Kj zBhgVyZb8fFSlDsOmD#%OWBa^Y$W>7~F0LcAt9mCJcYOdYB$wGcsc?4SvOgIqvTLiLh1QTqp|6!}ZtSvfe|=Aay!{o_6S=o^KA(>oLcf#ouN!y?Q3G z_}4KFH$%#v+#Gk!nSzUgd zW{BafgnZ5op)n}H5rI@!O@3+$!TbS`-MQZ9I23A}g{ z1)T2yy6xdtUa7u~T`%1XbLZCZiO08cHgRpV*Z3;0@oNX(dy_&G3JOU2tQi;*eO=rx z@njc79Ou^s7Lf6oi>OwjJC^jmKt(qXgj+zDKQm#Kp<3j+vl8U=ad3O}8fsfA>DtI% z$NH`;V^g-N(wteZdHIL>qPOPS+$GJMuwZrw?fJNZoA}}x)OmLB7H%0B@p3fg|IA^^ zQ!;4f+ypv)iUEe*H)Nxq<*_UM-tw#FJ7dl7nON{c9eq7@v2XDpabcN6Z}?tV_O^T& zb$skX7kPx^2K7SBe&y<9Xk~zVc2p1#<7iS}dkLjn%W&w!bZqS#ELxTQfyz5M(6o3B zK1W{aVO&ws_O&3%6t5OA&r6T*~{r7{_L|xxviE&zRAwAGC4EOID`8 z9CKB5QQ?54aG+-b)_m@dfA7f($CNIho1~9xW&is$XUujYC*jYXy^7gq&omK#?FeWV zOs9@R4Pf|r%Khj&3(B)oV6a0qKi43Wem$X0l_w`q<9E6Ed)7|WDY53(5e07YY)$a; zaTWri&!MZ>5kph2(;>a;PFhCY0JRhAY4^5k^ptEdtxXWYOoQ)q)3mc((f1rq zR>mFTN8iVHelB$KmU32W@JTdtp$3{sQ)PFNb5;UR4ErEEdCk68+716+z z8BDh=8Zx!EfbEH5>K$5uLlnbE?i&XvcpU}(En{H7=d{UABp?Asw}(A8=kgYgtP~{=<_xOHe;3vx{ZIr28H`T zpW;!LJU5IQg)Bk`au#-)O@$*GOHoes4-3(iG;c_JX}*20obrojQLW`aS@yjWNGS4z zYqnz8abJ>ApZZLr^f4fiSaD5GhAFO@4BlU%#T5*lvEt3LzWy0$n8? z2M54)LM-f2Q=rP(*YNw)60|Cl%z(Bh5cK0Tll}FLEs|bHPk1-+Eq#+v##9aK!mp#< z-?75CQ?4SdHW#Qr;R@{@yV?GR$1F=&Dc+)K1txDM!27KqNo7|Xi%qo`W*;;V%1&hx z<+3U8GW3tpXLQ{_e~DLe(ourux~`0tNjopSrY%2S&~f4 z3KZM7LrRJ~SInzo#>Q-n8prb=xC7Kc^FBGHwiL?J*Wx*wYwUx~AL^O@Lwr2?0?S#o z2pUIN(Z!WjY+2A6+SPuCd4E(ST(d45I90-bbX`Th+)73B{FS(5;XJ%EVg?gEbma{0 zxZ^yXcR)uTM#GDDXy}GF`1w>ZzVUy=?yVd^r-j?GLnk${zRU<+2k6qe??Ires7d3j zvTg7ylbdxfErBv_M0_-AA&7ky)fQ0nF`BV*a}Ja zV|4#OcJ-bOuH9-W_&-oDKQcj|zMJYTK3Sj2G?PcMn7c#yEcZ^f&Fci)we@LnX!qHQBkNYRvVlF#>`JaQt?oV`dOJcE6)zlN zzs!QDmYD}Takhaum>i~mx}=#@(n(nQClg9;=h7GVIz@TW`A*v=j9?9o_WaGHHg>CJ zIJtH%1fP5#%0TM|ecm&iUQutP-K90$l?;{5W-ZT4s3>LJ{V)IP@5dKRKpB1#2?j^?LtI<

f zTd*>j-WmQ!EV9gKg`P30mQF^u-lh0SD*!DzEl{-L0LpmR;10XNYge=MqjidLUA1`mK-OV^|>@V}%HwD~87edrZbMTyh3W^geKsODxcdQ6;aUK#1-GDd!&gw-6Ang=6F|o3Ih0-~gxSaLg2}e$AilT)j%B}v zz=CJsReTy`mvI+`a3y}Np)9}qele&XYJ(bkWq#}EZCGN#b!%TO0B!D!S`rfn@NPA1 zD2sryjz^55`8Qs8@k?SNeGJ<4Ey3IU0MV=dPE8+8LE)%m{KqKc`>&T!Pc9U5&U(CU6=85-2Zyp`U3n?{F*oM^{=ZciS>!;Oaz(z1IqaL4jUI-ztoF-JaZ z(U1bQ4@T$=YB>VZOH`fTg+z5{TtD_lH!|BG0 zR>XH$8J0Z@Bu;-`m#b@V{CksGxMyrR`sNqn-&;?yyr&(xLO32N|A;krkD|v4XDrCJ zWu|Q`rF(kB@t6A$Bcr#M+rK=by`_;n`309q(0W?7F6a}xX5(e1Fm)Msw+W=L_z&rw zTn+d(@E?gjsS3~DsKfRA2_XI446J@kfrBH%MAOw4WM_v%Njd$pXM8>>X7 zT#g}g-f*?O)c@$k?|p1tcQl>rv55-D7cRqwPlJRHxsK#ym-) ztN+i?_ibRy4XT)gZf%mJ^ zg_TH*ruzNKxHcdKp+6pXW^(h=*K@H!(ip$~Rl-*!>8JXK$LLMp^Gc|pI#9+-0>^+^vyZ$hd?qIId81E-E zAYhn=Za&YXeGZ|*({iz{Y(8d7nd1lhV@PJ`VVS=(I)Cq=Ti5@h*UH;zL)0Vsz`mAs zxx0eljmQ%DCpUQiGvrKK}dEEFh1TD!vWM+Hf)2=WSf6<91`=ZeQ)-Wy{ z$;S!1rO>4}9E;RhTE6Qn`KB=wUZ>1}uIhQ<_p+K-!?7V=hyJ8f?-dh|Dp|Cbm*v$* zYvF+=hAmhhPBzLM1IN!#iAxf|!lvmUeLxK?_RU3e!#QZV-UK6s&f!uM5j-&G4YhtM zhb!MaW*l?E7@x6ZI-qYt>ubcgecU$kvml(j2SqZc?+{3x)h8eJt|SwmTe3l~z7zHK zQ93qY%FPT0Xqojbvq94eTJrTOhMT0~1q)X$cXk@}x)67Kc!)<;O7K|KS6X&p5;nx; zV1JY$E*bxWxR2gp-mIBNeGF~!{z*A3zFkDA{0tg0tAjd(RN#qR1wo8aEe3M$h#l3l zn2QF-h;AVt%>Rv&RXG6RWl3=0_*sY>`#~0bZ6zL?@4-fz1(ut~@z06>2Vv1eAoAiK zoQuqd-^O1dW`6|i>ik6pDmTHdKTDwd9oKJZeL<2lH^8O5H}F{`4_Xw9VVYk6u9F=&;Z`i#`3>Rcu6o?l^#wP5iNf2<67ZZAYcqgWq^T;>g$hCS5 zlE01XfBwY8X{T|6nhBctzAkqgD`)E}l1VJrDQas1v`tV zxZYkoWMPkzQ>%#aRZ-H=5d*qLMR4ql1i#=z1H7H2$RDYZOvc&Thz8<2sE!(eTT~4op=tq47ZiRFuR~Yl&v6E2Mx&l26de z^~vn|x+S!6JIBSmd7X59GbbOFbIF$vLXgQ*2gMvAo6cn$f15-?yhj>yf|VJ=OsA-l|i$rKF_Vs5YvY~FjqHf?M8c<&4t z=_SLxJOO0RD1dz#PLRmW4@{)ag28Vi&~AQ3{wTcW^_54_{N#_UTA>f~a{6P^8@n7V zmzk1*SFMDR>nGc827vPVQX-@AjL|pJrqVs1cn5hEOjut5>piiCIKTIyy*WbA@M}2{ z5tD_h7ta#k%pa`%<4{`jX9AWCGsIXgkn(qSQJb+<^g#3=!wrjQ#j_T6(N`#YsOFPKbngiKW#^@bU8?2wf z<(C!Qv3;#Sw(qOq91DMO2KkLfS0n`AeTy(^`%QEmGQtGD1S&7fq+ypPkyZV=jM5KT zc)7 za$OTf&jo|>4sI9xA2(Y{w}q*5J;;%%S4l{v8=3ds7LG`#kmcoKkfZ*Ulx}}Z7PtFD z%8{8cr&$Gjgd$c!VlZa=QpmM%kS$^;p>F{2J0z5;YL_1w0!af znIGctZ-P15dSNbEV8J;R=h%RmwKZgtp=YTq5O=F~3K z;C>6dt-A5&g(q0+(2VJu!~_SWX;7=q6qJdZ}7zH~AxICx(|Iz(x=S8-y2tqC^_86BC1f z?{1QmWs}LOvnt>;^C#`q%_4)QHcb3mJ9rbh40IRAkbu!4;^}*XX?r@0gmT%)*Hzi1 z@Zfl;S(;0xr=Fu_gRxZTTRzt2oJJv;qeyd+&uyp`#r$5O&_M6lZBGh>0qzGaYNA>k{frzu~L2T z+qMQQvR{!MoL~E<@HaBReYQS#@`(F>Td1;h2PZDiHb*W5Hb^Ezm()$L{g4K^V z+K6~HzMzhJLa=4eGj_k=E!}Uajpuwg7IouVlsCMNGW%BGziUM(`*AaM557ldZSNun ziq)B_Cm~Sa_yRufl;o%Fy#Oay+})ow0?WIOWGPxi5+ zLgtK9T0KpA;Y{-b-*O(L{pe5Lah+fpWV|%c>f;`~^tb}o7Fgh3?c1!PwkviR$l$5v z0^C-{Wi#Xtu*mI2yYeri*!G{a>Gl%Nzurwx*3ZIR6-ktQk-^PzV{nU~IlgFXrEX#d zSe!B)UHv1_WuFXgh8=k0$ujD+CW_AJZZ4m&>JS}W)Jd`*D8b*1KzJRc3Emk?;C4YG zl%21KJ@KodxO5V*7h<9AnF;VKQ`yC5kK=Fy58qwfPuH%tp)*6zVvp!yTtyaBp{o_R zOz?o7xG;$p7{r^|YHfs#-qDQBfvrr@K24AnxeeaSk|0;N0O;6h5PfeMS(E`J@iT1deS$sJIWJQMD)E)aP53$eN7 zhCf~3p|w#JI@U`Ha`*nku%pj$a8yz7<(>f72TKTcE4)OxkD`M9)jGKQdj-vq(5DX* z0*U(VvouzypY-o4C9kTF68~&%Q2Z4SP0Nx2)31VDyd~s~1;Bu@0l4eT0Xc>F+&5u9 zv|HXH;g&avZV4ZqeUP5~oWg#cK8a)8bW^YIv#Dyw98zwm!;W8+UiMERk62vd`p>ps z$i&C?kXoVz1OB%l_}Vl$`fDqUYq$mnZIeOc_ImEjwx3S?xr|D88RGI;St#px3-|fm z$H?3=tkiGAkmnNwVj^9rdFnUjR!Iq1zdpS0Jc#?$0bk};;}^9BSYq7H+h4tc44v|$ z;+Lj_(%%~-x0=h9p9o!VQTAgiIrGAOdHV#PnYZB zxO`Le@>Il=jp6u}+xy%xUWhKIq^Z-qLqS9E+Mnb7rs zuJkNCCf?{n3A+1Rk~(|BBdX; zkoM+d)F&x}Y|-SyrSPrb&~Y3*tg}hTLSL4d^o#9|d(MIt;M0r?Y?t^fv<|pT4ctQU z%+EB;z00ya#==a}+Q;OZVOIG*aeZ{VpTza`k}%2T9e(&RRWSRoj9~MQF}x%)fWHn% z3&uxBE>Lxi8xW$MhEtr?0fp~#%j8(D~i~e{Gu5@Tyc?87md5Tn(Uf+>Ss|Lv)GxrpX+I&ZQM`I}kC z3vsk96W_j@o;%fQ*1-#=GRZnr^kpF9JTIJ7Myqmn@2OC-AQ{H*kmP5FbilgSVz9Kx zhUXg6d=o=O{=ISe@bYOdM*_%GR2(#r16wiGaYBNA3tnJqi6i55~=;A7!24PhJY3Yvbg_%!koixx zCMuD7igF&KlMArk>J0UmWkds?7|n9TNmrNqogw)_d4@a!FETJ`T7f_i+nT&nMcr4i4OfQGHVs4uXZj<{< z``)L~%QoVe*=B|Jw3@(qvamCLW`Ud)9H6ZanhR(wut2T_|_Q=RyDZ4=#Wu5yvsVFicQKU4r zsA!PVTV`bMj7UXDk<4>nkF*)t3 z*=J#*=1MqUD?uV28h~3~1@l-{8AI>Nk>qt^c+n{dW2DAl{l`vf!#w2-`F+Q0Q*W{w z_raOeqhL4E1zypL?5yY3?0h*>mX(^yDqOW-8-ALyTOGvNquvJWu~H4*moS~R+AGg` zEW8T|2lGI|y_;nGj3hyeTS()iS`x|$es3rt(7lT9JXKg>T* ztPJpcRU+Op7{&vU5AoMpVS%o}4zvuthHtCPac$WE1~k>+NYZJPHoAbm^WNdm!(N=9 zA|mLbT7vbXrh>=M7Yl|476|fdy#!8wK(OV9m7wOEnV?tGM9^dka@9S3^mk@YCd2cO-j$I&_mz+fQ?<< zG>_k@$j1nSL+gHG9{q$4uYW}!@I6{0yysi(r7=GTUPCUpnKv7^$I$E5Hn!s*L{sP5 zmo)!oDSlaT80X4P;557qP&dDvlWG|yonF2$Y~BD`j^9YIZVUmtFy@~BJW#@GDeD9()uwqGApLB{669N*-td>;dBgs(t@edzPNqrEiO%}ljgNpGP=LD zAgMHiOnUy6s4vL|T}O3z)!#>MUCbxzp7fHtmCa=Rmo;3@f64UIX=9ui@PfLw*VB!$ zrl?oGh?+aCr+vhL47w}ej2*MF?`{NrTr-W>8(HJ5nz2|dbF5i)bPAf5uS9?uD0$Bq z6Sfc{ay7kqVs8;q^C%^gE?y;S=W00d)e*opC7oJRPI^gmKCvd-5g79?Yux>0#b=>MfQ=eGUv!Wu>{;uFB_Xk`C}) zDisix^ai$nPXkBiCivd#4I|4YLxEBdlm|`*>!YT8C-P*tX#SUoKdFOaSwAq#odZv0 zS3`p&?gNOuyVWzYE|yT@Dnb$dxILvEUH1T1Di1F-*jyLVjz%F zK7`_Z=W!JqiqZvUe10VmIa^sAE47R{@%SR`OFB&?Z+;|iattBh`AW{*^DtFg@QdDq z4fLVAA70qH2iF=Nz(I#4D6u{UAJz<0wPl~_)YGrHfTYu$6!qf%uJZ@S(8CaHpUAU? zQE*bCf-1~4vsvC_M5i~8!RqS&s0Fhe^!;?9tn??j^L!PQU%$kkT}nWsr4w?u$HSDF z%AkCFGrS(ZgKP>eCS{X%kvSe>7*zhCsrR5g*S*k^B%8b?pZPPkU1~2e3f6*#yORL- z&8Fw>sX^}tUpO#fCLC!WN5=@qF}*Kd@a#(t#&=lp*%wRn4_J-9X=%8{ZX()=o~K*B zM^ouH9%R{;P4LcOJPrG(jXlX~xOdzv6#FfPLdp`DaD5FfvgP+I>#FI%!fH+?DTQ9w zm=AHS3h>Q;5rj!hgUlI{u=Yki{O9fiPcGWQgr!SKTTT|0dV4zq#iPMXzuxn?&p0k%(P7I#i@SMW1C5b>QiVMS4hwG9JQS#Ud7DG84EX0u7nu@ zt6{!aC$xt=g4L}NAaPRylII(9FB-OzQ>2kuBJNB*0)91V%BhnZ7c}7| z?;yV?SxDMuih_0P80HCeCtVK#diJa%PE!U*`sOgW?R*w$v2{ z8b2q{Cr>`O6T|N?v^H&hg;*HV0|=mHpEfAa=aduf)c7B_;_p@ zH7y*W+?7kTZee%xFN4oq?Bh@xRo7(e?)bWC{pTKX=|gz)mkc?aa;}`p`Zm#`tMSA- zNrig7c4H2$8|3bZv{N770cyQx8n#$;p?cnVp3gair%Y1uGtV7umfwomOP1q3&CRxP zifLr_2ii2*o#zduEMdLLIxwPd@Da}_O*pQGzuR`AgMB65OMZq$&+7SX*ioFv?{)P1 zm(lGGY1HwA8IJg8;An9uT8XyPv{@l2Iu_~so4c`Z&kqzonSkC-&iKG+D;h4Sr*$t+ z;adkWfqCvzYz>tY?5aA1nUCZIYKz4MKlePr!Fy7Ipx<@^jjN`D1UXZIldQVH+`>rk z#?PAEN;bV{O8!cp7$)yrUG@q75A zQ4YUTDq-lkG~3pE9G08^o|BIOdiTAMMkywMvC`oT$%E)YWCp$EmZDCf~s3xd}Usy0EK%sm+*&&S8K7@}3%kk8b6}WG?3~jomh%Sxl zcrd^lh2QuSm+5yo*)gfab14hMypOhh^HaM0d^~aZ?+NqsDBo~k*}xRe6M+KMnY5TdvfPUAnX)a6qUFs*Zhw1(TH79lPrwBisZ$*H47ewEtkI{85lx;{n?-22wi5Ll4z}4*DS&&k>p=3%xo)ZlwHo?{WwXwEPgDGZq>ou z*PAe~bvZ_z)5jUy4Ak!{!Q9F;eCMzRC!BCY%hnzE(KZzaDwAkuWD0gJKa20mEpZ0F zmy3`p!*NrOV&U%^%-&{VWj8*AL zxFK^FHnSekBhn0A8*Y>6Ef>iB8?96_sh=iI3q_}}9y)!f276qUFl@>;G|$e$*f9ri z*YDj}o4N*HML1zzt~y2piK0&WH}bb%o=D$J1YZ+N_$^ZiV~kh9m7_MW^UGqG`dF4{ zIxoS1&r-TgJA|0#R+6<54>_}y?l>^l6F>KzMkUcJD8AYQr#*7PHMg}fy&@g=JF233 z>r+|~{GADN-%LH9N8!Ne9Ati~;4{6LPKN#oo4pIs5xs*UwN&O`nL-iy%i}$eowSdr@^C76r z6goWluA`Yu7Bw%h@%*ZmoiJO^c za~$9M`scL*>RpS)+|;L3aS8AGb7S$ux|=lpv^?H=#Cz*xb2#zet@NbE3@G%Ef)V>h zvck@soGj|68_eWTe~vId`)EsBRQ6(wK|3C}5s0&-Ct*nRGK}z=jOR}7#rRTn+;d-p z^qE@{m&WPjO|vlZ(9QwR)@T?XdKDH)cfqU20uUKdVnfZ;*dvz__6&%#-KQtBYx&>7 z#7l`i<#Hd?M4v#`*+0-edIb_s{)3Y;;I)LoH>G+;V(?AmoT42t*NP85SG5ra2bHjUPM7)cB=t-yAdIogW zdSmS6JFa4Kw_`*|9cO)KHr_r(@UbUBg}1KEhk7&a>>C|Yq#lkHPyVY3C?i*F$J4*Nsb%{6c%cMV9Yh!aO<3b{}<1G-wgU}8xw ztRABedWQ1wCH*NQ^I#$upmu>XUK2p~*z6~MCj-bcWf3Zxp2j?WkV(TF1l$YX9vRG)8uVIy9AJ!IzDb;ZKPVN%&O+x$E4ZYjP;FhT+eN*HS5W zWIsQ5e?tRKr_&#u0oW7qA4-e-p^feGB*ktC{wH*w&nTx_?_X`id#bXrIMtrYD#@aP zoDaTSB8lD}-Hf)xH8QvGGtuzLgZ-UyAo6=IDc&%fL~UCDlM9R(k%u)n->i(beKW?^ z?*=G8{FPpEQ$VvV8hBVS8h>jY#MEs6%*s;CuHcg1c*P9C%e0bp!>5w zTwb{rgl;~A=5K+p{&h6S+|-9%YXV8$>&e7*`3Siel1^MJKG4Hn=1`zwfEnArk;g0e z_gw5>s^ghT`fA(27th1$j&KNViHCD9bKvH`U7&dD3LGt?Fz96o_p@)1kBi4J4Q8L2 z&F<;ceSZa+S6j#(TJ1y&ZeMr_k^4483MU?8KI%HsB-Jo3_`!MZ)TEQm8KHJW?fES5wswXT$6Scr zmkzg+*7BawP@?8s1UrY@_|K&b%(3ewCrK<#nrOnz>xi~BS6&Tg7KlSa#P`;AnkVq6x z`E!E)xHN$n{kce|3`yWjaXB0mQ6Xp7a*U*j6|?gw!))Iq54*WU_)t0vJa3i3r62ha zbfOF_=kpza?LT4r)pM|QpbMzeesJ{g1;4p#_^#Dh@M=$kV|zTIKinI-UHLra*diF` zaS5^<8^K#L43@_(hE~;BNV=Iol>htO?Bj~fZ1(d;wR)iNZ3k-G@}8f~S{QL!o9K1_ zW7ehLqt^R#>8-&sF8k&|=Ay$5DqZG}mO5fg;`%|xx=DpBHkGFP?T4H9Po>1iqJ|On z*vibYZ>BS5JfuiJkNzw0Lu4#owmC~b_gkQRkTIT_ zGDVlT!Z;u&Od07EixVwkC-})RtT1USti)?N$T@=c{Db zF=hO|!V|4dmZSf>gIK>b3w33qP^iBWr9K_Oc5_esU3d&v-@CffQV2c}I?hJE%u#!@hZ6n{5izY3#^ydM5HI^}Dg5x%@>voxH%1PJc3=e+El& z&%-=mev}W~<+j4hEB7JfS3Jc1Hx+KECcymo6zFF}c<+WNMD#Y2Ze3~eZi@{3`aTKl zqbZ+@DJFV^q3WLsd~;_SUe-8^RbfXle90b^G5f(CXmO;! zmtLlq5AYqZ$s^q2_7XDjnkWe?F6EBIWK)4>8eOWpgl-9wB+usmqbJqe85O6G#Oh}m z7;1&Xib;8-x7iZ1x~kz}_C6T5vkAz0QP`w%2I>sYz@cAPfxi-imV7zvQYeHA&-&qz zSP2wV@pp?)!{jg4bFk+znLd0J-(M@jig{P@Tg@RXoqPo|7Zspeoj->7yrzZ<+i9U* zFeV&JqUm?k@yV65IJ&PC*LR%4!~gOy{`O^TcdJJ6nfEc5wxVav5H4}Ogv06$Xgje8 zwajOse)@J?v4`K)M(d%*2DB*x3*cgB>&AyY>MXaEKx#c^(tEvTt$ z!5y7FWQK|s$^78Mxn(KA<~8K$=lLU=&2mnK%S zRBlfj{ktk2ONNc{-&%g=#LsD~ozu~Is0yq6Hldr-Da_UxrZU^wP|k~IS~_OpWv_WC zxulujKRKaM&?x1hV0v+0If^^0;yY(s3=x(>*|mwdXmJP=3^cS(U~#9G+-tjIE`xckUQ=yvucP$sLA}3Qe|G zPKG7o|3H0_4K1Gdl+SowA}3=Xk{rhh@^R;N;(sZbIA!|6MLkN~wuHm0C|!`>HPGA= z@_JXy<>fvx$7dHbQ(_&JX-oVJTn9^Jw zE0jG~Pakfaj(3(krK;EMar%5WOjEJJi0i4SfABbE`E$FJT3@p(A7&HxT}687H@=_Z z8hH}&ox9%9$!&I6#COEJZr0YSB<%cYbmH{@Vlw?Bb44PWu39zBw9YD~Wt+0;iv4M{ z_0}P7>RTJ!(J6vg42|F$?~wD@xd95Md>~IBvoKT~2VUOanMuDVGe!eX>BSFuRI9Ls zj(>WW#vhfTkxlYcU$~2Ezg5w3PAdO`g;07b^^^deMGn1--RF6E<~Nq7u3=vihf-`4lNRXQsaps%lfTQ|1%r{0``N3+X9lHnTqonPZHXJa!OXeD`gfb&12-}nc*WGRa zd`kv)62J4S(t*>m*3f+1hZGkpLQdvf*nhU11WbERCe1oQ%Au68j2KVnpUIr2 zYMMr0_!;r@+!>gaatb$W7^P>0=inEkJKW6gW#quK4_vqBLVPba3x|A9^LsP?j%YyW z9cB-W`!H%NCcN1eQnKj7Y13#r?|ogg=roxVmBcxpIY+9S4T*GzD7<}lk_5&`k(N#| zo|i~L{gDDRFh;61dYNV>Pr#;oQ>bXMreLcqBha}oDsY)Ii?6g36)c%&jZM?1p*HX0 zZHziZB8?)!?Fm9m_;zSscoB&}i6)G_$3RL+ryp5wErjgKUZ7hVOk&QB*j zV|}1Me>>SWaUW4vFNNP`Yv9b65+K^V3x8q@i8`=?lr)7=E3tGgjFrQP#&~?NrW5<; zX$TgX8VD?RJiz&9#s~r>Bn8JyrU|SQOL2qUSWK}ArUx?P;Z1TX+*kbtmJ6NPZF85g zTUrZQF(#FrnNY-wOq+iYH@x&Hd()(oQP8QwZOf^7DFewR!n;Q?2)$cM%=aF9WYKY*W*~^f>BVpd}GQ8a1hdVFxZsnCv zX)wRzl=G{kG_0U0v^SZ)j6cNZhP`3M*d)+r#zMl?3our(8-z=~z|f6fuxrsUJh*in zK6(9x<&Lt@%s(@B`*f2?wM??!{$KOomy4VB-yKUsLNq~tZW?j?a)&_JB<@yd3A1(2 zN=S2#BS)U4(1q_Va_~vOSRc!#u1d{BRndyh{cOY7+1G<*@Fl3Ty9m)=#zDlKL~>)D zCOVjQ(CJck7&GFIvu5n4SmuaB6>IV8D-+E4_J`{~Xo&qEVo_$-Ic)RDL%%y6I9{v@ zk9~5*Wpies#!CSW?vTTq;ahRSbq~zllt!t44)^l28trPEO?z}A;g7>+2u>JcYUK^d zc)f1!Z^ug8IeOb@Yx8va^`RUty0s9aS)>=@57Lh9UDP*A9&=2Sak~|ex7z3F1gR_3 z*T9u~d*c;(x_LI$U)DrDHH_)wZLgWk3DxAZ;~G9cYX-u+vtRDoVmdB3lK%HPyD8^! z15Ml9OI;e<$=Ja&bfdZt7u%+S3er>1%QTMOo41|FI$WcET5GZWNf26-ndHK9OQv&Z zCB3d=fXlQb@!t2@=|hYiA@=W>UXNpQ&?t30{~y z2~S+DK)+9BIOUZkO|bZ4{poHlBg}UdnrlUo(TB2x(bfi$7gM0FZ9RRmZ6jCwXgp26 zAxGoY&rubL1dQ*L!=cIwF7<;n`j;W`*m%yW>Ou*9v@3~*sNN=#m7mC2hjyZ}qmNK< zvVHsM1!r{W2)RDFhl~t~K;44+=A1pe7u07l%ugB*w=GVQnz#n?dFVK}$0UMV+iEDt zHh{Q&OCdHQhQy>3q8lQGpO-IU6w1ZmN5&^2*Et0`MD`KMkFg;9=q0lzqs2x@YX&(m zynx0%+d>mtMM(93qoiionM{Mr;PoQ~CcX`ToW1@aWSLA#c9oL*_l!Y!_C(M=vk~t0 zM!}(dd*Jn?Je#}z;*9;Ah1}gfaiZ~TE?V&S&9u}Gd^2`F&JjJ0tG1+|wOy)IwH_}kB%oA8@UuWxFjzYQe=l;x+{7tp-D5}|yDQ<f<918hn-~YX1ZLwj9ynRxjSDJc;ehK3sC` zAbsaNKpmBH>1rcctla*G`n}@W;XfN8B*7A%SY4o7GYpwFV=YQgWHk%!IgpPp^hjKF zJTZKk+5B~#Iw<~7fNL+aVAg*dpz=vDbRTR1%bXP0aCisVt*(xTI~6f9U^P{Wi^t!) z&!PA<8N}fzjH7a7>m6@goRERPB3{vtYp>F+0d}~r;~+|&%D{xNIk;_WAjWkZ!cVO$ z=vPJ@1I;Op?BuYbM;fKvZ_-1~X`H_6dQ$a65>w0WF>iy9(A7mIwEb=w*(fGWuNx_% ze`FLU`a+w?104m$XejWKTqUqpRTn(FKS7|qUs*7+QcqwpTSYK;bh5xv zBS3KfU<5K5xoD8J1T7!GryVNAM(``yI`^CJ1E3lmuU*eqqrz z#PGrnRFjIvy1sw(Lp1LTcr#3Uc~4^49cP+)-hdP>ZX<7U6K(hX8Y1?s*NHMJkpA=& zr2Voie7JKP(h7>8T;??F3qK7K^Q%CCpP}AAwgVQievq`*8C02R5ERizXY!0_)c9Pw zSvZn@yOmCpu8+s*V3fjCmgxTB|fsp zVkU)Fr5j;i*az;#fGUbxJf@#vJht`iw;eHBh}Rz=J)J+3P9rUhvYs=US5S$aH@2hi z_yBGP^=(cpt|Xl{+2l)jHGME|1{f4I(iuV$z#1t-aM(C7W#1Fe&86Tt{uG#6AViBf zf*-qsTq=ntCYmN-)lq9(?EjQ`9vDU4-s_R!Pj{JQoj4kPXAX&~naw-@Oh~iiSlHbg z2b_cwDmN47FTU!_D=N zB&}-3d~Q_;|Mcua!?HGRdT|zASm(zbuA2!fqt7xrM2MXG*2qb3yh^sLh$REVQ>e_> zXf!-rjXgWEv7FB=v}}pT-93k!e>nZ3nKN!v3$-8gNVg0|4andj5-yXL%i!NdeTnf%g!Wme7w)>O?Ymh#t2di(N6I^it3g+DXc|`BuVIxNVR=?K0w>0(VKk zm51EU2S1tDQGw*Ly)=Hf`IE~EH6iX1qRmdzKXbY2ir7;$3vW9erRINj;D(Z3to04Y z&?*W1YwL((4E$+Si!*WDDn!KZ1%g8{fadj9a;-JGv3XJ^Lzb~*YR{m74eh% zOKw>0tKpm+dCvI3OOkGu zPnLg?g_-^oT85WH*SJm?a4&%EvznpIE*Qw=e7JMWj%RzFAQX|%DV0Q`EP)=_Ckmc2tYvK?pb@`egk9OpJHmuDWF z+)I@^ACNr%|48$lDDG5P4NaV-hTB8d;-4-2obW?B9#D-!i^whb_ws&RGgO8v^n_6R zSpuq>2XaUER1k;t8=+O41?gc|-hYq(HP=;*L5k z8NHm7u&cXV&X> zLhR>cNRf90?t~(|i7@~J-n{he_CommX(KrRF190b&xy;a*&sRl52@^qg%b;ALBZT1 z;`2fmT0Z|H+m^i{n*yw0sg88>$D&^DU!oBe+gwQ#i4fdtI@lb$ONIvPwi8Y1@t{0* z9XwFFNp?G?^Yh#@&~4EO8BcD2u0{-ObrsM{KOQkTd!N#e*DJ}cH#ceJ!(Pt!ydvbz z`a!ZavdM*SlCVYd7FoIBuT5a=3*vbDHCd>6iNq8hWUBtgK-au^#Bpjlu{E0tWp8Sr zckE$O*xo{%FD<50##Z=mUNYIzHA*JAcF`$M8t8)b@wiqh5YOzmK@D5Canp7_CmObj zushfgb_n*v_AhpD|L8Mtu*e6EGUVSGiLf%+lRO)B<1VOqg7^J!M&LCGHF9b=x0xp} zu+kf=bhYuT#Or1gs|@b^%ujSnxe$HtUr7@p^iX8Q6ZBO+fSp>(XnrS#+S&C4!;K{>DjP!x&+s=u(Y|~rHlTW_>A0XNw@tqBn>t~CSW>#3PhjpAbTJp#Ao1{lxVEIdk}Rq1jx)f!&jbEp=m`V4)0T>YE2H9^_~!YwGN_{ok}!% zme8JkqNw^k95)}LnDej^ol?{Umxi8V!q3x~u!+y$EV)Y;`k0~dv=9E)Ktmr8!_It+q6M`5ezURWH= zvs=fdVTw;HOMguTijB1gO2?C zOWqw$B&O~{WM9WF*uH2p8TZ2oOv_xMGb9WyZb}BRhi#y%lmQOw2EqGtJ{;Bx1+k+= zu*v8)(KnW3iWcv{=MF5Um>tJo$GcG~`yif+>p|h#74VFIPelmCh|g_x2!H;aj1%?) zHz7ax>|zA_$#~H8b^vY06gZ<05;nV%O7BZ01^eyjnJ8IkYW&2Uf>H9{Pzs~jVuUld z19$O~6HY5Lp%0z4Fg81vUi#)myNs9PA@^C-%VRAOmOnyG!uE2Lj~6ikUwOyn%klJZ z%~jer?*={Z6F?WLJfcG78|dPR+fgc9A7g#o@ZQM^42<4}^ZicYE1zdHF?%uH;2TZz zcDrEKlk2o#)=Bz0aXxM@l}7UgkD6DF?`Gt*mGJl;AzY^BPMlpFiPS+=c&e*PTi!L% zlcUkxdG!&p>!Kn#b5f4)mD0zBFF#QUo^6TZdnDBM1k&Qb)pTOTBsxbklhd1fgIUn# zL#O_j0CwZflTO|(cDmgbK3@pr%p_)k)b1%{bA>X*J$5BYrL*CC%L6c>@3rx+q9+yK$u%TQYV z5C)a&A>zdyeh>T{!uAQV=0u6TI{Xo){rn24GAqEK_b}|5*$$xqk*^=mt(+^A0U1YYFCO{KmgyTk+2Ac+A(dV77-{BxkQkLCB+Hu*7Ht zj5ChIS;b3m$uI@z*w5r`Y8_0A;h!5Vt`HWxk!RB$L7B>GVvy{Hbvu?*kBH+~ZkmaI z<704&@;j=OxD37CJ|NePonc$nB)DCa2CA~Jpz~rb1neq=6Z-WK_)d&HvB8G@w&Fh2 zcCyHbS~&w8+XS1o$V6yn#{N%-Y;9R?`fqz>&u zI90&&AgfpL`MEjNQOpp-{QS}8>^u5$XD@Ni$R&GPtYFeZ9Y|Pq5?1ZgX3JlVVHe1c zVZ)ythg~U0!9xBkv2V)c^~Yk4at+s;;)h6wFSckMuLh#BSGXdeZfO*4Z$;|B79qL z1HX-_!iLcYxa?jj%AfSZrH;k4bX^u1>puZk{Y)f7eS;)+D1i)F{o;sjCS!6wk}Hjw zhkJrJYEL~-KX)}|Ne$zOv!+0ONKK&b^a2(7-lMD4eOx2liAp~ZCGV=^`+#f?y7J*l zAnyn}Db2bFU4bu({~#=_8NRGJ11{?dz`KAyM>&NMJt-c6^nFX?+NPjm{Q`dHGelRn zzNFq=8mQ|sf+h10;xm5cYqK>T{iK~RPR{`ErNzVYjw(oW+67rht0Cl*BwJ;v#Rj#A zu%iL$?4zaKV9-WkhGPco=spe)J@~u2=~t51kVM37VrjQj2t6Nkm8%(t^rBK~^L%d^ zX4Q^DX7$1dNOoCH!oD|vaw!G%kYNaYehMslnxXB=Dp+n53C9nHg44reAb3n+!m=`W z?La`x_AtzT9tX?`YcknN24!z}(fx1w=sek8oHd~y9Z%fG!hmI{YO99dDh#nGVjJ1g zl1$6m4e@yvLzg`Dp{YM7a^q);;^V$aIPFg+P1=0J#wxIwT2vb|Z(Mzt=)8RpQs)c7 zUAzz6crz?n&>Nn~ zIALtjh-eN?^vAgVMzr+hT{rJ)>HP4i_$I0x>mu{m@8PqY0^6kSLyPQbO6r zq2||bZNNkN6jeL)m#XQDIzC< zD+{jNnjr{$ttK!ORuqH=Y6@E0CJ6R>X$uTyP7rK%ufzQxt8nJk4xIc?n6X==Kt9VH zgNfBrY|xiFaNMZHei)d_I%v4EXWudGV$t7F@pcOPzH|UQ7v6%+x&zQ2&HF~p0{L^q zD@MM-0ly6u;Aw*f%#f177jGnSW%6Ofz4mA)rjAl6X|Tdyh+XHPzr@U_|vJ7+x96&J^hf-69{h$Ezf3)J%Y>+{<7xcEy9x*e&(1Jxziux&S1xJ%%#>NixYO&mg{$2Qr1 zi3MZJOi;L2#Czy|Q*EY@3hiHsqW~BZ#Nzl&cU0iHmWrs+5vkr_`l7jZ&O4{X`Lo(Y8sLSJa64EdmMg|tc*VYmka8!BtuR^OvgR%J`<=wwb?&%qT_X*QCrJ40tJK|Q zC|vIdZS_sC%C-p9HA}$pL=+inv;b4FI!GEUfEtHV=;+xCE@JmUNAL#RFaV~#)q<2c z&J?BvlL=-3-}j5d@#`sMpDORlSv3(Bl^s zf$r0Uu;IaZ{M+A(l6&0f!LBpp?W>7+{^tRr`Yx8<%05pQt($?p65qJk+4mX#B}8|) zo#ixxDEIE77pEQYjD-2~JyicLk^g+RLbU!P(3GLPe{=I{4jhC34CNpfg_|E}{YN>A6DpoK9_?`*H%L7Kp&u@e{%C)k@%}_@LX@ z4p)jrS?BnEkoWroZEaHQTsdXdCCHGqJTJ^%nU)8!v-5yOa#8jL~JqFdm;-j?H>@n0Y6U&vlE_e@!K%tg(+% z3oD>nJB7(nE4$|MQ}x7F#GTw3_9h>tn#lw9E|cTYMlbehVWMRkJ~D8_U#e-C8NmCk zHYy6vii~2-`ZBbgwH4ELPsYmK(HP7rbB(+UoiPN)R(6(hPZ2V*azJrz!zwHXR9IS%p4jhruFa+fJyXobV;hSD2 zeEcp7L6_fA(OY+^>V&hn<=tz%qpBqsmQfLeg{cbO*MOkhWUfHF%UY0~Ybto{DI<`d zK7tBXxftnTjC&_6LVJaHp7VGDbAIWu@Bc`%(hJmCm46f1i!w57$Qcb*>F5U-(c!@L z@fOfMqlSz2{l=uxTFgq57F@UY5tPMk74S9{!P~-0!79zI0*oh! z%{5bc^7c-u`fwhoKjHT>-bv)u)7xa5v^X{nl~X@~0i;;#(!jzPw9yz#dps4f@Kz)h z{(b}J@b{b(hri&Ka#O+01t8e)e1{;Ze4$`w(;l9+9WRKoA3&>ttGL}wRA9H#6=g5; zjl`UyymeNU2~$--=(=j20y7#w?_3yuCq zh~w#-#Huh69-tU(TrCTIj+5ZLu^R}{M|9$jZp^v33%9Er$L(X1Fmvuh`buUwe8|m$ zy&ufjR3}+Be7Zb)-0BK+mfnUa3t@J;Ju3fr~)Loqh{>uaqWOW^WMpc1IOFA6k`S>k{9blK)02X~; zz^~MtJsou$D#GVN?!E1ha^@HqyuN8WDKCwz>|R74hwE#2}qa&>2sfXURjjPZOygzW@vd?=gy>bV-b20j&=9#Qd{| z(JHVAr#-nttLnC3?zs*4GUWzZ)~n;^BgyC?l8qaWWZ(dw8?$>Mh3B8j(ZO63-1%xP zdQVJ2C(RI?nR^E3rdHvjL!($XrU3WUrQ^W4O#I$efL5t9aB=?;d?UqoJTIJy{_IM; zcsr9*7S$koxf&9*a12}w2;x#43o&tj%mY($#$M(o^zN(;KGVdrQh9_%4+tP9UV84`WYVfRVHluyJ$~s1|JiGu3~j zp(3EI0>;o66ru9p9r`gS5acEgx^#Jx6Z&AKVna2)jPbK|4Jg-Z7d_Z&k@+ z+(;3f13Qs*nIJgnIZlwB=PxLCo+U8)5i4l5$QQJjrwh7QEEFUZ*$VUmf(4?lwh4?* zdI>CjwFE<#Wd*iQL+F@YfZsON(*3S!m@Z<^`!QC)(^GtROehCwl0|Usdkt8gJpzhd zIXnlr&30#w6ButBBveEV{AWbaT{i!7bmrkyyW-4SX6tbVSiy}%iP)dYK zQEB)#YBFSqOqoS8RWv9h>}TziN|a0`8dOwDGtDaX?%(@||6JE`Ip;ikJ!^f|ecz+T zp4E6C+VS(k4or&d$4yeNXq2=Vih3$DWK}#Va5klje1w`X!jb*)DVDvHmB#Kobd23_ z;v5_Pqk(nvJFc*M)5Z{#q#PCiujodhxvB9 zIR40CZJHZNa=x#^$|tKZy{kfaJ`HkGx^swj?nrXx zvN)bL|3#}q1z&?kEuF`Gq4GVqIV15EWKGt5m|nb(Tpij9eAZfUdi}}f<&0_I^~p%^ zrPPp!cN5@mZ4<1Y>q71o+k(5eAP@cTq?uv+BAtR zaT&$>dJe$Pjgo9qTw~dJL{==w>$SkHJT8wJ^0M0`(^~ zFqZWRWZvp036PWyy>kaQx8^kwMrpk6wc?jX#icf@~ot}tWS2CJeLK&$H&^3X1gX*U~< ze?-$6>q~|(zswgDLP{Wh*#&59x&i@?`2veA3#NR33)c**;IY$xz}^;NojM}m!R@1P zJvJRAr|g7FpBjnA!}H{rXCsrbs))OOdow-%b~1i=Xh3gYxT@`KtUyqu{LFR_v0qgy2i@9yIDka8Sd7>~yt4x{JZZD>B`6t+-Fo+&$x zzj6+v_L9*UDb(1nYW(r(xj(e?{vwQ>kioTYkf&NU#l+OxlYE<}4*HhoL0X;%yA6Wh z)afAXUTzMR`!i}9KOQ9;CtRn=TR+l<($N^^+t0j@_n>zd3(b>9DD-xA*4!X_Kt#F{ z66W>5#LVT8_xCUyRyzW2-@D+M*Govc-3S2<`(W4pT=I9)Yc9Lz64nH4!N*hIP?hX6 zHCEaqYs8DIm}8q9nI{Y7q2ggT3Ed>*HAR*{YGX3o+9NX%Ov6 zhbF_3fY&;zKTp}rq!F3`X+%Q7gxi< z)#BVmHJ&WhtboT`2Eg;%KWON94%-dOK{4eB?3!}{p5-e-l&}v_wN8a>GfEZCr+Q*P zm8mg1eTSOOdqNcdZGwDn4=~)mnY?;!4OiTRU1KW~*tl1cRP=q~^v+I)g#2?{OWrSX z?8+qiV2O}vT;WZ{EX^^cX9O-CJ49b-ms5ZDXLPpQ5GQrGjb58N3kO50>3bh_R64In zx>496Rdm3!RiU_Y=|(j6^uy}+qIlzhI0ou{qr7|#b(^yUDpqFT~uH;M9d>B&pg)@B!ei{w%nv}w>?x(aSz6vi+ zd?)ndqww$G51cw-1i$kANlflE=Tpoi`L~`uC}LHMhc{?aYm3p8UG$K%9xKcveGftZ zsS{+jT`Ks7WRhMlA9zQzApY6|5GUpEChihUW&+_ySSh@-DS!*xtC_i?&den93}kRD zKfmW6$_2f`y^Wu+X`VDMn^A!+R}IlS|0Mm?d=-!R37n{&i`@1XR@{e}MKz3^8kCP3 zgtVLT1;pKOS^tB$*h9PJ(-a3BomTv5C}5d3+L}jLBa1W)OltQq*~2^T#A2N;!3}UNvn0IhB#9BsE(Yy6TaoPzk z%wjaxF7bjmt*NB->yMJ}8r#X6$X1&IZ%L?Z-4B=5o2GAjUMINOcaso<3L39}9Fk3! zz;Y=9>t3(n%vJ>BERPU;7J3tpi!2ft2JWb^m8BD=p2YXHFX{Qjxp-UhCOtj;mD_tm znYzp>p}xZkI23gdo6Q;pCcFX`kgcc?a2L~0f5wLq4{^prf=y6|nF2TaT;eGHQurYZ ze|HQI%e(>q@)s~?eH*AuH)icM#aX@d`Ru6(Z?@>DD7*KCI9u>Tfvx$J0-1gK^w!W@ z4Ct}sSuKVS9qYt@g+=_6N`F50cNqWfoEvX(Ie^zT0Y1Z3k-v~|9W`T8h^M(Zc(8jR z>i#uQU7^n&U8v6{rUN_BG>tvsBEzPClwlWcECf@KgU}W$T#ApedlT@Xm2-H%WBd5A zCGot@%?#en<~Z+VUCnoH&F6(BG=6Y>B(E90lAkzY6rUz`4i`@e$1^3Z)P2fH=uKB* zEuzM-ZvsZL*B0!77QnKe%C|I?@UPY<^X?JL#GCaCh4@EmRLS%p%j96#}H}%pOtLF36>xz&;tb2kX zd-O4DWDF`U^TOUZX_U#Az!`74&}pQQ0W}(hul7&J(vyMo7(d-+(S!_gURf3PS?Pl1 zh$M(y&44wpN$JKn+R# zYJ%$ZWANgGLh>Na5L817V8qVlB+pn6X15l?(Y3}9v9|XA_dGvT@1)fuDLM6C3ilW8 z!k8psUu9<*dVf5R`?rlh@Sk zdMa7!Buam4=s@U*%cSfj3&c$s?y3rIraP{LysRL3XU4&d6GzFfbb)zcS3{PZ6(tK| zl;QK89HKijHjh38GrVdT9QU&Er7pYO>AzHq)gT5MVinCXpr+@do zV!UlmkS*tyV$ib#csS}39U)yn+uB#-$hH5W?Tg!tWtcH;_z+j~V|z3)`;tHx&ic+( zPuN7SM?a^Ovu2JujfG!#6=B%q2u-Tbq8*Q3(cZuqM(pf4n@Zay6t5hkPkb8a7r!ys z@HPT|*KdQJ^()DVlzJGPqIe1C2azr;V1wS%FQj+#YB?Oa2>hNe(PU^Xb{RH3iAG|H9xqQ!)b48L** zt}gMU0i~X#_MS1-ZuO^c`{QZMw`807_?YzQY=Fh4L2%{c8`6JqHFL-KJ&4);2TPU5 z!~SWSpfW!d^7s@eU$Pw}<(r`Bl{!0S+#qbcC(GuHiv+g^nIOhE!l5t&a5$Y!iq|UR zkr$ox(=rKe_1AoA@#i*i^qzqU`Z~C$cQ2+{p2v}T$FZeZ;K4{aa4~C_(4T@6>eahT zjE-0bvsCRJnXD--Fas&+^*v7h1!@45)qvZoL1fTv5?t8m1~L0b!l8EIp3uFJHC-qTvx4B?YObol~h|;)PXGe6FqTe(KcN9xr{65mBTL+^649E zaT@A&nFQHCW`;9)ruOP0sMjk3@NIuGYZqto<8Fb#+LApfcTJ_u3!}ReJ z!FIOiGM&Crf(v}8Ot!4hhDCcMVUqk9Xz+I?dVXWL#bxO=3&dF*`=1dO?7K`qZy$rW zdk4``2_yXvJ{Vlwl3CsPvYM?`OOWcMT$I8#?Z!e^Bcako^P{jzEhuJ(X586PFN z@u$e%4GYQQKef!>smtgQ(Hb&PR7*3g9O>#dpPI1W9QAoqMxsQfKrA1@biEg4Oa-^q zkuSsK=X#b*m{i6YdK#1b5ijX1ozsl&>lL&m(2M?HU(jXYV)%E%7Vc|70zJIv4>Rp^ zJvrhN!R1~5&0MiN#f^4p(S}qyW_S{b%i1bKTE-FjSSne zTZO%!FUfu!+zxfZ!>mpGbLpx4(fglJ#$qo&}6F~8nBl>2e2bXEN0C@ ztk`?rc5Ltx3-;a|6LzKLD{}shBi@udhR4SB;8)Svy!se{k=yLUx4>ck*JBUf(fU9B z1#=e@7Rm5+pZk#WPehNmY52%`9Zr2|0zn_O*a^!l*vFR^vrkM9W;IC)d-BPjalt+nd`?Dn05~j0r_WuMIk2n~7 z7700JYUI|I^|xy}gYY3s=y>eLJOo$pn(~gsmac;N z7vB(*CKuQ)C-@XyKN16F!SUPEMtl#45Z|%cr0(4%;@j3jR+MCtLprC4dFc}poA+r4|~bGxV9wS>=f zsOtfdQmG&v!gt4T!6nYepn}5@Fr&$-1;631;usn zacVg{C@cZxy;=};&IL|yEC3f@cbNA5F{yf-M3%h%M;iCa!Pyco`XXc!GxzE{^6peD z^?dS$dKh_NQIR?t_dTb-Q$+*@iw^c2^}tvBc?`N2hfR_O>gO41%tP}g zDrG*ImRj$ldrwb?>tVsn*XR&3Mt&tJ)gBK<;=1HkT~5u=nkDqV4u)IkpGtDpDv{vS zDB@+el^I^=58MV@fuS7**J`#g&x}u1|In@_qvM_6tLoL7M;o%>XYB%rsyPMeg&)X@ z*j3PxG(fby?ZJ3yEEx!RZ4+WYNK9|+V?4J{AUo1V!L8U_QuFRC!Apn_s#?f!`A-tNb1E24S^#ju7nTUQ@k#fNfz%FrNHPzEM?VgLwtOtqyyU@A)(iaJ z+Q6)@Dd6&W5xjjj6Ix3e$vMBfL@y(XOdg{kaC1kI1S1*pZoj8PL0Ji$mPOUjJTXK8tTj9t9(T)amb~6+r$u7FU5>c zdZ?jtgH{I#9<~4IC00lV1KY@^Z)cd6JYN#3l|tfHcakL|Eg@#~96>$5 z3{sD33y#Z4Bq2!#o_*X%JeKxy54P&T-gUZgL~b&iKyx@Iyr=a6UF2tK6&a@yMAZw$ zF+sx)4LhUpPKXPR*&%pOcO9dOZ!FM$nJaT)SfAPFA4>LS8PL(!qN&`3-=sCgjMS}P zK#G}5%zcURv}MSgCj4cq`$rxZxNtUPt#cz?erP3b^sdA`ema;Cql~wdBystp9W`C8 zw`hk`DBZPH3Js@=V37^MP3rPB20!+bni6xoxyAymC3Nv^p*YNz4d-6`NTllw-_jo~ zLv+Xc(Ga}AguWDcjz7XOaoh4vTKs7=>fKlj>sLObX)`+LLcWI^t6R$a_ck6aZf-~4 zHQF}Ldsa|`kh^sGBxO9so+OhR7DG?zEoS9xee&zsa`xr%LIpGEeFT8Cy%sg_fAZasn@M+aZ*wDC;$Vl~5 znY^oX?ZaJYXLt&I8e*~V;SKB$=)~q_<4~eA9WRG}pvi~6qJ@1IE?rcG1!D@(`hYw= z8=!#F>zBi=?Gw@0WEAvd=0L^UNg!+9OyoL0lHRGk-0L4VoXeBTwUu=L&<%{g_Leps@Whlegp+S}M){bd_}|PExc~Pe`elS6)JaLg z>~))nMvgc9s)+=9lW3?*ZHHwhx}aq_j+kcWK~rN7;Ue=%-uC&}KGqu7C70Kz@9&~H z!u;26`n8&{l!>$~Qw5Ub7QwXVtKsLoIk3qoo48$5w3#vRfE;KIg(1%~B=f>avXk3S zHZEUDC&egGy)WCz`wfPa{oBIbsVyS9!}8cH=Sc4_^P`=AyO;>0HT2vLE8N{bhtxaf zQq}WHb$^t!Mn!%HrkFHahaA6o%cmz!$^IaE@6d)8-pa zWd(2gC0C9ZPm+Nfu?(zs=_gCGH^VreDPZ~jJGogF0o`u@fw)0Bh#rcClUI&GQQ~Rv zD(HYP+cX%|x(ug75iK0fRf!;7nYG0B zKo|MGHlBRCvJ)0IWWddFA`q*kM-;B@fs5->;8bfaNSwO>BZjTvfk`056~_}}As75> z_cq9Xc$V1{^P3xNrgXox9p2k7$#orhKy2gEh-A=2A#?0TpUhK(o(gTcNXHREVur}y zOcgwN{}8SC9!Klr?o-()2f??r4_~afi7$tnP^WDw#;(t&_el(W5qFsWH{Fvyaj7P& zCI4Pp4>#$g4KBE}YXipKI%Cs*rH$6x3N5Xd+O)JVlg^p&AMTM~jD2J7&~Yay@y8={ zX}uLLI(eIlUi(h!{{0|^)v~aeuY}-`T*x&WBGc@gI3H97Ybg!7?4cAEt&>O9v?Um9 znU6==RT%f`4{i%WRBt$p@yc?<@%bm(tENvUZCTGfy}OCNdoh+2PMgZ@V10@IT644z zcCsc4+#<89LeJY`2X6PQr}k%MvARACgLV7qGT+B|+-nBn)QM(s6kUsK;k(T=3{DMl5l~ti=u(eWnnn#2vxspHJed@SXVhgAsnT zAAvfB*U0IFH87>c2D&CnCePrC$$P1}TpPWVGYMTQ zpVB2R-|2tp+0?=NJu@V2#zc#R)r>ijL!^SnbFcM=n1ds)+w}HJ)W} z;tbHkxw8W(45?a-@S1`XVA z($H-UoCg1q9P){Q=R)@U#ta@({9Z#{;W*ZN^h9>kz#?GcVRTelWfJ<~ARBo9|vpGQT*K-?@ajE3|Wl>8IRi9Q&Gx0dgtx?c$Q*L4z2 zeb~(Xh`3R6=CvaopCn`^o7-q{gE^MAq~q^0iuGe#(0!W@Zl+Va;2#4%^!s0| z9@OL)bjb-kpMMz9ei@m@LQE)WL%+gql#nSye(OVwyIp|(w-4dwtP`l7GL=63BStKV zEGF-3pbcB&@b#xe?9i;n*&45~VyYsKdw$}+lP}P8NeA|Q9?9?h@E%<)MR`vic%v6G zyl&7RTs65HO-+i0UFICDHP%CRjV~RSZwDo33SeGXAIyr61iz?n@aEQ6D08?4Z|03= zn>)*4e|IQ2<>$fUBcs6NR31s3*g?Z53g*5XH$h-Ck#7^??G`1DYBacJT_bNE| zq82&?x7Qu5$2Gw^hiTf@Xj*ykBW;VzqL;F^bN7qfnV_|+z~43-#QT$Ab)_Wh6!{Un z_nw8&H~Eln90yxk1wUeO712#ONXA^2fZ6lzku6151 z*9}tc;MJAP#@cT-sS3FQ$7%(x#@WPk@?V=BkprYAR(M_ZyUhZDh4=KD40W#>g)tvq zQuT^q8fd76chmk+Wv-cC_`L%+-&jiK&uXF%!a3VHV(627~-G(UkS~7L8l8fLdlUq+EXim@+e6&8U&VyQWC8kHB4UEuOX{q z?h%GehravLa97B+RB2s<&*AZqWOo{px8H-O+Tv`M*Dy$LX#=kX5^PKIU+`~u0Amk4 zfFmoiAiDPoT-B+CJw{rv{QT7#$+~(HpKO8+jvBc0W;}{DJD{hYHHL0X!X*8_IQdl# zcD}xko6fwZ#+Sd-T7!KwK~@4vwZD=|NkeGdVj%G8uaj%tE}&2;4SUn~1#j>%!PThx{Qj`pBC&e<~h4U1_||17jIn2!f!VyWf+8&uas z0c@o|kcq2QU_+S>gomaGb%?-{fBTtqxGO>7%6hmi*$*CHa==_U7Fv#9gJi)adv?1Q zRfy2Tw%^Yv`&5lS{TjsG`1OEV_DrIEZ4vaO8%Jf-C_U)sLf36m!|(SP63`q?(yhxF zDOoATxUrkLq-hPgUJuE=+leG^UJCr{%!ZG5PLimlC#dfwVDyr72)9EDW(k~}bzfKE z2i65&9V{03@1yXZ$xd7ou?#Z}c3|=fM5pbiaoM{f>6!!cN#VVtb*1czkey!h5mcm?#bhep%XtD)iD( zNA($VsS+D8(^k5W$)odD&!aO{gqlc$5^sw@8a8-^gwbpgQ=UzNz9>UKOoQZSCh*4b zGWn(@4G{@PV7vY>(9m2cRr&;rtcyVM%pj2x?u#u~+DQ8vW2jC2M&O?z20pjML;ed< zYlkK3$~?mS(6tz^>VlaEZV3BNA86X+&-B_^jvm|-jLixPm{`e?(m^p;cJmi;P4k1T zn!?ZOXXIeUR7@)|rF(vjL$CYuP%m%|_Q~g?y=_(M$5h>Ki#ExU|Hw4>C<3tEi5u0^V|7LuugvT_8RVpGDuLM7x)A{*&<0N-<1d zT8M|NzEihKGi>PYrPCq|;mNiIBy`dO@^0UFoG3FNN0gpTm3=~(_Ci?`u^z;)1gBgR@ky#~X^ z!v6WJS{Q$Q1$jK?Bb{!$0vG0bVdI-6IAG6WV^$^>CUoFSl;v5CvHVR|kH6D5g4cQY z087&(c*XoTsIaIPm9Pm{&P_!V@dIceb_|D(Oe@wQ4;0=Ar)C>A35qF=7=*v}c}#nW}{_$m+2xYuaE*^LqA3&lq-T z^>~&qR%S=Px($2g3C~G``Q&D>32GKc(K$~G@Xd*O?A*8uo0TFlcvCzYzdMTmLQa$B zw7Vp^Gn#A~kc2%#)_oklgr~XRz;n`HxN~F%yVF~hZU1j9>-Yr0|HUa-@M|5s*rLvU z53c~(aq8?Zn8Z$Y`6Kv9++p&SE1cMlc*vWSODYXK@%wXI@>eH-gayunG7nWqJ!uGD z9)?h#ss@smEvac@0%<-J1o8Yq&``PujoHnhGw>E-^axwG!h+R&>&R+P(PC{^>a+6E zz3^eQ4r?Idz(&q}LBRO~kHq(jR*UN!KX;UJc#Pjgee<$=$J_@dr zInem=2X+O=L5_JP@Dlp)IbQ)z3ps%z?@nT|F1_-S&tD>Qt&WolG{RBygt^wz%d{YJ zF-{v67(NCrxMr&kCZ!96)l;MXcb&IO2|4 ztw*_4X%Fdkb`jbZi(uNA)0ii@8>N(IV(A}oY6{u(M(PFRFN)#1z6IoC=~$d5qKj8< zsA1N+|1kcv0ajo07IrDW(U&G~h*+OHj2_=l)&{o|mq8_XscA-}&t-7(Ngt^48yozw z;VQMcxdwj@{UW!;Y(e9uG7M$=0azzNn_&@L+~Q0IV{G6F*Gjhbt8;lCTWfAj@FY$1 z|0DUacSzKqDqOmj#>qZi$Y=Y^>D6twiX`0vWYpaT*ye8kE-tba)N2n zf6PsrJwPjC+=U)tGJUmZ9PwZyh=uGN;yQyXx~*9x3$ zK7eh~JkF1)LOstjcwU%M*&fQri!BLwW=T8FpQ3=#*RU}yza-59U|Bvkaj{_^6 zUfAI(&ZfmRL+h~~2=?V5DlZS@Qo@DVe<$o$I|xs#3JJG%IkD16W^x~dqiE6uyl~SF zUEYSGtPHN__e|i|{6vqPN_?7%9KY^zFTOt=j4QiCP-E*eJZY4V zs|Ft92!TtOU$hrzLse;svZr8;kGD=Hjy(z0}e39XBgKg^Z7=V0`W;k?4ZuFz71};}1@T zqk|SORZ$5Nh2D{-SS0k_+5oqMjiJI@@a`xss5V~efb30i-07H$U2FSkYiu^gXN={8 z4FqSGa}^%o4&cyGDxOokixM02a7lMK?wWEKmHtKH)#=Wtl5LE^I)aZRVh&m_-+|hv z1%|+m30U>M2sibfMDAh%UTA5;Yo+&bUBN{3w1~s_PaJjzy+l=$I`rQwxNXKs@yFEE zc!hxy3@NyXb38S$naxDIjAJM-@S0wkIAdsQ15PMt!<1i2IILQVzl*c+m`Xa1-5-q+ zt;-38BRE>jRC54pRJ z!<7Bk;7#it*fgOOBD9kr*6BRyGZbp>AxF@=+f6T?$)OKktfpzPvrx7#PUtH~V^r5> z%=WoNH6FyGSt5f0;#%C^4|1R<`-t4V_KjKeua)R-6M+SWhk$8!h7V4{JZFvtSg22d zKkIX;lB*_gg1>j(n*{3E!r~1n!404@0>7-Afw~_b;peU@yj3dfYA1=~yqq4!c=3X1&)x$ zA3Dm|kk0*+MBTU4(`?fqk|pa?&~3O+a53xwFB}d z847EnK-Z!YJZ4F=WA~TBlvWc5q&J73PNk}_L%&dp+bz!i0vvBXV2nTP?wGddC3R~<2hpjO8T<07M3f7At zZQ395{q;9E@}v%yc3yz+bWt|qPA~ZF6k~6ls1uk0J|JzD2g#R?L##K2ZvHiyt)>Sq z*%M%OSYFLcj?%v&hiUKC6lNDusJ0ezVhhJ}G}p)nwWED;QE3GBY_i4J>B*=#|2^(A zj=|5?+h`Y;ly?ddNlKs%;YsvN zauQdun$}dUSwRP`=Fx@g$DwTCd^|F73AT@$hFe8M(8XSZ-dQuisq82uVha^WxZV*= zEKo)L3c+z;nnr)j4aAKrO30zTcWJ1eE)i3ch7ChALAiMw^rfGG!1j5NEzH(lCu|mW z0Y%{RA2F);R~v`+titDpsklJyDY`b@K#4UY_`V-MgkIz${FkK*7xVtwIQR)XuUpFO zGK1;tJqdeu-0GR^O!HB!oZ1A|%H;!nvP=rXkc75)aHcxN3>?-JuTsp{ggy9demIUyvPjfcO^ z%V1m5TyQVZhn*?zaJ>M?WviDo?KH*2fjK;Ta4X-~GmrncZZH2vaXn9`$ME+vqWHD< zo%u7T<@rxZ7w~$qB3>S$OKeUU!wUCasByRqW$vczpYg7&>R3}2($}(I&dRY0L2B%8 z_i60d&#f@``fO6y7C@s0ZWANVO|&+;6ywwVv5}s@A#F+CWx{>jY&V&g{wBfO?Ei{4 zg!x$TjY2N!a2H|zra|cEBp7vm8GG;r%ZgjCU`5qduv?;i*h3*}*p;<%>?g|?;HV(Y zW-jf6R};fvSqBT!k=@)P`_t6q@<~oXO9XS}R^n6LEx6?VX;j@EhT9IAlJ>E-$C~)eUS@p*@>uXw3#TmP29UcknWi6*8t-kU8EQ zJ!4X9@XS_ceSM@6S&*59Foar}ES8P2q1&Y{vYq=h$Krf$d2i zn7b(uVc|?ndLu#S=OOWvZ>aW4E+n!;`9v`$02YKGWcI2;X6zogQ!o;Cu4o_+61Rfh zsq0Wtl?EG5FM|Y?WpH_&1$_AE&P3d$)V5y)*S z-$A$9|K;8Y^XA%z9wcHlPtuY?;PU0OaOP||8RHwwv^V9F)M6)mJzP&y#2?b{9RqZ? ze*_IQ>Y$}dXCYoQ!h{4T{Of-Xuicq~!{Ps^=f?lgR4M`!ix%Omur8FfN@GC<@Xoi$U2epS)NP+-Udzimu;`+4TngDf&;?Mu?yc3fcTT9B+}WArTh zNoBmBF=rQAQjv)fHV(})~|EF%Fo_F!DuY7UWF}Aa5OlA!z@0=oWuNRL(!AmL315>x%Rf3E?xm zWM2Sbu7{G9$N$j1T2|2QkVBd}t;xyq(JL_Oioc;;FJ#GK88(e*oE z$SQ<{tegtI>aOs`b_%FP-XIAlw&2nHd^+~vKD@VanBIub|0|eKXxec^=b-yX5U-vG*jkf``_W%E$SraveXtT8#I*vxww0 ze|lhcAnEUT$i3-Y1y92-K)z2CXZBGAjsE_iVY~qrDhOv~;Wlh;xQ{+>M)TPJ3@0~> z^RXw}aCBBEUixzdtu7Sd{L1q*!EBQA+zPVZzk|8UA2U%_f@>ya1HAUy0!Cu* zpy}~Rusx*2UXYMxr)6<)t$8x%AQl7T=NOPcLwoM(yF|gyD#h=z?Zlh4-_dww8nS8X zc*J!x9WXsYvr@}AIr&3qek&b^_h#TJDg=$K%?19jvhe%j_^5h6$|}U7t-xrkpx^0N zISX328W`t2TL`mj9b7zq72XDPLAA#q>=-qIJ%2)hP0Bn5&rPzRP2&<+t@9#DuSzJr zvY&*Q7myEsv+1i*%4owM!v%h6sP#JxLzZXLz9;KyjLlk^al1{pVuMur{GB%S^ys3w zeK)Dn*^~6W;xyXcJ&k#07fOeV{!-`AOk#Pvi+CC)5H_q8RctiSZL>exg+}7npT2lM zPlezc$6<$WG@K4v3qMCphqtwXr0K*Y>U6@9F5HvEIr&etncvEBfqHSIYGe^nep*U; z4G8I9@PG&xRC3pJGdU~t#a53xNOo=!1@HE=Bvf}a-2C7M@&P7r$V_<8b{~Rw(wT6% z{xp33rUtFTCNx^c!TxY2inTD=owrw78**1b-LYGkpvynt2NI$Y?xy zax1ERen9u%5vTj~+-yXb7Z6{E1Zpd>+@`m2CfWH#4&MKafhD_rA*O95MyzXI; zmZ>IJa?D`!f?TM(xgAP_e~|fU3LJjf0;jq)$UDhK<{Q748qU5>%dWhjgD!ihYfT)^ z?dzeRYOd29qc{{feG%oCKuMz*Mdu@u=5<%w$(u|z2j8z%S5W%x(@v<7E$97 zW>lxTn2ea1$;kYAMmoqe654xSgp}(^fjB<`alC~CGvtt7L+RlVctLp>@gL4og>@r(A43nV=U&)t1AJTO_ zkSb5DW`4Ns$Emy@^*>WdrT@&Lof>(Juk1uvEaF0>S1J&`iroRwbit1}JXpkB-e>}{Hmjke^9KoCJsAc@ zM}o&95y%*bCV79|;Yrm!^3S&d=9xKyW?KsB5dI%yc1rmtr9WCi&kS3nPm-RFixzmnisR~vJmURoUblX}fg z|A(O=>|^%ib$&UaV};>5xuKfK^+h7q{+QP|5{0e zI=*b9$xlo%MNQI(6Y8qsizr+L!8RB@zieNznd)?U4~#J+NmJ(?AB-|JV{ zZq2BeY+h60F6&nDqVQnFEPj2(h_^vy*iPNbE-{tL(NmvRw48rYk*&M3qDWh!B2dt$ zW3)><{=I9)l1p9q8+|hP!FQebA^U^)i!56C7sJE&+v=wA`3t+)q8-`%#u=)7?f1g` z{Wpe~uE83<%}j}FQ730s6f5tqkkEFj_+YxZLMk(>!lF93!c}TZ#R-+RYhT|tmufvL z!S>2!{D&L-aZ_^-E6SPm@#UE0?(N<2#J`mlY*sF;wDq@>{P|F|@SsL<-|UK=*IEzJ7qNjA@+ftmHC6I9UTT;G~;>Q=3g_1+ju=Q#;V?1t=X)@1fQV-{QUJrA~r zs$g1=3dZrev*`SQdG-|>#>(SUUOCH-Arr@^jVeqX~K`Bu!+$13n zZ+%We)|IKKDcOnXp$~E1$`*Vna}zEu*TH5yv(gY5YgH zJ3JbVUmKu|uo+^{R1Ef*&Np7Nn7>!oov+sY6MxIAGV4XwZ1UT*APkoA7Y|E?g%TirY>YGOxy&EYUTDHSwb$e`6Q~^#y``pb$>r0ZmNU`%Q#4K!$$*5H$02rHHjFi$i3hyh zuqHnoJs2PRBcky2y*s!j*$XX>FUQ7bL3lR06nn3~K)2K}?1KFmY&p)u?>%9-!hIWl zn7R%J-!H?D>yz+T)eUsod>4CeJi!atjyLRD@k@{^_d56=2``DH6+8VgSRgj)d94;@ z{A-DL_EECW0I5IUlo?wXvz2!(1^N^+*m80MiVs9$;I!Kq)i;1!pO0cz_5P^8rU11j z%P_V2LU5V<672dFP$XL!KTe#9sk{U{QS}=dvvg7Y^D^}QDTbYCdKmI=1wPhT1q)}S zz!_;t{Mu+rPu^NU7m3cIGL~zR6OX_%uVQfCg83M(xdo4&NX8zKAc)?O1S4TT>7hVF z*s~@BF5lM2o;TCk;ig6Gi}rH%VL>o05UQfbOBc|^OIvWu=^Fg3GJ?H_gYdUk1Qz%u zp#NBLJU?qZj#LY3Q12!|QrBUaTssKI_y3_I?`@%aOB%!_4TJO5FR(Is6u$R=NLBOa zgRuKnsFHQ#D~DOd_L$%w2)>u*v2k6C*yg{>+0oe& z*yVB%YemkX^n*vJeo&323i$fgv*kglYZe&YJq;QTN%*z!5h^-MF_Rgwxcf^i%10c< z&O>|guN!flBdTqPtIpK`#o6LuSYP_WQonW9@tl^30KZ7g@hJ&!MQzSx_XleB+gm@-z@q- z$4CwJv@Fm=OwSG>U{}#eRGg@x;lfA zx-R~0cf|=UYf$ue6x2`rM@#iZVe2DL+6!x_zj8F)S?7wc+=KD!=X~7JaRC=v39%p< z2No?{fE&Jbe?rr!^xQ*$2_|68J|FN0H0!}zL zjh%~F$h>pXF!4hnJ`=fzQ|IL3QeGQ+;4q%NT|gBbFu{!M_KCaoZLR zrn$+UiLctq47JD7tB0&;rC|Ww=b}XXWJZz0`;|%1C@(0H_XCGzJ3u@!k^X8kq47^j z=vAG&@F-seE63>K>-u0gX8H&^Jr3Z@VsB_TRSIKHUxlvITI|EEWlZy-KPyPxz?gU_ zQwd08D@T>$tfc1{q$|OG1Vp2>b}Y_%b`rO5RmVQW?8a|z4%4|N&WxM=5SXR*q zxVH8*bYdTjS(wM<-EXlTk6|_((GQOgSYW0#U}fEUkSnW!7@M&uRlXkY-VejF{6fKZ z-bN2izDJumX=t4n3BoHbz^k(jKz<3av%QimZCHqn`k)Fj3EnVm{95=j#{?h5hhTlu zE8M@5K+gV2Qo z#XqD!r5orQ8w*s8-i*p_3HW*BD={w>rRnuE>AcYdUH^FFa9j}P=+uFm+X!$na(Lb0 zJFan+Wfo^tSYlxl81_AZneL*vK31G+*;>-rs5;V|yqAs?4bZ+FSK&G}!*&NF^bp>O zeMUiW_<u^!M8~l3uNdEZ@GUz^rJ}76wcikAKkgv;*Y-y(2%WEn41<_*~w~>kp zvv1iVOk{00%rut6oyk*h>4%wqy!Cfc0r!`IWV&o!bxYBpsqzX z{0S1mpSOnKzQ8X`F+T>`66UZ858>V$0XQkl6-CDA!Lo?UFmm`U#0{>-{P7F0<+3%t zc4Xcv*u1Y7SpClpNveTvC9wA&&$W5!cP;t zsJIJ#epKSoNl$R~Tv28x6p3D9S5Z$p2frGIv6*^N%sw}Xt>HBI&lYL$)pXSP5M4-8 zsz%f5ojK%vsxZdPH^RCn1{g8bgf%#rvGZgGn|{C;&2-Fhld?T3SWW@0OS|C9cz>uW zNQA8pDUfO(3Zi?A1Rb??aoJKeT-_&(PR>&U<$iEDjubKlW}>{ zd0dft26AJsL45Z&u$bL}>l?~2W~c&lhP>F#Sw>EvVek?^ul|hp%Ti$IO$w;}t%YxjJXB0zD6)7ePLtN) zC!f{lzdm5X&p)il8f6Vxubw^&I^%+YWjj&ED+Pns$gte=%B(iejJf&hfc$7pNX{Ve z)p0AH*%5*R$2Op6{v3A1#*5h`uVcchjWlM{7?@!l05_ff5QX&#qQ{IW-x) zubu-%o-~_qX)KGD*I-$54blCvC1SKSo?mtf_sS(AjLE}YB{T5t$)$K8Y%Z3G{A2}7 z1ZNtS3-ccZ?_sxY`Z7P$UF_G)N>oz2j_?26#?sylcrV-tvUmSN%m98=EPypDp8Eg8Z2Uu;AQN_#{#S+6}dEvFWNn{`D@lYkonmzqc`Trx=P? zsNsUGCirHW44OSq75KLUXe<^0#t;S(yA#1fE*S?$rQ@9RR6%^`#-;`?X4+R>n8mYj z)Lb2e-AWO-Fj#|4H&b4;-`dyCCM*rs57V)aGX)1v5pw zU2cqbw+Px}1v#UdwI^1)AHk`ImVj)-K6qVv6wFl(nD2-2?1QrbJDy|2gaf9q`SYCF zG@)?1!lsE{bgHC%r*2}?)2Ha}FU&%`cA(wVBZ3_L7=HNW!mfqPVoUDLW3R`i;iHly z47?SK2ZQSAZksU>?y?;$cPGKQgGXW9NG-fhjex|;JTTcdA0+-J%F z^{);1F+2$Go#x>eS4SL?)Ij0G9&EFwC-c?a#}0Vuv+uViv7|d@Y{tny_%3S{D>fZQ zZ~ez$IJO#=E$x7#fnB(S{{tmre&d0%m8kk-Bf3HyzP`Q(=Orbe?d~+ReWOFGvP-xr z#Vg4R#Y#F(XFX~4^CbG?9q7B)!KB*JmU}<>GmTEyrzL+p$!EuSIPk@rCcjo9bCq>S z-u5Bh>fe{R-#d{jZ%?txybavXU)JD|JWOVvn!}w4lhu91s=Cosj zxerHI5&Lf~JmK{-xwHj!@NumuOwTw+E9z6p@%64;+7u5up?o{cN}5T-7N~F^p$0l0 zErZz$4M990oQ(3-=S*(KapoDm#D(PXinGRWk0$0q*;fskdcl)E-n|0d*IfYJml{-T zQ4MV^%%rn#mU7>wuBIt-tBKlyBJQxS2;O|}4@ufeaHC!sw*8Z*lOknFm47PCpttFx z%w(Fa)Yc?88*3H82 zyvHEC%^mI>38GcUMQKGuB2kK(j`Qp9!n(FY;P`F=+$}4hy|0eYhd!%__4;()^V(!y z)z)gN)DcUBQWtTXf|PL6z;@t*ZjtZfmVwO4C^}B{3TYR!gf44sQha?h9c_Msj#TK7 zso#~jQ*Zp~cI$Mar!7vmKYT`V>}|QLYOA=%g-TfAatvba#2{jhJ8|l>=iMvMhY2AumvLn5eTl!}U*BGvV+h(tWus!;4|I`lp z&Wz8U5H1zbSs4v?4=Pcjc_ zbF(E3AfRgk{bC(XA?PXyv;si#U=f|yVotqV>$oTW%3S|5L2RbxM@{(isf1($jJ@1K zZ)+dnf(?7Yr$-ZFTRX@^&Ww!SxQk=U8abz)Xb8?452m~Tx;{>xTJ3sg?dmj4HcTy{ zY8sQs_vO~)-j?IkFl#eW+%}pV1Uv4L!xrn#ghkf(-A2<>E;*d|pf1T>@SO}rF6L z$b^e>ljS+ywWhZWrAe#rBzpYTAlcTM%sro|%Poz(L+WLFxzm2h+y=3uoaM1L-WY>c zPTOA%hNZ?(9CL>=d>qh z(3$Ku&Ed^5xJ2TGgRwT|8$9ni3iE{T(@Qy7oX9bIIw$=xtr(2s>|<}!I@w0@ZzO|L zwh`uf7gp1m!*RrWeIxx5F#&d!l~C1ZUUX}lGSsAK(el@F|I3kbcnzMxXTuiTHN- z7bS|{!s8%K^A9XJXN+UxB%oLR86_u#A+~J{eAGSyMgROEzLmnlk{~duyAD^KV@b-1 zHAFpPEwK#T19cl}!1w1PurcaF_YITSLcn%OKh4-Y6Xa^;Vg9jyn2}P?~Vo&(JLV2QUepb`(Z_(IzBpf zndkS_#MM#J!j)_%LQ2;R(B?+(#;!w zP?6>4p!UKNQbki?Uq=IfqToX`jJ5fvP+@ls~%c+P<+#s3s>8k;l!L4*fi`(j^tTUW4UwG zx5<|V?aiSD+M{8^>A&2ZpnF`QPYDS-8AzoMSkXo$5&H4?J#ym8Yck`9GOend2(Lw( z>E*%}ni;bPD*Zmg_1gxRb=wS1A58#BUorHLDC4}s6}g!fzT`y2T@usWKnxNZNp49p z@63LAE+>5{uU{t~?jH^Wt*o=4K1~F^794<4qDNt;iUD2|#L>r9)}lw~dG6+f$(%#? zC~~j$7&Yd}(A=W$)ca{YtsE$(reAd+c94=&s7ccUf6^9jcW^$O0x<45>@!uy72&S9 zr0+P!ejkl1%2uF3;(9bsTZ-0CA~4}}3=XRx)st$b4|{BY2x7V&T}z;B5d+!RQQY;6 z6NJn0qzYf7>Cls#)Mt?v99@tD*S2lO_tL;O*?C{sW0z%5*Epc-KL;{WGUW)Pf;rL$5Vc^`XKy)UoWKC&aBO;L<18 zkOMy7Abmp;5t^V3R=z2u^==rv_`4QA-x>#GwJ0=IGSrJ!rWcF+sHfcts4p-g8S#cV zteb}q=LqnH@1~r-(Ku@58Hac0Z9>L-j7NW~(d<$BuxZ;4n0tQE>Wirn8!DT&lrn?@Rukhjc@etO&s zo%{01t~HauKY9#Wr*gsT%X@093{_ZYYc|T(z)K`x!fk`Q7EH5 zFNWz>JAouhUI?joSKHGNp6;KZzU4gc-GCK2 zG~N*A-?t^NZL;B-(haB!;qZh_9-ixNAux231PsN((WNC+PhE>@$7zAvBT-bd9i~tH z&0xyt_n`Jkl;^ZZf)2-RLWiP9)WskZ6yw6SeKfXXd$seMsv+_YZPl&E)H2^X7@6dhv zAMtxtO7Cn@#RZFku`}le?@ER?t?#YC00T9+DLf3v$_lMT&D}uCIuA~-t>OGvN`noK z!!dKj$cuhudY78Q)5;ZOpUepe`|Kpp;MxJxUoMAx^LOI-OXH~yznnZ$-;O%#XLIk@ z5^l#q1z7m`4Eea85zA0vNQrZ!!ODSLK=gImduJU!agD`={rwnVXalpSb#g|cXKBNH zA#S9-i=3BO0z37hsCb$Z)sor+amwjb-!hBzx>})y`UA2n^e6o?&K|cpL-+kk z=L|Bh(4qZ$v|AulZP%&6%2oC0;d6$>;j1wCPOT+hABAy0*C>HgRvdnQ*^JNLsIXAu z`^f9pWv8}mXYJhw>G+3b)KD*!&hjat3YVH_$LmsRrW($DV5_-vc`wPY{=<~{7(sPQ z2%KMfhu%Il3y${Bfyz+}$Y#Y8M7P40iaJcEI^(ry>W5&u^Fu7~zkPv%eRtv5Z-#Fo zZex7LR90oS1NfIuL)-GNaBgJ{6gjG+c=|Lnlvqe6?|()&wkAN`s&{yxPK=2xFlN!u z+sVo%BifS?L6fHFL6&_0q#TZhnxyYIH_DA2^}o#8FQ(Bz%Vv62J_;Q563IdD`9w+i z7n%NGJ6)=v2`L4iDJOdc9-JAAk~=++>vE*DZCeh)Z#tbvC+(;-y;6KzR6 zPfJ^d=+uRAWSd9_c?yAaI8g*sXHUXfeOuHJVBVFJcCsXyTxP9mflpP|VaxoZsFke^ z8p>wSeCRMd7S6|YTBT^wQHjyd=HeN@0K6r992LWrAY48Y4zBM8y;XAfOV1PMUb~5U zmlx2z_<&f{R?$v@&+b@9P$@qTue)9Wve*|}-8yk-WPtqWKS)}JUy$Cqb9A@G2dX7{ z8YHF0(Mx8A)NafmF*#ihZlaICc$E$wDz}2S?=C=;z`phD)?miM$JmF?zfA3PEYaO8 zMe}J3O+HWnE=y&Q-dl?COyj7WQ##riV8~{&7vOq-34o0N?AlCK>H#FLn#I+02=(4p?pOy}%N;}}bRUW7; z`UlqgwQ$LJfi2zDN0-BBSSRNSIJXpR?B2t86HU4!ZavNInN4@-dErU7v$*P_A``#S zPAesO@P1!0bQPb6=NCqy)9HEW@^KF|k52@XE(yeY&17!Eb5d(z2Rutv6s@ws4GvLg zShtKmW!LFLgSlWAEec;0P2p>YFx*U$rv^WE(cjPVsCBY81dX``NsBWe?%W!5{ZNCq zyvMUexqHF(@-xWH<8W}&e3D*SL;5ydB}=BI5S_(}w0Ex>&3V0<2DKE^{YI%&c=!q~ zF>k?HgW@dv5YY7Yd2qPf5Z_WS`g(!}IE7pTjNAad?vZfG>jY#jUqKItE~W?PGOEIp zhJV1pW+yw4D^aDP_snRx*D&R2-h<-FdKfsv44?O$#)Ebb@$*6{R=gk{FYo$;qsC2R ziN?3^Z?H9!NSw*koaDfcBIJdXL$BRl{HAmporT0$>gOADw_7ch>bplzd`cxt)Ay54 zms82SDSy$gP>szu;W6>OM^N6R3&VviS$Ka0(bJ712_0{UjFB+Sw`io7L!!YhrGVyc zbbv*@4`8(IIcQz}7FL@JV^%%_x7q{RCYFP0)NcB?@F5Kn_+r0v12l5#BWj>&038;! zz>^#;XoY|%JG}`_^uFP{Z*uJUb6H&W4)K6d2>!l)9D*)ZLx-d~9yt9M_td z`fY+1oHp_C7f@?go6~~XW^is<5!{{i2CQoYF^1h;EIVV%;!m~02We+K{Vx}7_Lx97 z(FKobfuK4`fKO(c<71yScwbbH{yx@1+@pf1omDx+>+9g1-a}XvCB7Eel5x({=mhn%bdPTn47f;xsk1OVtXhKA{p&Dqw+}XXUx2H6FX7Mx zS+pvaL&mn_t{?TND1D7qrrOYo5i__+p5j~!b!P44%AP6+>O!qXOx0yE^UG8~=RL;w zJkc4S$j88B;XJ6s7tomIMx}DoXwLjG5dZfpoa!2f!ewbFvOEFm4+v_>qGEXJ;}zC4 zPL%(M81u`Tw$u4lwX{t*80@~xfCH7Cu&-+on!WSJ13jr2=U9gMLoeZ&V;a1CcNNM@ zy{P(8C%VB;0ZJ88a6s!fs<>#d3hg|&cyc!^z9WhYubAU);Y8e;_72B%rISU+#t`P` zMaL#RplRE#)9%a|*tB{91nRs4o~IZ7E__7nehrcOv+8g^I1avTkiod?p*Uh4hzFxm zu(Rkq>Nwcbkepa*dDjD4gz{mF`ww{GwGjQ~R4{al3Py^|#_lXRZl+8Duh+(q9yz-P zzYNbs!(+$L=iE{Fu6_>sZ+F70m)k*@9|wz`YGUY_CfFV>gYtIj=)Z9?+5fRF`M`JG{?>zpS(~4^|KcQR9 z1g3KO3ci^12cx1)**c|bIQ^|*YI{gv8ycNcKzol_@*?P3x8-*tY-+AKv5iW4?JhG%qpB#9; z8wcib=&<@5i2qSw!|noG^`2wLG>7S{a(M_k5)JJ!mJs~6f@b@c)A7mH)PCqJZ5%0} zsndtyoT>`4zY|fkE*A>Iu7b$LW{{c|K<>NR5*5o(LS3ZT?{ZNVQ1lh6pZ%c1X}@U3 zvu-N+Vi)*5{S7t$WU*Vu2~F<<)>RoI>gbS95;dHiK_0j1*C`CYoQjQCcVKncXcUQ* z!s}{dv8?tXIEPxI{TMGiC#YX!Nf=dQV01HzI zijK?2D<=an?W8k)vxvh;xr2CxbH?RYkI-*pme47aR}igB7pRs}FOBIE2AeTuVEC{N zB!B*d35r4VK*}=Ob0CX~Y6wB+U=!^PmxqpmSUBGJ5yZPkKqR}2^)IYs>FP!7qyUS) zq}l*}ql~fM(G&lDNWco)RCK=hl@doAm|4CH65Vw1k;5pwEp!KX8J}3_-RErF=<7_p z=OqnJFQO|lAJDQ34Cd{q0EcU~=<&S>6{IfXp1TD&IMxo8j!#GB13KvBdjd~k6UKBs zz#NY{s28r*y-%E4p+xxchrmoH9MWzI?XX5MG+i(Y#`ZWZ`g~dZa zlNV_AItbie@-s z=FlD#ciN5*o(<7Sm!oN_OD*-0ZJ=Xq>gda@nY8njI}0tE!cfMT-SK-u1M@TJ>Sz~w z`*0`D-Y>%36NYf|$p;uX(uy9sy}L>aR3?wm$>rKK_r^ZD zhpwc@E=%ET7X?%n)Zcp?Eg|+z5AnzfBoQI$*to+7eKx3~ZE+)trPkqv4Hwblo+uWs zOQcHMLdk_E%b0fhDpva0k=12KP#vyF%FQRN(*{9+I@wapoPe}f_L8@E-N zvUwS1wCPJQd422>H*wWfdTDF|{2O(K?(_OYrrZAmlUbq+eY;>A}IRQTgmy6l$%yjT24lc%1g7Z&ruYrG3o9E!xSf&sI9|7Cu@vd*sDH1_7W8c%&4Sg{DHypF|WN=)(5ypM_N(?V4!N2Mh zy0<@)Zc>QCb?bZa`vr9z@YG=O&n`0c*|N~$Na6EVRTMrsjW?_6xpk^}BKO`Ph8(gR zkM{9-IMcKbihSm>lKL)oP-i)0D?GrK(~DS$?p~TwehIYiRl!67HoVa5Hk_UG18VAW zxvN1@^yM1|c<+A}dhYdO(C+70FUX?~NOrPkk`LLN-~eiP={;E(K7or)I!bsFMlBwVr6F5dVe;ac5H!6Ro-hOYd|xv7<1}11IG%dU?1V6@ChVHL8-}I` zbXjA!f`_UpHZ>l`)8j2z8}A)y$TP-cwT5hO!5y;jw-wk4$ze?&$CF9#q2F&=3A7JS zll^)HVEAPmdix3T1~*lFf8zp*T|7pk)b4;b8-}$7vr*sEgUz*(;9qcPBFmy?K>dg) z-f7dOpU@CWBF)g)wjDYHY9Vpx52UG11t#JJ-z}X5zSoSzxLQ!p>9R0?^A5bZGnc9M z2Qv`VH699VN9lt?_-5uzliDRvqvba`>Obc3Bf?3}x@lm!$`7)ZEQ9ofOzr#CPW@o{D_| z{joKi+L*gT#HMr@UpS9utiMAKG)t4npr=G9HHV&e`vOsEn;>f2SZozwYZgAM;C*5w zsH+=6(NHkex_XgblTM)5|7^yakpVcM`i%ODL}5ii65g`egO4t1V*K_PyyZ|#4<6nD zPoqL0=E_%E*lvT;GMBJr+Gvm*v_Kn=P`t5R73Momf$LKnU|Fdu{BUZuzu2foIjWdw>f4K5a5R#}~r2 zRYe#u>jl$&{+La>`JS1NvgVI}W6F>; zdoBEe#h+2+J60RA14_DV_BCYN`>#X%h7dR^a}NHw4Z!%H<8af$(P;5%EPjpgz@-8% zg`eXZiWXNj$-VKuL|3$v4FiRJi`7S`2tGUvcYqLC?+htgx3zg z!ZmlLSfS1hoS{AggSVfBSbUUY<>23gEarvzZv}PB9VNak2F7LBgc;o zw&TxP8HBoZ&~Ei z$83~ZAKN658rm6pmrUKah73;EAQu%4xMi=($?vJv;1uTzS@$(ypehh04L-mVacQXg zEy70K9>b<=)nZEL1lsXd32att8ar`l5;l$=i*70mKL$ziy}ym;=boI-zaGHn>nS?$ zRYVu@MT6$E*z6$oM>mHBHDsavP&YmllVMxu4Z@Tu=J=<=8_yb#WtIz^nBk5LCVlw> zN)}(oW&PJMbl{3W7onKHG`5((A$$W)R7uC+>rXLs?*-iXqZlPlb>W>u8La-w4K{D! zIg1zcmP@J{57o0p;nP7!EOZD(ld}SD`KAc=VPg<`5*E&Un-}1{`BU)mNejHSXccSa z$1#=LC)gePoAkw#yYSRnAFsa*z#mz`*zxW>?&cENTARb{Y|#O>ZnZAH*IS0JpUO~Q zQjRZ_X~2&K4=9pH+m_JM5of0K*p0O`9c1oK1S}+L$X>G<^v#WV5S$qWd*@6L3$G@)~8{AX(nnMc#j#0jH$LnvS<+j*RFFZd))g2)=hi^o`u?|{V@k`{kVclZOc(D zbrSnrWyD;^nlmLX1%t$F(4=54_D@h|4)fk)!Ht(#I&y;K?0QJ-w`I`ZUeCd*@*Z?A zGsO3opQ5Xd4AT?%`t_c_;i|Mf$ewh8nb%BNRcSXS6zs#&l4MqmMa<`U3k$MSqL;?r@@QGZu41g8yv!ooQy z)qDY^_8vo-f(tmfxrlkRzhprQV*ImM#e9dV*!{r^%r@f)E1K)UJY|EJ`{K3q*3v;T ztG}Mw7aV5ha_$T+tYw}l7jTc|E4&>j$At0|Fy-DLF5GR%PJOFD{n>Xhwpp1CDY&wk z)$7~rp+NYzDjHYLsR2QM8@i-P zz=i2LjNgB|qH*C(d{^}eCIsz=(@Yd9L z=z^~~bhzsPyg(ifoqUKlZb>nj%q%pHS7O%>cwx|tG$^KlQ1ea(hwo_PidG9;cE}i{ z15Ux`ph)Z(Nn?AKerB@|c=3I|Zw0YmYcZVJGqLbQ+>mk(QMD;^#-fksNSlTe{4kfE3h zV`R*6)Zql=$**A&H`1B=!#-vpJVLY2NW!4hD4h4$fX%l%h9@{%W|1U~_R1@;YuXKL zv|)Jdo-kUDFXlo0;xIh?(-!vyOv1w^EAX4W zG^aUym53C0U~?F{S0&=#Fb$kOKMpLV0Ai-y!sV5ZP;>GJJmoST z+XVPmXxSbtUj1D#XAz#>u@PSf&jhjMUi7V~41M8QfU9E`;j)?mSde!JB-mEE*Xb$Q zJ=KWuwk}}joA)!vwc=>HnZw5g=keIUY5dhNiOG%)V}|BUIOj?b{`|#5$2pteMEeP- zo?8wnA~L8`I0Tv-w+hah#K7r>7tmFE9C_L4XcH!iVdLt+?)pQN5?hKr&aOZnHPEAF zxn%d31LXX5ciejQ5-OrF`!zQPG&y<9Jzaor_w9!IdCuUdI}V#gjBujISQvm>dg`+X z`?+8?6SB-?>q_0gcgtuPwR8u3l{yH~J7aND&P2BHY9Mk$Rd~taH>!GMW6TRtH2?OP zp3~_E*+qshnDB_2PCib5bvz-hFN`24`Ua@X?}CRxzhTfl0?K#-F6WtDxN&O~Zb(pN z>qZHBhkB=DlZ)VgeV2iMF7feQ-2?P|HG^IB&}2bR#F%*uuuJU<=$^S7P8|)z2TRNF zU>jo_2m6>!XFt4okO0&Bq9OmB6I#lz!S+cO%=Gm_6pwJlInF2W_Ut^xg1XkYeRQvG_|po0C!PIY`!ee)R(^jqgp7YEfqou?ghja>fqu- zLoiiQ8TW1pXYR)iu$pEMrY5r)!=gO!i=7(=e7O$1&`{X7K?{FcBw~zM06J}5fiaKg zF_*cvEZ2mv@Mj`0t{{uH5e0C}Z3ejG2ZJGQ;B?=ZAA3i@;hr^$Ps+?#ZKNtYab1$- zj=xL{?}icSI0IUrZ-A2rodZ~n_M7iP103CG#LrI&#JCVu zBYYz3%fI^A7B2OAXR zP|f)ftY6du4pG@yAGQ%^%Fjicz1N{6;1Nt+cotMYGp2Ubf^C+Xz&0y|(73L(v@utg zu8o+GS`Xci7~@f%5q7^V0kv(*K(zKRb=zsq z9v?Je#-@6#pU`VLHQpaP!j$OH$3#g@Jv7P;5*Sm zw-mN+ViXHldyx5Ey$wCzOX0_u*Kj|r314ewp`vdNUNaDIoV>hIy*dViwhZE%>Q88< zmWLB=Fg7;GnDw1kWIyk>knRU+^kJ|l^eh*JQ`cpvNYNKsb6OpX-u1$?&lLhZ))LlV zXs7b`ondnPCiYlqCHpmF7URDt#S`}uF?(JevJ{RoRGSU>p8gxe<-t*HS-?W-3N1 zsb7F1Ojxsl$~J}5tn4G;Ti8L*MN9(=wiLVb0cYG4#qiyCz%YI}9BXrfmgVc&JQq*a zC%uGSv+`u_BFos)#VZARVC|^*+Xt;nqcAkZ9ldjBqm!Wt_RIgpO=a~sZ{I_V`j!c9 zcBjE<&n*a?brF^pdqC84K|a4sh>Vn1Q@AUSQ)6?;z?CKNqHZ0wf1GJ8WDsvXn%w6Z zuDM0doc95N_i!)E_tTj!@z7NvDmOS+lNxCm;BXB29k-Npx@X@^lhOZm*EKV_N z2hN#ve3Il11dg-*Xl22@%`oE1H1%QK=3n4^cQr5fjUCbdWClhbpK@28CsMEF)ll8+ z$<0XVqqjT1!>*-DJneasM0piQlO8N5+Dn%ZZQ)9~V}B~IsH%tL?JA{h7C*RR^GJAf zaSg^C2_+UErU)>K6}U%PiT5Jh*!o69IJaoYY3sj>qB-wbd#O^M9LN011z21p7~MKT zE2}6T+^52pCQ~j+zz&ZP*wfE-H#r}_@o=a}0w;Igy>dodEIhRoQ`QImlr?;XL_PIxlm2IP<|#nsm|biJ0VW9FYbkJJ&wHgkzLlNE|09= zUt2}|+Xdf7*h1`|XTuY1Rw6&V1HmE3kGDiag1){I1KaFMtyb3Aa6gRFX|&6Ap1!&$ zWQH7rN>3LuOJ)OXzp@7(n|N?Ok&#sIv=toxZcU==OdvKj7M*tKa1pAzX_ap!_?iRP zaG6Idzd3;IizsfP=~LS9_$wT+c*z@{rcP|j&d}C^UsgGuAFTC%+mPD1{nX-EDC*iM zvZ1-6q}r;R%6?e|LjG@g#clscZ{Jo>CK|lSiUDN*iVbu)Dx25vcLwj`KQmH$D3&+k zu%EO=9)X{mZFz08^Q|{)5*n~dfoD6Vm}qyz3iw>nJm2^y+|1xms;9VtGx3{CMV0J9 zYla55e1Z+NAMu5O_y?Tlh1nz_FPCmx&a+M$6J+I)sB3-VT@i2Q5-DD6t`tdK@zi?G zgGpp)%UepmEF?p5o2ip=D#W|T^K72h(x9~p82j%C$$!`cFQOu`zTAtGdo9GBlLTjET2Fg4X@Dn0YPv3zl*2{3OS__C+$_XpGtKZb)A zK{!h@5Njm`ZHNc>82SDi2pX|Kz_?^~GBWOeF?1gOTzz32M>eUfL`fktMI89| z?Ot33nr}Qn`Kt>2r+JW`*kenV=&I2z_nJ`bo*$-$IbmhpT%7rmXX;*0#Mq1aboXaN z`fGmSGr_rP4S6#CcJjYcI*qZwwAST$Sz?%YhoI;`o5N-fyQ8X$PozO)_17;t1WZycL54-q>)}9IstkMn4VDrJek! zx~-`W+t>3uV)s7O7ncyF9q6TxcRi$H+V`kWRw1=h&!W2*^Rv){MYhk%r2o`6|#(PgEVY9^sx=r1Qp7Ua8)w4=Y81R<6W>?BReD?_7zbeL~ zK8YAye1J}#@E=vH+(qHpN}PY)0L{-_gYon1X+ec5RZ4z|CM$F4tzAbcYn@H=ytUB! zyc$-l9EqhRDs(}P49)%f3H@9{Xi&o@dOUhN<#?XKxYh=oBX})@V+>uR6i%H=_R(7E zheo%yVx_DXN+yca628v+b#@5#HTlu9j5IF%)B&|sR?ty#n#}deCwje7qE!uIBKE@{ zI(NAT&Ry$_$39x%2)PIN>U<*}Nj`;5+b&^vTm(Ls3d6!IpdKEB*zZ=2Dn&=>6gGlh zHHxGKHW$gWO&huOPkwO6rsTs<+3m1*QaHH#9fOD~LXhiVAo=?sRNQ|CuO425z;_We zb8aL(lovyvXuU_rNw+W~<0`Hhl|kmP;p7}Y|5$Fe035=mz{eO#XxpfQuH0zUOjJNo zf|MvKZ-huHQCc*-zZY-mJiw=)FX4?bQP?xZ1NZQ2oVD;TY^9gr{PS}lZY2wAR@@~z z@x7#Bsy@p0OQGB4aabeZ_nI#LKBl%GjR)UAEQCXxx+fU8ex<{s-%{mwuV|CiewuMO zlFIDaMz=V9Pl;|L(1O6lGG74*`tGWsG>T_kQjNu)Z;RP_0$G6;<&KxwW#IIqiwN%A*9nBa~* zFYckdV+Ec_YeUJxK%8jofzr;~v07RTKKcA6Q@`fJ&FydDvEKzaDOU=H=O*EeLzY;V zH3PpLF~vvQRWVQ18N2nI=ywAnI%ny4y8NCvwNHPJY6qX<1o7L@P*4Vscx~15UpdTCD}*OI%3*c~AcpIs)A?CQY`@aXz%J@GMn;r(nW3kZr_w`` zl)lr5$BS*rxNmDcCYCLw;vOrh@5Fsnp{kAGbvYQ+F@kVq9qb)YM&r+MXg=PK4n4D` zEBumZ&(S3quxA`r=4j*gfMAmC9Y_2nyU3jXw$SX(IQo7^63w|goBmy6ODh)I(D4_2 zv5<4d6`$PjeT^NxHfBShD_hv`a7F;RKk11c_llm2wQ zNN-AvrB&f7)b6exRr`WCaitk{7_*oa7YT0s+HXr1$NcSc=#~c-RO#Uo%3p}_*AH=w zXqCmWD{hD`S)LKKzQ_>CFIs@vCk#=d(uL1ad5cGtZX+GRGaWin^qfK(?Om5gJ6q~- zulEOhStm}lO3EOv_a^jReFRzBarkd#Ij(*-jbj%4?o@QqnMStGrgNEL@CwX_@x11IQ~zeVUuy;3 z=(UcXDB4A<*9Ftm)q&J^G~%VRqj0EsE`GjaMolF7eD#}ZG{JES{mH+(Sxb;+^%%1m zr48JfgjTZsmnYqxHlMy4?L~ta8QfR&3I?}62Fv5EC`i1BCMO2*y^Ahpj}oE%(lDG6 zVT4DYkH-&vXEEdOU7BgcC+_A~({IH;z~H?Sc1VoH9a~PIO4%WdHBUzCV?5h(FBVg3 z4&axWPtde81La$C@uIjh^&87az^|W4!6k~^u#F_U1*Ig;@HyD#-G$Oc)9{aqB#zJc z3*9mnXp9};+g}ZJU)y2XeQWy7NQwS!dyRt7F?71a7ThnMsl|umONjGp$@Z9&XwLGVec!8(5CVq9{S*WVTION%j>*bfA`*jWtce$fX{&h&XqKbEJ zJ;J1nJ2a`biFy>@q^Fan(@YaXy6fY3ddvJg{$BhPtDd>h88S6IS5n4nu$qAwiBt1D zo~wzUOyd(WuyMhAnj%dQ$K{?9sC!{7&RwmIKTcTTJ-*jI-uf8csC$5|MTPk5!5Q3h zq5zq820jO-K!T$@Zhphh{B$!hXLAy=E<0$j(FXe5au1EQPUS>H{@luR6O#Qwh(UcE z2FEPLfu^r?T2veT=+{O=RgGx0voxKsp$qp6PlMX4=J2j&F@*Sta9V~VR+y~9qBd7i zn#5d@{VF@r1~8zR7pBk*MM6K`7>$z3mN@U9Jt_;u=ps&@Hd;@nNmdi!#f%rE)?E(l z{{><8!bI%^_TfqIUB+zTP32%6|pr-mK{Lsmy z$J|n=s4AWQt{#cGL8GvgSn+-lvqi0ymZI7!PPBMc7BcJNu&DYNUVir!WnHULZ|Wud z;5~;nubf6BR9DjHooU4DeG`e_brB$18gqW%h9s36a5(=g4o>=j8&|wWu4e~|1glY6 zZyCN=lY(0{kKp*xcQ7^WGCXQ(gNs|QLfET16h9e*DMK4@!{ZP-oaaYFW&G$<*$Vuv z5Q_U=Z9z@tmGphk4BGC#fHuU3;)mfdoPF;iUJ!o?g_oPb@0%Dtz9UO#x64xBB~tW` z*gbsM%lD-VTkx)f8{CB3L^7~~SjYUvB}Lu%_-;QAt$4yV$FAerbp8?C`WYve_Mq$j z_t;jo8z!kd16Sv%_(Ed~{@xmf1`S8>&Z8tc=EqJNlDvby8y2F5IpSu!89cL4h{{`w zaH7Ixj6S~`)wIHK;m=eI{mEEV!ZLn>m4|5AQ3hHoDlD%bVR$tLy=uch^AXUJIOB=Z3wO`@x{j3f}izg_1R@RL9wnuH5TI z=LolOz3XK`_o*qYnC6BF`eGO|MjWrNT}SJlZl%lPSJO#V!2mg7uq9g`YhykUt)L~O z)@^_!G2ppbrOEy^8oecC)0ggC4KFYi-pk-Q90X~9v-&~CPsb$ViSv}4J9zw zJpzyP|G=$APq-!P&yinuM?!ot&uUrELaR+#xWi}^B@HWi?&vhVSN|Ri9!X$Wj3c_- zdq(nlTF8bk)4|2+FCO=rKr5DtXu|MDoU5OJTN-k(?Q|S;{Cow{K|1(*btYfKmSakB z9RB&j!1cs=V7TNyY}6f3)tvYmM=+aOk}61C{yXGRWQ}B7w%8x{{y+k)Mk_yy)z?`4&9B$ zjE%{-)MhEhjXsC+I}`9gqbtsS<_WW>F2dBSigewhNw_^b1wZnA--Ig$;Iv=?*hHO& zAZ2}QeU*>zTvyO-l4p4K{RlJ%p9S;ptr#@t0_v_mh>c$}$mJ!@M5G)}vC7Vd z^;sRQ79K$JZ=wl6Ue0#nMZ+Oen-N#$t z;*E=Nf$!Di$_>JaI7Rf5b;8|`f>1_z9$#CF@VfCv_VD(t-0h5m+&+bm;1DT;PtS$o zN54SO++7LNyrV%YCmlQPgrakG6#h7-1e!zB@KI12J{~a%1GiaWbmDDHU!H_bqE7ri za}-@xJ0GT}e+7Y`2wftc;CE+U>*sq2bv*8aPMQ!u3V2U3pF_|PnhQdSen|UejkntE z@#NtwTvz>;8>@JM9I@7gDeYZgP%VX0tG)0}iwl;^Eysj}Ls;S|4YzF%Lng1Yb)1wB zhK)SCGO!uHEzH1OmlM$5?K|$*ks&!TuAE)oHV!qpYzoKbPY+Bm=qkIDGiBYS72iBXU@gFn%wm> z!s>HlamZ&Tx=VHQb9p)Rf8P&i_7U7_mVo>6TkxQ*3w!^Lrvr|%G-&!L+E?X`>rOV~ z_NRaFJMSUa_jw|Hem9;zjd%r3>f(5(d65h>>JrZ>PyfZ5BCc{`EO(1j!}fy?|uo4-NyUS)N%*2Qk;8nc&HeLOF8>maJ(p=Ta`1 zPc6>}uY7T#-Egw-fqDg=3Dgl<-mPQKd^k!P&Ugt5LO%${_l)EmBQ*s|iysN=Mx=vc zos-aiMxik0nj`n2zE&t&?=Hx(6R`kD7OD$7n6s?Z zEI+~V7cT@I>x^Kk=oqtMUJ?5*_byj||EVD0`UT6ch2wE%njT}770Gqwo+LMzbknc^ zf3rdlaqO+AvH0mFVVTOMllZP8MrY89i3^;-PWpVBdt=hWWt5e{_?RjS5akQF>hnys zVi#-1%w}R{CUB}VteA%h8p5b4CFJ*bj&<;Q&Q<+1Am3xMELJOiFwg#;jjK{z1P`Vh z6qaVDLVDvU!G)f+f<=4Rfw@V&b%rJQ_xUEzRizn+bE6cAlr`D$6`xG}&~ z&?hrdxOsCuN*&c==#CyXKBY@=WS$*cIaZTZ(MuIxb+5E+Kj~pEFHy*D$-iJ(bXnQ* zI6v1=c%sd&D}KTjf4&U=Z44EB&0i(xs&XZ@Mh6A+N^6DVV*G_qy$mf!uG15|wh|+L z%g&n@nNh**&x^?%^_4=O8Es6wmLillp5(seo#ZTM9Du2T_bnX~GzA;GfZg@i!}8RQ z6PBG`NMxsOW!LOe<^o;!0o~GN=`L$1JXFc+9mby#z-(#3@=5<;=ag}R^@k0Ff9@R@ zcr0aE zE049&mgo0qppF%}a&;4V9X*nG&s4%Ak^QhN=?w(6_|cF67uspY(!yQF*kELUZ-bvh zQ~5(8H!hwChfkB)MpZzAbKp|LMlip@>+|oheE&`#5A$=(Kj-V9CE^?;lxSmA>T9qW z2nXe&`xy491dsC^VO*p=ujiNmVs5>J?1{uFa$zVeT#kcf##42=sLO* zo8yl;wI9|PuxvE?trNqmM$3@&PQsNpP4P;D6^%^Spb91<=;VYwf7 z1!&_Y8(H*ywp8>~ahj;sz)%#m&xH2998dT6O421O!!Y~x64YBOz^P#~sjabyu32GB zb0R`v*-s6aliW)JH;AEadoirx@0!^a=W!MrgxX7{VzkFQC~9ekm$N>?+b_SkrCE15 zpFC;e^)>~gYf{nQGaA(kc>QqFVKh|9!fiYaQd9E}GV2>5^iL=lh~i=XU<8cI%f>{f z0Q9`J0?U%bM2gD&RJrT{9ox+Jbl)4&f~8W_;Os+uvHuZ9Y`>4BA|Fn#^@B4h94t7> z_hEwM@pE1)be?)jlfrJ$8?(>TPRUDPW_bt3d0v4FB~R!H`_nY?Z!CShy#cG?0$RBq z!x@9>D7#?-u4m;iKxH2t`s7aw?UqvSoR{EotQ<p12hS$LScXtg$Xlgp_n#3UsJ$)+@3=fkF(gb*By_f&BdSnbJ09t z9Gwt7nrdZ>(}dY#^kaSp&dBS={CSsPanM*8ZVw~bypP$4U>Q2HXaKqL^Q7YTDY9?( z6|%|0mX5znK2m^Qic7fm zU%p%uzg169;5i1F(fIGO6h=Nu!qMw?;@q)r*whtGMk@x9KRN%A(0L^gbvzFEY$`C$ z7t#uk}&fD{FYOX#sz0^BfmZ$$#L{torA4^Rp~j|Us%2F z7Y^1_g7kV#m}X44TYq^7y&b zV9_P2kLrJB(GrunNqoe%y2si1;=>yb?G-Co5rEH zNEfp-weaP}PbBSB8kwABMUv(w;%liGG#{0U8tbn>+oz47%dH7@}{Z#YEeFJR%m6bBIB`BKfs^0a-EKg8cm*0a2|F@XhJt_|bDe#-!QeCna~( zY4t(%mdW65Ou$|?5MmR~fcnlo@Z-`#kU8T@Lq%pZOnNJw^KvHoD)aqYwK14p*1)g- zXY4APNS&jzaG7{Msyhwf!UN4*30+0D%E^(M(kWPTeInM2okN?2W!Tnq2A3TgNh8+& z!-9;-wA>%)?aB4z-^K@I??XixKJ%597)}(O@6{K5oMA+@dgjw(&nUf>{Egg5c}?cb zG=j5#&cn|7?eP1a9VC8V2O)+)+U13)cXkH~{~UVN_PmdG1D8V+ZXC89R-qRU_Cmv<=b)0JfN5b@(eY_Dj(c+%7dWYqikd7= zdlcp3LheALPd|vukHAgu(|K1!N%(wT6Ba%>0yX_p!2d@pc_oa4(H(v;)jb4O^N0N1 zS`nRT#!&4uK3M3)_xqI}VB+N#7(1i`iau7bGew7L2>5s9pi0$tL~#Vc`a2-(3$Kz8B@7_~VP zU8L;r>In&Kou`I@(HWds!DZIIPKtcg*a0t}`G7@rBph#7h8w}w(A0AQeqFc%3+o-h zWa(V6e5%i9mfKV5BO7R%Rs`uA$Ru&m>E!wQUb64MLuAGC4X{V$7grOon0s|Eh1?4H z0!lvoy)*hN*aRJfp>I}jKcB*_o0-sFaSk3wpF&@^RA{}E0Aln0K*V4yStXZ3;-9X7 zH}%H2db2JjJ#)hM&Y5&EQ%-*cmr=o~lgZ-yu(C2F@G5vl#|K$(0 zmQWYzx+{xxon7(Q!!@YAKnD}99YoVDIassRgsymkPwzyp1GR)q#ji zSh7lbH_;fALQKDOkY%yQN%zVj?v7?Z@8u&wOr$$_giVHMFD+re8sB5tKqn|Av9J`B$4}@xFh{9F(~9MmgNufeE1FSjO=YL{INgz>1W60Jym8+ zg{GX#x*&|&C5A$sCvYTVG`8k?L+)k)4C^i@*MH9@f!(votfHx;b+a7Nn~x-XrWh7= zu7aWQ8=$4Hju__cC1M%f#5ZawxIffhLI9vA` zv|K5G>o0uZN$*=S#`+7HSECK=MH$H8b%w9Y_Hn``o4H@!S)41sKaPC!o0BipCcB|>NMq^Mt=|VQ! z*^_?(AHn4EIk1tr2638V_-CFhMwpL9nVeMeI75|~EEOXWn~xFa54Xu^iRZA>KaHH@ zJy~-093UmnkHVaGFF4kh0e|MlbJJQTvJLK+ndHa3SJ(1o@Z#VUIJbEx45Jz-@Z82T zFoIWJ#$_pOhUXPFl7P7bJgQ61QsF4HDe*nD8OV@OW7Y-zQap)LmVqGv+7}UlT`m zP$l?0;u7cUr$J=x=a7z&R_=J`OzvmGTCQY=1e9I*NW^PC6TKNHd0(Miv{;#r8QtO# z*lx%B4}RvZdxdb>+G9EI`vKgzu6}Z}eu%vBo&faJSkf%$=Z1C^l9|6H;q~@!#9Lz# zT=dxnD;M1%9^EI%TyuL8dU!FhNlF8I>3ICY9>CNE-=G<+7Ii@>vr%a6pU@=t<^UKvf+xFnYK(*n67jdmkcz*!KsJ2HeFeIxwsO? zy)Pg;wkHzKABg!Cmd*C)AZ+g=Zp=+nHtOgos--**Kd2qxo<2VYVt#UD>c;cz$Oob1 zp4xLZYj+k~{$no`uz$GAiu<{UDfOVmz9o!uCX+Sh4LN(Im(guX;r<@q18(h4xEB`V zNcV(e#L+LEt8K34bZ;lHpKjSQKB0x|pdy8oOSup`{W|n3U4^MTPk?>iTXJ6K7r8%6 zNc7S-v#HOHL*{iOve{!ZvuaH{Mo+km;q}sdh438}zXIB<`Hy@1l=mVt|3xgmX|ZcZ z?1biVOUcJ=D(o)hKyucsfV-G4WyxL5CO=vAfSL8Kgi{ajugKUfJdssel1Y~4ra-}^BdWFnl;i*g<~A;Ru^eB@t8$Vj$D!3L z5BU10l{u_o4(p7bk;nfHlgE?)BlA4^ShKiHPSK?WRu@i&6T!0?t5`j})!POoGHFC% zRvhLPPR6Rq*NM7fBkJ$kg6iRlVDaA#VjVP<<7z;ALLeD-?-tftB*4ty=g97L>YSdk8I4T|;gz68 zc;#FenOj~&wCdI~pIRJ=my8pR>*$6#{u3azUlo2xCUKAY)^T3DQ{d#3yWGz+=44^M z2gsP^laE@;tl6*CFme7NHcKgi3lmL+qThPN#VJX+_V+57>?=?HjPqgw-luY|1)(sL zpF@QjE@qcCEg*SaN>FNiRd{mlNjBRmhnu#qR@gg3oE@#TiF9U6V*H=^F$350xnkq1 z40CHYXMR7NjDNM0S`ttX8=)EP@g|FYzs@_Su$ zWC@Wld&btq{71g+62Xbm{Y?9;+uVgkyvNYdcy2>MAmmLSh4rTu*{M4%$%lZyMAJQm zEep-%j;$HVbLSV?T~?QgeX0|D%2Q)@evIVSj@(R6z0v0kWABnBp+1nheVD0oFlTl& zIgo$H+J(Y58C=3_KXTIMt`Nsw=dO1D;MBG_a3gmnajieqU{^s0qc{2@w=h8soJa{f zU!sznHI0Vbm8-eo;hAtSQWBp`StLBLH=MiKyohYFtz@p4rgNKr-ytalmLw%%036s< zyt^<)IDh7L_Vp4ka@%S;BYC}(E0Q`!-rNgguho@vBJ(~H_^grTerk}(zxs%O-!8^i z{W=@us7P+F-^3`bpFs5YKP1X0_b_<_?}bT0GNfVtT~@&?nmFYcgI2;1Hl_R?sX3hl zA2!-?9q|gVe9nF730*F%S<}qb`JE!)r9y=*v#S|FlQmh`u!t3@RWV1K3`pzgPDZ14 z9w|7SOcv{f6QpgBCgMG~Hx(J|DxFRiv|0|4ETc_cAzjg}D$IpoL^n zJNs^t1oT;*f^OGo!jtRD1><_u*+t<`*@cJYK{oIu93>ZpYJS?BT8=h(^d^=~-r!7P zR_cP=C}+XjN=ac*f*!NCp`F=in8bwSDU-ma+02{uRji%C3a)hEl8{ek6LeEo)^=PY zW2;}mMNCfR=dTmEr3>$pI@5Hh__9iPq_mkGEoMRJnrLSG1LS-i1IW9u0(R@z+1#-9 zUeXn?T6lU;o3XOk$c=89!>XkuviehEiMMGgy9g7=hqn~MT!xslA5XH@HHzfat`F?Q z4_l#smL`7NyolMT@5#wulK_>Ky=;c~0BOwm2p?&gFizn=E;no|vAbQ%wA;y&thf1O zw$EuM=z}^bJzGXnd5>&eFKH%AN{XbfHfNXoozJudPbA&T-!nGeH^_On)u7LKu^Nt= zq;%64V!wQ{aGIeEH*4=Ua&hfkw#6uuL^~zW5V5Edx5+T zJ}*>^?_#9KBym4-vxTK6;#iAa(j;V4im=~o32VMvnIx)j;|w~L!QplvJX`!(7+Flnb+Ij+S4Sq%@4F$iTiVLJ?k*%-noXFKZtm>)cZOip z6UY8*+aUr|IyBbnNe92Tm=Af^S2hl=zTqdQ=Qs zrMB|fLpjXFM`i5Bd85eEm-?LVnz1lC=^Rv^Y!!CSde3gV8Ohg0w!#;`zX`=(*l&himmREWp=T7?8ey(IB83FHhG>7x6p4Xw?IxoI6v>H z@U`y`VQcLy=I{6U%+bGQOwZdvOPd2z1V_7u7}M_#f*o=DgmWLB7V7>U%`BI`$(GN2 z$nEO=Cm7$VC0z03D${7|B<%ZBDIAwFAXxExDpCBW2&vmy^C2XVov$RuY1qcF<&8U7dccXztDwK(Uk z&fLKMkwj8Gj5O|-=C=O%OSYNshdiGwxaGcyyO|qH?)j*ZtLLu~eZOrcf^2{ph~Ha_cLBvRHy;=fsy5d6Ok97ftqM#XWM_ z71P(StF)V$CyzHW&HMJVqG*3IM{z9d>yriPvIYTp_Ds+(8q1cCb>+6ZHgLa}USgd` zs<85t$8$9|v)G;d$(lLOo$r-6!>BQPEuPy^%jF%M@adIufvaJ^;NDqH4oXf5#nu}N z|Few~o_laXa8@KHbc_!Z-c#d*VU8CBKI_AUk&$b;fQAu7`>Q7D%Ty;1Ne6k9H3}xZ zm?h|YEH2z-lOnvMc~^Kw+>*&CeZW>fc_>u%Ghme*4|66@o(RqF-(iL(CUI>Y5scx? z4EBweEjv%KgHi7sW()Mras8$**yJ_UoZD0Zw?iVHJu;-m<-gjVF4+pLtO|rbWkz$V3Z}&K zdkwca)L+>5QG?C; z3}+Xo9OBaP1IOTKt~gbmOl|NXU!;@;9t)od?EPy6V@c%3o!Yg;~|ys2^p5a$&iU+(=~xR40;EnwjK# zktCTMp3SkmANO*PB=YxO3M20$$4<;Wz^1nDpd}B2snOE?wCJXih`36NZmt&-$rOi? zZ#66_dKyL~UVVo25G~xV>WGqyQ!weIH>x~3h%1weRG)eV&i_@LCJ4{@~q8~S{iJ2jmo=Rq#G-($$jM=q^CZT3~toGB}@69 zv5-Yaqh4zGd`4k|#T2T1T0|X`!)cfEQj{$AL_9(;N8gp~T$D@}Ib4?A(g+cN5?nwGn})Zz_Otp5@R1g*uzdtoZF6iF85JaV-dcuUWU941sX*Dpi$e0 z>5#!!Y7;u14)qS;nzijbm%j`DZds0j6NB)%%}XM)-h&*Uok?tYFO)r9W$^pNACU9S zrhj8kQ`ez0^v$04@LcLO7*~FU4WAc_u3fMZRmM?~msBj~zKX|Ijd094@C0>+o6*YY z7Vf>tQZY##y4=l@cK4X#jLnnKCR-l;&8I=lkBN|>tq0QiE6_DC3BMUA&}_HG==pU4 z+6-^TulZg0c5gi%pVEj6?(QdnQmcvlybKa}HWhnUC1cl*aLjsTL-)p6P)DIL4R-g$ z5U*8u(AyE)jvT}(*0w15a0)h@QKgzK@^tG%B|4jDkp0cIkycCKje~tSnBIhW_195- zj1v8`Q-v;SA4!jPFCdPOB|!UXB8g8NLvL?Yrkf_q(i56f@aHH6luLUH`k&2c!x{&w zamt>qYUqIGi@zb#cPgIeS-27J&*5|JZj3!zPK~2aQsvrQ8g`BM%I+i>_X%+zdoONz z!t+V}`!U&SKDv|(uvS~dv%&m*x$gzqXf|L#s}4<_KZ&ZmR;I!A3Z&yr9XG0YEhm~Z z7TuLB@sEouDvits-E$`(#-|Tvj|>6H?>k|;_I3DDv5?lX_H>u66&*b^nP!Vy&=*s19kX!t?r^$j)qYwyejin;m7#SVGIVLD43!yIfg^1?F&l*R zDW{Di=g8xqcuD+SP>&KG*HOm%8+QI!#aq9wK%){Lw5(D`lkgtc{Z9$|mmY(4e|{6M zauJ@m?1}BC-@~mh+i`FgOVm0#xv?{hxP8;d5+}_F5+WysQ|@Guy~bZSym<@8+|nYh z9#dGC)iPifagq%VdBqBUPlm&Cqad*OFPHRjHRGpL0!24f!Abcg`};sTCy_ddn}0hP zdZT3F%Ez@_-R9G*OvqKzk$j0uDw{yYy_Nx^2Ul3L>^l(sGZU6)-X^Cy(m3^pgP@Rl z7+!8whpvVhgs$V68fP)sWtgmN zSWAxnY3CN^myikNnb04kNsArNa1VCR=UR%ZiEO+p}n;d z2GVe1!!KfRU>1?w>Oi*Yi$TU4b287ffc&bD0*;@Bl}5L4ZH^UiYU^EsPn(6Wb(OH| zQaA+M$mdG-1d{t|>D+tqF!sr8CqP8nz}IPJ465&6UU;%VJU~?Fzp*Ygf9W7vE{_7;UifSDM{?C6u7m+O>Fwz!$PB+LMHjPA{@P& zPhKD0OZM1Clf$iA@HJ)#?q|vnJ&UohDxS}3+{ov5Rm#GzuoQMGdyA`BV*^p<;ppM? z2u_F3g9lrWkax?}AYtiMZdRQ$m@QfZ3q@Wqi}xGYzrUOeL@SW^kVBm5rf`TUH6X!R z9(XBoI~lWj0oN%$!0CRIA*T(5+y#0NIu&nG`{Sk5wlSBsN6r*w7&wT&b0U$|-$dqg z;&t|!%5!c&aSwEltLN5_n@OB|e^Bvv%{2MaAv*G_2EF)0m6iz}pz@NFuwJi)lpMK1 zPCBb1d^N+8+&S1$LUDM*a$GPdLPbFz_CJZh3p>W(%tUuu&TET|+U6Er%$0!J>&5v1-q3VLcqJI4Tc^{c{E{WT;>=k#Jv}1&I z5MEe10vGQ7NZC#$QT3e3qRPwXX=v{q8rj@PFH$q26EllweIG-7w8u~x%No>;%)xJ? z_J|hgFA|NDFcE#euP3^8{5#bgDy1M}jb~%SpfTq0B>G&RhUH(Y!Jst}Q z4<+Gg?s=@gC5G7Y7M41{qfs@7=6MU%Uv-u(>B8=j6g-K(*W=WFZAB{=UlRJ|-gMfhqd}+jmf^L?Dr`Dgg3oXG;Dzqf zpe0!fdmZEGnubss*}I4y)4za$C1vQn{1|%OC=z`+aZ~imvtG2W{W#uj ziUnD{_;u?&23SO6->wh{oMB5PeXR68$3c>Eub?R9SZmjT$kDvnYTip5;dX3_Yv$r zW(||wazHOw36EncPL8dI4ULnDoc~vR?COgcp^k0G_2_rlOdWq`(+p{4wBz}7x!abw z<8KZ6UChRJhpRE`*$kT9qDe>G8%tee`24E(cX9RWf9N+@f?w0eQqc+__21FVzRjD@ z*4-%PE;li_tyqDCyQPu6C&$wyt{Ufz@j|<{O1NVFg0sG|o9GM5FgeB&6@peMuxTXvsB^NwQ}r+*%6bS|UvHxeer9&8%8cw#WCJHuyO2FG+oQ0--i+CxL^g;Z<_|Iiw7Zhyf-dt zP@x$GhSZHE)Kc(;{77ygw$=$;E6>=sK9<2hn)^WF+%g(4F^W3f4x%*rA>iSoc+lU1 zj#~7E_lqW^XV3xyC-uRvG{Ey~d*S;MCpvL)8kO`&qH~xujJqX4Gk)cucKvTqwU0#y ztxW7bxf+Kwi!o$YA*NltK{|{@uo6LavY*yEmSLswURWNN zjIzZ_H26PL;(<5F{^i}|z{7*+DLb0ioCeVP(NWl>V~qRu9>v|!FJXhO1!`dxDE)hg z3(KG4rl(D)%xAo~p5*JVJ2_Z2Q;Y8WYfnq}yVJ{t=~yZ+Lr1)jp(-{ma7y_(3|mkP z@*WAwa}!~0$7%Q;Uk(qIKEalxg^)Sx4&E1=Kn>;-YBsA5QX8bul4rVZS*6lYjWSw) zA&2G-MS-%n81@cJVe*1dRFzN1H=QT3dhY;iPv-f~h%o&1D-=}GpK`Y40rm!RFXF50kPa2 z2BAUo@K)nt+;`j_2PgaEXzvsFa=bh|iIl|{)iB&-#p_EZ72(R{V$}Ikg&#`(q3+37 z_%mt@6lNDfOQ$K%?-awGc4d5*1UNB-zat#%0}Z`k4E+|#P75p|-_!n(%2{DNOULt! zs7l@CJ>he2Jeic=Nj~`1L)Q*j+~&6(Z_ai_w~_-mbl3~BXBncyJ2zCnn-8|v3NhKP z5}%%|M2+wvqNbn+YT*OKi9b8z#LUrhj|A#JpNeMF*JDMq0*)R$1_dAX;w=AGo-pCU@x(NI=IzGRaTS!oXq?MZ`_L{usY zk<5~0REUJgNJv?~=l2iP>vcL$&wXFl=ktD_ZFggS+HG`Uu@+>$_Mmrte?suL4zRHP zPIZqYLE8f*#2MwVbL4ecIlBqimrS_ny&vsj?2yj&=H2VO8zJ>0s%|`r4xL-6h1qke zw?&VB8GQtn-8SdXBPw`3b~b%9AcpB9hT_(sAuKWe2U=T=WxlO5L3Y~~2;lh;8~tT-ZH}c?v3#Qf1!Yz_lsM6>h*!9o=bw_MK z>)oNK+vkWH+8e0md|h^GM;v>TT+haM_hZ9j4Q5eag{D0M_OEn4^V6J$+Hx86w{|KO zmlV=Xo2v2V*jISwau{aK@PJB*JD~DyBwi6^K-?r<%4I=~_TW)dAYu^j} z9fk_y&o~P2xBH7m>&sYKRH&i)k_iHEykwOt8H1)it%CFLpYUE<9BFWn62=Z(5?EAd z!Ip1}1V$mZ1VbDw=nlo}R%8BgR&jwBiNTcnoW`T)Lb2Uiv};PW8nyh4U`3BFw2qm~ zgBN4q z0r$Y@F*#WuC)92lBUGEST3Eg1hJbod!If*pRPwD9cYV!JE;ed8JTuP|wGW&X{d&8H zPN^CuI4}Z*f|ojwo#1QbS}H5JJZuQl{Psn7Qc6b{{%r=vS`8B(8@W!D8EXNsJ5cai zPDi+TPcQW>Y7_j`P!oop^5>=`xLMp6Iazpp$l^LKNePrcjuuGAnnRNJL?Jjk2rJ{& zslx~@k=`LD7@J42d~2|%euseDm9Ub+)8!UL{R=EQ>*VoR4YM3!op0&>X(#x)+Xxrj zND?_**XCNfzKeFGAK+)M-(iaTMJw&}O;$Ar#|S(B%M>NnnsR45M^X2wa|J_fmI~We zNP^c$Q{nMa4WVAmRH}JnvZd6uOpDc#-MHCaAat>OB(N{J1yvtDS+@6o6o{M-qgkJv z;M_4|tNFM1UPhU^Rp^!%f;Kx!-`He{dWI@+Q;e69Gaku;NA1Uj5q9cOJ7-XPq*Wjp-N z3Ko{iBnh8iLoO+N0aw~1B}(4bDKa$=7x>CZ2r^W8UWEEeQD2{ki)qy+McZVp6n_R< z?H^@B3ODu%6FOpqfiE zQG(yC5wx^tqQE_Vf$(ZriO_7mjn%KAvjmPKY-w`eE`k4=k5)FHOXw55Z31WG0>Mj- z8Q^(oJSTNd$o-2u4YnTjRyG&ESXEwa6DG(^uxzxyVtL_$I(;aCOCR8 z7kuv16Fs1nq^-$OxUIHFsAiIjFZU=2pPTOz$b5S&_$~cPIHb2+v|u0$R&0n8ZBH!_ zk=edN8I?$LqM;hvVaTkWhGq_u^|wmlO)=cVUs?061@ zIL@;1+;B_Jq;vG$%NT)K^d+k|;xzXDNh;eQ8_S|T%@psr4&ovmhUYt3anvMUR+=S4pr<%?GijCw+;{IuEyxz`B3`L1+>^IsJ{Oeju$_J znCxrtxb7gnPYFc7Pa(MJwLQ*~6Jz}7vAEA+94i`Yz)ppzuqB7bvV_(#?6bZZQ|K1K zGRf&M_1SzFm)M4v!XM*smm56aB#w=1iDyQC!a7i1w|N3Hy=cn9j3%Li^(ZvgGR8?y(y%Qf7Bkfnv79?CK5CM~ z&!4l!&)PFsx;T@mzE5Riun24S@b}>QV>sD*AJgyiX3I_8m@=QIU|$WvwR#FvZ7D}N zNW<#Bqd2_Oi=PQEW?84^vneSC7$j+qUJ{mAZMuLA+m*{_ABNFIf2>eX#|@R%dgDzu z1@>d052v<%!F7}D@!gwk==FXMev<%Z?r6$Pcz6AXcZIk>Aq+=6*^ADT+{9BR%ocl( zo*}O8O<;3RB(N5@L}r;XlAYNl$9x_P;N&tHZ0jA4&psPs|5$G{SuqtKCeFjJCEDV- z>s7?B#d6}Dog?r8pQAD{^u(8=v&Hj#BE{u*4~R82e`7=6JG{tebCPm&*xhG(Oiopc z-CkD0T+))+q_zHR{qRcqW>XOqme+y$8II=@8{weSI9%2pfz2mMv1`UY{M?#~UuFj2 z%Qf3^r6IDLGE*48if7H5I@lX3gC1+;FtT9)caHvne!Q2*XNnJNQ`*3m&0WrlP5$7d zkPm3<#J}^6Q`zFlaqLBVG&8+!%C_g~u{GD#n1t;T^gOp3@9)}&g%32@^^b~dot+$$ z>@i}aXv{Q4wLw__$%ocQ0Mm>~pZ{4JUoN?l-&29B}K4O4Mgmn_#eqvR~ zD|FNB#?Y6w7{B-!hW@>RStc?}a#{x}vRa&Z#gRGcs4=CQ;Vf+UOwM4^P!iX9p9JNq z;?J2{7>73edFBfmENa0a23Iie!YKNKcO{LxYz_l_bxG@|1$$h;8+A;Q+2!hJHtBr= zi;mT&Wx+;t{cI8KZpFW3g=ZeK!0c95Hd2krG?=g%zjc`t&($#Q+J~K8K^Uts z85dO9q0=W07w0K4=YJpZN7rQ>+5Z^lt-Fm!m)2oR%X4hl-HTS%P5AV-4qG3g$+p$2 zvCa+;^ct_nGr*m2iH0nFKk$i!UHDG)#O63^=~x`LYcwis31U|pcCh(BSFynZpZGpe z8Tf4d0Q}$1$f$955dn1K`+9bIZu^iw!O+M<-H`X`0xRu zoE0%)B!~Q=5R0bqZh{Z9#Y^I6iIt~1h@ZL6WD}0qutOq!HkmYHZ04I+fnT0kU;-4Nep`q3V%S=*98OxMUMNH`W0Em>7aN`$seD`st=w9Z=r=0s<9} z!o+M_oTV`yGe1yl{&t$J;0oEc&a-UQ{1CRQcO5IZ=)@-U>awOVbr#_=n*Hl;g4UNs zFgiI55`X-G{cg`e9^XLXhDtWR`7(2_E@svC<+M@xGJT@CjRwhd^X&f{Y>Q3-YaOx# z#gQ{Gev>OY8U2JW4N~YX!#jX=jF{l6JbSd=jQy2gh0iJ$@_v!A*cx^Y2Hnp?PjEO4 zVhWl#?8A!3ZYZzy7Ygip;mN1_V4vxXYb_0NSLs&}t<=IdOI1+cb0})QNyizF6VO*K z1`89v^FAAO7@w;JX9lYOqIEjcBFIZ1tZ21}WsA)iGi#*KEd`;n&*_NQ5^USqp=@^bPn27gkKQS1`22G!e(d>- za(DmWk467*h4}clCw`scfs3V-Va}X(>invb zDvX%L?;|(jgCIwY*58Ge8&+eK)ecOR{{pLyT!9daV(?j!$qq*4F|C?(HoG+u@g}_b_(iMS&27Xoh(-Fohok86SBLgz~pl~ zv9^?-%h-kR`{S*+d2I_uG}Plg%VNB9J&6|G|4mQGzob!Q2@dIW!zYhyP<2Nd{;~cK zm+j^~H#VlQ<;6Nk9CH*p_O9jzlZFvZv!z5avjyI3x54q|KXC6}5d2A@yc=o+lq~)O z_Af8Ol6e`BS;c2{i5kXhK7_29hPZ|I`u^;Z$JZ&>*qVXMY<&Dh7MkzD^6u+1_wPUO ztoRa+(~ZV7OJ^L>;)JT7#-qhb4NMPJLYaFXK=&-~cd(`yxj_Z!?e2np3zgA9`v{@` ziHS$cOYTkjZD{m70c=?c40dVYqlMM*K4KjxaJ=Jru{?Hs{s2F^+i_*uQ9QG34SKW; z#m)mmaFyH>sHiE!tVO4A#D*-4Hv7Ww@t;Cf%Nq!qe*r(rm-F1Vd;DDD7e1A}gzMiW zVapn69JKd`+=FTK&?psD3^c>F{LJUTs);;jkAIIhQJnqfDEygp6=LRm0FNL=u#+#P z9W$co=9J^O%VjUdUUI=Ydv#C@JWjuS(4zUp@;nDcm!6R_g^k=j$ZShxP zo~i2W^8yt%jqtncsf%$v??*iDH<$VtY0zl-e&RP)5C7TmbIo-p;OaPCtap;b+6~X3 zX7hP`{Ny6$C{?2UheN0yw+-_j&qc=-2Zd7O?+Cet+T0)YN6^Imf?dye9{R!A7*I^` zmhMPoxmu+DNCH{cIgV)8xgpA{V5h=)paz6psj_6Nl_#<0-I{c-cMH+rXBF$0$T3qZ zNjCi3AbJ;W!8flW@KQ+>mhip;l~o$xf5VkZSsUS3XIYdz(*P%UUXw-WC3t%y49?tD zLi;s?U?lMxj0($WoOC|@y{?RydyT}1sk&HK@)?@X@pg;J1-NW8-^VJjhf$w9>CNlQ zskCbryM8p0Z8{dg-tZc}@ZwkS^V59ra?`*%gVWG(Fbv*i?f`4$oe=$F7?|0q!Sc#K z^oooWbT|LQovVj2JBcAoW8fH;?n%Ubw~yeFDo-{!FNke_a)`~amJ@IG{mR0($cej+ zcGIK?te@;svCPc z2GGlOIQ#M@0#s68L4)xN*fBl=kg-I^*c4yTQkE_7^0 zK4yzfqgGcdvPXUFYSvp8I!#KfaCs54i?n7ZwG5c_WKXzsI0bfYD21!%4r39&yJMH70GDsX9V6TYmJX5*Fwv9Ya*Y}D=Z?7!Ot4Yyk3v5)pxwYHYpE`C5Q|6HSU zcpv?Ogo8+%=i{XNI!vNehxvGoVEwV<(U^BTK3lX5GlQq1=#exkYR};JI2q`#xd`=S zopD&aoVYDUPQ3P$n%K0xgpC`U$wr?%!D5Q$^WF=d{h#lO4x!nwtLp?@(0>M9??fo5 znvM-hJMh}{R;D`M4Jt_v#*Qv~0j%Jua*ua|ts~@?!IZBUxtE7^eP2lljzgs8l!?r+jh3MSMo* zp71^<*j8hhMg!_EK94#p?&13QV)&}G1B~o6A@8FVlkn1KW5$}X(De-G7>ID(;B*{s ztHo~Y8_gUGESc1UV-U6F4rn}UfH5hCtb|&z2-(?;_-$dQtTwRttX-_aI~qQ#U4SyX z9=N`BEBY4rVRJ(sl2-w^U*|CTE9IfhGp9eK+k08~u5yn56&qSp*?9?7dcKnS8 z+w^52+mRE-#vR>AH`o->bpJneYo8_)zcpYJ?2OpbVnS_2VmvDbf|az4ToHhVqsSu zZt;ADLyGQT6}gD@?G-r0cpe(G@5U6}3RG6R#3q|uWaCQf*o2QZZ2wp>6HReuz3&X! zP0o|iD#nZJN#YM|q#BPaZY)Q&= zcJ62}Qbxx zVMtCg#af8czJGBpxpwWbZB) zG0D8~;J#oSof#p)_q|Wj`o#ySJ$XsubQMuakciYc7m>yOzzUQ(*1<4)*A{RDMsGgCo8i zL#Z3i;vq39Vvnk`;{4r(IBt6?K6$+vP2YrI=f6vMhIbiljVfZ5SIgP^xo22;x-ERI zapD=p;Slv~5gxp-6$7*RpJ%s^HJ;F7p05w#Nd--88~g>1-IDli%WBr|lg1A8L`49t^z$Qe>fMNriY%9BXjIP`j)Q&gpn9GvuIOCEdmO&g zy{b}RUuVb?Kh0u(Qv#S#Q4du-8~{bK2AHgxfM0p=%qoY`5RlP=v*+d^b<#zh!EZQb z)Ir>&@fRW`LfDUl(@f6&H9MFUg-0GG@cxKwEdOlEPK=z$c)AUK(5%Du1A}^c@y zh2i?lDR}>kFOD*Oh^I%q!Gy^Z*nyBN8nfgi+4<};X)RcSfpbIQ%I1gUBah-fZab05 z?fQa5AdQpTG~ndnQ*_tm4cMZn!{)byuvU|0taOlLOPC%@4!%G`c3dFK_eOGV4hz`2 z8Z+i((}b5Ng!0@GZG4m$1G*Q>P~wPy$wr2=U+Kr;i&!7~-bLcCk`VM5aTeD|wqoPo z0VuSqg57c@@G!g-#-+y5e|My5Xv96VTc(U}cFzTq)ixM(@d=2RH$dyUO5A&`6?m2; z&GAlW^(%^*{G%c^G$Vv9>2F{^+S}NMo+6gGbqUi|J%`_#`oK=(J=7OVpwWh87IYw) zJ(#_gb?Y%)X)eSE$E$lOhBnZxxembt!!9g4lrA%j+zBP4W%0pLIkXO8Xue>a z`1UG8amjK{?D~&q5Y799Kd327_+fyXt}Q_yK^DG0Q-(zE0sefi#}1v)1TyI%jaB?Y z`*`+?`6Xw(zwR{8-{jwYuZ8SHLp1wqY=*0K^^u(zi&{csI&`Eg&1@v}c{T5?^9w+& zr>pRuO)Fk=xq^!88u9DbY8ZN;6awbGg$cViqO->|d^^kz4euVn2jvD>DByY7^BP#a zYdQ1JyuujSHDiQyzheLmbk>HdZw;V7OA^171Nyt1 zWv7*M*#h_TY^;6`borNnreYTi&AG&E8}nIJUNyhB*@;CR_Gnn{hMO}2uwn8nJTcD& zuN+>11JZ$L79zzAXItQRix+VCvNGO0Q3CX0D7fwkfwsr7Xms)^N*0OOIyp@Ywoyj; z9BteZnSv+z^TFibhj524fWsF%*yOeoW~rrM_~05`+p-d^q+Vb`N*yNG-bC^?4?8v= z!2bm6u;$!l%<`(n>y6#0y|EC7-O5I%=cn-ue@>hnD8SRF7ouNI1!n3M@pHcmShnRo z4i9U?*Re189mHHHdjE);uBfLq^9S(B(COOrNE@ao-~?_V>pN?7i2FGo4K7=x}YC5>IG%w*lH4 zS)tU;omh559Y-0A#URric>Rf^gj!ME9xsCVZlL&3lQ+Evsl<+@)<6p55~<3$D7X2u}N|N0-5BMB6R8EcR8K6FwvBzaN_5#-i$9s+&ZB@;|9Ow#5mA`WstBbJflJoiF(sweq3 zb{>(qV?G(no-F!4Qq;aHhjZC>gBu$oMfSQ5p$5A(AxgYfH1~ikiJP*Oe4pJS z5)?X;5ogYG=dRXrJ2q>>{@b!pXq7@_r%PdP<_x;U@0*Cp4J8qA9`x(C$Mm-!g+D%f z;l&#zQXXRqE1R5%Yi_n+vPgq-5pU-l{r#xmPCfnWbc~Lvjv#{3wWKD-0v7+Q6IGk9 zCPVIr(XU%mM4OI9lcWYS`YR}cEb1yD^2L2L>iB4GvF}OJKOE_=J#$HcK?p?s^M*EY zIpOvlfZ>rlV1AK0IW)7K){Uy5HAU_srMc>Kl!`muqOVL}?X$wb{?g9P7 z&tS~JZ_)l;Ikatx46T?|LFLbu0sq@DC_$2j_ZZOM1y%I=h)*J4t8fzewv=2;N+F5? z(o}30Mx635lI(Gj#r*))W+zCBtXh*9NSg5-&3<_XStycW4xd7D^^mB5!jzpoK%&?h_J!X5o6_^-Bt z^Yi&3Dy?ZDDwEa8q(`ekUwZ>MR@@NzdIXV-`%PqHtPZK+mVrz{9MGf(!bqi`!oSru zB2j2IX{!t+K2j8xc2A?%Qk_XbFGm)A4kzYK6Nq`p8uI+DGbv08=bW+*lRZqH3X9yq zY+E1QJ9ZXvpHoW5XaQ{4m%;V?*-5^J12G@CO17-8Br7qMS}AeF^P>e>G?b@aOp>N; zeui{Irz_lgJDX@-nL{(}DoCL?f=GP)Ost;1pv9J2q_*KKH1BZ-Co5y_dv!Bum>{Al zvwi6f-!xk9H403`Hk|n;S$fdajYtINaLYD|sIhoCaTu@4rRwkwpm&30ccd=~HeOBQ zmbTNXb9d;pwX5lmkcaSa_7Co+-Cp>6)RZ_(N71f@D%83x{M)qWr(Su2nR9+@TV;T=Qe&;2<)1832g`TRL( z{9CfA{R?NfG>+7}71N3#!PLX|5LB&YApS9edza)&y51tG9eg5OCAXd|j6TPmS)9i$ zmwHZH*G&i8OQX4;$&}l*^btK)El0}VzT`IluA(>R{^k1j*K<_m4pr4k=GMBOqUG}K z^g~M!)m2FYwa6b3ywrg#{gMXz`R>DrMK_3o?MOcRdKwN--9zT4NAcf@_vAzT5&~Qa zJP|0u%ItrlXWp{3-ft>(S6L~lfK;x+x1aoNNJhPf6wODqg5H=8mZnn5Zuv#C&Y(2n zR6CEX80JkZ9!21y%=uV9+!~W?uCt9bwQOR25xbT30mkrIrMc4gi(AGyC+5ly^) zLDnxB3~wogb*YD;ZbT@Y+IN=crmTi%zx(Mu_drY^9E+lkn-CT5jqQUj*wV}UFxUQI zuGM{Pj@xH8@m3eMcjVv>yUlpGU=H@oRY92#zo6kx7PQ7)go(b_V8sD1{Oi69bDE5B z(WH^$X06fU(XzT?lZX|NzUdo{RnDP0+No?(gE>p-`h{JNnkC0{4WM!^G)7M9z!`}HR!)5jY%3B=;81hrd^E02O?bzzN7_~{(dOU z=iCcE5|po~#}nztvB@eLZ`WJk)df6fe}@*n`qzP-{GHjytPVd{y+PgPyV$b41f#a8 zu+6fXEOUzu~>k(?JNq?cls+K3E>r1;?vC?6kZO zo5*{NJa$J>JHy%3G=t8#{GJKZS5=#DUqJR1e*~At3Y}MLAmX-Mo^q(&T#cWA<@jMU5OxcYw z`Nc3Xo&T(O2cW$AG3ZQ=#?;OgXzOo<>yO2u`tLM6{V)S(G!JLrS`}Hxi9WRdzLEV$ zoY71lH}WDMpng6fRWE;(H^B*a;Bme{R&@WVCOYFAGiQFc`rcmFME_Ilx0tN zXGd~`6np$3pXbi2qSg&H40Cn!H43{>9X?IT=m3!S5NGl2Pn7&vZQ7tHh?A3#CujmykO%r;%xoU(stb z=fF~q1EWWC@cC~eOn5q$|6el}Bg-|g<-mCKRoa8qzM+`las;J!Hk0h7`9$%B7S%D9 zM%=iFcj7L^D!rxbbjT8xzsrt=9t?qT@lWB|6*Vk(lV;O*OETB}z4)+Ro~;zU#b>!s zaZ@sZM*%0G*H#T5&U%0e9Ut*%;a^-hb3XnuKZxmLcVqLvtu#7aA3~eE==P6NY|?85 zw%zg%mQ_E7#O>a&cB3>bp&X{whv1U$by$#+MZJb4lB#X-B$AZzyw6_{H^c_JCmvyV zA(E**+0Kf)jZk5~BFbib0+GTXyngo;l!hqbRY_0$px}a}vpF_>_h4}sE zE8N^Ygv}`bij%)2a--~aa{uu>0Z9iPs8v`1<4%==Pth>&i_(EV$@1X4RDr5IeNVJn zbZAJ_B6jcAO!jf!1Xj~7&E!=ySk3sc%xZxwaqny*&x?1^o3h4`;0AEPI}~1iFos#& zIw&nEh5m}sESvwl+PCy$^WHQZHZcMlr}&_K{AyU$69cNc>tOK|3Hs`{9i3X4L?RES_4C@O_Pm&Q4;)a!%~h32>{F#uK9CgFlB1>k0O44OOZVfjBHv;8GzDJqPm zkDbA!XB)A@o!V>x-}Bg6c?0%UzJ|k@GPw3!2TYBxhc{ApFs3U5Gmk~#yd~e@-WheA za&7@u*iM1}?z%y(b2%*Q9D&MxNq8*VgoQmc#D~w1W4MGns}1`>mS&zHN_+p1`2JC} z@^22wo45t)PRt<(Ep+MaFXzBOP?Mr zsj49dpS6>Tb`o?{PzEIQo1>!MA-wi>BQ#wYgCV38t3I03{e7|Y<+^a#^F#wW<+?!9 zIu!SQ8cl*u{vir&`k*yxC=Pz^0_+mt_#X*mKy^EPC}jrkH#*`w%UHz%ISR)w#u z3}BI?E0zE=Toxa^*#^G|5n1S$z4L$9yX-)o9$@77T~qi^yRA;RSnIITB;j?qo9FkwAfK3`A&+hamg5(Si;>7bD>ztGl2>Np`) z694hNg5Xdkma)EpDxVdAryt)F^;V{XCfab=i*#`GO-ldbAP`mKWR$eK3N z6H%Ao#fQyUG5)3CMd@YXoVyd~=t?m)T_s0H`i-ZGjCZ(>_ybXgk}<_1o*dq-LoTUQ z5s5SDaH(z*7I@WTcIqP-mN5yJvJ`wjY9w9dx}Bs?8%G}TozEKOczFLz31_&*!kwO} zkZx#BD;zY*%8oci{V^SmUW zcIy-Ppp=O2JABB23mi2&&AYyg0wJOD1CUY4=$1E=x_1uH{dIRB=lw`}tHl$tnr(4F z(Vgnl7xI0R-RS+yiGDL^B7Ny;RM#7!HO3O+|15yw%Mrw6lM9^_!r#I6@m|BwSQq@L@vm! zBFSq@>Cu>UZoSh_@^pAQ{G4k_qYQINN}D^Gmaqv8$T9TFxr+6b9whwDNP2ZXLMX3i zF}xvyhgRKCH*FEEcxOvZ@+0YoTY5w}H;+o~)4;%LD_Zbc4BF4a@aOAz^14QiIG{SY zx3HF8u5RbM^1XOL$h-YUR=|^b-nIAa444m#;{A69%-zToJGKp=OU+qkK4%@-MfeW1 zsRTTjF&v-d7GRgVE%Vop<{bnvG*@jH6!o0tCcSz~u3tb{YWN5a&O8e5hXvz2{ctqs zB7iO`_KQKX(RPZKUXhga+=x7fpKKt(+^AucGTe zz2<(F?B=e#l7*p>;DW&kL`5o71QK{PgmL-Sk(@VMuR~Vf8yC@PKa{*^zae{yOeJWUa=N zA)+m``qu>R}C_AlsLM<;w+iNUUE6##ZcU!4=h_tO*bL;R5s}LTA=tsA56V}6$bjh(kS&Aq^agG4dl;ppRYXPr24bT z;*oDLDepT>atx!(JCC4R@L?$6-|Ge;68KYc9u-`Zga1ZH!PD1wiA1p}sHJPuRFQ_r zW$Fp8>cSv7@^(6T_U0gcqWS<-e+`E?=b^CYKF_7giYB|7_-f314VLV?6w*HQgI`)K z)n0Oqn=tW|@Z--WIB@UjyOM-t+LQ&K?t2dVmx=Pn&Mc z<)WuLQ5~EJo{x`Vr*a5ob;cxq??bw8`xN+8^qlyyg)}}Oh&ujy!+D)b6Q*ov!od7S zP;D2|eYb~TfaxKca>$a*9KH%Y#%=`9)dk#*;b&mcV+p!pmLl1mG#65BHQBY#*^N)cWG|RM0mVJNc8&c;mQGEy_LhU*uIDqoEgc+ zI;_B}vZLU2bh=16qnSuF`NI}FHM0D954UqzESOxKPAjEzM5QGW^tDqHnN~Xz4*eX0 zcDk*QFySQ6uYE)mr^~>ekRx=2I2?XW-z$m|pCQIFd+6T-%5ZzmeeUT9YwGDE4T|4+ zzPtwSJocZ8g~lS>Ts{fSW*2~qQ8;y;5XzN@y(C}7|A}rbJVst!eL;u*x0dMNn@FeL zl7YA}ne;@aKX~h$qvg&3b=V41HNw!6_d=EJI7OZvt0B?`rPQ}|2QDt$&ofwF!1B7U zFtJCO&siO#Pj_44#!W}5(a&n=@9W3OslVw(*@fiSfmdAmik0+){cWnC%&;DKS8>T> z?%k?X;&kO5EzWqyo$AOWGLEWZJ5d>PEm07MT$QmttyLk`&y5%V7ss=>CHI2tnMgQr zb_?#G$nTG1Q_wxTo%L@#$Yi(cvmLqZ_`cW$BTVCP`t}X*!PA`P#(f3f2}4<3MGyL2 zO+||tPVDP0Lk3!F@nv2t2D+cdAO9&ZEfyx8T4O5CZ}`q<%d6M{(>?6mR3$dY|1N|r z_yPwz3efMvJJzRsicJD}cJ^YVSo~zIIAEKb_)SL@EprH_$_tmnN%^VRDs=+R`g!A} z5vJBQCZEKj&!yt>wK3wKY5$3X#$03FbFy*jW*?N^ri|D03-EaVB-A!f#KE?9@kXym zai{Kdv3buYrl=dsPHG-!ch^a?Oq(k>Vz?UfF(_l-l|z|lS}eOU{V_;8SA)}E1xQSJ zgKeW{;$*&qm;dBH*5&yE-^5Fa-IlIpDQ@$a#7tc_nCb)a(^kW=B{6XKZYpz9k7oUM zllZ=by7+{@qIhB6OE$Ok4<^-k;nrBbf6?tER(91A$DDr5-u)|Ln(fEgxLE^uC$ml* zToW!<2-X(Qb2uYzO!5{lGcy!#4~%3%l@nRO4NdmjdOW-JvI}PgW@67a5Bzj35<+{; z@KUIgxXM9QY&5i%8NQ9i{gHlHHDN9)niN=zkS zKls( zMz^tz4w0<0W(}X~h{8{bbHsnd!^Nlbnpws+9r4@V7M3*o6RW#z#9r-}X9woB;?3bt z*aF>rcGP+rE0ep%7JYum`e!Gz@nSnzwd6S@)z5>?QfA^DOEvNP=n&?{*|J&pbXfZ= zd$vvTAsaqBk752|_G9}w`0IHbb~io1uUTqr{IC63*ffASxfjGUA_K)F+g|+jOgrA` z3&Ens0F+)ii|=2KVVe2Yth4Su7U%82RYvK!ZqXcZK&z?v=kbkf;jjd;)yq(EJ=e?b zWoFTNeok~h0G9V z8_0Sz?MZk6P*TwQB5JT9P<1Ymr!Hi<0=V>^Ih{8_Z_h zKFb1>Z{f!=3V32yE&ICNhLye?&g5=(V_!G2nX&pTP&b3k2&0UQ7{dN&C*j8E9=zTg zf}{2JW1r+KxOQv^eV>1iyyzN&2E8Rz%0r22Y%nHObE>!*!a|X&LOO94G;nesqPg>a z@o-zAnwBWv6)j%nP4(<{bB)tIMG`lgxdrtZqFrY%3T|lf`L7izG{(wEq(6KJZX0h0 z&-nL>FYlndbfr;rD#=T+^?z*u3KU=Y)lTI63#n`ud6P=jQ_~c=y>6XsUtY)om4nB z_>pk(Z!7AwVFf28ZX&mDZYOKk7?F&VHAG#vP$*Tww=NvLID64hkTUI|lB08Ju<~?j zKXV3!9*)MQ|051Bl*rp1p`wkwcZuz-Qf`f8xUfRQ2P<6aVC0zz#P5+M>J7|+wByp` zvaS^BJvu>KAGwl@Kh+TX>Ja^U4!G`)t0LEnN}|zPXDk(}!br&9ERo`>K+&u5%b@#r z9F_G@B?lav$*Jqdxp@~=M2l|nxy?06P%uWB+IIXPi)tT;cxx4Y~k_oA?yjQea#q9fu3ziuwWk@qqnrf4GZ{@O-^w_YPgKW>YT zSxI51*L;{av0Ajo;U0)LrqLT4=MeGd25RIto61;7(j18rGCpB3*{2c14HYpsu=odQ z{+K3OeW!%J=t-rC;rhInstcY4ilG188d1RGacEx`&U2&AlC-%8xQG8DxqUObMWRD* zsoY3&nzOi+d&im5pOfUd*8R<#lAZ~Dr5-QxyE033^oSJR`7;Kt&pb&o0#ZqNmx}0V z-7A6W>AhT$YJ)I((F{)Z;~1(|n<}!-ixBNxv>TW0?*^HouJ&A6*^7~A3=(Rmp?5ruu=*$uwu1^vv?n$KsyBx`w-U4B8o+NY(QKomLXLE{? zsu1CoP9t=yY4xN^DtbJeTAgYT(*9kXndKSbiB)q2^Sk|deo_+Yj0_XG5*NDV)(G1F zS5fqRMJLJZRON0T)a9mHcEPXbbD?BwH|N~wN@~@oa*>}FaF$Q(pxOew#RXbzib@(|TGxY#lL~ zmoKdSn@#s0bta8gIilrnuA}}8S3J>~jeB%znS$vd=5^bd$@vFShmj#P!e=45EM7y8 z&VK_x*Sx^Wf75W@55Bh&dk&ABIuApC=FqJhwIR(#fj%9l2W{i*arLPA@WEy*d^%}J zOy1AOpsQo?$T=%`-#UuUzc`gBI;OzNTz_=GdK9h>eMP5@nFJME55a^PN*J#8ljbb@ zKvYv|;YC6rE}}*QygCt{+8jcg z>AJXa$w}}%Rtk#tb}%Y^2MidzfX*AX5D@j1YLCpo)OqoE>S8ps^xuh;Y04~?v+S=@d)PX4r`!)!-TU*H`s{dE(rnyrN}NfC&;RtbyZ2(V)ldEVm% zRtL(^Q@ES%J`@aw^Y1}OF5=#GKH!o&PP{VP$g{~#vXFhhEX{8qf;J7fG@=L3B#eN- znJE-|5{c8>`XRbq8rQ_wAb08|Sm}MmV3zNfKjSDY&^d^U8>^sd^?dT-7?*_aT>vuIZG++hI`)3WkodA*-0#k;gL?fSUUlR^(5w)0p3emzJiSL&&rn7Anq_o+{!04Ru?wHi9s<{t z*I;J7m~}j?233vQ@L}#YIKcXVlXO2q(ab#{Y%PVC#2T>J;xqDF^?|`sQM#>^pKh#? zMIryUjOXeI#%Ev^*|ejCIbE>~EGI7Gn?nmB^x`o2uSN=!PQ^2-GUdc@LJ%Di!g1yP z5peo(60#Gfv$u(hhz{H7z*sD&;(06)k~#v%zlK4@k5HIQzJrHm0?4y>hZX7K*thQn zuDr^@$Au|GZQmcrOE?KWebXR0|0%O*Nf@Iho4_Mk0<#I-@6>B`uq_WrAN|N6~6eYrxQJmF44w3Pf|`pNzZfxM*IN}Jhs!ql!ec+fv=p1 z9FGH!Yc1G6s!#qh?(ii@8{ZU2Fbh7fg&iI|xG%PZj6JSq_toWWUn&Epsw_aZO%`18 z<^z>7M632DoO-s4yhc%$LFtCC%)UX!^>Z-pX^wZ2ePE1bL+lB@fvR(OAxe2IVKYeJ zq&7hIZ*ienGAH#hXaM$qkT7L!LOZy%&(|9u`+*YOCLVw|bh(B#-zdoCC zIGAwiF1UML1J(W|Brf(ii5cLByZV>$@s7>7s<0NOy9)-s)Hg_j9n%#rbXe+`c1f0&w?&Je-x}0s-St~399Y!fwC?4 zp!Yx;HvE`k{u7F4&bwJc!x}TP%)t*#pIfrI2N!(l%es~Q=2O{wGsv&gY_7@42K9a# zVAj4y+|W};{r^gW?GH)VtSXPr8tp_aV2C{a-3%r(KY{cS540;Xpc2yCFd#}5cdh1S zcXDCG>{%kYS1AR{!|mbVrukGa={vptx`%9RIR|#KBAC0S44y4I0_Xd7VDuq>rpMqI zlhca8F>ZvQta+do?+$qz)WNUv6&YZk$7=2ypd#@zb5&p#NM`Z@(bxjPUrSJM55?WZ zCirYx0>0bV2tzyDA#b5M3XBv%d#Vj+%(2A3+Sf30%QDpE>4VO=*F=CLLYS~A&hF7P z{KBJ$PhuK?*i@3rZCX&cp&SgiE8~BQXW(5W)(;l88eXTYz{;i5Y4*P&@@2pW%MO`R zh3IAYrN{tFHx=V>;S_U$_4;fse+_Xyt;A6~9_Fo#$KyQb$kv`}2quO2LS_s$tZO9I zu76p^NfbJM=0p8y%i*b^0NG!1jzk*Fh5d)DLHUIhsv1{ge3%PSE{cM5r375LJpuxs z3gWH?eLTOtn5^`CO_sN-;L+gakQ)$-bN3iiGw~LZlO_&-Us%G$t2>G6($mnhp%#@h z{Xq2j5-5^pbNKR|xb1B*Qqv^t2&f@Bi$w@+bOy_6QHE&lC3~$?!T;$ckbgB7zQkk! zV_{3St>!_CliM-#B?rdXPRz51%W2`JVpwk_1W_xp;n~eLBKEBvAL@BfqSuS*8#2+L zv;!9}PKF0%ZrH$UO84Ao0GDUmsD!dEH}nD!awmdl4JN{q+Wnxg|0322MbQ0;i@=$e z{hS*%K+q{gc+r~*?ziJ{?LARQui(Sq{F+#KUjx2>P)Cmo_p!fVEA!H8J>%ePKn!?S zFk5HKlkC}*+WohO;3d@L>NhJ9#(88IU;Kq>_ z3ku%KI%pNA%Q=N>| z-XRWq+`+X{0#A+$L5{!}qdpP{zAsKN>sQG!>Dn6^eH{bl*8EALGB6ty|K5hE|616* z^)A*4BS_4$dr29eFYxmHVyeP^lZg~NIK8@nJePS7O{rfYYnmA+ckfpQ0_8z7?IovU z_5q?i(-dTNqRI7~Omy1Vj`xNFN!s2hV(iC03twq-ig{DWMA0v@)^-3aU1ZSZ-AVlP zVL7NjDFowVr4ax08|QCf9>X8`nxs!kGB>6Far_iT$%XT0$xWeD4rlWb_-b+#KK93f zLEj{h*ed3|oDZ=*rw_-1N||i|DsVI;4St57RR%Xx8@mt-}jF!Q?QAoHyv>@Ey| z&mF$-=JywPx9KXU>~1XS7+Od!ulz@X8tUNqKL+(zOAgrQ4(V-J!F2Io;uLq- zkb~Tr#OCfGEIu270aqm9!@zZ@vJpnF0W)S{_yT63R6MggbAqG5&TwnZXXAE7X)=Cr zHnU@YF6mEEBbRy)k>5wJ!04}q%)8kunD%%4jG}rDxqjvcbIU0e{9FTe%Q{kd=+a$8PUEMu4Q{QZPRxPx6~KxzP%jW zir+E!W&V&|yS!P?su!n)eO78rUL|)86N&1^JIu{Z5n!C%2^mA|p8N4QOkca1bKP4U z+EYT|@sI=Y*r*O7Rd0!@Ul4PC<1+G1)C7iQjxkDnf^c<2p3q>DmE^2{Xo`Ew`A+9cN&}tWQwq zFu?p)am5X$7jaFu3R50i!BAxrIC5hdsmbbxg_G-X_`hR}QOQfrr!0BmDtL>@m+&Lb zhklb1wev7v^e3eAvz?XW6JXxD9;3Od=~OQ3R$H6H?2yx9rs_u^i*FhJCnrG7n?m6P zJq53}H=yv!MzV!R7+RLF|Azogcq_9BF0Rr<<(?Phr))f#!S0BkA5J6d%dU}1jcKr3 z@G^+-{e}}=7cuMUCo*232dlQ!f&brXG$`xD{T@SDn>!s;lCmLV=3bWjrbTkCE|QSq zPLerMjzMa9`13?Mp7*w3Z!AKj%-jS93^5j6WRm0lRUPYuKExGoRE39sR9;1D(Xh%rZ9 zi9jsfHjVnLN8-}+dH5>Pgx;DbM8+SCk?~bZkftny*(?*#Y>5qB$FdT;;1* z!F7DAR*lymHlh9U3Szp$0@Jfq@{Cncyeor65BjsQ+Y7^baA4=sV3TXhJ1>7CB0?+h0xZ?amSm3$< zo@OI*!?KBrVl1(ZoDD}4>WFeL!hz~245|pFe-hHDJ6{uAy6ZX- z?Q%^Lbgzrd$>gB*muWP}@gvRLp^C@rosf^~gO9qtK$!n~XjrI#pT&!recoM6TElyC zx$iE_Uh@Dpp1%N>y+uISb0_5cnWK?XAN0cyD84I!iZ&ul;5-QhiZ(O$PK4#P*4sI8nUssC|@xcrJlHThPUitdFA zDrR^wMF?W<&4HMCr$N0@fEp^8QgY-B%}Z8-3v%YLH2oV8r$q8=c@r6GD@5fPGI;o~ z7QU6PqE7ZRd2Lg&SeMsy%y~yvMv?{bj{r0c2Wn3bZawg-O?kP<2Zm z?t1?Ojl?E&8WG3Lv9owr%7;GJJdUL+Sf<9(Ow^3k#lNd+kkQ{u@6OkyZ9=nY_#C88 z7IDzO?+RQRJ&D_zH(&tEPIVj><>pjaF-{fjOeyR1U6u9@`flz)gC=Ln=}$zB*EuN3 zcC1!Jm_qzeDAd?w;#abk9%44rqdhS+?t>mT_PZ+gSc)1qy6hAYGcP6XLrKI#MGkYy zzQf^N=ir>{E7Yv$!gX1XFuXkrrho2+Y()utbSnz3g|R-}>k05qr38BTqM%x8Ep)z; zz+DGNpw+w`Mg}!7dxkAOWp^dQb&Kep!^Sk1v8Fi@zmdnJALT;c;&!!Hu(nMK-&L?I z$UDEV-1sRzz1o6ie3$8tg!7c=O$NO^N1RA**hub|i^8a96lNaZgW?T)aaX!BcYEn? zYI)-ueJ(3UPaS008HKk{`A{RIm|TJTft9fTBkSMfYk=*$d!e&9j=QZqmMe2^Klfy} zG90g*2kW9bL0e%TN)7tr3a$T;Z)_I*w#uBE30TvrMdq-jY>FH=TnnSgHMr9+54D9a z;GF*b^zF;T^hd`T`Ye1F&eYUF4;Mx3db^Dn1ABY^noHKOIg!bXQAnOIf@+&IXpxaJ zeKlW)3KZXf3x`j`nDrRww6nLR-Q9S1?`_^%tqp-6w0Y&ier-XV>A=-7Gs%Ljt0j894IpH4N!J1C=|~VACwj-89J0ooV@r z8r{DELbu<++6$7%-4VuZS-y%p96G8u# ztiR%51TNDMr2B*TXtjMS9#W{LeA~{^)jYXW_St^)yB&!mEcd!Rynv0TWMaJPFSOXAbSW<}h3NtjG3y0)CelDj*x zYO4fVsLNu%o+j>YI1AoV`EcUvb?CC+M*Hr}p~g?<(L3Lc(JicRLZIgWWwseWf}u2Q zT*?c##Zytkv=sjsm*9q|IJ~qn7Wb{$itk=Z(?TJZUv+dA{XKOV6XhdtzCb+2k4D3m z6$~tR7Xi0+zK4wS-=N3vFEk_?1K&Rt@aTO*UTx1qem2X0e>51+Xo=A1>AX}Z;uAuq zDGsU6#avY*d?H&$Q|IN=j6;Rg>XIPc&UzabpH!y&U#AdTCF$1!!XjkXVeTaYwqgAy1H7$ZfhD?H_~sLvCqL1QCBC!qdQ&7O zeAK2N)(bM?BKn-1g5At}0WUB*dl`5?^JBPL1@5kSkM^w^)MIxscrADb@lv)}u(FWs z-`PS=-<<g@JMtGeqJMuQ{$@0x6cE= z#$;pVNIuTZ%tw)sP@+6K4Q3|`;~KHY@Yk&$6xN8KiqIDjTGkAAv+Ka@XCo)dDuIb~ zKEe7bx8q~w82s&9jMb%^sPjY|&HYwJZRg*InFcLjyFmx9_Y&4y7{GisuVBoTMbJ&m z0`olgvit8Mx?84>UKsD8TbnEJX1M^Z^JVuWV(Fyyz6Bh4`v{I-NG1-7F+|hD0E8v0 z$gjFWaEs5_~K|h{xPp&*>2)ADAbm&f0sd*HFh%6HM5!h0|`tqIuFI{btPyGiI!v1Of`7qQj9oba7W zAs$B%+69&Atp9Xq?+SDJD>Vu?OP#^ThdyBb>RTj0?=tzSt<3IQ#qhj|{QtXPsvC1b zYx*mglkyW@cul7rzhr3rB^7#Ns{%+Zjf2n$0d^F!pga0K>BfZ1G$f7>4F;94Q_2Mk z*ttrPZp88GlNh+Kg*kcZ2V;LRgJq~?5Cf+~MtJxlS-wdDmR&OktLzA<^L9fKrACw* z6{I1XZPzDSawi!ZJwZ5Aoy%lHP+O9I%kU^&=Sod%}M3bq`L zhLpAgC?=2tHNwU4Z}TV^`qn|mFC!di&%n{ZFo1a_Ft}9{+FZfohiyVF`3Q7EtAh7=oiOR?z({47x z>XIUu-PHx+L)lPySQ1Y&syLgy4HaCOMV^oBC4Uzck(CG9$aa-IKwB5W@;BPJ!9N)F z=ft9b(*roSzYeDEO~N$&8(`VQvhp^Y;>V#Gbo8DM)IVuMi|y8M>75bmuiXwkfyQiydK+%ljzh`f z2+ElqM{C}vvb@Mo)bGp(n$|o-#Vx;)fV5$9bLJR%Ri%x8W@zIwVI>T~ag1FnMY$I> zsrSoI1kx>`EaeXPmN>$O0Dov&&%TF#`bDOEZWFsIWAHEa#w7(?u=eK;?7!Xx54^v? z674m(hh@ymhKvAfBEctXN|9fr0sD5o#ijN2@aF;BlX6CEKe88JJ`QEMi8**F@gsHL zJxYU*kJI?#0utW#nEW#mg1wJ7;o}t%_}Be3)`sq)aWQ^$>@6aVv0b?$`V6{UTuZJpfWfvN8yUaP>mOH#^%@ z+(DlrYig>zgt~byrv@L@V9hgQxP#lFV)O(mO*@4vxfhUArbR*9ohrJ;(5-K?u)g*@ zhGH5PoVrSWm);O8mi-!q zw%vSqrP&Pk?Tkj3K$fRhasq{_c({kR{-HHrf6&|@2YeG@jZWVa@s&go-j6QFg+aC0 zV#ooxufY&ISp;RW1{h_w5SRbZN2L$9=mL*Nv?HyB7PCIwV4J7IQ6Rn!?iEs{%_TocQ+F4S02Qa|IXl2y9aQmQ~(ccFvg_?Ml?LhjQ*Cl zq<%~wiagnc9MK3IpB{~FT0t0S6N=$Yjx@u|m5$1}(Jh;YNtVSulKycP_+(m9gLF#w z2D#C}#ccMx>lReV55f&YSE^IKgf3WTM>WcIaL+d@Ty}3iwmXkw=5Jm)qmG~2*(Otw zbw}vQHSzRcxB{N#{DD_SkD;&A5tS@Fai&WW3cnFz`3TCGH&Y9Tx8&f{9~tOS+Kijj z^=Nk-9=P&OJy?Ao1A0Oa z*X9Y}mC8?W*z+AMnzNLas;{8K{r2>nl?p(61(_q?MvQ;lV|OjR*nH$A+TJS0vIY65 zQF#@AMVzMI)6Y{8mX{I}VuIdl`7z9rA7k4zAgtF1mYgbwQX?gnaifX4?5%KqQyc{F z>;^f}BM_vJha%Ima3MDzx9@7embq2fuHTAsy>n1X)Ckp|vwU6aR=PQ=iyo4BPK`Fi z;$ro9ymz%8oxNY+`;XIU;SzglZmGe(gHFJJ(Ii~nc^T*W-o%duFVN`5K@3+9z#|u; zvF5`{yfJwc8^0&By`gV7vXG%?F8R>r+o$LmET3>+tO#tpZo@@CajbOy z4RSJeICAkcYIp0Zpx6mp8n*F_I08l{1-*13DB=%LUe|QJ&jU4M8AAL zM}H>sqpyl3elLH9ap^-~+*}2i=CTf7?$2&ss_=L_ z?K^pimMWiTnWAOvZ&NeYWXaK#1`FzK>r8X=pOJZSU1acw71-+XQTwkWxV&-_B@PEu z)sq?2=uk72nz#z5isnL@m?*7Bn-(@RZFy1Q&mJp1%ZY^Y6n8n?cI{@u>Au5w7^ZpIH56DYcGI zq@|YWsB8HQ#XN=Sk@XDzDGJ1nqGa|~bRIQ*im>T!F}~mRna=zBfo@pQNnLNgfJNEg zfgSrXD>;Q4O}|U6y!p7(IOV9EU4RLC50I;!hs~y^kP+B{#r>)@PuP^EFWf<8hVOtL ze>y~uZE z4;VeY1N-Yf!l7Hi$ZtFw<-U93g3}qeBI-TXD6#jZ%mXO+;V`ys&P4OrD13cQfP%Cu zt#K8^KD|xoK2n3SqKi?_<0U+cpCbGZrE$8}LVRhFhqLm7FmZwpWomOt=@MV&szwQ? zv$Krc|7ee;k1wELD-YFRZ&inuWsrc3m!zU$F;x~YqVhhYI7h?`b$n)`)R+f`C-0~4 zO8w}SNAC3gjB~iXxdR8)HEH@1LELd%0F%OParbO7v}0^AT{#yAK3d|2$fIoT=M=hs zF=L&RZP1$32FKcX@$Ql|FyCj6p9dF{U#m1p#;kkfwS5Efd1RxGRRq3ymxlGsdtBKt zogR3$7M?0ffc^K~(AlAh2dM>Wz460?ZDDlhhe#^OIZS=t@j=avLUQJjIoWb^BVF$w zM}1ne=$fLlXy%cH{qs+uWsNh=I=dYCrL1v0yczf(B?8`i3lS|wu=7+QG)oBK*n0td z>GT^Iy-rxs><#}!T)|*ZEBx0tiiYCj$O2$#vSA!9Qdx}4vo+bSpfk#{&s!;{D4aKX zkuGA&=#7`&r z{hYJ!d>oTq%Jxn_O2dNlYe}A970px1r%|dW=~C_0*c!0|2el$G`nn`B{CSb2uWTVs z3wq(CS~ZxyDu+Kx)+j7D7iE0rpnb|rl)tzF4?koXxN@(M_uW4%lGmUeQ7ZJ&8*wV5 zCrD3jNG7!(6=9M33$o+;1q=}>z^>zLz9gg*Hcj?GYc;{G34530Dj>RrL zU#^9#6L*@K30Lh}A~kD?p|67i=`$@`ln64xmH+W#*6l`YQfWQ+(}43ib+ z4W#}Kq0(&!Tw|+-B~$O=>wyOF z`ge;Ca?i0{l#{gMh!ApSi{JzrqED>}S78})4{u({J-qu3j)dMuIjiHCdEynmjHtjk zmL=uX8Gwerk7GL=LT~957y2|=ExwQeWI#PZ&|8S--iw;-(-UetW1!beGX@phT(+YA$%PE3YS~oXMcZsaX|kg zzMHs%FD_q0NwGatE^R-(l)U+;)PrP^2s5?okVYQE3V9H#@eBF zES@Eg%YSL$jYu8r(9r^D?jZM0q?29419+I{Enc|v9oI`olfV^;gx4*Htn8}BCrPJJ z(>V}3@2NnHuMg~6@s%8rdV$HmO7Ts~5!7OT_q^FD_Elv*s?1C$x2B7NO0_7&3;E-G z;mtTp*9Au(Cg8UFk*LegZUg?WaGd`#DlEyw_)j-cV)_~M`ku+=U#4@G8n|$F4HPj4 zqZ`07I0W>{#lYO79bd3cCvQ!b+cxS#WSqjull7;FX-_P+pGm=C$%pJ-ZykLbZA+ca zX47&us=oY1DYLbC1DgxhqmzyDbm+`i4BfE=@5->x`|rnM) zF+YfX-qiTxf+lGY)IUqUp3EVZ+Bax_b{XBDSw_$N`iLcRVyu5hfo^7**DbE;Xq>kI zZS=ZGYQz&VK6#V`nec!pS%`dfW;ctY%VAXWBgdZtJZc}&N)J3t}eqz zu5(b7{dd`_NKo?#1UuRG(*mS(~WC0c#t7GmF9n{L&jPuVK zv9pdE{nTtkmvx=OZc!@?pKFMV2O412ix2QNUK=}pKgAY{3hc1$K^K;N{U@}L%I>M4 zMI(_U_fHz(ALt|N3Ib7nY(3st7l+%6{!#bZ`?MpqhkE+YpnLrJX->T?-K<>>DcI<=gIl$B9rv}_Zm!-7pqZ6UG<5b_8XU9*&3^B|fxjM@AcOS6 z44~Tr87gu9H%XXMgAsFG@U_^1Zu=9kDEku*1nrYzV8AJb%!pU@f0gIoBtuv##LyXwVO?#A&g+&EH( z(urqr>&HC2d|?@eFYQJ6&y>VQs5SEuDUrc^-Eh%T^+#Lc_C zQEbabbZ@?fo)60KSzQ@!65*xe`F{~VenXXOmx*9v1u!d794d zQVQHfs8!HndL-GIbx()TBYJ7%n_?BQ+hY#?qH@&#i98Lq)1^|%FK|?(1)sini$k7H z6y|$TZ&gnk$@V+0#6{q`hzL~b6~VdFZvpQg7udJr9*FxLgVJ^n7}#SA%hgkWcm7(O zFIJARY_5>|REq9SZN*`uY`pH*#1?7U+vG$Tik%UpR{7^>e1013S$dlCI|$JQ932{= zYf3#V<>>3xwp9PhaeBMA7M|a~0*lgfLE=t1Ugax6iMhYgh2>iQJSl?b7kq-kYn52| z`3AP+RpN!HOGwx5LD#`!==U|6ZYzkSoi9CEX0AIU@kpO>jGV=snKuGYB%*<@#u`36 zRH4e90yJayPgJ^C4`$VYaMI!!e9*UofVtO1}b+KUgt@F^$QX}1NW(|plk)E>VKYq0n3DZ1mwdy4uG=&O#4sJG=NPUCMy^&fz{ zcITU0XnCiNcA#C6``tSe$CdhW`={LEPN(%ywaGz%v;Md4(VGx|+N6EUqY@?%&L z)(1pD49k4T{(cY+`LKM2JS$Af=3->M21;p-L4iU$oT{JBG6AaLvD00+u2eKVXH3uU&4-%J8)G5H;95 zx1&`Yu9k_zqKVy@y3z{HTg?H1ngJ5>)qH-0`E48s^9G;K8tJFdRMr?VkJL%sXAE zvR_52b`}#?^J=oORvh9jY++Q(nf^LyP9-fB=|bC!WXC57@&MF{yHy*ehTXztEI0T2 zGJCYYY>XcS#$k-@J~x&$k$lTV5M6o|4jP>T-T4{7f3BYR$1WkA2C<~dJQ?D9vq44r z1U#9$5{!@A6Vdg@NS9Frw%t*|m&F3OHM#>$SniWu8GGAoQ9+)=F|g&sF&ONPrX^JY z^n-;r)z~bFX~+1nd!-aUka~(C;@dEP*J8YIX9Wyf=m3BD2&rV&;GO%5*sQ>VWxL9d z@>HYY&H`Mp>>%3sEke1MP8j&v6&Kv$U{;_wnr89R@Aa=SthxZjG$)AZZw1g|y}@yA z4)EvP2>G*-2TanHpoY|wp>1L$J$iz6~Zr{ZOU!|OnP8&H(l7EwUabcL_ zb(LA#HVsuhUPGqaVX(O_P2wGSNjSHdIXUwwp{IpO*RAQyVbMsaojDIg{`Hddn-|Fe zLp!o>lLZs|`4XHuoCZB}!wF2&p?O_DvFOnw?3t@e<{G?WPR%;S{0(eo+=iBMCPpUA zqr7LKya>yRt*?aoGBrGWL>14?-V7hF|0absH;B{DIKs$rNy5#ajN+Df<}tfh6TNo< z27^zNuc1w3V(YGna^Jq3~eVjz1GZ`Y8;DU)RyeRM~6MXtZN$TQu4%Q9r;2QrwQPgYhgAnBs&$n&2Z8eBdG9bWU$?T;v` z@Lhl$VQp|u%OU4%RS6$|4kS)`Kw5|>WP7O+iAF6lhjW7Y)V7QGzx&CYxaGktu-3pV zhi(Y4)`OT^>Lita8`IVGnX`ZL7u&Uuh7utj7*)w4Dr|kN_MdfwqJZ?QuIAJc{`KoKJ85YJwHx3Y*%Yi z<0YedTp3^9y9K*yOrT&ejCsHM2gju$nA6`Ojhgl%*jC^Nn|7wd@tPDWXOfz~ zt1zwE5k)2y(QF!fV`dqv4M|xf^U(pEzx5E-y6K{l)=@Ai*$U4$7{XxI41DqLH~iwp zLqzRzFtW@hJ7skVkKrrQAUv1E-8#=KWO;I1S7&37^-knXP=cVZ_el5WJYt|00goDI zK~S!21IxYpihQq4UXmwnRsUU?jnE@?uNXazbKoX1BW?Qqt!8%*%vM~-T}9Vb3+ z2I){YWR^)=G78N+kR_Et+7?(52@wSlf8fUj-1h=sn`VRU zUl;uRbsR+;Uy=(?1xbh3b?^{Y z+mCY_9Val)xDGcK+hIg=B0e}a0vbazSpIStez@dLzn9&nto4G~A(=$#JmwRD=3k`x zML95yZdh_oi0U<^(>2lU=%}xQ_XS!YDX$LvmQ8@Pa5%coX+bWV5uMoB1;LN+!NyEU z?3c(VgDwJi(Nqf!JJhI2ZXUHT--mP;`&-S<|^~t_uZU|?&+{~-!#a)Z4ctV z&tr*)5aZWg!xXUG312=C zCft&&1!GL`MMI!{p&$wMhOowc_~Fjry0?;zL!`J&N~#j&tPe+M?iyR}o%7pc&@2840=Fpq_qL3*X z1+_I}a3ofeh=vtFX<8M2#7jiR{U{lpRAw`-N0_=$A-Iug4he2o;IR8`;2jKrp_))o z&=&!}&kvbLbziVkOq?!vE`#&d8E9!24e{dJp~z`IlAo4%P)-Kkv(CftS}U4tZot+$ zv_Nt5PjtJola3nt(dJ1x?iaHa)NqdlxK;Ol`({K!{^J4RXdyL$%RIgmxY^P*|Ip%Re>Z8FftPxSWp5j1{J ze*7IGoWMnx6*eC&Z06zpo(xiHe~XYm8Khq57|Dy;L`JIY$kp2mN!^zOa=A{Q7{yW0 z4CcTa7in0&QI-r3=#rKx4)ONnMbadQRj!YrV&DgpT$#yOcMB1P?K?<)9wnAXH7Zk!6j$1ghb{j9T^=u&rMw5tPsxLA6K8f|mTTy>1 z6&qY;f|VQVBj3PgyEY0lK?^Q%dUzf&d-cVj#N840R;G|Z2Th9aWa*>(5;XF&4*Fb? zLYINZP$Z;C68Kg!fr5_Y|M=Y}o@k(3yDcg_uY1!jS#3%D(x1B++(6CcgR z$gv*<`_ox4)wzmuM{B8$cC-%V_w!`+{uqGX<(WbnE>NH4J0l%5Z zp^mpQZpvAXKUjCXf1Cq$Tk>&ctTVSLyPRZkQHl?gpYnn_|8Mej*cn^Sd!R`EZk+Z( zhm7lgWoqh`$xc^&s+~E5Hm(<=uEE`)XnGIYT28UEKo9ISEr9*XTcPuYEM255O&>Jq z(zPQ8iG%EJB8lC^L>S40d$&E2Pe&oo*2xdYXgFUwOmOa17*6#6Z8|GW?Wp z!{&n796^95jx9Y2AED%i6Z+&ZxQb={sz4-X z+o}vQ^}>Su`|F2w9-(-^i1qL4vb_AW^WoX;MQ}*18KVC@hBVZ%8-7?*4%Z##LR z&#nTrS6>J5(8tjHU>5DWq(qxIiuCh8Bb@qA4-K;oQSq=o7G4~NbcY(yJ(ETTt{*2M z@2ttCgBkEs{tpP8XT1@uXV-p%HAXdQpstM=$FPOrkU2@5!DF?=aZ47tEg4IG*8KxV zzDdYo?|GVOTt?PAfjKeI#OxVcfRj%F({z;a4a-yhaPA`9pOy|YBi%X5Zo?K%hk7_O z?JY66n8CVV3(!z95;lmjT`se`kQnTVOICv{IFygq>nsu#f{eyV{Ip$L>7%FGT`;oy*BVHrMyiaRRDaUc>rp z%dv6sa=a&Gh;m7U^bY@Fmd4tVH(Iff{WTKiFW3UDaTn>|d1ln_nJ_(3kj3?M+srju zV#pO9^CD;ex)GDWK=OsX-K%G6;ionud|Ww7cddFtMF%g^Zt=@FuC9UlA^a%wU?;4| zn<5`qbKoi4e?DHd9iOa!3_)`LQ1X^2Ev>zQtG`5nUS9&xO(P)DnMu9c6mZCer}>6?t^r%RRaCH=p>q5x`?w?E})`cCHgR4 z=OJ#CdC)9en>a!F*IAa(`rrpKS#BN+DeU@sb9i_{J!)dPZ z2vLYP0x{+Zv8~I&fEah2FMAyC25HbeG!F8%|7LR}Ibioq5)DNc;Mij+rh3r{=4pfn zsn{P6JLipoO!;Ee?dPMs8(yHF-UJ@{!)CKsr|_sR59(RE)5FIVsPLz^==oNezTZ5P zPBx5Sj%F0suR54(8|2Il`7OYm2&$y<7LIhWP6++7%bhk_Yp^~xQ_LCq4hs*I!^l)L z?mcIW4|2s&^_e)AcS4!l*&)jvo;x4Q3I(vJr~>2_>bY)piQK<)ow@P9l&NDfr4K)_ zvts^gs+Z$Md5;>>ofeW@>Gk!rdHV@kc0Yg_b9m0^2j>94&30^5^hSw?hN!ix3EOSM zfp3oy7HnINvxS4PMadD!U{m(m*7N)q;HMNgv z50p?F?g2V_cMY|34y1CY7E{e-gLJAgk8VhCq*vVC>56@m*vC_X4?`|s(O3c2@fV=V zCqK#lx>SscSjkn=fh4~o_~z$6`sK@ zuL(}y{00^u=*IhY7x3QSWc;Dg$#Rywabu`4P76Fnzg?`PLxP9tys^vF?rgp``%VNm`uLxAn_Qck8VfgQ_0u8)1gfAOQQL*7Cvp})eHY{`SF=wCsb;z#%I#rcpxqYwMEv0%#ZIRGO&~M zBtOK74>7o}?-VvJx2Nra`)PH`S~|nv7c8GG0&zWI;J@t+eWLe{Dh%GCPpy{FXwP_b zFOI?1wp{!xBmjdqS7QCN4OsmN@m-EAp0v`So1R|BclX+`QC^?EH8-Tsdll$~lTuXp z##=lzT8Uq~l4+M&6g{f9pRQH8M*WUA(Iw)YbcXhHqJlc4YeqFAc|#NqU)V~Vdv)Pq zr8rHjQKwrrYEl(8EwrpJ5_NRFP`~*MY>i>}UTo`*@TD~R z;6nODy_Vc`JWVRj16W>XVE6iJvRzUGmTYY!9bqP*ARwU+)lbtg9T% ziH@b0{d-Cod43Y5FRP_p&pxE(z8BF4%>KcJtL0FDkrX=nu?p@msfInVr{QCjYV^)g zf(jA}P^xke>H#a78CWoKCDhn_5p8iUL64YBbYqkvD0G|vrWj8J zM|W62xxiSs{H7nYuHA;T>rbK;xuK}{)@ryt?*t5U3WHrL!g^QjJT&;i1~r&_3w@L) znyhDw4s5&*cCO-q)Ak0CG;<`p_-r|>_!t5Erfh+2yN|+!Uk}3l3v=n3EplK^Y9GCR ztPQ$-b`fG&fZViJz$v~Hq1&qtFigG_PXAR76&vor!3_;y>=_!^taGN-B9q{Q7ZGqI za{*QeI?>c+VJQCfe)OWTQuu=^aAFk)OeG`Gh|6-QX~YoBjCldaZ~X?vQ>4+i9RpBg zJQ_7UP(~B2=b$ggJ<%0;4|HzGhxR>nQ{-*4R8%dR2S2&Vz*58pzp?^->M2Hekq4^S?16sfY(|E|jOa6b6ID_nrroY*3i*dg zaOSEO*xZ^%-|1B^g!{xVc=_{KJZnzMIU>T4L+q9!GBpozS1RY`0rdeOn6!jZTTX&>FYw+Aol=V zchQF%(>hV8R0nEXeGiR)C_$EYDp7xLIvQ*c*f6RR*j9E1-jZDnp9`9tAA-hm-m^}) zb#o?6?971cH4eb(TO#4fi%IbJUki9B&Iy*>I0!!<9)m_p>7xAI(^13pF0{F}1+Dnn ziZ1z_rq4Dz(VmMDE%)R+ATN0ELuM}scYY1Gu#ceb%qQ^6DgwMKbHQ?fJso5Amaf|` zw1HoyfCHoA!Qx+s!RXkTz|L9&o-%F%z4fuc;yn-4=r&NV(*tIS>*3@%*-&L-8Qgi_ z9X_%#g*$Rg;Ys%l;BqAilwC9ua*5O7wdp(HN9+dG?(RU{jXtQucQwj8u12??l%aJj zE>p*zmIB335@6N%799P03N?L;MBnRTkqsP!$}_IR9m}$z*Q}sIpjle7zWTXRaRsX>am7Va0X#;%i69cA&EeHBKA)sf<0$3G6Ll-w!*z$NY zRM)eBqZ|gnuv?`l*E$nD9exTuHMK;=LQe465_x1|C=V?xUxIo^Id}k?0AC?XWSYiw zVSV)?a40GP9oqMSXWU{mI>83*+XImL&b7eFeGb@c83f#$Ot49lI-Wa65icDx6MmR& z0B>a)!WsQ*;D)++@P~sF)N<&DN~P^^{;j((>GMF))%^#13wq)A18oAI zM-dKf?*_(Gy1-_?J1{7?5dO_dh5AIE2G9-h> z8nnaB(q%A6)0(b$UP)=Jkfx$0)q(XMoj`4EF`)glp^5Jln7U11e8f!zhcXrbS9@d7 z-uxW?_+9~r(dVIo+BNw8BMXo7SK(go!ER{uVXa2&vmO;3s?i5TO^nwY(d#*S82(Z+10bYOCfN;#yrye|}eYZ8zSyqbR z@mNzZ^Ya+c_S_W0aA!C_+76oAoP@?Pi=cuA1BWZ!0;@vbfl{sSU{#?VnvsMNbY{@p z+YvC}Q6$_OnFVD!qT!;;EZp<`F5Hk~1ZU}w74mbZKzlwM?oh}UbS$OtYJ|Y3%IXC& zTZh1}P&@cX(DW#sn+>O1SwoB47SJJYEUX<-1e!S|m|7ACu0)hjM`OQJcO^=6Mzzq7 z3jSQ1p-mvTNCx~Y%mrE=24HmA6YA~1EPC}*ce>3h1!PEHrpk}crhu$DRUGX`Uvbl- z4VRsz*Y7t3K9^>Kz-Kc+#85eiwP!)`&kD-yUm^8ZDUiyotpsLeU10R~2?9U6fDRxj zbl+bD6e9LgGx?+RB+I>YbI(e+gRFwOR~A6mg{{`9sp{Z@Pz6gu(L=JUYPGeZ*QAUNBXqUAJU!( z{csmhnivMEa;L+zmE+)|NCYo0HUbTT$LjaPV_?p7O=`a?E7GacpiX|#0q+Hlbhcpv zeY1NET)M{&zGGIxxZk$GZ(cMQYt&C|I;uqTI-_aYb~*hy=`TIWC=<+ml?FIdwzR0qLq|mCoCF@?7GUOh57;HopmXk=1%Cx^ zrODY;+W*))q3)Ul{7z?zZVi-)q?Mi z-UNQxm4N(|a(J$)1$NW-;k)x0RMeus^uNR|dgl2B@WMiGcq?lIJpF1s{YpKDDiboz zPHem_@SqpK3A_b#-C0dL1#8d}w|^93y}|S^L10Q)4S0Snkj@hr73ym<=*hbbM6N1_ zL|(I#M7N%IQfUvSff$ooI?jF}9sT4g6_Ygwx|;o&72LoLtQ||>|k&!-H0;&{a7?}%w17zZ4o#qWFt;C$^f%_Y-nff zOozSAr6q%E@VT}e+?+cII(Ge`Rbqb7)z`-YD*GJ$;@@_GgJ4Uitm~)m%m@Ham8OG( zviqUIt`+btHy<{>m8P2(EC)=Iz-+e|6ul0WrKFHMwYqa8{rR02y;Gx{x|DYTyz3ne z))l$aNeTJDdqoV;{G9{(_GHl~FN6Z0F_~bijKKHj2r%0h2_9a^hUa=qpq8*UuYTJe zOkMYp4qN~Ldr}8xYSe=`z8hRCK0vkXoJY-Cc7r*5qzx}kM|YRmpthxRP=n!dWLt0+g(p;@=GjK{%@-qR)}2s3?;Akq zz#s@+Xbyi!g~J0b*WglrS=1Z;3{0J&3A--*1d}#Q27$vBz=yF1z^v#Xx^GMhol>$D z1k9EJ(DMj*SEnhkx>Vp3<|Sx!84Vj#L!oxzRrp`7mf{i@2*;m-;L;`NkJ46SD6cP z-gX$>KAMCasz$@k3nCa$dlhbEhiIc(OYqY>7&KQD0k%^Wu8Y-zt3Nd%egDsB{pj&H zO;``NO^`;<1l>u}?S;^EX9P?ayr7Dmb#N>G2IoD9q!Zmb=$=AV;D5>iept5( zo_}}>PWrY9WiHu=%+3d*Pvj#Q{5BG9bXowlVouN}+!JXP_bj^9Clb1hPJrs3ccF5f z817n~0Cje5hJ9WHDBk%D<|X%oyR&!0X^SJE=8Jf^?qUM?`O^>-4eG$lU3<{-ho{ic zl{zH<>NoVAKNhWj2vP8W0lez%3ICR)z!_gRfVcMPAmZvRu%F7!^&MDsPDf!jUr zLEJ_mi*B|T{AN%D9hY{%IO*54r{@wdN5LB`cB}?N>@DCFxf6<{PQyoa4`H+KC-^Gv zEI6!m4&-06ggTRjx`4t1_*U>^CCQh-t}Ttw^uT3!e~TPcNgM^WIRu@uJ>e8;7X&wB zp<_%9RDN;_I$GqxW!n~rw#z*bopAD_51$HxW9m}jy%YCg+MyIM;aC9JuwXt|eWL;P zyS#ycbz@L^TOarwA_I$^^x(;NF)&9u52gk_fSZDDK>Yzl)L*ZM8g{3H@Q_;Y*u51z zI%EMIbsb>NI440LKOZ{X(1+gEYoShF6F6`y1V|ftLX@{1p4w;&-#0hG@n(*+Y1c{G zYuySu_@AI#IwuV?7<0HK=PCHNcM4p5ofepHhd@<(Fi`sv0}Ru5!5%^X{9LUR#{Q5) z+JTzLW+a2MwLa0wN!GN_trogeT@PN9?1R(GFT>-vDhXffvx}3GW8Tq8+|Hu-NQ2 z++Y(8?055^QfCx&>RJRwZ`})y3|#^WOOnC3m-j%c)Jw3TT@6fomQDYB(g94%HbYip zC$za-2}dpySY{K{VR}_CytdB_o)J7i+77~;Ic_HEn!Ev}7^I@fr+Z<4)^N1vqd)F3D(OZ(wYf1_B%h{4S%}l>J{A;)LUghtWbjg!IoS2Mtim~#5@j$)v76A1ZDAE@VM z&XnK!oAjdf0(-?a5j4(t45sql;Dkp8{UmWZ{aR-P6*Y>ZI;N;oCwrb-)-=ARC&}9Y z>-qS+7w4Ru=XwR61->qAJza#SPgyKtb!&hS<%y7FTj$MZ^5={Le0<1 z7MP4o5b~vG1H%i$Y1uYaI;IuLWF}ETG$$uAAV^XsW)8ZqF@dH_TM5A4FW z652;2h<4gBfmZ!`5X2S;?fsi7`i;a#gk9c>W*9#dRjMDOU({a)+O1yTmuD<3-RA_N z8o$!{Oe$4;cB=4R2)^|}LY)w$(uF$4be-5A%-%i`EbnunBh@#6>_#{6@%0&c-Rafz z-~3y&?S%~b^&tRacC-TT{gEJ~5YgS0e&G3zT6)`@C-jbPAq#T8P&YW5NQb4y(`gG| zQY|7W;9-6OjJYib68iQ5`E5r)n4Lbgd*NxC8g5NTXe!Vd3zfi<*QH>>%{3rI^%u1j zzoy^JbfGse;q(Q^Jo@YHOH_vbbUMn>n~u>KM;%I@Ld$F|0Ur__fX?+fVE5$7@I-Sh z80?=6qL&oY&nc1UiF_WN(OOUcbZnUf76_Qk%cbXeUZnQ9ZK17D zB8>%Jd6LOZdf2XybncUMI0BtS@l)VyPAs9L zO$pHcTO=^g)ImpLHTC89S~}L|JM|$Tnlg!AOCA2@CmI;73tEEqfTizwx^VOyx@zko zx-GeeKFpn?H*Zg-A8h973sVB=gZAD6qec#xyq_%SW@6~wwQ;mn;6D%|aIbm-kAv4a zF+gkIX0YSHa%$G&5im3A8qhnofOg+!N~>m6(=E`AzU@;Fjw`2tHIigHSzytfJQYp% z4m45*6V_7lA;am6e+wyB#fwN_$e~j`*+P9qgm>A^#?)6kd?YS|c611!`;F4+;gizn zeTS+jKi?`UT|pVmipQgNhHqrM4>Yc;9NUpt=AxEXYK$ZxV$J%2YV#+h*7(te>biyMt*- zqd@bO-So#v0vlEEdFK3*L34{UQGHf6vUki!q6N|Pe)W86+%#$0aEmIOELR|m=fl9O z@n-nvg=u&yI}!8JlaSu0VF-LuM&ljQz>QaWAW`iLtuFLI-ww5bzRx$n%r_6wu=^!Q zDn1_lYM6vBY|%l%3R1{$_h@LevkFM`P5{yGS8&6bE3iYl7`BmnIdoWHNj-Fb&R= zQ-rr{^w3YOvFL*LWHf7*90=SL3Qkwbfu;?kP?M<|IyyrQoyv`bTXy^fADVUH-0}`I z@cABck9dR@4zwcaSy$0qP>TXuKS5>u7-l(NgRat(;Lltg`2B@EtUD-hS&23n^Rj_n zL^Z>Y6D9U4A&4x_ETJLz-FqUpNDMc9AuIef|~54Yh?&>@uu++LM|Ue8G=<&qX! zyi^D6bPPvdIDh0X6@ct)&1vVrakP=`QF`>s7BF^gDX^+O4UYXyg1^2lf@8CW!v!91 zLDP(z;Bd|}u<)TVN}p?lKA7sGB}?i-XMHNTSh5csGsqJ-D-?X!<_=Tu$D#ODX(;7; z3QEy4hdUQ#1E;x#;PmqYNP6!M6rHvaeOKOyUViozy!%Vgn?J>r_xV`S|l54O+6KBLj=bluk_%a68 z-MEfjy(LlJ{u@x7=x3N5pWKZ3Id83fkXCEIDXD2w1NADiVlavlU?FU>{c=&qlf% z)6kJ$gU~uv$d(Xag|m|fL=n~9mMiARS-RvU!co3naCN*19P=g(Y~K_(uxhp zbN2?cbKMqX*ZEcOUc7_;C*MI=VIKEj?pSzn#2C1|zaOkqDFv^;#RKErg|uJDYHD3# zEHz==GoU^>7A*e~2{uUpG!U-CJL}sZzay=BK|@nM%+bz+dv<5T%k|1=Em1+!A9lf&0`pdX<7?1&KzKen zWZ^)&47~H{99Zn6py;>=9+lD`CiUO8*hxlKN=i{t>i_tRm7=BA@7n0IWv`d{ruBQ) zn?p+r3yPXG({cu!2`ugBT1c(>pMU=Mj{ZB2#|=$2|9$^&`+psW`TzAezzkqNcl!V2 zIR1a`f95Ruzw7Wn+P{?4aqIui-Pbp#M7Z!$|84)TbGP`voI3zBAZ0(-{C{%p|J|)E zDvxlxe+RmA!41z}c8*0B>g3PxWb(B89u7X{j16U)xnsU6++;SI)6_8HRIV=K3UgB> z@hzTgq(zRP-O7QFUj%bkwK3G6=f`QBjpEX}n$RejbKJZEIWq9D2`7ix6Gu7$ANd_A z2^=?BQawcn4pn_XUE5FMqmPyG$Z7YuUgN8rvQIPYwB62szt&1V|0P`XI}ybW8v@+i zO!#}O4|(=5mUK_}Ml__V$kw8D{KwyO_`IhZ@iyNad@z0v_pjR+J&?>GdLQp{(Mjh8 z-uO6rFK0vJz@NDNpFMNJ_X&P!8PEOBUe8^39FX+glqNgogyO>$(PS0jbuDM&A33?8w! z9$b%qfKQBCfQMJe@GnM8B6sdS#X0Ok^47V9)IWPl>g>#kLG?NGRckmi>C|oZ{cd?$ zVJMh=|8^@{baOuzX{a*yH~Zl*!$`^N!ZgBWMAYn2ZUq&-y6iCBhmx~1bl4QlV6@SU zCr<}-cn>cFe*LELtktYY7-2JC9P>$nKj371>Czgc679y;?3zF>+La+0+bFPa${&f+ ztkEc-!<+2;r$Z+9?ZI=FcHrLjIlO!HI}%-5flu#!if?T?iv71<#w1q}`^|oe#+TgZ z5;irE-mErqbJ26IXN?2@QcsSbb5FJc@UrfwT>9QeoQnQl}V1TDybBQlaA$@ zxSk$Ou0EFMXw5CqHEtxSF$*B9vpKWQjl-W6gRy^`Ca1DwFRADs%O~va;Zn7OhgIg`+S`EqMj9G^STml zpXNRKylM`Y`qz!W_&bar(cueQa%AursswGB*n`4sR}syZ=~xu?1dA=KIjy!>ICm|= z2b^V@S4KR^d!)!F&ppQ3IcAIAS^2@!d<{Bm&k(fa2OjR!LrfPMVxKjyxzLPac4u}6 zdWLr6B9lMlgk35}#>Fxvd@wfMoQb9#OeTlBmXqbT14&zd9M^OCJZh9*#~d$o;o8zh z@;k%2QSo;MYp+WrKfI>$Dia&AwM;hI`Z|sLQ!gU>d)|v>?96%X<}lp2awd@roInD0 zo8s3`YfzKzCUU0d5L(q4fT`RPRBdR0)A!xw&bZj%XM0zXI?|5vd>FE*X$Vp7H#zB! zC~`&SDk}NiDF|pcV&j&ZSfwTx9V$9S1`;I5=4ljZ$o417_eK$^@73Irehz+Y+rhlF zjwAtU!}#@zp4{PHWv>2cEBm(gJa2fsn13{X32&neO^(i9SwaT87I2xBwW>2fOq_R+6C z597yAjw4y}dtmymZ^VpCAYuR)?3;+sRvI z&Ei*xY5dUcB)3QM3Tf)Dq^dlI!)cccBv0DNp7e^|O#lWD(~i6kJ6zw!MDpYSxAkBYg+Ri60<%tFFp zxquGXo}$1X9U8~`y*Ot&t>Mr{w(UCKiq=+JS%J4z_J#xcSn(VE+Lzgml zNyaO8u7O*@T_@A9(^Pe$_qLB%zEtIJm4)!f(+=VSfva{zX(8$QItqkX)p2KEzTv!K z5gs{^4Sp$xgIP*Ha9+g&Jloa|{kE;f@!vDZaR)89ZuA!NCc6Y3sAgeoUO4~$kUQTk zAIfV#QG;qabMX8R!M&XO@?H-fYv@ZGHV(sy7-q1X)hx} z<~={n+@JS4bc)+jbC#^InMbw_pNUtRzhrUzW-j1I5zzWl$6otxE4i;Tmi=(k8XV)( zB|kQ3;Lje`4fGk?tEABIBKdsQkqp^|l5fRvM8`LubStUw@`=lMV)>Mh&hL|~9e}rHAz?9pg6vZjH8*nKb)sROz zp{gB(i5=>AJr$rtfDVkfOn%1=a-%B~PH>-}kT=1(CmJroHp*Q)**LW4ZW@fv&*x_UoXZ8)L~#3k-AJa=Ds;W(Bir!t95=r1 z33+sOh+AxNirm}WAc0F-YSKTZ)l|=#$yHmKlOs3uc)QhJq_{PVjJa}z-%+cA``7&8 zx{c=}*P#~9?_fO9i=4#A^xoyF%`&(@=Zm;}t0AZ$Jrz~Vlj2^}(R}phL;N=GG_UV| ziCBy-B+9|_NW=qk?z~3`bF{g}XHrn#eTT@BqxBIL!SdNw#>?a*#EJ)wDAuh{c zD|fL)oBU0WB_0+#QB1};55WZDW|bGHjuE&o`eoujtE40!&#q@@G|y*7n~ud9V=-QHW*RqXWC(mvznk31?<1X0 zSMV!!hKQA>7Fm+yh?i)MAYVT^l7~lUp`Dv-h<|o5|Lj=>|MzPrpZC#$JRhCLT|C=? z7M(s%noAy#ou9mj%3VJkwfHn26&t`m(0fE41t^f`%F$%$Fh!>Myf3F4t;eg@t;Vw9 zshkg~XIsrZ$f7U9vGXV+-YrEL-)?*lRogC+kAmj>^x{{re}w^Sxppkx`OW}Gdqtpg zBM)+mh1^`r;{{~M?E!aL){_`qILCR*9>vD5)4A0Nrd-!#;eHi3;G2FE(R$}pWU3d( zT@rj0UsfdH2UZJZ+dP318gCd-sm2&F<*q?_LN%Z`GgUO(RWkhRQjzcBGhO zJUYw0>qzEy?~mb(+RjJ}|7P$jWB2pFU#{ofa%YhgFIS#J*}RRMh!5zI=2iO2NN{Wn zue&6JzclR~uVguo88>PXGu)10Iv0G!b^ivjd`BWaEPH||`j=znJB!H2>5jbC<59fb zgq3`IjufAAdKbTH=RAH+;wVN`^p_2~gvox-Y$p0{Ad~K~gE1ORfWt;Li5Jhf4r|tD z;fRdbKqS*E5%hAG`r%l$Nul?UT8!ncwPbMk z4enuBAUqqR1EqvDk?_BFt9DxHSqAl_R(0~XYs{nNvDi6-JltHyA8LhM!m3^D((W4c za(opNWEjmv$A7_Nd~5L95PR;N`2}RUb0OFT-mrGJ6eP~o@r+dWb>=J@!+h3{Lk}8m zqd%c-@NHi!$~_ZG9=OM$m^F{chrfelY-Ks|9Lq2{p1B+qbp@R&k7a&0hBHr31Tu$) z%*AaAilF%OIq@j^0T~}YmDf!UBC*NSm{wYWf4W>kI2&Dn;@Za& zJEi$t>w}LqgC2WvUfC@y8k~zu#xu;z3#!cQ!4u5ZXpWdHg8ak`HNKV6;5BV~$@m-A z{IJ%?Tq(1V^)7kGii6{?8c|ex0%tk2WVDNR7hJr=Lez;V00OE*pLn#dCX{hl5|%XUL|O1>8+jCuXh5 zPwXQ73_Dy&VaA8wz%wt|GjF7o`E~zJVei*laP`g~Sj)JB`8#U{GnhN6dW~m`c;vNi zpx+nAs`mR!TJEpo*1Qthp{eUR<(@O>T(39Cc%sOylBr`~&AGt^r7FV5-RlH8ttaM= zVLU;$56%8{Lo!)ijmYwe-1h!Hw5iSw&)VqBlnf2$Jd5(VmHx7LLZ~&K)a-_H#i7L6 zcLq1{Z7LeJbS^q5^BZrB3dhDDMlq4v{`}J57@~hOi!(lwgXZX{G4lRr@kW&Z{G;d` zhND`zdvC&POy&Mq9C!$T)Iu5N_N3#&nDscbpYLI{*My=Kjw-~Ucr;lXJ%n%Bo?%`E zjHy}|_X?eDvg8Uvic#_nZ|HVB4H!Qw5FhP1Bc7?$AW7&_XOt&gVA%S6#%v=XvD@Xj z+73NjF}#qZ`E4Y*ITP86+y$ilN;@KzkLmm?j^c>*2gLU2vRL$EGh^&v#PkX8h|hx< zaydL0XMD&Z%YD;01&avMO3lQtuIMwfzm4V2+-O1vzhz>*IU0DfF?PV_LkdndJBDnKgfR;lO%zn7iUDH>elQS+p5* z=ikp_$9vav+P?xxweuj9ueuwvGUPGlGQ6zY= zBTn{KhVRD;`nV|q*Eka6p1&-TbO(}sHg>$~v|Rq>M@v38ej%4laSggJIFWF@? zoGf^B5pTISk;yLUVyt(TGktQ$InRzDuCy*l91(jSkJ~Cj3f;v>en}HIIqwnc(hvk| z9A@+VLCg8_1s0t6_#$T7V|!-QFk7bL>PRNrdJ-d6*vKg9*5I2zwHbT+EBLrg2WUv! zMStrxpy=Qw92+o${T#lSyW+T#o%uVlh0UA^-tr@H$TFwOv=cUZQqGrw*n5Dp^r1SHFDJ}67g5@`n&?pw=yTxJo$JvUm!O&5G44_?p3zV~zR z(B7B0CuT&&vFOj0WrHUoK_y1Gp)r~a{K@()vsrqMeplDSmNp>QKIscjzfQN-Xy zQ=%ntskYPz59OQfFjNnH&jb=ihIp7fy>WoEv z67e6H#}Av%l5gV^YfLWgs0r=fVtJEV%AE8!!Os2??&$-4?gh6V6@BX=jWhDuw_XRh zaaqeT)frK<;C%OmM;Uh|!JpDl`+GfbYo9wg`_4xcH`Gw-);%0N`Z8AR@7hx{C1NyN zc4Z1FU0uvt9C-s4T{_D-?a!c#Ejn2)#f-am%p4g^FJcuH}SI8 zCvESB5f#^1+_cmgtIteEBQv|PyHFdwa>pL6e*I2zj9lTyWlkq^9#Ux5+Ci*&Jd~;9 z^!aUb+R1>d6j%D}7d?Jt1XYwTFj-zIa@Kp*NrnG#u{tM*Yp$+B*2RmFIBXhGo0dd+ zYiQz{T1{qWEWp=K1TbG8KPTUrETW-l&L6TX66*57+=_4cEIN^hoMKe*sBPMO!5t0D z?cRqa#l3o9;gMHd;y@7@3R1%Dv)ihFjo(LSZM;C2`DW4I6aqorg0<9$i{%!U?@vf_ z3KAs8b}O-!k#gddU*c)4yu#}E_f&Ol`5e0YhLASnQww@~`r+w@8PvI)FyLm|4_55M z5@RquV&hdQm^$>7k_k%yY1E4*wkIzVq9xB=xW4ZFs2y@T3PiO92H=Rmvt%BMnk!b!RCF;DFC1t0( z3(mi}v?fV;mY6b;NLC)v7Q46<)-3vA3pYe+l~+zFs(#=`Q|XJp&O&|3RlkM}u#eg% zvBsHaXT*LFE3&BZmFTT&tNP%UU30zim$kG@j9sQSx$6Un)}9f z^0Ggz)n}LAsM?iQUb9H2zQ#ooB)Ox@!rn)EH661$B!kO!VA~~4)=xDRl$b=o7t1m& zob`{3sHflP&L>dbl5e? zc$s4Hexqg8FKuN(bg>1iH_=agxwA*(*{0M z4~UL5t*P7|-CIRkgv<`5BVx@uUv`t9Hn+xAMp9kKS6$t7mv&0_ldQ6RT)8+oq|$Zu zVyiWY-d6T=L{ybz05zWAI$Dch0IIc zYg|#bgqv@@3|UN);MiRU@%3wGaJiHpwpROxbL8EymrWL)E6ibZrzkUb{^-+D`9Yj( zq$$^XvWWG`%fafe25?WFm{eeCKEZx9(c=BAVoxu(k_ta!5blV5oc&Re0Ko2f^l*9ha zdMJ5O>JQ5|=@X}7J1#J53s)FajQ4t7!%sq*ad5X2o}C=bm?Z` zYg(e%OTH{ugoohX;H7;3qX2TR)|yYC*Rpv_PD=cgX<>|9M4ntbij39bG(jhX8;8ja9?&ZUpiB?F^!**OAV=lzRFaq>Y+UjF%F{;dBhzH(p`E7hGy zK+0IMV$g(qxbzx@rTB3LzNeX0d<0Xyx0?AYy&BgYo5$1|Y+_7ZBl({<)%o!bHKfh4 zS(3lS8&#wZLtFQLLDi<;$AJ z&NBET!kVAE0rR2_%KSezg8SR4iS{2j0~da~Oa^B)5tT3QNHB16D~wf0K=uqazwR0v z+!&9xm@a^o54}m5p)HE2MskWpMv#4LLtW(w~`a@)>_kPZHu zxjXD^>{~E}%T*tXoPUpF3K#BX&Uhj{$*oW6=SmrpUp)MtN-8=Q*so)vp?dNtvLt8&axj za&Y!Q3k($6iy8OLIG2E6&i|vJ1skI)R^0R*IDZ+1Xrb&+?|I9)#+j4!iT6cs>H|gQ zHqkY1Z!%#~?^Mtr#lbDu82q%fXSoNmz_dFT{OkP28o$BZK954l+1Il0q#*@Y1doE0 zRXdoaFL*sA%1}0avEY%Ahew|l02#HN-1vqNuDXlR_gpQ3OYuTB;B$<)1+_I4(5a6U={jT z>o-ZLSvG69N1e;c2Isdd|(6a8pVQI_RfwI+~04X*_fNBVN}ooIHvUmlD9E{Z{8+=4U@v) z&T9_9vUV}F+3XHJ%DPvnJ4Udp=iZPQ7c0|`-ziq6^mK^QlgjBQHQ5qkG!FJybkYaS z?P14Bb2LJ9jP;#a+}N=mif|;mbOu!JX@IHJ=IL4xC*}w>)|eTyTI$eOvprLIUXNGZ^LIRb0(^bw}6wYn8 zn~Hk4_Kv0*nJXkAW~d~d04goBj*nk8Jp6C+$Jy8y?dbX+GLipRaajXnB0$Wpa;WL%Fo z88iwPnm#R*hNPjEir^pmn@ZzHi6Cjwd0U7$v# z9Z~VFDB{*__Oq=9xt=~1PyRy@H#LDta_2Sp=sp(0xN_`d>4rUC6r;6&ccK9%n=`lY zL<-xD6V=IwLR+SiUd`)3mSiRUHEo(1Hq zVgVL+ZY85$*fCa>DtMfRHD@UiYDbZZj7eJzcR5y=xcO?qxsnX-)IdL8wErRdV9aAY zKn+5a6M*`!8?Zi3FL3mTKy;k=pwRF6IJ_nif607|zIIN4zCkP=q0`8{O`FUY^iJpf zr`;f}>sTVUSC%yWoq~_ze=3_4>@FZ{==p8 z`O&PXQ-!}>q{C~kSK&*4OSt64kKn)Medzt?f5_#-NhEDJ0dtK5q%XmP9k!+wZ#}8Y zPV6rN(hpNm!u`?M==$UTMbVjtQ}wlB+?Y8eLm{FJDP%f(ueB9X6b(WW3WXG*xip!{ zRD@JgW~t2Q?7h~`5K$4SXf9KUQvDGX)qCFW=fk_KL&;ze%)a&z{DOA$Irw5)Rk8MgYEp;Q&J!dwtJu01U7(7FN`n?@5 z66$K|0PIKz{cU$mNq9 zi?oKa8pi!ItyPzWTJ@FjXjBw>{boM6)^Pwo3-|@rPltgUQM$~-S((`NT`{z&7N>HZ zuj1N?hn%%*ym%SCwZy|K4*0Ixd0aJLkUePT3|A%{fbZj<0jBzCC}_*5mc zWA}ElfgzggDaI2olbZoo3xJ5MnAG_0i3|+eBEjlR+)1U^UCfb&Gp7mY?#aFZ@EVseX(VJjcCFahgPsM z^*8Lzy$t7m(t{`0VS!(Es#xDP!mmqOvDgXkNPL5>K2A+ML-C){mb0o(I2$gd>>jIBP2 zNPe$Y*Q;jui+h<^Lid2qE57jRLPa*j@+7Mw%4Op`d+^Dp0_>*M{_L-#O}Iqv7kKvO z1==(D8B1?m#`hj6uy?%Afbb>{@UiI{V=$lsCi|j+vw$tQ7QBhgJ+g%@9lXqTOwoKj z4dd%M|6xfVsjd+DFkKScuMXvj~8Z}T+hdsWkgKx=ttT!IP ztIvzEh6;J;OJ6!HsdfaXR+NKH*+!;k5$uh~kNC~B26LnA z3@m$6M<>h@BM+uUGTUE%fje6E;qmfCtQWtRqSAXG_;dh4Sa%t8)&Iu!TxC`+T7})V z#F>4(Uzqj%#$hXDq}Xg$10N1P25V#Gkzsfn>^OCu@#k~k^WIG&6ZrzL;foS$?*9PK z4|7Lr!(?ImNEk};JO!THo(8JbLX`Nd5@skWmU&64;E1aw_`lKr5P3ZgcmCc;o0+Jf z!p@f{{evEAEaJd+<6dBQ=sF0bU%;8W<8aSi9#|#SfTHxL(9Gr$$d&S;ydT`Abei!94o*kBR+H8(u!W1Qd$wC!pYY=0EX` zNN1%ycxHfFI0*dmq=-Uf+dwrq#9BAc@A9X|PIA9U}}BH;raU}JU} zpU}BM=7l_gr`DK|>%CqD#oo8xin`CWQ70EkK8RhU22E zVZjf8fyd<#jqK4UPjIM>3hH98B6V*EfqPcB;N0tyK>KDA?}6ZDFz0t7=h-1dz47=4?WBY#b@^yG z6S*IXL_H_t4_RCKMgK?D4RI-{$rn&R-UUjgM!6by?hTe@l7IK3YP*KZ9iDL zQ5ZfwT0`31`3wCPegk{N_d;FjBsFF?pZzKkgu_J(k>iK*W=(yfH=H6pf|wzVP+#J8S8D{$4k^_?d;0+YxF3UlW<@A&q;T_d)H! zY7~4>7vH;%(1`XbG^NwYs883yzBO&|_x-If+*g+xi*o|=Sr6#5N){^W7m&V3vcP}d z`{2Ynz8;<92~BSMgVlcuklboZVqont5Up*3|8^cD&tn1-Pg96bgU(QXCLccTx{WlN zTj9x{n;{Mu0tGhdpuyRMI$W^;+Y)Wq*-Vs5f7k}p`(C2*$0D$OPC<_j{2(R|hkh9Wt&oSPME?P-GX4Zr_2Z$HJ3}?_ z6qyuNNh+kj34Y@w(#FH$*e!E6Y!_0YW2_IOd-8ll?wDdBw;L1xhp1TZuZDqiUBhu{goL@-V_6Mo`ZY~h;e?*35$&)>zpDkvo1tPbn zqhy0CMtygTk(YKPoa40;p0i$v-@H`7H=A|XFMV^_;k?uM_Pzl2?h`Y%)us%W<@|%2 zk}pEbn745Lx%K$9&PHUiU<0l@t(FL{V~liFyuY95f~m!!TP z$|cE2P5y7&&)8Rcg8-on=Bt+`QC5;ms?@wh>)$M9B^w3UF0l!$I>!*c8rTZuZ%Ltu zM0L)dEnz@YWj{G6cnP$<=!R9HG?X&$A#HcAX1?xU2?cj9#`~Xl5UcE0L9X&%%F3t# z9&9;F{aJrn`3+j9%A+AWC+`@*0GaRR#KB|@9bRd7XRliBP|pNN$* z5zuS;39-9L7+@(`6!oPQYH1h46(TY~r2dK-zX^~CK4FQr3#Ac9S4aW1ZXLWnD4x)G zK0~?*TH^B2O9X4PjP75O#M2Q`1&Oiq;JY7&P;EgZ*ci<3LRL$JTk1`~!gNo@=1Mgq z)9??R?VJs6<%sgXpA>i7{;=AVjHDWC>Ta!$hzISaS&5(39brZS_Ts;_PmquTLzX=e z!#`_%;J3X|g#3CNGW^sgbit7Y)_PUwV)Gj0G{%z3haZD6+)OyBiB znF%WXLVmoz7agN*QQ8+t#FO!a+D?MZ*02imos!zj%GVk6$;d~Xr`<9{sjdbxYFrOc zmNU42W-Wv%`>@lFDOg=6#mrl|iF+hi*5Y4o7CF9k7W~lf3}?c=@fK6zmJ!3Cf`;^ zFh;gsbfLZj6ZNnTxyOk?)zq_4D>Dbq9V5w4C0xvzlZh8;uOt?T^0TU@{M$U^Le5$z z!+48KAR*^xyxkkNGLAnk64yZlh&EZpWS@^@7^bFj&fLCg=d&(2=%X7iI(rt+|C<@_ zEYf8*eO^P>_*#)=8?SMpcn~@`rc1m!m&fzE9Rb*OG3+3fNv?k%MK+ABK;0V+z}R31 zv#oX>ta`}uM&4Ye1Jss-Xifqx_ii1nV0FW6Ph}@;5S@fk>6a+isu-BRZ7;O?@d2zB zw4gSz@v!PX0jEF;Gh3#McPvzfTpag^Y@Fpuw3ir>mycPImp-@f#J!gBB(WQc{$@f} zNbM&*s3};W(FvM0R|AzoXY9!LUhW9wIAnw76Tj)F zKU3j$;~@CQZ5$j6HRc}unZ*fOpbmyzteD;Ak`{*q{sVjUE`n)E5@>uWqb#i_;iD4^ zDfIjl^>9}r<<%EO4c~qQ?!Fd>xsIB!J%2Z^dVvvchinpj>dXloWn56FI{id7}D1Hy^I<6Ak!e<`nl5UhYxpAt~|Wv6oTjrJZSYWV@WSC-O>=^wz0W#Y_lvJx!F4Psv6_{wQg zkaLRl&awRrEbP6ZgM}j0?H++0w#t;i^Q%;R?juU9`W2Nn z%NotkTSx5UcOPUM$5UFRx2R?N&QZHB1&~4#TwdaJ3t)f92?ov8qLf%CSm0w0Ge+Wo z)8u#f`cOG>LB)~&HdfB8&XZw2sygw+!z4*u)kKPg=b_#J1?*vEW~$Pj$PxUlMtF=~ zLZgFEn5QWRVLQdrYW9&}Ix7+O(f(vdh&0o6#R!aO&F{R*sk-EA31ZB@tWsJt7$vf9VnAHaURK^(x z>Y}6r_0G>26u;OHmY2tqKeStjhRc)WcfDWCkLTy{O;Z;({*)sd=KKr1NYDZsGv)As zz%1NcXUEQ3&;Yhq6L_=UA#~#06J~Qp3DE`55Z3vCURRvJ@XCWo#bs%*`TKPeMYJG~ zgUO)V+7NY}TMKi64%{9k3TI35^Kbp4?12J9&g8Hfsyi--+<_P?QJMz6C~bnbhsB;h;sizSXC41r&K~;b#yV`xXXotyfO@Z% zQ)PeK@v-}@jBi#OzRBB-cMjfzVS5}Xjg$&z{1k;G{{_<$>B7`q3k^)Eony@lXQ2ZQ zk4fvIC~)QXJ$8%Abyg&AD{J%2nCz-5ASJGPnD@GS5zX!DBZ?o1dnzuTzI z*%~+jm;IHZ_6G^DD>}X)Q?3xsJ^B^heyqmUpMHvMWwbDTCK^B545sgyR+QVS5I1&3S*$FyV27)q!MCdvC z5O_3hf!&HxSja05um2?imJeG~HZLG=al$#!pZW)UKU)mzcl|&CZ*16dFKg;m;1oQ+ zXf_Prj7jaNc2pC0laQW!m!61hB$sBIF(>99WvivzwBJ@$Gd&ILEz;mA{t4(kmZg zxshcsSH=@`-n~eM-#X2@2sN`oYX{lCa*pt=u^`*Qzm3q#`kMeG=Qu##)C-Vr^aZtkG=_6g0*q~& z#sen5P_pDd?AMcpSB<7)fqB+ge${3C-S8?t`f@Hie$D}D70ocACoF+Zi3}S!lYpgL ztMKOEMJUXAHLBHFkFL4)qdT{D;?V8(_`x+*Q2)z?@n4`q-MIgPl=~D74X#{+=TuUu z^2}FMV<4oj=2UQAMJX_rJDSl}Pc2Xxb&q`h!U~0j-9l;0|B(O6mk{zBuaO00 z8Ay~Wh4b4cpu^`{_`!Y_6Zdd7lBqdI+MM@eqNT^_e=`rcmVXyQYk6s^p~oI(H#Gna zT^gL|G{y37qo}EuFJV#z3(GsrK&ylXDC2j&cUK5gD?DPTx?ReY#-Kjs7q*{b|1F|c zJXWO6d6iHWTHh$z{|9J$xC_0rTn2rTukv~ew$(fq)L;I1b_3PWwvH-wg_NH1D5uly z5kp0+16ArymR)A9mYaN6Sq|-4$+?~&%5nYh4@`xV=+fsu#H+cp=#n@oQvF5)?^h+a za+h2?Rjl4gZNG7pI`OxH$<3vYoYgWX zI%d=9Z&GfrxbTrm7{$~X_qWvJ#9Ao-c|BacZY{j1x0*7BR@6kO9#xgCN*zv|M+Ko% zltX+ZynH|sw$iEa;E&5x%!5ailvOE}-D6E&u=|6SUP_^LA2d;~7Nk>*Rt z`5Kk`$CldrX9@NYCFrPtc5n+=p&9u~+V@gAJ){u}M{=&i{-4@ZoTohD7O72sEbpe% zbQi(C71eyo>jU`iV>wT&&)`6+H9m2RR4ibpPz-4)FNs$(*e@bzo{v*mEy%zQ>zr!!72rk$t)QK|DAdT z9O`_aXs{&?l2;*%UTUIF2NiUxe;wsD@CRyL>xa)4Ou?g z2mvo#J)7^dT-HwA;}lKtcwrG;dcTP#S_|AdSU?NIK26mm@=nT+NGD57p5aCllx zY&@3B{EV@N1}Wb``uKc&yna12+cAj7KK)1hd~b|;k_s8Oa7U7o)nQMp$ia--GqBra z7yeSMikJ7)fxvy|QODm7=r)sybZt#7jz0cEFZq5FspOfkfs?uT*&onHq6fAtrS--_)h%YfeaC<5~*6 zHT{0(4A#(zf(zsep%%9e30N$~H52CG&!h|N`z8UmhPZ%>fx>X=>1z=D^9xG%ddzrT zzeLygeI*_4iLyuTD6uiegxP@dK+5$}3LI;B25;va#`ib2Gu_%saE^ilzH^<%zh$a1 zw=bF!inN0*dg?IqksTY@aUWW~GKatOOPOYQA+DKqJyRlJ&*~4%VHadx!YO0p@Nm~C zob#2WDlXkar_Ar7zt6gev8zjr&Us3}h7C*9t^GQq^N(?DbLI^20x29<>O0AWv4Sl7{q_lYX8A0l0# zYorLDT-plNI{e@>Zy_vEDN9w`wE%ZpOO*6i2)ga}VXTa$$e>;~f)`nV@!^}$=TQr6 zTC)Sp3EYhrDXzjR<==pH;cDdev_8UP%`Wt3nIP-bqQ&;q_;C`(m6 z(WT7Ns;JxEm~vX!K;Jm@o&H7dCOabEfZpn1GUl*1DJ5hHLMpBj7X=;(bB>hz0Tn5j1e z4X$s2(yntULRW*b^if59@|j>D?=iD{SjznVw&q$mAw&w_A@NyL8ECX69_oa3lFQba zk^;*VRri$_H;vSTjbO1ly{u$~n>nByk zH$sR0N90`*F4NI_3iXXh(B8_INuoT4h}F!&8-jyyyKXi5HzPxSY<@*HNL=N_7cK`r z@qAsn%>@cSPJ|Ux8sO95E~I;}g%pxrOsmKRAb;B)P_F-j2y;osHV5DEGuhR6CjJoY zbXQ~3!W`HWtSr0v#eJL^XNQ-5)?(wb_!+P$2i#Ee8Dsr-c>GNrp7FN)R}di`33_ zK!?Ism=%eTf`~sJSc~z(Q&m8EeJV;UM?f}4i836x4lg&^L#fioFzTf&mDybct0t20 zLH!hFiQH`vQFRu+eE6FTpF9u!J_T6REzPz7A&93n6+zsOJOF~Sqd{xRBJecmA=rE` z3eN3|p{~RTT5dTy&+=Iim$;RfU{TBOzw&z{L0aCuM~tq^C-a;`IJdmCIO%#<3CAOH ztm5<{3`_ol-DEa+^F0BTtwuR5AHnw9{aj8+f<;>4 zcHVr;l}MIRL^e{#@yKyg=%OocdicaTNor1H0(H?NkXkMn0FG8xl8qh|?Vi-n9BJ7BBq~hd-?LKq>;65cIlB=> zytctkrG7BhS`Ye0n80(+w8h~dJ6Lji1O%u*1)uF&!AYcNk?DwQOD8sSj00!W!=fC> z+Z+S45|_fP-TBb)&v*Fk>gZ ztprIc}a9U~ueo|hFmDLRK?Y+i0()T_%USA7e9&zR0U=v{ftq)|^!FG5Bxl(Jx z{ebOMJ4pH*4|5|2iLVizO!1yVAhLcr`!De^{xI#vT3MZd7RDXWCe{ZYZchZFB`(nJ z#}GdEBN{u*^zw!}t!v*L5MnxSE?^wjrvv7{)iC;A7U*aVfk}2!pl>1$td%$gVq(m( zPPRJuWq2HaSuzS>lQS~yU5z*0vjKl?n~>)&B~)sehQq#2pvb>Tu*&BVF4=yIG)o!; z)@m4sOs)Z~?_*g}jSOUU=r6Kae1`6R|AhQ8Cm7^xtB0LZD(sfcl5`nm#O|xxjj2U z>}##{Y*A?^c0s9}>{O1$fqQAZ2ftg$R823iJn|T_*^pv!W{To{+i`(DxTyq}7nC;Us^dR?qem zqaBP;^rQ^6w~j@I{MS19WGQ=IZ7v>uT7t#++_3YHOIfSJo9vEsW7bZILU;b=;F7l* z?5h(?*t6ws?4|X-><}?dE{?bh1b$QK^~!7T-=TQm7x0r5Fg%P>9177PiKm3C;aco# zs)6nm_JVi6Zi4QnA)Tfe6nf= ztyECKToqp=Um5@#pKb)v&B9RrOdQJRnquc={Quj!ZSdOJLfG!t0^LLRLeEEU@k4<* zSV*j%m-*L{LMq3p>6uCBMv36l{=Mkmad#-ZV-CDp`46wOmt+HTs#y2vM{MfT81}A; zAhko}4_te5Db_y|j~nM~=%>CyDl)f2qqS-w#xJLjEtlkNmx^$tzH31@J zk2B+}eGYjZxeU+iYT=)E1{jS~LyUCIU2H=e;zjTP)^iL%i*_lp+7WSJZf+*N`tc8a z|E`~f`gH-gZ%aJ*XA}!N@0Gw3kN3pVp{q3od_(ZgUsuVR?IfexCkTcr&6w<^0>r{) zouJm4-}Aj|3Mk9wfo=WIOpQ#on|s$AndPfSah%+OxlSQUV5goVX#91I$(bsq^#Xgi zvTr|`|2~yPJPeirw=@Nyu8R!v5xZaI6B)r%tGdWtlXQWb9wtG|5{M=bRNImPCu6~> zO@Xi>zmMGb$-Guf;5`{2=wyLP_TL1I?%d067(3n zA~~1jKy=Iid8AJm_#a%tXU_dchg@1-W{~Wk{X4mhd@?=4on9&ev!BMmO_`av#l8xY zN3UYXGYoxadQZ8)Tjn|IdBBoba2! z@4xvV=J^SWj->m<-qclqyKx-kni;~_*)K`nU;)cZwA6ySQdxwrfFP{g7DuSdm_29n0+fn-r|GMUr7 z1U%2EC3v3JWO&XJKy@HMWo{vO6YvUJRH7VQjMc{|U zLh8=TT>N}hGEQk)&X(Cm!gGDGB-e5UGElfc?mDuA`LIZhZ0UXhK3E8%3*r(mC_x>& z^~&P2eD2#4jk9F!DsB9T@0Tx4`G}k5$3u0CY7`7JST)Ld`F(uxY1b*kkKh zQ0LbM{iqmRB@>4pjyD6r9|F{Y&A*{eTRt;4R}hc=^J5Iu*Hfjh;wfmcje6&l!K!rD zvGOkN>{zQ5<+g*OYKzK%|Km_SIx>njIldxV{`Bg+<+FpPGv^FMUvv51Bj?{dtRWO zlli`LMr=Eu!C-f4J=A?rm-d{aTZFR!{+6;2#Ox6uCh2 zqz{m35CUgkZ$s|k{M+`4mF&pjBlym1HOw~Ju}{9)LvKe}d?0=eyKTj0YOGKbemKWN z*F~RWg)nV4a;_cQv+o+L&|L+0pTCA>i<*%geGyEK%x6wZRzahMFTkNdUG~AHIaIQQ zCbe<#E4VaiJNAD+jMNV#;Ou+Ou!ryKZ1g??orja)b46!3?9>7^mL@XV@6AZXPj8S& zLJ~abz6VqK3-Fp|XFNyE6sygzBc!DjdGj=8$k_X3po6Q9+a#m#jCDNDxDX7^ZNCHS zb|k}JGF3?9)p>O2&IQ!I))1C$Z3Dl@%Hfrx+u5%bHaMcG7@x2jMLYDRsmwnS6ftAW z?itNwxwB|ibcq%$Nq+z%H}f+tUO(Xj^8;W=68n;I5frHwmAzlZ}Ef&lGhA}s3q zl9@AZztO5I{JAMThQBIU&@#8T(8olS$%&okd2K5znb-9?(6YCm(LA>mFPKRMrlDfc zOZGVmtUd|~mrTOrg&k0Q(PltqZvk-|&RZ;=j${M`-7U7CvqP#j`9Sj18*sKp6rZp) zW)@e40HrU9_+Z;Yw5D_!3Li)V6$5r;k%AYx^`AYmez_WcFIolDzukv|dpcmtBGZHrsIIZne-Gx2S^^GB)V^>ugdw_cjr8Vuj&i`>%jKeuaFN@PKwVzk(dL9Kbs|%TS>} z9Nfm?0c~X)@P2`}lp#Up{c3 zBNbo~HitC$_zXF^2|=?Q2eL-p1ufUl<{u8;lEF3GK)S3y6$eDZX(Vgf18XuKK}BghSpTz_5cxJ6IyeuaEpMgZ+SV$1cK!|ebbT8h z+~a|}S69>T?$5(D%KTimrZEbwm!-Y7yVFiesiY2nraSyn21sw3MV7pAp#2_X^I7Nx zoJ9%_;kxTFSR6fUF82e0@ulo3h}{6cK4hjiwK zi>Pb40L-xd2i81EWxj`u{K_-CNbHovNb}5rpdpn2}k)@2RbSKF7DTjUHAK)xQF^aeK8*F`X zn!gWk!q1Pzpw+?+_?U_r?0)Ekws^Y36{1@?UhET&d{`;1Zl(;2pZ(%V5AOnv#1(pD z#S}W!GY^&a%3#hKRe0UjhJ6$bc`r^J;1oB+k+Tfvpo!&5Xw#=NsAsPqt;MG|?p~fl zUyRGer&^Xl7V3i~KlMvF{S66Fmc=*51nkGbycr{^hXik%iRKSj^Gvz_(gS{r!)s03^(?w9Z z$491V9?cscj79oCqHupB%Zv#f03oM0fr;xc(8>Qi(4FIc_=Lo44p&%@mfQ7_Se%@M z%$MacpA2)b)}|Cxm<`e2Su)rxYA+EWXwGE6Uj+AL<)AGjzY~oaLwkNZ0p?#bI;K+& z4VPv!+U^@s+R+NM!`BwR*Yt(GTW_J+(h%J(oDI}U8K8bSh)D>sryGNr+JkA~aPO;G zFs;iH+;cR>Q^vE=aF`|-@Ol)4EpCFu{D*BlCNRQQT?Z=PhtZfS#vOf_V8R~TdsnBx^E zCGdJN68tc)1KDXgXi)t%Qq}!SeodKSGEHSr&0+$k59z^%w_DLjas#teP!Oh{iKK9c z4#k;Jpzdslg;_Zp6Y)!Ey?&o5`7-0yhi<4i7ksj3g0_+4G5 z9)X|;9R&r>@sP9?V20h_&_)s)kdJ{25h(0MI5cE~CXX1nbkKs@bybLR3b24fOU0P> z0f6!Y4 z^?gH*2CC4f#uDEin2RjL)tGxpONjL8Ghk-1Khl&~OUXYJpfWm^GsjOUU|#-Zy!vht ze!aqwdgbFojSQWo)S4G5nDuJA^q8!6v! zm#N6JvDB}jFeun?7xjLVLfw}f(5KKGa9AuGJUOF>%lVxgA3|-2{PqSWsIQl#f99a0 zHvFn8RkhYWaPAMz_jUgn7z-BDV1v= zm$iN-Sq}xg&^8Ky!bbR(vBmd}%Y)CK_}-%1t#I=)R}%bHhxxK3<)LIlozU)vdxwV^ ztK+tS^JPA?g{koBpguXyNaOe&vne?u18f!U1vmZ{@%N`CB+1{C)j7#v#g!)heK8+t z>QsV(t*4nA5hLWDH4894SwM~~q#5aq{rq>bGOqtNi~Rmxj9UATf49sm0gBrKxXvw- zfUus-t&elZh+Jm*DXp%n|6ci56F+HV0q{#HbURSXb3EdvMZ$MJ#cI9zw-26lS7 z2!RY?`0b_z)E=?MIi7q*>;50uadIvlFZG59=qR_?o+gHs#r=VvY#+L|{We^qZ3uQh z`wH5rB%~bmnBHpro_DrGgi_dA2%l<=!Y|)`A~!QTW_ta3Cg^!7E(k{_ihE=TPDgpLWvoK56h~u!UF7=_TwybvD$RPm^;BSD{m3HyJB&D;)K% z2>yt<4X=krLEP>O+Hz8PjdAaJW|u_Si)Kw&SmPg3m|smweCr1S6OGvT))V&W?Ge`e zq2!$dot{u8a~@t3ycB;w&<74H4KVlr`7jSpy0BMnpJyHJ9%o-JxrT@rmVfD=XQHVq~lYB7%^?ZS~E&A{2xu0B-cCwrK^GY|~a8sChf%X=5q)QWQ z*qMP1EnVRr<F*}@*bnuk{uNTW52hWYO_ zMHs#CFW&xUjJY+U&gO1jMy@Hj!^|@lV%k1jK)q%g*<;V|&_3T@GHW`Ak?R3DFcEi> zCX2k7fSPAa(VYn}KW`d@QYJvk$Q1kwIEhD>E&@LPa*){Y1*SP|IYYMU!;GeQ=D^Z8 zX1<^RIU#NrY#TeBI@6R!YGz1u-=i!ro`_z8m_ox*>e zyy4nNEo^1pfo>2}uvK*<^6E1sr4+W{_xBy}?;ay8FzAY}$L+@60y*Tunip8%#6jde zkVFp9yYvimsz#HkWqYgkQtIu!Mu=sVkWAK{_4hmj=J`eNv(;b zzpOlub9J24BPPwU%ayCO%2{qURux=ZY%Yjizg1-9E1f~O>sPRxngr)x7{GUT8faND z4G_f4LFf1^6HeV5qVHiBaal^37vf}T&g}595d8Ik3G?%3<`*O~?z-hf<^FW~gm@1! z)8cWYUXAe*wMU3@s|KElvLt=ZG>vXnyoC1FW#cDXNUTy2jlGXV;U#wZc;A#J-$Px_ znD5Rvdv_y_N#fju-S%a0HM@|xwmj4Puu?POyF(eAId}t!i>(reO=nAMl6ubH!90GhO>(&|sXbjxyVML3HdB3k-$*?A zs4;>@9r&8)g|&pqCW!7o^y01e(YBbo-IE)vTE^#^W-(d&>X~szPYZ(}vx=j1Vr^69 zX(B-JA9rvGmnZWs2A@A|k4H94AmyZ7PX4o4uJiODz0J~`oJW5k42d(ertTDd^6U(= zTWt%6``rPUB*HUw;^~ z_@r4*{INbqM10)Fggbv{WFKuM8}?Q57T4Rs-t0mqw)YVct9A*^yY-U3o!LSf9rY*G z6E5()KgRGzqpIke_Qv?P@(wcW7@&8%&Cpw>?$ENiJq&U2JtvXVU8}^na^DTaC$iH2VN3&^&Lc1Pc5S%7>0JO zzXXov)e^rg7jusYZs%B2hK#J_CVF$J1E*Jh5#dsnjgAiJq93wZuqD_P#9xY}r(-Ii z@GDi=T2%|xz0hK* zM5x8SoC1qqAc7v&E#>>prC@++G;I4I0&hNg3KJUsGP3O#&{kD#_}Tg%{?RDS9toI7 zz1G={etdclIdAe{NbONOBC438hBHYlY*(ce`-bZsKPw{S1d_=E%QJ(mf=EqUIjEp?{ncOXs% z15t@Y6;KR41RqJvqgF|XQW`xMpmD1^2yQonNn5s(ac{0tv%W-9Gg*hJ)DaJ87C8u? z)%l=(1|g7c*oUMWBjJ$81Yh$VgtBH!slh}Muo5xwpQ{{eaNP)(c&SqH_yVjqn8Vh$ zLcC-siFHkR30y;K;P>`6&@HS2O}_4eX*=8CGY@4_CZHdFQ;PyttBw(;I1f1kNBqd= zuh#+Vxf$dx>DR;%XE*9qIt%VC>BjaBZ^6s239xyAFSW)*47B?effM1OAmDKw;J8Lp zAvlf_KbS;Lx6Xo(AF4y^gltBuh);B1$kM4xBk;K_DR$-l(|B<9R_tVhew-OQp_pSwt7nffh!2duS>z*E_V2M)MtKo?|wq++6=M5Y5*%WH1o;&omi-87Pz)7A8NmT z3YC7CvTHx_*&y#u;JGyD4glxHUFnWtk~wujJ&5=KYX0s7+}vPx&A~w z6`zp#9eq3!cn#0o$YwT+biq5S(TqmdKgP+WA35)NXmarQ<>fW2Z_!J)>QycY?5~`4 zSG#)LQW0aN^H6zXKX^a>ABsMAg-H^a%M=OR19cB$0ZLN@MoZg~0jMCHp4*bWleg)f zf&!w^y_Shyc9EU8KY;D9YsM9>Mr=UCeZ1*~44o`@825*t#nTNNP@vro_Sx+!d{@aC zuPr@$y+9$^Tqq0FnJJp1XcU8tZ)yvjew3<54pQ&IvaG zY}o&o5izggyf7#*JH6pOXQ${2V%gX$=5_c5epX(AwR@F-9Tnnn|LH>7kQPKP6F=zQ zsCD4psSEUyO=nSjCr!pDFDHAhjWUJ4xwSJ3E9g7k4vdl0LbUhoL);pggHJ}cfS$;1 zP+shbWrc#6lUWo`=|mZ+7qp4hoV|(-U8BZ!e&T@eu34m>S{aa*WU<^fWvtR2!~A>H zf-TGA@z)MVLn8Ger9 zP;worG+zZ&Q!?p8+ucbsT|sns`aC}MY$4vYY%UYU$A8x)tYgPcO0iS9QP@ea9F~#8$w_GE=8UV;xA%>sdP7%iuN8xr9QzKaT@64$ZUqw?`v$$m z3t?m3YqnSY5}j_tHlP``X42wnRH5ADvxvoyY- zWy})r^`#k8rL`Sh8_i`M3;S75Qs)k`c>;>-OBqv>g}jCPGs)kUMUeCT0SA--nc{6_!dox8q0g&e>!5?ZXXt z*gNPyhkW;V38Hf!5@+=(96BK&s|Nu#D8F z4qiD8&!#KD>OV@XqmMA&eq{y>YE%NI-5TC63IQqVLD-@73S#+;-Vb|C+3(`l+51*J zHm_?v7UBEr1=U5^Nj^7Bu3-kF?Jclh{VCJ-_bBP+myd2A5=TF0so>@3&S90rH}H*b zBFxzM>wm1h1#}fx)HXc0yL<5BL6Y2=xhV&Cm!id`NU@>?dU1CRIye+46i9Ms&IAtb zPLU$LXraX^rO*QZe$LDU-f#W?x7PQqf3H4w@BM5&=iY>nk{f;BqmtI3g_*6qcXL>e zf6Q;M$vedEIo-Cauw(X_FO2+)FAL7F;||xbFN=Y;nK}=j+2Cuw^KmWhc*(`~$~M2) zN2hJD-{q(nSu>)x+O^|$^N#w|NK3 z+5f>BPXFfNzS%*7pUXtp&HYYP&91i@SbS)2jX#rd_Pj$Zj@0&ThJ@GLIhInBKFyGH+Xat9h(T zMze*XnqwXu;3e-pus1cyl(6E&8+&J+|I9n)SM>faXLT6?Gdre!!9XQgG`w7nmT zSYJKqXC5k_AsCxslG$V5aI4zcZhG-c9r=;ItMr&#`TalUO77p;U$^oP+`~J~E~Mq! z%6VMeE&g`)8Xo;E<9C(>c%Q+Qc#0ce*|$FRmJ^R4@1x~q?lH?l6S z&8!dXC3vYXika0T^V%ir-R9L^R=VOunV~wNrNu3z4y<&)7X;W?g)7Ops z%7F3yRN)=0mX!u*S<997$7EZg^JRAAOi7ZbtiKagCRWhznoN z)lbyVtHtgqWUZc;puZn{Nw0jrqRs{k*Z-WG&wnwc;@{Br-`Lk10&&^b1+2U|muRV1 z-}0UHCyQ&fAtso0(TVu)i{_3yFf`hJbZV_Vxlmqyf967~SH{Zx=T7IXlS%ITz8RI1 z{tf5=UD;k#PuWkic6L#C>%A9Zdv_}nAFij4%9!`0zVg*Hf2E|k{6$A*_Ag$$T)*Ux z&}MFk*9TnOtsg1=NKbOSqJH;X49}&HuqyUFuH8PBoX_g8fiKFR$DfdIopq{rcK?#6 zMfk|wX|?;Eo=4r)lG1xAX4C8Th~jDBS~;y{_wMSKEA-d8R2-{E&_6S)7I!alNUIT1MwDS`!>4Mc#+Tz7AFnl?KA3kL znMv36p4Pi<)l}bi*VRMKxqnIMPydOTll^Tr=JG#X^Fi-9v9t9qQ&qcDi^_Jv1OM zQy3pxaw&gXH@m$mlVNuez;ibE$ZkevX%1+krRw zwWGeZM<)Ld&u{tP9?0U)yXSZRjI=BLhdR9A589WoN3{&{GaZhb^NiNkzw;`X+YYSo z=NvTE|A$pWFLydzzaHL5o47NVp6+!*bom-vquY;I99{8Ignp;smwfaq-P(0@4gE~x zMXObhQ^7JtdGKMF9n5`ZVz5#3MbU9(=|?aYCiP$O*We_imW&`%11KPTUxYFPV={!(dw_Q&l=+`%i>+sA7T zSjnGj-RXNJsrc=%Nmds6y&MOxR=1x9$J>8}7x!P>{D*FMle3iV9inuV>%2a;&VNZRnCquRW<+^xBkR(N^8j(Pv6+ zU9==%}b6vXq}o9#v4`Y#Rq0uz(2HK%&W~^r*HfAPxFgO`}`H# z&G!FA@9W;X?pCu@iiZAo%QpB&rWqXF_Gr53w}Z*c1b`r0Ku{IA6>|KDY1m@VHv_Pwq1Q0@GxpH-*O2lZsW zLGj<_C}JN9`-;CSz25il>)B@0)*1Np?CE)touz##il>NtkbY<6i$QmM6-JlSG6yGV zH8ND-8M4v${ae2>YfN}%CK+=pYR~gk>ic0Cc*~j1`I2o%tt3x&tA%^!GrJ$$8Z6f1 zg4H7Fji7O-kT#6I*1WrDxNlgMFX=o;3L4y2#FAd{kYLTg^MFll}Ne zXFKH&Y3vIX7i&4%oUx9lpQtu#et~~I^AJy=_2Czi-O!SZAEug>vhgePGpd)1cL>&6 z)7CsPe>vZnZYST9_8WfW%N%x&*i`o2?GN~WH_{|DANQBNVC~QLKX(VH3l43u-dCGx z?K$|p`qg)FQET~K-;&}_BdaE*p9StaA--qbMRtL=X>9r*yFBB*Qd;W;kF00@5?1A5 z;XGNs={$3tcx%*#EZXI$mTJ2H-kBLg*Ua^Q2Gp|U{xHwnSY%IJ&g=o_^4PnxcHk$+ ze93fC9S-KS4&dkjx)mz>#M?R|T%C92=G(yV^xyQ`dv*Ji}~ zrcW%RC8PJSN>Mz{Ds{b%J>^A1`~KmU_OlUjYQHTP%y(N$S!2>B4HlZ2gKytfO`G&8 zoz~NwX6D=RDynMA3R>0*SW(;3V-=oj9c@iI;Yz&WWRgqtMy-2-u zy_I@w>FB7QsWS0(od)nm6-x3o(=Pe?WGx+asgt8$K~&l{&}~2m#X;U+6~OC z{k|=?oomDcK0Q-4UTocX{?n<2d~~Z>=FgFS{`^2!tID6Nd4@4bcwD((tQ^ORSWB8s zG$&S{piTMtU#rcbuKcgge_LZpw6HqoerjbOFkkK0ZxD}Yzck{@FRoRYQNEV%V7G8z zmTR+ozqZesTVwo|Tdjh@5&mY5dK|24?y6N!4>n!u7YE)z}yed%EC)dZA1) z?a}F1@zb~Y;!kuaxhLJxt-e|Xeo{ZCc&4RYAB@UUVX5!qu!?HC6(!=6H$NQp@yg!# zi+LBve@J;H{>N0geb$d#qy8CKAnNblUE=aYe;?l~<=4KPyE>WgJ5N#%PQ0?S=I*#% z9eO@i=l(P@uHyQl@wZ0oQVSNm6}4>e&ZtoT!|~%^-;O=Lsax!#ns1_>eyHc$Qnp0A z>ibrmoF+$n(*NcL7q>{Fb^VaVTva`tdF0If_&!DF2H(VA(9+*ere&MdH@?8gIq_rb zHH=-;E@!173rAP(y(lVf+`nDbEEUhH(|>%arn>S$T{^6?*67MBb?O!RdvLv4YQstA z)N4BiM|`{KP5hhwWmLWUx;Ul(w%B+1!lLF~&!@(8k5<3`W_XmjYnW+HZKlOkN);Sm zJ-3?rhec}J4d;Wt-w&&M^X1k2?PB9rEi7oJt=cW>yOyQZ5$FE%HKX5m-K|X^j!*9x z?0=-WR_Z|#^FrN3TV2ywQ5mWa(GuvrTD2R;tcX*; zn`3_~u4YW1%G}&^u6eV3OY_n%rL2fN5oSWSEN1ew~kCB|z7hs=tqw&F>!@vhv#XKl*Hz3cl@6JK1iTJ5`HW!QPn`tZQw z9W#~SzSSl59eeL>GV*1EC04u6Rjd|AddAXQ z{QJ5t8)@cyJT!je*xTCX;6`oPlJ7Kgb5A`q;gT9#dWZJMxE|Vqh~d7Bs^8b+e0KfY zd;glt8$M7^zD%zdeOuZ3a?SL(iUk|#-zD1~JdomFv(Y#8dHEk7^0?UPyvd1@+T+5R z)%u$kTk*~4|Ans8^x+>1XxGa3)C!!>r}jT}-Yl2CqrUU(I`eM&RNCg94XpRK@@g|W zmW@Atqk>-Xd0{Qrfl1m9C6yN4A!9`ON>zgYJ?&_w`DS^r5nmq6usf(7pS{)G|J!rl zr5jW90oHJSab076?&mlyZKGMf#kU%o^P@)Sxr>%o&#ud2Oq3Z~i#m{P*V^ z>it}Kv?~RwnbR-5v)c6vx2)W4wfVQ=thoh$F^9ZZYyPt7lyCL7zxsMSXz#oH;U4dp zc_AO4=LYY-F($6b_VjALCDY8hb!%#2ga7h3+#n- z>(;APrXHV;H@0u^Jww~`tJPnr?esmqH!T)LUT6?wC7Im9IHz%+gT_)EdAUk?qQE}hK3XTn7P*A*K2=RPaoKiR5+{!&TDdrkK1 zm6|``y%sjqCw5h|uJn85yTqo~21Qg+Td;iU@yeFKj=-_x*H{+dJM{ zbyB6a+JD)>F4DU;|91Bpt?|=&+U3Ax--y-=td(t+@Vhe)YU9&n@|PX|#A^8Z6wiOM zynkbo@hamCITme7m&Mi+B09b+oJ%UzRYtUTw{@ z{=C2%RcIB@@hGIWDYeslU75Zo+}zBhd&$0FXkJ}Yi&rd1{^77#eAF^ z|DslY>+#9CRk}CG5cf1g^WA#S<^JDZo%H{@IZbr>%=z`HNm}tQwTPg<&IA3Q0WrZm zCEn=4Dzo*K(|hYZhljL=g=fTXnHFx9+FvL*D=Lk@*wE?zo1=d7&sGNUHuT;=>#_;` zw|LTdfaCH-=vj!p=$kk-ZV4$<8sCQ zt$L2ruUx+u{P)dc^_Vi>91)tUX6ai%U)k(8^_}m4UoA%;?5PXV?_%4h_wPT~UwXx2 zzb|(=f2MS|{C)eL^Y5>i%o? z`=yAf7ph_{DLPskJFrCj#~px+ssm>KN+q>}YYOW}hLzHCG<>i1>~q>zV9DvIGx>@KcRl|$_@?9{UN&0{ z_urs@r`2mdKiep;9Sj}eCD;99ZCEzR%6jHNFy_O1>-4Yb_&?33na|hdh|kz$w{K;Q zwR;B8@BbVW8Nt6;-i^Yu^g*7>*{ zljoUjmtHsf{`5oe!S8Xtv0G=G&9iIKIXBj7Wq#kKCdpILJn*`Lx@S#M ztwx>cX0ky`g5_5{u|_qTZT)&r;S+!B5g*q6+J<^lMppVm>!5F0wp9D>g=H@N<3BBH z35|D+`BH!R;IP*F^m(<{-j!D7fqC^&52ov7ixkx}H-2vQSpr_Yr;0$Qk;sxnAjx_aw_%M)%}J09M(&VM(UzbTzV-%ana^rFofee|k% z`tQ?B?P%WNJjaS-*6=c&ReS%>ni^F=tDl7ajGBE>t=;OV`7+{bOKBDK?fq@FmTpcL zt=Oa(YjMRiR+Ycjn7LCY)1Q1SqBR`fUOhKCOx@pekJ@(SIeMS=!`8v|2dyfTUz$bM zHR2;9N?K1=6y|lCEC}8_^M!9k)#h<0ChymL?dk0`)5WRZKAqvuT;pfH`@m)X+B~66 zEw|s|TjuFgpGMOAVV&pCU%s+>-x;jmSbJG(aWh~QIQzlO->;Bbbyeo5s+BjWMe@v2 z({@g)_ZoCci`!UIZ}jM>cA4HsU_|f7*5x9}d5;Ev@U*^k`h^Pxc#glC2RB?v!ZQ@< zZWbR>+#Z{+2ygml27OK)L;HFj{l4dkE&bDm=CP(8&%#g8pIvPml(%9!T(ZWDS#Mn| za7cZgCK(UW``y%Nn%`bh{~K$=DZk!kY(0Ou&<_3Pw!0d8?@Z?y+CBu{B$0wZHyV&C_#`-q;?eXZ`EAe|6=Q(Kq{-i2km~sn`hm z_pm*)HL>P}Z8X>Or@;@|mIt4n8B(cLy{O>Dd>MT&r)BkLPpI$jktvP;t9CEd{N^w< z^7nH3v$1*g@6~VhOMQFu3K_zAmt9qCBV|%6Z1#>lf$@bRcwE(B-wqS}hw9(Kk~~~{_p>+ug^^t?w{VI1K)Wgi(Ti(Df+@51ga04Xf_2H)N3dZ~pUo)$fGa zy3AZ(`bX`}>dPb5u_IKq_@5(!?+b?WMr&7S*Z;Ytw_LJaP4;{BV2`~Mv|HITnZ~7{ zK4xS&|Me9UxUqYvmTT5v|4&MUrl$GA8sF-;I&V}Xd%>i%`uV&a_^$bbc(|aX4ywGT(F<*Z(*v4f_uvs(4_g~XF0|K0{lym~Z>heXoWq{dr;y#`^*{FJ8yD?88^)U%f0}8n z*T-4s`=8Z&G@3!bFR+n)qTDunTqz|X#r;AFVZ%Dx*S^#3aX)Xd7tE%gA34624;Vg? z7mcoM&K^5XTQqorS$o=1zV>wrd(W(KcIl0|dEVbf@!xZ8;T1*?OjuUHNZ6FmN(e0a zLZ4BNeg|v231-^OAGA&*LRRj>X{;|7rt#SWug8b|HrIR@=$%mc`<4l6x$X&d7Ol5l z*Da?0^KPW|?B`VeWo!T9X`i0fjb(Z5><|9nU-PNF;5rete%5XEud}Jl$fE`Ae1G(` zQ@xyEcPeqkK3isqUH6$~+t*L>EujP4Tzi>6Z}!CAkS|Tb=oRG?I<=@~XH2-ra}PPF zSG>AkJL&7FWzMwL$}(V)vMT=V`U|d%I^@6LzkqpJSZCcjxu_ z_KClJjn{X!cD!EB$L8tALkTtPFSmT;X?v`)yU(j{_y3}3!p3q%6OtGa3GZgsL$KC53;X)C6Rk6Q(91PeBs7x^k_ z5v%h3g~8kj$*tOVuK22abxnJ?^Mg9+mkE{D?uIAnI&b9&@W|_aVRf1(Ve;>7P z>73nt8;{wOCCLNp!M4q2uPIloqL1!bvA@xKzdq??ojGn>)fWb=O8cipeezP^Q1sN1$qvHD(Z!OIqm=hYcg zQ+z#DOSsjUcQsF_PdAxW(!Ng??_1o<96I{GS*hznbHk8pJC8R@A9+6O4)x%&=UVm{ zzc&8h8()?~_jvszJFOfAn(#_x_FC7<&QUkKj0x8LEtXGORhH+w@@L$%fq8uo_GK`K z)%-3l3%v*P$3fY6OZ_`;d|5mA*WQd#CpYhnyLEIF7p3Y&O@03r&v_xE?;!oI>N3SQ zM@^U(3`XpW2sXO9#9EqqpHJVJR(mq`l+|(Zax3%Vo7S3-f-kI-MExP@akc%a6Tx9c z4{FH<&NbD9>{i2HnyAxXF;byNAx6PS+j`_Z9O`fOXT>fHZC%(CCdTrjd z$HBRCPMevUHT1gDox%&^^9}b$8ew<&%Hg+oRNf`A=??rD-YX*i zp7oden%eGP)iJMvYL$(}&G9>mYuSDorZ&2f#jIMnj(Yak3$x(16juMp^YPiouCx}d z8px*&?7)?AKUxV93;5L*RrsN1HTlroCAEhCO}4TXdS-t2Zkc*uX6^X29m<)_Yi;KZ zON`bE<{PYy&-jSf%(a*2y?Th>&6YEv^ABh2q+7Bj46Q#-eUYnOaNi%ftP%aQ^COW{ z`6hk1nl8h_V5f5f)tE9*w1>00Sc4lM2u|6XLhl=_#4j{Ep-1<% zUHqimh3NN1CkYlB8=*!tUC;a5gY*OOAI)N&S83z7F62+>cU1g&@3_5skYzX86|!f< zM8=*S)7MN^yskMrZytWA$0j@3n$7m1G}G;M`R?)~XEyMBmpk*{?-sC)2g~h0(q7Y+ z&1e#p;c%G#F4gd$m@v-lzi6sDZAUk|Ys+DFiRMl0LUXg|y$;mVzg^lyzt}2!LhffT z?djbLCKPG)#JbzIuhzOny7=I>(q`qup}36WOPDnm&5VEe%g4C61uw<-{g6H(*^s~O z+*^;>A4;FH>b%))4f`_-?~|{M)wRY@`}acw?e^uS@|u5@wJLs_k|&+_R2`Hvsov@6 z2h;2rYgbHi%6^_QN5V608Lu>H4)?zbvo5TkrH!uISg$@koiEQ%X%i-v&zA6_>fiS8 zud4DpKR2}w4$Dcu8$DA(?U{er6O+8Pd)yD0wKH#yzfiVzoa&pR9qBNKx0^rDE|m3( zZ6#a78?0$(e|vGI)qg`~Yg`7u#vhjD`=2MVmW^7X?^>VL-?!OGYuDf=cG%w+&7pfY znkCxO?*Z(SS#Q(trnYTvUY>ExSo&4iXRTqQCz$IyEeg(5&jyF=Z)IIbUP*gbG^M_x zL_RJ3)h0bOzNi}S@v9?l52-P?%!H`NXZ!Atuu$LP4sU(bIl%OEl)bx zY`-^~9e059>PG_n!C_l}fJoXWkKdrb-g;BJU-*Jn=fmo#`{N$z#h(mP zBmX`YdA<1PAg}nVerZX4|L-SWXoWMy>A92$yK2R2yviM)JtAR-o%8TYPTzD%NPFp~ zeZ2M&yIcQ~X1vd*y}7waE4JYq-YR+q|E19so<7wQQ_VAn|CfQ@k8n#-|My2X@qK~p z_Kj2f^zje3>I-kj>L=FJh|iv8jrnL)F|A0FHTv}1x%~gVy{QK_WYf|VNUoJi5itMw zeSdJl=yB$T%(u*fHMg4o#EiC#WSxT3@7%L~$vxa`*R!wn{hKQ0qm5s4WqwIsWNjh6 zPu~rClKe~c4IA$UBPXTQrmY^Mef#K{b~axUE&AhsR*mLwv>9VUzVkB!y!qoA`q25u ztlaNbn;&w1h|jxZq1F6pC*K+RbwnA)mFKe>J+O*=Nk2cHFSB-a&98c+2RpULnb%uo zFP+lgj%%&Q58OcSU!BAM!|AG#n=b?+dbE2Vx3>9PtwY=r{q*W7dddy=d{d`)x6a*I zY_;81)83sYw;dVE&NoyY7yn>ouekM5x8v#O7_Bo~(((%tBdj46mdAy~u86O@KBmgr zUG#6j&KJ|m(9cojUvgI+d0~@!?M)_MEWPjWq2<>s^^cTV{HiJRdqTPQG z4A%OgpjkUC%2fC4R4?wG5;eEvd#ls!Yd*E>I;$Ofp&h!o(faY9{lN&cfNw|sWqQGW zCG^{CXK6o=sKbBGlR+zeFNuDp#67d{f|S~*hrP9e_c!v!L*ld?Ws2MX%-p4BpO@Ln zdixdqdyq=@59w?0?G>M@H*>{kS<~Oswxxek|UGd0tH9(2QM_CwF$@gYg|;oX*5-!>Y{S0>Ek zhYvlp3YD3ly$QQW?|-?~_ z2ZFMn;M$ozyDG&P(cp>eOkm_Aj}c`4KBqnmF_-h~0DO`F`T%aO!4w5RJ)=6zkiFo; zCP8h$>&cuP#1i{ZKh!`_F9;A>M{}4Un>2F3K@1_4xDG&R}kT)DWi-ygQRN@d3(#NG(43;JhBf z<4J?_cuGlpE)?Pcc~<7w*v0ch6dOqQ@8uX2gB>C*p99i@e25|Ffq>^r3fT&CAot(* zsV1Cq0u*NisVNVy7YH~2GXR6&^S>BAl7<3tzz8AX*z_px*o(>2whW%&oXxU3Rd%;G&%tJo;m=PivghYW&kbR-yGsh>(>4|#cAeaLP=JbM%8s-8v zI0c#@byd<3MjY8fz#$QzdU!S7oDg&R$h)DM0DRI0Y2+ZE35O#_#u#|YftERBOD#yd z5@1fPaJAe|xMv{Fh8MfP4jT1-_hz z9OT0HKtAFjL7!{s_2$X{$fr7kQ3McFOTM(pmwD((TxYyqiSu&s8g7c7g3<{FLFS;I zYzgMX1bkN*BLfib#{l}1kB=$LK^)qkc!)5NW=PKEu@h=H6Yu3C51h9zZ03PEyh7sj zL0n4cIyKv?6I|t>(y**GXZPG$e0%-{L!hGx!V6IQvi$f!yfg^1|lfVxUAc8qO zX2O<28g*W+K|Wsl0pcKxNhNoZyW=dxNn1EEc~bDH*|>khR1KibGIm2sh50xdGBLk7g$B>e{j&Z0czs zj|8*AFrSe_wydQ%1To=4<`6G?5|?-`rX-5&kC@A$7X2Xd%+NGnanI}ZNQ^OxL74k- zOs34iKEz2OjlGzwB~Lzd0H9ok{-5niQCTBjuau(*2XYrOWi9HMYgYwH!rVTjF;n^! zlf9Aea?t1UBAi~{-GP$(fTvy| z(x|~5grml-lk?pi=4!bMrC?4X$dvUi7a;BJ2wu*jvt+JDiR7aPauR2eo!OmOZbM9y=`UwL0=*?ZpZPM6dZRyVM(9j6g0$>UH2Qgd zh-~Bm_8T1O;FY80o|r*qiPk zB%J3&o`H6kdw-rsaonS<1&^~r6OZRe&LCfM)KAVM1&zCqJk<#1)H341F#<#gvI)qS zv&n>wIAVZ7z+AaI@ra{`K?sp8&p|eL1|K#6VNR{A5l$|0@EY}UyfYCaIU$_g!*D)# zuECsIgJQCdH0FY51o@f=NBQV4^Ar+>2S?=oR1*+Rt@LR|hPi%yWyfE^?L8I2217Cn=pSS?|O{r!s0e7j84-w1(dLmy6 zG$0)9^#x}Tpe08&I5%bi*bRRU7-WYC=oug_bIFx+-I@R+f?^QZ$Uz+Wg5dRGWXn9# z@{UO}MjS*qb@Kjke)0u#=TKN7+v|-Ob3qYV3m?L0uIwqCez+q+wii#_jn0a^5XEpd z06TYvO&s!M4Ri9sh=*?wpuM|N;4{UkAzKRNBG1c3UxPV79qI-7&{CKi7tU)zAs_Pr zg*a)GM!t7H6p{U*1A-n1*ux+V2tt5>z2OHKX#xa!5JoYDxjbnIVuBE)7~<%G9B=Q$ z`NT6fh8o42L+4Y7_x6VtcPRL zc!(ft6cV!03mS37wc&@zMhsf+M04ahDJF9j(iCH^j~wsyjXL;=G3210v_dh&5raq` zce?KrWQ;gGPpCmY>J;Ya0C@7fo`|C!bsn$_1UYyfgmC)67tF~Ak3NbE^nrN2Fs~`) zgh(3%uLtrNYN^*i(o&e)lbnE%jO=5O?cG-hybIVKA7HKxjW}rlGnjBeK6dcVtdOAI z+Z8=N&-G%7c@#$u{@iDMp@%_$Mh^Dz?iL(!krzS?HgR6Aw82x2 zLA;l%xE#)oS_pCkb2RD!xrZQ|^01@)zCZEa$D+)kI^@ZFLJfdk;*+ho!YGCsIa81g zke~)U=70+k? zkn77{s3T4aX+t{LBPEQ$X5vRId*N-A{Tj#IOxQ^sJF}|%Bw>TYNVhi)h0gw z%%KCq=>u&LpcQEJgGLPr6L@exu1Fi?3(^6C$0KJT?uj6!n)7)>=1?r$<-PNwhu2pj zTY?n&c|aUJQ3D9_oP9Z_uV5G5+Adt-n;3en!;0-|>K)C=TTM9ID z=3rNu=k}9*C@yN-0Bgj6a5bwUgBR@b81k}r1ItP5&n>YnLom;j)?CJO}!LBARGAx*#WW{ zX^#t$EePNYSHt#d5tFmfixE#ygcFzZX_f(=ID^0d@{uD!wv1DrAR8QVJdr(Ur!efr zoc_=u*fi514Uzkh7J?q!Gk`(FZmO6YhKl z&k^Lw9Y`zUfA;a}O)-aZ=DK^Lb|Bdh^pJUE3qd|}&qy{y9mVAw@(luJ$=R-r zzLF#T`92Ih2|nJ@6JiJgdb)j4mndW-E;(0YPB`UaKkNdLTs1NTJnBXFB*7>qh$bj* zF7^Q@&qp*Kf?_lSy`c@#5FuQi`a)2TSZA6CjW{I0=m8*+GpI+R&79XmNPOyx zT;wHciiJ^6^z(Swm?Qf!g@T`ApYJI5c6*_QQLgL(+l6;F%nN}d9UvfX5D-U9ig1+V z$qq3mSAeH|kmu!leFKP*NX^`P5l)@l1u^VM9Al&i@?7LgO)>DOH%KEdL}-fhlaCyO zd`9qmg?PN)z$=6h`HX%{NCO5zxPTAa+aEr}>n9!dX{P57bY~$CHL{*& zcxQlCh!>;{(oz`Z5g>9;(u_IrfOl8qqc*XhaN;3z6-ePa(T0^^ih#IN1TxatG{B z0-Fgk7=Z*c3HLt_T=4YHvHY3yW>9U^#V5wK~#)MQKX zwCe_#n*;6cj1cB1013|JNkBMxh|Aq6z=-o^p+*oO-q{rKsW)=LV-|XZ$p0V0C?=eE zXUruSt{3g)%qLssLIV&XAD!o#OqoZs8S$_MH0|M?8+*wZX@z2pId?4k zk&RiI38+KNtHX1N+0rLo<_oF?M_Lev8uWx8g2oKyeq4KV$@lJtqK+_91le1-nka}N z=)nbJWRoyKU}TUE5ex!2*ve-B2OoW9J|mwHAP&6uy@wFljLw0a0Kp(%=8+8v!`WRy zZHPI!pb*0yFyg(t#cK(9=qEw8LhxQ=$c11p)EI&sLV$qY&4(#>$>=5}7@DTaDGe>(dX9pjIrTC|HY6bEfgp29XU6?gJ!(3WZIFh|z>~&3frD)@Mb7bNktewT@v=q`LGX6= zY+@7d1Dvd?8_cQyyf18=CqOx}H~BJ7 zTDUg8{y#=JI|(7-G7Q(Xu)fe{P>YQQP918i_O`{!o}5HE$av|XKeE}V}z05b!EYyx^1 z#DQnz!-j@Po4K>$3wi*UlOsq6$d@xn3xX6zkHX23vv5xY+@Z`z+_f3y8H5l4T-ay4 z>`T5tj5!({BO8L+00CO&P_E=~KFWi5TE;=3KO<0%0ZmYZtI71956M#unn9x)#pHP@fsrl0*T5)7fPgaym?is>4v{Yi z0Rr}hgqT|oThIgf5b$z8={XOWYX~>r)m~nJQ4T-?P7o5$1d;@`at7Iqxqmj(*}(&- zN3B;U?~Q!$-Y(EGmuSS%M+#}!4Cf;Kxda4x)rf{>pZb%RI19F5^g!@5?2sUeG-k^jH|G$hlJU50Q&gGFWa}Xm=0vrK7 zLf|t_aRoi(UW{@G-W=qIJRdee0Z0Oj9tg5l zxb;Lqgm7j_hk~jeSr&gYWc7%^wPvcy$v4>YHcVaG2H4yLohWK=b5ZUrBD2^TqdLf??NRSI4 z=FOKe=Ip?r34-9og>ds}ACHroeB{CoA&2ac>tlu#_NkXaHUwuuzCu84xO+$Fr4ShL zsK;HyhmG72fe`|XG(mp;CO`3vWP?WyW*`@~Cqj@#J-r@?2?DesNaL~rja#07{$ zU-Xf^6elLgM=dz-9s=OtQ$BjhTI!WJk2sl6anvA}5r@1GY!BqiJmip#7}C$YTTu!(KQybK;nT80HJu1O=KoS|J@EKkZY{)JTprdPtj66taN; zv@2dugAu=v!lY#$DddIGTtOPbgjDkGkgvE1K$9?r2ZC^BqCP|r1SyOj0f=z=8_+Zt zc@XT1d~auIlaD;b1DPS_yauozVt}{5tfja?2zd8KHiR*$oYyMNf`kM;0H29>FOp(T zZ<#~35RT8F$wyCU2ws=od}t4d1q9VHCs)>x|9L+80ScoWx)Z5gTlOIib;vPDhgG9` z0$v-5wG>kb0g8i{J>U~04^6yxKFJdmV#K-kL9YCmo=bU#aCr1%F3<;ko1#D2;882c z24FMNnK8%h830dE2#Dc(ZipL<_!KQOV7^4T_?1Y#vJN~Wn*d&F$}xoNL%T6S+RI}uj~*e|fLo6oJZD}V za^$_bcNZXy@59PHsK%gtMmgvs2myiuu+P4CO*>(Ce9lFlCvpz?sKY%Xhmj3|X3Uv` zI0QSPF9dOe5l0$%3TgB|hCT%Nh#Ao2V-{)?XLvbu20=M8_E{WxS1Bh9c_aZwpj?R5 zLYlI_oP`*3@6OGWv#Afx1~Hhkk954g6c2Z8xubCM;V zPLL2HULhbCd2(0OyE}n1D26!BD04{ToCd`h!K;CpqSPPDqSl zF2xW>PiR1pt&qlfB}abZof01)FyfE{TgJ#|WQX9>zTUZz0~pNZ$!67@`ouiqyqyv? z<%bB0r-j>-YQ4KMh?9N2yQA~~aj5Zn2V9%j#J%MV&liX|=VZ*;3;W7C(q1q28HYTN z#|};Do)8BQt+){6LqdZ3(T?6ddAlJ74mJ2(je1!_zRaO}6l8-}0GwSoF@-b`f@Tin z4D!7>^7Tr)L6~rA4bp-jecT%(o*)R)CJu4TXD%ms;^0fpjVmPRhhBnxdM;3JkT3gC zzX17=0HX(hBqW^YmyrgC9PEvq6Z63Vw2MJHQOK5OB%LVCy$d4b9FZ7tc0rur%@<^& zr#vTV1vbr3)XepXf*=+UWH8sJotW!;+UvzA2QrY(B5fgEr?+dP20gI{dP2N>)C36l zm>0r5J8AUCYrvpbz~x~h2Q>--IdWIhf_x8#aN>+KYN5UU6rn^1GLAW}A0i!a?ErK7 zD}u0ys~+KGJ9ONWd-+2{>hA&s2G{m4c=L~1+-?(+c7+Y_J5ygEZTd%KJLSzt1o1l>I^(8NnH;6~BCzzoK z0&@YI1iaTr#=uo|YQe*XNG+WF0BQ7+do!mVI|<@K1k4S9CqS6MgR@&_%1adD0)jkn zWE%vz4>Um#Af84HGZaDyno%C@2Oc)SpoKFF8egxYzd^Rlr(8xhd?3UeUm+VJ>zOm( zJF^hv6CluX9`y_(FxN+&;o5>UcnQpBPCtXRC+LlQ0C|FuCSX@+An`85r$3938vxvM z2Ph9cLIjBHK^i%Nefo2*GxZh3Da1(;^hgLN-hhp{qyq%`x`*xF6?iX(dEV~eLIl)? zz`MWJqsaC8`UV4w2QFF-NWB2SRbU{@tgwgRwEcLHDD2jaw|#^bzP zVw^uK+!*QwLB@qsBlq%pkd}QY2fdLmAxHt_1K8cmLoaXv0?sg(fH>}0crj1^KV-Zy zwq?1FWXS=v5Mqly1fM~`XW&EdA@~p|mKb7*6jLm*v_PknVu~dmvBe%j?`&&mWr`#C7Y=x-mZg-Elbi zb!x_*-v`N=clDz_=A8W=w0RR`+4oQK!7su&!*;gwNGyS?zD)V5$Kv9nEV7o&kWzo z->u{Y$2oV#B$vN)?5o=O)y3Nbin``^#m~b11u3*5KXdiT50@NDyESKKp7+xF)bOjw;fuJ8f#(^ovwwCw`}t1u)CV@z zkq_W!s(J3`-%)*sr}r6Zig<$P@%CE(PHo}UyGH}(YuSH?uV&SrrLOzzfu_jE!TSc} z>b2^f!q6yeca76hJ5XPAL8hEn*o=3%rw?AulMhdv0Ba8yLWGR_O+kz z&cCz%5%=!R`_SO|);0Cn1%G4bj{=yw)!RqsK9{2>HT*zxd#v9*faE`Tw*VM{sdAHhq*1I8rbDBKo;^ngoce%ZFc=cK|)YWBw-==V}sxsS%e;Tom9mdoOlY?K>9^*R`?+kTvMXVXiYjV!juIB31SfBNE&L9rxnbfS0 z6BF$>lPjPVI{|08v->Xc+)IvF>pEvpFXy4XJ%d_-b-dc2YVhO0&#Z`rwbgrG<{VA# zO+2Z|2jmaAcF#1ne~1EDX7}!u%Uj`d_mew!5vy(D%*Q{#e1U7$lEdkvg|qkmy1^i4N^P+ZqV)P3xJ20b(J@4wR-M!_T zUi%lgHfw7JzJS(s=Hl%sU|q{R&KmvC@1-@lXp`(84L8AGoPTFHUrv-Dw-$eL;|C1# zSnadbe>-M1f(Tvq(Cix=vyWpA5 zH@9CTH|||0+J6St&Y#}@y!TakBF~v{14MKDJwLN+ zzoX|(>*RN5`=W(&mNPVC`2|ux{L_1ft1arre|Z;TwSSy*e&hWd!V~69>RM{}*}I-O zb+mBK5x3Vl_kP_crY`4Pz31MZXyCQ`nUCJuc;a%V8shKvV&u>xXnfH|NA4g0ll7?`FQo} z^#1R6cjrO;6Abw8#~XKVjXi3eFK=A<2BSA}9(i;hZ}fgH=$!K$@0)OUSN=WRy;dRICI}pP2LU7t0S~u zKK)0t&^*Li7e53hulB+F_Re{&_F2}ew@!cK?1eZx;P2e2mK|q73e27X(reAQwqL~7 z?i_p4&$;r#H%+~^w^xgfdG<62kJs)yR`0jaUz@~B6b#5+JOgfs+6YfBpBUOb?k)wN z8SuW}BDW^*&iZTiyOWynfv3<3xOXmjG{?VtFylcNy)~^5aO#{B!$*skbGG@c4QCi@ z@wIi`GH}d_r%!G4y=Q#m9)mdNQ5zRwjz= z(brsfHRIWd5Swh`Ar#xw{T*7NB-*5vTt5hli`n*M4hv7U~P1Gp@H6 z#QJ=v@dJ3C_F0!-b>5r!>aquS*6tbAyI1|VfD4b;Zr|$NXK(tQMf|`|;O97u&l$!s z{N>`wOU^qLz~rAw`1Ll_?>oFYG~s${=gG_O264kL5W`pPv)siqwb?tqyFBFC<$5%p zx5zd1+THi(50WOYGgA-l@3p+9UOqK*?wCG!&0M5jN84ZT5Dj@9JosyO?tP0p+a0uj z(hP9>l%m&T@&aw7dZ;gD{?kPRFW`FceQhSc`?p@SczxqY2 z(f~Do_){Kr_d7R# zrlC{Y2Lu21d4BT`(A2CpI^U5#pHa!|+AIRm2RQQr*L(iEL+{xA`sgOyL#*C8)~-*l zJT@yXe-3r;ICBrbnLU5W_1O{c?7RtaW^%@TqnxM2nrCbE73Yi}aJ@Cr5#Kpf-uqb- z*Q$1{@nZJZj74njSue=_&N<`f{p_q`Iqy$`-uLo*BB%D=h3dQK)z6|1zHx^#b_U-6 z2Jm}*=UqO1)YZH*`oz8ud;`t;YCQa-C$FMF9K%~%MFy;?iS^N*)VfyfkK>IoXUKcT zI>>JDvDzBrHuj9he*h+Pwdcl+*5<)XAMC=MIZli?$Lb&W=qkCwlKP@d5p9T>H;rXUPvxaLMuZ zR+FQ*W}cWZ*8HraM`NG4Sm4gFr{u9{3~K-TboJSdTvI#iOXnOl_BR+HbI)0yv)%Cl zYP74X-u=#r;jQ_*dNq3I=YD$fsZRVkV*koDHP0%Lm^}OA<&FI;id>EHAfCk5HD|_} zpFo}5_0C;D%WP&o^;Ne=O=cfWk^6t!{LZfPUBUnQfe4$;kJ@uQE7s(|+5+F*7t6V| z&mFyU|J^75klQaNkdNlvOU-~^K#ee;dMv*1=E;2faNLUfseQ>l-}g^c7w>XW>V0o@n2}0Gq{+MgxpS=P2 zGxcyoV zI<4|^zgrUnb#>ya&HGq8^Jk4%>BX_9cdxp8CujVh^JceSE`Tp$`If75{_*azb9fZGm(+4i+{LlC)QtE0R{kg|?-`^O{dXekD$*U90s^fD$ajZUS zC$@fj>}&D`O7#AYebzmDz^l!7=j;QU_pwjAy2NPR&$&ZjPwg}AtVympSG-nxRybp8 zeLf39%~{S(FTRZbhU3haCl615^_gRtSkwEK*SfyW$}9@@Yv$bv_n9C0t>0aq+V`G# zF&gN>Kkj1AoYZ^A>v;Kx*qS|Mj2U-tpZVlD7k%baTR@FH)BJp1?;VO7ob>^U+|RX7 zo_QyZW-tehP^;%73Hbp2x2%>K)b01&+iPEDx)Zer_;*1yi&zx+MO`mHj(4V7 z%{fnRe{<_v&aQ;}p2VKtha2W-Gan!Q@zylCpYBje_wQq~A9;gGtVVr2`F+D%TW|yX zX3o`wEAqc7QJX-Gn6-6deBAgR4spot@n4z_T&~Z#+Qus;zl*O7=X|)#)|4uH9pwyxi3+ zYrykMkI6NN^_PASThs20_}cTbS6e-wZSDVEYT~U0o=v_Q&s*L9w`jQsXzpg+*`s!D z-TsD;;r+g8S7*(>?|ugO2Tb7itndRYV4vs8J98Cv5YzX^Y+~o%-E&_#`qbGc{BLPL z|Au=&&LsD&A|JJ3Jc&hk>V9W#ulny`qPzTqR^EIxHTKCo+#)ae+OfHoozYbv)Smf& z|0n+)1E#t6IvV>6H28~ozRx{N?g8GperzoA|3VXvIOq0=6}dKxcTnH^oN%58KfrxM zz5VJMgr`1;10C@M1?1it%yX7#eebgtv3lR~mwTFXj=BY3_-KjWJ>U{kQdf-!>Zf3i zCja?v&VE3&dT`&CEWQ6tubgEZk*Z1)Cyjk#SiduKX!cY zRsE$!nK^RV0k35?K75mNhVz@)_wyZdhCE#2yHAnJVa{1i09s#r*1~Idr&&q&pHofj ze7tqFo;S(0DtbDjSOHD?U+ z)ZnfavGW!%7Vj=Uh{^45dG@>g#sPEHxWmAwZj4!%w|9Zx^0oQw^f`mrI}~`(I_vw1 z-6y^FW3%`HJS%6A=6?8CG3y6hlMy@X)))CA7cDQCKhVpW=d2Uw9Adp^*{|Jm<^AuY z(Ho6-US<}0YTn~oH8r5EK0A1R0j)Nk#skmEEM}BB+OFN}pV97}UEuDFZepRKxqoIE z&)PGbQ^c|6XYDzjVgCbus>f^c@BNMK$E(Y^wfAcC58QXU{`M?1LbG=3U3+eJ?XOzu z)P8$k`~nYk;rre0ZaG_l#K(AbaEQ@A6Nya?UMy&(OY|^PMSY-F`LNEPkN*1+345 zfq!%6n!Edjd93)^fc{vcLGN6(MXu$%+BM_Vkl5Y%&U<=u^54(Q%wv{4_nyuu@>UNX zz2^e?Y&EI&CTg&^fkMpC2 z`}}uhbktfNEcTJJ@3q={%4^QK$Xl{UcIxeoV}-7n^<(>oe3H9&6Kl>HK>h*o=ehwpLrK&fockFMQtkoIllZ)`K5Ht35lhKN`)^XP3k2avuGu7p(`L?Hs=k-TQ3w zNvvk!^SL^gJT<&E_wmf(uHSWP2j|(^cfGpa^ZfU?x%NL!ELVkweVq54*u6A)_exy+J(+kl znZVw+F=^mD;CC@P`TB}I!yex{@9oamAM~iRR(NN5w&wZWRtR0|Ig9($>oS2;QVN6YZhK^3ic^-xx{d<-ulUJ-E(s`=V|s%(R-IB zet>Th9sC2VFZcq3*t2{a<4;pFgx2WF0IU4mGy3Fkp5bSa`@`qG;MLdh#^E_Hd*K)O z2JZRQJJ+WVKWp#&;N|AwJX5aCiM7>n(Pb{P)ywNM!rmj`i1{>&h%ZaEwsa!x_$Ofd~#BI?R#_QQO|qg3pgupMttu&FnbH=?NO`l z!9Q@?)u`**HF>pnCu3{c<<;kI>~pzM3N|mdjJ=mu{L^?~pY>Z?aJSZnpKw=e9MHGq z3t-L=<_kO^dGGs0ORRa8HN8-8UHF#z0)9u$Zw&LkX~FyFPNSQ3&cffB;i%;tHD}*R z?A+*xAFzPujM`Xwt-t0bugy8bKF_%4dCtGtkF9f#x;tN;XQY?Csz2%{mh*lS{~d2G ze#_A_h^@tnSa8NYXZ48@T%EO&KW8>QgWSDdjj>RBV(a$ZUUU1RVV^~O&Druojpk>S z9$oF)!aGmX&pJ8&bB21gcb}ZqTa!M{`^kJizk9EN-o6K&n%C;gX*{tyP3`E7g`c3! z>q}1-dw1j9nCCXpy<-uRJ5$VqKDo68)Hi;D?CyJMef~S5{(Ue1Y|=9K=$)ahKHs(4 zGmCr>i~DxYp5zbS*zc7#`|HfykK8ka{c1IP#%u5WPAq?W9%9XN29UpRI@#-g1OL6e^By}9DS^b zK@9KR&1dbMn^?PV_B1iGd?)9O2AH}!FwcDQRV2W?@pS;38JTrB>))8WdiUKL%6%4Y z);?dY^?UA5`aFNY+n+jT-n~ZV|B zo%j#>ns}Id4i;Yoy?+Nh$KQ+0I(7RzGrwbKt=Z>n`}*eiTYLMD<~fpTk3bKM(l{FQ>K+20k-ud*cPVx5l2lG5G@5ykB~IpQq*_*6tmvNkIC! zyVkw2@Sf%T?A%}T3(nsU&v@YNky{6i7v`h)p850p?BwjcHD{JFcxM%T=zgjt7SSGW zuXsT2C{OI~Z0?0Ga(C9NnYDAh`8PRgec&0b9&b&rSsy^{yKfzi_f#hyFi-qBbIA3c zqt2S&!(X0Hzi%Qp!4QWxf6OuGdUiF5bzJex7-w!+l4$h=(99=dp*g zwCsd8e}LwBUHIPhBKFK8UqH(nJFiGU<{0;RYi2ye4b^X>#VTzNfS8&G{difrUSMW9{yK^Hq=yFVIeNt@ATyJ;t?>{*y1f zZ`*kLzWvx|e#e_<9<}K31;TModdOqPTZ@L=colC53bg2rp%JSYwc|x@Zw#+ijpp09 zlYj1(yV)mSJD$DI$t(-%sKxLLIOD;8&a}@zKlE)4+&54^50}rz7*n@^<{r_ZwKsah zMhBd0T(!ST53!&7HG3D_A8_g#Xm`%+J;QUHp;!CvLw$hsQy(C`aPrQ(XMCs6_sh=C zQ}bzk=EA92=J;#5=i#%=X^PaQFf2_5PpS?fq~0qv;x%x%9nb=h>fJcVoPG&5s{&>~=q! z31_UP>dU`%IL+Ue&)&?h-E-gEo`183sjYi=I=V&8Gd2C&Z`?bd#LtP{@x3Q}?mFm| z2TSkl+WFO*hofd)?d^@X4rhN6FK{*HpInoV!}#_&2 zvl{qacaFUAFYpuGd4t;3EaC=onM03r3QQ1dV)n`Xi0@~o&VF_3&7)(+_0ApcBX?#; zceGTaUFK@=v1Z)C%`>yM-Whl2hqyo=j!Js<>fBLtzBA<}h&4Y!?nbC)71ZI>M~lz* z_Tw&v&fbB~?q~nZg17fMe&W^HGl6|~pJfb}8GlpL>ge#*`}2a`?@E5(_ph^$HG8zD z_yO*Gz32HkILkO^xi9fLIhD=V>BZmg$(^lUE;+G#6h4NpBEk!`nrHcY9X`zF?haRz znYC+?5x*Sg*zf$@o%qgkPI&yHel5G1_xZhN#y!(J6uby`Yc;mPYxk_H$v*PM576NA zd}D3)>8<^~3tzwryQy8^Ik|JR=#KyVKByhwv!1-*F=yWUCT?=L%r)j6E(rP2)@%3s=dOEtj-_>=_ zp&IR~-TP42s&<|?B98Up&ip&q+I=r~%=<9kvy3wnUj33Y7FcrTS!?i*Ipo(e)4cX> z@!mmeP`lsvTg2w!aW{T?$5mTfXneHCPkgbz|C^M*^Y`BBC@HSBD$)=6lxg#D%v{TfOn%ui2~l->I=aXCMFWL!LQ>*Lv%+ zC-HxCmh;W^t4{szvje$yKg0h(?>#T4_YT$v{ujNgO06f5ZSCeB;Z=HK( zpF$4?yqX2(@rmo4#XH?Q%4_Cb1P$=szlo>0XCBU)re3X@%N4P>Hi@DoK0VAG zy?gR`5}RwYHlicW+@?T}b49fE<=x4*cy0}!?+kZw&H@c*UmeKjE_%2H`au36PJL|u ze#bcHH+c81-CycG4>KMhpAFtQ*~b`-c@wK~594TvQ+E#Wf{)dnX+D8(G4$9hegOMY z|C~?Bn%cKcZd~|+_s=Nf2e`wH<;}G&f9!Wm5BZb>t3Td?IDhWrau53LHCFqRy`SE5 zZ{2xff^}e=v+GN1tw5jJovCi2bG}I4c+QLl`qunB9&2YW;~3rc5}6t98CJ>@}7vJ=XQr zYi@t6PLE#v3EKy_%#yoS1mFRAE6@9J-hlXNa%=YBtM{KBemCWAZq3=e&BkV|i1~%-OwXtfEN#_LBSkwm-9a zdwy;%!TxdY`QAlI9RV{Yv2z!2=IRli8QY$)BPAd*Yqf zkoA!}Ukce_-T4oe%W)PI}MQoN>>|nWrY_UYc`9M} z19I-IF=L?BzB$YMT7Bfi+U&1>`?3?tGCa~uxwTUl~_d0J-@P41wT>iZqv^%CpeF4n; z0sgn2XBc}<`tX|k=#6CzoE7{@se!!KyLaj9^g`QHjTc1?Z+~=m7Cv>H<_^^f_wR{$ zA&~?3e0@v~+@6=CFK4Q&uwVBhPk-;;J1b|a6Id}#%u{EqnX3s;e9y{E_$oxexo_x?yHlS91fdGmggkAYQ$B zzauwa@b=|>>AeR;`wE;AevZ^a;(HXWG+1-tS|2iqGW-HN4j4kBvno@uQF zc|K#jGhVA5sY!qDK3VVJyr!0U#P`h1Q-s~#QCsgloh7d>I}ty`a?ZGC$=!Fc|JIGw zTtCsRQ_ttaE~z8Yy5shlwRgUCdyO+2nf(J!D|UT)bJn9HufwTJAGHR_;U=-#7tBmQ z{&J|Jzx!qmRD6BS7{o$8wWj8dTJsNB;O;hweKR>x2k*V8Ip@xp=Z^0?qj#2P4q|N< zKR|l))_Z&7CjQNwmt#M*pZeS5oI-C;W|AALUBq$?3fvh#nJwr1|N3$6We(J+_4_+N z^WOYAoaaXeKY{jiT=p?4^6YF}{J_%JyT|W*thUyD?##?0pVSuqFZb~c4yRV$dg0rg zSaXKFW=*{JsvWCUlXt|MJLBpfa;*wYZ=HUSiv{ex{c;U{o;^p8yzj!_vx|D-Q&Ur? zF7J$axb!h!jh~mirmmZyfY$lB_aJni1-EL?_Du5v^AqT@Kl2_mgPJz$@YL;z6=g6) zf5BbujKtLJFQ7heGrs0uEmUSdy|w0AzVrBZE<3D-sE>1qo&WA}^UL8UurIo?t>M1& zj&j!DcckoI)n9t!Xy8A0`PcYiP7I&B;+Nn%aL-TPgW>;<33q3XrVcX4SS#{AATDAx z`E!&!v+)J&YkZ7*b`ya+&)Nh_$p3k2hd0-%&YV8yk;~ot#WO#@kDNJ(o%;uO*8jpU z^#|2EF7lWj+p7G%Q4vyNhCVs&c@OM0CFynFqK6JD-QPW_YgemZKy&V>-RHkwopC|s-TU5aP|tbz@AP`s0-4R-&5fO@R-{%t z4o9dt=k{G)3U$uFUw-!41!@sD_{sUkcYeGx;qp%SZ)nUfH?18kYD17~-ckLmq%`6C ztGBNndRfCfbG6ad`DczjnSnk!b7trBFlNUtKy3Xn1_M-YPF}s5BDN;{T-6Ma&lqlk zhdgHxho^TDt1JA%-}B2DZsLdD-Pq|hxb8hOT55|t8t!MWyx*6BPmI5Bo^y(tR$H$x zpxyc}?jpZHIBT4x?#_aGt@{n%yDhk!Z>e^lvzoY=Z+(F4C;XkQS8IPbq<6Q5voBU> z$0ifC=h>&J&0NnV_Wjiv=Zy+4v^ulU8dq;j`~r6;YZHva{l!c*0ln_Is29+>`~m+b z9#gl#Px>0Ksi`Kw4}8AsXOD2w@Uc3x@M?X__if~^6@j3>de2lJtFIpOyW6c5^?P15 zF?lo}UQN!Wo;%~+J4QUg0`|Ly`oyH8chBYYYUE!tuHEO^zwd0gDen99e(!~k6+fH$ zd2eIqtJiDRR&9UZx|)K({GMEXRH(iC*2_8`CW$f9#U?esj zS3BXa-MZZ6KW8{cO;H#AJ`cPab7QT0>x|{^{i2bCbH9Zb&Pna?1IsS;)th_3FF1QX zcOl#Nt*y7F#+lBEejF=2kh7L^pfvU3)~dedQO>=rXC^i0jK=AY&E7NQ^LFyJdnVrg z1@b1m+gmf{du#1-YA09)hu7{~s!5%`+@q~xm7n<`Uw%f|tL^i9j@PQbX5P2(JmbDK z&p2Gp@7?2054^tg*u9s(gYT|Q%)E(@uJ%3yFXXHlKlmG)J7d6U)p2`PQFErd9bVld zZ;w_R_NkdCU--M*JDSn ziT}}8i>GL?>|0j)SpHS?j1dEh&~*#vonYJb>fBBu3mrR@Xb&3 zqKTcM883M~xP8`t!uvKwJnQtaqdM!c+OeEk_iAE&>CsHL_c>c`S!0f7o;`~Z4Z)eM7j@q5GWSnL>^4?Rz?3o%4&DcK6>IKYJrP82{KBeUabYsKhd-_vf$XJLhzy(@yoI9x1p5C(; z{@0o788J9xckm4RA5g|{_G)V67nop(J|1WO&S?p(_6)xV19*M|W3^Ry@^9qcSDmq3 z>(OYBF}qxCV)KKix7K_RYtGT+RuS|*gDB?NnqD|VjoN#UB5!i}2cLb;UU#?rPN)(8 zQnTM)wV8YNn*|Ttt?=GkZE}1Y18aAG>&cm=t)^C|PmVYovD!sG@WS(da)xj0jzR6& z559mp_t5NbkhAg5zH^A)-^Zu-?3`uc2YCT_Ir}y926?kO=fOWGwjM2hf#2*QHx8_I zeSA0Iul~+dt47OvH@^0qhq%po2m9{4%pd#S)A~6_{sA$30l5j}tG@1#Gftmx+|;$4 z#d+zYt{=5$WCvsF3x6%&S#tSk&pW7h_MMe8)ivC)ZgS75{f6mhP8~jH8iQC-1L}ll zuO=3leZS!RPCUf+`?*-32lhFm&GGW#@DIov;LF*ybNS&M%)0&dj6QQN2bF$nTJcYx z?=b!^X4osI>ECx_##7|%oHwxs8a%-N-6H?ng=diu;NMlPf1@@F4$JvyJp0-r{;XRw zFR*&|$i4CS(#!eI4_>|Y?r6+e=IL!>5G!^AxeG+~sp-7~HEq?_glie{;)m!x*UvO_ z$(>`5pN~Bcxw?ElcTv*B0m_aprJJ@~k1&O%dt-kEq2n-{U4jd#iJaQCgKWMBDk)HM6F_}Vq+ zjRW^KP7m=f`^g7!>&ZnC=T3LmI{BZ+Q;+88?02R+KR|zXy?0zaD4ZH|<<-iK-fuPW z<_(O!-|f3Q>oqlMtM{x)d^J?w&qU2XcA@4>`?R;l+^5bvGzDO-;6DB3Z(nlnf?jQ` z`3a_Q-@*gy%NVYE`@A!pcC}A!eP-CN67zdipspyW>Xzt>AjYzK7iJlbrdz-_38|Lv4XRxI08Y=&QYt^E^{s z&SqC<6nW$29^(AIjLqZn-~4yBdUe^$eyt8?%)ZW;po$zAFW^4e2~Buv`q5jf{x`M% z00lfpuBhMM>u=A%UJh;MA5P5!mgj+Q&MjhW)t;{=v+xVd<45g!$+$eHxL{|=`% z#hriizegcbfS9i4h>&JoU$vz9vCdmengXUAt2vEFy}T)moAyU&?97ug^-cfNS7 z_TNMA?59r;d}fh*#yZ|P?)1*3tT|iF<=?&0E^6sxH}`CH;OE(wYBab1u}TV!)n8}s!a3L7s5#@#nQCk6@B25oW-aHk$1ujy>t3~Co|$cx zy4o@#u9F+1bCz=oBqz<>+PW!EzkLf7i1pFqUn?T8nwq$8NSv9tl6RgvQO{=uXWcXO zYGTDtu=K}1cbt9B@*cHk+WYOE3;mFPzK8Mq9@GZ7J>^>nOfG+Rv1b$iw=>P3H*PXBTTxZ;P5f30gix?Pdc@z3$wR@_`Kx%mB2(61B zz<&2Oww4~#rbcdWthd&|(p1lh%!i{Zyx8w#`UWH5|wuo?$Ex9&l|Ds~zCh)%Z^NjL(}y18>dWz1euo z9!*|NbP~k5Gk(=qzQ{HAy0N~%eWT>{HB@r`%)`&(BPc` zrKw-w&QfQ*^nV@@+dIJnoNevFk0L&2l(|-*!Lij_-+1E~zUm8?)2k^k4rUP&KX_+* zwwft8`+{oX_}a5`hu&Iw`$g%oS*NGXh?yPDi9JKB<_Dkuz+a8MT4oh|Z=L(7vqz2f zMeKLo@BUxFzsvvc6Wga=%_w@#o$Ryr6J2JIU!B?#aE+g6thw8xFY;seJH2|J@f)qN zJ!;J-UaQ_S>|bDzi$4CrmoYuxfi*eJp4kL?_Gr8_^lI(<BD*dd{4&y{G4w+`r7mD*X;jBO)sY=p4fZ*{(&dXOln#eKR{;Ut=aQhk%QGR zcy0F9HD2Ca|I|kIH3yn|topi}^Xw7cr|>`Ve*n+;g>N|L+&;ZE^_n&BI(pw>){Z?( zowJ=Q8kk$tcj4)`w!rT*)4meq;ulaq@b0Ewjd=7w_I{Z~{V(7d-PC}yCNNeL4o&K$ zo;#^IYXR#EcvkK~98bx*eFeTzS2w@|Cb|5mom;#7Aa3v-jr(f1?m5niwo}TCNgS^B z`Him=>+RQ!!+SU5-g@iacj^x}j;SBLb#090ZP@L-^L?vMIs9Yn{`dyYSzw6fAH2U; z4?buA4gX=Y_x!@E(VTNxb2L-h8AxAvH0=C}5~O>GU8{buCa>bU%#x<`yV<@c6-XYYCHjC0qBh>P06gTK3F zPVYNjK7Hq|jn?^fdU6Kwn{%$`rI&jZ@uVK>tsB=@*u9sr=zs1Blk==)4)vHj5W7oy zIQROeh}l;ZvHLe(OrTCQcXI|H8DWWBh)*0!HEeP z4CA{~?VO)`AARk(`k&t+bIl!|PPq3muFZR%=I8JfL4TbU4d*U+XUyZ}*U=TR=lJiu z^b%|G?!5H&g4db|+&hJ@{LJ4EZth$)fisNt^7w^!PR=4Xo`nbD%&=1I_GijApU-4 zK9^@k&Y+x`=sZhpX7tV{c@sx#4F1vQ9O~|(u7n!Tg?otQ23}6Bmfh`zvo4>x_#x_i z?rO-@hz9E3K4YLg9k;c0c)Y)R))u&DOk(#dd;@D*cKmyD=H4ZSH@3g%8-29b)1zH| zd~MAy^qG719EZPWsTJAH`0O<>Id`+}o;jb`e1YWhH^#eLfl=`8MQpF|ExdQmx*pzn z&QPP?orR`$HgHyLpWP3ani}fP$b4!K(5!3LstNGg?NeW1^*%r6F;AVm_1|j9tr_2) z(hDDh{{}{H{B`1J+3l%)W_D16&DygT{wI2C>hD?Ubr;N<`~cT`XLo$x=G4vGxOYvB zeaZEh^TO$BpKES>k@HXO)p%YXJ-&^>t$+Km-`?Kuo{QxyeD>_Gy7{lOTh8)6)a9M$ zY;83^3g8<*@y36gUTZ}zDXHAaU&7&c=$2ia#*Vg*yadezj@WOuQ_#Ip1fBYwTZT=z8n_eBfv;4b& zn$ZCNIp^}3%WSQRA59Nfz@F?)e9w2r!wyrNzJ*S$wW$Y#*cmNqbAA^0Ef(CT`MJvd zRD)Mn`W^%a)Qq3rvkO0fJ~8Pcw%^{jk2qYP|E-4E_Ni5)b`pE0`8~Vf9^h{Go%+<( z#qi(Vy~*&o<9V0NgLbxj<`mk40k2h`Gx1MBn1glvgZI3~dzO9zInP+cb#mD7snw3v zxc}4S*eWF8-ZOPHn!M(`%*1P-{AzY6$<6=8PTK5!hG+ehBfm3N?fm!I-(JskPQFK{ z?hN1Y)}6iJ-d$^-Xq*aMBi}#hg?27IL{iO8n4wp?;9G=u}4$?-QG4kJKx>@q^&}~cT#tE*_HEC zUj3;rs zIo3bnz3jJFjlFXDZrX!&#f8hek*hOay?cND*!|9l;hi(|K%e-`zdOS2osV`f^V^yG z2mb9T!v31+9WoPN@LKc+y>~IxW`Di)Q{?P$?-Z>&0nfcVTbH}}AXlp|;NAP=SEqmA z3RwRi?vKf9^7dXo%)tYvt%{3g5T_5X#&=nI?44J;$K2hcqu%iDo!x(Y?d1Hjr}tag zCmy(1lh+Q_X_?am-lYH*?(}x=8S?g6bAINSt22ksS;U&$0CLW~n#6}oo!Uaz2dO)Y z+?j>HyH)L6&iVomoTir-i`e=A_S~JviS^;BH&9#p+8CHUe&3$MrQQ_4y{j`8xPKp7 z)CT#)tG8}E1U8Ew7fsA(*W}KN#`sU-W1nxAKKa_3&+p!ZTfj5FG1FN;+5K~toWJu` z``spaddTgUSAQ|EKD_bW#rWx+7v48Yjo2A6xYhIn@21W(G~?*-atqk+oXmpPGE?9D z0O`luEABqY;jg{tajqKkMcp|C{5{Hh(-2Rdd9^jSr^(fu8{fHVowx9QFSD~}fdQxQ z#sllm;q23_UpPa)ftK^y0N|aY<*xcN`|@h;EP4AiXWri#>rd|)x%;2gFlSj?y`jb) zHQs5EPvEQu`HthWOOqcg{Hi?*cJEAX{eBmnQRJH5bLEEU(~A#x@>TGl*v0Sqf_J}$ z8^Ag4TE<|KJJ;Qtc!)Z&DY+UsZ56q1f!FMf2CfRC0rs9!(lbeLlgC zIh*qvjk`>sJ~PZ~?~?nP3;NWmyYpbu^VF%E5<4e)LT6=9d`u5MXH+v+{s$#uF$HHD ze`nXrFM00cjA0BK`1_YLYx~T1XY9YI{XF%mQ2QRvC%;_v?p=DpIg|PT_dbcK*<;-v zp{?Ra1FiO6-YIi>d+SqhcKXrjwfH`>@y>8&iROF3Ec=Zo_lc)h%>?HE0pIw1#`sa- zSoZL&{`meBIXznDHGw$uAL0p?{_N@JaqH1i)2qAhkad^z5CQ>4}6`tYP`MkaB3E@dXY2X${d@0>P14F zU5FR)b8=^7ANT>*@!5y??yp`;z0Y0n&b?N|zP0mi--Nqo_BEo1sa?k4nz8!QbC-cP7T3a)M5DI8X5Zar@41tjbuD^lVs|HZNWEn@ub2D0^XSt9 zA3Og2?wU6m<~MO>;oV@y7T2ogQ+}an`%@*z@izYoqxGi~?Bo?(=th5$AUrZ_fkNYn~DG`EFc} z_mWrl7c)HDbHb_lsFOx_>+#k>toSMVcw_9HQN#`YhCFd`X2IQA;jJJ3?&Hh`w_mT# z+Vk!i&c3tmUeUN?^AGxQu<%dse!XaOwOZ}?YKq(*>w5EXe9nijK7Y>RC$+2fKJVP| z&p(SV@7V)*2frJ>jk#yYExZWVoBM`l5vR}dYo9am*X-@y|IRWNMXrz4*IBdSfp+hd zo+dBZ(^q@O(_i^#ZBTbko%)?+{>!_4@M>S9mi^%FyOik0q*Jlz8ta=_hD&hV<8sGffC0e-b_fh-owa2;h=&~#Q z6Rv%434&*mW)6I^p% z)y8K{et_JK*`8T6zds0k+@?TM|~6OR78&EH%5a`vEce@$MS z#gE?kG6wf;<&DG1I};SKroS+spv{>8=`iPvkFA32bo2f8Hamk zH4>nyGtSJ9c)?{3XJ-bf_G%km6eRvN;s>M$uYT|ud(Klx$#W*LUfm+DlMjtpo!<+2 zO)uxXDd;m+bNNN?_jndywsUgEN3Pc1?5Bj>KlAR4K`yNS0l$pvaL2ogvjbc_T6=!s zwe+z=b<({Lq4}LVh_k02e9oM;-=+1|jy<*c+w1PR3w&lYLgLbEwa>6NjlH`u82Z|u zi(Xqri|Bo>K0A(Gd#I^>;Dx3>@63Izt=|2Vw)rMajkB*;GeL4=d)=+z)Z^c`F>W}l zdn0}5&BuT4|98>!*3rV-^Er1>bMA9|W^k@nb?$>#lU=B(UBK_z#5b_-`s^r&*|U20 z+;fXOv#3pD;JgRi*u2cuaAx=R`7XnalNa7tZ~tq(IeV=|XOVaTE&CLGO&xzNXOic9 z`p1diyiYGRXDr^K<^IMKJ3sS&93A=iM17qX-DeGHKI_XBIk7V*DB$B#P0g>{3ujHvSwc&c<&&uIX}L4 z|JA=9UaRuQoW=AtkDAJ+|mCo(pvD;0v9Yn923xqZ00TUl$?k1oTE8?)z5{q7y%r(7+%d~ zF0t@HA?{DGM`yw1{#-^(3%!Z+`)%gSU{24=9@H!lOwWs)oc7qP!r1y)8COKZ^PH&* zZGN_>qq>a2t07Ji>wfs~%tD>|33<-?xu9dg^WHR#QD-mr(MJS~{k%7|8HXA6#2$s1 z3Hn+G&v{F4{TVQ?wFa3^3|=+8u0_r|J+(uuJW^V?A35%(rOxYxvsvT-js4EzPS_ba zpL?%bs8K5*=ttzxr!&x{25Sj97-K$j*7M={%s8Xwj{1}_VKD<8#_CCtgHF-hqV9kM zh@i=wvnf1#^SB4lIylyy_o!sKWrWyf+Vs?*Ig44s`M(E-|G($rp3txNLJ#lC-P&t2 zZSL0d)8lRm)KY+ZVt_iUln#iQb|o=TQ%F;fy2 zzK^U29gV3AXg_M&M~7H79X!DIMfaeWx(UE%qW`@pc5{yP)O+^m`v;D_gadenh@OD) z42;8bQja|0KKPs=roMiEwAp*kGlw~9qyW!EW3Ot5ec*V9#IXu*@dCsE|xH> z0ugiQ(FY6tth1&%h_%ONbD>9P!8ssyk(x!EGA3YrR@V8iYvio6Mh(u0M4*4B1aGCIibh*z#-?eqCR=;v02Pn&MDl{m_6E$nZ)`Tl=)n#8G{RV)m_l# z9-8j32!O=FH^{Uca$0y+k<(P~p&U|cGmftbVm$}6X^b79bMP_2``C=5K6`TdphND! z7u7~7tJzG4b-phtJZI=5fcoTB(`GePG-sdId|ma?lYpmjCd^SMvdHOO2FC0SHL#jf zS9{@e8rTtZxrZ~{O^!Z-=M0VWvCVqa*Ux_TU@ak!+`?-N9cr;J%+XU%KQr*m*{^lX zvc^0DbT>kBF~A;;(Z@WP?%=Ei@4a)#8z7(OIoyNuDWE+lM~G?KgT4?a4v}RBvq!-> z*7Bhh^SQu+zUorL71@C}vDrTM<+M198lM66BJac5SyRA;IzpX~9Xvs^&~^;69U8do zd#UaMQFLCam^x)J$1} zS;rU8A!dzpv5&(%xs)70E`sL{?n@98YSw!=?RD_b`}8>O=EZT>=AkN z(<|Wg4)8R_)S?dc)m%0ccG2^kxjCElbA~*PHD_@qAaa>zh*RV`>v`6h>$##LG5E-0 z9G=iV-kUMaWiv57<~4_BUG?DAEI5FD>gW0IY4F_XP)k_HZ0h@{#oomG2yxCByKI*8 z)TL3wIt1|K^h}6pe2<;;b>w@UM-$C$i5>iK)VJ{x-S+zHUwn>Y<~cT)}2 ztv&kL%Jp*=GpZ9jIL>fqgfplfV%_7wF{USYMi^_(A+P|<+U)*}6T~sVn$97l>AmPJ zUVy|QhE9Sw*9U#pm4lC5)}!Y~tU1o@e4H1~8~QBzQP-at;c5I#vflzU)$(4S1^TpR z;Q?~oANiadEpd3U0E}Gh7y!Xs7EnuT@Y-jO@{D4ibG9zdrakkrGjey)dFDBD{(H3r z69A58`P`t9drmfssa?n&tUKQWP1R+OkGYe*DLl2Q;}CNu0uq~f=6vRs9KabGbryq9 zZINf^G9Au@eGy`Z9z7?#n#=U~GeF2waIwu&zVbPJ571PL9%rPs1ys`>t&xk^4^S%t z*ptBPv%q=D5!PlNueyi&+T;B5hL3?*r&fGtYx_ zd>#crIBVD&dE*ZDr2zG*M}5`B*@7h&b%zXafCZ>Up8(u=VGQG3 zc%If}daQ9rn5U1x>A4XGbIyM*Y*y1gUkuNHbCzHaf!(y!9i8DB3h`Wcj(k2J0VMZsM{XoMCJM_QZi_j2aOkrs+NAWu6?*$oDkj zImxrv`R_sS4xk=+jeTA34W4E#YCr-|hq`Lcp9eEqv(N#!&oTy29gXSH&w2I+{S;gR zaE@nXkM<*Hug25?B66_As>|s+=78$>K8sxUGk2Lb>$F0D2KMOtHxDFY=P(1*MLnEP zJ<2reJdarK#oX)X`YieY@+rI;yq<-7VEVQ0JZIs->)F7i0G|WA>Z3#x(8WeQdUyIdvSMcZa9N z4lmG7;kkpJGlX?Ve>Y{UX7Nh6Y&M+3f+J|;c+c2oUOD8{VC~oievi*1^f`c*+rz|> z$o{^PcP{{fh4&H0qIKA7df{Qh*vEr`K5<X3uuS*W4+k6?(c z&-FQA7j?)HDZuZ)LB9|+o;P+Gr_Ta2{9Ri$!AF=O!mb3d12~g2i^ze4PvF^`4-L-o z?%_SrIyla=ru@9jbH)L=yZCci^ihMg6rgb~abDZcbIyVzXzWQX0)QH7&U>5ra9^DT zm-CpV4)ZY2GeqE+E02(-&ky~|!zUJ@2UvY3^n4$jgSSQIktZJ1p?(a|*h?Mu)AK%# z9%o{WK7ps{b3{FI%(9$%bpiHR57k2K0GjGThgz}oI*9Wz7u$dG)X?V+Ud?5B3a>Tm zXK)U3stGObh>$x#4W6Gt(?5}^xRQ^8lO`hdoUZoXHKpRpjzmswwm=?m}f0= zaPUOX2b=rf0BR3vI`3G?c86K+2xrz9x~k=}{Rw<*F$Ck8Rmb|B4kW<2FhfscZ|pLj zIdv>RtIK;iKf^vi&1QYt>(8ansy=#%(Cd5-F2W3V>VDLz8&Tg|?Ed*a>G>>*;H|Hz zE;(4@)a70@2d6vW!?WUSA@;ByTYonA2swWrOG1oP&3SLpi7xe5?O?p^9cxQbdeB=OlYLM5Mo@Vt`-)rjedFu0bhzXc; z>U1yrxs!d=<9QNtxPyaF!BM9WP`xPY3wNYu0?^Z}Rrim7_mzUHi+P%!h5M{O2Rvt} z;Q*SR$7TD-#Wp(&d)Uia_K=TUk@o><9`>+D_uvk!hnQz@0Pn*4QiFHWeV9pr)O&Pf z-^(59;6e|z3;cY_8qWbl?BPCmYSWk}RzrP>5pzZjKE$fQS#TCe0G`YFoc{kBd&1sa z49?nYZ;Ut}8Wx^sqR;gqkC3yMp5`3p0Os`8!@=i4r!Fua;2o$(kSEy7vn22-Kwb7R zW?uq?JF-TEXO$vmUjcM@-WYKNIDm6B_FCsPuqQ>FGmFS&n)K`q+Vn}N*+|b4!>b;1 z_y6sU$isBj!ebxd5UaUNld;xp<~7GY>-^UVe9+)5d4l^LJf9oyOfDslS?;DMw1#@b zXTlnc`&0v-y%u1No;)pfcmaYAd4fFqbPllt$niZTM=QD~ab5$SXNj0&FZmRph21*m zKmf6Y2lTA)&Up?#nD#e7W&sDlEOuCjpvZZw&302G@SIJ+C*WcNXUfA&$f;?`!!u@G zy~|<;&wg6&+>9+=99rpN>MeN31pj!?KGx_t&)5NBHbG2%{tTTs@HFS=;ZuM)Yn-J~ zCwBgGgE{^A2sw3G)2}<+gZ(i;(|ftB9`2E=Tkxn^AVQpa&o~E!k;6Hx*Mesr1Hl~P z)PjcB8qUDIdS0B%f{$5)h3D)nVt`-EoQV)8MPL!LH+7NLR0CRi2h>Iy`F?&3;#(tM^=*qpf zVJ|&r6>zk~W&P9(?`!BMmotY5(7$8E9vaVT0a|R;9eu%bW6gqh@G*foa_YfHgv;tF zyq?c{_VcsGUhOnw|wb`-8C-S$`hH31a5FLtX61<=`xN^u*rhL7NdX`NSR7iV(*DdHV6_p-+Gq zaY7zv&^lviS@;6>f!Dj!dyf4Mu+I0=0};T_W`2=5>M-Xn_NM^*s7d|0zCX|RMDP}1 z5Bmf349Ei?`#52b1rLOC)01~V1TfCk5L=%|F!nHyfH0Tqg$`rZ*r(43DSd)GwMbyB zS%7Cyq6VOTFqYzufFx|-jI?iQYdvKNnxdrrnvDv)#kn=U}q3L;T zwws!IFX!js)$GypP@5VFxEP2KtHG%Hr_T4oGp24tU~dA?KAtI9cm~X4m(_DK$Z0`4u}m_Zd}xpl z?^B358xZ^62p0CRKXTdk74p=h>787rt(*mk9iiW$raZjfnI2l!f=8Ud2MwJe-yk~+ z?;CXT5GUxT>6!4%>8WE~2GwUBTnvu->Jc)wfb$;AQFDHVcl4%bLY`X80IG>R?4TAk zHTUOK4s@=@{-7Cp;Uf#chxav=Cj;IAL0jLw9#$M);$GhNT@_?nMmi5mHK4`P1b?+JHHHb1D#A+7(erL#Y z-q#|B9>nwsWA1|v&%{~Q)PpByeXg;~o&kH5N6w#fBIbz6JCBHlm|VCMb;;4J&Gzsd zoDE12C#W&kbKv;|Bd;8FV5&!tvqmn<`FV>O1I|VnPfx(C&Hmmhb$-rsG57>vJpxj~ zAqJ>LFt-5fdJgPmKljJ*jA_oQ)1FZ0oX)_LqgD(=gv~U9XD@lZ3$#<_7XQ!#{48;2 zA$$U!#+tK?U@ptUv$EFfgAeD{XNFcidwm2n3q6N%fPK`Y7Gw3;sphh?X&>T*=kXc^ zh&h)4F~Hr-sm)yuF}Wys#sIn0Wj(olUQ^GCT3u*!&e?3A_93Rl?kLY5y%+YeNBa>| zpGHq!=Y7s+O60WB&ohNS#?-ZdzAr(Syf^LjdqS*ptlLa0h1b}@VP7r>@4!);=inJx z=bgj-WAqR-y&uk?TG$!Rsl67o%d|M3?=^G=HTF9Alzq%O4^LyR9KFlHb4AF*&cG+& zBA0=j8g^+vbAY?lGq*=I$a5wo?-9du2YKb-w9jT-sFRDqvDdjwqX2tTaNNP~XX+8N zhq0c6KEw_&0j*peA@>GeYseFT?}d7cIyGU|A|@PSfOXALC+7f~_F)g_SkIaBGvonk zXb*ae{_Y)F&$Esi!QKR*sdpNpknK}Xjyb%A59f{%J90oclWMYu+VG*Db?S3QchfTm z&z^{I04%XtpQA@TfP07-JU@>b`+AsFZT66dk9=Plc;$2@R}^Ewec=Ge!c zumiMO(dVOe3k{b&V?Jl?M@}x7wa{~!CSwP%j~w#|2)f}eoM(-FtS3O^G7izvuOo1r zWxw7Ho_#c(%|h)7-xpHrb==Xt)VBb&Xd%|;;Ev|3=^i|v$Rc*|fCEyC7?!&1yzC2{ z&e*Jv@11@ok*m3*pG}8af}9$W2Q|WHja*D{KTMyS)lkpFGbJw53wKG8bB;XgmL3+n zjH7m9GpH_NT5R#+;apY+lRX>m2N&vyvU7n;t?#K1k6h0Ijz)cJv%L}ahX0)l`(9V*!onGHIu@h@V5}t$9=-I=dU%dIILAIc4|yQ6 znYV~_4qp4fv!=a_VeBDvhXnykY?jCH!N^5`o|mD+3wAqr&PMR$3Oz?Y1@vt2oK5Hn z>-9N@SUI>n%<6MM9D`?n45Wxz)7^;q^`bFqmN`uOoS(7(fo8vTjm7v1a-}QP4B|pF!B)rqbC3T(Q5#C2dI`k8teVg zo4X%(p-!+CyKIJY)aCP`k8O4iWNLa=m+eskeSu@1*dou&pCy7@QhCz#jUd+B~ae&1Li9 zd7(3Rz7SK}0XlEtG3)5b`Ll&{>U$Vv7Ndv8`P63n=xLnQ_u8K$oJY@LQLmr%Tl8r! z+Q~WOP{o|@#8cj)5`^16PX&No0ATmtAl;hFOpsWxXEfL&@9uYhy#FzY-lGnmCc z0JRc$o|UG%>_2_tPyn=--e$3$llzbp+@S_v^c*pA?IBP2vnKZFZgROEXj;^mvnQto zO8bI$zUF&W7dkqN9$MnE&sPM`-TJKX%u>eWu#3hTEe|!uMDBaFMFDvN@JwM3ImSFA zcRHW5M?r@;cjgeo)A;^U(*mmLW0aIvy->rR*vG6F)pa(TR~=~T9PSm9=lxvPA0f{H zTmtCX;5oxS=GJ;16l2bWzA&%n^*nnDvDdmtjaqR4cs^5|1IONQChkjspskubH^7;E zAmxnji(sLbvyQM{6P{c!-C;wPX>xWZ&*CyLw}=Uzk2488-}i7vo|iTDXg^|lhn$u= zyx5G>KE(71U>Q4rKA1(&0C0W>b02eOfCZ=XsHcdv2Qe+2Eq0n_=doEoIjx7>Yw8)m zg&K3!_x-vT_u`%ei1?gn)_)y1gaA$FECP4*XIOH4E(+|_Uet3Owb4UuLeBw;8th?> zAfM2KQ+N*5Y%c@nV_T$;bKqH{r4BWM+AwPoqetrwKA819ymcTwV&t6VTuRQHv)}`~ z9_RSAtnKnlbFIey0ZS>xYJ$H*NJ0pac%a|bL&U*b>($R_}Oi3R@s zVTxSqh$ENnVaPc(dYkPE&sKSGe7_xQ?x==7JJtdiGIsy`O!8+}{hGn^;yJ06qDOb& z{CoxqaGbTM>vKWOpLY}THk)B=nbTXZ1MdLV=*cJEhi9c{jj|fGd4>YN2bn%!^*l@c z4Ajy&dO&x>M?OcSyV%2hJQGdN;V=WRmmsHo@Ojv)JRHxm7`X)>5jNwPk7pl@ajuR$ zK^}W(f7s9afR9*%uEN5jPL6wH3FJIW0?%H? z5#kgvJx$MS8Tg&p9V6CW&<-)5OE?o}Ex>wyxG(a+t1f0Da2j)#IY6(v&i{Xu!ZT0c zweEACNjcIMFW^&n#tFc)MD!l4TYwhoy2mP&-PfO$x-mf08H>OHFwRAee~+RK?PqI z_1O7+=-2mY9vZ;W^y|rHJ*-)m&GL-AM*`$EKn$ZMVJ#AuX(jN?V?fQ}6`&4hSz`~4 z9Q(rASR<#;joxPcykCl(cXQ+c_R_-Lb=GEi&`BH`2%GTGtNz0jobp%vCivS z`{Z~Tnx|i(Q_}2G5a{{;1hs3_feOgd?ET%_&j)OurEd} zhF7zAby*MV`M!sL{xPN&XB=Pytk*z|vx(2?Q{?R9ENetvABS^A7W2GU0CX%l>cQw^ z&H&E&oE+k!TH0^XM{`-9r3Yx#4=)L`1@-~skfi+>I> zcR&Or79NP{QDcwR$WaHLJz*#7&({ti}tg(l)K_gcK$Fl|9IOHyy(Psb+JwG_kQI9^D_JHGa z%hl{ZKf@8~>|>pA&V064FrOr7!JS)+%>Go&u-=PsQ?A7`k`J4f)W={XSVxiQDwB2Q4Gr*Td>m(|%5 z`cl+l@HFLYwu5?+^8jao90xjRP&?ew@?6M?)WO5pr#cp%fLV(d$H2$%j1!mTIs=OP z*sJG4EuYWQpPjw@jOq+yfTlA%2V<*FzZWEk9iVpr?Hsf{s1b}a>`@(;>8J2PE1V?; z$9w0@B668#K{zLJ?LjSQ7S*Mu1tLIq`1?|aTmsPZp89MZVt~EMGq;Fg`dl0W!e(CI z7v?mJe+ZY&@Em-eClvkBtFd-7pEW*mT7_VMqMImdqXrU3gRdJo2yF?a>z z`jJ|WhfcUR^>P~MbL^)kjC&%0y@@NrGo~hLoR3`Q8OMlIAZ83ifOYUN)}71py1WZe z)DqOlb4JhZcd#eifid;M`6I-^IFndp4uEH_y@>hwpoMz4qw1iK^~8By>?O}$OZb}B zk%zNTo5r5V`B}uOiFpSu$0Mi3&i7kpUdv{jo(ZuUyxzxUzb9IF?uh}_^`3l$n9zBb z#dy_o=nU^O4=tz_;4Yj`fQYae$f^0evc`Kjo6RQh+Kc*Yz9(o}UyI?bLm!X;sYOgf zf9jw^;2i7=zf~{e4vChv^3eVU9 z`iyLpn@JF}Zh?s2!UJJ{0xm{Q9lbAN_OY%rNU6bdvX7v#9y7tBXS-VxkDUu z$nzZZF~AwtltWBDfIQ#30&3a}zSyBex0|t7mzc= zL6%3JecXswza0x)qok2^V2kP8Mjy(~e z_kd?k=sDq2muavk1!&f0{d~@mJKD=0y*IoXyzax<`8|uqo#QzNi*2^cxlE6`1dTr2 zk)AyXKyxZy9AV z>p?Tzkv<`h^Dx$PAm$m>99}JI^f2}QEUM=+j`PfeS@c-s`n3zv+Du>ffLG6*+=YBR z`su^DLO(TmmKeUk?{$c&&wkd+AWq?93m}dN+DHvmCI7for)F#MB@C17uJ-~C;^?gAz zg0aqhiTB}bs?9xsMV&pIr>9;-*v#k7f;`vnP(yCvQ-Cvhem;Vc(|I`7_*v3EoIgPx zbf|4D>YmeG$dv<44ZSl%AFJmsz`B3#0-HTg>Y|p|Y(AyOU3E9l;Q1IDG5S&<)FQ;H zZQ)_g`rYuVgZZK}&K>>tRy-?pQh+^tUl>O=`wZ#5nR_tT`=L(bF7^`S^LRe=lnFX4 zp+@9p$+>K=>LQO(4;l$->}QRAyc;>@+(DgOjGU%tqDL*nJPS38LT5oGo=?!DJHVw5 zFEHJ~Jxkt#!er+@_mCd5{CVN^ z{US%We?Ak-=KwB7jd~ix^SQB?yU5Wr=DA?(wJ!7Q;||_|TmtAWI-7CKoy!3F=sxIi zMxPxt`M#t!%c&QkMx#!+ug1u!$#bp-*c-cyqn1NW`w-KZhx?_D1ON+r>5Kk-K6CaY z_NY#R9n5K|Rj2Qdb6JmP^q$;N^ch);9`*qR(20EPtk<8HabYI$b@oLTfTsp6oSAo4 z&Z6x6KEy$TbBWKRBF1?~-rq5HC?XGL5fkKMf_?C*%VI|kF?SOBT+u_0z1C)2f}ED? zalXf9eXQ&3Jl{hNAo9-#kv^{oecVr<0x@FU0k8cLjL*VlpFP$R0(lJZIpu|Cg*>w0 zk+U9lv7cv&$RVbMduhK#&a=5p+ahOdfm|PA_9};%J)BV>W`C}iHSA+A&&RnGvCbk^ zebCzPdCqH{=Y(m^`mbkbjQ4jUpMdAx^vsA;<_V~?|0Er!=~IC#eNKmtU7^&nT&SZ9s*VlC{|9_H5TV;9eu zf}?Rpd%@{5K^|EEKJJcZ1X?lK(00_?unewXXm}ULio26MVV%5eu zoy`jDv5Z}oTX^b_PvEtdm2#g`eil-`$EefTPYYV~gqlMPaE9Ii)C_0;6+EAfYLE}J zh$D}L-r*m5>iDyTyM&pLb58eThI8Z-?{zO?OW3T&9@eQ%j`Lb~M`sgoK~sCs#yQpV zF&&Mw)S$7Kc>*{&phmk!G}5aggV$vi-80D4%J1@c|L1r zy%u8Xr^pk)HUJMFF?aAlm}Opwe(QNubr&`B!Rz@#4W9k_zS<}=dYAR3@Z6^}ycc5X zhMfCp+)ZBpe#wF$xC=acsgXEv@HECL5CN>QC#6SDi*2UIx&?yiUgXY#fv=0_U=eH1 z7#yEX;K=g~5q!cLJbM!K@XpkuP62XS3?Dl`Lm$osjXd_)j3ZB-)aQ(8st3?^wq^>LiE>3s5JbhcS;m9DD*4HNG$5%)EDGkz97FklJka zAMZ2IiJIPtx&SS4>azERF~A)TV2yqGnC00B-Zw!l%(E{AA`k7u40|jP=2%Z%q(;sm z*Za{s(10bN&$(dabQe8g{h8re<9W!@*q1mBwZ{GkpkC;4c!we2=sSE^+ydPbhIk?YLUBpcwuO z5Ze zzBp?6|KC1)zk=!2xT2|7p{nV4d^OXoWMy;iwsNMusc5$4mp6O2l{Y7RSk_d%;S_V< zeI?DSFP1VrTa_?1qZ7=%^-na7elB5N8dK8T_;xANXL~8r?A4Ra2?I--*TUjiW=r!k%qyd+m{!#*o6ijdPdCMD z<(YvKs+hTbs+jW*R54%esbcz_R@Gd)qKesByqfv_NO{xjk&0%);7VrGv8S33ww5)o zFDP#=xvYYDxO_jeV($Z{;d2j~CJSOyb@>o;OrsHI`1^xRym_d3@S=87NcWvtJu(#$Pnt*k&WlEnA0~s<#a> zqsk06N8WtUbRILz6l@u0hVK||4xc#G^xQtotbBZ=DRKEI^Y`&%%wOw9nWrCm*qrzA z!{(R#ht1Zlqs{Emqs{vEBg|V@j4)Tu9%*JgH_YVikIn5j4>3O;iOrYOhMHHW3^Q}C zA8Pg|gUzDuvFS6juX*sD`^>>hVl(jV*sObHm^nOqn5ppP2y@1rqs^u}$CzV27;btt z9B%U34>6s78e-r)9%?>p_K?{&YlxY>VX&$2{1DUfk%!Fkc@G-%V{ATJ{*d|a zYI@B^Ae;>&|@C zJau@iDN*4O^VE^H(&hf3#6yKXKZI|ZD6KA)UA9}Wt8hIV%n57-$ z*gspyX)Bw`i33~66Hm60s$Uk!5BGGEK3h6S!C9TXcUoT=bj<+yyz;%$y4*c-&vgT&Nq%=ZYkNQGy!Bp5|0t9VFL#p< ziglEoi#p2-1qCvG@ttyTXJyUTyZb(c2PdQ0se`^o)Rb(DHZC;6a9SE)C$rwsmIdzt-Ofi#WU$~m^PEc>yG z41A`wG`;d}**Ud~>^^w64Bp;BPVCc8cJAsTCx6gI{)Kb>dQ*Qnc~U3YU96LwU#pEg z)Uu7deL@>)d$>S49ceFD?e8jO#`coeFK;g|Jl;ocKe?~0zve!9@Ws0&`Q#osP`gkT z?zvY?zkX7oYaf|3XrMGGai3hbzqeeL*GD#O93=0yx=(5i?k(--_LV=UJ|F{^mXpgS zmXQs&pCX+;DJg54m6007%gP5UPm!&2PLHD@f%p%E|Jf6=m>?Wo6C9 zm1NcQ@-p+1veKjP>GIv&ax$WHW%=&+vT_Z$%a5xd`PHk+k)BoM(!bA=Pg>QGZXZ{b zlZRH4Yf4p+razq~dz)60hrg&QAKiMo{A|vUcIW3w*BWO_turgi@N=ule^#6+H&;DP z9;#JUYP?-p9#~jG4%Mh251nzk)IYwmy!TxNIdOfSyc<=MGW#ma_*<&VS>3D1@>r%L&*6{KM$JflA=%jfl~$<^;wk*mKeD|`2pmSk!vnS7wM-1$Ix>Daf5?0&Pn zG=8_594M?JoBymNN48dylWNzH4{xp^^{3R5Ll4!In!o-_hP_%-&hL>YTbtI9;myyK zN!@D7r1EFT!Gl%g;DK7wvd!tTt=O3|;fyn7-nnPUjg8I~S#plt-}WN8XxI5td14)r zO10&am+Q!r$JUXuFV&PkmeiD=Yvjo#H|0s^eHG=e5_g(K6qKJo6XDL-fnJ)Tbh%%+-}zNXkp%vrlxI`+sx|Wx0$ntH8qv% zG%@4b=bI^4-(v1~yos5?afyu+ncl3bul-8+sXW^V`o!$ zU@v3-=xpABUdQ*^ne$3^GW8F%H}xkMnE4lWG?$;+-hB9GE7SMtcIMGnI-32@v@v@+ zw>33abuefDudS)O@J{ncyEdj_lRM1nX)Vp5YAwxGd9BT_{cktB(iWyiy?nE^%`Ikf z{U+u=Gn<&QyIYwJ|IIhi|C*WoeeW=3_T6QE{j7!QQ+ug=-hZ(yK4+cuX}whT{bz+- zP<@uix*43ZJ)|HA1#zS{+cV(CM=bn{T9eIzkDLS`YwxqH2=d1Q;+`Sk|rclA!Glcv(-q*S*2wMB;h zXOsMA`g(a~+*%oR@=Ce>l*Q8W?bWj5mW|SR%{poM#2Pv8gr)NPS*zu%x0lP%WgF$Q zUK?ak)HdnUd7T_{@ml$1@do+jwUu%~@6}Rb;!0Vve5=%Hwo$$*wO0CU+9GG~ z-z1le+A1GT+a~MAu94%KEs-(DEt4@PE|D`Aej+_4ERsW?ER@P`ERpILeJXeTXQ}Lb zf2n-mbeX(x{4#m2*E0G3rsZk z*UIqCTjcsnSIXj_SIG44>*eOj>*S3ND`ZQP71FiED!K9NWpe9>E2RH3pGoZr8zt|H z71D3@R(b!=4brgd2C4YmMj19?yR3g;ldQ8iYmZGb=-suF_xxI!dEP2n+;oi;U%yhu z4__gVmR})vU%pyC{r7TdvT3>Ozo3n5JhQF5k3X+^RlZZEHg6#p?`tb9U%E@$z1T+H zEZs(mn*w?1=MIuTp_Qz;<8FB?zlHp~wwb&W=S#mcub0c0+$iHmTrZt!+#;84Zz3nu zxm~)>Y9V+2)l#; zm(2g7o3!5B6VIiKw5fWxjDM@U++MMZ6z|$qo>|{oy1md+j!j$3fG1kYfybK5O*QV2 zF=NG!lnE&>Y#~uDGFWpzJuh2)PozhqG{_ZcOE*vOT zuDDn3>f2u$U(r)OJHMMuJgVnPd9@(wXd`M`1Ub*wy!9ZaliMH63co^v&Z|%R|7gok7qha=>hE}ziBHuyJTCb z61S6j^Iw+AjbD*c-4c1>`R8TUxc|vVC;eBBfAbk>e9y}=cHaNwmbnw;(wZ;J3pFRm z&-cA9f7F^J8@EiBOG~^bwSJu>uPvP{(^|eFcZ_>Q`fi#kGy5lUP4CI_>rYeUkB6to zvs))f)o-VYdE*Vax^RYU?e?A=DmG09bpA-nd_Gey?>62`T{>HA?T_T6 zE;FTG-w&irznf(-w4qI7zEl60;#S%yr0T}m#RA|pCXkqbIbmJioVk$a0zmD|_9 zB%=!_$fUnsmZsmnER#AvFB=|zT0Wlqyd3!cIceGOS^2xsOVae27vzdZC&<8KC&`pS z6XdhCuS%)u6Q##(Q{>21ljO!tugKinCd%l`rpnmnljO91ugLzgugZYF6J_cDUY3t* zPLU_ie^vUPK1mKwdsF^=cB)LS^qSmt%v9N1`W+eY_H-FKa;}VMJV%z*{Yc)rXQsUR z{#04t_HB9Nz#DS;pYO;6ZQhXYe|tl=T{T?_&U;4=c9tw)+a%8xN) zzZqsuTZ-RR7LGK_HVijaP99>KJP@0&>J2g%z4d_UdUbzO_LK+BoRea6r0acV(2@RT z27ageUz0)R+WxU=^6(II+i4?A-*ZNr?>3AyKb9P2#;qD{PMS8-oYiHxx&FMNruH}e z%sXHAGrgDhHMhkBOkssVX2{|G=1A26=FMvNn|)_KWPUtmh&j1!Z01jmO{Wv_``dZL zP1Q!RS#-@%bN?myy|mmYGpON6^Y5>Ro9}v$GG^mQ^XX3`%(R_j%!@NenK3Iyn=79j zYw9f=Yib@CXI_}~sCoU%v8L&^apuNPA2Y|Tc*I=!`&e`KgAbeP&yO{io%DnmeZymB z@8qXU+aI4W4=sMe*gM9X=}$gxR*e~NPObWwdGLgB=I>WVo6UcYGCS{r{*+OsT!#^6 zW}{K&^)92#ko?hRW#dOouQ_8(rPoK9k$;ae1D1|4<*pxNW<5RH^n7)kncHTZsk!Ab zQ>psn=9?eKnZutxXkl{4n=mS6X+m0bmE3{8F+0*VKWKbO~@+bs(}+%4ysuVvzhb@I_A__^Zy zO|p2#4tc22a(Sxu4k=u@N1pGnMIOF-o0L3hqrCX&m-1!HCDQwXWm4yg)v|y57P;cH zz0!F17t;3rtx~XUudHvsUN(QUSe*&$1peIpmQ`9dl^`h(Pu_RHV%zLLjo+9^$c`x5Wj zcG>7Q$b=`iOO1NVrP=+PrB(lR(yVa36nkTp)R?+hYT##-QA<|I<;$1Ij%lAto!v|2 zVDSa==CFm*K3X8Zwf{tp;OCm}zgZ&Fu3I5H=PZ#%s~5@Z^Vdl2uNF!5gNx+N`S_V9 zT`9HRUn5gyERn-EE|PCoEt3)HVkx*{sjR*kKlkigAuUR;l~RK?%dh3P$@gZP{JUE! z_g(OXTyLo8*_>>!r)bpUW|Lp6?9U zDP_8+va9!IdAQbg>5^`eK0j`jK5Mo}og25xQ%APQ#*eqkzdLM|5&661;18+XIC#Ij zZx6`M)Aq>|=kJm$Mr@He12)M0#Wu?M71ztc^&6$l=G7v0ogBYxjXZMp61ngEHB$GH zR0xK)yrSVsb_sH`&ND@Gq&!Ng-!QM`8GRcfZZjt zw(XNoPWnc+9Nr}t?${;CfmAxy+bM^S+a{xXZI?c>UGDsNn-uT%m0a6Cl}kR|Er%EG zlDn?lE-&@oDQ!pZkaqL;OS{XrN;SJvDoxlSZI5h|L-+5HXOG(=r*GRR@0`6-PFlA_ z%I{n*WgqM=l~3sg8>6%S>pjB^Lxnf6FbXK6MM^pJ^RRrX&vN;r#r~; z7vCe>j_D>hztBO-4{Rg77q^z0)q6?FwfD&DZx_nPL;A^0)$f&;pS?$pd$Lge+|pD2 zy!c+(e{y$ee)B;20(*|D-BlK~>MH-uyIZ!-#Lr60?~~u(yI(#!_dfZ4Lw~t?T0bdW z<{mkIZ3n5?zPr3L=03Uf^}cfEt|2mH&R}WN@*(-_ouN{;!+mo6P4~<5yY7`53m=ja zUGI~9Z3?B&bA6=W(1CK|(n0d_FNLzRF@6?m*-y^e6HEV1_e+xjeWY?yC^M@R%Hq5p zvgOtOvhMhY{-jV=e9~Kr@9872E$%1R z9_S~3j2;`V=MNNs*m=NRhRXWhCdFJ!eNKylySexJL|ue z6U{zZaP0vZyLpdHt-e=2UA04goViA>dvv3WI}6{}I({x64o>C5e{YxFyLQXf(^HwX zYp0Yye!o=fvS03b&$q?*-#0rt*IIpQQi$-^h;(_sHKB_Q-&l+vW8d z_}+KX0a<$bPPyfr&GN>F+hp*y8|14JpUbqLzLdrfd?DNRe=EgK`&!O^;~Q!5`yRPz z)IM2UZNEHu-&c~~;B(pWOe(vl?U&{y_Q)+O_sgRn@0SzT?2-4**eR=z-yxeX+$~R3 z+%Bb7?2>aQZ;^}7NaeS_2c+=e*RrbGE}2?!uT-tSQ>xDYPOkav3;Fo8-O@h)J30K# z7t*=yH`1i*4>IP)1G4?%ujTm5zm(3Ux5?AD?3PB`Hpz3xY>=qSF4@rNb9w5`4Kn5Q zJ@W8byQN>T&!u-AeBT)Sl`N~ZM`||RAsdhEkw%j+*JYZSoM_@+qGAIyYyQb zHS$MkTmMUW@bPcu~8g7yNx#CcW{8+&cF!`D6Dl(&GESrDLhz zq;J9Ba!cVsaYqiy{Knr&uWP=Qg4chLfroyS3m2bgp1b*2^MQ%Xcb$uyimQv6>Q5Cj zeO5$fe!pYP{9(tK;e-B?y%Ube+RqNl^0vQ=+5MxOwDT+c4Ew!QYW%xYIP*_gaNOTA z`Q~Fx*>NRIozW+n##_pm!Pl2I*G(^PzL-+peEEGFb<-f`xldyd7w2RR~d%{(Hd z{`IHyeeX9}-0f$%t@t51=GDXU%iV{iO{v3j(`kQ7iOzq?@pTSK^Wg`i!!JL}@!$L? zz0dnm`i}s2)*(4*=3y!K&@b}n2|vh9hxW-&8xG2gjlY+cPyZ~3>whobPXAu^?*3NJ z>w8#EXm?1S|N3v)+Tss6a?UZPdCfni!}-NcgVX<(yeo>E|K3pCbb9$OIpLGy=A>=M zm||6nn|hD_DWBK=L#n;^hYTwDtNirKv1ZUyhh*KyhsE{$UCwLztsGkYm%Q8JkQ_Sq zuoS-Vmz+E4cr$NtG4trIBl7d|VrKD=#mp7uBlE}ae~X2_{q0!uR=4BLoAXaHSM4cj zYE3O^X7xJ74ER?`^VE%{&G~&wo4?zYH0%F&im7?-iKco%2~&6Fi6*Ucy!q??6U@AM zCz$-QCz#3amo&FMT+*z1?F936os-SWPn9vxA9!2VR(V@$zwnlve(-JS@#*_==-FBF z?|tvd^!@M1=(Qh9nZM`BSruo?y?v&O={ZA|&v7zw^@no9-5<(j8$ZDBmmf;Ub7siN zt!B%IeP_v`vY*Jt*JjG{?K9-KgCEPyUwt4KEq_;r_n9Sk^qD8^w$70{J>HX^ZJkvA zbiUm7+Nr!roH0jMPMaZ1&wN`}wwo@0zxR>6ID3}# zZ8%%boHk$1dhrwasLDd=*KDr*F=)0NyI_`lJ${CK5zmw-Pk&pkp8USlx$GmkvBgXo z^w~$!X#X5(chwxZ=%(3n#_YLr!ZIh{4_PXEzMd)dcYP#}pY@?MSTS2RUp7mAJ~TtB zTrflCo<3Wyobsu>zGj{*s4zzk)?Ora%FLHDW_>J=)qF>89s0gBcyq3F>GQD+Xg*(> zeK=E2{&2SRU*P25YyTeu=N(sb8-{Uv@1(+OhRCQWuDha?Bt=swLiT8pq`f=saq2Wq zd+$9ov{l+gwq%8py+YpS51)K=ob&vC=bY!h@9X+rPZicPeQk$cEskWh;KPY({FSdo z)b!ifaG)KQM?0apzX^9MYH^^f4u_LEabKeeTaQ-3_)P_}8|$H2c^~5jx>57G9iKyM z@%Td%9L2g|{j>q^beiEN+k+aW`+d=Cgt2J>9102$rdfd2>p56!oQtLF3Q-i1hsVuD z_^`DUzKLuN=c>S2QiAoSl zf}wdlX8(wX%70;)P@e>mPZ9WdB@MIWQn1q~4~vpAk=Rp)=i>@-xv>O;35A&0nudS3 zb1=}Hi^{K+AIgPHbShR&D?w>WnBP$NeHAR)xq?xBZS^XbE(3HM}Krf=~NnzE*OMUnQ=JH(c+O= zbvPC9qRDSw6iuHSOVgJ|lHSB99$8$B zq&=twyA_k@w}&k)*O z8AiIb!BpqXrN=#iH11v?Engi(jz2;uJ0y}?w4>=*GLI(Zg;CyuFj{#xlyaRSNJ1f! z)D>gMJ}Z&Nk4q*`fq0rc8bc5DV(9Skc>32HMKg3`seM-@=~~9nl8cEHr=CJr)}~PR zuT0vKmqr>{8Ps(vofask(wNKy5;Ww~{fT^XI1obM1n$BcIlaWpc#d&c_b44)Y6L@q` z-nZx83-MszmL&Pqeps|<`knTZ~i6nI>U$Fzzx=uFSVH`g>=O=Ftb_yjEACZh%E z*gG>3I?1undlZL<+fwjTHyWen(OCB?9xtVmaauMCHpjy;av=>=N>dHc^P66u9q~JZ< zr~F=I;HN+)a^*Ad?pHQ;@pF-NJP*77q@kVZkaCJ_ZOs|bI+u+bhf^^5JJau+voJ}& z2$5gQar$owtZtN{^GXqHxkZpVn1krIIk?uHg^OEr(6TENe`<5EIx_|0r;@OzEDfvq zDY);MjeDx8ICefB-tO@zZ%RbigGAJf&BPt1Uml8#!T3KZn7bzp+Z_{OW}1P!OtWl! zn2Off3D~nB6>&2%A;QhT7okL0xi(^+d^7XFSuLpU!1Xn^u`JX?Z)!m-Zl< z&5|7PM+juIXsY>Z41Ik7jjnFktbU5S{0C^?bPtuxcgeiljeyl%7%;kvju#!!ecz1S zGwo3Q*9(U)_ptQR9UL>hj|YOxXG!hF>%M#NAJ>V#Pt5;0-HDJ7oltpn2cy{?h!cK_ z3uB%#?dSoJNmJIaxc=ud$G^`KG;r+lP7v$VbBFp^~Z3oXIcx>&dzLj zidTIv5SG!4+f|QorROQUqx-QcYl!ulFQN7GJ-AbzVb;q>7=Pk5_MUl;Bh46>JzryX{{Wtcy@0^g=a}|t5R;6aV*B(y=ryqS$-4^~b*#SB)W9{ioP9P*q2*hR;er}G^;b2Z$$nlY8%jFX&OxF*tq zZ?BqhXG#;Y8XBN}y%DQk)nUw+I!s(y3n{akSXo<-1NYkS=0O8&ch$quw1(;Ox1cI; z3-hJxAS-zbx)W}p+PwjW6YJ5(w5Q&0t=PZ#7Mp+SvCO6hH<~Lj?>FnaoNH0Op#~Yz zt=J(_hnU7XtXSB9r10Ai{dpUT({3YrM+3O-tuWMYf^r7)&{no%Xm&I1eeXc?U>h`6 zH3E(;_-@jL{a#&I@vR+`@3x~cvIY4rx7a-{W1i$q3{0rUv4b~}ey;&qBQ@}Q_Xxi8 z+o6%(ir<#qI7e-Wy4#BHHy&Wop1Tme*@J=Yy-3~H2!*;veDCc>i&+o0Mz!LMaX0cN zHsfMfEv_wVhHG^>+z!;?CF=|N4z)p*eZQuy?11&@IwbbrMnXs@mTvEY=ABNYFx|EB zbT_Ut{jtKl3IFtK5$aM6>wgWn-&u!UF7@D!y@kKTLHt(_>|Dly@HBtsJ$WF^l7qNL zS3Fw7K|oO;T;IDRZ6K6=*Fsod9tdG+UtCV~f#_T(K+_ptUb-{g*9%JR?%4d#3*WWf zv9#L@{53APZ0?Rs0Ui$2y2Gfz6O(>QCtPf` zh36S(m}fadtj!fCvfNRY>5MCmJ~%tc4b5x(kigc^DC&wo91nPW_kzw)5ESeDu|6JUNgW_Q&@UKhy?CVA*;fSY-Jj zFgpNt>0WRdbY&iA0RGl`!+pFD{w)e(no}T7wD_Qi%fqi+FAV?lfeNc39w#|Co#caW zlAaiU#uFA(e1WP^c5S}cKI)07`+U%4>&N5I^j24nWm3Z=A~u1il61=8jMlPY8q1=`f7!4~L$TKUNBc z!?lVF_oJb>)*6ZN^;~?|77Te-({HB+qc_nTN9=R zA2D?SMaa%3`7d+n@99}IbJJ{^^?44P-GJoI~F%=F;`@IduK+ z6nc7oHl4c5uEl>gsZL`Itd0n+^%f?HF^fqmN|3yB7n7stBGR;6Mv{M5(xsa#>DiDd zwP>!SnSHBiG-?ff`zuO450=r`%$4N!W+jyzTua3^Yv^t08tO_|M)SvuQuz*1O5m)b zH<_aJ?#gNkuw701N^5An$vSesuz}=cr0Dg(b)>Abnu@Njp@Xl)Xh!22`m{}y#zctH z?s4_7PHDtZ#&&e+RN$dhCAfJvaf?@jMy)37H|fBUzT0SNYR7W5cEtSYf?Mq)EMC}y z!0hLZXi=}mF5OxzmMX=+xGHSfb`#rVs|jZgS+7KrVVYD z^{`ZHVfs}g+y@$vd4|;-eii)KTz2Fs^L0e4(b8}m5tS|2u%;E_&Gk6hRFBmcn{j

+3PH2eYd-+=#al>T|w zYld-Zi@SLyg8Tq?umcHZ{)tv=6R;Xw@39w>sKE1>1|)< z>(^Og&HT6h?f|K?$NK$8pcQr1QeT1}x!?P(-D6-6V(w}`@2ueuV*dT|18to-_EXMI z=JZ@ReT-+oxidPz1j1pR{pCLTZ+lvUt<;?}$Ne9` zTJsyI=Yt-=8G%kvd|Q0SpI}he(E0(&rf&<(jgHn2pzRLltV2K74pSSP-3Pq=J#)uS zVzmR>Ip$icPW0TiIr;;5HqY2jEU^@@aDbYM^ekV694GN z3~T-DT*>(ctQRLgh?`@b+B6_QI(Fv|_4P8Jz@B6>+g7jfUWM*Cw&QfI7s%sPv)H(*c2J|CYGdm!gC@*Dn$ z^?TjJ_gwc=JN*tz{Pdc{d>((;f5*}|ym{DO6ksHL$xiYZH zvByNehz;}%_5u%lW6D5}SiE)alSu%Pcc#62ZXVlSLe883?%|!tP5kS<)jo-xaOO`C z+R-XzUu)%*I!l&Wyghp0gPtq5?n7Q5G;nP7={<}gA3bB`$=#^WT>Df%Z0A^GzMOZ| z$W!F>J)gTM_BgX5cQ5jcPjLE7YC7@SgK*QPzcmB2-7^9k|9qd~t$U62ez{A{gFn5J zSO-VspY)qat625cInP?PGrU**#O~lmjiNt8W5>wv`oKJ67it50$T3gf_^El*2OTbf zGpCwdMLn|2sx_AQLCuuMH>dKn;R4v_^a%LMlf#xzImGu>sUES;YK}b;R2{y1PqcTT zuHL!q!1uYFd#U-(AgC2H@~X#u@fq7!3#86VUG<^g z$aUu*as0}O_0HJavky=^z_)j7HFz=W2@<<+&b)#BfUo;E#w;c1j*hFMhKhocYEg7dkq(-_gbzc+lV775SFX=og%Rg0JUFdG@M3%mvo8SkKe5-hlHa zsN7#j`C<=p^<(RkC-45HiKjmL^uVikc6l1o=Fhfx!#ZRF(+}pz+AsKtk)j9&ioeZ?-wV!?h8sz$>fK*=sF*>&b<+-8RiVo zfIR}#J3#M%bGPb^e|i|d+IP=4F9o$7yoh@@9j=q-U0(1{+*)gzadxssdBW?PYi)u7 zB0k?ybM13AhxzrK@Nc_j7|&H>u6gcb&-mVIrgPrF>1}=k&Tg@E?L~k68@c}ds`}&& z_yi?r&s36Y_C&j%eI9$3weHem)mF{6s~2{VBO3Mu)x$jyeftj3fjRac&KCGBm+sok zDv5o!=#u0ouy@AVX`=Pz-RlC>zIVN`1T2W`*gIu)aDMfap!Bhv;F`9lpo~JSUKj^z2lpEH6QN$JF%IJCT6Yk zst26sd;(`TeDeZy0jD;10(Ctr>*4LM_HNVzw15XKS1wldu->>m2e_#9K>&Sgwa0zw z6SH5!A93!Be0R3a-Yx0HUiyJ=&d>WgTfiB&{{S!esgDkHf-dkg7Q9PyjNQI3ZM`!Z zoG~S0?csel@;d&J$mRQ-+{bw%*16qN?(CfKiaLNZx;e}fsSjE#|3a^d;d>7+;LZcz zTK_y|FK1Z$z_&Iwr`DG@?*2~@+pgWHRc7)dKJU%g8+GadKf&2k9rz`=c4%np%}p?Y za|7+2IA78C&yeZ_N}WB>U?c?E8fSan4g%g9pR?NB((n@`zEb<&HGMo!d#CpI1)sSH z;vLMb>kROU&n0v9kdOGBW1ortgggu?v4#x z-nj7B%7!oZz)rNjdG;M@pqMX=dlq{OYt5}VcJXNa1RY$-v&KD#G0@)5G}#N8h& z_vEMtNO^GTTYke@%o-e>w9*&0B47gV^*Cb2ov+!n>X=0x}k+RAHY9z&GSy} zYn@_D$yhZb(bx?bo8U@KqPw8Cyzq|3=HxSpg8)q+-j{hTRxk67=hlI?Mt`6^r*S#$ z3EzR7dz5GsJI~NM^9E4#lem(({(YW?@3U<9asj;OH@S@tXM6&C4K--kN+jV}nCMQO z&%ir8KWB=)^Kmc<3KKR)6~?I5VoQt$v*x<5%jA{Ezo#SyS`T-osz?AF=N08@9g> zvG#?o=T7Wn#(i-O3HTSjV!nMnyS#N1f8YC60c@OGKV&}j&?F5@^*b}Jb{jN{+0I~HAZU*t+fjy9qXQh)X7O~>inWx|6 z)c3{N*L$rqqpekn%Q9>BviM1z_hJ6&k+FKpIV*woW9QXuIM45F>z(T^&NE*9uze?t zMXWP5;%Q7kN1Q%HQ-s@tF2a;vG~izO`TH>c8Dro;jYS z!D*Wp=n2j_)H7R2m)8bCoxb)J`|Or*@&P~O>1j=DZmPkPm@;zpPjuyw9|FHetN!Xg z$?xDnOl`gIR7-sB##&<~*zRqeyfcE_b5{KZuhcUObguvA`h3Tn^H1u`Y3L-+e%97q zB{Ams`}>J)-|A;R^7>z9x^oG-`rql9%dXlT4>{n>njL81?9|S6_H*aPM1!~Xjk-i1 z-*>U@Cg6Z|1MLnYmu;P&1LMX{t$fFy>J4K@?2-6MPj`(V7u5T&O)NlSn|tokyQ#Y( z){IeWpDwVQI@Ye%yIa-MW8e?#?MIHeDF?i>K%c#g>wnw4vkE9>#Oh)Gm7K&kKSAu6 zIhE%;AaDHaFL!gFUsf;c#fdlawf%3T!#whTxDWgwFVXVWv;(+_E&mVd4zHi^_S06h zl_$Q>w#9-w{RAWT{^WJ++KU~0Z?8MYozC+&K4R55x6X|Ijoc3Jd~y2J9`O3kR9@)A zXF%NX(N(u|{7gGX-p|uTJ~{W+9lJo^k{~zNy(HSc)~gl!m~X5B_mR_X^t>^{{J=lC z&YZ@9cI9s_cV3It%bdP3<$(FZ{`w8n{pC87{KSrpy0d-1XTSRXu>SYeE9oJu`L^pD zw)>xYcNIPpVQ>^_J_+-z{>pjh%(ccpQ#mh^U;uRJV^3^nTbrPI z5>ql){i$&u`*?5W*L|Yv-ryeK3hb$N&xY+YvPQ0xU$V8k%1sctay)bO#&6h3tnxfR z_8)N0=pgW|t7p@e4E&m9ANvFA?9+1j?ks-pUe-+F9SqcLChJz*zwJ+9f&tDKy0n;cSY@=ySFiHbJ_uB&(&Mr-RAadCTDa;z1b zhyQ%X&J~}=jln;($x;6W200D*S&;Yq_K<(Wz5+!*zzpJ?_hlaQ2fp)8Tt9bSU>_-) zJjL(p3oy@oV-x$FE695gx2N-cZxnOAyK~=(Z>^#>zZ17_p#KK-IU?4;1lG3u(&jtI z{2N-BW3I6P9Xs@ayfD^wzP|U+)$nESdujM5estxqgWAuqTB*Ha-NUJKH~9(gZr*cX z7m~fPoj0-1xyI%i@~luis_Z+ofjFK6uw zOzbYGSG+^t75NB=15D+O__+`Bv`6kc#Mds4yffF{1!4h8)B>5geqY; zCBq-&szC>#jf_~n+L{lc)*R7+@w0p6I?IX4|7Je?1wGMc@3!mI?S1VMswcVUd%!vNkiWrq^x47MK`h~{4{Uu!yXE4Y<(>n-dSLrma_*Evdxxj? z^z!{u%PqclejbM-mvX0|Ihbzr^TN>@aEdDYTyIk zGuXGqsx{u=^%K~m>|A*d;{SNxfA>od^9QjAu(kR69h_KV``pInGFLx+<($`hsC%XN zsck_*uAWbNln+`PCct18;U&-8k0t!rE zte)q6eV>_Qtf8G*IoK1JEANbf-LXB(#D9SE?zn#v>*S<)^f5NEFXV~3Pvg1X4gZ_U zmp#mfs~Y+={McM+dRa5%(I2_m&Y{O7)VwEZ#&l>?(_z%QzGD~(bv8b zGj9U@W-gMk_POCdwqyS~F2mqRkw-Hp&gl2TA-;2bYr#vO{sPxVJsS9Pg-;qxgKZ&Al?y`*?x}XmIvXXJG%%IJBa+w(Y#g z6(8^&SR|?#!`Q~bOxU#vAf5T6BMLhz;ntf(%bAvql-+*%y&RLUu<7#V# zcXaHw{f3yQ^jLP~B{u$rp3XGKojSUq^%Im2RXM@kGvjUNCAxCh=Vc9X>zp$;>P*-P zTCTGiJNIV2XmD@z_G}$nf9A6L(d?;2@U)VEhW~=^0+wgZKpI z+sAx45#SrVcW(T*`{J#=;2UqC{`+)%ZD**J8Ec=~NrIe_=neY@_5@Gt4(zKAa0SLZ zM*|)E+{t-P<#SKwbYOk;z?QSWxl@~%_URFD-kUKwZMl)F#~b;_w=UuBS3S*-buZRF z$hF>DYqW10c$J&6ey`wX_z7CBGbXwN#aicfwA>GDxpS}Xp?uZmi5Yu7c7O)zT<8#oqvU6^SxE)W=$mhzI>FH2y80(EFo5^&=j9JzZWBo4+QpnVVht_lGwZSDCw@t+-?Jon zUHr0Yof+U^O)mcQv6p%J&U7z*@4+5FK*gwS*n=EVHBcvJpHnY)?k@r!Uy-}g;~P*b zp5um{xz1&dv(2$Cz(pJBxqNMT=lw?Pbzk4v3Fo|_2B{6q>s#B)`2Y0|k6n8j+q^Mz z<=yM+-MF*9h@In{hOP8uCOO6iuxBCRKm+DfKHASmJx6BAJ$HZWCvkNb*v=f-fqkPc z(MQZ1xpEsZ`!;;@hxT`X_Dm@UzH-Lws~8K=6X#2f@vA<%2Vfsjl3T~8j_*5{oX4?^ zCqVlgrZ!@4&!mjx2dKM`wXgXDnCI*O%FGq7JpzIrz7NI|m|r#Mm;K=yw&)vm53%y< zUa<$Lvu^tvJv;ETQ145&5_=DAczf&n?j}0)H=43*&A(zBQ+n=SBe8vM*3OL)pFNVA zhVNO;sWYp8Uv}kEZ$D=}!3=CTWAfIRZ-09k3-m~OqU&tb`TJz=UoOx=+<*RFce3X- zxCgO|aq=3l|IkKiR+d@u87nX4w9(ofXis2mfQIdSc||>d=jyTE6?ZYO=D;0E>^g_q z^XBrcv$wj_r|Bn}KIZ6~pV(^epkZ5Mo_e7B_SNEWKD*lQcQjJ*#E4a^R@^NEP5$*e zM6R>TX)BeJi=XI^Gsfx_)OtT%pLq4gHZQ(#16vun_e2{vU)@3GdRO%8PUY$2+y;E^ zKI4*DzdpV2hx5s+&lN$>a`v>BJ2v32X(xBW+t<0Cv12RN`Ps83bItpkx)v+PI{!Rw zzVpmCR&n|`r)Fa({5NZJ;U~8HHt?o>UpX}^;E|EkQ)`bA>vP!AO2m54BlXs=q19U_ zK<@sS)+fIF&6(PWoI6bP$L`wTX2#{(uW~OQ7HzPvHG~U&i!1w$IWzQ(lhP?Y??f4(E4%5Hl|}I=*pptqCyk z{~&Ij-^ot!HNw87;cnI<_8q|Q=s-s-`-%So&M?QiF4#5kYbX2$ zej`V1zIEE-dPgNSoXtN2x{vh(u5wUycW~yGU!jjR z4Q`@)?9UiF;N1I)?!X-a^u$Ko?P3fqoNNDIRF;2G3`_fasu5F z`vBjV`+9#jFo;b#;10&L2S`x+-&ZgD$R*mk$|K(#d(}*|wUhV-JAgIbmC|GFW39Y( zbLT~JzvR`Ozhjj6e#6Kik{ZpecK=N%l@ z0sGZHJEv;eI-9eN4Rgqykl>mN#!zK0vzNa-m*FN5Lz&r1X z{RD5I;kP?{>^PifafcyC+_RW#59@-M&%v1AX9I2A9@P&$;jF8^*ynriMV`6t_O~^} z0%WXQYcB8);EXf}&K=Fuw#Ou2vHnTS_&4~<4eH>v2evy{a{>04z_@mMC;kgI^>W7^ z+SH2(G=9xRdk3Ne|BmAibeaR2ni}`{?eO&fiEU5c4|(&H5o@-(kdVOKD?QvtF_!sx z?;XFMF}iZlr$;Bx`ZwliiyOB2C%3^RXmaj*BrtO28Lt>RfHHFT*meHi#(c&67W@ws zmy7d#N;u&?&uzb~Ma=h3oMr!k@BNu?pDQu#A8^J~;CFPOjjM${&8fZNFF^kR-_61H zUcC>`p7??8;6K&pF5u^mU(pS}3vBD1m7w82K>%guifNnYOlR14VjFJ)(eL>s=ND@6 z?Q31dsPmcCb0apu1d2TmpGhnh+O|Do$M*i;u>S(?X>JQ*`sl#-v(^MMa`(l~SoY~{ zz3~a}GwRsZJEH|a^cy()xubn5jvm?xkQva<6}g^^`2k;NgO>lBxX;JhQ-6J5IUO$U zdj{$5Wxtdcx$bPj*~?w!64VTMd${i(*v1A>x-s}VYhbgRn7K9F08_vvaHjRu6Iy>Gj>zP#}PPQE&b86UtLIkk6M_4tq7 z=hXTBWPa&GzJ26AB$j2bEBT@Uedoot-KqUAa2NS{Psy(u{4Nh_kFCvn*$dx&C)%8P z?!*#%qWjKqRtMHuV{c(i6H2c8P|Kflo9C?JZQHuDN7_wt*S=d9j9cSjr%$KI>X%K4{YaIH-X{~wG(xv zIah}}wSVpXdagdk+J^QJx$*-oXRiE<+KS=ZJJA6zlo=~e-yK^L1Lk#brCz@ft1jrH zZ@j@Z^aO9@+QYjXdzN<1BG3G9=Ai34*&KQ2qTTfZCbkk;s&xIc+3nxRF=lUTuk>m7 z^%+OkJgASHSo_s3#A}{$c)3PD>$UCeZs(nto9{sm;(hd7ICHIg%Ck!gaQ|{H>;Y$v zXB)Y8adu!=4Y~n)tIbJuwDJ=^@a=6(UexYw#UAE>PQ2hQN~{^q32f`#yK>nn@Voqu zxucwC2vBu!Nj&8d^$d)S;;~15{@!t7*V*Xda|!$)=Si=D)>q`ryO~2w{ta6$i4FXT z77g_L+{L(idJlh5>z`HpwKUBE@&n&r2R!rJj@7S0#H-dkZ6NQiPY|HT`u>_|Va^5q z^?lhBzhPVNJLAqG@Dn&cW1XR5a0B1Dr(eTAF>B0M>{&UnY9{^w<_B6|eRCIZxF6Uf z)}Hac_V@;Oe(iVvPQ30KTWzl^x&!;WkA2*K;&&_);_H$lW!}~lseggg9fM+mo0Au!Xw>MDrq}ubKk!R0;!)Xs}`>H*w zeLH;M`|RCWQQwHk4|wNh7+JL=w$2r6=pe^@^9I_!O6l0o4Q)>;C>^ z{*9i-6JF`Lo`F5gdy-Ri+T`kcAH{LLh(_WU5S#jlXX^L~{sM~kbOBWtYj#r`&*e}2 z;@Hbx|9D@YJ>%{W#O>XzQL=unv&Hg5PMe&mR&v#z`ns2ZgFiXOs+RuO4PyxgC_$e) zHCE3#$r1Ps$PIk+6njRj^B5D>o?K(vC1~ra$FcFYqCXt^XumJbojl^xy{=AhZ)8 zl}EmO7tH8M0LYy@d%EXmd2PPo8&`)m^S?@+!M@Hi zjy-UF1#F=L_+Vsk!U&t_eq?U&eN`|frA?rE5dSnK<6y)WDP7rvO-&VK@X=_@1g zW6d|;eU3lj`{q)cx`7Ts&miU=auX!J+yEu9Y7RHa?YZjhrOf1qP!gEK4vFoY$9;Wf z{BNt1Si%L`x&eM*TW9Q{eP8gb-j8#x@L`S!Yp*#f^K2tY3wWUYg3~tNImQ(K8LM+z0;JB%xgs|LQY1j+ z&A$S_LwXQ`b7o*$Z~g?Qrrmd6VgtT|ApZ&f0@c%efmAKpd-B)+r{9;tQy zgs=C7oci~m48VEzKx>ac(^qY-e?AZ3nR0XHde;;EU*A`5Vt3*`@7Q{v?K#!JJMdjF zr_1H)8EDU`$a^mH>#o0O*!AwPf4$H2vPanlu?Mx^;3MeszJ85K__`0YGKu$G-|PQ7 zIVErdICms!_fT8kM0|dC=?8wG$U3)#Oge7No|RCVBrV&zu(uM&U9ycHR9$>pntogvAZ@o#zhBpmRz;91Khw*^v4eP z)=psk3!GkW^p+dsH{w-;sjeQxtb2c7H9ZJ|0cVfOr$@jg@J@V}z2__N?$8ZT8`jn?**^PGSGC6Ay$|m;uxm$b>l+vw4bI;~ zsXLr?V*|GA7qPlK&TB94eFj1F$?fF)1m2zRVQ8DHK6~Le5b;?LQ~Bl&cy;CL1NX8< zoI2zB#+@_KUBG9w^*wmzvX9UDif_F*`;;cG*du}QuRC}A0(rw8@a&@B8*ouT!Jn!G%7FZX$O-Y-h+#;Lb;!A8<;YDJcUbXu0aOn9B9N8u9si z7SG|G`tDgffO9@-iOE@K+}{JaLz;8)+VZc&?5#xbBR~6mJ*%^f+0XaJnvN~kV(}+G z@r~(gyRW|g?6oA;etx%94@mDt8OaYpZig#BKI1+Ub35Dg5oxpp(RB;c6od97)= z{OUu#GX_Zb+Hv>msu z^F3!r8|$DrZDzT%yj=0*jJy*rH zb}x4haw6ypZx3fX(=*!Vg|6>Ga!cTp|Dfmj=dzIyl()|FYl{I| z?E6C}{&}Wqza#zvJ}YB&Z)$vN&8d8BbHoeC8Nc4q@tJ4r4`A$qo@kN4SfHyf1UqYDq5Tyy~YfkjUK!eWh>pv@b9=;`3dZ_<>)N*yL9Y zJ9r*zl+1OX+CBJ@>pg1ki?^pa9cU|OZsgki+uHsAE&>nuBFj=UIw2F{x3nad7r`*^CBa{ z-H8os#aP73Yq@-D>we{TyeXWeohAFEd1f^1DF^I5fLsUWe50@8 zndE=sJ12ngmP=3U4&?nFuYPizv7xW`osa9BgfpkX8#iz2BR%*5>b@{XuJ14Rw{8OK z>i*!2%hxRYzkt${Hi3cmK0M1CFyGuKw)=K+JfD5#!}#y5q~h4am^?^qC4wKh-jjxH z-Hkr(X6)on>`ClHn3;iRE_TNj7i@dTDefTO5^T8<8@cyk*N$!7Yu7%|2^XODPy_Qu zti7J(`~)3Lc;lfCIC?50`5|J(-pIWG#hp&x%oW$SzcuCsw);=CcH92-%zR&U@^xpV zD0ODff6g(I5_rFVAK(S57kT!&;wRd< zJ_CDd*EzA(iaU(he(q{Dlr ze1ZVhXpdMs={M{E@}7O5_04J80$Y@zyF2VN81Dl9zrL^hhVA|nzV^i*V1km|^|9PQ z=jQ5M_~u^jtB0I9Eva*7#J(lDe?PQ&cQG%(P?IODoA~OBwc5Y0JiL+-yFc3bC2I2R zB}&kzwg){xg45sK!kn`cv9jfu*>}v^0-R#3g%*+^Hn2~gIRih8AytFU+&uch{QA8F z`VGujxZX*`F67b!-+q2R%u8_WnV-L&nscR>=kR>?=)j)!{fPZHyuQ49`TT_U*Ap#| zt=Pw$Kk@BbI|LlC|I{WY(ApR4#5-KBeE9_SKKN6MKc#wN~nB$*8FK|c9S9~v=*F>%y^CR|dTiZ(v@)E9c@T(8nGj;gDe%#l7 z?pWuBuYCM-#zx-RK^?>7)&KPM4y>J^@7OnL>dbHn z7r;59a(p&$-l?_P16Y&lgcA0_FJX>xq0HDe{BPu$f5G;is|H_j7yC3IKN1@Rwz_6+ z-QbJ|NNw=bI%eL`#!ihfc{%4!d}AlxwtwQWefN+PKC_87*IxGM`U0^g5`&sL7k0&= zE^rT!J1#$f{@?KSta`M(Xd+h5B*(ow{^9!R2W205C1U5L&a&M%@YVI%;4I3E9=_ux z!Ox91IQ#qFD4zW(M|}am-uqbE9MAX%7;w&-=q~6H=%&5r>Wc{)H4k!>I)?__6>h}( z?AGpy50`SpJAUmT4_fdUGLC)D>rUo8)a-7~1oG+x&Wih*)%A1FM(Ym{KwCY?w?;WT zs_XBMnc*%S-#%5x{Bvh7c>5$6z&i8lUcS`WOT4T?N-OvfnEazJPjr@lG z6K(yuS7QyX&WmlYMEki3U`;(gcz5wmU-I|`lsVcR3~b-i;=Y8E5SxaK=XJ5}a@e|vf54qRXd{LdZNHs4xn4i8&g67xMUuXe(9&oeuXkt5EY#&YNTZVY2^4Xu3S zX1Ue>pYOYdJv#iE<+Ct9h5XU>!yov!?~4z7Iro@o>y#PmnU+t#jxTp+)ZV+Fw%=*) zt$5b)edT6+e)nX;Z*;W%LLV((g6`BR?$~1Gc%L0xnD1xmKX;xz!WjFgs~22iH}u)d zXH|J{?(Ck;9{SW?cfTI%9U@ZqXI-5eegJzk@Xz-(%UH#sYxd6XxppvLyx`R}(^=T| zcD}r|9i8Zj7J=>lA+XK!+ym{t9j!Rue9zU;H)1UTVvcd^jp-XRw|dgMa_|SdImWdU z)O_@Vo)N2;@c@+rH{`e9ki^z^MrrJ(&mKq1JF5fF_m6jMW_V8@ImQwnEk-VFjkYoG zFty=Cg2SoJGk*fboRa(Et2VK{kKIdK-hHhf(Fm^3M8`LN&&yuc z8*BL55okfK_Rv24bJ@mH9;DpX7tSDG{ow0!LaluVv55Wtqs^1QITu?oE~nJ5HR#K~ z;GCaJpEIQX#)!6Np*wRbmmOzfBlcAyKYO&00OZ@hdQnpvUhZiO`wd+^u+_!}dSqMo zMxFcw-lOkm!u$+SavL%8&G9?!23i6na@|Mu#5S*L(dMK+a~fLTnDbKG`M~}4 zA87B^yMF@#XPq;z_|@B9@^JQcm%4X2^Uaw+F>bG`xw-p6)p_yHp0ywT2!bMsF=wiO zSv@vC-~#19cAlT%5XVe+Sy0A4K3Ie^VoGuLcrZZT^#(xdY#L z)$V+AV#9bzt{nTF`>B}x@ARW*#;S8p$G*Us->?tN{n&$8?7E7R<}<110zatx6YX=U zozX#!JvuS9IqqIF(Q#k?g!kU+T;!#Bpn4}UaD4l;;GHJ_@PTiwzAV=2~M896#vX&pA@m<@cLaJSgSv=?N?`03wTla_~#tPuJ(2&byfSzs?~4Q zDB8X=k<0!`zI#@jSOX(*(V~-UU!TJbC_VWR>lvIr<_2-!qrkSuL^t%=KegFY?#K64 zFZaS$X0F^o4{dU-b@u;Xy4l6pRaJ3(1EsAP;6^VxTB4)!2=v_Vdo|iUO>JWuaS{X~ z8Z$&2ff^V-EWxNb4@7-1@y5)<1Qm{niNyF(H9>jN&4ei7fuIjijLOlz5UD(wND&kN zzrD{HX3pNHI%T(O*4k_R*MG0Q&OYaUbh<1IJ_d2tz|Z-qd%)SiPkd1^kCE-y7Y}~w zD}?brw8WrUOJbb_OZmY1n&f9K_GM+~@`DHCZE5z8z~TEMJQ!mG&Q`jVuaM0;Mv`j| z4qsVGp?O^c{dDp>`M~99Tg_#%U*B)}$+HrF{iLY}{9G7@wO;wx24c)TpWLjG$+sdb z9A2tXp3l;23Rrjb2glGebV%ce)fAHgpTzau7L?zMacbBV80CVjA)`yt%su)=UH zTou=xI?0E-kkt1@E<2YWJhh^U(?9BkzwQe*zCz-MFv}7+jbv3e`{DheaaA5R!f+B> zX_V6s4g0u6n8JlRjpn{@(z$%(SZTkGaMZ{gJavKv!`w(p*-pBn7(TdjX$Es__T;3Q zTf`cliywYpv;3vF!4x*=IcsTTaLV?dwWZ1Txzf~6-0w+q)`63rdu}j_*POdHf}eFZ z*B1sX{8qA9iIG2(O>U>sxomQ+v_(sd1Rp{!y!D>Ettngx>lA2VyzN5G4lV!UQV=WK z4_EQopusTcw?9`dT!Y-c&PmM5PhU#>S`%I)pV7P)>g9T}H`aefnmhyydA@$x#F=|M zT=Esttd$ef-0xi)+qlpVxNJ3F0V@vXHMhr_sy=D>SPQN~ntUU>lm?5iE~D;q-Alhm z8~8KjEKUvp!E1#T9x^%&xOBeDvX?*qY`()xOscq%rgOfa8hvrr`gL!Vl8HTga zx*jKcQFe_JbQA0EF)N%fUiot{7YBPOt$a%oX?nuC#A<$Y>Ey4ZzD_*5 zT;=n*!m(G+A-~g{+`8}B2>dI`1&c=LBeC_`cHtR`&#x_P-N#xR4Y;hZMv_YkY>m?D zhYLB>V`V$$!sGWb+2B}%I5qhCH78z3=h8?iG2-b|T*d}07>2U}Z(ZPm2O+PJ&`&3A zYOx`nOS7)$k!`Cu>-bvbV-P3CUU+ZeV+54TAO;uwoBh$8-a2XUes9Vxz{rO-(#yhF zcw{4sRSsNM^ISsx40u*#XAv6GG;4t~lFm}dwX&VW`(5B%t#BSw#fUq_8LRayj8@QYcyty=dnR4;Xr9H-m-m;&f7R9Z^4{7i(R(SSTXwKQ<+_0YS zkK(Hmt2tv;G2;{y9jEch z`?w7paCtuJtT0BtLb4)kAz!&gnEUmljpAVas5E&Fdns^mFEFTsTx)}kW?qYv&A^{2 zhy9fNLdVD>N@;6Srs$ z;^dZ2w#3_-GnhZA9C#6GVO@Ow%qd>;!RdQ#6+@g9cVl_WE^gLmUb?WJlg)Y_BcIV6?u;5_GhZgh$*zPrd-Xmm zn_>OWkmEvatK{LACbz~H@|6;LQ}3ft907-3QH~?eN;C(TOEYpSmxLH~T50Pv*TD98 z`8&$zdRy7`HT4`dM%wR1DfoXSEdQ>Ym=;41V_BWJ8<$pBffrwa*VlG=Da5V(tb@M* zFHNp-vb9!O>5lML{v4qO;i!rJbtLex&qpNi6%u|&eXcoJ*7y0!HR2%7;JV`X{7Y3W zn!|ygo;Yc2#=3Y|4>3+s#btTS3Um3X5k1`4g0%rdZ&}Z1jxL4ECE(5D5*q{@F^`pt zktxnzjSF08;#S)CMeD9=&ZzV03H8FYDoqYp;@B2#NEZ?YYcjYNzE&5sk5{(GDEMeu;-;uxtH*sR*!?SB6T>DJ1nj21PSdqX1p0(>d z?jb8FghSwQl{T_-2^`>An}N+BkFhGvde-@=uRXU_o_tm|zvF?158-_?cqbpdEMyzm z@c4dd?$^POT?u^`aF7R<9$MM#-^eEJT$rP`@(JPB&TZH?9N_0m%SZR@{qd#8 z4w{V%^xLl&nB}7v?b7cg?^Qx$Twy+FZn^t2x88Q**#6Jna_=qs$J1;u9!Q!HdDirzSDQfxil+pAcQDW<(quOCr->g0nDnVG3oW2RkZhF{>}@VS7fT2YdCFEY`s|;RqRVq zjE2!gZKqgEQp{%2M%pR5S0<{M4B{$o`0Zm~te?EwH?o$b7!IRz@$&s=uQPL<3(0nh z&I=bs?dGgZJ7{NjHKrI$W}~Q`?RV;0iefgKMpd*^EZh}WG0Y}e zMA3dx*HaXO#68k}n|1Gws~C+Z@fB&`$XZM>(+j{fx~T0G3-?4-%zD#)wgZawh+>=# zhSR|gF6zR)%_>AO8jhl;);X?7iei$uBJI8Cej%!&*BcDu1Kv)to}`%Vtnzj;qY(a(DEOVMK&{^vdEXXnBvQxyH3+-CEaQWWF(fX_zzZfdFGOoBqU z>JFY-^DpguqyPl%xC63lPPUlw=RSc%_75T>xuIZ=W zD6J(ZCe!gGdTO;-(K+&_W;^>+y|72^{N|DFed&w3mZZ>o_~yHu_AAmk6;(0JdONyD z)>9Pyetfikc;jE6P2Z^t&nK!F^k+MGg3r%h(ro828x96Lpy<9DQ%nbm_gd{Q%WDaW zS?YPwelEKIj;olZZnHB_UHekY>5M5x*({3>_|K1*A5TAiJ5Rj1*-laD<8kw7wfAEF z)k`Ca-k`su_uyYkQHlVXtR(_QquxNHA|e@|1)_r0Y_G3=#Ymalr? z**}}B_1R3)_$2U`w>Bw8gT6jnHP6MF@->^Lm|t>5lVaS<63>f2zWd=5`j)^TjVdN@ zizud(ar~w+JM)oO(iHO(5ry8vCqB}&->IEPV~W1sYDHJ1{T^9MQcUAl_MiRc=?BtB ztNY8iiqRzTk*0kk^XJ~)Y^RQ9;!R`wM!Kg`6p4>_?G$TCio`pd+)<%>u<`oAMp8yV~UC1@I@PG z|G?cn+@zRh{aND8L3=OO5)^vo#NS`sdU*N%^y9aCB(7qhZ^ojHv{x~IErcMmffI`pch6q?@JrU79zJ2$s-O7SJluKg z9SA=;d-VPX=TBJws$;hzom!k&oIL)(L-#M{PrU#BM<0Ip9)0B0tshxEaI!l2 zzV$8}og49(xV}uYdHR`%fOf|ItTYf8g#%PcLSt7spRM_|T^==I?O-;!*mz z`}pa_Ll2+2^>}siRi__%_<`e#laHKwv^xI!m0RaezVpswPsv3etxml9-rNG^LPBH^|1bg z&!6Asw;I&rELd$&^T!N+Aj zbkCvrb4%tsy3Frx=JzO>-*Yh^Zp`27*n8fPd4BrDBa4SV{p9fz+@sf@@IKAoyQ{H# z(+(XS9X+}-f1grge^=vut?_=P#``bkcW%s|_qZA-9(c{^Q;Xxf?(7HbS>y1^fdea} z`SVMSySf@5XpOr|jTbEDqmB82WAApu9QnK)FVj4Ha=x->`qAoIIk;R^W_MA zc$a(E5jrrsXLQf%>U_QAetDPs73ThklKUeU^U22iqi)!nmAkJW&W)R7KCrTSWd6}5 z^DDc|A7kc^Etx-VF+a93|M(jY&dS|;9nMeKduI+EIm-TjVyW>-U5!t+#;24T#~1VI z#{9(NYV375Uv-ljhwnMEI={cvc%Z9s(i)#yYP@ zUT@4l`-anbaM$VloG$n7)48&GaCLNK{<$Uh=XJS1-`sC1x!=5)pV^py!3~>paL4KV z!kc6!@o4@W?IkYzGg8$yD|UT$JN;Dbbj4UY8<}j(2@Dqmm1&D)%Zqhd{e3M&5L=nG5?kuPUpcJ zPv^JpnSN#U(8}SJ`L~ty-`=JF4%2^UN&j7o`MHhxcR%h1TtA)PvsdPoqpPcvaB9mwWf=Jh*c3z&$JTpDww7rpx`a=Ki^o`z?!k z+L&+MusN$cPUnT2Wj=iP(ER61=3nSC|Du_Hsbv1;#r)#N{8w%`Ijeh~&R^YYXZSR9 zXq8VxzgBAedROB&tnr(r#&0d=+Z*%Weq4>cPUr9JRfCtHBS&}-_}x}=>W!!KkM~M{;NS}HjEClbQquovm;TR8|K}zBUo7VB#{4fIcLT1U z&bRKB*~SlW1OCd>xqmVLYg3OnpPbIWSx@Dq^KVOuzgx^NZ_NMx*t7X)^4d>3wb$p_ zf7q+W!O?;I!ugM-`+w@X{||QmA4}?gUd*p-%>N~Ke?9fi52635%e?zw9#}cF!pr7= zE}8#Jm-)Y%`M;IS|GJo8-I)LPWAD7H>FaOl|KnzPkMc?Of0n%e*5&=|-{Wn)dLnLnZ_dFHADXKtY4>w;wGXZotlj~Zn$?q`xoq_mxJ7jf z*Ny6_TyLnn$APVG#qB1QIaf!jchsIM-v-xJej2QmpAIdRp8>b123$9)XL8N2QP=O) z?vDV~JK?cc(E|q$k2uTKJ8REHZ-?uOJ`2{O&xV$w?*g}|p2KycdRMMD+{SB%yLvZ_ zd)6BrI?PK~^<3@wh*x*OS}=0XpY-PH-4WN9`Jh+t0bdHgC)}bMa^0xji|ejKzx%Un z_1@U`tj(u=K9^MQqdm{MAFiAAePNyT{h+1f`@=1&JGpLD&*Qr5Vc7KvxB39gUB%tc z!|Lehfg`Kc^R?%ScfoZPKM>Z6cSB3X7r-s55!a3C0M}i+yW?|jwSw_x^^S~=R0p-^ zdaH0?(P2K(Vi>c2iH~pP*^L!1X?P;6mC(y zjO#}AVO)1THaC89u09-(J&W?~>o7NVU3)J2a=5POD_|}95ztcfBjFa+gzHB2QCx4h zvDY7)4UBu$^V7tk!z1 z&F&A`)hA%zyY@XtkFvp^s6Eg6NpRh)pA74)p8_o22Uk{nQMzAyuJ{04SMemQ6+abPD!v+SQ9a0Yqk4$zuHD`7fxLPR#+%hUG8$E{ z)t>7;4A<3r9jx^pftGrY!Y!&9*NtkV$L7Y5^wsC!u~*Rpt4CM(&RKn~_FVMy;JTup4{OmkK}*p$!!4>aTsNvO z;CjQ2z5dvIA;vxH9Xfi?Q6AGT(w@JJR$mNj!3Xju{_0B*`w#uqmm-wPUk10RmRvWg zf6sN-bF=&7fA!_q_bPsHw6ek%jp{43=V8ARt{e8NU>)|Wp{3;4z%8n?TsNw(<+|&+ z+4ZHM`Z~;A#T%ZRLr0GsIZ}PS_FVBB;JS+62y4Y}f|iQk47aEnt{c_2aNV`NJH9GZ z-->b1dWQ~-4swRRO?$5Q?QmVa?|`-5cS1|O?}A%Y=eTZE-_3Q`>D}`SMD;zm-DGbM zagwU<)t)PVA6!@Y`(ds81JF|W2jLdgCfAMXhq&HQd9UZ@hjH7p^1&mA4y>+LKcYQX z{!zHD@{hq<`NyH9@=w4ms`Feos-NV#>$$n{%SQE6cAx%p}Bx#-Wpbwz&` z)}lWLEk)k~x2O`=jcSYQ4LA1sb8`XXUiEytaeIGWJ8w7Zi|Q9(Eg19fIMpvA_CMuR zzl2aK|1#X7y2y2-`W3Fnp1}w4PrZNF_vPwWG3;4&v~uVkJ^=rk_B`dU!*x^s2CP&5 zCbZQ0Ex1Ls&2^*tZLayJZ))xMmQwu=`mWRsPs$1p%BcEX?YY$N!F8p6AJ$TT04=5d z5N=Uj;<{1&5!Yk4m9egUZK?hk?M({#TUhlc+H;{lh3g9a8LWl=99jzf1>B-)xo%W{ z$u+<3Fw;kuH418d2@g_e?k2e+s$bKR)^o@-xnuZQ9v zu-~NQQNB`E|EN8e{3p1s@(q-0{YdQ-S?}-xemogf@2tJl zyB&5%y=Niz^_~r1>b(mbUp~3=<&*2_^`|{ny&KG(uemnxYi>H^;RA>7Ie4ggF8&nd zcj@X5Sc{gQrmJ^Hy!Mjz9tfrCd&2R}lPljmxxVJ6V;x*M%v{xbz2g5C@7jfOFUd;98V;x@c zUq`EmQ&(UNyQ9DwVqf4x;7ft~;1<<~a^0w2!u9m}dp}ma6z1)jYGs9g5ae#X4F9g^ zhryP;%s+p8IAZ_v$2xo|`f|8M^$M;V)kko>=~OHHG<@*zq3R=X>IzI?cNF+2#J<1= zd@1nJaEt1dTsNwZ;d*-gd5={e3-fkOb@0%LUt6n>!{4Hu!DFx%?Y^CUJYv6H)5&*GG?4 zCt=>MsqPsay=P@qeJcJG^%djQuomsUJUoba?c6_vP^!KLZc)9K>qhl3*PD-ZXyqP% z@p~OkU4=(rcT{*3@mhr$La8u^TT~U-jp`KF)9Z&Hs}?YC*Ib8J4;0CFe&)|CVxsDt-!oTfSpNUgf;W5}975*LKwF;kw zP%6}Li|VtvZd9Mc_4N9S9;-eV=Ixs62w(R2Y4P*$r>gJPpAT!*{BHdw1da&r)^CO@ z1<$}OsxRQWQGFrTn~!vqpSJmG^+h;!$-fx(TJkSJD9OJRt|b35IKD)3j)2H)C zkJ*Xp#Osc~x_b2V>{RvT=s&so3a*b;U&-}OADr-TRecror7s@6>J`89FW>!d`EOoc zb#igyRQ1(a|4+Z}`i+;q_@DpvH~;9_Z+`xps;@zx|GeZRS_F?Oqsm$M}!lvJN4@7!~@mWVfljmKK{@n4=o-#@$hk4`Fh=7Jn33_ zGEsSiQ`I+M_~hyvx!%bpKVkie{A#;z{TuHUx)wDqU2KVyBc z{>=4vT7T#D+qoIfUVoSM=d8c$`n#<^cm0m_cVB;x_4iyKuD{p%d#}IG`u_FzU4Osz z_g}wr{dwyju>Sn@yVgH&{qFS_@S=ZUeP#XN`s(^U>xb44uOC@Iy8goT4_d!>{e#zE zbn^6x)A^It^-oZXH_lF0-$ZNOU-V?l-6tM6@w!{bUaD`#mOpsj|LE&x)yY$*_)C=I zr|y679OitbnwrMUU%ZuYft4rFRH#ZTfU9!+x|nM>f3R7@NNDXPxT$Rd_?}1 zzw^(Lu79QY&TM}>{(h$Vu01{X`T&0}d%o?zWvRY<&&IBQT>hSH-~D$~H~a+uy|_I1 zHvcN5`o2A??EH=U`?G!b_77`|j~?_$l>= zae44<{@FqGBYRZY`CH|WX8Z2(_x^41$M*ExYy2P2p1a3?+;4Y3v1en~$3LI#yT`xb z7qOqj<-xc4@7mQ*?NMdtPfkCb?Yqa{`zNBG+0%2c@qadZ?jHYfKf(Omo{e1}|1H^m z^M7)!5)Kc(jej|{QGBhAzT07|9<)Nd-&}&^e<#TZ`V!#ZSWWOXz2RDzm%;z z2fpE1|79Fr^E|%x?)t@Y_dk$bd|Zj0kNU4<>&}_?e$an)55K);{L9Z{qNp6F1Jh`(HPH>v1J^KHuBfx^w2epYGq@!*8#de<%CxocVE2 z_V4b|(Dj*rFI#uce8W@y`#8Mj|MAQ8A3Uza&WHLBvvudpdq2{b_VC+l=0D1QJ7<2} z!(Icm*apstx!##S``v%Ks!l$#IKFsvcA)wj)Gv)6^QLG1 z{!`!jyT_hN)T>VNx3byxx4Ufbh^-HO-FQ;{9oF@5%s=1GA?hv{ou87jF8^J2_4jJO zD*yI-|HBW>W)D5`>f^6Fxp?r=2daO-bnI!kzhZps_}YiP^7u!;?89F2%IY6cpFVx} zWc?|({Ofmp!27)Q149s#oBqZ4bHlTC{l)kTx<9q<{^t25YtOvlkH`5-^8OFct@5Y8 z@XJ50EZ$a?7u>W8fB*cFwdX&+%C0|I-zEJ2SO35$|M~C#$NGG3-wGk65Ju1lZ6Y!V z;}Fh)1eRb3!dCIS3{4%*K{vr{%&vhez?c?X20@;D*0{74!c}m`g%Hj{%~Jz?8gLa1 z)(S4cTh~}?7e7i#?xu%3NwgCYLbk{^MVxSV@j;;cOZ&R>Iw zPXp|+1;z|OBJTXS#GZkk1r55Em?3uK&p;8FlRcxeFU$o|A+J#p4Ff8Eu(h`b4=$o+jcPTdvI!aMtI zYsEfA_nZsmURcAu*Ve#?3%t&Nh)*Ks4)g=4HTR)kLS6gNyN7iu_s9F}eYgtLTl)z9 z#+iFRu-_Kh{PwK>UFfeso$(g*tH9a~x+B~{jR zHR#OieX-sp^p`nyp<3_|tLnJrUdQ+g@eHudbAE1+(Tm1~Z%X%s8=v_T> z16p8DZ7kFp@|)zaM!pZB+#7w3JA+=oeoxeo&_(^3(KBdHmp#?7$eZZuL|t^|Nbos7 z`q>USA*(m`95BB&3E<)-mUd93znLFFyL|U(3hR zL!YNcp5Uv#1u>OA)Qz^fUN9RUvD>@v5B6z{ehoC=Z+Pc$_4}%9o|MqVQp0^PUjPy0 z`p`C(xI|2^H@Ch~3PP8!MVG)7S)F$CZo<2NE%FS0s7)RBxkhJR^-}BV?@YdSdatzY zUn}!XkQ-qCA(lSOW846YfH^&bSAe$5==#@P`#XbP$Zuowo(z!9*%F)kBF=sSi0G~5 zS+&j#+4vTlv#8TR9&3`LxF^;(_tjGS}rX z6a;wd+)3O;=oNIUYvuLHIQ0YQ486G${!J=56V$pby4(|fuIv*_9|7t<2K=mTo&PF~lsQb1Fb^cevKX{?^AF;Qo)M$v=MBc{7p4~fR%rk%w(16cy zTxL$tGw7VhXU)C_P%HRHvOaf{zh>Ds{ZS<~N^udvc~0ImbZF86rp} zSCU8Pxig=uPNjcoUfAkbbVGd9n?LvDo*;(~=F6|q=-msKvF09t23}-;XaqSQHBu_M z2JqT5g$r( zC*b{Yz9Ccupe|t3*E~~@;QKrgUYqgut-gk@w@MA~Th0p|bUM7UkN);z&WphF)PPNC zEOQ2^dmXp1gNwl2siQ4)2=p_LuaC8{&;+cp1QTG+n2P^&tDGTeeqPZh_!$_DulM5Y z_U(Kw_h%D1=fTg{?n}@Sx>sH(?``_3_x&*Ugl&M$=hNIZ7=nmwK7A5!PW^l@Gr#X| z6QKWobT0=jG3MBZzJ)gUwH?gYfQdAg`nTa1@S6}DfqVEebW`A7?d$Yc(3>})_dTp3 zUY&O5X|U-Z;jQRCd=ltqiDV3PqM=Z5!=+Hde zGvktacw=+lyB7K;6`R!DF$NfPPM@Yc+=pjVNvU~BZdP`(%FW2}Dm;Iq)83(yE1f&`!UfH?hU$O(VXsj~<8+trR)Yzdqa$bht zhBwa7fO)^+&6zqH@f*N9w{#m@K)394BV_p+ySYd3=GWgniSn!&ysyjl?CWfc-dO!z zO(F;PAjiax?|hn0wnhR`iq2SbCBl|3*ErH8jQIpY`ye*Fken zsPr?ZaswaW6Ey2M3ls8YJtp-DkdA~c}0?*^)$^H`$>se^W9SBxFf^&AP_J=K=? z7P*CYuYB&j5B4+bXvD6p-q;);imo9p+oA4Fe;)FF5Th>Z;cI-XX@AzMQ`xs9ub%wY z(We0+?bg|=wXN3zYw44h6sVUb5alC0$T^YCE3co_{accE1f)3*y*;=Wu^E6|8z(sHgTmk1gOrXxv;A^a}o#G1EM^5ukfwdw? zgF7r&45T1I!cDfs~+;c=o{@QEZjjhfsNK`k6BUY1TpQwSf+x+T>kB zgt>Qx3IAW z&|Z#-y#VUmOY0`&2+ey5uU>z3`kFseF=fyRE_I^$3=70vETuWo2 z`ql9EDO@SSry^?;&eh+`5xje|C%i%;hDVMTQAsfel6m%7ZST_kvyI`d3HkWa*dR^F8g-#}+uWxpc4diN^# z2FZJYyaeh8+Q!8{;hX&;lr=#w2Edi`!MjiP)1>ROw(vO*dS^;d?XHvxpKEu#dY3-- z(8N+lImbbB4fMvl+Ej9fhyc`^(^%!Yr>Xd58Q%VLZS3aIUz%fx*KRI#wR-w-2KCN5 z0e$X>tpPR2edE8x?rV4r>g?nwD+wi#x74Aof zWv?!e%X{TTXhi3pIa@zhqc6*zYv3L@TN{h-nXPhfM;#5&giU*nfi|hU7iZmt(%%%n zj4^$q+l1PSdr)`&+U?)(6KCPwCuQet$m<;DcJ?dS{BC29=5gNDwd?o1#Xo=sxMxFr zaossrwC3!OmiGoO8U`)~&L_YBmY0=)U`*L)E)==#_x zd_rdrOXvWchgau*8n5rLlkInjPCs?#>SwpU=gZpm?L0&6@KX?pt?@Db67mdMgRoX~ z=F?}SP99`Cvit8DHs@vZ=BV-2-@5kGLXBzBIYSK}(Ph0p=n`~<{tUc%oIAwwTiLf? zD0R$ft>Ic%6Td6K7<Fs`K%GG zO`K=YIQbU6eR|K#k#+Dj{0OL57xe{sM)}m;D|~#eQ|IrVL~O>_^5EXJ@*9Y~cvdb! zYkkP$zFR|^y*NXIUVa-sfC#TWp|`$$Nwsgno5z`Fph4DV4ewi?3I4o#=q4Dh724uw z>YYoUdGdE{gd(MR% z;q6fxwNppm0Gd>&vT;Jc(4mIvBU<2+Du*3#e-d)+(96!KulfK#1lmSG*)@S)i;Y>8 zFOzZU)p&buvFVrRg4+KOB=ij!!5gDqKA_9KolT!Urp8_)C!kNJ9pA^O%ld>5`yy}8 z5jcaq{kXIb)v@{>+f;NB>M~Bg27P>hx7Qd;eyMqTyLnrD8uXs6i@+Iu)_e^+2X)=D zR^F5KnW5W~y{|R)AK{Jjwd^7^;vZAVqu&&mGoc$|H_v6@zAVwZ z589n$hTb)RHnAY2(zA25@Mo|s!34OBb)Ep)SbA`;YIyt5)^v7t&soC{yYUnB`!8JI z>jvFj*zGUkyN%s_vWC4IW9}f`Di}cRYX+SH`;Fr=gEQ8#^wB_DFvDi95wJ$x`N|LA z`?Fi8vd_b&P#DWCP z`-Er@#@9WFu~tM*sni)Fn=dbDAKT|qwr_o`W1K7BKk}J>Kphd_U3DyIQt2^VD0XAd zlPdvpo+rln^`Ahuk*#6w(ONkV?(u!ZM`*&Y!B5_LIX`kRhCZd<#~Em2*)PwcIn6Id z$Sp93KIhOG6W}9$>JrHF1Xe2N+Wf{`1PwZQa}CD0bTG#f-M$N@rg@#$8Ug(_FiyXK zu7Re62(k`Yo%5TwMXwy7*6;URy>g39)W94=a_3m6&tgPBgj!p>d+DlEnb{ip%+T-5 zX>Z0g_@xdrb@*9p1aGWMyME@0Qc!ou`QgnqAU41^%J5l^Wxjmx%m?h>m15zIRcC(V zr^LGYy%-@|&%Aj>e2Foav7$j&1ASWPWUI(|9&-0N&82U`FLYFYm(;w=J|9$_y&E?G zo(=c7>EsB_xyEJBJP*cS1`U{j0Frr}d#jw08Fb%;l5Z1U+1~ZD2hWuEH-RhQ5>Rh0 zYud{dkbv)XzD}EC0{Xq)9GCY4-y4nJ1|#6A6K8g-JL}K}EJ2H249BHj9m_eIBIj6S z*JkHyFZOIqp0A@3ozONRHb8S<_-8*<&LZ^L>}U-?fETV9xpcC+7FmD$SKq*ks0<>! zb^03G0vO_3YfA-auki_8*ed&<^T4Nwj^%uV&Zf_>gF2VB+=~&=*ZlJOO@J%gVzIZ- zzPDUg2mR1zIl;5?UPTySj4Q?t31M7%)Fu*WIzFUuZEy4o=$&7m2;9dJut(+IR>L z?u9YVV*P%O})w+V#(Qum^bi9D*nhX87nY5^`SHYj|z?xA5lG9zlYSV1Qk|@3*(*vyPuJ#q1^DEGl@CPnyEzx-T{v)Yrsq^xCbb z?-D*>*Vi6>-!`Z532F}k-v-{#$~Ck`Zqc{!Ephf???aHVoreZ=0b2y0M4Y+XSZbQX zxCE{BL*Kwp<; z@F@c5412a|lBb3--lGY2>-qc4KB)V2FT8te4wpTfvjKho)~i8?06HYD>3H?6JgYwk z0lU1u#zd%o>ib>?)Jxb#oo|EPJl2$t*aw||jXt4s>63Zv>Yc${xu=8!atj@Gc5^AG z7`T)^)a5zx>DSb_h@QUPL#ckoL@>oyImVKspCh4DZ!g+{lz8{Nfp-RxZJ29x#(FP| zGj9YlV6I#jAE|LOFaZIZXU4wmHP4A(i@pY%TZImw<{NdeZmvBpInCAU`dSyYkv9<| zHJ^Uc2I%iT%MYQh+;=LrHeUCw;hiPtVTObpl;K0_dRKjuxO{et_$Opz zm!JV7;JgibX-H)cd<`<+P&+wW^buJq^qqEgW9`x0EifiPYtR65s&`)X(jjOPJWKxp zvN>bSsvOhn%;5}48_4xj=}TSiyUV9<3nB30Pfd)tNb+HY-Rt=o)CoOQRsZR9uuZLsOL z39RYwKK{1T!Uxb0Z-2%GV*Cwx0M03c$rd$$H7`R4&`VMfVwurce;>L6b!O2f%v+_V z@uC4|L5-~?c5__LO`Z{c1gvSCGcTbtFa_#d{XJcd%Y8OazF)}3#Do*Dzg!2tiAB~| zdmmFHC*%k*R}YBfWD2z$9)T(E>n1O8aEkZgDOuz zgg1Zo(}Zm81Z`8fUoCtMuTGnK_fDPp613^~Iu^URJr4o>0A4>~ObhgLpXDQbf;P|y zZJ`ZRowB)dkD>6+8nngYr;n(S&tOwOf(m;upE_d_Fjqw1&pm+8{!-wwS7SsiUVNQj zVpFG0zZRN$pHzIa-0^jnE7!x%o@#8aCUreUge@S&z6quvBDbkLyPjpAg)3lyO}}k$ z1^E2>Tv&IwMSgJCh4SoOg`a>x9&7JkE6>lJ&;ZPDOu*lmJHb9sqpxH6Jn8dh4WHEr zRqvkJOACC)^YaR?&1X{IOHg~XuJs#mehU^zr$GPV7|I?ya}8X*PJVF$v7_d)w$wF&I&T|GE~#-Y^)YsGx=yv_`Is@FkLCjAl{csRVsE`) zgDsy!bki|@+R)|vP<1&TeT%Hk-m~3)(A!soUfy~+KfFFUA9ic>`%&YgJnVb}etAh# z<}%H5QY!%an1KnfmV67Xfi+UcYcsdAjzEpgI=LoPeFIwbxmIUK%FBG_YQboXFR<1C zcxJV!%jY+gUaS%I-zxP2d_UU&ooCFmDb(vb9CyCRdH=}K=dp(8&pZuq>Em4X-2iD` z9kq@=<__p;V4nWosINgFuXCw)Iaf@b97|s1>`y&wDyI$&R0Qim_0yJT?|iM%pzrHV ztcUE(Ei~e>r^UF7&2DvU&UBs64 z#8_vtRq9G>pw4IA*9Ub2U~X%`vu`~G>cQNP^2pyK#LMr>}O!m}v6e-(`)lYrS105O%8u_i#tM(iV4Ip*MzNngn zy3^Ir8aaU$UYOfCvmbi%_d5Lob<~BPN7i4Ti8g`F_$IQa$WsuJ^|R0HuOHCQ6+1bl zvX3p)IDKL)eoOcvsIeIr!C7VN>xY~`0BuLTPa^6^kJ*lkI^_{m^s7ll&haQ~y^1~S>y8>7=rWo&CtgS<=_AOeaRl}ZGwFY{0(Y|+@ik^`vc%Id_yijKT1Noy>H|I^!aG}t0DC^m>OcY$eAF3V1O00J z`r6K}K5ApRXG37_p0){lr?Xaq-vrh+PF=r0_IM6hV-w_a0}c4*SZK}%b$_Q|)}e*E zr%Pb|2{6X~%ySu-)AQgNya?}ozXBp~PGLUtdd>rQ4%kMPMyU5JlCw$J$Fv4~2F!g2 zRBXdfwc%$E#+lRH@;*DlyqAG9X&a8Q1N#Wj0kH2t9{V(B1E1$gQ1}Fm!1%i3l|@iz zefhW}7F|LfVxN#xAJ+_i39Qiq<7#MtCZNBFphd1b+Ca@?{~@BsHh?!z1C5=nK_7uN zYW3KsTP61ZIs^gq@nKx0Q0?_MpzZ#!R47;C;-snBDo! zUt>4M{`9q$_hN|LpO1u2m{b3JFW~Jz$A=P^u-SLE!<%o2>}q3yIdVPqpg|stiGf-+RRVVB(?<4y&G~SiU z{JlQ=$D-F}PYL9?puH}RJDywvczsgpYQkHq1==>LH3jyYDSG$FT1)Ka)PDl+d*>!l zUkheduf2~qM+DoT0Yl*VQ(vd<8O6OJCV?D3Mh1Oe&#|A)lt(}x{f%$IWUJ(~chC0> zxWCpN0ef}^`?I$i48|RgZU%o5nA@Dgj@PHJ)1VjjZ@hWTZB6UMPR`FhvcB@t0L+ts z_aY+K@aiVW`dQ2RvC~Uk+S&vIj0;=%0&C~GP<=uyeLG`=pYifia|@TVE#)If<-WJj z8n|cfgTCI|8eG~cwh_Gk`q)E)H}@IzOXR6KWOYOMrqh{Mojn@U0(%JPB0ky!w9l{2 zd#&9ygP#ItNwM1tsLPy_t+J1s_{7vbhq3IdIi}dHVU7fK@026976kY~XN&MH$a@=0 ze&d}{z5Ec=$Puc)d5kxvfme2hfGo9!e#Rtt^%HpKv48WZxBnVHWAgokcSU4t&+uzv znRf}FS1P&zvOf0U{i~q~*%|?=-JE@HpAl=;0=xIr`|W4gb5Q4r*vx0Y`ndG(_2$>6 zJb@17^`(wA1Nv;22hOan1=2oe1K(hqX-}muxUx492K^Kmm-j1`zU(c*8&kt8JI4$( z=vtsoUVDPNyq6=;=Zzh&ZV9X<^s~1F^1XuQ9#Y9W-73%i5ZZtN&@WA4z}I>CfX%o~a0%>NyZ(8oA!owg z-#hDF1)E?3l=ZPMlzPcHklzA@*4_trS$2U|G6+v@~e z218)qF_qebaq+F8=9&R>O#Iv>hhaK%aF99(YgeErO=APoehN#Nso+ zrjLDC!@Ln&0@fY^eP>YfMC|UdzSisW>1#|2YV7W-el>o^C+zx05YXj0r1&&o2x|OB z*sW!ZHM0IKC9k~wMr=cT%;#%{`YAA{b=9@_1nuOoW-PP;++G6Ge1_3q!R}sr7VO=<=yODLEvSvBMgyH9M`(=YS+u6SE5{-a)g}D# zt%3POfcHC$eYv*BTw>WPWqaDUh6*~wCxQvQu>KTTec#Ixey04EQd7T>kUpmD!9_2 zf!64=eO&I3uXi<2XMZi2ZV?YgKwW^=P8avq{AF#i0q4erL48|p&{>FJ;&96Qr@@&`8lO@!16W8{6WmK5mt>JcXLyJQ2+BnSvUAgucah z05xW!p8h7tSK!qDEpd_huGBH*S;>!IRbv_mhk;Ep}*@qdSg=T z`h!UQ*&H|_r6Y{_BVldmMJvHlP{8Ih|btNBUGL88T2ha!W;u&Ui15J zD4WoWU;rQS>3dJa8`D5X*z?*dK3Na%GM9aY&c|H$kt?i~S+}v-hkKc8zl9vg`kQA8 zwWho^T|Et{n*-X%TF+TSEayjGXKtVoB=j@#HP9in-;ZoZ7e(sy(7dGP)E{2o5prM4 z9Brq!5A((zF;JBLkXv};Bbaou_0$Du?tANcuQ^vI8{6O~6)xlQY!M1v_N3k(teGN8 zpw74%`IIAYrZ$yX8fZUxy$?TynSv z4ZcfgjeY=HV?hM!@_Y#@*k|WU*z)S^gK?Qvo6Fc1)H|rU0Gz>|jY-HZ`96@xx-pf0CS!PP5xSHI#ssLn*qeO`b$K52+WLD_ z!<$pxaI5I0_EHP#n^XBLXyAJpF88f}M6a!lC10Ld`?&OFyfu{VC4wxIzd=7luC=F< zLtXYmPC^-2C+C8yQyy&ea%IzRX*w4rlPa*v%|-X8i{Ch*4QcxdeSyqBn( zSb8;QgRQ^kGvi#b+rw1q80XC9Ngedrc~26uy1Y_%-=X^3&j?scI7fiy8PHEb&c_^0 z?CvFYdG|3v`+I3zR4=0YR|L+oqz6TP3gWd*-(n-vjg?j*GA7#Jb8e zV2u$lhkO8=!1x9~Z3#?(c>?-?UO)AoL*s^E2Cu%xu3lPOHTA8Q*wLk{rT?4^abPje}U)ZG_ym~Vp3IC(!~%BNWBHSi6(O>9G84}M-* zLd|Qh_LFBcA3EzhYXs($H=j8I)SRw}JjFJJdRF(Zp@f=qd)(!Px27xa)rF!@smn*q z6rKJF+z0<0bTO8__AEKm<*jm_8+bo^wa_Ix^+Wh11uby_+2wobS$Xt+#;s$?X>Dhy z(aY!0Z}`cZL*Ep!vmgFjp?yd60bNKX&&w`Q3x0;}Hh6ulH^R>z8+?@`J{O_J?!(VH z1GL#HH73v+U5j1bej~h?z?*NW4m9YTZ3Gg$eaWlSzfSO5g@?=j17!8~Z~ht3rv_($ z_e$FmY<6@A4YBh}-JT-$XVJa}Agq-cmA8TSN;JTJ+gSET+ROIpUS(Y>b=^l{Z}w#` z2{fIoE_D0=D)hDgK%DvPO&Vh1N8_?@L#VuT0IXX>^;h301NGXQM2Zf<6|s}on#O3) zHKF<`C*Z16mrH6d(4EM6x|$6(p)NvWL zk9%fL^%Hq@V{IrGTUCD_&m^czB_Aau|^*eZD-aCyG$gRV`bR;~}Pz5z4xjEIxBw$P?7LXDGZZ{SCrPeQJN zb))F`fL%DRG1`UmWgkTMQM++D2fVAlch@K}Ar{}9gE(b*`%LoPlFv0#*@qf=f{(q_ zU~9CP6`fpP-9S^IO|I%2IPLf&Db<@HNge%IFHp)L09PB%eL@W$KoCj1f%*UE1o z0bhSdJBzMO#YbQ7v9f!tzqO6gZ{LLy)915x)Q-*A3Doz-T=y;I9(i6o2LZntK45oG zufVtHb1nE9K0uvAyM6ectKHc1$XB(2i|{o%??aTw-+4FTV=O)+_#lr>n!psDzi&3$ zSoS=F&u6uxO)7IcTOIWxJeB;dam2Tg{kl;4b6@PoJ+b!&TY#$feXs9v20uZkKH;O@ z-tHjJ{OT4{L(V1@vuD~bscM-RQ*VZ7rk3;!X zdNj{1_&C!he1Z<}kv{{?b5p~a+;5k5{oOxvC1`;9d*RE;<#GmN2Ixm%2p?j3EwJ`^ zWP4KIqHo}xv4)-l=2SO==7mpy&-X8ty7p0{zlfi|XYN}o{dzBk@KHN(hRaa%F2O$R z6KGBB6uoEQPIQ61c}95Y5?cd&F3v&w+5tXvx-c$r&Z)izZ7ln{jm}1drw zesxhD@{rgeH4?lr0cu|D5#E(^Bp6_>1R8kbYNu)bp_!hnf0bV-lXdSzKfSSi1^8P_1vT-dm zD#Q2tCf{roxv!i1QIEeieeA_}eJ;W~YqnzxAYsdSnWy*5cJwo1tUHq8+W=>AhOkxi z=5rMDPC85xfOI^v35Exi{F1%`29Oudx9i^ChqioZWfMeSWL>U547TF)nM@ zz&aN@Y7W_sb6k3}zktph=59a@&H($GNwE*d^agF9OJHr~(N=lJ zhEV&o&R|^bp)-gn2-2-m;|f$X2|YQ?;~Bk-k7vwY0`>%MyHNf=maFjQACA$3Rs4Ma z7=egvF6SI{{yr1>CvsoH?m5@r8k<1>8v8Z~_!>I{`OFftJude{Uwe0^xK-|*G=m)W z*hOvPC~UQ%mSYPaYz4ImfoWuG#7qYz?BX!>=@DqIP!&Rd6Xr4RE> z$E9!kOYjZ2k32JI4O)EjEbx&$=Cg)-ZU4^Lq7UeX*c$kVK0s^Gz+b|)|3W!8o)v58 zvx)xzn8#jYEcGs-OZW^yjqVKc4Cv#U!bjy;&V#x1$$i4V3_trja-r*CT<*)ukli!$ z?1Q&w_f2ZPE8r|hFG0oz429MLvsu35LL#ta}*@fpJ%WGnjvhZg`=*kF=@u?fubaZu`_Hz=w{{ zeWI)Jc@BA;(cJcDoe^m8)qW0a0&~^iJU0FM90P1uf&EOOiMV|i${zUsXN>pUdm^=; zJ1&$nW~~~zCC=9ZYng9={t|HR8L%JyE|0O{cM+<56}@|{Zyig0??s+p8ND^^)$=eL z7es7{cxN`xcCRN_gKZxFfRV9(cYQ_%pzs8hkh5eFle8a$iPojvxic(3z)3 z*G5br-&^F=A)+%^4GrT$&8MtSUR{nH8_^`tQG0Z?vGft4_UjD7UMC&&N#k;_eLn2p zt8hxW5@IECl1cR;a{zAira=+Di@2-M-*UJ0Y3_gG> zTjk8T2hOkW2)$=mxL5i{YF@;4890-(*+W8~&&9aJ*@O2;TMI_`IbRLbrw#!*_B^y- z))(fj6JS5}H7+PeD3E3yHu(lPk3OP}=%M!0KwZw2^_-EMHt$ur-OQy=^E*-Wu~mOO*pBJ10t3-F$ObphG;)&_nF@*LP!*UD#c zJaVojx`6IHbO@ZQ@9PS(x!lKwe9mdki9WzRGbUjR&?!1?_Nm^y=8EvfI+uHA&ZHe` zTq9-G@aAYDfVXe$Lr|lWH_tZI8s<-leAuSQ(ptIG<$Xr)YmfYG7rNI8WPLZWxrWI0 z9b?%OW&2Rxxy;=F_dSAvw9|!HWOepC0_x;jXaZqeW^yiTcrPP*a}40^q3QhWcd~^I zG#!eY^7e1O85jZU)#$BRgCM0wol4&=R5)ASTd4IDI$_Nrh~rWx z#gOYnDLU`#R0@WjPe?^?KdxL4zs&2uRr;?}$r+H%Biu)keP3MmuZh@zeGJC%No%3X zna9@^lNbI2IX`mhW%4)3A_42U^fSM-g}QGpcM@?Ie8U#B8lr-8SZs1Dg461=hA2jv`#ZWF%MuYfSwZ`#}7&12nK+R5q?a8CkI z@7(I-Gj_x@>gX}oH+Ld07{N;?P+u3+haJ2%bwgD5o6hlCHw^V8Z-nW zVC^3EnvkuTKm*JzwXXd*yS>}QXb%I zJ*hrZY*&CjHTEm$?Ii+pxi1$xcu$t-wTpy*gHOV~??UM-qVw4st|3G7eZ)2d?sdX` z75#qf_Fbd%z0~-YT<6f)gP(Z@Q0uQ^^Sg-q7SQ{-;|?&v?)O^lw*ceS`I+C(@#b>A z2{!$f@Xlp#+8cRvm%x7P6KEiZJ{LQrSoHeXt38acC18I6>K;x(=wJ^MVD8)( z_jLeoZjtrmby@d37y|p$F5g6C5Yg50EO~3xQqY2kY(4APYp#c0zaelni3>1KepcWk zeC%{#T_R6KE9JwCfi@1I#r8_GUi&R9~m!Yahn@ z%xnT@HP^D|@ioR+bM0Fr9(oQ8L5tm*uGtnc@C*+wlzcV3E5$O?*>QQE6VzwLJT(|> zmHwwtYhMIcI^KK}5Xce7#s4bQ_%p!y%#qht>BAgDV147AH=i?{Vl1!4#;9+!p}S`9 z{PSGcqCjU4`Po1=w>4b3PJ%(!*QIXI+1rEyOp!Oi2sFsK=P@>9^SYW?_EWpFI&-$k zyHcr<<#YXpAZv~hyRzj_Yc-}J?pz}O$reytQ2U|Bm-|COoC*{dw%4c z{Y3Cw>zG0@1Yb0^2-v>ihd@Tl8k*8F4C4)A2?lh5!kmaLH$Mue9k3fp;V4S3y=CCk zGR&@0dQ>QnwNB3;mN`R<6TzaN&vie~+B=LelaQHQlkd8IeXk$){XFlx_C6;8Ytao| zd_Uj#=B(NC9b35fAm^Uut+j4wE5!HDY>SE%N51tvXTALsf3LU4ioX&sTm8i3!kV9h z*y_>N2YRT1y$48rw7y~wWBSFz9-!eD2RpGZE{P#G{B3s6t$z2>)~!J8`_*##o>4o2 z_ZDbn#N?VcH-J)i8)e?SgPuK!S>rPi#)Ne%sDAE&ZCr8h@^ijR-8y1{jyn(nUl3c- zEeWCBDX>?dXpg|4ZfJu8cH3-q2S;N{%y-?`3Th8x9jw56Rr_9uNOJ6K-X7-8Th1OL ziPkq~3a0`4xc?bgb8%^w_{n+Sp2I)G1@O;mCw%j7CGVf-2DZJt?>CV@f_vD`c`f-j0ePPPdE)+=@U7qg z2J`F{PtJcQ`V_Wz;_s6acn;fn&l3MgI4=V)V0%V${`vF;w0Cqx{5D9?K)~sDHeh>x z<2@$dvs<_83+y2GAZOs)$KCCp=D`XA)N}K_#rB?_2Hw4S`?>o*V`Qt^M(%hC54cm} zcQAmn#f2Jiw)@FB-H*B#_`2lxka4QIYbWc9gFmp>oV(cfL_hN#?R@j@+0gbo5O>ew zyoWfkW}dhh_;nt2%-+TlTVBUrv5gJ1HHj^*u!ES-PfpvAYI2?>$Q^*aj>gE<_Y1bW zAHX0MIeV$A5All!K(u!1HA!+z#u25t2fTm zX#E5Ycn+a$TjXL2UorAcu^Y`zkGro0K<`W?H zGuFY#>7U?TfqEY7W* zXYK;Ho9C-~XKu0DiJbk-tvP$E?c*J)2T(Gnf7*H(ulJr@$F@hHjk}Y5o$Z|J#a!e2 zIn6hqK9K9cy%S$;o&Dsvv!cJo7QX{WYUM{J+x zeCGvPTx0v)xXJsdq6G+g=*>mT0Y;z4QR$y%ZU7y(JJH&vyIp1CNFSeL@c^8+H zt384__=@p^HZZQQK0gq@#&&KJKWd}xqb>L9&fKjb=IqfD8+|2b*W~K1%nLZ?75nJN zo%xwsH@iMN#MG-CH956C+`&2aP2zP2_$@Zu7=Gdp?CU#i)}r+7aUiCxw$3?2ovr+u zLoSHbyMk`H_1DC&9I?H|e%^zAf^IG*&u63Wu5v<~cd%_SdpXm%HSz{};CFPQjmt5n zpZM-?-nf0eBmGFU{q@m7Owq3V#y^s8Ib-%!)|~I4Jx=fvEpP9UF?Wz}uZ9lt1G~d_ zkD(9ty*ixQ#dFCaXWuobF^3$iiPhP`it$jM7$N8s!$;{JWde)d^` z=XfW01~}Wl=a~1;1!rOvM|Ze`o(*lEBN*6?oZ8Rf0s9)PNgRK_)BPN0V)nR#lhkzZ z9Ci?Qo;U&P%)j_a^d)(a)V-ISe=i(p=dQp%XE%8I&)%8;eKpV6V(#Lfb63B^Zi7broxu1N{>6q) zoiW88&OQO>DMxk8=j2(f>l-{*k~^T?Wnf?2nfL9ivEdoCai66%_kee+9^h$Eed*!( zp94<=@5*OaeaIz{Z{3cUJjr_aXV7Ern|EfP{qMy-8~b`6SMW~2XLB2TEcI%47}THd z%$e5M$330}?(gT@yK}enOH@EL$UG0swXNh^>JnPgA zzv|E}*c0viz)r9NYgS-h+jyV{P)cW+)i<%tt2-F_G4Id#aht1N@lX36=-AMVX_Hs3 zu;q*DhqmqlWO85V1CXy?(G7I$GiadWS3KrBILu?qmuqi1#>G4Jd9L;Qq)ol&y#r?l zyae7^L)*`|G6I{p_iyipZ{51Z^{p`<*gh|%_}FrfXyUN4S2b+{=5De~oXh?orKI z0&T8h_D^QRTZx$);9z{S?Q5=!Y;!;K74by-u4haHx^1@nkrU3S=Sb}8&g}2Z5afIZ z+~DJDPC-K#z7}1tgUlU>+BWiyMy;@j|9d%$AvNAZe9Hz;&<+!Z?75T#r}M% zCTE@T>eJ?#8~z~Yp3ZW=(wN42Ox|h^`?STb(MPZnQ)08-p(g=+wfTXrbFy!pk96#D zv>eZO0FkrT74<8&2;v>w-f%*@M*@Ghsq6DbAG!YidI9b4Fu51xp96kx3v29qLC){l zR|CJV6JP!ruKZWSYv3NXJ@y?AaQ)|<_b%r=&$oZ&9dVSD~-^6qd1 zF9uimK~AoHU)b#MJmlQrz369w+)muj+HGLX87Pqx?r)vC6H_wa0_SLt&Ci@WS=$6P zE82P`V&(^R7ocWon+F|?&9>L!k{GxK4V=`(<=Izm&Bi{G^FG#SC5dmn(t{t`Ikj

1=-3gB1iXLB{8a!I8P~;T88K2^`^j;rr6?|YWuhg$ncL5DdBtA2{>H{M6`@2A0=R}pY#c|HpL+yrm`?S<^s z%}860kI46kz}Nam+_0Eva^Qq1=>0oD=Pdq=rm!eV9_Ybc0hgJzdL7&ySqyjjmaCu> z{{T%W@sq-smB`!Z6}?+f&~-N_)I0f-gR{~}V%BurkvD*(wmnC7G0M!Empat! zd+CgdKp3iaxPgpP#f)P>f?tjN5#eABZ{xF~ zY`mu!JEt0G<2wM<`v!(@nKRE&s)0BMUSkF}sSpJvN22{Yk920bVXxaZ@JeJl`Z~Cj z7gHn2(;5>uQ-2%D?915)%ZgRWoq82s!(bQ@|DaCXml+fJQ3Dd_bRK4Su|4Ju9Xea4 zA9z<7vbyLa5`7(w7V=h+{&O>kj(GxjIIy4Lb^vLg9ShDK#h`1ELybT2f1KdP~=6qEa@x;He1}gmK|;^rSD6xcO2@K(_?`5TH*7|r zLk=$w>w`FT4=gbiS{J69f`30R#J(A|c=oAcLZ^Hn?mfRz{)TDDE_Wrg^VQp9_%fbO0S*xe~g3!a!smR*(QLCcr_%^&jprw<#Qmxnah><5wkIpjy*93F3C zkZk#Tir0BoiHL+xk>ta-$yYXC;2rBk>=w#kC5T0KS`~p>Zn9mfkNiG8 zo%d}<2=_tmMwpn9KqUQNQ}ai2V8;D?2%VJxv6-S^a&s;@JsM6f7ukd7kBumMnGk4( zu0@UEeB@@I4eQ5UgH{ihqJ|C4#QMNBB5bx9ee#z>4-cB47g6G9;>%BT>q8`)bFGFA z?!WQoH(|8FR~f2_lgBhv2$A7L0b*^_!BsgYOI^4iMsvSB!y9kc5GmP0a^RUiS>#zw zcFkOeNS@B*%LvDa*Okq=wDpX7b(~ z%ZK;J4CoaHdx_3cjFxE(kgnozmSrYMcjHNz#%`YXZtjF(vWGUWTS*2aSxg!^saCL{jJ6W$C6aUzl|tP{NhsFx!`C19_?z>B8&Nk zu|c97&F>>gLZ(u%-n%7mE5VCuc`k)gJj+pgNhy=(b((dt|K#YpnBe#Nx6!YxVf;&d zCV9AE65@Wu(YrO+y|i5-UT+{p8b3ZkBfaiW)b^5?l2&Wge#_^1@9`iVU4II;sQdn--gU*;VlGmyih?hbjFSm6* zv9>Hk+V2a|3?l=IQyLF9Q~rXRmKjMBIfUS(7MT)W#2k(rqNus?R6?WQmooR_HiRw~4^kGI#h7~~k{N1$NlNq;h?rk4(ui^a zN$+W_$0U)N&fWsJG$tSiy)u-2)2>POcT6CC%e?#MxMkr`{LGo59}C z_v2umx90#)aDq#oJSZaRG(-50Nn-)Xo=mvsEQj6au6VL9>>!rhCoQfC4= z^;U<}uQaClR#1fRP%KaB%KwK_x$P>> z=9$x=NTU_|y;LRJmGen<&lK*NpNms-1$pg8UF3-87M`*~2=DqxC9l5uIdj;i1i#2; zaR0$rq~oJYgxu5r_x|fY?;pOi#dkr|&l@Io@zHr72hgL*DP+EVjyWZw%^A<-+j zPNKr(NoB*utmYXGIUSnPIwo3gL>xRhIX8^1dUN6 z`2F|;x(6@t6-!HyJT(WkOe^P}rAv|V!I${C=Qo`CasXYo%f(;*_JMfmPUyU*Kwbx5 zK}RJ<&9DA#M48$*q0dbkJ|GRW-~S$B6(cB6okK~CZlaIpY=)ZGB`DCArcFX`Lqhr` za7}$ey&3U>&}+-!k@0ymCRY?S%qgtRDpjM$8a9K;`ZOqh4zy|H46rVlPPZ9%qwcY9 zpdD*Xr|B#K;f<=yuKqu0M!E?Nx~g=U_jkA&ZZJ72cR`{9*mxt4fV&;sPxMc za3t|O6snlxRj_gOdFn(BJ%6)LM55 z9laHYgclegXpg1*x9@_Jf0n?p3|Bb!Eek1}YC>17s~KTIQ92iwf{OoiP``W?@={Yk zE9edkj4VXySrlqqH;w}g%4o;K8IXhPV5DLLtlX0fL)G;lY<&yglG4X*H)ZLTN9X8x z`VOQ<&I8k>!F1r@9{S(UTzai!4UBZu!{2RL@b`HNepr7OpXtQZ^T}G&lGgzr=H8%= z6%V6>JGaA|b`3IQaS?9j>%+^ALNvKuk2u6?K~V4}B3CW~dY8Px(?$$7wL5|HnP@oU zrc1W|&c`=Hq?iY%vvAYmAbdG-CUP|RiPJK_laDhZiDHa5anSUjw<)UAYmakanSLaW z-I9kbq?L$AKI{EGlmUX=DkKc9^!2UbFkN{O>hZ3rDUqvS$PFJ9vm=gZjf7FrVGW!c zjj2db6T*A!WaLv?UV>*W#O? zFh&uypHy+4b1I;ZUz)y?{}%S8C_s~{J)Hci#VY6LQF*$9@an}ltZ3T}m&z%kP_cu0 zk-3u~qjpqj&<$oE?qGj7ML)wG5IScX#7rxq)clsy_S|&x$7vHNRH?#?<8q*EWekSj z%PDv~o3_2M6i(mNLAMSMfykvYfu87X;|6oj6!O=xbD{F%xiYP8mX8@j1HG#?>mRUSav1$x_%eM59h&q z-4%=qRS5rb${6ZI6qx*# zM7#ua5q&?O9y#?6F7H>OUk`4khfb=~`p@E_C|;BPpZW8D-#<01hrQ`m8g;#0 z3NapQ=$ggfz@W1RzI1zmoXV3CBgT%gM_ybJj3&loCaPb7fF)YcB3_g9M3op>duZ6&?!2^Jr`TgRHjnQmwn#c zcHP_5TjyXj=JksjuTTJwtxhz^7C{KjI;S1@ag%QuHs;^K96oXiX=Fb}wkZP8*l$Wn zM1|0D>~oH$1;Bx98C0>n%>2;lANWk2Gmfb&<=P7S(MdAX>9y_WP;W9noho|?s7W6X zI_u5oPOhf5ddx>%KCjV|?c>NtYCY4neTo_WxS1nSKfAhWriuhd)+cECcri({EL1K?T{3-Z}e)1193Sr@Bi$OS( z7AwzeJg7z&+{of6{&b*pwfb=NS%OgD6Y!mBZ`PF_NO1-ex$|QEaX`8V?V-4kmOuwd zu%-zryN$v9N*`+b7l`()G(ZJ@w_wnpj~c<{%!Y%~^xaj{X!YDRu(JItXiL^o)g>LA zyKDW?>mDJb({PL9^Y9-!s5gg{dl<1d%xm#N_Fm)Y?k&ukB|FScDL678;=fZvnzs;N z${^?R)ML(FjWo*r)kbg&Tn8SZtXC`WFV(-tj{7AspAlYV&G^2$Soi&D9P)p)0?GgI zWcXz)ko-AI*qIT_d2(Wy$`4yhshxJAl*+b3fIpk>8a*aso*`X*v|(bs33ro8Wqu( zB{$(gRWc6KZ{w(FA7`?94kF*KlhopnLGmG~0xz8{O-Fp@Lr(gI)RlRoREg1NV*81n@akMctQ2m?gAOA!7?{w$N*b_4xdy}d@$JSfe3N5N$i7N6Vs6QK1ncpHg zsOFVtZBCap758QbDzMUlsUQx0_}y3X-v1|h=em@%4WD3celkS;Cp9Sg&r@WW*a#}) zsyJgJ9#%#tqqx=9(C@wyh3ghl9nH2VyUw2SoLmTBBw~?k+Fwc|U_K|NBc5~7$PZd> zyu~A8W7ML*hRAhM0dw}zVoI4$1g?$FWSy1vu&r2-x9z?F>2)|l?jDq)Yq~TcyiOl| z{9uagqME4DzscO4SF|BP&k-JOXY+I&eDwQ)2q>HwMiU%1C;Y1unf(#PuR_A;9+TDZ zMnfFJwoPZb&y7e$Cz~2Hya4k1_d`N?0{Wl%&wt-PS}pHb?zu5f%Grchw9t)-+7?aT zegA))y?HcMVH^LgjG4y>MVXbPIQzcug9Z&siA1H+q?wfFG)ZKrkWiF4V?yNY``W0` zAVMV>iUvtaX(p+6KhJufwSK?%&-bi#*8ay@&a|$5U)S~hem|A4x27DH9_^+xI=yho z&)rzv=o$Iq(n0eB#$=su@wpu0NB<)};MKt7c~6*@}l~bJbCD$T<(D++PQcj^|;sYZksJb(t}U zDnaz~FxIF@6CBtTie&?|$Y-yA*!plS_WRRDpKRX*g_eKWdn+QLe#t`g&r+Ohsfr_# zdwAySk(1cz-YotMYd|Wz#*n+{EuW=~QH|;|(YvqnOH882mO9`j@^yOv$!uj=$0xF& zGqakO$iAccXViqh3~rLH(qX9FqtCT*mq2d|<__Jx&(409M88hnjeCDC1ec0=M8~cT zTZ9)<@Z#4!fiI9mu@~{z-A`OJG|6Gtjd)gSIlQ&`0*hXna~IFwB>O*~Av+zuBO~vN zQ0hGy_J%KjW&CVGiLw;ZPU1H8+VAEUA-U1HAbKlFXZ6#c>&|!BtPmpbu_cCVMZc3fDj97{``8YIE%zjw!vs z7^>*vZG-2Th`+(2vEBnHHi9E}jQ?QSPmAf?`>e2-Ay&)6& z<>q6TZ_m(~v2_d;S0FV<%c;IfILs*41a&zjSoYeD*v9bLqRdJz{>pl8f8`E5*e}l+ zPaIDs#>;?m|7K`7JxHrUbYbPQU}|4hM!%0ZM04mJRH^JjufI-(6Pv5B>&NG0-K~6% z&fLe{$3DnK{v`by?Fe!z-LT0nl-U~=&)Ocy$HT@hxbKt{>=;a^+2)N{cC-hTeJ_Gy zg^w7eEhjVQv-obbH#Smog>B7?=y8`wI{C(WoLBV+o7n1eYkMbh_Pzxq;Kv!77iUFN zdrzX*y@yGXR1S%iOM`D`^QhxPOU^FMft0&iiJa$)@mYhX;Iz_tv^Xal#|9t&&;0rL z|67BO*DU%>)95_1@V*48hg4E`K#fT0I|$p0UeJNJ1$3J05B$O?kn4P10LkyKgZacI zSpWA_>T8q>rA<-XT@xXfo1qIDRvaAf=mY7ZO1QeZj=tEIPwm};>G)0!D6Wkp2C?B- zGQ<B0q&I?OYVK8%E||0Fr{vE}f&dNL;+70Kr$$I!?tu@HGLnRWzDfaisF-0x*q5{7DhNpOFXkS&Cp@{~H1=&8hZxW87LM3+J}z;h+Ef;f$#ba3T7->V!(qyW5^$R`NHX1xhCBz@x4;lUlnxxC2%PImC1Eb0DjaXPA z!_NkEmARL7rrg6BBfwr{2}6r=A^yHNo)uEbCJyi#^MjdaW1R)hv+*aDQ6l)f@rxN znKxvTZxE16(RBFUKO$H37l*D{3x8Ih1Rcf4U=WoJx(T0QsCO)vADKiZUN}L@t9W*2 zvA8HuaVu$c8VQOQ2^_Xd$Endfh-8*8>>Pg&^7r3>b9Fa~p3(z6dH*DG_;3c@uJ9Zl z1log@*L*Cey;>;ys|}x>tp%sT#Yl;sGF4o;AH-@!IA@yGlaOkn8aM54|o>gx_20Dzm zS%Veyf=EEuPA!1^uw6{=yoXdcy9fM(l;NAi3m8(@;|81T==houoY~BK@U^cCZ=07- z%cJFlm9nzjw-@H5abXEnmzLrJy3YZ*ugBfLst4hJjG<<0DG2zTz_RUsMQuLKG~22X zY(LwBYn>jqCAJvaulR9_KJuKzW(5-bI}mjxC~z|Sy{Yg>4b?OCg0-*3;MKD+u&3}Z zomk~gZ#o1Mr=F!m(m@FQpZ1fHgRyAw`DsM4>^wT@FiKuRdveHn+ET!)*JB{(I&sZcwUg2loK)a&w0EL%xSB%c>4_MYx=_(&C^^f{{_PcjS3&|f}xsk75bJf&4us2Pw# zMlbAwotpnZcoxBr_rI}Hk|NdI`0s!9=RY;*xc<&-Q1R?N&EQVc_aP79vv|HJDf5Y_ z@27~m9G?wK?$|?I(IY$|PZM*ZN*wxn1g@BQj22a1Muq$0!7d`2XhglCeQW;U8t)7m z)v`cjV>Cn5BddXpXN1z^k0a@t6~%a$iVkhD8HbaaquKWN$?T;dEi&uoI=Hd;EPXi` z#P2y4W6!ZYxXA52c8*w&Z?0lR=TyzXLOz-CpSGCx#$O@tx))G`1bO;&*Gb0X$X#k= z*FqO9SUpS5pR4mAc!>B7qczqS5eo$17P?<6R)e1B-JmSVDzVGN=9E~(}$$Ejw%WI z!_tSHZWM=g`=`*oQzKDAwgu)+ufZt^#();pk%>}wNSApIUNrL*RvDZm8sBxCGUW?s z$>sMnPPvteFPH$^|BQzb$)WtQFdmX?m!r9V3t3C)RFFG6f$Zw~#Ma*rARDZ<;+sEX z*|)PN5ee^|xPN&P&PfuGnrpA<-bZiHhbvxeV5ho>?Zc~~UaoRhw|pZj!Ou9xUwy#%4W#4l^dWlp+$1vggDUA=`V{|K zr-km!S&z3ItYST<*s!luh48k=h(=zDXJoD=h^kB-sr}|Co@xIJn}|4UJzavImwZ6m zHr=M_X{2aY35`B$gtgRcP}?UXQghRrRC{bEFkhW)UmZv2fh|yI8%VET zyU%jE2hisC|Jd*io0!vQwIE{gIlS=eO*CL=|Uy5yBb7Cvz44!hz=EY-TWADzJ& z%%#-Dbau%eF2FYts@*H8mDe58!Ke^xC7yWk`v|OHM}f}YO$5L1VrBAfp?#TuXhMQB z8IjvV)4%8br~mqI?fK7D?cGTHby6H@+1Nx@jw{8w>S?r&_p_#ksS)w;TcoaSDUt28 z!`8(;@5e1KOA5i}y#~E?s-yMWpkDfl5Oy+Aw5;zhtdI<=@5O$Zt9B)nglS z(k-2)N6i4o?G|LuqtRq>v;kar6A4R)YG~C6A5e?a!@sr~8^V z;Uy==kj9D`bkQLLa^m^}I(63rQ2sCl{^d`G?vE6|H2MO0^Up)#>l18a=?Yw2IfxtI zS>Y_-EA+781Nx)F1wSIGutVttmoU2!KD7+sq6e3VhF>w2`V)s2%&(@d4q~FK+6V`` z&48GA37R~{ge+WCfDHmCf?i`6nsCSier73hHEp^uQ2hfpY+H&G{2Hv`g#ecP|U!joM5`gWD`BMOt*Bv zq?3*Bx{5Dr2|+#OFJzq829azgmV9I?nCn2eUd;tGvF9GG-3#nc_6bP4DhrD%df?@> zFXU`{Cv5w|v#XALrbNCDo%wQ&{))LncUZ`R?xqatyCDcrzZO^Za5oVTvmo1j%izMH zQaId`1xn7m->_yXtGxUtST9=zRf`WpV!k-1`)(IEHlw%HJM{&fcWom+wP2E9q~<|( zy4!2fOlt|Q?nneIUB3lO`@Th`8(x<73M{F~87Xk`eF=)ecVX;|Sv=Ea73}O)#96b? zk)uj1-Zy7CI7$qW04*c-YHjunWR^=y9h*U|Z939FdK&Kt6C-`|bI{XQTI6BQ5Z=`} z1)o{GflPcWgMSVEK#N-R$*a~ z5sWWYg+Ee*)6NZ>kpQJMlyW+XEq2Xgv-n1pP@ckV@Ws>MRwuc3z=uDRl?jZ10h z;#fROvW*@2kk1TVV+W7&J;B>go1M8W0&cl}f|c*fad2%T?#PnBUH%vGcKLB!@0Qy% z|L`f$mo`lf13jCxl3afKkyX1Y4*8B_ zAlvNqyEB3#gKcB5a@0jh`hr zkoT)1VbXFbzMs&6(7#4VpK zLbA4DH{Va4V5c_{^jlVyo#<1I@d$f7+GC2)Qz8a=7H1NP_Pe;gXC5i~8Ag={K9D(I zE)bJPL-bafIbI#yfG4_Yu{#rWLGnr){iD?Y?n9xp;etNTj_xT<`PmG)>z`AtT4yeB za4LRWUP>D_4WKejirbttxdMeexaaAK>#v@p_w$VTJas+ck(hZxp`eR!5wF>zj{o$O zr%}6ZKlw9m2Yph$0Y@dJ6NIObV&*l0$*17hkzPu;PWq|3mS#W5gt7IrNZQhKP@Jj_ zueMaNUG`es=kgit@n}GMyN$`2b$+m5?=T` za}suaFHU6EfN!8NjaZa}myea?Uc4vV!2^r9t_x1s+~_M7_*}tTiq+xcn`o+bA@cDnYahA$)*oi~QoX-)^ zlvV33^xwZk3 z#&$iXtvwun&7BG|Dy?*BhraONpXI_CKC;4^Z#vvY)iQRAy$#2E>H3C0WOTq+}p+n(Vv@$+f zq(&kY$HOh+KiX0_r}I3Ydo>oUmF5u2Giu8}?Ihdg8j&M$UvQ)SLNaIH zF`jiJ1>YwvLhnY+6N(%!2+tn-MZP_Kf%nIqAXajPxHg26jqcg_X-YjlG~i99_QnvW zk$tS*XJu~d0aGs6h|h8r?4brGayZ*JnoNr^hUjr2Wc!`XV9aN&%t^jNHeGy8-Y-56 zjn`)quE0)cah&JVUf)N3e1}jV-~W_mdLgx%<-+h+ro!vrlkwnXEAEg>C|n>iT&C1E zbT|An(^a5OYDfFg)AhkvXHy$?-+Pgqu__{MXXS)R<0+~9`vVU-1pL!hcwAg1FEa)P(1@cGG9#kx(^y0R+5xjU`47(qN^t@c2m^c1m;= zdfI6cBacAtL&F$SJwBFGiHSs3=7IRnO>HdVxt&ks4Y*_YI9XJ=i+G>VA&(qmNt54U z;;9fsV$Qn=b%Tx*WgT~(f7F2knPo!z2R$V9^mMX4XCmxzng|_Eb=W0^ujhL(T<_y| z7`&IE{j4TkelMT&7o8*at+L#;Z4ZgG$8FT()Pw(4DdFc0vGmTueW2kWCG5QJD131< z8F}zKnR!|Ju*Bb1 zXTy6wr}K$VBs_lHi!ZFtLXl63z%qIa=MaCHxWCZk_QdZY8ZIk^YjS{OI9w&B{7$BS zNP`nMiGvq54xA@*lG=iuaHvk6p4u`3*Ur1n4z6~^k80=ed@K`TxX~RX%a#+Fc5_f& zFQjXF?~|*J>xCmHsME!&Psp>A%Y})NrbIpdCi$2eNf*m~qZ1Vu zCxr$3R|+kgm#_<@XVHx&56GmfO+xDfTEy^?6&EmkfRtQpq?Z~d!|(SGXqNpGI-+h2 zq~B})-~31HKmFJ5#!^z^SA&u6IwbyREI2I`Q9p$`n0>oK^kK3dOtlS$18%!*#$SkNR<%PPAD; z+HzA#%gSy#u1kp&{oY8Xscc|vb?fPc`3I5p8yWCt@3Vijjp6aJ3lRHm9wp~v@s~kk z2&q3pl*KAZ4(wsS$|f+oiXue$kN7iZOe_`@C=h193tW963z{8D_}bUa#CBdUy}cxn z-0|)u?3t&uNwynl=)|$o)ncGl*UOIez71bLIn%KYuW?k@bW*2$j(q$vh7=zT7u9~U z2A`FO$@}~PFp0SD2xq!FG?EK+bloA~V;lA*X)aqRYC!5XFgkWKAil^tMq9t{qt8es`thiM&wSD)?KM}he5D9) zopTQE`00Y{7#+G(bcg0dSb%+h3SG}%=O3^o)-QgE-hH!%OY4HbFzq#r2wzC^Nhatn zjo>-4;kX_pAr0@NkoUbDoMhu*Vsk%reEE?6u2N-<{#DZzZp+|8<2G1jSr55g3BYM8 zL+g|%EEQl&PH(HDUs5G$LaQpXF;xq|^BlHRYQo>$RucW+;&AKp9G>08Go9K-;Pnxg z@CL&Oa^<%%J#T)3^^F@4mE6o`AJ5Q*!jC7xK;|r&dr3q>ip$AF2Me&s5~t_1dTB{E z!DAn^Fiq!LXwJI~`piNbdIClWL!IXkYm-78{!gFF6B_cq^(%0J&r=vFo`pvq>t)?O zFMycBcl1WlHqh#CV(pxrU~;&-Xh>_EV8L|WWA@?_+OGeZ#ztSp5&jR^CUFULcBBP1 zwLJ$mE3cAc8c!&(=JP=nB)LhNZNzvhz*QM{*t2^k=km#bn zlfo&S96wucdtM8*qFcCbhqdTc&?Q>E*o#c?*#Ot(wf(RD{I~zw))2_^-rcG9mLMWX zy-qSFza?i(tWa^{6*6AU3?H8N69sQBfUIu?bY`&v4A{p&Oll`SxfhWOQ&Ygx*%Zr` zE`)GvL`>~Zv(axxV)p2BtSgWaIz)TGP=bJa{7I6Icqt33ALgLtYGt@Za%@aJJ380_u02yAS^@9zl=mrQd}|ZFlblb>$}f|< zXQas^gI82fX#_cJx{CkK#>1`;IpoEEJbLjygqfNIC+&05WiNeNp`rt$`_wsw8`Y5Q z)d%I@C((U7@(9Z!5`6Crm_7PQJKUbolC&JS=l-6qV`WGRj%9i^O}W>+p1N269Mser z!lgCg@U6U(_$@j?mL|>RxpaWH58e@NR=viI{Stu-WX;LV$mi5F{UApbwAf@ro|$>r zmn~Woj^1eWlgfkVP;h-7HJE&yQMO8=SB`APjH$g)Wp)<6=R|~f!YEM+yL8j=D2`eE4NI=c#tlnb$+WNP!d~8^J}z_>>ial}*jbs0 za_`sEE}ud+D0ewLL-w%rcQ6P}e-r$qeq`eF3hJ|j_gc|YT=weYT+@h^;C=EYSYCC4 znW@?IQ^*6f{$>KJ(|=VIJl+GJ8azzTS(b8bch_^%Usgf-;aMaw?m13Py@BNv=96P9 z3dy>t6U6_sD*YrnM4Y26$;^YUF#7Z*NG-Zbw)Y<)FU%u^E=FHTli~xiX3-gPVU8y} zm0Ctew#$&u6BePP%3nydGN1A0`HvWOOk(BEIndW3cSx1^JTlgN49RcZNi5{-Nj;*Z z!Ed_ohV^_bNGd|JYIMkzsUx`IiT1R9`!~TpV^cC}&MK<^`mxB%cncogCPNN4O0pY+ z3Dh|KKx%*H)Al7VDf3JNIjxrg6HzNZl2t&r3e|*eH(JS(DR;^Ak;jK zZwekIX{;3rp)UgMsmywD`fkohu4aQiw?+N|1sQwxf?ffM!VLD&lEF!fV#rY1acaLl z7C*o>(Uc&1x+N7|p4duL=jUSi2V#U9FcN9&^e-P4D(K9%))Gh4&$aYYWX{TYP-|6lf z$!za~^VoRs4{hT!D&&@5Wj9YXAWlc836HP)CgPq;;(^NZ#OcI#^zf=8scu|`_bu^) z`ho`bz?M+j;yqQk?4gY?%FRxA?W#Ff-(tjlkP~ou()!%33^!btw3<3-?Zi(&YXb1)ysd!a<{rV7$l!22Q&m zRCW<_0dKIx<#{6c?0zhHhma`;8?cC6rv>{)!^}xe&}~%)ll?BxWnabV^-sYxqQV)v zzVZBz&+D;d$3@s2PH@>X1KQ}N2wjG&Xrbx^TxKId-hWQOU#l~aPE8BWwVFWU2cOW& z`i0B`jiWfX_&O1K~QgXXuT(j?lv6}t76MB9UAfV^4&36ZbhbX!1#*%mfVf`;-hk@yn9@M}RV5y3Qmw&a@-yaz zcsrJ!Gn1GcT8}r0Qt;gCt@w@JMsPN_g4(JoJom~a5d6-Eu+x6v@m36%r{2Ug{xt4x zn@YzY^MNJ)I`rJyt?1EVYfwwt4#L`AuqgP9TQ0cKrVG71&ulkt@;^tv9NCWkei=)> zHFW5OkaQf|Fvzx^H>Cl$canffd!Y6LBERc5qDxbCxp{vLxz^nMlq++G-uL`xa7YLp zF4p+s9SQ96Rgq5BUqnVdI|w%)y5M6oZeoS&$yDf`O!tp-=Dena!|Ystn692kr8cI~ zkK+_TaNC=DCjG(b`;L=);0Cvw#XvEA1Fc#j$^APX44b1jVCmx1V0-EQs^sEqG?`wj;R%sA7I~PR#u6X)O zjJf8QJkxDCcHI|`f2>f&CseGF^7uXUxl|(9f9s@**&p$WA5Qerx9(W)Vg{ zk=T~Ykj>wA;Vrj+(OGY;$am!@RDGZX6gNGk?jcjC5-9^#IqO!8P|MP{lP*>C%%6$&dX*c-B;BS{Y&?+_Az`C?51u_-pB9GD4d749>hq zhQ+7S-M1Rqdp|OmY0)O+s?Ke8BgbN!wbT)rTUTMCC+|}Qhk-p-T{_YU<{1_`V?{X5ZkPanP z>1z?=bA=U4>!x-S(x~7s;(h0@>BVRfowXV8>D=w?yzF{-T(lJY?V_m6`>V{={hHjl z1xMJkQ}oDM^-A3CBnbz-Hp7;zeQ;SKn9jLvMed|}(5Xud@i#RsEWI}cpR)Y@KmFl< z>rdYUX)ccUxa(~xCq5>}Nc>R^VfMamdL>1MOycvW+;^qZqDReC^=mrG>5su)nuh4& z%dNCwo+X;MsRk^oB;YEqaq_Ipp?@TP;XmC1+Gmz13LQRAAFptrV`d&gVxF2XO6d{Z z#h-CTI&Y@4<3&`_XD-Z>n@?0;xDl!6yyjzx1u0t`N2Xu zTl0N3$3KX!u$lqGTN;s*bTv8BGmSiM9K-|Hjo|Uxu^{z%0wlE`!Vxc4)A~;l%ueez z*qRd$QkN|7py6K{RBS>0SK70YPhQiX8(*Wk4VYdjj-lGl$MK*yKXu=rgb!zJq8q&s z-2K)7Et0w*=sX1(i&z-cD`LNPB;Yp>Uy}080^)sMlazll#xwbxS~V&`8u{G4ubJ9Z zYke}FS#}RIlG?&4Y$q{_Ux>{oYLm&_5qx_FOXdw;B-wM9<<}f z;^rhZsF0ue9A`Ic#6h~gIixqalP!1Niw0yVDn4jNB#nZJX~hiksmv7F4B8=$YBlc0 z3N6g%XAv3a8>sA714_KZ!5L2pjyV%CFYLpF>l~)(l7b-u*ICUt$lMhf1hk!XX&h=PMjU3n1@LJ;<1;;x9#~ z;mwX4SpKU8SSf}Rt-oV%N9H~f;ARNJC-=dc@ILn8yAU{r_TtLtQM?yG4v&?*#OAx? z(`CB%NkOxj@R8PSyj8cAt-h-Zw;eyx_2(1ln$}SGY}#0=zr>bw$;66hwA z=Omde#71kvsiH$1S?OHGhC&eZ-u{mNsm^AO+Xm2|vwZOx<1mQge|O?0?YJ>H2?X7; zWXGtJ;Qaa(8M4U5>R++YD_ovDIbkhqG^rtP0t<1*;WKzor8)7`K1Gk-%fi~{tHJ+a z3Cz0I53=fkXx)}L@}@JDxjHupD$JC~)4d8XSSW$_qP^gdZ!DAw`b~n3*O1Fw>#2U2 z3;dqG5_ue9=(X@G$eCSFvTi@eci-`B>)vb#R1Sx7|61^Bd`XtC_=0=p?Z6{aX6DOMjU z#50s{lZr>{NQbv1dG4-~Y3i6e# zCkJWJJU(Z5+3;J@j4dTFE+dgdY5T$ZJNHO>^kOnA?JxVwS6UD>WDMpRUU2`^Txi+5 zpS5{XO7{5J66HHypkr&t7XSQ6)|r(tJ|a_eerYp4uzP^4${Ph2y^0~b>6*ZJ zJ@}57WgH>zhASBT2P1@|45ksy{t9IOsgvmaN+Q4RhT_fNx3cn9;po6?Mr3unT6BH! z3EY!z#imKMLjQpaaBa>E_+$Ed@=Q4itCOZ$Y+ z(rSrl-9x(isS5tv+)e!Z8>lUJ918YzL*{uGvgRF2S9T_%#M9r1|LXZ9nI0$pL(jqN zC$E{5-9|n<`pCBIQ=(x<@3HlkmcrNTWratMw&1wsqsR%f89XOanthlR1K-;8;d|5{ zkUP;yZPIJ0_o7zzgVrjcmmWc_iJ+g<~Z|Y@hZMWhxTHEmI zY$eVOsYB=bU3lZScyjQC7l}}OhI6)s;$55eu$hUEk-u!blB)Au$GEXwG+>t< zJekr>W1a;-U~HA>;hT%1;f`VUY<@FQ+?hri4?M!l4Dt%9k0BZQ|BO zpXXLrWV7b?HsSG#&OFaM4!_;4L{sg@lsfIQ2Yze`vKed9vgI6k$@ggLix&!CJ?zHU z7iloB<;0Nx!!u-QQ6j0SJBe3487)*VKP$X)b_+S|d4rtc{*uk>CJLXAyd{|L8HV@k zCDIWpi$L$oJ^G_fi`)8a67)5yKJ{7W_z7#7$&$H#>rfDIq zs^vMrqkV+#c{$|&(Eo_JJ})K(OY+Ebe7IW9TIG%^$h__V&-2KuEg7+$P zbY3K=U;7NN^G=Y?M4lVEjQ8|rBtm<|LYT3)gYJtLu@f3z(yi)!(0%$G(5=(4!1XWo zR=)`MJutV}^B~t#aFP}bdlGIt$147w2Kocb;qi~@RNK}V@2q}L_y6WVN1Q)zT%p)~ zlQceiVl`Q^D;MXne`!`y1-%hDlfK#CCt90k&(^;EM++Rj(T)#>!f*Lu)FM+&XgF>n z*BB{)KW>Xrc!dHhJ7F%ZYK^7A>B>ZXY#+RTH<>Flh$cx}4M8?Rfv=4*Gj{EyYnd`V~LDKr{aO3gYuwQNrR@t6IV)Q4$2B|;Pb}zuU1ZfEI z=it58N>1TH5@+hV0phX^h3P79&<0s~?wZMT=(LmIwDrev6JI8CcGuG2%$M)*m-kp~ z_8g+0V}kI|19jL_wH2f{T5u9enn2pn3Eo>6vUT?hAXDTGsV}EdiED9qKiiETe3Kx{ zoHm2pZd1PhJVL8Ommt?GdhqUJFq;1G4leq+oD|yKq_dWN#~as#(N}wusPF6yFqARk zWS)(psi!Z~PubJpO4$?qd+-o@D0VGouE5H9dm?@v-J#O1!)Z?S*!RzbP%rXtM%QL7-FvZ`z(nO9=!uSl>G4O$Y zgqxn>V4$UiuiuJ+j!r`uUo%M5{yikg!Pnrx6bc5rM$?(=g>ZEGlv1n2&1AuF9P7IF zGWhk6=WcdSXIpMG(^)O|s9K2=-d#Hu4_Y2!H8iIaMTI!p>=On#<}Ski_QxbuJxKU3 zE|}K{=%7%=x9G^VBJRleyI>#R1~_gkyqnI!@;kn8<+mg3K4J*3>4X2P|8jn(4e}%p z3l>^)HyhTFn}Q%-yP(X4*~L=B+o#xnyVc<@oyksY9Enz|J!M1>bJ8)U9SOFVn%dLbGw- z#V$59<1y?|KMr9F8BQip0!C$o(mdTGWRhqj{`#f{*O`Z6tMBJlj9gx7o zGk245VWFt;NGk5WsYO+*&#(sAjA@+?-8Y`nwLu)e&!T8${wU>L2UO8SD4Xt&@W^#*A z!n7*ZdSDyguhB`rKRp9}7f(@*j(9Au>j)tRGvPYK!r9OU7@n;NZ)Ma;NNFgmU*buo z>g1yB7k=O(k6GN36SB1T^%{t>e}Ywqt9c(k@3X#|0UrNqanLavq9b#kSg)-|A=jMA z#pps}l+5q(JeSgj`*&gc=}2^9vI4HToPuv{(xGHs8x7OtGY}Sg5|<^x~GE@ZJY5_er+VIH4`JE%^UDiK8HgSPT>m^7qQ|W zf<)qa2x8jB$oY~yx?{X5>;Fp|Cul6k=#v&!Xy!FI9THr-Q6T%X?<|#B9dyHyH6;R1 zJCK|c1^Q8Lbi~p9H10zr-KIMk$;rgvueX~- zY=9AOQ>m2VFy2#EYQ21uIh9zj7GDY;Ko34V7JTywr(q?&DC*Klc3tl&$h4_t3s-+D zoz|W}-F!Z>p4K*KSS5gL)H{Qw6fLLAMlQj@2i%Fes*oIBmqy>>5=$@c|u&*K1H6YK#hv0pS31_1@4rs*`qb-TrI)1ov9l-- zk4{NomB$AV9XBabTR9e$HyJ{vBJZ!8mn{;xd7$us1~#W=IbQfN2)}+JMKou}QHOxL z*#F}P>art^Y+lZv$6D&af0Yef35|z&!_T01>0vfgxC^RkgQ>sb6`ZvBCFh|LYV)*;A zFRYZk%8vIK36olasCL~f;^`Gb!lRCI?Ft`h?@WLC;(M&1GDHats+hucV}If?U5;Le zJqru!bhr{_OB{P+1D1HYo|JXiACusc_FnkEbyVdZCD3Yqq1^ zJWIv3CmQ=2>L6PwLv)+ZIzGN-Sn%d>5wlFn2wASeHsV7uHXr`p7PKFh5O~X-6X<-I z#puN=TW`O2-^RdpouE)TPw>3zDl;$O80uTZS$8${*u0!cn7mYV!MXAFg6?er0mPpzxOaZ7Jgv# z-S;DDz4L;DdYR0$lCRA9qpCJ36Qyi^Nm>ggjZtBeJMS|j)Br81FR-~UuO@g~JRk^Y zye#-uFA~^D$)om-$u=?4a)JU~&Sru9MVk$hjy8I$2{ZTEeCwwDVKyCYU4rkfEhetgQ`7$*vxd_BE9Zs)sV~R&KF;(xT zAittp)`p|q1s-m3g1KY$85hGGX3x-C{^(^bNFA*$7`LAnZqIlp5Nw>qU?zcyQ2Ho1 zRW_Zm54C1;#Eh(Mp6|0Rc}*CV?-ka%lRIq`*}pbT(anOo6nCbgXthA@-gPFRR)}V( z+(wtx2W+CX8wI)-wHS#fUQFbiVdkuC8)IEwY90GiT(J9KF|+H^Tr~UTH?-S!9Zvb2 z&#d_uTBG-!}l$1ypH$TR0rrVmx801zncb`H0Kx_A2kD;526BVsoZpH_jBd~8ImV3 zGe0J{`^k!FoirAW>PfL_m^DQ((d~iF{CVST{8oLpvHlb(c&=AtlNr}!Q_1sk?Lzbg zLxBZ?=8sj(;>}l)Q(Z0cJTf0Q&K+%Ye2gjcGSV8UFW-t9McL>eoI%NR`_TKoujr?D z5jyErZf&ooFVHuUU?TpM+mt&_5p-lr2~N7rvw2*m!YJ1pq7iRwn5&PKn8hcP7?*W@ z)>|Hj2;P_k3)aq$6$r^5rsu~L)DkzDnXmCmF!jg?=4|97thLq><6>jH;g~L7lQJ5g z6}QC)kDU;_-`6kDeWD-;T(uHC>i0$-;?L2Db|0HXlhOpctd*D#9~zmTh8oBvC;?gc zrZB%YGpIH61QHqN3d|+q1!J5u1T`)sjnDab!I>>j1*?~t zG1eOfng7+E|6HfZgd*XgVP<8OH}krNpdZs5P{6=AWHVw7pGBjFV#KQ$^JEX^VRa$n zl#$7-&z#7V2YwNh9M)qD(h1FH_X5^Ip&yku)uP196ENpok{(W%}6*JBQiRK`L^p6bEJP76EoN? zI3S|GuCN>V8;3i!Fsb_ z0s}11NEmuEvg@1~%aucd{Ev>vPWusaz;+4awreV*sxq2+d$UY1a=|So=}#&1+aiZi z`ZfX09qWR8Y7NoLG3`u%@=`{7nh)a>Hy(XlBZgLv>tXzxe9+p6lc;X;9MosK5^XkX z7mWA)%q+MQz&x0+k@*}xijfl^!QWrp8I$sQCXEC#MZ0ebG>4`#AIFWs=j9%vvKe8> zt@sMZFhPm8+8BwsTKGz01iJ4q0sWJ3Wjdu@ zkeU2JlsRuMy7x`=tb%a<~wtrNjbL;-Ma0H z)N*>6#^WQ9LrWQxLcTNk71x4T-QUJN^gQ!$p%Pkps+MVR zJj8TwmS+kq?+Hfu`UrX|ehB2+Coz(?1Tmjxphd2wsOnH6D!z39am@u$X-)GK5G($PgJaHEGa1(I5&*(m*94QxORf zB|}0~=8_>~_&@)5zwe9QcAe+G&e`|cYp=ET>3K`JZio@DJ@K^N5j+Jf3%DtNo#AlhyoB{z*aa;>P~cP&%- zbna)wnfr5a^l@koKR^#IUBjaHMKE6f2kUz}uzlb|+&|z4pHK1lIb%06($1q-T|Vu9 zy%=x)6(d}%E}kCnmXGD%NH9zg@w@bLGXm3+2V) z{mmR#UrgmMohYtmnlL>^m4mC)IIm8PPnv~c&*rDdSt-x2Bg**Hxu@9seJ_64l(0`h zI(t;c@L-!@y!19jTJ$iy?>~g2>PK?rsKNYy>yLli8OZr4vGiq2ZhT_M-iwuaN@X%$ zZa3wVjY@nYZ4m$T@SvYxjuKrfppL$m$vvnKE(b_KUGYDjrq+WutsTNXxmW4Yw*q=` zBA(u-j>hj(VyxWeh}T&UgdQzo1UX0GzIha8-HSx0qa);c>hZu67JN_Hmenk*urtpa z9%doP+nptB-1%DYTXIP#+P;{6$nK+#uXWVex(%}f649WN3iCAw=)(9~nv6)&HuK~= z8B_UT>1y`7CC3ft75LuNXV}pGK(M)~LX*c%CiNP57!5yz{vO8MXSXb$H15OkGu8OB z`!K}Wu14aRg;-RPz|-Ut*i@2j;-_#7JFdRLy9r;gK7KjOHrZg!cO~S-o}&C?=V@ayYa%HE>FDoY-Kdi+F3b8qgj9?$PJO}KfnH0M-(#_^Zg@XY%_n_pLv z*+m(IrD$N@T?Mq2{-NKOE+e328{8fiVsc6b#(9LnR?!UOf+|pdOPa&ol2Cn5jr-1g z0PlbMAP#hcb*&3D8*Ne2I~SFbIc&9GDJE5d9rviP57lG;;z(hm{u*IUqYkojW8jnF zi!D!1qUH5$8Y1Hk>!cxEFvJ0CZ$H66VK4Sh_C<(AI<%WIkv*>g1N~MaEjtkhP8~t^ zqbF#-{}5v;iZMem02i(XgQg7So$c)iXf)(;)Ci0AzU&v)i6#B*Sp3A8+e0g1)Hf9A zyMpj+N&q^0f56{@UM%FN;(h&6KBleCHZmjFPDK@Btu)vz&c*C`k?88Z1h(b-F(&f` z9eOW=3&&(IY*7eKO}~Klo2|G#!4$FemiVBji)N=v(d7F|WYQ2qM#rzy!uboKc`AZF z?HW$D(qn0=W-O(RxPh@gX&5y>7ZHCKP{AG*obD%w_Q|;@^gD{%N4{ZuRXY|>xQUm) z=OV6q7@zg-!6i8z7&5RHu_un97haN|Zy@}?3OFtkhoJa@_lAygb1XT?T{UbnS$olD`1af4%XepBXNz zG}EV-GeS){D2_v0!n{V>s7V!oj8k3cWg5&a#QO<=62L=lwW7 zH5is@CAfKQI#m9gf?Rgz|KK10Ubhie+>;Y_m~qli6;>VY!*gA(@VOKt9&|;Q_r4y* zUz&XJ`_4ejh?RV+|D@<{{#n|(w3*hqR+H2Gp|HKN1Ti){XplxV)Pqvm`(QFRhAZ+LZC$=1cMlG}`n;=R4sR3n6pnkWpx?<6SSEUi z+Q$kkKQWZ`K8tx+HZ6&j1-E=?VICX~G!sU4m zyyuUh+O*j)P5p#6&u&b&)Mos*g(IXS6!zP7EZt?mPlirolkhCQ`cq15EHfP+x_yym zcNewhX(&`}g2Q&n-Xae*-r8TDxBOYoclA?*?aqJj&U`End$E>gIu6D4wPVmMugvFO z_roviYWketiuvdiDZmy>Az^0>;AFxe9lW z^=Fr1Mr`60#P*{mQ{=F{o$>op7odfK>F9V5w1YNa^N#;S?4VY_Q3_cL3^TR-jLcT#WRCiLU7 zIq^I;KbzI!c5;Z)93J7_j_w6IymEsz=T5dEgYhqfA2gZ5v_t4w^D(FlslvjfQ~zWo`a8hWv8_*+yAQ{k86p5fH$(cFD3lHPs~CDT18 zNJx|HJ~RnN%ouESbnP=*0ONB|cF21FdG(?BfuI>UY`jDqX^#wsynlbrlxf-h&`X{azSf zjVmo4+~iOX>7Q3HCHX4KzDx6+>R$Y*I1^hOFLR|~AF-*sp?JuUAexo+lw1?{(T44> z=?jw$>NK6T7c9o%%+utvMTz{Lmr{&q8!a%HP35b9Vym7r zwtHWpPp-Z~^{T-X-CrJse{CV2YE7oI!UcL+OFNJ0LO7dE>D!~}b5M-XFy^1=Wmh5o zHcmlnay7lOyh+h({|Uw&A4E3KB`LJJ`Y;M?rw&m$Gdf;#ZC`*bAomxbWJ5!`EzgV4U%T>_b<6ZaVC!% z-Ut`}d$@P;Cmi}+gw4t8=yF*HrQjlxvzK7L%kyZMjnwlcW7(M8yfAE zg|g$jAvInPO<|JydSM=r^9sekv7y`MdHxcublO@lFvNY$kRs$qUF2^ zYi=0Fhdr00&g>Kg7Ss!Qdq1G%=R1N{I;{%5O-nMJ{%ka>2JYce`rCQ-=wN)2eolIwF|;~OiU+k!XVXK|*jsrLO1fld@B2jh)3uor zGyJH(hB+Puc2effT+sne75*yrWcP+A*u+YlYA_YkU2I7;=O}p&$l!jFy|_Rz6)OS~ zP_wOrR@~61hmziKS+F$ATrR{f^|Q36@Pc6bwm}FVm_$7tl41Yu0{TvWi@n!Z^3FMN z>^aSqpIGEeatgll%>)JBe{4LIcBCUw&Vq7UUWuBPEfH<)d0q$}I*$~jx8Up_$*#f6 zRvf5g!e`wG0f2l2fH*KxMT zLUf1Spm*gZxay+CgOtkH)Xj_UMdL>RMca5MU(mjPE_2q8l*2`NQSg!Q#dQ+`b+*3*f6D+=yaMt}1ni90m0G0O%TzI#b3&XQF1&(qkQYjJvmH8#&x!`y3{l$XAq z?vB4sGoB5n9+L0f<86s>O#KjSdLO~DDL+u2wT`TZJ))ewL$O4+kv4a|r=N%3lBHZ< z485*~>An6#X^%r#^f?s0Ra{{HAcJ16kDwzOjWk93g;2QViD0ahOdr=N;NgD^vF9e( zkJv*;mOLWgdPPj^kVfZ9MI@h=N8ZknLYeeF%I*G4`)yWG==ssqr=J{s%i2q0R5#J| zySJ&N<^q+)U!zBR9+Rb~1+oh4G2ucDvK0H%a-ko_4f{p0zDwE4(|`?fomv0-d-~*( zK?M$n$!Kwr(05uYJzDUD5{++DuFhk6_D&rI9#4Qu5RGY(gIc2* zdrR=>;|?t@JzFNyJKc}e9!o*qcsM3$e4wx8`^jTnyrAHgMvJ1dsZrwy7TWKF*Xl^@ zYI2|@J;qUtgqJy75{AhhQ}FFv5YKq0IH&fJ$wXsulhCUszAb<4=!7EnDjJldiiZ_59pH3OpP?r+DJe#i^LsmID2#D6E)r zOE9ikLZ`bj>DB}>at2*Sv4(_1+}6e`@15nt#dkTbC5Zkr+e!_;$5Vw-8ht(Vfl_x7 z9*;dt&Habbr;WDMC+Q}&Ka)ejRwHZ>az$x22ZWLBpM>9iqN&`vmC8pBf`Uu~KE@xy z{B1=j@eB~ZN?#+sNde-q>pHk8_X20$l@S}zVqs5U9WA%sMtvS@WB&bD^w2{NduJH1 z?70h2>OTc>Vq3^>?+q{2L6W}$Ox(Kz{pTD<=JzAG{KXg7B#F@)GSX!7B8>cdNnywu zLpt>&Qt13!C-`_}aMVXX?m24@%f8h{@K!gx9%}>DPjWoUDiPzGfLUaM?dof>@y|}g zx<%s6vcXt6B^OO*@rXTA|7h}!Y`Xkcfg8L+`Itfm+b3p-CiKV4Wj*RVG#|??UO8Q<&95iRV9f%Z^dfVq4csyh0@hm#sa}FZ~}5y^o{KUbiUG zDvgYHT^BAT%tG?|LwLC>2jjKs;XF^$_nZ^a*fmcmu&b8tzUhxj`RgdjmE!p}S7CRk z4xv`3(etsKFimf;p!IgBaC~m2&~hf8TouESwlNjsObc+|{R+Ci%tM>>b=0&zl5o8d z=yud+yIL{d88HV-l@8$DJac~B^@yH?){x7u40^sHj%pT4YQjk~c=(j!L*+fFFH+{I z%e8PfArl6Rs$jTZ3AR#cSh-n|S4L)IMp+W>N>yRZyZ$H-9te|@R-mC$Fmka%a9@2K zd2}ANPTw$J!i&fKv!d|u<>dRdSU8g@&3oMrVfFU^xaR&}`Y%ihM{{M7IL!_T(P4P2 z=L4zujd0T_g#FF^7^!p^q2Ke7@*xFBwZmx6PhH{1dx8El8A->KkI=Kzlc-YZwxDx9 z0s9`wvyJW+>}=hM%I$v0(w6W$3)=Ci-`%r)Cj) z{Y}H&FV$GxIfXrrF5~SP+xSdfv=BaY7^1#@!;1ZL_>b&7ZrVrOmZQO2et*E8Utb_C z?#(mat>>BTCOqV5044@}gl2&}53$lCe0@Oe^}VqtX$)Hq9L=vA9C++K7LB$TOgB8# zVcO#d-g(P%&r#AGHM$@4a;?#zZUx<&on(4@F?1UCq56#)HdzE>ifaUBJyN1G!dfad zZx$TVi^y=cGblI_k4NXwR*?u1`xfEGssK#hcMH2lTtmoY9rjWh%F+FD(QFZo7`sk1 z9Fb--*@0XW^bXWpgZ*~P!KT|4#ZwzVlb6$pxSdq%^NSQ8m_WfX86!HI;21NQ3PKK% zkCFtxmukT3s0BoC24cnZaYE>|y@Ig%h_HUVF;~8^=4ivkToXH(KWv%Cg)`0hc0(A> zTp7xz%j0=wdY~jpCzl;HY`H`f09TW7uy?+N;+cwY9$H4hi=EJxPy?&bDSYAa3eI02 z&hL~K@Gdsztk@P@i?HT*vBo@E!n-dJ3w+?CI!{a2~|wZ*jP|bF1C4e zE^9Wo%a7q_+ckN%jylOMAA!d3+c-2-nx{{4;7;9b{9ud>&sbju^A*8pk@YeAb_siMJ~SLVqGu9*8jwNdB)WZI z6=|=k7oJ?J7Z$8Jh%auBNbN=|ofvpp7}p&kU}L&q*!@N*f2~0Ia)&5psVilMU7$sI z<+N#25-oO|Kw{cTcNU$ciCXLEMPvrGHAPaz`fPftb)5ccNKtH5voQX`eL?H#B$9nN zi9{zykzRr{bqtoG^QTQH`Lr=@+G9hCIX*Nudk^Fe$Rp?MN1D;Jh2|Cvp)=zQNh>Kq zc$*L<1dquSj4rN5Zg>_9PanYEmaC-cnn9Oq!l_;R4#gJd2y8j7aSnV)sYNyl5lj=pt5M1>+fQBJ1pSu?0} znLo`eOQpGqjoMRa}WMCyU-ij06`5N*ZxygL}{xaU&+lg+4M$mpU2lA-S zrZ%;Kq&!!Zmfue$*YAzO8S7D0-yR8d0};ah4S`MSO1fS#fJ7!bw6i)1dF6*O;rSKl z?b%3@OaNN=)QP4zDw6UCQz4y7h0O=F$&hUX&rf@W^jZ zIU#hibTiE!`C6Ey`%UoPSSq9ii0Q_S8sSH;-n6l95{17d>VB?CxkA-HOW}Q{46c8^OutWEAX={un~a~dVuBUiO!LTT?INn_ zwU#vJ9HhRtncm9pr#-)Rv(3t#oM_^~1%Jy0-@ZG9Zo^x``wJoD+qs+07n;+Vz|oZ7 zHI`aS_fX1&W0aRLg?gWxMd|*hh1!sTv~IwDVfU^iT6Nl)svH-Q)!78pRm?~IPcLj- zttK}6t1q_8GY}iZ&*c46PqW7U2mDHLny@=CNEElekDyyuM=1docr7;t(vGv?eyWvB zT0c^X#da!G*QHsDJ!$eeTl^JILhCOlB(`SK!CWR6`yg6-AW~R0Rat2FOA%6QjtjSP zJ%u$cuEIhsiRPx!iSLV+bLEPg^iO>lK0ouoluN7d*UvF+oe}U(BzQSemRgzJ)C55+f zbT8H&#zsn*klI22cmDO&wvblO^S~PaAQar)#M{!-`PX+vai+78q~4uEL}xR`#$Umm z85i+p^rHl>ZwV{9uM4J{hd9C{jgJ=S@c3f^xNM~c&4rqx?+?F;y6?DBOiu+k zo%=-Zj2<9tPdF+HmSUb&7i|*- zcT$=ER5}VVy~(C2h9l^CqnF_5{z3@ae2@N2 zD?n=EA)H+)2bC!i>>O|jAM!OY(EgFA$?%O}m%4$j2l!EQM_*a0j!?%j1FX{QF>V{VR*UlH|?Czx8@P;P1wsdoxkw8$0MrF z>LA~)?YPxn4j&9lg*WBu&=e1*Kafa-W&#Z?^eNOL?n$rEW>FpSMc6u zGq(7B9XGW0Q*m)UmC-0%K75-jyv^XXLk=Szhme=$VX_!K1M7@#K)lqN2hN;G=Et|f z*;=ZlrvfiKmn8pq}asjb>Xq zHav=+>#7Jb5e*nBxzo1<2a(?pf2s>AqQE{X_;aic3$0A?MWu@5r|%QATXKZ~%d5yG zw3O0*$I;rmaby)8NtU_|!v3NUq~!3CR;AgZXjKZX=9$AWe=Auik0mMfyBN-0G%TVI z)kUwvz0^3IZCC4SY?YnKrNIC@#;%9vo_}=V);azl ztY*3K190`>?c^XGr zRfQWzM+gh&9iiu++60$%=FqQH!YHLe^tBnpW7X%dWkDVXDNSVWW_ga@Pz%McmiVY2 zK^3iEQ0Y*?&jy-{yQ7m}E9#FSe1wKH9*5Ct5w_mRp~_>m40k^sS9pN8|5M=j3pUW( z(u1a5Sc1CTEEudj@IU?65XVS%v-So}B*Y6t{dRpy1@K2+@(5}x$r2npvJN24QO(ClY>kz^epc)pq=d~EfD!ok^4 z3Z0DT*GDjX-8G?O%?V*u^{8bXdo*@UF6PlFS-7pf_Pn# zHEJ^QQIh%!FTc2xnVJYe{>EHB=Owfyv*SLeTJXN4RVbHulB)R*zNnj1uQS_lK%ym` zl)QzS@2H~q>=np8ZxIciAfWL`9jJdC{nIp~7O#9tPaK3BLGcn+`wDJV#i8FP*-kWLz#XzQDndioVgB?)OYHG|yjV(PO{3mt z(?pX(8c<`%kuwTdVMcH95CzHIU806XI%zn#)=?;GGlh+WkNWZTuSng02X(l*z-Y}L zQBl1rEp0KxE}=KwpWGjtodWVhFJrK@+JS%t=tK;>{ zLWC{U5Pn4L5u%35qday1JuAzl;HzOcc_~VGc}$B$J7yuGJ&!y-GPFxbEN*Q*V&cs=h ziR*rJtr)_eMk}K9RWg)%S77e+CG^#Bn54fBK~L!^g1f#Zr6fEdQT+t;SWc+zy#p_9 z8=yLJ3uU^i<73TGacEyx@%YKx#eUaK#7096#NWqW=hzixSpB*lrsuVJ=f6kz^Xf6I z7o_o4=T;%)rZPoN3ZfI=kD~9~XLwv@zyWuUQ)ImYbGL)6|rIO0@B(g zgjOED!5sABt!O|;2FdlkhXdsy zXnyh>sSC#NQ=JDew=m(9U#2`c=maYIcF9ArXERA zUKi1=JCVt38ht%rM@`B1XvefT>Q?Ha$6Gg(U9J)`^Bq|GMMa%TAf!*?W zpTiRlZ{N>X-`!@XmTW#Xyow&3dMNz5xkWVUtO6Cx4x|ZR#pv@hfTf2#!@UDGsJuTO>Uv&~SAIfg_g@gU zCV1dEeWeFC=l*gQ(AIb4K@-FIeBnIaoH~OYH_P%Un+~c;HQ-3ueVAahk(N5RWB2)) zB(vli&H9xt=_L!$b-s%Z{C-B81PiuxTEXViny__I3^!WFaoG4Z7JEEdbY3<>H2t@P zj~Q0ThfQ5M)6-KVRxc)vXD9G_wLahUe1S!E)l{3&g6Fdr@yz5{9x-ShF0Zj+t+CqN zYVeMpuDL4gT49UfzD3Zh9nO~Pbhu}yHJ4c=vXAaQo;0k8R@dCecemMmusejU|5&n= z89{EwG0xY|=kb$WIXLDyME4uT2xN&ywHy0*#7R9J(xr=W4%a z3x_r~%t{g{VJNC&-q1MFVdTym!?*id^XpL&6!CpFJPXR8w)=$eh=yafQzWAA7sAm$ z9lL#$a4cXEmAKy!a!;QTl6WvVnzdoN<=Umzkr6M0}w2Hui zi|V-cL0;J2zJPuHyUBCRPV<$_7Mvbz0kxsUf_sa`?c4Z_v;=h|L50LHF@``k3#9v(>p6CehXHxRWis6XlEEGz}ICg||5MGl-R^ zX7Ty^k{xCTTsX5ViG5{1(?L1_?ajK}=foOu)Vuqf-j)pQ`7_|WtDRnHeG&rVLTG-? zSYh$bDms7rFZJHlh)5ZUS9;1Uwyza){O=gp`DXKmO>a4){So`j`$petZ(*3?M!ut4 z#JOL8a&^mVe&Zl5EWWu&xR+-y3^`cN_LJ_xTY3cD-Su|#ncJqL!w#xpO64v1tlY$~ zIL6A_SzI`I5ld$UV~1M-D(*;eW;{z2spDahw+3pO5VPff`>W?w@^{ zALoeJf}Hrv3TwXRZ;cJHh6qW!Oy7NSU~9J&vRlHjD%?;!zmKK3(d`{C6xGw|HpyIJ z#9bOWw2drmiUq#~>jl$7Gya&pjAvK$<`;cOaFIeAZu)PdIU0>@c&Z=woHUWTF81e; zE!VJhW36y2D}!9Z63ItLmUe2S;B?X^!RgNMC(2xI9OypS%WsmtWZ=hbTx)DO@;iPmMa_HcgaACGlYV|jktLjLyR z7b05qI9cx@E*Fmkg_(($eS5@@oAUYg_G2t_EQdGkGh&+__BiJ#jqZ|VWY}XVdR>ac zh&2`%;CBfyOXE{E}r@G=C;u29Dz?npfbH?o0k&4jh$MDw(5ypfe^VbZ^H8y8T~Y z`f*##6R+k#VPG(Qj6KZlGOc`Pp(P8$wRi+;^LCx1(Efdtrz%V4^MicYDY+H%&W*t~ z&t!5QpvLnA57r;)$}#hrVIH<0Um|x27U2kD6)9b`#qm*@BI!u@hkaxSAq-40-A)Sq@ln6kl8aVZKtO1osU<_N_a7 zbH+{{V(}gww|c|gZWhMuzeBYrk~!*}f#msKVC5_gzJ6fEA5j!>>Znj7F5aPr<0m{rE)NZM>I! z$75#vM#iGEC@hlbu4N4Pebjb%UeYF~3^~r9I|a`c?iV^N`mpgqH;#S1lVkEE3xG93 zpe?(DmM#^u*1uC26Ie~(D}#|5KNbZJj|tY2|8;#Prng@a8f}7UUFLG~dUab6GDN=TvXWuIVJDeWLdc}LWYC<&gv8(Jq=Ofh84`WrVI>rwW z;NNl;-7m-UD$OC{N9qRR$seS}xvH)3oi!Mx9(@q^ge{bOi7c;iL6Tm)KBpwqz zeYRD2fwxn)@sjeR{7myBso(obYXaYr{k?s(m`x>`Kozpk?Zou%C{B3wiF!W5& z(VdFXUs{E6FC39qXAilsIYOw86?%@ez{k>Htclg(4Wd(6lqG|2#U}VG+Y?s*&BWq@ zcPRTEz=n_SvCi((T>IoI-!GEj=0;D^tj`g`qY!ia86PRQ^`DHT#Va{-dlD}m>d%F} zwGn-65rV63VTR>ImOT-~W`WA$IgN(mbGAl!JV&2zO^fB4h+-_5RfkbIHFRcIPt1L3 ziatAz&ghd=D(W>?Nlg#QrUrY9$TxAc5g5>2bSw8RH zO&7;ZN0Q@lI+-fbC>L&oNJ<%v9LGHaZMiz~9Ad1C(Cz<}7Q3#*sDGn*b}w7}$QdKd zcr44qpB2F2upGxN(c=dr7V}x|CO@@)fu+W3^WXvy-{ zHLvkfSAuiykAjEZV16^8kXH@YV274zyseKtAC}|-4ozud?bpd{vtlNq&Y$D5%^e(b zRasoWc?Me_aG?}V;*j$JtlP&I8rM%I$T%Z&)?H07lduJ)-)Z_M0hfS=zaoK*W>biUawN?bOH zzK0z_WkV0{H_?Vq96SSsbM`_}(Opr@t_xHhXoVF*PhKpi&*yKA!kG4JNZWD~y;Ajg zaMfDKe@Z9sYuYT9pNA0}Wl7;{2G?F|<@gPy{P0USR#q(*#;vRxEyeCwI@A!8EB??+ zwPHH+HIW*!CsS3G2{ugfNBfe+XdbhGv)l~$)GaYg&Gh*6s7shA(@A|zPY9yqV4)gD zl5A>T`Q#G^1thWI|d&|%7ZUKsL@b62^SQ7y~syhr%SNFH3v*QvcyKIJ!Q7 zJjbjPe&)yHlln-wO)7zLiz=P2@52&XJ5Rj0hQlRuOoQy-!qR=mILq@Xf7{^3RoxCO zJYLJ|?>q6TN7cM~<~MF%{h2FXCc#Or9Y?h0!ryQuuU5EXo|ji-ZwRhYke4&PSFXR8Vg{;1zb zzc-B+CyDjNo*Rzv67^ZA%YROPH!nhisvGy&7t5D|wf+bHJpWUJPYJcKsXvZ}8TVlA zD#IfS{^LfKf!w`Wnw7)@+5LSEHjFMr)U_k{JvN$(hs&TuUM%56?y-GlF8>Li$g7;3 zB;FcLM4fz2i<~cFkg`5bm!2aXbxq#q8^T=&V)?4ZNsQjQ6pz%CXu{=M)b&lE3-4|T z-`k#GVMRW^tl5BJ5>4Crf+KXuc%?8iw+7z_6~VzH1`Q2;kmK!6u|}h){mpJHluJe* zZ;8g-TZV;@Q>a*X6fczq^F$|oSS=q-_ZRoT>!UI_5v&Aj>pOg|_Xl3L{sX_w`cCFj zrS#v%Lo_`ln62Xb^5f*)*nP*IyZa2~?fTC#*dUIB?bCRjel)9+!f`FRS`ZSzr^(2H`^Cb6Z&%hRWHI9|^T;=YADd34f3HnB>i(zzl!Hgzlw zyz+;n2Rxw>NT(HxJ_u_sl?!_&)Cn`AU!vY>7@kF`(}Ss(XvWloa5z4M12eoyX6PGg z5mwRlL?iC(nZzFXU)U0=9N(0TwGCBt>F0CA4im%vzbSN1{Tx!x*g?**KdJfzQK@zW zUF|azgX_xh#^xsdIiNsC=VtTouhL?Z9@^rJvoiEI(+Ax)634wI9r{E<}A=r0~x^U67WX$@R8Bu)b-9nP#DUzq!vAN^d&R& zg{ZAwgTpVK(Nk_Fezt4EdecuK&@!9uZBM243p!jU@#v+$zfPwv=t#KrZ`2{*PNUka zIc&-esCB79#;YLLp9{rjO6PDB~UrnOmgybXz3tt%5tow!}jlyzSopLi+or& zHWTu@7T|ee5t%35MEai582Z|tPB{OhkCAc+Ir5BTO(c3gKQ%~FxJi7?factYp*~I_ zv}2`+bp{W`#FJ7&=5Ys}TO(rKhU=(UJ{9Qcg`$pbe|B+r)4WU6_c9W|7U%D2w04ck!LzuUf zEj|8YQ?+d{u3o_DgW_@iP#3+F(#O`xx_BFJK{naMn_m)3!k77$Zw!}@bmQ#eH5{T7 z%)N?2*sj2XpWB4c=9CC}{;^b;JWWopn6j3p7D#j@;X?4l-e4^aj`>-_D0~c%7&6UZhla}%|7R4O?f22{BQr#zjv=TPpCac! zL~E9eN23JWg+`9V@Xw1R+MjjQdMH_#mbQ{kObVv9HG`;Mmj=F-BB#@ zgUTLTN_x-{zbPZWJ$N=7|4P8M@69wf;4tOgKSB2gDZ%eZ1gR&kWf_Ge{?jLfUH-c! ziA$H}v`HVZ+d+?e?#jo@J-wkn;~%I07syjJB>T$cGr3uM7aMNu!=@ws5#%C+0jnhX zA>RU;kkcE>gA(cX_p@k@Gm>~^yYv*f zr?MgD>VD_QUsrhMz5Co5Zbe_zf@zqcgoD^Q5oSZY(9u;+yLa{AAoIbvH2efz&XxFB z+7)@0=q)l$22#s<8`6$jLm9{2#NnY!#c~!S#T{Y|_DIpAp;zfo zr>4YD`;kggjL}%n6thn<_p+B%+5LYh!fF;7`UfJjLKfYZ;wP$X}w_oUYXA(T4MIV%YyO@CGq)4E%BL^ zb=12Z}DOMP3&YdkRxq>z~o^k zwt9zRXyPCmEF8vsnJT<#`GAAlYk8Yw51ysYN%SyyL7IJ@Q}OpM8n2zeThB{$(euT8 zHv0~CjC)P9LR0BYeg~32E`{msXu6Sph5EaSDK;=cm{zXCmd?}JKhv4jT(kJE{y}be zK9`GzD~Ofm`|@r5D1NAVom$dokY9NZ8Y#(~2+XTM*6CVo=%vE)W1i#c=j#~cJ%X3? z^g_X03-~rwaIDOD&huEsu9hm?`RX95T0OANb2r`=)l$__jsKayYm7P0Z!Ps$Q!LA| zq19+wREmKE2Jx@S!}*Pbe?O98fm`m&(HfkF+#n)_)P9e z#PtY8T5LYzw7g&^KOWzAxX|YOl~j3XEuBAn5xaW)rh4j$-m7N9q~a2FK58fZ_Ej)i z_KvzQjKqQwp&YMe!RJm+;lulj@H6ikdN<(ztoTXKnRXL!>-p#Rwbn94`;( zu)3Wc-(Ts^4gpgs$+jP2y7lnYB2eOIPGmVHf8G|c7oJsi=;b&7al2nY-{Kq$!b5Ps zMir4x?w5UOa}w>gO?iQv}n~F`T{U68;>O_z#~Na_+wY z5}vyZbH7M(9D18G|5D*|ny!erUMY+V{fF0!f-rr3BrbCSqEFmMOSubFLmtrg@e=G} zy&XT&4S3`S1$K5X!g|APsJL^OE=2p2zV~;0sp`!tQ}R$79EzQ0{`lg)N;0eHhl9O- z)5+oe@#6hZe0$eLmKRIOc8(ePt-eZWX{oepdpF(EJB;S=43y7z!J-I#Y?`-+yk3@I z{DK2;`&WPkdk5giMGKqcu37ROhW~|I=W}E@-4LD{nh8q` zQphw;q8m*KN3o708ChVd98kYSgV4j`||DCJP zBmD+*Lg!3=YN5^VW3mupmPip(hEkgM2_fi|h!-R$BI9}_U2@W7+4*-Nv!MZ5`>yfa zO}n{nsldrAe$)Bz`LrrNi1c6Fq1UdXXtVnziZkuOM!iem{rV6_CK+*VnI@01dxWNs zIh=CJl_#9_<6beJ;aQi4mi0;SU9*z&vWz9X`2YGZYF@WuXR#!=K)eHPFV~`IX#k$K z7DF*v9XB4t(fGAeG-E(3S+^RZ#qKICIBkrS$BN*E&1CDR$A6A2llb=BxasR(q4}5u z(>>0lvJ)Ffp0EytbRPN2`|F1 zb>;YZG#qvYR#ZH|n#2w3DJol{B^9O9RND_!y|0xf{$7pGiX}LEq?_CqKBM;y{ScLU z3-&*VZ{(=*#FqW=d@>bnx$R_pxLSBOu~Cqb)}uwI9C1DV0IDULp~_Xyp)a3-VvARJ zrnUgej&m^nqyrQo$B!Ew_^!4NKS|t#BFT}=eSZkxf*AC&UgJ^i6Jk?Fe zmgo$N@JcTi7F)~l)nFoYr6qj9leM^|6^-l4#}I$M7Qfz%fW3|*7FcY>-r7N!yWSlk zhfPpFU;zadFC(o_52#Ul0yZ!0rl1)+5t2C?5t47~f@~tXV_)Lc>JKoS{S=GGd*kEq zNQ_!=9--0i=#FS00&d3xe~x4A)-=d>%tw-h_q}R(>Hkr6-j7`Vf7CaUjEt;|kTj@_ zyszhZzO|5)k_PRfp{bHmDGeb+L6(*D6i5|leh@ZN6$^pC(4q`})*hQitv(BC%-Is@k6`wczl zZ(WVjVoly`@)C`;gRy7I3pyj}RD%b*z+~!Q3cS|?al2zt^Jxh@JmQ_aFXYhQ^xSkWGA4t7iTwiWAnYD#$8wif1`UizuXVk zZ6~09z7`&PEJi<12LuNnN00yd^Moud79uzBn!i`+q3bh22u{p*$xcgk02oO&8VQWqd8s*sXg%5lVGoS1S!}B>+-krCfEJxnJ7Ib}R1n2&S zoLJHUZFA;tv7B8B=e_+*`95jd3VULCeExF`KF{!;tMbCvlOgX}_>6iF) zOFea7P{tR%E7VR)yo~qIf02W;e@v*vbO05v^B2m8y5Y~OV-Ox? z!J^w8d>k~FhsY)ICRay3>vEp&jSzchW#+*^_ZGU9?8fxVNjRu}2D?Lc{LlQ=?p#T2 zal0_~?j(NkX*&7;EFo?Af%M{RIx=6xVxW2~e$`&47nXiP!|`;xcj~`|IlER$hJ|?B zO`f2N@XnjG_QD9VKOQ24cb_Ktbmy+!2j`uVgR_oE3apk02Q^1?VPFAzJ3bJ-Dib89 za@JBv%@UMayU>*W4>78B7k_x-1)1OX=w{YwF|*x8s9dNbTn{}Xx!vWxP_d_2=$qdv z^q+G`@~t(Qe01mG`2eP#SNf4+vNjpi$D-Gw4V1Wao^bGImgLTlc;V}j7@_t<1Z57i z5q)}lg&eta?AY)L?~mW5YqlTkQfqojmQ8vr3=9mT?>A~CfgeL93l@7K+Av=*Tu>wV zcQTF>_r1iv0fubT)(*myHe=F=zA16c>?iSGDdsx+#R((1MtCDKIAW^y3C_;vg@4Lhh#b-+ z>nC)PG{0~@&YLGB)Bc10NC!y*JE4QUoxjku@Eww`T%=lAKS}79?liwN97jr1p_=@_ZkKzm z_4p*E-sux@5kSu?_PQ^aO(|?7gNdo>Tc1)35rXs9AaXfBiUz}K`)-y z3;8Bxf`i&};pVLq!h=sIX;NkwK0nZsJmb$po{VR^N>?8$czjw&`MOG2z0iOPzeY*^ zEZHU0H~c63SP~{|k9aGz%0$zYkLeP*t?`nm-#aDEU7kzUWZV!=9W$oDmleXna2Xs| zji6y}HuV0`6XE6Ip)}a3Nz!3vDVZ|AlKgfCBWIz2sT(TU?r;u&xqXl$gVOD;dOnaW z*4rnUE2~GgyS0QIhxc^Kez2t6)Geq%@?l zmk?A?DVg@Qf|`!&Qdzq$MV>oI@74_Aio4eQq}-PM+h1Y0q+B#!f8kW09m0CgCv+<; zmtK9$#p%@{^nGQC5E^h3p8d74Yua?0`B9VFc{gJBjo|mkyOQbPuC(syP+GG^hb%uN zNP@g0=zNE!WblV%Nvd8yEbY-Ik!n7Y_1-k3OvJq{J(zv4J7 zYE8uM!KYwpe4QhY2Xla&BOldgY!iPAe^mEHYsDw5*t-&?holI2Sj&%69i&q{jioO$ z9?+Uc?ewBan^yLSq_qK3%A(am!T2OjZJEp#dWl$bjk%xSHJn&$j(le=c5>N+@1}cE z=+Tv}gVtfgmkpFOQ`{4g7PWV^7Q#%fLDBgol($FWMlWC7Jg7%|r`rf)1M(4)8VB<; zo|xcJ#67qFVwJ^fc)^~1IJW#0j%CKe|Ahgp*4`yoQNx?w>!aAyv6Dxnw86EoyfFYjhVOy zJ+{unlP)7Me`P3L6VAb4gFO;+|51v0A+C8QqWA9wBFkVUF8Aq$+6*0h3LeFUZL_fH zrw*-K<;nDDBWvy2$Scmir4)Tv8f|Y)2?Hl`<&Ag9?_D8m@LD8hfCu4i^AP;9zRWIS zd)AoI3W(Ys28BskR;zoG|5qe)*eOk`IcHLmn zOcggpn(}n*Y#-t5_KECi`Vn{ZbXfQCY-DctB+0F-^rxbjO}hMKyBr1Sa~BPs+G7qr zZF3g-?M>kgImUcs@qSTTzlGxl$FWt_9i0i{I~iIA1Ft8T;jxGd?tRAj^E;93uFchw zew-C|5kn)j`ETbzbQxbqo$se%epUsfbrKX$FBJR}w^5yrJB`xT=Otc6=w70Nn0cdd zaGDpMz7hEzJ^i3kTSgguB5Bj%DmW|uhN0F;9KMl>$zF*<~m1iaTOIAThM>= zXQ5_{x)7lijMZzMv3Ji9qz%=Ox~M=pudbdAWMl z9EWLy^8F1W&*@Gs-@FvUTb`=%F;JWX==`XHh&(6jUBNt8y-)pzpbG_DbV}f2EF-~+-s7>|I1t^A2Z^(;PG6x zUdo!^r5x3w!-L#ix%-$#eAZRwuk;97G4Wiw^F4Hrxc<-m=jgESxTY_|J_WbY5@o_I z@2B%X?S<^);>rg_9^&QIj@c=7?~e7yusOPmpNCL;` z30t2XLZe+dKnSK=*|3Jn zI^OecC~pnC3^U6gNIW30+P7EGbg^Z(=fpN^ty%eyHu3AmsDt2S2dT*YzXgDXYnh2k2k%^aAC^FQB zUgbY|BD=xDLEbqq%c?IlpSp zU{plGeNY4xTJGYCaT*L(UdPuR`TukN1Urd*6e}?gVZQ|He_f>zA1_QeHxi%2Wq6=k z5&Eyq#4<}OXikvBL5<6lJC{Wd=XtuTy_Z(ns7d|q=kd*rNj%}0DZMWbqUGb}QuaAn za$WRKQcyMDt~J}8bNqVn{=$o>xGwsiG+O9b@))`oug1STzmfhaZ%M#^@9|G*B1R}& zr_uY)OMV3{lx&{%Pttxo1g0;vG5+#XYVg`YTmD1|mwihlem#v~sc?a=%vnqezlUST z`7aoGK)e%8P-g2>c_>nUMEld;3IpS9X_RnD=<~3LbjtNKc6z#&JG3mtY>%?|?N!pa zkN}af*@x$u=Hj=iHjj;%$zxB+@Xd$SoP9~$M|zCmh2`(*Tm1p*C(RRvzP?4PbSIPd zhce;ofWuUq+nX*g%NA@MUt|BR45XdBKr07XqUXjc*!HH4mI@utBL4JD#A}bI$bJdJ?Ic~g_DG%9XB?vQ)@y{0hi$^d z-D#MrScrRNIha5GBQ+X6r`3SA1CDO~IrCkbhZlujKoN;=@UuJnUZR~~dK0-tzYkH$C5x z+GHsIiz+CscGb;9TEy-0ks@cJEV$ioc0aq@X2fOwi{8$4yyghXH4r$79k*D{8ID z7Cw|Y^0e?z@X#8C3W!{2pK>biv?sIOuQ2Jh8Oqvn=*ESyD0$_MXQ$;*?45+rA}OlQ z0lpst;eC1u^)s6$G==qq+d*-zNHnBj`j083w1wR2!s(m-SWNyLPO4__grE;%Rz=qx ziIc>f*S=UmVP!Y+c8RA+SFQPw*z=G(4CrqCK)AN7TlqeCq$p(1TW4}0*Y@-X{^c}iCoNR9;2_vJ6g-&{4kb% z5^cCzEf*Fi*YK5g1%7{;u;^2!uqtk{pow(BD(@S%E;C`X{ZsjTfhkY?GhXD8TJtG! z@38C30$jA^Sbw^(Kwc%>R1E!pazT|aTLF*R&c9+H&#E}=YP&0rRRDatkH(!xf7s}`<1)P zSMcE0tDMxgKik<)=91C!Jh*BjX3QA_Svm1eGAkB_;$CCx?srt=VuaQ=saVn1LF|jO zr3Y$3WUCiUP0~?#HBu3ZH*;y)s_FdAeF!g~c>zm&v(UOQ9|?K^xF|o2H*f8Pf%y{b zTcN=o2h;I5wS<-xM`1u;E3DFaL=!HyliAwuG^T1JXTk{GRj&&m9?lhj$I*%7y4`!VOzj4&x7LA%c&=6fsGDf@5 zCn*^7F84xn`Zu(OIDu!0+K#a=l6FLj-sac%Dj&t~?=I$4Wn0d%`ibK~-$j1XdNkj= zO#3>kXxNG=^e$hHU+jB^+$I%XzDpaPsUib=LN&&xZs7LI+j!+zQ|@tP9Lt%Pi@Rr0 zt8_P$YWL}6!_3bd`DZUK!C{Ee&|>DaY#jTf|+xAFGO5v$dQVUY;DopAK~AHibzLtTL(ID+$AmRXAp20`BWP zCKkrPTd$rQ7oT7!QTsS7W-4s74Y06>0ve8_;^^LXTyd{OS0@wp)cb(3#*^{G-d7j=LN6G0prA zuFg|nGaYNzw-B=fs$<}MRpi|7T#BZuMVS9@HnjD!$!FbmY{@x+I;SU;rxr`uPx`}g zrwl*VF2aeLS=fAT6y+XNgN55XQaCplG2J{c(Rd(c@-|35rz1}E6&^14!Pu(xaQC)B zRBkpomK-HB^O2BOcD6gDl5;^peexHBt+o~RrVO1tTC z)a(Z++cy_#Q_3*WH3rvb+w$1v9Q1LqMe_rZ>2l+cIFD!WeBC_mu0LC_+!F$~dUYOY z{Qzfz1#UM3Yi`<0eMRmQuY4@jt(nhXi}o?y9L%||cJY9=$$ajr7I&*OV%@K!IqF{o zr`kHR?awzjrIUiJOQ$&Ud%iSK*5_PZ2wCwU3#VNJq``014X zpc~1Rbj67;yRhw=Iu7KWBgIlvyjPVME-h0e#Y9*1-yO|X4?nV+*kv6!yASmor-4~N z8u4jpHz>$`BOiNr)c<`>bw%q?H1RZix2y1;w6UE3uZCS`jlhL~ACxx}s573AGb4xa zvadI|p+I~mS{}CEus9Qz#t-Mip?mp_Wi2nRm61M++>C3T=A2(5vT#?e7CG~8 zxWVor_dMi_{ev%|WXBuax%C^6X!p(J4ERI?AVuUMpi-;N>l77HZLjJ=3W7fgyNyz8ITzVzG3jnB$DG z=kll^JpOYYL1iT{ac&i!blTwWM$w0&uFglNFJk+Oqny<^jnoGW!@YA?pq}l^Mk#-J z?535{Jkg7j+x`#>E;J%^jL7^C4&~~sQobvum`ieXP#K?$^$K^6|{UXt;; z8Td3p8^IQjsdRw~wX4_Akc;Q&x_CeR{aP0eKNQ(h?0L~T9*?w-CsEq!j36T!{?~yy4X)J9TJ5`z$a+VugwYYM=77y6@O_&;& zNb^LV{|&B2!ABL|suanVH?nEOMsa^^Tqg2*U*V3cK0n^Go$onZCrQzDDteSpD~E^D z;`HY<)nK})UAn{PaT5~CHt_TCa+a}6;SiSu{$)KtD*t|y^hS1s)V^^UK0h76jyrer zi`H%|xv+>|898v*>I7P?6^0+9w7JUmES?)ps-((eOy|N>E`@Tlf z;<^0zx{{zcyelm#s3&UdMOIz%Xu9=CI^q43yR>;rv-P)2xt|s+Ru93Xv0L$D+(pdt z9>)Wd!`W3%%7KoRTp9nB|C4`3=$Q*g$!w@aS5UjvVXSiR!i7Q>G7gEpc)Q=MaZ*j{ znX*Ufp?gz$*m^7eO;O;sywyB-%U;S)jYn*ExBu}!-?;gW2bQQyRXQH>DUCuhOEN-% z?@`*c=oh`r*-p7#@<~IQKq;%%ll56s_*%tt_Vw+kPv0-{3eV%x)fOIEq9M)hp-k#a z#N5%R2~=M)0e2>Tq$!m#wC+?tp<{yBOS06OA4n`==jMP8o4ulsdK2H7(SoX6A988x zDc$sVCM%3KV#m){(a-z_Ej+e>YJ47`YTa6P3JB&xm8HCKjTHx90($TB#D`%T*zedK z(jGIU@4g#JQ)chxw|R3(tJ)X!sl%}I`+G@eb-F~M{F|Vu(pBnnLRYG->n1Hw?I%rc z9>cCzeChpw9WY$z!lMI1c+YNuU-Ul77Gpx#vf7XT<=x=#AOCR6s-dC~un&6n+o(2c z7?pY7C*!-j$!Dk{5A7X_xi2S^mw6PQTYs8Q1-bK#C~MA1_)b9@0rWdg7O(&9q@A6+ z1cjPVn1`z(c=in{>p77WM(GHBHR>b_R@!nmlCrsYf7Ke>0{yN#Xm#QNyG_xB92#~7 z%QAwgcgHN)erTrko7JfFW)bDQ52Q11B;;2fDm|vvO?qaeCx0+HOOnI)Aou+m&PQFv ztd@ElP!Hj0i(REgpAT~X-an90nkV($6DM8!{GQa}W<1;EUqIC}krTRnw`BC6Gc>_w7-88wNe{Y`fJQMbo?KIZ2zqG*SH4im#76s?szBWy96hPH(#DbYnu`XRoY7un2V7IS0w#f;d( z)2=9miNKia#J865Gzux+OC?>wE&w=)0Wf#HMnG&L>`ISHbB|4@q^lAC=0t?2>M~I|z#(NQ4K~ z;lk`_2SmGWBVYL=g3RZkbXxfeF4t(YMgMouKU&QN?+vBuh23$N!f9R6P`i7JukaeR zV4m>^e9`Yh8Ivt3P69Y@#dL z?yMRe#Ea&7v1h+0T(U5Wtp}{(!Oq)w>#Z~RZ}Sdv3}0rKyQN)nujGnASJF9N?bzyEoY zWls)#^bx{$IVdY`o~w{v7mvLiOfoiux8rpoMQ1C4PPPaS}87n z?~3<2ebDguIawAbQHJsX%GEX_sZ%f#pGM(pUIgZ=+abTi77a1|FiY<#-jyU^+1BBx z5$B2}0bSATbca%g`_b0{yXf`SDstXdLs|w&)cx~Kx;FMPsnc)zv`AUZ*seys-#V;1 zrUt{cos{{^2zzg5W7a1}o~=Nfwe|vJU%sby&oEl(agbE2Wbxoe7tEYG08?i6#hGD! zAnQB_DS37rmHH7~6pll|&w{ehk4}v`L7V6PW2@9x-1Af&uTu_3?wgtLNwJ2{R~PiF zX{Qq^k4Vn?EGf;-C0eC`*BR!l-*gKa`ExOS#3_>W6B%hTr8J?(b?Akiz#CCVUDe|a z2Az(FoS`kATq(lg<+a$RJDE3qAyK!Rj2%7nvAJ(5p2;7;?{!_VZ%-4ypPk185|X*Z zYA{?6=%KRR5E=74@h^G+4(PX2=-m?fx+b6Oubrf$@ z>A8P8b=$H8A>OW-9=Qa&I!@#L`6qC*H)fw}PPiC18ExBj5NmK7yC!PzRaj3{l+mIF!DUF zsWFrqpEyb`r9Wu-jh*o5EXMJwFq91RgI+*4HW;uJMdF$3HrEhWz07dm^(Q?)pO1Ra zP<-v!gOwUlIJ2uCdad>sc|-wNoDqwlmPN?697CjgPYCH=D}4T2D`tc*aDZ7e*Ln)v zV`o=B-#LJt2H%6P!WIm2TMqx7Jy2^t6%$<2pd2y=F+<(4dxAH%mi@!WU*AOEM+u&Z zJjey&n@L{fWcGP64 z&ZVwW!x(L@?Bl_qb)z`5Ap~zmeMLi1JHodW(&xcv5q78s-rL;S;$Iej&x_(q%Xf)> zhu39DC8HIZBDw&rHFmNBUTAx((w4k3)Hy3UYi{ zBo#fOsm7Z4dUz%>TWd(!cn%GkYd~|m#nXyu^XU1^dpNm7L!9$Q(8ceHbZDRx)qHv= zJQ2G`>aA@#XmdOqq!*}*{X<%?J%rZ#9Yu9`o^Zgkr;wR>n-?57$aB3Ha#BDXHK+U% z-2R%<@yk2Ww$vOleTHJfcO$mmt;JJ1ucKqnAU^e>f{(rx^Ivl-VLPY^=9!HM)btPr zKeWTGRYBM~@)(V`$fux*K{P5Wp7*^{;@P<`$z$wP+*Yfhyx0Teb+4A*)mzYOX9+D4 z8hB#+RepWZkDrh5<(YabFn`2rQkm{V`yVf&+)Wy!E@mwS*-?=7-zX@iq(GQ`1W&A@ zv3N*dsmag}oDi45hcxSi!QV#_UD`|kI?Z_1+nyZjIU17x_VbW86ZzUNkqxyW6#mW2 zuz61ptZ4HTrZ|=DO)uKo1Rv{(6lkx?-LHSfC($=KKBEcK zlX`IA|J~=@v*#!eUv4!V$tvex;rp{@vXx&haN1ZLwzq>}#97*=canBDmC(3$e_?cK z9DDz^4>@M~JXjGm=A2&yQ1#qisF1`N!MevTh%=Vq7oUq59 z?^ZpedV||EQ@0yzm(E5@*Fw^&8insu>r^~jn_TUO zOH+?dl7<#&Nt?85Y5w{xbWUV`Ke;GD;?l*W_wAZcZ4=BmvY*43h!kK17{A8 z1kNnw*8$O>-|j6au(o|k+Xh`DU%R2WX{jXYnjVyR)3lDjKbrPb0hrlia`4Wcn`}t3OSIxGLe;>U0D?2&X)oUpPGL z5pIc{2i?Yf;`niaZ23!{`*iAwny#0y>0>Xu!>N}seBu#QPWQm&!~*=&Y7`D9T(hg+ z70nx8cW22KXRL48%g?42qvn(x9)CZ;ZNC&)v9~Tx?0kbOwii&d@-F>-5yesJhhRGR z7M0h?aIS|3ZtNIK*{u>@`eGcPdH4gN>W+vMJ%A&Mj7T+hCDx7(#y1-gUlXvKBj%Q1 ztHg?KtR2n2Twdd^=3Ml3Imo|XfZO-?hH{o6-IpP|wa26sU#|b#X=5j^ix36?E<0ftSrXrnuwW- zH*`cPhc~q?=Zr@pPe6VEPg&ZHjK;yj^)AIL0br4AkLnwNw&84ns+`icZ|k+0-G zl75Vf7W$8WCWO1nLPq{NnPm5(NSS{Y^SvMz9LsF05MGjdC`ug#VO*G{VdRjjq!%#N)c4-k=7{Nmq~@Z$tJ| z)FE%W2Q|`2{?#X$Ya+w>cX7U;ToO+?pR93wr?FK17fT&`=t&(P8=z=Y4`_`HgWZ3| zf}K+mdFM^W1(h73HNy^h+fQK1=1-E%A*wKRspQKSUJF`Je$t7GS_BvS;j>u`eGHyX zea`9&;~h5(bux$P@4~B+s(bH*j(|=|o}xkBkCoHdCHuIivZyIP_=ZJmCsE@< zKN_?rkIHQN;L!D-2pYAJRoxSC^IaA>`+pbaY#K-jx5_d0Pb7z1?8WYw&v-sg^uVng zh6hEtX#G2ot^4}2@k@QS8ySk5I{D;W+!e{+G7*%cCRt@Pj}q6PhgIu7x^qJ0Z534u zoBw_lBHY$v%C@0Adg&lRXZt&e>L`EV$I3Odag`1t=e{NLOZqf5BAiZnb-}=8EyDV5 z#WZZ13Hj9~((}<~BoVd561_~x2$6yPuhgIFp6sFNu9nbVBYrc5!ytEG^b&vmNQ!mU z!k_=fNhV!r5l*SuP}z$9xR^Zy@u!M;@nDhNGw~k=^{Ep^9L%R{ORrH?leeIokRz#D zr-~8PUxmsQYGQ|+B5UY>r;2oAY#EY(=do{uz>L#^?9$$pnNdU)Yp>9mz$n;!G3Jog zVXS{?CcjB`mAfuf+}b6`zS9$} z>0K#l?;cXII!TL_Z_wf4I-F4x!?pt>aWvbMBb3s(c(*29Z<$3wiXm{F)jH(ia$CWE z?obSGuNKB1z9+?OA`a^@T_kLj_ycAOL3tP&P<`hbL_BP zaD}REH}o0$j?DE3HF_7#S<;WgKVHF;txIXh(o0b57X>-TBn+Y) zx}0#6nx1_kKie&2_Vo;nF_*<8qiASO3PfRNme8IfcEwc6qViyOEWMIT8Cru8_@X;v zMUS1ulo&WZ{X(19ACbro9Vsk4_DwjfGnlnf1G!BzoXam}N`5KI)4-xWkgm|jSyd^% zSi`aL+0$Igh@Jp#XyAAG-FPDt5I`=>Nm@-T7wTY&HL7DW^qX<`) z6rlC2m`jZ~0>vTcU??4quf=n5;AA$+=E!j78*4P?zNbsF+2q|_n-07?L09w-l9Ht> zImT=h%Ek^R%_VCPrm+@rac}6b$8Ky7{eq6-J8)NuChxt`Nba`@=O1|Ro)3X+b})dS zKeR;p`JwnI{#6fnKr1J%gF~$c4%g^CK!xBEQem`RreuSXtqWcfenzACg6m?8ed0 z@N>Xd^uN~&|8%!NslggXy7%Gnax-!t$l^ug85m5xk4?js*?f<;xaXhE7qu_odRQxs zZ|FudzorS#l{Hv#{U0>Qq$4{bf}fP;aIe+FIoo{+qE_WXulFUyoXXzr#hUjqQ0_EBVNwWnxc;-P3$-py-$B-cn7M{ z`?@=-20q50s=G)SIT?PNG(|u5Pdu*KPffjX83;rX_ zS>!{ScQ(*3@4Hl@sfunjrDT3a9>S zGrBM}ieyJ>lIn)@(m(G`OGgemEG;RD=bIJ*{HJ*hzkmCJMka4ZuY~T%lJ!NceKD5K zPldxTdEvz11Ul^=kB+|?czogkZJ8S*EMCz~)XmedVNWH-{rE_Ma^a}Rnu#V4(T5g% z76oeiu)lgC&AD<|ER;M;D|STi0^=yID7%O~%I(+&OV*xiz%Mp=h+TU-(M#mhPY$%i zh!3$8@1{jJ_ej7alki_ayLivOPv_^`;X}Qc=j_&mw#sZo4xd7SRZBT)^;lLGSXT6Ytg;!pc>vg@!lASd@Jp zd*p^A#k>{+&zN$54L6Drb=7m}qV8lq15eEz#jKqU_UB!oUY?`i5E+9yM`fZl2Dlt+ zhz9Q*^y-t0WTis_X%7?I=Dy2G^!2V#n$u^iU<84*QjwW54-s2v$g|#xYyGlWF#PBNm#A}?5i)?H8vf9)-5E6ZjXrDq{0J?xx44G> zQ0-BT1%JR#o^GZCUZ3cQ{aspO<_CiWk^gi#71i=xP@eajK8P93lph8>@a`xMjo0G` z@qN&Fa6f9A?_zkU2j5l-sdyH^AZX5>#>-VqY4QPmY=iVex%+JAMU%~<&r*?%0pUCH@OBL`o;vLO-tBbmQDx55p;jY8} z;b+T3ET1_R2MXk%bJh_{m-s?Yn2Add&kK?6<7rLdIic?B2h3j0Jb%Yn{ylaS-*7y@ zzp~EoDeng9uhvgeW$8s}{Nxl4ELG=Lo5lFoaaB0JWiO2`%%JgMpQ-N{J6zl#Ky`Un z%%AWcH@EfX;rBE+&+9TYpUx4xz>|2h#R~3RGJ%)Y^?{XtckY#J#>yF?xII>v(>G0E zo$~R}8@G!$*M=E@;C1x9AEeZ z!H+Iuy_hLhEYgSS?PBPv+QH?o73=?-%Q4&CVYm1W+=B08T&armbNdmF|LehZaas7X zEeW?Araa# z#bu|s`#%rK_Ng2@>@fG=X@$QFU*PHK**xCWSL(acLFzENfNwn?#$m%J@#$^;jF2+; z^%gt1m;Z&&isv-EAOyQk-h}VnwS3{i49=Nz1Am@m(L@_1Bv%`udR{W^>t%)0-R~p+ z`A}|Y*vJhEh1{z)h87x+7WXaF@UwXvT+S}z^?usY7X2YuY!ZqSPBZECf@yTB`ZfAl z9_Cfya!{Y7%jeEUa%-6r8yaVE@VI)0T@0-3&f@dD2Pp8=!5q0XtiIHU@mFWC$MV%Y z`C%S<$a%5tm2j50UFXH6;_f~%n@|5L;)(rbd4+Kx9~&~43(vb?;UgW)s`1DDcv)PD zeM$0Lchdwvca~e6%I~Ho^JVMZ+;5~J|LD5|YDbiisUo0iUXd_KyOHOuiDj84DX(f4 z_X@8MvczaNyXbDFUH6KxTlBm}Cx~Z&tq~i(=*dT7x=B{)jHG|jwupUJgDV&0d4Bp0 z{Q9s%aL6rzu;)52pKnMPG_R91!5*JF#|dMl+p*E31t$K}F*-j2i5W*=ax;>zT&m$8 zYdhHTs~x*kjOAr&llgC)sdU_Jed)>rKUlkdDU){!Z@$#dlk+D?n_X?CFXzwW>J6e7 zp_e1?Up|Wu*AJ5}cb*~*9oJpj{n}kRGWabbKlt+)uLRV#uNHeCPKcZ^N9H?WoZe-Y z$kB}x4s;kn?(02#cZ*}~Q`h-#!v^kZbXPE(X@dSLQ*ms#sw8H7jIjCEIfF`H7P364RJ*nT>bCjzei}hX+s0qw z3P|NCWq;t9D|%CoAK)$Xzw_Kx(OfEe>7|-$gbL67v{dX2>l8c4cCLCPj2qNKj^bV} zE>RYW>(4+YPl;D1n(@x56DaZ zyQ3_%^q#>(L&u34`z$u?ColCLFj@M{afbBcoj|ec>_5zUdy0n6*Tv+{$0QqdO|YSaS`x@s&?B$e%dYlS71f}h+ylY8fB3EhT(&c!($ynSA+-8M0S9!pq0b-6dT&n6h zPZ~V0o_B6t5Bp**yd4|`hmAqp(b&R9r1d}hcR@jkgQN!BxGiocrv{* z7!Ae;G4^mCf`9cThbbed_IoDnHtdG4w=LjsEHd5#?-d1N#xRo-rRVAV;6jSC&n9R080vdBjnsCE z`@irc9I%{LhkreU`?{~|Jdfw|7_E&RfrBZoFqcNWQN}cGr0ZR-QE|Vg7$nY{Q(EH8+Gqs( zPoivk5iZpxQT|DL9NcyRUgBpQs4lWn_w+yq z_mY=ZI4)M-!{u$QWb>z*`bP94gZBQk=V>gNDC(kKsSVy5cEqw@!oM*~k5ht%aj`-V zzUgI3{hvGImd}1n4RK@VpBh|V?}OQEpJMachiDtJ2Qzfv;QMYxzP2|SISs<(CGM@PKt|t(HQY>nZ7_ux%^PpX!AyB591O<9(#%pSv%p}Q49B1MNqT5C3atT#q@ji(6t(Z1D{f* zg?HLvO%Ee<9{r1+-I_~j9sTI-?IRR(JcsO!hrskq6mrsnXpcb}`G*guvN1(ivLpxY zBUiz!N29biT!~D!-ZguT#E0&(d;ud|xPIt$~lM~?At_;%NMy#{e{h#~K0Ox(! z@z#tuKOC;6yL_d@e-2i zy70AOt5`A;u2P+mc;Dv&t)DWEtTu?Ak8OtB@S+~4R~gH0-gzcv8R|iG$y+KMr-;m4 z2^;_0$w3P>97g*xQ>{XJEvu12Fv1}$`tBkS3Ebi3V9<7bFy=bv?hwTM>VKpX{U({BU^V}$87dMfkV2Tp2e~@PHdo1g}cP$=` z-vW`B;U=4HJhn|Zrwp!StK*Z=rNx@32QA|tKNa{==m;)sjpd=imdMj=fc2|bF5Yq( zuJ;sK_Ewjh-uTg_6Wb{BnyUyahT%g=$@Md=xuQoq_S&q=@9y>D z3Hf1olA$hBzBXI>Yi|Ur17~@0WI9iKAJ41(#~>qpKL)cCNB51Ot9KsLt*T-4QshJT zb{`-dY3gV(yF}UtzTofPeDu4RMJuL`Acvk7*w}rHxOXnb<$1GVnctUF%cD5B{s5N@ zQ-yhQ3}p5yyh(8n=Q(wjOUuJp2c7tVmMQ!8eF>ip>TK=0gO&HM=eeD?aW|DjHWW-l zUF~pA?p}@LK7}~zqt6M({rKk4joeiu7Y$E8(3B_RQK>tOlLmWp$MN2r`0qJIJ?e;8 zy9{}XdwaSxRe^$2Hc36I+R>Q3DzL2yfM&l;>Z|aUnw>Lb=>^$re$!WeJdvE?Ic$9*_<(wB%#+7Fl@Z-^wc;3d#Y-FY=um5Z--<>FY z64mMaIQlM>%Na6Y~ZrfnahKo<4si9NB2E~!8@J11 zu=`G+$OdeO^?GIDH5IOkf;;TLFpBf*s!{4T6X9V2+~{e*+Sxf+aDFm>%J_y~(YvrE z$Ap*uoXO4GN3xg9M4p)JE?=@kQ-1XR4u0A`isQHL6b^5}Rc+|V(|b<%pYw+bRrzUf zZ;?N>=4&Ck+`6&^wabKia`<;RCjG|vmz}w*Nh~EjcrE#M50TD!J|=_bV)kb8PErkB z!5f1A!acDJL!8D^|A{?m(h)g*v-v1ZtNAXeR!Y=V+m|*Kc89gX037|#2jM$=jdeUvV6P-+~d_GC{o0Ez#ZBhib0Jk=8&}@;Gps<~*qsJzpQ`r^X;?E}w)G z53b?T^l3D}TlB*v*<$_gNUr#9%ku}A@%;rYc-?L!Hq5^W+X1n%)=BNe{>qQGO-{%2 z7pff5wTH;B8}iWS7x8D?FF5M$5q_6J7&dMfb=n%qQT-?I--+GX@cJrD6|Bvy;1Cps z568aUYKR)s7D|czC}Bw~4b%BbE5h>-p|AqK*_2rPok}xyXN&t;F$TKj;GX#deBFKk zUDxeG)!Nm3NaVPq^mg*U6aF~)eE|9x^`>WnTd`l12(5XQxcPDl&TWcB`NW49>YRla zSt3L1y$q4Th3M6F6y|BKz?JI?d}W~)+gm-w^6}xqsilEW`|RNB{Q#3r9>ACpCsFQR z1>2K_a9lbGmtE6vKS2Dy?)VNq$(NDqnhI0v131=rld3khN5hW}sQGh}XT%5bl9e&+ z@g*3=@0TE?S2fK{k3jZ^ZHT+N3QPPRQG)j!x*l?d67HX(8QofG(;_1L+Aq@)`Qq}Q z|4C=p^`+}ybSNs~i*(`e4;t?ILpXrsD2p9HqvYXKf43_*d%u{EPesl*kxjl=MrX~o z(ugI=G_C(z1g}a)MwdhQ`G5a+>r)Y-DZ0${+VHr`5%^-d0~cmq#)?5_@%;Bo^s-K5 zNA+#29eaSEjmaUeM@Do!y@~cd%R^1|0_+aSp@}_@(HdKs;Ojfkr;SM>wH6O|c{Xa& zv?!|oY6{t@iOjBd5Z5Ld74j@}8aI?TWh!#?cyY%VHDA7Kv9rACcQ^UooM#NqFL0oE>afQxSl#lI||nexRLyR;VGHs^8T zSvG{XV?PH{x_IMk7k8C~xt|eQvR(4|=Pj+wm@fS-dC5Kch4Z2iWx4HzOp2Xw zl^*U%r!fv^k#g1_BMi+kEiM^%->2jEa0j0Ld?0cs?5E3J6yap-&a3BlXVt#U^93WL za_KGfmbi0k)?R+UY8jv09>t2uRl+NigU`?-efZpbu!<7#G>w@Y+%ALDzMd!D$arZ}-4hCUWKH9>eQCwhkL2(+1kXQL z39fN9zcigrK_WMIbz3|7>(Y)eX(D+xK9X+rcq@B&=!ER-wjw&_&=Gli1;gE4bO6mD zc}d^|d3xVm`QOgsZ%XdUK^8$g;FW^Z*{xV=J)SEW1RLXE#Wakl%tcbgUF01LV0Vul z{PNKUnsd&cjuy(OhxK^rkl^xLHg#n6A1CPPl~QUdj>VFlAEllLP06HT7}4587-O?d z^k0biT}}peG&u0E!5%y}=ZBOxHJ#Mdnqa7zL#wBqrB%XJf|-<*S_Sm%irdwV>)`be5+kV4}+*kk`2FZf4o;nACp@w|(g^7)s) zNK==_Q^DCs6gU4Ez65mQRdL7g;^8n}YPN}!gkLJ=q7SzEci^-wo!CR%5pPc$%!V}v zY;xoV&Du3za8!R_HddGYnoo02B}Mss8x@4j-bvQt?JJp>2^MoJdnML$(`p3-t^6u7 zH}?GXUK=SoeU9YlYL2~OlC0=JqV(dAEfuf^YI}u?cddu;ZYYwfdm$~;x51cMAyQ)0 zNLpu;jbmDe(4qJjb}w4Z&PA@Q+M>jjL%w1~(!juTlz9k$qsO6wt4O zq}|XL%?XZZS=F02ANAtLO_SKyXEGm6HpD<@9lkMS8eeLEk0#<5itmK+isc#PGGPu} zwi&U%+zgGaDfn<>2V&g_j$h8w0q?H3-g_@jDLp_yx)yh;^hM0{F1&V6FJ8SS7w?B< z;dA#GXqP-kPQ!1^jwQYtB0TH+uF}n$r>R|S9)hL?LZN;GzIna`o($uEYT|wO$&z2C zWTP{kU?IKC4Fe#rWxI;-03>j$u==z&i*_m)^j_ zfp(}a_=)Ksy75XEH`H}qhapQ_LG$|yxAIXsv2qh>9yu|yQBscL;uv(6QdJ_KX(w<>Wj)aH4C8?ddujvkge z;$gE3OdUVS6k2zY>QQZ|%rnKZ3NyjU{)dk5bZJ1o@T>g!MHe>h!lL^Rak8XUbbni5 z3g13x0wTAzInv|yU45#M{UnU!3!IsIct*kAmIHFh)E*fWUdye{G?Ywz&4 zGjH+pzn7>H@0>$sX%tfPl1k6a!j3-@mlx0AXPPs`=W`OKp0q~ix7)}m-5nKs4k3AO zKinBNmYVz5OBXDSh2#4GZN2dVJ%8GAOKC4|XEBW5%=O?i-A}Py_hxn*RLUuHJIPnC zkK&RZlel746lxWA!eg;9U!U@g3iPt638&EoT+a66kU=w8<>g=8y&J}l3e#CHX1rw4rw^SE*-OJZR`H#UvpHFH z3{rQmqN$qcv~Jo}^t-eZzI~i{%x*_Msr4UE{i?*r^M1h3Z93u_&e4Z=m&w^X4i=p( zV1C{Wx9^JcJ8}U(Ke>T#9r!MKYRlQzVJvU-F<^%P11?OR%x@aoV@-T12AKZ)pZm|s zUx(nhwHed(-oU+~l)8`9L{TRlxf)3p0G&T*8ZxjzPBZz3lhIZgEWFch+u%k*Ep=+}M!L3nDm@zWzMtn}>^<>Z_2YokTH;;@}m>g*i$(NdwreV&(>CiCNM2he`I(_Ibsl}ef(NJG(^Rc7(wBsj@ zwFuG-#hpb{I5FOh$6h;zj;>00@?bLk)s3TpbxD{QJOXx$ZRqPq7Z`oKkL|OYk(}Cw zH6za=-zQ&i4sGP60n>QwQU^?JS0nlU6yNR3i^yP84DAUwhr+Fw^i(m69o&A1+0H|5 z@NS23g@e%bp;$#;GRC1PB3COo>w9a~v-*;|{OepkoAq^}`%^|^vz8gZOI^Uj|BPW9 z&-s`X@|vE6zk83Uyegl14-fl1LzCDQGxs;OMiF~}SI?pla$W=)_ zc>jRWqW{H`wk_R9NWPBAhj+4Sp%(XOOs7kN4bK$aw4+5o@KdLFT7TrRze3D3PhR0UlLs};hhEnZ#5c&0m!-|GM{QxJl2yD@ zo=HW^34RMkLFd|BJNJG8(&VLz@Vt2nBbOdU^S>Qr8rGAV9u1-#i}uuXCF(V*f92?M*u672Xw;e2 zFDlEwO|@sYiAvbMxCh5?DWHU$T3JX33v~ISOGBQ>rK@>5^fsmqo;*_LeD5*z+iM05 z5PKMntA1>49>xx~7W_p)9cgNBVf8D6zf@ciT!seEaG4|aXEsv)eO1ZNz17b8?Q*uN zPDkUmkJ6sHNyz)x37-o5Y2)tzY*6@s-XD6iPudI~B69H;@5Mpw$6DHE?=4N(d0eKi znMUU=tDxen$46@(qQWi*xx(Y1*XtZ-csgB1jAF5posYtab`Tp(r*XL6G$DKLe;t8J!q-75M#*b06ZtkHk-TS`fH z!v@_r)arI7x11*WUbP(t*7GF0zkX6{9~V+DkzsbD=xceDgxbz3GPP-%(tjiJDDu`H zYCUI*fx;ExCC;pqQWAfBki=y+MR3i0gKqkITw>QiF|M6)m&d^Bp)G~JN~V?@pXqUU zCLXO2cjke&5jr%6+Q}|ZL`WsAJm>}QOPBD_dm0~HSVGPF+=UaN4s{8qr8lE{OH*^A zrSbp`?EWzx(<5tfV@L-Ms@cq6)ic@mZ;PFq=xEN)>WG|wrP2dubsTK^h}jl4boHwO z(p*ZQXtzmxc6y@Avwf&4E0PPZhx%z+`&k{!@2a7_ml4&kalr5MH}FzLk)K!@ z^7!CC$f)y1QrBUe=D3)@#J1pYGS0nPfhkBbx*8>`)bCV__!77UBC1Z{4p0+>Kp0$iLqF^z8bajV(~TX zCyw9l%+W3?GJlU_6ms$^GR`IP4%uOzy(peD^XH;^bO6@B&L-W5H_;N%jyL}8%JM1S zX~3pvsJdLnk;LQp^Pn9++iA}mPbtEC+ZAlcv1OkrM)W%>jD|h1g!156bRQ=2Flq}} zS5=iZwpB%}V8k28HpmVhkfi;A$E3sK49KSWExia>kCW?PO2zsYBo8A+*|@-NBJ1}V zHJW|d9f(#u00e^=hLs|M7kZip1k`%qGy5w%N(c1iq8^7E=FWQRR6CRY9?su za;WO7io1+5Th6i~%`9#5&@m=ik~ysQZ9%|HGj`JPMNLEtF1*y|h?jc&e&Be%KPa33 z{@ur)J|5wDHII-M*8pA1^B639M&@dQ0T?KpzrFmi`f4{i?WjkS#t$LQ+JmAyTV$l} zw&AWnoAB}GAa?ZZ%X&H=N%g~a9GcXXwZ-@M_pBHkVhwH#Z%-EtR*_Ci4m~(Ig`XUY z;V15AaB8(Bgz6XvKdnQ}@_SMMuZ3B%Enk0b$dgWX5{{VyR6V(jmz#H^M_)B`Q!v1g z(#^QPS&@sD7oe~6CuF=;=YDm?obUFQ&#yeeIVGaoN!)ua8;USM;SGH<%Rs=lPXBZM zyz%bIhQ4B67GFyF5vQeHw|+}!?kbUsy2zhDy+L#ThEm6VgOHv0o4P25(Tj6lv`a0P z;(v7`t2r$kw?N{HGqq8E@D^1(*QMN{(K03Bn-q}{T2Xz928Mow%Oc^wst6aJuYtHO z@{2XUZ%B9Env-dOOga~?DTDb)^n5lDyYF=pUYcm?c6$IRiaEsFsg9_#uc2R-p_FWH zPO>;7at^AHT6POa9@p~%mV{iVb16}*_9BRV zdI;WcO_0az7U?eK_WZTU)I$J+_RTJG`({>fdPh@YMm?rxnl0 zV&F5{)tpO5>}Fy`R&UDMlq-GDOyfY$9{k_jE4Z4eOGyQDr7it_N$>Z2$d~@JknbLO zgeR|mfe(4&{CVsP7ncF}6Bb5SYpta7I&yi`&ei-@~G0w?=4kd(Q$miy7{IwkkIMEY(XSszX+Y?3x@eLvJ#D@wMLL}o`LOQjcHMz=hTNrysi&yo&~`E{6@CC? zCG4;}CCv;Fvwl1#s~J)BqrWmrR}P{KpWgIgb0wyX6YuTX?sP`<+dCU=rYRc3DeBEn zC|vFZ>j+oT|My4I+G0Ytn>=O1yGLQC?+8q?oJ%fqFG+LC4q|bq zi{$B5D4DO&l7|(B@t}PL(A$4m)g6HNDxkA3kx z`81C_(T8tFkHEd_yY2paSZ&u_tSx(!AA^%t=P0Fh3S~CD$K-AAq^q{tlAYae%$ufz zlxHK!;ZH6dy>v;6{F5oA1|6cOKSb9=>`-Z0zh(S2+=cJIZ^KhAEZ~n)E%ueRM|%xN z$#k@fWHsxGEbU(r+3x5=+b$^3arX=))&xQ-eNJV=`wHd69s2l|mXO!R#om2~C>NgA__aqY@ZNja}G)$iUU*c5l} zw@ZObZ;QLe?8PYE+!r4wmQo+p>9lHfDVg1lyD(yoE*=SAQ2D|_8n3&DntQIH z&PRWf^528h+#E`yMvKn(B@zx@G>58*6{Q;QrB#kEa5BFH>HmGF5Fcy2q&EC+-w9|u zX@E|f1r#&sC>ft`plIJgJfx`%_K(d`mT{CW`Od(;;6>OM^i#^(YmJ9VA_E>50^O7S z5nG~zpk-o@mo=3wM!sb=*Zc5D?Sj5D!*G859XcpD`yV5xAXW5Av^E~6d7UaKdfRtY z>U(kV!#=ER-%0}thr>yIDAvo@(+6?iYu_mxX)_OT%+IyF*^hXo{2IQ`8HlVt+T^<2 zf`UJFK$||sFl}$hd%8x#Pn_3FdK3J2a5B6cQTqA%uMDzu)Rqz_?s@X=X_#^dIA zQ0;&=ZIvKZl~Sz8ht@g`6I{&^3er`<>QHMG+rOc*qU+LirxUbPuNzh@vtf4k9|rR)8Zx|ru<-jcg$(%6hv&oms?Gr-L`tEpt`2V|evEk)$NeJ{%$4X;w(Q!Ehqp!rGy>c-YYgfYO>o+oONs(S_FFUIIwKau>`8KH-= zvE}wnoZ2)JWgWxeB+l0V-@oCrPRt5E?nT|pHoRoR6J&kQN1vVA{8_j*#)|#{^)a>x zOw+(0r6E{y+8?%Q?J@f3Ny?tzLgfib_!RRNW0&`2wY;_TJL4&tO|B!U=OQSc%_g<4 z74+dzXOUN12fLBkIGl5t3ObqNcqbVg9TRX?QO>W$yd`>ZIM6o}rD=!p(diOxxfntr z*-11ktd@MXc0!l^GAz8o*kUmpb|*q{KA@WXwFl!v@B$1@G-Io&+9E6S0&~v&Ce_}) z7}dHN_LV_Y`FI7@h<%}x_ElVu)#9jLUX$$9wnHVk{>XCa3X>VjFxP=k^`=)?~C2cV-vKJ;z z%Y@E_SupY2furLCpq8Hn)M>MjSir1vBpMb~;6$(yx)s(L2J_tUVC5D{+R+|*3rE6#4bht50(ukGK&{(K#2lysrbEADKtwnA#mX`6 z7USCRnb>@;4*j&}abQLZI&aqJrFBKvWU0r`x_yI_iVl|_P{j6-A?Mx7!CTU{6U}80+Bk|OJ5`OYVOnK8AB_5k_dXtv$>$ZsR>I?MR`aJeT zrQ-PfI*c++MDm{=Nbz5efmu5x&8Kh4Gg|>xJ96k+`=^xV+Y_G*RmgsJBBf8eM5BYY z;OGT+9O*3XU!7`bd(mOSRTa3d2*rh>`zUPK1f?^35b<9V+S^TJ^T4*qTkL?<->1VN z-A`ohE8tbqN1UAp;c1)=tn?omsgn^o~ieAe_QVVRV0=?%s{04ui)%_@JaNNXtj;| zpYx|Uwgfe$Hmo_tgIk*R2>;M@9(edR+AXXhEzz;8p>zSBB@uXWRm@tth12{qfr!W! zeRB;?Xtn(&dmeldcl=}dO!OG464}rW@Bc}SYTdbdbVnYLUx_`X1EgwiJ$$mPg!kkG zsjkj}1`kc6+TV48SIlEygJbw5+yci}wC7d(`?8i+3Vxr8z|seuVU*QGBa5b@DX;n4Ne(p=)i$NtE7t{|s{brILE>ltKO&VewPMuD* zNS#Fz9Un#Ey&1*FRC@E^p9Wk! zu|Kz!C*bO;jWApO2&F~W=(44$=;jjMv9JSVJ?JO&&4tUutpVwxN5VLN5?hY3=F1gv z+!ANOOUBqk@w_X$FT8`mOQWz?!-?zrh2#Co*)(-mFOE0V;W4psFgM-EdrocSz@`*_ za?%|ON}kg>`}g#u;*DLe0#jL`pDI1xxrLK|#Pht25DvMlMb(@4z`Zt<%eO1z(q1=I zl(fT_2iK%TkzX3oWXQR$QAk|V1;>`8(y+=v_Pi%tycK4=!l*4a7R6!ouT*^KuO+-g zZTNKeW^6Do#kee2UOVz2e|K#s|Gnx2Z;*kXmUiN)N&R@Tbr)`V^PRuPUFMX4Dg1C$ z5gsT_;3K)EJhFBo&%PMW&7Wep&7)+_vpm5|FWzJ2lEnz0mVmMoB?wwQo2%zWb3*)D zo}urD1tBH4b>$%1TduP^S-+oNi?h&1IzZYdXTrC1BzjGMC`BDQOC72$kTv`&^z4P# zuXP!>zUYpnk!kRAoPf)>RQSeNTYh|Rl<2D2C-xO1_`tvoxG+A8&@vEfmv6(#DPCOY zavNX6KhRD=W*WOMM1SuE7@lN^e`7@kHt{-HeF)>i-`m))|4zPnNaVBs<>KbJ6XYzq zggpis^YHOkG2!B7>Rhyt>Stxi-W|xGF6$3s>b7Ipkr|A8S9)@^{!-q2;WTrkraZT= zru<>sUUEy}X0(bIowxt>NA|~y^ffhGHmu?Xj<27OO*e)2f4hXc75%yF?N#pnT9dtm z2YuUvc2Zfvaej8_4;xHUl9$7O zm}9TY*!;m^q%InT=qLFYVpb$NP>ea_dkA0Ci{icOw(z+2X4%oVGU)6UI(v z_fh%(Gymzj^E5jue&r4wSIbv)oyadgrm@Mfa8ABsi0Y8l_=9UGTE3G)Oa{}azUL_7 z{BkK|gAz7ew84xu1*GqiiT+}CX4XGk(v8rew2>8b&ioHe{b-DQV`Fh=Rg>JM4Roq{ z8n&Fy#N`wnZWkHIlb`1Bm-$~fl(~3P2^qlnW}7T zMPG~#m0#LU?h_u-*rDsuu8lIcIx6#1TXo_0`3Ae{D#Tu*XG@%N|QyQQC`^$Tjfz>Ml%c7X4qt zuff2%Ezf|OEdP7F?7yzxB%}4OWlo}pbmWj#Sl53F&Cgf`^EX?O9b6y_53-l`jfxEtMc2|1rF_%^eFWZ)5A28zl6TGugjJKaJ)e>nZh}tqs4r ze7s4V%;&d^VD>-8b=ZQmJvQ9kTh2{2?eRP_3Q7tSV6xyL2A&P!ptUDC>XV)Hx~5Q? zuy~gAT9z#GxEHzew9lMS?Tgwlbx3Pc=!R~cRCC4yc{kb%c5yua%5z}ftRi-bxQ5@O z4#V`32lDqX!>L}W=v`Tla+64SmaDLd+dy93PtI`;aa>h9n4kCWfa}#8kdpBqMlT5B zZxc#+{R5r<;lJ8lFBiF7MtWx@u}iFj>YIPKmKg!fvqrr5*lLdVNaB7i88q$71*!7= z4O$kjDt}&cpWj}d$L5EvxGMN2|8j-gBDuSibh*w0eFaLBaRYDd1a!Vuc? zk2L&VsBpdXW`{YWG4l0BN?2?I{ZkcG>VBOf$1j)f$TpDoiA>>S1!I0`k&Iua{cx=! zTk3MvosxI8q4^tkkWJfzQsr58`AmCbd56o7`OUKrf^nM;<$Nvdy5vDe&JMu-JOdPG zZl)6(+9TLqnajf!c=TIkxXu-aY$OEl2RhT|(7wO;qoW zr%%-nXh2+F`DQgsx#7J$_Wz#1{(Ii@?93m0$?X=a&+WtJnW6G=CX0ASwLh;mNx|@z zXY7AtHUB=L&3p7ibT=il<`JzCQA4>A(MRtMoA+Lx0zbaF?d5Oq@R+-?l@=2QCtcS24xjgtt zI?q0#!C~#EN!DqOal(&yv($j)iJ;%_ZxDXJpn##8Cq=dE1d<`r+bsU=tK zkBeWzeJ>&_C)FvRqS2xQqr4WAx-!`jxxuNLd}um2zY3Zj=Tf?ucYD5n{1KKKFnd) zbmZ&H%DCpgw$i;D+a&$g-6Xn!(Rl7SjCMVRuOxDB=Vl`KQ8-5*nFd?ejvT(^JJecF z0$PT6(DX@iuo{p5&i|yvi<%@GPZ$34ViGUOSjDTSAK)$hw0T)Ob6)J&k4rVW@rlsy zC_6k*zInNhyvxbU{B%qd8Si$5p~L_3U&a%bqaHuuUp$y&n?>j8^Fdtu#ur)p`jFLU z2kdx%3#o_BAkfkY>F?Iko%yrbzp59H9n?Ye-L#_(w=xlVbUr7{wWW+r_ej&|FwI@< z$12(Wd{na?$6Q|^c?*ANM&w|am+!_{YYW)V&!U_uqTew+AHT%AU#(jPw_A6DWmgXH z?BgTYbk%#b(Yy)6)L=TjBL>c&w0VS)=m+=I<2t)}xF`6H^|CWK;~yx#^ZM|=e|LCV z{zcZEHyX3HH(~un=5`+^keejtY^nPsckfVc-TR%#rI^bTSAOP3H%*Mv1YSEV&i!rq(yim-ZzwY!z&eD6>W$(d59%bO6RsqaRV&Hnvo%;Pqp+`mT6k$4r zwWf|^uUqZ;fX-aDzf;7_qAOZ)*+(lY}U@e{B}3p>dXF-z2`Jt0~tA>c_M0 zAEC#JtvGG*3-SA-5D@g4M>%$r=daa~YsA{&cv(F{ zzg!c4mp9b-z_Mux$5+4P6tfUEMi`G=+mSmOc_ZC4>=WQSaV z{-GLLLWjkSvr3#>$GP;=H#Ru)k27Z6pl`?4qRWwLdUWrvozu97WP2?d3RBNPbMF&e zezF`JU&l#Z#+;)67KD-dSTC-cP|P16zv8PCw~~@w0lJF3d6m;g*xg>sMU(e)ik74N zWI-wSxT?opJipTW!CKr{F^zZC{@{Q?y|{FVnN)Z&n!~?_af-)-Wp z2EVF~;q$_U&~u3upYklHk7}z2^qq!_M)AW6=;jYVd-E8IRy`YE^XFbCz3`wuapihL6+u z5Lkde9@{Z7b`m}R*@d+I|D#o-tE8wqVN$m~izSEsPq61l0v-?Lca;bg5ynx8$*Z zvb6oM5-dmkBkvduSV{im|5BH>)QzGQeFHG0i;?hU)sx)KhI(sEruNe}(N~2CYI`q? z5|)cT?APA(vWqF{XiX*KG!H7VnM`(TeQByjSJY&8#g%s92zE%LyRrz{(nDlYNA_e} z(Z%DU;LaVMXe0gaTKZUUoCfv1MhSywlj|)@y44aujjbwl=-xci^jwS+8!C`tWyOYV zKU2`;adi6L5_&mk4_8(C@k3h=9@@V*&ZwWHFH53m>+t*7u_y})-ji@YS&iPb%aiQ; z9+76>^~DiOk^gS11N+f2RJO>J(mxHPqy5?nzp)a#6hf`VP35xYF zrQO?Wq+{8kFg)E4MZ$~i5Py!+RU@h7=N+n9JA?23>%w!M^x{Q>AJWeOlj)FuS8BWb zFxK2pz&*hPkDK3*PE9J64jz6gxofS&p_4HP{g{RLvPpP!?g-!VN@mk9Uo0|>L%0At9DdeV?y5sjiB8%DN;@T11Wfz3X(&v zlI7atVxO#n7egATznUuoo~zTa)t{uY@Q&2J$64A_7D21?-ciEd44zY#$|D*@hp@wB zvNh;SyZ5V6_rTYpk48A+_mz@GT}LWSxhrX}x*)9!x=16JWRS|}1R9hOC)qWfm1-s) zkPdu!L>|>g$Yc6vTA6elA#Xp!uB-$j?yQnKEO(W+-X1A$z2BBQmbmbs5##ua+CypI zv(3`OsAW>baXWNLkU{%lM_5$u;NbE8yy29XznS$!gT@H-7kp8~f=2jQJj5$E(d(*j z#rN;jAb@t`aK}>0j$1{G#av);+aA(k`wmiGWE<(ii3Fbe^DviWr}B_}`RMt;5$aYG z#qQso(WV|n&jn+xtB>%}L(n+DAG1Wp`$*7CC`aUAjp)R19UPB778?=ivjH*775Kux z_8gr%oYnjL@|^dc%x&bD+r2mHX7$F1ml9GAi@f@W#{b#Bx9zqahR>#8t$1I5uP@{B zzOgJFTgWjwx8P;2%pJCgZV_u&oXX!SGRd*{kb4E?g3pY!+yy()TN0AF5~IF|o~H_u zGS9R~7WR*&e6go}{4ok~*UsTsur+qf6P*l3+Q=-{!9^YM?yz1)vqW!p>+V;$`!Jik zozmxC~|dGM}DLD^D%Nl~Lxfza#b)K85u3l?TN-tAR6ul$9S{8K!Nx5StS08Esg)b~ zl!i6;KF~*UQj^H*(pbo<&ftXj^EM9AVVUD+I7PJK)ZUt0webU8zUhl;EeV)dxQk9W z@27tY<0$r30IFXt#LPctcwlFSQd7|Z`atwMehn8rK_OgIr@+Cyr5!A?-wu@}CA*EHBQlGD2on=4-AWg}*Dg2y~Qqb3;f zo|WvQ{eX+7HSn^Z5r~Kbf{b$!UE;+j)U)_yU?Qg&m(gy?2yaH}@X6s9xLaREp71qR zn%1j7|L!u3|8@@IqTbeo)19F7=o+f+>TqC7AZACLA{~ur1g4i_#LsRpn>QS%vW#d* zUZP;8lX?F7LZ0+Ak^2~F;?bx@P%hkxhj#==+5AlO7i;n_!C*R#*XIkzWUR3|mCs&t zWGBH9YzuHh(3Hn@8A@(yxg7IAwi)+q7A)nhkMAKx z`5*3SyvBny@928vmj6f9c|YX%hHbpGv}uSaRHT$t&wU=}6PXP%BC=PqDSH$`L)vLd zi3(+}P-ZEKY#AjpBeTdTtM_`}pWg33U_ABQ_jR4;aeO{-UM2R6SsyT>t2Xz$AI^Mc zA%a~GqU`5&QZj!_<~4S>x!)DdiI;F?l??{m*Q8TzBdANnPg)f9KvM3%8#}W*3jdA= zpFHn@k^RPEkH&PwPxqw_yJcANT#2_j9pr=fSTqDaen_pIwqpi@@O&d zpL3R-*NBLGw`FwV6?5T zz^96D*z5R`ZrX@5Vw*_1@K|(rMi{}sOXPdvw^MHC#h6=R#$S4+vuesd?pSnK{BB-g zKZg{a6qUno8uy{v{uZtWe8Pysw&+!H0>&5Dag6C%_^Ld>>5jU5af2x=421VWa{#Lu zMDl^_W4Uk10{~+&@^&f0vK9(Q_8A&GqmD)_7=}!rEA(TsEj3w2QqG7%OdBwP>xDmH zw_#6?x%&%=&eq6CP+}qZ;sL+k3IFj4^t-9S4O3?G#WWwD{dN=UtQ)|cs(ax>^=G`_ z*OSMrjmQ3@iD;Zr4Uf3{96CU7&|K{0EroB;;j=j}g%i(=uA)_U3$Vdb`9J3$<@-0$ zW?U0oZ47vT>ul5vbw!}2GUTbrf={{`14QPwcX&J67n4ZJLz`&0su~^5yFhIlHc`sX zQnY-w;-s5~d}@m^Nx@5L?NwWHe{h@jI2SKO?MgP+Oylz7pp{BPa}}Va zZ;w-#MvC3_MnqMf$CCk(c&Zc&N97Xol(&bf$qBl;@C&Wf569?viMaE0CjyUm7oD;c z{LY9)Q~&vrZufIib-5+&oMsFAb-mHzACIicLh0fSL$tTg#+v>1xN+nXU2{`M-O&W- zx2%WNq%8;-sfO7NRyeo6A2#;RAls(}#Mf%*!@4?pRp5$OB0K1LY#4W5IGW=ut=agI z@LNS|S4Y;{B*c9I%j z7R>I$`^ICCTNJLeZq3`?JwuI>D%OB%uzZEtvj%oGI+rM6yD$(u}vcI=fl$E>2GwWCB1XDX6I+v8^D zL0S-gi3&y<;HmO@nEX8gi`lENtKl82I@}e!>=a~d7p_SqTa+p9!pj}?Xv{F=IW3W} z{8fR~@2()abqW3bT1aLO3Td*wvW;PxO6ii>Vy$iMb%Ci*3QFSl^_E~hzFubBHpw_?xjm6$O^ zhfkj?rd9Rru{U=l(sD$fN%MrN1OI|%Ato{v_xdct2Y!#iKO%Ad zTMJr^YlL}qJ(9%tdho_NbdJcw@3jYTuz!&x{~9Ol>ONNLrZIpwe(+;;!NAxx^|7@6 zz*&lqH--GQ5??-8fF#{Yybesn-znQ*XL=Mv>`JJn>l5lVu9Dh4Scl)@dF*BM7#i-&ilw7#-pZ)u}&gW27dIx`( z<`&jCSX^M$jn{+fICZ? z>Ezeh2$?n&2RH8EPmYDGJS&=~3-59Avc-Hv%*IcYMNzbRTReKBjcAAc)NfA__Rw$4 zT;$G&wus0%-CP8(y8EPv^ILfe^T@=dr=EVkS>w_71rffwUgg8eP@|$4*y*{ zihG+M@bFJAEkYN7EM8&oEMx?+ZSYYNXl5z1BpI-K)9 z&@oUeQ%dy_r`FiEb71L2>q?fqwaOBasK==+}$AlU&g^0@mJ)v zh7H2og|l$D#t1QAbGXg=GXCNp&FT|e*rReapA)+}KLKcK8gP;W_l5Acnml~^@e57o z9C`V!`_yzbpEBKcldABbRK@Os!Lx&~-!42U!>%Lm`$@bQP>e^qUr^S+8!N|+6JDg7 z6ux&7bu#VFQ@pye%JIRhEo*|&X){g<@5*`4l;qa;4diXLE#wM8B-f2&`O~(221U}a!yV!U%5-MfS#bAVHDsUUvP>Bd6Q=PaNu3)`jbpz#itM`7fj*3MLpO_cpa27& z+I=tdmpbs1dm}LP%ONx-yuiu5?dfCJbc9@S0OJ5$%Q=A zv#&h;o1Q%U<6YrM4kO(zJLtu<{Y1*?{BrqORzF(9Z(srWt5O_|P-WLYQ@H5rN7k62 zBae#c%xZ0dc(#uvpWP_td~<)0zEuka-4>Y;;V_;2?+v&mKar7ASCbUL4Q->VUVewfNpQb2c7* z23bajaoi(-vp?5xj?MvA&fCP@|Ez`On{J$Pdn1>eGn7B>^O!f3)p3}c14oBH!>5!8 z3<-N6or(V~JzyCP){*nzqAnZ~ZGsAsU3`3ZJf@C2L{nOMaCpN+zLT+s&v-1L@F}40 zEf*&e_k4FWq5B)j|CA zNCGl;T*tKLCG4g;SIl2b`FFb!eCl>*y3$S`t6x`9aAB~FQ``gibiyOL;Bih`zCxK?#=NBO$*~S0dzs4H;g;C9Lc6qXb zuUQq~xsD6G1y7X6IbllTVVa>?9wZmTPpfI<^QlQz7+~?@1>V(UFqkba+)Bac1u4 z!A1*-TOZhiH)(=3pm7#c)f(a1_5-&x?&0BUg86WM92B&KoBL-iE|2OB6b8ey${l~z z^|7;;Ds-ItB7VkP)QNXbf$J?8wA5pbnDyPb*oimCZe*|X{dl3x1JtZ5fmXs4ynk(q z_m5*~UZ>mipzsa_h7ZPX;k!6KQ5nz1n{&U2?~wl!-ld|^c-3P)9aW2vtvB6`>?=x~ zb;64M;-2Eb(+T*VXNbWQHL-d{IQ2TTPjXPwX0sl_{Kjq`d#>-#7xpxv&E`$8^Ib^` zcSg_=wPZ4Bt;R=wb?5x|&#`EgGu4U=?4T3&9CjcUISKvY`sfQvj+S6W&~ntj&7sM? z)i`vGz@F{KnK@^`U&ayfB;ZtpkbPVRj<(0T-Wb&5ym@K+dZ zQwy^V3sEE(19p!_;ozcm_;}v~4=ST@^Ri$enY=~k;5ux1=ZcpDd%?f!PAa#2C-vN@ zLH%ogNWETd!je)w*h;|Sc4AJT+l4M4Q>BrSm9V+`2|Ii{@)`HxP(Eda9h$Q-$94yL ziD&BM`Yt@K_f9BB35Iu+6JA)eU^9L~r-N^>%CZSJ4CVZwhqy}~lOei#5{B3u!9OwQ z`In&s6GbPi9c9EN9tu1m=_P#PmHCpz49r)(P2CHE;50N6DcgbQ+cI)`HJXmCJVA4> zY{tX0`>-?p1k{3;z{%kVZi;h_=Soek)tbiY4vTr+=|23mz7PDT{+6`Gz9eI92|lgJ zL)(A>WTc-cW;dc!@OLQg{QN^Nq85`!^?a;+avgD30uhxM!~;m1`N zc36jZ%=O|0)6rv>kxn<%8iaOu$p6*!OEh=%A0vb-IN8+!uqci2pX30`*a zW{LY7^x)@a*%&_T5tg4hhNo|W@Zzltt$8q2Hd0GZaHF+&%gr)u`|XRq*MKjVuOiBB zK6e*BJJUnObmCYB9v^%z*b$8g?z8qb{rK~MWNsC?i64tQV(r_F z*b!TRfQh4|+uwVV`lVxVExJmd*qiua(6 zEa&NN+0N=+vbt~6smiK?HpYkpMXa_=^GAQ_b5l5R{CaqV+=u<=HoU0&RgP-a%oA6v z5Z>=gP+8Rs!(1Is4_Zp~Jyc-!M;n*Vjwh#%_E=#u9(P8W^03$Tyn3WLmt?-AcNW&@ z@zWBWW&YB>(^{z0YQyR0H%g~hi;lV1SG3jBr$^Vyq~>+crMk0Yk(hrNW1DojENTK? zR|z^@hiD%Ay}Q(SM++(k)%n}4dD0_Y3v3v#!UvvsNV$EN(jRfh^EJpIEnNq!66Z71 zf#O}RFS@wm-xC-5zy0=q==#2y3`evP8E7R`iuZ@pQ)@~x55vSBZ*eqh2d-pw<SP|;^e1|I3S$w<2p8AvWeh!bPnS! z9W^=jXC53DyTbXX2ews*&_@m7X3|eU*N&;!kYUe5N)~XxXPL;(oPf;jZjdkN?s(qe zq@&4RvF|^ajkE$awy!qjhl&NXW8yiiH!cdJmgbv}U)|9hmpLFPEGj$yiW;?AD)AHGTqay$z*34P9`o_5>bw5$CPpI{f8_ zJD+(f&cH|Q`Nr98-1ll1SzLBJ1x^}{hsRGqPq^a?#|fsem#bq$mlNbI`X-Tw{y7A+ zOOwtOw!-&o1znmit9o|Tt z@13H>5AR4i>MB^C^9=dQd2rG-q{9Q=oAQdM{F=^6!jXFQ zQv-r*K2n0(SF&D8e^32OoBB;H=;IeE#n^ zoY4%1dx#l_b`ja;YsOrEZ#z5W2TKnc_DHE8$4XICbNP7xgWSCLA@t8XL&wn{r|z^v zoOhjULTwV(9bU%k)hBRGsGRFNe?VSrEb6+L!TZE{>aU$j$8Id8o6l zp|Ip?+T8MC<<8&30%(@Bd)s!cP2tfwCo6Oz$!(NfGTx=ssI zF44>Iz0}}r%{Ro%Wt!DGmRz?YuGy5sCtgR*)JL*Trv^$d`&&$MjFm}nPt%YR)rqfwkNAaBTr=)U|`w`z~yux{yy=hT-0m5qKCP_^A6vK|Qqvk+OrB zn|YpcTgQu>WHROyev#%KTP0y*G!0yDiCqB&;`?+P*Gt;+bzZ|`2IX=^@F;1<5Krtn zUW9==7x4+P)14Nmz)QXzMaMtt9JH(lFX;D=4lBQ>{!?#C*98mLGvO=x`fBjSTz`I+ zkie=<(QMrB2J++Y;mqo}Y^{2eoVVA~PTj6BhzsRqNqhO}J@Ge&8B(LwdoqnxMnPDG zRNAVH419c$vf>8SeQ5>dQGU>Axh2`FuOwZ;3(j=t$%D2!i%xMrZaDLh^u%ZM&Mg?J z!VS3Pqy{&C70kG!!twBGCVSoR;JiZNQGfdjK5oXW8I!`TtQWIvXM*snO@#HHCzKWB zhfQu@B?YervhG(-IY|(C_4~N(IFy$d%@Q2ge$+ALsnpqK4hDAG4UHblF@5KJnm%SV z-K*|R%Ii$|lgA)l@70N~WVPddDf4(9MX*`*J6M<9#0BA!XxKHJcg>IHrYk;dJ^L%p z7=J@m!~&el-c2cv!Y5X`fo8=Iq?s3|B0Eetb!zXSc*8sNd9)jcM9#5(R0j5|DROak zEjC}6#Qm=(^Fn38mmIQ@N2kZLQjoT!zS~jqG)|N~i*(@8#o2uN=u$QqT1UUlyQAGM zXOU08M$NZh;^s4F4h=M7&r!^tS8CvQQjbP_$)P%%UKC{EjxlE`-&=lQc*c_lKO6gY3N@P0%F z^XAKeyxnLk&&bcg=6k>B-TvjY_F6LSzS|y)yf)*3OikL77fESX3~+yV0~wc{C+F^E zbn;b zv$=}vR=uRK&3 zeLc}7Sd*Qm$B<_0P^70Ez~SZ+GMC$8sA&TFmOrEucE_QpAl$g*6Pkq54b=lApc z5Z}d$ZBB^U`pF~g)^jKKc{`e)ObesU@7lnuYZ@|l+>_FNycY}yW$s_?iQ9#$T$AL? z?Yi_ptC{DdZ~GEtL482MA7`Vkmm%L-)g%p?(20(3>qwiePRTs;5=p6WKW0p|qE}l= zs3PYV8GPwaiT8{!#G(`r4vj>AF+-aEtu17~eoIbAG-zG39kpIJpMqLK;P*y_9TPi9 z>Ss1f4m0#@zDUd%znkR!;jQs972(YT&4lka=HWQP|MS#PTXFSu;V zI4DCw1y)0jkEmG$AjwUx9tC0d+yGHGSqduj9jqtb6rb!@IH zlSYXQl&g_3h9&GH%X5)vDmD^R9235eJDd>NOaA`lG}+HD9i{2J6RB{q=(O2C;&Vxm zzl<@LugP!WDGm`_dU^~w-n&3ak)jv!c{j1nN?2|(W3vlKsKbnnV!vg`XKkCvab5xz zh+JvCTwf|Jcp|M_6;IV(KIFZ#kd~>QCZj`#@cnZ!6a*)7X@k4m+}BHP_1#_W_*&$# z7pBT@c)ykZnD2qoG+kbnK94)(HBkD6U%2N!hnrn`Q=hHr6z+C{^mlB*k48D4UN?>R zx$F~t+V-rrFpSs9h0nE{8l(lma47YYjnSPYtq2Q(#+@u_>qsxMXwIXWk$d=hrjC5& z3x9cyl_GR3CZjUv37PIslM2;)!*^;ev=8mZ1~qSXYq-qUzF4yThjy$G@E8+EEkR8A zV&Plffa2~6&}uPZ-wQFUdo!5Vd3KUl=OoJC`$v-9vqY>eaAZ&Ad$jwioye-F{Ad5Q zqs=j7ZK@MofjDuuoh+}I?jT?L(MxV}VW03r1hSjb68@W^#=C-ja67(%bhAS#Xih7H z)a6sX+cO^AVIwcz+Kf$AM-bw$6yIm;r>akLadE9RmELlvE3QiNZB>eLo0?G^s?nB? z?^l%h9~>x~e`^FSPK}a6FB{8>RQF(a=rAa+6O3UcN6hZifou-i(gQL3>~Za>*e5T+ z;Px-1po49tcg0>()#n6s{<#aq(O0>uE5 zZ$GLi_lX_O3)-0=&`TMEWl3~RB?)Vl2jJPH`A|@{qlM2y$fqEa{|OSmUY4s!G`XmVL-NuI=oJKAEvzd4G(!WiryTc zHvzt6J3EP;^Kv<4`Uk%0^T#oF(m}^((Vrd7D!0gY%t(}{H)Y8m&8&vk?9uQt?1m%X z2TMf{)1)5-snUYl9jvYzfF|FG(*8#a`HAK^+z{{ba?@lf_2_bG%cY@G{6{VBX?l^X zyKLpcS$*+K?2Ruf_s0X>wyYg84@#dT)J>7NTslsDRPAKH-L}dVHTuX`(N|{3xkHx$ zwEOmENpEdu;pXfF|Hg@AGrpJn+xj>j-Q_6lU(pW6WA&x|>&Hj~r;f({dADHhGK+5~ zc4O64TlDW6NRsy?zJBT^CM9(eJH0QoFzpO|o(s};sVDm!maSObL%K5A9u;#1yEiwD zo-}&E{c~6R{JNGVi5>qfi}z&N$Ber98SrMoY?JK@#6Pc3_%^d0v|A_8sk3RQkF>?t z^0Sm;5Nx2@c3~$U*d^*TRTIuHi@q% z+@r{VY)6M9d9v>JLnyMntMqw^hWya#F7mG1eB|Z_3+UkZ1+>2Hd+BD?bJ`HwO>%5# zBe}g6oce8+*yO*RrtIoP+TwoF^Yw1p`^*Eg@9d-<3M*)aV5G?pC&}W^oRtn7$-znC z>z&v(4yk)fSmkCFt5zw?55671rIwKU%-AD$Xz_BmHaSE3wlzcgyh`}<^ha{@fG+&@ z#YLv_V!rd2xW9G=wHNN%&L6hY+uT*$KHr~xEfV-!KxeELS&qTG+aT6$8qczkk?G7f z(yaEW@H2iPX|5eCJ^ne4Yn@_PXWJ#7tXs>+>JJD`-3s=9IZR%YoWxE)>d;|{uQcjG zzVxW*qEzMHB+ktpOlYr; zr)-RfHYP}ZH5U`Kd!cfj3FdgE<9kpVR(#aN^top+Hh2-LE>DF@(NPZnv4}OkOlR|d z87LoF0;kTNFq`#&`cg7I>b`@9i$AeIZ8eu{G~taWmtg-`otSvIrG}(%}CEK#>1-@anp4#R^1%O?K_Jshrw3ik?P9VJN^@_u|)J6 z+FtHhSjGE15Aln?wmiaEk8`V4d6jx0zJ!Dz^w>>QbsdE#cNqgjUhRTWFFuoJ%STpP z^Ob>H;I%vsy-prNM)-c#m>R-q!rwN*;|)dxWuknF;Ba5_;V#RfnYRt#r1f&P>@=8% zSgCT$xAVx)=qk>164#gqu+G$8{898k@3k1R{`z7lzc`P8c+tb&;lZO#XS3;y&F~#+ z14ZYn^iyRz_uDs>IZU5jru9I6N+;O%+Jeq(BzzBh>1Vfg_@%fKy*UW#77wtXG6Dnn zFU_*>rp$l8Q4o*^*C*bPWX?#9m;&=7B1h%gmglZ|fuT;fV7JGX`*+so59T*9STN0s zmLG+VPXLb2>o0mb=V5%`751UAykPD|o+5aq!+g9^7~L9a;}uZ5s58f(yp4@R?GSWi z37+^zA-?G-q9YE$_DwkqJ164AwkhzICzJg~AIiJ&2b0nrc&LKN%f;^FxSO+h_!-eH z?2*Ioh8A!;&q9tlvy6l6XK|rYqWF0iBldP4aAGTR$Nqr&r*!O}-W^$=hw%Zs2XHBF z$77BQR>mTIzPq&sJGKULk@{T0l}_Nc|2#OL)o8xmVG}QPx8%Qe-u!joKpuGMlVoH6 zh+YUt!L##aeA;?4FFtbtrgQs!b?n$c^h*Op)! z_wq0eZ8d-@sY*IG`?oay&>KXZK8oR9dyzQiwlr&Hh-B6+Q_6j$O{%SSOOH35m26JB zVbRNcI+!E6FNPne)u0=+_EJ1`ZmOmc^)`6exEoJmnyAOg(`0t+IvMmiDcyBWmuOX~ z zF~84I&sE21&5S8z?xzfg=6t&IRLoNx1878VW2zh`^8b;aP_8L{qTZ_f(XAI9Qt2js z1~pPyTUG4qT1NSm8B~^DL-Ko?Xj`a>hp&^+*WDbZhPTl>;T26>vz3zXpCsStSv=>O zKJN{%=G`j2;MlJtH7?%ve|cZMtLm_Lmrk);XAn)_O&z};q{h}U(EDeM=8sm$oUoI= z%t@xDO=D=kZBOKyJ7Pxn4lvf~jNHjxa3ezxA~p=uZ-Nc*;SAEd#ZbkhO|-GY6;j!& zh?7>osrIc0)Qe~E815C>N5X2>DQbh74Ngs7_>;O6D~@(R85gG;Q{$3#!|t8TY|TI0{2}r zvB#j6O?|cHx@+6Z+n(5rjoStLUvDapTYX8=UwMG)0`mmlSqnafBOvvghCMdp&?MMs z9+rW)=)Z-g^-QObnhqG1I$mUO_0dJ2QS@;joL2c@s7wO}ho|FhjXCx?9ih0{VkT=a zRHEZ?(&5iWd|-ts4~PrlnJy^^+%^NgPZPwsTFiCq0&z&YH%tcRlCt+JdOkfCaSK{# zw7)WZ!YrXtmI|HLF7R3$jJtwKBKa}u#qUFVW8oI5P-HFV;XI+IJ>H&gBtH*B>^KvP z(i1b$_2Z=f+`opcya+d)w(NZW4yNc-Amft?KfGNGi&x<|6cG+1k(0PSMGN6q=fdf) z1#%bfr#s)blB#MAX_qykQZIp%P6zX)mkHGFc{r^qj;Gu-BV>jTfP0fBT7T+GD>@~i z`By%ky2s$;^7UBP!x?X{Ir9VGIwYxWg38sixO_Vd?kBR)Yf}OZv0q2t`;O4d*>}jJ zO>fjWS;JZ(75yzYVgFLmeemjvR_kmLeq0BY!)CC5@pqipS74Qf9286x|K8un@FG2m zHY6L8O@kNRbrv}VldtIE?ZFE>u7%?9sW?0!6fe8&$LG8s80`8UUMlU`Q@GM6x9P=6 zr&b}dxh)SCXM{Nclc3u&2)CMj@L|;@>iV^qM$J4&i7|{v*QTL7d^--Se8Bb*@6jp$ zv)GY+!H?4C*dv$`H65~W$|DxXRjhF)?gEW?l0|$poeJ(mLtEu9{cJ9z*?%5FwOtvM zJ8!_1!Z3W!7$kCW!{8k?7Oz%qgrbWd9Or#T_P-ODUGy70UWq%WZWPq_)X{*!J2A%9 z7w;z@M3Hc>MBB~6^>Is~`RNkvQk_9vd#7V*(;U2W7=n^xHqcX8g;nR;vbD&aH)~j6 zU0-#`l_ucc{F9g+`x?z6PhhPy8f&^Ogh_56Jo6Atd1b*28+I8N{`7)xoh|;l_C?{i zo;-1s1CJY$iHIpnu}Sw3th?0UNBshBb3)t+)Rg)8;7Sboo`d;(hhkz*;n@O3t~yeI zUGdrQj>y6Ll*^d#`WVgytw!<7y?FQ7T6A}1>?X^?kcpOf(x8r8wKD9~6B#S@cNp1C z%nPUP5dJYAY*^3>nI^7K@Cye|>xjf-{e^e%7j+O_&!y~7PS>O9;`l&X8L=84E!GGR z*IM{ZTmmE))4(zn3_W`WSG0w{p<5DOrs;9kj0>nwQfA-KeEd?-;n-gaJkMW^AIH|> z*55l&TRR!o$_mlesS*JvoiMPQG5jZ(B7096TK_IYp>hhsN3TI|eO*+x9Hw)sD^O@L z4U-atZ)49`>Kio~CV!^aqpZn!iV-f2 zvgghN?YZJ?D<du=G{=t~F*AJ4n^7zGRvpo1t-g4}z6EnoeJ8`4#7>siFpyc~lNG~pv?uoHbD?5b> z<4Re0uUfMB<;D)}rbt&pr=in}BT(r(o2~x1a9hLve0@z%dKV>!((@1ar@Ti}nKVSQ zvg<@A)^y}=6JqcpQUlA2ow;J(Y+gNZ1m};o<1g)dv->VhF72xf|B!Sn4=%*Aheo`6 zZ4AEdPo~k$&$0244X>SL$|{{(<5#&Q?|ZY0OCt%7g;Qr+#U^_7%YcSg7Sgzi0+E#u z=0(RVIdJ4((SwiWSEmm07Tr94G9#aMu_}!A?L^0agQ)tw;53dO&WoFB$)x*Hcvx8T zwWYb(X>YzF3~I8ld< zFLcMkqfzK}wiHntG7!DeO`f6pmZ!*Oa;c#++a($DPt}7MU1`mW&&}q~b;29JM48vb z%)ycK!;x6ImFN8KP+cDU`g^F)$4si0fj(p?RMX3KYBV*l1>=wG86QT$4uFN?o=%~hL zXY}FS_sscNyS;2(xQ$12jl-eKVs~QF4*ML3V?~8FH`~64;qG22e+SjM{S!!N zQD9kcDvljgkbVpsLlsT~NpV9C6qbp(SVl6cTpr2Z8wq-`Ul_G}Xbh!cO1yv!*mX=5 zwRaHw^N#MA_P?3stn1Lxo6fBV&E-kmb-CB4!6>Y)rW6MkoLNmcapwtaqgRmjyN;Or zU^(lxtYm|WHvDIW3E~eO79IW1(29}xjZGh378j2*CLw(P=R7{>yNeUmcEGMqkAHVq zz-gZ~V0ED@JU?neKazI9MRUVCS}d}L%j zFIoNy)!pSO zla%D|2gb0X%UVie!$^8*@6aR`ZXa-YHc)~%|bo-{&Uf>aNkR7v?8ef<|uR-Y~wh0LU&os zvu+ZG4`7de?b)%*euOPpNx7mA+|Eapi>BouTq6lGt%GP>(hCo=H8Itr15`_o;ZLw4 z`;2&l=5jR}a(XoF6y2^+9VO(o^rZ{K{IO!`KqTg*(88bkbn)c{I_axS1N6Q~11kKn zRH+{;&pplFcg9n{kri~^wMhEf*-V-}x?K9W(HK+Kt|F_)Q>gt<34Nbw!ddKLoZ3Zl z+1J+iQx%Mkx6abF@q=iB3rR&fiBj#yDq3pa2995IsYzRh@7(FZFO$VHT+I8j=8i?? zk`GjTLvV5R{K-zgN_w{H4wWAEri4&evbFpzIk?EAAGImcx{Jr~c-D2CS8vOwHU{z- zF^e$jm_!@4CrO45kEQtjHuP~@Tlo?(lg?kE$L~7)#MszYT%6R1P{&L3{rwR9J!gj{ zO_xcwV>Il_-csnW4YK~XO<=jM2m{(>3y=2;G&`MxbH*a6V%cn11WaLluUC${9!;0^ zP!KG+Pg|(Dog%e%Uy9*FmeKx0FR1j!QNpuSzT18qyYJt^#~&NOy+{FHV&sD1I)Ybf zfV+wFQSJO{)V!I&>u!f|JDn`n^uEkblFx9B@S=}vN`jVNCK)~Z%#D|8_*_l}@7rG_ z-M?2#Tf~`T-tJPooG7um));QyT}5@lHCYZ^z71v{>Nk-VKSZ{ z?NlMd;?~$)4KpV#n+Gd!>!H_e*}S!YJNd_yJm`;k{j%^!?Hz>GQ2@?AViu z^!K;mJhltX_spV|FL%-ti{mnIERx`_mpeZuKI z3nzlI=txXDWR#3gF*z7MEsAb5m_tY8r&|6#kQt^;mVLSKN!pU4gLh^&NRRf%7)yJ3 zLWixKdQgsO2d2`lQy*#Siff1-vkYAhtwhH*7odByEf)3IM$>&JNck?RY-gV)<`bvs zUfx;Q`n*MIk}>~{{D+V&19*erN`9K^A@3eoNy7^VSQuOkcXK zB%#}~0LpO;vbI*A8*s+6r{bQjTP>IbCSAzPuVW;03I17%y0=|>kATdk1 z+X;S+Sx|a3Ms!L{QJ+GX-`WvbJGN8m$ymWe_U5P!9dUDqGb*>_QrX2e^rKZ5ieKVQ z+x-N?R6O5YG6UeAdPr~!hjaZ2!4=eOC8cSclAZ4>_P4<&q*D_L>9FTaY)_g(DJM4jI6rLhxqa*wLF#%W3@pf0`eA0JElEz=WsA;at9mu12Y0V)z(r(YS|* zV4*u3vXbir8^XDNBb9Oov@*Lxl?LrG&F?G)7J5MMfCNwb1XPGTs>+Y$7^1id@Ar)3 z90MmVb6m-7t|{<~TlSotZNcuAWt=zmHb?AkC3oB80aJzXu+H#--^zGg7ajD-_iK@YhZS3 zYdEj=C*!6jsxPs^)RQ~0aL**XXiS1pY9q2j#&bzeH_rF2k^jn$_s~Hav z@6(l^Eto=?OGLzQ`*MuyoyTkR(pdl19PS=t53lT0{EfPXziy_y8@k&LjIgY5v z--mmHCAM7{g~@?USbM;jQ?|v?MbTGR7}$>Iw#uWu%Vmf+vqa>kc#64q5c)qf*sZvf ze0}av&&GwqPnt$gpKm9(*$S{*;)8xa=VEL9R;(U+SNcS4X!-bZDe~n%DMrZ-Hp9*! zyRwdY{Oy3LhqaN?_=ARI&JelXbujV?X78l+?A|h%dj+kaLs^0hPHaKYp- zxlLVvIma%OZ<<&Hk9ilcAuAU>CW`L6!A4}COU1zZn`z204Wu93ifJMT;97VSFH2ry zM3oL3{_)`Ri78wWx{2)TLP=NTiX!Y?$+P!SGUWr*y30+nyBbE`*UwPy(<5B<<2D~^ zJirQ7EAd(P2p0YN2U7FqEOsWZR!6&|O7UHJH_VO-gI7HkGp zV)U*bqEq*V=B-(VUtTE~XH&=ln;N)%-B&icITK^=UdH}5Hk?w_lja=CqAKO~|2hBs z+IN9JVlHRc>#<9nKG*emkFSP>c-UqLU+SR9?K@@Qk83@3xY0}8MMV$no4XY9!I0X| z>qQPTwdGT$?(fiDuRUv& zf5DQg3HV;&!>fZOsD$2-mety!yW%Zr+p`tY&yZ-@)t>ES-MzX=$vu`~ukBvcdB)+> z+k-g&x|DYN$Y{#mJ2Z6IP((%d#RScj!nY~-N6|g`z}DeB=t*1d@bDt$2X&X$rHeVZ zS1$@reL#z{5)qL54atXF$aRCNfQ59V)gwFdV^!v0UmXth8N(l|5;(wj08cB@TvP->&4!TZP8z<7TiF7^3L zNvqP)cZ+a{Tz!MTc3#l<;6NC7hZ6V4(zJUgrOK_I(2uO9GOf)ta;Gbsl?(Ulhu+vw zyF_xSEGMUlH58@XASt%FDl?d!>=@ycg^BNN`J%lgkKZ;O9*Rk5O#D-neqh2bYGALc4M+avgVHQk%9GN0NK-?H|58qxb+# zJ|y-DzrRtmcM#w5?9N-PZeU4O9Aa`$V#2IRC{)*%HsrO#`P{Ea5}&Kff<(BCjmL70 z2sT`z&GEiBFxpB7zr)ML{A2{;8vJPGV`b_;@Ef^x>(4y{%b@yV2}X-2tm=XAjaFaJ#{uM4GDjt}wgvMQhK{}$mxHlztZ64CxjE`AqHfRf#2U5)H*Sq9;XJ; z!!065cE^{lYsAy5`(e@<7e(?ppg_$tCbH+@A2_R%fX|9aTvqMDWwXkm{84n^^|x~L z)?IA8dpCSG%%R!8&QSanH>s|dinRFp|4?<_k6ga*`$q$1mZ((9NU2bs_j#O8h0@Sg z8X6MOqI$Jx7cw#8iU+jXW$%~!maGiH{Ra;pXZOP*_tnRu;lY@jy&F}H+x(EMOIxF zl0PMop2+LRY!LVF4x+Q3Eb_qfc7cPYky|@^JpUkA{!XKL%$=VI`g0xWaX}PjGn*cM z@FIiTyRhfO4wNkIi9V6Gq^kLVDmyN~R5wpdHL!(mW*D{}3+DWb0bF@u4zrjU`OU0? ziP!_vrG039Bf)KQJzam1POp;~+L;G%u_6+=9twOoyb+cC=0NSU27W(F#PNK4!QOa9 zt7pd}Ss@5M=Xzt^ml*1Gaw!!W7D=JTvq{x(EwAer&$H9_^IbbNTJ0T-_aXLtdFxG_ z7H5;|3)-OSz-;)%T4SW>I8FcQj`=BKE|&BZZ@p}>xly$KNO(=OaDB*J1 z2MpGFj*kZvd64fiOnD@HrdwCx{;d}Bc~}n3Nos61c`;48A;V1vJ8VyVLmMg;P$uS0 zC4zhAe#;W84h+Wyw`df*MnYD!9gbs4@kKlX)P4zuq0MU51njFr!XQgd{*jO8xn82H>xVzx&aj?WF1E(a!Zn`-P*cf7 z>IxmUo>zhzZ5hw2-o|U@dy74c5B3O-knQYAsHl2E6=5p)sk9Z&rIT<@e?2_g{)UC@ zIj)H;QLaiGG>z&7^Vp$qZQqs$8$8A1_^X&=a7)@fYaw12-a(6c4hMb6O}(orx@>PqcX18v6n>TC!ZpUf=ZJgO zc2tdyr@5W7DPX)GUCC{v*1fhkUAz>g|MpWavpl-7d>vrD9ew>yqFAt%Gsmyu><4?e zv33pjY8!=fl^%#mA`H76z%lDX_`kA4Jo-ui7Yd$M&O;e5+2(+PJ&JI$Z-d^7({Wz$ zAl?p9;|<{=3v#^|Y8N%q8q2o$>F$REN_pt0d;6>sFjFInIy8>?g@XNxn(<1y2d@Di6r$BJXZ{Q24Pc;CZGD=H1W1B38I;zIowC zyL|j;JqugscW~;~o(Dvw2}Z_hltwLtO;;7}@z0sRiy3rFzbf2)Ux$>Y2gp;qia~~= zC-0$#^GR>%(g$T4VX4TUwh!WHS#K_gd5&1~GUSWw`lo3}VSQC_9na|TYwt_g5Wf$f z*66XG>0Z$Cc_>m3!pF0xDco;5PUpnpjI|Z#zZ%Gal}RGo8O^a`ukTlo&J#veqTb*v z;yNhsyrJ(gbG03NKAinO_pgQfoN&MJA!Hioac=W3R(z4b2Oke(v(3PyEph0*BoGS^ z9;UVHga@|GNy;lRBQu{bbkC&+mId6CDkiJ(?fCs{U7(6tQU@gFdkE&tQ8cZ5gXsr$ z;AH<1&~Q2h=RS29{INTJl-XdQMkiS7o=0wmD_4dm@pUnSic9`XN7T(QEpt9}-;6=t z8#SDplSEfVH@=sR*jb%S#08WK)=n;5MxKE6R7X6W^^a;Vj^}+-f_eR(nS4&Tw*&fZ zhS}4Bc;WMxIwy?6<0l>AiFkU`I2g;l_8_KWB5$>5$0NtiW?8$P(vt=4u|$0l7U*px z-MGted*6w-1sR3H!Vfp^)h%)d5mvH z%KUkgIqNiN@b!XdoPJ#+?sQM!bLlNFf!Jlq$;C-o(M%H$e-%xYx1+tm7WX$xr-YIs7-N`J|%+K}~;b80)$l7YKsi{5nI$;F=Sw0vzIuhksiP+HdENUjJ&`yzAEZ=ro zFep2rpu;D@ZB#_+#6oC^J9FmpOne@qjtM3?c(Hgdyj`kswNmUc?g;;ZTd}lca}zyu zU59&R*Rj6)aon^XfH%!Hyv6Prv;uGAfQbtFv>S&X?&fHJQ;BE3PC!|)8GAW)sQRa>ir4O5<82TcV$7?<0 ziZ_oj-tYy+C-#7bS|uM+yTbl$+sIWm2VkYeB)(rbj3-Vgp;fOI!m)DzT2*>t%P4VQ ze!W4Q2eZ*bA(2ZrXv&B8*W$PLZ{c-x0t}XW@zX%T@(6EcmqDXAuW%@9_^NW8QVFjd z*@s&f24RE7T9HA&i&u>~SaMIyZ(|Zrb_N(${z#2)S^Y0|bcROE1vO7iYg zpni8TLsCNYEPHCN{$3jTZ#~OaJMaoUEk1VWF!?EakdexC3Zi@(ynYyJpZ}#F^`?CG z`W>uI4aEVkDRiuCHwEmm!q;E(@aNfF3|GiUuR$T?`0Xsoyq98j)><0TES9)a?$hq3 zB`9lZgOC_q7+h4wy9GmV?dU9cE>**pq<7?NrHZRszTE$n9`EdM1gEe2Al>2}y+{tA zD5pH?T)2}S$Pd#EOObaOZ_1vZB|f@r4tkkBkc_5PO7DBk#B`p8h4V+@;QgWOKSrI~ zE(k*3*)p-$a-}LWVN|O7bb<$u7+63UA~B7Mr)$$qC_kx4aF;! zSgc;?0mmt`IHSW?bhjRfDSwa9O$T){`md05SDvDXC8H@v^#Wxc%%yvmN+>DjDqXfT zfpvm7Pffoe9GnZGxkVdK3=EN)m`8~vy@X@!07Bh%aB#3WFOjQq&Xg*$Ig&*aLV~Eh zYd9>%_eOZ06Dn%&JypwY>MR%63OR{D_!>KEm#TjxW&zo zw+n z+8?9=!U_E1W^YVz)uA7aD(se z1geVNpk;Fa`yp5-b;UI6KqUMO1RvxQsBA!vS~g*IvKlqb#r@fOn%l5v;XY1m>xy+rl5yHH4`2IYsJp~{dFTJnz( z^ZPq_JG#(>@5AxK`~rDh?ucYDqv@h&4*7t2kaCuwPB6!s%8giI^D*eDS)<+ivvea* z2TA?*qVvIxTv(>ddW-crCZr0RUAK~_VhGhdYL9#V@2IzP1SK^mQOlK)*i&nRA?_3L z26kw&ScDBFIxwlwp#lHRq+fo*6Kb3Yn_m-fM<#<$>U6x(bHJeE76|oQD7rDbIPq`< zCv+9Ls2h7=bhHb5WZBWWA$~MKBNW}gTVl|KzM|jN8*v=P9|lcjm&0E${pv+BS*8Nj zo(JKb?8-e;ZllN2`N+K1j_=y1;8aXc*vJc!{bDBKj*h`r-5}C=)rppGsgXi+YEW;Q z0k6LI#RTROY>cm?_18^kB_H6hW($5dp&DNfo}{pDXKA6+3&gxsWy_sMQC-vq3o2Xj za_32`eIj~RO%51dkcYGr*YG!K5qmu|=VZaR*r`xVHzqtHzh4uPG~S2{7nj2Se9iwn ze*#u%^OV_k?0YDhPn90WwT3+8kFKD{eZenpOy&!79J#5bnc_tU^_5x%Ef_lx4K3L; z+F&=`5@*y4ORU(_XfQYIilEVVI#AG;(UiZXjwbanMX1VOO5dn}`nT%HkY?dRdNL`t z--vX*0Ce>n#pUA6vQ&7vq$`Ec?Ya#S&cD#r@Bs}nyh#t-g){7QC}tcJ?_w(pOde;0 zPc7?lu-F;j7D{OQ_b#n_e~#=$E{yKHgKuIh3g!&Qo!ep1Y2FQa(>uwhYqa!isf%Pj z+lU*cG{I|D2p5^&M7`=|>=o>$P?3MsciM=iH+cwI(}qh+^?7Nu7C)I$fGM_(XuNt= zyeGvoFm?yiB52r!92vIMLuyTI8<9RQ+?6WK(2=JwEgCW_>ISOyUvIz8hw2 z+eO*?{BfaW7@oYiLfwbC@GoS+zSl=`{q%~qbZ>`eYicR?T`9(i|4T{y>|Oq?2hBT( zVeO32z2F7%YZO`gT{^cFXU8j>r$S3D2EWqJ@U`vTIIfKsyH4AWhO-)QQxab1PA>3R zxCiI;A`oDfPu`7F5Od-yofsm6*)usl-4%1VDgE)Z3*k+gvY4X|W~Ehj+*&)1Uv)LY z;;v^f;>TV@n;US~IdWcNw+$xM3qc--Fs1Jagq}PN?~CQwaQ+xfYBIQfZz!*E4(4f- zE8rq#RP`gTqNuP2J(ccZ+}N+s@V`Tj*iZ85W;k>96>Q%Yqw7L7J|lWpVb&UgOIScT zhbCh3=PGKx@PjhzG+?t{0rNUurmQS&oPK+YWJAQBae_BKy|P7k;WT{kH-TYwExk~D zLJs9=!qd1L+ZDFJU|R!jEcZuGr!tDR8-Su&N6=UKFlui;M`gzhcvt39ms3^jxi6ky z@H$@9sf?y2FQh#ET8dtzf#Ck@&^~)EuKoH)57xDZ>(DGZGT{VFGdEzoYcxKl zZe8xes|Vk}xLdEWG)3go_uYZh=X&(kQK#ij>*s;zzf)59kZwi_-q$y6ShkFW-s#mOvjqN$S>>}OBC9el7X=nt*ts`T>ax+CSKLcN zf8P+PzqbUJ+j{UFI|{b11@EF$JcT$ENq-bBO4U2g(1U};r2HcYod);Afg)83Yy`gf zE90E{5$ZE{CnwE~vj(LL^RU>7g09>Q}4*YKIT3%}?-2Xn^0Ep8W}UK@s!iA!NoJrcH)dZN3!zW5nObHAocj2C_OON&hyiL-fa&p~`--7V}| zdmQyyad`gq6Dhm2_0#F3Ak+k|x0g{=L}p66Woj!mLhpJFJS zReFM*zWP3_gNKW3;V9B~^+(cDeVoWYC%mRdF+1oi7Q4j3=9&ZCh7&>*K9a|?m1Gic zirlz5>b3e64n<}l-unw2UT!46{wmb+uQ!E%`~dHXS1@(#O(cr>iTmE4^i6F9B6fyh z$gKyM<{E-vPa`&XaTfcd*3*vPV_8Sz3-0}C#(5J>zP-Z{sdsd&W)2L-BlQqG&x}H3 zir}4{5*hlF@zU-5j$CCeGScZ|AIaY>Oc z*2nP9$N%vaW5GzzamKdCPifO{(LHYVB)=cgq$F>L)UuUWz4so*D=YDmOu;x1Ioky; zyJ<#VGrw3C$fq0n@ILK#qQkxq@gZYy^^gvix9iCEE;;}6{MqDPit7GDS;u-Tf6O|_ zfwu&AFXbL`CLAR5&v&GR)0RlK%b}3(Yq9>t418I-T~b-PfC9GJAaL_>8YlKKDx&K$ zcvN?qwQVQSnJ5}P@)4%HJcV87Qy8@}lFlC4NhNOEacO7(dH)GOL%HCM^xr0Oa|`&a z{s~+k6i=hW&f?TAGoBgV9?L^rQ78B#D~)bZ#Sp+`S1^thOyt8qgg>M>54#qhC&x+s z(PwZF{{2Y~ zGEe>(K+iNzP|2;66k{4k8!Ou)dBH;AW=?~O(N(*mx>IWM)D~!Jyj#gtEcFB(7!A_eI-){>Px|zdd8cFZA z4#SCTW1gGLJm}XZzIRy0Hu_y~{Bi_J<}T-pYkSM?y-20ocSiH83p*w4nM){4i2S#}p)sSwmA*XT0usfgXHO#r^bMxc|nG z+xmRx65eSGl-UyemBa`rHv<|P`EOCcF+FaiAjMS1*B6%*jM$I%+ za`Sp6*(l#6Th~Hq_{0|^%Xp8Nenpb|;za4z$8pqMuRY2I3(uyzIuGe_fJ~okqv*{+ zn5J<67O$%DPq<|^3_ZzKtE}0ni`bh#I83?D5#ojU4^AFgJXbrEb#^_(R59n?H#iax zFLmSz!*Of|oCfE~4=~SpAv&x|#5?2Js1`j4-}oj>E6qYilQzHEbxYdo zyquEV+hN4YKzP163BASjVt#0b|8_*7>x(Q%<%d|W?_aL!aFH{t+&TEx9Xz~#iq`); z0&}%`eEo0%t`pm{{}~r7n%z&#DArM*P;VT0mxZ+2uh{jm4+`|3OELSpar;pcm%Dc2 zg^w>_YwSJ@czpxT6%Q$6#X>}>_uoo;j(LEnargPN}j8xHNxZWbddrf?ckR)Uxz;-2f?*O zXQPp}$x-Y-tI^3+_=n9(cuB}nGM_=@(sW;z+~P_vqIx0fsOV`&@8BUT8+cf^G%oP+ zX2owIT-5}AZL?YWJ*$FDTZQZ9?`0_bbz|#jcTPxKfnJYC@Sd9!`E`e}P<#}G5tf1< zWuZx}B5%DoMh9c<1Z$#qF4i2mP5)heKr{BtK(5bFIzIXthbH%w*Sfx9Wjjp-9Ic_N z<6C5piXt(m3-B~w5z(G1Y;qIbrUAxbq5Q9;@_;l2tXR>C+o5 z`_h}Y{r-*g;kFXRh%8c>@o|n>X2{kacWKJH9g?!h%2gin#6pEtyhw1a ze+}uwV?3tQcE=LQ?y{qF&+h@Px3fp4!Yr6gnS{NIE+9hF6$e`d*S1o)FpTwi&6u6i zm?NF>>WvKRkcX(=~-3~Y~B2ugK$8xJ4`OfQDx#togUTj^0~42I$(fo#7I?n!lQg1 z`6P)qRL9Z8&A+6Zr!=LW-L<%Ej19jXbRG+AH?day37q}f7TqouA#`A>I9rb=ON;kous-uSX=P2XfA8GDiKdXO=5mM8z2Xw=zx7^$(mR|-`VVX}H`IH@P_)&T% z_FUt~P|3rMdCz1HWhco~D+q@S=cD(^Q&60v!o8;SW3^s+(h)OHSx!ot?CqZte$sS_ zUGDDS!rR=W3g0-bA$q{7b?2-tgw ztJe6ly|x1{KAFtZ9yRcx4KLZ}js}j74y3>5+(=D$yT`n!#^G}ZuzbKt1YR-X_??s4 zFVjyv1B0Z{^ZRI5`6JH$D7dlJ=kZKW2j5m0Vbl(1c#a-~9_EVp{@(e2_|N5!4fsl? z8GBCX%R3#m!81V-1@+y8=jt(?J*AD_X2Ce(x)4D*`{}RgeOZ!}OYiqen9#)#-g@3# z{aux#K2O4*3U%6Ddr0b{lrd-153W;`O(8)eI8QlD8nsVvfmFOICm`-yT)Qn-ahnN`?sv>E?x zX{1|6yWsKKObY!lm>%{RP8&>h=+NeE*dJhz7h^+d@A-1+`?NWfRjntu62tgpoE9(e zbH@$~1B9K`$MsSrw6X4lwF^q5nvKm;hP5xv{*gt-LB`ZBx?UQR)CI}A0+3g{3qP)j zywJ-)Dpvo6$9|2Nd96Rn7cIn9*HQ55e3JG*{6cJxH&07 zc=!%c<%l+PCVK?jD~9raAReoF$-}IFAnlJI=GWao$vDC6 z+VBx>D(>*~{YIX38?p974;%@xAR94P&I}iM%A8qf*tCVz%EwU8rDw_g=scc$(~%D> z8^IphdH5WYPG?Rxl0ktjS02{mX2U6PTeS)zV}k7>i}77yFVcEu;+WzH^vw>YbR}gR zO3gr2?osMf5{*N)t@xfGc1@1Ty#HEfez4)6Wa@a7I=t4v!s-J2P&VKlll!xp(Kl-M zVFB&RRlvl^DtPS>Sq=|7GMHq6Lno_gT#Xq;3^7HDJd?hw-=??=lW2waeX1UlCY*!! z>7%rqlKvLZ3(cN*{dXC)iPJ)p@K;U^s>ZnHUa(Cd^a(#g^6~+6T{ezSrys{TlP6MH z@>Ljk{DbHC9JDh%MbRa~u^pW#{SMlLt$Wh3e_SM{A8{Zb)#)@@|B5s}tRpUuT7g!x zRNS$Tqc^p&bRy>pogI)s3d)mUJERB!hm?4l`YBvGmWshnX9e9vgn^{tT(9nz5`_pXwu;w~PNZoUntiWd+*gNcg$Bm@ZXlNa>+!_4}O8)2m)p~UT*EgDT zr&n!xx1}HI#(#lpoW%bOw;@og6y95M@hwjU57*^We_dDceo;cDY9b!>8-(+LVN#uN z(0D%`#hYxOS)HQ0l24xCU7FvehY5!G8J0=mUYT8}9=WjDGx?#Jn85opdA=cp}b zc!akSdo;Vz+C~Sl-&&6s6MJEdy9GX(`(d8X6}marU&g10{*`(Z?Lpl*74~-yLcLW4 z@}FDbdyo~1Ctjt7g-o`-XVB@aF`i6nl;Q>}VP=aKTUkilUvwx}7BAwY>{wpx{|nbg zY(&a#TlyJxfntW=Cha~!w9j`9Tt@q1oz*XzqkCbLVIPQ23l1U7zH z#Gh@S;ZM{i`V+E&i=#5xs#_87Zs>vOf~o5G*%8>joj)7xtX ztM=32yL~ZisTGnn#2(?R;E{(I^DNC?h;X@x^1phlIp8_8L+)VA&|2~dvqXJZrQodi zA+$0BSuUQq|LZ@zNQ{8#o*ry&JCC2oZ4u7VYy@X@;%9?Yxrde>A8gwM=YKBJus?V3 z_|q0P{Mf>Oz4iD9+G2BYHLCVCVv*?;M6O=U&lgYRB`;TFZIU*k++q+ErOF4^FX9)? zJ6V0H9WPorfUSMic%jJw99>vOmy~zFH+(P#>Dlws5>MWAK!+n*BQY-fJbGQt!>_kl zqNC=3B{#Bh&QP13D~)+jbP>9ff|T$|gZ`o~eUb9X`S@*kJvZj1;hDnW^c;%fY|>md z414neQKr^~zl*+8_Q9U4_OKsrXh;bBsm3#8svL2CDevjHhfW3Ulng~bREchj&u0d@ zu2?G-l}{zTNq@<^zAcCUY{S078T)EW5vfnmh1MfiC{0&k#Y=ZkQJ#mg^F2gA_ZuA^ zFoVuqK8hW&U3j>`XSnxKD>isl!#Zly1^yKu58CbFA5=}bQTQV;>!Sj7mxarj@jyZCc zKI|xlr(kjOx*w=NEOLJ0{#ds6jnue3lOiXNpgvxa+%3?bqsE$W?ZpB-3rQC)wa4HG z)mWn8$C@+aIeJGfUQL;UD;ZD8b<}lCZ9A4ftu$n-cOvzuRDn}xRk+)RYp9tTfwTUE zHP7y_o%Lb9)n*lYd<{g})#(^ipX#LqU^XQ{c}7NNaN)&%Onq_FP*UkpB+7OQX4H-*!5?zb|Twx8eTnRq}{+d$`?c ze+p@v#0J$Kynn%b<_AF-cOU}GrY7QMaDS;iCS%FwCm3nijNk9IIAgj6e7|?356|N1 zT5cPJeLex3`F^}LXEX2lHIW~GeT+7$kLhW17pY&;TWQ>bQo6o6lN?(%^TS+Axu=T~ z^3LC(u4jhO%66I*(R&*%_ed4_<6C4hvIEaNt;h15Y`!No^XBSW@%@j1{=ftz2DZ?m z>C0u7_K;d0tKs=j77Bf>x(+8VkvVp@Bc3kTq? ztKtrJ?hyZ6c?XI1u^4do561_ja9+_RpraXO&e@G;H}lbZg(8Qh6rxjiA4FY$1KaYs zTmToIUa!H&-e_~+b1&4twwLvkwdcq3w(`VpbEFveG&0;?2cw&9U|Jr;S+j3)!}f8Q zm~tG(J~MGO$(s9mhI4GnakgGMl&!~4;X_Xq*+@!}m%8_t_vqrs&Q68cIrR`L9skIR zd(7mvr+>5Ebr1fs{2{O1GLkiKsN<}+zWm~>j`A-_8+clFM=CqdHrU+sNnp_2NID-8mvLK+M~WxVbWm!vZ()fT3W+#os9Ewi;gu zo5XkYooV9=!6sU~0sE(i(Pv*J^qaVo-bU5Z@qpm$9wE00~z!}o8;@e_;qwT?bl#;xLEqh4@yf{uLqkraOE@|`tyDbAt&%~XJ^bh|A6P2d85m09dzjP9V-{@ z;Z&=)+^Y1Dx8|G4+wHx{{lZRB>?wV!9UCNRTBOk7A9b|lu?lV_P9QUfU|dd6=AlIk zetKID2G)d}%*(`RTzYg-`SzOt!v<<8x&`;yH$291G`TNt6HQ`Gf8+vFnZz zj*Zc0+Y1-5uHR4OjLhJgZC-qTwhjM$Wy4+kW3jhZ2NxdnL`Bp|a&&r62m249i&mWl z>;5R(f9iv*SCQm7T9+b5o|10#7X93#Mbd`3)iTe?OL>9TAHlrXigM3w=y7}pea^g1 z>76sBk=CVjW?mCoZXN?Y>WaD)BYcsWlhdDMsYdsVl-BBo;iAKw`bT`<`;SPapYKYC zo;;-Ac4IKD;4%evdrq778`FqSHw9Z*aA}Q=Q56(P>x-J`XtxKXzO|6D`|P0}b34nO z&inDXtj?^ot|xW9=105M&7}p4E<&zyjqZNdq`Dh5^wQu8-3jkXLAXJ-)LCq4jTW<3kSlMq4?ZRgcps(-v5>fALnYxtA_zAE$lpZeIC>)&s$h0MgdbZ@s)Gj4RzMh+L zWAj%``ErameF&qPxYN}5zz@C|zI@c>Ha{C-E!Ph1LA%>kND+5C;Y7kWdVWHiH2Tky zT7wrz#&bv0vC2R^x?6=iwebiPd4nad%t_&$0(CRkyY_YI=m*QR2N z)Q^)amh-W1YP|jDbjUOQ(L2F|GG5-1&cqIt4eF{)F-AY>l~z32{xT+q*So0xc!6Z8 z{80AtgSK?lX0Y`Bl9$wU?FY@>V2r4`XY}RcWm@gHkdA)wqyJ)4>EdZGYHNEyiZ^JX z9^LfG|CNFidupNd@2@^xX<19L3#TFCsxBTL5IO4G&uQ}g2UK2nQ9SEc;NnYrN;|D7 z)hua)ziS8J$O;ACR~Ra97r9;@eC5+Pxw(R^yvF6Q{Ok2>d2!(_X-AGaz1h`3gLbI!=kQz%+y4>PbFU+7sbKgm z8G${of~3tupHM-;6U_47Cb_r$OJ3f$F(AN=3XR<<&T%v?Sv3|NtaM43sw7N0E8QLu zDYcz=N(!+&$>qu#@@1Zi^1uEe?33M>4KICx$Nf&Y60C|#f&(#UPCnU+%v`hXXja)G z^8VU}Jo0@kuiP_sKY#Y=X7=cTu$FCO3A@!?=5cu|rvo z6AfeVuJ0R4IGFiA{KK%}PHe0c%(3FxHBr+8`gaoWW4h=|%hNFTOg47xOM&zG9@K8y z3o@>h=%>R$YV-Ob{mZ;X%cn$P^nL?=YrcS;eXXR^wO#0#)iAnTHlK#74W*6QRZ`(Q z;e1*Dj=T%gsKbDt*t6g~hF4z1`I^JRx0=txUjF2D?cY$lJ^G?MC}HT)6;j4n6)ET0 zLz%w-QLd>bs1(Lygi0&&hl$LCU_A!+*hzz$U(&4q>SNX)n(=YK7NN$gaH z2LT^oJocB&e{!^>|Gh@i^}Hi>o_mz;*ju1-@R?lC0yEQPQg|E_eMxvp;{Pn}WYFa?w&8REmem|(%K&K`01!RdkiUU$Kvm%D8YLd z9naiSJW*~!ZfiTXQTh)rpZ$VqUN*Gn*I>}m6R1im6E4fqusl@^%i?m;%{8UH&ZXp~ zdy4LNks&c62#KPTq8u~|s>s#=i|uE=e6|0Ney0~_oDdp04ean z2|7GC03+&Wh?(IA#O{tm@U#`!Gp7Y=&KgAkEI$ zPJyes2)<4%s)qiBi_T*1xHF4artFh;1#KqhPQrN~Wy&_2`?Fc%Jv1K_?uZDX2Hlpy zC(k@WlubRZ`$phM-a(pk)*XLOEV3` z%D#Ss3ferBeei0PnxAPP-fkv%vOWsq)?yte;o+>pn=_b6O})&5>LOX3^u+Y$_O$2os&3IQ+313VZ&e z(pk!aHSmy(?x~>g%UCo!#p2sTG!lKlbVH&3QT;;xBjy8;CRSCv#%# zRy6-s$n}PM2uhJq38y45Bw9SHn(oABk!a5mnshEvvt zQ{VDs7`ELXEere6l2aSdIHeWO`xWDMtKfE;N!%&xBhia1hI3It2-*&|#e<7ZaC(@85iUoWNBeWk^-_FLF~MZN zG)y|Y1zXKJ3O)y2qB*K)VRunEkvLH;0fUceSKE)v`w^%K3-+@ zxpWj3o|p*jR1Yc^_hyZnR`RzFll@3L1;f?B;@>iu(nBB8P{lWhnCQk2Vwds5#i#i9 z?_A!nO-**~aJZGB{%2Xxxiu(@o6EC?KVXNR!}!}aZPpOy=!4;PyexB&eAyFadCxui zyd&~H8?+;N)vyE}w>6lRM_=LaVIy%`_Z(LAwMRb(RlfW;1#X)~HX*(YS6^g{3`#%Q z<5R>NkEZgwLp%87v?$8lYYrSJhf&uI{-_keqkW$s-@6yqG^^m-&3`l^NsAp!wu$ez zl&lV3WS5W~thL{sQ|3J;*`^BFGB?= z>Hp>}pDg7;vz>T$(+|P$vP9(AxjeHff^X~Qa**sC%2#G!T-Zy@uMM}#E-RMOn+O(X zuHgNh9$fN&zVAhb@@1h7Y!G>Z1DaK6yjePq9JFJru><+_V*`vHt3Zjis`8X|2l!>? zPu%bdhC>K2ji1v~{h|D=w379UYkA7y*;qFxh8*wRllm&X=Mg?ZJk&&ZiKZ&?fjo6C zstbls<2Y_}Cm1>n4DVBOP;0UsvT`fnem>z67)C%TX^eAGi zEiE1JK(f8{iCFm(3_}HDZlWireil1K<<&Ukp~B1W*mBJfGk#f8PW|JE4m5{JAM>Mm z-J&wy-u6Csv%3PLq;RV1FhrUeJ&8M%*s;F68(+GzQuxElk=Gar>)kp0VPZUAO?T%z zDOxn;%y`mBZBNUGoWPi#?*tdV8&+3cB}zOfnJAaq(F1ZapYc3gYREWD$aNz+w(_(S{mHYWs4r084iX` zOc7#}C-Ub%1vs5_R{GaRQ%V@WgfZN>a(~ zjy944^f-c)qK~d8!zrSmdEWJ$+qu1^U}15TNIT9P=v9oh~POBS-*NOg!yWs_dJtbx5UvzHE4>DZWnKwY0YdaMq;@u7e zEG#ehHFdW(zsz1ctd64s7Y;tD%OKJVHNy%IX%($Up$zS&BdVxN5&t1-kP zju!8V6aM$Fxb*#u$Xj2)kFDt2z6IU4Z^iF{r6?Y;8TMNiqu&h)D-Q)D zpr#yu3QX~zq7l1|4PdXe@pwJgA1a*-0F?mtK6ajCf<+#}^EA!g5`Y_@D$%A}B4Ql6 z^UA0o&MAoEJ3|if4EK|qv@U}G2*2;D8$G$WeE?jBJ%(COG2f0A8Ts3N;qjdD#zBi) zPkex3!&FW>))&1tekYmVce48QiO$Vkg15Fk5TP!5U+Wpxg8P+v@swn@a~n-NQAdaW zb?4D1%An=_6S0rBLIEmRwnc`pHA=kUTW>B378%2(6FGOVBcD7T$!@(IuzutNv^+Ou zd!sd+m$8(CeD(Nn`7WMPKc2U)GUe_usWi>?F%4Ywi&h>yja6gY@sBT}=OkN5a|i#R z=LQC7FPPSbOP?Xnv_def#ojA;BFO|C@4u~E#JltlE{T1thv>&-Jq#CYw{%`PaGJb- zSfJe0&RZ^QYn*u~mk0a*;{H`pFj+dGzmK zM6ddc*Xn)QBe54h&L*DyWEM}$RpUlUk850du|~HDWMt`b%Rm{QsZyq}_ii*}%PaD^ zw}7`sUE_@(+RE=cUV)x(7Q7r3x$klb?npU0E32g>Wk{it8rInjyk#z(TwA#$FW zTYFgi&+{i?kRDsPX!8D`M=YJu=T~0ZTr_wzzk3$Wcl$)L!iRXy$t|Sd{JwD9B3PsQ z{3+p;B?Y{{Kr?qFa<9SlaQSl!2d<2zCmXy-HYtz>-3lSYW<_WyWuwouQmNs6l@!}` zw=~IhC2H`DG+sjK`4P!e=}%SsIvKO%IOrMpSz-=h^sF z*4lAAjm^lS1h*9&DjUxw-aR>P(=b?!9fGq3x8S*%u&Q4dG-MeIXHX2jFV_^=-bFOV zMh>@Qs_3qHib6H4urfFvWFgq>54C9OayQyKK7oA7O5v*Bn=hTX&PP2>PXm#9!+ZU1{Ln0wu7B7~cAYO!#;U25Gw2m1 zZGKJNb`OI8pE87N_F^B!canknNSy7wM%)wQIQDNEC#r1W(8I?it5A_;NK3)UvGeiM z%?WS$Dh|CgMO5n}sI84;rCpUK}?RwF^oV&C!#f)A#8j-EmcQ`%%4Atv-2<#q!B?E7h`@T=qRapZw z?e`$IOGn;eU&L)?EATF@J#r46l-^ZtrK}aqN6iA|NzZyy)Be2LX$s{j@pQ*0LhOue=&oWI?Whvn^N8(8Ti1kh$%$ledMS6EG*#YQtV}n%45Pd9 zY&z9R0n;b!Wxe+s<)gaHkt;kmm8a(yaj8lwy`S@uvRA9&!k11IE6&lW>)hz>n>Uif z{bK4dHVUTd_mQO-jfCyG{QS%dtAF8k=-Zk^eH$elNOKbJ`!?L~eI8%gyN_FqpYV+E zdim9ZHS*tUZ_9O#xbuSLE%NMzZ$wSAhVzJ0N-(GoN%PwOyhyS#l9Yy&WTauQkdd}ZgG6bGG^8mzr4os<3Q>s4$QIdtug@RfUVF7CQ5z}-@e;X5t~72H zA(MHl@Qutq;HM?pyp)Q!abA40rG7|^~THsm>5iF=S z0IQsp=+k+ZTfTQEw@O=kYD-k0_}X~z`1w!p`?(gJ7#9aANpjp!&>?mvJq|kU!XP|v z3rO5aM=ifim@jn%qs}+t(3v8xRNj}H<1ip7%}t~A?%HH{HytIHY(m4-1iX^V`p!X% zC|5roZ?b;i=-eVMd0Yx7Ui*u&=v!!p-)LIDgaJzQYzXp}d-fRn$4;-Qz9 zv1Y+fu-~X9_)@Z%(;0CQ9=!Vv+qxWK#%y8QV4I0iy-K*nF`F9*UC)_V3A5`%0^Qc$ zLQAw}(Y^D|TvX=>Ze8qnt~R$86OWjo-ts7R59#OrEDz&i9X+^#n(b&-+Qelpyv@z* zb>=4AyIsIpuXpLZ_gSdc)FeSISLDOVFF z6`Y2oh10ps_ojk<#sp|}(8O+V1EC2GoU+XX(B75D1uoL(911MB8T(}M#O6WN7*P&o zLu__And4R*d_~tnD=4jZEv1;>0Qc#VFi-4{K=FYF>U*o0{fXFSyub67%^aU=7Qd`=p4_ zM&_H^$Fkk7PZME$mn#1>CXD3SwY5sli?$Vt(apLMc*<`y~Sc*$bCxcLBD?nr05n0e3fJG$Iaps%CC@ZN=Le9t-s zN0+>XGnR~@vtNr+KZs+if-xpF$zqB0Fq~p_4xX&g#(I_sIXrC-8oK+Un9y&uopBwz ziXY)gmgjV@wWpK&4pY@3=GvV+0NNkJad(+8pS(qscdjw#=h`XLrj6H8;^+X%Fb>fD z;F0)pP6=-MJrbX;zJ>m*r&6?JqTuE36oHl9HMlGyjiYWq1GdnB@>+4qtLsM@w?w>V za+N!!Wdr%$J3#Mq9;N17pxj5dDBFKJoSDWlX2QquNR9wQw(rJTlUXDld>+S6nuoij zb#RteApbewJDp+-vHgZ0z^yk2?T4Ml&SzR|&R9jCA8PWYB*n|LzM{{e>GU)7H_D%% zLs_LqD6Lori%%{=`8_o_Y}g}+{-=QDKcmsdUWdQw|A^uO?CGfNbne6{EBIkj2W7_! zXjgGPhW8nQm8ckvJFiAz5%svAB}nLnV=v;=mX*biIR2ID=} z1)t;b5Z(VyqF47X)0mBlWO5*h-uict;z?bWS4d<`r+HYn_BI$S2*MZWgu26LvVQYF z+_2VxhK&wqGqnJ8#SBE2Lu-1h4wo~(K=GSg`g6se8l~jO(B7Qt7D-Gy@GS!cpyzFY;+>yS)F~Pzmq%iHzY593YVl4cbNNw^_4v;t zttizX9M4Hq;FFYA%Dt<^J6;mwA6gT?W`Z^!{4 z2Y@~)-mUbC8(I{T68rPUDB9tF8MSHWHH9mcud$IabY*!!Rr;~W_O zWvv`_&%cg`MH6v!P!hT?T}r1W$&$^JI1Gt7iigyHVw2!C%89K(@xDytho+(l^Itm~ z)uM2J0XrKRkxo?;KH46KC3~t-e7qm(PjJD<-o0q%ZBM69cA?~zZp_N>z`awd@Xx{5 z-1^;)@GQ~}d^a4T9M9X7ekO-jx34DA55nZU{UYua$-+T4oB84@Oj|bcMuWA<#Lojq45{AD#g zC>5q<-|phA5#D%fs01#jTO;?;6Q>n##eiIPK4}|=UAy1lt*2I`eMq0A{bEtV;2!#} zIM4WI9&r5JK2(d3LeZ45 zBXy5fqJ=#g=)mrmZv7wd+Kd9?ZRXSRyKdCJNSTIIdht%yactNTiJMb5;-e|SDAzQD zKHtzMlb|;^{c|aN31nWA-YzJKVEsqm80_Vl+JoiaQxtUBtI_=5ZWMM= zjlv`3$nJL~9Gtov2DVmGnL-q0hYhEx!(PMk_ordg+3`4h8p{{{JcI@z7O3bOi%Mrw zaaE)kt`{+&7t=qWKHGg($UFu&14-1{QVesC$dTXw@}=&hNg=r$7y7Vk-#%pun>C$} zpQ^zdkTT!bGl{nToq!?FY|+Iq0^h$+g^IO5xv9lxp>u39I*Xh{1LhA~o3WH;tBplt0U6sGEA_q8UFBI_V~v`itnwyE^Tm8Xj-Z?W##L(~cO$5hL1m|1E_A)}IT z$+cLR#CEoq+MMXz0Gk!0Yf{I&9(*LV47a^XBPGed7@o3)WrJ?PnJ=U8{0x5_oO6KY zy**65a>@Uh{{(yz#+dOZQRI(5irB_cYkVB(EsCd5J%7~MQ3uQC$HRf6%DB6^7OUbU z$e_7a;8h*P1&!g*pd=A8zV<_6&n4Jz@6LTynZl`uYA_7MNu1o`NTMxuv{@dFu6m#}n{8Gp2yuyahau$OONe~bj(P4ZyLm2`azZPi zNK*&DJ^KLNQtPPmf=Y(D%=&xvoYnIti6!H{AZ5kOn zMjH*hldzvLtl@$>$@)CSdV3ifes3RKuDpQZ?we>vawC~f-$N!%jnsaA0knoZ2cv5C z+D2UAUgt>QP}LnANR>f7nN1ipopn+p-U+li%VBosEkM~Hf>70PXi`0dM*ms`x_4f2 z{nbyo`Thwu4UU@N!=CfM`OJYcHxl6B%()=E=>m60y%Bn!jz^&vx1r>DHg1uaOqa*E z3!J+|!OJZZvR);DtMyFi=o5!!Gl;V-FA=;ppUM1i`{}%e9FKdn`9p~_QK5MWDBnNL z&3G(KI*-=S_Jm#ZCE_TKN%dpg00%Uko51Bs^?}@pk>qts5(<)qp+u`r&=*=S*mD4e$y|Z=%a+ia`^QMdYB=BIbBY=?PLWe=B}HnCw8hNr;~>qu91fi^W?r-eZrPQ;oXo--AR1W%W!t9X!LU1U1lZB?J&B1X&Jffu;b?ZaO0w)nz`inyI`Hg zS@>x>8(jrk$?sbL?HxUy?U@t7?UW95YO0`QfeYq6>BkL!KVri3E>y8E#p=pps7(1F zu={rf>Q|NGUE?PBd)^NoiX^3Q^XBOKh~)VILgkd!2DKM^z2%KYK3-arS64e0~&E}h6&Y%mC@$U zlC;oe2#d43P^ft|SvT-?(_S*G*+tE)f8pM%L9?rlsZ6Z}v?5E=|>rm3_C?0qJj7E={YiX1e9&xwA*>mLS%%c?C zaYhPj{*Ityj;`eWZv&n2nnyp{kCVcq0urwNg$_wwSij^ozV%hYy)5TBY}IivD2u>P zN%L^C$a`>W7)FhgOlYGxb23lvW8IZZ{9>vE!zK|O^%=q%=`na{lz`j<#YlI#7VZ0- zj;Wh^QAyGhvsQmYMa$o4H=+pp-u*+D`UEsE--ryNYcr*8t-x6PIQU!nVsLOFwr89L zFMCmF+_VF3X(-aSH7l^^{zH^}*^IM$($W8K9j*#hpyeWfA}VpH$oO&dpb8I!P{>((i7)33e)qG>qzO zwMi_6s~)$?poIIQg)%?YDvu5UCLM(#;BM!!T0D8wUBX<;xXrNKaQP#4gbW6a#|tBxRBX~aCRUI6ZF+-7re#v zdN0WQIYrQ|xgMtFSm40Fd)U(zjU%q<;k~b$Q961PURxzi_s*7KnOGO*Hbm2dm;ccE zf*&M~kt8Y2iFC=vm=tpj=zQ1)TsEpD|% zCPB+}3{dUI1BYAC|6n=peXU7<3-91g#uJXy{lmpORl_N7WL>4I#n0&Z(({zLGK-oGM^UIkEesnxfx}l=qj&Qtxa+k7f8LS6KHpX} zGrS7UAtzwX&UhUC;Wc)paQLxj1sAUzOmmhNkjaKARM8!Q692@>^!_!x?I%VmpMB}c zvH7(Biwu7Bal%9U6j1!`WqO=8mQK#>fZMOH;ofnwR9n80T1>^L{An&Nw!KV#3&m-A z=UNOnb`}n8eU3`Er=z^=Fvz#Rk7ApiV$q{+RLQfT!`FsUXmRWR_BOmD+sT+uMKbHE z#XqTX2w*tK;x*c2~@Yn!q#a`_uf z^i9E$d#=C^%SiA$$^7M-A~3Kt8#craaYw_ikonScCr<04TY0m2>xq z;9M#y;Y=Rm)9fF?ZM=I7hxQm_r+x|etZV|?>~cX!&JUYb8wAluJhbJm;ClBh!_~e+ z;Cng@{C$nk>~$Qc;}Rk$apP$q?GMXpn}3({e)xZvBH(Z^y8zWr1P3I~Z2Tzc@Q;Z`(oZN-B1<=C{|8YhI@0I|wWL5b=f zL9=ZaX5J4#%^$KDs3D0$d>z_#B!ullc-HH0_9I^Zv%W{2D#Do`-w9>pE z6+G^PXQv!0oi@XFsr%t`IZw{5k@(U=0;3K@;K5yWpyjFw#(LFk_LhjZKPTWcfhy?9 zNP%V3R4!Rl5`9D&_j#2#%qaehag4coQec8N_;DDp(~v#u9fRj6+~*xKv{AGi zBa0(&(-UXf5o%47kBidb3wCtr_y{_B%#37u7t$`FDtg^#!hchm!kLGfFgVsaZvVaO zAiOjNEQ$-E)$=)hc%Mp#9tBaxCk5PaserqEVXHt@EP<|V?ZT5jOVM-DUiuLabpQ5y zT-BmP-nO6dolOx2SKkmMq&3>upK=lSH^#9UF6-kzI0X47%oA3>5|@aKM8g@S^r{(X z+pa=fs&9a*&ZTg6&u)lZngotjfsprNJ7kVmqCQO_`a^KHk<`o_jmLr|} z;)WTrnxm`B(|W^)poSNe!!+*URbU)6K{W#hpUNdg2OrmnDp5e<-C{+ zaI733lUK*QQgi4)#s=yTu#T8VA*y|T2;-;{)+}EG1C_J*!{Vy^t&|6JuhtLOc?ZI& zE5d@WR?__Zdj<5mne9?SX7Nv+^^)w5J!COr9{<=`kzd=}Ox7EY;sE`HV_$Yb)m{Tx>6@8x!vEa4^$rqIe@#@cyMhm{|g8QS>({i++qKE2=Zz1VhK_%|JT zH%Q`%yMACP=;lUGGer0JbI@>J0#1ywp?-bF%hvG43sFx5GH&AB67eQM=|D0Z{&bsL z5j9C**0TghZes4jaf2{>$z0UbISGpQbp)*{F0?>&HI^RIhC53(qNPak} z9PvfC+J!-DST@yq98TM2%>6pMoELUa;*-XP@C#V3rMyp=g!{i>t*;bs!MuzK$s=gN z!9kS0uoz3&bB>PSGEKEiA|X!$O3)lZBEq$(J;01ePBQ$FkN}d-?ZCnnqWsV6Cn;Oj zlG1n)^i}-HSR>Ygx{0@`af2pJytodvS?5gUp(x%A5`}ud21?)JMarIkaE;ABT>HKf z-*@N2q6r!FC#M1Tzf6DyWxXhSbQ-F>?S++f2Glw?3X6VShq|^fHpkY)jSFm`MyZlM z%nqWOZ+&RtE=F6*JyELF3}%Ei;%oMOv&J2O_bm}r@k^YXH0|)o33>7s-HH<)<-@5} z&g73;WVEjsH8+@ms(~oy@IzeSA%B5(=g%PU5+{$=!(6qAqF`X>3Y#?*rnF+SK56NA zP?A_N4*TnZ*^S!R9lDkFo_mbnEIaY{n@K3IHy6S!$~dE`)fiI{jK{~HhMW1jF^IWV zZY4Kyt5`?cu+Er*OT|C`p{>crU9(+ORRr)4wM2ov$L+z~sZktqrz*|)dGj$pF>Zu%@aeGQlcFajt zxr-VO`0@242PxyLH(4l;VqS=`aK2eaa5>NgzdKCB?d)@ROJWQ4c&Jjz>}sTaJL!$3 z2}!)!hgs2{^!(yUJaS13F>1{IJ#^pt-q0j={2YDTuKw_Sx3XOQTO4ZcRM)M z+tJirk>uy2Lr=EWawj*MMg~P8(tQVRicZM4V1G zE+@OEqv^ELb98ur|BA=`$6=1(OXyVW!(r3{0m4bZw%6DItMf%dY0 zc;$Wz?SA+G>c*_Wob<;~ePb)84He<1nN?U-xEBMwM^IXrD_MrFgc0U1@P@#b;%2b> zc)t!AH;R#zSOolVEoA$>*`OVG1<65(mdI}=uc^8u{>Y6sx9+E-%h|l`=@s-5*QR_m zW7_%1lrrbdqt-RbxJ%R>l>+M_e_;z=+{E@~-IuZMXg`Wqd(hUY(bQBIOT5Q*8n<={ z&1ty_3yd2WZzBT&yp{Oeh5CHJ_4(3s=3s;E5EvG=Vv8{I%zduKfRfbE`LIZDijxi6-o)eP3Ra7;$wX2yfq?+pCTWpK2D zF~$!$U~MIVWo#K2k>7+@hZkeWs(SP@eaf}49sc33dT^jq1uhSdz(u0J(6e$f7#QY4 z+@&*+YJ};%884n&3_k@lC=#* zW^2&a9BIaEnFO9w>Oua~88Av_j)wau@Vq`_K$JD2cw#n~;fXs>{TtijT}MuS*jPCHWH%hw}l(-u$}*cc`qQp8Vg> zq39>YaC58y{#tbzB(A*0bLc^`m$wn0Qpuiy)ljC0&4ZlS{c-(Q{8UngYbP;}D^8~7 zm`I!*JO)Nx`_5H#hjZsH-^90?a3D4INX2DYDK-d%SeyfdMc{*^qZsd{q!D~~9Y zX_O}?u}!p1tp~5Qvwg_g1I*2xg>S0w;_Sy=@UgOt`JQ7)^cU;GItlO*b9Kp^TT;RP zMBFn{h0YuMlBkLazyFRG|M77k{}x&PXh0YRZR*&lItmA8&&Sk%oBnhDYC2U;_WzXV zb<#1ESvVYn^GBmfD`UAWWtqjrGIaWaIbEObi@G(8yJ|cEV|K7kgh&Dgk1|5nMNYh{ zstB*XX%k7Xo>pL!FXT5H!5Kws{8Xro`V*a@p+SN=tV&p3a}Mi~sE~c@J)FHY7f1R3 zz)c#pc*f&0-uUtpzdHZGt851}Va`q%@K1%avoAsQE)~KtQTTbgD(Gx6NBhQguxE`m zmlj-$A5A6jrkNcSbhOaxVTQD$={5GRh{Z0$+ZdIlN{#C|SoLcwoO)IuXrFV5oHu0A zn&!hK-Tnyii#mm}d%^6cXiD~Xr(P*NQi$G1+Qz}OcGq0;@`<4CZHuYiYXMz#x`pq= zS7U;Q1xmTNqlX0R*(vx!=dw!}v?B?p#Yv&~(J@qDuSf^-KceEKT=biK2A|oBQR;(j zX#FFPowe(5l)e^vntlL({fQu^Y7e4oR>DM?Z7}goDQ>dO!xa+1&LNTTVUs+%8fZYi zW(&S6|BjanCg9KbpE$L)2xr% zv`>bQi|3L2-QUPLp1{QEqI9Q(@k75n$2ET6(4kwNM#hhzWXDFFrw(H4_>p zP^8&AC(*-Qr&!nA4xL6W1nnv@y4NpBuh&*F?@v4~8gT`bem&#juBbEyGo7|(@2G7*QZTz=53cs z$YJ^Twx1KJzdL|F=J??65zI-?@`;;o%Aoh<(Todz26!nwJo9!h>|bCHUmfh}sj4i6 z82I3&je8(;bSgx!++@ZK9qcagixka5fF-0~BKi$tki!C!@bMi1y)jKUn8d(LU zTiQ8pmAT+*Kr(I&-i2COV=&p`H=NsO579IA;I5|z?!FWNU&VFd(5qOkFYA&(w^~sU zap^s*Dt!v>77^evH;q;rPM}qX_cKQCXB;+T2_E>dn~RChgfF`d;PV7u_$PIlRJSIP zMZrp1;Jk;q$}Vuh6JiCnhZE86@fIxGH4}uk9Ve&c8PsnTkHcFI(9z+uXr$p6{5mEL zpRKEfZ;rRQnv*A?B|#E|Os2z>_E~W5$4IDEwSYN)deFjWK9;X;`J2rG}vz{bv_ z2s>6XziSQTr0gc+cSAVirxcBqf5TyP4i}Nfn5-nnhwmw-%FTXMGe(`{>{)L&vK1DD zZbCt$6uLgR$4SqVM9p;O|B~MXai17lh;<_hT}{a>b3Cfm%3@H%5GPq}jOtTcLEc0R zEVRZ_S<*i|E0==>4t03aq5{11=W*jN)^Vv@e{+J|8W{Asg|CK*VNqZYcOY#5AM5|Y zPL^Rjo;?anv}2%K%L*30xWUEAX0xn}uplnwGbEKj+gPSr zt%#fb$BE-Me1q-tYr(bA9)7=?kKFc&kUq6Za6W$yb_}SagU>W@N&QQ!#w5_>(xap{ z!iax2V=+IgxQHLqA`CHY_FUTfenD?&Hh!}$!aEHI@I&Q#8u_{kvy2RIZrTCnAzBW9 zx=mUCUxRd8nIm=jbWA(t3(tnva=#{sax3MoVVul1Tw!0$?GSwcBf^h?mq9r9s&*Vo zdPU&(s0`F~TSpl=mJ}DNO?9z5>1Nvpbl+)>JNnbPkSG&w)Nh{48+jeISayQOjHBWz z9Xk2>F{=Eyg{m$Y$h%wMO`BwJu3AMSeB~&k{TX)LipD>$HsTkLH?Ygo8Bd-rf?;Et z|I>e;JoFPKWHTr@`8!QhI6^0bb16QrmR?U~PIM(<%tyfG9aAu7&K|s^%|gaYD>&_a zB9K%oiL#^x{-V17`?;8#$Z;u=o}i#m4A(D*pn2aIbkZIS)dsMq%mQ<9Jp` zhLT-Y)1~wKsbqc>PPB+Y`}kg5T(nofM+}Et$#lH#^8j`qiN!U(+Vm--nfogl4hxbQ zyToY$t`d(yGcSD_emMeS?w-Y|n+5cw)sTDlIvi~J(@@C6oO4%>0qM#tJbs`CJl1r; ztBY*DDer?eEDxT7?zBtp0_QGq2>i76;Ld&-+C6C};M4M=qmc6yoM9rDJHU6(M*I|>Jr;?P3!HwLd~eKhHC zGH*_#{oy;MV=>+oi~B6!iD*SNr}2PKR5;)bt2r1AU&z4(C`W;%wYKK-24(pw2#wyTizP zL9US)vw`hGS*CGqOgb71-@tK&rW75-Q}KFl%354a?jOZ?rNs_#CB_yFxUsZGGLqty zZ_>E?HvD3jI!d~6j}}!W(p)P1TXAit8BrCzr9!XJGLC$K1Anx!`i!A9B+bsV;CC zO}$n`mZHMsw>ymLb@x%%YvwsFi^49Bu~EtnaHrYx!;L{zGWL7TH8C#D`>t6ism^?d zJAc!Q$QYV(;0*iSci;;*A^LV&8KSP;0E1mp^u)r567(F&#WI-K*SB3r>miL;DT+lhM|^MtU$rh*rVpxw^oeO+&XkrqspW zBmX!1`QO(w$scI(znk=MS~NB6I!u3D%BbkaAWdNRGT*Q=>`jisslF_iTAD;xQnTso z%F(oV0?Uv8F2LL|?tJS@jyEm+OYbWVlhtMRdfI4{`?C;w>tMng-Fd+{=RLIH=}fYE z7(qK;o~6AF+o<8lY&NSjBem=mRC+gqehxU`^(14u!^j9R<&W{0D9frtC<%LHiPx11O}``e~F{bbt;h8+d~#~gIa(v`^a^%3mIS%hA9b!ok22H(}}!rQL8OYenFV!lK?y;eU;Up#`z z^yzEr3K8ae`|R-di9c*FEl-zBMsk`#Q#j9O@o>R$G!89yp=10lTI8covvN$aA+Hbl zU{R_V-2#gkOQ3G!9z#FVma%_{Z`uix|P@ZWJxMoJw{X zIduM3H)X|(@CwZbsqYMP;e9!WEd?*gwc$BAODFQ$-EMr-^yE}u+ z1`qsI5>9z8PuP4{nADgvApMFG$s`{p-N&&s{#qN^2o7TN@cHD`If+`zjVXMh4k>)P z$-1Xv+>p8??f98YtFw>cCB{p7y+Vn$rX8cQgDYtm;}v{VW#9AW4K@}3#k$NkdJrbg zr^VEgzT!Ka!2CLC4+}`Ut`*OZ@xw05g^Z)Qo=O&8p<@-A{N5+k)M@gCOg>BTk2Z{? z&+muvd3C^FSfd|vF@PmXf&M)r(N0}}%4i%=@`VdNm zBRGql8O`GwP+wb!53~{DcWnPal>sMc*Nv0(#axA-ecGBob@U+}G-iAUt2I#HJ0I;W z#?eRpk$i>s5Q+FLr5w%av@LHB>7C1=v7LVWpy@HXy!R>5S2o|z!_OYMK2f7w)xD(l-e9YNS(b;Vy9uhx-#UQL{Gy^%ODoXPYq%S=m5RZj|NZwp%-QUE)!eaQB zo(_tFAx=L)f{zX;q{fS}^ix`!OxCZ#%tOO5G&UE`CNZ|HRyb<5#*=M=DRl_fV3>LX z<<8wr;k&&lr#2je7}w628sL6t2?!qLf@pvgJo;#aTf|GjSJ22+7rsVyw>q45Vjbp% z_>*UuD#c$+MTI%1VY*ETw`%J_LBhJ{pnLT^m?!Ci*{S=S*L)vt`{Q!K+Fzb1NM471 zCwk!F3I`0_vKyW}n9Ds}nv0*`oN7-NBf9;(S}62yb@e87+>L zWBiRW{JYW=bK(?eFtG+Zue8G)y31_h)ezR5z};mYq3wrs$f|w|<}WTrlNF2b=(`#? zyYebN3)jal8tKqz>Q4~~veXl#O6BbSmeLW0!Ww$;K35MSYxi?=ug~JbKU&jc%}1y_ zxdxVA(?S016YP4(^2}gD_QvmU)}L#r8|qIg$1dTt3mTYq>pOF1H^X1UbO>1PN(CN0 zxSsJ&4qd;4YR|0krTjuj8&ikx#>e8Fqf)rLX%oIJZ33~PMA-h^6aSum10EyKfzo(k z8i4nh>Ys!sIv7_w{}g<9dmD1~nYYuV4kM(y&^TC_-tFp#&Nf*HTxN;ita$Wsmctr} z7PRP9K%+7B-1l#@Y3bY*r0u(afI9|ikr!b2-f%dgst(UvByfJmUl_GW2LG&k239wX z;8?r}ohdki%ik}<#Wl_FT~H6(mJb8BcayP2XAO@3-3x)+jOgYlb6WKHJw8+3N2@2^ zq%EPx==vo`j2MxFp{_?!tyvzGf=pn`M_Yj={}P8kJ%rn??LhsbLcDfECUq5=Q8-pY z*qBw^t4Yq>-t4`U;jBm}D)-{c4=J=ql}P8rWt=MPfiZWj@psWf&|1&tj>9v-^{zPT`QdR=*=f!|W2SVi4=`b)y z1oo6oq3(M|^yKyjTwXbszHR!AMUAegP*Y9EXX(?gk_=onavItmPQmP+RBU=NBV!w~BFDN5(YAU5nQ!X%S@IG64C+XfP`H}f#A zh)Bf&=@V?$w2ZMtJz#E|EuP}l=~DbeE_%5S?A|m5n@X3H%2GdS+Pa0N{rtyuj+DXK zi^cHYHwlzKY5|&0>I8lQGupUP2)%X}b7d?OZXgnjxv^c?P%cC}mq?Py%ct18NeQEM zjWOWWHe4iq5bxUG1Xr1}@cCL6c8+BZ;r*kjvHA{taGy>mp0_gphaHw|Hf21=Xl{Gm zT<&+n49>fGDfDFd;=JFdv7j@E=D1xUd!ySl`<{UGq$HrA`v7FRZ-iBog{W(_C(S<- zMHMrIu=rU78rnJI{CUA>Q6GUW9TN0Ad=_SC*TUb%VURk(nC|bX#Ytl z72h=cdr+U|WUnISJ1*p=XHS#d9Pz90Y;gB2V)^>LzfSf0XEl zjtJQ_$&xVHkkflj+IoLI4pnHCjLnTt>}t_!h}n( zW=jaDOumGC{wnOxWemE5`H)hbjZqc0=$$XY?^cSX*&7q+zRpdY{D81aYXOYAu!l06 zd`W%IWJ=e{hVgh1op+6)SvJ4%+kwU8y}g9wwz=ay3}-pW7D$NE!1-);J#W@m#`{@- z$8_FcsT^~B$jrq9-&N@GEwBlKW9`A5yq5Lr3N!a z`m*CPN{E$Wso5Red{lxw^crA5bTzc*Z-yRcU0mEDjE~v1a&Vg^e$lxKFVq;5%RQRh z-e!=VYa{JC{F$Dq+mRi+??16tVBFY4P@?pnb9?s~n#}y+`*&+FpQjJ}?|&dPNt_0q zo1v{Rg7e56DOj}X5zWh&<_-34plRRS!K(QlsIA|J^6q<4b(A`G_FdvM@?!BMbDq}c zd7__`o*D?LjCY<@84U~IY;p68P z_#O6-d8S@*Z>mE%`+=zxJbVHLbt%vSw!51pJ_ct`9|yN5PsO*QQ}J8HA{=SG7agP- zXVS_OBQiv2<%iR3e-nlkmN((S-$szE+yg2dNlKQ7E%BK3V6?k9H0B^kM0+T!;nl-E+@2f|mapyURbkzpI#R1M%?-bsf%-$2@w6XcY zGmyG|4HO(0cU#DT0z@z2^5Nq#)pZpfc|Q+(9!5pvX7IV z0UhvSj>J$3Za{2TF=wCrl6zZP z2(#{+;~!>!l!7sflYtD`384 zA-pU~XZsKio(*U~fr<-uM>BR@P7lb`PlJgD%b;z@7K|pJ#dFWrpjeYJYAp68*YuHO zA7Mp{Ed$aSG6x0x4o7OA*MtYN>@YlO6Aiv$F0E`C^eIY0iMVi_-LeeN zj*$XV4(B$>)C=4_W2xpsD)HAUsBJ+zzCC>bWhWj%|DcsrkT#YiW0O%ivkCnCPQu9S zrLe5E7^W^(0FMjfK=-&htz*7m>8@{t~18*8gY!B^@b4UbTJ{&~%a(Zg_+7=FU_A9bmzvFrL z?BnnUxNYk`*L9ue=bVX6Un%0deO8#ZU4R&Gd!7Zf4px^dgHyK)=tZ1@nj1xM)v^cv zU0;oVN2kEb;X^QP?`7VbJ_oP9ejPvh_7H1iw()%?0l2KeA0G=b7dYMR7Kp33qMD(x zIBSOmE_P1D&z_XwDT8NmrcWv^@DD+CEmb_r<~gk6dzV+m$M8OuX1tp(Jv^qTh(FFT z#2q*mDxbUn`{-i$t3Csk^Lu!^(ln^(Ux07ToPp1bKaR(5YlVrsSL3or7Pwn}EKXg; z>y>|-uz8UamL1`O0}D@L?@^)fvLqQ?wVwi$?MAz~X!s_$4*BzyV8;1}V6pH5c;3o~ zHqTVh?5~G~=hN`Ss^z$CRM}>Zc z+utVR0lR(BXml3j$BJY34QUYf@&ycxaKk@8&xTmO!*g-QTX<%r1o>m$K;Uf^{OOnm zeA&bEkoNrs|934Y<>pB2eRD5f_feKCko}5R`Wq7a%5Fjb1iqh3vjk!mspAP|p7^=; zBs}o%1pFs89>=~i$KTs_;((BF{MI}f=V%O}1H~2KTbDLju`ykjQbA3Pd=Gk$=2hhC$#!4u%eo-#00Ta2r{rE$XScyulj z;P~2F)HsRfEB47?4W++OUHk*hyk3F){u@AvvkO48d^9dQH5#WVzK8UJEx1AV2;S)0 zj)i+9$!WI{B;r3=(i8Fn3tPV9Ws}vgm(K+llwX3UXsed1g(WLKQsCtAXI`VM# zV{I#{dGH674m?2?CYMmwPZ?NdvK8&Sa2)Am5(o&fhLJ4@#K{L}Qn&|qd+$NZFB6#f z-2meHEkSp{51mTv73`@T5@cNGa}~0>SbMA-Hk9#$;a8;sZW!w3B$uziyTcX&P6b5=+H27t5ggRs$4%`VRKz2y8cyfY3N2D7l-3B35iii%Rp5-RNAjC5)n{ z%UV!k7tccHX9%H%HxL(|4T<0OgF?>S%<^#a5u7KZoz4aKV2*MPjQ@BcSj= zHR{xCMXfbEVGo~q2`M}dvM#4Ud}2Lpoc|D}3^DZK?|j(cdk7BM%mVr7LKJ4mXF~HX zp;+l=)E;^mneK~)7t`!Pb2|qg`k$dms|(N?^@&h@aUYTs|HHGyI?(9hPL%O)8`Rh( zLzZ6<`kHeTJ(|Jm9K(gEaYqF@Yby_G8cisv?k~DiE5-Z6rbCIs3JANZ2F2s#@J*)` zIQ-{!EPVx_?u84SDAESUe1a?9dEj2ZaC~YP?+2@`#if1mcv1U5wC__PYI|RRzV+R zL=dpcJSRc?Iyk(qgk%3@qL6FzkTRe3tDP%{m%6@#!NOkXc$5Lt*ZD%}e+;R%SvhKq6xm$4c^Y4Io ze&SEv&EQ^inAhP+;4(DAAKE?3=(LYV$H_*JUi{Qg|s9v|#$GcaJ!;1`KAjHxd z`%dlWJcAW{1P+SgS%O=Csi>Kl6lVec1z#As*6T`NArq6`tz`KRzP@{VgT^YrD z8tUpGp+*^p`HJJZEs8`l)DxTU8ij`nB5cgIIih=~*CYA8eW-uwZ3yUd!Z-FF!{hGx zu#6PPF-+A2&_5W64)t4*r z#F%hsd9@e~E;I*^#rinC(-L~820(;RhM4htz#SiE;#o@#@hXQX_)(b&%bL^p-dS(h zz0(p5cs8r_oH!5*$btHBJM2168gJE+!_TrNkks}ZJp5xep3$(5?+{)Komb`YR)u0D zyRZsvZ@hs-5Ax8j0(bbj^f1JJmjl`K-LP<0H0bZec=66Mp3SZS4=UfHoBMV{cRtVh zdMQh0%je*Bdkg%=%m6;W+5{?1kKjj>87|tRhW#_8u(-!ur1z}=)wrBPzr@m^ELs)s z*iiEPdcQ4=N~$Q`rB<`C!ZBR*dKvsjXj8i zUfSTKU2y`17g1<_`+R6%-SC5~gxwo#LHITw%sx-R#nH2H!38BWcDot8iOql+Pi$bz zBx9bDHxoK-_?@xJVHnps2Y%dof+B{6Xv&P=f_b)&ArPJC$-z8tDM^5WHYvg%iTTj| zp&rEl6USwDTVckI-=N-i1ETXwVdi5QY`X3oikP<=hB{=Cms18>zM6|nlk1Q?vkXEM(nLDo)DaNVmA=Ei)5$);~0KvN7W4DW$iX;tuOO&b`SUWCkd zmat*b0NR*l0H6K~L9br^2R;Yv;lgqu; zADM;eA;H{I{F;#YVgl9Bk}Gamfk)bN;l%=wmzi z5m-ak0Tgl#u$#LoUitkSyqX&Y{_jQLnQICi7wb{^?Pk=~ss%k$H$Y^LH;h7M3`0&2Q=i^l%?T;o5=mn#d?N`C$T`v;d4?;rSukiMYBF;}z#q~Tl`+B=3 zHc^+s^#w2Cb>?v}S7|~QtN0FEV-*zPyB-y}Oahn7gQ%B07Wf;K0QlX2gMKl*e|ICq ze{q4umy}@kf=M_-ZZVt{%N8h|x5K;m=Nvz6Jifey_c6{_gkR@%V8+;PXj}0F%+uq6 zx-@{a<1VD4egHkk9-af3-7?v^@{k& zs@-^H(NSz$dI7D41@Lq7JlG<39F_;IgoA6Ipjrn@D9kehn;|23@p}SX*%$&hj!lA| zI&JW>uz;6~^~iK&6nY_PCg9oX+jY%=faO^LrB4Y z8)S9mL&J3iXiWQy@}_M?H!h_i50@NpcPoQce-hzFmKeUastcApItnsA=W)}f9xN3- zg~-32#WUdndp19W-NCVNu74}yrLxUNb9tVnX#oK|Pwwv_Wf)m=3To*e$XLAseNovU zm^C<3FqEo_TTFCttFRAF4!MBO^QW-v@F$?vURY!#jfckZJ?p0q3+SG5fynJTik`u+ zt@;tXljn2yJ~|}$cDEb-99{vN>m9Hi&khv&&B0%ng~2j^Jrp%}3>rO_&-La#KwS@~ z!Z~j{2ux~#A{GOuGHmeJ>lN@-)dK#O)ZnnkVSI*P5laQ00LM=&!JyLvUh{taa-%wE z^OpUe``4itc6eMqhshJ3W#m5ve_t0lT%0#;Vd$9MdK`Fp1yc5rq^(mI<^rsP)iZa4<*`1l6-2gmVwClP}9nP9j1 z00bSp0Qzs2LwZOx6m%ZNNA`N*bEA&pTD~)I{LmX{x^|c6fgV9W3%yW<_Bm8}aWh(c zA_E!RUx_|F^@fXq9w5at3`2`~?j~mmarb8PQHnlTHig4k6D07qOCE5}|0l{db%W~p z&jl80e2)9RGjieiJpnu$ekj=iSM40&Sq<;OJoq*=#6AG?h%DIaQ3K|xUtxoq7M=s$ zP!uDM&#YC)$6oNy;+q46SL(y1{5{|`@ew92{*l4I;YzRp& z2BEs=3p_q5iG`hep_ccW!Rb!O-(VuRe$q^!oX`NfE=XX{J4F!bbPm=~KgiT|#)iL= z@XZxoVE;B5Os8J~rB&q+mU#(&eDMXnrBlGhe+bEV9fm!r;_%>i93-{m!H;`~;aUF# znB1%YIhVb0{+TKGrkx^YTYW*qb2=WKiGkFpvABBEBy6bAiS-_g#0?{)@u{;PKrgZm zR0hsNWwILfJmP>4WLe?yHdAqjj1!#WJA6;vlZUu3=J>H_1lHp7gvZ_GVC9!#H1gXk z)SoYloj<*W2B9>#m8^yt@|tjaTLaoM?+9|91|xOS<5ujnASx)#C_tNZ-_ ziUGdZ%jfh8D&XSd5*Sy(cSwDM{M0{*m`f5NFIrYGSp-N1)cq-{S3;FEdG5m6_=fsMiUT|56r|JaM%i0!dF%x0@%w~Q$u#dnV`J$C_42OE*tjA1-^+ctc= zVGV@)uLa+G>%r=h3UmgJ;d`MKpiWzyE3kY;)Wr+Q_C|kHvb7v3J-C2Q+UfKyzWQ zO}%uy&8;a11=YtZ@znkxo~QQ_uwo1(%({z?tTeaLT*3E4N?w7duk3i{#3hjFo`e^F zn}MgezkpTyE%3X|Yp}vdeVeL_E&{!Gaf0g4viR-80eG451)d&R3qe+6;YEffO4d7r z?^cy#vA%nFchzWo*!CO_kg3IjMcU-al{B0b=8W}fW}@;aEB^aA57LZNq1G-CqHNR9 zK0P~F_Iw69JkVz|IQAG!o^Oc7hSjmzw?Ne7_8u7zO@|9^YS_d<4iEYVzy@<|I68U| z3646TSJRwf*|#Zx>=|0%XoBY&{Dg}eo`J7yprB&01Clzl0;xOt!kxS49_K zuR8>1YmVh-`HP|H>lHNjA%6xFzV}A@68adp5=?kqw5o;!yADM>ck+3t^Pdlwv)+Ky zYn~0LoCxuuIfB~nQUW)VV>TK$?!$P|cx*S<1+Q4yibhGV6HLmuW)pBif~-F%K~!5b z$;TD$pw`B}>u=}sUSz&+RZIz+%1GescN+zVK5RfPSE`Wx5?TB_`6P_bkmcE6VJJ40 z=S$nXwTXKxMV7%a{H*OZ-z}{}hDVMi2mI^s$&s^&y}KLfeXd1Rs4={bx`if;G=pTb ztEfLHL-6?ZBEiYddW5A;qR(p{Bl!jYk^5d9WV`Y~;txY`^6Y#;VeNC97U#{R<=B;rbrS#gAj8^OKf}80cec%=yjggu^Du0CrjI+* z6;Rcs-{`+Dj&NzzA#yWq0Wmx#MnX^QfVPa6=-yFdbpBKeEOb}Hfs>};f+Z9z~&zi`RjFh!A9F%>mcqy=Y*o zCu$2{iFdGE%noJY=WFK)l$JN3KBISN#HvSN+nxlqldVw~EkNo86e3I(*Fpf~6J}w;pAU+!wbPFK9bL6{JwG!sNWNTH}9j=U%=n(8zW(KNeF0Nu>^zNCkGo@j9wI|#-_g()ltCMa5wa#Y2 z@rrJN&oZ9p*TVZgMY5P_>0n*Z#ku(^Sl1~F)p!d8@&9V^rAf!|(Yqdcelv!#lz7`f)&X6Ko!=QWmVL4HxLA-$l=# z{X!<|ra(??nf2f|cY)F6aMb3w9<5#7j@}g6LSn&d*t$3o7hY<|JGvxb+D(5r|1bj7 zwoMT`!|n(N1HuryzK|GA}8W`+e*-6 zM^l)x*99W|6yVIq4X{r4F8n^G0_HW=@a$tF{IL55UAAI)l$Jh@o~8j+JHLQHtOyQG znU1F)8G-AYA|a*kI`UK*K*Bw-@aJ1P8dtjw7W1>y<)iqWV&Os1ZN31NpRU;y866iK zxqk`moKt|dDQCg@wF-D|um=h?>Ohaa3E=y}dp6B(PmrH@JfLz78?0s{a9eT?U2wi6 z2=`&=s5FAF+FSYVYTjej_8NWKD9QJGo<_4m+tJR3yU6s@c~Jadfyb>a5M=ZPp=o$5 z-$PvmH&(C4$NoiQMSTZwXg&(Jc}+N5%MXspe1~5rW@7w%If|VnjZQB%0iE*uNNoHb zP*|G)Z->bIKGec73<;hNm{6ze14WXPP^a*f!b2o_nRKU;mEp5_`>IIV9{b5JmCWw?Vz}ts=`P^VA z&;IU069i&NdFL7A;kz0M=B_{v4XJ{>iq&v$`y4Danb##J6rsV|2|667qekWLX!e*T zFvY{rCgHV{z$&vvV8G6z6vbhbBf1I-h4o0??j}4tItV=>7I^;rNAPKWGxVe;V8Q8j z9Q9Cvyh)5i`&;zkc*8Sv&+0JByJZPgCI`UWR2jY89fgY37sK^Oxsc!f6ehX+h3M4X zAQtlgH2vMM<$D#uDV;hb^PvDnNna7@jppE1Z5g=1LV*N7vo?LJi>kdW@Kd`;{A*GY z4qeX!E`G+Kyzse#H6AuLyJuCRJ&m0poIMDYYv$rJ^>%p0@JO8U?Jgu;HHYo+3w>`I z3BJm;Xw-5^c=qfnBz+!@_Z_#z6HflZ9tt+Z+3+2DclI&gDLMer&oHk^@}Xk0Hga$oiZ-nN5T%9GQ#Kh#}#QvYpXtC*H>jKyNZQA8yv0+{CItv5>Yb$Jt=Gham z<^R8*IF>AITr1B>u8g8@hb7p9d=>I^v=sO4>3Nc=YC+@l1S~5^nO!pVVQUxn(!7|p zEcuf>H{-)*-1creF_Z4ZjmIu=zN@-P=De>|>1!HWxUG_ImF5{-9h$_uWCS*L(v|pxI)BS z*4`lNY{O}~ek8S&lVbR3DQ(Lg!!~9vC%^c;=FGZtqG8!Lq~gCr_`Y`>lX)l-?dh1w z{1v)s$qyZNVb>4RpIuB;)(W^AJ}dCOf}8A=s~R)?&r`Hq>vKJ3dv9kfM>3V0NxvE+$^gtH5 z^tTb4ZhnCj%6(W@cQCWQKaIgxePa4uoF(p@$^}mbmfzS$JLlKXm31Fj(oiO=|2CF6 znf?QGK#N;z+Jq;L*+?DT53;BB_t^0Z92Si`F7z4POizEC$?Y(FgxkFoInSkIOEt}O zO7#}%&@T#2A_e7M;d}inB~jN-kcqM?Ouaac{i;`Chi9477AZRtsIJ528w9YK;pX%~ zzm4d}jXT7|c>$aFHikU$7(;h%o+D~X{73AcO{ZfG<}=F!E^P7E70l%DI+ot5Mcry7 zS%8`WOO?pRvkyEcIfvb7%3mcG*)pD))IP%A$ci=gFloeJ((29p_;4 znpjn&%h#Di9H?#UWr{3U=@Xl)P;rpU5ZyKr)T%zBaThzMd(t4R$qESXuvW@a`d z%la#BlfcK3EM$`vlMcHC%g@HLgp@B-*HN0;xy!STcsVBg@)=E>u#$G}8_C?~jwm&J z-o`41OPE9;VV|NVF>}umIQdFG{(I7uD#{j%l#Pe*x>9pa`bsaE*fw*C2k3(-@V^z4ly61D>SQC$pZAv&WXQ zk(W1cDTj`dcx`WP`{B8`|L0^@Ts^mRPoYw&uTy%dn_(?H&Q+q7GsK9EZVTK$l#a!! z95`nZP0A*(Bx{ykpe1Q{h~o!Y+rI1OY%EIPKDWLnlk_WzOOzk?LF@>p(5uaT{i;XW zPDgT{-4{9O7H{Hb@4!9Wv4U%y?!y%Zt!7s=Pt&Rc65QcHEv`7kjvGNTNRX`|=PC7# zoZsh6O-4;FwN-Uyg%(S=YxZ)sMP{blOW$JdTV)pa>skyqP}L0`4^J~ZJc@mrTugH2 z{D4YX8#m5Qi8~u|h8#ST$^D)p&z*=!=JbSdTL%@NUdptp!pRYtQ?w)}kUenE zyzB| z_l2W_s4)V24x&aI2-y7F`nMUoQ~d`MQFm7b3Z9mrJ?h zSGzbJZYQ_H+=CwfPu@0A;tbcNbe-6S?BzZ^4P!m}b!5hpN}++^6S=}E5dF{U#Ql;uTLT61z{!hyG17xsUtL7b2Nsi=O)EHOHF-|G`LM{cY!?X@8ndZ? z;Zp`Abh3LL9VPX?#%vyH*IB@ax#2e zuA``*85<;VtID_H(T5(AsSh>D%g#b>W6^CkD|HoV%5&pVUEh&$8Y-fU-f&X-cnKT2 zCCzR49?DH~mbUL;{)R|BS16Wln8W{is}RF}okoMY8)}h>Q*0Xx*OsWNx^p zZ4Gs?o$yxO_Epynwr7(Xi~JHdI=)5S##cIpX(<_`;}8jWP@9xvluHOxit8nbBQ`Rjbf z^$9&%b&ZDnu%v;>X~bo+FJ0jLf;eAzMugcrxwMtKr0V@*Ja=KYXv(H@v~JfTa^J+5 zT#HrV0_8;H(Nja43s0lD!3SDw{-k{Bl%FRwdjE#R-%X@BlGA8Q+8u0tPl*gi8gd=8 z&(nCbNU}27k{j*xm26q>$OY~C!2QkINYkI6CEM3OCQIteX`Sg3>bTI6yCa`VAMDR0 zD_n|M{ILqW%JwVIdD7%wxJ|}R1Jc}C(H^p=%bL1Ni*q5n!?}dtDI)K~d8E~T3z6)a zM&$Uu5Qo42(Un`a!1tTE!n9wHNUps!J&#J;B_8NwJve_f^6?8@DDNZeNJ-J_3N(L^CBt6c` zEU7k%)s7m;k`|hATW^co4(mH|#^xH_Ryj3teElevW-4Rr;i|x;wO=Gvheop75rf2k zu#@=wI!!**?dFu8ey6><(?olpOk?S?#;oF#ITt@vL1*+9aXTOX8(TD4L?m%_b?gncC_N$25X79QxFv=1w>Wz z5h*d2XTdX!=_kvV)I)s^doZtuOxNv#d$rGkMy>uk9|n{-yNf6i!g8OxPwoq?CX*YUMsOB3bQ$1yLRr=v7=;`T{C&lb%~~zT_#Yzh3+21M17}6vK{k%SfFbP zyFUIWon3rL)HF(hT#K+{(P85`p?*9u6UxGY)*xEp*hJUxrrN8n+QM6YTB28T%!PAb zM^d{l7|M@q>0wKh3UqE0i8iEP~QW#r$P9y)gD zGVNZoo(7QFbcfz+GTKZia$ltdZaO9$dH;$^iT|c!KbO-UQ7rQw?M96T)9D6-c{FID z3CoGQ%kBx%+3NTFJ!;$wI{$_=Db+nhm#scdUn-7fwW2PT?5>0Nv302twIMie3VHwFAcC3we6+fR>YQO#OSk} z)T8WR(kS-8W*WILF^-LkctsDLHfI=rMhfgsk_fZYE`tD<>udE_<$7C)?)OC!86gd7&|iDa+IF`Ggs6Q zW6tnubLQgxp6zM6PTw|P5?wb^E)gCmWG~aMvd6z9OBb(hr%p5a=z&A4+1iFou5Mc- zDZM<7R!od#$T5q3{(FY~hvVo-hmp+8{5cyxlv#M?&$~wRSx`b1lFn1Jv!|Hx{iFDIU?s`Um?JE!>!ov}Hn6a-Vr;+N0QGpEBq~7T zkly5d^lM`zc2wdsSjmss%lDWCd{iv;^ZJON))GcE7BW}!`{c+eX}Y{ikN$YOi^+U6 zVi6TfScAu~DEx^Y2{W8Zd-U$oTvJCH`$mBr-kHGA3mvxi{Tq?AuOv-=f0Ps)&0=0J z)M@o^H})WDGpl-bo8F$G!~8Byq^Zk{SdFGMbMRb4`{&%G3rD9h!vsxMcsz`aZp>$| zPUbLG+wrt>vO8VqRxe!qW<7bJ@IzE_-;a&ekYL7&1+<0NhgZzIS@Dt^Y~x#v(gSHP zsrUNRY`($~)@;$vPUej)U2VRm^!m^!Zu6xHEF~(N9y(M&wmthJ)OxC3(rOmTCY9AQ z|F1V$fAMD8nCwOW6>HIxeL5^)&SC0u=pmi%tHgA#TeF#)vzYPYlcFcD#aZ!_0($u8 zTXIirKJ7Qkp+2{tl9#-`{Ayo^9TpFW?nvh|7l-}q;l)V2q1utFNcbf3loph_8cZ#{ z^k;S{ga@+9fU_*4>=@G>I8b`|(Dc&NcQ2JXzMM%fosy%^6ISD^r|n2>ivzv&;1IFi z=|jm#{%*Q6g6vgk!>u2SSmT%_Y~{_}WbvY@xVLJMuCKGFSCynVom~!Olzto=+_H=P z6PKaGD-SUum1&3^vVmG(j(ggmj)FzjSvIm1jG-k#jzi zgL*OSTDe$hhsj4K=@v^3_A%0s^Gu{V!j;Twyg+WOxJO$1`8iW;v}oH@OE&GyN4iBN znb{W<(W|d|Nm+juwO)6DdiKwvwo?V77^@9L8if+u3-U}YqL8fJIfPYjW6pK!PHvlS zF3T?VV)9+L=-(d(Ea^fX=?zIF+ATqB=9nog!aId&R#|aJTBdV1vv0tY`B7Y(VGh`T zG$W&?T&DZyOs03b9f;hdt>k&47#D_?lhtb?*fVVz;wfRz1hV>EZ|gd4yi++5t4?Gl zt@_OMOgxKJx-E)Nj}{dwexW#L99Ljq#QpR?L*7;Bl(tJuEiGw^E%n^^kBGrrBJti% zsIq~wODhuC(R(sfcIh4VuDXkz^-X8Wv08X%tSMXe^DrZiRJaz`-Q2*-)#Oh70C7%R z$ud%|(sz1F94>AUMTZ?BE1u0@7ta{56KMu)(Z!kUY|j|_#M@msbl?mdzIc*}5608n z`?&PQ=nfWPs!;lCcQOfSzd-lR+{2RKi0H!yXFNC?L?>-yMW?gGNqjFO-j=)Rtr$-> zW~>(b9(0TTRJJB|-?!1|&mqiv&l#ri&X8Dt=Cy+PSyAVW8oGhUN`=)QWMM*u(xOFE z*mue4T;fY1cXjebdQeE%FVn@WY}R918N7)tdKgEXsvZ-$rF_RcJ}o*irj6Vr8dS>v z44XOQAYq?h(t`FRx`>Mt&5(a3GFV+j27lk8;CzApi_&KXM}{ zQS?s31X8p-lVqEyFqw*Zq|Se2Y54U7+NwI4IXY_5C#63{>+k`Ql$<}cjj=h6$)J578JMz*ty|ho9|qSmg*ltqPqPa+{A(hA-eik* zsVS0SGb1|Eq79tAQ#sYWCz&(4$n)>gsB2y`jh^eyLSNW2AIZZkaPvx@a;?NdH+c#|aOQyA@4n0WXVa=&G%XV)omo_TgTJ!M)Sz1zq`1Z)luAVaUqv+sZ3<2vL8>4 z6yozU=h+@lJx(T`>Ephgm80L=6-D7Pi||;NcJjtRpM&ZFN;Buui4LtoTTukL_R@lg zB4tSas8eL`4zRBKlsdcr&}rUwERXaK3j92U&`8UP;KMdk9uADh)rm)sM{;!+1|{0-f2egy&3CV8OkP40~yC#Rgv7r|oqD^SWwL{iD@1 z%vFc(E^Wt(myVOVPG{1jnM``!)Va6)c_d=)E~1>ALFI}jvF=a-H(FcW&)4YJFuGP zeRcwN=b2x(c0Cu3mbf4aDvcv0gG#oaRhDo|e^iUc3zF#y#kHctS>ht)0|&X~q=R%{ zU&E~Tq~J7>7RfeJrYe`ViZ_+bV-@yS zJOO`CvnCfK#Q7P}Araeri(WjUz=eAGkewRcL~?lmSF@sm7;oWcU!&!1cMsp@T!eqQ zuM?h==#QboO~onHkYos#c{ifR70S%Oa0c!D*d;11QKwUUj74vIN02a!ccPMFDSR+Y zk;N4mli4|x_@WtHYWY~sJGTxGsolZLE>@9&Y5JV-bPCzH>qexP$o0CrxK~46P)q~M64BJvCoCgL@GX# zTiBP#JssM}y-01~%1x5EtCwGs@AhYiKfjUuurY$nf00a*5{{DR?;g^}ahBXu?U7`; zlMA;%ql|RjtH*D4WY9HZ^oY?1L+((wK0Orkg_vB^VP0N;g#Vt~a!x}t$pym-k^ZL7 zA|LS#QHZ7lJ0f3Ca(2xSjeHeLyG~!{?)B~GO3#OIFl`c->p)1%#}o9VgsSLg&Y(zU zq!;OUV#8hYpUed$1&}`D%lMr4a|p)))HLW5hqhXA8?D;8Y;8I2Q`|UCBv&mevl&nN z<&>G%`)9Ce%WZPjEd^_)dx!>P#Ry%rU-Yh3n+92b$FjL~RBp{d?z&70%=iJ-tHXB2Uns?meQ8 zv5%-yQyRXZYs*f13_(VCSxHP&m}r8@K5}r31J|_U3D14XK+pLuZ-FokTvvNj^DFUO z*vuVt-E0%ic8x0u3}(XJSIj84s00TL$Iw;$p2+mXYr1Q3=h&siF7`(yMu%ZoF;6bT15xKJCh&JDX?HBl{Q9zLe*5vQdtxkrt7-YonM<&4{*VPMq=}y3x)6-_|3&^;om?+Km6B8CvUvMvx7|8l7f*WH9dt}*Lxgl zpM4{iGfxt^Yd`44lP2`g>jO|Na};N#I&qt&G&udk9G4=a!ns-Ve4_iOi2KJFu3_FK z(pxLRtt_;sW4$7%eD*AQ%=95$`OkutOIj1RBNOOXG)r`OSt@|;rZdQs};CG>sD zM=GZ2KsCRpQ_FL$ILSnud$49L9XrVsXYEbGmK9oD(4%3Yu97sHx6y>Ds{W*_u8Xm> za&fxl#AlKCwkqoawvYdo}Imudl4?nz7N~iMS(DE zjk#Vg3SAVF z!oK|_5nF7@HBXgVw{`({y6ijc&YK2zHc8X24fA;2<7?>MnWISwI|n?cAojp1pB8@} zL+r*|G1o9x9Np9m^S>$K`R?`D_3kh2KVAwoPG`U;y%JUiJmG~%U4zDN?o`#M3jJp0 zbEh7s!Sx-pKyqIM^ayi!cKp0U2P2o`?C7lXn~ZgF`ZRMAYM)6x!57=S_tHA292iej z=teJVuIY>&Zo1>ewj6nZSxT#govQ!1;K3Tfvt&^&w?A1lXj4fY#Zy7^%mE85GQ9zT|S z_$|(A|NVxsYpwxrl@e+H+RVFt{WFHf7GiK_kD$VCF@EA@(5%g8;7M-?9%_?f%|k_K zw7vPFpD7AFOfJxMy@V|(jilirKHwBsMHAoD@pPl#3ErHmVm*3`YYP`oA?@E^<3Y0y zTyZ%9&zFoM+e2oPS@i|1NLdvN|CGb$tQ@%UG71Cc$m4kNUzT@W_2Ka6S6I4M7Azl4 zp$$otj@sdZ^WU4I{EKvIG^~kNU=4hU`-F22xNr|r-KmIg99a}Rovbb%!Lz>AnD!|(j6maNgIXD}71*wSgqwQx z2{^C14O&xW=z{w7bawI^tG~a5yFk4k#xJeoY9leEf zcImOt6B0=B{(in+yw# zxebTyO`}`d(RRnQ;g1 znN1-(ikh)-oh5BEE`u$JQSg(Sz|I=q<@5`biDaNEiHW=ghLsVpe1kTKKDEL__Tfy9 z=7Y%cnc)82k=0viuzgx7uv>a2`#d%QoulNK?SuqexN8QHb(dhHTVv2VU=FjH!O>Cm zKHM{d>zs0kHRw0Afsf%gXpfS|=|3&egnp-YKks5PVRvwHw>K42cuVQwX>fbf6I5BI zOb+$VzL>md8`CA%Am#OR(jV9fm(->}d(#Z`eR-b-oa(082anMAsnJ#shK6ax-gYi= zsDM`I8G_&U1*rW@jD6}XhF|)R$)TZgVp2izwOAT&(#p}ad5A|oiPd3tfCYKCTNf3F z&ruJafHU)cPFucclMNT$Na?_1>b?F84&T|yeJk_C+A4oRMuP)f_&f?tZ#}E6QrL{2 z>`kfS$(tZ2c^h;mz7q@>D4?QV1m^PlU`D|f$SQH-zFie3_pRC>+sc$Fj1uy3JOyTy zI|jy&|7z9i)4?;}wUad7&qs1zg*IxP0_mA1Fy{9|kRDDV*Jq5SRo~{}siB%bra%iPKJGHl$IS={sGTJZk<70iC%CB#W5(L&>UIQoPgYmWX+Un=Kg-vT+- zKe!*lXY}C354PyDz6_Qbh~snj%@C+Im8vY&03BN)k7OBxS(O8{>)2Y{;2F)$|2&WT z7W4}bdS_wCjCUY4*A=)w%5|`@7G&6qP>sm z?Ocd)!Bbds%6Fb;=6iv&(P1u4tOYOa*$G-F-a=qdG#Ga#^H#G7aP*}fx&Jd3ZUv-( zp0qOU-pxgpFpKF4v4XftS#0!EWIuGBsd#7tW=%kRr5S|h3WPg_ zh(D8f6-&S7UZ(d#K4X1^Cr&MO!0O2hNS^act0bY8ZpYfsv{vFLEPj6oE?m|{Uo#0o zM5z?o^j@K7##j*LjWf|th|_I5(gA1xZ35zUlIN6E4pC*zbqt%7iq`=N!E`s1?=BVh=M8|3^$AzMs@R8qsYPZxIukr2J&Gr;L zvM>9o@uk*q4~51mfW^WnAwULK6cSp1=JG?Bg3?UC9Qryyr9tjnl`&wd=XJ zku%WWd%IvvunGGju1+?|nUkzJO%N+~1orMZ%_|z$jMhWT*BIvmeAEalot*eByU6vTxnr4k&I~CZ$ zvk$SRO_izZJ%>7TVCU3MaU$>&KXp$bW)t>-dEh;eNp0h_AEaYC0dTBrLQx?EKN6-g|$1iplhxNc<*lDJpgcS&4!9!(4LHbnM?-3IX4o;Q_mZC;lv8lApJ-mm|B$ zD0s+S>}{YXS|xDzm?K%SKLxh0h=Oe2NV2D`0sSk)$l%vDvZ`ny?U*@+MiyBEKTnm| zT`VHy=UT}~oh|q8$t$?AN?!Yw3C0yJOTMRm^19ARZY-T*EQdfRp;zx+s7SdC81~N$f|o_w!w#F z@wN&gTl09DVPf$6@j50VIR?DKFM!qiMexjS8-(sk1s^*@;`c<6mW(-qRo;^1Q~Dsc zZ2xK0ll%m7L8@d(&z+~1S3(a;T%;OSg;@V2f)**XLc8`vGUF5?@~05LjVDvrxRR_3iDclR49R}KncO*8 zLM{*HLu~$6uJgiQAp_@1-sP<3pD)nlyKA;^_Tlm5sck2j>#z$}jG4(|x}MTAzZY}G zUrOQb!g)kVub#YcXeNE12Z)->0GN4p66;g5$h!Jh{=e24pb%Dp#xFBrlZ-pb z>>N$9vj%w~>mtdwdCL5u-)8*h*G7b^bsHJ{q{eq!@PcruW#rZ=LRNkn0PzdsI2-u} z!HyJRZ<~S&*>~tY37qknJo|4uU#T^pKhZIp-=Fr9-2Z%yObC6(y~v){CF0j9 z!9V)yH1QO@Ln3uvK!9>SOt=*eY1`bWWNa|()15{Nb`KICkuCgpi!({o%yIl|x|MJA z_$}G@AMpQ9x8YYGO@*5WPJ{GoF|v2qh+~>8+D0X%q6%@K7vMUI9ZgaLOykUAs$4OlnXTZ3(ZaW8nP*L zha*M)q0czpPMh@ISxwr~-*D2k*I{z*X|TVR0`b8o$gUG%ByGiFJSEir_0)?I+fiL$ zPOr%w z4Hf=rVLDCfZzH;Tb1<4*F$6w0#p>}*QIg|w6aLe!h2~EdaM@IqtQ-g-#etJZ@3T^3 z^ZhJI9q(Wz@G~QOOOBJx>63`>vk!2~K9-ExoK99*bm67bkKxquOcKMHaA}&Cpi8Tt z%YV>@Rtp#5d7)NdSTqluMK*Dx&eZba|GP;_;_HdNs|dIa3%!NQrD4N@K~6G%G90Nm zLO$(N<}cdjNMa*r6E4fs$~aw$?9=xlqpjBSdmhx0mJua>u$uyZo~bB*nvo8_KD~hi zN{^xKKH4Nr$XU?D5_r2S5Z~4a`4{gpQg}*+uVXw&!ZXwGO8N)Tj@}A)PYM4&LS&hO z=sWy+LYy?qTq9NC8Dv#p0IaL2q`Rh=vRE%ybW{7wQxyHptz6ZHW5>+nJvco~1zIOS z4v&GstyOFizl)2L)uQPyY@zJ2E^9V=gzo-P>{a>0+Le}Sx!+xuv*ufNTU`9UOkysBW2kNMajU($gbr#;9*P^s~H!jbOz$nYV zf`w}q!jHZFFcMDaiQC!GHR29$uCIehg)(Tdd>l?{xsA4AwGbP2iTgO^J=YQ90du3q zlX&}aP}?8@v|6a^t)IwNY8v4$A4QyJB*iRr^XRc6XTkQH7hp}$bbLJL3bi#l#^s0< zK~|_7M5VQG-;!PmY)X#8wbu%8UttsFT}Jw3*B8OSRdMi1R)!;y+wjG{Fo6Z}faH!7 zG@v<3Fm;q3)^&GU9X+dzsj7wa_5=-h^-`4GQa?}gYUJo%xhXIZlvb-^;0aT40^Md#~ob7!4EjaIj zRIYFKI9ye5oO3x4MKvrWApO24Z^_YXxXOAGZhp3%h97uOkIZ#}`2)A9`yVxCY-z@Q zwB*qTbNADoRX)t>%sj3*L`pE&E5i23g|ptU3bxF07F^3Lfva>1HYp4V^CCsToyl?1 zPcHHVYisC=Q`;?jc6fr*QUk&HJvz+v_8Ky5U**h zG^m+8<9f~{(sbnmw0&(pSKjdo&D)bW-~SY06{pY3*b*X`G+6-y($esA=@XjArD0&F zAuH6mj;XU5#_~5~=&@w%d@qARQ+@~nhP`+@1H~|@_6!O`S2*!R#aeOaJ7{g6j{zW! zdop!-VV@SDeTf~KiDctIIZu%CtBhOl;U7%vl;;+%I!y<*NpnV{g!+q`ov89I9^NN> z=K@X%aSE9ZZvDx2yqrD>ryeK=LqQm=l8~kcu6?BnqQV}{Z~&1NqA=fP95X9&W+n}p zXtb`E1Gm4lsmK!Fzt*6?_f0_C@jGja_k=-!%ne~@(>&Z=63ITBz^n6=uNr;4jbXV0&acjEeb2*Z&c|Kc<_p`kXE<=ta8F z@)i9sa0%yWol1WbHjV{sQo)AGD|q2r4tDQJpj~$}@mkJJJQQxhynX~A z-{3Yb(Hcz`xA`N#+7RV;h_YQDD(H5pg-mqsN{Z+-i za08X$Kb{xlkGd=e6#7VrC>%`+-9-%9uUP9*Q z4*Fr{4XPiYkJhDm=rzlj1t7hQF7=%PufSUo0pd!wEjY5G1%I7A%OVY*GK+a9nP$FB zopVYU#*Ad+i*RJ^Ksy<%XI1#_BRi**anrw#ATo$KyLZCD= z6VkL7z^JpYX^3Y&%DW~ooVbQg*5zRayuk%T6EkM3v0I<&*>HOT3PSG+BKv3Kw7E~I z%gW1K#ne$+S6KCt^aAjM2vE8E+tVTwmomvmZ z_!wh*(-*Edf3sj;V+M^#9YULL<5|e+6bSdW;OzfMv7KXgbGm|vTJJArf)<>@VYKA5g=&-^E#3U<)pLK0>n|sj^-1zv)c}4>}>fTd3`96l6t=#_w^* z>C)^4IQFzET2(Elz6om?d|1UIMpv@TmGAL-(OzvnbAmjj$c7yz!)lvui>7ATgbDOW@mJh(4Z$0&6@MLFQc6CSk(*aw#)=4 zglS_!>S#8p{Ww~#6N8cCC!zn~Yk0T7ov9TYv+Qw-khpXrsgeIKXdApIYw)S7e!(~oAiXmSgamEwB3gnKPVa2mf-trMqLF$;lbaCob?qbMPXzvpGDy$U1*C`ad zPh5qs^|G`a3+Q553D$nxiQRjZN)3d5&-KciiPC?&p?6X;@6gsocodtse@lEByF8r} zxfw|Ao1cP)6T+OdxEquy@gdl%nYzaf2>eV|bNvdw=oWCFOY{H0v(u2kg+-6x@sk?p z+}}#PMK{2|Y2h$5dL8}mnJO{$kHQdVH;k6`#i1ZIwtqqzDsCzg{JA7f9QE8_)X8vI zzabdZURM`~jSkxX1d{@{V9dcy*m%1f zrTt%Hm~$$g;0!qB!Xt2HY(A%MP{aGCy98!mILI03+cWKDU%2#dIj|~R8dk16%Y}|P z$~{=M4ur z(eH&m!WbPES~8uSi7g~LG8)8g&oT^ElC7SBy}{sHu!8u1@kiUD5f^a<8eUF$oz z_jR>Z-li~gbHtR`WrVpA%t;>;XKdVUORA6=D2!3H+cj%zc}`6W6aYU@ASE*~{(Ki0v2g z++R`l@^%bYJEY6CgvAPCo1cI~&?#oWsCckJ>n!$1$QNqM-N8D$%kZ@M2x?uKihoC*;qvu2A#P|H>AriF)WvuR zQcvFJEV`x9aDNiDM4dXz8?W%G*AkxN#8|=n4bLwoiTKgO-yN9XwKucVSSiS>^1}0r zwqkJ;qH{+pwcfFmq>k8v_RMH2^Ywl7+Ri($@>&v3KH6(#!yBfb_MD>b+r?<-e=#^k zQVXuFOot5T`*EN+JsyYOHkzK_mF$4z6{uG~?iA|s6b zyf)0ja;IU^kBt~0DTeR5OJL>S@%VR7Keg@(!ou6VxXG{y*rcCo)k5~q0O0q1ATdb)@l&;g?2l9z|c+PnfW}3*e{E6wT+O`HeoF?O$ zze%W3G!`GK=VI&njd&^MBzrXd6>~Gs#;{qka5p0gv;Ks$MA13IY@4~bZ;k@{t2l*9 zT3q72voNat-SCu7H+;qYc>c;#cl3C+`(!Wk>TF{cGi>k!UkgLK=ChF*6z6^@Vu^1f z7{6p3NZq(EaCBUTX<3Vf){b;^-I$5T$JNoot}-0RPR8-7Ny6U7W}JN?mMIpMfR8;w zli_Xb2w#mCb{8{2?>1H>Isu&a?xp9SZ=xPIwqnGTI-c#-22^iMW8?Wtn7T#`*a}Mm z7kwYZhR!o=Av+<6p0xq!_7HdG?LMCG5@F}-hJaMk)5tf4$kbxcVg5HR*9}$D^lsSz#&5<|V-W z1@WMwGXq^d-R8EfPC-AVxvch(I9^cq6=uQ&3yys_P1SmYUVzmr+2fr@xX7qLVLygA z^GdDd>TCY;cILX`-l3bgn@xd{9s@``o<^q#IVt}C@ zv(Is2x3m;dzg<;0gWti+|BlhZj5vWGdm%VI{E`d1We4W3=2CBuCul!?J~ov4>F%o7iMLN(0|K;5-IZ;n|{P+!xU%+O^^U zx>c%jmz<(_^Jy^VoZmo-9%_@SQ{u4EMS%Hl?U|8f6|F6u!_(~oo>q^Qz~KB_2=Pfp zf9tcT)*g%bVXbtFgbX`$?H7vFenN$GTYS6v5pJUw(SNKyORSmA29Dfg$z3t5L%*53 zA=E=TyD6c9aQ<4`cazI&xyl>)Ey0<3cOkz!3w5oVENrD!;qNV3*mC3@?SHIayS%6# zROSxxEYkO}fyDu=&9IsbB$|_?uNOdIuTBE|-=P1)^pV{`~6(lI-Fq;r{iF2?v z=K^ED(;U$jn$zk;?m5?yi<~u)4$2TXYubZraU?fZej!<4H@>cSgCPqybzvc^Uf`R> zLQk*9K{_&RJgdo)X4Ajd!?E?}1&!le*)pClNUeCo?)iEPdF~O+jM#x)D~0FCH-9F& z+7i1;PSQie97HVCXV(ttup`!?#QmHM-%j&3xlk~Ttli1S-i!6@$>DV@S5>WUnv(_F zvqqHTEjF>4TA=_ zEpRI?NPbp()wK-1rKph$B@%TlvtroPSx)r7fM2MaJDzw~50d?0$wU*3VZ)Oe0ktq; z4*f3dhszRJI9?CPt?k5qZ6;|Fu_3=r_k+psRWxdzS2wD3R9)qtbNC_bwQ$!@hX#=} zBARmwUlbpPfo)shU(Q+fcV#x=HF%H==!EyD=aMgek(ho?gl_&D3;j#Cl5I`nN%}x6 z`J{B6SUkMK{nt8+*mf@=iJv~;51DTel@X4lWdR*3-$TbrezdewkY*EKD}q3L0`y-P zM;63K6OW3wWZVQ1f-+7h_UknIJSqlzNM>3o1pWwIH57=c9N;GY9$TUZPD7-t9 zg~)rc-!q@ml~HflR`|+pgjn&#o;wipbTfWUiV8F@+loo^Hp71*qU@-;DpAI=eK4>3&3}k=lkcUZD%&5AB8U@|t{6zMWLOmVg z-)9CD87qkPfH!$#Dm>SQM1)zNvVt?e!svVN?@(4X3g!A%;>^m!wf9RE+2*knTO=vu zPVFWc#_43;v)A0vWHH?S=MS&#$09+4?+SK5_z%7voh|rxOUTW)Sr7-kK2rZtm0kF9 z1x!A^#r>i3I7`ffG^XBRmSY~U$9JQMdsQwf1=+9w2U|AH&w^77pty8O9IahpMaYi| zLK6Z=d(kTL!2bn4kZodDKQ}{5##(y4MS=~dKfo{27l{3Cp)Y+tA#$e{ki;w>vUYJA zwX^r4=Y<*Hg3oX9-SQ+>B<+VP@4boo%4?*i{xz9+G@Kp6{p_&ADA4=)n1pYV;a}Pi zPiBotKm)&F`qykU`w|nu%D&6hDJVHHsR(g zsVek=&qV?juU^EyMinrJ%4W<~aDxlGCHUW5=ab{V4v<%)>gjx8?#BMyyG&zJ6!UZZ z&R&Prv9&gvu=Sm=?sWNh29CFQg?rC1KhbLTV5%=u7nq>p)DXh?=#zxGN~EKSPq>HM zpl$k13^BAIg>w=Jh|h(z*R8-kI?Cj&^VtT$WB7I~9xSUAi05^6SQx@*n?6?|?#scI z=|iYk*^Ocf{`j)@3rgIRq@z_$iJR{l^7Vu?lTlP-|ICD*kp5P5DT*hY*LwKXti|FJ z+?dirPnNYM4h-40`chU}I+089jmganD;r0|JgT~i)@X;ugOEx-hwO+}F)P|R^M@AL+PPnJ}^vDrzxE1o=Z6+%- z-C5f`87j3ZfvDtJGA1tU?MO9*U++Eeh-w~pEwhT6j5T6UFp8*;6X(~tFUG$9GXhh| z1KjZq1KgIbeol0PC2O(v5@VAX1)( zM`ssc%-L6HP@spG4IYtS2@&LsheVy&^F%g2e4p?$tJ$MDVN9f|gY#CXz!Fj{u1m>i=avs^kBxLMe#;F>V3Jg}9dyx2-A)+zB9IRW1zS&q0L6wb4M zw&E!HJ@_vu3tk4O!l&zc$eI@t#mbXRa={OFW~n|Dx)}2_i(ZhGVP<4@e-Tm8Uqcqm z8BYxiy6F?%EZW<49|PS3+4Fzebatf@my`T~w7m7gclA>!+kqvD6>P_^d;Iq;;Sc5v4>5`S8fRlb)O=1%0nTe*1pu0x!> zwB3%;6_aV3tv0F^%oe^G_hHrtG5(pQwq#E3ed0QMITNH+gKHL_NWHGW--n%ui|Ky+ zX8b{5{##S%3mL_F4vZs1zw@BXsRj>S?IjiVUF1$2l5H|pJX5gXUJt}@-c^x zx0VrwJCpf!Z4)tQ-2*g==fVCp0jzGwh$-FMLcBJP;A~ z=wueG5M~s8oF_x#az|sS%{#CqPgGR3;^80-*adm1E z+}k7xqWAy7){qT2chMDI7&F4n=XbN^m06ry-f`q+hoN1VE7bmyMkUow!B?w8AZ~Y$ zsJxG5QfJGp8lN79Gr4l?%gYpK@sPvq`%0<#j$ZDa+9+=C7afR{9^@H@lw-{wS@zU- zFK`p9;YPw&3>LK@E;?^HDGht}JGP&Azeyma`;L;xXS={yRUiA~1Q;uFf?hf*^l&Kb zX2+LD;n4GL!7`Q2#3b@Fac#KEopdo|ALkbc^4B|%j=;pD@GB zf$mnH1&2bl*`|6SM=s3!vHJdu91Kn*Y0YXR`@&nUaNaEO(M@;Rg#?cTq2^ri7hGw7VgnZy$@E9qz#og+s=(G`P9N-@I6T7N^wo{ z1$f_d8aXuWDHw<6!<_dQ$SUzW;Owo0ulAg%ombR|`O-Bcf8BQYs&WNwI-hX2Ud_j? zFJ|JFsC)R|sVH*v_!6{xFG=3ZKZGN3E<`8|W%t^fsL*rHDb3}`9jO6&pLv3 zF6N{=K!!h2t%Yt#x{OaR2=#u$O(3|OM^60vP33B$+0z6iGIQ=25@mfABGZ1;+{UTo zx@{y*`{K!F59;H%rDs5F(1H2%33u&JE14Uw6?f!n!?Hh5x#acEB(1BPTs$m?>c=&Z z|2G^(AHN`b891>4sS(X0B0L1WKsBG5R(Gx>;bGZjg1YKTEWT&Pz z1YJyJB4WefwLcrr@E&rJE){U*iwx*z4}j~2y~J2!G@pe0BdgXeMTfFZSU-0i8H(-3 z^eLHGaM_HgT!-}B8Pa+_56q0erSbMZ0!Xh9(Vr{592r^5+u{q*g$G4_<8OI(W0yHbYc)+QwJ z(i{4pjV%e2b|!+ycZqYI2tUv504c8e3Xk6pfSmVhn55E;cfBX0o7o%Odh;GHwW^SJ zBtC`|9akl5y+)H`YIfvw=L7Qe%5h?XhKc+V;-4F(@gp_h%=%W7-6C(`+cOU$ZE+m( zt_N~=uX^Deu``foUJnX#A!PZ+V>qqfmaWQ+$KVHNp+Y1Tbi8lTckNEBMzKP0e4r3# zRZbz@k}71&Qb%^7{0LDjY=Rx*ZsBm;B*I)7nP{<%7|wR%buCy6-_Zz+M5@`$$3v)D zKSZ+%{!mH1m28%mHY*ZwXL`9JAochvZn1p_$wf9?*X_g9xh|aja8+UU)}Gvnly8EI zBKq`OLo2Qyj5w1gD>?IgpW)-cz!O+?H6EP;?Nd@$!WR<%u|x3tEv(S4&Z$mlIfeVRk;?wlsU zhcrmWQE_;iI|2MelZfb(Nb>b>GZ(o>oBWR7%`=~L6P+&1fSTJGBt4F!`s6O@(*HpD zZi!?{*%-3_pgMI6eZW=k-9)ZQ?xhoIz9Tgh6Xtw0a>_2+)Oth%TJJT%{>~hDvGfNw zMK2l`x8y?8d~fb!fEAqn=K>i*AA@0N28{MrgwtNYj!i5g>0?re<7z+f(3nk6?lp#v zV^W}D6N5FebJ?MJhZz^wN*4(q!ubc#R-#X`z9+Ms z7Ae*e_ZC%3BuJz7b*Oo?g#7nQll0X*Cf!zRiP~&KmLiPrR!ueq)!+rJcjXnHh0hOq zR%sA-UFD%#?MdES1L=$4BaSi;T#yDTvD~WXY>jg@oY<@f!B=WADbJB`FTCi1m#;a4 zF91qvxzMXy0UtwRh`($u^%+%!xl7kUOkXC}jvGz>+fzc83UQek%X6spcMjGmj^oFy zQs>`0u7tkfve);ToIs1s~bpnd1BvCGF#n|kNWdWS?C~PzG=ZAh|;Ayjmy!sF%~_vOW1q|Zx*QT#O$UD znB3z~&|Q5NdQPe1tfFI>a$W(<3wC=TIuDJ-Qn1xwt!>3feb)PB^4 z>YMzS!-GH;q_q%FX4PZMO;>z+Wg<%Zh@p?rfAm^4n|a1b)HSMT)^Qy!bi9CEfP+1l@1hjYipOn0DWQHLcWS zf`9dlNa)rX-k!q@UtfjOEkJz2M+h@g6>1_oBek zfuzaqBD!z**phIEN;ek@NL(kEw0$pk%<2jD$sML+XXXMXC=xaBXI(yqY>{sX8&r)& z@1^Ekd7>yu=sCcQ9?H})jdGT`;W2O4+Ocf((M+atR0Fhq8`1Z*6OFcrVdhD~d&i%` z;wH7w$2u>@b~QNn%Ryri64yT_Pw|2>0-$j`%LCU-D3NV`t_P8188V#daP6|XCB4rL)+ z8tuNQK*#k6C9}SM8oysIJi!Tr6=X$vezlNW1}wnk|2ZWfp)Cu z-)Xky)3Q2Wcfy`tzksP~!5C70i;Bl63l18KV$c2&y4dSCPS+&`I>ESInS1RM6f>#YME9bzb?xFncSN3u=7?V%ilJE;q^thv35Mu z9!g;50b)#V*&XJS_LC)AGdgyeBpjdnqxMj;JX>u32HRy_SXJaN6zuk6*4xfO@vTU7 zaXiI~98kygvg5enwQ^K8_!ab6mSX&OUE-N8Af8SxENs?Rc=Vq&(b;|#GJ`(QC;!}$ zr&_|Adn@QsT?w+UwU{MJR^yTjjwEDqA)V7Qtu`Q8f{hRB5bUa(Pd)#-)22JJWX0bk zlD~8Z_i^eToO7iD?{5@`p)!AzyCcQ!yt@paoU+)V4Ij{8_C7G3F2@X-l$p(+U-X8U zP+w=H$6Wq+vFLIo7U|={%$Fn+&C5H%RG23^sL%xdlQKd1V+k6)l4nme!+Bw+-r?gd zE3rkD)G5yyV&Ry;mRyx!4)W1>V!a`G>E*-jJ9{$!x7DyYU4%TZlO=Ljin+lRi9mH` zfz6^E`r7#o7RKi@rkBJj4R6Axp38LWGeYZ36iLKQIjcptREYGhvBW|9h7c!j0C__z z_IhmutQ|~(h)H8OWh+;Zb`U2~LoK}Y5iKI2uSb-NMRAVZO_(}=5;+pl3@lfgto5sf z$7hq-t^XuY_fkFDUVZ|GCvQR1>SU~_vp|z2&Lr{SW$x2kCytmLfl2#{;W=Ly#LFW{ z<>62g=5rp;mUqzcZZpZYUEU;b_$=2Ia~iy73bP^tGsqoTap)X0B#Rn*p~Pw`dCGqa zPVWO@-tIv`$CVS*X-E;QBQt6Hm0iGZH{(ivPN1#pMfuO~wvnl;J>k<8Q*tCtiKzP= zCbnB*A@r&v(d9=Fb5TEbQ~DKd=yt5fA| zWc?*x*ur=`eq9N#ANtFg#iwHElwfpx zZ2^xa8nBHHW7vVJQn)=s6Mi4fB`P)YBw}0?m^L5gyw{XNO#Kqv*L+Oqp&QRUXJj*; zO*$JI_ehX(&6C-W;h40+e*%j_Jy!HmoSPch#~JATgpX^Iaej*;x_mo|53Pnl4d%cU zO);hs&LGB6pIk|d#I@7j(;T+%=jPOSNW_Ydu4^Tr?HK1oKAAYUUAz3o0ddm0l#znyQsT+7FuBqeotvPggh#7VjK0qnQP?{5~Oct+vAsD{) z7WF^%Lh&6@GCHuHgW4M2;o<|JJ}wG^SR+*@<@11@Ntp}p@|uM?Rp&WeslXgr zJ3Go3PJifh`>IY^C9?ua4u4NBTM?Dt#IuFq_^(d#> z76ZR`#0w7H-e|s0sO>EI?>OqWzM~U#)X7HQ%hd4ZO*kYhK8Z28D5&&Pq^nhBxjpBN z*`J&(_^V+Y7MM;V5AGVnyaTUrWJ)>JeX&89zcG$9ygWz(TSpVUSDLJ+RtmQ+{>vp@ z>criJ;#}KmH#BKYNA-oJloVX#&P`H9j|4gB6ET9jJ!ae)xdzxeQo-{%Kdn~6jo=|u zR}B7`LH&znV_}*+J7XCJK00DpbIJ?5OEs8skOSLz!k+10n+kXAJ0SI%K6?9Z!)~cm z?Dl#_U5mt+eAjZ$>54MmZ3v)h$^D$5KAvmJ+(GvI$>-#)^{HLx2~PI)f83;@7rdD= zkvp;+38G%x_U{&+tm`1XA)?ru)S%#VBUCdsCMDo=0%zgvQ7e(O1fIC1*S@QT3g&uiSz zkA~l+~BpPH&3KfwwYEqK# zW38XxYQ0yV@Ar?r_IR#V>w28$oV{N!AE-Lu4My7tm$4(7zUeeT*MV|)ahwd>#lIUh zyLr$NJVx22Pq}!^{yHso>W8gcZqc`!mf=rJ2{QVuB>Vl~RL+w>iRjrDf^QREQ`w9% zJj4aT-h=rNd@GN`{`%697N=m&>QCE1aNLG0(t+xSgmA|1Om8B6xr~vpU2c|*~Ofl>-%z(FC4_NCCkC#w<22+?t)MDTHyLE4)`!2h}QL5f*ET(F=DJe z-)FNuXNBf6_Oc)qC&aI3#&^@`Pdo-IP9MZcMg{c8ml$g7zFt&|Q$^dK-G@h~Vo}LY z7iD{2g~C3P_~7zb)Qi@@S3ExoG^BGk#?OQOMq$t_Zi59Ldx@s04(wVOj=$Mj6bh4Z za>Njty7h#he%&9D_r<%M=yRpukOi={OXAGWsfr+X;VsU^yotM%k_E$ddBRMM)vzV_ z2RyWKfzG<^sQTg_UG*poRhwhbVA(!SbFUb~N|RCWu?cItyhW)C2a%S3e%$rPhd@qu z5AKzXxA3{{M-NS#DA3%p2sL-y;zG(Ng6Q%T!Nm8sMK2Pzk=7yaNKwrMvTb1i(;V+D zy1H>Sa(wSi=A9+%P*oEyJXHvZE0*AUojz=W^)hOdy$!7@&QT936=rR^AGRih64k&$ z#!|zXyU$Q&JE51je$Q{_@u!?M?zW|^VTE+vszD-~qO-Va?tB*G{Tycms}q(SVgw+idXih+b- zW2n!F{ba|17TO*t!6^;q_jL;%q7DCCzH{?Cr9=JjO|?EZ?Bp!mbw7(fu83rv+lGky z8vkHjwL@Uk{t490u!Qa$JyImErJAGFyaZn;)PUq9BsPa~$Rl@4rfW13uN{?PzT9)s z)6GJF*XPLNdBgcx$xeFk_c4+DnVrO9sWe-#t5GzojWC1PmLypH1`)g~Vw-=}qX6RY z;FtuoT=0Y5UF6AKz55E=yNxOb~MZp=(rAMtG;`)c)U>-tyCJBu5Si&u~|iP>E67)WcHv^vSD^3*;+alpVoG3B5ER=iS1?&n3gsGN-( zwm-ma(ft1Y=?H4DOUO#?XVD#nBD`rfo^2h*?}K%ku_^5vDm+|)E1h0I8IO?_zT-0L zo6SO}oIngOGh)TXvxwyr0r{MeDtJ`?l3V7UM*UqRSi?wF2v!^ok>8C`)nW&;*5AqW z8b@KwrPqShH%%-yI$1FDq9?d9xr6Q-pa`IrjQZa?z}hAQpyVX6@4JjlSC?kR_n)JG+n-kVmLsOT?qS5i82=^wk z5v3li{K`9O`+5$>l&5o_|13hWtSj52U%*bQPhjigf3V)Qq3qzDZ03CXzUZvfB~W@Q zjnBsjamfA~V3E3wTlLPGj{NC^hm2G3uHSY{d1u9*s`!xf{$Dw_iO0FCZ(Z1XaS>k7 zjbzVm^YcF1GGNu`GU=N*jg+rfC-P<^aG~`D^liO^GTxT(S@$NHP=7$UQ)Rhu>ntU5 z<)j@1ClWQ4B8PpN5E=r7N^hiaWw6k zf!8IZL0dyeBx~HrWa|K;r{xE|GW)Uv2ZoTv3nL(~y$H`{XNg*jeTn405H9ef2k~0T z&#ErTAnq%M2zPXTB(dACfLq*jvhCYGvWpEPneXk%o@H7f8=pps&EF8$x%MRA<1IXG za3=w>>cUr^gU~-Zo*SU+$ZgPe09qMAuH8IGqL-Yfn=F3PlX?9~$I_j6^X3Q6;Oq}j zJ;l$`4(la!F?1rbi?!h^f4?KM@G@Dme<7Ti8wtnqhZDaf3+mQUD|&KV8=uB_(-Qwr zbpJyGd^UNE@PbX0@brNnLP6?wnC-BN0halNWdzAM#} zLe=AB@PMtbKg*g~OSj?nr_hhS>D-U2HHP@kbTn+JZoJ6Kh<%H); ze1u;L-f&H^bKd zqC(8hiipT7mX1Hj_nZkn;-&CfXx^Aj9@?k~ZP%yb`Qi$0N7X#C@lZ97g)|IbpI-_W z=i88KEjR3YZyPiX&?T1Pvc#-|$mg*yLkP~4y1m3QZMtwfxts6(C@j3pn>AHw5y`o#8kE-`vD9$vgSCK&PF9EzI| zt4TYg{b~WtZv_}!IhdR;94TBkv_E3baJV41f#?VZ;Kr1Alb-6ll{GSAJ2dAH-p*O6by zh=g?FQgH|FeTt+%UR=ez@jmcmYB2Si)|(g%X@$?X3b`MS_FUhhY|uQS%0C;jL2arm zPI2>(EUX%2jF%d1MKZlZO7cDLs-k>qp zO9ZjO4V;W~e`4aD3Y7;Z!=<{jv?`+&6z2|wIfkh;=!_dKF4cf_%9Wf>+h9f01V%B5^0@&!c|?jNxt-Tj7*SYrL#2I$w@QeXk01=zrG4LYjfz^Ge0;t ztvO&n^b`KM%=dAj;rSR6}7)Cj%len=<$h(PI&|%svdR&kLKgV>?NU5`MB>Odde)JTM z*-4ST!z{VU&byG4(jc?1sqh?3o2g=<1^s=rhI{om0N2dfi`6@xf?tRlOyZhwSDu+D zdD}}E_1qC!f-})EArE&OM{!pV`Qqa@ak!lD7!aQhpwFOh{HVGfd062`Ms2mn;^c5n zCw4v;dU2`XOJh;aG#wxKnpi`#>N|5Zqpwhnfn8ZSV-OnWO2gXbP_FapFt(7#*K?BV zq}O%VV-Qat^rLDh{4SJ5@ks$$5;Y#KH1CJtO+nl~?_L-_dkWuwGDOgnw-C&^5>T9w z%=3Y0Q~h8yCjEgz-K{I!cvB60D@x=J=6n_n(@GJIatQ!__KqGIew2R)^TZ1!cX4DQ z2dg`DSZZMt`qw29XN$2^>rpYO$ueUL#vDZdljBHfpEYFa%QSeiZaSJgai=>?MB`xX3>=6e+9$%9dAgvhJzHolgRM&l{|_{=*L52c5KL467s za7n0tt~VQ;wgJcT{n@EWLD=RR3A=nWFj{mRZf>25x6>xU$#Y9MXHP43M(44})X5Ew z#M_XkOT*~fXAilXBNX6)*$=9f%;ORdnLre7o+jFTf=Ke^4dCwhhFiR!A6uJ{#?AJ8 z1<9FdoW5WqP3PaaZM-gk$p$CTcgTRZ&-a3s9HQrjaPG~R8dy2&EnLe~g$r2()nZ0s zXR8mIKhes8w)>*H$7aE>L5HDtYz22b&6*3+OBUR+oQMZwZ&R{mKl^&0Fh{v0QREU&(X ztDm&Np?Mc!iTOTwaJ`UveXPXZsq?`g+7Bnre@JKHXu8K#nZ^1$;DkPp(5uEMJ0#qN zR+laT+1tsmNM;M9|44+;f;j`iqn`(OV5iuKh0OMCq1v%3BN#w7lVb=&`vo;82x{u}G|dqm=Y zd`?fiyRO^r)_;q{|6^TuzNh~0+V0rjez*Ov*neyO_PgtQq&wbickBPkf8*VLPwanb ziGO@fPpo@QkMzX;*4?>ozgu^U#DC=P*xzmL8Vx+nLqcz3QR z|F`am|KGI7KYr%l_1*byzeji5|MmZy>#6y_>Fzb%@_(!OU%7v+`&aCLt?!O^*LRD= zU(yr%SNy+q|G&TE?jp9Lw+ALVfv{=wshK4R`&4#lf8Z#L!4|t zPpg@8PLkp#Ns2`3HHYwu8YlK0T*-t=YHa!2{%lI~VkWU|CL2D^T)f|Pq&Q`LIjgQY zOP4k*MvFJ6xpi@N7%l4qb~0ahuD>YMF*wAQ@9fWPMy_WmEebqWt25KwT_QTVb{M-l z?KGAix2DjZjeCwAgbP}q5XWp~GmS!-L)9bf)#AM%{CY7u)DvPvKHS!_|iHBm(BQw#}s=p>k7ooEq&O^ zf%%wOrUo5WNN;pbgC!BK=vmc}%)?(A*qR{&#OC)NvwLNO;m|Sy2BLAI+$aJS{J%rj zgYRK(qD*q$Z8UwQhYfH2Y4VCvoI|9@z%xg%Y|$}xu+*FF)84BvCzT7S5btROg zyOv-}hltxl&(OlTce$YtKG2d;!Q6Y(wHPfq2H!s!&FLI=#5L;U@kRf&*lVx>jFIBR z7e-i%BX4Vohd=&+a~not*U}t8V#)|+aOerkJMoK+(yu^Kc`p7b%0zE1Y1W_~gr!x{ zC{!^O`B%*~4`i zu*mPO3*Gnf>lp`m8n>&SVnthaIc{;F%o5R@Xt*-3T;w43Q@=I30zbeg{3zL6zuj%L$?hBIH)M0AXh5R7WkXBy#aaZF|iR-6c9N$aPg)}uN) z)g}@gl=2uI(Vs0nQP0M|3SnV?^0`A#zvDVF|9w1c3vpqW@pUf;ET3(Rr&`XTPnR?r zW+`BacrJHm^;%jiPQWQ`e80KuWl_S8vly^Dh1&FBHXgDdP*r5$I$k$>igHLXGBdvQ)>AMecipR&M^(J|i8==D)$J zXfYQ1#RwiMyJBQ|8~=GpU>vqZq+H;PHK~^bUH8Vr*ZLSbz7@5!bc@g%J@?2-s zAGKGtGE*CMR`G2bH;`DA6#AFhWE=4a98D&O<4 z*&IXby6~)JKin~E9c!q{Wq~83*vEX+Xoca*>SKUeHq(Rvy3^O^<)z_ zRxuy;3?)o#nau@tCO=`3*lLKi*j8A>uD>0D&zEO`_T}?bI_E9EGi$^8QPteyvt z&arCORGfD(g(>7NVNu#zX!c+Md$&uItsFC!)tX7-l99FCv$!W1{~00ZQzVlPNn{@! zFEYC-o_kq$0b66>!Oq=QXKupz>_oU8TbD2byCUm3-_l8Rtf-@?zS6xU+=?c>GQ!#!AAWI9gM&%_wb6RdTMC9{D$xHPsmYflblej~0>%MAWI z#&dTWEU@Du9Huk92m%+qiWsgX)}Ei2S7Tp392V+2tzuI^YDp+0Eh9cMtLXzSs1-y)UfzexCha-~{9N_q;`~ zrob}s2(XrUCMes~COUBRH+QT+nThlb*``u`<~8>ROt(Ib4kwN9K)61jckRQg^$i#)Hx(ABOkomzS24Y{Ds0F6tJulk`55;TG9K0p4R5M4=b=Gt?DuSL zSzakUqqH3l?p9{c4ll$ak9zzwJc^xezQ8`z&R}O2STPB;d-%#znXTn9T?C$toABwe z=wSP0^f^8U-TP(qHxmV$UUn|wRt<331l(Cf=Eo{aO%KENTgqi&O&kFMtc4b98 z*hPeZy!;fh;F&Em>i0Uwx8WdK%Gq+>m&(z2jyLX&?vNPHv*tSey_bz|p?~OvCOlBx3KKPE@!zC8;|M=2@RhgVi z`APE@E#tsh|Oe25U;iSA(tiwze%%7jYz=^hK zn(7N&>=Xw440+!b1>C*VXVBX*1y=3)g2{a(*+H2Qwym;^we=EOfAwtjwr zzP}6w=jT4B-|vFp8YAFm(ad$u(S=su09Lco3%~vt29GbpT?Caa~MoBmnO&Nuo zZZyFN{@&q{t_2hyZWUdPTFBn;_kQ~()pHg7rbECgLslOf3S-o&VTC}6TvwS6<*^5# zJnjv5VYL*w^jwBU-{pB&KmGzy_79lw;S{c$Rw$6YzBRk+ill{x&l!uoi$3yw(oD3+ z%mD7Tt_B5s0newDFz1{};uu)l@r~X!s9~OVCbN=?D>x%*!Ggg-fBwZW|u^W_HW9>-7)7`rp$J3i2rFe@n$}1 zxA}wf7!{1vYn6{G-`^@vwu5}oP5nbo zGenK~Iq)24WeuW{&gQt^=MZ%4qGW<)3Z3!u6R1XvA8l(_A&RMf zNhLzZ1dQL@hsS%k!Fol!V2Lr6EXj`FdrgjH<6Z}`{VRQ#pfQPEOa*#d>1j>|&wIIl zdL~rn&t%Twx-cxDFCC(4MMgA_2O0H1y!S#Ke?>~+BW-EYcl%f}PRASeKG7zAt3@!c zB^}n4`P0=Ie2+!BqWI&R-eiFl-;Ov6IbZf zv(kDS7HqSKdGLH3<*%PHX(L_n78=I3*iT|3taY&8&uz-{VEzjhobx9lQe>s3obhw(s5>2A$<8Mw$8Xf z6m_!&GX~1Tr8PeAEX@&AiZpS}*~4si%Q^Z(X#zQAnu5y3mL!Bf-?wc0L#vf%L)YO| z^m^4Dws?R$GjEQ^7ujpTUtSF!Ys*_i?41KeQCm2dql@s*AW4?}sG9w`c?JgbokJpO zeuIJfST>>SHt5S_!OZ#UY@OQ{w)2P*+ciK*6mR8)29={l6ZVdw)gRt*ZLRt2`*%gL zP2Dy2^jHa+25jV_o?WIIzYA%8lMm`$3gpt3Niog6{%rG$H9Th18JhSr1=`I%ks_N> zWW}xkR&hNKN32&8-MAG1G2^;e6pzuc{CpDoz;mtm9?inq)lbEg)mdrhg}qY07V}A?{;Si5q;1Tib#sk)n*{KZ=oI6erE)?tJVMmq~gffl|oWb z%JV?#U7=qU2N1`*Lvh#YNrIZQ$4FModfeH|6$MZCTcr8ML6q%v!8?C{tbI3wXe*zD zvzOK|_Zz=qw`DnxUEoGkCf~v(T^Xcn*Ar6v;VPN(dL|ve+ys7%?MtejQ!GxkWYbNo z*x6cDwr1Kz8j<)Gioyr87PDZ?iY*0)%oI3qehyx4@`IMN974_gh(?+M3{nrJ9{Z2s znujmQYsH&nM#l!Cd9?r>YBxariy9jB{ycb26q2To_1yMfgUREx5@Pn~BUH+AXr?%t zU5VX5*1W!hgYpJ~LD(;-be)XpR%PhBWdrfmSPt8)0?EO5NCsW|2+6gMWN}m=WCb~5 zyP6{jFpR{Ma8HP6Oe92ZJQ+Ff3%6ttA<93OlBmPx)F`h49Bb0x<@pYrcp{Nhp8g6S zt^-;%tkvk@gmGPOEmP`Zk6eUG6)jmJR^nQ#WaYRK}ui1KKa*M}sHk01v7 z6p7!&6j8wD2oTIXEE=Xag?M_okkp_=5+^&Kd{#)ICOj^og4a@#!1u>|Rjns8_&cYA zbz6wWQ+Hw=V?^9*hvWDg`DDpTo^Ro{DW*x+QRAC;;og`~SnE{FMaRfc*NCa+lTs{* zoAX2J@XLf5jcp(+Q$3)lP@A-CNRelKpMw0Jg>b6DA?*Kp_bWU?%+AAEJWELfKF zlDfAaM8)S9=-8`6Nb$2@qOQj2I3?!**_UYwX)_zZce^7_?q7gI@4tZ+t^0A+_K~O~ zp9TAz%t+F5!kk9NgVJFg>^-xbzV8!?SI&rd9Q`2n@vV_CgXcQkA-e}!9xMP!gDqs> zTYX`jwWn~gbO4m6dvha(xd`UYQKjP_Hlu3a2RzxhI7J}rJ@n@1^hDZH?EFqo`qJ_pwha;z>rP4nFME=*pPPwL z_9Q0q^ZW+KMv!ga<*4p#11PS(hdcT9*a=a6KtF>@ za6Lf^ZJv`wQTIjaCJU(4<%5E^_a5V3`B!jf?HY(ZStIa!;z`o4y3mSQvTS~zpHR2{ z7hF(1Lyx2#hS0tH=mNt`oU6s?{%aV0)d3UQ9iecI z2r|WqMCbl}G_{(+q)(<}=QTSlTQP zE{n5?Heb?MOp9x+!1?GcZm^>^sXTs-irsAR%G}jlq09qP_Oz3nDB*-RzuXi2-2EOq zxvUz@})DiGzmn4;{LOIVmJLnx|ElElvYT!0ZLA>S7G;Fj6C4&eI``rLaxeN!+lTiBK-rn|+IvW)7POoAptO*z>=G z3%rs=+e!k-M|pW>vAu>#$IWA>TW7Io`75xmAO(UK^}+g?o_KY*0_|9Hls>`)d-3$% zAtSc*Ob{!`lV-p9zM)wLqcCGsD)e1OWB@8h$6hbiXdi z3QfZPl7=uS<2cU^1awqH1!UiC0JE2yiOhmTa;WkK2|K|vJ8d)KI#c#h*Cch)>^vJn zkEnr>(G6Op_6V=d(t-^Un$SAkl-_MuB!yAo-0|KMakc6V#|2Czp@C$JjSl*ff9gHIkla|?^``iK&4|lJvhS}HjOldrXQ_b z`>plFU!jglG)b`&N6Rsn$p~{VS+e(gwz9z|42hh90TXVPfsCvh@WIUw-!m!p-f){m zd%-YvX{Q2GWnW0%y7ba(r-Wl9+~tT zeD=G+%qUxOqb`(G)?o=VEgCb79r{B_xPJy!R4lSuX5 zE*irB-FVPji)c3Lj-A76)Z)x)%L?*0ildd5L=VE8!_FEJVp zwwtoinr<^@Z>2a3lv89Y|(LSJxRlOKpCN@xi5 zcTQ(gs=Ha!j|FUixek?6>xIL)QIv_lW0A5LokP@ z9`8YG41D0S_GtLnwT{gDJrx&+__NwFOLiq&l3OR>_bxv;_HCIuWOl`Ht;hCoCGnea zgQXgqrXtPlUZ>7JCxqhO(__%})oL_WtH6%^j(D{p2AB9oq12eZ(D#_TDCYim)H%Kr z+*Ym=8Gb3jdpiZ3+a!5b>iQhJarx?2&IO%gn#`iF7qyqvgx;gdVB7COXlwGFs_r{w zzOhvqlh%H~!_T}}*K9wqJh=~Cw>$HVPVUUxO9RebN#WApOR%u97rD>5p7>OE4O3mZ zfnM=`fR?|zXxFaKxc9X;?qU5P=++Io-EIe3%+|yFuqWugr3|0UM*J~Us7~Ju8_<45=Qr^RCDc*mgA2_ z+Iabz5%)AB3rE~kWJesIbN7V3Va=tFOvE+%%b#C5oGt5v=;Ec+SfH*ojr*wi6RmdBQcW)@5 zbH*eb%nVs)>kYj2zz6R{^`SElWZ~3fQqc0g5ZC|Yv3{<+f`Q-9z`VQ$?zB=csLYmS zZ#IXp_iFZhulO@GUH%w#A`-#=`6+mJWf_cmrw3l&^0>g&yYS|jKG-7Jk2ddmPW_`J z@Y9s(_)TO=bR?_6e4{@G6mG}88>H~*_W7bOdJj1r+i}=vM^SxRyv0K|buLS45dCCi z#`e+yu=ZjJJ>L*QfBRWL5r5xXH^q*Cm!=kAhjnP2OPeM60IF{$T(N-bgy>MSsx6oM@w@W`nNq%`iJ^RNSKWuq+&$W$SQK`oqAf-G!a zI|mmCInI5$J}lCkL}SN%!f+KH2O(@cEicc4RgD}d`kbPopVp)%{SlG1B&_}qLY-M#V#5pSUEg`EjX8q{g%#UH)P&ZWye5VIejqO zG3PQyZx3Rk1Kf*Y7Rxg@FvqYoyjo|pGRKX{GL=R0{DiW(QW4<5%X`3TB6r~GXc+SH=?}m_3 zd5I*TX&e^yD~AE=gIUxQZMv>(818-GPTwB4!~zFpxUzo;lx)5SHaA9zI`8VU-Htc$ zTZlIPiQ3FoPc>wgF@He9O9xBG%agCC!>R6Ed#3+6TeQ<(hWXv~W-lY+P^Pi3xbgTf zvAK?dxO`?9Hh1H3Ry~B6bK9_&+C0`@bqvOjn+4nYsWF+d-8e~M5YOj$f}CkR#%{hH zNSZ87h@sH}*6W%coa&m!<~gijHNT>m*_fxy_2UWrzP$xM91^qQ;X~Lmr3{q0^BcFn zkjpa8xi08$@EzYm2ujb4#YtmDXyme*g;gD9tIm&M{fEvGsVH<|!S{WvajH2R8|?yt zYZ3SHXDQJ$9*EvtS7DOcnmA!lD%g>OEVs^$t(or2mS`SvK- za|EX_R2v4}{ll&Na*P$tv1iX~X3>^i)5P@^i^XrMoWyh9RI*jsBU#Ei3yk`c!IX#0 zW_D|1n99VZWPj>(a$;zlNO{dG^bdH(t=qT~9M{La_RnC;u7xq* zuyAG{=g;!PKT*3rm5^aS65WHpqwaiDFy5g+?6d-5#Id==@vM+_S%k2W$*0)L&CP7r ziBrt?`4ATQZZSHSm7_W-gyJz@Kq6`i-Qjf|&EI8l8)_$VKQ(<3du>5YrK9ZrnopeD zK{q-_S&P+%od)TCiBxj=Kz38-Axx4^1M@Mf=&?=5;I`Cseipkw^$WPiPKbAit744B zk=rFfsXYeP)(2C^wUgn`nIYtrvmM?%kpjK%{D7>?nOu5Y2s2+|%@tm7!O!18JXkV)6_901!3CbH&r zKFkj?nVaMZMhCZ|rPov#wJ{P+K3u0}ZIJ@;J{O)RL6}66@VziX`R& zeG=K9g`d05bxJx?kA~w!CuuywfR7d~?_$|`tyuOg*^)(n`&;Rt1ErbnY{;9v?5?y49cjbQ9#t5iZtfVEx;+vHSe>GKBY)y5pHbX>Sx<5x z{ffwDv>7XH%)!)>WXxf6X!iFQ(b1fVtaaiA7QbnnI9N{wKBo+!m%q!AjlmU6BiKl6 z*!h}2SI?(OYwbmj@`ltb=|0p>8Oaz(GVLM1=mhUr5oxNv45*z0Ja zo@fnsH?9C2KfFYRcbT}OArVTe{9vy{E0<`%W3q&7fa}_ypr*POHbqV%w+p05=Rh4! zo?aHc+nNE7)_W20nsi!aQir=|OlG^sEWl&+QFQ*f4DON6Ihc{{4^tiD$VbaB$f^6Y zF6R*Ddz{jwXM3>j1A=S9EVAmX9+~_h5h85EXcvxd z&HBv4CYZdqe3su2_2Zs(T)`pyS-8yc=QwlnV%jC^$bG+alB@6=0-4(d@cpYVxVC7C z+ghvS{hLEc2N0P|459n&%A$QL!2~ zbmDxxJtYPk?s;Hnb_PE)dK~i|1dMvyvRth~lscb>D))g5Qw>H3t^u!vx8Yoqhuq;w z8QdF#7=ds4OH8m*WZFD#La&J)T%h#~I6qDwt@*y4Y0Dy5o^>IMnMVcHQpf3)asf-- z>%ujdg>!el_u+1yyo}{U?#$R@zuECdZ z&NU;pJM*>Zyj(hrE>>iAJ_c-8cqrUD9RP2=B1OZude9|%JaM`42}qZI3)2<|xW?jO zF3rIRCYVyv#%Go^>hX|Y`r3yv@PR&Z974;FclU}+QcdL+kj3NqG|Sf z0yB=J365&Lzz@sPA-it~x9!0!%yc-3-zvt#zGRNgKXpprc3~MzrfE>V+llNSWJqWF zSyIF5GB&(9YB@Eqp%)@D+YAnfC-gNL09r|g98!zV5>fwY;FS-TqLmc3yohn*2>)|3V9~dD#iJz2D!%h=L zQnNvYqrr1J)FFf$bNmE1?PjUS&VCVp{%Z=mS}aMwAWN`YRw%lqokP`cerBgO zmavPSrCZqQ0$tY6XC2#qA-x^q4L|I6;q|1E#7>xp;Ay6wMp zkKGgR_WzD`$GfFF)-4i$=^p!k#k=!=*Zf`gzwE#3yY2tV-|POH>-Kx{|H}2m{;uh< zdvtfr-}b-!?pRN}yS7`p{eQ{7a{r3;tnJSIzw$k`f9sw#J-R#hx837+$9m#D_TRq5 zU-|#q-TvP(kN@~R`M+!aj{P_8@z>Alw*Ri}@q2Vn?!SI_ygSw{6916@75`UlckI9Q z|MLGEllW_0ckW;LziYbff63qLdh8x8VR3yOF11+%J71J!2DUum#C=8+XGt?oLMD%V zkai~X_V_ZDQ{%bwTW<=gxP@#`z7;485W>5&JXaRyXj}DOBPAF4mtj z!JV`1Vb9D+%)9LYIVVD~kiSQt9ny@$AH+e&uMP-F6mc7OU5DMfi($r{NYK-%hx;k~ zEZ?R+OwF(nZMmON8X1Jfr?i>w^?aDW=pJWd)&^!HwTYbS5cu~%c+AX{ z&-1|~@*4N6HIQq!EGBMOJjs)<56F|#dtu4t$>1Dik4wv)h((7E`7^ISbUu`2j~XuE zo~8bDw%K61BkF`;Uu_Bs2da?z@!1fxpd8u)d`NSyEF9ljgijL<@TchjsNQDJ%JhlhhyfsfU=5W1T*`en zeahv-CWt97gJ)TvM4@U`*paOcp8Oth%djw9eBm%?%(_EgZVd$IP`)>>ktL!D1zeDYIyO*lv{Ar8lAUJfz$&<+;-blU@5_4z1bO%`Mw79rDp~G>X(ck zB@e9Z8fof9Y|ElBB{^w8Qhk94U$)(jK|u;siTq}XYtK~nZzox zkq#X+wX+{NV`s|m^9=<_lH*9Es}fvqwjvK2TWIZ{iL_&Qn#is>6pxrcfKhvsN$grP z++sV9uG}e))9U7vF>5x1T)-)^@#PL~yYm1{zwZSr7I~9jr?p9~(j_+HWr;{7)|Mk( zEwts$DzaB=Ci!kw0fE!Ukda?^391tJ5`o1O@O!5Ki^?;=W3B_-&B=ktb-kgX!Hpbu z=CM;pn~=^0TI4fCi;Ro{&^G8BbWV@vn#@k1R>3*);Mf%MQ7Q`ct|pV~8pcHBmOLAH zqJ?|AVm-N*tx6(a>IzfsYDl5=ZlcBa?zc{rVTCsY*f__K6y>ME487iDy!U+i=Ex-K zw74qA>f>Qz*>;$P%{dDnHpIel=|7^1@H;tQcT9$Z>r#j?e>52vlflm<`;squexfgB z_h9d&XQU(JCj90(iq>eogxhOVLB%MZ@9nN6{mDC0sxXQuo<9#)`8#TzUz%9fv4!^k zBuC!-G=VAB8wB?gfYW3ltf9J+xwS80&mH^SV?VhH)ZiRg-XZ0`*b7`W*=k-ecNT&Z}NTc9a|0*N}VQ@sV6jQq(O%PquE z(;GKh9w#bSMv!BZyU2)T`oboq`!IG6Ba;%;h30i5_;+_r;hp|K>@Du&^yxMvhP%Pd z3Vz4c9eD?8b_3y(R9_mZV$OD(O0YMVchRDcO32*nxYR3N^p)c`xEJvg+m^4T)joa5 z?8VtIv~mV^c1;$l=8vqWK^1@E$wKKL>l+|d?PEX$d-&kl=4#u$?A9h z?)B=Ad(XM&e4pp@dB2b1=v}Aq-@iF0-#1!JwWg73TCdfZw{Kq(g#sOCL=1l3p0S2fJFv;g>!opbw zdnN3))Z}r8tnuH3n_&339~D17B2+y)fN7;J_)2jFjw?|@NAn~seQJwNSO{9}sxM7mwAaTaOBTe{x~i*fElYA&!`2co6L7 z>$82UmXXH6K<;)bmWEHhBip~|C+nLs8rPibAq}6KfqR|9+3e)~cw^NoSgUCU8rwHh z^;J2p7`+J})m+E9Lp^ZtTQdrHxdHdQ3dEZ)f^n3IwoKmPI|$m^;(h6jF*mF!vRwnt zEv*#H4{MV2=n^4VDn18RC`k_$4v;qX>W3%i_2r+S-wQ0+kqt*DKA^`V6RG5vwsgprPV|4e5L4oE z;Fos?NKaPbi80DT%m?vW`JGSp)34($k$I{Au$BtL(%}A=FsM#ADS5Mg2zGZn0W*J< zK~;hsIzS(4dXKa@H3{^_z9x^#T483C7PTjLgYGNEPR)}ToNKZLO_>aOn(FcDg_HPs zO($&Fql+(0Jm9E_3_8Y$*=oCTO|cqeb2PAwV~ZJ+rD}{>Qw8 zH~-cN==U334YYCS>T!T$FTrUGb!^w#&gNV<jYnGWlf5u>Wc$hDFURVweN# z7BvH_HRglM%#ostz>|tkK4E^_x5JHgDRj+Vgr$RWQ8V%Yu32)3dJZn4z46Wzu`G+G zy)uDj15LDFE;2h6P4RPljnMeK312MzhV!1uP)_>{1e%AI8iZY^wNqlz_EIoLPFIzM z`CXuVzjncWM+(W;m&?}aUtswWKVM#)_9k+#PD{F;5U@R!E)8^-wjfKcj zBVk~qAr$qx!s2R^;l4*0joA7ew*J%wzqNjlsBR#v_pxN_#t$IZ(~7jA?GDWUxgKw( zC4pg+FHW&dVBXE=;l^4C6)76fl%yJ%oitMxe5X*{iEN-dZpXp(og%+)2GtSnmM}8?S8?A-z&n7@XqzbjXTt>A$ z8U@3?Te0YII(mIx0|pUV(yy0J;js3DIB?G;P+wus72Gss0pqGmcB``Hh_-uyquD`fC|GdV?Q2&bgRE?;IQw3zNbOm#X!K(ZaaN-* z)adoVCz(<-n)6QhwR(nRLa`Er4D_Hh9cQ?atb`*Y-U>bjtuVB@C)Vm&!I2fQvZfEN z!sE^&D0JD%oRWg5X=5;MfAf`=2CDNk8(p?3Hd*P_H-HwTgX501>{*Q!OH!Gvdcx&B{pB*!h4}VZpik8 zXZ~8y^P4R$KUq<_ZiG5T$+t1xdCgedk_7ex&G}CwU1-R6AbX>};8gt-O%wg`hq(^U z{&Sg3O{en@c3py1bv%@gI0NN+EqH15DXJJFqp^0u5U#8R&t^QKlm9i)k6#O6(ce%1ruJbfaild*}M^Rc#T{gO`I}9 z=$)brWkc+dy1Mf{w~oWi<#BYj@hNr}_fPhlG%?liAQ^}1)A3XX@by=adVLH;^UD5Q z`TGNAKI9k7sy~A6qZBZ?6( zx+oi<`;;Rv;OQF(-fTce?ykXq)BEwR|JK2-_)!qwS`IG~1I4Ubv}{_V1949WyyG~V z-JIl%sn1Puo8l;lE7=5w;E&Pqq0ri>#O`|?1LYwnA*I9|x5^gagnehB)>974_O549 zEjuV`rIcnyzJxRta~$)f3a?)-gl9J_p+RInwN{wmR_z8s=f!j?AJvD&jRo!%YKJSu zXPk|}GugNHMFM6nll|FIPM^<3Gyhm+F&|+_yC2)(q*NE)C$0obCVapzju8+(c@<{q zrC>zEHgL%x^nTnbc;9>@%vxFtb&H-r9}`O$Q(Y`nX?j5KgYw+&Tax7X_$cyz{t)tS zYr>5wqv4x&v@Fv15KI)`n|eO}2X*~QY3quE@PK zS?3EgC6!FWE+4-3JWF5tUBre5@n~_m3?)) zygQTs^{WAszoHu}J`VQw`wnIiXK}}Okr5{D7#4X11Gy)|y%F#6@U2YvGXE{kahyg@ z#bZ%Am_d(6p&)qWJN&t<%x9g5m;4v-SD0vKOgB#cMz_#C*mw8=e)W@nspQi$_D97L zGaEb$`DsW!rCPb#3IDj06=x(lrjrep7^-7zt90Tt9=rfD9tc)-vQig41S z(ON&?OG_cP`2P_MUW=LRoe^O9AQbn<$YHX3B$Y%tZPca5&zy~kIKQnrf_TfXatbxZN)!m<>TI6smO%V(oCe9oYvBnq9DexhF)(wD@J) zUBa8PP&6>s!ngm;!+jEY8Y%T;>7XpQ%lguo^}1kJG#*b(^MGUG{a|cz0DVVTgTX41 z@i)I3`rpXJVg3#%KduuMz4l?cO96X^ALnkqvo{eHY zZV^q{Zo;iz9bg&m!&vFq6xkPXXS3q69IUljP6uvHVHz&k@Z4!MDINP&G6PQ1Iq!+= zP4WTS;C_+q6FV=V4{w1&y)&eI)IuAv_w%fFKc(KA%T83eP{ZUyWEc0F?1%Ic=0DjX zT=&aizvm}X0G|U4zis<|F%RJ_vlVTYA3qIvvhXfNz`i!1ZxF zcv(2X`?14l?QUnXxLiwDwit7{v}pKx-j97;{#m$GeVpaLiy$m85qh=uhlg4RA#ik! z&@Cxkc)TrB`0=Dw7^#{qW}l3NA6m+w)et4P9~ZHyQ_WadA5FI{i5`=XsV!J~SZtl5|OP`w*Cy(u-o}&4f(PT!u#PSX9bFdfD)b z_O#!mo2vrpV`U>ORjZQ8rfy`BD^+>J&yiG{B<`?|XyVi!ax7B1WI!6TOnI>>>VE_c|`6T>!?Fn}4okqGJJ1dr5aMx*8Pu*|FISHy%iC+&=_l@zmY!ePxA7anV8vsH(4uZ2$`~beC1&ts z%0aTrUP=>ocW15-t};c76Ck-cklorai4=aD(7}#vq*UPz?Y(`;GyIvL7u-n~0tZ3H znLDr|%acZ%8PVGZ+SKFP0pj=7Bq4ehr1r0Y>i>%-n-No~-9?G_e7>CLnbfn6l4f>7 zUKK{3@nYkm6H4OF`ju`To+MOve@K&Vd|~7GX85}6tMGL3dLjInA|y3g(Y0O9Q1G+| z&FDBmib3;e-1m#J>a3S^Rp+oIcwsOvnN!Jre>7u5S`X3l2Txez>?$S#g2^l~k^0}y z0=W$`VS8&Q`@O4LxFX*ds)vWtqUaHHzIjWul)dHH_}G zv{9AC6FM;q;8#cx1?|p&wI5}uvrG*WMYmC0`zM%MCOY)$UeKeRXPNq=G(OC9IbW0Z zTRc}gP-l!QGnf$%^R@P}Ir~(|C2$5@j`xA%eM@0z&2hBQ{v!mP{!{YL(-i9Z7xVB@ z>SfoLp6Azk?q-ddTWQLpY{?_25d7_jll$~nlGeZkNWZ-YE5;wf2P4GKWY1Dw-SC6z zhis$RPinNjDje-6)RS-IXBjjMAXSM6({S#Nfi-`yC36H#in=Vzd$<6U7pZ~Hq6HAM zcQfYb2BVT;IaVKP!&%r3KNv;A2h|k3I_I;*tKhw4)coZX)jO76KHLj;Gq2#9`9oR3 zj|c47_Z-$58A*+%;wV35CXDn|#Bq2CgA6xFJN92=)$@a>&z>wetQ96$TZKUXlJl7R zvPl--IF-fr^`V~e18Bf!2g$2FMPgR7oxez&%=g{N!-C&ONbTe~Rx;BHi@rIc#%+<4 z->EM(+qREAT=fvvS_BGD@4ln^lIy}9?}IY4`eLygqR3nOifqP36|$bo-wD?v7mJjg zGB~s>94-vrj8oj@Fe*&!IFy&7v0DK1%`s+OayJE9T1|`DMd8nWUGY6Xffr8f!{xR{ z!H6IgP^hdCE~hj?fcGf+JpU-YSy{<)_F2QpfFDA5*#L0s43gLcw$qQ(TWF^FaeCaj z8T_~x4p@^e#7g?ZB6!rH%V5 zsJcK+n7{pxu-Ry_EK@uq-8Y=$0^G zIU8Y6)GMJ=>@HNFIs>2m4pPH{8#LZRWJhJV)6mW=(d~N%!#utTp&Bj1Ni|>Q`)m)r z4pZPZNjf}psSbr`%%I8lyk*h9?8***v?lwA9|)m7C2Z^#9dc+%!Gu@aq~^y? zN!KoJ6nRohL^qw8tZB&g!z)YN955&FOfdh zEruJty5YIgKS5b78O=Qnr1nKi_^5%G>3j1!ezbZ5RbO|fEiL&FqyG*~PY=aJG4nl5 z(~uq6r;6V{n(=)D7O><40kBy!2A03^;Rd%A`Mvu;_>rh3xLPw4JsvI*D&!8~#~42h zFI|mM#(ii{eIB#=sU}P47D|_v4dHJZqsZKUBIJAt!yrp_nMt`SckIXm6Y=kh*Kd(z zIV$4Ft|&Ida-85XUqV5ObNES>2=K}M3I)Fx~#5(6LTu$VXZ}3=)`^$~_+;<8j z)%peXBj3R69S14Z=MfdGe#f58eMXx{ounI&I8MF19e4G7j^8HbJ8WsWz_gZ#^E;Oz z^u4TDwtD+ztaaLk8qKR{Vz<4r->2mGW6^h1|6nM0TXh1b_xpoaWK#HTNn|+mUjc!4Kf=SmBZcL%XJ~CH z@@X60*?GNcc)LuvxY(E9ymf|Ygx3)3Gl(yr7|;I1T@l)=F3?k# zOze4}ROnI<=UsEMgrDlY=yQcS&$J(nMP0M-&)T=FzN?pHfahw_c`z0VZaYZETkof? z+b8Ii6+?fc!#Mqb3!nXCDzs?U(aC%ZaZc2R{RC6#)wC+yv;LN3s_8q?WBrsq*(V4Z z(O>DI>Ta&O;d!ayo>M|s#uGep!5jS=5`=$l2Hd*$0U@Byn!Q&!4*kZops|L8tK3{q z^A6QexN48#~hesUYJLi^279TXE zPr^!OW;&5n4=RECs2q_MGah;*OvUZvqIquPVls;!Pd^&OGg{6Hxb(XZ?>w@f->X() z<7}k7j9(-5v503M)e6bAO8jVOG)yaBjA!Sn?a);s}l=~`~_P6!shOAi77k{tEr)PR_mj|xm)OObytQs$P6&U1yoj`VIs zG3C0Jkk_^_esQu1-{H}(%qpx`nadcZvNu%&S^k&4?AKLA!J_M>P}UpTo{M<3DPdWXrb?STA^XIuPFb$e^rBEn_{DRX+RsO1TydH)F+<)d z`aHje>v8$1w(QP?MU=ZIm;(Fy(b5Uc?E0}PiXNoKr)L9qNI6F58ddrBc20VR(?Hj7 z3zIYLqyrLVet%sSjfyhjk6Nzt^0{_=hubBQyR(_kI(d=QQv7+JeuudJp5Jscdju~q ztD~6FtJzAaOen2(9# z=dQx<_jpR032(?!t5rz;V#Ac0Z_rDlTtR21k}&ImHXPlgzyoWI*+uImyoW+IB-!<( ze*dXS)^3s`E5#b7xFU@?dPdQ`wfE?mnKB)(I!<0QYG}wwIr{oogVn@yHYBS;xLxPW z+kgKiC2{um-?X=2ejlMsxRcsNespjr)w;Wr&itv|+O?XwO}@gj2Y%wW z{=35OTHO>ZDjL~pd3ClpZH8pmTy5c-l_h_DCYS$rSL}gK%M!Z~p|p9=PMR?93#nV% z(Z>O|Xo}xJe&fqQKFDS~9WEM9l^0~Bd@__AA1?sI8$XyZ^ zM6LHB`0&2~iN$C)cCP0JVXpT<`ZYTZg6CWSPqloZs$@G1ICNFA=v5&!NyE7y*-1C6 z2SMdabso7$4vvqpfwALXQ0m`4w5V%eiGtzb3C8-AYi69#Py!6Xei>6Yjjc>c&aq4}C7tdtxP z^xiq*kT<5dJ^BzFv9X1TTl)%|hsyJLHK*{DNf7pSO~z%x>NxbE1YSn>f@Y(2@Rh%U zqO(SDNZyIx)VJpi%mDP4s6u?(d8pbm2u2n^gNtpkys29tPL)riHNrvCkb4IsT)g<2 z_M?!hAAu)~n`J+HDUhq-I_h+cMs_X)HhC?iipf@Jkg`$SG5mumi8ZXyu~@K`4WYl9 zp%^*zIv)Md3(tMu4+SS4O611=Cex((T&>gz{Xee4`cOB?&U1H5)_*Dx?ky;T-hq>- zdh;(D+1*xZ^20|Ovu%^K>hNuFw(z3ets?QL$bx5uoT6x0LvQ*w%O=_9P)q((SX%xW z;AttGHJy#E&-D3?y$Mv5G*-O!!@269FmRhc9UZ-|^2@Hr=*!{(LY&4z*f7ciS1P)) zanBOr*HKMq&+5gkd|u$;t)bNGNEjb)qatm(@Ul;3!K? zm@2(WO8dH@&$SkKq@K;r&g)BqeN#v&Fcn-kz7#GW)a6NMGN|EB9K9KrByu&i@OLM| z1GzMzd!{8_tMtRueulVgxi&4<%f)D$aQ@G?Cmoo%fJ=p4yvMCoych$>biyY%T=D{P zGiyNqPY*UQt{fsVN-5`_5ub7H5IsGh$^V$m<^9zD6V@%Xr+KdnWN-X~px>LP5`}#| z@WR1`IRDmhUf%sBnch`JpCL#16^G&6r|2$|%;`@jF3*PsW09jJnT|tGFTrt6PPqP* zu5i{gN+>?WsVq|gg&bYlm2r}C3TD94v^#Xv_C9pH4ucd!eaz}OEp&f8ju&qp4MvWB z;Mov_a_KWbw!=?$QDjhPcZt2M{Du65g#;HGjD%K0k)nR&C`n$bpsM>6xOOca9d}P- z*Ik=HtWI$6DPjFAHo15H_hFJiMaKPtX5H%~o^y{Z&2q z^UbO>KKC(Y-x^HY4p{LoSV(u4&Y_)RS9q75HUH_63LjcOLK7Q-nv-|J^^LK(`-`q{ zGJOKMJJ++%R(h~)NEjA8x`(UY7)h1G=3t1uE8ci85{Je(gL#n(c`bZI)v7HtwPKF+ zZmoxO-OCBmrKdx2?qWG<@naZvKT=gT zL_64z`&Xo4bhZUn%Zeeh4B34%ps3~RMedf^<1m+@uGeSS zH0dj>evv3Kt8_y-FMGjB`>$}E>Zzrxr*z0pb?GHvOME_0zoN#J!kD(xi-uUW$4O)n7rK=g;*qijlR8x8rZ${37?s-e0=yfF?{xC+E6O#&0 zQtY5>#5&XnVJ~mX^ zmU3K(e#xQbas;djx(FtvW5wTZGn`}l58K^q(CkUDxZ@u!^#9kL=7+7qpKBJNk}wc= zT-t%E9UNB;9wR;X>Nch-@4&p7Mv(WKv7FJZpuWrohpq@>T@eYiE^a$6*gYQ)T}{P( zFH2E-O&0`SSV5zGM{N?*om@ZZ;#!h&*C?qtsa+u(ooYro{|+(fo4Gi*Iv6vIBXGyo zOAtCz5g$C+1)Yk;cGm#2bDb8;@?lINen^rL-8H2Jd0ClLGXtmS`Ke-3t66q8ivM(4nJsu%_>bL~0@g%su2!Zx7qu}K8i_G%h2)b%F2<~|e z=4QLbV#WytY2$rUx+`)v9-cXf`SLp0&uJa@2_B1fp+j-^qm5YfGhP^YwU*7j(pQ)p zEas|D$535=Ju%CD9WH5H19SbK_<8CNRFf?x+dxA7XYXE zux%!8{B;CAB>2L*4IjWQx)8j-9YfbiV*jEf5PLO8m5lSuV>@1t#~c`o{RV}RQcWJp z-+6|=6ixYpC8o6D>;)XFXpLL`8ex>-Zn%~76fW;wDDiv~hZ=ueY3i?~g8GL0vZ(Nr zIPcy=6w>xX(XB(m``gnYkM7EPM;)fBkE&$4@G_)qFlWQ~8Vu~}i5)u<`K2Z)Uu)NZ z!}ELLvWHhN^SmAQEYg`)(*Jv)C;ee?p2hP82ozyGs*l zH%X5ceZY~XS8=*ZG`JqhPiTr+hIM^eV%#&Z_`3 zZQskDrbeM@_+ykesDml}?eOm3BY3qXT(~g237QntkU4aKX5BFlGE#Ave8F$O6OaBs3gBaT4LrZs6HMoh!L{Fxpub`Q zUQoI!h>aUD*CoZOE0Jj7kcmt7_r*;?T2iYM-K3F`+NeC?66D6rgxi+`;JR@S=_lXY z*e}pVy88JN+}C3)t~uX}zO6rsj)qnEE4B{*MjfTNV;)R7wL9p=t4pgkPQx_iIMi9U zOh(%CurqQhK2>gJeqz34&3tWao30_f>SHQByF*v1cOxCDCLM)cb!s^4hBY2*o`lQY zQ!zV9j$N<#!KOBbif1w%^y{OK!v|f0lr2Z3=XL9(wp)!ytw_y)o3>BENzANIb5)Rf z&(e|VOHN?x!U){^=>TT_w*U)<#bKtdo9LqIg_pCJ!jVfZ_;jPlBD}o|XKKtt!?t#; z4OEaed@96Mg{R?casX~@+KLhj8S3q5hc5{kFheg0MO`AuJN(14$E{+vYnxPM&{1ha zw|c4DGFR#*X6)xJtiZ%>8Bja=JajJZh4Tvf;_JE$SXdth*=JRuHYo&pAIJc{JCgLC z4F_EpIgv>;2z?HI!K2I0pxgp=Dl7RWcnf=A>#nC5v+gz4SlUXL-IyT#dR~yaY#k%r z^z9t3tjNG;u6r_1Yda51`(44@_5bj~02wZ29vJ#w4L8I_!0V!R zT$b@3kHl(9<1_|H3!_z}Rcd0V;+HKhNNbb*zI$I7eAE~7w~j$0mtzpHU?6rK`zHQg zM`P#ag*Ywb7#ds9YW`}iBMd2$%`-oFa9tW$BpjWE<*6o5-A zMc3IkZ7^fv&dp*UZZDmT-@P=jd9b?lk*S&VLq`=pe4UCDuOGnk&XdJ`y=QUl_B>gA zhaTQ*G=TJ?zPNU_wNR9q2$5?{z@lNK)Uw-tsou`hQr-3|Xm$T5YOLOgBZiqv!-ve2 zelpxCtvN65U>~VM)0bG_r;V`gb}n`$O zd-K!J*jS}dnH(SZWe{}W3G{f`LC?30;p}> ze})bOuc3L#YB0N(El&DqP4eD3q@;2bH*7V*&u0&^`~f|=&qWI;+LgxxzGc#o@;JKO zejeUWn95xWGaz`(MJm_b2`Xw*dN!a_+|S3rB$qfL(5*Wk<1ijp4UT1Ni`R00<7_%V zxF7F1O_w&=$I9Lb?Ddaq(flIddku_L`yqG3MjRr51SgOTl1&4r_<+vKV(xb0 z>eCKt$kXH=@6)*FARC^XLzH}Y6PzC@K*hWjSmE$ml9kQ)hm5}bZEiYOn!1NO?|w}d zMw#@d+jxHe_!^4Mc4M(lUP6<05V!mgK;2|bv~Fplpw)dJ>sgmgW@8IwnTo?{>EbA% zXOtHIwRk8W&?x4PsG43j-JYtw};M z&G(=KtE;%bb|QZ@c_(*mt*7i)-Fd-9SK+{z?$kr{`0g054{sAP$>8S<>I^ZZ0TY60 z^sM2+jkE7rw{Fw$<)aFrMkR;5CC{kw_91RE;U+s9luc^mo%jpOq5Mv|Ej=1iC3=b0 zVbQE5aKcL$Y_{eJE}6^0Ra2MSE{ovWRRg#rd=c-Jp~5Fki=&AVWZnP)u@kZrPm z%FKTa5sVidqX9iQyLBgms(){%AkpJ)x}=M#>)sPqsVj1i*GBwYjs+cf7s_{9{DkS< zH-q`yp%lN(oJyanN^E=vb6vMM?pI*O{rqR~uJ6nEmCkJ#nYEm0)LxeL4bkKWQpN5< z>p!-zc8o0E)q<^4doO9(A4vYLZ|JpM1V8w&fWp5GM_GYkU!G;%=QYp zyuN<2@X&7ueK|jX*s2e7_1Jb^Ebf>;H{PNmqdc(1_*RSbCa@*t^u{F6B>3{#%aRW4=0@=(>to1`ZR# z=11YUb8{%*S1cDDrsUu({*R~s1h!eVO!43ce(l{KJ|I-gTu2=0l;u&{dTSH5$z_uu zpGf;e-jUDB^>99U2=_L9&Elj2ePeTZ`TqXmZvB)nz%i2vH7UFzJcv7N8p#XRxj}Sp zJnf!wnBVIPhw+owknNg4x~TdAp>iUw9FfoO+#3(LUv=U8P=6%z!5(`nvt@A<=EuI-20ORwa^K)s^A#Z zzmf3bX>ojYMkr4-Ukk0blW@-1JiM&=1jg|;QjFij{~Xf9hP%CZ(&@9@!@mov{p3)+ zqLn4dS#s@Hq4=_T6OM6?V5O6;z}>Bv{T8d8|tU`rRe`rLTBM#iz2WQ09K;`JcxVvv5`n~|@-SYujESZRlowH%^ zhok&Y=O{jF(>gA77>KQ_+Tn#qI(gq7g+_Y$P+AlWKR$ki@IM7`@_{*ZZ9Io12Unt} z+erTBvmM*P+8bh=si4;rf#2vR|jUm)sn^hZgPka zd%vx8+aq;epY@Mz8GnuG*NfcAY07AJMFSS*RZFxl*FlS`kL=+$UmWr*7H z;u_PVxa>jzO_}x|SdF=hHY;M-&jsGNK-hyZjb_51hCh<3;WJ@is3!Rxv%vw2bogeo zOWe7C9Mj3(iO+teVa(#UbZhr@%1)n-m3$Mbm7RdSj!twz^Db+Bw-FQCTwCQHB~DofzGEW6{^>lv=fiLM>Rm-zQ~ye)H};_Jc%HruI?rBycZc#E zS0Rr=;a_GlRms03?Z9n(=+;ECP-W~}Y>Z%-e1i2F(Ose_p9zJ=GqLXGTX2Ja)q1z<7oVS;IHqFK6JSC~6iRg@}=*4CzEuhI}ako6WP<)4xb-!vV~C1O8$ z)*zO|m4%tO8|;qnz(;cGxS(hrk2ilp3U_0fpUZl<+j58x%o>b%|2|AL`pf)gxS-DU z7pT)&z}spx%XX!$<<~!@K}+yt@^UjL%k<~mJEVWvmjShWmiHLU>#+)-#(fj!Nd$b* z?4%>1@gaQ{i&(nbMV3={&b{7~Ng$$;zAM`Ct2Rbd=vJY5Q@2m;5y-+c{+CNiX)NeHQ-1&`EhYw*gBYV8tEscWGen@Oizv0qRFKE`m zK)NzxA5?v}?Mha(~ zKS>Nnm$K#O)KNz~bIlrmhB6J8^O?i_dEt`|uB+!zR&3wSpN{y;2T7-wJ*qj+rE+Fv zD;Hhl3#$#dAab~p>fTC1Cj62(#LtnaZ}x>+)MN>~Auspm;=OEuJ!0mtWbxfWr5$;Ss5Z{Hx~#_TS5O5Vf&ca`2%X^k4pn z898}6oUWNe6X&Rx=^AU69rD!Xqm5Vdtpn%suC@YhfB(2JZI75cJ+6flF7M`dBl?w@ zKGo)OxeCZmmg9rR-caAKlA1HC*-V)RYGOO=|N4$jO}EEm;?B9tHyXQKZvubnh2>i( zLgbk4(mi^w>3W!UnY~vPU!2gvXWA|29hxVFN`O^>T%>7&ylz`)0CLOU}?jwP=>;T*uRXxAObR z1M$Cq>nVp{ph;3jyApCqZo_4MShbYf6l}n5;WA8`Xo$xh55xK3?f59BFTR+n$KCoJ zr8VOd=#~6vmhdT^-<&4qX7+6nJpX&aZT?%$PRR}CKes509K$dQ_l&0SJT)-LzfMiV zOSpG&BJbSvm!%C^k7a3D*blB^bHH>Ao*aZPYG2U$N%geaNdtdYbF7=VoM)x4gmIz+ z+IH(ki0Bc)@*Z07niA8pZByU!qsMpg>8}>zaJ94S$+fqoeJi@5&Z#@XX!9&VCrkxD zJwA++v@Q##fzB|}Wgwrt&xW6{c?;EHr*V{=AG~N%jB1&Q|zB z`E&b~6%;h^qT6@qz08oSDE8&8X?`VprX;c{dh!%8cpoupqx3pC9`VJP^eT$r;M^dg z;#wd0;d25^LwtCTeaBcp&H`+!Fhh%JcF@$hp)Bb_zp_`iMnHB?3HLOf#-&=vcz*VA zIHpyJGd3sVa_bDzsY^nox=Yx8Za%K~mxL!KH?u2VX6(!N4)hGD#@`k>yy?Xc8vQvL zV&8Y;ddGo1%_$V%*&(nU{F=Fyx$)$52foC-m|L^!A2G4jWXr9TV zH9bTZUPe2Y3$N#Gv&Qj&+&EgeB%gBSrCfEv(y}SnPnFpXK2)abuuS&&=vY=gr&Z+p z?dCUYgZZdVQ|^!*C+VBBgg)&qV)I9yr}nN&=CbrDd4_z34u$S?-lsuOx?YUqldi#O zuf9CXqz^tZT@6=e6BMdiv8BQCbfo+V8CLYC9`>{N_y^DVucu6AS9=04zL%TPUNrW9p~H9Rr%ItBiwX)9;QxdJf|8FE4>D@wgA+7}bHt|{F+1tCH*rzS65vD4rat-k} z-fVCkHBL(AP3tc6rlcTdtA0>;zh^0(8Zg-*_Ea95`Nf(qxRB1xFHPt7%J}RFvVI zv$rToie@wz3dvM7OPYitQ^qn=5{XKNXWv_qNCRmOMWZxGb5R=p)4#rVt#`fa`@aAC z|NmKQuY2!%@8|Y-*0atz_ult)U6A!C9?lh6Q?a6vb?BjUoz>KRb;+|t=)plmm$tv5 zb_VxpK!##n_%_A5T^YyV@>@}QpeO<&b_3ikTR?w_^PD=lm9S@&A2lnGqY^)_!#c0) z^v3!kdi2a1qP9;3GJY?l$!rRZe3(T~#h-=J!)4sqbe04OtRQMv7AX2{B}Y46hj&AsB(Jg1q8Ek8+b z+PUMm*qDN4zR7PFB&GGh;o3NHq zHF-WOfLygQME4t?z=gY;aP;duRKoH#LeabME~g*Z7tO`>N6yoEeL?iFa}lXidrDGo zd?j5oL&(RKN3hP?4QP~nFulL_3`Do8&|2GNbVKneuHfWUr1U|tu37Oc6`Gb!Z;o@O zPcx6u&A)&OYo*g5tG%f5dInM$xfqt3_k;eFnRJ@wC2H_Sf|^Z#N58+_Pp^K`1d0AJ zz~138UXNcwwgH7p);xEy`zju5&<<%$Iy7~v5iG6L0Obj>ux-nT-dRBJ?eoAN*Sti(Cgj3#{R(#b+=Vb#Ig0Gxl2iAtO`~pG#;7`nZKddrcrR>| zsYe>mP@Sjv@w!_lKGa35KFS$?8O^>3nM?MZIt0&;+@Sv2l642C#Gqy0o9Wl}6dCFQ z+Ms_GMZY$rvzH|x*D*j!-z#(0BBSuxjpDd((4ExQl^`=o33_k$3|ja_8}In-#%*8j zj<)#sK>Q@5 z@(x-iHUj(fenLBfSEBJVmf-!pLu6mee%j}xOGkzuBulC#$)QJ`5O&w1&R+!8Aw~H* z8}F%TFXVC48*EY7+e#wvOW|xTx{!n*e=6&`pBk2~piehQ@;r&Hu%=A4Zs5 zONwD=$~LB)ST&H<#hA6X;%Qa1s43)MhElrCMCd6@1Gk*u?JsimLj8}{v%1h8+j zBL}x0rI!uWX!IG5a(9-nKk{xM*R#jb!cjuF^-2gGc1WUPUYF=Ell|=H5*A2aWi-uk2 zk??&Yw0QknG&fBa?;m35{vLk;{S`-+N|lrGqxT_6Q!%RHkxj*RMRE=n6X@XW6_otO z?0uy^v@P#2xUE)3n;iU5MS(mH*|7kElGf2jg#`bY`ir~q;4BQyio~jsbMQpZCDf<7 zoi0G$^xF2DG&yWNEh{_4Dc+W+wJYt&EmaqEbIW$vqbEgchj<;P4QXgx24BbZrXM2v zHqwmUh4kGL5&G-YPHNbejxxI91d-cvar?=Y^vRyD$Z?%E{_P16IR6Egr}_ontdpYy zmEkzgIv<6kO~rKg3|MmH9C8`4nW}5t=jLqcgmF)gQ7xTB^66SRy0M^J5VO$<1QmTe zo@;d7r-}n4G?&-TYgq`Mbumy9uv{rBj`=LaElW*3Mn2vPQ% z9i7}*f!l;~`18+Ic6zd9`SKSoAcE+|BZJ$})J+ zClih1`B_SZ?YT1RW29KklwLhyM0Kat&?_}Vw4LYWsDFJJ>+!ja<>XrAzor62yH9e) zE@IT{s~&X;U5Ch2AKdc!k7ed+Ku&32v6P|+&J@?7F{Vz~IaeJ`YE?keW~rD{3&IN( zH=`Mz?%ZhMW(f3%q-QJQxUy+dkP@`DI!=)a7t`!xcZ&)Lu+^)eI| z(JSbzJ%dCx#GsVv_c^)4E)X&5G5MqG2QT{%aao5ZaN>e&)VDevQkM4;r6V3RXhJ8I zDTv}CPqA=TOM`-AyWr}hQ$)YvH;)yFMJfRjSTkxdu6z`UZO%$S*u8B$pG+%g-?~UW zSGUl~vd;KBUyFCF+6MakWI1gKO|*BcD!jeCTX1*i0*Rh31y$rB;x9#lxmoJypdzx$ zI{g-x{fVbaC5}+xXoH@f-GID#{-c4leE@SqxptLXU+=kfMN3naGk5YIh68J+v3Nt`-Pfy;~% zZv2-`bV}t2)Na5OAiPg!lR$g z!g8sVtefLclC(6nK>+|aLYlje3==g%JRITTVB<9dU>Cv$Neh!|@AlzB| z0BZjvQOB!u=-UquxzFPhxgR@N?u`8byyoZPsDFHyA; z88x2T`UY{k$K8TQ%0EHJd=Rc|Rj8Ym`D#GwBo z-d$~pvqtWsJHIVNtBMw(nJF4%%I-<{_p?T{to1sqU*8CuDxKh73$KrQ{hP+~95uUgP|8r|_7%UueiL3x{#mXno}{vF$qxH4@?=kzRss{?x&9 zd#0mFc{f4nvJqNkT#BL&c7gSK0g<#8MV!+Ca9Hw+dm0&xCGJ~-ec=)BVWKRoQ%HvA z9*MAdrUH)E$V4iy65z)0Q^q+WpNq(hB1+clA=0uNVV%Y3*p)1_V4j}f-n@^{e886) z#qI|u(O5KH?-sA^a|CzZf51V!BYu%*fMue2uFjbRvkxQj@Yh2qZfzU4_rna__i9k^ zcG+gA?NMM3e$ar4n=_a}a+ICanTl2yE8=K$4A&;!|DyC~#y9I`oXfys9eHZCwe|dt#C9=EZb@ zeFMHdHcwEinGP>`t%aF8;_#aMUQ)5^JDL47AL&(h5xIeLoRQvG!FG9PKCc}{`q~zF zW>z8knB+lLF1!F*MWztpm5bIMD#W)sAH(5*Q0|4zbL6=k;b{em_*r)-?OM$5lWns{ z1#i2!B}Ga&W6?8Un)rOfXcm6!y$|Jw-$mQTN>MLID`Gqgp<`+qXvNxENdMl+8fUbU z={IcYR$lA2I$aLUe0Yg`y?cY0oVbgM)?7!QY!c9^=n1H|s0pbkhvLs4-$ANh4XBGN zTzYovD!{e-IhzXxVnu{QzvS&5+V}+2q+bkmd@EUpMw!@N|7NVkB6Z+aS zn|qCvurSXrb3G`NoB3oSyz|qhHaUx6jiDp%xMYsq+_quG#u$9Dav}cnH5pl}iy-sl9m2k62>;#{?oX*BoTjbF8m&ehy%WvUpx)5DZEJx>W zo<_|*(rL51X`RlqS&)0RgjBt|!wRi;W7jTt2`%h7{#@CD6|M~+e|(n1pQB=QTNA-% zEUsCIc6@LKi)cj_Ea^-#uf<+HK)@X3AqM`N@hVWz?c0rgHRd zw+&kqCJWAgR>E7Mh3MlhBUFpVlCo)@XqHJa=S-|1R{RPw()308Q>LSdxB1U7;u*9@ zwSjr^Y_#U15f0fR3}eM6QH41sa7=Y08d-Cd+cPT$9B;S7=I;||rhO)NXOtm|J*H2W zj#^1Qcbm{AWQ-%7$I+`pk#y1>B@}P;k)(9ALhbZ!I1{lS`8HTl{q_8_%V~z1wHMHB z7b#lUnTW!|XW>sRF{pZj6pl1Hi5@mIqtGq8xP8C&(hVIwP`af9Is;-MIVw)@!a@|Z zY{ui1OLySp^5c~DFQ>at?1nL+af0tVpFy$jLzs|u0n&GAP9{|74|V2}xE6K?%0JYCPJV4h z?OryhD^eV9nYR%Yrmd!pVXd{~LkkIfPSLAHVqB*3QclLC3WbbyhSO)Fz_7s_rF}P` zt-S8Ssgx%E>NtdE%<@1doLK7b)&Tq7#nSyYny}Tx4~1Teqf7Jj(H1QaoKg}GDtyk} z9y$efZ{7owc4nj2d9zUU>j9Kh^d5$7I=D{r9o)jui&QW<2-+^OwA()wUC3{RG8xf2 zoURA&a?II-B{}G{OMK11+EBc>FAQZ%UKgZWXTaKNqo60>m0QuJhi1DwBEO1@s4ws( z^4=}YNxkgglFoibl{oTUQIau ze%^Oii6#qr@r!3OVAA+kXyLk2L3Q5`k`}%g{0EXKec@5JT~&#`$elu8-;W?&ok66H zKIMF;@R~VvGzsec0E%Db@Q1;rFyiPF(z_}HZVfBZdsDxou`3p%PNgWc;@m0fAY@Ku zb^8oUocMSx_E{ zj0&DW$~ZMzoUX=wNu7)L3|=O!Cw7tSDffAd$Q*dUt6YXY_9Q|eJ>F9FKEUp2iP5B02Ep_ShttF5dn&RIY54Mg21at{EroETQ~|#Q{_3?Dos>8 z;sj^-;U3r8SwOt_T5gr1Q~1sEbvWH8h36P9=XR&|a2|!X&`6!<=qk_ksj{^irG0i6 z^z^DDKSv90$@w?nV*d)Q=p^WS%YNjSV+)x*0o-1*6>zk`fYy*o)U>V(4rnZ==8M;m zft>@8q4k|CDOAGY5hqZntUGcSmO*--lEL^&lBJuIG)`5k5DG3=d0kZx-od@O9-kt@=2f> zzZ0FGHWzzIrGa||g3a3vVM$UhBrN9nL_3F3^YU&OQWl}F_(zt!iC~2HEdUJD-E&U#f?rJVVE!kTj zW0fJU?o!96(r$2_4#I-U%Ryvbp+8dI*F$U-meBUc0}!RJh7MLpL-B<1@bIx5f7nU{ znLKyHGcyG|tSF1CWD%CUW{ujmQAjg%K#1o`H^~%1zw9THX){yN@fq!KWJMy0_x6O} zb9~6F9)!o)BnW&1N=fez1**I;6&Wow<`&x`IL>1WS8SByYjC}=VasSV>-SyMq~Q){ z&Zm&pxAV}R5C?Sdk2JWbic_U$Uy0C;7L?UlinfLCVqYi; zN!$hpi$#dN#zZn_UMpu4|Ab3-FGur2FF>!K3J&@68)+O4qj4DntoLR&^s)Og8s@n^ z=JCIa`@zw8$X60gant`(3&Ot{U#h}1S4_Bqlz}mteiLsB4VA&xG z(mNx``qRo}c{PyF_51MZWjs$sgDYq2YK=;_7Q;d_UgKt$2yI`}h|`M9a7RTT@|SrI zQDTyGqsCs`f9({Kp{i6q>l3#-rimN=W6C}8cthqSoVVOmHHWj6N@69)IMdV1I38o4 zEhyJ4;1t^XYqnc{CKWH*NH=q9?f!F>a|VLSm~k{rA__!LswqhNHERGmt02{&(E zAzU8sO*`L|V4CkqKb-kZbr)N69%5o#THYoSX`Ba7yG&uR-7T)|%tJ;lYCqh(?g}nt zz!f;eazQ6Y&>4uX4}a$6;lmKypLB z6igf=(Jwy_&R%W~-7<3ooMJWTMB~qbcMu`48n#06<66*JjCad_B10==|T zLa6RHiXP)eWzT$q`s&4W{G|ir=$i>p)>{pa&-v52& z7JA)@h}N}!&UftsgdQTYIKPiresrQs!~^Ko;|%qBCjd#i47gS!1erHe(8cmDa%)#E z>^LF=%0idPr9ol3c9AhQ?{|Scp>|Ya$0l@W;(qR0;zBA_xst69`Hb!v9fb8UGY}nP zMX_upm3SaS!;haMp6l8;x2dxs)+YkiCN_e|r|E)~nPTYGFIPInZ!byFwL{Ye-=j>6 zI9|JMH7xTFtGzGeh-Q~))AJ4)bkk-rZtB}v+`bkexO*Uzd^wd1<#!)I@}oU4fGf!- zwwAm0zLM-&;=!N2oB|)e$zcBA1?(z}rdn<%xG$yGh~LIqGI66CUg)$CTuLp-wJH}P z8gLe>taD&eT{CRCQ;!lwXX0YPax!DqnA*l4S>%c1Dq6dvhJ;2A!Wb$+cUiem@v=hd zl~D#oiv8f^)Ik&~&Vg;VIh5*4!rA5Dkc|BiIJD6Yw0NE^)Al~{q<;Ysl6wiMrN_AN zgccH1lnzgn9uuPx-$=^R4zLQiA_!bn3GUC*=}y^f+Li(IPJSJF_99iF@zw!(n;L*) zZv=$iUqO`uj&qA5u0S4OENZF8SdfA+NH5Ok(vEGdMEWK^zQozrYWv4e(He5+L?}M z+v8K*WHtnBCn2h}>;#a4-6+BP9%#i5Kwt4(c3&Ok2Hwm^_Z&^>-K1Q)EBiRT-_wkj zZM}rrj!JV*clL9iE}tNuYi!``nnLpObi8hAAR3XUMkaKO<^q#< z;ec0vEEkt`K>5jLux)llzUwW}_tBzAc)u$=iJuKW9y}y%>*66-{uMXxq%b^LqJe!b zI1$I|5d_rwp>Xslx*{Qo<+RI?)09L`YojT=l}JRwt0$r@`u<3N+ig@dLBQ)lXyf}{ zTcBWhA2c3tgVhTb;En4w$?cvXTpCi)%vC ztv5%L$JEgO15KU(C(WII=Ks;$`N#g_|7-vH{%J!0`kVjw{@Q>1|L*_e`^WyB``iAG z|JuJz=wI!>zWfr9LSVm3r4XB*KOQ>@n0r?pXbhSmZvZLw6V0l{VrVmdHd^j&3 zFC4GJpLNO_ z8&JAULT$qbTONnYfP8!jvejJ(PhMug^biL@l0cX;{C#w451GA>(&_T;fEv`dqSuORiP=gUToJz?A6rol+x!kfg|0k|Nz#R~slZiA zdZTB?b;w7s3RN34afv5Kpuo8u#O7K*3?IoO&kCP1rdrvxbJUbcg+Uf~{VU=2`-T#$ z`>Hfyb}VA{o@e&e86&si7ideRAO53z9(4;PA|bbtba(I)8q`+DKVRPkk(meZ&kHeV zIPWIgbK?rvS1N~-ZM`AxUNTI&cm+DzY*50uy)eFIG`+Y0LC%D+So&HrTA6Z&D;Ji7 zGp6(C&-idQN;egaPX2;U$9+WkN=L91zfbzWQ4?>pNhi-=b|IIsp7@+<9~yoojAut% zLY9^!bex<>hidZa(4A=Xs-+aw%Zj5Nnqz3JtuCEyc!*ANKZr1oA;0Gvi1MrE(*?SX zM2Xtr^h>+YS_KV^*cr4bk>DHo$@ox;0}eQCOOrHW!DXi$&7JrIh1?9{d~0M8jpqL+ zr8ziUcRp9b^JJ^I&V|9c3*4EIp8~l-Nzi?-j%|i#!VO<9h}KSnhty)=;e54 zJdpVd`8Yj+fERDj_~nY|uK#R2Unv19qFUgYhblI%i`UkD*wO&rJ@WLjp-> z6#6h2|5-B~7g=Ay+5=0Fg4hFy-mFT$O`C#Nz0yI>m%nhYukSz+AC4hLHWQtEW(?4` z1~r-Z!1Tx;&F}-uFvbp5_4~u`YRC!&e(*n86dh4{I|WZwOVGWJc(C7^DL4_05#G258v2q^Tcrg$%9G^!@%z^k)qLANW0 zW->i`bR&FRf;qoGE@)+X2?YI2fzb20V6b5oeQ(d>pc+PCAX6Y$TN$Y<^B0lzp*#Huvhi8f2hp zT;F^-I9_}f1;jX$H9U^w>w)#?-47SGh>RpBBWWEnlez}!cCUuaMC@1vWdJui> z-O3qV?BMPs9VV7(WmM?gT57)*L0CZ;SUeOHj540ZrJC{BnCqFSG4wdN6Gh7K^@;bi zW6`81-7xQXGcafq_g?HKFv0;;6Wln%OFy_u9>=4fFM%89>Y(bQ!<_o6F<3tSJUp6d z39h4r;PcK2R8D0Q^|fpuZU;{D`F9~&pLD-gJ1B%Ezg|s!K4n9?WNl5){Wmc4YZ@Fb z;jaPyyE(%y8N6-cB~&*&AAeEYhB|U85xY+k74S9O>Oc4zCS7IPDtRB92Nshx))SGp zQ!!dJe+86U)NyfgjG%G)9kx;D6gjf%J~Yj_!A0Mli}sl`kmF0f&=9|B2-Ysfi*n<* zCnc6pD=vm7Uu>kw>lRYmcvA$n{8`2MR+v)P1W2O_=?ET^tUVXu>Kj+|bBP#ADZLB2 zcTDKC9A{2yPYe#PoInnRvq(2eMBq^H4xU7%p!v@hVTtf6ST*&fU`U?lzZ4yZuAV%| zW#!4hR|&bgTP8c{14mK%eO?J1h<#0mmuM%-4RXRSGWIs*r{d1zh4HELeQc zkQ&=*(>cv~=ysI_TGAs9%iN!m^+D!Xv34YO{*Z)sEggd94I&Wir2?Nn%w%s}=;n^} z7*n11b|meTHmF;_=Ny9T!E&e&rC;kOx4LZM`ulv?HbaG2|6rg#$rBv5))K`bO-L#H zMILm^)JnN_*7S~ThfQ~Gp+`PisOeQWmltuA)y{X~3KcDRy&r;FDzCx(T{5(%Dg-9H zzQR>buOXhz{GJ!*3_CRkxF^dCz)<5SEbfpZQ#Yxg9R@wb$toH&=PQEE`%>hyp$_)f zsnDGj1r~3=i-6&w1(59-L(I4Y==ojG8NbqngIR0n`i_2bBoegp@uUKM)G?X{KaS<^Z|&y^OM|=0y0^7QrExC#Y?3 zfID=%RdB3Mi=-9qMV@yHNn!J7syeTV+i9P`Nd$gmWq!7yn0yO*z@U&i+Mb29AGz?$ zpTD0S$fi+;nyI4AB&?sG2=5m4b3bGbk<8&+5Z9g$oSqq|dikRV^+(~_b0_+$EC)rL zspL|%M7f`~PY6xtU_yd5quiuK-L zH_+D`^C9?*4ISezjl}5?ULE=xKhkrfhdw7$H>bN?>|sZ|V(myPvSEXj{>mfJIwF8( zjI+5Y9+>BftAw9j$N8)uXO6~uA%*XsleC4xD{$Yb=xhtc>O#Go1V zyw;auD_5W0hSfgJhRDdHq-4oXlq+RS!yI>^D>*ZOevLqT(`lNFkiF?z5X#D&z~27yD~EILHCcy@Uu1@Ayh<1?-HXGv)Z^D!rIh1IiJ>9 zG!ezDTck%O7OFy4qrFmx(EQ`$koh=c^ig~%F1=`jzl94>;OM>lZ}S)aa%LU9cqpAd zlNX=|1`LQ?(Zs!mIas!xpuUw>=Yxt zCs*T|DpO2j8t9CQJM>EZEm|~vB(1lZM=hMEunI4yQZuigxZZf8m1m(Xe6Rlw--Q@z zD^v%Od>x@xa67a;Z-#>Xt*G(tbh?edd6>jcpvSHoA?BA1uQ6o^Ta`FFLV~{ zJ%zd7nu&OngbD_O*?5fHMi?1?jNUtKNL8)MVCNodTDvcU*LfKhj5zihdM~Y~nk_ts zmCksm6-lEJ^5f~fUv~t7ViMSIwFI8BYaU(ra49a25Ve|jubX?&7mX+LTyfVw?59uP z-y^Wcf!5(cI(kQ zp%!#X>Ta0K^O~*nu)?DQTFBTJ>15dn6G69r3{{QuBd@pA!SdT@Ir*$scl{E= z)nVPBW067JPh*tHbGD5rRia@zO7xtM0{28Vh3*?U6E)XsLzUeG*d>SusqS4kS$`8< z6jM`Ed+;`1d3ZNCE9ZjZ`zy5@BwoV-oi}WlGr`W%-FR=>9ylDl(DL(&5nN4RGV?5Y zIhQrc3GMn`3UWTH=&w&#(9bvb=$32Ibm1{$sF)y6(>F^~b1fs9_2d!2i9slel7#c-WP)3tNoy*Z$zK}+5it{ z@1U}~lCa6^liZlClBAqtp^|n9GG?2hev6GT%QJ~HTDFnr5Wk2$N)xfCnmX>)P~|zR znviK+ASC%?{BVT_WV|y$&(-_{U!K>&g~zAxDBC{#T)G)w*ov?}&$YKIaUpth<|{YFAiV~_v8sn+YE0J^rfPmLKwO3NZYfa|igWi*|V38KKm~)6yHx-iSHONJbn#yKWOF**U zMtoy@2RrH2cd)Wp2b;zOVAv{;8zKX7>-cX-{ML8GBqrcW&zZC#n2^3Rwy-9*ojY_a z4IcN3;8&+hv4ky)weK%Mwyh(utyvxCaOncd`(R4V7D<9@W);lvS<2?J=6K@d*;xCt zE?)ZJ7LuR06djq9M)sN(;gB!RSlaU|92wTY@?S>~J;h-5ti)1u(P2Mn%|6YkRgA^8 zD_wEGxCAg5Sr4O^1e~xy&$P*NZq$yp>u))aQ6)fJm{^1eGL$uL$i>=c0=s; z;y8S-HlurgWP!!)QTX!~ZEW76gL3SO*aGcbKSA4t7uJ$YqvM4*>69(7)tiggAS*NVIw^?!Bgu zM?c8s7VtHq;-`4rrqe5~TK6ha%T7beJoj9-$s9cPdJbdg@d6y5RiLbzdxEx8*TC}G zajxt0L}31GpfeSP>CSW`xY0KoH+>1g&wd^x+7r%0LYsh;&R+=<^SoeO!B*;+b{RfN zPoml8K2Y||gC1Wxs;;%xsLtFpt4{F_f?&Uoq?wz9X0HsUx4H*G>h64M9^XphjY25t=dTf~kjUt=qYsYD{aP?IP z<7VT|O_fNaM1xjUOsUm-RVUbUwuyaUsYN#_E}%2}I!R!{Fw9Eo2j4jbG{T9%&0F;} zJ~f#$=^Fz5!v*xBs1sxGfybPlu0lSocLlcWNfH!V2EFHn;OD7(@bf|xnmJXtPOI@A zJ$1F8&dGcYQkiq;aR(qRt$R`IEHkn;=`g5V97ImC)A1`I8EnY&4E4Ha;DnKdXpKn^ zEHqUlZzGrU`kgv>&_Ne=t1Ts|w~FAA_egrKeks+w=7`peHpgleJpQ~~m-cP7r-Qe= ziF8&QS`{OS@8mc`bdeS{TlNTwG>*|rF)CcTqdS@OT@noE)^YQ#nxRCSe`X5M;Xdqt z2!TWAX?|$~H*#DT7+b1Skuya!YUL(+wsZ}Zqf)fgwHqdF{0%o^XVC61K~z(_p1$3n z3sqJdLHBxvpswaT%1$i=%jXp+w!NBK+sf4~C0=!<_r2*~p1JYgt{^X4aoW)o$m5B9 zslzNY{7h*N4i^ez+3BuubzmfYd0P%lB6m}n)-dYfJQgCKCX<@bI>8Qgce*ZL2-Hdr zg4FxTbb3ZRT!?U_W>VIahgHHU?1t~y+v0Qnv+#!4dnl?+m%2WUg=6~*X#?F$Z59^7 z{H^1t{mmFU#>^3J-04E|5{qz(Y%skc*2yI=-@s$@n^4!$0%ZI(7LGJZP$R8$)UOgF zi2f6S#NHpGj%PAxQ6GZvg+E~DNJ*|U-2;wo-V2u>>C!TnZ=7xPIl3*UhA!CHKpVyx z(2Pl0s7g8#7JU?}Q{G@gZ<>Z6vqjnXZ09kgtndyUF*(Ui-;deThkTJ}RSFt&;~AH% zs0OuK`{C6iTkd08Ix)JIhE(1!76{(Hgw3hfKzoiD-K$|mHLHJ;D#uCm<<=)?omm|9 zDQ=@@CUy}v?kLC#Jm@T23hR!!ll24joP1d@mP$9oOYH5D-OoLAucjw`{zZo#eAf*b zo`E!l&-r%$k;1bZ69=PKV9l;qOjm zD6g@5^D2~<6Z%wLace zWk^@%U7#}8a;eewL3YcKJUwC4Lgtj(v+}3cfYhBEp!RYl;&n%{|B)-$C10M4@R6iZ zCq{$g)+S{1V=^k=l8eU7X~uVd#o=#$s#tec8oHoWgN_~2Kz;|4@F(Zr$lJrws$Wfq zyy|?$)SSP`2^R0=cI+^xy|rYdxv);lHo~z2E_Av>moQxJ_OeN-j^!e+x3w>uGh8B!+ z!<*0AW2brpRPm`ru#U~+G$sh)!2f4Wq;>8$Lvz zC-*Infk}@(aT*^95v%gR_NNQodhRRwAkmNRSBjEL?zhSJE6cg_rm^sT;z{m&zB5IU zIe1X(1nJzoot+CJ@cGj&=-Vhnz0d6e#;BNh+WsVVxmwWk*oK?ES%jo3im__4vrwJ& z5qR<73F?gM7Ch#kKR+g%;-q30lSMKwpn7!2gm>C9*NyzM$H z`;d-4tM;Izqt>*mbp+jHkO^NWjKYrVhEU&a4KOV)Lc4B@2riE3uBw&Ls;EH4KA$v1=}AM(T>^WNG@#$ccG@0s4r!C zzQ1_nn;(aE@PHDD?YaW72Qg^Ot8e6E))W-}vxI04l%tW|Igr0uk2>rz2d_uGYJ&s1QEx;}?mP=0vWGz9K{MEWImu1R!-BITJK=q} zFACNfKqF+Q;7fy9$kpyDl013HaBxver!iuc}4VetMQIkwGS!38Y^(Ob>_+jo*@O3WweLP+wr@$Sx zn~qmWJE9N4!>}%WEd=h~3i;8Mz@Ev$V>DGE|NLrb%bx~sRy4v~TUq3VIC!Le@gwn^p$|hORO7_1@xb%COATDQ0FCvuEn_K+p@~Un~m0VW#1l?XM}z=S}wJ@ zYdJN142SH@xekWoqLU2R-@Vh=Q7J6Tt}Ek?`IT_g;9Z5p39a zSyn-(hzY!~ml5hnXDqjhvjg*`+4{RdEG|%H{q)sYm2@%ogs&{SBcq48{UwiSYrVjv z*Az0%XKyiTDF;}I#$@)#wEe8!6c;vFfY<^9OE%z?Hme_L%F2cEj6XLcna~5-%=?FD znME%(nOjr7Th93{$;>ca&#k<)gqvL9&h-d7EFUa+Xc<3Ai}BC6#yCAV#MJHt=Ed}U zM&fP_Q=I&cnJewg`aE9Ae$iRWTL0E#4v$S_+KLkx?pkMk?icDz z*uf^wp{aqBw|c?lw4G!sZ(n7)9$#YSt1M)vt@UO{_Cwg%w{|D~%Vdlk01GLJSD78~eWJ(LOGW9qU zdHxQgtR==eT8cAP;%bcRH$%p`EsNPAXkreCJz-2nyD+L;I&&OnGRfzCneU%tnRC02 zFzWVNtn@`C_RYf4tef5(rb2BYQ_trWvdlEgfro*Xfe9X#s&fA9^A2zJh3b0t`&C2s zl$!!ubikPXDK>%Gx`8fNwy)x%Me#|_@?E5_<)xk^_EOn; z_Myu@cKafMrCH)5%g&v{mX}IYIP=K&tb~auSGpmE`66h*HD zFGprDb>UV_SoLDY`jR6Pa&iTuZFhvJXMI`Gml5oMV-CCG-E3xyz7xYLEM|TS&taNh zY-T$0lNmRQQEb}>C3fJUG5c;_3WM~bm~YD!n6$N;46h-<+*YY%v`T-nTc2xlBkyW) z)s^3v<~|KJ_RkphhPV)OF;<3IlRt;iZ#U;&6&rC2!_~N>4_sIq_bhhb`EoXH{Bv%4 zqA+?mEQUf`CRskK$Cg?JftJS09NBFlYgsPOkB#|$gb`Cb$0T~4U`k|rnPEM7_DrD+ zt2oMx+52iKGvT8bBkFUBad#TQCK&3o=lWMMZKl0UO_(_A_Mwk?|Ja3{x{k6Zf2*^y zcC*;o9rM|_Yn>Rqh+RyH#0h4!RXj7U;Q@2KRhE4}<{mrE_y_we`7T@V)0Xv5-No8$ z<*#8Yh1sD>Db~eVj=f;>nwk7bh!q)G&6u~PGZhb_7`44cjE8?Bx54^2Cmwx@+u?iB zGHA{@OTAkimWnGzu$vmh*sN|v*304pqaP~HUb)!EY&XzjU!Hbmd#8o7RfCtA@SeA!KInNLDEc_wzmvp)?Q~ ziAwpVZB#Ovl5_sgi}t z@h31Ow+HbtU9c%pqNuYbG%`GfCYcnF(aCIb_q3qFCKqZnTu0U00VIL6bsYq^eU)YURLC8B*jHQ{v+>}U3F8_KjT@`r*$;(f$s7{Eq zoP;Tf|L&*f4x?S84Vi~wA)BEk0!0^jh~-#dXv|{VSAEQhYJcbI^rbEHgug>uRL z3z6s6k7TAP&b@6C=iGGX!TjSxkZl*nK1_kKY!hNmG-KxRK$v|#guQF)Q1?)o)E2ta zITr&uCCSl?Gt24v=Jgak&yGy~GMeQ#lCIXKF=6>v?0T6c=8gYIjS*TLcWxZlS8u~j zI`71}ulDCg{oF?)aW`p2WHp;EfSDb7?&VD(xpxpX)STbX`Ox#Z z#pu0q0p~wn!ui!A)Tv-amlQmxEOi1oV+Kthy_q`VGtoQ07wUaVG*F%e*D;TAUFJDL zy`*T{AV)!u%}J=~B2q*2NjNu-Dlf~B+EO+02ziSIa+5)$-=4`G(k$ZsD=*@nZal<% zD-97oYzi7Y0#Uf`34|jRNX#ssT2{R#=fmIWTT3Ob7kEL}35L|Qib zG(}Sl=VM2pxW@z{#}{yGG(EV!c02A+-!JkQ)lb?JT4-sShUH>oWy_8ys+N~F8nZLw z2LNBq1{I-GA#>1XoLGCvyzyEE{6ObQ*H9!u^c7Le-hgQy*L5;&KPT8WX= zen6MX{ME@_CXDc75rsOfrYpCXQr^}iTI60&?R@?#{NO^L#tY~_;kz(?+5oHBS5a6i zN+0-KWT2@HnnM{%e$R=p^N8l#^PWIi;#ipu1-$(wIZ}I2KThvqI;w zdek)*6j9ASe7wM}i9F?sqaSe(x|_MYm|d(tsDxd8S-`rz_rX893}ctnz^wce>P6pU z@UIdb>eQi?x*YjyYLfNMJJ_822_HwaBl>0(#+4;vcv2*44(+0Gfhn}lGMWaJZsT(F zWt=d*i)&e1>3-M_dO0$k7NlIF4<;wbZ(IW%*s+(c&)-Yk-*!{4VknHJgknI>0r!qi zrNko^^q+zn6~7ZFn|>+Uo2W(ZqA5rln}YUz@mMVTgVZ{olIQcU^l5koN<<4GXSx@+ zZ+}O~jXoTbK8BqM;<#9)gpnGa$SW1YzmdZcUf<5Ve@j{hYDrs;YgM;&Ke-=*W&03T zvj=+rpH^3l8kB}H57ayFDD7^ zWVjRU-nD@A@A}Zok7sdi{RtRtD8R*cLMS>h5~4q((Rosp`d*EtT*J9EI_x55@BWCk zJM~!eZ6ERkxyVdUM?}>{>{f0^%FWSaR%b-EUj^hFU_nuxm*F<99!f%d51{vkG=d*d z-^}}bO_)ZNTM{ULSrQf7EQVrc7>@0lj$4zzz+v?}Jll8!`m2Uh-yAhs6(LEM6{q3f zQ-a?cFQ6lA25D_`Bq-^UmUJX-vsg{x-pO>=)`AQqt;p<*BRS5GL8O5%)^tc=feOew zWje8w)98N9AyT__kg`UW)BGE26=@$6Vn&x#tiUtirV|A&7%@CqOPLnbR%-*+ff)MP156sk;|b_x^pa$ zPNezK)VvE=vgRtz-g1JtUL0Kh#lYmv4rE6hpx+(k z!Oxo7Lj5mYh-qJEkHUU2{ey3q#PM#Zsr5j!xC0Y(j9_E`AI`ctVA|;geBTsAJKWcj z)y_K97-hh;KL_t_nNjJs@ib!DTnZ7=+wknUtE!{@XjYX;dtt`dyIk4MXkIJrU(X-y;8%;|7jPClHklKGI5FC045m`>KGP;4; zqb|dIN((fDC7}J~3;SbrlX(d72| z95x?$i;8*?Iv<#ded>+yzOo0iX8%TUKVX}UIf6m-6uqj8NC zTDV$)sa(5Y;xc&_bSMD!x=T^qwhh0|baFT6zvUJScXH-U6Y0PRO)5`QBDJ6>dZHCc z#Z!Z6_U$3`JM=(yCHQo9)aKI2d8`d84k2UpnowY^Mnn>fCN z`w`qiiTQewN%>Yt9!`MBwM<+&8%IMu%V;YXMcdSlQ$=|-O)x!1^Db$ShowI0^ovn@ zp(;HMP@(+z5#-FD`&C;sq5sJq3;(o2Zs;Q>ihaiJdr0^Zqlw zjcmp>;W`%EQpLuvJjYCBrjZ_xkGNdzg!$qLRQ`v@3AAUBdU-xt{}w=J?_E52#L8CYE)}uwK?2Rn-j@Kk_=gCy(KAU3Qx)B?sMN|G$r29U9VV8EA zd@i)m!W(~R&CI!^bK8jYj!Do1%Qohw8_8xU9cNEcQ|P*G8EsFkq$}@api!=c54OXw zbes-nJ7yARH)lH69V1S=Hz`tEr~>7tZs+X8{J0`F5AM*4W;kkJg8#rrq}i9k4DLD64Y)YJNY`8I6p*~?{%z5n8Tf! zxR}%KpThQfhWrrKO<5TI+lpj*+{vkEE*Zq=K=`F6qy`+&>;ZE5ZApIB&NO#d0=4)?(N&KC zda`T*^=gNa{l;p_KABHz-sBO#)Qg%pEjldn5047oVXBlSol=%2;V=cVo~^;1NmJw$ z#bvmQA>Mqiu$HFzr%<77Fx~X_rC=*Z+PQW!x%0iW{cUe54-%ylFDFpbhtXu-I?U3+ zK-f~@j+kYdi7-;G{$zy-|JcRR>J%jat*Kv3GqR3g;_51Vd|n9e>+@+>-D2AN&YvER zmM7aFHM*`SNll)svG+zOOgFB9Yt}kiRFy~yB|GWr<3SXy`~mZAj64Net>!l0hR1`cD(nxIpQG*OWI~h>G$@(1qB3=zFA7vml=?ByXm`>DF|AgB3N- z^rAb9AK=>2Dnzfn09QXlPH(Fr_e*g)=h?QMR;=DbLi?iV!KI&AY1E4TR8bQ9mdU)| z6|%NblIVIK4V#`NSQI@U4%X(D-6oSQ$KUX>R9zWC6Xdp&&~;b3`AMG3d?n2xOpUug z!H7P+9Y>A)tc|CB6G*7unr1{A()@A5=*Y7!Bx^mzy0hhc{Zd4{7oS3&_|T`B3uxyP z2O9I`6rBkwp&bfWDeH|Rs*MdHcGCq8eU@C;FMDqIMK><-UK!^fTFj~BlyVQgS#i6h8N@5`q^!I5j z-8r?E*zA8ScE@+tncu>mBqn3qh%{)&r(yT-|7e(}D-|oOp#2N};=xH-63SdmB_$S= z<2{)M{1^o)?xVEx#dQArWitIRgX}uSQ(1>0oqB44pbj@=9&!Y0F`@N_M4ukoQANgi zqG`uz=Fky(n`=e&n_Ow^n;=>{UXfWYzb7a-DZvIlm(i3*FX)N?aISQ;D?C=uLP?Jd z#{SOc((j~mC!Dgm=eot@^Pz}T2QSg=4OQG7_eM=-xek)KR6pO!`Lh;P@A>!u`V%`>Iat?PS@g{4E9t9e=$sKJg zBBA|$7t}vkV)O}XR8Cxg_H)U651dUML+5Gn{dRnr@Csquh3Qjj3jh5}*+ArTHg!@P zDXG69Eu@QrxtD zh%8Gf%={>VjMPyGvq(d($9}wUIfMpZ6J#zEAmy|*CiD4hO=Sg6(ow8?SA#{y=i%_J z33aB!N%QXjPKQk)H@!6Uh!;coWdRnlQpDYJhw9&8WbrX~-EbOCM=wErK_SKW7$VKZvq|~nnG)07W=tAm&H%6V4|-yIXG)^ zRbwY|O%v(_{6w{ZE-M_9T3!mvHM$jH<|LiJ1pRCMB5q&PKo>r(!!9!x(! zf~Idbry$`&xZ!^n9jCs+KJ_e4bWQ_2kG0Jka{--n@@aY z=W;7q#x}dEE0x4r#y4k!Rxs(-F(5%q@`S%4{VTz))Jju-y*PU0`u!F@7~hK`t)~hRyBBEa!illn~wSNXDaGM@Ha|wERKZEy>G1OyqU@otK9Sfc#X#PK_g^E((jk#1(t4H-bzUXeh1FhEz zFlohkB&%fzzU_+1ZSMTCXzad-K=g$T^(pYVE3#EU4NS8Dolg}8w z-f1iaCv^%vRt&c}2{g|( zhwX)NBoiZ`$8~B{@;U-rC+&bja6YbX)MvKw-mG~27AF36110U=L1s3)smQGZ!_W@p zdAFcor_bFoFy&O|SaQ!Ziv(rG#{_QYP77Ym%fj6=`A|P}6(O7C@$07mCN}R`+Ve0Z z-%Nmp+g_O5Im*c52W#=Y%f3n&An5lbG#bo>?d`3YmqW#YT+_CY5af&xqEZGF# z$z$UkpEJ2aH5Z zF`MVA4$O&T4TmqW??$`Xg{`e9&%6t>2bU2lWdOOWQV2B_p#9BJw#DZt3-Z3kX#7}G zm^hQVFR!F4U60spF%5Woh@$<6KOABgAhbIW)`@2XorP_JDxFcR;_V@J<8d)NgL-ym z(HL&sBE}{8x^qwS5EnDV#+=C(S3-|ncJ|nG!{)-OW?;Hgp=C=w658W{fR*cyS)*DaUsx_C}Xc?RkPP; z?y~+fbE(J4m#+9klSTz0E7<80%GQ}*U^VgA*tAD) zpnpt~YB!Ij(KF;pSf|1C z_;}d=AOY!f`F9!cmg!y^kJ2GSSe#8_hPSeqWaT+_E=>>RTSnrDk2#Es#$kEcbQqV8 zMcQa7RyKJJJJ}n+k~73Yf=V-i)3&FIaNc_7V&Q#0d*_Fv&>_mbFG7ro_ zhV5bu?+$?Y;aJosBxC6;Q7oEggfc-GdRI)t!{n9teL4=p+jtB(r3eZKFCx8C7QZ%% z!S9X-GN;zD(+#hgc32-%Ds{$VpXoTd)Ej*|e4i#(jM&)A(0AB{DLb;@^|1=eHv|hV z#bgPD>)s0t6AIAgavrU5UC{dI$21)iS+})-eX0ziI+Gm~{ws%8#)o3zgT2t2d<(yhv*0&c)KKG z)2v$7RwoQwIc41KY(?CD=cd?-MMw;NJLzci`@C?o?9miR@LP$$$Gu;?#=Jn;LV1eQe6dGsX8(-6Z z939WRo42u>{+{ehS~R|z?11y*9BfIt$CjQRhN)l0&>awgVKEs9*z^;jPMaXxxCG_0 z%h2jJ5obPW*08jW~PUQ+_e;l_BpL4`uE{X=kC8_yla!nHmK39my@tFs*YSs(%{ z<4`i%5Ld?Ng(i&HAdYt_Mji9IAT?IUUac|_Bd8<^YcDi-a7_9z}{Zy7VtP1QU9bQ7w{y?6>2PSm6MvH`B1{0ko;ngGp+l)z=8)J;;_bs5EyOe$NSj#>yG-l5Z498YC6FgR%hP~0* ztl{)Q7JS2kUD6h%lld>`VkSq|kkc+-Dou$e zo2a=^6^%oc<7&92dN8L?=B(g#oxnIMNpR5bh+xNy3GDo)`>bc=7p7JDnGJr9;#`Ks zaVA|^+`Z;#l+Rm>ch#P_-nSoP-Y1}ai!=OQmaxoaUCc0`n^ozEV#{+q9QiR5qgJG` z0)=ZVX7fGf=Qj#TCKGV?Q8GJXZ^tZiXR`7$eFFc@QF!#g23yWsfSY`Wnbw|UuJ10e zpof~YXo@zio25++(IYr^Pn4VbL!PU5e~7pv-3ZC)Ly_lbq?azkNVA>zGwmCTJS&U0 zJ$KojXeBzgz>Q8BOrc?Or?L;yVJtQzn)O~Q#;3IlXP(%FgJ1m^t3kp`SHewiD^^X) zhT^$o41b%1Jth0$)nCCp)4G`DG<{5In26H*t_VHuj(3^~IJ+Vo6QAvdL#q_3cyK+$ z-UjD9)6qX77Sp6R!#&Cx_TP*lvwAibIVjSLU@6+)ra<%Z58-std0e6ANLx@K=y?z& zc(rDSU~`ECRr)B>EGt9$ESJn;Hx)AH^|#pjnv3M{|4Ux_q4GSb6&8Yo+Gd-VTcV5<7>jgR&ae{RK3TF71znK6L!XI z{+ddrZ_vZCGzQqH8XF|Pvc4*MUiW8Qjzh2rjyVHmk`z2815_?+aY)PH>G-*=F zV+>gDgy*X)aQ#NKHk0T8r%b;2QPkI!M$=gug&y~zuAv1KlMLD(tw4hT@|4)FLOpj9 zaoTtnW(=Ld}=@Zwk`BlFB*&PY)2Z8v8N253wh0X=-99d>s8-Ug!wzF4$$YyPkmr< zuWHzA!95oD<38q{xC-}se4S_BD@a$+Wxhe>Y{~HB?0ir$`*d7gZpc2U_p5js~QN{!=0X;bTRB!;fQLk%7udNUI1w@t+D zlzuV*MnJUiOS8krXqUvThiiml;O2_Et20ARGdZ?buuu1N$lyu0leS z%iAf;-RzytJ#w|;{+7<;bk3$zKw}oY_?}Nbvuv>LwLR1?1mdL8Gj^)Bg=If%U}5P~ zDCVgH^(8Rcdf$W)uSZ@B#AwV6f13Ms3H_VAo)!*ngiy#lybI;I0MkfvdYDGrXPhFp zyscCby^hR9OGtKZ1f(l-&@kZ`b_pe7mCspJ_rHezS6?)mufWqvCn)-eQqQm9^l!a8 zk1?;trIjzPG z-#VN;REFvguV664jgAeXb1o;GNj=k@rio{eba?_-zAlj)ow1j@ zwrLegD@e-W}L)j|+kiJ+~Oa2c6QH}Vo_u3{Dmy?=^1Ue94dWq3Crf%|sOkT)|# zft(qs-_)Z>UKeogeIn-DBw)Pu3LaAkhp_EAq>j9Y(BwN*X7ZW#D83~b8byywcs%tx z-v_u_!ed)DRK}L$M5Y)wT~3lqNs#3J$qG@=9A#?StV}!i-{964KjnUueB|agb_)Jn z_G075tYyDfWz)C|MfA5Ioz`5Pfag)`P|KO%n&@5(kGPGB<`%3|ea}9PP=vCJC?qS6 zz}%0oi^egUo>7Fuos}q_C`#wXw4!gbE+t>uNTSQf(Wh;`H0kSdvh=>oPJER{zpp=} zbamj!`Js?sUG^mJ30=+oKr_#DQQq?BSoruYbnlI(UAbGZWL*VJ4!;KXeH8VNF`#gH zA)2MQ8;4#!!9LNic>8QCSF&s`=f%&dYq9*trq_yN^g99KkC$M-)E87gGNqP?+m=}(Wq+c*(^eLR*^8$)*kqbYDh8y>fdl61liTv1p? z1CvrnDQGT@xUmJM&&n`OMVvA`HLa&7e}jb_uuhq`oS zZz{sYDzJFhQK*$~V~;-VVeY74?zRij<&uk-*GHi8sFpi)pp%POB4U~KT9!oeI65pY zP8LV5QM2p|T5#+UJ-wL*kD!N;)_MmO=bKP37(pA#p5ge10?bYA!kdS}^nNISjt`5b zSQRU>SZ+vux|3*Qxhc(vT|+tNoyq5v32l6{iV2NVzZQC|t}{4Xmm zSbaX%?XZJ;^r?|X$V+o=6E!$XjUQ~ra6_a|)yBP%g-kEHfgxl+>q?zTk{6cJjt
QS^YpUywerJm3|RNSKtxuh9*-MbJW+imFR%k8vr+ddi*@PL}-zLSM# zKTQj)!t1F*v@t}E^gQzsA=8IpbrN*jB!;=g-DlFiqBt2ehSyVD(>9lxB>kxYkq?b& z`b{YsyxPlpmB+z3!n?3@d$&jG42kk$*MIuICDM6Xn@Pnh(6n1hH0iY{&HB0kZ6(2Y;I|PkQ?r;n zOJUm*|UiAiG`uj8l2Ld4(tAX?D~jDOhfD^)6>@AaV{bF z|GCdPKD1-Swby9>eFO7moxts&Oq>ecg1&B31T8VaxthfyMh~kj``C!e zr|jdDiSJXQJ@XNQ}d0WM4*3;pLWnQPNk z_M@?w34bvWSV&A2Y}P$1II?OXbFGbGb}2WbQYFW@WF%a3yVi&(-(*~>Ayq3SrUsVWIA3rdw%fDIaI$37Ohux zVQO#2*X26c@OuO@w_D+olPSUicpc%`bId?2ljRBRU`;vBq;=PvOtxs#e*ZWsb&aLT z#u@a)L!Q=L{D&axpXmIxft`Hh##B~$v8mo=g4?fp1)@>+1$Ud~VZ_V;1fGq>E9Qyp z52F##tcy>Xm+-%twa2N@nKVpc`b7y~_Td`%|bXk}BOth)YfXT^)a5kKPm$%h%K0^az z3tgCUoId-oO`BC8z0RWDdf3dT0tk&2#=5{(HoD+68`d7io{FWguAx{aHu(;ZYnEfx ziIb4891V-3>Ubabo5}0ivy25+Y@6y=!6B9Dh!Tv(H!&G_-cQ86n$>txF&9ZA%Ggwc z1a@2LD$6;!32M&qP#60TJI>@l@85nj5A4R`pEEGaaU~vf`M_~fFRNAUXOGh>S@tg{ zth_Q0Ekg`*ZT4WkVj-d$YH;7%2(w%@(QP*hxgW=|xw6u%VbT|YcV!Va*q_4N<%O`Y z3}NTF`RtGT0yZRKOM;O3blusDI=)$A%x`1-Oc{#>_Pc5J%|w!!pG%!twM;njJTo%i z&mLEQVzD3kSWVC{D4i6eODsf7zeZM{Iq!9wML&@li?`A2S6JPevj6 z)(Bki+|TBT?Pt%Y8M7bT2^wK`a5Y?nqF1q0a$qwRSj?hL1uiID;s7=DW#U?#Ta ztjA|6+iCKS%6lKv>4%MULei1fUVGD0pRqJUEQQ-%7sd6AP39a2w+a*uHwhAIas^$j zG3eP7jIxXj+>pmMi`Z5)xtfq z5jwB6sCpG2v+D-n5jv{~gF?z-&99hC|U7}i2@swAtsX~BfTm$OUh?eGe^h?h3m zNa6JWzBkk{wr3rj*F`hCWpi2l_INgNqY=rgYthj$ZZz;Qg9VB4xTDt&wo&;W!hfBD z(w02@TC69KsY(`|N3bZF35F98wEyx+nWSiD6W-A?BnR?q{d|I80?+do@TCA^l zeN2SbxvEpfLQOi;q)XlPvXpxGCCfT=j#+!0V~v}6EuW49DXpGEkKZn$=|^p;$=rdy zR-9&QWsWmdemjKZE@kv|%Hmw46rOxo&ZfS0VF!yG*%-dgpLFjQB2Kp;;=c|yWyfC@ zFMNWjTr6d#1vi;X^#GgI!efDE@;F=)h-~i+%#urhS${P4=}5Dmbw(`Ua#xZhDWq_Bs@N@g&bB4e1YbHR@Xj_~Z7g}i!Ctks%^*S7E3 zI&U?IEL?==rJlSOXgONimP0#0mWoCzQ27)!a;bIYF-iQRo%;3EIT=3Wi^qE%=@Fp7&b)Mozyl z1s|WmZq&uFJ4K6FzIQ!28CKB8SC^?*G6@rVb8%Je47_KJ=alMnx!^D(Zg$aaLD%M= zf}<63?A{w)T)pdn9vd&z4*bPrvjJ>eB}ISk@pjFjM`)V#7QH`KvOw7x?3lhEdwJ2G zyk-xl;(&)ZaK0FBqsoygQ;Ni$_SkgI2OGq8Aa&Ruc*Kd(xFtUz?=_05Ug}b3x-1R9 z^O5JF$A#SBKUePF-$mSnHN}D{V%G#7 zT2}<(Wdm@XB}@_FqU5?zmz^usVV_=K5*P?7VX?gyYLCvsTd|HMEFEHBW?W$M3$k#$ zs1gsC<>HB%6ZTJBh{-Vtyg#rA*CO`g!TFzP&Mb%d>T*oUI)$9@bgZbHg#*JD;4K%6 z{={_%PS}lWyr$U9*o>xojp1h)>EW>6O8CCpfL#l}VnF;iRMkE}R?Q9}!yU2Vu0JsO z4U3xmg6-dbi^<4IaM^is+#MG+&NlNB`@HugTgo1=#{~-T_&y)$yEj8J+yEc;+Cpm9 z23(%|2Yyo&>E`q%h)LNYxO6_;cCEqhvF7+^IvG8ta&YRu%D?kFm@&E#;irDHm|LR} zIz|z06N(UHn28U~#Tc|GV|~NY*i!W{=KN^~bgpGX=T8BgfB3SaGaOmlFtFEqmFeCA z1v-4P4f$R3pqRA)0;7f4P^ON8Tgup!qKpjvLQ-FvLI)1#(n&*0aDlw`VA@3d)x5$q zPM>G0dI{`M@@IkFEODlp7r-vJcCbqE1~%^7DQ0~pj@MWva;N8Iaa$`lU~bX#)4@18H{OPNs$58RQY)+NoB+-8S+EsTM8`e}s4jlVB)5#Cd_{NaJ?un9&Q5IY z&pcLHRLLR(ZeiN6ZoI$Ij)%<>`258P35#Z9*U(j5T=oWPlYU_2xv``qV@b=Fy3lT& zV4R;ChRvroBX=v`lfT;w^Qbhqy&R7n1&(+UE`ZDuG0fV?5IDeNpBhe(JK~O|PgcTS zwuPzumBFXoAD9m-Mr(Zm;%XXT5%Pz%=Y3*RJ8GFWf2R!po5+l<)!06%k)$Q{8{G<> zsE9AYyQ2MYQcTAssXgp=YZ0^Pxxq$X+K5>XgW*y(7p49?F+w2^3p;q7@sjPBF0vYn z+K*xO)!kUvkc-0;_&RdFCEmo&LEDTZd`{xe!hakdxK80Umn)!fz#maLI#ApagRro8 z*c_gY1n0H%qr-vfv^{BIjR;*D+lLjaUqdZFTj0*C^>4g5B3NUnMu8r>^vsQD3H!h< z9*{$CtP;%EpQPzk#Z;wmk+Rt*MEoAkK9wqc)GCpD(niHr{bbjk3E_sh41V33MMBNK6y?S15076)?7KTS?kYyBem!Ej7Zcgy z^%YF;PmT*aJc6_4`J3P?`7oSWj_sj2e0{N$8{QYdH6fPE+Ls|%ptf7kT3RibHns{; zeDAKH)s2j(A*SCx#QM!%u~qjgF`am<=57OIcc-$KYGrKC_%!x#@obbI^1_nCTQR5O zI<8$li1m+;!wn_)cPa_LyKdl+eg#?vPNG@03h$Ixvn|0xtRz+fXT_GH$#gneJ2oK1 zLqK28SdsWMbNbAC7y2r9VpK^1aw9*Y`Q>Yod)o66=2O|Ic(Zyhu1Z?KwH`n zBYP8)A+~tdFdth|SL2?+X_hp%k##D0vKhD2pzvuwlpV6sV5o#5f0FW2Go8-2XADQwCVnVT!k^W*Fdel1#}NbT^g3GJOq*=O+zw zt+8w6QyAF2NA&hCZ1L=3U+rEq6`j{?|L-)~+p&#m2NMXV`)HmDUu%_0bKzGED2CTt z&aYWXhvb?Bb0U41_`3~EZ{uYaWp_EvY{PAP2BP=I+;K<=R zT>kY7O%5V7g7^33$LYgngE1yf>cxcr-edXK4{*O>O7?qyo=dNsQsvqT*Kw+`6}+D0B)Yj-QS23RmRwm`r2& zZs;4v!0p#P6tB<0y)QMm>Ew&_C#!jXYy#%~eaaj{@3GL(5T>S(g$bjhaNwpNZoL{o z4fkH5OizSLr4lh`#ZDY<*@b&%n*~O$PJ+dYRthRh{=j;{D|~gAp(v3QX1?ec8@{KV z%?)ay^2m#%8-I_6ZePZNwx?(d{)r7eBe(~jCAf!~8rj+~z> zAnl%t4}oLp#8ON8w9oOZrs04onCH$J1f&V7QVfsQ}yfQCE zWpXZzR>b3+YzpkY7vue;!-y&|jI~zTLYv|Ocfi!hmwtgI)$}+;8OT5l*k|U1h zIdN)vmfXZS=A2C8aIVr*n5(=qNQ%}Fcl+5W=Sb5sXS6-Sq-(L&vkp2oH}FB*ow9?J zX&_XKE=eroeKb*|gImO2#*^-xy_9n_o?m2r3+%s4TuR%8=r7mNn<7dJtu!e5$Q#;J z+Dz9QpHS++HR>;FpazX5exK+dIC%v!S|US&nN0||QHwom%W+0~Ig}=vLgAMK3_GuJ zW7k~b%B`+){+)V|G#?ACekc6k{qg-h-;gLGOvPi>$S72mRv&nUXCj(pamAEo%T1(R z+nymftQ+gUDUzM)V?6!Zid5wqh{il1*4*FYDVZLiAag zWFz=pd)AWN{M>Gu`qqdJUUs5W<3h+&K_Bs3W?}4hS6*WoMZecDI(3|VemW>8~<*ASavKbV|L?g@-~Ev_9ISSjywaTNK_?<-u&D{CWVo7 zYxXFba8R4P9i~u5{bVG49FHy~BaD$w;Qq|p#5Ml(<6JS9tJ$!U8+mCZ7vX-AmggR$ zSzT8tZo4D?evX7`<9^ho&&Q_JB{&h!*A!M()c?$tbc$!t?3u?X>&0GrH?xY)ZoNT8 z60Nj6@D_b6nFOiKAcXAQkCt&6v`XY4NgumJD&Y#8=u{kJo7&T{D`_&bZGS9~H0-+*)Q;-$jc(A5ci-ZF*L|pVm~R zkb`p+%`jKMiTCnYK1Bm9k3`8#RfWWl2@}^WP3wjobM$ZEz8Em5ILH6 zQ;1YeCFsolBJwwir}Ein=*6cdex^w^W?$;Ui&tU1cPO6U1N4zTJWoG&-KWpf^sxK>ER^b- zW78`y?%3NY+=cz#+&XJB(p+UjF>SnGDDXPxVc)>b9Mi;+k|=g|4Y5PKW=B)L3E5T+ z*e&-OZbpeTV77zSA5W(;u{;P8KH*o?2->sIp4WlTg!D5Lr1|Ah=Azw{YVJcedXL~H z_5eE+htY}^8+g7m3+)QoDDQku57xaWql-VN%4s@z6cPRRllPu{b3|Q;BREYfq`o)c zMrZ4A=2Hb+dDj@Ky(}P~TUM06`!OeUtApDR{EqV&+{R*;Pi8R>cd&?>`J@}>PNFNN zX^qucQec7%O%2zgt-nUoTAn8j-*p!$D$n8kS%eM^Q=mNxU-95g1r^#| zrm#_0NiIp1PIYXgvG3+mf0sGc+?+%jqZr+*$%AeBS*U82;$J`jcG;#PFh2>+E7G{n z+4a{Rv+5qp-gxFb)fL2I2GneQ^s&tzIWY3 zot!i9;j}N>S8^}2%_bQ4b*HPH`)N?kZjyGsj}!k&& zK2g_1G2z?(>*&>ffCLlHz8N->UdRrlJ@sR0wK3R9or6qje*bE#=nN=zZ|UTTTc-ZpSfP=Em>``!d^>Hv@DE(Jl8}k`NyE@oTMr2%f`f!#gLA1p{|*`YJfx*%8EpEIZq~Me*K_Z3*tIQp znPlrt=GPiSHjQWKF@LYC{i4vZqXv@m-=S;lX4+g4LWSH@w3l2^mlT4Y!W5WoN}|PS zrztG^EDiFzOq*@8Y1F1NO3am`rBSlvFtr8$d>ZkH9w4H(3B8(EZ7#Rirr1}I61xz=f!^E!#WYA=!(?8#n2{7o7{^k<&>))HFNEGm8*qExNZPM!LXGQ9nQ8ML z_U`Bm7Iyw1%{ARj$?=XfZc#lO*CmORAW4k4BnV?fmV&g-9HGJ7m>hQ*(P_Vd)HLUa zaO+^0V7=+2(5PdNKu0aCsaHTvU@u(nS7OYjN3fTDP4(;El8Hw(H66IlXCW=PdEzPB z*frk!Jw(XKbS$yhOSiPvleNQMew8Z|D;`I)nHasRE`ZnUbnM(vhNb&Vgz`&fg3o!* z^LRF&RzKWAZ|+T}eD|5C97hQ>9;DllBQJ zCK`&vO+5Ui*$%O#%-xBL5*BmDL)!$>oI01b5A;FX#sIW*h2p$ND23GolU~a{`fVM8 z=aaV~edG$1d>l=4X3ysgn+ddU^jPj6vL;W}>Eyh2Gr9Ff(C6Gd(pl$Eij{lloR1S3 zg-t^5?ti%5AWbh!r05Yp`>mD|{N5;|yOVFx&jmN=qQ`aetIML-j;ASRKchvlV`*2t zA&Kp)hez2rd_5~k2}?up$lV*u)0g5g&xCDhD-hO{rU?u6f3X91AF@}g>evS(uG`9% zB+=q-NSqx^c_|KL@YJ6UogGIf7a3Bsj}Zl*8$?MZed+i^C0br2L+@=AsYtg!C77mB z(bNq3{_r6co$EuP7EO43`UxIf8cm&bqp0YGF^RvwN$QcW$-%9iY@gcD%$QkZe0B@f zNzQ_aFZi9f0vpyJrL1%Qw6^plEnL}z?$t6xJsNb#aX`4vbC9;SnY$;p|BH_OCyjVt`Ft;U=qRZy6lhoBAb zka4;QD?X=z3)tvYtt$QI%9u}r7d6lEqXCk;DSO`}I?nGx!`Z9o?y~6^y?!rZw{6GH zaWjMiQu77FsQJRgrvXBnMyRmoQjB2X|BO!R-=-w(8x#Ys1NRO>YVTEyd%7M6n&zNN zUmjl)PLdI4TMQp?gL)S4qO=o-h&?_+_o~B4Q~4}OAIhN6`V+WT5{1Or%iN2VPA0l% z$yqOty4>HBscbjxE0YtNs>hOY#d!L-S&s}Yn(-q0HPUllBCEKTsYw;HVUL;^{jj7* z)jR1}AlEJpeFcjtuW?Z09;7B6fy{vd?C**~&$BWgvZ{4|=R$@+h{b$qU!iiKWzLjp=yh1bHUMDKkpy$n_X{7BG`eHbhjc9dZ znSL`_URNxIbFX<55-IDG0(w4vV97fFSkI@4!o#rT!i#Yh!gucXa~vg0nK7DV-M2!h zzR@J)4)`VHtI5MV#snQ_OfY|=E=4SqqVy$VH2%k5(l_g%leyn%;D-UU)=HXkXZ}N2 zw-RN!eS~2`Bifz4NkjTLr3^Sjy|;33B!c_TqFs=f@)V=*U*YVtyBH!=P+C+K<&?jq z!AA$uzLtS>CQX5Sz8*!4{0@Y*Si+)SRZ#jmKp62^Mo87*-tG5?$z|XM%JHie7Kok* z+dozaH*My#!9y0Xoo}bJ)k%z!UM-@kJ^SfLuLmxicSXv<%~-$GpEgfkM}@QH=w;*w zRo zKn{}7tQQN>xU-xcasi?@Ml`C2Xq~b)<+`21lsC8Wx#}A_)Q)1x=Mcp5xz)H0OL)y= zLG24bo2Hz{VCQUz&D?}f^B&;V!2HP%vi=ycZ2IYZSFM)wAJS24m!cK?waf8S5UG3GG$)gqiG( z5ON_CQITmFIxGQu`T3TK$fS7lEA(SxBYI*#qC4mwhSV&D|Dx>(-{y@oYmPyl>nskM z9l_u9zSR79Bt^s*7*8sZ4gD`C6JmJw& zVAO)*1D$*6DuL>UVqE5|i9iWRKq)DXPB$#s>w z8^}G?m*Pk6p@i-^bakGHjLk;S)=oFN`qY)etX!zJ_XylZ9fG=F1YEn0QmM$F*eqYV z)e}PjmFLLI@jkV^n~Q^aThUk52QQMsv3O%S9vU{n;pIuVO77=1=qBt*Z^a}j1zMHI zC_6J6^|Bd=J9-oH@nJ}wdKSBphgXlPF-R#N!|r`Tg8F^f5ATMP(l1;as>NO>*|PQ9 zhOpksGqnDD1g{;BlYboVDa=(VC_tTjL#AVU?I=ur!v}@BH83>%jK-5a2no6a{Cfo7 zxR1!MEx_#d*U*a^PFHSzg6zm%zIX3{M#)anUOk_x?N?A!(oI-iJj?AlWq8zh2IhR9 z-C=(N@oUqeJmW4Dj7m`Jm`SxbM3W?aDY~sM70p#8${R{jZ|g;ky9TojW8_(IcLq(| z9Y$LMqNp`A4puAnafhn#I(DYv109YEE>Ly^Lw_@#uuK{39>T`PM=_17koZQd9v z?Pow96DHv4qFERcU2dKMz`-m>mPfVPZ&klgC~;HYZ*FVwh`k@7D3HP6)Oj| zLS@KH^p7ooT@uk8lgTt9*ob`WO{refhWZ`|=R7T2N{bOu5BJISdcMZ@c&T45*8J9OhEY1N}`lyju zZ(kZLrb%yGIpfnc21|u<ev zR^}Z1dhX+0SQxX{3S|dW9W)Lo zD~CSs%cZ_~g;W~-75joZpdI@Z#)^LUtmB6Rd%V$We+E)oSr{<*6!sfWrsCB`6!CKc zRYeb@pT2`={99dGId?oA4Pw;)6F=|w&*5o&09@})!68F6feR>ve(nmwiVxPp?%6X1 zo1;sFtkZS;ZmA{P`xP{$P#(uG>mc^$0yHbTGo{Oc%4AFqYH7%}@;+kT$dvKdyhv}gYaKGXau03f*id{A`})hi4;Aw?mhhEVPt6IQ8V!ki|Uu+_U}QtnsI0TxzJZk8=Z zZ=J+<2@?<~8YXzY94N?!3=+IJ?|Ue7rU86bF>}&JA=7i8P+E0ZSUHLFp>oG#XxjvY zy?+m-#qwn7u1>{nPiXM|o3u^k3VpCyiUod)a4BsrMsgqEqk(tvcgh2tKA}MQ+rIPO z{xb%hAAmj@hM3OJ(UXhjh+}*oy^V9;OcaD)KK+C(}mG=?EEImt~fTL?VHYn9 z#jYG9e4KlN=Sw6B_be|9&u(Tit(SSM^k*iEyz7YL3UN4*SAZ_%6#TL|!&wO-c&a~$ z+}|qF;OX-8kI!%h9(xak9b&Y9%yzV8JELul1^3R6r{+G3=*v4p+Q)NI%S}$hDeM+5 zpBqBjY6kRox(2QLvyj$vR@40INmP5ziOR<8B|V!UO57C)p0B9CBW{jf~WN$SzU~BfR?J{m}|mGT<)>E5wBfGbDxZmMKEbf%$^P zj8(#jqD<29xlP|5R?+#{63pbdD*N27%~G``p=tk?g)}^5g?%l_OU|AK_PUYP(EfBp zo=EAi4h@bEAQPuV@>aY+NA8%TaJ&O#r#T|B*plArd(jxKwZFc+mM*our*j=Glqu1Q z4zFgMk!wP*b0X)+<#In^35t$+vWqQ=Y{Up();Cs@#B|4!=({WVje3A288J$Z`Hm%_ zW^f(4jBB9PaPhym%6z{3^jQ-cjqYH|>W^62@Em2gX5q`g{aA8rDuQ}4=*Ev6(u%uI zSU8KW6wjc*^4WAnyNO-)`NM)vmazPU&FF4$M%XwTNL$~*?3@nVS>Ayk+x_u!}2*f z`524i=UEWMDEAQet|Vtr_MYo>Mdvm}+IKKb>z&y4a}&0Qq+mW>!;|J@u2&kv>#AXt z_gb2!&S+vP*DKlL@TY9TW)*zX9g6r%V(>6}jCUVD;>Xou*l%>h_XJz~9BYK-^OUh8 zRT|l8YLGPvp}d>nw5ZUF`iR|Rk;}Qhd&p_laCfZmx_z|JRH-iP3-A}=z;RcgBFYqfZ#@vPhEmb^`f?F%XXq%h#+Qy+ zuz#Y8ClB>7-P{4<{8a15Ux|6&U&TbTim#PK*Z z5Z3v+h`OhPsnOLU$-n|pQ@{+-QsZRgv?k+R&q_S=bif*MKUhfyLiyh_yvu4r=Gya^ zFm?#~FPn?fJDxafUdM7y*D>v{&slTjEb_`SrW7+p%J0X$K$)lDR2hx39mm*;=Lgx4 z5=Z8dbzfxeoiDm$C&kt`n_$Mc`KaE%01Dzs&MNFSC-CMwCMztWGXtKZdun`+kkA{B9+4^}dIg@eh&x{Unyl zM?+IL3OlVMxPN5>Y@W}+uJUipb@Fl6mg>m*7H?)%a}3Gz%Mz+KcA#&Dx2b*bHM$>t zpYjhol36h48joB@5&XFtFy<$-{#C@n%Ez7Oit( z0m?z_!0G_5t(8QwgA5*LPJvg+UUvQT47O%rlSoYAmT2mSB+)B2PgJp4lj=-3AInUV z7BAd~_8Ifh#kG*Dtt{Ep6ARe9X{PLk;#6ctGT6$AS&kh~9cGDG0yb`7le9P&=t z;FZEtHpHlkl}J>w`!^Qjc5pOmtmDzh^(E%pG?CIZ4B^#J*{rcu?AP@iRysf*lK)K6 zdT16bj&rWBc@8QSiZQZyA6)fzK`UFbimm@pQ#ER;D$CbRqY zui3vvkC<)91Gdt#7)jgjBJJxfNJ$5Y?yPGO=>-jBuZmSMYl|j4FARaB$4<7``O_`kEzT_VQHhws52qzxUF@VXJAyI|u5HvLQnzM^V1ZSnQ{3 zEOn6rqQ}5O6&htc4I{X(smyI}KdhcHg7No4Y>00fQqlV=7dHdn!FbIJ z{CL!XXqoKs>+MmvC}&CO=(@>*H5Uw^0$uECPqN_@U4K?)m|;$+fBr0y|;k^2~S_Jeg1pS3PEbH9zU(JA;kUB-+f8<1(g5d+UoN7=W-sQSZ)=9;nCHF*@iN!jBNzwe*@ z*+c;Y+-RADGi_HZVY|mwvBS9yEcodnde4^A?(XH}7*fXcWk0femp(D277fx(8b=d3 zPxW?IFVm9{Vf@%5xEo`Nf*>cjEZmOT+4rbO^EQ15P9xQ^k*u~_1^>ci@Ka(Y76>b$ z6YY#;j)jcfu3-H%hoMzImmP|{%UX8iFx&lcLacF`AnBMRF7&&iNb?INxKFkggT}O=WyvM{Ix+^`e?8$nb1@FwlOd&JsL+#j@-oP{}596`yS=^kukdig#QOo1?q!pnnd@P?VM2$BW zpb*6Ui?Q_Q`5DUReWS!&f6UCug?Q@=_O0a|6Vh{8hlC;ZOtztCWAy2Zy%;Lyb+htm z>ew0ChPH9-keJzx=Z@2Gv0)3GZh2$p3GVNVT1YiJT}alx6Z5|fqA3Tb(zCHPWPIO& zD(h#{&$u|KHx$ES@EzElmcsXxxjfHg1w7p?u}n7!rd_qTGlT2-+%vFiV-BVZ`gkl% zhh3v4mgt3|;ztbL-na{QzK=D0tU|jh-=aL`AS@fX&V8dXq^6pY+5sb~JElXLtAZh8 z90ye|KkV^sXAZmM@J3Gte>yIqF6=%^FGOL>f)oV2FG70rZyvXG90vje`AqsbXFD6v z8WQ+yU@%pm-pqCDez5m=fLULcLqgscjv?oYQ zv3b-xN)3||GS@2zS)RM;*r9c#p}m4SkNd_t4e}h!cDf`vLBWyje%0)K(gL6 ziMr~IDB0l#x!VW3g) z;Qa7ir-Sj4)1s&wCSU$;7_V3^KP!l>R&3dKNTZ( zXc=m+iKF(SG@fuDw&=_Nsu(wt6hm&JUA=+newRZt_tLdzXyY2kLH?~@hQwAc41M5* zwbPEE=y5Sh2E4}pt79nD;v;n0I}of?gQ&z-mho5?4>pNoNr5YFw0YyXtsm41%qeh+ zCE2OWq*DDU^kC5payrULral{9->$;;Z6@xO55yQNpg49jTrz@K#GWwL(eDa72na=P z=7OZ9rLf&{EW19;fu-CCW_5!M;h{VpUn^`7UX>5I^ZDqTmWO%IB+0&Sa~)KC5b!nThbf%Z~5V#@+>NyyNoJ)1Ujm*9NI(Xaew(F&UE#_iMjw-yz;~N z%)Rt+haKq^>(a1IXK-hAD)h<bHqzdIyo*%6MA#LmT@&C*b!e4{SHx2GgFUSo6vaDp%W$2W}iJ zvM%oyh0DgWQ#yCp@^gi(HtwjfLi2=RkdrRxJ)Fhs=>@Pnw;lVAdr($U7$uq>B?*^L zY}EBZXn({#D;KJn*6nUKK0^tZmC1_p^I1|wmwbed_TQfJ-T2p zgU0aZkoE`#oEmP3c3msDK9r{;!xd<2jvlFUhSS}EAZnSFND@xjY`^3m*5&(=IZvE~ zA|p%GU2w&{P9LULc$w*cIm1pJzlU*gg_zNzNOMBJvx;O*7}+uSDK5cAL#`1$Y>vuP z?nuaq#>$uHIX}!4>iz67?3^Q}Ega5b6O@?Rzy^`H)?C`JKA*0inoV1M<`=%Aens(l1Z}lTkyXiJw*(Y=E-!gYU?KX~u` z3?FpF$Vn|6JxP}^VoD_v+)j!r$K{GPyp-1i*Bxg(#o|IV$7XNI7&(A#AX%A1$_@hhv#fc=pN#yp}8G z&y_~3>DK|5GmIooG-BevHu!A)4qMg3Sn_-W(#FkzY4K4sm&D-dy;Ha|CyWAq9OoHZ zVI&v06U8kr5j0Dgp8B<-=};efenp(z8%_hN$uL_OgeNyQ;D9aXWX(B@{X;prD0Qh& zzkHq0F1%rJBcC(N%InPAhoAY>ckpTd4BH>G@ylW|bo`e=;#L`)|J2}9HotRKP9XVW z0@5CbLd$*&BOOoN4P%F`sOJ5-lj!84+hZ!mGOw>JGo-Y#qOwS zSnDXD{+$VyxulYsV;)Vv-9Qt6W#O~?S!|7q#g@qHh~?hZJ$p`JMhMsk-gB%i3ulW; zHzW3iGtR~wLf83F;Z4+eVei6wLZWgJj%PhWcTPU^nvawA^Eh(%2qW9%6^M784byW6 zp=!S#&*XMuv(a9Z4vfaOkU)64tVE2a3f1-WqsaA|yn4$w0(f!yvRuCm^)Y4?jA=+L(^%UP9C*imEwCY z320akLYZ+j%Z}%oe~Tk5{m%%RKbFy#SUVd3T@M4&4R9i79CJqSV3ykZvM^^TSKHL>mmhsplC7Zvb5HVAd4%^B++XzS3L1}y z(IG_*8s(RUA>t*_PkeyYH)cbNdk!w&+JjYn0&uA|9wu^+IQMi0nYS9!yh}XGAhZr< zeLtc6eGe?g^@^sR>(2)7vSAH8Z^qNXg0oe)21Roat{Ny}^@G9a-+qnM3!jmrLId4# zoJEr|Z78YOg}TP-2%ob@3w=|qgz}^P;MpVwrvu&0w}Nx?pC9Eq^<&tfdYG(qLTP1N zDDTna;X8XFgjPF9`piL@>^`jg8H_p2RhYRa4<{)f9X$5%f8t|7O^Wz`e$V6ARwMrN z^?yDuAJ|X*KVScE{QM--TH$}=>;B*Tn*V?MoBwyt$baW&{$Ic61?}ShSI><9$=BR& z|A_^EsAdtJt*p@784h1vQ7W?zX;<%%UDh>9RL-E$MQ>nM%XjaWYEf}U8t2YR;8lba zHjhsrt1+iZLqCdE=FcU+9$S*jo$V&s1KE9M`RN6=RZ%)ijWuJ-u@gREbO{?`QMgi7dy ze1@aKV~o&Jp-W%130nJ`q2|^(bUiFU*b7_K44sd5dj{*3?-11f9Jljo z;OZ8N-V>2{!gKj<{W^-UXWlU1<%yL3<4I$q8c8H@eb_Z`m?j*?T=76GNlAp^+$;o^ zoyF7hT6C*v02yhNV8}-mT5?2&BEp`c@3&bvYVQDRr{(bTl;f=YVRYbMe_H?KIBw27 z2lKg{F?Z4m>!jx6^Edp-Eq)LUp?Qh;-b+*`tZIomvgsXQ~7=TC2= zv8lVDd)pal2GelE-v#%#dcrYs7xpgZoM5i6U-i}%-hY3?koOz=LwgX}ZvcII@(U$# zcd_(HCu~zXp-}%1i~LvOGjG04(*$@({3gdkU9@LX1Lc3x6z;T13peJd3s+LZX^Z*^ zntt*)1q5jF{?US6ZCS(Q)V$a~qus3k`@`&(t^#fDqd}&D1F62^3x2lrq3vIP!=mp3 zO1D}{%VaD`Peu&Nd*z`QKL#IaMKt^5M6LxlBAVn)XIl?Z!}B2ecD)SE4{NY)+FLBv zo`)30HSir{j-0|&reYGv?gVDA--*qrKHZ7!)>70lFBm=**|_VQjT>|Hxz1e#izfnP zW5Y06+y_M)qw%vL4Yk8E;IPCW|8Dn%y`vhAPUMVDwx8q&d6N2v`Q%s-0juD6l=^AH zVBQlZy6}Vv*D~3NHwoCi@;ub*0=b_q57SR4VAqXe?0h{IRsk(6C-J)IwPYtg%6`Xo z%^D28G@h1D)uYs5!)em>F3hj$L`Bw11Pq#sivwlLjc{K81n7N-Fqa zO%cCsX#1OJ7}Z4~tk)VpPmN}d8Ut90M1SVrZz_!!bZO0JWAd7l!ctUI*#*^X7Uv`( zWEP4E&rfzx_noh>TvyE*Hpg&d#U$ZdorSPGc$DDCGm&bV#UQ^z84KTq!+ur_l-EX~ z$nO+s%16@~FMsOy)SmMfwxhmsDMsE);r`B4gqxj3|G5pgEc*)5T4`wVn1iq+5$YC9 zMC6S-h-!>SrF1&X!-okrwtO|cUqy)gr$Z%|4QQZ?I&D*(ff^Uy-;9YsUNI9aCyf^T zHVhI}6E2ffPcE%*%_hkJMU#6AB~11o{w(CTiE@`dMpi>w&lTq|IfY6e~)3a#D(&wUG&nD z=TaU&N83;O(`wBSin`s0jGClqqoe|*KIM#zU%e<>{}@uM-5~oX3J=HU;>v|0VaKZy z!Q|jo!RJp93vs@}j=99J<;uC}Kl?qV9sGx~VKtc3{TKVHpF`PAjUJ5Rx)L>IQkYtg zytEM%B(6y(?nYwgvn=GiNkZq@Ai9?s#rM4@X&jYd_?wULJ|2M^&!pnF|h zb1;DC-&tMf+)V>)>Td@PIH9vn5R_gR2+b==*rl~i=p5%F=r+0wduF_(E1iv$tY}B+!2luXF(;AE;+o`=<% z&RDZi6}^#ByZ-$m8>~Ey{vKUSgKs*L^qy`QnGB?wuZr~5SQ>Mtn4)qALrYTz z9=V*usc#kdcvziWbH>ubZl2BjdlMq99iX~<2e$orK{_7K=%?vPGK^b6i%M6Ky`nCy zSI~jZ$}vb4H-z!m4$e8Q!s$CNaH6&ZC!bb9b#EtpidUfL>_|jJzGZ#Zn{$oK66&d& zPVQ@6Y4{HhQaTq!`&NCx4NF&``T?}@TQIP75}n#5xp$zmW41y zdV#R6d71E|LyrFD%hE=^pPl2DE~F}*5yqtj2{~%RjWB7}Dv^GibOd>UMlh&n%JI-VW2Wk@6C&ASm(e=*oxWMz~ zhS_5LgvIEtc7+(OMllgI=CV zDkhYkGbT^&$r~d+o1}-H7hJ?H2`eYu5T>>bWWQx)Snh*nk-6G%`g3(0MRUL3QM(?d z_f`elddH&*XHak^AJu!45o&*x#bgaZ?^4d=t}X(5eF;B`UgG`jUU<1yA<4NN+V`7L zF7^%K1C8lk$4gil%hA2h;`Ar;Hl8;fhc{b@Q7NIwIT8hb`z%bq`iPdjt|Lvo$E41g z44>rWDad6IHD2n)`vZL_x<}+GlJqu40RDdA5h}n=_ye ze_O^z4;{y5iTs(<1_yXfID~Y=^H{kj5&4^vQGf9~tdwO)Z-^opx+>8sxp2I5y@_!L z3sLdH7k)pZp%tDFv+p{T|9m!yy;@F}R(qqGd(DThJ%zT9H=$_r5WbBS$U4OH#PbX( zk!Me9>JR04O^jz`4WuIQRH8MM@Wt*rQ^a@gW$}tqC|x;h3G| zg+?=HxJUXx^p(d7-P;AVjHCDwD?_0RRB5sOaQYf;M7VE4t}YOlS!t&E69OkcH!-`s!>GOF*`m)lLB={XZIQ9;DOKM>Moo7f*TSd#V4$`;)F*JQw zLf{xBSWX{8Nw4+DrECl(yXMp8DGzDy?9Y@^IuGVEeK`xL2$?6nkjeLOUb%iqcU;Ui z{5#1?jpCW;qd4^(l%_?u$I!T!zNj!w$JTZEc<&*PxGoK-RmtMlxm$>z!oUB#3XRQ? zSSfc0Z>zZPZNLPUz9@ki_2;YvofA~P_B`3|3Z*Htok@D{Cc36*LO!}T*p0?GmL(U= zW_fU5uS-16b)Q6|!EOw1j6vbXYj~u23_bNBSXOy3gvf$`8QmB2S zLS{vSY4o*@n$bk`lh8xx`PJrf11k5GJTJRMq> zNuO^yU`$aULa!zuKyETzVz)#2_K5&aPT(v-He954c}Xvo( zZ0x?`UByonANYn59>-vRW;T4z*x~&_DKh)4OVPqe+Inm)Nqlyo@gC+Bb?PDhy?Bl~ zp$1a6BgsFU=dEtgq{*}F5n1ntb-tW`YNm!gV}7wWDV1#Q9b@6(0&^ivdz@glB8Q#% zIsiKd&Vk8Fp7Ek^7qZT`F!|$H{9I*+uAF69HgX2n%+4dvDf7uh&lAfx$05w&Bvwl0 z!t6>oo?KpsKbuWyh^`sU3In}N8IFck+L-!N7a6(-U=Xk$6AyD1`%P(Fww8qX{Jz*B z_7b}tZ$RpAA3wbIz4O*%Z=W-|R=Hi2$LzGshyK4mAbbh0QhD_XL562-2XL^g8kVX}TM zl0PrS_Ci}ap}w3XG+Zg~lnagFPu>%K4^dj+Lhh%TfmQuA@jW&Oe^kzbdBh-T=tC^3 zY{KP-FVOaG2Q0+6SMbnfd>avr$&%-wSsjMZYnsrq9*@yi`e4@Rn>gHl0cUORW714j zNM0I<_{Bqb9>g$ayhVxCc2CRKS%e(doZ1T4MSWa*%}7c^ z8yqP{TE@Q-=UxwGt=CZg%ssMEQZ(nUIGH|e!_G!Y(prB8hW;0jb~pnQy34sgG9RzF zU(kA#C*0oq;X!*Oj2<1rnty>*T4PA=Uiw?Tp3<%#q4L3_>2RSA ztz2P1&#o)5J9!37`N%l7Bb0NQtX5NugEw8!3`1V(MOYLD0tSt=?R+U6j;c;uVnHHw)#(+Em+a$mdC%EABrS zn-a$%VCr0KN&di$4roHZqJu5qIuf^ZzJp(V6Km9L@&46fWRH)6+dzL5s)hnt0f=IX z!j7#9g1DoMV0Vepy0(GzNY#W6KA4O?9k!6TI|tYL*$ZpZNYJ$#DGcqsMUTFv(^uI9 zQd-g~Y&~5sbcBBvoIdVh0Xg=})yIX^^DM`Efx4vfRi9FF0^w2PjX#$9a2$V{K1Fy_ zjLcRlaovoW`=_JsiYg+CC3%kGN6apIg^Z7;G-9106&p;T=$Z5BhQw3AyMT0 z*M}zj;T&$B-}B>GF5;SfVPVbt&RO!*&F5eB6aSE-yo9i1?iY&N>`MP6ys5f;J*D6E zrS;bxs7Q@_H|+1B@pmQyIcr;C1VgEU9j3~S!v~Kd0WU8K>vK|ty^{vu*D`q+*S}_W zKl$J=XZL^UDMGlY5UO!!aYJqwOjY&hTl640;Gsbc(~8l7Pq?Cxiy_j{Fza&wc{>;2 z3D1;WaQPtRwrr-7AJ_3UDGlD$lX26E*L~CE$vRP<`mIQyX6$);Nc?S!`weOp4NbZ;D=D&^Oy|f9#Xj84QkQuLs?y)(VEZ&lcAx|d>?|J;qC~T z8HI-5F<8l2-AxB{>0dR`*y|mT-87AM#ZRNKKNDzZ@+3;u9Y`lP4yNx4m8j>7O6hz@ z?0-C8m^OK?kkDu-FrQ6=T+#yJ{49H+WzS36qVk6Bl)a>Ysb`o%s2$s`Jc%{znTp)Y z12A({FEif1n6qSOQrKpq%4T<(vSlqweA9&Yos|%6xQ$JFbSScII4vm{ zLOj)j&EAs1T=rzL>(2`5d+|jYxh#c35|-1DM=mt7ZUb$-JdMwJRG}{|1;-iJvBmK; zwmtKMZN)F_Pf?_!+;etF?G{?>FJee?HDYh6(W%99SCq~|=J zq|HtXP1Zp|TS2&BJH>}A{n*ool8QAeF4`)6Y(s$A%neiWBm%(d_LB}!BHVWTb_Hc%S8GaV8g|qGo`uyFF zM#R}tF=sE99C#1$IZYUQRX|Fz9-em%!Ndb&gy{;SgaTVFAvWp<+nOK_p{$qv=o2D* zK!~u~^;#7D=ZZ+?WR*y9t0U#t&Y@-tdy>wNXZ{9pY}besW?^>~%6+cmVV^8W z%W5;H*RYfuI)U$c@d=)ETzj6GqKq+1BrZoYW|&Tqt{G=ATNbxBLj+* z;hs5VbvnS+_Zsyp@Wo~>-gUb}#W4bt^g~eRc#HL&bDj+w?#X>IS8=>F8y;Kou{Cru zYRXNKVXKRvWGh-=IEfOPK&mBW$kwQYL7@zNu(jgz?^PIhzzU(7mq>!^83u}1)4K-F zoO$*RHUamMq$7^7>MyLRx0RjhP88i=E5;7F8L)4*C(+KClO4xaA^%jVu&4aKkm^$- zykGJRhrJ$RB71^mksIl#>tZq*VM%8nkH#NEUN1N+W02n(B!+KCn6*6~#_h#Y_0{l7 zSOAH~6-fP53%#yO2;935gCmw<0QWQ~BeLp@`kH#z)z*<%eD^883 z*%Fr2s%1fGoXPj}?_+kCGyE@()~Ct}bDHzkfbxuG5&D1Fd$XvX-|+pnXdqFVNQx#E zq4asK`)x{PC@B$ABo&b%Ltj%eHK$RAMk*yVQ%NBS6=f_bk-3y9k_?gk?1O#wU;mT+ z+k2nv1B=Cy&q{sX&vW0`^?La;BwV+{y-k*6^=2$}JmTH!x-(Q(e1$HDUnUPefBIED z6Haxy_&A{i=Q@gUH1-fS6k4G67$GoM2c`dILeljzwwhcw1RnNPzJ`< z;*hF%#`uK*(f(z4vf(o3MLc8d-!pb#(ks?Iu3c2QRg&FoIxc!*ZbM1&bEs&j5rs&q zldwpOn%-zrq0>1QU3`+gHFRU)@A6RB5sT9c&*IggADBE^mVSKd!=kGX_zWut!_IYL z=Zp}%x*LKmyyK{ouSi3>#c9w0SyIhTgl+c`=rkM-z5jJ&hj#S-F}&DgD%njAyPE)(l59Ok~Hb79D2(+PhI>8 z3)USg%=8!|OiED~#JkMtb(j-nMLW~q!xw~!&4of^>;oau;k;;|QoYE(J6}}(M~&1p zN7C;h<}_VJgR>hMW?r6xB%Zyx$Y<3}Zm(e!F$~FRJd4TwIkVncQdIqT%JosCj$A!5 z@fb{N{#8nJ+RD)1`4Ti> z>tHH0HYCpjx|C;s1F8~T$g`HD#X(~!_^uK(0*50;*#zpbPW1ZH5<2PWK+)6QLCIg5 z8uJHIkF76e|Ct4y?{jd1FJ$*_7O{pnG0vn^fSu2axa)G(j=Kfg%% zE1WTS;(m;r7=i59;h1cA7$3LCVR!NgXo`a1pLqc1L+`*vz7{@do!ERvjDp-GNFui# z{_E;+YwaT#P8mw?-dSQ&^nC0$nFBlNO7dRG@2~^zlgqRPf@9kZp)}P&82EEP4Y{4 ziZf8N@%VfhrjF=kjqjx~GUhhhllKa{?)G8zY%zMRWDN1B2>dnhK*$F>$R1yTy#|W_ zJ#lJx|AA9&Ln-@LHZ+&y;ljw%LLf5ac?nD?jt4$?HicsC-AkHVYh5&L$lQBI~B zeLbQ~23wa<@)1K4-!PF3UOAGHqXDV?8A*Pe7bT%~2CAI}m{w;%Vdn?Yu<87nYgJ0$ z*C%q{-65)yl%bNie$*4KPJ_IvL@#-+wKL<3Xo0;y#r|}mbLqY`L+mEozTy=-^f;0o z&a)S)rke^`i+RPMst<)SG8FS#p9W_;3zwtZg*N@2g8Bt3%sXL@FMnr4(&-b1NL|9J z@JrC%aEo4zJ5T1G`SkR}8BF3`+~3Wdv3BkPWOzRCuw@aXoJW$e!4PuaFo-5~?n0>_ z!~7LexEGg-?5IQ}s9)!d(L2;U^cC%!nNF#mOX$^UcN*mX2}kyZ;^nwVw4FKxnag6r z_mpQeQ8xyyj^GWdDA z!i?@;A4lgnC+MAy52frkq_y%hNz3a!_Qs#W)Nd6S&O1D(qK;yz_6wBCv;oYlqcWb*bJ9h-7K44@Kqet$VCe&Ryh6-d9NII5huG6wuvQIaf0)F6{V+jm<6F33# zxUfsZPgtIl%3{<@*wT;-O!di2$V+RGYVAl8f0Trw_EtD+Js7F$1lsSVL;-#Mxhvrt zta-G8{koIr@gi#d zbO4VX#=&s01@qHMChO`@lE|0j-dFy9w);NT9Qn-*Gy=)VVH@dj2F)XfduVU`f<{?I zI(y&>nB7BuR=tk9GbDveo+%Wn$r+}Ws`T{DaC&9chz1%$j|(dBz2Yx>TltI)SprSeO=T&~8aBa*enFi_!;ho!JEG)blWynM~J~h0y7p<#_YhmXh*E(>m_w`#n0G z%*=IZ=`s&&ezS`+cvBISpN2{K*T^sbEmbZ5PJ7?&6g))Rh2c9rgx97+gxCam;q9f{ zl-qlgasVxp!pI^!g3?=eQv~0mj@H8LOQY;XsfIxnR`gjcg z5>-&RK?^yh-E5@AMGP4kjocjrI7{4$4u5kbeZI5r)4u|7D@h#5N)jD>=|HE&mQjzF zJmnQGC3*f#lr!E&CBFtxUakdchMUrY@#UD>8H4%nzB3)A$+Ssh1zFB8A^FcIN%mSI z`Fh--|B}X1z^&Iu)dAx~>#@)spG{&()OXa1G9STZPo@X+qZYkF-_J6gCMvFwIneWVz z424qLcG%SZMf%uLv@Ie>$UjmdL|=U=yw;UwN`)UqXC|02pE579wGO3(PkD4!hmf>z zAXeT`Lg3Q7xT^mRvmIpViti03H>r+&OsHh%=JsMfze8VKpi3r`>JT^g8tx2ygq%-5 zP^%{a zw*h~Twj%C!CR_=md+zzjdYOX_#)Y`Ehsf-JEp76(B+1L=kkfnuDb9j(wYTFrie>n` z(HxcDwIuWYCDk5N7k;Z$BJ1-v#17^&y;Ct@e}Z9uY#u6ZS+lB1eoTH^I13S3z-RbY zRK!H0Wy2l8|7?pe=yjhUHb{&{XbqsX7iDOhaDu+?sxAkwfR@60iri@|8^Ayew)Ue81uCGNsuR4`Wq zKelLNzs5vN|96q+R4&r3;f+)`%1#(_z*U&IB}llvi?cvo^C)A~Og)MZQ!L0z&xf_aB2DK4JE#3Rh2l z0?DFZSZK)akkZBQt+mt&o=3B_g_ObvspS!aNadcSoWW;;QT~~zFiqh z2iI&TRy|)RuXPu49|j3KFZ8l7{YG{-gWuUtMdGz)0`L5v$LMT6r&Y?K-Gyh#*kK0B zc_y$(!w?^<9dN|?FuX6GhFs!nw7=!ODM-=(R2KG+uK53-|NbB93j9y~pZvf1aXu;i z&-5hzcdr?NA4mT0{=B-&`2VvG#l#8IrT;J#=G5??VyK5h4V)y}dSQyu{SpTgUFrH`64;SE)#7^3s=uZjQMWGJj zq{DlBVWqE-Qj^DGXUO5iivjriSVs7F>K{%1^p3hOPNAT76$(olL#Yoh2&)}S1=Wyy z!u+}uqWt)8 zv;76DiH*kXizz5PaRPmBe&7P{leg*iC%Z4-5wl*F*6*24=Ijqb&JG|$>wdI1cPUoJ z%)^GCUP#{a6K`-_A-&}= z{F0Bt|E~*L3!153v4%P=A5hANBt*VRMMLIg*kqbwMeb}Q@Xip)xY~NUH1=utm(mry)R(Tc>ro|d2IZ*c4nddosC`ENDXHFg^AxJ zgi1Rt;f98m;51KFFxCwuzw$^L*cVIN{j}NZR3FxBwt=1PZD2!8Z?eARc-HLp66^Fn z!FvIp8F6oU$2Pvh2+c!Y*cLi%vYLL6bf(>NrLat67@jVbL&)GT8a5`4BDBkB!tE5Q z-k(6K@@dq`X~fI(!XWFq7QW}w@o?QmlwElPjWO?7g^W7vGi|Zvzo%F<>N$>m??qnK zW?Z<(JAR)+Asc*!c`y9M#@ckSwA1Tg7~;j}HZl0jd)}tMKcnT^E96GsU<%^L+126% zw(oWnS*7o#U9ORo_-_=Qo2E|5e?j(E8SG>CdFHhF4*R9zjfT>hQTtdbT;K&9bMl&u`&S|L0I1_!P?Q1G_i3 zo{iacmOU{^#)_JF-qp*-lPnAR5vEO_yXR5E7fn(bFp#vDjiynnj*1LwFN)~m7&hpD zA=Uolv!P-K>ZnnKf0!oRO@~0TKu$1hlNaL0DGL7{%%?KW?taBPk=u{j2obt71*bY& zp?;=3lfUE2&ZHk@mkUdfVsIXk8A(|4Z9OHQ*+N?!J?Y<#B0Na$hEA0fRX$Y3O)*1U zsv8H_W8AB9*MgKZ%;|3APJD_efym}4b|{X*P5;rbNr)j_gi4P5vg z_0D2_eE4h&&D<5J_85vsGbRhm&W#fc_j0a@jtoYu(#EEB12OpCd1yS#!^a~h;qgfa z(=@g5V4({(X>kvRv=wPA)}VcLdk~`#$UQ4O=MWQyib^*etk&iCw%@oH)C#kS#c-&6 z&PrN3*?2+|AwvU#I^F~7nK zEYHY!-*9}SQ~VNUhRn>iGCThs3aPT_a4T$ z)kg3P_{s_&d|>ALZdB2%*Q z$08B6{PU~B-yfgm%G2~c-0O6Y?+cXoqw|0jq&s%v&*K0n2a03leHEDSok~Je4=doG zZ`)siSTJlp-u*WT*K71(6x0vRIm$d>Lx?Z`?d-5$%nYR5BeGUqv5d;0#< zkz!+c^nUm$R%v#P)!b61gl1{#`P-jrdrY8tl0jvl38cEZd7u0nO!;hX=C4GmYCp<% ztsyk4W)r6Q1;My$A)@&1;F8R4Xq;S)TmRIt`G+jFcnwFB?jEd3cg3kyoWVVLBXa9E z<4U&?T81g%@xuOCy6Haq{gr1D8`I$W{uO0`ucWAW$N<`XJ&N6x4r5by9%dbLTCk$=DVpq`W5549tTOnDQ=!=? z|D1r+pJMnk`ZdnHNJ3jIclEv7gT2?9aH6&xn%$L1o;Mb4yOyKZAQ0+ngYZ2(0(-{j zpxbvZ+gG7y5)3F)Ox{=ytG!* z>Lqgt`^O?Id;qdd2I7v?8yd0nBPne7PCs;|DE@mJo;+(sKYdAI>V{tOJNSeAtHz_X zWHQn>n<7Xs#|bkR{ACMZA2yC2*XYx>Tly3wHko~27|8aUA7^Vsi_vbp4pIIy@K@;o zZjZIX^#A5!CGRFkOmKv=j5zv*9$+F zZ*0%=My5La60^I*{U~PDl*4($ZAUdJ#MFo;mT*Vb%XEP~Ix3u79wF34RfxjwWr^aR zUlcuhXHGYHhV~mc)8@w_DEG~Qkx>^LcqVaP6@zj<_n+2q5T1`!$URt&)&)t@pmTQg*ljI+Y8H|IQ5}T&O~(lN z`|R3;+j!jX5tfa9f)|Ilw|d1c=vfFTna+ELg@eiSKIcSg&gNX=X{gSchY_RPX(Mx^ zzhmZ-YH$VAH#OpFc7N(Bx5sPwJ#h99K;o0zlqyw9%8D8MndJpf)g8DRz6{@|+{b9? zLI{qTP*d?|`Y-F)+1P!|Zj%9~mRKNj${a{WUK090zx024X@r%_W0A=_a&9xi(Ec=m z^5z_(?wfn*M@%%9{Px3XMMs!ix`>cH4>8L%9e4eBXYN5bZaW3R-n9jFnzwPD^LDg& z-eQTGG<}GEf|)aFc)mFqYch_*$7wsykGNt;nkk(2a0XMx8M5b$2CJ+|!s0_Fg6Z0+ z!rg%(rHR=Ev35wOKH!nbJRR0gF9+Y(4K^T zlxQ%396h+tP@@TNR2mSu@;naJ{$FGmtFSSL`-#$9Ilh1;~Do z0i_l5a7%AE{bE`)Sx$zUl?>=ev>eq)aUW0NUpQB4lF>v^cW^y)gI>Zd?p}y1@`e~y|Rc%UyoPqRi!ba}sp2*}b&8>UET+>GjEArHZDK`2-9QQnw>W!e~D>*+@(MZtyXCNGK(HACsS&h$9oC(qD z0o~=ncoJBI6Ff^RD!N3rmQ{TJb(McV1Y&5?Y0m3CiMelUp<4F_Q?0AXN_#PgiIY&_WtbBO}IXie@R`knZI)i>5K z%OS%sWs*F1J!uOA2hR{3xkItHd?;A5Il_fI*M*BpHKMLZcSWOD^BHOkP}qKH-v z^3uJD#a#^;Z_}Tgx}Tt~fsNPwF?)P_bM`IfXWv61L z_GN_Dr_s4DygRlonW8$+awmNezTGdtn^{scvQ>_Z*6}@|)f$BCcE<9ya6IAe)X3&h z@JSwvl!6U{Z1HBHY1w{ZM8F63<7pc!wEW8Vgxp7Pc{8G-Z{S5{d+rq zFOR`Vxk{$c-o;+b>}IDm5)rX62H&G%@a1-uF!kFjp?Z9;(75(CjMh{lc<4nWTLqH1 zR~XIAj3i0>Y~*d?U4g6yd_I;BtFLhwE)k0?JxF+?E(ahvm@JKYfSBe_4@?Ki6$ypaY*#Bs;M13D!k;PyMtk+nL2 zSu#5G)r~t_o>@})B1`;K@Wd$Y7CmV{jrtxqQBRyJ$$$5v$wvZc_0w=VxcMCvpNP@Q zyPN^v%YEDfYf!xW1rF_!M#=^?c%LLhSY;#S(k(oclc3xKk#OR<{q3PUA<=LMYD!+5 z>$VAR)_;Z%@8pO6&?1lc{}5q)5S?;C$mBki@5A1)L#hMyt*aqe8(c%dEJp3@9n2}m#Zw&yPH|; z`^J7gkQYJ%l!e+)214C}b@W%ypLX!C$)pi;g&`9+3eOFW2`h;6LOHjqAODfDG60|d zUB)QcX85kXM(*4Vdex+!45rB9Wh%h-*m7@1BgzY^u;}Au93EAN*!`TLyr%&%73%D( z##r{_-$Z7=Y7Ue48pXa$vt%1i?8N*>o`{~h0HLNL zcZx5d$v6*=h9|Md>lT|Wb&0(hb(d8PTu1!_T&PTDH?2PQ%jDRM0+TV*lT7lHU$d)b z{m{Ec8&`+^7MU775tV@m) zwk&6|U6w3zkue=p)TPE875c6hCt9z*Q#5D9W>NBwcD81C9&@EQCa<8yzIduJ$Dd1> zuGv#|BjYJ^d3}==a(>ahxZR9KtYdcB{h3?33VYs8%;so08}EFXU0imCy;&kK>y6{s zs{?OD^EP)d?V?9)^1fHBDR~6bsWD`;h8r`B&TiKDZwNFe8DoKx9ZS4!#J&k8Y;o>S zwxHi<7P9dh+x7h{3p9#h@im9ptxAD)x~sFWEwU`YEC6G#?Z@*SXONtFhNTA_V&k{3 zW^%O-oXaF?Ofey9u+l`I@Cx?QzP#J;tQrYc?)#r= zb+PQ6*Ks!etuEdf@o6Kdd#Yfn|ACvDNeUvGPTc?B@6r>~`4-(bLU`MTO=^MH8A&vCNJ_ z_V)b~=5Ti(6LUYo#(PCFk;4v=)3eL@NX1#@3P@)Vn&XUU$)9bhN7U1iQ|?=v}{N_KD8Ll*F3INWdaM_Yp!Fut1U z9z4OEdJixcJ3SU^timkQoY?dcpIG3r9F`Gwiq*fegSv||%odtq#rGp@PjfPpQh3gG zt*$j0DJ~|G+|*?f5kqKsr;c511F(6}GSMLC7?JhVzaUQk?fDC_m)zr;C&m_Eej*wZaaWYNB8@Gr%VMj$(wK!>5Q1)SF3gooc zr1Ov8akNu}zxz^!*}->&pXVb*w#MH?Cu_1qjo>arRhezw8v$Pv$V+WI?YCG~YBx&PmDKavYq#aU&XzYD8VkxVr z{y%X2HMe4lbmehRus)p82jqV?zy{6tAEFhcrr~ffSJW8?(-I zpm{|jYt~dvQADnS8xDHOA zYoJkIh}Y)alYLc{qJt*UN_}&jah!tKyNZ~#UW(dl<>{!`KX^I}CFu!+sC~&yQm-CI zH`a4@?64ISr=kGmGwPV{JPBn}M$kAV&XTz^hA!TS!xxh>+?&^nsxu;F46{Lr`8517 zK22v&C)28TJ=Bxo%9<92vl`kdhG*EK$3?7TnFcgOLvd~M z3{0DSL#TOtPZ;R@TadAviJNYoSh6Vu|3brPb6YT7Gs&WJmdPOZG@kQ51XsJg&{zj`o)ZBonGZ0q(OYf5SyQHOWWRn78omGli6) z2-nLN!18ho6%7t2+w+lBJR3r#Hxopqi-cygWt_Y0K${Ym&^+SMR)g58nH@^R&1F-3;5mlkx>=%SG>Pg1mk!IaQ&{lHnK|~)VuOeH zv8>FMYWnSs$)XXE5iZz%8_w&?sl^gQoHQOPCDRV#r` z-*wJuHpiMN)6u*41vBNmcoo{f=hTt(EpQ4s#|Y%X?}pC}NSqR?Jq_NSM0UNciMnAlTj>Lu2O{P_c$R z`79+tY=)^|pT1s@+t$ro?tf!$M@GY9Q5|#@TjAH%hLu|7bXEF1wQW92X+KV(JTeNs zpLe1)Dhu~yYI$G(2+v&o#2%v#IFkfLd5F=Dd(%l}?N+kmJd&?**YTWgBdaQz9zBa9 zd+r+>x^6KY4D}|(`%!cyJQTmfZO{>Jg{l-Qp?lLp;ZlLEP`uzLnpGr8#o!s=#S{qY zw+e*O%Wn!V(jwUF&5kVcR0JCmXGT}z7SQtO8FbURge9A`vE8o<*n;1~>B{@b1Q}Z@ zH#yHNEjrk-k9XOVX>!!Zy{BI)-ot(C401U(nP<0+DLY~!txuMrgGB=Ov+L1$=}~mW zZxqcun98#@Rd|?j0lt$BDB$})47D6aqFqTO#?SD%LA=!g`&vCM~|l@D8S@HR7uWgMuv^$8+Dqk+OXtCTIFU z@#{`l@_*Bkb|B_=1jGVvV?gC!l&e zih|ikZ-jmEL95(uG~}Jbt+B;mX4yF6dz{T$dz|Iosb(oYaxnWVgI^!SP<-mW@b*N9 zQ0>|$$R3V`cu6kwI=T0zGmWN5ZF}r?aP*@LMX1vER-M4&C_!}mpLJ+mgACoSp!%ROPlU)*E zxWANV7K_pF>N@_bzXqS*+W2!;9qDGO=t^lr{SpcKx0dIf7c?Mn=OrkQKZodD3ou4{ zGOF+O!`bb-DNU6-Tk5^Z>!<>4+CGx{WcBHnvjtiVXX2}-8jQrtVJkdFn{GdvVxmQ* zW=do~MTJzKRAKn|^N^7GKfCf)bX+u%ZtpN6n`lL{ncoA+DT8U%KMiXB^bH0Z{~(+* z9JZC8!i^Dkan`vRB}?j%x~m!5->T{Q*;sl~6hZ5x4CtlJ1ajv7rnvdu>}A+`?vY=? zWG8vi@0Dhhwu!UFxx?|hdkC(@gy6N7wJ@18G-R8Zu;4RyKkf}8>mU2baQI@OW&S2% zyhW_AC(93ezgVC@|Jtqd98Q7KKVUddmeeDQIG-km);ZfT-;TOCH2>T9j2A2F6THL5jN{8Oz$;)PB4l5L-GK8wTezY?~?U^l&65k$Vn_R|E96ny9GtI>{! zFqOYYOOBMG2MRA>&DmXHb2^YA`5u~wPC#Ea5cgEVaBg2Iy>(5e*nPEhW?2{}7AE4} zq9aHW%f!^H>1Z8!3gs~)d{_#xx*nvKHBNxWm*-XT2g*CO0r{7aZP;SRKW z%W-q{1^g0Ar|7;YT5S+YX7aK$r$~zC-<77(HxJ?XE$*d1d4$iJFJTDpRrDpF;v5Fh zfd!NJ`-mLLS52jZej{nhKY7}EK%Y7qRp{c;q4dBv4K0I$Vaw0^)4qSDeh077$p3Pv zU;IYFW{I_+TC_^28-Abu7~G+u*>_0Y?JLti-@z`dy~p;&e_;xF*O+}mDog(%O$Tx{ zNb|WI4QS^bsk{Z~`{)Vbzp3OdtwU>%D^UMVO&lDi3YjOmNGuH`{puy8IoXRM-{#V^ zFNIXUt&|)c^yrK}-{%igr4QQI@Y9U%pU-6Ae!Lc>pK0+-V=v2yFTe?xI{X~a1M{^u znCo{G5AVj~Y~Ur<*;T~K#vNgrV%L$q^c+s!`GeD&Bq?WO4`(QGm--iXEHqw;AH2gj zzB`^y{WwUf3I4Rsb1FqGpG>oSCezp1GDtA%VsmD6u(>|_Q72l0SreS`F{vGUYwIz{ z_7w`aC%1ZG6@r?%5AWM(8a!$|RconGpOh7q-jU+WZJqI-+Olje?rN$-&tMeh&gc17-QRqc z+>W%_g;3m*1wlF*{S%6?#)^BH{^jHC#Hpx1XMveKvm$%q1X*Mzk?pnve+j7V5#WhhK}X-_T>=TPK0Crb1@DU^T95C+Y-Dd>Nk zAsX_>UsM*bR@BPxLBB#&>2&Q3-lbZGi8cg@B^vN;xsC|q2iShE1k<*1&;6=N$a3X( zlI5Ji@kxf_mE~!DUk@AuBq-VJC(OIWXu(_+sy@7kx}Utkc}+2@8}S*R2jpPpveRh4 zUWrcL&)tx98xFa77^r)UUc3mWs6|0!_jLlT%F`$DW+NK+dlU_Qt4J&7e8bwo7QCxJ z502A$pOW7bo(>*~4_lOQ;lg5J?mH*J-rrd;I&6ryQgacjG7(>^b8+cM5iI7_V$h*D zSid-lzp0O5&OH)Vm;a;NQ@2vtv1^c(&O=c^93~fB=Fa4Ns1}|CR{Kzb%~tZew4N$H z&4HRC@NBIT98ZhU`TkNw$Iy4MbX% zC$3%W!McF^_%iAzrj*&TiMy7vgTMDNhvpD0`r*o-yJIoGy-LujX%adw^a!)G`JTLF z8=gJz!@_wfH11Re#mu@$O(%FyUOf}@Lj!Q@PX=zPq{7jCKk7L@R9Kpg;>sYDta}HI z$P%PBoX35h(fevYfKD7#BTqLi@CEcg%3O*vP=#+kq<0XYqc~4r-edOqHjj=&N=G zb?0{DPQX75TKpF~PCtgBMJ2u+ZGjlibKmH@&mIc**%G~cMD$eS@x6ZBcfAxFb3?FW zu`_2~$^st!AdJ(+`Z=Mf^eKm(pE%vkYeLWLCbZ3{!q}ovh`AlYIQ6y2Iub+&l($pM zmje{Ftp?VHZ;>Y}Nu7<|EazAOJ9MIm?fp-j9`ZBN>bKtzdCY(wCrqJV3Yrv~7KTy% zL!cAek2F?KpslVt6lbSQ&1Pz3d+P@*b_}F>8~W4mDQYy~-e59v`T{?bR<`S3zrs_xN>6rG(0`g|slRSIHqJed-nl37DJdShzR8HwPe1|5Q_p1H zG1b=RzKD3d$@53WSZ}1x+YjrS2n>-t4aY^gf{*1e;iN|+)n;=Rq`3*HD{wb@g&mCK z<|25AGkT*%!lEAr!r2Nth?fWS@8a

|Ux96#_Ghd)lG`9bNF=Y5;lJl>P{U@b;?IAEfD8GHIGpEcU# zu>B)n-~pZ`kiFNQ_}(|YYTHH^LZ;H2PrIn@%O{#1ZpzYcJY%e?gB_HY6n+`)K$QXa zk=hMVYW5!(S?maZ_Q!H*zh-bNS|)H77HRn8Uoz)CZatUAZ@h(1dGErq`Q*|UN*c4f z+4AA9*yi>+rd9czc4_c4+pTG&qM#@&3x3OvE=Xtd%tqq~KFhXmC=a8uQ`w<(7j~gY zi{;Ll!Oz@n*z70k*#e^^Oc5kw#cVzoxq{E;>U*=6-IZ(#-~V~-ex2)U73V(ukQPpF z93hNMk`ijwnF*(Foh-D{R}!lADzUP{VeE2$J&pT1T^RR8DBSKmPe>jXLf4cH!u-dP z!VA1lc4CB(&qkTjO#2VQu)IQH;lTsK$aXQXbMNG~pFaeaqeAIzLmO4B1oq-!Ec4Af z!3v^}GNCskN4^8Lt92CFM2x2ccY83$P>Ze<2++OkEQrp(0~*F_5$m1tzOprrNmdu` zh*1&lIWt;#`%M-5@zRf_Od7>fubdRx6h#RSHXaeW?GF}Cx_nsJSHDRZ!E^OFi(RBW zW+@p8TZ9%hw}oXRF9?nJb9MKzZ7feBjJ@CJk1Gdxc5GV|9_l~9M*Q()TU{2gqGEZW zZqP_!_;(56{a8J|r?#2S&5xps*_F84;0wC(jAFIJ7PKd4I^Dk^PE$G@NNpFNNwGXd zLw}djyO-a0UlDF2@C9S_5OgA5x3CQ*EtD*ZcQMS(w^Snc9#tiykpuqCM#UYK5j zx$E!4-i^QU`*d&EdO;H`=N)C;qZhO9({x!c*1`5qkKnA@C=BsP7d*78=6uw~<0#uP z!r&i6Y|XnCmhD`PzMk=z@?;pDn^1-kmQ8SaFbT#g$g^AT>Trpr9&WcY;~I-f1cwIw zxNGVBSuwPRD?4N)5P0Qd-ruP__jCtdTyl`BDNo=;Z;b*u{RMdaW)GOU_ksD*G{F^D zy-?n7u5kI1g~G2Y56HRY8nu6|rKpYwXfb7q4tR~9 zKb*xzyQ7#i*_~8>O{KM5CuZgsadPUDxbdu2puPp!E-x8&^kNNpx}2xj&ch@=V>J!F zY(hn&b-2%c7VbK6mkj4e(q|=ok{h;jla6AzeAc_s*7-P&^nya=?l{JjUXee zhIEsQn83R|)_pR-v45*zQFb`r!*_#5K|UJ$M`3oc8_JmHuw6fnv*iau*sJHIsIC15 zz7)knbz>x(rsu~D-?*^UZ^2ZXZba_9yL9)5*(_OIo=pt=P3oUt@_%DArQDcBk)tZ{ z+s9-aS$-6+USGpbxahOEK1t?k;{(Hr%ebnFFP!X-6Ra^dko`Fs$TpmhWQuD8_;-B> z6WeG*BCQ6rz)po~Rw)UST_uILgTAl}N%z^X=SA$QLLBoC_=yg2o*1PSjCmp;9F=P% zl>a_WSRHBy6VLW?Q%72JmyuYCye!MQ&_u3f71Zb(PG^7G61#XFoA-^z#WA&D7I_|@ z3XWpBcR25p9D-wgPI&h7Zzyf@#Zi4;Sh;i$He5Nvew1_Ub-Xf5>4;?IuMe>XR}a># z(Zy7{KC#30AK9OQy}}^bIl{f1meA&&7SHN=fJQ5QFsyyC@a_4z!dEwD3CnIPGIMcp zw)<`ujk(Wfg@2UbBa=96|7ONsOw(p&nIqYX?5*sugge{tdIOW2Jb_|rrRm@2UpN7E z>FuCAh#1#rVgw0H5}r!HC=-T(9ekL-t1a z?fhbzJY0#)6dKSze+j!VQ-#%BtRV9bkYaQxwZ4$oO)s_O%|68II= z_VKx`U`LQ2D-Dgoc_{nO7S|kVgYR2Mz{@}-&SIawU`2}w-tcUMGZG9g&6A|_rQLWY z@h`rTn1w^X6j9+wE1X+lK#RM7WBlcEj2)Uu#;?Xv&Ce-hp70jG&-23~&0TzMx|&wm zZzAt90~(trju%X{A?>dgh=+_}D;9RrUF`>SUSgMUe$z_fkc^p7Gr65|qRx@(lBKkP zKRZW1nTFQNBQaKEByG%lj^7>g@rdeN94~Z-Sy{v3;gCOjmT`dDs&8fuJ5`y4feb5? z=_BVG0mz(P&_I=NRDeIET`-5uDmC1O5h-j!NhDhu?8#Id{xGGd&CK#;HoHDdL+=!R z7Wa(`TcwzpPZe#MvW;GQ`_qzIf7-@-Hy6wLgIrwzr^mZUE}Cz}isx~(T_%uJl;X+f{Z#x< z{1~o!Euev0X;|hEjO9uD;nvR@_R{P-v)LjlRBb!}BR6P606yiqdqk+^@Mx0iTu83V zYAE|w9gVtIPF`VsDA8DntzGuKH#`iB=c^+1&4KOP!(maOHmcNSV);WQ_?`X;LRTd~ ziheR{sgGhWH+ivqV*xw7UBG(cIkr5ScZ}TlkH#NYV^iNXvIjcXncAf&cFQ`6wmrzA z;*l%po6U6A{&YWEH2*AnKB6Da@bjD9^)I0Rr6V5eGDbgbWz}}t+08C zsu7>@q3CRynk0rsnekxRuMcW9FX8yXyWAFuC~nM`$9QU{7v9|d1cnV8N4vhw#5I~{ zVC!u^?od}Ww`sm59QRzo`<=_Vy%E8jtK$;H$PFmw6@ZH6>G0(f&qlv8j4oaD!=F;N zNDsDwgxpS&EtpTIZp)DQ$Pd)l*G{76zS8wMwtRjs5uA>0g^O*g;LQC1F5|}-LByUJ z?Cx+2HrdvW?e?9DM{8HYhu-CoB6SN>OE>b|wxghbqnmUu@~);V1@9hj530~GXs3;txv!JYW~c)j&F{?43?Rk=|l zwrw>H(^DePkfppsTNC?@jbUrsSejn1Lf75K(2KpTyu{KGk9p|DvP_*wN_RFVgP>6qWn@KnOY=xbS^ZL{xpT%*h=DR` zRB4mt&qUJN=1Ds@tR?oA=K^#sV5#$_v(Bpdbb4JcH7n)`H z<8z-Y^vyk(TCOgmU;Ra}dviK}zxe{v>Go{RQ7snQs?GF`9hmzyVj}NmvwXK3xc6Qy zPVIb)AvLM2Rqq6=TX2-sTKAB|n`%~76J zkSvwP6G?Z#XUSwXmWi`FTpc+jUZAyVi4@wOLSq%C(&}-t)MckfSsn!#^*tRO-;|;F zp7&s@UxXg>I9zeUA4P4q;SJFpSbQght=u2T#u`MjEuTiPU-msT5dWD5#@n$RziEs~ zn6uPfT0;5Q@xtkXvBG+nR8m`hk#5c@qZ244d|y4lwtoD}uA~OhK07a3mEuL;-1ecV z#!B3M*B+&>mXP7di&U`oBJBzq&x(!5Ft1Q8wrH6nnjKh()kh}d_<*~}mBeFO&pGU9 zy@}NpWvDun=aV>w)2BNtsd#}iJsU7U#dJfQ_pAvd_sqrjIwrVhn;eek*@ZPL4Ja*M znuZ&?(RC?ZTK`g);&|@Jy-+)B+_(ns^m~$YQyA?Sm&|i28>pM-Se7Jzq|Q4M^nE|i zlut^-h=Mb)s(vh(>>dsoR(4F|F0h+@(^U#OEc$mw<41*jGCCGPZ#RUK!nnJbcDN0# z`18DHm@L~b)=e^F%E@BJLKf5S%yyd{VbfIQnSBwm2eZtX-6&-|wOSl|e!Yga>vA}0 zav&6%I750D&+sdH3-@^LOk|=Go7||#B#M>U8_8~3TltaCko=-G8p}}qz5p9;I$_GJ z2hjYl6h16D0YA@=Vc(Pu*`qmgnBnf_BHJE3tn>PGfS($}NSo$EL9@Vu`F& zES}jtQ(%M6?WAxhk9K|u#FiZ1;q2^<_q-Wi&U*mYYwy9)zgoCUN)qR-7r~a&})@cvECQ0x>A&box|7c($^(l_kg@&$V(W$8NfSRXAdFGuT+=?@%E2{K-POO=UPC|1}m~?8lyU`6%-5 zHWqCgLiZ;}a3>dpyHW!%s^lv4{`txs_;{I1tTBfwNk5n_mJDe={O=Q_H_FDVjOo6es_c#ZFmS zOnGcd{>8H?K*pLzBA@+>nTFE!HmGo6JPzjY=b*4Xn7D|~9e*B$)&qB;XQ&Bf>b{_M z+b~LzDZj33p|k3;SBHsA-zB)emBs~u{L3&kyxk!YwNj-APIn0NUQ z>aEz0A+`y)!2S?=Ur)e4xfN(OPAe*u$tKCs=7UOZnfLRt00Y0K2BpliP#B&vAEx&25~^w+}wMoVFI?q-aAb`V=D z{dxBGJ#O!-Y5{t53Z#rLV5R0+{LuLri+i*n;82qw`>hz4HPZ-l{j+f9#4A{Tw+bg# zyP@-XU36S?4le2ILrAkSEOBzd_J|0)FM1nI@5kX5n+KSr^AdyYQ}_;T1zs-f#5J`! z)O4wwtgbfEO5OptM??f;7ER$<417kLpa1VSx5AYD;Ur(MpUi}@bm93CGUNTQpmLAa zAJPRm(M#a{QxwNGccRX#UcBbP=h94el5Td?=iS+v*t97=3C$v>9`Xv0i#uw)LpNsYvXUqVn` zupj$VkK*t5*KsO86PCAqi|)P0QCM&e)v9~2@4`J4JW!#jU^F75Q3Ai^MfpUx0|4R^-;Wa zBORMNE}`48aa0?U%CDt4?(Y2;+#;uUf>>cNH*Yh+i2a(lEzAg;ejY%}QzBg1$2FYD zyLPA&zNW<3Z*=wNFLL$Ti0c=7V#rz%dNFAsE*mR_b+`Ob%D4{0vtMC$yb6`A9f$u? zv@vvI6><(I(Yz)HC88wAZSM?pbee@P8@;edKMo~+CZhNDa@=yv5}iW#qv^?$*eBl$ zcR&k+vYhZ)?kJk`P?D^CzGCpqT==5q2j#@baqK^!RAG!x1DMBCX9@=1)|5=ot;(+rzyip>( z0vtz}4@LqsXyXq!Gt);qVi@SNSe1jvtO1AKWp1lMPL%pU*St4w7_Gmtb*!q2SS* zTU?sP9Be%R7;wQI%T*%4VtWFtm3s(^)82!}Rlb98ifEum^BgQLT89QkB`9B7im4X&F_q_i4T&l8XN09VGT4A-O`1b|1DDjT6zH0| z9CemxQFM?UzBk{EZx&bJRB`_Pu}+V2BMo@}d^E1qyoBmsTXEZtI;fOy1(SkcygzFK zPUy72nUZB#WzTo`{^oP8->jhK=nt68Gr-3Gmyb_m9$@^7CX6d-#cqo22b#7B9XHg?E>NMxsL>@SS z=MY$m|G`72l}OIlgU)H}giVI~VEgeWuxQy_m{+(6a)$1~oKID7R_q!$P0+-GmnO9K z{x-TDS4}RPEx^sz2>is3!TmFraKqV4c;;y>F4xh+A8C$w&guuIEd-tU=`P>3L`=teZGzwu3-;IA%S^#0Q>*2@v zCfH)Qg1i^_Qe6HCx?wwneKKFWJqG){~@O|fZO73V5-wT8dm5}Z>RcGz~u*U z##Rcq59;BGk`_9gD9iTnd(NnU`BdI%Mz3D)pz*(I5!=6Fh=Lm3I{1v7R{W#~eX30H z-f-r{GpffFYcrv<8U?E zC}cw-&EfgmULJw8zx^6*SWrYWY#-7Rw^t;z&c%#5sklBo0|QGgQb<4)eXPDr`~9Wh zgh(4V=WIG;_wrnh#EaOqQjQ*`$MQ`33$(}n2611X!IQ~VP}|o8HLG{AFALT)+eK^H z#`8ZYtwNqjH_l}p6Nk~?kUkt%H<|>yw6S@+Gis_jVx;r{SPv=UM9@K}fuj@>G;j?-&ES=kF~f;ADs7<3 zpbXmQnnT~EipeqH0T#yB;C8EzC~-%XPR^7fzx$%J`hE$zoO_6J>-l@j<6>gtaw%wf zJbmdOfXCOvAnuVCUP=nZ+ItQdrEZ9_8h>$X@eBMIEJ};pwAgWLV#`<%dzzArPDjg8 zhwtyD9rUB#EAjNgr+}&*5AZ$j6SQPxC^>kKpv&d&@IXi{RtYL`)V_B7w0Rg=-T6$% zzI`OoJx!FwIr5xx4LaZQAIiGl!z=6ZaK+sPJk$Iu1?B!EX&X8A^F|+RE$jkocSoFM zv%-~G&RRHsR~lQ#M&tTGS85%zoho)j(-OHf+I}dX zED~Oker5~qacsa9!*1fqt%^)3MvFy0n#HpF@=$*BOWar_PhKUubSf42dHXiQnkN!hbe(3I_S2>J zpRwQZH)h(mp?Tm9dR%ygE_LP5`oY~eyviAic*fnaYH>Wx=hS!k+u{r9GTgkI?~L79 zL^+4oF=hTNz3*WNJLPzba@S_l$8l-o`R5L0U$3LIQ=e(#d2u%Kh!`s$HHBSWQi2(O zhf~VeaTKgKjupQZQoY4~Y%U6f5wcleJVO^{R`T!|k#4BckVh-~tMF&oF__<>1~cxq zawGe7!AT$wDOWz4_85to#6KKia(%o4#9J%`e>sAcZ&s6bi2^N@7a<+}M((=$Il)uJofyH-5vBiTz%Bxz5^SxF>{lacg?7azp7Qe<-Bm22o+5w>Qy%j%2 znBh6U`=ItbMld>lo@t1GuZim(K5y_xoO>gRF8h(LY^Oj3){zn zZ}crl$gZZRZT#K)tSV;Kbb-XA1UOT~!704J2KaASbFLko&?{vJ-zu;NH_B+0u@TBo zokEQv0@$BLY5PFaJD9=)b*A%9vi~~RAQ;* zxj1HgE9Sxsy}7+9``I>|6uKh!2+bP8U|`og!F@StLD`2&?opaOr+FYnuwo3)eBYgq zmnuGC*>EwOzv?~s%*q7Q2?w}l8zt_p^;5yXm>%4$rG~?nNPyRjY!diZ;hCu$u)E^c&odhr{?nD=;?hVIDyyOm=%b$#;j*hWe4Ta!d%8 zsm=i{b9I=p=o(iwR)Us&NrJsb&0OpTSynLR8F^*K(O~6K2ph^5%u83|{vG2va)V~5 z+oOXLA=2RMGlHw@InB+j*2FuGHQe@qWNv|BEGE5}f|{qIz+zn{N3M4TA+D{04j4g+ zwo}mJ*-ZGXpMZ;XT;XJ&DI6Q>6YOrU_FluR>e+aoA`7kn{R;iXN3d!P*TWnBsK<)+c6huMAFcDi$@I+I?)34hh>81 zVP{F^=`m{L&jb#uswjQQeVTK&gm%Zz!LIrN@T}zfNc?V8JKl%Nc|K3&?a|DpNtXF) zh%nOoirAQo9|lEeONt80Y#xhGw;#eM6K7&*wiQ;M8G@=t#+ir?2&RdZ3OeMf@PO_^ zH1Mp$&NgfIlxOw7E<+Zvl<(b$PsY8MOwi^02sTITJ6*L_WZsMF(QEoC>`&W|^K8;+ z+Nvy?n%PO$bc7UGu13m>MMyMzIJGMM!VOyWc*5y4)y_OZ^D-CGmaAUy&l>J`q!xR#OnA5>%@K|gFf%jWD8}BlD-j>1lI>V@9vl^xC z8AO-2ow(kA7ak#B(jqn z*W@$Ls~PNhNiJI;A4=B0LTQ*(2pw{@$8SGX&@Oy5o*#@LFZ+YkY_*Z@ZBk=u$)ni0 zI4#zxrhuXHA}FRSK=B!OP%1hTH>UDFcgu%hk^30FM!tj-ykmB98i!F6MR1|z8@O2W z4Ca3?0ppSXaR)?;xtAAXxTAri$*4$_W}dAy9>Y82L4ro!l=QP^jM zv@v)Zy%*a+`A0%&#KOa5$aiM1>+)X-MQ3ap{hMdtBc8UAM~j9L=y_!rJNCYxe%$&( zJ`$lA88r<9YBf+*#0ll2Ghpj8WgOq~iR69jXj*X{4OO(jtuwrf;rB=^4I`ZF834)Q zs%Tfb9`nl7@U7-v)XJU-1If!^;_@o+Kg9PU3w+Qd<~-i)FvH}l4ybAEgu(kqVDrQj z(A6jgnZa4?;0Am4!`+T0+UukDcPo^_typX9h|4DkaM-g}`0_m%3n$IzSKIb~PwYLb{Y4%=HB!6N1!oSqO#Lrb<$bcF+5m>k4s(AKcubG9&M zpU8WNPf(h{U8*?Z2S@W}z_!LW+zu&MR4K7X_gBkN-dUc%Ux`z~lRi}0>rLh-4vLp>Y5@;jtIr1G;J3jTz^8MPGH^maCmXqt>3`ZMs7PctbDucC-?qD=pN5PLO# zFI!yW$u8Dwk^S*5oN99s-}^JxC78~Bp^#bmULu8}2zt4~mr6rZ@df{PPL8^UmA z)Vxi&Bx(ezgpEQanR4C*(g98Hv#?=C7;3+d!(z{Jh|3KHnIq3(v#1>&cPNAVJdZbk z-+_V(P{sQ+rBM#30YXPMXqOc_nIA9+`J)?M+)3yM5<|%^jZm)oY@-d)1?+>clKF5cP z@1gkcw-|Xl9+#hp#>Go{)~(lCsB?D*y+AFhzs|eE*g5=u#2B{N4u_2g{JH)bb-MCh zhpevU<9N$C7qIwIBv6uQ-F{No6!Q9gB#WsQ6?X7E2i7#0L}c{=pAi z=xQrYs_Zk@5N$`5tDfTM!7wauP(iiMY8Z7q1b!+#!p`?K7~AKDYu>)5fSDJlF>Vjd zmX5<_>2_Eju8)eV266e?rRd@N0RH6~;hj<~^nNfGmrxhv94Lf;e70&$i85xnH-VnA z6MV0dC312%9?zc6r#MRc|-?f4YlqhB>%$3vYh8@mZ5E9^h!$SFoiVDp10n7VKa>@YiqGRL&|;nNxJ``JV;O2?RgrtiTelNNZtGz%T{ z3i0pu(>PXLhUbE;1e5!JxHpdcnb*LAWlIoqK2?fIX*>&%Ka}1-krwYzjCzW^!V4DzR6^3za|R z!o3$;_}TLqs^jOdd6BZD^J720;LqRR@{&>JKLqT*>n$i9}q2%(a02bYP2aA;=;d^2R?7lMuF0Sw4huK`JRuhoF zcOUxaWs+`gAyr4^krv;p%GEZ+*+(Ygsb%WS$8I8XT)KxnjjMuVr&MsX!Xb3Y&?jq) zmE`a>fR6Sp0r|HsustaO6Bm0@f748AnPf1^P#Zp7MwRr~h-kMFNbIy?BWmDFfUPuXgiJF z@puH&y0c+J&1Tro63C-y9sNEgOCj$?*eAW86dW-?hFQx=Rp3U$;uh2X!8#0>l!Bj-E)Ts~FrSx{7&uLTd7tH`1CQw0zc{+m zxs!?w>?v7WO<0wxC0sLCL)bq;NIGXsD3PBZe2EpHqw;Ee;S@tIoC+NbQlRA-Q)xrl z1r%P%#AdT>oO;@mykwS<;9*r*!VR1wQ zqO~aX@{FrkS8K`|tH3_0k7jeNMA_g3H}3S5%Yu5pTU@H#N!;@49Ns_4d;ew}MaAKs zXxM!OCns$sud@L(s4K$EzlWoEfbRc?_rdYBpUX( z-`DlLZWM_~p`}42l#!&8XxO7vWP}Q3RYpT;P|`3WQquUelm;3K?eV+5zkh${{QmfU zo#PyT+~?fy_uGBH$Mt$X9}k;FXi}zVE1ZC|@E-ccwV5iO&7v>*-JmTIQ=rx}iDZwT zNjl$`k=n@*s9bp%eHZ+M{No%^#cNeG#Vwh%ot!Niq!=PvWOtmnt(ruoKbBJc?lU;) z?nAVeEh2rK=d^%~#ISBDxH=AnMZYrox1*4D^l)FwGimZ}p&aSz2_bJ)T4`NhFFh15 zjb%DN=)rCkT-P$itMqR)H%toJ%T;lH8uw^5M?!PYMYz5`P1E{{XsE9mI)^miWLX3> zFU&>P`&f3acQZR<6UXYh#%#ATvLqOaI*aNDg-Ql0QfKsLV%cOmypqjfu1A zkW)r`_O&v^+9#f)3dmxRZ z5tJCrQ&z{G>oO6^+C_SB5)+7M@n*8m&0dXf*cZLQ(L(YP@rP0>1~@Xnd57tykxxIrld$ zn%6Fb*WVR=3b`%{cp1zpt9?0V#*=+an!?ug4Ptp=J+NvrrB&5W$=_Y|JZra@vw1j= z#(p@k zg}j`>NO>*yrQf?lgkj_97GW}-eti^`pQ1(&J~yRNKFT~lIfTA$oIp%B8&Q8RevP>x*_0|IJzs_DTp3u+Yox{fBQcd{zfSM) zMD&H_#IgM;=~sT0Tr!h2dO!K1=*ZY*!UXjMlsA@Q;L8&HuAWLKyi})oiEi}x+%huz z(^6qU`6WT?pC?9D!aP&bKGt$Uog#^zr;*GN?s^6z{kz z+$-%7$vygM)K?`Vlnh!d9mcie$gcg z+mlIN{&ykYB2}oMQv{vlwZi0=UP$kljTI@gv8ink^V!;gc2UB83cREEi2bX%!kSLz zv8dD|bo|RpbYOQT{caw_dJipNvJ>^0i&&FgP8iGnyFP|3(^4jbXJ?Am#10Y^PF@mk zwz?|*HltJ=s`G^m>EB0k!^Y5-A<^{PqC^_rzm7)4t0L6b0P+Wv;2Xv7F-BiPrF#|5 z2MwV~C!J`G_g>Cbzsq+0SIs6GRxK3aTW;fFif$P;Z$GOdmI->Ici-=yJ&iKvvCE0z~QEO}T<93e53Gk!y9?CcCG z<7N@@6#|kxK+5FuHnENk3 z@r$hq;;?&X#Q7JR*r83CET{h}cJkv6@!k~+#e+I@#TqMO@le4MA*Y65kQL8`74l~k z$7s4lB9hr(PG$|^>Fh`SD0bZ}57TodK*mm*Zk%(2ehKc6LtDd1t-%Fy&1os^`g)E^ zrt@pViLqF4*n&RF`pcbk+hF))1NF0(q6>ZuAUOkPQq$9FRQ8lBy{1qEug?+4^EeFq z$6ajj#h)x!b)fj+mGv0(Nd*O}KWR;t< ze!7Mfd_TJwDlcasBWM!6(7Bmzd-smEdd)#da|mua?#G8%2mJlx4ttX%xbk^oU|$J+ z@MkK__ibiRXD6^Br{dVk-TACHq?Wx@e8tjti|Cf@Tzb>%0@Vl&W4%fROmW2(<|wxr zm&!8X_9X?yS^HRs@oDBWqlGzrw-h@qH5XqprD9P@2eaRMgtgt<#58Oph=$s2qUfVY z?3yHNxll9?o9fh4}M@c%$dbXCW~( z;$Sd6D?N>B>{v{Dk4Mob5*OLcF$wJ5?>Q{Qr$#*e_D%7o`K97X4hN{i(CL`+rUX5E zikbNZAGUhfRQ6&|2@6_(pEVDY7Q1YwSQq)8s$?|ND$AW1v#}rBvDcI3wbas{mi@Fr z+n4^QJqt3?3g`}(LQwi*&I1{XQtubkRe31;Irs}!mhd@GMm8N^tm+;qP+M#Ca+=Rz%?f@33ZN+u?3RB|iI%XO@2iw)>I;vy*>B zBYZq)@+uD+S#{gUT0(Lh4o z2H1QpXEoj2X&X_*41z{6Z46~4W{2g97>(B8BUqf&C{{0dr)uT**KOWGIEz;j{^%)8(& zo2-79neUj(Y^GbVjA=92|2QD5brSL7`aakUP81K1iW8sh-XhMbPC)RjG)%Vd$JU$H zpyGxcYp{%CC0Q5oQLUiye%NDu?$`Jn;hSX>~Og*)WyQMf+c&!6#Ic3D^6XclfBO~GN-vlAe zHim4=8;vMk?%uJipoQD{EZ*xs>ghfi?klFVIRlonqp?xUN->EpJ93%II_M)Ra2q;5 z?g!1jj_)Z2aIL7q@&rj{5wILmxuyv8R0Cf*hlh$JB$9?BrrL|CaeHHDr`!%1UoP!n{GWALO0bdq(60L(xT@ERQi?%om#XOhTrpWNJEvqk8MV2z5$~n zQ(4zMXXgIMn#q0AV!KjLVNzNP{O>BVH)-S8*4;B%=T8R~HBbqX!wfNDmMzxIKf-?O z%x11P4lo%fT_IRWm-AP)356@Vh0oh6g%xW~3!nDvq0Mz+v~TBHntMV95z|KD^S7~B zvEm#->?7@0rw~7y0Gwwg` z>Ap>7={+Qc$7Q)sQ4+5GB(QF4Go(9PQN+EA3Zr@ED#r*Gd8+uD-jC(#_hC`obNG$B zMaQ4MOoQr5>9QO77!|q&{iK&5?Mg02iV87qALrg}GNhJK+C1CnKzFF0MS7H~^`yLim5X*iJ31->u8N}gH3}Lm2WZbY+ zy5Ni~d=e*NZnrmzjGeIc+-#hbI7AK4WKz8g7wEx1Qq<44gknVn4M=0n!`Jsmwm4x>kVFV3X)BPY{u2v)1gg}Y6B z-D^QM@w?wl&g}XO8PO9wPkf59yi}^Ae}q0Bc!ajq4y3!zjilu;q;e}MRfzYb{T5r$ zfZu=VpgW@=@p=M^c?YpZI*vQrJ!s0nE2PAocchf>5#y#&SUOJz`WI#RJSLjyXK!b* z+3qYUpp%??wV256+e*e>>!sRR2dM4YV^sG3A(FgqH>uvIL%g$xlk!b=Bq+Fs#7vc@ zQfX}@ZTAQAZ)^^p?oWs1`YZ%p-Awm%h12w;Tw1zu7IEWw~$k6gF@ExK2FMAROgO>cVk zqZfbmr|su0gyJ*Tgd_Jp392R===h=D^x!iuYT6yl-QjCUcYP754j&}4-)tb-Yw%WN zx;lrxJ6A#v*B8+Rt!nrqql7hclwf|wjsD#U+WUD1t=f`?t*yIZY!=JcB<-k8o|tx8 z^UV0^IdBYjL2Sb&ynEQmiW9rph&>(bWEZMp@m1r-_ zC!ar`Cus-NF`;TS96xa;yvYv~?R|;P=l3wIp9j);=qftV0D6Xvq$GbYQ9ZGcJdIE% z*9P<>(^krpfY*=kBB%@|TeER<@HIy6U1Svor+Me6g}U%{ze%UZz*8ljwM`CVu{&3= zMc3+)u$R;k0C#1Kn9M9jal9 zuoBD>c3F?AG)QAoIYFd4p|h$_Q#Hof2p(rdcR4%LfqR#3-5tO-@$J9Qou)DlNvLTH zL+Adj_++U?Qzg9U4AmC;)548?h6&3VZpenGHn8vqH`u=)r`R|}Wi055gQ?U-oP6Ox zHTR~`?cJ4h!X#O;tY{L=iIb$$ckL!0`z*+so!3ZVeImLKS!2ehdFWhmnElE=%2Yn& zG4o?8bkRmfI+>cIskxGA3_HayUxiHn-z0R57>E9{v+-t>3A0z2&boA#FumR|Tu`@w z|D|1695;=wlXRgwKF3k{&jw6ATbK2JAj|A0R0v6%7n9`0=|ncpi@h#h&Av@u%EoJ# zFpu*E%wywWmRoy)c351c>;ARSyj8vIt$!EW`};F1ocV(-e*KxfN`1=eIy=dUwR`B& zn)}q@lcM;Ozl^w9ZiF~$wE;Pnd6XMk8Nsr=3CGbZUw#8CkTOoKR7qzbj_3F`bSqepwjX z$o+uP$vc_#d|wuEpq@oml{3kV(`?MDuS`DvAM?#p66f|d;MIq_IJRgA+oPN&{^`9( zd{RAD{AT4}eog6yqp}p66|jpwIeUQTGcHr&qtD`g^k*6!Qf%?^4XiNBi^XbtuxUKc zVmfC#&F9`smjfBJbp0*r$-Ocu!3VV0=34PpYEpzTUtzat9 z3|vG9zmvkIS=0HNSrBntx{a8MmyyBR%n8FsX_%6C~a^31gbFh(_vr zax|%!*3Bh|nKS~g8d~U?n?*#@(Tp_x+edcJUkj&e=J0BD#ivF6n9lKm%wysRwn$Tg zuI*jKJzPhrRPra3)j!9&a%uKLZynW2pUItk&6Eszi%kC)$ksTG<4#PhFiumPST7^C zcsGQt-K@o8FKlK5jy@sJ|Lr98MGwh?Rbh1c7;pOADxF#{-;K^}Z!C?Cf#rrp%wnVs zJNs-Q%Ukyb`npoAElZXe&h{b06)7G5kUK;c?jk1lw202PRPtIOoK^n(AEtr}+xN*w zoMAOfJi%W}oE4&rJJFup<&=&S?h4|u3IoMY`v!^iuEsH$7JoLY)0uVqJf%*YTPZQ% z9X;7Aho+V=x~!{*yh^YYeVt~&gLW;#*~DCdHAz|T|K59I69Gq2canv=M zu5PuU!^%Lw zr;7SXj>ga*e?>E6xSw6;2$4BqK(sne5hwNKH2&u$viYV7>E&$Tm80H}=F21KZtV!p zRnP;C+k?ab)#zHHiIHREQ6{MlZ-)rF`s-lWPS-@{8lDf?!)WYsT`HkpMLhG`iLCII z+&hs&#Z7l;@StBj&+Lk#h3hbGRV*x*%A!h72l6KzFnYQzUHB%B_MiKV-sm|a99ei@ zIF>n0aNcS{hefDU3oAwX?585j4>4d{x-8j%sC-)2_Ki-mGQs{4Pl)RI7E&v9p1YL< zCVeA{nf*J?TAjz>Z1*@EHgH1Ad}lH>L5-BO^rPRuPoz)ujp*ExWO~$3lT2@qBOPk( zWX`8By8qMxy4*d5HcGjXnusi6%e{r*PnW> z&BXckT!fq`#*XVF=@RbPNafr|zqe8J{po%P`Kkz~AC|Q8hZ`N_>_@fpB7~ou!S|0< z5#nP-^+t1wR_$Z@a=S7Ti0cFFp?cD>@8eD zvwp|Y1%{EdYq~M9vx(yCP*LPqTN~{#9EJE81(dk+?*6b$;yhT6_6!~+xb#E`ih37? z)0cP9O@%eIwhx37{Q z?h{LV9zmC^T#g-6cc8lXFiIYJlj*5bXkI<{bI`-ob;<-j^i!v_*x-wlI5fYPY58m-p5+I~zsvuT{yQbH zM*A&&ed8hhw_pfW{yB{fY#T}go;cCw74vEJKofey{;06VQAN1ZnkxEUJ&djMK82m5 zPGG^aMykmjipiN4xM0*q&qf4O-8*J{t@H;D9q)r-#CxbN
p&g`k7GrOn#jAm%) zQs=%2bWW@`eHB#9*Pi3an0j@g{S~7giQE*2t{1!qKZVWTLX{3TJxjp?F` zE7_(5dltHBHkeFu!8$T^)GCJn50!Xb)GnIS@2VJR%o zkULqN&1uE)Pvm#}KyhPAADdhCl)1foE8IG-Od4kH5gyp5BeS<0$4YCFP<9nz^5j6W>k|k8@8&)s8HPw_UB_(!lvn!lXFrMWT(Fu zF-qKkgNmBC;?zT1m)?N4!8Mczl;GV!J?eaS0bQK9gi8KV7ja#y=uJh8Fo^qIrsynY z-OtRK-}Gm6ZOc2lcHCPU{PPEqiC#=X5;I8m*9gu@a%TqPEZBehWysgfBZTg&mxVhm z7fJSl?V_;TULtiJ2eL&jhBzO|BQJvMg_~2f1j$5Mq4CTpYP?TIcxH50sGqBCWOdy@ zFnyRKgv}Ia!I&e&V8&h2u}qcx)Ll*=*8HOSA)aLB9pi^SSXE)8bw6+;ubQzB9aW=cwhM1txT-K_K-o_^8~%7 zpTfQRq4ek?HD+2gngtJ&VZjy22=iOe*21nVu87ay!0Bwp+Bj|rdVLVlAMwv1YVo1Rr{;_!>iFyI2S2vlJE4P}_h z%5jXnDkIlF4tR$ahb#YvE01CCFaN+!CVpW+#Qq)|CU!Ezs-#3y>4ID>?U;0QN z_Rpn-A6@CjJ5C54Gz*H^YEV(vp%wL0s7wjyJIP%*xOy-AmEzIXs|-7@379E*PsO>? zP^s;tPX-;O|I)8fi`)dNJa8Ai^D7O3a{FM_Y>BP@UG$l`HSNgmAm*GI{H*wg=zW5O zP~W&;NSWh6J2s9bZ?@UftQ{+<<#^sJ?JcIy2c4zIu4L2q!%LulJRMT;(Qvhhz`nQ9 zxXYHo{W@n_Nr0ZT9#2PZS3$Y*K$LxZO+!lV(VP<*)L@B-PPxZ(B})UbcC;5V0#fL* zWj@r)-JHh0UIX(J-dHIikJ}})X~UXH^svl55EvGk2{eni|Ej%BpSWv82yvcNQyHh=)|1Y zWX-_w^kqmd>G$pck^Gg2F0WJ^pyS{(UjtLl8^V`~KyLKXmL+HD-=SA&MMe(oxU-0Q z8vZ6)>tv|+%JZZlVh&lqc`qI6=1W`Ftl~@s6&4&Yf^GXL#lEzM(lzZn>5S$XbdO0I zwTL}Vy_XH7Zzb}XRzo!NnB~B3WKTp_*DJcnzJO*uO`soVJR(n|pON{Ezv;RstyFVY zF%@)5$kAVqh=+X`S-J2PnPIb-cdw_=qs^*N+Q)sA4%qnpO-=hs#Vqlz!4h z!kb;33$u{!xIdT%3_XA=0b6igO@vJqXX&H0TWHUy0rYC!Yau`B7+Jhd4THn$MaKqj zB;tzk^n2=T!TCxgsU0zyemvtMJllVp9z9%z$1duE>NR7|T8cs4t2~1zJ@SItyQO5K z-%g?HLnNN=bz#Q4CFzL4$4N+GI*rJCX;c=+-E9GCkbJBxTp4$pN-71TV^*iIE1_4A zE#5>oaL-)l$Rt#)3t|_CJQB^Ex`n#$oQGh8bD~?G%Y_}4N#wGGhGC+@0%7apJo3A7 zq|m-SiOM;u<7Vm&vZ^eW?!6I$Jhj83UZq0v>J#TIge3~?SM2cV?hICS>VnZY32E}S zaU*}eI?dhE)o^{0$!fYvMNO&Cgi=E(^tVeC9yCoL&F$-{R{!gw=q7vNrRFti`D~QX zS@M|lJSnHv=k5r=J_Ln^C1=5@;e^#q0Gqj**~KTQrpQS-ArMB ze1gzeR77L94>Eeo8OP_$)X_ORNc7jalys^5q^3y^g(1=>Xx5Ps$SVyX{j=`T({G00 zqh_{|O@ag&QN56!eyb|HdvTEJFO^}hnlnU$jcyCcE5~Axnv5_%AxF5JV@EgJ{t-U& zwfeVZtJwaaIN^ThT-+G7gsn)LORgDAMwPWZD>qt4mUfKea}P;e3Go*VsK^q!Y~PX( z$!elOQpSQ|)(-kpdK(?_rJ6o3|3Q-vel)7s8BJp8Wopz`Dmqs%jjU)S^zT?pk$%lJ zVTOzgHCyN@N|8z-o@3;Y)v{K2&~l383Vgn6r)YHjM5gd1wTJXQSS#2BzZRlKo+1Xr zrt;3ieQ?b7; zlx=uMRtJuO=KVm?BjqsSs_sjT%HNB|P3QU2ijnM7!XF{@s~K)h!@t zF;$tFA?mD`z|PB!SesxW*gM3K{YyQl)E_yKp42i@UhtTPJ#V2pBU)+cwy)IKxLIWM zbu10^tfiBe3>IBmaF!(5@lO7_k%Ikgd1}OYnyvkR8jTG}A!P0e`e$*JV0U33_0$T* zhJH4psoQIX``0-~@X}zR?~#Mxl2J!uHkuPDdV{l09)X`ON#Jl*mLlq6E#uq_Jx$IU zn)Vb5V^bc94shAi^wu=GD&?!l!NH%#T5Dja(j39sRMn;zAqW|4C{NLXndiweX|A+VgpZy&;Na?wS>i=&4p)WGf z`#-$@|MB>X^q|LAZ2q^$|5t+#v#s@GUi*CbJT8D$y!B%xN~_s0%Qb9{o(CJ)&z&ti zv5J)+UBSA0+*rN5D?7f&kv)!H$y9rt*snRO*@R*4Y`eT0i)U-usMa-X%Ty2Mws9S2 z0sAn=A?uk(p)a$)@Zd|6}II(9?Po5_ywWVM6b+41G8*tI8a?BiiKHua(>J3M$b zTa~hgC9U&ehdun+tCB#L#KRh*J|C9S;mP)n@n#Wqe#~Y4T4p`bjg4C5!u0>HVCN#8 z+4t2hOy}(ywz$!odHq?(MxP8|E%qDOvG<$V!dsh}(pcV0w(?@XD&1I-vkU7eT*-db zy0D5fE7&J3NA_O~KQ0N*?0(uxwrhX~`!~jiHGf{mCfWG2WyzaZPV9O%xX_P!O!JUopac6DEUD!e=7nbwWh26?=V^NNttYWSgTmGLPBYYp02L~`$=U~=(K8WrA z5yXD(-o#vfZ)TFdn^@WMAf`Ifm-WhdGxJU#_Pxc2_3id%JEr=wUvcYM@n}C*k+qR6 z*YIaqG|#`#nmtjM=Va*I8t?w^BoRkNXYT8!X7Gx7P?Eci=W;(h6C$PYEg>ztXGyK**e z)Xjt8J6jBjox}4vHZU%LV`58;GY^)uMC(Xg%$vmrKY6_pI8JIeRdkbRiQRHlmP0DsSA+rLp50@e)Xc?ZD zbJyy(`S>?(F*fgC!RM8WAo0x+ezq>iPh1Y$s6{Xww*X!(ws2gr2x)MHr=}ZTdAnoq z7#AdOa>LPc3o*lK0lqgnApVUb<~1#YSPNva8iKMWVPV{4oL?9T#XkP5`DLRj_DZP2wN)K`OVlCvO&N~-b)PM! z!&yxgigX%Qf1L(N$0=}cp8}4k!wBi=Sb9zu538nPn~erute*nc2db#pHw7KHRFP;o z36rGMaL!*7Pm48hR8<=fe@($nTLXv-H4$=83x@-z!YEpUJLsmMv~U`XT-A`*tcEA) z>S*_vjLs)&D2SPiywlV0sCg2$b6(xE#A%$5p#cp}NVF03@W)^}dL=X=F=RU4WpnqO zzZR4$G_g2Jhd*NyxD40DgVlz}{7hkg-4q{x>f?HkDPCBbLS1fhzv}i2FMW`*w;Evx?wWfD2MzufRqn zCpet*!W$zuRPFV~!Nj$Yn7#(LZ9UPV?t-;}&M02vj`&k+A$h_T6<0lR#>5K+KQfa<6rrq_OS;x9{0fs zWmhdf57P7~AfP0Ex|z*|HUVf3_h#Fbqqix8b&VIIgsY zVEy;4Xw?cs)aFod=^}sj2teq(09e%dz?Zs1Mc)VArE9VBEMMFFF9;Rft$Nif7~=-} zWBKBh=n7to7(Y)aZQ~67&2I2=^u(M%cO+i*#;#8RSoA9xvkL?9d~5`MSVdrAPZ*LP zh2i!3aLm#S2C6pWqwP94?AU-BhYh%0>W}AAyo<6d1ABL-Lgo2kxGYXV|8@LtwIk>l zm4W3c2QX&Ieq3_mek02a#D}J!{CX;KV^Z*?{vghcNX6HG3Fxa&!o*uCNFKZ&C9(T) ze(65!P*24B2gx`TyB8`G5|MEv4o4N@FspkPj!5i5=-_zJ8~gDnDFKV_CqeI95>6N# zz^$(txNtQWnzwQxDVdD{vx?CB{5bFG6k&v6A-+o=$F15TEX?DcktKzA6j+RjZG1*` zB^{T}Wg&gRL0k_!gtmhj$p6Z>^*|Cvi}t}hYA;e$_hJL@_m|1-M#ZB2NWYSV1=ag7 z&Mq05FH>=CSpr7Q*$W5zLpb4=fpId2U`*4oydn*=|EBSdQz}gCl90%KPrfHJa8frL zk5{JQ41b;Yb_U{{a`9Ow51;2|!8tw;dsGUzTk|+v&*x)#+G$9tpGF)l<-DS^2ub1` zeA`pF=~{}*ch5o8ehgD3i!ns`BwSXPVAba$v>iBwLB55!vY`lOqmQ8?DF=HJa`0^6 zG4xkBfyAy{&K=Idys{ijq$t+>s<$VSg)u?gSBLrOd?cE&*M~1xP<9 zu)AakSwCH*og!H3FW^GDA>5V%%Qxs_IL|D$#S;|In1PLZ=Hv7$b2wilu!)$7tJjTi z&C3XJ3VMiBBv861hC~T4_z*?KEfMl3GgOD_qr;WJh|js3{}^KA34Qp+YC)n-2Y>u2 zX1y{&zd1mQq&XtDnDI3tGgLX7Bku@-oRff@!G^ee)fmDTEo`}A00j$8XzK`QurS7< z91)+X5wvlI;^Jb8m#Zl}G_5f$cQ(=$jZx=j0#Wfyh?ZDjLVyi6XINp>LmM1XG{ej8 z*_gP<90?QWV3F7oo$cVC69zVt;?H9fOo}qc^X2APT{9DAwpPe35yQ{I5@XGcF<|-( zJnfkQ$);J*IcSCyeG{-e6Z}vx!|rM`%vZ9)8dxE49uOUGh8O=VF=)?B&S9~}k|UPb z_>|9`7g%E^KYyF^zE)+lJ&NzmNB2B)XqnD|w&hHO8O=n-VRNKs^Vb}-LE+wcsEy$s z1%5yF!F4XK%2;CZA8XE_vW5NbnMjyojnRH)(471LHjQ_%M)fgfs69Z+-xmC7yN`cQ zAHcWdE`Hr@!LJQZ`RC#>Mt1PE&pl^2@YpohEFrpSW{%eMAO)Ctx+(&uR9mqTJV|DO9tXls9`-`7Jdv+U~ zM?c43pI7L;^afA*zr^~BFEMAx6D%6@2!q=0;m)$V$ct`9toKc1nbh+8l18|v*WihN zHL466xtHK7*7v)Lg%en)j*@}4wTXwVL$mcI=4PXSVk+%Z$E+j zj{E$(`vBH?_hEjl8GWYrG2gEhqc{h;XKXVjJiZNO^}BF0Xu-Fk_h4sz8`gYZg?IIE zJHj*D4;lcA+c=Y552gG5*buW5>dONWJ0Ki+G>S7t!jN<|0?*~5;NQfrKbJQlc}Or^ z5d^=5p=g-ri{g#j;Gq+aXWdaa)w~@)`fo>fzhEqO4TFeh-wMt~VYls8R1XeEt!OiB zB14cdZ4093ZoupOFbp?ZkI>>Ed^o%XeQrD8q#1_n-C_7{9SV!9+hN5*AiXOP`*>DX zSvC|8iz1Nj8i~QLqG2Bti_9}|c-It*pSf|ER2q)o38B!piNuYxVB{-CLLxF0zcz;9 za8U@B&E5vRh9I(tUkskrIZ{e7EFcdcLW3gdwEP4&2u>DX7 zylzL~jD9?Hr1@OxcO=?9cO&jyEEF&7Mf1u;kPLqP2ug<0>Afh{*^92vd$DixJ~-Lz z#g^bbP_&D}Y}IHCxE=$KKd~4Q7lF?oB5?nD6cU4?0DcXtm=TMaExRD85e2Q2;dtp9 zg^gpPIX^BA;`n{oye<|6b7IgS9|d1)e!TWXAgF&7y6j?bGjbR9pNqpyQ8E?{PDVq` zehj&qf(d8iK@<+*Pg*J>?(aw0Y<}&(l!kkq2jH-Z^SXjk@nv`-l1HZ_+~^2gE*ykS z+F?A5Nay$9X+ZUU7^v?>=eu~ke4B(Q&jWA|O6Sa>L$H39jBkrlq5JkARBomtJs<;t zdZ|!ad>Dtr4r7>FDt>e(@atU~%I2ov_xM97bUuVLyf6P_?*aISCgR50EI2qPp!i(^ zJbhDeMfV`y4^4xd?Exqk9l+er8PLCRn0w+5U{A$<>`YBXm(2m#=B6Qy&*}DgFjx_7srz1bpY~(q+Zm-?%ObkDd5TESw()=_Hi$OjnT) zKQ`?Lu!3Zwo%dkBw5Gwr}YM zq@vaDFwR)y;<4Rn%y-Gdu^0KM+m!>Kqld8TMK&t^GjV+Iame*%W5(8e*tiwJIJ*dU zhZkVq>%*w`ErNPd7Q7t~!{Yc6#B3^nLt{R!PRm4AWDfjXk04v87~3Bn!}Ezb5au6+ z$&x%gI++hUlf#(5EFD`T_CvO3KYl44!1ci@n4ztPGZz#P5IPxk|E3~s+cYe=Fb!*N zs^e?KG>q<4#Cs`4g*+EFCA05&e{~%VYvRO;Ez+3?-lO;MM9em{>XaxbN*0fftKdO{9OxWnY&|;>UQ%N*aKJdM;auHHOBtwz$Rf&)duZy%V{73! z%rR8q_qY?W#+g4i&YS>fsDLN%v8GW8$p@7Yepw0At|_8WUkwr6(pnkMzQ|*+-egR5Q$d-XDiU0{L&9DOSH{cX)-M&P5=Ceijl6dEbPYOXq-|EGlic27iv`2-|B znU0Cx(=jAXm7iNa@N@h(yxG)^tKWOjtk#1^?j4A7>E`$BACV&821oOE_@IvcVgS7H*gu*4c!BsnA!0K%f|nNFybqgdi=yE?H+91!?T;+ zUGU!93G}^2R=+nmXWEYLrSC9h(`$%(U*r7tmndEJ5?IWUt-Mc2T1jOh23kOLHGI#w3j@`tKyfqvgSR;C2~fUaTg-H-t!zm7bc{ChUSF# z(Et4jvy?soYrev>>=SHd-(zmmC*-{4Oh~B@FlzmX%2%IYF!>7xSboN;?61g6_=bOW zKQZ>oUwF;>jk&udnCM+Et|oj(jbsr-WEK~2{j*&iO-TleuJbe54|K-TvK9oJ|fuZmV*S__jApH-<2!BDse_`U#zj){V z6(8n*$HZ5EvBRkkQHy_}drBXICx69=`7iN(U@zile8bsZ3HI#rPb_)-7s_^$?9laD zY`|_4_N|7nUK7I18Vp&_eQkDWi4Nh82eF#9Slh_0f?1T`I6lSADi8ONVVY*JBfBOk;W72255% zmnqIOVjsSUSYNG({kg2qjz=4@5FLR@Q zu_sdcEYE8?8=I}c%-?FT%Ri>D#OKqQ-T*CDIfG|XmI$n6rU9#I=lwceT_&rl$;`E< zu@NJ5SZ$3iv)yIDZpR30zNaBeU9Qa}w`sHJG;MaIK$BIK>amr2+Dz=G$10{xWw*vo zV?!Qku^C5n*g0D*Her@Fb5)tbG&WCV4mw)wVzEB6ULaz-%nX>rYRb&aC_5Ho$gGkW z3lk{2cHWqMl{90L$BbE25Mf&%7_xmyBKBjT!0t~Lv5LP&tUb+ut>MR}|8!lZR;SDR zQ(A0C0MCwA>#=`l44Au_9((bOurZGW)+^1mHNk`pTF2Ph-HbVWBJ8!4F>90oo9U#_ zq+jW<)z|b`_9so27O&4*u83Imc!8Bh3e0smWf4bBn6HgFYaDCIl)FsW%hNO1>^V)S zj&DN$gh~uZxC^h2TUex152w+!$a21dx%2DbpHK_^<+t$0x*WYDDp75A4b7{$XJK6p zZ2R4U^OtI#5xWJLS3zmjWt3g6fgZUArG`s*(Q*aF_LuNXbQQI$FCcGmB|^ihaM8U2 ztxng`kN5X%`8--tR}XGi{ajN{|`my9ar=J#c_M@O;Xx>r~5iqAzLVvinM7c ztBjEL-h1yIl}bZXQs!4S*_)JPOMd6~hX-=+^|_yWKkv_ZpV#yCI+&z=3)>fWpd;cA zq&M8er@?KgTYU-Fg3m!`;1cF6xQHTYo{{&l9--A2V0rC4+Gf`B^;i#&)Fz}gUVvus zC0u%c3IFalqAsWvCmz(}?cZ8RdY{L!Qx&LhK7-QuLL5F&1vA&P2t8kouccK`)~oR86|TzVjC7H8o<`^NScUijPUD4VFdM5NFwnyUumEn0_8j zM$PQ4ZpHY_8q8c?2i?+Y%rWQL0qKn}-F5+wN7ljOV=ID!%HbtcgKbBf5Pq%!!7I<< zn$fMdPS zaVxnKqpv)I!=~qWozM#?0!G>rhp=g{@6DApffk(o0+MGwM2??)JrW zu66ArJaAgu9o?fn(WmQ$2y1`t_n%__iWhLl2j!`LNM7!Xd8YzFodHN2@Q1>WKnyAj z#0t$oeBA4c(Vm=zBjg7oaUTrX=Z*L7K5&@q4cC)i@LuhS)t#sCRNN2iCx>Ay&#np4 z4aI_(ApGez*{ezg5@l$8+}@g@pFGSOJs5sFiyVfghk6kUY@2$l-Qvk_sa z`V@hq-=pAkEfgKbVaRul!qh8)FgP29BAqD6j)_C}({QZq4o9a}7|tAw;udHO?p1~& z>~}QRcp)(H;#rp|A)rSASgjI*4B21^e>elFkX)$6+K9kHVe@Q7GOYhih%oc#@HV-l7boMVCgRYlL<}F1imK}g@Eb@#{K6P) z^i4n(*Z*HM;<-NOIpr5Kk?<@PX1mgm*O>;_QF-vWos5}>GVp0eHk@VBV0b1QTK7_+ z`6&adbz-pQza%(%r$Vc{5ci|A5n;pGp8OtNS7cz4cmi&vCSc>F6lku_!Tq5ba9W%L zSMxlqP02)KaT?|=;m2CV!>S=0t7=j)zBCa}zNSET1FJY+(sA`rHdY+W#BuX9{27x9 z*SC2n>PWy3hb(B@=0M&(9pwdyxXZn_6Gzh_?3xROnoP|2n}NG|1$f?{hN}^IxUf1O z_20S1U!RRfDn)48kdOGL6pUDwgIn3Ls2G+5O=ii>hNVGlO9o>0q(R;+8R=qq@IRl4 zpweQjDJw>DW*J^M72@>ELYTcTMoYk1ct#gtQ+yG|Dxb!ivQo6_RbX|%8CZThjUn@k zA$MRqiOkqVHS2cMTdrCo796Aj*pX$jEm=wIr99SdhwGbDy@v%EjW;95c~&HM$cS|B z8_}JNO|)pZKJ}$qZ$$vknM`}~okwo=Y z@;5Z0w=Fy9x}^oVs990ZJR3?^-A#k`*wE{(Hk9kOlSXBkQgeqHNsKn6niJbdrpJWV z_ZiZVpQd!9x+qTE~hRC8rLJ)O3hv`4O|cEfd?(Yul+4{Rc-uZDaahGgfxldAq$la2Ro z3aqyvW=?7N#;x?yc@x=B-pn&=*U^{H8%V5YEk(ZHL=GBTNcpfaxpeKIY8O+AEHtJ* z7X#LRx01!N_4H(_9vQ4zL-AAAQ0&{yv@&BWU1ANn_p&kB%Wt8&o~<;&co)5A9k?sW ziaI9kBIh0p%5*m+d)Zx7na>*c5^I9C32WGVEP5ssscJ#FKh5Z%t`Q~48`00gU6elC zg2dF#>9K@0b#jrgrqq#MdpeQxODoF#u$#>PTF{4RV|tdnjc&i&M*5XT6l{`!+Ea-* ze;@&O?-2F?L^=86+R{LD^9;J)Gu{vt^@5VO8}@P! zonFm2$y>KEo+mBS$MFC0JXLJ?aX z2(QTjh@a+SPyYWlf0hGh(%y-UkHYC-y`*T6yCjv!ie*c zJhvi>xxIKCd>N0!6Vo{-I|ecb;$gcf3g@Op;NZD<*er_V%;0b=^NK`5W-Ro^gdwOV z3eBEz7;!8P`a01#r4_|9IwBDEJ`y5JvLLfL8Sf6{LE>!@UYjN&XJr~bUQK|(wIqx> z9S5}=2?)QDgi}e;XxJEns6#<`RvU~5{lOR=6$H68-q^8$Gk_QHXIszpLb@k1nmmyH zECAx=flzV?!^im%a54zz=S1Kqe_yiXB5`9*G)6`S;Dnz)4vB<8aY+bj|1xvFfcsDm zk%-dg&wO4GZdZoE`zSxJAr#?Dqp@d9G)h7fF!D$$GOlFcmRCBeh9+ThPc9rCcx@k^ z4vXcyu3t`t|L9cs@;zb2pES%Y%)-XO8Sqfe!%+Pk3}~ieTvZO-|KuW0Hyz8@=ApYa z6Y<>FX_}pY6z?=pMFPz8;?eKHdGMXwv*SFj2ioDdH!&1%#>8R>fB$b@3`2u(1WJp- zai%vMy*EQqEgFr~oOrB?ON56{A}-gbV$6s%WZlTZ2;pq3iA_V472i)rrs6;61S~(v z_sN@4I7Bf(c`S^l$0DUK16!5zaJxANraa?9?@b1Zer0k!l7SS3Y;^re#f@nRcp^we zt7Hdp^)K>5#g3VbUcYbK}!@KxAW&F7miWJ{92+>$S;V3+k!~EABe`S>*27q zh(_nzFuddb-wmY%teBOKo6|C}WGSDw_ZhJ0Nkds%Jf1E}<+-;>NO_xp6ID?d6PSRR z0X(C&Ck5R$sgM)PLIvO3GZwJECX@&Z`!uw9rsAJ>IwtPS#^PmJ2>qCel%0J4OH6_P z&=_dD#bC&^XvjR`I%h*H0+^_zVI!ZzDdV!>ejo$Q3v&?al!B(Z zbfo{zMoTb%p8c6vYLN}~9L|#XlEM3&9H_fzV}DF8R?pz?{IWzyDkfsX&{SN$kc_T$ zK3{gp_{Q_dtd2!tM{XR-vtwYpA{|38Chm(!4cl81yF zr%^LEAF@0%MQBJ762ppMnN)!}h1Cce&AASG^>DAN#FVlcOc`H-)aJ7|CSHLe<0?$K zTZ>CNjfhFDL{4rAhN2XQTS_oB?=058F2>pSl?YTXfs}0l9;oNzm~%D?dJ5pODiaT% zE?j11WAoK&%=)h!Cbv(+?Qa=`+Rov0S{e5e3*pUs%m@BuxNTX314=D0k#0c} z&n({fAMa7;R$}M4GjNG1L!d$-{J77m$QtX=GwIL{$v~S*4m50X(EN$}vF%0JJo_yC zCUTC7Dr>Vw)hJ(k4u&mdoqYCm?RR}V!gmce%3}iLJ>%SJ< zbE=1PZaprpXvCYII!t=p1{00Tcz)~>9<%1s+S-A#$5+so){K!$>fywD>AW2^IG$07 z%G`4pP(07IU=8L>V-J{K725YUU|QQ19K3fEWL=L|qjPXfFNOECbGSRC4B>G_FjYDY z=ev~{dbk?(K8<)RbOC9st4V#oggC`(`26_}Ym05T@V*u0L$5&ZS2MnK)uBG18K)AP zU^LSnam;yKX}83W9vduNY=vR|j)+jTLHAHk^w;c1?Zl%fk8;J`_eU_g`8c-SaKfRa zy{MRc7;SooF-dMO+~(S&^bm6w7xqB(hYd>f%@Ft10yVk2FzAFKzV_}$?_DEo?%j;J z<96bo;&!BKnlguB3||LZILqz8A7-rLYm6{;qct98ZAVr0PKcD6;H~ce#bqPhZQ6yq zM@^umYm0wE)=1rMhm8t*I1kqxy&QPsBjtX^qtTo@uvrdmeZ4mP)HEziNaTK?5j>4_g8YYDY z(BF9!fq|a5vfLF8{Q7#k9PvWI87FT!V!ggQBFo${Y>F#_Ogyke*bNQaPa(R~6FXLV zBY)i~yf^cP)~w?gSaBHMqmQDk)e(Mcj^Jg+UaY=-4EL2zQgg~F8ZqJsWxhX3M~6Gm zmwYQ)S7AZP9;S56*MvO6&58c(p_K>s(mww~q{Y0b!uGw?wZNGk4E3OUu5P54aDooL zI6*38Zv)+OUq>sSttRn(1|)vOfD%f# zQS)gN`ZC;_Y!l4s&rU14yVsV!>~iLP#bKJ!=Sq6DN2o!>o$i+$qg5epdz7OWgelflmqncpfl}?x2F+~hv;dY9Z9_1*hO)?`1a9<-ybiWP3=z9801QyTHVNHk2j6O zDcYBNoL>I(qzCVO$X-8$UT6lcd4FAYJsbW!XKRjrA%b4tivNEG|S|X#pb66+$RC6Y6mph!@L&8Z#-us;PLy`{z2p45)}_ASfyi z2dCs=&G3Butt~;hQ9jnGq$8_57d;El;17F{_G#y%;6pL~o0NyQa#>L3J#)jzY>c-o z!1URtQOI17`tB00h05XhupG{<71+vaceq9=a@tC9vauAEf^ux~F2-L$9{&9*!j~Ue zsG63B;0X4H4M~HKO$t2B3bDGafOCg(u##& zhttzhky41_>_QyWDaC5LQZ%e9LagvvI8QwdNxidht1rSUp+Y3qXW``gTr}(|fYSVY zcJ%FN$wXz_$D(+`pcPrxP-HpBG!8+hCyVVH{KKnz(OV%uUI?2{5c5eRe|W@e$$Y{A@IE$4#_DIxb!;^ zi}ihAmFs~UTK?ER-xJSU{m~TehkuQne=P{a+h|{K00{1V^1)xJAROB3gQNi72R#i$ zzZv(uhQ@LRcoY;aL~%bj3gf58K#}V&E3QR%Z}-RDb-sxD?29_1AS{*Reckau_@4K{ zpemlZFg*}GtT8L(MWJRf?_>5yVl3~mlI6Kqm(24R?lT`S+ZXoSYg9_~MgIdom{kO# zA~_IM7lN@h+ZSRrzDRKoz~SXV5ElvHJ&-?6e&xNHN+3R$CE~3}49=;9!MHyZw+-WP z(feLlr5u7&Au$j>6$a1cp{PPIW;q1GCdeNz_XWdw z0qgCdVepTS!}D{on8vm9@+FDb+RNH}8rQnK=ecqsgL4*Bu;vbbXD{U5dsdm(>+4ZRaOXlao&t{Bs^NmDI`a&` z`}W1aTpjGNU&cE0D&(=3e~sk|NZr#xkIo`U#tD$QeL3DdS&5%P%b@sqDXb?i$1~fd zkWO2Gad$OQ)(hKkE8 z@Kc=-xnm*xelCXMljWShvk0hUrRU!#rPs~6**_hLMnqK*EQ zT8P_0cw@H+v7c2idHDi}N@!s5awS~jt@p+uOORi_9JMmbFePav?wGAcwAo_x<*mox zXB%*Q+(xVw-3YTzX3rH?qifYlX7`qJruJ&+4qJ&QnJbXAb2%2e3t;jP*ypZ=SEd{y ztE~Z3Y+2`Gt) zM^t7QhV2MN;qP!*JqyQ#xFi&D#_YsZDL5Y+heO*k;P*Hdy658fUJ{3%_DG0T2gBx< zA4C+wv8p=~f6^P!8sC=CSgdG zi^s`j={Ui?*v$jHRz*Z(FZb<#GdJ=117APxyvKdTdvKWoWbDnv4)s*lfKuW7FdmDN zlbO#-MU#CNM4B@2&y=|jp$sgWoQAd9%#Fq-!7VZYpLoCj`WwHe>nX7ANrm_36g1cJ zp3Wp0xxZ7eC6V**+LN)=Dh-WhSr}@Tf~UM6J!P7KdHIQ`>`y=>_nad}CE?YlIQBFo zLt{)l2D>s-laK|U%So_b7l$u*;xK*= zDE~D2ib~UNG^_REg3o(w4*rOv_0REhF0*X|&yXSi3OyxHU{LTFO^5GdyTvm^AANQg?J3%$r&--pK^f8lLlKbpS(!jL1sAzA+$ z!ruLuo!f`ehkik*s2~4+_hCp{HwMCcAbQ|Egetr7eRVJTTzirG>LZF5e}MYQ9whzx zh8zAr5j*KOA|-n9ddmm&g?vM%`e(F0@57*vtlhKb{!+Ua_GvvBw&E2WOrPS6(+gNx zy~eDIZ_xPnJ+hvE#_gI<=v&zZUwnaq?iUCz>B5=UpI|h)52Y8sVfBMA%w+ySkkSCE z-VNr_Q2%%)%V0988NgyC*6FwQqsQPYnuR`N`0YO^F8u`~n?ZEcbp#33j-<8sMw89) zQM5W>1XbM_LL(c7k&fh0GN>FvFK!K@p`!zi5C5&~D#k7Z#n9?VW8G~ivHBAZGvpH|VNfA#%WH7cs2I4ZJSll6kqYvj{ z$5v514HCtuDPj=SVupx$8oO9A$p4YXqUTb$R4a+N1(N8@QowxG`G}h=ii`&$oS85W zqyG~_=WGSMT)|$ERtfBGkixMMlFT$pqgrD=I>f~=MN14<_KRS>tt3*tMe$5(HbOe) zBKo>8-!mj}NL~!9rNmJ*Lj-5*=OC3|>x;W2COlI@;Xe)7C8@z8UjcS!rBG?EjDlT~ zP?u1~Ab)vSeHVqM=zW4S6NNT^Yckp`7$Xp-n}En=C8+I1Gt(tlcXZ3Ad*l`>8E|G&M`q;b}oWNWTY zA|h%Ot)xVIS13{06Gc*~l_OhY8QPvMO+!XWlFKD2di7;KS$>nC&ITDWStw0ioJ({5 zx*Y8ZmZuy0ZC)x##s@{I=$a&Tf0U%+3R${j zEkjM;i&RCrC80=*L)7S#vm(!`mM57hnq=ywMclTbr$-ml z#nTHZz-uXaY+lOQLQBcYXdw;%r9p%GR7r=KoOes5NY9Js4%JK2j4VmI`$CceCrFXh zWl6eXB}W|t{24A%q2yRK%CuCY*1;N-DYSs{3e+fynVmB+3iR-cDjm*PNYk&oB5S=f z%DF$$tKonP`*vgVKPUJ`T0`Qn8D!Zz^Kkzj%p86IDgQCQ`^^zOCmb@d32 z2^XEY&tvEeXMr7toV5TBI^vXpJ+GhpaL>&NITLLVnCOI+59|=b{+Ogu2OvMh7B7W3 zT$F6#_1zi^7CRyOBKvU4oDrgB5ARWyhzi>S(;8C@zi5p8Ym8yC(ikopW{8+#4e~ZZ zXRsxNYjz_`&I5p7O5 zxA`Dv7u%q1vLj}@?1e?RBV3~nU~q;bk}HoQ+W$Bftn$JEBS)Orbpi=Wu53DBhWW&9 z%;xi@B;|}9iq0s!=71wx_hZcUgYcPXk7+~f@F>~_yRPp>*Kr37f4&#L6ZRlIijU3L z2`jYrLbcBUZ}s+~Jktip`22OBK7?J0HW=ICgkUj8+-*396(z1XxZV;c);gm=(*~o~ z?uF@u1Ng0Yg#9M_5D>Eu>&EWKY6*UA%Y7J@a0vPj58|T3e%v%Th^A4lcfqq1{>bS|@0aiogVQST9t?RC5{Bj+EmW z?J{_KUj|o=B=O~qETW2JuGCCxPN{F+5o*1Epc|IAt%2#z;9Nm?`7i4n;(0DPXgsBIJ}bupmbTx2~$- z$~{$BRjELGqYBE|;}SDXfDwI~5NlM$D19~jkzI^f-p|y0U(DG(sxX_Nfl-q+F~G+y zcTyd(UzOp$bP+oG3F9jXwgD>4W-Nm1dlht#SHN6JMMN2DVqU2dVssYZ%oU!yyy4^U_7|d9RH_aq2kzPMtpsb%^;a!EoO1EgP%}8m|ta7A3sB zE>3cD#7H|{ica*4(WF0O)Fd{a&d;4sx>~{{+&PQxH_fF3hVx14iv-zxR-mssVkBY6 zbF=H_l8U$xd27t1B7U6Iv01cz_Z*tCZYp)&nM!qe(@1paG%7zoh0I=0qA@y?$VY1u zZQ4AZWZ#aW^a~Rx++-YG5F1U>eWS=fek|SUnn2^m&SdZP9CGWPO;)FbsYP0p>~e%D zNnD&FmWj~|Wl7q0hO^EWiIKd52#xL%p>UUZG}TXpJamNUq5V8co-~ipF`o{E%_YBB zAxb(smrTp&(aV){>1&S=%}|<8@$vIX?xzSTkt7v(%%{TH;#7B0oNhduPv2(GC#QQ- zlzLa3madSbU+(-mw}p8wG#}5q*)+jnI-SxLBHd9!G}B}z&EPzHd-S z7o|-rXH)pN8Dx_-mxMa!&^(8^q;OG~=GlmnzN9EMujY((1!0oOnMacs&8IEl5)>&e zP1AB^NLEUY=Ecg=-|5P9H%EqqkMmqxJ!TNcOVN=>GBm4eKKVTmp|Rg3=#`r&ZCF2# zq>bj%#RbEIfSfEWjeN|E*ld6GLVPqmt=w0gN5jrEbGRddy;BTt=trDTy_ zCyCeqVK|v_?bap>`w(fwJy%7Sp$h(0s={7h1N$2nBK;^~Pn!TQ?bT6sfWNy-7T|Nn zB8ZJp$KQ+WqEOVr)Exqrxm7r)Tn#_OG;sBtB5OICuvw~x_uA}-bymS`djXsjMjA6gjFx*cC z<1AF*EUknc?^O|cRs(9E*>|>46-&0tV)JQLjE$1QPOdivB#xC_n|(Vai>6>%%pWNQ z5f^z#-dDhDffAC9l(GGk9P&RXLDxo^JqU_u(&F`?TY$uDU}pFt43S%mV85l9EUyi< z$xHEK^&(uEtBEd!MR+HwinG(?U{az2pGqlA?Nx?|BIhY-sIk{X6+d?X?_AYUJ$o@K z4At-|RvAz9<)E#o2&*tne3-9^C&v^a#Mij#AD&I&#(rdRKx?@cjIYT<|A#tz;#Co? z!{>8}8ha=O%tR|;H2d$)-q1i;hBrd9{qbVdNpu;z!%ox(eQqAuzxovXnFrJ|@WuTE zPwX{0hE*!Z@!tkdEMLmA6rX$IVVyrN2b_Yfx+l_2PvBIBCzMKj(Cq7lqo2IEpBsdm z>K>TReji6`Z`QjHvR)d5VbXrE(>{R|We*g3__A;R2>ahT>#+C~7T;wrOO!kMc6egb zcxE)ldf-_*|LltoKF{^Stx{$M=7gZ7+ZR8A-SO?XKk|=wU_193$NXWwP|X9L$GqU9 z;fw5QZ_eQLLNE6W1H(NbcG?raO4(Cm7KrE{UJw`a#L4GA=$+#UH$~=9s{L?0HVCtB z`671w3Cv`D<-Dpp?yh#h$;Ixl&|$XW&v9HY@x&Z6ADHubI^^IE;lUp8zTpD__x7~U z`NCrf&uhLO46(!jY@ZQ=BPu~SDHno~7NOY3+PcGCUo6cJM)%EN#0Lap)Wirhmxf~f zxMUf??0vaB0c~c+S3vF3Agc&+|1_t6apfg*+Q&_GN5tX~Z09 z=4&ss!_@mShA(bKVfzid80dg)`85oczJa2E>*&646QYk>aQ|{6TEyER+}MK5r)^lT zcohc0?FeqT2+2>4FdxTkuHr?QbvI)A=4Q;UYJ=>NI%Fu`z%RoK%mH1)iRu<)J2NZ0 ztQD6U+L8CF1?Cl(;Ae0dou{r~MCUcA3g3symn(Sw=OV`3ZN`UPttc(4gL!Nt4pp`w z&!-t4x=napaUN$THo`2R3Fld_o#TEHIvGuPY<~%f)QrBd?eL7bgl`|O;LGO=u*tjt z@k19dvalA_H=2>W>>{SWZ9*XP(kr%KWUuxOs2^=d{s1XcAEJB3U3}($6S>j>k)?Ow_WceFRqnvW@d@0*pTc6` zIX>>|#P*O`>5B>!Y$KA{OhJKb==+_`REI63xp5o$ z$8Mq2@uuYUWG6k&FeA4UW>h@WhIDs3lJQ(SIzC`d9k0x3#|m?*zHUQ}(^#WxHm9-; zHazoX7d<_>m8Q%$rA7YR$V*`>nFJb;qVHz1c)E!$rLH4tF`_*?cT#T8E>hgTlWx4T zCQUsTQXlS49|xbJbxx>m2C7VMqEZup!lz6Ff4Qzb{Fd)`JfeNPRiWflToH6T{5gmVL0NV}kpPi>0$&i7@l3Cc(lR>$=mH9TFQ z2J0~z=;u1QwpbAocNXB$4{cO@&_$%dDxChi3iexlTkH=#cHG9fh`&qesPM%+AHhxu>nu*aayoisZj;DyL>#6)mU}45&5$l(0;QC1Gkv-nNf`=(G57-SC4+_1_+5Y;KjeQco9?q zIp(OEHdLT>^*J=CS7W?tJ=QkWBEF~!=k7IQ%A+Ri{8Edy`R5_Dy9S!d73kpSO^&ID z=HDvRFR4cQ*gDMJ!G0Cl7C0PjLHMET&@8wL%itS$yrhjaDfW~%HetAMJ1m>8p+CG8 zRo2&Ge4`CL`7KDZYQTKs8uphqVCTw|QeZbViD!-`z~)~Z)<>{cH!>M= zp{cA3F(dDrg54X_k(-^25`{z*ug}7NRh*$dG8u2K$3l5~97Jv=;q1dW+}_1=evZbX z>0%`IJdQ;0v@jfyiNl}{o=J8h4yL}bXn321xAuuxcq1Oe-zB2eDFewW%;3c&fy_Ct zyEFwUNy#`nArabEA$W2%gqed#_B1CV<|^ljmd4>feordFk(hNn2Ki#~cx0Kud}%6N zek5Yr=Tz7lq~cFlDn5C$pCW%mI2OpzCiUu8L(~To-$pcSfU7 zJ`U~~u{f<6#|&O7jEBad^G6&qw($(y14#%;jYB$ftA;)~5L=OjeZ~1m_sfH_Stj25 zr$C^VfVkK3&`3*$dvyZFOi0CmV=89oXW+-TJbWnQoDho~3>kbDF`7k)dQyzjZ?aK! zFdY|I1Mb&L!j|~&0+E4lf>Yc62~OAb3RHLW3LYnX7PL=zBxn=AAu#TJDJb#0CwR;= zqL=G;3jB@U3KE512&(;_3Fy#ELG^|Ag1__M3(B0o2$rdT5p<`26pT#&Bv7n*A-H4N zBaq4N7EIalOW?ihqhLyIx8TRi$ASmu?*uO{y%vzmYr!wwH-a=Pl=uLYknDz6cpw+QU zp#A!tpu6#&2p;S9 z3*rlY3tsqq6RaQiQ=t9*qu|-TUxJlg-vw(cdj&6deh?hF&@J$O@LAAzd_XX%>ziQ5 zPJXTRU4s8g-U?jw-wHhMJ{2^cdn(9ld?AoI^+M2G@J29l)jPrDKYs;FH~bQ~_720t zz<&Zcu~E3SXDt5o41?(B(U9mHiIV1F(DNCD8F_yNb6)-8JpB&>Z|7fvW3hh)wS4|e zz1|73kGvKv6MiBPKJZ8&|M;z-|LhCFEWOWym)m~|9_;KAbRK;x@H2TMSoraY;KAm% zg4REu1V1AC1cgC;f}dl*3h3Z3!OfXN5a=`#uG;?uTHRj-HP-zCXRB_(%hpcp{8ry*LHiWRhXImYG#%V)t$@!~&l@9J*hO`Ju&7 z%Fabpe;&+N6rj5)7YjqOA%80mVwX5GN45ZFMfrHTtQZGOiZQyQ9HSI+&@?F*f&25Z zczOYfJql2BJ{S5onNxbo?9bsM+)gfnR7W9h8lAy4hf>TtnT>&$d6>mIYRAMh&WlLr zToh)T;?ppsJr%>svXNbwgJ1lf?H2L%qLt4~4qsadxp>YjYJWopj(k52{rd$7?MjDe zS2isCGkAVt9<-Use(h?&z1asFyCZufIm zQB)~j@cW;`I&N@m1`@4v;r51oL5`gj@S3gQ| z)Sk~9v&{GNPII3xAA`?j^L{V~`y7~Eznz3#ONx-*e+Hi(R-=U1`CmiNqfX;IW*)17 z$E@opac;w=%od1$zQPPmJBIGR0=a?9@U^-K{q#ThA@Lcv8a~7A_$TgR^k8t{C)^f( z3GG$v1;6_qiM`#NKl24sq<$f6$1kYf{0#i%v{}0iILrH$+5bn?YL(rkY)Vlo-9N6m}6VZ)T`?_J_ z*$08?Cy;RuYkZ&J+u4gF9^Y_z^$#rD{R74A?6F?3bz{EFPbimvMZY=wxP`ys{jom?)c=c-HGeoW@hg@t`wl5D_Qv=9h6#J_ zjn;g}K-_0Mc+iFKCA|oi`Gk&+e#~IK_{_dOyczWsQ%4S9c++so6HFv^@v(G#%qY5k zWe_d09!9zB-%ft_3vELEQ2q52Rd)x_G3h4;v+%|G{d=DC37VnLy$hW&jmZ1Nb!ag2ljqMt+@uPoO=saT zr2(emWIT`UM7v}D*?*G(lNa)1_vAxF={v0 zZ$p!?b}0KFXQZLqGXV#rQn}|FhuB|yyoGU4iAzOqWg7RgW3ah33k~zKV9&p-E3$bV z%fN1*bZFS7LFZpAR$Y(5^0X+}I&rSu_-Is2jYph#BKEuR@60p3S5CqPhZy9JiGtp^ zFi?8}h7~8k_G>bBa@NcluA}d9t$cxx+r}{kD&+}~Da*j0M+wN>z}YSt%=siGU^DYg z$Ck0z&Mpl_BNA|%Ys1Z*>@9UnLbX9YzP>GnHTxX4pFM-~Lzz9jorc`bEO;?*HLO34 z`Ts0P&rHXi$L!M^#$MZlX*lhfkK}h{__^dPg5^)+`m92nl+DN1{p=UImV))wsmL0g zi0t5GR6NasRH)EZnZk!k(|Xnjew~t}vQ|-2`!tXAo=6fmp(w;a zfi}G3`7R?gX!LN<#b6ck-l;;1=E#%77Nkf5d_<|iN`w|n5ux@=!j#TBM|F)jRg{YJ3@rO7l{mZmY>8wW6JY}dQLWgjX2NDZB3P+C3mE# ztU`vSm&#JeNzSq0_pn+*i9U*IkV}#hB^%07hOrz?8LL2nY8qtqNSXEpsZhgMWfHrp zNIPfA(CjbF4;0AJBXc=&|0Yf?$K~kf0sb5(sZh!mMaq>`qT2#>+I&QnjPI+^P|myP zlvAe>XEbR2Z=Mh2Y68(v3)IJ0AakxMs=Bw~ocea8#qZ$x24*-ojQtY-wm~@F6l)WV zuy!Hm+L>prD{jJ!az71A)x1w5mGxtpS&&#%;cI{f6nY0COCvU>@t~D6B zV-p4j^wbc%+rmL zxNaLpM4MpMSOfe>G=!XpJ^~MK;Jm`^cr@G=TgM-R#=Mib@8yP6SaEjWPE5RFf!Hy& zh>BzXN3;c&8tFs4Wh?H*Z^E*x>yc2t8MnW$#kxT&@cWAnHcGF>(#N{kHNgOf%8lXW zvK^1@4ACQDjC<|+n6_pe%0gFR^LT!~_eQv#T@PnI#t*CY@$Wxl{HZm8<`qNS{bz~J z^>%3Zy9J{&_*_)4gJY>7p6Bnx+*Cb4dJ}YZ8DN6OMqG&Af@5`@`MY!tp6Tggfx}9; zCM?B$Z^F3Qd@T3YLcC)GMiy)`y=eTOhZ zlDUY?gD4Ym#^-xBPzz&@VG463O50KU-W>n*ouLzFji;9mAwt^?_xqfAO?Ab?Xb?Gs>gC5E$6lJSTZfTVhCeAI=bM77-#PuaKUB!lQ^kx6uk<0 z%W=m$Nl(sibcMl8ADEV(Kr|ggZS+YLN4R4DQUnwTMHgvb#cks44)QRLTt|#I3C=DfHlOy9y7!xnRVa`XYgL6 zg4|Rk#63{OPqJLC)#gCxy)x=_S7YC;)foM#21Ppt4=FWV>tM*dNwew)tI>W(4MuIN zP?DyGFQ1igd@IBGshYU*ehorKl=0*_^|r@>=1KY}?o-3ZA~o#hD5J7k34?UXryf;> zt`y+1RtK-wt7EDoY32vlU_-PPPOMyqy5Ab`4l=>~6jKy#(nLV47W7tXBVU&xyoTOG zUlTidbuja*0g8Rhk(Fc)#}s2EbQ@sHJX2^^>EM;?T4WqDz`7&@;>iHP$5tWMlyG0S zRk59^;^rkaxCgHVd#;A;lj`U>rh=|+H4JZJP%&9UbF#HKp=^N54%#^GstwPnhG6H- z;53s14bsr3h8y4ozYL<4SD<*ta?~fw;|TToW5;Eoxo8DMO{HKOAPxRHY0TnYhJ*w; z@*qoKWXfC!KajwBt0mxjBo4ph^N{jkF%Ejp!7OJ9;=0bqz?Y@y$y|Zxye0TECI#;| zOYm>GB$S>lf%vPXxLvXke+Cz$AZHOyKa{|v`1vqvUx-Au810ZptPSZtUrNIcOL1#J zf*3h6o$?_#{vUWj}C;&3@84*7$#z!xBfv9Ou= z{B!}jMqwY`-T?dSLFhX z-(HIP3ev&S|J9EYM~37Qd@`ob)u9OE5lM)iTnep1;)%~$KyyC{{5rD)ziO7?+w3J^ z*2F^JFN1Vjc^op6K`Ut6=?Y=g}QDj7|oG_)M+VPcq?w(t7gI zYeDpYDOQG9Vg0MMlqai+d(#c^ZL=2iW%W=(HN*BvdRSVe4Sh;G8LeE8HYr2!-rIoi z=JmJ~ZG;P@AiPCvh4yqXtAb zYG4L^#^Wa#-ptp)pW|B4{Iwbbrx{-F)5YaBE#x=pqcny56~Aoc{5Lf@kdhE_2uL++a0(#eDO9^p=+YE5O*bHkw z8bW`T9#jwOV`z#wX8hAb`1SP|IcS2R^3ABz{{Q{X9BMnMW|5-jSUY)Aimj1E8NOdk zt+8F-3SInmcrQa7e|q-MpRyjyo|q#4r4hcRn&IhGQ_P!6-jCHr(0puwXJ$t5`>KU! z`36|JOcxe|ns^|qfl<<-$?XWhhI7QjnBs-jsvszQ^Tyawe^gKLf%IlC^wHP+Tn}7U zazSI+aj2UfLfRlgrRl~4WBm&{b!VoABi9b|dwAck9GAaPo5rL3w3cx}KKNxsWcGwJm zVj1|LVXGfrPw~U-0Wa8jxkHM!(_YL6&H88Yc*j}fhn()O`Qqc`03@yQf$f+# zw&(jm<%BPBI0B*7P8t}eK;*vgga@BHqBnXIt2Pj6Uwje%!UrA4g7BRj3nQY!5U>}+ z`u$>PF`kX<4HCHNIgeNi#MB%U#RrX5s{Of>uz z!`!Y}u$wU%2ka+f&F~b2-xt7G+ca2h5QfIURJ5(0MxGd9Jm8*2d52RWzHk~WUQEXO z#{#t07s1nYlaQq#0x9|Fl&3HQs}~7lBz`7N91zC+uXFJ4>>MoW6oYg40`%>jjT1{2 zVf^$$4Eso+_~Qc9kj8(CoFryh$|5gz9##-v)cl?(^pa+yICcgS{bxW#Zw_U@QC69X z1X@$XV1Hv4{%)FwR!@4i^v{Kj!%UoNpMi-IVd`JaK$nOx{+ynU5pQ86Et`+z22t3b zpNZzD!bsUDPWOy?i0+t;&v(UfU|}O%u3sFonnvK$*;;{Wa8|eYFAT1z5xzVB+ zPZUS#Ju$q@nTOLimf((~1V%l@5%qf(($Z%`{jxaaFw7vY_e^knXCv|cY~sGnhVv$I zw5^zjw@*c}e`qdd(X(Rt{Kb$vxe(K1#o;??E?$?+MsLt;?7AUF^^hp!iWY#$&BqPN zc`%Tei(=UsbiK?%is3Bi-Vukx`HHZe`jrcT7kRix zER`OsJg6Kg!G{wi7}YDnXZ|wGs4qv1O(8a3$wu#33H6!_G5bO(_(lq_Lo6Sk`ikLb zR|L6@nMga6kB3*1;rEd|U?YjpxfX*jv2kc6-dc-mB;|t=Kj=K^x=w`S!B7aQCy1fe z5`{UtLh(8!8u?0L)Gv;JAI${62S>q{KL!I+<1qhwEDD_yVH6vK-@D@AyD1WTSTz3i zrQq=06v%8(p`4dQjE-awZzLTLbuw_iG==Ubagc9I#2Ke}2+`m4CU4vG&_rU6#lnY} zxN;-WSTPulzwR{OJCca!KCw`oOB~_Bbcig;LZ(m#B3>mTjF>PiA_;}>DJyeF5yeB7FbsLouNx1Ynjg=rY;%fz5p78-a` zP*IVC&%`t_^C#~XPX-)`GaRxq6F21OJtt=(up<*Hf;ni~mkHO!X%M}W2+N0A#C1r= zIScaM?azi=R0cxhvM_Hn7jgAD7~5L_b&(=EKMJw#6tR7*3Sc#*1o!V2Ler=a+ti6S z=2Q&9lacrl8H1%g@o?Nr$EG+D%EV_W+!=)$@o0>F41v?TV7MiP;Xy+vCXIyQ>ydEO zehft|y{@aSj>CQO(cPQKplhCH#N=tU`jUh>uZWj-DiUSWlJVtNBzC`|@dySxhtG z1u>APW0Fb7r$>k~BxXm$>|g{iJshix;t(em2`ygY0||s9x-S}nG`rsaED^2K(=na= zgfTkg10IPYtwJP9&&J^3n-F?F1mRb01mt<6AY2`RBGSnv5H~3xI2`ZPqHtt?5Tb4b z;rg6#JhBPFb@DP78pOc+IrRg5;vqW`k9g+h$CiGG<4%5X{UsH<)ciL^%j7o5i>?K6ZS@V$d{$_p)v_F z^AoU&uBD5Tq&<{QKr!tXlV@QF7K(v_MKsRd)q>A8BTS!Ij~-8BED2Z-6Ur}`oVkv! zF+(`K)P&w_%G)(3JmKm;X+lX`PtReKp9LvvHV8Ny>FqPZ{ zx!jG|*1HQM+;#}MVT$osb7)mtV2-^xe#V-?VYUhCNHgQvYl>UhRv0~KO?+`<9C&I( zJsN9-bMM5<+6{0zv>ur~#`rUZvZ;4iqW_{1isPy)}N<8zIcX5(_Gf z(Nv~|_fZDuxvhaOeQPnZi{V>?J_;Wi<9C)JWEs8h93yl%n*rg5bp4s&*&kgLZ(4^x zMtZPlTZ8Yb^wD9#$ltGrw{MKGm)5Ip9mWWDw1i`<0lG6yp+IZdrXvO@8C?(K4ijv- zV21uxCiuS91V7!ZvDw$0vk#Q_fY$VlsEam*%r|XJUtx&OUL7oOF~rlGM#PUX!q^=X z$ch`_Z}TN+rFLV3(s`Va?x*~tF5JsJkJII?IClI3^tn6mKC2ZMy;^Zi@;n;rNbh4t zS|P(ra31R?f5Ro%eILN2n0_qIy8@Ag{rK|lGCe;oBG9=Xb1z&(h|WcbMfKnb>3g=& z%u)YVJ3feYz_7mw#|7JAFue}jy_;d}SWjBwcKpa`!VjTl{CL`mUx|$hSE;b|n7I9#sijZkii-ekLjAd1V=T0@GcGY5JkT?^huW)qq*P6!jA3w)0X_tdOAZzeb@wD+&#tBar$l0#7I#z&1G^S#pW+VDU)Mq?{dt z6sUW}f#+{39?Pa9ST_d!Z)2cVNzVfiuZuLvTQgGe&n^W^o+sn?H{xr$XCRv9Eh$^0 zsZSUUF1}c7TAF~+RY_19OM^Z=JF|n+u!VfhlT?y1)jI{2#5Nlqq?ysTcubI2yqxNb z0{K`tcP2sZcoL4r$KrDY%~{%`vHMyABzzL64@w?yss}CvW`MgZ1#=Bar@L^lRtkb7kLw; znUJNq&jaG4SX5!tKq)GBmgC*xN^oYBLEgCln?DlUWOokbOQe#&l;%E5lHmB0bP#-b zkdz@-$m?B{AGI6n&+Uij#BMB5-iw5jJJB0<7{Ln;p|QgOyI(k>pmsYf?;e1b#9qjM zu|??njflvK{(#t18a)|xTJXsiQkVwPnEJo791nL?{T~qJ&A412iIP?Bjw8( z_)pOKb+#7_TMr@OOf2=1;J_17wUcfP-jj!*xGUwvAI4pC>isS|jw<1!aP2;V3ird1e|89CwnuR>_7Gm}cEGRpqlnNv zf@4MpG2QL}4yGO$k^I&JY`QzR_j!P?@hp7ToJ32? zDdNkypv&qsQWXxuId4A%_U^%Dv18aY?o51RnqD}x>B zZHf%aktSCLhh=4Gi>QFuiZbk8S^+7q3hMQj;iXVHaSMuZ^mz#+ohkvh3fxIAh0e!n zd~Ytpw1{e~dr%6kni6!MCMF#Hete`DFB1zCr4t;bQBI$~khK=^PO?j0XjFRIefSaRZ9!9&P-s5htIwQ}$LD=B#PMmE11Op3(wW z#RjZ=(*zgN*?wI{`iZ&EaP zQ--Z{4D?mY=sYdN->W4!5>MPiQ{wfFRbr^Hf$rP&5LKyzgMB@gx-}rfqXlv~^*EH+ zfQz=ZC{k)hzH$jFc9mj}a}g$z3z6eOJapYsT+1Pj(#;}#x>o~_h7#CW79(e(9I9JO zkx*EGDYJ_3}!=qat``)xh7n6k;=qv9Y-nynRIw+gXHd&3X8ll8YSDqw(x7hBa?F%2pS{=vE=t z)Bm3!R$!NZ35LIBW36N!Vk9cSb+QOQ)5=i$r2va8Nz*JyJSd-hEJ?`0zBM_xbgdMV zcZW{>@JO#A)4m$0}^Few5A9J3B{<=E=0XSHR^mz(Ydh_%ZDn7@mxv_bXuo< zFG235GI&p}!mT~TYQ9~C2lt85dMFQ7Ym4!W*53M_`Pdq%238%iNVb^2wGVQ7veI;QqA~`2i4*VVaCvrf4pv>}9G!K8lU9C(QzdzUbF89=v$CO^Go81Mv#PL# zb5pgAGgRKhQQh3mvHaT3$*=C_@H=*JLLLrs{&@~@B&7c1s4Cs!NNv5uDZDbs*)q{b z;m8j;^M-D6zRbSOIoo`fbLq=t&JzB+oY9U49KMiYPRq<|obcFwj^IQu=fAVVoF`2; zIdk9K<#5k?&XM!F$5}P+KBqGO3a9PnFz4>&VNT@1tDOA(H#lM`Bb?a-!<>-6Bb+5F zS2!wm_c*qxcRAz1w>kVV!yM}mL!9ERgPhhCS2*+UU*W8weTg;KIhKJ}I0xEqb0*9G z$0^x=hf^%~m~&0zKhDkHLmcl-|8X)MZ*$zj?r>IBU+36Ay23f^(8CEByTS45xXkJA z?c)Tfb#oLehBy!2b#i35x;RHhDmj~sIysM1~{`$U*=?54|2rSE^)s6 zxXyVHdXw|=;bV^7#pfK)xHp{8D-SqR3KS4gC7rG;MQvfz{!jR@02+zHtu-F)e z8^V?X6{jh~W( zQJxwSwN)@uu7tIF)DWhwfJuG|FqBls?lMKJmr}x$bL9P%S_{W-G`I9pMbXdISX90m zYWFqaYN(2neSmVKGDZj1;JyDUNV};JJ4OrN7IUEXP!)MJ_uOx)gjK03h!Rx6^kWM6 z`dI~y`&NR-SQ+NY@_6=}p`X0Exl0wXVkd)*_8Qoysbc;=U5Iqj40f>!?8xJ@$#OY@ zpDAJA{^dw`C6B@tDyZA7M#pd^c%xOIzhxDEx&ui^*1$uZgY(xFh!aL0p7TmjkynK? zH}P#MRgwLbcm>7EP`kDg2aYR2lYC2i{*sS*$0~>z0*y=`i??ddv`Xe3d#_BBq8af{(64mi=m&XMp^==ePh!01zdmPN_sbA+AiC;Xycz7)o^^3#c zxH1f4nGtaE3xmxwI;VaHV(mZ>qOAguNVTy4$v`|<=u7-NfAqHc!z{uV%j!HZ(dr4m zFMfDy>W#nu+)=R84Z&vaa2Y;$DSk~*v0$6QIWI?bbT)7 z4JOb1FbnhXyFe6yA}m6kx={=h#)3dBt)vhq1F(NRR^Q+P9_qQbBI~~Dj0KS z()>m`3_HGtB3vl|&o}tthp-=pv?*gRz!$6eeIe@*3|-ouLn7oc?()HoImGQtq0A1N zyL_$-fM8}I-aR0HaRX(igva2xRwSglVi5E!8j+%;tyCZnNk%e`A5Xy|GwM%2N zG59V5r)I@s3H5Y0tc!q}YXlbBN6`E&5>sQs;5(1*b7z9!Hy#BAS<3NSvXR}@+Qc}K zHmq^-4i-J#o&|bsWb&&vvE(Eh#?M9Ip*v{tQpe{T+gOWS|pqWpkbMnKW%zo1+QS zOgCoY8x5I4tpSTH&}Gj2`pm&oi}f1lvBlHZF^wpF_I&F)b~nn9ZF*puCyil zF>br%8vD3-^Qw@?O@X9ZCK$VYtk@oWKZs`XNM{_F!eMGwwTX~nRZyS zA1;<`(tLAfMEmODVrzDP%Ldl8Y6I&}wPEF{)=cZ54O_I;mQ_ocv&);;vo8Y{>~H=C zmOphH6V=(leoJqnOsXwRl=3bVbnMtA^3ysy*|6*<+gT9Djy+MaW3O)6GAF&QY|~2{ z#;vx69f`MMjc<+Escc=Q$F0TEcWX1njJ533X>G;VSE{O z>{Im^^{YSPRN5$H2Y*wZ=R0f{dWY{nUlTj^Ir#=2!fncJ{K&im+vaD8{B##VhhAVL z=Q*5DK0&ztE3DY_3;~6&u_o^o^`xI5WA$?==DmQy)VJ7c{{}}XU+rzfD8^bpp;zS# z0usiMZ1W8(_kKbE{Y>H32bxX1gZhJ)sN4Jm9STozm-7s^9sl9njk}m-dLOs$-^Y`g zw~ODsZyzaJ;tRTuzQSz0#jWRWAZ_>(zii&%?83Lu9Dae1?eFj~`YmGk-(wTkCnWHE z#5spg_}uyyX%b_Yapnba+Fm0-`xbnQAK>bhTaa1y0Kt!*BF*d-4#Yl0?DxC)d+a9C zqwm7{#0c)&Ud3ULVJMVb1(Ugn?!fD~r}r2iQyx<$(rXx>`GBg1w>Vbu5?otfqi^6f zbhN)fOy&(@Oh+-a`7_w>5Ab;K29@1!;j8-sZ$?Lv_~<1ZpFIZmkw;iH`yrG!J;CtV zM^JHi1YMn1czydRj^=!XnJi_|Re!{|@keY8|BNSsW3Ubz#jVFb;BZ%zO}#sV75x=t z6(a&{#E^Ui-CRtxk%ujv!Na7jej@VX2TTZjz@4R^adp~9OdB7A)6x&P_4PaUIPx(4 zOdhu3Jr6sU#m}OS@w0aGNo;NaKNFyT|1>2YU91qZUBt(htq@}KX9~0Z=F^yz`>HHrl&ofWe?3@ z4=0PUL}J28Rv4}KxfaRN0PKOsfq8y;wUz=qV1(4_skW9$Z`AcrL^8y!paF&bZS^dGT9b7EuDdq7k;bqgW z@iM+`{OqKnAe(N?&u*RJXF{6=*dBjAw$Yr6weI0(f2;V|L=-QZ$1lJ23@v)b^XJgaYr8*wr3K(ftUshq_>eEz2hlCwh?n-nJzqbZ zG$J0nzZXi*xMTi$ANa}AbDwGgYg!9r^M>M2Tp*OP127Wdk9n0|h!$~$zn%y79PoyV zzBlxm-Km#&8jnZ3q3!BVdck1Qe^AC1)zVMOqwrsG1kTbN_)-Mrei=l?g#FRet4?skH1>PD#`JINHEnL#F2kW`N)fY zdm{J58GLgmea;kbT%m2;cqcqLMw!r zyNDp$CBx6eFHT{1FV18ogCcBG_$1cOrn3vdQ<&^>I@jplabIOJYj`ZgTw3{9v<)AN zP~>Oda|BpH-6S?Eou8>I@UbtgJZz>M4-*&vjZimkX7)>f9Wvu#PRn_iOEeEt&gWu| zj=Zey05_XNo-MVzd@SIy0GlJp&x&vIv%$B#EXtXe4RvxcwU=DXxb`2EI>#w*Z5+2P zevleq9#xKv$TrLuSleWWbCJ!?S<6*Dp z9(iZJAbb6apAC;rVU2wvtnJ(s#??KI)y|$m`(hHia!ru6>+`Wpx`w~55oA{b1lZq= zf=q>%m+4grvNPNH87t&w4{FE3Ruhl&&KK-{HjV+!zc?ns&BCYivA5iOY-BAz+n6WB zj`vStq7wqFBvydgYD{7ay9C(WJ{~r8HxHX{&dVxd1(}2)KXbl6iG8aVVn^o+uwN(n z81D^kcCg_uxG4YgV!}AS-TjMEx<9^qSAqxblxtb zq`AhNeEfP>fY=9_upP|BDxX}K{>ee_cnSWgl#oWG06R}-A+?oySxV<9gOB>H+X^8% zkNi|6q$wdiz=67KB&^JUZC3_V%Cbj# z5ALmDIX>RYgRMsyF^Wr|>`;P!=Mu_aD8^t}5fmr#An}iKewO4T<8VHD`)O8_R)VD$ zbD&+Hi}{w7c-&tJ;q7HuNj+*&H}V-#FLUKp@;A|TDbCM9#hY9#ep-UxJ1X$-eGP7U zRpMwdvF|0x?{%*c>1SHuzqlRYndG;cPo6K)Zfu=sgH~HR-1l_gZB-*?O4UPcQ4_Iv zEATI~mNd=PsLr8&xLpe(sV{%HsR3)AcVgD5Rt&ss$NJO`T<5AoI=x@tnY7AYg$B`ASn8(+S#t*HyZ%=&uOBGN)Q;F7t ztw{B#!yYr5VO}q#jPU{_hnK=Hz7%>hOVE{0-d36|NjKMHCJ!;|U)I9vKqWk!8qhC6 zGpwz2Y*edoUZV_Q4=d0|y-nMrv>$eq!#b{lydQ-)`Jo7lysGiDrVxKtR6}}x18UFK zQhs3tHj~fq125@Nc2(kEZWjLguM(0w3h}$W0QYwlpvk5LPTR;M=v)kQU)m0%QfPXW z;`eYdVrS(;!mt#3bBeH#{IHMwE5WT*4Tup3pS-wcvzzc+zYZ5Bw?VVN8P!!y@c&&6 zZ&r)DldACOSw6T)kHb&rx8KzaXqRNcC7RCbZ+WnH$cEkVTom$DV#mB1^iHe8DE0T< zzgFU2c^yKF+i=gZ2{m+#ul=coDb4iqQu5I?vxv5xJX)pHPp`|vfi35-At8hKnpyZm z=koJKl(iUCgpB27_(nZwjh)5#{I(cFpGzRooP&n{YH-@53U6Li<9Ss%9?<@rX+U$A zo0JtA*MftI7ck{t7gBkuQ0&r*l(shXmYj$1zAjALG>C@d=dr(|70dKGD2t&3m9tu) zZP|)mp*p+_uER8)dc4rChHYafV&+nIYK#UBOgF&xiM6mv(!$F%W>8dFhei=IM2+iE z_L~lp^>lG?n+|T=(SL7wo3o`{Yu(^$qM@AR#$Fy)QN*Dj~bg@rX7qj=7qSj;`f_ARK zB5@sPkEi?2g^zD>&&Gfb-Q9Ze6-*oYzWTLt~zg+)$wxU<9- zDi!3HbVImp_kLgEy&uTN4}o-Ipr*n0V-C(`6yicoHX4hvF!yo+ zW*kaG%UbeS9!^H&y?C7Jk0(Y;3QliLz;Z6q*(D`|S0e@1OHyH$MXd74IVdIOM}SEh zdT8dZO016RfHYiHOUCX|$^eZ>$B(*H9QP#s!KM^!N=Zb;nG|Alr$cmlDzc8G;9!3S zR;|e-t$8vMJL&I>X%PhaLWuFGs!I zx0Ka3Eja+(-@p{j)K+WXas4*BdGq`!NTieE{o$bXVXv&01G?W3Ns9C6liqR2WK2`dF+)mOy8 zay9jtoFi~zH}z)plOg7p1gkk|C?XAmBlUP^?2f0*kYu!TCE;^Z26V=$ALN{ZR}#c@ ze~^YgHS&$xWTCn39L<%Iv9XMF6bH!%dp#XnU*tpdVmdT?h{@thIz0N@33`-s<`ap> zD`Rl+TQs+psc%7`%}<~NTmJ?#*O#UNW2$@7Glyl?v2Ia zM`@7BrO#QN4b#1;;3gfLr+E^bS4QE(uPAV=<8k>-IMTx7;1H7t`L%I)R2~Nr#VDMJ zjK==i5!g$N{&SBgOG7phnz7ON>6!@>nKa@mW#U{@I+8lmq0E&@n#fcv{g*_#_7xl};g7Ks<0!M5pn^!jl_RgfC%#1?S0m?Dv55+XI5IEA@ zAaft_^bUvNQBwkZXUDg=P4IXC9E`y zkFIAaaJrX_maj>;BYY0rCsScNnt+L%RALXGqujxGxCup(S1$>5>EyL`Ov6iy6fAV2 z41{fQXquOV!_lPkI7c}GZ*yQCMfnJN*?3R6bz+qn2y3SEz&i~RvYF`VCicx>Azn~T zHfEHKn2<78|b(o8uhK`@M>BEHD>uY5OnT$b-|CBHX-2b)N_^ zG)`s1zB!Bh7+E;#O1TfiS>WjAVreba3W0ezt#OWWtV)s6Q-Q5Bn&4zs1NWQcl|R&g zOKx>|CQI{lt~QiUYecOEJr~cnV4#5*f3A&~^`a3PlNz9G(u_+>8(}8V0Pj4S%L~22ISt_Au7}2|YFG+1(tN)f1NZ3p{+DX6J5>4esr1!aRE3CS}#aQJpjk znsqQ&sYEW#^jBC?O=4VuZ{MqN`&|*XT2lS}vkW~iYbaNlGU!NOu_&>IGP=s~wZ8&o zisguoE`709e^Xy!%dwZoP=| z&Ss4H{)1^;@)I+`APcoJK4?z6_yHi6uX81%7hNBEnN1Ed?tvw{j&; z($8f#tb%&!DwvEZA?KJ1ltww2S|NuGHcEK5N)fkyDp8+G9!vMjLOVqsXK${;T`?Vl^RSm0NGAzzkryM6`7-%UFOJOy(#jl1IvC`J6twgW=3M^Z!h~1u&_(1RV zg5FEUZaKn?l_2P?fFHBeVdJR*yG?7TPq`Lz&#B->%}Vr1u7tIwG8}g)BXji{*y`(I zt2)iDwy(j)fz@6Wn8m0d*hLj7Ldy8@NDWzk)Zkaf!MC|;$UjXw z)gSVhv6g00#Kb+7w+gEdkfi1{!{_U2XxpuY&=@V4CFrJO;p z@#SE|S_PTOs(5Fp0+|XG{CcWJdNgHhra9Z$mnyh7pblx5RSrN55^X~@X)a}!5`)v2_!9zcc(J|}=3gi`@kcY} zUu}kSa}&PO_y79Zp~mTg^NkMpmUJQZ`UQvyU7$HHX%_=~VI^|`pJe)QYT5%B#-1qOorHvh^`__)dGh1=@Vh1)JX~Blgjp*3hhEmIRw8nHo zy^RU$ln zur?ZiO))bC4`;s z#T2t))ZZCFN!ATq^ccVrC(?DUI8Q$Lt9T%K6&Gk*B!dRf`l1g$OM9`B{(YvEK3`)W zM$)@bDcXr)=}vI%zX;oXow(`Hg{PbgIQ5mdtHg{Gx!DT0&z%U>>_(R>?VG_4eAn+l zJMjSj>*z+k^Z+Cp&tt{E0ep`jEotEu_*4$ljCcTHHG>dNxCAf%3n--hGVyx=zv(kO z5r5?!9UpO?Za6>eMUwL+eCHcP@v43d*7qQG^F>&Rcf(*|5At{iP*%}}z%y+yH$RV0 zD=%Z&Ps-C2YlF=NI)@ZVud}BXX@^?zNQW3Jwg1o8B$f)N2jkT(u=l-)dbtj`-Ri{g z{0q1{=OXTucEituazKAxg5+2avOCV>>)rvJf8LKghhg-+xdQP!S77(~5;ktWglV?D z#5p3);Ee(N*f|93xC!60!*~>UgEBKmptA2jybvCy%)84NIz9mTxIXk8y$HT(y;vsP zMp>A(cxh9I+U+eUPHDobu}c`V>cc_%UbG$PhMRRS1YTZ5DsgCjgb$;8-xbVSat+*u zbi5Ds5kKw{(kEZV{lX#S-MNY-sr^vsCVox9WpH2UL4x&VG+rD;dSo~8sCqEZzlCyc zJ779egGGD|m_{05OS&ecE_YG}Q#XVnd+~492!{6e!?dy=qo1y$CG0Z37F|Qcwh`1l zx&(!*7h$d21!dPWSU@w=&z5e`sP)3v)YHV9JPmEG(`Z`j0ri(h;k5Pu1m_+>Y z4F-|Ua1wUG1wl8|sGo$k)d^x@9Y+1byog+{u@57_+<&qvY`c=(+AjWyw*5h8)KJ zV`p&dvM+k!&saX=1x71 zKc>e}Sa1S+lTM(z`v4}V?1dudD7J_kMCDm~2x~iGTiGEb?K(m0^i#Nd@-&_=qm1(d zp2%Kv3XdnbBjdRrJkOj(bO?D9+mDm3;RsYxj$lixGeVxYL3*7l>hydNP9CgE#SpwG zil%2m2%@c6QR$0bZ&%Wu_+Z*OA7VYZ!&|~1zQn~2=JJB(_!)R#J%fs1 z;t4$Q#Na8~l!zf3+%?~ekQZGBv86zF1Q61=w866&QBgXg8Lm%|H z`OtCo0qYJzS86a$x%nf6-yKfn{&+ph4>QTP7(g6@>7l-eO`ts0hyKJy^GANY2VJXv z$Wrvfp#|RbTnmN0UNE9gkdN}QFMi6Np$ywI=#&bOw_Fgt-v{r70+5{@jLA0x5&Jh7d$tGCyn*=UWq#8XF&6==Fm`g(ntr`M`uU4v$mZ5!dB{eDZnT z$p}QwNC3vF{Bg6x7n8aCkZI`)6*qSbE%ZmFY!IzYy|L^^0OV;#(qiHZ!%ZI0q4`bq zb06Xo1mhU(OLfw0NeO$Sp4eakx_s<KK>&MD<#3Sr}j?c!qW?Qe`J{YG;<7n8~3 zW$&^Cn1d|MPq?SB0-B42m5VTCKN0r!;#Br&jWA0Rn#w+hOk;O7rm}m+lUY9P=fNj} zZ24CKW`0(PDO}-YXO{`E%dUc~gXR|!zWglh3miG}TT$AQ8ha&{o zwLk$DLo?JcV!HjQ;b%rBeC#;a6lQDo4gaqTECoKQC!ZRw@Wdeuglqk>&uwS z8(DVKRhBK!m1Y;TB$-31Boi2wVHLb`Ol#6orei0^^um>y0N-S1|d86>LmJmK|6w$@Jeyu|Essn8qD>cIuWAbGo9$L{F(QjyUCHyQ-lxjLUsk6*|h zEndw0Cl)ac*#*p(M}jg(=CiS7GVEoA0=p}r#I*Gl*ytn`c6+`O`_rk&hF&N#!}lv$ zZ<`GJJ4KQ$3?g2YXr@Gv~x*?7fi!JAIg#PJyf0gpDc_o5EpIW-9FG2~`#mq{QBxRAoDcmDr+jMRq`j zGQMoo*zgNA)?uZ>c0@Dw^{XZmhA#7!tOgfpgKLWGuwwTGjIAXteQ*o5`E(%LryV}} zwb)>H9!Zy4Q0+tO>wV-~Dr`jkuS%q*H$tYc9&7%z;)-TFyxckqrH%W>ryNy^J_AxiBW4C_l(AJhdEc8ufVnA7##eD#GZJN*J0{ z<1%>=wmVbL{TDIK%oa9n!&}}hP;_A*c#e^k7jMK8*aqFcJhVMI-_k%9Yjm3 zAj@Ber>Cgj{pmamBbyU~EcS7pRIEW~&5N(9jS$&2Pi z%72;>c#eE&ZO!;N-hzJeqrEk%#}8Vwte;g6RnZ3U?X83DjAnSrHe(WNLce1Z5*D=K z(fN9KO>f1jx^D6cw^I)4Mg0ENME&zRoN=Ju1+DuRd@6x1_29d-DsXeUkg}BgIi2~MwZ$)i5sX^OCb#CTF1x)5NL8SR27VIIe=j%?a znAwdB&CNK@SApb&IuzMa-eB^1T^7YazNcda)ZPI_bgK1YnC?M9p9USziZzeaM!yM`0kzMk}d z;>Ypjsu#NSPC_)=8+N5W|Kseu zAv3!&N>M5_jQcuIQK^Xb-h0tdY0u~UJ%2ubK3*>`(d~9#*FCP|JdV%j{lR(54cK*E zFp@+abLY)+M4w%P!~h?h=^uzEJ^e9x%No>uUxV-60?^b?WTv-Ug1lv+*jE;b_KhJZ zxU&}3A4SGh+zPDgt0f;$GApxbH_*4*2SOzSWlb@Id0~ z69$Xxw7RJKI=igE#bv&@(=0N}My$Z-tmSxM6^O#fARIWo2v&PGBKD@pT~7_cU~!v! zV|}ooWgu=2+JOAO>#@@~s!?xmP&U)tTydK4WgK@|t3=bA~hTzcw(eqg+GIA~pzP`Nz?-ml`%j~dij6J56JEPOJ5pWiH zHF@J5VAGNic0z$y?ZkX0+6p=8R@n5z93$rrz=b{oFy@~bwmr2%>z@itn4}Qd($+X; zY6IV4_E<1#2+~Y!5r3{f9vFxWEFDV>5`80+c>_Qp*4X-S04}u^+Tx#9kaQHVH&kG| z=#g}};{?ZYM~tpz6ovcydhXJ*a|H-+d}i@FiZ)yMs>+h$olr!d(sZNPwhaG zVBd)2Gh1va@)0a?X~IC5mkhwvIxDD!TH>zb0NiM2jw4t4VfsW1WQ`dpLX^$Xa=bZ? z`wqawT#e9i_QOYN{}wFB0R9@5&wqwz;= z6x2t#ATD$)*4s|P!>r+0b;AL3o;kr|ipcn#FBox+gtKoYyp)}>({nrqR*LzUg0S$B zgr={A)?|l8Z6)-*C)hXJZQ%HD7}#}!p)8@U^-wHnut!iwC-gt*fVulTu;ZyOW@YGWiVV4w6^oF)3%gVSaE4G%qYb?GF!hS>c6C_r3An+XGKW%tv_Te9=Rh zhtkn*7|=WorjuOo!)hY>+?a`PGiGB?)&e}4wis_0EW*pN%hB{{F`5;A*#6rWht7H7 z{jd3AW;z$U?+I31+gY$5HWO;DK3LS&Zg3MmAa7iF<&HN^bD)^G5KCVzKthW-IM`+;nu2Gc zmx?Q-+Uf9pz6g5-OYXhZ5;Sh{#mO{ZBn=6Kwf<_+0}4RRf;G@_S&rS}|DOI?gQpD( zaVOSO=u}qWg{eQBa#v#bxE08|?hD_=^D#}ViO*Ri@^Y5oy5<_;`CW^i!vweHt~g%* z)%cIE623*T|JGe0vPl=Arsq;Tf9Q^J9*cywdBfyoRmBgs|(JAc0!;-XN-;O1n<(0 zA``j;%#6CBRKFV>-FsraSx+3f*#jHaYvE>fCvIw$ctU7gM+db)(_Xj2UlGzY3_ydBLuHY(iG>p8_Y6#V2ipY5;Zl@ zZh2Qseb52B+jYd4d;h^BRui8*G;!~r4h(;5!M>_H=8V;b)54yx2-L++Cv6tIjlIOt=iWe@an=pnKtJL1aKj-sE|3Hus4ATPWd!r!T5UB50!`Pm)* zH+rDW4IKoG?jh#Y+6c1JhFJ@J3_q&_$47#PqoNJ#Sbf}@qm8x!TG*|nDg0V`xZX+& zotAZj+DUB;^wx!TbytiZ&=m*9c7#T+9!Nji1uOit1)Hb`+IP~%#_^g6o?#&Jp!*>! z-w;;y;{F#3_ShODBx@SLG`TnK?(U5Xl}5Ok(+ds>ny6^q2d}p2W9ljcNXzwwKiLSI zoD9&gpcif$_C>3$dWgEv3p>*KVn%yI9AD4}GxixG;F^ih-y5UZ-53G=`@_b_6c;-6 z#l!%?a?3Qv$EaSyo6`%|@91FoIt87-3sSWaY}aE%o^OUw+hHOjZkGw2z15$%i3yda znbHAKgMMllKt0!(k-MJ>?Nl+Q&PGPmtA{Z){OC;;L|2MiajF zqbt`<$nQ#DvL0u-gQA`5VNkQ1sx zB=|laf;+9PP&-}{*AFqL_@uxBk!v~gjXnBx86h%1g=KG;I%fIX7uOK`~-KmRcV^rIcN1)l{mzgI?NDGf3{w|6fZw&pu=PU)(|kA< zIE_WR!vviE<$`0Y$783^z}TFf2-V4s*dFIBdbOi5Ji-}%-vkfxoE^^S+9O8K0g2@j z${#ww*mNxJPZ$o>We&gzv7UBuCmDjNd(-WVk;}CDX6@EfT zTF^`6tVG3QRCNM=#K&WBl*qCZ45u3}6VUxwBpm)lpwKrO%rhFPj!}5_Is!9VMdIB5 zcxyYwyuEj%(5MN$?tm~n+p_^fhXrBv{Y|(vFB+{ZLou?Os8vM|@4dLsJzL`7JSh&T z&tkFYw_x#zKIK1gAN_eEPDR9v>u##36NJa^?>4+FOF;+AL>x5PhC_X~VPZ%UT;3+( zLRLBo*QCMkP8>{L#o+L>ts<8w5mSPa(8V$V0Zr+UM+$DwDZwEB8H*K}$q2DZg2PRr z1qe=r>B3aB4^G4K^&;=(ZyH)C(=k(PJ7!gFN4J1flyuL+?7(#V$rAZ|2ezZrqZ}AU zW#Hh_9jM!vfpcfYZS&G_tb7+-+X(h}mC!9T?SjMZOcZA32!B8dn7v?N3{Sv?r-IiL zmMmuDIfzTng-2(h{}8$fjXOeXxpO;)d`-ps;VHO$B_2oG#Nhn!RJadG5Pr%8%ups_ zR*UVp*&`M2guVv-lJLxM8{X|n#NZ_{m^UgO(~cJbtX3@tQxS7&w@OS7Rie(QLij6Ap}5C! zSRbmukE#=xo+akmzbaw&qEytkN-UcqBgptTl*t^Q-<9K+W(g`cmVt}ef6AVd$hcF5 zki&xQ|Mxh&#lE+$FBiHVF{gZ_#H^rlEVL`dn?)s9ZCipEOQE}LEXCWfLX4hRh@4wR zxQru$xmkh`JK>Y}ti-@cB9qCr0s+q}5Fz>?CMpc&r^K8}a8`QVu0ij$7qGFzS!AC% zjR}vcq4!7h9xN*Hq>U0Ezi?c9d>lP;PvC9Sd2HWx9vgq36*;FO(=faqqdT2P>#=8$ zVtWGX0*(vDLj}f6ti+B%m4X9(0?SQLV)9>(e9=4b%|0oZgJP~DW^_jeRKnBe1TyZO z7JS50=sl?tn+!PSi8UV?U5@W3j>GGgnCll;!r!_I5B62z@lv4|6f7RsqVp*AyNK~E zujBZ_Ye=|I2S?q@BIE2R_M{)euB>B%rKrT%=zZv^cLY(_O0ny-cwGMzteqIa0PrY4 zc;-GFs?5dfUik=B7dbKC_G0H4k-=iRUubgolRjk%LEu}Nny+U?6mvQIXC{n~~8f_rjaJr8Teb9>aV93+b8yX*BlcsdFjUghpPF!=ft#C8l7GX>5@ZSV`mu;EZ&YLD-_qDByFIj=$g zCp2W)p`!gLtiEE822Xpeydw6)UmdXItRp(Ea=`h8PB1(-0&4|dD<#!V_^g0^CrD(v zlHimL$D(Aw#$Cd
6bZYdBZiAS2HnDNpQB=kJX1Lq?+ZrX60*9u4(=BeC`F2=wkD zbfHtl8oml|Z+mBiw6nt)Ge(`fkV0gK20Jkbz8* zn{hK=)FK;V5&AwJH_PKNqi8E^R)%AFhUjfA*n&$-a`utW5^NYnF zDI6#B1yk~FFydHj)f0ls8{we6* zR(NRZ1oyFhI`WUF;F?%xc6k(%3{#M-8v}W=U=@g-TcvXhg6G7G9PF*=)jt-e{1OoB zkc?YJnK*kUOJrT|!iLSsa2}tGr;SP2u}t*rTE(LHUnHKbiN;ZJZu4G6;7aFkMEnpw z9PzVBe_~KFC>*~GB2g~5LOYs87T=X^FnXDW`Ncw0IZALWYf|91Aqj7cQV>xtcmxNM z&@n0wo%W}otI+WE%1DI!ghU*RND&O&bfldYn$gWU@V~YnKbH#5m4`TP_Iqpei$nA0(4yxxwR^$boGCGg^&M-NwIeZy=3GV{m#g@>0h?ENL|Gpk3)&Eq)Np$fky zp1@t3dicmfhh%$E)WL%5rhN-3KdwTx{JOYSJj91#w}e;m3D(|f#Ojp~5aDqjM?YQ1 z%@@YhzPX?93mMbxawCc{>O&`I8Pf0#I%M8YmpVVvqNU%uQ%0&TtzFWazDM+>znA(@ zyktoJo%_-)D>G`r22iJ+189v=e|r2&WW^jdCxbU;bYO%TZQg82mhvEaSZGQ8MJ~xb zLsQz^svq6VF`(Piv?(RLJN50;jb;m9XS>W!)X}UvrRjC1>!W+pw`)D=Y)CiyzFnJ~ z{`97rr+uiKi!td%_oh(;bm_}|J<92)9#+GW&}JXUGb*g9R(exxJrQ=ewp_o81Sqo%_f zT}sd}pyJm0)FsA%;`V9N^V0@2H&2%w-}Dw9uHLlU#(=go^`RN7ji^b?0V{{~rM~uk z$YiPxZ4&EJQ`IBQuKM&Nv@hlUF`-{Gj49u_KUFrHQt=lfa%^uzm#!Jn_&0|1tE3-A zpXyDhG^CZ5`qVsLmwwslliqsKTUljDuL}&wU)h)JI`*NUbYoi7#gGC`d()rV-c;1B zH=S!?NTDD4Qu+)tY94Ax=eP8w@8*4JR6}oCu(B^*nq^A54+c>7BheqoH6^V}CbVUT zDXo9&hWQ=mz#?iI)I4TDb*memY;+YGad%7&n+xl19$4IeE__wy;?!CQe>f9| zdd^0}drt%|U4_9G^O4(c8akex4i-5VUiB^r%A0|A<6PnCJwf!D+;Ei5!Qo&xkr6I3 zWfXJJ@yR%tZyt|8kxi1;!3ED}IOBPxV0{jmfVmS#W8QyLpu223s!gV0_6j$Ie42x6 zr{H{{lP zVr`f^o<`3UTr05_%CX7XI_wJ9W&?}q{gR)uJ-^m9WM?JAl_Q6O= z%z&~Npnq3SJT;pOOH()88SajQu3oS+@D|x2UQp#Lpuc}UTzbyJ(LL>c#$2@Uo zr3b$HFT%6sqUXG36{c3LMt%#y2UiQkuUmfDF>4uWgdVE5&Qh!>TZXPQ8(VcfvD<0^ zc;r&Vn)<>nWmg#kixk)C(8Z&x47M7fOT37@p? z>Na%5M3n}9X+f{ce~X?>Gd^5+hnd<<(0lb39?5Uwo z7`WgWc9p$^rO#7jZ+MNYMNff=Z$x%|172Kf0Qr4@zw2k1&isgk;x8Ci{|2wyyv3cG z*ZA+hCwP$f3OD>3k#y<>rgwRR`4?ZIw&@A>UjBp{gZFsx=Ogy6YKEuHR~S}*6-U3w!JdJRyte}T*Mo}%~f+xR&D z4($Fs#zBo+u>N=pjr#ZS=le_KnuvVEexFch{Tb7SeMek=GagO)gug}2C_mkVlVM*( z2EaR%zj}vxOT~_0{d1uYeG8jCuMoKU1vH*W#@}g$k31~qYWB*{;pdZ-a-z+%iO$ZwF5qZbNeK~wWV)Hk= zKmP@u2|sXRMKkK6zai1#vzYsRhgaWk$ZGWiKi2)h9x-bij6~>7N`%Jt?Pz~239HIP zKJEPNsBM#umzi1UrJ9LrVQDbc6)e!eWc-Ou#)vxM0TDfxZd1~6$300rP6?>>i^tc{ zcy!tm2XnW0e7u3hhq1@D6HfWFv$oNY^i9>n;eBY$7tm0 zMxdTWVf(!ZXpf9RmVP*9w%ICPFI$CwLa>r5W6^4DlE`HgT%=WkeUvGD>j%UPb5ayE z+J*xKp|}|={N4*i&D9VJV<(~Y3JZr{R3rkA#GtUHVA^y|MBRAddH0URY`zVS#tF!O zyA74SlaMQFD_5g5T#}NpM0}t6JrmDVv1b;jYXVk=bx+b{c>yq)ZaVrMp zq>Fx;V2+E=__aImEhiP@tkRJ_APM!E+cBtW2MP}-W45T@cKV5o&=f(^ux@N#aaHMp` zWFX>P8jku3F4bk>YY{y$=N;+LcF4xDzd7)#-iJv`bHN7hhU4KwSTvvjr>hIF{e|$F zzS#}eGntrpdnXnNw)gF)8NxT8jT4=cF(yQ?MF*$h`a7}aJIM&QOhfy5+mK`@`f93z zFWx^Lx~G$INz}~?tkRG+C>aei1n=fxB9h{Tw`!W`x0Q=pTO$>{1Uvh$Zaf@>m&)hf z|G3^_-PWm)#5q}dF&Pzq#XjUxl3+-sLP=>NCrD)SwcaMQ9kaM(j+5C#iY?6yqI68@qUJ9h})Wj3CS z%tBX}98AsM1>?(Un4X=E`4OpjtCxV)uBn)ym4Z3nlCj}L0zP(0!^!3pY!&tT!!GHV zx-SKN@>8&4Z6a3a2rlh|1eD%M$IqVWXcRoKOXBe!9=#21^8^q5+zzaHCcat=50LPc z*`|x&zfrH2 zvZ323GO#!1{@)DW!mE1_9kUlcJ#!KHITziP+4%b|8{Ma8!{oAH)LP|W(%80YZ|erW zQK?l?`QPWHv8t_9RMgbOcmK~vca@0Bsawqj};u|lfK z-yy{szb602IR0#w9UnC}ny)rgsVJIfTTx|@!V^9w@ga}bNc9KoD;!mwB=vKVZ2Or2 zDSmDn-erLS<8-@K#qJYKJ&xy2d2a@S=(_2F>wl4(;>QMZLpxZfn3&bq{p_015T z-PZhE+hhE~l_T^s)0!9it-zt{8zq(B5qwnp0_xr-Q_^g;jz?twk&afsWY14}@S(-2 z?8v*%Ol7l0MQz{x+;OrNkGz8 zeGVUF5z87jJmf|?KiE^JQtrLsH#bdR&mY+tSIo>)<3o8c+xF`yyKJbV|=vf2O4uNnX7eh1)BnOV=j>uvnhMy=wFwmG&`eJXo4DJkR?>U@zua`i z0d~W&mZ$t2!r#7hW7{^na~rcyeAnP!Y=^I&6!ycIZ!aDs+5D`QtbJSBUCJ`3u!y_F z&*c;@ik#-Xht74Lst5wwncy=0+B;6}8*c-Lv?SId1&^(ri{c zuz;I{dq{l__pivCTFz&ryYZpdbfxG&+xXJPEFKcJLUP*JgLT{Q&)*L1!3SGbQxERM zqlORSXFXQ)QF~P@4i-$|n=B%DhX{RK|DnZiMno`8>jBcyj(=Fi;mO?H!-#i%yNO>t zwUeK<>rXm_iAy2=LJeO(G z@9`(38Mh3io5}a6#OMaR+n=Pkw%N)<8&zBFtdzMi#Xa|h|uwY5{}`-YG5n7&+j{@)Yj*qdDyvpu)Sz7^I~HcE%gwr0v2 zhlAK2O;sA&J%}yV)TWl$?7&6 zm~oS_?dxKZMP>X9F}GSMZr-{H$#p$7hq>Z)X`Q^|b)>@9X|nXiPlc;4+fFVazN8&# z!Yrm4u!Jq6*`__eXm9x&x$^vOIbii^IV|Fja`PW2T7C5u3r^N&`bm;pF*5`jT5)pJ z;X^EG+7wtkJ|@5Ox27p;edWkc4_MgDxbhultL1;+a+yxUDrRk3$}}c$Y?LOl51Z6z z{K}uo-eG<15>%5g;O=87)4LQ-U*^l6mi^&PX7uu-6LpRe4|Ls5=KFY?wD#vvcA=ER z);@&pxNc|PW*w97^oyqAe<~DKF+gd%z9lG^qsWeUqR-~lFl_j5HXCVw z9{;&C%X|DB<&k^*@n%E>^=itM*S1O~zDW2}5*Fj|i+Q%$Ir?mrQ;FQczcoz|9K-YW zYLw_5PA<19WmEMoa%0PH%BQ}GC~H{4)?6@<_RiGiQJVEqpbul+&bG93F7RhYwGWjQ zvo*1PTe>`Z;C8z3_L;2uC_>S2E0?zU8bD=hUwVA&r93J4qOHd8P!^jTBX3>Bi z${i*V%G2s|WbZy2a^Sc`2Gc-zKZzc4_}4#0uB#(map+H+{o$o(sX9~eJg7`;G}RFr$=`Uw<`_pY{*W`wJ!)h&GaMpvFO8bW&M>o zQjGI-N$Y!m-d;LMHUEss{Z|_r*}E^>mQ;!kZw4~=qV9IaPTQpi-*;2v{fo-j#g3Bo z<#gKZ(?zOTqc4@H{H2tB-$XC(ne<}1=wsZxfjvH7XiRJ_y;h$uT^^DrJ%90vPFj{x zyBk(gZI=QXmKY(;h|rbiTQ8S-UcHTw*x}OErO)W&baxE89ZBf}EG20VqgDYcq_usg zNe>_Vqs2XfN%2dAmTUEtEY4kI&yIRX%m0MSKSL(VPhMNoMjLf$?Uk*xJZvFN`l2pv zIX+&}QGY@Q9*J4>>}Wasc_FQTZZCa)a+V@@c9x2FmD8@jUXoc&qLi+6+~)D8yOiV| zNu7tSqKID6q^o-!8?9ogKB%2y;s`BR$NNeThK)tR@Cxy{F4C03e93vQl@xAvkd%V4Wc$%o`jM1Lmqs+ujD`w2@J?5{HZVi+-y?lU z`W7@>vkq%BMoJS5D@kYeHtP7$o38WcZn&$Kk9nBx4#%kM2`%8C

{IyF6r0;=7}=Hg;>+*LvP!{u=47EB#~Pg2)| z_b5}+6<=gCX}PXEKCe9u-_N1AX&MZ}=Y!$2^*ebgYGQ4dEzsad95eO9_1}uHyDC_B zr!Dd1OE2^s(*qOEsbQsdF*!7N!M@!SwNu5;U~!!4ic0Bul{ftE1fZ-W0}ZdjX?*x0 z`gOyH*14ZU_m(zvOMV8sL7HgjVkSHc7T9+20eRQIqxg+|F}v2aa z_#2Cu_Cq+8(}+9%B2#cobO{w5gm>&K%6TvlZtY7X-KL!()%GowjpndVl1OyF&Nv_K^2!(U&JXiw0e4h1CcZ zZv49+Ka~67uZAy1beyKod(-KGVD+ZCdD6l30La|!B%6&>pmysz>Vxh0W_DjTYqrOg zHLlqCvo8|OBM>7Fx@VCDI-f!FZrTc#*AR+RGl9344c@#Of1ZVG*qLaG@v=}O!ds63pD2_r%g`dpC@byH!hA9A=i zD-gbg=J@s938(L9;zoQ5J)5r!?eYv9JXwrm75iXVD0*N23JvYN7CMe6A!p`2yt1f3 z;+I(1>+S)GyvsFfLmIedJjKaIQ_xWroENXz=k{Re4IPWYU3;TIXzIm>0%+P5JGyix zmiWN}cv`zsxL`5*M1lhnQ@KjaAx|Cr;4wHCeI4}hz2-7`pS?-`$GW2Jf)e`-)Z$@Z zXQCxT2A|h|p^qs?DQ1=$-&DPUIrTPN>>`bue1|@Z@5PA6tEg(S_?^sMOu>1AZ78x^ zZa>S&_)Z*Zr(MMKmz`KTJs2M5Q9=jqg4lPPkyhJ%QMp}x!v0oI1^zOwQP7a}Mwr}Wd(P1&4?1Q|>PJDI#fT8JuFdzJg#t4SrQL{1J ztf0q+`6|5AAqhQqm7@261t{r=p!w}eWW<+9NBq`L-#LV)uTQ1Fg9e}?#}@nkb4Br{ z$K+_OkHar|V{U&VoD`n(l^J;mTj>bzC7Sp_Yj9q8;Kv1&;9h7m*|`{F#n(hKix`Hr z6}|}jwgvx8C&FprIb;lrfaVqh>?uD6$D3bh-SeKf+NzKL`^YnUJt9@vV#?ca zpMv|9kp0C2BDXxAlzszW0%YN%-AEsvSinEYPGst8sA{1(28%xMsDbS?H8T>i^C#f` z@J9OAPxKwHyiD<*dm*oPUz{Gh2GSi15vpm5gpNo=eb0x6@l-szH5>=bdm^i#oPI|L z54*@Nm~0Dy?x_uU=~o1cX}9RqiB#&5;YUZ;$fK9XCe#N`M9|Shstj99)8qD$XSTY~ z_=TsS`3*HJzCsRT+i7F0JPL0_T(EW_Scc}vMUfJ%U3i;B29Q0_HTVoUsebo zepLj0c~?i9Y6l6XPH#9Ac+l1C+0^Yu2-)5}BQZZMM~01w)Y`|7a#PPypSL1+>$iy# zI3m!z+`BJmk3MQtp_GZWuHTE7kV6jzbXu4;O%c!6Y6 zH_)yADs)0Gg>197({7`IRP4Nfnl>Br;UahL@3EX?lrBqdY2A=qaMGgAJ+>4(L5rTN zbwkI29En=DL`hEHSV`MSXKMB!V(V5(+1y>EzRQ{>?Cc~n_0y8%H^y}T?-z-IcRcOh zS1$1nyeqm+o=UXFeUTLN1$K2*su!B|8x>&6XkeQ zP9vQ<{#SB4PlGPx&7qIV28gg1eSiVxq%N;b*L?#hbn{5;TsQ+J`G=t&Z-V)@C3qHe z6G8P2l9y>8BmvLPNZK=uso4Ifmp5YV_GB`JW`!LZHSgAzB!4~ znkZ;3X+gIu>bz+n3!l(z-W$J>YRo24;-|l4WSd1cO^&qpdoUfn_aD8Bm{0zVQIwaT zNN-dvNUv>==%$V3EsG{{_Pa!WqnasEl;10P=^i0@vb;&MFeK;ao@(`vAaLQ zrS@Xu&snHidkH&FHlgoIHwskIrKX2*WVu;FyWC$=$M{t8L>P^q9ZZo4v#G4RGd()F zj$EuyP-?I}MIQB{!r6Y5;Lt4D^5LnZNvAg@sca#VcPE>Y5=uH@hfCciBI~0wrbHj7 zk&|jk-N_0@YZI{fgEc3N(%?&aM04#_>BeiTneyfCLIqw3t8ug?PO5zr%6G%&&BXH`sEN__j1EmH*r18y^)Oatc zUO7Qa#ouh$YEG$B3MC~4V+7}a77f{DMLMq@Ne2EBn&j_2RKHi6WO{^?)7%4OcJ?rl zhB9@zb5z2Y8zq_xP3h^YB9gsXM=4K@U^_EY5_%y{q8|TD0_hRZdi$I%d1@jA5m^7) z6`wL)u*?3gq_MCceF;{e#Ro#r>w_IG>-%Aid?=kt+eE9PlyTQtj~W%m(p)j82TpMz z3t6ENl=@OZX(oMab0mAS3UW+TrXBN#QT(%ul4YV7v-ViH$bj0x(|mrzA8d zo9M6HR@y5QM3LKXN#;D>FBxX7LAKr{$T@XKu+aPQ&bVm0bR>XWmUhRA*z=IPcp49f zrou-@j_RVHNCs!PP)>QUq<4CNnF zFQ978zVy;b9q|h$VZ~$%G}YBg9%bv&&vHXju5zLby+A7L%%<(>@${r(9qn7KMLW;_ zrfo;7$#9)2^79qwse&wd*6Gm*p9Z?_bBqp%o8_JE_N0;Zky4+0ASaLcY^z|*HX(ai z`i0<&FC2tC*B3N>t2w$Tj)83367>3KgDc)aaJACM$>c}$Bs+ud)NY_2-k#_xQ-q+W zU*YNQi>0j_;1uGESML`?_x1z~85fMs&mZWawFKA34#eA(KiE6QitCL2!P-g&ACzWb zL{MqS4+%zL7b-y;8z^f8IDhuo*!hwb#5O;D#h z7T&cLXmC!W*4G>9Ra!CqS=S5G&p)NHgJhw5SQGVYe$%t?4*K^=Q``Zb&}zr~H1OCB z+Tw76K77le#ACG-m(fmfpJW94<2c^j7Fl1Bwf^Ecoc}(W&;NaL;>!D0So>lXZr3k_ zN_`QXQO==n3El9zsW+sZ2wgh9(1Wpkv2M>((rq<>|Da5?2%gt(eIs^zX3myC9WQgUQ(m*(nKl1)?TNOHf&8?~Qug|}g?k*9 zXXhhLc%o*+r?#u%mcB8xx5tTF_-g7O+CYIlGwJyE3cB)A3%^eq;^VOvvOeDv*G~+B ziPTMc{;!ELOFhI5O%-xFUGQFe4!SO#jC;$)>l@*URpTFEilaPx8L6;BnBb~KT!F!p z3fNu>fqv#*j5bce)zWlq_`46Wf7W4Q$YAWcp)YQ6{_r|>97WxNu-fnymRwy5zr3CJ zl;nyXz5m1eehYD<{w|J1WK!+ZEqD?*2I}HDve|zlnw_H%*0CK|BWndq+y`z;OfY$i z9PENEamX_j_gc0J{VxkwZEUf^TL}hIM{zwi9ou&0Vf27q80R__&!2lD=Dp}Z4L?sW zPYKOpi53$3D&R}718kaPamMoz4LV?i;S1;DXOs<22QS68gMBfi&uwV8_h+q|6EHX+ zzNfR~Ff3XR#_k35w@DqYqk3Y9+G31-6M_d7CeTva1HH>Y#)YX+IkOpY&EqlkXeA^+ zt|D?pGvpewq1=&1tBq5sP)ZXcofNTG`YnwZkdMJjPEuR19!NW1O_kX@aQLAU9xq;o z4U>K8%7|2o4UH!GsCaQ}Itj_JUr5L{L~qaQ^!A~?c+UrOlVBfr939S@Uvjn} z=|Cz<$EK0bIYS(~B?k|mMX=fJfE%Y~LjJN0N)vpsZ*&A&tcIgZClF=^eK5*G3gd*% zukW%E?)QG;_re~0ce^472JM0RxF)NXHqhTP}~^?<*yG=Rtp- zR7jN1T_l58!5<3hjat!ztJ-ZjPN%uU%I~`5g~C$OdA?r~Hf=X8iX!UNTu90T9VpXr zA6ZWeqdB4Lg&$%chF^+By0|4dlxw5^*2&mu*9&Xa_37x#PgFiv=<`qSQL2~6L0S%KQnu(Fc?R*MnkA;&v8^EveEDNt5VN>59pL@ znIyMUbblKuq4qmdOCJqXOcDQ^W1ZCZ;41p?NxY`Hqs4FeS?ZgfK)aeR39pp~#ngYH zK%ZVBw=RMA=w+y{If%4rhXq%19mbD5Ei|1doL;+!48NbE_gb-(Y;%hgH%vv3d4m15 zGnM)ZeJZJX2kkvqNPf?R4yn`)HJ;jNU;KayuU#U8sXxi>V;88$DIm8Yg4QS1(HP6s zw8_vM53IIfe^xk+9(x2TXF?PX(MqzUaz3$(mUk8SB9!AS}0~-A{l?)NGXHT#r*vUlcPGQ zZhJKRJYJGxyR~HD$vi5kABiIKOQiNz1;t7B(C`VNZK}q|PB2Ey?g7+#zLabt=0V2c zE(NVngJ-{wm>zJN{v8g6DfPhG#q+4CX9W2d{-uRNNBH)%ni`ikQqUnVIZE>8HA$+RlZa3HN*w3ZNK|F+P{uqL$?o8n zlB3~|&^o#oK3Cks(4lMKD^)G2S~gzHKGLXhT`uvelqajO3-rjbUD9H_j9mO3@aktF z{qvmyBgGQB=CE3_VsMVcI`tn-I68?ub??$U(UUq&DnRmGe;KtF)S>p7J#|TUf%4`8 zMC{lIoxm-qjO)vixj*1Kx-a*<=g9}|#o_nmN4Ps=B)ZzFQh{SQ^?mAtkF6<^U#qT= zP>V=AQ=3@Vnsy6K^zx7G6h8U|`2`=s%>h0%EPPxedHGLtpC1W5&w3@WAU^p!oY|5eoq>ySv zofbCa8XJcbsfw8P&kZSp*K|?gF*R-4vc!f(FMxIuxldY1%7AX=fKe=fHLP8kS9C|L(!GaWQl`>pLwG-l@I!?o+SA zMd-Rg8CS|xP~Z82%05oS?kk7r#VBbk{aQsO3pNmqsGeM!@D4^ZaRYgOCaTA zcc`+90m>RP$m>5dxYemqdgM@v-49#xJnV!cC3<)j?S`?VreVXweA?bq_|sQPBTw6% zsspZ3w7xb@7UoNwlJ-%1gbM5b6>Q_*SFmlCI<^ezCuwb@n3E}t~$ ziv70@?#V7VU>pyVfuU$sp21U!H*&WTHe7Yho;BnSqb+U-_f%a0pORp_-5CiLwJ_{C zzY#S`dvVr(5R9bG?1$EiQaV#LRI5#WOfrb)^QKWK&vNK z#mcdxNikCXZpL-BODJyjMD?TYJn+v+(&0l?-d;(KdezY17=yCI^D)Nw7rL5s5hv)O z{J7AVwN{nk-#EcUopSvRgK5( zM^1?DGZZt=k3;;AkK}IF1FPMXMV{~{=?|)-uG99?h=^diZuJVg>&@8m+;J=jJ%yiH zN!TSAf595g*yrj9>*HfE(bR^O?q}o1`uQ9*r#s5W>tk8{H}4O#x5$$g;G4f1cS(d>#_p=%CmaU#TlW z?Pn=Qc*Y2JYB`o`ABO+$T=}8qN%{rd>>Pmy=Cbg6 zVu+t2^LFdOYfAg4jEw4e*do7=PH4&Ex45lE9@>S~F{|)tpd|_|onai?UvwfKgnxQ2 zYFZAW>`4)MANWGa&RsAoHU^LTT@l&Qy{LJ55g%r~faRu(7;#;~ixxgY^-LXhF_?wb zEWsvZ<6+m?uuq(W9-b5Na@tDB{jfmMhzX#jAE=9r=#HK=6Jfg22$vmz@}nb=aykz6 z)dR32doSEAr4dpQL_4~uqu4*2utN&VoiC7*lr`R*oP;s0#`w>~ozm8hqWt`IbS6ui zmrv=#-TglyMI{mIt{36MykCg^Vucm?A;NDN204*W_+(p(()NMS5MGAVkKJG$rz^7a z(Nr4JO8uPz>CfqbIPCD6jA{p9wSpVMo|LVo(bR@Xd(j61WFz%&1r>y*Hx|nv8PHygnC-Slwdche_r)82~*XLAN z|C$1Nyd|ZNW63h}7%g7234!&gs9o!hIY+P3rO##b^~5FGx${14pV=Kv?c;@avYNij z#?$BfrIgxJliht^AmP|K7~Q%@t1Cyq^U+cq-l>V(M(;^_#X6Mlev5L!GSnH~gaLwi zR<-IL_Qt-$nGybY^;?3`F+*@TU6v2X)Ic-lCHCgX@vBd=tmCN8I+6J(dy|Zu{>P!$ z8PEPQ$%3_hns*;e5ln;QTpWIcbw)2oNYW6PH!Q?~=v9~)XALdq<@jr`1XVJ5WN&%Vf^IpkUig>!#-c;RpXBH zoRUVK5_%Dz-P(}3LF5`I55eljd1#yIf#YVIL_dN6BlUGf=nD@x$gF|V_^G1NWCJh0 zZznXQ9lYVR1n0JF#_W&cuCzZJxz5iOAxfuY}F3z~quF4CY4EeiqU!Fgxmz7qv@E}*JS`BUw$Azs~=+<486{Z~% zTr#2a%smX#>=Rrjm&D=8<$Ts>2dr-Hg1@*AX=@CzvNlt+ikP5fHTdHyNUvCmj6QCn z&nXzyemT%N5e1F(tK4q*h`*ngwi?+#n`u@Gr=~sN1bI2pZ_$DEx`VmB;3|8}F69Kf zXKXDx#LI6?qr1^o|M5!DRul;r-VzQPqX6X6Z|sjB^Hgmg~zJWtdUuT zqO&b9Hm%2;Yy-~gYR(%T&EW4ECo!cd24{-Sp=I6_p7hX}WlpYP&GrmDJ^UTg6T5MG zn;q*UoACu%J3f8U5*wQxQ80S}T3WWDEMy5P=B>odsrGDJAU-GELRfi}(1)x}<6=b& ze!>acy+l`LK_J|0gzq*dT=3BL<5tW;{Fr?JP2-dCZ$~|v7GA)cm>00peuFKZ^1P_x zD=e3dW&fX3`L&fJKTvby+6jJa6XnWNlXfG!Zy3H#Jpk{g{kULA556OOSDOq*@|O+a ztkQcgk9SzY`o<~TCoP&?G}dC~fi!fvZU_4f5-za?AL$s+B|2B&5|#;xZX||fs&L(+ zrzl}nej?`U6R*Odn-+%$hAWY+!r1n80Tj$L;h!Z~;pPXRUb30TWdL%rd!bFta=&sUu*E=}w;B!Y!D{Rw z{88&34diz70!;U-#?z&BxKq%BxAF@-Szy9Gx?5nnehGdy#be9WMy%cV2G=7iQSLj9 zMX5K3Rqf_d1r;tA`M;Kf&v9trbky!xBfK`#u|v&=zsdNq@0}1%Ikuf`cgXMbOXHA3{bD18r6q$|~fxeH)ArG>}Qpzd=tYR?I( zaUDDyd;8u;T&3pTI91A_@TL=pIL^#N%l zi3$r3575JvV@SP6l5U-zMJK<_#((j=KlES=z8ihT`N~|_v^~Y(FdwMCTmbJ`Gw?IO z9kzFt;#1Wr7`%>$wN@JJyMrO~qz$VZD=|0V7tGWBs3B+vNew+fB>~#FHDNkZG$OI7 z%oTmZ_3(PjevIJVtsR}JbSqYil8Z_pJ>xY-3@StJ(&s41I)~jF%Ji+c01`JZ;O`i1 zGB$03uBtfM94Udd-*qhHu0ZWwqv@XacZh$`BJ)XaP@N`1Yio4KQ7j(={>AVz>w|Y^ zDbznzV!;I7Ar@)CJ(~>Zf2haUcWJ1xjzY!AjR>sTg^-H%oC9?XKCx?Y_T*Pq-SV5y zV1Kc5Qp0FNrvk}dRV9@<#w5I}O;2xXknq$P7{?gF{5Ob6{kIH-_)aQy(-um~P_x2pEg}NC2+!lEq(~%aYN?MD=XhWbHts43r zlKTdczT9gVwQr()=lvA_Z!39saqf&BVfmF+Xu0GAk$;YG9Z10iS82SNWQI-K9niK# zhOk$h#6Ew*J^l=R&U=%O`Y*wL`bT~*|KK^OcbL`fjjGrz%=Svgx{)nNDChkT*%mx} zHI$Ui2VnYs1Rdu)4DsHp__L`7vSJa~6@C;!vYW8O#t)@iRzSAh3bGgW@pDfo-YwaO z&)sgcHcW-`FK(l^e~-}kvg4#Ll0xsV3Gwe|5I$~lhqu6m_^OwdAD&MkYBT9}4)<#n zO{c58JF)47BDIfDrYZjgk>?X@G7`9w*@}5obkL11-Etz;T`FWVNP>p+{)b@IU(m3Z zrh2&^+}v1-^KOq((8Y6@i_-b|{VL=(W^v!#GCFDPORk&+WV2U<2A>MYl-q;6 zM|JpkMU0vbg;Lw`r8IbAG_B36LDSTmI9m1$Hgo2Y{fHoAoHp$%e+JPcrJ`==C*w-i7zYaw~U?;#p8V|B)0OGk}aI z(zHV0i|c!ip}A}$KE0mJ-9^&00#4YS|Q~^ ze@FV#+CF_6(lvut*BQ~tyQVZI#fh9pj;Aa6P1u(zN{LTCgT=_w)oGzM7 z#6%%s%`74QyP{3*d%ocN*-VLfA{!qxCQ6HSYKiINwxt?60Y@?7Ng2NJe#L_cSE1p29!4LsFk;0q#5F|V zdczzF5?evH9cGiQ(<-P|PDEo3hpY@YqH`)%B&R=x9IJ<8>~VL*?{dZMAHlSC%6>Yi z?M%|wcT(PhRI1+-2#%50X^0^EWc@o41HQGGrN#_3^rthx%{Q*2T8a3r0*wVZZ+aHB8f>-iZmoW#WADD0Oz zZ7vYd-`5MNr?U$qYIs*u>@Lnb@O!Gy1-Ugi5_$?1P3Lg1P9~S}9MW7U`Iq{~YPdI`AOw2z+~M5VyJ)r^F9Zx^NkH zAqLVW&URj6#aWNroA73)y)cVeoyj)Js+ zy@$dt8|M5j`I}QE$VqH!o4rscOgJ zvsIjL)`@4*%~&sKjYrN?q4;1wRy%OU+H6hQ5I&psY*4~ckLk#`Wr?L{BM|h#lfO&x z=d5Np&WeTLd36|$n|qSOr8KgRi=xvzwP;qS4fR~}ps$i+v2KPQ?}Zp+i_-<{Zassk z{#Ov(qfEsPI;8tSmmG@)PP8UOr*@$|cnGO!{zkyDDL4|o7In>$$h^Ie!e#c6 z=$JGTA09@N-yWg1C5Nc0)tJ)mm{3~AYLYoxM9bXLNM(N|U9=xc_OF%6HEcBH-8h8l zVd;3vWk!De-0S!L4;}@Hk)>BZ_Acs1+677SKXr_*)wz*!#3hPI*Z@-{Jv?7G6SIek zlUIc@UD~NhCdsLoGIbYqF3-aKlyelbbRR{2j;3WtM#H~S34^pv@u%<@Rb`zd<*%tU z^2H_GI&%{vj#lw^c2D{k5=tVcIPYek8hsWWPkKuPbaCf3Qi?9-@7E9M!xACdc1=Lw zBsEw%{l)R9=U9A9mPYTo3J;wmG}PYUEXc8l+~p6?TnQ-8H=}nA%Cz6ilsnX_&^N6J z)bj$@o$V;reIkXbEhC>c;9|NiCRe+nqIMQd=D_FPzq4qb`ZRLgZ9(gJA3j0Lo2J)= z(lE7PDyuf67ycG>O2wGFr?qGqXPcQ$k*Bx$sEI&760f#J|lZaqfX!EG49S6O8ECQA;{G4D@=K z23NYs89rcIxUHL33*`Dd`@MjmcR45hnczhL0w zC#2V=;D>QOVw7?qx^xgS-6q4VYYa{v7)>fp>ZHJ*I~zjZF!K$qOm|@wo1Xp)8~k3v zfWASnHWUXQ??*{U99oy{pwq3ZDeOW51-l(Wo$g#jHJm`~#2=`-T?OC1LFDq+p4RDD z@y_#7D!xCP)F=B;>LSi0TK*SdPnD>8q5&zKv%<~2JbNA@OUKI%Sj0|E7S&tH+^_v+ zstXx9yf1Q}q&o#ftVHJ347wC_l#O~djM-i?!rQe1j83#f=a2(1lzhqhbtYqL=N*i> zuMT~8HQ1_d#C5?LW;tm$i~S{rJ*&yBssKfv%pWA7)&rV#__Tb zm})J}#<~q<9nFtf#QtJ7yI2ff@7)ns<_@p!R5UC0!1!Y|%etn5aFMzAPkJdDB+XD6 z)&q&FZ&-+f7QXD^=g|D0Y}NKoR-qjQ{j}daa~6%wiOZpB+01hG0LODC!zuY7({hPr z(-MBOS8paWb%_Lap&${?gX;ylliv%(#;IYEeHp8I&K>V11_)Un$GoDYm~}!K%Plv; z&Gk2UPUbllB#z`xCE%)8A0bT2zr6^5Y3&j@pG7JT1OD(DQE#wx~Vu`PDhY{%`DxcT21CbmEl%27ih zz1)BazPu4kzRu9(-ziwUL!5bPkA=m9y=>-R&WxGRz<1TR1&%(VZ1;sY)~8#*n(Qi= zi2OLrblbo(5h7UUdl46mv}sE(u7FA0#Kk7c`50=WxL z0{%n01arzpurC_gXgZ?625+=x&u(94uT7f;N&YGQSegt0wwkJ)OMdO>bV zk04LYj?tl~%)^d%CXV$$#?}%AUlfsf=?Pp62UxDxM10^glNE1c*`a}Aw#-)zkC)W2 zX*@Ua^JqFoW_Jj#+DNhf?I)S#hZMNyd_#BcW6WFY0{I1|xM#*0Lt+=0(Hlv=Gnmc! z%r^wfU*8lws!wO?ll@rRPenxW-?@if3)tc>LUy~UmR;KOo~84>@S@;RuzGiyxeYTy zj5=pki)?0!Ntx_Iv@WjAW2`>(7~4`A&0MpGVqbJLiYM)bY4H$tRklG;;cLlEhD?E3 zqB-~GIBXrxr(*Y~@uX`vmPH+KZyhhE~o1z8-JV!?az zj%?Iv8+K~g1`JzR&76klB6Ot-s)7vJNo^sMJ*|v|+e0vWi7VpES0LH%J5$?zhh2GG z%Z8`*A|A7Qyd$Q!AcjRG1heV0{xHiwN)X>O zhaGkL!&1~Q3T&3t2sT=L6HM8$3`=&$!DjVu?o-|?7#SYM42OJTMK5faZ=f&xa8by< zsHj2r?_oAFRfaPcjs)c>Xz3TND2_ zykoc1F5#+u9EO@5!N+I%$ghlo`=8^2fR49p>w;($ta5`~L@kbVO46^f_XKKwuLYC1 z3u$N8dA4an5EFF?<$G}MteTh2P8P36S-KKSejP5j>e?;Pn*CBxd&QBtW!hoi3vEPm zme=PbBhChN#??#XkTriE62%-@P zBH<7}D_c>e*m&l@TAPJBuVJh9nX+#6+k$(41KBUd6_}Yc1rDbD_-Qta9X3_Nid(<% z^@2N7Kcvq#iKeoRSH7^rjWb}B7l<3;yn7Xt#tJ--!p-73tEjGG9-eVn-?j7&bJ%>?fznJqS!Qd-wn}jLMV#V09mg}tRY$JPo z;kID0Z9c0Fx`FNzM>K3|X9d!PrhVtxY%d*T?LLCD#){nCAdR4q<+%A}E@PYUY(t831y>q-eb3) z4Y0o__af)JChHJC$z&H6u^~a`EXsgqkusxEw`ms(v;W~l`GT= zWNWXp)G3n$L#8CNdY=VQk$5U7A3lhk4?HVa(UHvdE_%yGScssbww}G0xxw=Kl`-}5 zZssKUm0b=}LF&cf%;a^E;Kh*?)^m3X+7Df1S1d!I5cq^GD!9cOYm^};T*a)9X(DQT z7p8fOk*;DFDt@_Al4&Wsscnr9{64iidkRY`RLOYId)}9~C&Tzay11x{9Cs5=d|yg= ziMn(r^*6S78qsSzBU)qg8VB|@Vw0UT^{-CE8qqKC2;?rJzE_a<9ZNqxoTa{HiI~E< zeGZq)F=r}wK&jrxxRU$WHoXDU)Hno%d<9oc-?^X~N= z&N1B@%{hVNVCLm!rof$M`whA2W7nQrV$LJbzh(oH}hfl<|WtUJ#5N z+YG$-6DQWOh#v0@=j^BPG&*Jmsi|xw&ogJaPez6ksRT_OcTkWQMhjB5(NNQp>cFjzVl_=N%(*nsY~GL?*YfIER5L|1+TX<)Z4B>N5W1)>)RWA;d{XK zcXXh&^*Uzj*5em4ccRij3CWBQvrh5in4 zqu@D?{)$SFV&(@Z z#C}Hx@2)qfZb##sGw@aU%!;Pu;p6s8ICU-*Rx&-M#U(1p2QP01=XhW3c__q+a~v{&RUEL88I zzxNF`ml~61MLMph$00$v0j}4yF~#*Lyw&d@ZO2H6R7c{$4iA(r`-AF6MRGsAi6pxA z;@Pz<1aE(cCAEGqnd*-x0g{yMK9}sGCexy2+iAv1ZdQskqY>{8lKZ)}Sn(lfjo!Yl0bmLTu6L6Eq18hbt4 z@$O+Qy4}R7=;aH9-#vtu^Mz=5_yWe7gZWHn2I+6NpySIHVsLdQ>MH6%Cgw1sT~I%q z0mWJ$T#^0&gM}&-@+=PXyi-Ls#-^EGlP^?)Z?4LK`c&{jFrRA{o>#j4Y z0UD|W&#I1@ipcEEnQ zJGwJkFlt&0V|PQbrg9jqd^!S0B&aUS>b8 z=$}M?Wfs(#C5oCYLDm8Ad{c@(pQrr)QO7e;2jMil3^yI$qGQl@49$FqEhn>JJ0}M< z;l)TE`Ga>U8FyKa$Nb&ubhMuU<1F@jb9>l;RHRFf^2&#M2M&ka{y3HBzCt>kx@Zhcvthe~7R0614db zcdKPaqTA^jHus*#s*DU+>YRi5*k~5uQO8zm?|?(KA!lTT!KX2he73H^h^>#Yfpb$s z&krJvI89pUzKTYTo{8Ft8A#-=gQM3LVXPkKLvJrbUornYM2p}c?LrQt;}N#uIMyka zVncZ@*51g%uynq|pBspV*hqd(m_ikU`R8^m4(fWZIb+Qj4<=t_XB|#p{+0?_7ac$i z1!dIhvIu)x{P5(#Ck*!F4l8SE6y$27R#XIz1(&dM$vwn60NLc>;dx_QT{KceE_a<~!?OZ13c$@V(fCKMTg;hTJl|;ofxjMcdiA z_DB|bEdpP2cVaZ>9K_6Rhu1bAG*-o9;g4tdnm-H6jOOBy7vBYD9)oV|MZ~>rLfF1p zXuh}-XV({^V%U1Dd9of;7o5Q4ojGXyQ-Gb_pV9Bt$9`_P!%9X3!R7O6wsN8p?w#Aq zE4d7%r+%}irJ0!IKNl8idYr)%5BtJO#*g$EoH79iN55i4;u^@Uj)q>MHbPF_!X(u& z+*3b_CuT8ldR>AghXe7?@d-p<-a^2kqwu@A4`Y_d(YSwn4sQWW=FnkuMacj z9XleXk+wi=Vjq^7u-`!2} zlX>2!Xco1ecBjM6(G5xrE zlDwXJ(3s`~vayXO71MayTYev3y7??|YZ7Aix{+fKXOE^&BfD9rct^N`GJZ6XYDhPx z)qcRA@&7^6rWV7Zd+=~ToBp(mkVwaSKERTpABV4#OvDZ9zj&PXK9duw6^<4b`%4Q; zeb>_RC5!1vr7f9H-$lCx>*<1wB`y4NpT1?CA%(vMWceY6JJfH|LEbGc_P9-6Pns!s znS}5_{dI~JPo#*aXQ)s@lBS=%fg9WMu&>sF?mK(a>5&Sg@QBY5g&q{TY995vOOR~T zeY{MoLdTyqG;j4t8m?qb?0}*0*J5MgzX{sHK3i9ce&$Aty%NYVT2z>}<2$u>_ELcI zFk!U1l5o#pWloruCBM6~X^Gx63VS_Fn5Z>KsOKK;eOdzrugj;W zkx>-#GLEVTbs}RLKNH6Fq9{Q|=%4hG7I_;8Ck1`SmpR-cl-CZA)eq>FTOKW)*h!j} zmuYI(RrXi&5GnqX^mll%ddpEM}Bl1$`J6$PF zYU0;ZK_qu!tXWD}I*C5|gA``yQ|72_S|1inpI7aqe7#g^m~(&z?gZ0VjRll7$A;wh z6GeR4O{-pp(7ZxBy1IS^HUHjA{p%CR^gum{bQaUACAa9!Ok;X~Ulh6)o2;ekva>SdHp^-*0jLs49`GKAW{mJpf9gSlFnBbx;Ht1j8kLDC}Jl4 zx<8SA`&v`E>tPyMmgt9IxPY*auaC121h-zs{V zl0vyhGO2*x(W%cm^sIOVO?c}@-CrM*OUHT2lFFwGe}|IT`=J!ETZHC`azD#0aeB7u z6Skh;KwfJ5$b8KvNTJmA^f0n*yH}+Vn(ro^Y^E>YriWXhQYqR6EWA225+sreMRa~>_#R3b@51=99Xp)^A+zUO<5-!Y$fHv0x-XN)7C?7@^HrcO6P zGN^j(2|5zEi8idBL)8%@$tqf#S|X&$;uc;xkXF;Bx~t zwpNpO0-x{7J7OXCq{f%j;&J5zoK*S`vm$susNn%poSWgv-x zUV{0i{mA;S8R27pjt&|0|Gz`c^Duq;YRJ#!2IVYhAmz&?u=v@Gy$Y2$=c5e+ z3lq*pm<^*{`(Q405+&e-Wo}@z!5VS!Yd?KQ1M&ULiV8n(?^=XCqeqVAt>eV%wH#Bf>2MXY}qOIQAEo z8{dR|{$r@^`iM@WOH89c507L#kZlnUg{FhBd+{2xOYAY;(GlaX@jlb43e5Rj&V37d zoVk&Vgjw9FSfx!K^_JKndje8J?%{K7HnzVgK!e%~L`dagN5Fmj8u<`sQch!q|0(n+ zw4rW71Y(SwdG5R%Kb7a<%}aL(Q#o6b^MU5B83w0{0rsDJ3ktq|;htmup0iJf1j=$W zkIygQQKFk3@lbB5*Gp880{C>m%-Ia21`-d?!oMLie#G+i}m@J z;IL#kWrU5w(`PQo^bUcFxi-R-Jn^wL5;1aPP(6Ap+C1`+v$h1;H}6AT`aW!AdNFyk z8tFXyhtz3PaIZiS`yDLdE0>B9d$aKF%@s7wT#wP}gohi$p}tp(^nc0Gl)82Fd9xw{ zEln}<>q2boUW(u)Ubs_r7_r%$?aOz^XDx-W9iD;}bN4`C+lk)FYmm$Tg$T|l?V4@J zjE@}(Y$Wx?p$wgK(~=W*aqG6H_x!SzM^;Jd&DewDR2aYl_co7bS)G7UvV<#1|P zhe+vAT$74`OivlJ>fzi`pA#rLG7jz4d_VMh3yR;SV6j&kHtga#$7C;>+_{?OJ+dRC zysA&MJ6=OoadD%jG;izNj&EmPBL>1Y2}C$yi3TtI;M+I^X?d~y$po(31mUij2W%YP3O{nvfC+c#TOPS&JJSQ?m zSaN5MaL#9I;nEThdcV}14sfT+N0WzGY@tR)!7t$Z=s&8z|D9&rJ)$1fp>)3JIgXk2 zqfyO3sFCX}WWDo**?rehl$?W4`KegEECDC-V<5WbIEr*lgaM|*ge@xS!kj9BaL0mw zWLP05JQcwI9(N}i-)Bf8H?@(9?qhoD`j5`k*pocxzH0TFQ%ntU4ZBL0_P!-|LtUYBkgV|hgbyTT zBriPpTTb}dOh>p{Z8}XpJ(dPvFrek!!}HEYiRanX$)kHK9oAV+_t}2BXY>(cQ;U$e zpaIrWvBZJibXPfwX387UGsazS`%LMB+<7v}jG?fO!!$~CEL{&CO$RI{lE8BcN$^R@ z3Q^FPJMr|h&x6KWMA4n{LF8uLiDst)T;V=BiRPhnR^}Jndb#uI-yjONdx)eXL9n@T z7}uo3z?p7sfFNmz!{cJnkG(nv%u_iRk6j-uX8>*(QKC)&I@8$)^vv42V_Qg?^brzghb%IAYd zQX9qv>NwKez#!_l9!iIK=2Nx54t9}+Si{{n$HX>J z(1@wD@-XMXZ;e9H)Ii+o8cju!%JlJzI*pV+L~HBE(7L%UR2JWl;8T-n_Yz|Yp7<1Y z@kLOyzX=ngMk>EkO(QkxNU-`beR_D0_Bh<2or!7WAoo&e9fn)8faaAw`zuzQbc_H6J`b*PYjoo3`a{8&M)Ns$MH08yBw)=|DN*w7Px7j$EVeYF>W!RWfUm$x&B0aynh%= z69@QghOR2SOf z-Ce%>eJYD36CNTl@*4l_Lon%aEO*wuWQ!K-K-D|~=hfC@$Qfm9y>yJ;+6IvF-qSpv zD2axZ5-5vM#m1-h*xPJ`2BjMI-{MPHrIH7e?_;r+d#4}n>&Ke2vuN}_OL%i9{oGlJ zOfol`P99MsO~ZU>Y+sM}#mdm#`JI?#^Wb=Voiq>mjs5KU7d=WF@*7F3htZuRW8D6# zi*D;iHt2;BdJGzv$+!@9ZTEkWt{acl$JgM?s~EWUY=U09GltJ}#CC@oCb{tuyA`yP z^Z8z|dj|_ybKgFkdBC~osy89_ArhAz8CuUS$F9^&m>mxW<2$KkM{PklBcOTf87n!u z6J}%gLgCL^+|dre+XExet-Bhr!=_{1l1RAQaVEiOHw^JNM?K#=zKxd0&dr7}J@cMD zK69GszT{lHS?8FYu{8ac!1JRs%rQsz50aX$L3{pTh+nzSq?`uW$$7afqUi*`wh~bh zuLLVSehuCpV4GF@*f!Y|lwOmAn42NSyuFFoCSTZ0u*I~uQ{mq^2y|Qsr@^W)x@ZG5 za(|J#6|(hR(3EhLyIB2DoI4hUp8c$5T`camhQTXV0Ug6z+1A}9tUB=|izxVu3$i+J zbgsanIV&*Z=VaV48jRhaRB5ppBlSg(AzLkjU9D;m|JJ}{=JI!itF>%GS_yl8Uks)e zgJ7iil`Re}!OIz25hb-AUk9h*)W6v{_b?K71m2uWp9R63a7ZLL;uM+V-r9BWeH{&{ z!7`YCRuc(_R^Z%#8faz-lj1vwS1V7T^aN)wzMPGf@wx~ZB8BjO3vvIzA6Bxzl64fX zM*hb!unT&|q&JGvo_b@9^P34z_6?KQS7E-=UtIg=MN7(!Xz)^N3ec9It?xz2Lv#@B zGX6zvI`3#-r=l>r_9Z3xg5?1dD?oqW2cCUjm^K#3xn{9H1c{&+6HBg>CWc2Zq9d(#R^8{i$p?=Cc{+lprNO`%1p^J(elYZM)rMasID=`zMr+q=1>=wVC0 z{}hp*elyvWmC)`x#pIlQj)rEOrd9pXVD`tc;lc(Sb!fx6>v=e=*2d3KLNaIrZLyp{ zaY0A$m#$-l+*O>9pFuyKX;IwD$@Ir`oKXGqOyQt6D#Du$cd%tx5BmN3kuJ^XG zc==?)KPjQk-GAg#(@yFBvJ_=0Px%WcQ^zDtp__-A@M@QwuzT7UbR|?GY;7~fx)Ev(ZTF07Wyrd#e+e5RE~29sBir-Ut?|2~&K#SRvhJp4s19q;Mw zhAf$vpyQ!q=S0aJydbdm49tF}pyZK?t9jLx7FHe*S_ zk6$YjtY}^4a_aZnPTKvvE2kGiwD1YEV)#5e{T}KbjiO_JN08lOV=_CDNWTpZ(6cXx z$Vxha7Qg3brrG|a^I|wPpBE*CALiuoDS@1xN7389gOpW}gtXwhnC!`Us&mE3S~p*&3^;%RZbE2x6#g&M596UApSM`_Nv!}Rq*8%+tS zqtoXek&;XX$z~T(&WRFgIvWq0|JM@8{ULPyYvfEwE2g;G1Vi)lSpBnh zR8JR9a_&-0H3&q7(g(q_O?J%8%!p-WK48zq8U&r^7Q(Q6J7;v3vX6&VAU45< z<#W!^AD$spN(p4WXPnv2$;;uQz}5D1)XC9hWwe$6j^tbBuX`2vIff=7|ydB`}Gsazr8~+VZ8?n3lCz;q668~q4(Lb zA#0g;j1IdQFUo#Qip9MpTXFWrBNWf)-HFBttUN}KZI0oy%xIqDnj;JOSL;z|pv|s0 zs6g0uh2>R^V%h8VvUi-f{W3Y8Z4LBcUj3!4(@2}$Hyg#Gqz|&t6_PA`{y}!ED2r9T zn8WUs_OZRk?=qK9HO#_zKl6E}iVk%hymwHcD79c$Y^w!{I-KKW zb%K><-(qL4rLlNr4ZQOVV1>4)*gMW5?otV5?OqiEog^C+$scDI3T4qz-@>+McC)*) z_P|wN9MXFlSn-2;w%?NXuv2D0;$SCJnccyh)SZ~p#y*kKQMrIbsu^drhFV5ERe(<)@c`7T*!eG-0(E@Z=A zZ)fFGZLwNmB(~_+vvYf{uq{1tET=t=l}d}Ed(t{Ka?Wb*L{!G&u4Qa}Z!)t9AIsX7 zX0ZYPdba%7M2uR_U4#`~g6Lio76RKEPTvd3mX&N)xif2h!u>D5 zgPE%4Rklg*AM?{1g5g$!cpvx*=fPcOcd{+XVJp zHHxzq6+dCyx8anuNs`8R@P2X>_ozRzfq!~GTRW|i-Cj5!J|7}5cyTa1Hl4aWbNQN6d9cNaD$<_6IO|N@+qEeC z7KCE%15c#GTlmSWy2KxRk2lGC3ga%xH-5Ge?MZJ&fjLV%vBA z9L-TB&Ck4Vam^Chn>Im3H6OPnvSQ;zo)eq#zE_?3D#d;gFa_(6i3fRa^xncyJkb^ zsE?;}M#4|TiSJ1Ra9djrv&73#WjmV`40h2pK{Pq>&Tf~|aVTr0LbR3t7HVig%=$iO z+cco=p#<&L)hFwPocXrh9c7dJARpmEk3y5t6`4vgjZ?5d-3c38kK)4uTl6iPk0FaZ z@XytOV&tTWL^+>RITdA2A|$J;Mi=XJvBoR{m*t``<%kybZ<#>};`-E7&vy}Og-9N% zME<+V+4?dMw#Hi-rz7rSAon`%*l5s!WKE1!917n|E%<-rIsQqT@zzS2l>;Y^pTcmaF73QqhuMxYw04+~vvt<;olPHeYIDK(bBHUlHZb+PCqdt`>F zV%iE9)br;`q3kLsb0@t~?<3BVj>1ESGRS6lz|0{Om4ma;Q8Nn{H(rM2HY0@GUW5hw zy?6QLOl(wMg3kdS_|TM#50@|CkM$KuzAwiPzK`ke^TH?Ynai#Y#F<0fk8ZOXU1@vZ zyIBnlIYPJ!`TY6QNyItWV zUtfYnXU4!#aR@%n5yj)5EAdEvEzGoZ5nLII8r8GNS#=tb=l>$>TNoxpSK{fvboi#- zg=e}C>g$go#5EbK&W58cQ<++qnqvHyF_`pl6yo*fA;2yc9->nrw{n1`oV&x$D-K7+ zq-cmuI)X{@Au!NtVRz@x!1H8De7zBcOLg5C7d(_cmJT856NT9J;wNMk>yeq5!m~&> zG5pQ{v-9dOsTqdLOJ*a(z!=jvE{0pG9NJHQW12d;aCY#(Lq{u2oid)e@V!CQ+f=yK zE{DAg_kcAroDJQM=I#X;6EPodlSX6F8Us96R))OYcx0#C=ZvZ}Tpp!HQ~CEI{cJf7 z3<`o}m=b4!v!LyL4%(t?;e17zR%XW`nfLlO&U%2Dt4cJZS(;2=cu{842ehx9 zMrWLO_jy?cpT+z_ZfXv)Ch<9>>wRp=_zzdJ&1hkJ2CesxrlRq0(QP75cfK0aAdg1e zou)|HXL_JJ_#gHx??%D{1*%`fS%$IkNPS@p>)ueB&U1%BD?CYfa2r}Cd*WW$V-$04 zV3#<<-tjY_nH)^&XXjAouB8-|XF(zRme5`1MgQ*H!^7Y@e(m0Xe!4kXxB1aqwuDY+ zI^dDi8W=qELD$V$Br<&tHHK=@kwvXIDE|y|xg*?ZZXD_h55cm^2a9dbqSf>+{%S{} zR@W49hL)H+Zy_e;jV94mHne2&Z1O6&LLU@zsUhtW1%4Hwe(!fM-unyg6Sb*Xcx9I>mYrE3zxT*YpR-k_u7EOEZrf(bHDmVVl_t{n66&$NxFs z4PK_ejtlhQnI4HX*YaNSNE)5H9~U*aLunHCJ9M9*3sL1%yCa*Trj4ca{8@BNdkH1} zs6k!Ve9XDd*-_jLZm~{4C*PG)-8Y`G@LfR@`SLe)TO{&=!l7YkhXlzB$Tn7``Zi;F z@$4Y#3+|va@(hf%`w({X1*EzE?URH$Ez~ff8Yd;v`{PJ`KUJyw)M%=T9#3!Uok?!Q zQgYAB!QZs^h|?~>|0p`|K&-z%iYqH4BZ@)@p`|4EoF|o0Y0wf%MSJf}Q&th8GLmF# zNJD8URN6~JX-DI$y`%NJzd!mjJkRHT?!E8#`<&M?*!c*GlIt)U@f8DSJ;wS%Wu`P& z=kn@@P^^w|Kc<3E^EA55L-_0;ole{i1$x2yr9BnU3-Kb>8)Z-Jh zeWm7Y>}ELh9E%B=k(i)*2?cGVu*Ito8m1W-)OiIIYqr43Ta&s=H5k-fg-$X{bZhqE)#p4@ZOGnO?Kl$Ss&IMwqo~_zo0VxECTP9V(wgZ#tzy<&ExU(+P{;@|MnrH?kPUR zyv1DG%`mPwg)5VvBj!f9SYDqddOtoP3T1Uz@H2&0ZoB#4${Z|PVaDKv$zPenuM9^97a__rY06SvC8&F@{s+9D${@YPtcttrOKzJ8>?K5Hj7XJchov8K1L@JSsj@;;TocUdJ2X5GgX z?_wGSpJ)E&{|JY4j8(|>kaHi-7QnF(BTN{(7;!P7Xf%9`9i#O*a@s|4GH-`i?x%^m zTUR-A%3DtPRK#&M6~gr04_Tv~5k{81g;v!DStn-+pO|_4csH7}OGJ!a)Y19&4ofFTP zha1^n`h9acu7T3H6y#hhMpyM-sBRSv_30Vt{#~D26o0W|u{96Q2^Teq*}|#GVUc3? z4B4-hssE%s2ks5VJ?}i2-6+SU@$&ue+L?xzP5Ad%D3*AJVYt*OzEwmac%#&OEPjmz zs^-|Io+LTgQU`n}Ttp|V7lUP8TX#zr{@eeG+zs|v{UuHeoV8JWnO-S8w~rQ=R6NC> zmHow~p370Rz7XduO40nKKF&6ohRT9s%fI{qDLbh!;RZuxSAuu~Dm4d$(Khdq_D->j(6Gv)fzT*`u3^BbR z^OvfyBjcyc?c6OYww8($oi7Nrqh%PguohF?)H$`ADJ~dygQL`TsH~od$)`8s*xv(^ zyNAd3 zmx>-6D}FTIAc*H&Qxn~WS?%-v~9=yLzi&jsTyxTac6MQY&y2z!01aEe0(Q|w+yp+^3EBW z`51ELdG> zIRA?)tL_Zt>EUxE=Rk=$Eh}UVYQa-4N~tmIGB^D?#s9Kf)8xxY*}K!oRge4P$jbkpY&3pR?0_w)BXtC(l0WL#g?|odK^@& z#ib_=xn19%({1LmNmdxsdcLQJ)<2evz00QO+RJzt%u(bJ#?bq!Y5m?8rx!Q`fn|g{L zesa%mT1I=#5*9eOqs2-KPAO=?zRkC@Pv$uWeLTh`XC;SZmj$EdNk*JW8I$Jxq|)yX zG#}iRr41&suPNvAYHheFzBylRH)j*o*F58^t9a4dK+)(vf%;kN`15cm2kKi=C0gD; zoV@ttL_5W_UKi>8U0d-cD?>hqjr{Er#t5l{-%y$h-|46E+J80me2=nqVLr_c=}|{n zjf$0xXfN~Mk6R~mN7ivF9Z05~aSBIX+QYPq_lbnF)cdHdsOb^Js5fD3n>U%qeuB-e zxU(i`I2(FD!LdGSeE!ykgI2cTzI<~wPEb(y>T0sxbiQggRxbc7x^ENBeireG{}zTVNnvuQG%nCdBfCnz zjQu$Fy&u8Si@iy`&deU=CNq#{aevF1R2n~*moi$iS9>RFEwSOd_DXym{{wDDmYk|7 zXUcPYnOf0{v-R3=h(mjpuj#;~NE-%+j^smI1?P=9jWY?gsGNBkO8qV(agGJMPyUBx zlFfAdHf%>>6Nk%@N_D74h&(c-v*f%dV&S*>JeJw$(&hD z_`dNDMp>w`*x(IL{kVs72hU+=kJqsMFqF&xuH^%tZ2o5Q4?#z%y zY^63ou`pQP-E%EzTGgDF?on+J-Xi`HO!eh*MiAAyUIC>)Wj^bq1W~v(igUs-fOlqz5PC>hDXtI za3S5@cF}`!X49~b12i7-o5^sbJYSE#iG$FrT{^NJ42J*0Fl^tm1$`rSqTR$j7!mXb_gVxuO33Ht4y9;7QRuERbjZ(BXkN@o+Ula>H>ZY6ecN z?Zn{|)TlgoEQ=<^;nbsJcoCkC0WD-r)_MUZUtJ*?CKmAP{!}!X*awzr4onvPxOT^4 z+E4l=z8G7g-E1XfT=7JmO@9nOCi!!tN22P)F!(j=4;z;eD6;p#x3EqaX7Es`)zpig zQq!>CV+20;S%&^4(IYWpfuARDc<2&j`gUu8;Jo=9^ylk2@NIxWSMycU!FFn z;j-@h@Bhrkji1oI04Zp(a?sL1{-3K3gZk zJX7W{MC`z(`%`hx{32|wR7}J) zY*BYtf#-dv;M9Upc)9PC`bU|QQ>Myv}~HRW*tKb?1D=(Be0@$Rqaf6f5Sz3k9RNu3w%UqGp;)bqbQ zAhWrhP#NV1>rxe5kQ~}$X7dnx{F`tp)WGi=XQ?keji(0+5I(yYclwV;UhyU@-?J4f zo#gYG5{^!Jn_**@hBv9MxkvcR9; z@GR~zYV)omZe|lY4SOb9FT5`{>C}mj?#nUw&@R}7enh-dYne~+LqyM4;D><*#={ZM zY9*6eYW7rzr6VLU3FGt1&|fJYYj_Oj5rcqDU2(YONEEJ-`<1+Vd9QnkZ7VI&cDBrM z8Wo7s&MhUUw>gymj>d}luS9aN5pHad*>QJOxMr9whhyH3doQe# z?CEKvOKA1OKBe*4u|e`^2OFT|U8Bgi)knm`mRMIh0+;TL!z$C8=xXx-`d5;mxgrm1 zP5y&nk776kug2rB80odmM$Pfnc=RU?@pTNolQ;aDY{t1jUmVfNMXx9O@MCZ@=@|&5?WGt_ z>LAa&s3PX=-pp>{30$SRompEuVsR&O$M^~Z$Cxb-`0vDMLiX#jXEpl`dcc* z0cC~is!-m4IF-XYPNe3OSWdn&fzL;K^X_ps2AlNbiuj(Ka@#}^=h{ut$<0FH9=wbV zcPBGZX6jkXGczsDjA<(zsd97=RXgUedwC9{jja@8^34^$<~LKU-xely>p>haX(sJ6 z3pvwz7gMYDG4etSHhB=!RkmWcJ%1Nw+d^nc)Ea(IxXh8!)rKrX%6q7*}}P12{fz*=f1b+y8#3F zp=1+Bb(j3GjD_5OeVtohjL4%X0Z#)l$)8ouy*S z>Q;*Ms{TA+F^~t;rt(d{B?{9v5sE|OMk$_pO<#)rah(-SJ6kGt`&cVhzR*^Lt=CXIT@uSdS?f7IbrpMk9l)ALfo!sQC^Lrb zrIp!Qj=wd5rz1QhD>{fvwcVJaC}Z#D`LuD(p?7dRD{JFee>s&?j#P4BRu20ePN(sL zjU3We{&URa;+k!Yc)ydLhM8=fsG_I|sb}joUl~1pD4VwR;_3xKtbO6a88he7b^9Ei zx-*Yfo4V8KJ9wzpg`Qt$@~lrMsYMRvNSmqLl0TT)%OiQ#D1tjJS2Dt)j8nPlIQyJ|Y86}<+=)S%?RoW|2LD^Vf`f{ZS@V8A>u1RCWY$}{xLo9a4a$mQxBK#Q zyrj|2S&Uvkg@>##8aug&n%| z<&82qPoGi1c2%+T*u9+Rnm90IxCzH<>(a-lj#Ey(VsMjt3{Z1WY#w2#xbgf8?`s+; z-c*_?LOOou{rARvx=e>Ij&0euLpPRh?7*8vcJwgR=9r7B+}7?9Y*I3~L_Lvx=dNbx ziRsix8cge_{yY@3RkENL@T{21MWY9Dc-DAozIUe1zgSM1IhC!)&f%K0Z~RqQ%>b10 z$iN9)7&Mj3cSiECnGsJ6GUeO*8q}Wnhlju4;f?ig*m;ABV*Js6JYn#LOOl-x9Z#Dp zoQtd#ha|W1U&j&LKgpLvT(xO0`*H(*UB#P}Q)usc48>#SA$|Qcw0qP7>=THz^)I<( z(m^(S&7{s)FRnOf#dh(kQYYcX{VuYv`dN$JgXW;T`(P+FpMtJW$8f@dzC7@zJ131* zp><*|7F69t$XcT15_P04>ksw32<}=pj?<6!=Eaji@||tLs?5Kzu#=zlraLcQaAWP0 zp=|TPkaM~W4h^>SguCQ>xcGVE zjPEWu7_Nq^xgVx`Pe7IWRJh8Sz`JO$?3Li;@E4d89K*3!@_EBQD^q8Vt~JmoD3`KZ}=}yKtSyImR~MLF4e1JTZ7HJ!H@LL6-@PNp8(O{;F(C zXURBRh&Ew|NX@j^8DyNd{rozeBibqOI7r;s=*||Hnht%DlT) zg?G<>#)Hy5?A~cOza1XH$D2#g{?&4( z^f6~e)bQl~hXYCm;E-U;AtCG`DezFVZwX=vZl9l1pmBE{bu*Wu6 zE{HN`Ta{`KHaNtoW6Jr{rvl}hlQ2eoJ^oIz;ff+XF4^9c_bR_4B=0aPPM^RCeP;yy zT7c}|c4%pT3w~8$sOY;79hvKAqBs#cXLbUZZ3{={Qao$QS z$oP&S>o4M9=`o}|ABCOvT~R$&`b`T@^TFBO{907OSz$*x%{qls5Bw(>TWjD3PRArPqOG^X36ZtjhK|+jMMYG^TEj8%#k{& zf8QF=Z^=IljHtxDs=id&pvG2To%y|3Rk81A6^HCgW1DqK(o1;-LuD3V!`YsApksye z#XaHt;2wHuWMIy%)5z55gsT0LQGBTz;xAO9ibH*8QQ^rwg1O^}@$12O0aXh%e-R=l4c(Ch{iWR#-e1)V@H{ z*9-75xdxrMc-90b(DvReUj2QDNtZYCLgjugdA^yST$i)lOtMAyb?04W6()ITvvKYu z=|fXP?|GNSp9VDs40{aSzouMaQpvzsDb$)BMxQ=M&@Ji&QkD`JV!}6;`b=>7g;DXE zO#kx*8U}X|+V}|PTfM}KGY7FdHJHzohw!Mk)Lp z-^c96u{6?e$%ueosPFk0J^kJArKunCFDpqlLWwXMy<23s7_h_V-)QD~12g`c%wdMT z+5e9=N7;Bt54kssB2S~mX>XjmF#tzf+!aEj2`6`dh{Tjobn3kpHaYpo7@2~?Cygk7 zT8|ZfD>16z3d+wPg7TiZFcewPc=rhhVjsdi@um2xSt#y>Bn!J~x1kra5`Ry{V(LOC z$;`In@b=|Mm?Za<|6bu$&Q8oRTggG?w%p~Dj!@|voACT5S}fP)jzf}FZMhiv`7XHc za}!zbr8ak52)bWXqFQ(nvR@TrOv-gtYQO z{Bt+-%9sJWxoz?Fs5NXPZ)ZSlIL=i$AXQC|_ky%|E=q}K+Y4-U-!B?Glf}E~t$26% zNxU7FjjMJqC0FCRSpWKr*t<}{tiq=3df_M5XL*wQwD`CE7lfe|BSJsH?A{;f@*Qs8 zY{k%o8<^Lz2|w7rhNGtv{~Y*;<#PT`?H@QiMqT!0@8Va9ChcWynO*6G&gaaL{Ko_@ zrnQ0Fw`by&eFUu9b%k+z=>gp?oeZy5BT%UWtilpdl+Xc_U0dU{hBKaiC=sLQWQzs6 zhA{7KJ(ixk3gsm;*z~OxqnDZSO|c&Lr@x2ywzCK*FNOL+sgbW=gHqQ|NRl(t&?zTT z;IDsSqv;V5veJ@m^H_erS>NLHBOneUPX9NhKWY zwTI-aj^P3m2VNMX%mX>6p%Wj5@Izhj*myT|_O6F&W;$X=m!j!^6R_UZ1NqilF)7v? zd)hXN!2W?~yW9@lTOAN1yWYfkojX|VcL0~IttIoe3U}6AL#)?osY&`Hs$2umZ~Gez z3u#K1;|H8(|jYyaX2Ob?XIwE>rL*CtzP2Hf$hyc2#)=AH7( zNenqLfgMi^t_!U~>X!`kDa*ukcQx2=ZH3Mg|A{Vl_Tbr1o91ge-iKy^DgBV2s@I8cvQXe0;kC2|h*)UL@x^nx@pBDZMYh zqFJI7#onjvIAntbEfN}VC!sf5#x#SS*K6_d-6ljD7GjBH#Jrw=5Pcg?z;1FOG|zWK zfsY|RDy!mt(kVP2{{*i(-N&DQH+iJTeLkqG=k{-vnD_3c)UPCBe%5YW{+)#IqILNG zP-@DzpXclq$+WOehEJa&tO#C-v|q=VH?EA@7Rhw37>!Hq`a-Kn69#30s65#LyJj{+ zr=Y)dZ#>Kv?PNyEm=zp1U<4PuRnR4+4G&(ug$Y^bF>|~zKMycs-AZ+at+C?ieJ1kG zZp_nO8{yi{A43ZisPc{BsvmJoS~!H6j$X*!qmA8*J7H984`yq0qI(5+siz-LF789k zG!LGs*vI%Y2^<}e$T1HmbN+sB*<1J@hVM zYW;K6%KiLvsyaKx+(lEb&lqMec?70=c~W@?XZno6fJJ_|xUUba!*=0}NBq>pX4Ge11gz0Pdq1l-<=)8`@4_Vi)C=vWpFPYzuBf0aA11%R%q^`G|r+xF} zjF$b_aAy?66Eh6#XNMI54)B(9=piX?IR3X>M2^yghhtN`jF)`!q)QmB_Xq`+{_xZo zjq4}MgofsQOwG6r<@b^q-o^#vzBY)?&OI@Fi$9j$UJca?a}jSV*MQukwJ$hwk)OPa z1|GsD|9Lp`Msfh2kA{2C?zmUs!>uzG@O<0^CN0^>XTx;)=B8lOpIz+#<}l}EY~$X8 zr}Vrrztu}aP&@BSLahYzN3 zu5SS69DR$(R^L(e_$dkoe_@1sC6Apt%kOfQSlqmdZU3C&@-^#tz$$>VcK6}A6?^z? z@*(Oyj+T9@*Eru!o!<_=!0@~bTF*O7>p$Dsb@E=e|96-EulCZ#*Pd$@%X8*J0}cjs zpwex1TFQIAhNrZSU1-aXUt4g%;mtf%x{)*Tqo}9eO_86{LoqzSOcAttHC;O{=ZJ@{ zTpcR2ymoh>Ym^-a4vVJB><~^ij%3hvHHF=)W{T0Mk{KHfOrO@k1(P1|UV-#bRHm@Q z>c!Mc{f%A?FL8G2P4tpp4KK5q{I$}V4f}iWOK?vnr?;W5JR>^4X~UIk z4P=sW2z^5*^TmwjipNg>aZ$uyE_F7hzW&Sv~DyDO6;&3VbIpQ7DZSH;Sz z_6p^*dl@h_lcjliJo!m#%DSd8L3I{WF0AL{FEKo^V1S%A-e)iW8?sNak8L8GP%+~> z^j)6en4_8EP*Od|xxMGOO_~e|)u!ETHID4BqFB>-gPyIfF=xa}zQ25&Y04+~?piO! zRbNja3hUn-c)(K%sPx3PNz>EJoU+huO1I(t2ifCxJg#H z-)x=?nl5MQW4Nr#VHP-KGjeGvS0^Mh=wbqYZ=c9U1Id*670Pukq1=sQ{?Pl9n(z~lRO3oStmrk0)^mePMzERH8o{wR$st@NT z_T#~UZCSL@h%KZxYUV6k{^>u2&W=)3cui_~pD(17(-dB+6KuTNkKey2xMi@9%#j+w z%N`4<*TkDcgC#3b-j5aSy2#A5PF(D&U`p*LzT1(@O}*23Z&Cv{4!glad&^n=CW3D- z&z6}N-n3bMi#q9#_|p6+kDD0MXmeZU4;n(7tq$_IK)BVU!!UH11{x9z0xb<|SC%~n+m-Mfn)zr?e*%n{2y z)`~kMo2_W(dqhdP$EvmDwIgkKqRmt6*!&fjLTOyOa1Be&F@S9 zhmP#-WyY6>Y-r!54QGxs=TUPdzOB=dnZCAMKfNj6+m@~jBqaW7hYGu|m~ovIQ8j}K-=rMYBU-N!$<$Da;y!|#55;QObWm{R@8 z`CqlI&_-7=ZBv7D{*g%{u(O%4HXbZI+E|L($vXTJZNN_M8f@3aLd@~g5x>-$iTYhb zg+ghvP|eR0`N5k--zgi!sgUVn?=eTZ+WtkfA_?g))3~hvdQREAfF8;*F#79?{?j&q z)rG*Sg;=(_E&SfU#o$T0yl}1^KTq_BW1s1GsCE-Y9YZ-*HI9B*a>yMq-b?Vke+G2(#-k>{L5v${ zD&hx?5Fy>9&ZW;I#J~Fni>GSvD*q}*cuF40t!(rayYbT91@GP&ikG#HVoEGT$rCqm zq*D_S@bSB|N5?n#c}<65e+(GEua|gMFjwsHo+WPd)D%_y^u>AYj^fUvK(R8qwdk!h zQrNcaD_pF+MAM=W!808YW6=U#KX<|R1^X~-W*9!)osM&JLJ)jI8GBYo;VEPd5Y`TF zE-Ro}XoS^6F9@;hswhiJ#FOCG@K*Fe*D+R{xk}ch8kZ3{w*}i>Z9)CiY`pEzNo3TT ziRbH0L_p+MwC-8~A35{ST~{P7AJ{8S_BbkrH%vm?o*U6J<*XPt*%+@s=)-rW6*8;U z#pEM1#QWQ!qRh<|?NVDIx2Y*?i~qub=g=YL1k}}gv)|hnaIk4X^`K`cT67S5I_$-R zbyLLNHg3ZEdZZYx+D(j_uPz!3>_j27#O@#y(aNT|SmKs}ShWqfuQw2SGvdTGpCuxE z+a9qkY#iR@gh1`B30iMhD85<;iUX%ciYrG{ux?7Z$O!ujYfOLjh_%(a`f*`RI1&5wHU+9;{Xm^h!ioIRM);K`pF z!SI#dhO3DK+1QXNJfx0(;@*Qe>mQ2nExGc$zKu#PH)b5Qr|TzwYAMTn;I&$u_&l67 z-FibUU^DEc&wJPXVmPh6jHGV=Q2%WR!XoWqJShX}7sA+S_Xc_@yU6oi_9xsnBKhHE zwAQpmwQVnKwV8%4xkN*Xd@gP5cr&^J8`pK@uCyfjO?)HF%!4pw)CqWcC*a7EdaV1} zk>iI?!rTKr5LV}dpR0!zsleQ07|kAc5>@tV8?{Y+lPJ9m#z%`^aBiBHh{stSkm4xS|y z!sX6I^tq*xEB_rHZ=BLLVfuUTV*_{{RtE^XYIt3%PW0^<|Uk_(LWzwV& zlfN(Jjpfgok?oBy*RzoFO^u!3@5f1*>FHZP8Dq|W7B04VP^yty>6=7ZLMYDdxCv8T zN9vr;rrMWM_I=w1Hf518?Y|eTe(FfR_BtGtdAa?vI>W(Pk9imJcx{O#=g#lO1nD1h{)WE*^-HlzWLh&lZl-qyAawmGyXoV-Y zw4cP>j!iJ&`b?}iaUJR(ABcJ*fqAoDqb96YW_1)|OXW9YoY6(;swL>Y`~w;me-Zh6 zC0jMalUG#xp}@Wz8``{rYrV`2?O%uaZyWLXz*>2CKZbKMQ)g`WC^)Waz{P(U*E}j2=u*xn%^@p!fp?)uGoRq z-gRR8%{a_?whbNDEE9cxg=6Ps$)&z^91X$xG&*j@xyKcptJ#*OkEMSbN5tKmdeT2T z4@bsG4UcC8_8t5UolV-%ZX;Q5@pJe+M0$w+6(DU$BVIjm#%j$9Q}($`YNeuu)<9h zFML&+k54CcFtvrF)PVFs>cRVnb4JgX0hF_~POnu|zVY$FG=5{|WshN3{nnbWYQC=UfaIl11IS78O>fFmmTXct&kP zuTj_0ENdt3oz-Gl-$zVn{+rJ0Kk>_XOC~jM!iMXrysxn2f{HA(f1}U4^{Tx2yD1Hf znz7Dp61=)h!I7XbSa3%&NiQ7b*;!lIZc7U4!w<@A%R+p3C3`6sH(=;l6I9n9ME_Hg zr!ZtCzIBP^=+Tk9Qk>3?b1dMJ zw*y^m_h71hKdida9b;bBUjF z)7okU&5^OR^LEocw(6HQtu1k*4DKvv`^a@zg|1bt0&cf5A zA`Gt!L)#-$r5<|%KK2jgfVh$LIn zSxz-ys*T0x<-?J_Y$To?Sc*T>d!fYycYJAk4#ofGVcY&~7+d27FUL69vvrc0=|5n* zz5=^;yg~cV`kbO@z;P9In#w-sr9T$zq;1Rx)BV_L^CY^>UPlk-cwD}aEExbEIJ7Yv z?K;ZYk?BsHk-Sysyl+@P&X`_GTcw@gp?Q7_+r=^cm6qGcAz?9T6D*Jsb_wW ztHSXqb$Dfa2I~ghz|WT#F=UYvJ@$OU`5toi{c#S?ja!VEcd^*_GaGXT?}wVP^q!3F zhTV~#_@;5$4iraJiP1y^(Z3T$Y?gLzi#+c=8c%ZiqYqywiy7b!&>M4JA^B4H& zoWXq;!#VPnC1pHQIE*%WV&X2M~F z2S1(sDPH?L6tym!Fw(j}^ltG)c)W4KnbjtIHl{TP^|oP%`w4ODtt$%eIZ(*@pyiZ6 z^wwO9v*&)G_DKoKjy}Xr_pc(e?N{+8q&-%*l(}h3U*O`iIy~HA!X}b0xJr9Ii)Ne_ zmoFN^SH)d^C%!Wn`{G$i9GzaR43HNv0>2=M07Pr%H_dwk3K5`4PqYaH4p z%)q`0PVk;7(5Bo7;qmHl4jqOv3E@A(m++4n=OJh<0 z#8+k&9Yn2n7o4Lpu4QjW;b}`OHt&eg=n1%eZiQH_`dYk^x{}Un@-Fm%I6UnjYJ}7a zr$wN7^fqX|H^Ye=ZIIVD4SI_cgl+42;-{P;{B|kAHT61KmpO8CmrQYbi5b=&Q{}kI zCeUfAif#XSqs^O^sBme9oCG7;7cqhE&Svn7_$Tg!7m0Tt_6oNFaWL8$ghiok5xo9B zE+^RumH6YLFxVeI9?w9Y*&EU9g+1mf?Ggqfo`|%kJ&>~N6f&you_xM=fgSFOiLN>r zonMTZr#s?J*5U!a%t)+P4ehUjFOfzdk{a?Y-IUN`+je z`rJ9D8S`^*NdC$kwti8z+0lbYokG zMiE`6i>jfmpe5(3etl~2^I~gu)(OBe4|5EyUxx|pZJ{!)Ty)Z}6Z;$opmc6)ylSrx zr%&c6c@Th;CuhRbPMrq#y0Z9`VB&9I#&{iPlxiI{T@7el@eP&I#nE#jzI(N}5!o86)s1;(!8;f@TC$T>6f^dQ z@k{S&+N){ujdDjGX(joqy(O>6Tc?J5Aoyy^aODOnB?TWbW`ej16B~a$@5Ymd{$q(UFln@?{$9 z`ne&#`V6XleqrL{ozU{Xjwn?lD&H(eVYv=X96RyAlR0QD*}5-|NZoU_CG|e_=B<)s zj=&>4-gX%;M~q_Z;%8{qT%LQMhcIxCHuvKPJdSne>K@8`zkV>UZ;q0_`5|c79)>|@ zw!v_IEC!m7N9Psku-vB4BeGAoM9vXMRZA^>V-d`z?7`QXDRAu66z5Oh7j0JUfMvoy zs9bDvjhHCt8#xo zYsSfYv}V6;i0d~V1Ln#8Rems1qi;fIUoHl9%*LjYBgprAg@wn8v2ag0_HUIsp9C8i zm5fHzyuRpOp~O|X-FV!zDNik(hBZs)Ab4U3*1p!}T4PI&Qr6+d)+UHdZ-KxT&Cudy zFOIv>jp5Tau=&oln6Wq!E!*j^{)ii`r}^=%#Rd*dbiuC38zO6BHw4$r!Yktq7|=<2 z**||0r{Ag}ct~4pXnq@)A~kuyMVq4KLHv>XUyCwJZW5=_d5jhpU+h7@e)q(sXP1R} zMi8zpQs(xSXOLh!6Kv>TrD7RRYf9nC|$(8so zavQe&$;G$yPBgCBz^W(5>D)9P1u_3brPFV`*3{tF*>}+9z&DsKx`5sjQ?OWOI5|4I z;O(_hxX|E?q>*ZjTHK#rvu4pDt^@iitD`7x8H`gq;ou&9?5uAg`I_5tQ|l6Vm*BQC938j#!BO6I>fS_4zfLcV=_j>5Ipc6^ygkg?8bGt=Cl<)h(8GQJ zXWXyCwb^r!wS5tntnri_KwZpDJd0U2bMd|FB!tI2!`%8}Tqt`dv*x=p`pYT!-3da; zt&8X?dpqA-&VZ-85-qNqpvoZvb>Gx!Yi)yH*|lQ+c5|$wam6%H-1 z^vMStnkDPSuTA;UT$#O=s@FqfGHidRuaHev1LyPKb{6Q7AcCBXYggil`owac|ZD zES3C`tX*HwRP_MvtbdGOgL+VP?KR2pIxn;xUy28%JC|d>Y)g@P9Ka;CJpEicooH4j=*$?Dhf6>#YJh6NI2OZAv4E7bFBdlmN`nF+)zxP z^8%H39jIDnAT`tGe39*h_w%~phSW4%j_fAy&`y&3WPvBb7DeSB#DEt>x^fzPdas6u z>rE_hzJ}zvNpe5T$BbB6i~7kmPsx2M@`R?PIfC{5(CLaAp8scn>Ufz4q_q+&OlKf@ z)&uNbAp16vaflha6Dl>kF}!;?vd=ES1~XMS{M;etx$nj|B{Qk*vBlG}tG z$@1DXPU-mr1%)-xku!gbaf>j@It|yow%gUwT>@^)KaDpyUTqiH(g$E$O!$!|2> zWX7VxH)w7>ly0AN@IQ{u!=LK+4dX;e%2r84R8&?u_jMnN29?s6mP(}^rLDbCD1?Sl zRx(mbl$}IF+NGhPlvLW$Ac^1e`uzi59Orz_^W67!z2Db86yP?yCt6G1!StOb{~2^o zuxxkpk+qX~>15%p4k@Q?PAZVeT!8P-#a$s;ux;uN!*b?Clw9#Z;^QeW*45zr!d2{Q zYRRb;J@`=3X5OH_jW=C+3s)yC_Bf@+(R0F(-}yBz#$80_w2|CX@Uah;81WI=X&m%& z6#vju=e2rzY@|Dwi(O^;qGq3Oqc)6{*E$JMYc9h z!;7*s;quFb+{R06x@9ZZ1?=UNBO-IudQNo6eqfVzHMn2ID0L}sXmsWZCmlX?{-fYBDss1vQvTYh72K%H zi7QM+_t=6xi_O@l-x_S4;fCN}VMw2v!rh7wbHj;D9-$T?J_pN%Z^0i;c1kQ0;l}21 z!+A|F;Xe#&#NO4)Jn4=fOPF%{GetkCExt_aEZhDR~YV^%5QzvIV;<4fC#E!I1_<(alfBijT7VzVmtHT0g^JYdJm_djcc; zWjJBQFDUD%uvx<=XlwT7hjA(#MzY*xeP?bfS;C=reYkhxZvJs&2abtnNX6{KP#RMM z`H}K`(Ypk-=aP_GS&n~+hcR>XR7|wljhhj-$+mYY{J)7F=-EQBY!>!CtVdgbV9y*9 z9oadpxOYL5YXc2fZ@DI4JTZ))>IR}UEFTV29$`)74P3Z?3aXLiB5Nq&l7XqLzH}yE z=)8^91am zAGKmmJm=ghq{%wsS6^jrQ~is;ce-pNEkILg0OpBja$>y>cS^Qkvtc@%Su>ZtMds^G zcnPoXDt006E^N0;%IDk0;DxxatX!{-)vE$K+1iL?fr?`z$JXY?3zk-o9ry^7^FVDiYU>*lF zPGV!dc#gPQgRReEA;0(rx_HEj`QtR~SRM(p^k+!4cnsB8C0-b>&W;jCZno&cvb%)) zXY)L^iF6h-b6Kohs0`VmhM=^ElK7QM^mjvxq}{<9H!5tQzeo6a@7H02ODaFP-Icfd z4x?pLXOrnSCd-JYbiMUDO;ZWN@y@kWclSDFcJGc5iw-I%t0B!{q7!?t4l6xm`1s%6 z+_@=^vRCe*Q|lkm)A|&;8g`dHCpA%5gBIH5R6}{MV`)wKe^hQ8Olm&G)UwEfelFWe z3xAu_^+qpbC(p&bmy8)pRB$W5lA5Mnq=wcMdMo#lJe_RetHb1Ax`BH1aG-b{Gg#Od z<3!s>dL}DD-_VhTZ+ZIswt~W1tD`8uFg1&-y)O#SSN1ylG6JwDJC7cwT` z*|J`Ar|Wx3gM%N1EH$KF<|k;uG#&bTIqq$`Bvf?m`PXDo|8b?WBUWaw+nOtWK|I#B*8ex-k5P*7abYbsJ(LfkknW}QAbT%awwcoR7ptjXvl_+68I$Eic{2HHOUqmG zXkM!x>Gdq7#esvd^IZVVd=UZf;maucOgi-)Gm)OnJx71V|HC{Xmln*Oilh^&2s1eh z60FME2W8RzO^T?^-Rb-teain6K*KZJDZ0`KZ`Lbd{DCnPq`I0uAJidjWgB!7dujs_ z8L}=;q}H$p^n{mSQ`KUM3NNH96Nb~=xaF95W*q$O450a~KXQ6h)BI%yVwWk4%R5VH zIV)r3>tQeoQ4+4*fpExPLB6%K>C>B?bYYeFdxGcERZm5l7A=d!^dRbSX)IJK`@_w_ z2$Y|Q&$o1Gb%7NfU;JBg*5H@qd9yQ}=vYh@xsDVwdJ-MT9YMR+?IXLQ6IAYAK(Cj@ zlK+B8bh+J33$|XSE<=i_ZCwm$Tg1__W2b2KmcOK8wu&Z)UnD=<+`@b>+VkXdRU z>~I&Hy;sfjXzWk=nqopFJ?)6}CegXYr8GNqFWs*FPDz{1Xyc0kbokUXx;tPF-fG-} z$;dn`IPjGA_mf4^OnFqeyd^t|rx%+%N&j>;Jq)@|a^-huUw%7%HLNE6bzZP*VfdxL zCX*{&Vd(Xd_8lHPx^O4GbpQPcp44}e~Y5TJ!l;s*liJpDn zwrvxPAH1iBm%owiy#gv&y^mU38YGKv^rPB#73#TKLi-X9N^S=^)AzT#Xs*{_O3N>% z5E{pQrWx{(vN0@qGxK<4*ZTv{>-Xb^9(u$xWIlf#tf!C^A@|tN2c+QmP^kSwvS*@)n z??I}9Ay!9SCOKf@m_`cu`i5@R4uRvSk!(3zhu5nvEaPZp4DX_2YZTnoyTUj6#L*Q7EU}ZSL|*)2Vg_pO1NHM!rPvW*s-V5RVYAF7j;I8f*!NbkOKKJk^ zaeuJoD@QtTDY!ptdsgA2MFDCj-NuB>1$_8{A$Pjw#K!MDxS)!8jdU&_4c!P6r2)`> z*^ZqLWa^$W$_EC?fJ$Fkq6 z;XHA?Hm5cVCjaT->= z#jQ!9e#f5Bf`vs0cQVJ5!4sgYq02Q2ZRk+@g0y%YOl@*P=Vfc*QWl7zr`;iFz{u&T z#4+1T5s|LMvAMh1FmNG1zaaSQb$jUHxQn#q`e*XVlEqKyM|!ZQD>ikmqT3^*Xy(rA zbkC@Qd}qnSC%23ShxWn;S0h|sV+G4-J48GW!nOPEcq#iFGQ;Z8P@j$$^JIAC&>sBe zeigG$^@8!dqUU%XIMj~Ae@a>R$#hCm7sc_sRPA5n5%R8z{3A9Y3gXAblWAW zF)1XQxw|RtS0(~B-^0k{q4-F@}@vZ5AY6)z+>&b6#h4cG_U4~Gu(NSS&>beQ;W&A z?Hx7MN6`Ekb#z0aD{F`yY($9+TmH4>S(}HUR>qo}UtU4lxk|zI2*L4`t%$vQ8O0}0 zVXn<33=vtcPhyW&9CQ;_5utdn=pt5R>fm0x1

~cJtTXUgVE;+QPoV2$=0wg$v%=2=f&p=(6qxrWRo;%$s+P z6h-IRsR#qX`B@ZhpDFsMw~l63R`2oNt0r81Zy9tCccLC8OW~@Lu`moyv+RgC7XFZ! zO5P*>XsQ->Bccvnlow(DDXDPKCZ8sH=Am2qMRwNVlk{WIBYI$3gAX*SaI4yR`ZaA6 zXOOiV;+hAM*+3(*T|F4{)?MZwcx)z*IeCzr(wm+YzQ?}PohiVt9>y!oWJP(MkeH!C zuMSpoicR|%iNv~Ni;^(Sc22NGH`e59QWh!R* zl!K#HJ{$2!jg9U-3rg2trW=(Ls%L(-5XKaoVE6y+W?Gp>tkLTLRgP`M>Zr{^Oyy-1 zet*J$Va_;PTh}pVHi_(!F|>J?0k!yvtlO!5QL=I>HBT(1eraduZ&4-t`FINXB`67> zu0F>h>%WNH?Y@p%9%$odA&G8}s^oqAO2K;Wbg`$O3?YN#@x`k!*tPqnc+Td>mENj@ zU;dHgoFn?`7X?tgsd#=L>WoPpx$q#e7DL1C(ihdcymYdkF!Hb&7}OWTZPi$5&)`7T zdP0d+3^>9UYQ(YFS;hRkuuw|gC1GV1PSB@#DAVy2eYklgG$+>Ww%a+(>; z-E737?*iQ_xrJ%>_p@0tsVr#BZ$5Z>0p@=xqX#eIsN+K_%8Y-DCse*eKT{n}W{2Ws9KNOvZKUe=3$8931jL1p~MF=TA z_q;WpXbJCzh^v-H{B(@ zzN8n6Ehz_u6`8Q?_H>;62>9~yah&6f8p+T}yKqKHE0*0~gGYkWImIaj{FY}$`1^Y% zZqS=U7BVC7p7A7D^|={F3n$_2e|a#=_$|DnHjJ1#N>E#9B$%dVb7x!pxV^8la7ZQK z*A!d+;_Ww_cHgBK)~grVzP87#TnW_bBuR?n^x#3jA=tm`hd8IyKrac${D)b=cUw8o zdoj!<{2Z}g#jc=KKN^PIKMZ`&b~IZq!;gMt!RgMp%ip}a6;6wNZ{cGaMSr6wzqBrj z8`zx)()1%3(6);;xG@&H?L3+(4MMjZH+rQ#2BvNHg;SLtfMX_N!t6shHP;X82YSQd z<7p5*VF656=6U|`cJ^}DQ|?^A4!YUP5>@vPpniY8pm1d%&HZUgKObC1o04QksQqMwf0TZ!k)f_#nzZky$OAaEfqrHwV%-cwdZcp>dM56NX(4MkrIoFaWK<6t z|Gq9Z#Q}1W{#i(#SVaI4NQb7dZEm^y-k=6~>E-4NE~Xf}l2Ac951$!CpyJI#+}R9cXp+f6kJbCJb&D>(n{x=Qji=D; zjQO~9ha-PI=z&D}`&bO`bY!hw)7aRh>!72ni0i+59V*Ni0$yhl;LV?FlHuzvb3>g6 zuq_ul;f>F1Ox7vE#72KCx1L1PUK-N-Z(^pjB^#5tQDAga2SO_q#d$d$AMD(Z!F_ha zZ&6PQP}88Z{3LFNI0Gr~ULeis)1R6{u8@P+b$f{mB?(i%RDGfxJS6+^R3wKFcM-aT zc!F2yQhfKYka(9ouKCzq$PHe+HgP z=ECUvnoJhL#f-9!-)%`pdG9V1c*$6n#W*c=)bH{6Sf6)Jj3dbku zP-V;p+_Fym_E?W2V{;WuNLYm{bHi{Xw9>=80r2NiA~^ddgU`_mu&cX-j_nvvVb&+` z;j>z9rO!QX@>nOxd2}+p^kv(XSca1adM zvd-0&u3!9EvVAN2TzgoD7*sm@pib^v6J z`f1$TrK&=mJ*%^HF}>KA^{TXGhc*>4`r?d-;#6^;sC%5lZm7#M6?@fWV28;Mw!QuY zQ!x!@1M*yif9K+b$SbQsb-q5g=SeW_4&&+7KVQV{CR~f(8ScTrHu}?Z9h0l&$f)u) zSv|3&Rg)((Lm`&cZ@!P8QjXH@7(KE&bdbOETZ{Zp*TD_zI_`X>3~$))0lp8GN)wOu z;q#-^&^w}(e`V_g_OAw!x47?aeS3rA&OAWQQI5adc7s3pEd>9YvyQ?V6EN`hF#0%D zmG+I&#*2F%;bg<%%&kuxICPwbE$`yk=|FiG^RP_z7dbrG;=%c+-a^~XG`=Ly4lEr$ ziJf(OsV?80emv4+mBJ$6_tm5EkQ;R4Ke4Y%_A1Da(h_P@&G?LYz%&m}MQ=B;Gjjb$ z$gk99F0ZWF{untXvrmtUwhM(G%|_BEkD`U`1}2T#GjpNaDnkALqg;e=~jXw;Gj7TH_W16$%?`nv$OW#DNt zZ`Gi!V|uaq`TMAQxGVGNU%})~1k$@>dsye(Om?JYCQj7~VJ~)Du_15QgM4ls=Bf6B zEBc#6&E105pWVzAsaP{-Is)%(qAAHkThQ@Sq~LrpR~|or1w1c;XKrHG@_7Yn`C=e^ zmX&3bzvvTP8Ntn8R?0inTQH5|A33GDrQDBkO|(-fgVMWj zvhU%A|3;=LvTc%dM$xHWPbo!xB}NVh6M!p@;#4S~`?+Q+uh^_jPaepK+_EJ8{j1kFv-_0TrIamtWuL+PMVrCEYXNOk zSEGVJH(XHgkgI;Yko)L%7nEh>aQR9{GQ6>A%#W zxU4^S!8Z+yEv`vB*YsoaR)<37hyi$O)n0hlON}f2r6KuRVa$wY$*?(z+H6UECG9?H zB;0BG1OM_~(+c%GH1acL!DqkoD=RY4WQG#m3OtWCOI+x#ld3Sk|4>0S#6s90`X}5j zH}YdO88~k?<05r~;r3b|@NV%3$1hj;$9jFqP1Bw4dHRJaDtQ_*{Wcbl*A)`{BFS%x zGQ}qSg^KMFq_$Xv)Ykjp@Uhn*!$prBdFSptpfsNAG=9moq{gwsDWllP6T7MYuph*o zlc|u~Ao?1LUy#+dboTq3ExF05kVqinnp@J);M@<%ztSX}IVb{OD`nH{)zhi>oqhCu zmI{?yp5m;Zm5O{;HI^*yEQe_mteq(W@+ zcc;xR`Fvd9Qk=Lm1ocY2(ECjZsRovz-|wOP8#6WbU+?{rc`foRWz7hRO9%n$_vQSj zRkI~|k}l4zQ~_D75)JsLA?`?wFx4p#Uk}|!X&_7LB??qCd=-WUJK~npZ@C|jW#RcQ z6W089G9(+GrCz?5p=Of~E1q{AW-jSX-<>+0r^SgoEWN#ae&l;7*C>GWG*y~hKTf=h z&jn9QSt#55kN>Ztglq1{G0Qk>X1H=Wv@YI@TXo)p^x|JQa9@Rmlt1UsZM#gVyb<{O6%8-Ja}@6Fc&8mvuMy=Fdf}y|f?s`ReE$ zZ%tlfE$PUBK@>gbIu43|!;Y+pz*msKRUIhf*Z!Hy2d|BRbvGI~x7q@5@|{R~H7muf z%Mom{PebeC#~9r13ADh6xz64UZf>(HB$~FuuuVTfzVt0fOkd!YM^=>S3bacv8q~vV zKzdUmmFXG>i*?TMLqyiZ%BB4=b5Rt&&ss?NvVGX4rC(sm#wDzLwhB`n!@%699#(ih z!(l_3Bp3Q9a%-y(g3vh$U3;fO#F=dDH6;MFmPur=e%JM#)nMBBU3u0r**Gk)DH*xw9JD49V+GZ{M-2#qZ%=}>m7!UZ{Yqd zJqjbHo`IFawPEh<<@otRBaU8QN~Yrda9~V3DIG7vP>a|k1|DuCQbT_f6&TLjJ zbB8~;bqG`TeF^`Cu7>QQqkQBWfxjNMO7upj@Waku<0Pw>(wo#i)F+XUv+%4)(GWR=5eQWEBUI=6HxZPh0x{Q8$7g^u%=^ zVU({Q{dv}lo86X3DFy?mm&QGAb?!5)`Zb9zKItbdTVxACb|cZiT!raw{RF1_Zh~&f z4Q|1!NLIQukvxkA3K|l1u>BlMBa9R1x2Wa*`02>bS(!4eUSBZB>^w%A=VSdpS3&tp z53_4(WD-LQ>h>waHz8>dnjJ0nUo4ooBiChHKx^j9^oN)G`u z``O_9%3@q!w*Z%Rxl-k0j%LcwrGT?N_^I$FE;CL6-F~GU4iADC&xi8K1swi;?gq08 zA98;7>$rE*D)}uR%E0VQ8!Qud(mDyh_(X$9)PFJ#U+(NfGj+xa{|<%I>@#9_gyK;e zT@uLqE_%sNa(Tk5c`CCn5BFo{5^vNQ^AN8IahDoic%se1`@F%jAf94#_@&7=pm+9v z!lGRf7^Sn3b&dGTsrS9cE2)V(is--Iv|Zq%zqIi4#)a|+d&a{_KTB3=kjy{1wU19~ zj>mNk3Giiy4V&Qoj`wNlMvW?O{)N*?X8l7;*og7yk#tpJU^4@Ljnrn#{#No8#s~3^ z)mSnZbdj&_2&I?%oni98Cc0}|MgGV7V!Kxr=q#|}BU0pq^7s&Pcw)oGzL2uf+uyQn z6D9PpTuv~s{zU6+rgQfW^E7GRW?@{}6qm}zVJ9o z>~el6)=o*qU|(4QKMI(1Sb@?si`ez6lUSe9$!z@_6?nBNNT{5f!0bHbK{6u`$He!d zUW-1X!3bUa?~xp}dCeB)b-bXK5v5$j+!P%1eL7Q-m*tKvkP}Akx}hs=@owb3v7&D} z6c5iumpf{L_R|9xGD||?1FdMIq7AJ%8b!B-|-t@`&<`e~A9@ROS->~5*#ma0F)S2kj=TF4NW`}>kyz{tR*Uz0ld zZ#+Zae%DB3D(%?6(RplT(^h8m{03+2U;zDtQgG?cNKS5D5DRcV&pjPIh>c2<63tQ) zezeQ8;d5-L)}kBv$U2F(-6ZOmY)ct(%Mfq=;@51~6WL0C;Au}d>q21u=`mt>w|K_s@*z_8|*OTsoYJ_wHe3 z`Z?^L?>QJPYIBLBF4LBEb~HGw8%?k5#051wS)k}2C=xr~#uTr@@{+wQ>tZc)xxSOt zDIEof$e`aPbvQv%SUA-d}Y|aol2|>qLumt5FEt{ZE%G z99zI?j6TLrUkYYB-w)@CLSt}-?g-A*x`UrFQ}n5wMBuF!a&Fdo;u$akuZ-4Y<@e`- z%$RK~Ai#tL%yNKjTU}uIu3oSyP>YSss^;o0uYd#nZ$VU25+rHZp!KF9?AzjbT-Uk_ z!1Zh3Jk>5k^-N`!QFtA7CBL|c3EHeA3~|c$4jfh3o6^G4d9JTcg#ou18pW;?ee3&Z zyP-#FFD#_F^E)J7zEMyy%Z@EioyiSxipR5@I+#6GrP5WS`9CXV@pR-HiJZ9(8z>XO z-E7ik)k`jLpIv5Cqy1tGmUe+6%VfC)IZSoOTef?f51*Wxfp_1VVd#cNEbi?>`nqwn zTmHD19gC&=#v3Rp?ksPRJDS|L&mikI4Pn40iI8|pLwNLXGH?TjL1>>JlDaDy{Gtm@ z(s#3}r7te&;;wU_aItqO+I$~E^=T=@U5_J$6W^pOf87MrwGq%+V8g~-Q)Grg6;A1z zI@FsjLm4mPs@~Rf0io-0^Y3TquiJu`SM=r-mPNvcnsi=o`CS_JtBwASoX=(jYViqy zvnkAu2`d}ogde7dY2w*@$a(gH`#NtQh#MM69QuF`ZQBA`hI84|iP|jSV?Ee6lwvc9Y&6eM?HuJ=Qjv$qTu=Ff+3Co-_1ekEVw9>AF?T;>gJ zl<=S15Z>@i3x4ri4Euk2L7%HDSwM0kJY9PjR;|!+8L82LCcD?u`SohdZmTSuXdi<4 z4W9VAsUfL7B0RbJ zBkfaF!l0}{*t_f)pROHK@oI`YZoBye{r(j}@k=Q`=wt!6;m$CW_nHIWlmyXhaZ}>3 zp+CPAr8r9e6u6Z8!5ij}Dfy-lF!=(vvgjtRs0A45IScaVWOL1#Ui4*xIW-<@#6|Pt z!D#y!iVnHXKM2*wnuOOJ|GOHN?z$trYBdiZj`u@>^e7kiIF>%%ACVK_f?Tf_l%p^QeQ^O6|A`43Ud-I1@DAT-b zk*h6{A*Ji9@!ykde#`wDPWEaCPX0Sgc=vg*aKB5NMTU%m;>($6sZ&UoN8cdV<#jas z(|gb}ivwMs7|G3bgUC!(+oiy&nq{j-N>Yt>LWpuN+SNV@*2H|^rVUHS#xrO6TUQRk z$m71>%ZBp%xRS=`_oVEOh9SI!ZH*2=L*D+jRkN(W)PVK!uKo(OV|`nap-5b=U7 zy}s{9D)TSl=;NK74;TqaL$8DT+J6Rxjx?i7#v0Vat6|KqUZ|2Bi9N^E@nHXxymr=MFq=0Xf@&gpqn!dg z3*3#tiZkh-B4awO%IxY)6*lzpRQAc;PGlPua$O2~G-HJiKk?fI)~e(#p5q6l*Rrgs zamo;;)${|OrKRBfy9=?S`v+Dg6!W)iBB<{6RI+fs3HS77@p@|qgE1GveAG|UPA7Sq z_4+r?cys^{)a`}a4UBn(x1d}x2Mg?mv7Je?@kt?KW33@8|9)H=;G{*DkNCiA&*3cU zh8k{HYNz-_cYeBi0=G(e6hEoo9=4f9!=!5wtbdlZxO=pOwt{MUWt+~Y+n=U!wySC1 zuLEFdZ2;B-tN1tX2(Q^Lp)pknw7$s_J{e|UjBFazY#Yv&U*??;S+FLr(%z_e;>IIHY1{_(5%6t?ChygsT=?ynk1 z`OOoq;ovpqb+DbRr2u+}q6(f7i3l+)n6{I#N{7AzYyxi)&j~ zv5ESpF~UYyP}7y-yX_-ccQ4T^HhDY5uDbzAzt7=f-xvsuTLZ)VN8p}5>L^$91Z1Oq zSl^!G=x!cQE>#w^==37VwIO@>wu%+dvpZuKd5mV>Q>ICM z^PE@qE`q+MUx>RplUAn$V{5@-au_5lTDOLwQCTRzSeVB$hYx1n7k*POo4G>uAYEas zh6>oc_hRdv)_|?8A%CmX4T|=ZfX|MbyuIr^jBXFY+edY%{!wpQ{&*xL_SMC0W?OMl zwg&C+n8s%`n+gg~2hsJGLa>@)P5$O`Wb8SZ>bho<>Y+}4-Jq>hef=M~T}mNale^g4 z{J3PS*e^0YX&v7E??1e`EQx;k#j?edUb6WuD$K*!l+hF~sZ5GDdW>4g_Y=8Jp>apy z!-9Qq2#<^3kM*F#Oqe9chNWiBgGKH@!^sC&Z)KPfP>iQC`rvNO<)~v`493$=bCjLI zor>JfADB=pJ>ot@yyvIm-)~9ez3zlidn=2kuJ{e=NtW!Hp_B_7R|1auU*Jsr8GO() zipy;{#YM_#Li6C;lG}gMc=H3}QRDR%x;s&wR_)E?nh(n3h7B61eD@Q#C@7Pgvt^&e zTkLmEUcVW|-cs6q?mTYlUXDhWs!3tW82)Ed1uS|pfYk|_La5FgQ4`w9hHnYS_FLL? z%{x!**o%jRC*m{Uoej>cFR6z%QSyl1v|#5QIARedmER!8);+#NIqECe`+MKTa#dN@ zKK6>3rI5qo&~2hGx(#bjE~1bZx=g>K0BnPnfm}c>OdYrl{z?u*b9jH@NxurZ@5rcm zjsrCee0|haO97`pqUCG_Taa?{$f4Qelrt-t(&=nw?;@J zVymV98jj-4^Y)xf*94G^E9L8p1@u{S6GyE*%ABTWGLJnTu$=csgRn1rbK-0GwWXAc zo-T)j57u$phDm@W-N5Qk*Lly)+wp$;QM~7QjcnI2c2)Bi9Dh>?gM(i|TZ$?x8!;3* z&#F+tvRv*?h6!uhE6=ZaW)jR(CM<6Ej#l>I&Z;F3638E!Lt~D z-!&YrE2KkxLpkKB9OgQ5XQI^w50dLQgQf&6!5GCFe&L=~+~>6xY=4d}K5*T|UB2hX zqD)^yYNG=!{?{sIGn(LhtQW4omkV__)!_6M2bL)^s&w?-xGIYhw7PSUbF@{ajp?Q2 z>K-Fqb%f_j<PdI%VpjHO!1PzgfVZ-s>fDU0;K%yaONPcmcK)6v3zRJkTD| zKsP$$LA&t-nc_d*iM%&4U#M8%tN=@sa>RZ%8jT&e07$jSA4H^@OzF-$U5@ z$uB%1^@qPrM)3aU1Fn3Xz;7-{0Q>08%)CVluK(H%YI_5v?FyeIojS)vt;UwCa8`z8 zKbrB*yAt$dg(%Gb%)ge+#AY`&x@}>EDq_xM{hbCfs8!_OH>nHZ;b$Rr_k8EOK?^|6 zC|eMUUJ5qrJgI+uD7Pv70@T{S2Fs1%m{o8J^h((#Z&NzO{a@G+t7gpd$Fx#<{uH9ZiA6{^D zM$;YpiTZzi^Fw^P+L2v-tSsz3FC)xV|4Zs?_XvFNa9TI}n{%H{Bd}=NGpt^_8GCXS zY2)!({IunxaHv;b-sE&0{k#FpsALu7-06fOuPrpO;1j*PtHqqnqA*Ju?XNpvdqEf*R0r}SN3({^6}0Nd zayGJQ3GKZcD`p=5qp@qp(ZdHCG*xyY4)s_--X5kj%v6cr;jK#1m6ptKoHP5~a0uQv z1>*H{Z}@DF=N#`^!F3x%LcoHve6h-N&QIefS2tLRQDG>&^pj_SmG)#gcmWpHR6@ja zq&}4u_)|At+*#T?H(gc1m_e%8bms`4EB5SuEXaU>X@dnt`?F$Sh!<@bc%KTwU*cIy z9Tu&v3Ol?O!#kNi5Mnlly>}D)%~dXODLE#j)vFP+-O{CnAfD zDFw_F_lBk+?2()TD;j=IlCiFtQv9sYDRqQ2I{ZGDIc+bd9!bXIA_pra;2g?PDCE994=xSioMpW;Ja`+%G##T@E~|%~Ip*w#n=HFH>K<87pAIY2B<%_e^ zVE7SW&*l0nLa&k^s9e^7U($bp;@K5swPKp^=frtT56uO&ko{06Swvq&FLBYaW0WHH zMxMRc3@;;8m`_Tc)H1oZi{j)%?B`+eT)2=4YwnigG}}IW#V>pO<)j4EnI28DVUr-T;?lu4Q07DFDJVd|_+4GJm#EjWR5DVYb~)klXki zlm>>ADIY|7e>>pM)RTPtHV1t3R*^mVd4jXzCs4(eo$P>Y5;=G*gMir=xWz{%gMO7S zUHLbH0!1CbNc}zcpO-Stb+4qxck@ZpwKwzJR4rYkF<<1}{X@INZ`|V;HFo$?5$4E_ zMK1a(Xvx(>mcJ=``Ctv(6UxD}crE_?`G@c@aTpWSVak6w(>Zr>-u64- zJM1U7phc%53lC787Q*kud>`FrbnrFjF?fr z2|jPyOg=iXFrmnv*_o@6*-J4eovn_6Yk6+hoVj2YGm*LRI;?2)Lh2fKjlU`Cv!&s; z@KnPa?mEAiC3X&>$kW|ihDREEHldIY(Msey#-@v0Y*o%FECC`i?cly}1nn=XvvK>a zAk8(5@j;%T@TU?2LNb_r*1GN(TwU+{xPCX5`rz9QYJ0G6oVR*U+zHb`ap|OLi^9tDf+M%r9^qElAeI9<;Sz!Ku zOUXQJ9IF(c)mNxAYHfB4@M);e`59iCS}b(S}AV1vGJJttL=y=}}_oya2D4h2#^ zxCky^IZJ|cm(X=69`mfl4kmX+=@8o;6nbPmnJ-)b@HT~+M=A1OhZbQ~lNs%K9!~cy zPLZ~^DtBewOFXwRm2ZBS0#A>J)91atXq2cA9u7T6hur!?;KxRocjznZ9lM?KuU{vn zU1#y9|9^Dl;d4yuJ&-IPwvo++dOBUB&Z17}!hAg$-t5y&^6R#Dd6}o?GUBo$b7>n2 z@g{v~?P`7Up1ladN9^O-HGO8MZOQUTm3GE$hZ|pIS?IU7Q1Qo*m5u*FH*(UMV$Dlv z`c;kTvm0prhl4P%P0XRMeFX)}MhcDtzmn`UMaVDgMy0L+@pcU5VWes*M7a{HLKOS@WYKpz)^QQo^QLOd;45FB??w z+s8Mu9ogY*W#nr1$47;3x$YrL*+5!6z#C^No#Gpmi@A{-W?XvhwFn(2?ZoYY!)e-} z!wgD_SlXzK=*T6JPRVJm-Q11TYYJ)3F%`=D_mUqNZ^61W+9jM`2$c_27G@vJhYKHP zqN~Du{j3d;JV z!q4(Lfvvx9V(7VCbvY;9L0rHnm82qe3GRN8>Wz9-a>w|-V)GH{t0*V!}!~y z2H^;u#jwSEFqHKQgBQNyY_3^^lh|RBwyHCgys2>J+7zm}J(hB=Uza*g^@PYz1XDI2 zg+$K=F7m4@<=eifSoYix%|d2jhIp4fQL`A%t(Rp+-9p91jtZ`37{Y%SWVzO7)sXc> zjsAX`jHiw^K*!B8F1I99*!W6Dn6q*Xzf#SV9kZ)o{#93*8P?#I&2KPTV-bGa-H(nx zeuc*l1@LuDPR!uFV(9}OGl#J!MV{FxZqkegaAm=N!sEm$@($|*Zn?&g)qVjMz99Jg zGZBIp+QF~ySGbpxW4QRAbE%zKF^8+~p*80-YCP^Ew0}NGC0#jW`ExvOH+sxBUOoqR z&%}cAB9WJ}_#`Bh@LZkg89X+p7DE>_qJrK#J~5{j2i>)yZjX`7rzaLpMAlMzZErH0 zp(^rUcH&zxxL0BsK}DHGXeQ&ue)(84BU_Gr`y*!N^>49>TW9khP=!6vtmYQAbiwD> zHrPI)N4ji<3@Aqp2YysPSh1yoKhm?0^?j4cev}V^HQr4ysJI<|y-tH3dlO=VtZC}g zf6)F)kx57DGBf`QSaN(OL<|n0TP|-YeU&O}l0OZOXQFZBoXK2zNEK&L_Da%g<2<%c zdnG%y&x$>5z6jBC<)EiZ^bxFi#g&Ygf!J5WxJm_i(rhV%bCvPTRPM9%Uw1an=@9s! zyj7r0`4DRDCsm6C+#u%U<*Od=UC>j-|gF3WTFe20;HlH{tEuIGlEF zIiC2&{Dn(f@)Lj$)@|c_;mCb76f1iSTD!mIx_3 z%$D{3%W^F9naZsx?62-5HZ1T2?s?S3we8tLhupuzR6`}^W);e8`^&R+QB33p9AXJN@1XI6L1pY^o8gY+HB^!`!|e3>wa{TA>0PD1>hyT=5arer(I zpUa>Z=^(|<0rY-c2bt8pq1n5S&>4SUVfiC}A*?utHYRm~4tT zIMowt%_GA-AynwzNVz8au`Ilumjo=pj7g@Dd4C&s@4-jByNN@E#g%j*I1OFqG_y$C zRCY*pEiIk2lnh^L35HF|Y~$l_)~-LAQXP!QS1Q0cmqu=OU@JfOPk)*;;Q*Z#eO`06 z-^0Fh6D8sAQYlYC0epg!$*uohF*k9J{jNU4o_$bcjt3k-FL6Dso~*-y+VeyPwb(tI zu#;B5R~4T3+5p-AN?Dq8j7vxG88V)GiYi)5xq@*rWOwTRn5)khvjtwg*xUFhk^5)C zvaa`{;?+*V<42zubX=u#1{LhbzAr3zW*;s~a~)j^jRJ?qE@aVgi2b#QX3K8wVsAg> z(~Qj{Xxu{wA>UvkYq>Cua`y22jNx9i^4c^mY|S`fe2fMtLnYpt5X{Qw3(zulmt>KZ zr4YGhDR;%>IE0K|O)iVyLrad2n6p2{nw7#?_U!8vwQDNr3Xzh&+yCLbzlUH!Z8&M{ zP9oVZAIgl<6jJmuaNNyM7IAzE9k#BfV*^8|59ccIzeP61f3MIalv}-*CS+*4JR0NZLjA|Gwv02}`}N6m^p-#D*tm`*pLA!d zziGiJnvVDHYO+;kx|nAXN)DkpA`32oCiN;sQ^k9d{Xc)e$N^p~Pbq??x~tK^TXk@~ zXd37rc+Z=SSV~S+&oS315r>95GPG1?`-~9Yx~_)aC&S^KG?bnO&BBs*SJ2?!L(cf> zXWrg26ra9r!poy%_>U%sarE70^o}w^M_Utoyd|6}L{B_!SOOO^9zvJHdz?070`m&c zXGOKmsM!(>qx$~>t;%LRw-oW^=dold=Cuw@Zv=&2agf7ONc)qPU>dcRPB;u_!)HGR zJ^Q;{r^^leZ8BYIyukqbew;`>hM&=B<8Vyu_$}3(a)UcseH;zj1hOy8TB?jqa@^`|3L1cjSL?{>u$+snK(OO;`dAP~A$=KL4@L=T+GF@_Ss( zr^}$&&zS8GKgxn5cCdq^=CLoO9`Ik$d+v$1b5su@sLI*XM1E2cXxaderO#n?hVR-vG^2R+ubVfdw4@E>Oaqnb;hT1lI3eH%@GU-qYIclN{WMJqr%;0YL-E(e|5 zi=Yv56rYwW^2OIQu-|=Kv{monXITB?(_D7)bynrzH~kWK^(w=2xgRA8^G-{=>y_xj zd0D#Wc?N&mMPSC%1lTa&fkpN&r%miQ_B2n$ppJNqH43EV2Xbh~vaQtbfHplma0r8U z)L^IISkC?NN2%JLrMM?-EPvh-;bRidCVae>Xgc$<6o%9EC+#e=Dp zw*)O*rHfZ7!Dr`e>Z^VmyUGLz_vC2(x_nFuwWJNnckt-}fh%>?r-!~aj2?f6LtJl& z=zR;PR)phg{s_j8)MOi7WB7+(3|MX7NXXQ=!@pTr!n=JP3>Ccu^dDHl4>GvU{dG^r z-Ydr9pdFFoYg&Tq+ahrfZ%11SgUI2)4NRPv&ZTOsWU{(z@N@nHZur8t;*P+B1+VPQ zeH}8zS@UBVyl%M&g$ut*jZVqX`+pWfsC2MkX6r#6_Q$x$*7c}U`P#|nTBhLYvzKmY z_66lKQ{HE+8ntgXK>r#oN|T+9J>|=A?YJ)V`qjhFS$PyQbw{yE9Ub<_Gn@N$ci;fm=YS9E1MRI*oa(YH zcy4qH*GS*chSUk5qyCUz$PU5W0(Ie>fd)%>6vA>ncyW(%AD`tMp`bsxl$TUW&mB&} z$qS0Y+_i}msIVKxxL8#D@Vzg6Zet2p6ZfL5=SZBaJ)8n^B=qaE4{h0UhW+)p0i#!6 zz$2w2ab319zV4fYook=+u03i(XT^7vh8|`mXODo8tjPvEm1i{yx!~2Z1J;k54o6?# zlGN1tl1=w8+N)j8otf2!E7i5=c5i=4TAnvVUr7bGw_CCFSsdocsk4;YDqP&1i5Gqj z!-od-IApO8m&A!#)8FaXFvf&;sEC6Pf-G%+K3RD5_zTVX=mF89*X7`-e(blesmr#u zOEj2IAlzULjheb_vr#Rqzq5)0&NNWDhL|(Ce*iy@`HOl_b7^wnXXy7z^p4e)aFzN3 z7Cab&7xY%*pR*@ecKdaH)5bMyN%?p546)}z+oE~5#>uS4bQlvtg2H7k9%~o37G3m$%aJ|XWMTJ@cK|QRv0}K;*-QTXyZhbnUTbGtqsJ83}epo zUkaZmUxd}=4PZ7s62!7$C=j#FSIp#K_IVvpo30E2RcpAHWx|H)_4gV=^5h~4Y z!#ha>$@$k`3L6{=mJj3Ts^cx(TK61xjNejmSf+|zkL^Ph{3fWlXAJ)t$l=l`Vf2E> zW!$s0IH*6I4GT^yQ{Jt^+;puYP~vRFaX(DpW&T3eJWh^TguazdS{Djue-Z4v{(;Zc zQRVtqj|DR|AEzF1w$9y~NKSJq@M_;OZbg?aHvc=p@BiM1tQN*W+E8ub-y}6UStP#u z-}i#~i}e_BLPokzX$VX=mS>Wai7>acj7y6?fgjyN`D4-tTw_FaXuHPGj+-p2CNZSu)qx`xV{?P&V`1++G}mf!iqjb_*e ziCVz{$(^ofZlC#ZFzVZv#m>71hB7a}LQ9El`LG(^7yRRQg=$W>FHqUDS`;YsAQf%`_2~m# zRL;q|y#5dXM+=*Aj(GPUk~^N#C*24CR~N+WL4w#t5P_~i(KPXcxX&1T55Ia zZoN>(CvG*Q?t~)F)J~rrPZm2S>Mn7AUTlSrdESC~wi3cUqUNaJjt-^63K&V1C* z8aTV&3pzz_s_A1}xT9sxK3sBQE9{rE=d13*k+sQ`5-N+X5;eNH|15e3u0RbxHBuUI zidye!Q|^g>_+?-V9oV~0{3a#ff|(BFG-)EslQzK&%fm46v_7w|*TfH&wpQ9Hu1E zGdYCO9k=kd?ih5QvxTCPU!dClkDxDlqZ}^8(Ryu;Y}6xAVWvElMCnl088>#i%a@;J zuOZxP4@Z@u8__pw58g;S3*BYD_~dsRzVpst$4(_ceEJr;IzSWb9)!ZCHd{KRC=0P? z8gRFBGau#&``LI;^%z@-*}XOu5_9Wlp?2VZlYXFWi;F`PXJUcg^` z9Sznghu~e)0In(aI$Vvgh1?+#uu**xsVj8j<NEIz_69GXc4pD(>?Ho~xfdkY)WcWDSiYS>~xUoTv00!>#sU!D7Oe z=55?j?-+FGtBPuN|M+{c%Fd2^??A?jtK8u=Lt)II{^UKefe)Ub!LAP3K;QLN!Jx2E zO6=ZGPkOCm;de)}IlIMdioq+;vYo^>*e?OTdM)%nG8fiGl~MBHDw>{9NH^k+avKxQ z!SH+=v6Pd;X9e@IsK7O={CfIh`uO&8FWb|?gVFVI}iqB zR&WszxyrRE0-)b*W>~lou=wCvtCNh4j$IpoYCVBG+wk(dwCKpu7|_l4aP!f4)t!{l9doP3-1 zqxGUd$>zjNoanxsVmGcJb)|38@%wr}$d@pKITp+%rvbfNN~OEcs)NM}S+d%-l#Z-< ziH;+>v8GQqKdJTsF3S5YeKtN6Tqh}_S-Uu^ElQ%f-}X~W<1^T<^Z^EEJ%)|Fb=lOV zVX(Y-JVftE<#z6Q$B$I6VxAHHz+w^Ot8~sXuf{XMXVobUgIQoS_`w%f#lV?qWPZ%B1a3cRxd&xt- z8|OIpp{9s!nfaVe?5d@O%ia6hf}L#=U7h)p*D6XQo8n|X<-7yiAEW{0eUI{gBp&>i(r@_qj~aw*P-aboEVY)l)F3pBKaR7DBI)Fl z`|N4<1LiN@H@+^LNz`IR6+3^3TJlLO3VZ;|GoFCA%4jaA>OJ;2M6;dQd!+Y!wc`Uk z2x+NOe%WYy>d;+*-K)o8=UcJAQ+*Dza@1h9rB9@PjvCRo`4@3hex&q7;X{~yQI=Ev z`Gz0Y?<*XMNpM~jcnxn~vtq^lUUDjnW`TK(2Zr5#gP#{H>P(v{`C^i-6fd3aytG#IEpegdegF(&!F_6$iwtn2;+RV!j})dL_xL%QbR-Y9~`RwD^hr^rw!m`(WulxFZ&U14==DzX zoSsC{Q@``m{`rBTaXT(>SV=6k2qQDJX-bzY02mcLUdDrgHoW#1)KcXP#@Lr=L^ z_f$pcZ*#$hAAqLX@42x7+p%R?x7dCE2T^i_5!;fdLw}QQ@?N*CnBfa2su=1|JALwK z$)X7?eMB>A<;T!&muP&bYrxA7IK_o^>(N^aG48zE0$C#*8DBLS<7X9#?JVX&$-?t^ zWnLMrP;G~wBW>7)tFf#}N(us1TTokE&JBK)#$QZ3ix66e|AF8$JMPLaEEz<{{tjY= z8A?=>eTHUwM#7d23RHb`2~&D{5Z%(RVAI!Td>L&6XM^W~RIb1;K2ilkPddR(qxY23 zkw9Z_Or;&YQJm|8aeP0GOyR6j)o#ny5vX(JHoBIy@W1Mo!-Xx^K=;vN+&O{=Hpv&d zO(o(%QPOZeth>P`;{vCXQABQIOxWsvGHiV+&z*Fzz*}8<@GQIx`L9M?P*)bI=H|e6 z`%|3%-5Ri&YtB4W2rs?7ilHVufF_$$Xz(L?+gB~Xg5N=e#1gI*7Sr*}+c3j8g+&hi zjFHEGq4Zn>P<(iu-22I~^V7!*p2SYX8ckgHJ#-=5hu|5@&KemzA>R{Tss+QOK_yZz3q!@SPBz#(E z&fL!*VJ*YnlTl|J{OpPooO#pO$c{an!t?9=k4OK&(XtIbC68ot%_|}7c_y83u#z;k zdrB&G<+$~S_=Xoj>D<5%$FSzbF8+965FMQI3exP$z-0PTP|TK>Se>{_^Df4r=lPpt zc|1~b#D15qa$n_H-E-3K$0Ie2tGO1%GFFvYHtVcr5CM{9u$xh~v==bhGI zZWk{yZF1rFzB`4ZHP++b=w+Zjz?$iPJwiY zL-nNW5*sSOEaWh3xR}B%U85)}m@VdJ?lyqzDOafB-vClJ9Yt@dR;OjxrU<& za8${HT|PX8*}UE+(J$Fc_Ws+sio$e!J9a7>u5aNke5nwh|F4{%xa+KKi|S&i&|b<_ zJ${A;r^gX9J<1=RSxF6UJNZs+4cwPkB+50~Lq^(S-tp93ytV>y-o*_3qP2wI`l|p} zuU|)Ls)En5sfdkv^MQqOpXleFix75iANs#{;zIPbSoeP^e9*>noUMCV)YX3y?oQ3+ z&;QKBFKwD+8X<3|o8UtpPXaLQz+fS>m2F*Q8vr#kEui8>3hy>n2dfN*f>O^Yrh2Fm z#>cK@Lw@*EfuNEOiyw1WKcw^R%hPGpcwh2#y#PLQjo6v@jo^?}1kZA$VB^z) zbh|i-rp5^#N~fKqpnip<_D!MbtBT;yoDJ~LEt*BxJ`g=f9SsYM0(&Iya+9m(+d0vZxKk}6>l7cC$R+hj-c z&MWNEe20a^)5lm6(APw>#uu=AMM|unQv`^!mtt0x9<(3)1x*)?$)T@|+P~ybTz;td zrpb8BkUa&HHZ6k#8}h)atR2*UAq?6P1z&Heuny5^K4y>++-AWnnYD5Tqi>?J)^a*@ z`YLGkc|fE}3}zb&^R5|h!F$CKZr8wteE-%evF}S&d?csF-~SSe`u_#Pqn{ITv++Rm z@EHK>Li@xKH_r1b7wr|VlaV9;y}A;;2uA+bPC@&=NOra|9MyU|`N_FK5Z3=1?45TQ z8-A?7HKVKGIa}hT|DD60fFCfm3_K_YBVaB<7u0PsL2jKRjK=!BO zV4ixIx`t>=&b_)teKe57IeX}K>l|vnzmIqSx0fv+x0k)zJO$TGJi@|^AG665-RPZT zEnSPMAgBs~=l{)yO%7>%>C@#D`$0nQn@+K5S(Dhmqlr+ZodpSbQj%)-_pI*p0Ombm zGE?Q&L8^T?OxdA}B^d|#(0W-ssEj!OUOhj5*CQ~SWKCUT;;`#-5DvQRNNTZnXpfR5 zO=!Q(kDlAe8I0Kn7k9-#iS1deGMY~j@lx!T$A3bvDw8?<8imHQ7s31AC6KtxpIW8+ zQ<`S(9^ptc~eC6UEJjd48 z@>KBBhF`0446Ag<;o13$EUe-K@_pX4bfyDM9JPr`^ct{X#Wz?nyAEDNYq62sd2Uh0 zSnf|%DV!QGfEM4Y#s@(Jef>}4o$cT8t=f1#U+^gP`#O(n+jM~x7AfJ>M~Zxt;0WcD zPr#2!2Ssw;qu|QZSo}0L7b~5darg&KcD;2eI{Y0=yOw3aw7zScyQC9bf_SDKx}1FI z8#t}I2jK=A!F!WC`*Fk^@65KQCu@h%n{%nK?rIVL?x-^xI!l$sN=QDxf<7Bfh=u6E&Y=7KcHy`Kjb z2(FK?d9$d}zR=0O}my2{5pNZ=7m? zA18j~CYL2rcJ2NKeOFJm!`B;*C!gVLUu%nhg+|iHc|Tagf*>p9jjr+dwNK zk-q0oz+4kqytuQQ6!TuvC&81uZm2JA{Pq`*jq*W_SErct-$|0yBaV>jh#m~n&BO&E zv(e4|BDX(Gn~lGx!Xh30+3HwL*ryTACH`|_V-4lmz0@Xd%s(TM^q5$-Rc|ue?UF$+ z-i?v`w@qFWerGQeZEa+&mN6_~$9&pWHyHm57)NsnXO#`1?jOnYaoBl^EP02zor{U?6{DT{ zck$u)YMS3K9CJ=gV&|oQ!VM)`JRG~1Jqb@jMWZn^=W+`!F*(G4`R>g>`?DULMX~JZ zk%#bHMYtx`dj%Kb0gQNLkM!M_HV=G;|Gq?EWZFYc)iVh@8{44lO)IYumj@}%v-XG-Ap#sj=VX*Kko znGNS>e&&tD(`m8E6*NgZjE18>;I*9^H1XDJ!TVkd4r`jBbA~HA-pS;bFPaE-liT@? zwa;+Us%i*nkifM5ZCv|aFSO<#@C~MmVeko_Uouf(Jj{P_h6%ovPkJk9lJ-?xu%{XR zE|Oy(v=LVc-UZv1Pwa|v99-3kU_Sy*!l<}gkoGwXq#D#rS!vpX#?1F z+gzyJDaF0pkikAJGh+ky8j%81YL9-nk8fA#x+_mj9Sd-e$k?yj&W!X&O+^;t^c$?2&X#Y7Chy zt%u4uRrs?d1-plr!m^XDAi4J&zQ``;W=>rJQleP#`oE#*Fv=6Fw5O28*AriRHFSGb%XV~Jc{Cn9f)v3(;}0BKVud|f#!%l~%qcb1LEwF;-P_&d6kcmW#j-~5 zt9r$C+Hj)pe!v1V{n(OT9ag(jj^!_Egdm}Zo8HeCBGw;;Ke?@3&g>vBma^_uKC4r_-eTtYW`L7Uqko8l4pzXo}3gk?FohH z*(D-_t{V38u?B5vjv(L5Dfn~#HgwCrh`CmKMc1Ydra6-?;*zIMV!O%fP*!mlcc`ox zOo#U8CLL|!Qm?DByJM30{3r+5d+H**?pq3%1O0KKQZ~AXC!p5jcywB`2XjMeVCh^p zQVA%hC(5#9^zDuKtL}Yx_@xF;=YQkoX8%OL?SpZFeW1WOFh;s~*LGb%0v|l}J1X}i z5$FDu>^A&CWyNs(c~#7PyP*YTZmyzz)|PzrWG%S!pb`Vm`ol4i3F2NQxR-qf7Or2) zO<19ibA)x?ws#pE`DqPH5>9|Pyr4A1Cv?IfB&fn`47QEgJBaeULGRzl4g4bRuPTfr@n+0yl^>R4WJrfVk zO@n9IyvX%(9MnrSHN2()wC^2{+S$vQ<0N-hxjhS9FI<2WzfW-X+G%#!U_8ngpTz!# zB>L~VKlbe_;6mJ_n8Slm*xdB9Awy~ySgQ_T#WSMudU-7;J-!iI+b4=DW_+U$S5xWp z+#%c(ZEcL$_LG-0@`vtoPV^~Hoo3u8uJ-b8ff1xgrO)5NYGIEq>)9cm^kE=N~A=)<(L zbYihEC$`v&H_CGP{bLkh^}3 zj$==rzK3~_YiUKu1Dfp=%0?+0(J#XTymjC^uCw+SKG*RTdU`Tc^7(_1!*gNH_KJ8h zBp5dw`ogcRQG-e0X-rkeLo&$MkMo$_fwxcH#5c{8;gn$z46U6D(t~d`XqBoG|EnE+ z0!pFwnG(LbH5tqg+@oVj?I3?@G^S2y;eI~ahfCdMB)*G0nSa|rJbB(l@R!F@=EfQ! z*R_E?PIDnS%^o_rA`N^u3}>=#yRqwIIxX=lfK}v3p}%CAySG2RtuUrjlTL%lfGIS* z@VO}8CLU_8F2lzOulax0kHEB3nbqHtZTL|>j(*3RQGVJbraJ91=oJ5=ISJix#MgjL z+$UsVKK!GhC5^Pu<~mkqXTz`oG4Si?b(k<&liMu2pI@D}pBzUB%vpz2rg<|O`WD+z z?~q#jHTD&KoxPgIDW_q@mOYemEF5qHR<=EXT(olVyr^mWcl7!udA>;0F7S*Cf zJ|~ajGr^1NqkIxq{LhozUmy4mbry76_)Zurb1wd$sc*9@97!{T#f!qJcYHJjjC_GN zc1PgD`xz9m){m9@TfkDLL|d|?>AQUpj2JNm-fG5Um1zzobo?ZpX&XR6!Gs>PoE7>= zRm4vB!_mhFF_C#2sOmqVk&~S1)bXdVq)CT0Zp?w;Ve0Hb(NHq@f3V#S6def%UbHX!F+{z4bnezlP>;iK?^l zXW=xlL+&zU+W9zi`VH(k^&bwH%i+jN0=qD&9<7=KX^rnBg0&xTx=Ia}@%?C9tPw59 zh(-ISf`@SLBR*bPi;w$h3x-d-xuA72q~U2yZdc>@Z1*OCbut8tUrR79{RpMrdnb;6 zA&)kCL-0e(E*$nS3ZEnzV*N`YB%UtKG;0F!VM#j9@|y!Y+mxYj<#KGCx)4ja@BFd@ zf+yo=HNQVu;Md6wrzZjLX=Cg-tQs)_UiuFNF06?=wt1mA?m)Ka+M2=eutbpz2H!@f z^AoT5sN9=q1RkXZtIRv`zfhz*v+b7SAy|3(}pBD@mQ`TjlGz3n;zY$pSsv2f} zeh9Vivqh%1^0d>+oyOY;-orPpY|d0y@Y`;{k7;j(NS)P~{2+?b2dmJLTuqjxp~d<> z1%vyK1y~?3s9USIQuIX$tSa-t?d4DT1=X`j$Gm{fJ&-3Yhhk9LzY!W9Me-%%&!G25 zJKkY)H*e5l1MUkHx!|qlyltie{}-pA*U$=Fa5&ob@VgXl*2QFa{pKb2`>`LWFBwlo z%^LLi{wdsLosM~??V{@*gV?Zd&De3~idd~96K?1P+3IbOr%{`{N&I93&iIY|#nFrK zum40?RdW?9*X_W%57U_9j)in{b_#hnuci=VFHq?z=Hkq*Lw(~1I6G3EwJzNPE-7`` zbtxTZhWv*8r$U(dBUM&+V<~q!Ss$N{b>yryTe)WiL+tp4@8Hg{XRP{IF4u2PuP7__ z08Gu=4e3?+XqTcwhSiT@k**Rs)NZG8XO) zA=A?`H1~276csIoZuNbz?(kw3d}q1f=$OL-&7?6SOItvM%TvI|8gx_`&dxX)Q%w4B zaMFFnH7_jY*xxiRMHC56l+3+J8G^p&%t?Dm8YMq|2-Bs;K!CuzJ)&+!p*n}ejWrAL zo$YqK=9_?h2By55_zEPg5ZvM?=YwHd0_50i1RXg~Dk{Cf`+i7(e$7g(VU(iXo~fH* zc!wISJYhmje~M6YV>pd=J_g)*J+S|K5I^p&qDR{|f=re+Un4SRp6hEseP0(A1`cCy zKKSv;q49L*VKmool0H+Mn$E`cOn?Z*8KC5ThDWPh-f*@Tet0Yo?P@1crF4Q=Cg&}4pyW#zw8MG)anM7to@J8`?J|Hz6S3exg zX1HBo73!~8aBLXmn%8n$M-N4b=^aeh*TBtPz5FnxA$(`t4c=+aKM3{mfR*oOvMj9# zoBWf9IjNr&a9BzSuISu>g~m5wfx}SnJW~us(sHb>-x0VeaHQes6~T3VgIk`pocS9^ zvv(8zV`>UDLKaVnB@NogZtQ#oYcFaugQuaa^I37jM~7ql7Gbt|s->9Wb4Mnnwtzxk zD2OUzj9^;1enSa9{ke+o1{De8wzx7Gj+S|zeiu}x<=om;dSLe~y@>$d*?L|=? zG4!r43-8{lq3ZwUvZ6UhXzjMwd}`%x?)KN={J9HVd`ME4ko}pO(J<&!gTZ$h z_IkJ!ec0j!m-M^vecWo^`j|)@ex^TnLe>Qy8~0$!@)+KG=2Nz#V-hT?7)UU47%kMd z!JH;9uF@`of2%9xxPKglp$2x0x2k9Tx|VX6Js7#3naMvZ*&w(#+cD#vtW9R%4Y=@V zE_NM#4NT}9w$^u$+L9Kg|7azBbE>u~Gt6Y6tn>UncXH?xew-V-d8o+m5!~ zQm3~cmv9eM8erVMC2YsgQk38E9A~DlU~zdO{^6~)w9nF%-YN^;q(=aiv%5esx)M|} zl4xAQT6|f%4!CDQEV^_%Q&;BCIPASHp`-?H{iU*rn zIFMP%H<6T05{=jXEmn~o!4?%4uzB-p(BBv%3qOhcK98i) zkH3O*q5;$^RiV7K9(ex>;y0+9Q*)RF-%mVLw4*rzS9Wja_3BRGi8&s0dwMwcTL|ae0cTb+y5o3 z?~!BIYGP?{cQK5Y=`44{bpY=?c{g8iH3xpxRe|-4I8LEenax>K3KHdUSn=`~dgjca zc_*r|{JH`QEKlS{=Wn66qf3Odi_`S_w>&J(ci}x`_fxkb!=oGI@%e`sA@AGHvEGH; zalcHc56s6`K4L!py&br#|KgWL3G2m;L)_02FE+W=j``j5ft#PpP-b@<#E))3`-*N1 z`O?nqy1N%_ypo{g;}i6nTF=S-l@s_60c^Le43^E6YxM6q&po|40%cCxaqE81q#vFe zY3|Mq^v6O2|21FY9A*wD{k|xOdZEm8q>?DbDv{l-Q(=Y2Ptmqf(frM?LZ9QnERq_f z1cTRJ6^ETDVG#j-IQ-K+l9H8@sHe`SU*%8n-0TUWA0L-6XnC~n2WGXL>v#iJS~ z1{{Yi^+TDt^#DxT{tq>$k6|Nbs?z@KLVQ1TI$PDOE8e~G5C7TiI9%?$k6q@v%*3o1 zriKLI+x!mtGpU+|FV=+Vz5h{5TNb}x{XV1=4y6G?^yh8MR8}16N?)%JU}2*@xVnq; zC8mFt;{l}*>OS^FGB>Q4iPjra;v5B*t!Bm!KbOIPtHO6bxPtxJw1nF^rI*{0*9~P$ zmC5mHC7GxMu;D4y5H9kfBTr9)MX*2RwpCNDK_c57G!!F(yP+hi6PEv6jz_9n#fhel z@Nc^=_xXAx?n*5Yxz(5BkkezJU+EfpefBQ72k}@l_7ct(_S9^u3fgYeXG;f{fccaI zxMI2m8!%0q{W4iYnLjkynYxSQb#RZQeVQRNlviUXzO7@aZmH~0OdmoDoVec#$r0(qRG<;u72edSbjN)ISF3Z18;ZGJ9`<}(HKPiGY8?*^&{C3Z%-@_ zl%^p&71+DK$3a{UTz2$hSf6o?2f@iSKlBdgw>pQ$t-OJDCE2)c&I*dmR+aQwB*E37 zXucxYgif_*fJft0UZn90lTuPS?8=3+Q*Cf-lQYDekE6{u6LF=JvLrj`D47m!!uO{i z(L?v8{G`6+lAcL5uxsRftm=5kkFRtB_%(;r!p~Es(6!hW)z_`=P8~%yrlkx0Eu=`_W-Yl)pTQ2>K7;`GP&UXo zj0z3|)4zO@rsXA}%<5BHSNm z25qv=5aR14DAm^TCqCbU4Z(@fJFAWbTKAV^j45Q-H`b$d_h#6v7X-fcd(g~&A0(ZR zVDCheC04GZB!%HREXIBb8`3(6o%=2LO}~_b(l}*-BBm*NapekVpPtU_6f)3kvOaUS zG-JutT5M}{xM-iiC2<=+k{sRRx$8G3BKL458k;&}f{{9F`y-+PJu8Wiuw60(bM`1c zfmxWFK!{5eYn=EB#BmuIHlZ2!EX;=CFHG3r7I)5NhA}ht$ZdFeb`+^b#}q>6Tz})n zCEt5R*QP4c9`kngqQjedy|r*b>jOB{>I4A}JE`zmDQLc8=#%;wr#?;r&%XOGc7?zh za!P^7fghljcZ3d0;JS`-%;&z)C0QVFo!2cRoyH2GXZD%fXK@^S!lQY1y@m4H>cl+} zQe02;2}s|&iXFMFhtW&-V^+C=;D&M`mAXBoVYro}tTKTsZX}s7<0AId43vDC6D*m( z)f*1JNyFfT@l>}r498>~fQ^|2ysdFH>XJH3Ip`}I_x?6*ZOMdJ1@e-k(Z}%7{tfW_hzisDp~E~ys#Lk* z7f@a+cg4aL`!qY?jd?6??u@6d$UE@cY6^_idO*WOHsC3iWw!&9DIi9b9rWM9+3h?{ zZM)*>f!8R0)b&DGXlFuk&mKUuXpYVUCpuzq|cEct#Ag%bvywA^5kfWwg-<{?Y-`|fdk6gn__ih0j!2y^X z>E&D3XaVu720&(*@ai= z*~S+Xe8QynDokPV9aw%U7$(UI9rxYsaLqLwb_`xiWwt%G8`kWFzVC`m=CCQ7j$SATzS6BimuhZfb9i|bRaE(_UP2$g6;(R5Hpy%tITQr z@G_Rm9^l)|m)w&0w}cw&qW$$BW88cmi99C_u8 zOX5hGVw_(xo|Hz^(?0#vlosUA+h1&kf#2Klm|QbknWM^hBMm-t^A1ACODfC?v)UHLJ(e0nb2XPYr=HQWJj#37gW&Y6TtrJx$>J>I{>(1g zjXlfN?xDKq$hu*B}2C@mv}wvpl=_)^HMqKoYnmu%yOX$BX7=T#oS#aonQ=| z*In6U!CN8s>m8Tr{}oQ(6Ik@io4FFFe%!m0kxZpG12yDp>65Cxz_=YnecC}2L#aVXw-4RmVVTz(7Rej$~1)oIX0=d0q?m5<>`+$=V{vItrq z)sW40D|*vDmiso>6-&oEavv1~1h>H*c(Uyi$zI(@g~2~XLlR8bxSaE#aMF^rt%7m; zqsw^Oce5l*)03@iEdlkgvy^f3D1F#*ibC#Jif5k;q72{Zc=1L7`M%E(JPAc`vEUHC zOUlES+gI>!gnMT3kv2H>BbHy?^^&s1IqdsZJ@i_+k6JfA;a_jO2hO(pkkgH%m+3rN z7%9@x2j%qfoCo>UDAB>qTGUp!i(flW$i;+r;=d>()+%zOUTzIwa< z+7WDp@Hc;=TFAOvf*H444Su~?L0p&@eVCWb=IF#hsZU#KY|jc2`DH4zj-zgDt|*Oe{L!bqho5+d1NcEp4J3Tix=^) z>YAa{+8Fn}uBUN96L5OcSDY1Hh;BE18{8+I!33uj{HLkM5zj=^x7g1l;GxOty)(V& zI?wmz+~MDB6uNCj8@aD@9B8*@AQ?Ph#0EB?{-!tRdc7Z+oDIV-Ryxf4^hDk!o;YQj zecVdN5_p!^f%Elu@}{m8=-HqF_4j?jAl!;%H@j1RB}GX=XA4=X6H7jq&2A_(bGJ=L z;daH>Hh-K;Y1r?5l0SU|B}3+C;oqhYl2WY(iNg^eNmqxWq&(hUq9Al>Gc^(=dLFOo zh|D;abyy#+m<%DiDYLlb&~=oxWfx=kot2MUgEzjGgH@4`zZbbPJ|&;C^V6oeFQ=2y zr7W7$woOtzazFWf9mbZg+{a{MJ!s{a`8c%TAGai-9}DeH<1_~Tw9+d*E3mdJ#J4t< z;F4kKsHs*A8Fy-M@HT5kU#sDp-gZ3y$cm}Yt6+s2kAh*Kvm|C_MEM;IZi(TW*ei&S5oy}QXfKn;6_+1j5&B#&|Q*oi123>wOicZ}Bj3p@oZ?1V4 z70a)Ltkt9Wuv$}EbL|t1fDD#fxQGn?yeFBn`*C+*G1E2;VY>zl##@T#S?NM^=xiTB z*@2E^+Ga0wd7FXPxy`H5O?az)3lrwm(9vb4^d5H5gSXAF{mDmLAHyWR@|zt^JTV$R zl}#f@**3m%-GAIj*A(_UW-n=)D4^;63`~!&#emB)7-D+{gG4W>=kWyEFg6*AlqzvY zsI0_cFDDuJK@2mOnz5bd|3c$*U-nUBHKTrMyzB7+^xqj(!LK)#jMtulzUhgqZH}_w zzPvA**ZZ01G2WgK~dW)^zd)I?_r8 zLz7upO)PeA`VWddhp;3!X)ex=k*~lu_~5u&lNG{I2VkqOeHe2Y)I*Yz|V8?W_AXHS=n1A@S6{D?UaDo=hkpCeuCe5i4MKF zc9J`tyVFK{~f^LM{0)0|70q^u%CW5q|YcHuxaC}$4p zNt^RNF%O~ag20=u*vJpd_F$Y-8ciF-N&L2-rNzRWp#5kScPh()K5lnMqf;@kd_@93 zhl^z5%6V9}WIJ2D)c~)23dI7GQg-Ks4P9!=#Q6&AscX+S?!@lbC~ooKY&-l&ZBio* z4DUtdE1BGzz-Un2l8AfnOl6(tGPrkw^Gyti0HgD-^%-iLCw z?g8ORj|%K+p2nJPYEV(-W_YQ!n~Oa>ls3P}B*){8Y~8I)_QNfXseWh^X7*~5miMb9 zBZp$ZjX>XE2lnF4-#s75Ec`-A3x zfoxOzenEJihQ8X~Y~foQlp0V@Sp#$>YTQ^h$-4^mENA09=YgbP-9v`ka#>q$8$Yj8 z1IAVth)+&f!eos`(3hCWtnqv|eBor7c|jeE9D5kgxAj8Ru|WFQG?>ZseZ(Du>II&) zkEq8#o@5&C}?afv+fV9@f$`Z(@)q2 zr_JQL-=Bfvu0FcI;uCF4Pr?500!gHKfXrDT?0upICQFOKK)Am>4N0Q-_FZ5%d>{lo z`vFSpH?r~w8MgZ1G=A?o5B9V~l}*~L$-IvFu`ew(RQAG5QnY?E>&@}Rup#ld_(BH_ z*ZGZw7mm<#K8L~|)Ix;#l8`C;5C7eLigQ*wQdD3p|L65(R5frEI?^sE-|mWYv-`N= zmR@Z3mi6rG4Ldv~zRc+sYq1ka(MUCnu%!Ign`vf}2U~Su94B+joisE(m~=rWP|yH=?3In|e#IfwNs8r1bPwkL z4hdmB3)fO_+yQ` zYI>KKjEi8mpUh(gzmv$?HluI$FFlValeKY~{iwRQ^kYqUDyLwN8IlJk1^7 zx(E!M=gREhd_BtO44}3P))Jd@)7YJ*YV2FV04g8(oBoUoWDk^j_%U&}Xl`W%o>`HD zPd}M4d--~_35g<)Vk2BP*iGd2(S$9NOl6VhkAd0S?c9^_N0_K6FPYegbkAi!d;GzQ z|9a>WeoPw6f*(ESB}>$(7fIDVgt8vkn1 zBT)k8(Wc-+Zu}NKHu0?yu2~X|W^Kx3_2oM}pK={8J9VO`kS*MFpqf*8xs?=yWAK@k zKWykqgcB?SRjNm$S%H#7ef_tFw5;Etv09HME-WW!Zw*Q5rB&jrTNz}qNJnyanBXnd z6PN)hV_8G|NcPbGBh_s6C3pQ!(Vdm0IPBwO7`QzZG|De?6DGeFy0|u2u)zpKp&ESD zW-oBiyo;Bg=Hu46_uHN zDw?lggW2W^m=rai{O=w@w@J|?m0SURh3ODBg<;y_UpTDw1+O>Plhe)5gZXAo?BQA& zRvj{c;>P;Zj7y%ZH{TPFr7xwimACmrGv4r@p2l-8zl1ZvM#qfLOE7=!3>;PW553MS zuspTx!t5juC!AAbCKK;r>aIWFd|QLJ`u-FWHW>(+h;lMHZo)!e&H%Vk0_o%HfZGs@ zalx%v86C$>+@45B`>U~aZ;~+UTRP|6A$YT&`?AHKoVkakCS>~i7%Vq05f4~w2bx=k zGou-)T>Plbl$Lvsu|f^EVeVe1lT zyx0ZNqqWE@^Cy?>dx+OJPLRxN%7NEaVu3S%nJXSL9?rZwzz(bR;c2l6>&iRMCfyxN zha%7Lj~&g}@3C`e3^QVNg@tG@wVRxGZh>v_AGs~r)5)xN4fBqRLDQHJN?RSnt=neB zlHSBKOSY8l%5%mo_GP?hlmW{+(Sk;OH|gNqR9Ik@EOcJ$X-bonq)=9mwz)1OTes8L zbAAR67$d>&Q8j!-T@kqKm8E|T(&V`B5FPxWEcu*zjF)ryjn9KCDRS&)+bJ}W6b{DV z(C;~HdTuEgJ+>wFBQ;=tAqD>JmnNgAM7}Ph3sPPgGrU9aC}J1;7W)#iJv3R9;LN*T z_XG}cLicZ|C;R+HxpC!|J2=$M2}ag%&>iH-)|C~pAH7Rx;I+5dk$DVaB4b3q7KM?< zYA53Cr6pgM#j{iML0lEM)$9{1#Ked)^zuP8tD(pk<$HG%B16;X*heyiWdHb-#IAdH0s6@QLzrz*z_-_GNTYR5; zJbnT*y&DOxJ$ESC+7YvZvssCp9!yp?f~357=$n_wzdt!0)|w53+8bun-|wy9;OK%E zL$pb#VhPVrk>aR&XLfJ@3)_JkPGGcx5=)D7M*K2?zL;e(=W7{v+;Wy_o21 zPAe^iZuZ?o)}vty&&9HQz@0AeJLSm^l(zC~u0FzY+lAw*f6J?S2*<*jOupq1LYH-JjRn2nB{WayC$$v#{$4F z$pj*99p;@k^`~0Pqnv!oTHOA47X8tPqtpr0s7mV|H$6Op-yJd;7F(`o3*z2l-?TKi zW$D3U=E=f;y$NKpdb~t$PB~o&m<+ldp3F8S6Qq70MsCv%s)Kl2}Ie3@@<09)|mG4PuQp_Ee=>tRJ;+&zH?L^ZPm zE^D~}I}7@}>bbz!mEib}S!{>T5cV(59rpDoL)N+gh`bca7qw1>iI!f}rsIkiqTHB9 z%0hagk7zQvnL0*w;>$w^Np#qV23vbWdv-bR>@t{Do(!QuWhpqv^Qzc%mwLk~lLBZs zbc(ZAQsWdupYxwa1Tu>Unq+j)67Z)5h1)D*rq>m4_C$RO6!u+X9{P}3XBJzpyqHve z=h2IBF_*p495&Po!^EK{;KAN(*7$lAe^>AcgvXk)b&GU4Y3Cz&VqJf`c_txtRYLb@ zvM3c2e&oWo!dz_nDDiCY4j^_=!au7GagX2{}i*uysotvB1lO9A;jDjW4@VHpvkplSZ%VoTu8kV(6Yg(^TU|` z-)FpOSR$SuycqKuldyhw0uCIxlpPv6O<)9_$Nb$_*$35KTtr_O7F_`3Jr5k8V5$ycAf;Rqiqm;?YWC1pASGzKQaHXFimn}hz==VTFS2r^k>S=zM?rDR;=^eKCb1JFKzetml%4DreT`T z@Q{HMYbiSd3F-oG^ywAOX45e4s7f8k_&o(##Y8d=6jAZ^HasxuDjwcoLOX4Kic0r4 zlg$H7)_5wGiex2{v&XWa#IFmkO$uh=s*@#|Rpk=bTchxSL_{t_E%CE;1MQPOhb;!- zaD3}{iGu$ZIBOEd45LjYo82Bk!`F6PcG?Nd#t*}x2AXWy#38h~yc$ibms504K5ZDF zCTX8uz`^dpY+vgb#`7j@%z#ANsH(+u9<8Fo20`SN*T?ysSu{AFvN#5!P7e1bVFx^tgKl=p9to_$0a(WUKF7+lY*<#o@ z?hG9muv+4@=&B>b2e3^@vb`Bt;@t3G?S|s}RdxaIDlZ3T?4Ac3(SKtV)AXAioF5!Vw#Mk z-=rL2Z+3uQ{U1l?;gHk&{_&=YG_4{VI|HAV;=Q;O%UGMknH3*QI_9Gm363z{mV+Nlda=rG-G$>J; zc|TpqpLpMa|Gn*nfIDt%W3>S_mrlYQuNuDN@mp?Vl_sTm9Ea5d-I&Q(YiiM|#iRgL z5~5`IS1m~nr^m#@xwLmwr+$j|z57SgYJR}xbCcMvm4JHd{BT#HEqf7L4}Naeg5`na<%iyAG!45_byw+g<6I{Rd3C6a?`GQ&}|k2F~6X2m3=+$wqRC+qu9VC4pYx*LOS=jG09_ zK|gVKyzlq{GoA1{*}?lU_sT!N*Qbj&Iw9fEPh5Wg6UcxbTY{ zrt4;4+q7_~p7$2CPN}is?e3YhU+JHfE-3$B-k;Xf}#~DM(H&=xj2DFKTfFZ zY@$n^$H00|HQx7qAvswe!nUu-fenh@sBaO-GV)UB_P%;>_~p-(`mdu8cLxfxV~0v= zm#+aOu?utZv^@H^^y1w9Ks4W_M*^2q$G{8d_x9krx7GaPkbbOk`BK>JM$oDlmc%3?=aOQ6Y90Q z;>6ciA-i}4EE=W*yE!qJ?EeeB@&idNq!I0g-U9Ve#W=CKPIRVi=ln8o&Li;K5elB(mR)0WBUx_~^GWO2Tl&ko65iVbwC$ja1(HsXOF|T`_w%`4X zs<%=&jk=+1sQd^p)Q#ak4_(DB9}|0{pQBK5W*^!+SacDWgfhdh5^$bv$UJ_}BK3Pw zFhWKiYOIICfRiKn;)rAX(yPb8=x`<`m3`yf&JU*akhzd@{3SG2%n-6~m|*81LpENs z6FrVgekFy^40MA-DfWFrZXhk$Fc_LXnYX7 zmkQlWQBH9Qn|ItCnr?+sTki@g@>qq{4_<@I5L@`fc5(;%-+~6&X`Hul5*JeKOhGGy zx%f+J?A!5yEMWE|T&Hz_X4nLg5U`B*YnL!J-}f-XQk4q7=R@vOMfSBxK`@?Gk3k)C z!OtUuoEwKz(Y$eV+BJ=OB@Ng|M@NVgGikO*Zo!WwJE&`IJltw{2CzUUc?edNW| z&J=O3&+G8glo-}((Syrd<_JI5Ji#CLG=&+nu8~&27PfC*J)F$8r_^4t8}wofWoRo2 zoZ~Sz^k@#}+mb{hj6c))ZMirsMut^iOvTN;%ee8kN3!n~8-%mVW^fJ3N$l%VbBzAr zgv->XLfpD?>aT0TQq~*;jC_Nsiz3L1>&0rv4P?FXHGR3WoCFz%FRoihG z<)IB}#b)B}@F~^Rw4lO2873?CM;~3xw|+KH;B$4E^whBc3Ui<6}t@J&cLhVkM(X7x4hK>9WEFAu{@ z-(BHGu(~i*WhlHEX~E`;v%D7}0^MC(g0eGCP}+s@6q#DX#~scD%ZRt!p-zCF=-rU{ zLjeLNif#z43*02jT+$6#Nx3`2`OLVNcrvmURb~iZO`!+RvS~E7Y7MppP}#I zU|z#_0{)YY!o#hfFe=9u0~Hl1z(f`Y#2w;;y-(u+>!|WabJKar%bmj6edcVqvMbp9 z&7@zv7x_J!3NcZy;rQ}@&?@%OCZE>>h4c_AsVstwOC_j|BbfaL8=4@l5w#L0z&*P% znB?0)iB+=9sAwXkR#d=%zK`H`UNF76(8+JT)Qv968`vPDuW-2FSb5f(8Bm^M2I4t| zN*>;&v;OxbRrvaYXdW$RlxgB)njR1{b{~e8n2=$I%{X8 zp!yOk@eZXbdOY=o$`59=%UbLvN5=9WRC4gU=Wje8y^e*<=)fC&8NIh&K+}9a@(*{& z;opr~>{ZwTdK_^d*M6RWWncdHZs_aSG;A2X^=QEd?WtsVM~$rxGr_wxZTRy@4uARb zF?ttv6$We11{Nmv3{+n6TepU=o8?K|iz_^tl_tXce$~`_uYoI2?BoJ89l>asH?wX# z0)sf>g8_SU_ zl4o)auIE5|pElFEXT|=zR0^*$?t{jcW#s#NKEAem2}7E`(DcYXykzt^G_W@l!uK2! z4rX7X@&bEWx<+Jzjmw9;lNxMEXaFqq(ZjB&>lhpV5u)z9GKH>7^!QvBd-&XfeX%Xz zea<<+$763HG#O~&(odWn9+f2IJ`@67GRALqOJl>fON7b&MpOIY4wPEv55321A!S1^ zue{M-vSD~O=-M}+uVOs27M|mvez_W^H-Jeue(;URoh!(Rncuro?$qEXHCuq{2Ax*rvz+r%gMTFmb65NDeYyy}2umOy*MLGDvlGOEj&if#lo zHr{#+>sdaU3g6A=y1oo#KM&PmOSvj@?q|U?j`t@o|&XUHw zfjNrrVZWpr3^LAgMjn?*{qH)E58nrK8)RTey&Y{GdxDe~Eu;QRF0xY==O}x;C8toS z#?Ho{6SQ12aFp#9*3w{0qUnuW`(7Kq%4V|Y*wHNe>k+z_b(#`>i)SPC%WTX%V1#*? zvMg2HSNYKr&7H!a<0A>(s#_Hb5!7daH{4i|u56|nG-Fs5K&M!ql@9@kYv`)emWq4}cR zrq2nw6A@0c6wIh+4nUt+74|pk0_oJc3GxeKq4NDVwsu?^j1QL;X4V{})ze-}j+)(r zV$U2X?8v}0joaifnByaUDzkcDOUfD<#r~|)GMJit4 zT3p}o-{(ujz0Cym*TKymK#W>u7+1&!b`RbyaE|ehC936-i^#Rto%_f%<`_IEK#P+(L0@ zt?$E%LN@R%3&SDkKP$EsKEbLLLxmdyMK0afMQGb7dXJ1U`M+~7&}lKFwPt)gIKOdc z!#sAQAlrj=6%1d6JabTJ*2AeQhqM0tVVGy)38(ezL2q?3mUSzl)@vo2G&GbRCUlYP zvo5}o71A;*b9Oz!iiJugkjv9UbRYxJWK=!IjyQu0%bifY!VMaJ#Dep_t#Bg4iufu` zEOE|4^Bup)>uM_=EF*D`q=GLloPfPA5243vK)k5Ry_L; zS-~F)Xh3yOU2?9tO1zShuq#*e$DXp{BaT{-<{q2w3ic1HfV45-B_X*{5>cpA+US*8>Q-_|u zZD_je14JL6LnVuCaaHSL=CIzcyyV+qI5M{rwiqR0o?jLXxvMBV@g56VebeAm={`1Y zq&2S23Wv~(+j%!zHDLgE6!+!G^T&V4QC6rPd><;e*EvJL%_PZZy zhEAlGQ@`Nnwb{%zMF%?4RpkZ640!M{zO z^SPnQbROM;;WA-Nzeb;gs_)#!vyosQ&pTAl=*Cgwm(hP$a%oxZGQq|Qt1Wu24AGPT0@vlCgdhu9(fTL&YC@8zTXCNh_S0pOOI2iMlVfyn*V>`vb= z&}YL$YOvnPN1TY^8YOSR_}nZoQ@Y2U85BqJ)#T|+svcOzEE1}IO`%a4Vh&Zj0~8)} zqqE&E?EaO3?3TYFvzZ;iI)6W*gMmEp;!Ng!Pa+=iut3Y`dF-8%$R3lo#j_)Nps(LI zoS?W5Z+BPVdBa3jI!c>uZ>q;DQX*T)--96t@wA)su2sVKRaC4=|zahvIqua2)y69B0F4M3IfpCw!3+j5gH9 z#h(s^uU}GV=grZm9jwj@FVute85I~ZCmm{=%vrw+g@;-!-i-8n}{HybG^pUvaK6i;An?Pq3qe-eB0;4km!evrmaZ$Up-X-3-( z+2#jUthDwE_tttUW%cwIf4dp9#Pl~^+1Lb{#Y5>%y)%ZrUWAhuy?{B{|8W)iN|3&5 zknqHD2d_|K!Qw1KI4}7K7%O`pw=|WK(mqeRts=6$kGQeo^9?ZTum+3Sr;qJ>RA|7h zV0LiFe!4I21e#>jsgl)TqE9OCS(rpw3)3J*ryjBNl#C~)f{~B_ z5A5zqXe7Ye8F%T{bqQN|Zwu2|o()a2V`y*6NGb_F0s;2cEOgIy$%2*>bVPa*<@YV* z1817Edoq!fpg9^nrB{Pn#a#aKd|9?`;v0$Pg)){`I1ns*>`DGPK;6a(h%T0_NZhTa z9ddwlvw5hh^d28tx{#0WQ{28d1OF}-XNd**qQCn+C4bdsyT|uuu{M=V%4QRJ>t~SH zk|oe`HH2Bj@22Ct2R-lCLOpZuaOWp}1uu_qF6c)-(cKi1S(DCv`)S6;I*OjbKVhsk z%m}I%axn9)G%bzDr)vFH{QX&qjQ(w74`d@K=S3z>+ZIN5yJUreY!_kq#wyBPeu^$n zv!QY8Q_DveCJ8FKIl}Gw;lhnMUc7?b3Q}FZfPPnqj=&S=VP)!6m|S&;SFGL$-CiRF z)760_B{xQ}%KL+-uCApaKcx5*Gj1_eqg&MVt%0r?Pofu}qF|b0AYT}%hsr0b;g2A8 zxPJWwoo5_gi<5%#J4XQ&yCs6CtYZvcaAIdJPsAP3(2kZD+Krmq#(DAMjt0=ezdx5e{V3|D;fmvM|`5z z-0Re&E4o1s*uqEC6PT=aTy(e?vMJ|-FtYO^o;m#-YL3UC?bH!8v&t4_MaIab&Kg)U zYaoA5qaR$=UC&0F)`HjHcFZ_16Ng47<89k`7;qyImX##I{dRM1&DIdA+ns^|$9A)2 zdvxerT(IPofg9WL&L7%Bm6?3y5_JEj&5cu4BZUhIxH&7*arB-aY@oS}VA^8C98 z&D9lQfKE0J)?7%14)0;_tI6=BMO3`l4HQQ8mZSQ(0(djD0ol;sm{=vv5(Y>+X3J!- zLdPt$wN+%1(b z7+x_-*B~OPlx@qkEjAqt&CI?RWZx{dmLil1;OBk!E zjulR0@z1|H?3d)qd|l+2%Xm%4$=}sbcDWhTT4TX-RQj-o2kmHQh#Cu)>d$iL7_yp{ zO>A|ol;a*TKYRXV68#zV2)3O23#N}=^V{aPv&-h!nc>AxxcHC<*H~~vy#Jfy{I*R@ zal$G5*J#RV%3NVxVNFa%*vT`q92TgrQ+{siItDha`m!H_g{e$TptEIqUf+6ogf@O38S+&?ICYJSs_ zRrk>>t`{y|{e^Ygtx##AE-JbdQrDY@V0EkwWX{eMxov~k+VJnN-ig61&Y4+j+(*CM zbGX}Gg$el)oZ^s7m~c)!5G*o+aefZq7wW{;iXP|ccW88sTca6Sjexx z*q8e`eL=a5c$evva^;1g4UjR>4g3tccz@oDt&(%Y0rOVkk#KpTv9n5?F$VxF9f%vN zO7LEX8)zS&<}hxX8bxgUD>VBaoG?|Ibatfj7k4RB z>BYBvNd9$L|9K{El4y}pNHrSGR7chMzTEKop=GX7Rfq4N$~X*g_eSz1ob0bJ&8FJV)n5`9#8SspnroCS5P!uJz{X!r9wFzCxv%9{k_ z8)iW^=G72qUC32wAExo?vN)0x&+s$1@bzh{I60{h-r%JgUAP^@FBp~rn zmD6?FguygVPM;n7Z&P{D+(6V2x#-g>wAh^rS8mq)p-96w)30rlggE&CxGQomQVM_a zLpHp?s9m|(pyPv5PGZksM=fYQ4-|P-OK5o9dI~=%p7CPyx%ONa_@mwikH)otWmE(l z-X9OvM;n=&#xHimsFcl;ZRMZztb?)R`mw5w;^%cLgXX4&Q^Iv`dh{U;1ie;B3^2!s zgMG;(WB`7u%jA~*HpOAnlW3h)KU$-&z`E8=h2_R`*jooc5wS#zzOjv#49KL*V4X^8{GWMtvdfV*aS>M^^FDHca1G!A}9-&Xm zpFYB|rNePMuL(cAyxHB8R+1H}Px*G%&1ZaFEfn3>BHjBFS+R36HR#V^Str$KQ-?R5 zS89f_V)xBysy+)UKMY6z2JyGW@Al9ykzeH!&TSYs8?!s2FMber%8OE-m(C*MM<>=ZYuV;GHH zHk(bAg>>8Q(ILYJEE3(T|hZlD~l}`|jeB zy|VD5>;>#!@g0__JO*WzG*-K@KMR)mh8-2J_{e!iOeJ9sj(*huekIy`SZ)9x_c(Ba$zZ5cbt7XN$= z8=Q(E$l9NN8SaKDSC-Nk(+GA&cQlg>@52`Qo`khk8)2?~DL(g9q=lD*(U)t4fe#GW z(a{=wlUzIBDCWX0tSdp+72>{V-A`J4B#2^)57PAP`AlDW9=kUuiFLT7VNX>p8uMmU zA2@@4tq7Iu+jWXbm!`qX-+m&``vS~R?!(SFOoVlkbV<3bGjyD@gay$#Xp^r(&8`(( z_2=et2eba{r|kf1TNMuZj3I$PvbS9OQSnsFIpjU*T-36BZo`C9hsF+yD9y z?Vg-LEAkVmulii7qOEkGSeGouD^lUSbC~XX8DH5*pw43{9-n3j3)e27d15wh&bd(N zFLfU+%oFkPl}rl#wupY{I$^~V8#egs9lqsmFeLC|h9zBzTo)YYqNLU^>9z8BZI^}Q z_yY|x4*yWzxndsu_~nf^7rU^zLvpELPBj%gjs)kTDt@9k)j8XdNtY&Er+FWz2>*Q_ zMxGJ!LW1W98q0fpjPID}UpWNjKtj85c5$DX+UtleSD| z2kZ``>0WC{!sY;g{M;aNfTjZ_sGGvFr%G zTJTQfc0X>jmpBT?dt^Na1FZED$8v8wSny19#EV0kubCpbxs7Zp49p5K~50dT@%LY zI;OJ2>WuYHNyO!+pGk_vPLFVMCtDe_f^NI)q@kM|Fr>toHu_AbZ<)Lx6VysBVRxvg zLY3#WBy6)mD#mJCvg2ctAtmn&3m3hePpraFyYw(_`XI%!--k1=O|f()PFW~_^jU z`)S|(&)DW?Pgx%dU_o*|os|B=9hq{5^Sdo2_9S=FLG8CVrI#oBLR0Af=P}f+I0E~! z#hy!^3>6F${j+1DY1<4F()NBpD=Yk|c;4^wtrCH3duu#j&f$6QG=J4eh6(z-z&TfD`lV-8+8=G0i{B*jlaEzt@eomi8 zt`PU=|M&xltmx}T&sfu*Ijt|%|OcZ zx+%_5JTZOd2QV=TXDdX8-pKnET#0NauT$>IZrfgmT~~g>47mq*yVj??k&kCSbJxSj zxBt<&*N0h-mICZ=z08l2^(!~_?<1J6aN}CWnzOQggV-;Pv26aON*KAkoI*u@dH(n# z`0n)(VNzC9YobHr|Gxs9X2gdq-QnHXr8VuEnk&LC)KY=a@KF<{eDitXMgNjpDHZ(Q{2oO z>E4D+wt z?LCv2m$eCw+2IbdMIIzuH}~*XVz= z*?NG;7K!Ep?afhxzfDJ)VyXVM9a%d~mpDvr;b#`T18cJraPf=-I~eRngS=GH^~hpK z7-R?M)fSe|TphyqPPvGCE0g#c#}`3tpM4PYXc8ZMEFXqsuHoWe`EsRVPN>CkHc@^u zKPoW>7Hm$$uUG3~UmJkz@Lf19e=nD2tH5Gd2~KxAoYfJFD^BM+NwjID`*tT*5haoP^F;Rn}5A1#Uh+j4#_5ZxtHO%QRY( z;yYUeBg0szd18TW zexRq&c{>^D(qvL3O8Z|tWX}z7Y|ov!NN3*JSXm|DrRGh z@CC1zDzGWv3-~4}XKc3X!ZEjJ(GlK_opBn@RF8^`*F(O%fynzfU-+_|Tuj*5)^5lU z9g{ou%D}Wyi+B@F3+C8098KPulfQf?)arYIRt?a6&Yp4lv)I!M%FKWC9ZIrl1g!&U zFypr!6Bgd)m&sO>j;D(dT9-=>7y45E&j9it~AYqbgu(APwx(SwnvfIn4y9sWi%NTOEZOJ74kB0 zrKHD+7%68%6wrq5(~QCScLTHzjpmPaYthebXSuJBvbk#o3ZjD|jC()q2`3lj&fbW> z&wRc8ylH$rN|_vj2_YOVx|m9KM<3AcZfQ2~T`IG0UBh-7B;goMDdrDpIC#Asj=cJn z;_d`V)aF?;w~CwS+^xXOM^}OcAE)wXSDJ5=KL);8Pa&%!=m%R3p5pv4tb7yRA2bnd|o*(kVn@ko|prLRKZ*J$fs@=K5$r%&q&)2!^Jd;LS*~3uO?;yT< zHj2#6hY5#5(`b@n4lEtSkzc=G=zsMg{O$H)yS4t{;~{a}r{&6&dHy^4O*f@=5ysRZ z@?7>7>j^q7-*}5-b@1U(FMG8~nzT}sh3W&71gjRnE43H?|9)Tc-Hq+}I~gy1_2#|o zT46!MbLe{4SU&B?P06*$9b7jz7mZRM;oj(Ys83R+RGD>XzfzM_=g$Sl&t`0mvKd;~ zdef*sZ#m`bZJh1!_t<#xJiL1ANB0-{)4|?6PTAu)x6At^6Y}np((2PRx?>NUH?oRu zm4)ytFKMu!3%s~{eKYa6^<$73`I4<3XwQ!IX5%d-Wx;z%4()w>2?l=D=IoZLvSidI z(^qBu2=V{v;toshp`isG@(4sXD=oHfgfiWfe}|C^h=w%#@p7J%Smlo{?s@5C#=YdQ zpemIuKb%BHWe)ViMNyc2VF{hoI|`Oc1^hs@d>YYYz~XK;;+NER?&PlXl6qfN{HYc} zT8jF#aV9!{w|Jx7b_K zaDQ+auF!L5!}DH{nVdcJeC~x6&yKPq(`#w&q&`CJyBBcSJrifIi;_HQOD6|4Q|2e; zm6tqNOOEAU;NpA(9IbL$%rpzoG?ZoyneH^IV*kFQ|v8d;`qSO6L)*Rtuq zV#uZW5~Uupgoj%uf{Nrg+<#MvTFtI3Ui|6m1A6xCDJ@$KIi_*G#6DcnutmV`{AbDbd+v8S50PYj_QXWc-DGZK71 z7x0B!f5M2ic`#tvXkq_}2=v)F0v8$i<6W(<`2LfqiWGN^O|?F_{Pj)vqE^CBQGbZp z8awIIzc?yVYNnMwmb6Jd19Yc(upu_%(fwT_&XK;v?`ru11LEI;opc@aovO+DKGMSO zS@W2WW*R(xz62Ft$8#M&ELn``Yn$z%O+V~?n9lY7P~%@u`y)-*6Ipvyi)65}VF(n6 z%s9ydS*9HnCp2c=qwou_%3E#3^QOpOn(|qhojo&{35#>+l=&svT4%$Ro;ps}Bc=!s zl~vi(xUbkHQO!Swx!#3Bt@;e^M5)1w3Tu<$K;a;9HTf4}E%~;%* zJWrj3OHvLHpsPk7i>Bkya2c99+Lg81i;jqDBf-8)8rKXsNXx6cao)^ZsP{39Un6hA zj29+g)U5@4azZN1IBCzC8sl)}pG-QD7{s+6JTK1OrU-^2gSj>O89p<&W7?gXRFiK9 z7nMW?$ER%UlNF1Nt?%$le@h3qsPkZ#Ah6umk$hCeRPeog9ev$WpgS=ESGg{w1y>44 zJEsP(4^3tt>`vpe>(8L~?+MmwFcLYR&tmto0F$?^XCJB@*pzY$dj8}El=OXrZ@Siq z+~*r0`dKM-+a(&XEQ*&~+le+W^@OZ=Ino(h4_zLEgpuL>nf0F%K0-a0JfHnW|Kg+k z!YFz8-I)c#`?>7aPaF0tMVWkzd$~UQ!+9V3V@%6%JMr591O zyyI@!p97lIF!(e-a>g~RpEQ{Gy9TJ+)Qmmz67lDWX72K^5VWW~P9qhw`NsMgcuZ;- zKkfWBsDF1Ae>~5p=ObR=CInPCa|7Jx9OQ1C-B0TS9>c?%BiJ7GTCTCq1SBy}%g>zu zLF;`EkoV7zbn<*RJr}uc9+Qe;*|4<`mlVv|1YhQF9Cu@eHDmF9z!kjsdjoY}&nNeI z6Wm-=K%NI=2^=nv%jU~?N@TL@{T>R7nw8+I|6Qn-*s$k;T8tm-%$d|*#?0Y9u!7%9 zzFB&LdH+9jUFsc$k?u)7nnWWOxXKP1s_=S zg*&42`_}VjEKPU;x1U-v{k`6}`p`9g!k<>ux4q0+sDy~ldQ0-H>7t-J@`C(pL%R0Z z2Gln6qv@?naJp0zobHer<;R?bJS8__{(dFiV&N_LJT4I$&ke`m z*if3#W+8NW<H zR;V!z;%_kYBX~k~`i~;m6JBIy42wW&cM?A2jKia{{cD)#O$TY2xk|WqL7})Uvf3%lCVf1=-V5uT|*K-^*qTjNy zN$xB)WgMRS5Xmjvq0O2!r?aQd!C3e=4fcJU0auqI_b^_bTc14v`&HL*>oiTVesm!m z99~Gq-+y6Szv1i}w}zeYJH>@YJL1ds0{A@Qr^rS{Xj|})41QVhJJOP1#@unVedQGP z=U)(iNIC$e3#BnuWcl_#aEcs8EAn5)YS8`UHev<(!rul3D)aZD& zV#4gQwy-~UhB>af)z@*;iWK%RWE{i<8%UIVWB9kpxxDNqZRnAG1$*9aqctb8Nk`6; znlEp~z=`gnSFV6mwMCxwiCx?@iwH<;8O-0B{*U}FyWygnP4sDnD<=It1sP8qXvn+Q za7El5mS(RMdxxA5jK}Cx@I#a{1v0DhAVX;be%e4Ib}B%YhL+4^KH_(w?puy;uPB4u z8a3J}=6Fvge#SUS5?ix9g!|920`FQhV%ym7a2Gy;ZHXwlf!m3Vo=D6(spXUA3~^0GUok$IgL{k^=D4Q`&v7WdAk zh6#pPs+a*L` z2l3!<5`r+`H+!KiFYK5XLEqFB9A}MOhVs(6WiQwb8en*gR{CD0r-1`mr?}r&^4KFf zb%(H|!wPhK-4J-X--)$<^C4aS0dDcU%Vgv-F?@%ZN4Rb&&Kq2Cy)Yb0blovbM}a2S z?O-|@0{ApUQ1F@gu2^I-9|r%KqG(?YMT{H86N*$j5(4c9e=f!gh2gQ_MeC+PwNYYD;7CRmC&%2hM> zn4f$@*nOP#w1x&c9uZcYmKVmliL9aXPB2>0l3jbROxwj?BCUy{Ad!W;@XAI!idM9% zcnSTBbB5Lc4b)T22Px-xYW_Jz=vK95ZfEbJV#zqLRCz%|Z{*OFOV>H4MTemN<^xI8 zlz(u;dI+tWd4?-adkQx|nZ=CUgj*xCIoppQp&g!HNFX!??1Iyh>!;5%s|Rwr`Yoj=+*wO|jre4k0adqn=*_d}e$?g=P( zdxb5aDBQYOhdEB2ihj~t8GEJ7Os4IjCIuJv&{~fEb;R+?!95V|s|SxX5Ax%`?&o_w zWvFIR0bTrUL8h}`um|=hsiDM`{n;iuZp#?^e9a!s-TKk)d|S9U^FQp}@*Mt4TnY&# zYuH$u0_>1}OY`HK*%`|?46~d_i}c@J%sef}7Ua0H;(n#5J-J=HC(K~p-cfXP)J)++ZUTEfv67y=vlEgQ zYqOg1AuP*EiQT>t%`QLb3&WaDg81U4E&s_f+rgJm+Dw_nPdo)H8y#@awFcaz`X8j4 zyHfW)RrWwmmvv@dg2<(Hc+IXJ7YumE`S2y$Dqg^-A8m^2|+*fVCQkRAL=e2=%Vlzl5%qS(P(G4ly$v9F3BmZitkrw;?I z3rDfEXa$X4`U4*veaQtDsjvi_;dXfa2>)C-1V0;;n5QyhT|wTk(2iqu2Va2b4#wxL zc@#VUJngI;#X=6}^5=irV|`m8S)6)D<2b&&PkJn!oMS4H9+AbH#N_eUa;({oY=1V& zSdA?m=SIt?%hJDfdbl^Qi8u0-WoFyH@j8))*zG?{$hxl!Ccpg&@_mK&_x*$qx_9}D zhj)lsw8wC;@*vIp+F#gEYC%61N-^B70sZ6Dc%PYHz~%j0I9uC@*E_>d(IN`w-FXC~ zH23i<1D9}V@sVW-=6%_VZBekfP?q)XbpqO(56TWBDH?v^f`nh(nwXcg*xH^(&B-LU z#jn8M^*C6}-7ox>eg>{yeX-*2U(Rj)Po^6;jiz`^p^2aNA-~8|I9cQ*X!||ow=hfL zp1wWh_sJlnJ7Y*hHP@{n^3_y( zxj*5$bnBT0r!`ui(;1t{8!Whqw_i12Yb;0MC4Io^?IGUT-Uj~K=Ys{FRncQoirHvC$10{(Z;FpBi0Pa&QdZBa~{9d>B-n=7QHzWJ+*>K=T-)FHJY;94mX(JV+WkyvP#S* zhw?tkwcOrjJ?$nC=Z>g=gV#(OejHKOO#jS`YMh0vgnh z$MZfruwYA-IHSErc3TGtGvn1j)8yWG?*nhZ+;a-9DM<%f+=smX)WfnX*|(jN^az7qdK9MGkR>JG9-| z#y?$^g^IRc`EBxh;U1;(veFBoOJh9j^6$qzoO~BcY#7eduwtG?A35)j7jRT9 zv9u^=3;VM5F8$Z4!j6~3vIA4o_$6Z}z`^QwsBD z`W|GVZ{-EQe_GV5zl0t3&44pqvLrujB}?gOh9z3Etaff1tzB)(?cR47j4IvOpZjB3 z7+tWiTpHy(g@ zE@LS7`y$%QUxAO4bHTn?mX^HkqCE>g(7)$Z)TOb7ow@e}SG+Z*We@)Y^N=(+k`Rot zUu1+y(pv@XKOJ=8!6dr#eGkbSox)QmE$MrX$U7{ZjLY*)SpJ7~ROO{BY{|C6LH$Rt z0?#g3_u?X*oBxq-+h54cJ%TCV&=4Bee?IkOjbaXKP1*Qc-IB%+2@dM%`@u?Y0cWKC zLXy2tz|tpw#EgKB(7m!wn9{vguxo0eqwB};?}~cBz`7Eqi@XE*+bY!Y+=<75Zan|A z8_R~MqKcV0ubJR2x{SuK`{GPuQI#8=5AUUGO{0W}o(aau)AaAphqtUhZzhZi_d2SnV2QQW|h(O^nePbUrNL=-s*H>loh$ZP?dD~Z{pGu zUGRiB|6I|&yZm8Xm=IJRN7?(0NPF#5oHU>i^q$v(Bv9(|(uWCPDejjqf7~E67#}9< zA;s`xuo+lN(jZ6A0dse#u|B({gk>rjLQh648#^F~?ml-yrSfQ$J?@M%54?v3J-X1d zW(vML8BgP1+H$Pi3J>Jmqb{ADtjKQw=^c`x*h#j`dRYP}Tr4H+*rjaH8B3PuC@(yC zvy$yu-3!~zF7b^QL`M))B;`u{jP@30*I_y0#4+9{f%skAgT zuJeqnkV>+$DU>oZk_e?G4GkqNsU(St#`QWcWo54rsfZBSd%b(=kxL4zg99@d1N#m7T?RWs)2Z(f6At(TxDj_WAX0pr5L*XsjK7tv#w`9 zj$-h;4`fFjV#V*>Su3YTL)I0-j4{(OL;XkN-SyH!xR_5fymcJs-5rG57u-?pUN)@D zH5021yt41HAAwNsWu6d>~r8-bhYTL~}a7<&-+Ji;LwIR#;cZ5&t*^E&q zCbAng!(1DaK0{7^Tf+&Tr8sE&AJ-Gl>Urh%e7N^|3obdhhmUI+4>tQUneLAc3O*$v zRO`+X)}>iP$81lq`Mrw_mUh6Ev-40}(g?q=4afJFikaSo{?rW#U@{d^%F&(5I(+fQ z_Hl5(M@tw}Hbi(W@tE?=vgwWdA%1k;NtaohEqK=fyZDwTYfy2I1$C-FVTKmd=-cIB znw_VK4Sl9)`b(Kir{?e<-dRw~#S*C1_**-A%{+2C27K;X**IAT%CM6htqz_NsIXlnwxUF#% z7IZ1o7KtHr_UUJm^>61NXI|#k{65YUT{h94Rk|>4d?*yixntU`omg>1W zJ0bM)5E!QGjYGHHpkA@7vev1PEA%))Iw9hYdVeyzp3==Gtk>n1nl7Mf^`m^v>w5NH zel_>;n#6H3|b&o`C=R3T|rERCdr{XX6~tFs|~9EaK=Ryk@E+dWgd@!$}(` zt=9`}r6oi!Wiv4CisF2i z#!~f%INJ71i|I=pKb5~b+F0&!L{aVx*@h5@co zb!Q*%jHYTyIg*G>MW@(aRE#yIr}uWFXO;=+8MxC1_vc)7Q7==kaEFt;J}OCWrM2~+ z`8tjB@F{30i?*CaC;pkBbVU*q-HTjgvjsHUwt$LCADBdIlZ4f6(lZou;l%}z$To5> zZ+)S?zu&TQ-^a|aWHVbXX-ThK=Fl85SGoMsLQvD)!*Rn5Tun5MS?kDqtin_m zuE{OsR;6`1e1@vLCX*1uqS85EW_|--fGiD zCindfOm6a}@3EIy_^?PimXJo90?pavEpHo*0(80Z?>ne#ez2-77_Z0)m8>_1-ai@@WNy4SOQsVkU^8|Kw%X*u_qG{qn$KRE)wzY0 zC;L#+0ASk>s0+7thtP!IsVvZY7iGQM0(qK!%;jJvyC&xPL$&pw@`^Ll85v8#w=ziN z7vO?b^~_bt7{hIg_|fJ?>;O!Id+r|WxT^`L#UQ)!6nl%KO&jw0Q$x3LK2~Qq>8BU?%w8X~RC~wtUrZ$b z`8NrDWrgw4RTQKzhgI71el$g~}RXUF=X|JY5i&GR@sQeF#z8uvkF z&^j_5Z;6h#2C}pubiKd%3~rtyMJLpYNl$Y-KR)z6oU56__NSI`iV?-o`A!0@_jW-@ z$qZP!eml6Qex$6~YXzMvGX-2YnH4x(r5o$SK3YbL$cE7tx|%Nud!8+4^}jx{tt%b4 z#5x&az@WjD{$vG{>c52LNi_4m*MCFp_3>!ZPzdvvZlT%7;&Jx0;}D%~f#EyF^Z&gp z7%<+6?H66!{{^IQ`%ibm67ea0l!KSg<1lvRWXN3^NFQJ;<5sz_p{>t&r&UIvka3hO z&rK3WMJ=F{2i2kTPm#zkwZIFWSD5r_(Nnd$lfBEXVpU63N%7=3G*LOuDMgQfh*w4M z!KDSF`zX58U$;$gKFU{{JM25Z|zqIj-_erb7+ifoA(3eKPj4j(vpF}d+TV}>puD< z`y9F!6uLYsIKh5QnFad?h0-gf5g6WegU9~fn3NgTRw>;U8sRs$G}SmuJFYH>8-Hd!j>v;XVC}$%TQamm)FgnOKYBS zIM<{QGTCUpOk`Ho)<1;I{&Bo%o-y=Jk6~;5PjGJ|ck{=;Xb6(;?}B8_JJ1^B#9d!L zkh!0TgfVtQ=$`&&@bH!ru6KQ5Uw(&x``Fvu4Y6-jk!Qnrhgd8a<<5?DdD2Dw`w(}d ziGypN{NnTqRxBfO?$7G7rVte|=X8T-F81gX<4+;KKEt4^R`_IaGIuBRDr{&+l@vh+a@)$MyFBRP9?ZSsvsrcl5674#1muW6Q&U3;7 z?totjteWzaw_Ejxy|VLV=}z@vUbP5KCu*@{b3!SlBBHUDf5dMdD&_L>;&s*~UOzhf zD+`U!M}Y0*t!%NW7c;y6hI{S#l)}`b+08qLd0)|0GF|LCX#1XE?dyIs&w2;Q+%W}L zs!iv;V;W)Y9v}K&PI2D93!L+&WOhFM8q4QaJH%Sa}dK`i~2H?4Hk#sz!^>EyiSyv!EjPRBdk^%^z$ zlX-&m^*GWumsfD|W<6KJp0HhihU3buE=<`gjh%l~0Zu(CSP`a-DH2`0n>ZITS@H?u z3~o{N+liF2(G?z~MvDH5R#==9CEPVQPW=ugkcD(Hr|Q#>!U8neycfI4GUzAozdZ`W z*1Iqj{w~PukU?CrLddc;XU2~e5mGiZF26JbR8^+ar7!-p^2#GPe<+w&PLQS9n9rs< zTwq@sTS*>R-lPZ0lNz`n2M=id zubfT}SxHVI%?P?v_)F(!v9GeD>1@j)zRAi5W|SG@S(kYb<1`pKTaL9vgHU@#mw7$@ z!xrw)qK^}Hl21?;$qkO=iporJ(kE-IR9=L`mlRXVFAa=*Z-TOen_VMzJ!W#|`|0Bc zvBxXUy(wZP3-oMf2J0QjXI?IBnJ>@I)GX(UhxzfIX{%|EpETNP=`i_9LmIf`66p$K z;FWeDWbW1!-m)H^v&n+)8y89Dva-ObG?SEV0xrr^K*=S|a5zMZyCBANKc zlbhQ3sGxS(J7+ilGd7Y7bnJxXwaakgh+J}ZF%dRu4-gLYmt?~OYF)$Yo#9UUaw@I9 z&%DZ&Xz8&nd|a?Pn`|5cOS)yrwrxMV{Xpd8tL+5{I1A1hQSjW-g$|aELx+Haut!&h zJo%U4wCW}JcMs=Wx+NjE>oMI55t$?flH5G!N+=dz7w8Pbl+ZZdoQf z^Q;NvE+Py1)y-t~&Y^{E5BQ6^@8B01lG3m!TG3U=|6cfp{iA-LpzbK2V{KEDRc+8(LHN>6rN4clSu~3=4IMi(xdbiGN9I$o_ z{FrBm8t1j>kj`K>U`+)z)Xtq18ygE!qBbCxN8lxEAvEGDfvrZnBQUdeo}Hh^zk zI~=lfre~s$AZ%eCwaSfQJBm({-^c5uT(k?e9$Lw67I^X|W2Es=eJ)gl|Eu4bm_rq1 zVt%=_5`>rLsB`@%DE#LQsG!0uyw-5@OLx!+WLTtWrg6=s*EKAJzVd3Ok~9`#xrLzJD24DtI>0K9Okiuth?YD zteHNZwu_#u(B7}&zeEW)DC{M7EL;z`ULBCFO`<<11peGpYn-&?Gejl)WREKP;A21{ zG;w91bMXTh9(8AGUhQnsMkjF}+Y6=m9==<6zA8$AmP8oD*n7}`q(_uHOofKoSMA$E(PI>iN)*vJ|IMZ(1B@r7#&)6JLY!U|m7?#cYb(lmUBVb-9i);x1c%0B+SzGGV|4^XjHKd+Zun1Sxp)W6Q+y<)1o)rKEpGh-0u*$PdbHr!lz?X zD6r?}?y>8C`>}nKPOx&VOE9G&9n@PjX?>-SuvqROjZ8LW-SyJ=Gie39SRt~Zv*&Zs zPU5a=$Q8(bqlqOqRY2$MMfS`jpKVUJVP$<@G+gmEIG;a&3unxM5pSoF!;wwA(>NU= z>Bcvzz8k?%*OOi(&x372*0_0I8u|JDq5OS3JGyi=Rfw#Fn=RIGVR#JAZF6SDaGoU& z7)Sk-E7_IPW5FzV61gP&qx;O0y;BjHMMH96xP~cO1ug}RZIfwy%1QFqZ*?vB(7@>^ zykO;9>R|u%TG}+SjIDh?l9|n0!|ojtd4ywSDPX1w@)O!!DvfSaVd5ejI-v|542FWu zk7W8_y_}xU^rF{NV=1j}4eMMnmlQ?Uf>)t5dK%nfcG{Ut=Jx{b+08V1bxRrwlVZ?T zTaUIms!&yCR^y2XE1VC()Kz8-t2C>?z)}S*wpOvRGxOQQqJzvs*&2>+(8RQq7V;ES z1($Eu7@$9jT-0BY!W2K0xR=ASOtQhn>?zRuIR5Cs?<`+rL*B{MK||L<`0r#4Fnb^T zyHA?^{aykK#{cCdKYnHY87gG{(LlKSbrLPSXi9lKv-#d{eXh^s0A6V<;j6dY$I%o~%?xXZYO)maoVX{nu}n<)ie?7z+?c1bP;FnW`dln6g3+eD@edoy`s0Cb?Sx7a4O^ z&ViSAW>QmrF0Fp6B6e)n^K&C6qh3xDe$Mpe^rEdXY_1Y-1AhGdo7!laQ^c(BJMFf$ z6f)OJxXA`4LH6CHtn5fBeVh@4+P?0%p)!^HMJAo+>k_K8SEi@?zQf;fL-6E8;$IE? z$-ho;z>$C58s_z0r=Df|sk+A&ww(`wgf;DKsm&H{fwwCek24Y(3DfDO+D>-KB>~2d zSV-rFD6uKvgwfJQ^yo~0P-SN)3j#-9a+PIaT-mu4_qf@y=yAkwx{UT`pc7Ts= z@@B{UgD}171ymLqLeG3XA^v9`)oS_B#>expV3P;)yS$Y=hn8^$bIdXGb~%5(yBr!; zO5vdSdHk2kbnxEmfW>C@pqFnA23;Q1y#5&Rngt-wK7o`?BB%B|hSI%OP?z#f+T!KM zZkaq~$Im@uR%w%j%>P=+&g~xMMt^6XQ>$3U^cj?5MRcn`pE{>sVs}6F(3PttnU*<*5quuD(0Y4^h)eKwhgQ>Z> z4~}e?rhfz1L9pZjkbGXnJH_>Y!E*=b8NUQw_WDqM*>qm#el3&fmw-7B&his)Yq6=n zB(YU-Fy0pLy&R-X$S5_5{8!hJ*~<&;#4s5#1KB8EqX*$#yDzY?qa2>@%ocqFhhf5L zGbS->KMm*%=Ue2p@ZQYhr1(>Grp9{<^Vhk<>DWfrXxoqG8SBAMIZd{)$(*%6l43zY zYPhwwjmaGeqP-oXLAh=tG$qvF+w+^*bfXlmt5kuO>day{#nNW9M+rKJ9l6>0W~93L zP~+~RU>+>OxXlMr$#PA9=LFf2Ol$fvQtNK!O(jMCb*(#A$DZSC?k984M@|x+u>EAS zbc1mG-Df^9%7j(RM6rzz55b(ls%ZE5FrO}Z3#0VL!>Ns}tY(7fUXryFc3WnX-eABh z8&V<0F_AOsUc~jcJP8iRjM&2og>Wr-93(D!#Z^=`Qb+6%!Q@9YbviD=2`T5`$Bkkv zD|*6j*_;41qWG+Q@_ropL6uGKO{BG7HQ1UT1>&WEy zPRU~zJPxu%*?F9s`cVAYqRh+f90GeSezB%PBKgIq=-s_*ku3~&=5Oq%rL*u4!Dc} zGFjYAd+$Z^w*N{X_Q^MX`Fkr=Rr|p$yzlS&VUaUAZ5&8;QisuVvmUnU42BdHYpQRb z%XigZVC5T@an}xChx=zL*zuFbu=xBC)PgeZuxdH$ugVFndV2(r=Yqg6PNev9J+^JFL5qv+209#uqN4}1RGViYK6GoXNkWz!P z5PoPkjXku5PdgpL{5#yR<)LVfVS>1e&(i!FU`Pl==SgM52RS(0v$uD4v{~B_5J&(Uy6vLPN z+DNx8^Z5D3!v%wb9jy73BP!*{2#apyQmX6%IC!Ix#S|_kMT<&))e%4B`X@4{hazji zq?X?F7C?sVa+J9&gYWVjsOa-T-n_}4Y+ucBX?*OC>j!nf4Lx1_uW%WQ8Go8naQe;a zx8%cQ{d3GCCzRX$-HIM=70<_CbfKTPS5x?^;F>tfOEM4Sai$ zBDx*1Fe4FETiT(1X&kGypTj;DW&W&Lx%67u&LdHY0NYU!NF4<)sp*zYF#HglkIpm z^I!rStvwFE?H|INo2)UJZDH#|V^MWzCU~ z>D_?SG{|BPdJ1ZAPW(NjY`+25>{2MFuMr;Y4yQZuV`1a80x%wamNR`d6*Fe%v8#a= zbT~j@EqAM!#p`M8R(?7gsC^&$54p{ztU1Qc-WUK{RhwP+O*%_okET<|k!luvw9s|+ z17tB@zKE=785n+h0V&L{;%cQ68rQwvMfHx=+yv2+S6@?2t1{!U{OJwmUVn;g-FS%n z^&Pg{-p^KKjKbNpp6PiFB-bQ?T?maJ|AC>xk-1i6W-*eT&Yj4meF?21Sa-bn%oT{XLK+ z>=@Mrb(b$`5?-pWPt5sT)8C_CF!Bs|VPP*c6bgvxDJ_T`>DkAnUnk zK#flx@VB&QlJW2r^`C5{@V}46Fk)RC=c#_UVW4^^x9PSX3ZIp*bao-QHmrrFGc9<{ zFO>#*93t;bk+4oZ*!KymP=DZ$z)dJ{fQ~5 z+w(?NHjUc+iFZ+xp@}pICmgBa9z?BXbElT_FJosz-@w5v$6`KnlY#s8SN8VV_*nODvl=|5BcP4_F%@OuN^$#s`ZWNkId2Y-T zfT@lNU|}!{CmWA~KTl_&waf)}@n0ZP3hIVbrm>+|lje%TAeIi%Wmd)F3H~?zLlUc=7fifN@Gp&=oY>);= z&25S3KiL_U7mj2{rGj|(rw9s07a)E1Lug*QfoTOdiaE5!0$VjfDDOY;thw4W49Pvi z57pYi4O?^zHl$r=@~4$h`PU$(YAG%D5~s5BDkh*R&c#i*7j(t{EJMw)Y^FF<8u+$9 z_75*)gQm*T@bb&9g+Cv0*Sgokip%3sN->s|cg(fa<&#TEwR*Ra*AnxtYxd7H?Up7hHPq189Qpx2W9*9h++1Y)>|*C0k0m>1hkthjkzw|6lKJ!xmNhDqLH15g^VeU<6a51}wDsxI zt52L|mNmU=3*kKH7qb%q=lMDH6^I`epog>}t5(pY;z*IPF|UHoUCy9A$49YKjjP!k zrHR<66AhXqz=ka(P z#V5~Wz1wfIS0`j)vxK#4@qbEi*R&SK?5M+iCV|`qGg%6@P{E_0J4jP8Mc7-t4u=^e z&=QFjI_9b)T%Ny7n0l=rdVU`PJ+|>Q!Qm;Tz6i#giQ2+Mmp6jx>s(eB_K}tPnFA?(u#A#ivXWoGH1_vIFryflauo{qtFA11*Z!G?Lr zZ9(UI)O@<7)r7i$hnU>JDt_SI&Nm z9ip1W!(m~s9PR0n!qI)C;TcXt{%(|8Xp&#$Ao<6hDHj@uOX z=o|P(R&gWL<6)lMV`gY6X1-&BQPw*e|1F4vs;t8_`jez^WQhtszs@m<76nY&riAY{ z#F2(kH-*Pp(4=-pAga!Ttrq*u|82d<>eQ}*{Ep@LP4^}k7W8q^W8Z<}DMy-} z8wfUC+aR}u=jzspXMV>_3jc2ml$YkZ=8Jhd8wdH^I1SlHj{LgY|R1NRwvt6Zmiq;gV$m3|ufrxGUre&bzOW{j_InL!K9( zqad)2eQV)H+IN1lts6{se8jC^p^7Vmo7j?FzXVmyGAeK{g%M43q6mf0?>5kE4 zSJ_7R?IG@I#7yD}>sd6)yqT50RYEzf4Dt>-Mgij7dftD<%AK?6NIAdQRw9<>t-gDkY?op_{7X`a_&2R#EK$_Q}n^3eY3FN+^f_& zpaj@DQ@ZJ->mn`Xiq4iIpQ<-akct>W_GaVg&A0y=_6~T@#{RHC=M9h9*n(@IzpRgK zGSS3WIUNuf6#&Jb^VwZL3)G%?A4;CZv$m!fNUB&*4@KUVd9ui_OHYU3+IWcnph=C- zGbt^7JiWC)&4rvCMd9nj4y(>=9Hd?#x>(v>JD-GOS$#Iw9-IM%HveRweZ2y8`rpRgthknz%u9# zD3ojS_7@DF`foP<9WL$~&22Enz>lnyf3tG212e9cRre+qsw@fXe(wpE~;SU3@=K1Z9|7&s^E&wqO;ORQ~dN860Db+)&k=GRVOu7?{=z=+wyx%&H|l$7*^-1YOox6Bhvyqtv3r#5o)m@kd`FjdS|=W`9ioI$qt zEn9iElr7%S3$-FwW1ONUUB7>b3VQNrlcShdt`R#$8@=dGn-OV`2;)BGXp-7?j;-mE zz+o~=(P45ZB-C_(-U@s2v2lj$s7^!hB?d_zqUgjnre|7$S+`bW;siyMO+F7>TVmmN zP#3?c?>BslG9xof1H`^FH2)+=>qkzZZ%N7|yZ9C5skF0jwf4rB1D9fr+)9jIq=2>a zBye95jkg#R3N=i5!v+INYWCv>y$_8ZLkR=evQ$9oL} z20P+m=|~p+VmOq(n?@gJPR1^$B5vLEAKaw=y0p#Ml4ft72OTd}@#Cu7sON6uI?O}# zj4zeOci%)`Qe`q7yedaSXLt!Grg`y>iz?ZTptrQUVMTp+v1>S({;)>K1IIL~c~bcJi9V@arzoE!yijF}#f7m@cKZeS%&ul$ivrlpSXp6gm84*4Y$#~W z8Hb(O96q^fhjsVzA+=WzdFNotcg>}686SR6M<=~_6i0o-N4lNV`3JiegPrAF5^a^SJwY^)xW zLut|a0-t!7&3==JueSQJW!CvD=8g& zc&p}2bDEA)oaQL1?9wCsaeinyL(D!M$l{(oh+~}|?s#X}6A18&X-r1Vb!*$KPg=GEKqS3@56XvVX=4GpA_V1u^5zko~49iW$6n5~ow%U+wt_M{!e%U_mb zNBN`1jysN&(QbnN{wt|2=sh?2lLZ;%RrA5M?W8L8hN51yv-ey%=nlwc3rqV6my}*m zuH<*fxPJ$y`q;pA(;`lCdNdqeGMZGjiu?bI!?}nv_bJS0ieMLzN7ipepGiRi)u(C* zS60jbZ8pP&eZKB}a&Xg2<+-{rb+cQY%wU<~7Rs8 zuIeT8-y0s_6%`I*tBPomv7TU3I9hlVXDMJrAzbvGDe^qzu-(RoR60Je0|rw``D}2%Gj!H>1npI9Y&_g6=cam6?0QZa2%G$V;*8&q_|ppeDaCI9DKuT+ zf>eEk7(FvsZQBVgsmdf~>A6{3c2HqQ6#JMG%aTuxW#b2|#-5g=)VF?-FnIWOVNHG> zD!1&zKBrzRHtC^$VIntX!C-1nl;q9^tRYUt0Y9y5r4Pra)23~{RC}zNnfl8Mp)+jp zxArZ(*H z{umnCu>{+%NaB#Gj;wk19Ashp;co72@G4-S79`&FY)C+f1D4`_$xrBiTj2j2GDF-k zOrkBS$KmRAfa85-?6CY=nzXi#IefV1dZm2>RgcS}(F?oz!xN^_P3kXbdi)VP#?7P= zkjL+IKh35L?ts!gqTlJM*po09uOn4s(0Oz+{t|fA`-4<2Hd3?|Tr934I-RT4|T?Piao?^1HyR+@DB zCikmzJRX_lMCvq5NXqyHFN$V>di-@3U~~k+H0IE#&+Yj9$YY2&Jw)_)oh4<}VX)D} zk1NcIM!V&WXj6~)FUbZC_)JgRb~+m(~~Y?n}3^&)e!?vS&J(_b3SYBMihD%Tl=dFA{f-_)J~y-)Y8; zOPJzlLE%}K==tM#*lpae(QR-b4wlTMSNcA9s>}tAnzrG+C@JBS$~{tv6`h8cFY($g z#jyCJ$nvWzrLlGWsCx2Xp+ybR_U~mPQ@nTwl2Z${SDlkc1KRha2fO~ zkz>b%5dNEKHg0~@%WA*1v+5!pa(=r26V>d|Jh%$I|FcF%u%<%KAugdMw_zBJVM8TS zS#Z7&-tH-3N;*ZX^zv+Se!79)n1^!Oi_Vkf*8{Y--%8q&ktO;#_LHaDGJ1QpjU8FF z2Tt!RVRx&X$Y_E&t=R62o_+o4bnz(A{xuaREq52L>q4)AWe7I&DE^)ixsK|2RXZ1#sHx${F*82cv$Af8wb>s9zGjg*q zB$vV$l+v~f1LwEFTe;i#Ic5YqcHV`K+*}5`t}PQfJ08;x@3~ak_Yf|5jY4^!Hk_bT z$Hml#k%Ve2hR2)JI*EPEwn0Zwx6j1xM=7{);S79Te4K{9i>H#2UwGSZderNj&!#Qv zW)Z`rap9dXRJdS|jPGWivkr+4S3f-NZ;ewo7@+hmb@Jcijz!aIXi>p1q3ydo>=e(} ziCV=jja0zmvW=P2$w%;ISt?C+F`;S3O2VZOHNjl;xA{#m5l(4H3A46MCFkx=aC4Q$ znA8)jsizrFo`}FPe}00?nSVr!Z*qPvBgA`CZz%3QO)Jh3!KNgd|9mC(`nR(JJ{jK+ zNZ?y@pK+FzXVJ3LRG6$1&nErm;JU>|Zn?x!&QLiTC2H%@zQ-Hoza7EKi6t;Iq=89V zn9-=6hj6=;ANBj3!SZEa@k{0?qs@r7{LFi~%w98t$$t8XtL7Zw2ltO6-T8f_FJVIc z4_VOuBf%_M(i;=R&zA8CF$nx%3n|EGkp1BEw64aC_1roFr&312#gAjKOK&n;rke*> zeLCn*YdQHfFg8NQ8mb;&Zqy0Rh18~OHe~uJYJ2w#j>&6atY0zHh*v|iyeF*FrIW?~ z&VYe(d1(0D1%J!k1XENJUKbyrN{{O_rFlYQuk3FY(td0K1bNd6~kUsMv0gUI77Q)2~0db^T~BKh zv?#>lAWBUv0uf!!=GDZ~yyM5g$sADN+a>+J;=FucyyO6vgX7EHza=1VcR z0l8HzdU$GMA^R6Ck6|`F_^Z%N2!3@4bQ`9Dab^m?JkZ?ruZ^@2B5r$jKD`PCFJ|De zExR!1-~iaaFq(G1egp5XcC*CCF}PvjFp&MYvM!>nUi>@0Bi}3aw5G|9jK=ev%EO)f zznNaRdiosLb9F4+@iZ8G{XBTZ_s1HKCQISYNf&6@sWSwb!%^*_k?{D#Fs?DIhqEfO zrl_7Pv@ETgPg?T zk1x6(FzagZjAAQKzbY(gN2)#4#vDM8GXmeB=7Q;4C$KHrB(e^Hy|kV~nc+{^l()~= znGMG%LDNWh|NK3zQM&0mZr>)_?OsbxsR6E9(@a54(gq66IU)1hSR63CnRDLm%DOhV z;4{6Syt-Z?zxbY-F#Dnj-M%47Cmi~d$u7pL*{fq{b}1D}%!J@mmK0d;22VOB(%>m$ z(4l(|$r|0~SEXJ7hs#@_dKqIM2ZRaN9AcqD#fVL~yhZdc)bYiu_QAA>j2>1-cXhN;WNj(*O?X57^Pf)2!*8JndL{2W<*T4YP7=B5=KTQA1 zL3lmSj215Pp*K~rG+*K)z@nK5>p!y_Mw`b5%J{c@7f2F6Ur)Z6I2HKCg1-oC# zVl?wdo;yhn5nAy3wmHozujJ=mGp9!f!bs4nA=9c}+PGMY)(m{fW>oK`t|5JV(5autD&y^6L<9TH!5AX7aN|;qRUH;F%`pd+W$ToC+SGKDW=u%$F(Jd zQkQpdX3154YRYNKcxwD+{cJMJaxnnP@yZ@0{Nf1-_Q55;(;{**)V;r6w&vgtL zDee~!vC6ueuMtvU7U}< z_&r?RA79K&b!m4{qzz9;v*TH_;MRg5RA2Im$=&c^?bmj|7Rk@-;gPxg`NA0Nys0UC ze72ayN!_OOi6ImeGLns`>wuJeZCg1_LJn-THV~A>?Dv{!eKfc_m`>%LCa&}*rFPe|EAyD^ygoS=R`ZOO-^^yR*H82B z-;NdEhd1-xlfqdi?nIrtfdB5gFv%ynFqdNa!|`Ib?eI{yZSg8NyGR-(O(jt#R2Rz1 zhv4a<>q+lt3Li1-B;C+R0soi;nEej$zV%Ud|LstGJS~aFN4LR{zH@Y_wS=CQjK?<# zXTY`Q38*z2P(Q;CHdztbio`)ePEr}!=8CL-aVE5`LyJX-`~>%R#bj}z0it~+>0Ztw zI?yO`vN}W_`E7N!-|#ie{xeW${;eTse>V`!w3Z9pssETk+5Fsp_~*W(f&_slr9uLokjPS4Cae3GN7K#@WN)_(WIMck@I|r$jXO~Q zMd!}IcG*j~X+{k?4bDKL*M{Uj&5=}p$8)lo`lKg2i;~~1rib6W1gN`KiL$ zkVn`Vr!Sn^S3}y3ZDQ{I4bGZxPjMzY@GR4VytR>ZxcvZFuUd~AWe&5bUs@p6$(gQk zD6a0j4ljagxF4q@*thTpXz3w|m1BRfhZsw88dJpcRTgI-Y%N?=7W<*mvl>55dCSaF z1zO^Dp>fJR9qK~y|H>_#KD6{eq-7LWWh3B+a%rI=Zw@^_k^@7GH$q4rY=f_uuQ2V}JG{iy_ zyQ&SQLFH}uW^O6Wy}Op=ZI|NXm|7TlGyt~qD^TrJ9?WnFgZqou(na+)=v8f_KkWda zH@!j1yNq58GQ!nb$gP=?gpt?!===JCLS9@Lb?hHbTh5NcO=AA{yIdNYiaBBhcSMO( z`|#1|Rd`QY^c{XnWkO{UWi7c)-;}mO%^P28{#s9Y^4Dq9o88nF6-&KQ$Kkqq?Eg`8 z-ho)XZx}~Pl1)OPveGb;d7c}QhJ=PRX=lbvv$`ydiZOB99nR?_@Z ze&_e^``>%s_dMgiuIqEv#pxvAAIEgGcWS3+JD!q-F%H0kEy(q;d^VcbO`!BGJ5#R) z-@o`p3LdEQ-Kl5r^tkGapPmKGF*#&112c)Du-yv`O+m)3RpIL5TqymM5SNhxTJL!Cl?=i3t!Sj~uV+xnEn)2Mo+*U0_6AeQ2%_ekM~{X$aUo{= zxQpJ>BvwNdru~^t=ikf4{?AtA_j?bvoIej3$1jHWJ2K7lO%8KsXNBQ{#h%D|orPPS zB{;P~hnw~BA~DjbVg6IR!Wyi1pdD7*`A)ADIL$1IPCQ~)XEsFqBd+Dc+E5DXW)yoGw|9BWw7gx0Hr`*7=EV#`>GIkCI`^A$D_1n zRwVwceTd52)9IfS}`_*aXnN`ew&HI++}BBr~E?B(RYO1tSV|T z^>?OO7ynKKJk^9g@p>wEekr`=CL z@y|Mo#pkf8a2>Ar8ILBr7lN#6E7eVtLp!^-v|lX+GxMGlL0u8SN&Q zjYiD%HwChizOB^K{wxmcG{M~?k6Gq;6WNhlO!gHXpc{O|p=MyZ#gVU1aLey;oV)jC zZ0Zul;~RG(rx4CIG-`suK@Zf3>t+5Wet};Hgy3ZVKAg)l-Cw246L-b$%wGoy_!+Pi zo$ozodbLzhFGP`>`n?dPG?UpK$_t=gw}uA3bYVClRYtcWjbzKF(`%+uI6CMHU9Ufq zeIFIfV^vv_;wy}oH>Z+D=>%eXY7S8@4rBz%1L-omII6FhOqIvyK&xjt2*z#0?F$xR z{+K%4`E!H5HowWzi3%umco>`RBoNCBb0I$RB>0pk!L^Cu=yQtK*MHd3*x?}jnm2+s zYuaf^xh$083gRYaLbc*Es1mO&+*K5ZNbzU%w!>s1C!LkDB$&k{fc^e*aE&+tA_IcxnswHth*(#>{}j&)+iI+KwQ(?-ujZOrN}QTtx1BC*b%avM9e0>EDg! z7_vf|y@7n!_Sb!QO<9<}oqPceY#i}X0-sgQ(80_tdzeS^_9T3)jy~JDlD_!j1Y);l z5lh=N>bEU~XBA68zP}h}<`ID!vo?a0-37ANBpsCb-dE-Rv#6lBkPDACCh8x{iS5#Q zw#?%UD6d;X%q}}XFOL#<5T`;6KZy{U*$F*sC7|onB#RX}N#whdDWf^BlfE+H{latC z;PI{c?CIkMs8uyWGB4g>b*AQnozn&kUhjchBF^9^<4_)8bpc<0PXfsumted2Q*x$n zE4F`c(fTqp!LWAH1{4DGq>0cT{{YBPMDTmEW$~QUY{%02oLYuj9FAkDfX{unHRX`N@ z{844WH=@4Uj9Z>{6Qh?*MnBy1B&Vv=nEkc#o>? z{D6bDhbgyD(EQW!B5>vzBwO@X!8h|v=yUr^c8FBgr4GAtBU(%F)<__g8L;NUe>&ly zsQ{?7c9VJahG@>8jpvG~L4Y04hFX&gcZEebyNAkLo3$L5do>GY^L{UDc{z-9IYTSH zRpXU&&+vv-D9qJwWg3*v!|A~jc-{3NUTz06caAPsCn^bR|X&zpRXUJsb zcC5A8jsK-AWR~juM)M=HFkCeWVs;~#{TUA{Crjd1>jN0L${6Z5e5SS)KkMAj4pO~g zzF(HR3q~KW&<15kRC+m$J3nepH+>kP$Hu%rMkVcoxxY43@Ztvr zv&pd+aQiuH{27~snxRcpOss^h-;z$uBM)=i@{YspCkC)w=pw%1eM}#uPeJ5d6a2K{ zJk0h;MXf{fz$UMy?mDl`@2oq{iPXKuRnuh1$`=`QQ{hjvKXwtVD~mD3DVug~NrpGi zHUrN~$K4(^sNZ-I&Kmz9cV&!Vl6nQxx1C_E`7AuKIR}gug~7WLH@a2BkQ%!AlkbId z;hU5cnjgCew)^vG{m~Fk**O!f3hzUEHLn4`?x&yl%+M!ICp2#zfQFkt%)~Y~FxPhE zqmY3T9{bARz9!F+EL z#lF9#boCzxw#s$^T2DW3>L!fTt#%PBeC0L5j%_gg;$+xqI~xzLOrq0WV#wh;|463a zc+9$PgEQOZxu%bDAi}SQj~^ru$=rTsfqEoJrgCRz?!=#| z!nl9M9{TX54!(*26mys5`ZN_t<>nz`eZrUY>pvl4B@`2^tDxetA$!dAKf0x00-Uj( z&80T~f?PX8cs%n6jb0LjF_O#bDmegkJ8ir-l7QR>T>xV3wU@HOp_exrrRi(p|ODb@0;3b}ZE=X4I*pHiZ zkCNCdK`hoWLGvpYsN?nV@XBNzt#{=)(vQUO+uL&L3Bu6e^NlKLzM!{#eCVr!C_1QO ziu+cb!wN>07MJ8;x4AlZcYyyqND*_T3rOl&VcfH<98Ij2;krc!I4)~8m!CTmCHM?b zV?Ll}_&Ba*i69p>w2=G9iSXG}1K8C(5kI*k(DI{V5Wep}W><|o*CT3+weOq2JMS9I z{L@C`<3+eOD_Kr1Ym`>_mXLVSE_2a0Ep_+y2EpL+H1>IvFYy!2<9oCR=p?C6)K6L+ z0@S{vXlOdR@tVzkYa#S3`%VXkqd=(UBs08D7o&a^n``wv!md>>Ksosr2)Z=jbr)n8 z1@)4r4g6Vm;%i!;xr}D<>|I&!$&9-BSy(*CaIJs7;K{VR=n=Gm)*i0L#w%{TcR-h$ z;d7A6ZwP_MdXl(rK?!;DCkJhs@1U5;VrsVC5`>sox~))_CfVGe0+|b$;_H*SQNP95 zu+ae1Pku(Pcg1kPY%^DyMQ{!g;O45hV|39VP3)e4k3}MA{qsY(WEJoG{3*-*SyqUH zqwUz*ACIe(Zj-H{G9)^SfVaJ-!}@MPi;=HcVDXC29LHy2l}!Ux`PslMng0}{ zrzO)!wG{f+iRZD`l$bZQ*JG)nIZB#uH-B~D0eRvZOUCO=#UT$uE*-1@%e_l+n4g)? zOh_Z2|BYa3s|NRV=ooE`}jm_sf`2L^;I?AS#yjWiMyk&yUWp|)tbS7K; z)(GI>J1lNA!EC;ILtoMZ1AonhU3;V1PQxv*FwTj=Q3pMcd^U&LN%0JygY9^*N*_}sUSYur1-Ljhif7F7ewqGB z+`Gksn5>`#vaSAb-Bz4Lix*%{*kXBe#%bG-QDJ($cp!0cWr38_oB zVWyrd?fJZ%Mm?0o&A*hvCQS+7cC01L{6aL1a^h&+{JNEtpScUB;QfDvAR4n8q@Q!F z;kp#?44Z>;1*X{jppV)abG(3Mo(p73GI_J`$ z8)YaPzLQS>_JS(Jor5dtjnrq-BYx%*1)X_noLX`nPS~7>@i~iWLBxJW2=Z{1Y%s1I zZ%F&JX;=nd^d?$-lJBKG+K0)gR|rTv zzYbf*M=)j0(IjXIp?){#bGz*u@a_hlZDlJC@sELA^Xi4l-$H!uxPe_L%KvWZrF5bB z5YKcJ=A?xmkj9Z_CXCPP2X2UBH)>U(TAm8t^7}%*{4ArquUk z2)89DgLr;4#dD%}NOy1)`7M*eh-Af)o3`g+wfsEbOC3nBp|3fKw*nJ2HQ49-%uKy= z3QhI0v8M9q(Tvy1BbyT=ANn(`vZc@bl%J-PIUJCurYT@7L%E9 zj~Zu35VMv{T0GE!84|qiF?|Qs*%?BFn_VD#UMFPV)d7)9b~H-yA5E-V!N0#pz`5ca z`RJ01hCSP1pSK|OXfC0?hpWhFku!MMNYJ!+e|9iS79Gkiz}3Y^`Cd{bSp9k&%%3A+ zpdwqsCa7;7jDX_A;VY<1IE>93orJxYCiqujH}lMUdW9 z4K|XO=-XGe=o1@^YnMr5;4=!jbIqaO>>}9DnTj&gbMf^^2AOhl5Ph^akQL>U@YJFR z9e5AXONm*WXp$suFVY83))R0>5@Ce?F(MrnX!W`PN0*(!4T3?Wdd4+Oh%^ntNqrRW860mT}% zFmPuFjx%^ftA7vCyC2?=Ss!-8-HaR5yjPC%S@xLRQm7|GUS`CzGlE5e1Fu8bv^Wj-Qo6?wA@)o)^CaFFuvrxu>n?m$HjyfX zoW_j@XLGP|6Q|Uo$9=hG&1^H5z=6C5(z=YnW?i1;#_vP;G@Gs{8zARSD8inL53wac zhJ>8zg`j2*4jPn#u96mv%@Tkx{bV}jR0yi{&caz=j&OIjmE-db4m6YU_7@Bvixq7rR+m?%h)|qaW|c$7A}W7uKjg~4_+eXQ%^7}F6WS( z{{*1q!Z$3?Zl?N9(?EKiJw~6BqibxXv4zh+#2z_BU!S?fNRBAr?XECdac3L|YX*^1 zdzVwA8FHXey8+bTKPFrADIqU^pr6-i`n-39u%e0ZJb4vSQhvnq_M55L$3f^hk_8VZ z^LtitJ=pd{LaC@Fv59elR6gG#y7m{T4d`Yyo0H)*pWC=2rp`4^<9#eD0a(oEB>U!G zW3BsaF}aYXzZ-7j25W{a*mQ;5yQT?-cRGlr83V4htD)3Ol`i+Qz)QuFT-@8W=;ZyL zjhcU#v=7ba&PB)4@S#B56W@z6PNv+n>OxxLRE<(KGki;7-h zvUo6>e-(lm7n~ugqYK=VF5n_2kCr-4!9}7LQ1mN?eI0(Q?#1vn{N_`S;!E?W;2A5h zj=cg3(vVTH7DYjacDM#X$JcAj61PHFyqW(`ZoZC%ra^cwKpRiJe@XBE zn2xIwhDmeW40`257frcQPwh{;fc@J_d~oe4(%O0Csj*(^)Pc+!(0A2E9=Fhg|5M!His0v>~Yc;Qsrv`zX zp05V!v6r`)>Y|4-{;zOc{4_Y$WXvhl$)U1rIKJ1&#BT9;?i`cZ5zzbI5^-w_irTej0*xkB(a zj%VGS@1V=oj9|_p-hcYol5{%eleVz}H~Zg%rpv>&qK+6KZoo%)H~-Gk$~*A=(O`dcGp zzf>CYYPSYSS)h$i3tCC<^2K16X+X!-^F0rNm+|Blrf%kxFmSK$fdwn>lGV=VKqqw@ zY9+0qC9E!%O1WcCZW!5PpF!oL+sU{1gP<5IPG)@+$HT@ZWW2g5Bt$x5b@ThW_*H~d z7*!IVjBU)Lq&~WAFpQk=y90*Pb&*-Q3KPtyTAY3QxVpetk#xv(!263INa)q2ta^M7 zt&p09!qc?bl``jXisNdwqep~snlwNnb#>uWsuoJMwh_-Cg*?OE6Lkz`a(}rce2#1_ zyv{VFdJ8_%bAPVWw=3kyl8-as^cp?-&ZU}}tR9L3mxAfhfNqE^9>L`9eY{kq3~`Ak zxNrUL)TS{27CbP4^_#-UuU=b}j6F<~`SX^H-v@HW?LEE5^i%KQG*Xsa&n|WRKpuFF z@^zNF7|d&Y3#H{@ecWM_T^-2vKAb@g3rB11TdM%)-HLl%w810vpm zUqxRbb7nP~9hSAoQTE1$+orf8w*Y=+>2Z}+ugRRzjdV|a2aVXP!41ULnAd3SMW@~8 z&@mcG!S_gz6Lcg-x;pHQNC}+x>LO-p=Rr}N0}Q>TSRTmxL?2d@S019s*)-tyxr~Bs;&e}y{-1dUBqtO!9!=x^DaYftms?QDr;m(P+rsvr`Z!}pF8)w- zAl}?Zl)oO#-f8Cj2d0kPQUNdQb1%k0{S3VIZ67#U#-oeOJG!wm6#FK>r(TOXh%u7{ z$}(GE`;KWa<3R@1ED^$-fP*OJ-$kdE_JF%j9?@_%0g2E)VjVaHc5_Oo;8Sm^aX|}q zFVtgII}Wl%X~m$qVg^jye-JHg=g|A@i#WNHE4ZXtGSpRDn`9i?&Dxy~pzkzd>56Sp zOn$JAF6iA0H`MkYa7}k2UQsKyzaQ%EN`kayD z#y^e5X0vv3L~9BaJ93^bwzcEpl^-)T7Bgr+pZQp-v4q;uFl5@oaqz4<+Am#!DPwk= zY41IpA=!_6{PW;@nI7yt>IK09|6%Lw0IFhR2|cRX#K*M^=j86jd6#q`T5LKoJC*|r zj$DTgL-#>j{2n|Qk%FXE3H0QD86+yn3AIKKp_0L5JmqT(txxZg_bthA)u9o+zMUhR zV{XD6DMhpsyGHfyonbdmnGQ$%6v)O0Pw4}PX;{^97#=;nP5;)efW)nrNkgPB_z5?f z8)#M0Ba>^%Q2`dMAe9R6H$HyJfO1IYAwaxgbu9X>w!O~3Bb zYNktMG5Qz6Ob zk?e(OzMz{+2)-DE5uN>ts;!Z4^c}x;?*VE3 z@64oMV>GMu8GTwl2V3}=O7{0IdaB?NUC4L9cD~jH^RPG6f-K;4>&qzKpMX(i6Y<4L z1N_t>f(i=axLjD4`+isvl6q}nk>Ut-US^AV`+=(vYlly(QmL8TRJ@fDf);$nL4?z$ zuMauk_2-|ktKkwhob$%tZw+wSH$mtuaAg-tbBy%0eQ@1C9qNW2k`s=F5HIElH|~V8 zsv9zhckgamYi5petG1y`n>WoZ-Uj=wq(igJEb?%^HS9Qj3k6)x%)PTPitxjD+4$X_H`|J0Nh@O(<`(`v5 zLoS3_URF&81Y%&PkT+H@yoL9V_+wQ3eP(~yN3v&uEH^dUj(%@m26wg{h4l)D;l}JJ zcsx)JyME`Bt&0ZfvblT6j`-t@iGvC(UA&2IP2xM4=6|F2DrF(-Ml)#fx#1aaBvGj1 z11jf=;^^!UF0h|)A6^A>FI3XuVpk>PS?1xA=}kDM_?vOCna{~+xo}TKRk-LEIoxVK zANup_nL4+>5~xuaLVNC%GW(VXlb$7mb#Lk?!s~D~zF)V2^<2`AGkeoe(DEb=IBAJi zvA4+~?>jVgJI}lOuZ`_=8gEf(>kNNR>eor`;)sjZN%S3%;=D_qqBvvDO}^^E6;*0- zpN`(b>xEOe!n38gBk3zsw)qLwn(7U0uMBG4>L1W8C6C~2xjoFe-wZ>CvOuHGlbO=> z8)VhCWA$pD@$jw{KW-A_)D%+i$(0*aTTGg%&np3LLpLM3triZONy$+>)AWp!7EwY!X`G%Z+;YbJY;0ywE^{0t58ff0Zem5JS?w&qb%AcJ_?oJoA}Z zxuiep1>=0y2J3Bf(Eefscn_q(@zD$DbaE2Pdy0d4MmJqM^CoHgssks+@%`-YD&UXH z1hkL2$7^3pX-;=H8_@ES=u}=nmGwS!>kBJ<=@JHha}&`*lEdRUOJJ#%Fvy8rzzT_` zTE~p1wYs80cw}oc**#l@{jupcwd#q(%|$mzN7Y~YQB4*Cf|uj5BqMy?v=FL)NOLLQ z=F!XJ8o@ej4Vk4BO1}=Zpu|F799dU}ZsL5O_}DBs-Z+=5vHgbEPxe#O=-tfaUyc}L zJQ>yP@1dt)8Cg9$&`f3CRL=8|0$y)egb&{BCT(6T`LphLbEm8(7(VJv6#P}Ffs+UN z9?8P3=2>{1&BV6*S7_u5Ka^NgiR<@N;uDTgW}PI*i@5pN5X`bNNbDY{VjZdWtYeCLAd(6h}SJv(#fxX(QSXL&?umVT%K`_+G>PD*)C0x;WJEUB#TgK zp*B64bp)hm#gXX|$8fGpA6W^LxVG8$oT-op7-1zgh38|N=~r~+a|PD~PvTF}e^k`@ z3Q_)gjUAQ|rH5;+fH@-p4r5}h;gL)lXtRrmJRfi2oA?+4d2hzl(%E<-CxaG7=E2{u zVUT%|zmJ+s2NR_ew0*BIH+Ml2kx!owMtVEQU2Yr7J{9GtJp(@`Ilz>$H$r zXAVZ=px{epIbTEPEOP)tQGQl6(Zlh4AptRdruzl@7+`ZOBCqrr*(Co=)8;m zbl$}6T%Gt-bX}@MKHP}mIVMT4UiJp_SyzSLzxkT3<9U;-9+c4S6W7p%>l86_P}Sn- zsi}DQa54U!aR^T?jzO1cX?zb6?@wD0NJDCWQaA4(xP2gv#%{^rb3L!{(O3qWNzb+z z$MJlY}gZ)Wt<|fe9XTr&b2|qCHOEy~G z{8js@wwsEJ`QZ%58ovK-1~=`V3^#n|7^FrZF<5(?B!0VxEpgg(T1hDGSt4K&MpG#* z4W>fL@6gy}EAA^&hlM(k@YHQSRF`eWa6Af;11S)gvlI*-me7S(N@z0EhJ0#&PTuDD zae<=?FeFqEO}=_yUqU>*`W?+&yPSs4Cr;vi)lK4fJ_$xW4#n2DH|UU4GFhc$ZDE>p znY}V@Cf!<-Pd{v*Nc-Ys!HxGkoL0X~oSf&vhlqUe`#v8g9qFcHp|7B5yc%X|T%)Vb zDdD=clOaRXmN}lIM&7a0Aa!yV`(X2Ln&VT-8pJ;$7gnz&|6;>1c6cp$q2xy0V*kPg z#l@g5Vg+5dhZ%S6)6lUyl5&F9u>P|nj0$Db2JLa&TC+X){?l}%p09Do=Wad^KSr-F z{XxW4Ly#Tkhr%EI;dg#GUDo%HxgNHPw0887cVrX%Sn!L;%6G)(C^610 zH`u~vJ+%0F40BbT;iy*vIz)cN?!WC=H2XcRRtch`0!}dhT?cs>FoE7)wI8c~pJl;N z5*Adfr~dXYh|bv@)-kV&zHrRq^q=wUJ^5qsJ9`Coj!}$>SON97yXf4-O>9T`Bsh2L zAbd}a!vpJ7InE~y5>oA8STGzLleO4}<{q+hLOSG(@R8Uvs$BIERqjTrGGwM06M^j! zL<3AX_3k~ycJc~RO zHSJ=dsy~a$m))b|Gs4lcG98O1Y{y;gg?Q;kKhb&_0;iQPz_d;SI@5OopO;Vq+gGvB z`FcO7t`XqAMsA=BA2!szmJPz*M>=p+<{w&zui{*dy>LZ`KaS7Qguf~u$lf>0@p!ue zXQMeAwl0o@Ct|1B5?)JCQg0=O*1q_9+c@wSjbwie=D;2mFY^28T{>NH9-NfF4na3d z_#Uuym@iXI74~N`;aw~7z*`xNt6M|w@2x?HJf4a0Up3tssfUO4@~K?z84~kzGpI8c z8U3?;G}}LhW;saIS(RL6cfLrc6Iyj?Cp}1J+YXq2EDnSQo1@fX@nmu~(+O7p(}h*a zdGO;PzpfgF1L1ulhy86(VX+L{-k(dFGw1N$m2XVXNgeuj%5nDoeiycJU?L}fOB~h= z@5Pfki^+G_KspweU%PF~Mwqer8Fn7K!<5eoh35U`=+yRv6zKhe`A1FR?r9So>@UNN zVo})snD1NZ;oW7T@%W?ZIQ!{EBoof}U6v$ibE9ik;ry*uAn+gs9Nxzxr0l^fT0&gf z(?H_dahQhc{-NItpWxboQ0zI748nR6_~K~=2=7|KwV0Q}tZ)Bt+{)>w+GPcj4=SO@ z`!-U8<+N$ecbcKBOFtiIfg1l}?ijwIuRZ7DznC9nf_fTCwEU%w6Mqt;8V}I?Dg#p6 zS5S$nZzM%00N-b5!%^pp^o_7A!d`iN>%)*HW#E*;d8936B`td81XT37Q`98r*XrbSS0cpV0a>!%(o5VwHQY@**ie9pZsLj1z)8F=Z{0*zFKVa zd56_6Rp9n1N6^$6AS%PkkQiMApZ`-r<4vucx6U(|sd9}hWR`Ot^Jmk-qpOI_p5M%g zIY-Hc&%^k;q>1h-xlANmo0$rmgCIH73(J1+eRoCnY`M<_i`#c|!Nx8QXU)uEmGh)A zIwu(V?bc!RmN1-i;u*ge*T5|?ak{cZlq!u~G~=yMG~_PYMcvOaFbM`{EcXI(#8d3yx!V)dzQpHeWmM; ztc6S2yJ53c40IP?!DT$pa_UP-?(V1ztDrHRT$0eFq4uft{Lyf{4hmdQaxj)(_8_+& zzJ*bBG3e=@$z^-mqwb{<(ziQ`tvq~>M*B#^sa!8~ER2WEymzB^u_E2@U6p&c?<|#A z&`Z|~-yvhauhMb9r(l6CN0u#nPA~Gg6tm8C(3Bg&?5eno;qwZ~$?ac=fr36gkQ2pb zZKxn8UO}C^u@`l(+>BEL3yHD77)|nK!M2Ql_Sz+2@;o0l)r}{2ES2$jYY7wKZVCmT z#N88$aIQ5$q&Xc}|UJqjM^R!Q>VxNn1=SgyEN{>P~VebB#E5+E?^B z8Hz?eugNpN!=zN}G+Yv}Wvj!Jaape_cec?BR#p#@=faoiI%}Xazt5(Fj{3N5$_8%7 z4rBbU{RKG`Vg=V;ZxdJkYZ=Oo$a4pETkxBzC%4VmpF6&OBVOwk|L`Q((Bj5l5=67l>p77Ydl!zTK}zztrC&pHHZfz>SM(ZIdNip4nz5x|1X0=jY=zRdKxRHXRQv)kXm$ zbJCb`9&DVIFx>!|>0J}Kf~_6Qv7C2w*{?VBOU*Z0|FfS+hQ@(Y#Th);)q{I{N9nV@ zGr8@t2cY0TAP`eFdwJuNliTlKKvgnEiBpLQWvFSrP`|2TZ_`v&wSa&DnB5Wv?HWLwzfZE1KM8W<8HB5OxmR~$ey!OAut#km3WPhWX z+gDhzP?j`XO2gHm2k3av65j_d#E6|A@v5#C9&L=E^NOlayR8bJNX+J@9R5puu3Vv~ z0|H5??5*0>r*zmkksRbl4Ur;G6XJEZ7TozR!T&Z|;C_A`E58|o^~wzGOc2Jqt!^;y z$T~0*@Wv|=?R*d072cC?fD(PV{cHbdPPYg#f1wwDS|07$Tw881ht0e2= zF{rv@%H<2i;@b%iFv9mga?;rcHL_bVxjYaT4zA$#R88QfUbLoZ1v@2l;{;=FYQwpWQRbLXPED^%hVkM-F*X<1ygV^8{ zPyF!ZCB4|_M_%f^fU}n8=+yIzuy3y|uF|+ke}xAj3dvi1`??HNbV6_>>w?Rl ze!9^pnvNY;Bdhq1m{)5FxqhRLiS==1UoDz|l2M<~^qn2?xFJk5st52)elfgVcL;Ps zz2TLSJ&ZlMK#iY^(Z)aLu+!uR^|w6^^FNDnX{Ih%A)X9u_FP({I*elrCgBO^e^~S* z10z4T!GMo2_d&##`%sjQ(cc9uZko37{WYnewA&bB!lg0pO(gGqP6v+%x?tA5oDObW zhu=&*K;JPOUd#Ymt1+1r7;J&v;c7VKy@S{9e9^E-17{z*L$!20Fnp5|m}Uv1;*)L6 zfpsHHt;{Ow`vENG+rFS-5g|0|?syXL@F(3b_7j)yUycty{2-h8`O#dSn^-bF5&H}) z*cqGvDZiruJKZfYf_MHs7kW%5cxR!Tv>4A(xJ91u9$n+}Xj;BH8XvAdh`n2OLTRo% zud_tK^rH8~)sNSd_C#Uwx!st4>NQq%@yzKZh0N!==Ln1pUF#7=Y%;1aN?DGxo8-ie z*F!enEt%wu`I2H^QxvLdqVvLYv9oano&1Z>oxNy<{6a%+ZW(_z9e0bEgvprueR80& zI^vu`97hZk1L2-d2TMd#(CZ7~=4CRhW~K$1E}2HU6@Ab(dK^4y7s4SpDb~5n8(em#!O!TZ)G0GvRx*xi^mT zDRvA~?nR3^cKG<#1)RF39!j#a>C&=4;N~E2Veox2cP(rarzUn4%bq;OMy?!3{4?0= z&Y$aM+FnPZwunyfOQ!P@yUB%yB&;bG<|I4YC^geR*LiN+X4prn)R=XHyp;_YDH zs=z&t+=BN%iE-tn6{IWlAXB{HJf!4$ktsvzbhF$6;^Y@gCT3dD#z!1EIkAkie*Z{( z+~diXZ*k;_)lBY+hAOvT>JYp;w}8q7cd$`i21LT`0e#Vuh5%9J>t3|snoRzLAI8gFsrwJhXYl+zC;s`!23 zEOzJFS2Sb5i~SrdjD1ZrKuo-moQtY~O>_yDyhMOgv7duAi{+_tHeS3NsWqWQa0~ z)l|z>hsF~vuHxt;rt8HQ_Mz2Ny82}R8D1#D20h{Ru;by7_r!qIA8dbEu?M?fQTQ0XfSCmnk9WABC&n+Q1=pC%3Q~Y+m4Gt3uhb?K<}CBnYwjDbX205ZSUTL z?*aly>%Ft2)jN&~7I+cCp9ni!u40e#ZM@@jod(1WlJ%$J;l}UNJfq@yUF9nwXjLH~ zZ}o_t$kK(Zpy#Z$Q69d3lum>Tdg(eE11IJ9S-e9zZdXrWFRWHU?|1>w^qEiZ`hUNg)G0=;J`bk4_K<M?oVe`>^UHmNE$lxC z2a{yLx!9PvFFM0nbWYBMK^MqOAyNQCR^W<|h#Ib?f1n!DfKug>^zV0^s=Z&5*!z4u-m-X!2E0 za?kWJiuML`3WW)bw@e(mFD}KG3Rdhq8=kGNW`GYGZbHSl6uR}#E{w@Pj*0O=bx&qs zdD7J>S})>jq~~{;SKGxzthL zhy&D8?G5QW?!oAJIicZIeInEwLPYC%4t~2P-E(XLs4Y){gwRx?pFYaY_iZGfeyj#< z_ubSZTa@Iyl?PR;E=Kre9{4IoQkQWdq-jeT`?j|T^KR&XbMzVfc0`dIr!j*BteFYk zdmYG}ob$L*Yy$}~^2Q=L7u@b-NMk4Kaf>3gxWwSCoZ24=Zke^c^K;K{<5!;in`Z@+@&wNTf*%WMHTFG+6E%MPAKV1!oNT`5oUk{P0UU zQFnPolO-lWaQ+tFvu4FTFiOF@KbtWxMjwmWI7;^yQtrz(uH~mYRTZs(J$1&QHg=L6 zm>A9YI=v)QQXTQczA@_iXzBlZim5)|J=t<*75R5B69⪼mWZjXtc|SnXfgSbn8iG`PDpGp;hMKQp?lPG`k%rk^f)~Yd_)aN z6zfVuO*+_&q5EW6j2Lvt`}29Sr)2K_zqoD5WsK`hgV)bOaHgyZwv3iB4#6V$+3`8~ zu{4#}hIYV1Q!Cb+U1(7zWdw`fH?#SVUxKc+3iowuHy(48!he+t@C~-nO-tP{%SHrB zubd-AH&P+0L;}}%tj58jG<>ql0~J0;QqyJeIL>beEq>RAsk=;Q<-b7c`aF%cY#ksC z1_GQmH^wr}^74go=2|D^@D{E1|V>_iy+ zy$FJ%E9q;WI64q$N=q{)V^66G>GjHn^K_VgOfIqtp)13Z;X(G{CXkz_B6@~mZ8_M7us9?hY6l8T*b*@>Z*$D zaB&!I)9fer$Ig-YA5>6{+k|Hqh2kuyKj@z94zC{{B3pY@ELK}7a8t_buyxiSu^DfE&}<`55W8kd zw?4=x-IZ6NhwtSWcl9_)mYD;gb0k5W@0QQ0l!vv>YwP6aw^3Bo<0=9Ixg81-TzbV- z(Aw_-dMg*1Pkf_;)5ghRT&M>|uO6k|r;lQ(wk*8591iiZE8zY$L5q7o$8)LsJ88c0 zZ_?GC&RAr(GEaUzA_Je3@Zael^hfam8YtLK8-kXjpG-t<fL3j zuVpCDI`;?B18&R~&*>0&Y!!Ny|6@9$ztQHT2h`H4iLoe)0h?Llxr@8&se24d?Z028 zGtUXL)-8^-r|k&Me zq7eP0Xpl6Nb{aO>d#j8R4Yb5}-sdz?QW~_#D3wY}Mf-RD_J{SI?|a|px$o=x)@?KO zL>h$n`(o*|W^^SBpkL=47N0prR^EIK22SF(PPtaLd3QcxUa=(AT$RQiOx=b7=jGth z3L&_D$qUtWaI7+ehoAZ4$!YG`vRe;3#-1Rjwtf@*mdPPDicO67cZM51 zRX}GvO$Mn6%AolnnvHvNgKij|Pr5ozqG9p_^gfpbSDX3#yQC|99{Pz*%_t)_nuuf?YjPED`T;ff29$d9?nX3y!#V^U}8b4CO`*$zw zvd0er?=W?wlZT&+<-TZMOq7IjdiA`t(?w z@B5sw=*UM}Q^8H`jl?gE6Ykg|$>M?g%*^Zs^j^gz;`1jE9|&zm?c6ZxD|Mdq4i_?! zcUwS{&p&>X`~mN4Gx*GtBL9&v-$q3*1GU>eQmC);oWR9I?Xw~dK1nAt z#J0lxBNte2)&1Ba>4*;*j?R@YW=$-!`Mlv+qFrhU3C~}!I@-zX%I@*#XkAH1O9_^z zE`%AE?5WFWK8MEG1|;|#Dn7$9hE z;(y>8qDvnr;TTLpSp$L3dcd#LM7gB@SJ|qGa^3Q+L)>T zUgNG}50u-WKzr}*C-pPx!T9ef^55lQa_dzLSLCshBe||y&S{c%=4k9d*wjRG64NP*`h1&*zu1@ zKxY{d?rHx`9FdR3rxCK8-p`p>YOaFfkIvz(_)0;=$|>N!yPcjKzmJ}ITZAHub7=bg zYxrU2d-PrP6_W&YcxuHq=I+EE@-Hip)82iS`e{95i-NnT$5tx{u}Ok5%lo7*R0dvc zX(tNiSLoi(!{o;+A={uE%4qdEkUv)}Ajd{k$?+}fq;g{(ZV;-(UF+NV^Q#%3y>-MN z({9rJa${(d;Ysify$Ai9qo8b30d}3t!!do0IA`z#{e8d+@ASPR`~9YK711T^0rwg3 zmuEGMi%%i#SALPmy74#x&fuTdKE&}r5`AHHlT7p9grj~Au=?gIEOl~++cWvQ!+`?Y zJrv6CB^UGXj6A4V69_BfOK>>J7dMIvXoxM}Lwa$Qthi$VQ-0MlRbNwK)4}<$?w>TS zDpJ4(Z#DAR#}8D_o5J@J8yNl^4eH;j-vJq^?Oh+rJchrR6gc%&LCV4C3>B?i4RNZt88h82gccgi^Oz0T} z{y8(hCygCt6=1(uldaAph4U;Y3WPRm4` z@5vSbLbg|jm{#R(r5$*?dLhvO)E&4Nh#S-Tu3{{xnO07 zEO#R|7hFY`LSR`JStOr?hVumXKk_ZL1r#I}io(vn1iwvN2fJZcG!{PHJSsGkL{EXm;=xFdpX%i#i)& zMXw#b{38TX0>a4he+Ib3#R-M=H&BJWkI8e7M<_RCDtFTZI5qAle&}){?JwLQEl!8K zD7^rMD@AQJc&_0~J_p-%(G<3XC_tK38_7Gt?>c|_laetKY~@K!NVG{Mr+wAYd5a3H z`tXt-?MB)%QpX;9I}>ErYhjyHA30FgN$EfphCMq$e;wAanU}+R+@&8ALzP>^Q^%cF zTQ!03+=sAc;vx*YbQ~o3nQ`omY}B2*2?{4|1C!gaATWvq^DTYYo5G@He;MCT3kESs zdEAsJ&-Dub#OwU7?AxG{(VJ1S@5cxPj85mHQnTQ#q%rnQ`b8qvX>uzQf>Aq83V(UETIV(#rW0;I$B>#@ z+_Wc|#6K^miVDjyWZ`7|&P_)lA1}Ht`3@PV%*B{WZ47!*PyCO@v%#CrqS4J5CNpyt z27GL$r(_nPkVGD8ZF`GrdlB4O9cbVAm0k7D8#Z~=;@D;#oP5<1RP2+$d;eUr(%S|V zxkca#X27kD!>Lzd@sxQt9biN_x+?&pj#|+6UDi--{ERrKU8eexH=zA0fVQfYAk0Y_ zikD2r-$o7eg>(V4y_kR9rHHV>DF%2WYAwbbYohAcqUg3GUZ|@PBnZ2`m}=kVdw}=5Iue0Mka?oRsD8009G3;h7 zP(8$!yWc1ZAByy_>OLW_wpNj5=K&fV;EPcPQOwHBEo{k8Ym#QVmMK#^ORpyLUGF0! z^zQF3)NHCeC)bvZ>N$XB)4J%NzmshrS)U<8!NYX)q7Ja5>u6N#WXM10fwhthJ#j?{ zr<^gTx1_YN;nW|-PhX7votHp?^7o%29qx9Sh^@J?BRFL=(lDJPECj#6xm#w?kM$cs zaiS1>zN!w(&suUW2Kzbbg}qpM-W!x95-@&kI~Lu)4>h^jWcI!yblLWlJ?CRX>W*fU zH;QKD`N^5+;b?=)&b)$M2RgWoiH~u%=VqFEs{j*CC(;0CE8A;|HBg>i3Z6na^yn$d z7W4BKciR#&+gh4&xi*P)`gsQ1#(l8q@pdGql?xfc!!0m&@;gSXx{VpwJ(KP}r-crS zJ-PG$rQ)<15!l?hgwYV4PPg~5JsRURpz^Zw7| zlC%z#S2@wBo?lqioq*dvM_{0}DcEmMfV&U()8gqxGG%zhdDUVPzIGq5O z?S{$pBgr5!MIKUeJFr`~3>PKlQ@f!d@>mElesK~iYra8Sy&7z3zrf$2rV3`wkVHMK zg#V`TGv0>>$dI!otgcrDjqfMP!vn$iXI(Vc_oWzLL`?#j7B{HRpJ0poR>DuRndgX@ zOgW%>l0>-NOEfan?c=aJgBUzr^Y6i zz~%c-65zU>jvvtg(T}EZ{rOQ+G^UiMACks`c|1{LX*c;;`Vq(V=%XK>AzB{Oi)-WZ z(f#pvmWN{Fg;`Ay>-dj%nwMZ$S2~Ji%3;%~8K~#~i>TasN~RxwLJX%i5b3j}SpJP6 z9p1AU_+F3dRuZ7uEJE%Yl+hjgOCjsoR@|L^8e6wtrFE}vvhr7A@Lzfi`YU8GYE@Qv z;Br1$Fk>3Qq2pAicnG|utswS87X)M#V{~3R#;tyZVS&~#wskdAQ=COJMy)_}i5pru z?jpLOS@2Pq#mO4UcyKTfZiLN)rbY?QB)Nx5e%64Q);xE8axOII8v%R68AL~pz{c^) zaDeBeOvo$e{e+|J%iq2@F=YWc^Uskodor0deTg`(!v#N2`#|@oEAzl6ZEBYKn&^C2 zK;_`ij8|?n>3ov_1HZ?^LcRys8s1GVdOZ<*+xmp=*z17jHm>E)eHXQDIi`ZjmGU56 zoI&qO#@cx3q|+-M(Nyj}zZ>|JhHKXBq!n_RBz)u-v)N6VKRXxUa7z*@-J44LOfDh4 z`UzW{&oVngK0)7vX7VsM0T=chK_(!J8Mpr|$y~35&sh(Q3DV(yTAHGbcro@rl7#dx zs<5!jgBqnD#@gCKcoN-DLSzcLHIqJ&9~Pe&>*J>QtTh3@Jhp&l&(&1hZ7ng&GGp&8 zIt3T}hXj)z>T;a?WoF{oE!2MaGCOs$sIAxwXFe;Oj1NVgQ_bROupnj@^cpb`ZmCRa zsw7c*as)(a9N>D#XmMjKdg;K)HmrXl1&2K?$ugf_zLye=!cC{hhwE-c{m%^cXy!k5 zil!Ws!0KZ`G!l}yhkWX+BAK%*LAWUeeLhBEcqzh-tjcp9^osPM#>ddmk%e9}DLNg@h{SH&RObvFiN zY(}9@NuC|kLt37hL5=uPGT=Ui&tarv2s=nb^rPurRh~yzvy>5Ey$-YoVyLqBF8pny z#D#qdrTR~%G4Icb!t>@+c*HLRMTWz;rZa~GA}#*JqhyqT!E{(9f0UYNZG`9-CFIH8 z0Xozl0pTwi@%LdLqHh$B>&}XBWj_cm6~2wfwT{z)@9rS{zMGiWRzrZyQMzTzQ9N73 z67Hcb-MyfM_2bW-A+O(|?}ab8XK(@9993h_i!sbpkBfxlV3fZI_PxOR@AnivU`RVqJcsy^jnYhM~WQRM_sPaUNu`uA&4DMoxS}4W_ zpMWnJW8s2KpWwAfHqX&HOnOy6l0Q57Go`~{R%MeB6N(qvRZrG{q}pkm=8}s&jRxq= zKT{ggCpX~BcjC*a+y z6FAL60nHu5_c{mH^WRH^8+*PVjchVG*%y!K@XQIA4aWv!KjzZYvQ}6!EQIf7Mc`sHRo=6d z58-@f>(rMFa!LX*t+kEL2wjE(Cks&eWp3Nl5(LYx38P(sJ1kcfC$ndyW8b0z95c}y zp9dc&!OtVeIlpf-#CIPIIYnI&Ei^15|q=Qpsc~+y|CK&%?6;W#QhM>1{ zbmT!FIV!6{BCIYTn-I&Mu)jogF5h8Cw)|!12pu6h672$w+bV3A*HbD{euV6a8iep< zN%s4}S~5JXhJ@XArQ~`QwFndD`pF&~i;FS;dKvXm9RR_e_pAgh#>AgvXn7%j_O^dc z+;bys#%$;#WkeYC*)gE3ZURx)ouK809vJr9K~_vKyj8w~E3QY_tXQ)YxePOAfvPHO zJ$;Pm%rLUG?m1|~?yDm2e;8o8x(B9B;#ms1PuQTy4%+))6M9VjfNPJP!vMyJdabtt zzd1s-ZP!o0WS#|SPtt*6Y8b9Mt3h{V8WZ7nqPDG%o}k6gIEMDhlaltSurBKX z88>W(>V?7}y7~|~bvK(4N)xf2iv6JDWrl5IJu#Jk1jWq0UjiZi#=riTtv`D83 zBh0k8(Q0M9zt#fxHuc&_j(tU>Kef}R2NK|y&N28h5W)UaF~dJs2buGELbj)N0rRc= zrS(Wp98uvj_gXG#IQ^3}PPzR=P-eRWMPA0krAH>*TD9Bg5^$E@7cQnN)^rio3Omk6 zQI)g1VUF_hLbey)n8K{ua-M-Dh1xv7KYpyq5oyMHib=?mt)^M9ONYXX%}@dss({j`1abbPw(25q@>(K_t$NoK{AXLMR<98_Ix zprP{$>CXUT?7YEJnFxK130_L<6GUybZ#VFMhu<{US`AfS)WQ1iX{_7%lQ{QnAtoDK z!nx&Mcygl|l#E`7IBfw)S{w6ctB)k@#|5VHfeYw{DWXQxA<`Ckf_mG}<#G<*r^~MP zQ%Q{;HhzsLypP{wv%DjW@#j56-0uW3Q?Lz}%Z%eX?ELVj+*gR&P)OXplIV{k-Mm*% zALT4&6aVkMH0^o;>g*Y2T}E5z-bFlPpXcG649LU{?G02YL;x!*D)Daf0+f(5huSYq z*i#*aH_9@}p2KBODA~bmC>OD%hF8Ijb;L&wD%{=5({vuc1O3>20E2BDh{_3dOqHbA z>>h_}cm5>;_cI{xE5+|7jX-|fX_DN|+XwcF;l#*=TRcbdqGY%@M%!!Kns zBI)Ok$!N9_U}?!4Mmcpo?)!Qf%O|P;B@tjcrjDA)okiez9=A1msb+)_4vjrVL(J~c z<2x^s9|MhGuzCbqL~41%vgzI<_m^3ts4cABj6-{oZ(^!60|{BH@iT8_cMk1EWsMWc91{0$Ras|{~a7(Ryw z!O>6d=>Kvob}r?8Gfk)HGylbGZM78V_0x>Y$#$lCBnpCa1K`5bA?ju>#^pFZBp)ys zrMg#>n@{y2uUecdcXeR*-#rc~^MAmir{$z+RRSc|yU+~DEc{ERvT#0G(E3xI1iU=Q z?7YHf)3h4tT9NtOnb)4Mc5@1S>{EiXd)-k-O%*2}v;@-|svMN&lC@|C+rKjoX=kC~||K#hqQE!+w z&v=4T`C?r5&Qyr_Ltv757R+fGPiMF05nd@qh8lk&anZmaT#FMc9--5v4qPYl6%P$+ zL8N>tRe1Z5c6rEx%uqKQnxlq?CWzTSFF6ff*>UujqbrhmJokI^HV*$?fcvYaLxlQd zERpLX%gz*Ik;z^5wB1kIV!9iyizmTw{~sv1pa{C@snoA7l*#Is#-z&W*lwl*)tX~* zA-^*Sn`U9Na`ZIaUJ*^cy1d3q?tif&)Dicolw0H>MMPqH{ix4Qe2;k7R)q1LDIY9VAi%VP*_(A_vShQdor2M^^ziAL;H!s zL}wUNqR65k4nl)+sl^UEqQ`uN*h>rk8Fu2KC%L*&vY zd)&5DK+>Ee$*0|O(A(+_dza5H4DC$7fA_mc?WZQvxZV`|^^e2S#r*p~Cd=-6dzfck zMq}>q6mIi_WGojG;trt(s?A=;>>X4f{dxKLhvzPCT3wFw)sEuSBrjZDkW2@CUQzKD zS(G2sf^BYdu;73K2K-R~;dx8J&}0m@hkat%X^%*_+8DYcq?R7&)4&CTF0eV|KCvi2 zO6DC(;7&c|`H8D@aB_+)x7kSx6G{%#8(X5`&ZH~kv4tiXntO)i?e@fh1IJmo#cj8Y}-?%X~49#Txl`{_3kC(B{zOahxC@pQ=acc$~-XUgl zPhlfHrIm}3@0*C6hz^84m`h|<6k_@H+suiz??|w!5-3}rx0yb&1m8`diq_(mq`@La@I=3z z=5)TJxkHz5`e`{l@#-$NrKO`{!bPU?f)n36EG1w6o6YZ1J?Y|h0Nuw&$pyt~+&wrz ze|`@k)8uD!3l-ZiVedEQx^O#;P6>m^VP#thm0x6C;Bq)4Sr4Wud$}D;2eH9n416DQ zMyXN0b3nGD-HekUH2(|n->z&c^IeZCI`4og>n3rXYip57iKVWsr_i(_lD()l5i|#1 z5Jh)A^l$aU7XHjUL9>H-ul*RMK1{}In|q1emd7-6JHKaOSHOccMMT8nD^YlQ41Dh2 zVchFB!Of>jxthlmGXHA;PZ?zhNU9(cq-0>MX9-G~%)%~y&lkO|ovFR+Pjs9P(6gn=8b|yTDGJF|<`J65lACVD&$*A=RTE5cJC(MAJG*X{{u=ELaTTav|h$ zdnnVFKg_KC@tF9FPXbNzG`zcg6}DS?!(ET@q^Vhy+V|RH=DJSY*m#C+D%(XAI(IYS zojv4-xD+?WR2*(gOoumfkHW|qUE34wacFbB7JMG@&x7a*Qn2V2n(i#Y3Re#(T`CV9 zvbxNXx;}oksLM4g+mruHRKfc5MfguTpBgw97b(Uq_s23$@jQ4X4b>~=)O1?69PO@Q@n>}sGMSIhp%AC2cA_uFN!qZ ze#6$uOOa`da&c^34mo{rJoz_n9$c9+h5Y(;iL_4L!ptn445gkc@$<2hAa8Jo_!nfr z^pE4Y=?AUR;bR;$ncicyZu)}FN(C6WahH8X8i@Iyb=0SJ06#gsqC2K3VAK2uAgX?p zKmYl1Nv6r@>f4LS7rx<)_kg)am(tavG&a-NcxsoBs8yq`M-M?Ie5@uhuawu~tR_$ZSM zdIYEXOL2h%3S5D33G?@Q1`XO=NKC$DVYB@>^7(BPtksjphd*|r^F%#3wDmq2BZpY) zJPoC9GPZMfxa0MHDQ?X;FYKMa)Y?jk=c|Qn!^5Aesd@KzLEVab&{=JOA3j_{Rfl0B z^~|0GZ2Qb!F{;GHf67Vf8d37E%8=OrE;zJd9%d!JC%b|SvBjwoN4qM>zA*+kyTF#{ z%EXdvO`i!xKeTsNhw`K8jdC~SMHgidE9vPu~~)>mlFulu-DYj?MF|a@Z@pk;Fu-g(Z8um>@rWd=c`ToNbtfWv8mxh+V2|ch4?ZYkm;^ zl=MJZ-9dQya|UddJ%A%z*B!#BCN(hF+sAr&mXL-i z&GgbwX^1(kK_+;9rbesNsL_@rvPKz6!~9&>&fn$c?2`q}E3$$k2UVzXxH%@z=p=Vf zoP*0c56C-DTbvM)geTVlee`e|964Ny8_(~cO$#Q{yrd3b4vmsS1@GC6Npsm9#Vjt3 zxQMP^=V{8+N#MGCCwb|A6+05EaHfJRq>J4lYV#k{zdJ)Q{EIs0byRVi_5LK z{Za$?Bcm8BQ?mHKdjkGoO`t4g|in|Z0Jpb zI-_W>>=v$Iw;L_m8cPoDT@FiMUn9z~$>iMUUb24LTu$`&b2RvOjUG0RBmNJY!CG|( z3KhL(Cx>f7!#=(nBlWb=JvD*sKc9!AKT?R*y34>XgosSvG}wGT6qNW5xRjR@vbNp0 z{X`{PJL`ba&dKmrpZAlRYBItbo{_IDb8%j-GaV-@f%k(;;MyEbJ~QeAwBZ=SadpD( zyN!&WC=^UhWdmhW$o%)GsC%guzOvcMhVH+EB2pIkU((*j#oLCcpT7nq#I;q#GW19|_$ zekzlriwVi!X;$Yru6zActlc&R2Z9p0$Tz#G@)a+}`eG8BH~Te}2|q^g%L5Dy3cy7U zI(XVrixc1X00a7WbNY?HKA&MXVV>ifso+PMPw`>hMPws~~_hBkUX zZ6g|8^?2*DmU=o(IWDG`!O`# zTZIuGnkXu|15~rGka~Rv7TN{Fnd=kbjgtgdHCBr#8aASPDbFVU+)2BqYtc`R^>E7d z44YCPg0(MsC&-ObG`X@Bvz=%1jAC{2K{}gm`&k1C!;k2ZGfi}dR{{4_Zv>yNilI(o zRWwz)gh^WKO2zKaX8rXf(Bo0*MpZ2it@35Lrpi(r+A4!8`)APo5hm0;D}~7#sK!IG ztyua`x}o_!|Nh}uhykTpbXi+3ouT#({+2DbIcOOU*2CK=Ur!+~|3)w$!}(e5Io^$s z;!j`pm9x^T3hA{Yxe(?FIM{X+Ujz$dnr#*y{-!}u zGHm)14~vD3xQKfSIBC{NJbz{)UbA9ApTDmkA zM<_G09A^pc(d5hm_I3I**5uYV;`&_5_PvHRzK&WCGH#z~@R93uWcDH0BHfGkA|se( z^X$oeV>^ug&_iUnztk&u0vF^f%$4ro?+?fRk#%3L!XycCa10mbN>1u=@_Y1%y{{a) z3C{5x4!&&Fm5ahINBQ0}&}A-z)M?Bfnm4wXpwW2lYxa1q{m4a(|N4V?Tp@G{DPb-| zuVrV7i8Hd{La^Ohh9ocfF37Wpz^CKy;OZTnczxnyoRaw%I*R4sPhcjo+f%_wgw&#b zvl-0(HHb|eCCtKL7osYhO~3r~fU#w6FusD%Fqrw_e_yk(prDTm`Obg^g!b)1!y&xCh1%@C-6gxpEZp!$tX*ktXEQg0{QNZgRc&ChoA% zARJO>KZc;t6u8#(oHSKG;W?3ua6qjPPh-GDN_E7BzW@)PL< zwSzEO@(@H6@jRMrANVF+4AbAuV?Ey8BoFzWW}?&*{A{rYUbG)0PC15Pbhw)3`=r?E zy_q}zGza(XXJBmRRn&@*gAH1{a*fQFl zbOnc`W0;4DzC@&TBYq2i!*`lWu}u3gU4P0Bb+(P*&GIcE|FZ}}<=>HA^Y_3-IZ09> zyp0A{`asrRY5d(_$n6_j1=>XhaNvjqrdshcmAU3bgz7Uc{*Q^#)jkr?s77l{D&gPC z4#CeeKq@Z|QKo=FG#p}f$>Wj+lh~K z3I0BdRrz5klKYc9z7|7<_ou>42@Tq??+pE%x}JI*iAF2sYgmL}at2aENN(#JHd3?GHzp4JX7!B2-){<(yI&gahi}Q`Ph6N zhqu{K2c-tIZ@P$$l|#7hStVYao`V|nKK5}O&K;>hpJ`qg{@DOWeksC(+pA&QGiPX) z@JH9kO0wSmD0R**qzy+5&~Qs5bykdnxWo4v8R>k;H@J%R_d99k^l9XEu_Q^PB!(`UzZmxfGusa|LaS793-o455e4kY!W) zNoz(NY%w^9D%ulh@bVM%?oQLqY2mf1+sljn%<@-)SNZDbyw2ji7lFN*~U>_9{-44D}{J>*^ zBOFhBOLlxWqc6)pkq1ksgNMR7H2t=L?(nxp-5E}(;gCTlO2vR_Zvs9(ehJO1?C_ag z8QMGUf~YPl$ez_fcL@CH#-Km++nVd-u=fhg@jZ?)UWQy}-K7|IAj}Il%umCW-fzgbb)L3T3Z$C6y$oDvp`5IRTB!^QrZ>9=ee4O@@d3 zqhbG?>1EUZNL;cd=m_=OZ2DS6l}%#MGIlmuU!@D5jlMu*pBEl^7Ksi?PqFLPI{53p z939oXQC$BWsm%Xr(`J%MyEZ1%!<97@MepFBmsVI7GY zr$W1M5`C2zhYO#6!Xsl7QObD+Tr4)>o7s$~#^O+TXeMoSW?kwjhl#8Hy>0&iEe$Cr|3pqP;h z;ylNG_s|(4I+k||39TcR9_sMVdIIV@bzPOg$^kwTP*q@%KjB zw(K&;z~IGb^&PRf2Cn3)`DTk7Mttp!svQd z0duM)G1FO&e6v5qG`Q$c``lq0*Em@iu?r$%*Malylr30iA;nFaX@&N;axv{}Cd!Q( zz85Q7UeD0zP1tgn^VplLH|Cw+ImHd;{C@WR*|GFVbr~iO2DANR zy6HCkquizUyvNt30>;V()2A94u*jecEEhkbiba+*Y~35W$!Ip#IT1XbGLAY|sIW2gtF>6S>pcH<5?AQYGCkD&;&M zr|lb}j#JL!e$5~>8+}4t!s;23wrr9P494sd!bd%8dEXoV`YsY;Be*)65ho7cq8F2& zhU4hQq5_P$&NCuK6JfpHZqSI{&HwFh5G-d#j+U?g}_!*DWvza-eyt`Bdq$8g>26a|t|hIetgOk*lw9V@V+xeUpLB z%S!3XH_d1@qk>)1dc5KF_I*@$l^ii&-AnG92jbZZJy`AQ#b;CAk@!_6o6m9}s%KYqjbFXh}|%0x8Nm1eZG9@8RkYcgArzw1PvAb}-uaB;al z-^r~+t@qDped-1FX4?dOpdX0|!^U_eBNl(_C}GiB7wTN20B=35u*P8@Dh6$b3RM9N zX^3&@ExUkKnQrUEyQB=4*ik9#G#IGA01CTqlQp!M$>F2#G4|J^$ZyUb+CwN4X3q`VGY zu4D&M#gd@ncpukQF2kEcJlnor8IJoNU|~TcE|W?|tNuUKLv=p=yM8)q9M@-_Y??^x zX5FOb%ujkgPJz1{Vur<%8Fa0hC?4r|AP@7+&_i=P8fFX-`BUro|H)y-sI`pd1^6^&n%Gf!6QPi&=C2KDSI6*Y~y*I$5UYD&O*3lmI3?LkI+|P ze{j-&vzb7jRdBF*C!7B71mj^D3q|f*=v?oWxbZ!qYW^kDRV_hdOFRm?W5A5>>;CLNPR2F8p}Mw@sq?1@Tz{YMcdVJq zbd@BsGhc}ihl@+l-EkLN>=jS03`UTs#X7M6?hk6SK$^>LJVe^pr%}BwA2c}>3&$4m ze(00w{4CiK#CU}Wv3Vsam^BKjf4!i>BnIB^_eAAySC~}mP0W^5E3)CQ z5!b%C8+|LMV2#8$TZ55QCMSjGdvuqBVtE<7SDFKb^A@txBua3b@qRL3lMXtgEkvS5 z5e)uypisU)PAyO6&mqxx$t{se$CSfy+fVHB=eFeW%Nw*Om*-NMyr4hrBgy!_T;_CV zJ!5=s4w?1F4(v$@kqME4<2|t`I`#n1IF6;>=mBO6i^#`V8GK9T*mh^e!`MqKw6+~b z)f98ooor3zUp;^c_U`1Nq8H2 znDs&gU$WhV42Gg;T|b>P&?UI|eGNOtFPXY;zCdp6&Ii%OE9sT+CnUF7m1;YW(ioF8 zk`UTQt0H%?v1)`qQ_^L&FYBS5FW%yurAOg)&u(nq|4vYtvkXp*@GfNeOOSO~1aGyT zqOmo?Jd2(|u3L{07n;fa)6&2KKi*?fH4mTu6VN2-L==6RiF1dnfh-)LsWQ;Lk(C#PyNpukP=Xa=Bt~g$Hl7kgqX>iKh z5{8>q@aCAaWNEhn-bEy?%k$lmhcu-5AC_;OkG^){ zgpuEe70r85WMUQ*Tj>eClDreNPK>H49>#xOhp{j6A=^@ZhW{PU!P2COw&#!3kcOKH ztig{KFzx7}+^Q3J_lPWNNSuZl4++eZeF5qGj8XH2D`$G65UXOt$SN)jW-XAjWfOdG z_MgAh>5mAPRIn2tjU8quE=i#hCq3~*(F884={44vZf4>;dsy$2j$0=So+C2e9{ z zNZb}KBD-V9(c*or_`t3XU1Xrf5vI-AIC`WTB1riy%F;t}qeponXF z_M@FtBRyN13A+vQ=tFKR+pH|c=sezsUp!n%_1n`t$D$tBC1&Ed=*2t(kihcB^OzI6 zfSq@xp65afam8hb8b z#Gq^z@w}0UkLE202~EDrS7X4Cck5}F_C~C3J%JZ)S0W^z#SgqM#lf@)M=my!>q<^= zC^ZeVnZDnxmD% zqtry=j$nn-KeU-8!8UiO<7@TZn00$T_@DSr#%KS7MK%X)Wdmg}x!aF+v_~{qnMpnl`a`Ow<*FWcLg3tBwK_RIvt z_Q)q#p%aQJ79Qv!dl2%ke<1ed_R#KJPomWKqK}LYcvm%%?1u|*mx>>ywHvT~BnSB; zHgo#jE)dTfM2GNGu&BV%*42b(*Yxwui6fmP`}a}eKW#Ht7$D7!*=>d`-Upd$ORkXz z@JiT&7{Zgy`nMWmNQLiz4Vs<3G^|~WSZ?a;u@-$W$W#wK_UD`;@EOiC@_FN+`6{O)-f(d2{EVzr2=YZ<+YbLYM% za+}6;QMtCrRh#0&8RN10ULI7PD~07(W`cuVD3ly@qVn^{bINUH^lNE1IsP$@`MTv2 z+&{dQC~vp{;rx5OTEg*|Fn z&ZW`^V!%7xn?G0Q<8->6TD*zJKMSwX-FGzc^WI!o5T!vTU(CYEOYX5dkGax@iI-r@ z%6@kJpIo#wT#wU^$njraic4)~;_;v9u)@ZJQ+MY*4VHG$khzI!26fS|Q9tN&!#%9_ z{5G7bmj{nO6p-^zbU~sz z!Ulu1Zpmg8@EjSv_g5f8I~ykJ{si$qwpcV~27J6*YkgHgiECF0#I^I^(>2Ki*6_^o zQ6aA7Xn#3bRp49NRXhb7lVw@tSWbr6=Xl z;&cvvdQ*fuJ@`3)rwCdX>f%kg8<^$#k**G(O&iOk{B_gecZx>Cr3n*dBGF zG~)x=8+id%r3vGPjfNn4&zQ)~Z3pv_F1jY}4xO7m8$Go*BWBFPL0MH8Z7pPu{V1ik zyPaSRn*wn&@@dJ1NNRN@7PK!;q^U0z**l|knD}WEHR@2NPMC++DkJDWKPU8hd7irc z+bo#pppXAkbl&k;y>A>xM%kN`ossO6;@sDhQi>2#G&E=rtF1CKGLn%*MOj5OjO5(c z=}V%hG=z4DmWI+Kzw`UoKVA>dbDsOYuFvQF&N=>)pVBrlzjp^=d_X;OeD)un>7Ri&$PdWlsVX*&L(dU{|8SK6~X%LVPYX6AR0b4a9(3BiTiXC`%2H^7*nKk`lK*Z zOP4z-rN9YIa>GZbe-n|InPhm{Hv03XDSWH4gt(9tV%BB>n|Xf4w34k5DI|cDM`y@= zbsA9?|bpz^B^p# zc+O6)s$|#LE7P9qwb;7kJ>9ag6$|9A(`T+4;H!KR&b7M0c%OxW@QO_^IY*A~*TrK> zj3QjtbAj!FF(~GlOJuirkx!gA4fB`@YOdkX5VaT%x%lCywN}{L>WFO&rnOQV&*%tfJj3eM(E(AYhQ{eQ26GU*) znA?`=gcjkvb0FS@z(@xEc`Hxz(koC0i*~H(V3dxm}Qcy;j-cy z8u{xsbNinS=fm1D;nymewz4^7&b8|pw&D%>5t#y#UUp>I;0D#HNd%>^Y<#4#oTzM_ z3KjD%V%mpT1pcfRHZK|@3TEQGZC!Na)+DTCkal0aM;@qG5o1M3@Rv2fT@~Z-rO{2| za%G$S5{1W5mAaRN@2ZEX>ryDz9;P$g#&APh&ZCKD7aO$hCZ7KqfU_sGlb{zXm`Rt* zFx~zDE>HPL?~G8$Fc8O`##!XLj3KvSsyP*F%V4G}iV7||P2rhn)4_u8u%>S@hZgPm zb#cKjaCqV&8aVR=4SxKHb<9vgduvU4db<@4Ot7JMp2xyTtEaHvAf9ab+X{uTcW{>$ z&_u5Udw2g9Ch%83{nX+LA4Q^p@s_4fVszl6Wf927O^5o@56tlxQNdjGK63UapAQX5 zLB-D(&|+OF8CN`yQ_O0ki<=gZ1L}N6ez6#R8@miv4;&*$KM>4w3L*-UyaU^1KY6CR z9h_`-faB(TviyfW{)7aKxT-*fd=oLTZ9a{7kVPu@=)u?-QvqiNLBHK9QgL+8s^~@TiZY)X#zcIFjb7`y<{tzSJeQ$t zpeK$!8G#au@6#1ke4Z#I1Ggo8#EmOu?akYP-w~OzLG(Nn@^|bRNlom_)@zBK&Si1T>1w7x zQx=sTIN}i>DOmkU24`GvqryK{@p{Z%G(O*hJUAPHPP~(J*ff}#Eu#**TLx*`1`mu7 z>;@>X9#J!Kn(aVu=Y_N@twC&}3 z4=%??oeNowtU>n1F-30Ya5fbSK8__jDAqZOKveb`_MliVz4X|STBm(uCOte!yBhQ8 ztY`O7P5S@}{W(+znzKn_XAO=on+PheMe)YEH|Q}&6NmSY<>WUQaFy2jIAgy$arf6C zmCOpb{9h={Ic^P)uf z9JOs^B-~e#hAubsy)czL-VloBCstwQKoWXuHqe*m|LB+ljTm$I6)kmr#4ev$$L6I@ zA;Iz6;GB^SNXLAk`!$(@L=Ww+R~eC4+$`2iIouJjQ*8`0C#} zMq$fS{F5#Y1-J4^oG|YQ4me9?d~cwIXg=pRWY&_R$U^dRRst62#)Cz;IT_e`2CW{ypvz8wCN7I*neF>8(=-E~ zlmF6yPM6~|>fc4Uff=XjjM)xyJ9zf|IK@<1gbly z(7F8Yc{9HSVuw0lc#jmF5SE3@&90N%A#Z4X?ge^pcM=`*ERns@Z9xwFGsM0mAx>U@ z1?Qb5!*5H6*!89H_A5RdEal%j&&5vQpRx%2Ja#M|DXyl=7xMj2#T+{F z>;h5v`4Zp9m17u|(|Jz}?fOmT3#x9Mf>*EKktc~e$i(v#!EdH9@Dg*N?;bFJgI9yv zAx3?E9L(S>hBxTYXO_#JSU%;R*3c1t_0 z6e@u;U2mw0w=m6IASamnM~@q$oQY==Qkl$7C!`g6Er9fzvNb1=tA1)u4}pwIns{5zk| z5=$&b9sgUb+J`h)>E_1OXbsbg#jBYt;YRYL@(c;_9E+LT9$}$_Jc{r=ZNpzj&`qWh zi$^;!v+XX4n70<^IbKB169Jq@M>euaW4VpLI*IbqDC};Vz?}YTgrmjRiOC9O2>Cb* zvLcVtX+`3Arret|*!PRXAAU*em&U>~o?|FEn8;ikuY!U7wJ4J`lY7`5i?uhOkxOfY z1+xQ{!SZS$4QIFGVT}S*D5FFoJPg)9n+9|*f!o&e!DQb-`ib5k+B{D)YO?{gG-se+ z(?Zaoa*}*4k>l7ASjr22`vOL!G`-7@YW?HtLy@){7k=T@;6NR8C zx$;yaEGvRlUL!CwNCUU935PM z;{$OE*a2(jPsZ+F^Ks@e4Xj)q%WU3rgWf85SXX;B1XlW*!FM(TZ3K7dGqrtnU(rb;kS?}qiNES)7R!daG`LDBX&$e64O8hhu^fA3nU)DtB*A*=uMH(@U8p&Dpysjj zc=m)TSfrSOx4;CyNa~a2u^N1rypC@Adw{TVoy>XzF|=Qz2HRVr;H~^Nz8^6G^$M#9 z^E;k4mga!>rsKqL({$8uvxJXM-t^leD|%pMF|m5Wf`#b}PJZ1hrZ-uRsL9Ny3F|t@ z#;XJ5%!>lz6B{6ts!0#O(H^Zzi45}V|&*R^GUcg#i^5mU^JI!8NE&< z6J|Hin5+ie^=*NmdbJDrWw@3JO&%rRyq&T6;5@o)s}6==;CIN!9GL{i1hjEiL;sFj zY>dG;CTk>(EkDyv6cW6_U}_%O`>lz2qu`GB{pOJikInJJ;92@azR(PHXo?29gA zKT=ltKD;~?2ha98piF%g+;ZdHH&ZrJmE2$$T6rGcJhBD4s1rU#_Rv{R8_-DhJQXgD zL$7f=nFOBkrf~5nd^FW(zT{g9o=WV7^_9Y$_^w{;6Wpgd_a-BrbHdKiKSWfgip)|F zCgYEEVt%L-d(&_haL1m}va{VZdzUGQ#x9_dd5UPVV+?ZRgW%z!544d#Qy&Ubpb{4I zS&M_QAm+rw-Pg)t`bY}B{lNt_#Wc`CUKqCWnZW(QrewM&pFg^Fj1_sm)jr?(2~BI~ z^PT$yOw*EXx_vzIOh5<5A|MewNju=>ae_n3;)n@9pQ}5$nSTz)G6Vk!=+%Pf)ctZM zd>W1-=6ZJ2{ni=6i#4d=z9LoI5rXy|JBZs!MX-{biZjyYc1&TVDq`mvGiT)F_{*xj!M|_d;Q~ZUT34Y!B~^QzTB03c>L1``Vl%JWr2jmI|em zptPkTH-G%~x)a8w_^Ts`ivJ#uzozC=orZVp7y6ks(Kqgzcq+!w*t=}Rzwbu8Kx>SNd{f9Q;v17XgWQK(`888p}m z`&N6x=_Wo~#P>@+21~<*3JVZByt3|>TsodPD$gw~3r5579BDTa&}T;RR1mz2i<0#M z?y4p;=C1~$_$?E|56#2eimN2>r6Ol!rAzjA*U*&JBdq@XxAY#ph|(fF<7fO)qL`46 za?#1Sa6=eaZV<+KSF4HApJNy<=Eo!W*Z@VbT6oemTUvJ_&!k1;xw_*erK1jv8Svv6a)CF94(1@A7?N*(|m#6UO znk==wR)w3l3JC%n{otUEEP9OdK^LJvjIB~B`~1W$EMJlbqbZ3v)^?PxdR#&iKNZkJ zB{#6L!Ib~39HRd9K7--&M{1K836qppadUTW<+L+<>}H2us;e&I zNXV@xq`NxQE?D+Ab=zZ5mzZ#aE_d`HChmOJCM1g7)3Uewa&9Jf)507}Wy5h*N(_Gc z@erF2+i=tFxsb)K>CD=d2kAqb|8T|LHv8N8x2WzSQQT&CiuwKE6o!_G;p-t4*f#4J zTeQ^*#dnrtZ4B=O94ky?)c(?PWlOAH^Z_@kCt_00Y070gp;3GU4K+B2UT5;imikF> zTkw@c+a5rjJFT?al0c7*Jja$XXp*;_J2X7PtQNH4;2~LVO4Vjs5q}&lKW3qang~%p zWI^n|DYCg!F4Ef$RaCyifa^QK_vW{k)9q4I!Afg2De&&4;V(U~WaCqu5aNqn(g%?# z7)Nzq$U*678629gPB*L(AwzR};E-z&v|OJ|Mt{g7ZI^|IxmlpRRv(wWJ%!=@{do4} z8tzJ*EBq(jNQQ0r4ozDVoe}ncB%EnROT$(gZ4gf*YtC^(b2l@8H(KMT(COGSxCiA5 zE5R`Q1Xc}+2;{{LdB)!@_S*FdcB^I^yOb+K>3w5x7y1fFhpz*&)6usbOh-BT$hc0OQ!+}ll`|2Sg**uI#&y?BIw z=G5Sc=q@lE9-wo7Cg7$G*FZfb5dw^hs08nyq%$LlNmw3zCh-7v8_fkh`W8f0WFe<) zFI-7ZhYf$6K{n(Js_tCHb$!YtiKZDaroRl=K6=Bvmhpp8_Y`tceI>aT7zGpbJAf7) zW`qi)aQcrV{BYZqglBaTx1S1vPk&@lVp|IJKYNd6_$NSIdHAKQ!A?f;H)7 z;2>InYX3e_%E6|CwI<5{H(IK$N+Y2}hP zXuSLZK64P|dwCae9sl0%&`BdVV?FTU>KN+2RvJHtCDeJ=OoVeW3-Q~9Fg*OT60bVj zlS6XWsJAAKxJ261*;}-)n)Od5CZ8nj$6vpOdd;uw&(@jI4N-HMXnvl;e11lk+?l`` zT~>f)>IWgaR|4FA=g@JA$z-LE3J5osvQFn6?VFndP<~bkm`(78uB~+-A0W)#{2PxM zV`5Qq*J1Km<`eab=l3~Be9$aw6Fq!A7!H=4AhxaYoXt6)7sW?#P5NxyQX;~|{a8&H zsaja4n24Jdt+3h9kOaRfrr+Cg$vDvzy3ne-&MepyrD6%OTYC+}w2kO3$urn&Qi&VA zobYl_I$7r5ON%F5rD_!-96vY4W0DHc=yREj`D+g?L5dLaCKo&pDuL<=A2d&AFhsO89bNtw(@SpnSHS5&6b90v6(Kx$76ycw%4n0Ge5u5Z0F zy=`NM@d`-Sdi%4jA(=F-ho)%9ysud+fvdTGjHH(r(XyYa)b^VM`Lj_KvJA`d=r;@4YoAPtdiY(8&3V%I^96}k zwt#Q;iWonl&P8cmz=)o|Wc1!nHa4}KY`VIGtm}yag$y3S zzu%Vj`eS2pV_`ck$O}Th5CM*Nn1#&_5m>tG3J97f0M1DVALm%oJEsWcW^YHS3?eV}OXzxcj!t>|2IFQn*lW(zhyCK$h|I^y@bUUtGVt;; zS~lIg!{UU zMvmsP)9&cOHvLIVfz&?ix+jhhbsLrZ(y{9ii&o$7_N#uOdRpmpugNRwS~&&I zL_A|}W$S>*{!2IkV{z7-MHsYGm7BIHfU1Y7aBoWQq5o|?YVlK=^Nh-b^^Zf@_B;7( z(~rj>FyOzt*4v?`AQ39>M3QClFG;C)N2v2u~9S=96>*ajl1?nX6@e}-XOjH1+ zWkfva9h}aI0=)shE4YEYW@X{PY~@uPI-)jf>rSia+a zC!J19Nhd0gPQ$!NFQ(WT!06iuxc0@u(I0kL)uxRl@6Pfp@l#~cUk;6Rb-|#g06u!> z(W7T0=mIJrmx@ct#T_PiCW=2lM&;lYJ4d96Z)kX8A6eY#$E^4rfhulwA4Ai_iW!A_j&Qamg7N#Bi<#o1q%|Oy6dUgY;NQzZ z@p9a8<0$S={$Vm+P(qV7n!^@3M_fC86Qg6__El(T9twVZbyTFG|{6=V^Zm*s{T!{9-xsAkK0$W4VU>&IizfGQLQET!*!AEBI)1=px&O#I9TiJ|ge%-$J9 z-`wiM%6l8}U-e;jg~U{p3DLoarWqJER|(Yir=rv5Wn9j(nYit)500H3Nwr#dr+f4Z zY`h?YTa@HU`ixfcH{vx0x4feZ8V1RRSF7NvjUpToivx$h8Kgt|Q{9v`!rU|cc-URS zF#f5b+=Y*E7&CN?Ru0S{2~B;>=6U?O%(b4*9di>j43y!k(SPK;xE?&Q52O1?blt=Q z=Xi$UPjbt4KH2RdK@HVd_Mp!Ltme<03f4_nSmAC!&_vCCFhH(y!^iNdJ#6oWPvGh*YOK z?65rCAkQ* zSM!;Nw;R~+yDw1d8HCn;8i(&?`PkBcHfg|zM58#3rslZAT+yW@__q~&$y-9J{CPj) zesx&&sgU_nb%d_iJAx)1eI%aWaiv>)Bf|TyP+v3N6QL&08SW}2)^cXlRdW$(czcP& z2FDSZoU=G_MkYOCjkN8?ZL%=52-tyBP;Yq%@|THG`%`o2-OhdVhK~Xs?60IsDGZqP z%)#7{VWx}}kqOg!h_+cGc|7Ef$DT{Vb+;gVTQ`P#rlyV8mH4ifiw@6a^G3<>E6E*q zFRr?WBj)iEaJ5m6e3bN}j@8rQtVSHYP#_Kg0mYEY?^ycs%4z&2cj&Y6pgXQ!fa%Ew zTtb#8aIv1)-Ob;lPRl^l^QWxeS%Dx_c|Es)wAz)rj^~POHgJz`r9;v83*>;yH0rgu z970XpAxZ86wC>L*HznM$L~RVWcIP0~9cpFn|Hwl52nQT^UW9v%4Y+u>0yG#H&w1v{ za+5Y!V#Cq{FmbCTiaef8l_VIjPJ;x8u+4W!2dETLxF%P=>n`|Ip4!wjlR75GF6ph2AR* zXn^}Lf6o*{WBC!@WOIr8P}ISQK40jAf^pLJ zDdTy+{FqPp>2(VhtdVC&|H+`leL41y-V}6ry#`la?x&&RW2o|ybn3wO&!QhFfyYV} zx;9rC{wXEUDJf~#7qgmP`Q||v*yY#Ve0+~QPP~leb0pE&-xtnlJfJNyn%G~vgT3+K ziG8Emd-83OD*l=sL5`$du6uoLH?>?F!ROqqL4@ZYyyqDt@2@71@oj%ebkJ|AKVO&b z%lJzIJhT|GJE>%s%XONtXFC|(d_!t7i^x~OF4}SGDo)o(1q+qA)PLYU_4s5Bd$fDV z9EEkzJk*a%HqF5)^`GhWO)2o^Kq*YUW-U;R2VJVk(|Em0vB|J>qBh7U{e%w+PM*+#(ULtDtz7BN2UH%-)j@ zMkk|O#(ouRZ&Q*7eku#O{|f&@C*No;%T1SNPYl6HH^n)-$QL}rSp=Hoa*4*Ca#*c9 zP9QNo7gP^v($Txy8FlC1boP}J5WCq+(g*bIPp%b(-)Y+z_qc`BD6E^XE9`6eL$j=YOu*xK#Yz?f$cV1J_r%08PbK8ku;~ikbuGu8b zB9CZ0dV^9j!A|~j&^zTSSoO%kgtxQkE_)j&&o_Ya{>d~hIvbanzG1(<|3`ZMEF&RN zX4n28-|@OSmU``6g>rf0I0LQo@FCzCY2VzAdtI);*E?(JXr?@gTHj{>WvQ&-)VF)! zHSQ*zb~+BXU*oetOBT?{sY0-Ng*cpB!e=;spTf zvojIyN2Ze1pXI=2tQYKy)TKqbcSu5pognYsMht$W1&MibxZPKkt7`v4y(DTV$KR*) zu7#1+6?2HYTn1$3YQR}8ma}>Fg-+UD0G&5wp~fl)yXKTraW{@EbSlPTo!8X6J{gZB zTw+7pcz%FM5nbW*fp#7^3U(^HaeUqavTENP;?*?+#=Sa<60SGtD{~)I{vC)NmrgU? zD);f<0zbG})(ncqm%vMXE;$zKg9)E)us^{Gi9BnXFd~J)#$B{TWqUHZ_BIuSkH)j^(hvISV^uS^C?cl}vQLL*`uoc2;;2 zj`xcpE8Jb7u1H$o_^Xw6@mab*lS84JQ6(#W$>4U!eQ5i~o87lm1LC)hJ{9QQ1#g`Q&@FiC$72-VNTerH|I;HEi!{p%8yEnG~c zq+;N#PbIOwCxeP9j+oSKjkQx^K;~gTyPU}&SG(eHYxQ+#`Zxl1?{x)vruLj+q%J2L z$}p2OhfwWAGJXqkhkw%(a3DW{`cziom<~f+d|nw)^#xT6y@R*wkF$;MtHJD23WuyO zDl}Q)VB8k&YvV*~zGgP-D1+uVzEe6h4RtG$ ziM{YT{CF>gt9}wptoOd7`RWTe`}e95hHsh%mJK(`dX|zgm zKoEP&l%&{E=Pi18{;m#4_s3#cttm~7>LJPh$-xvKbLuT!Sl4N?oyaE^K)%Lj%ADz- zYfG&_(KHm&cYYybhUYL(Kb6A42u~=_d`hGS!uWpd958Xugwu)-sJMg_jLcYz-fB14 zZ|wW|Yum;zaTr&-VJVyb>OkK}E$hswFG)Ya@HPU>$! zH&GSNeBvAYyy+}m6(fa%y{D)z6Nj73Y;lKP992x+LC(H-LaJ`QBBrsC?17>^;I9+L zbQTV?ddeT!fb_#~`QvqF&3YLyC*r_H7Q#e7KI@*b6xGbOf=$s$`ug4lFn+y*Za&PP zCHT+ztIe}Gk)LaEfAug?{94XxyfMIOu7^+|h{bv>9j-T)=h3EpCtZ#}rWAM5ANnDv zx+(%4{0eZp^9(xqzBO%`6@#TxckQcBPasJl!T4*X3KLTiNUqg3;P;K|q2DsaE>K8; zyK~DE)qWaN7!5~%5ohjK@+nSj=@ATH9EZ=xNnux*JgSN5avf8L(NzjrgG>9kYo7$T z(np6i@XLgP_7~()TL_b`$nLD0Qo zIV>9u2Ct_^7}z)+pBJu1e|g>y+Pntp-2#b#f6mu^qqJw;62@KC7^RNt;yIZ^>?@gE z+GS4wobSVv{W5}q^u<6!_EUMOFu1u+o5_lN$Rw^afCWbfXp+bgntCDz^X@Zbe~&ee z9GFKPv(J%l;SMza!xIqUIifP;{qlODE%8uOpx|dmOoKu7tlMa-eB)7mc(`f$A6 zzFd@)2w4BNEwvt@x_J35(3+u!bzc*Dtwz(gk9&)g8gMF44p2FSGIRCa?31fpaXw)+uTBixt(-ycnTVM z)S%ee3OYP8mJ~el1?D!vTc;KAFs7rn^=7MGtC*a_wk+e(1%1Qei;&wbfE+6TfgVjQ;zTbb%xP92alrkTD~VTQ6AUIXTvLF-s|*z z8(OF*;9#RL{#%qmPix1Ko1e3(VHw}WxNs5X@cf=@FOP!FK}MERpAD zU^Y>+$)15Fq_o8l$E9VmuG1|!@sYbUFV=;7uj2RUHv5gaUl@ghl_j*IcLlR@vORZ*@9^ejh)~z4V|3whHM%R(44<#!aOKHx=Cwy0 z?AbRD=eq7kujaj|*ZYfRPZFbT`i>ZV*ckgNKiem(YYXPM4b^Fm(rj)MxauT2L{NFWeTVUEWcamFm30_785UH>tvb9YK zGe`Kouj~Rq9dsaWN3%AI=zUxrr8+gs@jEk3`KFFG&4n$?aWMMr*b?vJ;+*5W2QHee``U-C5!JZp+D>$74aXZv#Z+X66~4Kcg1ORGGZ(+g zz&Y*|^1?5WPrC-8N}uN>lpG+gYT^+5L=7%wuOQPCwy-MuDzV$@JbFDehIf}s(8-gf zCJQIv8SU%%$MOZSkoUy7Vb`gBuNR719OB9+Q~F(IJ6!bdATM>>NR^o$jofk#zqr+q z!#RqiFe?tPEZmKf-mAF-$80ftD2F5|x6$Kl1kP7i;3B6#r+F21MDCF(Y219A?&_<9 zh4SWjZ{#Io5vGMx4(a2n*=hD4B+e1b@1D#N&yKpG^c$?P=1MN?-x@M~zCBqZ+mH8_ zw6INyk6~G3Hs@>G#0p-nXCK#Hg*C4#VQmVa|F9LV^jc15kI|y%{~ZD~?j7q^HXDpl zgBw%I;i8`l7uEPPjggnm1p z!z#~shI&IOm~b~0U;b&NNBdV$A5Uk{`qxe8-{83>AvvT^?in~Qa%4x&#lyB$6Va$k zhjW)mC2a$BICpFUXnc-@+?lhXf;mh~94t8Lk$7;bJdaU5yhkc}4w!xPV$4!@kfF|x z%pEZc%r?z|_Gc=b*YwkL&6_I72r?ustP54!_%i(16Ck)~8VLHA{n@>}OW|pU1!jmj z!Ppaj$-MK`+@EQBdc zHBn8zOMdapj@LZHB%;KZY<4lHiH&Al2n28z%2(;zo<}5*=O%ylxr|FibcymfN%)sq z#DH84mX#zjzaf)*8Kf53Qvx_BKrKSxi8=yRyL+V zRzVUeSyjQzsPCa%XgpK!K9Vq*`DDkJD|C@mA3JWvEKY-V;FcA~*}DfG;aOH3cg0Dd zXkAhr2{hxe_17i*Wsi-eSD=Jw!8c>>6izt)S%?w0_}V)2<{o zVQMj$|$~pti?7{+Q+&6ookL%Rbbp|{2a0{7id>Bq@2hhrPaTIl)fWut`jZ*pZ#_DxgPc-1#j1-tKL7!F? zh%w4OKj;$`C)jW#lfL+31eRPJivKXg@lxaKAnpcgUr3>{v=O2=FzX88ApX1p`FEKnpn$H(k)~Oze zWj=^m!){?A2<;o8i~k;^nNkfrw_yuw7=9M(e@!MQ&W6EH-s@_3muKP|tKp}l!_Zt= zK!Zk=`E12Oe43!a8XXy^^ZuDa+mp9I?D#BLohu8fi(=U0yzeNZ^H$x&pJwz6I)K;4 zlf>)!c4+MV#}CXE`FV{p#s;V1#{d~DmKcXlAGAS2;{;aUHzDN7NA{U%9Np_Lhnjb6 zapU4nC{gGmGHGfc?ce~)T_>sUI$?Zz&@=iB?U1+)Mi= z+WVk@hHhUBPrf7V%#r34et#wFcOo5{ew8`N%3f2b&sCzlO-Ob z!g!az0}j6Kfd8L23s27Ep2tdYh2K@VuSf$XygX4&JFc%|$Wn!ax&d(u70;i%ittt<_-s`J1sWey1)(W4vm zJ8^VnBA1n_#q=tW;r z^pDxVO?`BTdw)NT?=EkH0m&$`E&C1(%kZAND;yK4cNX);^2}*m1-^_09E-1F7ptaW zVg7YWVlrs{p38i0DS{?v{s)h=_-^5V0Jb)i5f$ZBYP{>3{Af8Q z<*IRz9gE4kE}{Q|&20OuGzd-f;-uqhII}gXoSSVeNFSbprqg1n^_hRf*5U+txlf8r zF_gjk2N!TF=1$=ncmJYe8V=ybhEehk>+t%5H>}XveE3vt&pP?;p&cc?xG61=-ksM% z=4X$PnwK20*&Gio@k<$ddpoXQ(tt}Z8zAF@&Ee)aee_;^6jv!(!a1Kb5ITR9CZ~Pi zXQHC?PU>_bXeg$M!d>KhS28_Z-prbY^wPf$NOg^OfX&uBY<-|6{9Y`{U9;ke>au1q z)o>CP&Xy(WJC(tlcLF_ylk_|j2G$D^w{0d2H`|GBo_K^CSuW2h6`9+|F6qIVYc&{B zyA~JCID>b(juqC%_qv(Xs zGicV@k8pLvSlFD$Ga)}lW8;1?E>ZIn2|2L{62>{hf=dz%+c^Pd&)0+0MqxTZO&XOo zkPHZ2B$`Unf+@%SNt@0R=uxj=qzHVd2|glc>+CDYC>bGMscwj-)m+OYq>F9CCQw3>x5e zoJt;!#9oF_iJjv)uZx!S`<*gUA^Vd2T|OC4x&9?0yg!$fnaIgE&!e59ZpgHY!Xlkn z@M6kQ5cFiy{!JR}dzUiSd8sRADUQL=w>c1$*N$^k~x!u~zFedoF zXmGjqfN64kK{h*;a&4!Cxrj^uNb#NLm^<`=wx%~TTIU<^*0_&&be0hJL4xP&e!WBX z_^DAFDPP(Vc$L2cQJUcUm#Vpz;JOBWR*^Ihgn~}P)$5bsMoTJT6);*5*>WtPFI{P#e$)qSSC9G87c4@g8mN`|cic z%r6!e%t|4x+j5!Txtmbp{3)EBoe4R=`^XC=Z#wy`1NZZ*183kN!Zob@iHUQ6p>>=I z_b%6uzUa-S&a)oSp0l!;&hpH$t#NEtyaME(-irSYy&`YodkDw7KB6`AA%23EVAGNX zu($9hJaxPY7Tzr+UP~7@ax?Lf{4kzB_?rG5qO>SHntdo8K>PwBf*w)k-5Bp7 zH%t_++?vQ%j{QlLI&#Q{e<`d@Qw2==|L*$2T{>da43lH`;1Y3B&Z0XSrC+^9@AyAB z_u&b=KXnyjdD)0$^-qPc&kM-JYcZI`fA8D1@EpTv6Lh%oQtE?fv5-$Qyz9(eagDwn)4 zk4a3I!<@ICuwvplb7Mj5j(#pKggDHD^v??Ianl>|rYUJx#`) z_pej&GsR^4wZF8gd@4B0DPZrg0+B7V#>3U6%=u-%`97inNnbb(?+U9>^-Txi!2ad% zQ$dd`I<*$JjVmAyYrN20!lSpVfjVR3+RB1I#UO*}_wF z)O!*cn>0!d63o!C`7_z}=NYp`(E_Bd4B9ul|74RCopAfFOsf034{HjZ*jZ>@!7 z(k;Iiq0&JqkY5oG$+hCp<|syNPwxP2kvNd`KSdwS>Z6NNL%3_UQQRgUH7@7Zdg9l- z2!1vEfZ#Vd_)97Uk89q>Q*noZ<_EB%U7P84eFKPGbBKs6kAl9c6gr82=1+PrgPpSK zuxX7Z4frvEc@p=S$dtY&J919j-m{twF%2S&Ny=R+#?RTxho|!APzR7I*bY_8`5w4% z7&3W%x@^`B{9dDk`z%hv$)OhV;rw#2u}OgkqCp_3dJ&%APUrK5 zYe7oX9h7Hh!4y4Ncyj(F)0{aKehp}$;)9#?gUWjlj($Scbx(%OBR<69nhrB(?Xfxu zdo`W|=z>cP^|`d|7PxoOMXF<#iT;&8$d+%9shp22ceVRF4(u>T>oazfaXo*x*Y$AXPPV*f$R#0oE20BOa@+a2wE{eM zasij{RaiVfmTn6Qhk|ow@CxKm+sXfE)=xFg?N1RkIzE+Ka$}SkF}X^fe$XR7r+4By zAuYP^k`~q_S)jlw@>rpv#kd|32i18~z_}zIC08wDMfVGHHdP<7*z^I-9Dfa6HEVGFyeN8iya`Gj z2*)4hQt*QxHrz-k!u8R6>4%3dm}NU2JKfvSDO4ISPf+E$w{3=9-(HY8KJoT;OH9b* zOB0}X;$j$z(}l~`e)b9brm}zECa`)xt-;*K5*wZ?aVhuTP>Eg!?#AYUcDD_VoyvPt zTO+_*GlUlN?5AC}31sAotDFaF(3~zO)>qOFq}*sDT7m_H#A2Gu#8}8|Pu> zz65$|kFenGx^1YwZ7iF=Vk)Y9NhNk){t@kmMU2bFXH?s;g>L3^b{+#yn9o9oa8t7s z#~u@)NOUtS@N9;vPkYI!m?dDR5s4l_sOO9Yn$f59VjZ{c4R3RA``RMsre>`kRSHQ-;b^D3u|Z?tQJ@sFJ9Vp^`Mx ztWb%b{k~tHZ;#{f-R-!Ky{~Jn-|svjeE14YoVe-#GqabJcQZSLPx14T4w$3uZKthQ zOs&NvP($Y{RjU4q3s)Es?R^)?zti$??t?AS__c}@ycUD@Zz?2e)Ju$8|AYpOYGstW z(;#Jf7>vFzg7T(*+|#My$UJ)w(K{%9HXZ{8c}Hohryup`vBvYc+qp+yPom-85d6JN znp^xj4aS~31qN7lEleZ-Yf@DtIU$Ogp<7 zL`Q6JK5nL=oEDrvZOQHD^LZXahFH_0Y1bZ+LEM{;!G6opf}X}y65MNz+jbSxN0uIt z*9 zOn7xj*vu8v@zxU{FEN4*cs&sp*k^-cTD099t$X-kM*yBzlx9ht0KefFm^7aE^X*$n z3#3cQr^syLATI&=!LIo9cLp=$+fEJzlo6HiI5IV?i`A9p=QuiEWH2lRFl>bQ-Abg! zx+AQ{!X$QZMkg_j)S@iUmq>mUK)jzkCZ7AB<9xYO_~RJ=PCDRAz8kG&|Alv9!Mm|o zA;mkJtt7be(@r4U{+xf`j^>Kk)3{{CGCH}@6BfPL1isJC^81bkX4+8GEW`B|sd(5A zy84$0mou;vzv!+ZahI!Um7PA6HpD}fj1w8fvveQQ5^_Uu0)=xb?5bvV(y+5T;N+e# z_R6MfZ1VA)XcZ%ZGP{=169xJ>XX8eyd*m|qg{#B178|;xpa5b{eQ_U1j6Y63I)$LD z8mE9&$HSWRPU6Qulj^$rX65%& zsQi;d4sO0r-rn{?5sd@n_yJq)mUJ?WShoV)3lo{y%2jZ-K~!+;L=zn6c~&tUPw2Y5 zspQy&u~;5{g?D?)3tU~(f$%w&!J#}bDt08LI7t3!nUN>EwZJpIL-=CldEq6s40`nN zRowT)12@J^f{j|uaAIyU-uR$`QYS9LL@zb2_RJz~LEuDuvqe@o-mQ!du5Lof6o3Z9 zIZVu>Y?`yIosD{#!f5gTo8L+NIZQeQC8T^g%`++ZGW;VNJUGjH3cc~mRZ&>>d?h#* zoPxL!S?Fq*#`6vxasTtL^y-p&2=a`BKW8mz-J9!JxAYe6WNynPt<&yeS|{@X|4z7O^6mi?=Ic|lXy0o!JfaooiFmR-Qd;UT$q{~&BEP$%=p zDd2(l&DbOPU>?j21M-~I zP2AWTANp0Gv&q>Y?F6~9$<+lzNNC36jW zEB}Sj!DHy3=DS4C*p$vFyAAWhb0KbdDvoTvL>I(;qQlb^x%gr?&db4?8(H=MFQ<&= zHmmkfNt;VJ{h%qK6K0aVnL6M%@i1hU*b%3NOTcEIG&YNk;l`dlgQZ+PExh%IdLB7O zE&7yEaa20webeRLlaio)UW_69eer$nQ4Hq$3tAs;;-#Q5@QmNb26QS5{JT0~MZ-8j ze}ykv$&|zD(3^04wFQveGpwtt2cWqJW!KN7(=*3nsEIUAjY=fZbGFhe{>7*taRh(7 z{DfmFP3ea9iDa?-7vk-^n494!$)z~B&~iybB5ipJ!{^_En)zqxI{RF>zw;F6x<=qt zIdLqV^Okk@3E+OVr12i7YZ!m15Tc`W=snW|C{Q+~-=mZvnD-?f8L@-@TMUy@djb7y zTcPuHJafT#3cSe5gmj0+%-%O=$=O}@nAjghp6-7~KJsUJ_VjU5)msITWA<@&6GP!h zv4E?4C(CWWpiH{nHo;<)=Y*?@F>vx;tecxo6|K|o5&v%8wsR>zs~V)5DP5#=k22=e zR>1|!a`<+yN!aHR#_H?bXZl3>Ea?zZ(}+{h_j?7tzursTO0Mv|^L8w=bi>5)&5X#( zRa8{?guOr}g2k}Qs3S%-bpWZCbB7ya(IAuu%s*CaC zy@QHmO_PFP=k?dbGt3bJJ*9}T>K(hNdp7jvu8(BPusQxtlR~x8DI{R`Z1VE!C~o#9 zcb+FhsLke=WW}@$R!3Wp4s5*!pN~twlJ*>2Ph(xsh4JhN?76uJ~0!mnN9V5eC*3GJOjeX=!C zU~WQx8Qme_Sp`%tJ`Sf9mDVh;%Eomyp7?Zf1<}5gL}mLO;oA1suqKvgV*i>7J`GF3 zw%(g*%Pt}tAAF(dM#ywE>5_AM_kuxC7>>9eA*E?W5Z-nhr^$-*cZ>*>TWo;2vCnYc z-)MYi5y_shHN>RF8FUU$f5*OeLaDNTa%en*&%fI9z}Pe7pr0XTl@`)l!=UWMD? zx6N*SgcmIE(iN;wDrExlW*{+|iBc;pxK*zoV*TT0G!;?6Bhpp)#!ZIm`<_8J_euC) zj6N=19E$_;ei)}bfNK)1x%Wle;d|^oD7|LGJ}+2;{nkT_i_LfJ+hfN4lH#*xyEo8F zap$Se)YH(r*%+$d9m1Mzwxo7Q9(>cKF;^jt^$41X{(ogm2MBPh1pA+lD6T!*1!OB?>Tf2m+xISZgvJ-+s2xjy&-y&dA^g< zL)b7~9rF15=UY~RoAXEveXKUXjutQ6ZcXHaab85uF?t zgKKp(aI9Yuw*S44!m!hD|K|`$t%zglpQZ~N&3lN1+ZE=N;|AL4yctytc}JMYXd1L` z92)L^$!Kr4eYdp)@?NgQLN@{YSI2Pbr8PKKF@rfXNywZ|sK9|> zcRbPcmge8x$_+KFV&wB(a6F%N5c{%%8sBNcHANL9ioav$=0%~W`$HzFO@U6gRv`8F z)H#>1r*xUYE8Ckx6%ebkgc~s*VXQ)5v$ZD|aVAei1ic~qU{7ibOr0SoFrStHp{nJ0 ztwNofm%51rwQF$I4mBuN=YvTKi}CVc7d>Aof>-t#kj}}w;G1(B*ficJv&{zx>Xxz@ zU%a5vRuYV*mASc5ZP;%v#nlf^!w1f*g^fABC~!H=pNSgTEq`8;RY~Q{#T+vrJja(p zIsKRMmwNZeVZ+HS!~!nSFT78)!lVK=l?H%a$qClpL=%-ZR+0C1(R}_b6dtF$vhmGx zY5BbM_Uw1_dwg&&r`H2PD-%;|i8Xn~tomQEjs4o{oM^1Ip zGj|Fx@Zfo@bJN6?1J6;!#E?6i=?q(w^zo8v2CVMlJ;+Za1j<40+1R31+S%02Y}jH7 zftN1TaD_3#>z?z;=p#Ka+9wl^mSlm&eIZ+`@r$lJwix}-ETo5zX+gZTDwe!=#>>Jj zjI(S#{jpAxe5oFd#XfoHnfQ#%c@j>)<}&nrU^;6&VVW$1i3pW)9JTw zFdsjyLH%Fmu<6ism~bSO9G$VAt1HkHNSX57wpsZwrSBlQt@o6TdvuRPP2~ITFcmdt zkKuO9N^{jo>rpF521EG!xV~x!F_O=~qcy7_Y^x``$?61YyL$^FPOPS9%yz=;Q2`kG zUJ*rRt;Ja{AF~fy8tAmpHb%Z+KBo3Al;ywXW#7r&vIM*=;($0mBsXXN-dSl zaU^d~>2rZPW4IfROEHO~AQhhOg{*H#9xqIA1!In)&P{* z8bNF-v}@!oNe4=3Ctq@|~c!5p$F`gb4c}?DIH^gp_atgm?>Gr`FblwjQdgenm4fMQA zyCQ1Y87sn2bImp0&p8?6%PWNG6CY5&8E?qMT_*TgyO}*5;K3ythM`v700wtwL&LUX z@O#T`65w2q&HUeBy-S(X4c*Gem%pQq-EL}t>A`c@(PE1lcnTH-DMiNJr2`yJfGyt@~6*ZidLhC&@q1HnR{Fa)Hi% zuY|XH*Dx!aU8riY6MeIYXV2YHhO45z^xn)x_@>|-oEQ-$f6tb)8X>E(+=IXh@d=qRAdi=LU%SlzmL*JxVt3k0Lf- zv+EOma_Sq>_}pl0jT?w(-VtEalO%_@I}fJa--!fo|@-KcJbbnBY#ih zwUbAoO(%zmF+WFwo3h|!QxyI!Ig2V;Mg0AdU}s_;e+JzL-PxmHckxR2G)aTciyniI z@07Tr|6an8MQPZOEyrg+P7#B1fiy~&cVGl$pn|#`_gq|)b4jj3nUp7X=bu(H`>flc zK{Jm@lKBmePIKVklW(LsUj{d6_@MW$4|LCv7+HA#1wRjI!*z$JayMNh(NkNU{#z?b z-R%D2g`?k~<>nyy7I_t}yzGD*kt+nx3eS^xtCb8dN+oBt#i-igL2^(@2nE}s>C%a7 za9HII&lTd&N5;}zye`yP%dt*R>p|q&b=upVj2q<2=}5L5IPEOJ4H5QO zw5t$_Sp<<+J_xlRlsHATnJ9K@2mDymN39#eS=b+e*WbJ&t8d9eqDPKh5}#Y?;B(Ex z&K>lox&hAE;sP3vlrVq7FtdDTG?r$o@OkyokZ!URbY6~S;!1052Xm)j^0!@hzo8KW z`eZ=DFqaj~n}APxR-ocmZ|IGV0l#iw{0gjKd%6KcG+KkjN;7t%P!e+*_?fS(4e9MW z0S=AXQ0^29*?k)!*0j5N=*Jw~pLzn_L-Xh!K2P^A*?|i@o{6WmvO(&l6Bik5L4MfZ z<9E&)oM=`H&pSIr0@XcuN5=fIYa1LS5*Dw#ECkLjro(V=-1zU{j} zm^lg5`o1NX%y+OpYqI2m#8_;}Z>Q2XJW$2M5A*3b+LRto$F5FcZt#16Cc#)7|0bBU z6y2u*PUC<}93yajlEBbZn$wBWt#7g$qlc^Aiw z9F*sRYTaqv4W|aQZi(a$X6j&D zo;SM2j^oyN&xb459y7BXy%_ri-qh}vGCW`UfqC*El$!75{fyJ&DPNZpw6*4tAj23c z^2Y=V)nYJW$sBq!q?RR56zIW>IAlMZq!JuQ-jx%Y+!%}cC-T7LaXVG#q~QF{NZ6&g z0^-g&@-wzU%xJxYKg<7sQO9%e5r%-Y&nfsPcuP7?XV5sa9eAlb1Xd}i!`Vk`$>a4B z1O{ydZ<^y!{Fye+9Q%Um&72Bv&)%o${C=XRe>R${68!P^DNZOc$CgY*XytQ@U;RZ; zt!^A<6q~}?QM%}7upMRlVp)xpWpwb+OB!!dh5cjWL2YR~hW+*cr#me|+ugUx(}F1S zb^RpF$>i{f;UJy_X+Tz=$s0|2vpA2wwl5&rPFI3mDRmS*A+h?!$ya3J7YxmEYKjQ|4SmG?G20^ z>xUhBhSaIa44wuh!A6abWSOC)fX(oROOwFB8~!Np9d`&tkE?YD3sd@GU8`T_{oghr`z z*npg?sQ<``ZO~e7XVbPFKywF%-`&Sesc)e$Ui zB(u$#73Tiv^J*Q`iL{c9#c3EMx&-9B`JF;p2-BEyn{HkbflFSz#>i{`Xu{T3{%-pR zlMQ#0YxCz*|J|S24##J>b}~P|a=Ut7YgH7x@8WM06Cp5lCn@wfpWaUh`=>aHeA*X zX7t}JCFYf$NMx$8@v;w={L@9xXldAI`I)?5H4$~ApVQ5cc+TCBIBpTYM{g!d;*06? z=%{=9kc7TKInnDFF8zv3o6ts&D8`d>vM$V5c)6R2| z=&!aGkDEGR+Txe+&@>JUMAzWMedqDXtJ@?-(+y_5cjtm$iITL5-4K%S45hmOeg|A( zBuWnxzlG}fqbms;b$e*g^)t-9oeHS(R0P9&1(;HlgBsS9oNEw+_&d2YD9M?dIXDOD zDLrn*q|>;gzMU@kZHSAU#4z&rFukNI!f7W(@Z8X?&~9l7b^PZ^StNjsd_OH@`+byO zy%|?#$$-mCarU)+0M>03V&J4{_~lP3s4%)vb;E({9=Z%W@&zP2LyklLlTJ6ku=F2AGTD{zDAn;H<7xEQ$SFG$H8FFY$|jJI(ov3*)h zixyiU=;;xmSuN3@%{z=v9)SJ(kD_4V3u0p6PV7Ro@nc*c*3Pdd2ffvJhd>;7jM4(p z@L{z7vjT5l0Nmx1$;x6);D`9i^{`8V-uW0wM zFVOfNQq!i+RA&RWb$W4$jXP-EjwPTyH(dBf{2`mX`3Ljj;3UrR@Oz#sZ-ncQ&%hdesut*t-|yB+7&;zn_r>GZVYmtwI-iojZbKQoQV4_Xzuxlhf`vSnb{gdJZo68?S**;H z3Eb6t$NAxYH1+zkf_tpoAgrtEfFGYHvTyU>W5B&`vaNVLMb}JX0kLjPx-(H99_sU6rkw#K=E6~U-~EUzsa%H% zu>o*pg9)xY$AP+RB)PY;oQyB;CaR$!^bXI;>Q8V%IX9s00jcEtRXM0`Rb$iLW?(Mg z|1=@r>A`cInB1q#X-Ae5NohMYa$1EAt>3VJa0gYrR*5g4Iid5zAJmJ_nV+e;g*U5` z(0^nPRdzf^r>uEIZ<=XR!n4eeH64JFf5DjOY-D%K_6FH>Z9I&0u7E!E9Qg0@KK#jh z>$==lawm1XXpDarPI@T^7CLcow=xYptL<^JbPt$0Il@Bw78qI3icxxPG=89$O|b;B zz$cz6$Q>s;_WwcsF(Noy@*z92JBSu1b<*zTvml_f5jQsf!mNwqxI-zku=!F1cQiN$ zKWCf6zv0W|HlLq)G*=QthlRw;bPoP9m`+p24V5kRdpL92RI3<@DkGCXg3cRmYJP&<84x_`)nINo8BZ|ja zVO}wR@7GF0?@xc&qty(3vCV~!w11*~dK1t!=NfZsR3th^ErXO`EttA24<9s6XT!JI z;{4hA+*E-OCOy-EcU28gwa*D}^;ssr5kGM=p7s(^13`R`_Z z0e%pFMXLp4xkjG1rQ>CaXI)!p_{(F^!!!F6ZY98h4U+IYI+gW5wtzg}9S6pNvW$7m z0g`LHfCT>eNi7z?#aZ=fq)*NfrvA$X-J`~M(_$Pd1t&u@-_g5$)Lv+#YAE<;_>h?( z9SJ(-V=*CfDeSlzihp?KfpFq0nlwC@`tjd`#;0Kzv%8hP_vKwQH{|i|M-O;pyAWzW z<)N+XWW4JB9VdT(i%tvb@yse6oaKIn*aW)5#jqQsX@$NZ=;UD{S|Eofx5Z=5MlEXf zuaxQYyNYjpk7E734oC?0(IrxW|1u%s3>13wHJ3m%Y_+?(Ry+m^6{y>orIs zc;3?LQ;(tFaxSy^coW8cdQRof3-HH`{lwA!ArAD5ar$}p&~>d8?DlZ5J9B3}O#J4G zLiy$FjxDm>i;5N4^01kdG=#yHjVUnleH^MZy5rDz8DjgXgZ7y%hrheL>EP2M9OfpG z%WcuHDKs5JYW`8}St_9KbA+nB`9!%+6)ts1h@8Q8^V!yCFMf#Mqx zjMS9i_+2a1juS)eUGXSil*e4*Gg@t7&S;*P3=aRL;KVFH>Kr!?UOT4HdB!HVB&LfD zzc0cg``tL(omR|pSqo~lzXDG#JB67KZZl0`Hspo-G+fGVCZbBqq2rGj4RxaAwIzSg zNIDJYwT?jR!aIzWd^p~ak>wsvxr{HJdux8bC&DhJlQWnDjfFWGd03( znl^$H&%DLwI)`l?8Z&rr?Lk^2sz)Bo%0olmC--D(Gp>v@q=l0OJ$ zm&R3Sz3&i64TQn+q;FKJdo}mr`!h_Lgt%v&709WJ2;#P>3DP@wpYRrWEY_}}<@!6( zd0Z>~+!M$g52t3q zL5rE7DAPsd%MOFi6HglQZVHwy3!|m?<;maQm&vk6ve4pwolf?i154MQB=YyZ*_9g? z65hxT-sKT^Ms+@ypMDjK3LUZKO+0ykArShSccFzGKvn%4R4Qi>Kk(l3ZKmoZc8wk^ z9QaPJZAif0`?pYfYX*7W696vS0XQjuAtcETbiCG}X!UDc61J6QPe{V3w@+|Ust^S~ z&Y{t?<#z6&?{H|HKg|45XLlemj}~v839o&Jm^)2oBqiVtITSLRPFFX8PK{Q!g>fMp zb~CWy{wTbzUxR<-JBi%2`{W)Yir;=cK$BJ(^q0^CZ?*`C?Hs(r)-!Rc_ zy3Cc>$L#db#ZXsDgj)7{upw3!<%|i+jV1TOTa7~`8b`6$h}U+tSKknX@&T0eKOB$zRcQXr}A8F zPrR#Pg~j~+`(NXDxa`_RoA|6uoaJp8QnP`*x8~8*t{fVqE5%5BPo-04i&6UrH^>Z= zF{qem!>(K>i9HX;f`rxyY*Mkos-NB{wjh$$?Q@1y`b4mddj(p$w`f1H5uEcb1FhEp ztD}54`^WPzRQUs2z2QE+qwt$1bjG70zbA^ciian!J>am_KlBL7MytALyeRwxg6&6O z;&NGdY4@F)xPQaj9VK{#XmQbYPPjvmK*awFp>B4?~Vas$m zPWj&wR8se$pE+Al*c(JH?KcK(v$IrDY$eJ4eIN6WNaCdCr}WHLDW=(SJXbuCQjNcZ zFjD2g?farct|j)9t5;v)gg7UXC11mIB{brlURBKdb&?Kj>4NF^PlCu8efW0vGAHfa zL`SZQa>I$IP|RGM^X2Dw5u27m%9uaQ(1u$ONgeUStsJs6a5T47(+Gnj!(e@7Crlbr z<*?ZUHU9YHxb3rWD9;&Zi8a#E8;KxJ(q3>rTUMYNA%+JQbYYa~TSV$3uxxz~-!dn_ zmQUaDKc!#zuG$-w5^tkf!U4|i%yjs9<^Vb4sL7qn_=*uBW-L50V;8mwNh&GFg9fIs z_E8L|z0Sk&`)#Si%p`1W-+&IQ{o#VsAELH*GK9vQCl}63;80u_HLg-7Ypxm5(Q}m0 zxkUmGj7qXQdRUPo<0qk1^+#rO-Y>e_qK^3yr9_YUOW@p}v*`uZCAh>R%Wm!5P`exh zby8CMoQCqO9o1{@RD5g*jeb^*Tl&rsEv56A6I;w4e&T}Jb;fvZ_yRuOs>KOyW(wTX z2jPA@-yc3`%T0G_r(Ne(!LG1+w6W_!w{8QzAFK#+A&WqLZ5E7+$fT1F>*L3Nb5Vb8 zE1ju*9Noq#!Rd}?=yW9t``=!~wl9ZC{kUK-sqH7v>@TwYr<&{>Zkz`74U`@YR@GP69X9g+;}==Mjmlovyz!r#6=uCecaKKSl|-* zyS#UI1h4FP!G0)J#xvi`=$32y>6)vfIma0ixZ+KZ-QM%vu%SUnB|eJ7oR}1Ne)}3- zB$Yr{?45v0ZQ@k+ksMdTx}`Wu%In`0yGP zx5u)6HL~cUFT!d1NT5YtGBNlR59?{#xgHR#LM5jqsp3m8qyGys8^AUQ~jj$1wWJJ^ZKcOzL7j@H=-?<1IeQY znFqlfv3h?AQ@)4O3j<2*xdw4)X7gAz<=5=0&qa1+2Xg7_FaP1qgHti{_%PXU>>jAk z%Ap;VP3Zsh4mnx41vA_XAwuf{vtD?f?0sj1GCOrR*~70v>~sY5_%8*^_yAC=^B@%& zU)gnr55SUVJ@`)kM2@|t)K}dYQctOX%AW5u|N3-TVl|Q8PP;~pV=GYd=L5X>E*$i> z&43plc{fd7EG~ShhTF@csLQ9Tc*yk#HmuC1S0?8`>B2AUmlLXFlO*4rms|`=>6@T! zAQAIhGVphgJ;JtZ;+Xjt%wF(bnTQ1XVnrY^d}SmoKJbQ78XrJ6Crv@!&1SH)bS_=k z;7%QF?&8qm5h9_R3qE%WVf(yl+;{d6Iz)BCMf0o7)~_mp!IN^}{&^z&$ErZLM*=C| zrvZkP|6Xr2A}VrU$-Y`mbe%XzpZ1)^h=f^ibLDz;DL=+%$Je0BdvDH{-=lmgKg*0h z@PcM|eq$~&^QlOdKl#r&6%8!;`(>~ygb$piYuCDw(UCf^YLx}?Ib=Xr`29hHg_7i1 zTrt`x*r3d?HeG9YRH5u6+F6$_fz zqSd%IesA*%;|rrf#=?`nA3v28cGl7UYZ`RNvL)!d;vs9q=dQ21zXPAi)#T7eW9(Pd z;)2HHaGhBfaD%cEKkx3K9j6p9d%y*rpSe%}`#PCBv`Pw3h8!U7zf;MiFMX^^T{^DL zAC3JV!l?U>D0(<~3^%~2mIj^3M6Re#nGg?3n_q!gwy@HdA8W9s?=Q{bBwTB>h`hTqa`% zx1UEr+rLWKZMcZk&!{D7=kLIn$HjJc4lc&Q+d1UTe@W0|smZLF^qy9_orjHMcwf)w z`}Dd)5DpYBM_cnuR`T*hkY3eKZkoEY(LE_p(maD2E*Hc1Ro|I&Iw#0xxlybubCx8% zuz>v2LvT6CnC$bC;pgCYVNdfyvTSf8errgh%OkBYJ=6+amwhHTlFC>$>E(1nMkHD+ zKL$nmnox2m5oYmRovfMt#B#Ab%pc-CZ_d>7;%(TA#DCv zOl#vi)RTqqKD!pjvXxXl177qxjmk-eEIW}S@%zm@6m z>x?bz-tm;}7ukeX9n)a}-=9uDdY`P|T%h%gCFem7L)O=Lp1CQ_Eq$GVbDvEovFqBX zQ|W6OD6dMX*J3zcbDn-NPJ*X>B~+GoLP+1UC7Qc?N%9dp5~<)xHtkAbp(L(w=dRLCAp=dLe+!?~m3>dHdWr9T#?<)@KbAMX)nQ3bZ_ zwWV{eod@ehZ%B{lX(q8*jc)w83`!?a5G}Vx*Px^5nWMsWahEwqcb*w*6UQ#ea)UEF z_u-W^G4x}7$g|hKm_(An{B$1$&*ax~k`6KS8J`V`IxWKONdJa2FFeNAOdTqUTXP+k zTkxA%7Ht-LgFhCH=DsIdgVlb%i|`{2FBu)7a?L{Oz3DD$+h4-ENnMOurZc9RRZ`_` z5n!N`0o?XJqF}?Hzo%BxnMp3p#T%h?@x>h6dT|US{@X+Et9yW`?nw+V^GBT@7wIG6 z83=hbo^_C1fsOg9*rGaFpkjZHe7kiVwuP)CLj6_zd^HrjWYa*gLS5i6V=|dPLq<^W zXfcF%9fzLhTk+1CO*nn(T5vic%IXbIr;0vyDEjyly9Fok-TcKcZ&8ljv44tK_COm$ z|Fn=rmrs#BE@SX%#dWM{8cn?IJLqDeI9|Tthvs8S;n0&2xK;I?bSnsiPnv+tYd?%> zIsAP8kStvLc8^(>$#7qOMBv8`Yi#|t9yLcOPOIOJQnv~Tw}+uq1C7aPpFhYrhT#2o z)49@<&SX>T4EXTo6ua`eGJfOds*rYq#?+gTtSWoqK-^_=Jg${Qf08AkHkY9(w}hN* zkpr9aOCVu_3%4%EhATPz5?_3sj4?Aa?93BplfQ{IRA24`y;RbUcl6pw$tnXp`jp>` zR>#1Tfe2`Ms|~jX8_DVuRpeSm8tC>3$jX0XfxP_%uS`^+S}PFL9*xj>v=`saB#!pR@%(`s@L4Vt*?l(Iw z^Za)@rnDYILp+(V4GSRNVg{(BTk{UeKSWisit0(X5O=Q_b^~*YCe_5i*<=s0KKd+? zJ)}rt7PiuE6>Bt;KZ?R#GI(1mo_wBA$86J0X3OVo!{;l#>6p>U1a}ExXTR*;MLBFT0f9B(t`qgKPgoR$s%I3y9!7 zE<5Tm*SLaxyx{`oITLVGDZ#O)L~y2b5%$U)26$7?GwCew$ELB|To(_XzkD9Hi#gMu zK7WM!-xd-hZE13;!;iH3c)=c~M)|!mVDi0`gk%gLWu<7s zs@b3w)Mi5hW$PWVS0od~Kmq*BB(clG9*XPj82QXxoIb+uwl2v6 z7jPXnyoul)T2b&`N7l~&SP8TCL@?J|`50HF&gGsAv_kdWRdiy`1HQ9gPjFYIz{G)<1g{-BfDD#SCyzS9}%9H5>)5 zU0XrFG69-0)A3P89NoG&8kt0%sd23r9hS&(?ZKnDv+f<3bdm3DEVrke_8MF~E0apN zxxv3KDXw?^V){Jv9DBWfCI2oeCINT9;nKq?c-s3S7Q70?1qmDIDc^YP)Zra60jktM zW-c!DDtL83;oTjr_lD`Ra2NP3oJ&4#Nrs+16efOo3ZV`H>^|eo zKVPTP_v9)`8y>McwBQ7k&G`UL%M78;OBIW<^M@!DS z_b!e#I{@0A2Jq^8I6m;z;x-;I=2Fx7u9T=hR;8#A1@khh*T?(I)y0GXE!T)_eKWmg zmPfuOq{6O%L=wUGQEIl8QPQ`D@xA+9sQD}vkGY62l3Nt8OO0p1BoPMZwByvN5AaCc zJbJmOk~;q<508KGJdooDiQUr#6lSS(}-pDlQ)9h&BzC=s)6X-T?7H4`nblF<(ewpxHZG)ahAOm`8~z~vYdCqs2TV1^g~~)J1I{?pB!Rq z^mg%C#t-E8xpUyK`~VsE`Y;@r(1pA7lpy*<2$NQ7i6yI#+C}j`nirbggo~X(Q|(&V z#$Wna&F_~Y0?*LjOR~wzqJ9Dr!BFAg!TgLnLDCJMK-K(wbX%gsD$Wi;r?aNq>~G_^ zbBm8-tY!w%!fqN|Cj&=^uY=j{b~5N-jpiH9vCmB25e+?8Ac?JL(mk6KKcvNlr3=w{ z^-NS)CJUY_5?I5;;+NnZoX4sSj7s29*iy)|*AC7n2Z!YF$j$F46(}yy&pt=r2CH-T zyWXMy{ROOFC_`>Pe2L=g=HS>}&**y)rzOol*)9J3InF(oJS`a}JNI@lDVu)a-7+uM zfL#nrEcyBJydSi;#8xJ$e3K1${P?1$s!YMkZMS$MI?3p!VQh4~TuxgcsB z@w;k@JAA59yd)E1TXxegi*;G|#Z6$N4v3XmHhK^3GMNW$!1+SjXx6}zv{-KSPU zw%2@AULS`SybiG~?%UCKrV98^$Rg1spMHHgjd5EjDK%Y=-V@Ol75RU^ZVSc4MMR{e zME=*$I1xrg{Gi+>}=FAjX^?&@CllO|uHI^q84fm-} z;C`C4)SQc7ItrUD6kx?WT@q1YM4xT|Y=15bSqhFc!ZVlnRnI2oKFFFXdSiLQZ<6W$ zlt%>k-d6~n-+ke65`A1f@dO6q%CiFfmoOU-9a1Sm&5Nn&ww7NNx zO&8~vCeJU&BgygH_pQC~RdjYqU3tnrES|$c)L!Fnc*s`UAF_We&9Sw&94Jlfl|;5H z3))^H%OoWoH>#e3=kHWNK{ZpMz?oc@E3RYqm0Ve^m+7EetOCmI8?h%%sA(2ZXZWZzRz2o6sMiYJ8Wlfbfr`Y(O2t0c|RQR%axUh7>F2Q?K zKCF1ni~bH}ES5ebbWXZJqq?f#$M{zw=kX5QDh|aNTE>E}K?VDBzv{K zrPNy!h_2CZ;O{p{ga>J`;eaX`eoqDaRtfSOdX#JKdBphILUNt)nDP@2;@Aw)DK)l$ zUc2fDdPP3e&W+&>=WB^|_~|$!X#wYOB*uRE+9AR|Zw*K|xDE3)!>Df5Bm6IL2=;do zdF747shgna?&}SMXhY^a&0amSRdT?ZXjnTUChQw5bKf& zUl)8~j*r)hGeeo;yh;phUWBb8!y8}nbssi4N$RCfLG4K@}|vjo|D%)q$? zWHz3Jz!VFiV^ksyw4FkKY96wIMgC0X*?qRlH$?Ooegq4ZD>y^4-Qf(+prX+RJDz`I z;f@NZ#QnnUTL!_>vNQ^+YGj*mBDzed=Pz47CigKrG4fXbibA~u^z-8_xP0FnKV9!n z?;3gLv@4C)>J6a{8(y&)ihtSCv@D2pTS{ZY&T*%H-e9s;E~q;p9scnV`0%eLo@tZA zpfLcuPEDZ?UkYL07$wK7kV){w+zao@e`NO@?xK@RE=DSP(Br9HDJLm2OIB3IIdSD(%g}k5Jzn}Pg$e5vgo}@Pdbz(J>XmH7l=8ha z&EFlDs&~TskN0TOr)qrq?LS(l*$#^%K|H;J(eLHK@%1!}d$GM~z5 z+SvD;VbfH4J3RruXoyT$%VYdV6Dg)SbSX5{+OzjpC54Mu{ZT`GGHhD85GJ(eLf!Z` zP;U5`J{a1O&qq5UH-7+0eOih)#P7;nY7QD_o@QRZhY2o$M=9C$2h-nU58ssrG7Gy7 z*1AZbIa>}_sOlZV;a~UjN#;|_A009fu4Uh$s`@0+X=#e*66RsG#}YxOZ48cF7fN;C zuYkFfF}(BA7yfK|OOlNuZ{bxi#f}kYRgNo>rs-`y)ghSH&sD+8zG3uu*#h$Q4nvoJ zS!CofTBw;di{)+Khh8`CU;&m=m&ID)*77~VoI^`0A|*F-6))fN*CgCn`~+3uXq?EP zHdDnDvnO-Q7FMv=ky668HE-!uc?b95K{b=SJcG`h@)M38Nd{@dAvp3_2xN5JqMHsD z!dHa^NW8Ac&3-o=%m12+*Tq4S`f!FWCu$3?zqqsFy7#bba}ARfpYvJrM}__GhYPhu z=PPDS5P3zl1tcRn43b6faoG1Mh;3$gTFfASh&;qksIaQ&++09=m&(&>2?h91%mq%_ zvaVv@l?^y#Vi_gOxQ16RwS(QoNSe%7@$CT<(Ml$ZX^Jl5tyvZ@^jIZyj-QK4(aMlE zCY{CvEyXpq*|6Z&33y~?i%JK^3A2?33Q`|3>9^E>OgDTn+dI-2w-`Tw-`iu^woitz zIBPat-}`|SDk8YjbcAjvz4CP*M&pIZ-P}QcE!?cH2GZSRzqe!~rCPSJcgpc>$f9r@ zVf%QzgH>T!*wUjj0TJT+Mg0&}WXjhon$JOX$c74#TXZ619rC9`&nZLFd_MJhZcy>c@xE`Su)E-()YSZN9`7&r%ip zqEl#z$A4^-bTrA&pD5(Ln$D~*grP&+5g1_}hLU4n!w1!WV7c=S{hlXg#!oib+bYs}o|sGv*LVBGC+ z12nHG2u7>h;H^O>^*vIc)N-nDwXeX7*W=;%yL|R;%>i-@QRC%{5AerE9&)%bNtZYu zJ0UG%BX9Ub8H4`FV9q!V_^zl=v%WrO(>CrSqfR^E4q36o_HVgrdo4P0xIY`w-yH)= zmhk>AV`y;OAPV{KKPubX%9e(g(T)p|WV%$2)XS3Ls9XVmY@7^Pr_VyOHzlN~m5e@P z?!ysKr@WRu{G6CM)VUG`Ba@%p42=2L$%031fV> z2)!#3>GD@eOjnm>>gDID1%k6H&)hcW1f1E=r>Fo zC%f5GrwfD6%dfCDg%?zy5zAJeuLH+_SLpt=l`QFa0>!ACQ^&_7*e-LJt(rNSLi_u& zUNs4dtP=U}Q$K*s&q7N7Z7sN1CZK9|3@`QHPJYBACH8YcwKxNK1%YK>vTGBDePryps;0Ey)T(z>+&$#k&4+%&zmQQ z@b7LNrPFCQSy<;jdU`<;Ex!MUcAulTlCPOCK+z6L&e~v?vi(n2@*#ZZno>yJCuXpV`wJ!x zzgh2tT=p{Fo)unm#p`yfVbc0&;hD=?c6MMI_pYgpdHN>qg?dacwgGBbLp%rJ;W=zMob$|-~Jj&d&Tad~7Oo!D~&7JpP>E1zq#Wo83#O4$OOex`}~%^QZV^dpeg-G*Tz^LFHilk8Qg zBla!~VuO}Upn+c2JZv=M$7iYtJHvqdqIivrDmI1iM1SWmg?)aW%wxD3JPmlm8*K@vw+S1tsl*CxXBI+cq8xY)F{0zTn<-~z z3>g(VL7le_SxsoCa;I%nObNK$QpIuln;v#eB@^;*Ht{cq?{-K_4x|Y;Z_^Ih5yGfN znq*d`D@^y$5^CpY2@6Z4g}t!?`_JVs_tSDNPEozjhi*N9vt%ZtvIyY_t?wysEoz9JaQ%UGWR(fhla_%b3VfL%NWje40y(7e?JTtx@tLa-$De%Ujty^vUj|Xo z|Lt9px?2sePTyrCM{BZezAfCegpSW58~&rcX%Y$ofyF>RmQ=f22beRVIrJXiK36i zqu^i7OLljH$nzR%hy3M(tofcLCdr*+=LdSy-uT6F*9@t&6Y750c{V8J}Pu zI%^s1-hrcj8$j-f*<@q$k!`QP1cMJ`&;ZR@v~pZ0JGRdnHRjsGz9~6;dAuuD4Y|Oy zC5K|gk!O(HaSF>N!z+4c%oUaxtP@_IE}}xsIPQ9t4Z`V0ww+f3pTq01B`p=6EpMW% zqnWIF{%o*I^TAnX0+~smF5xm)f`%vj%{Y-Km$d+b9nZmkHVQC)s0joOvY`NxgEw#b zC7Sr%5T)K2<9rW8^fsD{pU;Y{&_7sRj$fFKZ<)3qTxpLfSG566BmQEGk z18K6mB89#iO{Z3`2S>M3_*kz2Z;Cc>hvNtHRtM`vU*|w3le3p=ZjXgwuhg;3R!!j3 z_mWBSFiO#u24(q3JfwY+lR7pWi=4;eez_@-r#pn-v7?0+R(xY&HyWU-t`$d^a?Hu; z5c!$(qnyH{)V!fbyw0`=d+KeeIYvBl^RScuBvCFvrXyZpRRcnneTF^}rDGcNKr;oNy>y z7YBxEe^`ZsDf(`g68r)cs8;R+>wPhXW@qlFS$79v_ALoK6JY~;zbj#vodqs@y^ISE zdQL;r!>Bm!6u9&w4Bq?)rRBstqJ%Z=6#vJzBqgC`!UKmxpZ0T0lSfkL*$C)%>E=|r z#6D2A5=zXN2(ue*(vNlG=kOy5+uuy3pY3BRHdMsor=A4<$-+k7SKl8l4pZP~ZLq`w z=Q(W2#sd|LgEHCF0p8_or?_{k8;ECD*$cN7gK61A#PVZ1Si#{>=%~GeLY%L224TmU z*1AdbAjpyi&-nlgMl7Vfae)pus!y@_h!ya-AWeuk*9^0SbXdPIiM{pM$HJ!nhYKea za??*{R_F+|xF{)6?3WucN6%Sk^)ZRA#BLO{W52Qwdld0!_s@#;nO3B+O-?vAO6zLr z8gtRP^orF#{6PU|#960r;P-^w1Y<6kN&Gni@zqaQMRp%e`;bRFJpIWwwG~$COk{)h zSqQO8tI>>a2mW~^=AAo?f9;3hCP@uPIWfnAb&ttdPM{eoGwFz!Cn>0Kpdn{Fz&m#( zYzu0qiAG}2x6F*jI}H|sZcN9u21BS=(%PXf*c^2%eQ3RR8yyP`L&bqkwC>w78hKrn zvOk5B@2WUqns}xzofFQ6PP@Y7KCfqomcFNgv<&Ji4#p8x$N0&AW}@cU0m62jF~ajV zOK6$z3c=8;A35DkVK%Qbz=j(`zm{(1wUt^xRihtPJSt-gB!b}Qk(C&I_!7-K7)lS! zlc6?gGlUBv*mLI`{yf_S_`HEFSrTD?alDC8v#W@C_eRn1v~RTa%u>$cU>&$zG{aYw zTDWiFaeVqU0tT4rQR9#(dKo>N^vkyj?oG#`zxWJ{lNgDkX2{cjQU8(W5mVHxy$H9> zuELk$KVgDRc}4%1ia7VKJPnABVxu1Czf zK$Xsm`?|JA36wPQoyb;ME~wot!Dn~eQS~cwH=S)+NLM!GPfq8{3~ZR7n#LQZy@EFj zmXL0XC8gf4WThq}E5?t~gP}XTxV@J<@t><2>Z~&4%g=?P|DFnN^eAQFbpQySxoWs$ zq6zxmeae1xY@os~we(qktss3}4$qXn;b)tjXXZ&w6@Tr=PxP4-Ko zxG4kZbaw{R>^0#!=9}Su+eBFYxD1qg#Z26sv3Raao8`oZ8xm~Gle z;sC0kH6)Rmf)G!LHLXHTsjG@skKoL0U&_`hV`mks$nuCCwv>CJ+wn{&^v@tIBaY6# z`2;1GBY5TD@2tzB3ub=R$H25+ZdHyNs%dY-=4%H8y%|TCW~u0q$p6Q?_^Pu9ONR-d zG(?zHC@rXtRCQd~sw_0TnaZYo+eCl#&rrdSMsi(thxPk&8J$GNZS;{y_%@S4uVM^5 zQ@cUKESCt|#ouk6VaS|9ezA*zHbU>|i^AigYC2lC5A^K{7~d1lp9@ZcpvC4ivyI4U z!FFu*8ct{Sec~oHSfJ`MCqd<|6!E(+=!Nl&t*d{ZQF5T^5zpSJMg)B*7=s;zm zTV0jhGG=jKKQ+>)e(Uipn+;puZRhVgwNd?#oqYRxc})6zoWFlt$8l!xC#EOajM{uA zj8rkjJ70ElX?ttAAh|;{=;u-}f4CnVKFEqJ>U-Sveur50KarW9)B#gpnJ{A)PpV&H z1X;IEvx$RmvQyPM_%|sM2Y*zdowx5}RGldkLHX>!lrJ=8j~eB?y~;|t3hI!Z&s8ZO zVXHO|#0gWL!3|5s9&Y;s<@L>AIKk=* z7Z)YVN%g#9p5r&5p7~1Vwy>7=td7Jth0SD8_mXYeeiXx8FXQ{Q3puIYZCsbU99(;9 zM*eE?^sj$2O;S0HXFjUpxsou@>n&h&zlPz^PshPlhGW}iuM$o#R}prdf5uhc8cXeO zYbdcRf)1Q}0Tun6gahS=!EVc4er)`0hnUg%_`Pv3sk@7uH<^1dtuvHDw|{}^Btv*t zzn^xEzacXDjo`j#4Y!i%2+wm0VY{6V_{=Q^*|!|&)=y^ZiZWUEn5%HP#0{R08jnM5 zuCm}C$A#FG%jC8yAB;usK!A`zP8x&Qw-rWU?RXw~%@gSQuvf5Woi`Za0`|_lA30vI zr6>QU!P8wkAoxu^7j?w~)r(y5=(}UEKF*W9PxhgW%!Tb+R+ol<)*@^m?(BkWQL8pI+Uax!dcI~J{Zz5 zx1up(H?K4E0AvplxhH9g;5F+Xbx(WG{I+}HYP;zav{eS}rJpg5(-%fZm|RiFc*e&> zWw6TBqpaLki6U-w(oS<-eC0L)-}&jXh&AF~r}zYCxc33K>30H{s!KuL$5kL-Gns|@ zC9$zFZ7^zKD(G04z^0c~@XbqPwVXVJ8NOFQ(s?R8-*Sjw?qQA#)9$dwlbJ%M(mcV& z#11?+FA=?t$m&+w^(Wp!~ zbjys^>@Fjd0xRO;BeC0SIH$AiCag>8Cp?grMG18S<`%aJQ|GIpakD&|t5Zh77mktp z5i4QO@m~698V<>2sc`e)VO(@V6N_wQg@$i!v}cm9AfZ*lf`{F#C?h4T4f5pkVs#vc zeG<6a22yye|1E5-E}>!Zsq~8bj;F_*#tJ-$d_g1G43*=qd@kX2<~?LL3rZ=$4=}<= z z>Iesztc3-cEHZ_z8%ZQ_*@{NpILM^Ex6m&2|0=pdHE_XjP4a9q5zcJa6waMm#1224 z&jw!{V87P!09*M+S-5&Kj3jS=6@6xw^kDm7+;C?Q&QyqD3&pvkxGjH}=jE-a?d%OR z)7Eo^>rI7_w__f~@3R^CT_otr_ zsc$J*-PlX|aVJT8ogvD0YT?5T$Wko!I9TN=!)5PAUVD%ao_Q{2BhJ>bH5PyYXM9;z zo*Ml;kp(TTsgN*KiR`~`f?%6J>_T7zd~M99L&q=k>L=ucJ2&5oGiNfwpU7~U=AKE` z-w(0t-^z%(L*d>9S7;s3$oK9!%x4|p;OAjA;mzP}oP#R|`isZX5OfFEhy+-+OHL3+ zsF3~HgbsC0m|c6AJyKm=w4{5E2D}a6HK0U22@kOv(FLxNd8MO zGg*XUzF5o~zDXA5dMXI-_6!g{Oj9LKwO9P^xE`i$n#_#m*wD&3ah#EjG)c!Lhz#XX z;vTy{t-14sbDpM*$0s_0&ZlGe`SAw+LP~l0(WDfLc-_t(M?B(d)^5Y=bKl_gG<7=R ztd8E*JIQC%bH+ZH9Nq%Q_Yrs?NfkG|O=4-%m-&8a;V{Ir zl6(6{f*z_>Qp7HGK~w%Nd$n~tJvamu`&_JHe{OM5T4w=v4XQN0Dx3BGKFz!@{$z*y zM8@sFNl>I}gd1fRz=lz(xG}Ai<=p-NFUoWr}AMU-s2@SdX34B6_(y6m&A-_V6Om`iE zq3Q3T_v#W9hSssXzZ)QCP$jl5JBn(K`(fF|7))F2NLA;4L+_hRA<*?3#c91IkB@OQ z|6?q?EEvfgsDUD4PtoSaN|Ks6o*gt*K^?bR)~G*Q&^`<#b=-z4961VY181>Hmpb+{ zI|D}O$&j=NZ0!F`o8C8vD{nB(b-aOa9Mjk|V-tgDk*l8+^vj44Ei zXHnSHxS8K+F&-8-U4l&GG}bv%lIh1*vIo0VFf!;QR;5+23kBmzeRw3Nu{)o_4LaED z23dzaaZ2E6ewv<)d&LjhdLGiYTML%^_rm;!8dfjX4^KSX1uNa{xtP??;L#-|_S;s$ zTO}vHW$YsIy}GblyhGp1m#hK&RSMlx5(H z5nzo^eda-|xjCwt^Q=)xS@3x_5mLnS%;I`Qu>CU~J$;WuRjdSPgqc(Me6gpm@QME! zAqTHU@8c%t#)J9$PLO-F9{Z>13(w3oSzXOwJbi5vEZP&nmfXr?h0iYX1#tmn68D}f zY>9{VSQM<*%LrqR?ibys(lAnL4NWVUM@6mg;flTs<^8%wU$ZBW>}PkrsqcwHTGT2C zYcFLjvv!fu&06p@n1+^Z>0JJUY?kd@M1OMck*S6n4g3@cnU!L_Bt;E_Lgh(McN)H( zY|jq&B(UG%W`ats41|1+i*R$e@vksOHB6ROj!~JO!>5Qdeh2_ zadj6w47ovzy(ZJ)-7XOIAqeZ9$YWw+Arj>$EVxC zQ9T3HpQSU~F{>Egpo(+ndWx**50DW&o>ba9F`~8>Uj$yF&pjUO2U+1i_hn>RlupTi zCh|A>715^w=J4_dp>>Td#8RaSHFZb`fC=iFA>8l z&-es8{#J2a>m^Ymp;A-e*-b~&^FKR~6&l)`tN;tSauZDf; z_Xgg2%VD%iotQ`T1N7S7A1JtE8O&vGMHH~$E;`7i=BtJY+=vFbKrMVhh&ORK>kii z!ODF)#?f{(QGQCrXAgq1`bLnj`AybKb%dU7N1;b?6U{SC5VQCe{DRDUNLDJMwWGe! z@`Fw|d2^)T{>M{z^KlmC@lse(ltn)t{GhC*Y3#bg4p=r@n@&gnWap1N!6KRA!bSc% z^f#Qwl>c*~c*}#Va6|+Q={^jpE#Vjy=*qcoO2p~@{cyBNF1xYpJR6+vLb4am3c3UG zSX`bVh7}DJOic^9b)}JX&sCe&tnq+*rn#(W!D5`?6N^O`weZ<_ADo}%g^RSsjNjw4 zv_-s*kDEQD(ci7u)Q8QSRJ{zaf8zyxLrW_9c^6(k6=wu>QUzPriNeT(2f2oRDKz4v zIxbz4MZ5f7)3W#Gf=yOF_AZ}FZZCGz{GDgvkJ>JJ&^`t;N6&-Tm0oP@Tm-j;9?U3g zEHvC!r8<)h-2Aw!OkwC$A@t{bK_)}=6Cax`+_+ju)nD5{b}mFJ)Kvc3+q2w~9o1NL4A~g@ zQrK2Ckyr28#?v<3Fn@ih2-?QCG*>YT{d;WCTrrpU zD*%E|)qwQYY)I!?Sm1k+p)phR>YULMYMk{ETRL&|(L(%j?>B_@-oc8UwqPUHMc-Rb z5VE|wX}W7K(;BNR{PVMbs?{P#{NfA>S<=deI1I$Qi5>9A>@jQm4rqDO1dPnxFm|<)wI^BUbCBKILg}T(XX9Otxs)ki(^ZEZ&t_W?sw4h$fF)dvKT9UM$*6f^5Mngqr z@&Qw^7w1XoNOl903d8%(60h1Nxp>C-j z)iur#_8M-YHgC~gGCV_gaV!VEpZvvASD6U0q3Q5>R{_piD%L~Bq>B7?cZ%y-&Zkx# ztvIjPMZ2X}ay!3&grw#GOcub4hR@1l9pqXpC3hdE1w+YSXPd2B`b z`^)`aCZLCqi-$72neyMSFtpf@{FP5I-_HN%7dv1qIiTdVM7S|{0G>6~0;78uq5u5@ zoVjWcxu@KRCkx)guLc_!)mlcMg0Hdk(A}ah`5^OJ*#S}gTv@cH0WO{40M$-UnbF}* z!pPR=-v^snLjZxr<&&72J)Jbuzd)7kSac~5V{>cQQljqxwxA-2 zJ5oJ{v^sO><;Rs|5G^8Ro?gbSi}i6~b2?_)*x=v3@!0CxzzU=t@XpA1{?0%ZNSHSc zyZ=Oy!M&@ZucIGE#<}4W%_#DSibqM=9~EWE69D&W3eOJx5_#tD;L)WB%HWRj>%$IG zZu)lcc5jDb=w{P3yu_I^DTiPOPUIvz5dSBVjq@8RbiFMgP2*yw{PSf+ijOLk`wXGP zwJH4OXVaL~YYDV5T!Kc`Q^dcK0bhIK+0#F-?Qbrzfva{MFim_vM-R)g^P(Cl*}R5K z27l%5t&UEM~e6Y2iEdbDfMq!%)y$-wdub4iyGew7UqUbx5NOQk(9 z$fO-i-lt)%81cKkdlEDl-N1blBzeKk0i(q_;E|UXdAnC3>}$9_YCccrrZ?4!-|=Ov ztp5jr+RpITSdS&GAA^lM1u`KP(lza{YuMq*h;5LugR2~{(~DQWfr*7~Z66kfmJr!DlM8AEl9A~!|^K@ItK>i$Gxo{8nA^E5M zO>t&^gZePcsK0^#q7N{o7tZupay-*{y@sqzLfF*ci!er72Os~<75S2BY$WGTo!5>q zsiSRd;G%qdGRP4}7+X@d%r7Wgy&d~NWL<2$3UzT0U`AX9Uo5(If7aPE`{6n`T462~ zS-K1P)loud_7O-kyica#>v+d=?(D{N(bwj4j4!VmO1WOgs68Mb4QDGLq@;oOZ6n6} zhJo+beDza-&`uTu@gSoiPx@{I(saX!xsN)?yR54 z+31XbDM{mTiVULb?^)$)-bNG|CFb|Pju-u5{`|1tLvY?WD`Cmw3Bqpqc;={C3fEJL zaBs6U9xSqPZ<9wN1r6q?m(%sDT~*Q!*|}Rsr9iHE=V)Oy|IfaxNfkp zXSo|TTHX*{or!F^$okp6Ra2Ov`` zNiK;yut&xcbF1SjEaVP|YpM>zO;4E57fm7L_8a`%o!bX;+e)rnk! z{zv->YE!?1<6eEX>$*GUJWarhtA>-jvVkCd%Mt#D*`TzIBD=HUJ2Lc$xNUp56^>O6e{GU+0o?h{RH{z z3c{x_GrA->4f0hMP;c}>meV(q{4-au8n>t1$VvU6peBqIw%gEVw?42<`_0di$YrV< z#<3ahNoYVUt&+amn{4qStrNir#fcc%fkdM1P;nhQ<%5SeUdM>lO^= z0wg%La*!<_x%L7+U&A25IUFDPXrjrs0xaL?%pg@3?VN|>{EC6(Wm15#*0F-h?in=Q zcbY@fsU-63xksgk)47!`cUj)YbX@yY1O3=|$SBJLeb&SW1{CpE4K#3e)Nl66r+_}` zKEXAbDNyugHM#}dXKtx`$+>Wh(4GF8&2E}T3X-`nc=a}DY8=k?c}@nCshMzecRB{C z#eAov6t=4jqrh#i*})wptg%uVhq}$LxY*qciLc*5^c({WJbg%H-!Gui{B+0+ zQ6}IZz=Psmc23k(?E8BbvT z?DpA8?cp!`=g%m8En>5+elwV29WMhHIm}8`DK#syv3C%toZvw zl7Au2eO#^}!DbppzWK{M>XV_%xDBqH%4V(x{-E~b7q{bzB3>PWylt%L(GBpX_}7=1 z_R&Su=IKGjC&e10?)u$g~rCt?A0b?uU+K;c|@M_j@R=8F3sYX?s)hJ~fu; z7tF`jdeF4*$~b(0470lOnoCi?fHzK!rKOLp;iktA_;tYsr!~!G^FwxG*9tM;d2$WO z$jS;6z4D3sjZFKVI`8Q_4~&l+vnAaw92@zS-OpbLySAucx33&}BsV}o$8daKG6iH_ zNC}mjO(|2dl|6j&lkM*(a)a8R!a%tVq|}!Qm6O69GP@CmyyZA6&z&^(N+pXljE10+ zRiGErO**$qY0a+y(r;J6<3sFV zkCBJ4=x;c$sjkg0bZ+L_`~q#6!Cjh_bBTgUntK_$9`DaKa2)-8Bu<-Qg0`>Bxe>>k z=)_+=thDdJ?k&OSyJG>YmEv&vLn}<&T!OY-1m3kQc!kKQxR#gGa}j;TWCKSTHjf?kta_x9%3`G+-{i zX}bt#L$6YBb)v&zX$2a4Yc;5x-oxsuw0YT0O8DniAC;9yKq6Wr3_Y*>QGg-n3~FXU;L&j*;OdyZv2 zIEjDz`-A6|G@O&qSmVg8;M5jC^E`I4`G=-)JFTK|T-<0*i4Q8a(*WQkTVMRf7{YxS~-_nG03fpk(wRV2P0b8`WGMO8A%FwZUVt>cT zuNf}$d2H9v(H&rb6FYRXBj z{1&;!hftu7Dr)@m0N!Z~D!k1E*_j%o9;e9IL|546GM^?*ETXf6qbX#WDI~5n#$v~P z$PMp8yWI15!?u!nuX|f=(%eQZ(sN46n%`gE_sdFcPH@ivai`vZ#nw*Dg^gS+`+({9bz8jAUAU4EoPUXhuiDA_*#Ku zyI+Pn-qHakHR>6qE}A6Vi;RL3^odpPtYPowD$=~2<0*J}JvV=!B3`PT$z)7M;H+X_ zx+lMm^V{0Z@{{a@f1bVq|HMyN>^K5eHPen%YaHfN8$HOO==$F zPxLK|LSi=449D9z^uY~QUm7dszLoKjayytUGQ?dUuV8F!5PoUi2c??_IWEwSrhhRL zxpK*M_+HFcpPMi3*e>=i(jE%*;^|g$dH0WEv{#k&3;xB%&pe7bkw=(o;$41~*xzs7 zP)fc2D)g#D6Fic(2(VX*p0$^FfrG>(m%f!xm%1QkIyYct&uC_UTbX?Oxv*j< zWnvL}us=bL=2~)8v?`D@D-p=-{gR3d6E8}Tb`|OATWC~H}A(l!M!>~D3)@kv)6}XP5paj9e4=yM_j~rQ_PwFP_eG<*Gj%9o|iZ5 zr=IA3>?pe_^4p8woN zLB5)pF!vy5rh5r|ceKOa3$=7^gY>wZ&2PHqUw9XU)XFV*P5Sbs7sybQnQ&&R(P74Wv@G^V{UmW^eDuxsT{7`s9n z6CU<(UJ4TQVD@o3H`bY>!U&r8UK=+$k76%P`@x;*m+9-$3*gDmfdjfzSg`MOv`+a& z%gauY#Ysi@Pi)>eUcJty7q>(3{#H)q{9m?r#uyx>cZMlU9L>9(e+>F%p^eKlFWV9$^Nqt?g5WeTcFEN4wUp;`TJ&JIA!HPk{E0z?%T_# zd2bUN5t7fXF%4l_;ojiu6^hbrk&yV(2G##Z(Ruh&`Mz=7jO<7#Bs*j!+(InTN8>-v1&?>XJ1 zbB;7!*p$TYnfDX@Q}f}N|8eT0?us9~B5=EFBa?2}O&8ji5Ho=y&w{sxmP!N2-#Hr$ zo*D7k1XB!a)*uGYVo3P29{M_88Sef0$@G~R(L}L$vgySm;y65wS-r}D@IQO3xMeXv z7xBe1{r*34mkjvbF@d=`NJ zFz*mHISt=Z=i93)9${sd1(Kc4Rm`*18>muO6j>prh*5uy=;VG5Z+|PGTMuo9#~+Rn zt^9htvh^tLd7#S07A)iPUan!zj|&4)hus*O6$Vj{S7OG)dpLWhG8U}6L9MTxVVCMg zFvmsgsP?bD*n8*qUtc8~=l!>Vz5eeTX`Si;`@$Q+LSi+v{8|Xx;&t%6 z0+2gF=5Q;y5jA-(tm)V_U}K;U6TOz=F9UPx0Edayu3L81OMjEKb6V;6ag#ak6$`-Y zuLt_vSHOf)iUWPG=&h&?O!pN>dbDOWb3W!6o(&JBu;btQ520S26$f(7^W*D~(J{dn%_pE)>JgncOm*nPbe zw!OE-IF$ywO~uRU3Az2`>KG^76et8bNi(olZxTw+h{anhLjN$oi3(hP;-7>;ywLZXx0}@+*IZp$4GT@{U|k_J_0s zcYs#dl7*X(qxao99fzV9k0i7cmHW~pI-6)lT&N3mVYmx zpp2>iF-RBnz6Y~4*)T=V8VCP1qs@O|B>aUM?hOj&8G3~vQ=bLQsYhnh>T)xh_cj6U{+kO~QrWOyI|uf+K0?jr5}M(of+v>R z(G6HemU?KC4~ND=@e&=_XTp2>?lI5Jaa_K!D3m8A(Wf0Z ziNhydYVG)tPVu~h|1zFjDtljWIqhl;7Qkw*eJ>ThIZkW9~K_R&mlGZpzco``goCJ1X~9^TPy9^jfJiZYz4JI-oZ<~J5<+Oa<~ByI{LbcP1(W5^gkh_W2y|=p(hS{x`tRvB z6iKe4Q{xqI%$E%I;=##yQ@w}k{=Q0dID1l_c?B0Fn}fGe3Ml)=(F?0&pjl@%Gpoaq z{aaH;eXB0fBZsrpZlu8ifiTEJ=BlK+XAcq}IYy+=XM# zFsd>Iw)q}y3{wdtPvh!|Szs*O^kK>H&k)F!6b5VU({#-TEjlIY1o5do3hTW!Kx&gC zctj{rp;3nEayx)SvLQfIs+hPXb~NqYEb`x?IClIw9v-+$10P(jz<8U5@X7Hz36L$L zS0edN=MoWa;3SW~txAP^?^SWyw{4L4a1~K-M!p}~Njc$6`bj~PoNrL4UH7ZW2TM~% zYt|1^r}>c9AJs>@@@jBO9>c}S$>R%yNIGk`37p&9Or}H`LQ} z6CHt*UOZ{=v~=c--V11hc_lo1Zv=X~Lh1B`Xzu66eXQI!N3x>K4g;E_(I{C1a}~|e zYPkVM@iV8@2Dcc`4}O!jQH+>%9|`zO%O1@SKcFy*Ja4S`Gfq7+=!b!CZGJ5tOXTg4zP7T-9#US z$-lBQbfNQHny-~loUc5m>bWB*Jmmtatl5QE^~5PVT1=<^yNIS2c0r%^2pHNvz-JHc z;Cj_2T=#f3wQbg6m*r~U81^!D3u3TDLzDZw>lU0^)j&u3&AA7s7Nd9M5sa7nMDlD* zsCp~kp9_konm+oBy;BKpRyF{C(PJoN7=|%TzO?qTKPj=)W_RgWqEAu+!wPQFo#$1k zC!0ic*4}B5H1>r%-IKU|lJ6Kt8-21+yb#mRSwgbx2s2*%9EmyHLp%acv%PE_i4ZBE zQDU)V;?aA|_jlnWR!^F}S6)Et1BS`xmE-82`ZM@K>@%J3)J5AO73t+jo>RE-J1Oj* zOw2CXGHp!@p=`kms`6COU{YE{7p#=SiL=beO^*UdNU0}XE1r>)J8#kajKk!O<_UIE z>^?9YyM+ulJraEW(Mh6TZg0F8=Z_5$XQ4>P$-XXJm~FM*0TDOtF^W?L`$AE?Ypp)> z^|E&6?3!gzHQYra-|a&4FGirYNRlL+nGPrBNpKMwW4RmEn{iK{2PaXf$@zO`&~X~0 z#5A}JLayt=YwvE*la+&OWd+o4nHU#VcM#)v|K|^#Qgjtw$~kCnh2-uZWQB0NU}LO+ z`RS_0Fi$4Zfx9NCQNic9Ki(k=7XHBweP>Qj$BFwKZVG4j2cwQzA-}_`ARny^z_a}o zX`1c;(;Xh-uF`JPcAeB4$_>_?`7N>v|Oe>s&T|2;{R_of z=`K3r@sl}|Y6`~NrJ=`752=(BjsDqB{Nv`6&-VQ!tG<&)N|w{Z?vuEVKk|70xgF_^ zC&Z;C2L<(Ouz1iAW6LJsJl?VMFwzWXsZZx7ytKe6V@>FpFd5=%ozDJDcY}Gp=h-R$ z@8XU94}X0U;wFr^TMjRS{*kt(Y{t)K0=HNYM?XZilS)3L_@_IWn6HSVH#67Zr=ya%?}jz_Jc=bj zKI+hFe2(;J>Z8@K>-5C#9LSY8$ZX7%MEz@?+~n>t7?JXh6f~YCf-{RacHCb)k}JU- za2U%yk58q`CM>5Dq%Y7RLv6ZOSPNWcoZ(bAUM1GTnV_zxPShrNkVWpJBc`v~BBa60pIUB;MW!7k^$OrQPb_ z{ADJrIyVD81Qo;UCs*kD#feN-wz|Dq6+w&EgZO-S1fOqN2F_<);2!_?vjaNWgvJDN z;gc}u=9P?MJ6hRsvn=4i^g*WmwJ18?4dEWtmXR-U>JT75mTMEc&w4zN#C|3k7DFgn zwddor3Qufh9mv`W`;(>cjIQ8>-j92bp77vAji)97s@2U~}u0B9S7kNSV znNy60mJ^B?%HXss?@42ZI9;f;8$8&4LFCXz zbn+1UY1|F`Gxt98Vd(@TsR2x2%S^bFH(y{Hk2JuiER|07uXcU{jM3yjeIr5pj%d4UxgmGt!WI`)31Q5|8|pi3i}8-^pvc> zvjSBL)bPO0eCl~A3m#bx(&54!VjyV>2ZPelDV@(~PvTot)0;4RcrSXsbb``7>L6Ic zu&;{w@3xv26cmrpf+x~k<y2rot$ft1jXsrXuY{o;3+wd`_bMx>e20J_$^MVF4TT>9<% z%=AbP2-!OoE??Y=QD0u7+DC1gd~KuP)6BK_$~yy3_k3j%La$(o`EJ~Df%l(@MR68> z{2hpj0Z;iLJ7*I}sT065x$WRmpTKNvxK9JWYeD;q7v!Ce1q!q$;E|vUFtqj)ZPbZ{ zuxpvbCgnHzv|X56mytpCUzMgEHO=_!A&18Y*W&Nh9GX7WrG}2?pfkS!*2z%3p?Mht ztTeFtEANFHKap5p${?$T>xlNiPYByD2P?H5`J7=cJ-z7y*nrhgQ=`nJ)-?B?}_L1@zTu+?8 zj)QRP5F*vSl6xz31y4)%Av;$DS31Ts*-i=6M{GK1mdzx3F-h=q$#L@bm;v*pyq-8e zTF%YLB3!{L>@mKD(QyWrVm|>t)J|sEL0~6(4 zoctYisIwBq|93rVBn8aH1_2nZXk`>3vxZ zaD3fK!ZoMT2an=uPRb!PeQ}8%^=N1O@4u((UX4;C!w(oVa*Fj^8%ts?4bX9OPSTd; zN5FZ~O6tDK1j~Ir$YPiG1agm|a>yof>%?95@TKdFOHv1|b5CZ%raq>7XO*GX?2{;E zEejT>wYUf2E1@mAR%n zacHXak@&yQC(o_;Y=Ce9#Bc5*!hH|%%(1nN&0Q_{@nIgHS(1Vv&ld7)&s6yPYlLU& zCxPkJD|m@h#z~_pxIUg|Yd9Ul=bj#%^|wNB5Ke&pjls;B`K|PVx*?f;_H1KkWGFN8 zSD9WP|B~&UAi!|tbL76}SPbh6af^^*8P1&`><}{a3)dS*uCdP!GOVO$ zE^i)m3%@0KZAC;@b_vwFP5_C*LSp*CmSnrukYUb^oklf5h3}-qi=D<_Ia8Q8;VDen zqC?d8W)cS0)-$71m*b7}Pvk>iC0iWx5abNfNvm8*W7PjSNhzjS*d;4if4>NBo9Ln0 z1qG;<=I^gr#bjJ$GWWChGQM`N#xHz6dE=(5P@{rO*iaK%MdfnC4HU$jKDN@g;gn0T2;YM4R5m*4bPjV_l~03_p)7<8tMqc=N} zapm`!7%6W80*Q@el6nf)AhsOo!dl>`De%5NkBwPgN{z=93hwUGC!%C3JF2S%vmXe< zz*Tur&l?A)*GqxluH}5UoZ&odgt(HYi}2@dIj}lxN8BoYki7?l?Q6D0@;$tz7gNpfiO)8Sl3PaWXh%ak~27bkAZ@;;hllMKam%0NzkB6lTnXl?&? zT4B|}o;g1SwA{vW(*K=7&VCdF6m(hNY4`aK!!dN?UE?-Jcd>O_FOIhhh6eMaRCnob z_V@j55dTLWUdf(GG=1R@t$=suJh|&8>#(aDz%*+{&R)$G`}ybAGTuZ-fWD{Ek=h0Zd{CeK2frlV(tu$+MS-~0h_1f@!87`7C-)<1AU2F%4=Xr zaWpuh0}=jviAZ;y$K>Pd@muT*lFIjD11k=bJx=H8Jm8<@%MatR&Xe@WZVUXdB87}S zpii!JE1;A2Uc7Uz256QStr%*@;%y3Ckwq$)o_40)^71H?nMjX5GJz*wbHO028B0$5 z#b<%d_#%yG{$9O9)5MQ+R~)ZWn@C&YdjB8sulPeoRcAofN(hj4&QX@v3lm$W6 zMleW8pf82a!oMy9SUQwX3cSL}#+3>5i1amb*CZ7cH%H^OEf3LWi!sqHDkkeK6G>KX zFBvOzip<~{-hqBQnSfV%tnP>3>>?Kh=v-|A-8&oDBQxJ)&{H3JRyK-Yc`Hc`J#42@ z-_HK2ze{f}t0fcfg>onBc_!CmcM@HGj24xqac?k$>~Bqg;E5keu&oUdGtHyU7d@z6 z-#*+g`jbxGR!*zGG&3f{xnS3j$=oc`pyKC`Q%(Nf^3GQeKF52)*S08>y|EEDIdXV` z-|IPTtN_)paJ2pM4Ab^nbAivpxMjbWpz6cvxZmY{LvPk>`liGRby9V?7ppt5BhHt; zYS}@5Jrjkk@&_0d`&St4pb5AB72?@4ZxnTt;%*GwVP3{|HnLhA?$mVT!p+BHl$j~` z8YRHK+1G#%lG)d6n#ehnr(yQ|NcQX5AvSV*GkKzyggcL}WUaF&Q^jL3bnNCY*#Fdx z9>3g9c?~zrblML$gr`Hq>i3l1)&{AlOd?V7g=j=q5Gm0lIMyzM+*%8sU&VjcpZrD? z3t&1vhLXa^8G^{G#^}8HH2RBtrE7XOol5D_>R{mKOnDzf58M1 zKQes#7`}Qro}L;p;AERiu_L~cWLdX1$}i&m$kJCx=dcSLe)EhK9&V=ki>s+*t}*o( zF`(Z}UNED2V)mnr5AlkU2G0KW80B-t(NRMVqJLO$U8P<)b;@lR4ERgLY=vnzKiAvc zp+$eS{lKuE72NJF-uoGP4nr?!V0}qHU7a(7-nh|8bQ&C~zwSil_Ei!1v^102JU&ay zW+yP-yNX%8e;xE)NHM0FPNwujGSysDL<^m>Xn$oE3D%F}UKsLDMy3_@wh3dbs4`sa zDkSSqiQ`+rW@`Vci*%-^lNXaB*@a5u!GP-{$6x-YgR4){b?qj!hUayU*Ei($?*ENV zHox$~T~o-3xX4U6lSxmu<)F>0GgRC%ha9bwzlxG$)io72+cclopmCPoM z<{3C}Aeq5iYv_BC_*86FB#R%n3`x-j{2*AFGc*{Igpiyvoh~l${Ru zX?`IM+~tldWWKQh`ud<1WC3N#6QSt59sc}oPcB`Kg2!w$lgG2#Y+?l17287(wVh%n zy1yg=Zkynztp;0hzMLA3y^R-^?WfGFgVbB%A}g%DW`8!*C>ui(7P-?`!WWo%@ka^c|F*F!@D(dQJ`B#gP6HdE2sXOtE_q>k zn)WO_%03?Vo!m>xBxz(EE?JxkHWg`b|K@x?ud{}~-^{?uSyEhv%w+s0lLWt)$HLI@ zIy8xyi?em)(EINk`oSxamY)5IN;Z`^Z%GA8A6SIq-&D9Dqm4*%D(TP64w84Hl8h7; zHiYfk3n#>^-*-X{0q;TP&tY4Iba3`T zC%iZ|h&vW%hE7wCk|#xLz;IS7EPcEYJc6@PZP5aJw<-oZFK@s?<=15Hz;n89|3#8< z_b6G>an>BS`-z4_ikg$;HPiSW_ei*8YAVxv2yH zCLg4}-pSw*stD4D`VocIp-p)vdV9pcrwf|I&;2Q@w+yi&+mcSa&JvH){!qL{gd36> zi>aHWsCIY-6ossSjY*Ms&q4`Aop|oiAit-%R*wy1hH)VNGWfgl+~N7r_~TSK?%8S0 z9+s1bd()QT_>*(!hItuSzVjh{mt#wPr-=}!WHIg%pGzsc&GSrZWWc}ClJfovBJQh? zsoO4Ml9d|dS5L!DYS}R3-W0ff{|efvttZ!QZm^$|2I#HxVQ7g@*hcXKSYVk>)_t(V zCfPAq!!Dr>3G49GD;++2v>pyd?k94S=djj`cjNBa&*{CL;`aTgVxdIE3>=@G=hiFk zxAT8+g-G}ekk)z?>^>id3Z@aTSaLr#J;d);zxd&6*Q@w5xq;tB`j88UjEH}m1C`Md z;sQ==z{OH&Fqs>Rjg_jnysr;+_1iE$SRTW^Eun`~E$FqXAojN9k zgS^LWut$6zciYzxXYYxnIhK>T334uio$BJ?HdTko?p{R=mx)8CeKzTwJ;L0*HJ|vt zX#;yy=KU&<>BG_lY8W$*Uh15O30FnoX-Xzs8Xv|y-p+dp6+H2ZS022MC}C6-Ka$I{ z7va+xGPteM5bn3igSej)Iec?3ifJ0--iRsC5!&8p^i~pTPQ1j`n+{@1;6j)anN8-$ zh@izRDZJhw2}(ceSxYlN&_5!{P54g{cUSw+lBm`6!oKF@^do-#xZ9UQ?ivdvUSKL#2O#gTn)lesxv=EVGA zFHDq`M=LECD5!~nd3P(3F%05Z_ZBAak{qUIPvnY*H{-99!Cb6{1b4`{ztLU10`2{s z;AUzQo$ddP6b7_I@at>%xlA8E{Hp?ZFoW+|IFmr$8(f{jA-A;$ixYBD!v7!XazDdx zpHGmc4Uw?pTp?B*QlqLD<3R82el8*GJbk@7fy~hULRKBzL87~2@FJdIhfcVI(8p=; zXW7ZdrCXhuTE9>9_PQl_<8dn7YQ6z>W6t5aQ@O;YG!JFh?q{uAL}2IKql9FgM#(Fm zN$(F)B04HVbgW%ruVFk2qWdYldCm6Er_7G?j@{y)8^EkNXRvbncIFO|Anq4T*tsgQVC=b% zney#1-!tVM!IC3n_1J*FDXIoby1vIIf##bm_dWnJh;5KoH$hHz^cYS zq?`9MUk783K4KP3yGW;;zResM5#kchoy2iv`*hQ?PBh84gEWWk$4%>2d2Z z{O>|Oqq9mJw)v+K|KLl4*v0S2W?MV5{5$`=*<_B@wU?>Tj8nK)O9bELH6!moqruKS z=%rDO?mmOqu8@JHM*YmP(cjeNRX#3O$f9+A^Jtgie40@6fdpK9#!mQKk8fuN!QtL! z+-biXtKabsW(9SOy7vv=5;wATLMFt<3pdQ+J%fR~XQ-@t7E@d_hKri;30vo{=7bg~ z6Q>_eaIU8Z_ebS4zMLovJ?7WRj)gAh{y>HsTGdQOOX3BawKjtESp}Hk#rvS^4wImn zJm2_xIC(6833M+mhwnSC;>4!)q$uPqiP3LobaYF|_{A%^?dweJZP!a++s>n;n!62Q zx{7xB7vqR{EsK3edQo)>;OnP%DM?BJBQpoO>c9ZLVfJC<)OloAxhMM|Xbg=0*oTtu zH=%P~2)^riMcy7=&v+E);ej+^ZsqgOq;Pc+ZnYM}=EFVIAR-wpoDWmU$*PdKK$LTfljQDax>ENR z9d5jT1Si$P;{D<$v}W9C2zsOlp39f=d+$|fAM%k#Z&HF}<}jq$eP@Mt#3A*Hg=ST8 zXsc}{7mID^^tcATD^LWbWv%S5Yw4sueG;f2OMq-92Fj%4=t13&s9t;(m4gXq^dkjJ z`1jLK^*kPK;+-en1!Q94R+_P9BKoZNLJ|CdS1p~mn5q~Uocj=sET52FB|sbspJMq> z0awLmM@9CJ^N1)Gt7M*zan?32qoVa=n82Gb)_AlFt2KZ6>GxV;*7oZSygJZFL1 zbtT;JD+%S7{YAZ;G*mCrAaO>nsA!}Prq=w# z$6SabnwM9Ac!mr_3(tbXMSXB!pAKQfU2(5dh25lQCY*d|EbxLgNO;^wS0z6rE&s`) z=PDiCWa-4V%4M>O?KhCSCBY6TisJ!R2#$-bpmkq_i6JcE0?Vf0bt!GCS1^W)UQmQ@ z)lOV4R~-YE0anb#rg?C1!epo^F@;wnb71R(6n4lpnHkib3QuJvVM5+J^qpx(W~QFR z&F#Y^_jE4EiVTo)-xMN#NlnI_$r=dh&3fVdH3qIp_{VJ{->4p`Ai0*4ZpKLq*pruHw6=rDp6|Ck%HfuPiv&PuM7#ARRa z7chkmk*^Wp&T8BHNwmoAcx0q_H(~!F(PA(yD?h$-Z<IDJ0MoVpl}9vLFs_4S8P*+!iU zo373{B)7bk#6>}0MbMD?ww0%jVpev=7|35d$`)NVc zXoMw`3J94IKMs_~{3DwECdA4!91B~@=$L;yKyiyMYumAZgJ5HKAf*(2C9AxCu-jj&rV85qvg?{v#z>v zMdb-P8m9

?Z#)kmta+Rj>!ddj+q_-I5@CioTID)n-rBH0ICI7ZZgGMO|%ommfd-J2P zt91%Y)O-pd(f(}EYBWDjb}L>>I0nse@8Q+j5Xf1QL3bN{pf|4Vrdo@Rfn#4Dbogzj zSRiDuzNw*}x&zleXFPwRemE@HHImxOHqm7#o{&`+6{vIOTuwK=fqWM!lKGB1SVtX4 zLFl)Y-?s$!T5nA0(jc8{9&*>C#6cu<^LvF|N`+!D`RZs+jv7CQ%d-R)(cZ&s<^_a2 zILLjyUIc*+fcF}BylxsK$`W?KW~xH2y2g;Lrh2e8J_|NK?k6T$H?eB8Dwv;qQM%!B z9{aH%kOrJhh4a$6Y=62GyVx|JoSkogIS)6`H>*A98=Y*>v3dh$C6%1khr{Ii(koB^xCrv*0e11XGF17Qvn$&($f4R`tgN=6aX+Pm8AvunjCl_!Nqw}_VF)e$ z&lE~LhGEq1C1BpAOWzfHKxVfx?A()5{3bXYD-66~VvYfCD2_{Z#;=gvdZ50E3-fy`iC1Y?<7WM?TY0wc_ z1(U0csKbugIBwt_)4ydaFf!%%H&xrLMz%cQR!v{WD%~bw)h~B?Y}*LX>$`~aIwrx3 z0~6SMjp2}$GMQ_TJ3!+mM!~Pw*PuPhmv|{7S}*Tpo<8At=iFB^`cyqTF`|XF&r_#Q zb)BerkRjEIjREJIb9s~3gxLzwoAv)afp!xQ=yHlA(I`&eol;<Zj5mX`S>rB z9pScEH$*%k7bA37sZkQFkj`LN7v{pzcj{1Gk_B4UF^ndaf$JHe!|*X4|FSk#WOjqB z{XLs`U9zN0ycBGnDn7TWn{*$?UOfi2Tn5e59e@v!RrsMm9fI7143&`S)BPKU$u-VY zdfj~LxM&#rZ^1)sDu`nrR$Ah%78!c!=Xp@Du*6wLh4`OW7{)Bu#RsZWS$Mx1dbZq! zBEO^1W~$6D+KgD16He-GxzbIlj>4Q-n5!TB!&Mu&(w(>%^|a-=eI_dKU{(=z*{Mej z-x<(x?HcrGfCC+TC9oQzpP-R^KHKQMf-aaJLAQ<9!`d(jRGCnKed|_ny9O5npRGi9 zOuI~Nmqozt)s@y8`}L^2Z7`KBCv>@kH1Tr#gdVdS@aarfvM*2^{@F~RhMvZtlVuD3 z?*k$6(03Sl@B!ElPo)dY9+9s(Ye~LgFk0!Vl_q3zoYQ$L+%=<)E6!er&y1_MTsUVv z=X4mnq-wys`J$=7emhl{Cpz|3mS1uA0&*qedE?I!On>r*jMA2&u91d( z^b$SM+}c*+qUS|!Ozh;M7YzaR5rf>n(+2E~_a~C_Qy&x`RpSh+7vx&IHwHidCNQHY z9AB*kPmh+76RveI#;KcFxaAT@IuvKfHG@lhJ8fPoN2d*aMvkm?VpbL^s5jvXd~Wxq z(~s%fgaod@zxKflK=w~1q()+6qfas}-k-$rgu8A`{lHpLuq17SY5 zn3>heQFYCFkhyM2?{77SX%9@PagjWX_ztJrP2uyga#(mI zfs{sS*v#(Kw~5Z|pnbQa$lb-GX!dhA@=l2DSCj5N6CIYvU#AeAVE(w0&vZDeGZq$HA(Y-PRYIZ7l=rAR~3 zo?4P$snqZN3-0A!-gBPk`~7_UChF1OGv>0ZL)^Hj{W}QkXdsQjW^}E6AXQv^oSNS1 z!GoLLq2@(} zYFxS?7DI1^zzrEGW|rCq5*BKc-9{p-Au<$NLVaO_gE|Br907+wOmb$kUd5k^ux%G@Xe5Ipo4|UUBoDKJM1p4gfCZDNrKIuRyyt$?{^Di zdD^~EsF+*@x!=9u-LPur7t{r>hhG&MZxt}xx&xrWC@yfi#nMW?;I=buqcI(YLs z8kn$+W+ZT-#wi0fTQb@qdx1K6^}tJiDSB4zE1VoCK$#K3>kJn`x6Q`fUoVEQI=+k& z69uRp+d@yp+R%lm9dvHzcL@3X1LjF&FnH4;$t}SKhO|3Ekb0{yHDn5nH(W%wm1f}j zh#HtMMuDFDbpr;f&!q1GVB1!(f1gF`&BtI&Q;P01iwz{dDXkDm*58JOo{@x3a zTKEg>`e{L1|An+sHjBc|mae>Wg#IqBf}5vK;$GDo;5_&OT`N0@s*FBDt+T7Z zHFgT5-%ckVTdQy})RLZ?;dD@~4W0RY9QCvP08Pu1>64N7=#`Os#XjCV+H%mF&Un(w zS|?qEHLBKdb(k4dRgI!SWe9i7vT=OS6PS4p@MotQO?Isj9bd+<%X%v2I4IEcOcUA} zJsK8H_#^Z*q+?ReL*mvQ0g-zWSaeu9q2ngNZ1t7k`ECh4B)*S&RHJ)7IMSaxPS9uH zH$t~^Dcz%K(Zg3{^~71I(N=)2zfSC+)U5AlP)P(uXOgRQuUbYV77mwP&oN{yTNxp!pEkHh%Qv!6L_B5EC>3O3&!P3+o(7=dkHu_#z6lJF0+H zd;ljqWlUV(gxM~taAeXceDz6%D-8*U7l$^o^9IE*Amtkg&AbPBj>FNRWeD^;Vh9$i z`az*wKUjOC4;2!32k`zHBFq|1%ubS4D!i5 z4vJJA>H7EmVQ%AWFuWBDB^`mxe(H3jbS3?Bu~1Npe9l%pHU%@?Md)=Tlkm}gkiW|W zJEo2W=d&VD@L{v0qO2c}+xC~_k9ml4*Tr1+?)z--`byNu8NiKmmH8B>omiO>32w;| zP`^@@Y?v-C@V+*}%r$qxN>2g2XQq>Ja?(6s!i+Mm-ER}86#mpPsLwHBhXuL-(x$Kj)z6l>gho1Ew>hsdneFg(gvV&&(A34_{j zp7^{?aS%KEc~*4rjZz4U48z5@%9!?^a9ZYR1j(WMumKck?!@CTM)4wDL`xm{E>EoJm*Z%j_eOEx1-VL-v~ZJehmM7d#ey0_nSDk^^#Q0aMmfy zSyF~JJ$SWJ*e}}&UuSBPaf&C2^gK;E`nN8m+4kp!pZ9>{h(G50-rEzM9jl40!*qOO z+a_rKIVkyl@Ji+0qIOa+Z3qjzHxdUD_sZERq4>kZLF{#>!{enpA@PPUm3}siuD5y$ z55|_jK&KH@CPbc|&k#94`qQZD<2gXb#lf;~aq#7kB`wW43H6^^$$pDWd^Agn4BWjG z7LBlgOZI)}lf8~KzWD~s-WLge^Or+Jjy>J5K$c#=U`9KiZG@)^Pe8-Ej8y0Sgv3v) z@cE(xF#YpxUcF79?zp{~O7;F==Pi8LaF1-1{NBU;J5P{C+hwpunl_(&1ou0GXzTHI*fk=ZUa^~ki%dSk1m)8-L*_o)C>_ccDjsI- zI|tLc&nL|Tu9$(}0Uzq(rvY;c62$vQ89p(YPoM9L!%WQ-n54La$Ir88Pd$9`mfARC za`yr@`i;TZqr-TH@@l9IH^li0=jojTJ4itJBR1jsF8)DQj(Hx^XL${oIP|6>`*1Ub zxSO6PpXRT|=|UPFs0ak5RDGIJEeNv9bg_NQV(^kmK&`U7ctA1?pRLz}fNuv$^0mW| z+wUOcwJ%}@shX6%`2+R_!=ZWJeez7HgPe8Uj`zyV$WJEEMu~fAm+$?=OnE*t(9eYY zE<#qQ$HFO^T*;PgCzl$!|zH!jd zds4`Ba1-aFOQF^)kd8Eq!IYi-QF)gq|N1=*#{0f!R{pyMzoaN=`Zu4(2KmtGAND}z z^Y2jQHCF7Q52jZHX<9l|kA~D`5%<}4l0gciVDd;ym|=g7DIDDlZbQ1s_8Au-sLw_! z94?2kMw%do5%9i6E{y!MA0F*L2|<=s5JC2WeBCBg`o51jItQ__a~?6ZWTrdg=1T+Wnx)nMhT@1);qZD@4ei({>?fX$58 z?67yWa3(7QHb&&5 z{1l|i{TmJvlzt%TTIEA+GzIdeM2aU<9e(gi0b|mlLuP%VPNHV>ltX_(IIKP=gIc({DY?U&Ma_dudwFqX|h~H7Q9|*;}eo=(PF%iGu_ z;twV3lW3{bAKG|-KK<)bN3Lc}$Exc;@zn%pSQj=A`nT5M5?6DKY1n}-_bcgO=eyK* zXg&V*)~!0QL5DwnD@`MxcEKStp!3ZqgU9H{k`Z=~h-Foq#O|&y99i8*NbBk($x^ih zXD+2xW81;;W+e^b_4Hg>F)SD$Q}uS;WswuV4=xNh!kn9JFd?@WUT421=2{QI#xJz; z?=M2%jLN14-{fexivdq<)dt&+QRXYwPv#ZRdPRR`ChRZHf<5tXps_Q>y!?eHgcO<5 zPu=R&^!ipPUNnGaS4Xq9E3H_eb&n*bohPYQ``Eqw{(R)xL_B1`A#lTBI`gp#eRyOT zzVy?C^td$6>dx_RA^Z6!8(CiSBAk2kSQa66ob*?{fzG;Ay8UiGeQ-vR9(fW9QN(v^&l|HvF3E}+Lg zeSuY$j&yf#f4F+Q9Saua5U0;q!ScQvKiD*r7xdZ)NB52bV=YbGA1Q?<-ZMm8ks>eH z5W}X+XtIDe;ypCI5Npe0!D&G~2z5sw^2A;;bJGT}-H}f1m*mhP8`jbGoBiq2FNZ-h zqlh||tbwy~^Z51pX}o{#H_5Y5YrN28hp~gh=t|FB^mc+G?OCw~_I`GS-f$%2rza9+ zr*mXV@^pxnpNf>HAUUY&^sNVv(Opcz~Jcc&^oFu;9G3;Rl>|CV=^xN5w_79YS)zjm^ zSo#hNT;fSn&j!*yZzj-LJF-FZ;v&fTuZ>I((jx=sYp`RDox)Fp8^Y_OP0&BzA5Yy` z$vk#R@fX8JVpZQUaN@TbPdk^yHXVG2zn;qRfboiKz$rK2Gbgf)U8AvMMlhT7d>rna z>_+wHj;0G1QhHe0f(}g|C)hl=gUL&GW0+C4@7Wv>M}e3;Yr z2gj-6>~nO&)?hHLdPd$)Rim%QUIN*r9c<9V1F&qtC>HlZ6K>^Qh1AuI?Ed2j4j$o{ z9U^iOKfi|8y?yAGqFl_ZbLKaV@3Y3nGvs8X9ABQ&$yQ`q!hsPkT=(!WZt~-Yq)f*W zHFQQn?#BqYFyt|`tY5@*rDoDmHwx*tJ1=Q!p~(5Ttjb^X?jrjyX;S?qSLuKIF4MQ0 zOQ^=nZ07Gjp)&f@Vdl`d4@YT-k+Jvt(R(+oXnNHaI&82OOz{2#;ii!^==oUs@U$yU zSS@D3jlT+3o3wGALIwHq#2)7yyvkA+1+nl}J<>kJK|DXj5Uq6$*yUx$(;l3}ffWj{ zV#owce4{~J9=s&75%b~lx3hdkYaF7>Ol*C5pLt3$*{WJ2-sEszoU;(H8QK8F#l84z z@C@#ItQNP7(ZH`hfh_X2A-bn5#y^(_GoJ$^P z6z0*8mHKq<$c2KwOg;`Vp2yV}4C4z9MWg2TG~se&GYh?F3YU6DiT_g)V5v+E8}zG> zAe*X*13rC*#oq&HiQ!PXT`>aSx)uCeP%SL-Zzg_jXFx5s3GWz-e$AiD^I$VmQXG$9Bwa0m|;gLPv5;Tu~o-E$0JL9-<{ax&< zq=!=z_u{Xz7wpM7d7OJYo{stP3>3FM78;xmp@n#!92*b`FWtlWRp|!vfT0)3mmj~% zr5AFzws8(5ti6tly#BJXJw`Cg#FABu*~QOO&M<}GMSMZaR3ZOIF-d73Ntf+PBiB6J z&6OO}!GCTrCa)L-4{jbK&e2M=O1nS%U1-LSP0wK4+U2RL+C^$L+m2dp*iBbFl*US# z%cvPQ5bn=2=fi)rij2oFSgJ46Pr*N2>Dco#)3Fl8JQ^!Y!B1e)5J%1JWozkdPw%ebV zx0!Os;*~^m#XVtU)iVg0^pw7+{YtyOSXpRakfBy#$LTxm2aE|S^hUobxZn3IoG`27 zl?Jo9^0bq5?jJ@K<`#ja`&9l-R3-T2kAd{+-S~D%I-b@!fqoZ`;nU-=A5; zD&MVyS6UJ{YU&D477jvUZ&PVmS;O2;%27`-OQ11+AYUGMT9Q*S38$(^gZE7b{IV$A z{Qe4O{@FgC1-fRFc)duHBU)#`Bt~-5&y3|Ao=q(F#FRS)my-7?S0&4O3}I`p1U3$T zP+7dm4o7ZRVGoo>3Xawa}8|7zhbmHli= zP$7h<_k*85yhZ2KB)lN?jl@Oyfm34)&Y0{-ZWMfj+Nb;2j{bTWW^x|2vBFz~cfD6(= zd(9e2%7zpY`*%3edY}PuR$@Ox%xvx4xj~Q`mh^`Eb6)V~ZXf(D{|geu z{gs#dPU`CCM1#d!PL#t1esx_p9d~=9#qneFEDAu$BIk`4ZF(}2g$|G>rZ0Bjnl;)G z`0*}0?UJRs&z^ztsxT$6Gh*a)n&^3Hb zI(#NGIt~-SOKHnq_f@i`|!|*am!SlT^ru;olKXZ}zC>gWI zoj1tccsux=8U=1i@-T6N7OLfz5vM__l5Kk5N$i3@=;+&y`o;?&Y`&^s0gp(nm|Y*W zw+N0!)Ci6zO|ey53%7=j$I9KoF#1dqY?=E&vhnm*;+Up_9o3HTJAWvAU)6v*>*@(U zr^x61ki?!c0cq2ZFkrR^rvFRA5vN^*D}iDz=>AHUG`_FI4_1>2O$Cs!F#|)oPGINb z@yyT7Ub3$&La6;+4sDyw$ksdAc>3{rvM`pDIg#B0yI6|w{t}!1HxGvP)x>1wd**i} z;;g!10sgyWO8@!hv#NeY;Hi;HJs+jecU2>)LHt}vb6_4S%UvSpSA%#+8Yl9JqPa^#(m>?bn~cgem^C>A5{MDwC%Nh8KcHO$dGNqe+XdHiOOl zQDp8GebGZ41u;7P*gn%iXp;7s1j-#@{ccF$>cJhXN21MUA3edm7M=mw&Q$Z4UUDea zWJ;%$H&yn;m_Wj90+*eWvE9{z4QW12ehTuaRwxa3j1Gd2l_%Jy&0^;=8-s=gGoqeUmqbH(u^uh^mkL*K z1JnAh!M`3?VsighVgFkpc>j(I*7=O4V~&YC{=4^~JWLte7Az1w-lN&W&DZdZi}<;G zzfN5Bcfsi8yRhk;4%w?;%$f>)uuJU9=jw`nxjp?L-S`Wiv=7-7{Sase8cL!zg`rSr)=Ul*^(j0tU`j6Z-k>x{lX?(s%(?i7NZyY{jPuAZE+` z#N`*lSxay?t~%9^2mEKveKu$Dnt{&vY|2-9yKtCAdi!`LcQ>SJN3k928n6e|K260@ zN@wBb@C&eQt}*}hEdZXDIm7lUEjrgW4j*sv;=aF*@rp;6!GnIl7{?BY#_SWqpj27d za^kd*(SIBM9;J`dmK9>rv4`YP$2(Mecn5bIFBC#f$Kmi1)xzMLwJdp!Ib`wgGpZ-c3)Ey-)cmy#Kthp;$02d>;q!0?Na{6@SEZa2At)5lK6<^50K z?7kX&_T#I>;-?C{y%otKKd!;Irq$4Qlz4X;`w54gp3VL3tH9H@9aJ67=_z59V7mI9 z5ZxlsB{#C^#)B2mTi?si9ly>)#Tm0n(Icp9KZ(bNyRci?iy(N%E;P(?Hw$TtMfX1M zg`V!aEHTFg`)-^qtZ`{W+p2{)`C2=U8m5lB*T?dR`N#NQOpufhK1TPhm4lX?`xx}6 z6nFno;=6B*MIQw<{zqg=#LRpQK}%l2wb`#Re1HSrvGg4+JeCWAzmzc~>#%uMvL1PI z_zH8MRENv9C`w*V$*G)LC%&f_3ni9rCviYdF-GiXk`lLvkTaZ6`*{<^EK&jKc30rZ zw*4#)FOH@EuG}ZHYjn8oi5w^iQ{;LoRV+X07BspnMwREfv|H^FaVS%<@EG|SqE!-@ ze7qbs*4)GU)l2vbn;y1!sH2$Q)TbBU#t`qnTX^TdI=JNEOn#;qz=bu;x^A}9aZiA`yCTM+eUUCI(_$1d? z*c=^BE*-EJ`Po-+L8S$MU3VI%7L4I~S2HCp@Bfi`0Zpj)s{n%zT?3`d6WL(1NwEFX zQD_RS$6GHV`S<7TIPBI|$Y|4ndjF@;d1e`3v~oFrd_czHcKBG>w=fbcA0>jKvb)50 zx`#wz%yN1VIcS!>WqR1gW}HjFm8L#iGxIO%in%D61|9CClumSEBXhp?kHw37tZ9L} zu&Qk-aQS$%%>i8yof41lH}~Z~?)Tw}MN`0|U6FsVS;N=g;9nbG z+;TThupT>;x?G(_S8VtU7VT5X?DS#0y{CyD-F%x}zkA8N{)G*XE>hqY1$iv@h3FNz ztRQsFDTfT3S>Q8x9)?|P5FON(RNjT*yQvcFDLlkv@{WSWq#((S!FyruQcvjGFo#YF z=ufR%C!vh&eo|VH2T50ZC6)~f7?IJ0^Ow&_j$M8$nJ4oEEYt3@RSAXMrtp(w^`qa+ zF#agl9VGhCGOUQ^5i{YUe;l7RA`^|qc;eXTGLrJ|IQXpXfVMl4c<<(BJhQ$HI|8O* ze&%728a|as51ztr{+!9ZyKV62sTJ(9!d&<#<;3obv%#mox`m;6eZaN$Bv~VN9d+Om zEAXzzE1}Jm0q^5!UH(XLtiKDREM2JT2^n~}>|;jV)kY7`Z(~m3KDW#IF50 z?Bjvq?8)~tOm<>C*00p!-Gxh`eC1x~x-SJW!7aksywhZWRF93{4kK| zw@Y~PSOG?c1hNmcHf+}XGBoC=$-TcGfZXD0+(>^Ul=GY}|||%;cvGtyZ}X4zr4p8iYXBEFTQMTLov6Hto4>XekRL+Y}(U1p1rnlMmMylGg=Rcqq<*rg!frdz_jfp}Gwm+*_bvT?RbvqiW&f z83M1@IJ0rL|4IzqXW+uM2{^C(D(Tg9keDd_B)j_w%-Nt2J$nPual~g5{ZW>ACKWN` z8eRU?Xghz!C*ahWw>aTaB`!H$g6RWWLH>ZfWd6|>Jh3Yk)0UpE2(8Xx&MuGH>@VZ_ z_9QW1;1S7oOi#sAEB8a~xr^-2eS`^R6e7Y5x$yEkfJ!2M^fcmYi~@vV#_i{j~Ygo;>z3@$x_o$I>D)5w02Q?YL1B{A=r#%DYJ#%Y(<(W(Mmb)^edG}#;i;gx0-gjglKA;iFb;nP{r8HNf z{yPFr;%}7dP{)!+Wmq)y5K&LPjjz6(Vz@4mV$OL!Ct0g%Wm78udbp5#mL8y!mL3Cz zK{2SBwvT83Qs(K;y79`jGE(wmF_t~8;F&j_*+r#b7}FewH(Wv``#yxDSNIvqAFO~G zZej3LcQd+8{U-W$b0AWP!>!`pRlL1HwHVr}hRL71UO%FAomUfE~t*O5yV<6l?;v5A7;3q$zr zfn(W@#fMSWV+lO_mjDZw9bzkn$g=Uzn46U}QMPbH!{7Ua(-VM4(aHvVHd`!76< z{dgE9j1}{bbPGe*dn4fb+IIHN#DXP1QehK^Z6k_)eQ57lIoh;IQ&{@JijM!N2o;}z z_E{xn)sjAveJVSt5%(iA=bXUjXXNSckpocpZy>cfkpfp8rwH#$`r#qBc{IS}JWgK4 z;P5;>Q22X-6*(Va=VGd;N{I^{v-K)p*mnRQ{xOMbYUlGk&mMx^plFB}xw*Zzq0}k6 zjmXH$!uAtNk{N?H@cqjs<2!i^aFml`uP0}TS^X9|Y4~1P+ZiDldoPp@SG-PiAO=^y z?+XRCYiVUv6KtuQkCrWpl1;y|B+Fmkfl%8E=smd~KRrYsV|wSauoqdC-rJ7D(0Ru6 zqbOB;zc~(187siTVPBZhKP||doe#@e93_6A)W9|5qcC+t5BqmG4a!&4LEV@cIPLjQ zh+ZlOmg{58+g%DF<=8py?{SYSu9e{#3lbn-r86SYLoZq|F0Fv|H5nk)&>y#)jv^{&XR=A{lQBTdVOj^p z@r2zEcuThl*L~lGPu?B{6N@!4*LxLRdoYT&E4Q-S8cOWh&eM{L|89fYi}7^ixPFi( zy$J7iMdR&R%!0Avt@rQ5e>hi8gW4^puz(q+sW0ZDj292L0T{VCAM|B zWURUo|N2NEjstti-S$b6KmS_T?mMTU(@NaW?Kp~FdEbPa2iwR&y$H!H9SxzQRGQvR zwSx`u*VrqM98^~qU8`;5naREC z#=98ewb%7vAaxjTFNj4V-b<|?*+BhN6TD($f?r!qg|VyOkV;!!2)jOloS;E?A+ZnQ zCR^C^aJ6_#3gIVyhQKP{$r!G8foi?FDC=WMDfPOeP~zXGPu!JKrLAs3n)2>YmP5R z`N?%KWAzxaLBmh7f0RDGde#rueU71NUDHT!qC9a^S;7?bML%^pfs;qy5to0?=)J|9 zL=UeKcFKmD%ay&y7YPZ_*J(C{(j>`wElS?71z5aIpGS|;<6)P2aHhQx6kZRY34xxt zCN2%ycMZi?4Y|~_##VF|h2XfBiJ*4p1Xw;);oZ+lgvFi+a8-0{$-H=|v^Aj1bGi0^_>BD{J1mfY1&CXRj6b3$yP8~u>EhaC*k!C|gXnWs}BRWBPv|EUJj7BY@H?J35^qcm{5ZxW0i zsK_$=aZH!x{Ks)8iz&uWL_U!%&>qBeF zSx*g2d6&-I11s3P!Rs(WeG>O*SK{m5Jrq8F)U$Xq(ox(+_viDoKGPl6dKT^W3n3w* z0YdcbXqWmu&>psrIzHV?RG)QJ9*x>bw+8)yP4az6zhT`?uSve$^uuf62IdrGvw zOCLJM`2*Rn&4|{^v!EAR4}I7B(TT&A=|!*Y@Obt<;$)71pSa&vnoD(FQOf(C zXZ~yCz_!5v7ky-`=kXqxY!il;giSk4Tnk{_SiQi>??hkGs`Ier?0)WR z9nRAmSCCET0bfj6j_ET#LAZ5|WP^SIemvF(+e7xi-u((R^`RkO@BbSQi_E??3GR4O z_A^;`=^genH^=<5)o8*jaMmn;GGS>Lw*3lcrSFq)(^o0_@xdC&W`EJEJ&8NxZbKS@9z1>M(%aS(b{+Ld}0Bw zP(-Y?cn#)06QJ+ZzUc5e2gAGO;jSNQ{GPbq(_Qg}IH~mmtru@#S7$k{zmU#!A829r z9kJgw#(|a|ctaZ8jOjq<m!bMzZR>+xpUvzD`~_vfymnx&kgxEckU6UoHHS;{laEGX`=vi5 zP`VF*_jNoUmL)vY)n)G$O!O^Z$om2Fp-5^ek1MLiT|@6lG;c12 ztM_s%yn6G{lt^*X{u!%AMv_N|T-kK>MNnxzg6hsbNtIuk)4<=qgo*xLh+ZwKeom_8PC-1+6G)fvsF1m3|Rr!ruqz?eaPFq3^gjLz%Y zh6>lR(W${nnE2fcyN*_}(F#(KF7HfJrEh!^tP|pxQcoiLS#yRx(0|LO{xGDfJ~yBy(~1R%tj{2A zL%1qq&Mn6*MYlOpyr<5ia_&xV7HO3r?jof~uJ%Wael3TqViv1ctNPR12Yt7ji0U5_`;kHJaI7| zhO8VwSDck)OB!_fzl^Wgv8I^y{b5V)oEB$Ip%!#S^F`)xUm7B(9f0a71IS!mTi8A| zk!*>1g@ae_<-;G}!tW2~vYVFA$%oU6$;G?H_%1aJ<*hO?bCn@`6KI6CKNgwGe7+#W z-x`J`4vB2r#W%$G$2GxPOXT{LSL0S~u@5uY3aP&Z`kEy|(Z-wR*P{enyzd~uuR_>8 z>kFQle~4MoeW?5ShUCewqe#dgo>Kjf+|Kv|N$SJtQ+09QJ>FkpH+F(p08|B&ev5I* zQcH=&rVC`Cqa2^LVmSZ(SNzO8KC#}I73lxKh^M?>&-aGZAbb9t{1ZDn&qq6hy`LOT zes}{%hi*lM@XM$dKNjm3^|1TmKGl1x4+f>F;-{aAJlg0V&OaL~S?fIyFD%P~A3asD zt|0~IIU4c1(>_4;hslttyNs_h)aOGlnsAd^YpzRNL3w#rWtHL`IHr0B)E8!gu-=;7 zGRa_Pw&)6)OZ)JgSI^kp>D44JWHKurbWq}`T8`JUOTfzR4hBv2A`6yK=3&qelMRgd z$ro+t8#{(y{BJe+vu(BDc2tj5G#N8C`#c%9%7!(HKgY7_x2$)mxL^Ir#Gb4phJDxK zAIoBhdFM=UJa>-8l}ka5#xtz-J}mO}ikbGj1j#Oy7WRDlTf7`)z#CjQ@`Z~FL9Kld zG@ZC2^cuL6MeoOu8s~BN%{K$JkMt4`r*U{ue*mP-RN9N4K%hFXI*YLBN72ERb zHojkD4&<0Mu8zBiI%-|4Xu}OCK5s+i&coaX$ZbzUiIJ&^HRWMImqN^UCENZK~-;~52hn0wWbZ%9H^Crg>$lz| z{*IBj>U>{Xx!js~*Jv=S_0EER{cE=Nv=M1Jb^umKeSt}nt8n1o3=(xvmCx^2gqA+b z;rfV)P}}A?ItENx&Ca-RBw$Evgd~x6Or9pOnbmz`mHSV#}JAuN0al zd0_0{8i~|LC2~cPN-Q5)z^IA{JU(YG`3q5O{1;c&nbJt4cTOQeWgD<}wJYqM(ntpF z&KJnQ5Y}OGRigjMfVHfX6DH63LClXuVYc=}{5~&^uqih1F!~W(KHdO>Un0y3)k0@S zMY`K!05u)Go9x{@N%G+7eC)oX$+KiG3+cYYY565}Fgh1SWbW*TIqeciIHEvC{X|UM zod&lT=!5?WFPJ1PMbh{FVhL|*VBYB27R73lET+!fvnpFyy4&T7THL~<|lNL2AI);I-Uqf?<1+*P8sVom6l1HNl^KbGV zyu&U}bly*cA6fF)F|aTEUU3We>^+S&wjy+f3hMMqWl` z-Oj_T()tAG*QjB&;!J#ES483s2Ed~CAHa9%LF{h*h|}9MdEL&%7A5Mj7BAZ(ERr7v zSZK`tMFj;di`Ar#81K3P;aP>y|JnfJ*^`0y7M;Si>xM$;$@RjJ3M;(p@)hka2cgN# zeB82TFFpO=XZVt`jt*7&B)S0o>Cs9J+DgOOr{yc@q+_Y#{N8|1F3G`flYCj(j{uN_ zsL(BD`>>fW##HnoAk{LKI{T#4nqMRN+2=ljjPERB9$U<2pE_^ebNywLar=blUO zqVXw~b{quD>`@ZuDG_+!g*VF^{Rd{8b_W$pJ?hTyz_Qv396jQngpFN5cYM0e21!ey zXn#eN7oVd>v2173JoG5YVNomd*@}$*wDmV&PE{S-S(%R}#tpbs#}JDr{UgaQ<@lXT z$1qA}8V+ZD@nq0>(j78^O~0NGAFy(XoRT zQh)UVy1CPwPSQz-W9zTM%Vn*oAYQ1R4-|Jxf)4dh-AC;Ft#RG8TB<9jN+#%DgrPbm z=vL@mxpwkM9J@J?1U`I79)5R)L|bJD+nPttJSxFazu!rQIBsOq&Af$mV*Y99W+iIo z=;&1>u zN^GIApM-w<)eW_pvuHx%WUR0aAZug1nAK%_7XNk#&vtErX(_+({Z2*x&HN~w>X%-r zJjRK*{}=&Zwp+62!BTJ*l_4^(8H+1yG5cf;&L1QVCv@&J_$!e-j%C6g+3&=3)n0ft zc_W4`ok4y$3FMDP3h`@BLlgZmNIlvrl$;ECX{ZLPNa>ZhtPnGVQ-0BwFk=>Vs-5Xk zWmI_nj*WEeAW3@5#C~)clw6R;j@V}QX0|e$9Qujb)K(Ln#zgYOieQPNG4UJME37VE z#F}fgNMFOxgkQ5K6B-SL?K}H3l?8FwSuV?$%a)VUieN}hN+sa`io8v26J+)sqVR11 zAG1207HwQmx&7Z0>eeY|w(4L$RgKRjVV-h)$|o66inn9N7y98+Q!lvHWsl0w-m?2uULCYNKrN~ww@ z8o6WQ%Uo>M4U&Afcuy_|aUzs@lFQ!biG${F=J?By%$rmMLBenvt{z8K59|X?b3NK% zV}!fLn|%L-Fu`gvY`*8i9$c(owa*t| zmaQY3U>6DHc@dD=F^uG0RU&Lnp(Igh2^%ymla-FvCPjWtl4a77U^82Rua7TA<+>4U z**b63Xe}1!c@s(fgU9Cg=e|U#g_h=)gNulciXz+(8UUu#E$B^Kk;R5e{GQh_>@`X? zXPeG2zZn;ZYMU|p@z0oSp1BC#D4qv}c{?C*&@DLDmIW+W!us{KK%;1iVPVHHJYhK8 z(aDBQX=7-PmJ!(fPR3zY-^}v%+d`gGFw8z*$bJo+B7%NI9|SkWudcGVeF&hI=zg)$ z9#5B@6PU_|L+p_Fgiw335#BB|pc)MqaD=oKR(=VfC!CLBcKB6L8C_Q-WbgR^_zfTrkr&phM$RA@R~Uc41>QF{CA|;`T(` znKcdi782U0+YK}>q_L557s*VWkt|`l!QeocMUY*rMaa@B3u8EB(byGX zu}aOAhI^T@>Ig*|SfB(WXPUD8mIdrmu^)QYIl$Dug|K5#ks?tf6%CR~OM9nHh(wCYE)^|B%lkgp(cTeFX=tac zP^gshyT8BxJsxlGao^`Y*Ll63Pdp1zO7`@Up$v97pJpfit0%3}x%K4G7^s-M7hcVi z7*It9Fg(2$6bm?p}G=Q*f(Qg+_^E@?V}Dba&hb^O`8cg*cPQ4kf6V z{((f<9mMddO|Vga9C?0cAG=m`7oIr3z_b@CaKTs#&IWQK=dxGgQ|K|PwgG(OjQ6-$ z@*nK%+>b52+DOtNd6eze#H}{@Y{udP2E_r`0%-sF9Ec?IH_{U0)xX((H{Jl@P{_@2nIdj)eJ%5uatO|?9|)^Gbn(e*FPd^miMl*YCl{C5;JF`<$uo(2pJW*g zSEbh+`mYs@|1E~Cop;zXRXx~lu@ldpFUN~#BS?992JRhOL=0zIks(H2HmszntLo38lYYsv61xw!R><>0+k~X$QtI^wAtDt4*EU5UBf`$G9G5@1WQ~oOe zNAIH~c0d^VX&&c&BPGT-(e+zV{rN*3m)mC#BcWN##PrOvDFRj#CY{r8032%j3qZ89&m@+ zN5?bntTqoSPltO2C8}it%-LQT}7K|ypiaYx%;;$d}bVE@*3H~Y2xQYW9 zW^`L9==8+yi9nY5=fTDoa`azCA-qZY%brX~XOpb%3;1jb1WbPlD{F%h4_2bF$#1x5 zo-949kAg)vk4xDt4O0L59gAsR2KsMDp!WLp(4^HWrp%f~Jf!}CPQ!jseRc!;Z8!?; z-zV~&Q-%s|&tAYPhz#PXAmjpIuVvCnNW+FTGU-GSMmi|QPr{A`Ao|Iet$tvzG`6^ zDpqUa_uNac>F7XwwlNn&u4}-(qZi4^)+S;1&sC(XCRVf=z6%YO34G~|bkuMCLLB3? zd5w7{H(19pz3~+@?RkJ)SoVx9-u?kA-ksvZ&iuy})KXxN8KHqot{uxHA)RxnCBlqLu2v68tcoCXs1fVoZ zg>m;g!8$M!bY5SAP}><`FKr&<*ik)lJ1EMz z$H+t$vprY%xK753cRR6F!>1F={t;|gs2Z}b<-*{pXJFX9M&h^jig328iA>ybm29@K zBgb~l#y6YNVbMZ4dYcE)#{R)rS(Zbe46KJ^<|f#DcbU+=Z9MMu_($G3|AJZF9-IpWwpO-0vRjF5^Gv+zWlU6t z-;`Jc=TT?pTZmfu4F`@I$%ikT!{v1D;ip77FxMGP*Nz(ul0=WZUlzmcyqA)q6Y-K` zGY=G>h0w&_aj<`o4vg?`WV3PrCmMB1JiUQj{f7=WUUUxkIEduB)lGcgt48uuSVLHr z8CBn=0p-Cnh!{H`t#yBhYiTfV9#aHfFRECu@-;X#D-2B|?1@FM`NEvAW^}i#7h+59 z31LH)v(HcaaJRs8zHa|1a&JPoQ1^2KWSW;FKYD@i<01I6MIOJIXbBgWCCi#36P@M4 z$HL+bg>cr$S;%s9Ld)&1h3ykHh@ox`zMCQNV@`1p_@_4vD#{?U4%*=rJ9!eiCW+4v z3Fb#~Gr8@r9L)B-1y>?m;hDiDao!G1ZeD+fEM6AM%7y}KxUWzD`WQmTfm+1D0r%qn-#^_VQtLbE^lI&)hcjl!9qO#t%tN94aPo^#&keuIc%{~5gYrb z;xC_Vbnf*l+$Ul-U#d%ae!V^)BaXy}+LHTdSwFtmxfuv#`?2R8}ZM%Pe3=irg$pa8y&%+ zUnJM2vo&{$bjClid9dN{NAh&%WO!TDDh$U!(xKJ^`UD+ij{?Ud7%I~N(9<< zi6CWtYr)}DEvl365W03VC}e#m72~GiTYu^LS)ol6V?Ch9pHAV|Mm2cYU?Jrf`hcPT zA<-~ipS>Dp#~xU4Q-AM88s4XIdv7VbNlvFIV0q|-zJT}%(*QQ-p{*GqCm1q{YMej)gA!cFFJ zWB?`>TZ2!U4XS&NCl#}H{Ev}M{d6WUTZJ*g&h1x-+pv+)C%X_L*j?!L;Vd?1Heu-h z(i*{+agD_!I+cJ9t2`2pc{{wxLR@3Nd0B>iLJ6Egrwzzm{ z8GGLNfmBbNEA0LAj;PpwXVLM}bL^{)O!g-aKTI*Y2E#??Cvw=w= zRj6>46Po1LIp0&>#~{oKYc#)zTiWH}W9VZt*6IYTKIDtnl(jI)+KsrmOD*E@Q}B@7jkGn?O(^--?K0SIo)KqCw?d(sY$z1_Ye%`~P zKKtV6KOx+F^)7brt1Ihw>j0cGSE57C1%dc@2kPmW!jh9BK5M%LPY*PM?-mPiYNOs{amN8U^xwKc7SiJp`!-n}>z9ULRSza!j8$N@ zGma8+Ya#cYl)!F>X|a^*7NXI-AH#yx@a~^LZZ|~>7aq_@*V?Zv{g0F(`(4Z;_Z#yR z*S*+7+K1d!>LJ9cvRhuUAoPGC5 zu#}y}AQ5Ky*a7Z%S^;3db_pwoyJ z*fqT;wV9fTPT$mU_IztTCOMuj>9HOC$1a4^lTWaThlj9!FK6J?tHa2X*;B}snGwSJ zE**SV^OS`iaAhuiD}YoQ!C=J^B(rq8ea5?Zo@@kTXhh5cPfu*oZNsXrz5ia zx5(a8N2K*@3GR1TC4`u~q@$i_iAtH-@I@&Hl%2;x#$641w0$v7TV)SUw&&qp=Q0L% zZNi}b+t6?DKBhP<&zbL2gxdD!tY6h? z^OaclTxo7)FqSKNY~m4C8hE5E0`2FFpkHG>g+{4w=WD$h);VmYN=uY+*=%c!)|`sI zInk)l$C}?$$iTsq&0$(nf3`5NkX_3&B+a=^0@<>ewZ@s?&srVvOkfJWvz$ZpW{!aq z4>O7Nl?Zaa>J8haxLlZj(E$g<-eg~blW<_IHhWS!o)zC*E1Vx)MAq#%h+~wyORiZr z_I%Dk5}pwubvP8+jVT|9oS;rWO)sR@MWbkq-ym{sw}a@uJ_e_ZvIJjweVW@%1*&w? z$xizVq|0J7*H1i&OB@o=JE%Y1GEfEIZ`Ky$`<@j3Xd$dNy9vEq?hvc9Q}AbACU(r+ zhhdTDAv$vb6scWjDFu3TR^CW@$anz3rE|b(#Bkxu!&&UX882qNejy2l9YTqnob$;B zGx%jXj^xexMLsRI#cf{eWH|{BB-Z^o$ympGc1$1zpR6L6gH?aCaBIUln0RXsvz*`z3iB)3?usk$X=etE`&$R9ZUb=mk(XdMm+<-} z^4x0PAN)3;2fb=hP6FE!NU_s#V!hiC+}1t<2c7QNy(XR=4nKvTn|85n`G;8F?NX1; z`Xr9t>?~Ao97X+W&1hJ$3H5kv=t2sdTrPnootvg5W=_!O?oaF3-bq>T_1k*7GA9-f zwtOY3D&e#Y{(@M)j0TRWgUMFLLe~6z{JrTpejBUGDo3>n$6t7p7Z!I|gMtp^&Ws?J z3mb*>sV?Nt@+fj`h_l2;dxhSuCYW(onfl8Yu&JdXG<)7|mXtL zhU_Fp5_78Bf#>!1EM&ZljjVs(xsqvr-#=*8PMrYfbr~ zN+*7K@Ct6a z_9#ors~OYvv%K->MCm+j(nE|MwT)j=Tf3ge8U@DncKt%_y%GHD0#+dm&C zB`oH$U%+D=3`%@^zTwFOY;oSxa@-L50T-2|@P#=Rtf!|d|M3hkaYrGDs`1QJbp&p? zdXT%F?Box-sc^5^A`Gki1Z_Je@*(xlWC_PxSoa?{ux)__9y+nyX;Zu$`S8RYL(<}T z_bD-4u~Hvr7P<4SJ1?WUjx{QDC2>=iXBfJ-H(hOfoq8{gqNbZ#`1;YRB`>9YV*S#& z@T#;5&6F$oqi3J6>&GOx>~@gOw%$e4yN!WDeE}x@Y(ck>p?vNyOMbL-1XUd8frFpk z#UH12xxZ=>&noN7|B7#5n2HZQvP>V1qZjfGevCiv<%g-kmr0a9p}t?O=s1rRQ1l>! zee9*iL+zDlzq?xCvE@BmFhh+Nw7!M1RnO?>KiB9VgDF(#@B**K816qa2|c{c;O(n) zy5^Q9mokEUXYE07j5|W=;z|w=%Z4<~gQ&2*2cPY_mG2*&fP;-f;AN1+ zI?(LK&C~1g-JUG^j9#LbHTv>#rK54i!RKIiek%qgoAEd5)A-7>a(L*)7g1m8ELZJn z#}ki?z+p-<-Tz}4E%H%7#!fJ!>5J)rW2rRPz=WRH&7fWIA9S1<%)4esNoTJK+%vK# z^?iPm{&Kkmw^B0smCH7qOZ$=VUE5&A%_^34@i+V2?8bNeyvT#z4&~q47vR-eTOJ!Q zm|q{?#v}UoC^@ib9-kxSb=3d#fNt>?;{3i&;_I$@%+4&5dRT*n+7#)mZS)4nmU?Qh z&)${#9>0VwjcPQs03cb2gD~#`W~2Q_%&slMyT3f}X3T4`eb`5mo~9EFg~VP=+f#(%UY&Vi{fFlBlI|0p81XEudL*;mXo=O zG#`vxe4Ny5n8o|=)8tcbI&jxA9c~mp6Ti2PfX^?Fu#o(|I6Qm~Mj6Rq;u4m&m7l6+@hf%#t8g%~f2QV()k%jztg)lkq`97{ImsmAJl6lzVU_P?jUAAPlbpCdIF?X*7R7yC+BhQ`TOH|c( z*Jv#&`+b_*f8K)Y-%5LTp8`Hwa-UgVp2$nyr{UBTGENp(&kB7}w<2Ww#On~v|AE(QONNmZQ)#DxD}5$^56aa2K$dI=Ro7dn z#>5Vqw%`~&l{br~mKos=`D#3W>}wd zyzK47_USq3_wyt1+QpGKtU$Lw1KMD@5AVl^@xCKhGljPc(V;U6J{nr!qk`d_PWTM7 zclH$Cj<&?i{zg1%_7wh7^BoSD@g2?VG+}I9ci!VzKD(l^f|jOsipBa9@pOVJxBcmk zMJFRjYQbX3;aLd_`YwW>S40|By#&ULdGF+xnhi;>ka-8~VCQ8WqDkC2vHk6B$FHxV zVe%|1S@_InC|f$8sM|SG%gdJHm@5o&)4k!>!606{#+iS6Y`}eX*izMkqu5`Q%i6P( zF=F`_yjoC*gYsX)+6-5E*ir^zBic3O#D2pvq;Ne_@;I4I!iNCK|1f3T<^gX=bbe z9VC=7DI8Q1w8R%Oh8&^FDS48==n{-Q7lk(quJiW^LrM~V2b4HX-^EK3p7D-Ld->0iS9xC_$#tiBlzYfJg~z^g>9pJ(bpI=D z-mRvDKeEcg`70zRO!a$am=p+~T_f>Wf2m)k+z1(~b?G|RPR&jQ(^1w3sbpSf3su%h zuICWWOC&e?NNqa%TNqyD`$&P?a>2q%3tv3VK!2^RVA*jAj15}E%deKP6CIpgv2x_c zyUgHQ&NTR1-$;rrr7X4KT2{2#iT$#9i1YW4gwm35T&c7V-&HlC+1b%d=WL$T;oTwZ zt%+kVclQS8cz~x23Fg z(Uk7@y+F;Z^l8?9KVISgR?4;SMRL{!FY1-> z*H%+`pTZoTXMBt(CI6(>0gGU%AXXXw%=9=S&v@k-L5lk;^+ugMCe z{rdo%@KYa)<_@Nol7G?U`Z+e_`DRRy`HoXcQ}Jf`42h@OAAKxU@N#LY7}`%4dwo^J zckhlnX&`yQ68}1?=S2s~BW&!XNq~GoS7t&Rd@k z=3flx<{D${wW5&ded>xUiB7ND}c2VZXO zg%-0HqiOUpm|ZiHR({fj)0b}HuC)|<>F%5nmprsFy;r2 z<1_Xy#X{jSNTfiV?{kj*cRZTqx7qWgxfi+K_}Ms(5;~;W6LvgG6f-SGQnWD z^XbAhr!sDy+lBkehO)JDXJOHRg}hNCii>~W5&xk^s8H-l{v9^PwoomutDH<@CN$XiAg@>f@y;IT$IK+AZxZ^P3&0a~@6lx82>vbZ9Ogvr zft!>o=^` z(C3@8tC^XJ9rSxq!YbJ*95VVGXs%xcx}||)vW+Pggr>4jx6hHhahB+QN5X`w&4zy_ zZ%~+XPu%XU%`O_vgqB^16P8}WoSU|&lpZ6P{#QyWbGyMELnl5q4ROAfGss$?7%N#JAG-U7u!- z<`0@#+KfHyTGB;u-xowL?+v6oZ1my0Ln^abZH;Ys z=1TnJELos?w&4xWuVwJ#8e} zv$lX#)UPA2)4!1E#Y6F!@l(jxcZTU}V*z#Iz|Orxa=5%^11|LA!yW$Nu=`uY1K;<; z>)z`0YY%gHFyEDhzkPxVSwWD!GG4a-VGIsX(c`k=R(x3OCDJ?7iTYM{k}RWQuC!Xc zB!9U-zd(!8{+9)gHqH^scPyfB(niqvTh^l2q@nD<`IBOq%WT{*Cl+rspCL0=4`GdS zf6JP-w2ANk`-<_GPGE?~5E>V85YDuk^BReDyLsw2930&j?zJjH+V-KmVrvWg`RD{` z=?KEyU?oshi>5E#Evey?D%Reo5`6Ee)6>Ia!H!9ugz7la@aro0dH9pFo3SaJn(qpy zFYcC&EFrubZY9pCSHzvYw{wlSWFD$Flz&w(#~um?Wm}4r$eo0~Wb*4ymb~YIm@!VC zJvJIeMyz~+QoIe{)R>_2%4OnM-3wS-^$FMhC}O`2$MU9q^_a6Z2gB}EGOe6sA=;oF zEVigaMRg!`+G5D#uK!cJ#`d$bLci!H@!E#<7o;D^a57%6&TQ0ou4EnLGTm)_y-d+}IbtV(|VFk?l!3vk#Hdwg0e z$C`ij#!=SON!x%?#koA5TBSTC?`;S32~SeF=G#`8PFF9wPm$8P?T1;}=#Q9~mCgD* z?IC>hbC$iBw+~Y`XW$^60#q6^T2Q+)3d}OzlYNfgNU6d!e!qS^PmKPHCm(cy(eR(J zeU>_;%e{7bXxaPt`tT{$kFe;UI?n6!fNhz0M;y$*ilH~IvZUZ} z-1K!1u4)?zZ@h04zVy6cEBD7)Mbm)xGmycA_c6|QeygE@!A|VumV$XtCxgXoZ$TwN zm(h(yI+1sCoJ)+3dLK6l(;ZL>3ZKfe*5e2M1?Lwmy_)kR|0 zyj(Ih^8AH0I$JT&emE~_oXplv&|#kj24H6WIMS$*A#hQfYzpf|&JQ~+tUjd6ZG-*! ziR555M}074OWo}H>zc6mjv^bV--36B+{6{hZ{XSYHe#*shrdo4^C%r%@pP;Wz5aCu zarD!}#T^5HIo@Oi%L9b47qvpiYXhFu=M|}Ss{pG;b9z3&fL4_{@Lc2RbnxVMa9(nq zFPz)AWblxeeC=u*nyu|8D%#w{;3db{fY&WhePkuf<*OiEqYFzR94C*Fy!M&n;mN&N z9`WG~ntWdeEuPQV&R5rQ`u$vfWJx40iBh2%9mX`<^Mlw6o5}>$@&QSTw~2Bg)FLcCbMnxVjN=cBV7y3 z_pFKJ^|$_#;XO9Pn`%{hEV)$ ztx#jC3u9}aksdFD!9{g0l;^&5ejj|BC|9Pj56kL>dbt(Ya4`r&=Eno@$$UliZT39D zjJ&=qbvpjGu-VsckSEe}s=@i0;Mv~^(`W1OAus#zq;+j@(7_08W>{kCtV}X{VGipE zyo3hhJ5cZTc)C2mg&MCr3TsLZ!O68LP-*v4_}AwOnN?~GNu$;Z(pzUcET@8&NguRw1V8p5cC`%EKf8h@=Cz%7F$kM=Yp_NAo`)XrGaheM=QA$Kz)0ID&=XOyBi1 ztfC(^n6(>7h${bRaGam+ZO7kr-%Yg25+UW6DHwFiV{H?%*r$E_$@@Y*T0Pz!R*(l` zz4}klSCfV>|lZH_{_jrs-iWMhbH5ap9I`MgToVhoDC((lC zXpfZLLbd)u@$dIYTy9SA?H4Jl^IL~%Im*#%%0%|gY#gHB+>p(8e2J$># zvFY!U@L8NS7#=gjW`{PSlz9x_c8g>}{#ZObDTu7xpv3su1W`F70cAC|yr9biep$^Y zKZ3h6$Hn{D*Q(Q~k}lvV!>PD*+Fe0xT(6w* zkkB#wrQc8}Il4wvkWGNr5_h1{elxTW4Tz`lNEV1fF0mC;UI~9y_ZDR4;AOv-cEXWZh}_K71eT)8NEzZ3@M|iMEg!6;C(5P2`hV z6kEWeNw#4w_9`5}lh;V=>p5i@80Jj-$xkIa=bu5^?YCpuyH1er)&(UK|G~d;CrO4+ zCg^Tn%@-M*F=&6K+TaHmVR znBkGYbkWhM8*|(;2)FdJ;5VuXw>W14hDE=MBj?WGcA2l(3+;5#UelBE+T*yH-4q9@ zSD~-HAGiD9z$bodXXhpd;nSD~K|A0!xZDVU(Zyxr?cgc&jhi*z>X-<#y+zoue1K@0 z;fHlPkz$(1P?8(4oH@A+VV;Wyf^26zSSHN{RaaBe&#jsn9;o7xf8U|akEQtV`#Wcz zp-#qM^k6LkuVJ9yL#X#3z*jfSg*n4>iNSz{Y(|~*KYreg+CR7AknBw7LZ5o^^|@%# zJUN0zXGE}Rja#@*@BRG9szf-R7RCOISAc0*C1k?=JXV_eS9JG`CJo2KV540dW=@c9 zqOv<|=+i`Tx|ul)*t-qaRfh1zd&Yuh^(ESHV;HU5vPt|zqoAk6mVDvgjox0^ej#w# zEg@r?4wZ|)C%j%~4S_mi$^1Rt$iuMjY~R)K64N}9-78EHw>=Gze7r(QOjC_Z7eOVgWxBG>5t!9n4h^e`l*6Gw8l?Fg>+34KHNJ3$A77!F~Az z9J+EHcAqkw>-YSNpAQz`UMaN|*XuYR`EVJ(lkpQ|_wUjl=O55Mp|99gl|uG>fhPR4 zl97(?&)Jg0Gl}b{JT}hl8V+1@8?uHK!XAkMJxQ8V*ht)oOT&ucPQEwH&q?Edwk+qm z;S;FAuGxZTcP)P2jImjF+raiw0Swt$2A&a?P^VQSTk-dg*jGpsd`gl8?Z|Faqi`ts zPdx)Z=SE>-r6!h+P-fHii!kYzK9u zp@Wj>cfu9#wW{OL3F(BTts#T{TLzk1AH|z)4~3xydzsFh)u4Igg^)eQ2^XmT6h5|O zfxdYH?E7~bAFO!9Vvb4IbaaDY;3&wte`#ZRuS?nP>sf3@zvD#c#mPwX-Y@$a-0oH|Jt=O+AOF*kM#R%T~#L7M?H4&`XP{)Eg(zEt*~8q~Sl;Ag_xg%08{2_wyE6b+?mfXUxT!TV1iSIhn5iMWFH1UE(+W zJ(_9j@`U5&F0-qt%ZLN3T=qOpbGb1i(#71gi(XY%=5-GeK(})k>Cx>!7`!AO4v)8? zAumm(KJygmjw!W+3LJ!8!7BJ|f41yi&2;|vW& z>S==>kbvN5c`|xc-B*R{|X$D7hr1 zgzFiD5pZu!H5uQ?QMT(2tdOsu#)aePGKo>(vnP}`9=Sx#M_i}&b2m!;yfk=znXqw- z6kvK}cbweb2Y+|HBB!n5aEA3c9B0O`Bw5x@rL3X2Ib>{b7jMXn{wrI-{3` zCAXZ`hjpfzfx_kyWY6&;Cja_7D=!t$d+IIFG*y9D53_Kf+hA7I?=aRj#^cr%=@|L( z8q<9*b&_+8=-SdaaOklVi4EcF!z9n!Lob-V(g+`)d;#Cw_p?uP?vePh01t-F!EBFI z9H#Wu`EyV%L^c>vgFmb3pVlW}<2M%MY$8bxL+7SftLWbq*C6@Ea#*ri1sf_U+H|}k z6+#w%k}_Y)|9Oa4_3yx{kRxolYq;pKAqB_nd&HWr_ks1cN~r6#m9D?Gf~qA0|2iM| z3gulGxv?*{-+f6`B6C38gxKFa7Yicv+4cNZh>P~5Yjs~i#ke2radoq>=)ip;B=8Jd zQ@;S^#{2SLI#XcR(c|o0R5MXK9#1}A?jkk&;vwQ?2-#<6jkER{;&Q5k3)kL--Mw1j zobO~@*j5Hf*2mzLMjgZsRHGj1Pe8TwFvykc6f^8>;84U`h=1%xZ_V(aBaNC^zyV+6 zYo_2O%ad$tiV+l}5f1D(UND(!0UPBP;!96uvez;NDi#$;&n~;bJ{H7HU%aHUjqO>i#l{9pZriIw$k>|Bs@K>-%CKCP*Cr>%g{R@`UOk1o zmQBnt{UOuoJ_t5wd19+qI?TV)7e2OU!H9inB*;D%UJo2W*RUhs;S_- zcM*=4GQFvezsPB+2Nt2?iDQ1hWgkBtWx=1+VBn2EY}C~?u;qk5ed>OLUTj3_Ffxi> z7ka|jGBK9CYI4}Fwt-(I=uM7ezGU9KTmqCm~`lTU&M@#c;UGRF>Gz_BCvAp zLu(cfrMVqf;O$c%NM6!Sa=<;rnbYzxakY%5ol3*}rMF4R!xKXN!$nYTTMSdws~|Z_ z#wrvVvHPVR#LY3EW#1i%BSJ5U!390ZvEq*6{M=>Ct9?7sZTiOg=eRhZJA9S-jOkBY zlm>`;KbOhW-5q3!K|w4$dl_7bY#=_`_i^67A)pvr19h=K=m`1cm^rop7A7~r)1(gA za7>F%pQcWZe0m`6_LZS)ryYNGMnu)g%Sis?Lgx#cuQ3I+T)ccTgpAlc7wovE;3s>; z_SXhdt>$QYw5K}VvLzq>b82LpzV)H1iQ1SFnhmE$=HMQu4Ja#`gx!)&iR~Z_ejw3A zywT$lTG^)yzZxo7d-^N!WfValu@&DQx`R26({Za-H*hw81*11*V9^ABTo$|p-&^Xj zZ3>BSJzbh7zEG!6C5FHog-ybx$`bLQI2iPwcu}Jx3t$wv2wTdHap1xG#D7W%bNv>M zmSH{Ewo6rLoDz;o@>gMZaJmq&<`x8wbw%x-^KoZd4gPG_=d=D)V&~iw=-H7@hDdqr zZzcnXkD9>JPge;6lWZZpbtQV7TZy$f-`UH!c{n`i41~E2r6sQ0naWQm*1L8v5_L7^ z+GR$k#jKDVmR(qEZYp)W-V(1o7aD6vsatgbm<9L)ras0K|J`K@J>tnDzttq7Xeb@> zhzVy>eWY%NBT?>WCOnm;g50NlusS9cmbO=sSfytq>( zZo=?0gJ_4fDZQDK1WUi16}>90dCR^49&Xo*Z&yzsWBF;)7^;HZx2m$7E+sn0QibX{ zwMp0K82p{E68Had!kRMyGi{k-EOeKF>UOR8IrOLOxyzPsixuD9Qg<2_Eo z&5O~^0g`oh!P`L# z(eT9wr1f*TI5Pp&`uIUrQ$N~WQR;Gj^}r!Z%ba>{8AYB2I?@4ce_8sAY?9y4Qt)q| z&nzczg#cS$7(2Ov%@jwFN4HF&#kD}_oAUvj*1SVA^d@an->xn1c|Cz#k7Ke(<9agfH_uZgK(Hz#c4}pt2j^ZGph*+e@ zqoURm(eYI|tWUWw=9gP?tu?_oNxc9w(F?o(jTY2{zOi_dzhvyp?_!?&QCZ$2KXT&e zaPEE3hEE)X10_NN#dy~h&OJ@^0^}!jmsno@Al@#%AoicZwZp~i{ zofxbobq4NUC7Zk!Fst#4*x2T+xY*ITq^cyF4_}$c%|o>5^b$YN+#kwQ)wlA?8Lb%5 zm5JBgy73VcbxS7M{^ni#%}Z7ujp8vwW^uEgQ%FxKE4S@f8X9~b!Ydqm@$Oy=M61+o zP<20mZ@ML*lT{C}cs&C6Xvs;mMn~fQ-QfCRvD|jHB9E4MbIdQAUDMX5C*o{b1AR-x z!#%jOix$StzrbeKNST>W9=s|;C~>-3#v8IyF#Ci(hY77PAnT*h9K8i|QYOI%Z+q+; zEQhx)mSVJgA%6Iw#-BeXvY&BIP~)^#2ukfvExe8C^HulJ!%u2te4PFN=goQIFxqT5 zi0=(Kjls(t;I!{qFt1uejAnq#mV+VmRaOn2emsTSEEvfn0{ifxVd*focq|61_LTC; zZ!v0~Jl9EZK;(R1o= z)bG&~y>f#19N*Q4`Casf^vqWcao*9ebGip#R}nr0|^~eSEhOvrkEL zT~dyLY6a+fAqMwX>hc@<{kc>u;ip$M;5_$1c;WUfX&-n4TrNGp?~SgwEHWGQc_iy~ z;~2XUavlxlUd1Wn%J9Dpf3bI5hwR-yPwq`F^THE8JUQNsw{@ygUU&Sbl!MwZ7=7g{akUnFbuU@WO!Thimn=SgIz7zC)@J1j>JuW zEqS8baBr9|&;Rfe2Zlao3OCQ=e=`R0N5vEqYRzz))dSgtI2&yKQ%l4rGlV@yBKauw z5^laH3TJ+r!S(W%3-Z$6w}&pJN3OlX`VX(!f$n?Acj6A~T;h1hy2tGC$U@Ay7KLpm zZU|qVz7mrl0av?o%urf^FfbfDU;Sc<{g0qY;~zAb6V1I@;5SMu8!nDmJ z>HdC7bl%nL*xkLH#XVHy7q)y8`={GNLw6-OcgqACwx;9Ja5ePrZON|u3SztKcJSB7 z&q{d>%0~^1=Xcy{(e=IpKXu;}?N*&Z66B8G{00&nRRGsIXY=JzFZs5TE&Sq_A^e#= zM{Pbxh<*NuG;2G@%*gbejHvDfYx_KY`%(I z-}WL;yK2imbQ$q*%Z&nW`9+phR^hX?h?o1?^Pu!7Jn4UjXh_|z_PqvNJ7y{06xf#= zc@<+=Ll4qkvX>-&+C=`X(87XK<`@xI$BbUhaEjjK3+EhC`Ix6CaZtlp%=Z7wMhjoX z>TwSs#PBFN#^T`p@;E4}O2uG^CcVbDFi)k^*skS{ulHO<`Q0Nh|Lqf5N$PQ-A#^eB z_|SoxzB%Z1ZUnowI)?ms{hmC%sU@zKdTWYcOYD0$IS)A^*?7~RFe9BhwkV<<_k|S; zSB1r>U|dRywSE(YGmqGyW!ki#YZ7f8a2XE!jD`u0k?>OLTlnZ#l4T$4`M-|^n40X5 z+f8@C%pb-Q04IoB9GcB_Zj|8p?OkGK^lxbO{3+;vROdNX8a%{d04mG<#Jz^$@ID}v zliTn3+An{2)RhSSy&y{U#&Ad`ni0@G2hvx;c`{NG@2FVJ-)9xA)CnuNP zIW`D?+kL{ff9jz0=`q-$a|;z@h5XxSyOQ4XN_f-S8~jhNAbxj@EBpDnhHN3`oVXog zm9f{^>QZ~S?vx|0{I#Ds9qnX?dIWQ&q|tbUTOmDpjok{~fY!&(a%I|&>+bv~=GWX| zv3-Bb7EV0`^nx`V8_~cm-~HyHnv!p8s0m-uoj%?8Dfvoi?J4DHqX+V_lRu(Gt~EQjvJE@=N|argB+joZCtiDx;xV_Oyi02| z%6ke-M@bd+>iZTaj8rcP{y3v#;$QcYsb7Ebz$ZR@Uj7+yDLagMpN2xJ{cNtNg*@xl zac;KoMDIK=lQp%!@jsqT$PjS)Q@8su~SctBljCO-V ziNfKBOgu0TWovd~qxh9>6kpMM{(5x!q`uT!+nEP#iov#J0_-hRfK%h2pw3id{qw-auHn+DWoUM;pNJBcwRQpkLPxUimnWU!9!H|Ndtz zSkHefj_h}c_d9zQ!)Nv-@jZI6wcbv`@Cq*&t>#H}1by0S?TO<`Jm}i=?JPx**dAG{ zz*SlDbe7+TDc?7+HKD^H=O<;+qkHkshgRWorB-IJA&^E&KIj^}EU1n=!E~=RLqb<1 zU3dE$t@x}5|3}ez_~rP$aU3O~sXbIiA)z#$bDa{UNC_omq#{B@$Q})q_Rv5Xl_ZIj zy3ciz6_HV9G$@hrHOf|@-~Ia&p4W4ab6?lz^M04#SS>p89Ozd^HLBA2P;|Ntp^h!t zXDKT}m~vm?`%7J#r;|*tEzYKQf=1F8Sy^ndi3!$a z|HF#a?@*A@rjwl#*^6JA7<<&49zB&oA9u*mYa(y;tJYU>DA7qWDk9*(=8b#_`-H1) zCSrt6u5jnnIIwr#ivyf5lR|?8wzuI!d3fU}8g8k|m4DsEclw*?2LB*>A!HSGP>iMT z*Y4)>ErVIzrtui~;1&Ft-Pig;u|I9Nf11jF9R>BHtgLNz`&ln(b+JAcrbyq%*)ae9 zp?Iu$2-xi`0i*s+Fr?)N@p;&QJ+s$|-{Put@ckItb=Vt>`}~IHm?}7WcLWcrT?^kv zin-djDb@xtFQ7}U3iCtVuqNv+K0W&ov+W)T1G%wK?Snvr-YZ+Lu5hvb zwj+f0o{L4N1S5>BOvVh0Zk(|7BiSMs&CdNgggU<3Z19QwaAfH{=)1TYT;gP@Wa?J_ zct3~nfxmi} z)01na(y-~LXnCBabyh^TRn^2)+N*ID9bZNv_)MbPUN%#m=E>HXp(@tz({jipk1Z&5 zLhOg;DR5_-)1rr@8c*b{<}G32++kTEPYPbf?WJ6BrSonQF8*h!eSN_`?hb8V{2%R1 z4W_Smr_hJ)8tAes5)}o?S_-@*Y->;gp&}$3u%~>5jLPHAo|2Dw6VrFFh zH9IiT@`YvfDbResk&RaWP5M-S2D_AL{BM#K_|+sboR&mm+8)u+bv3X)yaLx-*0I6% zm1Oehmy!p13+RYj^?ZA?6f1u-gGQAcp{viop(C5_((^+W&;@7PvCU>0x8E&h4X^mn zH3xpvZtL}Qf&N}xp}QRSynX<`BBSZMwtw{Iv>rNeZX}NBiN~rd_QJQAVALr5P8^pi zQS-mKFbo6fhNed7ot;4&|9qlTC%l5<0An6FJBL4;T8t;I?iLLFqrkmzh;`Ak7kEEQ zlbg@a#%WbfSkgNOQ|5n!UCOnjc<%}*COzu{l3ezx@Ie^?QO$X-Ui&@RSi+{+knf}=HiX% z!7${{edf7zIgQu)L#GZ;r-!Xe=<|QG@cKNl15q81 z9zVvjqDK-|+MLfee$0dJF{VuI*fMJLGMQ zhf8SA`wU@G;br3XFab6CO5>O0yKv5?1>&ddC)q3PXr1w2nl~Z9+Cg`Y^}V6lG%kA( zzbq&)XSKs{>#;747#2k{yBg@b3kGoBgra@+W!4~G1Ku6~kyAg?$b!s&@TfzSx71z0 zq3x!8X;?V3%De+gI3+uh(|r6&3RQ z`38F~7qoz<8BPVLlF(yz2e8g+i1n9_Mo_VeV(&k7lPAw2sr=k^bZ3AJ_1M0G&X=5q zyo!s^V|a}!em+jOtJzu)vt5DeC*PCM^b}#k4+q@p(=If%WRS@%vq|F0(U>k*CuuEK z$ErTj@Fx5+~_ExZy+iqdJu)OE1p?IQ?jjpfGn+5FVM@5J-LEZVqyI9(eb zO{-e(!`#MH3|{%0Et5IN+#hs9WzTeyy|*96$^XKdO`^9c%bt#Ve38~n{zV;^KBX#6 zi6DD@J-h!w7u{VK(Tl3n=?7&e>a7(iVTOz7ZwD1>%qG$E`Gqw0{e65n!jrDo3860e zuj$!Q%dL~II#@ri(&lxY_I#h}2|n%3S{`I3hwCEF!haf(B)T>MFT|Sh(6^~<`(I;R zb8|TDwzIeXDmx9rHYQ;IW_7NfKZcum#(>g_Beb(Yho^hXQ;)oEdM8O@-SyerI@4Lh z+IeMvzD2DCcQ}@TX+#-Wm8F353x?sOB00&a2U9Vl@*8>&K0|xz=h96YyIB9XJ9&R~ z!-|@}6VOsosp8D)!4FoL`=3!yprm+9ZBujri(TGoL+awvHLTyASB8l8<~8Vh#f zLAfrtv?{PX?7v0aQgH}>a*6Xb+vo9$Thq{Mdkw3fS1cKxrN%wRKjzy?P5A(`eW;ey zf?uA@JhE@8xRX<4 zq^i!9?7681f$f20o9j_rV4F-59Vg<`lR5ajN}Li7>W4qZenpS2X&90cNGvTf@lW1a=0P? z<{yoZS`YB)q=Q=~tMaCAk8#G~FRbXrIWBv!l)G)$;gjX1aZArIo-}w6Z>TOL6C+3P zy^E*e^gLM}@rRSld7=E~&IRy0DTa9rienaY$H608Z#H(#+KQ`vGAdd(gjV<{=v3sJ zz2Z8z)hp&M3=!_x=<_`XXLFeoLs;kQaGdz2hV@kMWXV@_aO`kfa&F28X6M)kr}i_& zgsK!?&-E%6Z+pw9OibXH*02BS{zrDD3;{=T!*c82)@b?PS4mT!t&){_ zvn01OBAL$J@#x<KxT|GmE0-XgT~fxL# z?t6;8e&vs?nb)CXaU_0=>%eQi3{C5Q;hD&#e5|jS4^f)J2dXOKA(aA*k9J|cRgSPX zX&~Q_GzYadT*pcp!5~#284+7yvaSrDzVIe^Y3^LXD4 zZQOt*@&0$;Fr%qCBtTZ4)ZghZx^3$4UtUES4)I32b8W)apc%L$@1s@rSW`ST{T_L# zXNZIMXIXXY$qM>weQ@N4{iwLmmVXnD@RNR4cr3^U1GED8(Pt)Hb6OJq`ewlQAHRgD zJGY@}Z82`N{Vo}Lb_ZU4?*m)L>9D_{I&kPgBj)M6=e1W9Dk4Xo=a0kQ;+D!qSYIs1 zZ*F~rdU@mcur<%Ie;<3^Y@t!nx1$~pEkYdPD9;NuhVfmxL&?&}1N?Wv1Uzb@$tNuu zRB^s$CeK*%6&r2uvV7Sk%;exU)P41kY~Mc%W10uUV1AgrRGNxKE!9?YwZi#IVGEz3 z@(Is(eZ(QLAJ8&=JpYk4M&b~u0oJmS0_`ji9Nzq!IpB&0s>S|fW zw?Gg!st5^={jl+I4!jv>4SlYr2zz|9$r{)Rhg3h1!9OTFqdJ^;dM?Evtz`DA?wZIa zX+tB?8~JyFH7q%|mlb5yusV;4MBgfhy?xup^s??qraIf=H~WRW@$v$uU-cX{Mw;?@ zpE6lUYaI(zK8Sk?wxF-g1)$NQum8@}Pp!a!sH5DwT#aA+p^4-Ani5yPNTM+;8g<8=XA1pQ_^0Pj@uf$D z$T^4*uIZcNtfa$|3GXID!tZj?qr8ciw7tZ$ulnNS39XW)#=G$HuN!!>*bU!r@`19d zwYcS12IlMRS-8?!?oDJ=Z$Ckc{Rw3T2`8)lo{DFe#I(O41YyK0zIlc$;DvvY#s{atK zXN_d&%^Skb&wYe~*PkV$d>#^qTOTCH{K8nGZ!w{!eUjMx6i z!Bmwz;fB&+wzY6Ln3rfl(~%QgDxC2niw3eZC3`S^lg>Xi#BqKl0#BCA2dC>RC7}n- z3GH$Un0!ohpTzt{-F+GS_-8At1mgoZ_k}faZwq5aVs3H%P(Q2H9XHrBb$RA^?UGRD zHWUM{z9Vld+@br3g4m^SK++jVRK0b*%FOM&iZrj zw}T-<_6f;&Xv*oePEO+Qa9cqt}>dTj}4+nDg9NC?F zSmNpWf$cmNd*OKOdNyoKC#$(=%$~@Ypp@GYqR+yxM>d|h{FqHjmi1-Y7Lh`#=~Pf@ zb0;Tf4`%ySo(Y~wmG}g#c>4wmzO~O6^r+d#!>-3;{xPxV^q>>{_b4JC=>)wVsVx4} zD$ZWTqyG1&TD?368ddRAY-{<9B}bra9X z>yso0&jq4skQ4h>v>HR+n6keK>9{~agJ;B8OU9n8Ba18QP_5}9Uzc#6YcJZ0L2HLX z_q!ByIID&GLRRoI^SXGXmcW;O^Wx3o-0x3=R>f)iTsCl_xF`N7&j-sJ@_c)dmEKw- zOg?VR^$iDdOJiMrOpp>CY4@;cgCgg_WBBIFLwWkEA|7#V9(SEE8Fk%5uYEsJRCV(4nWlE{03vd&vHpR^d+XcrY(| z!9Hg-g7;oSxH~@z_o;<5n;Vr(-lG|2hU;KaWg@&umj;)Qw_w9nYg!lmMPgy3L6)sb z5pG-EBBarkycin?O^05?$kmUaS0P)nu*3nhUiKAsO*@M&3!apkv%`>mLkk>sI$`hN z+pO`?BC<7Y1Uo)x0~{_B=OzmraqwpYh}jxKx^8n}%BUcAZEP6HKe-H+sP=~m@LXUC zO1Ml)j=lKv0{Xu^LKL$Mu^&;wm_N5bD6EC61{BsxJAzNzE0X@ulIq+jgdpdU%qg=$ z655mtuO03ADAhI$6wmZ4=T@`yRr_(b|1)+>=_6a*w}VLAd}rQ2bIIt*Vm{?{Cd}6r zdC{jb>Fth7WaMd68n`uyKFKbnE|)8r;VuudysU@Z*rkT8m!GqLyr_J>m#(mLgBh!y zK9l^cUJH@YBjG@k3O&*E2>gm2=;rqyAg$7bzH=Az*}`i$VzUu%l@&vG!5Bysa~4Kx zDZJR|L~}>fvJ%HV`1IyLRNs(IEY}Bs{kc-eEX)&~Y1i515rg1y{TVVz!-u>s4F^pv zT`2nN#3Ij0B(v75!vK|iglu>Z1&ak-;%_Ku4sK!BQ{N!JJrGAWdFCN3+7zeQ9)2d*$&tUSO zoV##y(I?Wf<%6(nMlAVecZ*aioMZVRy=Zr=mAIHHg6-I6R+IJ)!>+&#Dla>q*2i|k z{Kd1GiP}Y?@yCpkn{J}t`ZAgD^QEMF`9)!znFfsbG6|{*PBKHcUN+3@2x^(j;wxzz zRw%2(l=n9XwIzCj%H0N7({mdq1|_o=Yg2d=Tnw%YYao69P^y1<96k%14rg^Dsn0H7 zYI`Z0xGyLYo^KpV%b$l(1*ugq+Bp>~|BRsZUp%Nn*HkF``VUe^{E_(Io6nZ$$>5ET z0(8!9!&$ZM%*Jw!;AyiG-~3G=j=vN{U*#FrwRbwM*lL8|w3DFz$6$EdZxQW(auChk z@d?J&ZGzCZL-E4FmGD|{q+>7n!`gc}Wc&{`_RzdPZ121ca>F&~<5WLV`pJUYUoU|* zb;sE0Kr=kkYO_rm99(r)dtO-^X$>7;}C2W;P91C%eWdH5(!}485@U3bEl$M-i zvtGnNmZ}LId8r>!{+Wcs8oSEZ1ox#+pPqyho-ZJx<~XyRnFBsy%Cu!=61INHV-q8X zqt4WZ@{dDiz{Iwpw0XZAQSn-UQ~g6E*Sl_mo5Oy|-T5+bTgW-U`~ zbb{rZZos~Ni^!fhvBy@qjofTs0K?5Ku(`9ZfXT}wc0IGp?`OKRD~6^R{457t4-*>J ze-$5gdMHO353GgBY=@i&xt}u!XNv5e)k9ve@N{c@^E{j=?6n2@2;t~4S^lo`k=R4r z&91G}!d+w1gz7u0s5w##JVMVv`U*>OI;}r_vq(G}+)bhLY9EkhI-F_b+L49wo0;#{ zC&=s6>3YRN_Bqpus4Q3o|B6PEmWgh#xZ40XiCnnLkH3Ly!3mIeNd=P+0$fpBr@?y3L!QlVXij-b` z3|X~n$*qafH1FXadc*y@VEOta>5b5%3h%$Np3Wg6LrZ~_T~i|GPg8KzULwg|x{7Qo zQo$v@=SZU=BX#N#sPR+=74JPHOYG~|tED-(RMr#%eZGk7CrglayFnJuvn8RsEs2d! zGVE$?Cu?4Ow0ikuEX#Sn2t?^4*#C7z=NoA(JLt3EURX#DOzKMx40}ng`t@On_T6Rq zFGi9nS4xD26-&s{F$?F48uJ>+S{Y`nGFq@X&PdGI=8z z`1%jCPn!X6;`&m#=Req~S+$mjMlB}(^C!~Y^FwKMx(7C&>R{=gPD_q3O}wul%Pcxh zks%e*-0b>0Ty{7@$o-sT>35?-VicUu-imWArL}9Z;DEK@^}SuP@K9e86WtE$4obk_ zqZ3;GIB3};e@NnII-U&LVhR^7myqcvmr2%z6oY@}Qdk`qUB0+(3yf@0qiN&E!IT;9 z(57$+CUsp0-HT(WMt%(Oxpxe=CwkDtr`d4tNs;jR+#1$5dzQpfSDcB7ofZAK=kR@l z6m6^%T{(m9lf$buAbsN#LI05{{!`5&KRmQSQ%8lS-Wg%FNVNq-tpw5Yya+Y3Ea9q{G6o!4ivB2_~(X1Fk1L zU~=6uGSgxw1Sp&G3j>v~v?P&T1O?dPe1O=+K7busxnzN}JUKRFArtAS=q)mG;@-&8 z1g$oq{7?gtvup=tq}bje;;s#sXsI(pB5GQnQMgVuscT?k!qHtt^ruR~Qb9 zbtY3PX42oLl@i0yMPN|iK%1PW(hZT%gj03uc|=J28ma-cQCu`R7o>uIZ?*Utr#$QClZAc$-oTrMbcGkj*O%nRS0VA*El;m!jyuq`}} z8@w%f?gM-NVTTge{BV{vJq;=gpFM-H#U@0)dcUM#*a>{$yis`I8H1TG=c8)wc^G!) zoS1Qa1jd3Y9@(zK4d=FkpVBPQd^`bl79W7*?f??FER|RYGqLTr=x!P?N^W5^sg1wQHOdlu@OBJ65*M5U#Quvg^QWwxvouipZ;__#dY9_q>sRXyO-2Up~7 z{dn|+Sjc>52afCFm_t|}-r?1V{VM;mx`4B=vY?9e`7@jj`>9EXe6eQdDwDbNW+`4B z!eK<<6qx9{mfjq329;gq`6Kfw=s87?D<)0n`TD8+YXHTT;1{_2m^51epYTL4#A6rlbhUc?|FYRf9#!wTyY!^kcbf@z9ag~y$r(cL)YnGrKaZ)gR zt-}@9@8iF`l=!f|y0AaE4n0Bo)7n8T3@nqH`gTZh%MfjZ{IH^&I)INe8m zD>o_#gJ74u1AVYSWcK}CilZ&FC2R90^RUxyC|#aFT+VA+{k<&(^`Bc<)Alr$cg74J zH?M^Bwh*}9n*_!K4e*J$*5s!->T2499xP&!(|=a#FZ zoatx2?3U;Pb2v{f4E+I7<(BZ{N)H>;*O+w8`@&v1i_Amq5~ew%3%q(9$#vxvh~B;u z^EMt zZFE2Qc1#*y-OHd8zdeDr?_Sif|7;TAwHw~%wdJ<=_M&pZ1SE+*J;1Zb}P989c-?%6{V^Obet3@i`-DQ7{%P!f zXe{#me>3CyG#uU7%}7$Za4sMNk54&F?3XK9s`ccuxbG4S9k++Po7E{4t6gHb8P&M| zoQkk0=PWB2T0&~=?J+=PMmz6$fT|D8B&)r*U}n%M$zg{_q^M*N)5nhymxRW17No0o1*qpnkgMYPUN!F$DR%uy zO!r@8(i64FE%B%2%sxV{+&V|TnCS5qeFYMu^qvKlXQS@*`&iUpgKDi+K%dIl!lAvV zvBKAd7g|@5km~|lQ?!pCeO!f~sz)R25t%~ME2wqwPsuPAgguXoVCfD=+9BD^vlOJM za@7ztFs)>Nrx3wCTkI!YYDTB-wR~4=Fiey?nqG;S(wV9ksOTv)i&0Lt(gF*I%!iO9O_#pPnZCC=F zh@XyU%*^TMm-!eu^(CsW%Yd0g3#!`^u`AP6_y)LlDuuOddVE*`uYQsablCW8~gQpv=72g3tB;MVQ|2m_>H>{5) zs_Ka%dEpI7@%P8lj&M%mib+|0G}@+4=i$edx&B5OGCyhz8+{{;>{6J9PZJhUMH3&I z{Nf63s>tKjfdF4NJtCLma#*L{S%~Xcj%lvSILCMv-PG_BYGcIlZEG8>?kTf+`Yer< zzHK7gjDlF+o^;SHSOsWaia`TLajy$c(D%Ykdcp80-S?#oYBjC!g6LY+tU4>CKN|@P z+*P6YR}kC67P5zi4!DMWCzInJVpsJToO8p2y*f7)TZ5ire1tdsU_Ka`t1moMx`Z3O zz6yhi6nK2JE^>z!a@JmKNXN*)$?>BgOvZ{9$ymW?k30z7rpRv$?I`!QJ0V%TQ5B?j zZ-j1RTP7PEhYvItz}Yv8=@OGH_|f_aZVq0_pFNXi3V*(_lqc;_CnvI$i)8rQ0V%xf zi^w>x$Rlg|^?`#fOW?`Ka|mI+s4!tESs3@B)Wy^m^!HhTUU>sp#)zTQw+kuD2?X06 zL2%h7n0$?i#A9|P=x&q+MO$2niYnBon1d27%}SpOTBr4L@loJF^)xf++{s>Q>p67d@8I`rB(In@L;*D z^^*clB~teUpZj%O=4d+~UO52&>pCGaab`pCynfg+WH=nXoCkMf4w2(qLnZ#sax_W% zM7croMohK%4si!wu%vWf`nq?Jb?u2dxIV&;d*l!1t;xmAujxg()}R7V{c1tSOeja! z@Dj3cfD_gfrqaaD4Rpw{BQ*0_4Bb+)oL1S7;8JPHxU6v;_@7)uYNCVj$8mQo&zX*w z4oZ<@v%G}bR}MmZ%V69ra{QIIWH5KL(KP9Sg!Y@JMz!qjL$ioHzp>~(Y)g)%s=xB6 z>8!JKdx{v)HI||IHb#88o(_+x6`d}osUUS@GI}2l=UP`Y`9dux^0D?1$a!p}8)Rb8 z;g=FglADVfI~LOIhi_9iYES>^4CR;9WbwIYB3DhgiaPVgWBjNu;GgjUOm)Z7e?!h; zU_v+CZjz^iUgprqDgM;+`z(|U8ii4(!7TAw{$kOy=`Ru{LdZwF6{wG#69QP@A(+Aw1LE?R73J#TN?S+3ompi^3xk$ z0gwNS8QC$YuKHP=`{iMt*)_Zp^GMhx`AXUq-inS|4X7Ai#)7H`W0u}1Y*#je8?#2y zX$f&O%GDpf70<`IfrIW|sA%3cE zJT+`RBfF0aS5#7QU*T7h=Q18DVhxek9!3&8JC2B;NkvR2JTFdKFpv)k^W*|cvG|FjMGC+!0Hkurf<9kasL zlYv}2@Hbg#I2b%n#lrFl8Q8F24;mIu#m>xV+E&*=rv-ShDG~ko!u2t{+j<*WJ>@X< zBg)q7_bGZUbpj3Dwv9iz94=WS?@YZcN})4bWF;JK#0nh`ws_1qvT)Hj+WE4YrmSwJ z^|I&TNRt)R&pwA?0oCBq^p@C8iDRknu3`h5PC|bau*WCN=`F<+8fJf!?VnSRUuPfX zhmuVB;wjr{f!tJjG2fc`4Yzt3&RyfwruAoBgy{w z4EX+JDLPhd;Fi~3B8+&#es;fQ?P06ID0c*2*qP0IA4W;~cUQ8P&nw8`KBFMBYb99V zSZY$W8E!QD(bxrxNO?mAo0c_%zAw$7SK>F(UA1j6MAHK1dp>}~kLzjNjd?WSjVhgH zHjv_TIkM@$GTIgV8s226)8p2KtmVj18va`-pK79sQt8v^?-*Zrs~^Ly*U5^Wkpno^ zs#mDIFawgecQE_VflOIs+_ki&a_bpCB`ziRppV3bZn{-XdW)wKDSak1>^ueLT2at# z+8=MltD))6CW)Hf23DQ<4rHEq(~QB*bjbAkbmky&?mA#D?0%jP8IPw@!^)M=PjrM^ z44#Z%`ut!QA8dy#t7D*?S0ExI9#eySCwk{kKI}gf35OgwJ?(vky11^VRo~Ll>5~^~ zOclUk&2aiXQj?F8%0f@KV>mas5FLsh!r866VZaqjy6MQMT}}ejU{;d&~~Mab)Qp57{?MWstpin0CLpOuzPXqwT%p z*^;;MJa_97N@15xta-n@r&pOU9VJ(>LQDFwdA$v{{=ZU-vm z$W~OATJdrA8h9wCp16#R!_c1pX!+mA^t*O8J+54XjUO8M716u4MfE#~mK}IkTSrX` z`&+MBVnoet`m@U=4aCORmLC1Dlm>L@(4r$t$Z(xh7)T$$g`yCU=`^IFgJMv5eIbji zn2XkL7lVgF3~n^arK6@s(3FXx6!(^~0A~q(f6R@_@6@NfcLxmpd=+!wiH_wv>-gJ! zM|hKtEmx%~JgVdb7Pm%2)Zi*QZE>`jY94YP^+F$!QOQxL$nK-cpGJtBIvUpPRT1CCWkg^47OXs? z1R4vMP@U1~beDo2llp1PU499u)~(9R`vc32amSHieZaDBs$`7$dbl1SW*G9T$)a&< zh(ClSn_lN&UEh~x7>b#?FKhGO(y|5Monhga07>V^@GHB+l7n;7cG8uuaii3EyIAJ zIb^M=1&%&?2&PwA!t)PCm~jM&N_dLZ!Ji?xZNgKy|LPh9Z`Y>_>iXh{iI>@?JzL4J zmx)aEZ6~qXdj>;PzTmaZ_l1nVUt!bpT-sjc4Ie(vgiedCy&zSu3a>~=>lzxr8LlG9&HQG!JU$ecf*Uv?S(nORNx zmp&nKk^1mJb13~~t4y80ek7m%8$zZWSErA5AEHb70h+nKFDTVn;jX?h(6nb17QVVb zo}YK7zxQ4u*AGNNPQoQzYgomu=U;^XW+tL!%QKR`X##kOTzOp=Gki2U2E4AEr{?{u z;Fp~fbv>kkBV&i)qm5$!Hu443c6??C-S9+aFLRad%at2<e;bRa@z^FPZmOz91c6zcb z5sIqTp?UL4EEUq{%{tWRo>!gO(U|jR+GFNGnU@)T0sx* zlc#&ec9dI(&t!A%`{2P8e;oMmsKhK&L!5c~klRBYMBcFt43B#aYIj8k#DDq_|1B8k zwS{o`pUAzTHMBR~mKOOfh5AnvHo3ik0-a(AR6I9WDHv^m6%mR#G*${Iq+`7tM0S@l5Fc_uoHp31}a zN~nU;&+HyhArHk-ienH}}Iq>oQabf{N<4_XtXOH2MV!FZo^^2o2AF%Ng*{8!vN z87{=vKenLk9ZUR>a*B9qrAh4lNAb&}&Y{Izzo*@mL8lPd;nM~M|z~O|& z75qFV0&i9h#Fty;Y5KzXv^nJzylSD~Ue--2x)Na9<);Jxy&L^?198PV@MJax)gAbfq_9b`lm3v7}d-E#31$3y!UNh-ZExEKpUT z_r<+%}v$hz6Gvu&logA<3F`&OrpF=-~Fkz;W1I@}F zO253VWEU%2Eqlj}K<~#_$oarfs9!k`rrRgr7E>QsQ;^DbH_QPCO+{)PpbzWj$xzLA z_faxMlOI`f4z}Hy#%Y)6C!RYIowrQFBR1=qBw3NJua9JXM$du~d5$DZbtH@#E6&NQ z3rN(1&t$sZ23T#R%}(m?L+|%qyv4B=n)jR}zO0?@AZw|fn*-|i&k@Ss#nLq!#0=Y_ z;dHEc)||CJ0iw=&q3uOI-k4-e|NaYs*v<*`Mc-mlRC*kPoWEn>^jCOvu9{#jM?}}! z4AL>@uH<3*SFmtVrdRAP!40D#;37w4fYC-=m)ZqXSc58{KqFZjQ@WPG8jJ73t0$vC zciu0urYf3fSv!%n8B6f1Ukg-tj;Fo#R&?`}4fN@j{&?nNHsID&`g!*S>K+$Je`LOg z-P@HgMS2jA`mRChr>W8o-2nW*E(JiTKX`B9fIL}@SrQ)u1 zW+aD<$>))~Ckr-xWCboi0PPB-actEK{CDpxh>}wnaiNUOm69jN#dpus))M&rOcpmh zl%+$T??auMbF6<^71WlO)0=~H;FG2!d{b8dljIy+xqlMBup)%}FB6%t+9?pRXeo_u zylu5PtfM zIKICdUhFy!o*iyPp!~lYNId-r zqi1Z#Ew*#;&!04SlY9^pOxIvRNGQAiJ{MQ}xuJU7Rrqztg^mrir$3%WvL{xYIog=Pg4PofKLsVcvUiY`5ONZ}rc~gb zq-Nq*ng<^>Wuaj5L(m<(7bXln#dxBO^}5VewCkM)f1vyV{GAIRN_HH5(R%#<{zrOe zlQs-(J&J=yr9oW);_JF^XgNb3$%G1Aeaa5Di2KU@Ll}I~3WFJzCg3uy6%H>sPlo82 zz>Ae3k{K11Z1DvRuo;|81{vxA^O}H@CFa6`?YD8qWdU;v#Jv0Y3%HP9Bu{>8^R=N` zFt8$?W^P=9R@?39*T1>^YJ4!))Ait86G!vGnHu!!^H8uHkVac3Dbu0O6X@j5cG`UT z5!kQp0-w)mU>I#(KFzxw&YW?iE~z=tlUs>KV%OPt$~>^jJHds?JJGmsJ~lWdW6S0% zWlxR`pw%O0(Zw@UX`a_kIzQPK_b#1X9z)hsnj6IBlk5Nz!XKx>K`%5UC-hhm+`;pdm&cx@NyQdX zKcv3D&XW3%%SAUv70Ua#!+@5P&?;-kKc)`jtRs$$UV9OayBR~eniSi-a}0Ei(Ji;j zA@so;4IDnKFTG)T9epaauqZVVD$HZxk*3HR{T{+d>QKuwi@fQ-zgvMXRE9@4&j?Bb zyXg~`WAuHJ67|xa0V<0HR)78`TpX%Lr;RFyrQ2u0G^2R<8rwu{H5EzZFgbepzxmX- z!GSK2mF5X!i&&?TIjx(&oXTiChlqE%h_`1@T&|9m3-#&u<>}aR&=_aA5TBD|d*Tz8 zGR_CK1}IVe0kLrER391kPtY{zplbp6Sg78*f5@@E#{AY~Aon9*F|5NNT zkMx8?Uh*(^o-W;5w3L`$kq20HjRZS*S>^P4kgnP3%zr`|1`N4NT(29lB}JlJPk#dQ z)Rq%-5LpoYH4x{;Zf9#Pw~^ip@#Me$ru?RPEPvfr%Ilxq=5B89_?CKKzQ5a_#y^gx zcO{o4CoO{MyW)McOJuA0?7Rt{W{<#G-pTcHzE&_rlJh6--&WmOXx- zPF{Q;N&C9|NAi?XVCE19I{D89`pR-5yE4BP|4W_^w$0;UUw$lUjPn4`d)92-FdOub zyUg4={*u#|rob6n#Puzycz#Fch@S^Z2I9qu6%N z2se2xg53AhsijdkJG!(V_ndf*40w1%vU2H!@+T_hAW<`cQC`_Z>83OZ>gXZ4;yu~t z;$zhBD1?%bX(I3BEll6{8`irWfhmr+_?z$*;HYUu!){4a*XNwA($a$6A*t9lQ9L`^ z*5LBbcR+Pif5VWlMF z^j;>tX9RA%-w$k_7h}nbz1WL4g{%6rp?AeCV(Mf-uhpi(h}-TEVG%;ct-8XJ6FtFR zKZwQNR+i|kyKkxYzP;>m^g@x@S4z`IInjd3LDt?sWURLyO`>zod?z1go6(amuG0ZZ zbLqYQ(b!Q~f!gh=e7u;^+NW@fq=a<9ZJ#JQwIFUo(*?y4WNW4zJ%^fR;QD%KV&cRUxC~R6_ALztj;k-P~CqZ*Lwd^a$e{oUPq7NT`QyJwNa3# zAoig8){@l47s2A~1#E4z7M)=Oc~$OLlngB|zlLk*B9%$B`h*vIvQP`mzwCqp+o$5p zHCninqX~^q4-(7y2Jj^>k@&?N3ttZP|fC0S~-at42(2(uIF_vM{EwQ#^we-l349nw`!FrK-JjzN8pWx# zE)S?ZP5#H(dB^4S{*S*!DJ?2(X-P^$=XK6?ibPo%MJOvILP9Fzog_+okoI0G+T%Xg zl@O5-Q7HAQ9oebFe*9L5_o*eti_a9^VUKUM~$${J5 z8APsHoro-6Lj3qX&&;3o==hAuctN2)v3wCo7ALD?dRZpEwBkCNoVbkq=&MF;%p}YF zNdfG_?kTL~?@IzVfv8}@Wf^n?UFY;4PsW9n+fcmn0(8GPfjKE|iI;x2M2CJDv!mfh zP-w{&bn|r!{ruff)itt`2A}OEF`iKA-!7WsSS^t zIpBnLA81Fmd~CmWO+9jAKQh_TUgYP|8zj=yl+?|Rp|7loV-ibMiSxrLtZo3$-`zIO zMg;bt7pl$Vu%|njez=7guhqb5)5X~J?{}d?@v2CBbp>;}{|riqcSdI%3AE? zXNMmDwGqs^mWoQ1{7CK3wFE!;MJ~wYW5R@@G~HXwMy!YATGW_BBKxt_$yWrdv&d65 zCw%gbFY13&$~=u0LqdFS@%4mw^jY>S6BcO1N^3>2xywS3aZns9)%@4uNKqwvNo_^; z_ZQ-4mPK6py$eWb<`-tp4h>W?`yaC_x1E`OEu3+C8H#=s`IDHqm*|0vlh+s<@pZLu+H$IDE9gQtAZIP=y|A zsuRmySA5Lp?KIb*xQbD~*M=>!E6|C!I2_ihO3p3Xv6BP@FUwhkrRV04Jzp=~tqMB$GmkV9Y+a_0N)A`@lF zP5W}Pj(3;8H9ml=4-PSJ3>=tK^}?j-P%|@fYXO(xGK!4q&B>#_zxdzKD)JtriT1or z_{rmBz*NTc;!z@OT}+fdq+_F;t7r=~f!SSUix%xW3njVh@xT``+@G=v zD;yMK&UZfIYabMLR!qmkHA}E#OARCXXa=6Nt&9nM8o+HkV8blE^pjq1lEnmElS6{* zVtAwbdi=w4J<{4Yn`y5sK;Dw0%$czUY#HxL1}+AX$xl-lN3k5H>)p*-TlI*l8dhF74K1OvDXJae2;CeF+Y`SnZJF=++hkEWN%YG@L zgP-FBzm0tR&Hk^g)C?#i(oR%(2MwqhVy)pf5hQ!H=k1mbK!AM z(fai@OwxfMF2~x93GUWrhZc+qwiwpqF^h}X$lI3ZqrGEhq<3Nb;SPS9N3rg_d&_Zu z9*#1vVr!yLv#)qh^}nYJSXJpPz68RH@%!8SW*%T@a4NWaz zLqAEd z(5lQ&jCpx5YowpfO5|K)-V)yYd;1)ID0h_o6}*aVw3b1I&!sJIzbRo$PefRzx{9+q zpG~RtoMFnIEM3AbpL>C6R?a|5N1X5pWdYK<&d1B=tC4}A6HT#vin=3}SS{Za_dt*SNJKT4yh+mZOs)mH9c?JREMoJnYhk19^8Zega%#i020yV$3t z_tBF=QFdv|eCEU42b@SxCUQ_KU{h_YnAz8#2@J!Zb1HigQZ_Qh5ch{uTDuiVN9y7s zBYn>Eume6QAx#eGUBbZ$47&Tdf?1@qi=d%x1RdB))^Fc}*ZbBBmW-652|l-wRMj~~ zwsb4&8ympyO!;T9JjV1$*l{-MvW!nG4fR_}ATiCdQt6=xUX@~oas9bRu)^kS4%7X{ShZ9(iNwv*%? zJlob{2&)!K;Ds;O5z&M>0*eX;Es|-#>S}S^!-SWpNxYDVngN-Ue~h>)is3%n6|C8S zq~*l%P60S;;ae{+F%{`s(XU}`?7nX@>v2MYs2*1qEZfd=S=Nq|@TFh5MWMRr#hux> z>QW%O=Vgw5zSd(u{+o>xWK&Uy)_=6f-wPNGOV;jc79!yOfa!h8G9&XNxW>4Dc>bg? z=JaxVvL?EP_^m9$p-*Cn^++?>eL|eOA3KIn%>pFUqs=O<)v{dMC}-*Od;k~f=d<>m zymR-V1X(yVgE$W;k>KwU_Z1iRoiTDwZ(&rn|jnO$+#qb{2F5$v1%=6%OUU-R0 zSDoSZUDqXwhc9CBtI_Q0vZG`}%Ma|4qK_*5_&NC=D}u{~c?SM4J~==*R=XQ}XiDGg@(1t%!fcg(yrw@9( zCyReqlIcGelID#AjHz!PiajmN{?>NpP92#-E(pyaC#A)Zd+j+qqe_&_iB%PBm3QOL zKK{n2ACzZlKHp{^eN<_JTMR85a*x7a+cYPeAT!UNa(3-i( zcho>^Q!xnz!A=Jwgbk-knQef3r*nwDyT`o@1T$9zM$sfDv~n#??Weds)TTW(3bzDkgEMHxKQX^vm2 zHDf*MBo5Pmg*QlRbBd*-%n1im%xUUjG!(+=@8WaPiDKMwN|8J4B||>leTpw|{XFwK z9QB`lE6__3$Imxak@)u-T$e^OGfz64WbO1Ieg2*t8&ZkS9xKKBotv@NR#g&aFp*4_ za!21O7cRKp3>_yt=R9u*^4FY&G}a4~wI2E;E1KU=zhcZ~Hf_SmcbAaWE0V~6-Fvw2 z>iO&^t19-R-~nHXI)zg9ULc~Ao|Ev+>0EA^DUvxAK-cj15i2At(9n`cOMhx$qmxes z1(oAi>D*7e#5M#+@Hy0m3&Y52VFvr3R{%-iy;A{RZ-~EGFSD&whxObSf=$~!h{D2& zq*pbQciBqOrPXarlLL!y8y0b5R!X>eFrV3~-h`E=nQ|=XTPwhLj;eZuVd*>!NkzT+v2^?u%X;BjK{Se*yZaF<`@i%(K%Lqod?gsw2_5!m$PMWn}HX`_2 z@(O9rGsiJ^lCbZQiQH=2JQUtr!Kz!8v#OUz1t$BB;$)S0tTf|0cv$9H3@fM6(yBKE z=^tINNsu#^9|kuFROWX*nLf1sfbE4D6u zl=Im9fXoj3LF^npu^EphS(=KPS)TmDGxdx2qV-=qP!ccGzUZhW9wg+gwAUW-(5r z&z~ms`n=N{Wix+oFT&!td3Nf$Q|LeSH1zC&Dhi0%&xWf_MM7$^knk=Ezs#va3#wX> z?DaO1eEuPMvuqVg8()uHW-{#0*QUgKQV};bX+6rFD~TVI9q9hTFkDUrqR$cL$U3+W zYy5TPc7?r12?6`rWo8nVfwv@B)e=kg?Z9d__+va4vgHU^zb zvSf|(1jtnSFN%_wf{B?Vx}X)zc{?iOdt)B>RZu)OI^2i_7awxpHKLgbJL8ablQs5x zy9E!aMk9yGwv3eGVf>DF;U0?|!$kv6Aiq=qw(j^)>8MI|8rr|L-OWd(ZgUx19xS)h>Fo9W5Y zF5GkT=gf_G4lUi$z?F8a;mRt_`5S(AwGt|sgxtwlOQRmhWpFicFVk-tm{ee%Url($5lDe5T1r4mK>w(Mul z!Sg%1;BCQ1kNKkd8B5rfU(?yuRoTetUnkn%K9PAk^akzt<--;(OJ!cC&*3}_=d<&) zBT4u7TcpeC8d25EAqmw|#Gb$Re!M-39xOB9bDILHcVj3P9L?h1G`r!kp@Z1VSC`p0 zYJ%RCDY4nd+K}|78ni1hk-0Y204Y2=j(;}3U<^;I;3FF|xxW#K*k1b)H~HBd^xMuA z)tOB}Dk(6;*pOzC?=cESOI(th#btJQR5C836l42zLYw-lyCeX)6<0t$ltZ2JNk zwqW5Cj2e|lXoD*j6H+8D8`qFEr~NVeA`ZP#*JitU4(*Zs=kb3b>Uh|?6`x(9fIR1A zF@HrE_Eh%=F6m%BS0}^ojg4y)qqakM)zb{_g8DDIW}+CHBjkx}ru(xNFE*gaE4`4k zZ9N9@r}#bZcz^VCE$;8B=jbiF(BYryjL=14=7^*aT7FmrFTN9q*Y9uTIbb=oo&<}` zOJ1UsUDMcUPm0i$p;*TJ%mS`vrWNLr1x&=6nM!q7Y_~*k>{vO;(-uG-H zd*bfmxj`|g@z(&ex+oLR+&Gnp$2D{9Pe-|P#UBNs$|1~;%u5_wRF3waGC)(k&!Uo? zx5#I?pf>B?N%ZCF9yF@I8NG82Vy>KfO#>B=@1JnMmyF|)FrT}2Rm~;C8~<>MFdaMB z93p}5wv(|xL1>`T9r;XK#C-m2kA@=`W9^+^nAxIN(A}>C+@Iykv9q)t?)|fo8Q&0! zQ(W8dg$2t5Z;yOnL^NI?KiMRdK4UdIio8%o%?4UDa4zBuQc?D?0mf`xgnh|97X-)u z!z~SkwaeWeVb8}GaQ+e%vS6AIezM>QT~)jl-!^WbBP%-5?zO8>NUA<__oNUT>$n)1 ztF&`(=jGD9e(sd;`flFw0bG@%Z-B_(Ur~O|!$9(mQ)mheikz?~!H}3*X`P`tm!lai`FqgKdK6lj^vl zpE1m^au9mg$UDkayRhCiA2bl0&m;)zG2)jS(Sq{N$g}wxKDPUwAojn0#^q`pC(ZL! z@;8?tzo#nL?x-Cj+#=5&pE(CD`1cZB+pNJZSS(<+j*BBJMhc}|89@Uhd+?XqWwrSO z!vfZ852HSV@5x%%(~GQK&}i>BPU@5>UK`|0y9F&p;t`*jDI=};##uS+vons*g_3Z| zD-}9DQjQH{RMF%Ou}tdSdDIiLGt7Padgf%47M^c(0oe?PGp7c5A6bSiF09y$_C==R z4M#Ib{n0+bzCw2tE_xreAK1;vm5-u3EnPTm)*pPWpodFYEYGYt_(UK%i6Uh^0XQhT z5hb5@#{N!QX=#_+0-3~qT6dN#+t-|qV%GOCcWUBL^@k{|J+F{!8PdgHPuvr%Env~g zbG^vu&sDz0k&Es2yQ0!-^96RASD2WIf4Hfq=d$l0d=bX1p6ySd==q_fHKc%C=&C`&#$)zy(~sQ)DLMn8y@2^N#w z(iPafIg;N^s7Jr7dePDIm(lap!l>AGKMs7jjtf)Z`8UjOEGIV`FBe7lb*(slkd=yT z%u~?^IEtr4#o&{ZN(J|3JEPq5f0>s%bx~Hh37R*5BRZS!${jFUgDu=;anjk@7-ZYw z(f3rQ$>0XlTzQBSpRk!bzQciWa?!%|dM?bvxOKEo>>RREjwRJn53r&LBK_-zu!vd+ zu1YpRZhPNwXj?iO=eZ)^>#j1ySVhqAOBB<^>A3N)9A2iPg_-N+r13%*cCAT9o1f1@ zH_CUQkNoTwt)7d|t~X+$a_i{{I-gPIw@av~@Fh+^Y|c%1Hyt@X8AC72+vy#Pl-V=? zrlO09tFZ7#Wt?W;#Lt+TA^lhbde77{`l6C9qhb7w*_yA9#jfwAohnqZ!qO1DZKYZ5 zh8@qjRMBY8=07n!H(?sP@G;Nv`IUwj?aRVp|CwWTiys(k{pI32OHmNXM@&K^*OUAL z&ov(A=e95>%hy!PmTgCEc0$bh4k=Qbtwz2nHR6WFS2+5(0-ARG1G?WjfPI#B;E(n( znCnkLt;?(MgA{%Sjz0r+4$jBZ49}1UQyuAF5h1wbMj7s#pukq-e@7o)-b3F$=Hl%x z5!n1jD)T8*lcSwRxryDA*o7x8k^a{(yydtMsxETHhg12yymJOANn>bZ|3q&6p?tKx zO&L${O5l6Q1$g@@8?&oby&^AXOEo3 ziRWh{`&aGwSdte%-(CZu70-~qdMdN=Pbn8MDn&NzOv6g=d49#cZ)9P4I8n*n$XKKg zlUx_sI-4dHR`~$Wka%B)T^E$&$qG&Q+9e+xmq7$BmrgORh7*`Ka^mFh%sRZ~=_%}C z(aa=M>rs+(4xKT71ZT}^Lq%a1P)yr7!PIws+_Tg`#AHq)&3_z8xcD0G+kQ=atW1ru z80oG3SInPXs$T4&ZSz^FPg$IvhcJR55YR`CdPQuv@Im{})$G+Bj71^Wd zPS$0tAQL>)$pYGr*m&_Zim4?iKgX0?bFvD@-}S&xhfUGrJ;%|Ucj z`SWHFY-o;mt`EiQk0!CUroQaA7HfQbx*a*e=l;ylXJ#xQ1)XBwG9!0_(Vkr!P;2G{ z_CoYOrr1A}S$R@cyZL`kv^~CeYc${PIyGoOW zy|QGN$|CZ{V+|P^x{Cig+u#)ze(aRn^VmGw#kkLS7S>cSCPT@ESZK6_snv^O>f;@W zPyZkZsOux&>YB)B?FB?(LK;btrtz5(Yjk)~1i5KYj@OCEkTjFYBzEH?1&Sh(^Pw{e3k(f5~NhrPp4+&U|6O~H48(p_63yK|5W*te1^Jp%k* z!4kG8*_>_uyaN@z?82+m{gL*Uvt+HJFgsJE4!!Z&i#P67AwSH2;K?4TL?{xWC%YfP zVe7TbP{B<`Pc=)h`N}6Q>(C9fYr_W&hca+O&}MvRVgRD^dB@>zPc}!;hOUl%LlwU7 za7AY~enw3~p^Y|-UPlo6I+Nmk-_^|0iwrV(?aR#z_(xaBIdZl?-{Wi(19IAuXVhwK z#eRhou|}COElVnYZ`YV`2x{xS8U_sOv3}vP>nd zGM;4V&^TIhWhv5o_6yZlPek##D(uefspO zY1hZyT%E`)Qwm2eM-s8l*M;bbLp=WSCV}Z_kfQc1`^`IMHk=LU|Xh=T+x%6FS z#xrgpr$dw3bt@Iw@>O<7VnVB6KyVMOyPwSL2qVllFH1UmjUl!xh#9>Cjwvu+ zhH9@&V*Y4)aq?bIP*AcpTVKX|=G=Q&`>?I-oXC3SMz}M2L`R~| z8^Ub*sruS$Vv?xY_zWADybMJ=D?%d$SCQy_Ptxf%o6LN8mQ<(3lV>SLs7Lk+{dS!t z?DJfWY#LuOm8Of3(LyP-_ND`>`Im`wYiiIllTcRncORO-tw+NA4&HUj3cZ{dj5h7% zJ@DJgxy!S1*r#hkST(W9tj6K_%%50)mPZ(&f!A3`DKd)PqmYcUE-8{G9}T>8X(jpK zkw8CgI)!WAe8XP01A=H5N%BFpo|}3^oLo@ff&F!J@YU~0AmgOVUhxa#@BgKEyMaG5 zO{9uTzkZd|&^}k=eZ(9|2mN3~?04gNC)~NaZ#vMSoSFPgAE+0_lb-1nI= z&)~URdI#9&U&Gj>Nps19;bC%ot`kvtH|O|gzKi4uzR8#1fO2s$CZY9ENSdOQp0bMnbI7xFX9dwlka3^?{VUS3V-9W z_Y2Xlz3&cuHuu$RDmMLQYoyI%H|`hLdQk|AcK={7aWcYLT=+(=&@gLiF=!qwZ> z;k0)WSj#g-btwGG{tWF#m*uvwApXC5g=ZsOam7tAgf>Tr zb2W!~SNpm!%1G%xXD8o*MVrML-^|b&*U9E6Z}nv+v%Q`_yA|>Mldo{(p8>qk<~1kW z-HGI5f1@j*fHgKAK?N_DaSq9IxMSXn5a+ZPZPB`k%u^<_F8*uKorKj$Rci{{+7imd z+zmkXUvvaTN5zoSb~`5Y^8h#YX)e>C_^MX_*EE6L@7}SCDqR*^k_WOEq zs!?jFEmE0m`{2S{xvb3o2r^;D?cAAfqoO$D^cwUjcPpnJxD7doz82)$Y{E8iV+_qI zW4E*IRO>t)cJqQBG?=G>y-%tk^71xkxTp`UJRr=bo|I=dCWs@yt&Yehb|Sq8yW_NY ze%|N=!&cPsyuC{!5FNW6k(KuZHfgmup=L)*DHOaK_-{_);I%)`i!S#oK7QvSY}NGDX|QD$%Mh+F0XP8P1l~;AG90;Lw_4+_GX5IbtY- z_Zeoe8sgHdf3y;wK5&@PAN51lx_6OkQ7cRzj!MXkcxQ>7pwdT?I};$oN(yDrKdQdq+V4SVK}0N9 zEc}IAqFRnxS5720SIh+Cw|1d1)1~ytcp>VJZ)BE#s;JgU6h>CXq3$#Htt<5_sm ztPSYj79Vq>^)e&4eV9(*;a-CQq?gLK}Y1hj#juC0R{ zkrQ@3yOvx2aSYF1It9sE+(ZT^dYC`@G+ICSRgmC$7q2T$;Eu16WZzpDp~26i0!7a{ zROHRi`CWHoch*Lt-$vWGh^KmFtR|LTPcCrpoW3!FXJK@Oxe(g6yMomjyuv!)-O0wc zKSWZRJJ|j36{MI@_ioIh zUWx}&ecCUm8$Z5KZ>MR2mY)JF)*YrcNt^`v;BffieG#_Hq(k&BbGZD@kNPyfj53q^ zKpp)k4{_QPz{u<})#^>CzhgDjP@Oh7=PrWYNnfc*sdTC@Ss3okodRwr7J+cJ35aIw zhg%^_V4V9(J^HN=2HQ-5XUxDH?So*|UM^kuzQe2Wj|7b zeO(%ML_W^WSVlZ7$97x{~?4+4Yb9&(59th9Lgp{AH zusm)eJyl{D@Pjb;a2f$CoCZbTM_|_2D3m0Pf*|oMgrC)h;R((laU~GmB}PJ~=`rw^ zoB@e{UsK{E)1YKtB+OEafCt~=V4b)QWJIojrya@Q5*h_-k}`nFc>t^LVpufZ1UoHz zV1bP}{kUcYy|H>9eN@hbo)Ig?{4QR_L<;X>3h#L`uMSu<=iQ#teVK;v_e>(3Q~U&@ z`^4#K(?x0Djwf(E9NTFL;bS! z&gQ!?Zma-5OSiyhp;ORz_8w^UG=TC@2wdvQ0iRjVfw&0MPBQPIJKht#|IL80_S00g z?P4&ixCDA}WiYk15xxqYhO)F#DyDrQT=#K*8$i@Tv>Iv*H6QUiS z8_^$@8POiOe<5~G5xhP47g%{?nl;d(KiCM-55HAH?ok9UAGkw7a3tLDdjf-ptH2_B z14v4n!V=MRC>S{dgvx@3@KX4`V=a7ETnmlkG4N$}GYqIcgq)NpF#9F~?HYIC_(epM zUx@BnxsiTjyo5g1Jc(W@8Uq!Xf53RvWLl(Fl(t@WAAaeT!K~Mt;9yoT+-4s_%Dz!> zsC)_CIU!*AW(I9MH-O%`E0X@Z_X6F&(Sc6dMbRzBXJK>Abx5%2fXbBnVD$1NOzpM> z``d@8=%`LY zYO#Q}Tj@Zw5LA5)Q*J!yVw|dx5$#uac6e%b;xJ<-qB| zBG{Ox2w?_NFvXFDu~jAz{ckD6G&w-bQV9r{|Bz~qyic8&Y7B38@wE3q1~lIEQxDC; zs7YN_)Xj;)P`%$67VMq@hyN>}?A}GfylurGxby_Pq@F@^Nh7RU;|CYzIzY}@h}O%~ zrxmJK(Rx99XidhM_C4qdy|H!BB-{!&HZ_A?e>*%_=LH4%O7P9t0)k49fnicK_!r%T zZ7GScb-545$fWD(V&&tuE_cn;r&n}&TC9*M8 z+2=ZH#Jz$ltW$v8*G6#Swh~O_XWb{6*wKk0E9nIl3+WF&YV_1GQ9AN_FRcEthQ2v` zh+cj#h}M={Ngsh}w2inrJkpsA?N(DE!|^nKrbdFCvI%6Ro`!49M_|0J3C@oXKO~7hW&73WE*XG?=Y>bypCG6_5roi{1dfjsXTOjt)iN)$Wme3U8xUC(kM%~ zNNxI&LpjjCl+Dr$lzZACs-gM}HF;77)pRS5`fZ*{?WrlIGK}M>8TM^ddcFj-zLJ89 z15;rCb`dCFJ`4WLb^$k2KggST9p1E<(UA%B=mlo(a4Rkm4tGYu&R=2h-@L8Rw|^BZ zjOn8S^&V1DDX*xT#0Kz%F!=T0C|r|W2a7gjf?Q20>~|NVGo${&k(23QE_WL8mX3i+ zk2JlDQl@R4-hua~Wss!I0#kY($|fX1-+{ky-baESo>vA}nig0r9W$nMI;^RNV?bTF zJDcigy=3tqAso&(Rl)wl60~iV1-)j`YPx-e7Vnwb50)8kz-aP67=HF0=DlnLF;5-X z`mY??cD;cWQk~$CGyu9QROwlb6X;``oxyD=6ACr=fGCawg-A0vRB)JDcm5x>&TKLm zSo^_Z>2g@=7zKwTRpFCYyBbn-PE`R+9Mc4K5e5$Yj)b80Kro0*f`dEx-?Ui;ufos4wr5wt z|LR{@9jr|MdU6YNyFXEbK334x%!0y9EpU7|6B_%5sg{L)peWV|TQ-ftor@goUmOcB z9vlD#|3W(2_9xvoAkN&2{Yy8Ts-VxBhSD=@h3Jge+O(&$9=&PHTKd(;Ao^{#BYp7Q zbhzW;2ESwegKUv!@Z`!nFjP;6h|13vSuY+@AHL;MSI4$d4q`W`gypuBp_nGUuVD>c zyLcs?oDxWT#U#*b(%Wg{nSszDQ~*t@(m}pE7Y?Nz1?}*qV6q_(T^$Y5 zI+@Td`50DbKY;zZ9Zz|M9VFhA@-NoF!N)e2FEZzfXt3oEGm zbH=FkLf@(NRjXlf&KAJNJD~Ye4y2wq3;mfhz_+Q364#}mWUm1XUDJV22NdAHb8?`C zWvIdKNa~>LRmw8gj-rQdTCCe!WU-_G;Dgc{IMurevKP#UBj=;w=BOLU_1&fBS4@DG zQDIoUN*xA!=Y#0O84!*SQDGeg)S36`RPEJ0RPg5`l#j14rPwJ3I}Z9m)6j7!em5UZ zKhS^$s)N+t07-a#brsy!c7ZLHv9S49B)ImkgYKp$cMS6K}0u(GyfyXr-;Gf^X|8KuQ>zy1(@eOCRmVal; zibnVxypwS-D`C!rFJMkCm4%;{|EP}6g%DJ}4*ZnY!k*4aU~@1OTJDsAM|v})-E9Qt z@*3FHbsCOv;`Dt-M_Rw`41KTJj=tofM3?3!LSo-!kkZoR4X4t`qb)Nc&fuuVT%W-(E}sYOX@z=Ze2)i(_BWS7Egxe+Zu2r zMgwl0+75HQR>J<5Un#l2(O?#w0&SjNFzs>;*vbvS?VtR+RhAEX+^>N?7NMnX|Aqvy zI6fpQ}Cc&-jZmQnG4|K69)I`RB^5AZ;ne7fXI#RUE(UtUCmr3-ux52Q#=mc~=Nrp35TOm{RI^eFq;JQVT z{y4J|q*FEF=)#>aGj|(kd9Q@-JUy7Pu$vkdoP+P4!t|s3e;`qzO)op6K;K^11!ak0 z&}=45JAC{C`B~Ah`;RxQo~8qjQ{>>OnKSI#dJIZu?SZvtonXf@H7Mre)kKSND%NB= zjEK9zn^sfMaZjT>jRN7?=DX0AsY7pY)2C%mwt{F!9IPlSfQ}|9`fJlaJ>eOI#&VP8o)2tCd%_<5~VOa7nWx2fL3BcQF z3;ld-9erIOLLYgQ0pd~iuuTL(K<#btjh{#-JbME>VnQHt=@MA-@f9UJ*AQId)j&8! z1uP$pQ_%)>RBdA#_3(=jXm8hq7<&WAG5k*T+&n_fz28qcdM|>ABnHkFBmwKm|ISqj z=v`s}k5#I`;cz%uKahcCs#O$6CtKv%Cs9q-l92u5J>`B$7JAnYQ)(gSs8*ps=s#Wn zQXgYr`|xq7GL8YWcbZ_Qs}43zcHrd{3`3utU|)*^>>8H_ug$7ZaA7&T$}#}!!Oc+W z?gJ^=Q{ms-^Psf-DljP=#MHh7=dP=u`q&KiJ=dVCj%=duT%JRp_g_!H*l>zI*KS3B zX{X^<)=7}FKL-u}62Y=&2ZX-*OpSWWgKyYc7`vkeIo*4}DL)(T8|>rnawEu$I|%jV z9BApr!bh8M$g2KHd1Q2g_R-(av|~E0RBu5Wi!t;Uw_3heTtZ!Dpdy*sG%`e^ECsxI}+kfY{qnGQ9zQV{U|3pLX5j*2}LLfykY z)W(nDl%sVvb>eC)rPz~6Ee{B%#y2)oUz){X=gM9xDP|_9rAmN{*FB0|^MlfQwE#pb zwnMd2Ih35ANT+*=&`ZT#L80n0yug|KyBrUu5DObVt%1}Dt<=Mv@}Sxv1Fc`@!iO3J z4=y-Ca-%l1PS_7lH}}Ds)A6viZx2k&lY>>&6Y29~D)ft*ne@@+&h)DdG_9YJ3GdtW zfXMj6a{YK1jOC!W?J6ATNrP33W>Qy06Q~UrVyTd>Bq|z5Qmrkgsc$lgP<%>(Zu+E5 z*G^kXyRHMiIXs#6yF*}mX9Co$m7(2TP3bq%AHmgd8Qdw#25Toi_k1o)+Z`K(&Q29N z>he!0ZjFHK!XnUrHVhiZrErN^1JBd@sHl42QkC?9mDdz^M&rjN@xG53TnD9Gj zvHc3)HXynrVm{4fe1qaCTR}EAz`|f^lf|4fF_ilYM44HY%r#h?J=bm89ZEZj0#}_W zFjZm&xO^0Wp7Z0B{h0@p>?t3>m4T2+D?(#ODdnr`OgR@k=FhQTP&xV=?!Pmm8`D(i z%YMm#owBJ{k-w?S;^vT}{hM0!^%x~{O_I7>l>wE(l62Un$#f?xM=KuJrgL|WL+7kh zpj2}aMEPen;c^6M{$2!zl`62^L>7z(&%?zT9^kWNDx7K&gD|u0)U0zsbfoW3+WVa> zW8LtauAY=emnhDopWGb;_dZFwt#cafl(K|AExUtG_BP<>Erp;QkJ@n_rg5(rOC z0S~Vvko!9wzUmplvxvL|LWGl|xd2YSqL zBQ&S9!frt}M9T*P^U^-Qjo_($Y1{4Z_ z7Lx|cCFh~I?kW6;D2Ll-eITs;3#?2z@Y;I{7(TaNBr8l`PUwV&#Wk=d`ZUBOs6)qx z^YHkvJ4_R@f^vx+5NIz7P9kMg=a=`?IM+*=>CJdm2~6gOHz~4^2A}G<}dn`a_1S zbU!S(>WQtYvrxM|6GqiJI6fvCmM=pgvqBN?o$N?T)eLb0C0N#2jNwOVkUeaS5p{FS zUM-IL+JjiKXbukI3yrK{cw^l{GqMw)E>!@V#}x?PIEszi*axSqT#R{Gg;+-wc5>Hp z_UIR%U2B=n?A0f-W6e3}O)0}RiOZ1Sj`)PbKNj5bqVBM0vX*% zWvjEJ*~_W1EOe$1^SVBW>GnBjm+8e{a}m~V_zdq~)j(1*2r6T?V5MCIUN<=5yW1LQ ztU3;bU}qSfu)$`#8}N942d#$LSamZQ???=nd>+vq%US5vwnB1)EQ}5*as7iZnswgO zdNmt--W!bd?;?=dXb%+|7ckyO#GaLAqXR@)qViZ)wuIQumEGK&70Y}>p0KXaR`%WU z2|Mt!m`(c`z(_dJZV#?8(mJiXb7;Ih~Ful^uh z4+%4=Z}XVcSW~7uX+9gRp~5DddkarXMJCs=oh`_YXQpA9Z1lc#=5bqs>6fmBzIiMz z8ehZ+SLdt3_G0&Eb)3Di1>0q=;Pl2jxc5C9c$|mQG5%1^i^T2gRah}pfXV%5Fo)Ax zxIjicaaqVhY=YP(8*e6DyPgTGHe+?;Mln5^*YNrfzZ!Sa zv+JS$l2?T{u3e4msp6qN4((W5C{SFF^J`~y$!qs4CoZ&`t}TrS3H4~3uB;g za5gq9@UZE^)ozIG+Yi-0&bU*220b_Zp|#;VB|GS1`7>kKUI!9qsiI)g zH7Z)&N|(OPK*@OpP9y6eyCQYmiyx!|;}*j9!A0mx-oVaPp;**&34dm9z*3zeX!#qk z3!hxry(?zS^{p@~lb_CNdOpD-;XRZuN8rN#7+h*?!Z`gk2n&p*mp1E=b*B{vW>sUK zLNJ8vqA<&99CmJAf`H&1P}vcSF{SbNemapB3Wj2Tc>x6Dhj96iFk89k6$WRyLUrRQ zD*5n=Tt)<7QYMJR`f=FrnMh#*8r&zV%Xo~K%Gw#sJw==SO?`o0o&~H?5*t;6m1q}qQYz0-$ZkEAYF|$zV5;jVFipTn}N=gJAm3Z z+`G&MZ0ch(_PY8xc8AHKr))bsJv|}pvLD^Q*I}k^Cglie(3#K0^ft#4B4Kx-D=EQ@ zYMiiQRu;VusHTS4B5KaLMz+&55TgE#F8x)7iTf^)(=Pmx)5g0tC$v7_hCAOQu}1zn zVg$P&={%09{94VH&-#i2k6l<(eGN0bpFlpU5ylg>nR9O=;*N-6{>tC9Wb+ubI;lWK zZ5hPFR54{^G93GdFxV~64ygTt;K(Sp@9;6`-|e9g3vKup2BZ8+2;_ZuDAD28{hW@R z9`O)v{ap)iJ_Ev=Ne{T2Kto3hv!JDGUN zPPXosG>h5OgL5resFFGao15uaSK5Rl$16E)G6o-tsu7ViiCs-MX3wm_hE1&5q3PEk znj(%o8x{EKTVkgFBrKn1h-lMkcpz$oIQvW3Dw2=Z+F;ak`jz1QJ~}ok093LRY*Yqn zqOYLT{TftO`on8_7|l=~4VOFZw6Hsc1dd*#Lm$5LKGm#3;`Tr&=exov=^{1+ZO5{^ zGKgq*hl5fgN(I(K)Mypn#LvU7iP_|4m5Ai+BJ9&&EharOi+PEPvJ`$ALeJc$OHvw; zT>FXKjjqx&jrY`|8BXraX;{9q9SVPiSpU9%_!r%QWv9zgoEwVz%+sov6)0_C27i{}!Rt`Gx4sOCD;c={wHY%4 zLow3~__(D5c}co#O^z8`r0K(2ZIs!;OYZ!XfCT=+33>eQs~+%AO{(U<_!`C!sC0wv zFHR5GmyFJzT%Fx;1G5DQGGQI~N?qt((}Q>A(o9c7h!wp{MC{-NXiBz$x2GE`K758^ zQ62i$2V-WUG%IUZ!78L&+1*w*w&0o>dw#MQJKNG}QM?P5DjBmo7xlmUC!*d=BqjWAVG$A3Xm@wBhJKy6m6Sg3#wTRK<&-ebX~ZzOjzRuKY$J-+z-->lAG2 z9*r}f5~w8a4@E5fPMiN5g}b}Q;J}^lr1wb#qJDwc`8gh@Z|6f*y@tegxl`}J6SB$C zK>HqJEEqcnI>N*Be8`A~eh$${#axWq6b%Kgt|g|Ph8kDbPyFEH@Yw`}3ZH*XikLMV(JsvZdakLOSc7O*plNj8IzKDUH5!h%DjKdBlm}nQl z|2y#>e|Pr-e*53ce9;;YzRC6~=2R9>@tePp)#xr-A=OP8UP8!BjVAq#O*DP~D>D#0$ro6Xdjk(gl-RJS z6Pv(!4$evUGO1#Hepo2umrvE^KR4=Svx7FUUr*}j==bB0t^0*`RV8*|V+Tr_P9Wrl z1AK-vv0+U;jEXUxIvH zKALS4pnAv-W1(yWpP11fE*vR(y4#;ICkPL-d}r)p>0(dBkzIK z8Xo@0M4;$y5zb1Sg}_s1#6%C1ZdME)oNGWJZwR}&yRM0wuisge;OVn8{818M??p7( z{VXw7sl5j?Dv~I$Q3|QMW}#2V8t>bdq5R%_oIYGiA}62F+0#K#b}Wa+vxoRF>pT{h zY=X(56oiC5MbiEu=!#Be7aojbE4=a%8YRY#T;g`lJ|@iYtqz--`2(_^k}K zU}c48mEll6AB*V>_z`x5-mQ8`vx2qp{O&Rg-!jJLya40+bE$N* zBOSikOv|j==|Ec@Sp^H?MwmTSdKXhMHIwV80Wt|4qUsu5qzm7sC(ozDId~gBS_GkT z&k6Wgtl@ghB6@J@HvT=ThSUaWmeXa+x{5cj*fC>Siq$G^S9}>aRu@Czc>=1x*u$-4 zJUX>&sq@q;lIhh%#8e-Ym2&-)-ZW%S`9K*n6cC`KfLC01w98_MjGKr#@xN%f^dzRf+L76J`ZH^_jqI_7 z09&%=5G*HrN4eiTW~jJ@wJ+bumgrAp6(N!6c+p08GQQKQ2`X@GvZsA^+!Rx(nAt6-Je0I`z+OeV;h{ZO8dx8ILp+_pd@^TTj7 zI{?@B_Rw03LaIrgg#Yw~xH-Zc!isiK93_T*vCE-3x(d#A!mQX{}6+oROc2TG0Ov8YfC@p{TQVCIc$Z_=^BE)aV{IL^ggodn#DP_}h09eb5Q z?z5uFIQ}0`AT}2rT5oXwtT5YYGm{cZ;nZR=7O&^UlI}A}W%>Z@Fttn(fDUE)<2&=cQ!{+Gqkl&<- z`sxRCuzNhRgFLZi?N*c>KZtzKWyq7!rDppG+&R~ZO>$zar*Skh(yGR^QVSd!YXzTQ z>A2i*5n0wZvC%yXIVziQRb-eFR%&96!xCghad-6sAM`1WM(js79B@y-=a3s%=JFYh zP4zH&9}DyT`B>Ez2)7qo@ULq(r2nbowp|Q;*mM<#KMo*dMiZ7g$K$r396U}tkyDI2 z+x6x;GhR@@!Yt!hNvj+yDhtP&UhZ#fUO@6&D`M=#Sg40C>#!GMgMSV2%r+Fs%Z?*S zhtqUl6e6m6Gkj$(&=aPKxRMb%X4Or$|DMr|qw4f!e*<2wQ()^BO0wU1x@@ROldX6p z$Ry9LLi7hO*uP(o_J4ZZuJ{6Np54lGYg~uI%K->KeGDtAm9bIV9Qv-2WZ4|bza?AG zcUOJOm(3QNWAGeVT09NwPJq`+E&O~r4m=rgWK8`^8S#hc z;uQ_dI6MydLyySZaDe{3Q^lwSU+4hGVP)K(kF+x##G5x3KjzFpbMH7L&-EZn%NVj3 z4x-UzyU6F43yF)Y=DC`V$M&aM7?&FX?L%8|$7Kilb1u`0wZXXJv<#M3hoJF*JD(*4 z`z%dd;b!~$a@_CnW-Nr?tVeyoPLyoQrEL%EanG&_|9O-`f5#)t78YmLgBhqc+k=h~ zb%?jEMQm*-%Ew$ok?ulRJ1X;!<{0n~nk?ooxMRv6^~jw6(y^LdkraZX(|KBQW|-^+ zR%5OHE|dq~r^vp?a9#Td{z{)=tDcPo(@*1lqXou}TFAsFfQ?CXV5SO;{eUQ2znt*5 z{|xEHuR{8tHE8>_2#wr5v*wTxzWTn#yyPqB*cFeNOCO?g{B`7WHB+jwp2n%XC6Oyp zWEB5^!td46!8}76oTm%+P7RAUnc&S46(sCYhE2~YYN>X`tdl(SB-c^n430k=2!%m9 zLHTnbN)#(`T_YQB%kSXm`A_ihjfIcIVrJ=aiZ$94u!^(E>{sqZ_TU%TrT6=wbLtFA zJg?%{lTBDKE&%R_ba1qFJJzn=iYohkn7B0ohNV>e!FE=5k-Y}6kvp;w1jVS&64uGE+#X))Iiq)TC=_zhT# zoI*!L5PnOo!-6wck$z?mE}N)e(bQ4>LxNCos=U3Gjb6o9|->1FgRM=gl3>7 zu4=0=>Bi~If5|FV88DySqz+slt6UZ;l5%%v~F#L5Rf|9l)t62;kQIoM=PaAfh zM`2=aCna}`#<}7|Qkp&%dAGF?@Mb1rm(GFYauZY;rI4lEWbBhvg{h7#^c%D|mdXw7 z&$_51xDtI4wa9*54W$?3*qWDLadX5ObyKdB*u!?Z|1*%*ZO}y6+8L0jd_?{m{cwmG zW1t}vH78f1Nh=Bof(CGkF~y=(CG2e3h}6$!xSTQ%64Cv1XBQ7Ea+Ag$7lvpv1=H@JphA$HL5 zFhKHYSxA~2z_$$8|7jJ;Y6>GO@dr7WO5@OXTeR(4iUc`1)UaSu7&u4fZ^uBeV1$HB zZqv4P`{`{}0{VpAV1M;0#yonA#>hS_s0oLhNDeuqd?o9T??|L~1r`@ta~jWklKoGd ztu0Vv^J8q-$Ct~QYq}C^;O4L&KlNGf_a6A0=rfU5hD;=T0-Nj*jE)#h_yjJ3h>|=4 z@9)D+kz{1J$ic;{8ZFZW*qDQcCj564plQhm`0Z6;Zo0y3!)m99-ZOBTViwi`#a-%iqN_s!HO zzn^NCo6^CT3TO=OrdH1mQrQ0yZMebalmh+T=AvLdD$qPWwoU4_PA7LK3EGhA7$)jvmuZ@T)t5 z+R_WKPaTiE3?HQ5`$(uWgLo;o?-5&sSBDgErpFAYa+aVgJqy0~lW^$2G?Av(6;9z>JwB2=dpfssx4u=X-W-t6R<(g!rDqJvJ@FMz$bEFye<(chVm zDCtf$6^2UV{Am@K{MNy_A4{-o@IQ*U5kqT^Yv5^E7{ottJ9jw^IAo2`zH4KkC9MiG zMSsc{nt=G5N^p33grj=MZ|bibvRb;A^PGLf7n2(YtU4#M@IQOrC|hGXxB z@Z4iHiehf#Z=ereI|o9p=mNfcxrW9EOHnc}1AN2H=w1{G=f8I`_SGqfyZofvg)d_OQc2~82+L?(#|m~NvTwtoY@GEP=DA3iRo!eunPNFAElV+w`Urc%1z3{F2-Z$M z27iTcw0J*)@_`&QIi;aP#~Y^~_R#A|e`s*QIeH~=jwHs0(*vVOo;%NgEfSr?6xR)+ z%}s;p)|oN0jbbdB+n4#7tHR;_9IS0wgmXUf&Sax&?0+S>jdwI!NWt z$L4E#{EBpIetq5!zUT^je$pa;{%&z2eq;D03@CA#=GSk+N;Mcc8!{o2G#MeG^PxP& z8C|8Wur}Kdr+{CS9k7Z1S#tXR*HcJRi-*fEj`f+Jfq7r2p-{LT)d_iU@5@H>1R3_h zdpv7dmlIB7pBmfpZ_fmFQ9`%V?cI@sSm1D?~= zP*A6f@@b!``@uMRuvQK$mgOURBn5lf706WZa3)3_-!%)dVPh%ADPG3iDeq9SHW|OX zb)o0T_4scM5V>rKevi~p+4vrsHzNjGG4ik7MpNN3q|e%l>}e@@UgXTKhPtzz^VTvKj}`1h znK;`QoQN|1WR}lq@_xVC@aVt0sB{j3f~bn`n6^J<;U~+az5ewyL=12fzo|k zU)6!nyDp;3AQHbG??7;f75Wr5BTwZPUgjJ@VbNh2YMw^$+Eqv~$!EcP8`vqgBqqh> z77PIZ^`Al_MD%*_@tae`y^Z3!uS{H`0 z0_mmf8%4#KD1C(wYy(7Z1T zyTn{{e+*?;SsFX0mc_zy0@&t23HFa;FF$;VV7b!*SVpP?+q!2vyp5B$)8L<V796 z?JN&M>5^=MgB!EA3TC1K4veoS&P=9o^OsN|9__EijG$IKs9V1d!u$nA}V)EWtPds7^` zD|KOBCxrB#c&Ik2un#+bW9Q6N#Cx1XYRd(rH}EmY@d|&=tVQV;f9Mp?#T#dF{I}>K zT}-S&g`g-)l9grlvpW$`p9ynLd(hz6LCxt_`1*M}7MJE><=kA%mK&ro3*^uc^NXrx z`5;pd z^Js={9_;dNAs}Qgx*{deb#fFuj~;?n8Mm8^kAsHhIavJIkE0W3U`GL`sWDadOVEP7 z(zapR!SmT>j(;u`*n|01#n8GQjK!%A2M2AoJQpcH+lcRym8~5O;fH&AVeHc-I)3 z2X|xnej^xP|H|nix=8sq0vEGR=v#HdWcC#H-!nB9e_xetcrDB#1Ew(9DdX6!$CKG1 zff-CgvH>6OeW1ICPGHl$eEhwYjNNBFu=USs>^l&Fa_;wj*fE-Q-u;A7QEnFF&gJAJ z6YNq*L+be?d_HsmarNB(^Wxaad-*a15d3YQLS(4~3O(LacjAAfJ|+cw6f%%?%nYp%g5M2pPwY@a_j>)1 zaQO*JFMmY2uMCUYCBz15ZIRr29xjuUaNINlv-TFj`&cB}tCXR4qaG*rzQC9dQP^#< z2M>3iftrXprsu_B&a@0TIyS;Y_Yl_Ko`^fHYACtojNBV%xVvl`dvWP87U}7twfzg( zRXs$i=MPL>B+4%9^}<~81TOHRC_{HN3hl??zP11k#HEw5=2dd-KL+Q;^H}ph0<-kF z#NO@oW~SLvY+j}uyJt0vi3rYRtCf{mXwoI$Re|=>>af=gqC;;4^_f{ENU~kvdx^Bh9`m39&!% zHP~5y4{A=YVRNqwOJ^j&Vs#A8$i2XeVUG@9h5o5& z_?rC?e-I-R!%@t1oPf*Ro(tw>m|F4`=M2S|;np1PJO0Y;)fJhB!4-UQ zwZj$BQn-jE<62EJtegJA@bN3;ret$`OdI~pwZ%x)TPB(;%kL^zRT8$lFug~tEm0_h(0_^hMCM2iGK$7!A zxzx=-LQ*)z)qSCt#}uH?`B?^w;AkGerq2#uxjA$G>{vu;n)X=w8HG znMDwsxdL66l%X|L2ObS_U9fBJ2Z$>Y3u z6$=qw@Qmj*w4P!ME|39LlB$9@a^t?zdVj7bp4|w|@+KTO_XI)-XYopGBVH^RBKoZc zkw`7nE|$iPnfmx**-uQ`5a$=UU{}*gjMsC8>*QK$XPHDn=5)L?o=$(0!JXy~a+eav zm9Bc4I)5uYef5&2N{>d`{fUV1mBi%}F{Hjj0nd$BVg6VLgzLy-)Qu!s_%MR@jW1H6-IsJwFl zb(&E~6NyBWr!ltvE`-+PFTibC=G896wjRiYbkagh{&5zrZP_rlev0?p8S$=9$2sS4 zzPRLbzI543{#$m9U*CFy-y&wkj|=c2H}7VusTl>k?rUtR+-U#1z0so{TA!o7_JKamkI;fv_G1-gS zHWpJ;Y959Ex=v~56G&R!iq?%pDAYd9*6a-?{LCr z{0(@F7{||S*Q3RJZ&>nO-`Mi!d+G5#Vm~qWI!XM@NPyCZj~EscV)66eA=_d%&O}Dz zjlx}I#@6Dpe+{M__u@1*73l3v!!zga2#6PA9^1LItXIsvhqN4jr(U7+-TT2ZZs5bX z2PhaHi!Zf{aQ1f+M7H;1Vtfw<3R+O{s|f|7_weP`c$};Z!5q7LFjM=7pI`pK?#^Sp z-edu#ut+=>C_t`f6gs|pBScIE7TO-1cXc%Ll$gdGesjCJqM0n~TOFQ$9LAHZMQq^_ z6E^On2Fu9hyoxc^Shd-Y9x7JS;w{5e>Gh4S`Cg*+wioE2@idMnJC9R;USR&FS1|te z2xHq1!sRyS@ktqgs^4U0QOae`?Rbo{Z~xQn$4QcXR{ww9Ftx10}{0lF}vj? zobuj6YgPr$y9XfkhXA@xjK5TkFgWxD_QaG zVf6U@hrib!V$^asZ2Xdl7e$2_yDJJOen;Td(|y=@xq?19aEyZRZg@;#nBd(;dn}8o zz)BS*dbT(#5QM9T4}ZJC8BqzTF5R(dV&g7j18F}q_U zG(}#~7wcg1k9|N5KN?8vEFalvCg_5b>WmU0lB;1l99ME-pibU z^211EY)MCoNCs}3+acWS1Qz_`SepaltVeMHds*+qtnYKov1SQ*C{4pe-!-V8#E04U z8}!+JF%1r-(|9ETxX4e(%@_NTH~%P>xy?Ybof-TEW6)I+h4|^NQ1~2(rH*{O51E8u zodVdN_zcT=L$G<jg=-bLS#^dQ z3v*FtV>6GzI&A}zEsoRV>g9N6mV(umy(soP%B&*2m^b~VS+2Ht608Q_@zbH6RZC87 z6;u=Qh=hMn#Fs=_a4%Wl**XP0-28z&+qTi<$>&K*cq{cbB$NBYYm_}~MvaxakSlHAPW> zIPMLk!?9vIbd_$>G6QcM%Wpx<#&XcqXEotbAG%{54Ku%l!sk+Qb1~g!>%2ebYT6+UGui#<2p3#)jce z=c*G7q@3 zy!)EWdW|?|@#&$QO z*kviLT@y|HbtkFG_CFdk{~M*gnSiT&N8ZNc=Xsqn>SXio5OwXpNdCtP>BrSa)Rr#` z>99t6!OhBzo@+TT_*Z)7r%S$H>`1cd0L7e2Cq?^6TALj}6^5V>%AqvbBa8Yt{&jlQ za@rpKiMPxrmKv0EDSOWga@6dl=g+5ey4F|9PFhWQwaJu`{Fjn%4Aa;i6|DZyM*%@Y zyu9Znygr-Bl=KE30ZyY4_J_$?ZXdb4hiBidw zq)@3zS6lAVo!C2MWGaNiVdD|GL=~ zJRBD4GI()h66(9u5FDk9@FUYOLu4$zW$E!xUJU13OL2SMZ_fOJM8?<39mW4>@PZVx zq@cKQgmxCqz*QkJj*t38gA#?b!;<5R^;9??{~(teA&KRVmZZc#gC*kEvG60O_l^35 z+|tuHl;8sC?SIK`!eVS%?TJlYx3F912!#H-Pu_l<4tN##=C&N}4oji;c`82JMYox+fz(+dOImF@i7U={-xc| z1#wSC8#_$bV`tJhv>h?WlY}k!(`Su{{%cU(`ULUiA22uh7Avw;<2xtLr1~1w!>+7DmsnK>ps9$gRFbqn2#I z`M_ZKWd)!`cP&=~mLmMwT>M_vPTgYKaNlc-p14Td={$|4c6`o%;fI1#`Cv71aFQ@a z@{bKTesl`ltGemr#bJ7#IUY@u1#vM#3q~R1kRouC&b}R@-v7F&GFt%BZK5a-xSGM^<8#(F2+@p_s}NFVm@enI!wUo&K%9O6&91(YLemIAppCM>cGO@bjtY87&3( z)eW@4^)RG5YEc&54UNwV>}38YEK3SU@?bR0d3l>WR6FU#{!5fsFqif%QKQP0A@p0n znH=OZX!d<+lxav{f$?dQSaSeVUR;Fq>=Im_eFB$9X5;2-!rVVr7<)g6(}2(5t(Gm` z{hWsF*L85c%9K*uQ>egiG0j++PUieq^!BkCEtg25H#!oy7H0$+fj@|DM_$;#we1Olc4zxUe%9ix^Xj+Z2YFu6&Xh=Je*EdJ?+$Wcnr>u z)xutdRua+pOv=BeLT8x)s(y~gk;E3d*v-|nDnA@M_y}8ECa{pVLag=FC8(ru^9`Fw z9^sCZXmE_wV~>&=r;oh$I6%RfUSyZ}jXazuV`9#HWPh%uxI6o3u{3w?=S_r-M=K)=#C zQrLKc`WAK3ulr9(U1==$)cXt4m=s5n8OxFOIt9bcuTiSoh)dSGW*W=YY6IdPg(3H@VB&5C+8OFZ|J2hcE?^IRf6+q{ z7vynp2chJcFPsGiU?(KO`eLP^zg!uk)<{CCSq%$qMR3DHnDak7Q*uu-EzUVi20{s> zcO-$bB*RHS>=J#7?j=#1PFj^G50ykIG@tsxGx`^Tt7-X2TlW-;!!KfwV+hoi@lbx_ zC;dvaMee%u+*5u}VZFKrQ=QFG-F=xCvf>r5QIAhbd*x{S)vZavSHW22I|e3cgO zWK84u2a)pMurw~m%6b>{bYFq%iVC!zc@3enJGkr)W5geH!L}Lkh+TF8{!6&IYGDg- zxDGE%q_HoH2Yb!A$lR!iq5KAF;J4Gm>8%v_;t#nUeo0e2ERiQ9j4ShPXv)8=V3@Gninsh!n=O&Cr>s16E9L|E-(6&tl?yrhcq@>83QeoVcM_> z7f;TFSN>hf@Ezf;n|z<=mBuLSlsvTLfmqf-M7Zyprl4fN*lws(9ntmxbgg=IN-X7vsqc8RfCsaJ^N{3~tpkyNmO zJA*a}^_=N@7CiWD@wG-F{Pc?Skgk!?@Hj&5nneaFFLa*Ir^col8^;msQ z8xyBHb_T=~DCh0!bfbX%C+MHie>{ufYS@Mj!+q@pcKEL;tB_dEo{s5&a)AxzxP&0$ zekjzdgP{FO2j7;qkSj(*qR9qv`{qM;csArS=Rnn22K$cwW3x{$<^MP0&OaU<#kZe# zhOc0>lW%7mMT5>C=|_7%RW^u2Sy>W3UyA6sp)l5%4v>D514QQRM&#gFl)vubm~KUA zUE<;ErH$~~8wKr&9uN_gN9X)4sEA9&&(sU>_N|A<=3r#~_|g-a%W(V8gQIX> z={uSK+z4f*C5Sn~F^Fo~h+8)bp`891sdk2Hub0vh<8Ydp=Sxlt45_*WBwJWO>&ugA z^#+cs{4@dWYqBWXcm)m2I7=z5M=8kR968L+pnq}hq*hTt?~e%Mi|so~F|H=5qe&#S z+n+Rwp7RFJO4AtHLJwCN(eI;^$WV1B4IY??%iJ74#jgMw?v?PnRe`gA*WuuZ0>0)+ z0-g5Qwq+%%e>);$!$brrtjD&w7cf1g1i#YWW4hUI)c?E=zxG=4TOo(M_gz%&r#UW@sh_u(&P8O>{sJ%Bv3VIe zbgw4;ElIS-l&j$?9dvhZExo+jL^~!8z_!znM{xr2R`q~4ywydH8o>v|)$Z(BAqcYes8%idup0?xB@d)Bg{nSLmYt-L(11Wh$%_ho;A3sGr!5%u7-5k35aKJIi2n z=?h&An+%U^f2=voW%wM?!ll0PsG28EH)JEIZ^mfE{3@b})wSed_kkLID^i6>BIR)& zv?0II(9#;FTcfU0_N@Xc-913YoFCs(LKz9SWKb&|O|J3wu&s(f>8f*ZH{`PSu7_gv z(N$1X_CZW_9{diL;p%V{mWiK%u6zuHx0ut$TctD@eugYp1k#Sb4&?m7mkL_;li#g! z3YQ)S524vum)=j~mpmfT9dqGb5Q4)}_i?=J9{#Fa;c~KeV{6VOm}x)6y#5NDwBfi3 zuS_UQMPtIxP@MLygq1-TlzzUz%#k9va$3}(yvdl!Wp$O$k4Mg@GkDb;0iT~rFr9k| zFOG{d8^`Hvvd~wktS-aOm>cjaji#B^RrI!aEIxgIMnX$okVVHE3SKPFv2?b`+HZqO z+elnU^~Lq1Uh4TIL-%C2l236yX|)#6nSJq8n5se>mq*jFz)X_V9pSPkM6hJyJKEJ( zMh25Iao0td`@Kcj@7PK_-WLT4t01VR*7GJhGb+Bbg|3HRqUMGuYT7r448`ZsAK6nR znh-_Bx*_Bx5J&&YHk~tOTQ=O(~y1| z*&7|BpYd9h6c<4rubRoZq=h67wb4uuVQ8Grp^wuKW8hr|>^VJW(U@Dfws#QygICb@ zXAaa(p2NB;src8F2Z5{395=%;VsZ8~d1Dya$`;eONCl32GJx#Dx0K!3O_>+JQtK^_ zeNyB!^f5}9{jZ1L*U+p7m6EEPW`BG9* zDJLzpeCn8Yp9aeEs4XLl0=H&Uar6LP@s_}-PorU;okFn_*HcSj2(8t5Of}RUemgot*7nQts^DROMBu_tOd*e@K$v zsfp6+6$WIkk;|KTVjcbRxJ$0-eWa20m*yxb!`m)^#z=mqNp73*u=pY}oLV6ARDsJO z;IhMPCGomZi(>{DG=EM<#!3xzENP>hFDuBdKavi#T%@qpr({%FN$GC&WdG&_S&NTF zbLd`7ndk<|o_Smy_)M&H5#3zkN2$-Y(47YX^f7lD?bj0_$J<2bRrb@cB6nBJsi)3G z$z&#wKyhWt^nKJyDyum{wh#8vyElgPZ^AvE)9P4chQ7h6!6vjdPh{3HX%JgrcVr(eNyY44(&WV@t+HWk%UY5ZN1xgSOWg?DI4efBt-WzN~+#nng1 zJ=_Y(yES;7nFY;OPtMCAgWbtSu(I8S2h%w{Txb!;X9&X6&KRG@?Zu(uNCG{Cy6y-8`Wks({vS+mK=w zi1yBd7uz-s za$~t#_cso2k0w*t*jh4>mBgi0)1m1oh^f&%BsT3Q$xY?*ouVCa{rY^kBuqrrFvl$T zpM$}P%P9Vqh$nuPc;Z!uj_w4kn|^`Z#{Z&8r^Gn6MG7)vU#Mio0EPR{=Q7e3V9zBv zC{9)8m^>romrTNROCA2+tI_-yE3)|AEAH|gGOPKjH*5nF1f!K^}+JBQclrduN=Z7!iljbm}i zWhPFK9*3eoTt?2(ZQyxxcaV?^24bgS%+_Jrwp0@3R$LBFTMOlcOu#yUiO?KT#eYwo z=;Z=m;(afnlV&gJgLM!I`V^Aa5@+HB3e0tW3}073*%m@d1i`IU(OR% zD2F!_2z$71@A%Ao5^op}(UB?&YB^2ma^|EtZznB#umrjD018-v_TI?JgRrE zaoy)S=Xpe#tsg{tk2Q)e9Jnew7$nd(%qwCgtSdah*JzZ_x|!bCjIi zPQn{;I4t0LNZx#OU$8`W^ca{I4rWi*EMZsWJXz-p8=4YlWI$k_N^ipQ9=(@g&=!V%Z_Vkt<2H5XL&8w|2u$zI5 zV|w_?HLdAlzsXQ*E7I~bAi6sb;yIJ&+7@+WV?yfFl`6)t18)lDoK@DL?$OHt2r0-;W7 zEat;(rgCBk>xk1}I)A4!nR#X`GC!L0zgA(((7`zHQNZdIoYf$89|L2>QTa+A>pfTU ztg8=Poo8WEmlA)*x>G;*%kpbg)(@iuDHgbk!$cMe>(aGtFUj2b=a*N)ff>FjrM3CJa(}| z;wJ}etLN-v)#=F567if0_pFbxBZ-|67`^2nVy6!0o-{4&UeZirmZxE|)ElIy4QnwE zoZY(&TTYJWJQrJ5k>;upQ!GaqjC z3*i#r$~{;kvB_8p*SUpc$y6EaWzreC-C zy-7oGOH@f}bMh$FY(43zHc+Rb7}g)VK#G=iBv*Ww8l3BC;DL97Sm6%pjCQ6d=@k?f z>PkmGno;(_O|)$21!`R(flGh8D3~+b=6(nvrDI00J{SwBw@={zy&He)IPY!b9W1@A zhxLPsIVbit^sQp?;D;j=vgRO%?=~VD1|h$jd(+1?()Az8n4rezfto1L>{b-mN5N;u zE^P0A3_}t}qtk0C`;oMe>FGXzUD_c6``zQmu&yCA)5 zKGgN25PKsIL7(!l;>%f7?v`S=>=fDGa9u{6dFnajCJYqL;@?n0L1_n7iz`DScrz?6 zq{2=8Ee#$ki{Qn5V|i9ZtljbsBb^rJctVP8YWHWuEP`13&WSAi_;eOp zIgwQ^;93Sh)BYN_4V!e0vBG9H?20sT*EpBL&psu+nbL^P&m%YOvs85O5N#YHVjrI0 zV8j0AFpr3Z%-MevTVvDkGoDeyfMu}A38)HaC94_74gaq?oywFI4M28>5 zf_9+!LKAF!4}wO;WA*wNm|5<}4bBBSl4-^!D9SOfRDW!Jl8PtKL91wnY->%pVy|_Dja+@YneD z@fX^^a3)ys5Bef!N{0FluqbUp%2#og!x>78HucBPt+$1mP3;u4!~&fO8^9fLv}RkZ z;2rziNVC`sTfGkB&fIHAihhE)+9XIH)210?WtiO8Rm||#W)>Nw%g%|UaCJ@?{?th` zlcoDu$>tO6RbmXgRXCPi^dG>IGs3WT{$w`5A&l86rLg{UwVBGfR+>EOIvfY-F{e}G z*)Y#CyioHO4t|UhCf#0x?q^T1^KU25lBV;H|Wrt=2v%ghR ztY6t9`ZXp3^P7sf&mj=@i=vUfRt$K@Y^DQ)7+sIm?kFyrpjeM8yfloEBg?#x#Y>0a$n>OB+HH-0m z#Io6x`sF&=*j$3b=T`XNS^-J9Dv&|MO39G2{Sw&8YK*t#ABayH1EBKpq64ZyNEi?Z0z<& zxT*et(lv?%m23aev6p*L`MDbZYWgrXNt{{U(L==iolqN5f#jx8tW;{GsnKn;@E_kv zE#HsTpWk9r&U<9l>iRk}TCm;Wyy#`6ZZoTod*M=46=@$qp;I&`(iAgRwLA=$o^K!NKf~)+$z#kb=I4gD9!nk7m!4#Nt7|DCfEVUz$fbqg@6~ zyLISTc{KE%KgSnQJJz49<~h`EN-{|nbYDfGBCiRytQzXB3COVgQ% z&Ls(5In#v4+6yq~KptLe^ZC?Pg;ZC|Q2M({DtTx{rL{LHsQDW`9k`c-yGQ6M8BRhPVEQ2?4lCbTL$7n+_HgDlS42YQx*IA2crF<+T-Vb0ww}|a=n#{yFx3^@O z4xQtkK$C|%pfhMD?j8>%(fur8aoQ)rw{a+#eGKv+-+-HGJr;y&qxA3~AAM&EY!xXhuYLyXD56fJ_~IV#<2Rx-K_EEJ*M!ko((A*$evBAzyc#@_QBGPS;St$ zXyKobFssXG^nD*}IQav29~ra!cYbWnk5s0zYc!j%G!#v9FXEqv1A^zS5t7S?3JYF5 z6-K!F!~En$#9e=mJd3?JG(n&47DpNR%sm6!YAIGSCmi4A7t_Y?$EaPqmi*&ONYY3d z3VtoLO;kaB%IAcZ@DGNDRhIb9{n5S>+H6DnXPkH;iGHzTsLk&cd=99x-*yVDZ|yvc zQRq+Q!HhJz>rr+=gFUuVXQMAmG9|f0bm+XJi(ibf=w2|6?B>2eYf}{3y3%7=H=~g7 zoiMEah-tbCEJ11j`%o$3d$>-)P`?g?j&ZJt&2V;J`52UA>u7#PAMJ?IW#5XyvgPwo zSeZiWUiVOB^9-(Gd|{_nX^GsIon--Xr!Zi!EKFiDG0Vahju!GP=h$?n+`5j9FI>-_ z1()D<5~1Q(f2L+6u&iW(d0mUaL{(RzHMEg7ob$tforkdO=K(}L+DTs|ECluIgW2|T zdzsvbP3-=VICk@y4|`ku3`2ZQqFrtfb5hH~7QF@J(BdoHoy&o&9s5ym@B*^6Zy@PN z6Ygw(hq)o=dDdyVNMEf=)NggW$fN%Q(dX3%MF07ziM$Hd;|S+HT8*-1fuH*^g|sK+ z@qD&0E|0OzW_wxNvT3aAc_%$4PrPWeeLVebq z3n|VT+$*`1ckn2~!ZV0Ii?xvHAZ;A4`a|M&>*(Ol2BGkiAGS{3jAv_dQK`Hi?Hk;9 zE=e3?%@%PUu?wP4_+a>p9F(SWhLkJUG-H#{#yOv%)pu~etrb^0Zy}XuqCZv?VzcBu zbk)_Pe8LFEAY#qyf5Kt*M9!REfo=XPFuvV{=SgS5V6Gx!?s$o=pExVJYj;`nB`Hhf z^2t{;f8$`$q>Aqpkf)0$HMaO1H4;Cn#dx;!Hu<=#aG&NflrIZ~!n15>)@Q(iO~y`T z3-(rh2Qz$nmE~^FVxGE-*aYDzlH|1TpU!LyjfjJ#dLf2P+KLZF;?ViR*+YIINQ~JG zos$*Fx?F%mkf zkxZV9XvBAUkyG<;7FUtRkAx=Yd`|eenkvg2wUesywOX?ox)ep zE00G|#8lXBJc@B6y3z9U2P~csWV@wb<3v_8KC7oeTBi^re_w}cSObRN>%>vMUw`*~ z79v&rV9#}$V@CV2Vca^*%^ZjoGExYTwMT`^6iBE9;HYCF42^3pkg5-`$bSJ+ zl(;4;bDU=}V_|uK^Pep_yZy#+);(NcV&3X3I$w?17rnxoh0fTxZw{(1RU^9Q0^F4s zz~PA`?ta{j)R$Sf+V&06MJ8-pxfJVB=ZySot#G|Pg8k~5&a$o;vx7Ertk=s6CGQua zWN`*sR#rl{WCQflL@;~(l1e|X#<`}eI34^JNq5_@;!q(TbN}GI&IXdtzek^q#IRt@ zI`qH8v$Iy0D7 zDHPgG!q*X|xS?@~rfn~!3tv0wWx_yom-o;UEp;r%so=8bx{>?80Lr4uym#;ChywBuZ37FtlN%v zP3N$0Hs`tBn2jx`1|me+6=MCfuzD=t3EL*Ysg$7Heh|AdPvGpgE<7#b9`|Fd2>#5o zKX(6UaJCnU(}Ll%U6H*zsm+=+RoO{VGif!7K_Or#+!O*aXrdKXA8Mo>4{K|V?(=1vRM&NB(jlZ^b_~vlv&KqZd~u13Fo9RTr{o2G3f_bGU*(2`h2-Ydx6&R z-AO5D++A9~32krpW7ozpa9Z$%lm;}??+vn;6Ox1}cE@29@daAl3jUoC9@IS8v zm4p4cS3HYeB~3$oq#r-ygh9*F51*|3k-vTfX3d>WOq=EuJQ&We-c_lCx+X*fJ@B<>d$6Iyd=@W5u;x=fYto2FuI2xr#jaQ^9S zK0hsV=QC^qh6}f0dnpV`9zzfr6-VJS6zSfqA3|K)O5yulPtJyurD3nX3U>Ydka6$~ zTuxracH?rMT}j4`gUT@bIv+JDyYWD|0h+t&(BCEyHJnd6yW<)jdfi8+T@j9#=HrBH z0nT{1^Y^h9W{l~F|3V2>+yiz--WfT|e$Z_dH+buWz6l@>|clR;GVbeQ<)BCz5zo$c$xn$Sh8Y87Lxvi8ij)Q^dYRM_8f3!%2M0LZ(5 zZq5X#_gjbGYo1ea)=aK5)N=ln7*m=20)-)txIghI{m~2-eOsR;I-PV|bZhSuQTXO7 zq6*amqIcrv=;fZ{O`H+B^La6Hzdl7sgD19UW@1(2K<274kxidEh}D+eM8k}+=$I6M zf!`ZY;(H&4$J(Gh_z^UU4xl5>2ZoI?n0BWZ`_`H=`>Vs)Sgv>cJ=#P^4(oHj)d+|U zo&x8!8_?9?jvI{@v@9inTx?P(qOhF41@!XF*-#Aqq6&u%?lAfjhgVw?px?Y67j~-f zm~bE!>YpXr_KQ;cS5b3FF*!^i+Pdo_ed^~A?Qh0NUAX|;Lex1|M+`_fL&rW{BxQ+r z#8+&%kZldG(-J6M=0bsQ4|9*`OS*p28moH7@$*kPSsjk2=P;7-HH)FQRSK8AUEq8-7};ez@KY)Va~RO!B8i`$7PuVIOcce-r*RNC*Xi_M0sfZJW=FJrCu#F@|eKn-2Yhl?ITj*`<$N7*qsG$FI z+9jOh_vt4Lo_8LL>awv#cQqUneq(9p5~lIMfxWO1nB6fA=Cr&Gwokihu2u+soGw8y z4)c67&ku3d*S@zyFx0dZqBXn^=Kp4c_Z83_8HOn(B_vepAXmi^o^Pik_QwExGCo0{ zyKc~C{y8Y~8RkyTOz8cz$G-t82>bGr=4~E>wM{!QX;(3p8uC7mNq%ULUyZ-NI z01CDpkY1$2E^U%#53koS!$KEP_k-0U_k3s3y(^ZYnoa*$+pD(}WV-E&3e^G4q7l-3zSesrc0v0t< z%$lK)ci0Q>glt58*n?f?hCwH&6$`#7vxw`D5t>_!9RZ1W%YD1&TQ*VfQ5S6IddA4? z6PS_|gP(ppqhQ!aFTaWKZ0Ay3iCu_AKGP7(Sy;IfLg2iv4jLU={GGFy)dY@WFIVt9 zsCGGa@Qiy~)HpWLiE|U4-i7AiBrIO6ftxcz5i+M7O3|74b7dXguA7Br!zPne^k6*f zAA*tMK5*j9DA(K=8mW{8*Zg5D=Yle8P*7wWct$d+@hVRA)vxQeq`&mdK2A*#0xgXg&rm@GJt#LQy83kt$iQ&o&l8p*Ct z+QKAtO_=H>E!KK$C^MX<%nEh0k!s%peaSSWj4r^g*Gmzpv==k|Lug!jDuuGE^tSB- z_b%2^=ZBS4v{@XTb2cEiWDk1n`TvA7JTuOWMxNm^Qa>F_=NydTUA+wTQq%D6@jrs= zcKC+#4xgY4i1)03gIqEcDx45oolCKYwK1MwofE1;@cPS2s9&AJ`)rgkT0aJl;-VmT zm!AW&cwf$+-S~ayJ=x!!jt4F=xaDSn;YH3k(4&bErK42grO5rN-cXbCmJ*7@50%6hG;pcj#t0d!Fsbhtfv`4;UXD^EAr2B17_u} z!ztsH*m+qS#XL9JCYy}2A3CvML>-h)??YCC7u@z{lER}ly36(1-0@=&{7?}Y+Ly?D z{7bk+*s;H@bD5mvcotSB%bISr;QgpIJk!LpPUr8z!C0Mr`n-}YdbNR7u6L!KPbz8b zMrD46dP$dFq|m(~GicAeYQZmXKULdjQ4`Mu1kJV~`^llgmneajEV@F+W}Ktkkb`vF zwwNYtOQcM5Gg|++jt(r+Mn$6x6uxpkg3BpN@H6Bbju?!x;#}YDAE7?XocT1)VH>O& z&;Or7VRiv?u7|58p2o@MqFplF9>;PmVsqDPoA>oX3_rdXdjXbfTt+@C^gJI`G1alwAkC%}obpxkQ_hAdoe zv*SC>Q*`)sAE^%SrSS*4Xu_D?^xSbX9u3byuI)nvnT%rhIXmP0-`i+D5D2e6HCFme ziKTZv#DHb8`$B$gF!on0#S-H?a2q>|Syam~sTJ*zD9J>;`5Z967CjT1aKNe@ zotrP>VGiLM>`?S54R__0*=R{IRx;rp3SLK{YUW$op0)~2SC=5Y%m_#}gLSeJR=$aX zPULrtbaG<9oE=%Wzb^CalVr&<(s;XgHhlE8`TJ%h*7DrMg+U*weFQ&S1(smSx>P9V z>_fG{_qaT(?O<5R9DVwU;vJfq&fE*EJT#X@eu-kNoHOhn-a?+365DF;!X(x#Wf?qs z*|xotKL6!=_8+Iwuiz~Gh-a(pcB86m2c)(Iq0RCf=;3p`T{4Kd4tR=JvHQ`RT#3zR`M&p= z5j*%|9Q(q3Ec=BcO-V;N}s!Yc9{f9w@yKy@(k)~`*r-zznX)e#*)he3e z&9K#Yb8QbYGk##WuQ)sYyc;Q!)krHc#gW_ISmWh@wK|)y_`4^Z&U+%smUj$WrcsE* zPqNFGfbl9ZcZt=+kFaWT&z?6btSN8A(KaRy&5UB|Gf1H1!69bb2^;$FfO)c2l1 z-^gq@_Fss2)8{lTM-!4#5}328hX(Y!O=IWUQO#lj!l0v=W|V=G9h-3Tj}{a!6w$7N ze5(951G&+mFsR)Ir;&THy>uS__;ufvcprT7>03TfLbvGAWGjwpnS=52m0+HUbiWO%(^B!2jy z=w7+HNIiZMmOF<1Ka)&c2t&x2X;`)51^GQ{AT4!Q*oZ~p*`XLnYA!%x@E}wwE@bJC zPBW+71?*pA3_HR5E-J@dgW(z*YRDOn^yUOqm#jm|SW}c0rclhvJOp(-L6@ll+j(D% zsZYEI)ARQ57JoxeIn#Ymn+<-4Xh3;R2i*%PB{A&;n)&Pxjk~79^H}OoTH8ksef@vBd`PW61vbKJBhL5^O&K-KhBXE#df?r%C6dc zVxy}5WB2BsWVU7F*_as#+}F7W-vA!5){$vZw0y?4xdKeop}8$%IGpyKd}Hi_nZv4cP}V_Gh8*#p@ZKC zN*^O}u#R^bCy!#4x=Ywi={>AytPdNg*Mn!y#w^4mge5->V+LmK?7;YutZUdW^s;KY zI8F=A@-rb9!E-s5Gmw*7ME=&fq?>Di4V$fSMtnZbT~*}Vk!+f@&kffTZ((uQ4_Hr; zWoy6R=id2g_;dUNgix?Ziy5pSg6r;2O<2G8=i%cLhR2f{QD1Wq50ax`Ua0^{p@|0V zO2O)(+RQFcpLueASWu2Ui!0d&<1Q<9aB~#X>Ym4xzUVRKkOpK-OU2)7=a4u-i6#7K z#CGn(G2G63uH<>nrD77ieGenuHx)Wd52EhiaSX4Y10%UdbUA+#T#ooaMjL3`poW>E z95OwlgpYFB*xT>|-Jkx!U%U+af^4y+%M;DJ(lP87zc+K8Fm|E=UKYQnGwSJ-w5vZR zDhMC1(6 zhQg{;8ujZ5iHa?;KHGzH-Q3XiosV<_T6spjosJC+Ld)Qbp!y=@hHk^2KzXdouEfw- zCFZqkCVT&R3(NT9#%_8IV%x?F$c;#VL|Hse`})Ac$qiyQ|44Q3eY$<#97TtAV61Q* z$GxwkN7)OPqx~`8=@f2`iRbr09%jUv!Pc&w?#EW5^o9Z(yJ{Bmsq|!;`HCzvX+J;5 zN06@?@2Ux#kB&#Suz9o;`r+enN>PTLS1@4R-HPmMzBmiun(|t%olQM`2NF)>*yRr< ztVc3xbr(Ym$DeK2FZB2ocr0V3-P|z0JSeI;P5mba!0H1Cj1hDzQ$s^ zY%kSW?&f_)HJH8eJ6waGU`;IVXNXUMp6*$EuoGwIB}y#9O`d7D=5vnoe3-O&;N+eW zu;pDf2IJ*fZ+kdPzT(c-Hyhwz*+$NI2*BY38=<5y5Y8VTk?Cng9P*TdXx<>asM3I) zcMBc2en@W?r_&YPTIxBT~#T#^YjlDZIHl9;Q(eUSD`-Y1r+|Cgvq!iXzr;;fV(PFb<}4=dCg2juM*>q z9_};0K*>F(NPoQ_FP5Ky=c!OEKBxhMq;wi#CWgP(oWnF=BwmNN(&fY)+IRXdWeh(A z>zPy7g~PsVRA&Httf<5+AD4pN;$3`It`x4=&O?W|ACz}`V#+&D?E0yUuA;B_ zu+xf}bNzX&mlj*!b^#9OTgY{NfzXxhO07Ot^oQF4``IW^!#{JuAnFy#cDf*$Pm8(! zK^WFP3pM|8Xr=mOo@L`Y$!UK`wRoapmKtZM_0rCLzi~8pGRqa&Gig&VCY3mYO}L+d zX9FAPz)nqA`Sj7db3QY*3!im8?;FrK}tX!T)CgOdfz}+VRRe^LQ^#?BpFyHHz;qlaqm~hV$zb`(b zy9FEZ;PPSYE}Vm*v*u&_gVCs-pvOe-!`ZvNc`T>xA$u~Tff+fjW<_)qo(Enc^bzl@ zdk_X4M{(3B9VG32sd7A=jg?_BHVg zmIkwt)?{H72JFa7Z8m264a~IsjDwHOnPbRWrrXSWsC|uD*4aB4J$*g}y5-Sjog6wg zpnwvE|7cM}0)_umpiv9$$a%<8%35}c?oSni^{8r6UizH2Kb(jYZzeMFzil-|>@iF#qQR+sVqf z7FkD|{%PW-ejjar?#4Z^BDmJHQOMg3^qFU${SCz7K5ICl^SBPMjqh)K7t_&!rDVp^ zNj)Kk-m1ot(urvlGDA|-mgFJId+#Ok^7RpE|DGWV&HltfXR1)Z&N-xG?o2Lky=khR zHof0Cka9EH1U_{O4N~pGM?iND)c0M z6wd5eLb2--=-cicWG5X-ax&(0+CM@luURb&`{gRE;e6#KYn}+(>_dfTZN5VOCu^!o z7ExyHY@)emNciVYL7(;uO!lvk^DS8@pHv_ye=Qfve4_=wOKCF*f2#5XuCv?vYBMKE0?@^HZyg~YZApCCyCx|6dEvrcIV8Z7O6-oKOaSxO=gkr zm__6swLs|NpQpTRtFU^>H^Eg=ksiFL7IdCQ(Y$%t^iJynt$cKw9(CWL#o@1MSx7$| z7@~!)r15;8r-x56obUU7H2N9a;>oZgyb!yErAoYWe$snvvkS$nJ6*y6F9~ve+9pgq zT`xrMmZdw#Vg!|o?~M*^sT00!eUDY^yHDr)iLp!JShqHtoR<@Tr@l$?Mv57V=u2DnT&eN$i4{>>|KmdQ1vBC3=aWV< zN>hYef3FI!AIb|kYs!T;1;Z)kPOk9gP`a?=#{l~F=Cbfl>lDqjZz6fW<5ZA!kjiUU z(@SqPS~L3t{q1kXeRXs3Rbe@Nn)h(#(PFIUJUp|5(p-=GOGBRPW3i1U=L-HL9ghJN zpgfD-hNRQg#G~}`>r)bGSJTFa=jlT3H#(l&A1T9I>81K*I&>v~g5IzSc&_GD>97cZ}TV)Q~wH)tXjlnO@UeCMrSIFBpknTAiqMIW^ zNVj(Y{pXNRUuM-(-Z)vT)EJ7Jw+GaHgb`+Hbui7T!ytb2vZiNF4>8G7(<3 z?-Kk@?-43y_6Rk3`-Jzg^8}fkBf`S!2EoKVN7z5mTljrqm9TKZQQ=0!2O&Pemkuk2 zk?*Av`aSFvwW>`gbCXqgXmtYnTu5UkGOU0cgy(D{B9)HAhcD@Q)sv`nVG`M&gLg$a$~i~-%#=Zvn) zP7vZO(uD_~22jaL(3^xR;blw*B`av*_#-KF+VE!(?{-}JVijeKA7*4ZI@##Ik1d8L zmxLKjEhsVi(`#d7eKm)?q9-;u2GQNe}#aP!BP3gzEID1JZ$N$!;;(@pBM zbFT^c->@ai00Ro}8A2z+W$9Y>V4UA-jon7J*!kH36IQ8VQTTQ0s~L%_aiK_#&ceze z`B-^jEIJDm5O()G8ixMD^V<@vKleWTIq(n$uAE7e-$iQNk8ndoc&TiQ+ap48^S{j) zcHfWs`uN{=VK;T>93#!WffN*MOqg6jZ`Z4!>8ljJxczIcpt^e7<5-u z;{JnDv@Pd-YA1GLTw)U$_8W#cjec15Qx3NWb0$W~Nh<LMdu`n9${jc*{M|uL*;X zj~Segb+c`$J48cx*8gZgvgp>^6C$~w&7%8<_?>-v`w>!(A z_+KQH(`O>`nE}d+t+D9kUVd&{iyhq8^CZs;xBJ)7kg-W5+i{tyR%BBW?-zVBR2QZL zOq+|EOpF+z?!g5MX%=7EpKZCj2a}T%c$ZHx&WVz7JfC;G45{X#033i;meLp^$;-e7~nZS;S%2&*$@;E-_=j#bnmwe}Fa z9WBsnSVtv;XJBQHD^56Oq43QubYHARxw8pot(%RjW(QE*o{YV>17Nx)0<#{=V6o$8 z-WfRi8#a=s^Fb~u72t5#vw%ZMum3{@8 zr(faH-g77^34m_yU>t4P3B9mGcrfMy)>@_^Q=Fi^^b{7H_`&zInymcSU?$-t#q=K^ zgD(H7{vE8u*^0^6IkI=nx-92$A0Yc3Y11Y#yUuZJ+WvodE!PZ1*;82J7S6dC=b>tG zoM$c@k!bo3rjZW#c~xL)R-r7U#-AOjUBQ%+MzLDX$G*tD%2VqikR))nV|EPmrb}Vf zvR#y|C6BS|oH!q5DAs(F#3kO5L^9$?Tp(ahLo!+yMBs_qQ8=dazvHpT`b0yH7Ky}iefDKaspnpRB`{II2-z9JbQV< zjHO(3WL<%MIQ?W1L|NIWU6qOPQt>ESxdyF&hv4&ucA8&ghp}s&5qx|XqSoBN;}$#o zs%j_a)c$yWmUk;mRL8x;Kj?2rJzekl4fUA{OnQ?nYlxh|qTG#H)XidOtZt!_bHB(w zob%Qm=%A{10B55;r~M6fOxZ7heSbZhmA`XlOJnWWc6ApPSEIp3bXl_8_3PNY;Q4IK zV@)P#-iP*gQ@m(0<9sMH96Az*cRSOdXdsWYmIeqL#MtVwVyyYq045st15-Pc`I;lHTMq$;@qTln)z zY5qqHK3|U6Wy4v!hbn8G9|2|DK|IbOyXQ3W7qvo)zZ;g_h@+>R4{&p27p0b|;ZmnD zq)(_|PO>$8cxQ%YKpezM&T}u=DO@j;;XBzzoG^HaQ9I7#uI@m_dNo<`*aX~e2}Y1d z1?Eg^!#3G5OjA>ZZTG*$y{(sNxuXfvvUon!i*wl4jpm(M=cr@vbC{l$VE&lR^f%67 z1zM|F2G4euZOFj$$j4Z*xEq@S?=~ z6K*fk@OJYXGW=3b&tnEc&{RgAngsSF*U|p|?ld_00@d5_zM$7~*u8Qxaz_`@f&w4X zpKuLvFMBa&z9ch9e1%oN;^Dg|7|VY>hMMpJ9x=Bt@Pj(j8O=G#+UYtvjZ|XJ4?|;2|q9L8GH8x-VLpfpOzJL z$>2Gyba_R$erx0Ke`>hmrU70g4T+8Acu>ge}^{2kuX%twBd))`G}wrbJX*-=#R{3Cr>-XF%hlrdXu9L6tb zrMlc4Quk;;gO-xyeRrzfp+nh+In$?5vXg=`5P zT#LXQsa87qav_4a_Wn+%8S*kupf{IiSJy|v#LtuFEK8ugf4=nM_W>Fo7eOX7_fzYz ziK4?2TSaqpHj93~brIEn(-kG}Z(>h2&!$sL(y8o12(5m7oD?_9(?k_XA(b=Kf4w|U zH7P}8rk70FXVNIsv_mkLP-Q)iR_ykU$!vz@DCYBudoKc&u}M~wraTFz{$n_Yz%ie4 zKCYtkx)W*gRCg*$;{NqSHSAmTk94NKqQx_KckU)<8hj<2daw9Q>=bGlA%bv5YI z^vhIysGUZ{*VByfvox$efu?PmMh*8u$Wg40vewF?{Hrd)e8wQ&&KQ$o$0K)O2u#b8 zkd(e0DxSAc`}_hLe6`tD;}CW_FoaE<7R-#-JF!=t-3WV_PtSdIa6*rJ^(BlU!+99T zIS;XnciPQ}2_yN*_o>M93XObhNE-cT2)hsIz*~1V?3L{iKF=B3R=Q(rbbqYT2`1CA zC&@hW8cmhFLLVo`lj9{Zl5QSJmi}JUt-6@%^fd_28iZV)6Eu+*gs*$M1$}R4`ujbX z2L5+h*y~Z;2ek|rKl#HgW*$Zy?`c`eb9qib1$6 ze**`;wPNdwXV6Q&hj^Q_Nc_qfLwBByQ=my?LrRIem z`wr5Hio!7SdKC7ZMZ&Wn+{kytn&@lvAY&b3Qx9O?%>6Lhos4ZRk&vAvjgo{%5S1vi z?Bl8XkYA%>Y)+Gs~n}Wsa6OmjnoIf8@$=7~3vL~3~-qCS5 z;?3P4m1(s5usNwV9HOE#eso=aJ7sETlK3cRQn*6-6JU&Y#LiS>cVi z;$wKTpbUR@KSfz@64YuOa6=~wHfxU~TX>eg=VKAn(M>@$|IyZuqBu4}9=s$17t7M!|H76lKIRnUC1WwF=3xKbGFOY$9O&J4F+8s#{9l&GihgO z78YBL9w$S1ihiSuJhQ%ME}t#k|3W?=UUFhW4n>=;1TKW*xneFB$R5I?rMx>Sa-Y`a zXrQi{JG;l5!Se{uQwL4M?ipn?qGJ(vDMdg^Dgm3^#v-+*lg_W!A?3HR`!NsOm!;l9<`n>yt_`}ldqGi zvILF#Xh2^Fy%q$zjV6&vzXjGKz6kt8+m9zN>KgJcY_ZZY3uB)ue?5GCGA zn1xHN%i;3J6QiWk5b5{}lm7F7Ug0(iHD?D?+-`gwIExh!|2Iodnu{^4Bu^k*O4x#~>APlV!# zDfbImZA88Ae>geS0ncJT(y&jHp!$9Rw$7V@@j8p4V>t@eUvANvfm^6q`WNGS>M=0k z2&w{q(i{n2YJa@~cfMVNSbr53NY~){{kv%H3c~HTN#s5D0Ex|r;68wA`mg>f9SxmD zhnm!ISWE?<8|tasQ5`L(IWs^ah+gn3I=j)$(&P|FL9j`&kTmx z-`Dhq@3|+RHlcZa0|L~!ASp2!Q?I9D^8@Y&8c~Mr$}?fmswzA((o4ANQ;_icv|Yld zt3!kiR>8vCZF0!2R!5nc5>7@+W9V8r?3J2EH?HPj^rtJB#k-I~uLzhvABk|yzvO5? zmS-H55oA9Gkpq?(uRj$o370ru8@6jTN$Aa9v6R=%$hyzgnFj!siN z-m)9}*XHAf??_Ct?GmWT9H311YqaqZ-`#%=$E|YCIes<>(-)*c)v5#+rJK?18Habx z+*i|>LrW~b((ok`xHg^m8Ql`+(?k)w&Wzs^x6n(i4tjcz@Z0<{b&nUN3Bg%Z_xT^G z*Ct|FeI~~GzsEb5hnRKDm~(~2k@(dVi)1;+>uNfx6Jjtueg|3S_tT`yA<&J|MEMap z`nWpLJoTmyY%X6mCQ0f(Xetrn1ZMzL0?H70-{T_RrD?A12L4s4PMET`4BF?f|oY+Q4S@7PBtv(QL<*aoAG8nXndj zuq@&Y`nOp!4-09wPDKQxx4NU_Y%k8WIWp~MOW2eXOFQ~U*HW3J;_kC+kjf0BsVNvTwORs><3;jE&47#HpZW1N~Ro;7n;c9J-2y*q-vXr9MX zIu0`TxZ&(qeJz&x*f5z*`aq0oANAH670Ig;A)qG+((A&W6(bQiP=OYZ-&#m0yVfkzKWxZzChj>O!V)_hsDUJls|YBHI6gq z{DCgK(&Rnw@W-g$8H+GuhE<#yac)x*9>o{nx~K^2Q1^gw;S-9Oai4a3{)g{54shjO znxw05`R>L8;&Jb2i2r&F8@V5yGM#8&kb`i3KCFwmhK-eAQ-^u7MP9i~>P8bgGvX+# z;++x)o?oiFq{$x1Ie3 z87%wH3APzesdI!Y)#Up@R;bJ}23%Q*+$J`sLYCFt7>gZJzXc|xyXoGjVoJMgfcgYq zD4y2gxfWBLmzv1kIKC`BbO0-rxR-I`G|GOt1WnqlNSGYXR&}jq@%$MVS1U@TyR&J< zmdQw47=aogcXM6Kg>thADw?~fGcq3Ex~pKdM4kD{y+QsjDfno1lFDvi=blv<{PhMh zuO`61N*5Kp7foNfQQl?ETsE39HI-4w6en`=COj;ek00oU`mBFAD;2@cX@|0uuI)5( z=s!ySH4`;uhzuR`Q?SoL!`uSrhO&d=~6CpFp!m9(?#*Y}hGv+{t&xIq_IL zQ|1m5pNEK_ZHRo02ppX`itXCs#~i-SWbYqvS^ETjmsdJNyH6d$GoDS8c9LdNuPoU+ zy%K2jETu!U*O7C}2@0^5z)4pzeC6l!NmImdZ9zTk9i-T1tpL{Ky`Qbh-Oa`bcYyWF zvN>nxGhx6SHse$m)Fr->l9@VGJ$GRmJwfe_pOAjQ*iyTR%zgS=+*Oog9m-NnC9V#A z&f;(%IhtaIc2MwP1Gul>f?K=qL3cc7SaUbZK=o||yx7iawRuM+cM@B<=Lr6s@1f_# zUj?3Z0y?fdfsC@_s5eX0ptSvh{*^61ZEZ{tglxHS_5wp+`=bX2bn`02( zq`iUP9vx=bd>g6D79ehy2HKr3vLdn;4%j;jQ~KVqcf@Di+>QI|X8m68GeiztqX;@tu%9r5M@8ce!=FFMF`Y*#(A>3IK=f=#YNoBD=R#sjIOiXxK&I;mtSap}hgA)sFZugAt%O{T>mh&pH0(Vf#Cy+v zvO^YaE;Zm?2Nz7_EC820{x{+{n?ldq3)*LjasPW1=Mh501A}=HXs?P$-9v75B*AF_=8%as(gyFG8lL4{=ZL@%P|V&JSj= zd3KVPWDdr}*iaPpC1BM9o}mw1il0&XShyq`QI*Qf%efb~Vjknxs9a1sJ{cA3?73I2 z6)zuIGo89A{Cz%@JuKqB@Y$-!+Z~TNkFMfE8^0?zT|+K+rO&KiA+(V@D|Bhj6KX9; z6z=uhEA&~cBHUNYXZb-<2r#|?V-N0*%w7+t{qHDnr!2C=_TW-a9;OX-!*x9)gip&O zpXy*X!n=Su=Oi<;b@SQZ30CZ9Pcwuo256XSDgqjBaGpXqeC$ln86ifR&p69=x;9G| zH)bF5Ca^UrgV><4hcQfK9$nFq;69sV>g6-C+3(NL?4m&0zs?pn1LGjjibU`4$=K+i zjHNpZsPfl1T3X_W*9P2&HSioiIIH4hg9grz&Bdju+DzL_nhof`fnXpSYbFK5ebft< zWW;6 z1*ST9(?y~$0a6(PdCXWHyk{VG=C?asEg^7AUs)eqA6f?$u)ERUJ6XLhR0?#&}; zRP2EB>%oxcoFA*eLKx48#f&x~mff2}p_fJRUYG*oz4^RDm%@D?)8XLNE%33($IWLS z@oEF_Mvkn7rbI3dfAYXU)pP#YjASZSKFoKkCNs!WV71@8(R7^H<}zc}v{8Z;NZf(1 zMF;ln{STl1?ZLTJan=+d%52}aA^zq91!UU_I0OoxE{`u73$WXLGfeA(as9+@%<^-^&Bc9mV*5s{jMxs( z5`7Hg=a|2)l4NS0O|Aylpi}Vyrr%pJ)jtmh$1Udhf^Sft-wT=DrJSE$3LTw3OtkUG z)`XeV*v!Azd(81u#|;h3^s!>qcY1rhoM>4#E@ei;X-+iGtaO2fu!*A9R?(gflQ2R) z4BG4FVUm<2+}`e^_b2wz>aoKieZd5$11zz>rHL+0dP+rkd~g4ov(4f}*l_J1IIedE z8hf}K{?Q{cI^7S6x$79U?`Cs4-5K8R#=OE^2-7Kp>XLL=Tdcr9kQ`Fd&Qto+YC-OY zAS@VJhPhd{F>1p_j86SP;oehG<(-0=m$qYcn+!sY$03=n(KqWg2yK&NbH3`aBMIWn zY-~BS-Ux6^_B_pv;ys=h<}lJ*faxVBxR#@iW4V@0b>BQTOll@m8x2;oSio)!w?)&- zRg7guu+qfcOzyTNd#1URwF&Fs7j>D8Gxoqto!=`zoQJf^ER+`gB->FTIL0^+bnFOr zvR8_Ibl(Bjw_Z5po=QpXCa4){i6T!SW;Tz4)#IBqb;C!D|1^M@6TJdQCFENezz>fmVbxQ_cSw95BPAMa z({2eiojXxScB#Nisf{!ouR?a6BHML9j+M?j!M)xsWR&ws5V<~_irRDNhvE?$>FiDG ztod9tdjZ9FaF)rPX}l-6hPnKk%{&T=P`>^xiZo5wQ( zIPX&5#OPjmroY~r-CunX*V>b?Ge(pp7*(P+ax=z!Pfi`HZUj5 zR*{9!=_I}S*ZY6oIlDKESia;-obEn|D%Te@>{vR5TKAF4`Fg6Y&Y-LrM8yj;1Ub4I zuyRU5LslkCmPNoK(-FtA`e}{&ClsIU=KiKf7`HGThfm8f&8|2cdnO7mZB_P(_gGb9 z6<9{q3oMYBjCs7nlsvKyvit6GZ*eN_l_%m(_!-#jiJ*$l(SV^0^SJ+tvo>y{!dDB; zYl2{6$$1S+c_wD+7M$+&htf@1d~rO8>VRBGEWf~W{F||r&rJ8d_(F%w{;&jT72)}; zRyOWgENk0s#f+toBkgXkplbUE#L6sVa|}XQ|ELA*d1*UFNIFy2ND~~adWa)6-I)A3 z6Z0=5PlwCHpZNTD2-A#i!f1@v)u9@UE-qdi}9=;h8>dbdvw5<}W4eeetvsHs43_aZ4)ZlGTI zW=fqahswm+xEHeo+7)em3Wi6H3s#KL#ibA5>9EaEWN%c5 zO|m#P)`rsJHM6}9q6|;KrNcuTb?INM#9~+@*zOnF3yNGbv;AHmHPal){ zzPiTv2*2kfWACs`2!>n3;Kf9qO$g?F^P^bWnT`}kA2>#I(0_iLv2fB+o;Rt$WxrGy zOL9j1Wl#J}tfWslHvD|D7vH4LVUO);v?@4r*QXDS%*&^@+Xvxt=};UEJ4>pvI;1t~ z4HcT3!~WQ0oV~FJ*Q;DHpP%#f&9$&RW*z={ZbVzp28eG6z_#v55TAXG?->->ytc9I z-}nWrex5B`v_OQdO}ma!Z)KUZ?HIN>U^ttQT!;B}OL)FYmiOaBu|7KnBP{cgUwjIq z?=Teq=I*p6V6dMlYFroLik=R-ViV|2xjjzSr1O1AHO|I9!FlxzEdRi}TpA77m}kXo zC2d$ooh7ULY{*RFpJ3jTKUiLC#^M^CSjM*DY{s)wu-DY!@04}8UCWty&DH30--F|u zr@&B26kF7+*sAn!HuY#QTOMY^0)F*k9)JG6-d;@Ot*=wjo0s(IWHsLlO{VG+&J4}c zL*)CZ2=-Zss1d8N{E`DYPpM#_yON&xi{i^zTYTb<*Ze`R=+3GLaul(}+2#WXw%LKk z&eIs+&l=x!7tV5cj}N_LS&Yt5cCoqxmy9B@NNP0vjW5ypKZ*#tVFKNF3zUu-f!A40 zlso?}$?lebGd~-bM(A^Az;6P}Mh7NW6rj)l9f9a7hz(rB)j{`Q`tk)FZsj3*yfq#QPQZTCQ-~)>vema0n87x6 zcJ4(Z;tLPr$=`HLyq5_Jhe#x<&qn*|7ZjXcO&M9-`;jq)^Iy5++d2Za26Hgwj3Wgc z+CURdCey0<>9o8nlA>Q7r2Vxb_%(AOMh@&n>-}q}2+P5*C(f{vRY0oV9Be!of}v9W z7^G-~xnpE_&)q%ud_8wvPm(F2lW)5Oub1vcFh~aFI|90kS9?nG0d$EUl6WHH} zg&3owhkpihvBBCLfz?8+DAdEBWm0I)zXF48;_P^`2CLa3Wbak};%8bkD$YwnkNXL3 z)hlE4I7NIs{f(p&Pm^O%6C_S*vcxzGR(?;NRp?%V>3c(-sp71i{#ht2ihzbl2cmv+ zZn%vji~o=3k=C`6_ib^={=QE}LvK;kekC$_#ZOYhWa;0O0>QJ!TtRn3wqRLBvcPTi zDvGa^BIzCS6Oee$!Eg<_&IMp(qqms#gvr;~?U2~V-g_#-TiDzhDjFFW zHrJ_7yaGj&wAsw5ChV1;2y;w7j&(yVFu2MIhT9IIZG9|?cHct(=|o%)@PNI?b^L1R z=Ux^$wzTIR5+g36s?!eknf55WQ-Fx=6$rBX37dX(c6P>7Ji5)D!nO7|@WvCdl0fPL z4X7^rOrzyZQ9+?_Q*=Tl=a3oswb8_qqg3WPn0?P&!183)FbNeq_ND&=s@plgX+$DA zj=#tBGI3_BtjluBEm=m#FSN!!r5pLPkTZA*hNPdv8r}(C`1Am}p5LKueXW!-xt3OX zchRojk!0$=msYBkVn*$7wxV$;OLH+}&*zR|9kH)*&`TSIUzQBToTq9o8a~B z1blNjfYD|i=)ATPtqBGY@r}mEKbQDEAPPm>7h|uC7UFJpQr!2k7*XQ`m!WnD=g&RG zMYa%K{GHCHJ>)rfC01sj#Rk6=Wg?Fr;E62fajv~gZh6vBel`{zbrx{kUxveq=kUd8 z7tQ;dLifr)(Vm#e7(H_nQug{IBq5VF&lw7d{oWXvehhjW!jaWD2C4d|sCe3Avhx?B zLUAJkE!W_#mk$2UaS_}uHKK<<%<$b(hVQN4QRSjTwCLVoY}sT?c8Br>A-mFO@{OtR zdNu_!CN)#tcL&mJnoj{2xvPF$B6pB`pgplx@Wky5mFNAVkeu^mb}fgjD~?cSS|oLE z*2L8H#mLC~3K5^v@L0@yX`J`=ewj86EP70Z*R~*1D-Bbhui_j#XIf(J1gr1WxHl{x zQjJ-#C!X7nJAgIMjzjSH5zoGwv%{a4v4QYb*au7F{JNmN^^ap@R*OTtL7-N z_}?M$EL}?7cKqEQQ-z(z`mE);G7FoOjVCJ~l1srQS{^wBh+c>oouBkGH-@@78+zg% zZKSU|huErHc&o4v=JD-hd90X@r-lpe>H#srFO^Zk^*%Ae3ttup{~o)~<{a2Vg=%jF2|fH-bG86_D+{n` zJwNL{^b-8hHl#MkWN026&(2QjhPc`mv@eupf5wera&xD#^T&+Xy*U28_U?dA%3~UT z_pN#Nx<7)pFV#3@Z@`{jZp4oqBb*s~29Zjf+csND}m0`9m9rN1BhXmXV!+QsGAH{Q|xwQW7JK2-@$cpK6D zIB}>ugdp~+5Pc`AX_~qf;&f->CBGXR%#&fmq-EG0uNs_sR)i@kGA#1kC^k^ClO2hU zV!wR!n6Lgms>-dzt@a6Q&~|C2|K&1ghAn}T>Jd_k62t!EA7GMf$bJkAVVBGE(J(a` zo}ay#eAE@Te@;A$RGY?z6gQ%Gp$Wc3#$cRB4X*iK!~L`s2*1xA#3x43^bY5Yp~#l4`qOzxOg9Bm#fZcIw$`ye0s3Y~w0 zp)FZLs`?Q~{o4U%#W}Gp#faQ$hZi^A3Eo{iFUTIboUWLThr4+n6?;`t=UWpNnae=s1{iuSQ7# z=V5tS!G>AmYR?tg!tcxz-RdySdlJ)DG-UHU-=nlTj%E~|r5V!i$TEYWb(sTxl1u;UOkWE#Np?pd8Nl@rKQ}dJFe+rIaSSj0Dfc zIIAE8H?;O6Kvx2~$-$J8<4p_Xi#P{(GW!*v#aeFWCc{i{@gX z9rvi~r_e8>1RQOg#IoG2*`6KmAb79{&DF8EU2vGbNAm3W*XL-z9M9I-9c4kjZ2~#> zLXyZs$S&8z)oYQYu;`&+m5wNCCH}+oLI=!FQ^d)khe?cQS>={Q(cZir6t29G zEJqsA>pe>-zfO@R3`(cTQi~9Ca|O=5vqkh>5tujn;!5g8dSS#JhvD+f(`P^XC=$t* zubdCj+u<16e+rw+Mw?qq=Ey~?v$eEz%EpO9zhOS`jm8! zXS}XO!y`-?zs9|%j>$fB)U$)<2UD={jT#H_RAN7b$1o}|f*wcSGaoQMNADhsF!}K_ z*{O*iaMW?1z#;UJ;N#j18Zt))X%3?~4=tKhb-v;6bl#)1Or;C^v(Ra^izy~AVR379 z>0uG?y4h4>QkW$3QhLtaH+vzLp~cRQi9(Zx3g5(pqHq|WX*@G#|H<5^5fjHkEslHf zY!2d*>{2{3mBg-T1A=M|Pkv4ug)@gkP?u-`$AM1zDe;!VzNO$^;$ZfKcY9tOdXK|j zv+-1Qwm>g0Qt-+tg_;}ZgEpL^dujQCmA_KR*E@#tT$?DrBN!=}uR*!|&hdU99b{>E z*{Q_DB1M?RqQLh@y?4Wtv5Iq+#&+z)`=!>NW=Mc(a0**6pPT26q} zzx&Ys9l`X{U0I=>EK2TuF&{4fQ&5rK1XezewYC1^Oc^nHR~#is*nWa^gc4||mch=- zT3Vpl1c%R!IN`--U7>SWkM3bsxpfB1igLl)?j)2cUqIi#YE-y{WBu&kG-t^n*m)H~ ze(GTyoHYZ_YptR2dkHqV9~Nq>T@#-C+$|LL>sV|!01E|&FGBHHdD`LsP0%sV9N7jn z_@4L!k3Sb8sR*bu`h;j-DK@h3FQgxp;8hf`g)=k*sEur*<|1uM4J=P7Ghz1$G%8Gi zJkP%Q4K`xqt9+Q6l?nULq8l6k3xW2~IDCk+$EQ?p1l&A>iU}9t{ZJm={$1pD>mV5( z{!VuCu~;|L0nf_{xeqxK3pIvg){w(EWKxAIqa$IpdKR`#}&OSl66Vm8*WHt>`{zZn~!6fRZhRh{$D2Y?YglaK7==w(cE?uBd)lTyL`jryR z+Nj(>2Gh9H?9GLE3f!Aa55r<8^>{z|TyLar|B}c~Z7uc3Ip99$didKa;b-h<b1H6T+0M4u6G2KzzY5&cF$UWb_aW;vGA|xoi0Fv}2y-Jp z3-5Q`6CSziAUvT~O8UbulXPJ@tyd94c6|$ZYz!dZVAI@bu#c#uzthFBH()4UN%qp2cxC=>ub?75>Ls=(~F#rA)Oz z_|P6|Tq+H>mxFQGc@(nZ)zHCNF@kPo+?-vB8!2azS-S{EoY_<|X$|vqw`cnurU?~n zf`zM{{e%Waoku4@}q= zQT6^J?1GkJ_FJCOcKwK*`z4rkrXu^+F3SAe_v7ChC8jFQyNew~ob9&`4?0d_NMk#i zMNZ+C+jVRfi{~>|?!6B`g%nSo1x$WRD-t{~Zk#@MA`gSI_87DC zHy)$1kv|`cLs48EfeQ{sa5%mUe_t%atT3KN-ujU;8l{jd!I{ttddcw9Sz6r~i?353 z0*k8P(smi$k#+ddQ;8#g2eHspBPKn03`;3DVU3d(vG3&yEbzc;#4ehRi17*B74j43 zv%Bze@pisXNkGTGd?<$rA!BordkkKq`$;NNvun}k*Nl)(P1dz^J&W5xtfub-#_uzO z!-5&u>}iFwwyUsXs}R|no>8~{O`KZ$A7-RWBH@l7_AYEeEZ@0)C~u?b3lw-yshf7J zdr6HC3{Z3A4;9$NV}d;Yz0a5mbNgTPTX>f$uTRG*BOTb4aIf2mV<@%ygh`x@mML!s zu??%}+JiuvJvE#D791ld?>ssgEeZLJqhUI77Xqe?z=$Am1bmu`sOZ-e+i{CFLmNl7 z#?$(}^JwDJVOZ%T2JxN@`tLukTzM$Qd97z?%}y~qeJ+a0w*&D{y9q%WC-MIKahTOz zgL}h4xV3RN)EW&Y#@!y=hSig$Zcvf z=%>%-zd5f$1wUPDDB@%b^>821Nr4fh2W`Q!jx=05nurgVMwGe{ab8ZTG0#f%^E*+rIo=PSkLpRBSGf5k_OH)`PVgb#jmtqtc?aCK ztVPg-%doWI>_gKA%pF|-g(dcEm%9;j-l2t#yZgu}_yT!Pn+u~s^I&;rB#wsU)53Tg zGSP1*>y3oX;%3PIHwwnEqY_z4pzWZgRI9&1LvS8vhNk8QMJbr{9|Il+Bt zn%r;of`-Z+hgee{0+EjuYg%#7#eu00vS;x{kn>Rs zFS|<(54iKUC>;{!dr_wphWU%5z^;43(04Y~RAthP&zorUl_=VwS41HJhe@aA9ID&A zP_SYyM5MX%_rGBL9bpI8fq788XM*KtgHe!H3K5y-aPpst^2M2}bIH=K|4>neX%@!odc1>RY?gZC0% zu)b@Fug(^5uFHee<6<1SHyi1)LQLVm<8!}mK>Bqn_j0DfRdF=P{=_;hJ+^V5E@MeNv%_;8({`&ew>~#$y%>%< zA1630-U(G*AI@^@pn1WO_-{%gB))L=y+}D0+ucFOuOxi3ZW2yvc`i(reknA~8DwGN zp<}V;_Y0xhHQpte>xq9!(~x)b1Wt}D#nM@!aPJMqMCRx6&VR^Q(BT_%$Z+tj+?v&xjag#DZ>DLubWg`g=K%dXv)V zW%6rEO4r1f9g=u;rA~17rxL9jvYr;`MUl@%HN21Gp6Grf*!IQY+o@}Kyz@M+Bp!o| zt_TJR|Gy)VLj2F0l(O~$ZIHM_&Zk|;ATN|8=guHqYYR$gG@-`R4wRr#FWBYSMr$5= z!(X=@HdkahqdGZ#CO!^#&Fm>464(M@Tl@JdMc*lPw`=@-P1ymawq82HXrH^ zV08NYc*z1h@_{-(1+Q( zQ0LW%^=?C$nn4eC-sAbB+%1^%={^?SUc(ZuUt${3_t@sf1U9V5m>vFe6e+$=|^);S^N549*;BV*CI^@H&7glod2@MXfk@6s{4N`lE>Fk-r!HJMA! zFBp8>hTgwc$eGC zyXOA5!0WRF8XlsKFr9+8#T78CoN@OJw_bi6)_3ySIJ>{!k@7@c%>S2adCj%42UKLgIlYOA$p#1U0N|^y-!87t_X{L zr^|Xv9GH#ba8|P-7i(%tp)^#9<;>;$7Tp$X{h0%Wx4gf!KZmNP8K7A=6fHg*V7f&c zJ3BRz(z6~BOFyD#awjyW)?@nBG;F>;6*=QnSjK8Mc73lan;$Te1wW8xp1Mi!OuvHT zg~QkukFjitqb3VAl4f0>^3mJ70n0f@rm1Z?@@Cj#>+!KLncho-TR3C2Dhh)Y%Q5s> zGr|VaFe%3nl^xEkPCc2OZ+2u&?jzXMu~xB#r@m{ zr>q!cd+o(K_oX;$G9H#aX=u2p#ICw%vNiEi?4sHOc)D(dQ?wZKU9gh96}ilQpZLW3 zf@;~+J{y*wri{Z|BOq^Bf|gB(p~!b|&pChL;|<=ObUlP)K}AsgqRloxS7ps+i4bY! zb54Gj5GwL4^r(HfHP;rJ^ZH0jN{N}59A;};ZZRwU!z^pA4SVyn1&^$@Q|y)!&S?Bh z<`cy*uuTFLA|I(~VlgV+U!u075~Esubn$+;-bOi()EPaR1L+l4XXG#qn+l}<L-JO*vZb`1bQ6 zR`MNbVen3donSGX+Yqz78XAMXBkDsZ!lj!qX@WJ*IjtmP$v-r(ZU){z-;UT(k0I1B6GpVe4Pv<`dD z=W(0^;4Wkt+cjC+iD;b9SWa5)SExAB2s>Z@X zwb-)5`8Yge6nid+VGG`!V6J%%%rIMn*$0&&b|468vF8zE&K)&v@1SEO#qxJOfm5y$ z+dgg`d;Z6axmWgMg7ia-c$1CB?LVmD)?#!zPsD*-5ANq2j^maoG<{ht3;6woeLU65 zE}q=Ms=}RF#w8_|xMs27bytR2 zoayQZ1+p?Or}ygZboc#BvTtmrvRCPJ>XH{de$#+bzX;~O@+cc^lfrzL3t7HVD;5m8 zDVWmwRiG9yf~p^TQE|~?8ZLM$sA*1s;?n_?RL)_+%D!w+fE9DxaSjun42D*$Ck%QG zVbmiHGd*qibCz$Hcjp#q#ehxPa@rD zRH?*<3qgF>u^7+i+9C9K6#m@sLn#Y|iGDgn3UwgkxmeieR3_XW(IO!4_9wo`5j(+Iha7d9>Up{mZ-r*MAs2z@r=mI&J;pN?VlSx~UhfJz-XvvX$@ z{kf@uzBk0VV>*S$%kmj7$rs(`*f^ZvElFv7!`1_v(KfT7#jr(bW$NKWv`Arkc zN_`+|9t2^MKj@z-JPc}S%z8;Ic{Unx+WZ+~WPz{G@6+p+F2RgGO%grvLGb+b0t#I> zmoBX+6llpjr~5l6AUl$GOgzP~qvsR-J>O2b-Tt&eq>I)l6;SN@1QNNrmZFB*(9p*` z^Ec`MOzqsjZdf40VIQ{5)q+PrKF`d`viwsv?8HR}<~U?5yLsmguD!~~FX1(~HZ`Ju z^>4&a`~eMd?w`DQK-gmXOc>Bhk~bna*gtvs7X zCZjKsEpws3$S6|S`;F%9{6c+W-qSmW9x8s;K;wDNbm!K3tZr(?zvlPoH2aI77YFe& zTpQb5=b-c>f6rfT!s9ar?BU7=zB`yGZ1Rc{&ghO8&hOnLJon5-=yu~Z>!XkKTWt=$ z^SSfg&?!*l`|!XR9o*V0ir;P#kQU+ml(zlwF5iSfWm6%3tPk4J6WGH69X7^8jjb@2 zW+!S>@Z(8}pr)x;uyOoy%F;5V3XdAWuf;Y3TK)x>bpE1#)D+g@sLV!`?MHC588TF_ zqpe$n&9Cc3M%HVdhw6pbh#2Gy=s~}0DJpgaK(f&XIZI9OB2Nw*{^Q@j8EKH(-iXMl zSy1ba#PEhOQ0g*9tB5BYnKOPL^Me0MRh-GJqUPJ2hkEBGU2@E(!+{0lJ6r^ni=wC~ zeIr(xl;QQ3)A*T{igEk77l4h#lZqy6^!$z~zdync((l;HM0L`#jHG1F*C%!)?ai!wUkbjw$aYQ zcv>JF#Ac2bV@FKIIWu?=)0`f~onafH@%AeQxqgOM zqgBN|6fxR~`Z@Phy>%9~9-lz-o}Q)f7d0d}@{qbCZj+gj9o2F^aDco4=1lW}!{{ln zxFUk*;|i#XKf@3H>7akt39Xl1OL?&i$ZPR_D)N4iCpSDD2D?*`nZ1O3GK=arYQlDa?! z(Zku(_IlbHYKmJf;wXE=`A`Qmk-(kPn_47suGR~S^b*i=^&C2ntwW!U07WY_A+f+5 zYYTL-`8)4fw~4_2Q8N|)y^o-2!`Zuc5cGj@*EVk8l$jP zjYDY^{R4blDuR7w$`fVu1D2CJb$)$9mESt7R?;)S{fmFXtipKbe!RGTkm^bLd zB)W$z9C%lLN(`+RCy2|&`>u)aA)89t@eHG0T8xdMN4EpS!D40E}Apv?LoEuN)`(f$+g#$X>B zmk?WGD9hT1I^t+4->ZqoVCBXL=qs(l4CCeycPDt_TounkFI9)maC3IOViwc?X2z1u{>RaIhhzP|Z=4X7?AeycNR;O~ zuSbcbt$a$0wu+LZBn?fJ5mH7GNmkJyQ3-9*LQ93BqNPE5`rg0a{~d=8p6C6(@9R3x z*J~c%k1^(7%fz!%aVlO7%7j;y$Vjdhb0?2<+)%Sc-R&~T=5=p^|7)>d5G>1k zk78lFG7YOboxst<_wb*>MSM4DKu^yXqKA`@BWEh0dS)TC0!JXIemlJwstY5r-^m#y z7#P{V=;ebP%At75EaO1>5)cYPv>|FaqQijKoJQhXRI2bf+z70-e`)8P}Xv{CTlDdNpD;aaDz+EK5RvOA`b)#*WiBM(~5JscfI0&a-kh z@>4JHm*ra#JO3;ey)?mn;cQv{R|_S(tI2gYm?GX8Rf_bw$+gZ3xo&2N}aAO!@m9y(7hQ#5@tB zI%&hoSRb(?1ot2-8{NJagWn{IXT}j)(Nsd)R%*eLCcPT7$Nqz=VON=`A%2twqb+?M_|hB4BTnXKxY{*Jb0;s zaRnW8?4uo&UTsD%moU7KID}%aRcOyYO7Yg_sChIT|CZU~s^J7&E}4Wo*-wNgxML$XeFIYUw=BeRH?^;sGI73SRY|$XuiQ(gNuxCvg z%)_1Vqf;qHobSnl4ht{k+=(pHSwlFD_F$sO<-NBb!GliBW|xk&>^jzucdr=CUQzz| z@GuGLUlZ|V)jlL`iolPai%`3zK)Sc)G?1yVSa>+9G z4(-WZ57v>q%XAD}H5+3b{g53z9U6l==<2f+*o$86qXT_-q3<{@e4)(w!P}v^M-_QX z+)*Yo69M_F@m_H}J`Q_DGesugT*o%pI3!|=StcfyreI7bd0db;LgTn_R9}sUzw`=L zoq3C2OLK5f{W4OgsIc;YUVO?VfPI=*h+MlKueNFA`_vv>PTQ}VV*4gK7u!3}EC zm0N95Zz0&9s}(uxqz+G)^x_$^MVMe?hvOF;2oUTuQ8kvq0b+k)crdj#`M zc(ha_5tnL#uF37B5$gek@!_al6p0YAZ|<+BCS2yOxb1ToL9);B_qZ&t#~Y-^#^PoF zzN|55J1_DL<@IGtSZ2HxhX!}xdyme%VBl&#qqvEKUeDu~k3J!1?HSa5cLnYiA+GQ@ zwrmyo?O{s1z#$JlxjyV1R?d;%ukqch5v;1zgOe9%@+R@~zg4`&_`^+j_n(jG6=l&k zgKAQY7Tl=9O=!O!g3xu*7*?|f8NEipuuevJ4F%)xn&5!`JcRm{Q&1~*h7sX?p>QMw zom0+YotQ;_+W#CAr-*a!cUj&P=fIQNCvp4pK5Sh54EkF8FfF|sj*OiP@8Df{ej^>% z#2m={+#KN%T8nNYlA(55@e7+hEi96B1XDxp9c`9}=ihSSa0~%CUo9Kjdro(r4ifzuON{Y&PqV!nDYjS<`RkRTZ?^>NO^b2G>H|(WpMjt082o3U3(3dv zu<8?sB?pcn?Cvu>pA-UH-}&58ah4-Iuk&lUB;I<-jkDj4U^T^_+z}DWTTYg+SyC>i ztsBK{C0*FV+6l+f1=rW-7?PT=LH0u(GDR-=L+2xi`t}TiHW={Me-l_^uLIvl9!%Dc zMg2_S{kfBew0%X`nOlmJ3#OnlvoqHnapSnft9fv{A8(bYaIs+m0<}W%`tEtG)Tu|f zyD9J3mdQ_a_VY3^$4vNQfJ!T1-)@mpJFADQff3|9UHE4$?2r&T9@no34^5ps4)?uG zJqxE(xN<3u9ki#IUBjD7GjT8AMm?**OUHVTp*{ zehgVhda(1(H12mii2ruPLObsRe71ICht$qIA*dbQjRYS#qJiqjq zKT2oe0ZkPC!n3egbR7?W2>)xM0&g8F!ZM!v!dclv%t~YN+BYAY&FA3Zgb7sVd7V~^ zUX*uU7>Pw|=vDTBTC)Ldw7fu;Eoz8fE-Toqb#$^kl47*F;Ngl!Sn2Dq*;5BT zy?HukiCs?dL*b4QKA62%V-VD9BQyoipt^Jy-CvlGm73+am-z*be_!L*V_We(u;jIR zo+!va0>_I5(C)krkCw`yHy@zyTP#ss9t_p52k@#b8LdCoprAkzD?(<%_>nhu>1>1U zZSjAf??TlwQi*e35sjWvN+%+PBlYhPOn>o}YT6c((cfDX;n^Lp*1V?Z>bo>nKA3d6 z?4ysf_Jbtqywb&v=TzD7mZS!x3))WKC)?pZ;w6HwD)WD*6uHr?6R+|r!eq{{8MCv9rSq~~G6F&Bj?szF@pU&H3Mci4Ja3=h_PfwuWYxb#yz za}C?@HbsM#bzJzpNibj68pVqXzR=RM!%_ds3mNNN;dyu#@L)0s$p|WMX=4o zBCvkWUF2wgg)&u&U2h=1&u$i6>>gY=$%so0|6yHF2y$O1!EQ-7mdU-u(N9lt&rFe< zQrdC)%WfDY{DQg2PTSNj^ST{9xXl}AH1H>MyruHPGiu}Z!JDGq=|bG@XzlQ z%?Yfcej=Oa@c28mL@J0r)g{u97yY#LJ8@=AI=D7qYz=Is?$b>1>52t{ ze|E=D;~n(5o1A2BWd^wxKc|DmFUjkA2?ci*e9YIIB;V&-Q1ySUv~kZ&NMc4{KynZi z-gL)ZISrB=>M0xxW085;6?4*d;;P7Z?iDV8xR!&o?qfFb%=yA8a$e%TMN{&`CK78; z-GzM2DO~?_3E{8K(dO4n>FYohn9Q7t^E;O!JYKLQ8?w;S`5$HW-Xl5_Rq#J`AMvja zV^Ou}Fpf5+UA-rvrH34!`en`aipyAjz+9F|%*4oFo)~-dBHpigi0q=L*x0)grP-lW zx5``UDU%^RF27$oipEM;#FVjX>`ZpklM{Cs4;*M;gNv;PQ5HD{{ynW}@Z}-kZzpi6 zR`82=d_(T+VjTQwjtYHCwjUbK$*UvSU12hNZP<@`xj9r{yi)W<@4-7jgB4B+U!;)& z+ZQOqtJf{K#ti1!=SK4>%LzPLTX@2AcawYQ5gN3+oMKwC$uYQuemz-4>q<)`Ys9(k zv;9%Qe7}l=n#bT@BG@y=Cnb^}U#M#2UW}B>z#~23U%h@8uT{U2*SoW5ecPQ?edJl3 z^f}m4m0wQVhsu;dF2w^5ReH@o)jqQ9nphr^^#X2*y|L)ePB^cxjkv#pDlb11iSh<}-nDcc=SjwK`O^v<4;ResZ3F0E@j=`ce3kd-2k;1sLA*Fa zivx!Iq<0qbs5m1sy7M&gqhA{ZZ?2*@2VLlTT)f~j3}>rnuAJO6oR^MV$*NUf5$)TV zSIuT#@zRoGwAwK1pD`N7Y@=TpK6q4d8#kxDLu;584{i8@kyrbmO=p>y8C*r-g>nRW zhsDfRr@OPp+~S!FlIAP_;A*tWL8?5 zz|H5Zd0>(kOec(>M{gcL=7u7BkNSbWv-B~o)>Klz@3F)n)B$SlZt(236ay`Gqv7%- z$n6%4*G)a}?yxiNs%?Z_V?4IIeY>L@T}yu(;7Grs1WRv3^7AH z4x3AMkbl>m4#a&TnZ|9fsM(1=>fyK%?}RJ8gC!NY_o#oHHDXu0<7C||EKU}mr4?f4 znkhP4Q3`BkrOJ2PyR!el8tDEvidLTSLP_ONR&aFXZr6PIvrIT|`}hW4OEfWinDA}) zjzIaul{l9!I!-aSC82LOQPTM-&}#@rwCN$-$vKJ3+l{c`xEt>7I{@XTc)WHqhg0u` z!c!AVzupOV!N~1g9g)D7K85hHx5^x;DEbk?D|0&E3iVN@m{~a-^EU|3>05CYQ`F!K zA{Lwj{BwI8wMdzrr6*JTHLmU!@?>qzg*KUz^d3M21XA^p^Sdgt7imiV5q z&aAJ+qEl*o)Y*VLrpWT>sncOMwl{4XD35?e`DmWeot=_IzTx!&mK{8avtoYG?XP-p z-l>CmqKEBPk|cTB)nRD6&RJ?U8IFxNAJL2*j?}LB$GX!o7rN7OL=t#lKYf$#B)dTx zbR+zn(Q&k{A%b-i0L(+UiZbLoiovMfED!`cZ25qONL?N4Hz}HVxaL1Hof%) zCI_Ubdy<0dnj)hcy$0j*9gtw)O-{>`XiAMf0yiwi@=mibVrUe`pE^U;2`yB5Wi*O} zXTszAM&T#!ii80}>EtFUH?N80RsDAIICwIY-v~yiDan;AfYBj2Zn!y)lU&?+)^!)2 z7?qBq(0ODuwVa-b`LEZW0xB!?q1c6oB?mqFVpw?({(O8Q7tPqlTRmfW|6VEgdE1qj zdavgo+YolvTF6F~M{sMfDtQ#j6@@q>=Rbr ztcRWOc}tE6*2?rWiEO9y$S=^~)sB7mdM`6>?{*1Sl4ptgdjiZ8;_+(TIqZFK36quz z&e*ZqGOcnGrNr^0uE2G(p5d8G8U zlBP+&aM^MLyEF!~%7)!se=D4SToWw)vIs=0G{V#3CQR?|K;;I}_jLIvId;tzrU`qo zz%P@Em+p~tR_{fcpUo(Dn2vPe$=%YjUYSy&K6+%{b%MAFTCAL+2~yl%^6$yAKsm`r!_0Y3YWx-h%{>u$_Jc%%QM_;ndPP zhk6A(Bny9C)QxPTpCYTJl_rlZW$NgqY>o0AgK%|I7VZDBi?U0_{org6Wn1La3(I!$ z?pj8hO*YVk5(B(=Hx$-}eegE+Gkt2(qbUU@6n0#Zg0>AuBekIRYXO=i4yfI?OOifj z937u#0i%zCQ{QJH3ZM1ImW+7neWj9$+Qc0BjV$_^MbZ6`aWo?PGC7zJ5!q8+e75kSBuQI$Nu@y+&UE{xJ_h@`qlf|pmmZjT*2y=(GM}Q|@~s3`X$N+#x$i!^254d>XZTeJlBcqCk^0-tFGaStOIuQ0({D>#;6?+@FBDdC+{i4 zNF!P4$Yu{|>T5r#qK&I`Z}U)T{DM;Mz1fqJdT61q$jj(B?S)Qt7$&3&XVr?yl-B4E zgI&=$xMml6h)mJqsSG|dS8Oha_zV_ z_&J;|RG@eFK$u`GMJ>H2xc)b($D&^pDV0Zb{0_RLv_&%R+cM}r5ZSAmBd}S!4ac9$ z;#zYU)CV}T)ofE6l-uyIWj7XxJ${*CCoVFOVRhs8@L6~P##ciyKCcW41xh^Q zn=Ee>?3G{BlHoVU1?O9eDRnX9>Z2LBaDOKD2)1v-`esTW$81->o!?FK;HMuQx#nF5 z+$M@U*V0;?8feDdrnMp7;Tih39>7j>FUUMtMk9PL(D#SUWO48*Ny}>JTE94vqa8&3 zm$%TT#Ug{-Ne8<5ziH;%ZDeJXXkD%LkNlG|(d4-mJJ&~}&suFHC5}R%;Ld7J&cpMS ziI~(Bfqa*-ICaHKlHdO}`OO%OUQYHn_tylEmYt&0XID^A^LG48O2vH7Ks+wqj_y4= zL}qdT-Fmc#uJK7Kxb>4>Ot8aVkIQ5(+?K(I zTQPF4IoA1~k$ee%DajL=SDtzWj*A}Q^mr+J^SUA@G6@c<;W*q|%pk42v3=JU()>OU zJ0=b1KBWOH|9BktayH}S`iIySK9dG_c}}Jgdg6|4g&Xzem=$q{0;*p?=And7?i|jZ zeGIvBg$9>jt;PAkI`|zP&%TeHc+>Bxta)9J_lRzGTU#eOxaTo_mj6l@h7UvOgTAOe zkw?`@hNN%W3ww`ELE|ACte9zt&qtrq#%-%8;G6}PR0hEQ@N(!*b%gpjb+oN>gN|Vr zzVK1ZVRq?pf^H3}Hx5Q{=x$nU@Q{W-iAHVQM7AHhhkN(3WdBxc+LIqidDYjbR&;2m zjo(L|E$t~=Ihs_SC6Ph%S!&oMvPv@PRR1-NWba7njUAC7z^qr-Jyq6^_$7`iH!eBJveR0Qj>!gaUc8VPrB z=u13ZScEBqHbDMMprq#Ic$#v(jHWFU-0lir>@)sBgPxetbccPg>h>LX4S%6^_YLUI z*Z_l&RpOZ##EF{5f4LSp>|pXTao!U1^xc+z^f@n{BZ4MEMAfcqc5h|(Pl1~ zRa5A)niGy_2f)H33ZY>kc=PQzt(A6W&k3n)Ztc%`5WMIE3Owpn6?)7}mW0?hlKd?> zJgvMz_VbcxL8S!+YAH~PtUJjFu6oVIv(z#)lV+^&p(Vf` zi$8Sc&w{DtJr4C5ELA?yOByo!BUgu}vj3j|UNt$Ndf!h$=b?+ZqCA;pl;Zig#$fij zI!4l0+l@9ZRDzo5_1(4I3CkiKtnL^@lSfaYIa>p1=&PMnA$Z3Vyn9i#$Y0DHV#ptl zUSs{OORhN52W{;OI0FYW2e~x0Ag%|Y0T2C{)Iy@5l-?Wm$ z9>(D}o6(*Zhez7$MQ3XiG}Uy`pqN7IRAMo}X*}21+HsuA2c&rAW0t)WWc~_%nuR@Y z9yfviT{q@Iui^pWMvR?nhMVO-$yGF;dhT#3|Q2f54 z0N-hA$y#By$WfmbIqN*E85)7V-Lx@!toZvk_du=a8ifch@}t+Gh})k+0dCp!@vS+w zKkkaRb3W12;tv%1Zv;(U<3f)+3@K`?U?U9i!L6L_7?$&k7F(Oq_)9ZL8Yso8^mr7! z-wT)F4U`{sg2orO(lpJ}WIpr|<(dZ51Cg%$r>M6|AIY7M| z)zNrh2j=`Mh0Hm5zM=XV!*v#*;6_i}Np6IeSs|v@mf+Q@Blx0IPP-l6(t(a6fRj2a zj5X%{<(>IWR4P=QyP$K+W(epx_I8bc?WtfKdEyQq-59A_)CH;f?$gqt?+;1eRt89| zl27p8h%NMLh$Uw9S&U7+r(?Ib1Io;5X+d%qY|T6XgW3w*JzM|-+Z?#OP!VkJRjjO6 zz)4e*c;~NRzCFv5yDU}3L9t(5me!Ap0-X56GkrGcR}IfW61*B8g{2*T@i|YK*JM<~ zW>ytUVRJMejt6UaQqD$fCBFlYwPoeaC*@)Hc4B>OEWC_Qhp-b)$>u= z`W0hV%JRr9pW%5w2WvMC#JyD?$unj%%#<#{O*|)yEeo+~h%e^dIYCRxztGg0;g~B9 z&84v`Flt6KskV2g2CXZ&**%d>?xe8QfkixBD*AqB`?Fo+F1E^!WOwrC(G@zZvQ7n4 z3PrD4>o@YJ58ylPJ-K>!F*Fm*F)C&t{c~N7BGp2uTgGB(>twk8R_<1IXioysrML{EQZIBT!BlU{qI8bNL}^9w#{N!x)BX;gE9-!sfoYShz+C?T$(+jmyMIV? zMY$XBJGB~H(u43;UI7-v_3+w%CB_!)$Es|>;(-)_H&2s(Hxtag(S*~Xh8)@=%V9eo zpy~c{9O!9BzZajU_Is~M#di=M3?y{66z<(oDZI_&82{at&hJ*N=FTmRg6o||TJLtV z{rh}=GGjLn`R2z~!9&>IUmu+kObAgDoG?nmqRd1rvhqb=&AT-3&1&JGoX888x^mm8 z{(M}V(cZN7L7$D<6s;@vOKEGdX^`MUN`pa$7s)SZ0RQ)MEjzDWz?%hs+t^EsM+x58 z3a4XqFsaE9;DmZ5c5ZvU_nnUv`ZRj(5n*?SHU;D?c;*j z6MOMmUF1K;YU9n-eiUv3tF}f=aoygaKKnw?!IseHt5A7f9gJ@jkto#-D}|d@(XPa zdnY+&^OvSgamVnMAROuv4$rZ2IN<6{EoQ=fa6TEzih&|)HyAHJiD##CH^_F~g}?KP z5z@Z~n%)H%axnybrn}LaKX>8wK9I}12e8pF9nM^wfj6&&dwt9?TA$YqJAD-eYws-C zAG;`-g@=-<#tvAc=+8at&+s{ilkDWq>~TUB#hTwSEu%X(?fQtUF_++6HwJ4YPEd^d zP3vFO;LBTaZkf1_Ef%ZuQT=83J|r1B3bDAZ6omfYXQJi5Ov(zsP2R#6fs>2Ss%D4h zqX%NN(QY~x5eBRM?=Z)q8S#!U;dkIPa;AEimfAX}HN9#FkmHc;(}TC3^4wS={S@qd&2#`JlUx%Lj}z7aoTS5k$k{|noD@z!x67@r%12bNR>EZ?EG>xxeeoMTEr$gj{ItbA=|dZ;+Neq z`k0}DdFRX#f7lgQTUX=JBo}m5Hzc>w*U6@S06w1+bLCO};M_|=%=$mj3)5s+S)M|9 z^J*Me*8_6eBhf`od>$?(WA4{{Y*7~HHuGro>arV?Uj3#zlSlO9WfU3rOQjY+WAdsx zK-(^Nri7b+$VGTD+|^tV`Chocue~Pyt-HwHeID+2+9LATZkY61>>Ez?MtV*@og4a# z9ERl6p>%WVp6pFOU+9!P70+``=mW``=*rCnjR*$y<1M zdOxb<9N;^27fzK#z--c5)D7#6?LY2I+@j{geN1oc*ABv|7;zR^b`C4%?t}6~afaWy z996b&FkLwf$?Mi*T$=+5O*No?hoo|m5z^rg)=D$I&7`yM=u6*Sl}b0Z4?)<2(eS+N z3!7Vupc8F?G1|jmYYdo-h{ie-(IdW|2MJ&ZYQ@>i(8 z*W)$0t1-{w6v`$XML&;g@JqP_M z?)Np)$v1iA^<@#IEtjJRk#CtlTOFl(*7)Qvvi66TB5QFS{pyILrJb$e>F$p-?|#_o zse)q}Cuz)8Z!F5%D$b&r=x!%`y1WBt1y8`BV<$fg$YQ6kU^dg;!EsT8xhS(6n{H4; z9c5#X`1kR@ON5K+00R1GVdnT1nC6^|la?o8BkqWeNlIAu%?dWz7wEK_9(sTB5w4_Q z82g6_HmDUeFTSDOxvB8*x`0-fM>zLgu*?ppLcgncj(QiOMkx@fDPlhpwiCb9Oxdwx zJ+#AwPt$21Vgh49;geuHM-yEeZj$E+7gVtuZaMhlLVpjuyfzRgiVmQ=0ullt(oO2lc!h>O=DfS*k(Xe!EpcmOz;x6@p%t!V>r(c5SvT7AI zuMfaKr3}1?cfrKEK**m`6+Y+HD6UmQ{~}o&7*I&hNEh>K5-F>~0XcqJ7~qzK?^AU# zE1(au{&=8p_anMn;*S|c`;ps4yq|lC?@GrqC_lMOU#@jStDYH4S84D>yOT)2zYukX z1JFHA2l998NV>xXmP;mMitP&u+m=eV*o-5MSeKAS+*h$nvX#XLJ+)(GVnQloObw1j3+F!F;2z;nq-`aUF-I*GYK_K8reIvfnIm&fqV zs1g=mbg_3v5N4VjMry?*;g#{zFe*$M}Fb2IXst|M{g4Sd-3 z3|m{o>{Tg{v^RaA;-O_UAuNIVuc)Bo^CBr^NnadY(vt)JN_q82a~{4&m)+mAV~6A{ z9t=?6woxife%zV3I1A zf5iQPze#^=AuawSj}sqA;Ef}geobFVG`5KuUQ=PS3e`}un`%~_uHgGcbX z-{QG>%ZrPiS4%CJN~3 zOKoP)F>7Ktr<-nPKV>gI*S3sfKFG7nLg6Bh6rHf{B5!WkA7g@rLrU~F%YF`p&R|*I zCOCMP^{&HDrV72H&q7i5H@|E4kv3&}NL?1PbjMegYM!>27Iz#~|K#6fsmM-8XVJ6@A_(^*{Ys=yjg*CWMe z19l`iqQ|^@G}O=r7gH}{aNq|hZ4xf1rDpiE-vHK?s_cAx4&PconAO+5#nBstedp{D zWcXI_xD)wZn~HQ9w9C-_BLg8gZ+ z2R8(Ru*Gu_=Jwx9WAkGv$fp6{!XnsBWRh-%zr^VSiCDM5ijArla8QF0XKYg7+K!?8 z>ryRG9J59mcz>U?G%-Rdex7u{oS}4O`b&v<@CuyJ8_j!6BDvggHs9_o%Ry@!$>eo5 zBx43}?H?1)uE@a9O@%01BpB))eOYE2@yk{@&b77Xr3x+Loj%KY$)UXzaQzGle*D6s z<|3>!mf-m3dJOxN&0XC*IqIhpKmRWZr>i4rs@Gjfdsj2;2>5|d&jhz>p~#c33>QrN zE^yOb2#bR!uyCyc&yO}|i;*ADeN#93!A-RFoiR`Dn$HI&rf@<;Z+;;E5aUOV<%Z}( zeCOL7F4K^*meF;{ZB3(-)jn7q9E{3?f}c_~9LZyzQ_A=tNv&2m&drnK2_uH^N|Avv z3aE$U?VUKgaRIvr?d10t7V;=-F(b{f!Yb=<+Qem(f0`xy&T}cx`KiJ-1{Yy17s$RjI?@eW{_=;E zbPl#%$?Y!Nu(x#&o>w{Z@H8d6c!pq-ZYdOlG7ZJNBDU=7+>zc8r1C-XSn=} zRId7%GzZGE)H|I`g7=DiWQO>4^_OgHUeEx|{tXSPw_-Qif{YGubj_b_z`ElGLo?Ft6m*ndr+`cV^n3Y(L zMV1?IctRk>ytW|y2W8ZCRxquwJT2I;jb!ol3f=B&K;zssX?ENd3KV^zULgbFpwtJq z9zT^NZ*3(r`)qVFu0_hLEcjQA!}$AsA=MB}L(4pLmVJvzrzPBJy8$;YFv44(3XIwi@8H~0+O4F8f@=l_KUHCkr2yEe|xa( z1!dlO&2dbgtz4vJ z#LCqPuz!Aw{3_?5;7LEM*bxK83%$5g$!2bQ)?c{pXTb5~a1L!1zRRdRa9%kbp;!9T zIkQ|UPzyz`aetxJ+Mi!fYQZQ?v7;K9f~6-ic*v+UJ|cWcXPOUSRbmNczgMBaiJg#e zZvY0a2P%L6rdxi#^x&caeNwl@r}JANy}loYuN!n0$*lQ!$I zVQ-PkG)X4Mfw%ec&rs?9>wePHqgAD$9tU~VLXma7xe0bD->7=7grk}tahF4X_}Jt< z?3b4(+1Zj#eies>xAqIB<^RBq?RMDOXD-bZS>*_mlW14U#Hg=Y7`MxkX6E?fVEI8< zA4%lM+ZNJ7UmfZF2Py1XBexFu?mTj^FQwKW85jgoinA$woB(>_FkG zsT^+7iC=WzC>`PdTWT4lXQNuAZ<8zN{M-M$lJ4`#rPRimSXQUW`y`!2$1w~2Qo>OX zroqYE_w(;3+xhmO{;Z!{k97(OWLc()h=;+r`GffH-XuO6yp6AkbHii7fa!OG_*36A zJVg6EYmErvt-~jX+-N8KSNe&L^*6)vTibE?>|s3IXb86t8Pwcq8(GwzrX^K2*muNG z_}srx#}DB?u}Fr}!7{k&9>Mm^NZji6iNb9r;i>&Jl&G%6*A1g_#(5GJ6=c&WrBpg> zEqDtab+oYeODbP^iXQjBBl)J8P9eh#DENIVWq0vF`rTj*eqf5wPxX}YK8b#B6+7Ql zbM#&|Q1HjjQ2Y=LaVJqhq{e%ClK76k?2o7M*6$@N#V#;ox|(1Z_J+pdcG@!YEu9tq z^N{Xsl#u3({U_rvIw1?+Z|*^9vS7(H86omV0`yLmBF;7i*FNvT6O;2u*$MWXWcs~0H(q`Yoe8Km8J$;{aPW>zC%RjYJ`Nu`l z--mWdU&?fnp1kfuOPoJbr0IXC&=KAI+itMiTSspbw^Q@^hcxfy421uS#bcSZSS}}v znL8)4_uGA(a#;9=Ka1Jt@}3;gdjyPSuTtcE(eXZf91opSaVRMe&&SFl;c^&-=ng<+ zUM}9{okifElNh3HhmRqJ)MMo$jMm(SUOKaoJIV;soMc)aWkCme93zEYJy8;=4aI(c z=<~3(G70qZR`D@Y0>v0L$PG#Z@G85I0?D4ZE2Xdl!5h~mTI))ZJb5avVe^8gI z&GM4I_ggQ`HFTG1=8lx6cDc-v{g2ZGw`f$RDe`;QKHRoMos}-{7jAB4OdIcxvrA%e z{O?3WI1IwPkPO-wegihXGF-k`fqzEmuuf1Tq91wU^UZEZi5ZD;Mwu`keF1B11@H31 zFRJEcc$Y21SJI!L_{cRlROMrt&mx2ktCXlHrO;)AP^5^i$Azm;;aIi-O?Kw6)L()n zpPlh1asjk_M#11(EnSZXg6Br!)xlQ5c9`}=69(VXXu)(J>X|qZ+UW&wNInmnWry&^$^d?zg7L7U zMtp|uVA060c(eKedS9CXuY_wTNgK?)jr$4rm^S|^Xc0TPE%0Bu7uP-<0t$}f{_*4Z zKG6hI6uOY)$5`|^clZCZ`|ITz5&Gi{Y@HO5XCcoc(kJuOAtO2Hrw)G@a1b;|8=VxC zaigjPJA0o*RdWbB?4-~eXNrkW?I0(z^73JY*y5W5y)Rq9H8~WTGmQFASI6UVM#vmx zjNDHDXx4!#BEt}<3@E^K4WtQ!ykR2YpV7% z^P%u?zMh7$p*aW)FGR(d)i^ZA0PD;5LDgRH?mgz=%^`c?l>0)9R4U2ZZZ|&9u7;E2 zC3xqazzmlqg5ke|UTnBP6<4ZB?}{D_Kg&YpmvGzpeZf`*Gmc$8j{AnF@|9lO@vCzf z-CHD_n$x(CLBPpvl1pc8k+n7f0~|jHZ~JU^tuo>xubfadaSz5njDnwQ3Noc47dE(q z4((Y^kDnf=SJQ=O*tcms}_5Wgzs(THn>XlX!4T;np-giWHkexn&8(L_PX%{0^EXvP-$MlL_}{ z(YzU|bno+Fd}-T_@$)tKR!`N0*v3bFHCb)tM^MF86wT8?i8x27 zI0xY8zG!%k=*IuoJ-6LZjk`GoG+BKw*47*qj)O5EvO`&ytF&mv4Ei;qJ8GPU;F`KN z622;<6S-2MwJs`O)T4BO47cl7VyR*#Vov=hcswce^WO^^6LNz-eEmifx|PzEEfNV^}uMD$S%RxpxsFRorDX{e<*27uB2*{9%el`ibrPc z$eDZx$6|vq>Fj9yw(zGjvb*UXim5u_AMMRZr*@fx62p51G)3kcjbGxAjWPN7G4T+l z_whzZ$p$Eku9$NFeOT9d7ZSEf&{8DaR^Dkiky?VdZ^A)(@DSW(^&z+PHua0#!Ls=e zd5G@=KL0bF2L-HP?3CkyLS#+| z{)l@^X6tkeUa$jIGs*?y^)5b+*$1@%!BGCU0fUx4z(vsk4jUxA*>c6$n{J3fX=mww z9G!O{mhbz9jZh+lQ1%WVG_o0(Z z>nO@c8nUkX+(G}8{ymDP5BF!0OTv3H=}|<<3_C~%@8bOUOlbBRBgy3o`S0tci3b1c z)s8`)j~%kU-65&tZj`Vko9=Pv^w}4lWN}u9>W=FZD;|N8wLDMa6^H4fAbj9C!=DH4 z)AZ4a^!e5$8d*_9vnx5H=|9e~)=Q;%e6AE;{0uUa#hLl)bMWZod4|b^3F}XB7Tgnb zt8sSY+;MF9E6$#q9>bYseD|0UkCNsixG*jkkL}~I`PCmXQw$fKeOMz}uTUh~pPwko zQQ9o3)*LFbx0OKEu2`5%$ z#dj?%3g!E0t9=lg+l6uFL)aPaRFt#o&(4a6@%?Z<-dvPqRxAJEhW~xcNjksNaBx$j_75WI5~k@=-T*tI_Z$J>rU`SvAz z{aJ;L)u#Y)7dYN~gotJXX2^F=jsg9d|DlhVAfCpVT?diwdk?=_p5c6SCafHHLpE~; zCY>~f>ao2T*^rH!1GC`e;(`YYo9GaKu3tK}lTB}RWLFww*;&V01j`my4JECsr9mmp;oj^ue7Q267EcOeIGVf${ZX_*eMtG<{~z5v?O^GSJ^ z3J#yw#KG(HQD`|0+5g00CN&vl{JRpVRE~k?qwykW8`OMish@fq`Ajr}{}*qxsE@;6 z8*6CYtENH860k6HLGQbjc-}h?zx>A_@Np$cwC}>^H%~C}(;J)^TZyibQ81Mlfjefc z$gv*69+ZNu9k+#fx^7@Awzk2k3&=G)jK#xFVeP|3*s-UF3?d#<7vB%rEc{IlZ>*uH za~LVn$>M5K{I8SiX`a1qYQWQ;f6(3h0qgTC_+H+V zbF+KMUEmDB;BMj?IY>0wV4?JLI=lK6|880`Nd+|)AbNwQDj9aVYykVE_7JCK4Ov5) zz=qA2WDfVABkFS+UdIeW-TYA)C98$_zIl+pJPm7{#E|#i1)t|7!u7!|{Ia>s_w}1_ zI=~P*`&?;4)Hmwq)klpzQb_8|q@uIV^zTCnMwp8;)$9G);?>PK6LSHngB@^wfaqc>SFa(Z#zOjY@+=*i7eLbs2chI2VT$cWG|f$hX8Kkf>};U7 zU86DMb_DjeW?`t^Ahy?aEGrl#%jS*rM(e0VEDqg|JH;VzvRnwsmUfc1ucRMS`=R5~ z5Ny!W#+>E_C_E^Q6Reb0?dSLG+e@*ojpv`wf276X>9lD^Dy}-#z-Ia%EK2`?(?%~a zP1XxP4&A4))Twa4mW!y@m8gnN;vGpNrdhGm#)bdBoV($QUb0|&Y}W3E<; zE%q{Jax0!Iu|c2?xal+|Te$B-B-Sy2P&sfqAA zF&`S-hxYXQ11eK;K*q8#oaXb+eA)l-pehtPY2459!&dY_|A=T?UWq6*p;{!CbV+nE zc%$g(obR+}?kGIC8_GFDb8u{>IWBfe;M)Fp&Vp=#NmdRPT?|3I`WX1?kHogg+}rpm z0=dbDkmym2@4@>p&5JPleJ!{ghizz*V_l;a*r%!WoDsASm$V|u_w^hq<^8i=<#)+r zvIyFXr4im~Ntw5b6#J<3Gx~9ne;!$Qo83^vK*67Yutj#&4G8P zUx!ob{WM|8f`yp4_W&YW#L&%-KN^PSJ_ zF4M`R{p5V3m8NGtqmtbvbUt1Ve!q%o(}Kh3bJxT9`K#e{Set$Il3+tSwjg=`C%S@c z`rM;P(Sz(MJm@pk97&{IjdgUp)dgiz0;a4ZOf>I+T+B_J-djL(&FUyjJ%r*-x!+Ma zmzvEc)85OT^ybSkil{wKZx7_q@gpzj{mLA&G}v8VDg2#Pa)OQo=pw{04*@k^ zh)uV^o0sjv`QTshAFsm-hvvXH@)T2l@P;MHUuM@VGuhrr7g^!4$t=N}_pfhF!OPq> z5`Sb(N=n~oj`wf6Qj|{VkDt)VXhzfAm2m8j8@6YtB6Vmq7WN%MLQM=j`0QzD|JC@A zF%!eYgQ;hEIbEe3%G{_#^*_zXE<2ghMIXtR=TJ7}EkwoUzfc=khSA?NuzlTDlDi{? z!S>3qDB4J`UAFTKlPg}9-lXr(o5}cYDYYL1q#$wwCEzC68 zhFKZ25xlMezY4~&_@jf^L)Ap&g+x%rJ}LfA=6?^pEEpy2#?~(>bnVJ{a&|AKdB!XbB1^q(?kMJYhg8%&l7 z&9tI<0VS{A!MO~|SUU1Cb(=|(o%Al67W-B(sI(wYZz+sASV`^$wzU1~e$pIJM9+N} z(x%uf@=QNQ%gu*k)-q32O?BiQ?vmzR%YUbkyTJt?l0%^4a~<|! z*1*?JT3)r96myl)d=%JOa+c=u??TAQ@ial{ApP#UDEMF8PqvqoF~RS-5S)}xr8jO7 zg_`2%NC6EmACkV$b`p<>z&(Qvu+`=~i97mivT_qD8g`Pu+(p_Ze~>meMbX8vnyByV zq^+0zuzk!t95?I7H1}Jwn7$z_Sn>>3aOUNot6n@i7l+H7v-ZMUhwaa~frUdx&}GSD zN`ES%0jnZt0nf^J`f}#CRV$r*=7_8Nb16MFl?Dg$yUtcg2-^jck{eB7ky^sxk<-Yy zR~+ZPvuN>!6?kNBg1G!;)Hm(`o;^DW8x2#quiH)Io?64{#4bqcSYwmkPt1KK&9vXY z6lVU46Na3ywP`lpOIqQ9oMY1?Y+|UabiW3eAp?kcAy;hrPlBFf} z=4T0~W=hjf$)V)x^hda*RU-T@SVq3@FAH`H)k!w|JLT47QHAL-nmbDp=lnIHDrJd_ z-aMCG&ol%r&Ti4XtL#v$tXRGmiWWTVL{b`4l|W`}6?ZnBzvOtFketBnVwxKNk)>kfo0ofo zh~%@9kqTExOHLg3EaUMtF$gQNhr=<^gQk0j;aAC0-e;18(%pY3TK^0=if`yuLqBZG z<=@8x<1n~$I#fP4(LU?rbj{ZQ3ZJ;!x7-lg+9ue(X*2q(MNz;nJMummOiOAw%Y&bP z=k3p;%wgP@>R?WN{4AmVJ^;g}n4xe>BH5ihgSY)UVK~H@&3t`>*^b}G`g}W(-1Uq6 zRgPfz$r`K~xrEOrE+AZU5VJTMMq6nw<@Z!mhH?h&qGU*-}bchO24Tz^Byjo)m6n%ZdJsG8ZkoYAWjLT`DRuI4b%%*I86*bBrWP zztIqfCls|*6Z1UwLRufF7#d1@|MjD@+<)|B-)YcBS18OZphHDFX)%kCvDFuD-5sbk zF=b8N4?*+S;HW#OXrnCokLLa|s|ZZu3~0TaN;>tO`{?@}p*c<8gonA;>DhpAyxMXS zwyj#+xu$}JS3I$0&LP|yp~+MY-(&L655i;%V8*%oI8-nK*E}4lIwY0mXUb9jtGjgL z^9|lFNyeH`HB8{2gZJz?d5mZ$i;Q4w%q&KB*k7FacpU>T4=2CZhHxv=WEbyqSN*&I z)Ha=_cONgHDfbXmvVOxkeJ882EntVQN3rUJSs1CD2+7_mjO!C)kBi5$FYlY6y;4Us zKrUMp>e3}jd|N4Mq2;2Z`v!A|^EERZWs2>aN zT7enI711$Q3bTIKL1N`f*4=2zPRX8uTKINaFu$JOJT<_pjD2|Ua6ayY%OJTmljgoi zpjggwjSDJ9;p{bN^t?ueuJRaVJd^u9c98ap)x6i=hu~*PsA%jV%XAgEP82cC$Oxu& z(}>OJ$Nf>dYRt^QhAsLP$10xovEasA%wmuUyOk=BDYxP|i+LE1957-h^f$1p3-y?b z;uWGFn<@Iub4*$0#UcuR;CjL!Y!eru&?69i+?O1GZxs8%RN287t+3$#KL6!L()_%| zln{OvPbwy}-`5nF*5Z8X925uHBNfnWEQNN~S-e{2hwGPH$xg$ALiM=|Z8~S!PnKsk z>-c%8TM98_20^`17d{HB5U{2ITlH2W%d{D%W^Tccs)ev>UCvq>pD^*}txT5h(uJ_G z>_)$8b|Nm5S#r;f?_thc-_b#s_mp0m4aSHKXZifCho9|Mft@}<4YovE3}kTa`%@SS z_i?ez0oFd>=;}LdEaP(p1--j?`{EFW=eXl_`5ErI)MA@fc(bQQL9FfLE~d6~Fng7| zibjjwfxCn90ZPk3B8j&JQzl&h12i#fogmnGCES&P{Zz1gq+W7yNbS4c0&6|vV; z*=OCs?9aG|5U-eq?dN~c8J?XQC{`e-^h)FY+)Ct)OGmc;RdU!Ki?8P^&~D(05FbU< z)xM^bgM2P^R6syfJdWBrGs8#Y*^uK?u(0SftzVc&TbT_`#Al*blz^ZA)bRIn71}k9 zvWSIg+%IquBeL4a;jT8cS|8KK@K$);JI>C~PC!Mnj<|&= znFp|uwL-=Q4HzHaN57S15Mm+A#{c5Jwqzyjjw`1>-iA2IGbmFJ@^0(ohy|Rl%PvNNQy2BB zSJJs%IW(eXC5g^2!Ycdki0gifD?%>s9QM*CA)fsC-QW(-SAPr$WUkyD;#l~B7LT67 z-6AE(KYI>Z6HjBxs%@C;dzCWGk|^X<8@(%ifMNfQWZsq~=slB9FZNW>zx7d=G)ctt zsubC_^UiP!uAp^g!Q{D74$DuSMyEv{6!R{S>Yom>^k@*)HRD#YAUb0kNk zlGLWNl)j_*SzCu^haGYA#Bdnu6(PH6FDt6u$tHCDM=n?Fp&?<# z3NOVnzi>OY&UhDke;2{mC>^6#@H3-D0Q6U=GxJN%Ow#=@wb}$i=lEH8dHbPWGJ~x1 z%LK9iEVyr4jW&Od#xwul5dX<@ASwL3Gd`aNj|zvmh6)Q7A zWo*TxzCKbQ{g>v>npP9?fC^Utzawd4U7m=9)eb*iwH=R$TU9434H#n@bs3QLP5+z1TC zQ@&TpSJ;p8kObzHj2)CIUwqrQ%7HP>uAfR>tGW!n7@e(+x_2p z>LqVa`-VWCzSb`M(4Vt6PmR;{|?y52Iw0 zzx49J2vn||hl_S$nDuiNu0_p_ovu0y| zCRetR`As;)`*CxbacmJ9)Rb6Xg90jo^K`T~2Scwv|2#w}uoKaTVcnRG_~p;ALuL>grr4i7PI?4o3mMV(^~Rz= zrYS0so+26|0g+YyPPRdf|5+~9;JkY$p0>!di)SXX;If~X-oshJ?>n*2_8t_h+?bY{ zA-i8b8Ue%B;MCL_SZ9=RK@ z@|=-#?k0w4cjAV?`If0hsG1mr3soi9qFaD-f8M}2(1tbn)WA4*6;cGgZ);3JY{CaT zv&)BU+y(^7DYM<0uBngS&KE+0=aTC#OR%;ObN1I}CwF~b;w=0kJd6B| zo)}x!dh9Ru9Nfe^eY))F=(VhP$|e>)6U_U^BGzNl3Fmu!e-QZ`<5WG^SMw^)aXC)@ zwkj0$G8Iu%Gf}@U4-@{VGdehcZF|gr?=$~kH*Olcbp0sX$lsY-fd%+)?ho3a%00h9 zb@;No7d4hm7_FTK1E}$UhW*u zHkub>QvGIXdT5KAHM=0@#P?KQ5xDdCH8!q`g2e0@C?^$ zYT>=Z{--tZz5NOO8KQ}0cLnsG7=VC!9js{Hhkzr7c$2mt)gckcb9#XB#zR;*=Y%t$TE~up%T$cZz?jd za%d?6%eJsFx$9YY?-`u1c}=1@(^0AA1&4i{lO1y%#eepqFXSub+;PVr>kSw@b0efL zyvHD(?+coEmRh_g!d{Ad)cCzjO}z_BZHMuLd7w6vclPzPng5zXR`I2nE&KBX3&dqG z;-(XNgX_`zuL1TSuHu}|1h%v75DN=eLaavzjb#Zqd3hJUa{s*I_G9!|zl2nZpU@e! z(a@R>s@c=bYx+5K`jZ&%h}Tf+1cuN5oRIO5&&vCsq32l=825ochxSj$KlMN;?^%M& zzS8J^b`xVKwV-X&U2Nm;n}-E%Y{_sR7BS}2J^;ga7U+N9;rGmkzn$>8(*ky5O4c@I+8ZDLA~W7v(48Hkel2+IfJ%??Djfmw(IC6{2F{8ioQvRKUR(MTbu!LfxAbRUxD|3a_otx z8hbtd4eE}GG3$rJS>l%{oaY^-YawT7O?W7dg-yrCdTmnX`*nw_oGG+bnuUkXWX2YC zNDrx_nO$GeS^5rz1>x{E;m)P2-FUj$5x4JMBboCS^uF~xeZDy!JM3%ez9>@2vKB+| z0S^r29r)xJSv-C3gLIie@PQf*|JZ|jZgTA8*t&2Exs zSB!x1+Yi8mzhkDFtb{i2fynba{tv}tJR3ck4P3FEIo^MT&%Z-ZF)RvMkwt8>d>wmZ zq97VxAST*ZoWSN<=`sC{1K5G9t5}+)h{-?i=8nu&Ol3(7*T>Dl0{d$y)=I=IsW3>+ zlVD*+oMZN_gyzTpBk`2mR4@5~KJ1V|-j6*rZ}22m{O=$Oi`~R}CBj)J_lTs`C$r{P zu@JXg52@=6S;P7N|1Iu4zp|LVMP|eGXD#1j6=JaqzuV83VyRiJNKOe8RassUof>mp zgdsH|6Wv#$*ydc(#k{FFkiQK2nn9cmbO!(8vf;gOHAY8epsKDPi*c1_rRjOtZg&bt ze7XNgr42zwcC2vISZ11U%*M_f!J>GUIC11$w#Q-&Q@c2y6-+K+dq;XPT#-<{hP~)&VG4riWDrmSRno zI<}mbLBFl(6nHwFX771T=foe;s!L6Dt)PQ`U3@?pV(K(wS~~fytRRiY$LZ*If-e6Qr5C;`nEfkf&yO9@qRj3cG?0@R@uPS z%mMCNm*~E)F_v`sz&zawD;!n8jMQ+wx|e=yPe)u{8t=K^#O$SQ7(C+&7U;S|GP9g+ ztRDlL3_b4RQ$b-5&xICd&`N&J$Pe}wnO9B|B~3IIrMAk6io5Ty!_9WoS$LDbXOd{y zspWLlbR_>f=+dA&&*{<{O|1Mt*t2IiT4T-dl;<}b6Cd$8i#hA5lVZd94)^7y9L%iS zjlRfTWE)dWE;EMW#AgL8d67*nUAt**_EK7;XpCcP_4r(^mKLvhL0LsbbWigT6>Bt; z%!S2nqr!Zik6y74qc4Y}**^s@-*R8cxpJJmJCY?mT*E3_cQSF;6>RN5b2izh6vsX) z;`ecP43vz8(Ze)&7xKRR=HFEGq?+E%;taLT`p`bF21lOPKgc`TB@NQ>I>{jF9*@{Q z5nc}Sz+&rxc&uJV#@lbuwP6i3B5oi`ZuG;**4?z#uZH$b6vqkQQ`D&!PNQCLq>|fw zwr8=EG7b(!vbi%Fk~lZT>J2S6*hjy(CvRu}V!H3}iiX{LNfX_kk#oQSGA~$$KXD0| zBN>L5QfIL7MG4en7vYWOM%<7nf^pwn(DE{jGpUH~)7x>M==zL{ zbZ5Q`{k%S#bcduC9{;VdB+(<@jKIR%OhZk;b=9B#EF!}n9aEtt4`S>yP$^to0&)cj;F}=(PbLp zcAoB3h~qcs<62ZpVcfrGG*#My>Ww|=&yHBqHXR1-;s-RzC7FJYF`}U%zVuEvll)g4 zpry?ZD5U;2g`M%IlyRf5*>5INvORFJKpyXJH&NNlE*dshfu^g>C-J|BDEDDL>BJ|| zt96E?9gs!E{LXsok2qQ{eI~7e5~w>=MW*fN1W$zove{EY%j$T)qQ#jsVkT46^L6Cb zcb3f4Zqft=F=)o;(@IANBF=KntNBQ_vu&{a=TJO7_loA<+RHPc+mWAq9or8!K#jX{ z&dh5=VR$5_z2|qd`a=l$o(l1FdyJMh!}2D+Z(bBi`nhkZEv<$n(}=@|hM+MsMS(XcTuc-Lv3405hyx@`dJZK1qH%oXC{tA1ecM z=}Su~1+?EGIq_JU&Oa~u-yXx`{!Q*D;@v!@gP3)l`!Za2Vxn&?2BzKQ4c;_9R|&;- zpXD&x=}NQT93*%Bv(%QCOB)PIX@F!h^_M+P{WpH4VNHfm-DZX-m)=u>PaI{bH`7Gc zN|HN_F@A+Q$|Y5?pP%cNdK;l*b1a6n9mn&zYq9#F71j^#rjuJG5ZTJ_ANS_KOFsr* zwfi%dYqz-v;}9imXr!KV->GNSXHs-aqY{Zl^nGnUv1mn1$?uQn4&Q0g`LC2Rh%;W3 zxZic41?6lSN(!Y~^jG7%FyFpKV1*)j`tbp&9C$;)51-Jmuu|IS!=EX}L-3YYw@i-O zVXkHbie@RX{J9sl3X^#zGf z3ZYeR`yt6e4?*HHApL6%zRFvp!Qeh^SeZge=Y~NuS`Ita{?O=GnY8rYAo|?Al|Fn* zr2g-|Q^rmg`1b1J!j)W-{1QV0!n0|LXD%JvS4^W0pQF7J{$v!&_uks4&^;v`7uHtc zS=3{E-xZ1Q7q6)?Y&7;3jYqZ4Tbi7(k~S~)rK;7=m@{Q3cJ)SMSC$`^FFJwHu28(* zRz}0x-_hH?Aez2wJV_*pk^C`dI##iY(iYw#y=F;F@fwV4+;3wZcA3f={!(KfXQOKk zhH~av(&Kw6r;olgVoiY%U4C7-?4?8-Ci0xis4ijFlpvw(kR`cpj3cL|p>%ZRLh8Id zfwav=QH#r9NZoXVikdreBKW>FVF*Ib)zOm*Lxi07Lcj+<%(L8yn3EG=6Ulw?o0=$l z$5+xF@S2{htv6J#N(#$H*dyY)7Gggr zU`YEVI>&iyf2M(Rs6ugl$9kAP^u@X@Zup^Lf|icI5cW=D0|%M0k!qY3CvhDku3o|X zH4d0w;){nrR-z^12;APBgK3f877y5N%-1Cu2MZ=L_u8x9^<8*~*7Dk%hr0}_q z=&3;s`6(|YT{e`Sn#Q3w^&WKBw{ZWlI2*)g9Lx75K|1dZ1Xu7 z;}G@LoANFzaNceT=faM`&RaZNLh5ickP_vqyNcwyW{EB(d5Ltlt`r3iH4))!DHR6w zQg_EN7!KA!Or|1cCDoB!yg$UJp1{f(OK>q&z}Pe$RAdf-k@g3&o;x4!Mz}&UI|SdQ z#=tLVXMkR>dFR zWArRCf{gtyP=t0ZWjMZ|5l)|}(lC@NJr|L2Y%yh7XOWnT6P#vYG*@7Dcs*_$5o1eq9>V&AF8WJa!sf|ZyenRZs&6at_faUmKkK6F z1LZ^$CXE$U{j?R$oFy+BtJ}(!dd+7~tRxXwHymbUghzKYF=K!nV)rbdoyW3-F=D#3 zU`)Lb{k2=@+xAMxjG1jy{_zHaIdetQy&s$P;3DMjaz@doX5mlMIZB%Uk~Ef!;qSFR zTK;bhcW93lPLJXLUVe6vYoCXL*=}e)GahnON;~ddrojB;^!k1!9ke|{zxoZMb+YNg z?d@U0&xkA`WaDT-_e`bW@FZENxvwBxTx^T09afk=(f~8(u0*9`DB?D@(3U-te81q0 z>b4UYefTEi@2BG0n&Hrnk!8hJR?K_#IA%LsgKc$b!MKWOd^yGu4o$18hLY~!B#Pq%qc;~6dnc5i?{%9?g4JoBG-#Y5pbd5gvexa(l#_<2+LkmP9fsH1XU!VPTm%l}K5UeB)S>uX-UAByJV%C&**0w?1}V z8-zKd^bq3INu#FDCY=M%$>^6O=I~jO-4JuU{Nn-xje8V$a1?@F!*P1}Ztk9Sf&W@@ zy#4l0u=Be}OnWd?4B99*_B_QHuOSQBM}n35Yr5hS1XeeI^lcj5r=x2}iXQsGibAV_=D#iDENnNuCV)wV9 zh%CBG{r?mTg`IvhDS8f-H7C-VetW3@?e{`PpqZdgS@dsQ3necZ4DdPU?ZAQj4zh^0 z&B&l2i)$3A5<~A|CerUDbKzK2FwbhuBIms$DPqeUa+BOmPwn+c`$UP*^71NW-R~g1 z);o0Psu*^>cuQt-UbNw?pAcGkTDbSzg1*nTp@glRNg1I}-;`bm+0~ky*RM&TJ&ZOx zYSHBpIYNA8I7#?sQiBjes+>b#KXD(m+pM9|rUKrZ=z(1~=Zq~I#1gx@@qYUv+)f{e z&&g(dmjj%fERQyi5W1PHB&fu8lk;W`7~CF$?M{Tvjg#=-mUDD^$5oQytR;(2qj72e zcbYxm4V8`?&s}@c+|}%XP2I_SPv?t_0}l8zP8Co6P3h^dfi!)HG@aRHOKA~vsXew> z_+2^&@3?<)&ea`A2~ov@J?ClB^bNFPjVYoU0^l`kJ0jx*I92_m7sIAgMam%3OMZ+awpLD_0L=qcl@6;TIy^puiblLaXalH*<{MI=ZDVq`}-cem)UrPfKO98SsF zD6rw~+oBBgX5K=#V;U^VJz#S*kF2c5aR=v8*nQg#MXyu@ZmEP^{Y3n%G(nb_H*{oY zLSDlH8c*bLBRGhBCd5(DuS?`>%I6C?1yo%dM%Krl31VK!sNq?pX-&6rIw%Q8o#Qd? z7vX~QLgD)7>%#HHQNokIV};la{x(Og46VBy%Z!$Qg2!tW}Q+ujU)m zl7*nZi&v70_g~?f@*d&)$pD)fE4JX7>_$8;&q4l}L_En`gp`sVnmH*|c$?BF{Ce_F z$nABY-QM*=-?Bg(LxTt8^7SoMEq+H|o@v3gv4QGULn$~w99J)xLW0lFPKlS0Zqq2UO{YnNh6C~iLIF#lM*hSB~+Nf=F5B-&? zpdI0sG|FP2p!V-4Hl&EN=g|@@C{2e=pV9}Vf62Ih^CvynY=q-B{Wu%&4$XLMMr)V& z2q&&TA~z-mjYlsjQSk+_j`wuMshAceWU*-zB}MhN!$l1zokSs34x++&C)tv%$&|b5 z1w9aZMw7$aNs8}my%m27Z~pVZx}Y>H)lTMYeNjr5)~%efsMWx6=vFgO`)>jcrtU zfb-|$Ib-hO0@&^2y`lxj$m+puI=R*pj_R#+=}Mj;fAcsMKAnye;rci|iqEQ)j!=zU zkDw#7jh20}hcJ$NIv2K*_hCg8MJ$J*fi9+OD5Z7phogeK*v8B|0wX@`PL~`&qlZ4m zt@=Uim%>e`^w(y09cHpQ%Ds5le+$IK`m-TEy6k9s0hZ`SU}vEzJT0f=Ye5Mf)+ONa zIycez^D9I@x9k;NeRWy1usut(Y@d_J$=Vo)rwqfuyJ}QfJeYg-Cc<{YR4nqf$EYOU z6TS1A_O5tB!TD;TV7b>!a`F-FE=8M@q4%< z!2{n8jKSJXwP@t$nToEHR8-E{tU2Y-%vg_o+-Fvb$ryHaFZI9ri7ZtYL!Wmc_WS;z z9|$I?MY>ev;zf6r+sNkKBii*N4#}NoAaVLN$*DaN+>V4`w@Vh5zPOI=ax-kTQotbY z-fX*}%{Gqzjp1+KvD`01M6O+OqLbN5B2GmSDUUwOGCES=vY{R3+qPms)LQELvKhl} z9K#s7B(!XMM#iRRg}HVUp(S+@*|jcc=u5}9$z@1iaskdxXR*hP`|dua!@%?snfGwd zcb)>Oj7C!M_C9(%(ittiRaCX?0Chc1!AkL5EPA*f1L7y+`FP%M-n|1Pdyn)Acy7fx z14nG*k^5K?dp#E8k=GpVRF%VmJZmIo^Vwr%3r<)VG4Hcc81Jl%8RpG66fVv7$DT&I z`A9@w{fi!jc$T_u5sTp7nf76ZoCEuW{Ky%&IYIr<=*KGZ@S#J zgbG(|gOBAd_!n{(L8UHgxhE!4j{llEJJ7zEXOKrVqdKn~j`xpZ@R(0%kn7JL-tWM3 zb8FZY9TbLa*aWG{TYN5+!rvVm@g}w$x4tCu|KojRGIuYg+K*roEpklHBptd+oQL2n z$2woTvz`c3wti|0KEC=+ArfQB@8ljR_p8CDwj`t-H$p)HXD^$_VD9W3GBWQECVK)( zGfrTEwi*80{F7GCiu`{Ljm(!dNH*p>c|NZxeYyfJi`1$59B80H97f5c@Li5H+*?40 z(s|bWKo}$@ti^_a{uS0mAshZ|m zZN>b=TL>8+&AESrG3`evhP0aE!ZIb2v6nz)?IJvQDaH439;DLkxlY-a`?@08>$ z%kq^R*x2d1Y)E`J+QmvSLexm6wUbeK8O*s-n?=2Rf`fO5V{Dxn4u4ork4z5H5lzn5 zt@OZE)um`Vtc@j)+k|b_d`9ar6T8;RqiNoJIQ0$Sy{#rX;g*5bBfsG7{$04r&)HYU z2a;@e4P8_Gi=)QRG2`bt$cz1ld;ZB-=@^NMB|iAMq#A8Dj%>x8|1ft!5orw_BpfYG zfV*}TWa}2=PfZV3doab9pM$}0McBDTP}!>qB%hU4HP{CV~tI{NadM&Se9d~*$x zc4naYNI6WZW#F4qBAh$Vvru!6L1~gaLdL2hEN=pWR?H`7>peoHH19L}b7w|W8O3_K z(bHwl6naY=qXVYkTiI*M?h3?4_hHN=Oq|^>U5A%^rYv3Cp9K!9;aRQ5ioRhz(sKj+$`JLt+hRa8%XhsUM)nDN#E&PUFWy?KqmQs&xZ zsjJb>yj^6>`%IZ zr{djgNoEw+3&~>6Kb{7)6Gkw&t$IR>2n84jrp%X8n z`mZV~kEGHvy9r40_CU$?Rp@N7K)-T3448fl&B!`8wP9JZ<2a|H z2zz<9xkS={Eh*4K&+H)fNUoiI`gfL%@EOe}%@b#`6^Su;Ve>|DMmXMslyX{8{wZm;<}Uf$;H55Gudjr%$TR*xo$|??$;X)gHbVOqL&Xbm zjOmIH4RSvq+VUwge-h_8n48zc!8Bs=w2)a(T}5=ZvYjoDV*ckYqlYLi#1~PC_b`$W>x!$eWt1|raC2Zb8Kc-9ncJxR_oC>?*+`L3#$eH^ zPCZ<(I*nTk8u2r*KQo*36<3r`qE)e%=2+~VhxMuJpwC>jD-#86q@qCk4hWMAgtN_>Mdazrt87JfM&3oSnDq z8}EwRb|Wol5cAI;&RzvS#NtpZ1lV7MwtXkxXDG2hLnPVt_nCNEmB519RYZEmmZE)~ z_M(SDaw0FiyKI+R4)uF-AC@<**=mE?OmU|@J1;52MtRC3KqeQvd7u7@%xm0Ukp@Y{ zF=*@>jKItsw2XR;pv@(iSa2KKZ86CEEKPp@9@4=>qcOboIu2*`Aem=Qj`10q$+)4+ zM{gwCv*#y{UOkGGps%!3B;eG!Zg|-LA4BIIj%6E$aeHrxLXs`Yc(3z*sf0ddOG85% z6iscVX;Y+RMHEqK7!i@Av}GhM?L8%hCiUI_b{ysCdA-kbKi74hzn=wDR1RhhQp;KY zZZ(#s+fI)wc!$`1CD!si=PQ>Tm~}i1TPpZ_U^;gN+JA%WTQ9ce>Jp~$nfK^NAIIVY z&-v`?5z76#5YtzPtd?2$9aTzY@?-F)I|d=wY7pvN1pNRzjQlk~BH~vO?XShcvrJg$ z0TI@>`XGiMT#l?#X{Hb`iv4F{&D@e@nfvb)jP|;S$aXEZX2E#Y$oZ41c@x>fyiRlt z6yu8YAhvMxY}P*Bl4-2ZgZ4bmc6%p|Q#MJ+&1isF!6h7NYeDVtSOkx-XU!LnvrT&C zOl*EMb5a?^&K$UaM$;R(&oiE}-$t-T&TE}vavaC5TBA0;o5VE6LtlCb=jZeH{NFqa zSoNKTdUufc-IbUe9?cm`o3VM19X4*2N6fo}^zW@Ew2tmTb!r)!y_#|2dmIKNBEUnd ztf$3*$u0SV9A|fg^0}!)kRAnF8{?LUGEVNE2I~Xi7$Oiw&(_iS^zAHmZ@CXArw@28 ze-q6*F=+A)r21L5D9+ymbBVxukT>Dr88-WH-94+ELnBBO)XFIl>a6|O- zTw34nf*nITVaUd_*d0?@jIAD%)yPBSQ430Y`G&ezT4M^&b#|7_=9zDGY%T7^_kaZE z-k-vp)cNaj=_w3apM`McCg_I!$1d2WvP~PJ*|;d)r7<{#xC&SD+yDgLbK=||I~eYY zfRAMXuRt_+ToK=;s5F>^WAMqkk243 zo%!To9fA|aZmv#hLkz zYUumzz`^rF*=8|sHbm&dww%#m=A6B#yyP4`e>fI1hczN^ur2dDDqu0Kji~NBCdinr zN+bOo>BXfg`u47fZaPk-k{78oCDIAG){ju%<;Ij0cClHq#Admce&)Hzs zaYBbZ)47g&Pk)fzG(GGd`5&%bIRneV!`aAm6}Buf6@A0M!6;agIUSN`-iK3gLYdDg zz3NG<#D&lBLlN|IJIsv6W4vA`jTrU;75rS1F_e3+UrbvWvmpAD+?H=|;93}(rzLhcJc4}^ckaS;LAx^@KX=l7ze z?uPK(-~=CsCPZ}Wvh^O;?2(8T^BKs5{*?K2fmbLF?A7|S! zzxER9-S&n4B_*Sg=i1Br)mdD933B^al3wXU5)ts;+b9L3?2$(1{&O_qTROJbYq7%U zX!du~33gxn1~c>B$b1Syu;uYJ+&$2O+YS6(>d9(kY%qs(n;sk z4xwrDS7hy%V0T|V!gM}2&Fx%^ZPC6^PI`dAY#DaCWC*(n11960%|G{a$Su)f%i9c? zV(3_wG~bX}TBTrkRRJAZ-A=o5hw%*j0)%?=^ZLihn7B=szFSVDj}uC%_(}&&*eHYM z3s&$=_(Q+%FXH^RROrmw$@81Z+{NjI=`EMYOo$^Q0xst{pxxTKQ4o!w8%TTYYR ziMKSPqJ@elY^R_D7wF#dd!)3Xmcp}gY0h|mdeK}&g@di&_tG0DvLjG-&j%I!Gi=au=!LBpBWkXM6Sp zGrQ~Yto20-la5}^`lk)>vnZn(Bcw68bRqoe{PBJgpk&`cQC2)Z!MnivIcew_UeB4z zt%$qJcX?_(f?*kzbmH78YMXeCRL)(Zy2czjZyZMU3*OKZRcE9>_kiVICmfTKfm|@( z?Tp>e_aqTCJ12_n=(W)T-ZPH~<~x(?t+<-^4B@Nz@4l5W+k4;xRx~_g$vZTK3$?X{ zX?e=RpgZrmt70Kb^EyCzk8JTSZ3{~J&*194^EjB52Dj2joDZ-Di3L|;0H7c?Zw;FJ|Y&Dc0e#MD?8J5)~!VdA?=M0m5Xjzm> z<%w-{>b5!LTD*~B%yZxWTFA0_1MPR~;~gnPM&M;<+2ajN>`$l@n7;VAu5`aNBh>@Bbya6*erLC zPRV#td+jmmXcfVK2W?Rhu>{s(G0^{Jg}ryHA@<@grdO-8_RE~{#K8XR;@2(2f+|-< zb3S$!ZAyGfH@tf2!|PPi@JptfSCr8g@`_#tiev0#M@)41OK!%2w7%*Vy)_d-x4kUt z?pq=K{Vj5MJ&Q)aeL)5OFQ{750#B~HU{J(Ani?&QvnkWD$!{j~6}aQnV?9P56@!S! zQdBK1g1GBG+&XXzMQ@atlJ#IV&Hgtnm^cmh4@99gaVI3JBYD2w3u02LbnEF-N>Q9k zEh`deUWq13Z>peuX%9_5J02!E-Z)^(dqzHU@ZZpnWYUyJH~}Y!Q3w$h5oILIqD4L#%sb>N( z`A-mr%}qz|^)d)&4`;5rGnkupBvs#eL`l7J*q`*3q)wO8H^+E-J!b*EPdZA`s3#A{ zCUU<}Mz165N%^%EC3Py25wjr<_MnRDXqqf{j6PWV(`;q_cbm$oH%S%azUn|qsD=kq zO_20;E~dn%V&RJtbhH~W^*Bp*Jg`f!=sF*?~e5WiZe(0$1v0A+&uhM4ik5)=7<{J)vYQ52Uu`Vl9)^th?6K%qA(Yy{NdDqas^DO!cl<>}I4@Nq3XS8E0ZqNOMbH2<=(N1C!^pOEyOdC2-F?lgah)%w5ez$*^BaBdeI4fuWqH4zW1nQ)P9O};rr_e zfl%P{k|`|(XiGeciId(_X;dVwZW+S89BQZ$wZYcCW8j?MK>J5MqAy-9*m5@%8Cj9o z;T{9W*hloNWDizs`GKFFywCfFyB3QEu~}Q%5&K#ZqY5|SzDEdFY)XgI;yhUF@yG9U zNj%&ckD9O~d@8Jfd2ACZTrzR4>mT__PsDetDD3O1L~C^&u63niz2STqK0bw9K><=a z&!TVve=q0Wyz2qnA)FG0rsCa5QceWhwgxBf`$2QP3Df4!tsP~DSyj;$rno1DT`Bni z$#-G!Us8$HAq^G?LM0G!^I1?zt?>MFXG{P{S zaZp#9k8Q5ui1AiMadHGcFX@HSh6i{aDaw3(ZJ1fPA)C8uEJo?1V~#@{?CvMSaO@P` zg)1S8;l>#M)B~IHToHI@3~pNOfO~rwLdLEY8a5membYIO`n>5DhFN?UcFjE?OjMV^ z*MW4H8lQk#3}@shy~DBgXf!8tFVO-qCNoi%6>iXAQ%=7?=w>^l{Tl>d>&=jAnTZ?m zDX1TlgBdy7aX@b8kuztswD^3aT3`17Ds)R;=M}(n)i$bR$2{#_OT7=_9#Jv&WosiPYVy%Kg3C z*fK4eHa3h#=yO-RnG;7Bblqvh-)S^CvPCfd(ka0r*Yko+yhpg$*$}T^l%nzdeb_u+ ziIbVh)av(ze%-M{r@?ew6;HtS8g(dH%%}6yys6bElC1p4kxq7=?6D) zhuT#1@y=+;xyy9m*8^HQs*ZHb`K;`>1}>k}!BKH5WCu!P;fA;5m(qZvtJ+}HaTG=8 z+R4P&4PSVNaiCO$9h@G{_DjaFo*VA$>qd39e^?i?7aRjGq9bf}J<981p=9EOl8$k> z;qg=$o2q1I#5#q0y>1KFl{E`%+|q?RYhR*i#vpd?%?$|Md{7rY7Vbj6cmDGTCvxw> z<=z!UP27$Vcg)e2t%MS#h3v~Ae>UluBHQ8|g-8FySlo^$n5=Jz6q`**SsVko$AjS# zlSD1{*;L)hz0fB5tVB(R#n*4c`n^Mtrs@i}bvNm$;eR*YkP30QROH$6J$Ofi}7$d!#{bL=D8UC)IqcDy6Yo2ShrJ<8$Azr(R( z6L9lg0d&{yLV*l0*VRv^Q zgUdT}M3nL!>Y(Y|cNdR>tZqjk)!+?z z=_SE#V?7@GmcYC(5T;R?a96HCZ5Z#a)urQ*g&)P=ghk+OioHUZDc7c5m{R^!fO=!8g z8CyN)V(}p}sJ(tpsp|>_(>|9{-p&fTq^pF4=m}V#8G%r(37jGMh%O}x5#wjZ@_D~! zSU@^57R*Ma*KUZoE`(RwP{tfi!gF;r#-3gYx$Dc&JkAW?=kWKyLL;>H`QxmMDP(hA zP?6(Py5G=GCl8oneW@z!j}}qE_ZmU+nyu8hK?uo_OL&i=0Kw+j2({tcDC`^qF4hUzCMBz{{ftdDh90>B!G#Lp?XMzWv-u_D_QO?{0;{$ymN0T!X_( zi=e>o1rP3;*oDf76hBsf&!xMABED4ZK_9g&as?`KdMvNaDQ^ z5kA}h+{$_Eifjz;_Ju|Bo*ri_^F=T<@jjiKyC(izG(%Rn4Fn{G*)htzx4s6ac9vqp zF$J`(v4+;0GZ@1;IA7wq&x$(;Eo)C>p6YYbZ-~Go)nNF^DoQR zE2cfB#?#ADc9-7^&A9uLJf_lBRo`DDiIx zZ2Br{_;{QCp3f$c@{2UHYAy{{NTaBBGb|ixf%*@kSa@**&Q-~xdGb!`d|pSVPyVKI z;SahJdK{yM9_Bk;C*;lELjAKev2N`s_z&KML+bg6lfMQf|8L0aFlLJq8;}xm4Pisd z@OG{@wl(kL@xDn|BU?+k9-KXWS{xUGJNV!zn}!N6(k?$&{_k=I-!?OZH4jF{l4Z2{ zWh*6)mtvRVB)Hr78XjEt$F+%^$JOIO?_xrr@%%FEqJCljs#>f#e~V~yF-=pLPwz(^ zBt?T3%Bo*Z-i_XL>4H0HSf4;grX;It9?bN{k7T+l2RQ3^Cw{-xWp#JmnT7p5{Q5N* z)m{-;m1hUBj0%LP%d(8g%524B8TM#|3i}z93)zgp%;&#g421X61K)Du1aoM56wseG zDU5y`h+T44xT3O$Ozz82&cq_Rpy9(tB>%xLt10+$eH_fDEX8Mi6->+Wq#qR~)Sq@% zAe)jXm^@U;_Qzdofr-j55|3tReC>cyzog-EEuDOhuBGjAqFAHTNzTI>Xxg$M!DQ8= z^k8rb$#Hg#OSB|w-r5g1wBf)xK={`bb&i!Z*gcP=70*!IcqN)O=!2lFVGsG7mSXhEc2XZdY|qgF!6*#BD9!nwyra%@y8q%jsrjla zW&|d3e>i{F+1th@{wZdSd|LM7yx{euMf z=Y``zW)6OAIgYJgQ(@&V0*5p2s88%2Wgn4(P=enxhNVzOxd~lZR!bu%jAM1qyP1M+ z7W);N&c4WvVBKkA%q8+U*7#n+gG;(_HW>=tE)OgT*T#=G3-B!eG)y?NsOeh`0vDY_ z`lL9xs5}#X61BEluv1{?C1YY&;ycn#FIm&>@@LL~jpwtgXUpKK|C6q~IY&*4e^OFE z=Xy=P4=N8qMV>W6CgmX3Jr2GT(%>um5UDvLj9sY2y6E}HAyrH~@e;X@HQ5}wJ}g(_ z=hWF2c-?*sMu#p_?!3JeM-Aj6I~KsZ~oW97}2L3Q4G3vcz9MBX~J_BXwC56)ZlF0$UmO z>#8wpR+!CnjCprA^b}oGvO`j@71sUL!P9pSY19cJZsxwCJPTF4TWF0!xoA{!=hDNG zuOQpE0$O(_QG%~2T^;gE5Y*yJo4CJO)c&#HQ)V9Z%W*$rJUhq8wBeGvI<%>)2 zD{;H`9Ad|9My1(gbkCK5JwK0WKU)d)=qUKFJc;c1LwLLK2+grOPLFPOQQ@NxRLp0H z1~)2bKR?rij;hDyXOE!g)C*&=%kWro6xl}?LhocEluSQk*8ah)gl8^->SJhA!2q3{ zZvm5hLwu>6h`r7_aDQD!lRG*o$VeA?(UOn~9R+FMsc5`pgo46ATv@_y&pTyK8N(an~>4Hhzz4^C>ySZ)|A`y%iaYGE2am7t{^Gq@c)J)-qv!*Hdzdmk z+dYOpwO^%scQmnZ**4k{_#dpcjm7&cNlfa9h4sT&l=HdO2U98h(R@$#%Q+)``d1q2 zZ;7s1&7=^XNEMdVB+mc*U;F_5Uh9T&=^yCH?~z!krUI8P1zc60fqeH=NH3g&=KBTw z99EB;z2ERn?IWJO>g7Cx*VxhbnUXKbVT+M6&&fHV_t{vC*cd`_vd3uAoST%e{twwb zAD|me>X0rPjH~tH*sZ4r<4+=R(b30-Tg`N4ttEx{a&IC(1B93Cz`njb1Z&lzC@K!G zRjf%M9zwdUEflg=hW7hPJfZifjpi8gGw}jFWJ#?=`;8bpzZr^x z!y=gJ!}t5cKzXA-l4QyPJY6~x5f8m+M(|MljTFP@;d>CeOqSWbuR>MxI4W;dK|wnA zM}%|m;Jp*Lq-4n+ND0|ny^HiRvXgE^rBUFuYPyk{NMS_*0vEZ-bkNF(4E) z(PteMZI`EK2jqb6DFWfKN+{WUn2Bq z5l+SaqRYIWVl^lVx7K*#7H0-XoSu!?&IZBgw9yngGK%+AcT>1aI2}A~L+i4-p>nky z*S%ljfVe12sNn8xryQge$6+I9Ot?`7PIqSUzGW-o@0(+!&LMt%E8#Nu1ghqp;2*>d z{2am=yXFyCRPN7RaNI|r5`~)u)3N^BA$pOWL{Xz7=~d$t`nyjW74J1rw5vv-HPC>i zRfkcUkdMXO5A2n45$@MEVA_hI!~(MgYY&>zy<;jk_vQr+9J($rL+0n=eH zD859OPQIi+zRA4X6M^CPcEL_DnnpHi;_ZS3+}-7meWC%VH93h$iPd=SeV1+*+M{Nc z5E=J1p+93BWTF|_BpsooDH8a^c{?BaIfJw5EFHbKlGO4yvjopb_UC;L3vWKgUZi?5 zSLbSM*&dInm(D_I#T`iYNHFg^{jlh)!`D|*h%%moQJqU*u4f5n+hx$H6rxX|oAM^! zB+1DNcs)Oy4!wU%Dci@;4AJS(+Ik1=Ys}b*WG%K;QJooG_>4{6AyB;-h)oZz;VZQP zq9&3!{J2Z7;*>Hz#{A@+wN-}jeG|(zv?6ndc2s=H&XHYULK;iH6m_4gI8s_z>qUV?h840 zDk%W1>oXCf7K-Zk{QNZ`24lyq$Lp7?@q5}g?s=1B?&ZU38digZ63xQ+>e5Z zhj7PyJAclpP^4@Emde-T)wUvJ@I8@H#(M0zc}9@@YB+9L1R(Wo7hUSvO4gg-3J&ea z6)30d$HU++(E0B$zMnlv18$~*+4@6hXer-e3aX*?=QyHsBCz6<|c}J||l+gK_j9t{W7U6ouY+<)Sr10OtvBJB2FZ`;Y40>7H&~RiJQjB~_;I@i8 zeNxb!|A=#^ui^dHc%1wCk_JW9lazV^cQm|VPkNTI6_YO^No+K{0+!&$#@$%3Pll;& zQerQMxuG&5gP3t9-TQkCJuTgw754%c$=%x$X-Ec<+~r%UwUxRX$X6L zs0N=G$54vLE}G50VxZ8Cjh?N_>Ye?OP~<~mQZEE!`nqx7&YP{}yX$Mg|FEX^I~t^S zp=Kgyo6Znrk`pZ1h#h|rQ@xiy>C6-?9MA3C~uiV#ad*tdbdw2d<--fBa4+eLagAUW;b7 zd(-fFR1=A+2eL5VT!!rdEa*lJw(#$azEcE-a8}Y0?;faZdJeH&yjOf^10A2hI|ZA0 z|NI%ZgN|&#t*JJk!&VgY^)@LzujZTuEfy24!K`od9f@crCif1)qc#^dZp&__BoV_# zJvL=sE_u-3l7Qd6Lzz=O*E@;rz>C%`w9K}aO00zV)Ad`()t~yc}^e_}leEE6q zK^F8mFQR-_3*{telWPk99ysP<(`W9h9ru5$3iL48X~?=XtAj_1)SXj z-8G7AuTKRkYNw)ESr<~X3&|~74Y?vw2s>kg#PPAY&OU3#GBxLC@cAN& zYR0<}7sXk@&;m@aQbgUgVw$0Rnr>~?fuhwmSbsQYM&ASiLd`FQvOK7vBliP7$M-ex4e$mS@15{}0jd^|3u=ULt*lKfr zc4rBxmaJti1>VeC>>6BuPC;1y{b0M9H%N58B+9_~2H68xHvhnA)HXbHTp_rvSqw(|~c{E2*&v<$4 z^nXb4?lIICwT_r>7`ZiCg)wY*Rrc@0&fe3&Qc(5-im?fNT3Upv3An#a=m2VyZDH^yhral|z}7 z++FOPbyuMKtxfP?&rN!*F$TtII-DVuPs+7jf)76CR8X(TimX<%vT->~=HF44n=i*6 zY&k%WIfwFn?k}|5ZA4npEt-FK8}A%yQ^KZvI3G4=F3Cwt!gI0l^jXw}Sn2h(=Y*BEPu`XB>)=%sUZD3cloU;=bUPNH%1){>HHL$+*MW z;}3ptx5tL}v@22_{-<*A>3{>4-nO8x&7Slncq`|^6k~K^BMw)8z_geM=w4h+aD68T z4I0W^+bx-m`XRjb9s%ucX|O3nVa1&+e^*D^D=6gpNypW*SV)~ zK6JG*XwSu3s#(LRC8`w1*D5j<(Lt=oq#N!*(VY1+96mnjm?wQ5O5Z&2x9kTE8hnA) zze}UkApScYAv2=cEKP&`@N}_=Dk|a){iULZbIL7d*fp?}}Zqwy}<; zdd?86_fHnoI0dlNw+fk~&tX<(a)H^JxHCVXWju7LQd)kSWa~(>OD6H)lcc-JAmN{B+xBLCsg1&7Crg4Re!cr6Il0 z_TCQ7vghEa(0QCw_+haL*P_nX<51%3TqE`*8Zr5p>(}-phY?dFD3&3Y@FA=s_(C?%KdI z8?Z>;fgdO5qWp#_9a;KE;G*3ikk(2ktGPol+9{7hcptD)P8o51lAPtn9biW1$mv-C zT^?tJI=y*#)1`~)`3A7kHb%0=5T5Vtr2LFHdK*# zc)0=`-?X8O`_uJ;bXZuSILo~L5Ko1j_*7wxv4O+zp3lX|k1K{k#=>*|Y-mL|3SZ6} zCtMsV5T-s45ng_}NGQKmS@>v}Cfrp|IuP5cQ$~H(v@e4`DQpfYy|RpkMr3}0!j0^^A}Nl z7)*_$xH0Cq`6&oAISaSsN+GdUh`Uj>)SjjRoo-Wn$bLlfCQ|6;Y>S}83Yt-%i;cxP zod0KwX(OyLX!LE$IN3xKyT8%ZiBkB>XNaB!FX$i-c|9zs;p}#vX+HXo)_zzG>z)5l z@u3;IKi*;w_fhCv=e+Xa|DfpN$*$Wwu%@meq+8y`v{Pbi<=MS>@AiuCQG{YV!123sLBi51P&Bs=06eh3wqsOJ~Nel0`Yk(WsHFGcmWWVl6MAb+W2h`6E5Cf6Q^ z*&;VANYTTgC8z1qZO(d-+=+=%JQr|Y9?{}=pu4^uVy|;>qB9efwxw7a8~{!JJ-gQa zg|3U1QdWEviH@6ti*vQ0(H{lNeL8qD!kBaF`>1-J6%6;=!msca$y`sP`o{(c^Bsvr ze5Sa2_H-)oH>Kr27!5mKK%=kbl7*%X@?(di$8QEUAQ46Uv);IUHC{gEKBOHBAZL-v znRo3JD7K3}PLani-No4YaS-HBx6!>wdCu4v37_5p`td@EbDOTwmXTp}ENDM{Jeo_V z^bbKuw6c=Op=22$!q zqx+aGmaJ6*RZPPtEB<+meL+2XkLb)SKcp-Dq8ofSr;-|lp^B?eFl;d2>n*@1-VMlR z{Qni~0)M+RsE*o!fKUU-7z86YFNJsP7GgjphPynyFw;C4lJ%pJC*Xd}9o%s)eH?-O zyH8PH8y)-|eK)dj=!7iZ2R)@JJxVyBCW*6^VrXzm!=-D6uxZi5lJT0b zy`TX1)|oK9cnC!`VYwCzrk^hToiKFrNnIyG&IgZ_(Sep?dqkkkIJCT zyR8q^oiRFnBld>v!8Pj%P;aUe%)xkibLg_*3iB4+H!Y;QwazrL>JCx6J8mzW!S~^u zD}7)bz8@W+e(RlZw7<-Kc2)4?F3FZ(JP-U>p2?frLAYfXE{z)m?^*qHV4V?Sq;&D< zDDOrFpQBWxw`6E2jdQ;loHqjUA+w<)SP0jVo6*Xhn4Rg-h&Ww>-Zf?PslQn8uYEM1 zEm&dgf`vF~B!|NaFKD)+BBr*!qwA>!bl+Sa6FqtF;KgnJyl{YW|3At^8?{eAPj_E% z&W9P8TD~O{@ib-M!$p|lqARfEnZ9KC3TWkIV|S1SA_it*?%s0TcU+4HrrkI;M3S{_ zF2pXa0<3S2K)YQ4tR#maVr`{h#`;(?UO5zhv?Z`d<0Tnnh0*KRpD1OBC>GZ%!t6;D zMx_VB>sbX7_Kn5D2TJ&6#`zT&Qn9FK2)<|;2$HErP(M;zV6iGgz*N@JaQ%IP_iM8u zDiV(_BSV;;XOQ$rLU7_#C@O@Y^wJ(^25&)|!#Vun``GeT-$}lz4RZWEx$YIvJ$VoI zHk<<|D)4S_0Itm01g}~8IOQ-OC2aw4Els1`>)Z?9r3Xo`7qquV3nvb2!u0ONc(!N& z?O$$UZb>HAJJ;d*tV}E}vcmf4Jvhu~H>K0#aqq(&9E>bLZ2c5m+~&m;lKAd4hS*Hm ze6}_+m3_Q2hV4Be&ooE0LROP!4SEkkO1u_t*d63+AHt>PO_1*ML^SUeu3Nnmvv@X4 z$BuKw)N(NK&U};_-=+X>b6A!s!v7ENmt6UYovD({@5TtG(eK4_Ik$c%-3+a{I7ewiE8}jzY8ZCO+;I zXW{1U2+=r+RYu~hVq+BA3Nv8Zvk31#`rxU17T8%6c2$DAzvq_FbPpix>(j0p^M7!(4wr>O8>o~%Kcn<#hF08sy0;%Pzc=t#e_g2V5BtsiT zi(}ECavo#89KwCc{ZJgC1C!T|oW(SfPFo!%3-f&>X1;_(Mpp}F71WSL=t3majK@K2zBtc77{mA1X8cj+ z{r;851Ve!dOA8I33wX++1_vwVM@fOX6H9=gaHJ>PVoYju~DuBl-r z-+e5s*(LPKdM6Ya(INbJxKTKx>$Gq}^h}`>_c%N&dr$ERMSPxefW$_NA~?EBM zjWqSnDXM!dhRHe~X}bP*&WyZ7?)Ll1F}#>cHl3g@e%}cx<{5Q~$2{*^Oj$$6V|{EW zuB-e+#m#?kDdwHHUICkX#aie#I9IsmQJyf@ceijx+DhToGyUvp&pyt{;~v&u<#^#; z1CP8I{I>W?p95~w7AIheV=69-&cHWwRlYa8Nu#edV}tTg_HyPw>=%<|vJd~FU~n4d zRwq&eXZkMn$w5cS1sJ_s&sonRc$HH~#;LwMf8>fSn^)k7o+p+sm&I@UeYCl{pJKD$ zkw4DUnNN!F<-e=P6ZH7LEEq>F1!DPcPYl^&h#3=WXjSDnOcor*6rST0Y|VYO8(G#_ydGURP+i zsxlmG7DHNG6;3;hv3t%oy!>nrgF6#R_GKW^<%N{;u$+{n5=hQPht?`yAaLIBo|`*K zPyZqfvFfHztqpWeOr8eBMsv=dA}vX=C7%g1=+sa?>%6m#B=C;3hj!B9MThC}<2G_T zeVMxKwjflY9J6BfqsM6ntcI>akGTrA|J{U~1<}}BdkpW!U&UC_(};bWhVNpwc=TZ% z+~@8=rriS6t{V;MZ^LlcC6cbW!D-NRBi~95z5$P zITJ!N75pe(N$$P->H2|8TK}t%`U`eb?A>UZzoCI*ruUPPg$SP1Dq@+94d)N^(D$Kd z;Nq*ms-7#eFfTbKInJ6jJBqNf;#g=Ul;gsL&rnmlhI;+0NHWht^5fg|L_rowr&OTv z$`N}XtK+PEJ=F$O(6ZI4keKq6f^PF?a}(!ToJyi>?w6U#yESWL4DmHcA9oELpqs_> zG@n$F@y-+i`#1zd%!J_@3+}r;M6x-KcwtvVnRdfrd#De#`HU$a(q%QxkLms*3)lq+ zVK$(PP05D%QC3W6EsoRMIYTj+PI-#r@HtJB8TcX^WFzRB_A07`-XaNW?$~8HMHuuU~@S z2gU&66_njLiM(EflgPIzG@!GD_WL@~D|r)&%UMqMD)&%=X)4W(>7e_U%ju%TA?jQs zLEb-HDaDLC-s;L|utqp#7(EtreBJ@M**CC4RF2929m}35@Z3WKcWBKwL|}dvx+L52 zaY7PIxp$;VGak<7rKlLyia(dHAY$fkSUN&zGeBq@66z4}_zl*x#27fQ9 zI;sxKE@|9~5+Hkb7?x?pL;QUlg2ks`WwbiB8=s@R8d21=jl`Ke&WVWQ9E>x4bon*+ z$F0`It%CxDWRFJLY!NrEtOMrdM=*3> zGCpXBVZup&1P&g?!ly>FFO{>|;5`;>eUdXPDKcXn!Z;)@OvDbCz0lRThBe-wQQ4Bq zecYmmObP#FhPrmZt{gpdmf^zDHe?n&4HoS|8N=NrXN5P|E`yHGnK!#17i z#cc%#=e^5B`iKN}q%(@(<|+#$7LE1@>V0x`J?82y9KQuuq9>{v;deocqH z&N^%jGN)(zE(rdZpCaqPXo@Hthk8E0@Ys3@RR&e~7xSCej<`VM+(NM1Ap?22C2)Kj zi@JLoX;z#QB-Hjo=9d(oWkm}9TRDd&FFr!{CtI;4Yci|x)na1N3Aj_1L+NhgF}V0S z33{g^zH1&L(zoO8`VVB1I+F(bZ^!1v*La@$CQK4}zVzw@N=R`+_Ny0Am2+V6+_zzI zWhX0d(qma$E1>^AA2a+v;abft#K>LY9+!1+Q{wlu^SMIFaUyn?i%$wA=dKnO)Wr%< z-wYQvRvAI4yAHP1+aad88Lx+GaNf%+dVZ37fE`khGw1>=da_71c1k#NCYpD{1habe?tJLo2s!hTDU^I5k%jhC6G>-$@)h zKlG4WO)M@RdkS$T!W_1(K+s`-JX)NM&Ym%B*H9Z)ykaq)%WcA&Lmup{m$Gnrd#o@# zH(j{>(lVj`kqN?r3|(P&ryn9taNofyS(fumj?J}R%5%RZG&H@4pRrxpva+eHpfn8n z;~mI)vMv_X7NLFSSVqS+*s{T5%+v4-Cf5|>f>;(}`x-H6nE^XoX2-l%&B2J)p8^B^ z-ZHR?^NqatcjavhN&?^DhVT=2)_QXOPXopzi7{2@R+!oFccHJ-G5owGTyB-2b`5v- zsGr8?Bl|cnRSTBa-lFu3J9DZ{W*zT}*`9g48}jcx#r)>oWXt_*)#gyvlX?N`_0HnZ zxi9dLi^JcacVO+W!&>*fL%m-XRgXSPM_g^O>Q*kMN8LeIS3O#LPtj^V4>J6AnzMM+ z*^%Y9ut6sq^y@HonV*K!Rd=@Ua3T{f-o=`%0@>}Umyz|L28nKO;IP+{E#EW=4r69f zWYI9#H1EQn-SN0#Ey10PYD`~n1V;U-wEckwzMYF;N~N{T|8EZ}yB*=TlsmU<<=Lty z@9}D#HcQdpgmu!^Nd6g##a0!D$x?ep-XtNfZ zAD)hIZ9f{7Fps?2S0dQ*A;ya-uxU$)rOsZ8;1MwSX>U|04fI zK3MdrfpkZ>Lib}78e&ZmT@p^tt`QU6YPju<;Vo)0lW8|Tzy45UFl ztOX~SGCxnplD6_w^6o)u&b84R(f+GkSoHX&{nFjH={n&a>~gNgW>j7 zNc~rdDHER}eri7Kwml$&TF!Bq8;^Czl_0kDFBN~B0N48lxV7s!7C&}o!CQma>`p~i zxbHfaT@`1Tu$yi>OC#Fc7A?MVuodOqh(D+4jrL%c@Xw6d4spV6rh*9WgE`vAJzB~# ztk^DvY3iD=$&OvLq&^l)=G@1ZNkj2uZ8C&;4RAVi6tnt!sbly)@>{+S5&zWL$;0k! z@$D-JZiyy?qnBy5=?wH*o=u=0ZO!|c_2G$Z)cE)0`}Z07 zEZmQ}I)0v5w*#}opW{%lCv&j~VsE;iqgA$-WZ6=ju<^$B;VWswgr@%~I?uQq-#3hx z_K=oTB2tnjJ?C*fWt21s2^q;&M#-v3NLqyUrf8@XQlT`-=!X)itWZ)$8QB@}zy5E% zs1Kg!zOU;%kK_9l^k0}wI&-ID)H#1dN*TdyMLr#ESAgHTp;!qA`tNNErOtUo8rL`@ z{q=4tPRghB6Pqc~h|luJk3q;@TU?rzMXfWFXy3C=y62yan|s7q?Fw^tS5kvz`N*-l zpJBY4){K`rjhMOTIPbo5Z+1XD3Vn9aT(y6+WRNFkqGNo)bro)NF?3KX1hUe(} zE@TtrO<4BoI0Qrx#M)!<@M|!>Yl^Z(4|G_h`$Bfw!<${6bOnPYj^a%JQ!G*01-%*D zC~db*Ba1fp5{J^q0zJ8k4QeU$WyYnE2R*OkUvy#@w6%kHR-F`zg)B zhF_pZC1bJv!9;YqMG8v$O3Zby>rs1IIL0lhM$aqGyEeFwZq?hg=|DQAKk&v5?;rRk zZG$pnS=za*j?P;-;ELN9eE&C_-Lsm6(Q zf+gbj=yS{!?7ytRVr=#?-9Sfn;P4)Vu{qq?WQzus{n$9bxmZ2L2E*!D9Q*dIEL_bcq!>)TPR z;&v&Uxig5FM^s>&2>*<1??eG-Y4mn1W~*8b;j>I8DGpqq72U(IP(%c`G=gcB!b$Vw zXgTV3Fl8fT()oL(l%3AG&#s2Wu*D@BEcn%Ttn5=|=_;IOvS2SZn$g||{G`fKZPDe|yz{kI;W)G!H)FH75n6Zhu}M}*V}@620@FoHz~t1JnX*5 zA?*5K_H|wi8$2(Wc^+*-$f$XIpV7@(sB+kP-40G)E%4z_IM0>DV%_tpxFLO%p6zL+ z{@wC8_jG_1^u>@>S4vklaK`1yI(YT+cin_g%vSV(J?HlB$S{WAv21u39YESmcPPHM z!84UM8Zy_GRz!!;KBbYQ8qz7yxu8sL-{olj=PUG3u89_`k%I5_$(Z9|hf1^kFE_q1=LYe+s8{A-ro*dyKZ+ zO(cbF2{f-Uh&tC!KzjoBbKc{g*PII&YLbS31KOCleF@g?xs8qyF?e0=$p7Ap@lf$L zW#7u?{reZV96W(N3^8R>KYzuN?MLuHUK}s}n}&^j(@?gX!Q_k-rb?b8gCr~A-oXcj zD+9cQ$`ibW`M(wjO+35U$L@P%zuSd>Hi6gI8Y#H0n&ulVrfWj(30D})8M`KU@mLhH z%hT!WrAah>?-XWPoyw%NHZs@3F>Lge&ye>GMdi;DQkI*KO@dIo{WTVXm~yHw&Z4U> z*STYLI+BM6W9>?7#NShbVm0^1Gz{YBCnG3XtDx11KbMQnP+?jO$>c_px?VFKca_BK zz!TJ~Z$!1SBS?JIX;L%!O9#dnVeU#RB-#EWzhN7QS?=Ijqj&I|R}87~yD(d(9FE6I zFnx?88`yZ5O`9Ig)`sq7nx>Q4VV#$l6ef$OS^3ZoHo0l?H@?2 z+KpWEYcyf|XiV3N!o^WRSpIDV%DuFqKOhCw<$1KRi~DOm7a=}<1b$jQCzUH+ByxxM z*e4G{`8FxMZxTgjN-O7og^}S=16X?eY~KAyl&SrO){-HWIP zdC(r@z7a1X`_L797N-V+&|a?#gX6D}D>s-OA6(Bnf~PPu^q<&Eah5Y%N;d_eKoN}hR&EKea+z9CG9fqq) zDzN{ofV{U?s9!P#8Bf|U*0l*yReXy%Zli0 zbq~#sETOxz|D(Bkj*z2oIrU!PnX6)HbUf(e%;#71V{ba?xZK75%_3~;%}z`*Z^H%D zqF-k@wnzlQIHVlW@x^$a6M_S?4r9b-V{{hfkvrdsjpf-q>(Zgnf2_#;SVzb($&cpN zNn?ki1~#O;C9`31R2AGSFxS68jyvSButFP`Vr8*G`VB!Mooaqr!NN40?c!?HI!ccQQ-VAH!{dKqKwcS`jo1MT+Un= z)~Sjo8u8SY7fRvwx^Nj3h6jr>z(nG)(5@K2OZsrzXBa-DxMB@w{K_13#yl@$*yo-k z_lcV+k26jb)t2B*w+_-D@;QrY95schqFH`BCVtAslIy`Z($z%{>?fVI5h2;=skACB ziP~uyZEjpaUuO7_&Cm%Xb|a6X3VCPZazDvlvn4F}BRD&4GNob@X*tcL9yLq488(4r zU*8l|d*ma%|33`=r^_CWUd>EpEZA40%XpYD4jQvxV!_up*uc95n>J2|3-^|&jmw7F zjU-IyYs9!=a_m!|3=8c|0EUjhIqp;b;64&-f83<)?mcAiIEdc0a_0FLb-dG3g7W7d z#LNn4=lN1P-Rq30&1I;x-2xMLZ7e*XgA0CA_;*#6+RwBICTax=G*f~EBhTpz9=V1K z@^gnF^>_?gwjIHNTb{fZsSL%xyXdj-BkurzC%=>yG9O(`^{39$taZ_JXtptQ+c#s9 z&N0Nk4&(hPZJhbjK(}{ilbB35QPC~V(|AC)IO}7Gz7x;W6qBfi7&0^s@a{UpukMdD z^7D4OetHK!q&~p{mWM3P*%q9$M|EZ`P5qpM4e5!zNAHc8X91Ahwv^8+HE{IxTsCY| z5_1{3jAgayv+-@R%zDQ|yfI#jD>(_UX{bQ=xK?DewBsDln^$6h7CFsFiWcYBa|gD< z+Hu&K%{k886ZO?Q5nA_;@C-iRfkqAIcZf*(Q&5UOoo{gCs3Pmu(_sy19dK3gd2SY-Ix(5Wp6!AuSWA2XowtRQ8+Xox;TyWXqHuZP?MDJu1CdKUXj=5y*VETqa|6F(a1T2E-=)QNDE_C)9PIMjTI!0ZPL z;WQy$y6Dex&>{CQF>!X?)CARAJMwWFnhRvM4fK}d~%6#s&W99M-q;as5ii0?h zn*aNJHEE#_F~>-!+!Bf^N3b;EB)(tT40oA0sJV?rRh&QPd;AAA_1AdXJerA!awddy zG$i6g*h7ED)SCs&eVIRdA)3mL1T17JTs@{ZI)UvjE@l}Qk21}qIF|ly2%@jAFxg3}NzZVNm^foHOvG*w|_%Hb_^A zweHnn>ARo7CF~Z3_>V`{p_M$hxE|wPj>O%kpQ)lso;i=TVxvQ^!sgI3oRkj5caKvP zwTXLjF26>$W*^+Y{YHf7I}BIhU8>;^sa$3{wnPR%k@HzM#l*q*hz88m#hB_(12%op z0v0*mn+Ra}+5^KqHsHd<2=pup z+w!9{USx$~)`W4%l%}iw*rrlu@i2z1)9JyI8@F-oxH$XI>Jox`uVMO%ZHW3L zP5OylH2g9-tLq14u(8j+{cd0_(kaLG;;Qr_-sF$|K{K7h78+=fd&ihBZ z#o0Y)S@vUY1+JPpAXwx-xS1;k3S2a!J-TIvmybwL{D3+L6 z;E>rT+AMJZ6Mpc%gh3I!9ZTRJcL(96i*UG$vz)kx@ugk{yv6d7ADap5i(~Q2OcDEk zuYeEl${glj^fTQ3Q-8u7BB6sYKF|#rbA#a1cmS`zY)8hnsoc5o4bATJ3L&R?#U(OMgKkn204OcKeXc`mh>|>*|#;`{>`ceK}05QwIw05}-O8!Md zJUkd(-O-Sgn1~J4F3fax0sAXg#G^AAOylz=)~l0=WSa%dP2&pl;Z9P8jx+3rSQ6t* zLDty34sFBgVXFQaT|A>wFrLo|{Otr6lR0osAE1V62GuNA7(9x^!u9;_-W3KJ?#ixvCxXdZX(*kSkIOyc zY(vChR``A$J5)rt@+AS2=O*D;!v*BejpbP~p3f|`z+}Tp7&SeDabY*Met8aQS03D* zZ;C;3fjHLj1I7hUAwKj9rr58?rP>z!vTLr@Z+JhJF|gP2UT!JpB>!-d}Kif&tsB;Ke3Ql4YaLML}WC z6C?yQpw;mpuHOhnY-kX<{T4%E&l-fy;{SKbwI~cZf#ObeOq3L5Ulwm=I?6*>x5s@L z?)nc}XSiQR{gvSBlQ{Bukws0_DfDV|0ok{nAk&I|`p)OkJzAc)kj(i|R&z0ImWk2*S_7*q zW6<4m5L4$jV6SX12HEDJiMyuj$(63wiV3^s*z%6oXkp`=ezx8>jty9TKuYdwh;LoV zg3XfIfu+rC&!_Y3>s=e>m@*Yhje?=U=e4KrpGV%4JPfTofZ4Z#=uX5{(tDtSx#rv* z-u8u*cfO@p-)6AfXA$hp*fwUb-_Il#onpsMsI$q*nYf$q5aBbbF<_pI*js_f+L=LL z)22XHFdK1!AsAL!i}gE7u$Z5Tij%E{1KV?iTeja8j_G_VoYpE~@vL#8)QeuTJrtDbY7u5gGLxqJA4oUELrjsH&@d?ap-F&uo zyDT%)$%j=}8Jg$YGi&QK_5cZN%l%DkfOlMzECTplQ3Z3lc<1wkCEl#r01@-|wBeR0 z368aqyuxuZ(H(}W^Uhe^H$Y-uZ%HYh^R3?n<5pWZj9d>w;)o*l?n|NJe73Ugk2J1+ zRm8JD4m|gy2Bpj(`l|YvJ`~NP?%SnQHImaJ=1d}onp^>kI85{Woe{n>j{mG|xNKa3 zW!p+AL_rM_+=uFa<^=D8XW$smNcNqNr_)OssoHLUN;4kN&T{_yOefOt_!AHwegwh# z7XI5RP%$A0A|1bIUguB2rwk>|uXe-83EU+%BMLn(!_c9|zrX4QD8F8c6r0Pi*inf2 z|0W>y%Uoo*$+F*%4cMan^Vk!ERV?(kF$)~x4m0a|Y{_ZHq8)AcupEc^RtR;3>epTD+9Q~rRJH6kqFb_OEmr(x5G zg=kuxhR+R+up03UbL_;}#KSjXn!gLqqc)QopGD@r<(?y-3-n!(OS6n(X#3)7Dixi_ zoyLwRJ|>I)URBgJpCAnTM3@yNWEOr;w;fg;tdnN?vi^ zjr|9Rj=zVznClR=y9{~Ge%KJqXL4cboH-DSSF>ZG(tieZRSWP&;UF2$=8WcoAEWuNO`-p*X+z$`H9?2WnZ*dQ^zqmt>|JkfgBYZBOM~SaAFi;nQTtrtGNi=bs5LfYvJ$p5<4r4p!~@mj~$Pbep)!yTF=4G{~V$1=Erjqj%c;$rzKU! z+|lF1yIFIwC{7++iVDx9V3HP=ka+oibnty+ZO}(9NWF~J25I;ym<8$Ht=N4(94BT^ z!Ob3V1n4x=l8ECZ=XsqbmPsNcWE`e#*N1BI70Uj-8N&WTDDceZThBDk;9HGUlP?tc zhx>I-CPVUuK1MAXgO=gr;B+s9qRcLn<{w++6mw@s>sR`A?*-}YuBE1^S@52843#|p zonWnqkuw&deSJG+ZsHllo^EW=F=pZtPONXue3or0$_h`*!uBVhPwE{*|Md(s+IXWO zP8Lh2Y!Y;Dh@vH0dwABg1TB1*UN5xY=Gi<26A4t4vcw+{p&GtI6SDshXPI36~VnBLCS21 z(@msv#&yBA+ngJ!#ReRPv8mH_STxU#IHoOz&)7#KmZpvy%8M|rLI-c;h4?)2BYjB- zf#a_)SY*{2Wdz33H=U{UxN`{h zNa%#plD=7p2K^nI8zI?rwP9H zlfOvrhc=oV#86duhXSG$SxSosJ9XKKefwy`x$C##dfprt!wLnG_su9^~%n7U+?aJKm`_C79()twx}=H5Jr^#v-d+a;7$o=aiT!(v(5!8xoiR1#+o z*21nqnq7YT8R8GBVD>E?Jpxx6q&z^Pf$De=riN!uugUTJFF}6FNBonY%Zl7n*p{C) zOyo%xdl9pOJ?HsLCF3?MUdmZtXA-b0DjeEbqS$PaM+K5abZ3+y?u<^yu9gS*_ahE# zJynFiTK5VAmnRCn-=+!8{^bcDbgvLrMGixHbsUVexr4(z1RrX)Lhds6DV;inutm>N z)BhV^Yi{6tR0!hVOvl}rZ3r4y2eDvrwz5i(EtgPdlTwbOX>kp#n^akVf);BI5ilML zVkda==DA=Sb#r#!+0i`H_0O8~MNE-9y@s;$1_WuJLaEZDg(SC&;?RZBIPv6#VCnmD zSpDq?Hpp?lX>AE6>l>pwX`!G$qlPNtlra#)^KrcEOO`{)!6a3nx_1RVz%$yl{V(m4 zd`1ZDlh_zcGfEyA@pH7T{4? z9sM|HLaFY_l=o^C*`?2=m;NUN+Uo~X-Y8K_+8>4U%MK!BkQo*anMxZKPLsfA4vk$u z8ZjehLY!wrE-qaHQ(q67A9fJxIlu7XgEG@u`4&zlmnpIRK8@Wz4Q`WFSVy5XTbi_u z{hlzFO@7sl*$@9vd>x;oHC}>Q=tG!&Jd4CB&ai2jBHaA=gz#KqvhaHDRH4j|Ys@(C z5<6w7hQwEO_&7?QO}Y1%=k0uPS@8gEE-R;Byo*11VmYpHzI+MHT;zX02OK zu+E3EZ2W1){3b8P^TLBLnHE7sWxq)S%iyt>&+cEoqon2DWIyQ*IoCdix{eY15pBxe z=1Q^PZhQFujG=dLzftO7O$^@OL#cUb+}&wOc2hZzRZ9n>Z?4A>&g)G+C`6Nz1wB1- zN#M48CY(3L!FD!xhx*K$_0 z;24`x;>)%V8_fKBZX-oA#%?D|m(AF~=UeVl zPhq@fE!?V?Qr`1RIK7inD`Tl}K`Aw@T?V&xS;$kli1l$xQPXpW{3hvR#Ec~D>3ogaD zBkw}i492^8>Cm_(!|nzSVUO*raZo}M|GCUV;-4dUxRcLYxohE_5BC7euBE(h;wW^H z!9YJhBfRGK`Ac=U-fV}Z1t+P|R*&hf3}Yv^Y+yMRZJ1sx2d9t<42j9(dFvsp?D=qZ z&xm)|f1IZLGd*-9@I1v|TY&I;cKEhJ4<0eQ==hK>^C4xK6j>{W?}2Mz7d{yxC4}B( zV@W>#iePcOIo#fJR>rGR-ea?cr2Kz$C?OUmPF=_msl*QtWyn2#Ntq1zGgzl3Ygh{^x`NSXe~VVcE6{~~qu}c_14oM|VZ~H+ zNDgj0|W)zHE?+GTueR0 zStKnNv3b*2taGR(w^b9cWxoryT+X2O+<595{G9s6-J`H(7nt_vVcP7I2n?PJZQpJx zY8pZ7JN{5?`UosF)WRbfO?;S>O8M6Yvk>mIwe=j#0{q=Ef7T%G=XwbDXaC{D7-Lq< z9ZEY^NV78^3}GfKjvpOXq+X+h=dP=9wlNyt?kvXY*;zc-FHOUHT;Q^+fU`~xplj?j z7{5-XzAIwz)eV97w0M{}h9f|?gS4WH1?J%?koQn#0XY)P>#ZnDWzF-F*dTL5W+|0KC8Q1S`dK&{uL+-W1yX60C%?)b`uxNTioCa{8dX9w zJ{}i*tX)a|TMtmdy?^xd3(p)69*)D^Cuq)NE!y=rj9&Td!>aW;i12MhSJg#y7xdAW z*jZG|=YYD0fAAijG25ZW`|5Yg@SL9^W9DRItIic@wx*%%sxK0SV(8x^K@tN$DI;hq z!oDwuZ^t6sIV}dWm1n5`+6tVSAC9yO3FtTxjLq-=(AR|$sCvj}NngJ~?{gwu8V${v;nrwm7N5j*?NRKH!eaJ))fkrk?=S|xFDGli{j??84Cz0(TjACZ z+&Hbzihf3;;n`lCy#Eq2OZs8twh}r?2Wi&Hr=)u-oEC0Z!rsX%aLaxn{@XGIjh8Oc z-j&C&XP6QbsElHwhONjJa7Vz019UdLU2u6_F7s~Eh&Lv z+EFm+nF7}PlV-(fkS=A@QSL;L{q`2m>hmCN(@8r$w$KEHI1K7dX16v=3DstNWd%zj zSo=e9_8?4}jV=tNl}9gQ+XyXIk-L=H_&GA+#al3{?iN%G9|6~00oZpZ5s#%Oz~Q$B zf;u)+gUTqrr<@DRS__=qX+=|1s?9@>1tLG+m6@o9vDz(}%;aPYvzuwk^z%Zn`*;Js zbew|yAOU`?_M|Oh1vKR#q3X~MEOdxOi*P9{DpScxZ3|60BqFqPOcNGq-w!<6u&g__Nu&^5yz zIz0Lv-PBa#{HhLWF({{5vwzW4PbqYB9)!9hfhrg2z#&*RgPiUp>`Yzf6ie`?*^PQyG55rs6js9In9-o zLeSiim@fa647;s`8ubanDSlDHnd#BO9E~vHv<*yXv#x-4^Df3N3lStnYa=;A4rRwH zXry<)z^cKM><&fHch@MAlwC+tHOl0CQIhS>G-Gb>wb-vkoSjsp&aQR+#?!APc%k-F zu+MG|z4$tjzB8=UXc0d$ovZuhCh@sZ;0^dIpG@)fC9U19H z_dJy8(80d~)~HJ9nWyRXgcN#sC!XAd2gu;X6q@k6gQ`U*;@2l#)RYQwbJ%pA@zaAT z=M%~uS<5@Mh8X4VhlU&9F*p7Z9GLa1Y7dxa}5pHFH{r<*~d82{_wnwNu zrZ3=6%pR+k!ar zVH_z`rqMpPe0uTm9JOwWqU$AtNjLF3?Y}(+QqRWYu!Sz(`0~tZ&;xqG-yKVI;&6Qo z_sv)wMa0T1m>!Bk;3I8(^14& z=jWj#k)L?>=Lc+CyKqmm1$5{qv^RC}d#wtBI0N^X4)2+Y*uvr9JUD(FkM|vgR5oCN z&uyRSm|P}ZG`vCP*ZWCFC?I>zcq!eVz*%0|6s>ZN$}B9Ye4#QnZ&-`c8{xRS%#(Yj zRzgL@6nAUKLQ2&I!c~Syj<>@heP^C&oW~h~(bV?qK7}2+OvS6kU=SdOOtJf97<8R3 zU6VjanJ9)9Y2x!PRqT=JqLr1CsbI@u>JHdRx-67}Hr$}LA*r+?Za58ZP@!k5Z8$Q+ zf#(0xB$X+aG}P`Ig^Zib`{7eD|Ee|WU99kk&!dwH+^|ACoHNm*(5T4$8Vj>9@>l?L zq`nCb@AV|}71L;o(hO3&Jdcu(Uotv02!T$b zxVoo+I(8mKX+;L4EJ7i}_m4|V{=;>Po9M9%K*iBI2$t1gVDWkU8D9;V1nx(@D2;bK z@1>^Uh_&7Vm{v{ZOl%PZbJsC@Hxd^*T##pN1>@{7C_OC3&s)5&sLFj2h)M>|Cg>Q zjK|j%XUXGa21yR8qjUXR>HB1$=P8Ge5*3tPHHtE&UJ3TM@1R>h7t<`2 z8j^bdkhW`=kwMKUI?B%h3njKv@ckIFnVL*V;#TlW+wURi;yZr*ST(#E*$o9K310$HznLW?YGsg?i!A_dOsli*CzRPN1Q zEraEwW@6d&Vc0rV6vv#$A?42+wBOr+VCIXhtM(%*Vg!nptfdu}r|A9vy#gBV>0*c& z{`wTq_0gSzz_+IajU#kvkLy&*sTe`CvVsMf2R(WI`55*csY1%VRJ`d4g6OB6*gbzD z9?JWoEo2hBwyglWHVbcW{UGk>xcreZzg-{A49kDn0S+q_v2#IwoOQx*%gq6Qcn&)Lxg&9SH0G)C^X#Ee?uDI&@Z}+R`5_ZG!_%?+B=3lgx5bO)DV!_X z0Hf#6aV<@h-71t|BHXbpYLH1a8k{+EXDHH7s3Wt|5p41h-ks1O*YXHbEB;O88H!MU zp#`6!XXN#KIEwGZ;^&=WLQZwU;7wP6)QXj}f%p4u*E61nXaQ6K@J%QS=es z^S`Z4Ye!tBV{iNsk?4YykaaLxXoe-ZmbB|-8J2ZQv)MdrxZo1O>8TlczBHn}+L!P< zAdR~?2AKEgLBcsjr`Sp4;C*!?d|S4&hhh!%X~Mltxcfj|yD_eh6(QHwf)} zyo3@ZO2T2o|6^MpM!@r|0+UbeLEG(89PA!}_5~7T=Q9gchcBRMe+fQqI0S>`8n`#| zwLo{*I<~O$0?YLcW8FOOwba`J+h>VG+fp1=>sR8-Gwz;~U527XmU!}In!xm}BOA1> zj+IVtV8+QtO!2B6pOIGmpD{)y`)r`JU<=0O00aLe(4JLy1?PQCF=ks4q?et7{=TVL z%vk_Eb6yEV%cj6+xD`8Nd4g?wm&y)t)?D48V$$tChlvji*-uY1=4>d-&Udt9V4xbo z?t_F**|x$D&JZqG@{27?31`V}&A3u&h>tuIuy>y!+ZLwDmfw7h^6<-)c_b9C&&#tF zMoMf!Qwbab4%1zotAa22n-Siq&K@|8V53shn3_{FX1TA!o#!_6Y1TV(-Qk80i&t{? z>r7o$nT@w7qE{Dd$)NHg4H~P;ny>hd$B2-t1&j-ZJdabtODLP>#lNhOAFh zoE_P|AL_Ri@L=C~Oq(Quhf`Kz+TPtTx^0c-$p(nKy-n~!?;NZQMziO}D$G&sHeQ{! z#)6>z=7m*@nE2cC?4JD+=DI5zo7F~Q<9_Zce|r{h9&54akupqO@(xOD>S36~ouZee z*rw)beD-l3bNRF7=h1w@qUdYnz-MDS3m;&G%U^)MpSA1kp|LxGCQr|05_$L7(-$sG z?zIa>7``EC-YK~!IiE$p-^E0Ry+dqD1bsDCCX=H_=~4Xwk}c%njC!7zwpD<_V}Gja z|0j@N!|z3f;~=l$guU8x^GisB_WYTLo))kK4jx>y#ej2aM^C?B`yTDzx92&VwY^?2FoZjvV zrKhs6R@5YeyL#y64C1U{f0WzKLd_I6k_>$*SU&6#8OrR%7QWZh4hY7o1U+QT_${!_ zI){T&GE7vf8)L3UV|a=lT4o1RJ)eaN2IaD>+CCO`v4=&uM6p_rY^>^wM|(g!r1!KT zdPFv7gf!88>sx|(o1#gY_sya#%f-!l>1#J7w z^RP@6n#%b!_@FHmCO<=%pD4TguM%ptqhPgSGA(&1f=_v&sEgT!+b=9(FASxX+#71Y z`#W?ama(gkOjydC1Wa)dr`)${q`T=S4bN?Zf50-fG%=2yEK6kL_dLVjXP4+tq&j}Q z(8r0UHo7oRjIvgW3;eUns8#X>bzeV2)dpi}(=;s_{dG6J__dHq+lt8h(=XE8agqLM zSCe+@DO$#R978<9aXYmD)_-^}``mCu#EhZdfHYzm=Wu00KMKC7vlR{|tlg>|p^F{3 zH>Lr9UcbYg$6vso*)WP#z>cC=dOnAHF-vo?>0>Om%35$YLlo5+1Rz{wF!PHrVMAVy zWWLkPV0E-uurU8LoBKvzSmz}z6f9W5^d%Q#h4utmV|*C(DzYqVEzeQ?evhQKNNmgW zqffcd1#dN-Nws_(?K!JO8(eM*>|JC9n~WBc!=#U>U_;oT@z!k44NJCZQ5O9D7qk1z z(^=`etxSDNGuDL_k!Etg`EAuHbnb5kE~h9n2?K3rEN;V=mp_Hu87p?jIGC-R+>J&3 zmuTdRc`|a=!!3m*$c5HIl5_Z3mMe`=j~C?o-DG<+RfM-+W-@!1VqEpQPIE_} zquZ8cl>44xNc2}y_DH8acb^JO&I~hOCUKG+`b=QgZHp&cMxrIAhx{`pQ>aoZ9J$}m z&#oA`N7P{Em_z&jgwf-)p{(V=O*Udx7t2e^VL3|NjqY=ux=%Xe*pO?eU(0)Y)9f+z zcQln9noA$P!?+tW6LZC4*rxCg ztWZi^IJ=>M-SqB++`5NUua?U5IW=hg*or^)Nl2bvN{zn~xvPagn;Lkgg3sH-Ih#rU zU6`P0|2Q`6-5}u^a}t&)>kE~~oMpD%A2BDt9G4tqShp_c8QUlzGb@aI*QZjaWIdfK z7@!eXOkk+)2lv6;gSRpP4)3Q57u{|UuBdMoijO)goZ}HAymN1?P;@6}l00d~!fmfG zhxbJ$9#lZsuWj_K=`m(hX|c`=Ptb8+8>@C?(8`_`ig!=NUh&<`YV##FbWt*!y08x? z6r~~GuFBT^4Pi@f#j|E@BNkG27&eAJUi?*VL%9>C{m7ok%38)>IcAX51hK6B>JCw>llq?iQL zml@pO9*=A4|LEEL;kZ`sjP@7aP?Va3@!L<6fo(Dg4;f*LY$ztW`$FyfCZ5ML!*(+> zglBQTRPkbb@!@Clbw_da^fc6U-lW@?u27)3G76NJVR!m)EaAN2uw;uDd$FE5#>~w@j~&>f$;vG};LrH^;&mYGLTgr6D89?Ixden`J=I`lhsM@}O_L{8-`y7G! zDNXQ{e2l@Pnz?NE3hoTLi256f$UHU;Z~8aECe#LBOL!*6kLR#b_EYxapLCwjsyepK z<-DiK=$x+v**lZ5#>_*HA;H2I zMOVVHyKp|l3wal=eiB5)B$!`C3nzvh zZSZ6%+GAPEWCoXnwK$^{jBm}UnA=_kk2+^OvXH~Pc?(cBe=8hf4}!({qTt8`^lprW zrJN$G3Le5lRAt#|-2wi-e-4;Sv7^&;S<+|?R&{1GGy1cG1sxyHzS>;mu67BwdxI*Q zG)9|Qp81UC#dokw>n@h^T-_YOLyZ5b#Z>>PGbghKXc$jrsr~C&(vl$FH;87dj<_+o znFcI!?NJQ7a1{%NCh*KlH*T8bK&a>lHLnB=voAqphYGgwENj}gjeI|_38qGT)||^Q zuWtoTW`@8}Hvu~=`P`zE--qM5ORr`Z-hDm>G3P>r?&lrv#wyhK^kAEVBpa}jU=#W! z*o*D2p%C>llyW8L?RS;{fi2=_7~Fb9yGWqH9SoW^6l&g;%n$jeC9!+REtR zP|keZC zY_?`sRw=U1&vs1Z;UcDNAHrU*@nZe;jLnr&hi;c6Vq?d`MplRehmJsf{xod<90c3W z6W9}c2j^==*)z}ANZoT7hWC?Tkd)0?gBy^ke+KfI{JU-)4E^1MSUJxvJ54iYvMHaP^EOPO)s?Qz5o{sn8ldOmxZ!X9r8VL8n!5YF>uC8~qb^VS}b+0K})IvW-) zfsp?*8XxZ1V&VN0xP2&+bLrPWPgDkrFXi&6KnN*hzM)qqH8K6*b^4t8lUgs?pigr( zCadx8VTT_4=NY5ok0VCyzlq7Jwb(fKIHpACG5MbPtfn~$C4=kW(3Xu$N-YSpY{IVU zG)z75i{7-#LgL&_dcAHi~OFX!r> z|ImwuvTpv}=Dj4lRBYk>V&AM?IK}z=E2`h&XtN}9{8@zf5hGC=VFbPANf4f!!8uY6 z2yYpTNv`M7t#8L{yOy($j~!V0^$0fg!A@3p_71W&-e8SNC59+9!%E~gR%qm-v;Q7< z6ezQVFe{e*QHEV}7|J|<>$7J~p;+V=g4PetUFg<6 z3+daLSm(Y3o`defai0|PO1g$yYb(&iGm{hb_zdWLDD67YKo3-}kQC?oH5}lX+*$Fo zw`?(@j|O1SJsB9ml{?uoprO7No6ep_WkVl&Qh&qug9tk?;xUd$wo;r*vn`5zBG6zb5+~1zV-<9+KVcrxio-3Hmo~@k8 za?<-zGvoohhP}Y|$|&w%e2c1OA}n{%Z3KKgh9Bv7QMh~v`w=yo8IL;wzijR%TyTzw z-+IPw7`L;dS44y_6YH7Jryw?D>kaN&o5tQgpT^{-j%RZsH?q5P?U-ZJM$A#;&+KuL zP#RT%f>C0uYSDiv{N)DQf%Etsp^cW(CGg|9$RNLP#EqE9##^PZzg6*!H>%j*_lf*7 zzlOaX6o`Ww2N1U+4I%y|P+xNjb8fH4p?fDW-|8MzCUzk^<{LD|@{IW!zSBG#B21Mn z7xu`f3vW%lCe&K>PuTiDiq69w%l8fAW+EaZdz7>Y@m}ZsrYM?rMnh?8NTs3D@NL)< zNmhuYB`X<4LqaCRS`7_x<*DeZ=^(M%uO2fW|lDS_nb=b?(8Rc)Zw7H%V(?vMj(K9jGzIJOr|M=S56gm8 z__x(U-n$LmiM=?#{1I$6cp);)8hvjU;QKlqEbJePw>o;*;}uJb>L&=5p2Z8jr=1mc zi>C?KSOyCpC5#eAzb&HCSr2L2C^62~?xM}hdPqk!j3m$fq*o8b;d##p{ceT`wwjDh zQ$J9__fB|!ab@0bH?l-S0aNBX-<9&^s9e!Sub=d^CL z81(D4FdVVCztFUHOvEJM@#}#3$7DAD_Xm9*tQ4!(_5B07>_5BdUHF+xSSAbq~{Ki8=*P zoMz8PSBJ7mrb+C1ZxUPDv7Bi=?ZvtoQ&G~l3FA65Iivjy-vOFq@YW1kcQhKKy*iMw ztsgII+MrPzi|BG=G}WqcKEDUH*sa8^AAAod;mdjOqhPyt7&f_kV)v?lJQpO*#5^uy z(<3b$xVo8~7tDp3b~xTH+kzqc$HMuuBHCl_lD1hLg*3HOwc`i`E|!42d@#Mxcu1#0 ze{ya<@2rk9L5$k~1;2VrGUuvs?_CpehbXb5@29dg3wYk@@pbI|@B#h33z^LQBdp0i zo?W_lfXPH#G9%ZkRCH<)L@aooW6dGFwlhH4!U|e`B$5X1oufxecj%wh2l9WNMW5%7 zB*StQY*rYLSrxAM>BY}nFJ|DaW)m&4JcW&B+U$D{F?+j_Z0lhKmdi5Ha8UurGv=c* zVH%ccXyA!TI-Tp_-pDKie!jd86N7iy_gRE3S9t~}=Ws-2T!)ms7L&d!&BVno!nJ-Q z+^jYuuipgCH&62WQy9kl+=4StRDsgxH1Tyf8B7qsf$w5-m3AN{X(d(!E8)r5N}BJ^ zvtsXEF(hy(oWo_%#d$N~7aq}n?%b8|@)Vx(8I;IR&f;G(igymB(NJrT8nVIOerar% zdP|QUq*B@4=``waAgT9HCE>78$~-)Y9(I77YpW^X-6VYcHy@45x1ilB9!&i&4X<<} z`=>pWF>W$G@IFix&*Oy+$|SYKG+OYhkpzkFN!7ZA7CuNPi_o!T!E=wM{Ej8*?8xWg z%do+>iPE2E(faETxofNe&u?6S#nHde`5?_|xlbW`gA4-K&cUZBJE)0I;91c_r1^Uh z{c)4WRZ&$W+DYKwT+Ys1FTkHCDRge{3o_T9h0>|hpe=5JVnfb5IK=lSPGa~lZ!aFt z%0a?`LY~W*&pA&+@vh_)>FmEnN(B}4Ec_2$I6fG0OY&&FKhIbjx8kG5Exh8~<|l8L zW9yU2a7c^95F>ssdvOKtb8>OIeHI3m<fivL4@>D3>&l>rO7gQtRRip z;3JSuI0DO)eh@?r#*gr;H16Chdh>h>jLjPGV#7~l1$;ul`&^XmjKhlDA6Q->!c;$9 zKvEj-WXN{X2JTQ86&wP}n1#7AFL6m`7qh&o%PuZRqBX9P80W^lzb_r|%YQy*{h0-G zBMIIqIf02;wNRcU#m43QLSFR)&LeLThz*#Kj>B3KTv$rK_st=-&~Jk47f%!OyFtNx z$M}58HEN7VB!z!zg4>m0xbDeWZi)Py$T@iQG=Rz7(PU$!RiM@VXrU)8;6TSsn?roi#y59#SY@?oF7nMfM*p)5Fo*Otyx*rsLI*y55{2c218^< zt;g@*opkSJC-=`jgnF1b_czJq^aXr%!nUll=Pe;vAAuOfXi|14F;KrNVO z39p@k=&#%GR8nUoNlae+E$*Kwr#*K^;QCwvB5wDRQhyFT(TSoaUokYzorH~^z~IK= z*dIBFbI^mR?zIHg4zb1guoT!1nTL1##o@E$42>>aKuHJo(4XhVH7+J5BUPzln=wJO%>M_^l`Vs2I(9TF-t`X{ z>cY{@l5E${uduAKfoq5_9p)~c3Z-oNoqv-o^y6vQs2u`_^A_0eoeT5obBHb~Kw3d4 z<|bKy9;l%}F$`73u~;FR2E*PET$V7!Gu|6;;y(|G!!B-Ehb2FQ|IFf*p5W#I_CD%T}D*&NklA zV2Ye~5G$VrF_SWQTJvm}O(1?eFhut-FIbF{Yy1JM%d+tPjCgeggu=&YZq?c}j|Fla8R3E}x#fGy}=UTA$ zohVxrQH}}A)R4(LL!NsNV53?t>dkpRv`z?*fNO|dBhAk4mS)Z#nrw@mCcDRT9KQ2X zX;rNPTBg0F0sm&&uQ{LJ@n-NDRRTRe{etElzDa-TBdMZM4xz8~Vf3toPP^)(^?N+k zavxsWwmEpG<&0L5Lj;}A6yQ)rRy@CTk~7;bp4v+@_r{Yr{~2fbZv5GsBP7xMnm!*= z;qIFPI%4ICKp(=%0yS82j#C8poJVgep_m*EEc|tXI=Qp`aaua{>)X;)tuE64P(>~; zw^2i75S3k<2B}-h*r1d_)|^}USmrvtn#r9gYuCf#j~nLP-a+ZE`Q$82qLehwUam19 zJH-g}PY`&&0y{U-$<(H3l|UxZcD z&9G_SW!mMehJ7z}d1j688(x*OzJ593h`LVJ;-w|r9=Sni7~?IBax3I_^IG~XSWT?R zm+WTM(fQ4jFwJ2Iz5Bd^+Sk~Ug8y{ti?5V_KX?rxHX$*XV@bke*qbtZX<=ZW~!9^LVZU#zk6RdMmt=gncE^M z^TIM}?4C^>4=l-CQUU5C#3}6F3@WdmPm{EVl6q1kO*d3T;&ujyiAFGA-bv<*y2;e+ zCsE@F{PlH4gCG_1lYEfRSst@mB-yHDB{ut+DZWNIqkFysn%rCIghL6cu7|SZFMHu+ zau|!F=fk+>JKZa95X}F(7U=qG(PuNOFOi=bteB^2N^2)FP4phT5pIC_(@>LZ%W zAZIdZk1rL7?vkQK zFs)gqig?>!6z35|-SLNL_e2kxb!7y(@w4*t1B2<{ncFn2gL`bheWkHt_i=iwHnaVB z2qr%xaa3@|Lp7lAet>Km`f2^_oj7tsjD_%d zfy&Jq3VKiI{LbeN%f#`ac?5QLtil%3pz>57YS#-O!PY!MRN)=UJ7@@#JN;ClF2E<> zQP6v9iDSVhXyBX!9!t7FDt`?++Gg@|(ps#D%Eow?D|~l1o%SAeq0PySbml&HEi^8q z<~h7y>G*-%c7G(%L6$Uj?i1>JWrl@MeQ;`lG4;fU)3--FLpXIct>f-PWp`PO%9uyS z?>tE9OtPT;RIT8`Oo71spcVbe_$FxSd?9dYA4*T(Po)~wOH`NVg5R#oB7K0JK+X&nA%upL2h_j8n;=eeJR`yJXvFUA0*fE@Yy$|4_j~sG1o2c3BJY6t9 zNX^eX>0#AQzXbZ>|z7A;sM$hvjb+#yz3FsE0H=JvM>+NDzkNg|vTS0TaH?0<@) zA<-!LID&h^e^FTHM>;oF8vj;4Cq20)`nJ9RZ8Nl3u51Km;XC%a@=no?^Dy6e6a}w& zo~J??Ck~X5%n?(P$cUjIBeGztPy%n(PPw*Uh~j6Eru9u?RmN~BNQSh-1Sl>wN4w(? z+?g{G$N+?5#^YC2cQQ2~u{_Xlq z;yw}>HbfE&Ri{FI$2j~Mh#^CbU3idj7;~jJA~h@vYqo}>YSwCqnw8Lu6zhw4h*^@?fi;v@cb42zOiAj*d%=c!?)X;Hz|C7@(6C(*b4(49 z@^l5h?lp&-17~pD7eHk0CiqpH@xsn(q+dhr@Y=jvhd_#pdOAtRK%Aurq^ zxCrsy&-fl(g{TuJFzd@|R4>l}%{)m-3Yrk8%5b;z8=4{9Nbz!Iv^~0#441zm^?*LX zB@1bqBkoGNm(oa)o3ypqyzUSn9k7(I&Hy zq3sNhvsp0WyNunZl2QII1Xl(rVr=tumc9NXlhwY!=CoI`MTc~SKbwTYFX3Ov_Q6n` zeB@ILMcz-JfO*#^ zB6nIM{_@$*zG+wR%f%aJqZ4plyas>b$1)XBJ+dF=Mmv&{1y|m!rN1%5kaIj0>R)we zs$`eotDAuGAFrjf@%(&YErxD)d+K{O9d<2K@FaTyQgRD1=0XspUyP<%8)cy@W{&TN zEU=BUjM@riQMx0H%(kSGv5GFWeb*F(+U^$wM-)=xl=Ia5PXfJzhXT8&ljQR#QdT}o zpYN2?fyLb0aq~OnFAAervwP%ieT`1dtq`0a;6BwsE6BY3B+aclN1LBtq1uor+MwJ+ z>6{}G!QWF_mUc+^?29JnXu<7&`Sj86A!%4mKzQRRNN+E|-F1x=pmBv--zC%icZ=Z{ zegXXGbsqVBm0 z!beYc3R6Sp3NJX>3Z<*9gk$WjgwK_B(Hfp}7?x2;AtG+1S6WSW9eMO1-;eI!4yO9z z162Aji+ip^)Q*b#|r4O+ZVz4G0l{_VLfDHIrnFB2<2X3G(k+CG_1|3)TE7GoL7Uv zvJzhn=XefRv7Mj&x}e>mhYYJOTH?mDN&2;9>GKSJZN_YEXEzNx>;@H|S%}vNqTX?} zg1oRj6y|t^zRlFb<4sc7@tF5>3NO+mqiQ;~bvm5eIMYQ-1y>V&VW9j47Y#q7ke}%y z1<`Px#61h1OZi;b6Tkk&VbY(gq%8M_(pBqepmz>;nsUGR!9H@h@{BI4a=v0lHTOK_ z($cCj5^+C5L#6`{4&*wS@V(#ETnPNbUl`2W}?N&1fq>bI9F(l`BoNK z&pSNVJzL?BWzL3|4}o{zR_N>;j8Cce>DGEv)Zi|kUr&Tu8DEvO{F8<%KLg(gkd)J^LSGn~A!lU_XG^WPh;5aFFNs~eLjxQgEu z(oWDRKkgdh-Txu8cjMWHI2tLb4CDSvqBYTi3TuI2oV=7^&F!z`KQaj3?mUlM6G3A= z?^Cnhe4hEVqoaXllo(k`hA-`D#2ZHn%l$~vhNGx1VJAsNjG(Nk`{}WXF4flzaxjV>IkkJUNH(0{X5Tr8AGpso8K66)5fz*v*zF z-?|7g<=hADX6}E??4)N)wDDTJj-IqWB6_;*AU5@L20qsl&tiYKG-TE zZ>JpgDBY#B$+M|f&_#B-os_e>i{^~jPIlr=`LxY!UP||zZh|fZT%;hD3GViG5oF^d!NETq*5XGmGyVp1Lt&*GGsW) zm!^<{$Z@RK{s^yqJ5Z6BLBIW*1c5K&q5fG8t?us775t%;)Y;hQ7ldir2AEWwDEM3Z zN)WeSgfyyDDce<)E>CR~6rbXm?8Z&FwhNGcw;AqfW7%y*TV|y&1^4|=V!T!^_k%Ry zj@Ku=i@J}4oSE7?EgT=Ld~vWw37elr@ocs@dlAu1zh`(L;rJA4Sfz=#$upqqpAXXy zo^_s(dbXlNFn69AZ7qt$(ox;8KjRAdDVj*{vxV;Wq3oCXTfw7w9|S*p z%LEs6j|ry4Y!qzYQY%neeUQG7=%oisCGlxU+92wMy-yi02#0@K-cl zlDb3b*7IQ66N1~ackt`s3}lU0M*s4iJkv4i+$U?SS32YW7pk}5hf}%#CX6;a% z_IXHa%hb{GX93*8)LHzwZR}N*E4#FF2|J|~#`?ciF~ltN|3m;9Ju0y_h+*BF z!SG(%O~*~=V$O^n8e6{?+1FMmSo3n7r ztQq35jSwas#E@%4uv|Zt6YM0Qwn!5<>fJGu?@}#)8RE>Wv(&vo0k4$A`2BAjp768e zqnHuMYxBZCpK`dozK*i$B$#0;ZvSg0jdMA)Na-UT^$ozdj44QUnS!y#qhQl1z*E8TO9&e>aP&eD#rr34zpNe8e;LkN<+Ip? z2bWlJ_*~ZDX${fz*Z64l77mv=%f?in^&O~21BDA+Z?_1`#(xuTGq@tO4c{shU2G;? zdbFIbdjF!lHJ;e9E)r2LDX6+Bg3#q9w75_kN5&;%vPc6Q&L76pH_22X7sBlBwy;43 z-7M@{DLZ&_B-3q5=kKETG}a-KJ5Vme-t7kjW3GW!|D*1u^4yDf5QEKX;Zi@8-P|M1 zq#h-p!O$7wCznE#ze~q=NiwAuml0Dw7$=^@3VPI6)1`UbAH2jFI{glC2rVQ>-Iru# zYL6k_6R;wZGap}nM%sf?XsPER(@=w*7^ltd{uXCL@4ZA&M=i|u*s_SO0;ccM&Z@_Y z3QO;PWotYSvU&MiF!Bs{;T>koOV5jWPFv32&Hso(t(EXgJ;L83m6$943;LEh2rTCg z5E_H+;&Iry>oyKLKEyV+1V}#5MbbVAJUXx(GU4Gk&wJgkZv2OhUVo_HVH|#+c!TZE z-N;Y5jn)gC>BQf8M*Q4xuObpFYeZRZ*Gx7t!IkZEEl2Yxyv zFc{{*`?0#{SJQ#fWpku^Y zb=xScPCZOxhE61{<@(TU(7^cJJ`nr29gU(_DO5X-CYS1m8D50OvILTvlt*5-XJgZi zbUwpr$M#Fyi`RS~!dV&+^9jb&iU?eoVu7^Dc{KgjFT$@`)SL2=wrFfY-KZir0q;QDN44E!)I^ImD$1H)- z@KC5IOvJ8z;^b_*g-phK(2I0_4yI0Ou$6?!_a`JTmPd*rS7}d|1_n-Sz<(oiP`|JQ zioJ4>TWgHsux4zTVayy1Mzgg~t=Rbo;!IlC1!w-IBW@&r-n6>#ZP!Ii;~D>tejn-n zX)Tmq8;#{&t8lX*1pjUOMY9hd$BV7%Y;x)XHc%VRat#)+gFT$x)0)m#j!RK+GX%?S zHKSJJE#RC9DkrXEgr?7vINZGB2o=7{Am1LTL zR`0~>Dp@wg8!Rlrhkfka#B627n29;TX3rs*H}+#@*AVu;wGF#P#sdk2vEy&?9`bOO z8LYy5ZS~mKP*B94-w8Cbfp24}ACLywIs4mI`AUAr+jG%XSzu0H0yuK|?Y;f#`{ z32;;~z|vjy$X+;{HFCa|#JO9Td2<^?B(3o&_ci?Q+(0&GM1)8iLN>Bpuwc_m(pYT@ zU-OwL;(P4PhlaxVpe;SGU`%EIZkBXp9b2!i#*STffrwW)0Zy+E20-SCY`vC&lJ(G+>VF+?n^;)oj%&6K1^GlY|o8+=~ziSL-Z{xS5Jh zWnEl*u7`Qc{4qmv9~1^JhRB~@n(^0IkZITf%U71{PQwZ|o^WfkYkRE=CJx({w%p9imi4qW1}mdqIlOFI_C0~ z+UJgg*@~YOvTg%Z?^$Ea7bUWKcaqX7wXoYJl5(5wl3r{BJ#gV1<;0UXm41i!>R-S} zJAm`uZD8~=pEQkzq0(w3;?FnJFJBc*4Y@{sx;wb9Ig_TgCz0~C<8*b*RC>yd;khzj zXkY$txLBJY(Qg7;9H*g@-^F`khT?jW2C!Het`B7KYQk*PD(OHad4LSdw(^;nH?|m$ zg!Am@bjhrgMwbSXRe>RjY}~PL^*F>|;(mpO3|jMFIDKt0!R#WQbDrsi>nr6Dv&kJU zw+~|9v2&2<>_x8jWn|?jpmH;nWWRb^YAn-M&iUq zbfnf|)alcR9psL`?aPIRp6SB-x3Yz!bkl@$%MS<-Rw@gfk~jl8Yku2rb@F^H*x}Ca zck{66wDs2O4SaXHLQU5>jEt6^~|00yZ?;qZ1pB%aPhMpPVDPPl@g6Um6b?TR}u+)+@^ z^PrA07#L>?gWbHd-yeh}e27#>k}#TVgBM@TA-KpO)-V|}T4m81UW>Qqwb;d}gIGw) za8}0I*=>!V5v5j!2JS#t-0~X(^qw=VZef;u2rd-eq-q;FpyX&+C7un{s>)wIK@QGxp=up&d}TeFW2=|A*A32fVAqGwhWk z+4kR@uNFF#iQl`A_T|%&6;{Z*iRbZq-a$;{9yR^a-*o=zNlaSEIbHT&A#CIM%+Ol? z9xj0K)-s$;vSEF{U0Ae<8q?SM4DATc5U863mmOCSu&D&|>tk@VN&v}DBP@Y5@-N*b z!@u0mbWIis`JXAJUkN?O%E{=btvRqmR)(=Vdz#$r&D=m;a4z*ab~qnTQt1xOoZ=@hm_miN_`@lJV!7R_MgT= z#bYo8oR2GY-jGF@3s^kQr5v5m&$B%OOo~e6?=`;@{A3RHH36Y~9Gs(uMKea?iXak8dp=`^*L0C*0<<;hvenFjc)wg3+$6L5<&fo@MzXXC%! z!)C{P9KU~&zw1t7hjK1rB{HDnpvBK-sr-iejO?1^pWjhmdcn2)(-)(Pk&;dW|~%B$^;f%L%8`XCrFe7h2HJLR)yp&~t_tPQ*?}%;*Tn z&H4|43*|WHO@S_Nw-`=t}w+Iwm5|=Eu%pqS4muyOSMLTx`oOogK%nI&H%7 zt~`7X7(nph`?znKicOqZZ|#>2dx1FXZ6X%S@1QZaS}-Q#13IsrB=4*p*zdoZ^JWFu z%vqaKMj@o=kV}T+#$((HA>=xc__B+;!Kjr|1hQPBILJ+ zVc3m8$ht-0MU()J5sTQ+o6p&dw0A77@f@pa+Ro%P#xODE2N-#I2rK1mfXOkY>|nDE z%ZcdXIUdd-{9>fk|Hm115qhoC% zOwB_Ob|DFlo^iq_K1D*+E!D!d?7PB%RWF5YCwB^qyM*Yg-Hf8Yfyj7y1f%(#Fwrdt z!$O16(xt*2-x;tW;g|5|S0?DT6bcUWzkR@796T3++UvO}x*3GeO6Ev(d4-rHITkB2 zgtZt*F_+ujNcQ46jgu)Pd6hF+hIqjB2A{o^MWa-685S=VrT1gkkVai2$;rPV-7ERD zE3Aaxepv?B+#4_qkzlHEGOSti0rsr2g@w^b^ou4SHK-cZDUmq3(gSsea%lCCTNKK~ zF)d#muAJjD>r*zBAD&J<3tv!}$`HKzY=*6I+9-QJ7vV;ukP_BI?W^_ClWu~&oqQL< zze^h52Ep>na3p%VL!`{ld;xX6w`jB(W!tZ zRM@=|*;?n|*3gIASl$yHbQnteXXE7IgLq|=3-9Ve%uLDPI~NHE5|mldI0v@E(vE4y zZDO9*AuRl?CR6ACvnCZ>z>>8`;T3iS9SM;TkFdwgjNQVqg%5=%S2YW5eXk46T22ZB zS5FcC?cscTRb9mLyyVMC6A(2^8dv5PP-#XVb^BSP+>Ouky6y0vm@_JPC;nN94eK{b zXQ@87nZk+_OzExz_h3FkQMU!W=M`X8`+Y3o^Rj*G;!yKR0v2o`YR~O}r(FO{vzOzS z-$+cwb?Uq{15W$=(RkDu$rGay;5Qe}WlyL+<34R@e@^al^3c89M)jZhJ$?5SI>q<( zM*Pe z3$2BJ9My$H$CyrP20C5E+4|WEY!7BKv%S-}+vg+RFAs&|sX}buUxR0K2^)Ctdc6Gr zS^IZDX|)P__(_JP|LDR@&dG2+>xZP15j3MY2<3ww;m`RlG>Y6p^5qpM>EZ6jVLPz3 zjn71L1Sscw?>tLobR4pR{`YNgiTw`;>x0l1?g8!3A4sRl8?&uLFkLJXt4@aF)|&me zm#KhfJ!u%;F3-d-3}w$Q8M4oFCNNQtYv^wIgp$c)*qs48CclN)1*0bc0b%f zSWOmwo%KXxdokoJESU0GE4E?WB-S{749oN=gR71Q{>Em&e@hqc>`3BVbVsN}byL!| z9oRMfDm-7d;SSFkmFPA@HFZAjiu}a!RkK+3%~|Z*?LjP7GoACMS3vW*FEY2KpycsB zJXTZ2uswe$?OPpPUfM(J<-F17Ylrm~X2`zjg3I@EDK1q4FCXzahJ!V}ZSq0*oDnG5 zvW4#Vw$gmZMF^5Ofb@T-V0SqY??w#J?Tx>9hx{U9Wp3f)Kr}o}k)P=(J8 zI`Q%YrOmOxQAy6Vs69bLW(u*^do7Yrnn3Qi5vuiuA%Nd6#(cj-krFDXex-)m!OtjH z?-1S3_#l{{dy&o^Q$UePC#jx)M-fkXE@0PTN|y{o(Zn;5DbB}#lXIYd`7~B8G=klZ zzYr!1S*0JbOY5ANTBH|iC{SX9dk^5w-_uCpu1CkIaxA#*CN{pBf_HbmQ>Lvu9B-}T zo#qHMOw7lV8`JUa^km3TDwZ1`!TFbg7;L^67E&tk?A(a?ECN#>W?}S_RHWY9i{j^U zP;6@_g%)M}E9T#6&huWgRR#H+gSRoQ5W$LW%pzqWdvijM5qA|0j5&dfk>}Ce(G9Oc z6Xq*5jB~Ylf2?dL=aZF@zoZdf?-kYZASuf|Y%aP|fy65ASm{>>G{E z8E{ZEi-;<$P-%zbU}dIl)5{t37jck(KNj_A<72mASVbh=(B)f6eVZk=Krtp`xcsg)j zdlyyg9*W#gV))a*y;o}2@VC1Hp#=(T$-Fm^A0C7Yu6B@h;Jb+mC3M~WKyl{}(J`C5 z0*9rCDDj~!2EPl(VVfB4eA$RS0ue|QjHSch&QpHvP4ZmuiLS(!(W5L$3RGNzi5)|j z#KfuWvW_3C=u2bT+rWB5j$%Z4E4(+$u#F#@5y89Ot5^?>d*g|aU2#xOAIg@Ni?a#t z(oFi-TWGg)59*=GIDg6st?#u^5VD>Ow~wc$5&ZMwJXWb|<54F*75?KI$^GnMDoRO2 z_njilYQBcaD_20yn9zFlD}}|?)40N682VBHVu^OJE#M4(!G8KrZvr%b!4Zb5S#>2bcp+4>dra1M(x#2uMuAhoS z<#8l?@G7YrHPB1ba*A?!Ozv9)$lZG(5;mViy;>@o(+g1`nSmBtKD(AE#KwKP>}%n0 zHaJv*DQ7*!lZHV2U7Uq#wc#vm#Sk{`JD(w&^O>2%c4Ur9VN;~;GfDdfX5o~|+{2f! zk{29S)RuXfbY0Ux$`p~}OJ|%rR1Ot9&&JWLo!s#fUo05m^xe?sy zunzhYFQGPxyMN;g5D@DC-Gq^BdcXn}Rke_n{q|;cQ=FK(+yKH|j`Cf|RqPgKV(o{c zSQ%!BpOI1I8Op5$z6UvTFAJ9+&qr6}aNMmPhw~@3wH!% z3hACaKHfZnQ#W%_oO&M{#IHguZ3Rw^xR1rlh&fg|F;l*``gZ&_>J>-h`b&S@Szd~b z`M0pvv>FlX4#Gv>7**q5V4bQAE2_|8Q~Ole>$X3rh%G|m)Yp{NrH_X1wX;)r)u0H~S$7`zePfL02h!O`l-Y>KNK3bCmur-$|vN;-o%n z5Y;YuMZvRN@bS-Le!g6Yqp!4(Z81RJZ}KQUvxeli$-s8rCWKs=gi$T$w`IpU}o3Q@AE%sAU2Ui9rtqv zr6TNKI^%^+I*R%8wUOTyCZ=(hPSI`X+MBYe$_v;Seoj~69tfS}$J|GAgwDTwN88Kn zkmJMmX00P|#$+hOPE17u&(dYxDMa+xJpO$>h`KeL1Ci=3@cSS^<0kY95_HoAXX?8J z%M--tTU-E*eHcz<(I2VpR2xOFE}>yV_t4dk7pU{E2)=MW(ZF~!6z$W%fC;Xf(hAZV#;E8-Jt`EVcVQZi{#RhV+iqL&%f+hO1@%Gdv^t)_^34hiedJKm` zwK_UZaWAlB7`~lt#gWc@-O5&Uoe@rpb!mla3Bh+hwb55b? zcoh!)1HQQTRRkhi;;Cm&DFt7-PSf+N==I0nBsw*j5S@UClTRV*^OSo%44B}6CVQ8Z zgNi+z#s9b-1Am4v~&JTu(_B%|>?}UQ!dq{TN zMasegY>GLE8%{NtB7F<4iRD=Otq@LMJfT%C2Hg%jd{E&&r*9{?GxGpss*Uk!+e7#i zPhivC4cH|wWj5r*SLB$UKzC0!9m{ivSi)%d8mQuX;SfBvd_=Zy8Yr!L6qb%#i^}bX z(NMPq=VuSYQvQEZt?C372Ye)jCFS&TLlk{#-a!)#Z&R#@1$?dedG_fMD8m=2d*t8~ z9S^B@9Y_vq;y$ixuz8scr|&+9c{G(grbWs$1|qLZS&_fyWpzOK47pS8=_zJ22eoTtHzLV|^U1YfpMo6tSx( zhu-aKrYjy+nDx{OZ(h&Gkx)4t;hd;RUlO1lnFf2Fb&(kV4^ldRv5D{Y&z(O?l@GgV zM???(?s-UKn~qSNVic{84yMc5CG_q=4V^oniG3qEulQ{;eam=2CbiD!44%hl0CJp* z{DD&DOJeAIE4*7Aix8(um|rZzkt^|dxpyMp)$OL4dm?B8&q^7dsV3RR7i9hL5IOUG zRQ)CgmPZz>=-6o1Jwlb4CBDYH*mQVJ`GI*h>YT|S$C!K%#{1>LMwm&ew||jN&|`{^ zAH(^BW1*Dwo%GtqW5x zMjtw3U{@NBI=qBs>KS}lm;&vQ42kc(ao`8{B0pV?C3oA=sc6Ka#~QKma`LqLZU*@! zw$RGi$>jQS1?6XJ(xWsP+P$EJ3=&V$&DHnGB=|GwsobNXS-k@HPy@=1v!?n|3v!-6 zf_6_gBv0Q|a;)K!Rj}p zETe$zRw3T(;vDGR6EQkS6l+C`Nw4`1+4Xjk+!5fjI$@P(51ok_Ll@>;r46;u3HphZs!j}mUuw$&5R*t8OJ#E-&EZyiP-l||(q zp4T;=ODzliX~oh#RPfuEB;6;I>z5R4-N5%sd3VsdLX%mA7_qG@Ie*ne0TnVq@H=$_ z;n!m!&U==6-^5_BFdwG}HRJf#Yf#HbR*6>=5&wkd!?xr~^cYdLY1aTC8 z=Q)rR1JW5ubkZS;3YL4&_H&8!#L$}-Z(}IBk$}}3qPY)nHEwE8L7(OjNWHP9-*4im zV}3GKOIMNS@j{+sUO^IW`83^&cd5!Y;R5gSmAW`#S|RTuKAr*L^)yJ0xQBcCWpLvi zzd>-uQA-CrJDQElO=Z|7^9Wu>aTvpQ&bdE^^Gu*7l>HXs-)$GD=euIhjJK2`Wx>DW zJacXohQ+@tktkh=J_GIx-=z)Dn*I1a>NLbQZ%6*l`EWCufu&da>BgAF5ao0CCkDH5 z`28f_otT8i4Wn3_%t5B26w6k$tYR<1XR;4}R9QwyEYDC)sz#ib5rM5UPa!bG8(}W9kl83i zlfzZa{dE;I^T|BcltJrIHrs5jt-D{os1vzE@Ni) zHAJXig2wM?EIR&&wl}+CtCR>wCMLDfEScLR zIy?(QTSu@>2Mn0<`a$gL#7d<1@_UJ{45A`~@M=#4min(l!1PE=dKAugpazuboIqE1 z))75AP1i+|Np8^wk|+^@<+n7*NA}>q(SzBIrVKF7rJX2m3P-11_$hJ@8%&GwS#BXV z#|}rTgfphD%tePigME$);^XC^Jhqj3HKXv>u^1DK0yuYeIaEz2LQDAry>k6a3yqR6 z_LLIaDJRQf&fdbUIZ04j!ZV8nquGx2eX!yg*7rVVaV69qXO=S7!m9ZhTS7RfxRTv| zyoOzDRAIaM8Gf&nD|>v-fu+lIVv@%lI9YAQ(7+~P`Wux=2GMVY58Nt*E_zu)tKo}< z2HTI}yNd)XPQ8Kd9G#!%3wim{3YIL_uGYZtw=v}K>NI8)2?g)+&yWGiP_ai9jU+qg=VbqPb-ef@xIBDXqKip zpAAG;p+$+$zj+^4Zysl1-P(qU3u++G`6E$24LnyLiEW;t*g4)2t28sPGNBO5n~D(g zKhDlOo~!r&|74^>c9fY=c5ywga~&fjTG~ZRTH2a#DwGkj%8p1htSAYua~+kHk&sH! zP$C-IMN@sx=kMQtKewCzy1Ct6y{>bv=XgBs4|Ez{OKgDVSm7?-H5VH_T8HE1f^g}R znb@UK4wK|pptfTQwz{E&hnA=anch7@&({ncdSb9;NiftpN5Mw^2#lWU42=rXFy*-w zd^k`5-Fj`XGB^!9%npE0j})vstcVZ2o{WRVhal*~T^RE}zKvBZjMEr~`@dJ=hl|$Z za+jr8GTQ;4wN=F>8_FTL;U*ma7Y-VlX0T|t5qy}X21oOYVc?Yxwy~LrwY!8lKXN#u^Y&JZt5j^If!r;C|J}kR0a0LFELxQ*@%rBG_cmgkwVzDqs-n9%C`B=dfe<^t7 zTY|*SMnTTC+aU5g1oKY|xtEI;K+kGzP}npMuEiaKewhXEvO&;6+BBnq{AN`2;GUrG zu>`e({ZR5b3@(-lOx;(49#t<3Oh-y%qf09I#<5@UN=jg0EewMZHs8QI!4C@|8u+wP z6|US7fLB+k;8%6^Fn^N=o?xPn4HjI0F&Fjw}qathx&=IwSBK1_ck$H!n!@m;)ceCxiBXlfeIM3`lQG1qXSdKGE3-GxUYL z`|cSKv(_J0mI?DoB?2~?X3$kE2Ai&50EzdX;r+5MVQyItD@it7b@PYiyQVKpU zHV#V-JL2qlOYy0Bj@U%P0KN^R!c6B7u<@M=y2(Os^}Pwjr*On z8)X$lAiqNu=&Sliv}R2(TBDtc(iZaQndUj9Ag?btp^ljnu{Hul=>!RRe z)>?S8PVkLiKLX|l+h9?88|c@b1H!S$O})>3H30V|;T-8m`^40>{`4g2cW$pcDJH?ytK)v!ruh0_5f=4FnGZVg+X4=cC!ZuW&k zQO%$?+zR&`&ciW*^|dE-ADB2Bg0EjBq)z=0RLo5wLRDZG{QQ9odS>A>8HspJ;0ZkK z?^N9F_XT>tCd2buZxpKX2knqAM$M%+g*j<3TCTPmUELo6_A7sY!JElAI>i$2j+DiV zdm7=_tNS2#+Yp~m_yt7!J`~Km4qxZ3g?S5o1s}CA`&Eg9;|C%jMlk^vNgjon87rCT zJC-vSe*eY2^F+8}#|2#YV;}ys1mg^qENt@kIxd+Livv1l<62ckyf@VqCY`$k*ZiJB z@6}hr87hY~Q5dW!6+B5@Pob>-D`Oa4 z4?aVb(sk7IIRlQ=2=g%^zd=kX39LM)!Bit1^#1M(G*?v%cK)6Tz144!<+~UZEAUE= z)(Sc8%dUgH#7j{0dyXbYu0tOLzss9%0$*KYBwTd4iO%N5p1=)E4i*6#*S z!3(UAeGHXF%Lp~Z7+53chwR6EunuVXY~7nM27G$Y!F%a4*l4*Q?A4_~_qZ}Dx0Qna z87JUtNFf-RHGsEACS2|Ni_+~6LurL1zSBGoTgZL_9brcoxTW4Y=lTq2w|xPgUo^2< z(QrUr2xGeI&j3^sm=haQzXDDUnPG`2DX3?%yC;44*Zxk3|<3HkuC1un3u$r9_o zsm7L?jo4{u5f0Z01cj-oNO#mXG%-8~5~rLNG`Uw``|~8|ij;-{>xr;>W+E7A+=uVK zByrA*4#>(bKt8vxqK?fI;NtmNP?)qF;8rh^Ws0mf-g<^QUNP`5!4EuhV-_4pWUmCuFP={j(JRRp~EABkO5M&bhLVSzo_58C63!J$P1_b0?) zso#}2c`S#2&Pc>N=IG%;6BT$q5C?8%&tbZtZwjs$g)KiV6SPTQf@b9+L=-%SqD!|0 z_MmiY8C1w1R4 zgIKy2HVWTig4-maUZK!#q>7D|r(mu1F(7YhfLg;WuyQ~#mXtV;JF{i+HzBvB<7Eai zi?)YtQn$cet_n8(6Niw&)98=gE$a$%DY*AfU}bJCgn=_zD>@0=sVQtY{mHsht4PQrO92Tbp;vQw40`8sq0mvt9uPP%h52fD&EHp$@GAuV zJaz$3V<{oGB?dhdG~q2%)nKCG61Y9a1PXudM;^o4P*)ZMc`F*AS-BBvw=9O0=f|N! z`9zeU^%_kQ*igahGvVxUTY+zM1dUSd6W%ixFkxk|z~hd9pwF?e+Vd}JoooRp(jV4o zltJv>W~iEW6=oa|vY#(_3trPH_=(TH`4FHjJg2$`u-QPIN%0)Q})BWn~f+i<`y!|Uw{}F zL#Uja04Db3kQ;CtqV}AI>lY-2T%K^Wz|{bD>=e9hk9NapVdm>_eKCr=EQ?JXeeva| zv+%NrzYzAY5)P^#0!5Q&kSpw;PChw^M^@|O9INAy@9u&c%&(z8bq*kXXDzIL5e`LC zyFu@SpiPzC20N2(!0YRupwmR)Qk%U488IX9HETsO`@>+fbSMN2Mu5w`VC1so2-5m; z8Qk|;;%b|vSVnvtUbJ0c&b&^yer8sJTDn4Dc-A0D>1pDusmrj>K66}nz8x8K#emI8 zVTUK2U+IaqR%5>9Sj!)@LtIo0+TEUrd=%2HXYU!0`t_HgWFtKkofCjeQ9trI+P}Yl+o9HNvAB zBQT0R1-&CP;O3n(@M(z+ELRie@cnC1-<2xVVBd*W_iICC$7mSuosMiRM*3Ka3CY3*?S!X;E){o49u%wOaf{0y$STi~+`O>jZqJnRuP8ehrU0Q-G!f)9SDNO1S66L*~y`AxGN*9{F8F zAzq)Xch%dX(SkR1gx+^#S2q(a)i{dePf1$KPZU^*XA1xvE1_~Tg_A-Lm+KIRCM{Y9 z`=ft@#=qwfX;KPfE4ARqkFnNTS;ipy{RO=F{SS6Olf)xl+yb|KZqOxguDPWm@HV^- z&IyI!b#4Ys?Cr7MzkDOSP8fhu8A6t|rV92{)xtKD+o1Y~0VJ8X!p5G{ke#y-R z(Mnhzp^m2b3|U|32!Y5pU2K+SjOVuP!(?_OR>;xDnX2y4c12*SJWha$Id9RL7bnrp z!_p`%Sswf@5Lh0pENIrI!S^!WWL&biFn6GBhh zRVQRGJ{EGO`eD~kZ}{@fAMJ1v30kaKxMCv4N;TThGdmZBMCYO>?hK@7^n=p;`%qT; z9cG&xhvbTX=$&ybTyC|%qr~T6gQbo*e9k3sTzUoBZheVD<&|KoiZ=XyZw71XJit6> zI;^bNj#?zr(Tvh=^g2$+wmmo%r0UI~?c`2Sym?;ms>Q*~bD6NZC>~Z?%mPbcudz0S z0ao}PI!*lHTlgM`y>JL-t1m`0?jZ8;E=GQ{+mX}vJ~X^ufGwUF1CqB*Aa$iN#H<#$ zDeL25y00swz4(aE6i3>|j1er^y%8F1jqe*_?dB-(`XMkVy3OI0z?M1BOGEWKbBJo43#|q6@R|yF@z#G)PV9LQ35@XQ zpb|Ly{2kOwq=VnX09fCd48syFaPw3neE3h8sR_@}x;f=AY$tf83%sECs4LtFI|g|U zr{H7uA$YnmigBuFW?Hh^nKqqT#&>)^qnEgn2`>GFUio-{(l&2U-+BVX@;89{;YH9D z5QgNJx1#p8_vq8Le<-)`8_GW)i1r$};g3n{@%;r6*d=g3eyvUL?F$oe^;@B5)R2W! z=VjpQW_gHKzK3Q^5WL|(1s1Wj@U9hRtIksd{Zi?3*smD_Wq&Es!8eiLhCCE3XguCM z>q5~u8ZBIxgJL?R;gioa`0wm^a5a|!zw=wsZ4Ek1L5#Qs@jD1wQ06 zl__xLEDN4V`@r;{kn6i>1H9Nh1`_7ZguU+;g5!5rSg*Sh%mU3}=+1qx*AT-B@?GH2 zC4uuwyM;dCIQ*E^1mm2wasL)moPKu#cGptJ&iw*&L#v%^x(IEdi;0LG!fz(V7S6ki&+nXj?`n(zXyGx3wZ9X}B5WBEp3X zWMO{&q6vH(^I`op3s_#V5jd$>h!y6K(xyA$c+zgTE9f~(3#TAe*Ko8yJqNiM{6v~Y z67ZXzK~5(uq10>%T&^>Ow$E#!(BTjuhwY#vwiGe5640ncxkx9l6RrI$1*h%KAvfz_ zv|m00$=jr$)gj#|ty$nKO|eF6TPDH4#AR?W${!wX-wPjec7e{tiBMQn2F>xmpg#B& zEXf#wNA~ODAl0YP@!$`7Fkw2}pW z%`gC~8{cxF)gjn>svO-D zwD{h)zo62^W)QQ<8axADAl_48Q_fYx;r_aKhLjXGvw8=2WnO{ym-}FpuY&CdreYeX zk98OI!ngA`;FYTybZ*`x@SMD1Rq{!g*mD-9@0NkONdn9DtI)^CUlhEx<=|Hy1Qjob z(4HqM(D2?Gn)Lv@qlM>7W0Sxebb@$~HQ?G91B~SrxH9`1ymU&1(+!^Bf7}Cdt4_he zzry>hx)6HAm*Aw&)3DhM;f$6`LFU6LP`hpxEOQeB>+MyjXZjtqYo{WZ=?R>;V-~Qn z$OBp*8N$MS$5Fwv6129z9kmNP$&D`*!K3CQvU`+>)GI$C&R+{YT+{{!dtEr{^%7Oy zvx0fO1O!-0v+pCSmKMxzh+It(I=;v+-84(O$mj{7n&qnyY{R3JbpaiYI zKchwRLN6q|Hy$pd==pCM$gb)|2ItSCx=bOXQ(6}qx<*38c7g4XGYM{o%0tqAL0{b` z>}{{ifm*>A(dwcM|C^uf-MShwOMV*dp!CD^Z%17ZS7bKxXp|P{*h* z@VQRNNs6D0y_^m4S(9ma--G*L_9#x!G2Dcsw>se8+Yb2eO$ltfoB_rU-@slkLwsY8 zF*d&>us2p(;MwmL@CO?^fM6#$odQP zLHS@R*qCjBK-EMLJL(QAj`ty@NMo>FC1g$dB*TPFyJ1qk6}Xz3K#HHhq)NR6S_U=3 z9V(9bHP2zT(64A8o&-^Y(ePqsIy9d>19O91z)kTkiqKUT{Fl?9EKCjND!V}Q=6Nu% zZxm?X*$DSL)7gj3ceTQh5LuUz}lNCINM7Jr`8+d!EAS|@KhFWun~u*0RmDM`e4WH zf<=k0@Wps6M1D~R6tNS!1?{r>f0N-sgeI6+3w-}5S^T$XHeT_?1-nlKEDu*XO_YB0}U|cQ9c}qTm>fR160^a;Z#3GJZ8>I zn6H!ri+7a3w37!xq!|mdgnzTGJrBx8H$o9U3K#Z01+SY!aMA1s$bOoNPsmQh{gZ@w z;K(`@FwGba?X`oklO-Vh+)VIVuodK1Mx(9UzoOOA!n<^a3Zx{s!1jqMLaxLH=ol3P z$q@zcchgBw_FfEkPQODx2L#XV3m@ng_6lM{VFIt-6Iw#lU}sAlYW?vLB|Q3u%-*!2 z52K=yQLrjnv8@5s&96kx(l5~oMO83pkcXvzYS9Q=5mGLxMnOei&{)w3$QE+eQf@vI z_}diu1P-A8Cj3Ks?Zs&D^h(s!G-S;P^P-Oz?F2oQi4YDb>~wl=p}6d&sNPB!!rx4V ztb)a$qHrAa?ni=rx-e@`QG~?@_rdlDi9+Vn4#884V13*>loOc@3rQQCTq|%GY?N_{ z!&t1u7r}GIS4ek-6|5_r0J(NHpm1Cp6gNCa2M6MrHlKcGwDnVFFsGJTWtPR1dKojH zk0_ypAKTD;@Av5Lj4q@+AqovIu|UICwMc)BE-<~OkS_3mRDG9#zl$8`{SxlYvi11n z)}6S|(*%dE`T(MVP*9EkhR!y+gTDJZ7iA0Y7 z-A3LEKBJ?)zmcQ17PPI>26@G=DE4JA{N$S9p~gteM*Rfa3}vi1Z6eP2JBguOI1|1s zih2C;0Q1PzjtO;Y#KZhja2ZO1XHI2sTD=;sTAT+K&4Cty^>+E(dQcE{K384}`bZTP z&h;dtzP;b!fSv_jIY|dQjUSCS>b`=&v)QmgcLGFwb%NAi>){xQhj@Qs&r~-R?lPCq zkK2N7V5b#qT+xp{>Q$qQLUy~BWfq#YtsCjdj0N@V1#s)B3B*S)}bRyF8p95UP z9$2$87*-1$wb4?Cz)Hy+?mg{?VRd!9e2lj6_k}(&@h_A=jE4o43!$+(0s_Mp3i={N z$b2;(o=KIV@Bw$2V3Y%jv%W%i=vzUDkOxT7Z7c5jbgWQy_HvENGmz0B&qu1iH%=p<m&s>Y> zDMO^Xz!810Dn>o`t5D3G{XoWyrXDxV<1-FEZq=iDCHt6^q%4{2iy4wM+)^ww$3#yQC$N@Yi%>moUQSiw) z2dR$K2cr@5VX52+5D(rD?(Gwxe(WgN(Eb!1TRR!V&ILhA$y)f!j)m~dW>jSo0(OEg zZ?g1ju!<@K%ayaC^7Cbsz>C33ule98?8$sj{szzI3^s@~!`rS1?0MlCtG~Y+qTdVL zxxTT`EE9!VPfJ0~m-#SmC=j6771oJ#g}e}k==-820XoRd3EEZ}_G4+w~GGG!q>sX0Q4*W#>A`J!Y;}+=uZy%^xPK3u+ zb5UGaG73v9MUqO5DA6Gn^;WMy-@NZY#_)5f91_Pn>~!$#X;W}xtt4)C8wWWXmxI~2 z98gvb0-Rw0*Ax_T_H2>hAObr8_?rsH`vqZ2TCAh zcVh#H{fN!i$XSaHZ1xb0~*3}J0Y+IMQ2Qa_E2>Q-NPY|k372A zBm=Q|&X92-81gpmh3%Vj;IwTSdrK2@reyors=8}Rz^9*iQn2yVxzCi8%Bv@2_8uAYmfvIbUkohf- zou?__ovQ^sO`5_yqIXD|4JApO|DhM2YETos5|uzuSq#C$LYC~gI8^^9OYj~& zgmZaSuy*%J*r1pKs++=uY=%hYz1L&r#D}kpN#b9o!*UeSEvsi@GRz?>X&o#T{KIXx zf+2cAFl?Er52NvEXew0@_*2Gs-Tv2Lu{sP|+I1nQ-Uia#!ePXtFxVtykY$YA4o5F* zLCzc@?@a3%j31E2`Sy|+f4m0iaob_x+>tVI-zHoV^B5uazu$#(B_zefWMsttpC45* zJF#_pH*DGJyUAkXy8Y`ch$R+0qXJ3BiHNQI|NU9K@;8m|jppy5!+i30bADaHMZPqC z9)Ih28gKh|6(4mqnbw3HrK%bBxbmq3tdMr+#UHQYN6B%#Mr8zVdgea2xx0~B*mjjJ ze5)in`Ds6Y>)tmS6*_|++&hEw`e??hJ=@LqE|20}J6`azTX&Mo#CpC(EQBA?W563K zUFOv{T;uCrwDYFnPxvW)PpHhBCCt%?1hRB~2ECFOE$Z$XMJztmLyM0s^6Xwag0znshSW5zOFXA{VpiVw`^Qg{B4 zB;orthUkORSnkRD5Pm`CW&Qw(=IM9|=1%J)?s3I;I$-*g`q;*Dw=U$-4Go*PT}MWe zYcZyzV_pPbzcqp1+aSrmDG}r6PUxZ10f;xRI>W1~+4GyOWb$KQoA9W575}g$meF&b z$6T~IN`^)(qHl`MQ*0N?$E;6aM5jM+X7_7(hcA!#`zz-0;_O4}@9}~!xVMTwI;Mdx zbdKj82lnve=IHPfCk60}GiCWrl4B|!<~`^C%^0Tk|Nbz+pO*0jEq6h*q=2l?l%h4G zUh@h^+j*DiJikWfC~qX1!_Uin%C{=S@Mr55@){!$2^{l`h7?|++iXLaS-;CDHfpE% zOE?}L?&0R{nZRo)uI1lfP2qPR&gPxR#W84b99?@%g>QB3q2h0D)412?iT8GAqV(z+ zH%o((0Mi{j3BJnHgZ=y>ZUR3{o^cVCQe^HwLLOQL^Gd@z_}@+ieEpau*LlnNEJv$V({j6U7>jFX+CM40%e+?<*%TvAdGGkB+!F*cT?mW5;ZhmB9D zL5>_B7rvX{DdEJ2ElT7fk3@4nKaV4L;!dh}Pn#dAoyA{I$mUHV-Fb_oK;A7Wi?2N0 z!z;ZW;B`ta@^`lG<8ypC+8_FiDnDOlXR42)QvTnpVHGrqin)W2u(7GE1j2?7i<9lj&*5B;21(c86pU zm+|3b>y>`diUMmZ6s^O> zZIEU4&U55lj2hYZ(UjFPEFwPGi>&xFOd1MJX=*toQln!@$bkVu`?Ogvaa%SeZyD)* z-b7kzq}fM0%egV;spQd;k!;>nT~>VSVzPB3M@l!Tup1}6Vakf-Y080_#By*SyY0zl zc5l~1;@bV5Y|I+LI`Lgt)Z@WK{V--TeB-1I7yMJgYxuW%nNcQEBEqxQ&^{Hm`wsI|bZhxJ; z`q)nr`xV)e$T_TdjWg?WIG9^`OOfnXu!-<#wUq45oe9{#96~956Nk@9_IQbJ9^}& z39Gs%f&D0P7t<}#+=NwSjPkr$r14<{>67vxJ54FGaLOMn3OqqomnLApPANLu@ja=? zxm+%PCx_7wu;hl#=5W-(kexo$jc{iBh{u~E;yy8xSwE+j(R?tExF2j{Dwggd<(J2k zUz%!cp3-S9FzyjiSU!%ex~|SX?H$Wr`hHF{>+!+zXdxckN9irIVyZsfeeOJGrK?Y_ z-6QPw&Yh%j>^man^oSVFw`4D5b=f`$x~KkqK{uA`ll3G>n~=kqTWH))#N|cB4;V< zmiLhS9c&{76Rwa1zXzo6cq~`-dl?rQd!N?3hccgQYKTVNSoU|QHdzu;Nrrb$VIS)# zv0|1BSYubk-1pUgENHPH#>_5ad2|KC3JNMZ-Qw^y91 z4IiL38q-8^oIIZ%HG!vHT72IgWB$;~P`>?$CjU=9p2|2Bihh0@K@XYN(%$6b)GQ*D zw<^29N4z@7A5bpjt@gF^GdJ|{0Ri9mED`0W9kJ)roc#FFRVBQ4ekv~t`AH9%MA7hF zPq=YzGx)D1Bl#UuMsgAZ`tViyNLcgq`=R4=FLw$u!hec9LERG zUCdwMF3>BXVsyQ+2B#(ZkKP=g&V`vpFeNvP_$40k*nqC%+np~`o45XaLb^PC5ztBJ z2mA5AD$1#<++M11WImtqK7@MwSH}2W5$8FlA9UY&d!Dh|%jdPn^16RH-nF}wSN}4I zv(8K*H~WM5lo%tv=*E4zdHEDRSJ{cSPF%sKX$R4rX5!3&Ig6=v_)`9ry)U1ZT1j_C zJMm=63|?)WJ1KwkmQ#OU$yiOwrlQbdZvA9Ss=u(6X?v$bz6WIvX166{(a{EyiejIHu~}v;i>#ozJl-Ge~Pa$TgKZKMe)sxdii~$&hcwEdhzJN zLtfME7Qc3GE59jVHP`T|mup!)#AIt9;QpAH^H*Pt=5?Ng@k<^ary9Gq^Fi_HeBIGh z{=cRSes#40Uu1ufZ>h-QLk{fa#S3e>f`B^4_10i z?K}_9Z}78jadh&GNhDLZm5~T`6z$TtWs^=C^WQ9V`0s0f(13?dOh{loo#(DXf_gK# zs+o)E$dmWDOOhr0$7{m*)eGl$+>79=70>ZkrM3ANTMM{!tF&^voiFH{EFWH}CyO3= zn?Wycc+DJE>mc8%)Y-8wx;dx$&iv@KS^T5U8XEYYFEv`C$p_a4F=vae5}&K`{Nc^! z{DNE0sZ8=QUTV@s{+`kadS(UBbWAxzD*F7G_C+IqbV|9RI>oUKOlCl%)N*P2h$C5yYb@VAuS=Q-+aJBh5+ z`OY-OPGD{`XhnKarKAR?*@oQoO^>rNrD|Ip@zt zVB2Lr+`3{y_L`MeU_vN8WnWpVH5iNUI0z3Q=wxt zs~E-OPdHnh2i$Du+tlgqd!qa~n-p7J;XGWm=yIEH^kQ!X9X)0zRY|!(bAFj}hwyLO z)6zz#u2+ zyt%FLJg`L0+f@{mTt^#(+U~}Qt2F4R6SG&hiy1t6f{HF3qf5_Upwpu3s6~Z89qH7= z?eYG`&HZXZ+3~NK&bkh=a=i{&7o$SgGZ#fMb9XUCBZZ9XS-RZK)?oU5S2g827f`5g zqUR1?W**6prrFE;@YKpYVyCCXEqXCYByF3@WS1Q$+cev$(uS>!@svR>zBIi2+JbTX zx5*;<=#LJ6;&~69w0I3I{kVycsdXga&v=?7#?Wok$#1ESOUek44vo)ibAl2Oy9(^$!qwA?<1Ypc?rPj9Xv z$tIJDW`ibQ^1uc67}ayjn&yyWl7E@a4})khee??&XdN2q`gcOsvcnv|_ir4%_G|`8y1bWGM82as4buD)=P}&c_-C|6 zltItmPp2-?U83^YnE2@KB`cpMGB;iTCt-R*WS%BTz9fHSCNYcY&VgmrvH1;MndrvT zaD6tyX*qeiG=)>XO$i^=E~>i~#w3?35nV%$xV9_PFZw6wdEXMwrei6QwEIE3Mc3$- z)D-$wMT37)OquExq14uRD&_wBj+fjGq4|wjbV1ZdYO%hInc>fopc$#;LUIV%-5bxK z*4gAnhz@BOSi&@DG&9O_U#L>9B7ZJASEMqTFxn|DgpQd{ruCm^lN!c-;WXM`un6BBFUfS8i|DKaR@D1(AGQ4?&7U4Uns?Wd!!&Kh& z0=4mt<>IF&P_pSW*E#K?=)#Zd+`9b%v?BvRzNS5Dx~ ze`xUa)83TH$bM&L|8XQszp0QjZB_Kpt~2!1xb7<)oDK5>Q!%< zE^axI{h&;?_VCn9Y#vqK%2EG8F}`xmJ6hI1K-t_uTCY2z1o#)1jdH7SN( zVy81tl;4zpQJcx=SITi|`*XC{&GP~%qp{vk@rw3itWy zR)+t$nW)~WXTrC0-018d^xXTA<(KYmBYJ=CkTD}?GT$BB8TBL;=H1rAOm)>?QR{D@ z`?VA4E$w*GwR`7i$C4--c=&#&PN`c3Hi@;jXG`AlZI zqB;E;pv9>^S0*8EM$&>yx2f*3A+E3@jRrWEQW;o5n@WA@qq0jhX^k@97;M1HDn!u& zvu;K?A(vV5^ss17o)V`fZ_5qMEup1$(gXX_G$2y7ltWw^d8mH z+eS0gYq?yxNZdFO%B-9D5$8R4z_lKfq_>uRWVY8oVYW_4VfIUGA{DbJV;`kWrKty1 zPEVj^1ApnQRBzha>`lr(JCFr!vh4dwIwa+HJh^dKg3Vq3lX!PNBj00gkY0~-I569o zQ4t#?)`w3q4oihQ`xisL_lJ@SeL2o5C6+D^*g$Ut=n=cH47z{JDY|sx1saxJPCxC~ zLwZdn5c$ieM0o02F7&b_@3xPlX3wwE)5$h;%T`|ia1dm6p604I*X!)x6!;6*<6mOgPU<( zmL}MRaADnb+-x%+QpH>+k9`o4DIHJ7Mkb^9yXA!RVdmKCXkv8m2Gt(>h?Y$g`tqM` z+@g+b8decZE~y8R{t-4LKm8zSLn2aq7BSM>C5V^!CF=L;4=mh}z&M?br+fJKJ?;)wC5#)(e z08>@1Mc9-s@<&RX4gGYQ6jbb_$~P71&a?M9>(!dHC)=L8)60n3tHjt3&zs51%PUxk z^OM-(-jQtF>3l*|_cK$*+Oc^~_>K7*v$MoX-^A3edb=ItsA^R^^KhzN10TH@V=0 zKsQO0kssT|2q*i1DI0o9lALyNvFDsZFpIRAYGRNN>p;y)n(c$FVwBFQ&&zMopymtyEKPO=Fy|R#ro3fL0 zAHK@%qU!u^#{imoI(DfbP#b^zm2=mI+vxQ8VB+fFWr_mUNCyeO_jthz^ zD&JnJOUU^p<;%FGgfaZi_{Vw@m$>C5OiP+9>9Aw;_BfLxCk#oP+)i)CtOC_)CiS@U*J|$K1?&4fJJ0P8&o@hn>Y*b@4syj%;s?ChM z;!IJKK?V0Epii{xRS)M}8%5r2drnrR)Q|uJ3DS74nMhw>0Sil#8T%1woK>Z;S}@d8zOf5 z77(NS2C}+QmGoNeqwkkk(Zan~N$;r>TvbaTiMkNa1f~CC+&nVq7`rIqvCM?NJNF*< zpI$?JVp+0x-Y1-VCYDpqyNNFI2cw7+@LjCN@$oxh%65L%+6sy9>jI@VjxNT{9 zN#qgcqi!}49ausq&792yWG$c$UPgq@dI71%(pYZ92j=2{I2rk~O@wC&yV(Y#JvayM4*q8m$^sD_*GO$yRo$mRG%$AJ82b&{^$0vUx zIdKC6;^xfU2l-5eO*~z1y^UCVG?30yspLX!c;Yd{1{tB zno}%^`7v4MM_IZ^)2NL3aq$i#zhpa^+v|({<-A$P=wocxs_pEBIX-N0gc|W(@qyUt z&13INk7N}G*R%P*T*$!TNo;wlHIr>?%J!qV?ABaIcGG|!dsbqR=!g^{^CBh5@$cJ2 z#u6s%gE`Mg)&UI`r(GfCdd95FtuR(=l`q>jVHNA7?Mm|Yk0-tEn&mqd4lvfikBH6W z4sv~0ibz5tiCkwP=Gzu0a>TBLC^T*)^UV#I^#vt3 zSEhyOu@q7!Ojna2;h9t$@vnT>=1Q_kQ;F5=)?)q6vTSNqI5TDERW5(TJo>#cgf#iY z5kWOTUJi$mV-@=B+EE)wZ>~J+vTX$08>GO7?kr#u7F9A^-|t~8i|v`zi<-=Emrx z-Z(?%zn;K8;tvufxichj_aK>6zn^SqpUyh$b!Af|I~eZedJ<^T&$u}rA+BG1h#OZ= zR;**#A1@Rb`(<*pK6rqe{A4V1F8MZ-wX}{ZwHI?2|Btlu{_F7z|9E?7XlWS@nkc3F z+~+!0B`rdcJyXaY;ggIa?Y$K3q(Lb3zR#UDS(T!NqM}mB$|m3U_fPmfzCXPGfcrci z@9UiFdcCgK)6tk8yGEXGNU`F#|2II5s!!w2lV<$0CKdj9vY-^}#&cC&JL!F|5-v1A zlE1xw2|wN-nbM>SQ1^_`vPL5doFrdzP zRy6U-ar)lc8C(k(KF4_+f9AF(JThn}sU<=8jfP?-?3zGz z^Da0Sw}3lssLGBvy@;O%e5NnWt+7wiVwz?=lRM{e3d*u)K=~OyHB-Jm!L?h zdQOD4_WZ|f6J+7;bGNyK3(Z!E>&8RHM+GnIoYFp@VeV2@3YUCKo}PGml8(&x z;ZmK_=$Q>i`HkY!==m4#@SUgTLMd+fMI2dVGB~IanB;uYF`OiL%t2MTxU7N@9hHvjd z@H$iY^(Pf(#I|5P;s>4{>3F<1Lq*m0)0<_6ydq64E&gr^zei5cPrLo83eKgrw^Tr7 z`cs&5AQ!T1ufg;Q^0Xpl19vdw3GN(n!1hJ?IQo1VJI8AyWL+JGz2iB!?;cKNi=J^R zcS@mry^94TFwP1DS}P#0&x0#fYYBJLcL#aqyp(JnAzF}r$ZIU zSQmG0(Df!337khwD+&;ry zHy?+#LL*m+ZRcEuweO#R=Ch~3nVq2?(Osoh=ffaUb|2KY_`;Q7V|rb(i97x26gMzb z#>J;?qCKwZc>li#@b91se_7ZUr%RVp>%~{;)FoC>5GR8+Jon&Mq@1B?zgAJ#p$yp4 ze;YTJr{hiOw=i-#4Ug60*xTt5bbo#`cgj7E&b;@Vc1quc1Nj~BL8=wp-u4PA9wc(r zuU>P<=4kRqM*`{EQve!L7T|M;qh?VSRH`qLhPahr)lxY~t=$JsfAnzrrWr8azy~k< zI}B1K3UI29g$w`U@Ssv0mKvCgKTgQyg7Z`H`fYK-_n0=BRP6|sliEv1LWj9sFVm=( zi9YSx-a^0EjOT1`&%le+k~!(GRMlc8(&An@>mC-oQ?pIZ*Llf-AO9*QNG_I-yVx1|M(er$wgC1sp~*by@0 zxd0AqjD_KgqU2R?Fl@P^L_STg2RWKbUerY4WisVNxS3CQ6@St)e=$+<>W5~jmE_(( zO)|+og{+vR&CQi)ppNrg=)Z0w2;ha^ZFHGw?pp+r+lnBH3oVtklc8_J6UojU)5w&# zN1)rFKyrH|;9-h1jUICb7DTC&>Xi%0$;RV!yl4xw{o6vG#z&F57e~lB16MfTU4qZ9 z2XbP8BNOD;bMCyz&Wk|dD{zz42j{~l-{_vBxb_p7&&k7^UilA1;$5TudlmJNhI zH3BhDHA%%TFLHOwf2`-4S>#i>2tm)cQgLS~y0nuaZ~hdJcMr|UKWAyORP!D#jV}Oq zX%l!ibPJm_Hqgo?^T>2op0xb%qk5hqbimgIzEyj}^bb?%(>iHvcTOKbf0&^V$%b}qb7<- z1@z*X^~>Q(X$e~(dx$fgZ$)3u*+zGv0M3fq(!*~k-}1fyJ=`~qL%Mo3rl@KR=40I<2fJ}`kwfH!NKVfK0I>fcW?6?!$3g`iErWASu z7gN>rgH%MvmVcci%7=N3p*P9{;fljVBDqkM$h_Q7_HSt6rl*>a4+nE0<*gmi*#X@6 zqZCY=cHx!0Gu@jTL6gkCaXU6!@|%()=|1I9(0EWl>L#xziOKJAnf^4;m{HBC%?uXK z^+kx)6nT>Us|e?ZeZ@7;HOQGf0qpIcN9tY&6U$s@g6CyujR)=3lH(s#sPUUSi)gg9HH_AU;;(Kq$ga6TM1AK5GHK~)^7zLYP_X>N z<){hptMiGJDKB6nH5F*Xvc>!_p=bF=PORYmtwGvw(T+DMxJIkTMS@A|Nov`j#YqL} z;Cag@Q`x_gJQJ?L>r7q1?Ja%5B@YJEZPGp?!sGa>f>-9j@T7@tYv8O9d`<{-AG+P5F7bvV7M_ zhd^Y6(j)o-n0f2~CQT7ckj!Ez-WVi^-YV>S#MZz&?SpjR^%ff0m`Z>3uEP@D3Ana4 zgIm2mlnNG@kd33)sM3jCYIb@u^g261^W2Nr?qdioQctBFyYF&OPm{?>2eMFx_>q8*{_E0<;B3UfiSpRv0QMz%@LXl=EHxtQ}A%U zDE;j4sdWBJEjs68G)_KJgv$zFuie@fx=DPHiN2ELaiTUlO>imJMrUPQjqj zEqvKimHwB0mU}&65sWxS)2m^zG-+H5HtlVN&Lg$3Ks1{&S1G-r|A!lt_k;efd>FGN z2_8OvP5pC1X=dDm(qqZ-!t<@5E49b)(kBGe?&NcDyDU#Hzlvb9lQ+@`?<{KgI1W#* zwWBGKdDP_5b=Y-26EEvA!#O68Sh4jyEw_pR`;tBgkM0AlkPZ-a&ce$>I;s1DK&~TS z6+WIV!HIrFaO$u)OcQm6fZbNqUM!j3=Xib_$KM}$nI3U_RAZA{qHjXFCFMWUIbTuOqL&( zT8UrDSK^D-Q@P9X-cVRQ6VtYG zDOD4xLrn_4AuW!dymp|@+RyR#G;jL-zjrWa#$&j-ZZ7m)`3DEXZbRiiC31B*mMWeZ z3w~weslv)r^i)VRO}-<+OTHR|DykTex>e z`f=I1M0#gDNA-4;K;uhsu#`WI>uk@^cR!?1uTV=_^7Eb`cC|k=34T(i$cs3B+6_Uj ze+s=W>@yC>ZYBT4$&-*((loHTkj6O}^Eqaj{I4a^{8If;xW9m((xV={y!$&E)#X|7Kz{dd}$ zZ``-DqJQYZr%$L|VKyyUnm{!& z75J8kE`kLgGzGfOF&y^S~JHkm?Od8*9oOs zFIxr2ldsXM$Lx5cJ(VMlJHFYLo#^EUxwW5}6&wdDsQ z{P~c0Bi_N&f!9#g;nxoB;I+Ozpvl|(Xmavgn!m@8*X!2f)#vrm^G2I^PrZNiaIZAk z_ox8pw|a9a!>x24u@uO3?Waap;;3Y-D<3pHl8Fn`B+Gjji1XH0F2K&I_vpsW@_b&CJU86_g3_|3 zwB%bTTsV%Y)$lElRgmHTCM@QICiQXc)nnnGjVj#nWNAy<4uPOKkPWG$Y<6Qi-jt?< zy?6G|+Rzj_E=7W#aSGx?1c98)14(`deND$s{=k*IcIV>sZc!K0OfEv^1Ko7i42BbZ zAZlJ0o-03=YkKO670s_ukEaZoptg}XWi;Y5DdO-tMV>Dh*Fk^3`Nf~e&EvoBYU0I5 zhxpAu9@5fPr}?3_SNsJ*6z%$wNcTODrj_}d=zCR;4z9b%=XYlD^4%BsB?qqYZ(mQP z`Ifc#P?{3F2{^$K6IWQX%?c9Keso?~38Y4~22fX-0hW7!f|OyuP%zcm}_Ng!2m$ zv-tmvOLn(I(ly+(Ca`{@cGI{=$Pzyz&zj#rYnLjYD(fWN3!|8?fdwr;#vHvsGsz9dJQj@kir+Qnahe%Pd2`T zV@EbTr$N=ue0}L--X^e+Kl`bUzi!*io5zTjjhPtEH$A+|ckEun8>H*;ncsr=7v;-& z>EW|{qUKh9*8Ecb;qle{#Kw63W==mJxavOtd9h^KtEvUO_wt*3=5KHQgB8a~qh0ip zVj`dlb%-m81AD!Q(#IzgxQ^}5sEyey9-Y?bakC09-LZt<|6Ye5+O(1Hts3M#`fl(y ziy!i9TRGm5vyeV$oy~tt{zTiwhk51Mv-wAV_RvhV?X)TMIvlBtr$M^0{PWrv zzF>Pfuly^CUo5ejpOzKQKYiA~i)F^}0|`mI>ZKd}X7jE5o4+PhBe))m?n&Ycz_OIIljriB}lWqlEYxOf{sa_$$bFuTlU?;b~uU8nNDJM-wKL*wa(inXw`%ZR5q z1Gn2B;aN6@KRV$peRC?7hNw>B{cl_He>@y%>+3oE7UQ|}cxEEqwn34fDI3QpXbQb_ zL-%=^CR={ONj?6^@Md1Rv5hzH&!m#KZu86iB+At1*754HWqgWMGJo>sf4p1nA|BhE zmG6A% z$xqwv&5MTR^7acJ^TyM{_;sn%`GwY^{0yf-+Wgvz%vL@~?w3v?bHp5p?~e##suW9p z%7hZ{#6J-5$`9E1x8Sh2666FAz)~!n$la?Tw#Jvq*DC>Jw}l2-XZRVCZBk+Mw<2u$ zw*=Pzti>O@w!{2ip76&og*;ypNG3aq5VsdB`8ojbYjFHC$@djS#UhsdtJqa-2GkVt>> zCac@};d1?LnB0CD=Cmn-`Hv_v)O3M_tjQrUNsEYs#U-I1yN*Bbwj1JVoj?)_%SfaHT(sF)N>G=jF0yr#)STRonHNk@{qZ91ge zTIembjU}G9QV7@UK-MX#le{juO)Oo!Ozk0oEu zNRkd;h6_f|?DFF#9&rhTsdWMHG)Iw`{f&VRr;Eg& z-$lGUm5G7pJ>2BJf%qR-LsHvbL(@1JqWNeh>AMz0vUgB$+0h9XkB;JF!Brr;W(d-c zj{zmeDP-BrL*!=iFR0(Ok^ImPAZt37l2v2f$)(1LglWzqXZQLOMcZKF@mNJTLn?#z zbzi}ta2nW~6@tTLUpjeFE;n>9lOu8(w1@XJvXrmf#CoLPrLTEzg& z8f*qri%Hy`=c(9e$`n}Q=E;ut(IoF4xCrbSLOXtDLd8x?NV+hIq|sxG_WOZXf(8k-1T_Bdhn=!@9xp#jOZ zokPMOm=Wc*isVj^9-Xsh3%6?O6hYCBo80KI1lca$3dx4`!Wo)2IjNcir{tdCp*XB6;$4RbJ4AK8v4<0TJFh~0W+@lL2glofUJEEz< zKOGwGznz9nljpX}F+l3hG%Y}@M@!(Y`JOwr z^ep=H@e+6M@I^QgaTab>9s(^*0Q(VDx=cMA`c~C|%uPwsD>fd^-?XNF z=jL#E2K!jcrw!2Wr2-~gEhs);0K`Cv%2$XJdFzQpHqf5qysyU(lv5Pjuszq4{D0*#FXi#1LKLtvn0NFX(_XZ^Z519t%lRvhkqTZ(8l8#-*70 za|v;o0(aqmt|`hHnxD3T`7aT;E1a8s`2CpOX&eBlM_uUly{^>Z;RrLX)R?td9e}$k z_e1KoEnpXZ7an#@2d%>#9vs;U%9eewdD1()*5@sjGe`lHMPZ3?3*PiflUTGMZsr0z zTE96Ds`VppwvIUuE868q{E#9+P1&Hi;4+ThAVnhnHREjo8Cc_vC8xWy9HN^#u#9ZD z;Oq${oTV@HCJ(G9zENHzAaWB~u*ZumJm?RxGa>;Bf54j1_nz=ZX%YE1lCo^D1o*vZ5oPajV3t-|R1wRXtu>0y1u2NKjzP_c5 zqhhVG;)Z8OrT@Cz5b`LhgkUI@SU39M* zi4%zbE^#>N5Ct37=0kjs2)HlPAZ$xLJ|gM>sYV)PS7!xu32Tz!?Vpggq7M2x1EH{9 z85h3PfQ7={rY{;pFwM|faPgCaJe!x0%VQXhU5V!XD2j-@$Sw`>Ag+P)Q*yj+M) zg#JsP6^r3-$xG~eqz)?f)w3=tK3xC7Fer6b0Z|1tB2a6DeBmAJw$B3e9+$%4h%=nO z7!08k;vi;bqd>fS73Z?{Ew{*3lJhi`!utDixHnUyAT#A4e3eloM=~D5$vbs0iZ;W& z>o&w>l>_-6ElX}D%8|U&Kk;|ry?$|TJPgZy!gJ1FfwUi@g0(^TrZ&Fh2%p=C={Pa1?$nvP`c(5%xjLvPM0r$U`OrUv$z$dccE=h(yhZ($823o1M2)9Z`Q!NS>-A!7b*P)MIH>@W7f z(!wA(=$-+wm-->GIt^?Lm%;uw&*9FjSV4H16)a~jan4 zCm{g#A3lq3es{zZb@c%0uL1kn&cII}#M4bw@ZuFt_*UR{L7~}P=!&|7->n-JC^Pxs z9}~;zDZLH9MS{b$A(c_6+}Ta-3cJK_s& z!W}ML!7Y%l#gFG98gnEaS2m8)9VV}-i0dKhb=ejcn?%63#QpeR^e`?8?qU1lbiniG zDq5Mjh>DOt&buiR4nGQ|F%5b2+fFk+V^kKzGbZ!lgLzaYl%uv&75Q~9E9h}ypJul# zh=$=rYW3tYy_;T)C!F%5hjdeD!P{y&W_1ewnyJVK`*+a)&V8gCgEY8Ng#)le=+QQK ztbn%`y~WZs#_+mJo1d@+^HEX%(T9J;_#-~zd}g)^KdwX){#g~^BH!zH+QxVI)q^T( z*I7qhZ7)&B&trLQ=SlSH!b&beHkaO;kU&o^C$LA%5CTn7aQL_v*!8OnpL(N@_85%k zPkd&0nYC_o#q3;aD))q{s`K=SQ6iQ)o&>$$+bB+01X>;Obj(*dI$~EY7J~#(Mdl{OU2^S>M++am=0V$h^tMmP-~J&r(Mv& zW0L{0_HtCm7V|C{_I#oJM-<;F4YQ|8K+Av9;HSF+hfG_9%^gZ`LShEqy(S$e?#`ok zA2@-@o@MM>w-~VgGZ$umETW&<6X5T4H)`*=651sqao~@1XgD+#GLnkuJ5f#E%s!0% zN{ptr`|CM7$Lp-<-WG1J#$vYmdJs&0w}Cp!JJDB1edtGU)ps{~8AK{{7K5oSHn zf|EukXv*0Z>bmAFT~R5=M=w_bg#a7g@uLI(ojJw*b{#GK+$_uA{wl#2?@yz9KHFo1 zj0n~sLK!?ZnZVJLarhW(50}c-@YMi@#@|}TiJmcH6OW4VIeW#ZG7QqmafWz^r)Y-1w5h|9u#>PSqk?{XuZpi-F2d&yy3m6A61<76DnDR;u4rZLv)^T9@FHAK*(i0cj7F}a(@V8KE%OM^_e8)pegackxV*f z2a|v$M#RrN95&y51k19YLC$_-Qp(rCS%pR5bzPNciA*BDN6cZ!rWZcfD1%#`j8Kb? zz$=_tZsyE~RAl9R@Sg7od*|(j+8@TwpNF8t$sAL#|H@ zMBs;|_BZ133*{Pa<-ssPs=x`ZyS`?BPBSGtLlZ<9p;mUHY-wcXBD~dSrZSPSVaaJ{mFR?J)&%EfTyV0ahJNz zW9U|ac;hGn z5xJN2c-$^JWr9%S{4$@bs5Ky0mSthtFVA7dFCC)uXaV_+97&kv7kKLuN(2(oWOH!@ zq$#X~n)W~_s@x76?aqU==^^q*{VWNn+Cn_si&?Kz2#jKs$%JWh$ca-k$)V}(VE;vl z{N*LdoIW`+?JY~*AKpsbb*!meKr8INm<-pJD3R^`Yf0NBBPgi|fD4D_!#yuWvgGk9 z$bXm)LGxC_iee`sbV0Ks5tvM$E7apR27>Nlf_K-n$&n~k66q?;lBipYzj|(foVrHt(#ipD7Db}9 zN}lr-S0ZcQDG;aHGT8Rb3l}7GVf7jFAW~6{n3Va@+W07V8|@EIi-N#r`ccf|9ki-a^^Sm}G$s66c)Xvk7kPSW3F7Ga>vRs^Kv~+H%sTrR?VT~w_Tt+guV06(g`$l8StN1 zgmJQK_S61l=ji*ZUj&c;gma!JZ*h+bx54w2LWpem590u3PJOd7|IxCRddF(;g(tq< zq{p_fes0QqPvdtQ^RkKtyI-QWUm+Sidj~Zu?!xT@R&=(m6@SHi9B-#0!`oL|@}HI! z((qaY#q0mz_Y#gceBo-WGb04s+KJ+sKceyb{V`BBAW0M~CeVsW0xoQl@Qpj8lbfJU zIj9l0AhsJ|VB=KkeMgEv{uRMMeH^#)PzL7Ja|Gdb!o9Wo@zlUgmX{nimw#Sr z!uu?kAz1Mue3$rIM1_B$A z0cH!eo!eQGZ5I^rS-YLu{RkIZCwv2RhfkQX3=x!}) zZxY3dJ~_>~-am&Ys4LQ#f{XMTmVh(&o$>wVRlHk&GnJFn;Ds5G{PlhbJ^;z_SL?j^ z^EY((_Npw%+@Ooc|J{Qp#pbc6#f{)zK@ioM%dyR;BH(c5a{j~msl5Bk&zw(E2X)JR z!KEg@5X1#ydRb>GZ(P;Dg>j=?VXq`_8LQ1#UQpxbJ()vyT<)O7`+m?nN84DtZTiBu zB&CAmlHl#F2mNlVKq@T(ip8$rJ(KjQl6VMQF1mwv&Pf$~y{bs1O5OSJ`WjkibB}(m z^nYxAsqSQv5mDfrFT5PW$J4xq|DIY#g$akMv~uEX$96rOxD~nSAbQ|4mcRQD>zm)hiP@dlwk?dFysb#n)i*$z!x-+Bi~(&gn+M6E zYoYk69bKL=jISx*;Ia-`K{h`hTkiCr5(h6(xjP}W!r}Z3r@ zU>tnV=D6vbgK45i300qSnEu=Po{KOOq4n)G_?3|j)Y#d9aeZHD&eu-1;lTs^(Y&0i zpMQ=!GZ4r9V?WXAm2>&6p&In&%qUJWJc-J*XW{>b)!_8Cskni33$NWTbmR4U>YEus zCwwpFI(IulSM44clNS$>`~2V`cbs#bq73_8Z0PiYNi<~VKhA2JGCS482tRx`D3JH< z;hZiv3WhuX#c@T`h477`t0~w>@4~aH=g&UiR;{{pW|uTAb5wjU)#fZugayH zWIgB$ClUUZfi2ybB0{rYsZ$63FqKgsqPJ=gr@p!vHohCjeX@GNP5fp;|HW;_-xe<8 z*oiINoNwCr>eY!{jExGNuVMg;S})=MJ{-fs>cM>*Bgd|Y@)Uf_p3iMI9pp48eP&Br zoH#5Uz+gWHTl!xW?D3I;YfbHVRqrx-uQP@&D)8m<$IDR9_33n(N;{n&mQ9z4lv4G1 z$7ufKN;)R2kvfU$(n})|)MMFG`d_qwx;z}D0cRI+Z|qXJkn0!lPn~9*zdVhuooB!& z(FB_3cZvpfeW%57Rn+vt7rHy|DqR$%LW{@C^U;|a{8=Lt>UKR8M_<3q?&=T5hB=B> zqCc|n-_dE@@ru*5KjDgy+v{+H>l4^nVtC?D5uN-xbebmn>o8@R{N zN%MWqb0gX{H?JFnArFHd?ntW*yZ&G~^Coyaj~j2}SAvW5T)&CX-|j$7_{X#=+?>vm z*+`kROv-GSPJfyI;>5U1obfdl4=Lv2jNO&&&yp%OtN1gnnD-duj7mTz*#ZWP4q~Tp zA2|Li0be_0M`I7Kq2kuBsaEo1`eBWLE)8~pian87EO|OaH)z49%ZI4JM*}*gQxBpQ zjRk&JsyV}sLF_ED8dlcCamDOQ{O8*;!L?cbc(Xm_JSBG0!i;gkUHvOq$w3~f?Oj>o zW)30v1$>G2;^YRu;)K@=xb{{<`cn5c_upm~r>)P2`h-l*oOglQbBgJ%cYgHF%TL_* zlaar*W3@X==zVXxMKvqm46cH#zP zJtt({l(X;~tVp?YHvV09l$r%hz@v#-^x%%|^ig;vv|Sm-L`#>I5T34A=-X|diZs(olDh~1isH+bHG#=kEhJAD|J1e~SYy^^@U3Yzqh=vlht zeiGgnXOE4yID=TvBz|_qYPi>IN(921Mih1ngI+?(c}1aybagUu>nwwNjSskgN|VXq zJ`N6ir?7L5P}jWi59Xvq1JqAIF=a2=^?J|ni{Sv+aRbQa;NS3!uw-8RIC8>E29~+K z#Mh+?01M8+Um}2Ezew!AZ?oXk%j_FYVyY0a>loGo~i^%xsOi*fM z$)@-@ifXg)U()adUzhS@A(1TsF*rU8^IZ zbapFQ2N%Eh$9WAU^gBw$-X+>-U)w(R;hR$qcs|{u0*DgaE z75W@Xk3EDx>kQ~|kq6ug;mq-zR~d+_-+{9yXA}F4`obPjhoqn01YL8d(frnGy!v$z zXVjSlGJ1+6KxsX+(4*kn-iTu#2ZOVpJKpZL0HU^Rp&iX@p)dX))(-l_4XCey)!(Y1 zdSxM=lTif&&t)Oj(gudx^KjjpYl6wWUuc8oTl#3?7;e`4KDvCD4IJ!?gktGPxYi*J zdDB1R?)yD(#B3IjQ-iQesGIl4sR{L*Y*IHR4*V33a_v7Sg5?oCBHQ_idsVLuDs?_xIBA7l0%dcYJYUSy24PBZt!6PW{c#_YO{#w=5?wi0PwmLlcpH_!#Eg zk3Oq6EqAPD3^iF<94JSBs%ns`kh^^6;TcqPy8spCS0hn1A+P9II{Kq$j$~y-(cE!T z=!THB75EXcf3)pcJqZUk%4I5RJ)@oR9=*;SdwYe68F=EkvU5e(TEOhl>BlCLYBWC&74(4mW z7Fs;^oR!h~UsiE`j?BR>%+#4^F@c}57(dwx=E}Qf#$;>}vP_CW>^*NJ)@6abUcN_t zTk23GcO9)U5ptzA%P}gkZcM}LAjV_Jleu($D^s&*FQXo`p3%BEfhl+~g6<#hMB4Xi zkZb)SR9)JIg64^{TW83#QqeC^ukB`J@jx5BeaN8P`^qT)!AB(3*MdT9ijZ;iCghOA zAct5V)V5qdlgK;WmXapt+mr-8Y2Jt{cZ5{He>P4Vbf*DSMco8}b>6zA8rhUkDR6 z^S70ogf5ekI-XfwJD2f)5`c~$t3)XyFOl2eGqmE8A(}B;7EOQN##k&l#netSWj#*J zWQToq*`Xa{*p2@!MXe!5Xv+D?sP)hobUR#$EwNBzyDH>atp^?Gg;gjcGm^<%TzG-` zulWhHJtz>(Ul@yy2Rou!mz+?sVhtnmoif}bVFtSSIp*ctSY*hAp@7+$=!JC-TBNFn zrs|oXZ1LIXw;7-l9;s+|YZ7Yt8;{I0yiiGb74xMxg*pE-h2e)!F*7YH7zZJ%!C>DM z#-Q|}RhEV?vsN^V(QEt3eEw^QJVFc5FcW~(FRwyt-cCce4tF5CVj1?(3UPLksXV*< z+BDWyb1ExusmtbD%Ci$>C$RCh8m!&9acr})xG+cMB{Qwx3jNe{MIHgpNM~CTGo>ek zIXp9$xxhv;ADgsLs)`lLoPH3!FItEk-y5?hjc2fZ88>a#h7LLY2GJGbvd&tDp%ukI@7io`the^YTVNoHSB`-Fd}d%P&CH~R~kck2v_nV5l0*5{&% zWx?o%`4qJH$5FIxO&l7^&O^%z64AFWen@=2kYSX(9If&?fbQk4MMpH<(b-Z<>YGvmt;Qz<;BLqny^{*EM+Jk=j<;C)f$^BJi6l@j}^O_l95 z8_V{oenClXC(x5^bCKWvImq324ElU-6eS10L3Zb=P)+wKbklhSW1W}H{8bS!%KeX+ zeIf@Coq7Z%{5go4o_eB!nsIETg(5rTtiX=Z`GNHPCLw;(I&``iqlV&*sOC;Ql3$pD z-2O{JJJchPXuc!b*qV*j2stFit@Y^d@?$8d8$q;fMc;6r68mez{8;qb>e|;z ztIi+&R%_3nL3^XpQ0hh@mr%J1b&5%`b;9#nsG-H;^Ga-TjWm+KsDrZ2mLO)e56W_% z!pypD$#f@cG8r$7n9C2U(0;q?XyN!Av}}GH+LN8l)M;8WTYbB&9$(6^vS==2G{p}v zIT7K^`W3y*8~;XTgVK3sLsBU7hRI;G1Q$?uNh!MQTYyfWYe=f14Q;;Lf{Jd{BTTL! znH)8?UtfcD8dYQe-j!xc%O0UGTb?6_!%tCfZ3%MMXtmN`x`Byk+{ui1<}=^)lhB>A zEOg~?Dw=*X0F|ZPMW^f=(bj!W(7PK&$dqU1>9j#=&jEGbU5?^@ zFG9sZ6uB&FKt7$dDEny@lAfx8E{9G+b4LILc^*P{XGNiD_xzD*-wHH8_#1P!Xn6R_#@1kCYj*F^;8P;2 zujE;u1&JsiH3(Jd1)|B{Lr__)70Pl=1t=Gd)|zpm~B(^u}Mv zF;);`|JF;gZ_alkEB8V4`Pg4%?E4XUNJyjFq}eD%nCq{hI}OF#?_$JmB{QyZ#Z2Bk zQ8Z=G3-st`0rIRXM5l}s(Iu5A6m>ZieM|~RJ~JYbiNKqgUggLBAba80Y=uROcsngV-1L5?lnbO)6XP*2)9lswA^9ZNif z>@sGdrge?Ntno+a-MMn~)Gi#^wkR@1-z^xmsOgMYZZuOf@f=z+zX+Lbd4vppT|g36 zlB_|m0vk1;%5Ew7ftG!YKrQdnkovP&u#Rb` zqf^N_Xh`ogy8rMBa{#KD^nJO^p%qok?IX#;H}NLsj6@uhaIBqaI6nhj5wf%b3&$a6 z;d^(dY$BR0WNZF-aS@G{mY}EW7vJv?@@(hzfC}K-3w9FLqI*+J}9Ks2d%dcN1`1UkiEzrBq`)lv==#|Z7vH@ zU#kcjzrYmb-?2i|m7LJpC>z8WMKbd3C5*tmh-u!J&xq9qp_2OdS@)vaqFxld=q6ICT7;OFtC98O z)#yNg5xTbL3;Gr#!^*Mp?9iVPH2!`H8rfEel>W;?I+jVOXFkx@ixmzMTcY157n6{%!Ip~y<4kS(L(z26TVeW>H;={fG_8t1r9VMxS$BDc9O zkrX(Brmzv*(C>rp;!XIq&>03X?l4+06%t#csAg9enLHaH!_H}_82c4T_kQAl)i1c` zenGv;01icr;;P6m1bg$bx030Uen^zdT$IHFH!hzre>b%$m(jZ&H>vY#2Wf`#v#VAe z;O}Zddub*1&NztI0ml(p>VomjPEdbpNrCHDlFe{1O?L94gvD#2&32-r&K$kfQ*r3= z3yj^alR{aqs*`nnYdSJ^^+lR3pdIYu8Y`%&ihDiYsUk6GsJXqa^! z3Pm|MxYq#l|F|M%Xe;))uH^QWi|9XIgVeqNFfki74>GVz>l{W-hmgB! zB<*hqC4a3L$`@Hm+nt8hk+F-=D!pV*p2He}~%8PbmL%!oWX&OnwlJD5p3~ z3qFU6aa?^qk&mXTWXvrJg53NA_;ltV8b&r^yO=i~Z3cjOPb;8b`NdUN04?4n^*B>aK> zhEB}S&Vc2@3@-Dt5S6a}`0xH_a@{)y+X8uzTRaDU4Fl+PMI6Q74^mv%Wi(MJBQ zqpI*|D zr42O4xsB#`H<9hqbC~}10hS9rfcTLsSY0kl?&Z3)=$it0uUJY~8**`=rUHujSzLx$ z0Qg2_>EMZ}qY8y*75!WcU zhMVJrGU?dI&s5Z}iXAzc;B?OzS$7al)~B%`D+rPA<8au)A8%KNbMJcuUfl|ULAN#S zEjvZRl36t1m`8#0dwF5okCW4NKZ>v|qJ^$9&@LRJ&l=z8YHSU$_^GHekVNDbA(RLG z`v0uKThVZvmr!+;XPQyR+v}}H=RCEjc)UDy#mLd>xwdfeu!7O+HMlJNhD^Dhy>-!&hww_cx4bUh@^ zE+t4Dn~m04y13DhOqF+iNpaReIxy3mcJXJz;#W47aZiE3pf9AA8tC%X82YqiFO9Wd zN6Dqf5v>w|>)Fv*Zm<`1*S_rC--Mll|(Cb)x9OVI63T}$cM3q z`sp^Wh7|YMqV&8kRLp%Kwb%#UahdqfAstHfDG0dXjj{vfI9PHGAJ1Qg$FCF&t`)Wpm6RC9F)##J-elbgQQ$%sCb+ zqS+ANehC{iE@5}hJ-BZ;fb3pB=zfpF#ut9rtW`{pa|B_d`j>bqf*9EI1Cz`}nNR(A zw*R9nQ&stZjWOqtmlgygyY*Oc-yTZ7>tW&Hh&Y!ZC@;`v*`c$T&7@f@xl@_#Q(3@P zOtN6whs@a|Wla_u6@YL3so0^Oh1Lb>=;!98IXVmIWG%N-E#FOL;%hMc;V5=a55wD4 zr#Ku!01DB)q~$U~t-=pU%+U%LqURuWq7_CX$6)(6e&%5_3R^#JzPQr?`Hyqi68DuX zYpp%Ae7BelW__TeNqywGyMywEZ&95DnA6HR?7NW;Q_hlR{mxNjf6tya#O|lMrzc4D zYzr;FxQqBzP3YvzE<GI|}juhoF-df7Ps<2CPK={#Qg`$*pU z*fTsmdoB~~d=k9;JkZlj941$U`ESFZ9_fqPDt;~-kJ!3v7Oa-z@4ESQ*vj1rIBR+Y z1CQPDUfKW$YsRw|=fs%Z`1`oOI3E07$02GOjRg-AaeO!(6Gd+!gEJFpkAH$Mx+VDH zKb6^bDX=p^vh0vn4+?ejP;jacp=*+{ha@O`V4V^i&? zFjYfM_R1_Bi(Xtq`e+N1-SZ%(n?z$?a~WvX6;y9}lQwSap>~1)@M296R-f}lv(`9r za#N!6c6Dk#!=vH8i})^8gox5KNWXQ%i1}G4&b)~G={2}@wFndD-$(4bW~?*iUdP&0 z%wJFd^&bV0t}2J$nrxIWoQ|-;PvrBtmUi3SpusI4;8XAi-TOt@QKeyI+8VLNF%~SW z+mg-Qw2Boe>_UE77?uvjVy;y>a^|O!!RJ`A-VjG$pRbC-9?(lJ{@qS1nEgh&FZA3s_EvCs)6U=_F0>*XMWz+?2^|!_WMIW%GQry*?$9=(cFrugGt0C zR>QbM1CAw&pfU6kS4BB|YZ@PGKK>XRc0a)r$ye~`c!~|`jVNg4vcaqN;bylhuCB31 zzrinB!fgq2PKn~J)ezZ4>Y?P#Lg@4vp*O@7`T3V=&c8=otsEx1Rf<^ay%cxSXXE2l zbySspqYDc>$o}mvdabpQb~^5)n<6pjNJ>Her3|#5N{9Q=^YET}0dc3#6JwmCI zj#5qAc3ezdhd05i5xYzU^LuyU(~JY?@jnFHYs>H}^fX#EOR*;QD()@MhS5`5_}`Sm z%-C^QV=jkoV?q2__lq1yJ1EZAinL^Uc)yjU*FaQcpMs6Y9g`eyTE zhNC*vJ5^wDOan(pCPKlqfGlH1D9uh1vJVPLqp_0buDO)l!Y$~mtv8KnO+dy#DpXq2 zp}sF0%}$<3GYCVgYzSVh*n#dA?)_Ijh=ME+xR)=$@GW)vyv&3K(st84{Y*-50n7^j z(8BtEBpC9H>YXNF-vBA*5^nl2}MKueE_Ch&_-m@Z{jn5OtFe5=}=ZM4)g}WVdild zOU=e^^&h;{XhpJHEl+7tE2&#*7Y^U>M~6%}%u|}0QTA9e%ITRUVCgE2C230TM1hN;2*EO3=z`FiTc+_PJOOuuOpy5r4EoJ0zvWO1$^wHQPNu=DK2{~J1 zyc(K@S*vD}*5BRa(&A1#A1tQE?|kgKiUJ#2K9yZi9LHuwJ)l3T@98vm0o}?>7*o1A zJi_$~hTmPmlGRsX>sAN9Bi-ER@jE1veq%v!0ZjcOar*v2{CH)6)N(KQ@7RE+0$b1= z;)E#&X0W!8v)Q#q9rn&#k)4#8!=y6InAPHW>`%THyVM_oBR!`v>>r9{x8rePf(V?A z{?LbA&&V!S8A)rtl4#2Sz5DT(+9gDx8k0@C*OXDriL>PO^$eX0FvO9s`WSh{)u^lA z=}drRhBeWuGiwk~3R0Sj4a1COb1^FyyjB<{8OV3B<|Dko{MvyuIi+C!3diTW+G8)GJbRUw7VU-UX(4v%9?~2uekx* zhhCw#;~jo3>%o7kT5<9~6xP)(L&*IV;B6g;Ckg!Q_}~w`@oPhhe-Z|V{?gemayV+L zfS7n~ERc-DoR)YvaG1SIb~wTcr?J4Rn(XNnEoNUhfej4ogU$~hn6|p0Iny5xPfoz~ ze;;YQ@<+01uce}hOYk)t!<2JHnP9{OcBHcso4(w{iZu=R*O82$d&5Za(P=vA9YQyx z)9KEZKO`Hy3R?{KB3hTL^@Sch;ovx4WO61iamoa;y7>n6aoyOVbO&9>(@}ZO8=9_3 z5OEI2sXAv!rwt%#kq~RyIEaNd9e8b=ip;1~)J@8Ur~X;Yj@d$ac}jF_{1=|(fF#+L zp2F{w@vx3dN8g)7yqYf0GPcfS`r)(L^*S|X#p@vXzuh$QM;Mh!U*|eL8mjHSaDb9oK6JrsJPu=zW$m1uXI*(h9;j zr4U$Z218;v6yjXHUUxbZu`lEB!|oy$KC(enl>&B;*Fu(8KZTxUOIDkV3U9i*44+&XL zIQw3OwL~(uDT&ywA8M>{!%dPuQb6>%kgjk4O0~DNS^FeI=598fDeqNaWfI=V`w@$` z-?CA8H~~Yw!SwjJJFOD=%Uj3cL)l7NtSm~O^*5`r(4DeuldB&VeaeExsuUO*2VqS_ zK92AnBRu3KhUb36*Hss>GwdqLZ{~2bTQr=u#bH#j7|N07;Ps%I>+Ke_Dkz%M4V05u z>J7TUFOTMPW+>-vgwvbj_$_t^OLkSF{98I+XnEnJ*&-AkQ-oK#GzwPSr!8)$DCTbm zE!r%AUkf;Hs$2vm4uposQdr|N8H#2W5c#kiizoB2{6#FS(Z2$@E8h@lIKc7f#khJ+ zA5TI@$!5I(YSKTGj#3Y-gF9hiRgCek7DA1~{B9Wwq3)6-TtX(HZmTkD+pNcSKAX+? zi>5N!BAq`%R8*IYUjw}O6KOR=@?%Isc@JQF!O2;*^K5EO_(?TQH8`)@N0eLiAK{y!*2 zh_drPAEVPU8KqOB@M3Km;zYR`b|#6gEl41{d9HN#V+ExOS<=#aA#xDYqpc}Rso<#~ zvsf(5#=VqhXCf!Ek_j3(5xNYoE*?Vmt6*gC_oIj7fs)nl!*oL~JZ39HkiV05Uw%cI zTe$q~F)idK8AK`*{_+kV5~VOrXPVaGNE0<9Xp?Uqr4`i^MX#re(dE45<y-2al(nU0jN6KVblWontdkrH*c zQYrP*d66Mn=fGk6$r3OrEJe|+%h)^c3L9E)qW7v2n<1vjw#2A0z3;Lt`MDS?(vx92 z$x`f@oE#h8q{cip&u7ZlX0sdLCoK}7^@%AY2&5y^m=<_hs=D3Ofg~<@i$clO5PL>lSV@-~yB*@n?6TG}zxYJx`c$Fn)k6eru7&aRbIACU zj#Ag1h?vf0>bSYF(cEL~-e-4K5VDQkSZ#+xo&K2kkjoAKdJ>&W>+$gK80Jtnk$tL| z%Jvz4gruh^qa$zOUv(P-@sgN#P7*IZs^bvHA^%Ot<=q&sLwb|ulaXr>oem5j?Wb4h zf%p?D7LiAmTN)-8S|jziI%X_=M+;k}*tAa^pLqW*`fM}tdFnXQ%icrUg{Nt$cM=7U zOheaKj?0%)Lvr3YBtIL5i**+^7}r9~aUG-zC$bN1#4>i8vc@isvpX{xlST$8BTWS3 zEQaZuof0$jkYXZg&yjp;9~y2?Kz#mQe2DbM`vGSZ=R33A4-YZv@dw$d_${nkpaT(3 z-%;i!&d!QFgHg#W7%!fTr4MZ2HtQV1pakV8S;QGjVvT_~P7ez*GkJa1`^AK59dKar zON>b}_5|H-527+}4)e_@p*!*KspF{-EG>SJI;VGV_*9GF+a37!`URRy+z>ET2gXV5 zv@vljH?3PU4?kxnc-@x$`N^0-Z58T{_A9>_>}T8e$dG4HWV&;g6x&waA}VK zJO4!xiC?6#ZRb=(Zk>R|g@Ua1ya-eM_YD@an~+sC79Gu(X>;Bix{;Ah()R^%Z1ghm zT{NImr>#gwB%2z4m67^>L9CZv2*dUU+_6l@VLwmIZ`g!nYiY(f%H`(=Zf2fp$o$4E zWAm|$)%BS${~7A+!E9Yte`+?1Q_yCP>Z0sR$y%y>oJG7@wZ!-L4fWnAp~e}tWb(L? zr2G0Q$${Zv$6AP9NJQ2^G$Lj>G5z0rnaKEEEauO0#^)IeyOIJ(nHQimv>20I?o*2G zPr8{UfSE&r2sK7rp7X@YSOldp>tT{a-H}^4jp~dJXP2yhN4IXV}eo zkGM-8q5aaG+Q!IGTwfp0@tQ0h2yLc?{8Fgzn}mT}bv#e^h7hNFTkRByZzp|lm?ws= z&Ar4d_>jIjmWqN>up%)7RsreA+Hn#WkMtqzv=BRdTYw3kd4@A9tw}C!8)=sLQf5I4 zZ5!^v!t9UOdh{{c7gb}cqa|Hb*-H1cB56ia7zuX&<;}QI!#lV>nP<1}K5wy75F$AI zzGqE1{5YM$JY6?P4g2D9#a`&d>p;fy63rc%0O5(_P%6#sgg#k#YJ3g~|FW@xzZi=> zIn9v46qj~`do_z@W0@q;Zus#BYmtw}Nl^E;00l9tVpm!xV?Y+OPa)pR z2X_BdwNKCleqfV}+2Ui3sVL220`R)R9vFq1h?W2tSF4CLI z2=B$L@#~ESg38CjGeQti<~@`r9Y`oVx7F}@jW~@86k4tT?YB^t2o$Hj71-O zabUJAp6oEe#z`iSO_f6or^A$eUxMKJ3}h6A;Wg_#&^M9XOh z{M+UL@247&jFiQ`BU8|3B@46d(vV_iSoqf+>K}H)`HuIRtx@yyD(KLfX%}F zIMuZWr+P#2FUJ{s0=}UnL7eU0GLDHkh_L5>@}P3(2DXm7io@s9u=&p-`qy)q-s`(l zWqv9dg~p-ArwIJF?xQTZACE0hK__Yfv;ul*&v#E!E3c;3kaOhHP)qebujy{!YY3*j z!06tOnE15;&D+J9VD)(RaBvKhz5NZoKhNPq?qxhHy^D8`uVW8?0up17V%v5PSfBKV z>J$UGy`7B@C924io&$%QJL&c^N4o2`kKEm|sbFa-^wyomNJSa~kM6*G>wdcV=M{|^ zeMP)$)s)$A6b?>>crddRSL;%de&i+9|7)haoCNap%c04tH(`=ckNZ+*@o}OpVw#WR zj@*7&ggaxva|yn#SH=bLTXgJS9i?Cgpzp=&o7aCu297gCG zEI6`+)2OUOabX9UT~~$8oeg-tXD#k)i?N&MKjUM^PkcVvjCms-IH|SDn3Yaz828ig4ZZhc!GapE<-9K9QUFEVez>HbGj=bbm z+C!X<+4w9nTio>I!r*bEtC|*PTIUSnhj5^5^sE)x$X2wevj9x>>|=j7g9y zi$SK!9>na}NR5g^yz=okc>9`F=^#^vv-)`Gcn?!+WH;To>VtQ!0Z6zL4#EGz(K5Y| zSK7}{mwO-b=J!vhoQpxQX!pjH4*@Xsu!ZRYId)`7kyW0NXCF3GBl1b*hF3FL5}F+2e|=7v7k! z$J-e@py|B^YP!?mX)J?CfmE!}N`&U09jG?yqi+{|Al9%SN{L7Db-ObLIetLt@O3(L z%ZK)UsifM>Aq;v8vsu%Hm_pMaDpt=$wvrE24{ZW3-w4x`6%h2#7jAM#@#Yt&G4=X~ zTqOxM(LUMS$uVM%-ukww(FetOa!Ll?!5QAPZGIwzDxWg|`W zw(}!y3W{$`60~3(^WiIYC z_fr^O61m^;p|o8l)NHm5yO;0fxU7{p$o59jnHfM;E(?gv${1PdVCVpHzwh!Bc~7Y-;Q(E^KoECD=A*wPydbINHd*+ z=(Sim9hOR?6ZTo8{p&ETyxc;6bk38>luC+Qe3K?bo#x#)f5Ed!{KNA#QXu{MJksa1 zNgFMr$l;DN$vAz*!`+`zx8eu3OyjuC){`jUu*1WDe33tCBhFWbk;n6Ibm_%p44B*C zf?NU2JzDXn?hZ<`tI_oS0zM=)L3*?v+e308pc0D`!BTv`ejdF~D#6OH!mXzhyLk`M zV{;L`P-w=$eoYHUDL(!u$oC6=ofl*ubo&t=Q4Z2fNIvGTal{ra2h6x< zgRJhcAopw}HkZKOydI-*Wq94r$5L0zuy#o~HdG?OmV135JA)n~({prw<#np6_=EG` zL|914Km3>6j@A1Epc&>5Z>xB0xaf;_7XxUUiyC>E^znYF*pkJ40am_pJUcgypGmFg zMNvaBatkW4c5N0Ob@(8r#T>Wa1VG9s6xMC0@b-`lMx{*=F@6TZE9b+0TLCOos<4uB z(M0EwUld2p;sc~=pn%D{ba3WeES-pGpw8{XG{#*VikfF2+#SgAD<`mHj~o0v7hprT z92OLe(reNCq>~@Rt1dpqI}vo9_h?5FZ>%IA&Z;V4&MN_gynRfS_e?SWpBB8r6!2uM zD^19+#=rTE$Ub}nH`a5UUGgaPH*-31`$tsbl}DjAcX9mpRm62kA@)fDHe|+PNyr`!vkpP|*=%gr z=H`CcR1D|`W5et)G;+P~ST_$Imio|CHGuJ<#ds{f97%KL!KGORts&!aZD=y)sBc53 zxC?ylpTyE%?%?)9{L%r=X&;_hPl7F0w-xJDi~I!RO_9K`ilFU3%#?e@gs zB2PHE7~)s&Oe7f!qRKj*9&p|`hZ=sWt`(=%El#90e-U|4x27eJs%dVV53)UWA;NMw z1bt0Fe7^8nk%jFXPto%%ADv6mpsyW|FMM9eI}c7D%*Qgu3bLw4BdG7Vg=bnm)Ks^H z_HElmrCK&5*Q<)fRW2Cjc--=9+0YWbhjGOpkk&SU{Gbk0S#{yT>bIEr?F0C{Z{e!% z2JkwLVBOZE_&L@W3k(IYr%@9}&d5RI^9!!_){u&P9Vx87PPYz!p+8Q8++E07Fy(r_ zN}4sQ{xp%_Krns&#-ps%Y4oTig1VG*DXZ9v@>S1Mc}5pT-@gXGOb;&4cntrgT&;Q$ zij5-MAu$7xIVj046w0x}CzDuM{&|OFpdowkeSKPUg^% zV63n6qwp@mWkhRH-l7AMw>!@qV% z1jqS}FPNw4QqR)z(o%j_UE(?oizdLH7d>oDiZW%R8{!00$Jc2HBA z$uE*-FP;f7k61pux+a0|Kh-$z#WCoq*E+h9_z?dWH;l%xrzxT= z&(#f+=0@Vl?=*;?I|HW$(Rej)2lO;t;bLh6)#Z0c^LP~14hB+~h9j9R`AtXfXVXv4 zOZH38hNc!gq!>=qv|r)_b>_UFv?M)JQv1Z)_br-7YE8U(iQ(9&dKN_-KiRHz0a14K zDCGL~zt6eIc8}8H$N{Jl2XNq8K#tMpFr+_8FV_@AP0pZ{BV9~lT(Ceb9jbCYAz-Lly%*StT7xwYX zf$NwgXp6t6qI?r1SX;yTrxf}c8Hvp8qQTUBYF+b~7Am?Tn4QFnd|ylqh{5`mA-MK2 z0dk9yp_y|UmsY){XNu|&*`f!JO=Hl$ES*c$bqMJPR^sxoHBSFf z!QzWQ$)l!^Oc%bSpK_lm`%w%TUf_;W8m^;i=uy{OqT5W9?yi!fyPbS=X}%vWhJzq zSdn6mh(t^HzTJqj(W5xy6#|9uSfpfZL(IWAsB+KiqI*RU8aRq0?`GigpgtTg>EXN9 zTAX`tfbSdUU{RAH_K!6|)u<*mDDK72l%1$C^uuW0Or%Jiz_+ZOP~>KnX2-2OkL9m< znRXLt=}ZlBz0^p{-YQ|;8VxuYir}g!Q><(G)XT(T2Gn;gfpXJE_}<+IWi4O&-Y|`VUntY1(?ax5xtC0~ z^pdeSKSr8GutNScU3u?C+MZ#w*(;FTt;f>wIUT&|n_76T$mkzuMPDXJRNCdV{C8E4-;>QBLC+wsZ0DOb*o8;d_79_wep--cMnDu`om6*>m4hOp<8Az zp6P{xkLx8DIQ+cA!vkxZ;y1?rmNpcU7q0=jR@}!Qo=Z4d_pCxoQv4Ez@ zR?zXw3Dn4UjK-cTp$gR(l)S3qCOsC4a71UYlOQtC+Q2Tc= zj`giX-(x+_6A^kM2#OHzA@_@2WxkVT!~Tb`;aOdSuJ>Z21nrMn?~&%miC% z7eH{QJ~tzMqN#slXjxPoX|xwpIj-T}jV9CtT|frsq0qb|il4r7aLQ;NCPne%Xoer9 zXI-YRx@mM_`ztaKl){@GI{5oyCTt5A9N)I#p9eCc-bMuU`UxSD!=*$G`AdIKbf80%+(b5&zKyY7#v~&Zpu?C-Nv|OHL)7 zHcQ&QUz;>VH0j#u+4Sqz1Tr|A!4s4|427&M5c{wL2D`+dQC^8$%?2oNe!I;}E}+S+ zfP%yQsN1rb1Z!-ln7hgFVbK}v6yWson>KS^^u-XX^}-cV4;Wn~N2orNs&CAUC+YXHRmdcq*jA5$8)Bl$@@ zVw@x4&fS+N5mEyGssHd_(lS(K@!%;^$osfEhIc{c2~SjNnD^2~i{36iM~}*H((%W) zDQYeseQ1}bgdgg3S5li=yI12CZzbxb?J;h{98PcCN39jlDC2Jlea(xeY4z3+8r_4_ zV>xa%Xf>J+yV8CWJJK!KM{UfRN}fnUiPI)N%YHgoVb0@Lf{aA|QenjdF0}I75$&a@d#NgiP5L*bFD&ar1k66!wLlUinJf6}#x2 ztsu6{=d`J$jqdS_5iENG!Q8x{U7ZQv0#1(~;Deyxcx1m%hPaC_mhp#Tc6mBpX=cLz zO%mMF7C_;a6~>INf#$q9aL>KRJ7!VA6Ha=}t2?1buLjg0=+4~}xHK2G57{Q~S*tA&Z)?)Y2h1KsT|7|7)I zOR*|!(!B$Ig*#Yv^a`FUIKkukI#^{aLi84EjFx0mJv&W>c2P9`XdyK`^{1A=^As@d z0uAqaN#?e`*vIFI?AOQeJ9I9Tk{3~kjyZ{SPNh{J)riEWA*;s%U*P~ndvk;vO5wyu z0VwwhVAkW;bRf(bYqj>FvtkjHI-b)}0V!Ph(m;n;8!3o%(raC5=v6$Xl2e`3_-vTU z{I1}S*Udn3|#SzP1I`n8A33lzHko8{F6zfH`az(IO z8-i8Gg7BWt6kdGOVdSQbm+jNBo+e|;JWZ-j%I1kZ8seGpiPQQm8Fb_AUDAuFr3SAI z`Z15Y-SONVzK^(D4WiOex8eM@MP*2dufeOfb4dSML5_YtwESQ+?K!VawtT?|o)!ch zV=rv~ppO?v1i+1Y7!;b1^vVSg;&cF^CSkC&3B)30W3+QxIs4Ld)W{{^M&>%)8(M&9 ztT0>lHTZS8p8IoJR@H_`M(w{J{5A#)%l9?ga^i0Hm z-2UDY`iA&7nZRtVDH<&eu~A0|2Ud@PYrhiiUQ*9&*ARJOI>(PlLx(4ag?pyq z^etTu<5k9t2QpB+#*b|uFHrofQzTj(L~|T=)91D6v?6B{wI8#i1mS&DDRiF-QuCWmW>}M?XFW-@#*_E%1bXo;m`?7j z!LyiaXbHQ3knic({(J=*3*4Ywv<<3{79sY8Ag)$1NX|19T!4ex@0T_%;7akJ~I(=^-e7#&KF#j<;a zFrJx2F*m`JUR&?KBc-yC;j!HCnLMmZy3a;mR?+U8A9InFVM~T42y01QE_xLjQakA z{+#Wc7r2=2$~90%-Ca@%?;~OFd|I{m7G2`LmW@0j`LRJzbMVE5t3gmH-hk`BJ4+XE!+eFx7Ak2noc4i;B({F~1f zM2_)6V4M>ShW*j>XDPO|@5Jf%%h2)mA?7+f08hFYW<_qepS+Y3CH<*)D1bh6UZah- z++aP`8GAhEVI$Y`evike=L<+> zc`nW54U&P=14`@`p<#b9S~4IkYn!Xru& z%FkNhCdm1cv|nRD=_0=Lo`oHM4vvcyK>m6jBF&|#La~Xbnx4W7kbcE`uIxjdZlN@F zTO}2O^RxJUfoq~LyE9ReJy20 z3u+57__Gi*)uNHAaD;SBMd>@gAg%nOMZU&~ ze{B(rBV(Xge~e~d&81QGEZX3Gp2E4CC-xc-z~9t?r3X1Y|EDrKW=ud~kuh3NPD8o2 z4~Cb-W1L(Pu1$_dSk6R<#67s zT7Iy*?+@vm^%&r8SKRpVk4$SW(}xu9=23A4Ib`0T_&ewEdqx?qaC7IAO(hs--;9%5 zcX3mw7FV6}xEWgn)xn%*Em0LU8OA8RqXnnC)1j*)g~d54kZ$m#leC?TgO1UGL;2jz zfhlC~WlsHOi|Ag-Hk#7phX)lWA#v{jbe+tx`Gy5b#Ma@6uoY5gXyHH)cZ+MV3HN`t z;pp-T)JA5~>2dkFr zBAV;5dzV}T6TE~wQCFa&x)&O8d$4Eqb|`i_BCyPyJMJBWDA_1%_qF3NfOL{PFMzvo zvfyISFdX8Enab;s!@n6zoR{LbS`IyVe3mX|og+TG5*nU2nY8o*c;aRuyr5_sUPk3Q znv<}Pat{ZSqhdI1czy=SGTCsAjDzjyVn|!o;6-3N6cyVLxau-?bcjM@_A?T(Xs5xz zH1ZnFhZQS=Z!D)jZ*W1pHFwKvQ~?G%rz0>_6(0Liap=!wSVmq(u0$#tEbpQ+`x!or zea7hsFQ6y!HW}UNrxR7e@OU>KJ@<_;R?i&1JLe&DmI1ux+d$w;AmX#4P_Zi!Qis)` z+XxJNpM{XMx`-O{fmg4lL|qf@X{nevdF4gY+(&P@{aO-@!d;Z7>&IJZ)yWG!|Bk0= zV@S!3Ek!+Ux+ ziC1$ZlxOYT#;cnok85N4NHVFG!c>dNZTmSSEH*=sg*e$be&^ZD zzsP$Yq(+uC8)<2`KTUh;Lw>L0X;$}m5Vs?iI7*=AmIiER4e;(u0k6>voc7@5WlFxZ3mkVo<$4WV~6(|4pZ zM%B&6UdMZ!K6rsOtdqv94lA)q+Gry$$=VcRBah?-TC%_cCOU;c;zAVEAO!{^&Q{+z z_yP{TLWDB}C;ougTp++b$Hck(OrEFe`*!xP{Gt7ho_*hXe^=GlJ$H6iGkfVj;Lp?F z@qM#T*t~r9ZT$TW_fKCo`|NWQ! z?b+}An>Ww?=v!{`|8sxy^=tp_-@pEur_Wx0#eewZ^*_hIAHQ$Sp9TNPAIz?Q=O^qx z_{MLUz4o>5oW1EYZ_m#EE&kic@B0q^ead$B$1gpY?SJ-<&-$}Su;?Z5H=xI4Sf|1bFmzx$cl2R`>3X7B&6o|>)rz1|;v#@qS(q+YZC`Tgtm zKk?eN{SUwS3-*8GLtnE0Iq&%^``stR{nssDy+8Tqe|G<8|NJZVzveft?0?Jq z^Y5qrkpG|h+1~!E{?~K+|M15?ZU6IL^;@%>Fa7=5`H%g(*?02m`EULBM`yqLNiX03 z@V8vs|H;32|NejU<}cpA@B7|9`xO2R;^$xT&HUab&B7~g?Emh6@tXa`=U(1_^-q5C z{?^+F%V%%=PhT_pZ+`33X1{dz+h$++6W=rYz)RjY`={^tXS45k z(}n$){n4fU&-?1n+JDvU&))yd@A`uMCqDg)_wWAZ7w-SP&wKU$E8h5y*^NK*zS$oC z-_Jk(l{>Q^|K#cH&ZmCk?C1Ulf5*$O{lx6r;==x?eLa8XbL~%O55MPsp8eWq|D)MA zz2(W-_kZ?Znf>DD{H58)p1Cl4`4@lA?A*btX8kAr`PrMFdfDt_uY2X}`~KiFXMZ&N zh3k+0`_J0{y*r<>|MS1{C$rDE@xj?Ue(B==>Z@P2|4qMsY5$wv{#pEgeP1{G=5P3v z+08eE*{kpV(DncQYtLpMdvZ7XufJ?F`;l+_3$yKO-p%jPJ~aD*H~!S@xBlJF%);sq z`0vd2_CMnrUb6p=<;m=me)IMFzxVZDy8pgc&G-NFkMMUr{wLRF|JU(foPFEhxH^0L zC;x-lm+)uRm*4YyvnRjzS7$H#jsI!((eMAo*^-}$Uc=u@bh!Vy`^%60li8oV{by%a z|JKK5fAcLrGkeK@w?BLJ2mjLSgOC38*;jqba`tn7aCrS={Qm0Gewcq(#LsyD@((_L z_IG>!eDFKIX!hp!{KvBo{Qf8JzwO6Aeg7<-+rRj+|8w^64}A0N@4n()vo}1GXYYK) zx6Z!e$3AucQ(ymQ_W#LC!~XyB?f-K2tyj0RUw-a`v+w?npPc>rzx|2X|G=MT{Hu3; z&+KDQzjOAP@z(4MU;gE@m;S;x%yQ!Y`+Vp9vmg2B%lF^;tFPSu^!MJs-`#%Q{*8a} zH)rqr#wTa*zW=*t=fCK?XMgR}e|q+zU-+5XfBBZ5p8ez#ADV?<{(b(giJzXmtA{FE7w2#Rg2j> z_-`>UzUAud_A9T?KI6Cl_3S_U>QCDL*7&Qlk39W5vx^5GntkdgzJK;1{=ZQNfBEms zzU&RJ-2V^0`u_cY@mH_!|LI3wvH!b2_D8dqzvm_UH-Ge#_rLVt{y(#S{^d7jFMH+R zoBj6sowNV#_D{_|`gOlQ`}IHf3HyJ#3;TcJ%YJ|M8=vuqv!D9wpSb^~FMsL&yT1L8 zXa4=8-~89!6F;86fAqflZp`*?TzgS}|0sX!X>^8X-*fvb_gKjP-`|1y+VC~s@U@S> z@hxxo;>W(`$*=k1*WP#keY4p${tndZ;j2ELCxq=hguAeW4)QL9+i(jG;0|nHbr?c% zJ;w(H%(;wx0#8E+##b3Ne!*ev3SYG_0NBJSfHbWgwBa-nIP%d>01)0sahAa6j(%9CC^bj+e_aLBW?8Z#NUJ1O5M6CiV zbnVvg3?G6A!Tp6Dao8jB81}HaC;fFE@|?rDGdSl5;2C%XJcCQft6<)R1#^HiUPSiv z_dKqk2XyDyfIXda?0NH$`uG#9^&@*?ta%NbCxUZaU4(Ep#GLKgzLm}~agFr72w=Q( zOFa|b(So;<7y~l*~tRI7C;W8Mz1#2eoK8|w3}7;wtit_5|k4mQLXr?f`TEE9F4{qi9?d=K1& z4xc3bV4i#M&fN1Y))$EL9NmZWekVKwPlJzlq0jxAe+nnC1^4NGTvV7#obhDG8Zs=2afTkOHH8q{9R_r37uFM``#nOQfwB5LbKj>P)Fbt~dF$P=_PbKQ zJ>5g3R_!q-pV+ZZ&sv}J_%2#If&Fg58difGeJ8Aa{H#sDbJVs4bF5d8lSO+^tY;1L z?Qskfn1W|J-66^8SVx$!_T4Gk!Mi!dewBK+;4)apeB~NEqd<)e*1K}p=JNaK25gDx z2fuS9bRT$l=iuF4nGg><2LwC9kbqm_9b}VJvHltOt7A}v)NlTd4arh z&=ae47H4-Z=ikB+c%D7@jvOQF9Z*Bt1d{dnT-bL7_VzrM_yVkyJkB-+>)b*HFxI&z z;6C01?+4F7dk+T?!FeA8-?0bbGAxI+K0jaM`|mmTvF&F)Z+8?~7wA~Wa_@6)pTydI zpP+kZ&OSvt{|R^oXW)5cYPdgpI>R&Y1a#oM%41|A*8JXhq}an;XWApq8rQ%&cXsW2 zQT@H(d-gW4{tAq}0++zpQ)&cicSvhT?5Eg08*@Cfb4cTT-+N*vpmZOB+|^)H$Z@ZAhUZvfcfZC@@o&JH zA7$NXgSSGAEmqmjhdCC!pYS+%uTR02+**^EW4Mejz!KeAH?W4sh;@I@fbX@RsmZQ= zUb;n_CDF^-a%pGqG;!AVz4N|(e|}-%>P)*f*Vr#$-+=3$IUb>hpD~~BMy|l# z>eiaV5^CRFdl%iWb9#5a8{T_8zah^0_KC0tXK^-XyaevoeV54H4n8vKJL-GjJ>3Rp zxdk0~7T)tQOkfJm*(2|QIrds1eG2lF^)V<#Vr*y9BdzbAPOhK#$$e!c*`#n18lwJu^OY zJtuwc%X3>{i(vgk%tdUsS#JlN+kIVzWE|*MuGHu4y{LOZFW8)IsvTYbE$}R_!WCG; zd6AJKJ$K$R%qzh7BA+A!oX4 z_vLyGTkw923%m9?dEl_kXpHB+M~_=bzjr*JXW${|iFe-*qCW=4U52_>0eYJac<6QdE;17fBk-@yu=JM?AeHE^N=WVaYAuNXPE#rqA&oS&+gR`5{ zqkE6`j{_O$d4|aUz@GN;j=Diku=&(;Cw7DV17Kb6*?8x9W{}4Ak$((%0o}98$SJlJ zDDAz3P`HB~YvdB?%;tE`KDX!XJk9a_6l;(5v|GpZF|vmZOtgVDy&pf1?*Hndeb=8y zZt-Vq6a4O@?ipKwH6)1CT_R853OU|Mz?ODMo``(|J`<#8e~iz4j6IEY4nMcfC8U%b)RD&bL)4G{zc1m&(o{{Lz!OV5H_u>2 zjBFvma!wd_V*|Hf_c_IhAs5FW(%?H^DZ{$Dy{|HS?m#&CDeWK&w7qMHtU&pWzl-Q`>_39`2gvC zJ_8Rx0`JUvz89B3+Xl>8L5{7SdtXDw=FgN*z_XC>duCaQzZ-DR9ropd?-%g%YL7%* zci27$SCGCd7m=Qc`jq@aOoFWB9-L1uFWS#U>hALbYh&ygUmUihlGAz6a*jTzuAd24j>xF$p4if;(^k`lraSYrUM|1U=Bl z@0oLOivJigEMX7Z!|=GiU8|?vIPc2(1yKX6o13mq2k>}4*atPLqpfrC;+nc?oBdp>9AlX<9jtIzz)*o~bGjAtK8+voS=e*%`^eL91zpo2@;t_;5A zuFd87PQhNzeiUvw*7k6?jvD~ z*pz{FMy`mRLdI^q6o~lsIj{8CJl}ZO-aY?asl9y{GQNy7K7ccH>e$_jy#w+lJPV%t z9ynJ|+{wTlE>P#8-h zI*(^|1)R@!GZp$Fm#_x!%z3SM#`<*L<~_G)-(we%9h?W>1@k>y&!wQdC+A;+=T~;^ z^Kj#^eQ%vdUIfp{8SLYp_mIZ+L%m4ugtfIpYgsXC~;txQFoj zE+)7J&gi}~()xYG2Z41)_ShaF?m;*}5A$XZ_|&K1UK6;-<*t1$o+18mbFg_X9Takm z$?6am_ug0I_E{35xNc~r+S?5AN2m$;|J_$=$5oe7K3j4I5^$Q$>XX?J?9P$2H#tI$- z&&hn>4SfZ@?j5m5FrK&H06t}f^g7h#Lt_iPH*b&ey*5J8)>IBSgU z?7U~s@-}w&Jb{eQJG0ISSl4--A;Ee;`vc@Y0^=SswiW)K8XKfDPGJJeUF*{!)wkFp zaRHg3fHP<-+IQ_tI&5oLf_0_?+9KQn`>l}XJNpJ{9Z8@+?gE9|`B}r*JJ5sYWRA7+ zpl=R|0e_zG-8Xl-YwtiX4&8Us{htTlm4}d?sW}}n3APZ)P1t&3tl@dv&pO5|2YTYp zcGx7g7~}MNe@EDh;B%in(t74~_-~=#0p;C!BO_Pfyr*ys6JnLV&)vMeV{6u1u$DEZ z*zIj@g2#ycY4`|4FvfYTvw|{j^Au)^;Ny(e_k1E!J;4?xi+lGON}IJ{x2Li8kGr#H&Nc<(s*U^={>uw`QD=?r zx&z-2^)#&2w-AU6$cT&uKhisx4tTECeiZD}fp_4{;`v=-ZH)KqT+Vv}zN5FnvlQzu zSzjaOL2!2SwPpGiq~BwC^YH`3#$VR^Wc!{}^k$^)t1_-b-TqIq%(&;H}W(+YmPeXYez5 z1CELHES~`9ajqW)&-^mdb5M7_9-PPCp7#-UWkI^vGW4ji=zU=(IKaP0-Zi)h0o>DV z=pYjtkmeP*fPcJ8&*mvOk1bLw8uwhA`Sz}TXVmlKZr!`KDLh8blkhaSH|H>aNxt{# zIdtIJX&>KX&m!R;=iTCW&-MMGuRUUHd$mSE_i?5`UW#poXAaxhTtIqv<{c4#3o^Wy|{Mfh&)C5e%ND=I-ZNqhP9u0 z_d4F8@g25OhR+@`*7I2+??CT*(P}!&YR9_pzu0ED%i6oPrth3mn|*t5PVcpzF?r@L zu?6sa&yX3Ld5;n2-UCcwYs}!gg}z*Hu5dhOZlrq*5s1l1iO{o7#INvg!LxQ>XK(`7 z7HvPy=nN;w4*L|0lXJw_yW3G7*5G-VZ{I>6?=NFd=wYyDq&0oqlkdW1xC-_#HkKxj zSr_z(Eur5eCc-n|T^YAQ-UjPvGiDDC^dldFB|HH3y9<}mt(&P^doqvpR}diMbFUNV ziAms|0y2V+z0FnD{p^tJ*VrZYJV?*hxC9&6!W6VQV~s}!@Ovc1=hxG2CU)!_7?Zf`3`IMsw{90pYxe>~ko_SJ@b3_`i)zHIocz1yY^mc z@4y;9#;+j7_8#do=JH{y?|z)44E|GmV$KAPAi#7-4ZdT%8+!#P;B4McA6t(&Z?)W$ zz7u$oUTf@2n1V5JM=f$P@)~~o71jZi9rk+f=&=mFk&SYmBsRr%FMbZ{uY0Ty%mKfLB0K<3K*1NF!deg(06N)>Ra?a zwi;{d*#8#NeHs_Red=@1cVPkr_7?5zH|U=ERtXu5?eH6)sc($W6zMtkNY8tXyn?i+ zH4DCH4%=tPeS3e-qP_z2J-;5t`B&KNca=KU^o-np8DawRE}X(u?XlH~$UyCD3)T?9 zc?0%@JOJl9f7m{w4S@=`?F1R{7koW<{?6{19)Wk^V@)4>TwJs}nBsTu z#xKz`n5XXka=dp>kvij_H$RDfjQ;|4oOc4&bRS#rJlx~>xlQQ4OU`v*RtI1fFxZr;9oBXUjb1orT$ zJ9mae+(yYeFxAF-f?UHHOco7i?cgRjQ@|IAO-Rqd_rmy#$V(9L*>?*Cu3+2Kj&7gR z9sa|1#_rh|drqK(dY07ie!Tl2*hT*aPQczr;CeHfp> zIuXjQotyVF1#6wc0jwb`TE7cOd-T-oc5Tij(z~(#L+}JtpMEg5lolIUpMdxEIJ_G? zQ|~4i1Gm-Z=rHcjb8z-4Sj+j1aZcrC*UrlN;yY#UHF&;xPTmV}kC(t+D>%aET}MH8 z&mBy`nVupx!!vNY!^U^f2A_F8Io`Vidzp6v+5@b~%e(geZIJjO0L0r){Xd< zyS4}OA0giv{7i1(m>4Orw`lp^ku`ea-Ty^!pZ%_#<5nNsf&_Vx+B)$3Jumxu-Vc-O z-CPFGyr*t}39(aHf^}Sb*6yp|_q;0T#!az%UivaV&*9c#`~AoVkiP(Lg-786ICp*j ziS38@NY2DODA>F=>nIBZY%Am`EC;;zyO6;5WDQ$bfotEbaqs^7xhKKoSszf#`xejlJhro4d)JNY7TBPk1O6Oa&Jvq5xKH;GmE?J*&UgX8 z^LhuK;}Yo^_2>x-)Ex3sYAqtCj|aSq!tG4Atm^bi?| znIh7j*0jE{o|Ao7;MzSouQeV7?|qF;4!}J7d)DsLx`nl|Pk`s@`~Nsxft%3NcS>$T zc1Ue2cp4r9=iEcOCwn`~2I(Gr{H_eK^-86ER_7I_Uk8#~$ zJL~Bnoh!pBSl>Ck$0hjtDQ79L+_n2F#9dz8J2P{9E?_GW1%DFGKD#FGfOQYfY+v_j zecvVD6LsxRgLl<~(%Q~$zdg7B>NnsFyL&41a3=e%A@AC+ho0$Eu)^m1;<+5a3L;G4 zVf;$xaDSflcvfN2=G|a7-}mYm;=Iuz6S${q#9o3P{Sq}F1m9=(s*8{k_F20gtF zh}lD5Aze?9Q*gb8W7xnQ@O?A3AhmDd!Nd0ZaDUzPeTvvRaHa%#*YN#uAJ+A;rg7%T za=<-KVFJ!;PbpwdiPYt@716Eb-nKhz;91mn1^f7GP;KVARISELi0jDfmEimDym^S} zV>?^d`t`{a-TrH2V(mQ#*3Q>MShV+ei7xu4NN1WTncIK==)3EAy&IBt;(AaP)*HxR z4bLrN&!9YG?U`Q%`$u9{U?1nTPY+iY?T(%G6binGT;n^BuOOY*_}k!nqYSat>d^K1 z{&*L@ySLSM^a1xGnRxT8nUMNDEBgiX3*-biLib)ziE+M;cy;~m=V^Ehd>6D`0egG? z_O^!gtb2LUzE?KnXPB^-HTpT?>gSf&K&&$rbkA)IOV;)`PhZ^!_p{zLy~8#IYi(c) zx1hitJszdLJ+48&Yqgxm^E1YpCAJzX^en8ksXK%7tPVrnQ^a-*3IE=raSj>5_i96| z=i&L;_r2izyS%uHfvp2v|rQw?sfhE55W?;flOW)a@XW| zm)1Lg3~2}6J#qrp^?c+AeAny#cDxs0edly$-yiSKxvc5^jJejXtmf?;w%Ar+j8BgD&Px37xDxHsphv*62xzK8AZ zE+g+!!?mB~Ex2EQW^ljHfbX!Z;1F)Xxdn5Bwe8U(Gq@LVUzuJBxq{7({t)TyUV3cK z=viDq8sFoq54!j2nfrT$i%?>VJHh8ZJiiFmw!by+paD(qi0O&81#P zoV5}@=gRoh6F%cSlLKVd#{1A?pAs*wBWq(bxVJmto!IX#bcbyp);XTH?+Isg?tp&+ z_P@4hzxRAMaS*{jZVM(P#Y(t3e(Nwg-q`f!|L`_ZN@_ zxq-wUCdAer?9CoMoL@8^B-R?vXHFr0gWtJ22(10w9^-Sr#u@KDYWMxIw|msTA})jb z7-zVE?%BELQ}7J+d#>&&6zbAvivAFG>)!T zbm!>~TW=z2jByF}7VR8-cRZVpwKn?|{PypMSYw>Q^%Cj*J$u;p@64VHZW&HjCbpOT04;M*=zUxwEq!A?AG(Kp7)lJo`-)fRKPPe*ILs>+eaW~ z3(pu++I%UtIQQe7T*c?zd55Ritl|95U~B|uQ+GedV9f#-7Y*~qIgiorg8AONeV54k zSwlv!pPzrvp8d#55tm}yRf$)j`w+rZm$BK%h|4-gL~8GXYLO4;H=KH1#<&y>plY?hEvuV zj==Nk=SWC!3sw-Y#~nVTdr0(r80tLukFi-}3iclVeEhxCe*%`+%-cYs#tLcd1OnL4 zJ|(hN$EJS`_H>UO+=K_gcj78sflF{qPJs*&*06#!=!v@SU72Bt{Q!FKY@R)ApR2ba z9{}rFV*}Q49zP=+aAtGQE!te(L&kOtr?7_c*QY?QScZKKzB2UJmtZtH7s0#p=eUo+F&scxw0Ft($@A>V>5wbP z=m9*_@m#MRw$GxUUC(AsobRglvVmwk_g(j8{tmE@Pb$q9kxP7!Q{!pazzw*(X#4aX zy8s~)5B2jyoi#ZLd?&26L28>0oXuOF`R>iR&LdBdMX2RjWtf7qyKnb9f#r@o>PF%+ zL~O>~hAp@sXSeqOdWX$AH3yrY{RqzJ=h4sJh_(B9tCNvp+y=Z;>(0<&1|LcqIKf8~2;N$N9 zV$c8nDbI`O|6=!l|MWHYznK0ndjI@2#4qOlWBvPw-~Q2G{eh3~{lEXd`)|CM{nuan z`Ci)cebgiU-i}DWBkzFUeHj91^E=k>zYhF9tKY$qQ)FGg+mPDU$P?(%bHN1@)`fLM z_q%ijwLa_VVe4(|dhi7vzdt9e{VsPNQ8)Jp0`bmfpD=H|+{50Y*&=p*D|}#cDGn*?-oows)l>Iv*)56{gxo#zOmak0f^WOB`; z-MkWAwZ7)Y$f7h7x_*0h#5!M{Ym6hzC&!^4vM(*h+L1V)@g4BoyqEKb)GgIj@7#XY zu(mR@PT=D{67;d*+VcrJD9z@+BYxM`=#idFK~9ko?3-C@D>1ZKbZ1-6S!?4-vHg7M z&szSP@47$~P)-yvITXRtB*-^va zv-mTOcbds5NY}9pYi!24UuO!eee9*Zk1X&xPlv4ihWMVC@tu#v=m} zn|X_~egW%N8+r!&DXWiobF6JGZ4tRrpQD34r^tjfu0UVfXEq^&KBV{P981>L8RH|m zJ+uWuH+D^&_3e{m`+4ek(4E;{o=?qD$9Di7^sG0?DHzw6dvm%E=idx6;LrGvS=T+X zE?|vY^Lukzw?2h`#`bzh`#sxh=tY5^)70IF|%`uI(L=>KR;{ z(}T8pe(3IdY7TnBUcesCQSS@;YVZeq&LX|~uGO@5&3~@l)7V4Xk{G1~u!p%B3OI+4 zdSBYvXd7o*5BZ+m0XSCxJ*@;i3}v9xuJZS4&9F~)iI>(g%S z0OPuT@35}36!M088NHBy1h=7s6Yw7GoA6u1+BFxO{U$I4`&U0P3AAUVb+rd%K^o_y zT!XPa)N^KbV_bWl^~{ldX|`L$i)SRpIcJZ40_NIxtf4*2cSb z|N3feWCU}^*cI{i_biq5%!inwg!Mes&w@GDwnju|@Mp9DTXG8=4-)J1pA&PWk8N*t zjVZGppIY1cJSp-Ia2EFyOOrj)Ty5^heDe~vf?m|gaV>#3;}i0Q`^(tY*v?=H`lE5M z8rBE%_MKI*1@sNrW2|X^a{^@aYTsc4=empZ9*xn~D`EWHbl80Q*v{9S8jB2&OUu#j zI?daj3--v`cgGnsbrKlULn@7PhnTQiXN7dXaUSYpegb6>o({if(ML?Bt^G{9mL<9F zxyM&n2Z&&8^-zYINc9PJzjNyM&gJ5zy;{bZw0uJlhRS!F`n2@~jmGU3rSnTI#tV zi63=ym#zmUNI7<0T3n2Tv+j}hk&Nz));nfhphq_+q00u$?dHvHo;3p8!IohS_UmB- zVb|UP>n7p~Xjd-vBdy~+qpju_eBhb5j>ObFY_5G`Y(HB#TL(VQXkLL8`cmeDUpwm_ znPQtgBi)br<|N^4?U5h|KGDBs?VYWVePjk~HIJN#ZAzR^MD_&*n^@1i_)L)rR)aLB zLpRP^6V}ESuts7X(UlpL&c8u+>hw0=cdynS=FhCR_)o$7OfA0~v>7)+x_|Sh+8{vf ziOpFytZ%~#^xJO&2T;Gq=B<`zA^MFC5LpMXj#9tp;4I3eIyT=O{UH4=b6>`f zPmZh$F|TiN&T7Ae)b3j8TGWroUqS7IEZD^UL3o_a+OBgU*EKRiKvx>$S-Ahgy2fL( zzB8Mfu!jL>G$x7j!jjxIWMjc|aPP)!kh!2!D=NX*9%)QaLWxM$5joX9hthIfdmiE) zh1lxU{=+&GYmMm~A6Tns8|+bAZ2PS`>z=hWHt6OV>-{*N1lCKWXRNRGQ^ruADp~s& zZ_J%R*4WtQcP|Iv8Dwn672+~-i|mmBB6^JI$bg(6SJ1fzd${(oMgsFP)OBpKXYEtZ zg8kXwdIgLhW7pX2=d2lNys_i?q`~G|zw=l-fq7Hb8&HqfoKf4D=lgE2v8MYlzr=R- zt9g?fq`$YbzO{lB>bM5i&gf%I)yute(Qcm%5himCkRvfj<7-`f#`);8u4{7wEUCQ$ z^$k*A4=FZjZ*9(%kO67V80+~x|DI!a_ANH|r*8tzHNFSVrp?)$M|{S<>V0f}Z2`UR z2e~BAn)b-}A_S=Z=USfcoN?CGzJfIbm_Y4Ej*s~zwx8>b)8Fle?-%UWt}#e;>6OGy zAx8GaIVYsgc;?P$ozbSWkNM_!rV-t|jyeIk#&!l-JJLtJKw!O@w;E%QS`XJ2``!X2 zw%J@?bzP|Q0-RaOu6-x3kk;*qQ9qlbgY!&Yphx_*m%d%om*~C&DclEU{yv-WWSLoX8<#WRN?qFLYQ*1p>(WismJ!I^ztrd4e-O_qYu=V); zvu|fLHp0eOWI~R)<{yE3@Eu)(b7t1wMMRpjBqoFP^t*P?#?;x7g;;GX1HdpP-z9V+;(O&hlMQ*}&8+jM|33wj~jJplm`*}Mv^R4gQ2W^K) z{Pyp_vo+uIb(TQ<61isWXV2IS-oV~%>Y5wnD-+_hgj?9-ud!C24FJ6nT$GJXnSs1ud$4`M-@Z>)CK)7`z#<7m^L z@q5Ps>MZ0H;%{lgXRfu(cQ()8ya=N&5M#c!pq&|fe6Pkc@5^A1#KDC44OBbz)3CNr z1ot=gDa70B!eRUO0nQUy_ry87YwLJ6wSPno>x8a9v-Z3a^w{((ov$DxC@ZDOiSgKs z&B%guKGz`ve)rh3uIG;Kxw&?JbFw}#ZvsA}FA~#fM>?Z5jZ4UaZSSzX3jtd!O+V5v z#24b$#AW0P8Hh1%>=UtDe{DQy_cJ}Nr|6k@^YsO!H41Evo3}GECdJlY+qhS4J&dzg zNv!oxq2Tk3tf}nOd9=G{*JF)>p0WGiNuHulVGDtEkKNcB&wVNVcYAAODs7&GZoIxg zylZQi8zF<|IL_*4X*KxGnLr^%X}q=Ex98;lN(&H-o0J~@H+PSRVqi3X_i-?pkZ+(1MjoBDm z+MF5PJ&V1wc9>{~f$@|=1ni%4Vu=si^baA(#S@6Xu2v|b5mj(bnW!1!4j z?{&00TOckIGtTEa?%MZDL|>sBugt`cGt?TagAkXHVyzyW%`;O5u0gwfe5^55hxNSujN2jYV_zTb&YB?&w%Tvk&LtFlV4SmAyY^uwCnSuBh!k#@Q2j>elsMdSpG@z=Q5AzAG6$5?5=YC-9l#bA}0+>*x9y ztPx9#pP&cOW`A>h>U)4RSKOzwEYXdzo;8eHvsNzh`sNoq?tO z?d6lPhuC7Z_oWfg>l&YR$M-exR2_dVLw)?#PT*|Duh6$(OlBRCqP<7@xsLNzOMfsH z8A1O6Yq6Jk?!$QJPQ-XN9e5@_=GJ`HJ-YX0&hx(xQb*lB&Kij+Lyma~dWiT^Y3sTt zYM8SEWf*+M`sC93sN1i&j;+?3_$4@N2JK@^ZyXe3i3v*95!9E&`+N!=6gd%ro{}3${#L0Bt^fZ23#pV$FcGmg^0e z7YBPr@9;UR`V_sunmF?&kgywXoYHkc2BdxOz{)sm1zSe%z}VGp$YX81ao(}9agcaw z{1ji0-8knl-?ecOn|cEM#%6UMeKEFuC-kGC&S;CQ3x3a1J)zh8;jFFYJ5-RhKhk=XF8^ZUuz6oEMl zx<9{0Vk0(Xfibu4YdAA~nV9NF8lPY_)KvP;+h3fq;P-sYImXw4F{=S}eStX1kOr)$ zZ;Cty-_gK&vO~spW?OY^?#)`BLG@#H-4pl1`r!pJ+PnF1e#Q;4r`XJ^dq7&hkf+~0 z+3N(9`m8yHgkH}VX`Hi!dDD$svyQRZja}oje}{dE4EjKspg>lqmQRc=-kK{g#u~;P z?An=*Yr@-k;Gcp$jPnd9;5`02NN4KNog>7C@d>-Wf=~SzR*+fy)EdY-2lH*_t;Up? z5?jAoXPDhO0TMpi0%$HMP6W#76Xt-Q14#hFH(M+OTJIXDEXn(2X&^elKAU=*EWu&+87j&ZV7kmDoIQ z>9ISv`*Pk0*7Y$qfjxa9_8MPUfc1ThNf8CIx_ff%LyZ1j;tTAo<6v7F15+h2_HnK5 zSx7x|*3LNgU{${Z^lP`?N}XqviFH5N#v0w6@k^_p@s)_~y>Si!d;J*{oAv9cNtq|omp$^SsQ1Mq;H_^ z2iqEq36O}7NMi#eZS>bK)=glHcKc@ZAUrWPj47qfs?A5)Lw%=8tCd+Nu%9)1SfZO- zA9Q`$HCUsb=U_8mdyPTw!PrhedVq1C+9N2ntNZBpF(;JP-+Inbb7Qk>n}T(~k-St<8wdfy4Zc4Ip5@gA+yA-yAKavpPgWo+xs5bru+H{PfA#ivb-tF`g_e%Nmc z_EE2K5y{*U*_YPWx)NAt;;a$T{r+EreG-2AiuuMbvD?${4&x+ZpCZi-=(SgA_4HZW znLM|#kM+ZVHf!fOG2~@*dng08fYe_KD)wGEFxu>IOnsEFBsQuS!0%iu<5>5x`RZL_ ztA1=-a@o{}GVNVl$Ww6&gbiZ>GF|pu7-?+veAc8sSwN7b$-17u# z4AS`};eXyvP62BeH=gyzIA)1+B)A7-(h#@CU!*i!#xC|RgT5jsK+oEK?%jB8N^SaT zPozESnQF&B?j;TS3cESZrOnu?BaQWYE>S0pap=~3{vAYjCT*Gg<&IclBDl7uz9slR z8Y{=(oExN1jtyfHjPI#^3%*zgLQm*EF*2XA%-y5gBjGnDqH8m@W34^AE^VfO-`q9! zJ5cml2fuv^_+33EW`#6Qn``U(7-!unsE_unJ%;umtsP)Vu032E(?dib&)Rbj*s349 zIo6hhUH^%_;Dx!yg~3+ijD@HqCnIwV^|9B4*ihO$J+fnM4{LbWeMFzPd-05{uWf~H zTtv5yvXs^*Ya8v@%oBUrLthUCf|O=g8sna1sl=b5%6W^6#5tdjzB(6c_czYfV^73S z;1sst+8)MSUZ5KT+6t(<=Zv1w$9a3!_A@_X_g&OK&d}kr#>$uh-y?tRG-nFtb;va+ zC)&udwsY4S0iC#rFKI_N&PUnDAx;~ZAFw%RKzE&47Z@LVxsI?JbZd{f#yN5+%u4CYv417Xp=PwVFv-QSn2^po2mGcrIKP>*GJzjydy z1NIF_AJ51>skELcHrcbz=+QNLmFUKr@7(o??dQFE)@ATvi^OHt$|=+*HvjlOSkot# z7P}teQfzy&PwmIrn1H{2F3{bte$VPIEWxwzjFRhJi?@fmbwX+S3f$ho0d8Ry#tm>$AobTPpDSdwr~7oKMsr$%p!|UO{2)uWtd_ zgFUPfc5S8?p1(fr&XthnRGoEk4erZ1R>S&?^#<0Ev6~Z+{eZgX;+_k7jKd79!;ovu zy~Ebax{-Avc8Puj+A_E<jx#5osU&;~56l?n9=aykX?IM(W<>*>#!-UVQz^?NT6l&+MuHJa(V|{lX>aTGbX};?X7#|?V)^~-iuy!tW>qmUXSu3z!!V1ir zDj`X1HH^>rYW%#_x`T9&0jwMK6W2ol?R8&+ZS8v3_GCSu4qF0e)88#xzH4JVr-Q}IaeRschd^p_j+$(K&$(_#ANJbh*vuA zIG=IFwKaFmmyxH$C45t)_04gf0zGyy#~9yrr3A2!eGr*3YY4PRz$jKAJFOPnKl z>}6eJW5IxKtbKZ`Q~XiHXwVX zdWOi_I`*=Tn3Dv%b6Hn^H*W;;oZWL3>rC+%WDn*W;~DuL734Aa^V0!%hQ`$=^3T%p zAH#Qs&-c)?ogxpwnYNIL&BO+EB{BL#gwkdzgKn%dCG7gFQP<3!@I`F(9YcDS&YaPW z9qkirJ?j{;>qmMX1?j#gpsj}vy*{O_%}Ze36092{p!>b$JnF{OIF`nY?}TSyFCS%! zt!^$a(6zC4hDf~ioyl`cq8-dJE{)Q<`k!llbDz({=2L4BXKvhK7_gpov_<4fJJQ~+ z*YmdDW9$jC{y{G1*r;hvM(>F)*F-<(7H z0qcZxZyB2u&^GQn@OaKWc4rgk_kOJ5tR1@Z`;Iv0>97vuL}H@r;mNVR2L)-Jh^+mP zxwJlk$J)kN-yF}vy;@Iu_33AAp7VD26E@cYscw&l7Q?T()C%YmY|d<)=VFZr^&Q6- z(2eP1J2NcL=eSSL(U=}K#L0AE^m%6WiPWdBvjjNXAt4tePwmg_fq3KWSkKhR*o?JSR1PuTZ-h1W+5=tNigh6_mS)eak04-kAGHSS4qVsykr{tNuIJ6> zUVCiDdtTP_?C)SVreJqoV*)<=xJS>$y^Qt8yh2Q!ecs}&>-_H1T3d8yneJMy{#=^h z`t?0UTF0Cm+icb_E|O<&XRt<(-}!1T`UKq?0o~dOoHLeIt7yZYSX;{)7Y|v}E8uf& zta(1U4Dpf09FTQJbbS-z?b$((?*M6BME9AZ*Lk$_xPNQ-Gl=g)y|dC{>>bqME^$40 z4=YHq)wZAarD1HR=FN@YwJtM_yXOT?4{3n{M=>y|`BQcLzAF)*_9Jx9{qiC8S<4#sSnIE`bBYN${(r5^u~q=@U+HY-*{5^; zdgL4*PZ{ietjl_)b>`CU;}&t=fpW~%Z~TN@`+Dy7Jp<=0*o?RL*ppSwN%)*)1nHJnMtyJuD7BDN5*6Js4~YcoHwj;!_T(_Wy~ z#O}VmlZ34wFvoY?J8;Gg)CYM0)^i?njUD@}Sl3$iW#1WJ1Z9oGW)FJ@WQHZW`PvJp zkMo){p=JPUtKWqU6nw^MH_qIQR8Q#EIl$MmuFsHD>lboaTQ5fL1)0EeUqis}(`k<_ z#@H~d^=a=53RtuH)ggoLhP92+?>;-o*o-Nludp^QL4XX-rqrhHIiG@$Gd{TBJM<9W zu3?=M^bT8(43J~%r+x?0ywxh$BV^XqhHr^(ymsU3eCXbxb1HrG7h;_!Ku4_mTtPwi zopGPmP_AJDD+uKJOb5F)?UN!#q_Ngh?<4vg9a-ZzKmFEMn&apGNFD0_O3NwuB8+o; z_RgwpLTtS!^b9^5eD$4(#1wQewo3HzagWYa=fa-RJvZ&fTIVh}pM88vX|+8M_mN+? zMz40Ij89=1+g^KgeVd`4pFR8Z`1OtNo%i9)wGMJhoc$))tB=@#JtC*rv?tc~cm9H| z-Or1;*6Fdik5kC3wT(XGt-r>%oVPjhU~?_@P1v*>Zw=p-YTGf(7(*RvZ?KK^l@T)3 z`RBy&Sj&7L??07R*R}DkS4i^-gbRo5X9w%K z?#WS~pxdXuU#zFZB_*~E`FGj)E%pG`=%GX3Kt^9FWAi1}eh#%cw|3*F zto?biphpOTEyWfyCAI@|^hNC2dPuCbx&PV|UoZ~huGOH?;ecrT%-5c z3K$zA?3%sWh;hBrM$c3Rojmv9nRejg+F0L%CDM5Jpzb^P|8e!Uz?LMQDFM)MB81cx#;rBCY zKX+&yQQ)sS`kkc$bFAr%<1WNj<2e!YjxEo5-7Rr`4hx())^?vt3E-_`v76-Ya)!Yy$;Y=emQk1Lrip!{n~j$E~4#u{htIPw=_*4yz}*Z$tcvm!GmuKQ(B63%)``#eB}Z`gh&KCrE)-_O{N zy#VbJ{13SHCTDKo-_E#k2e=1zB1g^($!Rgfh_8tYbkM*;Oa*ckD9!aA%26lr@gCga zoK4-_ia$XE>dyD&c0O-x=evU+&;*Zf^3kbc+8 zd8{GUGuQmU4kZ}B_fW_01MgyNg3<4^JGOHb@VFs80dmHVTJQ^g^R621z&P{GmDf)2 zfvxPI3g&f&1kNfSKt1#PbA)lsI*z(`#P4T*2VpF{x#kFaYB%JU*Zf`NPF#VL4`2=N zZhphQ@Y_2+&X_A_jCTz5C$aXgF=wv@>teywSadPiE{--*0iAtr#bCGtFS7o^PHK!x|&x|>87 z+D*xE2J7^gvEFG1@(XcyU>)l@t8-h!{7#H|0lx1QyCOSwAkEh=9mBkl z?Yv`Or*4e<5!MP|-|;)<$!RBi$+I@nJy$u-xO<|!<{a+8-FSE3x$QQ0JmQISCXfCK ztULDjH*D}A-n+>g+lcuFBH-Zd*U+88bH``DF;mW2t>^oe*v>WT#90F>PM^GSN^5&Y z{X97d^Z*@HV(ep2XO(xZ4jyp!b(WC>x7_ADa>i9T{lI*CncKihY|FW0YwgCskLL$( zFHh#I@slU+E3lU}B4$1120!B9t!<5jx32LXb9&C$16|u5&bXoPzc)`#X`u^}oQKTc;#0bbH#TqBk&fxhOx6@y{Fapel2^KB2oKYg9RO>y$V-<|ujg zfOD?v@2QL(aY%pejC^Bm(8%#UMEuzQ-_BV5=!tB^DGPGUpbz-$Bk^mW`92qYU5I-64>^oRMfj939`gfpiY$Ej;-;sa3CbuFtbvRFp8B>roXV1}xn82>^ z2fFu2biV_Pbte6nZk)EWb$L#Y`I+<{#x>ynQ=2&L%jZ6QUYT=WN}>ONFG?_GgLilG z5$sDp*nl-X>udVlkvRi7XDY}6g!vVWd2;r|8mC$V^h8o;?K z8$2jLee44NhTcI#H-F?ZLq(VKnCH7$6Fneavo_k`tz}>5u!gzb$^0(hz0(~e_7k7= ztYyuPl>4u5^EsLS7293>6P)&@&7IU2b>xBU$OH{oOWBa#Q@(@ixhd`Mu~y`axwsND zK3?Zeso~wPKKAvj%9!t)vSGVZ->Uvtrw$BZm*^u6j;RUBAAoj`FJE9Rk14)g15T z9vbmgAKa)>=KTuxxC^#2V@m(KrH)?ld#q>e!y0k+U>19s_W;&kfPIbG&;#Uc{yR!z zzJqUIqkbXA+%Ye;oNK}Uki%y!dscK$=6o)`Ump87SBsf@$wK~vm`}(<9jqXU6Qdt_ z=8yVk4SQGo72a8t){+X4tTEWQWe*?~DjWHvBYc{YF zQ-O9BQ3#x$Q=iG;?S*~^K}uZ#b@8Zb_S84UHke#GKijlnkOGq9&U0^M59Dn>1&KcjpWqrPzHkr~F1tN4KAS-R}x^@B?^i(DUT9A2Ue0b`?{Du7G;HrKFXA@s5=em^NN4;6oVVwc)@a~)JM-YpagIc)8|!Y46ZUAp zI@WB&hzcGz@__mFb^jm09A{Nt`y9FW${RWQ(VyS}I(C4JGV0Wm+x5R@uLa%D?TWP5 z_1!SuQS!dRSvOv@PDQ`!S>O4CxB|>^o*>7*o*whL==WU{=6MhIU(oHVy&*GZJ?r}X zFMlOYG`R6hi8J<^vy*Et`?Q>J4(}=V6Zq`qyJHV)dLrjs9l6MJmm0rnj@ZA?xG^^| zN$eB3-sbF`cOxB2@) zLl0seNdMcW6`9xtxYLsJ89ZLphwC7~j;|u!g}t*T4+cXabB(j0$2UC&F-*$+71?+>J(n@>RB9r*y;YsQRP z)J$@kIQ@m#1^o-yfpgm1-;MQYe*kmLHAcOI`a16r^Z(Xv!jB}b_u+=NN^H=PS zp2%_ZamUa2ihlGA^duda>mAgc%N}DsVvRenM=WxMYs3|Jb!T_p15{vNpWO!ZJ5Rg- z{2g820>*P;j!`%919fM6;7{}>$Jw#v%^xSHkhh<+Bya}%sGI8@%;~4D?K~a7^$NUq z52U$YLB*%vV}0k9^DZ8BV;5kY`J;Z$o=whJr2M!ia`b}NAILE~eA6eVj_knMs*-t? zh4@6;Q)$lff=8BMU+5R#u${@i2dE(6?AcH0os28^3Nmx{vaY$-J%|bUm^*`=-vQV) zHRc@lxf8nrW6e#ZXkdd^x0kiO({-N;U-8?^*iZPqmo*!B0zdEW!@FO#t=Yi}Mm=&< z`!!>Xv$lPkG1&e&?hCrlxWE_W4(`BMZSxvP5HV+Po`!DDhVE?JJ2pq&xyC#xGiQh( zoVW+L1M8V@ymvWHzGGXnp1d-#Bj>X!9AnBE6SwpG2D!l5`;a^Jk~25SY36|)zcPS+ zq0hY~(q76$?$q(T$;Ojc^RyX!Oa3Ekf7YsLgJ$Z6zQSKVWs5o0_tO$n5D;CuKD zByq-?Cpsu60{`d>W!?syxm|vAZF7vP%2UsiAF;18j&X$=`YPONJ-9Kq95v*O5f^{W ztzC$-=8oUJsMibF9sPmSKX~K@j#G}>#CRfzNwxW|L{83J_fuZ@kqs=`0XWvmm~~h6 zobmR#1LM`r(XPN}VK47pPTRa9$Gyi1{+TmR&K}B!w3hd^UjXAe`~%(Z)hpH=b$r?$ zrMb!go)+`x=&17=)Zhxd@3Ze>=C60sn&w|;r(VZ)mWoW`mEO^~kjvR;JVWC2UHv+` zws$D<$m)O71{9F)A@bhH(;$h*tKui`1NVKla4)N z(61d8^l>llY}_4Ke?u-{-#MO8=Ik+M zAjbI5$~Sx*dsxfb$_@_v9otyt#T8B?egWeIF<|VP8B5MxDBb6H9_qyVX9H!x7v;%2 z;|^9J@0^LxJCB*6$FqU6u5-G(j^4m#Ud-4Lk8Yj8Q6sTCP|5``&sh?D1^S&!pLg-y z)wV|k)=PXw98d@8hv%4U|NJv0Gw1V_$jp3J735V~o|l zdN=Agvwf}aEMTO>2nJC*}-(r+M!O_zIl+_qX};xu73% zTNQjo1eKJ#sKHJ`Y7#j$KUYjS+AYa&vxdFca8)5E7H&7IQaa#5aZpUzC^dKc^zHfhA-ssS8VV3kMHx(6b1bO z%nf|*%-tHVEO4KZV>if#oj_@>_pj*B7d9tiJ`-o}Irsa}hc4HU*YjF$=)t>yG*;hu zR_FeQpOBPikNO(1apv`yT){8@UA*Rf{db4s-q`lL<1^1YjXZL8bbCe+!)Y<^R^a{n z!f#0Ptgrsx;r7?eah_2V-xXs&d-=NzEKeAXMWKD(Z9 z$Orgx*-*xrTY<68=nVGVK?C2E#F=l5@c|M&PTRarOow0Kj?lm*suF6V~%iA7H&vgIM498gqwo_UQ1|9!Ck9GG}a{8`F{QzM&hlp%;Pf zz4cXeq3ulLox)DwJaP@)+68&-K;3!tjXK;nPyGphkZXQs8M!;0H3C$iUgNBN#(K8~ z^e^w{e+x-W0DVK=W=(Z#UVAS1okz}{Ex={`gr%$fPW3Nc# zt(%FCZBPBi`wT?Kr`>=#qsH6ZdAZI0&Sf3{eV#Qv$_x}_AQ$AW4m{TJ4)QG~Hhnqg z^BETS4y@sEcjF}ge9g~6MK)~rAZN@{ViN?&sg$XHJJR_Y>SV(9y$KptPp7mK(K^D|Q9W z<)1C?=wGyfymR`UWleSG@z^JF=IiSq;SO`q_1Qz&PF;S~Kt{~?F$cQe^Q|efdovw(ps>8o1_JjL`-OJoczyBYp=fFs8xX(S2_ADmi2IS#S8DOYlGVt=G}N zpE=fipjY55&S|Ypj{GqbIr@!t?xsBb-;lqNvm(t~PH9cwEsyu_F>9?aVu8;@*>g&t zlke|U$6DUu0l%YL_W<_JvzB!Oy%AUBkk02k#s;v?7a-S(`HsE;@7lq~X*jz zfJE0xc%I-myI|i1GkH}9B00s zQHMK>&0p`9`2~#UXDmML72O%WBAu-wozI>PWX$@$BlZ~copFACf9{1E>eg`|l|17` z!Do*q2MXvppTnq!K6s=x5^0ROxz({0-4WOEdp0x0%18k3RJD&rJSq?TioL&-zuH8Y^*$bcTj}Vt)fWaaA5z-`dU}F?$wH z;JaX)xk{gN(-%`(JJH9v%^2@Ae$F7Bt1!=oJk*mIa<03K8DrKnrlIRAa?01-UB$Mq zXMuYFzmL9v4diY947;N*=+?Ss8+S_*;U2&o^S$>DQs6iDpWbJ`9>fD@aG%!j-Ft$B z|MS%Uhd%D(oX%$*+82>kRHBK!;1DJ2&3g!0&NRXR^Qf6{)?z z%dNxiI|QGd<-1@?ws z!4rJkrr($X%=J6sc+I)ZGiM;?Guh$pz&d^=KWpD+oef!uuf!Il?{q^7XZ{S*81`U4 z&mXUu+hab@NNhNB5;%kN*&~TDcSZN^6~9M0bfh`fc9w{lW3HbY?eYSN4vh6amu+p) z;GIF<-lHG7d%1qz1iXD(%)Jd8y%Oh~`YO_WDi3AOXY4HIwO1L(y9Bn!`*!fqr=C;S zzV4iG4P2+*W^HZjUdKKgdDbj3>zFI2^z7(0=FH#Ft=oZnNcmIuJ_S9rW9~)2b|4Fo z7YB8Oxp{Kte!vBF%xRa8yZd=`&-O6C19jh@<#o1-+{LGG2c4Y6x!ttAdlC1Ph^k({QHuA z<3GTH-@evwV1e7@a@H8LxFc+JXXr}7*@(AJ#QA;2ykjnGXSY_IIrgf+KWjRJ{y>^r zjHi$FjWbWb&wK|hX-SN|_4}P-jJEq$8YB1b)b}1My3+Vj7yUcB{OEV;2QtaMBWp}O zu+3}OBNtuniTsWF^~O%o3*4>*qXvABvko$WcW_7M1h${qJ2Qe@B@`3(< z3vkV0KjZr8GhW@=1=w4^aS3nkT^-+!3^;WWugN(}U|UDMo_^=-=tTq|e|$$Et2QuC z`5?9eb9{F`0sDPNdKUB%n=?-T3fE%xNaO>1?86wrETcC3@9*=wV0L z5wn-IM?d5NuJ5+0|26MwOk#IxG^Ba~=3SiMpB3Brha@+MDRB1f=y}2zxz8X@zwe@X z#w%-}1MgYzeF4s4?zP*e{A=F#jx;`VeqVItRm;8|1Q^eaSobsDo!hUu>}wC}+>v94 za1GmC*gLTOjQQ-WrO#umzC%7BL{9lx?z~6JbIh~few*@P4D|wglf%EmX+O2?g>9XV z&pPhGSpDDeJ&^JZc+BxTNx!k)&HOvEB>^#Gn}*3&Da6< z@FA_!;LUC5fgBvVb%H#5$h#|jKf!&R)w}|J&aD}f>qz%D?hd8wIp3=v)U$2@6+}*H zjtJ1ybLt22zSkFD@Oi$&*~5J-z&ksS_Mr}}<2|0hIeiA^yE|hORAAgwj#Fdy?%4V} zSkXlT_Zu=T2Ks4Vd5awR+KEq?W1PBk-(OQ7KW~_| zz<(#NCF+S0i_1G7pa5qm?~_x0x?uEH#U&!!Z_{wZN`rrbBWvZb1tsOd}fKPg1cPzS>J!n zUy|8?z0C3N6Fdt&^m!uZ^VODbG4l^}hPJb;z}Ubykh7=y9p0P-+7Du#;|{dV*VbQ={&xiS3h;QH z`$am#h>2g%p`gnfD?FcwH7=;tkTuVpz`FzgycqgH$G5;4WBvzvLHfMpci?;p3UFrk zqdnfon9nYe4fqaJ6a>fl^9{XVooTtZ}#qWmoHD_7S{hV26d7Gcxe?V$S&Ylf@ zCq{WE&-cq62DbZhUXR?EA?KbR*yi1Vb32#))V)InI!E2OLq2Ao5r@wihK}s|kmk#Y zQfK{`BiH%(0*0OX5!$U#XEu>v?av9XWa+ zty|%)9Vj2f+WP~TyJ+9$emb_bcF<1f|1Nhczh*?YUnAaL<`rTJ(pvhQ{edjt0N=sC zyv?4+YxYaxja#wr_=P@a@aML^QS+2oqu+ur^6ZQBw3z?Rl=1fbnTs9ND=~Tg`E%di zv*)=Za;)L;&SKO_{OHbT-;S(s9%aM{cux&R2EIQnfBCN3_Oq{dElSXdPk3`jUtHEM z*qu22;(;EgZvBni4et63tmw|Dw4QKof98Az|K)w&@rw6;e(xP1K%=(s*xy?AxMs4h zu{-Ga)$LJ0fGT*$R`>$s#`h}b1KZzmt(WlLBh-;TYyH||2K?4C-yC^$kA6Q-2e^X| zkig$J{(KPt4&aQx!5QN{@=jk!)nitsKx5($LeW0&$*gH~x#?1SH z?$5~v?()PKul)i1&ee7vd;b2$?*ri5Ywq-~d-{NL2IB^gU9ja+y?B8EDj4ZP3TS_z!Tdgj!aJM@kf&KP60SKw#45M$iP#rAIM zo~r$t@8Y%34bJ&LK*p?VE$i?4z^IAs9^Fwvw^txr%-BHg`2DlQ1Nj8@z24syd*l*l z{&;_G^Sk`Nl4C95Jv|4~nl*pz)mr-PGjz;`Z7pLvu?xO}G_S^t`ybT*E{`3=i$uR- ztr$KKbwmVph7cb)_raZh96e(nU*J&!xV z8(WbHthZVhSjTykiM)dV1zUXwAKD4^ntM{V({@kl_VMof>2v?iEN_hs6u7(oIOieP zcRa{9SKdAcy0XO7tMZ1O^A3Uii5U5Q;*P)Lw~nz5U4KD$-il1@@xI{O;1X=_^E-*i zQGevYn`g~&kmKEo+4npC6?nG+_|@2p=UxYPB*3RENOPReeLTPddSt=^Pc5ldDMm1@3X1b4ASEG14SDo`udI@(>BKV2HZ&%=nteZmsB^dA`>)Vyge_@=jn0J)^kVxd3@A$ zPWUU=oK8$9#<-#m++T}nTf<$L;~eHc@Vj4q#&zJ#>f`wmU(i=#cKGWVXHG8h8GGQj zR-hN+c64`P4QFtcYu@kp3${ItU$HxIj;jw5BYyaZ`M@8>D6tQuGZdu#l?7?9?QMP* zHn{QZk^cSJ9bW+FikztE#(f6k-NWwq?33vJyWs;Fz@HWVnXJ*rSaZu+$Jut^XS3tK zBaL&es(gW*y|f$p1K$FSwf_!AU2B7D-WA?^4Ig?ul`T$>pq1T=e+Wr z$9@|a{bJS_bMma@6O3nn=6VlfyZm+UR~h%QuDdbMdo9SJQ)|^nja47Evv{7^5pXA8 z6B6C{F)wz4F#}vfA17w70%y;Tt#30A-98ttejUGai9jFo5_k2JH(z-_^Y7TsQi)eS zfZrKOjQJk-V6PU#ob#eP&-I*+Gqw{~f&DvpsDm9?Lth8h3Q&mg)Hp#e9sP+cBIiAfzw)hZ%mbN7bIR#A&KY}7pZ5|gG4=Gh2W#DtJFthJ z-wi#G2e9UKKVuqr0%QFATT6Ja{Waqg-P|hhtw_013u$bJFTj|<{&t&l6l~`XkofIm z+_)RMdFGFUUE`dC9S8dGz2+=EN7Q(C<*o4rc%OnDpn|7dOx_sLu${#i>l&|Zy!M7J z+G%&BdTImfCHQokckr{}{^b3Aw&a|9#G-dl0zc2=PvWi`9Zq=cmpOm`G0vQT`vk0Q zoP2@*3HUES+xTmS3ikjR|GC$-`wH(f@j1A!AlDidq%zKDjK02RUt|1Sjn5vsxsA95 zSwKB);{slcXYo4UbL^EE{RRE{41d5CIPY4K9&=kxY3)BigDY?I|4*-V6TdM(pErJM zbhv^nNT0Q}g8YJX)`~R8xB!W5>`v^(H+kT`<^PYjIfHlXF>CDBqRxTzL` zpGfae^HF!8I$iwd+L|7_cwHB9Pfp3z@DBJ-xFNNchveDvLXxk zxSjionWwM8-{Hp`@S{Gxl=`h}Z~G*%Vn>%7vz~f|x0X59D(J>8=vN)>5`eSA|MMFP zIBs))&SBl2^LuH<{toP?Ze8cGo^h49Lp{I^KeaxP*70-l1RdW4`3d}vUuo~_PP;ZJ z2|TAhX3fEjKK8etamu@KV0q128K)m^10TQ{jC+7TK!5%Eyj!P`*!x@ut@jTkREg7 z5*f%K#ygwieDWP=Z?BnW?x+Db?m%2a_kJN4bARrwkTXt9pLzPP7__ki^E2j*D|zao z0efuLLJzP~(|P=EyMsix*SKGqKJx<3+=`xH%!B*^_vJP3Wp1E9PW6`49x{l-cg2>|ws%J#yD|rCr!>Z1 zD>0)sd}9BJ415*-(1%`-*7)sh-Y?;dv6k~h{+fa4F-OPmJm2wqNB^GU;}33Tgwe_`A>IiIKYKfTTSnQIMa z8UJ0x--(a8Bj>2_P5*1|rok2b*5BdXlevLy>;dX)`g@{-0`!~bz1+=?6b<~29k=;> zlb8k)`vX{O0V^=qW2|x3Z$Q7t-PrGwc1#5Fz&6g$=M$LcbGP=0LzriMInje2a3X*^ zTjA9Uvc;@%bUm`2;IG1yFZ@ ztTT#L`I;VkCU{^wUn5T48tzDL^d;OH2?EF)<6PPm*@5+bp1_^cZ{C9MPL9&qoOQ$Y zxm+i5K9e5gp}X(P?>uq^?BvQ@H_rU)zUEeub4K+7My|2w<9)NXhg#0$=`nYpG_C=A z`JQM8$ej03TE{#2`SJ7QXU%zjfG1dQ)IbiN+&i}Y`k8C&3as<>l=gTS1OD}GzC#G< zvzrqF-CS!vfN}Z~D7`}oaO(a$x{9tHr@VIc*O~PDe)#;{4!&Z3?<_Z`RrykApNQ`xm+?(~U1-ZSg~#&=NB?K{42 zn76`h#M;AI@A!;g&KPIz*xye(iEr5AE8P7x>o)imUD@Eh!)K7ht=jO`mzPt2ypZt2 z2N%HE?7ahj2i%FVrkpb;`WLuDtkN9!(2<@RfV=3ou08A@==L@)k=EbjUvmcQ{;AGf z75#zFJkiK$$RA*wocY$eo{haK_K1DW|JKIY^*ejVpZE(HeX)Ii>>23hJInqytL}4oO`r~d7ps#{Wa?r;yv4G zo0lU00&MFp#N5f<%+j&F*{$2ccN#Vrv4WpAe8ct`=sQ5bjXFs4jjgXaW1{Qdu>-WzHl_n- z7oULr%Wc+3^bX8*SH6QA81KIvK7UVk$If7kpY<=G;M-1KJ%RoQobwrHZ|mtxbo(^) z9bbVj*gi+!ZMlvwkYecg11`Y|w}I<9n(LfOeLsM|zbDeYIivrc%Y5?|P}N^k|3JFG z1?)in5Vv`cM%{`YpuyY6I>uPXIZ_F?o_r9~!2?@Y(BGc^0H+ zyz4pCQQMfnb{2gjANT#ntoo3DAkDXTN7^ruLj9UoSp=N<>)Wi^-~u!_pRdo=<6PCa zi+jzQk;otbyP~_R1^oaX^Q>u}u=a+00B3s20Z$L&!S~m^hk2tew(p(q*AB)k?g@Ro ze>r=bTh!T&K6^Ny@yd4U_7MR#V*D)R$u(rn8K20aP5!8fZXNyB@7UKFdBpXE0oM%c z3p)U3)Je=h2OX@`@s8>RSU?wa zOZ+|)^UX=%eXLc98@QkNxF`jckDpJUxExRNiN9AN4Uey2s}s@At+6?B_0w z(dRr3t|3dzXJOAozwC;?0B1Ho@ab22%Xk8@PuyWkAh@%|$p zX-(~p%u~i~^42-f?Qe{+9&7SLcqL-zj(WPhjN0t>}-FbKY{BIp!vQVK4bR zu=WO4Q1I25`hwj+yne;#cV8X(b1veotzAIN^LJZhfN=sIjCqtkubg#*9$ z#7gI zNc(G-1RQJmj*s^fy`X2zI>r`s?{1HVuC3pFn%mJm5%YUz_)~@xGsn1w?sIfMm0W$! z;GBb_zB6_>d1Zp_^c%B+3NmKDifqKlIjf&*WrZ`>dis6x2YQ&K^(jD z?&|;eKJN!JYW)-TzW{$Hdy?yk_;vROT*XeHEII2nCHE>Z`Gq)RzR*XXd-u4@pLYJ* z!Kf9?iT?=#UjY?(*8})$JaP$qj`4_jhb$%2Thy+xGx>V7zq-FyA`* zo17;Bc1N0{z1^sVv`?TLzXNU0&$;nM%vy=`Wd5?PDHd~iFN`tQ+ACZ`Hfj`P0(~3& z6ZrS7@+~H;;hg%-+i&w;A`JJ?7`P<*eb)s6@YZX`DIcijABRi!{#nTr^l* zNaO8)fC4_ia+{wY>joIL*nbJ&!9RGop47m$_tY=yzrk;R;WMdwB9P{2XP|SeeLuM|f6iG$4xjf}fw|tZf}-TE*G}#I?KSH-kM+mh zk=OT*G1e(?%DXxU_yVk_ufOKcspIk?t2(@XJb#%;vw z^L~lXcU;@qDz>rXzRm^b&n#yW&TgFX*6GFrk8$ee2Jmz9`P55Jzxf3iQ{lP@;*slX zatBi0XX9=fGS!)Dx&4~8a30vk2PnY3nDay?H5zgS4gEkqz;Ey8+|)D1x_{i@@EI#t zPTo74(k1{Sgl@=TQc5E@!CdLo&a;oW&h(Ah4}- z2i|Xg&F3cPOrAp@b*$%10X`cCoX7dBYmD`bx4)dR>uG;O`Ya2OGhW>q+RoI_&1pbs zjRYN(7aTZ<(^r5pPrc%M;?utY@36h*&t~&?Y<(W>mXphi$F`SqxCc+qS>GN1-|zG9 zbF5=s^Pk4OF$X?-I-{q?%ynM(T9X*KA&Cp1o?tNtc_2H|oa?kWIi&gHoN*82mm781 zW5<`kIu$(Nb*KMuNiMG2Jd$)3h1 zuvP)~&_8ORYP;uxbZ5R#9eqQ3oV6tiu-1n>dcREsoO2vt#dpnCi5d9+-CMu&IE(St z8of?^ziQxe{?JeC0-pL}_OOoi?XB)?=329(2RP90_{|+0@fH39`+6P)TW*6({N`M> z?cIPe4P1RXdT8gAL5{gWje<130(Ixn4s@aIoCP%OJJPzoKcc*s^8bPd*nqW5%-AHp zf<;cCdm`o?gLvy#Y;&x;-)63P<~M;a!1_LCF)pzWaz{UM*rr&M|YkJ zh4=vLZGI=Z>mR_~$c>sv-%n?;?gM=TzI$@c`6s%y%lq{GK(E@sXSD!z@8{1TIi-Kk zdq?_Cx0v(W^Kts^r(d7*D95`{<}B*N&e>Nz&^w5j^Hucm{rQ@lws8&V9!5T~9`oeA zpZU(LT#?_wemi+&h3`n;1O3KY%bM14cK71n2Ya6d9KiTa?uN9sy%Jf;vBn*~plj=I z*m76Ad$3Q%*I%>W11^jQ9sY@(Kt1J=&hyt}of6Y$-t+Y%hHe8?&Hy{QuYclS>&pSD5p^m-uugC+e@3U8j3t(Ss`T6zPM9etv z)bN{I(S@Jc4!#593%(w+_m~^Xo>jnm+(8BkocpOcd-!>_x1VqQ>VdS5z3lC2@hd-a z{$1^zSaS;4(T!0$pLrelJXYYbwr~c|0~xP5yLBosU%$4{z60wWNawQum=CG$JKAIR zD;{j~GH2f4{0FvZuj%vdW44&R7wjiK{nl;p#*Y|OdszGG{e`$oH$H*!?$`c5=MCyz zdCYH&u>oI*^%*|#Z{R@JAIhK(@)H;{VC(<)_xV{=w^yG06|=xO*Esm>{cq&h)7T%_ z&bEU4izs+=I=%v&UEBCk>+kUB!}pHB=lkUS%Nc8}7IR*EYcIgN88c4IJ9^I@zGJt< zgB{3<{RQ3U;=lLv+|ix4yk<<#*<1S|k3DLjR`iIu8*3+a2cGeqi1+V$D!F%{f7}JV zz#mBMpT3xPYDnw33v-&luE+wc-A~zwsn&@ZyP}&{W}X<2(Bu7~A2* z{Wbl5PQ3qu&v_T@g50q$UD>e5K_B-bWt}*}?RiHgX!>t+XB*N!8FTg^ZscP( zbZaW@Yn-s2pHX#V6MrM_s%3m&>-TQj$}k7LfKdzESr&1F7aV6X)|%#x&*9&u^bSus zdMmd+f;NggM5F-`?l54LMMW@x6M0iod|=GuK)@XG|c4_ipMje^*X) z_hK!jws95on7X$2E!Y((hY#s{^$jS^DZt(GuOKto1g(} z`0qA4zJl#BcZXk*BR?Te`2OAT)tK{`XPo#1tiS1}_jlw97Ldlng>_!@yV;uVtDnBa z_O8~xj`a>Xa;;H7+@?3szvI(Zcdp^n4{wg~o*E$MyT0K212pVV0(*Q0*75Y1K4<*> zea^ez=6&q*4H(<;7v#7b(s=8dUx^X^%<&!at`A_ZB<3r!fE`q9^W>H0xodp^eE#2o zJ2vmCVZ8hgV67VD1JBP|=kB%}wZLj@%zvMiNROPki+*74*dO|5P~K+D{Y5H1d#R_k z60U=a{@eSUVbup`kDvP~aJShc5x#fd;sG=!zD2dXgmLMa?W)}kN5NaCFTL=EY>xyA_KU)Cw8Li zZ^#0CudUMq8)*-DkF#XV_?|OXUsE^s^pBZi&Y~~Cfv=xZ+xpk;S9}YoCudFt`-?d6 zzuxD2<-2O0u>riRu>n_*-s28$-U6J*9!)=b5YvJBfjy*B*KTx7+*_I5}2GwyLF`7wt%@Xq@u@EB_^YuLMjOMk+f zSLFXbYhjH!n>jByaG=}6Jm(vAPTPF%mB6E~B=N{QTtjc@=8tDXzJCuUZ=7fF$evm= z_R2H2pUC7c*2GRwfi*U22D)?D(>l&G?gVe%)%%~{=T5FX_w1cIadH7XJ^4of>TE#Y zsE4$d@ghMX|Jt#fyLV>|+=2e}H9y;dYtoZFlRj1&4gvLdY!D1kA{ZNiwux4h5VS3Z%W$3m?B8rb9vo&8yB2iCK;FlN+47QuPqmv0G$ zyxz+$2duZ<=FW_(=;NTf5B)1O%r!2N75hN<4tMYXV>Wzp_IvEww%3% zbw{&6Af;hnD^8ddDi583hckmX3s*- zfwYgY)fo0}?Xkncd8a45{uJmtIod&f!G35Xn;fXXdlu~g=vO^`?#;e(epejYCo1tz zeMnL_1o;x@jbOcLtm8Ua96ns#`}(p)32`^L*H_n_c71iUh(=9SSu4=9k~%N zuFp_;!>8X`#8{(a7vS9zX|DK&&(E!~iQVOM-dX7^o^dyHdpy7btmTgO(~g*Zw%4rV zOfBb*&ADuMP>8jkIp$ybg6?sD`XVvltkbcJz;<8$w^#1Hov{_4GgV+6w;>$lEVc<1xkX;(0OnC91*cU20x z0~kAg2Y|a|LpMKw$5~3uoFUPz@l$7R;~y_L(9ts{cI*n=hdK7_NNao7MCwy-=+33j z{XBtp)HmKKc#qG(Sf%+JG1gv2t; zq04)EPJN^APECEr6r{HPhTf223{ba!!q<}B-z5zM z_z!CMcgNPWegoGGNp4_QWIuKNtNGaS3AQ)-!riG+u+48sbH|*>4!nma6CbhWilQyh zJGmQhN7@Y(Flr{cdB`f<<6mN}abP>&lQ`eaEB*`mXW)#^)XfLx82j7%{QR41zp*1^ z;6S4 z@4jote7=E{vsTMltD&bcK>fOhHNL-y{o8!kln-Oz16X6n@2T)LXRf{y0BJ2})K2UM z-0KD^`uO`Rb!uBrS+E1$Gwhsp3kZD24T(PD<*=>oZ1M@z^{aOfIq`hKgHOO(#*F>k z=JPSux!h4h-{2SGR-|0VSK#HwZ0Oo0u}^8Nxy}&C1m5>w-)EjRI-KzA`24vfU%^#N z|NlcygR|xWlmSKzytymTXN^CBTq8H|8z&+rr{8;<({A&*Hgt20`GL$U9=)NDXT#k` z?c3}h^2CdUxQ_fls#~Y#tU3JTC_V1enj;^1*?0H{NVsv(3pl_J(32Xk=`!E>?PGjH zUeCrjeF*&kve7|^Ctz}Q8H41XY zzk>}JUqImd0{nB#fwXSI-*3NWsMy}u{D$9WG2(K@H25IKXXV_+YMYmEf!)#Vn?fJb z8XH)FvFi2I3uo+ryR0|fA>dtnjulyu-YJxE_Qv+u>-zpP#^Y@N?VbPoLcn@8=Dt4A zox^(eUe)ml`#94BEVp^DaR+>XU4VRIH)5T|d~0`jYgl{CjJ<>L&T=Pz-pCtcUPoT< zid@ebZ@t8p8}X{>1JtmOcRh$Jz<6h|mi>x4FkXIz3#4a1s9OnIGeG5fCu)N z8GiteobRP|H>CN-*sFnZJE5NQE}!A7ZQly7{Q&kbuLGqpry;GK0(Aepy>LOb3F5FPyTF!YZ()mhEd)$E->xF>t$h$e{4Ly8&Pv4?~+fB49GpjV`M6<=Tnb_3QNeQ@95 z?PEXBfzKYJC${&re#GQj&K^nJ6BPWSqW|}|c{h8FxyiG)Xh5zb69l^7kJtCTbpw?5 zIlG_L6$G(Y|AOD|EbIS&RJ<+J9LbWV>7c-X1_>H87!aVrfCUK#4Cr&pQfI(|0SN{y z7!Y8=f&~c{E^t^d;e`$b6N+6+A;Av{IusNvxX)uAQL6QGU)kSle;$$fsidmg^ka*& z{}X;Ya_r$u&Ugg_Ti;Lg?GySNF30_id#rWx?Cpu<7dRJedDhep(1FkJ+#%s!@twzh z<~zUn9eXGKQx5)$A?>*nZ=iZ$cl>hi!1x4-Prtr~^wj4g*Sl~1f&QY-9k_#iy-V}9 zccba@*PMsn8W&=6Jpb#{>R0a-u?Rr?GkBPTJ%I9xZCqW{T*xQfdD`fa3k`jsJBxC_ zan_E#1LgJW+ELC+A2Ymei|ZYk_Y+yq5#2m}K8J>sKZx(Z84}bk#F{#BeG@yxk$*=T ztM?86DFf~YwE}pTNzDDe0edT*Q{VNr=B&Ep4RGG0J*_uI8z|aIrq_b zr@qpAn8tobM?Ub`zX9X^oVo*hD2sO^2Df&^C-3AA{4aS28hHtv;eyZi-1v5D1XiB+ z{qLW!W4&h+=^YK?1HA)dJ&!$z8TXEieL?=~C*}su_>?=k195B~MN zWbH;Q0-KneBxc$*d@7RXSU9!AHD7Xzw^saxEJI^-jM?| zQ2XVwE7u%20jw7XZ@kwo<%cuLv4=HwQ2SzA&mH|c;3syX2hyMS{|k76=^?hlf2)PBYQ9%~$Tcp=wVv~;4`9B>+RoO3*^t8eb@18KbAPYd&6Tqq9P7&Q zy?ueF{!jdQDEQByy#YRX-i3Ccix$8W)L>5UL3DA<-E!&H{{}i-=ew%|X9;}PtC^4w zPCw~Ke-={Hr z+Rh%pXKRjq?SDsiKKnV7`XI-=2fFndFsA$ozJQ5s&*J`0%o8i$8piF{!47;r9_v_J zUm%C}ZSDO)eg?*z$J%m24p4vMZ@}L686bfuj*xgGcpH{w_1JoLo& zm@DUgBMzL|?}~a3=t}Dh^cV6!e_i)tp3hxr+~dEmul?JnHP2boorE<+&$Wx6DeIdf z*YiNy-*-)E&qiH)xMR(ITlXJ9?C@QQ8w=24-F4kFdMBn|egbu4-|-J{-g)!H>eFuP zz8=KS_skyFa_{PgylCH6ZDZ%Nok<;Pbr8fGF>Pfa)$O|-F>~$n#Qp}%8^Bqe^NLT~ zW4(yQvCp5x8@BZZ@GiaArhWL$>FAmGfpZo|{k@Q}_PLmc&)z%ofqmYkee@53UOnKP ze(z!ulW!fl2{v@O=P`c<7Lcp%9CBVreOGjO&e_o${_~yd_>Fo0>fex_NPYp&NdEuZy>HE`FTORqaa_xCr(YgS|& z@UHB6J_q~RSDv+u-4CDs%Db)Et^G9qb@hCp4>+AX^*~Q_b2_%~))o94{DIH&ArJIC zdLre-ZPij*$31t@-oy`|F=wb9W7YId%Ex`U(^;F$x)b|UZ{!5lGsl{>5ArwWb5_4g z{(~2u6bXuGuDg)zje*Iqfh+S+OQjbdl;|T0vGyO%NpvHgD%HYT>ZK3*!wU1sPTcbu`2-^{;0WjOzU#h_chGWu-mAZLW5vyt|3AL2TKWr$+uHR9 zad&a1%cu2wF>7CmJFn93>VY0$g7aQ3n{e4)j4xp3?aPq<}MwBdsrI10A1t{RRCS*ukIr z;k}U7cz}tmZoHABOyuRZV*0$pfvo#S`n-+n3$UZh)AlHx!TLVWs)f(qPq;o0eHUVZ z&;AYRvECIsbM4--C$RU!9AMqV_d*}=KaidttH#;G z`Q*xTx7rVB{lM@3S5STMb@YcedK2iGYtG{NUSj(ktXp-WL0a#JcfOHho^Xy%Y{a^+ zK(|*S%~f|Vd)?umzB@ki3it>51!+y6LmllnfUsBbh+AiZ1p0kPFcBFkStys$m?mFwB? zY0Djn0~oWOu?Fm^)Yk&smM(t+{fXcDfi2`2^Nd`77OY*p;g2A|alRl=j%U-3Rm0D{ zb)8K)!4tjW=$_0~Q-1>M*yE0FZNCTPcn9VU5ZK;FN1vc!PwYJOjT||}J$&A^JZmb2 zvv#_$8fvU}r8_ z_eGuW)Nns*S@Vvxe)WW_zX5v)?>B{Zv~`e?tB$$q?mO|@w;@I4F|IAU2Q*8t8f4F@A0D(f9_JcPM*I{4dVJ5QfPasW}?KZ;r#aa0LD6gZTZ%d=Vz>*%X{sc_%?jX z>W{Rqwd7ym)j@WBz&rC42kEo$&aAT`d#?AZ$I8FsKl=pweqa#SFRXj&el7yoTWL){ zC)eBR8St(71lDTU9mrF^;M)#NWBIGU_hPSsKfqb@>~FsJVDChqxq4Z%V|&!?J7Uey z(Ibr@4}bBK9Q$$1^&PgK=kZ>#noT{+T{BtByr~q-b3V^I`T7Pq-^~N^{dWPHHhKbm zWyg{Od%G>aaq9)Ks;#XK{H!#jx$WTVpSkX1*N1I=rF;798+KyL9pH)IS@jF|)h-Ub zzhO^fNaM!NS+3}wp8OK)dotkJ$M@g`oHK}ZFo3+sC5Q5F@v}$834Ok;=XXJxTYb4> z`)=fU>b~){T>4ak0d_db1lBgcp?@6uSkgK)JzDk6d>#;acCC4RrMT;j^AR>tFCq^gwo`r{?{A@$D%mbJZK#w>5_~%r)jo zVwFqX2fAor1TkvV(a*ecYG}SZ{)w1KD!TV7(+)=|0Y3UI33dGf{xOoZ*bi@lGE= zUH>^}zpYw$s>X(G{og(v`@@}}-mQQ{T#j?tBatGMxw!JJuPwjFnn{kiKZ!k&#wNaY z_}zJ+@5bIM?}6^SE3e}-?z1#s8HcV6V&{8#!M2_-_l5LCt{NGut~}$;^xIc_>@LUt zzMC(gE-IJUj=q5rtLMPyjN5JPX}yN;0>^s3H;tIQjD?OA0msY!iM51d@%_9`3BhHl;ONdKJOnG)N%uhgQ&GN|E^6QJRDhlYMdueg4Er~jLAFk;=` z25*24PhY5WkKRwmwr0H>#H}mO8HDi{au@G)U$EBuL5&{#=*w_3Y zV1U3*eYe%ezH-_L)RF7mw#FaiSnm$h{Y+nPtHuL8KnL!9@||U1+vB`n@5~xc;OFqS zud7CnwM*4Of85r0v9g26_3v*qK5)ya|lU9q51H z*O$n87R1zjmhR(R&X92I=^QmHa>6M))~jiaMc;e*JAQlW@45OA`RhHa z`f#lywgKn*nc;2grGsvlX;~w-+d_R%?o!Evn=Qs4J4t8K|``dGp zZ;iz60-y4Q40U5bPRsQ@|G+=cAGa0%z?N_PiccA_W;C~+J?^SU?6VyGppj=icd74? zuj?J`xq1ZBdc{Q@>cF#+=Pn*UpK_gP1P*ljPIY*RR5yNdyc=^G{yTo_3Uhywa{-=_ zi)RmWt$QbC%>g`0XW7A(n8*4ZI}$FiZt?N;IDpTZ7k$9EFm8_DPtGB{KkXp)Kn~*V z2A4h?vN*^_+`H88UgkWtpL3S)@HauHRhmCefYCz3!j`Gsk9#uD_ywNWK*JtjM<2K2Sz-sUr~V7FBkdz+1IFszLw*DP9PZ>8Q;vkgsoX$$ zQ)8vxt ze}m#b12=QSZOQ+Dc~9^su-676(22?O{*=xTplfsPZ+^-3exLSc z41d)PW&n}+z>eP?2YL@&;`EbuzRUGlKX~#l#CFgG_6BM`-W_XN-#&8f@3Xj~*Gb~& z#yfiT_U^C;uvXor{9LuP6FrGPjyl#b-*?VAy$fqJ{PU23w7zhTKsw7E$k*RN16OSG z#v2~GJwK4vxMHEyRgR4PJpg|qCb(wE=YCr2lfQ|ojugE)t#9?K!{j1CLH;~ney;EYvmob zJ*;n@ybF2Cj;-(V6^ZW(N8gVA4S4K1V)>k@4z@L&rGbeZz@G*7l`qsAGLhbi9QzJz zPwhxf*~UTD#pkIS|6IB=-GMpQP`CaB0TTG#V84d$F4i+H9>nxzkcS@lJQwxA2h6p8 zgQu^d*F4z9MBtNS+&Qc{k{GAwF$1#4%GFoBPKn=lM8AC!xglS`S(3PQ&zQ6NeK+Ch zm+M^mCtT;VhJSbL{-;lVJ&%7LdBh$EcaT>xWCM}M-ssl1#=y6eqqJ8dKalozXJa8^ z>CQgkG_W0d64U<~)ZB!vt#7|!p!ZyP)~kGcPhx@JedW4;^&=LX}Gfc^(?wyK4H9KPopK4{qO)0(03 z;0|~JI+)-E&l>6t1UdG#pSxYaPwXqwdL!5WH=6M?=O?izsM(1R>nOo@{2i(NBH#Ou zSoKqR*bQBnW8D3`^PRYTwUd~&Ch-flHT5O7zK2{8VBl}~`^x$1mD zzkkJ!mA_-l4{T?cxpw@wPsg0t_IeWA!5v(&{}VOLt={;4>Ia)X@IbewcIvYZw)q3S zq3avQ(Kr0oJvo7`E-&z#Q=e08YnaoK1K6u}cw^(gVt+$_f{q>LpbKlrnc3=o?CXs7 zkh?2^9FMvC4JD5ldey-8{dmaNhkij0?3$H4bA4Z`7yQ?u%k^Ba?cYGgs$addbJeW( z4SRdT!{_O-_Oq_Lw@(0{xm6ooj&&P(&gIOBPeiWStampCtaXKJoq=vGYs>#@H*{yd z;@?2WR({|M^as*82R@~__UOQK$LATba-B(DfREeJQ%>?fVD0LIv{uzdn&1M6r{u%QRib49nVKI6_hkTceIoO7rL@R;w84gQD~Ysm8*+K>E%CpUmKp71LV zTUj&zbLE?R?ql9CFOHeW^XTib{#j7{?5%9rNe|`4J^`qi$Z`1f6&Kr`^*cw9BRrY- z!2~VV^VR36`N%0dSAA=2Ah6FIx!&aizjx#u_L;Gyd)wbU=P{-<|D0#235sG z?!des-~s}8?63X->@7T*tB(CTF>Pz~!#@L^`ON*x?>xl6kn$iXl;y}CMN7pas^gF9H+#!i|Fb=6b^SFDYy)Q?u^}J7g z>H#jm9M9+Xl74<#P!oARCwoP%-761n)q!WNE}yu0o``i1-B{xI8wN4!$yHZcv!3 zISTywoW#yq&b#RFd^XA;-#hi^xpnnV{2l#;eg&Q#8S04yWyh*LZfhTF4SmRj^FU6d zadm4MYhVZ4A>DPrt9w9iv7Vt)o_QBg9Nvq-|AD`LFNfa~*e{UaUq7uskI?=N-C4X7 zWkXuayExx_XR~Gla%v8wzB{prydaGW{r%9(CT8Bp8$U9M1=8I5nSdPl?Cs9xPU0T* z03P`bSZjb8*vKHyJ2Ix8_yd1KzOXxZ>c?kn*N1dRdG0f?<@=mB@Brp|tSR4oLC_c$K{{Vg;oHJU(BgeXd%vg8u!1fM8`-thY{%>E``{tYt+q?mG;u|?HptNoWcl4$n zYZs-ta<<#*0b@N=>?hcXSyQgklX2YnA>G578u0Gy>(7qjACi^mNo2sAy)T5 z%o!&BiGD@*9deh3{Rh&Y6~@dr=gf^a9DL?p;0&b4?+J73Al=3J+}BwH81LALZhnAv zz+MTu(0A(ew&w)K%&BL9{RBRj?bF)FUe>+?@5q0?_&4&llR5+Y{yui&emXvD zG(oHd0vXubZQ1$*e@Du9mMif%Y-9INho78*?L8Uyvr)6)Gf$sr*riivYI_n6$XGQx zx#~}3M>?l-$#;hxzcX6@oGFRfTaLB%_v+(2Vs79&XE)#HJn(sM)=Km6sdwOibNqa- z-YxLC1JXZ-_U{Y>jyn#4|KvW%x$8UXnbQQmM$9=bNc)K^`Wtj^*K&>*7^?oTzAlva5`M?=6}JRM{XiFd;@8op9$yjm=kft^rP!@2c`Cf z9Qo#-cjF!p@CPX0?bl59ns6UT>sZ6xl*WX$XVM!z@LSJ#I}o{i_BP(d@AQCQytZuj zK94+W%Cm?1M%+2f_g)6RhHiiF;SQSM$Tg=p*w@20<{b7GFJL|MwomA}dR$IFF?rV7 z->Z+hymrWlHBTaUxbpQGS9gbQjQokTb`qPwobnOtU;y(xz7u}WX;)6HT7mqAzvEZ; zew2Ys;P;~My6^c1_zk`8@ePNZ4e1?N_d9WOtuLn`)z5yDnEZwud2p26dGYa2Z8+X> zM|Ylz6Wh^K4(Px-SA4!}H3Kf^2<$s}89N|9(Dk+Vid$Eo(s|4=KLZH{c;Pe0K2KoG zJ^jwGr}drvfi5R<@${MZ19*3xJbm`Ft~}3-^{$+J`}{0FH0nNk1cu+4p<50UHM z?DZMgPguJoarbKY+&yF6uX@KH*vgK5hqHkIo*kcga{T#Lv*haMKK<|qzALZ##5r+TWcnU=lah4xaK0cJ+R*S}m#lH$CB0 z4!-JnW26nqKAGRnizNvoryqmx$>S&Xr?HqBy-dFUF-@hj?e;f#; z-*3)oUV;hSx#mFDUBY|9bEb)|?4O{qKH;~5dm>2OxC<3J2JpTmuuYE6Z<2MzrQQRcwLByEMdVRK6T5>1<8QkA zyPI5Rb~ia4yCHpF{EQg$a}oIb^BeQz`b-8spQm&9PS&gm$CJNi-H0b~`^lfgwDos= z?!hxDDl2x?kSe@S)Fi3Prz z5qZwt(G$2!sR9F8$M_JKd=MN1lCNXu!eIEb$3SZ=+%#yvXSphfqqB6;8;Vh zJ*{sod3E1NkI$+nF>LE32y|^f7ZEEi$KHW-{|=Pxh@(|cdCvL}@0B|bpWFe=bHkx0`4sCSf4OqJ)18l@k z{*EpNetDVea}Xy7J8vt`&*an(&oguJoz-V`-u=MuU0B0oFMD|8Y;fgR$G(2%^gock zM^~_e2k>~G{+Yi0-F+afeF4_$@wXk41Yiwg=6Yud=Y?#CZk<#2`8bbxyKyk^wOG%| zzH;OTa4zMU)6wrhp1vzGkQ>rGcMEFC@o3xYg00W_2S|K7wp{1Fzt z3C{a8UME()KjFsL!D$N(kHKWydM^Nf#Jap$qFarf!!L7r8 zIqkM~GZyG_J?8q^lHcN(4c`;ndw9U@*ybsn!o_eeL?D%>xtk!!`AP!IPbt2 z>|yTXM*mzK?bqS6hczeMhEKU+w_M+O)~G|Q;kVuc-5%DHH}Kp4^f{ZeYU{V3CvyFH zi?P4=)Bew2*FMe?A$h$c>NBY&=)DuSuDd$V1w6qwY(HxWKj9i5=yJSE>saH7-+2a5 z{|P)-WPMiY;W4-F?uI1h3=^NchE%tYHLb7hxfloLI;S!N9~@_QUh~TbZ=e@1@dLka z9>5t-uDzewGmd)L#>~AVw*&g*I>Yr-`|EiI{`1a$!=G?G`XZ0JLBEn0vHDocy$5m8 z@VkT3os667$y~O5@4)>%&Ss9h9a(omOpbjz@F)i`UNVne4_-5d?@A481%B^G*ei$+ z{1a3Uc2T!a&4Vtt$GWR|ThH3>RrhdC??TQv^iVR7wZ&O`ha+#|8~VWoJCXX$tAnoJ zya)0p{s*Y<%7YTb?9uR>`$YQOtl6>k@4)%2dj;~X^Cw8)=W7SuTG&2cxs4hLHuRbc zy~$(GKsGr@XLP=r6Q47A^jX7RBadf{ZvcIsy6b=j>|+hTcV7qWVIA|Y=pDK11OM}D zqv(oxCmV2Hdx{@$0+^%U8qTZl4y^G6S1^hFe6RQjB&P?z7vj`PTSOJoH8T>&UU5vkd$l z+nvP3*4IEN@ofiBf1vyO7iY51j(!It*ZX4pGclie!go*Wc$eSMKak#8$LCxRbZdRw z*8S)c+7tOa;umadn5W#p$yNUXuS9}%2Z?1ReJjPRA!UGMx5j%a>4t(xx z-8$Giy1PWKde$_5600LOSA0{Cbx-n~rFi(trcQ@zz009HgU_ys^PoG6T;a)Fcj=rT z2iJb5KZslNzkXU$zdq|Y!$dc?A3k#fIHTxrUP|KkH+2R2IBb0l{c%YB&TLKfALzal z`w{nj?t*(zo^zBjte?c!Fx^jR!KIfXT;?~oC!MOtG6?Lz~hVfgY;mWs0oE&vx7x~oa+Q7Z+ zV~@HMeE!*?T$pQL|D=$cEDm!)^r297f&ao4-XT$HF)-hMVJZG4| z_?=kINzA^UNPZE-HI>%i@f3DdkvIBbxdBz7)j_*MSza91O=zgvOYzGFsE4qD+Tkm;WH3Gd(tQ_q? zmz(HA8$94Ut9l34p1@wloa+nH*?#*(oOzO%+yvGf+Q`a>H}N~y(1$Ij;Y*~twVl^^ z10z@N1E0^<_tg9Hu9fE1y8+KS>h}2UD{-KH`WrP8dq)nWv8y$(KY%?pY9AF^@eUJl=9_>EZY7k{g*8^8M- zx32vIS-Kv26RQ@A%AdPq}r_|NZOw zyx4OFNFqr4c~7AZths@jLmPh2g;W{?Nz59~_0)&7M%@)P2Qlk3;9kYWFVEUne7;-O zG{;=g1jqP{U%o%!T!Hm`C#(_ZLp^^X)ahZJ0ptf#ZuxTg22$Q;Twg3X^;28hcj^QD z1o{TP%WdiA-0^$#&G?0uYet`IN4KV@>LN4v#Tfmq`J_(hLwf3tg8pzGAagy}D{=dH zH|LDrMG&u9&^^ZNq15jlaymZi1<-#-?m*uO(jly!L^FTm917*}_`x&w4i=6VoP95ZTaK-}t4qm`~VL#^yHI)RBW8VB|vS+Q8g~ z-(Gsv6B2}f>#%@{Iotv+KlyFIYIvwH84eqTny1MZpTpxeu%-q0_=&xSt_>m5OV z0%tkzx5di0SAYhp5B>nvml^K(C${~rK)*Te+H>XB9iV%@!LiTCb-(61_jzZ2rxUiT-|*9-hsVCiO*-OJr8uG_q73M6V^)npYhqd z4z_vD{x4tg#KEO!;P2Smr!~8Me*)$8ws;Su9MRCrrjNPak7(#Zy(6bG@Bw{KP&|Dc zbKSY&zsps>;RhE`{pc~^U5G~%D-z_4_PF ztR7EvIbT5gy8f=_zfrdtPe8l_dFqM3=|`Th&(KCrr04nmHGc*UupRmCu72(^a?Nlh z$6n6+0M!S362BvzQ{Rr%C(mcBw7$8XiCz07M*P~fWB>M*9C-Gh#1dQ%yCLO__piDg zy#e#?hwgi6e$7b!Ajf&mTFzn3PArge8~E|Y=kvs`t?sNI_w&^E5W4n$$PsH+>m@!p z?!5u$ty;gYKJ{L~Ggn^CC5I#TiqCr1QaASk?04?qelHO42k`tgcVG~!&oH_*?6c!* zxq28c8=3fgKF(ubPsTB8l26RMAYL-ouJdp5tQ+_`_zX9}MsClAM9Q<5J2`_f{WbgP zkF}HYySIJRtrc*Gz&DZo(9IX#QQ|vm|8*z$<@z1u&+jMFy#h4++wI5;e4qn+*Y1rN zwspV3ulcbb=y%ZJ4Q#ms{Xagf-|twnf&Yr_9Og}G$aPLn&6mq3*BTx_8_wp;@*8mf z=WX@dkE+0T&@sZXR=4{5;{~OY{a|N<$p$~Lxy2Ar>?0KN(fZC1SmcA5SwoIHH6`=K;op&RjrG5A^$l>xSI<=mr{Oav zk=G;kLkX;Hzt;h|Uv6ve9cjIUQ(QQ5t=)+|4?K|WDfF9HIn2DFPx9r;v!69Ph&MLp ziS+x{=d+>PYv$sOSoQ2_{X4M+)QwrU>oFPKj0`EaF^i6gkxQOr+i{tvudH=!6xT-a3hCU0(nDQBr$x>rQC@HGK8N4y`>n0Iss1N?8Fj(Dy){clzpe(xfHb+5oX zv(ELlW>Idy{_?ylPtCK`dj320>?Pm)4m|3CZz6^Jo9jC=v(*gtHCA^)fBAA=Nb{e> z?QL!z@}a!JlW!e(C$>-QjIX$GegOZz***MzXvAi$d;5v~Jp6&L6H{LCsar>``s=-( z;RUJxid4TJwtBELbr*1zazZ3F$NUlNebCl#Eq&gp{qoGg*YVv! z^}#mZU9|nobnJ$`p({rs(e114JRSMK9?0i=y&LkZWj||7pf6(i(xF0_6 zTNrcp9hi5yEj{pG(XCVYNb@&+-cv4HzL>xq_j2bM%hu=I58%9mco4g46EoMizCbsx zfeXI#SR-S_%{R6IYc%Yi$Oo|ggzJ2X9QfQz+g%IJbM-yMc*s$9C2i=ts;M-S?`EF7`v38~E$FqT9>U69fvdPsORx z(Ct|s-+rGL!p7??`*-5B%0v7SQ|YBR=%!nrkE0&=a3Ien}^c zesYxqpU1w|JL{Y^U&Ke?NB;J8^)u%KSgRS!#j$tU2&H@Y?$2C#$^_;ayTbQ??6p{Ft;a@+k|Nhr| zZ|31o;(k`F-;_XK0CVgq&)Lms_@8j(5A=a_hAVg-a^mYiP7fp}W7YKj?x14Sx8Ee@ z&KvsqEbXzsfBCFu{g8WL;2ZcJaIJ40VE)p!yNA6KfW&AJFrI01Yfxw zFt5Y+=x^lAM51e3^98PlUUh=FNN_uLKu>MX1+F!BWgcLj;$wfuZ(cns^b5MZiXZE_ z22!|>y{zYbME>HryRpEZN@7(Pz48t_u&rMlut?7sPA0XQyObXYd?Uw z);xR4F>d`GynuEG?Y(-QvAWYEZod)qfRk{Q4f}i-pTvKHEB?T5jUOQ3TH9P0}&U?*f z{ReuBUp*eg&8aj8Y`+P~qqws#^w!3Dp^U24u)bGv5;1A9l#!|&dn*KO4d zWPKiy&MVJ;PoUm_IhQwjE@^Fb?{f#fkNzIj`SksKlNPJ@nJ?e_>PR_*cp~kyfd{yQ zBpyid6YSs$6288gAKiNz+T!>1dttdj{zWPOy`Iezx#5#L@jFB9jC7_ZV&x2c#wW0^ zd~=oB{u%2Xsh;=}xr2Zo*bO-2m-pJy8H`!a({uSZ{9?x9`ws1(W82S}t#v`~f*k{U z^PGC|#2l8^o% z?rzo|_>6bC=yzfrX`e(2XCATqO-XFxuUU|u$W^QSv1)p6@&ni_krmG+6B+gca-MJ| zQoTUDp*N(qb9i?X-QFFUppn;e_0aEb1OJY+uiu3YTTCz#UA%E2t($?(aTeqH6UZ6H z^~KTG-pt|gzIO1q9q&i3-Xq`u^II;yHPr`jE;+(E8>l$y#OINxPi)k&UPss8&^`X0 zv7dQ!{Qckm0M;{Sf*`LUJ>E^k+C^Uq?DM|baXP*Uxu1rv97uE2y*q8+n*hqh z_KZ04(C@dSzCKX3{<-c2t7=)V=c*S-XSCJ;jk?a`%;uMmxVnCix;r}4M7PFO9hpGh zd0*x~h_%~M3)TGxoIsjuFZsgS)-rDbcXg(Q{u3#`5J-^N#s|{7I%~LN61P@E4lqId z4L$iTB;G;&9-#gDcQ>jBaqFu0IQk*&BWiB*y(>BNY)H@l_A7U!>p-4ge>u#~F9r}=^BOCUH^x2zh zj>sg>eUR$zGXu#v@84%o_nE6l#bf2SLk?ooIJS^uzV+_FXSGwu@1r2r5_#D2wG(Nc z(tgj|s@1T6qWkmQ`*i-m=W*w<@sC*hTI+&c^PtxqE04HyeNZE@6TfrF7ji=1feSoQ zcZ<)O!gvDx5$p4%JZ)#Po;@lb?q_^H%e#JT@51M*Pb7E(;~RK^0k0vwM{CtNr2T3J z_Nixx-9h9+`N&n)z2o;C5Y`&tp$(^@+e4qTmyM|2#Ai|m8N}?lqn{l8KS13xy=tad z&mC6ve#qxo$64G_-F^XTR_yz2&8F|1)w${p@i|YxtKRs}{Hj6BoKF0|d|h{`U!U_P zsON({5C4E$->JmZJN^Nj!z2Gn?3~wGd}b}zGxx4CR!{Tneb#X&_cnKjH;Ks)V9o98 z`nhbLbqna-IH(!0pKz--vs}6kSWBM% zKhfno`?<>lf97GMJHr*omlHty1qQm$SHCCX;OIO2=2^p-{JgC{1KsgC-5kyp={x8g)^K)Z2lqEUa=k|{^xEa0YfkkGJ~{vQ*L9b2 z0%WW|Zy8AQt@jPs^9j5+kF(fYzB|=@Qdisiv9`H)Fn~LMKJ0)a&l9=ko`LbkkKNEa zaGu}3u0G|X_xYK-5_69M|AOuvY`69MOnrm=1p3rlu3i)Ayv}t6{{4l!gdC1Fo#Wg` zpLKeoz}<<}-q`l}kc+I}8{JmEb*l!o{5!kvaI}@?{scjer~2iZ#h%W)g8=5tSo6QI z`=JL?uJSXSC%A(P7zdKr6Y7&;9eQ)U8ujmMtAm3+&Uif|ABr7zI)0maMqvi^~{{jI@S#^(4DOz z^_32%V;l3>V>^7t^(WYF>$}i@Ph(- zH>}-=w_Kc#bT@tVOpxjuzIyJ)I6HYgR?azZU_bDWga9^me9kHNJa_yp7EZrCU%)=* zUBOK9fZTyUw9yOb702%`i9dk)l*6<2y=jNP<*!=awY$kR$9|>5nfU!27SeFI7h>_zINrpIiJIfHH-7;s~osHdVqQQC}Xb2 z8GZg2Fmv@zC2*e&*zXG5tv=Jy>I`HgDt>!Aa{~K0gP%eD=C(Y3R-sxmKnHToaXxE( z$Jfy1JC~3@fbVq2Z_FHL_n15L*avcejQV+H&n1`fq^q zMDI~*%{9=)MqD^Y^+-7Q@7VGKTi)xvel|Yvdp{GO&&fXiS%!88#_gNv<~ATloVMI^ zu7NJx(VW`F{9Mnq;nP2XHQd|&`aiInzFheo-~EW4_2muVbF{YeO>B?+j{Ottpw9b< zWBa~XFX5T*G2cDTo;!YV#eYWzXv8NN;E8Qq-5wjd^#+h<|3L5B;B|1SAKa7Q^LW3Y zUEn0#JF-4MUsrGIUC8mh8^Al2d%i2iC(ys)w}#T6--&;xj=qy~#Xqp^W8EP~-{G&? z#Le;ZV$6CW@Y%;&?Y+M5tb$ zj~r`nr;i=XbM_Mr&biBlyu@zgbTB~(bCksWGmj7K3o!n`r!;RMC-O?nV_$hn=d*VP zI@0eo?ZCE1Lw0b-r#^wXJ+Xl`g~xrLAOZt<2l4_~TfQ{{y=t)U1&(t}AlLJO{=(k_ z18nI3_6Zx=kj8!j{eOZlzw?Sb?jp2j&o zi^$_0Mp{?&AO|>8&4F&6I>E2_6WF_h>UH=&>!WXfuR1LWlT*Er1JvE%ulE8e$GiYf zu!B3;fU$x)XXHW;jMIniPR7n&-h=m+2d{den`hlF7kP!JJn?hYd!YeCxc%=2b&zwMK0)bHtm4pnaMBshcWKCp3i&L^bXGVsqO`;_8`a4m+`83Bk*a> z@~_DAc{zu07jxX*qwk@Pj6?$KDdl_v7vM9TvHG~Xy{vOZ3UdRw1AEHxL@Zz3xsK$2 zKv_F+2URQZng2eXA+|MrUheZ@47`Bv=AXz1uzvMAq_wWVJo(D+z+L{j3vA!fZ@``- z*DOg5^&jswgSNHr^6;s5;vx^}XUTI#jwrRhhZ?%I?CVbMYTOzDUf4d14y^S=egNk; z&l3so_4rkLqJO}NPfL2Q*82&5!*@lx*Y`JdfwitkdzF1#zD?Wu*uICUdJbz&CRucJ9y25uX+b{-*D+`y~_b-`?#$-_MafJ-Ou;& zZp<3^t?z#7`Wkwmiv*X#!n-fc^dJ*;yxY^*Nt+!RTtY`2U5&D zX2M-NkQ>Ckb7OKF(%b~KznmM0b*CNOXSIWlPriBckp8S4zU2oSR;)Q$QUC9BUoc#b7t=D%oyTE7Q=F3*GNO4MaalO5=e`TLS!SDU>UrXvud0zi#;l=3zZDU0yb&kP`M%7* zs&3zV{qLY&$hBB`_cLctb)_fgp1zCfPFmNCeR6%LR}62QflS~(;U>PzUY@hggs)ZC z!QXva|9OLXpXU>t)iT2`*zfhPtKY!Bfiw6oscXBJ^SG>Y^<3vpVtY8x1)JEDF;M$6 z@SRofEb3hP#U+-!b#fQ{5U<8B3w5Wt?*h&q6Wh_K3w+OIpQGkxjLw+`R-N|frLLYw zW9jGW>Ha6sfcc>%KglW6PX&u|8wUBA6RE>2&qvv@zg-`=@>JFw>+ z$lbsQa=;p`{r!IOY|atbAHd_f?$3O~pIGz$S>0JWFgM^_nfwAbfcX=6f35!wO|7$y zSn~(At7kb|1~Kzn*K%4}}SVO=x!ELsq`-o>XHV2ySF_x^k(W<9YH z>-}@DhHw6aF0l=mHz#)n-5B1yFjglg{^Zm8cgekv8E{yiUBee6;ppdiz591i-vMyP zbDwCuL$v|iS1zNi_X%r{zy`U26-~Hu>CIbbkGndM##pu1OpqY1tLLG=pmC48e6BtN z+kjei_O#CvE38rD+1^0)C9eIjC-OIVV{5H>1J*c4<%xs%j5U_`eri5^??}CUZS$SoLa-rA`cTGZ&7}&2dM^tU=yA51qa}8@S;= zc3;n@)|{U!G4ZXFPy7I`5%}m^t~omK1W$T}{SLt0Hs{@Z)Q9uTht3(CRqO1ohurns z&&Q*OHEQa69eW{mFKeB}SqI2ywd~XIGZ&{db_Tx#{49GP_WVKr(>^nKwf0Z^2i85j zqYiK1&p>XH54es!UDtc&oIVflDB!Jo?)ZeY$N9>v`0itmfo<5it?!HdguUziF|QBY z!?U~WC)ZwJb4`wOC3eJ`=Z(Cxs1I<#f1>xfG&Jh$n^=3PJNEakJky_jSkL8*F1B9I zxE31s6|Na&YQKShLswkF6FZ;Qy`1L;yqD_19XM|S_t$UeybHAh|Fo?=jK3%D_`Yx2 zMm({R_`vf#K)^NN*?$J|o*_5i@Gro*(Wr_Elqze&B}|*l%E6J=W)YVsG&7 za|iPFE{i|F6aEdj>Y2ZA)G~+j51{{$*Jq9l_9V8x?j1YLp>rmezI)r>{fpNINv!#k zmVWx4^+N0()~PrDAgu@WQ-3 zv=J-Uu+H?|ht)$}J#Xf=r}?j1a_86LZrZ)-o^riM?r&_3``E96Lt{VvqYv3S|C@V> zPaqcTbqC(d(X*4Yc4Ecl${s-fIIsKJZ(;|$=jhk~t{~Qz8ygLq;7Lv!;H+;g_zWgo z0|P|D`)s!#Soa;^@xp=&P-E<~Ha^L1U5_3A6VZnKmXd9Yu~c?$L!u?w-5c!_^w~&K<47SC+nQ?s5j?c);-8O zZ-;C6a{70;hSd-JW8NE@>J{X{nVhqm51^GFiGe@PXrB)6UTQ>=_Y5M)Pq@Ecb_2&&(bJWWa$2oH2~musnx4 z^9i(m_8K}ha%01s>#=KE?D@G@_x72#T=)@xf&kWg_o1byoYq;7Ts<2!+6G^rfAr=* z5>IP^%bBe$EwN|ppaswZW6x@RZrtqyZ>`Uz^yKs>*7*Ed8ltbyO}>j<&uHi~f88be zBac^SO!8_XCAOA5V*Bz(?bkWrtZ|w1{@kyD8Prm*Zx7=}oW%N`r=yp9(}#0;mJ7&$ z<6dV%6005f@|PVfNfY`y!R$aSoBYHA)i&Q^Lj=d1nj)p@=E&Zhne)Q((p+voe& z%9(dgts3V#;>r;^L(NI92WeoewU^(m6I=7-`aHR3z>Uq@H$aCU*y23{d2>FaUxDYS zKIl5Ud9`Y^wa`3)_pDVfZuoD&^Q*n%%USDr&jOS?$|eDog>H(eD(Sb>z=+(<3$|%-6D4G6I`N^KXe_w-VwM-?o4cddwDNv z1I`(L27~yyHO~IP{}tHJ*nZa6dt|LXu&@2a319e}8&I!x?h6Ps@mky<7I$Mo&fdny zJKh4GHKR=3?VTFq8@A_~+un6nVr}h%@7e4(@oOLRpx$%%#CH$%o%jh{Bl#s(y*c9x zxdC^^>TBOY)v~8&^sH(h@V>jo58%x98`$~UpBJ8i{Q}tUNv@kGc3G2h=vvC`{o#y% z!F~buw?_h>S0~qA02^5641D*{e%<4`W<6o8uN-^%>~6pw1HJPM;40nLw7HVJ-I{^aB8&^TX%`ZJze$)@P+&txJItoFK9fEeQx;PpIicKp4*zU zcZYMHuK!v;Yt?v`hnD;RF3;9t&F+48^mn+kI%1da>1QxwJ+I7_JE@_kdsGe9-mc@m zyc6}_Noj*5{^h#**<+G7C+|7k>-Qhl*@E1Y*qpu9S|gvJ*0vA$+828~j~}VCw!yiq zu}1^}ymLESgPU;j1Gst+llvaLfp>L4e&n+Dd&2phr{)Kc*zk(hn%(bj^Iyp9mq|!)O+?j@zEp5Cpr7c*-yQk`g$Jt_Ur;p!`f%SC-I3` zz5W;cA@Hp?XTL^%!Ua}d?zpcT{^^B=C4T{pSp5^<`_q^2*aqC&`}98Kj1#zyXE<~1 zVSnGz4ieZWVgt?8H;-+^@A%f7K$B-)?TG!e63w4|T6>NcI{Y_m>Bu#79qax+{|S8N z0dH=AeqCpd{B}QQIBUEM&*gjkfU{SE3%;1K_O55bY!7s5Du@3WJVB!K9-Zs`y7p8v z(Aw{TUwJf7eD@3yYX)amTc0=b6W>Z0&({pIZ3N zAgoiP{zNaIqjPB|`P0_8w|Db1xZ&Tip4oHWiPg#5>kaE%_GnmZ&8cy|4&Lw^)-|$B zog-q^4Q%W>_44*`&NH?}!aJMqpqzW$(0K;60Xn|3cXF}w4`b@w z=io2sdH{0Hd92HO%3OD35AVuc zhkM{VkM+Kb>P8T=Zp8W*tluFsm$vHQT#@VZ;T|mj)*cgIz4O0=?>?>g7kvM_Yf0Wd z$LHKVyo=&_&fdcyZ$Isfr4M}fnpn^N*frK&;E(q#h~3+|Gnx)+$>r*C!#{64ut$x1 zH#FvtXP3lJHCT17=XKqYbBtL23EySk3+TCe*w5H{&vd(D^?|(ljQwt5T?5YEF6#!k z!|BW4@U4-TJHrLsJJ2>QdEYPRnrK}1u+Gn3$Ceget+90v^zPw(K8Rhv5IeWMj?dzc z_#ZdCE8mk&Jc2#2_H^Gf{FAs5f5AU;Fzb6L*EtgFTqiZLa&P$V>*~4s`>xveNzVD) z=L{FX*fW}!_xyLD?^!$6ebm^y5j&f`8}S)bI z-gzeWW(-c7KeVOK{om&_e_h&uE5Ny*ymS0OY%jkLCJ2ydJ2{tmbMoHfM=+BF)_Q9b zJApO&o>RSNbKVB`hToEZjInA)tasI3vAGMrJr7QM2Ln8Sdsy!otZ|Rik)vnA*>_^y zH_5e6`@4$R=UneEyff6Sx#!2aze}Ba_oR(nto`h-rtTHZ89(4|`1aD)-q@t(jJ?0s zy{Yp)?OPh$j_CC^!y0=J_L|h#Us!+41ygarx$LR7 z^w-tHJsSQK#I_yX07=}@58#Y<;2vLq%R6bgcy*n;d*9(Fcn9VOT-`s`{Z3f#`t8M8 z<*04ofq#P2A3)w--m&%0HQ_G!=Ka0YujD6uZ0?M$9`Neyp>3da#0i4@N&dsW|DAW< z`rI<7y{vQJ5!9H&mj7D(O^epE4Dt<|XzZQDa+BEoe6HS|R-O1B+^;MD!2bbsvG3`a z<%k=-ypJ;)dt2|`p78<9+5azt$RFy{0Y=E<(#4Bwf0M_cW?tiow^$+E%ptZUTaoo zGVk*DPc7DbkF9Y|dwe1GUficqJFy*9oGaI{1GqBbs2Q>E#F#nlq0Szm1aJ?Xylxb@k$R*kjQm;Rn*5I?b=A@Cos-59(SH+*AlCnsE0qZZ8_WG-K;c3{tHvHI@+wRga?j%U%M!g`l} zp_k9ynhEcWKhsCwJ$>dkU@vl?Sn?MSScx;LIDlUoHFw zf}HPj1MYjg2PgOpVl}|}#2ODd^6DSJyP9&qnG$O(-;86SW`pnW$DWPYy!yU#2WR}5 z`agbJpU0XB&V7ek?jW>i>fH(Y;@?5S%S|}@G_3o}Ib#6#?Lhq#*f&4}w{1Pc(MNsl z_Y<02zScPg);+wZr}bAt;sF|d*N0DR&8?+R^~XS=HRDImT=inJSxtcRp|N5Rh>fr48M(iCl z{G-2h0R~9m(!bb~b2Ke8TW{>lelL86e*<@L=$u!-fZm#!YfrV#D*`UC?vuc!tr))H zKd>htKY{aJcCKkz=3J5MeQALX-+c@C=G?Pko$HQo>}P+%`ME!TTJZiOzU^tPNHF1R zZEO7#SYsdSj(5Si?^o3D=Dc(3&)CEcVEql4f5%oWHi_k&bztoyyz?No$4B71B;RA{ zE-x(b=h74FIXn6Uaa}c9_jzC&sP~b5PR1fpl*g9vM_-Cx^XY6tJ7kpvfWu38mMXdK` zzeaYM$g;F8F?2K+O{9H?Mx;ztKaz^J)`L{UramwwgyjKohz0-pRms^;mawgR6VT zs%zBSOC0NgpI|07*1M#(&Px1LOWXx#@r*Y(=d(886ATgS&!H20K{w&55B?LF@4&q4 z4ri=JnAg_5vR5tQ6Hx1WYp!W?-GTcAs8}0IbJ-g8)+c;3|I@PW<#KL&oY4Ph9xL~C z;XmP?#a?CIQ| zbE5UXi(sw3Jw*dSJV0)Hf>YOt_02VrtFQaW+t+oR>kR+(EV=6Dwt=;uy`0VU#M;Zc zk;@+O!Ih9-l6=N`f7H5|yk(~0Ho^mC~Td|}RY!B3Hki&*cQpD#ZP^1h?iI7{E0 zvwD~I)Xr@$@W(ya%URCAHKWY>d8iM6`LL}19lQbm{R)4MoL=jL`#&N|Rp?_LdidaXNh z|A3poc{{lVr)~gsa>u=VCg=V~?uoC)8kcp#r6$Y)@56WJ0cwwc2Lsr9ZohitU#_is zRvv%mu0>mOVvl{#YRTKL_Qtxb$$$pVpvLn$t9uUOC%iT8W1j%}!kS~RPV9c-j;81C z0U!7cYhUd{n+qNP^4fZG_7T+^J93@1dJ_Bm^ndqZzYqBKP0+RI@2~ZHct3v+FyVsS z9ZY(tx#7P7=Ra$~55V)Q^PP3h<2$xd7jVuM)Ef`1@N;FozB&61pmk37O85qT#<%BP zE&%odTr<{xcHw*Ko!IvVo#)x8i;;2Ce=41(C+ned$>{5ulw2aGkJAE zz5N^9VeiHXV)Y1h__efIURn9Ceejc<@x<@fb?2V_4n6|+w|;=z3k`RwuSU3+bDXrf zfByt;Kkvpq_6Tz3FZlg8zIAHV+51AQR^HfmHOYG~em**UEja5A{Y>h}TkAa54C1c~ zF5u2^1KY6r$3D(9&^*8y$lFIQ$;GxqKbYgXv|s0gzr&q?XRLFgbB=nZ?O5Z6R;Uv#*Q`&ti#+h1RPP+VhwlBr z-o^D>ZNS^xJ*?M1lNU`Lyn7tqUFyPIF6&vXF?N}Ewx=2}h^@T7T%fE`k&9*Q_g-OL2+n0zB9SqgWPrfJg|@7 z0sr^|I{fcGt-JaWoy)xYP3#Gs%YK9W8}Yl`#*f(B2V(LGB34iLwNKSy?PI?IBA3oN z&!7|A&po}{JGja38SE269kv7al{dDx+yGPLs&|Hg|HOKaO~le@ zdmOV_-+`Zt1m^5t3v0g~>zy1}SIw62oaI6O3@QgVWBGw^?|RngeC~2V{!KmpgM8vQ zIOj<%OYO|nw{rLwb*vey|B2e35b*YOufTUUaR%<8Z!bS<>YUkH`?#Nd^{wsL+m0K{ z*dFHJ;0N~5S=Ye%g@rfQ5@4ObV{hcFx5nHAPdKf-Gph?wG1mE7f?aEmqsCn99E!Tz zB$m@o{n-2O8Sd0O<7d#x>xVwQ#-YcySoLaL)#nNq>n@GY#K-xyC;He!&CmG$tV;Y7 z*1Yi^Eh{fPtFUL^aVK`Bx>Ia}KeYY2|6LWVKLZ+CYZKILXsw-C zIJ^YS8V$Vc< z`?T)N`#btfH5<<3@?Kl6y6>q=eE0HfYSd3T;5|B@Q0wd$(8#&e2k~9*wd$P7d|=Jj ze)vAazxhCp)$!lJQ78AOA%o|j&IIoZM_%7fuFhH`7UtB z9jNzw69i~MjK(#ReP1i*=lueFFa7hd_u>BD@%O+zoXNBJz3VLE3I6UwYyTGf_*(jb zw)P+w0$yzcom?}|eR(Ex&R|W;HOmQ|tF-vmI=k~dz>IY#p5J|f_y;r(Fo1nJnE3Vc z4Qmf){}nj1XaTV5JgdD$t>D*O--Bmz?(e>JuV|iNf;)Hz7hu22Yk_rIxxn7gct@VO z18d~e{BmVG>>c|6U%&-er%uh-`MLd(y$*7)vpeD5NnRo_4VgikD|<_p$;hjXCwJm%cfSifQQXW+AXuDHdr z&#f_MPxlNo_O`BpjCG&(@a_jTfV~=e^RB9ionr>$nkQGUdM0r1^f-T5@76^Bw5J@s zeE00>vep?q*9q9K#j@^g?sVPzlDBTc9kcg9P8?Wu58!eRm+yq1!;A%cs{Nk)9SnT$ z@`k3#V`KYg@-B5VR{bRR4%B@Heph);_j!VjZh$xZ@8I=aW#gOmyqT-c{3PcL-oXPn z=LGd`5j>0%(7WPOScn<~X-{wR6L>1n~Ry1Um3u%v=BKYyCZ|&&Kl(Fo82&K*!gb zi(EBwgIdqhcCHa8uRgk&`jOb}<5gd|^9vvTY%R-utsAlKEU~pGIknF8 z3#dEL@;rUMPhf54>WBPmFZ)k$C*K8rOTc;>)5AJfu+HM_0k2k#^#k01 zT5Fxd{_VQn1N*3-;0EWN1Q^8PiQlltO!krQz<%|)*|9z4oa2JM6VI*d#2^=RLDdy>~C)*h~tnpp3J@xixTxq-DVuK_KzR<|DXc*D>?85{Epb%Kr^rj zj9mfeS=4Hak97v;ZTR*|tg8oruY18a|FG`0H0FkUuJ@<~ft%`k}1N4&Wc@o^^m9`ne*f!lS>TW6!jS%h=ix5e7`jIQp3xMpR4 zbJpF_+pqS@WuL?!ftmr#4RRAdvCd>q_p(0myD>}svA?#1+8g^odjoQ3Z1s)Rcft*D zr;qpN-CfYg)jNiDSsOr21_Z0Pz(Mbfht2%WKKS zweKKy?v6dqY>g|>``uOje8xcTd2q46hp=6!$#p08@JyaHUai)T#O@d5F8Iz9#GawS z8QaJGtpC-Ro@(zqcf+<%;I+)^e!x)1X9SF7e} z9@KphtMBBhH~U&+KiARMy3BPB{q|wa?S5g*k~hA92N=M;+_#fE;ZLk9a>1TmcrIYw z8QAOS*S6nSH*$kmt@%cLc`aYwegl{b&;_yZ+1`QuW56deOx_uPtH1X zIumr^`9)*OMUVCGZp5(QCb@fen#x;7QIq zYjEz<;2!F+ZQH=A5AqXhZUE=dI;Z{hJ=?@D;CFp^ducnc-y1pSG4Hu*<$AVt&R9Lh zw(eow(Z{`f7XzE1?j%;fv%R;ru_k$E{|MLd)!*>7F3+tWvHE!*li1nnPB(wCo}6cF za09sPRer9TI!}-wRu|aNf&`x1=OxU4MSgpZH*%h1z-bfP!Q-{g(OB~KcE)!#vG#Bt zdpd`3?h~+n!ka&m9y>gbQHq7x;m_U_Fby)T+rK2JUMw z<67ivZ{i8e`y4y|9pC%4z5#!q_67pDe0~q$jPBK9S^HG1?z(;sp0G2CS=Beze$^Y! zd_CvQ+be+cd{xq1$KwZeJSdPfi7{pjb59O&dqm+M~Y znVKh_An={lySMajk{MEAfp;$vSq zmpXm>+Eb184Ek$5Blms)ZIXBX7W{*Cw&OEqt^0Ix6JHE#hy&j{sybrV8?mrQ=9)Bd7Pjy~4;j@5G~c8wqpcLujl>&~C(dcu>t!PhgycLw$1?>?>Xlz;VM z{cph3GePU$9r<_npXp&A_qP87f8x9BIb!uUej{$ix%+FuMczg23BTv!?d^V^&)Ew2 zu83WyMdNzYzSg{I-FpI;GuhWYoFRdC>%3mpXajy-0zwm{y!J$u0Yd}Tju0(tM=esXtm_VT$pi@uyQJ%E~<_SI?` zzv0_U{fwgPj2%DVC%B*)#O`OTHi_N0=EV=VLGFS*bY;n@y`lBIsRrw^r*Xq}xBw4O z?*+U$XVrEz4eL1y>WH1iyRmDxdvt!M9!u>68Q-j21J;N^pC{<}Cv3y2@i`~{z;{LL zygoV4a^4!Pb@okg^sxS>25!o`61-O5|MbH;i!-Qw$DZN5&mgvL0BbIAXE0;s?A!5$ zs}nnub=I8leRkIPZn&qhdUv-ZTdPib=9fqn4}@yjBl;9s4$T`>OYxcQk69#oPeKEmr^IPD+o){p;D0SF7&$tQ!B5 zPjD{!dMDa00{_Z<`1TU+`6PB`pPl>K!@K>Ju+FYk2%XYAJPGsbJ}>73?%1VK;x z)!O_Z9=URXpYTcST>kyj8SC64K}RDRzB9Q`ziy7Y22RBG8rb{J9d~E!9M;PTwQule zwZ!h9a04{(kpE3Aug1Oap!&giKgFZ>j^9Deg4Xy1`mWQrK6c$%pWFlq_q;ZomG9(S zH|$89So?Suxd(7=Yt8AqBA3p&F2GoO^mEUIbH>lWeXXw^e^|WFxy!lGw}lr@-k)oi>pEk{`g^o5*pKM!bpt^kefPLs*FN&j<30mC;C*K18%SWU znwdMWM}mpAo(=v9jWdZs+(FX^-i`cS8@ms&eq!CrnVw*7&RXX>0d)^-eKzn1I1~Hs z>YKB7^|&s}ueGndJe1G@wZxNr;0yacvCfh5SkE}+K)`iDKl1||GmhLm z_V!xOWMC&6p*Dd%)jHpWp7wo!GqL&ru8#E?My#5nw&sC5;-;SXd|ld(oz`IGJg+r9 zS3b2s%>ceXPx6T`cfDXeCr?N{I2>4tPzoGU!rx7 z#LlVqxRay)_#B?hzbF7_piaM2+r)L%UJU);{)E z_v>qYFC@7M-ifCiIW-roH4W6P#1GJ3YY*>8yn*jNtQkfw&ga-bkZW+qdoyC+*}kVU z8uu5jo~u5v4fy@(T<#-h{(}EAa3(+ZCm`pY-$57jx3{yqX9Dl2W1Y(yXH);LKdm#k z-xuNsxB>6y9h8=w{r@+#1HWdut{(1ZJYw~~!+THe^^X08ynQAb&*R^DxK8BMXcOza zzW4TTuXQKxZLYNN?&IpYbT{&H)+BHi|9-}O9>8VZyHYdYI<|o~{D{@xeb2a6X+J!kA1>g{pgIp4*O{+R7TT>D{N?m2+`&0Owxl(U?P-7hxhJxzRY#v|! z@`7V|rj|?dhVNZg&w$5Qdz`nch4-BD?YiC@>l1zgwazK`r=J!EdiQ$=0XjMBf4HIp z)=XmGXFmh(8^pfPzKix%cLE(hz_0l7)()(&-)CU&4&-VjzXI<$V%^sut~}QLZop-~ zGyacY#?rfQor%1??IrrQL(7^87YoMT<8dGEbu<2Ho$WYF!bdLcv7ddaC$=-k6Z`@u z^*-OA=8k;=`?`ma_nzE$O+4~)u|Ht}I z+JDjh{(t^&|A+tafBo3{e4|I6S1)4%O;`2Ao1<-hpwx9z|Fhb}j9hCluH zfA^oUaTEVvGyLuUdxpRI>%U-zfA$ZXf$hTm|M!3Q!+-S;-2eB#|NY%aZ0fBqk^zdx(|44;7Seb4pz@jKjiTrRbN1+IFB zbJ_X~930k~2emC0{EWKftIz%%bm9Q^9a!OKTFlMo<{Qg=%@^c>{R61+b7kJooArV3 z=Sy7&>YBdi!>S$F3tH<%2hJT>*NkuV*4WQ|8h(P~zFMyDCF|^Q)Hjwo^VTKSSpS%J zz}asAzjs{v9cXKh+~41A^>#*knhWfVWdrN-JFAvDU>`MFpJ~T7aKpc-F}|+Zygz4B z=R0u+Ni5WxtA%eL&k^_!VBWnG-k#P*;sbkiFo5|3>IRr-&g5LyoxsQI+T&OJdTu|h zKF-hsAw%4tJgZm418AR+J9n?_vAR+gB}6@wVwD6daU?`qdf?A<&z zuI)EJ1DDUQ^)u9dYCdA^sdZ258*u)i&+_MNCw7jm&s?6Z=i=UX&Y9$D!P)14XSU~* z|FoWk@5uu=ulo=6teKR>x=(!(;CnRk12ka2mTLxi;fh$?#Mfp0Ej?Q^+O4<~N$6Pv&p zjPGC&+q=%-{(&zC9_Zwc^YvKy2X*!dpdIk;rL|YV9C5(A*9lyJeFv~dBj51t;p)LJ zx%5pf^Vq}uFvhnghyy#Z_v`Y9+^%UZd0_ohn+x)9ul?s}a@7;LecjVOL!VXmlB>V- zKX<;g*L4QZ;XcRym7{jXIU>@UKjmU^iLIrj&kdL>VAPAdJoOQ1u+EHGMx%ZlfHQZ2AJ+RE;aricUj+M@Cw^jueLe#3 zG~v~BU`5;;)sdvXHqVLyARbM;vHnwR`>_tv^@aJQ|OvyVO0 zIalJJ;jDLd`wsl0UQW#fF7xI(_D-(eH~a?b%(%|ma_v)f=<3|?)^&28;kb){e*kTg zlebr*wXeO(VqEr6djSK#W{{_D}8DCdiJ%~N8 zef_gZ!`7bIqwm0Xt|yRl4{Ox9GQRaQZyZ?j9q4P58hah@hk5a4FIL3rTk~)SBUi7g zA!pqaZO3=n^SH+uE3Tf0kH$0l`v&t3Cpmj~zF{7&VXX<{Tr+DEn74=T zWz}H60P`;IZ06o0IRAC^wWoE}(><&~>x#s;FZS4X1_?9+u2!tOwRhEG&8bUZ-zIYP zlC%H7+EZ=4BW*3u>}=W-@N;6Vx{e=O_=z?D0Pc~*Qw!A8BIh3B_`Kcm-Mi{Pp~CMV z!2mw*6L6*=_WbTs_u#$2*ZjFS@7rE=XIe00!roM&`qZ8r}e*w1qK1A77P{UFY*sWZSg_@hS7+6kw=G{iUR zx-q`{r#{*@>jlePY=7VESccdULMR*w86u9>g{ct7qTH?ieo_1gh*^3GNFf$x0|^DLjEbNO!A$G#bBhI+>M z#`<&nE!X@vYv7zK@GF)B0q;8DpN(^I_A=*r>{I6^-{t-Qhi^Y?F56Gydd9JIKg$DW z@@JQ`tL^BHK2-x(3ysGkWXXt6@j3xb*heeVp68{5w6qfa6SRoZEb@Tr=qZ0`J{4?R7o3kH)d< z-8UC9w(5W})=cS$%yj;Q=53E0rFXpLFasl^s#fsH=x7NBM@Ka~u=-uQt&gd^- z5+@k3db!Uaez2$Y^7dBmxkdHBb`ZHW=#Kl5@5JVgy;Gmwu80Nu_*{jv)EVI1xAfdW zCmz6kCpLh0Xb<-g`U5=SGuwL4ubd$^cg*74 z>T1Q(xBS*mzn(WapK;aUyNCB_tz5m=d%n6Oa%vI;{&9z%rNe#w^MSHleJ0kGxit3i zPK$q$5V7`iZxO^3sC5r#vHpg=1NEKw4L^W24|%w;b9FxI?ImBp)JpoRkEJgSeg~e} zp0zizy%X!dSMDrB9rmd%zwO&tbuRJ~Sl6*G_sYZpi9fOC0?519!1uXX=QHv1C2D^( z16Y57yMuOJb6C@et0!}LC-$D8`sUKzwgz>5KI~`j0O~ujJ#zbvCFedjb@=AT3!Zt+ z$+cK>IpO$`?>?>GL|@;N z^J~qfCSv_8u}|Wk$W<@n+;^p!u za}B@djkS+C`M~<`q&e@6*dB5TCb(P6vZs5GTsieEmhYa<;Q0>tIaEK*0qc(*75_w> z3)TdF1NQxsPg_r1yeWLw5#M+39&09R8@_SJSMR$!hyzrP89cK&HS&jUZmwYmkbhEN zb4ROs^`7%W4_8kR`|s!Mi*J2H<1)6FXR7y=GdH*i0=Q-}6Zyotm-V3^>(5qa`VM^0 zYhHaL?%31^YPIU*1IX!%wvkx(au#R$2+R!-v9vxjb-(QVk(29i;zup|+%?UdKJ$6! zYU8@@De(PVKCO{Q_jRE;-bwG&ycn_GcV}$7PTptd8d+w27sEY3sN-DrpXAjy;CWrv zCjJB5q$Oy;eqe`b8{U#$=ORZU>v}@3%G$3a4+vo z+lYtpmHk1-cMj{+-0|%ttaSx8v)t7q*8Qk=CUf<@2;b2j&)j_Br*A!+Aq6!zYF);U zy+>jj7;xUj#Kwj-7jkO*<{I`mhw~<}b>e1B%^f@7eLmj5vvznn<2nbN^8{S+#P(CC zroq?3Py7zd%O7VvG=47)_yARFyC z#20EOIeXe)ZGr&ag{VBScl-{#lZd5JXWxeHpmNw78vQ{oz@b68d^qhgIi( z3Cwq_u{!s^gVGWQe5%L4Uza`pdsx|b|t#H12`0GGY3vHt^DH-WX@>lw)PTp+5y@A7;B z-x6TY)OM`#&-&C0HJ3vRUuQ;>@K1I4)(xy$t#fFF`&S?Q7W{*~k)OsQ*1XO-@Z~y4 z(BRBF+r)a-NPvIDas$pgQaeBgPhxGI6W_WfSo&YjXixfEAF*}V1jpy>PClt=a4z}s zbM0a6MUJ?1=m)-g+sC}I=Lq8CId%NY)(&h(qy9Lz_hvseO~mpKO<8EyY3F;aeOe+1 z*4=@9T-7ty{hPPPle~4Ecz`pp)>yrFY=5~MHjZ4Z=V0v5%pexlG%e6(t{&cpJzVxY z@@fuV>mKHv-S7Gd-hiv;+9R=QFYtl2*8o+^{`V_9@VUsFZwb&__Y2nA4kocZo>yY_ z)qmhy8?ia8eu7%tRv!OBAL|BwV(l}qa?O~TjLQ$K$fck7=Iv#lLvv<-b&(6sB&Xi9 zop!!rxGvcD(A3>g>zR+|n8fzGZ7yQXpiMA=J=8tXOl;i={2Os2w%;B6Xdd2G^CuSW znBm}WS3F>^5#M0VXY^+f-~`Mk_P7u2g}C;;F3la=(A8|@GuLc)b)cIA?rYB=KS07c zo6B1FG$-$!dB4^UV7|0+&Y-2%8uzeY2W4aDhyzSwXTAe#8?l_epOcyE*}12=TFI{g z$fuSSvG#4b{O>;P9@HhA{RbMK!yxvoXW*Ybg1F}0w(opBi^d9nS@&kn9@+suUP%y# zap1Ee_$2`=6K(7};!G1?y_`DhI?y_!yfw!iI@cr~@D1CyuWsN6$n7_*dEX!NrJ<*P z-m9OP^;G87ISsZ zgJTc14Zilm|6(oqUpMcaC$YUUv7mYqcl-e6Y7Z?v2V8-;;d@^LKY&*4#0q^iJ`erG zubD8hXRxm3$n~=zuToK_d~9OH)hCKdwPEZ-~G<`K3DsFhyIwqv7B3t z|Jw-eZ%_5kr*B_-iZgjpE0;g8>JGhrZtiSP`WS2LnPPoc-Ph&b5gW-fBByr95u2}h ziKj7V2=dwsJ=EWUxhJu;?q}`|zXu#~fRDs3wI_V{tvRr+o{P6OW9?IO;G1*b#JY!d z#{S+~eE{!oz}ee9>~9}y&TtL?fvq#)zsrFOFhAj}buW3XJ-l;s0|cD>1p0S!Y7%H4 z#_$cOKbA8ZTj&1yTKgaOTRL*qwb(Q0qZ?T7JK+Q0GuTIM1QrZ1foD-`U$wQaNb!Zd z+5ksQ;(MQiyj;f$<677N-i@~O`2LJ3E>_=0?lau6|Ad!QBL;o~H8*ev-i!TwcAZ$g zalK!}lX&dd=MdnwA+J8L>O=?D%axuU&6t{zSa95ddl<_%^7_`QH+IR{ORnP|XOpk~ zyYC>5SF5G|f}d~`-x})&s73CqHWnvuZ`TZb;PcSh$9wVTW+QI7_$RS@)^o-`ZTNY( z;alVLF>i18vzPg_j+xAPPW=bgx>~U~`v?579^_~2+9U_m*iT>hY)7EMf5Vs47x^3i zxpf)eatY_NZRiGQ#1reB*1EsvkT>pu&vJg}a^C=#D%(Yw>ClK4NKWZ*t=|TI(-xrNeg*dkn06z!}F;M{Lez&idLf z@L9E2mb>a=y)WL&4QTDx;h)CIKlsinPQW?8_A$pA!7s@_pabJTr!{`Omi1>|;(v!d zlegD}*!kV}4f`_~;Gdwe)>Sr_riBcC?EXKk?}IZnE!PY|F7PMt-u?6Ngmal|Xg%K@ z`1$w+_%r4PeD^P)?lsX~0n2_j{0w}o^Se6MeG*vTfwQQ0dB*CcjTL)F_qEP>>duMn zVV!fiMlAivl~d>bPxAJ4HfM4k`QPf=jy=CCV(sG|>Q2;nhSLk4xRVd8x&h>Cb~w+~ zu=Nh_oNbZIv>zi{AefP0P0sl$PJv>h%Zi%rMxPU=y%^Us$1hH%8 z>UWb1ne`miC#YV5WgWfiUeKR^gX%}@UZS4MzQ3MpPiN_|b8s$uxYRhG*7r%i_PMUR zE+3sU-mL*o;63{+0zavp;DPVms+D(^hHqZzi@B|umaBGP103;#KK8ny(Z2w59eB=& zrB^fXJDmJ6yZf}Q`yTra+I4BvJ;2DNv8Vm46{!zgk^Ztik9X>=5%SLD{+@H7X>jhP zzGCd$T;huktQP@p=+$)aw3cOj!cVMdpzabofVty2&scuVf$xl=4!;3?q5iv1^oH~M zBEbWgYhZ#q@fqj~@6S5di8dr~XSDD}TQ8(#3uz+5_KxI=nqX3)Br_t#eQ83EKi_bJaYyZ?8N48Te=U zFi&smomaj4yRWO~=7RuOs~w>3B(ZSzk+Z+O5`TgVvAG1pTHu^#AZN~1cbGdSXWU{v zTeYsz;>UIMddlNRm|X8*)xmf8mJ6PJ;*VJ8(i&TP21#u1A&1{V`?TikiBEmKm)w)H z_}LJ-bJkjy*EP2@$Y}?B^~36WA6oNweXPE=bPdZsYOL?__9JfK#J<{=@YE;1sCSB3 z-gtt*@8~CXfII#TboprO{UB~vkh@2$y{s<{e1g(ooTXyw)Hsj#d-S!Bd;qnRxXwXr z&YpE1{6G7&boRDq&?m_U{!ufD-D7~Wx*a!G;1`!GuV%p2U10q=Z_fg;`v$GgHMQv5a%rs@-~xB2XPujz zedPma?OQ$CTHbMopGl0hpKHeIquzTw?(~75Ee_D0wI=V9&`{+CS(B_WG9kF`fEB97!oioC$W4@XT&f4mQFJ}++4eRpHBhKI~*4op3Csytbg4kU3CT+hbymjtd zby({ja^w@Z-v!;V?AgK4($D*|zgla})n3%it?9A$Xz296TJjA9{0%s(`2;RGwFf_f z7`Vs83g3}48u0`!XRdSRsy*IYYwzegcA!o2`f3Az;Maa?b`NzmBl#d-y6t!Dh5Uq9 z?+j|Xpl9{Pua)aQDv$3Soq_LwuztX6Gg>WU``sm<&u3qIXw4;<8|s}=xNn1ZKCRph1pdrDJHBh=>Z5JIyfyYyAN2Aw=?pjQ z1$;41G@$MSow53_cS3AF!Q+ZA@B=)lpWAn@nzMbI&Hr7lhZ{|FlfBH@%bKR2D_1`GJ2lSL@ekcZbM%x8b7)SK1`p$3zEr{)7f8uUTzMcuz856GI zyCN0_*7MnK;;X686S4a^gRAD*)>?aO8*s*&6Z^Q5V}=3mT$u|G{DI$qeZ@!6W9@C9 zWBH8a?NffP9w+p|v$+36Bb@OD&cOWw&RTQ&3C`C&&*u7XtofP^P0+(VX8vX`@;7tX zhV@-ndSk-7qrK-F74(!8hPRB zxjwHg7Uv!p`r7Yne>9%inbgaBj}!llZ|<>k9eV~o6E&ZA+==DXJLjO*e#V2i5!X1lA_j*nMZL^LA|I7!3=kh_})v< z*IdW?S$M)t;GL+`);&>ftv#wwtTWc<6+VfNa|FKgI{WdtRaZPI-=BuYeYN7aPc?mW z@`E}tw?>`&SbK*5zo~j#Y)On{ThoCC5JL!XVS|7TdM#PJ)dL9dK*Jb948fsbz=941 z4-_lQM1lzg9SRBx_V<~)pOep5ZJF7&Z9nc2=cGzib#G!lPkPFi4}WH4z~|bdVT0Je z-yK+c4B%1Y(N-?Vqx0vZ19df99yBm>zcIe|^R48~-|kjE!s~aCz)ncjzYc7O!2(XoDK8oCMm{c8S-4T10P5q-IO?^wT| z%pJYHyC*qk%?+aJZ(N^Q65~JV^~?8FbHTp>kJkBqhZ=T*2It(GPpnmAZUEoW9pwV+ zseJ5zE}jzfAok2$bKOI(-Vy6(R2;K{c!H|qUIX9z`TlZ8=f39L(H-jZGX~yEYp!O} z_XFR3CwPM^@I3EpW_@>J?Ia)Mt*wJ!yVA#*&hoPxi6wsHcWe*`R<7mJ4t+2|-7}Y_ zWBcZl{_f+PdV;weFqJckKUcDAJ*Qrn--1y#Cra~&scq)_yhd|STo=o)|u8dU~MOsb4C;R z^|@lLbtgGtU56X^iSOrovxcMXx%>OMdE%!(5Am&!{Q7%N`*@e9J@9>hV>Pu4=4vkB z*+bsBLe817{hs*5_5FcUQ+?oz)6)CgIf;MHU#NA~|NH$*uVzH6)3Nq%pze(A;4f{i znuZN@4GjDV)T^6t#oxbv->B*Eq2+!UquBPaBG z;?qap8NRhO<_GA&UG+WYK8Sz${`LQV={m>~GzT`oMPJ z-QCYzz0J7v&P?f{+!sm$I98e!`WZ^ zFc03q+yFN@&IsBj4!Fu;1N+)Ix`_4f3)DJSn{f3V$2wb$e8cwpnq$9;e}w}V`Z}lT z$tBRYr|blpLYu9${B_+u7H7UEDR92CYCF7~_tJW__7Y8>Sr3r- zSFG~h&Y<_GsY{a^&oW5_azJ6xq zKmy<61)M*&ZerD1oA~bAu?Vhu z*xF;;q1`>zl+Hb<&)9R=@hOL$a2?p|54iSy<>bB3L)_tsz5nN(iJg>!cdavz z*598ua0T+#mmkFY+`ermR?mjk-Uaj{*FS@HpSHF3aaVg5V(mdxGpy-w1DNmBoqD|4ZvU-@*nMWU+M!|H z$zJyFU}(|w*gpA&e|iRa`*`#d{7rp#4!(5_TSqHm%{YF2FPL%e>pbuDvJdlo6KA-W z8aZS8PcY#0D<==^ZR~wJvH!olwAQMvI})pFTFwov$A0R})!pF}-6@s>lMXzb<8>VfXH zb0LRjVB5*Z>Qynm^?k!VX0F(t=Co(FI_rj(GlRb*pzi#01ET78+r9q5)``Vi8@X%c z@O{rvM{KVSGS?fZP51!5gYlKTv3m(?UK`eZ9=J)|1Mm*|1Ne^T?FKn}HQ>(1YR-&? zf9CkE>a3Ob&!z$V8LT~a&d*h-5!D|%Kmxf5jNPT~kFV~`J@@Kr=%rTQ)2XTVwl-uq zi6+(_iM40t2>bi9dikV`nB&iizH z-|mI)TRnhv-XXBH4_fc_1pYH3c{TPfi>qJ2$f*y|@TYO$b86RIJ9KPAqu$RXf*4pA zaMmU92f2w=Z|qLq=L*yWT$9Hy3$O2Y!2AHlGg{@f?$NdMx~V6Q+&sR0&XG5+@0c#W zXo&@#dhaoJ{JgI|?pODXtw+}J4+egMV<*1v@3EK1T!7!{>Cqp#@@miCf&T*ME{U#F za|8CWrglN2_1@;6#`@sQO10EC@To&9Z?AxNmU_P@i5=v<_hqA@t2xdRXbSijYPxyg z-b3Ig@TiMiy{`1L-rU&*$vUy;?S4h#H)=kx4`81LA~3|YFL6gR;Hp1y;!j|$cWC-x zB!;^BJ%V+Izp(Z!J+?GhkNHR(dX&XKd96L29iZ==pOJUGeOupY=S^~nR-Lf66Wd?@ z^a^ON261wiu#llVoy;_#nv16yy4y%O8Qek16Oy}%i}hk3aq7x+!!U&&Pu ztU5KmwRHm=o&KrovA&-j&RGF!e`0ff4i9j8Tk9?fw9fm1ZD`w_nEV7IM9CwVo4*faC%=dt5^Iy(1J z)4>Gxt~|AY@A3QRZY=@4IyLfvKk*xO06EVUT?gjf!&=|qocg?&ow4TpNp9k+yW%Hy z0P}v94Xa=8uvKdC_~skvv34dZ&R&@YtG?sQ8S6KCm&Fgbim^kV z6G{HRfH?zSjhMjuK6d@@S~xWmyo`5UUd|ddfxW`je#DKO^Zgu8U)XvV^>Q8C1SitJ?9YfkNnG`C z6Mq0_R2}c8-{H&&?`uxp*&Tburg0KeBUdq$8u=FM_lkLY1ibIzJ(7HYs?$fSUf#U9 zi#2d!5IeJD&HEk=_&KWUK#j+F<};VopZE>*4filVM34uSbGH88!F&?S3IBJ^>YVuo z=9{?hI%cbza1+~sClY_>>}^g?d!vv0$PZ#a8~Z(g{;_H^SKX;Kwr|r?Up?@vUo5>p zr)nnr5b$sMPIym7nd)(Y3t}}5_;%Lz!$D7cPngR!tNg9|+?{;=Z@b~$uX_D=?PT1^ zYs)t#=Y2Dl@4bgUI`=Z)h}G0xx9;p`p9I?H)<+O?BB=o(kM(`3r@L{~{Q+lf;v3(# zw&9zz##!#=9sOr7o!nH5|L?c`dxSW1-M7J+Q>WhVt2sXt`#Z1ZaJPX*-GG}w>pbsL zKDmIir!yN+f4wiA++XB^SPbGg8u(0f!aKt^eEVI={qmidGjr+WdjR%IEl9Y5_4xDb z=Mb@Y>+GXm|AlYP-JR!st(Wud6009c46G|(j=UOsJiwp8(___GE3EM?$`a4bhdE{% zH^DKMAK*gG%$1k7K8c-mc4!>+-m_z^yWX+DnSQ2nYOM3_6TbsbPx5drmQKA|W3|pv zp6OB6ksw3x}I(C2uI^WLkrnAhwK=mNsV(GNrv19EMxq3d( zo;qs=xf|Rs-xS?s}>{am{D@SZi2z|IDmssm3)+4Vrv>fjx zXH5?X;2T`k=sS~>xuE9g&FhzD`;*vRy^lM-fc@lrKl@iNY)5MkW3>%ye~;ghimA<5 zJ!)_GqmSI58}{s=H9ySBQ!nS7MyxgGIs440aNP^vymfWl89cXjmm530iS1==2l{@m z&U|OtL%wpJ3A>pcfLIqy2L_E77*I>BGT%yr*JZs5C%oV~2?)XP1; zukU;#S6Xs@E}hs}N!-8$&N_VL>i;{MiC=d?^QP9EI%mia{2=G)v2t?ty}&kf=iN$c zj=qT%)f?M@_pbQ+_J3V{%sr{Of~&J+-C54lsegM>40_e89G z%hm6?bJkv+#ofGfC$2Zd23q@g%pJey(rbr4pQHV2XZj3$wE@g0arM=r)t{i{uE)OV z?;EK1|8IghPw8{@cZZ3uX5btD3;KQa^iI}&{#+6ot1oWvcKUQ`&9wj?dwCc7xQn#` zCh*iQ^oXENBpkS)y&XOL1RiH*E`9YzqrQn)JxYt;LA!6eV}E`^BwDP;`oHlLI|2eL zs*l>(yF=>-IrkV?WA}Wk1C98Kl{4>rH3=pd!0(my>eW`i|E`}&-(%+5*<2GG^(~i1 zz3~6$*US8lj|Qgh^ti9NgM0_4pVruO?lBXe*$sQyGb7gS9oshK{v zd2MO14|3kAdSgBIb!;&klXUP5h2E?~EtmA5-a)DMW=T6p~ zK5D#+^J^w?6S3a(4SzF7TzP8GDelB*++VoY^S)~A^FQCe-snZ{?ChLLK7~HGn8(Tm zIDCWu0PbIhoF|B_8EF7G?peL=%lEB(gUaWsG4JWQW>k+LN8IV>Y&q-XUiep_zLT3+ z`DPwI;q;%N<`F;TbJhKgZ%>c41NaWsb$q$@9iWppf1_8(Z?45xop*Gev*ahZ>T@Di zkLn3$Jmjc(d|Q8pdAEdf_us(zPV`**#5T|aD{^ViXE%vo_jNaOX-pq?a*wLhau@S6 z@Ueb=-fIwBqb*HabFFjFbCZKe*HZL z_7C73`wY-P<0v&gA$F@0j=8da9u$~0y z)Z6F9yt%n2{Ddv6J{l-aC0R{L9i>r%tZ+hIfW> zV1wAaIq&K5{(s`19fp3A0M2TlYdIsA&Ym9oUBBZH+s7K;M&De`!sf=At*-iFN8o~r zz5A~3`L(yX34)yc?OiAME3m&k)i@)yV8VG%_w{p;*A965i1X&Bhjl^TqgC5t-yQ#d zzOV0*x=H-}_VxQe;6(M^_IwZ9j_vuN#*?{s#m-o}_?DgcrUe1i)#sxI`2DON@b;@b z)|x37==hACd2;d{cp_IH^Y*<`cM*4z0{?g58Cj_upH2 z|0nP}(PPc_PNhL!dUiE8lX+TTJct9RX|Z>~{zLr>_*tm&sH^Wab0Q$H=Wdnbo~5Pk z|FF~z^ojM})&+2H{g305_kC`_{?4+mHRn6j;5{{e+n(fs+8GPYv+H!-GXo1A@H1D< zAoloP7xhU1hL&Sp#2Q^#iD{&lkVpmxg}k)TDMFwQ`;O#5z}f?MtZjXXOb7{F}VomH3J; zPCq$&`96Ui-~m_jv#on!tvUM#vAcHjv3lEIByn+CG=BKMyk?&3_2{pABr zy?%WbSkDK#k?^tZVcva3!dc_4iLHZQJ2>;ccB{9^rB!2HpfhjZmMb^${aglCzj$o* z<4r&Bmeip$wojr7xGqoZXPw6KLVw484(bQ7IeU~(Jc*rOJ&AAjfm3H)$0qz|4S93^ zdmMje+^G>eD?o=+FW2yeZ!wM0+UE)<>Mrok_jg3#%RTgS?P6c^N$!btzdwQXH3RMq zCOWlU9@wW5%lkG@dCuW4_cc#$#QOQvx~qNFO{~3~S-a=bmj9FPu{9r9XL+nCA3JjW zJZkM1Sm!t6bKsq8ow$IR>)ys5Z3h7oT#eZ?1DjZ2Z0!rUf54r-cr|y#dgIT}1>QO_ za?K4i=IvE|?ub0O^6uh%t^4R(lfb*38td(2--fTfkq=-mxfc8jzXMN~>o&@T#GnURe^|86w`3CpGH(%dfHNgFTsJX9M=KSx;bQ z)Twt?2j|rNbM>%K^~BC->)^Lov3)vmCvI5v_D!HhUO&Lq8F#>uAD}*axJ)$UpJ3t( zwdxWCXymNXS3d-P!=C#jv3ZaEW~{q6to7EJpIGl%=f2+fhW!(q&t9GQ2ipXpgFoNp4(GkE)&i}4PTrh7JX-Z5*FU#$e`icM zYt1<~;nWB}OXnxK(!u?_n|BZLjl9;H20F0D6S;QMsxNIU{`QR+K8W4A7=N zgP*{>_xg+4y0sZ|DIZ zz+-*nf-@?Hn{wFPwqq-g_1wPiI%b(GJ=VSJK3Hq*=kfQh8ntrb2O4Ah*1Mo-@W$Rj zYu^7qRRbHKVsh#~)WB74`&DCIu;YDwU#spXu4>5F95m)dg1PO8l`kH@-jiGt)SS1x zF<0(`{3@*^9}4{u7l4V;0Lis-k)9L;;;z< z*vC0qd#iI#xd!~Kj_+Q%d9|W?BpUoTuy03GcaXy5s+0Z%(~u(1Q5%my2ApuIk7qxaou9u+C98 zwv8YL9msWH?D1YL_?I>pIz8;^o`Ie4M}KBj4{Eg5d+)j@JNnP_2V4TJ@I29c&Y<_m zRg>HL_D|4a?d3b$)8pNpWlweT_PyfEnRjm22N!DHHSqt`W~;uXrFOzM;B4&+jkfqZ zF8;``&(D61Sj_+l+{4&f-`;udp`GwOcYmxJ?_c*JKDqPr*EkpQ&h(rak?SqI=L`Ie z#@Q2IG-5RgtapybJv_Jj(mU53Tdvr>-BEOber#*3Uw<}H{{igNb6I-?`Kr+e`saSu zs}1JK>!+H-<4^pCZCZ4x<&0SRnFPS^mHD`@pS$yW-}|_eh(NRLzu)$|K&g2X+vkd{ zeS(eYJe`#|fy6Te%>vCdu8+sj;2hZP@kpneC&%B!h&rdPvPf7uuT1?0?i z?Bl-RY<14GcYxBs*IN;*pWqGDdrr;F#b3!a^ThTzchElJZ$RGT-$!3T4+1bY66?%~ zeW5q5?`vX#`~V*NIm;6PpR1>R-{$D!=X$}HYiR71=;VxFSasSoCNFQ_hdKNkoOkxU zg7{)gZlbZ5cW&7F3zk;g6-he&TJInol$Ps(2tG#mdvCiBDO@PzedVBp1?C`^%S){(^sMQCAP+xUbL3f0q4$CP5?ua@7mJg9#qwtVvL{^7!VOxgDE# zcEeU4Yd&)AHH`kz0y4u6vi)ZmJ*-uE+SU*jJ@f0*OU z-1G1UzPTGU=lE`=Np#fMW5mAK7Cq6cyOHbole{yljyE^=Aa^Bay}39bwqGMwR~mE) zj1%5>u}+kZp4NDOzi+wyqYGj&@@t2%7TD8s#kY3Gs#n|ayH=k3Kj56x)3`L)NdL0; z$9nu-S-q?U4+F5myg83`;&*uK^zG9?$DiOyPU{?Z?ZFQB?bGLp?;F~4pfm4V>pQdR zvGxrxwB+>NPrm{GH=Lf#1?M@(9vy34&36ZIgSSW92;z~O+g8rr)~f6H#`5-Uz!~l@ z|Cx`kdDcy!-agKDZ|B#Y;p|~fZS_HOdiu7JOKaX*HTBQ*ybIh1Tw<+Bazo!bP@ff? zCt`hu&VHipx#Gb0&7EU?i{(Gy?E6xSwZA$!?=yi~xpSXMELVWHhdZ2E@&VrRi4Fv) z9&qk8iS2j8caHjm8(8Z__1reG>Oa(Gt0PxF+`zZb$n}05t3QE!+qopZ;9u`6A3+au z*bY2?e&@gE@FzKYJ+Y7NyN6th)xY6a3}v4rc3$atmx|%->#5HR>s-&_t&QA&uZ-_! zQ9Hpo%ly$!INxryjI>G))&XlK~BDEF*DK1qqF{s z*88bf*RcWY_W}bhKmz^ieOddQnt|WR74WN`8B>nfI`8Q2L+i;^ueJ_W-=Cp>fjbUj zdw3s@dp2ZEnxP!e1xrTM$I%wS2dt`py$-Ah%VD0C9rg^MA zUqF2TWB-}dz}902w;u7s?AtCT=4_{3GdtKzp#ED#@37a%*b87t$5%UPtF?e zWUoP7-zVaRZ>+7O-SzhIe&)3G`35m3WAA`HdG)7H%}e4~Gx4gX-*G_ZntQb-_!o9y z-7m;h&tLK32C-JHb$z2W@Cmwr{}YXWZrASk|A2s>AVG_zZ}{FR?z@&*=H;|w`{&MD z=R3z9-k}k{!7t2{i(GfGegb!}PXu-7yOur^e}JliJ9RaCYg(>e)?1grd7jMGw_}To z^=@kHVeJRn0BY2i7VGKq?4j=5*Y8_#*nl6v?}&X;9r}*-ILG|JHtb{TN&{zq`6qT_ z&kSqoO=8)jW@*WZ$`NNYf^CUA{s3y4F@5B&%+T7) zUEaIDeVaVc*3P+fa@GlVEG;?v4q&egfIT~PV*3=Q&m4D9e=zaY{D2RvHD}+%Z@D<< zPU0Xwyl?zFb=KHN=(}5z`dhA*DpV{gJ)|uu!_VA3{Z;U4Jr=x&GeW@IN6s1LV;h|Msb#Lb|9$Y8YfpQ6KW+8E zS|cv#2G)NMV!qy(xMmW^_RTrB-d`RRN4469AQ5JuIm$}W!>zAgjnuI$)1K!U#(G2VSeDRyO3L5ZlA!4&K$?b?4+g&0_~#U?0TJtGD|8^?Cih`Rjenw?=)34{CjHdo;KN zEmp06kD=dl{r#-ECpDAUzqfUsX!oVDR*gFh{1N2H8+ZHyejQ@>v3GzZu0C3hcbKvE zvxoT$oLc)#Exokn)ylU7#4VOScU(VDd$x5iG?$-rR8;K^&O4bq`H0oao*n)LBGK@a>#4bKTv3zM0m%`uw)? zr{^H&%m4%Qt+j9LocXJ{J=fTy=J+jEkBRlX(ac;mYW@V*@B97moil{Kp!dJng!9zr z8tWc%?pqvwy#uklJ=#Xc`ljx1-ol)FCN@C50eL^KI`^fw=g6g-Sm)WNfzP)fR3qHC zVKew{Q2Rc8;A%gNb%T7D6WJC1$mkiZP*8}Ul;7>p4K}vfclBGPinyc z_Hb6!<vO7i;sAGcH*@Y`zq$F2RnuZw`5SrPt#t4c4EUod zyE*Uw@omlXXG?7d9&7%BGnNl}CfvZcPR=<(U;Wh>Kf!r}39r6m5AXf$<@cp}1Z8kf z>#>8_dsx%)?G@X%PK}@AQ!TNuw{gQ)uO{$4f8hJ)sC!f%YfY&4+}A$Vh@MNESaqI( zrse7#fe-Abf2GFxfyP+){+`G(U;mtenkV`WKVtRMI#;c|%ss4+{dr>TIg+;qekb(B z6(TOd!f%rfWG=!2j?!<$bD?icR%})soZ&wD|&zTxPy0-*H7Sy zSm@Y?eaJ_y&r(enUotKi!UU#k&%h}I*V_|LNdJ|)hpIOG%Bm_|7v0r_rcdW2q zu3gpWH#DCo*Bg7+nu)&N5rMVNX(ecF~cjnf6Y(BBi-B(S523K|TYQ**w z&h#^wAkc_OUi*gY+dtYtY~H;^qLVwZI`bZL5esAYADc5bL5o$N*ltYEjvxA*+Vj6S zxpZ247SAl>1orV*>zl~c&h(4zcP;q`ocBI|PvG~{nn5l>>6rNe*K*BE?Cs85fx08{ zlYB(0+4i)z^$qxWm_Pr$d+UY$#09Qb8?|cOZIF}mo-~)@=qC1GsA>J$~IhxdiT_m3Q{Q4`c8k=P_;~*B$h0e`0sj_jAyy6@y&I zsvYF4O(2)xVsB!Pb(PEY9rLX`YJJ1cLyda|y#HB6C#UAA&7H%48H2xo+Ow;Er~}oP ze8Ty?9@qx&8M*eGv3PU!eP+)_jq?(Xb@uY^_URjewZ#LigtjWdbQ zu}^)^^NT;6y#{^k?{~jp>$A=ES^K_~BTjUpgY!1t#~va;*XQ_FBi20Yg}v&Huu1&F zhQJ>Q2kdQ+(D$7Ji^iRNA8YL8DH~XHH~4{X-2=a&^}DWp0{dQ#IrhrH5peqEMiK-2 zb+`cT>-mFxY|V_7Q|G*P;%%oKINuHD+3VcrLZ41tv*h)eQx?}cHPamS3F>=8A2mJK z?~cIQ)8A3fkhiA#65B(~iJf6AKMqg5Jw_mtBd30Xt3C*P=h|D1s2JN~X%j2g5{er4 zvG#rc(wnbd&cs(cfp2^Pv9I>(e|Ms$8UBC_Ol^GE~X^C&ZI_neYcdTz@Y@Iz`@L>#^*r$FcCj7TAt@rG)>fZRT zojV%$a(D0SoQ8kyTJxxFx%y44GyYPuIjyzsaz1Bwa;`R#a!@hcKcKYC^;rLa>*~Q1 z{AtWCJ(f1MP3+88`#AH8t`XbAy#o8OYpt<|{k+d0CmJy4TTg5Q=bZ&;H?aP=J`f1^%RU$pjXTHga3+aFqX9qR9`-om#rudUzr_%HGgP$(~d ziyK(?G4684<~i2PfM|FQh_|AFn*j@oQvF7V{ zJbC#CnlP_cAHIPJ+`D3Uk3Ecqy&CopbNCa#J`*&~e}D$;T{~gD%Ny>xb7id)>@~2T zCzswi^7gw>KjiVf!;c*|j&r#ZI1^BwH{hHvmf{%XvzPs4wL4E$u4mUF%r_7FX> zv1$T;g7bFn;`h}0A6WbMXzl;|zWf1ij{#@xZEMx2HMU=x2Q`n~tdp0s&b;^O#vAT% z&^SS_zd_4Y^J$@KU-%YZ?=$e--yN=Zav4=jl*e~vbjdDs9+-g@WQvwCSk3xIVud%S>o`3`;%pYMsgtFfmw?p5yx zXD{pg9KFL$+;=a0bAx^l{KR_HJk-$Foc-$HXX0TyXz28v`>#GWv2S3lJ@oBY+FU!h z_r$N=V!fq3+{;t%K)tg9T;PvB;-}9XO(S1-zz(my z#(nkk&qj~hA;*sP5g+6p8`cNDdtI@{qTx@hGt}GXafj!~+wbHqzIazI%an0Rx8eAL>-(8()ojG?2te?%mdhDTZ z-Ob)uYbMrH{jutVJqI>(|Bk|{o!9{GXHE5#(2Mi)cWeUztapc*>yGyDxKnWf{|zm$r*JoO z_WTFmlezjetb5zrUA#}#(O2ttG|+gQ@#Cj_`Ec&@0B>*sw{NS?od>?V%-s88yP&sq zPjK#aK3DJfrsfI!`Aqa4d$^;zs-edS5@>>4^~JVW@de(UC$TkyoOSA*T_^c#?DwIc zGq1o|wIjL8YkxY<3Sj@5f!|}lEA_99nrRL+P&MS7Va|E}`H-<<@8=uVcfot2agRa% z^WeYGs6Fq|bH5W-t-TX?P7gKCjex}WO+@+3u%~w^o_M0MSHLywq;}v>eCNuk4XhZs zJ}*BrYdiQG-<)$Nwhp#>CJ}vV4t~|x58pfI=CmE1oUqS;^8|g5uk|zcJ#X@X9!`B~ zL2NWQb!XQg?!dg?`GK{Ev3ImbX_)WM{(YWr<-BGNo8SU>PU)>pm{_be1D&~sbxs27 z{QIj1xB_Rqv0YyT@Fa+}o4q>mpX#s!SesaPH+E0;^_j9u$CnRKpWRRPaLzmRjZ=KA zI&1zSujU1=_qBVFe-PXIjh$HY4Xeh_&3*k|&saJ8Ch)HAXngs$toIni=RUqqY2kb) z=ZSN##1}o+XX*}>m!tMZtk#}&w4mn1skY&YPoaRt`9%4)~S2m zS54K!sh{wEpY%Ka=bmz)I9zKQ@O}e4%sD^d@dNvm!=Ak|;P3mNUm-w&< zzdf|}x#IghP#bVp{01j)U2cs#ck&YqIBVW$o>;Z|_H5*y_;P=M0RniOpUW^c``#6Q$Wd2$W;)k>j3f3Q#O@q$9j@U|Y!bi7-9U$X0&{+@r6WIq+z2di zH$P`*j9fj<*;}1G#bf8I4!^?<5z9{e&o{#Not$gmiN+J*U-q2cv#MC0=&rkHr5VcjePBjJv$_R6Y2xYQK#>2zO}s@v38P^8|Lr$ z*z@--)~GY5Cdf~H`Z%ZK`#VHUf+up?c2{C{zd1W*4oQm2DWWZ z-Zv;ce3u8-*iUWUYsd9|zpg(Y8fU^=8{oS8G<=Uee6J_4=R*rBPwZ@S&Jq5OldHeq zyrTo&*WTv!8+8|a>*U?XoU=Od5cnzZ>tJW1L8mpZw)FC#k?Y&nHUU?2%*EDWJ-(xu zv1g_1>Ys_8nu&F{>K7!5g}tmB`rJ3L&YIW+Px^XC<102|>Hos_zUJL4;p?-(4z$*E ztemlag44sfLF{)+eqbx75BdRoyRVtfP^*=DkekHz^5_Sgn)BVV{|&8s)P8EfiAUgb z?*Z4qc}IEY1eoxZH^xt3y*Ycki}y8dK<#VC`e&C8H;JuRmmAi%T=5gX!=KNh!%t)2 zvDUX&FYi(7KE9#4kxO^pv%%R%UDe^cpE~!x-dFt%KLZW1ynjZO9&YBUS8M)va?Ve9 zbI#ItmV05P%-v=Mh~^*<;2W5~VdXlwssZP2=EEF&n{&^m z<=AV039g&(zD6qrx9xx zckbIaZ;i+L27-LVg8NMT4nKhNZ*Zk;%bIszYt(z6H}I&_H`mBDV*UD@v1;8l@tfet z70+9qyY>8<-^kBcZ>r9d`StI4Myz)`pL6wA#L;*5$n`!o2j3kZKt90b-fCBEkG%)B zcv1~t^jzN={RH0M9zw0w^SgbR>l;`z1OMCF+qi+c3tZsa>*%U4{D6O9pWt&>bmjv% zqhYPPpz;3odEr+~POT^4^wruk?t3S49o~C9fX7*p=z;x>J19%+nX&WuZ*nt{!g=5G z4sz~eFZB(3v)6tX@899Z&guIdc2{efJTM<(^ZNGkocr3#x%SUMo69!*23Ho&I_n2e zqrP&~Cb50w%y;77@cZ^J`tA;|)_HQq>dWG*x5p2296uLhwbl#kTda36ANcn3bI{ft z^lxCT=O(X~xIXLH?@ByU#n&G0;*0>b8?}Ds_O7|`{eXRnR|Ds9u5q{I&(Gdn zz!|&y0M`EiS76Q>cedA*)8@M8z&^;mftu=zzL^J}$n_lv>KYh8oxI0b{Z;#YSwGJi z3(i+_1HS?LnzLTN!KrD$_kIxz`}rPfynEvNo%sL*&bMs9zJprp_1|D7F-SQ1$`k89 z$anlO=G32yxjx&!%yG9K>$|Vs*qVXHp3nPwD|`CCX|hize*xalGh@{S+DtUW=Bfvr zJ)Aqq2{j_LvNve$oE+L9?g*R4cTLo8^Bzm zm-~Odw>okW>+>o-am~awc;CR<>#nOFn`1K%3?p#J<^b~r!l z4sYKJzIAH4V7Gzh#PxQ><{Gf48uRvP2{`6!|NmCFJ}>glH}{|}@E>s2PW*vAz0?hI zPjn4Ta-sqIT5sHeeFj!^{M4e6cPD4qqj+|0#P)B%`~|43yHMv2J(hlA&0j&H3F3ZV zzV(gRV~z0;v3%p8t`pn4-j3M$PhxBQbIiOoRkv-u-QL5V9n=owC%!vrC)WCUH{!EL zgHvO!`jV5=S5y7X2B(@QI-sIKR!KyQ7P1VKP$r|r#zlQIyR( zh*jr(KWEFKH78eZDJO_MwL`4g#?HKwt6lJ&*o_IrTL&WI3_+?dI9Ve9eNl zPXoDOFMG&2$DRYOW@9Ulb)Os1dNRK5;tW4~-^V%LH&WGCZ#DIPx%L^@hSq-ea)$}0 z9~=4~^l}&B=X(YAvVRxURSkaae0VfHmbPuH-W?{fbL}y(?veN&`;DX*I`hUI-`Mvt zXOBAANvy_x&I_Pz+gJN@zCI81(>yhqtG9n&RsUOac=hI;*U8(f>VC!IPxzMWyY6l; z@Ywj+*8CISH%n^lbH$erfnU4PGdJfu{Cp2}b+Fbt+xm&U5KnwH)_9Bu@XzcBH1b&a z)4P6`GSj~gPI79@7l=LHKXcg;OD8|zguU(C^}k{17x=~#-00awv^wV9H_6EvCvn@H zHTEfMp856OTjSr;IWKo^O2t(#?CkHV1ggi+v5qWtrKVx4UkuF?s0GRR=(GO zlNY{m?S${$%&D>d?EBtv@#e{SZ*2?yrRCTsW5w=ekB2&rd?DfStqH6%st+?7v9OQ% z1kUN$!>gahn|o4MJpu_(t5ts?H?au@2$8Gj!1}Y{=cC`T)}}tNt{qFCPAq2+&lT?c z|J9IpmYOCwk;Lq(Uah`=*4W=SF}@j-*E&;cA8XaUiS1{7Blg(O_mQ6>iLvztTQ`Uw z`W$;!Z7eH)_DVU>lNspj6X2!Ke&&s>FAcT&3AE#<}X;edvP*d}nvDCgg~n zr`Emf^Tu8_jQtKZ!7&%PcXQ_E)!Db1i)HQkg424uCXK+SrOuq3z2&`^b1&d#Zs(2l)#cWyeZZ*;^YG>eyuExcb5|hOzzaXI)dRmL z{I>FHE_V#@ZPnS!zCq4o{sl&?d-`+s{6-?zyc@Yr&f2QMx1W8C6WF`%j+GlAa@Bu) z!v~F=y!#E{&*uf$$M1>yX`W*bW3Br&tb53vW36Z8dN+Ode#pbA?XkZL*rV?|XHVZ$ zjXkW@PvFe@GY-D=Rz9eP5WRdCvNJ z)<@O-l1V&Pjc3&!z`Qd% z+I#d}aJCr4Hcz~x8~#_IxUabuEB7Mq zujzuA^PWKQ4z%T(`~DoxuikL1S28v)^^jPz+*jKjdaiTUof${$WB)7piFWPV_>>-?{E4Pw|~A3DRJQT=m}l8!+ZMLru=|BGW|Ih#a|F-KtqUArr|4)A4*DwFr+~3N7-u{R7_y5Pg{onrm|M|cE zlYjU3|MBnt$v^+6|IL5pR~b4VZK0d9U8(67S&szVvxdjLG`~>Wuhj*1+~1N^~Z9&N9&rU$M3)270LN z0VnXE*o{~Sdem8o4`9sPlX%&<&Q4-i>fXTAHXp0@LQEOC&L{V>^E1BPf&Jw;zhes} zWA$-I5xHhHbc;0`tIkuk>{GQn=e&XaqSjv4+h?NP-Ou7VySCr2ufV;l&wagj@*B3b z_Aw`?`e7Su>O>>v-^e?or;%^xw|~WdT+iDY`*<%ue6TP1zk-Mbd->mQyzQ92IrsU= zSnfL^cAoXr-e~G<5SOm=;@j^D?DLM+_YNjGMP5dk_2t~vxrYaJgS>II{H_+cePZsX znD_qVHV{Cb^RD!IkZai9Lk78DzTM}AoPA<*+*PTb`K{J^d(9vR?A6frwx9Vv6i4J070P~8RI%5*DPkcl5kweVh_WA^|0Pobef9~AjwcYD>ZzXaQ^xXZ4dp>8_ zvvTb6#II+7vWIm8SSL^ZK^K5pE@Ic*>I-0jW7|8cH^-T1AJumO7ZQth* z_+O2EfGxF0uJ32|?y(XNDTE4yOTg z)~l6FzpT4RIr?`HOLw$z?#wmAJ})?vxV#&7kkcN(9yd6F?Vl-jeBa;8j#>ZQw)~FW z@2lU$H)cPdO+P;R&+^b$HF?En*<$hStDJXW^T$PFOJnRP~7^Ulkyi-T?b`CMBsul6Kv z&Ki4ZE9Mk?+shhzg*Lf*2dU}Czy$7K{)N^b*lN#htmB(Eo*OM&$>+W4Z|;n>mo?{1 zNxtI`W9(+VzWWCD)ZcrZ-K!o>6Y%QJ;G8+e6+8|bZI1>c1Bvx_kdiMCmzc1grQEJyjgP_*B2?kIqkG5xvEAI(BoBcvy`#jC{Si9Wr>wV}u|Am~r?Psku^(tWkJo~Zp@{}hWceCg7zIt7DtTgstLBi1ww6p8~W9<0ElKP3h5$ndm!1r05 znvQ?Pe*=lF@A+#kv4$TT1D&7|@3H#3^Td~#Lg@nYno7_$Baj&FIX~E+6flBNpE|9Xo;YBxgPX+{pp=$XxI~6zfJ}#Iogg zpdWCwPaVufCU({0 zKIm2N7`qt1GUqGkx%P%q{r^nt|Nhu_Mec=vBmye|+PR&aeJ1wFz3Y?jz1N*+wGl6m8+=l-EUqL5UbBrS+*FM&{&);^*Q^6lgM;(b|@=nme2y%MhR|LbGbX%Rub>nbx+$K z`rLuqIqueyp7_OM{-E|Rz|Z0voHw+bhQ0x3In(Fj*;}lcrO8d8edc-=@65X5@a>`w!0&-@Q6;ugXKQ_k<%?dEzTQ0B1UXH{f}PLTmuVc^A0>$Q7Sh z%f%_5`oxzTXkmRlckG9CaIAUJYoLo~UnLhmV#V#LeIq}~HEipwN$jz;^OT-^xdZeA zo^w5)aqSy8VvY3|@^Z{M&ly3y`e3`S-+$h~sj1%j`~lpM-aa+ zZk-E%0Qu*f`Y&cCx)ZN+qPsq@ZvkcQ*qLkJ8N`?|&~M}i@wykzq-dLKAc4=XdcZw7 z@7kZIaw3=3_AVyAysy~$_78GhFr#eZO3w}W?qrXa0I}`?uj*pIi|7F+xj%ih&)*k6 z?^h6)KY?~bD?Nd`*bjSxD|Y40d1iFe4)l>&z06x*wZt+PFVOZ>>iHxFkoO`!0vDL` z9;=UMHMX_X+NEGhw@P z0e%Ac4Gi!`uI28_&dy1EWA<0%ja>KqW(>Q4Qdf+)J0*VcW990(5c~X$VEVk2k&Evf z_tLjt#yYdLf$jIwcWb~^9x+kpBYd9y-0{fO!(R5ZM*|)EXk+%WhtJa8)n|sap%1TW z;K_AXotLn-{SU|b4E#IK7gUdg2O<`VuMZ|LSJ=JfjSc%Z?8>9v$DABv?%L29Yj^)Q zSnlA@h|iwni-WFvLSO8oe`3*Uj_=5k>vugo&l6bVeCGzX&%~X)7k}0$#^rR-uvLz$SxH>Z^<%wnFM2xHyn8*co#P#Szo*}0 ztW`$vCD)AWjytdT?CtIYzc{h#?5l6?cHcV@(hoS5`(^d)_*XcQt5%M4I=)gp^ySC? zT%zUIePS#2sT#EMpr?I3`#bedkg;c`jMdYacQDb~4GduaH{$vY+dWQRdk~w(*-Kx@ z?O=jIt+|11+#UgpRo&(%y!t+HUp>rKJ}9w%`Aliq{zJ zT=I>a{j_UOw7Wmedshni2YtLF-wXR?tp3(%*BQ`t2k@rxWDo3tZI3{gPfRf`A{U%% zf4QCb>8U+|=d!0T*3j1c49cg@`P01j4JW~|2l1v4$GCHz*v3xWpY>@B)blk(-m|~V zVf(*f_<~(?@UJ`PKK5}=KQ=k*&80CB8+SBvo=qevAn*M57e>&J9_P6Za@Nb6XlvzY ziw@k?{U6rx4C}daAKsI_WAps?wv6EaM?4h8=wwYyZoAae{b|m#6-hAFGGE*vor+0q2a(slVq%PL6Tw zv>WzS9=S&B_|CAl&PTqU(~ikYY-icaccS{?Ti1YmdGF+ki!Zlw=9xXo59}|-!Nk6P zeCzd-@z~GM9KOAdyv!p#Rxi1yuXCn8bJc<0f!`yX^#Bprx%RTRes0|G8@4g`sPhmS zx%QmIjh%Da$DB3pkZ5c5-E-#R76;qBQk*-vSbJRH4s4O2_D^_l+}*$T^#;fN>|47r zPl+s5?}?xFhTmh&)(&j_1Pu&O&v$F@CA3dmKU>|`>81Xx!~P2A2?q9m_*k|a`?y15 z+s_)~6R4Gu>zNu0diYtDS9RzFK~7zFz3=q`SAO3deLowD=Xu*X^XBRqp&P!jp6r2E zG8Y=!ySxEsD)t(=>(#OA(Dt%Mkym|Vah>;YE^#@|iu*cS;`_bGzQX*U;SNw|h}FwG z2=n;XDj9nY{jD4N%m}o1*U`-w`O0D2Lwjo56FrDo-vay!AN!o4b|B{WP;>I?EcBe% z?pgO4YhTY|PhlVTp4bCh-SD5M23P5^cB*qFBY-{a-+;Xau$TUccWh%xF0t)ne>v6& z`^Y)-y5rMu3DdA)!^Ug>5PEafb{{~zs`tuzP2?JbaD1TSIv(3TI0|F#+lB07Vw2Vl=I#=tNKvev;z$~b`qNc+vkv32#pR?JNhv2b`3U!FN-~!Lge@ z{mr*rIGhVHb8`LnGR0i+RJp#dbY@TNy4%>hqR(^d#>|6I<0}!%9%!GVJsNgw+q`_G z$I55ofX_Jlz2N)IjVtyOwNI`)KKEI5aBH{wcQfH9@P5?+?4g|6x{KU7bQeEDK-}Lw z`&d8d(|~j89JrMurti7^oDK8Xop^$V--zp5>vM9K)5D&}?&Qli1_BIF zF>34`KwiemH{$Y?x_b>{iH?PrKA6NhvDy!NYJ=x}-Tfd|b#3L&XA$?a_k%t+pjbbs z9crMM*Y3nUcf-yN@5i3*@%(l7#5V6v6(fH3@Lseh`MO*501aH=S!XT;zS5H~0n|tj zK+gQYHfJnBe^5$^ew+$1*iIZMN?o$z1OYS+%ht-0c# z9`cPPxq3GA>i9z+ZJk;KvDeNg{u8^Q-BWGP!@KU7+B)-7Es&}~&sei7hkt|LbJch! zb)H`r|AFsJ&z#gL1KzG06?;A5D?JySTRWny3F0HLnI(Ss7fZWKf@ThmJaww6>(Ovl?L;f){k8D# z;16KmxUXkrp8#t6j$C)_z75C=V(#4GhylDWcec*IgJWKqvCisy;QTNS`p&z{g zE^Rr+-T$xKhEXwjVGgc#>c`H(eSp+Id2MiF6Wd*DCiWHE*y&~6%(cIJJb^jSW}RaE zC*Z6WtAEe6gZDGE+5Ki7`_w(j)iW`UT+v-S)(#yVB3F-wOnhsK%Wmcs_pUSAgS`9A zM8`tSBBpOY_Yl^M*fw=t8;Rh{?%5L?sB2$z0>!f{Gr#?w{|xe=3G#2?oXdOe#H$ap zoiBF)?{DOScjSyZOI!cGRKLe+K8U03CC?uAd{SS}0B!7ow%-l?4&IF8Pq^j-{W*iU zyH@WXSI(XfV)BHx``pkCU$MVDKRfpF&Mw%-C$Me;IV~3kwti{)b>h}2RU5?E@5~DP zCX#>9-laQK{!WT`H@(}$WX zz3lOh?eBNie$)k+AUA4$wA#h_FZd z%{6p!cTEN^_`Z)Zc?lY|)>(ge<_9QEe1PgrOrE*VJ&Xb6#PnwpV23Brw}gH**O*BXT95x%l?WSoIfp4core8*QC2&s2HL@{Q%{k!Z!58S8G{;|r*n z#Okc%Pk-a?Y!7RU-}Vfjo1cM_J72ZYGuK(26X2YuG@KrQxZESg5`^_=VZHlQoVs!1 zYIEKbw$fwG8t^7|=Bju1ngyp3Q?A6c)pfo*nmNkI_2*N5Z+{1#?}v}|EKWXS-GTLm z`77w&>*LXe%>UzmON-f%EGJXt{ZC9>Dj>I-z*Z zLCjfo?`ZoeBUXQR>}h2ABNxZKFnh>tNzE;tLr1)u`&}omS3^$R}DR@53wl+?OAI# z{S7(h8aeHuPpw)%0-t$7KC#`e5x4$G@p8qgFZDhrxg!>9l6QvBLjDYV^5-mP<{;O| z4S4bcyXETR_k+YA#+2B3`ULg_-B_YMKla%n%@I@Ft+<@4V)0Eo|90}k?d8rLDAg}9 z?jZPwxSR$&%b82$bC>D~-_KfyCwByOK%L+`zsOeS=Ii=1e|El6E6=+2v1T;<)Tie_ zYad;E;MdvE?tFG@NGBGt`1;zB8jFjy?#KJ0_5tMBW1{8d)@N_ut(NN@bB}Wm{#^6d z)&qWk)1&I_3s*TgpN&5gTdbPXM}Bd0I`q}r?U9kso^=BMk$%Mc6t2+;)>#qJ{KXSs_jLk)d_Ae08HAf#_`6y#O z7UyOk{4i#XzB&2GjFs0ePOjYbj!&*YtYKF@ac#w%GIPCaX-+|B?&pA3oHt|F{JL|* zl9=zeu+MMiK*R3j++(1hXno^y&wYD0zAMhGJmwMi9l5v>HRmgKGe=x*19Ah{Yv3Cn`0@@%El1S*XAZvbo+kd-$gMrE_SsXNv3$Ai z)i>vYwol*uiMF=ILZ}h<@VTp^}zt4PBf?{Iu{nbm{XFoyP7RWE$&d_U!z&kgr0j9g8TrU$GmZ8PPBJ!Z+j&0Jo-W0yt_~0H~bQ8edF@F(5J6G zm8#3dk@uih?dMn9nQ}&~nf4p_2_`wUdo;A%h}GYn&OXL2_)12p+QInD_3O5pgFo;m z+I}5`I$HUBt$7V!p8f6dWzUs6@DrS8t$ljeeR6hdtoIJIC-xuo@Xn3@jQ;66Urqyu zTea|N2HM;Jb;j*`{swaA@ai)pujiT-*bQ5Mf&p5rU#ottsNrEtg|<{`oJAU3--Ey zvyFLYFM4{{`Uz)%hVP#BZsN{dXOI)b&3mT5$|2W-7_eUWEQ_0~pLzEg;ARe8?|iJ< zP9N)?_rP|>%%zRl)7_1W>VtNMbpgE7j@ECvVgq}Et95AQyi4Uoef_Ml&Ux0=*{O36 z?b6Ou5(M~Iy*vKY-$-^`MOi|02PnDFIw*Df3#;2-8``8Ip1yfKU{YhaGLSl&p_K4 zFy?&)_Jg=`v>f9D$aBwt@sH1}Rq?ZCt*;_B?jtn=j7RslK<8NYw6Ic2G)ef?-d5YoFLJ;8{93^(5Z3 z*>miey*seR`~VHqGtfqiMSgovjoHWZRbT8LfWD7AIc&ulLh+8vkKGULXCmlv16SY~ zYc94w)66|U1DWe*`xDz2x8FcN zf$`c0yX3xlWa0_pGth{&r^epeGnWqZ8-4<9_Zx6Js6ND;W$e#zZlD9@fxc?zva4?_ z-x|+lAJ3YJkJR3=dilAkeo3B~@|BqVBi4@Ybe_NStIx;!48HIiJZmSs2Gp_R9eZqB zp7%A;gZO2~wHvnn75uj2L`&>lwcdAPf~vW%c*HuxB;K**3V9vaS37}mcNp5ly_bgl z#C`+by?zJk5$o&^avi8ekh^~D8T9M$3fsQO0mden`?nH)C?tWm9+L zya$7v_RHD?_{GIE?tc1T$-9SndH(rb+4SthoLBYa+@YhbHSRN4+fUA!oq-Q`60^5I zr|TZkp5bTUT=NM^(Aqx{e}aZB&JM;#tmi+8`<~Tq(43LxiOcbw4&dKCaP|c#b?*3E z-x)9S=m51lb?S~^{j@;z3T??a}KpV-N!!8 zc-y{x`kfx=D=~B4m9{xWEfUmO6CY?7>`#veHTGA#XOH#!DbpBv{i@&Eef)W9{)V=; z>gemdhObzs9pJnd+GpRBQ@2>Zk8+{k*<%nh-V+T1TVFA!Z!UoQCE7da8*=I{iPycL zt+&Sa&v&BoYU1Xc{}m`zk9`B?i-SFZ&)xdpz`$?0=DUY=2}%<`GaBL^?io97&j}P` zp50y|VuP4Fm~X^7`cEJGtUNi;1Z-H6F`$I`^7@vS?#jvweI z5Pz*^kGcc52G5u~S=&J5x*K^x z-u~tbs2=+qNlp&hUe5FW?Pra=ilYY{?|W?f>>R`zvEspThPHi;C$SD1m|&=*RgSNV zpX+;pXSTopttKUcT3}7>PToE*dfUs{pYz$>T;I9cAK%(1u?Ohb#bH0iSc^5&og?uP zXHMW5a%=nA=RB7y_VvE<5A+v&XW4&XEA}bLZ3o&t(bmYXnP}@0SmW*`^fBg~dKY4y zyAvDO)gRs9jUB_QxVyUl3F-hTk+$Upr$w!U%gItyn_;|e1^UO~B>|QlL@s(Jd;$zo6<=QFCF?-;j^Lxj2 z2jr|(dVGtyTL9;F^aHJb?%{>+tm;Ahirvr$)rnC*QMG?txtrxa6GaoB*{a+E^0- z5c}!RTqvK=l~@<#ulIEZYX&^;&AvC{fo*R&z8~h*avvY--{ zi+3a56CXIEp+~Geywmd0e+AB!XAe0OTXdlBPW2AKbDvHufH7;YCr?a%?M$rYdUkRW zxQDadt#*4_vkNH)7Uy5vy)sTc5_@eZ{su$lKp|Lr>6w zvFCkpE7#V&c+L>HYCC!W#o6{OJ6E5kX5R0_<;n3~xgK7$66cK#?Q;*tI(Wgik9GQW zzvhUUQ#@ab<)6=jd3SnR$Lu#XF?XpQ(2x829zk{GW-M+uCm(&p9xd0dCsumL+{-a_Kr3eiyrU&7_x-xp_fP zV#UF>KG72`Ui2OK{(9**;7(zTTq7<&&_5Z+_MLdc_Iu9%__3a~=aFk&*>`+I{s=Va zz+MGykxR=N+DP4bEN#u%+rG!HUgVS~v#h_8_j%Z7g5r{QhPjNjPuavu&`)xK_Ix+c zf?9m*6yL4Pb%)ORcOOQPShXYJV_Re1ee|^z^#sQ3na9Sh7`lYI#w^btxpHT3Ym(eF z4(cx8=?8K5wl2;4E}#c^SQGf>(Lqe^*fDt>H2wRkmmBy(vBtdq#D491>+CPw!MXvb z4vtKGT~gDv-2a@9zLV`Jy%J@eCMvj6ldsHZ|dCX9b~Rq#mBZ+#_HG6 z&Zr*5=Z;$=r=#67v;|!6!rsP}@?*_$Mu3K|9%#>@F1|K#_Y7j%>T}LXz7dL)Z|sI`U5ix{=)e~>3(gDNfIW>{+qa%-a@JKGE!TN+ z+`-=F27ZFNdtwPXxRTShM$W+2zY(AKo~hv%4?nPj*aJ*#Ym~^v(SK5V0nQ(>=Gx2I z#_eUi6KnE_C%!$@bKB=Ejkx=~fag{8?UlKH=4iix2WY`RKyiQfwXc}Dc6OFO!>u1c zk>@>pA^tO{y>jhS&yzi^Yry^kcvgAU3t!tEiie(XJMn4m$F)xyelOR&XxGjU=E)W9BZz<0&n zl9(`ec62{`s^uDYhtuD_H{wq@O7aiz_m1C*&w$VFa^%{_yzxMr(>FiRx#JC6DULG( zC-LDu?yHVSo!_0z*((Bj%N4GE^~m*Zo;#h!K?mAnqoFf*A9UR(`s4@p1QFlr6T1UB ziqE^=P4=9zaOO7r+6#_ztc#ED-g{yJypQi|ocTtqp%ru5a=Re+ztHD}j#%-AXxB62 zjwbel{KyqoTkEU=>c8m=Cw9DM=mUB3pYW6(t8YUKIf{P<=lL?pvty5y^G+J5UEH1c z3$C->NuGE8j_vmk-*Gu#w~tpnFZh$VI?%8Cn&WQ17jMM>Y7XDe;vb;?To7M*;J4g+ zZ2i;bxu13QK0!OrzGC~nUvLH-XRELGZ+v;~Y@aVc?(y9z$xYA&&!cCnynPz5=ENcv z&v<|vxr*&Q?DIVR$>-)ww7uoZ@foPyWuUL>Ty~lV0a`9D)+evd7W?^7cb@0b&Z-`0 z`<(mIHXgud83MaI+dg~0MobKK5(hqu)P`eD?asym|K!|FF?V`;NB+K7TkihbwwK(2zM!9S%^}9hM_cQ;2KsVe zeT``g``f#KYO!KBawQ*pO=25?wpRjc^@Y7Adi;uSj+qVp1;6FO72C6Q;%BGx9F2c+ zq1SHK-B&-XwMVS;^)#@g?lG|8CVeizod$4E_mJ=BvkC08L*-)mV{2!!1FlAJsSS=zG}|h4Pr^oK62dkMU7m~ z?QHWC$Wi`GY~;EJ>jRj#ra0)-zwf&>4)on+Vw-md?Xl5LoSb%XnHA2%H|FOc(cXu% z{)%=_^Bp*&B_O8QL%#Y+&bbpmfIZX$yvqj{b?!Q}c}0%%5^c=g{rm(m?I*U+@(sv! z5B(ST;bZ-p%(@2D)+pA^Tz!m-1Xt?S6L{a}98DosKp7AGkh6R0Yl}=sNco9D-x267 z)_$}{I)*bUS(fPZ{>#wSqhV{F9Y8q>CC zq62uwp7@~dl9-$ZPQ9O7`#5i)&Ffp=(NiB>;B{m0{Or~pn!~>Sim#pDIKC$h|G-`a z>J!`Zx}UwzS?w9b=H}Yd8vD!ldqLn&v^^$yYqaf=_{Ii)plb$p2gc;tzr{D-T*vp` zjJI6=IiJ3FWo^c~hk+eHZOwyx$9B)L?ep#q@^TZX#SApO_OZT)4PwqvyysKXiD@_V z=kv)m$9~pDG`^7kW$h==vvgzBJ%Rhl9b5Zs#08wWbzbw-<((bjJH!4@e0PwO_zgJk zYAnC`_IG}4ZO(edUiFO7C9!fn7ym`g6WG(=m$rws6C|*<0AJsJ?{EfM`GQ^fSh}Gj z_cK9jTceI>brxsV_hWi1Bi84#rQYHio4IN(@a@~3g?3lx8ml;chIa1Xi@2{{27cQ$ z-rE!V1|n7;V@l+To9nUj#LitiujY`O;M_^|hvy9KszsO0PIB$7&J)Kz?^^X?hae`W zZPZ@a@~$AU4;NO+%+*^*y94j3$LdG8Y~P(A)^eS3V1FTYb{NF1vHu(OFJK>gs^h+P zZ-^Z4P;DQzcazxmZ)y-p4sE_;|5qRDv*0e1xG288e%bpIyOAHng!MPH(01Rz&fN9H zjQbwdJmS?8PCtG7S1eh1Dn*1)HYt6#O~4kobX#66>NITM_7o#!-&1t__HyIc7q z_Fd7#&wg>?R1b2VuK~4l4?nQ`hBfv&BG1`!GuJbiX!-Ik^zy#s4Sc!cLEKp<=iN0p z_Bb5ttqI`w{fTzBL}z^4uOV`r`Mj^V&$k;Bd#0QQtWER)&iu^NTYC}iMF(kdIV8)?OX$*-{PW#?V)C?P9L( zK#fFvgTB2xG3WR!Ja^Tgl9+fBS1O10S<3Ml`0T9rx&5)@{#{V_AH>sq@@xA$T(sOP zw!IQt|6vZT_#P-TsRzY@Tl^if#?HjiJ2Kow~paHovu#pS? z2>6MER%WjH8$9FAvfjM8;$eqAs5?r0ImpPB>#=$_d^v&kxtOoB1@*hHvBdYD3dX@C zZf=lk8})8rzmut)+$VSd;}%vX*^d>PSZx$4%k(XZp3_c)+hcj2K3)R`RD*mAQ|g<%2WKi zx=P(McYWeBjS|l_SM5CveGtHV7y9m}MEGUrG3~3!FCO|rA7!9tpn+#ohc-3Uf7_XV z%e(G+<0Y})8+*CW%w50xcj7)D`zzkHwZ`R#HZ$F;YYX;ivFwSjRNg(4cx-4>TU>L* z%{LK?b7e-^#LNf2y51dPVI25pI}dg!4te7@VxG6S#3oRTow|sXPqeWF1G}N+SKb}* zJAMLX<~~!b`!Och`~cc=>W;SEnX8y{SAFkA=}8<6&;ozo%iMNHw_NY1=Slo?j=-*3 zwDD%oZI_>G$DfJYe~_E#8-DG9U7u5C_8o6XkG&5u_p$a!IQlAsrmSZmP6WuoK@e?t9_pSTDmN(_ZLSX9$en(qpZeY8^z@EUETHjot zMcMdjYfrq9douS5p7c36dH&i<-oy^jvFj|>;}84<4f}Vw=n}O1i{k%CH1@MWTjLyc z=IZM+@va8Gu}_J`D1t>iyyeg&|}-i70+X=;%Il938>YUr%rSO_Oy3B59~jEth^k> z^Sf^VW2aVr?Et3-`CPjWwDN-guqFXqInI-p2l(cjh*f`h1K-(w_i)d$iO*cN-y4QD z2%z|T5zhMpKS4|W3GyJZ`_}RYxWG}jAjkRqOxV}_+&F(mbk1Slx(jOrJahJF*aNt4 z;u{bA*fqxdJP7%xkI&uD?u1*|xrU$Mja&yod;n{&+wa&t7cbD)kHvF#nsc}N;y6#v z1cO`?O02!BCgH-fwmA6p^J4pfPG@Ost$T4nB&A|R%jJfZK?{>wuhdKMJC)#(ZMC}gu zZ}+6{>>jJO&P>d@nt}Gb)suMUvIlWFZR;l-V}3tSBKC7fSA8zudy%L7gnizb6LbF- z>-%$}Z*V5s_?vum;5*0XGPdu2O4aAu@hLwDqU&75uhjj4f27VxZS6tcS=NZecTQkC zr=jg%GjJ#NBzDg3eiwTaPhedK@|;s=F%IlC&=tdWKlfDZVXdE+x=(cF$@N^{?|zem zJ%QRgPBj>@&NFO$nWrssbiz=273r{PtanWSzZIqNc8dD!=9p( zbG8z(b~Wz#tUY@*eD6U&h!408l#jMnPFEx6nK(0n+Fj)h?Be1ZQ|u#8Idd}BtjB%L zv$kvJ;x}{&dIT1_sX>Zd_vO3v1^gLwVv5hE?(WQ2$C`KUH169^G&%RG`-SShlbGkc z5Ox{mUB9~f#=8i z-+*;u?&Xg18@4-UtoXpT##-%Z9Q$@(Jp=m7)xc$sM z!1w*+UV!_x4?T`kVvp`IJFUZtb)Cw_LuwH}Y!pCEDOdd}veCuqSp&?*6g9 zcev9eCO^Rw{0w|=e**R$+- zem{B#`cL9-+MH#g-N!x+p1oh#58(dx`!ia;GGo=*r>Bwi%$BwIa=rJOjrQ|mEOXV4 zSZ5D(Bj&7nKJeuH@bPV*jJ03&BB%E6y!-E)7wU%j&G8QH6JQcMJB0BhzGLbJJeNJ} z=~1IoPjT{aNYm->I5CwC(sRie;574?`voGl;auP<%(`_ zoMRu)Ci@Ut5zNo?c${U30XcuDXT*ekczEZ_e??YQIS z2XG(H-L-R};+PX(o*d8MUX>$ezcdHD7q!x3?R6toyW$5hKXTdn=Pb^2zO&?ePVI)i z5qGY2#)S8i(MmenIrf_yEmx0@9>AQo{XXx`7=5jO7{^Xy%$NHcv4P*P<+P2&_YM@} z@)JE03b2p+i4tS6;{R}-HG|xhe)f>}hF!am3w--rg&bmXj-1(T$6j#C#`ovd%+<@g zJxFryX}@!BYwBLm+Md1J3vrjq$Fc`n)LA&Y&&PAry%GPNI`^2^Kik8cJ+K4#Gdsv7 z@o7%brxVk6PyK;9&`yB##0>P=k@<4$ zBZ2d2@-~(E0;geeP}$mv6nb@~am*h|7P6Kk|xR2ZmT4;}4yt47Fd)(JwV<((9a)X@r=UyFKab`(;JFmH?Ja7RcmtAwn zeSb!K>ic=|a~t?)A9L>Yg`B-!;1AG=#~qyjo%nL=Zcx^nulmG~bq1{3*I7aA3CBGr zwj5`9o~}(ickBvJG45Fs1h#0{4Q>29C--z;^U=6o!l-Lv5)647TB@t zM!>;$M%nt<2Z8Tg`%iP|j?P?m^}=q%P9L>*;Q8F|^cvV_z8rg7_tWNF;M=bWq~^rp z86Vj8vd7duxgD4DPOf3Mx;eXnlDf%OxfXt8==)$WO30?%FzxPRHW&h5nXBi0Ub z#JPX<3!Zgy3$f;RW7q>+utOinZN!AOa}vmzxqf!oQ+w`Q!*19ju!nvuc*ZC3ck{#& zZ7+G=^>xQDcnE zt@`8t`0Y;Y?|z?LI1Oa3GrEWUlUVhn*T9!EK*#o6W#hZQaJM8^TtdacpP=@GW9;-% z+snIY=nFc~10*meysOx?&J5t$&1qXF-x;3kND%WZJ=UCtHm09oYfh_p?)Dq_9mtbY&z@ejEAdXu`*ZFCKa9aO{}cAlpw4>VJHc1vT+N~Fso3jDEYTB~ zvxmIH8QSz7UKA< z3pm)-3p+qbuKC(C_>%bWCt9(-ddGhE@s*09t$8xDo@=h%lY0A`3uC~28``@1`#01C zvAQQxJ=UK7etrY?ueCM5*_iN66+Es7wSkL+f zw>uNxn#}cUThHo!Rz0af?&s|%w$HO~?X!3%=NarJH-Iz!+}dX*F<|cO@pfN5Jag>_ z|BC$r1MZ16r1qdT$l2$5U%eVJd5xH;x*r!mWA(GAJ5&!fya3isv|_9{#EhHwjOqk4 zR&TXGqr3~>$$>Bbft}l)`Q7K@$JN8Vn}9RHz_zzB`@5TcYA5oY*i}BF?%MYS=SHj@ zMiRrm857$tP0BuR#N{}{Jv;Vm^R;Vkwk5>B0C^3J*zsKTavQdoz+L18cHGz5A{KWL zD~ZLOXmeNi)|LccK?lZDAL-aP&~x3py;3{Y_p!i!-dpWqZNm>WxST%3p7tPSZS9I4 z;CsF``iYJWBE_>;12fm`5pb}TdY?3M*~P)umt!yal}CH0ncV$-y;}oYbl}+=IzT;V z^m$$__AEKrsSo688?#opZ%6-zzJY-~0t#?O;M@NlyAu-$-tf2m0-_YdzkO|0qiC8z4w}x#HjmfEwJup?iZjY z7R2Nl8?k!%b1JZtxX^^6{wdRbsTz&2R#IN4Ol=D2CG2|*i99XAq zoxBTqp=2z3CY1ZKU*?E+V!~PW=xBSG(^n!vXe@Zg6Fb37`r+4mO1k5FM&*jOuY2^U z(E2qu_JdgF;t#a-a<1rx|1UrA_q{dZL2RIvpf*ASD6#JSMttV_^>s(v=Se^JbB_kz z)Wp44<2}~y1PVL$OUHto+ZOnT! z)_{Giv6nRiecAaN+C8g3xw_xY>-hFEme>QHwO4(#Gwf%NhL)o|$=7?pJ-m@S+?Tvu zd(40X+)LfI9owF{`VD;Vs^Y|)S->{tE|nu}@5km&a((-K+s^t4jLnfh$T`brY~J(Q z)BS_Ez13Y1Q(NO%&KVoNbz$BAN7frUNtz^YdI1Iq2;`rTV8MX{2@V`Mv++DTaNs}! z0Re#o2M!1%6bzFItO*s6$p!{&Sa8_*7HZgVV8DS52kLn|BC7jm3fs5b{q@&9JmUK* ztEQ)S7u$&+___I!a3H{iI)AUaVk>V?b!zY%yzh=@*51Hhp!(9gp|#FuVqa19I_I|DHDYN4d(Nica?R@A-5kB&@F%vsmXjD~ErUd#k%(J(u^Q@2WU(i0xl{FheIc=d79AlCOIuw%@=WSYPv? z`5V7_kq^+p4Sgrq1op`x^;?)dbjdR;`g1{eGXRLd`cDY!+tgGjR z#{3gZIF~i)bgt?(58- zoO|1AYn;j4zyDbLAoh9Jx#55Af9~o^-uXtX8QlNq>9X$ikCnIH{ter();N#0gVdsZ z!o7nx;EIIDH(z&VKe&MNoX${tqP0(ePG0{;ZsKbPae`V{W8a0J_fl(Z17l;x()o^c z{E(-gzW2~Jr|*10>^m^9LjHl?Ijt=nzWtqhZq8UV5#+EHqfO#TEFYll`U|#JkO$Qd z`?#-qdmUci^Hz?1gE)cnbo!`kBG-Ls)0i6T&Yl&c_kETN&~rZrdesWPCcqBt)OY2c zOFtF@>kMjM_nn8UId`u5;9tqBFXXy!dp@m$_x!@V{RYtYzgzJ;bA~ZU0bl3BcF=(O zD>|3Y%zVVs+gFXXUEn*ndG93PypNhej#ZkZZ>)aB5g(oVRWGd1(jN9Ko_G=`XdCX; z;fup>AadC^>xmQ8tng>Pj&Cl|d;^}%+Uld5t4?l!^E}R(Cwc1|R*m`Z^cXw#{4RAq ziy2!l$+A!6sHJ80|L#_Ev?D;zyYX>$!0`opEeG!QT!x(F>V_(#} zu2?n3H_&6na{38pZ6j{jZ^YK9w@<*ir+MR=A1+RepMvQ>fmqaT;}{9cmwX?vX6b`X0G!MY}{ARhA*Dv{j84d zYinWcU-uBJU&l^(5%>w5#bvI*zAf?Tc>%S8uUTHbSH4HyxB37BU+bE|7jUx{UE({# z$mN?GKx_ZV9Ye`=5TF+J)M{Vo>we{P*V}{GmAUTH_{!er_cHfz%~;xsQ)bOHV)d;) z_+@u4cdjM8AQlsJ;+m(e^9{JhJTCprm4AnO$ir2QJm*PcFo++_>-qjvhhJJWliUsN z0<;g{p4NZf4e`<0ziMO8N$k&qe={ds>9B#$cR<^S?c*NW+9%dM2AbLv-u~hnI^ToH z{rv)GjrVRZ@1nz9w#N7I1&T*k8jSV6!(X?~-tHB|);{m}aAWh<-Qexrv2VcnV9*OTv>Jx4Nb5&=Y#9#t_&)2X$7VkW@a`%kgSId6( zce!WJ)hDs`GVgnERTC?&J0VvKn}LT_O-pXxpE!*Moz7Bst{Lgyw?qiP*T=uVpRaa-v#do+`N!fy108ezA04;C-+D2Idmn@CW(F zec4I;E=O(x>o3IWy$5R=aar7|hqFiB=eF+UzvudHEi<2b?Ev1XJcoN|eO@ie0cTe8 zb@#e%O=;Q3J!(Jf<@nUe*}qZiGOoQ6Ul4!dkM&$1)MqSEQhUp(b6@9~z<9*I8{#H+ zV#4#j%elIJv)uK*$;Ix)-lzZPv!wqd?>jego$K~f5BHNlb5t+%!x^y& zPkOdk`Sz3EI%|AaBiH^(tj2xpGmWX2FHS3`rGH>6hd(!WrM~5U*Rbcka}Ry{`|k#x zSnK6F_SAUp3viWHmrL&qFLT7Tuz#8d8LQubYxvGRumPqTtlxjyx~Dz!B=!yl=->+4 zw(gtc6Wg&JYffHYo5X5r);qmG?L%&m8{`^(V4dGpy#gl}S2^NA{6KdH*Zy)5zT|qrFINulbEf)??RP$E7ey$lN z*7N;tO@M7Z+sWB~lFLMbZ%^N=>J4Wvd-z-vowb4A;JxR8jm^7<{i-MYxx+Xbdf%); zcX~C}J2me24Qt;061)L<_ox-{iHB|5I*aqxGsFJZk99|rd`IUDFZ&P&Nbn#&`G()% z1_(hsV(r^l^3LI&6V$@m+n+z4wI}I28T^TFY@IdU(FI=XvTgw9O>Boh&%?6@oa=Y; zE_+$u)#JNIyAuai{Iujwef7%yyFJ$0r+wHXvA=xab2q|SoOfdV*;9Q3%bK}*&e&&y z8~7cls~+~ubw3Gb-a7Yk8J{zT{aF#~d+l{!pBvs!)xkTNl|N z*Rd{b`9UvzZ1vXy>*cksy0ctX-{*9rU&fl<+6(dZP6?bEbH-==gBtHmt$FW2?Mo{U z_YM1hqHEXzQsjDW4}A3^ftsdu#o{OSrY`9x*lS{4@)zLuhdJ+k+(`r=HfPT_P|p{8 z1%q6I&-Ph32?(vow(y0cj4kq@|U;5U%K zReg^`7*Fe$CTZ#MZld@`arBBk;i!-}9=s*5#fv)@&mo?kiV$_z5Sk?m?Zl z!`t6k%mrv*1eTStG|o2hZ&=T2Z4ir7pSk97F8A)_2As2+v!*P3Ep@Kk`!GYt7qRV? zK0`lNz4zytoyQf}ItRU5>~mqQA6g*q%!!q=z81ce+C$D|FZoW*(f3>&U+Xz}C+eL2hR(AM;6DpF zyU)@dBeCFhCnUDcRriI)zV$p~)l6bBfIV_see1ihw|n5rHE`}weft1d-vMXRdS7zh z`xVTfjJWEw#=v}nui)*z_E+0LY2nl+c;W}vHDba2JFxfEk5=pK+kT4WtW!TgX|ZR9 zzs(c7W~|S=fOGwh^&U>%zL_Mj9h=yRy`ht@JA?BKzPDOA^Z)qZPJqwNzgueWs!>m^ zy&9O?>Whx(g)rp$fx$)ocKlIl9 zA0P14`hNX){DD8=JyT-8V_(2`O?bvYBRY9|lu!R4=lkNn>rvatscC@=?;OtK-2QH# zMxqPgeuUoJ7yra(6(Y@WV zrOTG+aGf~$4LNq}|f zN31!`HR8VG3wrl|0hfL2eYme0x&QRB^u?1`^L^JitLJtP^X^kH#{Xh&^9gtE{p+qV z|E3TA2L4WU?zf>&dYJpi$9krXWq!an`2Pjo{*5@~z~@ZpYHxa)yP$u4tn&`Fxt@o) z1imx=-rDdz)5vw7o!Ipb=lt$}&NOq)AF@8>gz&3FGSb4cW z?G4w#)H0Ln8~o|(yJxNe^VUt^I`6Z&8#VF+G;NS)R+s*pJ+ZE%ZS<~th^^h4AXaOg z?|^X^?C&1NuFO?u&OJ~3pw9)UORRgBhIj(~k^A{#8*zY%Kd^Pj%zmZLzBS{{CHdIC zwi8#L+5STR4&2{knmplfqM z-+O+NH&^_wtv;L6cWM1Rux12S5LX@>Nes-LvrT;Yz~0bw{JIm)Y|gXzKH5i3U=xis z^$wUfN37NwV|%=?L;gS>jX8TYk$`1i_2x2Gzf2PMwf_^F!+JJz0je*& z=d_<(7}KZn&}tLzY8_UczWoNa#nu|*Tl)_GsWDGY+xdYF;GSw-wMVR;J(pEqI&6}2 zsSBV+&RE}fqWTekWpC&5Gs*r}a;0TId%9+>T6>8G)OtVl%<~7A_|rUgU_FcV4SdCa zfjWm=EdPx@*N@GUm$%kl4{}=L5XAPHz;z)Wi393f%ysT*#_|(uy&7{TpTxy+)`2hV z-?5&{x#aBQb96=Q&k%fj{(Kf=XQ=OE^T`)_G^}-H@&5^D|B3&?4{R;$llmS2UwFQT z6%*9_)LQS)#&aL)^YeXitd<(kY~CLBt6KV-8|bVT*1Ctb6F2-?)EGak z-OJi5v31XXkh!SIDKo>H>|!jGuB=;Gu)Xm;nle^ z*WNw$=Yr>OPU`)fXdl3(4{pFdBNj}#>|>p)@(CB~Sp~KOYbU&YI=F(%wgcX{nP*vR zzp7O@G z+p(^Q&7m3i?p?jH30eU9XsxT2EBAnZnqzrpd-+b<#~Dgb+=a{?tD+c<;`qZh~%(Wi0Qi^Rb7yi2ZqEj@k)lopYZ3OD_)sbhIIiqw!_U{|0xH zCmtKutxaN=^L_`Xw|n+jHShRm4|`ebXJN|)=df?Wz1{g725MX*SC85!n1fuxHS*^4 zJ5V3`o`2s#iywWQYm$%MH^|k(S~u(mCb$AWm%a089&0aqxRv3%!q-$86&SHwD_IrTSda`EQWy2tb0AXi%9L9u%9{ibJ<7UKO6BmdB4tZ_SCwMcOdGyQL9E=$o~b7f1XLknmN#jwxO;Cdf>j! zQG3#3l6!-*R*m_Dd&dU(z_;FKs^74;jdNz-uZRU}8@R!_kD9t8IQw|ETG&6qgqyMU zJpHN%BJH`eV;kH6&MG&s?(JP`i^tkasF!vE(0exoRhI zfU6pL{3Pcr_G`p)wQ@Zp{fyPiKDCFICI3zz-%;nXul-zgzn(W%uIKW7FJGV*oOSMP zZ(+YPgR!gUn%7?ACm5R>z%%%ansKZi+P2|7nQM>svHxt2T)?@^*E7lWy!7pL&Z}<5 z`rfS^egMyDU+d3aY9ID9w>2j>h~-_ zu#XsU!ko)}jWhSpfe%PxVXkWhwco&H_n30Ia)Cc~>^tbP)}GciYw#<^s=2{cJvw`t zw}+@4e8Z1i?}0sB@7C?PX6~Q+*ogHU?Wx9oa%vLPUEbH;_Ency{Q>ktXMiVr4>;G!h5B6iM&4)O>=9Ua{kkJ!b>e0oowmWNYvAXzxX%sEdG6{;i;Y}o zNUZ;^N4|qWY~QX0&ZFjc_IUvJdj}C~Pi^UtCcZuVIWbdW-AC#1N38zZ30yTVcUbGb z&z?m1);s$Pc>bz`i&(kD>RYxv}G;v(8nT z#Dc56oc+Gw9P%xIKGs;xgV>XRdmfI$IK(dsC~9 z`1opFWnM8fw~E2pTHL#xMnWY{^YG~ zvEF?-q1Jx~mRNIDtM!cN2fV)dFg6D!kh|DZ;2Zn-S$-_t=_425%qee=1_E>;m&SnGP&Ux}D%>`{OdbvidccgkH1gJgGJ-|&}tiD=%x}O^7)&_Ry zlhbd&`N~!Dv=dur$EFcjVH4 zF$atryt&V_Lp6Bo5>#JeXFvBarS!m`2d05xsBD^`_u*T0N$TE zYySch|FbXoZj3#5H;Aw3>YUJ(XC`~ubAkjd0n0h`0~CMX?;ACrI&v)*>U`Ar&psVa zjgTK$&muo^{S28{b6WGx^@gSqR~uAimW~_cA*V*LtU7p{W1K9s|U+><iBcW0DJbW~~_-3*H(0^Vi?8Jtcpf z=}$TO+siq$Tx6R?L=SO_obyZ8O#(nH9*MNPr0sn312C=I& zSof*t#r{3ljLuTeo;U*0wsO{~oAIOngC6g|8PppeTfOL)@ZuXeIeQM!V)bg)*aPl~ zwMQp6fUE8&*L~j9#>V#5dX7PCjrt3I(@((cFY*CuCV2B{4(O|K+51ZTb?5#3ulpdc zUhTw&Kx@C^?^t4=@c`$04+i1tF%k=mt#3fiXxJzUtM?z-P@|b)}7c z2UtI=og?uBG+>{R|MurY%?ej{viCu~J|8g7WQ{uaaPLNM`750JpJmQ>_Fcs>t z!N=C)4`WtOzR<_Js`JP4gC2=r3mc&J&Gj6c9R38_K|Fyyge%i8`{y40?9m1J&p$s< z59cy(uS7fXzhkdpz}d%~Ft&#?w%9*=f5pXDC+9n&-$4?;6HlzmyUE0fHTS>|eD&I= zwSv{7Oz9Kfx`7Ss1nTR&`1bX=pL~bYuV=x$T_4U>T&}s~DyG(cF5{6nK+gM0e4p(g zZlG%PT>;2BOaG})z(;PLo>iv?B)KbE-$(%hLOnzL4raE}y`#I)M{eMM{(B+v)k7XX=%ePL<$MD_fcNTt)mzIMU77o= zSU*#o>jm7yvsv>Mf5z%_!|!k%`#(O`o@$?;f+0`ef0ldYJEmE7uH(W z2zm=N&S;MTJXgc2v%mhtZ}^E-=WLIit7iyjKFEs;{_p1RKlFbRcXIZy#@>IaVO`DX z-&an~KL6zdJ-ofF@$c@+*E7L?!uza#0nY9@pZaK=FYxUd#6jG#Q$JTfpOI%tIpA5< zYwb6|zx!Cv=Zel+ZHI5zMto-bztPz*;m6K(^7eC&zrhoz5&nJ~*&xBA4ges9bBO#Qz0d&KwIJJdJl9-SVY z8gU`6zQ*vb!?z%yE)tH~Bv#|ySmz%8_f5=u&WZJ|^qs+)riJgZ&xN1h&#kkUeFIF; zfcXLRo0c`v+0%aS;0n}ey@z_{@uSz5omW$HVC~fd3!dcvr~kb4No*hCxy);QF4nir z1%3xJkO2F6Ui}2}=3PA&FK4fMR`}-Ae!$#AAMR$14LJFuKkt;9*xJBXpIGxFR-Jsu z7fs}j_p=MQ4)>i}@8nY_&mERt-u#ZO$(_$!GY+hMoae#*@}8%9Qg_vtJ1e#ZYd@F$ zJ-fA@<-F(aW&H)z-sA_UXF<+A-_7lOV+Hv^u3@cjz?upA_RYx~yCU#8>sd)GPQK}z z$2Rurx%%1jH)H1a9rm3JIJM4Xe0t1is}`;M4=~AVjlH|l;&-uY50`S(nsd*#T|2(c8P`#Gdz|EYHAog%35!e?SKT{CVjf5sM#i4fuQISMp-u3@Vf0w?i?uMCN=g!K){qJadt~0!& z_b$|1ThEc)47MZ|7hulc1J8LAjqh3lds+VkuE6%t!ForoT=)DPeZ;Dn+rGnHfg1Z3 z&zysN6RaC~;TcK`S7(%qwU7CcBv@s0<<88VIK+L;;91|SXU6IgYwy|<-@Ow3+`c;B z)#?H7bA7iSoqLR}_1RTF^lBTZJ>A=5ajjTCBa+yDlemM^ zH*)#*E?z5#RU>b&u%0-HJ)6An`AqnRU%;Ga;(XrE6FlxapFcS3?d3jKH1>Bd>-6jV z_!sal7wz-imU{u~ybrYxY*6QZ@>=@`bqOXI;48db7u1!_R`Xl`PL1cXcA{x;?&VA^ zu)nVJ7$>n%uXQDU#817w_uTIi{8?3>Yu3N50S!1)z_$SRmAC#%?glPkl2h|L{=j$E zB<}c43r@ZIQ>V4RHBbD&s;_;q!@LN)z&W4V*xXAUXj-_MUn`i=UgqrM>aqIR{|iu`aBugu=WlTDK>c}Fs~&15 zUs(32{<;47wDyM&;+l_MHOJQZ{OxgStUpVgy!$0+xt^bAwcmgXP_t}njX8U2#o6y@ zd%y|qp!Uaq+*fb!OnsdTf9BHqJ`Hr$6Z`p&!#j_?x>}%Sf?Bb1KgZ^U`{eP_*}u!v zOW##{+CLWRuAGl0XI|S5aGl4#bq?k;7r-Syf%dv}Bi8%+=05bUvz$Gcr8we5=RIlF z)VZm-0N;%OEtfyAEjBQ&h#&WPdiLBL)|#*M_`FN@a4#`H1Apy3I(}@f&Pn=qU$fND zsDxmZCV%vNr#_51Pw8UqKd?SidH-G7>4~b&Y2IF**?aS>&img!m=D{@y*8)Ldh<^> z*N9awH$m-<#(QfZ1Iw~c?Vant(``@R(YTvs;hPt4`oHe|iK|D#fwNzWt;gzLsB!-@ zhq3wwBCvb)fCmG(j$Y1tZdhx+Q|~iVSK95XGvC1r&e-4M?Bl9u%XC%MmAiRJ9=eN<0uBc8FW`5+!xb)gmB6R~($vsc|?5CC(21~%*s8v?!)yGD|V z)yw_WP4ET3_TXId0cYP@_}1CydS7?gi68h4o4~#P@v-(fy4>1#a^_w3vH!V4Yki+) z{5UVn=U(bgeYcJ~=xDXpSg$p;$e;cY)7X@xz`8*yq#*@g)BwCu$L! zcgdX|_8%Z(S>qdd_f~JObC3F-RqgAi9{yR|R0nT;!%m=nfRnRN$b$~=a*hXZX{{4# z`yFsEE#b&nV{LBhIYh#BFyTvUOr5>2_hrouVxiV`#c#3lT4BHc=VR?9#QWt*7^YUs#{u_Gz2ll0J{?~?Z`Oc~D_m=PK z`F8;F-qjm;fsy;3u=3x5c2etpLkp~%Xae8or)`4eeOE4*HUkpKRWED+KjTv0di_}M z+TZxqi+BL%>iMImI(vzFmc*X*cHieF4ss%?t^t$m^$plZ>` zdsl;8fCTn($<_S@0lxjNXbsqcN zu;2CXD{lCaOXq&pCUAMS8LLLVfk3bSw2pksJwI{BcYitis#lxahgt79_vl_p&Xvg* zG!g6E)?4T5xn?U%oz`3j4P-7mu|basS7*lB^BeHbJ^eoP?A9l_fem;y)dTm6AMhi7 zaK4un!`WAjpS$*K>b6#$n8e=81Gtx(0>1U`>m9kzl*7->J>+9)>}8$S_p_rBo&w#XN#Ear`GZ6=iM==L&)&{h zeaNY)Ip79y%?4NJfa~~bKUb_8`xZ|eUDq#{r*y~E-g3TxCs)} z!pfZ*_1@!ou1E8i+yVLk=RL{T^Kj-S&$6d{fCl>Z2lhfA^PiuGJ7<6EJ4pA^=T4mE z+-JZCynSlMw)eoghs&I6Bwy}p4&!D|`kA*!U>mk!^-n+Z9K09tA$mQSZjsn zYFM>Z>`vee}by# z-sJ3S{|gx8?Bi_0+1kf?&d&G*iKZ6*>Ema;{O@q~s@g;a*RTmr&wA$ath!t5jhZ(w zfcl6voAuUywv~hMzljAA8kvgu4>Q&QE)g^9~2L z{_ci)^0i|1F}BB_=CH1re2F!Cy%+X!1^)qa+c)?d$Xt7vQ}4T1GjSfzqHljcm-T%% zuF|7X|J5A6XR_Z*%^l!-hQJnL^^iB7z@BRDYfpLm=ns7J?tNzIaGyG)>fgz$`5Rc{ zye(G$f$f6%>}CA~DTkF)llTMFS*Wk`w8b~_uHWR*8{hEV)489(WslN^!I}+keSQ(Y zZGD6!lHkjo`N3D<{^s1|%QA3z&&*oIZt z;z#T=bB3cGrHAUo z?vX(28ACso&hOR4m#ceXFV_g-)MLf|-|ybJnj3oL%2hx5d?W9y9lf~(fgjdkh5zm; zfcY0S6Rz)k$G7(td;y-zSgmX3%H1|^{}${+`~(-!;X3vK?46+cVC#HsEqe|*b@=w_ zK-=Idmuv6p%e~ph_xK%HR~qiRp}P{_fPdbWLEdLTZ0`WZ?r&UmX#9-q#ECBO8?ZJ& z16`ZD9)E%)t`*Ca2KPKy?I7nqfpsSRA&48++3PN}%$4v7ujT>Qz?uCoV^^#`mpe81 z#`dyC*vFi_duj)G!qx9!aFOs}!grv?k`e3yMktoOm0st!KjT;Aauyfb&f z`Dd(N_M6}eKd=dZ>g;RIp5}}XnmV{AxeKVz#QUm$Xvx{DYk}X->a{*Y>;3um(r1}# z|62WC*sq|+;Mcs7jQ+NJLldHP@lnYH}V5( z&ldYE^l&!k4>;>5uulI8+{0X*@xJ$^&R(7&u!Fps+8fU0Jl6CiN1utEv9#*s2G-g- z3w73Zh_#1*9#(&cULBnG7uZHz%lL5E^BnD~Mr)mS|IeM5vv*>hBaGQwltx}IvE-+g z<=)@yk?Tx>?XmLqb2*DT@5;Vvp7-^Ar9a^YF#ir}5!c*UbG}>F&Ro43@h>0i9JMEN z)-y%pz8%}py3QUCd*9c2p7`S2jeV_2P<7-k=HXo<_A|II-=4yAjqRU#wSm5RwRJvy zwf39%t`U66^?gyF0p{TXB)kaw1Mg-MH#MGh^BHRnbMiwA*RcUSuW`e#oOgpx*y{x* z`KlqO=0#1&5#R7@5!d^T{wp{=+@k|~xn@#Jjrpn#wgwH@)4YD)C>~C2aqKDQ+*+S| zJqJ0Kd!6^H;a3igwN;1ppJU$O0$lIwOl0dm)l?7Wvp#{#dCQLq>$CBjJ#yWTbqPBC z-2V$X&n#!3j$d=hv)-`}@H?=_6+5t=tH+P;JNsB?p8@pEU(i`+-gC%#M;Cn0)X4>$ z8h!H#8oqj$XELu&jmy~OK6U44)itcWy>GR0b)NgG8Sqn{I-h^T&#h~*V!jqU(D{^X?n5 zYHQ9NJC}J^-N|xGdm-mDu~$pZLF}?mf|{MUx}hm z@e{;G-}a3)j9qf- zBiDS5*z;&zSNyt%SkGDC+8}nB&x8vCF!38G9dXzHVqz*&xUZ?)Q5+WXQs{0>4c`44FH*1F%D`!I*P z0cu~)sJ>>#uU^Ei;?ZBIb9H+K&-lLj*tZkA)JCqn_UzN~tsmGHOXn;ud7);+dS8C| zSk`$G{zksOSIMh!@3!lL_zhp3{a^U@?(jmrxd7&!zj%H6&med7a8n-6*{x~#RfpE^ zj|TVoY|R1n4|-UuJ`RYT%{{FwW``T*+G5aSrbM>3}f#0!h<7kW%R8M;Q zSyA)jr}4&T9ojE&BiGD>*qk+L8Y?T5a4ExI#<`~;EftmdEO zo!40sUd`{s58#}y`7QPUF1Mk3|&d@<@ytei3TxvYG`JVfC8U0xM`twEKy#wq1 z69n*|xe{)GM(kcU5b)|7{M@Mb6#hY8&YV5UCoDbj8~J(;_z~+o&M1Z)u!s9xz!SKh z_aA%OYsiBU@W7tF8-AZUcgL?BTJ`py=;VAZf$aj_btAFr%=td6Gxr2q`@5XW*kz9f zjGa3`=HkoJ&%AxDo8%g(`@(g$jFng89)tY3FLTBdYn{Ccn2vwWCRcGTYmM{R(_9ia zWA^i0TF>a*Ux4dqwKE}t|Dd@6_2)gZ*B~e7K7p@Z=m)lA?bGqU*`Ih|y#qBh2eh*o zJF9CXUt)c)KjGfNFCXi@`H$En*K*an&mocBH&DM*QDdJ|KZ%9s)T-^+a|dc&k*n7TgOx9C`=zJW_u&bycQoX_;rpDN@~3isJE`8KYhIib#e`D?A*kk`pnhH zf9Bv`)QtF1t?)L2bjo!6#O}>>u!~$hD`s z4zBbnpZv>SaITp@w4T}J9Yo>)pZN%4P;>6^P^C!D-^C$4u1|Ds0Vx7B^IXBm)!*77PyABPUdFB&tG61L8vAyveUC2iGZycj zzUP6)^EzMkBv#jq|JXBQ2mDjs6{|M-|$`DYdyCh z_|NN|IdO*@`Qy2r-j6#qgeZ|2Ee?>}nf8nL|oOC7M@y`K2qgSiRpGm;qlzIo?0J{sR|`&>W|cvr5T zTI)MV5YeiRuVszvjO^`lj*O*`gwWqy@#^lYr+N3jfr;;2TJ>`FZp5D7`P}D$9dPD~ zL#sc5`G5b>swQ*I`;Pt!PT!8s*!)FH?C-X|rvtxMuDLp#`>AiS_BplA7;`^g&zEQ9L zvJYz}Ujh(2>!%)WfSRAWswc1hs+OEw)sa_McVg}iPfkwz7o1kDJp**&<4kiR4U(a1fAHTm^0G8TF94#2& zfnH>;pWh8@?mJw^Uae1*m`!cdbp1(_g+2LESY;Y?9sSK`EYOOU%)w>$$c+aKc`M@;C}_zjfOVRxn`_h zBY#+P@(pxg&fnwe?|9%lcVGv(03AQI#`tHsuX%BP#x(pt<$=%7S#Q9end=PplbhHX ztFLu~*mGH5Gr+mLiyqv=y+RuO$jhfOnm6jsvbTI`vBo3TyAk-+^PH2|IfJ|!XK?=k zZ~QdI_N|%t4c{eizXa-Y$Bp#`%~jn`ocft?tPyLU3h-O58A?wa#NN|8ux84ETK833 z{d4h^!?#b5HB-ZSme-!c+;@Gfz2z^U-Zw2;@xUKw0{DBt>Ej-6TW`)agBqYFw+-w} zVryB|Kh{k4KY91|fB&T=U}gGcSvBu_uZCYBXRp#h>3auBet=0{D%OD??daj0xxu@f$$wUs%UJj1@2AF{I6xDubACDZIJ~xlk;|r*SrRV5 z4Q+Ao)~VC_PBr|R2RHCtJ^4b;Guq>CY;p3$*1X9v7txbMI^zo%|sz&oc_eLVwuH)4IQEAD(jSMx~Zs(0^qcze0O zecb;+tUie+XtAFAfKwyX2QXL9J+a2F*ncL6^=?`&t@^~Td&9Snc|VKZ=v^yUKQ;DK zQ#JV73%YZ*vhc3M`Aq8$bIm?ul!`XCJxTXt~d+7C+#f z%Nl#AJ$-_b-wyyRfuO`3*8S5@@Xd3?Y{$qa6JmAhOudT60tuWLgw4LN(s#kLpf27bVKrUy_fe=~<~egZY6!#{T+ulCHGXwIFTnu_7Q7t!#| zH8Ak27HjX^y@K2m|MZ(j2Y*#_&Vp}UGr!M%_OrHT2=aS*Ch-N-S?`p_Pcv-aT4y-0 zzX;@_8VGY&ogtKA=gayu-4~t>Q4`A?c-ACQu8x! z?ddy1!S06*a51k>|IAf;c=t(i+5tRkSvc=M;ZD6hoLkQNj&)}FhOeC)mH&0=FAg88QuKj2H=sED` z=B+iSwXVL`-CMx9w8qwT(8PVurS6Xf_sm?nhP6hWbu|;VdXlRZt1htt)M<_D{&MRZ z&xAh3le@6bchEtAfo=jd?m5xaO1=VnHgL6uSQx((J6GMU9FQ{~p!D=B4RHd04>C7_ zR<3~&SXWT*xmjms??7Hyesf8S~)*o{5fT9X|NOCo^k`ffpgbH>-!CBznP?g zngFGPH#YyIPx;2b;v2JG*w3E$Bllct@U2O#eAfqYZ`M+|w0n_Q2XlP35r@ zF5w>R_YDMaMzsUzKimU9a@~z~0`1ZIv%wm9wenimNOOK1nIMXx+<~r71 z`q}{IM9(#w@0{m%=65j2={NFIADcjZ&C^&rj#z!=huoew;q}d1=Rao+eUO`TX6L%z zTP^oInhX9naC*3(IJFV07i{f&RlB)2`{csyjsQP^%f9N`hW&-lCgEP>{quxrey^E0 zvCiW6X~&myPkG~7`J?}cr9b`MU*9_Q=EZk<_{{9Gw$`RLG zaEV@xIdy?`PS4lmbN96#- zV+V0y->FZ0WADZDtF^bBIsL%aIS93Cyu%E9P-l_HpWsDJ%k{oBV(Z1#Uc~246t)p>Wt#xm#a87eoA3xp0 zuWJ_d6C`jibq%<`@P2$xpTId1457w-w8u|k)|p*>2iCacGjK$%9%^p%9@tYi10R?> zHJ-y2v0&_b?OgU#f9mz+-%lP%Ci~oV79Qj#eixi0u-4eqHDc$n_IG8#0k!%a4B)$J zKl|9zr5zxFef_&%zW_DQI|4ZQ0V4iwU(aS;2T87jhHre^dY_A&n6dV=w_4-63v$K* z-qgEd^%=Qr(9gOTw*HO@yqx!74{NIz@fA34fYi!Tb zZ6A95Jfqi4f>^qNAHcojJolNm`oMjs#$MHvJ)P4Ye_|6m(1+|;aacIdoZwK<8+j;HjGw#5Z!55LM*7r({+yn#OT!H|W`!e6bAore!Fv~V|aPcZOnpF2IdJA1r_gr~<{0&OWS?t#YORg6ATJhs&zx^Nb zxq5eEd%4%obGzr8x`X?g;ZI^`8q}M!pFP|^fqgHVC;B|AIkf}0oN)#Y@J?$VjDF9h z^}TTYPQNGq%(aI-oYkHWc+YcwzUwFX3;e);VSOK5H{zj>ja)Na@xK6j4d8kaJJ-zh zxn1{pU!1=C+bgif_P39|y`I1soW*_ArFmvn=X0}Ot+|HATo?o6zp$6>n{$>Ki*q02 zC;1NNnF6S5#9DdxPk6P}*BrGC?yTxbY~Frffjvg9_tP1wpS@Z3s(AtpIe$O@hE+G< z{5x=0koX-$u05^snKa@S{0IDRAFJ1mTH{YG@yMlr;MdILs}@u9&@0J#hm&09eg@7- zd?T+ufYur{J=VFj=B#(FF8*k>f}DTu^G42Po$IqluKvz8@$FMkkN*yAU-Pb0oi|7LGvV7>c#7GaIgV_<{4zdQNdI`Nt7 zN<4|xbo_y?`runreej)CP2AVJ<$mtp;Og#(t+S?~xslVirp}N5PdMNIj<3cZ=C!5A zH@4psoxNvbKagNMae(sK%l#(tjb4fExxX*qT}cmTlz(8qg21;&2hQM3<{PndPW(Ct zvDV&N`&W*9+i=DKAK>Rcv3m}6k8j?u5|EM z^3GtbJ+*;#_R_S?tAU^78wjALYROsi7uLHlXYAhk4c}hs-G35m-@y}@&sh4`eLd5J z8*nc5$FDy0svhL53vlMu8V7xP?)~vy53Ksv$9n%KzWZGNg$~prb~baa2>O7Q>#ocN zzH|JW8Z=LPU=wIVn@iiV@*Oza)Tc+iXV^Dt`>y}2;Vcc^*;C#-wa&eKcAtIluW)t8 zl(?TdSH!x%u1$EXGYqUfT{mk3C&9;h?e>sY+ueharVToXKvo}z&pp>d}(6oJdbrQd)42i!m1guX4DUSpY7PN#y#a-Y8!Cx z#Ci_FJOO79Y{&!Sv)^s^9qTYg(>7Jd?iX(w@Bozs2r{ox}--AhwUQbgZymK0pJ%eEj$vshR8# zuix`W5BIC*VA?#f^TpN-xVHUo^1F|BI?1WKzUSLWK^KIcyCQFA$) z2!x=+PjGsAUbWYc^}lUB&^T+o|N5?2^&_#I#o8+xXK~eeb7^|4dAz&CzoD5}dstWZ z%RDaox$KpR6N?+z)CWOLVC%lIE%$!R(TrI=*Ut!Nwx(|c))ncO)l5 z=k!d84dMYha5j6Wv5&ra;~^3qu&3vK0CVC>ZRX0W5%!zp0(d9ZIMX2Ks`+C5nQqIq zM`FDLdzX&bKL6#T)!ckt_v2Y|_4ZsNR(xve&ec$xg8T&E@MprajN5(17x*w1EH%!N zvF`kzdr!mk!bSe)dDip%pZ|K++Uxz8IkY`z zqo21qRSs*bSv1IH;%a4lW1iyzxKFKf-ycgJj@rUeG<9-)Ux=u&au#SM=GMEPvJSRr z6PLRpH$SNy&!{1mh^eP zU7N|FBPXIl%`mQoXAeaSqmF*D`K(@U*T*w)Icm0A#sZB-uvV$XH&@o~eb*XB;b-#b z%J-i!(|5GF#=PG{j|smc?*8NlaDNAKh&vB#(W3{4{kXTHn+bsPYh-dM`aNK0aeROOpO4UxH*TtrqIJg&6hd89AnMV2nE<|Uu02o>+vR#zYI&B! z{R?u3%;gpz{;aQ&^BH_~JwG*a@3-*fxsICVZsTX-!`U?WiMTxXr(QWL;&Oy~@&fqG z%X7zXi3hOGn7W-HHYaujTIRkc)VIFWV9jTE*30o;$T9BRbv*;)r6)PL=i)Z{z_sPB zv{AW4Pw$Lc0WxUdByGkzqOO@v`uyigwDeWdJAf!6agmVy5705#3E#L78gL-a8@PXR}p`}OeEaIJC{ z#LXS-_kCl{so}`EnA3aT1^LgN?MYrktPE;>cIwJeSN;)k*E}RHzXN4`)@g}bKZj$^ zVNT8g{9C|z%Y5i-xC@?xoM9j848wQ}f>>a4u9}{8fMkwXSsP!?BQ0wQ`EpY*$4{We zcW6(0fg8*R_-xfl#PT_w=i53Ltfju30E1Q2b<7>2;rhNW+rKqoF8Z2u7Hf&Qu6fZ4 zYCKalJABVW&RpAbie}V%H2hp5p=_N0cK2(&>1oV!+w4f18gS|v$8MG zC&<~uZT$I3-Mi~}CM`KRt}QS3K9TzR*I~$Ofcn1v)LUrj(K?5*9L#gyt3D`qz4sHg z11)HB)wAvgPne@;Uvkuzlk%CqxlBwOpab^q{17XSfJ4nD7q@66{BoYy2p|t%IQJZH z1J}&N99`60gKaXv<{tK^@3T1oK|T>Hz;A*5bWt_r+Ed%lJMwbn*;hk60qvZ4lPkB> z;F~MMHTKD))y3J})6eFt?ezvu&Sz&6GZ~8rMnSz#KJ0&uy-AvSoei zjHxHr`&u-p@pWJ7w0%yV{Fb?|2~kGWHHTc2i4yZH4&ippMLC_1 zxt`}DFW@wNEbZ6lx;W|AfbYn`T+oNLEq)TY_NCwSev2)6v`U|;t`TrDIqOR;CgIER z42N?x+(f@7d(w~GTjToF2>m1&`U`bV&b~(lG0$b_qtr3aLBW3Qv%{a^o9o~^eBQ&q zhdxEW4l#X4`K*B7RTs{@Xg)nZ6cQOy`RHqi5a(#1j-zDh3|Sn?f_e!uV=lF zAH=Xe=Zx>XSbBF)iMc#=wQ9;~L5H7#9OsUb%XO~rTETqa=!;T^wVw{_nb=1FpJ!_h zo3M?DGnngu-`z_5$=$z4LxqhxM$Opb7qH3RiA5cf+3<3LPknqbL!@8Cu2hV|A zGvqlV`&CO_YfGMa(I?jCe(j#8?i{~pVtr1TnEW<~gBfu-4RMG4-)1h)KFgXx0{Gvb zD(8S5&e~_Kj`!deP|x`he&kg8e3qC)En^Gp%Wj-~4{D-+-o5oXC3xx_!j<1JZv`=X z?C>+t_jM(21v0QV^Y)Oiw}{z8;LnI10{fMtO#$!^@aLdB3#`0!qd!+T@;y5tuVvm5 z&pjG!L(I_z|KF(_XMDM?ov{tpb4$Xxh+4BE=3ZJs%v=!D%00x-Soscn$oQ9bD zc3-0K$;mU9LD$3F!W#APBT^X+G5KvjHz?O(ZH}rl@Ll%++yTbS%U3sH{rql;CD1ar zKXC*aVne+gYgXnNO|EmTbu;2Qh&@>K1Ha74sVAqwcGxZ|FJXh2{S9$nJ8}c~o|E?b z*FM}tOwBx4d^K7i-*x1)J7VhudkwYJQ$b7HP#jsEF58v_B#K*L-_jaz?VkoWA= zZp;B|hk9e`wb(rPtQWxHKGjdz8GbIN%u8SLhz*>@K661#U;VN#=8iVkezXA`O|IM& zwT}{`U+3qC{0?NGZrK;TWqdV^mmJn%Pp%zN(PD)$d583FZ4!(&^9N`^(0hk1dH3U9 zs54)HiSQ4=DCb1J&7?sS{qx0t!^zoeB(Y{k%-W)t>gRFsjgiZP#h0tr5p}KpgLQ82 zEpdnWp*LeIdJXI_yoa50tZ~<{U+0dwHiP~8S?U_<1hGucdSmureaT^ssd2bp_w3!i z1N{4EYviiifkwz@6y3}`NP`xXk?ZS2p0(!7eL`!#f3~(-r}}VGKlcpEox?i1T-Q4F zqxO-B&+HF-fNq2B$T@$Sc#CbYrC;McZd)VBx5N^VBV4!iO3a*^);aw5e0fee6Q{Qh zE5bRo<{Eu+IqLd{#N1=U+>vu%AC+g!T6M~^H?Uvtw0jecFdyUuBf>vGu5119Nqw$I za9wW>PV3>!_AA$CE62RGfjwQ9d07XWqxSC_0gP)s=N8*CZ=6$qs1_q~)iR&)h1v}^ zML%b+?pLmTCaia2;1x?=4QmE&3$Mhma;!HeR_K@ublguzZqT&O9vkv<f9MWe{_yAzxmIz?XyXIX76>a@cDZIH1eb+Nb$UtijL3XV!zEUQQC^ zT2Ka7UV{x#)?j{q$n_a_SnG>kzy)*WDFb|UN)CGj%-gHELwR7BH<>VTIp&MDbNDT` z&Am1?&V#vJ^M`1=ujuE2FW=tf$aQ^f2kfzM@vSdD+%n7&xA!h;UJcO#>xX#4ca0#P zfLi7pIrsx;U2lb+eTH+aqm8qg!8h+{a?Lf-TjuVu#EE6&%&qfhFs-^nKfzqslXW@Q z_nh<3(=+LaI}Y&M3_ogJsnsV|;=OiaZq8WghyA;5?&FD^qv+)Rtgy~KQ*FoG5t%u$ zgmsv+599WBJ|5({9vPbi!+4v4798wTh85n?CRaEwdna#9ZeWdfSYtjXtvt`ox%o^i z2}To@vp_p?^#?6;@`hZ%aSZW@z6Sk-pV0U9VNC-veC_!W%spRuGwcGJiRFL=)GS&V z%dox$!*d_@>{%QD_v9HR{1!9;=*M-`Dz*3-f6z6a^4Xa+_!*dQfZ9TivA_;JtKIc+ zxx?AG_srbUQ2R8H7g5pRyWfO$9O1hsx#m0va~T*9;;z~CdHX1Pho(;J9PXew^z8g} zO?n;9Z9!apH9RvpY90`8dcUlTe@HCL2R^5cSnkmn&vkv*DtkaP$e*4Wl%fl-1BI_; zs3mU>tci>UaCinsAjtW9H^&h+=i(N=eb|?pTI(HZWajeQ0M0=fSmy!k%erEjJIont z0uHdweV6+zpSkPGD|!3`^c#JKXWC^P_*(Tv5O0{v3+A3z#-HQgGEdH$jFl&+AS$oJ z9ue!vS<`?-OpaDP^G`d^%#8ti*UHOUu#)4zI+tIDw&!_2?4#5M7JYMbumFjewUMhY zKZqZK!X@lkSmQ9(<^DXQ{=JO37T@Qsj?X+ZS4XbHoV87^vEdvK;FLl9HhzmOJw*N4 z+WHRP9<5D+F>sowxjD6Eugo3#IafW`PEmQr%3VC^HR;K-wU0Yb9z1lzh04o6E(UZ|tQ< zp6_6hu!rCK_|2bU;Uu1`hrNj&_LjSEzOnuMdT^7jf8W>M^t$J{{&Kmq@&B~`E7F6f z`+x7--u=X5zx?={Z~WNv+j)Irn=f_uU-|eiz4Zm0ggAKUF3ss3_~1t$edIk4q$Fl0 zE{yXx;zy6(`L4^=|M>YY{nV`wasxCdpnEm(OH~-yW!@Z_k_gma{N%QmN=f2*cm_6^KTp%H?sEasV;VRud$KiNr}Ck(Q#h=GF$tfQ5SoQ z_3iD^z4+M2KKSox#X|)bH*#}IV&hu3bN2gpU3{_c?(S{vtuMAN-_EP=#qos1a%*e- zI%lyOcd@&^u{C;`U45K4rzAEO*SVedet4pb&5i49$6f5LPu$L1 zANZj^o|ahs)eA-!e2>qfYuwJ+S6?`m*x0zv2jb01iQVgb2;BSk6J6}AZ(QS{9#2W^ zt&iVwp5u1@!$cQ*yW``$`tA4TNMdh&cXQ&6?^_>u-&d#K_|~4f+{NBHBID1Im;d-@ zAD)(2z5PX(9x4(Wi}4e=wf6A0rzK|J8cA$yZSGv>HM089i!XPvv9;V9zhC_2w^!eo z?qcnCCnR<^Hm>nXUA^k=Q5V}A*ZR7)_UZ|V?d5geJ8w=&EH}sR7gxXK+X zAAad*Bm8_jewkhUmNR?!Wn+o0iL-c)Z#h#E+Y?>9{%i05r|FHXy>+6C<;2zc;NSh$ zZ%@D2SKs&Y@kVwgz9_x@pCA9%>DxKGbxC4xvAMa}8sCf0e(_5`FfFmV^NNdIAhF!p z9lzx~bKieHF?|-ZZ_X}DY%R9=zIOS0=l%b7_{_A#>Qmn{lGxhV9^cMC`P1&B>8JVG z8C-ms%^qL7EV0cOl<~XP>OWrn>1m187ru8S!T0C!2jbU%@@H_bL!PE_Txzjejj%EYF+&yuy+4c7Zbk%xmx0QN@D5F&}r~_oR`;hraUlPfq^|e|mw zN$`v3%P;n;C61>g{N3?oiSxgVeZ_@enXVn4b6V6cN>jSfG`*!j7n;ICDF~XL&aRc-;&W!QY|ozTX-_NrgzQ{?&&7M+aN*_W&yCLAK6ve= zS9V{JedA3}@$83&*`YJfCH%tR{(CMCpSAd&Gq)jK9-SLqxcmIQ_l$_U;l<9x8HwdG`KRl z`|`zmzjZWx?meUX>ErghuZ-@!bosWs^9!%Oa_^<{caJW-{__3#-EWw^ZTP%Hht52| zE_#1{?ll*N&rcz@z4Fri=SE*cILh0Gd(ON7pF8G`%^jURdi>b@@e@an4?pJ2b5Pzh zxH7zL@mZtc$A0xT>GIny-20k~qq|>y@96UI zg%|1J;`1-wRHJj^=#k?`XNR9~vCjR8P40c>eo@N(;?eLWOT+zVUUW^yBl_YZDQ2QWM`E%RLG*ootJ93ReJyzN!l>}Yt@ z)aMtA#b>=~I46-gmP#BS4Noi$=Wn@&vo~MEJEm3P7M(aY+ZldA>i-Ly{!iNf7p3H1 zJQ}`YX&7#~J+n>o>5Lbq<(@qDx18+k?bB}0S5K>P?8uy}87`$7 zU(?jM%Nk#sYJA;j`1MP}Gf%5A?e=`bv>L~cAK`*_hToWKd{a~7tTnzl)%cdt@b0By z|7jb#etXVM%Y5v}k-6E~x#6q1J;h)&yvNji?$2}HG(7*R!_QiLj@Rcxs=;T}YnFx= zZ@E54Z@xbFK0}$Kcg)WX?@I%`wi)1(1H3LJfBk59|I%=9%k?>W)Abon&3$72M5i-9 z%v0{mP41DoucX{>7!AK|Y52e`%Q+@r&##=|jA72`-+;V;9ZoWQ$V_KQnqn$gB&&>~iA`S32n*knofWMWJ|Kw=+ zQ%l40mg_Tj)Ajl3X}OOcotr;4J3o9P<^J1E?zfry@1)#Mj)s4CX}EgJa^`NhK0h-p z^W3qc^PMBdhd-M#|Gg&j+s*vlK% z(Dm!HF)j04XYSaE6UT>Fy*>v=!(TFWhYQ5@`Q^o!K74*9m3Yr+_%D`*n`d6kkNMYr z+vRCrp#J5w8h$aEr~6+`-T$km`}f-Yucg%Q8x8;U(r{9De?9e$-+KPdw9H3m=Vs^U zJHua3ng4B*`Tb`8cPaC4jE4VyX}ERf6K`tz`cJLjoSJw3#EJR2&hWQV-v7|#{eXG@ zW6Jx%(eOVl4XZORy(#aspIQHTYK1%aWV>T__}i(%e`zZGjun14RrtNp@V_n%w{NL1 z?auuEvdL;d#dnzY&Gj4e$J6r89iP497(Yn< zq`5Kw!^D9*!;SgV#c{eZ|1%Z%v(fO+mxkAFx-qwW*ZGTSHI5&h>l|nQf0=s!Urp~H zw)g*?QvcOx_{(fM)1|rT%TE3*%+vB8yJPkUpPTu!wfluRe-2!#lFz_WD1R=r zu9ZIzZj?Wt*QI<9uQx61hOa>R$6%aVZ+>>}cxOJpO?zGMW8s>5dtt5janMxn1#qMM z<9S`mTfE+M1*U#6%3p}vw93bib>@%UF`Iva_PX*X!Znrm!CLu6&{X-waHISsye{SY zdA+6bl*^ZY5^mEfGw>Zpj?d?xti7)MDR52Y1F%+pDKu67RJc+8X}m7wZC-EL@l%&F z|8$H~>vfKI=8hcCKSO){voQZmSPS;+&$Ij>V)K!fe-?Zy{Mm4${Bw9+$`A2+(`CK+ zn^pcY?9*!V-TBBJow@vTwb!$L9$Yi)VOVGVd}u0pJKQLLIj>83hu52K@J-*h@*|k1 zRXo-?GI!+2Y(A^Ku6Pu#sW=B~#beM^@i^QlKf&u#KF{k-+r8miSbhh_Y4zq#@JV}g zHvacpy7HI6 zHI?s#wepujQ{`8}jq)$!btyl^>n)Y1-QX|BZCYhMgF469@mFZCD=)$|l~2Q3`Bl(V z`77Z@`B(9}l=pbOWyjZV@Ks3tfH};_i34{A&?X z<*$Pq5Bp7U&9G--9rl}{spPl7jqLaO{I+$et&uS@wduV+4SQ>{0{jq)Gjbtxb7TK}z%S~q;%%zqgDv{L*=arDS+C;t)c zb*UePYf8NZ)>1zPO{LxnH_CsU*QI=g*E4(5SWkWb%>M@3sfCX7zTJrBhQ^}u$8|D9i z*QI=o*WGnqznK3J?X*H(5x({1@6gW0e94`8uyx&hQuP`cKvQKSiuB&bfQ%kMS$ZZ0GeCMsGNp|AP8`c-=o3-TONJ_?h2* z{qNKHKg0S^{w`kAAF^vx{?BXeyVWi}_o4h3YqOOuk1h=Ii1-!1`@-jB_1~Q@JTSQX z+#6n#Z=jtv=+W5|{5fnszp9-ZA}m-0%Xiw#qPd5%>O02Y4!Z^%pb<`U)7%K{VUiT>b)0nx8AS8r+V*$8|D9+*QI>I>y^ci zeJKAoFo*7YYT)~xX~^S8W>4@3E!fznlLL_*C>y;YRuYcvDuPnawq5Q94KANeH9-lkX;fLM-!M`b*J$rYT{yv;P z3$gi?IDa;LD*7C_Q9i@#QvO_CpLwd;qj&hj=6N_Z1)dL^4y!J(2XVK+$H1oox516_ zkL7hK-^=Tj#m{^w|2UYBW~$la$7hf5Q{V;oTa-Ka@vs(cevfJ)?ymBM@Tur0z>V@x zaFuh)-4c9E4Qi5Zow#8LvzE=kj`G@#PQYp9k~N%*EeL@0h3RVfV~9^x zIF67ioPZnU^SmzQckp^;@z_K87r=ZpbIl&%%MV}az7T(_^7Hy6tX0#`>n}oTeqMhu zY%2K*xKSQ>UCI}Dea4|?kI(bBu}wo++AFlr^ z_Ny<9&Rx!5RqH?WUY}q4vRC}kpTF-nU;L)edo=$_^!0x*zChNm(&>x8^o?)+?apIg zP#0~!te(Gc&$&0|JlpZG@>XBUqy&Mh9haOK>U`b*UHFH5604ld+hOC!yH zimxrVpF4l}X> z_HzC20`hODE#JuNNB$Q8`8VNm@gx5C_xvm_Us?aYd*i<&xc-g)n``?=@!!JpZ<*>j z?fdiHQ)^!T+S{+~H~-s)oBz<7pTp(iNBnQY`KzZ?x$)1N_tf^AC!7A~$n#S@r%iTY zYR&7DeQRyMd9qvnD)$;(E`G%SY?)u2Qsu@!d)-^xZ=P)WpQ-Mf>N#z)*G{c@eX>in z{pQJT`D4xNaJl#q|BG7w`YBaz{A`ZN#J)*Gj}<(>6L9PZ2b5_r=e z!EXL95#Ro_5;wlFzN5C@IMDRB)q_*~rVaGalxo)p`p(*V<3P8(p}q@;`#QG{bn_pk zi>H;i@%{AOwe`kX!FZ=&TXe$xi}zA4qN5A^-Db=sf6egMmTV<+(+nBDS6tHZl>CDGrc-6w0ySrb0*WF)p>dRkwSN`LuuUxskxA^=$x4-IJ z{%r10mWkdo{u{(k$-d;Ke}nk)<{$4h|FG_r-B;Z5&k=9>w}~H(K#vRd^T|K=Q*Hma zBR*0;N2m9b{yE+&yD$HL+t2?`_0vDq`)k!NUg7-r-S^XP|Lf1+{9}T@Uj1LR{#X3| zCoTMctIvO=`D@ibcs0wyLYCbHJpksw>PD88A|+0=EZf32Lnpr<{W5YzoO!Pz2V{M;+1Dz_u-j9GkFZ`uSNBF9b7@~! zkG;Q;?1#y*iLL`(f$r}z6ZV2weSQ`F<6s+9=!~_#&lp;U1Zx)X=Cei{?11$J{_SuV zXZV2Kes}d5@;>}ef#Mg_K2F2$1z98x>TZ4#x(%)HFVVMI%D&(69PvG!+A)<6qX z#0(bLE&4wI<_hS34*o9CMt0xVfHg~Ct`2gY1G_#YvS;EM%z%h)M?I*Nv1jm7^Qred z-JvrLr0N@68~ZPy^WM1=cQgivkln{J)P3q7pgwJQbE=CW-Cy%4_o2q`o}czA5`^tZ za#^FJ9l64$e<|Qspc&|Iq_gy#J=?k$^qy0JpSit575rIfKwn|E?w+gUpzlri^YFeu z9t6&Ks+;O$#Fb4R`&q-c4JI4ujFboP1wK8fcf`ITvb9#>iwKC#kezRWPCw7`F<>rj z^5zp&Ku2%FeVeO~-rR*0U4hPfWxw_Qm9f1;8}*#syyn}Ft`BdXN8ktOjFE4ntB@za zGgaOJ5nf%0DVKT9bd!%g1o#-p!}I;*$pSz>2A+qrd$*%{9(xGJ0edNdPaoMhb>2JU zw?G7*T^m{j&JUY*@BIw=d5{6;-v{;n#h7Muj;x9C{`t6@`d&cS&~1SMXaVEL(E5E4 zQm%k(Z~FNdt8ZI5rX6ikvjyzYdFz?5SMcubsU0KRQyKU*2><=vG|wu$@xonY*dsU$ zKfu3$FJs!T`>`ME`1}Yy=YhUE*g`-F?7M`n}w3hJFHdSL*BcGUs5=t%Wp?JyzK4X&t=N7<4;8Tb&nwAA7ZC31UchT6+e13+X(} zeJ6eeG@$RpJ5vV6`tpRp3;tD1J2-{UsGIyBgtzZ^0q=}xd`sjR;{8l>h64Qn8nCTFE9~~;JjN;;YaM%t&>preec=l*#=g5J zcc9+{Y7cL|di@?clTWLg&Y_KcAN(Gm{~7H2;l*3w4?)#ib2onv{R~j&d;B~IxwA1Jc z=)>sj|15BRYn6eXpqF-e{5r%P>ZYF~0oj^;5a>nOc^}8$r-ZLU^S;HWr3?+|GJK3V z2;B?Tux$eKd9D+vXKcO#U@(jK^>SOxsGTuBb^v+_e z^&)(MUO)LB`fXs|Iefj3=G5mTa?PEC?xda{T^6YaT|-yrN4AFf zCm=RO38E)>#JuUuh;b(tE|4=CJNumtA4=49)}V*8bSr*xZwQcNXtm zg+71|e2K5~+l%$={Sox;(Dz~l-oF9*F|e-ugXk)Fb9x^S19RB>7WOvqJu^o3?ADNd z-xL8rOQ_F5c=J{v`TKpbhwNEqQ2Y4}bO##!HDqJ#+4tZwato9()d}$4d+XOvxZbNa zzax5wSK#~b?$2D_cjw7K0q-vBeRY$~oWg#L-2n%%d7qrO1TjF7-9u?Sa)G{$&OJ)) zGoU|@y@Y1Kx@VEymGP(HEA8vF0n_ z?YR%H-WdaOofDn=>>-y>?S4n9V&ZquJFC4H&?ljXfj!>|y#zd`44eCW1il9^?6V}_ z1il6Q&YE@8oeSvo+1f~FI)kkbZy&}vXZ;{&G1WIFqPI8a_TK1Qq3__g+~n|QaT~ue zD6qA$Wl-<7H3R%W9gp`|-nwN>{aVkMb#M^7GdgI>dl0XRcP3VQGraF3CkK1EDB z(;m==kaqu=wXp5(d>gx;Q|7b>eYe3Hyg3H&&Lr)ltM3E3Ds1NV{basJ$YHL!SL`J^ zbGaAe3aE1(YHFQ=PLMP7=IKN4MAruTTGPkA%xUZnc!Z{~;vU zV+;RXeIMJM$Q6ELcw-~JEvT|*Q({x+Yy;$CBmJz;cq;Vz*>?}>qmMh0W+A;7)cYQ7 zqYLUcQjF*1JS956^Sl2TQ?6xn&SP$E0K`8J__bO99bKGrRPappbTWlrQLz@~oc@vYAuzJ#|QYeuNKtWyYN z=eACU9AkPfc`nXve*1KPCAqZs;7=1{4fmo?#MXiKkcGaLI(X->UJGO~?cUx3SSL?w zBjwHTkKh{mz3BVU4EUaWJE(bl13sf}v#XeXXRGfV`V3U~ROl;W+xV>`XASH@*wYr& zdG_P87gWRs{Q6)EL}bs_yDx7~@}5m8h_$Da-E^l`u(v=TU4_4AQlB|;U6&Z;eckl8 zYkMxSS#S4q&wWPX9s%xdcV7jv@7VxKayyrMvX=tcS)6%{yaL~8a0Y+R(LCy<#tZL* zJ=mvxOn`klqtyHE<99{-sY0q{PVE8NeL2%(=z73?sCRGXcCJ(1G-H9S2L@mh?7>#y zYhT94NGz}y=Xe`B&%r!BXdiv4J)ncuHu#9%T@~;F-g(`lx(uDU{O;_TE06f*N$e)nJ!DXG?U#qz ztFuS=65Wh?XGP~36xu*vijO+`)oyL)FTn)ZpJ!3wmjU}4BgtrB0Y7--BK8i{yZUZmTm{_M7I+IfXSCK{ zFw;$QnzJHL0W$oIk@pT-!|&fgJG{S#dk5>Cs>gQ>-V1MEcS8NyP{5bKS?tFp0hL@7wUknJ<76`5>{@tdRG1lfQRi4D7KYM+8#Oa0gwT zFD6~Pn`(M)?k2-8>I1bu`>}2bb(TIlZ56UTRPfSkz&_@&Th}?Yt-uf93*esIK}28I z!X`Dg1I(}8-0eoE-8^k--CCZ*W&>yT zck(r82HNPIZ3Q@&In>wh3+M>lSzs;m_Mv_B?l?dPP~QW;4t*!E?+&O!^7q``LqOJV z0Po#?2fp4L>mNk#%<}eW9p?_9o{#sawh6G8_6D#zI?vBNd#?HxA-(tXn*ir2 z(e<(UtU-6c81%r(M(TY4^}gDhk2UPY-5&-8He<}Q+d~!5<8zvrjy8Nek6KS%=Zolf zfVm3%_0!(McjjvPnK?ig)T8rEoGaoV@w1PB-p{`te1@(B*0Oi|=|MAS3%Cn=a<9_e z`7?61(TVyqi5wktJ>auLoH^~w^BP0-4KejzLEoY;pBZJNx9-^liLQ$B6h^NhB@lh_LKXg>h&+!;Rl*|%pk0VU{T_h-SQU&Rzdu#i44c55HR z{)b>M7=Vb6`|$2P3_AGRi=T6my5{hVdg!b*Y3vpH8GOo^?o6RCa)E50_U%40@2>52(+voAz(NzNd|;=h6p%_4)_=0`f!| zpSmXg5nBl&I{O@hWqb-~hR#0auYql#zc~v*{2VZT2U~#lyUE8K-rGEddS)H8GL1N38b z_V@tyHval#%9ld*4AekFnPE z?yLdluXi3Ae|+5G4nC{Ui^ONx-UYRn`i^0X__d+tY5`}c_p^|89FR-myysHiGn?o# z;;e0+9yFp`1uO6a(190`n8RJ-fczL;CE)uDX*VTwT^%`{-#ngGiL74>yEE!zOabhx z4edkE;%hISleK#23h;jP8S#B|es%_U^HflIds)7kKHK*|J#Y7MsGHUAqW~ZAQP+l6 zA$`WXPd`%*0_UdcW8`L+3WX`_p$B>@iPB{i)0Fah?vmeRSX}{0roX zJigZ3fp@2#skSyK&^wPb!?vQ1`&fi@4A;X;iLXBV!SsMm_sM;&UT;^ppUipqH{le=s_^h2F+qRgCeBdJNRD17SQ|n9@2Ie zUuW|kM|9fz$Q4w53%CP+kGc!=;Agt&?<*DY{e`T)D;;#+f6r|hpMv-;+^7BeUU5&GV7G@o z>hN7bR}yy^`4Vu~89L+L>F$m_A9oP()fQvZ2ky@5?}8aN?^>|{3ku{6**mMer;9EU zi@$yNezu;wk=n0)D)*rWh%??8sry*LX5Sfh_tJ;2b?|4fTi^Q-Lds>n3iKLGHmDE3 zijT8zfhXZhYK-B-MyjoUgx}ju=NYiK8`~IK|2)B`4eZ4|>uV1CuRs9?=slaBGW-F2 zN_sP&d(7m)Vf5}Xq90#HuRn1Cs*ioP&$RST*BW? ze6RXAhxTny=g=pn8GK*t2U2qvA^EJqx0Q));h%xG;p1bDtc#7=v}cid(VanOZv)^i ztf8-SkCDw6fj*Ct%YNEWbDPJ#43Irv@8Th771G~k?XQEs`VMFVd$jH*ys@53gx|*3 zy(~k0Kd*xifq?x^sPRG{&q{h8y=QX(>NAgifL!k`rtb=4YX*WchKHSeC^wwAH&Cf)Z1sB1DkgJ*W{t*(WiqPKo8h!MK1Sa{urR}bv^hs z{PkT219ir97htZy!91R3!KwgEcx+LwJ~pss~HhTnwF zbg4(q0#xYCJp=zJFgB2m5?x>ZbeOf!_vC;P3lDTirh)#+AfBir)Nx3e4k7 z5xu!4$kIM>`maGV?C#3``XGW9`jf`RX20FEt2TOfU_S*m`;nIDvH*~+*#Y)7K(^lk zc?Vdxz^?~ytTFb}My}xXdH8CY;dGZe@O^aJ>$|oAjp;kL@!qwA`1Ihd?KxdSE};Qj zg|EnKA8`RdSHT?mO%TZC69T|)->1>*YyZp8qOob0A8%y!jP^4?SE1X}O?#Oj@9ymy zyg4#x2fJ~xk-d8*`arCA)E&5gANiJi^S*NjF(4CT@8(6y)0LHC$Q=bL)`o0r$9+;y<=e;$q2Q|jN2hche8j%AS1MiP_ z+?scFQ!jOG=VSAyASNoUfgp={2nlYKZ8#jTR}`4`5L@2+QwiT7r?b zclY5_g|wT2G4SRsusM^vdJELMzuDMZ*t~1z$wXj2=*mLMyAAIy?0*Yg71PhR$B0=& zZUJ}d96O*7R)PJxqY1EIQIfkLzJQv)l*j)4lU)!PP zJ=w$uWOMX@{o0GMGuZWSfeIL}k8_34{Oln27{5!v^U2T$?DnhwBj^Um_E&&)`1x+q zxflPIAnQky!W+a7WAN2B1Q(-FR*KEMxbsccAt8 zU=JaE?{L>^_!Pw1TOWRm&V5@m!_VF`bS1p`+|@F)j46-xt!tcjd8SJ~c4@EfW6=1n zLo;&N!yJA)*jn1qg_zF0g16pX*yS_y`bBhy&`sbo^ac7p@&FXbJ{^1lc5PDqOXx+Y zXWj$OR08|oLe7Bw>fb}>EEA|T?Y|A|p$4|YAcU05dis>e18B98;>>rgv3bV|>@j8p zruT|D`#@WX&K|6tfiZlD#Bg3Ma4+_Dn1Cw9#3AjD80k=#q*y#Bs*7X6c~2 z#~$3_K6C}V_Eo6<)-!$?YG3ZBLhc|7YxaS6x+lz@jZmShdw~%_XE%6^i_O0H_xUAI`~+_J0Oi9+kk>GC_x}! ztC96Fhj*(@%qn`%Xb1fsWS73jS$<}|-C z?}eX%FR|JCN%(E}2{HPcv%YswcVgf6)gsQFc-E`f0_dY}qjOKbSCz}q{NCT_`_MZ< z#Aa^?yD5kD?AbGuw~qBQXaz2zxBgz}8StyXIsu($Q9w)J4&_5gXA+_I-{~g3J1L2q zBi`Beq4%8hi^Ls3b{FoT1@sB%*EZNAG43be>)btb5#IXlU<=q&r-8knB_;#DzuZYg zXWkYxBi5evm!Ci<*!sX)`be)q?QMX5rki{=;VX3ZX8jd>4&iHksWCo1XdxETdG+Cq z%Ybw0?=14>cYh^1@9-K3$lmV(yiXgs18+U|Y9HpxLYmJ!6?_5Jr-XNh-mNvLeiNv? zyLT_vEuaOq4txZC<&CtzUZXqPP5JJ^=HLD^?C#b(?|GW51ur7GGGO1L0s%N{(6*5D zr@Qc=1;*$m`1FA>KG)FQh3yciLfXImtO4r>DHy{S*u4XN&;c=~&mr%I`PP83caqDU znXf?hXJa=5#Qu7&R@>=pbre86r#>sjCTqr7!Jn>M-(-xhQp+a93a zGq?A-F1hfv2YWPk2DSHg1AEY3LakqfW@f18;P0~TE2xKB%b(HqTa zuY&J&(>*BBd0#r<8rT5^IS*sEb_Kt-klvq%@I4Id(|mrv@O?egg~I#!)WfEq`p3}Q zdq+8>?}qBOJOxt}|=LHiBA39X2AuCS2qoH73W%vnlo_OcJzImW%sRy*dPaT zBv$}+pE1SRLjm@76J5dIT<)mQ2SjA~$H`$&5fmYvMTX7#*0HuVTc85l@*63)_uPHB z*NRwa3GLvsd6k}kzGLL9n`#=bTt#$^Z58<>zRqu-B{}pd@o^UKjCJhO8Ub4moqH;=}d_Z~Eh={{}3AH>!{Z_mr{&fq-WAN%t?vjesNJzZjvx8Tj?o+sF>U4`av zoapNZYL5ka=aHJvJf6Ef24wq}cfT3vBTJphp3QCD0^I=ke(`SllriZl^yV~w1kx&` z&kXz2{{i5>N)WJ{YXrXyKL*D5Y@+uZjB`&_NWJ%v*YRl)v(xyj!`r7l`+J?edJhi3 zSH#(q{YLl}{9*JFYJT@FRX!t+PX#*2?#mp@8%=#=?^y@A)uje;1-!C*QMNy0PGYb3 zL#_cjYs{eYvA0$I1%117<95#{gmfOQjbvXXpFNkzm59`bTK6zH-IX&wj!yvs`Uv`9 z8~qpz;GM(T6R17euQqGjoBdAUvt2N&dwDl{`!#PxZvl8tZFI{0#=eH!L%$7HfVJ$c z0|J=GuZ7-u?aSHQ$i_KS0Ls2+tknnRilDcU&dxdXE2(S#j`l{jwzXQoe4ef6^jDyd zZV&QFV18#ZE(@ujF}eq_mFTQ#y)kfZYwBC)hF?ZE0bAezG1}JPt?$!;iu2THgF^j6 zn%BLIK?(K}>x}!%0qrfM{g}51sh2vRc4TX9;`h`}V|owlpmV$y&3EK&-n-Nk7HABys_%mK?R(D z9jKo`GyKd~2xNVvS;UV#1-KK~<0>|HqJ2*{QG3de1E_zdAr}F#PtPR-d+~KQ%h-ap znCK32C2wpK7oicI@#FfUPE3%_Dr4aogg9?=p!(8FS;?X?|L3+g}w~Q zPo4F4(EIy?@B1Enz;3+=-QUgX{hG(WA$kU$p)muH0cW*t8}xxWCL8Ij&JyD}L~`k` zUxuxOs{bJH+_ume<9lwdOP~1fAp70eUAd>uLfV1*Y9pJYfEsU3?_Y#IhP?%K_itGgt$@SLJ>5H?RF{$~Srad^Ns6 z--8dv;OEYE!0#bXfU&&=^!RPV+rtj>3NX&|3h2By&hcL4nJzm(cNpHfbI9h{0~N;G zZyyXc$PFz*`rc%oh+H8X8^HOiO`rIxYa!eJCbD;ZVm$g~=6900eCHJoxM5PR3vrxd*g+hbw$N zzXI95tmzy9?AEe3_v3reS)BDj_yPPDXanEc6-vsrtLZWTn}Zq+-JJ@wFSgEgQ&K+Wsp@7x1<^!CvKZDgrE z+RrcGGeKv6AtrqWji7@*gw&rkTKMch&0(z`bQx^HThlx1=k*%0cf@}6>qEDp&aeXd z*sS5B&3@EJs5N~L+OPN0C#VM{a5v`HuLbNus8`lEgWvtkJWq~Cz@s2wH%AOgGyY;iobF+ta@Gj&D^aI%S3-B4av=tl8KA{3>?m{1X*+-rOz0vRcOE(e+}ASHSq9{BS0#3PYc)BQd+^t=JLmass-?em&EaDX>l88NayKzxK<>kf z4w%EoT|WuN$QO~F=`LW8+NJ9t!|o2EJbH7*koLZYt%P?D;mqc-zA^6Br-z?CwxLCX z4ssPSBA>z6ds&9-?;Cvsa=>Q>yLkh;d3a}tV0k0$!MkZLeFyM;ZTLB3M7KU>_Svx>|1@#O)7w>~---pjY8`~PxI>xoYHRMRH!^C9B9jLmx7xX2( zyYXJ{z>yE*h3K()UG+j*dF9%`@VwGaEP z`+*kNi;&LR@6#Q05xE8RsVVxO0M3_%G_!Xg;^U4g=ma>s`#6a20P0!4r$H5)-Lv;P zr{8I@jq%&tP4DwaZe#VGpbyB~$j;^-&F?JR$R%)|5wgDKv4;t=zP?A6K>;rYQt}q$ zl#lTGx!)3MZ-wIkR~> zpbY6w+xPCQ89H5QvXpN1oUWL{Ne@eiig! zH>>r%tLV|~M>cL9URl`tdF;M7Gi;v4401%S--`h~arQI@-o+j;PaD}jjawzA0A)z; zqLwyr^c>g{EW&i$A}*>l;0tnZk3&vgsxZmgF^PIJ{ z@hza{SOFdMcj4#rHh62f6ZhvH+-bdIczb#fTmsI07u5SA%xA2hlSSa}p_>3{h$+tg z+>gDU1sQ(X2D5abG1aYik52}Q0H9Cey8Ea>gPFM1^PS+{Q&U(*pqz}=qq3yKU+$4eeJ)P)%{y9LtX>& z<}UGfSMEH*&p@wfi)pS8UrnEV?%Jmc>9b3_dqbY4Sh4{!Poo5 zPv5PDbkEz+YrwOKU=H6t@LV=Q2c2;h@Nb`MU>TVEU^m^R$B?U-=5?NW2Idavc95;t zftCwtPc3K_(w@yz;#c>EYeesz($~HU&FY|3-GhYVnYe#3S{s++wpq`7qtLTmI zA#cHt)dO|fwEU6}g0;>!zO-#{2inF+LOUu10R-vkLUBLb^wuO&bJs zcS8Hn4k$ptN2=VqntEH?NIO<%Obe`m2%O1teGuRJy+*uu%o*)hf9F&V(vW2LxP#r; z3Vr=vz*fT7dxP#Jk3JJ{3BEvQ54AtOWB3f+8g}PB4BYnseTjVr`7WSO3w;5U12%m+ z>fkeA?zdqp8`+$PkkxBf@8|Ola87rc;bYz%;2g>WU?1Ldzwdb`E`pQ5d(}p7UeD_w z@E)q$()<@t2b?*>$GYwyi^SloE&}JRpaX1m9pnO8zZ%GGV4wC|z#Fpy ztZ$ExL4EAa*|vanjq52x*MW1oPwP)W7LqN-^s}T~Km{srCw}(S0{snqdCYB(0l9$I z?@M;?8LlDQZxtZGdF@&M0k9wY8Ut z0r0h+dzKm}Eui&&(60jTkWUP05BBC+n&&Jw|8{mzAM6F_Eu=VmF}{A#7x zqGQL-6R;fsg?`8d_TN$FtbDxd+8%>u*p0PL8^6Qd)Jq%rH1-m|PQ%-)y$z5n^zOxT z)Mo;3es_~0`~I@_nJ&G68Q^`jmS^^Mc;nUe!KQIwe>a^)*~nz!FAO0OOLM!kn{5I3J=oYBR>)f89^Lfv$rQZ7XS-`t%XFk}aU-V@_&)L;86F;>=~+h7dLb8Ulu zjR9?N4W0E@fO&*}GasOLziUv>vV|^2a&jkPXg&+EXYj3V@|z%+_*zS9ubqwbd0?D7 zb{C#WMw~URA3)jUiBRWsjtbs=Xe;p@G&qSp!ruwqA$F#V9Y6UD-rxJgSy1;xyk~a^ zTMv2$eb`9*vcJ9P_kg-@Wc%quOXBRYBDZl7x&`%qL~P!n`kc{g+lOpT=iA+WLXv6=uvCuW- zcfxyy8GZq*pwr)Z^fyNfT-#`VcIp?C?=re4h^^4sLjiiA1j-eD;|=tjv3t*&ICI+L zL3Eyd9n1T|Gun^r-x#!O^GvdkdMu#cInOSGM&!PHg92R(ehuob?Ae`omNVqp*G=C& z-2GS|^v?f)GJFLF=A`t?JY#oRg>)W|$;UMR7XAm2 z12*4F$_0Fz*gf6!S>x_c;$sgJ>=CrF*^j?JbUImV=OZ!L88XKrWq42{n~ ziC>AX2dqDbzdN%2Di{OnhLG|^=xKc30^Gq*LhUEQSJ*t$LJI81S!*79f&7Fr`OO=U ztuX<09I++3dN=4Y^^NatKl(f09?*ifw;sqMpnq#^B0JaGMk7PVV2hZGKwt9(U~C_o zPl&#x^XfGAC-v8k-rn{hZvxM(gIjk+D zdhdXDhb3|btW(8wr|l!4b6)M%+}(4=Qz3W27{ra#_YV97)ODfuWZ(AH0oTxZCzqjh z|L|D^o+CSJqaUENha#kXXUNtVfD9Da>`Pr;2i`haH|@q11eURTkK14cnAe}5=MWTs2c=z0d$1V=JK6S%qKSNOB9_`{TJr>;vQ#kl`26ySoG3)UUle ztF|npICUQa0iC(bt!;<67O)p{2lV#r*{QGKGxY946vUi`+SdRS4J!0)cym`V#b)Tt zX>M~GFR#wC>j3+<&N8~9@i(t~srL$Bl5f1w%-c#lgSW1C!}p#2KY+hI2l#;f5oG&1&`tV2z9srhpm+8|*sL34I?HW*GH?>T zcd5eHp9^<_LLaDk?8kY9J1OxqPWbcPy?c)b@aDE>`?P1D338o3ru@c>0(%9$i?|A( z9@dXCkN-Z(vsm6pcRoTdnUmbB`0Lk)wz|oGfV~ZG?h2cJ z0r<4>Q!cS(G1V%S(L4X>D!G6^`^~M*x;`B#di(a7K%LF{#-Hk@d=a^iZ$M`+71TRn zJ*us=l#HZ`|+`_9eiRF7x5|JwRvB(c?UWm zV*9|=)R#WlM*6;JE^E32^Er#3i=MwWb7!E39I%IQ{b!0fyd!Pw>O0Ue@+OdK4+~J9 zgTzN<=l9uxwt+QEV9(=)la=Ch0dTj*JPG|};P=HC(q}|VpM})3wd*rMc29mUc^H_#r%r4n zR3Xi_yF2fA~>bnwf!pC05#Hbq}m*}0LM1KHO$P?@#CTh${^ljkYddT*<+UTvj20Go8 zSGgqC8Xd3(dh#31S;AXGeF2t%wW^JDw_3=n)EXcc=mWAkbDTnN9|g90-{{;`gwfU9LaJkIog4p)__eWZ1MgQ|8@n+j{>Iz?LHr`H zMu}|x5vciYy8rg~DExk)-zDq=dGu>wTNxV3y9zaDwvl|*`}Cn>;BI^m#7Hm5J$QS{ zLaJrY`bytM9%pVNKLGFDSwkO@3;cIujjzy;!43#uw@+j0b6rS(!?(9(;O9<(UrBzS zKK$Np+GD`C3TX!|^jRb)@&MldzGdzSeBB4O5?h~`Hn84@8Xx0!&{d7ym}}Sy^df-A z8u=PF&&toQF<65S#7QHx1R1c7dhfCIGwhwK&3lQyTFB;d&PTyvu!8){QsB?$43y+B z-g`2mjE}st87DP1##GO{6^QYC-M4@HcOL6h_(&u6WpttOTZiufeJgaMF1Ce)4s;Ll zUToIt;FrZj-NPF8HdH?`hIjYwr!9~(^v>S`)}4U%Li*X`4rjV)u0H%>_!!bX9pE>I zt_SVm?@R}Fl1{dF6Es9(y^7tbZ2M4*VPT z8g~1t&krBZO}PxIzrL}?K|4_2AMU{Z2Eh0MwjR_w;{Nq*A(v3+>I2`e=6Zm5?|O+Y zP-higkOvQ=SLZ&hHHH?*=01$R^H(9+I`EO$F*HMWsC)fq&|~EA z^R|u7`|2+Dz(-?=sdJ$(;WKpRZGi|Xbnd7^-&tT@;G@m^XRNc%!+RcCNV_oJS?zxr zANh4~23dcf0$JM{csF|Ybhw*#Vh>p)4qXf>XS?yW&vo>BfjO+I|z9k6w+ONLL z_zaZ0_&4@U9au%b27T~q`W%k<2W0I5UwbI<=_BvQM|~gJUJCdL*oNPO&F_eQMp{q% z7-S*k_pU_rC31Su+9wn-I>S8-f^EEcIR^z>0l%E?k?@6 z119J*s4<>{we8#Yjz0F$LN&0J#TzPy8Wx9C+{UgqGL~U{B7}frf?T>pTbXah4wNejmU` z{hy(CmJ+OPq!{&nw?B`M{bV8CN%!rnC$R|y|~yba7}UnTw(Xrp@+ANMBjU9bo5nz`)JbMbxX9eTeqdf}aG zb<-KSn=+Q{IF&pxeuKHbq7HS?6m^UXMEk?0gzd~&%Q!PcVGuT1KXfNr`?%K_yFG`M;6n4?Vvw{ ze7Ku>F~9NN2m95pL>?pCi}(I?<1^DuXBR`-y>x)DFvhdcrzFQFvX8nJdeLl3^Jx zFUw0BFU$KpC+?l;!~XXF{TXpiW~zDy?u7?re$e z`N$0Fecxg`t3M&zzxOktdy1_84*8$q^<5(Sv-uu4r}t7=d%o9D=hFAL&>nwtdKT`} zm<7<^yZ669jBSnzwdSL-*)u-Q;CwI8UxOR?C2;q?*M*oR{?dSM0`qx63-Q)0-J^Rd_&8g? zX}QngzXI#|4n$xsXK=rP*h-A&xHE=#VGXs`gwK2QgE8>-sPDrOJ8|~lu8m31Bm5Hj zupu{mh8jD6lhp(B*q=MI*9Uukhu;3h9EUw)kHpxIb#8#YJBzUq**Ob7_wW&&d8OLS zxdj>jJM>pz35@qFZqHi1dw9=zu19PoUj_--&pGxLRGsv6q%NqXeIo+AGh1_c4aH9+ zVBhY`ID4?J`?y5*@3xA(hQ9*8Lk>{$_Rt;jEjR%QIR7)~OQ4^1Yix2N<_NvA=WM)i zA04uP7rodc;uEoDXaxnIAVudsJx||FpNwvW&VJp+7T!4yd7W!QuFy+j4!&!2?)3?} zCAJ#d?#&&Ly%+9Ee`}P|@_ILVWMjX9k9+yW*y=lHBG-UD!kg;~pDpqcKJM0h8J~de z4nBg4Z$|c8Ag}Elcpl303=uugFUY5^hg!eHRx_Zh@EN%4hhRzk4bV0}P|wT$e3!Rp zZKv+@0$lsDVm0_X8O%vqH26<5lo<#fhl z;@*J?n{|5l5F78=W-#A%Xe6$O9^Tsy-W;!p@qIl92_Mh0Lw}1sf*pL3$5zQbz?)aU zHMWn${R+GWKLhtx@jHdT2FAIEd-N9|p>wu?ZUR00B}l}GAT4bl6@G`0^#VHoJt*Vr zezs6+*pK}xo6lPE=ACmv?KMZ_(%wm9%wrw*G|6L^zk>gsygO(G=J8$gy*7SOK5M%S z#4N`4_wF~?7uYI3^K9_W=^5?tcmB`t0eDtP;A_tXtby;zCr~5bgt7H(KmNO=IYmLv z*geMxwU-1IAQpJ+9WZA`p8E;S-+C{}^#ky%JhS;hZ|@r)caoOKeX2)WzyD`?hlZ29l^`ZxG^SIr+b?HyES&+e=c6UKIa^1XgM zTj&}Lz+L+d@_s}qk2!N}dvz{%x!YjCX9a(S-Pt`~@8J#dgR$KM`>uQ43c3Z(<+Fsw z$Q<~r;TLCZ4Ne2-3l&&gRI1ABBf`}9sOkdk3>fSG~&2@;=euHeh^_(%F_Z^tv?JddUa|d;vC+Ll<*zBj`@7x2t`&_~o zbP;@m&e)@|{T!|67TAps@GGc$`3XOJP*>EWvwwRVKt|_IB7A~+cf*JbZ+_1u#g^j> zaRHsZYIE1aS^HTWplh%NbB;~>_qcPY^K7Bk^IqE1GjIyrSu{V?cV7Q^*1pkyi#@7G zu0Z-^Z2j8*4w%DUdVS4B;;T3cX_9y!6=m+#x`i>$66@kLkg zi?R88R}ykXzJVV$@W5ItWbNMPdyw$S#Or6T)^pzjI35w9o}JV?Zj6ul_4Pje6Y=^y zgxY_N)PkC);PYGbemi?)^Xlh}0W6UV{t3ue{5$j!xj-}2IH^A7?SVGy`&_`+*y6)k z@ZS{OS&0pwk{*ntybwdOU61Yqe$TiT#~#zB=`N+fTod{Q(B@8!>%ba*L9PMbcym8P zj_9w@Id?_p`Gyf7n{xvGe)k$w_#9iTxwI8z>n*{Y2c7%z9^}&Q&!6DkQ^t0WpM4f! zJ@q?d;H|j=gEG$+>b&~R?^;BOPJ2RUKKrfsbl?KHf(icMr~C*X{pa(9?vU*{gMvLr z{6WOlql@UwnUUSCv5(EcJ0V}={{N0#8~kcdo%_=^o#7m0VEhgw zusz$)%J=v=xtvQl{~9Dg5BBJL@*^- zHTW5QAzy$u-`^v03YNG0vam`kVv5iS9JQzlJV0ZO^`ck^Ii|Xly(4+!H$Eyc^mJaEI^lzkzn(>a6_^ zSYq>8AfI6OOjg*}z&xI#k2RNPtv6@47xQ?ISI|@9{O?ohN^G^&`FCV3=@-c8H zFQEg}v$TJAc09H-Fn542#3bO}b8J1?yM6S)ov$|df(oCZ2|v%!XF~n}Z@vLIQ-U68 zA6v~7TOR>>hkXHV@HzBd&|7B*{O`^0K*8q)v|?MLH*W>@=^gDsE^U^KkG%%u2+Z$& z58x~CE=<_X@4nroNU{B#at7~BKo_8%v-#bjc`v~Hu){69aA(i(IT_ozgxK`P_rN<_ z(H&;_0ok+ic{uK4$n!=Wh{V|EckoBRGv6WGv*)&e+Mm=NJ{j9L!uRPtyyxrhmIb?U zIilln$0hFX2h<$;EWxz-um2syK#mS%^u|Wuu2%RsO9x+#M>f{on`;24V>@&A=Fg66 zkkH$carz`|&h`{|MxMX@`V?&L(Ox5XjNY0{?Z^v|H~V)V8bOaPK&|;Be!V&n#z~3n@ z(fRKH&R`GIS*v5eNg#Wt?0wEny!DhlPtSM07nC<9pS3frvGv*^FVRJF~O{ zH}KZbUa)6$&S5_VOkfEHusCbI*3$gvICKHoTp<#VpEd1e1^;T~n*-$a^NH}*KSFPf z0PNd4?7ixN{rK&WI)^s zV6KPWJn!(chPwX)?~DVy{awJDvxB;G`IoyRl^2<&89nigIt2AmSejyMfo|Ek5Qv?9KPc-i^~Q;$t2Ae2C5XJ9O@K zwxQp_FYwty-;MYx1F7f!!kqiq3EA^>FO}FF=}$k9KtP^z;b;6@8(oKN?0k;s3%CO2 z@cmg~OZa)_*7^WWfpP?XHxhR5w*7c!y}9`1+9VU$E7PowIj;OYG(h&;j^v7+cUK^~AWFCt&_25T`C8+plp2Xe+Vp z%G?=Vn>{A@3BIc1@my{|hc1G7&e&>dGmkxc1|2ZY8un_;5}k1gTEGI_AnO}J7#r`; zq@z8b_c@%&yaTA{60nvC@a8a|`7fZeUW#mt@5>eZe82a*3Fud#7?157A7jn;4Bp3> zd2dktoZr10SD`!f)-{)L^4@*>@P70lfeC#C*6}H^o$&&B0!R3G?iGCi_H8cbPVo8| zul{uJtL|td7xa5{<{HkLY+Mg-kJ%iYtv}zsf`UGxGhYBXHoLi28_pRY??pvlu@C6& zQs3D|-F3*V1gOGiJsopqsGG=c5mHW9;cIcJI406#SO(Pshe*c;Ov1 zMq7gJpx+Ybe8#4ITtI$_{y%^*?jV6dA7sz&64{xS&>hf6EI=gBcGKQz=RFE9s-Y^4BvZ}f*Z-+T}7IoAiwdqkhH+eZZU67g9;-{^zwpqn=PcNd-c z>?^=0^z!y)9_!h|3cf(EfH@Z6(b#x<)33v}piYIK(3RNk?Jc~so6kIP1CYHh`X&6_ zf%jz%9*pb=_&&b}53WP&kG`y%@Yh#b<Mz2qR-)*2{+hdPx zo`TK4`4;eN;N$LZ;A=#{7NqpK-n3lKaD_dVHgm%6jP7sl7pngXpCx<;+-1bi9`twC z`6(^dyJi19)U&i7`#bDwKCAt%>}}pD)Vq@P-|YK?Chh2a%n=ImHF{{qUO)ifq5C7S zKkM7?H^_%t1G95}XiIk$nd;wgC0_9{ad}26X!T zB+qj=NAIk4e81hPy>5Yado=zY+1;qS0EfS!uFVA=j_nK!zK@X${x=}Uc8Bnq{{~9f z{NH0|a+q(0&-YO0aSsK(J6M8`W6RmYzs0BD?B5ss61=l+u}S@g*r)d)pmQfj#LqLp z+t(>MVr)JGKH4IfWAHh~?)zWi1OD#Ay4pRj{|Rr;9lUqPz1g?_ei`xq0qXz#l)T8` z5wda4wjjpd#Oc^}TlVrV@v&!n)Sg9YyR{!p zA9nE=yFMMdr}+FHY~kI{8}yNUU(s1360e-0&ScHHhrPeTZ}!1{1aGZ7F!xKG{0hwP z*SPmRBNuRA<_yGzy)6F(%rV${^d+{xpC@E(#tgvv?(Gx09+=O$GXCC;2u;9T1NuvF z3XE~4fIeVcA$y-Hws+_r1AU#@+U7i*TL&MBE%+w*g592GAE>#FJ0^#`^~~*4 zVvDKB{tjS{Exh&RPtaX}1pf!9ciMUjko5uPwN@pUy_(;AU*LTYGx{ZV^SRq4wt}t# zWqVr#^W6Y{UK{6p=6Ckj>hPa|&idxGj^`@x>;e89*h4`!{(E8zu>t5C<+1tL)8Y4d z_1|8%Z=|vB;XCLA=I;qTSuY~{&HIY|KY;J_9kRa9;ph3u6+lHd*W9$a&SrnLG%Ejo zBo_MF*t3Q;vUz%BYfZ@3Sc3s1aIgen zQvC*~zL)4SyuR9YP;Cdh-%6h?Gy>l>_u+l{01~p_KJ_o5D}3g6bZl}i1oMusu-oH& z$Dk3s#Xs-$S5W&7`1j~jgre)uxNGG444n5N2<-Q9U_-ETnW z9i87jd_KV&`@f-E=q;Gv1@>Uw0iP7x_jbbPaL3$}&pUK~2L68FeiHt#$M$YjeA=f2j!eE$YKlL&mL z7Mr$XXEVkp@Aav)^KzaEo%?iWD`E$H&1qbYt!L-&$F>vqV}1P}?yy_?2;Mw5@b*~Y!`SY|3ADs^ww{&u)@LF%~s{?$1Kb)_3smAs@ zzx}>44xfb1b2sJ&A7}mn+bM9z*7xk!dz*bYkG`c4k3ExPzIW)hAfdm6T6>`m8u78- z3jF0x)!%9d)^cAxw9}6~_TwA{c=qmMK-VK%vxB;afWD$PkMEJxI`81!k$0uU7VqCZ z*4XKT{ssIo$nbN$y`Qz5_u<|i_C5)_bGQ@xG`IIUpm$dTas_kJMuj{ywf{;24w3R z(<3{N{^|<4fX;g)U89So{k|RG_4hGmg4fT#X=+5bmz}+0UjXk}2d(Iyj*Z@b`YY%bA9LDg z65L<&T>#I*`MOQ(kqmQ>1>M{mJ4@sY$RReL?}!=jujneeJ9yvi0$o7ocXn+3X6!wF z-lHq*$}4C@w*^ah`;6#2d=_Jv>m0#5kFg8w`27N1Mwg+=bIpW5ZkA7lHS(3uN-(2qIX)g`>|;`6cX zZ$+H<&9l?bpZUgR^rZl31&8;dqVt}Y*v|Q8L%e5$F70{WdGEOUwob&?9L^y%uixUQ zz}og;jJbMv&miM-LTrG))DPKbKzFYlEH=&Gd-)7_m+Y;FE}>ZpocR{nT%Ym#30T+u zCSa}t%x914V`=-g2Yvlp?XW|89qyyWnbbSi1bY0A##ZYw{2MUu8lNq8_c374`U36lDZsDx zx)R%-udzjJ_U1hbz?$yz9Gg2{Kp%o7`d%M&&hYjx>@UaGmpN|mdxZP}*oQqw>KT80 z-Tv=WOKEj8)S92roA(MjfGxU&tpexB>d{4HcXkEep^vfUiSTdGTVnv;^@44M&b#Bj zy#t&hVbeBYf;beZVK8U!Tzr?`A}&uk%-^d8=_}?R<974*30BV=KfT zjcxy)kv$IBmq42ES%L)bcWecXt329Xii?LRaB?pw4@e$zlC@ zeq_(L!_S=Vz`6Xr;|ibooUlJYuEd$k{1XW9=IKEP3ecu5lf#_$qa2C#-F=U430>}O z)^cuV^;`zfBWL{eE5Lo(zfXzy0&^N`KgONn<5Noe-()VZi#4_-)IK6~0u`T#zK74T z-LIF%BOgIM%YgoHeutkcE9_sfdp{EP@8M6!{kuU(Kqt)azHZN29cNeoZJF4xm*vfE zkM@1ILsHMpnO4}W`>*ics{x#Y9eVd^u1sF z0FgM)cY$97-@^Ymw)58iGpNMdxy-e~)hN z5#2NR9_r3sz#H=c8n4^W{O9oNP21BkI(O_oBGmc)-N9U$ng=-F65RqH{pK0viJS0u zh9&SUy+`K12dlIF*+K6>A?Mr+v=X~UXU+xm7W#qs4w|svK!1%K;lCZ*zbynFYrDJG z=Rura@YcQuk(j6G+{+`dL+3e#NIWxox9#gYbmmwYgUwkUqO(uW ze1|-N1^i3U!<#b^8}N6A01~)7qZjBK?AFWhpRe2RxYn4!5w;2a0PMlp|NFt;eE0|S z5k8hSt34T?!4>@c`#gTmk?>og^PXDUKJ`7sZqWsN7Et%MhCalngX-sQ&*2|kw>w~- z2((?G&-8FBMRtEn_yBH@2WTa?bcw7!f>Uxn0QPqx#m9Y`GXU@Z7xeliWM^{c*0!EG zDp&yfa!2-Z4j+i|j_7AU6>4tpM}+3s`gx3x`MoFi>cJ6wf$IAL{ss0Pc?JI%=;QD2 zPthL(=W?bkI6}UIcG!)-2gBLk2WtETy+QvPz6Z{)PoXF0TVpfd1hnhx_oozk2R>OF z@T_udIo*$PBIW_G{t9{46{FB0*Vyv>jF0gK|}uxpFNWO#o^R$q{vSsG)j zaSMNp-JacZ2X^2kI%nw6SA6YDzbkB8@IT=L)H+9Fv;Xy8_Z-=?HYP)tz;kuyo{K$1 za=II5wYES1tUcjV$f?eqCv^6b@t>eeeR-T;^uRo|%<8${kN8FXVmlvacgH7Vt81J) zFmC?2i(ijyk9W}3-loo8J;#VoME1UV?(*th5wnK3_iCO^JF8gQeR4j}(c0EqAxjg! z8NIO?o&IZRL0{1&_y8ZTn_b!6XLR}*XRa4x`~E2xZ01U(z0=mz{{gh1o9jY72kRAV z?!|e|p#zw}m+NL%wx%=m&|7FC#`?yo_uiYQ0-tn7EZAbxHv{E+U=JUz+perWJDG_$u0sESKEYpLvq#S}q2D5V9_9+@0=fb`{{dZs_uT9)fX~;h zuZ%upFTffrY(2cQKEucJu$DD@XvAMx-vYmcw{}3*X8vPh3i4yFCU$=b{{eafo}#lq_u^dkyn@g8RIu9HcIceh zJnqCjm_z@FyoCP`XhLuQzN-;_Lcaj|Ux6+BJO_NjKNA0L@9$mkeey}Az0=mT&TD-A zeXydhkvOo#)}tGs&SGCZdD5_h?hRjOh{%U$ z0q^%`LVq;2+ygp$e`YP{659$GcMsf6Kn`)w)9XupOZ&MZwf9fxGQP%7`xxml`iQ*( z&o4y)^;|uBe}A$6$9wt%RKK}D`bo$~!1=$7?fe6|tl8tIK7bogz};D!+nuiA?O`JR z8&IjO{tLW&^^DYIeB|xRoqA3kyt~!k^UwGtV7`@dY4hv*2VmS>3u;gOo~pkCmuIan zdouR~+{ps!yIRQYUVYU0eyIy!y=nE5azTZBwU0X*+j;q&c>pRl&oN{7u}^pU4g_t? zy~O73)wz`t1fEo*1#g z-zb;%J?XFyz*!>t!?$017~5Ia*z)eM7yOhf{2#B$54E1PcVG>S?KkZ_ZlS`T#R2~S zUt*hc0q?okllhis?cChmdwipQ==67J#!P!YLU-~bzJ(U>zudGu z`b2c|9P;QB_5ih?4DX$l-b0O_W3XZR za@fm*v3+B|LOb*#;qTtPSN5|;=QBWiP)j@Od3VV6rk`hVL`;s&S3h@S&H&6~4*lYu zn$NrCncV^H_UYMW;?LoGutlGt6Z8W$jL|RRbBj)d$p3>D^wt&${@?KbGcad|UENdU zfIS2E=DY4r7T7~<@2z}-AMn2g-{3RLaEp3(+a>Umj3bf*5Uh~X*k)m@x{Q`ay z)H|cBzk5{9K)FDTH|H8!)PV@JzqxK_Yi<3$Lq629KjXuu?X!ov%R<~k{N}vSQ*e); zbcua{pZCPBBD}Z*o|*H!!q51vI?k)0^L+Y!o(?`x!*Ae6;Co{3c^CLz0CPCAJ=)iN zccAtg$L3q%l~3>up8n;crv0#j`+;y0UzY?{R!U*=(h&my8-x^ z&zeCQ)Y$Ht{uP^k_8H*6pt}Geww>D7F|s+F&$;Y<0klQ@Ca5(%=ML(;YmkX4*zI{x zK5Mlvpw{cj8`R@xZF3s;7JUZ37x&s@+shx2y+h7!j)Goa^V^^Msqp5E*zWKdfVPZ$ z5C0e+B`uMu}A0t;}_v77|Hsm1AzG`g0lbG)jehXlI_prq;U@yQ}Yt4C({cnlZ z@Art%8{q8DRq;`_FL#@fGx5UnTA@!MBD=pUVE#2asdoEbq8AJ5EubCf_qw!?eS&QQ z;jH3?}m3Opo^v5i&NwRzrgN2a`r%+J_9=W0&i~N|y?#}y#66F7{>FTZ-CZo8_S}ta_aU|&zd`>mz@Fv3 z1MW4L2goncW&EwbLiZW@7CJzk%|3fj@y$?UGraTZQ=x@;XMec&w{PFeo4swP?^*j^ z`)^T8WneyQxF3DJe+%S@>>aX?ujp3LLd|E`ylRU z`>_xEwf9%RzZLxNay#t46V|+kI`0H^<{I0+e2?ydIgFKe9}!=vei5B%ac z#eSCXJ+uJr1$hPRTk7Mi`diEQE5!D5;0Jv4??6J=kJw7fXAc26BU?uwzXwl&J$gPD z@NYncPhbH`Y`k@>-=XtvodE3>dm%o+FR*VxQl>X;##z(+J^U$g_A>7VUBLFQ==?W2 zYnwkr&SPx%Qr?^c&}MJWzXlnbwIlfPy7d#$nQOpa_CDTMW2M8`{Pl~-7vK{AN8q=> zIevzE2kbqL?R$2C-W<-5K?0Q=z7qvFZ$wVu2EQz{k8Rctf9v(v_<{-_u+I;?_eGz7 z#`Xd{1m@hL_ne%^9=`y43;6cH`Tqy<3cK^2gBq~|`6F!3m*mIxO>^h>$i|;yTW?yP z3F_lLS655(x3_xMW^=A5#Ms+oXpF5l-y6S;_Wl682KprY1F(L=*FN3BTDi2p0}H&5 z_K(=M_;jEEYyBIz!AD(y+Pio7(8DY2C;FgUgB|u%$+zXRHDvBiC!0ow$x&jLiyBfBSOtN82FfduB>FnOQT*?xEU8XNG<>SF7!M>qG4 z&huX(`xxgpYTkWpXX{+fD(rbewui6q5g+T@_W)I1LlZQBSlZ7l&&(dK@H2;d*~)`U z{Ef}nFQDG11&FcTCHVyJ`{f>$eR}vg-`LJ(+RzL99aO(mTF#yE^lOja&8STM9NTxc z;%Ci!pss*0!Xqy?Z5Oi+wL3__9dytKBRj?CEqsKo&^v1f^;w}S;0RrU=GgYNM83qw zUaheP_Ioli6MV$adp7Ou3ur~>8NVd9KSRd<7~cK4lY;C%GJeL|zxnLz9B6m$FT@PM z9$!Gsc?x|qwz~RiyMb1)gSY;_7IR#uy65Gcj(2w z&1X;Q1BivQfX;V8+g~;P9sH6U_Ti2ba=~|i-+=-mxY^Tw)B4iBgs)KR`JL5I{RMIa zB{rXoE&+QGD||a;?DKAzb->^Gja>lyI3CH5kMR%SSL*O_E_KEfsIwIO0RoDCq1>o3UT9 z-2>0UJS%MGcLsd}I`etv=1S;4;qS~xz~0Rpf&LX_FkQEK?a6!zo%xseEs@PBIZ$K{4g8k*#z8}uIJ8OBgS8TqM&*5KV>*3Ye$G7O#z#aS&oA%$qN9;LL z8~Fk9NAMLSbnYXPW9|#x1RsF)1NKvN+A`FhJCF({Yz1oFZv3zFG?#k`=sl||Yyl*6 z_U%4AcX@OF3e0;zp5R}C6>z>2P^mMYBenE1cP74vzk=GYef_`4cgVNuK#$Hll!-6k z)!6p5L}x!c>~k*Y4OpO`=b^tI{tlhrfi=8!v^^v48{j+g^t%09Vn&~UcShKICCQ9%vW{}Fok zEmEW&`T;mw54FzR7rEv>iSh2)R|Jo**$HxmKf(8k7|%YSGu|@@*kWuui|Do4!%Jcl zRDW~cK<%f8SHA{0T?&$bJOO@(26|}r=Ri?*jIMd=LCR&^w&bxexRC zJRh5%bG*i0u-X3v?8W#a;J%f^ru|I3gj&=0#(iV~f7qGtop-_Wy#kkK#NxBSCqZj$ z-!;!keZ+45HFk5I!%y(vVz+(=Z_Ez;0@#l>dwvIc?C#vS4D=89%bRC`k2TzN0M1{q ze}lek+TAg)I{Pc|cOW2FU_Ik5KrNizruj|C#$@0ez848NhdvRV`Lu=W*4I)W-VJ*& z&TqssP|k9cHwutQx7-QW9e85kede6nbx%AU! z{3GP$88+xJHlH`}&b!2CfM)#6=Qm7W_0H}b#!H_-SLl7Dr+a;ft_5gfzJKI&u5O^9um}<3Roboz-N38cdCbf2_p6f(13q} zE}?foy)zh_m4UW%lmpZeO1&j8QZKyt{FhJ9IDLGkhud zLT}MkVzmDeTT%wf&ZN%`bekux%OgPomgGhamKOpo#Tzn~(=xW7BlfUNHX&71bWb@6+AiCnQ) zbnZXFJEwbfj|Dj6OLXEC?9e%fIqe}3V_tQaVAhRIwuZl}hSK`!v72A4;QO;?^IKN& zDG?c61=f8rwsS!HpWmTesDa&mgxK!g5?Q@@dth$QaRLMU^RfLKqB(tcJ&S~Y05P_A zz`j@L3$pvTg1?1&zrKQVcx8KYH||57@DABi0Ot0a@)%rTi`1wEVs}JK&D~ z+lZa|89o5*OMJZxYv@Pp8Qlbw@1dS|iTD6}T@dGYCL!MV{v%~1DhPtI+>{|Y_*3*FD~=FX8kQ1`mqv^(-#KlJvaU7Ndg zcm8|1x&Hx}yJD}{Ixqn5k2VO^vw5wruemSb?>6-Co95TSNBn)vVSnz{`=b8Q z*zW2Haz-xrB*6?H&{<3UmY5(V{ss)-;fN0XE&LU*2W#7t|F#!m%MUl#wrB5?`@6-j z6b#JsC*%c~&{@NJ^Y{@C3~B*ve2 z-^*h!$jFIayPr((w_7L#>g6ugh&=uhRGYFA8iS0-~V%@E0a;QI_ zG5!HOK%WJ1-$7r3P)HHmH_*FcE&DU4cJGU^KJI!L`5uDj*b;ICcVg4{3NJom-$Ao- zY`&iBd-S=qvrO1}?VEN6_F}CFPVrMN5sFWbu0U^~zJCia!3XRSblAK{5$fan?Y$VF z?!ejJkMt2+Zoj<=c#g*W0$awm+U#d0>=AbH0GoGuf?s0)Yvi}!7-V98b=`KhK=xca z!!8?Jwv|YmHxjhA&9ymQZcNv;P~Qk9W!X*0}=b;8y;u>8+t&8lj2!1kE6j zBcorPwRgoD+6QbEc)vPy1)n2)ckmUvk2QQ}7U=B9`Hn${-gY)S)H}~=TKpuX>9Md_t0}prTN>h^O?(Uv1jE>6(nF@Yx;9kzYAmO(HzS9_>|as zQ5WE?X|1=w`FH4r^_Ixaa|hMe9_`(kR`Ble6a25iOE94m3*bJ6&7L3IZhW7&0(~Tp z_q@iI)3bbuU($xp9kAzfWPLiQKRcd-`JLcB{)oSR3v}NDYd=PJ1$~I@S?OyZ&eo&1 z2W5R`3jY}22tC@*4j0R(vvih^;4QIP6fHHBZgOKF#OOZj2-54g3$lovzTC(-@x}yw4gOc6R}|BV$8pKF;YJ`dQ0f=lsa-%y&Vb2|j}q2?xg9oxwxRW&T6YTX_M!bHy8i|;ah{ho+}!{^p)Z93d%Hdn+24P| z2zVd7kp*gm1(j0%*&g@V4{tH;{?NOhx_3t?nsM8AXgopEmg z-hS=f-1eKu^B$e&?=Jm4&KU1v#&mQe%&hwil+w1>h96k|D_zc8K7r@;W^cBA)vi=F3z00rA zE#ZGpjsguLl4EQ$MfhA=Up+SStf3Rg*tSsj>)$Ni-wyc+ymKXZYxtx>T4dp!wx+vx zPX&H}n%gt?JPvoznIq_dGyOY$zXtC31X#zjzK0%a1Z&0p{Xy@OVw-ujX>}eG>(6rc zV}5Hs8FA$9`MaQVyhpaDL;W6|Xa5De_tPG(kSDM>Ykq5Zd-r{kzQCv3^OYX%uv_a8 zV}9+{EBHF6HJrg-{y%j7EQ@=;0h@aWk#j(w;6((^;vHVWzlC}x`X^$wdq=Z8u>t)D z>>?xE|NKpZ`aWB)!W(CQK^=b6L59!T@iW#v^!V)HrT)#7@UbW1IXm<3q54N`>Mr1W z^aC`3EjsUlKGr(CU)uCF?j3exoy$G1Htil7x6%h3k4-n*k&Qp(vfeC{%Y7`7&AkQ% zd4l%xVBQJmnD7mP=W^5LTw-^&pr4fU60hjP-t^aJ-aEYc&%p}JxzH`(t)-t-y}DC! z821GF8JI6_+L?O)&Vk>{H=x5W0sWoLer}-;$9;Z$+?jjYVS5X2zs~0Un1Fq~M1KQs z9?#o*;C`&d)FayBBxi?*o2^19B-)o>ch% z@N+K(dWX$;b;frfB0I~ovHc8;@HOI7n(Y&^`JL?udS?vwZftke`RqY|>mBZ-`K_~* z5^F7W#`-(?32;_pysMs3!gh?_z5fd8UJAN%WP9)Nao;=e3v3D7&0e;*H)Gq~J$?)H z-X)(HTTXondi%Hn8N|}&&*;q^Kt<;+^$|C~d>!ykxw8RXBCq}X#MpS}RA2CO7WWlO zvpK_ryg=_axx!a`C-fKa3EqBI$UeURzXjG6_G|ta+s{$=@&&v1PhG&~&X(w<{n+m1 z1-xhMUGpwDkF`Idvxf67usN$fJ76D~oEd7~HSmn?u_bhV!z;XJzXhK0dvr(7ru0qP^el@r8Ek$oJr9WaET=8SnYntFwQA zckb`uUxAGN2wL&of`q&TD_{==IGa0qMBE7|vDG|=ztN6tZ~7K|eGiNY$Qhe61zKR?^dC$b0OJsX<7V}u!9N(ck0xRqh-4)cl`rDKGjNAj?vrBwmgBaUf1NsWC&{;d6 z{{;UExL4x}`Z+IhLe?jv^UQqQmwnCqDA@2bE=MT5kNxN)Qf&Da>WG!vLqwn83$!46 zCi({8F1Of>|08^aw+Ejb+kVgS_gvhWI(sg_URKbb(9eI{6XV_TdvF6ak3OE?w`2Po zuMqFKPuK(4VV}S){Acvtbvv(l27Cjk*pD!@k_d8fv{vPWxHH>o-7+UBR1M`wQd&+1YNO?j>P!him9La>P%c z1$Z~Mv$ei?^y&A!_xTO>y!WqAb&1>;XYJh0V?Vc0?Gb9d!}<2eD`fjC*b`Wyvxa-{ zE-v8j!5XM>8$PIXM850uR9IY3D}MMdvq7jLVQw> z9Pt^@DLZ>sKeoSX0{k7iC)enp9eU3}f9KX;Y9H3`@iTuWEaxh7vvpKUO+GQd_dOUJK|mTe0%&ee8J}4A0k)e`HY||U_bg+=n8r|?seFl-TL-^ zj6UEKkiFj>I`?BP?_-Ue2R<3QXZ->`K?C{%GBC~>`Z&{agDv?`l97!5zL2$2Ri=^`*Td z=6i}<@R^`5fHNgzdz!F&=iFNdZ>@rD4R3$;KfqVuJ`=i^*XWVe3wtzv2Lg!bjI~$& z<=y$hxY+h;d_XT&_&HC8_wNGFT)#;@)V#N#0Q)olSABRAynfD{V#__D%kYCfdtHSV z^r`I61-=L7G=2yF5r6BrUwtdS0l9&+cF z=LyiZz*b88-gquGwmQz@9dq~hPY*7}&;HdvAKTpCYjZp69k9Q}9`=^WTahzyzano$?vX3h-puWp9)W}|?saSQ z&ie%Y7T%ZwRkmOKzJa%&3~x;z<9v2dXIO#x9Xf00@3}e06?jGbJ@ob1{{7M3KcTaZ zJ9rNKcjzT@imi@u)_1lE{Z{@r?SB{j8Q+TTP@@pzZ07Q*vBA6fZm%oGqb~&t24wGs zxo?3z>AQfs-%EJUEyKTu-l9+N`Yn+Q{2IQ4X7Z|kiT(n(lR%9}@Du2j$M$_SpEb0b z&zK2nys`5)6WJL1R5!ruAMjbgyYC);00H|2vQVy2Z85f=edbp7UCL1N7i{*hL^tOp zhr0~ut$E<}Il&gu8RrxA1=hL&QvV&`3)pVR3oWsIzams#pB*--JviGIx`v8lVrmKV zd!xTT&iDg*`;O>dzz6iksk;L6j<_G*6Ytswd_2=NyfF!C?&YTKH<=rL7~6aM7@M{9 zcRu+J`z3I1#${rT@RNTCeGl%>+FakjTd&~bjtZFbKt1~ozX{k^Ko{T>G(meX0PE?O zB6e-eGv|Tdp^wmWe6%l+{VrDcEBLi`Z0`IYfO!fwYeuNh=X7lM*>g4565ZSjw17bw zpR{S`;Oz45Re$A5E}tzr^DfYt!}t~Op5)Sg#?AMKTJC?;rOu1zk6`Z&k&|z1{4o?}O*$J{M4Ho2LN#w0C>c?-ZM} zMtF6`_{@H#-Nhv~doh=G`54=76S@onx*jaRB+lp?L?CbP_TlqqDRPOeZ+rL#Ty9#u zBY0!ajT5Easfd2gh3p+jo3@+dksjjy=iP5Ocl6%H=lK5&{u#K_NSu3pJhnUqKl=#K zg1(TWhq~7X`k;3Y)_jDlEg}26uD!YseU|cjTFLQy?6=sCh_R-A(mQ-2d!0iSZcn_rN+|?SY&V2Uhzekxvxx+S}7xWW)&(Ztt z_v7i<&MP2C{C$jhiQc`g@cDO8B6*PQy~0;~enfAaytXC0a}Pj&_xTjqyY(V|Cu5Tf zvc0R9R{W*L*#A$^x4?bL`>sx#R>vJDat-7D{D^h-8QeRmkNv;K$Di-^Wvx=$nN;NO z(RaZ9?Pm+rzd$#i0dl}5j>dL|&Uu909z8>MIH3!%#d@Yu9{eLVAM5JxBX57^%dz#C zh#By`z-Av2z4u^+K9<(+N92ml-qcOd*$16F`Tzp{_U6B*c?LbU0=0j?(GfX}^SvNf0zI;^AAobYTkE@bb9du@ALt&zM`Bj!-Gj0A;jWI5J)6hi z6eRo}K_k>NS%MBl1@`>>v8n#g|vU&y0RMoydk*}~t!dp{<8>^mcyzrY{jv_;~b zdxxzb+nk=)8hHsF@bTRBTSMoco#=b)3E91TgIuX;obQ8ovBTzB*`IzN(Hq}E-GzRE zn1p?a-rl?$-kVv^cXL2z-3awL8Qbg+@v$d&r0*@%dd_8h-n3n;;N7df&SS3w{A2X? z<-7q}uz7BNm#sBH?I~khLr?KDr@VJ>2Nllejy{7P*}MNU2=E=y*I5T(t_htn%GzF! ztT@CnNBH^6iCeYU@QRU7x4N|#5sd!A>D!i3V!zd4(R`9Xop^( z!_Q}bN38gH4)!$XJ0rfdSrSw}#a6%j|0#F>IM3?2|NoSck)?N9Sz3C#Y+Bj!PIJzC zU2T`wyj$6{=^cxtOK)t1l{Z#4ieuYqWaXW|c*TDw0{^*RN;1=$+qJb9JJcf~xOM+2Jak4-;k z2_O#@pGsS^JHg%`uI{Yy?qn-+o>%7{k+-3247MR;=WrfBdGzaIE6~@Vb@WTLBhSE>b>O`NIn=wJ=2hQ0_6%tA zJL+8X;5$&yJHy{I2*{!U?rH{IIgb?oHq@E)cNg~20pnm^t2*1UiE4*^LrmWFZ;7rgwGUuV~j13-J|Cn&C2fBxy_-UXKVbtRu!YKvHF`oBS-n< zqz67lsP=k3zS_6qYhIr;=l`HL?>2PK?S5;}E-;5N9e8!h3-PbR=fG!q>&ab4@F7&Q zdyef19Jn*1#sX`rlPbIGKJ*CUBJ>?(&)B&NjK3ecQFwkUdlLSfc~%S=$&o<>kcr zyUd6-o)D@TJ)5Y-I|`O!cgGnv=Y?)DNYPyl^&9E#{e5u;o&D8;HH}-*s%rX<)_~_%pQvaAokr(- z6@iSHKGgn|&6|db&!j$q|A0K7{r2IPBQI!0Fx6}wbk1s@<52IG^+%9b0`2zDhU#}4 zx`NojRyF%ZWWO24>084m1@>f&J6Z>PPqI+;TR=VY-mGpOc5AqIdkCNl23pm-0CZp zC`oxoa%_EI58nS1!42R};EeY|bNpxE%imq#htOqEcTw_rb!SG%Gk~tp2Cq+_xCrV_ ztfRj(=ipml73k^HqQ|`Qb#BkIn5Av$aPzGWZCDY@*o3%oaBqKdbU`7>nW z(^-CevU(>nW+`V77>?& zhCH_U=Iy%pA3hEs8 zm*LZg8ry~jWc#YOs{NWryEMn&VaQ$ z$Psi=tEyq#QRK0Idq@Lz9&;}SOIua{3*kMxEX?M^*E*?GppUToZ;b1~auD>z=2^~t z(>wa+)%#?RllVo*?ja4;o}_EhyZ8CP+QzSfPRa8)zdd_?J&=b|U7mA+z6)xtYF=x! z!B%ujpw=~K7h2@i9)^&w2jx43J|o|P7CDGX;hoXFb%1+07FfHXKCk?y;RAN#yf+#8 zJfOqZIqcIrTE;;2jnJj|srT&W;pZ;wEh099=Exh->l4rwvs9lk;GDig1^g7Y8Z?E@ zKZ(SL+wTVDtyd+KGfQM_&PABeyh&J;Uz)M$mPD@$R7xAF!!+ zcQfecLl*(#bD*Dn=FqLk=8i%&TNl|L7f)9Ij+^7>+2z<5lOrJ8lQHh44^nIqvUjD2 zY!B|p+8MU`WVPctxrVKSu6(f1{d?&Q`T)wEV6+Ekw5D|;d>5#j^~;e9Foe!Hd$NZ% z{?;6UHqfW|M8LS~(R((|WuF~r3VQH0U>@h|K#iS-H@|!rq|{i3Y>zp7iq7*Z$lpNc z9;|D9`w1XIUW#1)z8|W3Db(F((DH#VA7f`>ALiantasblozEWUq0d7Fd$m@C?4z84 zFaeLB=RSY3>c!r3ppWP3?%VLnQfr1#-Io;GT5vCju=PN_MedyQz%PRO9;jc0-n@O_ z&t>(_?mg@ReN#}wwoyNHqZ6|`g<8MB?!D^y4nO*v;n|6u`VusB(K&)-8~&WADc4Bf?}Qdw=`L zZQvcZZkY$WIYyApk>wS&RzNn!-EYP(ptDzZ(}mjerRX5y z;zfk)+1Q)CG(ecMPgMEcr+)S}1$8$OK5h66YW^s%zSjZy2;@FCeH;1!-HU(d*1J+~ZcuTkI(q52L&1GEn8L)+9u^hR$V>7l5Jnx0d zMIS+zl5-=lUIgs90m}Owsu}&akT!ANjk)?AP!mkqv-Uv zFRAAd0cWxo`}7^qPt5fo^_=z1i5qBDXVgay$m_8!109f}%g~j5z(?pj`-Z;2{VNCb z>bjr+5xOX^b}_f#JXE>}TGJ07@69C0@ol3E&L^TO$~QehV#4}rH@9b<l7C9p}0C$}c7eSrleCTT6`!|I=3f%1scnIBG zTv=my_GcP-tkK0Mh0mb!4d6Sw5ZioUPd(6vSKo*Cy_3&ERj2H2m^|n`p9mlO_I}o( zd0zdyc^BS2q|jB+`UE+MF@Hd}4&InQLH)K{!@AxL=k7zjdp&$pVuo5(oU;8oX9rqf zbB+#(utn%=(EG8CLj!v6PhUP%Fjn6CKZQMqZ^Qd;`h{$#{w_$)yPLJ>y^9;6-uo<{y=T<1_b61~s8fw2K0_{Y@?rDmtT9tSo6j`7XVAs&yXL;r z0A5vnBc{*?BdIcaCM)%vFc)fdZ6ypc!aSK*4M4po944==}T7S@bK=*MM>E%d;?UDLy@@ zdiC3EyZ%&_p?q2f&=hL@~-5S=n*EHw| zWriBE_kB6A{?b;ZTY@}=9Oac9;oqM?=5e?4fWGc`KF}|OYOnS?I*AVd6kgf6JNWhB zBQOFM=>yIO_O=bA1jVj5E}kGWay7XX32kee+JN#I_2I!FNFi^np6hs5epd+kt9}!t6bV zmihGo_TmmoU0(IDo*X?8(EDvLR|DANhFN(Fba0k0^HU-sw>z??_jMs~*ExD;2vT#Q zvu=B`g1IAXH7QRWx)mt9ry{T7zCfJ5_Syh>sOniK;L}DYP0>%m&(|N=Pc&;U>tic@ zq2|_S#bk9p0pAS14{~hNvp%EfPgI9&YzneaeW$%s-r0bC2-tfY)MtHaP<@tTtGB9I z7QlO+<}=r7Wa~M54|xh0cY`s&*lBDzH8cE9MV7C%Dp`I8do-&zud`S~|3UoRWeV>- z_xB6`HVMc_NP)IKKKA0QYe9i-E4l1(0NLJ*@fm1Ux-750Rb6cEp*Kv+mw|GR6P52O2bQ$spkauSLac_;`>OF90+WZEs1o~vy)CG_e ztFDf%j=q3bo|}0gdUxUb>U=3N1#k}QI=8jlVTN8PcR&qnnW%nWaen9QAm{k>utoZ! zYol{l`ldPhR#nfs_7VwrXIKO5*;%6D>b@Kc|5xB{y6DUuOA+KyzaT}2D1Pyf7x8G5y@pW`{ zcz3I=hFs9cTI9J}_9pCS9Q%Bzb@U6M>>a&3Rqx(rkOQ(i6ZRQE9i6?p$3gf!RK0f5 z*;^k>!8^Bm^S+k7;BTD_f9KJ6?&n||pK^A5B6w%=_d08suL0b>ci9~FwGAK7*gWpi zTIPuY{{ap799;_4$DJA5RR-?X-(QykYnxx+K5$-pTZC;fev#lA!>5P62nf(T#~0mJ zkixtFF0dDCly%4<)$UGK&&lLZ1OL|1ipriNc=JpF`*m+MV86~95htzybzm>!=nLYe zp~kI8zpz#P&1xSJut$5fPiOL;)U=_qPvg}efxeEcE;9zbefT$s_r5>;zpqoX zd;?k^u4*{L8f;mh5A<7rb8Lm$Pk`E=vwI-t`PQ!fxCMvUkQB>gwngAnTW+%K~}f%{>OSw+<-K`z}e_$aCM@ zX?al23w2N4dGiKxnA5nrl-L|ZK%Mo?Wlr_pk=3B=16vBTc^(11eWc37Hppkq^3I?) zm**L<`wZc)?=(7nmmqr|?Ws_w478QKPH?8k1@z76?8n{Pk29~5$8Q7lN|1&s$HnmGvWEFR*Q3x`F9SKc zDeQgt2wKPHE*FFH+a9WKyS^RlW9U<20=zr(S=6d%iq8BU5I`H4y8zy;vPX2DvpdS6 z5$H>qC(EnYoA8}MS0Ap<$z8e6HnawuBMRJm{G6o&0`gXTw3*vI`t9rC@4Xi}wl4Y} zu+J%A-5KSHYIpwrVx2Mvw1&-DxRVNlj|6g?7=I-+6QLxu1+$*7LEBv3*bg?JJ2} zKbzxgQ1=$KDxU$UXR}TnzRQqv;-XN^UPpF5bJVfVNA}FE=Pq)nygWjJ(*7192 zpVl1(o|V0XoU=x+uXbg3IR=&!V@_*yCMp=8(~omzv((-*-SK063hYm*b8DKdXO9e90xoMQ;S^vzMIM^^)NkcKDEZW?(EcqjC40CRTGW%w;=RWk;3 zQ|JOZd7m8FoM{fw`TODuVDI*%U40&^ICF2rZV%29&@VtPpdD=H3E0Nu(dm=H7tlU> z`--6EkI;=GYcsEM`TMHx2XHQ__dUlJg$ix-)*OMZhT2PQxH_L2yfyq@8M_TxyFIN^ zM-I=ti|qShE^Ws7guJ>>z7J*WFiK#L)^g8_TC+LPe*wOO&K;$p+MPL$#aEs8ZZq(% z*yFe0>rj0Hw1;e;Vm*8pU1Oq(3jrUHgGJb!*PgtSDY^of-#MJw8S=c+TVsoQ?D~{3 z$mhdbtC)@N;OCQJTiB}RD|doE%`r$rbx!tWE%!CI`wX7~SYN*gTLB)Lto}O*@2zv5 z2(|8aplxJpxMS&JbUk!A@zcoi)-Av|viHr{h0sA@eq+nKgO9v@WW-ym4z$&cLvIi2 z-0c$hfZsf*vuM|M3`DJpSHA_h4|440uW3g=1?_<~_@>C~pdnPeIqJwAsJU|V_Few| znPbD>+$r`Hc?{mZGWaw^Bq9bmKto>HRPIra>F2icnX8*iWPqJhqP;#0$C8hLK7X+X&wevb{(db={r&H&-BU4%+b9tZd&S#5tpRXMlVZsu@nl7R}1maNcQX4SgNj1_57V zy1?1eSy(T?&u>-lVS(Jh$1{zH(Km-SCXsoZaSgn`m-!Ca*J@x)4c{rS9C<6$KJ~Yk zDCa-W4BAJZfizU-)PQe;99wOc>L;J!Gqn3=D!c@(ecP+?)4i z2)%O~W1Z3l_3YiTd)x%HIa5k(0jwFo7`6yM5JX``K-S= z7a)%#8&`*RC#v1gK-0-;2j+G!<~V0`z@D=}9{lYuhwtL!{V-1|;C*V)Hn4y71vG>D%`-knZo^0T zES1vFRZ{dlbR(_m{HNffKn(Agea>Ld(7Btf@a78G*CJ=|wN^Dl7a#j-0Br-Hz%SsF zqg#ZZyEO0CSwDSq?CSJ$@8-0}DS2{?K+9f2b#_DOyiY0m2;Ty|vju2?I=^}!^IwT? z9ohj|08n?CLcOEjN9S6G-&`Noya9O(+66TbAlJ)<@q3EAeeZ&6i@Ef_Kk8i?G>O9q1#q zpZQ<`sH2OZ?p{6*m5;T0z<&Lm)9;OT_h%3GT0^H#LCh31f^HqI&MnU?-aeeob9PpH z^>?@$@|93~3c%bmz@8ct)f_4L%9`knjgaSIkC1ElXj9jR&!N`r!fP*}Yw+Es%$fWW zz5}m5#n<^ZqSH17-yg1WWYARLC8=#KP zxB+N}-n-}wAxB1UPdW5d^yYLYCqg}g0)KNXhhGX(VhWI=H~$biXLp}{5TTb^&)%1| zs+c;ot`6!Q5cZ(&6udpJo3&-+UolzrVy_Xh`*hZ2z`cy%zX-kowT}Dp@6Drt{-fw> z@RtJj*@0T$xM+fzp=lr%bm7O*1>(%3zdG}7MW5kge-TuBSr1=lPNB|_VpDG8r*4}R zU+ZMZ0eb_xJMC-xezl{Br>u>wgWlZMabEYmMHxB>ioDuMy;aq)ZV$LCcj32BT>zd- zj(;&+`MR^)&{-pgZ$OJs^PLAMZ(KPazRvHy`uOF@IkGw*&(3qThaMOM#@a&-)PZ|%1AXhj zT%N>+UHN07-p?93^BYruHRwGX^$~tEQ2Pt;IVhh{&0L=G zL^X$JX?z#7(V2Tay54LZ`zi8j2G4gIImO31&SiWDSTm462lHA`Y|3KMWc7V>&sT#K z-B#iwe9F5x>zkpo=K{O=?8(|Wat<{2U zQy>S+(Pb0W%;uE0UI9|#%isB+-Whuwo2Bl^czYS;C5S-4 zC(Qvq8GHl(Rp@=EI$#C7cS)>hRkiKQ9L6n2*C9sT7W`BA8K^zh!O&bh)ZNVtlMA%L z+`gQp2FzKT_4mBS<;m+#vj7je_*qxp8290f_WE7?3-rdN&j>02 z*SBWR7PWPqL+Tk7UuA7=tbcrC+GtmlLsX z0^ad~R#fJ*j=yJlJ~@8Qy9ItpYxe&)fxdh2bLX`XJ|k~LT-i6geS0=3@c|6tqr7Oc z>dC&$IS&-reFE|BVt$MK@b0?{%+&^-a|(=ecX_DJxQ;G@_pBrQmSA_D8F=TD7TANx ztF!15a}+l76zHaqjj^tM{1-OQZ8g-I>YU*WFaz%+)yLg8CaRs8@3+K^qjx8_z(-K| zD9}6l2w2nDY@+Hr#pce9SGNj&3Yw!Y=fV`hZ$szr(;=tka5d|^R%Nrt90cTV;WPK& zajZbjiSy}WH%?#sawcmEW3{ayb_x30MD@GJBKTEcA^JA<6b!f!4>(3X`lQHnXS;5)%Aw8L?!!6sX@LKI&l>0hx)7@Kw71d+-9zww zpw9DX>kA6%MfiK)mZI>rz6;Gj9er&UDfY5Y>=_s#uE18uF9q(wIBV#aLw%3z z#oRg6ccBK1v1j!&@XGGie#*Q79(x@`zqH6&6=swlie|c|0)k95x_@IofhEErm z$3E+=N+0o*XNaFU?8(>+S!#agOOds8fw2X0J`t67yASQ^1KizWsB?7i^}SBi~1=$-?*@*Tj(+MQXQ zwXG$!{y1phI|JGv!j=K!m5(B})~b56wsGdI!@Em;^jEe<0cxNJI>Y4S3HbNrK?d~e znFBfnf(Y25#*FH(4!#Y&6nP`CSAA3TDYOn1*2J>_T?F8K-=^z@BQxrdul+ng;4eHjw1MF#CEhP zgA834n&;K;I_B~n)!tV>T;;68n|~?szLWa9hn#qG7Eou^&scYx10U~;IorckKSRj& z5VflJBLsNl)6sjU(urz@ZO}TheH0+du>t+foxzJ4d_?(vAot1Rerr(oGzN_ALUZT{ z)SS{{xYEsQ5yKffi+X?GvF8rF_a_DBszH0eCm`$NUZz2Lrl_p5iTEjK9;zPgB}HD) z!U5j~uovf0zXgn-%b<((g-?m~(XTeEk3zMRLGtLo3SWB=&J@4^Hs?~86eYhHk}V^{zCC=>dHrboOAsqNX0X0LHl=^Vy$w ztxyO2zI3pSgQMW3ePE4fxXRUqpZmScJM5DN{tqx$O+R9+cPe(zG{-ir4LO6?uCMw! zx+(Y>bl$rKAXTqDhllon{T84N{JEl!_73vs1bIOKM_~J%c5L3sJ_zvJpc&M=G6Nq0 za~tnHs6%_e91ZCC_zi-!z+C0sg70GUoZNMayb@duyo=__LuHSk<}vT3*lz&ZBIFbr zfal_VGI)FSoUPvmD_WI4!j_}&qbrPq3VSH>s*f4$=1H-4puNc_?^FZ%ROAs*;I|Fg zJs4vi&)A+upw?Z4UxvJOE}l~qDvY5s#{PPs0din$`Gnb@ z9mw`MMC>#)LtoZ}M(8#H`|~`ae0E33TR^~PK6Y~#*v&Bx)_^;J^$L9I*!u8+_yU+q z>b~4Vj*q_HANwu82~jocTpfM;*zIu%)SPY5LAI}y9L66ti*4A-JlL%@3hkj=+^Y5! z@Q*-Av8N#5<9_VrMEuNaY);(L$!brF)briU0@MeaKPQ|i&*6!U@CnG%ppI-^^Jh}- z>HxXw@b1Lj`EFTjkhmUl4cLp{h$}&V_{lxB(HrmV_NGmLd$OiI$d9A*@6-m=9eR$Q zr*+M3&rx76>Xq~A-rk1nIr_hur~`ei6|h@R-W;p3E!PgLslNQ3Xri*0`RT)Zoul*b zP5taQgVvx0`VNTT8=%Y)py)g==lADsihlvtx2m6ITk-X*-T8c|{fq%;wO?c8eaicT zErhDR_h_p&dK;7GRUd1JSq!Go&EOM21KC}BX7*su_EqjWK#|>(yRDcMrQyitN1h*oF7o z;0~;_sD%S`6|#4@k6#Vh`o@GnPw2fX)~KT|&k^dLGweBfAN7myn<1_Zihz#YIGgRC5{tFbSi&D%$wfxib>U%3eQ5pSGlQbVWQ#>bzj&R>IWfi8sF za~~{FH%xEv_SF{f(oxWZ??T7$>%a%<__$kZ=;M6$?|r-0&wdi(J4{ny8{NU`To>qS66cjR3v0=vPULABf8fO`DzfEu@| z1;wVWz|WrDNdwHE^{vAjr!L^r(blTs)%p0Icz3*${(hC^RZaU@i=65gs(a=8=nR9v zINxJ?a34`?_B`RY4Oi#p-mPcbHtlb^TTf_ZG^%K_6zm3j2q^$!2x_QLr0YA=f3Y~t-s8K+B$SJ(@ ztcOldRDIN1)sEd!G;43*JB3^ao6-4i)J2%RFXT2yS6|S^X9?69OBrhJRp@f4c^APK zP-j%9tp=Pe0(JQE-x{%nR;9PLyN>c|rZ#@^9dv1^=0#I})1A%7{ib={A# z(40K>I(PS+ZySDn;Cmon1B+Uf?_#Jn`|CozZ{=g2yxNg54PY*3v347pffT>mEUiye zxkk0YYtMnt0_qs2zxQP!ymOeZ4&05q^sae_oM$UE!sfmfX@mZC)~5|M&iD>~?x4Uf z-Q1#Ia1IE_QK)(h*sbGPPeI2(if%cw?^gMQsyE+Bo7M?L8U}WhJZ#SA*?O1s--cY^lSA8J8t9{bu7A(f*@bn@y#N}3 zF@xC0CaRhBai2$l2DVXC|Q?PC+X{f?mb38A|C zjaGHfGWD~zBk*g$JF7dD*WbPLhpV%)*8<)-mba?jw!qt~GpG6NIS}WQq3;3jcsV;% zpEkY`dhrlC_nM=3f1}u>)|!%nX?S(@R`ou0k?q}@gXpw{0H6g(fj!m18gh?MFpsgw z*3mBmQ`pRt6RX~s8W;!mo1u@OQxo(^d~eQQ8~)}smwg&%{8Iet&_HYoy@Ht91iZP3 zX~65Z5WO~awN|wweF|h@tnvnY>cAO0(DJHiXsRE)JvxWGX@COT=Yj7=15Be^-I|>rU!NRW*Eg@|BK%VL`S>}Td28sp@aDBA z{mmD_tII%5Ap6*p`O;Zi2q@vZ*tbHJbM448$Rp4(s5t`q9&&_F2ijKG!k2S#x9*~b zoMX>&^z2%lyHRIdXX>IeuW>U_pGkDSHvu%z)n^ff>VFg0XjL_rW49M`+Rr%h7`%1t z$C=%MwNhfuRX}s(09DtXjLNgCL(NkID_Year-)sEKH!&u)$-WQJp*3P?^YO7ycin-S%X{n$BhZ|fF1mU6rO4K(A-6%mmLW$V!+#|{ z9e8)`IhM0Q8^HIlK2iP5Q9cSEA9r1Z>Rz_7s}ImL2a`|!jS=Cq0h>NK@qMWKT8VB1 znqhYa;{rN+aEIPw`)EM@t_J+N=#Pc>{@BYD=z|D9f6rgus^r>4_5117=!`daHXAQ* zthtSI2ll1TZ-MWi`-|Y6TRFqF4Fq8SGoZ+;-Q=OtMNrS9M%)%?cevVZ2EQ4dat?KG zi=^D;h?oNXT4ZPMzz;yZx8}&O2Z7!`glBBeSx&84A7=^h1H`qV^IO$@Spc=4nc=Fp zG*mU5&-e_z_d7?gtw0~pyYC)+A6Ub8p{@;|`f&8b_eWa?w9##aI)4uQ<~qkXym6Nj zUl1F?N6-M(H&chrn$|0z)-*oX#N9wS(K~AhO z5mX=N&agYz#bB8>9`krE1$O7Omm08NW%(Y|d*a>#h=4iD^F;J~?Y}Wm-6MU9Kx}LF z9Yb$#<}zoFZAJ?0w*cmJHglH!;NMfvGf18~y!q|H-p%39z5wl_H*X5uTOa*2x&Zax zYV`LVx&XKX_hs!>t%?pJj~NH1fb*JjZZ{D%>@g7A!PW++nggA2Ir;$Yf)w6=WAQxZ zwb0{J_CgMA1vJ914vh5-t!b>kN7-{aQN179^w|ow0Q+3ns_xzS@asVxc;B^0ehTVb`t;!aE<}0Ni@h#wkq^17g)PO;Jnl^Y@4yd% zOM(3xqu#pe8}QzP4CJ67-uKrXrpUf?8M=pNbwysi6W-H88*+qQw2?jQ0&J6_%X4(V zd92l$sLsuE^e%dKU0@xb8hjo7O0WcMMc)NIU`_Ada`g39Rl|Hypx3;@mf`Atm){$7 z=C|$?e&(^J)V{3Q8)gO|EwDNFg4UBgW#o2F_g)YaK{NT`N)Fg(;QQzz;ueEZWb4<} z@r)yv_Y1iX-v&K+=X4&=*ZsJsIy6AFchH+-8*~v!!)(7$dr5SIA{>?_AxV2 z?Zf=`p?{8T8+;V1_x@sXPbp(FMtcqF3<2E~u*Y$zd3>LiwJK^{ioOf&LknZ^9RVr% zds6ba8_%)<0=f+U92EGBK#jL&Yo+k!n)~;|2C}+r&OV#Z-DSj=XAYkO^V|PCY6$yL zXAhq3Ty1lCrV-Q`>%&i;vER8FWb#6` z`5mfHR5jY@oVSB)jRJrDa;UqFKpl8zoZVg`Y~2Zbf%WxCXQ{c}b2(p*ji+lYcI()u zdvISV)Ex4j$714q7yTA?hB*s#S*ZShYwcTI2c5Ouv-v~{daY_Ed#-^YNOdwa~kVt;p7MmW{;a@XoOf^x(_f@C{JR>fB+5Z%PjDR{`y#>tK&S zsy$S48@Ygwa(Y2OhFk;2b&+f2(bu09{tT!?SAiNf^9Ou;v&fKh(8E^FjozA`PaQNS zh+$@R*3zdi2hi35Q2^9QC)$TRZSb^b?EQ&oayx$8+%9S#@{|G2aS;3e^XwKiqql$P zoyWfC_+wta`t57q{_*8Ohj}Gm-pdW|-xjX*re)E!J&-iM4JW?_QIaexo0F zUh>GfA5UI#`PY(*?*4M}f`fKW)~{$L-+cHB$<3F2IZ5B}h2-kdy~=*hx%2R3&6yjL z{!hk|v4_8t{PeCLmik+hq2#iE^^;eg`@`h<`+X~U{hCje`aj$~KYmAgZM^ZMKgBm* z^QU;%$G=+E{MPjs#c!VeLwxO1ACGU3o|^n@<2T}GEt|LSji)Ec3om(lve*1?CHuYW z(xm^4&m>Pf`i`gEyVncLd`ridCwK4mxA^PF-5K9_%Fp5*{y7t$bMC8>i{JmAGe5Tc0H~p#R+5Z@6PCtB7 z^R{1)CxdU>HTlXVA5D%QIWOt28%ef%@kh#BubG-jcKAZG`HN@2vDvzAc{APi+~!?B zzb^SoZ)ur-?Y~_ZFTVd5@s+QCF#hThhs3}7VwxO1^I$UX(Z`baY`!-6k4IX`;kE6P z#SiV>eEvH=-+bu$Gn==(_}u2}ciO-C+=C{QjlX(+S!e&tmn7Ziyfs<#xs#L6KC_;j za$qs^jn?LP#UZ;V_rLGe$>VERCm;OAjqxj6m&DO)29kYucu6vR(yqx5y1$8|`5!KO z*|z*ON%F}zCm*}G+JpDxl!trqZzs1)77Wf$4t>>|k~0sQmrVS9O?=gPkHqiz(t_lH ztxJ+cS3D>A>9c|3_clqh%$zA_?EZHyLxw*3SjOI}@pDg>kaK}yYeFH})U;FHK z(ay{IN$2V7lan@GTJ4xRXWg?le)wf8;<4S9#+P4y!_d3OpOL)t%b!nva`#2a4bk_K zuRXpdIpo$eOaG0(|59?)(y8QtD}R`L;PA7`{Py~ztve?V4Ii4Eas4}z+kRay_4y~C zA78vnHm7=H2fu`KSGC z_vF%%7bcfKbX)xGfN$s1@OrCk%CGmlWy+3aJZ2#o;W$#R$`l<~{ zZ>J9>ji3E8eqhsM@vA<*FuCCmyC$E`FNsG7oAJrde?|O>4}UV=VfpS!bFV9tL$3Qn za>DSg&42vx+H$tFe|~Cw|9`KESH8O+Z`ko;PxN-qHE)jpw)@ZH`TY=B)@Qx@ zl~0ZT^vsXO#c5~77aaV~_^cNn8UO5q7Y;ecn^)Z)-*epW;vJsx#d!POK3T?`_1!nd z2aLZg+2MT$CI7kM&B@c=b3v&e{h#&CPrvYr=HGgkH^08?Y0cA*-@Ey>6aJBOPq-jH z?C(1#>pt?AI6v=&$=hz+Gg))>U1iLV-f>WT=xaB|FFEb1xO3nM@h8sw{LuG5xBJru zzwvPVlenKJm>B4ZToyF{_Lao#Xs2X&+)b2JS{G6J}LR?t8PlR z`}t3j)_bRtYYtkOoOAEvWqxP)fAO|kveNdIXWbEB)Yul^vwHWW*1k60{>+z`wQin# zXL7}}ljIfMJK}$>9gQEmd{gt7JO0`{;;zA^e;fT*v-S47n=d=(qs=9JA9mjF!eqCB zYm;Q?3s3a<`J?Lb)q|JCfBVB^{D$RQ;~N*=6TkoW2gl)SZ;4NQ?_c6u_Iy!tV*8ln zu6LfDJa*UO{);JQ*if8y$r?lW?Ja{X_PODA%4?Q)!lKg`#$!H_~gyMjW0Xxy~#13dv7xQ=U#H!r<+N8r>o+z z4}PlbVatd1f93@r+&dZBZItiInPu(`<1dKYhioeCub=p8e9f-A$IDLL92Xb-DjwVa zSMi&_eqFrbfU?e=YRVt$qQeyQ}WR}pPBsf#ka)$Lw^w8zhOAJ|LaF4&pmrX z@{#R6mF&IO&dE2fY?gU$ANykbAAdPJe)#6&;>#Z1HQw^v*3e7$kD}8auE!g`y-$4S z9z*e2!yg;^?&ZVrM?U?=-*W1UnjgG( zmu7GA&E>4epMFL1^zC+U-uh@gDoeuif>G_|jLckH0eI=ZAGMNqpZAjFvb(FF33KnnvF}adbEcP<@e|L! zHhHwMXY)n%dh_62p56S>Gya?`+c=V(@cPSlTD|Xv_|~&7i$C>~>*D1<|3RD{zUYa5 z%=Pf^HYMq{Taq8&^TlNE`8UNk{dm7*yGP%c-1*UWC#Pz|BPR^^6{a2zB2Um zx!NaRSC1FH>W}4jM%`QE)8hwLUlYGCIW>N4*B8eN9@?d>aq5Go{hwODZafmfF%6rq4+;GAJTkQesc4p>rZKZbbNkuT{K$m|HJP~ll!*)ce3@i z+mpvPJXq>}`mbB#qYgSax#EUZ$u+l}k^JY>`;$#4Jr?hJ`aX1COtM7JjF`L;Q^2zA4`REtkc6 zK8K&7^Vh}q|MJE0{kwc5e)ZIq@vn}1wCw*!zuWs|cfWi%K4!%a;#=>Umwf!5JpR__ z&yL6MKPW!$?7s}1aMxM!+U9lf0n2|FzxQ)z#Vam8GXCuL8;9ck|0{mztY5^teFodI z@iN!=#V5tdoAyi|`M2YeBM&<(x%t?SCHuc-?{YSGx#5BLBfv(Z~E)j_!rkb8h`rRo8y(ooE)Ee*guE% zy5_88#~1x2*=g0&%DcVsH6Kjg``Z1I2anhm@4Mf=$&D91Q05%{*>~cDzuAg^TKjuE z^sIJr`M!H4FFosr@%B%@BcAxuKJoLvb$I-a=S{}hPxAQaoo-15_Z zBxjuVs^*vWJ+}FUSGJpv{bAqcrysd6>0hvC(rF)&3|_TQa>EW$(t7o`gQ zWz~k{j$=lXXKwvya^Qy=$y3kxbNqsrOvcG^%bU0U&&AEF_S)2ZY?n7Tce?ro%?~|v zVsi9855{kw_kwcgpFQ*FDuImryrR7dY>P~!^i$xnQPsdQ^}>*?A`p%g`xSn#V=@{-+XuSmUlcae%qZd zNzQxMX~|cP{AzN3?W*MayRA*~!RJMHezOd2qQFWdXjC;GYT?md&2kH0%v zf55urCl9|mIqa!V{=M+PzxGZ(|NB+RujdUV=RO=IY4+B5W){u%+C^x4)buOTW-grVjjZeEI#=?L$AN^0GcP(gUjFy5l;5SjPJY94 z*L?NCv72sk2xy2>hsSiYpy9yX+Ctpc=P=IPj9|+=nc&s@_!~L-0`Ml>hPB* z7wz`u)@E{Grf;3H3bYk&UzPbW8Bxo>jJAJ2NC?lE6~ zb^P0NH^g86;Kq2j({76EYtM_1eb2u?VgEsVMsmW*)i=)?r`>u~e9h6P#DAZjh>y>< z#BY1kSK>chI~;%DhIf@QzfQgtPwfB9WXH32E&F=$uiuhptlf4dnU9$6`FH0`C?TWa6 z{!Yon9%m+By6St$W1|lzU+p}l`PhHv$uHj6=jZjk$xh$X~;lU1+!RI>E-uS-6%;_vZ~Pq;eX^?i58FPe8}{M+O2i+8^0*KzCc55@oFlk(xO zd@9-Lo)eSZ9^Ehb-NO6h-ir^6=O6l!c=y-+FrN9y_u~t~^6k%E_;7h%@7#P}eBK=g zCcpUeX-VhCb;%{a+#|UsJ|^C@A&qak>z4QrJFSfW{q)KB6;rQG9{cZOlfw?-9lH6e z@shhwk3W4xFTUcAGvXuPx?McH-LQgSTP=- zxzDZfS5CMoK49g?qjvP3_=8XRQaRsE#~%}a?~lvk!}q;@X!BD)J9NpCgOf+s-JJY> zaL4A7U%jX~vG?kzBd?6UnM?oRl>7dVMnSvu*J;k6jpVzxVdd-7ox9 z^BvQdHYfM`LG!%+mCb{mdRTdXnwu|;@96f*S-i7z_{WtK@n0VK40F9d-gNYac+FuC z4vqYFk7Va(o}0Y*lsn6NcJ!myC#!aUXL8pYH^)!;V3fRK!Sj;%B?lz?uG=x0S#VqY zrJKGSfBUfN_x8EG(icsBCjR*Q=W*9>iO;?IS@FL@yhpvoQzv8p;X@qAFuuf zc+JtR`1RlUOnm*J7sm(On8yEk+v3vqr5`yr{*NC{#J?Th5MOuYA8Y@6hG)F;`gr7s z=OpuwIUu?8U56))!=97$Z$7kn{3k!z{LLeWHox?_N0Nb?o5{M?_Q{rgMv_~PemMF5 zUj~{#JN545wvX+VJnQK{s{hAr&17if`N?aujme&atCH`vfA+*V{Pjbhh<88eoAGN; z`eMB1w`u&D&%QAJ@zf5@_Ad`=?)!?+eD)vyklc6m|3}7`$My7laknZ(n>OtWN>tK) zpBs@SOGrpjk%}Z+%D!t=+BX#KC{&^%lC4CRY}pgpLZq@}DSGbbet!Bs&+GZ)^SYlq zcjjz!X6DS9b4Iwf3g|jHcPGwDh66$MDERn^hjZR*Kcup@NxhFHTD!`YPHFTfy(Q!D z?VK5^qb_0UXchJQlGkz9(Gtts&Pl{+s-WtC2b88HA+_od{H7d+X^A|ZWNyadzh5w= zQi@uiwc(X*5?rQjL2X$!E^am8bz!+^6#cZ^PZbRgbSJkB)BRm3F(-)R)(j@k;N$pa zUB+ApRV7pJ9n@C1n@m&UXpUYA{qA(6r@ih&L=F|3;(CD?FLljODLJ^L4MEcP6y}hK4&1;9ZSw)*O ztmyEVUc&OE@wk5T99C8mS-1L=y3;7KSO0_Y7LiQglrh*-bP=O;-l6c`M3T-ZBDrnL zX>MyAjj(@1AAei3`o=5S<{6V%ouW-VydzEwb&oKo)-!rkdFCr(o1-!H$!%IFKbCcq z@L+GPabf4rmtaxY8b8e7?PYk%}iQYc!mvM!qELq+@o6t+IWp~kxivCc-|3Vmp`AHc9(8WhRQqSZ@GY3lj+ zcwWAY+56WAFD4el|7QcPHCd8}sU)49|A~3^I2J+A`rj^dwL4b6d%Zhp0^IL$}v=ApXfEH1?Cm zU^OpH4i%xZ4zp-;@HCRTuSx^=%JH&zp*e{16X(+1Y17HdyBjvUqPacM8<9 zH1%g7^?p8|Zh7zL>#JGSU`)vfnq)JNT$?8L$fl)Jo8&qNQC++SO(~3kjdEP|tT#El`hw?>)^p zn4*+rc)Bi*RYzO=a+r^L1)a@<>Cpe|sN^my9-h*Q-Dv8}#u2H@$Prlvhkd6}v%-X& z2LD0Z@hgZ^2uFs>S0?HBRJKX=0z0kcEqnd)Q+9$=DEsi&S<2@2%=LF>P+4INmCp;K zzOT*ch=~Ps3dL}J(;D8&0cs4X5UCX*yZk2GOYs2wf#Jh&;Qn^rS9HcFv=Yw_&tp)tH{X+p2LGetAzZ zq38kbP27p>6G?E870%nnYU<>`a`Q8XPsOc{aWEZZMZtS0X1&_x$kp0mLTKZuINhVA4?LuXbqO(^KB5NjM zma8!ejNZf2aU7-JQK$U=t@!iz1dc`dLg(fb9MHXu+C@rKdee!ts(i>(U6pegIw1dS z4z#v^Lc}#O8WsBj=f*@M{P{>MFA(}vn@68Qf=mT|ZwOFWa3T#(y;M10&H|c{lR{N1 z1L^${c}g$)$OO(kglMxSOs{x}3l-`3TIUMY%u;4(XD1V*JP;FQLSWYzfG0cjdD;^j z&!Qz)h74#C_1nIXp2}CDdmJ~fYfmN_rwLS5s80?Lvhh)N8X|&%Xvt%4ZEi7TXT@)1 zmko(v4X^64>ymCTy~i<_KQjcQrDNb`C|u7z*s0>``JKqU8UoF^!8|^J{Qv!Cyc$F= z3^&l1_FyWlQ{em2c;zE}+WiITMdK)SvlspKR-k0PXsj5aEHdl|XK(2WeFsa|-Dc8n zcn=7g-=BWB{((_I7rYiH;?L#jP%j)#UZ<9lH)kIUaQ$tUrw#4Lqdov=RzbPau) zh$8WGD4uW+F=|re7h+7OOa4M{#%rRW>FjrdA?&sII;yxML79)7P;dAEceXB}X!`^* z?yF7Nvr9p>O_rBV)?uZTGbwhDq~WD{bkpb%KJ>Fgs`FePxB8bWxMolfy=|okPtiw7 zQ>ND00BQFA9T8Sz*a7M~+rihXwkTwFv>Gz^-quxL?xH8G&;}IlWrhMhP$@%{CxBFW!{d`VZ0h7L{hH^_q)SA z6>;uTGSh~^Fg5v$8}+UjHsudvoBn{A_@x)l`y7YyOl}-cyus5dkjrzw^^kt2N$cBH zD7U#6$+Sr^HmBY(e>Jk0&f$yUT+xmU3!%?C<>h?T4IhA}hnZMi>A=HSv?Ummu7AM0 z;A8xIn(@6`YNw}SGL zTR`!eD(v>DN163hMyh*1GLG#<KMoL z;JY&|5o#`eblN7F=IXie=^z2PB@m7i^E0seBG$kRj>_y z3G@1L$S(VXi=D!GZpVH*zd6HrZ!Kh2tPf`1J&$8H zJQKmZRib$By@z3cgfS-9e^^e>Si_9_$zozzB3f6L@c15HQNoTs>3Ev{0IN(3FwsSU zrYz$2cC*d#=jSE->C=RMMKaV~{?W2lw3cskX_79}G3O+c`RX43et5o|TCap@A~%ATRyZg=N{hS`Rct+oeIoC=v4R&8aDJ7z07lDzh5@v^-z6+Hg-=;!m{ld z(8=D3$=B-G66sejJ<~@yfFAS|J{RE*+g2&zx(J z!}s5Fk+(_$iYcy1(FYONtBqt-e1?u=c@q9=_dyC>mWO+>>2{G<&mov&Z<@*d-MI0VZ_ci^Ym9UM5A!}Pu+i@37QjOozb%)@iSvy`Pz=U|km9BN{1 z;q~JzMw;gFu${H4VL4?OqMV;Flb>(m?1fyM>h}n=ISpMa%kbIeJ~~7nV*U3)R9ile z8U_+m+&zlPjnZHa=4e>$>zAwc`yhCfCGx|?8Mq9*F z{(Jp~^SZXjUE#DUjY+%frt|27K9+mOq2_H0`s#-AG%b0kiR%}K;_%Z_M0;Mxw96Ox zasllB`h(BT)(i2q|MN5Lk3d1i48HE4;s~axNR_Xz zlc|E@ZN_jCmxop48Ae1~IJOV&nTr$0H1KD!a9)xRu)>JB4UE3BGUMAz42zNid+4#s z+J&3@ufk$jW>48!+c(UZo&B-7UAQg}pAn4kw6(Zsx*R!Y8!=)3Vz5dx@#F1%IM@p3 zfj{GpGi~C{j7HTPW?G;s_B{IbH1;rmRa9bU^SY=rPMiYLHEEaS3L3R<4`1dMHJ2%f zYqnGf5bh^8pBT!+K4p-=%O`ZkdejZtg2S;EP)04P7KsNtRp`$Wz?d}`XjT5n&k1+L0(m%g-R9V&Fal+l-ZH7vCLq^#GlEwI;=;WZ z=vEm-nvF9VvDMG{x`9#$8Si<*xPrS4t;`hLZ07Tl-9GA-8!VJ=@ksEr)ejtMX5@;nOmnmP!jV9=@~JGCF?FnbHP-VK$fu=Mw=9 zeP3R|-s8=D9l^WFXk#W+-H>m)HnQ}=Hks&QF84m|F)bxtMj`;zo)49fEcSu3wdlOIV z%l>N+=Us;{f7YTce*u)*ym0)=PAsoKh%JA1V0`u!=1|^f&Ih>|J2bDNW}FtO9x13t_hU5weQk_V6X>gCK1+M7ev=wTLBjC~hXroTp26+l{H*B#Y*k zc93I873n7|!JnieBu{C;xC8oIIE!R1L|=KAbjn zDY7LSwAra+)aYq(683NS$^3Qii|P4$dD@T7+QmO*22N%YZQe0&rio&v{OpUVBqImD7<)s^;jtmiF7mDT&tLz`&#gL(0~d*OKv~?Uo41#)~Aglu%I>r)f;ViT10i9Sibss zmG84xd%~Cnu0v5gP`KVn)ygB=(;g{miG1B-t{LbkIEe*z#kHVZujdyPXkUD5@hUe=t6d&*gDs2Us#kRn9WEWsmdkRk!9d z`qo4F|Bhxt+h$w)bY@I`C$nOt5uzO~B7b@#emE83k@Iav=1>!3qbRhOKZLJCV?`V) ziWkw;hpXt4emD&e`AqvXY*^)2%DjEGcCI*Fq)Jd%B2A5xg#Mhj-M+lMT4j`(L+<;T zTeGFHI#GD$ch*=O|z8A@iaxBpW|`Wa~eWg2V!^7 zF(&i)RQ|i^xoQ}_mZr@5Z%nn;DxN3zmM}}#^eIg4jvb6!i6X>et}#`a0~p^r6^zg> zL$GKb%EudF(0O?{e%r^ld006GyL`Kt#u=ySYEvDJ&e3Lv&kAJ!jEQ0ME&8)VT&D2& z1o|NODL&rI6eKEwNtuXOZ%UcbqE(FSmOvCK2z?OyLiI^vyC2_1^5F|6PgD(}^(vSh zXGHR?*YMs&1Oo<4g5yS68g--(MVPembhwW0h0-IRu_>`PKiu3-wtj77l895g81$D3MX z4DvNahx;`qwj>(wZ++y>=6+&=-Y`rT6|Rr}cpGIF(ZlQA*f2(Vc?2xx9K;*(Gfcd7 z3N!xRKwNN_g{^2hlh$3!B=s_arP6G~j=RhFulHk8`Z?lZb1Yq3GLf2|@8{_&Idq13 zJxC3=Z_bC#ZJ{qrfN$lya4bCd6{8nE!^(n4q2V!WkCbmB{#H}$G@$2ymc-O~pa_o<|DQ4KNnvRyv z-w61vNM5N!FsQhLIa(wE{WcAp9OH(N-$96bFV5>{%uO@?DK9aB=`%Bli4#pjnyk>L zFfOzl>K`9Kv+@es>m^BZ*L1qx9Y||ZgJ`YmP`1`ekB;jh zp?N0W4LyusHwR<*%Y3F$-k9k;J&D=yTn%epkH$#r7tHFe&rCt6@Vv_V`+2_YB>PUw zF|94kk|z@(b1MvH19IVh^)?1n=TVW^W9ob6IjIe}PWQK_(e)L=*yPmVsu+Ghi`R3* zOA(mSF#}UXci>M-1v7l^SLW}h^@s>`BrYtH^qC=WmN+OS=*w8E=@7L}hmP@nR37iN zR0_^!I%|dTVse#z5cVh%)8^_)XzeXzEK7&ub*eDt;#aE;GP`A%$`Tb`22oMNn2PXu zJpJR+*1-O21WrsJ0{@e};q|)+BhiVjfDJh7ehw*4l4N&0mTbKrQRMU^d>>d;C~}z( zkC<}NUU0g(42u<`;H0LB^S>sd^n@lBPTI)D633u@nFEISi7*+RI!yWIemL_d90&Jp zKzmLwBc>;X#w2Srmyf{I)dP82&Q17>N%F$}sdX&DX#YEScI7G-wb-Hc{tpB>Dbvjl zhIHrhKsv%KwKy9n)^-T!Cy1ekb+;6%b$lny!u=~2S{ z$f)E%%zpG7doO^Tn4P2I?s=tpj6Rp?l#@Vv=*pc-cRMPTQ$G{o&ah=jash^2a2+AOT9&pa zNYlg;9U8giHSXy(;7P0=F{@1}XHkDTw6O)n2{AbH@i{&m)26N0Y^c$BGSzbN5(*)= zkz4zpeV?u%Lgs^`X#SHebT2!R9%_rRWf>;y^o=g;YpbcO+39B_p>Q9vLH$U6yfIzb zs6kgbf7<=KC-A#Tg+24cm7R6Yk9|9QEqg6!3R|uGjAF8tQG8?%@}J*A=$8(>ydg`0 zN6X;R#HyO{RddD7q{u!`f^mX zSy0Dz1KKAyfqv{;OfeT2athqU?A;`d>htOFoYRd$PZ_#kDD?ZC4)a5rLKq)!X_2)J zOV`Mt|43^rQz^qcFG-rNE=#kn+~@mi&wy@ps5sN;w~^$R%pEc>9YHFOU-I7sW6M8h z2(90x!t1x-z3@yQnm5FO+A~7QC1W*hjT=S*xk8__;CuL}W_;k{M>3uYRbcg)3 zP-x=>X3X4&qfawnA2(Qk0XjiqL(HIz~xP4hEn(un&tm|@inqe=So^|lpFyk|-Z zYJ0`6+8%&pX&drvO=;1~X;eMlp5`izB_)Mx_+pTV*@^np@M}HYxS!3_Ionw1^WFD; zIz@c+?x9^kAMoVDl@?7*~czLzKh?uo}kh#^mv-@MuoM6}1e^i_DN0lxNdcFb0*7o%R_C zLmIFpdnqjYsN$J0rhmzcwFo+Y z40$PBTzb{-L;mV*XxM87DgCSb9Hu(6i#e*r!sERa9OgGO zyDSPZu|u9B5|l{)sti3^#@Qhk0_hJRCcgR*kkHxhrn;L$g0kb}YcQQQ8OyUxG)L(KwZsgTWa> z+g?DY$iflC{8~VN3!JI{-hTA??FpWKmvigUt(OJEdZFG}lukg@>}m8_KaDOqmy(Cw zDXL<9sWMKkr|)JxzX$DC*L&n9fG_yD-48{FrW@quY;B*XY~=D5c1y2s>{Y*e?1gs=dAl*E&Vbx4EGUIL{|advOv5A#U~L`_ z>l0-dQ}`BR*NM~9H{GZkm>!$gbXIwQ2d3p^spv|=04XU`Nb}{`gRwF%gIn* zJuZ$@C5x8z@}*e|#p%+a1@z_gA{y2hK;5DdFopOM>&cg|DfUFIO&(sXKZ}EZU!wS1 z8GQEX@ZZ+%sldf=Pf#~rg6xX-k*azKJ@V^=&GmtJFvu6K)myMkA{5Q@HKAvliO~HY z;oDn;+NVB+*vK1boe=~J_d?yOo!?+Jr!QGWTT#^INff2=3(n>H@MpCol~q1LQfVK$ z=(I9%1&h^_z=&s|xQg@jaQ*F)u1o<(*HG@&{WS6X0czb9OGi3X>7M8Y zWijVSRPRj22QlFuR8aqkasxjuvI_wHeY1ENf^x%#qm0GiB={BeqZzH!Y2PM)O0$wB z&mW+RYzWEN#&hw*fz;q;P8KFr&|f)={%$)>QGaAv%@qXIqxj8|d#;7VGv zp+7rQ-H6rHkzyyQB-6z&UOjw2Ns^_6bN|JF3fi4qBu_dEjp-W~8+vcwExZUm0&T7R z7}l>DPxo*!n3ujHYR)?zXWBv&Dq6RhcBrI~R1g=VQz^8cuTGf+gDY3y`BRb-0~Ben z`Wy6m5doF;Ke6e>bZTq}qXD+g)JuE_QP>p>oN^vYCndBDMuN_{Rtu1=ynPaZp3prn*cI?fL%Bfh$a<vziuY8| zPU{jf`4+>M7eA=PP&*%bW_pf>d6bf0%@IocluW;LSUPmeobul;qM*Z}^!a)iwc4(x zZ;2YTeM~M|%syeeiUtKup2Wks6MhRUwjjOX(~k3Np2z)7ok+-^CaDe%`o*4t?#tHDOG$}={zh4mWj%?BffNsJw^a3?~g zMYz4N1wByON6kzov5$r4_33MZu*tm>>+00#%IgLLUzuAxU!sY^n8tqT9gta^!j|I?FCt81v=msd12_cdQfP;T1i1R6Q}_LIprveMp58g{8vDJH>x|Cw_FL1n#Y zO_UmKgmC}W(P2PFs(Webg>o7^zmTfq^T|WkoE~V))6|-kv^77R=5leo2c8l&<(|V9 z_nR=hu$|&Gb7;p8tsFw7oW6rT@j0HAmLts*(c@ zHMb&1m2OO!^A#a~gR#K!IK;gE@a;+#3S-0n$w%;P`>sWc)|=Ca2MRRmXE;u;`-t`l zgGeg0A3ZITAp64^kodZeZ{JsS0ZsdQi_{f`IU9oZ13&af@8yf}gx!mRIS1h67mcck zL!mRwj&IX6FdyG^KErljZ~EQ47$R3zFspRSn7s`H@&2td#Ec`Mlx&9Jaxt2--hm7z znp5xhZ}G?f2p21|orkgVVqa=Xv88?|#*)Vdp-zdjL9EQpbF5r!5$jzY&axUVY~yk> z)=N{0#)i%z=L#btrNOj2OOLc%g?8+=h5LDUx}llq({~2qhp1sse`N|)6vm2oPj=<< zX=NzgNJK>Qggpkt?Lo%hD%ihzg~uy}F=29ct7w*D3C&oNO{ZmI$zo3;uR8_IY8=k{ zfY(}TH0_QsCQBfn6F$2Uw&fuL^}Eqn_YGZKzS-DgoL_rd2V}pyQkV~y3+p$M4twg- z!j=mBHCoH|cB*CDoF1`p*~eLvQ>pB^-E;V{>OG8GtDEjo#`RiKFRAF!=}HQPL_i&; zc*xO*bHW(okK>c^n~QPXBeRt<8&l}v+HlJHVnn6Ws!_8`mv1}MGlcvG?xz9Sl7j@a z8aT(1x_&lUjyg!YUU7L)!H#6KN_c)fLOlxO*5tvb>NGA~+zGAOl~{Og1`R9=qr8#* zDK|45bIuIn<$J18fi84e(y$wTRFED-Hrov;{`m{w$v7(SKZ=5U`%$X(Gwf4##ite- zl0E24yEX<;OZ05Ia-5634CzL-#bjul^Te*KnRqhxKYL@YO)YfH?KvO(db(c_#q%*S zs}XyTaeBt_VC$hyv9ap({>n-kvCy5SJXnLt>}gzf`3_Ic-=K4PH)7W~z#{$>x~3^n zVulE9(d;WK`|=d(2YaC6;(8RWZN-<3JLy+`E^Ska=W(?Dk*Bj`t;lq=1szKoK`-qT zxjyU0#o!8v%$6s4o0quVbqeAs_tCzh4Cg~$;!em8$~t_7BD>G`^p&I366~K;jS8uU z=$Cs9#tXT1u5K=@@5eK)d0|+P?qis`xC)=8r(nNc4R7ZN_=>;Q4nJqt9ysFqp>*!+ zY1;Ab1YI3{l&%ce#(zuP`-m30gT4L2i>?2Wz-C^GV0%3PuRj7f>SdX*DG$KuwE4We z;^L#Q*+iMr8yjJ{AO~haTrQj6JpP+%?zu4M; zWjDNt`h7B|Slc1wVbPb)vR5!+b|@aEO(8MK1=Ky~FSf3T<6#G^ok_m}{7G@j0!mi& zrB!B2$Wf~WLx1HW!@di~JAPw9pC_1kS-AEy2QT4?h8_hrc#`bV9QrorGF9cLQq+Sq zY&O#(|BtK5e}^qO*K3l^v%^TZ=!l;UEr^YuMnjX$$glJ%B-2!R+<(@@<5zS!CKjE+ zG39%F**_ix&+!@YknNX*;|ZQv&e``V!{6bF^j(w>_zeYZHR=;og``!MZ1$BMtd875 zcHurtRzKqx4LILIo2<2H)vLWU-R%N>jJZIsUPjR>aZwr~qD+gqoEF#iMb!Dqrbmtf ze3ia;c^bm3%P^Cp>-SpD*Vp$NiKuI{GdB-q^USo_Wn-U_e{LyVUu=!0{DY8uQ-L>H z6-e}mf>*8q-cC$I#+L7JzdnTAl10g`@I20aNP^W)8|?oT2(jzY7=O_cyUfC%)Ggc_ z32=X5Mk01V0vdJ;`(@OXJvj3;9~*9G^E5@ZRAcPnvzq(bzoJ-aGIcu!QD@#hDqxby z$$31_N6;=Y+VJ=h+OBHRj#NoHe&8Pcs zI6|JJp!c5ZNEvn>(M7wFmi`nW%SDNePlfZEW0;|oi$6J)oUfyuhoj-V4zYH5NR3Y5 z-vzuc{guW4-_M=-s^5u5@@jFq`e<`c->8aKQ0`6<_Cr!PZA$z>GPUPvu)#IFv{xjJ z9s1;vuijI4%QXd7Z_spBN5z#raeW$lBX1Hv=I$5{L$62kP}dX+-A~&wVedUeo{hw& z+uKP;?hHl8AETFRV(3)gaQb>6jJC^GQ`xJ(0T;E=--@7-|l`Kiy!;@vF zXbSf+IT3brBw3iB8?v;95%V9`!&Rr^_B=(ZwbfG zX86XoFagUXXzO7Mnxbq@^HQhNp^FTCe(@It)`RJHNdm1=+D{{l_t6iHU>eh`MJo-b7@oCu9@(X8ur73!6U+OhajAmLULG$GRyv=t=Y!o z&cM)HzHpGUBJAzqe?o=NjGDy{Q_ ztlByl9dt!YfG&E)RWi@EYhub(MO^f?!k_8DF~gyJ-J7|K8K!OtzjlE{uwZ<;0>6t2cD z3sqsm`%AJ?6XfX>m!GoRNr|U#Lpej)0iv|*6+v}+JW5`bV*So&th_iEx4Ar7L4VBs z{RLmDjp)7PFGv`j!>7#qD0GwMzvVHeK+#52l;02&)KxmwkL1QKp>MX+DQmkWk3;tq z(IU%%kZs7tZn@i-n<(_D3+h*ea`BkMg?^ly#ZfS57utA>k%|{%t|4>DVPqW2g6M!4 zY^aLHP2E_w{aFM1PU5rlsSMOHNv&U$x1j+tWEc$ zJZ`K@duLH~+9~?`BAfPC6f&)i>KGj%hkhn3PG(qQ9P@&CwAUH$)mGx8rhxipvH~&Y2`tX$)bteLm6At#23BI|@h2VEmA+|OY;cv%rC@gu1 z11|d!Et-M5Tn>w%PVMq5n8n$rmQ_EYHryNcYZ9^ico012AAl^EdnfkT65E1y^KcvQ z9l>PlFEHs|LK7ErwrfoiuVg&1vNF_5;22>ELB_;dT`% zMhDY4&hE6aE1>5$i|Iz+mo&sphP5}nPtu_>JiKS6lGrq+tVa$4`CjfTidn5^`Lf4i zEF!9fHub;06FkMgbzw&9JxF#eLvPz>*by^`rav^}aR+g64j;IExxcz_y`3OoMs4>; zk?OV){CD<*2c`@Xj^lqY>%J6|^#U1w4HA`?VC{2mQAN5ZrEWjZRGyGR>mEy<-rr)& z`KOxMZVZb1ir1rM>GDi(S{{`{>a#<5_yU;uYy!@`TZ31bL3lmGfp7Cs*#~yHK`>vF zgsth(FdC^OE2w)zY%R=`!XW=_KQ8$uL1)<*DBKjsx;NZfIA{^V*9pguM3XIkwPc{% z;R1XjvoJMY7~3s?ainf9IdVDblkVT)@o4i>JX12Kni-wA*$W#gCXz;Y-cP zKP{byHQspvY=6p-ceySlg$yFG^L=}C$AI(84_fjRd+NrLpY;w}aO(@-=KRPiGTmp$ zw(5g*i6LHBhvXz<6?aaSwsioPf2TtV-~Qn51TW}rE~Z}Hc5KR$5O$ivBKBIl1gj}J zqzA{k;pebR(u{VESD>-kI}rEQmZ~otp#k5EslQVe4brmU>-r2Yq|cjXvNN>Su#zw4 zv)@M%>o@1WI|~FdvH#&m?Tdx`K0C{EmO1+#GYjGZajf_rybHg>aG(z=xwuY2UHwK6 z%-%T{v-?`XJ*1GSJvk3q7v4bUjREb|;_e%f<6<2gzwmXXK3P(2(Na=87e$*=7E84ih7Xwhqp}AdEk;yX*#;@dgxFxSj^5giz$}+>KIW2}p ztqiB3n?eSzib&(z2xL+A)re*vsK?>&!gVqB(ol5&48@5tIk39B0RbjAm|7EiX5a|! zY~vlse)%Yp_+S2ps+e$n6&>~;JONC>uWrs~X47l~jLW@cYKnpP-9$aa9*h*%3jh z%~g!mFe{u^OvTU9X;AsQ3X_+$Fxp zFp|UHFvZ19jPHjC#`NeNIF7fbRezO9k&BD`M|;+kO~@&!#;U#FkuBMu21yA0=c8vU zvZK7VvxhVnu{IjhS?~RycsYmnOTq#hYwCU3n%eqKAyv$OKnyM(R^hdfz7(F{#gxIyjZ*S^H6MR@eE&rU!njD+OM&~7xn6@6) z-g^Qq2MyZ7t@%r~_aV2e3P#EJ9OI)Om5?{9%#?cdSI)tBTGQ|U7a{qa+ytm%2K|$EveG}Dsgg*On}6XPfUx!4(wb02HH*^(S25yR$@(BvC9>Hh|^-P;j2^9}~jC}+~v?Pf~*y1{r_CNvi6(c9T`sm*5) z)hyBF;f#DDw2RJJm-6pujYgTfIQ6%EjCQ3M910Px;{v@CiFwH^=k8l6pE80D-!!6i z23+naH~)53PauW8x|C+ti?+K9$JW&U#o-CsX{;$k$3dYjD|pvS(k6!?N#r6~OVI)4 zq;+5_4XNvmTMrXZxgrf~t77rGczRD?{nPog*Ccw%_9vDwkA~mGQ)l^rF+xi9GiupdjlY<9hlR+6~jG#a@?vY%_<|igRIIEW7z}@DF~V zL>AWO4S0??%>NpL(0tmi^kC7@DS6+ zp{LrI^u7pI>srvQt3{2ch4z4e->YfUK({h@x)0jM^UuwDmLPTFT|^Gc!h$QpZ-TmG z$JbMB4%j1qEZ7w8e#M0cg|Q1PmoIqh@GSB_=1x!STuJnb6uF&dnd$kLVD$07b<)hJ z0xK%@c^vuSfAKO?kDdo>lk||Y@E+U2_>5$!!aAEidY6$r=U51{5ay;uI!VFg{X|-M zB!(QetL;0; zMdlfmjdG^;F;{SN?i6y;7)3JNou}I_b@TZC>Ce&MoUgfi7kOI>ZQ<712=do|L3xUu z^i}FDO>Qe8X;m$HtL@0Q6W`hZb6;@^bkL*^zX$PUs=+fLwq-o*j~o&IctUuV<-h6& zO8aTk4~aQkj-W73#mt}t(h}eCWP=6GVSR{ zoH4CfJ`V2&TwoLf`->b(74~i5*3~^@WpKOHge+gmBZN1nAX|Um<|M_z=tWD^)*Jv^cXLzWM8XH||?wmmx zb=;jj%Aa}pWPPed?)_j&-oJ(nmxs}Wovy@KlpwHa0994Yr451ZlxaDY{B8@+%q~3+ zqxD5{>_gi=?DP*$C@}H@`8*TOeRAF0J&EFD=-?fqqFMc^tJf~%kKM-fcQ=OJCTply zy5dZgGmkgD-WxCd#?z^#PE;oAKpo!3Jg?@5CW7Jav*>)p&`hgATpp$pMSabIL0J`w zFJzO5W)-c=J5M)!gtT9ACG&W$TdvZBt<_Yl_|qB$I<5IE~$gup65uG@~~7})WK%_|NG=#4PvA?vzCW(YN0FR zyLvu~AL!BH3U{*iwxp%gjOf>CYus@)Wlv5IW=Br+XEkUZ>l$yy4wjrwGH)*<*x!am z?Y1G)4OTR{UZ1o}*W&k)XxOYDL4VGJPPE$6gik_UHs2%kpI+u-g$pdK=-M%1jBO5i zG7~bx@b0z}7wV@%>}n0VpjwJ;>i{hn8ADCkGikU^3QzB%#HomS*q3(d z=+n{}ulX`EOkwi zz~$=+@z!AsN!!~};l%-Dc>M`qF6bK*8EtZuJ3>bBm83Y|f&7E*>Gqlf z@$*COz37cx_&8u?(E!2T9U>@3nXBw9S-}|#i&U*&)3b!amJ)R_wZg)l=JT>knUC$ zI^gn`|E9;~#aL;G5W;%XgjfU85dX{BbH=dIvP8qXS72IxKzdhi@^Dh2pEeOtGu*(~ ztkA@>heu((_$Lg{FCx~rgS#^zkCE>Gn=!aO4ojyml1})!SZ>;p21bHAFGP5B|8OBMzRsqd!I)Qckfl#$__MloJPpN zFpy3+ug|B&L_}15A|YGxpUKu13vdVe>;EwZ4ssfyc2(Y7Ukd=QBoiA+Y*>0iBy~VA|MY_!-_hx2k$KgRiW0<>WKL7jh}$yDJdoxo8|Yu04r zDt7PTNo-fZK(@K&1bs`&hhk+nPJc)vRmm8NdF)LM0n^DWe*{v_E7Q9NCZwA(getcf zP;s&tT@uwpq)ZL^n^=*@T1z@Xrer^_mB;hKtBA59rCEcr&!ir4n)-43dO;s-znhG$ z2hXE^eK9hV-XL>GBVT`qR542YA7f~3BR+lYM=RI=#T>_Y#2oU)mxf?WH9iG@1cLJNB_NdV4ISLPwf_1@qHS!wiV)KVh;X($;X|D6tp)?;r3Nre8*Z- zS~$R(zK46!r#oTvxK!xBJab6sZ_552A+>$o1N8pwj+ZHt1SMp0$99A($wD@XluHfQg7XPuwx-3Vt&4j)Kf$T!po=0i% z2UOdB!G%5T7%nCBd2bwGi@;Y=2x}H4v(f$OPt{0@D2k#-k;&w9=n~Ca_=Q3rbkUFt zIb@i13To*hH0h%#nfUzTIHinc{(LF=)6K|E%>8;x07raUh71xaYq3CHX0 zV*Feppq1N<0S9+usmocu&IKhs+^`eUR>D5Vof1VlAJU4-&MZv3!QB_|APDQf9KcTQ zZa+bL!Ov&m4_GC3qd02>rLA$KZ3U7f7A%B!W{?}QRE?m`<(k~^4#&J1a+oLoj^B-!7g&JI72V&rvg|*J~R( zbEgrF1Al=QDpOyv5Hs%!W^~L-80#|5Y?YAT@Jehf8$!`5h3oh~dZ|&Uci-qTUvKjN zq3XNidi>u1o0LSQQW;5Un9=>bZb?x_LfNuKvMH48N(*fbAq{P0q(P}D8lt3gHZ-*Gt4x*4|%t#XbE% z$XxV48!qJOZdQQd<4W7}yGsPX>_#HWEMuWKb~Y4DbuhKIR(SsYvN=61bE2{7fnguVEA4+*?lBnPM8%DCI(JNI@;1?Muhk+a*K&P{zU1>-F`Xni5db)SdsfgYv|25houXjw(}VJx{0{)8T(gEAc-z$nZMFA`bc5{ExUhzZRhnZ(u^&MR+dH8FVvYI=V$HL+%?_3{6v% zh+5&&PHTBlh{TM&FJN?~44h>ew#RLSgWeDB_I3lj-8LT~pO!*yoF~eg?P0&;Hon}E zqjzRAsmOUPSyxp`FN%7NUlxk^GIkgy?Hz-RLoBDhp z-klQLl;gbzL&8Imv?vj04((+=bfW%eOzCQj9u|%0=F4@!t92 z@q1j|B+3ky7v|!Q!Va8!8-+Q?u3*v9NJ!mEZRf9T?#jRJhC1)1c)ZRQOZQ7-u!1a> zRt(0&F=Mgq+In19nSh;jzc}+w%qE2y9yQkvvAF0@t0VElZpA*b{y4kODMCg18?s8{)n+Cj`Wsi7JT;};+x zTGVaG7vII?fHGKo7yB~wND6GuoWQEWOvv}n!0-2>zHw(MNLtaBEI!Sr%T;6O?@3Kc zJ6FRs-{^=Yr3DE4zW9IIINN#B;efB(Mq|uAJ7f>)jL6gJ*sXaLc?J*hs8D4S zi^NwPjDClrO)4~F&1aM*hT*QScvsoyAV&)izsISc?n3+G2ksfzR;SMHi=5(^`Nz7bS0PBQIRI=Cct*(GKwj;+$^d@ps&RYtwatwopdPBpgb1sZhCi3NOdqg7Vq3sGhc${%pEQ zQ|vyIsp(Z((B*)@%Tx1EY|4v*W+n3}D->k||N6UET^vSRK7oD7WweCEV9V&8?L3X! zv7T0)U0$D4F^5vJMkMS!eQrK^Y{JG1un6HGt&Q2KjyUNuT zi5MhpbxqF|ZPxrh9v0|7HO~{u{NtnTJIA(BNIP#S3d>d9XXvPcy zr8tC0r68>8HA=rvr*Dm6^xNw!Ego%0Menb1`hUIgR!@dL-{NzFIx^ zyfz4vV#d>n0fR`*x<6cMnLY7>FuJXx!RK_)<=w6%(!i@%xGw3Z@a((+RmOCpKfBJr z`jDI=S}qFhcs{BeJnqrJNWpZeKg^|C(EWS8`HTb?-gnJZ{%u($wRYB!t=Xl32K8jF{M}KM z71+_an@a8Ldd7xeE-sDPPK+Chl5$17Pdm#w*TurwOVr1F(!2|cQR4e|$n@Qug4cQU zWIn?6+g`NK!Q0FirMa*ZIlmZ8vgRzFiIo-3l-P62ZzW<{k_64mzKt%|qLH}3SLn-0 zfpWy_onZUUed0ga3%pF#9%rH_Nic6*N$ZmVW6#vPTus*U1I(M%eb-q<-S&vS88@;I;vv&V9u4>dUCT^N zwH_$6S!dK2i4T^tP-Xk+_Q_@K{ot%{p6i_NiHFRV`(2hCg?_Fq3Fof&JH!R&{$7CHB$P)kUlc3^Pc94`2ALeKB|4B9Ue~n?c0m_QE_i*$wsDM9Ffi$nI6KW4?QU( z@jJBaf5O;Yyw|7XZshiF(uBb@Z`9oz34TK~H*Uug?!2BaPL3QxBkuH}eeC|zG|iLS z71Z;#L}9W8zC z!@2IbP0-woUs3E$7TT%3uOqYkcbh5x;GcE_1n4wRPj zofhOS!1;OPBbhKLZ6Fk_pE7f*iM8^A; z)5M8mg!Q7TEXf_zoNHo1Y%LtV20gV0cs^-ldmZkT5=2{>?ZB?@IC*3rvu7#eY~x=E zebH9`i`40K=Fm+HjZnbHSOcNl*0txLe4-GON0_4G_GE6tKz|sSUV^QHE*YLzr~6k# z+uHm++Ccn%R8{^bV;J|^noEBa!}VZtN?Qp@D#??lL5-nMyX1lem)yClHis2CcT-Gt z9ZpC3-^N}2$y^5sDY(w+2G7moaSgGs@}D5^@z;73hF<;v8JP}bHMtRct2B_Ts)+!G zjZm|)n5%A)rHl+0T3M$+-5)YNH}iT?FM5ge^EKEsV+n#^ariR(8h2J%4eP7JpsJk+ zyZJ@T7S0bH_K0iDt;c&j=sbhclEO%~d=t6!*v-{V=`DM)G=|f3b6|KZ7X@DaTnaM@ z&=Bv2D|Y(et=jGOK8~>5Ym(Y08v7GBkXEMyWTdE08+E&5f%i+K`HiJ!m8lfBSf2LH z^TB${TiB&5A(0wtO+Py8@ekiQ@(wyx3@a|aeN5t;)u0%lgN*tZu9@@Vvgb>{?9O4V zUm`(<>%Sn>Ee~gXL~JCpsuXU2xf(5-WJ+J=Wa8zLQ`}%FNpAL&X`G>IAa+HlQb*sT zus<5XJslH-5L*ShKgy1N+%}?#-!`J==S1#zl@q><>qM{a%%u1mYII2>0Z)^2@Mc^$ z$~h#?Vakzr;LvZYinkIN$T*aBL&Olph$M9&=+$xMZMzT-%rt5i^UcRok32@ zOSrI%`P}ljO$;BS56vAuhVE`CLH?6D0-e7nbMa{FXgX3dny%?K<9J6cj{ndLSKf|- z%Ci12xXkR_cLxgehyKvwemtw+GMwq}QxB=nU zTcIK!iDr{R8XBU@Klz~`$e7nGTOiP>XmP_SMITyxDx4hCdr|7MrTAril{-1&4KD2J zN5e<_#dojE7#R?aiknj0;w*LekKcz5ZbwnX^o!cXasr*&U88AY*UOaEr;0L~tf@7V z;a8VL!PL4V-FUu$ZcbfDcD4P4b^3N!G7iT~rgsLu^h0S9onZ3mw)NNcSzum?Rrccj z=3m+kqZri9xQ!!*kD>ee4kFLGAU(*5;kGYEOhObk+Z@3%Np>$klMHi{eTZQiUU|8R zq*`%=ex8h{!R3o+%UBiibKNKKdS5peW>b#~zuW4VoNjiTY7aQk#aKQ$ijXp-C%whvwDm6{yXH|u}PLt)k(yxU212}f( zIQC3U6Y6OA(NFkRRGW;zj$xS1Y;#&N=VGc%Ewa{1(Q|1jDr)_KDLWns`3|3}0B>YK zlarK5si_0gfBkCb^+{tMb`QRT)VV!KS#1H0Vs=$!GsDR3$~TJC@4^p|eoG^w>S;&c z5L%h&hqUR}+RN^kBlfG;+)ljaPYZs2&2TzEZUG2Cey#MagMX|1|*jO5Kq9r4Wh*=OC%M6C=HM2<29LuE7TuT~j~a9Mu)w+i4~GviD?T zS9-YaH4atfW8#^c81uCRgO(gNsr_`2)9ZbX>p1n9NuzqTuC(gEFBU=>>&h26*sC3AZK;Xs_o*hmLfvZy~0R3q#bz>8M@$66%T* z$@2X^%CK>#LVZIDiK=PO|JAx1^t3;8`Pb%=TQZS4x(*(Hj$+Zb063oTfoxAv4)JYe z6~6uKO2#oyupytxKkv>&2fx{9Vlno5G2f%Kqfepwxg4(!ZfvJt`JVX~?^2+s?FFzO zyH!X}R_uiChMMhp{?+|FCkq3VzA}uJKTs^J!4$2vLY^n%i&2mp4~fEfcwxDa_CQ`j z;x)s^TA|#Bithh+FZf4Gc9J(Y`;tAkYIA@v#`}7$<_tE+b1M!>U`WF!u2p?6Csl0N zKK^fKf8i7}POIPC^qA}AvX9%kqZHCF3lX&84Jx;YGQ2k0hvw-ZvLORAR1)FoZ#t+g z?fA&?P+b0u?q~MlcL#i=4~zgYwh%P_VQ?=d?PR znEsbw`1c!*S23Jn>0{UtHWjV4yD&QTBHY|hK&@~)`WcCPWLrJ6HY~;D)j80Aco~bE zN>KlYttD?II^*U@fiJY^c*S#Q6}6&-6J`2sWvZr}K(5X{^iwDn+G^Fr&WP`majAzZ zA{Y7Kv4R6K*Kk-7KUizYk47juAHy6bch%;Dk!LyuG9TUfHCYAxi2FIbfleB~uis%_ zOLsEQhZ7c$Xh9cSd5YVg4}I4FVO&DCDPhr$5aiy@z;4eoLR!59yB}^JiBD0!@P2s@ zuL?R)XW1vhv)K%HM=PNSRbTF+yIv=1zHtvjrsg2!ZU_PgK0xt^oA?-_K&?G&m>>3k z`+=uK4U!if6KJ%xGsozX@cTsAN38rL+8|7NF&+{M@h}Wxu@7RDs7A|*y3UzM&&U6F zkMGjdjZE)J)4;m__Tp6^M49S}PU8EZjegrF-FGWqb@PKmoOlMmxV1kf-}dAdGYo^@ z_We;L(H-`jp1?=K@pbUt9|#BSz3^2l!uDu(SC;5c7QJ)0F#B%MxnYLtXWbCwEaIM) z$Od8elN1y^FG6;wKpb?7#vB%l?rzj6B(~~O%mHtbFb|pC;0{GNoAVCaZk$Gq<7D_~RC9T91A#6Fu?Evn z*G(NdSxV^TXijH$ZKundtmv;oe^OHGLrqU!TVc;)%UjT)Kfs(TT;g-5sdg zT6uDn5`7JV4krox_Po-O7KWJ8Di-H$xkYN{TK`$XUSHFwPl^2p(B$*|>A9PDe|&f* z6|PAg>Bx&NG-~1*o;E$;PXiKYKSw!b^ z9@0C%G^&0SMO(-0722;9r~$L!_2@fZ1JA9QZP<}sLLGmey5d242$mlRgt~ojd%YzC z)?v6sE}qBVg4uXx3;&hbc`;pfk;@g7-I1n-LxGxG#<^OPLUlcytoVxnGM~wqb4N(jG31)o13os zg(z0&j;k)7C_A2qb*8x(UmXd}(x|_#{**) zk2IQ;D<6^y$z=i87HZ6 z<^>i5Y#tezPNB)g4```FC2f)`BKI>l$=5H0k^*({#IgzMy^QH@=UcD{P+~R{M^Uw# zP%^Lt`bL?|$EF9^>^ zLN7X%&f@A^^XAplxARlfgZOYYZ+`I(5nrWk-Zd3-0&gWXEVj-FQGPparUd+bU2*?s z2_hXDu*%F8mEqyI<YDfZmQ@MGyXV zAuT(3{2u6wrxuMEHdB%I4{C<;y%q2`ABKj*r6?+Vi94U~qGsiFG^x!N>X&)Zmt3AO z|Kimiba(QI?rpsEk$a3dTQeHF#+hP^{ON7rCbHZrLywpqu50!vvU3|t+tXGs|J?cH zu}h7N%|%R^&V6j?<*E>Ryd{}N?9(K@eqBg@{9gKfk@+XLuAr=`B=n!!>cKd1R)X9& z=};=ucg6qt4Q^is>c&KKdy0Ew5Yu&KZ%x2-<}*KA%MCw*%8)ea9_! zbZS5(0)`Yo@`nPMT+tz$M@HoBE=B79hy8S7N>^HCKZrhU)T4DaqO5U6#03nR_zMPC zKjQJzyYO5SBlJO7Rt*=d;((nh{pnz460Kx0ax|@DX=3m)ZiLGk%s!C`YiB0gPK#rH zSyt`imd^Z<+WPC?r|+sp?vuL_qPI=N>&si=ksOIW#%qxfAVG~E%os++P|~t87y96R z%t7QVk&syCQ;wz;|6v}q(OPq40>5~BG=H^z2mc^zJ8!+pimytL;6Eq7$J20j*Bd>J zR;F&CghO5etqHQauwwg2C+8EGdr~}W{Nt(ZJ7lIU)m#zn3Fd~SqWPpe{6B`n`^X1| zVb_s%N=~L++s8s%7OzjFp)an|y35z;RX`c7^<{D_M#JJkBNzEnly$4iF?@`z2XLH% zQ2+H6ZYF!;#K~l#e1U~O1{q#Lo{TD0G%`Qt9~z|jxfyF_WMHWMTE2M6eg1dCAAU_x z_u=kedJOM7Erz#~6Lm?G*#7fm^iWD$)<>w*t=m|PvQ9-|ek-6Q`d?WEY5>G zg`eDxvcG;TT^x<0}9E-EB> zrWgl)Jiv`N`>}nJKi1SI;_&ry#6{FWIqDTI9P@@5LXlMyjPlJMo53%g#V_M|Im%k?1byBhS-GXY0sYPsQiD-hLeLfOOZ z>ER^i2Q_Of-L_4{<~(LAIbDxF)b^z23tF^n&VT34d*Kn7W+J|S(`r+&b$tQ)z5}Tp z4W=aH9C}z-NkQsfRCGWdZ?7H1sH~UxYFrQ1dhsryJ82`=5~qSM14klOX>)r&|C^%+ zlsc2)l7X})ohaM50xxZkleE-pT7L5rJ+7AF9h$CCr}N^y=Tu}h-Tz|DzxcR;|0dwRWyRe@(%AuEQdxFLye`w1~&y-B^sV@t2^ok=+T{98tP^ z8HUp)($iVH>G`pE($f+B7&h#(5uQirXwjAg8w$PJk?!@jqcx_<^sOw2e(gL;&Z;Nc z=gtTpDay~)pgsHb$ZU)<+2kKV#;|bmn{b6PVy@EP?U%@-!vR{v;>kRnBF|)EM=;yr zJdQP1VV3;^q%0Tbe(l*JY+h_mN^+a%{`XY66B{Pf(NRs0t{o!U``ntk4REC1!&Rt6 zeiz=GeuSF-Ys_Ol#d}RO=r85M_yFPDxfrZd`;E$A5tk=-N~F-9?ib3D(w*rZ^L}Ia z*$1e0^h8+ta`bwZhI6~}ar8{Eurml61R5E!Bv$mLcUO?30QgK zCFUIJPSsiI=w@F98|K^hr=O_%Sv$rD1||wZ-nMZ)_iQ?H4v03Rkv(QXf5Xu~ zH%qANj%+y$ycNheKLKApiZTLwpG{D{y$Q_<;(Gb@)q~kt79!S<`IX)7Nl|t8P``gE zG$TaYomDn}@%OxmE0{_aqG;$NY%bm| z)ZKa5G)h`OgOX2`(Ty4vUVpe0uXsqAZ+bh3k9U^fZ!JBJ301WiH{=Yo0;Q;#%_F-w zhS%9$hkyBJAb)I1Z{bYE0OR?~OHBC<(|qvYYXufEKhj%2#C^#r^eP5Ut%D zNts_O7yh zJm2=64%JtoxDuWNjb*|3y>p51?8mDWNb0VK%fEKD=PhNi#26S8lplKwX=i@F#KW8p z9g3Ju+MH8rWXNPySI#$})odh%zLufEf6n3|vw4tazC=gVn!5O&FMb?|osSQ&*s?wty6L)5k4>OTx;WRc>?nLhcEGEIw>kJ3_9X@oohey_Cw4P;gf;W}n zNmV*JWr^}Ai<1mb`=>tcNX>`zyp5PONKfD^uR|r0EYD-$lq>|-e8l%L-;ki=jFnYU zh?>r@@3F~G^CiA!devI{=#q2!_xkc3Z}$)r1iYpKqUe64IwC@gsNl;Eh$4_>|Ix z{H$3k`S%5xlo;C#A133`nRXZM$v5%ndMDbM^bO{JH1KAZ74)6%a*<bPn@XZsBhl_p$%>07_uM^tq%{vq}3H;&4Y2U%z=tFtogT7CJMjwCvpFA{rE;|T z>wo8{z$g|QJxZB^H}K@7G?adhoJ}vh3`udPh?iz69f6fQ&qG=z2G*}nV$aQJ{5&q& zBW`luh@&o-aKv>Oizy*N8a;yHb@wUgp+0#$8?IRYNuGSVccc?vSxkiJB&@4Cim$Z{ zU*d`q$q!?{SszDtdxVkczp(ULE0(NGI8z;(V+Mb`jDS-Od`BUfhYE0GaHu|cXl=gjkGMh)a=k2B;gR`=ySr;Wn46;s>$;B7*f z@H?s}F<-j@xVzjI@7|ur-|-iO^w!}CcqrM8ZhbYUtzkO!s+{|5$173&ow1UcjgHkbh?ex z>9^4MJsbWvMSc3&4N5}(&s~_$@>p*SWpZ839z!s7d^Y+f%8>c023$5*p{o5`sqFhM zs$u){J!UsJpd||ZcD3Natij~kwHlIF4?wp0EW(YaV`aWOj8@g+;s^1*9=1CegSka` zRk$7jddbYzD+1r>h(>J7DxvNn-K;RRi1{OW1PcF0xi{mNCc|#r)Rh+Mc4a#3*ErX3 zkqVBdk#xj;8r!uO|9Hb2>JrPaT|4_je|r_Cuc^Ytw{_Tbs#&0MFo^k_+Z3SNaHeDL z`U)zmuVe3xzQVID`4_R@^*oEOEGaSaVj~>8uy{P{h`L^|B{Rb*RJ>&~eP_N=ZDT*r zBOT6nQ_%l=CeE}J;`ymKfkuPrS7cZ9qN52$w0reHx-V-)ul5#0{XsZt<`?5r;CaY= zK8N6@47fiX2Jmjm{S2Zxlu7!luH}-|A>;CZ*gl-1+2IA!^0h> zB$>B`nz{whsI*A34?0M7at>rLr!TE5?L>`Iuc3czAg$mPXie)PEMoDVm8Ydp$+mn7 zue?l_`PXUf9~(+zK6EYzeQ0W<7u~q%L@B+t|ZNAzKbvqU7Dl3 z0e$2;)AX@TIN-tjG7Edrp(S&vR+#%4qSWD@-TX)}L)3#h+PFDV1k-k|q@QY9D=V z^dO@U3Eq72NZx6~0v;W0_zR9x`A*0G(zDltsp8mfY91d%=@a~^=G;26^M8SvA$t6~ z-Sc^caZC9lW=r_9{b%z2M^3lbkzMVFPf6l?zpc)qJ-eu`;ye|{7gKD*DJt9HN*C%I zFg0}yuQkP*m%2Tg$MGfn-zBT~T)$X>PU)RNR3i>vKvd~6F!7(QEVp|J5CQZ~!d?a{ALotKCGAvV~yw*yUUsTStU%;gMsKTMe> z^c;XqFW+Ne|1NZY%P6{I(3`xDiI_cmDoUX+`Umq3QzpY@o$1EOazwoNPha+Kmo|-w z5YOgq?OMNATEb_vct2?SKUSt(s4HVc2xN*%ky!T(iHfpR?<~G^BbH90kd|3ALU}gz zo;QO|&vc;^8$TkU!GpFh+fUA$eQ40qU2J}A7WyE7#jEa;(}A-8$dS5zXR5Vq!LDD& zaAwFKY`ewQW=Ot?I$WYi-78CP>5tVlvB+$6Ase+r)&uYN7E_Il6 zU4kNp-9)X(0J-^S}iKFsom~G}MoQc}Y@DKE%xAi1KB)P|S<*#~h454ZTknY%kI@$(Hs&BelO~2N_TW2o9tO$NXc3`gOcx-AI2py|njADK$ z6C!*uVzD2ZcTB?QS-G6!4k`3|{)yY^s(=??hcTbW$DG9P7G&-Jgn)c~>J-P~)jm55 z3DxPsduA2$i&^*hCG_Q9!<}yu_^oZWy^O$q!#|&fl7{jvQWn$ARoF zv~GtsNhV0s&*gHoB>Mom#Y+g~PRE(T<>#RGHoI#*#Ox=`UoumN>QnlV<4WVY8oxmcX9ZDqD_AKPiS{kP2A zg`3Q$Lu;aV)?RzOmf3F{1h)KvNg|8;cd7!ZW%tm0OFWxbjfmz)@43fMuX(}Wc7M(X zUN7Z0^&7;iE*nSrjpIpjwLTdX%_1$wS>#s4WO1$2pnLK>CY{d4&<@o&;$9D}99K*_ z`U)p@jiJ_YmSpWPnr5_)qhH3oNrSB;sb_X%;>Xb_qoL&5JdfVTi8yR+{7$>0PcvU? z(u`powQSdKpDUZKlTdLW66FUXP(R2CPWi8}PyH7nJvvk5rtj$bGau^;6>)D|E-Yu3 zVNXkUDzt=9{@Am($eU9NZDkc&QqhN^`Nv*kxB|1MKe5SFiT;-AlDFq? z{B+90=L^GxemqB_tfY-M-N{4$r*GTR{(a^ftjCa?D8zj%LjOb+DqXyc#;AMJC08eO z?tU0?hqDl~z6=9Qiv=2Ti_T)7sx%#*_zA`>H&{&WlS2B#BTHC3#zv$tKO(0qhP3^; zGF=INhO`)YT3u*OvVBb`_SZn_^hTTBT|6t4S#{woO0S-081;F=|M$hda4{cE=8u@a zXG#$1`UjD|TM#{~m_P%*hU2^cS?qXIfn#x+Kw-|-JRD@RKoU(3AjIh2KHo*N8)V_f!^bvr#k4(4@c|LI3#vHjC+HF;W}sx za$TjUmn2Uch8oZUh7ECT(okA%(SbY`U;^OddN~kzIeB2dmehcg0}0)^QoWY`TqT+_$fZOX>ZXSYWlMh zw=>+ab!9YOPY=T+*Dy>OpaHd=Ptau5j|TRdMs>aCkW}jtirVB)Ssx1EZ_$emsCOaz zoUUX&)q*0zMLF>f!^IfUeH*4()Ovd*HhWQPU^L>XK3lfl1{DJn^f1)uLrTI!bOVL^r(EC%TiG@`)~ht3b>#|Iab_F}E=LJGDy{Q?aeM$I zhwGtJ{6J*?7VU6ew0O~TuR~<^*OwX%?8w4`BXzeT%$qx!{J8BU-f!}dg{;#lFgIV!l+`ev?WygWWKY>(b^CNue!sH54P;*aufIe1pr0!N=} zx)J!A;bKIS%HN)payJ!ywz~;5o^6Q0{PFq-xnczF_8j!SGz!um?0N2@8}D@H7jc0) zG@1EV{u`^V@~4GzZU19ROF5%zFI1bwQk{Pyc@3FPC#7~H*2fsy#=9{tteE_!$B<`) zGBqA%@q}c3sJx7w!R+Pf{MupU$BiV*gG!|MklC{YdE)e+VV!@j{=?UnHR*hewe zSzCQ5ox#1#Ud$PFF688*#kgO;qeek(fUQLBH8X=A$FB|ymM z+cA$jG$U51WB+U|3{tqtmFbq5c;##s;vfbcpF#Scrjc*l7p(X{zaX#M!Q8VI9m%q^ z7!UZpShB(&mgmC|+^P#dPo{IolcHmDd?7P)F5+}|U}9=0oR{rn*kUC#!!Mgg7S5*d z=RfiLn=iI8{ixsia`fsP1f9bFbZJWioe^&ojg4d6u~w1AVw5>#vgqVS!B%X2d_LYx z-4Ej?Non{hwr-MFVEXGYwA9$5%D99(HBF4)>Ucr)(_8mUjf;Ab zgd^u`u}3BV3nb)G+Eop9lNX>zjR_2EExFJ|(L$g6xne2Mt?9Xg;fOt=Mms5f?!59WS6@5}H#=TWsR4c$L3L?*-V{71Ww%?XMg;zAu3GTETv zG&-F7f@0%aXusWT((4gO@|NT18MO%cjk-^SjfXo{M)|<#{e0xu4HVM9i`gyjg{leiOrG(Ii)h{{jF)Z6cdozdW$v%%TW;;a*+SiBjqikSTk8_8 z!l0Y*dp+~I_T2RuGR2UpxBgSTa5QBJh*p?Ie#pP0cN2ovMQePRAs zFDMG5UA92xH;>>8Ib0W`vz()$HVyu;ktB2uw6F64_W3wn37UJ=i`IMYrOa85WF_5^ zZtturB#2bekSQHEu07YibHKMn3<-CBKry!}~k& zf?;VTFBD}KZGBx;Jp$b&`!XNcZpbmwL*@Mr^gFvhUHGX=Ura=~%i>g^CaNCOmjs}A zUO&*^kEoLY)!llfYxGLSPaX9*J!4Hr35bKyFBv0?dQ%HcM#@u z{lKkUOYA^&{Nn`WA*Z>Jol%^klroFG(j+|pDk)Fq0To#5;VSUGX0jK?PYlF{DdPQ8 zMj{+erZ3=gCJK@EdMILXluTz$K-8jn7*ux`oTe`%p6;eohFO$(E{h)hm_chCSS+4i zXMkvBbc#{JkMsym!77P!l03#$x~WPWRhL1Fth`XazRS~||K|B)J!Mkgv652uSkd*! z2$b~_aoGxZUs}Fz4n4^B#^8#pT+O6IP+e$_z_~9dbF(48jx*%T2fd-==(lGwJ#Dz^&;&JuCy*BnL2N< zqlJ6wpt8*in)OX6dn7@&Nhffn^JCS2d|IwrLFNy~P`Qezo7htxh^KG;kg-w;eX3UD z*m_@KEkrV3BCmNT(UD`w!OBs%9J~ZtR*vZPU>Ty!dO^8x4!i=Hue_r<7B&nL`rk@z zDAGO-r$2Aj(A|NHsN==H)J>JeKyqOFAdBgvQK&8Ct6E-dQrst%TYG*zR@VE#WXKTc zI;3;Ylr6b?U#)Ry++kd0K2OMv_xs%^;Bii_Q7b0gr6URd5KV5!( zXIE?)`3?6Ls?fc{{pc6?12XqUQ16M>!rFH1^T-@eZwh`y^+q z`wNSe`jfuj9gOKH`W_zo*a2s~b8xElA*T99V&bJe@Qm**^hLm+4E%gjEBrncu=juZ z=lO^2wE9sF<(X7cO0S`GY;`Cn`+6KzHD}Q7ezC-DzKb$`CAYvY5s9(R^fmD?EwJ{X zHF1Wt!S_<|Ebc z8j?*uH|+QUm;Lxj1$u(cZ$8^^K5BL)$$Bj^pVyE2f6GScR&#+@^O~-F!M!c~navY< ziwUopy^ueP`Ry%?k6hyu9BMS8PW#PhnnzbEkz%;snkQ+D&A0Zwxvf3*%*JA>4bPWb zrt;UNVwlOT==a;2c2UUptNJm9g!<6yz%YVK5b5ZO@3#vF_M&m24mI>Kp{7N7SQ1jr z&AB1kO@}Ha;bpl6!p9vK=2dNQEhZ0`KtTtmQP&&?&?_OYO}+2e&e3XJaK znO$#6nmUQzJ?=v~@#}^3FSFFi|I!Nb%}S)B_KWC&lekA)m7IpU!Yq3DBA9kXFQn}< zH-!AnMRM#eFY0hSr(EPdIgP-pzB@3?$P>Bwok*F*tlCtPiO5+B;4@xv*DOLHdHXNc zhHH@FC{d@eX=xC5VTgDhajd$I9EndDIhOhIcOTcjH*AW!jJs|*xK-2>U(0JbpXxx= zIQON)>#V53YY4qM?j-PY(NTqZ8uK)v+Yz*E(7=}IGf^<)CVp&Qi-3Shq#f{1Ncy-Envjs{4@$iVgj?E4Fr&0P-FT&e+q5W-irWRGfAf_J$+Aujx zo+uBuky?e?C7vuk$1x~HiS`Wt>Mdcq(iwHTNPbEzDXop6FwYT`^2iG5Uqhk&_yUyA zOVSFTo&Uq@hp!f}N)mb5bzH=NR9i3?lPC9~jSoHP&0}wR<#-KC4u9eHggDS{op^d@ z$S_pR=0nD2F{ag?!&RTj(Pqp&n`qS;` z&<@E{f>AOk2mO_I;k<4g_amq~j5f;)e9ax&m!xJ!k>B}r3S=>y+S=ySIsvbi-^PZ? z9q25B|Ego}v;!YW{;WeoCNGDZn>MVP zdqBCR9B1rKV}H~=Y%ckM-hua^(!33^r^msrIgXtr!ckB!uJ^Wjm77;1%C8yT{A60< z=}D!>i7GNL;@9*Cc(YxDOo~U-iY9l=?75DccW@}af33mu$I>LzO^?OaY6gF5oWR%N z&mlO&^e|ESok(rUXRO}44SsKAurzffj?X`W+vktr(BG@=>vUL{H&)AuGOj*7nJ$;B z#~g-bp_V;`-|X(idq|6RX6!Ichi$s@nssYv>)SdC=#fkcE8`fhDU-^uxQUV)R^(?C zD)gb^#|#YL#C$Lsexk5B9gns@!0{hl$??{86sn1Lt)IOgk@saw{;kCoN(j!wvWAn` ztyoSwemY>6{dwdV-eg!UY4|)%#5T_A>Wdqx?5trwip-Zr;bfW_XS{p4Ov$lxcu2Bj zJ#zp5%(XsV4RFpsB&5}!jK(_cCx|O8LKl;xSg&`C`RK5?9sE%=HI&0-@jfA+!kK<_ z*x8LdqiuzDd@0T63@Xhjqi8WDbAMrXwI4<$%L;v$7JrsbQ3f3vzk#klbECsPfwWCC zobpyrrjUNov~lPh8q#qU^h`w^MH9oK;%{`Kv@Tt#GQ>)#?`>@iC%>`@U3VMNrd_IZ zp~L~-HXGq7!@cqLyM|YGJiYj+M*|f4(9?cW45Rc1w|S;Lxj2Y(wS!SB_sw(-j_lK* zH+Pww?XN%WO|chvUz9ESsorc2!GsyoG%lLiL$Z0+_KAFU9|i;O)0eT+`QIa4`AtVU z^KKd%RH-Mf@uSl$Xm7!I>aO+z{rppL({d3GTrlFT-+A&De~0j^wF9ZNUk$ok-beMX zwxh7_Jz8y;y~6Whln|YU2IKzHdd0_SaBc%tj-NwM-W(A6%yE7&vJ_nS1F=c``fh9Z z1w+f}yN?^y8Trz=2OscWbvoIe-AkQjo}}Ua%UGN=@qTdnO$J;Q-@w)JGYkVod0<;R z9#mE0z*SMko~HClsH69Td+;)tK-0bZ)8o=PLYspV&$sIV%+_gCiCr2Tps^?RKOKi6EL-lIO-;b^2yo+JCx zwe6xzd=2vpS`k`-yIJGtp#E&qIM|V-W8QH?0;0IZzp}7ooFOUqupp0pH-&cTm1c8~ zbVgF@9}h})`3=dWjzYTQ6fJOChq2OJmb!fzPuV@idvjZ#oz`0ihp<*m>7zj{E_pE8 zHXdi)o^ikP7vgGPUo1EtiW(BFl~-zDC?J~6+pb=_W|_N+6V>X%LBy^EQBwHpo7bHj>n z;$G}uI98zhZ;l$xyT_U8t->VsUh|lB5=)l$K-rvFlkKX>T<;ORP;+zyf?|(z7q$|Y zF+arQRLv0jt2|$r`!AE<3Vpoji)e4s%ef;36m}&AwM{TI?j+Q+{moW>R9OsPKcFWc z^eut;WF*n?aTjU9fzISVTZfLRis!{Pen#~EPfqgo>m;H4qm)G1S|b&%+Kpj8)HxC5 zSkuf!hV5$(*}xnZ3W zP?aP>XILz(H3Pb$HYL;K^iCD>&zVD;dpt+yi^p-gcZHC5AB%yJP-{&6%1dDMf9p1- zJe6B@+l#;Jc8%}Q=+93;FaBhHBQ0()=F_c=$Y;5g2i9EM&)tdq*2@+SArjZnp(AtEvk3 ziN6b*5=*+Q+mk9f{y;&%1H9YgjER|12paepKE|@7WUfe78Dbx)t{Fr{XTQxM^A4VJg%%o592blcHg0xJgVNU7KHstbr3yXi z`vX4ZJ!$2X+qn1c1-7zxYa8DmUA7C~<@1wq;`U-;KD4FZ**=ZM31og0?5@!JEQ>4t zXc)YH)p4mG*5D$;hBq&5#lob`FdR?-#ngOuCw9R02?g-gn1%t9MEznL-D%3ayY2N^ z(+vy{{z3^^F02#tHi_p4#|~q-s-l}%Yjj6Q*PXTs z#%lBMbY2d|=vU!!s|j`K;7>dC{oDJ&d+;pyhQ5cYvj*|Gqv-R-YeLzwc+tn>|L55v z{1}#O?nFCGorLnkjYS*lxdp$_wUWhqh%uyCV^i8xVM^=1jUn@8f7{zK-(nmLW~e|p zJ&&8An#hHEJK*5U-W0iScsnm`b;%d}!O;RsTEH+y&fd^yFW$jdXrM((N+n4WWv1*@W@IHZ z^LstddF%E2+-|==p0{(J=Un6QxUTD5<1y|JIeOmkj@bZdP|E-da$BrO6}?UP^T#ZB zlLb!nVd6rJ{47Pk_Q})Y5-oD&c7Hm_Iuk3p z2BZE$Ij%VVXR{*7qmtoE{4FZ6@_mrN z!{m}sdTkX?FQ;vw&|YQ~#IQk(&RElI*|~J7Qdwv>eqV3lDwR}%ylH>2Q27*G!}lSj z+8!6fK5#XaX0)5djy(QGflj_xCim4M=Dgpq=@ctlM+aTCdABZd{G+`ELOJ!8y~s;D zn`Wl(q=OywsnS7;rv4ZT_wd7HP*+2Vor9>R#Ef3%|3R@CvlFSzpkAJ@$$0uLikK2h zJxa^X)27s6;RS36jYjD(9lEGBmi{brqKK#NbgsEO6?K&3{i_`LZr}X*bJnByl5~AO zzDkVebal>BDp+Mj*}t_&f%)m^tLC6<_mzDAx##%Oq!awYVLSLOFMQeASY0^#Z|NgT z-*?+m@>v%eTs58^|N0H9_y#&GrO2Xz!b8)akxDRMdu{-ILFF@!&hw8J5EKjsv2P zcw<*XDk?7@6UIibAC@pXS;JYq)}pIhc97a5F&=5$Yy#)L*I)vQNT9y8MpeS=nK#pTle@nEWccc^JiJ zwue(e8w}Rty~% z!=0|X19gi;%9_!US539x6H5B=nI$&`x|diLAZTAG>VAs08B=s7GMpJxwA4o7S?AM2 zn=Gi_h0|AcvFH0W?v}khQq$e=bfXW9(xZUp-msNjLNgz*v;N#6)c>0zDXR~c?Yf{N zX8*o~vpGE}f9Wrr|G5$d=hDc`^GElQ1kLr)Oq!o>xvHjau zT#zI4FT=@p*hETM%=}9_O3~Y0*Kl<0dU`1Rj8<%u=4ZCgrPNKJGp9OJ+1=F?6&XPn zRs3j8{#-g1qfPzoMHxrnCKiKse+22uMAD!}N1^Q&`$eO*do{XlyU*gzW#VnGsRF%6 zMn#b8gGkC;Y(Y;isL<8ENy78<_eN5#j31qQvYJZHn$se;G~s#0!|ynmaRosaL>bLB z8F|Qjm_-f;Ce!rlktF+W6y1IO7pa@&h4kBA-olO*MR-tm6lX8z)J~)pzlT*)t}N7H#+Ke~qIavPvg4{@!thcaFxjDm9?7r!odx83{3CT3uE?((+>W1G@PPttkJ8|} zOxS+@2AA9pG{WN!`jiI?dAi>)!1i$g*jpHlR|(>Nt3I}syWWMxC;s9~n!7Bh-du^Q zM?A;*>{c|j9A&n>&!}rPv$O6NN%`;nDN?>1?RIPM2cK=^qbtJsNrqc^+07RG$!}jN zD39T&7UW@Y$s#;n#KB{h8KTm&xyrP?c;r_J`yM&Whu{!yT^IG&A2xo&*aaKu>&Ls) z)HRl*rRLG?lCETEt4-@CEF!nFA=F;s0G-_9Nn08_Q2V5fbUCw#+;yX9!xdk;dt(>9 zSFFNG9Zd=|T}ycj-6*B7yD*+6z?kjonNM@c6q3_o`;(+ql-FZ0rBzg;i)8?Mo;VNb z%u`q}^)z0rn~me=d*Cs%i{XvT$Sl~8R%9}tlI#?rtcP_nq^G%s=8W*C*Dlkj&T0zv znezk_SZvA!x879aFq7#~`v_@kn>WBZKM}D_Run$02dM-Zkl}RZZ~Zhw=!2_C+GKSs zK#1=u=|ksb>23XKys+NzU*Afu^}+M_qVn;)^*x61f4hQ?o0ZX^N58P_)j; zD9~&-J7_;4mo6uWRBZPSot^Y|ZT+kO+_zN6u0T#b3=-Dt{HD|-7(pR{(? zAWta}TfUv7O_v{1yz~FHyYft)f%Z)|vBUo|QkT{tlEuZ88yEs_=KDUmNRg}`t58rK z+k2JEQqL+DKX9%mk_JcO{?%NjC%X<)%EwBxfo**0%llwfK{00EbAYR#HVnEiL;Y7X zsq=Fl!mR!xbba!$>q0z~_efL8ou#x%&yC4s{AlzWKPqb+OScr=gtDH7$VzciePh6IsNIc+yKN~3-}Y#ue|FSeOm448*;l10G5rUk z7y7}s@d`Y)d_#W|CECjWK+$)G`T9p(hYp-SBjmki_@}LmI!}AVw$H(OhM}8kwT>Jf z&!_q8Qki`C56K^H&ua&E=PeF&=SLj8L+14|G?pgQ3l?+MbHXrsd{dh&dtS!SBj2c< zQxAUHzA3!TJtzKI=L!7dGarO;w)Y0;-{V`%?m z5&vZLCv8M@a>0G;;x?YnZrg%qwij`swh+BE8RmFU4EztxfLw$s`rciR9_rFk39eb# zub9T>6B{Alq(Wm#zS56od}MmYvuXnEz1aHQD_xd)%+#S3v)YrF#Y5C|h=T3MZwS$v zM!`~B>G6`aGI1lh7N=;fS*72U;e zuJUFTUImN3G1C`}p*@zi^jWqE%X)EiZeYktCo(`_b0=9kTImj=qh!ItN-GU*Xv$jcYKRbj?Ii?2pR1J z#l9uN|Ed1>a8P|4HV@Ooi?Ucu*w>8)8Z0I0l0D>IkxLns2PvV0EMlKqLW;%nTN=e| zC+4j}&ICOn|D&mzIB0E*H|nO)H&~91dxm4X;x$fA`@XQx__|sOP76#he%fDd@Z5Dc z>9tkJm&Rgb`KBHh?i0!uL!ZTKYfY(P<+LN&qFYFI55w=z56Q1tn{O&M=k2yy^3FAS z{Mepn$a4K<#1zRUOJvwQqD_Jvj_PYjnag*%HIv9j_! zlwzwGhM*KZXPCXC{uk46gQ+b#{7pyZfN~g?X1A3wd5twKo$o_g8)HZ{X*2bFVJwV; z^zdvf+#$*eC3T(sDC#C9_clvJxs2(RBgkzx8+$t*#fIln^!f5h?EIdKky(ZCCTl#& z+rqbuFXO)yyye$-Y2aUXy2n2;4d=bbr=sg6(QdSTlMeIw7j4g@U(OKdiu-jO7Sg52 zeRL6W>qCTRgL}x6Nq{Pu>|pcutEqJJr=dV+xavqe@?r7`KU4bR$oyaKJ5fvK9Rz>q zN0pUB$h_2(ygF_n&rv$PCA5F67>|UANYuFg$B*Q>;pyl#P*`2 zkXtE7n~UWsy4jTeG&Ewf%`Paj=lzFW#`?4G*?y!BkCOL7RVf5!JFmi~^*6(NdI^v5 zchG#>v#qUX{V0YBJ3suZPjY?fI7|4Swgr|)na>YfeQhExg~zkP*YQwZD-9zSW0LgRM%D+y;3=rDKd-6?;H~Px3ALMA_vy12+u$AGw zRiYO^oTAkiP>Q=KKWyCbpYFSBCmZz10uUj_J=q$uRw!$vT__URxpq{0QJ*kv?PTb0 zzKI`l$8kj|6mvXRBB{#}?&Ow4tX(~T!Y?v8)v6{T@7}k2@P6A#xIL4nJMC^`Wa@Jk zMn&A~OL&?SHIx#2vl!f%M-Ae!?ew}y z;pSzu`F#*^bEB!$eFJH$4I@Xj!_+G^hDJ4Qqpk9y9w5BpJ2&2hgHEq8D7qoaJS25$ z&$ww)%={w~TI&!!@g71h`ry!+ms;%%+|ZfDeo1&2$@)dKjoI{O8*B_Z)0OMKRQX^k zZL0afd~KdFTZBuPSP_7eTJ4F{HJM&gji%e*!0*AkSxm`r-0s|wI(IRmBiDwKk$eOx zZC_3wCmbZ1u5x^@Q9YSox)%TD={NfF#+L@I_cPNd(x>}pb!l(N6S$;`bMm;AQhdVv z5I)B_omcf-$8Y{2FPw=7CUjtWyIfq`PISU@Bb}`sO7iS%KRIj_ZImsg87(;!GuDkf z$C^@M`60L|d?3%^?)>R1+xXUbL-+_61AbX~3B5WaE!FD#jm5Q@Mqh>{li8F)lCB&; z&C+thJltW(U`VfIGWwC~u-)^xZ5|mnSzLE?e+S}Xuqx8;XyJT?D{d|k_t<{=y@ftY z=ov3u7aUa-#<1#jY4Xqsz?1Ha(doqjJRhBdc60vkJp6lFAfC^)m$E6Fk9n$x@G)DI zqdqFS%^A%q;HuO`JxZaxqKu?3?{w)z8(7Sfo(6r}zLSmbmZPd7hsF5IMeB_sq;)!u zLAPhY&Ybzmve+3}oAqgg!E>lDP88^yeee|G)XnIZeQ(;$FeN1A_FXYuxE_%^jS~Tv zu;S+h;lE;fJd69o@C}*FxpaY%Q0KVjtHQg97guA;?HBkft4s5D=+L$MmxbrCKW%ZO zq8-JRxza1R(wB$t@G*a$@UGQQnR*^Ok0Eb@u}pb2YId3l&o}lAL!D0yh8??z;@81g zG`gM(pXZ1JiPw;@M#O+g`{jr283XXGr#U=6FG2izV?452j4mv%ga^qlN+lOk4o39d z+C`vsw!ImBF=*mqj*NmVZ-<5VcBAK@3Ha1Elbe4wlsmVzg|o4^!}Y%0T|VH!9j>5H z7`L;*k4s+ngj@Vp8FzOM!kW@4m>(X)4UUzkPm6=8;*BTWT6zUgA0H4lh$;e#Xp_egJ1DTj$hbEm3P(7r_2X3ls|ha$uaDUnVs)p zo@Htq{gS-B>%s7n9&s+)?{ZE)#ath~El>)tf+8>KIVO)~aT*o1V6b-zT5q1l#*Sq; zmv{o-Plut`V?E?9lBT0`#*tphRA$>ZhB_?JA={HX@US|Zo9vdy%?Z24rH6_(sw@9Q zaap&!qa|evj<4Pg@5apnEhR_d7!Fq;q<7rMuMXa5r?LXA6Sj+Vu3o?w48O&P4{YMC zuV?csudEfn)14jK`1muWozNc=eyW3}W6kL^xS(c0-{Lqjx-05>M@$ZY&4{;%ahy&A zy9Sc}uTAv!o5-UipRy;zgZbQtGJ6aa`cmzTJKi4>z&OyM_5v?Ex?e~6N2Uv$m)pjtgw8=Xj^YeEpG)o~ zzjBot4qPqwrzwj|^tr@seWHV~<{oHY(HUj!Cu04zX;5M^EJ@n>j^pTv%50dYM`O$| zvD{DHhp;%}J*AFYGFwi6SNaw^jLL>8(G#}zBm|lu)XxQb=eELT6pPKCxsE$IA{HZEIs{FK2S? z1Jbmur30DRGJnEvCor=QVL)pt7FJcFXvb%SysE+1uq8s>{Y}ED*PhjMAzq)na!TR1 zE``&5qfR-@&beFV5wc&cOs=DM3u(Px^`-%uBdG2TlQAji(zHcyaX@J`dS$2xS`jm4H2*bN}+TVYQ?@ay)g`5`vg8*R-s?j#&l}#2OL}C3BA7= z+_hNWHr_wlnBbiLinjlfwn)@6#`bQjaB}E+oZW0Fyt}(^IaE#_!?Ijvn{}}qFNQ6K zXXkM!VOUx2&Uq-jC+_DQyEbq^cJau_zln?uA8>p@0cH+$#In2Mxw!PiOYHu94+$qv z!Sa%L2H>y0M4!*jbeY*-?S6WgmK;u?!c~)*pUe&H_wGZ-7L2EnmEFlTzfPbf)N==2 zs!AkN%`8%Ud64Wo?xl9Tw)Em?8`?50p3YAyqT$OgFrUx;^l<)L*h~3f#gIH0*EHbg zCPq`^AGnQNk0ExEnBW|TDH2{XN-_O$sZjpH7dPO3hT&O`H=t|xtms>vCAD_Fh=vUe z|2e%Msc#-m9nXId(zb+7Ly>(p7KcTkVoNcWdC1WnO(`kQ%!%;5)E|DfIVgUThbG<7u8-S+)KJkzW0}uWvgmn(YW8fW zpR10rxT(s#d@nOzwZ??+pULs1#+~{4;mPD}%IsmcuO&>JOdpb$kaF5VdbgXWfj9PZ zFMQ>sB8=j3&#DsZ2Z-{+kjsnV5EqL~=6BcSViZF6E*0A1`tO;-^}^0w_{{L6cMnOz z{vs#z4m78!ChN)I({b89J&vAa`ZF7B6*^$7L*Em((}%ag6xV(uomwUOSiNU9Z|UkT zWSo^p@5hzV-}RgdF<&L?Djx1``|BZle0>|md56Y8za&qw^}J23yA ziI#PS_|We=tcr%yx)eLoIJudQxXvW!MV)C@>0sIu5=rkz9;B5w#Xk8mL!W%B^0AlW zrS^^Z3o90*sQnIIQt0^$*Z1AS*l#>(tl3FR$`i>(V+S4G%hBRH^0el8H!5f}r_(p} z*!iy!!*BM$q8eR+9yRlc=vZ+O<}9}5kV;X%q+xdny3+Nyto|2nswcoXe&*KZm!ib$ z4HU23LhYdc>6^~XLD}F$bW3l-gV)Rk`sx{6TIYvq*?vOVUHv05uvaP06spkp3{lqj zZ@f5#6~e_Q9Gm?V+Do3Tvp)&FPNtMC8$s%Su2ONW6u+>gfLf!JQ03``Ih%ICU&9y8 z8(d)Wsvkzb&gY~X8r$X;zMYg*CutrdlufwLZ$$sWt?Op7k(NBaei3WOUq! z+*WB*K+2LX3cvnxA7(h?USTNC#~gyX%z3yeO+b~u8SSa@r)7r8G@aRu|7$PF)gZK! zP~J2q>(f~$?zav145i?}nRIss+xy=t#q%^#<|0Wi6LJB26|cdJ$tM4mvwn6Y|87AG z@1&+VK(4RqfW9a1@EaUQ3wta{n)M31(D;gB2F>~jqq~>!BS{m>g66?8Jp)A#7^Zml zJSg}1Z!hLiX+b}p@ojWAzrF&0r{B;$_g$!y++*qc2yt!L(Jz|%%{fPdVsFuHhh~z> zTS@i_*R^N6h~MPSqN(V~9T4(qseXrY-_CUJmN^BMo6(fUbJ#rcHn(_1C?0E_#Fp!9 z-aY>c*H)@fEW?dj^0yh&BD9%IU4>TED^km*W}%#A-wxt(>_u4R$Py_wqQj1dc$qBj z^Gu>YBZ}cAtv?QW9?kmayLe9d*H&4swTM+0&yRhNWwO(t_bf&BnBhFQbD+Pb+VH+Q4By-P$R}G zcBH6nvQoO6J5xQwOttx~OICT`VXxeQzIVTjC6S_B^YX0+Y(JEZ*`8N$-ys|!FXm6Npg46U-fjuQY+{07Im|;8_nDoQl zgZ_BW@DqML6wh{ZEP9|wB?qc({^{cAO1&BG&@O{YvQ*wrP3O%qa;3RY=BV@pM68Sw z?sqbMMCAEB*sBqTTc?w-{pC*Vi?u?;hg~>O`W0{LP3TD=i+B{p^lT%i;=R^vJehhH zrgb@(z%a2)bpwTRKAJ}2^rJdVQ$L4s=gy1I`mDsbYWy4cHoU3`qb$AjA zi=?Xv*#8>Jj&(Sf!RTGUbKUd(Cd6w^quXS z8-}nw?WnFL2fEwca6zRP)8{90wj+$lWS2i}9}`JW7sb+P)(2%RGSql~BL*i{VVd*> z9IgHDys*Dv5S7FS5pU*0%FlXIyA%^5n+PoU@&PV~Txg}IE8QB-Y-(8Du8)mjtmX+j zgG$7dzk%=f2XM&RjmwXuq{hZ3;z*C9NP1X?0plNH(Zao$;hn?<9dU={gWLFIcn42k z)#Ju4hO6+*1|tX8LcK+Ujz*}{vk3*v?p%!j*Xi#C-01xmM|)b+?H{7oOrSxpiIR6I z(1E~JusL;(hK<+a!#a-SZw%1oEsFDK0mI*y)Zu1lhgU(&U#C06lg+t_PqY-h<(WUk z=5Wd=*xNQYOVY0Ws?0xsGnU`2XTkeV?8I+-m_wuUa#H2Z73X^3b1S6 zMU2^-g^?b<_z-YZ7$;qH*OI&HMmj#&p8EJ}P{HFB{Yo|` z!PTjg<7Q*p-1QeGM8pd9To`PJ+{O?VV>X)UgHE8TAq}$@4#&vm1dJV%2}`MZ^x|a6 z^eWrmG5T~EMos9lZWJ-Zj$X5P7X_t{K|MI6%#fo^c9!HfZvbhpv!O;`j(V~Eb4}hI z7`t_*>mF*zx{yIEWO(pOG=dYBVPowxQW}SHf1z9c$c9ntO?{H+hLJPf6|#Q zwnfvUjohkX(olb3hzO=Dz1z(lqqRCBZRRk}(pR*7$+V5cp9TM&>o-o?gmo)V!}im4 zY+PR^(5|+M%}e`|;koKDQkfj){PuDtH#v##F9r$kOYCA$QJ;n_C*C7xWrgr;cIpne zJT8V)(i-@Ai}RNEhSl7!w0aC}-a_qurP4u-0t&c#p2hcc6UN5k9S8B4`JHsF_P~t9 z`FO6Zi=Tt*xdXR4L7({(&N<%^!@FAG$-INy(a96It@)2Q2a@6#q4CWveBCy^lkmoA9QkD>c_Y5@;|wJQQ6zMRT|N`C!KUENpsu4f$(N3Gezg zD#_n|% zzsSn8t$*wsPGHz=u?=&zBDg2nlaQEQhAww@LTzd~U>R z^f%!ztvW^7J#lDizb@q_EP`2)JZ>YN`!guHc z6!8hz(`y+HzZHGDek)pHNw2L~R~ClG&L&8ZAWZFilCpf*A2~bW{p<6SUU#M?29D8TNe7q6L)B2Mo z!+X4NCy9oZou`n68T4pTHMZ#{kk740gnQp8l8#Z}^S6jE59Mt99LXj15=}~YN4ArT zN$x`yRc`7^jyHGEbm>eASaXFkQboPznS@*%ajHgzi3O$f>OrqJt5Z>Dcf?H-F=(;} z6TO|QO)5A0l3xEyczb^WTDwf4=eLT(IJYP#Ag8I-%&uTo19YW44m<{ttp85t? z3H-kCKFf{C;A#D*bre{$kdAB`P6_OrKBFs+#qBbsSn^7aUdfMQabJ%J{b?jMhW|1; ziT59RmEW=P4DZn}mbZTGOCMY{=ydgD3XGji4`nUezPsc@efjFXGx@EHXYzDfgHKmG zM5f1>EfMRlviF~`chqZC`*$L&lc!-?o`_GqNI`2q(UiBc{8Nu*l=4kHn>#vh`L(S(c+ZWsq*Ong!h4JBv!uLUt9|HiN(hx6GNk^~kKySbj$y4wQjFsvQt7ab zS_b9`eCzL0BVA2d@^LE0bSY6@)3RtL&2x;W2dXI)*C(|S1De3%PUcV0p;WtFsK^%W2AbBk(7a7E_|Xf-@|!vrQcu5CST5_% zb@5zG@t!3#Hnbh@5b%}$_$;OY)5ZBf>3u#>a%!frZ3RATq=%YroJJs7%6^o zQXL(sTuog!Eyt}vV<<3rGqv^&qMqbS*Snd~HJ4tHwH`vNPKVOv$;sq=B#%@tjwN$V zHQY9_qMW|5#Od6j&K<5&rj6()Drv7mDLEChvgwqgd0(j0X1oksk2c}vBMsU$z7M^( z743D^?!V*4x=h88#GQy7ACIicG)N}|AZyP}PLqp*kBkgW9-%~Ues`hj{-S(9Qoh#Q zauh5$i~RoaSUG7ioHl20e^xv~khL8}Vq7yH! zNxnn6bopC(lN`qN;PVs5@O?SH?M$q0u^L@ZgrHw>AU@jmgwgat8s8c2*WJCP*gD0K zuG}6?>o@6>W1*A4pFy85T>9Qla9yu~d9Utr1q)Ymz4M<7?{*B5p%16THA+Iy#LL5I z)X{@<)8;6>xD_CbSJUGr+{e)edA|7t|L0FG-&=1uf9`84y^k@b1;Nr}X<|nEq|B)J zB#T$hS7C+CU#_E}HM~pn^=5rN0?#ieFyB(t;jP*}mY(<6NnxSh)ICx3JC@YD%zPW_ zuib}vL>=07`hkv7qF(x@Su%OL+@pW}{>9Q%_C4%T zYWw!CFkE^`nOojU(>bHpaAxPcjd3d}p)s49d;_UdzG(L=Nz<|-jE`G+gZGQ8Tg|#*OMw(&0rWw?FWN@7zc%4 zR^($imE?omNk(Qd-HZ@rHR|odsQ1)YwC`&te)#MzyrWex6<*1s&!ao?j$3OfJ!1uZ zZ8D(m+8WGGFb#a#P>MP?lpfBpqa+1+ifNLCX=^l8?U~(VQazkpPr&4nj6jR=nK&GD zcP6!QvDChE33bpcqq6shd243-pG0gZl zxZC14v}^H6p{(m#v%s}a#9{v|EcoV$=ew2)bgtK~!%Cg*q|dNuah2JoCyTO@2$?r# zLtKn$x&9&==o3P>+&ro3l?BkUzmNJ z{rntc3=!pNBQ6*)yx=8tBEg3SI~}09FAJ!(G!2EjV%z#m^~Y!7e)cs#;aXG2=o|Cj z+AL`Y=No-UD=CNW&XeO~dMWVMT~5>NL-&OAYaOd7-8-K)hKaTXeS^m!AoV+K>k!Uw ziY?~nKDfsZf1ky>c2f|>k?y=F(3W}y3xfl2xiJ8nZeJ1B>yJ)}m~-U}{KErrcYwH0 zu=jk0w^PDt4u6H-JZmIf)r)k0lQJFJnuoaw>ZIvGv~08wO>13FgKd~x_QoK}SM;Gz zy6foCuy8z-n$X7Qo`}QfK3Lp4OL!W6r9G!|&x;n;MpM*;P)b|iLq9{AJLBC3G_HU8;Z@1zF0Bq5Gd@JyHVu_RRj1Qx~6dq{OaX zlwFv=aKcy0xXT zURHz8f4T_m^Ww1~Y8h7V-R(6L*zY6-uW+EBuf;KVvScH@%^OBdW3*|?nVkqwTi;b7 zd8qZ)$Rl**u`Ivw*e5!q=}mbSGX6KyR;`2E9z8T3?p#n{tVkeTz*bX%4x)McOF}vYO z?S$u3Q~#SwC445W8pma-%3<11XVRIiMo$(tV1CV8#4gc*?(1ky#y<&`1GVX^O>gpQ zWxkMwqFw8Kw#GGoSitX{t-w2$ZJ@FX`b0nC@jALA#q74BA>krFcfLH}1~n9L7Y@(n z>sw~>)o(0#^RRZj?beyX{8RZZ0E@lX4|63l7?oUbT#@x2~ zT9RgMfgg8%t|eK%vLcHYK{!<($u(civewr(6wW(wOJ(RvH?h3AM~elUEX6?^C^EYT zee0Qq83n_z-r@kaHgOG?x3xd@IXsGT2C37I)rSRI8@%JW-G^q-zJYPnRBlg&Pdd@M z=`&Dt-+@aSx|6C7htj#_Yn~xt#!Tij!gQB8A|{?~ z*bL!&*|f76S(bUQJiQKi?}tGn;?;f>uxn zo;yj?$UUzZ9^Pb}ZRf;!{Hhe%xy)-2I=-^U0c&wxk<@);fHc)4Dbkc`ISMot`!}Tn$ff*#hg01OCc^% zin}IX3H=U+WU_Z2{m2K4c~FQ6KF>KH2Q``#=t^m$yOUG&QS`dj349nNjRUw@!&SI$DoNJM`+b2*)e|TcYk`ZFf?sX{r`_v|wSQdz zG@AzD{5%UReZHIFt#_yB#zzRf5rKl&Ehz7uia~S5^&;KtmoN_fogX0HvAnHLN#AJZ zC{u{{ChGbjh}N6UCpVA2^w!rO%id^G|12BkU#&<-rv#w$hJl!9yC3dTvLKzvY#^7F zW9>*iOkFg^NK%el)gjvI_mOh%$?&z~-chIZhsi^o+4g_W!r6{_=yLWf8e7)3)qTKb z2tGHSLgbig_`ZDC_B>$mFnX7}m(DE|?KLE2OMa6(`%qYlI9Eyjx33aqPA=#4`JK~z z`BcjwzICr5A0)Gv_US%DhRRx$cu$A6cMoVMo#2f61aiN&%oKd(Ry@{YXVX4(xK>=d zB;`dvpH2o76R03EjKDAj_8cc%m>b4jEBT9v^%Lo1V<@>eI8n(V7CZdc6Ru$M3_7s= z7#tbib`-;N8Bm*Hy?u)^H_nXV`fLfXEY;RRpO7+&F z^V>yT(X2?3cb~BTbWQV^{YJ6hK>A|JlYh=Fp>15akMLNt4HHs&P#yETbI?`0kCM66ZWk8YQN#$i(#;8uZf!q z+APLO1f{*yAB@!K;R@%?&Y+U$vRCsWBHde-0s%y&>sox zzYq-h_u0s-%o6f`D3QjlO%pITdJX3HjDhawVBz_;E6NB?bi)GgEc|M2M~iMgMdhFU znAdEA2_buNepCXMbiaf^y%Nkaj~4Q*F^<`6AebeJNRiZX=<)Pd$HYE$ba0}7wa(QR)X+P~u~H>yYn zKTJEq>ikBm4c{)*lV-aT;}e_P??1l?ea6*7K2p^2==@cZlezd9@0r|-u8MNG+m)j3 zs%wxOwGW*@k;^=2O#dO&YoHvheHww6UQw8#P>CVO)M*dHib@%hE7TG8d<>4SWO0pl zh%$-G_P)YcUGw6Aa1AnhM*F9>QozhGW_xl0D$+W%cw--`yVslkv=~v)RPjvrz48v7 zxURrou<66c-YnOwi>1b8Md3QGIIMgI!IO_vOORXK(`KZkafLD7Xhm=$VGW{uq_ zp_^EbUT`?-ua|k$Ct0^ z`gew2#h1D+^wB|=s5%@|`?n)ok8P%(D|ov5&4G3&4H5XQHQ0{qrT}C{gbMM$GRyzlqN-PoOQHWcT@C%Z~*Td+lRVcK>%6DZrW3yMdKhg6g_jKq6 zs0`gF{2$(980>UxgnJ2(+s~>tWHh(*0g`HKP@&LCGb4W{uF1|tc~-kn-KnUCL$+`Ye4!^6%1Oh ziiwF1$XmA-U8l(kYeQnbh!-j0;omQQ;YtkmIS#jc89I{EpC-GB@}fm%x;X#+G?!w% z6!xh_4BuxKDxaitHa~As(G(5dKBFIB;xvH2aaotocDhZkwVdFXkb#%xrTD9(Kx4{O z=-tty*v1=k17((R!B!`@AWg-#HtZuW>IUX^4!{pylwtg%Dd&S3rkf7I&AmNLC z4m_^ImK!SM-i;%NG27|Q6&EsD)}B^X48pQhBTV|$4VUIx3FUJ}^1^ldCSUXm^2ScT zAQV5{jXkyN5io8MGRB0UWzK21hQ=a}`S42Woit0GpPlW)JFPj)H_ScB54lvtE2IYU znU2?m_c=zVkg@3x*80B2r{_zcbMXHkGs#ab53zl(vTrOI4GeGF_LnUg2bi^elin3c z=(MgJi(C6rjg=;yc0Go6$;u>kYy$NtRHE!RYJ>@5dRO|%&CTMWuBXD=25 z7YdOczYc0~AGpa6C*k@wU({_A^@_f7(#Gm@9q7A$E;stq8?M9e{!A~s9S7B%aLz#o z1wjLZa(X{}gYkR++o!0Shau^~0E}~;FT9u3IeG;fLu!lh>eE$(?EH+!YnAEwA}RX7 zWb?WWx+Uj}}wL{=rlJ~61LcC3i#kJfec+@fns&gN-r5k6)6C&d1$Mtf${_YlC zsJ=)Gv|OmDYCp4)DWLr~GbuUmEvYXt=HJ!-YU}q;F?+ccYxi?wvvWDw*YmihwN;$_ z5p5RdXc_`P9cwH9LC0LkKg<{Y&)!spPUfI#RxjG#FL>z&>vnA1E;$AH9>I8<=mWb; zui<~L6~p#PlNz%ZkzJkBmj2Pq>)1VLB+YP~MoPP%q0k_l?%(@D4!W&0WyUv3Z@f+y zI<2DuIp)9Os{+^J{)qiD6@~pKVsXKLXYIaU{80F%3A5v!$=hw7z?T=pM7t*Khn@Q_ z3-w)%NNlTD(vDkJvN`wZdJNS(}?v~PsaRH>aj>QJua2giO zM*dTkDkU!XPad1QaTZ=XoJZV5O@>dPK!qx$7_ePDmrH2e7I6ZL-j`zRz|+uXaZ)O? z>yfQ&Oj0on)5>}sHGTZgW=N8L2#aI8@6cOVviMP^SuV^sqXV8W97D|mQ^_bz^zSG? zl`ZgftojOyj_pUvEj6U|-+-(GGqGw}C2mw+z-X_ls0i*vnzyaU`F1;Mco_r3nu|!9 z5rwO>ehc|6zNTZZtr`j!UdF}dJ=FU0FkPN%OwF3AQ{qY3a|IH0IZP$mvhlKRK`uE$t)XjlO6Y)Ak)m+VJ0R)a`E=l&C;T&vfZY zZWw(`6Xi;Ad+S(i#jgDI>+<|ooT7mYOF?JL12k|esb{e-)ykUF+C(kt^jbBE%O^LkizSD^I0SA;e-FEgRnokMBT6+1G2Ekh~~ zyAxS-q<*J9P{FGy{0==UenVyhRrHhNSBx~^clKAHakT@4^46VMM)hW!siSedP}V;_ z4)+^{?a5jw?_h#eGwtv;`W?p9bRnm%=a4ox3Zt2wP2jDC%uYj*D)nW^ZloKeZ`-r8 zK^`1TwP0v@4#m?<=u+NYWYxrBbVvkZZcl=H%5H?11;U6A$Hb+6*nf2k4%vNV_P(2G z@Qr!oe$jyH@AM{3bbOLY{=c*C^`AaF0S%sVd+FJOV^}39VH@xVlc7+IhC4-u@0k+N+~qvsj={ zVLo&>v%zhEds#QZzWrYxO0IiWPZ0PqW$_>j>~)3pVEKxX80VplU*Sdw87V82B}p53 zNg6v?{FoI>%n*A{d~T;#!oAu2fz$rg$T?rS%yDC)xY5aHP_RmahU$uQmy%u%#%`9S zi%(*tAIW^f-65BeaP&BOmYl@)#9YK5)8N19Eazum(dHK(Z6Rjr#lOt%#ZS%JAk^KU z9f568g-E_v4-dbSEFQ&Mu42q?6#qZA-a9U*_x}U0v?C%)3Z=3_={#Qt35kXk3K^l2 zj8KYDRuTx>xtFMn7z;qHtL7joAm}64Wcb>5sS0V*BwEn zYltR(ISTpht+=${svsx7@HJPvVK<~6W+3H;I$m0H`ws;(S`i-rv&=7hY3Vpv}9*dQ_ zqwovMZQkLXL4}a_&m1?nD0HJ^1xso8YkBHtl!&6-0vytL3i)r9$g3A^2$HsJtaXKR zc?Oo+u7c~bjmo3X-$u-g4T2nhgGMCmUPoD*_tNi+EdJ7+ab5rBap!Cl^Fw8NwQ6r< zE8auin{0g8F3%14{DG_Jw}-`;i)XF0qW|JVC3$6-zQJze!wA-4Top|dso>x_HpgC~ z4~#3~`pig@q6yTaE|}5eo~O4bp3#$lk^EQJ-n{LK0vf8#Va}HosMUJR?LF_$+1%7c zMGq~|=Q(ic!D7o|0-$_d8~5v`NNL{A#2>vrLLH(r*5j6%_-;6_UkZy|;$G16&<9ML zY((dlv?uq%X!Q23g2@4*?SB9H^GM1(=t&ydHg=-Qcy&thTPMh@h}EW?yo(g?sLvnv z>CbOpt-=IfMIF#WZccM`YWA5Q5aew~RRE=6Pq7?7yWpYQ7 zfGN($Fyhq-gnaz(eUp@Z_NSM)9j`?SQa^?EQ|SEzo18~ccmhu=CQgE3>Nz|rcq9C0 zJJo=cMk-L@i(<4C*PxetcY4kf6>aN8fkDzVeW|CBm+3CDSg?;8LTllnq)Eqj_|h4# z?>OAyInZ|j){T>=w00h3#P9%%XAhvxIiO_u5W1Ush4iB3c&?&>9y+_xrU6NW@hwmq zGLlmI9i&^M_S5s+rT_Z)TJSt}Hq@te&*aI5#hJD2y@-3UmgvN=6r^n|h@aGr;mw@H z@-w$!*zgDyyfG~v$9PVzx8di8-n1pC0X?_6_5kRHGjr4%G4ZG)li{$lv!L#ZNlfjvu4cjrR##!f(Ii#&5Am!ql@raC)IO z`RJTO@B(|NxZJ?brF$t!@EbV;+;A+(cHZjywA}@ z-adCX-;?Q1`@LB~v)Wh;;p=5W8C9kd}&YF@j)W`ZY zmJ~ifqx)OonaF_#WV>2{9v#uAol74hV!1OS_6?(jcFAPv6+wxe0^z~%X(Z2X923K> z*!2NH77FAfC+d$S_az?_1@XMIuR$D}l6$J(zT(DUSJLvgA(dPmI`U&CRmWHhb@jbq zPhG|jqUYIVIN&sz+}?<1sev-elqzFLI`&`Dqs*MTpU|S2&R>Q6?Omr*fD}*j2lSy- z`=wf*SH{!UpG#NAJY}pY!m4a2i_{ZNfH&&Gxsx zlJBAhhAC*V8|$Bi!%_YLwlja^1yzcSzx@r~7QJNp?3MJ>Xa{w9%=V6&R^*o9PW9(| zVPeY`bZ#@p`3Gk3Nw5@Tp6~ORi#WaK%NZu?|7oj4A6hPg`$vqr}}?;9WK!=kDC+mexeVwt#U_dB~CR zM{C|B$BO@DD8-kI4(1D@PV(~{!ueK5P3pS57k!#pgSstR7Mq$U@{BfVWBsP{&$ZTvCrcF!OToEL`L7mqQ9 z;cFc-|Ah0FMs)o|JIen02@j`tpiB1x5gl5C^{d~*W9DmA8zl2pD=zY$KcacXUd8;A zgTMH%?OyTUUe{qb!^E+v5yz*0r7~rP_M}y&t1&7q18pG}v2E%}40b7ktXc_1?mUQ= zyaBZ5;v$OkH6jP+Q?TDIo&oMoH=;wsRms}_ITrO##}lNSuI zqQ)^FF%xq1kA%azdk8V#0@;(P&{2}6Ma9~*u*+1s@v*lcf7*vaWII+OYoa>!I{g3( zSo@5aXo2r+ZTFqs9o*wJ<7ei56n~3?vtcLd zIa!%Bocqys7x65$tp6P3C@w;ut$naaEfSMAustZ9#5mno7!g>4U)QeSXxe>DzTTP4 z`~MN@W;<4$8iN**dQlWcUK;W59f?@0O7j>_@Vr-B$lbp1?0p08-x|=0U7JX%K91TtI+FH= zL{d*aL9{c1YIYD_|7SPT z$iv7xSDDTA2}~GH$A0x|nB8d%tzK2??ZlF<&O%6jEy1OE25Su=#fv zm3Nv#57+uYDr6ttc2|_zz}ASgx;%uxEkm>RHiW(HCXAOJoj+m8l8F@MK8IeOWn)E~ zXTH?C;ren0mi-t_LyacWgA>d@V2LzcVX=skx?VWaj2D?wG(LAPhM%fJpR{Y((q$j6 zO}Z)kpT@@U-*&lrS@hMCTn{#Rh_No6>369qan`Fy`|E1@vO|xa#+Fj?7Ju44A(C1` zrt{a#7V+K%WBGQ9m;TAwmsN?i46E`p!{T#aUX3xS-;pwzot;ma(84_fY3q7%9Z)`1 z2urU>#Bbb=hKvGyc#sYqhPh~0%XlxEj$&+|AnNT`K(BiyQAEc~xXXV)O}q{{#i~)4 zGX{xRb?O^@|Mf2rp?QJO#gc*RL0V&(O`(03kZl>mF4{Vq zl142g^I%7^FY8V>537(__A-p>aTp7o4j_H+GTaRp{X1G64`ay6RM;PS&Uj)!3Ux5p z<_F8SQq*(XAhN@BVN49I>qo0fvoYCJj=C|%XO50)cK-@>or^VBvhj{tc% zF-%kz8{jCbCOo%y)FkL!NX4vgCve`W1GPWDgZ9`iWtiXPjKfZcpIJ4Mj~qUpZ$8(F z$BzOEcVm7mKDIO`e+H!lS&;mzKGgfiJIt*7MKh-v@F{18@$QDB`0e3+_>EJ(33+r` z*OOw+4M}0^Fq*ngwE1XWo<~woQ)%MyW3&q!Xi7sJxaGED z#`%XhnaTE!78kL^Wf{hLw87D52;I#YO^Zf%rXP&Kv5;YC{C4>Y-{s<- zOn&ud;hA9l?a=C#3or9B9GsF5^-F8PAG(dmautdirbU(O{u9`HiOfXeOJp594}zmT;1tw0%i@QiVucn48SUWOn? za{v1n5kpM!J7L^^J_nNPWj!|_;`t#gzfcB~FX0%w@+{ZSY!de%Z!>4zua;|D+04oO z7JXL(^V`#aeG{o$Un?r?X+TA5pMy6@#afd(n1{|b^tT*H$$dm04TYc{465H4Nk1dpY1Mi+YByJj zZYxXEB(qB>Jl7v;p*!)aOCiQqTt$Apyzsn4M(=lR(3yT2(c7=1CFL6Q?OCHPXv3=eb;4MKaRtoTnP<}`qAV{*Qg73zh@$H_wo8pH z+s&tUMs}3uy`3HvA0!tKQ}X1jX-~~6+7UdD&gdkMi003$M6Hkn+MiH`WG(e?F5g}euzHKgBm{V62XgU?OqcPkKD$=La z?LW3~F_{(IxYTNVywH&pS8GxCvTH_v`$?7Iv}<%rLo3t8%&XRLB#<|4yRQl#dct?+b`qLWQccrYjl7NG%9pL-e!ALB83Rs`gKKE&6iC?Sv9b5=tB zW;a~^y)VBbr@`;}E_5tBf(+|$Tx7nEA9f_%mCrPQ?yHfBK2# z!Ncfl*XdMJyNZU48pAm2FJR8@Y}_9B5}q02S&t8DPqDv_!Xf?-7BY^AK84+B%H=DF zi@GM%Y2nsiC|#>Xd*`&LVi{>Nxm_dNKWVs+>dt&7#nJ})rBzN1jtAIy9xb#}Rx0B} znHYi3MJ__xWk+<-Uw0=~e}9RiuUgm{2{gWE9DT?R64Gy0{exa#G-y-zPGk|1gJ`b9>GwjRVukwNRJNv3cX~*iCm7ZQ_#a2hrmAJQ$Zv^P0Yp z^^bexqq3h?Z=Ucky(HZL2VJP{6ZHgPoyECIa)1BK{djld*uOf=E*OZf35-wUYzVe7 zjDk7R;yb$CF$nrAj^bo%D&w5aKu)w5{MDsVH77uL_QZ*cIKcLY5}5;a#-S>x89&Rk z=*#JWv}f=f>g&Pq+qVv)0eStX{Gbh8HQPpJv8hz%b4V!XRWD zc?j{bD%9#EO^2fflGmGAG=E4CS+8R85I#M~c-lmqaScX9hbZ)A_#ZpkE70vgW`FnD z4a+m}$ST~9+g)1WUfrImqFrHm<2deG$VV3Qw{!@RNf5nV8 z%bQWi7O|aM>fR$|>Mi*Hc#A%ved(^YCM}l^#h+{-*PHpJ4lKix8w03fN4!w}-F{a| z|64OT3}iex{)72R26lYYh0nr&s}4$01jDCs;y+_yii%3!Th?- zQyC7H72nPNw2*$m*cj|+If$ZfmxcCRvoI8E*9@jp7*0uI_7lR+nG*X>Q4u)X9zNnnFpcQ zh-@tRKOg#Ys^hUNV*#|}7vi_wRv32i!=fb$*w!lA>{G4;U{>8*XuMOP@mHomm&M*h z)f5Y|Zcfvooa`aYC-4!x=Vzg@FMH>1s?%KRLJCzMaWK>uRyQ?h@Zzo9`w%(V<{aH%W zUb2|j3dSEX_YwBi9Rq))1>dwfP?0WAT~hzoHWMF{e9_u?~#=-dXaStZCg zergMQ2N-X2=#elRS87979#*TE;QUb?)NVUIdR1}DH}s_s@1M%+IYzP=y=eMGkLj65`v0_F^H>}F{4z$6 z)f{Aq3S%pbcxi+hrw}x*K8$JB;yO~;BZeype!|5jyyHff1klmX$Ta` zGkef^m@KP+V)ze1hD2`d$fkdOEu;Jyx8jJY2={#X@17)$3gZafd0B>jy${ENW?9e! zeMEON#DJ}nvCr5`kTrXN1s%zpO5UT_Q`PlY%BhVZ(_QsAklz)J53JBoKLfAWnw!SB zVyi|=NnJPgMNn7C87Y?Pd)Welu>Un~xAG^B7d7LH8AJ$jNn+rO~%kFxktM z_i0V%lcrqd&-8l24@juv+s_y$*a($}X2}gdWk%IgLaF!eSfQ@L?`#m%$qVn2m>sk9 z2%2+>alK|IqJD(q@$&C@bMZIgGG*y>q7j{$C;GG%F?&OxGs7-QVS99o0LoswglfiW z&}-9qG|4xe*7kW$9oMDOs6S%-`=}Glx2#$dPfU1p9XSc=gEWM;nB?*vRm)^)a&$Lx zNRXqJS`!kQwb!H{v^*A%rD1IsFO#?YD{?oA&+Sa5vXar0M6zBY?q6+}#<`B>$gJm88F)xgCeqQO28Vn1}OB zkJF%|BghG5T!L{dCeHDf*j|#fR#}CpV!RA~7oNa*=U0sU@&>JJKXu4F7JFJ!;XCFy z`h2NH&#&+Bsb91pr*S3otuS<=?XTyOE5oZ9aL1VRl!qd8^bF|rTaT#WTNzH;I$Rhh zzK@c8B%fu|{x{a^9^+V&+?Sd450NW2ER{!wFx>j)hrxLmE z0nFwb9|q%1UxfLgv*lBUtNZ}N8n40VeG_uy5>aB8PTq@t)6zx5_+J?#`PB6{h58QP zqeiI`zxe^-^!np9N>FI)pRSESx-z|#a$yH!YqsR&UgT>WuWL#B|MKa`Hk_?xjbzIt* z1b8eui=ad0h$>?8rNnc+t_(-BZXBWm4cE!a_7uZ_bf?op?svuiZH-n-5Db2D6 zE#5K#JDO!F-BFrOf0#vMZFbR@ZsHlEJZ%oCSw>RuysnH>qzFySX6Gj7jKI@(u$b-H zXZPBH)=sauT!rKKnwE(E?luTj*$?MabMRutEevrI^-hxfM*CdAx=l?`nD7n`lXqd` zV`Zq!Kac#TrO;-!(2e=tWVoz587+Ivtuk-@S5A}aA&fU`9?iV9oAI5`$NsgcFtQ4^hXV!wp{46p#q0B3O`VVui_F@BG{QdyLuqt@9eTE{#(UFvYon9>6+c1`T zmCh$stD!<)8nh_zy-m#dcMp{KUr$BcLz2p%+j9oMqfC67jVWaM!~m-0^E*_!^f(_0t;{1et!zr zbZh#qYDr6jME`C{og|+LBld&q5QN@mdkA^{d>xAf7B@{%L0EXZCvu&|^}|PPJ8tdz z3LWX8q#6ALD@H_O-EBiU8D}VzAPRQ&4@7SGjTbYeiF}A+Dy0m)4zp#t@w+Bh6fJKn;?|= z*T{ZY9cU-~4rIPrf7}1>_x<2W-00)=+?h1d4}2=8hdr^i=)YF<@4A@f$Ngfs5~J=n z<4naZXg=yK$ngGXjr+x&`Sab4`I_o({HA~fe1pqSe)`$2#5qe-w9!)9ZT$>8e<;#V z=99YBk6|_)K7>g_EAZ93g>lzAA$<68ZqCCdZl3lmiq&_fRcHQS@1-s9ZD&ZfN6um3 z;9h7>OhW#x3+R}61Gbj?@Y#}a34e5;`cqR#FCib>rVODQn!)7tPP8$M{Tz+WpG#m; za|jW=hG73OYj_V|kCwAUg{_6?;By4yDFG?km$++j3HTFy6A1?W4kNdR z@t>sk^7|VX@XB{R_@7$C85IgtigA1Zv@jlCU&O~JI1q?W5UkEO7Lef=Z=b^{z1|g_c~-^S8gf}G494T zWhF}L{}Wlm6sSsz#d&i|I8e74hV{<4!fdRP^j=-Mquk>)j3+)6t{rYl(S-#fhm_9I zSKMZ!FKi#LMLnB&YQ8SsKiw-Fkr7gKH?svz_nR;>(hC!QRB<;M_G`;J#)0Q7PdCii z{$q1{n)-8&Aa|3MH{&i5_vCMV+QD&fs}T1Ww0IWY8?gA1#{c$9EwM>hSf-0Rugk!V zIDxQLvuV7v64f*R%`L@k+*~(f^k~S!{P=i5cFeh(c=>W0_?2-;9C{u1V?|xbRTjf{ zmBmF)Uhd5NOy^Pbp?-9S>Fqb`XX8+tF*VDzqwguNVD@JVw3Nd!fBJvE>8)cTk(zuF z_LEpU-JU~cMMo)W^eO6K96(sUk(ON!r03(rwY*W_U_FH!<$%9V>Mu?ceT}3f$lK(f}Hlj<>8dMY|>H&s;2Mn z3)jD^3QQ|F=D%@NEZZ6jj;gAp>C*%;zcoiI(Cz3gNPlL&s4ucm@$fN@SIrilo#lQS z7xM~*->J;UssoE(Iyb&MRlHNByF1blHDxY!bze-|Ix;^nPJ_BWW%i7@C3w!xVE2NK zV(;ur7}obS>eCpWRq6+5-jXG^?czDKwk<}GE2&={!*9L$au4>Leu4DIO^9QBI<}YJ za``hap=F5^_B|YoPS-C$uQ3KUa&}-{Od7s#8jNRUMdWMTLXQ+K(z*9nb_PzIwjJE6X`9097S zXiQ~i0CToqJ(w%Zx83V5BPqHA>F%0GbZIMH?!xdIH{A*2ww!ey4)kd- zFscsT>zK~uM>))b-Ep<_J@?Z2GHiclqc$KON-|;@o3bY2+T$dQh>)i4d>?ABu1d|- zIaq9b3-LSplk9Xkx|aVQSxw>?GEiNC5XK)+`MC!@59~$rO@+`h^Zh4A^ zSzARLYRgGx{~ywhCI=Og(|1FD^#W^t^p;Wl!mkf0VO9mhOeU&l_UVVAj`TS}jXdTi z$2^C&akCf1xkZkF&*&RNfyaPVzMMgvySXhS_}<>O7c z2s!3&mJOwP8L4~WaR~MZM2>a8cBOF#Fq6d(raPvitb9ETr4;ceeGzKksWYrOMY=yg zn#v2Sp}2jq&<=@__wci%H`%_{Al^ff&c7_d?(kM#auqqx`55W+(o^)x~1ZF!jqU_IuN%8%4p*+_!Rmfu>!yS!hd-AFQ6rGTa zh}2qvvu)Q4<_A`yPi?!uL*{EFs-#jl1+{ogxMxTyA4R@VbvB>O+_noV#)`2I_G*uy zb@eLleRIRHwPnIuBI%oDc6U)D&3HoE&k^!XJ9<`r>#+!&9a(~2{_*JIV~mMUu5iN` z4&*4q6sRI(cqv{n-Ui(Tj3dT;A{O*pgXxd+ zQ9Xe9Usr`7PI0r4=l8X7T$}MPt~c{xJlsB%WTXy~#|T%U-?yH;!%bVFp|8P0xy#}w z3h5>Js>+88^F(R)jaZzu7~hq<;6S7lH)D+sw`XHIH*TUTBKGTG!mkn`UGVCjl#{DU zLq^skcfxK#PV@>H;d{Wl4%AfjLinBe%^yE9zRv+P0&~xTr7vdM)~*Nb=bHs*H2LKMaRmkKo>U zQST?YFZswBviTr(1S$TqZMS%3yMOQL-~YQaZky}bq8|6i<(C-LSBuYXmR4i)mLuf2fbvK?uLp6HuV5@rnbt8N(A#RVA; zX5)C!TTUsbH)>YoU}Q)h`YP4p&$e^8u|*SlO#}YDcQMSqFm0ZAuI$}yD6ZthV8riy zTwp$4Ez?_}7PJEaYQ3RYIRz;L9pTMlabNxv_ePTPx^4=Gh4mQPc_o0}_n1cY1}gOH zMka9a=|L3)GxyIxY0Ln;q>|>_+dSUz=VwTFgD!;(@N# z($xRjJG3lJ#7?%iN%9#%^Du_2pWD&slv$K*)*p3^ZrqH6acEu4aH~6|;fhBDmW(>X z)t?TeTNe&d#>h-^3D2YGj=v)7=PBsIZUeDh@X)=pk*NHjXmXUxv-=~*!{|w zGD@webMz`|@3xwBy~S~!wf={!yYh2P>d=VkKezrHZ@WV)QLcXj(^Qht}`JGt$sPuhR5O78WmcypYg(bE}@LTv6PnDla49qQRCfqbmfjZ zX%4Kz$E?S2uod@X^UcRor&ast+4?X#GAxC%{EyJPqHVM~ipA*FY@&@h+7y`o0@Ak| z&`Vbwi<81GG47g!Ffn}wi=<;hc{aA4#Lq>Kq1N>f)_87)>({&Z)iE1!?+)Oml@85h zYpEqM->XZL7>}Pi`Gm*_?QP6B$IqH*QvMUsC&s?98V-MYP*v$K80h6;+Mp)TGI=5G ztltk&?EL~E6IIFk;SAwkR#iErhIHkVbmVwU?ZLMk?7&ZrNTAg(^+>*t4bA^(LE&Ru z=yA7Elz32@akeV*`n|jH72aS&&%EI%|ATSo8MDEkspw6D#&X8DTuC(oJ+AUMp38Yku>M9 zsE2s9E(ql^mjQ*Zc$IJkrO&rvY;Y)y#@)i_dCl-K_=qJnC-F%}JVSOlsX(5in7w?1 z4yitAPXmA4M3`Prnv=!!ahW>gpgWq<+{Lz$j48?I)aCuK_o%|A$)Y~_m9Zijch#eG ziy2h+b06Pm^8m2m$+Gmww&DAZil+~i2&jYARbu%?Y_Z0f1!}TWon8)}(?^=@m92E*& z`3372w4<}vML%+dt47o_+L8=gH_(H!|9H`3tvWEyu=b?5ZZN&n7IA;%t3Put9<$JH zfj4%`t$^LidhS}cD6a2Owr}aAPOtx5z%q{ckH0;GLo+MTqjNlq;rImAxHs^h^Blem zeCWVzrt{7CDAdJgj(A4Mv^a^6og*3VT|R5`QY;KPfQ4$}y6ZNV`C#uo4Kwp!c)axi zMq8Z5%unG+VYtGT-+$p+WGro3lS_$1j?mchR9c$0O{lk|{R{j1<3rqbS>p#ok)S)I z@5=-o+|!?mJ7Y#b<@#3q@fG*TlC--fs0d|xH(JDT`P+w_R3>T5dvbNxOnY=9$>!+m%mAp38gLkK{du&F1T3ocO6Sql9rNk*7-M zaq+$uwPSKVHypr+!lBp}9tDr90f_Pu&#~Q34MT!s5!C#YsPdgL6}NY!tqX?|IsCxk z=85Rzu?t%UWP@`#&p1|2;z;fW?8!Bu%0_RRq-#&k%$_gF>*J)4s4wY)*)!K;SXClq zrud^_g$2TM>$rfq!RYUJ4jm3hB1$(72kcJ^&%}*p_7@gIw5B=*-(F{-uKODNsdL8i zW_y0e+9-bYnP5JBe>|V(xr|5sZ>Y&MN zQ_q7jIM-9SC*S8PN_5sxQ-KqetXW8U`>m+Dgvg1-pshU@Mr$VyqKaQbN$zPB&GZbS zL5yFo=7BklGPIzKev^cL)!_8$Oz%ljrP|_p^S7VQc`;id!gGvPUjyyHcg|9(rZn`$%Q z=~0Sz-G(9hsWxU5C}REG$uND@UP$+L?0WbNxP-9GTWCFYhT(?fB50ntt{u_HhHQQ@ ziZ%7cIm90Y)>0vP5J{W8}oixSaV+ zIB(3Y`^xpoKLD-$210)gK+V?ic4MIEK}MJ}{5N1l=z>nY)$ z1^-o{2QQ;Pn6KV6mk$p&=F=1Hg*>*dFNKBeK}_GqIDp;O2=}}*Q*dvSKh9lcT)DAP zP^^dqBgcWlHw)Uk&7AV4Ge5TSJ@jSTSRwBQ#v}Eeapm3(Ig0}ma*%p93$yQr3$i3_ zb;0eRzS>*+;QnH4;-8*Xo)FDQF85ZLLQb4ovO$aP(Whpx?=I(I-h#_STu1 zuzAj~T0YugoZl3ote?v@$ti;}33A*z--q9mFW79NO4)Bpke>4raZ)K#kGEXLmwfRYdu*UF zp0<{7=Ym-5O7b=uJuQJcm29BtpZn6cYHND8)|t~^vV)s`QUOyfgP~ut3!fE+AWb?$ zJj+VyD`h$;lh3{yIPdU8f300y1mi&-eCjpsUwE(fe&7^2N%_`!Pe-t>lx{$SA|)+p z5z04s>O^dant;UC3?$aFIhU_R>G<0W!+kBZeB@y@V-;SOb)fB~eiYI(j9lOJ5$c;P z_eOYrsj>z&cax!RiTjZ{{2f-N%_SSHSPE6irT)u{Xkp0?A^oec-%@}2*wQ1_rKC2; zRk&Am)kP@(aJe{WIJ5b%pSTuF$~s-v6z4+@;a1P;X!bQrx)O9jkaK`>w`Fe+ z#)QR0ds0N-;={U+kx!;bbR zzC|~y&J^caO{;4A@ttyJ@H;-c@#7NT334iHXTUz*0iHD#n926H8aKo|ceY+Itqp(3 zwcjc3ktO+6SzF@#zWLB9pM*~i>tLfe2}`uz3T>cN>WO8#EC%zv9xZSkNw=7NG;ZP# zPFHg$tU4M}V3awt!7==MtAof_o`+9kU&3rj7b;uH(IqP_I{(!UN#9rCq8j7=h+IUM zyZO>B^H{p#$$XXO=W?j~$-OPTj2TC+pr5@Ac|OY(+GcToMLf1Sj-8L?$xgl#t?IZP zbT)V?wYhIk<Vp?7^N zV1HtkkaqI)FBoa2E|f)gu8DB(yoL!m@3l<4=Nqz{F`EH^88gv)1Z}f`=mC_b=NrG!A=; zc&|@Fev)>uK5?G2ko_jelU#=~4h_lo>Y-Q2RL-22PO((1J++2^sqDs2tS_L0F8XL% zQiQwiQdIEBn09;;{YKKt{Y+Eiws3vVv~kNqrr}=Y$bWfAz12*> z{d)%g@>IFA99ham&~g;#$t}xesrxi?Ndd+1&%)=h0&sKOa z3o|Xfpx40#-;!eRZpazPmtBVU2>*Y1kJLMhtsUF`U29#uML%t(@~$4y{9DURKBhIE zf7{rZuj{cykbisQLnsvM(~2$zG;*bQ4*J$3m|GU@$uN4-G4A?*YsJzFy|F0531eAY z*Rd=!8q_N0b9TIlC$(VjSNz)SLjz@}$@}+-MYg^g^%-eJBj*Owi3K}p?{RV7l(f&^ zPs*Iri0b^Gj?}*@4*fUfV8YQH9JR5-kTS-L6HW&E_Kuwm_gYrFJQv1|{�&l6Y@5MZ_nZcLAB7K>({RW9 z3aVa;`_hUniO8Oo4y_B7IH>vkUmpl{y)ljX+djwK#Lj}8So?ScPC1KP48LW>oQF6Q z%>3~q4?|za7>0>!7=PMi__00O3axTjB?k+#Tn-$Dydtw%hq=(%P0Q$ry66`pkzaUw zA7WTcib-e{qAR?Sqf{)s7ik#<+)?drICHC&^Jt6WtiEm2FA2H9rGE3q=6x*~zsHnP zw=oV8-7476=tUPh6J@L#O{0$MQOejTgiaVn?aU{Up8IaPGOPzrs^j@xF53K;Sz6qa zP;rl1=5#>2{)Y!vHU8#|%L_O|rkl~$9fP=#9YVQ>KN0mgl5#4_n_<3kF@M zaCXCFI&@T7iq*HTLf=?{jBj2Qp37K$4Lvk8$*;2-Y36CrhDGAMx|#XZT&~-N!)-57 z<9S?2yJ7VzdbDjVz1u&D%Cxm;zh5KVfA|Xj$&VU?)3Ol=yLAy=A_`F3=|8>A?N5(! zsLGj+&B&yz3mb%TIR=(-*CwT4ee)A+n*SRKRRP%1M~5ri!2G-=&ck^Bf*11DsAtyx?Ri)x$M{N*9)Pm#P!w-vy3)a^!vDud zEJKI2ci{IU2QLja(f6JruGpC~gZV>aTF5%Vh@Ni|_2SDO4kGihBk9_tE#$&=mQo$N z0>jt+!G|^PVRyMJxn1f;@@pfJEgjDd>MH78ZP#3fUCl9=xEmqk_)%PpT*`cWQYrK5 z4N`rOPRCrILM2hu6Su2t1KGVn={9knzin#=%AIROQ<9k8Y>hUp@@vANIB#@#w*{`n zF_1mJ8}5u>^2YMT*j*y4w4T*Paqm%B-71B7=0iMvk&X7>x&scgz-yesanC$FdtHsv z)GX|B4uxsT4;1#FNF@tbQcc8SiZGZ#MKw>6E@vdPjquK}@i8#MO}IAi-IK*xyOZDX zdE^@&PPO{`>B&(`dSFTv-Iw`fsM^x9FKRU9HRBxl-o*T_7bC551A0`ihm+$P*e|ih z^A|laBzZd2ZY9BRM>vL+9>B{@N0^N;7mqUs<9S#Va{OX&^Z*y3FFT z%Rj>P#$BvRe~)+F+4)%h8@hebp-=XDG`!+H4lsS3RMk@~jdg&3ms42M;ha!zoxml; z!t&|4_9dz;Zle(+r1`32&uCX$7MWLXrBEft0X5%<3T7CQexf2BnEw(-E+k-7&lfZ= zNQXa}pu?v$5XO0LeH$A8%94T)FCw`CljyOh zhH(F|jEF7#dbS3dV{dRzZ`#T>>r8<0*A$^nS+N^%cCk0;sWIMeSdJMdLm<601b^PI z#lDw%h+=D@<`_r3J#+>ad*|a&_Y_D^Spu8s{V;3wZ&j_GJK)jX3*S`-{7+vmk-CFc z+b{ofPcjaY58W5msu%h47&T_A@LMs+0oK+rsOOU5>E@5i(?W&kdi4lFP4h=AjQ@>e z4C}Xa%xWy{XUecfA0j^6n(YC6bHIZsMLI>{tP0_UO^zEN8M^x1RB1c#+jg zalIJ#yMZ&fB@esLuQ)rEJWfwl^zV>hDf`-X7UZ5jtPayPA2>6U%iPUu(QkB-Un}>> zy^d>64CjnIPT|?0X4u3ZLaJOe7IuFGy^{7vR_>=C`rP7qgulVen@vHqYIL8pHqA?i4LI)ECK<#TuRx zx>}Q6y&P#py~48D4DWWbA4XJUquBC_AlKs2VWjuaMeG-Q1e+xxa#AdedR)Pzi`9r; zw;!YC9K!05MhyHOj}?FHncqhkX4HJ0OeSKQWIVJ#arScVa(Z)wq`bbM77< zmh~X3!(C`w%@3>}eBxh!U1Ra1hRpZiSBK+p91-&G{@=bfpWlYQ%?(&_aG&s>N7Vy# zzgC3$_a*<*c3#i+-IJa&eyBgfy>RUgRFl$+`k6?9Gjc_{o31d+utsg`V9Xug$VIjj z=d+B#mk|{<1!lvU&&K;Ps2r~#75=mh!%aJrI*VNn>e7eY*&4aPDGlSc)#32U7C5|D zqW~5+J<&<@1^nbt#(?c2AntEdkmBJlhlhPyk5%zfw4)dS+%dDF`UmwL~@og&em@a}8`d?Do+sitrs zk@=ABWH?cs}C^HcE|fmHFp=Lu zA}4T{sc=7g-)Q72Pr>sH@tw(>HVB@{j2CiPDU{196jt9V!`LznnF}rzLi&Y2`_bD% z9WoiqVwnprVe8sYD4(iHyMdvHC5#v1%Nr?$aVtCoXSLXju znH8edN#_4Jg@;bri4eDgjGO!%S2Jfl*Y3+6uEQHO>kGQX1V+A{QjFA$R-Ls>29Ql}oQ8Ntk(7g5ArHFbhK~#jZ1x|?8erp|virq*}KY^3)zC&~LT{u0shqYCaFbuSZ z{eX*@eBv4oTFv;^o^o+M-0B;lNN?#zIkt+_-EAW1&SUPac?@?-v4E@gv%=A7-H>{7 z0hM)W#O!knBRr!MdFCFa>mz>A_=F%@$@F+v^Nw;_!!3|o9D%i?Vo*AGBh=%6a_X;Z zp&LJ#(%Ql(d0h@Q6f&QFR$t|DQW!e|nCxrtWo|8Vu@aXmKQ|9G^Lw4jJowrnLe&vUh?B(j%mp(5Fpw78WerA<-_WeX)yNN7)% zY+1@KYj(;myCQw(bzSef`}g_%aXqfrYp$7d&MenlGiT0u9t18krjk#0XlFMKq0csX z;o6k-WRlFt=dEOy*w``z*6V=s#(S91;tT>ErklSloypF}I%C_I=a{i=Npqd&zFo-< z-qFCb_XEVVGfq(~FMAiutUr&}92ZR{pX{1+--9h_dxfpv9t~Hme1vS;hi7%t+4YmB zZLuoV2Rk)(!hck$I3E7dZQ?Oza$sZBhZam)O0->`f^X%rA>2OwNt7Sk*ViBJXJ0{& zfA8FS7K>$y^3>QR%X4hLh9?>mOi(yJRQc=)SJwHBL4~i)GqA z?u~jLd!YLFd1OD|f`VKIpH;d@K0gAjwyuL+@IY!cuQi?R`+=p8l=PqNVNHEKIUfEe zFC<5nvIh5o7#t@@hvqBOS(30?I^U4y)Xgue&2t)`GkJq0Sbs{lx!tx_?_&RGEMu;T zt0nPXis`{`QF96BxIctD$=)!z^u*C(3mZE|C0xShp`p3=I;YS}z? zKTKNp6vysGj^!_`apAi1eud1Tt{?Uo7vaz5X?Q$k9qU+drKMTF%WQ9G9b1-sjqT#+ z;`x3Kw9Z)4+k7$6Vst8wwTrsLrtI0uI_4i?kycXtUz36k?7-F}b}nx)E_}1cn*U*u z%KCJ|j=8Ku$HVM?`&D9FexLS-Z4C%SSh%E%N~ev8Fie$9}36n$RlQg7qCJQk(Y zj{2vPsRdtCeXeUSRQhL9&zs|D-SXu$WSoQzf6>JXtXlb8}m1EXG&vqeN8oNQGB#MmRAE%Poj=~2!W5Gu8gZLgKU79J%8jKng zM5&UtGe75S%x3yqt(dvN29e&XKAm22-ipUzR0n^^I@r=j&@KcFIMAuL3@i0@dO=f%bk@I(13hgPi*wHfrGXqwj zecUeY2dsw}Jx$!vQ^C=9fsD25jE&{N_~w;_rz1=-{dx#y@%Tam$DLz)%w5Df_>K=$ z+uc5t6`oHK|5u;3$EL&>6g*ri@~YHJXU6ASW7`20EVAy2X+NZIVsMDZfj-MpIL^u~uCo3wwQ<>bCcb2T5oy1!{wdxkzd0}7N1pCT2aow; z;%;AL9xR^D;{mpi`e3Qt7iMpAf!)3G zkbN0gz-CRH%x2YBisN6!aXmWDIw96&_kt>DDF(BLm9JoGT1^jIV9=6Wg~p2(ZQ6!K2np!m~r=mu8t_-?)^iCm8YYd0KtwMdlX-+t&bss~E_{$Z)R zId*Q=cuH6uM)q^0b>pItd-zuJo|WBAL(jM6NU+9cTH2dxwh4pS7=&aR5YCTf= zX2%S0GB5qh?K_;z$u5-J&s8lJ^UHWCG@inu*B11TjU~x*oS(i!^0C9|7x!W3I^8?n z$Y)|5jE*HBuTK&N@Uui&nfe}G$)Oy0vbrg~OSiF~-7NLS zg~9*V?q^5XSgbnPl3JOMphLZ&CCch__}FlmYqq4e3k_&Kx7Sov z@ukwFBXBb5OXu%hqekrj;q9mx!DpbNa4w-0rED~Ua>83Yo~uFMnH()UnSy<8Lt2c# zUPw+_6NF&4M3^|uP?+>&hd6J)dpfGajKE?Xs#T+r4|_x!9@M-h;RD` zt-@PTbjM409>0pUSgU|^zs}qr_BeG|dYF85>Tqtzai;5kh0QyjfOgIuMcUqO6*xJn zimm4R=wqu4N&R6Exoz7{l@F!8Zn@(EQGS>4S82`S@3hWwA_Xt#hc|n+!>-ng7G|!c zTD60SJNQt{ceafIrj_l4_g+VNbTn82Xii22-`%}JJqP;#xbbgxdGnURp>aUg04PN6kK21 z(dViKRP6bT>RzRjNB(#0eWF78``S>k?=5zwNEvoxld-AQdBhl;5ov3D_CYTnZd+P0 zg$#$8(Fi$dt+R4?cT_}`U`!A?FYOdX!z$vbb3Di4c&5FjpGoUrEJ5aRZ zcjT8UlXKHQWK=dt?7PF4T8ej$4J@`@LDDzq?hk;1{S@(kQ@_V-LN9fiX2mi2(wwPQ zz7|!(=b}9D8EYxKmYwE_{O!Q z4RtMO+SpYn((4B2IaRF6YZnyV{$L@GBYHz&E2V5$Lkl$J(`&cUEA~$Tes;X_5GScZnQ^S3lF}k8W!r zr!t+Mrz{l9nLXSipQot!G)@6Km>1rztqaL6~%5zwl;plF)YdM{@g- z!+gCX5a6=|p*-%@_Noy$y zuaN9;T^s(>W1OzwHuhF=m;p1WIWL1^t0a+^`e8#fO)JCNL!ZR|EWuN}r?))95y z(>HW@j*pL}Zbw8)EmfAbDk?UG8MHCz|?xT}j7- z7LvcyAWD$Ch3u26w08XiM30aoxzP>HWo6}4KW~EFtIg0Ha#BoRcWVGv)NO@X)KWAu zer~09yRR(YnYiv4w>uIxmf29dl!eZ*smPzraXNNnW5exlK&N^<`h1P=tLEW$Hy7M` z7mE{r-$T*rHoQN4Mupui)NeC`chv)SHpmk{M)EybI>#v9(~}gZYts1H|InPO3W}__ zPM;s2rPsDaSL4#(tyF1PL;cGeY0|XD z@W;T0by)ss2ef>5!M)=~k*_yDdkOSYpsRc9F`nysmXGaAwkIl3(rY4KF4=-FBQxQg zdkD|uw<6*EYP|oHjLNl1P#U@s<$H5s(M}pGGXCB4d{Lf1MWpMJDd{3*>BVD~;qLIb z=6cK0=5j2fR;BXvyZa4HZM=*j!Us(BI)dKuuW>4KD1{Wd)6=d_bY-YDHSAX;4;!w> zjlKz!?hhdJI13k!_a>urG29XLIr(^BQ;gc_=Q-Y?DpjU7aZJPG=$)rT*`FW6ebir+ zESA=(`ZG(|>Ny%%Cf^IU99wXk-ycl%*m34I^8>qd+YW{UmSH@P1LWrUpB_9J>(M!v z>&@OQMBM#$SZ>o6Pa`;noV7BU6l>CzEBw89)P`JbrSp#ST+f!{AB|&^Ly-Ax6bg^G z6=k(d+|2#K*5mr&2xtXO!NTIzcv-a??SIuF)@v{gDj7m*8-vf0ZKpP6$>4VyXTmDu0G#i^`tgBx2OlEHkmC2PCF+5V{I>&Jmn-q0WB1Ks5_Fw4D;Hu;zc3(~cP`ALSt zlv+jMa#TJgr6j;*lrm)m9l_~ocj4Qo^#And>cDz-{nR~C9vSUv|D3=izCVg8Q>HAA zeHk#~Kg^?={u&s)Y%qprjTC8=cCTYe`Ip7tDIAksR=&f}yW+i8_{(m1dq78XbBmQN z7osB0403$^sHndhf4Gi8Yv}u$-BXG;z6Qv_{G5L2- z)?@l8s(k*FK{CtlrrJ zDC_pJa6{rz3IwUi{6G-e12IdwPAT z3ti|?$y%Qo2y^Di?E>A9HK#W$hjkR`uW*RQ#^iM<6Fi#v{Rz>A5u1yKDLDvryU(X{Kcb&B9PUs*S zHH9d?=UAFm+?Ki>y2ApSjA7l;4sj(GkSo+f+T0srA39|nVYZilGS8p2Z1K3AOg~AJ zU0ZcX#A|k|a>YR7TPoq_B^|e)75C#}9HZyQd1`y1oVwT?6X)Rf{v*iHK8IWf-KCVo z2P7ByiB@NqiO-fVwZrYPmYDuB9_{u^I;ZtE(pZtzp=P{>TzWujxE@t(W#+2^>)J@j z-98M9_gT1;z909NhT(OkEs8S^LP4Xv4_rR6AWHrg!yabMcP;9 zZN}nFP3>fKS8fE|>}EvDqtzgE^20IB4|Mh6w80Gk)ONZ{oDy?6+BpcK20SC5eIXbb<%7UzB>RFnnD4PxVN-bD6|lh8Et#tw^Il-5?l_TU*6|8v`M-}eE6lX<&b&xT=M8a(?3i}I~} zu1TexSCQ_?-Q;#?Fg@L7f)4+m*qye?(z-Ni!XEZcFOGHx?BMH?(bRvgBdPycLmJhI z6cuH_5mT2;yjO=lzpb)4=Zx1gZ zqgQr(A6dxb&8=m4P{I1HlHPBdxVJc7)P|m&4yL{}jSK7GPJP6N?NWfB>y=hNMaOxhAygoe^>tiH(s=Q?#3hR8(<3kQxA zRs=sIn{*#Kq$c5S%G$(1cMugB1yYRiEXpptjes%IdVIpC4ftJFB+4{-zzOz->o#Qh z7mkWT@?~l4|J$Y>jtz)zDNn1WxpR!5W308$GVy&GF)bIJcYVe8V190SL(=t*UFU;! zO8M|Rl!zT(6A+|urup3}eKeaE&t6EK1~>2+%6cL{8ILLZ7a(Hoa}3>+gMNk58LfKx zakMdhiz$z#{rNxaqquimRQUwzT3u+a(34)bQXpJh)=VeXSceAu9Y6ydxDUY*X|I>v zuL$|)I#AzmcN(Q)N(G^jP}c3we$35->X|D%R&j5dyJfIAkH};x8+igF)YUkRE9kym zI;#)g@KUTp|2sFZSxK8bx7FdF@B%T-@p~5f#{R(J5_Kv#@c~C#jE3@RXSVa%9F$+) ziGQBVL|!K^C|&O_$~k`S5{xRlg{Nu{kTQQIcG}!$PkL#w9~;-1Ut)m<*O<+$PDr&QiXlN?6$g&IC^d#7N1{=u%%nD{bPz)uF+={y_}Ilp95m(;@s<4dCUQ}I+DI< z_;o*I%AG^;8@c9sX!rHRR*w6+{aYk5k4Ga+={T|{lp)#nJZ5YfOIgFDyiXz&uv&8} zTk~tYV0t}EnBkr*=soWv1We8qX}!%+q-6`7$j^Hm{j8NIo4;NN{*fw;*&L$RR!5^5O0gqX_j*;QCv9#dVJbFBnL^@M`Jp*;=EM6E-8-@ElerUgkcTHe#A;X|LuwemKFfNY0-9 zzS5FSipYTN1#OzH)|FPSb|8m;9H{X278F_Cz&PczV%aFK!?=0mIVL@9OGV1pF=Kg8 zlIJ#J2e)L4&t?50yS|^8gok6|vFxZK{*;qg*P!JLKW}V>7sp^YGJOkboW^0OeB7 z<;@Q4!E*C@^m}Sf2|_r9OdCY|ZQF~_l8h^%Q+1GIHMb%Y8IIS4Be<8FkM7e?iO+wl zPR8`z+33GxAqHhC6Wi)e3C@Z1EiHj=9&Jy~$D@>wkn9)EnWW_ZOzL?ikDRDP+21 z53}g1&8Xb{3Tv;1VU*@ImN481I&4KC5vKf`fM-RMag8;X$;1^Sy!>NR2b*pH>VM$M9P+^OAuiwKc{Ri}Q^l8|~lL)4+C~HtE_GAmz0M)ASC8pHP5TiLc;lXG@Ml*3g$pEhtSd z2ggqJqqTBlXkmws7^Ssa>|ZUGg*=BAwB?v0JxbjH739IG_R%8 zikQF1to@8dF_MqAoPiZBm`>!GA+5tzjqaMad-4owA-1&hgd#mj$-%=mbCEOX1$%rf z0L5G{v%~NCRahl{8FXN|C)|sN#MAOE$Cn(3Mp7Qj5(DDuu z)HdJ_rrVDprNXV`Kc2^g+s@^3$<+LNnf-FQjg(N$$7Ri<@M-KF7vjNLSiae#yXM*x6_fyB)i_uF{5u^XHkXQ zR8oF-3yzO+#P)o_*Eh2cyv2D-OLF(y!(*~7p&mVMQm-}!!if2kgr_SbgiC|QH1Bb* zt=~p(cOIp}A=zSm^)|l2jj75)-!;pGc9(Vu)vGrPI%?eoQyzoI)_(-zC&XiOV;K;y zMByzo=_a?wc(L1uF8;hqTQbknx(^8yYAU@0>tjf09ky0@)p%H_dzB(2Sn3ONUh#Ms zwFY$3$(4F=J^v@4)%1JLKDs~1gVrwUF4nLAmvVeI(5Lt|!|3j=i|AY#E2bMxl%tMa ze?jADNpm~+e^aK--yO*1%oHl0bq+oB?OHqxHl$y1TS(6O3e}rFpp{w$N z_$-bK5we$*_HU&2s5D?Unwjb)r9oA6%*yHr&6Gd6I ztwBcpK0w&`9&5)GV(S4W%Gspdg-oMCfyu+k)W%30 z6EgZOci+byyEl-t_omcowUB!x*?2W^9GcGE^~l{M6(73>;o0&ZeBM_?ko7+a1JO4kM+XG`h!*1ts_&;a?3!QIlkJSvEJTeL)+;36FZ`V01^cgWu z>;o^`drW<^^xgWTVqt`d;=G8W%Lu$l)EyEYFpC-^ii#UD@>dZaEcYMv_~1C2HTdJ#3f9!O|cF zd&{|fgsZgQ-sEY@eXkQ}r~Y|*aA{97Z%@aW*!lYeT#Aa&_MWr`N_$)-@|MXpU-K>B zJKv`O&uLH@i;QuGS6c8 zqTZBw)QpbDm2e!GO)&RuU~Tp$3d08~Ths)oTiA#E6{^Ce_|EzBp2NGTH=UX7L{+v^ z>GnlwA7Hv9R^jUFWOVC%2gl6iDZ^Zu&lhQZvT|J(ZdObCQ(66W zZa>3^>8A94_hK6Nu1|Bjg?hZfy3XzC)sJ@M7Gz6PV}iwJ$2=FJM`9&LY<`WG-{oi@ z$Nt)Lp+9EayMR|a2GeQA<2PBqM)`mr=&6#1zt(MGynHCkM(5yw;zE?WdG#%dc`2@a zW&B^BZbkWlwJ0c+^xNx(9)QjvO``Dkc(Jb!ohaZK)4QXn*m5iyr*Fnhvn-T2+~$5q zl736|>2VByXGLn=4Vq=2cCwzyX)ncw*7=w-`4e)d*ztY$5SsKn3z>}{FrdVeKDX#j z$^{)MbMb0Lt87cC~gqs+A;vo5F;%?941J{DL0#c9(Q)$w^gEtZPeqV(Ve!nnd+T z7kRui=9pUP&*)G37m7YwMIE?qK*n3mwuV$VX32F2LAYQVB`6f63R&AMgrIrzU~wTI z+2bFfepU)DznhGVvsrAU$w(+_or6*5OmWQk)GoqTCr$D90;Or(R^EdwIvgYa#hH9WQorfWw)a<{q5=Oq+wJr?l~UjhQMM@@8-6Y$s)VbCLDn8do414WJZ=lr1wac z9yPU^bkCXy_}ELZ8r?-0J>e^5U06%wx&58|sB0LeBTqvj+mpwI+Z+#5h5M(-$$9?m zOOsbFqK&6ZX-7)^|KWAT;14GDwZAmt@1O z8St3LscK94Uk#`rQi;BwvxDQM+xW7M$1Z!hf^2R}wxu#TR}~$_jyp;;b*CzIS5l`b zDi(BZ3dem>_(lm@-onIGZ{gOS_p~HOl^zaKN8G$aV*6-nZzpxjA+$eiE}|>0nI~#h zqpqbh$F23IZ#j=KCFVGE3mPNuPeH%Srl!mn8|w8(uc zd>fgFFDuLc|GySxr!>DCUE;Q~(SsG5Y42kXnCcq?=p{?*5?NW})UG(JIubpzx5NHi zI`*Dfh2*a#Y>?|GHvF+M?g!YRw5Kzsv=&4H_Zv*>*Wrw5fa^sjDAn0s>-#rcT*B=9wWWq|F-?9yN>u~(G6o0Z}1o- zMF?6oid-sU$!p>!vfR3{d3@To+RncYI`K?Prp5Ip-Z0M$b+asBrx^xIU7!KLT1 z{*knQH*GPG)K6ZZ*;?P|mCY|&ntquM>bTSRpK+LX{5TZLwj-gC&-K@C;?jVvSo(S& z3w@w~-;3R$XCkdjlC)Zgb-Sg%fyqC;!ybDaXNz3Z#qX|9cpv;cX@#*KaWG!XWBtyR z&c7tGa4=NIW8_kGe<{!VYc^7W6*lcP2^mY1Iw#KTV zWt6-M7XK3urHfm!WTm8gog8Zhf!oIC@-sR2>%VW#E5tv&+ma8O&%i9 zo7EvS`s!bie^W~f_WSNEk>9`a^>`h~2Cdr9_HY~Ey|<+^i2jDfVjlk`j?Hs{EdTRc z^6YnrB3j%?$I#{5(Ro%5%2ig2w1+y%;eBg$xlY`kxuIcybA44WC)2=h>Vk5Oy3l4x zJ>9-vK=aqfQ$g%q_Ni~?`>p@fGlCc-K$-}mdMw!yWh2u{;YsK{u+)y15@A;{SQjrxSi|H9f(?L!2L;D zLvf%q2BNe%p2rb+4D+@S>EwR-K_xv@1-Vc?A?1#mxNrLW%@k)2@5Z6J3e?Veh_b9r zXy2FnkrrvUwC{U>eZBIW4If|$ch52STk??oQPacqNh3shbTSTEx`$voWr_ITttHo) zu1!VWUG5t_V*n|+m*AG459n7LT9i_aJ$od%8XNmkw)Se)SP{df%&K5kLJjNgQNvQE zC9|rqV=Tqa4ts`_q4U|D*j*;ATfVi_M#LrFhMwHsIx8GX`soltW{G?auWp6r;O#hm zy9~2`AH(>1V=+CpT@HTjTZ3Qzy%9LC9gndi%@eZz8a8GVYm~QW9$SL<1?F`o5Qd>A zuxwB+w&?jFrJVw7&v>9|mp#h(yz%cCl&fz~W}jxWN1?Y^<^Sxt{(WXS=^;C-*a@HI zhT>A{BZbe4`ikW?PxKRcwXbQ#V=fF8(`=qQ!pz(SWgm-Rm!AXHlmW+9(mMZiggRcT zFmxO`6cbYW!`}5jed#tULquJ4!I|CbaBj;wSS%Ev{*%YMU-y@N=&6LEbqoc^2(1h+ zGxGH}t1xJ3o_B7z`O1a|;`?C~c-#C;3u}zH!egBKUIi9Qx*}N}%4W~U{;KDQQuvHQ zhYRSSwnB_mv*OQS>Tn`LoZDOeK2DxT9#L*uAhqz0XNT?`W^?WRuS=2$h4eh;xd^#E3O$<$RU(E2_p$?yLV{jumxz(~A}j z_=*dDNtoCtnhBau*^H5DaPVpgl}^{$+53CgZ|8SRLB9)P9|d8d+cVsm+?NWkD2nwQ z@;9Gt>nV>DeeB^EHByxIU;f5%`nb4%5CR8D@0KjxY;`cRycsXeMSD*%mkEJv*pdyb z?)edRDC8%5oTPwLcjQDmzPD|mbBV``b170;mlA_v=Z=c$sm}+J=9^O_S6f0qv=&gO zW4hGZcP%CBj7QPmKbYUwhRPHqyVZ^?UM%BS_y~hLy~7e$X}{S;emmwX_h9)22Uw|I zCF|GmBa3+`on35wbr1XPmrx9k1(JQ&m`26tB1d5y4Gne2y}c1UCfZ4CnRg7y8a^Cz zvp3Sc?U37M4+iqsOslWX>Bb@|}~&?KhKrF*;n*-OAd)*K3gYEUYSn{jiS^ z+v2gD5eDmy6Vv}~zklzSM>XKhu-rkG#N&G)nPnm{VS778R9v^H$DSE2&^Mk2c)SzxbL$58O6Cv|BCgn&vJEsvxPv~Lr zRLN#oxzA!O5}ZL55b69hjK`s2Ca~yhF3NeSzbf}Zex>oRY!$U@@mMzx9e+k*h~g%= zo<4$Ku2y1M^4`JM*jw=4DFG!WPx0z@Q~Zq2d>XRJBd`{U!o#^`&$>@;kNA z&=GuQsS4MpKA_5iLo|NHN9I-`>6l8*`-wabD!yV$jV4^@H;!6a&!vqAr;FoQ#k3U- z+PPo+z2kinHMJi|ZEvrp8kaqM-Fi$c+y2%gWOr;t&CtJC8oL3SVXff*gY!4d!H%kZ z=(k{vnD2mXNBm$IoZA*Cb$v)lT?vfn@)BI!LE?eURSp0R*eXF0jcV_e04l=04f+<;R*y3&sG zI+V9uoq}4or}76iXk+ySgXZ#0YaFB34%!qL`UjtGKEQ>FZMg8|DqctY!ULxhXqveb zKLWPEbcYr#@^Ys8@>W!>Vn)`D0?jJ>PnYz|mB&`8@y6gUNzL_ETBm|QpRuTElHQYk zvp2J@Zx!KNtB(Og-mz%6yDWL85>|iw!s_LH(9PZk`<8`3>om7L%LvE(^Pb4I9s!R9 zX$ae%jb%B~Iv~1TTdcl45JQ!MKvP{Yq{$60I`{5krNOZO^$^(4?Sk&NC$kYRcd{GC z>Y_aTg5GdF%tY$)a5Y_Rzns3M#!=j4SE9;Z;EG{{W=+Gj0|6L!ek`|x)4}?A#y$EN z*0YluUs<@@CW(VTfdJE9;|xhi3H8%@RsDagz4fa2qq87Rv3G4XHK# z{n(Lmryj#>OA|b4>5s}kTae1J4r$k9{FLhfqnsER9lVV`4-Q~N43EuLqK2md!MOOP zm&oVwmXXMFkHH1|p|H$0$Nu^r6ux{GJu#V0?c5!x?u!9ccj`bV@-MRFD}7-NeQ~s!|k+(ST7e~k@P4h}a>YhiG;0!m52+Mo>Z3A2yze+R3#BLf!pUVC4Ai6HKR5(!TTDl3 zqB*p?pMzytSE}HCk;BqWX?<-6%J?biM*WjIkY9)rO<`)(!qtHKzb?eChx;*OZYlJ; z@p#-3Jg!i1FDe`)VX?{TU}DxoGgwv&^4xQhQ}&*s`bWokh!Fs7@tzd5C{6T0gA zu&CfF2HrV_TN`?zM{p1x?d^?If3#sBXNbaa0#5$~sw`*W#W3k?eRq?zp839AIvaXh zC$*E8_Xb=z*u)a)d}Mt> zA!aVyf$8x_k);%kwKdj|i_s-xwRu$7JD#3S_oY0gk(6sJ*|YrZQbEH89wPhAJoa;T zG3E7&BxU*P_B*!sMNOrI&Co;RB)fYppUvcRN6WDvMc$Fw|Lu{?mKS31IpP1b;pC(* zm~OQIbzWQ0eaT^tUswT^(meR}T*lXcyAiM6f$GOtQD3jGVjTifO2zy5Dr4+ydzb}r z-&*y_kJwqRM^dwyDbnt=Y$xd{Z=(65B4`@dA8oHxgHMS$hNtSGQ@jeYN?oxz(;Au9 zyYcr(ok&mLyCZ3*I+LZNq%W)9y#{89Jm%-T?)WrDRm@|zJePH5T`+L&dp4{wvbmij zM{zt(?IJV}eqS1@T zL0pZ*=v3_6FWH(r$!sg9+wKfBs;@Q6aeU!p_-Sx#vBF)b(+$VxmA`P}O8@5m*s}g1 z%g|g2lW`Y8#W4u6zQk1iTx73RD>2l_iqbX^t!i7|Oz&vl*ZkaEomLFjq?vB zY1l?3NBCO(elX~;KuIm5s5Y_8sO2U;w_NE-rjg6&gA0!_uB)Dd*KsmD z?EBM~UNMx`-<>W+^`XJ1eMKIN+b3aKrU~uJyGTlNjx@LP%IeqT{A!WV6rUhmPnGbi zNm+;19-B!=zlYGWeqq#WrV2$4h$0V{8>IGRf{-1(LufMEE#%ou5_U{G59R0EQMym^ zr|Y&7GdIkneSM_$lZ=k+()y)~L@d%Ct{cN8z2xR2pdIi1sJc~8I=1Mz zsPi?w;!2O_&LZtJYck{ZA8}REn)gc@j{_2~Mh9Q?pz+-Hpk&?bW_dO#t;MPx4G5i7 zz-@`^8BXY9X>JC>Do){VR1-QloNUfJbZa(LcRq#Doj+pQB}CwI#ub*r{jGNT%JEnZ zj^t9KMp-A-a4xS5HC^SXZ{Q=S96g8nd_@W}>`AtNRHvgByFx=&t=AT9vm# z@HI&mzs&o0E&pQ6nq7qN^}7i+{NaqJ$)uXr;41gxhR;rxncuqfX|@2))~ z%sE6=o95Dl3ufr=Vahh?9AYttMk6Qa3HZSS_C&<0>5H=dUUmt&fvu?W^-bIx zuL(1b!}BYy6#X6zBl%DtdgQH20}j1HTdS+k8QvGIY%huBw(vdbByVS|G(U=njhYnH z-kkF9nvlZ?M&5M;P?b9c{Zw}2NYZ|UL1IiN*_7VW5bPc|Cj zJ41co0xZ)$h_dko7_sImIt@R7E|>Yc+$R-1&bYwCygPkUJ3)iqNoQuVc8|0R#P!Qu zzkawLwH&Y4e|`Y8D{eu(qyb9jxL>oXE$uD)0rhY>(tR};o&Oi3&%D}#zKyjJ$9cue zPq_T>6SS>*QOB`eTE8o7Mjb_}?OXWisL`*LQeC@Va}@PY^G-cBAD-}ojjNBq?%$oD^o#o> zShc{sTb4-YxT0qF*HX|lL+WyOBITO zJd=meQksC)I^pb0-9cp5jH73phf%3vz1XK!8)MjDeQUJc-<#$S9Z2i19_%+XCX6zK zv*hMfLh`{J@13{1?6Nwri(OZJgR>e#=#Qg7z29EN{iG^Vw^0y=`6&wXDt^#EabGC3 z`)WGB_9HWBYl$283b8Zw8D@8VfKL5N;rwGQoKvPDnENbUPH?1FW1msTe(5{dG}sQi z#{EUZ_#k?5Xbc^UXh*j9`(VX|%Phgj2aP4K5PY>i&GD=fWtYjc<+usD-`|dnUS+uV zG87-$DT`~cCG&^j5nmUsJh26nxzF~K)F&|TRHvq~O5``LGmW_``41j7{)(CXs^~-Z z1flD;A%eq#9zyrA-)PPTCmg%#ggj+MVycYp=1CZVFZlfMf{)q2#Li^-PJ!Cq>_7%w zAD>Ym?I-^lf5B6}?Tbq_CBN9q7&@3^w{UDRzoFUmEBqK`YR8eo3rD(+Pi*Plk8Hrw z4#?#89EV~RFoEMAeZAv|Zu|@+?@4NN9n;)@vh4UR2oA17{KAt63w?l+`H~)F)!q=K zj7o-C2ChTR;SJmmClJch6432_3p89mkGp%CV7%|Rn5X@Q z#~5dvEdK6bdmO(9b9~PMFA(l`4@3Pjap~cIdx9Ug`%~vuk`0-nkqRkn8%;qwyVAj0 zZ92yH#GV|h&?aXMqF$eb(vJtQbRR|$cCO;MH4V++_5487HYf@~E&kG@e0hPk7t+JS z?I>)pGaa&Wpjml?$ehOjJkaqxJaXg(`)FUGd+!m#2#tZlBF_QBvDHVZf6HvPQZ=7r zc+1fym$vj!`GhFD)B3*BoZ>T?-Yp$S{nDhf$KqL?1l#jZB=Kz=mM`0huRBkaWTleVPdr&j?ev>FR9yccPkC>}w$`v7=&twHts2=r8& zgGCbtp=sv})_veUOnH7ClP{h^kNf8^^ja()Z(4!3+$v$iw+ZCtHin0~^d#*O62^y) z#ZA2aW<=V?y+~WxfSPh8yO)38?M-f%uzurzwp}~@EIHn5PwY2bjNPuOI2Sog^v_8u zPDe=FpNO4)5w;;maYU(Flr_TaH-g`YO7+&wAP#dut**-yVI&X}gF%xm{Ll^2X&5VZk zejv(tSIGyj+{2JkzZQ$fd%e0hPJS(O4UUDZdZmok4`pp^{L-gRR zE$QWC^n2vq5`Eq#_fVh-ccuBE{jjAdh|k8AOz!hPvIyaKuOgx#7~_3skizT(l;OI8 zY|cf|-;rzRL@!Ba7JjK44g{`5w}J>!&Sd$wOm3~DYg(My2bJff{VFN$#Oyvi2E}Jp zGSEDR+c=0k-3<5-nu^o@^MxPUB|`Lt6yb_xim-U;CgJAt{^I(|sO~MgRh<@T=ap>0 zZe0NfqdGxpq{S>N5s^$w6e_owV?dwk69~p`K8g5Aaq<4gm z2i|6~tNR%b3hJy+m%5&Vmb#={@T*C{OCF z?G2buBd&&}q5MQ1n$Ew2`OgCADcwcSwwG}ubv4I^yWOmFknwux)kmcDY*9}v`qo9k zceNYzH%a;~+4DJFVlgW@66LLU93e+(uIlulH~A)q!Jg0A%0Ibp<;cS#jZxhVF=XI5 zY(8JbbuW+c_2L~E_I?4yD0}hw_kMS<$Jvn1mUGNCht{Msbw9kuY{Ht4C$NFXGxIOe zr0;fmH1%h?nBOVH2PS>CV$--%m}I??(k8a0E@jEII3kDk`=yg|=svopGMUzE?SP%x zLD+=H;Z^=YF|SPir_YoAPkm)+;&o~x>_;boj-+F-we*eYsIo>;XJ$4|feL*cxte=|=R{qQG_=)B(v zOMAQFqULg3@eRPrl}=I~w|=A==E>ucNEi$)UrFoC|Hsvr$Mx`iakpusvDMCf7Qnq#_l#~!kAxf4=$i8Gt$okyR%)|HjJ%4;&GoP6| z_nv$1ojd2A<$dJ7+`zc3V$_bi2YcUZusyR0-kopbX^uR7dvu-QV66fD1j@h%M8IO`$ah`bfWJpE?V#MY_N768UNI!5v~(S+sK?g%;`-rfwy3l!erF3J1BXm zH*ML(@~Nf|DfXZpssSkCM%Cvt+Qne77-9|*b!uLN}*4Zevh9rCn~L2}uc3YbG? zY7S)4(VhF^7QxAMOX9xSEaG%jw^7WkWXzw}hui}D(5Ffba!8~H)ff)#6Es%@Vaa%=b6(Yn^J$jihA~XBsVpaumYkR9 zT$Kvv+%=3-T)&rlS+Iqhq)|ZO%x^5{tS#S;Wx^?hpIS*P0z>Gfhc44oc1F~vWK3gu z_3p8_x-G*5`KP160@oNOP9K$9XpOlB^W8J?YH^S#zqA}T9x=Idf&sn1CdhcJ)*b7K z66Pnl%b&g%?I)7~ndExijo#l1K$x90NqkyJ?qgWK5}WSiwlIKsMPcNp@5p65doEXH@ZW{Xqi&?Il{{4E=jOE*KY5y{Q z&AS}9g{FF+p+zOfX|)B5DasM{2*>>;$n&rfMQpXAQAc}Ga*reontDo{tM}M z3Cr8X+WTtl3h90DXqj3}+t3-!OjA^9XQHs8lO1~BeI=M+sg71|0 zS@BO#dNkd46yA-vRjHyF!_COQ^9nPk{f3Q1E`q1DiN4A1ffCmO-NEe)+pZYrSx$q$ z&(}U?w&x9U^ub1x#$LLItx*d9?!97j%3ITh5?H>+Z*H65_eqei#n0}so{c+0|I@A8 zBpQ&;(OBx%|26sb?Z!0@Eg}WOOCqiJ=}4dQ23P%AZi2pl48+ftCae)%&#Dy-)jNj# zEWs`{c%h>(NBUk9O=wsLr$2{L=o!oMJe(Cpd{n0D4nI2m(2OQnq+@^NL6OY$8uJGs zDX5%lNbAv6DrQLS$&C&mKB@CXHs@*h5dw3qBKmPI;pB z1|pnQiHG^JE0C%s6*v?b}AMF_G+rXqg+Kze_6 z8C_M^Cn+gcJg{jrS1hzf&#P9r zG#1pJ$B4dNDPWpDxjFZxOFduX&&S2M7N>;QyX&!1ONH8En~*xs8KMRW(WXv2@FTAg zHa+j-xtBC)wHLyuUcgiPJKry*t5cu4)%?4cYcmjK%M;*`d=g9kWTBd2ewICM60IoR zho1`vGMpD>>gHC4Yd76^dP=rl9N@SYD;-LEUNd=T4Zf z*@%5#+wh>DATRx;yJs`=!MVSYf8HR$99Np$L_;Cy)|cBv(c8#WrsqkdbCbK0QbPyE zFN~$M&JkqFWQEgGyAY64D>5w$rF#48q`upN>#eYzo3SF4^9s=9+*$%K*ptbP>ZhRV zeGF66dhzY(G)G{d{6n01n1kjn;I9p=bP~N?z86(LzQA_fCUO~ckSr$`vZ%C?)YC%p);%*loM{Qc!;%&sq}urQ>~2Hogfr(<}oa#7SP z@vCTg)g`>Y-j`;3n9*z*A^xcB!FbdaDblqLW77GkM7d}DusKzTgZ^9J-=p#T?&x|) zkNapooij)q!6m#rF31ln6mZm%=>3|xw9!9+c8;7&yVy73(vTsdi#=pXGEIjp4{_8d ztPdR=RSMNd2FTp@O*DNH!(bDc(!?=6$tj}|8((QMU-V;qopbiKlIizM{yg&canToM zA8|V0i0{k4;7OY@-A??-_t!qNE9DMm@lG%1kmsQY>e4)dzAa^XUE|YO{O~1YS*+#i z_pfmL>>2dXAC6Y(ulg1*UZP=cHfm4i{^MC=#2KW&zKMRN64W-U5~_I}{8(*HW8dEu z0aUOmf||M?px)WxJWabM=~VPf|gQ@`G-I)`jMh@47duBfOF+k0P>UTskU4j?+sIqSWn`h$RSEy3qGFzH>C62VW zY@p4lmLwB&i0^k=FImboRiuuMbIJL4Bz3qOlBM!~YTL{5H6LEZDW->TWf|^VV!%DV zz9o(JG=~J)J|m|$JsVhps6u(}$CK@x?cG$)bG94zaH%p^tACR(A2WWrgsX!B1-lv0 zmKf64&ym4xXEK@5gZ}&;z}@xQ$`#Z^a}VZ*a2fZfaJIdx zY0$ZBWGt(MN}M%qV0O>wtb=rIR&P4(cvSRUO&1|m{z$5-!>`dFp>S*+2HIr8HI=<@ zHvAlI#u^??8SU~6jcYybN|_u&&BRVjn7X+ zZ+^~~Q{L5uKYlNir|fx~DMme%g3JBLhtnV{zk_JBkw&$F4N|>N9DXgSI2%%#Q68WhX?AxQI^3T2K+{fvp{m6qj&9(I0hUK1AW4SWsu)KF_`ItQVGhg@lk$2&5y9f%~TVboL zLpRjNl0`RJl4m%&Vp@*Q+krufGSp%;f{rH7rD;98lHzp1|F7NY4h}I}jJRCP=iPky z>US4lRPq|Dr<5bzpXtqprXy&L-oJi`UMiC1ysyx>`50GL43buC@6O-*#`X{WFX>XR zUvFS&IRll>(KwJ%ghkUbuz~4Y|MEqAY-Bl8?ta@qPcL4kNg>bav65gn7vD2Juf>d-tiji(m7+d6oiW9GFke2M`D2Tpe_Sg|rfc2F z#?gcZwP=x(dnVMI*5KllC}cFUYKkPF1)-IlFE}q40)%e6eGkD5cTJ9T7p?`!up^};zi!05;)jz7FUZF-q z^6L;1v*=$RXGEEhVWmBNNLomX=Lvg4+0m6yl98c+rG9j|EtBMmZOF$c13Q{c=%q|B z^*R(wqIDx^NtCe1@46r!<+VL1`SC!~2oij(n+9CQ_gpKoems?QSx#)}#$$*s7kr3U z=P+!|+r7zR^fVgU5=^)J85XXkI?0deM-C5PBR{7V{m#_js_z!uVmhn^KYFq_Om(_k zFHhY|-l0i3o~M8F^HNlfe+M7`Oq|(chg-)4-OH>ViCFA@8kuKr;yUB!p;gmh<~AO# z>4LuO#e}8QYtaUlCsahG!Ga%*cuaR}cf|QpL617MC$n)nX;M&P2kf7oMR<>+XwHp9 z(-RJrdVT1?r^OVqIhuOD2&7r1SMYnHh~lesm|m3m2zd`<@1q9g|18AD<7Ke4Vt%vQ zEQfeTE{;mSha$P~^gCYJO7D!0l1@t=4V$!pTxQ?Fb+0h^1wY1);#X*Yd<~WT6A)@P zA6>@{LfnryEQ~70>Q#c?zt+ZxS|#t>CT3`uuyWRv(E~t!|yQt`O}ZOSSs=FMPuS`JT|qZzGep0^ZGlcSKW%M zeQj85=0^IuCV_&(<7oBTV2U>se0-lRUxv`OV&uHt5BJ+~&^!O%TrR9Shqz5_4H_-1 zJ7>oT`m}78Z?SV^7@6v9A`>SjSB%Nx`%pSUnXK+_~-izmeY2sN?&(%x$`%+;n z$HPx!?Dx4TI_Y^;WYI7N?M+)CH^Kr-^M$##)+itMhxZ`MVkNShD@C80ggt-h$ImqK zyahKa+KBU*D$C8Y+Cv$`R**L0QKq0sx1vEy8SYf|Az^=4$Mz%roOCI6zBKiitwv+} zmcZ|30AzEmaI>HvxlJ8L9w83}x>#(8c}7o~-DeM-ugImXNB5A|_g*|domYorXwhUA z+c6Sd{ygb7THg&z9RKk1M03VC*z^|Wx5>jOzOF90voKGs7P_udbZ&wK?F)a3MS3gv zduJ23p()`7^3@(<@S8M5XRbzJdKoku+A#9029=x$*XsR-!q}AbHm7V%uv2PG5y1P-eWPT zo#~))PfFO{pOiK;ebU=6{E}W<--Z=j%RwpNeAhnjx(K@RZZvfyIerm90xd^Sg(Y|FW9A!`7 zUQb{+@au!P$kSfj)gjWH%3td8a_B? zBjj=#c7+Ri{mFe#;NIvn;I@}wL_W)JB^{1W4=b?GsukMLUZFOx9X5&Y;IM8#W?fXE z%TZ$}B;Jj_DKotGGIPq9t3;+3s}TCJ2SJPJyXN+!>iR~8&08rItzFBHbA2nr+^`)@ zUn_;V;%M)Ladsituzx>{-=D)ahI{w?FvpkuV;hHtaSgaS@jCh>Jmas~w3Xt3eOKD) zJ%)OjTa(<)zO*d!GW^=lVJh3x=KN~MTGK9+e*6{l=Z!?y9^a5IU4_T}S}}TkAENL- zd_9k|@)7Y&ohI~fAq{DNx|B4D(sHF~PMkW4mQ5q|i@`Map))NS+J{V=3$dz{kQKQO z#%8CWa`!am{y2yp%(j0?YZkT)KY+x#7f8??K%2DHXnpS-Tyv_%lSK^-(?q;=I=>(A=sTWuSnuf6U#3k zy9mV>(qO#v0><{+h+cQ3>GvZq3LLtTBnAZ0QcDM(cj<=*V5cVgPL5`{$UaN4&~Ycu z*^Px_cnYHKv3Q@ZD!)#q@-2zJOF31Hoyrm9Z1MSEzlWj~@CE<%%E3E(kX=jKHzQV;tA zg{AqJ^vjsQ35 zwGdk?ZtIW#fA4HZ$`Stlqnyq(@18lm?QcNm#!8V#WFPusGl?wE3Vs+L-)iu5=DTFm zv%9L?ut5X3(;Duax1$&LGE0r0KdxglG4|dWSkDr8^+(lHLEPT^^|4Im&quF0HB2W~ zh^$>Mn0Lz$9(mGa=cP`RX+vw?xbXGQO^U~aV@>evXn^hW&Lp=;oh;Oj;T`iw&ATv| zx-AB+m+eP-Z@SX~rb{-|&te7}1u~jsKqFrbA@!buo+?4TRkCfmIgLFhO;Xvn;n|5{ zT1sc3Pu4oVzbVXKuIZbBzB8lIseBT4q!(Z;!&&L{v^!0cX8DDGo+SNrXZbN(g3;IW z2o%RY#F4ME%x3ouE4RjB?jkotneV~ggO^~cE=vWV3fPt3+ykf!^-#m zuzfiXyIW77qqqT~DR(i$y#kH5#_`YnrS-{KZSqw)&0mYwae`|r!{E2QhlI+zD5|`L zV;SqPp?4TwOZ~)^mqv7do<7xkDbpv%Gc4Cz6i%-_kM}(j5!kDkuS0h582+fjd~%gm zo;j!xiWX}uwmb^I` z81SVT!;PeAs$6I4nEwM_t7@>|TpdPS>c@Oyd(r9BLVQk*(SLSnF;BnB|3OUTn}0l( z9AHA{e;a8j>Ha>HD(2c8YSh!DQCnTQ}vH$ zCswIZp|mIMiHo9zd0up1S>T(v?swUHF@MHmtTX=ao$DA8f$h0b*qh4asVS#0?`t(u zE3EnY*6mOw?_5JNcsh(+`}Za9#Tf{jq(E1chSTI}lc}WCmCDTY>Egl~1ZB4(cJNTj zvG$|u`?pentToNvDj_ksEfH_ryOZfuRnjf|jJ1kF+rJ&Vky7z++JC<*bvNuxgV#QX zjN^99k64FyDz4bx8V&7&*=V#D)*X+&DX_{f$GKVZv}{0MQvUG(J-v2ffK(K=EdPvG zdR@piREo^?9^#d(Md$TLLy;X&4&Q`qzAs%C?!nai7W{c`b}kx=-s6cWQ|H@m`%0hdd zIs9iVfUc2Xe~H)~EqXNkrRe3zHzE&(TG4=IDlB)2y{PbDGw!KRAX6=CDi8M^yz|6c zyq9#`i!k88g%h10%o%H24r!G&Ly_$P-P zV|KpkPi&p+*TmB^`E?K65KVyU#VF<%=79H^ZfIGc&ELDfU4uS2^r1QBZ;`ZKSpRLe z8=}YZjp$b+@cqL$c{y?a3oZ{=5vU3wwtk1=xQj$TV9#scD#Bpawd@qc}Pquf7zb5z;=({=KdyQxG->2XVO`J9s>JM$GUViZgx0t!+ze*2*i1pzjNkSFTSC7=V)RAaay$Hw zXUpb|1q~5wQW>EKL}6cj(Qo<#^cyKlmrwMddEMXRnCe9de@{VYB<63Zmf%zTm*=_* zS3xai3+#veHQU=^ANKo%b)zz83_e{tj7ekU>4=^!J-E4<47$3I$|4!c zu!{!^V!+PcHJHEdJWh-iWY|g5X7aR-?-vNE`+_VaIcF;KITg{P5y5mg+kCofqOC%SEe$I%!Z(4NooY6MSf&9S4YB1Mj@ zuP`04{Xtw9Vn*{%CetyQL*$Tqd`L zKH`&kKI9!{;rFHe7}+-)LrivJ`GUi^@lx=`d_7d~n_RDF2-(rg|IzOqF1%N=FOT5L zyfesQYvP+v4fH5XlQa7EgXZsVr+E{Lc>3G*ZeTl$%l)V^nrM-Tj?5L-`EB{9@F87| zbnYpPW@VcsN?-KvfH%oR2!!Hwb_0MJvquEDCkw&&Oo!S@6ml-;09Mbpf#_?Gp z=;Ij$sW+PZ^^#*9nB(1u$J3vp`(rQu+SB0gzyHKCPx&=676KrcGyV$kK(R53mm$M)&3pq>6BTN_0(tv(a)|a66SQ6&xg~yew+@ z63p{wTFG{XkM#t}gPFf~`)3%L2|7C`iGh3@wOj^%UWsv$`-xRcngDeQHDFr%p6$K8y3(5D$}9O|m{Wi9W0!PkK*A(WCLTbYp58IcTzt38_z)##3I4S!EHAPDL8as0Vvm^**0 z5BMQkyKfVGUQfgMBa<;xvlo1Q1)PryE(|Mn+5#$bcBGAON74GGJUqSqP&7xYY5frYhgm=*eEN(Q!xUPG!p-GZ1amHlatW4z_jmY!1 zEH6tG4QND_whTF0eMa?I398s4>_tyXX;6SsU$V8ENa-tvlCR?pG)o%s^kz?~!Myr* z{#@MG%-mQ!Sbtz}M(6XGBh>)a0-+A^Jx`0F_%d@UeEpLc*7P`hEov04kL!u6nrZmb zGaTb&(^yWb(=b;W&+^_0a?X(FxA1er9jFh^fz9zi()5S~NzH{OW}FvF8H~ZvO>v zjl;)t`L~91fj2dH+3I>bP<(zmmNMMa37aFJyL90{URy=mV)$-fWc1$zt4s5+AfYF2 z$H$`6%~yEVE=Q3e%A}b74drjQBR4%&RJd0Yk}RH}+He)Ba*`pNG!u(%X_IG(BdH&p zOItHX(W?eQe-Jk%QZ(*i5=J%%`%H1)xOh8EXY%k>#biuUnv9EQj$-%3{g_=;D3aSI ztmQ5rj=|gEDh!>^W3rhm9-nwY%_sVCXVivqo`8yH?;9Y!l^F2Z_S zD5}MSynMWBQ>9^zk`y`P>(FY_!t(4JJghZJ$%68x1J-iM4f_|4@xDo z7q&2;Vy9ebe0Gn*SRHS znYf&r&Oc{3aT6{D&p>-}D27~)hnw439GYoJPfK2)$VeA8^9b`!`og$90UnAr@C)13 zuNOR^_hT>0`$XflOCF-8yJPPZL%5x}Ec)Q_9>yQ_DZs>mo<=%T^Nmh)>Q)I>anG=9 zbq99)cA>0WJNdTH%<%8V^U5F`m!GHc*S70UWqEJIF}6aU=G^W_{mh5bR{~wye*@RL z4Z)a(jo?@eL__i>^qoBlo?R}90_LlW9tKUtz-jAopnMxXIeB9E2l{%xERMnO}7mp=kH**#RW`nPJy+C88q)Nq2(UeNbAdM(zVp*BE5!i z5;Nb?utG&#{Nc>>966}C9SgfIJ5e2|%G2weu?B<7H^Z#cAhn;vLoi*rNKI<3CPo+3 zqws|q9b&m?Y7b7P3!MbL@_m<5xLx>w0DDc^^>5Y?`6AF9eKpZl@#$s7nD8>x2#b#4Hp z$d#v2rb!!3rzAC}?R|ul_w4)NHk(ps+R@z2@yH$d5Xu#e*fI14R5OmF^!H*&*W03x zehy;am*Z_*Ga}9~xvrnEUyi%055pT4XpozNm!lVAT(K12hM3;fu|wd(WX@67I`h{? z#QVXy(|K$N%SCFp(+E_5im;{ga9-ssWz?#0$=&t1O#|gQ1;2yTa}0|$Wp!AOZb3?- zpii`~VZI~mT1ROz_Fw*vKQ|lTt|>*!b#LQdh(C(H34Ur?vePj-X(dc%+arCHCgR;R zq#xy-rcc+elH80uYF(8;%0^qoqU69kqUW0~icVb7>suXrOcZ@}FqXP$ zVN|`^zy92RsD~pCBVjjq8_O3Kf%Q+fL$AdOS9S~XvV`>6+T$#V}f* zO=LFx8(Xn)`zV_1xQfhr&!&JGK{R^nR$3n@;Ax2Yp{<*Wb+#3-KOn4^Z@LTPynNY7 z$i4Ybb~IU6jJ##l2)J|w`#m%9A$C7cXO&7B)~u1Fn!Fwqxv>Ymy4IB#cFHy6FWo>*j@QWZ^HTV)u>??0<_$gpx3TBTasHaa_7IlSt`V&}U!rvFQ?{oTVtXR`Kf>_2zj0*U zA9U*b2=Ub~;qa^on^?eqW=$0m5*|QZL4roONKx*F7}WT><3sUY-1rcO6#2avYP=Du zUDXgZd&(R{KB25anPvzs*ie^5< z@JVuX;UUYLSJ;5?F@hYr%wHO#SgxGHvoA%PCaFQiS;$*2o}UJtd*kMJj;B%lf1bYv zf3DJLEY?jMjxBd3`2TZL|6BXkwDu-VEnz>U6cB~01(kT?#KwTl#V%crB08{ER$SNJ z$K{yxGYxODl9+CA>%TUp9q7g4UwlZkK8*QJM$w8i5#M*im~l#Z&xAdo2E&(3mEOgF z2j{4DgPfWO=g630R8Rd^kGPHh|93Ap0TSO!a8&yOy7VlCY481b`;g!gkdN=JPcXOq zA%-2kiN$furs{kOH40i>+_>4C*@Ma4uYusYhL7R8c}j4Da>gS2##Q*Fd0UkL!U))9o5CA}*gD$R#uda4Twda_P27+^$92x#3F$&3B?b3p?Y~c|}YbZ;aU$hH&zIBYM_(7a~XBfNb+K z>^b)lKj#ViIFD&Y8qMu4lGB~^=;9Vf`k=g)&SdwYw+u^Me|%TE^G1VyFxlj2xD|cc zE9ik%Sqwo@xiMRp|I-mo%`Ab*x4V!|xs1W4r(ioY5hmWfkRp+cPXlV8z3(P$zQ031 z89BOLCA>?2$NZMh0hGE4vXS`Oi-FZ}%x=Vpx?8BJsf665a8P)P$bR5)(XEx0BKla# zaHWrl3Qvy}o$MWq$EgYEFIRvkPQ~~ZC#=PpbGxIYIRqhl?_i_GYm=q6VdTN`W_a2S zfmN4fXs!zXpD_`)o0GZ(g~yhlW^O2dueY8Se>~tPtOwbiOZopPg~7NP7J*$qTd+j! zG;HP1%gb=3UV7V!39gg~54m9u_=%gm#AM;FqjIpSG)0 zJHw*wS=AXAl1&iw#Q?K(g+1lt(eFiT<+hmbRh})1*OULBccX4xBDy{K!@hq^Z-2;~ zK9!o&csW5QyJ}83UAl9b?tDK(e#;J!4)dAYeLIi(AL+z7uh3+6b7QVW%Z%T5%#cxcLmW6_iy@V_m>)Wcqgzgw{-B85R` zm9LuUVE+?41eOTC58hU$h;1~*!cR+3-*+eOs>GoDM<2{#*pWw<7GlUNNr}EvhZufX z6|@-kQeEd_<~!Lz)id8xq1sbA5zhxwH}ChPK~$J7_=o*1 z%h%IoJ4`NI#?lFbzD<1X;%qr=Ti}aMRu5qj)&ev&z?9`UnmT$fPv^xg&#;*BRa~~j z?Fdi9hzr@cG3FU$k~&euglkCLH-o>QxaS1c-7%t$>P@ueR0yfq8&GRu6`b2_>FI;5 zw3)@_b?kMaowh74VZ$Ku%3e)74hPeb&kWPZK)~zh(ahqce^}G)=5SgXnZ~Su$4Q^% zNZvWKMRe1C9q#Wwi=K`ObWmw9oi^Cax2saZ-l+r43?oGFIp~rm=;cmZZ$vlOdN>Yb zaeV%ov|-RS1l6*p%l;M6E0LJTHGqZ2EoxrB{2+@T+yTqn($95?GHolx)1KlhR4`98p~JcroX zk+oZY{u%orexignFMN30jI7k&6uIXSUshaSpXP-aD|rg8ziR(o6JP)4(GFb`eezt% zuyE^ywYWayFt&{DN;h>~DVW<#uQDWQPhT%&+-ZlyCONuZI)U7_2zv-|9i7RXG*@h; z>puHvWl|=s3+hY$d;CQ%E&-^!(1ADo-Ki$hm(~^ud-gE5rKnwAhVMp-q_*x0wnVN( z^v5m;@%#fjbALJ+ew13ydXK3FvFw+*fgwLXa!v8Sjvd)jd z#_V&*_+Eoa#^*5e+j?|6bVRi8tToE6Mxm;7_y6!LUTOtm{Fh_xErDMGW?g~CiQ<3H44LpobY!3k-e%|Eu(m8MPC1Kr-FET!C1q~F?_@1jt`^o|g#&e> zDa=>$B*RT^mdT|4&%V<7J!Svq&lm4;s2Gq6jr3#ii4fMTg{kXB_ey#}DsnHbciw}g zyWP+h7j6DdYl6s8M*bfiw|!oU*5;hXSSbm*qQ!Cx{gkC|=a2ID$4@hc64UjhJ3U4j z!@N{}GKO!n=G8|U(`g{r*}{`s;qA$7+cA-|%`K)*i-djIxfcurz(Jo5T3ONO+roO& zg|3KZevVK*x?>pH+FVd0>i&-gZw2OCC^L^b^;q_=PbF3NMBj4kk+ozlT4eHIZ61gd zbuW?iZarP^eUx6>9HT?O4pOXSZ~hr~19dF(%0@x(Gb{=_%-0>WmgW0UAA>Q;{cvoD zKGZ%7Iu6^c)%^XLSK82P>j$)rsz-`z7L2`HNb+$YS8?YkH@u{dTYl^ccRp|iH}*jY zMo*Wf3uP|k);fyV+7K=OGej)XX z6XJVL?XThIqWanU7*iNXyP`Ib55wm95~)O~C&to@!@e}`&{n$8MUxIMA7=h_#3%DZ z?6W+{EV8r?Axbc0?(iKMU$ep^Yu%+@5PJ(CNx8u z=t}?H*qQ$W&jbI!!c>~nyIjDh;>pMgX5%%g4naEdv}0l)+Ize=J(vB>*MD&6UW|KI z5A&PVm^8H(SE^j`aFZfto!W-w_GQ>0@d>)IiY#wd686sW=bx#O>Vuk9y6|!CMp|rr z{WY{f6q^|W*M5mC|K=V@_3l`4@&`v^fQ>v#suSTc=9Eo~5WS~?b8@0TKWKrV9Mr!g5#kU`GNtmgT4$VrN? z<3z+OeQ`bFAE#`JMZt|wJUM{g4zi|IEIz@-e+wC|ilz}-arDv9lTv5=_bnGcv$wkn z_oitv*D(*A(c>^t-lh&wRM85!X&1t(;JpQ6;Dgv z)D)ag=|X|VL6lZx$#5*>Osvo^z1DPAK6h``wTkzV>NYoGv(iVB&Oxg z+2{ZG^0!ao`M>vu^h4)2VwkUKe!x_4P zsoRIrn{tNz^=vo!MxN*CTxMH|y7o@gw~I7|x(j>trN5#OyeJt}?z#WdcG-v^6y0b3 zy>7qo=uji#S8s;%>k!dd$v2|jlOx%B^8m+IHlzEgSE7&hdFboC2XUK)Z~b4|d%7N` zyj&$NZm}(Aa?X;QSoVWT29Ke`b4!q<*p(DUZl()iLF8@k$@Age;UXBjccJ0SMfAmP zAbq{{2q%XXV!cYByYG5j_*1N^nJ(^)-v_;_zlUcg$ zlBDiUAJC6slbwsWi1Ekbv9vNC;dRw$8~GgDzF*+mNbZ)7q@UL@|Bfm-4eUp!_M842 z+xIaHr({()tu{;~TbT_cG0%XG`Sqacj|tR!Ln9r_c}Z>~Iw;WWHx;EGqf)jnHSsoQ z*rI)C?#7XHEQF(w$R0GY(`*!fVmVuG$MbEj`&^19Yk!RN+08KTf5V^qj8^N%kcqvG z9V zLGk!y-O!V_XAt>9raOYuJ0*$v?mE5>uqdb2sN zw9#CB)>HCewn#3>oUW-_P{Q3|lzqXBrb#`;NP8*n%oHOo+k)fH``U5^nZ{gvXBmEP z)6#Yr2mFsA_47$|KJ$Oqwjytef4rS~OxVKHbiKuauC1{lwYn)>@PX~z9LHUp*_*9g zMe=O!V8#&6u1b%jUyP+}*I?|JdX8bd zmqGVk0lb-yn@{E(+;rLkTH8J1SJt`13mcyW1&$DZH_CZ*-b1Tv0MOO#%joU zY zShs%U$>O}!pUTj)p>?p5d5VVZm-u=j-g{DTM+z-Ca)r!nuJiQ&NIN46E59ikWGRm% zD_!inCyk(;mqpPrgVVL)v9ww=heTxR2p7?YAUV zi-}ZyQ-$_qC{X{G?0p)gMO6v?Xl%*|Qk!B%D{JrabqzQg$-VZ=<+2hEaBdFa+*WH} z?z6cor_Xev%lFAsgi2@X_L}W^+8OT2`TMAz`(F(IHI^^9y=Rz|n3h_}=kORHLr+bC6lVO^&5p$hma4Fmu7kBl6_v=BlX2>?GF^-}F7r__)Z=Xip zSb@D`*jV|dW7X?q)OYP~B)&Ik&19Zer%e@TxQSpJzF}faa<}cM(XS8PW3h$B1}!K| ztV3eXMs8Dd2A6TLn0wn;#7+Ic{M)Cj=AL~uBZGG<$;>!_cGr)fO`!%fP@@_}eVx(9 z@;iE+NrFexaXg%T3ZW8H5%}aF4$W)F_{06_)xIg@)O!R^!$OwhX~qGT&tv6bl0p^T zT>P9yEnG<#(sHO$^BkQXQ9$0?POyAim#L-l5d9e(g@m^ISbXsU#@7kD{41_Oa9ed3 z#iwQHpsgvXZ<$F^o=SAm;wI0x6GwC?mf=1hS;^!ZJLaz_@dWow8j;|nPZe%mX_>wp z`B!wKNz+flg@#kH%vjQ4@wX3GYtXTkohe~rCsMZE1jjaS^jjGXn;7BU*D2qM>}e|G zt7pge#Vun3g{BUq^CvnG&GMtRvu~@!OgVa1YfhD>Srk=aN2lH`0CEZtu3wDHiWi_` z6ss(^GXhJVT*U3@cAQ?y*4UY?FsPTIJ9QVay&?jB-}a){_HyJi`@`DxgGljdCiE6d zk?G>T)S75S-sk&L%%(m(4Q4wdA#vjf-afy9C0}nKem~2@z5XQbDQZ%?mleHDv!=Gd z74%(xC5`W_NTXaQpxci`*jw&K_|;e(8!zk!{>~xsu`Wr-6(&Bvy>~L5>3otBt!`3g z@m*SEQbKk}dBHjEX(s{g=c81QUSK%rONgjH*pMt@l z3ce51nGNRt`Y;S~euBOyfAZJHPMd>28FS&~7>qlelKE@rnf>W6@2iTn@q6z){{PMk zd7|<4M?@pmKEcFv9g-{=L|OB6XvHvH8jyDjk6Nuc`NvM&^maQg0IFQfm>yh1l>*m% zB8#SnR?z(HYcyDUAG0T|AuG#qH1{>b`+mKV>OO6tZbdE>JasThyR)2q{(>IK%7*!y z-MWBvvO%#b-0UTD=5qNB$ZYjAcN7xYFcYzu$$$;)neb!eoWt} zc6JEEGG7Fv>;hEoxrEUs8z3cbkMbibd|&+5grWX|g`_9*q5Rv|-eEB~xhw^WD}te% zoqzz_EZD?5Vua>+gnm7U#F{KvH0?vVD4Kuf%ez-lnx;;%dWMv%NMtKHfU4TQp(x=i z^tuRk;{#@c>B@*cv}f~6Y+~^)k~2IgbP>#_)B*8ZnwwUOS8xvkIO_Ui(!+SH{DwzNmGfk)!Zm$@p^h;>Q03LEYE1_J3N0=Kyw>T(N&)W zT3vXGl+IRCf_^Mr410*zPu^mqQ$7M`9zk{x!`FP8j@bkL+c!L)UI4Z5JUlxe3y;7q z-1Zx0T;gv_ZbZjaZs9t6?!|)+n&zQLWffZ~Qhy{B8Jp6DObZ$$sYA6=W;`!;j~)-i zuI11F@=|=9Ue}80dW{IoI19sNtNz`867(H|Oxt15s}-M~3VVHVdBd8~|H_E}M>3r1 z57TTPqdr^()f?@g@D+n8rq!FjH^QCzgDwREdBkRLxg-EDZwaA=pU-v+Z^ zVHo~lDxRMihSSr9+%d6-4~UMJ6^gDn3Yb_{l3i(Pz#BL;Rq<_!=|RjR{v7>F(91`V zW$c{e5WUlxzwVwOBGIp5)MYMhjeI8p&vcH zwT|pZ`;ct;PO6r%C66*6PH)5|PTD)2+m?M+cqY!g^Aq-U^Cd8c`LDLQbZ@Wg|KERY zzfT%va!Ge)3F+JZJMVap&J$(Y$9CJXt3QKrDIp8;{?*uj^f5Bo8eGF&o}XkpmG6Js zh0h^3{>-v5ywrXfVZGpyathE0z{@8o7&<5lVaKg8<+3!64;}&S!&5QM^$&NpZ*TN2 zi{@tawg7+g2xk5c!lk;Ym>s)VsH>%&0=CzfU~KQD*zji$4A$5S$9o!lftp1hvJLu) z;3qY>G(7`;gW`m98~mjiZ<7Ud%8lFG`yx363xbZaxTr|%JaY#x8J2IS^iI6me+4Ro z`cqSK7n&+5O*NHCh>MMbd5#I69qhti__>7Vvb}`%WN%XCzlNm4ukHhK-+jlKte?>P z^a?klj=^QRDy%PS6eu989-qrgVmHAj>5r{r&|vQuZD<95R@{vagWW z=D=od&rf5R_|1onnWJz_`iG6M-&)>@7G-{h^w|oG&0=%mQ^cq>bU95{HYw>>R*@(Wm@w20_DJ=kA$D;A6 z^#?*zUt?LJFIIUKQF)#=pFUYb7~{wLRuN_fP?)JQd}{SD?Un`9d-x;itVMfYBzp(q zXphYZbAFAH6DE;$SJU?Qgl$)GuGrRRZQFUz6g*_yUM&^T=p>)abUNPRL---AlePwI zjw0@G5biD!*U!CHtViD^l8{`}0r#6^(42EzxF7bH)^HKUC$OXXH3H-+F{JZjyyTw4 z?n45SH^))S$={US+Lt#^Q{>NVXB2pk1PWBYk zEE#v&Gi_Qhoy9imcOQUgTMNN8`CTgx6uh)kEN67vF8!$_O^`n zJLgg3!kskPXPMC6hEZuKHoFd&i!648&Fk;4io&-3?_tQ$YA^bj5DOa+(v`<%LBEJY z`qY!S5cwW^)y4NzTmKx?YT)F^2ThtATWP~0TW6~2F&PvJx6GfjF|pVR0W?Tw8{D#SM%%A)j1Fcj9%<%;mdYTf-+O zn)BK}-qXba=TH*#5xb7|rr*yO(!Dub=xokEdsW(X8T7cZ4{P^{SoN+Yb5R=e@-*busS8N9F>|NMKNw=+hhGd2;XHOOwf1SIw3@Fp%`}qg6Bkg!#;07D`;A<9m;|lQv!{;{ zBdCkt8KKPilV(C;@&ODwF&ZYhX0S2|;#w9Qz~!`W7GUw9je1LIgNrwr`psiFkS?@Nw?E#lVtnT3WGK4Wf7EgJ?`G zvp+Y#!jb(8QDVMGC~v!FH?qmvO6oHQk!N)srh2PkV_Xw*^1IM>6=tJjaU{Q@{UN_6 z7Y%As)Y>$JY9fLuD(?Zs7agQbiMyQLyT9Dh3R#q-bQbdb?>QN=8AK}60uhSwL@P5-6u*-MZ9Oc_pZ26UjR z(tU6rcZ)_|8prP~2;i5eujFU$9>H&zrpO0oiff#<>jt|<;N^-f?RENoXBjr~mk>GW zEc6E3;7F|@cJ1AZ-JSm6CewNds(A#>XZx{U!xww@h`JTmo2J5gRvKo>oj^Um1kvgB zSny*YlUH3!2b@;Y!m^%p@nJ0b%WB~Jk#LOpkP4|UMX0uCadNN!tqbBtA7nZ(jIV{w zdHi}Zj9+#v9KODXW3eVl_Uug;oo-@j;~9ju&gQR=NaMR+xy!#ay2LLW63g#NaOCCV z;?erFhGCmVQt>`7x^ihcjeOUNk%=!jjjto&Z?GHc*U#ehwQU%$G8m7hf93id4iVaZ zD_@6f-u0(%{jQ;SjP0PI=MSP=oiS2X`qTZTSrp#Mmwq0L$9d+z8Zz@N)8INwr);9h zCV4A`U4P5{ZffL$zm34z0ETVX>4Vb-d0a=&?HJ*4961%gVU;i1wc7gp@q$RkJGx#d zw|n<~B=z2uCNa)6>*iHRZM=XgQQhWoYxgqq-aXhD(Ayg zcX0&mll#V9f1nE2Zj{7j<^+@#5wUvA8Ln-w9 z3`yEMNM|zR`B`GCOFt&sBR=^)MwNcR+9_X9J5%&uZ<;*_vN((#Pj=(&-EfRl%!7L; zHV4TlM7I+I=y0qD^%}SgPdPM%>PGlxyYp;Ln^tO7o?m0?%9nXZ#Kn8OUeyco}14DrBH#F-gKjzes{7CM+u#*UG$n4T~WlDaN9c&#TY zET`k$yWUvg9fza!cGPig5GB+}lgYRP$bQg_WA@54C5dQ$*AcY5{GM>X22F_;_K^-^ z8`|m8B42wO{KGwy-_?r8?%qsVPIHK>iW3^Rf`yA z>XjEx)rMf|(oabG+lq$`^+MV6l+iK72$ODi zMmOi%T*duZ?&pVi%<6FikJKb28W}%sd8r{8*@-qXi(p+o&(fOj7Nf)84lJU=)-qba zG!s`koX78}Cux2c#sL{@OXhn!P#*KyS6g#5=!88Psfza3#IaZK(R&J(ei@Gy-9oS_ z-y6*djyQQk#7?|M0(CBH`XxmnX8(dY5=d%Xmu6Z8QgX~1U z{hDx6u@5zG2xjrfY>pfzMOqvC(-ni>wD+zyRSrtPSFRuFXbq*{R#z(79ZPY#jldSv`P`T_aH_PJ` zY`=cN#s(Jxca!Q4JJ6jRBc^dF!}|^2&#ze$!;E0Iwo#3tki;L zNA;omea|9Q@e~zGF5@>HI?5-Pgz=uY?D;Ql$J+11aJNug=u_K1{;$olr=6Z1`Q^Xm z_`P2%Xw-}oG@s#wx=VHB_s9hCdv2ZJXHALZlZQv}Uv69{n+y?eQoonQ5}Qn;+pQsF zcTx1yCYuFfIJ4Vm29Kg~QRC(M>#jB6ZAD+CFniFPF+4eznJ5`)R$ga*pFQ*A~ z(Zz^jdv>9_{S)veEgsV@xpiz@bea@8U#2XBbTaavLXRKSvACy&RM&e3jni_Z+2Nzf zxHJOx9%>R#SL7jm%3_jAw4<8`iPU-(3-?E|uO?1hyNi{#x{?pmS*TvOPdHwvD$Ysb zpS1}8ch-Lp-WzRg^Y}B8rpOE;lg#cUVbrU4TBmnhwp9j;srR9|&TDALacM%sDR5f( z`0z@KzOZj`*x|c~Um1d1e-Cgmm&SH!VeuHpX9m+=8EM{pwhpiIBad#Jt>msGj>U8p zADq2&2He?;%=YDoGp@~?Ph&rfu2>+YucAO}@2Su%xkRB}dNu~=r?OP|{a?QaFdt9$ zrM2Yua~_=-HnV+PF011ihUFWa9+*;-bq88B`zs`@7*}m0$3N;7$Di#T%#Xeq%o}Lf z^4$k`F+I}V7_xg14SJv>mnS&Ga9-sty?@KfC%Q#WQPJ)by{ ziXWQNrc1+w>#PbdbE#P-jO#i8PkKJZ+p&+~_0WrPcm$I5A1S_fxH2EQ?3a*MfpK}O z*&g6NEgHvsq$M!oZngB;| z&gPTa$KJ+y8nn%c@dfQ?{P@+PjdF80eO%w8N(NI_x3{-xLO7;<`HrUZ6RCX7W(tZB z$4RTrUW|9GLIbm3sST8*CraA1%U|Sac-g8-M^Zwm(Eb)lyWAyt6+fz#bOU$)A<{e4 z;%eCsc${K5&*P#`%ID1%E;maaoi$Hs^Hw*wpJ}07|1GC6ZbS`+1%^|BWC8VlyhEtd z;5(B^UGDApsgOhtmv_?e>;JgC-Q+%D_GRY(T``!p zdo~H>sTcI3qZ=mDUwdP!IjBJCYsL9b#Vc1VJ$8+0D>2QaY0VhWGaswBTMO5GzVsGv z=Q+~wn|WmUu$A7cB+%!dUhtZ4O!5s|skP2YsGnl-zwb%gSW(qpYLw4Cu3CMVC=l4W|{OEtfubn7>`>i$MwhYrG6AKRsd>67e3x^fO) z6=>1M5Gmr;>+uWH19?@2x%{*7qj~+n7ed*~m&uX6&tX!&6GU%stfwT8;q+{@2`;)x zP|l2bwB2eXHO{amm!Zl+`nLL&+^N9Jf&FQgaVJV2_8y^+r^2`9BF+7+&M$N6%U{^k zLb~;*$>hNWp&f14)Lde=>)Q{}!)-T0D@Dx#1^ve;9;YPa8~L&T3$M#FZGk!1y)qx` z4PWB!(7kw4D$d!L&G_dlR0^4k=bgnl_iWQyn8j>=-q+7CT*yicA0yhv)4%tJ$)c{T z?^lv-pS{#jyO81z-o#9UCs527KpP8O=)jqO_LN(dL>9F%v?=l^4cH{E3n;$`ZZUTd z_j?^VOEQ}T)8lyV`V~iI>>y{LfazbC;I{EfB=-)$03Cle-_nKpaWfoS6%VVoFQ95v zg3|aM(1=|xeAj$t*pc=}bDICqm7a|-p=av1F!XXgrqvL&%o#`*^xvVc_dD*=F-vaW z!9;Z1B}Y=-%*o~QXtHGM?7!-jTxEP8ta@+{PBD^XUh{7*Fr;KK4eDM_>FfLReOD8I z^HFa;qTw7_?}^4frp0)yup8}b5!2t&E$$y=nM6ZnEO>>i`MhTHNWMTei>j722={B0 zl`P!npT?0+k~FzmgU*j<@dNoRKF&Xl%3ozs@Z7Cr@lr)%)_6laHmacB`)&Bae(QMi zplQN$dv>sh18uwSZJS5wIO^}amX5s|LUqSPU9C~~Lt(t{pY7?tw1&rFoSr)d z%C1eE>f8vto!p8WwfZ!3pd6{^zJUC7C83^6T%IF(QU|(`*oo;Ki#{=)#bW#BY0#{c zMl2tni_$zED}G48&UYeBKKXzWQm)XoYQ{zXxIf8vn%O=+moQF#`(5Ag!r&!*eZp{U zvOGeu6&5Kck(TvG9wndp1=P4GbSlpk1rwFM`}o)hv=IU37J9A@~CM^B-W){nk#d4*{Q znmK261N5K12lF)3v6}HHo%psCb0$3DbPvd)F61-GANF@W~$zt~@`Y>TJ4gS88 z8=@%#Q=Ng3Ij$?@pI0vW20Gm`#)^$@n3b>(LtnQ$I$dNxwUFpV^-(Hi&k}FFrASFLs`UQd6eQb*2+Nk$-{lM+z}t*9eaSyYLSe ztl_(lUc&c$*N<<>I7IUy`i$O>YJi;S8~kn&ao+!pwS9YMNX*ytz-_M}=xQn80rO)j zEt`$i<1V4*YzzE6U*fycd7(ag<*d2qU8QmB_%`UW`LT3T7>w3y;*4c?@+hz)H=`ZY z01r~F>e&9CELiV=9~KetNeV|P1)+KPFrX#T=+DXET*gFVLoU%W zFFA5w`3=hQJ0We^7d0nU`S-R_{FjB<{8_0mUT-$>3rFP&Wo%$`k5F4f8lbE~Uq+U- z=l4*?0-EyiDAxXfiRlvbbf!BFMO_l!K_)#q(}1uM zdP_6~xD6&>77JOog8BEyETV#^bAu-Dpo#=ONYn!*~YOY|LJ)ZTukJa=0kjo`7o*)T|WBR1W3@UzgV2)@utqg>}!O$-hurUxKX*u4Dh$ zo>=&p=`#FE0PWe1(Q_{f@1gH65|A7HiaWV>9`vF^@iS(Hkk5C)dWr>4nJz9HLy5leA{+tPY@cVZE zZQINIFP-kVW;~U$qD{rVl;X_d>=(X<%yVO|B;N<}aW649T%Gp6>_g{f=739MaeXHo zsCEmJN;DGJfo=IXOKLD zS1FR3C1^)sGwP*sFfUiUkA8dDJU+M+bzY!PkNZFRpZ0b7@erf7h zP}LOkRYX5ZTN|!dUBvFs7jP@=QhT~9$}h=c;y^xZS0_He>@+1*i?)di%*NGys~^p~ z?N3|3+0yo@;u=MJ&Jz^9wWedAjL9NZmlpREZD3ZXcQLJRHF~ZxgH%swkx^z8jz(s3 zZ>}*-ncy;Vb3IRo@`7lEzW6O4*k!L!2iZrRNzy(`_)Urq!q|~Z?m>#C18CyrY-H~E z!98&&uHHzMY*qS@wcC4)$#HD2KecI9jAiS>2Qe5cGm+Vl#h8~%7G==1k){flBvg;K zh4-{#uKIccH}bPCO(|GHQ56w1z}1VzHYw7v>gDKtNSBfaY^CULYIIC*EBZJt$F_A7 zv2O7xQkJNs<$qI2o{iC8cf_^X_qj&Y{HH4`?;?5>(C{p97h&i z!(`Jwa8-HCB~CKp#uXmJf_XJ4zkE@s`;pD=c(7QBafl41O;wYHXW?fhXnBAteHs1% zJ+3*BCBuP#u&c+q(hxitzX@GUMISRUd`s?3dzxY`+8^ZWA8@tW5)w3Co}{x(NPdDm z{k+(RT0&1CYpNn&QR~K+t#RWiU5Q^pzVy(WaTmXO4*N-xG}h)ePE|+1xvC#7rzdb1 zT$b|bLyqu2Jf`!>DGzA&MMwHLEFSY5U8Uq6GhM*x8uTu@0D7D486Gi`(>-{Pj*jlY z^Vj2O?3TZ9csUR1e%{=JIS#y(PYi$Yk}V(5KUHYo2Wd{&FWR$!5+&!;tC0?LNPRe6 z_f136pFuC7V=oT3beg zovmovwMAHaZMATJ@)^(DsmsGD$i$QyPhEhL^kl5O>5QG5RcKm@3H`p@lVKQ-!g`}D zM&DNC#umy``LGqNKN+tK<6loyyaJcrb4&TQH zL&f(j%uX_2qK6}!-}(sGEVf#RRQW*k+!zg&@us-%AIv=%sfdfGe9$ex52LF0v_J0> zfg|YM0aJQ1U=DWqY2%rV60?I;bE-+-V4XOP{QYFk$)Ec9c; zv;}1JHk6!YrRmmDW9sN$42NC}V}3^TEkBX)#>nlvaA?;Y9PIX((>m|YUFz@1rEl*` z;R6*Zr6{7kE^Yn)Dd-UTcYO?*USF_$bDnT+>6Z^&{m)4HsnAH}TZ^c2zd8BuevRxN zga7B+@}uuz)OReT?)T!xjBMgGRp#Py$rfyEQWMhs*162p-ZY?FiHj*h;trbmFm%bA zgH?@#P_B@`d%Y^>FC~TZ_d4->3;&27{TA`@`7WF3l%XRv)qKS5<9pC@_yyOg<`8<+ zDpS{qBL8n=$t=uUCF0;LB)#eQ-Tid$O*r*oYv=C^-$Q$niJbfJXu1%aMkW3?>D8ZL z%Bku@`f~PUnzDyB+*BbI{Z_`8*N6J2%h8)o8K|_R}?Z1;M1KC z3CF)h>_(&dRa_jkQ~b}$Tz63!!=_E7Rg%Z4caIbb9k++fnKtbj{U}s!48zs>pF-ND zv z*q25!T+w%}oBN>ukKTidsnDSi5(?uset>fAZ%Ut z5sfmMBsFS0^(?rN!`ohP_X|b`&uzF~4zzX_@7eDUvDk677b&Yxp@sWx zXcLR0svN39Zm)XK1?eG-BbQ-)k7R(`qKMaLW^<}D8sWy~Ht8KNV&Q&0j31W6rD7i! zTakjf>{~ae%#x~&G)QtvJPe;?bDbPxVbHM%{T{CqIPL!%Z*BWoee=+GqBFhzAxDOa zm$CPd0baWr@D^c0S4_2Fj@2KYxv0mo`%AHZi`eF}f6;Mhp?& zb)G}IV`%OgOu8D6*EYVWEEtJ%gB&^pKH3Y-jo*-O#JG}VnSRXYP%N2a348Z!WI4!? z?fZ{NTfumUXSg6}8CzE@??DX#@^pTQ4tcOSacQ_tz%`OpXUz3hKZ2VYKJfpV@6={Y?R(zW$j=B1sK?EIDTEF}3e z`yW#N{_9k!y`}wIZ6oPLElG>XYxe}o&#lEr+byu|cN`&Q^|-tG1$wDHfoGQ>A-~;? zA?OfP3`x(vG^0S|w~)A=iouNY@b3DnaI1gXrwRel(2n6wQ$Rg+KLOG5=i}5|_PUoL^mO>kfOmo7$CbSNW1_ zToomE(dR=lC-AEdoAV={{}Jk+`%#6?4(93atl8Apz=n=88%UOH8C_VQ&Re-U@axoE z`O>db_{%+1`B7Me_+nc;p6UXx;zX<;(T83%l`{S&9onJ0j-FYbpuYqC>2M@_zuHOA z_O17*b=FY+_>Wb5hk?uZEk9Q9y4_U-J66)z4B5Spt_ywg#KDq|OcG;k_D0F!rsRI` zlj~rlcpau$!9tpA?kCYD>@dt6;?Ub%^vnG>uG9Jtrzi_WrhWAse*#5Zo~>?o8mZ(! zBVATdW7#;Nt>?B}!7H~ubT4Hv4e>K2rD5WFVW`yy1bF&VsP+!BPPQatcVn8`v!F~Nz1|`aPKb-;?)@nmV5uDxzVP8deDvFs( z%Ch6BkMaN-mifGW9L=4cg^b$Uc$ZTu9IH0$kLOSJ!)z+(Y57FzBr$+8t*Rg$avY)G z<>|D^VutbarRKHE$h_ld+O%7ND%7m$Nz!UM5HpRYg!d-Ts&}a9R?NMU*G7hqCp;}8 z(J3REJOIvM8tl7Y~FqeJN+)gV@ev{>o}mNY7*Yd^rFwx)o9)*(Y71At^#j<^k;V0AMi4M z1nnvjr`2{}UOrAjW!VVgPRCLTi-FXbyM?Sfa6-Em4GY1qTL@ zGZj-RqY?jZE0(?=-`;O6cf>VjgUSry*tvt2*cMcQ+}=d?Y%QU@RYJJGOYbuO(WN># zy&6c>+dI>>rIqkkjlxK??{t6d8h%Zz7w`3C0dM(f7=Lg}C*IpA8()pI=}np;UEDT` zc8wTFVPTRqerXOjWfPApkKORtQ~Wk62B~0P#XE$sxJB7nZlsaBlI}B1&wpc3?br@1 z9di|9tVX51rr%9J>RVGBndCLKK=Qe5UhjC<)`H z{zM1sG~}dBz>ZNNFf@3B3>hOj7UfJz>SlE6W)cp>y}|mRPV{xrY`SmbL!US4(=FRV z1ZFma_sqte>gUk^*_B?NxQiq^WoXZw3&|}jP${<#k#Tugl(rXRQ^avmG<7iD@8d?Z zHcS!vpe$xJY^;)?`s)Mp7x$xwGtJ3+t`<$%&a}42uc2#;?C49`B)S>egRT}_!;aQ2 zuyDvhp2b-V+y4|d2Y$ksgQ@M~G6_$LeYL&*yA9yHWV`rt_o&9 zsX$m*4#S%*6zXW=r^0p2`NM_u7=;-QJ+S-N|Kb^!7>=caY2r8hzcD&u4@d51#%%pA zzW0h1&QR#7dTN~Ai=Va4o)28%%u6lo&AW_Tg!g6n*fu*K3zl9&k9ZNAms{zHdbkkgL~qh9*QXN8a$)W_NR4r}ai3tk^$)T) zideL^Ivt;?Mu*RR1OI=q{RKzZoJ#Wy4EByCgES|~HXA_$^k1UWydN;VtW7%{)v5e3 zN7GI+)1w6Q3Fb-)eeumR89JZS@qyuoCNeIZRocE7#O(YgePw9n2Rn*n{7LESm`=%) z*>IU0g0jht3{R~>RthyZnsOX9cV@uY*dG4n?=kBlM}t=mr<>QG;`JRb*ahUngRR2~ zoi(W^)1H3AbdsG%?!reFLzyF8htJb$5I;5nOtJ&RZ(oP>8YT%h(}av_`_Sia9Cc&d zMN%6wsOsZje&``JKE(YUMe80S&rvSIGvM$s8x<3nChnmqsB!7Y=`lfQ)79KG{5GtD zIpdTbNwM0T^8q9UMyB%Ymdzm<2<5KQf?un?``N!WBu(Zd1XJkXemSUvJPWf zgAzvR6yQflEBdm^S%dCAyl|oksT(ZEtJ8zTo0=V_@d0C_8Xb7h3xKIzGPJ*PeeBTPN(k zkVyAdU!>oE6KSWy5Hfo85RHY?Y0ci*^f*ved=HvG5IB=K$6r!oOI5G#9&GC^Xy1>QoPV}Z4y^@wkPb= zuVD6$8Z5v09bdzrAR^%#H~Wz@4k_Qn&d_d@^~#9CTo~V_71Or2x}c`N?+{%w*g>CE zMp60cJX}6n$SG(&gV`>g>{T~Y-z&bf<(@kE<%#dG6(L!W2vMc-q?y!9C!TsS?y~B$ z0aWo&mcMt(iPv`V**8OQlF`m1kBn}Y|^yj)qzP<23`I7$Ax@gBZ; zXA)n_*8Npq-^7Zaf+O@KZ%`0Si_pk^d-CdLZ zsB&y!^Zv#BEA7L4@|Au3Iz|WaXWc^bWSW?5`Fv1SA#L3Va$1~7p+*zP`Naut`>tBf z%_bCwkJRIyO=rq+-iA(V`oeKJ;{?5Y1|QBGK)+TI6S+ZoN&EfWV9}qd+`bF{&nhxb zc-lzz?g2twMxX9VLyCtp?6DLL7%+{}b~)0knKdj1Qyf3rFLb9b(o9=x<3S<)z>I%u zS{LI3_)}Gc-|sro_XN?;QeG-P18r@H(w{}=wljXT?i*0Kr5@9J%%rg^z36Pm01DaD zi%veYK$omeL#uup($ zg^0scSQL$$wOuL9@-LFUR3K#2UU;wH*>BN6aZb|KhPJIk_hCZcg-E8senu6}C01h1 zgaX+16l21FH=n_SL&mg`@prkM7VWne6Jya$xg$YGj%JPh*JsV)2{2Co%T-oZ_7s3rG^e;B=!-*s1vi~Pwn z*t>Ez6n)%Lk7Gjit%B+b9PYTg)C)Qyh8%&={M^QB6Elrhzp#ijOzc~KejF7WIhiLp^)AAk2zmiVlofaJT8&lzuQGkpC^T0OL-!RMh5XZB^`WlY)>71fZM4|5BZaG1VDMKK zlm5CNDY-~f-aXN0Iq-8VE_ApKRi8_^Geh(@wdEVk?3#a)57Pbl56H;mC5?ENPAk)g zQtV_4jLPss)Zk1u2HD(drVo}WYYFuVUZ;;?pN1pq)^rTtKNo3!Z@34lqFt)JPclRd z6pdO&Wmoc`LMC!Ckrq$8|+>EC{DS5hek``kcuaT@+6yi!lQ_nrG}WrOd9$p{bKLdheOspVV_X~bFz&(yl4 z6ye%~hN55i4U5b2?LUYP_=l5gWVDc`t-i-tJf6jx^LWHfpxz4e$h_8+q9fm+Qd$8g zUrnG}VLQl~z5AkO7}MMBov8X*Fuq=EO2Rt9tm;_I|Nlq$!Mr|N6cuL|hBD9JIxdJ2m90V9tk0 zj^{0mUQn`JfpDEn>1=r3|B3kQUUa9Y7A<-ZE*uY>pNjsMl_^3;pAsZRyTkMUo9mo- zFj(j_ox?rJv0H+W{!;iaR1OO!?~bQQN5-?gPh~$<3UxBJ9Sz4Bn=tpZp^$!sm%Xsh zAI!#i-Vfn-TfIMhI)^bu;=5h1K$2EfG^4+~=o1PqI)eH)AxKtCLhPs|j7QZB-TZDt zxqlxTKH?W_J~g6t;1%56IRXhx`{-KAMsn|WhI%b8rlm}Sur2>n2N)0dWlK8yFpw5q z=}3Q)crsmBfQG)yC?_wItg3I*QKicieJqEvRD8*)iYJ2{U($CuMUPGmBgdb=F=JgU z?pee`>%j+Hm_Cf^kF*N)ckEdvwAHNNJM^@BgeURExZ|CRmsybepfZW>&)m;!xQ-Nc z>qlfIdvetG<^%JT@7eZ3Z!Fuu296pW_lOak^gL1K&Sd=3ia7xl|%2?oa?QDF0<2? zq9j_xyYAH=3fF0o7%<}_TptCt*UN512$qLOF{};q#Ycu>mc1EXG2HJw^?JllKMbdp zN09t^D>h!2#?~B7IPPI<&0q02cIXbSZViWbWd)aaI}1;J`_Lba^`tI8l14P@3U#qt zS&G%nF5Hk`hzqLh+rLnZ zY8mZyZ0ozzGnHVb6N_6r%kjaQ*(7EJ;%%`Hce{Iu@Z7A>(vo<-^dmQA0l0{=3>>#q zqO*(Y{wI%u^fHt(Zipo_2UFIs9(0lEJYHVjjF5bL-r>SkzIV({e*Jo9UPo&eKkd~| zDttGJI`|x?x}YO;t8p5gw-RmYZS|3UDf&pD|# zwQKKaK6_6tAGO_c-T?I3AMHNsD@_eMb5jZ^j-!fmYyRcS2JonHjEww?IJEMjQWpDqnd^i z#+7@VYHs%>nRFMjUgn|JsOCr-2X>I=Mgt0d@qo)1r;h`_4#4{08KIrCIbE!nfE~(S1bC_4$FV`nH1f8q1BO*ud;=&g!#-3pxS-EB( zM$*9TB7V7R?q95b%k*~qlfY|eP~Gzdq{(oB(TqcD@8pA!vOfu}gIlohjxv59j>gaF zt+;Zh99DnCdDkO#Md&6N!sqp2Y#)0MA6om8*5MJPA<69jQ=Y?X=pcHU(F&U(IyC7> zC%QQ@9=7+72<^I_c?*fT;YjcH3{kzTgky8J2a>PNJ^JSUnVR2+Q~q2?KGk5RB zy+uX1F)bCPk-kFOf)FE4uYU-4c6AY_*+(AVH$WoJ{wNpz&WO}k2T^@XD*f49M8ih! z724g_c8Q90^wZXgW-jkd3o_~;{jHJfwWpEvs?2#K`_=r0niG7`w!{2rA6H?1RV6P& zuM8O1*HUrb?6l!3ml-w&jko>Ks<9cXv$x^dt>Z#{PBZS6liIQ5_U0yK?33aZV;|DP zrhPQ0n-f+lv2}t~39MfBAcNj!RFNy9ArJdPkSwz)a`An{vR z`?0C+(KuRHB>YdC+KDnO6lktu5Z=zdh_%ytl1Xnl`qHNiE^iN_#m)nxetkvLYfH+s z4yE_b5#*Z^MBY^^g}(6E6OI{;HBjjM1rbVx_`2bOUeI(s%-y*HVIE;v61)%(Pm36c zb9K>(Vtyj0mt861+ZV{l9LI`@xzg5Y*(FkHNI26A)S#BFS7=T| z4;IVz9cOcv)1`g`DPe{gEsV6LIIm;)b8$4L{yvDtC}!JkISZK)cJRM?o6|77N>>I; z^U_~(Sp3vo%6s&h%4dr{MA_0vSU)`juQT`1k|AOjW|@d~rY($v^m6lXTDG%NXw#%P zcj)gdMe3W6nAGcqaLnVqFJ}^S2z5gqVXyCf%o{fYK201I zf8WBjFE1fKK$|+NtrhzGcCTQXqh?A~j3;r$+C}75JeRT^htQtg4A<<}2cIt662(T4 z{fuz(x!^^%+AGQCVi)SWtqZL+^`qz!LG;OXHO=r4V-iX}c4bQ#Uh~jSa#eY7X zNgYHBOGRJA&+Q_f*`pRT_9CqRBro-9~Sj zu9KepReX|Cr#VihR2U`B-v_1Gkdb;O(|*4|O9!XYxQfj}dEKT-@CzC|_@?*%yz$2s zeD~OKywRpe`a91aX}xu6*X_|X*TbJOC+wh2-({)Z_7=B!Mp6IhUnxjj6^C9;7vU~_ z63ccCf}GJBEO$DBq0>X~Ip0aB=UT;Uxc9R!)z@|>?~iOOGfcR}_AD-Nh%d4k|3kbl zc^Ayn5lRi{;!L7Y7+tNN^ zK7`R%PNSbj4m6iM!qa$1NPixJ4~)}T(jpx}mQjoYW+ASA6tTd{pSzHh%{1CrWzDq1 z9cX^cHKDE@^{t3o?MmIdFC>RS?7KI(0c!E(Qr8Z4qbd!Kj*jUil)E8Fv=?QZiWII3 zDep(~2aTkj=Hj|ZAyS?W+V!MOpTEFbh3SIpcnR|l-Htop*tY>q52b00nfMJbF?Gae z<0{yP8q@6Kov29lFg6ZWMb3;haB+EzJd@p+U2p)EOdqxM%NPu{bwt9aU>v-64o@7I z4Jq;@Mwu$%LRAogE3RVQia*Hg{ux?RQlPD^C)AI7S%g{B?_tgw77O@xqHt`9;yPhJ zpT+qqe@dd;M{dy7x2LG5(>ZS6`ET3^H4Us7$RWB`17qj;a9hsSbJs5n#`q`~l>0;p z*WNC)5%y7U4Y4{_w6CdZYmHn~!j3Tx+?{1y;DMxDh+f z`5@?86oNI!GL2Ss+GTZu6jY<>;pH&eTop_wUu=hUyab-ON=VojrQyZ9YOK<@&vcPu zu*5A_Vc5ba_;$&sy)Mqrb_u_GtJ6XASY`N6wxqehEqaI#pP@9v!`so*vq z@b7S}^Hrt=bqI>K`|#sjFC2|yYXp{l@YD_z)A$AlAJz!}C#~YyS~3vQU$Y_6Tl|K! zrBiw?kC994SuCM9ejg3Sr}^`Q^buW(xhZ`gB6Dsb)^2zv9J6_3LYm{Xh2L%0u2x(o z{Pv#~%w1}^g_r(KxcHde@2K0r%LqJwcv*PvPM&&(vIb*v?BGDnH;ibX(`%%B%Eyb_ zUm96v##PGj=aRbdE0w16JpyL&mvt5RRIiI% znZZn44&0144>z}$*_Kb+=2mq8J)T;ll-X!TEv@H5G}a(Wo9TRo>QdaS@l?`#K7F+7 zOFt(KXZ+reQ1Y_@1-p&O>Jh_;GMv+77XNO0b~Qb{J)IWCF^z|<-_b$y4!1tx2M)KI z)0|#|=n2zv`mekxs!R*4pCLWB9YcGS8Me5kEA75<1uwRFD|&77LHrf*`?+FyAeP<} z?cH%)A=6k>q8H^OD5WdI3M$QLAGa1!LupEJ51P1K^ifqMgtJ)62E1Z?woA-1+RMGC z)(aD_?FBYn#fjU+2$k92eoooK3@`fhz-x^$(A+f*J1mY1cxVv}<`}jj~5Y ziBy&(70;PtDJ`UZ)hg|%q(yu9&N)7QZ@>5d|E}wMu4}G2XU;wMe&)<lXgMVd}sF zkp8OwI9L&uko7t4-rJCB}fwv1Ec++B_WzO<~5 zxVZ2s%2h2zKlymbTf4RRlwhBsj~Y-~^DQ)5b{;9^JQHyw&kw^Ud6)4hRD`QKvp6I$ z30E8S$7kx^p(K;uI5?;sCNT-9)OiD1r7}&dv%xJCch_0Y$2;)H&B>YA_)!4%i1e}pLn>0WVJEqc6thW zsKz4CHJi}%m5E4h{yns(^fTHu;t3kO=C(+4l%66!G>+fX`m6>$x}Aq6S@CVk$p5(fw2fJhZS5?YSh6i{6Fcks*N3A1%ZBg+K8RCnd>> zWJO8jx@UOo(?mR8%LeaGkm9labMX*0M||g?g=iNWeMZux7$lk0)l<@UhM7cll$=D$ zToj)M-I&`JSA(Z!_9Ul7RJvtrL z7x&f#JiOs8a$kQ=#5*|b2!gjRXg~Le%f>lMZDXLuK9dG1NlAAS9<}e!Z5pyx4n<3Xj-m?{GSu{=7$v&B zM%!1mj4=!Jr_%)VD=Hp!F4)hHz0XE-kGtVV87FYe^eX;Lw|DsA+UJ>xoe$0!99Ya0PohST0n=W_J0go8;B~2S?c@L71(rFUK zZM`HWemx{X3wlUm^)w}_{J!Yt`wrk5;($Ah9)k7SjmF>Bsp1X%9^a=<1!!mA6X;C! zb=0#m2w7C?qB+X^T28)SzDN23Jx*vj=Vqk*ShT<;7P&vVj%M=j_A*rtohhh7_pPFl zx~mjB#phxr6?wU&sFrujgZxIL-iD&rx%sGYkE7WOxf7^-ay;_Q>V`u4T|g-t6|n8+ zqiAX!o9=jkW^I(iUQIWVN6uR$JCu)Hquh~k=>dsT@h8c-Ue_ht@{da# zO2Wjkhp^~mbhYvT+FyGJ`E4o|^HS6oTc$&2ip5~#0gQ`d4;CcLhfbXwR$|ED}S2!MD zI=U90Z61Z6%xEvd`^OcL&jJJ1@D?cug-CT|oRu4o6$HyC(ny8}`ute;(miheL@VRK zE<+n5&Y*xZDd?-jOVlfC*dI$zPQ<<^7vNm0?s(dlaO4nn4?XFpg5|2K(2aSYQSYR; zXx-^u$kg*R@?FiZ`zS6)`hh1=Q`{?*-1ijXy6wbCC6DpT8L#I;PhAn8^!8NhireY`&Sr)2|2Mg0oNXu4P>$=a2W09G%HXRz0?pxQ%{~ zNlH(2XU;{Np259qA-Cv-YUQ`Et`16f!l=>cvk`-*i>l;M2}8Dja|l4R7;y5)a# ztE@kI&qfDGOFt;I_6w>Vb^~p=orOA?$B1-C_y3K;Pb*?K*Er6Yo{@K_Us6HJLCg865bs@QXPun!5!3IeNLMU)qUh~@~^Hq&p{J|+%vS;Q5*kQ zzXeZ=+>P(i-dKHhOP{`r;T^O&q$8f`G693*Lj3$Oe~y^P5kwB}M4d`MqM;WI@uh3y zu%2hED3|tHGbiL*=>YefuIa&Wi!|t6$&lgys zqbLx4%6*L<^5@0ayYg!zem_G#<2H!2A2s&CpS`=a*p=V^)@s{Q_h(3Xr7C_>b_d;g zn9Z-v{cSTi@*O&edgEO-2H54UKknSq6EA(&3HR(_fe5=%#8(%63iX{U&&N^Ri=v;L zL3(_giFGHf@er-4ID7a9v{I5~vxpu+Yi#czll`h#z7uc9coXccS%8jS>xL(P9)|lp zT8FuWRaoI7?=KD~QThiKeW>p)=i}QC8)Q1*z^1|YIX~C7ZDRp?u;4lJUZ{;XFXs1n zT$bSFEBM$qd~7m7j^Jt|ys@l7ye1DG+GPH`W)re|3Yz_uKMUuaCz{(Y3vG2a5ueA8 z`-r+m>tMKNg1cYf=aD;4MYoRd@2u@z)K2*e>hSXhTAOeZb$J$nx^+8^RJ?wmGs*JU z)>{kT8ll#vH-WD8yK89Lqp!$i;5Sq;O^V$4HS9eGx5S$0{V)MZ8dF7@0_rqc=79Geoem2#(dlu^wfGD8Z@2X=iWOWHTCK&*7>q`G^*Iq4{cpB2~9ua zf$H|oLzC!Oo6%YyZF)Prwy7|-Mk@?=qR;&P9Ra_C(lDIFk0oE-XNEnmbVu7#Dr}yv zT8ifATtP^_<^RX;qDLs^$2OFB!9rr_u}-3Hx>YjMbfzTg-FK{RRE=wjDOzQ(gG+pS z;hUFR*8HB;ZyDndXsogEK^wl=qMe%tw8nO>t7z;kN!Y5iK}w#5F^F7JXq=I%l}zwF`9O?rpwF3I5u%yg7G z>odCc#T4J><6XF=Ova!1bK%cwx6GwKI?bOGE&mBMUR1*lTP^Xd^dTbtgBfOc#BzH) zIc_cX-k*ePpW5TNq3X!d-5%*4vQg_+5`#YS{rj6U9Z}laj`+i^p;+~%Ck{1Sgv;#8 zP{H7LHa6BL`E?u>=wfFbxtI40@#QyzabALH*T8BG96oUZ?q)Lr+uwSLR8wxEq7NTL z8$|r(W76x?qoKYn{aZorttrB040@aDD#ATJ<0ATrY_UIzR8B(A^*$l5%^JAwT@pI9 z;EK4#cvfsIwjH+*%gsN8brksZ(NVq8WV`(+^l%Ca z{Z3JJO3S>zzlXOdTVES6OCz_r6O)SgvEkX87K9av1!|wrji&W%W#uU^)?!?y8 zSlppF0c#I#i5ph1uuPOW_Mw^N((fJ;U7Hb-J}bIQjQZUX&x7)OZH3xxT#dS!#*4TY z&P_l`mamcd?No6IUZZDfOmi5W4cd~A+4=;LQ^QI6tjdAUPtt?(M(C0L`gtH_(6 z=L^z(*yZ>Ge0AalZ1pT2U(oWwJ3Ds5$z?;)B*|{1z9a@!KMX?~i~w8g0_tShRE97xZ3D6Ib@Nz)eScV91;%)~SnFiq4ITL8c?dp{!?FHoEKOZE9zn zK)y?AQLNh!*Y$MQ{|(FEU?Z@*LJW)!_96^)F(h&E{)5bO6JTPo&uKQ>?u)imrH z;E!u}_U6|ys;F*%y5CjKg+I82PtM~MmG~I|RerYkzkXfiBy#gsFIcj4o#%js^VZB8 z05-^G@Swqi1`ZlX2pMfRSkCSL|Hl&OgPxWgwKL4~t=2`T4J0I>9UuYE#dVrNvTNX1`R~Dt&#rx7{f6 z^s7AeFCD>EH|leyll3_7%OWmm_2{YGy<{Ek&C*qL_AebSmbT}XwwuX)J~N0bF?`=f z$6B$LzV9hvuA1K>~@@+ezp3DyP3fU(!?8&(rs_jxZ9i+M=HUv%f9``EMZ z*FUB0l27Dc^tb2UnfK#VKFDzi>(Z!Z*?6?$swM5LVGJIN<-zvxd~ORfmNUsQ<=9gV zbe%?LZenCF&gH~FPA$%uGh^FvgJrek$oy~YuqTQnLuovB>a`^|F<}(l(W9CTaHya! zrv0RqW4p+jFL?Z=AA#4YOT+$>VT8p?dNE~Mo1d&dY=e({))D=peV>-{%(8oJ&(~EA z=Ju-p5p*owv#?ZlV3Q(OZxc*6?2iy(b~YN*-Szt1jd7c}Dz1>`B!{sA-FfRu*@E{8 zH2H!vRho2}3|!WZD#pfwfX8BsHs}7nnA-d9qbT+#RSQ<-rnU>Aod$gsadr(1rZ*eZ zL|GD-L~vEzHMxiGgdRJa#CGjy&&8nyoa5=WT-qaNF61*y-enro(4a5G_~nkjWEJ?3 zQ#uW|o=;|0L``69*@4Wjj0(8aKU6Yf^%MbCTDj@DY@k|4D6A~b54q@0mK)_ToeK^y z@#z!bd7}ffE*3%c?7mE&3dFdL3S?Gqc4Y*-LIP_`X2qjrjNjFr%-l=|Ci{gRGlt8^ zYke-{==4A^(HH@X>K(*-TV$go_BAS^oQ{JXN&9}=nDZyxnP%;!j8=dnvoX|zIcE2U zesY{d*?<5B|F#kJX>whUeX-`HCeo8}uTbPn&+VdXHXBn}<_WM|zJ|$D)TNW}4Pp`oxiH(v zesXep6iALggfh#4VE3mZ^gDJ^mRap0+OJ_)oPbxl?4bb@^yxV%zLp3V)%wyvr8$4m zYg9Q;vnNew`Z05vwCwdvlj(B^*Z4?!eILT&;uo^qn(b}y*EBD)6Ft|mBXs95jz>Ic z{0LR%zV$Px-+r2p04bxFV+JzQz7G=VPW_Wh9&9_$-r#mJ;f14_^I`Igmny@Q+0SUh z=idG}jrB>Sy$xK!Y{nG4YH7B>yIlHbQzj+ZkQuvVB!gYgi~7syT`c;IO~6hOX4a<7 zVlKc7ati56=bMnK7$njc?h_R5$+Wiww$T${?Q8Zi^IthJbH1)(f?fwRyA8K9U)IiM z?&NQ0o=>o3wz&ti)o;c79UCz=(}S3;UFI>3bffq@=0Ob@cx$)9$_cnzQ@q1dm>=;{ zH*ap{%Z~PkZ zos^p=Um@ZwP;p~G(~;@wq|5Y99m@E)S~4SF3})u;>&F;OTf`LI-OC)eN6>F@5VNk% z0@Av#U`|(=F)xPdF>h9g`f20!>u*2eGPw+TE;8q)c50x?X+J?Vs28Ij63w`L4rV^A z8Tl`M=HNxJ*X%4YbLIyl~%CA)q(ht3c{|mlVkFEKo;U=&>Zw3i_UP|uDHnHR9 z4Wl!)-qY?~FOnN`bws<)ALB|ZQq#%m;@cA0=6-CeE&}jq_PbNZf7)J7W!5T4Zhl4N zmo%}1%LkL5{Z>KE@H%3p>Hy8}4L~+z8*IEE1Na|L$HVbnQ!`QTtvpoiOoP|C!}DLh*)F!ZpHVZ>d_i3yeaY_k=nXWB55#VdD*s{ko(9S8{h|Bq?U1T7OsxAW zMv*Q3Zu!@Dx0RlT(Iw(!_5|*;*<_?rN2u4Xmt6Apq(1u>fKdjyg#&AoU25 zcgn-9^;+|&560k>TqWp48vgVk><%)7Ql^MF%A2r-v+70q#@3DOuZAx0y7v({0uyNv zxA!j{UYF{DW`+hSNv>qiA3G@45%4=Lw~|C8I56GhbAZa_<#%f5&)zY2VONyRmk7_h zR))j3jpOL}tzlGmY>6zrWB{=}9t6cZb(x`a<(bo$Mv1rzlP=Iul1iOV76|&5PCHCs zsMBns^yL_+q&mQ?Kl@?OOD(vm?@C90j)w~KT`YGx4DP*(fjJpgfAMOxbR@gzaJcM5 z-&>?qmlN^F4l8QYEzbW8)px|UOS*pgL5al*4E5u+VG-v`P25*g*D?EOW>zRGvn?eG zH7|kNo`|T zeFxSWS4h+D41rEtoqONZ;ea=PaPALt@OY?1=bcvv*8mnQd+dgH9S4%fHyp^q94Y-+ zZA*_C@@=`vH%aoX_fWFb7dG0(x7oH8r)QWO6u#I8mOiDRRiMaR307xz=E{q0h*w-Z z39HFr$5mRj)$9B~Q!;J)Sw0r08T9QxntUqm*7i)mX_k=2zP1Y_z7ai%qE{%9UK|7i zWk*2y^ETLY+!H$QZTTiLi?7Ru6)a{gG)&vtE55#TylyY?K6A}EnEp0T#66?(?oxi?4a*5OeH~Qv-m%}g6_4xtBAs^WKb5Pf(2;!y${Kac4o`2@LKX$_4 zrZV!rVFS7T$ogOK2Me0OLfMw_@i`Bf8}~p>njG_V{x7JTmjrfGtI4YCa!h;Yu8dnZ zeWtWa1{94P25)|pLZg2SL>TUA!?%RZfQvg6n8rT2k|}ncaAnMX(8^L}POP;duCLNz zpJfEuIZj6E4HID<`wVjU?{MjFBWAg4B=Dz(2zaDI8eXKxtX=m4(#K|j^4jyz_w-9x zXw<3goq6wd8Vo&GfRA@RxF1S`u5)%l!=8T7RQs5uZehgtMlJIwX!*SdLU)3M_kprC zcVP^yAc`C0|MERSPXq2N0&^3S8^tq3{}*^roNfj-M^z*%aspxb)U#kXr~-Z^$uaX@ zKZ8GAyrATIN2bxS7qfEHM(_){2rCzNW?Wz21h;$9Q2)c28CTVd$^6`dsbktPa|$0o z`#q=|kFU!V&|IH{c zx^W3UdZ;nV8~6^=!48bt-b50(J(y^EOp)CD*q%J--UN$NJ2E3e6&R~Zb=YW>0qlq) zF!nM9U94s5nNLwryO@r*hswn@%O2ic>^qyKuZmd>$&M5#Tb&Ud5-KJB5I z^E+_@tG9Yn=KtsAsZ8{b5?>{vEYTu4#C~{%;X|#Qp4Rrf8H|l<&k`4+g zl(`*xMc4UhagRFm74S)gRNUzaeZQm=H>I+HEMwI;EuzCkZtfw{^H@g&{BlA{X&lbF z+t-SElT`GiQF4~_RoFI~l&H*2@cltQZLgwN9w=}{Yn$niaAmQ+Q17{%F0{v zDdVcIrx+Q8FSLZ&8^PI4rzVYtJk@EgrTxIVz`2-n-WH;FY4k2fF zbygjM4_|JROW^UJ$yEA|0| zYq_4MONipbTvn>k@;^-|>or4>%M9~W2AgPYrDG&gzqC@zIR=JkcqQ(ErT5!T=NLVdQ>Q#Hd{ zR@3G@omFsz-0gEt=9V=Qs&4F}1-VD5?d<2Q;>m-wAuC1n8-btJ)L3I8%B&ISOJ%jn z+{DUjls_qm`(^xvuDIJE$lV6JKD$Qng|>U)`otLipzPe1cBxn2E%bxNZK`}vlN&SF zkn`x7PV3(1^8dx0xwtW3sat+KE@J2&T6b_H8Qo!ctL%c##1z`~k66Z(^nq475yQ%a zGVyss&1q`aV-%cvZco2F9wz#BSCwGVKW`tarp2EWx%TN&y5H*>rJFueqY`B<@tGEP zdck4ZA!7u5eECzv-;?P}EFAT@z0t9BaCibe?*E_nYyJHN^x>~NlqZ^sCeInn9iFy|J7QzQDX%l7P4mC7;VX=257+5{sxTR9o6F{>oua9?uW^%q?>3 zbrAI$P;!ONYFH~?4^<4|W^Ys}7<8eW>vN`v8wdNj>EF+C211>t;*ukm-vV}{wM9$ z>w=@z>|56k)bR0HT3UFCJEBABfIHi`!**-AhIU@?w;A20pBxpLu57UZ}o&boY0edyeOButFV(x4oc=C+o^Jz z8aKF}ztXvYGYdo?=(j;ZGQWK|H>oL&8&hw{VTms1TCwI|{Q5gTqfgvbSms`dsGGs% zS=?fca;|pl8P1)blxx+;KXR$*jC@JW`UtMf>58b|&3lUJ?#wIH)+dU4yssx0(su>t zLku`}RF*(bdZ>9f=e~D0*Fotx_t5S#mu&o*Ykv8fJ9X>=7d-co2$QvZE!XGq22OTm zhj=g4J3pYjjehwi{?uaLH{{k&mG2O!mn|uofEnAaYwmXCB@vbgB6^{(h6?)jTA0@ zZVda>^$0t!F%nv3&i^6y{pE!FMU{12zb$9E#uvA^+Ou!CO=;QOJ^4#qMQjYWeMh9o z*FKB=^v1g+@jAP*hEvgd$?Y@S%Y94U!L2gxM`JeMqx+J|>GnAnsK%gp&bYlh?LDFw zeQCIavk05Um0F(SNO&JkA%qce1Z)4Kd#`V$q&SnSyg8rqQi|e6MW5mJ`tIc}-agJn zd_2e)^_RPJ(RG zf2P|h;=9;dCYLx&mpqtiMk*rwC4BH260iE6oV-*@?rncaBvZzLPqwo7jhtWkhM)%} zq$#|R1Q*qiER}s=C!>(Wq(O6vGarJBlEc^ALEw!OWP3kfc0cU~=a&bO8)w5wu;mWY zIKT`nyCuN}H5HL3Imbg}B>lzz zbBA-pJF}FOD{W+ZjD0|!-9}(;bB3(`A|)1@2W38T!(_Tq)v`T~b7i&{l8NDsI?{7p zuxzQiQ~p17kkczqQW@(+1UnyYk`bqw!^mBg;5I)H%KtkpEPqC7_Xfi53`;1Pve&$4N)b=zo; z&|M;N^YsANx%1(VVI-Ku>p_T_9W-l;|KX*!k)vQwagKy6suOuA`Iao^n~JkYWvMB9 zE7n=OZtN)UxdgLHd`JsfY3pNGq1$#j2{3mih^cvJx9s|4DAz8glfF7Ch|PrIpV%eF$=FXzUGVy0jPokcm zDi0F*krix){39=koJPHPzy00=$&C&DNNDmp*{>;m$i8b9U=6w;>+c6QJMAQsf*oP8 z!+KD7FX~#r^_RTrj{QkqRSd~l^_ASvbOHaUEOH_@S>|ChmtCQeO4^ZZ=&mErlz7QA zZ{mK#&mTsNwfaSc0XQBA4*c#Ryw~2N+B~Y8Y*Wt zgXJu9CVROmR3&Oc?=6Su?GiW2!%4N{H5sjAbzqroA?mZWu8`j{vV>H5Cvv(%9Oscf z+gMI53oFI8*9 zV_C8?d$Ed+09>fT5okv5$7YlUP=- zCes27W##4R{NAeXu;y-vXv9@uYCppkEn0-q#FxjWYAR zk{@ro!b+CMykaXNn?xH;kDWr6cRxZ-jQy6s;9LykbQ%S}1}ngjQO6~h*S!(>uy|+A zln&|0@N}eSpZ8?iEwpCTh7JTb(-m+gpev)d-kj0s^#!(87K^-P*oy6lsK=LqiLA<| zGvtu&5=hF=huhVKB-3jExGYRzV;T}EZGHfM_G6fAdzkGsaXa*QF{MrCKX&gYWzpZ@ z*PA%t)-GeCe|};+O>a*&Ug-o`Zb8s*;4Wgfq9&&RwB4+89O4!v^pUJQN1Q{Q5A%Vn5 zG?5ORntDm7EA8izfUVy35U)QDd%?HaCrEOgA#K(D?yaE^YjFgY9z6h^o83jYj`MRx z-#31B0Os$FBP#>8!?b|Y;I$$X_VwxvFD_fbj9>4dwzDBqE74;5OkW_>Yx6(hdW-fS z$SHCq5l{u!t#*jd8xAXRRg+rgM6T`%B8JCG*|KF{;M&4NurOsy8&8!J>)>K(DqL9@ z)aEO>0pD=_8)Lz)BAsbjtEu-q7w8fjMaoL6M4xDV_CMw0PfmvKn-XbN9WY*B%b_dx zlzp1~O~j{jw6j1*nmVwCuFhrH35=%TGi~z3258aYTRSizpXV~eFE}vKQK!I5uM?A( zxsaLPqZ*Fikz?YzSTpkqRsw%GIH&T(m3HrJ4gQDy>D%TsI<(%hRW_+W=bx!+cqAz} zxRI*8iKZVaPmmzRKww>5{-w`1Rx@S&%bZ~A+6H(e(}J^M>G}GXyFljwo?!QL9gG?= zo!mI6&$v|>Gk4XV2=%4f^G-qedoB3WbrgiftY_{P=E9$hhY)`M@9z_uoy@)*HjA0M z)|Od){9?Db10sm8B~l)Pq_njEBR6&!s$u@ul~DbsPhE?qfE zh2l=P$nK*y%#S*4`0zy?JjSdi?tN20X>%ei9T)=o{od2^J3GmOpI_lu#l1E^X|-Eg z`Da3=D$2&5EueY@PPF5Sg*1@8m473n6XR9o32E&W>ASq{OjD93W1|%V;2gkQ*a`4s zO(596Q)9YXSJQgq-=Zvyo%_;n<>O%d`zQ$D=U0ZhMZu1jk#upV9ppvm7qT*L5%}&k z6Wf`^>YAed*17y3S)Ul@>9}xsqD}s#lM(v=@#X9M*yUE~V@UVK)VKT(JNSM9l&Pe^ zwDeP8>`($tX-#Cxlm@us9|#%;N=aGOTpAs}h6(ACCDQctm5KN@pFM)9amq}PydjK! z90eNpW8rHdhFRyNZPg^f>-;&eubw{zatV}=n-9C@cap_7oUfZtS9a29pH)EN+#<%-)Go5kSV&o0Df&YXKI~W#`Kdv z>oeJ7&cF`2{>Y!F}1teWghjcS_#$;kV<1uR| zGdOoP^EO_MNiLcVX$jiwUaeAa`V3@fEyS67`{FOkulsfXIj7uUAiI zLQuEE!s!jYcou3nnD$OAA^z^evFyRTxMCC z3v(lK1hevrA5&-N$`C_MCLr0A`8lu>2AY0_+6&{EeP4S1)t(7BHM|#y^phUPk@?Q6 zVEoo7ru&C^Otza1v*L!7(H*mpIiu5q(QF*X{JG`EY<{4`plkcVQx`;DJ_TAZ^_&-T zzS))ees~&ld)O96|Ly|jPy86>-p;MeAmhw7pPIS;@3k!g{nj)lcq2^qTmm;&5l~Gq zVm@UDLy?sCxhp|TuESpD#hJ}a+0Yrx_-@}M88vTM#ytuubFRoD4Tgi_fZoD;FBQ^n zFK;mEx?H3=w009DmZs)suD=1H8pUAmu=FqakJ@PbbuaMptYR5sQ@xW}8grB>)e2^s z9V6lNt#nY_S1uEta}N`kOow^QYdvex=CM!x!S+QutjO=isNMPo9_J=9$3`kLIj zt2PK`PTj^RWld%FUDsi1H~j#oGXDQYmnX2X(`Hz;Bl*D3c`bi67xu<{Ytw1Y!*NVm*bqhmWrOp}XHfg} zB8(55-1dBpw-sc+Z%>nCK`V^?`rkX}1_gHFs+Shy$?4AqmNvZNQY_-I5MZb2U(l{M+nea@hBJRbtutH7!b z@r+CST^LZP&J6MB%>xVTPXxU{aqh7xg`$TN(K*=fZaA=L_zBVFEtUc9X-p zvoF;5z+M0I#AMet+L2!)H(~uTD$o(qw)z|5d(e6>Iwxz~)&%#Q2&;4>-_JXN)s zD5e8#zOKP0S=EYkjNFiDcY)@)rx#>kBgZA*+DoS|o7Com!n2$Wr$qRF;ufx#OBPEG z&JHGBXK=*o&3kf!bEC7mDnM@UM7WzW8a~~*!@_k(vZLlD>(G5FoIJ&X^+ya(yNh$# z(%0V)Q@yKl^q88fNUu1of<%rzNzb3TMdwwWpsQ-j!27~#=rF1sGr%|-?pW_2QJag& zBe(wS#>lHQCbk#TkP;&3Q7ZrDr|8qIJbpFSfp(E`WX0JS2xy9f3G4{iuDBB}*6`~z zEoU&jI=Vy1ftPH5JsHi6>;%Ee@vtt%AJoH@n5gqF!F=Z)YS*{~$oBxzP7Rt_BL0zY zBgh=JFYMr)!PIj@H^yA;3F}%IPP=Ut$E2l0yEwwl@BI3-uuG(Q<6*qc;3=#b#lXs` zPyc1l_@A0MWvY^>+YWbCIP25|)cJ?TX_EzKAjsMZ&Og)VekB#Mqpp7hvm?GDAA)?! z6VH*+-U>8EHynH;otbGT53|&H4eK^aM!fwPW9w#kQ`UNM1I%2h5+sPc;UuC7*t$D zk-i_D;#C3WC|BXDZugsGp20S9%wjY4|2ohnf-|Y#LcV_`zR(;g_{f5 zAImO)Vdn&qp9y{bcDN_RA};8`!*i$ zcxIE-WyP}o(VeK)w|v$o$&qY-8_M}Fze8UwtEUTsv;n2K-}4gxtw#w zZ0>-gGMk>|*=i>dZchCpy0d;Jd8*#Po;^N^K4aH#@$vV$-E!ZkW@&C4zTQQIJNH9h zVz1PRYHv2;jIKIT*Hag0ipDaS`f(0Bc+cPKkOV$Tu@T+1>p~m*V`JAUyxw zUXdKSx`drk*hm*He@->cE{o5@BMsr*YygwQVKk&lpKF>hht~hDrbBkTq#K_#a3Spu za!bd$a=I$!f<2_aX9Uxw0fr=C(jTGzU-oIx`avg;U|52rX9aL4EIx zXX`xD>6qbKV6+JNrG8zU_0I zx}RkrrfZ38vE~SBI(HtY;<}A49(0)Ap059|_fe^XHmEx-W#^eHwE20gWhuFo;Y{E3 zaF^}#J4X-Yccdk8U1;5-puhC9T4AE(pZp5feXj4O{YQ!G)TCM;lBj`~C08A}pL1rr zb5qCfAw8dXa!s=qawoGYsM62ff8i7AuQ;I9syAu(N14KN>FV9SO#ChvCZk-35nx6| z9*4XA5&Q_939pA8BX^&M!ALU=CUej`lGAfBgq+{Oavu*90k+EWQ$FX|OT?41sYw>h zZGgPvgwcxc!K}ULFejhmiguFHN)K+AU zfBX9zVHKn=$Rs^KaF^s_;{~=}qQcnSoWWeX?#1YC-3s~o3yH;m&7e5LNHX-DKHLu< zP7J1OhBHf)7|R=7MSZm5w^lTUt`FY{G^KK@STT>VmNK7p-I$bE2gXVM-|I0=ZpE^8 zOM4UJ|OgWI(SJu`eUav+rdUWMBIL!5J-M z^{qVI?qJSzS>MdgTqYCzNtzR-z@!d%4{ub*fmU(f{4={%ndxig7>Vh1NcpPD)M}QI zgExan$ioW|78=3cedhzqvTC5|S_J5&4Wo_fC%{GCh{}Y}6xDTOG|S(?sMYF>5{zbUWg0U-GKMpiO0g2-jJ;57+lSHVnE*R< zEFtQ|MPQeIfD$i9W|OxEld&8#w+{Qmm*V3fyIaa0)k~Ls9_u9eWp4()orBoLH#@^N z%Z1Q?#wob5eH{~+W5|?r{0$j(n4!v(7{x`in9@5cOg6s;&n@>R-{-Mo&iW<6v0E+c zjazjiv>=1QxiW%=-~vvAeU9*&w=81Wr49lI)sFLBIw|2=)0&h zopIO)?A>fZ<5A;Z{v_bj()MIjUk+iUVS0?^UPoqy{6?lZok?{b*V31T=l<%`y=eaiY;2Uk!?^}Ft+Rj<4@a1xa0nu{ zU4~EFRYZR?Xxc&AE#4v9uSlHA31KX@&PVE+6AYbFcs51%#gHNh3bm4Zm6gn07je6`EjcX9Fx;+iM z)&W*G$I@MYN~pcwJ;{l{7&;;QfP_&m0lSX-Sxvp&(1V|w^wf=%DW->$4|%WX)~vp4 zCw3BPI1oUG`A?=hmzl%KZWSPHN~THvdhGluqeTAOeK<>dbTFW5FWS)=7LWg%Z}DBr ziSm;LEXXBy{Q9)n{}tcga9om2d-m-K8$z|n?2{eHw_^tM^MakkJEB42pc^GY@eJKR zB^0LEf0b$N?$7T2S|!*^`qj;qE_>)ivf3Sy)!&XGmIpUHNQItIU>?SkE@oUWN5ztr=<21&=mfn>wh3q*d%JKFo;6`{VguE_|_ zo$W|h*$2ofFIdu=t}3v*mo~tbn`G~bLn4i`;g?u92SrKmSqXGd)+*wAW+Tk$FbXmX zKjeMyy^y`)K7d|!Z&`zc(M;NIwh7y_T^F{pb{rwjEIse~5YEqN%u|+_QHMuivYvTq zH2YC0z5KwMz8SlONUm)s=C_U5)81R4kNOT6(QyZr&Tsi{`f7Hd7b{grcyu&v@*~v6 zb&M#Fe#Cipj`0o2B&A4NRgggA`WBJ~)ll|a@JDv**7n3KG#0*ZtB1=+&H;Y7jXr?_ z5;FS{eIIz7owMsDRXIDHZsmQf$bU6UkGAv~1%2ENP@rqStF^t~2?Z1B8>@0w@!$#B z)0Cg=_(>}SJ^y7-BNr{&d$g-)=QrK=fucuzzN*J`D)k-CCGE4|1iV7J-yB2^ri=a0 zzv}mMXkyhb>Vd|~I?33o7_i=POZL)g351_J&gLAIveoW}qRhzyGx7?_4zhc+H>_S0 z1m2Hp*`AL_(+SZRXr~h!XmIE>_P|qjHq*$D&8;6!uiUC&qq0ZPCyO=V_OkPA_e5u! z;?#qh%pM0DuPhXCMjhBsx0aPjls+MWf9Vy)RB|oXnsgf+%5r|4MLl>QYldUx9U$$8 z1__>#M>_Ch=c~b-I(_cJ_ObXW+x5!Hza2_18tQ7oK%BR2N zJ^k=7v0OZn7;9ez&9_SQtL-7#d&`@$kC!-d!|zmHzC|#&vf@03^i?+<_S0NvC|jD9 zKZfs*+T85~``*uFmo<)zBj4R}eakS^^}W6y5TBy5h#T(bI9 zH3?1J3FZgmfgg8*9lsM{XGtK;j5Z@jvc0HBs@QiH`HM5`Mn4=}N*%A|!OVw?A;xAj zG~N${lduwuHbs$ZYkso^8c!s~dnNSb{gWj3>3OpFcnHimeucD`&4LpXJ)uhHtf1Gw zlNi+F~_&S?gM>?xmr@T}iQtE=_;;h;(|kk&MskKn*mk1bd5gm+!a)UfeHM z`BE5Nn6aPko4uEH-!~ibk`my6YZtITV<2(*b5b_+-gEY;{#**}`}e4L1CIJd7AJ4ek}hfWq(* z5PQTEl0IhtOJB}Yv_*XL;^QE6-EumA`gvfioZ)`5x#*+412;(0W^acfKAN=dgbC+t zkqZ-Uo`aD}&Eyb0NV0lOM^qK%H(p{8lzlN zIQIe3*%cBCS^2{7<>li&Ahs^AvpV#go3VE&D%O;WWRi(9N$pFiUl6 zAmg9`6dh=p!)|^4zvR-sQQ3;=J5KID3C5Nsj(iL{arf|`0? zvi@mjc%Jx*b-3&yyD@1Id3?%1?2q$sQnz*yBLADW^=*@*$S{<=R*sNO`1n%x)>)Ty zi{Qs!Pn$C}zPilorSiGv>+z9XOzMN5m28`=32Q^VKTgw&r7@T*Q%^ zTrTsmahBcfB96&PF^~Ls?Fw)5TGl``4%P>Q$ZJ49 zj&8%FpEy;#-ZWJlOOk7qGq7<4e1EqeEUU%&b^ZV;9}(`pnYflnI(5x=cGio2DIN`(f}8Fze_I?>Fedz&->t7oUa6?;2q4#!H0Y{ZK#K0A9D>*hW{NGpE-p}Ma(yg7W9)o7g91|z^d zS%37e`t^X6>`0WTeV?U5x z@5MbmBH!zMTE-GzA3Xupn+#vZ-K?%G~h_hH^`3AWE%dwhGRQY+h~WS zm6D7hmY{IyD!XwZ$L{_3<^K@(=HXOD@82+rj3q;e43%U^AO_V}<|3JrBx9zt*E&ilG#VPHM59baWNLiR+5LEa-|HE!>v`Wl-app0*k`SK_}uqe zd#~wUL`#7W&5JANWI8N+j-kDZ2^?Dk!N+=85XMt7^?fm4YC0jNzKOW`Ob~n!Y9X>W zj}vv(65y)VJFrhDf+(3t2D{5kxjdOdSGx)S*-{`Eq=Wm-01UA$!I6!Pgp|5CuzF#{ z`4&G8{uJLQG~UgDA)@i|ydGMx>0HvT8 z#7b9R59+PISA?1?z#Pv5Q1}A_#N1Jb(bXm(?t=qRwp#&`GL5J3KN-FX1jmNaW1LCd z{(EjnokcZk8qi!$98LEbinKq}+7VOrxn=*zTw`@+c~ZsoXTYpo@4(+qEpX9z5nOcR z5je(KyZ)_z1$1ZJ1pUW<5V=;{L0N_}SkoH_!nWCgNU7_b{Y2R`y+5gI(IoF7UrXxz zwnm`9cOSHMeVb}OA&RfKWtlk|axm0d2byh}50ytB|A)U^INu3+4!s6j_xOVYR4jV= zX}5u?p)maYY8!mE48R(|8AAOH1G8T}1$KypsT&pHrayvQ-DrP%Z?un4h&~NwNr!`O z-$2lkSx58a^5v#4pOvA2L_d+@=|aoRt+#tq88um-n;-aLP1s+&#W*ZgM8#+-aNnae z?f;h3w8nC8fCs5BLFaU_iXY`3G@si(*TDM(UbuOtAqccL1K>(8P$UR=-ro{_oTU%$ z>5D*Lvl+1a%1JmurUSM66@)LpFa56O(5LBg^G2pBm}7k$i0tSGHot`l_q`E>YtK7i z_i7XPq}IaP-Mj*EOpBPw$WAdEM#9+Mv{ky6!0> znQ6gPK9mHSkp^IAl04;Qm_YL+O@dKK=Kd+WPs)34=_`sW18Kx$TvFl(Z(O>FN6gK@ z?etbktmF*1t|$Zoj!A(Mdr`1`@C2x>e@y5kKB2ZwQzVvpWumC=eMJ9pX(~Ss6FBczMV{?e>dIz*pubUY5)&S- zOnJ7lbb0c#E|K;r7X&HR5;G-&X&EJMXHbK;Y1`6ex#figgNRGD=ZRXD7cgkr%~>kvMc4~X;gX&)K>65SMEZIH zj1TL@h~jxPMm$Hvohi3XYFuo1YCnH0dlcSB_f4}lOmJmjUR-d?;K5u7{vK{2B=Unm zyy9Ep`NVpfuDgyeQDGHKNS!@QSaeAf0&C(K_iMEQ{|S3=_102w2EHb$M|Oe$&il4K zHG^nh`Hk}b<4rt%^`GAwT;7tqV=1SJoj{Gf7FlJRQzb_xSobaeP)g4H$XoT_zS(uG zQy~1rC1Uod4Y&v%)9s&Imz#19iy~WIOF}G)mn$1nvaQu1$yv6 z@JFEoNdH}jv;!Vcr)!#t71D>n!Bt9NruZlJi< zK=iYf0W90S!B}c5QOSRtF+J)NZMVsCley}7e;n7H%lM`sLi>^7k933aXCLvb&@sH^ zLN_MoM&q6qO-As=1UxHs9^Ta)f=or6v2Qj5TUE*-Gnd0u{Q&`3I`MCgq(Bb-ulO)I z{FtM!lNm4X90X2VAE0DIH7u4DfLPf);HuXyqCNjDHGEMC5A*-~X1#HHE3IqUYZVOo zGXOr30E&{5K+~5#5Yn*+bLsuJ#51YStZ#o=p7l!Xyqzb2!+io9{as1Znf#rbm);g* zFBOnu-~Z%|?=*WLi_~V^nx2L&n%?1X213?DZ?Mz0#o&8jHKo+?iddr<0Q7p>u$S36 z-0f_Ea`fo4FjG7y_=yWWZlPr?AZ}ADwzXD-c^ZLXzuREqm^|2- zBFvtY>G$HtjE?GY9CL&p9+fx)W~WE~2Zl&yF~z@UF50ddM-;k6Q9t!Qp|VIBEaD`H zJ<4w~JkMUBa$5FLR_i^$>+K$(NcAO3QgOq@>&+-DLj@4)F^+QkuHi!ha(LBuCoXpX zU3Q)r`G0GL!GZyV*NqTVRiy!sc_b13hlA0w> zp3ikT=Z(M9HW>_l0^AO9zR}Yla3A5t!wK{rFeYnjAj0lsh({BsG zkg6WAEwBYgbHh02*3CiCJ7LPH{yi4{P>enJ4{4DmC`U#?mLAUWKS$}i-omGvOsmg>- zYQ!>v5!ZH+(wZGjJbm@rAT@OpO;^*VjCJnQZmOfOh4m=(BHb<~G26B41bBE7OksGQ zlwo|k0DbtJpxgTYcuKLPr zw9R8S*t59LKE$r3|CiQQM&^Z_^=5Gce7p8Gs@}Z@9~V?-gy-`yp5GDR%ErvQZ;rb^ z9H9A)xoY6Zp=dPsjS4QJH{0GIGPsTEvn!%rAH0EPw@c#O zumb8^bTqZxi$eC%8}RpC#<>1UIXc+Cho+Ow7hJJk`G0)deim(->dzEBchLHAb=BFP zhmz;4Lx+8X(QcomC|=V5iy5!QVh1I$Te1;O@%T5_J{%-R__wB^msRR0-y{^(3|&C4 z*8b1?HCnV=Sz`4JH1<>%-M$fzCNW0cm#E(IX;c4CKL77|;?K+IqyxQ9Gvzb-DiEih zcEbznO|UY>z@TuDVw+c@UA5A9-InJl{j@u6vq?HDCYGX|m-nETYsGPPz*{aqCO54T zxQE2P#v!E}26%J*7&>-vHa@Y2AK#s~9_#fn@O`K4*r3gW_N^NK8B=m0UXfoXWu4OfzjUJIzt^3F{7q`n<>ygTvi;ZZ zJaS(bBlV*FT$?g~?FA@k=Pndtkd6eGIHQEc#glZI34EPM?&eP1#EifR-Wze^a}#`} zTmy%v{yTTat$!x=v)dt~ z7Z@syp9vSue(G)hZR+}teB>2=9gPJiqhE^>QPn%GDPQK&Q5^h0;e%^%YY!hgQMU%4 z5Q@i3Qv#5dln-uyD1aYFox`j%9;oP0A~xwI^i|+pR#QZjXc-?22so%L+MOr!IrQS?5f4CI|m&*Xp#|!c1y)3N3+kh&L zJVg5bA5h5MF#O9u1iww#fpymF;q4H)@yS{L+=UZ1rX=RB3`*Cl=!K zT9Wuf(^-s;RO1hv`D-qpNi4Z}Hs3iUFaC&iGxHL%OQ}X{LT(^NSSfnCE&=}<(Zz-9 z5a8Ciz((M8n>*f+bq8-ral^}hdmtVCU{ErX7uvAJsmi=Aw6RqQ-)n3~9n=~vEt_$uQ5|K~@*U0c zZz5JCtpT#T`jD1tE++C z`|sS)Bu2{=74XUPeQ2L|K0c;;@;~gwmBU)#9nu=%*y2?-x;nlFKN|dlYk9BXdy(gH zAZdvmGVh@xiy4?adlqX68sbelX24S5I`K2~4-m?EjJD6R!m59sqr|FUta#x({=7FB zCp4_Zg4%(!F8A->!M{4vuq*EXe&|1lt$dH;z!83&Az_7lH|wF!=3L`!V_3u3 z^vee1XvX3X#ew)V^B&s5*#f4qstMdL`^0$A@)(?XQB93II}(EF^3+Mro{DFpW$4Qq ze>kZtH@_9{h5yz&VR%XyM|@GmQD(AO_2LB-OhlmQE%JEL_cQ33_%SeB^eOsUqm1>_ z$IuOU7#WzTVAH%Mc#h5)TyOgd-;l^ftG991LpZTV_p9sJ@ND9M#Yz9C*rWfO_@}yn zFj&6q|KIqh92ppT|8L@-av(M0Vz6gVWa@fSW2SJejAxTNDxcH8g;6DDSUD#A|hUm?eIIb_w>M4&kz(TACXbiX?!wGwGCkAEsEY zt3h?%S$I-W7za6A#jhn}(c+zjc*o^`W3ciht>t^(6j3FX&ibhWi}YRBikBUbF#xh- zS>TaD9$@KZQfk#5grrm%L%=|QS|Pd%?P{U-HZWhWs|R5ZtY|r|J1mA8bEGJTiJwUP zP9?Ga;2nZ@ZXji4`vV;nXk}ge;f5;IUJ@q{O8~g)=vB1{^F)BD&?jqd#Z;@oJ+0eE8vfiYL|xg)Hr&akEz5LP>WX#^Wl- zh_dz?>XIOXwjI~@gVIlcN?tN3PPs>yopB%q_?mYD+uH_IXXhq#@pTJ6FJVvn%w)UZ z=0Pui{fEE0mmeUi#~m2W3j~Rlp=iRQCI~0{Zo_Xgj^G_f9N zUEpEDLvZQXoGdX29(7{9fj6Bi%8R#D#c1;|yWQCaLkpz!-Lu<}dB>AYgFOG*Wa ze%{PZ{wBlrZv23~!U$T|i^~1PkCU`rc?LOi3ME4a*rx@=X?hY_`=})B0<&WJcPWu8DTpfYe@X3Lhav5l5 z11}zz4kj#oXTW#OUQ@Okbr(Q7{5I4U$NxIdwzC}FbyCF(Y;AC3*fyN06~c-ABE+8g zq7a(~pP~w#o>D(Y^SO3nuDr|U%E&y~FNdyqI)cARHHhEw48hZ+1LOM_!{x@y@xb|q zSbRl!h1g*YBF2b`JO9Yx+lM;wPrzg^Mo+M1vk!ZMv6J1o+8uChzhW>Q@yM96r-dl9 z&tAU;7T4{?V%Lk1S?+!)*Ve>H(ltT78W+$zc`01=mzQ0z90{Yu5>{2ThqZMokF@?wPObR*hU@n7uVyt);`E!+cSVx4girJeyG#D z7SAr&H8rQQ-R~Aq9#zFAMFF_tLp@{chyX~LurLsn89~OSko;ctl=fMzpEda0mR)SA zm$TT;!}H1G?4{_FX*gw3REEaqh@d?-qZDW3InM4a!kZ=7xLrRUP4s_7mt%equJe{+ zbzNijuMM&6wvs9;E+bpM6_A9}kKQ=LjJS4IjJKMAPK{KbiQ<+sLG#8ZyQF z3@K8@OE6?@shO;Y)CHw}EG-s=+t2+3p=z(etCV6`=yw<%C>6s8K4lQX8{7#gD`Tkh z_9|q}mYKB*OB%kg2J-FVk%{t}yXgt@;ngZI9GdXhzO_SHn#G0 zF#vCaZ(yZAHu%=|mDsdR4Er~w-~jj*mP|Jz!$+OS!-wXP`u&}p`KdJ{ORyq#=~Hm%v-|5*Sa!)`0LwXYf!?8}{cp4QEM(phc1l_ENqc^pk>k zf$U1!1~C^nIOy$s+`T1-s`@$t`ai^D-@EhK!Hlms?DaVQ-Q$hhJFX(X*;hER z(0SQs_6f4h{1s@NB3^u>=}6^V$KiI7ctNW*KFrjnpY1WvC6{z7k$094VUt!R_M#Pe z>?%KNws)K|TPS`2pWkp1eK_%%Yj@^rWHE+{sk0r-V76eXTqn`vQeG>b{Kjz8Ae}9CqG8JkeY!<$nxaZ z=+KtutS}bB_v4q*{>j_0j%|L@4c|~xKr{U`aLIcNGiJ7+zV&X`JPrHvE zX{`Z08+cLGO!^)I9+ii;!8O?fxJX0*ZF=@^4M|W_5}rz!#V*atW!rf)!Q%q=;m_;{ zC={&@jAbvw=c|vSY3=v16ij3{k9l(Yd8U)gMSNcVDBbtO%uyr_H$8>FnKQ_gIUiw1 zmI`^xQ5?Q~Rty;Gy|8oIa#&v;LP!dJLtj_1aK>N^zR25(s-+W9L{kPz5B7p?4R@hZ zWh?AYxemiD<-l4WV|a2&C=5>Wfa`m!8BYs0vJO(R>??=1&^8Txdz5UNAINQ6%x6{y zs5@RJAa_)V{!RV#ek#7*5gz_|3iBUWja}NCV3bq_eBb&8dU4{RBpUF*R!uvwq;(DC z)<5UI4U36oQI8`8pv~iTK;_FG_;MeU_;~r>dTwk{Bvic1*>BJB??b)2!1Qft+c?}KoQ6a!pyZX%Y%CBZ_b1xen5B4$-7XD>@;!kgTL{)~cmQ5S+<_;)|A4pLm%NNWyjQ}4&OT3Im8=f=On{*K%=#vNM*naw(rOlgfzp}Kg}e+K zB1WKSt{jj`A2N0&>pr`Ew zl5NSu4f45o->);cu7Ck&?a!v|hWh0<41?7#{zzcZ#E}Zf`J4yk*cL}if>k?^ZRc^$n6eJWR(9B za`);LFh?~8p0D$T9xNmrGjwYx|Q&m({$+Z`4J@~ssq1q>>kUb17-75f#g^*NcnM= zmOpTC9a_G=9choqP~wq&OdDkqujdSg!bF`vjV4X9%B4Pjq z%iF2oEKg|gdl#(CquT)ww@g%y2~5p5ri45BiO22YP^>NpEq2xba<(DJ#bY~)KcNqq z@20{03>k84{1SwZEa9eP00vST2|?QgG_WI>Y+IyBZg5*g&U9Y@42^ui@Z9AzZWF)D zf#jSQ#O-KhJitgn{eDhF{V8%@i*3<9?C!^5gqX>7RqY$_*V z@5~BNo&NxMYi|Pewj%Iq;6n6mwg$`9;k&-yyhV&?vEMZQ^S99|{5C%@W4z`$t!;P6dP-1!*=lop(49N0<#}CX!&-Ycb>J4+yB|SE^;Gip> zraMmkdbbj{+r*-vU%aSNJqbSUsD!JZ^g`RDX0ZPJ5u%ZC6ExLTBabbu#DZ`QYI(&A z;BKeDiMhbpf&c0_O=Z$*KveUzHW{;WIlqa)0N>; zJxk2NkolnB4y?MThhGd{LbrV#@SRF6WW2JGsM*L*Mu^0bMQaIMBybAFzt6&#FW8_p z(pwmEiPsswZX1D|UBR%@Mj1*y4S@Dvl%e4;~cD9e}HQXp78JQK%A)y?A9^C`V!-WPx)Ff(zp<`CqD+Go>!r+PA;yt zSjyIU6-wLMeBCry(&_4V-JB(*ZjcVA~E2jdkA;g z7gM_&gy=Fuzx1&C;%q|VNC>U7&iPqr9zH_cSS>=V8mol4dlul@BVI&&qd#hn8Du;&{7AH~Zxd2y6&Jv_9b5ICSjv^Vh@VG-$!Hiya~1-u7XN^S&;MwUVGeL--Gj4r!& z-Z##foqvA8)oEXbD`0Fp$fd{JsT+%$5?A0w3gawW?|qz@8uDZ&zYr;m#L1(JY)FrY zeXyRj5!BCaM!tn#i2hg|xJdjqR6Zg|?g;tD)J@Fa8~jtHu*rx8hjC z*GE{6(n_#8!IHkmrrG8$5bj$uK(goa?76=&VlL;GvL6*I?C&{1|C(m#y&4zQ73N9;KW3{Pu&hyVAZ6$k&*i= z$gB@6QabVzF;=@8xn6#tTsD=b=~K(*o= zsP?{?rX5@`9Ud#50KI1~kvWSnX>a?M>~^|?HJ3v?I<_7U@3o>a;L32t`3ktxsEcRM zxC$c7w!jBA62K@ABJOYRO{HNhA%sUZFNbU41ga|dLqE6nT?TH%?f}c;8Kl<81z38F za|cTILNfo?DpGlm11aH=O6IO;Am2PIC)ooR;B3QGxb>kTzUH-<)Du@EHSA`Br+b8P zG>#VpBVmV)Ff?YgHjK2xb++CaAQN83`TWQRkU7DDpFBEKtA^Uc?EwVZc$sLHq!l?&tVIq^y5XE*13YEc*6b@ z))nKSG1~9jhd|3|JjsKbYZLOYdvyvrJuwS!n)?LxMjK((hc(#lS3DB9PvSMiP3)3T zj%&?_aNcKe_JPKaXl;5aO@D`D0Q=N}7@U7u1uRT51vCEK*p%{pEy9r#Pv(SID_2jR-lGKormg_`^7_#D z*%^55a5?pXUlpv6UkwKx3XpiK1GE$z1_xiL;lwA&EYayd7%zm6Qgi-vf)lIkz;BU5 z__t>OHa4k8!_J&JdcO(zG*a{bzzRQ&-~gOOyztw0q;Mb#2!xy zP~LrGbb9(~j6E7~=ZGJY-j2~{tyJo}&H$X1>jjNJorD4Y@4(q!l52k^SLOq6&*JV+ z=Ws-7FuHQ(HE?;HPZb+{BaHa|^@*c>-Gn=9KV5!rH9^gcJqzYJIz#p+GqhyEN+{>m z4(e|Dpyi@s_~|lT+;#8*+FzzU>C-%0mZsr1a#yhR>P`4ak1Qt5-DrI8%yXss5?$!O zxpHvjsp=60EiP#wL7PO9bs5m$R3Kc23(yiXN37hi4$leT+(WfY9F9sY1nPAJ5~zxY zB?0;HkIOw87m1h>VC%jWR($8DMmL$G5bHYzyL&9q&S)X}_i9ml^zQUUAZU^f^H)AX z+YXAsV~?(}^1GbqF%?&i#XFc-K9CbrBeoLkxoI;cpX1stu&bgCR z^lZ^|+z!%DM-fD-IjY;(16sOt;F@)5aQ3{Lz-%C+LQy%Mv8V4Nc-y)j`nA4?d&I+` z?X0_S%Y6g5amQV_EHDE;^Sw&r^+J3V;>sWjgvrh&$#8m@EIht?9u7KK0e*ET&Ot{dYwGViDMN-+0L_u&Ygd#;}$ls zZ5tbZwAqMve0WRT`oN~!&Y>D(I2Rv+#y6e;^RQCtfZIlDz1jd_84GcsPXG=wttC2D zB2a;M31$4^6WFSq2;=StLB*(2+W*!U-GSsOY1quanUVFxo-t3$hnkqp3sO&o5`#;B z5T9OMht->%RLWZ+Fpf3 zAL~$7moLEerHA2`tiRO9umA8YxO|m5ng0ZQ>`{b?&n0Pp@KnzN-oGnZiYdbQS^v^0 zA5o6}3XW+kKvc#U68#f{YEEU~ZD%0vad$<1@9jXtg$$S|Hj~putcHTBKjHiRHuSi7 zQl9z&b<%X3E?IV)mwcLX2Raf8B>#{UU1t74SxOulAvv9yq@Ttd5)Hk8q3$v`gSQLy z1p3fru7%Dd<`GYsu87Bpl6{Mlmsug%=q8(Xetqm_(aC{-@3ia{Q7LdJu z9Ly9FrhU)pL?IdoQAOX1w!vHp67dzhd7bIllGd-5aj=u8H7k_LF8;xeFMd;>>|^r6}n zmvN`iA?V(JzhV>n2Ut?@oQhsjL_A871*;CcV%4e61xlNeh+54mpxj2EC#3y}Ye!P8 z0IXm*f+Y76aI{AcZWGZ4B})JL?aTQx@c9Z8n4f4}@!^Oz4qTu}`?XH99_)4wAx zBFp5q1Fr=)VblX(IB#n?v>DHYPh9d~Z9onPi7^Kt={#+()>17>^2agYcsdze*<1{i zedB?r@J*JA^KRO|=j@TB{hcdQjS&T1+6XXsriYkSmq}Qv8lm=`LqN6qI&~&)j9_^G zWKC08023ZrLzN}>!Rh0_@ym}_;Y~Lya@Sl-;#pk~F(~zdu<2BdlrwlJUA>epRHs)g#2D| zVz3r}!16=?;~p9=zXV<>#LXwuCB(x}n`MrqTO*N{S}+ba+=Z8aVHgs80VM>-F=>c%W=YKZr@wfqa(@ zx%klf-i`WB%g}bJ8l{i$u_vF)y*bH=lTLwZ<18xs;U$Bfy8vBu*-Dq4e~C>k|1Lt? zPO&7I&RG|3)8h%&eIsP|=Ogj(a2R|fQ3bzGO9Sh=o}*F7hkK0E@SI*lR8nS&&rCN& z-cPPDR(Tu;ci{&h=RN`L3*Nzp{?Xv?*fr#v6@fyOe-Lq95K$6lgm}nBqG`q= z>aJKA%IrQyxXOwVPQ^Q@{nn-6?T0Va64k}jmEai4?a3Rt=2C*iCeypJzeJNr{s#a^Rqy#mNw<^ zO%~sKeR;|@l{|T1wcmLnCpjC*R&vfkolyY2w~B}thrXfVWJS(ehZNE z{BN70l##!Pw@j0^7nklI%Nn@9Paei3JOGZTY8ht&ZIE?gE_h=93&r0Q#uonK;O5s~ zNO^D*9=~dVukGW*^JdlnwL|k@r$r4ut1gT6GOIym!*V29v>q8L1QKS(ci_U*cSje_b?)g}oOEC@npdNN{O+0~ zUyXbU#@2$4-rvA4vKY3Q`>{8Yg6t0mQt@NmZiepzCm`GD1!f-31w_FH15e2u?HOw5*p;utcab%#7Lx`=YKA{;oZ&T$xv_z-`f}?(A}S zcA^QK{3!?a`Uc~{=Y>>%4lkU>@z=e`F$edYt3;2tMj>9; z1%!3=9klOeKk7FqhO^Zgq4c#xXg)gy4tu)883!pa)R_X*=5`ZrI<4X3cFz4zd_0s) zcRmoQx{5RE9I(4#Gpdnnz|TT-u}fk*7)-o{4)60rUe&KC-A8TEqxcehQ8WXDapFf7 zpHRY{<~h_jpBFH6b0LoZ@C83gZxTvT4~T9#J+L9j97t;x=p*Du5cQ*MU#Q71-L~4drGOv%Vd@%gP(4;}tWP)y7x&>ZyYl zL#J87AvXB6U?y?HuNQp&W{UTz-oZCt4^vH?bKKIAHdFpCV7nQ(ey*qO4Shn1YY{D& zb-fHvj1F_lF#i_w6EBaj={bVQcJ$wJX1r}ZR(chVC&cFCxnaFvXIBWiE)fQ6A9i3x z`Dpy=WC~t6gWeBM_m$x>Zm_IB8HpX_v}Y41MpJeevb4QQx6>e{6?Ff{8~kstL}5=H z+5(CxO;a&~%QK|B6NNt##^pkHvC2#}>=p4FJxx17EcHp6nup<*ljSX7UFqS(l~2e+ z!;hz7#eOxk_U}2it=4t+{lkT9SFtFTgq!YEo%z8sxFp~QzULnaS?lWI^Pm_wuuPaU zFS8xAoB9!_Id=~5^ygvU#!-AbpAVPvAH*)l-r%Z_A%w!~xzvX~Wg5?+@%dEgYXQ(G zDhH#~`+(H!Q4sgC3uzsiiD!1(pm?XPP$zXI+-If*9;ms~m>)Q_5PW?o0_Q62g*~5& z;L8<_Sk%%VGZbU7!lD&A7`&#;c*(QEkZjLO>jyZ3E@5xPIxrGk! z&E&`559J}g9wt8OW{A_w=3>#5KJ-F$9<~^2rVjQt!{Oq^aLX==mRnnj7n!=Y0iS$v zu#tZ=+WzhV*qkHH_{d&~+5%=`l}+v7&h0gDjeI=p+EtI2W>UE0%?12%?iH&2nFijH zu>{+2VyFL%XahR;7og4+l9cn2?}T&5C{T=&#v%i*(9YYOZnJ64X>jQRN%HB~IyO2r zm+frJIfK=z2p@F_!sT~^U`XUTTokbd4(#s)7sAiLZBENMYs6Miadkgq67v!bCdwvP zqlUsgXxiEvXz<7?{IfHc${N%o`A#dai!v_Y$aVT?<5hn`wL<}d-VqQqS09Ai|HEyP zj?#Qj_U8qg*wYn1aq(nk>w04k$!O%tiOC@HHK6C6T79FOpZfgBYKW zI94QV%V1^C-onmp=A3sOC!o&8aP(wbRE65cbYPz4!S-3S52;7j(AXMR`&GoZ)>Y_9 zDX?>cSK!FeZxy4sm|Pxxh1@^0oIDtC4aywS18)R4_qjd2MXvuePR4HMTT%7b9tsj3 z&^@jS+O|qUv2k7)tDXnc(h*6jeEffHhgY^n7sHt$1q0ON~_# ztN||saLyAN9A#&1X5n!D8z9N1l$c&WKousZBkKVve0SDo)Nb<oo+Nk ze;<8>YBONf1H_{3zt#ti0v zL>*eBg1kk=kmmW1plL}QPFr^AJWwi%t*Lbt0-yjGtc!*z zKAB{f&02Eo&GTd9xGoA) zA3vZ}IX0Q2P()_m4TD`zPnY{fJ}}_=6j!#Vjz>}VDT?)Mvkv>4UnE+%vJ0s27XrUE zF7RoMCDdLh2^ZWFCh*NB6wNb@E#V}Go5KX)+<{4QH`axgwsXn%`+q>* za{_o`YXvAhstncl#-Jsea`6JOSe$s>gtpD&^YUn}Lj$#tCkqOAD?;fTW{@ktQP^c* zETs>{1s1|Lq4YR_C+OyFT>e4XLT5B(hW}PSSW%de;opM3xv0!MJF%QNLp98aVKH{SDs_X^xJ&7y9 zUxD$(@9>KAMY!-RA9}uHFG}9I4;`4}0be{AW$fIUPPZ?vd&r(byXSvIDY}CzxhCClyTs}h#F3Xw) zC5G}SEytC?40=4w6XGzR?!TGfLcby(5?(h0 zIA`p`U`EGM5NEuQ+kTjeSrQ<3Knec%)(SE!p1@W4YpLn7g~am3vp8`y4Zxw~tJK#r z5t8|99%PTT(sh;u7IJOH%pDxXmGcp9`4a_APe_uFD$+pQ!$OM4y$$Wc2e8Jvqg-1v zxiWTin*x=!*YNbS!+7QGA2?t<-$2`v52cHDqkFzuAa$7rcyl)#dyf~g3ca2Z+n!m# zRHuLI4=YB-X#Dpm>_lUU`;h6vde}Sm5Wdh&24*e`CUIaYlmZHBEWtv7xAEDPoO3H| zJ=)j46wjXOE3USlz%o}uu=)~pJext_$KM@r$en$3IhT8GgmZNm`oy_qeTiE)>Vp>; zeO9iZAX*i(SDnPUtbcpTxcV3tKx*KtGmdEaf#mr-@hsil_?qw>d@I5P7h1?*fzU|2 ztFI6*%QD36<@<5EvnpP=paku>v>3k=3_#%~zBrjp_fyPQPcCC8mD6~bbA~1E-VVGg z#2;tJ#9&$d)i_?>5BL3=OFX%fi2M$vp@eu-WMsL4#zX1(d9>n_2YR_d5r-|IW8v~} zWjmU^A4@-K2Y+wGG(LK?`mu{$>nTst)py3`CjUD zxB^}rx)kdzC`0!W4DhYvi8!HG8NP~oM4Yp*LN6a}0Yh(B5`O*N+Ycmn}+*5N^wL-8FHzvpJ?|yepS%uf z9G-=*jJ%~%bA_h*Jf&-Mu>6uV8WXPm+~mH|iP--+6BW%KM+JG^s7^QpnH&iuF#h*_ zVW zJ@=0aUbAd3mbDB);!XQe%dQ=`tRe;51o$BtZ7XzMGXk(Zzu@bypW*i&E;v|F9ta;l z%{cD)fgv<%flm}KqIG=~H5Xr2i9>uAeHgRc8Pm(>6PHeS(mH&!J&T-8m(xBgQWZph zo2`CsM6G|rq+IL%#WhZI?x*C!jKIMrURH&GH8r%mn_`M~5-0hk!O{*@e52iw{Uq-k zJI$J4=NLbw$^&iDUYk!y0mVW5VjO(xS_};~AZ*&SjlJ^nD}4Av1p#go@L-t)*4B!r z7MW%d(YmLp#J?Y?H}B$Dy8aOi<=?!(;nrPB(sR)i)-z`OTN`-zT?opk2XNxBUH}Pg zA#m%ob@0UAh1@=p32Xnw2qLb&!4dnG;`!st;F)cyU`~7n*jvU1cU>H@+qqTPY~=+w zxF!fc+8&0B=7oYgN9|}n{VhVU>gy#KX8sW@85T%tDM3aHyP0*MVS)+>qU{%7HyM zjBYwt4;LW#AvA&$|hi#nu9{ZS*aI+$d>w7%H zB`esmtCA|Rv|EY0LYd&fN+*~ed4W-#6^Shd1JKglx2O6YZuxzZA~+vD z>rmq|Z7i1V3?{K^teYUJFOO2^LKx^`*hi@P<~)2AOcBbIBG|rQ1*{zK02+HdvInhb?J{8v3k=w1iP}aTvaX3IpOeI1Ts`=Ys83{6J6VIWT^`9efJ=A8UU?eq-p*pNrVHX8=jm3!(bnc&;AI zY==wlS|p`H zi#93Rds(t1M8aFP64}$jYwms7q9_%qM4=s}P}=w3yvC=#kKZ5jnAdCO-gD1+o^$Wa z+F9F?>m#Dt2-i-8p+!wGXv6+vh;JB1`N!noX~X9Wa_4TQQrNwvD%1p~ z_R~<5Jzg9}E@u6FS=~-aIFCFktC3jd6d|1+DjYZZuCOr6m;~N;#dH6(;R%aBps&?@ zkW3r28&RFbSC_K+&z_Q6HdbO}Nf;z-yu+({)5789lA=LbCF_I6i9;ZJClX0p*=H z1Kxd)W`6y$pdP5QPPVmNoIO5UmcM@5F>Y+sS#Sd_>DWSzL(@5#rq6n2 z)%WKIv$Z>%_vE+~?|v_j+W+b(a=n(z$t7xUYi$);ZJ~u6OI#VZ$MNXPiqkm3eTt~Q zs0}eA|2%=}wrbp~oZ0;fH$xc7-$o2CcNkS^*v@<( zHv%i1*hx)|nuXOyPOwpD=Ka7x(=gwR`k^X1_F_@irv9%HP%4aQ!*a^P2yz zeJ748Q{`k9@hm#ly*`Pr`V7Ie!Y7We8A&3&WY% zBd$25G5`;oT!Ra?r{er^TQJ{T5&bP6XH`YDKfU&k$G_Kgrn#e*@@!s}x5>Y@3M%;j z>LnLR@Q##vh-Bbf4_#*UU{y3Pb;1lr!sj{qbb9dnyY$9M>XGvh#zm@!!{(P;QR$^( z1zt|+DZ!di*U4)IOllu|B<159;ECo$`j)yAUBBfQsD4qSKP4;C;%AFFz6_^6!M_!3 zA^ZnM{V##ZAK8jdIMAgJHZ0}j6!F}3T~Xj-G8ErgunxCP@1`n`>NEM`f2g4@>^o22 zXq=Jxg&Zo_Lw>qE;(b>NL&gW^!K-0k1;*xLMEbKDnQa*;SYr~;%%7`6m2_otW7#gP zkuYJvh1nMq@(-Ugd#*CysRs-*!-}j;{fJl9e!^ApvpF34M+~J7>g1q5*YBe3D~{uD ztv!^hx-vGk-p1$~7GvY-6+98I!diqFqcNq<;SMDa)!CR)XB)8%3BYv8IoizhG<~@= zhQ96?f)w&uUx-F!ST(Acgu4BQnf77CoPVrzr$;Dwy^7*^yr-5S(h1-6L^J3}VfeGS z8VylQ0q5QmFtcn6%->Un7dy;i#M!9~QQM-!t|$dRS$poEE@zLd$Je*BE(A#Z^?(H^(Ub&eZ-@a=tMKbey6S`Ty&3BaP`{{6&^`p5LHi8Z@tH8eJCzAE1 zZ?Vh^RfyW61l!`J=zxA{0Ew7?au0h}jV&}z;uVI^ah66gI<$WzY-*i~GMg0513g}& zEgO;{{p~nOD>7cA>e?$&dnFq<-f?#5gysM8H+If zdrZBYDzZJXwAqw+j)vD2LhtWNe9$E)0hs*$j}C^ zs88bP>{&FLJdS?H*|+Ci9XME*Do~P?Crx4<$gSZQCU1_TV}WI8mC7YXpSKHTFR~>8 znx)i~i%C%UZ5;afKc8c*F+9pgT7{#NYGlaqG~h#%no=V$WMg(&Ne?0W{>#8WWj2FM z@$b^@tdH@ETU<;Hu@5VUBE@y-RLJ-X^fAQ%`p&mSw3b#3R@Rcj@&~#1BL42p$>iC` z6=eJscR{zyN&HZDHYe*}97N}W7sHrEEz*SX3qyzepK?5~*ZF1V*9nQ{Y~f_}X`Bw^ z6%R5pw)df~_afF{*VOroUNGa>cQIz{R5nH?o2PDA0UIpFnybHO!hfjpjWkoRGoJZ$ ztDD*Jqz0FlC~Zw%w(Z}rj&Y<6+h!6kD`{>JNcAxtOQ(h#}{->{G%@t%d z`hYs@o5(hwlZ5|E8ZsC9-P7{GG0A}b%ZKQEF7PJQ z$i0NlxfY<>F$E^rKc^&qbN-8b(f!?1Z!x115AzPiN;3ZTrvzsY261DD=-zTwZE)iq z1LJ*4v~KnY=$fC0s$c?a+5Z`KPgoAPxsGyDI>eP9ufGXQo;%QaM#VHgIf9lNKNKb$ z9gHv2kS9%^-FQt+Xp#i}a%p-c>ziXX_cW<|e~H?{=IOk+PZg#bFQAuQ9Q0Ar(z%FY zF9zaqZ0@pI&->xH%tE@nWdxmgSe$ZnUy1^PZQ1xbL*elK9>!<gWk)K1ti4slwBO*q%b_Tc{Sw z>~RA5XU??Bk1RMAzLowRefZRIebNRp;)euA|+kKI=^8(uEH z2m{6Up=s#g{->uc^0dQYHV0(pB#V0}1??}*pd_D9L!JqrQEH1CiCcOMKl+ji(;be( z2AO>5{S}4E_vmwW16K>so8}f!Kf0Dyw<<=Nc@*-o)1sOy-XmkPk+>zyhWYP#RU>ar#)2A{K_NH^)$8d{6UwQjQ>hO7x}!y7ahVkI?YF@+g;g2^TVv5%cV z$vm6u8@u+7LApj~&{nrKIA4AUF$!_T7seXUmkp)pCjVZvsc|KY6!(C0EkilHGcqR7 zhE4y$l=pFLzE(N3{;3_~_)rSW`fJf}gJHq}->0OhdX(@K6)v=vnJUzJlS%@wUnlVu z=gB>#-DK5CS)nFh4#Y6y@LbD9@cF$EsY;9_O1j*7wb;eh<7V&F2KTXxY`%!U z6rXY$jQiGU_)Mt$F+&&oVfAs+nl6R}|jOlM%|Lu7LuBU^MqX zBb4xKo3N&Hl<Z3;l#Wh>Vu&on>FJ3JdW(-`S zR+);U&iBQarQT1&%^~;kwo8XFFQ5hW-oAt46Qt0Yd=HpaZimjdjwDjwj{F-RGVaEZ zw!7ESk3bFBwAZ$*?T1xa^+e6GxXC-YapQJh5b`pO6G{s7h0j*+5YD~J=3^$anF{*{ zIC@C~a!~t-{gT6QX1;vc+ZP{&uiYAjR(Od}=k00y`|Vg>!I7CLlOIJntN-Tp+__6S z7fwJ2W_iHt_(C>b(tLbVyaH><8xe_PcbKSEQpDYT1nKE{AR24=RU;JHIPh;o@2>nG zTYK>Qi-yeZW5q<@&siu_^qs6#t0431>V%i0-V1HFJrurMkVk?`syW%x=V^eof&{eb zjzl+ft+;CwzF4Ezf6hUrLp5Ahc7dOX);NImy%y=~RjEIzxGyhMXq6R?NV+ccIzL@_ z?!E$PtAF_Sn}KhdFcC(xIRkU4erkx+Lg9qlLZMD!oKS1X5Zpfh32{=U2zswVMD#?* zk;^|Rqfxi08s-^uFGi#6N6T;F^t4K0K~#XSJ9MKkCh4fKeKgG&?~g=t*!-%a+5JL! zhvYe%{Y%3|asWFBvshn@7Ci3mSm78)Y0-Utr+OngG*<=0+xpSD?a?TRIY0vBzmxOr zV}&NN_nC1`={Rt24cfW)rtrg+pTcE#ZVF{0m-Av`wYYDYGhe@<;vcioKH3EgY|ld6 zj1KgDHajMN2qldx9PqK~gAkOw0J(YvLv>?3D0>fw*9)#A=|jpS_<%HVX-O9(`Nnd% zW|@Xli{2JfK{EXi#l9EsG4W^fTBL)~)vrwIkr5>QLIyINlt2}wxueIHbx162gQy-q zygdzCU3P)5j1owczGt3?OLP6(LT=F4LWIG(&Z*>Q`w;T5d;ztA-S@lpxhLH9e8u?n zg`)PrrErIS4zYTz93Rort-QHkeX%#Q91XAD$l>Veqy>vsmry^hCSyaNa2%$KNJkiE z&Z{rMmaMO6z}a`WKQjjHlCMV_Q!DAy5f!v@eJVF@Ykz1&vh8EwY?B0i-6Muh`*VW6 zXt5o|>|6}{yI*tk{Z=KS_&xW~PK^%;8R)~4`4dU~n>^&WR)JWxhmpx%NB=#~vyJal z`#rBP2|Lz6tX%=Kvg$sfohz}qm=6wd*Ftyq93@rzb$E9rcX09){ak`i95pAOO~aV7 z-(3)-jX}xryFl5Z4ew4%A-|&J$fGR}aLB}AFe0xS+dD4jU8AHJ8Sx|FrF0tj4?Cf} zZyFotCIlu&4bDB=x?BT=vJU89o(fwLK{bwpIHhQ70`$&g7HrIh>*88m`?t3{VBV2H zHU~=+!@p6B<K=jejLtuYJ=bAi;S$H69hkf zgJM0Fp;c4Ypab6Gxc*}y6(+&v4iw7K3rGaouM)+*v;6HJqVwvDh7g=y!P7|a=Q5%Rvp@oN6PAwIk_Ps+ZXjs{@KqZc)a8vbb0eWXloNPV(cE{ zMNimu={{4s>%9fN#kdtp7Trbe-MXk~p&7bya2o6$Gq``(W_19vX6Ici`}5G$_?0m9 zH;v6JFNyf#i(6G;ZI{W6zkxCNUt3~LyJ}FI{8^|BkfE(kXpwn)wO=;F&p3F_c(i+-@B8}8U}R0T^aJ&UI{zJ$s+AZWlY+HTj=G?07}^}3(JiO z5LA8o%IWHFThjH1;nCwh^qW5pXB7@Z7hg<6k#oOO$!62xF@`Fi=4?i_ z13Q@(YfXwDJqfJ$(CE*%{j`2>7L9ciME1{*=QZLbqzLbmP^7od3q#{=JP4)s5EWYO zM&t?~#n21z;-~AdsO_Z>b)bF5CTMg%1Jjn8(f#qJw7uT|*sKobxjKbWhaFy{1G`6( zwfFP6@qGGQQ`nHw0MFX9aN@zS#3h6#9l|{97QPX`Y5ENdW@jV!85+daISk?D%h7k$ zbHvd<5o;x$rdq1h=!yC&wA3#p&^NmUPA?EVdhrlWt?UDzNL|>xK8TvVBA>%m#M^}# zQgq>~TsXhJ03Oz_1}*zWtHZ(&D6`7}IsY7xI4FddCUgy6Uii}Z0Q~Cu z;d-GqnLkni;;ypKi!sL;m)xVs)**G*Y@upfl|ebo<|ev#h`1XbE<2RA)1cU-AR2I~rl3vy4(R5x6R_LtBD~(V7H^r5#R%G;;5}Q$LBGTVureGAo2DC}_7fOYC;mYXD+gc# ze{fIU6uvxUeXv8jO=Xb6o(>>S%RqM7Cpe?)N9NrVW8;sT!o@g0@@(J=sb!>v>rH(K zHW-Coo6SX6UQZRiGOZ_iXAhCHdP4m5yY!t z6?oyM%_95b|F~2txOUS5O{>%4>_H^^{|}2w zO`jIn?T`jRNkbq?{}wk^YWm3hqp^hMqi*k4c>aiPd^2PKIfa)YJUkPV4H{(bOday9 zW)rS`s03o)Z8$#CWHO=TQ!)&n{R^TFhKjCLW5cD#7`37KOxLP7h#c`B>?lh3XFI+3Dkulq1vUo{9}*UH1g>W$U1M@^~;GqQ}|FmiEKARk<9QOW#6! zJlH&$S2LIqW+luLH**MGqX*AQ?bxNio9IwR7V50&w6d;{BnzKNAw8R5w9(@;vpgh$ zOsttgvNauvSC$G)mfguqzhr`ao~B^Ghbhn&(e;mKk<7U!T4)<(jrz8zqOu!rn2Ao2 zIKRh{nCQw7%HSdXIxUCvogD?GEjExVKLJmRngxo!))2boF?2ofBgb>q2}_61;b$9; zUp8lp`t4txib4G&X#GAL>Qa89TbBAXom~W_ZZa_XUJ})qe~jm0pO2P*nT@`_%b{lS z{$rLXAA%JxYT(t9O@h96rd}i?|Ah4mQe#y}q~5L}J>648e2TR^T#r_LYK9P*I1Yy; zEjQ7s14iKED1^f^2IozaSIOa)%@f!db8T4s&~u`{$3m!|fyvVgYlxh9Z3)H4et(7h$w=?YkSpO% zevDOTefB?Msl+8HqwFPE&(yFEcgF3HwyOWLyKI@xECF z)T1J6lq@hMx`)Hb(#CI$*o#&Yr2mEFR~;iyY-`9%9TRkK&PwX=uU0(lN*d+dAd4GL zdE@o2C20R^b9CmxW6o~8uNNZk77NDCakWt!QM$B_3`F-( zBAG9zF27(WL5-^9BRgw7-X!S z2ri}KWHz;r*p7RM{9F&9Xle}Wv)#bPz$hYVno+D9+e0R`=rVQb$!jVsr5C-}kpgjd z?!btRr%;-7I2^UO4hxEgLwV#l@SD-d?nx*oO77d)95JKF@1GAKZq;C%uba8YQTdfb z)OVzuvEUgBY`@hCk|Rn4w`KR0=8YfC;qu1x5>tI@9g(QojpZ#v!R2rW3hXbS-h>6B z=aQ4)tf?lN;z4^Ks2Mj+U!dFbyXgF#K{)VI;^N8JPDdG?@g`K;Rq<0M219R`UAx0|%#eXkA zza1n<7q1%n{W>8ns{viUvJF)}d4?7nI`eKU)`x&?r@-B2Ews*%z;gOV;O_5&t{rPe z+mG3Ey1%n{1a*FkGkRI@iCowh1s`1uL7^}QpJILVLdi&ouvLOk4|Z%^ID)Q1uk2SWr_Tt7Rc7C>%4PyxeL-%hst~`DxYSs8*+X`a&5#aMd5QF<&o(-J)G61fV#cb!KU~m81K3QyagS+ zTe|Y_^2+qT-@jr(YN1q{W;ND1LF4nkFHm%lH%>CP!Y1#!QTEvDDDrCxFEuHQYiF4D zFQz3d1Lrw;qbr40XjcDb2)?)$4zRwn-&StIAH_O&FN*t+SwPOfel)Mv1~y+Wxz{Ei@e0a zXM!S!!>Hw7s2Y#4q-FFeHr}QpOi*0`yA7UT@A@lDz;X?d?)hvuG;sIsqelBIB}32k zk?u>{!k$UG!ltv%WW+!+DLSK1B5g;K+q0C37=Ib*9Xbpzo2r4gNhOvV@|?&H{xG7U zI*o{Er5pbdi0N&@jc--S*!nkkX%QPseTS{+9ygAxOy=U>bHA6>@d?k@R%EaIV+MJo z^y_Y->FUOn^$1^tZ!aO(s6-5INK4>@2dYS|vL0n$*TRIGMB^S_H@)vFQ1FDV;?i3 zoRhUC3((~~XONAlEIy&(htI85!ZvKKhxu#_IT0p@=IO(!Er$5}5^vHLFq@qI)QkJZ zH{ee5;iY%Ci<7-+He`3kQ@p&5PYhn}An_7 zEH2omj(xN}0M=?F-r+gKGXcn@up@X>X9nIWpG0DF5(v_qMTU=(At9k=g4*(8eAK-h zmu*a<9u4Vatm9*;Dcg%MSrN?XsBQKvPTpIF;&^+_AKde5D2XdACiCR=$me@voD4_T z4X&GB6W_)39FFEN6yderJce2#Q^|a`*oqXjKj0JxZSsUCAcy0|2z3M0gmp*mp+=nq zs^NS%jxTvhDM~p)u~q<4?O=1Uy^ZDQ=yl|wp0qf4*l`vu7S5s`MTX;LCnw_tNd`n! z^&pA2@F0&n6Hv%0jI)y-3EEy3;{R^?qq=r?rv73f&SL||=Vv$LrlB8DbG#gB69$t% z(}O{0Ng=on`+#zfzs2{KOosPyy7ex4oj#`CDI9(747pDrK;zKhio{oi-k2kD_^qZ5MB6H+( zb|JOywlYTH*(AjG10xaP{SV))DHXWyel+n2isi$f*i4=zuG>+dDu4^IfwZ;YTb z@+r{2P=(%WT@@(m$&xram6~a<3u{A%z|;39fxZL&E1kMEHq`Agf#^zl+c5#Np;!-R|4whDWn_h5;%dBo%9I`G~!nby%C zjOP)%G5}wl^^4?(uO=a$3(56m60j@2l#@Gor5EKg4sg@*05oCrIP|$r42qLIP({yR zu9Lsm{8Ejk6qO6{hTp}E?-vFxrar|(X8Zv!pAUdm=0l3vEOb|w-8*=v8+~d^0%wCf z^!AA<9BX)pozjjFvt8nlv}YHLAP1pp>ltdXh8Uf?r5RRultb_w3-U<%B9@osQ48;z z!F~D)G~eF{E04BQ6)mQS<$)ibaGjShrVgE6I0GVvZWPE(p32y&c5^Z;vs#KYVH)~% z;|L53-VBZ*`p~E}6a_VSQ0LF{v6+H2ycZmTt&T;o{6iG!tW3fA18s12;0sW9OktIN zCV3goCl2|Z_;A>ef9oce@xkOrY9ZMz7s>FI{(~hI9>Ds4f~QY1&SiZo7H8=*B78cI zvM~_RQ!v^hL%YYFVkfp{lS><(AVI}RxXQnRZ}&|U$-vo|NS6AjGWMAz2R3e#>HA^7 zc+ZYyFb3}*2o?mg?~LM)nNI0BOi6$Wto@S%(ns7xGV=e{C&}!7!skH{t2`3cr_E)S z#5q%9CEPrYcI}ZK3s5Gh17oh5Ujsc&Qq#L=4_}?(;pVgy@1D& zXJEU(_P;iUEn?Tg?- zu5zuQ_&?I9iY#xO;{S?~n>icx7Z3U+yvf&qx{7C%R|i2plkL#xVZW)`fQcx?QU<(K zWpH1x5;8kCjg8YQ&XYAg2CDNE=^{TpTJmc?G^ss?n{8TPq%a9iG7Zq$HPRkTMl=#l~(H> zOo5_Zt*oDAEz%(=MjI;JK%+Z5KA};O zd@RzTm7t6(JN2;~I+(~u;VNs$O&4==hNp#&e_Dg~ZJEa1|Eud+8U3Z>jysF$id`sM zDbgEXEJBi5mlzA56|&%s`(UgGW3N(la{p)qcH`h+`&dw$BZE)N$1pQ)C2{RC8)6{I z#|-+cwdlqJH&Ig@4H2n%aJB0jn|mgZOewfbe&$GVHW2M5fX|Pe@EYZ{WU;F*$w^S4 zTdStiY1yOc!~HL)*r*HmSC<6&%j1TpLFo3lGSqNgh);gW#b@P|u@@DKhYcM^kZvj* zznKN64@vy%Qz98U6V`%4ff;?}XeU&P7s0e`qdB<;F8bgFyXN80ITw)LNDK09VFSK( zsmbb;!{D6E^`s$`tx6ENp45XstiHs__LqhZweYXZLad{F4q`>2A9hWqudpq7

c;l;fqjAGzoykH0FN$>A`#pUg%;u!o!o@QK{#NL`e_K80Cvu z--2p$Nks_Izp2 z4T0E>^E<|BEHVqR*1Js6kgA@ z6+|mv(X_maLiw=S!lYUs;k)}F;ZUb4y-PNWmuEf|FvjZA&YPQs$Mu%NY>o&Q#myDB zmz@=+nI($7{~Gg69czT0*|EaKE7t}7K@oqC*b3(N9@1fMS;C(5bisG4E8mv>M_imx zE__X}7G(Pl@vA0j&}S)=kSxv?ekTnS1{n+%5;pk@KG!esi!aCt3#+O}K_>Jb&O`}(@xYMT}U(m3q9DdV+!9vbH@L`Kgx%*3HgwKa=iI1*+C1mID zLQYhXV4u=0tow067@{;qh<))D6W1mRMb(kxH(&BuL%{7 zOZW0i)*a^CJzVJhB}4h3Kq=wiwq?SLa|L2kuPfy4=Q%jE=$)|4szRuLDsgE++&lhH zOR`XD*)K#EC|!CaGfjAeTf(`7XTqC=Z$f>s(WTB;-F)y*8Np!YR({{$b%LYaV9rM7 zI^EXbDj0P2@wy6*!dLAj{OUbJh2k?wc>Z>8dMiSXCqBrHX~hhTe)BuQY|6umQrEH&ghut{mahwBSqo z6k7$$w_C+7hr9Xu_)EOumic_{ngc@Zutwqch$>ufwG&2Oy)N9*|0!&0RN~AI^Sty9 zSN?Li5@Q79iQkP*7oH#T5!zA@^39(|3sFIlf{J1~Irm}_owjrkU0J@D->K}&*ZAj& z_s)&VVbF&{eR6|d`4%73{ZD~`zXq;-XR_|qzx!mMZdg7n)nf|XJ?zc=U}Rj@G^ zlznxCfKOY=f3_0B&7qb;lSY6re$-Oo{6byf+R`*KWVA>~PEp|vld^av`HjLP@=K^q zHNMn)_nDA5Tu-=JX)Qc1(-*R~Ef?PAY!aFm{}Ga&iG<$jetvxC8W`_*C~h#HBdqs) zF8=;KNq84}nXetJD%iOj3&s0Ag*CG(#M^ZC3Wv2g;pxSbLQkZnAdH?W?5vR_FZz^( zF2%3H{?KE>!T4u}78r9M8xQh|;|+v+NfN@}iFHo79XJb9kbQNVAF6bd-)nYOco3mRRnDq&ao%!tM9?|z@N{j_)K7-A?6V~|=aeqD zdtj%y|GYP+kX;}gHu4rd37RdsX&x%l^tr&5)|(2Q2W_}}t2IR?^V7Lco6|Y(>$SMx zJeZ3Om?rEwe}!wgWkyxDoa0hFGDIa4&xv$vbws!0_i}~}8)@(@2`*{JHEw>`9qy&= zce=3tC!KQmH7zu0r>Vo*Il~$Y(eYa?Tt&M;JEvt+cTobr>jZ@7S@&pI!)<<~)e5Sd zYQY`p)!~mUT+gFiOqba{6W)Ib<+5M*(ED)*sl`EO&QLXmH@(_qm8IdeJ@qF z&F4Y}j}l4HV$l~hBav^sm#8EvL3q4X{*r{tUZLPiJtsceL%Z#5xfvU@MHzX{q8-Kd zqPDT_oY`icx7GINPAgP%Hc|&UGIyPjw`K?Tdajpf#q}W~_gy~1*;j8cbTG%o#UG_A z{1NVi{5alug@UNF_!}4a%|(<)YPh4Rx40R((ZaH$$(;QS%J~Ze_{+`{Spl_<&xCjCecCfwy?2W;YEa$j&;uUzK#rh9YK%)ZgE zKR(D;fGFS(V+Bt+WdMvJbBu5;?gQ>k-=1b3#ZxbA+@cZO)&+Cu0L2lamfRYySdZg!-+PKST~QS+?^otl-#x}1WY61mdXFE$kkuf<{V98xRoJ$xIEit?m}y^_(!IglPUmbKYS4PsOmNM zVpuvS-~EWY_Sr?SdLhk+{Qx$3x+ z^zrG3$hCK;WZgt1VEF3i-qH zZQuFbm%SM0aW?E)iuvfb4``Z$EB|KLNvc>HEX;0b;GeQNw8^t&#W#=rM`QdR^II?n zvQ=7CyG~ZHylg9Y@0=hkFkT|Oeq+sB$Eol)J7t9viCg#uw@#uu2trZwEphz!US4H} zq0q7|lUkl1!(V@Hz#G2S;NpT`QV*ZC{3OPeoi`&=JZJy?a>-d@@s*Cb&^n+cbUhg^ zzO=NAFKHdj6`a=M4?9OuqlHuWj_))1`bn&>+LO*#Ej>pwK3fSJ?^Xz(rS|hnJihRa zJ!8dF=R6jhg>2#f`>ZX@-+T_QU2AynZxt0+78cOcF5AWXv)9qS7u&@3ap`>Y{W>8w zc0R5C{ezE65`_MaLh;q9n*4;3?t)e2Zh`Ju#pfUTCpOqMfm^XUUYzTHl~;-L6i-U@ z7w12VqqXUUg6ZETKHcOSzxsF#eb{2ptN-!mf2$qi-;FrO59KU{bd`}p>qLFt_qes7 zAEPYn`Y@SFipTQNDiipYkl7XPiQ3dhXDj!kZUud;EhDHUABEf88+_(WJz?IYHauf2 zyt#ea{2a*#V&CRj^!-Ff!O27(be)M%!QQ+2t?F=UT(mIlNWRec^AW$=Y7d&2zp8Z> zkDIefk!*HH{J}<9AiqlZp_17`-rY&UKHYudhzSb>$srfS0Yk^Inx#s@xw*%M z)g`BdiXQ=V{ocLwtn3b?C;jCI2XEz1K2{ejx24c}x6%BYlDA~PyPa^$KaKx%w}$oc z%PD<3jrE@8g--(%xP7YXJpGj^#29bp>4kB0frGY?K7XFrJfYudRJ{woeMtx(RTj?g z{w2pBJCMlxFFDEIoWDw}zEg&mH^}E#tSzc=jepKtec2+OA@UVUB}{pzs)H5FiatVB z<}&KtrG)b66Ld?xj`(qlDNWlTCm6p;s<^SmMTkilEKu(VDy3;fpF2kgPa9Q)-lbRR z=Z!^t#hImaY;!w}x-nWX_xM1!o!|!?hk9T`059#%rsV4I_Jv2|pu=ASqk@-*L zX6ex8<1$>^?dN>XUju%|DGg!h@=eVBQzuASNC}gdY@@d~rt`P19O4t?hjZ$Gx6&9b zJDx8H;A!wmzJ8MgUtV^HKfOzefBEqVf9Pd9?3YaC7ma?%>&)8257}MCFQzSG>Dn%; zH(OO0Ib#JBFI?me1a)4&c0S5?ZsE5_s_~t=M`?Qbc7B7wDgIq?BscxnS>CdFq44yC zyioi3j(8B)%6B@wA_IKR@S!b{{E&$`(1_0C^#bkK^|^;GQz#Y>FSFq`9#a?2*%$H( z&V#6)D@D{>5e@o2T)6o1Ij^&_TznvKjIiy$nHaWzDWCsk8s9Tgm##V{!`+CUz&#H* zhq(Lcd__|+@40Xu_bnicWdq+kMzVYV_yaMp(ku|7D$JCzl>y6ge}d#ekdwB@<*FHKR{R&|K~(%Qpsz4wk!sPV$GrZST3n##W`DdW#H z$njgYO7Yh^s!8%>NBYRhjK*)uy{M-y&xaj)%`b}>#P9UUt|ZLwAFD0r4_1xgYZ4~$J4Y4s(x$A=;h-wc zyr{_LIy#6+?i6u|YBp}0W$`CJZZEGpJVCsB&l&N58PdXwr%NwZ^sVEk8@7YHBPV>= z`-zWQc9@^vW-s19N}lTXN(h(tm~jDT-|(|LgZSDf;e5duX<=5JuaGw+jaR*~jm-Y* z&Ofmm$)~>sxgv3!ekg6Relc&=II+S`@{RcHVGZO2t2kZ^*h7yRwDYZon!;kGAVH+Em|t~I@=})OenN9= z#d0UhvEtQrUh<_5KXB!9zV4M9y;+@zs6j<^`|@r|!j1Uh`b0c=jsX|!8z9)J=nG4( zmQ>^!+VN^fP#k~u6aAR{fENFjr-5z` z;^|dVSoHF|ux7w*;cxU2!7(#QJjQ_Ke1Em^>*T7$@u$-H=GGQcC8TM&LtE-PEP+<{NprXp0ftmo%yOeSaTN=dH4NZzS{FQ zJuX^&(MjhHZ~W;tUm3&R-z0|dW|4cb?9~=t`*)MLWOg`ztNl1#pYT{btY{LyDA8TW zjF%D1JxQ;4a&KNm@gi;Bd#WD)I#9$P4!z4yFjg130#vwpy8Ytblw0DU7ja~dv^BSR zc{$0pZX{D}A6Ix^Hxt%=KI%ke=kjwBhf->4gxJ zo`d4F2ZTSMRLz@5PZ9Kg4X0UGJ$UEBP+oR^J8}>0#rL4K80Pz2T<|xTA0s6vWYb7~ zOfd7b%g^VN%-w`oS7ZKlp_DM#JO?eo_7!?B8X^0smA+kD%vaw2&d>DzA-8?@;6UU z9kAuo`l@-A*ah^J#sI!SJ%SoG-shj5DCK>d@ABXLM(~3>YWdQTO*}0cNq^ZdVC-Q( zdZ?-ZHK*Oh-$GtMLuHQmyq1&@drp_u&C%h@nWK5jv~1x@{H*a(RO&^oZ`9Id-3{SaoZ4XVk z>*xHr0XBBr?KkK7-o9)uBKsK~=_-YW>z>?x*<$XZRsk)L-%c+pd2@XR5Am|4j92S; zMc@3Drd-+@R)@oK{#7m#F`TgoVm+^^<49!HaRTqkZX! z!kM(ZcosKwMIq;zXw4l9cH<83?4gR4o9V6se=g#)HaF^&H@EidU~cXeHcwTZM^9fh z<5ac;a_2{fa^ts@(y&hhIIr=?xbF}3xEUwSxz>L-nbFvaw#4qEa~`Y0Ec!lqZ}5Y@ zJz~Q-(Zk%rBZXYa#aPa`Lz9!9u#r1>Z2&jlAcK2dGnPBEsTP(aDQ6vJOqX0;!uG;5 z#G0BnXy5t<@zu~4`u2Dj{g3so#!nc+xo#Ec&xx{}x+(KJy>b?G8a}g}ryVD%oWh@P z-NUVSuHxSOr_HVW)GD61+LpVaugSZXexzQzW{@fqwjk{bRXBctbB_+EUM{zwrJ|CUsl$jcS*Uq%Jz! z=tg;Y&TO0%oqcaC$Nl{T#p6@ymXs0P$Dh*Tk#i+EjgRqMgJK3vJ6*`0+k+_la+%Xi z|3)Q#iD6N%!{w*xb4QMQ?-nmmeo8FH9= zL5||BQ5YxMZB7+>q_`W0ZqeIE4{$Cs=J7d>)!a(1hNk;Uai@0d_3YYc{WXbt+fHT@z38`6~9Zxj+>rkLH#ZHE^%m zHggZNGPsnXhTN(*A*d|zqHlYu=|0) zMXn{bABu{*=;{fJxmS|QxC^&sIQu9O7r#lG3-K?b?xCl}O}az4zwL<&|5|Z>9u;xL3pQ|?wrlA!(ONp}Gq|&XX(ae`9zALy!zE}ba=QNQRJ&;#V}ftu zvP6B9+h)eijMJv+Iaw8p<>PU8cL)7<^f8@zU>x`EW*+zB-x~gP@I5+Bj&kAu9#Gel zgLt*9I_~8a`9^l z|6p7#PW)n=#g`4@)D@4#*{ej<``*0@_m<=QmAwb}p75s?Ph*qCrsad29L67_$Mh7q zFZ%A{#4)P8X16o>n*5f}3s&R*A4S(4h~?XbWh9vqWkf|}MGEhI9wQB-sFWg=Qjw&+ z{94G~n<6xM!#*U~aTC{U0XTDOL`?W<9#>YVb`;(!F4smc(^JWyb_8$*Q*+}FerwD3{Mf4q{HC3isJTxN``N8`7iATJ*b z565qTvqLgrSfM4HiN6kCe|;rNnjRoO+?<}(kEK%`0+YN~*~Zy-410OjX5 zlb$+rxHb9!79W*l_T3lpoZ(aInBRfxCKjO15-`>8g(IiwVJVV3m!pNr)?IMFR zJ$BW8IIGlbf$Mz>07Iw3Iu#Mv2bTk>P8Kc952J^vDIHfS%VkKI<3fv%BGnCv)>Ys4 zK)hiYdDhp8RR-ErZ+|#zulx`AGg_RkP9*sKI}Jf0m-$B*lJN4nGvrXTCy6dLL;jN? ze0?m16Ka-`rEPz((I$|6Q!!>@s#YR6TuJb;IVHz?*v<9shx_W3Oacpy;A?7zppEK7zcFdq{rN?d4UJV?pJW4D5W70)OX@ z;$6=I=Il_x-3z2xtd~?IB67$Iw)mIcs@K)nt76{VcuW{F-ixeMtvLo#LCP zm66MO>QqtF01KZiB-3v1poUG-T;tx!w5wVYUieqh#vdBoeRW58USfi8r-otEa}QL} zUqFWs&*#tjPr^yzQZVFA8U1H50x!ME1Al!vCgW8H@j`IKi;w@{Z)p!WY9!u44C$k#H_@JSXM38k}Az632Dw z5av*W>-IK4;DMPe=_JR3-fH5om4vNpe+3`ER?z|pZSJm41XA$;oa|o2Hz>bHJkkXH zJF>XgFRM{*R3X}4@?&lS;#ht*3S;8ii2sr0^y5Equ6=3%*_mQQWgj-f{jt{g`@TQx zGV`n(5gmmImJe&+)=7%S<>tT?%Q3L3tef;He*@h%0b5$I5El%+2ggpx;;7AIxF5Sl zK%FHB{y_pivJy45SzpOImvOKOguQ0Xu-uD$=Dcc!v^F3LEfa};7Fud z*@YU>(680duh>jdO!e64_A#u*`#xTgc|}IF24d^?NIZJqj;)`h%IpmF@#7{b7I1_> zUtR($jr&3b*F3oO@Fz_C8OarGZ2{ZIg=B}oMPygBvAq&A;oD_DDzhaB55Kp-GmA@5 z+J6n>nLK|rR3KN{wBjBZ;_hHlC5-H`VyoX7<=K#=J;ZtQF$ zZuj0X+-|o?q`vwd9y`}WgZJH{HBa_|;e$l@j^jXjOcFkyBEVo@S1_+3E4cMV#%%Mv zwHWw|u!E$A!RrI8c0(1e9ms|=QySfS;}(jY=*OR<7RN5f0U(<6ZTa4zR-|EYV7N|DzMA% zMyYKXz^b*-WU~v~JF6KIQ%}R>v*k4E^(=J!oeSd24A8u$6bHj~Y4tQ&j?<6DSU)|M z;-7*y9zQ`TA#VJeMJa30If0La5A5l;dQM8unEL1$pHO}3J zT1ENr^=&*2xQKZAeFt@&Z_63Qnv%EEzri!#5YU$YL~dT!;i?C}LEOY}sN2|rn$GsP zYuz|@v$Yp(V%*{P=-FIUdl$U_Pzq-p9$?6AMHto_c$#NH zJ}fdbs>1dLOZ0+nYS$QrAum9VyOUvvb` zeyee7ti0$^b{y8dctX{@KY_x5$y}C6B$p@5Oyit3a`tAUVZpvnaPGlWR$FI>Q=N}l z`;U|cso(8j@9+&)>kB?K?$2RJ-+-vY{1JF71W0}y zq*6O3;om$rbakVA`$Zl#(xg~>mp%D=c`F+>G=){NB4{;VPQ;p8sBC`{Dcqa}hYs(> zbB%JatL!Si9ug1tCTOCzj0`8UqZdj>JO=HVD)hxu1@6vaUZdZVQ{`G`oEzUd8JdMgko{I+NBnuoK^UOFsnu$MkLF^)6s z8jD{Be-JnEV7y%|PTomJVv62Mw*RXXj_#d_!FBhbs;rKlA3cQowCMyXInP5|(hMB< zuFu^G%f$`t_B`_jFJl^$m9|(&09Rc;MQ0T@1VL#XTA+ z%gtP;fZJv5*?-qHxtm#6VPaSfn)aXO42C%2=KwKy7|C3EGryoYT%@Mpd zy?{49?_u2GZQyd@4;kZX$9gZvaO~I>=%jz?fSU~7^XR6t$Jk(Brw&Q{GZT~-%;8ow zOTpu$SajB~qT_z55`Is-*yeS3pbP`ea>m72-!x@@vV14lCuTQVP| z4#vR+v4{ARmt^&GW0YOwe~#dq7H867!>85-pPvHTrbmG9g*R49x?FJTqiB?st0oyC z^H6nJjDVf1u#|s&bnY7$fg^Tdk2SY(<8P~QTkl(Pn*_dRFscB3K4_O0=yF$=q;nl*ZJ;vj0H*Ak4#oxwOsP>GLzEuDr}p{iJ6?wK$Ue?)zPm|gXQ=Vx zcD#Yo{AX}`Y^|tJKb(6fvz+OESAeGnQsDTBI;?e;!+qRSkXKF~QAS6@q|YfM5x zBh8j|&bQ%&SWt+|tmADDuE0}UR^sZff?w~xyRhv2O>*5a3&wfH2>RdWr0q>0$SExY znPsi8&0P++YUP1-d;uDENy5_MCZcCU8_Bvy0JZn>!BI_)6Gcpb^pcfWAb510QZa^D zxize<=PkLQ=gpm3B4T18etcbbFVqa(gIj-{rqDeC#a651QLm#gWn@0|l%+yp)&`U> zp1{3)p~keu1GpstF08PA64x}N7On4TqqA8LO7&I1=HqRmT6|C5FS<-RRlk!Y!$^^- zQ!?smbLh2mH3_~yC}@qZL)u(ZYx5D`MgLXp#I(XQcz3=c4EiL2gy(85cIiaUQ-2m` z)p#F1?9t)w+!zmq8^Uny)o@OKwi?<;7SLO!QCK<4iG3dG#yN$nGSk|v?AXa)z{lCa zh5IYGy>K2JKl(sklqH5sir$fUAEuCtj=bm$4G6Mw=026I*`eoD>q_~b{`iVJKI5w zlFvfc%D42?@hI;7i4XWaZ5r`?d=x!wzra_WDXd(1HE4!@p(^>W;cvqu@T)TuI7(fT zEE&X2DLn*FK3i~)ocimgEwSd5#AmP*o7~aodA%s)%3JCZ^_lkW*5GoUrGrhyBJSA< zV{Sxz4WUiFkR@{j#LU*9#LZrKX==t2qTfLLvT5Y?VJU19e12|s24n5>TY~4PC$O10 zl>c-WoQ*a>W`-w@>i!PHzOL}&G$IV`ZH>Jr zJPo{!%RV*08tXgo(Nc#($8G4hOTwz0AK7!*f2XpB@s_fY*Z%`AR#8!u$T$|;9)z%R+S?~i6)XwyRMXxiusO7T5 z!=KZZZINuo;wQv+iy!IvH-gQw6l2D8BDZ+md2aBtFudBd(80nFXmx!Jn>sUzYYdcO zbq8hHJXb|-V68Q$yE&Zgyi?9*1aOA@v zT$!+fU3%XKZ~GsKE~KplJ!;Pk?q;%-*PGZ}*Y|j6%O+MEdW+nMo&y&aOJUIMbo>L# z44)3bv1gQbsU3ssv?hRAMhm*dst$C-TIeVGuUV=9WDa8xW(oNU~G*s%bKnZ z!OfFUf4dCJ^T}nuXI*Ds$O`uSP&%mO7PAP6Q@A-Knzf9SViMN%Ed5|KYro{eG-MWX zyGbb~Uh2n$$Hpk{{2R`VaKuMg%iif7!s+_0L{nx3PTRVj>F#!84%RPdiB=UIef>By zYAa*GPUUQ4-U#OL-j<Qc6Jt&P=n|s;sK6hsK zYc6DJM6z*ePndO$0h?sd%6y%2F~dQ#KGuIdd;8?CNa@*b&e0(X!j{@W(sgkbbytD? z7c!dMm(_-6+h4MG%H!)FYI~rcSP3p&Pte!q5WevENv+l=Gk;4pc2#pC!;uGQ^kW}R zr1qLMi03gA%klNvOM>~as>h+Az8wvMW7r6(DsJBBY{n^T74 znMOqxbY%gv+P;-$zE^_{9`BiPR}cGfxt+;w@_z#T&TsGX-t#JZDzCHM!1bnJm-IQs7DkzEOL-F?6ssgep7RFo{#S zc%?K9HP;S6vQq;_6kNpMn(3%=GLjA6`($mH$5VIp272U`1=3D&w6IqK@v!AA`=%Y& zadA28b-spMT?2`zPlJ89oy(S;JHjrFntfXf{cnV*ixQsK^D%tOka!e_14cj7roB8Gs$MCuV2rW#e@4d8GjZ7UYiOOOd zaRqD78stYfeL}+yp~tsbfGu;jv3Tl1&s0e9OZpN(;@>f>a`s>@FRXCG{1fb-`;hw9 zr>$)MkTEPZ|1JCCQisXHbHlidsyzMtKR{G6CL!d=&3t~I{w$&Q&!cItoZc&<9D zQnv%+HDcVP^EY7QIR#jf--uz;9a%&cV?V~rGw*?+%=<|c{2MZwGx=r6`g(?Q<$;cv zGwuQ_J-wJM%86sMyy~!IPZ7InEk<{=mGIB_Fe=v)2TEx=OyjByH&mjB_kS?5{@jKA z$g1k_0ZU@4)_P3Iub9d2RjaG6cxaul;2I{r=%JQ^UiHVlov1TMhni(oQl>3|Dd9(j zc_oJ2+wOx$#a>gNtqbacx_WSHNI33D`b;NW{Z7kFmeeUFkHzo>a@27bProdfE1G8a znyweV>CtPuarzod{Ia{1j($O{d!9=QI1k`MnXA+xvY*B|44}!xV(QJd!hvId@v(Fm z`mC1Zt~I~rcih&(@f{nbsl5QmEtRaouMA*eY|pYKZ3#74guj9zts-*jae{ct!N#cO|J(RL$rI==&( zAB{ond$&amj*Oo68-wk|C3IZdUz+ODgmptDS*LK%OG=KR)9u?}mRSobD;+TJ`F=c> z+J=fDk#vdDhUzQ!MdS%T8HZkv!AELgm|f{2@We|{srxmJ{CtA32^wt6j7re?l+K&( zFCqbVyy*9tYpCrV3G2ZC*wS|KK^pS625F~8qP8fpf<-MFl^Ij>Uy2= zTV+O2{R#eLdF&<8^6W6Uw8#SWI5{fcvyD1@4<+BH-k~zHG+El>TXb#mKbqPg)RNU! z*kQOFySyHXp_eI59Z*I=Etb%VM@>Je#9%qR%4)LA6AEJv%cdt z*i`BPzhCuItI#TxxgSMMrJSr&KE-2n(GPk^;j8G!jA~@Tr8Gw?4!0I>N5L%(wWlPY z!M-AhjG2fF(@yib2LIKpzOs>?3Gl)-nl%`@@*rKRV@@Ykr_m7uMs%s!1vZEV*{Fo)qcsvK&+9FZ++haN|$eMZXcEt;4+=aQ&95)$dk*#`?==tIf z^>neQjh^BPlAWGdz4a3f+&2naPIvKcKc~@v^jvH{rH(TM-C6#|@xYMhKsZ_L43~#+& z$md-8NTYqsQDgi;`Y&ca6$KonaZ~r=#$-S0c={{-omz`Quin#7{e}TD3lIQ)IjxZCkm2JQAWxp#17#nkfP!hGy3YxM>j%>}R#awcQ40m^nyue{8 zleSkuY>N0#8oAJ$*)LOL&xI~!*Uh8CepxnBbukjvb_xIU`vd_d(_z#-B{nWujy>Fv zPHUGtqS~TF*gkDJ+g?y^H)gZffojG|&dNO;ysybW{(-t&4#@rBC2pMFHx!gs`gNZ7k)- zEm7R9Y{(xkvPIPGjHUr7S=^8z(t^qyOGou+ouotXQrPW+=w7)gC$6G3x}k;3nd& z8xq`sd4N9sz3g(76<7VLoKbZib+m+8?9nfBPi-HJd{>3F5B<6Kwq+o;KL|e9EXTd; z&Dq4XpO_$~3TeSQEPniR+ zfKfeGI7z3c__AKTN$iA`BU~NTk2a1)Y+>SYCcj^SbB{d?!D2B`|JZ`v7;ZyOw$7*0 zK6lVo<|!t2p2o&>bLJGMgmXhLL9E?Ns6Mof?6TOwt}knbrZ?UgWcNv=w6v9e{4fZc zOH*(lJ`{$3I*R!>M{^^kkF$9>=OO;XenA&w!l~4`(4Fs&z<pwkH?r_pDGO;g=0bWTSaD9K5G<1i_7Q5#GRg|{tSZ3r z`zX;5sb_FDv72B2JRIjf9)h=w6WN~LD14P+iYaLm`AgR;P(SJcT=)}${QgQv?h$Rm9*_SgR`tdp|kxM)+BKhR7MyJ|AC-#0K|h%vJ| z*~6;glu3aB=-Mx5UuB{(cIF46r{%aKXi7on9nXuS&h?2PtKrY!`!jPhN>HfDP>-+cwRFWH#& z{8@osA?xvZ*h10Y{kL#xsRR0zpTXff%=nVUc2v4(D0?$jpIMqEqUEP(Ob*(^YHUU` zvsw4?`OV8{6g`|hHa&`0-fYLkoIi`{U(6JEIVfE^6u$&aVh_yj;zgAJ#+3>C!YyM^ z>vRv2XA{`l>+!7GE(PmK{n*sSx$O9z4Q$b81t^dDj>gF&=w!E1Y<^xZzUUgtF2qk1 zGAKiy5XevVm9xbJ%ec}6ymxUrD)D&^1h@8bWoa>tjxHJKcRni51>UV)KQBwzc2Fn9asw=5#NDo%Ga0 z_b&%TKRgUXqgH=K-zhEl8f0093!F}|YsqVw_;WLM=ZXfKcio6>SfR^-fHgTwK7-g{u}t~L2igb> z)Bm>M2um~e>wOs74DCnhf`cqepRi}jli1VHLDI2hFYK|ahf$M^ndU)pmT^a&&6sJA z#y`E7W{ousaz4bqkEz6Aif*h)@dN6NNX0*A7UAk?-T3E%E&CQ5!wxx)XW~DE2;Z;8 zcw7E0$|xG6-lcE&F5?v%SlmJ%&Bw@YrL(2sUQBWAes)H*ozYMGSiSFid{r`*Jt<2S z>W?jR`n{BGcqd{;Q?_Eaz$YL7T8lnwACQ2l7wf9Fjljx3$63|c6qahBgG-VR;4G81 z?4)8TYTEcS&p=rWx4nf@hLLz9>Ky8Poq|(W+p`;X`&r$?0qivJVg`vmY{=&U9Pw^6 z`m76M)%tHxso#=09Q9@@jj=3wwg(n=zGNA<>)?9zbsYbFIexsdj19k6h?!bH(fZU7 zG(0eoN!KW_#sQ&t8yvvlHY3@%DMe_eavPm*hU2xxi?P1V7)Ps%vDbgkV{%d$e!MoC z>94TC!jQk{HBYGPEB)EZVsHN2_#{TsHJJ5fNv4}_$sURAWpct%c=}o}XOfZtdpDukP{r|pmM)=3>4dQuK^-kV^hX1t*Fc3}_xUPYN1 z9Gg357Hh~I%G~E{#^2$xY}=3NtlPE>v*p#9?KlNI{q{Hd3862f_}(}ffx}NCTYTejeUM;MWWTxNyXn{p{uAb)m7wph_PBx<8^>YE`3+bN zmr>=nH-cg$D47pJS&#+|I8h1tz7gD%xt5s!&<__~_)Xp<_F}2Q1eVbCoOpE)qZUz9 zc=@rTaHo5dsNCx_4Y=Px@} zLx0Y?Az9e^$6<6+nxb^N)g zjx2no#p3lgGUo@y%=AkY4(*wPGK)>nadHGyNgbx?Gu^P{O)GY&RKafU95g>+5P!+>o!w_=a0J?q*V1pTj-8mHwZ&wieK#A_Z`hyAxk zbMS09BuLs~)1~CP{VOxckPsWP@f*jfzG%iJ?N7;$rM1vA)Ii7*Ovix`T`0F+%{3$* z;U2l>;}a^y3}h`>@}eZ(re+Sh=pTe#y1dBmUKr7<4nV)t0{7PXhxe&`fJ2^U)9s>C z(zL_^Cfi%UVm&49M!Tgzn2X@?>=RV&>O0u2yab2pz9U9D9G`I`j5bJ*!gIbVcv8w9 zX78djcda?TdGL=YXO8CHA9}?_#=BYH?i1#pd>vBspPs0)wUU=h7)HX+9H5@pKT@^r zL3C!TB;I}!L${R`6I*{H6yH-q|6D1?$FG)PV&oH;64gzS zfq(sqs1b7r<_x>SuVJOou6hN0*Waf>qdgIYI(DW_1($4I&XtUP0{c}baBa++44-U^ z;q#hAodNQ+<9ri6-u#1{ZBl{6;V+4Q^k#ZKubP(LG{Bhqg;Z>wImsa>VEz6o+#vi< z9xmSxAqz8!=2uGXgBGLYbAOb1GX$SU#1q$Bf9h5=XNkUc2I1%824ds75Dst8fII$H z+{xGGd`M;#eP)Vv=}(4`y?v8KFZG4Y$H*u+tYv}TTSNKqf!mmCIgu+#jpYgr_Tw#H zi)8Ieh1L@@aQYb?7PvQqR+tEWz{Q3bXq!W-me=#cHXXE%brUcXnZP~`xQ(3HP0BT9AvTc$w zT>hTpRL9U7S4l=fmsJ!a2=P7(W9-WrqFP)cv$h^ z7AYQ+SG zxM*|)?(D0?Kt&}ytI^FrxGBMJTXBXSG@K;t1&znuZnbE4kip&`li28#b7;M{5cd>I zux;rk)=yR?)6!-GJo?R+Y*w7ejyDVU6ZnL>7Zxz>gd*mGw)nw89>a1P`B%k^OdgbG z&;1&)Z`u{|^#JlA1|Rr=FM4e6(qXWEm=eq0-A13L&!KG&(?pxg3Ry?mGj@7~ZvEit z0HODNf;*P`;G?q)*X@|aK1AKc{K{t9lfMfsmR=)~qwKN7aTIIGpTKNAr(ojXaM8h~ zs<7JjI1aOG!IJEUpf&w6E}Rx6O8Tfn&Ir1T=EUQ$_r!MaJQzvu{yoIsJ28^H*?V8e zbGRTH8QwuhW}DzYl|vX|tH>gU=Az|OO}6#)Hs;a09HPT2_{C=8jG4%@lBgC|A@Hly|0TKZ{SAts*9L7AHp?3Be!{JhSG zii_0Vz zjsNQ>+HyS^@4IMVa2CPu<$tVK8MLEp_7UthM6zYq3oxD-gehMJ(c`EA`*AG--(6dd zKDR}f8z+nH(HZ2B@LcrOrJ2NSU3R==6tlB6K#5L-gl!tc!$T22SA}9mTQ1~9sIq&t zb3rEPDvTzD*i~o8H+t*f$2VP~x*%=Vb;E**1*O3oo0V|PYbT#+^B$i(Y!VIAT0wup zEsSndhV0u<`K8-7V%LGmnAa#qN^=&VT~8q@d)tvP&&4ozO%c8w5yiyfjG_1-&u_Ka zhfhyyunC*O&?~hSZL~(fP8;z2i{P&uGu{E{H6C2Yaam zCtt9HT2+QoBl8#hyLVUE0Z$|D{614oPjLwMrZovJeU$`j#e3xIJ}o#l7!FQ*GMMC& z%4g@a5^1ouGH3i+$c5`(3O6R%l3AXqEM>xZa>~hqQ=d@H{Sx%8 zmP6KY{juM`T+;;YheqGYpw*wm zoMgJ;z=2S*bkWGV)|dYHF?cvePgeuFPL}33)evRjjA~vzkB$?twM+lCvGGP5!R7NA z2$$yBJ>ywevPZ*SugJ z7jW`3)V@V>&gm@7`H}{Q1x?h?HTC3x;{+nw*+MV8-Uo|?UCG$Kjkx#ETvlg%5M(B6 zb2rq~No1V~ck$yY)Dp1jj@j=Z;&e5>JQl`>1}j0r8yR|FL>$b4K-6}9ZXNiiK=dc! z5Y^tR1hqFV!R)+!puQ!ay(%2V<(F7+9`n5^x(33tb5G#D?0>MwOk2ob8qWZJvAehLLb!M*4O}rAKVlaC3K7vej_2uGt2+A1 zDiOr52|F!Xi$QE^ugF(73FQA=05jHrfB$sU4&QZ~em|%L^(MZs&T$!xp3+O_+iikF zmkl_(RTH?-L1V1Q-hf(_d9Z8DF!rr#I9)CgBYJpwCtjAag)3{%fvoNr<`ps$WZizk z#YNrx@3aNzd%Xk_@@BHuEtz!lOmosZ=`x%C&I{eY?cwqd1Ii!C$Av>gAi4J*v(gz~$&4wu&*B1j z8wvQhz5*^5?u6Gr+N^gL1e4NPvh2OH5}q+p!j4+=|OF?hzj?jJ+P zwQIQ-lOuV-+fCa zt3KC}rufYey5Su0Z1tvZ9oE7h9Utm>v;q?EOM>J09MKrxd&Fs1EO{{RD$Q3j;nJp$ z;Ra3(CFI;sa;#9Aj(xWY(k~0yh`)})c$ZqDUhs%^&7DupAI|04{$wvW8x=EL&U&SbFiHg7oAn%CI?kfeS^$i`KqKcjWXh^e#Pj zaJfjKdLAv=(Pz0$Y&#jDP)aM-7>fdo9+5d)Z;2*+ZKFG)HR0#3bc{cr#b-yxl9U1? z82>q1v~uiA(3zeJJ>MsCQiO-?8gX#qqXCGXNPuFaKQt~I4{?8tVE5c@^y7_mI{WuV zQQ#CsZq~^CaBtjwVy`ibYudIOGzXRwrIVNG6sKS^qk9dF?Gijet(~cjga_o_d=Bf? zcY%k8l4#A-deQsvRFP@IY;ag^#JMcb;3WRe4k?+`BnqCoia6Aahr_ayAzdaK?hI`w#>?fYgGmwT_oyKwR@%UnUob^XLLq1VO_`By3>IUq~*X&&gAVluIkM; z(f&1YWc*PdP&60t^pBZf+NMH+PJHFXvXjZY)qZeeRyP^bS3;z_eo~k3`$^pfJ6t;5 zo&Jd44{=F`bhnupk!XqonITun?Xw}+v0+Qy@+ZHD>Xj?}MA!BDN)+~Hw2MB4|JqRQ64pn1B9$dAq!H7LxX zo)u~!r8a^)AJzb`70hv#tt9A79!5^7$61$UE0O41FL<@~Td-`nF<3?|gCAx*xjQ)= zLhr7vQ!wqKG6&6j0k2bb zk`sIii4}g6I2B4&V<6&V8s|J`D0Jpivh}zQyqY=>^3r~hqFu3|9=NaWqOl03sM(U| zkujup#sINxmZSqSb44-xuZZ6L-U-Uxc9eZ-LamQiXormp4I5QN>*fsy)1fjjSK8%um_(Fa&YNrIvJ8uPbWN^gzGO^z^edXveI}gH+k4>F2$pRvyzHOsm{sV-eZ}V zdHNDPx~YRdQa+kCO-&L!&L?p_f=}U@=Hc|W^$A#a>HtphUyQdhi}>9t_V8WWY-(NG z2}qcFnOsyFAV0ndKI5}Z;cL=Ikg4&)E8IuuZszd7$^n71@})=m)}f`Y0=Mc)25i^! zAVGf3WMj}O6mR$~G73(_rvav*yzw9oq*dV2Yhj?ad^4OmRAJp-YYl&Qbd%%jJ>lwv zQjDq{R#)gT1Ww4DB@1^lbbXTvQRDYPTxSUJdYFto&6(EHqar|GySZmh~n@E@6r(_B|kJ0md`_VplWG7P`PZ#WYNMSOE9VG=c8CY*5#{Nj|?nf{O%QLQjo2 zH+EbAjQzG8I{qe-+M^U4zW)*pIjfD|kB*|DSC3M$5I2mU+k%Ofd1OFg9c~%%h4wxo$W_ch4~OU2cy=May*3|=Jtp9i({2C@=7Cpl9InaY zQKkPdsk|d5V90^=NBjjk=kgZVqf`cOIV<=$ypU4IRFNEwb^@6CuI250tf_i$G>E_ZO84+ya4NYDF5k7o=W|nOrE3fMa&t0132A~s6+hbN zsLC$Qi6G+})N!aU@JIF=a%0DrkX?aRs8f9i^sYank*b|^^~7WN$L<)mtjdH8uN0W# z5ds^=PKUSil!3c`4C7-DK+a$^y|VWVUGnfVU2*0OtT0=KM@BjY3gGhkN9kbtcU;j@;!IJ-SwjI zev8S!GyA~5b}7>36CuPx*ohOno=wZWecQ!&F> z3c}7`dNH2zn?%YV27^oYB#4-l!|IQP z3%h5TOxdxQ#@6b<=Usp^(=K4>hB7isB-FyQSMgV$4O4&k2hDdmGvc_B)o%%7?rTQk zhf#L?Hy$lpI=L(m~;vc_I?!lnC&p{Q#R^P7>h-zK2-MeZ@$3C6q2@k z(T>I8)Y{RC%4u(ft`EogtTzN#Tiw9!B|~Am{!7v+QsYwBxPf`3DmUY-1sh|b$zBQ> z)t;68u(m~qC2I+1)}c{sW4;co?zaHzc{2Ee+k@BL1$^gCKN_DJ2^qhm@XnwMYv{1W z&8m!kpSXhl*cK*otUL?_i%dlx@fWQ34@TghnDuOgSu7ix-vxVxXZ-PxK8qIg(gm|# z5=DVwp84qxHYbLI?*j?W*ff*=Y%V1J*@m!f!D8k)<|dz-V~HJ^S=6t)9(0#k*Oea* z#SmR{l;05#S@)M?;JQ>ydv>36W~aeZLlwMu{1PtPa1xc)pAi}CD#Q7?E$n#dGB*Ef z3+sQcftH!~;fj(BH?~w4ZC?Je(#~(78vnULbx{rw#V-0cS(TVQ3M2a21|S~i!xf(X z4vH#KkXg*bsqc58aqVgd_sqezw0zv~>^yZV|A46{%9+y&MJD+%j_S%z!BZZqNYS4% zJaRAs)H>zy=1v`a(6)oT5%7-I`|Y&Se*@KT&_cycC-H(=9GE3El2tXqpjmYbqNC)w z4T?>o;USi|ffuy=rF}FaV;4~l*on+H03MZbuw-{G8Bok&TA#PG7(o+b|E!WUe)A+! zA{%)6@dAe2ZAV^FkDP1pBX+AE)hQekZtbuF)&~K^ER#i}q}L?tn+${p%o7=k?}JB+ zX5qXBO?(rTgLyMtu+ZtRNPLO~J9zyBw)A8nX7$0m-UeuW{eli(9D!4D45+j0&aZ%SLcE~i7ZOja3lP|;*htFcjcSRwUN+o}z*TAInB8VKSjxC0@lY)P18ag)5r|^CTWXX$x$)8>_`eJR}UAbL#J&6Y7 z=@@@9Cnt$QWddHe=_bw6owRbNH7WVCn+E0-l6kitQ?u%1&=-G+w*$*z`O?31j^iVG z$Uq)lr8SuR+!|8geI8tgSaG{IMuKiZAjU<>i@tRT8e;z%GWC5nT=f_Sea@RP+A9|~ z2{_T&CIj*&N~r&*^r=VvZhAO82%NS=V{p|e{A_p?PyO_Vi>?yf+m;n@_WLW|!%dQ# z_&Eq%B=3;PrxUqziIq@N*#jGTi`b8?A>`bam0ai?Z!&g254xYPlNav(T+GH)2;MLR zcI^p+Sidl+3fzxv7H6pD-?gGAQT^~@XAdT&h{L@8(;#*z2j@Mvffj&xni zhQ*DmyBudSXCmA$8&3B3HiN$4^S_M6oj}xR+Oy(BEz5rD|8TbBy zKS-Et1;@R0LXVaiU^6zIj9kgH@y5TnrTuG2_i6=lezFw#?buB8YO2UMMhI5xy!NN-x?!NCcx=;m*|4OS2>5g1EgEq6=KgYysW(k#BM8-ANB{y?y~u8lE@tl z>5pY^9W!C&u6|*;pDtu~1+dtYN7?CoRaR-N0_xf)VCZ!yT>M}|x7|)>EtfV4n~Ku~ z4_Cx8KOBIIE9&9k<95#3z@2J2mkXj^?4n-vax`>{GZAl<(xt>VOG+IbW`r4D8uU0_@mz{!to9moq3pe6i*^%D z*;mEY57ffx&m|!H(;P%gFACx|rSV;PvCzGG_f?1F9!_IjKlbP9L+Ua$mbUO17o=fH z)=0{M*1KGRg`76f?hLBgqsGnhkLQZKzob;kNx)fo<4xx@1s^mE{Fd zVfzy@Y0YysZKoO$S*A|nOyudj6D#OtKDRe`)t6X1+{K?WAER{MOg6UG1azJcqRq^W zsK$4=B}6m&;{vQjPh&-eo+msMeXbCH0%n9k2lZ^D*St4J+>?LQZf z1(iQhaD9U<*$+AFUf_8wSrq`zb!CExQ#0^+WgT3c@dO?Cd2#QC<3y(GF#Hg7!fAEGTWJ$j}vah zVCkV2YV^+3TymQuO>nt{kx}OCHm|5%8vYh;;Od@mkdbsukMHo%Yk zIb^5m54`Vv6EA#AGH<+n5q(y?$H^BJnCs2{%5BGmsCsNR8X0Ec-?k!L^FtrwYi{Am z9v6D&-W-GXhMJ~#ag;z3`z7r7KEqJ(e-+KfEXu=%^;WFUMZnfQ z90|FlVr;|L|4>nvKR*>cGLs8(WeUxK)Yl-H+q&m0=ady9R9iTKIh!nE_gmJ|A049w z8Cxb`uIM&oe4cB%>J*k76pot@YO_7BCNS-k7KHnDOySUKob5CXzPxUr{)_&gyLBlh zO55ON*EX!@pMS@12vBWw9qP!a(46lPcwU)T0!J%mle0UL*|Y=+mcG-F>G8cCbG?hX9L}Ep zlWW64ze=35s0;6UX>eQaMWge|CY-e}9F#VyQ=fAd+?0oJkA1l$d;@Irchxu=&|{F@Dro_U=8!W4EQKN!A*? zzxK5-P~8qc8b3$nxKM84usOVWWr@Abv6ySa_g1fs#kJd#sP^(cdN-k$>OLL95B3xB zt%Vm(@JqlWZzfYcXL)R`r#O1*Gt}g>U&D*<f!R7_B^lI(nan9Zva3ml$9}e@;f2 zlydGtVc5QTFDG|hhfRwg!3LJZpwik=_<%&wp>r`zbw>lH^1R}IXYS+kcmq7p^ix0lj{wpJ%k{_`Ri z6!M@+LXLQ?-$)XErV$4}IpS{;br-TLd4<*5h5s)D9??8AAa&A%HktfOt8mQt*>ClyKU58M-BRaN?SZGt07Z0Ei4qYK0s99Dv+B( zh>pj6zJu=}OqH|cO1`LD?2npY@$_v6(Yo>$L>??8?y{p(fSb>_A-=0}>g)AnQN?KZIXx2(H_8&RwnP%dpRrTxZo;&j>8N+z zmKMgg!*elF;EpL-w3+R)(AT%Ni1k+|33m>YFI${IdH+%PJ!%D&yz~&72Og83s?}uo z(Jb=%);{uLr8#KV90aMRGBV01lI#&G63@e@gl&bM#IAM$De!y(26lm1Zg7MQIr);4 zW5P+3^9>^Ruo=zT=dee=zQbkn8Q`F!NfMn;k-f^27CmRP$nZ!Y3$Gs`Qbn6cqSgjt zw%nPVc+Y!a@(Vz1ayT?CI7)O=HOX(2sf68{PQDb6A#whfi1U#w68-xMS^NDiIq344 zyq0-FMp+#tL6wy>pFir(mNz7~WQ;A`M)VTPVG&%pEPzvU(k17krLU&NB*D)_SBSc| zl(bJfhOZ8JaQV4*#M?B91TKBco$WA&z<{eb?a&4yawQM1zE*-uErT#>WhqGwxyv?&n)2AckiEk!3eUIDX<7u^o{}W=w4&d-tl5<4Js%^h-^6uKt&+ zFFk>!c2kJ$##tm*CX`gpTuDSTazI8=mdO88BHtd*AZgAQ;HgqEtT=4Nlm*C6tKr%>WvtJt=W;K9fV~kKBYnw?LB zsnOBg(qlVG;)@NWL&uH$x-^B1_Bes}a|^lIl9{CB@iAH;3&djY11LN^9UV9K!#;5x zHfHlf*q#?eHXK-H*3fr>bH6HY5jU}w_!J!{N1SYkZ0#FP{b4Aa+#e3(Uq2-q!lFpf zcn|UzM-tPYLekk(NX7|{<1gN+EiZZ$H|HBd$%oy<-^+ltceH>-(`gzze-f!YzlDV8 zio&!V=i%184KQ((EF(3NssD0;k(wh&piVjFU%$!ZjaIW&e`QeGTaC>8Fp?C%;rZF1qh$2E z0&uiYq^fbF$k3CCByp-UJGkv4#fnx~{`fsi*!&Yc{5%n(cu(T0Pqa{eI}1CZK)Sav z^ZCw+#79()Oo*eL##U!$aeol#v%PF}6+iDXIL+jwEXcL)3$SI)6_odTiC4;AVW$0h z?(DQkc#<`i_gr@10>!DMvpEqgB6qR2@Pk6T<|KAgMIC4T>4DzY2gu)bFY#H;Jy<&R z15Eqd0A9Vmu*NhAOVxeYyir?;dg&KV?MVbACxj8hrJb0(C6cM^KgPtDEN2?xYAoA6 z7{nH9V#;$942YYCDt0B%+s4oD>-uql-v)9ea3UMUvlQ7*E7_Ota!`LcjVMN#&?VfY zs*2VQOkQ6C4O3(3yYOy!ee60n{j)pA-BU$HaUpm`XX4}^w{gtr8+84+eC*#^Dzq*) zBAUDhM$2XvZv9saKeiU28=s%=X`F-#?`DvbB2vuj>esP56ymP_}=c3`*CPi{~Xd!vG*&pi@3Y!<+>msZ ziJ)t36uXeQhz)mTv8`XHv4wnw*X!C`Zf5`?o#GRTkLMk{e^G=zXQp^1e*zovKl}s zJBQ@n-^ndKYQju3cH{b6q8QTsmz%bEJUjL?72ZXOvXkY7*#2xQ;Z(NZmZot`++!5$ zm6l}&3kUJlhA{5-c~J}>jA#1?9f*4!>he7z90dy3u#90;fsG9kLdlz++zgKlh z*24nka_%tu7A?(&ARRZ`y``&c;?VJ`FW9xLMWu(wx%+!uxnCs{$la!qSf~^UW_u6f z5>rVg!)J&6b}N#+&QO8IqUZEe+b$sc`sinAcPKO621_>8zlajzDh{&2== zT$a$R425Z-XW`t(Xo2MZVY=u?GG|Sn!KX)qfXYL_={1AnjV@ryh-0+R;;*a|IQgR| z>`N(x%Z9};@ANh@)8hm!>@g&Z{@S2~&R%k2l_?v0t%!|UuEJ`B(rBtOk}AoD^1j7s zI7cFw>=h(Hz6_7m{T9Is9jBPI&SYk>$rY;acEOU3uc%GA2nlI#gSElHY3eS7Dc6s{ zHNAK&Sr-diG>=p+Y0aYhV`t;lWIo3~FOGe3b!5+k3qi4q&pG_Ez+#O|CKY&znof_s za(4YAs_s{d3l2}x zG5C@^F*)bRcB)KfFMddo;hrzhsPvBOiqPeJROQ&wg*SwzdKN6lT!np}WC@)IgJ3u# z35v(*gVSpVHg3--Vl_q-f9?7WPu~~O(_2LG`tg~}J+Bo5<%{5og%C=*)98@>FKjFQ zhemgfW8+t6XnOh+q@FIpIVUQG2BIg4arpoYD&2?RHvYHIp2ud+cVtb?i$O_$A;eE_ zq;GvsVRrp(Zrhye+@9)K*m-*~#Pes=Rb8U+&M^T+=GkzQ%V)yAmwn*7=pPL_(JRO= zd;|V13(&UIg+q&2C?QfT{`DlzCuKE$En5e1%LmcdK$(0`oy|G;9pc^{GiT{_2CV)f z&y#*Dgd?wwNjBe)ZXm11q#mZRzN|7h>>tib+&*(xzRf15L^pENsRbD)9Y#9nNS3!T z9&|Q}!BmY=aMSPrB%h9FkM?S@lAi&@{!b&89}L5U<^_T?TWa9xZW(C(IGTKyw8Ov0 zx3Y&>A^7!yD&Dy31?yH+)0-iaNaOK?d{##lS3kE#y`sn5lvZsnw$l=auh@Xvr~vMM z`CMF2a2X(kM3R zL?3QEt3)qMaA9H5rsUVH&1Bf*5qZM<1s80Yjya!FiKtHs>$F}Coehp`MXI>@+Lm_o z%zp-(?2>VDWgu&@%m-t))%@--8~qC>O<4QObXW`FDqjxO@9|x>9J)+5!Dk`_IP0dmhZ~$06Lk;ucE2-pfW^Gh7F|V9tONw_^L(}F@7@!AI8ylP;JJaep)1H*w5w(flf8*%thWx(?cY{=h@j;l_=Ppd^}QeX{DM>gWO!Xi#x zRFeA|b>%)(EC$N25B0&MjssC+4(jy+q-47#VV{DfC5dEzK`DCQ_iOtHZT&r-<7w3wak^Tu|~b$qvrHn-4a6HMB|yC)z1gZj}MaO(2}oO#t4zfV#o2Uo~~ zLfR?#JoP-C7{W(PAAiIYb`-sQBru5l#Cd{1W+j`*wpY%E&<1BFT0f87@@~LY(#9-# zz79S+JC5r8xC@K5^vJzFJ94M-FvnU~amF6jETzZ^A9AJKv3Z=}n86_ZuA3w{dPtSI z_{_jDCTGy;)=?C0k7h~It!#cbV_*2|%QO2C%bHw?0e?@iEu0Kf9Fxq|?*2~SrW>-2 z83yd%*EX!Q-v^^doA9i!EU3Fj(ROz$=r9f9-e$&P;DR$Kk`aKxYgRIam?#$bjd%6L zw(t%MV|uc{j2y|K@N%*%bbhZC)a5+Jvf@zK^`D*~;9DOm-Allwt`93$@*c~PPik=C zi*}@+tjM{QAE@9>608sM=FD=cV9watpu9&Jr4v?zhMY5Z%T@)gsUH0rzb!{N8bp4NUQ3Q%*P12P5XdTt{yxk8px{TGmWD zZzBsXzKpl_Ul7LamOvk3#ng}R=eOK*g0O8yWazyG+3&rKb6@L1j7wG$-ykFI>d4Wo zGpLcX^*+I!%T$03U(C4jmRUGsl^Qv?wFE@7XK{H7H{i~`-7p+63FgiZgw%fm=ub?f z*79Sq;L{(v|L0PSTXz+WQJTP7o)8Xpfgi;SAVF*@^_;s9bW-L}uYLcat3@H+ioOUn zt43q%n4MslO*BS#TKyG)DPUj4>Gh_ej91=8iO3_&`m$(*=)}kI$O5| z-0%O#iTvfy_tmGlUALCw?mr#ex`b6A!gH4sKBeJkW$Y943uB9_ZE@8#y~ z+KlV0Te#3`>NNKIJ9Cl2BFKCkM0BSJQA}qHB%M$tZ+#0ui57D{e**;WPot=sbTPN4 zZW%<^?!f7FY0%u=%k@Jt{A!$so4)08x2{;gmOGhT?8{s#sr&@5y_4eKDPlP9n-aLW z+EcSF*SUemO<=xA3+oT4gUo}o&@fJy#C`JN4)|N5gCL8~2TKzVCr?OR^OQDvS;4us zU&4LE#jtVO1zd7O6YC`l`A*q2c+EQphDJDYGH*n|bJjxW)0>IPS4Hvu84Y+n#SX?V zDx_ml`e^%Ap27O>I+O&@#k{&{=(^y8P;~Q0JT0)smlb(X*ro?&vbEfZK4pw+jmP$m z=au8;D5JR`0H=;==34%~MX_%cxTzqBB~=_n3)4i%+vLxA*ex+n{C2EzkyVa)_wfv& zu0G{zFKMFT`4w=i-$h`Sv<)7Lg$s5pSp^3TFL4`$k#y9v#oV~sBpBHiML)&Y(1JTP zf@9;OX!Yv=*4bFg47|fwNyra@g*IPt6mYm zhOp&4hq|Z=k~c}iRgWvs;PMl8B*qH9e6WFoMjg1hrAIK+cOuc4vjO#%t%a1K37GZC z8`?_~F!_ZqSD9Ic56NQOzGyk@BA?;->?pjZ_K+T!E(uq3syOqiqi}l47?SesfKbPd z-?gdpzU-qe+(eV_oFLr}-~Js7b}Ke;OFr9Bxj9GajNb|Pc=B_hhhzq1-Mk`r)+-># zf8U@s^Yidp?{%~e?B$%J#6etAhwFNgf~|^2sNiD|-urnUY^1Y6K4~%%Wl5o0sT`QT zxd`U=V{uE>RPY{|2BN=LV?eS3+aoVSdLtUSTE$^*OJ*z=J2zUWVrKxVhHK#ZnG~!K zI)ta1=u=7u+t z-*WI+hhxtQ;hm=^(QNC5tu8Qf?3<9 zqMOVi?)$fFu1$G0j^Meyu=f&dg7*fRb;B5Y;@6{+st3R8E5`4yb%f)3GdT69%jlar zKEriyEp98^fH`mOP&LCzC@9i|fU~RV3BD)f!W9D$|8@)eSUUOac#QciLceB+(Vg2&@j}`Q`Y~OAtBca<$O(GbQJ5>d z^}~*59#^p8pUJq?J(^Cpb*6Qy>Ck<5FTDJn0wX7>BK_wkxM=i>zUlWd_eKFO9WjDA zD-;Xgcjj?^sV2heGb8bbSRp;tqk_Wv9m!h?WJZ_2lD|nJF zizY7!o=CohQ&x0ebuP8w)(hg%QbU*7R;1$26$WfYVm+NPrtM0^s6;BGmxX(_rEzx@ zrBHej#p6R>SZg#3e+c90D-&@rJdi8gSiPBQ+sbf<_l-o`X@}8jqyoE9mdCj&{zFLz zV;uT#4JQ&*O%qrB=0Y^*a0m3RP^aZyFt7AKZbEkh)qJ>Hpg)|#xm)+p=GMh{!0H-2 zc>>(*XEDO|wg^Gxp>(Qmae{t~3PbZZakwW<3~T><=DGJwoPTCMp3oXi!!-WUp9kb| z)X}5VzGEBh|JiGvK0ypm4~&4sf$zd~R_UD2h5(!&bRA1-fQql>cf*>i@z#GD;LS6E zLGMoqUJ4Y@!S{Wo8A z!hrC3R~SamG@|`QNw`Ba2|l=~p{;KT-L*v$wGJx@Lv;*b$}?}+udfTgMQpK8MT_hF zy%5(t+{;bm9Y$pfzeph4j;Tz?& zs^O+!-mNrz%J0ulI5%?hzGn);-x}fTtHoUYdx9&M2J^1y2;65~#$C$4!d0vKp;7lg zVO{2XdU@>x3b!V6b0d-jfd-c51O^oLtCqO(p|w*t^ip`B~Z6JXWRJ+w;G3)AoO{Gj><&SK73>d=@ErADbj zdSe|19hQY;5fyxPJqYvp8T-%9cT^;9Cv7!;VJ>|zn0qizo%{3gE}h+Ti&lzUFe_OS zi(ZOZ!XM97u=UI(;jDTwYH-AZNqDP087Rd49;Z@t_NfjwDX$^^>-helo9Wo`yByTFKOn?U zlC*H#F&Y5rq@L9G0(%^zTFPeHa{@Td8XOXsqf)UZ#FD2 z^d_&&f=FWbd0};(7##H(pjnDXaPgiwOi@Xmbgw)^(?!1uHwG>xr%yPMJ))z@m{p_b zO`SAqp3{nv?~OU3tptf#Ig52@b7W#_9?CsZ!<#k^WMOy+8Pv=miH}CHthP2Nb~!<$ z9|<6MM>;h>Btd%2ZHev6x$NhMR#27-fN8Bsq~Mz*+WA_3dPyfs7_`p$(8A zx#DWMNDas=t_O|gbD%%4pN^0#gR^n(Am~K}gdHz{=3o3Q^}P$Y8dqY%LpKZ<_CnWd zE3xU~awshpC+}vQgo2rb4vGAOFS;iXEuGPIhBwiYI}Y}`>)5G9qlnM58RSZvE1Q{^ z02_9F2ctc<(8K58d)y|FYjW}=>A-YS&k-V8yr1td+sIul8i)5c+`<`KrwTpm-PxRz z5zPPI6KuS@p2TL0k&hK!a5Orckit;B{-A>18O()uJ5|W>&r8Xi>KM{6WXNQcMzY4a z25hBfA{WT#Qn&CdRnE&rmBTyJNOgWN$$q?!RVdMa z!FR40VZ7rarXVUbSKL#ASL6A<6Vq5si`$7CHnBf6FNdT zmBeJfg5g(&(C;81QlrJmLCZFfobiv-T6YRGv= zCQ1{fKz}$79!koSIP<$O@n9yn%^FE!6oj01jW^Z=@HyD;yYLefVd4Cz(7rnbrA{s& zv2`h=_-{C_y>||#f80i%0_5a`dD=)cF1a;&<`9arZT{RsRk>>`(wPJjZV7=UUn7x(M1H6S!G(3$6r?U>?1z z;o>Sk@bKmREfW7=%mWFwmgf}xc4`W>%T9psSrc@1t|KdYYteFx4OuK(Pb7TGK;0!0 zcE8hQdz1!oorx#6M1BtuKkW;-(h_icu_*k_Iu0Jaqln7oGvKBlgh@sGImtzw81;u^ znX4a!@O{!y7$VxX%RaoZZp7&WtX7Lb;f+rRe?sEkD~z!|?1JY(JvNa?Jd}T5B3UD_??N zYla|aD95~4u#VKuP$k*Rop}GmI&$Jj3jJYGjh@oU*ztwn>8IJ$Ud)U+&HT%aG5XBU z`5Bzi7{ipTFVTysS-|t!Y=eO;Obzj8N2(+6$B-eeHSfS@ZX&ETO_>!tP3N|GiLtqS z-_xrAf1df}bLvv*xGl(tl~3_!Ez)`DJyVBEaTa6dqk3mA#VfR2^%E*QmxGvkZI-vb z2giNA%M~u;^VQ8_*j;BuJxzP5%R_z6{dX?xh_^-Sxo^3B4*dRYB;U0n`Vlu@?7&WW z5oWs#Iit>%MV- zimZ1O%r0!CTh7Wdo2biJp0yl8M=P<=xlP!l{95SR`~`nMvd7MEt6>PeX^_-1dd{y1 z(|etWcXSeF3iytuuTi)l`ZFi!xg+EhQjxBj!sN#;5FRPGOKr5zfYK@@sLT-H*LW@F zqj3rCn=I+YI9q{vc^MU*-%gWcucN1YmN5CECeF`PfrvDD;e6g3tiP3KTV`f*kA0Uh zIVccLSf2v+Hu3cOL20HUnhx~8H2mkJ%F@0jxy>HSz}sLiPHew~QLzRzTA>+72Wtw4OnBEM|KArb zHN?Dsdog>fAKo0JSo!J0ELf`i4wT2Ago+3e_ItcByIqve;9L-ssre7DpL4^2hPPYjdE{R_{ZfwQv5(}Be z-%k3s?gy@w>B0vlujttN(Ny)o4EWfj!dWdhVSm?k@;O3j_Ho}SVf9EqSTH^sM_kdv z{bfUV=&l&F?_0(_;MssI)dlqPc@dmBrVSYA79fMB{4VRCtra`bmyN6C)>M`C<&vtFd^RQP6`V|%PF7ERC+L_o4Q^Xm zpxu-x*k~PI_4|hpHy9VsMn_C$Hy2qFf4Mqb?Qn!MS}Ot{rd`LsvmL6g#N}{H!V)1v zrWmdlEQJGwmE0`1e^8`;o;&(|Eq^UIRJluuurE3mjzGOH@S z{RK3WzX`t1vMOSFoQe&chja!tOfGWgd%u;eCtkBxTqa2a4>_WA6sC_rA1+ z<2X&`cxqeKcrhE$Uu(z)&ycFd7D@7Y!B4D>Tvc`0p@g+msWH2yR#0(vGkZTtA4hmM z)IJUnh9$hA0rpM--fE-;uW2K%(9-LZ^TW)Dq6{VCx?&a0&@P}j3$MmBbnRhaaBVVc1*j#3Y^-?**=d_VVl4Tl2uQ! z>T&0o?%T6494|$`I`?rg8css}Iy($qHGyqEWyl#`bKqIq9PBraCav@W%e$cm870P~ zsfN;hj?;*%*G=3n(haAsyOMKf=CJcmo7l|WjX3{zGy%1G7Be`T?fk6;HGeL#DkpoQ z@_r#(;9ti|w#U;j2_N?BvMhv7{DWUN@p%CIV%9v$7V@OrvB+#+)#c0Az;bCW+rIrU zxx7n^J=GqBV^v>?OL(Jj>zgHCt>#% z`LR#K%c!s8EL3k#;Y>t7qrS>5F=Um^HJpPj<~8f zI+yT{e6?EBu%fiNF8?=oLMwvS=idj> z0V()7rX91g!_ub6HAndf8b{Vy=t z@&@qm_l3wDq2Ov+EVp6ubh0b_E4Qa!18b@xVeWM`c3(dPl%t=3(ziLm7ix7kbwQ^UqUs7D4I6q^ULvaIQzR5R57^a~uW&O+Cr zCAjvlJ9j}jZuSSa6%dzZLS1BA1qS(<m$0|_6@8}Q6rZP>T&l=Q8XQ#g<9G_z;a?# zWz!dHOqkVLc`fZ19X{hqS1fSGh~-UmWcMf9#WSdjx<@c8TaH=Xvl5Qxy;=G)*SK|? zhB#%h5-K*Bgr_d%z{L0k_|BzH7~24Ya z?pAje+<$k#+^{eN5AfZ77gSfdCTRi%iHqzKIWr?GFP5n%K^6A~sklSHwp#KqnoAAgl)W0jJaL)&id z!_EqLeMb~>9*kf<_7Y^3Z8TvXTz$G2?YS_~6g)z4lF}ZkFkQR$T)&{c|Lf z-@$>pZWA^HCkf)UovP|zRIyW8HyImpq75P9B!TaAD*Sy0t#2KJcRn4cGP8tsSh*3979r^d>ZS1J(bmJ7Qi!&!ZN0rQ;Ccit?U1U?F>Y}rBsHr!`StZp44 zyAP}+(+_;aii%Gd-XThqd)-l-cfXhOT|vah1;XPO6K#`07<}~rXGuh}Q+G_*y8jjt zkthE_r`cq1+HC-D>OGgp1a=ZY>N=#t$3QwT$wvZfv zIO29u8xJkmK;&(s$$+RU`I?>0n#!KgN3Jzo<+3z^%P)Wkk ze=cAWb7yJ`+*$4W9Om2Ez(RVfS;XryHiB!x86K(ZYRG+7C!0qmEvc@Y`Q`<&c+5NX zB$r^*oj}$zQn6~S*B&-He>=HzrUNT{4B1_$znqxkaTXmQU-jG9kBJ9dz(7Ae7V3MM zdr5ef+Ps`)94g{|jwoc4V-GQMWIk)xZe|y!-@t-L@+?8M0vG%Z!b8Ch>`fWMky2B- z*LDd^-_pjWr)T2*OKZ`$^E0N*^dsN>Ya!*;NLD_6Hha1>2Q9g&?7#$D_BD17%L>Yc znFmDK@RE73wD>%kF|8idY9uW55=WCui;}tW6~X9nTflwL(;;&O)A>C4GdO)fp6zMe zL|g`JAV7ORF?gMa9pn9}X7wL>j>+P`6w0D^IkA0{3k46kwA6zu+Bz*@c?Cbxnj z=z76?a???Rxd^;q63-NjJ`_nW-#SK$Ts=smLm;U#Sj4;AZ=g|464w;4UC<$0OH~(o zvzzN~!6s>AnEw3>Y`kX3T#9#~N5^|^%dxreHRC8O6qDllrr+FssaWU{kz=mqx1pr; zGVHLthEc)%ye?jrRhpTT4YT_pL;fdzITr=tDo;%~#0!dOAsJTvmv^%eSv zpB2>2n`iD4U&i(1&L>@CbnxoP?O@;UjXNvu(BVHJf&`ykblyY2kAW?4BUQj!EJm>1 z)raB9LLWTRw-)v8~2c0u{0m*uiNTPC7$xy{lSn*PoS*7NaX3t*E zYVI?5=Nu1R=8pyT3vP0HUmoIw|Fl{4z-c0GqC%9F`*G5hZt`&`ftDYeKw4JTQSsCa z?lSLFIk58{9QZYan-siZJl)7LXKn=bgJQ5o&4aT}(%?Qw-Na>6{6RmjNuVG#mKYo+ zoQa_$QH}gd1sN_(GDD5$CmSGiaRVOvoQ86%s$r7F7EV@Y2UwJ4a=I?+pw<{p!nW9e zq?kFB7HEUX{SIMNs0vXvu!kN84u-$Ig9!((g7-ED)KUwfjYXGmaN%n9rhb6}9sBPlKnnL+3s!J>czbVA{VmabI z6iadg_v7Jndtq*{7+EoC5Z{dbPA&U*C-Kr{q;4WFvF#Qkw`Wg=A=M=IVvjQIzue7f zE_sdv2{R#JQw%(exMAM8%>u{7ctCqWU(alu3d&!Cj!x7Oc>dAItPOJEb#3Y&&`inmSBl=4P8_+0J105iFaBq zREMiU?Yv+pxww>vH2PPJUFWp_oD@>}RyUID;=7jbgudj}FXP_bs6 zt0$L*(25PqIk}!xtOs3k_jQc$wuB|q-LHC*bPM^NwncXtGp{E%(6BG{YO4d)3o@TK}n8vC$%>D4S$q( zvKRcdnmpvomTXbyd)kf=x5Hyd!INp2A8{XgZ6nwn%TOZjS_S0~hN-LdBC;&>0sN1m z^YEwo`{Q_Ik0P>CiIRp9;hy(fL)uCMC6z=fNu()dlZ>nqvPr&@B;$V0QBm1Sp&=?M zG&NM({XV}x;65Jre(vYq&v~EodOe>)_lI<$;alPOI3c{h9u93vx5zk~J#62@OEfg{ zE4Rl#fxbDPMt`b>)3j|)^nlR{rruFbgZ%4AFOFoVA1l+o*%=&vs0ms&Td;kM1MOQG zPOC(Ga8vJZo>_B~&0lE193wO6{nli(U9Je(ex0Bk??`1d;%TU7FZnlLhsms+%3$~~ zy`j_ybJgz<4YdW#R98>9QZkKQ7|y^+{@37Re?G3y`${b*_t8#AYZm`!Ifk0f#+Dzx z7|L?+95;bY==4Dmi7(vrAU&oeX~+Ai6wxlapL4g1CE$C`Rx9G%PNQ*Px8yGYokD9nan$YcbY&G)>8xP8Zlj2^V>l zU}&BhwuT4b=kc$hEu@inW@~UWLoMhgqX+z+-zauI_bkWiGQq zhPv)HV4M5B;c%Z5J3P{q6*R6v_YKVcX_W=36zO#$L+^YlT3ek!MzP`bSNm6tIbqq4bCS?)cq=WYSu|@o*2^^13!sxsmo{h`5)Pr0wZozUvb4Q|c5nXHd1 zp^DCZ+)3X-T3F+Pmz@l;!%y0x+p+|G;+w#AVYeV)dMDKS5d7WHM4!Y5&^FJ1oZ@j2 zs*)H0X0{QWt$RDM7_=kxqog6ext4^dIr7Yx^RQ8MBs|_yMGn|)25sXT#OlRy`ZT6a z*jektJ-#Nzr8<-lsq!*Dhakhr-I)Oz2Mf8ZvK`$1X?wUtrMGl2WE%6EKM!`WO(gNm zHh3L=9z0K1P`i-=`rRi33}%Z$m}emwR~Jd=)u?f|mI&$XOZ*;+s{x5njUz{MOSvcG z^PqEnAuJnpk2Acc&ADe3K~DNuZg@cit<2!N;U~)>b7}?=zVf20zK?)BfxU1^w*kM_ zlZ3k3Jor2BIz8y{&ivjJC$h#vnY!#KAm6#$f;jCOVZo}0T*B(P;MlAU9TQ^Vx#&ge zbjpeII4?#$Oz#NCe-okkhos=1w;E{d$%iMi%R!j#K|b^M_xe@5%Q;q?i>)Z6mTyi_ zxsO*#gUf1Y&i5cO-Z3OdCY)$a%HU@HJP&fLbL%fvN}Z$sGq>m3E|syF)XzBH&EL22i_l7)G6`C(m9m zLGt1g-26OG@Y*v#)Z;Heno%>zt~o*XTsy`5zV6}Fmn@{S>oYjtL$gV#T#fKQb)KWw zsVh_;T+huj+$z*E2%$sesRE1JZ$Yp7I7An&p^4u#Ky+=DaP?#fCRewGO$v_Ye08UC z)+#!Xc_={OqQePG1jh8Zx)kT;{gkYbok5$wG?Fy6>F^~?6a=J27`m;G)O^w=PR=R9 zd)+}qZGNrbfa6%mXjm=i^g976|M>oXpdr_pl}U@747oXDT7@D$kNATI1<*=yfbn^g z=ryy=V5Be$HrM{(*l-p(|I7kr&FFzgOenk>uRsd!9-_YUj&KQaYeDNKe;#X>$R)nm z0Go~tK+AtUP+$~D7R$s!SWp_6Z(2?^SlUrV6Dinc;YKe02@+WQEFc%pzajd$W2k;# zB5ArRgJsi=g((r`@ZzC4XpD4&{_n|T7Vd;I$(E$DZ|`%0cCk0DW+rc_PYk^X-FKseVTk^FSAg55y{nu3=I~(Nr z9@PP_9kkyh339b^1i^k@Wclo^R9}0paM}EcusXbg=uF5Yrr#5Z-(-M0%@#aY#{s;L zt|0E0(z#B3ZK3AMcw(e#!siP1!i!t{{`D0Tn3HJBGl}{Hf3;7OAJ<&LM)w!^FFQn* z4| z|2s|3{oci8KZ_NV<)~9ZK|H1@c+*#U^NHlV4p1eJIi*Nfu4{`b{H|G1`F{5SYW;61 z1e)xH-LoE%HT!D8Jvxaq&}Q5m%Qu|%r-Pt6P)wfIchCo``CO|oo!Cz~K^OS)46Ez& zx%*>Yb7}Gl=yJQ1ya@bDs!iS!_dap%6&k_4_6U;KGm+H!h{M9mKe>=yvuS%#B)7O) zij(p^Lp4v>(PMtyFmkAfvl*1Arz{Hb0sk40*~8csFrhBusIt zSCVk8?FI1HHKzuB`J83eS-#_{N_AC6l1F}7bRQ=TGabBO>Ty{z|4$%DT?qj(G>GbGq(D=B5Mcqs^?}({^0jL0FqlrsNTwS;z`;0YN zUTX{^g^#$<2kl(y);o}+GL0Uwj3b4Uec`s>HS&;W5eHvXsW@SygB#x(QTtFU^Sj3j zVDE-XnDG4?{74od-)2pw#<$Ypu;vNQ<*7B@_;;otv}7^$HJ-#dNMAK~>Y5;Yw)r{j zwQhuir!;uHx;$IjV+O^K17Z1%5U%(9N3!Sc92)m*J~0Uy;Jl{s&hx0NWUA5^7~g#e zR!luY#DXiRxY!96u&{}?d=hYZ#}`uX`%ZM!4w1?o?WJ_sTobR|*MV8GiXd6PmzJug z@Qj2OqM5!Br@eM3^N(B7hI)6Rs#8YdZ|M_*mJ8e>s1Vkw%%p)0j@%7-X=qxfOzn;G zz<4IlWG?CuI&WD(W_HvHGk-L5^OQdkIsG6KZL31WpKpLU&f_3_Z#n7kv4OG5&U6b{ zLC(rsnkoKYcDPzZyivoga`Z>tbk2Ycm|OR0rE%`=CuOiTfOt zOvN<9xzf-^!H5GI@No4m$f{B2Oo%2abyEbx-m^4kjvVIm_fOIBja05mn^QjL2(}8_ zLFaN0%vqZWKB=PY00zUsc@Ff@7elCYSWaK8_Xg>ht6XdLe&`n~AeRq~r?%oMaLp@+ z^w=E*WvfLzvo?huPqT)r4?mC{vZ0VOJeJ+5S%%S>FQD{|KD~I*n*`p}1l6KB5PMY) zx)T~fPV+zLI#WXz*{INj1)1EWJhf&(>QV0LCQ9e-K@{$va22zj1^ zIWvo#Vx?3mNr&!VFdi&iW}`*hHkh`mkB@y&LEJwdO93SMFODX|NU4nM=O zAi;%yTZ`?N_`K9Hb= zO)g-kG-~d+4=f`X58QUf(N{;{VV5T$l4y@sDSRDAem>P}n2M>toba;S9@Jv{u}0>8 zdD>#Yr2b==w(t-1oG-_J?Z3b=pPzAUZoo{Fhj8fhPe=@Dfcvl7fV10&FPGV38=%vWaYMqZ7J} z>Zc;&+IXn`6fPa>jE-OLLe=_vQ2Mf&xRhQ5>$6p;5p9bbywy-ay$$`R%i)3hLFkeE z8-6~Lg79iZX6{!Dm#ai@zB}(!{&){ozn)7EZ0`ZhwXZqPW;?Jk)xuS$P2tvpeV7oQ z21k=;V546udbB2?#HmjBIco$ey*W!aDaYUpZ)F7b3~B~_@zcN|xR<#RJ^a&QrdtVV zvyH<1j~2M!?SsG5BEy3b(qWXz69x_T@LEuYOCu=1JkV zZ*RGDFApqSQ$U!F9T%`M7LQavg7yDmN#Y6xOnt76Vb>$TX0a;X%u&MddO4hYVn2<& zuMcnLW{}(beZv!r(ZlRMew>wp8t0l=TYmQluTm zPnrb(HMk(Be1Pw9e*?2KTQJUf9v*7yf?1O%!=-s6@U$2I+ja0nyQeiUdUZW{7!*PO zU5EsUnx*)1j5HR0y@J{kldxN@6z`@F!?CY_V4lts+&l6aq^0ZPm8r+LeG)Gqar{<% z>Rrl3{7VFj9~MC6Qm9ezUaaiS#?Zajmsm^Rm28AgbTsB zTL;ZsCApKOCOo&r0)_fKizg@pS6<4(iGQWgY35y2dEU#ZT{#Uu*XrY|@Afpea1K&F z2T*tCK2(M-!HBSIYS#aS?)=ehsQ~!=GmO{34g% zr;uhR{;cbK22BYtq4OJ?&+y(T{ zn9l+|88y@&KZ$J#vZcYnSGdMEGL+~waEA}}3lkFtInyzsFuS>vQU9R>)GHS)aI=i@RudkP8uHA0P@z;nd^KQdUiM*kaRs z`r)nvyqvGZwnoljZl`X-)Ct*w7~cr$@FJc2R%*)9AH9du8oW1s4`CsfPqKGYOxX^D zwX8W(z~XAHX+-`7x_y5(-Sqh}Z4v9G6GnJ(PA9EtZPXgd>2q{?c7Z_VVH+KQ0;;o9 zhQ8R-0^to);QQBn&`AvE##g12sc&Y%rout)-l2M$aqR=Of9izO!k$3SCLJ~>*Nr|N zZNkirvS{TBMJn%blhdDK$8+=i*`pm|OyPAXbyzJ;qf=*qnb}W4>^2uB_A`R)HjtpH zBN*8j5>3_Oo4AoX4$+{8l3b_%7WnS|jL9CE&WwLu;3SXaf_ml(_Ac6pO?hUwC zlf8Vj@$gGJ%~O-Lrt?|g53!tO=pgM`>Oy5M zo}n)%S5YhVLN3YYA!m{_hPG{)B6#rkEWP$^BpdDON>!G1P~FcJ=Ev+$2{!$400((3 z_`N-mD{vejhda;FL+jM&^hxoQ!X>v&gq^@TS3rS+FAP(MmfmTqOkOVjA+6DsE89y&~qF5E<8G=9+flyF+S?>_8Ja@v3kFHzHuj?<>lh%V- z^k|j@)wmZdsQBd1J>8%IqjYu&8^w8lO%<};uKZ_yoJArFCP0p)AshYFk6%AWGV;rx z9NIC&6`n2T45y@V-ZylZS9UNByS9KzH0#pZfGW;R^ADBO8{pna*)XxCBfz(j?^icp z=Mvwj)2@gGu zPkk+&=>L^&-N9#iN}ifG?%+MGslh~Vuo0byGU3+0Vw|CRAN5!a+VAtn*6)j<%RG&h z%!m^H;=NuYWWy_8dB));y-vZU6^rr4+=HxriLOdZ&`Xd=t1}X)>F+{t%lby`<-U?*Q+cP5 zLLRtp1Jr5y0u!uObKVx=uq7u2Klz>^0j_1(U2BUb`+28={wqKQaa>mxg^F9F@yJLQ z^!qRg|M6LcT`kp+!U^Hvt7SA@#tyjIIp8!R3ZJ|aLyX}$k&Uh3Xl3hW) zAE`6fuY2&v-GyMWYLF{FVFs7F@}Skl4kVY%feHI4lsLpd)~QHb%S}MBkG;_4{u(BY zPK9xWcfoZ1JUrj(14nd+1?1i>?iRnkYB)ZQ8&}-}pX5)Fd&QEtGN6OJ*wo6e`^U3o z!SUSMw22mDPbuQ=9%t+o#(?9wVGLDSNd5GD;r>}WZnO6o_On(O9esp2n$Lw*)`(hk zO^m|o-ZsoVf0h64{ivRYFYMV*fvxhSXT{rjcl}cIoaK(wXVvrhSHz3zt4Qidy-nv06h9N_hU_tWOiMzn0I1B%4yvE-5rU zbsA2uiN>VTRFvhJ!}I4vqfCx17OmFDtgE4DbMg~8d1x~3I1)*l_Q;}Xh6eV&%}0Y- z(fDzw5t~K@;dZAAF!Nf2?G9O$TB$Sf=Hc}8 zVfN#HKzDN)hPCqNN}D>!JISSlXOwb3@^^8Hs-e95p$p`9AZ9pdQI8@cmNrV7{pyP? zS1M)@QPNKbmuNzLZW=lD_#ta-7c%{M0nAFcgcf}I2hV3^(V!T8wpQ$_FxOOrUVXv; z=QNsNXXAhH+Ws>RHQlfqnXsLbyWO#7SYK3Nyp3>ruO#JXn_4a znB?e&z4PKo=QBQc@*A+RsFSRoxrDu2zZ30GY=SQ*JK@9Se%k5&k6PJnWku)t{>+zf zDm}a%PWP3-HohxmzH%(rH^+dv{GLwhX1TI2KeX7zV;1OcvW><6=-ug&>`u6b@aUZ?I?>J=wmf=AO;XFLY`0hm)ZF3gI^}cIQ_F3l20O3A z{F}z~{d*xba@J=D22H5mv@cXmC!X!RyMR75bf6C;lZnYtIUAPT#gd#987e8V*q%u8 zW8FjAmw1X6lrJSeL+iO0Bkz!D#vdU$BL<4kE=Sc3ez<$}6uhPy43#GOobgL(dSDyD z<=@L7I-#5#3AN`6D>Yf<>^AdU`zUG<>`SPOD*bRT8w|{gskqk_aFt7kD>|3C8)cJ) z$3K-*GVT`V`CF6CEol=hdT}0xr(GoaZ(PXqU!ipJ$zw3&p+)_@Ye;n7XdFCxfF3-m z4cR(bkbh8=eSL5Wl4n+fn42u!wdgQiAZbOkMxMat6=%8jgdz|x@uk`4i)n507%H*q z5G{;!rkT=dl(?y|tQ`~B%UP;O=FUVbt$FxyYdt*5^k9eITQa358`;f`+DQ1Eoud6p z+J1c%yJGMk+hpKF6IQfR&CM;~`(7XOPo~h{^=0PC3)ge+6=o6Me?|2deW7xiQm|=i zBUlfH!Q6GHLDtQI%eI_H=L%HVxSzXd(D&;C^|9ya7pX9|?BGM#>K9BC-iNU*-+tiH zVt2%XKq_cE%^kj1OO19b;@65Gy8GuBP&pDw##k>9I4+Dw5&mpPGA55HB)&0k4{yfG z_{-EJqM!OmSJ3(Q>Zq2@Xg1z%4Og6KF0{Ej3;%ZAKv!nK?Bpfz?WE~!>Lz#k@X2es z%y)o3N#=7E<_*xPYt8KAJE;G_3DjF3ft-IVt6061ww9=JQ^x6Y=~BB`x0Ni@n172U z%$)$;VjEEH%s5QX-;E`2CbLPM*34gw@74S|&i3u>g@&IYXz`#7+!Ex_Y_|cqv_%7@ zOf~S#6ipV>N?F+oYdmJ&LX?B1;l^&h&+IW9Spmle9qUmcwTxxSM$iX;e^T;s8j}^h zN?N__*uH)rZ01=RDSCHd`KK@RqxJ@(Y^}rO+DmYM(qrJr7iRp>je4*<^vs2&?AxlX z)av3<8u0QFEzjPLErDKie7GI!RDDW~d2i%w_i9!bxeV)zQ()7Rb#Q%JCfl}u9PX3Y zLi4mMkyz!jKNZn9wmD&$@14{(Y5OgV5cC0$u1Qz z^AUF37)+*?L-y=&l`l>_Y(y{26a~FNAr{=cg5}>MsoZ3WP<@wD>1xQfauaH{WzoKj zlXxrT52zOZG%Kh#V3Uh%SzbpY6>QngSZ2h;yq(WR6M(Y&j?(i zIivFEII~jv;*br`D@tITYX=xORZ_1#JTLawc(x(Xn=ZDb>9&$O4TOySX z+poPAkm%m{KyDA=6Zqa^fF^{$KH^hmCx9!(RV;-wTPwvk--0Zc~U%!@v=Xo zvOM>F!fHQGWN{jwgVN-E2wrrSzC4aR_8PJx&EZpPAX~Osk}WI?W^R>AOzWKrD^!=m zwDm<4_C&yE(`vzsp)2s-Uv)Ud>dwFl_>-ewSH3fOhRRVUT7qGUz4cl*Rf~^Iksd}VjmHm`B z7S<|aQC4_{9ozH?XM9$*=x^lkg~~%DKgQ$yawGh-mvCXM4A#%pgQNCSu;KhM2$4Aj zVh5Bh)Z}x}Ej*4ba##Y+1L|-oNyu$@_YfqL)$shk<=_yg1DZ0s;BjsQIKP%dF>n6A z;T(gnJU4rW%T-(~W{4qbGT7niPWEIafwNi!b}m=N4k40dV z<_!BT&4sXK1~{mi0L}*EL4LP3jy!z3avtyPlD(p8e&(tX-tj$zUUnJiW)KKB=S5|~N#24axp4F)?<^}IdT8Zzd&)}DPi96ZW zLK;^*hJGR>-*ViU#hWVfUx^y}e^2594-A;czL*6G5-RL#Mis4e=Dps}yI}qac}zOG z4o$4;@Smj*YIs_+$BScFgi{ZVkJ$vDrdu%anr6DjZW-LLmgTC%130ad30&lJUA(K7 zfGrPmpmU)*`fQ4)tDO1f#I+m3+Q;Gid1MavtNSyj`{N!+{YfE(VyoeE=zP?2;=ezM zTv#@o0b`!u!wPT8xirR*kE%xK@L--GfIn-PRQHc-TKo=nIA~$(Em^pkbq>D24&z>_ zpMcjg?w`*mfAfV@L8c zEo?9-3^MAB>DYdI5Ue_m4eefBS9mftvqx$eD-GJm4&cqhSGk>v#x!A(G(KxTAPjiZ zP5-Pb7CzG4hp|WEv6o-F_{)EWIQ;_XNPR~>1S_Nb=5jja+W<&DKZ?aaj}o;-QCP|g z|K{Npxcqb`?B1t~*OY?Du4})bE?kt|kG1f^M_ZUV&k>&-If$Rkqj2pRTYR+q zHf^@NK+_`@QR&6UaEoI%f3ENnF4=ZK>c+k18RIjczThpGAGpr7C@jLPG%2BGt28Wc zupyJiwQ;{s)M4Gjep1-Hv$ERkH;hs*qNba2>DKZfSTQIC&#%4$fxaT92lj$n&UAQw zB&o8tysvWK#y;V@^UF~;WDQ>JLo9J=gcFlRaFvZNMp-aA<9RAQ_ag<$X&uBb?I#sS zf>;Yb-v(Wd#Nk=faPg)o*mHCXiv6v?E9L4iE@V30a(V@<+9gR>Z`P$dck*lnfdZ?t za^|^Rd#I(|1j=Rvva_Knka;VLcK5`h@)0j6uv242PsegUN4}y;N3ysLHZDy1r8k{w zqKAst%xG=UD$eccUG~>_JM)Y=$rkopp$9Jt*gbNc%FW$CZ#GY4K6y3Ny9dBQ>N5Q8 z`bfW?9}Qpj)^MUR9<;}JDv`QeNP}%wfZqt@{x*$e+G`eaFXC@-ke?uJJ&Vs6 zrRidE(P3I~rIHu5bzP)s9ogUsk=*-cdGPv`&D7bTb;vvbFHrCs9<|5h3)Qt%c!P^V!@MRrHv`yGk-{ z(YOm*AfauHYu^~r37-zoWsBD^+m}voKck!TF5!fU3X|b!UXtMZH>E1s70GOyatwPr zX+B99RZ6pqW>Xze7d)OPhTEK4xo1QI8jkT!m5;hy)NmC0mNA#5W%@AX#8E8AzxBM$pe*Z@@h7)mFk`A~^9fq^vYFPK{0~vYH9gj?!LNy$~;YYp8#4@CtV$K&jSS`-#f26`evsd)ra!+oo?0o1u zcaJSGI7kOO&T|W561bJFjWmADXZrDT6m{ET%%07?N}o>h7mCUH(i;boAkV@Xri=(i z&Ga&;eY}r;8Z(g&#GGbUn=RP=oBGVYdl+S8ja8Uz07? z^!YIpSorEF9bS6?g8rq#)G1IBlfWpEdy2`wV{uExI_7)4cuxi3~A6LhFAFp#s4I60o#$V=}oG9_TJQd@2 z2XIei#UL%k8e&`ZF-~&?)hL~e@ouxprDKKc!-cK*sv(8eKTScA-*bSQcOM_eY2aAj zsc^ui7tb!;1T8my;)YKbX$0>%=@6O`Z%lwS8JFqD>;mFYHUvhttEj}qd`|N8d1yXz zSdcTL8V2GiSLV7G^h0$pZ0!{|eqslV9_$kYmI3u%6^f7kCXoGSqw)H7Gd>^tp2YZi zvZo!J1^TInQBP|m>u^y+&Lx}nE*XPCFHHm!-Us3=-l_1O?+PE!3?uNUj2i&Ye#74$UB$ zHhakInqbO56lsEPDiixHjVC79;E@-j(Sd)>@`H0Bt1Sv^Z!CiyQ+HC6iE{8`#!pfu zE}&x%rs8g+c#zcHOwN6NPx?1$Q*ogTi84!vovH#h7GTw(51-{G$bv*T!J_r5Bu4QK0Z?yMj=@ zT8WM6)4+AcF}Nq@JLu>Lkm?PRuIGNB=^Rgsx2(p*jB!-_-820DYXZJ<3&fdm_Auw- zAqZ}?!VfXus94%~uASf0;OG@d{gQ#Ntz==~xf(9H(o0p|a z={02`e7ipmS1Nf>!|RJ-1mDTL`{<*Y`p^!_o%VzLaBWz$;X5aIx*el_XtRsH6X4Jk zA?AgSA(!5t0o(lwc#FRl2^z{UG3PDt@M|R3#MnghK@gjiPtFS_LdRuaJaSeJmn$yC z<6HDG{$C(o%j>2Ow3XO6h28YUK2anxS7Ex>eEcl88W#mNVAjb$@Y24DG>06a|GvKz zh*ub)&gU`sPv#7&{8NQK9cwD-kRW{JeS$up^%00i4xYVr2wpzfiEDNYKzpDEO`C9t zCZ7vIRc1#nh@0T`zGpbblFxyp3ov$uD!nH;9sfNDflEAxM@zmHF6`M4Ss!IdRdPM2 zyQG-@*nEJDu6qS}ea>K{B7^gbV<}M?$u`vdg3zPK@p7UH)+|ecb}dxG}fL$su99!1!?dG9*5dp53U>CkENUzH3*G-*X2%Lb8J$(hWUPvy zZGkvuObDYhR$4L7sK=}?rG@D}n?b{DLIi6b&#KzFc?w90sWV?gAuTf&kha7EVpcf{ zPxxw}?Z+f~(fT+Er?0skLz=YRd615I&WONnKs=L!(~B%Fk1f$REii2!%OF~ z^z|2sn|3}c*0@Zar)jY_%i`FPEZ&1TB*)7A7NfYFIlU}Bfp(toWqZ;W!8yf2x;5@Q zfOrs^FP3KO<2mj|zI4?nO`dTtTm~V>n`zj(N95eVPWZd_0;isD$;tyYm>|oMnRm2v zhd%eQRaZ2sRtGI)ML9ni+1blZ1fOK4QlsF?6KS@U_nJC*Co-9Hm8|wfF54^Z$>fZ! z2(jS%j~8FygGV>u+q5xk+E{JO(|Sw)+Y!oK29k)e=T`VlZZUaLm?H|>9F~;GuWkG zbEX>1Gx-+g&`0;|p==J%vpnp>v`CL2v`LkfPnlV@sx^UaPuFHg`{jkt(@wC9T{8H9 zGi4VR6tHV&{|b7l=fJ<$PpC>zAl?7$1B>`6!QAE+uoBT4y7c`F*uF56HWyX0k)K=G zOEC@BH^+oV-sCgo3p41HQMYN!!Y-J+zMN}0{G3~=7S9g8UX5GKMpjw353(N52s&!Y zFo($!c0($# zJbMr-Rdsbo9_RO7mnaSEGCxxbW|vaQ*`HCx(@Wy`evJt8wLHjf4{yPHF*%@8P{>AQ zJf{|6y-@FKiMvjWVVid_X6g5UDwuds>E4U5#>D}b`i^CmXFOO`UllBzF`NH^O-G+A z>*(KCacG3J9*%)TQya%7NBeloaN6p)Gt2ARyb={z-F^N6m z*NNkQZD-zRrC7&;M^qT9#vYfLv$L!B@q4k#?4e^J%Tf4AXG@o{z`3n-^Y5K3*J2y1 zJr)Dmo3rV*kHMHeJ(N{^-HgUE$-|2`Hn1O7gaT5QUEUbsX5$W#0rUBH&xoMnF5=2f8{-mLb-C@h`4g&FK# z&qP=4r{%LauK!XM-68iDwr8Hf4e#8kG2h9Jda?vfSB-+<`5U4B{wkW4#B-v=H?rw( z)Y;6=3|L4{kykrc(Ost$n4e4|k*zESDyS8#to#c<|EwjwdQYk1_VvuAljjT_E+W1B zHT8=Uhqg)H?DL7&kfZZ~x^FQ>?=$aZZNrHqPGNvFSWJZpZ<8U*vXUrN zj<>izJrNDut#E<*AMV$zkx(!Uko!iFc(w7o!8`7dnr9$f*f1LX3v2w#?v{(?Z5Gygw18tZfVD%=tFq%(*pEK z;pgvjO89L_5~on-N2FDapy|^kc*egc@$+Tm6VJOfdL9fq^OZpD`8J~ILhzm6N^o9$ z6t-qu;yw-7!rf3YOjdfux%^NdeObJRc3&Gr55x(U&*3=`UGjJ&k3oZN2Y&33#CK6Q za1GzXT)$`#ioWV&sE;medq17~5p@>6hT5as`=wZFYbJcX=piJu*kI5+HOzS0i%%T% zEJFMaqqspceE6k_D>jItM{hRVZph`#_Z2|anUQ2dMkE|v?E=r{#Dm0x+wkCoCoXRp z;vP-WgoR_aL3)S-TJH_RKE1KHUs2xTPO`nlB^cNt~Sh$}`X&jdVKpM%v`vKvWaEl41oVCNqyc0Njdm27n8jA5t^WfrRcgQxK#~BO_ z;k`SqsAv$6r{>G!_6=*`TvZcjH_2K|&7VtJ`*ZPDz-e;9IF(G=6hXV&ALH&dr?G~A zZ<`G`Y&Oy`gSkY8V5ty>>b40M1Wai2jW+Hkt_6PQul<$(+{C zxoDUef%$x1FQ>KyRGfJ4lv@WBj?W{We%ai_&O4RACm?#QXoM#`Bi6HE5?1K$!_gsO zJiCDJKpJ-9tleKQOJ17aBbLIJkZ$rgsfC>Py$@{{Lj;?P)yTlb3s7L~4$q8KA?43X zb91fBWWhEOG;gZ_6LnWouqlR<9si9^v-!?FeUSrmKWU<;A_u|Vw?Vm;XO&;@2Gjjo zaOqAIi1aI?_sRqKts)Lr&uPR3VK*Ty=Q-ROECW`NBJ< z`X67y*b-! zgMm!c3a*61RV!HKLIJ1b_XSdxi$m_7qdX^f2qHSo@p;)96#iIGb&fBhl07S7r;|9& zl5ipam6dZfo5M-3xIVU2&R}n+q=DtwmGDK+5JyHfa{lcN5WKGmY#!7Qn~YarG87Ch zZF8XQ`EN)tv&86V1#bJV_R4LuD&ZE*;%7e!7N-?ZcH#uC3}ZmEG9FwNR?)NvQY<92p1!DY zrWK9ebacXFa=yMEOnH`outpnu&NQQ$?=sq2b{zL-m6JunZ4mB}$GZr2LSC8#IF3_* z`bQC15mZYgXDp$iuoTVBW{`mHZ^F@{qj9F?2)rJtg7=@5L1)%QE@tjOeE(6Ed%GwY zmbJ^lv~w9SJXZp@{kIMK7Mi1?;TUvv7>O@={wU5l0?Vh`;eykKxT(<(9!G?dSEEi6 zZLwSuY*PU%oK)$_Ee^1cjKsmfIXHGi1Ww+%k<~c#K;3RXI-G^v>q<%bX-zzF63qh3 zvP7ubb_F);F2V}W@9@+#7&8{O!e56lSaG0}Tz>hFtluWi&!p$fqclul#+C+Rv)GzS zT+btN%kx3vsiUyx%vroJH)(Bga0pZ^dF9<0Mstv0aEmBtwX>5!av5p0h<0;xU? z$h&r(SUB2HHVc zr62lM_@P~l6LcMkrCMSAWcHmj8aMC*N?RnEu}Lnx6(&Q8`C<%8R3~vod%1?jeS(cd z9Gt;b5cYT@&}IJcu&D>E7Fyxb@LeQ9d@n{m?-x9%hctK`n){|gW>+pZ zAT-6h293o1PYirJ=s-Mi5$O2;f%ChfK<<7Ls<9W;eCk_a*X`NZl{gKJSM#0yhGuf- zB^6i>S;NG&HN^8nIQ&XUgWVTA(X&Js_jW}<;IA53`e6i1-q;KKqIgE~mZd1Vq>NZf z#Bh<_U!j+OUfiuy5((Tg}# z$9ItGwnCS)3DzAQ7Tz}ePNeO=5Qms)P&!bAXU<6B>n|J1?mjff9j8)>wze_*5wV=6 zKeO!2_N3;MN{Pt<~pYsJ%SPJ6 z6dJIwtIxP3kJYSMTb#NM*s@XKdzf&q46EE}$c`nxJP*-V3fG`j2#jjoDi z!h%&_5taBR4)zoG>B9ZKRbqI)Ldb8VKLRjdo3+&;g5;l9863yJ%OQq9O zY0}~prn%mhh3%aU1>@dwj?ey>`}vKfkpgR`GXv<#$wu73kw?_-C*S`*q{Cj`&7wEY ztih&=YKRiPWiN_U*qk|8boA+q^zsWG_POUXyC9$YKZ?#XkgNBN!-QlDWn?F0l#zU& z`&1%rlG3Cynj$JCMcGnjWb>1iqG;gz+z+X!R4UPwwn)-Wsnq}c-+PhMan5<}`}4Uj zmUpp?T|H+m%sy}!8>(v1|CTp|elmx~7Y5wi=!-bcZaovRtsp>=N#6j6{h#1-d8t7`@k$Onp97u+U=D`mPK+ z+%a2^Rui7%+QFGD;jR|Hr`v#yE#uh&fe@r7XkhJIC8!_zM+<*dVzGQ9 z)6UPu#Q~y^e`mtD%BimU4mY#9qYt9pP~F$SJtJpl(pI<;i_6w9OCod3)b;DgrwuR z;LuTe#{M*;a!c9rm4mcX!yi%}pT)X;s!TS)52Fkwu}wa&u)}>2lDo?w`?4N8Ug60G z22?S8j5<4ZD31B|xiQtk?`ZQkjB8Jjfw(v$R=qb7b@p5(o7YdrS@z;ksk{^$`bc%eq)-z;z$mizORBR)>HaljuDJbXoDKXUF^BI zDQht@hYNBOQM+CR*F@NX%TL~UtZd7Un;+xnUJ{7;YoV`*$PH#4=Q*Xyg%=2`_5q3g+JJ**L`48ZaI5wydPfYSMq(r1DtWa9+$dEiXE;Q$!87A zFgup_A2d+Xcy-mTvq_*{0T2#CrWc5_ zou)F(<8TVREYM{=RGd4%)`Y(Y=D>enOoZ=P1o}TnC*vLRXp)L7md<~Un;?!#D}RBT zE@xnF#~HNHI*9erm%+XGIvn}+5&oPXiCtqCLdBQTIxS^&ZvGVuT+?goO&r|r;WE{@57m`-_czoD-6E>S!(NWQs?BfRkTNDsZH?18cfk&^w@1_Ny`K%n` znLd|ZBhJ}r24Ru50jg>r!v&fda5{Nd7+e|!$<~uOy-N|CO_L=WxuxK7&kX7wKAWqW zmP3rZO+fnYW4gzoOj!HoB`w}`v(El!2UNCckyrupT{}0QEum2zo>{-CY z_MN25wW@LP*<2#~FpKQhd;!g(d=~M88G9_F&EAZe$xbfV$xUV>vG%el6Y$r1L6bG+ z>Nvxgg*VXRz%^JX$Dg?(Msr{1xY3NUn_y}FYdBHcf_Cw%p~+_nJakoH{+pd(YKa`ZonMy_!10bC`X2; zj^!RORVusjyI_QeJ^`@~;m)F9xM(&6gRA}V+qDebmG(%mea3uY`7{}3zH|ZC{pmRU z=+;PVCMALJzB82BIdFMS{bgL%J0I@M z`Hn5}4mkP5BGj=~5&Wt@#LxKUK|OW}1g8*OdU+nRsR?1fjnnXqZYa;4a^t2=8Bfa8 zR57k<4(1$cB2y&i;o$>y@O5t=%Ey1hclH<1ZGkDI_+``QR)@g3cLRp-IgayBZwfuU zrRzKlf>+81$6YUR@`TmgG*mbt=DV6Cpd2Phoo$oi1WR6P^0;LUTM{B zHlSJyZE03a_TxqL(urZ2KAO0E;%N3#NeT6~ABHXY>BQvLL{!@Bj~UyP>-)zk(EPU* zxMuwi_gL(F}NYzX`YIoAEw~ z9^9t&2YaTOz_?}Jup$}IaWC>@}K<050S<`Y1Y!y$z1M$||0tUR^R zR^UVJm`k_}v^}5g&icU~|WAMeCcqb9aVs}(=w$#Nck!7N~GK7QKr9G;mSgHr1| zWSaXVxV2E1?b4Y92Q+y;(a}*LxoH5KTBZt;)Z8J!G8-4je};x7BCy+8h>un$!h}s$ zY}KiGP_!`wg()wA^((>hwrFtVPO)PuwKV@QKijZB!0!F-BExwfvHx`pyWsu{og>6p z+owD>yV)N~4=%yCrB}(PhXyQCdIof{Gc06@GBN`*>^ZDa?^@an0R=oO-Dxsi{O2-l z`;?9E2JQige}zuFG{LnoA3Y!z9qcSH{^E%G!mJ$pZFCxj`EI0hQzB0JVSxr-Bbi5$ zJliI6hiJ7Zu;v2_O!w%bx<^?ev}DmDR-0dk8-m|3rM3JwqkaLddahQVEB6+5jw%$& zCZB^~;Xf+wngE+bCSrE#a$M;8nXY(F5p!s7yn@h?Z$b-S-%^=GL5k>V=uT0FJW_Q8yUG%o4Kca7u-};VD+9=+_JTW z+_M)YT(-R$rxwu(hkC?Vvzj&&``e6mMY~v_mKpcs*hY*T9KlAaC)=`1ynFNFc8uJ9`6mzkEEiVr?`HMOJDH1KGBv*O z8dRi3x%-9^%*%NY<9F;t*>W4!dm)(lP)AmJ@F2MTHD^h?KGHlF3FZ)E&xooRTa%@N zPxRH;oTn$qi0J1~WK|3Y?N{(Fk<-F$=|-&Z+zz_kAsfCLX4E}DU5q1wp3!4*TFiea z3=Gv@LDhq?Oi9Xxj+QFNsxNB{v&3X__VZ>8h9 z-LA}F_$n-%^Ia%VJB9_X>)?Z_5#ApnY~$9exIeR%HYNM8=8>CW=)4KH)&8bn)qyvH z`GwQr{>fAvoN@}&(?4UB%6yE-{)~?m6q)_6Ht3&|4&I7W(e}R^AX=Y?m(Km6!5*H% zkAqD#r2Hg38WfKrufnLP?sj}S|1*8HJb|-Xdk}oZFGHYvBkrH(!t$)Ma7A<({tNWQ zDf3Q~z-?dXs#S`@h4q%KiyNxDv?qm|^=vM9ym$@c$MBgWG-Ri;jp3TDqJVRk758!-P~ShQ0sj38pFRt-y!(M>jGod3}1CM7vK{8YW)Zsnk zi==?!xefUHM=RQ%*@lL>p)Bmk2|6&!1a(}cn8e5-VUX{0+`aPwjT%)$UKjJO(Y|}I zu-_hf&m`MQIy+$1;#oBMat*ze;4U0mtVH5#G?BAt1-+IWbe)g{XQ&3(bJ-X&rzfJ! zWJB)dZE0Bi$P1L`oy0qv6UafHfj&Qe3lsW2z`U=f=%koMwY}ZBQAT6A3z?GWCEUyz z9J&Ff!{XeeUvA(NQbDJm{Ei2f$b(Un2IgOBre{LUnCyCSwqe^5Cfm-ztmZmWSY*m< z{EKj&+rj4db@%eB7HChL$HX(TF`G1E=Mf;JOpLc!ckdKO=bm`gXuYrBi1v z(PTeAIItgz(wx}ko3tU>2)e5HZ@`9VwwkM<{o&VfQ_V5jEn>*VH{Zo!$6Sn)swPg) z#oJl*m%GpLV5b~6dQCs1x}QPa0u|1)_%!N#yoXK-p%6I9hJ7)!1>bR2 zn8UNLX6-j;%6w*XbAmI5e2PIO^S3x~kiWJk6KH*-IPN`i3_nXm;+x@7?AWc5+_T>i z^ad8g(y3;+fae)YcuBC=hcDuGZwMWIVddnE8qi zlli6*Y>urt8-FW-`964$Ns1@2N>@cF<4J|rS`?vw{eKv#bQdeCKq}!cS9R?Y)ArA4DSmJYJbs|=wE_Xy=?p%R*x2AA_@LgriEptu=xfe z#&RU~lMavzZ{yX1edh4;7S7lR>0< znm98|1k;6Ta6Q3{6=)sDs}5hV`IinGs@jZSk8I=JJX*M0Oc#~UOR~&ggMu0UFHlWT z%wBdTKxct9GfawrAgdSXd|Hh4Tz`OD{Ux}R1)He+TQ#oVYd+f~AlegugFj;jS?}NY;bjT^u*Y?`M^{?+ttKo|h%>v-^*FT>3#?D7xW^=OIEj zr)3yY8-R~S?m_SOSLmW8TXB1kJ0`hlGR3XyLCx+U&g7pVi^iI>r7~NgZr=gC_669{ zS8+_zY$tOHQf5Cq8*pC59kPFOE6E#qhenUNj8Fb(FzH{`Y;Ug|oQU*f21_2)I(7gq>Wi}%Ql@N*OER&uzJ)lxO!!s&1Uny@EzlNKz>tYz zV4o|^HmU}&LWBK+oudNTos27V)$$bl@_Hk=`&5t344BNew5YNao_jf)-ow}TQ}JQg zJk)9-%=%S3ER|n^d5f=5mT(Vqx*7#p8=dg@JQsR+cnJNk#Nf}-JDJ(~7;KyJgO1z3 z2h)6fv39#DrvLP!Ull&^{EvJpEgQfjd*eu#=oy?Fx&za~^qKKwq0p|a1Ua&liEpT7 z$?--kaBC-79uP!tv@XY(+HN%M9m)Rm97ahIJILvPW(1Opt>p@U$BpYjn@z0 z?};H)aJ>S14X@&uuzOU0LIT?T)`gdB5_*c((8XJHP;c{T>}ztz-;e*J>l$aG!d4He zZLfi9B2L&X@eTd?do4V8B-^?v3g?-!kF*_jj{)y7pb)L~#>3R+kl5grxY#Qy!ZV&>KUtj`Sj+(-w?#pt5ldV3^a z4`SJq40OCUft7yLWd_ox&}^kF9(f*ul@k+jqi{fBLmad zCt>VgIo!MNKRQX6j1^mX7yFD4SmiIpCaAO_k&t7r#{6t)xh<@{u8h@JDzLLX z&Qv8x8Mb{Lg%j@H!=Wa=>(=gqM9!7mdNvQwEKI;xeii8QwI9RxZa@?HI?76})9#H( zD-K+w?YGsip)QJ-21KolHNl|#Ez6jHGMp4n_y|#w8I_d2e2|PctlWsM- zEj)Jm9u}E2PQ4Z}8M}_WCR!KOsMqCojFAgMot4L_-T5tow5&O}G0p?f-n6p{8ui!7!FM{2CVeb2MkT_r$LQZ(a`=Bs@}-85FfUnfiBc6#$@6DlvY!aLVm z&|bz}IB}#Xs`!Yai^>#Qc{Utd8(z`V3+m|h%toS8Zp5j5Dgn#hk$6|86z=jYycV%p z7$M2%yU()1HmbnwWVW1+QzWc&AHfAMXSgvC1bQ>M*#YgO zsA)P4xllw8-FioAti0J`-ApLCQU!K;(X?(uADO&M4n98UZC`=~UoN!E&!+3gQo9CUHgg44uh;3JLc`~^Gj_cO@vUwIH>$G)e?B|YWeNJ0mP{pbBEGDgB;6(mppHo=qIgTig2x@m1;Hy*va18geJ)@`tC2k@K7^>_?I zI_rb~WP>>o9al87UQGi(bii)mYB2S2VXkX$pl3?3aG%r}c;^^Pw)(o0QvTd%Cx3>s zik!$@p0*jb3rawuV;Qb@cuM@P_`>hbLtu%%E?AOxS!)F z5uKt>HApc@em0K;xN7na?i}JhJQG*Xjze#Lm;XL&EQfxFxY()&$Z+f@ukw7jn{I14 zsb?viEBQf7A9RuVnlqs+e;VxVXam3JiGuAWSHZMq9rxFR_W}+W(?FYY>if_SuHT!& z&0G5q;OQlzT%*R;Xmo+*CPOwy%8pBSx(XW?^^+#=W^Day11&j~yidWL`#r^gJiDOJ zg^thUUJaMy{22z^mf1W%V>plJC^pjJp#u^u?TNn z;ymF6c{{_IX~da>k!1z z8E0$2^4B@K;L}k-^)m%7w_qGswbTxuzaPn+YdOxo?-O#lFH5*B&5c~XWGJ_yU_3b8 z(twjU8F3K^vHpuxcA9UuI17)?(xl9NLgNu`^vtP5$4)lU#k~M zb&%$wg#%!Fc`QHMY806NxJ7bJ9;4)(Y%t5I06Et_NSYsm5}UojI;4Q_)E(rmPL<<) zvcqZTqDFZ8(FD_09b(zbBH&;SKmRlSK=)5Cp{;3iuxscjxW6>x9Mi*Lzrl3D;&gKo zZF>ZxCNJUl04CfjA)jZAU&^GkBgvOcM%5` zk@c^Ivy}nH`)K0`lLM3&3zGGHZDg{~BY2mx140ze^1FNil$!2?r+enJ?$RT$Y4|g7 z4t!0_+;j2sh?gYgLRM||BrnL!)WV2Cb=%h|Z?NF)augpmg)ZmkLE~{ZNDkd4b0@~* znOZHrFU8L>BWpkmC-t!T6HZf zg#GX;&QD#Zw;$JjKC%=IY z%{&uvq$#+*Dy`9-9Srwlsi4KB1wNNOBF=4|H1hTl)ah1Y^(Q~@K8MGkA6re$q&WDT zIEj>|EW~+RD@caw2%+bWd{FwSNT-bS#V4T;>CP?JaG8}7??N^FVz?Jeu#qm zDUL|zYjUQ~GjW<{4$14BjTP7ANlQNe4wy5UJH0_0YdkXHj^<);_0-4v2?BVUJXKg< zKY}SKW?`YzW_T|8gd{izfuXe;ckgE$?0NmA&UwfPU$|tzwaq|YcetW=-ZE-=Ad25r z@#m>vF_>jw3FYcV0xJb;*uGAg9$M%@oT661y>)@Ke#UW1ByYi1ZY1RSErqHPS+HYP zCaM`rVt9@p4wOBE^lnkmx^xtc2aUPzUHrzb{uX{bok++3iibe4vEc<*g)el$H#_jTS3_)c@6+jWcKCSe94@4!5}Y^iF7@%0-g$3MZ)@jK6X9aG zEujO+Z+7E>A5t*o>;+-E##q=B)GSbbOz7g28{i*33tIOFkS>$?;8%Q#MB^B|9I8VM zZeGWbPM$~haSuJS=_Kt`v17k3sNQnJ{Ssbq4;0w8*ZeW997S1i1f$JQi1+FW4s85j+6(OQH^HeId z>`Q~U=cVyi;2MZD$s^D1b)dt#mpF0@C+Jz31;y2#f&-65K=_sjzMn~k=I6V?nSaK$ zJlG4PU*?gq0b1O?lwfF#E{3_?Q=orL0X~&)ffefcI4ID-`0P%6pI`}F6T)#gG>Y0z z>mn`v<%01J#TdXfpvA>udgQeRwzeE2V<$^;<#+5k@!UwlTH=IzMUUYdUwvG!F_%1B zwi#C+?;|_ZZMm)S^}@9YBhhYGrZ8m54h za+>nNFuv_S68uR7b~rZL{_wY|Yd4d?6)T&F=lOgxUz&jO94$e_mJaB9Iu4}rd8fjn zMZ$GTUt!f}o{v0cHMcrq5AdmI>JMpfAYYd2I%tWjAH~oOg#dlIb+GBB73VtH23yP{ ziSPZJ;Nbj$EPpBnUhi&VL+m&Z%W@z=r!6?Ov9hp|<`aW=W@NL`>tg~ASFGwxHl zl)%VI5*#kslXWqS{_T4UEiLDW+sbCJ+A<6Q`C0JA{t1~X`~*D?d$%7n37 zD{YkpP0$n1?<|rM;L7bo@Y-n>v|W+m9b$8EcDOpY&%MkuS)ai6_pX8{$%Ei0mrAv! zq`|v9E$E-F4$aQG!ZrFQ1$Iw#xfy${VcK{TqULv%Om>sT8`H8OP|FJz)c9f%Jx{Wu zN?;T<;1W#w=*lUgoTJBKxLcxt<6J%iv`axSzr#<|v4GWV50`h>5h+$C~T3H!=jPrHWvpX`EoYuM|{2ssOPHL(t6@_#Ku$?N zfK}^@;l|mCD3ix9u1QqT@n#LQXqCh3ohym8HM@jtIw3u}zZAu~yz^Ew#= zj(Uk?%(oU{sG~j4&Par?>D%E)YBE_~r4Oqv9|V~v6|kpO2_hRV!Men$u%kznym+*g zo<6Hiw)fl7<0c8{(VR%d7cGJYodMGO={>B_Xdx3EqF~plJT#l>3s=1?N8!t*2UoP(ENduj4;APi|7hJcCA@NTI)5erO%dphc% zK4?L28ZSh_BxoIl&Yjnaio(~`QUmkqvx+EZpieMZz3HIs4bN%yTYd3l}(*vJQ z;ZFO9f`UbRVF`(!dTnJD(RHwZuKgv@&|wM&<5Rhy@)YoysK(t78biX4JqGKgvtjbr zS7b3)EFnSo?Ue-r;a{XqtlKL~5BzJRgYVNRfUO!)Y=E0_4HnRj9|fq9kzHz4tVF1^k3 z)~bf7Q;9xTbtN9Z2hQWpI9Ce|)<|%*ul9kC#UXH2mEdx6S8$yk1sq2ALUvpzxA%Iu z;EBG4@a*Kpkg7Kgri&j3kq29$OIDeFh;!sDuIz?uE27~-wF0DO%@Ig`zYbeQw8EZh zepfQ`7`Nr~L2iwkDmPxypZ6hJz{+JhoRPIMtWdcR3W=@6zY$@P<2~RvBwY3G`P`6XKrzT1t?1r2p_%yU#&TQZjcA_CywHJuNhP0 z92@TKv1*ub@D#V?#aJ%=YY3-czLZN@;{~~$rLg>I7k_?UgX$f!T!>&Ump-+VZfH>D z^h$2Pkn~%48Q=yZzDdKAgxf&OcwXil#wA|V!RlBCxG&xX8fQw$!+Lw}`i-MxO}z#D zac`k5%kRUU-tl<&&jvV_x|!DIh;sv1Mscg8B)F=Ashs}qk(`M|91K;5^8FuwPFLip z@OgtC_Q@`Rg}Q;9Oin1dH(DItkDU!intMsVXdD#Pj3WkH`yu8{GfXj#7MA8DLGG$A z&{?vAo5i!}XUE=!a=)$Ir>R_BPSs-2yz~=R^Gy1L&~At^2QDVpnp^5uPM$d0b6Lgu z@YB`;4&7e?sz)U_u{l@iNhKkAh7)d{FYg>`$%ODDi^wvO5#)mP0r;ol4_#VQxLZ>v zaSdI^;i&pFLF$~d@QHjT_xF^7^Q;VfHfM;quQeu)e8%qaj0%`F;K9xGy$G61d6&je z6KJGXa>^ZVfd&bc<#Oy>M0f0WO_$Unbfvpj%9eh&O;I0N@?Ys2o}65M}D5}aq} zXzqodH+H|-0q-&!;l-RXxY9OA9<6u>U2a7XZYU13Du03AF+*;df-HAEP`^$oW+%M= zdILn%HNlac+y9z{m{ekI zeg&gD!kPE!b1ZZ5bhgi@Q=nR2%sNVn+0K~~^^f`7x6iM~!WRd;_%5gxmzzHspUiuN zo_yAE`*e}|i9K7{2(xPRKEH_Pl4vt$-ly!Jzm;w#X{0&kDJrJ9FiXpkY>^nBp)`HU zv|`7xxUb__P-_qCrUc9WtVRueA9g@XlDl;(jy*ek3NL;(1!d1@K510WOUSOi`9zJoirDxG0c^V6NmOfbWi;-6?XXTB8#z4+wO5O=!W~!Gw%NBZ z?Lab{eOHAGm$9jjc1~i0iIZ5ZQvyrW2xH5fm$Lw;ktn`5N1!)Xf%R;dS}!)$y8fse zVS{^jfS>gT_Db~;-XDHI)r!k;!T|3!T=Sc~_!!P!q*|~R$JK$htSc7InL{jaW#IqX-$Yg0C}o|bRS-s%E#ZasxhCk33s}+0m>u++No-cK%2=OP#ot zsqDFehaSlYe}DeR4t=@I+Fi<-r@~=S+myxRpB-kN??mg_-(#$M&>I!<)7W*>&#dD6 z5*F^@K)!FxWzF>n5?$X=!u>G2n<390rQcwqoG-Av%}p$TXY(0eP-TPI$lTo0*{yNv zY)SA8_SC!{bCU8w#>0-)Z|||K)v09Dudid@|8=3lmEG+A76~R--@``iNoQK}&&V@Z zp1ml~XP(qN&|Aiqor;%bPiIxIA?c|Y<#!FQFAQM)_ir&XqgrgVa%QuhShGJ)Z`h(+ zE16%AJQK;B#a?}1#@Lu;Ec?V6bYE4?*5|HZpSxNyGkg~N>vx=)oQ`5*fpb|!J1|fE zP*y+9gGH|_V6vA_V8YykOvlfQnI?Z@DG!Em>Fiu4o2pcQMfC;>EXvsV&N5~%X2KXh z+gh+XmuZjxk45adgZ?^iX>G6!hRu>=0TDY`-o#yO;A$>YSaTX0v$X1zEso(NtstiI zbLhDtO*!{8lOvDS{|MK!Ru{a4XuxCj#?ge>y5En3~q#nqD! zkU!fUsOjS>!G6{ALa!xyq<#Gl!RVEn=)<~j@Pc~59)%tfZ}W%P20yJ;yWJzqSh1MM zj+{+IR|eSL7ZHNeJr$CASQ6C6pQB$d?jWP)?;%yY?$XOAlE|4%BWyZi2lgLYNSTql zt&eQ9@SB+iNo+Vm+tsA;Ow(hsFWr#L$%_#r@Bc=e{>2LlQncaP6FuHrU_nf4{t7Y$ z4+Q4#{t(SI*0tNNN6?TjGW4d$4tP}bY};=coCQ!n*w= z)#JSIve*$r=ZuEWM;IvWq{X^eNa5Z#AY33CRM<)G|83NLE-Bc#IVJNCPJy zO}5R5CVS@I5m+xdO^lwdAfKEz3b!wJr%CVlUa(sgG3XvG9R4<$Cfs^ptMPLs@$o(_ z`0W})*OfgMBt2dXlA$N)9|A9d-O@i*whkBN1u3Z`V` z0uUbTI4e-PkcOfY7Yll(i4g5&XN1c+8DVf?lVG*05JD1Lgl82Ad1tLaKJAGU>{$0x zxJ<(rcJ7%%Ms2AghqBb*jYz%ilFP?zR|^NoEbqtkW!x_^Jm;9;-hWHzmZ>?yM*Vx_ z#QSuz%{Nc5f#03t$68Ttt2a*ptksZ^_tg9#lVnvoJh+ zEDU^k3^f@8 zQXzdQ4?F!cNUC8S$?=sW8IQNu)o!vNqb)?q>C1zp>2EtIY_ugGZU{j^I*uGzxPbUe zou@ieuGM*YwhE?y-bu=|=R@@;Q~G1>HyXd{u27vVf?uljb(KHNY5Vpz!NOM$iRo3lx@`WvCU z#00YVd75BfhrX@sby*nem?YHFE)ZHB@exd^^(I#jUL@sq%jwCv(`+5DT2rmfvRr?$ zB^WOB6~+v{CN?h`1=~MXlh~8W;6E-~5SP{=Q1~nlWqu=sTZ|=WX!8v6pSM1d|2z)L zPQHM!hpmFB!?`rUSDj8vzec=ULIm3it_c)y5wVEjXqQ1V`FOgF_zk|b-TZa|k-3#@ zyVqUbHn*oxm^Cw5_`qi&c`_siYGHqcKjYhlXU9AxB|9&X+2cSc)3jx z%so$$D|`FNy|1l;)spu~V95%>tu4oFT_+vI^)f9aw)7G?p{ht~rIrv?>Fa`9;|FXn zE>Wd@ij>@3bDI32vgF_vE!vQO72XOI;pWvb!oSx=VDjHG`pf17iR_Z8llg5=3a2k7 zwpQJAv-QL}npTH-smr0&UW&~)Qb}zRx8ULbEb%xsWzp|FS!LpF>a}|UDf$)2%0^Y= ziy!UiBXb_b2X-*afOY7TbC7-SwPj~-D`HAdH3peV(93yu$%sKP*723+Z?z`2l7$Ze_aq7qD-FE}OeelAQ!?=3%v-nS9#7yl+6qts11~v=0YVL}49)zCTf+wNAD^lJk$+E^b%F2njdWtFDVH&_yOgCT zYOz?CX~LI#t(nsibJp|ZJ$@>0#M8qgNP>?ZKGl{I=sm3>DGT#R`_A7&+GNTcl8*we zio{t<4cLf$WqM#ZkfhRExcS~_a){6AdpPXG6YT}qyj%gtr$kY!)*<+>el3VpE@wyW z>*gY530F1s~)D&OIhW22LvU>g5-Vv)eJ>0~Z~ocL=DzD3xKDTA08FoKEv zA)9@M2c8tZ9KO2a}23a(GUQLVC z%-P(}i|U$AX|ZvQ<1l1^|5lwj2E_fAqs^DMSi|4{%eoF>dzvvr7ZG;Ps{(}qYcYuD zQKfy^jj6>GS)jiYn=Sqqr+t~iY+k#uVyV;o-{)7Nr@jJHn#W^e*m*wRrol`<{lfA_ z8!Xx*%E;##EUxng>e*kzj>u-#+k1%i?5(9IZkyukA%CH`#wV20yM~fdZ%}TP7?%^L zL>`t}F^5$$?1aQ}hT5@AX2W+JN}10y7kcSY{fBr~=`}sHdy!!5-ciI~@(?=Rv1AoJ z7VO7b9rj+hmnoleV1<(gP-9~zdTOr5l7HIh@UsJj(xEIkD4uH7JcT9t*Wg19gCkL` zxbxOLc4pjSbT-$k_ffscB%(7|g-LA1Mn%yax&mMBlEa!_7tJrHnhmy|YbHjHid;1?g z8F)`4_ithA_- z+q0+;+=`;x%+SbhDwC_6!5WoVV#jwr7a_GCcuj9``>2vYm%VAqz;*X#vB-%N*tc(&P@?+-76`v$ zvSuU>YmWh|#oHjR;x{^NTFe3pjKTe}EE6l8#Wr6a!)~n2z*y5N>{Z&qUKwSfQ9v{H z@!rtrPCen;QA)VWZkBL+z(d@oR*3K3H=x0TrA*cI5L+i%$kb=l>`vAcXAu|iaP;>1 zg=a$i=j z$1__sF}8QC!1}NO(>5Sv1Gs)yJ*ICcVx7IE z#IU;=eKr^3#HlJQm{VlKk%j1J?nwW1{lvXRkvJ-0HoKU(2$G9m(oXw0FkPvJ85eZe z@!+{^bl+;0Tzd(#QpK3>3<+j?J_#$;?%)UIOPHntAb-+d_-(ZryDaI_ z*KdcoHr^MW8+=+gZiF>n@>opI?|KGjRc+Af{7bIpn#8m=`DAYW2MO*}!g1R7H=BKX z`5wyeb%N-RE@9A;CLD7#71e?!aPO=`;a*Q1=cu`!OSxde?fR#H$5!}4yUI@P#MnvP z;CQ}=s#OW(_;&0RZ%31L#t_sY4hw(2!sz66aCpH9wBMTve;sGBp)yTotg#d(-jjyZ z&CR4cM1c7xu94`;_Uz8_lem7=53)|;70KQ#!3{q81J4cna6#61w)XECf!W+EG$*SD zoaSu=mj*{X4jRmOaXJ?`qMNg-It{C~XW(mVzE3{Wp3OWyjOOosIm=Z|Bs{X5tI4|~ zi2arX+cJOR#ga}h^|Zf^fjZF`rLdl52kNrUvOVlf?`V3rd<57H*)Y-b ziCm7GA9uxZ3_H*^hm*PC!u35W;C572YG_HYygOe>*x-TB{JE)qLoPCkdzdq9W4#Y@B8wX|E_y| z&pqcnpU?YUK-U~HhH4=ia%Eo{{*G>eqMdJ`TIDSCWZl8`($VKy&#xX5^1w>G_47I*8+VBMR>t7;?htWP&Sl6EW}wZg=V16o zL#i@3f^JYALe}nTpnCvo=HWy9C4X)QALM;^j z>ITw{(iymg-v(DR$`@)ogO{BG_MEy7&#VJLYX38M`fCDMdSv0M85dbynHg;wa36m> zk>SzvKH;O>fpp&6$FQ^K2+pqb;tmJ@!(Ihj7^w7{d3tm3_Yg8)?Gk8wy$V}-0ECxB z@T+&nz}kuhRKZXmq6SyO8lyK5Rxt!N9G^@hjvc^VMU;A zg~}*Lg6E*=tUc3>)~*Sm6_q`((8C{6z6qVw<687jb1FTn=m0~*|3Rrw3V0WL(>;GF zExK`=#>~GA2~!v0-&6O*w}Kv*UU~P4DX;2B<5$rvH~J_X7@rB@dXjJDg`)H0l<5gaVLjEG2OZx$>E;#-c)i=3E}e#uxXBTU>W-5xjraJ|<|Mq` zvJh8i-bUT~LHu360^0sLfnRJrFn4+q)Bb%AJm=?8je_0u=5d5h{Yzl5_6Tn2dQHZr zKWAfw{8O^MiI6oLBGN6@0-F=*AfbGoW=>3`!}do)WWZM{PsDG{zoMiq`BslXr8>IO3z%TGEH4Hi~jfe0NPe``cQ`kPQ z2n}2lp>Nw7u#s6wTKg&P^v;JRhMh!f;u>n3^AK9%M?>lG4cryFK~m|GTuJqPs+LU>i6d z+6xjkYv}nG6Y$nwX}-*Q3bOT)tX(@ChD#2n7rqw3Swki|nQ@Oe?%IrhLiU!PTA9UO ze{YA}^lCUzsfkx-c0uE#`?$4;V}jXm@HlM>KimIY39u_GEqi|tYJ%ce_n`pMid&^H zudNfrKbL`qc@qTdUZm@_f%xvdUwUk>Bo;0{E4)TGR8kFr^=sCNPFZ{quWj~%&xOfu z`1g3nT!wF!lEgRB*8IaHE5Va^hR(dVo;J4pB?WI?g)Yn}H2hW|-s_pfmI*nA0WE*< z*UM;fs;bqxNFuy+Q1xy!+Y^A((;dYwewwg!nIsS`O)${Pxw4Szi6Z*{;gC2fK4Cxr z&7V7oQQ!Me`)D1#meeQG(oF~FxM#%wVG0cR<61g#+GJY)8LccRPZ9U^F8^+*dYF&yEi`nCxdcYQkXk!qIj12JTian0ebVj6O|NSz!NiSaQ1qE zH>6xdEh0n$zfB(QR}6-`yK@AdjlhA+pCX>TTZg?{ABwAnuESf9Nmg&{6+bKPhW7z+ zB9~Y-m{DrVuZ60K7G-TC$?n%dD_|NlOH~SoZz0>~_D}SzycZXnE@66o`|FRK2`(R>v@)%TKeNQ#JpD}zkiM(610}{qpvg`&6NPqg1n5+@pHX-?tux|o5CC#D- z%VcQXn-661Xuy zxaMm@Cmxd}M`zX3O4lKvr_zq$R_`!fGaQZ{%MdNqFD4)3Wnpf60c<^zgH9I?LqeyL zz^EC?k4(rAnS_31U!qR(;XzmN(J)Dn9c+p6(^{B*=*}T0FMAKqxoDQI zwwMb`Ka61;4bGFb&CSHxqn`XSdqFz6Em$6zFN_)DVf%_OSeG!A`qtelwHl{sbxrpo zK9*d<-?)vS@o7V-kE#n@ReKJ4UL40gegPPuX+qq$j$`YSdYDD$Lea+izlndy1~MT# z23l)0Xr}E`n0Km5bh%0B#g}Nn_1pR|Z$}S|*smctW+uU6$vc?r)DN$GX40C;YM|eu zhO<5^vV-}-F5eq;ZrtBwA^K>lrTm9d%@W zrvi1kmBkYGyb}i)-zC9sDSq&ti1I%IMb8$!#KG1^5UCXeQHRsvV_F!T=?y1uMJ3ke z|Hh!-s)x|g5KrHh>}M0mVX{U#hWvXY0Y=#Hp^AClqA zKTof7=gQ0%rA=<2Poc;+L4Otelbe7Fa9@Du9ydMmJq|O%j<=sKbXRM%0 z>Seh{_fQBaGod1d<1kt%_!`b&uusU`nzuRe&~1n4xr>MBn~#rRMxrlOKD`>A#q2`W z2|vMaJ*BTB##4u+XY~1^0y-4)sROx4_Xy09BI{*z(1lg}lk+{aQ?sDs{cUN`K`}iu z)rqc~ozBi_8&D8((uzh6p!~cF-s*qEOY!{z$A1@X)%gThABrLWhbP@BR|UzBC&5)= zy_S3%3>PaE`J5;7z$YzTa2sr*FPAN#TOPfqd!8+4)dGLiaoRtKaPj3Kf-m4kRR|P! z%F?S#y&&`4A6R9q2|ivqr54LV=%nq3z}d=#HkB2F)q#;zKs`};9S^!RHja)K*zQB6 zO2KmVT-stEMju8wiuHDn!vUW(_?G)srD;3PP^E@Uy7Rn;%>oh(a>+_IaZ>qof5138 zGvNZbWXz^ZEbkDh!YtaM9!>Q!LqXT`G7S6`ESlCn0otah(WwGo@c4hW{QT>cped`w z=f1S1hBtoU$~Ba_m=?2`QQKhEu`y5^vl~wNCgO9k3h3?{hz;#&rPFuRk`GIl(`%{& z>2%qfB#@cW;h(;XGrQI3w4E@XAsE(yFKjWy>fgtMU6@!Y_l5}h0Ig%pG z$M5Gw5sw!K=&@jJ>hdg`Uj66+57s8p|Jn-RYj7Re@-3bo`56ssdIRVM&1`|CC-@l; zpN8Kje#5~&BdT+I42|8got``&4UW!|RQ|;fxO^-DE-Cx~$%Z}w$Kpb#P8}lZp&GQc z@0O@jWe>HmaiV9Fb6Dl46Li&L4QRdSOT&fk;_GiF{Bw>jC}9)0x!*x8;|;WYu@YZj zm<^Y%et`qw@ib&rA-I=U!?nTp;n?bVeEw?+F;0JkI|V2DQYVf|Owp!=<7R`8=QOw^ zvsv_Q@B!#sQUifo>*3ohAhH`IVanWz)aAc?_VCp*e(m7_3^?r$`%F@)TTmpN{NhPd z``+@ILtSW0Xa!ZP?mf0C|&?VUn8^|c(x4EHDIP7V;Z z(S%+Jnnpjh%z{b#jH!m}LE?M$22^`E15c9z_d9=K(c5$;p`;Ji_a49!sWb52K!x7T z=!UOlS3#=dH9Z-o%)h)iL%*N>0wI?&=ySJdxX=FrOH-l8&fn!j{fE(M_d77#CSUmO zmgNs3)#$&2FX3-TI8XUzNb`^T3I0tDYW?|}z+ViZ)!K7mkGC=}oHw0XA9SU$UKMoW zn$xtproH!d}>R4@Ipfy`*gV;-h6Du$98#4a^7LQ_1`@lJQ3-wtba^P`il5#bPK(` zZy@c-9BcFSPXT?NPyn93tq@~5mF^h3kxJZkg}B#&uw3Fh>5Y`M>9mZeqkWFj#>7jg zpJ+s%Ntn{RkGgo8=5{>Omq1TfMbQCHi>ce>@pQ=Io&1SMJkASzfXxf`(wxZW^t;Ju z`o^*drZ+{2)~}GI{hv||c^yO7on zodrMAyU9l52&!SWfV#Q6)0r82N%qn|bj-&nddxnaoe}!&m9lYL$$5E}X zlHzeJs2z`@8J)PIb0jRu*-BldOR0HB8Ew65KsKEJK$>T|)0iMTwtr0(E?Ih$&$?z! zt@mkAIp1H zeg{4GM&tLeKpg*UE1d5uo``KPsUwel(97Qp~&x`By1RR6V841z|K*lXunCZxW`oJ3woN-ht|ib zN5pp88gdSvJH|oXs}@{owU_#_tE?pQTFE=1>o8pYD|mg$;yuswXzQtKJaJV%B=@X@ zG2@@puGb|fF((}gi|Xj%V=ribjus!=m_$RpGvVOUV!GL=7ZsA~_*nG`eD>%L=jYxGQY<@0y21|3*4lRrpxD5@1eZllP7Ktm*ZONez5D6%f+2%J^4KbzSmOArja`3y6~9|w zK>NalWW`Shuo3Q~X1?0a|HWLv$pasv%o)m8+-b*N;hcS`B?l9}s|)0WH>k5WO=NgT z8Rwk-gfrh>CspSJzp=sleg0_A7%cXmhu00F zQRRdzrZfiQw+>Gn=9faE#V63LWe9IvI*-S~1DtmwltbxyTx%JQ7bf+vxM_dzttn%t z48K`@ZhnHZyd}7oTp1%z{@_;9$R2LKjje)XM>0@}H~Nj>b7od!&ZMW9`l_124KdQ0 zuDI}24j*~diW`3r@n6uZeAHTNj5)wbkAKi9EKL7L1h@FWHw?Gj$U9RF`e zF)xm5Lg~mGIQQjKoY?ap&)RIrSD!kKO}LU*z(-s!?Zhw1%J3H-r146z90rCx!{CYT ze8q!$OxfB<2I=>4c2|X;Xgb9wha2IcK!K?oe;q$5W@3l+I}Cjwa7@}%aBONS2JLvt z%6j+Uy3hB7-#girMV`1L{)o8nw>tl3TY>hcSCbie1vq1t$!m=16zFB+^FTEJV zM_0ec@ljD?*;OM@;l&TUMi22D3hUXM%m#73FtdKJPM)uSR3e;v``G<>buL-1D=<=; z(SP$s@!M^sd~t9fUq1a7FIcx5gSVuiDB2EnuV!;^`53pT>XNr5>r|M3k)^Z3!L3jDys0DOLX1gkEK$M_Uoer}=? z9kSekWcwJyA<1Hp7-h;+<1e92bOB6s-;ABFF7U>gi@1l12{&4H7iwZQ@SRQ{aLwfw zoZj_>ZLRd+3XRUV@B+s>m;MTOU3p*=GKwybT#7#$3xTZ&B2z7&)9^z*r0>8Jh<(&X zG;*@I-Of?mZ{HVOU4MvOdHfvJ5{L5Z4(a^E>}Vb-g6LXSBh+GE<*@Q&l2e69T;8h5b=wuCFt=N{kjexx+iN=#x~HW_1i{)19*^erh6IK3H% z&+yL^FY#{l;HJYS(GkZa;OipBr$t4Acl!YdHXRHPu>j)DPT=Jux!l)wCSN$>2t zsadEwRs1VQ5AU9ex5~{yVaIvs7P3G^QTDw1kPjYxqYZQJw(!A&M)MIeJs4)0ipLtV zF>lKuW_x=Tv>A4wo>4XYTzr}ySP(>8O=akUihOv`uS%n)jlkcbZn(K*GTNQcrh`Z3 zqv6Rx{DQ9>_cu6$E~awW@GFp$wlQ$|lM|W0=o@BC9m~I{inwmb6Iiya#5!8I3mZSz zAFo{K1@-D64666#ekK}_QDVyr6k_?~8==tX)s5p0R)BSyAO5G^kN@7Tvj+iR^xNLv$=eW6ZfG~@MU~6ZoX23k`qjEUrdvb?J(w- z3o^O(vSduWEk=1CBiy<}36D+Kz+Gwv!s*$0csE!X4t~}LqoNJ?X|+5pzc3v|7dP`C zm8YoVpHRMSw+ncr$TIbrCE`VIrOG5mE#|h#o53bWp0`{+&9oNX0a5u?e3^a#+kR_< zg~T%c%h(gg`3q_o<34n-a^lLX&G4#a2e~u86+PmnayGvXlZ>4C)D<`I$T|_%5$@k+ z9m&A(LE)%fT!RG~W^|Uq1<>Di3pZ3W^1%zzd2jI^Ztttc#|fO$oH5^7#Z-c_!Z~zh z<6s{C#+OS}RPlj#bwEyi2!xLwz>A*@;{Jv+>6i*Np7UlCZeDnqJ1mdj#xpulQ9KmK zg-qr~_tS9HK6fxn?*hHub8+;hc5H1FICkMT*#6^*%%`}3J^0*)J5I0X*K6Xiy)H{k z*#MqF50q)9b4{0O>e5QQ;+xX z$r_)D!S0Yv=C~K29i@(d{I(Z_qUfW;RCh`OjwH$F0m&a z6C&!-T}}-Y{D*SKITqY!i7Su4E--p0dSZOQ3rP7hi#T4;=8=I}aP*Z8SJ-`&&z0BW z?h`lirPu%CvrhZsI^_o3^|}e?u9V<*iGou@nE%)|jAw@5mT(f$#*c^k`l)Z!%BVYsXWQ4x-vE3!-@ZC(biP99Fgi8!U_Az>gEu z>4zJ=>U9&B*WW?8o$H9}&rg`~&zj1+9T%lel7KbpjOax>;ctNlxmL*gx+opT zIsVac9{%c50{!tHvGCGMxGJB7mn7HHIQMUG|Hl>{c={;6cxy4z2xGX`+sICazvVCc zNASL?`y@F=#^%DRCV2CXpsVUs8gV3!+Z{Y3@CK8}^17A$jBbfoD(JANph1>zzGcld zhO3lmW)3Q=8Yyr^R$B90GvxWcg0WPhs+v7`S3rArAK-dQGWNtBo4wSnIRmv{-^z~-MW(%IUXApc9vO_x?X0aU!A(%TXkxxFWPdmG|QbX&rc&qL< zp7*F>TBUnwKTBXw=rWjVGLM?-3k?20ruf|aJ}V1cLz)H?USUOH#+*g4<7kF(2j;@F zPDNwsum&6vF_C8mMX)bRj=`$ai@9<8B|cI;f?rc9g&*G(;OzUeOyz+(K5=NLYt|HF zsrFo~?S24vLW@A`avLhTyun+g2&V0K;Sm$QL#KK=)ts@IlE;_0%oD}35XCgUe#WS> z$dxsGnu@KkS0{Ys8fgLkGsar=w+(;x=n)Ctau%@In+*4JsW#}gr~}%3qrlko0hNpC=lWiPccs_`jh!h5I~MRyfqneE zMWArTbp^dCg?vWch_WJwYH^94B)zQinS|LG@m+a-bY)!^xjoxMRDNhFuswsQ%W5h7 zcrt_Z$I9{$IeU80J)h%+W&Dq<8+CnqNH|+h;Crj%;ewXnE_0sE2jBR}nk2T+*zsvF zSbHtCY^|eX=q?&z<_*EmMuW`9Y`BVoC%IxV_fi`O2j)ogmzSEbU+FUV=o|4((Yx`R znzU%>!eOw!LAY1#l;KZ4>QnoODk`00LhG{3sDZMW$aKVsM_4sl&o7nWn?@?~m5(bh z*>XCf+YhqEEDR-`%=v~P&Rp7jCsnpMLn0ncN3Fa?eE-{hfHgP7$Ft0Fzmt%&DYq}} zUmi}Iyu45{QxR=n&gV9Rt)Z``Rp^EvMx8@W=zr`O-?(Eb-#yBZKNlRWn#0t&@g`|% z=X47Thc>ZO?uS`QMk2o&-AJ59j_2i3cD%Ks8(sF^#G!kiz-mn>(CYzG;;#5)d;~wt6Op{I-;zxDQHxuu?RJ0)-98TYBOrNN4DVR%#@8%+At=q_aYs%sk+YDp85Lp6Yby%T#+;G?_tHrDmvB0Gdk&3} z%c4?dzEsnqo402CK>xy(d}-Sjz71Y;4TXDrZ&e!PLlN|Da-;gHrR1*ZGU0pgxzI_k zp%Kbr>f~6AeJzSMCsWdB&3gr!ThdGSo|pyKr)(qJDsA`NM85KvM(PdQ>9Z%He$6;pfxob?zzT7czr)4JQ+(5>~sREktr74vm|;~ zZg2_d*TCk5(kz{YH2U3kC}%cw?2C0Yq;WQxYDP8iYfYmIGq2;bW;m@N>koWdD9b3H<SUPdJeW=| zn?!5<25@#<4)rHKrFNb#>D*Xpn^cM45VJ_0u5IZO*RI$}_DMb@9hK{;=l7>{r?m{c z_VA%$!B54ZiyA=sd>q?!vYpzjETg%O*I-$f3H1ugBh`=Eus&iOwQ)YjhFI>SHiL@j zYqMbZ5>$g z>b~qI7!AmUjeP+qx2zosoTRBi(rQWt&)uOZqv)ZOhZvwflh&dNbv_x5qlO32ray;i z@mgW;diyzq4!%gs4Nb7~)^%Efzo9>8BYFB?(=;MKf#QTXl+He0sbdKC79Scv<>%#xmtwLcQWDrI-hcBUB>m#Vs?p*Qv z;9sB|ZidzkQQV>?3a`5jq#D-~c-wm|I;q*1esfo(#2hh@1yL!Dv!Iqah?g(vhJ^0Jj+z{B47eb#r)UX-0 zTNdxn=o7WQz6fe5LG<6r6QaFFfXd&Mste<2@yIB4avgwmoi@$u zRHdt0V<0arj<)x|C!wm#m487U8jr)n?nrE?lcuv5BtV;eJ{{}3h<5Gz2=ZYY z(dVQjP3#?y&KJCC!v#+&svAztYu>*b+u*g$?U0>ut~cKES-Hny$W0c!^*f#0yP z#J)L(5G!l;!b-W!?pda2zQJNRc;QKD#jZ%!aCIFCnpuKIuOBcqvzzSH#1Um-?&)Nk z>~@&1IEZ`}W{i5(y09!Yl?04=$5i%Mkkl;eGPQPBvh`UgS(lOyx4g6AqT+fM%Y~Vq9H?Epkj;%hI3<>)}a6f>t;jw zGbE(oF#2w53p!^k=7t`}@!5gV)`e0T?7O=% z?P$>{0~zD&?{UwK#}r62Z0rMRc_h z_~eD#AU7tDOc3V(zMsE|bOm2vtzruE|8o>-oD9l(r^xch^%9V?N3G0hztGcsuFhh1 zYn1UCf0XVOdMJ{y;;Q}8(EN58#Z!gH}ZwZ zTCGXoOgFORubOaW!xs{i<;S)MY#(_reV1H^9cO41ylO@5BN#h#=r#|;Z|NRI6& zm|bXAcGsv_WPE6~_;$lqh@1X~=tdDTQD=epQ(_GJGxwNiS;G-bS=~#-A!P_E zHr8o13{<}gjEXdXx9QFz9Yd3{GYM^?6%({zl0!1C|6RvM->NJ5J-2}S`PY#7>$-@V z>`kr~zZcY%H^SqPRPOL=6RDN^D7yTs6bGuFhPj_^KygM8XudL}b_;Z=R$VojsM{cp zUUjxK-gFooU%V5JEzyEKee21~Ppcs(G>m7)S+wUd7I}0J zFWj&+JIja`poMp()FTfD4)R)Q8mGSHMR$ zvrDhMujTnGMuJMVJd~UtU1r|R;X|U}_t@9PN?qH?G#U(+E7f6==Y9Bi#Rq>HPTC4xIM*tLVS3uTM<23yG#+$ladUby{pmckQi zfzavNAWr&T%KRTlmTj9hf@Q=WW00j&wkGQlY|?rUs|6M}TQ~sz)ZQeHKjUHVX*IFq z%cms1@S=!jWs2`gH;XLNck$CR!bB!d)gblVUOsGeHd&l14@KD%%Pu=VAWw^RXsZ7P za?Ifkdm!zOs+EWE!Oa?G+A@}W%Jl$6KQ)+TtPcgtj==-@XK+=^jx4HEH{^PCij!3rkbC77;+;Dcv2^P} zCLi)yl=JR6b21;!%=6{omBTPppO8xqyHA5_%@;)x0|_}Vy@iGJDWUdqWioDj6RH2J z1LNc6@IRj%*tk0s27c%xRjv1k#zAReT{TDEH4V-h$K!2MjK}9E6M64!QT|jta>YT8 z;Eh!A3e{XPX~+QhQILen32|T(eb(Z#$}w^{F9*lm_7WvUFGoc$Z5}cFy7l3c8&UE< z3HG_|0ef39m(ln?#Jk~>xUOP4vv@g*spuw&6w=`O1Me%dKE``ssc zxcZvt&j1N(@oPDGw*61vq04MI~De~onT9K z@>z+?ccwl*M-<$63i+9dET#H18J}?*cic8YyHEXAi{TXcWe~;24!9=%^Jf)E&wI#j z+!=>A9uq!${SM5j^%ePVUWc=W?uL6o_sOy2da&h84h}eRp7AFMuyOqrtg4U@KO7f9 zK6*Z86Azy!`ZcfFO4A9%X6riQU0K6^CtQW4!2zP-?`C1@q%M(dXBA8EJSqO`Z-yX#W=XJln1)(;^Tf1(=-xQ`qkpI zVX;_akSzY%V}*@|I#?NSo)j64z)+?@pUtbr7cttr*U=9*rHv&%p1u&-8cE7`-ym_* zE;09mThXE-h~2Cli2vj!!=ScOqB~X_R^F6i{;pk+s;ABy23%)N2a8GP1bdA7n1vE& z64?1gMv%9~4@>>(S;cGu+oFS5Y={hCIgah!B?H~r{$x+!bCxWn z1rv9i5W58(gr<}m;Q0GL5Csh9^6w&1Cnj8cHg7tRb+^T0jZx4NCk11FUWY|$GfC{C zsjOqkZPC+HxnR&R4bR5qvlAP;tcNbCCx)^*B>!X#8aMmnwLEvMF0f=dk_#bd$a;aX zejQA#NistmBs9CMhPn4H~`0Y{(+3E#_+IExG%nui&s~^fwiKI^!=X~khA$Uo0)f- z<|T)cpV@oC*l-`z*l384kI4|3hc4nWrH^3okJGT~4v^|eQZ&KT6YI7BCPw&+FZgtl z@+o29E;SEAZws!boGjE}vMAZn2G$N^tw+3i$mVK=(QTy*Xx^?``a5b9PK#ZE$p`z$ z^FCMDKVv1vC3cGszv%*uB7E6ez>-b8@Lu2;swT_>?}u!`kDmwX*)@Rjuw^q z^&E?`^~lAYF)-HT04%(68Si>|Vt!f;(ec|3?_%S@cELXIpSXtX6M4guJ|=dqNx`CJ zI{ZmeB{`!gi}O~_BcXe2MULwX_*&CLtf{RK;`$e{=~E6tWba4Fw?9XRw_TuFy5nt9 zyh7pixbMXFbQdh#_!$<@JpfrAwQ%Q(Gl&XnU|4nz=Gywu7Xgdus=kZR(7p%<3LNG7 zt%=y~KL}Fd1ipI37-GC!k#}ac@O{^s`PtDg$d~4EqN1WqFw!5z3V1}ROnNk4k!S-K zJ1zQU(nOLmRgn&#ei2+3r9#JuLec6+i8$XX0*f2Xq4#GZ+!Zp(UlNp{>GMFElV(ca zWGfPrBtLxl>jriNpCg9?pTlry1+n^Iy7X*07-r3eb*8;UrdM!c6>s2ax>2;Y*q06u z*+h+U8vwGL*w>0A3{y#DbKeMTNuPGOd@`FzMSl@*=#ixR=D&m~zwgsid8gR}r<>5L zvjLiS+tZ8B!n%xHa4lD&3GqdsJ98QJXKk?A^axDqU&GY% z2hovBW?}bWq2GD>Gg*KVFuPR+G!tV)M>CD_`-@W8o2HIhNiRXf(#h-)3DFD740y;w z;K}9;u-$JCKdiK2joB5LG2H~!rltcwQAKX-lSTV-7oxCGMf}V!18O^_z_CsKl!aBm zgBNP_c*i8Xo;#V1+4uo=(Lp#P;IFlJc{fbRt%kP2iu~2=QnIxy8^@~EFybu93nr@4 z>lzMhZkZI-Ei8w1c!uooD2BM<=f$n_%t2{qDqbl&1gfiA!MU~%LKetE#NBcHC%$D- zzjf&drVL99Rp{ae&e&%t&D%2$z>-rr5L2QGwZRcEbz>wreAx>Zg!!q-f?fEg+JqR6 zGQw4(7ZS%WCfr*0H(uO-R`5%zDsBVj@?AxZ505 zWpCib3=bkH`AU3ahYH(1y^Yzg98B~^3H#|O4{@h@GHfdx&L8clMB%)I7w?Ki@01>~ z@mjucaqw>vvnmBOcW-Bl6KX{-y$n(H!xXr6SzpLod9guqO)O`#7s|f0!X37|vC^f2 zq|G%%t-w#Lx!s(-J7b7-1@l?sgi3bM@i|NM&1LH5OzhG z6@LAWIT>19E`L78DDA;7*8idH@ zl3^aZ=9Pm(vZ}>Lf1SVq>xQG|j>TO6RxRGKc18b9Q`w|}Sy+;An6=G#Odc0Zz%<)A z804@L&$u69i#~Tl_RDtIxAU~vO7p2W^KmV}Gv*6L^#rH?lF@S$_|quA5W!_v2#9)|NXNgn=2y7yRt}}>QDloQs%M7kt+C9sg&pm%$fQ6LXS1% zBWZHC!t4}B_RL@zJ84^p4Z%W||A-=bRY=3Ty&f3f-b_BF8X}b5Vz1vcv9k++1-?ne z8UDYRrmT%9@A-A{F3To%Y|2%!wBZq0m6S`gc)IcgK&15#a2tG zi2HFrK3ONi*k<7j;}`)K^`fx$!v<`9UTpnqgeU4fbzm+Vvf04LI;`Evh-xhSgfZ=( zL;(rM(7O2nxlmFqYEe!_(>yVz$r$3i`&zvAjtX6UAV<9Ar7X<e{pW6A>X+bJzhb}+_sf?sZcc0TKv^^Ga@ z^@=xb_5qiF*YVK8(b!h%3*Q>7L3>`Nz&j3Ozjn`JH-dY}`ffG2He?A4Na$s~|E@um z!W1&+(qN!!qsY|J17T7TK!-4wycE7dO_rD3E&O@=H@9&*p^P?~vFO|$h zk}(jO`GU;YddB+whVii4XCb=2EyBu_8>OSm^F<%`ue%aVf>Pk>_B=tn;(32EAX5It6&KY!#&m-bvqznu5#$b}y zQ0#k}3r3y-leg6oXGSN3?1dnZ5S-=HcQ1j@vpeEL`Xw9fdW}@OTS9}oCmPM&2fZ%Ja6ig~f!_!mp_xwt zk2pZ>$(y*ZqL|oT{7d{MJYtpspTQ~G3O!AyK-jTl*eCcwx~A!poB9*!(lR;v)$e8L z`U&nZqfP}~&yHa!?-H3!t|jv<$|mdAeP1MQ>}}@V;w}2p zD4W|a>f2hw$}6InUcD}KtGs1BHh;uvd%8#=c~ts1=oZOT&t^{^6oAT(1ES}FF(Ti` zvaqpZD|u@>j()76VEjKtXBtS=*TrFE$W)P;kPxMjxMyvO6s1Wi(Hs>SoBmNzh76g? zP==H!N}?pWdutBmF$~{d6&F~ax8A9~!ILH#RB`l7X)t2uX)`!7FPpb~QdM(ORM zUEd~BwR>?wx1L&#aJRWSCc$&wv$(pz<(RrnRj{>vw{T7I2z+?mhu%{fg+UGGu;B7t zDE7Z-p<&JYdE)t*LeVM+Ta_;8F8@VszP+KQqu=7shbzmD+F9W=mjtSG;3_S?l|`-2 z27twUHE3CE2`B6_@a+qCc#;^($yR>gUKRJypu&CmW1V2Zwv*K7?s1SSn*r(TltKB%2t334d3*{g1t0$IqWgBt z04I@8<~vsCg zO2Eoq3RrGQA2)98Sw2gVz#21j%FiXpme<>bvZNp%cGhGFTZ$D}&x~5G@vLF-FPiLn{0d7(X0n45Bf!wglsWBw&4MMuS%Je#9KAG>%`FMw?tXv6oV|FK zpu$R45h+D%?}mVSaI^4NaR_D}H%5gNPcTaN!{p(U?8~}8nETpC=oOKM|58@K@J}Q5 zJ?|FNuYQYRR(sjNmWj;r#%FAJGsNKRCuVs(fqC$3TUD1LTy(bxgD$M(;uk(bZIz$g zp?qNOrSF-tu$~p2i>A&lmgOok@+^FND^7U$np)nQgm}^B9k#W-9*i z*mSjnEgY+fZ^kU)_Bxlb<&JCM+hIeYUqKRz$>rflDOu_kzLiYN<0A^#}ixFbWS+08~uFrUcr49qI-q4_Y?tD9@cXB6- z@3dmKRSDi%a}hl*&7%#X+nAFu3V-)DV}I{UcK!4VSU-9QrrqS-y2C#O6ZG3?nYK35 zkp9cEzxS{o>cK2I#0+*{=Ge}rM7HnVI`*Gz39DKZ07px7*f?*0obtzy9a|d1j4HiY zr;}KD`He8pfLzvMv!40Cabr@voAJK<91`cE%VIX}XXh5|Wtl;9Sn9tn&OKh6Svske z>xm}dj9_iDZhkZ^kM?0x|GZ*r#~xu;e14|z(=66H+o(L}O+Ie(S_h55nDgZ-D{dz2_n%cv zU-cO>n=$Ol{2wTI()Y#YBY!>))3o(v+!5;0K#tsT+BZ~A2hWye6 zo8DGn+qI=k=EMT(tf*Kn^Vgf@I;t{9pDd>GPLqb--3UY8O%U>te|MW|pzqJha_i<9 zWFUAFlT$5acV9&@E!!{bPJTKYpOe9MWf`!ptS%hqrbV_yyP>%aMN`*mwyLKe_x-!i z(lu-unG(gy>VLCS+w+;?Y_sy8S8UkiN7LCl#|kv@8fKlpMzT4nhuE^^N3g(4jvACX zF`X+R;C(oko!OE~rb@lmjuP(QIZ+xgb7l7M}3q znf2!1csHFs=~+|AGhq86>(L`H=uM{wz>wDO-;c{SrJ>@!oiJP-hZ-MZp@N7BQgWn7 z+cE{l{%l6Q$^G1qd8g<-&WT0^T;sG4Dq#6QE2KAD2+9{0!QM(mn7H$vg+)RK9la_G z#=qMG-qalGltrO$_Zj#r7R5ymmvZX*B4mQ9C!CsdfivCZ3U|Eqp}p-Aoy_BS8dj!2 z_LGrV@#8$l;`VV(PEGLkrVMR5B`^FYFGZry?Bd3K3WFXY@t?B`D1q1x_zQ%Q|x+aE);ao>@ARc@2os7&a3cCktb-4nF>#>N5H`- zHGHkD3tv^kp?q5!jrHPPDq-%NUEJ(4ACDI}Gkh3rj_!ksi~KD5?eCPwwbPp3Jg8r1 z#651Z=FDT?!Fk6Mc*?+tl6C6XWoH5r+Is~dm6qhhoO=4)x^h5}q$$KB=U%>*!o7B4Al5?FTf|HA4Xmhy; zMC^Mkuy-hcx%L_mnj(ZR>v!Oz(;v7G{~2<_Zxisxk2ad)uocD?@LtPbv2eGxfD>l? zpxe`>vAX;^zBc0Rc%jGON0JLw`*90YbH;O~uDgV%jm05DITK>U_dp=;&e07z3gKnn zxyF?ZaIMQ7j&J_NHOj}KkyI}IdNl}Yl}&K#qG9?i>lwz@#^Pr~p04VN$f>F;SkAK1YtB{hp4rR=hKzx(jUf=RJ_@H#t>h*>nQYOS?h5nD`}n%L0!dwD z0tpw7f|MH1^1PlwAKJ>3lDoi?&h#f6)IJ8ZJ zEpv;5%L}xHtIh6lJBFNL*{1-W&#*+$wA=+vcCMf!@5~2@)nOo2AItrYj)QL>M40O5 zU3kA$KypKpaBIbVxEFR9eH*9ahn-U(belLiH^l??PAJ5hd95&_a}!fKvYI3s=5XcV zYvH6apNZIVjIYi5lZ;36*}m?>`BeYv$($|o0R@4pY>%jLW|`_FjRJ86Z0pqyRe0Zi$syW-0`ewbtaKG9s#X7e>e-9ftzgCSl~IwnY-kDYS=$CNy|G|i=S}SW4RGJ%dz`Oz7bW(p*R~eyw2lCB3mE(Fn#LoKB2OTtNO$9F4Zh zhGzxJ?E2(r>OT2A%=;ionACnKJNgz~7pgG3pHFcgdj$tp{6pC)Y1liu5X-EZ(T;6} z5!o(ykKRU^1#KvD(nq-Sj5|)59!hoYo1@;?ayr;q3Vqktkf~F2AotpSu#E^2uI&5) zOO2Drm?gW(@}){(T)%=WSKI^U;aA~hN)U1Rssj2)W67!6m)S^DUxD!wC1#gq!#%9u zid)=&;kf`C!c`x~dmCcdqHpo=_x3Dex3v)G>KB66sRH6}GnK3gIzyty=YwbHTnOx! z!IkBia7yqGtcqetx5ip%I^|Aw>&oFn!7`Y_^UuDLIt(yd&)Pz2Fhj?X?D#So^qubj zk)KB1Z+nN+=lS5X5>M9R=mKtqT6Ep6P-3@P9o_UIQTI_MJ8~}z{KAf~h#PX`(Bo?m zn>>!p`)fixEtWxiPb{guNLc9VERMV%i&Sk5UH$StEXn>2_WFsE!oGNNNpvyUxnw==J?u2L86L zBc10L!HakSJ`yb>zvgX(qk0EOjo~Qx@0v5$wl|W@TW?9Ob(oNovHOYa{p}=@XBZ7? zq!OLgUtxYbAxNA_%#X9=P3I9(xhbF6eJUd67A3H#)R_qU9zfB%am49)F86ly4X~XN zMH+vLkd2O;i9%*LX}>*>yxDjRLh446C7YhZzVklNJIR`ej&1}|r4YVfmrlH1g+d{> z8gz5*$d=^GoTrTmWCa+Jhcg|C(NG2Cth@xXW97*@xJG_Ha|DHz+N46pj70Z2k-WCi zV0Ezp0!?H{muNDq=V$Mo)C@qc&4?`BB~79gs$gx{IHIff4hn-5iD$-MGXGxz(akA@ zSi`U6Y?C<2&^$t>CU}q)pQpk0L1_qEd;}&mdXk$41tjl9DCrqFi-fjVk-;~q)cx~b zvSGl4*mfpCjl@WDIwq6b_aY7ctuq4!>kr)Kv!7v|wi{RQ&4~Qj^@#(+4OsDouW_y# z3GCb#p=+QfX)seKu9@$I4U3;ZML+`Vzi^JI-Th0%z5sdDCku89^GKu6mE71^N?jHv zanmF(maQtf3#waopmOUB;=Ck^d~tg~TK2Cck}H>x)gC@X)}>!ywetfMt#&2rZhe4EvmtVTFn}X?p(`#9Q*9`KAb*d0jWl+F6fI?zhB}=666Bws5s7KOj&;lPFY2 zl2tEk;l{kMBK+V3W{diDNG$4GGAVt9j&~(2QtXco2uhXSvAzF_1M&l)M{k zfmF$0X!>;+wjEwT_NF#-qdYc|5?58rUs|G;U*DZ1on@QhZ(tiIZlWa!`l3j*y(~z* zmLECr+?B}ody)amY;tAK1k01}YKXT_Eb$lB;(016+=G+_WOm>mm|{PX^qx3G?xZaw z_L_4@cZv&nvwjI#6uy_-SK2|mj!O~!eMiWJN1ep){y3ukIMia>RVk9Wbdcv5Y$b{( zXOl#pof>F4k>p+uAZvebh1gz8xH@qxw7-1-D$DuI=P^;zr*{fRZp#D9`cw3Mhbd{V zJU|o^&Ty0T9|`(j`4N@4r9^6vFKp{dL#b7+r0DxO*z{m8dASOR3ZJ!A5IadWbk>7o z-+v^YpQCx)=1uBu>XKXWZ(w4s2w}P_aLEQSa#J8nZt0418^)%PfjR?{S`r2)b8GQs zhYrhB6(!<_Wk}!oDMV@OO$eS|42KMi*!`_zNor~=8J<%@(#5LbM(J9*!{jDZO^JdN zu^Y(Mpl&#SWEjGqsF6F*?1-h4FS&oL6XtDlBw7nLla38Dh^)?PGNVeL$PYP?(gWSl z(H%xYel8$2DJx0s(@dTLCq_Cuw?h7CYq;q`2%j7w?>ugD+w8Q7$a3CS__~pN8C682 zdp^L-lrj+QFocfuXnuj2PXfkRP;;F~@?Er_>#R_Pk3K$76*r1#$(xW7m8Xb#Q$0cT zOGLnPT^(;maK>xpaZYI>=A|nUgP|lcRS*gh=hj+6Z4}YDy@r%toeKAkM!*i$I*Z)# zJJ{FLBgm&KAGoDSFX4rvGo(6=hWx@Xh?u5LUimG9avcXUs$7$t?%Ypqh@1m8%~OBQ?Y<;1qZ5YJzwaAYba^zgHC7!vy5Y&1?(6Z5p z+;&taViW2G!{baWv6?Nk!xTuA!J0GTJF#ukOU zvF835HfrTk_^t8?W{%Z_t8E`)gw<*I_n#^0*?n2CNctEoeXd3(yL}-#_hl`~&>6DG zWg6Kf7D^n>8<5FMG+=nV9#N@C1p7c~9MLwK7%H5`O=bmfyT1TQ;7!=DVLnlGn@6Ul z+~L%^bu5q1Ft^O993XBx)A4t>2;8W?haR3K#Bikvj9uMA(uesqPU|3<`Jsi_{anfY z_#R9o<~}0PD?-Wq$txhlc`ND5&?bA%Diix82lB&q4cT@tlDzKV_W)nD$+y81WE}dE zo|byL^YLnSalRsop3j6?|5cEf)fH zvUTH4*y@1Mh|h!Q&)`adX;%=a4TX@qE$Of)V=V+mhSLT= zh_l6K^61|`)R{hweQnnwFhzott}!KBA1hfJC@7N8+Pv?5HqSbX9Yu01B}kgHBI%lO z2gm&0$8zdxa6#y0$lU8J=>2L!)OV#pOYL^DJhuYc(|pO#9BWu=a2x!eejxw4+sMV; zBQ5oI*^yHJ0HRZR4{VpMfc;Y?$+foSVBmWKZb#_j-sS&+q5TXt`y$VKkiSeD&F_JT zuO(S%y&Y7Xbx=(!731G7WF;9z@bgK_j%njT~ z$Gfm}S22+uXolm5d!c^R8q(5N3s(nQPaqOPDSovgw4eajNuT1E>9&}9t(B@jRh0PKh zmi^8Gjl2J%rCJCx8ZSxmm(2x``MLOov2tK!n9?! zvw!M0FvV{S3ovd$tN1ubem(2@uyr=otlY+uYHniPPdU`v9gM-NqS;IKl&$N2$Xv#s zCIa<0q+ZFJtrSL+-Z_J~OPPY!$cr>uc^8Y>r_FS`d4I^kX)JJM0k%%z{W-e)jPSe| z>Yp`(Mc)XH0!Hj`gH2c{woh5gTU}5ctap0B* zQk`NXopbR@s2-EboXo~)oPdk(-Pq8mO(aCyigo`ON%Xf)W{x+Ovlp}Xut##+nVL0W zp5H~;fP*u7rEF#;>nV%SD`hHEE4Z~`R?Po^C>zx|g^e5d%VcLqk?Li)%gi?1#qw+^ zp3^>`sThLLH24Tk+wYANw%$aYI#+U3Lxx?-Im{~Lis)CD!}Aw$9r{1%?A^5(Us{KDmmfeI(>f9aEV$bCC-@5j5Owwlf(||ZDPU%S8n%^DqHV-j1;+NV}x@X zem$SWHtb!D8p+2=&DG~jCVd>6enOeq+{|L(Vl^yr&UDxy>dum?4s*NQhQQdb9Utcm z;!T-q41dy$CrAGUxpCPn+pZKB{~ATJEeJdR#(;GPbaO(zW;Sm247OoV2W|cSBQ}L^ zX}QTWDziG9%jlcHP7KZ@#-A7#FN+1y5k*{7v%T6~TSCxT@KO4|^(12NQ z^2a?V$Ac!%At)2ZvO~_Qtl{D$vO(%CEIFx0;trZH0|j-<4Z$(6ZVCVVRT45k&lrAH zcH`%o6L|E{0DU076dskO@r+a}Wg&zMWs zvo68FtWIwEj0*aJcQ1LpcvM!xhx^;M#t1C9mH6{q9d!Rw!UoNSOw#BB<}7T*dVLwL z{g?q-dTr$v@VOK1H}^5`)DZYi4q&eoM&aCz2^@W#4LRLR_4&1W#tN#*Ac1L%X~eQc-eF${Y4 zklpLLh4SZDVWr1G)F@G4JM6XD-OVdlLe4ni>>J4LKE6yFI+W?GH%3&TGbFs4R3#h$ zakkx8i#5PZmOn*;l{%cqA3;{Uo1lya1#M=39$n#ve{&?bUIVGm4?J|^y5Qih<>33v zfTgYp7q-Sm0yicGO|R*Z3gt&Ic=8b4eD586-f@+?qI#Ql_$@}Ov*Cg#*G3Xgr3LtR zMFF%-n@rBciIJg)-}w{66i({0H+-;~#nOYWL$_u;fD7dA+W8u+_= zCyC+DVQt4ZlNvV*vLn?ID!wXnK8sDMczhNX7wv-w_ZOms^E>qKk0#|)4J?#=Hj=yz z4!FbeE_gKhK^_Yra?aT>_LCuNuxPe;AQs078~3n%QX?4pyr%LS1B8a=#_XxBBzxZ7 z3R9nb!_GaX+(8L%dgy%^mv#0Q7zgga-i`opE9d{K_!_FIxDp>lbP&vhCQFSV{wVN2qMAQRfP0wy~b0khPdpE$?#-LE>=|E z0MB0yP`O^0ly@`uwtgS(YF&*Z&&xqlj5um&9^>~M`b>7yQ5xVl0c*zSf?ARJ+{9OWDvuf$t@hkAwgR#`g*o)qYjHP?458|K9@l4}q9UWbg#`cGlGft$0 zdNr>QPB${Zx5X3DpJyjbx9S$`-u;cPyrU;PdwBxB;F-e*Jw;Rg=Wfs@D%Gfgjn)%| zEs{;#Ufb!M#_utxB6XS8$SM$SM|0DfTW9yFNLJ@~!Y>`P7_4q5vOmr9H;YH&FI}_$& zv%VH81X|OFHo0^~M=chGw+oM!=fb_JSzOYnAPk7O1cpCNF>tRW%_+|m2-RiTI+uTR z;-DALlQG4*7CrP`_eS_#^$}hUixb-Q&c?It1JvejI-S3t_q8vLqFK^fsGgrD@G!AM zXX9y@C3=%e92kM!g*o)~#RmFI$qd(e$l&gQW_sKIitvq-B2(G#hu;p46SO8>qtC{q zaHFTDQ{~?txMA})>iK38`yBKM*=AL;d}jq-56H%gcley|?*4H{l-HlFHyyq$@s1E z7QNVXi_hs3+<8A;P^V|h_bRS(FT$0CF=rT8;&+!ekKB#o26AX!{2#h@O~IygF`A!N zN6oCG+mj)oh54oeKAGoem4eH9U~!R<1NbU%<;V-S7s z(sHqD^q7{qAZXJ+p~FWPdwTF3t6nQb_R1AewSoY}Y~lD9lB^U3RF^um=FbgkJ@ z?wHOkl=@MObId~N?k@tOaKIAG9OjaWafwW>Gmg3+*@ExJNYkBh6ESUiHZIeYCvKI> zv?UIR^!&<8EP<56}C2r%tv0jK`XFZmi2nuI%&I zBjluVG_(em(nEIh*>#y^Y*n5m+vcB7*0m{-ochZk$7#U5$}ngMmE->048?E50=EBJ z8MiclG0ZvX4?dFS*iruzcUly(T_Zljbs|9&?*yT=nFo5!-wVm>L+FPk^9AJnBJ}Iq zjq)#r^~zh8Ia=JD5|?88K|Ln?;7ofOVGjE@QYl4Zys zf1Mt)#LJyO#<84|Gi=4}B<|&KCDjSq%d}RXAT6g0SR9w$rHc95)K7vVi&7c37O!PW=*V7Ox!TDVA{?Py1;HUB7Rh>Zogp;06>=7CV! zQxm!T6|lYP0+_XG@T^oPL7#*h-6mbcRoKc9$4%7u0#+~VmxGEs!!TwUpF@6m113JUg=NBptZ@^+FDLrs$C;}zfAT64xN!lBgx|!Ut2V;} zw}WJ22+wEwPlH^y-iXZs*J;U2A10}x40#>TxY+U|oL}y37<4G&>!|(sao$t-r~d^5 z3YAz!Zza^2M#00*Tx?r(k%T>c36}i-OA4<+w0sQP^Lrv2jxFHA`Dgm`(*m@t=Y7YO z>O$;@gad_AxOAHywI6#68?z0mv5_wuYbVA)X$y3?H-c(z1y26dgc1TIY(Pi!0hf|T)e zqJKk`O_*{WH?FG0G&6rRiC>H_`I&}tc^P~lf1b9Cl4Oy80F~AmK|p36evJ&nBi8Hj zde3vLF3iN2r)^lm%tCZ3a>Il_qu51zcL;lyLH6|TAPt4CILhr9SvUD93Eq)F??`_@ zv8z_3sKEqu?W(AMkt)e?EoX)XS>_d?Muu zN%nv*s_JB&XC`UmnSIlC-{KCLcSFn`un@iNiK9*D65DUCM7l9g*q=3#NZejcPEI@m zf43yS(|w;{fnG1b)EgGH5j<lM-qnE=-?k#Re#zd87PAK16A-Y^*MRpr(sz&ua@Y9YeD~4C6fC}hK$(#2?B(#xd2f; z5|yn@{@Iqnyn;J$B>ox_aZ$^id+Lb--{(2;(v`fb{!VnO=8~@Lct|NIh9?dCXryKz z#O1Cf|2??R_i4_O=H+$7BF~BZ`W{F&T5llAZ#9WV+6?mWtul%A6C=90BglBuX+-g= z1KI5`j+RImfONoAI4EgBx$-2^^!Xvo_;3(cdHzQZ%#|jaA{xQRb2(WvrV}3ITfvtw z1#;%rQX(BJO;*OKl19yaP@gU#lfT;AWQaM!s2SB+L6)3tQT)7uF=gT4TA^aY%s#y*%UZNYu@&?64g{2DTU zJ2~~=QDXCb3^~5xC+&G*56-qNd`={aT;^+>tBt>bNRt71(U=F95@X=X6+JlF+J|#u zBH^xSGAs_s>3=cW{?m}&`EPEbUk1q9yKkpGqe znT73T({E%z&CNx`<5>VXqs;$q(89fyjUzAh{OCSwFLrOyGN#NXux6QFjCMYNi`Bi* z=}Z=luS%mwPoS{CeOPesNg7wuuLpls*0VX;LHKFpKNQU}!X~{!;WF*1I55`~y%H}o zAJ>iS?u65r@_ZSVpD0G{N7i`ppagyH8X_z$%tybi*EqZ0qj>kuCz`6L1<`7bLbZhp z*xREo@bsIdY|}n(_I%KUg_t#Bcg{XcHaLqr6SvT#e%kEI^_|S_KHpCi#3Ebs9D59R z;379&K9BCnI=|#H**Qz7+4m7_nqxGM-0qCO;$(1V`B9qot_62YO2RIUA#8a4lLpE1 zwf&z<(f^qt^NlisgtTFNzI6=#SHXJ`KN+w!3r|BtXEey@RO5^23U1|3A!d)yMfT(c zK5MSRMVfi|uEB(5_XRUAlT|FZ%Z?S-WO1%Diut)2HT>)D%Z&Qh(OFBg(WQ7aZc*1O z8-K+W?S*-0AoWWSn(0W>4%cGGLq^rln&Z-rW84LMZ{el1k&t3_l6y-$EX=Z;*n{FH zc(zK4)%o!}`%9{9UwQ^IQ$1E)eTuqUi?b>xOZ0wZ#KSi;aH&ibs~F+T5v2?O z3dZ%gFD?V4Qv6XSK!Nq|3dK8?xU!=dP&F_}tia3SSJY1r{sAKjPLfrCq~VCK+6y5r*> zw%gx|DNXyrjav5}jE8q%?TYbq>~}+eLv8SL!;Tbs0x}SvW#6{(tsK_8>x8y z6S}~48@(_#kFH5S&bNVF@UW;Xh|g@t)VEbA;Tw+Hj*U21-HOeMU&s6c<}i^OQ`THK zmR;xB$*&(vbC2gxSo!`tE}v0H)n_lG$tjyr!jWStB8PB6^$Dy}+(N%}A zKU5{#m3mw&$1IBrSRHp9--!OgAe~o~lbR!pQkjUxRX4aNb)%R(mj^dYmXx)3w&3po zKQMWv%62Fl;oW!-^j2;X#=TyIvBFC?%nzrzgu*S{u3vnf@8J;~@Ai-ijiRv7 zB$@91?>WhaaM@kvWGTlaAel>bpc>5siQ=oSu_ z$$NTXj|dYD=|wX41NUW1pm4)peU{c0PQ`lx=g+xAH5c#1YwRRg)d+A$?hpLBM+`r0 zFT)(PB8+T_#NAKk61!==0Ha$Z<(=EdL z!ymcfn<7}h%Y$39Bo6mQ_HtW#s%Sv07bmOQ!X?DzLeC$0;@h+i4}?F$;O9wZW1o3FXMXLF5-nFFLB8I9Mkri!Y)rvrP5{3h5i$qahqNN&2%!u zrd?5ZVdSu>KA`P9>(Rw4-uu zRR17VwB3UJ$?72X;VA6)JxN^>6zGXog6BVNrjm8auu8=ataemmx_dvpJvq!`=yg6G z+Mh`~&+4*Yxna1-ypPsSUygfw{4w3<5|+9b;Mfx@aEn(g+CIug%RvdqQc9+^C%Ud0ubK}xFz6dT^f9JvypHX+2Qb87fJyWBvpoITaT&kdluWY4IPq&6n6CWrG z8ljE5)!xu&YERI5*%_REUmh1#Z3j8)9NHms0Q^MUsoa?hoMW~cwl)>uOP9?sw%c83 zTXQGCE>^RyELyam8!k-zz3~Sw7t?-*kV7Ot57%O zuFGqn!i?RZ&d(>#P54cp=S0y?F1U9&Yz@cq3M_u0_~EQ)Jf&i^KiEsQC!w*hdOWoUWilBMQgb zS(h2cM&k~#Cn$aC1+{nH$*m|d;bw2n!tpFA{^`jC&J3_Ck+m z$0VV_jd#@N@D8WC8`jNxPmc@^3w1wga0bf9F?8}-Jf&ZO3hVFScdX{}21|wI zj}^c^cN)r141($p9k{+*oqeAA1wTJnkJIB~@MpI(i`*2C8>F9NV1hFX>&eI2Zhc&R zxfHBS-@ux_8#5pAE9k6um8ME4;a9s}&gOzK#(f-1<`wpU&V~YfT9%KUK|Nr({GCwo zr8o8uTGL6}4Pk4xGRh21$7qi=3 z)*c6Y$Al}~oZ-A=t!cy=S!-m@SlRg0+F zuIWrQO@qz2n2t^M=eUoS3Pf9^8lUpa{)c>ypyRbHhT~(w`e0>}nSTnq{A_-*s zU0{2RG|mhAg3ChtX`Of&YPtg+T-6sldIA3=gm684ZEKQ# z6vv%TqIG#9cwa^e54A5sxqwoD{wlm`#%Obd0IZaHxzQ>0;w??z5cK@x+%E`_{^6Wou@$SD@R65QKy2XnH%VBpOTddlmUZ;am-_i#!I1K73bFPF6?mK&{h6~t4@!SiT4qYQwHPJu0p;hIVPE;NCx?u;gJjp8r8$-Q{Q!I46$CkBedt zBGbXsQw6%(BypyiDeFID%T(hN*dLuyMB}6zJUQ}{s}800zC?yPZ!+=l z4ausOT$f?f14L8N0p*E(euxm%RR{kDrhp96lltEC8Eggi&_m{Q34 z7XeSAKjOjGN*q1Umad@)*vUGt{k%dKHJ{kkr^Uo8%{Lu)|} z5=fUoQ&^<5d9@gPp%UCg0wAHk9S+ol315gfvz_hELe4UbX@*9DjN&8iX>QtzEwH0^nJ%Q6hHQA$*Yxvn*hI@J80Ld0VOv{6}@cn2z9Qx1-vQvG@mt&hq zgv2xpZ;4i(Au){|zFR1WD79tVbI#&Hi7qm~xe)&~I+BfhN5i@wqaY{bK2}aVP8V#* z13S&lWY230cD<;Kg^!S9%Lm)Q?eSE66!4TAG5$Fm9xI0?pBZj@Zw8PNk41Aw;BVPX z78I$$